From 20ef271aa2ec4e7e051ddc722bbf4c1d5021d833 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Mon, 10 Mar 2025 12:49:30 -0700 Subject: [PATCH 01/90] Uploading Michael's notebook before testing --- .../validation_of_application_scorecard.ipynb | 1448 +++++++++++++++++ 1 file changed, 1448 insertions(+) create mode 100644 notebooks/code_sharing/credit_risk/validation_of_application_scorecard.ipynb diff --git a/notebooks/code_sharing/credit_risk/validation_of_application_scorecard.ipynb b/notebooks/code_sharing/credit_risk/validation_of_application_scorecard.ipynb new file mode 100644 index 000000000..17a1007e7 --- /dev/null +++ b/notebooks/code_sharing/credit_risk/validation_of_application_scorecard.ipynb @@ -0,0 +1,1448 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Validation of an application scorecard model \n", + "\n", + "As a model validator, your task is to independently assess the application scorecard model developed using the ValidMind Library, which is based on Kaggle's [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data. Your role focuses on evaluating the model developer's work by conducting thorough testing and validation, potentially including the use of challenger models to benchmark performance.\n", + "\n", + "An application scorecard model is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", + " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", + " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", + "\n", + "This interactive notebook provides a step-by-step guide for:\n", + "\n", + " - Loading the developer model and verifying the data quality steps performed by the model developer.\n", + " - Independently replicating the model's results and conducting additional tests to assess performance, stability, and robustness.\n", + " - Setting up test inputs and challenger models for comparative analysis.\n", + " - Running validation tests, analyzing results, and logging findings to ValidMind." + ] + }, + { + "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 [Get started with the ValidMind Library](https://docs.validmind.ai/developer/get-started-validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models, find code samples, or read our developer reference.\n", + "\n", + "
For access to all features available in this notebook, create a free ValidMind account.\n", + "

\n", + "Signing up is FREE — Register with ValidMind
\n" + ] + }, + { + "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.\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", + "## 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": "code", + "execution_count": null, + "metadata": { + "jupyter": { + "source_hidden": true + } + }, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "\n", + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", + " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Initialize the ValidMind Library\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", + "\n", + "\n", + "### Get your code snippet\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 **Model Inventory** and click **+ Register Model**.\n", + "\n", + "3. Enter the model details and click **Continue**. ([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:\n", + "\n", + " - Documentation template: `Credit Risk Scorecard`\n", + " - Use case: `Credit Risk - CECL`\n", + "\n", + " You can fill in other options according to your preference.\n", + "\n", + "4. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "\n", + "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", + "import validmind as vm\n", + "\n", + "vm.init(\n", + ",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Initialize the Python environment and import Model Developer Model\n", + "\n", + "Next, let's import the Model Developer model, used as the model developer as the champion model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Preview the Validation Report Template\n", + "\n", + "A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", + "\n", + "You'll upload documentation and test results into this template later on. For now, take a look at the structure that the template provides with the `vm.preview_template()` function 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", + "## Load the sample dataset that Model Developer provided to the Validation Team, that was used to develop, train and test the model.\n", + "\n", + "The sample dataset used here is provided by the ValidMind library. To be able to use it, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.datasets.credit_risk import lending_club\n", + "\n", + "df = lending_club.load_data(source=\"offline\")\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Obtain the Prepocessed dataset from Model Developer for data quality testing purposes" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "preprocess_df = lending_club.preprocess(df)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Obtain the final Feature engineered dataset that Model Developer uses to train and test the model\n", + "\n", + "In the feature engineering phase, we apply specific transformations to optimize the dataset for predictive modeling in our application scorecard. \n", + "\n", + "Using the `ending_club.feature_engineering()` function, the Model Developer conducted the following operations:\n", + "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", + "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "fe_df = lending_club.feature_engineering(preprocess_df)\n", + "fe_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## As a Model Validator we want to split the featured engineered dataset into train and test for Validation testing purposes. In addition, as a Validator we also want to include challenger/benchmark models.\n", + "\n", + "In this section, we will to a train and split randomly as the Validator want's to independently challenge the developer \n", + "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`). \n", + "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Split the data\n", + "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", + "\n", + "x_train = train_df.drop(lending_club.target_column, axis=1)\n", + "y_train = train_df[lending_club.target_column]\n", + "\n", + "x_test = test_df.drop(lending_club.target_column, axis=1)\n", + "y_test = test_df[lending_club.target_column]\n", + "\n", + "#now let's apply the order of features from the champion model construction\n", + "x_train = x_train[cols_when_model_builds]\n", + "x_test = x_test[cols_when_model_builds]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cols_use = ['annual_inc_woe',\n", + " 'verification_status_woe',\n", + " 'emp_length_woe',\n", + " 'installment_woe',\n", + " 'term_woe',\n", + " 'home_ownership_woe',\n", + " 'purpose_woe',\n", + " 'open_acc_woe',\n", + " 'total_acc_woe',\n", + " 'int_rate_woe',\n", + " 'sub_grade_woe',\n", + " 'grade_woe','loan_status']\n", + "\n", + "\n", + "train_df = train_df[cols_use]\n", + "test_df = test_df[cols_use]\n", + "test_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### As a Model Validator I also want to investigate potential challenger models - Let's train two challenger models as basis for the testing" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Define the Random Forest model\n", + "from sklearn.ensemble import RandomForestClassifier\n", + "rf_model = RandomForestClassifier(\n", + " n_estimators=50, \n", + " random_state=42,\n", + ")\n", + "\n", + "# Fit the model\n", + "rf_model.fit(x_train, y_train)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Second Challenger Model a Logistic Regression\n", + "from sklearn.linear_model import LogisticRegression\n", + "\n", + "# Logistic Regression grid params\n", + "log_reg_params = {\n", + " \"penalty\": [\"l1\", \"l2\"],\n", + " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", + " \"solver\": [\"liblinear\"],\n", + "}\n", + "\n", + "# Grid search for Logistic Regression\n", + "from sklearn.model_selection import GridSearchCV\n", + "\n", + "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", + "grid_log_reg.fit(x_train, y_train)\n", + "\n", + "# Logistic Regression best estimator\n", + "log_reg = grid_log_reg.best_estimator_\n", + "log_reg" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Compute probabilities as this is the raw probabilitistc output from the models of interest" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", + "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", + "\n", + "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", + "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", + "\n", + "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", + "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Compute binary predictions" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cut_off_threshold = 0.3 \n", + "\n", + "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", + "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", + "\n", + "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", + "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", + "\n", + "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", + "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Document the model\n", + "\n", + "To document the model with the ValidMind Library, you'll need to:\n", + "1. Preprocess the raw dataset\n", + "2. Initialize some training and test datasets\n", + "3. Initialize a model object you can use for testing\n", + "4. Run the full suite of tests" + ] + }, + { + "cell_type": "markdown", + "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.\n", + "\n", + "This function takes a number of arguments:\n", + "\n", + "- `dataset`: The 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", + "With all datasets ready, you can now initialize the raw, processed, training and test datasets (`raw_df`, `preprocessed_df`, `fe_df`, `train_df` and `test_df`) created earlier into their own dataset objects using [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_preprocess_dataset = vm.init_dataset(\n", + " dataset=preprocess_df,\n", + " input_id=\"preprocess_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_fe_dataset = vm.init_dataset(\n", + " dataset=fe_df,\n", + " input_id=\"fe_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_train_ds = vm.init_dataset(\n", + " dataset=train_df,\n", + " input_id=\"train_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_test_ds = vm.init_dataset(\n", + " dataset=test_df,\n", + " input_id=\"test_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Initialize a model object\n", + "\n", + "You will also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data. You simply intialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_xgb_model = vm.init_model(\n", + " xgb_model,\n", + " input_id=\"xgb_model_developer_champion\",\n", + ")\n", + "\n", + "vm_rf_model = vm.init_model(\n", + " rf_model,\n", + " input_id=\"rf_model\",\n", + ")\n", + "\n", + "vm_log_model = vm.init_model(\n", + " log_reg,\n", + " input_id=\"log_model\",\n", + ")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "### Assign prediction values and probabilities to the datasets\n", + "\n", + "With our model now trained, we'll move on to assigning both the predictive probabilities coming directly from the model's predictions, and the binary prediction after applying the cutoff threshold described in the previous steps. \n", + "- These tasks are achieved through the use of the `assign_predictions()` method associated with the VM `dataset` object.\n", + "- This method links the model's class prediction values and probabilities to our VM train and test datasets." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# XGBoost Model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=train_xgb_binary_predictions,\n", + " prediction_probabilities=train_xgb_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=test_xgb_binary_predictions,\n", + " prediction_probabilities=test_xgb_prob,\n", + ")\n", + "\n", + "# Random Forest Model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=train_rf_binary_predictions,\n", + " prediction_probabilities=train_rf_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=test_rf_binary_predictions,\n", + " prediction_probabilities=test_rf_prob,\n", + ")\n", + "\n", + "\n", + "# Logistic Regression \n", + "vm_train_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=train_log_binary_predictions,\n", + " prediction_probabilities=train_log_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=test_log_binary_predictions,\n", + " prediction_probabilities=test_log_prob,\n", + ")\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Compute credit risk scores\n", + "\n", + "In this phase, we translate model predictions into actionable scores using probability estimates generated by our trained model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", + "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", + "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", + "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", + "train_log_scores = lending_club.compute_scores(train_log_prob)\n", + "test_log_scores = lending_club.compute_scores(test_log_prob)\n", + "\n", + "# Assign scores to the datasets\n", + "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", + "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", + "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", + "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", + "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", + "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Model Validation Testing\n", + "\n", + "In the section below you (Model Validator) will select a series of tests from ValidMind in order to Independently challenge the Model Developer evidence and Assessment. In addition to this the Model Validator will also configure custom tets (tests not available out of the box). The focus will be on the following testing:\n", + "\n", + "- Ensuring Data used for training and testing the model is of appropriate data quality\n", + "- Ensuring that Raw Data has been pre-processed appropriately and the resulting feature engineered dataset reflects this\n", + "- Comprehensive testing around Model Performance of both the Developer Champion Model and challenger models developed by Validator\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "##first we start with Data Quality Testing, and we are going to leverage ValidMind tests for this purpose\n", + "#Explore tests\n", + "from validmind.tests import (\n", + " describe_test,\n", + " list_tests,\n", + " list_tasks,\n", + " list_tags,\n", + " list_tasks_and_tags,\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#let's find the data quality tests relevant for a classification use-case\n", + "list_tasks_and_tags()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#list the tests that we want to run as a validator for data_quality\n", + "list_tests(\n", + " tags=[\"data_quality\"], task=\"classification\"\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "dq = list_tests(tags=[\"data_quality\"], task=\"classification\",pretty=False)\n", + "dq " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#now let's run the list of tests, first focusing on dataquality of the datasets used for training, and then we will do a comparison test from raw to final train datasets, i.e. has the developer addressed potential concerns\n", + "for test in dq:\n", + " vm.tests.run_test(\n", + " test,\n", + " inputs={\n", + " \"dataset\": vm_preprocess_dataset\n", + " },\n", + " ).log() " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Let's do a comparison of the raw dataset and the pre-processed dataset, i.e. has the Developer adjusted the processed the dataset from raw to pre-processed appropriately" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "######next let's do a comparison test between the pre-processed data and raw datasets\n", + "for test in dq:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Now let's do some independent testing with regards to performance of the champion model (xgboost). \n", + "#First we want to test independently on the champion model and then we will move forward to add challenger models that we have trained and defined before\n", + "list_tests(tags=[\"model_performance\"], task=\"classification\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "mpt = ['validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion','validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion',\n", + " 'validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion', 'validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion','validmind.model_validation.sklearn.ROCCurve:xgboost_champion']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in mpt:\n", + " vm.tests.run_test(\n", + " test,\n", + " inputs={\n", + " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#excellent- we have now conducted similar tests as developer in order to verify the results. Now let's provide some challenge by introducting challenger models\n", + "\n", + "#first let's provide some identifiers to the tests since they are now challenger model tests\n", + "mpt_chall = ['validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion_vs_challengers','validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers',\n", + " 'validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers', 'validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers','validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "\n", + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + "\n", + " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", + " - The metrics for the champion model compared agains the challenger models\n", + " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", + "\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#let's do same performance tests as above, but let's challenge the actual model itself and add two additional benchmark models\n", + "for test in mpt_chall:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Now let's dig a little bit deeper into one of the tests that allows the Validator to custoimze parameters and thresholds for performance standards - and let's diregard RF model as we have\n", + "### learned that the RF model is not a viable candidate based on the performance metrics.\n", + "result = vm.tests.run_test(\n", + " 'validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold',\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], 'params':{'min_threshold': 0.35},\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Robustness and Stability Testing Comparison Between the Two Models\n", + "list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Let's see if models suffer from any overfit potentials and also where there are potential sub-segments of issues\n", + "overfit_testing = ['validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression','validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression'] \n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in overfit_testing:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Now finally let's conduct robustness and stability testing of the two models:\n", + "stab_robust = ['validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression'] # 'validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression'\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in stab_robust:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Let's verify the feature importance and inspect differences - different models might have more intuitive feature impacts that might lead to decisions in selection of a model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "FI = list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", + "FI" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in FI:\n", + " vm.tests.run_test(\n", + " \"\".join((test,':Champion_vs_LogisticRegression')),\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Let's finish off with a custom test example - scoring (customization of output to a FICO score type)\n", + "import numpy as np\n", + "import pandas as pd\n", + "import plotly.graph_objects as go\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", + "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", + " \"\"\"\n", + " Analyzes the relationship between score bands and odds (good:bad ratio).\n", + " Good odds = (1 - default_rate) / default_rate\n", + " \n", + " Higher scores should correspond to higher odds of being good.\n", + "\n", + " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", + "\n", + " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", + " \"\"\"\n", + " df = dataset.df\n", + " \n", + " # Create score bands\n", + " df['score_band'] = pd.cut(\n", + " df[score_column],\n", + " bins=[-np.inf] + score_bands + [np.inf],\n", + " labels=[f'<{score_bands[0]}'] + \n", + " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", + " [f'>{score_bands[-1]}']\n", + " )\n", + " \n", + " # Calculate metrics per band\n", + " results = df.groupby('score_band').agg({\n", + " dataset.target_column: ['mean', 'count']\n", + " })\n", + " \n", + " results.columns = ['Default Rate', 'Total']\n", + " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", + " results['Bad Count'] = results['Default Rate'] * results['Total']\n", + " results['Odds'] = results['Good Count'] / results['Bad Count']\n", + " \n", + " # Create visualization\n", + " fig = go.Figure()\n", + " \n", + " # Add odds bars\n", + " fig.add_trace(go.Bar(\n", + " name='Odds (Good:Bad)',\n", + " x=results.index,\n", + " y=results['Odds'],\n", + " marker_color='blue'\n", + " ))\n", + " \n", + " fig.update_layout(\n", + " title='Score-to-Odds Analysis',\n", + " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", + " showlegend=False\n", + " )\n", + " \n", + " return fig" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " param_grid={\n", + " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", + " \"score_bands\": [[500, 540, 570]],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Finally we got all of the tests from the Developer that was provided as evidence, now as a final task we will verify testing being appropriately recorded" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.utils import preview_test_config\n", + "\n", + "test_config = {'validmind.data_validation.DatasetDescription:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", + " 'validmind.data_validation.DescriptiveStatistics:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", + " 'validmind.data_validation.MissingValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}},\n", + " 'validmind.data_validation.ClassImbalance:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 10}},\n", + " 'validmind.data_validation.Duplicates:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}},\n", + " 'validmind.data_validation.HighCardinality:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'num_threshold': 100,\n", + " 'percent_threshold': 0.1,\n", + " 'threshold_type': 'percent'}},\n", + " 'validmind.data_validation.Skewness:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_threshold': 1}},\n", + " 'validmind.data_validation.UniqueRows:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 1}},\n", + " 'validmind.data_validation.TooManyZeroValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_percent_threshold': 0.03}},\n", + " 'validmind.data_validation.IQROutliersTable:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'threshold': 5}},\n", + " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.MissingValues:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'min_threshold': 1}},\n", + " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'default_column': 'loan_status'}},\n", + " 'validmind.data_validation.DescriptiveStatistics:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.TabularDescriptionTables:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.ClassImbalance:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 10}},\n", + " 'validmind.data_validation.UniqueRows:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 1}},\n", + " 'validmind.data_validation.TabularNumericalHistograms:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.MutualInformation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'min_threshold': 0.01}},\n", + " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.HighPearsonCorrelation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}},\n", + " 'validmind.data_validation.WOEBinTable': {'input_grid': {'dataset': ['preprocess_dataset']},\n", + " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", + " 'int_rate': [10, 15, 20],\n", + " 'annual_inc': [50000, 100000, 150000]}}},\n", + " 'validmind.data_validation.WOEBinPlots': {'input_grid': {'dataset': ['preprocess_dataset']},\n", + " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", + " 'int_rate': [10, 15, 20],\n", + " 'annual_inc': [50000, 100000, 150000]}}},\n", + " 'validmind.data_validation.DatasetSplit': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.model_validation.ModelMetadata': {'input_grid': {'model': ['xgb_model',\n", + " 'rf_model']}},\n", + " 'validmind.model_validation.sklearn.ModelParameters': {'input_grid': {'model': ['xgb_model',\n", + " 'rf_model']}},\n", + " 'validmind.model_validation.statsmodels.GINITable': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model', 'rf_model']}},\n", + " 'validmind.model_validation.sklearn.ClassifierPerformance': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model', 'rf_model']}},\n", + " 'validmind.model_validation.sklearn.TrainingTestDegradation:XGBoost': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'},\n", + " 'params': {'max_threshold': 0.1}},\n", + " 'validmind.model_validation.sklearn.TrainingTestDegradation:RandomForest': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'rf_model'},\n", + " 'params': {'max_threshold': 0.1}},\n", + " 'validmind.model_validation.sklearn.ROCCurve': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}},\n", + " 'validmind.model_validation.statsmodels.PredictionProbabilitiesHistogram': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.statsmodels.CumulativePredictionProbabilities': {'input_grid': {'model': ['xgb_model'],\n", + " 'dataset': ['train_dataset', 'test_dataset']}},\n", + " 'validmind.model_validation.sklearn.PopulationStabilityIndex': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'},\n", + " 'params': {'num_bins': 10, 'mode': 'fixed'}},\n", + " 'validmind.model_validation.sklearn.ConfusionMatrix': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.MinimumAccuracy': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.7}},\n", + " 'validmind.model_validation.sklearn.MinimumF1Score': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}},\n", + " 'validmind.model_validation.sklearn.PrecisionRecallCurve': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.CalibrationCurve': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.ClassifierThresholdOptimization': {'inputs': {'dataset': 'train_dataset',\n", + " 'model': 'xgb_model'},\n", + " 'params': {'target_recall': 0.8}},\n", + " 'validmind.model_validation.statsmodels.ScorecardHistogram': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'score_column': 'xgb_scores'}},\n", + " 'validmind.data_validation.ScoreBandDefaultRates': {'input_grid': {'dataset': ['train_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'score_column': 'xgb_scores', 'score_bands': [504, 537, 570]}},\n", + " 'validmind.model_validation.sklearn.ScoreProbabilityAlignment': {'input_grid': {'dataset': ['train_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'score_column': 'xgb_scores'}},\n", + " 'validmind.model_validation.sklearn.WeakspotsDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'}},\n", + " 'validmind.model_validation.sklearn.OverfitDiagnosis': {'inputs': {'model': 'xgb_model',\n", + " 'datasets': ['train_dataset', 'test_dataset']},\n", + " 'params': {'cut_off_threshold': 0.04}},\n", + " 'validmind.model_validation.sklearn.RobustnessDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'},\n", + " 'params': {'scaling_factor_std_dev_list': [0.1, 0.2, 0.3, 0.4, 0.5],\n", + " 'performance_decay_threshold': 0.05}},\n", + " 'validmind.model_validation.sklearn.PermutationFeatureImportance': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.FeaturesAUC': {'input_grid': {'model': ['xgb_model'],\n", + " 'dataset': ['train_dataset', 'test_dataset']}},\n", + " 'validmind.model_validation.sklearn.SHAPGlobalImportance': {'input_grid': {'model': ['xgb_model'],\n", + " 'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'kernel_explainer_samples': 10,\n", + " 'tree_or_linear_explainer_samples': 200}}}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for t in test_config:\n", + " print(t)\n", + " try:\n", + " # Check if test has input_grid\n", + " if 'input_grid' in test_config[t]:\n", + " # For tests with input_grid, pass the input_grid configuration\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", + " else:\n", + " # Original logic for regular inputs\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", + " except Exception as e:\n", + " print(f\"Error running test {t}: {str(e)}\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "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": "3.10.16" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} From eb3ada704296c5815ae6f2bfaa6946e7ca7c0d0d Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Mon, 10 Mar 2025 13:43:28 -0700 Subject: [PATCH 02/90] + champion model file & moved draft to tutorials --- .gitignore | 1 + .../validation_of_application_scorecard.ipynb | 0 .../model_validation/xgb_model_champion.pkl | Bin 0 -> 117919 bytes 3 files changed, 1 insertion(+) rename notebooks/{code_sharing/credit_risk => tutorials/model_validation}/validation_of_application_scorecard.ipynb (100%) create mode 100644 notebooks/tutorials/model_validation/xgb_model_champion.pkl diff --git a/.gitignore b/.gitignore index 0ce6f7c95..a3b44c3c4 100644 --- a/.gitignore +++ b/.gitignore @@ -191,6 +191,7 @@ lending_club_loan_data_*.csv # Pickle files *.pkl +!notebooks/tutorials/model_validation/xgb_model_champion.pkl notebooks/llm/datasets/*.jsonl diff --git a/notebooks/code_sharing/credit_risk/validation_of_application_scorecard.ipynb b/notebooks/tutorials/model_validation/validation_of_application_scorecard.ipynb similarity index 100% rename from notebooks/code_sharing/credit_risk/validation_of_application_scorecard.ipynb rename to notebooks/tutorials/model_validation/validation_of_application_scorecard.ipynb diff --git a/notebooks/tutorials/model_validation/xgb_model_champion.pkl b/notebooks/tutorials/model_validation/xgb_model_champion.pkl new file mode 100644 index 0000000000000000000000000000000000000000..bfe7349b673d72f09dc01d5d1640bd3d62a5aa07 GIT binary patch literal 117919 zcmeFa2VB)k`u|U{fS^jcN*qAPISv~h=@@5rHk%X|=_=xgs zbGIffyqbG?d3!haXxYr&-P_GeE^eF<7CAUFF<8@j`2qH$9D->}k1#5<&<$Q4xW9}$w66dw^B6A~Sf(5s%DZd_hjwUC&Yq>!lK$e6I; zVX=SfmDSLQcv@SS=2U_c5@|IFzbjBJB07$CEM`FBz~7~tM8?q4QBlzmF^Rv+FrWhy z{ks(Nfw9pM!Lh^Wz$OfgjQd@lQCw1dTx>$b?-Gq;<04{$L&ARF)vAfHw88%)**KDR zfL@BO|^B>P6~DR*0+^St+t|q(P)%WR*yx zNaM(=k<}tiB26RBBF!T$A}u4UN7jh6inK0oRMY3knEtVUtInEM_fP#NS5{He`|m0$ z9{}Uf5SW+|kpl)MCiJQwUjAp5A`SjhL2a|jpB=Z?UftI|MtWvy!xT>&uk`$QTpfLX zo4U!3TWZsGyQOg-KNBW*4oh0xV@sz!RYu;hRwsG&R{N{EsD1jLQde5sP+jXxMJaIF z0BKgq8Fug6bcrcx%62;v$W9dPl)c*>V7ZohO2rCGWk5v><+=?=)#;=AsH4lUG~4m4 zx{dDvHCH`FZM)~2+M~o?+CJ;By570D(m8_->WUXzk-gK$NNY<+lA|vorK(kTDh=x$ zlddc@Bu`x%OD^Z$u&a(VWqK66V2eJCRC-?Ts+@K88MC*nGrM`Bcm%StGuY>GwbQeh6zORkFBE zmK$#-GhM@r77QX%wM8G9K9Ra?yBnvO{QM76o%_$pcHK&3+$k5ujn`{Q@r*P@VV$?W zPv3814V;@0=MGPqY5mVI!|YqJ)wrF^wIjJ~7UrEe%JDPQnYur|4by0iC$!QWcdJ9g+2y{4UCKmkB^A?ehi_rx)gm% zMM`CgA;pMNm106MqgYU?Q>-X8DYYrKl)4mqNS(v;#(X+~*IX-V;> z_)-{(lEQtXfyS63Qd&{kP})%fC>zhM z;y9c&dIN48H=JVz=%6=?Qu#$MQuzpTOWbf$CVy9q<8#dV@(V8_G&)>I+4jB;G6>f} zcRFt5zRl3#7Dd^=X0kyP% zIRk98fN2BjXsJtMz)p(;je&Yv>eCo#pv95KfRh$y8UqcrxX>7I)zXN@Kw~XUXbd#f z;znb@U5f{efo57fX$&;i(t^f7OD$eB2E4WS&=>$S3@EgKNd{Ozo)FdI;c@JTw-1;D zaNXfr!}W!03fB>?9b7NCMsQu=TEO_jn8P^3*xFHG9AWHWykLxATwpBV{NbG8JmK8n zeBd0Q-_T#^BckwN!f7l}{3xv{Z7IxV zM2VurP~s@@ltju<%5cg^N;0(jGX@w~2&M^fQFI9%78?^D3Cr@|-$K}R`n{wMoS^J6 zp*@?uWuuJJY4S*EZ65~!w!cR%L;BjiC+?#LvzEc4s+Vyb%dV^oX%I8pzI4&#(|x7h z&{M2_YP2$0oT?~kGJttFs6>-*AJmNLyzUD;utOTO!BqODfEenPk7s%YCO|o8Lq^jE zlr?j_u7S1zbwWPGa{HQNVISJ0bpV96s~i%S^|L>)SN&~f(Di3z*zcdlQ|P<_10frK zm`8E|BLaBO z0C`aEd!P*{3o!t7%l&{@ZeMQe`#R*~XtAF!$OfazZ_@%~pZ|ReCZhx@s|<{wN?!4d$>QoWE5y3_eqv7P z7tvyZuQ1NJop6vDB{nfyCDg85D)jN5k5g;x5a-mXBvjp6S+p8+8tYZ|7SXkL!pZ)X zg*WNz@qT4}!EyCryr*5BkiX+TE-?2OdL|nPPcH`WR;B9&vusB2&_lS<)gk-`9Dwz+ z4~eP!Z3SN43S0Pm#3zqT7v9|-i?2jwaSsMQ=MNS(7o1-#6>Ahs!AD4tu&~7@q59EK zVfDPvIAe7>|Fv@lzPaiH?%LfOd+X8~b6W`AdJ6blQ!C!K3BpsN#^L$9I||0N%7hH( zFG8n&i}CJ5<{0H27u@;;pi?6-A9F07zu6_9e-v98T@?m!(Mff9+O{xY$$YWZk}lX~ zZDYaOQo-Npy_?tbh`^kfhC{pL3$32s5N19x$DX!NaC)B^T$?z}>=d+Ww)q`jfqQW$`*;ph-5KF{L%0W1Nh?%?cE{MjhvCI+)`D z#;F`VmVa`17=J!18LvHXi7#?<=bw*8{4&3JJRJjmzN;ty^_v$q-O^X^XdJ~K(2c@N zV^ru#QXP(_^J{xO!hUNk%4e)b_WHd@X9W5iRrDefBf`rSI9)gS5KXpL94^n+bN7(* zHS(QCr6|wUclT&6XPbiJ0J3!d;Gq#=RE91urR$!O@;?h@W26>67f7F`e`8(y=BO3@ zhDbYiKV*_>4q-1hXsY&4yQ{7ias(T4&bEYISA0-fwcL}e`H)IJoC#$6OtNBvo7`Z>H}+w#+`OuEX(FoaCNyT( zsOOQT&m38+;M=UP<9zmZ->vNDd$~$xcTFiRJXV^#BV8JPXNDviY><{seJh17F_*fW zLXwA5wuJUiA}+HkOO>L}N`AZBOMx>o$-~SW64$aC$@Y0G`DSNH)puu;GmE;aaviLt z&biY_@r}x4u-ybj^@3yMm`flT>f4%h;HLXdSbd&sPc&gZuUn}&XY_`7>y$&@pK~Hv zOX`yO7T1f1M4cdy5+H35@xyX=hoXM~z z5k~CboFpkd{3uh$r=j9YEW=i)ev4!~FJpEe$WqjE3uZo?eXJ_ndzjgjf1g=BEQM`S zxheDT`c>9*;6v8-xWJBTe@vc4e=Wgm%zyZfD32OQX1emuk0mqcu6)~vTRc2Az@q>> z62O`j)~T>Ih4m<`L1EnqYfV^R!kQA+k@l2&lm-+hN<)e(1=fGC=7V(}tnFYu2WvQ3 zx4}9M)@DiyN5K@4(u&fC(vA{9=|Jg3=|bsB=}zfE=|$;738sWl!YC1x0hCC}U`jM4 zmNJBrKuMwuql}=8qNMzc>z(h2uIQuN|qzkSLN zo^8julzjaC-sa}`&lqwLHJ+}M>@xGm`=?9^#G$ktki z$+Oo{#&p+Fu4|a8oKU3%ndOx!-}ScREZ3HxITAwAP6#b--ogqJUkY~pqNoZrn(0TOT=zs|UKt2KVwaBX;Ff9PcDgc-n09F!! zp%!_?1m*`Y)>4(m0LU-^m?Z!t830TaV6Fws6#!BV0HzB7*#-bJ27rVEfJp;D&H=!@ z0U+%FVCn#nc>pkbfSne58Uqem>d_deucZNv0Y@!PGzOftG^8=$qQ#ZQKqD=UX$&;c z(v-%4n-+H(10GtM(HQX5(wxRX3oR{a40vhrrZM27#h1naNK*iqY5-)ae?&?fL(9QD zfa?#}8?G^2TezNZ{otCxwSsE{*8{Erj5~}qjISL9#uCO4#tg;@#sGF60s&%3lLO?*KwA(X27?7RC@@}d=r$ITCgJ78N{BOw6^bI-2%hJs zQ?c$0_xaO7{8?E?sNLv2u62atFK=+=-?bPkMhvB8E^oq>97FN3J{5!qc}MVtrG?z> zItBdYhu(rsMn}h3hV_Om{ z=ta8V%l;DtwnMbgVpId1{&@wyyLqHgq_*epJ!;GuHtNIeh?s|0%xr=ib}tiZ=GI1s zpECU2CwaVW*Ry>3nkBqPWpl1Wtq1(oE{;Om{!>MdE6auYncMjEMHhIJF8%qvv_D?= z^l{kfih_5=3}JE&L+qct6New!z^!b45(>*op>?? z>7@&NuRX!I!Yd`#xxAM59oL?};`xAQvmfA^Rfppp?dD)Fm(6%|Oh3FnrZzt2CSf0! zSfTRsiSp#A^6wJcTFNc{--~Tk9%o4xw!C6ioikR~VYf>QDvu#;bR*fY^p)x+a~nvv z9iB@ikFQIO*7PQ?9~&@BEjqE48s20V$8}J*$SzeIf7>9vTYNxT`t_!CVsHj|EFO~3 zhvzK2rIo~%`YC3Y&SDfU&g}TeWb&$03%21N59U*t9lQNxnY#61Z*`NfMQSXLQX5ut zSL+t7l}b&fN)LDGN}UfikRF7ukam9PrM6A+BkLTeNUEhtBDgzmnTrfgY3ig+8TVd#8v;#!pCeA-5;(kPECckU?} zO|h4(Ud&PKUe-pD<#3BEZy6&wyqzoQPT3(E+`{WEtKdDK8`)0208iERmV43j9sC?hDND8IxuItUPcEVc;+q4jawjE=bb>1^KI zpdTM(n$P#z>dtf55oaB$gU0Hnau2QZxxm?FyvhCsT&D-YT^KjmpwdPOG1) zTxynzx)yCjS1;)x)yP$x&JKSxIm?A>rW40yRvFA$Oz4a16mCSRk1ug^r~pYx4sPoejG?#>^$~qJhXOq6{?+FGPE%hN7I7 zb-8gl!_lPK2I&5d zLx6PH2Ehy3f_>Nq(M#U`UI>!~GWmEYC$|B@nGB>sfV$*1pez7sAjm--ve*Xu5FlT! zTa)lJ{!ca_e_63@aZIwGQPeuWmT_&xj&tqB^^@9)svZx;%^7P%>zWIMP(4#|z)*eR zkyuYm*q9?$)JLM{*a*>W!VBTetmXJ&d?%rieIMLC{jk_!&|+ND%S5<#eHOnz=%iTR z|BbNl04rQSVvkoo7>PIXtXOB1t)O=biza9L3DxiK7FHLC+^xfp@RjB>1cP@W*m%VY ze$S0W{!x8V@H+gEchKPtQmzk8I8pci8&yy|Cw`>Y$yqY1jWXy^>V<~ruT z3{(gfyG(>;K?!)%k|BJx`i^+&2$8Gmm5ok}yv|o2S%@7fY{Eu{HMo$}A>55+m-yKo zGle#;eTBo}k2t;eXZXONB>qvITX?PadtvtHFSyP{Z+_d_JwnIh`Q>x>uP3&VlXmx{ z`PQeHf)RVvdQtzJlaZstv#|}|9aql5r=@=#*uxq69AIwNaE0;>Or8H^rmQ=~l zue(%pQyH`Wv%R#BM36`CPLT_pIo96zn6yOQL~=MdT#DSVUBRtCEqQnI-1$}Q&fE-X z$-b_tN7Oz&nLD?#S;LD#Y)+t~RQ;U+TgmVd%Ukqf&)?fG#ZPD;wVilPazE9K#3^n` zIj#q#`qNKJerpDjuQw)5UPyVoz1c$|;WYxo<*NE?Rm*MS6<2%Z zGDXHa$cxo?nN(duj_x0%y60&`cCBKHPk5|REh;#|1P`7^#s|Cm$ZL<`d zg8?|t{M1379lyl3Ut-%YvF(@G_6IkRKTeVj$NwO<$uEn-3!UG;EJ{7kz8BjxLKX;E z;8{k?fuN-k-Ke0Z5y!MIi)utJS_aZI-eq)K6Kg!!)_u`#^fi)|pN$zEaK--#jZd$A3~DG;z^feQpH2#^lq7>HdU*vVxe4f15s4Yr{Sq<=4h z$@?@}Qx?k0Vi}~#0-1cgoF@xtPzJW49su$I*oI>uK)a9zu`J9%{8IaNFi{!~yHC3VQc;;mnK)!k3ikLWh)(LWRVZc+uWo_`>N+ z;#8yFV$Iz7xG?R#kiOd&f4p%{m@)4ar!oz|dvkjUHQO}eDqS+1ILnZA0&(mFbK4_B$)_)gK?>LswShL$am__RqSD zH5!y)hxVrU9d5?&j>UM^{fS&7DV4vKbC}P+x{A-ZW2;);$PjjZH|y)>vSe31P8Klbz?>LQ9oaYDVD6!(Y|WSL2aZLbyKDJ1ySVP z?Rrw$p@Sr+bEeeTY7dF`I4!k(zLc~wxIhwac{AaGpUIw+H<@v7?kG-EIS32%?Uz*!zw$$C&N-`p=n4>+tq$!Ui^7gW^ zq*Jw-&)g{nQdTXN| zTql`%xnPazvBd`F$kC3hzg>XT<=!o+!iCY&@SDpdkHooB{hg_jp5r>n{n2&O>+>jP z;(@`EkU^bD8Z9G>44owdQJ>v#J;e9T;UQ9o+on1Q7DOWF(b{NZ(yQtrz--Fs+n=7^O$AJhA9@>tWeD;j$^pCZ;Cc6 zcQ9eTrzML~nM}&W*5%Xom)Q1C+Mb{)|rM1_uFw^{f==~5hKyNOvG`M;*=ru zm)8Qs5@hI>iYC3WK&J~4nmoiCrHxKSb8Q--9aXMyRd&8W6Hmr*-N(#8^WA%KRaMu~ z*qx4?`MMzGn+qk%k_9EG$@NrJ!TvegdZ<8|qfABQb8S?S?S#e<#OYTvKwO_>G%ENd zimK>_Jc4vk{ViwEfXW{c+sqEx>1{?^nMvqNe+Se;w-O4pzl*$tZJbBId?bvrQGT## z%^8e-iyS7JBAbYXh&eou>+QUp^Y{dhz&~}6XU8wG?U&g0OKkfkw*A2k%pFEa2_0}%{f@ze@# z8c#R*byF(Rfe@w<)wEu48m}|D4fVse2G|B+4LsziD5fb-kAr?dnkJq03+mP!4|Rf& zrWN#L&%EEf+kWd4)Ti-kqh&PtR9yV^X#d-t_0PpN5XeAO``)un7R=yy5UfDB0-*}F z;f^hfSn@W+Ac{d55WZj=L^C)}4&NUGc~BMtl!tBk*zfn@7`ZRdhAf~#x?CTWhdekQ z(&T!ftSqd_b<(8r<3LROse^1l{<30QUi~NH6Yi3q_sS>2=Zik#dZS6=&c>I-qWU_* z))5bcYEpsNuVkL^cI^kz!wL2-Kx)%xt1(moxGbf6z9z4YR zw@wt;y-C9Cwrauk<{;r)*;!%yv0Gw=Y!<(%xk5lI#|Z8Cy<+^TkGM`*JRZL>Pl#?b zf}gW}6dxE~0l(dzjXvmK<;HiODUh9E!l_zA@btmdgDzEczVa957UPkZ zU-JRJZ}8|(5&WSQRrt3}Z{z6;rt|YF7U4d}p5iP09`VzjUJ@FwyDW5_?9J_%ohF(N zs#HFA|AJy$L`p5GdB@?5-I9vZ>RMmP)3&F*BNMWilAvAEzA64vcfHlpu*Ux6fOP}1 z?0r0gQ(iLerK{NLM%~n=Qzl7L=4WYqVSUMg875`zOD7(6)l&UyhQu%9y5znPkq&Il zBKwvhj5^Lx@?f)>h=iA9VBJNsBXGf1Z~Idf+`Ghjr2a-h%?s zdH30>m8*gjD@!k`UOVM0F6(qud}N+6n;P7e6rv$naQ+(Eu;3iCzpgJ+dU+A4RO@b0 z>!V*4&(<)ir=d%j?w1ZRdCL+Mob4p0{^Id$@$prRSv?i=*<)AvwEZQv{S&uddDK8` zt0?dM65D=>ZGW&ZlBdU?h;6;Te&!eQiQG11!VMkQkL&-Yj4N4qlJgF|jS~Hk@@e)o zl-@p$OYfYDj@_{2j9RJCf#`Q!;9G=xE!ROY2X#>YYXgzXGB@PCBo1x2P2(zG+lb~z zo1trW^Ej8uce&cZ#whuK7g|zr9*VO#g!X)xi!QaXQ?^@^idxSZggl4qprK!eDl1hj zL3N%UV$Vx%%DIm$k;{2IG;4TQWD{P1qVHco-RA0`E(Y(>)MpbpiwEB5+MZP8W1NXH zdyGS)FO{O*Zs|u2GTXc81S=g6EpH4UH!oh;&ef zigz0CG_7YFl+heV>xXSC`lx3fq=AU2DMz>ASdDO} z73F?C+W&UP{ByBQ_H6rJ7z2?C1S|-$hy?k~LEoS(oC9nFP(RcOG39^hK=XiOGXffYjR2>KrcVAd0PQ12DI9jEl;Mv&gC8n$c3-!^I|NCy+~mXFtQAxRV5zP96cJP8r29?ufJcPNBvZY9En{_lln zUMk`9>NNhHrL|aL=XzmqyE0+^^P&9mi|@GC`#K5hClAHC*I(m~5659EBOhLWaT;FQ zrW5{PpNTtdtIJ&-(SlE#YAN*iq$jLjJXQEQ=e`iR&JHhY^AyK4br4eT`w30HgyTam zN(EJNTf8)-wQxGyjo<#QnETvm7|zUoBHWwS4}TbLkFVBUBseJz1oOqW`Irxp97!@j z=dQKj)K&Ki%?t#=#e602z++J-ai9U!@y@-d%0TNj)3$l~*7f-J}YB7&${&`{ong zGG!MwI;FPhwRHr?YIRXwVrLxd9)#Deu_>Rse>t%&g*bcOk}OZZQOsL?NSbcj zgBjBTGi&dVw6W2xP`b*jB~y-Dps>CEk#R#J*yl8=5y z4>JDzH75JK3;W)9iPT7;uSiPi&o1lcmvH zn6`C>5~fZGxs&OmT3E)AH7`1_R?Wi{RlaqXj9bhnNeerXHkX>KZZ>z4>a42HI(!^Q zS{3wDESR;CBusj#n()#|b@Y;vZ|<75q_je9X4|$ds&_htOrBVklqR=i{D)1I^lZ+O zv`MeX$d4mc*~6czTIAGWOtRWbHC{AUz3rnSx}M_{s(bH=qgj#SblsVX)qB#IG4v~~ zV<#Uc%k$0`PszW_Fp*J=TeTXhJ1@5TF8olact&5VtXSy*ely&bQkP;+sYhu*aiTP&xKiMEyy3UI;rF_mQQ&vE;kUTq_qX9U zw_)9;q;M2W5h<-GZ7A*F>7B*_ln#_mlrEI6l$18;Y{#1F6{7_m;;U5Veyz}RNT;3@|&fsuIu11alDs0PgwQKc7UQ$1$ zj=@H>^HeJ8zUd>HbEG|*=syu%oY9u6YRD=_uP#B#OGsI-B2tdZor`+!EkSKl@{!^A z3(EKFwjrg%86*zkk?X^cXaK2!UQP@|2fVwX?#b!s{WVWzQ3a$tRS%)K)BTX?%Q$4b z;tEo3?8k-;>rJns4sy)why=UmsLGWnG-#y{^6q>axwKFzAM72gJYKOrnldC@SyJ~s z8gJ1JHSC;*{JTv-UC1Sb1A8eS9&v(3?Ts4GT!r|jJy98I z4Q>34L7o+VB&q%SlIZ`x&me#~1e~>iX#`xffSCjuX#tZ7z)$}JU_JqOlNkV03cx1; z05Gcnd>#M*6AQqn0st_#0DLw80MiS=Cj15CJt<&J`cnE)LMh>t{*-}~L6j&;3?+^d zPf4T;^KKrXBEugSP0t z#wU&*tMQxD_7l=HG8?>PO3P@D2e}N6(Y7f|aM0Gz{=oiY=?n~p5e1BdA?5o3=@7t} z$mr2+U5fl{YCyLuYxf~dp8^45C9$0UP|af}a@h*hhGLWU%nfb%2nU=_wvRyIuGgy%jHC zFiPmC9ECqtb4Ops2Z}9>MJ#<=EnIsV&0oEK56`~YQ@H4HLeL3y5Zu zwSqriGS>%>-~JQ_^u2(eBzDEm5*!2@vw>Xf$4_X-kt}XRVS7H;wmEmF&MjWiZwucb zwE+99OvfYI*$C}Qh6xqhSz%+_M%;z%ru@n&Cvo3x^Ki3-0&INqHGj2hBwjS|CN}7_ zLHX@%A^&;Q0ny>O3r@-N=UYzC!fT4sg_%9_M7__c%93N75Q)y^Q}PG!xjiTH=Zp-n zIzv~8diWWis&QAI6#ur58~n?@B$}dno%c)%=$WQ6KAtHh2302SO1hCi7eB?j)IuqL zO_0>R&JZcOjXk;0V;Y%jH$>$xrVvvDU52e=C2iF&lzJ6sNb?WemzsQzmfqU-lMJp{ zOLe^3kd#KP$<62O33FjEsae^EvGGnO+f8RMsmt7UUO61WWF0C|SL$Ie<%LZpGivvj z_AXFL*B?HU#@sz6`99vp*0LWe<;=s<)CbAZtqMyO^P4x3hE>@?%uo0#KAl=X&goy0 zID>Cw#px81^T^X@L992)J*lh0dzbkN>VvG~0#|86wasMQ=QKru_feLg;-J{wP=|!< zyvA5m9?6>6?~!7yu1N-y^oet|f#l7zEN0o_aAITYD%m6l6+1e0lsw!D$!_z{l8;js znK;3Y#NQsHD(W31RjT@mxiI%5`TD9hvo-9H)aLbi-&+f7Fkz2M$?m1~$eJ(y#67>3 z!mfG{>2C8;)opzmo7%91ytr~iHR(wj*;X=F^~LTg8TRsqYQcerq@8K5YI;F7u?@CT zOb@TAI(@mA88df*YWEf=MZ=f5s;$OeiglhR*}4Yz7@hkki2kIbQr$|+iR0Bz#kF%& zRT*(9iiNKSG4BqiDb6GfBhshIBXGTe&;hfvp*fx9$<%DtKLO= zi{2x}8B^ui_QO##Cnai8@4a%NzAu`8xeNtOk5k^-Sb{Q6uR@oru0-AYvB+bzf@^9q zR5|i&8S45qRXOL89kO0>M!BM}qw@5v_sa2}^xL0vpD8zXnht*s%1ud4&lhLd|y&}TbjD0M{l0*)c&qt7U!U@IEYdJ>w@`4bxBm50VMw#t2X z{7{d@H&N(C2_0G75)}_PgwC{g{)hk6{>uu=v*MRf_W#cJ?3YmXZxYJr-=;<+)80OQ z31$BeZXbW@YnB3^D**8be6%#86ZIMUO-!Pql-9osL?_u_OZFiHK}zFSrWL_J=zo~|@@f5WE|3Pt|LhO!e=d~C0+}3xumnLCrC=Wf zCRxyu1tz(ioCiV|2ySp32vrcvb;u&x_ktMI1N(41)CI@NA{gw;_0Y}VwGGF~eS%{l z$Z3!cbwc}c4E=&_$dl`bH2GM%`Ev#N$>J|6ludiNoPMp-&oA}PRbk!O$AaT@rTFUo z5WmXD(=nOl310){7ta46n3H@l@2Hp9qFcOhe)lHv;`$(*k<^#JGPVbvZtEj7TW`;; z9?_T|@TC`?*f0~%>$Foec)w3{Uc6R#J@lavH~5V(@$NRE=Jq#Y1wB7;_PXYLD%%PB zAD+p_n^x!j-Y(=0XxrK0PX%*3uvV1PbJ(#z>y%yK4lEy!rUW?z` z<~pskQHO~7|u@8Vin zRfO%_IKe){TlbFNYq34K8cE;uYY z?7z#S>|Z97C8##NN|$P#>{NX0K%g|L#ucgJ1}oLsfrA*Uf@M;4Ag8uAT`Eb{^SN6hTTq1}Q7@!^vU# zTV;ihhN&J|lqinY^!9Pvoyw#w`K-9OXdZLY?UK5l?rtgV(_5(^!$wk$NSD%PG$-q- zJ(h-t#Y#q9H%ftStH?6vm+GdzXBeB^Na}rEN9psyTk;KmPR2KWO!fxt^yy#Pj-1;) zPPO3PD^-uZrdL`;B=62}vfcX(Y15)JY17k2RWi7eWVn4H<2!nw@7O&V zq-MzwrfB~IY4VgBs{3|LnWKx+*`yH{Rr}rJB#V2WNe6dVQe)^+pZB((ijTW1h{w{0 zs)6&v6tylGk)g&pzRia(QRRMk@2%gXgy?>&?b~6_Ek#S?`HbuQ`LXN;U6gOxaEpdpH9U^MztvTxm{80p7L@7~D@sjD zZHg_W+&@gazGy(xoG7rq1IGt&b^r$laB2WY25?>ghXrs_0LKJ|Qtl%rKbq4tk>&M#-_+J zC|vn0F%@Oqe!!XDxq|pZiOBnqi}FmG4%@Y?pEBXADU)$Dm0l4YRO#F_<+wa+B(AG~ za)xz9=Z7bwJ8uJ()9>Cu-UXM?z#Uec!=sldzxx=}ae^_sr(v9Olc6rcbNef$Byjc6 z1m)y7H)Tdg19ZI64pe=cpYp3`vT~Wv3CL6i72QPe6_A6H)zzu;Tr%4D!|fBbh2; z44vEGoi+T41MsviKmEg#IXsO6@TU*RD+e$G06eJ!U=jdWDF9#|0Anp+Du8NQz-$1f zTEK+JW&}(L_zMAGQULHB0>Hcg)>^>S0JXG$*#W@62mliVfUgk%<_NIUVozhhK}!P~ z1CCn2Q~}Of8qyeW(b9;<0GKhLi54(vfSVRDZ-9puFm-^Z7BG823oT#*0r(>Tz#IZT zTEH{{;0L9UFLbr|$N>WrgTn?!Muo>m#C-pyodjAAt}9$ixQ1}e;5xzefolTS!JGnP z4&w~t3u6jn2;*i)fibENCXvQ4J}@RQ4siZ(-f*sPesE539?);-FZ7Y6APP?rpbiLh z+mF(k(v||J(Vo&#n+JcGfs}5PAWBb4Z%SWEKT0SioYJ2%kTQr8MTw!rQQ|d?{1^TY z1`5KT%1Q|!EB(Enl}1*g@{R2ORPM6`at-y@Qc~Y6jZ8z!Xrv|ZaiWj6koKEjRaYBX z>XD@?kanP~<`qIJ?Z`1mNg7`%ka@I^x0=@xL5c!szQ*s1Zo}hkxuit*HFZ$AO7j{d z)I((^c%894o%Rd*4*MFp367z@UjOZnw?CGP;13)OFa+6XLkvNW0>&B)I0RXKk&QIu zK^X|J4FO^>5@0Ok#n;eu;ZVfRoMr|LW!lxEi;ObW~572H6D=EouV|k=XEl4 zY;zKqpI^+Tw3HUd)+U8IW68$6evGr_dexJVG-i~Oi(+p#%<8>)Ev-6WCK(ydV|N^_ zBQZCfrRG&05TisVGNsKD_UapjlxSflg#|2RtBvR;*+$$YuX|P`ibMy+jA7eJ?BZr> z`*V&`^wZv?%GJ}-v^+zmQN78M#{~z~l2RMCalPJT^Zl8sgkhIS@yWW()*6LHEsmXG z>@A9jX>DuP-q%a2z2YsIJmsOPU;H-7`|&U$G%aI&RPC9K*+yhvy;tPQ40qCM!Wcz? zK~*w!>seCKBh9x@j{>R59Se5by->-#`WIry93l-y4`;4-IZDx~q7D>6vm*B6TQ)2-;;yP9H0Rt4ggR+qfU7)oj{yujEbc@iY( zOEz^+ky^9*GcDS=OP1X?5Nq;UQNciu=+qj;JTOc5O?60TjB1`HCY_tBF5h%1n)dPn z2~uol{BM^j()*oee3$oBB{{Y9ef(yGYTS?tiYd<-)jXvId#v<2llA%;qx(p=e7gRn z#GjOZw+a7Ud-9lp3}h(p{8$Ep&VGOEh1)SaZonf1tm|McC$HaNy#{MESeLjiK5`X1KSqn5K0wB=bWrQZS;(`r8#pC*W9E!$nE1M$w{Ef<0i+7=rJWplh#MVkaLO{6Z3+1kElTcL)f8@Gs zhca-N4(j-MIm$6CL5*YgBfO_bd9w_mx|4RH9o+|`%#KmYPhCo({+}_(v*M3rrC+ZJ z{<*va<`1Z^1xz6Tath$A1xzF0q6N$(&`1lIOaNpT0L&-wZ;_j5J7AIlkahmjuL;ur zx7Xf2I@0a`Ysrb$kw_Uz8BQ5V`SqIMKlGX)NKzmNX{0IeUZRqZM()wdNE*oqJd)to zv#9(d|B{u)O9{NE!25|xUQiCCDb2CqX{42ne)~17zsXk`&nRu(AaBVsmd3k@N>7jm z5A2$9Adf-2;Q0i7)8@;bRPbwBnmpM{iyYIoh@q_925if9 zLw!&e1jvVNNP__7pe_iI2itN!v2m?2-^Kk zIJN2)yu{+UVCB43oO?D?EWW-9x7@GhJ8u~+G*~f4&=09XJ(G^$p|%&qx;1VHUk#1; ziG8o()MRg*ld}-dQ5_MJolgn<4W5b1mEDELdNZ+Jye;3g^8_w)d?UWv^ZSDSoJ?%c z@|vJ~WenOaHs@?sjN{LJzRb_RJsvx)$;DS<97X*Jw!)5gu|k!4Ex8VB4)Fa>9mZ`< zUh%yaf8t-hbHGj4t-&3S8u9yj)D#4#V6k1DkGM)LCjqBN2;Y1oME|l+{F!bG1@mEQ z-kTZDuM|h~EtYpjZ$f6H%?5;PQSl>syYv)S<~5JE=o^4v9}xL%$=~>9cShk2cB^qi zw^gW#jtlBGQORBAAMp8&?D=Yj4!o=SJ|8+sm%nD_g7dAu@h^fF2>uo;g~bLD!f4$M z_}--uT(8Dcyu0W<=T$sFxjp?GU%lioH>*Bf9!8GgXZqjg&AR(?Ik{=PqERHDQ}&cA z-Xvk`mI=7hB7Humw+Y{1i@=9}eZn7bu@+o;8+_VR5Z!hR;ZN+l$yd5^5q~~C1W$DD zDYzQf7FKNy5)3!A7hRH^_>*7Fm9y$u;?x-|UfKuAlj7eclzIHigt7#c@90Fb+AqDh z?}aqUW=l^pso^7{pK7EqeHbK}9G{}r`OsI=Ej>tjcRxg)?1)oI6E+cNrja7&Vc{hiX^Zpjn0%EL~ z5O7#BKXy&g-_?>tTO=vgju|0M+!rBr8qkzYwi!&;1nnlVE=nfi)fiQaoky6_mh^YN zS~Mf6YxK$Wv!8r>ES;@t*vwY3OYB^nX?>BIVKGRmeCG+%%C&v*xG#+eo+lFf_ID)4 zxPtGJo5|9k%k>mb4`z|FWAmf|P3|bN_6}t_E$KsE4{~KTH_0Yd#s;Tmr;i~Y(BT3nVy084ii#>V zs;Oe1Tu_W(|5$Onz>DmAe}fb_%x4y74pg03^ig3vaT)P6Jx@?a(s^KSnid*=aGMY867P@-yM6?4Lf0T4t5vv9gwRLnW&oH1b5QNgd? z<7q$l?l61r?%kQ4aq4+~R(Ex!4!l+M_w7DtMtcfdE1IC|k9kbhypQhpN<^DSUB*pj zH$|`8xZr_rcH%1Cub~O|HltS?R$$%Yd*n|3hk$iq;BIDmx~@Wuf^Q3v2A1E8)0;H3l9(E=Vj z0I2Q&JPd%XmU<)xK#d3NwAhmvaM0pNVgMcr0Hi43pOvFz8OoEC0lc&{Co#}MOG^?1 zt+cczG2pGmhr|HZeI|tJ1Tz9$cevJYec_tI zb%bjN*UOFo*QOz%F#)a#TniTh$Zv4IaGr2(a6WJj&~NB3^br$S0#ES$M>9&&{0IRA zcxKxXIuL>hp@hy_^W9Y&cPD8*2)zh>2>l5C2?GcN34;jHgrS7tguk2@6kH?-|D5*| z$TjlIW*{wTq#^hPLXdvQF_3v=35g~jeFqsw_AzlhwW*@pf>8^p^>Q=V-9Rj0QV%SG@94iXOpT>|`}!yh@!*C2q@0`oUG z25DdfAb=65qy>%vP!94!88`;pa)5dv9Rk!L&k3O(xh$0b#~`->F_eQk1JJgd2I+D+IVMSejUfMlS0Dqyz5cy}+@*@dpHaTEzEtBM7%F`ex_EJ`27B^i z^l(+$)@Ht*<+j|?wF-A;R0vmXWHBd{Oyef4 zU&|*3*HsNTlC1LX?7~O(Ojqf;UT2P(N8&rDkFjqf1a{$#olLu?5zHO~JJz~Pft!Sx zGxmc{u*o%UG0r2MSw+WJc!MKiN*WumPU~y3JEzy@%#SB=LC>c$6Qq7z)h^y#%CG{q zJW$CVKjX>;L^WWZjcd!!Wy{z(mj-jQ)f4%V^EUJId&aZOo!VT$u#SAfkt4kM!$9sJ z$DY`Od6id8m; zEV)~IBe)m8T;-odd2?>-%H^B=dx<*#KJTXl!PsGiaIIdfQjxn;Trek7SaEQK(8j4) z`Fej1QP{s!)MNML?^rRYB#)lIx@h-;^Pkm~QlqHsI2|E9Xc)EF$4okuTqHi4 z+*v$XE*maD7C# z7#dn%k<{t7RLk>-(qe6{n0T@lwe|f8rP1)d!Ut)Q`a$Ygs#?@&^_nfi#0S$3tK}eV=bQokr4IVI2suSBv44Uvo3i3QE zR41a)*vgsIyxikrk#B;!xWxvkjmcc9)q?_xX_rMmDY6otPOGAbYoVu@TYm!;ls;FO zEbXPP>opZFEpk*GZuV7ubH1fAr*b9}@>K&tsH(hEu^63!ChTX;{L@3ISKZrf?L0q-#F*H|yA9nPfizhg|#!EY-Bf%vfcS({^$;U+8 z`dTwQ#&t5f_3SVXd9VfRJlujOz1xVZzn_i=mh{2>HYA~e)0*Jv@o&++#vRb99vg6d z$LqM&=w|rvzIc@KCJp;dxj~+gpE}4-$FB)a2?^wH<%g>yuQlX#0hkj2=J)`t1prtp z0AOZ*ocn{x0btz$fcXJTw174ZfVcsGP7Q$20f1%=fan2OXaNBP_`LK#bCeK19sYh(sV5X+L)k{zxE7X~Zrf zDrp2J5WE2JVS-odi0}lW64U|50T8V~MAFm;`9N?2NM4PY1tJ%u10asUF(N=|`T>HM z_DfJ8q=EmFrafqr^i3mR5#j2)sPr?Fp$^00RRe z4+t=jKmdaRh6MIyQ3_v;*Eema;$J@G2d3@d z_EkH?&!~Nv_pnIh^SsxyS;=*{nOm-~N!$YVtL=04!3rxb+1rTe7!ioyXTIduQEB|M zs*&uGwcfmq)f_%0-kiJNFrGg;?Fu*PXbE3!@NzEDumRKTLNSx$Ofk>v1+nw2j&XHs zWN=x^Ih@;>a`xC2Z4rMvQt%XJ(b@9`@Y2wru3YE6n=Kact~>-h8EYTT~5u z`fvv(jpg+({lbO?PvWxH#c*TZn)02urt-$Nj$HjK;e4J^06H+zgN+_Cn%V0;fwN4v z;VKm`W~(n~%Dj2>1;3v^hb`S+%&vLo&Kd@tW=E7TT+oL^y!e?PqhIQX`?qd}lcrzc zo*1v^JHGG9Kk{&8>$Wi=-ePyzR%_>RlNv4Nc2!N`;;irT!7T=JCmYOU-3?dr+nUzo zeBRq|=9?a|cTQd7#C7_d&%l0cqocN*{lXRU!|{8GZJw><7R^RQ#YXgwj_5m}_t=Q= z(ZfcD$VYTNCQgi#cTD>A9Tm|#CIUXKFerQwUld;z4wALkF+~?w9p$>|n0jrX5a%49rJf+Xr-aql z1?-t4K7DpmY-8S>dg-1ZjPKfn4ry0KN=)w}KI#2Y*=Js?n2R%&=c-oZ>?i^I}}6$z5hh&GZt*!S<>2kcS;;+v=;N z$1c;vkQAQMKS$F=r>&)F9=mDk{ZOf3`XlLvVH?ry!X6>j{RY*y#V5(()O2dc+nvdKy!03Yv?>PGhMaZ5N3B^)6AX-b@qT zCy!7cF;1h;)LTh)-%w3#etQ%Z_TZ+F7MLk1(jE%WA3RkojeRMk)f%kK3#dyEVO~=! zLuyd3hqjSwEbdI3c0VY&pKK$pe_caLzLhGa9GxLf>zGEZ>$*y~z#db+*zG9o&DJDwen2vPI6Z@36ZSRU-w!fvD z)U2r#7=JpaUuWU?rbSez{R`xWpvvE3+rKch$X5--HWPX0Z?WxfvF#t1L}XunJbpuL zi@l9-y%+ZA+JebA{dz9$qlm_>$D3n*fPfE}X5mv4KcZ8gvav(=nK*4k5!UP636%r} z;sIt4@kPT4ND#;4VeCgdsDCnE*sdcU`$@!u$_r3otzOu+N@u+0;zjJ+t0OA2-G>dk zvuK<8DDK|Y2Dhnt6|ZP@7w_*g8>30{P|1`Fc<$C^cxLSqTz#s5lk5BA)nV?~*-Zyq zy^6<9FGiuv(nM6RiGX%H_Cz_gV^Fv9acJZ0V!XVh5$-*@2R5Cw5glHXgY8!2qfaiK zkw7`1ds~iT>o;ZSka+~!x27-d(d95MPFjYWj%e^N{L%d-v&v7$-(uUp?%f3s5Ae6x z_W#$de~WFuVuJsQNwVSiFT^$?Mrl4@t`W+<32q>6$zqu7-33CKMnof?Wgxi80-8n? zBl%?!?z`6(5%InWa70KW0-DBSi-=$Hx3@s#gR(!?`8_{~Z$ub_I%RQ9Q$G>#K%mp) zBXw!SKG+A(Fc8v+7zg#s>EB+Q{8pEyJ*}skMxY~t+uu|BZzJY!i)|oAfgmRHt;{#U z4FoA!!~%gz7OWtTEP{a`_G6IqfQToHT5?_xzd*19;5d|(^OMc*`Jr7%hX8q@90XaQ zgL)wi`V4vHdSo$9PKP$+x?mgn0YKeQ$4?z(1M=q;+o~S)^_#z4waz%tw@$4lD(ba6 ze{Zt4Dm0i?-A}Vp)on7IFIUv%$mo3(7fTryzb`t^1ta(hsWU4m79dJB;g9{|X;a zJc3Uy8OmSx6I2lq@myAq6TEYYk~1FW%V#dmVs8{(=XNKy;DzCNtoKoCuA%2=#(LCi z?wR#`ZrjQM%=)iw@w1c7nGA=??An>Vn7t;w*}FBS@V$ad`9@c-u(#6huw0Y6TttMQ zs?o70?B=TJY>P_q%=;akIppuk<$U?TJ)K#?JvsM)uT}Ddc{_*YFON-NHoc#t>Tf)c z8**|ED_*qZ>>I4-E3XM=Ypxm29TV^W#DOy?8l*h8(3aJDT1_*VKWIRn$hyk6%w z>}ubKOj&pd)A&?lw$A8auEW_;Ovd6IF3o1T%KUIwyruRrcI_tz-1Xip)^W5iQ#2zH zU(W5u85ZiYy_7q?KfAxz$J+mGVq2Vgk>^*DsMJrdOnbGkH?gX_tgc zug=n@TegZPsZCM~i;v>i=uyZ`DiTuiCklZ=XXVX(orOjABb9{%FG_Pp-=p)oEZk!o z<1ZH8nJvC=v0gN+Jb-Ssxz3)t57pA3-ihMa>5oL$?M}+edl(@%(}FHYcu%$Cx9%}N z(OdeQanjqoldja}Ymrcxe_gEIY&%LR&68dyeki;@aGqFWf3To?B3PLm){j1SqfAV? z6Q=xn)k5fU?=)q*pj^29E?jDVXSo<1)?e~_8%Eb2)j)h#GmvV3qODkdq@Eb`?y17r zd#(6p%OHC2nC;@B=)qE)^9JftXo%^y6or#R+CU5@YpN=~h2LadwfpXmH6zc+_iD;m&!k>RtJTKKDis5T*sZ zP!3L{B!{@Av~Pb-cwRQ1;=6^brFusx!@Hu;Xui2}L46gIt=(QA%KA$26~t@fZGVTihz3vxP^c#2)t_q?-0o*RY)dJirz^wvYDd7Dcc&A4}fcJI~tSm^(5IBO0(1zeo2qXj% z+7mhwLI|A*6?(F1>xv+GA_-B1!Gs}%VT2Kc{~)kI{+}txk3)q*S{$~`4nbZz+p+o6 zAT+d9p9aT#+(_6R%Yj8{`FWiuR|$%qWFx5+;Boe zN9xdn-pxm^To^P{YKxSsbx>KH8M;xa2%mN*!rSY8Sig2TUU~I33Y^*!cL`XAeWTvs zpbT?VZ-8bhOr8sb)CpZu2JM_#&lKQ-LikN#EDk)-KnT-(tqWd>1c3=& zTO{6G8V@nz5hi{rKyK_zG?E1vYJ;T!848Y>ANQz*#@yo<7Eb41S9!1({T85mn`yuPzL=W z`%q3JvXOeA45a_8)9{LgFfb5c5Fmizhrt122rw{!!H|K0L^dLj2Mi0OK|Ux4z`l%J z4$|ei$fl+~h@lR-E!c;;<+2dVeSxxIz@R?ZhIRqihQSGr!8t&F2rxjwHUQ~zT~G#& zK|N3g(toNT8;A;oFL)OIs33Q(A`i1T-_#u&_>6sBeJ4jHbCKP8^3yIg<5v{A_>NE% zGsnILazR%{@e{^2XP546%vatM#<$pB?prw`PIYbLe5O`RCnj(l!D^6E2HhpZT$Mbnt|cGr2A zanpE*=i~XyVJWJZj<@mAj!)3N_;|E^r!7-+4~0%on1!RK7BGwTNgL%)*&4mBurD7s z#K!bPbnoynys*kVJlw1`-sEM>CcGWYMz21{Ts<|7x4Rk1Sl&IAOmczAe zZNWZYK)&%iyaY|k{K_O1O~E-;F7amWtJ!zwUUL^M5A%WM+f=;`Z>!vztmB4G%;pVN z+v1oz1x!lAOulx#N$iHx9^4^xpPSgyhR^;qj}xa{RasAQ;|o^yl^=%R&7;gqZt;KW zQ5L7}+h(G0vul92Q_L!)(9u8NOw3bAt@)n}u^F@UV z75-EmYd7`uC;g?V_Z4)PR;6N#%bv)w^){+tRuy5%!qyUU@)J(1*($EMS5)}kNKog5 zkk=C522e#yd=$~82I7=WQA*csabga$P|EPgp+8odA}xpvqn3~9Ax%BL81+BYPz*Vd zCe3T%LD%{`fJ$BeR`jlpXy0z{r4`lF#ADAAy%#bK#ZMobimj5*iG!0%sCqVWg6*2u z6q9y|Qme;FJ+t3Y(KbPJtKrsSt>+!6)^r`o&@D)~wmgIC-R+>d*1ZAZBx zy4hT6(ReTQKG#{|iNF}Kap4^G?C5EV^gFF7#~wOriz>7yerWz&9%TO)tUQQZ)#c0ge^}%qqaj1!2r5Dw zSWS@FUmFLK?I1#XLPx@XDRhxM5rjxW6k#x72w@mu1mSOyi(G_?MJ`l4_#tlA{v$dX zuY-GB^+MTp9_X~iS=`e9EcQND4|hyoh%GN2#Lg)hxMd4H^tyU8yn4byJTZ(znL1Y3 z*!mN8ZI^&PZJ3So+*Rnq`dC+Oj|MUn;rl4hR>N+3>Y%4og2;F!h}i{#a`0pA`Y zURxlrf%v2m#E57G^};k9%EL4q#4e2&8hC1H+k;QrYw9O$!?xyA_aK0Qcr`&aMCUtlTO2bTjU~tQ{P1)5P?Aa0l^7GA`pW>9D)G*5I~rM zJRlZ<$OOkB4a&kkhwdJWxNBf#dQv>_hn< z`vc{m4736HFRM-8h za#fGMW)1otbl<=OLXs+OW zy7c3=ZkoaC7TfbWb=vao_8(OFDf(RLx@UZ#Wg@ry z-G07WLKYv;{*uZ$xQN+3b^!hy?7$55jbK-24aYGifjHf08}7V40EwMnGkvEpTk&aj;1T;NO?Ho}mW_tUor?+S? z)CxJGoSccIGN(*>#_W+&&h?$5MfpHM|CJ;zR+rEwWk%9Lh3y`@oTg&Z!!bgbT1E9= zWhw4R{o>Q@R-7=y^QH3a$d%%u+r@jZSDDmxwLenWZk8tZTrEB^z9GKI@fZ5p+!VIG z((^G6D^fO3HaN3Tg6Qdj}@Ie zr_nmzf;dt7Ah{X~%7d}xlwHY4(V%r7$!uq#V%2LmDsA9E<^20Ll5yH-di32gk-w8J zO>1^cyf*01ZM0T(r z)qAroRbNa~3@cO!do%l}XN{X9=54$t3>LA78uV4H?08%eQ1cDtx!TSr@APG56#nJ= zWBI4@+=}^w#y9zDfuLn7?^G;kK{sJ=hRFjA(DHx{19dfmDWL|zf&gw2wF%(ya3@d@z-<5#7?@O(nA658vfYN@PY5Jb zC}_!(agryD(38-c(3cQSh#*7~q6mWt|4G=A%l}nl`Sw>RV8x)BJ5$ljM*1k&vNbwf zyCq6*UKbs`(HG@s^~Kf)$D*C}+MqRe6j)Jd2yUd?9&PLM0lS=Ahdy^-k1lU?$0`>S z^oXm1x-3dS)i?LUm%6V-`IDdFXR}|Tnw2bZ6`$w0aaKHbFPn~ghp6!QIf-~wXl?XB z#}c=kR}VQh00$2pyskDlN$8*__VL&%ayDL}%0@SDx}vj++IaE&DR^e&KHR=I0qsAi zpfmbmT-EG7u~RZIeT=6^VQFaZF}-2pHM0IWj*Fbx3A>j5wmz>o8P znD4`y008p=m}sd+VxYPfFe`wW7BDdYc>VxjZU9RyV0wUBTEGke@bCa&k^t}&0>C^0 z;57t*sRF=*2&k{cj>Le2mWCt-8fj@vV!%;L6A}Yp;s7Tt&Ljq0v@|0z;Hm{oAmFJ5 z%pvec2~;tpZMe^HZ{dEzeT4f4_Y3Y3+!weHaQ)$$!!?HM3fB^@pB(|N8C)y4K5$Lo zTEO|kIlB|!T;cp$5?T{{2ucD?zyub|E{S<<>`S)&2myq)gm#1ugkVA_p);W?p*!S< zK(>1jdJ*~%`VsmQ1`q}k1`(nOLkYtPBMGAkW51RD>;4ND55jK>QZX72DfoCjyy`~; zB#o#;ystp?(ReZu!Ac|g5K#w2CXGiH$*1v>BI1sw9J~?;o>c%5hBRp)RDl2mACo5{ zmFAU4t>^>-7yu%XMra~aW_XPdrpy}eC?YI@hm~Abe*I8W2WdmDOP(@o#3#}}%}4mj zxybv_M~w&tF{vN=4kDN)jr8qjnKD-_K!E`Oga4049E?7sK@0(m0_@8H3g zmiHkI3=06p1?mCdIJ774!(ahyxTgXLD2@y7PQ3 zKfPtL7uayGqgJSveYRuY-L>GBc->Oft6NhQ>a<<8=k{!7@}2#7@qxZfJe|*##AK=F zHD1T9Y;%ph9eJ2>KHC{TTQ`_nFsY-;*s_4j_PoQ1y*sLu@kjW)^@XYh?^olShbu9) zdQU~Uu{Uv-6QOwQmM#pQ-<>^bFoxUp-k85VC0ONIbp;pgOg^lCID`o(RI`W7M&P?= zH{z%aU7Q>2iYu*++jo(pWMhr&o>-LdrPll53hK}QWeV{=pBPC zZ|z~-b~D_Xue$uv%!ZuntFe6fjBb2jojsi6vkPp>#9>TCV+%I1!94s+<~F?iR59~) z>jZXMgR?lJ%2xEb{u4CY5n+e9eOc34Yq^eTyIBi|r%d4WA=q)}5$lJj^K#xCf+h~3b?lv|@B!e2}q?txnYKj-l?cFm&s zYcK@hI2l z$+<*!6%!3!C7Uwb~bLzC>@fS}qqqM8S(ejMaCeKQE#xcrsi&qL; zd%G)A+TWoZFOQL&m+TVgSto?kdLGKuq;9l$%TLnXR3x3+(hXJf8Guwy9cj0w
J zbfqJk??|(ne-*v)9ranJdq_ikV5f z3SV~zQ%S?x)5{Kic>S&wO%Hb{F(d|BrlJrk)5=MHKN?jj#(}uN=Qq$Zv(WBBV>FRIJ(i>`T zr`H~Ntb976xftncrK~e+hR>84NNV(Qt{8amgjBPdITh<{Bv!rQCq4&S z3Zk0drD{HjSJcn36z#JvQ2ULXm9Z0hDZAG0L8o4##q|q1tJj-17rb-MDwc(pQx;yW zh5SQZq?9IE!h)3A>PL+pDA&!MEI&MdE>Eh7e}q~8-llxbKzcHhcPf^iptJIn0ET!N z!eRJ_;TqP8zl-!l+TjQ)LK}iVA&?M6Xiw-!2qAPLbRl#jR46rR*CbIS&tSq3!Z5-J z!YINR0t~(532}ej)yHq2nfzE(C`nC3{(7I#i5*5LrNJC*l(7#p-BQri>dR5%OjF$2 z@hiTe7{wUct;Hc7&!FcQx})Pb5?#Fe4rThCMThujs1w%5C+tgc@u#iW!Y~adHX4b| zR7I%3HW@vy9FL9?S?j_ybDZKb8Y?Uspbdq6(8_RcY;o}%>bVqSJpMG+38{w4;&X6^ z+5&d+%EMOtYaEufnqJ%yC;1Y|b&`IdRgOfNo zd;wBNZ$RtgFOs{hgR&Ekqmq(+a$6H|cOPuPu4$3BL< zKVy)eieHneCMJ+4>W7O0CPSXZ!%SXYQNW4;08;|M%pZ_f2C&)yz{CKs8UVoD047?% z^Z+1(0APjykVJqQTFgldSZJ{%G4Q`FRoRm^00%7%NDMU8(ul-BV=ay(2AXJTN@BoC zi!+G<7cI?547h4>BQfBv1!OG1QwzvgKnpD(_W-T5fb;`+YXKPuP-p>32%xlpJoL}J zNC?9R+<&;|aG&Ac!u^E%2=@){72F@VC$$Jxgnzq)Me5`Uz63u)00Exmc7zUuU_vON zGeIL=k+?$13JfPXK8O%a7)lsU7)cmS7)uyOh$T#bygy@riwog5rLBpw{G<62705$G z;`-*Nw5bgIus~jdeUPF+8j@utke)PB7LmEYp9%6n3^JJJYgc3+WG=`r`!<1ert!hj z%21Fl%Tb#8z(7@_+&{KfYxdU%T#b?lpH!HCY+O@0{9>&&a#TpJU&uthzks_xGyI z)u&C^3wynJhh2mC{nId8X}1S^Gc*q0-{j4@`!~hc&Xlos-b=W+HHK_*s1G}LsSSJJ z)p*Xr)S3?+)sp*sFPS|O6vZ8E7QyCsyQrG%d==}()Z`=gPvQ0bt$F>S7kTdGMW$8l z4g8t&2ULdbj&dpf=IkwFiQh3KlC$(3!nlq9j4$c!Vh?wCj#nBR^Y#$nwG%lUb)Em*6PnL{E&d-*T=AzEGKir>x|-; z4Eiox{a!+uXAAi;{GSSC356H->Iu{RU3{9}&JhdFtAv)f{MACdTJfT&j$}?_>gqEe zrNdfhu~PRYf`@O4a%28MVf~io%6`u!sm0~7w85G6^pg9%B;V+4O7D(Rv^_XPs?tAQ zSfA@Fwy(O9eAjA{IJUGh)yHw2^62iDV*4(w)ZQo6$^+zkQ>E+oNLi^@(z!X~q|@#4sl$DTDyuxS61Obzpf)xS5$89y*<%*DUfjHIkov*VOX7lz z+ti4_V5J`wE^fWwf^MPq65rRqqujaHr!b9MwMU1Nuf@5jYamS=m>eB=l*Y(uH!F?c_9W%V?w-o$9hb=Ylch08;F zia|TNp=FS$dSWF_zgI5Wj&3bZ?LC`n7O6)yJbhoi*gRcKDCLEY#&^U*kH^#ii)mtG zD?_SpN{Fz*w4oHfW&riXDpIVub&`6Uej4?_sFUFMMiARQ3|6nI>r2f%wxV$Q?5|2o z@0DV$Ic0^pwpS>dK8bXiiHEqtiWah~@C9S4+@KV@T2ZIcrzlRCf$U% zGhO6|=+7mXCH!&5U(>RD)j%?rRwNY=4qU? zdkqpjkE2khJhX4)GF&--D_XR;2YU9YGg|+81!~#M1TBo}g{t(JhjJEl!fvC|(Wy#m zl%4SgPe>kvj+tFXi%M3L`dx6JI~(wb*JilU;SiKPZ9A^s_B{5vT7Y%SkK%@$K2BR_ zhX*G+W47rET&Kli?D1hG`lZYXjnyBH?ijdY>y4kjU7MfxU)gy4nuPZEFGYSceRkyW z0KfzS4qCt*0u8l*X#{?639UJ46ZpNQG#FN3g$OIgzl4-V&f}kw(MZ0(e<_k&1PH$= zsU?8a1TvP!J4=3jkw{P=If0a=k;sTF1|DDFK?Zpt)_9Y_Ylz@=MIM9nrxbVMSP%WLE}BGG+&CGy*^PikZ|QeN}RmZXoe zlqP$Bfm{Z%pC%vF4PI*6kJ*23leD4vwih|qzo+)!o{);AH83C$DiibxFdx+=fD|Pg z7Lc&uH~{ljIfiXGCQDSZdy|}rK(|ypKH9Q&5PkrOePtOrlrw3f-SN3SY@7!Kjwbx?> z`=~`K(?9DHleO>`6Kv+gx9(WV>#X9& zl*Y0}#Vjis8#C>rM{rFGQ~02VmHA2WZMeYM`l@*kD4upX!7k|7f?MEjz`KmR$@!an z;GR%N*=H-}a0BR6-agBQO)RRVG95BOWqm!8&u*N<+9y9_oiBZ0Zp_<@=R4%$?WyZ= zQm=)$^?^_5RM}NrbwD&WPMXSadq?1uu!h*4E)jQPt4jL*TPT)pb`c|XMzW=H#X zypKsdH}AF`S1s9?9oOKMs_xVpDkFX%uOC~&^*c11O?|4yZS$PjwdQtQr_NvaDwlV2 z1_RUBC5mwN)6E&&h0PsUo48Y~RsX8&>Os@lL~jGGv4b)D_RavlvT-SEH?JPE_ogk^ z$iJFO-|8v9Y5yC3+S_Y9?drmt*KaI89KV~`_V4?6ec{HE6N<{iuwZ}pqk8O!-pUD< zUxf5C>y<-mHV~FrpZAIBx0Z5Q8(O$`Qkc+Tk++z#I#ck+F;x3kuj#CfSA=c0>5@h7 z_X?X^OQ@{dIm#vPCkv-d##72G2ZhL5t?A@9FX*#}rid9~+k_`>QFMRT!PJpc_mm0V z%cNEvk0G;DO~npDDL#oO+Ddr^q13h1w~9+`y9-^nVe0MYuF97yzEG8xFO}#@!F1ze z>B5Im$0dICC~9K$+QRi59mU=unNn&-6Z+x)`P8w(6{vCpJ83~>9np6cB}Glo7Mk{2 zMBl&ifcm)ekoxI67dmM8Kx)8};o{Um{iX0RXOdWgHt%jn>_&qcixF9fFhEFqw1 znX=B6Ws?48LuycSXUUqSsVfdMl$piFK8t6Eiq2K<7oPEZrt}I3pt^Q(6Q?z^R&F1> zg|c0$P$+!gD4sJNrO+8q=vfDbD;_`67m{jCF5GsZrEv1_S!w+2+M@9y2jwpAvM{`D zUv-t`n-$4UdBWl)tLfsj7u4#KNYOoXG`&h^h}fK|BRLp9SL}9m5%t5Ti(5{8q&m&X zqwJiE1wU*i&O5$_9(Cz}Q27>waXNxRat{6T?g!2vJ_34eh@di^&0KkI1h(8l+btXS)tjRSL23OO0;{46GmOe zp`azZaBRc6cx6XD-0Iw9?4EZP&9C(syRT`37Ce24P7hv#ZcSDqJ6#ua@ZLxCi^~_h z)OsTde_RjkkaEyIyAHT%oxSKyxQL6d=HcYr=4eV}dAjKkHX+n9Sk1@it(L4F#3 zO*o57Ab%S_TnsP~0IU%JSPuXooB&`t09aedYYlk~^5Z%L))fG(9RL$8U}^yHUjjhS z2f&vJ0Bs)tza{{5e!x--X#N2BI{~$|fbSCk{}%v&9~4kmiw%hZ_&)*wUhxC!Yq29S zV6Vl2#6SZrzn5_4N!kOvwEW&e84MF}-~ZG?8JK%=4*dxOwA%y8_8>wup+dhi$on$| z*)aU3Fc$YstkQn73tl%Q-`0|S!$9x?;MWp0!kN}j3j`<-x-?=F5u(2Pc7g8~*_Zvt zWIr|dt`-PzkPfeHLOKXw8ox4<2I3b8X|kZD$qV^Gv?HRLMr70anQ6o@Z9g3_@QHx_$9*+vljPbndci0sIspeBp*6BF48!-JeOu^$_2f0uVNEMbN> zFJ#TbTe6=%Msd2;vbo6F=h;VT)v?k1R(SrwFjN#1z^KkF;_5}swh^$m7dz3&5CHmuRrI^RZa-wuz4hFl4{NcysOMLYPFpA za<9xb=-|L4wA+E#70zJ|{I>J{@q2jh4)58fZaMgn(GE7yY6kDxCyVdbSmI;#?(oyS zQ`wI#yP$mcP=@k|XH0aKqTQJ+Yni@@Sxr7h-@Nt^&hK7lw%aEKqqAxi`(Z;P-po7` zXMF0z*zTHv3Xb(>O%`5b+xVBVOF~AnA8y>o>n`TAjSe5;Xq$CJ4C}zQ6XWEE;dc|c z{%zm&N+_IuJ4u+)VWYQ~Yq~hOcd($>)*5^ zdxgg#A%$mNURE4%%n~!_-lP&g-j?1+RTXouoRJn&Zd7io%LTdyeMQrE3k7HUN|cHF zT5&=29mVR_m6UTzOoU4Liv;_G5~by^`I7Oh!F0`5o$2F!mI^ruH$>5J@19yOo=C?h zIxE*&XYZ-=fTQE;-4xlk#dO=0XraaWOzOzBlj6?a2b57;Z3^>R9hV$yhe&me{VC7M z`y|WR)kX7L2dUm44vFngT^EmcjaTleHd3g5-k8qI%Ty*k+%C?r>ZM4U@PMk+CrsSv zpjUY7{d?N$=v=W~j0L^0&_vl?Xh!S2l7x+=;Sw{k4qaJqn3QO*Cms%dC54U{K;`N` zrflp2q{`L#U*N` z@Rm}FM+)i-YmM}%$nMV-hIj~-Y_?Y@$vmVS%Iu&Gj?55!9gZmsJ^CpRWQ~xCHzq) zU)!#H)j;^NkasE;zM#7>L4b(^4C37!PopP``xZld#f}oI@H*$4CY3`wD*WIo3*|?ABa_>O2^sFZ} zt@CQSUn5(zrf4lT?lB#i?|qG(jc231Uc<1v>KvM1Z!>mVeH}Haeh&$a>!Pn7199jz zOB^@f9qXn75>dRY<^yibO=cf+x)A4J<*xw#y ze|wbaR38}?IdF9E@PSd${YOR&4Qm+PKz?RmSozZnW#k_}in5 zTm%TeDV!z9AEnoPPmBmpninA9*Axlhk)=tOJ-uL`2xcI70U)yd=s5<06r_VtCW}-W zu}o8s_N7nq5~s#%41_T7A|qlQpm}i;>LkLO#zPFsXuZO|_lfmjFWkVp1pgL-6f zP2*9f73M&Y(+Gj^5+?{}&>rcV+{WKi`)^N3#X=ew5D>4xh`|4($U+kcRRGLa<#fm+ z&tD-1U``8jS(wwxKsuz!b-}irpKN|_4~|0_s2_j;`5?%1WjU7n0A&D32O!Ax$ZbNJ zETF+TKt7mf1JEAigCLjxw)YeMD~}=-2y5_c{E@hCfx!LRn-lq%Zbt*_tYP_BO247SPE_U!Dg=lKb?$5fS+kN9&7tyMe9+pDe_bZ7UJ&tnsQ zF=E=z%VnO{J&di?dw$q!oKu)2~-p$B#|%{bB2vX4iA^gdUd}osyb( z?)W^GdXj(&KlDL2>=xm%`DJ)l@ijcZ%N11qQpu$0yu+h6A4R)@n&QegC!*w$W+>%g zAydzKJd-o-6cgp0!lXwBvv1BQxMvIQvpuKoU~J<*vPlbDvWN1vb5%1_*rtnjGX8$o z_?g{DvL1y`na!P&nS~=G81>*MoY$L;yrXG%j%_fCRgB)xU0!sSjaa{mD|TAQK3Lp| zTQOu3SIsm4AJ+3^N3GQ5r!HN>Et{-VO|f15|yn8FCJvp%=1W~tvVwoofB=qvpVKdPe( zPKYKCKML~#4;8p2J1N$?I8cs#8wwfm_mykwnA3U9JBcn9tE6-*M!+ZXseHdrh2iHG ziEj4imHw6D3NupXQRfspwjavk(-@bjL zPCNb!UH^<7UFH4`Av$uMxJi{QEa@{^TJ}T|;@i@~&eiv*PUkw)Ij2>U|IQ`Ug#oXm zROj)OHWJ|7CM=_1s=H#e; zb?znAPw%78p6n08pvOa$#pdzilVd^R%%D(ZS@v2 z?i;rYPYVJSJ9`<>gR`QAV@)P2H>WG8tXGtBtIY%IO71wR>gV@LJ@QRA^UoiIHT$~= zGnR--N6(3L`}_IgjVm35vM(2vOF!r-u55cE)y{rFIZfT5_;`AwIOj@N8ZF)_R`O^= zUH!CFylz&NvN`5R&?_EoNA-h`A%jSk-u&bKIXZB>^+Ij+4I&Fo=btEr7 z9?ZsWrYdxMizzAyJ&OFSU*PeGk_O^qyGn57Sw+}y(pB8`f+3#WIUDl_@8eFifO7X;!q2vD z#a?wfptIrd^y(nXv5nAKbrGIYeIU9w$pg~<%0a&Czb18!Ng&Vr4<`?Q@&K$lxCnb4KcozR2Oi_nMAkI&2t!*MzcP@uNFRTeiEzb| z6a2x0AqT)98);)O_CJP7WCxBx41jTm0EQpRfpM4fKv_5r$lFjB3<~6dARhzc0%;IK zUg!q|s1Ja2xnHmk?SXNE@MGIhKkP%0%Rw4!L-?^Dkp7nr@*jK!Qqmv0(5Og28sj^2 zMqABTbj(wS+JZjzJ8crkrBuqyEclA92bk5>^#KhSRGgGEvd?N zTlSRmc{_}sadH8FDDesoYu<}{c%(XeYCs|vY}A7b-+7#CHDwxKyZ#iu=&BVPGUYYL zOo`$0jui2WJ47?ORbo)NK`~>oA(U-hcAiUIu$xOFTEXO*Nqwt~NL~M1^o?p=V z6uWrWK4!s~O>9=V5wpu>p0f`<570Un|JGJVfDm)mrlOv!LkA9@5*lLlJV6zBJ~r6os0$*IdGD%MS3kv+kZRxR2k)-hPB?i-XRj(A^6n|9qU z^||o>+B+AxnzFu+OU0-ZDHVxGZlMgN&e?mdeK|+DBoYz2sU#_ePwvL0hFmi)<(4}S#x3bx|1)PBGw+z0=Xu_b=N)E!KHt{f>%P|B{r6hG&)WM$ zP#q&5s5aOAfvR6(D10^2NopuFR*gD(nwm)sQ_ep#MkuYmC*13=q%vAA7RrsnXalG5 zLjU5g=&aE#CBpYx>76%E2+y9ZrPjSDp&W|((;g9Rlt`~k9ldwCFmAh2oy1L1_uM~- zT2wGy$l7z4%Dr4mx2Bp3rB5~qDQ8bA+s#`;RdeP-A?B3ZswYdGH*sq7uC>&{X)BfO z7VZ)zZ5${$8t_bcbx^tBWVnQCVp&BMWFqD1?@TD;%w586yMxl(P4}v)P5lL<4Jt|T zpY#o8QU;^*CRRAQ z*F97@vlVXAbrf=qZHAVQxs84-G{vSmk8z25m!^?lz#sAI|CVI*SNU~CFgXCo7XX+a z;6Hs`5KI=(S__yj0HhEAOc?;O2+&STd*TCN-hhr;z|;W_TEOf9Afo_a0s)Xz05FFD z$SVMtMgXK10L&!dtOZOa&{GSTPvGB`j^asPSjoZa3|4xu%7WFK5y6(7*<1ijD2GD2dC-e>a1O0%uzn6}*=`jgo2|Phga3}O8z`g59@FEN(3?d9B_!9n< zbVSmPBSaGvgjm97|Er3k>4(7s_q!5OJjgB@iG|2LAjg1=1CowLTGH$;^0C%PJVXN0 z>|7$dk~HaIrx3_Ou=9!RL;`6?vm*+o%tTrezk8*TcZhTZa*igyR!Y*ugPcRA%$m9& zjdoX(raq9VK=J{ZibzG8G+NmRUQ+}KOCv9lFijk^1JXhna7@BAJF7^!;A(#DQUfc_gHVW3g<= z1Lpw5frI>T4rPE3^@(|*EJFgM5sx7r9E3xd*e*DRGN4>IhBBZ$@S)9O969{_ooE0) zA}NX2{!GEO%92P4LMo(H2kk;}$DD&bt6)cBI+P&7Ah4Vyf766bDk zo-ZA@RpvEq0=Ir%DQ7cRDGL%(WednIp}KAv{7zhh{pSVYV}mQ18M7q#mem$!K;!)^ z-ZY$j9QvG{{&o&t;CTz1EU01~<P zAIR>18p~GP64;($db5__hOU@RXzO z>@4ia-CR71U7D@rmuL3n_pjZ=*_?@Cv({f>Jxm+0Ze>@Qjh7g9S5#-VizSQgp1$T5 zyJhp8laBM-_YB|{Z{N)-mOteq5@yO;^`F6~z6xW5$_Dd=$KtuL1C6+%7@Du^{F1wF zznkmP_c*ukySw7W|5Uq<{_lQWF|o+?+&tB?%z=CB8cBuDO{P-v#gkO+&$M>yM42g9 z4^0xDhWXO%9O?@TnuQ3KFA9}gJGYcHy(3i?uR(OvtmlI9L{w}RI!MS~7B0Q>I+cFt zlCHLhHAK<@-zw+%ey?u*Vi)E7xU;IslyyaiPQ^;DOJ1syqZHB=X-Cwj8#~eqTg?^* zKmD5SVBVfu#CjEXx7bA)WW7;$xObcO=C7!850xO-!7f6Co3%RW@J*`qtpa-PIS=Ka z+FR;vTOLT6z&GmB{=)>HsswdWo5s}IJReE)xSc}%$mJ5(zWY>OFJhIZISZx91sQ_X z$R$$UF{_mA94=BxgVTkEb8o8KTUn_%<5Ifb@i^&~m>*R~rr4_;uoZO>uMn=E+OMA3 zd?&55Z>Z|Q5LWHHznLoYNfrII<1p1f9`}%*f0Clc5pSuZ@BT@vSc(o_6et{Ax>q=! z|CZX3FwtPyORY*Bm1R?zK-HCELWq)C&WY*qK@=PYp5oOIyP6T-ewPF0gwsdnFf zM%A$F2IXCDT6FSpKWXWVOv!|rN>wf1p{!iPOWN1HS@d}1MnUhvGs!CF*~+|=>C)c% z^_4vupHd}6-jd3{_ExPZ&y(s64-@R1S4yUZ^p}P{DW}RK`n|tB|574T;-@y@@9PuC z3uLCI;>pKlCTJ);hQ&uYJeFZv0FUv81QSAILK6b4_bdojgpbNhB(4Ji)?Bd8g0&T_ zr(g{Q>n0ZhtdBGj6Y+l`FKO2vy-2tRp%0-Sp+8{&!JFVi7$VN6$?;Iaza=%1eDPvA zpSv&K1|OB15@~LY9QoU>rj7bVp)3O*^wxVd-64Jt?U+9RmGo6m3rFhE@gjz~1)*vHbp4w087i$h zfL9fmT9k$5t5@U5 zZHd?_{Gn3+xE`Grx?O zNYcUehHDJh6|N;*Ke%pio#6Vw^?>UD;|^mDV+>W`AI7Wfz0&N3wBzmikw3ljT}Vk(8y3ANrAj2?vNsrYmk;;rxO5@ z5~SD2QBWuOd_1HTC)Z*e)GzLi(nwuI!h$x6yRI}}*MfGDwn92hc_4FrVUz2R%TZtu zz=06}M8jTJd;Z~r0*nO!4+3~tfP?f92Ir6#9vW~CK9m7zMMDGUVtJ4c&LIwfV*ngH zNFW~oh6?hCbwE90I&%1nI7km^Aq^ahY2g_924w(nF8bu~^LL^__=qee-UFY0AjF1! zOqfcPr)Bx@2GTTnGC!Ux+1rO}I5msU9{o=Kjo$$-bx1Z>@8Bpl(_tl>T9(1l_U8Pw zbrQMz)VH!eQI_o2zEUo%yfxFFYRbNv+LrZrQ_ke84zfw^-{HLFK3vndMy$;tGj?uu z7C(NYuIz{8MzWm=RWjpzUp~fAhcTG8f^D?>9Osk2fteqC39XJk!kBe=#|DpxVrQR< zWb0>|Ff-3(F+ryG_)+OW)^6NJW{>O?o_1^pTGo3y7G76yjb<@ScdJn5#y4?z*>aYh z!IbefQG2*tMV4%3`h5P`t!kOuR12A-VmH6EaT-6PmmS|)Kb^}lxQ&PGiNzc3YZlYWn_{(CMRg;79 zUEdvy*}Kz>(RLTSNw|y(eX8(*L=U!cxe;EJe+Mrl-?Uyh;VjPRB1gqtMziMrUhMf{ z<+$)t9Ny(SgxS0z4qr>q#RFDNLnpfFGtW+2vqsCB^2U{^Y{K;$T<^y_IGrC8_)yyj z*|cM$_?wo|?Awd^eC{J>zSQJ3ze8Qb=iBe(XWZgAry$G?9h}U2^~_~c+VtZaRoZb& zy`q_T&&%RX@TmlnGW4c3o(BBdDkZ&6$w+h34Fs1Vwoe>)9bd;W+vrCm&Z-!+228Gmm&wX`^qO)r2!j;JWWGb~P ze7<@@Qkd%4@T1a-Zv&`(wVhPe=Spd%MXp2=k*`X+&{|>>RU{Pd>nUB{dAAUg(vY^I zU8PGF8LMtJeI*FTt*N{l_o#v7)jGSZXm!&1MM9U9W`g&DXVkEi7u1|od)3#~2~_4h zbJgB$ro#0_Wh#Hq7eb$7snTy;rRwJ8>4ICc8uhG+70Lx$i)q9BL&A)e_f>WGrU~kJ ziFDSReXbL;a+IY__DBvbTPWF7U{F+Aubpb%?RC;tltNXu;H;2eryDiP&03f{q5(D8 zem@;~*@K#QXp%JBSuQ*&TqwPC-$G&>QzGbQ&yjxLZ?W?6i(>I+`D^`p;eX$zgFs~= z4&MJ$feKavf*_K-< zF8M8;TvP|^IWNFE0h3S<7lh1Dd*e`(PVyX%l*wuwoNtc;g95P0-AQQsp#W6y;0pfgdI#Dr*@fK4o1lx^XX2C= z!*N{H9kjIlx8zFHK}$9)#5#90$MoB>JX90tfiGpGV(Ydo=tD{0K^ly51L$KGr_kdy z2hg*nW3b}&5gfs}ptGaO(AiK2Y;mJIHr(QcQl^r2UB$>{Xaef7uL_5d6MCrFmrz_B;{s1_UJ$jnD(a6bL&+L;_I<#3KmT>_sABAW&)cDv@*`25H1D zcvX+=W6}VxxRE#zg*1Dqw6Eo9Ue6<6nS#l+MuZ~56HKn*bvP&=ehW|(r(kjo!j#BQ zb^r8*C`}tlJ49g(_F`$i0wwPAB5_bZlm~mLG=iG8J&^88n_T~~KqU%GAl`_G0uP8e zqR0f|5rl(q1YsaFLD;VaCJ=-G5R4!W%7lE7R?GuFI4BF^zOs4`=Xmj6As)cfMlG z7KgI>4HTT4$u{o3Y6usS)tWtX!Bb`!RwHkG^#?xPZ30^nlgFDM9L!g2-irqw%Vw^G zU&TL0&Bk^Y>hQfhv$^?;>amL0Dcr^E1Z=Cj317^6!Cbqz3FqCZViyFqXDih`IG6rm zT&CGtuIb%}>M%2 z+1jLhwpFKBT!M`U>#NLX_fODerukX16LsBjvzDFN`L^enlc%2J&3c`D)us`Mp+V8Wa^36YC$Uh>G+NoDiZ24Es5Q&ihmi#>GYY2agSmiFw}~9Y?WYh7cJV z5*QozL1lHFx{2`y5FZ<;2n&h*Ik(gMel&;*8WR#68$KcAed~YOag(6%ae<0Nx3N)S z;W4q{!D6|_ijXk>u~9KGzv_FVnBc&%A^y=(G5+x(@V(LxYn}YQ-uk4sG2v0;{Mv{G N>W0@1*9q5~{6Chflf(c3 literal 0 HcmV?d00001 From 17b5da82a9fc6542ec670fade8d046500927c781 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 11 Mar 2025 12:16:44 -0700 Subject: [PATCH 03/90] 111 draft --- .../model_validation/111-draft.ipynb | 20 +++++++++++++++++++ 1 file changed, 20 insertions(+) create mode 100644 notebooks/tutorials/model_validation/111-draft.ipynb diff --git a/notebooks/tutorials/model_validation/111-draft.ipynb b/notebooks/tutorials/model_validation/111-draft.ipynb new file mode 100644 index 000000000..f4372fd53 --- /dev/null +++ b/notebooks/tutorials/model_validation/111-draft.ipynb @@ -0,0 +1,20 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# ValidMind for model validation — 111 Import the champion model\n", + "\n", + "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, you'll set up the ValidMind Library and import a champion model for review." + ] + } + ], + "metadata": { + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} From bc4e985a3618fb46128b5f7f17b1a66d3acaff03 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 11 Mar 2025 12:20:31 -0700 Subject: [PATCH 04/90] 111 notes --- .../tutorials/model_validation/111-draft.ipynb | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/notebooks/tutorials/model_validation/111-draft.ipynb b/notebooks/tutorials/model_validation/111-draft.ipynb index f4372fd53..178d0bb4f 100644 --- a/notebooks/tutorials/model_validation/111-draft.ipynb +++ b/notebooks/tutorials/model_validation/111-draft.ipynb @@ -8,6 +8,24 @@ "\n", "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, you'll set up the ValidMind Library and import a champion model for review." ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- About ValidMind block\n", + "- Prereq block\n", + " - Access to champion model export\n", + " - Registered model in platform, assigned as validator\n", + "- Install & initialize block\n", + " - move LLM context + add explanatory section\n", + " - Adjust initialize with validator instructions > need to remove yourself as a developer!\n", + "- Load the champion model\n", + "- Load sample model development dataset\n", + "- Upgrade ValidMind\n", + "- In summary\n", + "- Next steps\n" + ] } ], "metadata": { From 29a6a075dffb1d9ed5f55b8cbc561009a7dbdc68 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 11 Mar 2025 12:25:21 -0700 Subject: [PATCH 05/90] 111 adjusting --- notebooks/tutorials/model_validation/111-draft.ipynb | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-draft.ipynb b/notebooks/tutorials/model_validation/111-draft.ipynb index 178d0bb4f..ae2e3b960 100644 --- a/notebooks/tutorials/model_validation/111-draft.ipynb +++ b/notebooks/tutorials/model_validation/111-draft.ipynb @@ -14,14 +14,14 @@ "metadata": {}, "source": [ "- About ValidMind block\n", - "- Prereq block\n", - " - Access to champion model export\n", - " - Registered model in platform, assigned as validator\n", "- Install & initialize block\n", " - move LLM context + add explanatory section\n", " - Adjust initialize with validator instructions > need to remove yourself as a developer!\n", - "- Load the champion model\n", - "- Load sample model development dataset\n", + "- Getting to know ValidMind\n", + " - preview_template()\n", + "- Importing the champion model\n", + " - Load the champion model\n", + " - Load sample model development dataset\n", "- Upgrade ValidMind\n", "- In summary\n", "- Next steps\n" From 2ab9c2683776932d3724aae98174e953144e79ea Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 11 Mar 2025 13:10:42 -0700 Subject: [PATCH 06/90] 111 +intro --- .../tutorials/model_validation/111-draft.ipynb | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/notebooks/tutorials/model_validation/111-draft.ipynb b/notebooks/tutorials/model_validation/111-draft.ipynb index ae2e3b960..9c0361f57 100644 --- a/notebooks/tutorials/model_validation/111-draft.ipynb +++ b/notebooks/tutorials/model_validation/111-draft.ipynb @@ -6,7 +6,23 @@ "source": [ "# ValidMind for model validation — 111 Import the champion model\n", "\n", - "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, you'll set up the ValidMind Library and import a champion model for review." + "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, you'll set up the ValidMind Library and import a champion model for review.\n", + "\n", + "These notebooks use an application scorecard model as an example, but the same principles shown here apply to other model types." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Introduction\n", + "\n", + "An *application scorecard model* is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", + "\n", + " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", + " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", + "\n", + " Model validation aims to independently assess *champion models* created by model developers by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance." ] }, { From c7fcb9e30bebe69b742f869ff59ad5f92ff96e73 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 11 Mar 2025 14:04:07 -0700 Subject: [PATCH 07/90] 111 + intro --- notebooks/tutorials/model_validation/111-draft.ipynb | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-draft.ipynb b/notebooks/tutorials/model_validation/111-draft.ipynb index 9c0361f57..15d449298 100644 --- a/notebooks/tutorials/model_validation/111-draft.ipynb +++ b/notebooks/tutorials/model_validation/111-draft.ipynb @@ -17,18 +17,21 @@ "source": [ "## Introduction\n", "\n", + "Model validation aims to independently assess the compliance of *champion models* created by model developers with regulatory guidance by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance. Assessments, presented in the form of a validation report, typically include *model findings* and recommendations to address those issues.\n", + "\n", "An *application scorecard model* is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", "\n", " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", - "\n", - " Model validation aims to independently assess *champion models* created by model developers by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance." + "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ + "# WIP BLOCK\n", + "\n", "- About ValidMind block\n", "- Install & initialize block\n", " - move LLM context + add explanatory section\n", From d942bdf66f9b61968d93da3f57dc0548e73e96c3 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 11 Mar 2025 16:42:03 -0700 Subject: [PATCH 08/90] 102 accidentally deleted a header somehow --- .../102-start_development_process.ipynb | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/notebooks/tutorials/model_development/102-start_development_process.ipynb b/notebooks/tutorials/model_development/102-start_development_process.ipynb index 4cdfa0247..b64c31426 100644 --- a/notebooks/tutorials/model_development/102-start_development_process.ipynb +++ b/notebooks/tutorials/model_development/102-start_development_process.ipynb @@ -70,6 +70,15 @@ "Refer to the first notebook in this series: 101 Set up ValidMind\n" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", + "## Setting up" + ] + }, { "cell_type": "markdown", "metadata": {}, From 511520266a3801ae52c497b75f20ecc45c03e35b Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 09:49:49 -0700 Subject: [PATCH 09/90] 111 new draft --- .../111-set_up_validmind.ipynb | 256 ++++++++++++++++++ 1 file changed, 256 insertions(+) create mode 100644 notebooks/tutorials/model_validation/111-set_up_validmind.ipynb diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb new file mode 100644 index 000000000..f3bdf68c9 --- /dev/null +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -0,0 +1,256 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "34e60911", + "metadata": {}, + "source": [ + "# ValidMind for model validation — 111 Set up the ValidMind Library\n", + "\n", + "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, set up the ValidMind Library in preparation for validating a champion model.\n", + "\n", + "These notebooks use an application scorecard model as an example, but the same principles shown here apply to other model types." + ] + }, + { + "cell_type": "markdown", + "id": "0426ef4f", + "metadata": {}, + "source": [ + "## Introduction\n", + "\n", + "Model validation aims to independently assess the compliance of *champion models* created by model developers with regulatory guidance by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance. Assessments, presented in the form of a validation report, typically include *model findings* and recommendations to address those issues.\n", + "\n", + "An *application scorecard model* is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", + "\n", + " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", + " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants." + ] + }, + { + "cell_type": "markdown", + "id": "about-intro-e542b60e-207d-49de-a380-4a6f99f1c392", + "metadata": {}, + "source": [ + "## 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": "about-begin-5260219d-b4ce-4a0b-924d-7333fe5d7be3", + "metadata": {}, + "source": [ + "### 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": "about-signup-9ffa73ae-a0c8-4c3d-b307-08dbb5c54194", + "metadata": {}, + "source": [ + "### 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, find code samples, or read our developer reference.\n", + "\n", + "
For access to all features available in this notebook, create a free ValidMind account.\n", + "

\n", + "Signing up is FREE — Register with ValidMind
" + ] + }, + { + "cell_type": "markdown", + "id": "about-concepts-fa466c67-c594-4e29-aaba-72e793f905b6", + "metadata": {}, + "source": [ + "### 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/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": "install-library-2450d0dd-fd90-407c-8cd7-53ef0973ffcf", + "metadata": {}, + "source": [ + "## 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, + "id": "install-python-7a733859-9daf-4528-8128-c5d23a0fd6ef", + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "id": "install-initialize-0221421a-9308-4d3d-93d3-0514d5fbe196", + "metadata": {}, + "source": [ + "## Initialize the ValidMind Library\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." + ] + }, + { + "cell_type": "markdown", + "id": "install-snippet-7b7726eb-26bd-43bd-b1ef-504895bf9bd0", + "metadata": {}, + "source": [ + "### Get your code snippet\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 **Continue**. ([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:" + ] + }, + { + "cell_type": "markdown", + "id": "install-variables-b606b626-0791-458d-811a-1b3228ce85a2", + "metadata": {}, + "source": [ + " - Documentation template: `Credit Risk Scorecard`\n", + " - Use case: `Credit Risk — CECL`" + ] + }, + { + "cell_type": "markdown", + "id": "install-credentials-90676cb9-9fd8-4484-8804-bff9cc2796f6", + "metadata": {}, + "source": [ + " You can fill in other options according to your preference.\n", + " \n", + "4. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "\n", + "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": "install-init-0c822102-b258-4404-b21e-26b8540f611a", + "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": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", + "metadata": {}, + "source": [ + "## 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": "upgrade-show-752d8d8d-55f7-4465-aed0-34c5a3be8663", + "metadata": { + "vscode": { + "languageId": "plaintext" + } + }, + "outputs": [], + "source": [ + "%pip show validmind" + ] + }, + { + "cell_type": "markdown", + "id": "upgrade-version-d64591ca-3073-4b3e-9586-d3577adda203", + "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": "upgrade-restart-3825c878-7301-47b2-92b3-2bab240f343b", + "metadata": {}, + "source": [ + "You may need to restart your kernel after running the upgrade package for changes to be applied." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "name": "python", + "version": "3.10" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From fbab0ee2b37caf9167397529e89847d95d587824 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 09:56:13 -0700 Subject: [PATCH 10/90] 111 editing intro to be validator-focused --- notebooks/templates/about-validmind.ipynb | 2 +- .../tutorials/model_validation/111-set_up_validmind.ipynb | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/notebooks/templates/about-validmind.ipynb b/notebooks/templates/about-validmind.ipynb index 1c135f268..e982e60d0 100644 --- a/notebooks/templates/about-validmind.ipynb +++ b/notebooks/templates/about-validmind.ipynb @@ -31,7 +31,7 @@ "source": [ "### 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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index f3bdf68c9..afd60e0bc 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -34,9 +34,9 @@ "source": [ "## About ValidMind\n", "\n", - "ValidMind is a suite of tools for managing model risk, including risk associated with AI and statistical models. \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." + "You use the ValidMind Library to automate comparison and other validation tests, and then use the ValidMind Platform to submit compliance assessments of champion models via comprehensive validation reports. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model developers." ] }, { @@ -58,7 +58,7 @@ "source": [ "### 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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", From a9e1ddf34da8971f0c9a983d7b4c328c1721dc22 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 09:58:09 -0700 Subject: [PATCH 11/90] Edited the About ValidMind section to be more role agnostic --- .../capital_markets/quickstart_option_pricing_models.ipynb | 2 +- .../quickstart_option_pricing_models_quantlib.ipynb | 2 +- .../code_samples/credit_risk/application_scorecard_demo.ipynb | 2 +- .../credit_risk/application_scorecard_executive.ipynb | 2 +- .../credit_risk/application_scorecard_full_suite.ipynb | 2 +- .../credit_risk/application_scorecard_with_bias.ipynb | 2 +- .../credit_risk/application_scorecard_with_ml.ipynb | 2 +- .../code_samples/custom_tests/implement_custom_tests.ipynb | 2 +- notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb | 2 +- notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb | 2 +- .../application_scorecard_ongoing_monitoring.ipynb | 2 +- .../quickstart_customer_churn_ongoing_monitoring.ipynb | 2 +- .../time_series/quickstart_time_series_full_suite.ipynb | 2 +- .../time_series/quickstart_time_series_high_code.ipynb | 2 +- .../operational_deposit/operational_deposit_poc.ipynb | 2 +- .../customizing_tests_with_output_templates.ipynb | 2 +- notebooks/how_to/configure_dataset_features.ipynb | 2 +- .../how_to/document_multiple_results_for_the_same_test.ipynb | 2 +- notebooks/how_to/explore_test_suites.ipynb | 2 +- notebooks/how_to/load_datasets_predictions.ipynb | 2 +- notebooks/how_to/log_metrics_over_time.ipynb | 2 +- notebooks/how_to/run_documentation_sections.ipynb | 2 +- notebooks/how_to/run_documentation_tests_with_config.ipynb | 2 +- notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb | 2 +- notebooks/how_to/run_tests/2_run_comparison_tests.ipynb | 2 +- notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb | 2 +- notebooks/how_to/run_unit_metrics.ipynb | 2 +- notebooks/how_to/use_dataset_model_objects.ipynb | 2 +- notebooks/tutorials/intro_for_model_developers.ipynb | 2 +- .../tutorials/model_development/101-set_up_validmind.ipynb | 2 +- 30 files changed, 30 insertions(+), 30 deletions(-) diff --git a/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb b/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb index b019c7b45..6da5cffde 100644 --- a/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb +++ b/notebooks/code_samples/capital_markets/quickstart_option_pricing_models.ipynb @@ -82,7 +82,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb b/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb index e25191847..a8548826b 100644 --- a/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb +++ b/notebooks/code_samples/capital_markets/quickstart_option_pricing_models_quantlib.ipynb @@ -120,7 +120,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/credit_risk/application_scorecard_demo.ipynb b/notebooks/code_samples/credit_risk/application_scorecard_demo.ipynb index 5bb5985f4..1bc8aa931 100644 --- a/notebooks/code_samples/credit_risk/application_scorecard_demo.ipynb +++ b/notebooks/code_samples/credit_risk/application_scorecard_demo.ipynb @@ -86,7 +86,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb b/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb index 349cfd30c..8025ece67 100644 --- a/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb +++ b/notebooks/code_samples/credit_risk/application_scorecard_executive.ipynb @@ -37,7 +37,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb b/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb index 2c91302c1..77e5b2ba0 100644 --- a/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb +++ b/notebooks/code_samples/credit_risk/application_scorecard_full_suite.ipynb @@ -37,7 +37,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb b/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb index 0d6f4e270..0b909ddda 100644 --- a/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb +++ b/notebooks/code_samples/credit_risk/application_scorecard_with_bias.ipynb @@ -75,7 +75,7 @@ "\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, find code samples, or read our developer reference.\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" ] }, { diff --git a/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb b/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb index ca1bdb4e3..961a92c5c 100644 --- a/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb +++ b/notebooks/code_samples/credit_risk/application_scorecard_with_ml.ipynb @@ -37,7 +37,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb b/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb index 34ed27e57..80393d5f3 100644 --- a/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb +++ b/notebooks/code_samples/custom_tests/implement_custom_tests.ipynb @@ -78,7 +78,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb b/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb index 0f88228e1..055028eca 100644 --- a/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb +++ b/notebooks/code_samples/nlp_and_llm/prompt_validation_demo.ipynb @@ -66,7 +66,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb b/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb index f6942033e..dfc9ccc15 100644 --- a/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb +++ b/notebooks/code_samples/nlp_and_llm/rag_documentation_demo.ipynb @@ -33,7 +33,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb b/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb index e4e48884d..04ca7aa1c 100644 --- a/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb +++ b/notebooks/code_samples/ongoing_monitoring/application_scorecard_ongoing_monitoring.ipynb @@ -33,7 +33,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb b/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb index 156f3fb14..5396ee1aa 100644 --- a/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb +++ b/notebooks/code_samples/ongoing_monitoring/quickstart_customer_churn_ongoing_monitoring.ipynb @@ -74,7 +74,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb b/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb index edd3ca9b5..c674ffd42 100644 --- a/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb +++ b/notebooks/code_samples/time_series/quickstart_time_series_full_suite.ipynb @@ -77,7 +77,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb b/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb index 8873b8524..8418dcd0a 100644 --- a/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb +++ b/notebooks/code_samples/time_series/quickstart_time_series_high_code.ipynb @@ -77,7 +77,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_sharing/operational_deposit/operational_deposit_poc.ipynb b/notebooks/code_sharing/operational_deposit/operational_deposit_poc.ipynb index 9fc708325..b98aa9d67 100644 --- a/notebooks/code_sharing/operational_deposit/operational_deposit_poc.ipynb +++ b/notebooks/code_sharing/operational_deposit/operational_deposit_poc.ipynb @@ -37,7 +37,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/code_sharing/output_templates/customizing_tests_with_output_templates.ipynb b/notebooks/code_sharing/output_templates/customizing_tests_with_output_templates.ipynb index 980529dd4..c787a8831 100644 --- a/notebooks/code_sharing/output_templates/customizing_tests_with_output_templates.ipynb +++ b/notebooks/code_sharing/output_templates/customizing_tests_with_output_templates.ipynb @@ -85,7 +85,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/configure_dataset_features.ipynb b/notebooks/how_to/configure_dataset_features.ipynb index 9bf927740..2e381bd89 100644 --- a/notebooks/how_to/configure_dataset_features.ipynb +++ b/notebooks/how_to/configure_dataset_features.ipynb @@ -67,7 +67,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb b/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb index 8dc4ab10d..2529727af 100644 --- a/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb +++ b/notebooks/how_to/document_multiple_results_for_the_same_test.ipynb @@ -82,7 +82,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/explore_test_suites.ipynb b/notebooks/how_to/explore_test_suites.ipynb index 7cb5e2e49..4fb36d894 100644 --- a/notebooks/how_to/explore_test_suites.ipynb +++ b/notebooks/how_to/explore_test_suites.ipynb @@ -63,7 +63,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/load_datasets_predictions.ipynb b/notebooks/how_to/load_datasets_predictions.ipynb index 81c4e0e98..d0f629307 100644 --- a/notebooks/how_to/load_datasets_predictions.ipynb +++ b/notebooks/how_to/load_datasets_predictions.ipynb @@ -79,7 +79,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/log_metrics_over_time.ipynb b/notebooks/how_to/log_metrics_over_time.ipynb index d551d58ff..bbf4b94cb 100644 --- a/notebooks/how_to/log_metrics_over_time.ipynb +++ b/notebooks/how_to/log_metrics_over_time.ipynb @@ -78,7 +78,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/run_documentation_sections.ipynb b/notebooks/how_to/run_documentation_sections.ipynb index b7b43e379..9066d637f 100644 --- a/notebooks/how_to/run_documentation_sections.ipynb +++ b/notebooks/how_to/run_documentation_sections.ipynb @@ -73,7 +73,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/run_documentation_tests_with_config.ipynb b/notebooks/how_to/run_documentation_tests_with_config.ipynb index 0eea64a46..8ad291fe8 100644 --- a/notebooks/how_to/run_documentation_tests_with_config.ipynb +++ b/notebooks/how_to/run_documentation_tests_with_config.ipynb @@ -77,7 +77,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb b/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb index dfbc4a0de..acfb923f8 100644 --- a/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb +++ b/notebooks/how_to/run_tests/1_run_dataset_based_tests.ipynb @@ -72,7 +72,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb b/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb index 9370fc98b..ab8880c4c 100644 --- a/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb +++ b/notebooks/how_to/run_tests/2_run_comparison_tests.ipynb @@ -79,7 +79,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb b/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb index 9fde46220..184594bf2 100644 --- a/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb +++ b/notebooks/how_to/run_tests_that_require_multiple_datasets.ipynb @@ -75,7 +75,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/run_unit_metrics.ipynb b/notebooks/how_to/run_unit_metrics.ipynb index c6469e7ac..163306ae3 100644 --- a/notebooks/how_to/run_unit_metrics.ipynb +++ b/notebooks/how_to/run_unit_metrics.ipynb @@ -104,7 +104,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/how_to/use_dataset_model_objects.ipynb b/notebooks/how_to/use_dataset_model_objects.ipynb index 44ef1b151..abf449cb0 100644 --- a/notebooks/how_to/use_dataset_model_objects.ipynb +++ b/notebooks/how_to/use_dataset_model_objects.ipynb @@ -79,7 +79,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/tutorials/intro_for_model_developers.ipynb b/notebooks/tutorials/intro_for_model_developers.ipynb index 95e1b19f2..a1412da7d 100644 --- a/notebooks/tutorials/intro_for_model_developers.ipynb +++ b/notebooks/tutorials/intro_for_model_developers.ipynb @@ -116,7 +116,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", diff --git a/notebooks/tutorials/model_development/101-set_up_validmind.ipynb b/notebooks/tutorials/model_development/101-set_up_validmind.ipynb index 9a5936350..1a316cbba 100644 --- a/notebooks/tutorials/model_development/101-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_development/101-set_up_validmind.ipynb @@ -101,7 +101,7 @@ "\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, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", From a1568c0ef7880896e69792561247f6b1be934bae Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 09:59:07 -0700 Subject: [PATCH 12/90] Missed one --- .../code_samples/quickstart_customer_churn_full_suite.ipynb | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/notebooks/code_samples/quickstart_customer_churn_full_suite.ipynb b/notebooks/code_samples/quickstart_customer_churn_full_suite.ipynb index 21c1aa17e..d7a9c6b87 100644 --- a/notebooks/code_samples/quickstart_customer_churn_full_suite.ipynb +++ b/notebooks/code_samples/quickstart_customer_churn_full_suite.ipynb @@ -75,7 +75,7 @@ "\n", "### New to ValidMind?\n", "\n", - "If you haven't already seen our [Get started with the ValidMind Library](https://docs.validmind.ai/developer/get-started-validmind-library.html), we recommend you explore the available resources for developers at some point. There, you can learn more about documenting models, find code samples, or read our developer reference.\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, create a free ValidMind account.\n", "

\n", @@ -545,9 +545,9 @@ }, "gpuClass": "standard", "kernelspec": { - "display_name": "Python 3", + "display_name": "ValidMind Library", "language": "python", - "name": "python3" + "name": "validmind" }, "language_info": { "codemirror_mode": { From 26ed5505afae6ad908289c47b564d3e583863341 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 10:30:39 -0700 Subject: [PATCH 13/90] 111 +Register a sample model --- .../111-set_up_validmind.ipynb | 74 +++++++++++++------ 1 file changed, 50 insertions(+), 24 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index afd60e0bc..da96f6e53 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -98,6 +98,53 @@ "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": "8da00e8a", + "metadata": {}, + "source": [ + "## Register a sample model\n", + "\n", + "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", + "\n", + "For this series of notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\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 **Continue**. ([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:\n", + "\n", + " - Documentation template: `Credit Risk Scorecard`\n", + " - Use case: `Credit Risk — CECL`\n", + "\n", + " You can fill in other options according to your preference." + ] + }, + { + "cell_type": "markdown", + "id": "ddbb615d", + "metadata": {}, + "source": [ + "### Assign validator credentials\n", + "\n", + "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", + "\n", + "1. Remove yourself as a developer: \n", + "\n", + " - Click on the **DEVELOPERS** tile.\n", + " - Click the **x** next to your name to remove yourself from that model's role.\n", + " - Click **Save** to apply your changes to that role.\n", + "\n", + "2. Add yourself as a validator: \n", + "\n", + " - Click on the **VALIDATORS** tile.\n", + " - Select your name from the drop-down menu.\n", + " - Click **Save** to apply your changes to that role." + ] + }, { "cell_type": "markdown", "id": "install-library-2450d0dd-fd90-407c-8cd7-53ef0973ffcf", @@ -129,7 +176,7 @@ "source": [ "## Initialize the ValidMind Library\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." + "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." ] }, { @@ -141,30 +188,9 @@ "\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 **Continue**. ([Need more help?](https://docs.validmind.ai/guide/model-inventory/register-models-in-inventory.html))\n", + "2. In the left sidebar, navigate to **Inventory** and select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", "\n", - " For example, to register a model for use with this notebook, select:" - ] - }, - { - "cell_type": "markdown", - "id": "install-variables-b606b626-0791-458d-811a-1b3228ce85a2", - "metadata": {}, - "source": [ - " - Documentation template: `Credit Risk Scorecard`\n", - " - Use case: `Credit Risk — CECL`" - ] - }, - { - "cell_type": "markdown", - "id": "install-credentials-90676cb9-9fd8-4484-8804-bff9cc2796f6", - "metadata": {}, - "source": [ - " You can fill in other options according to your preference.\n", - " \n", - "4. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", "\n", "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:" ] From 4416aa0465751952bbabf07812d6be5a5c5562ed Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 10:47:10 -0700 Subject: [PATCH 14/90] 111 +Getting to know ValidMind --- .../model_validation/111-draft.ipynb | 57 -------- .../111-set_up_validmind.ipynb | 132 ++++++++++++++++++ 2 files changed, 132 insertions(+), 57 deletions(-) delete mode 100644 notebooks/tutorials/model_validation/111-draft.ipynb diff --git a/notebooks/tutorials/model_validation/111-draft.ipynb b/notebooks/tutorials/model_validation/111-draft.ipynb deleted file mode 100644 index 15d449298..000000000 --- a/notebooks/tutorials/model_validation/111-draft.ipynb +++ /dev/null @@ -1,57 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# ValidMind for model validation — 111 Import the champion model\n", - "\n", - "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, you'll set up the ValidMind Library and import a champion model for review.\n", - "\n", - "These notebooks use an application scorecard model as an example, but the same principles shown here apply to other model types." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Introduction\n", - "\n", - "Model validation aims to independently assess the compliance of *champion models* created by model developers with regulatory guidance by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance. Assessments, presented in the form of a validation report, typically include *model findings* and recommendations to address those issues.\n", - "\n", - "An *application scorecard model* is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", - "\n", - " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", - " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# WIP BLOCK\n", - "\n", - "- About ValidMind block\n", - "- Install & initialize block\n", - " - move LLM context + add explanatory section\n", - " - Adjust initialize with validator instructions > need to remove yourself as a developer!\n", - "- Getting to know ValidMind\n", - " - preview_template()\n", - "- Importing the champion model\n", - " - Load the champion model\n", - " - Load sample model development dataset\n", - "- Upgrade ValidMind\n", - "- In summary\n", - "- Next steps\n" - ] - } - ], - "metadata": { - "language_info": { - "name": "python" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index da96f6e53..c89d4a9cc 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -219,6 +219,122 @@ ")" ] }, + { + "cell_type": "markdown", + "id": "c67460a3", + "metadata": {}, + "source": [ + "## Getting to know ValidMind" + ] + }, + { + "cell_type": "markdown", + "id": "7f4f9a30", + "metadata": {}, + "source": [ + "### 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", + "\n", + "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e63a3025", + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "id": "b7347ccf", + "metadata": {}, + "source": [ + "### Enable custom context for test descriptions\n", + "\n", + "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", + "\n", + "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", + "\n", + "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", + "

\n", + "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", + "
" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "87f7fb2e", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "\n", + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", + " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, { "cell_type": "markdown", "id": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", @@ -264,6 +380,22 @@ "source": [ "You may need to restart your kernel after running the upgrade package for changes to be applied." ] + }, + { + "cell_type": "markdown", + "id": "3469ee6b", + "metadata": {}, + "source": [ + "## In summary" + ] + }, + { + "cell_type": "markdown", + "id": "4ebb0a11", + "metadata": {}, + "source": [ + "## Next steps" + ] } ], "metadata": { From 1dc92cf03517e014b7a35eda8f1fb3bf56e6b0d3 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 11:00:29 -0700 Subject: [PATCH 15/90] 111 Set up for Import the champion model --- .../111-set_up_validmind.ipynb | 77 +++++++++++++++++-- 1 file changed, 70 insertions(+), 7 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index c89d4a9cc..bfd9aedcb 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -98,12 +98,20 @@ "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": "f25ee2ed", + "metadata": {}, + "source": [ + "## Setting up" + ] + }, { "cell_type": "markdown", "id": "8da00e8a", "metadata": {}, "source": [ - "## Register a sample model\n", + "### Register a sample model\n", "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", "\n", @@ -128,7 +136,7 @@ "id": "ddbb615d", "metadata": {}, "source": [ - "### Assign validator credentials\n", + "#### Assign validator credentials\n", "\n", "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", "\n", @@ -150,7 +158,7 @@ "id": "install-library-2450d0dd-fd90-407c-8cd7-53ef0973ffcf", "metadata": {}, "source": [ - "## Install the ValidMind Library\n", + "### Install the ValidMind Library\n", "\n", "
Recommended Python versions\n", "

\n", @@ -174,7 +182,7 @@ "id": "install-initialize-0221421a-9308-4d3d-93d3-0514d5fbe196", "metadata": {}, "source": [ - "## Initialize the ValidMind Library\n", + "### Initialize the ValidMind Library\n", "\n", "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." ] @@ -184,7 +192,7 @@ "id": "install-snippet-7b7726eb-26bd-43bd-b1ef-504895bf9bd0", "metadata": {}, "source": [ - "### Get your code snippet\n", + "#### Get your code snippet\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", "\n", @@ -335,6 +343,50 @@ "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" ] }, + { + "cell_type": "markdown", + "id": "9fc515f0", + "metadata": {}, + "source": [ + "## Import the champion model\n", + "\n", + "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5b9e47f2", + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4e1640fd", + "metadata": {}, + "outputs": [], + "source": [ + "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "markdown", + "id": "e0a31bb7", + "metadata": {}, + "source": [ + "### Load the sample dataset" + ] + }, { "cell_type": "markdown", "id": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", @@ -386,7 +438,14 @@ "id": "3469ee6b", "metadata": {}, "source": [ - "## In summary" + "## In summary\n", + "\n", + "In this first notebook, you learned how to:\n", + "\n", + "- [ ] Register a model within the ValidMind Platform\n", + "- [ ] Install and initialize the ValidMind Library\n", + "- [ ] Preview the documentation template for your model\n", + "- [ ] Explore the available tests offered by the ValidMind Library" ] }, { @@ -394,7 +453,11 @@ "id": "4ebb0a11", "metadata": {}, "source": [ - "## Next steps" + "## Next steps\n", + "\n", + "### Develop potential challenger models\n", + "\n", + "Now that the ValidMind Library is connected to your model in the ValidMind Library with the correct template applied, we can go ahead and start the model validation process: **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**" ] } ], From a8b6ff99e97e1ae1760ef928a9b6856b6287dd86 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 12:15:48 -0700 Subject: [PATCH 16/90] 111 Shuffling things around --- .../111-set_up_validmind.ipynb | 146 ++++++++++++------ 1 file changed, 99 insertions(+), 47 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index bfd9aedcb..6fa6237ce 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -227,6 +227,102 @@ ")" ] }, + { + "cell_type": "markdown", + "id": "9fc515f0", + "metadata": {}, + "source": [ + "## Importing the champion model\n", + "\n", + "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5b9e47f2", + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "4e1640fd", + "metadata": {}, + "outputs": [], + "source": [ + "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "markdown", + "id": "e0a31bb7", + "metadata": {}, + "source": [ + "### Load the sample dataset\n", + "\n", + "Then, let's import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle and quickly preprocess it for data quality testing purposes.\n", + "\n", + "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "99ef5ead", + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.datasets.credit_risk import lending_club\n", + "\n", + "df = lending_club.load_data(source=\"offline\")\n", + "df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5a67644f", + "metadata": {}, + "outputs": [], + "source": [ + "preprocess_df = lending_club.preprocess(df)" + ] + }, + { + "cell_type": "markdown", + "id": "6e5606d9", + "metadata": {}, + "source": [ + "### Obtain the final Feature engineered dataset that Model Developer uses to train and test the model*\n", + "\n", + "In the feature engineering phase, we apply specific transformations to optimize the dataset for predictive modeling in our application scorecard. \n", + "\n", + "Using the `ending_club.feature_engineering()` function, the Model Developer conducted the following operations:\n", + "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", + "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "d58a1eb1", + "metadata": {}, + "outputs": [], + "source": [ + "fe_df = lending_club.feature_engineering(preprocess_df)\n", + "fe_df.head()" + ] + }, { "cell_type": "markdown", "id": "c67460a3", @@ -343,50 +439,6 @@ "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" ] }, - { - "cell_type": "markdown", - "id": "9fc515f0", - "metadata": {}, - "source": [ - "## Import the champion model\n", - "\n", - "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "5b9e47f2", - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "4e1640fd", - "metadata": {}, - "outputs": [], - "source": [ - "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "markdown", - "id": "e0a31bb7", - "metadata": {}, - "source": [ - "### Load the sample dataset" - ] - }, { "cell_type": "markdown", "id": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", @@ -442,10 +494,10 @@ "\n", "In this first notebook, you learned how to:\n", "\n", - "- [ ] Register a model within the ValidMind Platform\n", + "- [ ] Register a model within the ValidMind Platform and assign yourself as the validator\n", "- [ ] Install and initialize the ValidMind Library\n", - "- [ ] Preview the documentation template for your model\n", - "- [ ] Explore the available tests offered by the ValidMind Library" + "- [ ] Preview the validation report template for your model\n", + "- [ ] Enable custom context for test descriptions generated by ValidMind" ] }, { From a6446a88a913a9549e4760587818adc20aa0a607 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 13:48:55 -0700 Subject: [PATCH 17/90] 111 Importing the champion model done? --- .../111-set_up_validmind.ipynb | 20 +++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index 6fa6237ce..eb8c4a19b 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -270,7 +270,7 @@ "source": [ "### Load the sample dataset\n", "\n", - "Then, let's import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle and quickly preprocess it for data quality testing purposes.\n", + "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle that was used to develop our dummy champion model, and use it to replicate some steps that would have been conducted during the initial development and documentation of the model.\n", "\n", "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" ] @@ -288,6 +288,16 @@ "df.head()" ] }, + { + "cell_type": "markdown", + "id": "891bfdf6", + "metadata": {}, + "source": [ + "#### Preprocess the dataset\n", + "\n", + "We'll first quickly preprocess the dataset for data quality testing purposes:" + ] + }, { "cell_type": "code", "execution_count": null, @@ -303,13 +313,11 @@ "id": "6e5606d9", "metadata": {}, "source": [ - "### Obtain the final Feature engineered dataset that Model Developer uses to train and test the model*\n", + "### Apply feature engineering to the dataset\n", "\n", - "In the feature engineering phase, we apply specific transformations to optimize the dataset for predictive modeling in our application scorecard. \n", + "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", "\n", - "Using the `ending_club.feature_engineering()` function, the Model Developer conducted the following operations:\n", - "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", - "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." + "We'll apply some transformations to optimize the dataset for predictive modeling in our application scorecard:" ] }, { From 2794c909e26178a8be2fc6dbd4a3f689b89bc60f Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 13:57:18 -0700 Subject: [PATCH 18/90] 111 Setup for custom context --- .../111-set_up_validmind.ipynb | 84 ++++++++++++++++++- 1 file changed, 81 insertions(+), 3 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index eb8c4a19b..c67e9b456 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -378,16 +378,71 @@ "
" ] }, + { + "cell_type": "markdown", + "id": "57ad6e51", + "metadata": {}, + "source": [ + "#### Review default LLM-generated descriptions\n", + "\n", + "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", + "\n", + "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with later iterations:" + ] + }, { "cell_type": "code", "execution_count": null, - "id": "87f7fb2e", + "id": "5bb4450d", "metadata": {}, "outputs": [], "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": fe_df,\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "5612ba38", + "metadata": {}, + "source": [ + "#### Enable use case context\n", "\n", + "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", + "\n", + "This is a global setting that will affect all tests for your linked model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b6105468", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" + ] + }, + { + "cell_type": "markdown", + "id": "0d1e90ba", + "metadata": {}, + "source": [ + "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "87f7fb2e", + "metadata": {}, + "outputs": [], + "source": [ "context = \"\"\"\n", "FORMAT FOR THE LLM DESCRIPTIONS: \n", " **** is designed to Date: Wed, 12 Mar 2025 14:09:16 -0700 Subject: [PATCH 19/90] Moving LLM context stuff, changed my mind --- .../111-set_up_validmind.ipynb | 164 ---------------- .../112-develop_challenger_models.ipynb | 183 ++++++++++++++++++ 2 files changed, 183 insertions(+), 164 deletions(-) create mode 100644 notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index c67e9b456..181c7348b 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -361,170 +361,6 @@ "vm.preview_template()" ] }, - { - "cell_type": "markdown", - "id": "b7347ccf", - "metadata": {}, - "source": [ - "### Enable custom context for test descriptions\n", - "\n", - "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", - "\n", - "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", - "\n", - "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", - "

\n", - "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", - "
" - ] - }, - { - "cell_type": "markdown", - "id": "57ad6e51", - "metadata": {}, - "source": [ - "#### Review default LLM-generated descriptions\n", - "\n", - "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", - "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with later iterations:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "5bb4450d", - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": fe_df,\n", - " },\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "5612ba38", - "metadata": {}, - "source": [ - "#### Enable use case context\n", - "\n", - "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", - "\n", - "This is a global setting that will affect all tests for your linked model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b6105468", - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" - ] - }, - { - "cell_type": "markdown", - "id": "0d1e90ba", - "metadata": {}, - "source": [ - "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "87f7fb2e", - "metadata": {}, - "outputs": [], - "source": [ - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "id": "d6e38bd2", - "metadata": {}, - "source": [ - "With the use case context set, generate an updated test description for the `DatasetDescription` test for comparision with default output:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "adede511", - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": fe_df,\n", - " },\n", - ")" - ] - }, { "cell_type": "markdown", "id": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb new file mode 100644 index 000000000..d116b52d3 --- /dev/null +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -0,0 +1,183 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# ValidMind for model validation — 112 Develop potential challenger models" + ] + }, + { + "cell_type": "markdown", + "id": "b7347ccf", + "metadata": {}, + "source": [ + "### Enable custom context for test descriptions\n", + "\n", + "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", + "\n", + "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", + "\n", + "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", + "

\n", + "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "57ad6e51", + "metadata": {}, + "source": [ + "#### Review default LLM-generated descriptions\n", + "\n", + "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", + "\n", + "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with later iterations:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5bb4450d", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": fe_df,\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "5612ba38", + "metadata": {}, + "source": [ + "#### Enable use case context\n", + "\n", + "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", + "\n", + "This is a global setting that will affect all tests for your linked model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b6105468", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" + ] + }, + { + "cell_type": "markdown", + "id": "0d1e90ba", + "metadata": {}, + "source": [ + "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "87f7fb2e", + "metadata": {}, + "outputs": [], + "source": [ + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", + " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "id": "d6e38bd2", + "metadata": {}, + "source": [ + "With the use case context set, generate an updated test description for the `DatasetDescription` test for comparision with default output:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "adede511", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": fe_df,\n", + " },\n", + ")" + ] + } + ], + + "metadata": { + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} From 0b1fe9fcb05ddcbba7b28d809a952498e671ccc6 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 14:24:45 -0700 Subject: [PATCH 20/90] 111 Added section on finding the validation report in the platform --- .../111-set_up_validmind.ipynb | 84 +++++++++++------- .../model_validation/compliance-summary.png | Bin 0 -> 61626 bytes 2 files changed, 53 insertions(+), 31 deletions(-) create mode 100644 notebooks/tutorials/model_validation/compliance-summary.png diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index 181c7348b..37118c750 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -227,6 +227,58 @@ ")" ] }, + { + "cell_type": "markdown", + "id": "c67460a3", + "metadata": {}, + "source": [ + "## Getting to know ValidMind" + ] + }, + { + "cell_type": "markdown", + "id": "7f4f9a30", + "metadata": {}, + "source": [ + "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", + "\n", + "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e63a3025", + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "id": "2dd3ddaa", + "metadata": {}, + "source": [ + "### View validation report in the ValidMind Platform\n", + "\n", + "Next, let's head to the ValidMind Platform to see the template in action:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "\n", + "3. Click on the **Validation Report** for your model and note:\n", + "\n", + " - [ ] The risk assessment compliance summary at the top of the report\n", + " - [ ] How the structure of the validation report reflects the previewed template\n", + "\n", + " \"Screenshot\n", + "

" + ] + }, { "cell_type": "markdown", "id": "9fc515f0", @@ -331,36 +383,6 @@ "fe_df.head()" ] }, - { - "cell_type": "markdown", - "id": "c67460a3", - "metadata": {}, - "source": [ - "## Getting to know ValidMind" - ] - }, - { - "cell_type": "markdown", - "id": "7f4f9a30", - "metadata": {}, - "source": [ - "### 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", - "\n", - "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e63a3025", - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, { "cell_type": "markdown", "id": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", @@ -419,7 +441,7 @@ "- [ ] Register a model within the ValidMind Platform and assign yourself as the validator\n", "- [ ] Install and initialize the ValidMind Library\n", "- [ ] Preview the validation report template for your model\n", - "- [ ] Enable custom context for test descriptions generated by ValidMind" + "- [ ] Locate the validation report for your model within the ValidMind Platform\n" ] }, { diff --git a/notebooks/tutorials/model_validation/compliance-summary.png b/notebooks/tutorials/model_validation/compliance-summary.png new file mode 100644 index 0000000000000000000000000000000000000000..a8abd2dd39ea4ef683e4300fb90d80f309b64e48 GIT binary patch literal 61626 zcmeFZbyQT}`ae!eDJdc#At{X@;n32JbRz;%!caq}D2Oyjr{vJxsVFrH($Yx7Fo3{- z^zRIK-}k-lXMOMc|KGZ^)?sGOIs5FrpLjj5=h=G*(@>MYg-?Z#hK6=aQ9(u%4Gjm1 zhK5dsdjmLAij8-UhIUigR$5v^QCgZ_!v*}p_N5gXnnGBD9*(|N-HFRl1jK-fd>q4l@u^ulUh8{sLBn;FE^2?%qgLNM?+j?u6O6lw&5KDb`JRSkL9g+ z=x&pG?LzZH9v7#BDFFkIJyz#L(7sO}YuXYP<4Kk?sox>n8l*~k|CIrPahn$n`+FdP z8GVvVPR=KEPE5o+tR)TYY5l~#_lB2Smld+g58?>WBy;H1$ojFHi0QDof|OmM)ApURu7pYsGno7$y_<-A2QF3J2X=8j)DGRoR23 z60a2J`^CoQ>R4yum*$-M4GN#a=)RNm^}Hea%y1hUtN3|OAA_QA>AN0ep4Y|0;;P1s zN36y3v_}*xSR7vbZ7i$Dn0O%AFBK|E|P;cJpy1`{|{aM5n*C za17p#^=(Z4>UGS-&yvs49=^USQ~TwX6Wt5}QK+7TeulH)_Cw+x51&t3PpE&A{JilY z!^t|XQA6-EtSXyk69>7*is(68(A5|rKk8}w0MhPGX+(GGG+7Y6H$yS7g>T^4G2dTW z8DewE^@=W2ql4y4*elJDr)LiYrRD8@*ftUtKX5&_@YilHxV4%l@`K;af17BM;*;P@ zD04loNMJ^n+H;CZmQvcv$|Zfw@mDlDiyz64^1xqieZ4cXrY<;2+5Dni!oZ4UMNL30 zL(NRv&nw5)tN&YX24@C0nYHC=mGEx({pY`!FnUNhty!)0(B99}jlmJBZu1`4=Jmd7 ztCxgAP>mpTE*)IDE_hNzOM$lIdQUR9uWU6J{(u zz9kDE9h;Oc!%FX+JgUjidvaXH?bQ&~b8h^a+DazehxKQ(p+4HgKn~9%=zwCcYw^CoqWs}^* zJH%4Qh~|xokxEv2!0?{5o{W%WptEgM%2S?+G)2ZfM`R{k@iXV*%|+718;i7yc+aVS zJ**0qVc3Y@Qk=_EAkBI&&(2WBh+v>$HDQLth{ufevRGQX5I;`gyiXCQ+$Yl4)2r7v z*qg{)3ATBZO8zZ6Q3aOEm9I0R zHbnbf+=lnjJpoUuCy|e-$f+PydIF}yY;MY+Q1Mt(NuGSVY_kFnTg^vhkD5|_zfGo+ zr-llEZ4Yf8*2>w++O61e4NVn6q~J}0iqlFW8b;7INuQEKHbl?O2=jpjKCFo22!{y% z2tt~is5(M*R^>t&>jt$FwTANg@^Hsj+mB}PXLFkoDiO9{mLIw%!;()^gi^9XC~qy@ zB)erKTI&8|%zOtK0zcjeJ9vjYT%Y}Znm0I1G&(qL_Q9+oU>;4VlK-t>vR&oSLWIgyqwEK28)^{z0yPQhQ!_A{(tDaWIj@o1lvE_1G6jWt7C;qZqvgpoW zxk*MxR@G-0zr+$y>1J?Gm#`!Pk~-#KXG8V8n0NO1l-uAxMSG>4Xthwo_Y*6fmO^~^mcY3n>{D5k-cU}~wmKOa8f%W?V4g#*Ef zl5Urd?2?u*uxPaSV9~Y&i@n1bFJFGw%W+wVRj=?~p@W)-ylpf-jgD}!^QgVGh;@+b zn|^ZqY$vdh&&EA<>*kW%=Qi5C+Uj@yoW4TcSB9t20)N677lAE2!V zhopyIM6W_SdhSKDL`P>jW_o3wie2ux?Dii{Z6JLkkavfY9V}>l(!BQD8ax|LD}6Qn z1W#*EQ&0B%IQ;02Q%>>D>rcg&ez;T%w+Qo`XPjAGgr3d2iZ#_VvSL&QE(9(Hj!Pms ze|PS6;z+%aqLdPln*CV*k^hzYYvL!Lp6I@wS)}}h)oN$z0rgog{XSy8X(FuYW!G#`$fF)b-2CqMpHhgS6^7Yxx&D zz1!auzfQ(052$_<;~Yw~R8~_L%oKHSNgFjaWOB$_d~EgYy`w?$%a(hw>M=sg8=Z9L zciu!UvVQ6E%{qH`Mx;enK$~13pvqTf)jWIfWh9Fni~NiaGwXixkm!ipYHJFSwet&y zl*q4YX|tj|z5}lw)3^q&>3+){fMp)pf6)5b|EXc0Y)y!p*O1ra<5JeEfMT}c2XW-H zhU;}sBMiQ=o=@S6L*Wx#7rLCB?fQ$_nudE;VbA)$WXYc!AG+=gALN(8W)dBY)r`3f z{9WJdyem#HT z&;8W#e9*fK89vc5Pe0=X$(`?>Qk6?n67&Q7%hyMFH?di%q<^^VbJ?*`mZS0y_ic(p zp0Vj$^+_);o29FB+t>>>-RgWGNVTpEnB8|OS=F`j_a=JoI9rZT2z0hOw5iaN=SRtp z)1LNGKcS;g1X#RNT*K%4-olA=g!C7P!t6ow_Q}_i!;rj?`mPV5Y1D00L^RH#)wzAO zMT^kwA*&u}&z*a4F;l13wbutS6MYf!(P^v#^I`(qqsWn=LZiXg8IJ=5*SCsX(_MQ8 zeX_vK3tm5WWe>lcC_5%?8X6lBo^PJOKj=1$H^a7*sLMT#ZC_H4xNjBjjKxo2lxMUJ zwD@nxeU7Y8`y~XPS85vbH}X6^&PW%j@P}O@_Gg=paF5>Y*w69!J^P)#=P^BhRNGQV z=XZCva^>k(Zc8+Fw3%qE#PM*|0vp11gFoJSY%n8B2yz2rdTenz~VQ$A|kj?$~vYpY0V9SW#fDuYe0W%9DE26_N9w+GpY45Qv05tFv}9^qUW&1vk)I=g`r(@X&6E3TgI~421e&Kh3Z` zCH}3h4++`ga6}vZ6mYW6{oBKpAh2y3t@72$!ELl08??_}Xl@7VA3;uua1jr70+P&I zg7SW|+3~->@T)$(+g`nX8-$;SOGaKVy zcjG{_G5_@$=m&U7X-O+80`FQDE>>2Kt~Ow|65TKl-~^tNf`Ka<+8t)pfv%{@vIHn#*nx8nA6a6&x?P1F+vd~~pKGpF}-cSFllg%E=O0!>xca7IWktR7B03cqbLHPX{;j3{zgs?hEXe!!rhmKipG|dKtz4wR4nUu7;{VpypT>W` z_@|*L4{GXv6UDWhug(I37RMLm`Accy_;hr1F91E>vz1ZT2HpWHyLv?ee;)t`>K!=b z-IZ@i(V?M9qAAKqX?vn?&EoiuBF}q%sUL>Crs=FC!SmU<{}y9WrtA=$QKtK)Ubm{Q zytuBbS>A!>3&^hI?RZ%44a#@8{ifzSXpz1qKE#eIVL@{UfP>g$tGb z^i}pBI!BEU7wUEEpPmQw_F5(|5XAJ=80(*h2@G=LzmMi-7$PqrcXKS};qL;s&m z{v*Ep&%6H{SpE-!{s%$-4;T9Xq2pY(C|FtLYimO*`tgxEY$u0`X2m7#s#5Vi@4P7%$pBg3 zqnbHE`X?rm0TU~~kBX_!=>97c1jL=rQJPX6anru%w3ZgqIZ_WA^gZ!b*{vYM!pcTEBH@=JCDj@r(u7pXU zP+*Ciiy%SKlt?w z^xf-j=sz*Wk0iph+DRMVr=*_Ilswt|VMDH6BR96j`Y!N9gq=ny>3uQzhz>Q{?=8Kt zr#veYIQejm)QQXh_XD|q&^-(G!FE3=k6X*^}yXd1v;#aY}|S+!|m^ySA5RI3pLF zlhN8d#Y;F@aEiOw|6~!H1%aju#S|@@75IL-mC>?0&85Mc4>6Tl z&4S)SZNgMk-Y16?k&e`+$dv3jCZsPix*+~WXo%H|(y>TaCkrk;gimUz1GW?(@<&o;sj`+pd(~P`2H_5VKJF|w% zagj&xnAA6wmp)}FzZzm-Hi6*op0F^0>dmK!rPJTtkQV z#~tUJ*I98VMbJSAN|=vq|^I#dSo4OR}3kg>q_UDVFJJO&rw}!2vxG zaq?*XtmTpIxO>1#fx#uOS00mFqT5?FW~S#bM6)g$A${fKopk5{(srM01Kc*kQD&lH zhJALc2(J(h=Xlc-i8`}T64-HH;Ad?`^}$sWCuw?akm-QJ4r9uOA9IaWM_}RTv_G6C z!;kG3ds2jRjPi*~q|>&RHKrYn3OsPiXh^-?RG{pr_^4qSJk2l(OH?>af{jiN9bXU@ zN?gN2Ic~c11lVtC+XxKL*k94qEAL~v4vC|Xpd05mj%#Nd=EhG;+6Xp5IMGDc2+)J! zH!T1A`pQPrk#IXL*>zS<{5DV=)@%)z8iv>2x8vB2@>hRqV8HgpPN1%WGM}&*Zc*e4 zfYc%ZGI4G*&_!JqXZS>s$WmmdOPK=p^UHq8K;og>t*Jk*_9jp~zPU?PV&A){72Ll1 z)wV*(?OGoT9T+Dno6Fm6tuCdHcd@P2d~|X!uP4L+T=@FpyiN1WVq+=&F+wMd@M=CE zh;4EQinr(8^A7s%ZXc~2)X#fAEF0hwOMpTjC~A0F(G%E;eWTuXfB@jKrb0nZq5LL^ zRU}4dBw_Q7+p_`A$HD>i?=9&FkJb`SuL6L^AeLIkIk%!6_)^AfvMM*OWe?_E@Tu!M zes`uwD((1_&hn0d6xDu_7}SBQ>N0Np;RwP8+1BA702n|l_wUS3Wwt3^G%tME8QW#N zo`e8TOzIE)8x7PP(x3<}Yxz@unPcPIZ z63?^z7_7XS<_ZdFu3eFe#KwJ9*jDTqXWX8`aqfX^i@W}`?X>{Wg0!#~jjR=(^b8aV z`s&pndkrTuYR0D2{zwrwo%4IKl#gxJI1)9&lif5S%U_8p^Cds!ucz=yo}P_cHpJr| z4@cTCMMRWN4T~uG9o5}4d}Ma$ zw};MA;D?=iMF%7!@t9uJ#&UNw4|<=clf&{wg?j!r4tC;%VM(e38YjCh4*`xN?)R;6 z7US+ze-LTGGHCfn$gs$4yXt1m{&xJVe2p6IXRbeERFSNJ(#{@>)}0?c$l5U&H&eaNIq=>ch^Mo{!tq zTkIko>3`2eSe+dPD@8o|Q-ta)!XSA%6bJGhExI*7#yR=HeX4l*sD%9OA4M1A$;ivV z79w1W$1ZkO!4i#9?_lpP^ggOzyV&chK=EAfElLW-(1OERsp5IRHtO0r@+mDmOCcNe z{WlgyWrR>ayh00$=S*sT$C~V;;m?mE8kQGcJ#Kd^B)Ya0bLQ`BO-W@(YrU<5r4IiZ zS)d8M!>GRf(kBF+{rK8t$?h3A@Qfnhjc1*1ta?!61#fkjxsAEKWyjRL+xzl?6 z`o>yqrO~89%?asELu!pvQnl0lWg3&FC+?~N@rYKj^ASmERM(RF1jHJl$dRs$D|CY9 z1BABjAZ*F;3XETup`k;5hzra-yix`ZhWGSN*lc;s*Ti$*)Gvtik;3qhr-g=;8mBI+ zwGudCJ7-657pJ)`UT?!jj%)G!=8=>{bWk#0>w1?3_@?^S{aUwYWc1mGgW+>mch)IC zb2_c!E9yxQpS5ipYna=c)+#Eo)|_Y#oJwMaO*l0_>v5m_)y<(<$p{;9RlgcO8b;ld zz`$l*QPmkm)H8z&feuVynY`QFSF`(pn%%k9XBrt4TsuAZ4n!;C?{KQ2`8IMPS9S0N ze&L}Zxki$nwu02tSox06jlyB?-tGO67os(7*01j<9@s1`J(xzqT4vAQfsO9lElBY(l(eNX8bD#EYPaQD>n{+K$F|;*F*~hTVA| z`uXyYcNNp#)~cszmON&5X(8U%O0YxexXf>}ap`Ij=A6JJy449iu2P3=U3_AViT=V! zk?UY4nLS}Ey*%75m5jr#W8(&ch9 z`;5>(USwB;l9m5q$HsJ_sNXr#v)UXExXOGb8p^^V%7ND}Wv-wV=d@0}5o|KthIWbfEn^aw( zZ0-su6y!0rdYhVz(RmZ5#nvazX>_rdcYAworHI<$YAU#+-hfpYUf$kAyE5`m8bIOt zX4dU((kl}zyF;H{KBi9RaaC{c{2&>Aqp&af#^n=DGAf_+qH8}bTi=J~GB7%PGdH6f zM~48mxE?d5oiip5_qGSC%T}b%oqe@zNmsiqT*j$zl)_lJK4>H*&>Wd20(P9D^?#`TCR>LZVR z;66V=n6imf+4QYTZunH4ETl^lbWcxtFqaS(I5myF(9UTz@zZVlqz-am1?zfpMWGh7 z0qJR}La#I3&LSX%)spJI=7LVuXz0j3+L^06p&b}BX)pTgqzk>VcHQq)&p8k`GOf}k@S0a-RGA+Cnt9&}Q;GC#gx8xQ(T1t+Me3@n(Mq7JTN3ETR|b$hfCrd*Q{j7@*H`pV z!VYJ?Aodya<&*{;2Ch^2JqYVRu$jzeU z1tUv}nk(_41n4t}{Cczdh2m}`4R9W@7ZEMuS__l);=*ZBo8hMg(awSzTs08yiywdPu! zH0U&H^P-y#v~)3rW$nm^eITu}Tg)#pHd{YEfma_`LK+(i%i4hbp^(SIV0VYzRq?Bl z(i1-dhMGOZ_wdRs#ZYe9qb^Q#MYQN204kF*x6-_}1_s{7bN!UjB_(+7kdZH3WlY;G zv6z2sO5+nlHa8nr!gHh*yE8jtBGhNPe<*>&5>7BUm0ENKCX(56xPXGw3R7|3p%hH~ z1n8Ccxhvb%lf6-{$&>klmg&mQFoWm_lNnODP7SOduyMpM#*te17^_5lF0OFm5unw8 zGv@9iH*Jv}o6JDvDsFM9L@%q+x`Z&KV68Hw0mxGe4NsuN8pLDR$8MVSCp0g!ZsqO@R~^nq2i~c zrLQyUDOskc1}7gah|4D2kJ1hGp6(wIGK&u;SpD(WL`_kaTvC{Zb7ghJ45*1pxl&$F zlo6OHIMWW;_jiRsK~jkBUN5l#%Zv}Dt;YCcPfWFjIG{+SO($*vDh;^6h!-}t9NksM zqw6A68n3D@{@lLcdWkZ~n-zH2PNoYj@0IFsQEQEXUzx9rW-|epYMXbnSLKzh5Mq^6 zdl$OdOpSqp384@T2ll?OFgq^D??kFwP_YcL3~2pfl%|pIbbq%zU1BR&n!~rzYpl`O z0UisV)f&LO79k2CaATUTFjTS)Q@qxueamwx!7jHEaHWr=-`DQmE)M7;zjH^?%b9b-h zs7n$|CMAUyv`Yd+sI0YDKCc3W(9h>PWt$$y1uR2vudFHhx>Pge7byBFy_1o7y}xGE zCdM~K0`)HY;A(rp>s`=}_s|+JYbkt}t0+<%Ga+2jT&osa|LF z% z>&rBY#L;xyWQu8Cr+Tc@)7-fEJ__?h@6GtwsS{wxPXG|0F$9-~1WI7YZU&EXOq3;v^a52?Dl|vQ5RIm(#g&mt+4=g6nw~gvwY?4R<;r(zZx7zI(zi&W% zaY6-Dx;u=F8wt>;N!24d}Hr~&ke|_q-eec+5?j9Ql*S>v!l*4rY7=;g{p> zGSPzJO}#TVUz#8Q7rNUpcZ9Ty93Hmp?evdJo%%j;A64T>0{qD2l_eFH@$DbKkJqYq zkrR2n)zo1;NlGt&XvmpD>S}kUCz7@gfQX+2z?R7kD-hN_d`rX>GfxMSXR-}!*2ynh zq{9?HaF3BpJ~~zb8N`bZYzWW=%(!;W_!c@C0~BK_3Xa_RQ9o~_7mYxOc5u1c!_SK|T1V;X29M=yiH%xtyys|B{a`|+M$#Sc+2)dBH}4jK70 zkA4X9xk`wn80oJ29f+cB(d`l*Ek#9@M>-QWCcT)O!__Wq><@YP)pbh>9kL9?THUH% zUiK&py!-Nn1R9QEKLLJU(Qb+ukyh3)l9DMQioLx|0p(v-7aPnzVPfl3blLvxoHLkM z2xPdsfQ}N&G!QXwE(>rzKW3VqzN*h^Jf#OA@t5ru@+Qh+q^lg6noankQEjBa=s}) z0#s5s(X;{Zs0%!mZPzxSsO5Ms&!hiajc+o)xS-nePg~NA_lclL+I4T9gH72ik+Q+8 z`h>`6aZWEI!D&<=4no5^U>&^N|Ea>p|FS8-w^;`HG4?CJTGAn6>g|Q}rRQVM^~!Z8 zpSAMYo|1@pA2&GrNjOlk{|Im`!uE%HlAURR7ZwX;r#tb%1_tBh4>zM56993B` z=R3!fox2dyz8}l-W3hh`z+#oC-nuzZ*6bEb;O{pP=6nW>{39x}D(9(TTnO*%5cf0_ zH)-s&m_Qfv2{XMlTMvPgh10nywC#qk@VS2UjsEGbsTp^6k>v6ALJffC1t2}gnNQ}R zUr>G4Q;)efDbFG5Za*ce<9%SuLy1Vu_&sa*^r-3=yfN7k;?|UA?r?tp#XpBa0q|tx$IHEvI%%( zP$vUq+ig55W1KQo(2+A&?q;m8Ivwp~t4_Oq)$mDQ6SVh1$|Li1rjoJI+d!}a7Ste4 zeknBQW?182GdT7O&S$3ts?4QRe(#$p)Epwp{0<%FIO)WX8*i^H8=gkAY_M!Ra*DYs z5;~~|22P17zj#7V95j1+1Upw3Wg$<@P&|k-(KGd|x~aDY?2A`(EEj!@updf2>{wx+Z1kJ6gF{Lk+w$oD7ZTT&mOl1e<2Eya-9xA_(=KA`2hbC0sSa#aJG8;j=1eqLD_^dJmw^0Ud(cqrB!OA zz-5Io;0`jWeWGa{FrEP*RAd}Kq(U#m7(81dH)vWlfSRFGZA&K2PEz-)j&F$_9-3562D<%NCnDh%|JLQ7}N~$d7_-M-V47L=U~x`;qYJuZ& zs!_XbbH_e|SK-$ETTtZB8BlMrsi?z_pW6!lw z;*xy52^IU6qMzmR7fh!_DSF-l_QNv%0X-Q7=A!@@hXb8*Kj$OI@W;RRZ6My`qUk$` zy~f&zQh4ho&`b)9Y%`BsxvN;WF<~V^#iiKytKb9BPBIWf;vTOopV1+IRK3GK%?J0N z%4!O{QAJ_>G1#Lko2A(p%6C_7=l4r(1LC(Vxl5lnf9y>%N(Lx~1{_-~%ao1K>Whah9}LXB3{l zo2;#1hn!2DGRW1bC69Udk($ZQz=`(U<;z58H50ERt_G!+5w)875tQEnAu)rR4-??v zrTWcJb<2J{rxCZ^BetDt{ZKbMTIu0Y?|WA6_fxauexY*o970vPNE!$=*xng}GTaEC zVCP!X)fnr^Rgs6{ra7Yw@0=doCx9xCra``h2yNwz8u*4eJL`t)G|V!sBtDx4p(^RS z4|b88;FamRRiPGSC&6GEz2c@SH~{u(Fsc8LrX`{)CpTxMx^L~Y6`ATx$_`NJTmV|+ zX{ktXi#ydIf~x@+2dr>`?m4I6ba6{)-R$@-Vr!DmETheqBWcE?Hqlu&)R^MqeVjp{jqeVI1(P7Mgtf>{>)7*wb*gG37{v5&Y^$Cko!lm{a zv4H|EVkU-?7*~5Cctd%mvz@W~C)+L3l{gN0s%cT3%~{ZVx}Y9V<|-I&R@I{PDEG=Q zERui#IB0@e)!P<1su+*L{_>^U!e#OKb0%HCH_~_pk~Fw-15;qDeg5b-!DL71K_F&c z>kEd-8jq}B4+R{C^#^FFf6y(7xFZs=!xQhWjU}w9iW*Vn11gl8*ZP`)atRsxzqxic0D-1x z zlF%9Mj5LUl8u5CL`I4nn@l6BC4C+poJ(H13W}Fv;H;1LI4Y4*WI^Fy(YNy_fYI`58 zv!s}`B&W8HCzK%CE{}n=h034OHzi z<+j5e@7Te(P@)pBi$2-lRpad2@X*3?Il7#${wa`dZv$nRmZe@!+}|9P&yM-LtWKu8 z&8r?(?QLNaDOy^-u9`JkIZzi*j9wjh=&lSNH>Sm3&?j87)Mf|_dtX{+l}3%0yJ`>(~1BBDVk==;*ABTI+r7m_&`^ z4RU|X960M5N#U$d85aoBcncw#XMM9tRs*mKC-rQ72gGh7h+|4Oy!yTedbuCec6+jYBt#R2pHY*QbPkgq0tP0sow-I2>gS^*9~JC~@`BZxpp zkqngo_#)@iPBQQ0$#*xpBI)dUfR#-gTF0E48DZBu%(1#!m)Fjn!^%7e{LRO9o{@z;!x8Nk;h z`ji{!64;sPht;K~4Fkf(esUpatE+6?f@bU;#%+cddnYC1e&S+1*!|Y;u=iyRUY1+Y{OcvKJXBlnWA}iaGgin{@N%}e=N8SZs@hKfKx*bq zTWNsnwRB5x^*P^)kam#g>FMcKQgN$wR(Zf7O{xJ$G19Sc@wY~Fy$IC3!W57V0 z?(@?Fn+Tww^j%XSAD?)Ud;Z zTcx^WQqBSxfo8g!VtOir4EncFFaC8TYr9s5v_p48Kiv)(a)=*#E)i&7T@+Q z0yw9hv3Kmr8ee6Z@j)I;uk?LnTK(NMb!F9ru(b$*EAabwfX7{?73aKcL{N->SX0el zFSK&6srRw}zNQX$g*vR9bcpgehlTyAT>(fVEg8Z{oZXobaMpIOe?TxQ({uH1Q}4%Z zCNF@8*5GUl4U}PZm4es2jQ;2uvp**tI5TZeC8O6e3NFtV!f&gm@OHk-#u@dt4{iOVg?@0Z`fFEi zYO_cpzPfb=jqXktK$=)8rQYQ3iG+({qX zPJch5$N+{G05rS&b~h-H8pBoGfX%?@kg~AfW?H3N?Gc#l<-KRnGn;O{$%dY@pVb2< zMzG4E*;~0#pk~5~q3F!Em-g#W_EwYKgaj&8TOU7g^pzrl!eAVE*^TS(dn)v7SPT+l zl4wQP=g&^+wYoKeP+U$BJjJP>9M1fI7d(zQrSe zEzVr*?kvCc4(Qb$kg!ir7ZTic~3(JH|wLO8e2y7nk49qN+|QXQz7?NK7IfP{Ef!$Sh<3^`p1y z{AzI(QP0YG?|n?aOTqK>Y0yrdeQR=70w5A)SbrXLdDIo>U~=T-rkz&bY^9o@*zwz< zFc4~)ztWIVA; zc*=}RxZAo!CrT!WSI2xu`AMa-j=mdKVVW)3z^|<6g#n}b=xn8n)rvP$Ht=??WZ<1( z0zVE>ES;&-2b6&XPlm4O1)K%f?Kn)F>+^|t{2wSyU>|Phns^#F36vT9brr zc@bFci0`N2Zg3g)CO_@M$Vhbt;)5)pF9PCkxLAtW#(wns$MEtxpo%T8r+DO++CF zves1Ar0u%?Q@z%TCG~Umer&~RuydLN{dk8GBnGVMLsIE!5k)B zRu^#>)H{6}*o_FJfLNpGpB~;sr@*?}nyIx>q3e08k#|?eG~6`6MAr{A@~G~!kU}gx z(|`Ob5H$kJUWh2h8v2EbP79tkdV9gt=;+Ov_TU+>uE+~Dubj*NI{W1my@HHJlk-p< zaEpndAlS4mbsCU<900${F)G1uYMVRJ8HJyrXYK z_rB+E-v7zNL3|5WnE39PS6l=fG$|Z>6|@0k?f?D6(wZ%dzA+ zof(y(LDjc=zrm7@#DD()!FKOs=Bswwj=O_zjm)2F;1du;&VoEP=`l|Z1^)vx@LTu7 zvy@HU>zzG?>b$-43}tC-Qjlk6mP>KW-9IA@>*8Q^7j6HUJ_J$BpA~W%949#Id)KV) zedhrS)gNm%{bA(eb%_vOSjH zUqMVz=!a%{sex^$KfJk$$GFDXN4C-{aqSWh)v_Co#veN%7dP&%Zu$j7MAp$EL7>2v z88K%$YU7l@3SG#_rvY1c{zR}m1HD~!UmlPdq z23XZ|?};@a-ql-(Ki*$1i~iaIXD-_P%|_0EVmnc`mrP4*Ws^1Y zrcKGCHizFZEyvinR~Sn((%da7-s;IV|0#!O-wGo4euVSrhl(A*M3~wP`0ZN?Lwd}- zpSI67e*BSdl*GLkZeM;KvHq6CX_)GAU;duvgfHgYXXMXUZs@H4Ic{wP=Py@f2C*oW zpZzjGKUaO38Vy7MVFUU$^AHuv^M$^I`PpI(bV&RC>?rN&6IN!XyFIUq`2fSn#D}Vu zu*HvIPOMQS_cwUV-)Ej&-TgY61+?E)e>R!KVKuVK)hz|^kUk2vI2#z{uWqS%9slF8 ziE96)7ZyplVKFf7rq1TN4rgK%Cr<+W*UZ>Wu2U!I8!WN!>)MvJR93)RMIL0L+~3R< z{*HX_mzKI=k81NdVc+`P?A89rzcJUC_$<~A&xHmc{6tZ5P(+DLPdVE6ej7}hruCLQ zNP`sc`H$M9i2A>}4swbBSTidWNYSGOdCciQkdI!bl&@D%kH@3Hu6MHSP77b|{b{_^ zs+v~VpVt?Bjnu9|qo2F^ATt)-VQ%6ju2Y+`Yc-(4vOdy-K+r_>RF{@y+vXvgrfG~> zJhw4*5QwW1Vw4DQPGz{oJs=aSwO^VqTsi1ipBq|JrxuUQ<$0etrNNTvnuttA)ks|d z?z0y1H4t#$rclCu3IAhd_^t}_#SzrLysCbh%5A-@3#=8U(L!Q!Ito<$;$GSn0PNTp zSX_6iY|kn*CJ|uVA#z* zW_~y%%%Ytja7V+(dw~PoCdCd@lD6uE5JNpMnDg`Z4kM1rvhIS%_)O%Y?`|f++Rt&8 zYnlj3ExeKIEVIq-*O~8MbXyH;?g9a}y42k#WOi2Ub3u%#E9XrWeV))XH0I=)cd2=w z@cwS7r5l_(e!H>jI%NE{B3#iqO&7pKH&{c*2-zu$8}7%ES^#^Cf~z6R2^>0Rm?vNC18-{p=_zm~Y?j^i%$y9PF%_$Bb*5$t=KjPDe}K zZTm}bzCpNl$7&w1cknkiF%2anQTkC1B+%i0Rbl!#C?))sIi*Gbqy_}Kf~v`a zQFUiuU_;Jf%DX7sM(#?VdGV&BU-D)9nU6i+p6$LOC_{`3(V(skQtH7GfA}T_z6AX?;>qGB(om z7<(+O*s<`D!h7ZNd$xbD>H9R~a{J77ybl*zGUTbW+R?l*W!KgO2-xvY7J4!Ra_|nn z!}ns8*Z%Z~`fzKK)%6bNiT{|QgpmD|Om`$DM3Y7cp#Oe=MvhcdNYDeY-nqE0>;P4{ zZH>6(RQ9CHq+hCuz<!#s?02cYUN(5lwo1v z`~q6Q_H=I4+W*@=ZCosRr_1>0z5y|xNJd5m2CL0TA*HqYIX!P3ZS8)#(ajC6>6!@F zjeZe-mt41v?;o$A0A zg03U52NrGn0y$g;Yjn&;^u}muc6q@URje5bs76-e>F$yNsxkvQrY!V`6L0hOJo{|y zen^rc77%;8#6iJf_Pz zeqj;aA5?ZIvFD?n`~Zj;Uq6aj0Xv3%1G_^JCAZIkB)xM_#|&mR?>P(dvg-OKB|E5qH1w}fiVQVZR$89fcyPkHU83! zx5u|wYh(9K1%*!lXBKY5*O2pQ=%tAOvf`MAeFs@sI0=9iYL2!bbB%qlK)`}<)X z^KQd~Wkz-HpLT_&Yxmsxd$j>DjJ>_(x+1MIjx40H?ZyV*@*XVTAv!Z>4qM`3+eNPb zbbp2AFKV#Dz{gZ7VkR^p(*QXK7KaonRl)=5Fal2C`_<;~JMg!j zb-BIEcx09~^ZumNWImc5r6w)Qh-mTm5Th+?i_L@|M=B|cM_%KsCU2<$@c^4~>jt&m zWaw=$I68iCawLJP2P-bL5#u!(ZacHv?*ku8R4A0C&N0B-tLToPthBsaHwEO4P6cd$ zO&fYx!ogFYYzUz7W0UouF`)#V2KCszSH9Qmyt*@IFi8PfUM#FWC zjqRdemd?$^g%svy=Tl-)z02Cyr$~D`gxcfsuTrfr`FGT&FyJQ$`HEz=GsSgG#R+D> z@_heK6A@+}k?Cn(N{9&Z+rEV!JR@|c0^EkfAgQ6MKG`hP8M>LECVJqOlEs};fl}t z=ES_MoI&`r$zayYRE1SgInoV8GZ4gU{YaKBMuDCQ?UOJvJTUzzypIfg`hkv;oupy^ZLLOUL z0dJe;xP&;jqeBf3UC(gYCfn$|;^?>gdw*~x zC|f$D{34)tU1%d~y3zxd%GL%5L)!O$;rA61e0V%6ws)Hw9sd0E_R4b@xXpHZrwdtO z38ZT1Mn;M*0QFEd5|!yLT?0y3v5p`2hZ})}{%y>CYhM;L2&@8K@ey8(6GoWy{NkzE zqV+UNw^ey$cWg(4^GP)(X8vY`U6o3>3l0)jNuu}tgnq%?o{T3XlXw=7=QfaR>`wqE z_vE7leSWf;&;HFcyoOxdfrC0gVKo`vhl^m{2Nzz3h02?UslIno2BH-qfuNZ*-p$D5 zAQt7)zrQIce4%eXno}p~wt~25j*z!^$EbdtXiXHMljGZ!qJyPY>Ww;bF<5f+BuKT; zz&c!*R1<2))Rbkf4_8-9YQ<@r2G8wgW){sA_K5i_6BK-r! zV51+dOdm!SKm-Tl>K;ztr`4@HE+wWEiA=@bLmA@^2`6+(w#v(b(?EMp+hE#?=8qmwCy6wX-R=SOKo$i1uF93tN=h!4UA z#!G#)rH`&Z{E;kZm7=gGqtU4;a?`42bFk9#!|a9%l5HUhfS@m7#!S)AMGVe-p;@dR zbZWTiMxvBnPf8juJ6xY&@Yugs?Z=b9l?U4)QrjPE{tUJHX7@W`eMhESEx^^5jKvD? z^o+^!h--pnEhD2G1*<|>NI#$U?_*uIz0_Lfr^@Ur2yg04|1c_-@Cx_t-Ro;rZ^^>H zujw2EKX_uz-3kctNy5WG1)!56oSs$$HpC^gfYtOL9~L_rDo;;OD_<%ng~M8P?1D+9 zY=xnE5pD(7Rz~ezSZk952Ccb*^_|Ts#?sGFd3nLDd%W@EHA-K!t@Vi0`J76+@Svcf zXu2jUg?KP-6$lmJ3KclUVnqQ$=rmq1HMI`kU6ffUDbMojawYgPzW#h=;tVfQh`<+D zdUpR!JR)5{N(vvx2t_3LFnkFku{TyT8jJj&-^yoj0eIEv3)AV;|7OUg?vEZ!Cs3Dn zcm5`5Fc+x?wkl7L0|%or>y7+=k}7(`gN_B~FSn6mskC;bg8nmU(zte~72R{YH&OXY zU9!EsLz#kRF$TN&d^As^mBh-DOBVi2iB@7f;Tu6GmFZ3G%<4MZw&yzZEpD%s8`A|jLy`53D5@}xkC2)9*kO_2lEp~QsX50M<#cWS&4 zeO34_66$P7h~xz22UuBX(qEyVp|L*_sh9K2l&S=G-{a9y!NWtX`u;=*tumudApx?! z!b~Z%Vo9aPdimAV$`@^tW%Dr+Vo1)Ed{B`i%SZB8y!_zH0x*z$Of8m?^_h0?O@oy> zv3~#Zf;H%~W1cUs*yDk(2Rhx=$VdEeHR6|SYPlqgaMoe{9|4d;s6pgl@J0})OR@%{ zynhC4iQ(Y^Uc~ECY7qUPLs$oN!MrCck`ww(@$W!Z68!Sw(MJpX$*DSLRJw_F6H(5rw!3UvsCHZ@DQg|#;h~oT`Bt!pQnj#^LFm&nU7P&xGS<3Hps}v1(t|VHu5|b1S^$>`p zw}e@09ro#89U2uQ4RxApxZfe1wFB^UI9oM-8~@9V5PFr65WEGF0fpATtKmX}*p&d< zdEWnoVBiOl3RRY5%>PIu!x)ikPcpOo-=m?mpd!8mcW zMrU%^kx3w1^yYBXdSQC!im(#N&Qn9`0rwD zMIfu=6xvq&cYuZp>Uonf$FRIVc~<<`yf@3ak(SVpyZ19pt;fO?A0wt84<}QUdG_Fnc{FMyAPe@*tyfB_ zz>e@9N~sIQ;f~s`gU@DK|3?=d3bF)%6;vn%!hguXL+yn?&t%0ai)LgxPyUc@Ui!+6!nMy7_uia z#E`u#yYgE9>t%XT5W0GTqEmqX5+V`GayhoeUwE*X&cK_~!eK{!n?_wH1qwnL2DnvbiYCQVv@Ea`4;#h{mvu?NL_2Rb_HENo= zdMgEU>##@Y?J589JKs?Mu4?0f{rSnQ0EXl5FAZjzNj#`qR+?k@0p&@g25+306-oZb zclLt2{+uiwblqAxekB922Y6|=-~=%w{))k=kU{#cD?W*mwL#h9f0g_hoJ)g+yUzJ4}p>5WIbXZd?F@RKYZ$u(V^_x zZ|@^YKfuAp=*QDmgxpIE%$DT)M@Jj++V;Z4<>ES@g>vez2|*r6X1M%*b;M4r+1+D&U%YV$g2pt|PIBAwiSJHwAWdRQ20wZ4kWR$LZaz#QT zZNbuL5Dz}Ku|E7^D08GeHpGzDz5e-0rSRKN))sO7790Cz+}}(!KT62%mwCH}u=3G& z)u(5JS4H>BoG5yxx+dK+Q~&%pr2?3nA(g@}oRy?MqJEV2v#hB%>=mAUG5teWYAN-P z*>>%rzvFH<8*9Nha0x~|5xs2{jlO;s56HTGW{#tSBv9W=v3J|oagA&AzvaW9!2T39 zYm`u?&77ZMoSM;m9p5q#Eh!Xf;o9qMFh+BYhfR~ZNv90VP|Xif_`{kW?eTvME9%|g z!Hx_#K~kdi<`>4o83tggmEd626&<*I3Ogcira9;V_W>rRGqX9R-qbih39|Bf>JPBhYwf%<#cHkdR%)P7 zN-24QPY6<0q#Fb}5WhE0%S-g@y8mp;#t_?Jvno<=p;0(*hQhUeeBYpXahOQKZIwk% zQSU3-x3anbmI_6UVs96sbF;y!7>?{-yVzIFCq?l21ikxb9LD_1h7y`L$8j~OI9A;&;n+@Ar zMqgGu{&o@b&%DH_mF|z|dn-ROE1?_=d7PE*D`MS0oO-cn)qTOxaMgO5~*Da=4@&9vxKDYKN;U_|;-);bWtW_}uQ8x>$*= zl>UGkb1%#KG&)sQ#p`iPAMtAVKA&VleI_XKP*y48^Jq#_)^vfqG)?>wZcK}+!A+<* zLEyk7kymEbv5XzJ<_N3|bM7dD1;lpY$0JBr%o18GnJK*<^R*K4)V{)aKR7`#8FAs3 zwgFSsK?Q0^)1`dFEJLw{WA?j1?pusY+$bP5aegza-KFF3`Uj}0KhO`#CM!G^J{MEi zb-Z7+PB^2Y^2dpmO1COxmn&G3B(d8mrv&dXB_DHC|0gI&J;~DhUGfgJ0MWx8EV*fK++TG`N`Rq`OBs0p)Dxu51ZNk z*ZNL_KKlo=CmV}SLr7zLv~wHJz#!-ZbdO|y=ojc7v(o|KS}MoHo&I&&{-Ql{OVQ); zmVP|?bND>$x6kb!uI4BkT5T_|`HP*?@f$``v0)6|pOShu%cHMNcNL1Pz&UEU(!hj_3zhnjmmLg6X+a!PqJ zK=XrmS9(^3ax)^3%?MttQr$K%4Py8%9JXhbykP~e zrvue~D5*C1z0>*#qP|b&^|<)V6XzZPd$TX5baRNHeg6-U<$sAlwYZQ&h`2;4 zXx)rMgrUygyHyMqaXeVnA5{_}R$1t5pv&D?4{7 zN|vg@@@GFL~l{gzu8OX z1nYVKmfuyFx;~ptXC!8cvru)AwcfLbuPMITo1&nqHvA>^JdUZL^!{V2q^^*^;}L#yJR6w8X1#6i>x^(X z?Rj?0Ya2~jTpQ=<`d#wtOu0&MiJFrvUlW^lfWJtv%Km)yRzOGux}GnFdM=L`$XT+- z)nfPV!jZ?5BEF`!Hr_($=5VG+&jd)lkTo|pI-1k-B~mknCRlT#DFXt~RI$)bDbb+L zY0wL-XZz5{;986wU|Yze(@uy2*``brX}VUii>rs zSuN}YO}r$>o5F;NgP49%Hx^kAONE`RQFr8mx7_k(k0z0b4=Q+K5nMu)>^ zfYN4J67Sd)Yk19#^xcn)sKSqeBezyd;>8}fit^$-fCs|W?%&nbuz3kO(a&%g4JtT- zaM1Vn8vc(TS6WciIPR^qweR*%Hrp=v-D#F5D|=d9RD=}K#Tv~^D@Klm%is^Mh%$ z`0nh`j<-0bZi%waVQ-26BkBd<#MLBkef8T?S@zF|w(SX6s`Xj!k$Pt)s|eD1jggTg z0r8HFZ}e_hgWr8U z#u@Jc8}h$oSafe-7?j@QB{+05cTQHT{kR=P&FgLVFemqs2)J@Uu{sFqdmw1A&hL3z z_IDBC@}cI8vXcI|v3vIW{8GS`LX7?6S`$K|LmL_u1?3bfDiM|*dH5y)XTf6H>ChB6 z-!$wgrsShIN>0*Kn9^GMuWV1h%=;=-2Axtck;2Koo2$=qg9mPBB2kv3HSQdN8fQ$( zrHh?mLPoq4LF46BiVxiRQPH7Ksu_dqM3hSh_7TyUGTGDVyl*B$lIv1@nrJE}M$!8L z44b94K=)ww@zk+1+f*0J!d`xlV${yu#=Sl1o05t%Tldza;KSJte+w5ls)mUyYM>f(pxS37J|F$7%`y#+ZgoJO3qkNec(~< zo#|w?Bp=A(55r&KBPa-WboC@2HqWKb-Ejp#b{afR~^T=!N-Qy#sS3oHLmhY=;J3WUG&!QYfGM{wQR=V zqcY5cBo}#^u`Dmoh?f_&IAqUk<@4|n-)D2jxTE)-e=PunYi%jD{NNv;C_BdW_*gID zK2bQvQ*f5vs-++UdvF4i)96U*j!wRBk@!kZlB*K#2Zl370({K~Vxn7Ay19pB=WDDh z{$GMtW@LPcs)4v$?y2!e02Fi~MgJmaPFcplbD5vKA(M4m`NRxAI}#zGzR_VAfF~Ky%wim_+k=Dm8U=?S>=7x(OZDyxK2|5 z3k#8GnI0@AG?3RN>-;>$y=D=Pxz*r;g^HJ`n(PMZ&lBZjt_N<(glMUCh?K|oNo6=# zJ2?WC5uCB_!`BP)Oz+~#fu%So1k3vTy||28yA&A_&K6%UmMRq#aFZRQE4HZRmBr8- zJx2S;%{Ml8K0Ltsxe0Zy2D}`Yi77I@dSFu?FPDdYcIGnrMG%A#4!K`AJ!^%)XxAU% z={gKPoSfOwbr&qX5WVTcMajJnf7Z+3n?q>=?>^?24CO#8GzGIjfaySjP;(PTrtlJRBk}#Agh#OpzXr=*GjZb#%TPO=QN!92tSh^>?0oUQZvSOM zK|o5v6$)q8yI1niX-2Jp5+w&c`8`!ATbR#72*+p|AliL)Opr91N8cU6sqK4QhAe61 zE4cDeRaR52w72;ihFP<<_wUptF7g zIb+AYi?gW%lD5i053oBK-xvO0mG4FCt@8t;RvB?%4XXcyL#{*rH>H>ve!9iT>*sc^&0KOGpL0#qxr2sJ49L11pK=RlP+GB2$;w z;{`QB=j#eeUiF2iDEFw$)7E70N*H$(-1WZH9FRwxNU)!lD)lqFx)x3Bmz^YC+P3j>wnm4Uf+>RSw*on zzd&yC6+iQ|p`$%}CTB)S5gzt_MW2&|OGb7($#f;u&5+1P%xLUV$0hb)yTKj`eA;bxhSzxM<(-v6+5U8I%?UiQ7=NhLSNw(JSWQ2 zj|8{E5Q8hY?0CNuS~Pog{dg?b5mo(jGwNVkw}Ls`M{vb?XGT_a;7~37QN79d;&y%2 zYhPJ?!?z~8l|iRA#XE`17kylYGklIxBK!?MK)mU^*fGtO#BsD)@ z(+yI?cXG=*_1dFn1Xzi&5NN>LyFX<5l!`lZ6DYvMU7_o~Qb zIgVg!>P(kD9nLe1U#FYRMMBp(zEJV}KB5{J(eIN>na%C+U%K4X(GWJjXp5+`nkIHi z@fKP9((Sq$?!ag~9l z&H|kwsXgQE>{@X;Vd4|aSPg3NGJa+m!gN-5&Gl`=Ec`yXi~@r};LzWaZDYrfQzWsL zz&L-fXE9qs=5&ldFxM0=qh$`bl%!y&Y0!0M+Wt zlwfZ>WJIt~yD9Y^oB#Ok^RfxouIGAk$hMs$kI2ctV$n&X_gCF&UM}jb`sjKU811|i zZNKWk(lpQZcZ$W25E8Zfn z7)8jD1VPP+hsd(@q_9P1XRM5u3o(Uftxo8*IN;JVo=2-h?D+i5J6*|59qEyua9Nu$ zVp=q>-BpfU8cEheWwH=~y-{Ns6X(&3+#A&wg+t?3-z7MBn5_&YK_Q;kaFZf9fE)-` zzJ_XSn$~Ez0iqVI(|KtzyvH6Q#MKfR_*#Qe1i(3!n4RYd)?Ie<$Z0^@%lCt~+5Qi$ zMFd#ES$Z|cHjDY?oxqvDfw zvT+t3B}As}ej=W!poQAB`+j-Ck^BZ?zF$-wPcIc$$2{Xcvcv69AP%F?#usjzbA*Fo-6QRe~|l{zj&`_BRQ~rm&du5*Dh;b>?YvsBqJ8%P0?`di(mXtmoN7-yYL3tqQs zR2gaF%H_W8m^X(%X%~n@)>4bZ_bK8KX>`%F1X@lj8ZU8_VC%dtPfbK1Dh1XisyglZ+&FPjB@Hd+_pg)v0mB=zV%thxsw~_1M|gef5cSK$dY+f8xuPq_4*^>S>dY z4W#IK^-xu7xTNlWcJ#`9(aVZ^g>zoSVL4j0!Dg{ey7kxo@x?`O{x)D1VRsXKsfDk} zG?*e+_bsXEZh%u_HDLVQZ!S;joOk4-lKNRE7d^6${Pw(tzOXr5TI_*KwV)%6kCa}s zPjq+7oIW>O4N%c`u>Mk6`MTyl6yN@RXUb}rS7(CtcFD;?I=v-R=Vw_XlD&&d_xqPI z<#TS4_+PNdRto#6hCJqkL#w^yOh?uxzp(*c4d*DQitn znSSdo?&E8R>`W~@Zbt+C6s2X;Bkzd8(e2AvZ+-EF<9aL8_r5P|ms8eQdXLN9Z1e(2 z0_K!N1`0MkLi0!SJx5;KN!AUL7Oo0QH~+`;q|3qsF;rk*Mn5ksyalU&9p@d1O`Sb? z(R*6Z%vNWTJX2jl36pGTut^^(T2a^FSFl9f6AV+Ip| zg5d&-ITX$D(|6e#XDJ|GS@`?J>T{jhiuj(4+SQ%r;qv5itBl(Ct5Ebkq>5e@0C}tL z{lQYc3S0tWWeG(Vzzl!eZiVL~n~8zoH>Z_a?xf0MktKCm#JR$)HVbzo53I9=R)zS$ z>*ZhB&}ylYe-SL6x_+iPh3#I*ZqH@`TVqv$aaVgG3o3;QN~-=3AU?(fpSI}S6|cwN z<(8mxGMELcB>ATjn3s2`7u9lF6*Ez{HEessGe6RsI1RPRGmbZ`G?fqzK_Ipi9*4-1 z$fNz)d~#pgfkbkO(dy`Tm^4)#?H*kDZ8zSVrk|WPl2@WDStHsX8>o*}Yqm>1`yg*> z?%d2{E`7yacFXCMm>Y-rOaMmZ%Z&wdsjuXc7&8&=Q1ownK((>`^vh0{x1V55ix&@! zO{Y=%OdDVNSF8sPm-Zfr<+ljTX)U6tM+f!q>$^f)8g#o!J#=?-g_)c2BYWy5qt?qcfd|+>H2wc9>!{QLoPsFlnONYYa%eOKWvqA3u`Qq*2>$thAU7o5Tl43CuCy zj}dbOS*36$4voGR_Z)uwcx>naCw3XY!<Ms4a=qaWBsALrNkLDYch!b4Z{j*BBlD za7}^y8P*iNm@!;)ro-3Mn177IL>f-0s6;)eXue@o0l(I z1@lRlfHtQW?8#<`$)~nIm9D$yB}xO;G{tr2lS}Lm^UnK<7Gs#5?+ML}z=!O0zuH@o zD{FV#J*KmKmaVZU)7ctC6f?JRlWa)bf7(wIbz0e0-1bN)va5fSG7VW=(D#)8a^DNv zk8W(9aaIt3zeKLqOuD7n)G{rfje{a5Dxc)r{TT#Rb-l_u zanWW;=^b30Ly_=18#9SR0KE%_;E28f(*$`kKql2R3y9$qs>+Z@m&gVXLt+AV!_KpnX=x5{1hR< zperXKd@qvMp!>tQ%y>RNmX$N;XMrBCzvVzUC$#Lbd%A?lwWHaq$6{k)5))=g;~vLl zJCb+)@>WYFRa$qQ*=ul`s(;I*FxA-IUIpCozzmtty-!YMic^r%`;HSnEF;x7bPhRvLX~vy9*smJvdTM>vnVw;jo+_)*Qh@sOPKxLc4-Y zXh?*~#CmLcO; zlut^aU;c)eq}3Yc!dombil_T*EduMPnZ>X_3?ZI<14UWQ{xto=s^&v#j)H`IS@^?- zsOJwo*w+EAPkfA^jg;58T{coTWM zdIJU3{*jHzb+y`wi2+!pRUZ)rnQ1#tdJe(j8XDse=+2g_ef*~_IDd+nDu|Q4t*Ub4 zM*uZ1a?QEV^KhPP#~z!>$VN7hpI&sxzfz}OB?4twSilP=D%tSLtRfeGF|&rU23thG{tw{QtL&4~dCE2c*LJj;(T_hEGkIj~u(H(|H5vN7hUUp^TnjM0*)TAJXdq+QagI59ROLvCYA%aFr zH4XKMb{ohQmK>92{*NG14S-O$5WU}VzmH^(Gd0!prYRkW3`Z=y^+`D~{!pCT3(3gi zC5Mt0A!aTyoOokXEp6Q7N?+w$$+Oq8UgXA(eMh=pZ@UE*9Bc1gmtWOarx2))44n|@? zlgG@c(;@nVbF-`HDk_V&O|JH*$ z^|{!DlSL;bc0;=zuj~;;u)M5G4lnH?&$3~n8eMtd^_f)O%Yka|6z^Pabl4krQMhzn zPNGgk7vu}4rVGXFbbrURjox|=Y3#LZM!Ib-*0ZG=f5)fD@y;g=-b%8yUg#7A+%Bwm zyI&$m%Q?5vlXQQ0sSk?}>nY?>z7Y)(q5!SeV^WJOBUsr>8cYdp`A zi1kFhu~>^|l=K`)LmA({x>`ng>}u6Vp*`y|6Y=~N1HUdM2d4|_xray@51ciAy6x)x z#lML%eKxYHp)Yrmgf)k<*V{}6UbVaaZZT1lTNNlfUYW@(a?-O!lJVM7zQu=r_Z9qd zxD)OuGJ%&#{^?Lh-~}}6mKW^;IV07(L`W!k#E!(WuT@zLASM7g zSbJq)FXl^)si_Yv<9LVHqR`r}{p=*)+JFvg^OGSWk~;pR{g08>(GD5)-+xO(g&+@o z14`pz3}>?bMump-mmT(uOphpNDAM#j_YK7|O1htzeS2J<`Zib$08f#jTuVXEdpo!@ zIeLUqTWjxnmMOY?o*Z&wdN_t5LYn)UjcRW5iK0^9v>tV(^zPk&i1$N!BgkmAXwdyS zbLq^nU(9=VrSZFhcTvG_XxJ&u#Am3eOCJudXcdx(>n_*ODE$KB|eZhIwr#xTtywh2QOg{2q|i%{r%*Q!sH#qvIf zq0StbvncUoGoohCVxRv91c7udaYjkJ1F9||k?xON5s{r*ou2YJ;%G>hIISS@dg zou?mWFyQDg9+jR?1D@sHj3uIu7d!meAr(B`B7Ab`?{aA`2D*(Y|e3hz~_rrB6*CsRR&0@58I{cVR=&&>UEt=3qJWI)yaAgvV zFv2K&flU@aMjze$XeVndCO{?LY%)4?C)digJWM9CM#e8vGNuKBspW0tm-P+nkt{vgM>dd`87Q& z=z8uGRiBeiqhYMdB~OEt@!@KHY=#`G*~B+A(}Bb03VLYb3BB#FE}JfQ(Y}mRn=}sL z7wLH$VGg-yRh#IblAIwpBex!Z{P{oYS}TnjF7;=t6C+&e&hQGz`vtLmW1B_`rS0v* zvvzRBl`?4rHy>kBR;NkVQJ@c=ukJ!Dls~MV{+cLPAk9u&b>{L|jv(CCnkp-TVIO}V zgI}E&m<5Y(^k4j6hYnRD?!9FSXYl$Ba`(*2dI;#~@JA7mM;AVg zi_PmbtDkPYbuu5|VXX5^Ln_+nWk7HPT z_sO1#r7E_J-Y1P&IiXRI?VK4Xw}L9n9Hx7mte`u(`&IQT3xuvznVtYognA zy#@{C>+{Wl&x*q24MEXo9s~=jKggTdFlC14m#8}~NyrJ|9k0#bR9X0}*Srq3{Ms_D zgSq+oee^wMl1+lMULxAt3u`bCktvu5ja+yK&ob*>vl-fEQ2Yze_!R^Xp5!hk9hlIK zNRA4&P?C?pZc12GLn|TM8YCOGL{EBF^Elp&Z5PQ~T10;)9yEx(L4(RmA+qbQL!tu6 z#9c1B4&{q3HsA{n*U2;-SeU@&i%_rQAXstj45=Y}|CR3a5~5ecUP znp;2TM*Yszvul`HDRoEy?R8oMSOH7q4JVHM;UY`wI*#KB34Bhy5Ljl`yg8K@CDznI zIdXNh1(`^z!=%YhBG0dnYlOpNLn=ei5{_9(RH*XrGRBWIcySz(hkSCO+Dc39pRD+u zZ}cL$`N;(fxbw!~kR3*OkM?Gk#KW!WHP21ND2pZ&$Zk-~ZOMxGvvn>mtxs;?*&}QN zY*<@)=+4aJ%7?E!+Y)+yl@>xGJaOnREbDbrv!*I|}1IIF!{;u1I|RN?v65 zh-p)K6Npf%iCVgPDf5%$zGhJN3!bF?AqsfAtA_ z8K_mlD^O9K0W|R0O}ls8w+)p(o315b4|RU}N~`#n!SK1Cq0W1YDt2fpFzV9sn>nzjy+LE zxL%=Ghp3a*e;!ON^-&gWl(!}BjH)ce@-GK#O1m;*5dVPRLqouR^JLvbWl07Q;uxDzn)!jmj{)23_6G^=SL$Tpi zKa#FXhw?IcmE6H|YfYZ7r)wP%FRqgSu9V+{(RX^6R}&?2=~=UnFj{84TF#tps`!^1 zWPE?Tu3FEUaqYM>&WA_pY({Y6VLub@)f)R4f4&2+sZH)4GqBODsmN3!nV;)txJy%h z(5M77w!L;NGQI6i*(o*gGMUaE3#W6vAv@q|+!kXf0Chx|u%ZZ+7o10V{;_;sPsp*{ z*hDwxQ~2x{CctZmJj%B1NZs9TvDTdY`F`;SEeaUxlp)C*ZU2CFx(AMLQLyvMnI|H| z)?1(lX?0Y{vQ!K&&F!SLQb{w zHu*iBNX=6q+^fjVA?S;cqR-3l4@RV4jaLpwY6&?(hS`5q!0A_f;`kIDKE^j6lc@a` z!wo)&zj|+I?RDR_R(AUG)H@fIH${#a5T_a*5{@kByw;NVC_Y|95YR=rIYqy|PrmM) ztIS<#Tc;0Mi%^+TPRH00=T?BJU+`-9V#wRxXUJ{v0O>^@9rU}EH86r11Rb*NzndkY zvu89jz3dr=f^b+{5}87K8~Q!*UDmO-eb@ zL2%&0$vDA;(o^9>eBuFTNR4GH50OZ zn{JHIw9nzRiTEvTn+)>kVpS54JBQcR0WvDkPCFlh1}x|O`e(7WS%#m`av^#&bqMIn&}jXAj^bBl+B+cP|}z|>y$Kv(SHh`RA}(T11zZ=AEmB3AP!79P*P zfGK}jj!jdA^rp0GDV4r!$J~q==aV(w?>C_^@hF@4QyZtSj8I`X26CI|yltc*<#S)& z*slC9+Yh@99$NrMyH)~vUbF%ECt_AsSNR_Q#nkd;1dEq)1x+V}nt>G|;UMiNc<}{R zr!S~-8e!PjtOJzkv7+AJU}3I6&%np2cfj;KFUY>>*&Ym9F=oDk3rUYsptq zvnq$TXa(oX7<>Nocy0|Qc2u}^wcPcs(16?{k`RGBUIjt)BI2znz%{p8Ov64r%;_U! z6jUu~Mdayt4EW7`#soF80#-orSwV4EBpj;J9dQ1?+OZ! z^jhv`Kt}qke(o0~@z*m2Vibh|l*_wuu%?5XE+&EkB@H^pT+* z&l}z9y@p~d7(}c29H086s890;zB*kofEwJwSjdX9f8KtHLRpPiIblcTt+twzkN0FB zun`ZpOPZN?{ai%$`{g-cExkHbP2|bF+c$~Vd19u!4#r>t~-i;Sn@g5g$(YmHGB z;PT(6=06}zC*uCYbYND3$2<6a81@cG*J#P$lbNk4CJGDZf^$8yYmcX#uTp!AjjIw|{a)K@O3HQ%n{Jnt=gn>tP# z{?Hb3!Q3Dukn1E_UT}?%+ecfg!_@QQS=SkhDZXBw@V%kn`+|VWf%vxc z=DZCCx9*W)?Gy4!YW>5|KpV)-7|11+MqyI=Je(E5b5q1~2er=0k#85>673AE=uumSIMBW-V02scHAWbP0BT%%7KMYr1 zHVP}e1))gJ0u878wqnVjidW)wlDLWehZO~7I1R1@-xCqA zS+JvSV@1WKxpm)t6~+v5snS<18L8!T(N;{%`nYZvf`MK|J4iNj56lEetFl2Hv*J6s zG>q0+9dslvnWRH%UsAtX=BCsQ(`ierhH$1b=_kEhiI|xEy}}Gt^V97d>1{OEcF%{X zr8FlS_dTHBb%T6dj-}Mcsul_R6)L&n*D+24AoJQ~<-sC7+Q|@AC*-*s`(x2hMnLl2 zdLo?><~m#ldMnN83ZVHN{|D+3v?_|@xBkE|Xgy@>=yK@wpA(JF>Z?$np*qoD0dLRi zkA9ZiveY?{&2NvxL#fYdkKM3*X#;-UcK`&0dEpmSfYpD&I}J%quf&g)!w#LnW#jpX zSHai-tGy1kMcai-qs9pclE-b%zYXqQw^;el%#nLK{%t))-rxMe&~QVDK97< z1`rY0Lh4(46)!!S7h_k!={gBYAVNJO>;;JPnIvH&l(5|Q4EwvI36f_FCB9Vr$=I;r zSF-)By*k9#auhNUQ80)Kcz)s3pVn5l^98zm@o;w_H04cb^C97XJmoDZ;=Q*3NgxmK za$8qjc?*+k>|Hd7i=&>36FlI6PueH=ER6RBh=Jnn&)bkC{4xYdrd{!Spy6%s;?7ZhUVu zZ>j;#t~xLIP;!{@_gS-K>71Ae$q#>Z@qE#llPyxBZ@INVDgq+X z4N6LPgY?ib)G(xgqO^2MGjt3+fFKNwba!_T4eyPQ&vTyhzUQ3(m-h=^*C2E6z1G_6 zUVH7eetWdaS$0X$>~9tTSAoW+2KjWFEFEaL|KQ@ImoU+O#$9jar&jWr6cZHFlaz8> z5vn=_p!8%JBZt|Cm^u?jdLY$?G86p+SNvZZ#v6KV6Grl6mq*yXq)F``i8-!sahAb? z)>cf|A6KlzP(3qAUNGrT`$T_Eh?a~A(z(l@Ja>wlhl~imhtEr?#yyO+K6dcfA$j5Z zK$$i%sHwyQ+}shd_mGiqDUX$e!}oK)0z%RU+9oL`M>7v)F@rzFmgv-UHv^m}Mon@t^QGqpkUXGI{#C%2^ z#8Ihw6(5MUM&a7J%6~OYX~S(jtc-K_;ipPVPv0rj&Qo?&5wLpm?fU;DHC_h!2j<4? zuwUpNvdjE5Dn*8PRNCVuSG0VlvnK3*S1{odh0&z(Kz*}G3~0;_;#%enU!fwDtn|EK zLyEqrvb(8mKAZ31OQK&GWlYX!tKxtef4-UFF=P6RB-XY_Fd+JZTT2Nr#|-s z3flu(i9;rcaw(ha1*1Fw80(r&3;n`tl&2qjbs*;Mv~z;zXOP#BzL(Oglp+>nJpM{m z@~#SRmC?OmuPdG?f_;*o^|dSC-Ttkd@>T)()MIl-_n;f99Yx(?RozNE*(hQ}ry z+f(1#_OrocTuYuDS=3jTcg`b?GPezP*kfrVitvL@6%)Dp@Al~o*3eyFQokENeDiX7 zNfNFk3W!DttpYcGsC`gI)6?xR^+%T3WfJjA+2n;-Wpff@d4lCYk7rmw@2JN{XD=38 zv0-IMZIHY3dIl;uG?KsiFeOIcu0yh6zb*aI{cTr0?eJ7_U7)#S2JRwoRK+e>9OtsJ z54T)pS2JYm!WR)R*(N6y>i(`aoUmX$(`8>dg{Prbq!mnhfL3+}ri&ZR&+v?#Z2qz= z9Mr6Ly|eIW8`YB*nag4Mv5zd2r|kf+oSiC^*82FcYQ)3o`A|C#JF(XBPkX_Z#iV3R zzV0NfS?3qo*J7%z4Drb)*?H>cAwS=`pBe?Ci%Z=ftRlc}7O>yZ(`o?YaLfsEgakRT zEV?e>=)L*TfNl^lS;PBmS;Br^lI*Rc;>I+j8U6_Qvmy5Q#;N{3^2@KfbClA+qBuRW zYMFp9nVWya3)zSRP}&16fS`biZ*S5=RYdj3eU^6pTfvLBis znhNT$;VKK-E-@(hL6R=H%qHym>NKIchr8obnH6EOiHHJIy8}Z{dvJ*68y7n3nK}tD z++;$05D_XSk_SWoN@d)W1Ki68o*|dGOhBM{2}>$>h+8<;(E0LgoxM01 zq<^h}H1YLw(@R0-Gq933UfeQnW4x|^PgE8pQc!7V*1&C_ftOaD0|wecY1=%$oIXYq zh{w(9vo5OKd{gh&m>W;Bt!)RZH_#JPSnz+zrLWXU!R?j&3GahENWS7IYULRpPoMJr z_JjL(uz^$r$$(jJ`nN|&G3f@NRYqLL^ru+>#(|fltD>08O!1H<*RSP*i9tgHgIl_6 z%OB`E9dK+dU3HsxsVMeR>0G0V4bQog;d6b#LoG(EB=5}0BJY0sY)y!u7>8%7G0z8s zKpdAD&)Jh$H)>p*fql>Jt=6Lx%O}iwC@#n)L#}5)KtO+NA6+68I9S8mv2*?0JlSqn z_eMU6Eiia9Cm@2Lo)8mT58 zI+t2OUPq2|iHD4lLtzgmXGIlTm4c0@y zM@&JuY_mQW)u)Z4?|FE1P1-ta-<(<3RMbJy|`S>a+y~C$XJZ7Y2!+8T(|#X z{gp$8VsL2)g~`qwGO;A~eA*5jV#e1N{5w|B;;#5;%Ugx*>*__ z5U?O;KqiVCYT^3}} z8H!F^ffdE@k(un8tl;cG9$tjxRL8mqZ;Cv3tiOl=aRsRz3+G*-e|_0mpc`lVGwd!; zL1Y(oF2U|KE1%wIrN>3C2^D>F?G0z67Kv2e$?8s#XpoD`5Bj)!yo&ywpDP5ks`S)i zM$kR-GdJD00R_)ty*SpQB%mSVdU=PgdD?BA1lGFqu7+rNEU470E4GJnJY=W;mEOyv ziw3GRpkc_2{Zr^y?)GBXIx?ID1 z%J}T|SN_U{WERet>&$#EzsOQ}>U(#)E_ae#nA&Xy0}G7O4iMMSOw=V%{-C{y5$qy0 znwR47cEauUgM4Nt=ZlRjZQvYgR^YOOk*Z%#&C3T28i?Mm?_qncu675kf^c7V_uUuk zp1Zu1iI$SccZh{-TIH1hZBvXueY#h7eD3&qFI#KTipE0C<*PY@_%f%@(k}piAG}so zX)&F+-EgY5VqqW5>ai>MbQ?^Fe|C9&k99KkTZ3J?Os%ui3dab~hT%PIjO+6)^(F+y z7|JuhQ7LvT>Qo$eZ4L^l4Fja_vddRR!Mb@o`cS{?kLa`W-PLt`-KV?2<(B83L*1|; zzqNjQ#B=IjC1HGAjG+PvF$!u~epSzzfppo{r7Z=h#?etf3q{#wC= zqt1$6w<>CV39H;WbAf{YBf_7x78Y7|{VWwi>^E4C8W!m>?etA;Z3kG_V)otXi7T=| z3r2&d_WD=-sld-iRURJ@!0`vBYuI@#LF>htDi*QElEWD03Pc{IH?qm&`10+B{u&4gy8muOd?#x9wUiF@A<)-HDFy+7gglqUfb_lz|j z>#)4r6}{tHo8Z<$W)gg3j$xhldvF~|)3CwZ+_8$fVu`Oo>~hbwsYFsmG;pT|t|$3E zX;%Ko^i}3yx67SXf&QoqNWH5uMy^cCs&4wYt2FwPvZcvbyHxp+D$09~6`O5Fyb)gn z$}f<_AfJ#)X?!-c4r~nTQ?k0_B}h)~gK75gdKVQ!z;%HpHOHG{D_^10dG|UY2X9Rc z1Mc$+;YH2EbtkdTamI_9b1FK(fV}D*dowXQT|$gnQRf9 z9-tUB%J0daB!x8YSgaZ=wqxKAeD!0s_u=YIthMJ|&=W}{ONz{KL{CS*hYu5+O}AVL zc|xdCs1!&W~R#GDypuiB6G>JrD zv>)X>W?s8f^mFQ`p2G$SRkL(NYHQjjwBx5np%7ZbH6`A1%WPg|Pj+44qzW6I{#ME! ztKbfK%xXp5iB%&t_T0>?doH_gdk@!oG{%|FkFsx9CejIg-LRV(Kl+GTynQ*O>r%zF z;b7&eTT9>J(|bei>BUv`387W#E89UVJlNs#je0x#R5#A+xH<3iRi4t5fR>uD*${(w z_nWuf&ZI@PwP+uEa1oTscza-D2MM_Bxq9vq`97k1j(w02V>mjCX{*Q}_T2|jtj1%8%I!Lo=G zadguv7n=z+=D`4BM9HVyy^||c9-nAWuL5o##|T|{D=7^O+HaV44;~qX-yKgZF{h#6 z^VC^7GITi0Dv8O1%2NSF@B(haGshP)$#NnABXFc>gy!~5Kl!%UED^;s7Lfv^SF69u zUfvCs%uTQidu!TWAv{SUPQgElK?Hg-%NEaR??3RRMo)!#TS7qZ7F4`NO^#>L?&lAk z1(T6OMW{%Ma9T6n@d+C;k-qJHSiEW;hH`(Q*t&c-&;BEmh94`72*pYuAHz>z=P9Wv z@pd>BxXHEL#Hds`x=y@XfmEO(J5q%es6X2H)qWYv5I&e?WCXr^9vD)Vh-Yo)21}%uC)s--wVimK|?6#@v$QmUMGtw3h~r=ifG(QQKND66g#sGS+m^5cU>= zZ%lq$AI4F(FQs{hk89Aly>EH`;IZ5z)4v2FpR}lPVhd&gyub_negwg75B0@9}u z%VLymJ;h$#M#Gcn^jKMjvt_MYnI6)Hs29OEQc~EIVHr3Ig|0nQel~r04ki zUKtzc6Wbbp)2>a{x8t&9%S-DQCL|#1$}1b)7{76lG}Q?AE%$|T7VTN|L?<9=g-1(i zWcc25)p~Ajs$2^9>E>u=giJ)LxS#)A2)Cx#y(6PglM(Lz-A@ve_DAUoJH-_$-y9K$#ZHLt6c-S?_=`C|o>0$M^Ji@kO(;9->X) zk@X~JsXspjA?*6kj@~RKRVF1jxl^L<&zwa_P=2qG?G+%n*?rRI*vq+c$S-MpyL&(Q zjf>@UEmtS_x_Q{;<-Dq#`vD%>9LVH88ykMa*sgdFYYt>7eyE_GQZ+{4iXKDN(KR#u z{cl&rA4t5?=~HuDuoXr#$T(Bm)pj%6Nl&+RDQP1DQFco{r<}+P7AarSl^2`!cNEzd zQ4?-HqO386nU3TNu2*Iyo3N-MQ~5%ftAkgeaT^`ZdHc?FIddX-YaTAmv;?k-+Hqk6 zYHcTSH2{K%UCmTr3SmZoK#ei-&&8TWZ@ov?ypUH5_vXs65t2QNG-c7rVJva6B|~i^ zL^6rQ$w`T5h%Wn)RJ>RMbhANq0Ptp8D9qOM%YbRl|@GD_1(QB~Gge=`g)Y2RTfG2bV(E zReterCCF3(-V-Gk=*?x5p;(vI4L75^C8o+)^ClC8H<`SE@$lTuzBUeTc}C!_lx_s& zAh)_<)k55tmOBFb#JHrFGEy?>>UJ}wk={{r*SLdNlS3H!9TR(L0r$Z9&IC`aw+!BS zfAomrqLZD$<|#;@n>znpAN);qCrC!YDA}g}#Kf4e(cSBQgX?aaT+hj`Oeuwhy0Bew zE2GY1Erz)B*X`t;Z*y$oNO<0SmRh$$)xb-)q1s9^A#VBFnn2a=j@&qSbj`moMd(#q zO8Xk6N|F(TOpV;v9a#Y_H{k3QOE@9*%>mFVy7!N`C4k+pInMM1J3~PRIz-vt90aEY z{c4k`jqvhLk-6%~oG`xxAt&K7;#aqq10X0{Rr9YPL?Hz$$W9_RX#DP(En$yMhE{Gp&C%(Ud<2`uCygu_eg znG)sj>bswDr7EhjWK|j_u||PIpely1Elp25ljxcFBSo+)EeE87yD#3!3}(i>5eXLW zwnK8X6t8$p3^Ow?g!*OZHQ2LfW>w4isxM&eM0d|WlRJ2!p2WJ?Tu2P2>5eu1QeOG| zC9z5x;UE#DDw3OWXXYV8|J)QG5B_|0C3jdBoQyLq`tnRs=Xg7sI`zr{^;6tRe`=k} zQf`X1A#uG0xzU~}-nh;vrBfMQWZ|paG1lgXv^$@p%Tc|)>PYC4WVK2_3iZd7K3b`0 z`J(aoauU>k%d^6_>s^A6O|NEpH(WWpl_Gj+^#hNGG9A=kX)#{3sNz? zr#&r!V>a4!M9_9u3TqHDFluhstb=A?lsZU(XO@xQmUN;wLgyn@=--ESMW%zE^Bdubm-3UQU&wr z=WNvIa}*0-`I0X6yMAY?YvJ0N{4tUus~CRtjqxRa3zh3!HE>#G&K`r{v}2Z0FLH7> zrxW0Q7JK(-y=A}a(p3mk3d2eua_}LE@=w8G=43=a>q3jOJzL?zbb^dXrk6SuTyyrLqr4jWJFTYc(H>gE4QA@jJ zO$KvdefAsWrS6nkb%%r3Yk$!qx@j85A0@w>^uboo3-qh_F??(Eo0!1eU-IP5Bx>Wo zYr))sLs>H&tBfDC8$%iRQ}elc=p!yt5-d!r#rHTC>u8t@dNutntUF8K3)9K_STkd&P%2q5ZFjYdJ^C#?!53lI9_@7@sdrs9ZX3YY z+d@3u=SNZ$NjvF2W1;b)X`I|5*;2lk#lpi=Q<`!2v`4X#-1@va`OuAA?CS(BUdhr@ zrdmz03zwd|2#BsHmS#;i(eqKZoSYm>a0++3Cea!|CvQI!R@wk zbu%Rlrp3+bA|Y6harlrZ z+OeRIj{FK+r%@Azk!}={6z4j~C{siU5`n=gN0ljedeiZHI;*E;)^e_m4Pp1Q%2&@W z+&Zxes|bkiBk`2SF1l}eF3_~eTY9PsfPvmbcXXxOsma6Idj%GByYf9OGUwvA-S)Av z7f~D6P@SXfXDfc{yZ7}bqktl8j=Ygnt&7^9No{Zt!B2Mn&imWxoXH~>o@&vi?>rWc z8x5}t{EwlMO)=uQLgk2^Un1B`Gi1BdtTOC&5+2>W7j>0dlC#r{bq}fL_BsT^6NH8h zTRfMe0Sv9wOzv1JPspFXN1m@nHyE)v zxM3i)^YeLOOhKEHD?wnuexlKye&AF>CD`hnhY>#UgPBJ4(1+G}iV6PYRZB{)Tc8)H zvTo3dF7EUo$=X<*>n6PsWe$Do)&-^VK_^KpW6x8e=$ODn7i!k9XN6I}$5U5z`OQ7^ zi|~1XfQ1fY@Au~2TCzLvI77-i*n{<^dQu9IeuaCD&Xa==ry+i?E3{APzC^s3KDvSwX`$_QVQxBUk>LjpG!q(wOO!IuF^Wo zVy@GoF|;7=-hTg~z)XJXNW0?KXgHl}mj%L6+hr?bMCkJEQ^UDKiC%7-VhWx z2vaXYj-2WKenvg#+1uYmK2o7foM=FnsihvYQ(Y55pX;B9O16hjM2>z^bCO{V@zWQ-L%K8^BC|E}sK{bGs0yN-S(Qyq%{Uq@P+7O6l9Co*7+DM_XJDZb z)86!>B69#bj0$WobnoB6;90jxhENx z3@OU-&pR~l=aJfH(@hy4eCZGPcEkcC15kx~Vbh}$v*`eZK$VQUo=*F+2CNxU&BtLm z{;jmMjL07GajBa9b(xCi^|rv;8Jp4ImH$2sP&NPc@HeRpZG?Nr84{o|)O?Q00``%c zYT*ugH}p@v*M$9eBGgpCQbsJON&Y5`rQq;!MP@~B;G#PuYd6g;?;*XTZrbv|%qs4OUtyxXlMv?5`4XaOG4$+(6!T9fi{Q7X zl}y0u4D~rob;ZmmK12b zo!mMWTBaW`xlBJxdoAYQUuds@3gHbOBI}nGpvqVmNp40g0xWJu9{#U!s)iP}&*aWw zZX7_h`9w`p>WrI3Zu!B2i+=`ch)?BA7j&~V1^s$lYuM&V#pj+_>`&85f&1~z%deG& z9G0Ju<)UJOT{k5nxI|J9?|9(c$kMis`W$L-&dD1amdYCleS&GBVXTulP8@21A>n@k ztXon14Gc#`5^~4$-Z*yiv2l)X-O`Q$B5^FQX?@SY8*)I48)pP-4uFcB=9&zR?RKDl z{2=o9-S)&o$q2^3sJOESH_Xxlv1i18DATk4Hy`->(dxc*^T34T`j3v|r@3~u6%bkHAguL}1a>J2DBA;vyRWWll-ocM9z?n`>hJxC(h zgYS0Z{Ydwb6)L6H;qDyTe^AH3-vF60JFyqUe@M%-7&imJ@T1oK{S0Bk9EBTYeGgD( zxuL(_@UM&_>8SmknDjvx8)(cS%UvmEDtgctsq#tA(zt@FiVsq8_xa{@X6c%^=|SBd`7do)ZbsE*B&QenrO+kN>n- z7AkXeWbOnA>v5uJ=($Rn($OX!&zsFKLEhL7r1TtZAok|4JNGaYi6}FgUV_IyNS|#p z&9?$YD5w7rhvjebV)D3=PD+0Ug}gCvm5y5~HX1gLZy)`;IVu=$0K5 z`0KW^xg~YBV{Br@hnIbi+bqHciCR2@Qtz(%S68Ng)veyyT%^D0n(mO3d&|N0&%Cp5 z=A9iV&-(|$Jezs*h)i_4sRw`Nt$8!=vG0)~IDZkRN8jG$opr~rOq(iRF-c9A6+5~q z*Z1BZyj7HO&$5?QNTH){Z2ApLKk?rWNxgw(>l(LZ`2MY8g=c^QiLt{U{jngzJ3$`- z-|@~@2K&!bPM-X8>Nm6zvCLdhNfT+n9YLzw7`R-oNGd-}t{bN+r3OT-^7FkUvU&blaymOv_1e!!k$p zvZn$R^`_TiBSTN6%?t|w>zM8S1qI(6B?+-b?1bvyU~~CeGQa*kuUXX_>8Uo;B;Ef@ zdO5(N710&BtHdJqN2iuIW3mho<)!`0&=f>5oox9(1yt5Vpk1I@!(IPE{L8C7V$_k} zaiw+2w!STAAsK5@7mQymssOa#2((e!N67qefwjcgm*fKd^&tY2Cue4U!L1p z&FSz$&dUEA#ame9&WpWTV~v#C80`DX(VKDN*}aLST`~VG0%1bCcd61QLhEC8RhDVa zhnpFEt3PvfnBO`7S!Tdp0q<`5W$iTwb#P0G$KBa}0}P)xD|H()691Bta5W9#gQ-vPLTBYmrvvKD*#jgb(WenK ze~IJ9$zj~jQ7Rx+-5D$}5clkeN5^^2zEkzzrsBC9WgHTGKn_1_5p+ajsd zXg-R4j~OQgKv z@hhf`-XG-JPNM~PgpfJ`f&91hEq`G-^g3+dtouXpI(4AbdXpN}$@yt+)!-AW+Pj*A z&5ou@9T#Ksiq^z$p8t7gD41ftfvmD&TR*ULt5-&bDz?T){eMI}{GI)D2$Lx`JSD<& z{#gNE)#!hlQzMa&b2uW$-9UEa?bfsB%szs;yS~x1e}mY4g-_{jg|1Ns-&5|*64T5b z>vayCRr%sNK%FB%qLS*Ti9tGn)A9ny*orh8G)IzB4kd4TsL^JyXur?>vLf>pB!5dL z`Hw-w(I}*Y`9SW+4cG$u%K>nw*GH1ymoomjSAiOn@N@(wt2$B6`Y1>?P+dW*XHK+J z>Muk*poLjc32Oc-d^l_XuJEK@7sY8UP*w5ELFtcg2ME!?bu&gaGVg`fpIB2QH^B&; zqf+MB@Qzl}xQDco24eb|$VIHZy?mwC>qOP)Q?2ala)TX>x2B~mI%AZ^r9(F2l1%73 z)&ADiGvzy4{HT{05*#lN6SLXbpF@{Q|2-aILTcD_HJ@s(7F^+3>3+Y(kq?G{bmZT_ ze&0s_5MiV{^!b}q_wTRBxs%`T^g>Jj+j#Nr_+EZ-VaW{Mf#1!^;?X(de*j=WjF zC%hNezJ}BH;-yINkMi`j5C@GY!r7f7n>*yh$dCffcgW#_v;}}$wKh39|Aq$qw~hMH z(h)i?SU;LB$lh{yyGNifJST)+G>`(b-=lujGTbWxDKLZ|8e6t*IC?r+Shbl>B6|nsd`C;^dgad)0dMRhK={} zMH>S;s$dF5joc#J{NEi_DOZ((Vlx6Kv}T>pF$b|39D^rb;IB_u1! zd2&o`v1%T2M6}+c{llqFGXs_tC1tu4y@n3*a$MNj*PdWXRZ~c!BtV9AH`J8CytEJe zt#vC~F^2gntWeo4k8a-~6u#y25#!DB;pZdL)}aCyt$XS&GYwv+Bt}Zv7(X3;|NbRo zDqA*zz*2ct&5<0A#1ypX^wg*|Lk{QD5-`MDXqwW(MsmcwFb$n4_iT*x?fMnZH^FSV z185(3e$PP%%?+Pta-T}EQH~cDmXx?Nn7Xk&uvILE;@|w4j9)0M^`jhZ-Bn9%yPt2I zvGdMgjhz7Xgzr5QGxDLqaX|q;5Jhk0R>Osr*B#w_oAWuy`2jK*wbGmLzVngJH5{C` zLF-|raCCXB&-ZfmUgJ*d?|D&w!8Ycjk6Gfqa^>4BPm8W}BI#7--Yb#sPn)=ZalFzy z+aST`IcCh~cYXJgV&ytkdc*al#H|~V%nIM%K9QQn3U_6H+ZGUJV{0xKRvmyftgrDE2DlugYQz7s~biuUlR$u<_p*CI-De&g##Slqu z$iH^`9)P39Ejt0+mIrQ=BuMcEA?}B-`mGQw_^zii+4&EB+ky~f{V$XL`#BdbcixZ} zi(1G&Eck{}|1PQq3_3RJ(@@@~o3t=Rv!uuoik=D995~AC`U%#zp zUXYt+DLZsA2%Tm950WL6q?v`h9aixv%fe2dQV}^6`r#CmgqrumR#Ii_elh^@d9n$N zJv#*rB=n@lYjj~Us{20}l<@4WM+%C$0Vk@@x_9f>9MQ3IBJCy%Z}#5k zfIH*Ss{!9=NHTIe~J(k4fG2Bl2@rxD&lJC|xymo`&> zMR764@odMFH}3rv{LbK-tKU@PI(EL_b|aD2PWSBknyVH$_J10SC>nMuk$B#OjSyae zDJMDxW)3@{OSbR2*OmD+K}WKWe*5?P|LE#}o)$HI>iVkHS^BH96ruR+EvnmS>(I?Y zn?@oPref4Ve2vhLYIE3f_BOiLe_UVKw=sG_@t_{rYpitLXA~cQVP{=B+ZbJq?*Lb$ zo?kw;`wx-?f>Z+ZYG7Qy>1EiI2pHeOTA8dwD*q{N2P+{j-yAkVuALZZQ zt0@NDt!X)FRtQ|({XZ-1AK8rl|0LReS%dBxaE0lX)__X4JPy&bSwTy!90QSqlqu0G z(SIIyBX%_44n#`ye+{hkQ{x$){=~#iezD7pwngAlyAiO>FOGQnWuotNzm>VzdV*8C zdR+{1K1EbvGo|i49M_V+$kS3_-Z#K3f!egEVw}dSw4Jfxaa{iDa*$CODHB3Iww3t6 zj1l_v>-3&5H}Wgz>pGlP zC#`3bBU2qoE`@wO(}_m?j7=feSUx_XxMjp<-SDj7jfvob!b|zYhm3nOk2Gp7tT)Cl z1^L!8E;wxli=BkpjeXa4r3jS*Ib#2c4t^_&+7d7|v#s;qA-caU3e(uZr$r`OyOV5u zVwqXn8Ip-1C)nN^>^UjNcl1qV;_`GQWw?GpS45JNaj0e$V{2mCz{P3Z;2wpO!XZq+ zQ}1$Z#o3y$CyOX!<5u=#7Gt^;@&^a}$0ryTRdT@V|bmX6AuZK%C2kBj!~;ncmM3b%vD zeCs)MTW1@ErKH5_jy^jsGuH+((e)-`0eF==_2<)I`KzDBKF%UhL3`x9h1St5kj_$* zP8%45e8yIux7NzHO&$gZ0Zl_~ReA#QH(l!^c_w1_1Y&-_yW_5n0zBtC<8j0L_t*GSI(^%SnY=JtA19(xCEkjpe|K$x!y&DUXYR3jGdTW`w zqnQ(A;2;@{b~Oc zT-3>b4Lc!F=;~8mVhH&&km>NIn{#tF#Z-T|Jic7XmbPt!nSP%Q>2N>(lGaGFGdh zdsNjd?MEL(;~ZsLJp?~Mv!fBOuffM3n%A<4@ZW#QhHDzvY8-Rw=^s7Yos&oz-Hg#M ziCO8XRC+4&RGT+io3@LB;LN$Pe%2<&Ac{h#4J2 zuc!D>lO3v5HI|cv7krLC?+0qP8zi2T{%|a-&6O)B5256CQB@~ z8Jbn8JXTR9v&Sl_jn|uLAXU1CxpLs@2i(r+2BR8Oq@N+~;Z9R>egsd0F{+P;5h8#1 za=r2Zyz3D5^6(;ADJRwBtuVr1PE@LPtoHx{vU%_FaBD5T(o}<(}42AzFsLkKbM6J#S(oQ`8mGl~1 zbMB!)NF7SS@zT`$ysL~o-zd8<{8A5Ml10YKTWY|?t`0Z1#=b>VjPF@BV^c8+U!Ojw zEGJ@`u6!-xGO4>$$59G(3M1Cs>+ND~xB!_&<pE2y3?PwZ^g9pu%X4pF_7 zA8M)g-BIHL+eJ;-q$v88-;uoBS_)pjdaPDI7Hp?37 zHJn3!;+_+QO-&X?Zfx_`g)Sc@;(4sruePuW3H|a4!*g@0!hs*{i{)pXtHG>PM^rCQ z`a;M2RbAXHC*FFT6Oe;KaozZb)vGq@);k#3Ml9x*WDDV$oiicTf@hlQp=cvIBdvm( zmli==M`+^d0Y~}Q=m=-Bi|oMJiUU9oyN4@-Yt!{E~UCTXW3m-#C{w zvy(IXln7x?IQ1s~NB0(QE%E)dkzC7T$|;*rl243|p6#3Ij?$>)uasPU?lUt&ubuot zz`sQ9jU^@V#d>axlv^$6eUuDk%UQsS2jlk{I@YO<2o4Ujng2UB4Flgqq&HqS< zgpyBpTVgEprQy8cOS;>u=GL}0M5p9C?`uO`#vOYbRyl-~ zWu5Ugr`uCyUvyop_lKP3sEZ&YJ-KoMao%TnM`ZauChD;;wNM`36kQKRRlcLari#~? z<9dl4NR|S7`;0SQ-qfq3k#4o3xhT}wihfPwYEvr2Np%h8?xaPyx)IM2QX;x-C&Q4c z*61Q-XY<8ZdaILc5PqHHl!hajEqG9OPU+vxo|5qGMWY93{!wk7HY;&N9&MFU3qVeO zC3u=2vl0iB2Ll(2#V4ojtE(Op!3N8V8cv5*t%vI};)!rz>#A@8NDmXkiYwAwtp&1# zV_Dmo!Ub{fS|7%W%oZg2}rm$F7~)?QaH zZ&_?Y=+VgGifIio$zo;yt2+eD z@M@zd37~(ek6t3KwxXzqkNsm#S8B%nReg!kq9?GBR6EJdbw4z=4)k@+Z4n2_8@>%k zc+4+{9aFjl@V&eaFS2BXCR*U6JFqM!b$+3odBG7?uDsW}$NIge`EklG<+mrp zr&JMo*`Dmg`LGTxPEuRFs7BaD_v1sf-Ttrb&y#)j9%kK{b7CDq_x6Mk-?@$WYStTP zyHi)&giZ9)z)>$gdOq`k>R{zagT#+c4z~l!rR}ZiOzBbejfM)dtZqBiiFHi_XFE?I zc%IBVkXepD^lzVzbUV;5i{>hKIX)k3^3QL5Iy7agBJd{XLxWra>GljAo=+I%ZEzh; zS>bbXg&o=A0;9fvZ#1Y6ZX(DlNEgw+qlT=;)bH}6V(0;HhkIR}eWO^wa*K}q?AUQK z(YYVAJKcPDax{jWL37?(G-E{Sa0gvT-n`4j(k9a1y zW;k<*l(W7y8e`5nAvKL(=l)LR@x}za`p0(7<>N|MTMu>+E+|QP!E~fLRUs;$^?i!D zyM1G&U96k@G2dpuM$ulLGv84!oKxuu|9rd1b6@?Lg0f)(io?{?U%Dxg2;?|d0qSIz zZFt8g(9fn*zL?lIhVpPez!sJ-ROG>Ntn@P@JD~R9XyZ)q8}XmwKfyT}v_(EX>hn?&6c0)*#)b z7OKsXrTe0;TxOuQa9?7%H;AohPG@Z-rze-p@xYlqx;R4tijCcs1ey{|8`enK11#%S% z-PNje#)DX`dbh0`qpvdt;FXWZ=sn;GN!gouND=fWX5WI@m^_|(l=0kH9Rl9Q9pLT* zt;f4IHiy>O+Y)_};0h+r?-+HRSWLw^m2;W2X1wdO($BS=g~_5?ptI*2|D(9stM--7 z!X|7h*cOPc>$%XtV}cw0?LRbDv$JVkX)#?O$7?$;O-SVrdf}dQMZGC&uz>XAjo}4B zGQO*|)SxR-GNL-%IqlNp(VdiGm1Eb9orYS-V?=OGteSaY4A!BNkF9K>)f#NGnU(Ye zOxo)qqflLBpAQ`5d zokoIn`I&~-PQeKLi{r@u`-+!iAC>hXHv3-tyomjPh+SqoFfm`L%g;#XZwgIW&S7-L_5 zfXfScFqMwpb6X=)kdNnQesX!+;afbvU;XXNapjjQG-{Sa`5Luwk%*ZX6p83 zh`gG1-f5$8pPU#G500Cync9ZTOo?!<5ofN3Q_MK81|^9jSHaSF50@9qTuG?vZHRMjP@w}t zrG4IWKxKfnL1(3P6OpcbZ{wGZy80Y8Y7nvSYoKGo5v;6RKk25nUDt18wWtx*oQqEm zpRJgvo%m*!?5N8jG+3l=T+ncgPF2$z{{}{-xn1;}M4xItm&pqJIZ0Bva&tXdVED79diGxej!)ivXT1&0kd*^Bi3-M}>>5A?iPItb}(Rb^adV!W= zwDv^(wDZkwf$H6Pchs7heuk?iy_1xUGmyO6jGm~;kRhJ=d}pvv*NKr->7jzejFjMi z<%p=mfTGqM_G^J9h*l{)h}odE_eZNK`nN^hWQv`EsYSDf-5$|zlm*i41<&W3=W!F@KoK0rE@jD zC#>8+7>2Sbe(AC;Mu!|FoB4{JbIG%{?be5(+)~=IRcM5( zqGs6Lsa9aw^bGolMQhx@u->T$U{3>bg zw&6DwN(LV`$aqeaXN3y!HUbH+`x@|HSV8J1tV|IV=B;`j zACE53Y_AV@CdYHP1n-@9bY43LZ~4^$homW-)};?JU4k9Ywo5U>gSyLMVPGJV635%9 zCKMX}?9bHLd>cG-kE%q1ip&t$*_cpFREEGUgG{O&C?1eN-1P4ZHe6o?5;5rw<6~dR z&sD*yTt*xg+bGM_TefP0PdU;=#P90O^SN;8^RV}N{4%!U~v$aw;RIwKpt7*Ejg=CknFv^k}_hiiH`d7g~Z)EUXs;EEbcfGAb|h$`H0 zQte~92pwvbkNp7bPsdqr!+Ve}M+JpHuyTy1XrljpCgV*^or-$!{$Fdi?I#*p{Z(u5 zQU0F#kcxdy;WwhIr~P*cp#7BOySCm=KqB6%9NRd>2YdO=*x9R&7=+`RKeo{e_6oF(RbZ5sjb#! zRK6v&;s3lZ&@m;&%P<tT5&H9V!^gQKbCy#5Z+>I@oQ2yB(xsguvf$WgVgN|>~CFj#8rYgdh4hnOS z%U}u}uA}urLACR%bE6shBH4Ci-4;+C{!2rL;<_BXVXyzI+v>LVZU;^@9R$_J+kFyb-i>^ykk_xH ziU6hmKkc1qIGgDj$2&TqQcJf~scqURCfZtBr7D)TXw}vd6hW&!h`mSUr?6QZkTc>=f+=1pa=XdZ2ts!%g%9lPB&-(bRXI&2PX1irLu(uP+ zcl7%f76?rB0cBKBJ7S}#_r&pSH=n2CVQ@%3bP!P2XDgf{{JW$b7VCFe-E|~+B~Q@P zlBWgr`cgy0v951hOg7UiTun%QT@f?M`#Le*S&@T}{Jp^wHy9;+RmGssJ* zB7>%6oktEVSLDQ(c(v=i$<__)s@-WJ0NbZnywBE{3iD+GX-6OLrkiymR3)7JY< zJpu8OhmQ?E6k&fx&|P|*SA$AelUyDs!~Gzgz(uUindrY?r8s5o%t3}RLtf^bI_Uc} zuXbetYOTrj5zehsf6I6IMpZ{%1SmE&*j7uC?^XTbygsA@mx^%1KUrc@Fyzbeo+&Gd z{$_Ggs;>n_!w(b@@q2RwX|2ZN7bK4xo;c@c&}f; z#@_(oN+vn@;$GlMO|*AJXjUEbd7`ZYt{3Mu-uf67w4i|UO(Z)?P^LoBwV8QckwcbN z)!A)x-*;-G00LEU^cNmS!tLkUS;}A@a!kS-=rB&t&0AL94bG9TfI{7V`^+S*%_t$F z8D1g=Xx+TteodS~e{skKK;LU^tvy( zij0xerzfgJa7WK2P2i!nK_SuiEx^#093%Yr=#%%Boa^E<_v&%xi=(k)-Tm%7@sGFG z9&~ap?x`3EA(VK8SLK{nGwZUYRpIApFUi;19c^D_3|Es^-cKfwy#e!X<~k_Ikm#n? z9|oZc*j05}R#*!!tH+)`!|dj!a~{uqR!e-7uPY|k(#DChS8@E*5uk;%%N$0ymNT+; z!Wpum?niCCWl3=sOR^kRf(9E>Ev2#^t-E{{{~M&Iv?4Et?Axm%ZOK3*N{0=7`aN(f zI)Sq>_ax5oIQ|UlRCPT}E~s>svo5u515<$e0zlg56)+aZ zm(X;E(L_zSaTx2T^!Cjj>{=RXCaGx}Q85w|Cw3Z65Zq-@*uY^*Y#4!9rS6_%L^V zc5x`}5oFU%ke(8edm$mO&fh@Y`i z@T<^IHc4aXAjV{EZtzwls3b|#`wEL_H%c*iK+gful?ohJDBDqTV1Ys(*bEvWYb?|W zuBCC8?_gW}8ZtBQ&-Zayk)K-63+WuQI}uXbxW`}yNH;cD8OwOTT?8zV2pe-h zr93g^(4D^pK`R8ia0bxm+enUuO(aq@)zOo>Hcudv0$NY@&C3Q^tESKP-NST0rQ=Y> zIby4~Z>Hq;7PF?8+N+$_$COe1{tfjyMnx3xPc*a2Rz0A$q}z>6naQWm%s|qW&)Rl` z$jGfp9)DtydD>|_)6LG+pCA*t=5KR2H{BU66A-ri!m+H!*QsMb2{Y!;AtN z&;it8*MqEOvuCFD?oN}yQ2kZz_ z`Y>mVfMZ`C00gB+MkJ+vQ3OAE=$7zEda)H45}3htrf7n~yDFa5z%0dg$we&Q;zZyw zU}`-oI%)dq<4JFIwR5hqZmpXYSW%M+54-#W*F)6|>Lip@Spbfvw`aLkSX(L{ODGmk z*-bA4+}La%ipQTJ@S;c6AR43Gq~@#&hUun^c%R9wbv>+jLps)@42{faJnG{R+w{D_ zT%^Jvq6_rP`W7$+Z%uO7EUD;%!Vj|~a#Em~Ay+@UW7e9~mO8S{NcJUwE=18Dm^muh z%@*$x26e7t<*-NNAYpCE%g7&)}VbTuAwu>)^LQDvU0unp4A3kIlV#zT`KlZty)KK{Q3-Se3>v0}h zTR`hvxe2B%Ij>$!Znp7x_IN{CjJR$Ou8)YEer9OC98b4~K}0ZNnKSNf02A zC#D>iz^Sep^c-FV&!F^b6i1h~AG06%6*)61jO&)6?}F(uGv4aV}A)Wc)xFE1p)!a)qa+7;+o6b2x4 z(f;O9zV?nG&~Y0tE&chH<>R74ogYnJ!^ek-eUgl21a^|0*(M^5DNe(dZ*~G;Bkz5B~cqN`Qx_cqRMNx=GX-8lc=fgjGvy1AC zGxAHHC+K_tXIxVKpn(ZawOarVxLw0sOCSs09&X|%Mj?Apeb5~RF@x+l(ZTYhW1R;L z7K7*7Ie5(6efJul);(*Bphn3`PfuD9;u^)e2+QiBsfV-g7pmStJ3TsZQqFI|o(u;p{e1zyAq?5CTG>`tQQu zh6>;*-m-TLNpnvJomP?kHJ)Afhi~uf(G-}ozDU4o+epw!w?j1{JvTczW3ux{_)h+s z1i%?K;?c1In>&k&9z{Vr3o2IyIYV`c#15i20^M`@PE?EGke=cWUr3O;qIvf~Q|=3K zK9@h4*DDp!Z%h*S0<3_MmORKPG1VQ0VF$RzFa&@tZgfhUE0&3{Yku)cyfJwB&c+@7jB=?QV*n_LT?}|JB*SOXLcNW)Qzx+ae&8Pz`j+c&BRmzBrhm zwN`!7qjgX8VQAs+F7zea0)ITgh9n(OIh85aXwJ8R@_yCe9hP#1WeL0a`hNaU6iY?V-8byAz^_OC=0R5Sg zMlI3&uiyV%^}gQ=~^NH literal 0 HcmV?d00001 From c988e688b0fb9ef726a0f2e679c600a5e288a147 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 14:26:10 -0700 Subject: [PATCH 21/90] 111 +ToC --- .../111-set_up_validmind.ipynb | 138 ++++++++++++++---- 1 file changed, 107 insertions(+), 31 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index 37118c750..8426035c7 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "markdown", - "id": "34e60911", + "id": "7ecc4b20", "metadata": {}, "source": [ "# ValidMind for model validation — 111 Set up the ValidMind Library\n", @@ -14,9 +14,52 @@ }, { "cell_type": "markdown", - "id": "0426ef4f", + "id": "42e1f410", + "metadata": {}, + "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [Introduction](#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", + " - [Register a sample model](#toc3_1_) \n", + " - [Assign validator credentials](#toc3_1_1_) \n", + " - [Install the ValidMind Library](#toc3_2_) \n", + " - [Initialize the ValidMind Library](#toc3_3_) \n", + " - [Get your code snippet](#toc3_3_1_) \n", + "- [Getting to know ValidMind](#toc4_) \n", + " - [Preview the validation report template](#toc4_1_) \n", + " - [View validation report in the ValidMind Platform](#toc4_2_) \n", + "- [Importing the champion model](#toc5_) \n", + " - [Load the sample dataset](#toc5_1_) \n", + " - [Preprocess the dataset](#toc5_1_1_) \n", + " - [Apply feature engineering to the dataset](#toc5_2_) \n", + "- [Upgrade ValidMind](#toc6_) \n", + "- [In summary](#toc7_) \n", + "- [Next steps](#toc8_) \n", + " - [Develop potential challenger models](#toc8_1_) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "9e4a8f6f", "metadata": {}, "source": [ + "\n", + "\n", "## Introduction\n", "\n", "Model validation aims to independently assess the compliance of *champion models* created by model developers with regulatory guidance by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance. Assessments, presented in the form of a validation report, typically include *model findings* and recommendations to address those issues.\n", @@ -29,9 +72,11 @@ }, { "cell_type": "markdown", - "id": "about-intro-e542b60e-207d-49de-a380-4a6f99f1c392", + "id": "69c913b4", "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", @@ -41,9 +86,11 @@ }, { "cell_type": "markdown", - "id": "about-begin-5260219d-b4ce-4a0b-924d-7333fe5d7be3", + "id": "d302994e", "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", @@ -53,9 +100,11 @@ }, { "cell_type": "markdown", - "id": "about-signup-9ffa73ae-a0c8-4c3d-b307-08dbb5c54194", + "id": "9e7f591b", "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", @@ -67,9 +116,11 @@ }, { "cell_type": "markdown", - "id": "about-concepts-fa466c67-c594-4e29-aaba-72e793f905b6", + "id": "2d5a9c74", "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", @@ -100,17 +151,21 @@ }, { "cell_type": "markdown", - "id": "f25ee2ed", + "id": "29616ee3", "metadata": {}, "source": [ + "\n", + "\n", "## Setting up" ] }, { "cell_type": "markdown", - "id": "8da00e8a", + "id": "15a805a5", "metadata": {}, "source": [ + "\n", + "\n", "### Register a sample model\n", "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", @@ -133,9 +188,11 @@ }, { "cell_type": "markdown", - "id": "ddbb615d", + "id": "45c6d89b", "metadata": {}, "source": [ + "\n", + "\n", "#### Assign validator credentials\n", "\n", "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", @@ -155,9 +212,10 @@ }, { "cell_type": "markdown", - "id": "install-library-2450d0dd-fd90-407c-8cd7-53ef0973ffcf", "metadata": {}, "source": [ + "\n", + "\n", "### Install the ValidMind Library\n", "\n", "
Recommended Python versions\n", @@ -170,7 +228,7 @@ { "cell_type": "code", "execution_count": null, - "id": "install-python-7a733859-9daf-4528-8128-c5d23a0fd6ef", + "id": "775b0796", "metadata": {}, "outputs": [], "source": [ @@ -179,9 +237,11 @@ }, { "cell_type": "markdown", - "id": "install-initialize-0221421a-9308-4d3d-93d3-0514d5fbe196", + "id": "2cecdcb7", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind Library\n", "\n", "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." @@ -189,9 +249,10 @@ }, { "cell_type": "markdown", - "id": "install-snippet-7b7726eb-26bd-43bd-b1ef-504895bf9bd0", "metadata": {}, "source": [ + "\n", + "\n", "#### Get your code snippet\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", @@ -206,7 +267,7 @@ { "cell_type": "code", "execution_count": null, - "id": "install-init-0c822102-b258-4404-b21e-26b8540f611a", + "id": "822f649c", "metadata": {}, "outputs": [], "source": [ @@ -229,17 +290,20 @@ }, { "cell_type": "markdown", - "id": "c67460a3", + "id": "3045cd0e", "metadata": {}, "source": [ + "\n", + "\n", "## Getting to know ValidMind" ] }, { "cell_type": "markdown", - "id": "7f4f9a30", "metadata": {}, "source": [ + "\n", + "\n", "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", @@ -250,7 +314,7 @@ { "cell_type": "code", "execution_count": null, - "id": "e63a3025", + "id": "b5a1f62c", "metadata": {}, "outputs": [], "source": [ @@ -259,9 +323,11 @@ }, { "cell_type": "markdown", - "id": "2dd3ddaa", + "id": "7b996415", "metadata": {}, "source": [ + "\n", + "\n", "### View validation report in the ValidMind Platform\n", "\n", "Next, let's head to the ValidMind Platform to see the template in action:\n", @@ -272,7 +338,7 @@ "\n", "3. Click on the **Validation Report** for your model and note:\n", "\n", - " - [ ] The risk assessment compliance summary at the top of the report\n", + " - [ ] The risk assessment compliance summary at the top of the report (screenshot below)\n", " - [ ] How the structure of the validation report reflects the previewed template\n", "\n", " \"Screenshot\n", @@ -281,9 +347,10 @@ }, { "cell_type": "markdown", - "id": "9fc515f0", "metadata": {}, "source": [ + "\n", + "\n", "## Importing the champion model\n", "\n", "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" @@ -307,7 +374,7 @@ { "cell_type": "code", "execution_count": null, - "id": "4e1640fd", + "id": "b6887d38", "metadata": {}, "outputs": [], "source": [ @@ -317,9 +384,10 @@ }, { "cell_type": "markdown", - "id": "e0a31bb7", "metadata": {}, "source": [ + "\n", + "\n", "### Load the sample dataset\n", "\n", "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle that was used to develop our dummy champion model, and use it to replicate some steps that would have been conducted during the initial development and documentation of the model.\n", @@ -330,7 +398,7 @@ { "cell_type": "code", "execution_count": null, - "id": "99ef5ead", + "id": "332843f1", "metadata": {}, "outputs": [], "source": [ @@ -342,9 +410,10 @@ }, { "cell_type": "markdown", - "id": "891bfdf6", "metadata": {}, "source": [ + "\n", + "\n", "#### Preprocess the dataset\n", "\n", "We'll first quickly preprocess the dataset for data quality testing purposes:" @@ -353,7 +422,7 @@ { "cell_type": "code", "execution_count": null, - "id": "5a67644f", + "id": "805cf3de", "metadata": {}, "outputs": [], "source": [ @@ -362,9 +431,10 @@ }, { "cell_type": "markdown", - "id": "6e5606d9", "metadata": {}, "source": [ + "\n", + "\n", "### Apply feature engineering to the dataset\n", "\n", "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", @@ -375,7 +445,7 @@ { "cell_type": "code", "execution_count": null, - "id": "d58a1eb1", + "id": "eea535a8", "metadata": {}, "outputs": [], "source": [ @@ -385,9 +455,10 @@ }, { "cell_type": "markdown", - "id": "upgrade-vm-4b09d40c-edad-4f40-8884-c0f0eb014875", "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", @@ -423,7 +494,7 @@ }, { "cell_type": "markdown", - "id": "upgrade-restart-3825c878-7301-47b2-92b3-2bab240f343b", + "id": "6e12584a", "metadata": {}, "source": [ "You may need to restart your kernel after running the upgrade package for changes to be applied." @@ -431,9 +502,11 @@ }, { "cell_type": "markdown", - "id": "3469ee6b", + "id": "4849b3cc", "metadata": {}, "source": [ + "\n", + "\n", "## In summary\n", "\n", "In this first notebook, you learned how to:\n", @@ -446,11 +519,14 @@ }, { "cell_type": "markdown", - "id": "4ebb0a11", "metadata": {}, "source": [ + "\n", + "\n", "## Next steps\n", "\n", + "\n", + "\n", "### Develop potential challenger models\n", "\n", "Now that the ValidMind Library is connected to your model in the ValidMind Library with the correct template applied, we can go ahead and start the model validation process: **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**" From 4b5f31852ac89e356fd3835ee71a0c004cbbcf59 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 14:42:13 -0700 Subject: [PATCH 22/90] 112 setup --- .../add_context_to_llm_descriptions.ipynb | 10 +- .../112-develop_challenger_models.ipynb | 404 ++++++++++-------- 2 files changed, 241 insertions(+), 173 deletions(-) diff --git a/notebooks/how_to/add_context_to_llm_descriptions.ipynb b/notebooks/how_to/add_context_to_llm_descriptions.ipynb index 6951e5285..fc66635c2 100644 --- a/notebooks/how_to/add_context_to_llm_descriptions.ipynb +++ b/notebooks/how_to/add_context_to_llm_descriptions.ipynb @@ -23,7 +23,7 @@ "- [Initialize the Python environment](#toc3_) \n", "- [Load the sample dataset](#toc4_) \n", " - [Preprocess the raw dataset](#toc4_1_) \n", - "- [Initialize the ValidMind objects](#toc5_) \n", + "- [Initializing the ValidMind objects](#toc5_) \n", " - [Initialize the datasets](#toc5_1_) \n", " - [Initialize a model object](#toc5_2_) \n", " - [Assign predictions to the datasets](#toc5_3_) \n", @@ -230,7 +230,7 @@ "source": [ "\n", "\n", - "## Initialize the ValidMind objects" + "## Initializing the ValidMind objects" ] }, { @@ -349,7 +349,7 @@ "\n", "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", "\n", - "Let's generate an initial test description for the `DatasetDescription` test for comparision with later iterations:" + "Let's generate an initial test description for the `DatasetDescription` test for comparison with later iterations:" ] }, { @@ -431,7 +431,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "With the use case context set, generate an updated test description for the `DatasetDescription` test for comparision with default output:" + "With the use case context set, generate an updated test description for the `DatasetDescription` test for comparison with default output:" ] }, { @@ -474,7 +474,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "With the use case context disabled again, generate another test description for the `DatasetDescription` test for comparision with previous custom output:" + "With the use case context disabled again, generate another test description for the `DatasetDescription` test for comparison with previous custom output:" ] }, { diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index d116b52d3..cab9d9a27 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -4,175 +4,243 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# ValidMind for model validation — 112 Develop potential challenger models" - ] - }, - { - "cell_type": "markdown", - "id": "b7347ccf", - "metadata": {}, - "source": [ - "### Enable custom context for test descriptions\n", - "\n", - "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", - "\n", - "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", - "\n", - "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", - "

\n", - "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", - "
" - ] - }, - { - "cell_type": "markdown", - "id": "57ad6e51", - "metadata": {}, - "source": [ - "#### Review default LLM-generated descriptions\n", - "\n", - "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", - "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with later iterations:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "5bb4450d", - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": fe_df,\n", - " },\n", - ")" - ] - }, - { - "cell_type": "markdown", - "id": "5612ba38", - "metadata": {}, - "source": [ - "#### Enable use case context\n", - "\n", - "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", - "\n", - "This is a global setting that will affect all tests for your linked model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b6105468", - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" - ] - }, - { - "cell_type": "markdown", - "id": "0d1e90ba", - "metadata": {}, - "source": [ - "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "87f7fb2e", - "metadata": {}, - "outputs": [], - "source": [ - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "id": "d6e38bd2", - "metadata": {}, - "source": [ - "With the use case context set, generate an updated test description for the `DatasetDescription` test for comparision with default output:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "adede511", - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": fe_df,\n", - " },\n", - ")" - ] - } + "# ValidMind for model validation — 112 Develop potential challenger models\n", + "\n", + "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this second notebook, develop challenger models through a structured comparison process.\n", + "\n", + "A *challenger model* is an alternate model that attempt to outperform the champion model, ensuring that the best performing fit-for-purpose model is always considered for deployment. Challenger models also help avoid over-reliance on a single model, and allow testing of new features, algorithms, or data sources without disrupting the production lifecycle." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Prerequisites\n", + "\n", + "In order to continue on the model validation journey with notebook, you'll need to first have:\n", + "\n", + "- [ ] Registered a model within the ValidMind Platform and granted yourself access to the model as a validator\n", + "- [ ] Installed the ValidMind Library in your local environment, allowing you to access all its features\n", + "\n", + "
Need help with the above steps?\n", + "

\n", + "Refer to the first notebook in this series: 111 Set up ValidMind
\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Split the feature engineered dataset" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Investigating potential challenger models" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Initializing the ValidMind objects" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# WIP BELOW**" + ] + }, + { + "cell_type": "markdown", + "id": "b7347ccf", + "metadata": {}, + "source": [ + "### Enable custom context for test descriptions\n", + "\n", + "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", + "\n", + "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", + "\n", + "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", + "

\n", + "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "57ad6e51", + "metadata": {}, + "source": [ + "#### Review default LLM-generated descriptions\n", + "\n", + "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", + "\n", + "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with later iterations:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5bb4450d", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": fe_df,\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "5612ba38", + "metadata": {}, + "source": [ + "#### Enable use case context\n", + "\n", + "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", + "\n", + "This is a global setting that will affect all tests for your linked model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b6105468", + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" + ] + }, + { + "cell_type": "markdown", + "id": "0d1e90ba", + "metadata": {}, + "source": [ + "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "87f7fb2e", + "metadata": {}, + "outputs": [], + "source": [ + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", + " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "id": "d6e38bd2", + "metadata": {}, + "source": [ + "With the use case context set, generate an updated test description for the `DatasetDescription` test for comparision with default output:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "adede511", + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": fe_df,\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## In summary" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Next steps" + ] + } ], - "metadata": { "language_info": { "name": "python" From 9cd9f70f6d76883fe0bb656a8b8ccc4fecd7e6ef Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 14:50:06 -0700 Subject: [PATCH 23/90] 112 copied info over --- .../112-develop_challenger_models.ipynb | 333 +++++++++++++++++- 1 file changed, 330 insertions(+), 3 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index cab9d9a27..f33e832d6 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -38,21 +38,348 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Split the feature engineered dataset" + "## Split the feature engineered dataset*" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Investigating potential challenger models" + "In this section, we will to a train and split randomly as the Validator want's to independently challenge the developer \n", + "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`). \n", + "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Split the data\n", + "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", + "\n", + "x_train = train_df.drop(lending_club.target_column, axis=1)\n", + "y_train = train_df[lending_club.target_column]\n", + "\n", + "x_test = test_df.drop(lending_club.target_column, axis=1)\n", + "y_test = test_df[lending_club.target_column]\n", + "\n", + "#now let's apply the order of features from the champion model construction\n", + "x_train = x_train[cols_when_model_builds]\n", + "x_test = x_test[cols_when_model_builds]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cols_use = ['annual_inc_woe',\n", + " 'verification_status_woe',\n", + " 'emp_length_woe',\n", + " 'installment_woe',\n", + " 'term_woe',\n", + " 'home_ownership_woe',\n", + " 'purpose_woe',\n", + " 'open_acc_woe',\n", + " 'total_acc_woe',\n", + " 'int_rate_woe',\n", + " 'sub_grade_woe',\n", + " 'grade_woe','loan_status']\n", + "\n", + "\n", + "train_df = train_df[cols_use]\n", + "test_df = test_df[cols_use]\n", + "test_df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Initializing the ValidMind objects" + "## Investigating potential challenger models*" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's train two challenger models as basis for the testing" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Define the Random Forest model\n", + "from sklearn.ensemble import RandomForestClassifier\n", + "rf_model = RandomForestClassifier(\n", + " n_estimators=50, \n", + " random_state=42,\n", + ")\n", + "\n", + "# Fit the model\n", + "rf_model.fit(x_train, y_train)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Second Challenger Model a Logistic Regression\n", + "from sklearn.linear_model import LogisticRegression\n", + "\n", + "# Logistic Regression grid params\n", + "log_reg_params = {\n", + " \"penalty\": [\"l1\", \"l2\"],\n", + " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", + " \"solver\": [\"liblinear\"],\n", + "}\n", + "\n", + "# Grid search for Logistic Regression\n", + "from sklearn.model_selection import GridSearchCV\n", + "\n", + "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", + "grid_log_reg.fit(x_train, y_train)\n", + "\n", + "# Logistic Regression best estimator\n", + "log_reg = grid_log_reg.best_estimator_\n", + "log_reg" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Compute probabilities as this is the raw probabilitistc output from the models of interest" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", + "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", + "\n", + "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", + "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", + "\n", + "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", + "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Compute binary predictions" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cut_off_threshold = 0.3 \n", + "\n", + "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", + "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", + "\n", + "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", + "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", + "\n", + "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", + "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Initializing the ValidMind objects*" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### 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.\n", + "\n", + "This function takes a number of arguments:\n", + "\n", + "- `dataset`: The 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", + "With all datasets ready, you can now initialize the raw, processed, training and test datasets (`raw_df`, `preprocessed_df`, `fe_df`, `train_df` and `test_df`) created earlier into their own dataset objects using [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_preprocess_dataset = vm.init_dataset(\n", + " dataset=preprocess_df,\n", + " input_id=\"preprocess_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_fe_dataset = vm.init_dataset(\n", + " dataset=fe_df,\n", + " input_id=\"fe_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_train_ds = vm.init_dataset(\n", + " dataset=train_df,\n", + " input_id=\"train_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "vm_test_ds = vm.init_dataset(\n", + " dataset=test_df,\n", + " input_id=\"test_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize a model object\n", + "\n", + "You will also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data. You simply intialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm_xgb_model = vm.init_model(\n", + " xgb_model,\n", + " input_id=\"xgb_model_developer_champion\",\n", + ")\n", + "\n", + "vm_rf_model = vm.init_model(\n", + " rf_model,\n", + " input_id=\"rf_model\",\n", + ")\n", + "\n", + "vm_log_model = vm.init_model(\n", + " log_reg,\n", + " input_id=\"log_model\",\n", + ")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Assign prediction values and probabilities to the datasets\n", + "\n", + "With our model now trained, we'll move on to assigning both the predictive probabilities coming directly from the model's predictions, and the binary prediction after applying the cutoff threshold described in the previous steps. \n", + "- These tasks are achieved through the use of the `assign_predictions()` method associated with the VM `dataset` object.\n", + "- This method links the model's class prediction values and probabilities to our VM train and test datasets." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# XGBoost Model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=train_xgb_binary_predictions,\n", + " prediction_probabilities=train_xgb_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=test_xgb_binary_predictions,\n", + " prediction_probabilities=test_xgb_prob,\n", + ")\n", + "\n", + "# Random Forest Model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=train_rf_binary_predictions,\n", + " prediction_probabilities=train_rf_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=test_rf_binary_predictions,\n", + " prediction_probabilities=test_rf_prob,\n", + ")\n", + "\n", + "\n", + "# Logistic Regression \n", + "vm_train_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=train_log_binary_predictions,\n", + " prediction_probabilities=train_log_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=test_log_binary_predictions,\n", + " prediction_probabilities=test_log_prob,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Compute credit risk scores\n", + "\n", + "In this phase, we translate model predictions into actionable scores using probability estimates generated by our trained model." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", + "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", + "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", + "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", + "train_log_scores = lending_club.compute_scores(train_log_prob)\n", + "test_log_scores = lending_club.compute_scores(test_log_prob)\n", + "\n", + "# Assign scores to the datasets\n", + "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", + "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", + "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", + "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", + "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", + "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" ] }, { From 1de2f5dd2ef145af109acc2631c9d0b12497f369 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 15:08:27 -0700 Subject: [PATCH 24/90] 111 more context --- .../111-set_up_validmind.ipynb | 12 ++- .../112-develop_challenger_models.ipynb | 96 ++++++++++++++++++- 2 files changed, 102 insertions(+), 6 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb index 8426035c7..178cdd8db 100644 --- a/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb +++ b/notebooks/tutorials/model_validation/111-set_up_validmind.ipynb @@ -416,7 +416,12 @@ "\n", "#### Preprocess the dataset\n", "\n", - "We'll first quickly preprocess the dataset for data quality testing purposes:" + "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", + "\n", + "- Filters the dataset to include only loans for debt consolidation or credit card purposes\n", + "- Removes loans classified under the riskier grades \"F\" and \"G\"\n", + "- Excludes uncommon home ownership types and standardizes employment length and loan terms into numerical formats\n", + "- Discards unnecessary fields and any entries with missing information to maintain a clean and robust dataset for modeling" ] }, { @@ -439,7 +444,10 @@ "\n", "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", "\n", - "We'll apply some transformations to optimize the dataset for predictive modeling in our application scorecard:" + "We'll apply the following transformations using the `ending_club.feature_engineering()` function to optimize the dataset for predictive modeling in our application scorecard:\n", + "\n", + "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", + "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." ] }, { diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index f33e832d6..3a3821e7a 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -38,15 +38,103 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Split the feature engineered dataset*" + "### Initialize the ValidMind Library\n", + "\n", + "First, let's connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "\n", + "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "\n", + "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": [ + "# Make sure the ValidMind Library is installed\n", + "\n", + "%pip install -q validmind\n", + "\n", + "# 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": [ - "In this section, we will to a train and split randomly as the Validator want's to independently challenge the developer \n", - "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`). \n", + "### Import the champion model\n", + "\n", + "Next, we'll import the champion model submitted by the model development team as we used in the last notebook (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model\n", + "\n", + "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Lending Club dataset from Kaggle\n", + "from validmind.datasets.credit_risk import lending_club\n", + "\n", + "df = lending_club.load_data(source=\"offline\")\n", + "df.head()\n", + "\n", + "# Preprocess the dataset for data quality testing purposes\n", + "preprocess_df = lending_club.preprocess(df)\n", + "\n", + "# Apply feature engineering to the dataset\n", + "fe_df = lending_club.feature_engineering(preprocess_df)\n", + "fe_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Split the feature engineered dataset\n", + "\n", + "With our dummy model imported and our feature-engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", + "\n", + "Splitting our dataset into train and test is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data.\n", + "\n", + "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." ] }, @@ -65,7 +153,7 @@ "x_test = test_df.drop(lending_club.target_column, axis=1)\n", "y_test = test_df[lending_club.target_column]\n", "\n", - "#now let's apply the order of features from the champion model construction\n", + "# Now let's apply the order of features from the champion model construction\n", "x_train = x_train[cols_when_model_builds]\n", "x_test = x_test[cols_when_model_builds]" ] From 357ff98da5ec434d4dade7c3b51d824653d29583 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 16:03:25 -0700 Subject: [PATCH 25/90] 112 Setup for challenger models --- .../112-develop_challenger_models.ipynb | 35 ++++++++++++++----- 1 file changed, 27 insertions(+), 8 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 3a3821e7a..dbb8e6ab7 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -132,7 +132,7 @@ "\n", "With our dummy model imported and our feature-engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", "\n", - "Splitting our dataset into train and test is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data.\n", + "Splitting our dataset into train and test is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", "\n", "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." @@ -187,14 +187,22 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Investigating potential challenger models*" + "## Investigating potential challenger models\n", + "\n", + "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", + "\n", + "Our selected options below offer decreased complexity in terms of implementation — such as less required manual preprocessing — which can reduce the amount of risk for implementation. However, model risk is a factor that is never calculated in isolation, so we must also consider trade-offs in predictive performance, ease of interpretability, and overall alignment with business objectives." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "Let's train two challenger models as basis for the testing" + "### 1. Random forest classification model\n", + "\n", + "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", + "\n", + "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in terms of predictions." ] }, { @@ -203,7 +211,7 @@ "metadata": {}, "outputs": [], "source": [ - "# Define the Random Forest model\n", + "# Define the random forest classification model\n", "from sklearn.ensemble import RandomForestClassifier\n", "rf_model = RandomForestClassifier(\n", " n_estimators=50, \n", @@ -214,29 +222,40 @@ "rf_model.fit(x_train, y_train)" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### 2. Logistic regression model \n", + "\n", + "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation and the logistic function to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", + "\n", + "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients, but they may struggle with capturing complex, non-linear relationships in the data." + ] + }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "#Second Challenger Model a Logistic Regression\n", + "# Define the logistic regression model\n", "from sklearn.linear_model import LogisticRegression\n", "\n", - "# Logistic Regression grid params\n", + "# Logistic regression grid params\n", "log_reg_params = {\n", " \"penalty\": [\"l1\", \"l2\"],\n", " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", " \"solver\": [\"liblinear\"],\n", "}\n", "\n", - "# Grid search for Logistic Regression\n", + "# Grid search for logistic regression\n", "from sklearn.model_selection import GridSearchCV\n", "\n", "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", "grid_log_reg.fit(x_train, y_train)\n", "\n", - "# Logistic Regression best estimator\n", + "# Logistic regression best estimator\n", "log_reg = grid_log_reg.best_estimator_\n", "log_reg" ] From 94bb706ac719ece636bbfaef99f840f9ee0c0631 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 16:24:33 -0700 Subject: [PATCH 26/90] 112 Challenger models v1 --- .../112-develop_challenger_models.ipynb | 48 ++++++++++++++----- 1 file changed, 35 insertions(+), 13 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index dbb8e6ab7..537c07a22 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -187,7 +187,14 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Investigating potential challenger models\n", + "## Investigating potential challenger models" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Train the challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", "\n", @@ -198,7 +205,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### 1. Random forest classification model\n", + "#### 1. Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", "\n", @@ -211,14 +218,16 @@ "metadata": {}, "outputs": [], "source": [ - "# Define the random forest classification model\n", + "# Import the Random Forest Classification model\n", "from sklearn.ensemble import RandomForestClassifier\n", + "\n", + "# Create the model instance with 50 decision trees\n", "rf_model = RandomForestClassifier(\n", - " n_estimators=50, \n", + " n_estimators=50,\n", " random_state=42,\n", ")\n", "\n", - "# Fit the model\n", + "# Train the model\n", "rf_model.fit(x_train, y_train)" ] }, @@ -226,7 +235,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### 2. Logistic regression model \n", + "#### 2. Logistic regression model \n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation and the logistic function to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", "\n", @@ -239,23 +248,23 @@ "metadata": {}, "outputs": [], "source": [ - "# Define the logistic regression model\n", + "# Import the Logistic Regression model\n", "from sklearn.linear_model import LogisticRegression\n", "\n", - "# Logistic regression grid params\n", + "# Logistic Regression grid params\n", "log_reg_params = {\n", " \"penalty\": [\"l1\", \"l2\"],\n", " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", " \"solver\": [\"liblinear\"],\n", "}\n", "\n", - "# Grid search for logistic regression\n", + "# Grid search for Logistic Regression\n", "from sklearn.model_selection import GridSearchCV\n", "\n", "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", "grid_log_reg.fit(x_train, y_train)\n", "\n", - "# Logistic regression best estimator\n", + "# Logistic Regression best estimator\n", "log_reg = grid_log_reg.best_estimator_\n", "log_reg" ] @@ -264,7 +273,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Compute probabilities as this is the raw probabilitistc output from the models of interest" + "### Extract predicted probabilities\n", + "\n", + "With our challenger models trained, let's extract the predicted probabilities from our three models, to see how confident each model is in its predictions:" ] }, { @@ -273,12 +284,15 @@ "metadata": {}, "outputs": [], "source": [ + "# Champion — Application scorecard model\n", "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", "\n", + "# Challenger — Random forest classification model\n", "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", "\n", + "# Challenger — Logistic regression model\n", "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" ] @@ -287,7 +301,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Compute binary predictions" + "### Compute binary predictions\n", + "\n", + "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of 0.3:\n", + "\n", + "- If the probability is greater than 0.3, the prediction becomes `1` (positive).\n", + "- Otherwise, it becomes `0` (negative)." ] }, { @@ -296,14 +315,17 @@ "metadata": {}, "outputs": [], "source": [ - "cut_off_threshold = 0.3 \n", + "cut_off_threshold = 0.3\n", "\n", + "# Champion — Application scorecard model\n", "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", "\n", + "# Challenger — Random forest classification model\n", "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", "\n", + "# Challenger — Logistic regression model\n", "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" ] From f755e5148f4d6b96acadd156b9c8865cceb819de Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 12 Mar 2025 17:00:47 -0700 Subject: [PATCH 27/90] 112 Setup for custom context --- .../112-develop_challenger_models.ipynb | 70 ++++++++++++------- 1 file changed, 46 insertions(+), 24 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 537c07a22..af190b3ce 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -132,7 +132,7 @@ "\n", "With our dummy model imported and our feature-engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", "\n", - "Splitting our dataset into train and test is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", + "Splitting our dataset into training and testing is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", "\n", "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." @@ -275,7 +275,7 @@ "source": [ "### Extract predicted probabilities\n", "\n", - "With our challenger models trained, let's extract the predicted probabilities from our three models, to see how confident each model is in its predictions:" + "With our challenger models trained, let's extract the predicted probabilities from our three models:" ] }, { @@ -334,7 +334,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Initializing the ValidMind objects*" + "## Initializing the ValidMind objects" ] }, { @@ -343,15 +343,13 @@ "source": [ "### 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.\n", + "Before you can run tests, you'll need 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", - "This function takes a number of arguments:\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", - "- `dataset`: The 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", - "With all datasets ready, you can now initialize the raw, processed, training and test datasets (`raw_df`, `preprocessed_df`, `fe_df`, `train_df` and `test_df`) created earlier into their own dataset objects using [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset):" + "- **`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." ] }, { @@ -360,30 +358,35 @@ "metadata": {}, "outputs": [], "source": [ + "# Initialize the raw dataset\n", "vm_raw_dataset = vm.init_dataset(\n", " dataset=df,\n", " input_id=\"raw_dataset\",\n", " target_column=lending_club.target_column,\n", ")\n", "\n", + "# Initialize the preprocessed dataset\n", "vm_preprocess_dataset = vm.init_dataset(\n", " dataset=preprocess_df,\n", " input_id=\"preprocess_dataset\",\n", " target_column=lending_club.target_column,\n", ")\n", "\n", + "# Initialize the feature engineered dataset\n", "vm_fe_dataset = vm.init_dataset(\n", " dataset=fe_df,\n", " input_id=\"fe_dataset\",\n", " target_column=lending_club.target_column,\n", ")\n", "\n", + "# Initialize the training dataset\n", "vm_train_ds = vm.init_dataset(\n", " dataset=train_df,\n", " input_id=\"train_dataset\",\n", " target_column=lending_club.target_column,\n", ")\n", "\n", + "# Initialize the test dataset\n", "vm_test_ds = vm.init_dataset(\n", " dataset=test_df,\n", " input_id=\"test_dataset\",\n", @@ -395,9 +398,18 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Initialize a model object\n", + "After initialization, you can pass the ValidMind `Dataset` objects `vm_raw_dataset`, `vm_preprocess_dataset`, `vm_fe_dataset`, `vm_train_ds`, and `vm_test_ds` into any ValidMind tests." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the model objects\n", + "\n", + "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", "\n", - "You will also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data. You simply intialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" + "You simply initialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" ] }, { @@ -406,31 +418,35 @@ "metadata": {}, "outputs": [], "source": [ + "# Initialize the champion application scorecard model\n", "vm_xgb_model = vm.init_model(\n", " xgb_model,\n", " input_id=\"xgb_model_developer_champion\",\n", ")\n", "\n", + "# Initialize the challenger random forest classification model\n", "vm_rf_model = vm.init_model(\n", " rf_model,\n", " input_id=\"rf_model\",\n", ")\n", "\n", + "# Initialize the challenger logistic regression model\n", "vm_log_model = vm.init_model(\n", " log_reg,\n", " input_id=\"log_model\",\n", - ")\n" + ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### Assign prediction values and probabilities to the datasets\n", + "## Assign predictions\n", + "\n", + "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", "\n", - "With our model now trained, we'll move on to assigning both the predictive probabilities coming directly from the model's predictions, and the binary prediction after applying the cutoff threshold described in the previous steps. \n", - "- These tasks are achieved through the use of the `assign_predictions()` method associated with the VM `dataset` object.\n", - "- This method links the model's class prediction values and probabilities to our VM train and test datasets." + "- The [`assign_predictions()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset.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." ] }, { @@ -439,7 +455,7 @@ "metadata": {}, "outputs": [], "source": [ - "# XGBoost Model\n", + "# Champion — Application scorecard model\n", "vm_train_ds.assign_predictions(\n", " model=vm_xgb_model,\n", " prediction_values=train_xgb_binary_predictions,\n", @@ -452,7 +468,7 @@ " prediction_probabilities=test_xgb_prob,\n", ")\n", "\n", - "# Random Forest Model\n", + "# Challenger — Random forest classification model\n", "vm_train_ds.assign_predictions(\n", " model=vm_rf_model,\n", " prediction_values=train_rf_binary_predictions,\n", @@ -466,7 +482,7 @@ ")\n", "\n", "\n", - "# Logistic Regression \n", + "# Challenger — Logistic regression model\n", "vm_train_ds.assign_predictions(\n", " model=vm_log_model,\n", " prediction_values=train_log_binary_predictions,\n", @@ -486,7 +502,7 @@ "source": [ "### Compute credit risk scores\n", "\n", - "In this phase, we translate model predictions into actionable scores using probability estimates generated by our trained model." + "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" ] }, { @@ -495,6 +511,7 @@ "metadata": {}, "outputs": [], "source": [ + "# Compute the scores\n", "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", @@ -518,13 +535,18 @@ "# WIP BELOW**" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Enabling custom context for test descriptions" + ] + }, { "cell_type": "markdown", "id": "b7347ccf", "metadata": {}, "source": [ - "### Enable custom context for test descriptions\n", - "\n", "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", "\n", "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", @@ -544,7 +566,7 @@ "\n", "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with later iterations:" + "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with a later iteration to demonstrate:" ] }, { From a8a5dc83e85e1a169e0e5079934f2ef23599146b Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 07:52:20 -0700 Subject: [PATCH 28/90] Changed dataset for custom context --- .../112-develop_challenger_models.ipynb | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index af190b3ce..5697954a3 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -528,13 +528,6 @@ "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# WIP BELOW**" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -566,7 +559,7 @@ "\n", "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test for comparison with a later iteration to demonstrate:" + "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate:" ] }, { @@ -579,7 +572,7 @@ "vm.tests.run_test(\n", " \"validmind.data_validation.HighPearsonCorrelation\",\n", " inputs={\n", - " \"dataset\": fe_df,\n", + " \"dataset\": vm_test_ds,\n", " },\n", ")" ] @@ -699,7 +692,7 @@ "vm.tests.run_test(\n", " \"validmind.data_validation.HighPearsonCorrelation\",\n", " inputs={\n", - " \"dataset\": fe_df,\n", + " \"dataset\": vm_test_ds,\n", " },\n", ")" ] From 4fb7c24ab427572d58e8c580fcebe64a568d646b Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 08:05:53 -0700 Subject: [PATCH 29/90] 113 setup --- .../112-develop_challenger_models.ipynb | 17 ++++++++++++++--- .../113-identify_validation_tests.ipynb | 18 ++++++++++++++++++ 2 files changed, 32 insertions(+), 3 deletions(-) create mode 100644 notebooks/tutorials/model_validation/113-identify_validation_tests.ipynb diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 5697954a3..586943fd7 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -559,7 +559,7 @@ "\n", "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate:" + "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate. We'll go over running tests in more detail in a subsequent notebook:" ] }, { @@ -701,14 +701,25 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## In summary" + "## In summary\n", + "\n", + "In this second notebook, you learned how to:\n", + "\n", + "- [ ] Initialize ValidMind datasets\n", + "- [ ] Initialize ValidMind model objects\n", + "- [ ] Assign predictions and probabilities to your ValidMind datasets\n", + "- [ ] Enable custom context for test descriptions generated by ValidMind" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Next steps" + "## Next steps\n", + "\n", + "### Identify model validation tests\n", + "\n", + "Now that you're familiar with the basics of using the ValidMind Library, let's learn how to identify and run validation tests: **[113 Identify model validation tests](113-identify_validation_tests.ipynb)**" ] } ], diff --git a/notebooks/tutorials/model_validation/113-identify_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-identify_validation_tests.ipynb new file mode 100644 index 000000000..5980e7bf6 --- /dev/null +++ b/notebooks/tutorials/model_validation/113-identify_validation_tests.ipynb @@ -0,0 +1,18 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# ValidMind for model validation — 113 Identify model validation tests" + ] + } + ], + "metadata": { + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} From fcab3f42a61902b8dfc65a7a8104ce63e4375b6c Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 08:07:05 -0700 Subject: [PATCH 30/90] 112 ToC --- .../112-develop_challenger_models.ipynb | 93 +++++++++++++++++-- 1 file changed, 86 insertions(+), 7 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 586943fd7..cf5bf1c40 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -15,6 +15,48 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [Prerequisites](#toc1_) \n", + "- [Setting up](#toc2_) \n", + " - [Initialize the ValidMind Library](#toc2_1_) \n", + " - [Import the champion model](#toc2_2_) \n", + "- [Split the feature engineered dataset](#toc3_) \n", + "- [Investigating potential challenger models](#toc4_) \n", + " - [Train the challenger models](#toc4_1_) \n", + " - [Random forest classification model](#toc4_1_1_) \n", + " - [Logistic regression model](#toc4_1_2_) \n", + " - [Extract predicted probabilities](#toc4_2_) \n", + " - [Compute binary predictions](#toc4_3_) \n", + "- [Initializing the ValidMind objects](#toc5_) \n", + " - [Initialize the ValidMind datasets](#toc5_1_) \n", + " - [Initialize the model objects](#toc5_2_) \n", + "- [Assign predictions](#toc6_) \n", + " - [Compute credit risk scores](#toc6_1_) \n", + "- [Enabling custom context for test descriptions](#toc7_) \n", + " - [Review default LLM-generated descriptions](#toc7_1_1_) \n", + " - [Enable use case context](#toc7_1_2_) \n", + "- [In summary](#toc8_) \n", + "- [Next steps](#toc9_) \n", + " - [Identify model validation tests](#toc9_1_) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", "## Prerequisites\n", "\n", "In order to continue on the model validation journey with notebook, you'll need to first have:\n", @@ -31,6 +73,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Setting up" ] }, @@ -38,6 +82,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind Library\n", "\n", "First, let's connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\n", @@ -82,6 +128,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Import the champion model\n", "\n", "Next, we'll import the champion model submitted by the model development team as we used in the last notebook (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" @@ -128,6 +176,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Split the feature engineered dataset\n", "\n", "With our dummy model imported and our feature-engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", @@ -187,6 +237,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Investigating potential challenger models" ] }, @@ -194,6 +246,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Train the challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", @@ -205,7 +259,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "#### 1. Random forest classification model\n", + "\n", + "\n", + "#### Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", "\n", @@ -235,7 +291,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "#### 2. Logistic regression model \n", + "\n", + "\n", + "#### Logistic regression model\n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation and the logistic function to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", "\n", @@ -273,6 +331,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models:" @@ -301,6 +361,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Compute binary predictions\n", "\n", "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of 0.3:\n", @@ -334,6 +396,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Initializing the ValidMind objects" ] }, @@ -341,6 +405,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind datasets\n", "\n", "Before you can run tests, you'll need 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", @@ -405,6 +471,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the model objects\n", "\n", "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", @@ -441,6 +509,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", @@ -500,6 +570,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Compute credit risk scores\n", "\n", "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" @@ -532,12 +604,13 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Enabling custom context for test descriptions" ] }, { "cell_type": "markdown", - "id": "b7347ccf", "metadata": {}, "source": [ "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", @@ -552,9 +625,10 @@ }, { "cell_type": "markdown", - "id": "57ad6e51", "metadata": {}, "source": [ + "\n", + "\n", "#### Review default LLM-generated descriptions\n", "\n", "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", @@ -565,7 +639,6 @@ { "cell_type": "code", "execution_count": null, - "id": "5bb4450d", "metadata": {}, "outputs": [], "source": [ @@ -579,9 +652,10 @@ }, { "cell_type": "markdown", - "id": "5612ba38", "metadata": {}, "source": [ + "\n", + "\n", "#### Enable use case context\n", "\n", "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", @@ -685,7 +759,6 @@ { "cell_type": "code", "execution_count": null, - "id": "adede511", "metadata": {}, "outputs": [], "source": [ @@ -701,6 +774,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## In summary\n", "\n", "In this second notebook, you learned how to:\n", @@ -715,8 +790,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Next steps\n", "\n", + "\n", + "\n", "### Identify model validation tests\n", "\n", "Now that you're familiar with the basics of using the ValidMind Library, let's learn how to identify and run validation tests: **[113 Identify model validation tests](113-identify_validation_tests.ipynb)**" From 64cb7f36a302c1e7537d3018c4e949dc977e8a66 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 08:40:04 -0700 Subject: [PATCH 31/90] 113 title adjust --- .../model_validation/112-develop_challenger_models.ipynb | 6 +++--- ...ation_tests.ipynb => 113-perform_validation_tests.ipynb} | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) rename notebooks/tutorials/model_validation/{113-identify_validation_tests.ipynb => 113-perform_validation_tests.ipynb} (71%) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index cf5bf1c40..0faf74cb1 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -38,7 +38,7 @@ " - [Enable use case context](#toc7_1_2_) \n", "- [In summary](#toc8_) \n", "- [Next steps](#toc9_) \n", - " - [Identify model validation tests](#toc9_1_) \n", + " - [Perform model validation tests](#toc9_1_) \n", "\n", ":::\n", "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", "## Prerequisites\n", "\n", "In order to log tests results or evidence to your validation report with this notebook, you'll need to first have:\n", @@ -49,6 +88,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Setting up\n", "\n", "This section should be quite familiar to you — as we performed the same actions in the previous notebook, **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**." @@ -58,6 +99,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind Library\n", "\n", "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\n", @@ -102,6 +145,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Import the champion model\n", "\n", "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" @@ -200,6 +245,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Train challenger models\n", "\n", "Just like in the last notebook, we'll train two challenger models as the basis for our testing:" @@ -255,6 +302,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models and convert the probability predictions into a binary:" @@ -305,6 +354,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind objects\n", "\n", "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" @@ -381,6 +432,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" @@ -467,6 +520,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Enable use case context\n", "\n", "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned in preparation for running our tests:" @@ -544,6 +599,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run data quality tests\n", "\n", "With everything ready to go, let's explore some of ValidMind's available tests." @@ -553,6 +610,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Identify qualitative tests\n", "\n", "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." @@ -618,6 +677,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Log data quality tests\n", "\n", "We'll store the identified tests in a `dq` variable, in preparation for batch running these tests and logging their results to the ValidMind Platform:" @@ -671,6 +732,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run comparison tests\n", "\n", "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" @@ -695,6 +758,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run performance tests\n", "\n", "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", @@ -740,6 +805,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Evaluate performance of the champion model\n", "\n", "Now, let's run and log our batch of model performance tests using our testing (`vm_test_ds`) dataset for our champion model:" @@ -764,6 +831,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Evaluate performance of challenger models\n", "\n", "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", @@ -869,7 +938,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We'll challenge the actual model itself and add the two additional benchmark models with the same `vm_test_ds` dataset:" + "We'll run each test once for each model with the same `vm_test_ds` dataset to compare them:" ] }, { @@ -882,7 +951,7 @@ " vm.tests.run_test(\n", " test,\n", " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model], \n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", " },\n", " ).log()" ] @@ -891,22 +960,36 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "
Based on the performance metrics, we can conclude that the random forest classification model is not a viable candidate for our use case and can be disregarded in our tests going forward.\n", + "

\n", + "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", "## In summary\n", "\n", "In this third notebook, you learned how to:\n", "\n", - "- [ ] text\n", - "- [ ] text\n", - "- [ ] text\n", - "- [ ] text" + "- [ ] Identify and run different types of validation tests\n", + "- [ ] Log test results as evidence to the ValidMind Platform\n", + "- [ ] Use tests from ValidMind to evaluate the potential of challenger models" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Next steps\n", "\n", + "\n", + "\n", "### Finalize testing and reporting\n", "\n", "Now that you're familiar with the basics of using the ValidMind Library to run and log validation tests, let's learn how to implement some custom tests and wrap up our validation: **[104 Finalize testing and reporting](114-finalize_testing_reporting.ipynb)**" From 071ce0d6c8c3eb30caa2109d5fcca7a0924372f4 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 14:37:16 -0700 Subject: [PATCH 45/90] 114 prereq --- .../113-perform_validation_tests.ipynb | 2 +- .../114-finalize_testing_reporting.ipynb | 37 ++++++++++++++++++- 2 files changed, 36 insertions(+), 3 deletions(-) diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index a7edf4790..e30b4e30d 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -6,7 +6,7 @@ "source": [ "# ValidMind for model validation — 113 Perform model validation tests\n", "\n", - "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this third notebook, you'll identify relevant validation tests, run them, then add the test results to your validation report.\n", + "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this third notebook, you'll identify relevant validation tests, run them, then add the log results to your validation report.\n", "\n", "You'll become familiar with the tests available in ValidMind, as well as how to run them. Running tests during model validation is crucial to the effective challenge process, as we want to independently evaluate the evidence and assessments provided by the model development team.\n", "\n", diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 889d0d459..8043ad4ef 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -4,14 +4,47 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# ValidMind for model validation — 114 Finalize testing and reporting" + "# ValidMind for model validation — 114 Finalize testing and reporting\n", + "\n", + "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this last notebook, you'll configure and run some custom tests, then add test results and findings to your validation report.\n", + "\n", + "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to put together our validation report supplemented by our logged test results as evidence and findings." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Prerequisites\n", + "\n", + "In order to finalize the validation testing and reporting for your sample model, you'll need to first have:\n", + "\n", + "- [ ] Registered a model within the ValidMind Platform and granted yourself access to the model as a validator\n", + "- [ ] Installed the ValidMind Library in your local environment, allowing you to access all its features\n", + "- [ ] Learned how to import and initialize datasets for use with ValidMind\n", + "- [ ] Learned how to enable custom context for test descriptions generated by ValidMind\n", + "- [ ] Understood the basics of how to identify and run validation tests\n", + "- [ ] Run data quality and model performance tests for your champion and challenger models, and logged the results of those tests to the ValidMind Platform\n", + "\n", + "\n", + "
Need help with the above steps?\n", + "

\n", + "Refer to the first three notebooks in this series:\n", + "\n", + "
    \n", + "
  1. 111 Import the champion model
    2. \n", + "
  2. 112 Develop potential challenger models
    3. \n", + "
  3. 113 Perform validation tests
    4. \n", + "
\n", + "\n", + "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "## Prerequisites" + "## Setting up" ] }, { From ffb887b9bb88d1f1e4b81c568a4497c8b9a7c8a5 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 14:50:41 -0700 Subject: [PATCH 46/90] 114 transfered rest of code over --- .../113-perform_validation_tests.ipynb | 1 - .../114-finalize_testing_reporting.ipynb | 821 +++++++++++++++++- 2 files changed, 819 insertions(+), 3 deletions(-) diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index e30b4e30d..798a1149c 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -471,7 +471,6 @@ " prediction_probabilities=test_rf_prob,\n", ")\n", "\n", - "\n", "# Challenger — Logistic regression model\n", "vm_train_ds.assign_predictions(\n", " model=vm_log_model,\n", diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 8043ad4ef..cf0143941 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -8,7 +8,7 @@ "\n", "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this last notebook, you'll configure and run some custom tests, then add test results and findings to your validation report.\n", "\n", - "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to put together our validation report supplemented by our logged test results as evidence and findings." + "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case and was eliminated as a contender. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to put together our validation report supplemented by our logged test results as evidence and findings." ] }, { @@ -44,7 +44,824 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Setting up" + "## Setting up\n", + "\n", + "This section should be very familiar to you now — as we performed the same actions in the previous two notebooks in this series." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the ValidMind Library\n", + "\n", + "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "\n", + "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "\n", + "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": { + "vscode": { + "languageId": "plaintext" + } + }, + "outputs": [], + "source": [ + "# Make sure the ValidMind Library is installed\n", + "\n", + "%pip install -q validmind\n", + "\n", + "# 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": [ + "### Import the champion model\n", + "\n", + "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model\n", + "\n", + "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Lending Club dataset from Kaggle\n", + "from validmind.datasets.credit_risk import lending_club\n", + "\n", + "df = lending_club.load_data(source=\"offline\")\n", + "df.head()\n", + "\n", + "# Preprocess the dataset for data quality testing purposes\n", + "preprocess_df = lending_club.preprocess(df)\n", + "\n", + "# Apply feature engineering to the dataset\n", + "fe_df = lending_club.feature_engineering(preprocess_df)\n", + "fe_df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Split our dataset into train and test to start the validation testing process\n", + "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", + "\n", + "x_train = train_df.drop(lending_club.target_column, axis=1)\n", + "y_train = train_df[lending_club.target_column]\n", + "\n", + "x_test = test_df.drop(lending_club.target_column, axis=1)\n", + "y_test = test_df[lending_club.target_column]\n", + "\n", + "# Now let's apply the order of features from the champion model construction\n", + "x_train = x_train[cols_when_model_builds]\n", + "x_test = x_test[cols_when_model_builds]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cols_use = ['annual_inc_woe',\n", + " 'verification_status_woe',\n", + " 'emp_length_woe',\n", + " 'installment_woe',\n", + " 'term_woe',\n", + " 'home_ownership_woe',\n", + " 'purpose_woe',\n", + " 'open_acc_woe',\n", + " 'total_acc_woe',\n", + " 'int_rate_woe',\n", + " 'sub_grade_woe',\n", + " 'grade_woe','loan_status']\n", + "\n", + "\n", + "train_df = train_df[cols_use]\n", + "test_df = test_df[cols_use]\n", + "test_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Train the challenger model\n", + "\n", + "As we eliminated the random forest classification model as a challenger, we'll only train our logistic regression model here:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Logistic Regression model\n", + "from sklearn.linear_model import LogisticRegression\n", + "\n", + "# Logistic Regression grid params\n", + "log_reg_params = {\n", + " \"penalty\": [\"l1\", \"l2\"],\n", + " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", + " \"solver\": [\"liblinear\"],\n", + "}\n", + "\n", + "# Grid search for Logistic Regression\n", + "from sklearn.model_selection import GridSearchCV\n", + "\n", + "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", + "grid_log_reg.fit(x_train, y_train)\n", + "\n", + "# Logistic Regression best estimator\n", + "log_reg = grid_log_reg.best_estimator_\n", + "log_reg" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Extract predicted probabilities\n", + "\n", + "With our challenger model trained, let's extract the predicted probabilities from our two models and convert the probability predictions into a binary:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Champion — Application scorecard model\n", + "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", + "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", + "\n", + "# Challenger — Logistic regression model\n", + "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", + "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# If probability > 0.3 = 1 (positive)\n", + "cut_off_threshold = 0.3\n", + "\n", + "# Champion — Application scorecard model\n", + "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", + "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", + "\n", + "# Challenger — Logistic regression model\n", + "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", + "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the ValidMind objects\n", + "\n", + "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the raw dataset\n", + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the preprocessed dataset\n", + "vm_preprocess_dataset = vm.init_dataset(\n", + " dataset=preprocess_df,\n", + " input_id=\"preprocess_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the feature engineered dataset\n", + "vm_fe_dataset = vm.init_dataset(\n", + " dataset=fe_df,\n", + " input_id=\"fe_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the training dataset\n", + "vm_train_ds = vm.init_dataset(\n", + " dataset=train_df,\n", + " input_id=\"train_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the test dataset\n", + "vm_test_ds = vm.init_dataset(\n", + " dataset=test_df,\n", + " input_id=\"test_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the champion application scorecard model\n", + "vm_xgb_model = vm.init_model(\n", + " xgb_model,\n", + " input_id=\"xgb_model_developer_champion\",\n", + ")\n", + "\n", + "# Initialize the challenger logistic regression model\n", + "vm_log_model = vm.init_model(\n", + " log_reg,\n", + " input_id=\"log_model\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Assign predictions\n", + "\n", + "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Champion — Application scorecard model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=train_xgb_binary_predictions,\n", + " prediction_probabilities=train_xgb_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=test_xgb_binary_predictions,\n", + " prediction_probabilities=test_xgb_prob,\n", + ")\n", + "\n", + "# Challenger — Logistic regression model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=train_log_binary_predictions,\n", + " prediction_probabilities=train_log_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=test_log_binary_predictions,\n", + " prediction_probabilities=test_log_prob,\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Compute the scores\n", + "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", + "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", + "train_log_scores = lending_club.compute_scores(train_log_prob)\n", + "test_log_scores = lending_club.compute_scores(test_log_prob)\n", + "\n", + "# Assign scores to the datasets\n", + "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", + "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", + "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", + "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Enable use case context\n", + "\n", + "We'll also adjust the use case context to focus on comparison between our models for tests going forward:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "\n", + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + "\n", + " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", + " - The metrics for the champion model compared against the challenger models\n", + " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", + "\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# WIP" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Now let's dig a little bit deeper into one of the tests that allows the Validator to custoimze parameters and thresholds for performance standards - and let's diregard RF model as we have\n", + "### learned that the RF model is not a viable candidate based on the performance metrics.\n", + "result = vm.tests.run_test(\n", + " 'validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold',\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], 'params':{'min_threshold': 0.35},\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Robustness and Stability Testing Comparison Between the Two Models\n", + "list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "#Let's see if models suffer from any overfit potentials and also where there are potential sub-segments of issues\n", + "overfit_testing = ['validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression','validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression'] \n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in overfit_testing:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Now finally let's conduct robustness and stability testing of the two models:\n", + "stab_robust = ['validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression'] # 'validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression'\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in stab_robust:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# WIP2\n", + "\n", + "Let's verify the feature importance and inspect differences - different models might have more intuitive feature impacts that might lead to decisions in selection of a model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "FI = list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", + "FI" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in FI:\n", + " vm.tests.run_test(\n", + " \"\".join((test,':Champion_vs_LogisticRegression')),\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "### Let's finish off with a custom test example - scoring (customization of output to a FICO score type)\n", + "import numpy as np\n", + "import pandas as pd\n", + "import plotly.graph_objects as go\n", + "\n", + "\n", + "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", + "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", + " \"\"\"\n", + " Analyzes the relationship between score bands and odds (good:bad ratio).\n", + " Good odds = (1 - default_rate) / default_rate\n", + " \n", + " Higher scores should correspond to higher odds of being good.\n", + "\n", + " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", + "\n", + " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", + " \"\"\"\n", + " df = dataset.df\n", + " \n", + " # Create score bands\n", + " df['score_band'] = pd.cut(\n", + " df[score_column],\n", + " bins=[-np.inf] + score_bands + [np.inf],\n", + " labels=[f'<{score_bands[0]}'] + \n", + " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", + " [f'>{score_bands[-1]}']\n", + " )\n", + " \n", + " # Calculate metrics per band\n", + " results = df.groupby('score_band').agg({\n", + " dataset.target_column: ['mean', 'count']\n", + " })\n", + " \n", + " results.columns = ['Default Rate', 'Total']\n", + " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", + " results['Bad Count'] = results['Default Rate'] * results['Total']\n", + " results['Odds'] = results['Good Count'] / results['Bad Count']\n", + " \n", + " # Create visualization\n", + " fig = go.Figure()\n", + " \n", + " # Add odds bars\n", + " fig.add_trace(go.Bar(\n", + " name='Odds (Good:Bad)',\n", + " x=results.index,\n", + " y=results['Odds'],\n", + " marker_color='blue'\n", + " ))\n", + " \n", + " fig.update_layout(\n", + " title='Score-to-Odds Analysis',\n", + " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", + " showlegend=False\n", + " )\n", + " \n", + " return fig" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " param_grid={\n", + " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", + " \"score_bands\": [[500, 540, 570]],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# WIP3\n", + "\n", + "Finally we got all of the tests from the Developer that was provided as evidence, now as a final task we will verify testing being appropriately recorded" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.utils import preview_test_config\n", + "\n", + "test_config = {'validmind.data_validation.DatasetDescription:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", + " 'validmind.data_validation.DescriptiveStatistics:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", + " 'validmind.data_validation.MissingValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}},\n", + " 'validmind.data_validation.ClassImbalance:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 10}},\n", + " 'validmind.data_validation.Duplicates:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}},\n", + " 'validmind.data_validation.HighCardinality:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'num_threshold': 100,\n", + " 'percent_threshold': 0.1,\n", + " 'threshold_type': 'percent'}},\n", + " 'validmind.data_validation.Skewness:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_threshold': 1}},\n", + " 'validmind.data_validation.UniqueRows:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 1}},\n", + " 'validmind.data_validation.TooManyZeroValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_percent_threshold': 0.03}},\n", + " 'validmind.data_validation.IQROutliersTable:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'threshold': 5}},\n", + " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.MissingValues:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'min_threshold': 1}},\n", + " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", + " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'default_column': 'loan_status'}},\n", + " 'validmind.data_validation.DescriptiveStatistics:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.TabularDescriptionTables:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.ClassImbalance:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 10}},\n", + " 'validmind.data_validation.UniqueRows:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 1}},\n", + " 'validmind.data_validation.TabularNumericalHistograms:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.MutualInformation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'min_threshold': 0.01}},\n", + " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.data_validation.HighPearsonCorrelation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}},\n", + " 'validmind.data_validation.WOEBinTable': {'input_grid': {'dataset': ['preprocess_dataset']},\n", + " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", + " 'int_rate': [10, 15, 20],\n", + " 'annual_inc': [50000, 100000, 150000]}}},\n", + " 'validmind.data_validation.WOEBinPlots': {'input_grid': {'dataset': ['preprocess_dataset']},\n", + " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", + " 'int_rate': [10, 15, 20],\n", + " 'annual_inc': [50000, 100000, 150000]}}},\n", + " 'validmind.data_validation.DatasetSplit': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset']}},\n", + " 'validmind.model_validation.ModelMetadata': {'input_grid': {'model': ['xgb_model',\n", + " 'rf_model']}},\n", + " 'validmind.model_validation.sklearn.ModelParameters': {'input_grid': {'model': ['xgb_model',\n", + " 'rf_model']}},\n", + " 'validmind.model_validation.statsmodels.GINITable': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model', 'rf_model']}},\n", + " 'validmind.model_validation.sklearn.ClassifierPerformance': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model', 'rf_model']}},\n", + " 'validmind.model_validation.sklearn.TrainingTestDegradation:XGBoost': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'},\n", + " 'params': {'max_threshold': 0.1}},\n", + " 'validmind.model_validation.sklearn.TrainingTestDegradation:RandomForest': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'rf_model'},\n", + " 'params': {'max_threshold': 0.1}},\n", + " 'validmind.model_validation.sklearn.ROCCurve': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}},\n", + " 'validmind.model_validation.statsmodels.PredictionProbabilitiesHistogram': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.statsmodels.CumulativePredictionProbabilities': {'input_grid': {'model': ['xgb_model'],\n", + " 'dataset': ['train_dataset', 'test_dataset']}},\n", + " 'validmind.model_validation.sklearn.PopulationStabilityIndex': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'},\n", + " 'params': {'num_bins': 10, 'mode': 'fixed'}},\n", + " 'validmind.model_validation.sklearn.ConfusionMatrix': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.MinimumAccuracy': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.7}},\n", + " 'validmind.model_validation.sklearn.MinimumF1Score': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}},\n", + " 'validmind.model_validation.sklearn.PrecisionRecallCurve': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.CalibrationCurve': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.sklearn.ClassifierThresholdOptimization': {'inputs': {'dataset': 'train_dataset',\n", + " 'model': 'xgb_model'},\n", + " 'params': {'target_recall': 0.8}},\n", + " 'validmind.model_validation.statsmodels.ScorecardHistogram': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset']},\n", + " 'params': {'score_column': 'xgb_scores'}},\n", + " 'validmind.data_validation.ScoreBandDefaultRates': {'input_grid': {'dataset': ['train_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'score_column': 'xgb_scores', 'score_bands': [504, 537, 570]}},\n", + " 'validmind.model_validation.sklearn.ScoreProbabilityAlignment': {'input_grid': {'dataset': ['train_dataset'],\n", + " 'model': ['xgb_model']},\n", + " 'params': {'score_column': 'xgb_scores'}},\n", + " 'validmind.model_validation.sklearn.WeakspotsDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'}},\n", + " 'validmind.model_validation.sklearn.OverfitDiagnosis': {'inputs': {'model': 'xgb_model',\n", + " 'datasets': ['train_dataset', 'test_dataset']},\n", + " 'params': {'cut_off_threshold': 0.04}},\n", + " 'validmind.model_validation.sklearn.RobustnessDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': 'xgb_model'},\n", + " 'params': {'scaling_factor_std_dev_list': [0.1, 0.2, 0.3, 0.4, 0.5],\n", + " 'performance_decay_threshold': 0.05}},\n", + " 'validmind.model_validation.sklearn.PermutationFeatureImportance': {'input_grid': {'dataset': ['train_dataset',\n", + " 'test_dataset'],\n", + " 'model': ['xgb_model']}},\n", + " 'validmind.model_validation.FeaturesAUC': {'input_grid': {'model': ['xgb_model'],\n", + " 'dataset': ['train_dataset', 'test_dataset']}},\n", + " 'validmind.model_validation.sklearn.SHAPGlobalImportance': {'input_grid': {'model': ['xgb_model'],\n", + " 'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'kernel_explainer_samples': 10,\n", + " 'tree_or_linear_explainer_samples': 200}}}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for t in test_config:\n", + " print(t)\n", + " try:\n", + " # Check if test has input_grid\n", + " if 'input_grid' in test_config[t]:\n", + " # For tests with input_grid, pass the input_grid configuration\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", + " else:\n", + " # Original logic for regular inputs\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", + " except Exception as e:\n", + " print(f\"Error running test {t}: {str(e)}\")" ] }, { From 4b71f67df50a44c6d11e83b62d1f42149ed69e7a Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 14:51:37 -0700 Subject: [PATCH 47/90] 114 transfered rest of code over edit --- .../model_validation/114-finalize_testing_reporting.ipynb | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index cf0143941..7ca0d79bd 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -69,11 +69,7 @@ { "cell_type": "code", "execution_count": null, - "metadata": { - "vscode": { - "languageId": "plaintext" - } - }, + "metadata": {}, "outputs": [], "source": [ "# Make sure the ValidMind Library is installed\n", From 3a88f6be4de2263d4ab03fc247461e508b9755af Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 14:56:47 -0700 Subject: [PATCH 48/90] 114 cleanup wip --- .../114-finalize_testing_reporting.ipynb | 53 ++++++++++++++----- 1 file changed, 41 insertions(+), 12 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 7ca0d79bd..0eb606f9f 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -478,7 +478,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# WIP" + "# WIP\n", + "\n", + "Now let's dig a little bit deeper into one of the tests that allows the Validator to custoimze parameters and thresholds for performance standards \n" ] }, { @@ -487,24 +489,37 @@ "metadata": {}, "outputs": [], "source": [ - "### Now let's dig a little bit deeper into one of the tests that allows the Validator to custoimze parameters and thresholds for performance standards - and let's diregard RF model as we have\n", - "### learned that the RF model is not a viable candidate based on the performance metrics.\n", "result = vm.tests.run_test(\n", - " 'validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold',\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold\",\n", " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], 'params':{'min_threshold': 0.35},\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_xgb_model, vm_log_model],\n", + " \"params\": {\"min_threshold\": 0.35},\n", " },\n", ").log()" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Robustness and Stability Testing Comparison Between the Two Models" + ] + }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "### Robustness and Stability Testing Comparison Between the Two Models\n", - "list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" + "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's see if models suffer from any overfit potentials and also where there are potential sub-segments of issues" ] }, { @@ -513,8 +528,10 @@ "metadata": {}, "outputs": [], "source": [ - "#Let's see if models suffer from any overfit potentials and also where there are potential sub-segments of issues\n", - "overfit_testing = ['validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression','validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression'] \n" + "overfit_testing = [\n", + " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\",\n", + "]" ] }, { @@ -532,14 +549,20 @@ " ).log()" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now finally let's conduct robustness and stability testing of the two models:" + ] + }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "### Now finally let's conduct robustness and stability testing of the two models:\n", - "stab_robust = ['validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression'] # 'validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression'\n" + "stab_robust = ['validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression']" ] }, { @@ -591,13 +614,19 @@ " ).log()" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's finish off with a custom test example - scoring (customization of output to a FICO score type)" + ] + }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "### Let's finish off with a custom test example - scoring (customization of output to a FICO score type)\n", "import numpy as np\n", "import pandas as pd\n", "import plotly.graph_objects as go\n", From 32542acc2217c884a16a31e31f04c3a5b3c70986 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 15:03:52 -0700 Subject: [PATCH 49/90] 114 cleanup wip2 --- .../114-finalize_testing_reporting.ipynb | 48 +++++++++++++++---- 1 file changed, 39 insertions(+), 9 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 0eb606f9f..de3659ee3 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -562,7 +562,10 @@ "metadata": {}, "outputs": [], "source": [ - "stab_robust = ['validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression']" + "stab_robust = [\n", + " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\",\n", + " ]" ] }, { @@ -889,6 +892,13 @@ " print(f\"Error running test {t}: {str(e)}\")" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Add test results to reporting" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -900,14 +910,34 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Next steps\n", - "\n", - "### Work with your validation report\n", - "\n", - "### Learn more\n", - "\n", - "#### More how-to guides and code samples\n", - "\n", + "## Next steps" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Work with your validation report" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Learn more" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### More how-to guides and code samples" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ "#### Discover more learning resources" ] } From ce256f9773cb116051ed0e65731f31675dcc0bdb Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 15:16:26 -0700 Subject: [PATCH 50/90] 114 cleanup wip3 --- .../114-finalize_testing_reporting.ipynb | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index de3659ee3..21d5f0a60 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -478,9 +478,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# WIP\n", + "## Customize a comparison test\n", "\n", - "Now let's dig a little bit deeper into one of the tests that allows the Validator to custoimze parameters and thresholds for performance standards \n" + "Let's dig deeper into the [MinimumF1Score test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html), which allows us to customize parameters and thresholds for performance standards:" ] }, { @@ -503,7 +503,16 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Robustness and Stability Testing Comparison Between the Two Models" + "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "

\n", + "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Robustness and Stability Testing Comparison Between the Two Models*" ] }, { @@ -598,7 +607,7 @@ "metadata": {}, "outputs": [], "source": [ - "FI = list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", + "FI = vm.tests.list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", "FI" ] }, From 9a690ec530df4770ea5bd08a00f4ef3d7c9e91d5 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 15:30:56 -0700 Subject: [PATCH 51/90] 114 Adjust a comparison test --- .../112-develop_challenger_models.ipynb | 4 +- .../113-perform_validation_tests.ipynb | 10 ++--- .../114-finalize_testing_reporting.ipynb | 43 ++++++++++++++++--- 3 files changed, 45 insertions(+), 12 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 46d84203b..f66a98c05 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -636,7 +636,7 @@ " \"validmind.data_validation.HighPearsonCorrelation\",\n", " inputs={\n", " \"dataset\": vm_test_ds,\n", - " },\n", + " }\n", ")" ] }, @@ -755,7 +755,7 @@ " \"validmind.data_validation.HighPearsonCorrelation\",\n", " inputs={\n", " \"dataset\": vm_test_ds,\n", - " },\n", + " }\n", ")" ] }, diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index 798a1149c..40a782c6b 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -714,7 +714,7 @@ " test,\n", " inputs={\n", " \"dataset\": vm_preprocess_dataset\n", - " },\n", + " }\n", " ).log()" ] }, @@ -749,7 +749,7 @@ " test,\n", " input_grid={\n", " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", - " },\n", + " }\n", " ).log()" ] }, @@ -796,7 +796,7 @@ " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion\",\n", " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion\",\n", " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion\"\n", "]" ] }, @@ -850,7 +850,7 @@ " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers\",\n", " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers\",\n", " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers\"\n", "]" ] }, @@ -951,7 +951,7 @@ " test,\n", " input_grid={\n", " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", - " },\n", + " }\n", " ).log()" ] }, diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 21d5f0a60..8131a7329 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -478,9 +478,33 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Customize a comparison test\n", + "## Adjust a comparison test\n", "\n", - "Let's dig deeper into the [MinimumF1Score test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html), which allows us to customize parameters and thresholds for performance standards:" + "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb).\n", + "\n", + "We'll run the test in isolation again for our two remaining models without logging the result sans adjustment to have the output to compare with a later run:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_xgb_model, vm_log_model]\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "`MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" ] }, { @@ -494,7 +518,7 @@ " input_grid={\n", " \"dataset\": [vm_test_ds],\n", " \"model\": [vm_xgb_model, vm_log_model],\n", - " \"params\": {\"min_threshold\": 0.35},\n", + " \"params\": {\"min_threshold\": 0.35}\n", " },\n", ").log()" ] @@ -524,6 +548,15 @@ "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "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
" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -539,7 +572,7 @@ "source": [ "overfit_testing = [\n", " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", "]" ] }, @@ -573,7 +606,7 @@ "source": [ "stab_robust = [\n", " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", " ]" ] }, From b51eee72cb9413b606395fcd5e49b9027dcc0d18 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 15:59:16 -0700 Subject: [PATCH 52/90] 114 Next steps draft --- .../114-finalize_testing_reporting.ipynb | 38 +++++++++++++++---- 1 file changed, 31 insertions(+), 7 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 8131a7329..d775a9360 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -6,9 +6,9 @@ "source": [ "# ValidMind for model validation — 114 Finalize testing and reporting\n", "\n", - "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this last notebook, you'll configure and run some custom tests, then add test results and findings to your validation report.\n", + "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this last notebook, you'll configure and run some custom tests, then add logged test results to your validation report.\n", "\n", - "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case and was eliminated as a contender. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to put together our validation report supplemented by our logged test results as evidence and findings." + "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case and was eliminated as a contender. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to put together our validation report supplemented by our logged test results as evidence." ] }, { @@ -945,7 +945,17 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## In summary" + "## In summary\n", + "\n", + "In this final notebook, you learned how to:\n", + "\n", + "- [ ] text\n", + "- [ ] text\n", + "- [ ] text\n", + "\n", + "With our ValidMind for model validation series of notebooks, you learned how to validate a model end-to-end with the ValidMind Library by running through some common scenarios in a typical model validation setting:\n", + "\n", + "- text" ] }, { @@ -959,28 +969,42 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Work with your validation report" + "### Work with your validation report\n", + "\n", + "Now that you've logged all your test results, head to the ValidMind Platform to wrap up your validation report. Generate drafts of risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### Learn more" + "### Learn more\n", + "\n", + "Now that you're familiar with the basics, you can explore the following notebooks to get a deeper understanding on how the ValidMind Library assists you in streamlining model validation:" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "#### More how-to guides and code samples" + "#### 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)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "#### Discover more learning resources" + "#### 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)" ] } ], From 6c6e59f6f4e8e0ac661f5873017c45858b5dfa2c Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 13 Mar 2025 16:19:35 -0700 Subject: [PATCH 53/90] 114 quick code cleanup --- .../114-finalize_testing_reporting.ipynb | 248 ++++++++---------- 1 file changed, 106 insertions(+), 142 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index d775a9360..830dfaa4f 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -8,7 +8,7 @@ "\n", "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this last notebook, you'll configure and run some custom tests, then add logged test results to your validation report.\n", "\n", - "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case and was eliminated as a contender. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to put together our validation report supplemented by our logged test results as evidence." + "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case and was eliminated as a contender. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to draft our validation report supplemented by our logged test results as evidence." ] }, { @@ -765,147 +765,111 @@ "source": [ "from validmind.utils import preview_test_config\n", "\n", - "test_config = {'validmind.data_validation.DatasetDescription:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", - " 'validmind.data_validation.DescriptiveStatistics:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", - " 'validmind.data_validation.MissingValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}},\n", - " 'validmind.data_validation.ClassImbalance:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 10}},\n", - " 'validmind.data_validation.Duplicates:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}},\n", - " 'validmind.data_validation.HighCardinality:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'num_threshold': 100,\n", - " 'percent_threshold': 0.1,\n", - " 'threshold_type': 'percent'}},\n", - " 'validmind.data_validation.Skewness:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_threshold': 1}},\n", - " 'validmind.data_validation.UniqueRows:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 1}},\n", - " 'validmind.data_validation.TooManyZeroValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_percent_threshold': 0.03}},\n", - " 'validmind.data_validation.IQROutliersTable:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'threshold': 5}},\n", - " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.MissingValues:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'min_threshold': 1}},\n", - " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'default_column': 'loan_status'}},\n", - " 'validmind.data_validation.DescriptiveStatistics:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.TabularDescriptionTables:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.ClassImbalance:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 10}},\n", - " 'validmind.data_validation.UniqueRows:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 1}},\n", - " 'validmind.data_validation.TabularNumericalHistograms:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.MutualInformation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'min_threshold': 0.01}},\n", - " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.HighPearsonCorrelation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}},\n", - " 'validmind.data_validation.WOEBinTable': {'input_grid': {'dataset': ['preprocess_dataset']},\n", - " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", - " 'int_rate': [10, 15, 20],\n", - " 'annual_inc': [50000, 100000, 150000]}}},\n", - " 'validmind.data_validation.WOEBinPlots': {'input_grid': {'dataset': ['preprocess_dataset']},\n", - " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", - " 'int_rate': [10, 15, 20],\n", - " 'annual_inc': [50000, 100000, 150000]}}},\n", - " 'validmind.data_validation.DatasetSplit': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.model_validation.ModelMetadata': {'input_grid': {'model': ['xgb_model',\n", - " 'rf_model']}},\n", - " 'validmind.model_validation.sklearn.ModelParameters': {'input_grid': {'model': ['xgb_model',\n", - " 'rf_model']}},\n", - " 'validmind.model_validation.statsmodels.GINITable': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model', 'rf_model']}},\n", - " 'validmind.model_validation.sklearn.ClassifierPerformance': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model', 'rf_model']}},\n", - " 'validmind.model_validation.sklearn.TrainingTestDegradation:XGBoost': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'},\n", - " 'params': {'max_threshold': 0.1}},\n", - " 'validmind.model_validation.sklearn.TrainingTestDegradation:RandomForest': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'rf_model'},\n", - " 'params': {'max_threshold': 0.1}},\n", - " 'validmind.model_validation.sklearn.ROCCurve': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.5}},\n", - " 'validmind.model_validation.statsmodels.PredictionProbabilitiesHistogram': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.statsmodels.CumulativePredictionProbabilities': {'input_grid': {'model': ['xgb_model'],\n", - " 'dataset': ['train_dataset', 'test_dataset']}},\n", - " 'validmind.model_validation.sklearn.PopulationStabilityIndex': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'},\n", - " 'params': {'num_bins': 10, 'mode': 'fixed'}},\n", - " 'validmind.model_validation.sklearn.ConfusionMatrix': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.MinimumAccuracy': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.7}},\n", - " 'validmind.model_validation.sklearn.MinimumF1Score': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.5}},\n", - " 'validmind.model_validation.sklearn.PrecisionRecallCurve': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.CalibrationCurve': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.ClassifierThresholdOptimization': {'inputs': {'dataset': 'train_dataset',\n", - " 'model': 'xgb_model'},\n", - " 'params': {'target_recall': 0.8}},\n", - " 'validmind.model_validation.statsmodels.ScorecardHistogram': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'score_column': 'xgb_scores'}},\n", - " 'validmind.data_validation.ScoreBandDefaultRates': {'input_grid': {'dataset': ['train_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'score_column': 'xgb_scores', 'score_bands': [504, 537, 570]}},\n", - " 'validmind.model_validation.sklearn.ScoreProbabilityAlignment': {'input_grid': {'dataset': ['train_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'score_column': 'xgb_scores'}},\n", - " 'validmind.model_validation.sklearn.WeakspotsDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'}},\n", - " 'validmind.model_validation.sklearn.OverfitDiagnosis': {'inputs': {'model': 'xgb_model',\n", - " 'datasets': ['train_dataset', 'test_dataset']},\n", - " 'params': {'cut_off_threshold': 0.04}},\n", - " 'validmind.model_validation.sklearn.RobustnessDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'},\n", - " 'params': {'scaling_factor_std_dev_list': [0.1, 0.2, 0.3, 0.4, 0.5],\n", - " 'performance_decay_threshold': 0.05}},\n", - " 'validmind.model_validation.sklearn.PermutationFeatureImportance': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.FeaturesAUC': {'input_grid': {'model': ['xgb_model'],\n", - " 'dataset': ['train_dataset', 'test_dataset']}},\n", - " 'validmind.model_validation.sklearn.SHAPGlobalImportance': {'input_grid': {'model': ['xgb_model'],\n", - " 'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'kernel_explainer_samples': 10,\n", - " 'tree_or_linear_explainer_samples': 200}}}" + "test_config = {\n", + " 'validmind.data_validation.DatasetDescription:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'}\n", + " },\n", + " 'validmind.data_validation.DescriptiveStatistics:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'}\n", + " },\n", + " 'validmind.data_validation.MissingValues:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.ClassImbalance:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 10}\n", + " },\n", + " 'validmind.data_validation.Duplicates:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.HighCardinality:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {\n", + " 'num_threshold': 100,\n", + " 'percent_threshold': 0.1,\n", + " 'threshold_type': 'percent'\n", + " }\n", + " },\n", + " 'validmind.data_validation.Skewness:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.UniqueRows:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TooManyZeroValues:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_percent_threshold': 0.03}\n", + " },\n", + " 'validmind.data_validation.IQROutliersTable:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'threshold': 5}\n", + " },\n", + " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.MissingValues:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'default_column': 'loan_status'}\n", + " },\n", + " 'validmind.data_validation.DescriptiveStatistics:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.TabularDescriptionTables:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.ClassImbalance:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 10}\n", + " },\n", + " 'validmind.data_validation.UniqueRows:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TabularNumericalHistograms:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.MutualInformation:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_threshold': 0.01}\n", + " },\n", + " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.HighPearsonCorrelation:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", + " },\n", + " 'validmind.model_validation.ModelMetadata': {\n", + " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.ModelParameters': {\n", + " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.ROCCurve': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}\n", + " }\n", + "}" ] }, { From 9f1e8d5fb8896bd3183e7f195edd5fad66e604be Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 10:00:47 -0700 Subject: [PATCH 54/90] 114 Run diagnostic tests done --- .../112-develop_challenger_models.ipynb | 6 +-- .../113-perform_validation_tests.ipynb | 25 ++++++---- .../114-finalize_testing_reporting.ipynb | 46 +++++++++++++------ 3 files changed, 52 insertions(+), 25 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index f66a98c05..6e8fc3eb5 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -255,7 +255,7 @@ "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", "\n", - "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in terms of predictions." + "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in their predictions." ] }, { @@ -285,9 +285,9 @@ "\n", "### Logistic regression model\n", "\n", - "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation and the logistic function to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", + "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", "\n", - "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients, but they may struggle with capturing complex, non-linear relationships in the data." + "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients (numerical value that represents the influence of a particular input feature on the model's prediction), but they may struggle with capturing complex, non-linear relationships in the data." ] }, { diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index 40a782c6b..1fe576e0a 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -630,7 +630,7 @@ "metadata": {}, "source": [ "\n", - "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type.\n", + "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", "\n", "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `classification` 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 `data_quality` tag." @@ -680,7 +680,7 @@ "\n", "### Log data quality tests\n", "\n", - "We'll store the identified tests in a `dq` variable, in preparation for batch running these tests and logging their results to the ValidMind Platform:" + "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" ] }, { @@ -698,9 +698,10 @@ "metadata": {}, "source": [ "\n", - "With our data quality tests identified, let's run our first batch of tests using our previously initialized `vm_preprocess_dataset` and log their results. You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", + "With our data quality tests identified, let's run our first batch of tests using our previously initialized preprocessed dataset (`vm_preprocess_dataset`) and log their results.\n", "\n", - "Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" + "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", + "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" ] }, { @@ -763,7 +764,7 @@ "\n", "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", "\n", - "Use `vm.tests.list_tasks_and_tags()` to this time identify all the model performance tests for classification:" + "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" ] }, { @@ -780,9 +781,15 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We'll isolate the specific tests we want to run in `mpt`. When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. \n", + "We'll isolate the specific tests we want to run in `mpt`:\n", "\n", - "We'll append an identifier for our champion model here:" + "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", + "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", + "- [`MinimumAccuracy`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumAccuracy.html)\n", + "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", + "- [`ROCCurve`](https://docs.validmind.ai/tests/model_validation/sklearn/ROCCurve.html)\n", + "\n", + "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here:" ] }, { @@ -808,7 +815,7 @@ "\n", "### Evaluate performance of the champion model\n", "\n", - "Now, let's run and log our batch of model performance tests using our testing (`vm_test_ds`) dataset for our champion model:" + "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" ] }, { @@ -836,7 +843,7 @@ "\n", "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", "\n", - "Next, let's see how our challenger models compare. We'll use the same batch of tests in `mpt_chall`, but append a different `result_id` to indicate that these results should be associated with our challenger models:\n" + "Next, let's see how our challenger models compare. We'll use the same batch of tests here as we did in `mpt`, but append a different `result_id` to indicate that these results should be associated with our challenger models:\n" ] }, { diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 830dfaa4f..d8c2a949e 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -480,9 +480,9 @@ "source": [ "## Adjust a comparison test\n", "\n", - "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb).\n", + "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", "\n", - "We'll run the test in isolation again for our two remaining models without logging the result sans adjustment to have the output to compare with a later run:" + "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) to run the test in isolation again for our two remaining models without logging the result sans adjustment to have the output to compare with a later run:" ] }, { @@ -504,7 +504,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "`MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" + "As `MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" ] }, { @@ -536,7 +536,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Robustness and Stability Testing Comparison Between the Two Models*" + "## Run diagnostic tests\n", + "\n", + "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", + "\n", + "Use [`list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to identify all the model diagnosis tests for classification:" ] }, { @@ -561,7 +565,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Let's see if models suffer from any overfit potentials and also where there are potential sub-segments of issues" + "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues. Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data.\n", + "\n", + "We'll select the following two tests:\n", + "\n", + "- **[`TrainingTestDegradation`](https://docs.validmind.ai/tests/model_validation/sklearn/TrainingTestDegradation.html):** Evaulates how much the models' performance degrade when moving from the training to the test datasets.\n", + "- **[`OverfitDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html):** Dives into detecting overfitting beyond performance degradation by examining learning curves, model complexity v.s. generalization, and cross-validation results." ] }, { @@ -570,6 +579,7 @@ "metadata": {}, "outputs": [], "source": [ + "# Store our desired batch tests\n", "overfit_testing = [\n", " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", @@ -582,12 +592,14 @@ "metadata": {}, "outputs": [], "source": [ + "# Run and log our batch tests for both models for the training and testing datasets\n", "for test in overfit_testing:\n", " vm.tests.run_test(\n", " test,\n", " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", - " },\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " }\n", " ).log()" ] }, @@ -595,7 +607,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Now finally let's conduct robustness and stability testing of the two models:" + "Let's also conduct *robustness* and *stability* testing of the two models. Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n", + "\n", + "We'll select the following two tests:\n", + "\n", + "- **[`RobustnessDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html):** Evaluates whether the models are resilient to small *perturbations* (small, controlled modifications) or *variations* (broader subgroup-level or distributional changes) in input data.\n", + "- **[`WeakspotsDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/WeakspotsDiagnosis.html):** Identifies specific conditions where the models consistently underperform." ] }, { @@ -604,6 +621,7 @@ "metadata": {}, "outputs": [], "source": [ + "# Store our desired batch tests\n", "stab_robust = [\n", " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", @@ -616,11 +634,13 @@ "metadata": {}, "outputs": [], "source": [ + "# Run and log our batch tests for both models for the training and testing datasets\n", "for test in stab_robust:\n", " vm.tests.run_test(\n", " test,\n", " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", " },\n", " ).log()" ] @@ -629,7 +649,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# WIP2\n", + "## Run feature importance tests*\n", "\n", "Let's verify the feature importance and inspect differences - different models might have more intuitive feature impacts that might lead to decisions in selection of a model" ] @@ -663,6 +683,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "### Implement a custom test\n", + "\n", "Let's finish off with a custom test example - scoring (customization of output to a FICO score type)" ] }, @@ -752,7 +774,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# WIP3\n", + "## Verify test runs\n", "\n", "Finally we got all of the tests from the Developer that was provided as evidence, now as a final task we will verify testing being appropriately recorded" ] @@ -763,8 +785,6 @@ "metadata": {}, "outputs": [], "source": [ - "from validmind.utils import preview_test_config\n", - "\n", "test_config = {\n", " 'validmind.data_validation.DatasetDescription:raw_data': {\n", " 'inputs': {'dataset': 'raw_dataset'}\n", From 63beecf0ce121c096a50d92c6dcdcbff156a6a1e Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 10:32:02 -0700 Subject: [PATCH 55/90] Implement a custom test done --- .../113-perform_validation_tests.ipynb | 2 +- .../114-finalize_testing_reporting.ipynb | 45 ++++++++++++++----- 2 files changed, 34 insertions(+), 13 deletions(-) diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index 1fe576e0a..b183c94c1 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -602,7 +602,7 @@ "\n", "## Run data quality tests\n", "\n", - "With everything ready to go, let's explore some of ValidMind's available tests." + "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." ] }, { diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index d8c2a949e..9a747cb58 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -478,11 +478,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Adjust a comparison test\n", + "## Adjust a ValidMind test\n", "\n", "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", "\n", - "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) to run the test in isolation again for our two remaining models without logging the result sans adjustment to have the output to compare with a later run:" + "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" ] }, { @@ -579,7 +579,7 @@ "metadata": {}, "outputs": [], "source": [ - "# Store our desired batch tests\n", + "# Store our desired 2 tests\n", "overfit_testing = [\n", " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", @@ -592,7 +592,7 @@ "metadata": {}, "outputs": [], "source": [ - "# Run and log our batch tests for both models for the training and testing datasets\n", + "# Run and log our 2 tests for both models for the training and testing datasets\n", "for test in overfit_testing:\n", " vm.tests.run_test(\n", " test,\n", @@ -621,7 +621,7 @@ "metadata": {}, "outputs": [], "source": [ - "# Store our desired batch tests\n", + "# Store our desired 2 tests\n", "stab_robust = [\n", " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", @@ -634,7 +634,7 @@ "metadata": {}, "outputs": [], "source": [ - "# Run and log our batch tests for both models for the training and testing datasets\n", + "# Run and log our 2 tests for both models for the training and testing datasets\n", "for test in stab_robust:\n", " vm.tests.run_test(\n", " test,\n", @@ -649,9 +649,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Run feature importance tests*\n", + "## Run feature importance tests\n", "\n", - "Let's verify the feature importance and inspect differences - different models might have more intuitive feature impacts that might lead to decisions in selection of a model" + "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", + "\n", + "Use `list_tests()` to identify all the feature importance tests for classification:" ] }, { @@ -660,6 +662,7 @@ "metadata": {}, "outputs": [], "source": [ + "# Store the feature importance tests\n", "FI = vm.tests.list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", "FI" ] @@ -670,11 +673,12 @@ "metadata": {}, "outputs": [], "source": [ + "# Run and log our feature importance tests for both models for the testing dataset\n", "for test in FI:\n", " vm.tests.run_test(\n", " \"\".join((test,':Champion_vs_LogisticRegression')),\n", " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], \n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model]\n", " },\n", " ).log()" ] @@ -683,9 +687,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Implement a custom test\n", + "## Implement a custom test\n", + "\n", + "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", "\n", - "Let's finish off with a custom test example - scoring (customization of output to a FICO score type)" + "The [`@vm.test` wrapper](https://docs.validmind.ai/validmind/validmind.html#test) allows you to create a reusable test:" ] }, { @@ -698,7 +704,6 @@ "import pandas as pd\n", "import plotly.graph_objects as go\n", "\n", - "\n", "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", " \"\"\"\n", @@ -752,6 +757,13 @@ " return fig" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With the custom test available, run and log the test for our champion and challenger models with our testing dataset (`vm_test_ds`):" + ] + }, { "cell_type": "code", "execution_count": null, @@ -770,6 +782,15 @@ ").log()" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Want to learn more about custom tests?\n", + "

\n", + "Refer to our in-depth introduction to custom tests: Implement custom tests
" + ] + }, { "cell_type": "markdown", "metadata": {}, From 871eabe21213666dae9406e0f2d725b174fb077f Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 10:53:33 -0700 Subject: [PATCH 56/90] 114 Verify test runs done --- .../114-finalize_testing_reporting.ipynb | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 9a747cb58..f44212041 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -797,7 +797,9 @@ "source": [ "## Verify test runs\n", "\n", - "Finally we got all of the tests from the Developer that was provided as evidence, now as a final task we will verify testing being appropriately recorded" + "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", + "\n", + "Here, we'll specify all the tests we'd like to independently rerun in a dictionary called `test_config`:" ] }, { @@ -807,6 +809,7 @@ "outputs": [], "source": [ "test_config = {\n", + " # Run with the raw dataset\n", " 'validmind.data_validation.DatasetDescription:raw_data': {\n", " 'inputs': {'dataset': 'raw_dataset'}\n", " },\n", @@ -849,6 +852,7 @@ " 'inputs': {'dataset': 'raw_dataset'},\n", " 'params': {'threshold': 5}\n", " },\n", + " # Run with the preprocessed dataset\n", " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {\n", " 'inputs': {'dataset': 'preprocess_dataset'}\n", " },\n", @@ -869,6 +873,7 @@ " 'inputs': {'dataset': 'preprocess_dataset'},\n", " 'params': {'default_column': 'loan_status'}\n", " },\n", + " # Run with the training and test datasets\n", " 'validmind.data_validation.DescriptiveStatistics:development_data': {\n", " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", " },\n", @@ -913,6 +918,13 @@ "}" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then batch run and log our tests in `test_config`:" + ] + }, { "cell_type": "code", "execution_count": null, From 8ff1773c89d2670bad96a99fb978063a04cc9c5b Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 11:03:20 -0700 Subject: [PATCH 57/90] In summary done --- .../114-finalize_testing_reporting.ipynb | 93 ++++++++++++++++++- 1 file changed, 89 insertions(+), 4 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index f44212041..37e16d7d3 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -15,6 +15,47 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "::: {.content-hidden when-format=\"html\"}\n", + "## Contents \n", + "- [Prerequisites](#toc1_) \n", + "- [Setting up](#toc2_) \n", + " - [Initialize the ValidMind Library](#toc2_1_) \n", + " - [Import the champion model](#toc2_2_) \n", + " - [Train the challenger model](#toc2_3_) \n", + " - [Extract predicted probabilities](#toc2_4_) \n", + " - [Initialize the ValidMind objects](#toc2_5_) \n", + " - [Assign predictions](#toc2_6_) \n", + " - [Enable use case context](#toc2_7_) \n", + "- [Adjust a ValidMind test](#toc3_) \n", + "- [Run diagnostic tests](#toc4_) \n", + "- [Run feature importance tests](#toc5_) \n", + "- [Implement a custom test](#toc6_) \n", + "- [Verify test runs](#toc7_) \n", + "- [Add test results to reporting](#toc8_) \n", + "- [In summary](#toc9_) \n", + "- [Next steps](#toc10_) \n", + " - [Work with your validation report](#toc10_1_) \n", + " - [Learn more](#toc10_2_) \n", + " - [More how-to guides and code samples](#toc10_2_1_) \n", + " - [Discover more learning resources](#toc10_2_2_) \n", + "\n", + ":::\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "\n", "## Prerequisites\n", "\n", "In order to finalize the validation testing and reporting for your sample model, you'll need to first have:\n", @@ -44,6 +85,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Setting up\n", "\n", "This section should be very familiar to you now — as we performed the same actions in the previous two notebooks in this series." @@ -53,6 +96,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind Library\n", "\n", "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\n", @@ -97,6 +142,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Import the champion model\n", "\n", "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" @@ -188,6 +235,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Train the challenger model\n", "\n", "As we eliminated the random forest classification model as a challenger, we'll only train our logistic regression model here:" @@ -224,6 +273,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Extract predicted probabilities\n", "\n", "With our challenger model trained, let's extract the predicted probabilities from our two models and convert the probability predictions into a binary:" @@ -266,6 +317,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind objects\n", "\n", "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" @@ -336,6 +389,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" @@ -397,6 +452,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Enable use case context\n", "\n", "We'll also adjust the use case context to focus on comparison between our models for tests going forward:" @@ -478,6 +535,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Adjust a ValidMind test\n", "\n", "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", @@ -536,6 +595,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run diagnostic tests\n", "\n", "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", @@ -649,6 +710,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run feature importance tests\n", "\n", "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", @@ -687,6 +750,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Implement a custom test\n", "\n", "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", @@ -795,6 +860,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Verify test runs\n", "\n", "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", @@ -955,6 +1022,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Add test results to reporting" ] }, @@ -962,23 +1031,31 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## In summary\n", "\n", "In this final notebook, you learned how to:\n", "\n", - "- [ ] text\n", - "- [ ] text\n", - "- [ ] text\n", + "- [ ] Adjust an out-of-the box ValidMind test\n", + "- [ ] Implement a custom inline test\n", + "- [ ] Re-run tests provided by your model development team to verify that they were run and reported accurately\n", + "- [ ] Link your logged test results as evidence on your validation report\n", "\n", "With our ValidMind for model validation series of notebooks, you learned how to validate a model end-to-end with the ValidMind Library by running through some common scenarios in a typical model validation setting:\n", "\n", - "- text" + "- Running out-of-the-box tests and ensuring that validation is complete by running all tests provided by the model development team\n", + "- Leveraging ValidMind's flexible prompting to guide the text generation and insights from testing\n", + "- Extending the capabilities of the ValidMind Library by implementing custom tests\n", + "- Beginning the compliance assessment process by adding evidence to your validation report" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Next steps" ] }, @@ -986,6 +1063,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Work with your validation report\n", "\n", "Now that you've logged all your test results, head to the ValidMind Platform to wrap up your validation report. Generate drafts of risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" @@ -995,6 +1074,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Learn more\n", "\n", "Now that you're familiar with the basics, you can explore the following notebooks to get a deeper understanding on how the ValidMind Library assists you in streamlining model validation:" @@ -1004,6 +1085,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### More how-to guides and code samples\n", "\n", "- [Explore available tests in detail](../../how_to/explore_tests.ipynb)\n", @@ -1016,6 +1099,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\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", From 664591b826c5e54bf898ab61622646f99fa96664 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 11:48:31 -0700 Subject: [PATCH 58/90] 114 Add test results to reporting draft --- .../102-start_development_process.ipynb | 4 +- .../103-integrate_custom_tests.ipynb | 4 +- .../114-finalize_testing_reporting.ipynb | 39 +++++++++++++++++- .../inserted-minimum-f1-scores.png | Bin 0 -> 110789 bytes .../link-validator-evidence.png | Bin 0 -> 398858 bytes .../selecting-minimum-f1-scores.png | Bin 0 -> 327835 bytes 6 files changed, 41 insertions(+), 6 deletions(-) create mode 100644 notebooks/tutorials/model_validation/inserted-minimum-f1-scores.png create mode 100644 notebooks/tutorials/model_validation/link-validator-evidence.png create mode 100644 notebooks/tutorials/model_validation/selecting-minimum-f1-scores.png diff --git a/notebooks/tutorials/model_development/102-start_development_process.ipynb b/notebooks/tutorials/model_development/102-start_development_process.ipynb index 45a259115..68c637d2c 100644 --- a/notebooks/tutorials/model_development/102-start_development_process.ipynb +++ b/notebooks/tutorials/model_development/102-start_development_process.ipynb @@ -674,7 +674,7 @@ "\n", "2. In the left sidebar that appears for your model, click **Documentation**.\n", "\n", - "3. Locate the Data Preparation section and click on **2.3 Correlations and Interactions** to expand that section.\n", + "3. Locate the Data Preparation section and click on **2.3. Correlations and Interactions** to expand that section.\n", "\n", "4. Hover under the Pearson Correlation Matrix content block until a horizontal dashed line with a **+** button appears, indicating that you can insert a new block.\n", "\n", @@ -693,7 +693,7 @@ "\n", "6. Finally, click **Insert 1 Test Result to Document** to add the test result to the documentation.\n", "\n", - " Confirm that the individual results for the high correlation test has been correctly inserted into section **2.3 Correlations and Interactions** of the documentation.\n", + " Confirm that the individual results for the high correlation test has been correctly inserted into section **2.3. Correlations and Interactions** of the documentation.\n", "\n", "7. Finalize the documentation by editing the test result's description block to explain the changes you made to the raw data and the reasons behind them as shown in the screenshot below:\n", "\n", diff --git a/notebooks/tutorials/model_development/103-integrate_custom_tests.ipynb b/notebooks/tutorials/model_development/103-integrate_custom_tests.ipynb index b7e48a9e0..31cd0758f 100644 --- a/notebooks/tutorials/model_development/103-integrate_custom_tests.ipynb +++ b/notebooks/tutorials/model_development/103-integrate_custom_tests.ipynb @@ -905,7 +905,7 @@ "\n", "2. In the left sidebar that appears for your model, click **Documentation**.\n", "\n", - "3. Locate the Data Preparation section and click on **3.2 Model Evaluation** to expand that section.\n", + "3. Locate the Data Preparation section and click on **3.2. Model Evaluation** to expand that section.\n", "\n", "4. Hover under the Pearson Correlation Matrix content block until a horizontal dashed line with a **+** button appears, indicating that you can insert a new block.\n", "\n", @@ -922,7 +922,7 @@ "\n", "6. Finally, click **Insert 2 Test Results to Document** to add the test results to the documentation.\n", "\n", - " Confirm that the two individual results for the confusion matrix tests have been correctly inserted into section **3.2 Model Evaluation** of the documentation." + " Confirm that the two individual results for the confusion matrix tests have been correctly inserted into section **3.2. Model Evaluation** of the documentation." ] }, { diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 37e16d7d3..54c0ccd1c 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1024,7 +1024,42 @@ "source": [ "\n", "\n", - "## Add test results to reporting" + "## Add test results to reporting\n", + "\n", + "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", + "\n", + "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", + "\n", + "2. In the left sidebar that appears for your model, click **Validation Report**.\n", + "\n", + "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", + "\n", + "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**.\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "5. Select the Minimum F1 Score test results we logged:\n", + "\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Adj Threshold**\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion**\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion Vs Challengers**\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "6. Finally, click **Update Linked Evidence** to add the test results to the validation report.\n", + "\n", + " Confirm that the results for the minimum F1 score tests have been correctly inserted into section **2.2.2. Model Performance** of the report:\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "7. Continue to work on your validation report by:\n", + "\n", + " - **Generating drafts of risk assessment notes:**\n", + " - **Adding findings:**\n", + " - **Assessing compliance:**\n" ] }, { @@ -1067,7 +1102,7 @@ "\n", "### Work with your validation report\n", "\n", - "Now that you've logged all your test results, head to the ValidMind Platform to wrap up your validation report. Generate drafts of risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" + "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to draft your validation report. Generate drafts of risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" ] }, { diff --git a/notebooks/tutorials/model_validation/inserted-minimum-f1-scores.png b/notebooks/tutorials/model_validation/inserted-minimum-f1-scores.png new file mode 100644 index 0000000000000000000000000000000000000000..52ae43c719a2621e56360d2f2f357256253a4fe5 GIT binary patch literal 110789 zcmeFYcT^MI`ZkQBARt8$kX}@f-a%TVsPx`@@4bc)0!mf7^o~gHp#}&=1Vp;j(2?E( zks3N*JkR-+^}TC7=e+;Df1Io}naRxFvom|&``-6;-Pe3nSCu2gr^d&^!Xi|Vf31my zh3AWfbtm`%4(5z%4jvB{)Cq^h2&QIH`=s8f>$nSJg)=Eb`I z^BsX1W2)zuFW>HPV(%>tz%#Kl8WFTfhPTMu>NiR+5(u!QzA&ni_1|wMVYr9oMk;Ru z!Xhb*7s<_F)mfP&+5Pc3ghY?cBOG6Z%_9TfY`gAb)UwRgJ*-->T%G|eseGEliU4Vx ziM%fR_ld7xq(~ugYt&r#t(m&Lm>S;;yQ{Wn=4PLUsD57(9LEtlzlxTwh%Bs*$L{uY zAPTy;O^>}b@$`C(?WNyUmcvHyQRDl=qtj<#&ot6<*(>P;ibr@g!dNf3DNbiN_yn{b zkyh8rx!m*Q9;7mtk$9U!Hn6*C6?*lK_X8o5YUkU7dD|l?}`4q?5I`WAPCk4sC>jd+lWPol4H$mh!Y$=_C;RBucC$^0@ z$E)R_bPuL7V+*x=SAbg!PJNjCw~q`zANBQy5qC2`zJIUGvbT>}A)q3%_oM)H!&g?* zlr6|wwnUGnT))Qw68NyXc3y*Rb#(M=MSFYo%~)&6%j8@?5XCD&#dYF#%1~_X2Uwh+ z-w6>a;}}TI&+07QM`=q+zazG0#QyM3_LTrHb4|6}-RhdpSXX$T9{j-VJx8BJN&(h7 zh%Az`{NlAt=%oCEay6gRNQAwl41G`QSG}c`@X$XVD@W?>6>;1d6yN-zD3ty3R`S)? zK)aXmcqqHa*aEd%*eTsomRNkDPhQve5V|wW6A(w}z0%M25Z>n_=>>wnY01zmKU&7g z%XYU*Xi^vM9;o?Dw}X4K$-38j^-EWMoE+WTmM5nDJ-z9U`&W}+614M_!$^Ds*N&zB z3TU|fEmtr@j(P`O&&OcRaE+@M!m@IX(+*8UWiPz0tpc^%iwHL|#is?l1NVt%D8C6i zmppHLApS19OVyGR%2GiOg|6yjPX*KItaOs23*34L2Pwuk)r2RgT5Q{28Q8F_s|u-R ztD0#;Kwlg{{X;*pIkUOR?5sCxUjB|^vOIWpxAzgJ9jl!lR?-r~P;%|}3+As%a8-Ui zdr83kp5qrz6M>!@uHT)OPVwJV(j=rZvQIt$EQ9r#Pd^BJdm9ih^^EqTbV9uLGwe@{ z_|`1^3~bW6%Q5Sb@r%2AomfTTe0*AicdV?8IZ2hikMU6n({OxO ztWm6>a0(qcr=E)Sv0(6E9{y_3gl*i(h*VfCkSEPel4EDCe7eU>$7=H2Cms+#`GdvU&XYtmos)?&L8(u?ulI*u z-^h<#wv5M0k*v=Xm7?B)02E-lGm;ciMYT!5#MpM7tA0Zic3)*$mc55 z8CM;p{|T_?6{HpNqn3#gr6#BLq1F>J9b@xW5{m$QGL_;fWcX}W1a#08tQ2g{2pF2l zAkT;pa&tJf=c|9~@Wyf7k!y6e)JJ-^Sy*9CQC!`qq)iH3e#*Aj`!L#K_!s}Z=<{fo zXn|-Vx-YQ}L~5)`#jovPs^zM%s>Z4)*Wi7@1^k8lmObTYhn_V)ue5=*%XE?S+;A$w z)rVw+MiLdi(~}mclkna1?T^QiC#PErKQ9YL#)v0Irp)rps)LrWM4$p6gwq_Mqral1 z5$@T`rd{{Dq9RZXHcTSQsJeX7xbewwh)9%aktwkoR0^+-8zT($b1 z&GL|pfvl#_F>#e82t4mZK?UOO-;%HB8S;o6yIqP^aGdRn+Sn<<2ZU*92CORjv z5YPas1Hr;R_UK2YQwULX7-(|Nq@<@fag7+X!)hB9>kaL7+)m~VM>6b8%uTEyBjL8I zP69neV;e18qbV+#F7VFNNIJ(&$5aPB3*ZcW1F(*A&JF5DBmHEVPlo@CXEztF4{oe% zyKGFCtXz>*lU1Ho+v>n4il>QkRZl>!Ya*?oL2WNtg+F*XLOK5;5MEJ7) zGUMV9!U178Prt;wZoHISo%XDK34h6Roqc6<6LGcVCD~lp#CjL{?$^7OcT-Y(U(dcC ze#Moxm8Oywl3wVn>J$i83nh{HCZiiVze06zuhr2MSOVUv_&IK|V^L$VZV~rw@0-nX zjMZpOo|Rrq8>Iym)5nsyd!Jf==zK!wSL7GHru)e9R_=}7>sK*UWY#X)$=RZQ><=jn z&Lk=gs|-nUj;2^Esj3O*NVs@rPM8`#bIDy1wHZosHAr)Y(|%Hm7g^i>%5Y5)7PG?I z(-n|=6?sLhMOH+gRwSgtUun~_aNIMVOMZ|1iXS_dDQ#3@+|D(a4k}|CdyzoCV7S%bKF%EQ z$xmZtW3HglnuS|CJMdB4R;KCW>0H>*M|F0?50#-YUUUO=6~z39P#fu zi9&QNG0waD`3t8%w<<&a^&UI_~1zP_7 z=CtQR$u)Jy(AbFRdgo#{PZu`TGO(XYQ{`vu;7l{_i!4J;CL-=uWw#B(1GnFH$24Xh zh`22&Hcti``CXo8XNgw_4&3e?Ei|Japd(REi#!nXv(JaXxg~TxynzAo|7Ba93Rtjt1AXNo&p{Kko>yzWb-0bn1}dg9j&K? zZrYF4tJQ0(PHi0xs#~obnA^xH$-zZ=EdxkNnT2v5Ns>W=*?Bf zEPZQV*d5y2OFp!@YB#&{@I{>PHdgE69V{+9EF1|D&EE3i2*`bnY==vdGc|pma3qH- z*2K46H_mo|6AUnaM9716^h{}Kp_H9wg zaKWoH$@=?>I+VQi*OVc`A!IN3_$V>xBGN`*!B#~Diyd?P0PEfzYOMR1qdS-laEIo< zkKf*Tfrb5#^LMeZA{?;p{j-cJX8-3CkJ4z5RQe5`9+G^{663PuhMwFpRxbz z_)ak98kV${tbzh&uVv+FW8>;&@8(@Roo04%+(O z`YOs|R&Flb7S?W-Hr#$L?tkusCE+KAIdrk{wqW#gad!0*^OI!$M+q^^@t@N?%#8mi z;_W2KtgoWZDC_2F!zj$n%gxISz-MG+l<>5+71MnE_Mg=;za*LMy}jMVczA$7AU9Be z+s)IChfh>gl!up}ho7GdQ-aG2Cljj1Z}=d75zgP)DF!D|N> zOrK%e00_JkmiR~c|KrO4wD`Yj>iWaVm`5$L79Sy*j;Q8-C1K=k%KRCyXBdx=0HEqluV`cyNbYi|= zV75Pd%+?s{x1z{}g(Zch@LF2i?+$VSuT|NHq4yvc5`~;et+gKDR#GS|vF2b?NQrNI zFe96UOmOZad!rpn+ONrOQ~E6Vc|iij42*N+7U(@J>9s$f-i#))ts1!n*$fBXY(j?T zo8!(et`oCGRO}xxzQelv?|wac8Xs-0*F8@l7X0s5dZ2&z9^8+FL<;N9y?^iP{R1+_ zX>tpZ(Eb0LKZj`q=ij^kKmNym8u?Fm{a={=f2`|2lkwkiVE=I||8Xn-kGPfBDPO<4 zh_8JM60eNa;q=WunXNKi__6b%P5O@^%WGnm<%Eg7HOqfv(fs!Ovqe>@QcAiyB1|f^ zW3{mySWNQ!khP2d&yxPW8=nNJSiyMt^j|;uGtu{@N&w!xs`X(Xhh%QJW-7OL_0#!v zUcUE#LJE?%cY;kUzDj@&lOBQtqd7rA>Dr~*(mDH+D)^M_Pr{#ieXyta)%;-tF1PgP zZzGB->pQ>_r7q*Wok_mDD&`vb`8k#C#Zi=Sjc?0qf*Ub!{TspCmT#8&(Bo7L5B|K zXRB)23oOwzOt5+5g>*aSMk&}n#QG$op++whx;7+Y`{Sf1DjaHaCA{4HA(ztVUvw%7 z-rap@N_b0=H|jCeXn#*cnV!sjPF;^4;A;$6yvnWv^~u~{o1I6zJnCej2x>RY$v!E{ zk_4){a%*XR_x0zT<~}eCIy`nbcV8751{C5`#(!&;LFLw2(ULI)G%E}@@!#`s-l%I* z{AigVa5M=RAQd1Z>XPk^?PJY^i7k|XuA^7Rp>x5jOn+~>B!qYO<8@!;abIT4#(@-sUFRHR%$G}NPr61r=)g&5xox23 zeX^BnM?SmRRl~VQW!mM2GwE?DALSqVzB%%G#D&fdWd0Xj#{E7tyyf`H7g}!8CO!m> z@gZ9%?8bo?ao$K3wZZ0h%mEbO_q6|O_S%aR&`CM+RX8E7~pb% zO_-?0YgGJ*XyfujsgbqB=@KB%ZS`Aj2OW`s8S=~N(F?JeE8~m~G27#oxVOYHQ%cPR z*RdbW5gJ;7P~*Z3&#~kth^>?xsN&?QG@ohCMcG3zcafN0kuuf#?upYgpPdP5G@(z3 zc_+8zaJqoqngCfmyHQb+F~p}xop=u@9&8e5aF_cS2ku&>p`w_~N$<`8Ubm_=3VqC6 z4~rXdss~UWG{0?p^Y^g<vYjy}7&=ZZxmwEhBcX4$fGls|=>l=>$Ya(~e2w(6 zVgm0LziN0=XH^ANH7Q*rzrItDn+cl+qZxEa#*~3bHrRy83wK_<9I*|SF}g1xu`|Kg zm5(E#e^<{x#AG%udYiko`q<$7=%>Z`dE6@q`_OX{i~em%_zyw^h2k%R#3@(917*5B zimE)Pz18QFhz}2Z*<6dOB-a;ul<4f}NZk*DMY7JDSayU0B=hx0bf6 z2b-LpcO57~Ic_2s%@M!zunXQC6&<`%zlZW*%Z++q``PJVK*&&?uS7Q6(qz4c*EY>l`KNun(i8X_$E^bWADu@%@D1Cfe`5)C(p}oG2@Vwt*eZB~8kkSceMxC5 z=tiF|@QcmvMlpwRIvEOQ;Ond+)~k9%Sk*MQnai1Yb-qIHb)0GzlvpORu)X+ZV>pLf zWU^$ZZZJpG^SA^#eUNz7hyscRUWKmpWh>#Qb-RA{+G_6%Ho+yPPiX77a?E{OYu(3r zKA0(>KG*0HaaHdWmvw_SYwipnA?J)=VuMgEVW|aFD{r?{q9$~F z0rA^vY)%01sdIujLTFeF)qQgvq>uQ$>vs9fWma%Xh?2`c+gd(%5nejdir_#9i9VN$ zMNSF7a-HKgQue%B3L23=anvc0`d~F2{&aeMPp?e7bnRm04dh}!`(QuuyLKdpZpDX& zbl^f^Aa_04MaAv$xa+Ras*avQdjLrJdEM@t0RWEUML>=iWOlZD7up+*eI3%Q)*DAh z*|zj1%GilentID`(P!UIE;lv!{*h8%@cm~HiQB@jOo{%BjgaCi$3TiSk@&T~M5X>D zcDC)Y!jqf?(7yoeSNPigt?-6n{=vWd_#QJXX*Au$F+}Ct@R)ywS4nb!0kqL7&X%HcGY(m zZg%;kkbp8tnd)v(ZpGYGz?V3!^B3)%RZA+#wYhTSitM-tkn*qT2>M!Z$e(ex; z>vkm`FBfT6rC*Y6FZT=4X~jeKHfE^l=N$|$8GDr*8PysqG^f{a@Qo9@;0{?w2W_!U zM91e-3~N`-NBoCF_iNE39Th{7YeQTnG?R5jb!m?lR?5bFe{b=fPg(OHgI)&Rw667M zyg5D}$9KnbSd28e2{!RJmekAC;OXDUhG&mg(2LK{tKBYHFuV+S-H5!I*|Ebvk_>cR zt$YWXGB*6H%SZZ~V6YttLflH|BQbj|LTV^c{@p{VpB z^-?kc;9|(8R{-^4hj1eT;gkNa=FRHDK}A92k~hH(yVjEkzOYOdZ&xRm+;3OWEu$v_ z>`6DK3UjZOp-K@I9A&4Iu|~NzRJvmg%b+Hrlfg?zat~h`Y@@t3YmCN&AwQa}kNS`))6+-U4<9 z;jFsV1{^%mIp*KAjTWMwN8oA1&`27)@)7%#bXJ9$B2rN8sVyfNXIV>2${bew-tJ5K z~*Y&dRUH5)!KMhx(7 zq&dGL0hKsd*#UmOei>{6M_+M@Ot#F4#>Dbn_Y`wfV12IpDZpp`(;alukk;=RLp_y9 z*$Z%US#KLwIx*hrLj+uvDrfM^0aIyOI;ffbIE%nwJrrlV1yPSQ~5v7~Gef zuU-eQIknY)Ntdtc@Z1_};n3B&R{17w>0%$$nc@n-c z^D`Ax`fcemJ@Hi--efqAC#1 z-<@8=Q*@ikd%Uc&eY$1^Vd3AsDXv8FAb}S9caH{59z1SAI5C!*w26Iz95uSnE1~)9 zIIq9f)(a6)m>n2aJ%7V&LJwj?$o&gk*WMB+at%(wO1`W z;gQ#33OP+X1z~BkQCSk(+9|$GAvy%D@ZFDs4o_&ha5K|3=bEj`?v`AD)@o*#J2VG* zLn{tClzAl*NViKZkQP4y5i0LnJwYW*uxZW0^{W@ri$AA`l|Wk2g_~=7>zB9T<@LZH zsZTSH`O^u+TOD|S9W37muIWcfw&w3hRt5&No{O_8)p`!{Zi`EI@pP$m z)>WQ0&H!CrE4)853_K3}st$5F%^a>|GbsxVfLga>{8!1+nZ?ZUq{#4!aAn58ulH)= zM`ATUlbUx&o!({fyDgem%EEQWs@3B3Gs8HhPj+riQ>A6!>MQi=v+@ztM~-zL@ca-b zpor4tt(zOvGfkLui4R^Y>(C8gO$hh@yv6+AF3o!vW(pkhV1 zdfBQaW;$ut*|s65^q?GBBVmxZ@2Y+?`J_Q}HmVt>L-sq5?Utu>a1idh zalZls9?au*RFAgt2M&7ezZtGys8-LRtK<|$pKNQMhK~*3?0;Nms7l~o@E@pI?Olkn zQrQ)F>G6#~^Qedb{d5Y4^KT&^Nz<46NcB`^=QN4$z2DkYWzDE7K9>9I1<05KSn+th z*g{MabCEByS(R(buFiJHe2PBz(p-Z-=#tfDiWrZzDabnbf>po`oc#A+2>!N);eU6x zM4Z;hu%G&)oGdX;n-l#LVxSyWWb-uVqYmR+SZ0G|a&!>5M%B>8F-}#rhZ|_$LlZk5 z153v@iHpmA7BM6*cawqRnd9o);gPmmaY~7twPROopyxBK@7MUe!Si{8(D(V8>FnkN zNly5|T_M7^F@dcOOP|#(tDaVQeHWQ7WP-Yq75AsU>ejA51Ugr8o@l~)S1Tgd-Ezg@PMi5 zwy}m0kbCpg7WsT=rg55ICgU}%b&pNrF#0}o8pR{>1EOLic61s_beS^#@VYn;T4y&h z6IF>=mc^Bi+E9U{ggu>Z*pDI{Tg%f0Tu&z2hvk=7^)t%F(l!i(8rVJ+%VL~pmvAxh zqPd=BwNl(dQ(qW#?0~En^3eN5>*GEx{Y}@!ijQhP_b|}A?}Nr_cHj(Iy7H`k#QL~r z;C{;NL!~C~gG|jJ5ixaN1X!MZ47Rv5J8)5a|hs=6KQc$TBk$Is)aIwDnDW7+k z2rY%vsvl=lQYSZ|HF6QQ^e{}<_;BL~`>XVAl_JA*_=)qZe8!@GUEL2FZVz0z_k%`y zo{{*vTS#%S?50XknNpVQs3@+F$}{0cYIzvJTF%@2vnA9^Ke3){Z*9IaZHQ{ku? znF{nzeanMnYqZbjk-I^as`nmgeN>7tug@fkC{F%VZ-{`s+IvK|qQSxAV8a3jg_uOd z(j^K%HA`BsJ$~l0Jzn&Q1vWuKIDTytY>KL!GbhAzKvgIA3jVSYaf5O&p*7=9HpRQ%sJSU5Q}_&j5`MQbULT0P+#e<79I#hJn-~~SiRoYT0UEW~SXxQ~+_HwG!So*lR zCKKHTY@ixdygq^Gu3uRCzefpPYeTtm@YLXQQX zdT*2DM5xqI5dMU;f}*YtH0kP6PL%iucWF2md^)1ATHP}mnlc7q#v<=28aNAcey zuMX!wc<^ry!74%5vJo~DQXk>Zg{KSOCmJ^F4yeI>#UmyehmN{%6tl%$$q3(W^S@aRT^`_aRnTj$Qbj(ncw$npcWAw=O{u*)CSv985Q}m(E6Sg+)R(^WQi$I&%5h+b%rMpZA#QJ zS%BPIs9wx_VkCJ1ooHWx^{0cu;1<78=08jXcnW_yxjZ0BYgr}egQNby5;S4a;tAZ_ zk)thlXM=VCoQ?4TO}SAnMVxkY0^LP)&ST~#L0a#tCi91~%mCh*L?-_IadVcW{U+it zfi4PPpNmSeee+ZnzmvAoyuKSqi_-Sk^K~d&=f(a?;grBGJ$}tjO!8V~vj8M^9W~S3 zFg3*Q?EZ?)ob-FdAkQ#_&e_Y|8X2lu4H!X# zNx!+C;ZfU40V_S4%F=Cdr@_|#D==%{@r7cShVN0h64D$2WO-KDrI{mvL37&JG6McS z$yr5E*?>F{e3hl{;iDl9XE}0?ruA{Y#3r1CR3UjG6#}-<e@2Q-pNtl{6`&_>XEGIQ0zlSZ4*DE})qRhscRbyoyKDjhi#ydF4ZfrAhARMS ziaAJwJz0HOyhe~kut2w|FPXOgZS#B8nn!=l-3Zk9GV|VDrkxo!g5 z9^Nd`f^`ii+f2lST6@#gkUlSQBdxxfyjLmjP&py1onU&_I~d2N>*>_Ye<*XAG7Mbz zhlJP;e?qHh(8GOrW$7B5FNwgyznm~YVBPk?kHG3dPYIzdbWXFKudU^9f1J&6YuIq} zcW>vXVr!-FZLnG{1AK^P!l-Hp5+k$=L2Z8nEAuKxkTjIR!Au)c83HE zHyhG#m~^dL)y5j*9gH(-C#k$^4a0C=S31m@l3ukFE$b|okEhB{n<2&ffV{^h%S66m zM=i(Gc{p%?faK46zbpYwLbR}zYv5~+faXloU5E;UkA%-cg`OD&;oQ5 zG5CAv7o>|UK=DZ0^O1*id5OB?_{NzckUz2IV&H;)VM41zxS{ek6~VO89hIdh8boCh=WSv5 zl0nn!766sNL>Z2do$?V~yOvrB+b?lcyP9CPElh`lLO~kqE=eK04^|0gu9CQsQWl_sqQEh?Hvn zuJ0yK!Zip2TgB^h)k^y53=c5vtc-acEi^40iw*d{eLCjjBBzp6?R#bIwt)dg>`K5> z+BI!q@I~I``i)VU_}nbnq6HVLw*%{>l8Ia8DGr>}lSjz9z~U*q$A@e0d+Rf~*z9=C zFamiUt%W}*Yl?fhiRg0Pqb-~i>3E8K;pC?*86G|+_;8(QcKa}&Rka^GMD7OzF7b8+ z`4eOFvih*G)=uuqtRiV~7eg+jn^x$WoL0Aed3oB$s@>^duDOod#Kbf%F2%e1ykQYJ zPGF0Ap=?;DF3eHTy@`wnTtDnGR;VHA)PQa)9@)4YAwHz4(h!+Y^Z=gFb7HVm?df=Y z;iV)k(VxIj?J!_YE_?HvSm`ElunB!UiQQA~=?0fe)#jm4)p?Z=6YAf!bEu++7VA4O zapD?ov*s%kjsQzHvRuTD;q$>z6J+DnHLS4UT>2mL07D@>*D4ySCuUtASYV8zZ2f~P zywkeH_Z^X@Yh=hngGbgZ(>~k#7${nQS9?ZLsFGdflx=e`heZVOf$hMa2Z6W=3GW+3 zSdO*C&`c%bS>*zSF)?4g&I;TQ*x{dLLR;GeT z^8S9fTgU$5v#Iz*%sTTjmw>Kc9ra1KVyx&+dG%EF0iIK{_$H)ivyD%-Os{;!d4u_vc}w)`sk7`@VjFbUUj3M%i0l!{sK2E-3HqWz$F-@38iU7~6UbNHp;?w?1P> z;^)A66}j^tCj>o_g(!Kv@}z<`;9xnL3S8@iyM16?oJN=?^S6X?)0cZl#uV`^a!2TZ zNy9)DuT?}F_uCRBx52`PzADIB)3TK4^T7E)UVoKzHnD(gVY>Cr0fl^|&I)0wpg=xb zC7T$`3`5o-)|5zR-Ki~3h8WN;^ zDp{nO?*+t+l*7i7;g+ee9R$sFZf!W#vFN*WQ41MgG@-w$YF3N-{mQ!1q{%Vu!7&Ve z0acIA{lz~k#;rvhFA6FSbnz@~TDK|dKuV|P+gZV-Y)cvq-6qA9pB zCj6`!Pp*mzowXGo?I9Fw;?Z}{l-{G0d(YKGAZs2&Hv#=`w9&(duT#@%V{bEs^ezYk zrVOf?I0>!7#_CrlF+{Fm*Zuw3HaWn}nt(ym79`h1$h&Zv0}hhqwUQ%DWI^|Yj74 ziORRD7xY(&M*D%jUV7X|r%KpnM(r09cuq5yp@)WL#y+e~DO1s0*Tb!d6^Y>yaL}hy zI-V;60?9_8ey{p)ukhca6$RO^L_XV-O(FWeF9+yO4}m=xlJlmKV6#?<3hAG3M$Bd# zzI4>ccSPAw!(eijW5i)t83d}Y`zT_(QgPrY>Xgjh|7ZZD&-^Dje-aUWk31&G9p@+!tSL)8#1^q$X$d=~0an+wY7@u!rT-+OQ z+F*xyWZk`5%quX3jgGWSQ^!-8JyoiIf$b^(+G?FX#mhYDXLQuO04`d@BV+)^Oy`sf zD{PBMQx@(L)|_GRiS=400aC^0Ld&GU_|v!H0AsQPJYO66OkMhFGrf#?&936FF|b^Z z>;ZVI{Ng-6P)g0I2IEp?Eq8vV-ONT8GV~V!tqyrlkCyl3WZLew$z7d9n5Vu<{x!hN zKV3LkS)cJ~p?^``a1jiwlH`c1-(f7*?R|)zLT~Sn)j*u=+JGx_Tt=0Q10tEQv^=Vg z#HTEkR3%~J(>}m!+(8gbtT+n*YMrQDy^SfP5ds%f2DF}f*YnsXo3O8(48QG3ias?@osAYuj&v2uKYmuw9AYAp`D0uY!ER6`|#5VyWxeY+a?pSXvj?4i-PT}^y4| zkwoEu~WnwcwGF;q9dy^sD9?&v~X~RTFSOJStUe4*VBz{9{=62~y=#S#y0a zytHD<_<^V1c8lV9=o0%rML-lVphKg+KcUh5J!>wVY!)Vd3MZyxz4Uu3NS2vJkvgjLehQir4OS4vHY!Fw4Fm@jHG)bH5O-#BP*00mR%JxI*f!C+CCr*_8%vHa{rX_m`s z#*a*{OnY(x7fi-?o6i-+Cq+ajUX-Z2&R%Iwo=XwC`@pvylPbx!4J|E_eJd6LLgt9C zAAonoj`ly!aOe&+!6-$eY>#)?MU0Jpq{lg7CThkc44(8Neq`oqk{FPXO3}VKDAJLi zBH>{Mu`^}$y=qaTr==br!H5i!NbetAI@*%8$_?9i6HgAhd{!!!L!0{g)hB zvcojfN>xLNO5fJWw^t z&Gp*CiTe;s0lT!f7%9y9B6#QJ(ITGPbUz&uf753h)V7)NYE!3FcM{fJ0HfB%Bqy!+ zcG5f{i#K3djnsTgJ796v_3g=Y6-dy(psZ9QU)18 z{}I)>*;H8eM1COhpmHqNuF0Bc-=nZQ`eR5Vs?tY!YxF7Oq89OZZaq~PWUo}7LaDhW z?c1QcIrz!Mu-EERJ+|*(EOwu`NE8{B$~{y}YL)*eK}@1M+*9d@j$r$icz)Ol)Fh*y zD2TA{swo{op!-J5FgkaX~{yTZ@% zTLL#I>$GYViRE;hu3-Wu$supNANueC+C$4V1*ij!%eS)YvQ@JSUB}uapT=6R6E@fz(Y6m z#1450r&cU|dg6h#4uw`B@WK<5hU|0oKY?V%KeH4* zvoaf(3%(~iic8y+k-hd?wxiISDS3;t-NTubeMl--N5f&!ynnn{h~)=d7U-i zu_Fd}f^POWfw=J4Yza1C+Pj(Upb`CDhAK#u1SkmG&Il}nsGdG+cs!D;_lJrZ0gf*} z@C8L^PFOtm@lybbd!1HD@SffOqe-d6clNLJ6)JN#j<__wDZ;$8=lKRds-w&j1F5}p z#0ftS@lCG$NwjaecQBmw;$JxAHki1vyABB_!(<#VF`9og5O8iB-T_z=onH{q73}i9 zpy2e-i{fwy4iEqXk7xpn&nIm;UKgWHIDG#^B8nF6;z8({z=Q#E#&H*+4+!WjFY8zR zLE6ojK)x>5@Gg#@;O_nE%SjsKh3D9d+|8Lbu5rG*k7$r$o@*sZ7^+fW?Tha0h3|MV zi}{3l<1Lygp8pO0OnqR3z5T%4_qLY?Z)4K8IHxQ&yqiRWDm@$KY?gu?3{4m?E5WJk zr4@|1=-zr6dDK-g-O~SnE*x-%fNHK0- zmdKK?uB3aAES12afgM2qSKkaTJsIWGX`kP}n08xkuhC}Ztp05w7wCix*rb>LRB4yh zSTTZZxE0Ji?Gt{~{1cE;sjWrC0I>9t<}lm_%bJmmeIrLkqUy%;yok@ct}M0MogLWh z5_%e3n-!|~OWj@mibKtw+!}6P>!gma3;qiU2=cRTz+D4sobe)^qWYOU?(9WgP;|R- zEhg69R(B$vD0}q#YFo4T(?qGJ;oLoZ;1tmYF%Vkr*thcKeTg|2B%?lo-6m2$^V+3dwrzL2bC~3l;j(jwAYhKd z)99SWJ9&J5&i<=h1QBNnPh10l7VU}34eARiDw`?2ZKoVb17s`jjv8QO9WL3jscJ!P z&0FdrzDK$lo@hQMgaw006@r_rsDINa2&!AGlxPT`6`Clc-&ni};So`|rK-z*V}nnb zRFh4gIH+fG)_(rGW_1j_b8Fqk=4>TGcTEN#1W#DUcJd;}FIBg7`(nhxE{2^6~S2DpgC&(OkpdfEwBE(I!Vw`U5y{!uN2OKkp!Fq#GAcLtR%&$ zr@TcVL?@tnquF`~>v78R%4IE|nK=RqC5$o`Iq4~zp1=|+s#hPJ1tVnO)1}}vX0}n0 zs&Gyp)5PR#G?R$$@sJ(;9X^&yAeWbi4#X@I`Sh^TZUEj|7Gc&>_1MD6ocf)_MydBv z?xF8nmq~$8rIA|mVDQooY`i$W~zO!knV{9!0w;A7iSzWB1!oKPsfe2i!utOW0Wuv-) z{>}RY5;KsGBjqB~Iv9m-p(QQD?BeQ_NTKji2J!PCmmazD!bvMWhG z_Ibjox)rxqCKUZX;k(U0kFMLR z70X()a{~OPG8D+_)-_Q>{07L<37fMvBH}g9hL4 z{q1{+Qeza*5Cc3so0-cpb(vG*2%Dt4jwDNTZ)Y3P zdUoPLWV$*gjSBbEw+w8+A?*dXlcHSSlJ2x&Juen7hC>$|{f>=5CsPzFV zH`;t@b~6MGyE;mEPe~XY&rqH+cU134wtE?k?pXa{@x0}g37@n?ylTJ%5S1(T5_IoQRC?CZ|GK!H4lx(?2M+|08%++U% zh|s+H=v5-h<)Yu>>?7)rOsgwzXH!aUHN~g+!6WXwUmMKZqRjE&8l#rAgGW6=UJNJQ zZ^el7Jx`sLzfFqfImv|B^(b2&YawW+|AKE04P_dstX6tHV?LEe08Uw)o-2|pz9S#1 zmuI3T>*Heu3Qvnw5~eS)0c~ufO~^J_fqPyW1AmLv3k*j<7RtLy=oH-Hq}gXXx!#$^E_>glD!x7+!h0Jccuv4M`-JNl2ES)hmoG9EnCu4A zMp@4sP@9AwW(Giiuiln#x$E0g71>jj$hIIJ7LfOeWM^!cCCnW0QLDJ{{5KzfygUbV z%v7pZ>G|{W(Y`stL7+?{+caeDM^KlnBQB1_m^Sr96W?%qD5;^S?_q(m!PR?Cqw9^vYZ)+FaDhKrahV*CVW $v+SmSk%&vPpl+B=`J?|ZXPgwB#x{#xSB$=prtJ*=z@%secAtv#!d)Bx3oenu>^GE5B5?+%bF`+n>E)oON zsZiHlgk1XM_Ke#N@=wOU{f<9%=4@>%$cgTspsZXzY~7W?lK`RdgoA1@bA8b+JeH)* z6{zI$o%?2IU{dutCt3_FHu0As0x{euzvWufXS94!JTcAKlG{tJc}!-(sAMEIynbUO z+u=U#o|9*XV`1AL60I{NVm|6xb!n-oJ=))*q(J!*nA4?097^){^p(N`=lMq2*|>D; za1jE@X>zu_s%IY9RnVxP7N|hJH^QoE!7ttnelQW^G?mG~`6;-TznraYs}G>Fq5hYhCgX_@cM(Y*68@5zy|{_W!Z>mQisv>(*!>kU-Eta0n#02M_M< z!KKmQ?ry5`;6$$5-rA{eWX9ljD<_CQam2Z+#;$o z>Sg#h8W;e?6p&u3W{?!txU5@_Bnd0e950$J67jfV4&^VNlOVT`gd zYEM@Ugf8T)U+-T8Z4NUs@=%xi@+fYm^IM zvv*t~KU$5;P0l;{c-A$rad=pTq3ws-? zE=wxXb$xmb@&3cHhFNg#>T-}JgeMoJpf=t(HpqMNPc48a?boo9iX;lJ&wf>vJ>uQ|<%V4TNc%)i+``mqWTaJn#${1? zdmx6p7O#aV(v&JH=x^l$Y>kj+RT`DwmD(m)vpDrJ&(&Gw2V%`QtAWYXBA>xVy%#8_ z{dClod(ex832AW5Nk#nIE%T4cDhUQ$9>Yx=Tx~ozXlylpiFMHEWrHJLCWBUx8XS6v zGRzlHl)DLoPAh^lh({~p+uwHZA4DhHB>1K^1{)U|?8f1MhJA@{1JS0UL~-ET-b?xF zsfjE_NRQ)v0VM<^#r%P6fwmkBimO>wk)zoUfTRf5z_0^aTLk!z#D4zbk$r7RHXj>V zTYeT!(x~+S9Kk(AfMDr<6-)c~W#hl1WBsUb01<^Pj_*f*pIZ8BUj@bi;f1*KBWksP z|2!Z6WPd3#0I#RthGgmYRQv}()c@_EUk)YEW^YNU6yZ;sz`$i`1G5~vBG3RViN6cB z{`#q;JkVymF(24Bf7%2=r2biZDOIbVBk`a2Ejk@&^Ir_oKbryIUx(NOYf#bq`5xb& z_l=AfXd?+tw-^26;gFzj0Yu-d`xoP0fB1HQzDcI`ZhigZ;RulQ!y`D0-k{{% z{dwQMpTi+IlQ(nVZ6)!&mP{@Qbr|J^|UyMcZitpD9WB)$Uw z|K30$vwAZ>C-S$h`(o_Jo6%OGtSQecK%aDPyRG@&V~NDt+K^~AJ3H}O=2;Q1(MmI9v!m#$=6sA zF6Z@td#B?cPHf3P8fm=TB~8X1e#%BO0Cge6R;f#ozwPJ09N1*^wnvMcoK|8lc=W{( zdED>oScLR02b0K!8ypo~4qszJ?1LLvN8`G4wh$adm(E0j|JVT5L16JuZ+pMt7r-!D zjo}`B4be+LZLNcrwbF^u`#4{ysA){x|3iS`pQrkNm=`kT)zPtR`%$>2>aF2)-ohZG zZvg3<_$BHuASf=*#Hj*#ISE0Z-~G8+e+J73>qE@-S>;G7fg*f!FnP)ZL^h&*`Cp6S zs`V*;voBh5Ih7O7n<+X2`}Xf$%D)UlAQo}A{9EheNpLE+kB+_30EOA7{D`l#b&&Q_ zA5`M_&fW;gJY+mF={y-lKQYQbo>Ku5cepI3Y{xl&(B0#$wA|BOnYQ7$tywl13Jz~1 z8SmdG75?ST#~3TOKL9_#x{^GC%w@xF0y+tokB^Ni+jLKCGarx*m3f#QG5*va@!Q21 zUBkUPT4C&qA&;?A7fT6>mo0yPiex?J|iWRi5Z}A)cGF^XlKmXk(!cqG@K(A~Ry7YwJ?SJR&zkbS43b>}Ow_*h9JdBY;f3UhsKlq*Hqy4&>c=~nY zgEnweMJY0zhRx}7^h#;f{&k%YTr@e3WV&W_t^{S@C!^@kKdTa-Ds|rB5PK!yS#grn zIE${Zh(-Rv7rmEVg-~<4tmB-mv>*wVN*oYB2tMJlj{aEy;iUYgfPk>*rz&3``c~U8Z5fNY8K)ydUziTR>Eg$7M*Gb z+tpFPl?~k9w8XN~lj}|TgB4h306&~@D?R*n{=NOO)c#jrv5qKx|C30`?`W;c6o|=RV z6M-m!y37dU(&WwA4!K6<=Y*pGoHZHAHGBqGq~LEtex^S0+09Ri*S|)$q)BX{3bd8p92#Qs%7M=2Q zp&j-&Ht&(|II>vVX-(6n#Hs4@NIcsl2(x*K*yo!xjH#j6A=KoK~KrWPOXPB9Kx>Y>sR_yjz;9 zZf|yx+@JiJb|cLhg&U^#z~i(ZOQ(72SECUpg_Wb%e&6m~FyBb3 zUKQ0`7DPEeUYW+zC2jD?{K%-cJKEHI zJ;Sm-A0SC;dkNumGg&j}eXmxU`nyIt>`8^W0EWieu%}>5Kh>D4&`P&7xj{4#XYR&1 zf0$&QVK$eHjPz>qzh7!}IxjY2PxQSyW0}2dcx(~7>N#L(^BM#|qWu{@4_5T!J0ckg zUDaN#P!4|GmYAWG)|gK52q{2iXAtU{n`0VvfbpOaljuzz=!KIw(!sJ_rLX|7(&CEB zhTnh(UY&bHJPwamU6J&q-k2wGnH9gG`p%eh_E>FEYq@wXE3x3tF}`ATUJqW#D{F~L z<3Ei5&E#*~p9bSl*3U>>7b6E>pz&KL*wk9qcr(;P4Wk?NwbB2_QtQ%Mh4`V^KdyA1 zEHlocYSnh#r@8ix=lWX>-@3lq<-cF7^KK^OwHGN|a*WB5jG;&u*E^HIzBcPcy*UXc z&e06VkLHw)mpCe%s{Qpg5_&h(~Qx$U!m=a#R6P3; z-G8XYSXb9P{DX!gO`b;&@b?{e7dyGSa5752H@Vq!2U{$G!|0xDUiRJ$h67@|#hr zM5$-L%VXQkOr1$+>~^>3W2YsIsa{-|Fb@54&sx%6;QC4%~lTr<}d9W-MFvI6ZxULe9N)si5D9 zVS3Cryt1z~e4`_a>I8Qh=)ej!nV@wa>x;?HkqG@7;TMFj35R2zg3MMN{ltoVNfiV zCsqYbirSxN!~=YK?z7M|gmaJKY%4CVr*h%WL3%Bw7r|)o1^hRIbRC3ZH`t07A?*YspP zn$kw6bwPBZgK?%Lrh-&V9b`RPSHW}ri*#)A;p3e<5ES4omSv=V@X;-m751ZF0Eau= z<&uOI7JR5zflP!ZBCIa^+}-3k5%|Z3pPmja&g41 zu9LyCg1>FMG{42?S19n~R<=^ zgAPqv!{zxN2f!sqRT{HuUHnZw(B*n-KCdpuM?9a9kMBNN9u3a)zmqD$9`Zy>^{uaK z2|=Ez@?-+g+*1u&s6>30UV@Jz3B86OuaDm8U0l25I=1Uwl1kmSzPoo)S`Ch7CHtJ3 zB?rDg<*qxyhpKvNGgu7D3OY&5I$yD9v#R8f&RJ&WsMSt`?%-A%uh-D)i1m*~888R$B}5$Zp(N<^hweY~I4 z(=D9*2@BPmw!Srp%aOwJj=165HFaoNM-?cn21hX1tP>CS+K zcl?|3+ZGpSS(eIX^&S)!#8nfQpAJb~ePuZ9SK7w^g6Ax?>9%#I`$}GKd^ZS~2XWmN<48zneVWi4!eH%#VCkH5mH-4xlIJe86gKSw60) z;vv*0-O+o^^BOTDZmpbR7ONc%4iK^@t zZI{^0evO#(C70i0nrRrjRn8Ib72G`C`ixhLbBPSDSEla(FABcISv);y%Sfi<@q$fD z)Y=HohGn782%p=w%vt22;Zs^}^Oy&;#iB)vTTiUL@BIZndo_+$*{w^hgUJXuuZ&Sq zh(@hszv|I`2yXw(4+!W=4;QrbyE|q;l%3+?4cbfJuH2N2%icwQS%*CdlI}uo(^k5Mf!_m$=k#`m2k zbh?9jYI~<_#r?LoRWd$#Hx{7}ZtKv*1bdCE^arD5Eu<19zT02^;~RZ)d{>}O5dil@ zDI6SG4J(L$t=V|%m9axTJF=)ZgSm@`2!_U0FB3s3eVl(Q>0q}wtvw;YRk@zznlT$p zx@Kem%;RLYjqVx&{t{)MKw#`O|fohyo* z0R~6C{Z9No#|p$zxQPF7p;n&V=ORueRZxx9<6NJ#)oT)ktwBu1%@R zz-?m()vog>Lko}o7tbd1xXoWlN{_c^@?-BP7n~qewaa$msds!>QvAJf8X@SlC zD>?2%4bzO=7Y9>uqSAcjV9Hsnl*2>#uw1PxzUi+&V|x0RT)H2^f~*~XbM27>wpeN`Bs`s!a zzdqz_ylyQ}pp-!ZG6IyOzje7ODuMy}2e9_`$&0lgOW)AqUneL%@Dp+@xz}FY0E$$5 zjI&rqbZW{jOBope09B>ZlC4N6gHN7Op3n&1zi` zz?MvwPW?<`lmLnuz9WqP@Ks zkfqbxgt;5Zw_W&lNKO(-X19oGfpvxGeACK7_dk*L)_ELrkD`g-&xR zph4RN?pc-0Pvm=ML1$;+W{=BXEXFr>XD->Ai2HFfjLMr=*1h*zuSb2JJ=Oi>abE92 z99tvjNZAK9Iv>SiR5kpQLyM9cgX=tuCO|Ost4nRAg=0}ej}X19#@x9#y7QA3-3INwnj4K&6SG` zRIpC1i_ZLnUiduTB(OoUt<{zYNoDb$$gj_(k1q&Xqtm!uQqadVSgK{H4>YTz6@M_+ zRcEw5g5{Zw2dOkAMN?FP(9H8BZ(tyS9zR<4(z8Z%qIXK}NYpj#p( ze2c77@@Xu8U*X4D(14IXt7hD0g06QA9YADCzgyp)o7w5toj4p&DKsm==WwrE1n>{7 zL-!05KU8n}Yt^yF_NVJXdGc?eF5DiO9w7sLjN#ZQ5v~gWT%uTi^1Qj=9cD>`Qub|l zG_{f#G!Nk6>54RDs6neBko%RcOCtWhczUfZ01>zCrLGM`Pt=+f5?X+U0MuueqS5)V zVmWwT7bWAC;H%0?B*9Qs*5Qd9FCBl4>3mLR4IQAu_d77N;F`>!Y_Jnb7rfi)n^H50 z2E<}CcP5{NkEdo<`r}Jh!cfgrd(K7uqY%w*9oGDeOG-dbSN&x54`8H`qfNsc;watHM(}qh{ANZ?HiZ<` zk{3#)^xD5Pi>-CN>-%06zoEA;xqz70vDtr8>LflA? ze`U?AxmWk9`vu^S#!nj%!73J=nE{|O*k;D1kw#U}IP=)NBUSTLl?#O}mx>Lc4_&5@ zk$jkf#3yI!EIQF#A?LY)WOC(Kb650Q(~vt1t5>Bco;}UbIRHxGa0FbncfY(-tda^yqwQS{~YY*f(7(d5=E z<`p|50T|A_Iw!C9CaS^?rdatpO3iEV`;u!8F1ROyp+iV|dk-JX`eiO2!OB-fZdDY4 zsTr*c_x_OCdg;ibU{;(Vd&;{BZ_vTE#%cQb`u2!6i_z^v{2I^XN2zjpay%`0T zgqqb}7B%XslLdgTI|pi4ycmxWa3-|fa_4a#oMd?-Lt#jNpjdC7s7U|5mcZ zdF_>vM5FlDTsDsp5wE*4v(a!d=Tb78MSDKYJNXLO-bVs@K&(X*X{R-14_ ze!k?LLTKzu@CG1h{Dsq@+J*5$nNuIPTEv1)g02g6`ifO!D{zZZT|tWXzMQXCWB=KO zuI0i~H3$rugV&-k@G{!DJUpF<IMs)?slga{0bg0R^x9h*Pi3BG`dO)Ort#?Kr1F49zvVKVu6e!SJ(v5#(`j{7H@eEz`+IRnO{F9owr{-a?38kUEWt zo`VMuUC4E^j$`dWkI+SVyO*}zS0a<)jn1~%HJ56fyjJxSKc0mjN!Yx~t~teene!tb zG83pkiuomKwRzYYdN7#Vl#q@A%qHX^t>cJvxWmed4@h>2^n>LaLvCd1dVUoaSKgJj zV;=5dv`Xv|=ziyl3!pgqvocy$Oq*IW%Rms-Mt4No1zQEmMxK*yz#Ba_vw5+2pVK6{ zv%~q3=M)R1`=X|c*VN2TJ%op21CwatA*4mB-a@w*-n_d)p5=iOn@Bs$myidL&f#Jw z)))yf@`?`bXl)8MY&RL?_CBoa^Ktpumq5=l{nypvwQ5^yFueiZnu-l~Y?NA=4~r3HyHXW#=#2^{TQkw`T@KgDB`?1pdA@m){UI+u*S; z2x=8N{0ejhWbrL4c)U($e=v4#yY;8j7ZvN&4}K1HrYqwSw^{;&3YAOAvV+l(LpUyM zZndB99UIUt5x1yiDH)=hoI|E1@j1S^i|PyiY>!D9s&E z6~UzHmY!9q%Je>Z2dh)HzfmqK=B+)tTz0ifsRg!_z z=6k2-eMZa32jzq|HHv7q&Pw8hv-iZOoYY(+G^%=EW}bh7232j-A-h65wdXX$hYsgi zBiU7TjxGbK&<#5$&8lhyMz@^|w`JxhG(Ws+#q;^#oz=WRbEI3y=rOy2w4JE4pN66H z@qrElN}k!Hc>WNTO^L9rvv7PjE^&O|6?LRDFz^#$`4(U5Um&X{Sz` z%;i$b*Xx5Gj-t2@_VT%33tEGNWuL6Kt1>=qCDA<%Mz@?S*Vj}wGU>uWu5i*jn=bK6 zvzXm$-VvUBuu|x#vxtic4V}mzVC>gY<8VUzfB-I}QUzZqFVB>N^gk5oG=k}>kET-c zJXTtwIaXS^le=@5WEV2g;=DFr?`6(!bcJFb<+f$7W6DcBP{=N}P6QXSh`5YQ7XLg; zsjSE~E%&Za2%THjYEnOj*IX!7I&hIq=17Rn&GHyxHtNZ1ah_XCr<^{-tx#3WmNRUQ zuhAQhwY3r75!BfX4WpuiAaygn3j|fbCE)t? zrMhe?Jon)xiiaLGohF>V^6&m^i0&f$Nr-s?(oI=5+8fcC|eD4^l2h(S9Vzf#>ayO@w_z2m!s{XJxRAI9RpI+177?433!>Hy(@3hZ0 z)!oc%Q+V2S?5~3>?3+2`a(b=AR_*czGoJU(Dq7Vo*Z%|SQp!+P|jgEb)JWRc~$%= zg;AK;>e@@Bxnu8_&tnpF4@-5ib@pae!%}3tY=N?W#a;0_i_RM)8XEzAo29$~LG@(T zpKlu(~GuaR{^I3CO_tSz015bDiRZx@UH<*M)7%yELD7<)tbRl6*8B&508b$y-8< zB!0Km&-X^5Ntc)@)OawNvCJv_cI$;cx^SM-a^@%>`gPkXr*(w6IFBw^wD8H`o9_&N zh^uY*pgBZjMdP9S)0O=;pC+pnSy3zpYfj)b(sJ4*6{8DS!fboeZmfx;0sba+wu1qI z+M!_QcrIwj$(*0jBP3wvgDX_{G2&47`{~ywO!=Adn{^rpvrg*=@eT*9CY6WN0c&zk zzOsS9onja4;cx6u1Y(&1AlK61NJ2{Ssah=(k>v)bU|MuvoYBZwQ~WFy1gX`~mm6cX zDeR4B1lxOa6O=DPN)@TkcmgrVNI)Nf9(=ii@)vZMTJO^eRkN)1eyh#?hUV4rY4+)2gHPWt)1!o%b3@-}ePVSkgOW`@)T&-;dHn?N5qAab|Xlw|>O^Xib@ z@jAsFwLLLjaR}}BeHguYcsBs^Hwth_WwD7$e%`p3Guwh;NP3i8FmHh4V4_lEbVOC8 z)879L>LF)<>8zpv9Y)8#8H#HM7c`|FB4s-8vRQWyU?V?VKdyIgmb){uL6vdUR&;r| z=P%EV?~7G0s4uo+(kgE}o;ZzfH<`w^H76dUTp3RY#1Bp0&U8WvT(5#~F`>_Aw=Y6_ z-^Q(F_bAR9;RlFgkih)+r>-AS`{QlN#~s*K5^Q2J^wB0>M%$}ru3=x=V-jXCI)0n@ z(kWdp^6VfnFa2trjh;_?d-+TI55+w67tlGSG9N+hSQa&g=AKH!J7-QO=6$66NR>h| zl#EEt=d5@>a}~lQvwoe3xLpc!4x8V3`wicKVd?I6vw=bf@j=}HHN;&Yv}zR|DW1V= zaz(dSHWZ_}ZDQ&`iWMGefNsB(&XWK6SEAy~&8pvXF8R{M`rKd)tu-5y9+Gbaqm8vX zl)@i*zPD@5Y|Ex>&gL=_#WN_0Y1UYnR0cQ&d`DgQAha3eUhA<(?iH>o^~OnN3M;)W zDB>-V!t^F0W_54*((z#Kv`@ub#4@!HorFG(OD#ZDAYIl{?lU3FvF8wy@y(^U4&3-! zDy~dj5nRyn+`4wzIpv5p7>#Irrrk$2R^Vu?H@oExe&O&|?8lLiN13_y01$E0S;Sy6 z`w%_xa^L94X#9uj9a>PMl%G)GBI`${--aFrKH%n^Mj>+a`_5KW_m08%)xN=m6uWl_ z)g7;01QA{+CRB7$9Z4Z`k#|xSXKQ|L%>v&=@)O=`OQ!I;WX4=vdYnh@>j@TRyP&}2Flet3;ke34a(WUJt?)3Is)Izm&yX~Q-P`==w} z{*M4O{oca`m$v$Y^mQ)Ah_IELHpOM!#d}@P`uY0Cw&!w3jbRO2WuV%z@ZL}i;@vur zS8s=e3sS+dAFSEL7y zUZiF384fwCl^rUkh$^Q8XO3`;j~~mN_x85cmb(@|I$$8a$fIlB$j>NoJSaChUmnT#3d!$g*q?mRxo}tXAk7; zG%rK@w%?L=8;pLBy5-s}+MtlJi%sll?jgRrZ-aSpH+~1_!^bK5mtLA~?;SQ0yzYMzn8@|> zoW@2#kL}ekb0Ra3P%gahe-);|gDHvJoE>W<$K6uY3@M=~Xa9!RV<)OFkj< zS^E@mDwB72&%ZO$=p?|;a4Vnz>Ro`+ZMya%ap;G{sOcPJo^9|+^C&CwBVDoyeRj&e z|L9wV@pkW&DZYr_WQpf8RF_x~f6H#Dy~^*l#;qbI8|-9XWl*YMKveq}lRxaZF({dH zYjdf6ljaF8y=&5aPU?W-z zsO*r9a;K>%#QW&j+q5>E zSIqz!yke;ExMiu!5@CRx!s*eH-6eNKPczn#-Ey3%krkO#(Mk4U11T;p1^}8!maFEp3oATgG%cL6k-&+iNgoyyouv6 zEhp^+iYsm(Yqy1P?#F>Uxe@H#CC?QfuKsXt@CIn-hSGkw4xfC2Gh+mLlzQV;AJKa< zG?QU>=GXm|6SUM+;8OAl&+)!ojRrn|fSp**Yi{UiW5=e!%+`7qkZPbd{kSgfTnZsv z2G53z*ZV+0`%T(q<-;r0O3t7TqBxvAnp!$8QRS88i6A<_o-~`E z8HR4NoS1TVni*tpjjeCYA5pDT6GeC_w{jWSiE*bK@IAR+n$Do8m^512v|gVc>Y(1@ z^&gGb)t%kBR4rB3mZb0g;C-!DW*P5$SuyZU+6!f*Ji^>l@*DekILA_J?z&B3=Jc_m zPghE7^LogA=op5-95uW4kJ+ZIcwApWIvfm zxv$+|y`XO82|qRO8ZMGh4OYRX-;Q~z2CfianQ6avn-01-*J7}hGoCb`wzc@40|HSw zu78vIXgN-xtije;XEO&~;+!!fpWA_Hub^*`eG@i$;?K;1;PPP$XFyRbrJ9N9!`^+w zWk>btkIa~#R_%^ylkDl5)muYFO}S?JNOgi#@cszM87CW9&eZZM){729^X_3285PG! zcJQadBso0na5-UcNObqfToD-`*?xGQ3A&j(baTeh%&@oLn%4f*^?>_~aGLq_W32&I zUAZ?)|AW%P$sTJzi}W)egx%voR-38zD-E%X4p`zNOr#(k7Nn@m0K#RDsthF?KBdRd z7|ILSmbIcL{zNLT0AF}a0yu+yBr$>1n`cbVPB%1=DW*ds%W22G%AvmcdGj#h9`B~O z<6n@S-}pYN-OcT=z3e<)?m*+}CylPUULuF-Aj)|E<*Qjm;==XKbb~g+yC&H8jdZPi zU+PJq%~uD^$ArSa-VCA#HwY){^Y=zUT3x-H9KvfEw)qrr5D@+GMDl{F=szLFPZJm; zHO7kL6P19or?&JfJfSwN9#y*|VwP}YSe6e#veraRhZyFA6&cbYMPCpnE}{IZuiQvJ zUnc8sdX*j?sJdop`cieqGTDtg)-JaP^hxCNeBNV4e{voAw$~1a=sa&5079p5U13ys z(StwHf|lUJyzA9;wj6AYfBZFk?u@dBaTDzVvpC$qfj1{SvK;B=gyFM@lAgZ_;=@jO zlpztq#Yt+<=YGs$hwpK|K-4F}m9p>e zS`1a&L=CTfA794TnII%tR;^Tw$=?t1U{ZB*14T!s<}-2%eVilG)cWze#z5XjGRM1Q zM(R~O1mo-GXHe~H-d~UJpN%JEJfq&=dwBr9RM&b!sn$evsc$!ehAlEVM2H7Y-`l8%d=cCE!u+EAd3)T=+|`4& z$wY5wn%^O9os*TymfTANGKZ3S(v*WblQ@#?w#c(3Tg<14{t9cIkX)JTd2;{7W9o;4 z`ZbETGp0)PWhvYCuNt{%*CM?l^W;+0zBwM8T{J#qP2qKvy@*`Qny4;L<^J()TYhVD zt&UPDoily*!mn!MpfVvX2}jj3W23PXHzn9uzVN$-gQ1++Y?cmvNi<8&R-L+_043}Ar^y-xdh>8KP$5C;479(EW z5gO^eo+4elp*Bp;DpbbA2kKb6^uNsw{ckE>JS+_w#=*%|B7%>c!X-|P!9qRgsV$s_Tn*G zVC|ci{Q{M>}Ndim}Iki9j~47qEge) zb`rUd2Y;l48ZJ)64n~eCOpiztYk&PnVt}7N%cmxtJZbau&-{u+%h1xnpbaOa<$BL- z2}lk);Po)rcqH`;mYa?X7eGvKd--%u+^X$6h9d1{R|3V^t8_0RQwjeFC91)GM27=h}(qsNdT?R<;Nt1}_iy^#ANShha zQUfAv()-GRu#tLt2y+K^D`}S&IDtkRN2lfc+qcd4`zHn^9!~fn9Wu#o32^$q?iTVi ztm{I1?M~|0OJ{#?LF|?}8dI-jK!0o|-}@%zxrml?8s&N6*0kFb+S#Ig1Bb^bN(S=} zZ-v*XAi0k_=*xJF$`AALE@gpD=%Kv%JUV_hF};=b%cBr`;s0mfc>((lvypFm@zV2! zjV1+QuY!|U6ZHbNvtUjz`}451GCnMYbP#wiA00HmmN!^8Im1Aide+6$&_i}4C6&cs zD`(zO`nkYS-GiM%A*-v>C0Elk_JyqGUZbbQB7$47f!kFCqjnvY=vRI^AT<%!FVD+h z6F~^+>2|c&IW#H4Mi6!L{VpghzP~xbR!Yg+*t~D-<1{G zobP)y`PYH_xfp-)*IQO1&QA^E<}<}K!Y?kJCko(c6=7K)+^$ExKRqP*Txo$f+^<|^ z@i{!FhfFe>?jFTYg%%Q=MrU*xVk~QorgowX^mJ)w>orZIZ^O0r3$}<`=x>f%q+{#k zzb@;l9yWSt{xe7Zs}PKbK@J-;ju!u&lZA4>Zp@IP-?yFg{AN2`S4ymgxPqzX2X#Mp z+@fw%;=Hs+A!3Ac*M_^x<4X~GR>|C6+F#d9osP*86(24NfKm(c?gIV%g>2i17WgM` zc*YoU0=)*+?*!w>;cv^)M=-4D28+41C2};J0*D-&Dhk8{?oi(W@9jZ3qpBxKz5ViA zbm$cf1eNewY1-;qRmXkH=|=i6=^I!iWj+s^MAXgtyCz#<|ES>Fg;7lbgOn{r?psfR z?P1RW28LD71MuOe76TO`>gLbH?o)B-1>6Uqs~zh?`r!){aVn$^A0Q2A~lAX_D-=-O-RD$zUV%n zSmoFZ%k0W`eersWO$LB=$zMsH%km31pW#sFE)@12q;1|%Ts%ZKtW_Mgjwri*5^G^< z8zP2uE%t6i^1jFHMMaAjdAvXSUO>O)NH{xLtV3?vpyy{YoQk!DC@QW$+`0I&NVO`H zDyrhQD$ePU+Q4t2KrX^3YBc?mB)chWI9zC9_I+Lfyl0=}Bxud#!0Lr^q*0>(8FCf0PsvRQ7^MbW%;a`r5&s=qPCs z^((0y?CX6ya0NF7dC%l|P2RQ5Pi2@=P=vdsSDFma?B;t1F~J*3m)HNABz9oz+0=`t zwNX^eGTWPdgg0AZH!rQ4I9%~j_!>u*?=OaYz&KUNXCxlXQ@bNbDzE8 zM@g>kkyDI9-8ZY5J~1uvkY$7D8EToFsF#HF`**{DCUotZA`5pi-kCDD@L z+B>G}B(njKv6gv7Gb`%auVnz^6OmE~2u}2~E z>1Md*z#`Zv%+cqNjp)&HApTQ}&}6c|ZXGpR?`z?pUmQiMk^pvQtA|blLUnb2zBgeH z*;Ra~ANe5|eK1^5gBTm>{L%*Zhy$oy*zbT`Bs1TO=IE_;BnL(ypS8N+&_~y^81GIp zHeF}Mx?ivPrcNxoOUE;4WOZ^m9nu7T6)Txb6e**zZatdMJzz3*UFCW`51>@Cr;TY; z3(3OMeU?8@^wxp7nH#Ot3S+DHaC4)|k!lx8uiE^gx#iz-a4Dxf?+4#@Lt7sx7;DV3 z2MR!C>6#9O+wmqD-*9=8JN2?5252av3~g{c+(KG@$qyD?cDxAFVa-_aE7uqDCc#$KQrdf^~E~=9_KQA?@8h-0JSn%%j zJSK9_fnOg7j=*J;x|s4b=)&S4&>*dShaUl+1eZ!P=SS1sC0!Z~1|7sbN7ln%8{kmh zObRqt75b<{g2Uf3{Q}K`aR`KKe%iDHccw9|e>Do*D&={4#Y}SSJDRQM=V8>>=-?Ih z0*Imx`^nJ(xKxLa_=4m^oo3nQ8Aj z5?hanZVXqv(})a`=O%g&b|zy@ki1D8L5<$o5nreb>M>~619@VTy3NaG^W~cJPPYon zE^3&KC5Q_VcxP6(mgEu8_Z|s~nZ2*{^HsDg%wFSn+{4K@yWcI!&*nD4h0b3pP^K{% z700T)he@1ARYEi#=zD#EB>!MyA@#(! z?2buF>yu=;p8r8ejP()JcyCvO>0?*z;57563IA9a(G=J7^xy@n;gRBAoH$@J_PkpL;`AeB*yQAdE*>gegB*B=?Oh9=<5mChqud18 zQWrq&)J!{H4p{KS#&qgGX7ss=DxD7(sgsyZ;>R#4UkqKIMe<8Al1qFcf1$6Ikf0l{ z>kL8AoYC_ubAmFUcj5Vmn1eA49_rkxuyd1a2`BR98HcoLBPd!bCtOE#&1U2-kLI%d zl?qx*Ji6_!xe2;#%A*BEBr`=m0oAP&GndJ3EN6{%+5m#E>hgmH6oStR8GH?{HvG&p zU3RydeWzo0(1Ra#z?HRJQxxUblWI8VM?Tc!68(dUK09a}191{NST1e{-`|LU3RnU` z%kg1r4hFE(RU;%xAI%zcVuQuWNiCUR(+OBiqMcSe@g2urBaP7O?2_DF5{EeVs?!Da z1{o3i&;W9HlRDDt2NeU@7R{}$;5Q7)B=b(PzVj{eh0`IY2xqz&&Z1xa);3+vF`eha zF1_5Hqr_0MbhKAxU+LsBQ|b4oh0Y()_(er#jU15jAFNxezPFU=b(G>&ZA%kx zdBHDtS#}y2lxlP4>OFdhSUvxl{hk`~^^h0tz`tNJ4%->ec=crkDxA&-*9?CgE<01Ieu_)tOfD@8ta#Z!7}fp7wW1>}KCm$TM%tJYWvGpnSz-28<2jQnT<- zg=2e{`|HqWBmU?H>{d5fE%yLiF9G)EY|r6U<2tm0mI-v5peX_t~aAGsZ!S4|W zUQ)>$UGw&VXNQ@<<+PbA(gtU#KdnS+W{h4nLZ-L=9fU)>8IFPQK{eGuaXE2a>Rl{l z>1-?>br}d;UL;XraB82xV9i$k_T|Z_jmOpghlS^Hl9Q#PNSv;#zB~$*Ig=KPvv7BB z-)(*gC*u4B;8K#XNo5t+neGoqEs9$2KZ&6b0O!mXLQ*hwCY-dDzpK}md>Yr{FQWa1 zGuLC;TW46bWQKV%{_5;S7zJJIp4`o?Y|aQ7?PT;Z!(Jy`{+^sSTj9NJ@1R3%019#{w~6K4en%jZ^TE}}L-iuip!<~t%>NLIQn?j9sKDsucG#XX!2OP;fqwRus0%gnFE3M8>Ik-mz}1vGqT z#!xo82Ech$;#{G-JqLD|SkoX=S$t+4kBjJOy|QUZLp-I&JBg(`S>sJLJ%AqkR>a!a ze9y;d9^d+d*ZDibm~W4>ExBCRjmYQ`M#3jSdalO7r!=()A*GRxeX_nE!0)2Ox6P zxr~H*x#!$}j=H8#9*7iM)_Ny(1Yl3@3eIhZv2-sL7Y^IC5@rudpqe{Ma}n|S3Zkj2 zd&?BisYK6J*}WzxZg>3hFS{jR8~HVAWT?XM^UB!Gwq)wxbsdQ>dz{{fruvxBp+**1 z9Nt~#oUx7GP`Y%Q%BM3O>z&QNj?^1jtY6~1|5$<+?sdAG$DJQAAAYiG|6I!0BR&VT zOcNJdBaeYo+5TW$>wd(=<-QSwp}L^ZV&4y+^2L5vAP8q6zvMkKF*k)+2zGotlhrJ4 zp-i+uOdxACRf^HszP-fRzC%FtTZQb?;Krz^eE1j9NnrRio6`Klr*&j#fv?>IYQ<|G zn;(h9-4MrBG;gQPA8`!;mxc8c`?NEWHQpv6{Uv$cd)wc14b7n^x$G zlaPhVYYqj3IllR}6+bHvJ{(NW; zB%U?nZkb>w(RNF>D=Xne3?MY6);?RcuMO3^e%N{sSV*n!qy4`!)@5Z~Acz7g`Q)5+ z>&A6makpsugQp6o{j1MaS<5Lu;~Z}KB0twfAq(WL#|W~JEZz&%4?5a!GGgU^Wg9QBv@{N^cyn^VOLqP;uXE?d$c3dRQ!$NLz6LwOLKQVw@_Ew$H39`DQf6k+LUe;ZB{yuh!6w+xJ(5w)=HJF|V~N zC-v1sK~vn9Xv$@?A=d}b5a)6_X*aTob@(2fMF$xZfwDd$gvTFMyW8|?i9f*+<^49T zD~Aup@g8BechqRTE<^UrIBOrf^YrR??~rh|JgW}SSJJzWnd~Yyhsq}DGi(J5BHc$d z9!8+8y@bQyA+RG>{mhPmOg*Z|K>p6$NbIsHPkGgGyB~;~QOm{AjqX&b?JeEW;&4eV9Q=N8X3xvO*s}I-0XQi`SFQE4_oD_-YdP%uFu7F!7$IeK6&vi{gxwQ z5U9j}mE$Onxd*C?aN)Fca--bR`XM5{b+>1I5v;@AtQ5NXZso9uy1~s9!Fpq>dQxUr zZ!~eND6Z1pK>j)n>49@4vACo0ZT*VIyY=Z$w=-UxbBr!Pw0L~&@XQWLbvUg_WZSJ~ zC~TxaPvBy>(qc)u_TXrk2tnf7zPW?Dw0p81vcoPcx-q`)_Pui{0V6+P{wpkE2fhyy zT}kD9+9Z4GEMRV&eRA8J`}{UzkD%}X$RO>!v51k`w$y&d@Lgu(rOPBUR;V>D3UT^f z=7BAUrG+((Tzs>1uB4YRa!i)v*cmRla@FUPXGf4lF+MGzRgzrYJFUYwR|$F)mQG#0ICcH4@-9 zVY)k5e_GJvWvz~98tq%E?H3wO5vyU=Q&@y{C z$9N#wrJPyj9HWnh&kmiV;Ty!CRpktd+zS!>H2qa}Q{n@k+C zuFKZ6Ms^S^N*sD<=mb=_%O=%~Uq^IL3tSyuFMGFbN9W1a3ukDK@?)cHbCe1PWz`H! z&C;BH5L@Bjm=zym9{+N~>RbL9UhnR+9WMNuu5{nielr&T{Ze4+jILuuHW+^P^Eb?5 zlI$<~@Y}9mSW4%etKk|+zOX$?y2({W)@|e1$l#UTy!_f>W(3!mX-g+B$}}jK>d_GT zDEAtH=qR^#rGy39FjTnFpAOnVJ7W1|s>{kVjeapK$Mfwub_#XT)cCB=i=G&4EKsq1UPI-StY|VXo~cYkcAp zQe?5!YN)^!kkbU1(LG@Ot4t2M8bi`ob;!j_A-Nd?bi8A5R6+VUZB;;Hdk|XXWr^uGPm3PDT zxrpXFy6D$doX^$UL);=oGrF}BxBNS>;tWgAm&n+Uh|0v+zGa0{(R>+PK@7wy-lC1nP<>k47? zlG9V+5|R~r7yvKn4>Ww1YL(1J!Hk-p4n#DVo9!^Vs1LWPc3E2 zXxhDV6`2TZW6Q!r*23}6SUC6^T|aQam>2T?**P}TreK|+7KOUy-H}Hv(XzLBX~0u) z&9$BV2GdSkNHXIIoD1uZ7|h9Wm|?IhA83{5Zuxwxb%o1eOJ=OtS?dCf47*v>B_7lr0!kzC-U8VaGx>gtV2WTloof-&AQQ94dT}!|>ejwr$oSG{( z_V8NX>Xy6Oyu6*JJE#D%Y^bl;g4TxEgbdU$GV0t9I_~kCvJ(00t?ozk`U6a(p1SB$~FhWjji5YL1^Cp|$1XbjRb&q_wp74tvW}jhZ@2{k@r4b@f-BW=GL! zcu}*M*8)43Cii%w{xvoE0w`6jCl|ax&C>5Yn==x98_BK91XL@O>n)s>!hE_NnuWw6 zcrf#stgyG{Ac7=wjjA}VG!2>19Qt_f%b9=4=JG0=&5dNt2W80f3a;@d?2v288?&Ku zyY7XQoJn1~^Y>D~nMEXYI?qX?)>8ZYM}-C?f9FdqD3PmXZ{DKHmrS|gPDBG3KO+%V zm7w3j^cii}j`lhhZNE*R`OcbmM7EA=o@ilh{>Xcc zUa9N@QKMm@Lw*Hm&`TEwW@ir)1m7~Rl~JeTl__fFL<#i5mfX;Ua2&Ny@$rI&leFEA z*ZtpqN?CU#;#`*&&*lf@TXAgwHNhVNe=v&I&Hi?)Xw!b5kpo)n5{OFX0MSYAQtQLL ziF_f#dLA1h1q=VP$x*G1;qvH@{o;d{-i_v~AJQze<=veBG(-;v(6#9@+(abMfA zb=~~3Ow5=Nh={)`T^Y2s3tn3k-4nIyrv|bgvqJggd*%#i2r?Sp;?*?t0MKV@d{Dqs z>ukwOO;EA=F;}@nL9dpXC~v>4QSOlAF;RQ_hnfas)2~FOkpIAWm|2 zRFovw~j&IfOt~~nnwNUdM|>*K5Aj}l4v>St@EUb!t&_bh5KNvH~<&= z3rO(sIS)7~Fx z%r2!0uU=-wqW3De+%!!sp$~}Z7L@e-It3x_XJ3O0(?C2?| zrV2#e(4yH)N8t8mN}U|8C;YyIV{9hM2n?3Fcxw=W4p8~9ex|i$SVVYNIBYXFEcac^ z+@Sh;8z#G{PUc9{8@)%GF-kb&rMW?XXP+VAZ~2GHq17udv;-hi5zsi;9qiQ-CF$&& zT=J&0-vrW|=4x_?Y-|+P+2(Jy8MTZLZfNf`gfamE9P|h=+1;Mdrlvw-c0OP*ALL3U zu=1~|7%qz-2|JsjWH;x)qI8{lEOHqgQ55AX+VSH7{a_pvxRtp>_7vfh$KH$s0vIZK zo)iU&ITPK$0it@}W<+EuP%?{GO!A2==R%wp*_Ek-{o;Jl1$(-mb=%;m67gDV)v z_8O>Kaa&Jr58ae5&p)jVDSVa5mc7rv_NfM-v96+kV*RMG3REu8mIRKopMqz^TkjUW0tY&}I@R7QW(ki=UN@`fCqKnv zQ7%GWLv>IAP^)i)wlaG4z~#0^Q*a`ou~HQo{K3~#LF-_b)p2cZwVZCo8kx5eV6b40 zhykhFZ#IZ96>_~?2XM(VeLDak#*GT-mi8tegv`hM16whzAUoKdS+McrNQ^=f-^vG~ zB6Z!-94SFY>39&`I<-tBV8;3mxhe*?>6hbUXcgB^o1<6*Sid(GX z=h9ZUV zWi~m`{0dH@dcOMACeqP@+Vo*!%R zLoU-Mj0YsDHLEE^dP0AVBY%!T6D;U#y<+VDg@3_scdzBTt9bnx;^XsIJ!w?$5X z3f)1*I4gZPnM=v}Xl^y`YbywgPEjP02#@T8}^4TT;~56r>Ll5*koi)TnS0#DodtqVgidn%^010BMfy#LN*Q^EeTN2F6>p7XA^x4g6re?>7e}rch4B?}N>KiaNxkj2@ z{6)UQnSc|3Z0J{cVgys(c6I=$C9GS`7^PgIt9RaEYV>n`nl_$MNiz)OxA|qijHCa8 zl?n-kLP!D%0(an*h1R#VS@9t;lB<%H{G9tJ`1T$|GaD2JeSAN#L(f#XB3SSzvo`EB zXxQn{Naot7>MGMC8;I8HS?Lvt6>*Nb^&VCLJeIb2`f5dBcoY&WtBK-j`kkuw?>2^| zZpV)~zWm#mAQCH1D;a}$S(T4-C#=3)V?1e4xoGW8oM&MW<4`qJggtcCOZh~{vN&60 zLHuD$ahIx>@aGi$k>ts+_7STat+)c^B)>U$g!=aw=C+V>%7mxOG0R@6Cm%lG8*|_uw-~*6O!7fDS zp+9w;1XqqA^p@M4{?K7m31Ik5AoSZGV^*21H41Um{L}sFSYdOJFMAtQT~dgpceH$Z zOpF~F_x?8rNDA9av^rPwcT68L>w>B3W7fE8wCxi~xxrh}IGmvP^mGMk`+KD9RCVq1Eu1Dy|6PRu$ z5^$E8poP%d!nZ*{X}q0&d3~3Mrj_$SyTW7;D~n~YEJfmrA8`v5k^*6uNsUmx%gq?+ z<13>oGnO7ZE0=rjaz130?SuhgU6^H(cTK0c$s6vYnuapBbN2K7XW^3~dAF>)o^jzDk48Uu7^c>iEo&i#u>1+aK7ZKD2jX&vV!f4;c#_M{E&Y zv3;QvBGf#Om}K;Q{b!>YVf21%NHAF%^;>dJ0d1fM=&g@gBjz+txdF^xPeaTA+3 z9Oe*lNgq%*-RECHC6`1bTL($1a_`KzIeuoG8EXYC8r(g#>)@y}W@_d4t;pb!d!k7r z>yvK3VPk@2{Jskkv+9Y-t3hXc-a$shyAJ*BK-)ZzG6+o!SR0)!WJX9ZcB)f=X70wCQEqIj_8^1e&HU z)k1PzmJ$~(Q_yAR0xnK$XB$^g6c)&L57cd!&K~g|4bh^%_iHnkdY#w)LL^l^YbLzP-q^WXpRYvA!#l9f_`iX1n!fU>cvzQz+2w|lM*am58`<;#5B`^=DSSxlny7|yuDzdu8M1{&V(ci&PE45EIe zK=82*(1CSuij#tbl_&YZZbm8AV+ac()g{IA>B?6MR+XgxOw`DL6Uo_KJ6pbl#fz}{ zAp``}CZbN2e55GdWPC~V)mBH`3L;U@O5saDYtir%J;Kinc;CX67*1$m6b6a~EypjD zrIGqRPY6EnO_GmUyp!#XFrx2S>n{-UyyM9cp_Bz4E-J6+zA)Z0&9Xs6Fz`$Z)70n; z~@P$!4k&8?3aTkXyBIhW1Xi;QNm*UfWyJw`*joF*z-rduCMOGz6L zXJwwq#l!oDIZVfOsYweZVM0a|Vo@$(RZO&(9ngg5Fbl)KGBWWW!k^xj?P;_%CF@w7 zd)IF-0Ehi{=aJTBb*klo z_XQT~n%GgRwN!&c#V9of=E`_b^d4Y8%wbA@kwjEyu!-pZQ=SbC0%E~qPkzOVkUB$` ztPWAt=XWmya&sh+D7GU+QSiO*sH5f|t_y_>WIKW4RDM z`=m9ZeTStzkb05Na5jMcIj_|5T5-xLK+CtIRQ;Wx;!CW-1ta@Ey^CH#e+S=YVDUEO z_#QUcQN%{)zR>XdsJ<@a1Id0yzPUJaf1nOjUqVjp@C4Xn$ zfC9#EjxgH1)zVB))%3f{3G9?W>cSE#O3x*bqV*|!5NM1<^A<24!slPHlBYh4HA1}p zQ@bDvO~_R4>llktN!sh>X!$}l4^72fiO5g;1HqBMixL4pr?7*>9vSGKY%NIlBEue~ zJWV#Z1jJ%ZjKqyj()*BcZ~I*&M!~uY2Ag=CpyDL$xRKh&h4h?%_qaa|2)NYV{MEmx zi_1mXW_(ARlaTVlo=JC8+4Y{{gqZRhLwyjMJ*Bi=iT+?W(C#-sJ@k1zS@y10%s`D{ zfu#TWuoZc01*E4gMYcC zfP|Ek%@d8#`TBhl4EHyt04x0&*i>h-=52;<9x}fR>i>s_euIIbU{6KkbEfo_>!kai zX8SLL|HnU5FaYvy*w69*zaTWF~Q4Y@!`ukqq}^>MxD;PxJkcmp-EMFBVvyyidMsI}+4u9MUCle?6dc z?GDiS^utd)yl=PCbo1`{Q68Z4jo-mf#{5m)<9~XG1RfL;u>e!v)00-IY^r9($DxF` zbS37AH2yl+MOJ9n$_J;Cq?a{z%#&}n_DXdq-wVWw@fzry<(1xWTYC|0dFzdUsgU8 zSdb_VsQ?>`Tsl|QD;;b?aWzKT;D!&Uq5uRj0L)0fn4Ze-fkkZgANK!mmkh-MayD0v zad*1ZN7C$G?4Ale6+EbXbWrbbR;xIgs?d8bC25M`2sDQOi?Z~GAh1AzMf9br+4Pf? z3pPOJ@oiIk2{DZ0aE(!KlxI@^&XTW*09x>7%t-g|3`gPxB!)BP3uUZ-XYzl$ME`yB zlvL0_&2xv9`7eJ#Kz&lY0CXqX3%ZcMv1dqpZcqU0!+((w|96%Q?eh!Zu<#;Q?6`nYiUVkD}fFU)>d7hBz=c({Y`mR+u@Z3 z>a{dAox!YLp8Q~jQSZf)ppeVs^Aq(f-K92NPPR0#^%sw$xc<&OCjT%U+fyT-^QM`6 z0Cl#oM9SiDcVWL&Jxf*ZA^*06No(qrklO2g~>hQ6pVet7&TRKLS%?~p4K#SOIWv4d9Ao^!hB zg5^4$)r-1n*ryPguabns`?tRR|L_XH?WYh6n|OwJ@#6UzA5D$vzxiYjQeQ1%bmR9D z3s1EM$G}Z2g*mzeIZC;>NVsrJNVxb-NJ!K|N|6`5r6CE1yiZ?g8sV-zK3rb%xbC>` zVDr>3okl?9NFA_E;Xt$$uaJEeP1MGSiEOuL6)ZZHaXIOs$ z$W__Au8z`^+1|RIzxj|GiO&H#Wt=R;$-U!=e`Wpt?>u2%LaEPfuFJyMw0T+*iJSh- zr9wb~vEim%4rWDg#4CmT3t#g8jzk|aLmHS4{~6-{7elF;!+gz+sOIxGgbNsw&lqJ} z2NUt_7Y=>?cM(}^e_25Pm=?-w7=MmbG_3=;m#JTh{-TKg_reJ7py?aLM|gJP|Mk!O zlbu}u!7C!P^*H);^yQnhN6Cu^8?yCVi}yasgHXEYkf>y6VxhvQWX-au%Ry48nWESr zR^8~Hz*dtl{}?#PE7W4>FYQf(u(nDdAvS8UcI^J`8-4L~#C|7KLR$zqiN7xDN;vt` z8mgH4d3P*;d8Gi>I3h@!+@C|gw0fKVFDU`nC0JgNV`SR1-KkS=+aVygZHPWW#}v6i z$5hVf9lFeZg07Ra<=v(67||e! z$YptP)aB4k5wdW+Pk*QgtVYPCmz)BBX$r16uo4QhHobbg@9ftB%?d*Z8JTIUZ>P5N z{gO@%%h@epw^d>laSp@|vNQc+@WCGjRae_5WeqGc!$-qg*)9MQ?YoB3+miyA#o=VKVHJ-BO^DEu}kT{H;w zqu|KcT4PVio6cmp-RkN&r{gED`39@JgI}bJ*;QtWZcD}&RkTVJc?i3cuBkIRT}89r z8IRCNgWuoe1W!IcUDx1RjV`~ODzLB##=GQ8;x>CrDV5NMaDKIM)O=rM% z+_2~5UBSw!^-_Z@VscH+K8N+#_b0vr(ee)OZQLk?2Ls#n9>)1<({v5XJ4Z7bl>wjx zmV>%3qE)xm;60;S4TjiPwC*PFNa`o>iSj7G>NPfN!7=dZN2PIwKkT}P6j^6Gb_)6X zPzlU|*huj*v|tH3#;yp=0WYaYGO>$y1q-gpHI`Y>$V3CChtZaa!<;^QgY;@;Ii{m& z3Jp~Sv`&Q>CQr<36e#QcA^CZ=#@}00xxM0mu04`@qbWHN>s`4(9$WGCY3Tb1qTzS= zd91VVBC+&m#j9vM>0o$0u2NJ>pL#x`@>~VRFv~NY&nHJ$ZZnOQuXp%gwlb-W$kXU9 zMod$_+N*!dR+Um^KC99q01%PM63*9v{m*i!A!n}h@@`? zF<+oxV1pJf@j1G0&d>=3p3wPRZ-OTBq#3t7&1l^!vb|d^9tzZ4B|T0~6dF8l3yZD8 zKA0v7r%pSimz9H&4%?yBdm07of0`+>^47%y8^Lbtb)mT4J@WU&w;zM)#i z4t#}xI(n^yvh5;uAuQ)xS2aB>78uAkrpz?22!NWYEKF$=51sS=Nt5yKq*)&yfF|$g zU5Zb2HZxY4TP55B3`hq*gzW^vjZ9|0+gZ#%96Gx#)u<@#EtXf~ZXNbVX7%Q4=ZqR( zs0?mJGO3=vi}==)%x}bWiO;D`6*y+PSy4|qJW%HN#4^`nggsrlvdAFdQhHG3xGHl# zKiR(mdY09&$H-jqzF!J;f7oFPK8@0)dIGcH>pR$QAMp*uFt#EP@|d%zl-?xD%BUSn zDw&M^G%(7hp7Sl|I7$aSWyqvsd|v^91QXxn=2aJAyG0MN%4Ha)kJq0IC{@#8ewsRg zFj*A5-0Y%#gTA+vV}x%$S@<>aIfN6o%k_+GZ&wyqDQ`5Fw6k{tKu<2cXV$Gk3B%{> zIy9`9OxSr!R=+HYQwKp%*M*1e}FCTYl`8z=T4pXM9Y+Zs& z?ibj?9n~hNIZ{|TyV1Q#vn?rtA1AWiJh)HSC2RXh6mi+?QDPZ%yR0}pUZb|LU6|eC z*{Ad%)6A4;$hgL(Xm2y$jb&2Of5g~7KZ+SiX89$71FnSYf(QckLr|Ixmh60Y>k;}e zBu%49l;mDq&IuSEuR=(8Zq+oOZr@ByQWZTw+nmMfpcj$;7DM4;V8-y_aJgmK!a~`i<%b~Sf}^t;HhiPDElh|t z2d#LTp}(Yfm^r3d*Vt1js%0sOdlwn@;%D!r0`JtMcdX6VHRy*>h0KAuV0xSOiHmzb z=qa!_qxo!u2KYU+Aunv)F8tNa`Dm*@W~Z+IHr&x+?Ml>)gMP_=8d%Qt?tqJSPE37G z6xOU~$}WaX?2UHPh}jT3IakWuC2Al=nejK3xmPeK(!y_d!~ClA2;4%X2KDG#RCZAF zw-q3Jk02m%SuUB8dvNI?%8%Yv?%#J6z` z&wJD1S?pFM=Xg3{yh#~BiU|Oao8?<7k5`^$w_X;Bd_|}H=G5wbiS3KY#X0m!v({23 zb-os^Jb*QEoc?Jg9zAao9J&6H{#18wDkIh^q}Fmt=o0@`K&L%^SG|wpQ~0A#@!;T) z#p>FoB2E}L0cyI@Lph1Z?ipqwJyab|NUUEDyZX@m0YdSoiUsHCKX4!j=?w-c(~A<+!F!!=BHw=HdmoanQlT^B?l(2-2o2b(ZXzK@bQnl zp)^Ks%V&YfQTGb#VZ8DhXZZtPbe#u@+9qk1P-TWR0lm8%snjI?{=`Yj2V))1>D%R{ z#;II@-OfRAEnCv;q%V@WD2YEFpG%b}Q^Hca&D7SL;+IQ%qDuu932e=SQ}I=`Fbusg%NdDY^Gnnx0)qD zV-gx~dNfA-`mW=GXKfRDkVJjxaK50Vy1=^3V#ZoVoyPlk$>mj`_!}hnM&GPLg`)CB zRn)i-;e^Y()51nYH9DVwwa$6PAYee&940lDOFOdEiEDruP~f1$<2JTQL{AY?CqAjd4v-2 zY0Q-q+-!i3`N2>eJ9>#O(S=YOIy@!Jc&{8#OydT%R_|B{@1JL92-?61=(L2{IUY@% zY^6}2fX@cn$D&>mzZJfMP*=hqtK|v9<mnVEzhm6VaZ8m3s zT0S*{+20AAwL%>IG-rZmgvxUb1*W9mn*SLo*v*g7GX;rpv2{6mQ{R;v``1xaS=ywhP6fqcP#>vaps88jl!K0S>oJX5 zd|!3n-7lMohtokn89sGmpNm$*y(esq7TR!32 zmk7drwBz!+YeWI$dDLQml=Eqe4Gra|7f^(p<{MaGUM zVt7E3Tt99PCZjGc=fmFc1Hx&fMT+$_@4Q3xGFKK)BuDy({9BaW({+LFccHhNjFq}D2WakG|S6GSxXDJak2V}oa{~H#)09K=Fjfe@S1f*0wy#n z48#PS@tC~Gp_WR<- z&~H;ih;JXy5ud;{XCu6%JTRZB$Vx%a!cFrgad_yGEL^F>TxxO)$$ak>;S4)idxy+_ zv)q(D($@W3<);rxN_KK_Wu3VV3ZyB~A>*;oq4WL}+z1%zILjaTm2)Gbd!@Q9GF}1C zMyE?O3_`=|;I_O}?Bd%~Cw-u%Z-HAd&i=aUv1neC6!7SU<9w<524x!uAZ%t z6;d@Sh6R6jy*hAf@Z2o)Ea;9PjyT&GU@C#rSYMkrV_m;G_$7a$-Bg-^hn>{l#)*=%Rmyye)nmWfWR~Z02W^r&fWT4(po29GNi_sKTFBS_- zH^-W zbXgH{Kf4wXwv7*+fIUs&SkU7VO#X(2eDKhb+PpQtmm?oYu@jxP9QOGoDp8d;iNMStnA9c=$YQao3V?wR+v`=GY zi(Fl!zg$0Bzf^YR@xm_yb@J78H{-GQSFL% z{6hY!hJPQFs>|8N2q`vZiQRN{>L>BzvYD3RiNbZjrv-R*(y~{)pLKWxX=%Og-a>_l z*)GF1%BAv4V>9cJQLF%&8L>Z=}y=li z@q@)^YWYvk01D!AkGMTu76EMlu8 z`upWbZYpcTC`D1oQ8c$7{%p@u1Vci+zGGrA3FrdCk)$s@CVv`a?XGr;Ph~6U>@9UV zd1zWZ-6Ip^4ovS$+F1I)?0ed(8pKaJ6kNNg>-Z*q;YIV9=fjV$ zP;N3>O|Xw|kOdHalxbdEn|^|Al!~R1dU=1zQj&B>eo&}fLPW^vsy>o218VLH$hkD^ zDZVv)s!_P2hlxz7spN5f5a$N1W)i3wh@ohgs}$GD<@7pNmUnwF=*%mJ?_NG73R$iq zLLfqwyQ=AHOUJ#1UFEZIdJ5N=dAQiGKac-#tt+%{T-kG(Z+b0fa&i`{r_>m}wCSXT zIwK9wMs7x(u&!?npDUmKWHjP-8i3Z5>UEDiFqhaxfVBWgw{3mUi%m3VjoCr2-qTIc zyVR(H&ChJ%NzjQJGH>yRO&z2Z5agN6>b9BVVu4ma!AN z&XY)lcyYIm1Rk5WF6vRtSmUzfwOhTF_@NWf0cb+IPEBsQyQ%+7QyW8&{BHyB{zi~J zE7v23E3bJ?hYsTxo|76v_12mwzRB!g=Vnp?2s%!CsR+1{`OP^kCEF=ym$iU~Y z{(y1nMYu*GlcJ6@5cDx@&mOUTZq*NZ_o_q-gWN<$D(wwrcL=eb=k;R{aCWCyCo79E z?0q+RIA1PO8qq0?2_>U?z0ce_c3WwhU~kVTnb6h^4QTw+OBK0wPV?G`1oj%MS8Mf zOat2er^kB?HdTiH-Ss-V)F`emhbxtVJ&%otB>vh0L;`a3@l3HRdJ*Z|>3k!M&mB%@Ao6Cw}W`(A3-Oy0gw z{*375E%vL6T->*l@`{z8?|oU8xTd)xcZ^^~n^%>WogX3Xl!@yY8Io7;6)uWoG%Dx! zSd`ORrm#0}_Q=FWMNu=mPS}6Ie{BU&l_-xSJlVUtqFK5%a{aUKTCT7fN#q7}zGrXD zWSYI)dlG}z`f_96w2`Osv93`s*Qgw4rfJ4%jzN#Y{k{5y)p&7+_Snp$-H|n`NVgrF zk{wuoMO;1CXH3uq;QivVYEr*Io3WyMl_;fMZJCB*X#6hAAQCae3dOt4S2(YNc{>Jsz$uarxiPDS<#B1I`ob-h%Dx>&OUgiKXk;}Yjb0BHw^ntyQ1Roq%}-n0 zCAqnPe)M7lMvWTH7O2anKX|{GhQ>Ol-AL5+Gb&Bf+x@#$Q}&TU_+EB9s|CP4nFgx_ z>|al*8DO6QA=!a0$Xn#V76*Xw>(inplbrCy#bY}m6Z3+HA=z8wJ%%Jco5-r2oV`Hu zu1Fw!mkTh=)awM*q93|W*3hegGMF2#*~Q)#YBRh;CY;iO^WlYUboQ*(vexi_rC!wO zYqGMe_rb@rewak3&dlAM8tnNR{*&!;DD~5Y5c@y4zJLbXRiFCd;ynPaFQ<^6sazHG zbkK0Q=`jRrxh2iQ8&3(nLXGAM-s?*Nc>cA|@-4Bh*xA=4Q8Ie0r7nyym4^W-VK&ndPIsC`NcfbSGr#P?%y4U2|kPU&mB! zwi4!ixJ+LQR<>g~LO{vHRUbZ^-Uaxd%mTaJcvo1vbPvUmEFGGj!OPI}-s(YLzNWw+__Rs# zQ;X9k3B%S$?C@^b|0l)H%lzP%r2UJD(Mxh)rq5$YytU^W9ipYs4%9`}!?_)<2x6Jl zzEFh{ZlRU#t<)J1?v0fO93WHs?gz?m^EFxRuI8Y2oR! zR#Gd6i{$mC8mxN3GQZ-+uL9JuocvtTe7UJ33H&2dTXx7nf*JAGLd6MhbznM3g@;lF zBz6$n3A+{tc8ws9n7#AIi;bb9LxSUlU^H39KeG(=4=2_p^n|AqSjq@J(yP!&0~Bwz zmLUSiT2t_be7u#F+dxYt;LBB!KSC!3iVoiQ*GZ-lcO4V$jlRB3N?zyAF_iqgUA{& zwf)w;?hkVy$Q&~I+kB$G8e~Q}qEQ$eW>W{MrfOporn3Q059yGsAB9M%Ta^jG|LAc&P|)bob*JN(D@*IL{8P2^m)chw zU065=HxQD~&=pSQGtXAUF(z$VIef&+7E2nN%jw(?o)i6O5RZt%a}0h_Y#7TuvGxk! z6&uScc30o${nSU3#A_8tAb54W%z$mLLZCiFqf?qxh4DgQBa&ao{B-bUB^xZ>?H7xt zPCuSH8Gj*X(Dzd7zS?YdT6@gr6PZ<#@`NT1mI(M0?{^zy+BsoQ9s}$NL5gvpf1MOl z%SA8q6^PThLmQ|0M;utzDpM>rH` zFI2I|f4p8H4zj|5?~7+zBPq3?+bdB1K{pWR%J{$FY-Pu?y3j zvPZM8v%tqfFM_pDtUX>-=^hX>SEun+;pNQkG{mQZ*Ws>-T%%&E;M8bxjJ+-HAV%*F z?Z6u=U^C6hA-`$9oJkgDc%!2^-hSDyWIDviFq+20e2m=1InW`;Uw%_oOzws!|1Dj> z>jSKUJ}Q~?76MerI~?gKSRuS(|2E$*njgC|NuerQIMvxD(>ql#?kt<+=~v@dt-@LBUYcH0l?sNQQma0Rc#h#4w&eN#3bb9FeCn zFTMJMQEVDu6#MX=p&TZ>rSiHl%*3vc5S&&it}Y^?9n`A7v>Obz&kKp!8cjnRa?eVP z9RA95;449v;fh3h@VT^AWy-Y3RvA?y0mvAjR_g4egAG{Ij1anAkpj=k`L^UpZ+7OJ z)2j}A-8=EnL^%d752vTI!0O#9;(wSBCY5mS3-P&WX_zF_VI7)EzwanoEnlYE;V>BR z+V6-7(gf58xSj4oZ-0@9T@8JmXo#vNXuQBVKi#L~gdrZcaVkS2ft$f)LfVngq9n*e7CJklvas;#` zKR19yT6?P`frCMhS1VWslbpaw?qJCf z+b#1xo5}g9H!!p()Mb&_&{@$6bD2ua_48)ZDE0QDJ)1W*xsXMYmKMIfh~3u(VN*z^ zD0^?aJT7s%pEK6(m^~r#&RTucR_a%S64#k6s#2WnA(9Z(>j(6| zx%j+T91G431c^RS?!A`Q%cYdD9kAT0HJ>z?qv*H^g*^4Rlb7jq&}>~3{$84|nPg14 z?sya@@**E2@m8Fv5^`USmoJeBMRU|RC$PQa(BYX^J6dTX8|v_R6Ch9lDbD)y=+T0N zoW@sTH zxekKJf9SF$?dD)XaRPNBRCaaG;E4^^&KqmstCj0mB8DTFxSy1Ik~syXiB~B##w7O%RN?U1w}276oCa&vgCO7!L)RC#Z*mQ$n^}1DCsBexeo6Wq zcFn5{sMZYQs^`MLF&Clp-7FD-;cERHAq`Jo_JNYKoTDhHw~1DZQ=!Q4%K+>PGaUtgr15`DyyT|iazvuaZXeC zkmNAM6b?{!Rd0f=5CsE#O`plin7yguEg2{fNgiY{pN7vc{6&w29G^HeVO}OE;&YdU zCpT$2YKb^xR6X`|c3!NGaJ~WP1Or@@QW*o#NNf|4;8=#`+re`)(~5hFUo}~YYu|pL zRBt>R6>=p%>w_@Vr+t{(p>pyH^KMOur)nh->Xf%n*9 z(NnJ9D8xx#UVTDzHF76eiUW5LS@(IMZ+z^f-~9e+v$&14o1V6xr(~N(t!=#^6Ndib zN#L$bcACiPzSi^2OT#RCAuSWcj+#n}LpD=H0(_|2N@qN9aTj32uKE*b$OXRs=e(qq%gN>sb zhrUkuaw7`w3iSAM&F9Pl^WB36FRX6wL+lyo1K+xKqzltvAJLY>QCj=4<}_9Wy+OFJ za7fIEK|s&_Va(U|%${>o%G$nja+;d{NB>Ji-}L!NKy7w(qul#wrXN&{o*wID)rtQqQ=tQ7 zj2J%LTfXRGq}bxXR4C{p;2H#Wdo%p0T(*@}$Wlj-KC1Gc=MwQ3PfII?_YQ`Z0*{ei z+@nQlbQaKdQN#B`_vxK4R1kihZ3FB@f>#X;R*Ut5<*$=ZBCT-DdttW^p1Gf^qKqL? zuQ($X3)PbJ+n?^gPE@M%u`Y*N>!#?pr|TNB(Wc7>sSG@{RT-KSZB*)h9J0lyx-2m( zco|+E?w(?rzAKj;!B=($))MR}L?t=Y7mNJVBv);f5)?uc+ z_p^trxT>(<;=SduL&HqdScMzCoZ7=I4ah;XN5A;uKRvN(@Tzb`5?GqV*)6Ea*%T

>5?nALj(>}qhDS0f>Axo_tKn*WoVi!gl0#AE`LkuB2b=Yq8E7)8 zQ&oK}#IRPMGI);GiDm0A|N1Q{lh(!@k4A z(BGwd%ucsfydGy-5DV5GPUArp_6NJ1RV85c81sFk=i8d0F#)tFGLXkb)~b(c191q< za8Hue#X40him+&k>-Aa)sTLi%RJ9PRE-e6mrha#<_(S?2QFs@GTFJA|n^P2ifD8`o zS|vzKS!UfBP>dY%{2Sp|4tG@~e`kfjRR_>lQ)&UjCpnTwDKjjV7D+`uZ{N#gXE_tl z0Yqu$fK5!p=@ci1PJ&eMQkXeOZ}DnT{4llsKf zKEtz4nGS#0Ka@?1s167O<07mRl`0v!|9M}pf`~q&erNc#xg9>tbFZGD8HCW{X?J^v zzJWc-{m=vxZ;K0Vd_;tOBXl6q#t#5`(@cl22VMo-y2|o$>LaPz_dw|Dr=8vt7>Gmr zT>aZhxNRo)`K3UbGrI%YD?K3F##CJr-*&?U>1;kNpvG?Ez(wy_M*bWm0GX z?vY`I(Z?0bcUl(tQAi-0CiHq@Uq0QMFz5+`+QUn)k@VllZnbEgIu};f^UW`H3Xr?+ zR3BWQc}=Bvc@j8~c=Aoa--hf1wOd{qbtB8~-YiY`#b$cjwBS~DsJT3PV|hdya?PzR zugfY2pX8gQQ&Q^gzq5CgoniLsE@B*U+)FQI?d0^%%Hhit7`JrVOAqD>>TUV1L8$wW z-eSZkxW}fZWek5;iXR|SESPKS`|XeLV5xq;+*l~`8Yp0X6Bo@#dv3a2^!ZCax(kZ@wTtLjCpRHq@hfheMfUO`m)J37oN;0AyvvPxv65QP7i}$nUGJ?L=-au zv`Jczsnw8R-(P!unpP*!RFN^Cd1#|pDjs!TQPo#lEpRi=0Da!gVNjv$4%_2Qql0rF z)E#3bznDHdab@UDgzmLqx&&e?U~>e+r^hf}r3Pw0O)7@VfN+}ba}Vc$k&Z--~#DTQQGnv(4F=5QArtGWYyo(DYvA-^mYr&0D5Q4 zpF2t^j@Y}Gy4;0WbHs4t$Dgur=Hg55F9OXe0HO2ebPiUxJa9EzRS}nUgGvY_Ofj7W z*`E1?2$ku(uLQsKu}KK$qQ<7i8r1m4jhar2EI!S*gs!cZx2m-NFf+xS-ldpqz->1e zp{GzK@bxi_<0wGA1FiAo3npBi%hf@VC_NQPaGf=TN7NgHLsQE8IqU=RmZEjdYuHjx z!z}^#V{9^^0e0z`_}nJI0M?EJ!CU`;mv-l^GAj$VJM#fd-4J1LB{exR?Yf|jgzXhc?yrata}?wHYt zv_Js6&T(ohZ+8FCc6aGyc(@_tbQgjealJ?M{dvw6pgeam!lkwrp~KYseCZv8ubQi; z-<(H&UE5Lv0#m%2*Ra7(%@~{?}0wB=xj#E6W5M^#p zOuL<9x%l8Ig$A9CdKMkmTe<@wRM0}_dZIz@EN;ROle7A9VhJ8xI4xhmp!3@e960_N z&z4V7o+_(>fh0@Po98Z7D(Cs@Kd*Co!L2fBUF+z*8a)FR>O*)$hHP*aA-XVc2TSNs z9oi^qMJ#kEJ)S2X6@o>tE|`F@R&ezDQSD--YU*)T34~Y7wZR&6^x~vVWrpIUsmFXJ zJF#^10n<5qmG)Iq;;YWxV`H0MO9(MaXw4nCfS)y+x4=JrZg8spHfg)0O593v7fo*4 z2w<*<(VX_eP&UHT-}A4v6Qz7Vtv>qk*O2S3?m7C8#+Vkg-_*mlAp2eO!!jo~vIE6V zMqYlOQ77QI)z1--tW1_GXKO)jgghbV)urMvi(Mo9kWS?^il^6|-!2#4A^8lZZNVr37RD}7$|S^k2(2dY3AN0OG+cxS@CVfC9X0+TQ6r;Jbgc5 zOz)HQ&Bl(hX+UXd{|(4-Q+$exmg$y&l+=c84)p&1RpT zk%JUS@wO|1Toj|QT+7jtq`3zwz-(I# zO-1(7*Fb{lWeZx*{`!1J2Zr-w2P4yLlOHH9dUjd_#&!vPUGPR1!q|M#BQ*U|)hG?W z%%JG8P&0AzxcI`Gd(-rd5|#%PYT&Ig@&CCxl2gfIs~hg<`v?}{(a3v{6(gz=+mp!l#>&v=L^*dr{^ z15tQ6w1wFH`+D>A?glL!WU%^mK+?=%dYQSM^{n`hnL4w~SmD;`RzDRgo3zRY>_q{0 z`w{|K%^7`Pb>4j`vrQUZ%p*V#V>14V_ySiu!jB1TkjoRIHGS#m3vvcL<5=@Br$GxW zI6>7y9L}!@&I#L&meDixhANzMva%<28q~^vEzFL}qKmnxvnlGoIc=X?&n8sfFKAR zw~VJ??--wO{^NS-w-m8F7VMzRNzgSXK$D~4PRXtXHOw_^qa0f5b9Vo#ezj%`VFI*= zWGMy}+dQfqZO=(bfj3Cr4X&r7oSN<@no=dNd)g9e1e zy)Hha==gPBL1W@k1y4R%?D1MqH5d;!V!*2u=T3xM|4^E7DD`YJRwL!cT6dBg;tttU z+$2E5nk;kU3e5wAMqsm}fItz4x7_3iDL9Y2rRPAzBftoCdFDYzgl@o z4u+2JPMJ!E$nNxxXu|m{e<;%pzrA?uJij$rrbKJsN}NTww90NIdg|+BIQZVtrChAR zo4i#@vrz=tdQ7}$&`*Zk(KI6@kK>276fdpRhpB}?f}N-66d^>3mq$xRxBFCo22bn3 zbUJQ&Z4RY(EMq+%6&?TBVRhLjr;oJ8LNZreF}0|BQAxf?RgSW=;sO z{Uz!EAkx>^)UQ}rgP*!p$;rVY{Wz(!m04`XN`nsi%VAiI+M+-lfRcb0n!e)0Ujq2c zsn9e(44z5qM&^z2(bJ4k$)v~v57F4|zNn)doQd2a2Od9zj-(7eFDal4sZj92L&^vC z)cu0m>#56b5HO9Gt5*q#A;TfJ5-Q}ji67q|`#7r*3*W=+HWE-WnLcyGKb6(;xHPZU zCpRG_eQ+VU*kgqRSuNLw0Wg%$y=V!DJN*-eam>P88Vx2@yEeS2$o}M5pCv=aB_K$o zRnNpaY15_ImZX=ImXYY-$+4n~HR4~hSu}39MQn51g~^;B%sW`fA3N-=7N%yT4>dvX zDM=vF-s#-+1@+}A%>f%iA?juuTjk{St0W|Fx#8H1Q^3;Azvs@P#Pui~#&^|efD_sy zWHH}@>a;(}0KtQuh_qjk9Qsoar=Gx_gvAvJn2u(|ZanV|6yRLiu56zd5K0pTf-xjD zBhkF?>C?DX_zFw#(_i*>`3FZ0W?|68Ol&y~(b)g$r!85`)g~%dC}loesF6#+By(qL zFpsBN`QW<6U#Son^yVnHWyCP+t}?mG*>COBPJpB)JUaXgK2$g`}h9cDXwKN-3acGvU#7q&O*h%J8QtqnhK5TkeAWg5VBgx-|aea3*GFqvLmbceDV%C-}VVLq*jk$>*{2mok6OCr&q>cBAqaL?_1WXjqZv^_q@E~>KKrVeBUK~u~%xAOFCg>6`y8B z=cMJ*4vBKHkUH<&iI$JgFm^rmJp=kwOZeTUp_b zPB3ydsTb^!Tcg#4!1~dd8@!^JCxL3-oaB$msL?-cx)e}MUM~_m+ZTC|lf>+|a$y4Z zj;bxM&lBB&tr6a#Pyw(TY&Nk!prgLzG#T(~y?*Dgr{R<;<(&lkyi}`Fus2!trp_Fb zZMNQ|#^#9xji2G$N=;jeKu;J(tr&A<;CM*MtOa=`_SjPxpeUFJ?HYCl*(4|e$){na z{ZPJl^EHMqDP(YT#RONV$wW`{YOT-kELjrwJGy$u^JrM}R>w>u72~KCj9?GxXC%TA ziP!kVaCs)lNopofGO*{ZV~rl_uaLO(Xk9(xj^Z-#EEmx ztU^cGwcFMcla|Mw+;_YEkJcU4D7&W2HW|qRUw4n&&fca!o(Mpj_X)3;{KxnKPqawz zB|I+~V#7v&TLkFS5V$}FUo^0F;j(ar1-IeL)s}9Qi z=jcWsT*k`0^!Wi}y9U(r#NZ_T1Z|OrR;84(D{Y;)Ge@P=>g+egGfT|6Qn+$UN+57i z;!Sw&#dFHxamz-pSu0_Z$SOx@0c`E1X8_PY_H?B}SI^i~PmA%;FD@CE=-D1*Y-*D- z2%G+`W5a;#!#VoJF>dYWg_ck62X%Maj}0tUD!?;a845?9H$_gtZ#yP!<{P6;WYcLM z?%v=E(J1D}Y?~=%eipRE_?p6ANZR8+gMV;;5#ix@B*UM{U+KY_;j%v!{MWv|3Gg@5 zr?E$Qb9rmcETe#^CUQh`3(7SV3d#icY2A~S-4g4qclYAySU6&?Kl2I&ATlVN@P5SU z*c?qw)NnzWBrB58Mf_Z(bRU3ZehryDo+9G+mU;7Zh)gjUm+k)5ugyhN!{>;5^3MMF z70P?D0g-m)(c)2OHy@v0h;N8E`JBZTq zz&NRi!Kpfab-zOChI=9wl`hf=`C>f*M9;F`9%{@51;>AZ<|v&eARxoo6dzu_rPAjW zWf&B8%?+TJ`hEtuD;y=jq>Y-6-JK3K2?RU0UsE+A1q*?0G-`{f1Z(oLzWyF-pFr(q z$bPB{GL1t4O2?oIp+PD}iF|&2i)Va*wyyzT93TLOH|@E6#@{Fhv;KS2NGXkaRVd*m z6M@i|;j)l^{~C1XU4zUc?M0<cQ(-wvD|s_>GY5e z=3CJ}05I-b_ZlV>$)iiO&Peb)vg8_aw;a%nh7|rWK-5D<ccU+=d1lFO~U<@bOa+_DM?7gdE*!exC^X^hhlTT=jY0RNL)j>*w9?ctko|s2a zEXA@Q&o?%6w811kdL&DI7V2-E47!UyE8gWOw!Ky$K1a!eN;ALY4{;*??f0Imh<&<5 zT?HNnj_5=eV(_%}70wh|wYPUh&g`rXI#81}FV`1+yYcS)Ig%Jh6&*nBQA z64_?LInS_S=hl_oeSCJfJGVmwT5Y0fRy1P^hVw+FRlm)3PUX9L`o8Q~lPAF!Yid3N z3O1@VA?b1I^_I&KTuz9$Cm}jPmD|pK6qbBaA^Wm(n`3|OPq=lbTl$k z(4hme&Op?)?%^k#?D;yU=U<(Vy{|5Fx(zv*o#*x?U7DC~;8By#S~j2=_`5J3xEzMoAYC z&HE4)a+LCRl3PP>CpYFoVV&VXt(5xQrXl-l>F?biJ(P1~jOSlmAUi@%{d#C19aS0& zil2DBw(z}Fs|spsUp^Px4BMoZJHn6CfMiH~BhH`!4bg)A^qET){1<)zO~dg^qDFcT z@y*NF*P!yaN-gOvX76G763C+xEB`){hycjod#e<%cV&j2R_iPx!9a3Wo8r*2Sb9rJ zjaJXL+J(3e;)3R)z3!QZ0L@g~Dv+XFa=J_>Iq2OkGNY#2ZB2{?Y*1@&yn^{-(16j@UI%AASGPoUG@YBpOvt`r^!B5d_?q!ydpP z9uUt~N`V3S<=4NqoiAe+fFB--~b^8w2&g&s&S~dYhsqsxG0rhh^CN> zdviHQU%1)ln9vvtYohty5?*#24%Kj<^3%>yI(OU$59ht}N)`4(9u@axt~8a9f^}Di zh-+D)umiW$ldwbk=P+{ixX)CNKENSh9pTGe#A zj;}*Am#Xnty{mD>fl2*6LWViZR?NxmInf^4_I};_k#hqWM`dS)R;x(?Dsni3c^S4r z^mqgvmZz2URgf~>w{qE&nJ41=0t!|`Bpsz(=+$vlLW5Y+N^ zimXDjeSY{ajaxnoCkb~U&1$i23akq@XzEVkH1$OZPvl4^DU-x?_W}B7(xjIFlP~L8 zlgDv}H00<7g>e!HvH4n)F`!9?%)A85@_FcE`p8d#Dgk~0-PyB`$S3RFyWSqrOB7*q ze!aga>dE8ug@&mg0yf_!Htrcbb#$5F{jPTqVx!!^WYngC#QiY|K;Ndf9uEP&h*78 zs@_$;5GhyI3}T(Nb-|6YxyW~kI@~qN`y%BbMM zVv8s7+OQ; zJ>9z5L_k((uR>`M&Q`MmgkBU^82aLK4#QrqkTj#Ov*RX^Nb4+f&o#LpY1SX2fro+V zPNYXWX=J)}7ZJd(RFW|BoK_ zw`XB&gW3G09?gc^gqw#K6N%@y_eY&$*dr`0yO|*e$4SOmJ3XP$EHislEPHK5nKYIz z6m(TO?M_?YOT7&E3jd&p*H{%E5%|0v!o55Bu6k2XG$3A%G*kb&13gCWT0{nTTM-;<5iVT@)9ao+Cc4e2wTE`_uUSf3-}7(M$N$bIB2Ghdef(8%GWNBcej6*#c!vSw}fCY^2iT;lL#ArDbqaw>oQiIj*y7mqa4d zucx2u8iq!FF}?Q+Zz9ka+t|8%miRhkW`%z}U0v9WFy+BKoZ}S$z1t)W(AS#-TuM_% zP`GcF$rGWgqm|5faS=8_e+lf74n58rOBI9KHr|#dN0qA>RpV%j=WouM#VVKE#%ZegECn(XG`w zB|O+9)$PhWrjJ5rTO-*RQZp9Qe4QH+oH&!b+v~!J4d%(aB5+e1dau33+bo@|cSGv= za4>2zU+66B9J--wTG$d?$H;J27IWvdu~RU?;6_T&Sn6G6CFGE3>$4Z$()!SKjkdB> z#Eib_63Y?9jy>Zjm2s^vU6Z)ii3g>bG8q_6Bpso8g~jcDi3_eQ?|ep}vHXhO2;YSi z74!I4F)PbkU^!q=ess{5CyrmpL{WgInv0@uEIBsGgH5X|3@mG_bt&|*@o*7eQG)+u z`4(s|L5mYfLUbW5-3i>Md@%HD4kVZ|zkikpcBOhLc^OmE17di4GqBatx0Bq867A=I zbS*PIwttKI>alDzHIUg74`!*~+Ar@Hc!FMl3+mD*!F+(|ip)-xZeyPW#Pjm@^8|Cq zLh55~e$H2MU9?=29w;KWHu``N=9(DY&smQ%RsPi30iqWFMT^PstqrqtSAh*6@>jS9 zip*h#@D*Xvrw8N|nMnj{zrkxrh=7hqipo%3{hlCGqEU;-uyPh-+@{k+ z=%Fy+D|$e?0{KWWZ7Ke4$Hqyl`u*7{d)}wqcx|DAz+KA+%QhW&BEA$-h5CKN^P^AO z!0l`-fz1NA0Oqtkxla3($ZnoK*W7g|lnzlE_`jX&(F4_%SKgF{u-I&}-jtJzCd?Io zoLpPow9^jmisLW(y(u;y8zVlf6!_epCzI6U$88Ew=W*Mt_5jf`_hi{^(0QIzqVP_RYsHh>Ox^WcGkR_h*C&~;0>14f+oCzEuOhn#x%yCN z+vHm~T@VXnr&m6@w@N2@+W`{$OD9onCh>7?;+2N}cW0ut^XXmOVi&?8YEIaQ`q(QS z1B#~cheMjs+Kn`Y`0_2pqC}?(qbnU%^JxVnJ0R2H!yoQ;HVmgYXPo}B+8v-G-!1ArW}~+I+xO^VJ6X@>>pZNYv<2vI&Zn z`nc8A;6VAXqt8S2{FiYFW}K-DC7OG8ayZexhDde|H7VoQUqF4d2Py2;ZOK@6S7n4o z_nx*1RUVtqa;@Xp!~nz0@NWSm*myxzPDYX3G|vP|2<#|Boo3PTFc`i!ZOv(fW#V(s zO%&bx2rki$ex{J=x8?6Ye}DH^cOIF4@=>B8@`xjC>jffT*kZqIK2PuQZ<|NZ2SKBP zQ(=wxu}g~9uYd^%=c?yHc})s`wn+KKORs0s_4;$Rt%NIZYta`k$=b34EtRHuo&cqibs0*DnZ#o`L5y39!oao=bX35>e~V0z8Rf z9=oH_4ygoMY;ms>)RSq_@K_L8nFi3G_p;;J4F@L`;MDKf1I-yk%Rh8_udQQtR6C+M z|B;U8kM;i4=cyt99jX#r`t;TmTzvIN>bY6=^$!b6z~`n_Ua`@DMf0WVlA~`T zJF+1ykxitOSx}Fq@!?D4{N@W3UZMvbb?i4PJdg_cF_A3beCOAPsq6#(ca;pTu)_iw zRN}z~hP`+iHSL*FuR=oF&>_?-q8CakE`XK{!54A`b~##;CCG*F9RlZqHofRSDDr?m z*ef0$PnK6MZo}Oe3MyR?bYpAkcPu)};-5eIR=s;z-aEkJtNPARe-!ToR;e#J?4&*7 z-BYPGg5Q`sAsRUlu6arcG@}CECvtGIX%770nQXrlRY{&*p0+r_D%?*_@11~ki;zx8 z$P#DB`VZEQ&I)270O_IsSRf*}6KguAZ5yn0tKXu0Y?xQo78|D`vGjhJV!3IJtS|NI z^FXxn_Ayh3+M;N3jL@Qpa1?O+V z73@ZwXm0ianQNv`d8kKOQXQOL)g-Zx$q)x>C($+TbMNR{R3PY>f@gfvId%>7xPAG{ z#7l@IzvrY)2VVX+#*<0_Ai{G@di$GtZXA3`jru7ok`d_J`}h{ni!q*CXRM9{b^tE4 zh^@_-nRI$^1`wpNzIj^UDH8>6V9ldHS;+g?OwjNU4d((HN8!=*X zm6>vT+5^J~ejj;>2X zr?Ej?|E^sp+D`P_|L6W~XphOAidWjH2!7#h2@ekea|P6+3a&F1O84ToYn`bfwdm_` z54D++*$5!&muhb`CWi2qupk7LZ1QnAb$5fKYFTlv&Xj1nXc`jqXZYWnE*30S$|tU> z5C-5I1);H~e-kv^8EOov(g1Oc`3!pHnvh zg^3?&&71FbCcA%gWAHAQ1DuxGdn$C}{rWFue5zV;Lwk_jc%6Ufm1v!?R(1|ctcyUz zvnc5oq#1vpI;k*#P?U9?#jB{Lfl|E0j=%M?W)aLq8^u7{y3L24jdOtN6!K8eFvXqV z72nfI_lAI3wt&~$5xXaZaG4?#*r&+@s3nQ^n4WO?Y1&O1pOvNa>#;s^tZPXyLk!4w ze9@+z;#hb~23KnmA*=^*h)af;7_AsfVHQ@TFUkWrA7dXr?{)D%>rbTUcw9cGNwXS4 z*{j8fUtI%N{PHvc1_oK+tjlNSRQP=A!DSk(}TtJh@JmWX}W- zr(+p_OBHYUd;#t)@7>!1WNhz*3$dKaf=IG+^yOmH3jiN5nRi0=3x^W?4y|G>trsug zZV{msS596w8Vz%2c;3OD=X>MNa&%I-n=+&!Hx|jDbMqqY)Ghm#uVlTZOB(=sYy4Mn z&953i2^vN6Tp)SzZ7iR*4?P%&)W*&gNQ?FHcD7EYYZD3L>zn(RC|Ak}84=&j0;naL za97m{1jc*M(2qW_^dOY?WEq9q7*vAui4lX@e@e)7OFf06=YMTOqJ`1871l{tuC@d6@P0 zLC!r8uZ=$Y*865(zHV@Hx9c5$R?Rl!c;kr@c`5+O0;0WRN;me#W}S2Tw3cqQrR{_? z<|SnT)c|Qn^04R3$cg>)QjS&AYw2UV++g|q*T&@vR+Sf5+rzuMiU4!YYU35H0OF~n z)h1=%I6x1(jVt%&|2u2o8@guhNro z7al2GQCVjg*;dz}CqnQO-gJHpMOeQ}F-U%@cVy9QuU7s1e4bS*j_LW~-0hn`K&|0U z8s6LK-cOq53u+EV?mX{esUqUMM2vradQ!G~_>(J@?F;pbpTm`H{(}j)0IOLo46yj_@GaR{lT?W%iI-#o!hA$OCAq{MCXLjrF%=X9A3h zNwbKhN^C_wawiP*;MG_m5h=?ybvc0h7h?`6%fKFpJ{RDtrMmqT37INVoNhkv+vtn? z_=%SJxKgqw924z0`Dvnp_s;w#tWpv}5@-G|$>P;lU3bdIHk&{kt+wrPgu8GSCZg}$ zn*dEQC+djm#P9`|*H&7Z(!H%2l_1nUr`GT9uDSLCPFw4EJsGdzsT=sEatR)P`- zA48~L#_djIwdnY^)404;w_-@$6jMHUy#TbvRSLDg7G=Ar z)2?6de2ktxviARLR^@8=76DAFaI(>>A>eI_Tyd1RGi5drt7Y1|BbRl=E4Qz-Y*x18 z+b?!W9v#*L?nu0?gy?!a;4`)V5DxRDv2wx5tRYwZA2^R>GMUx9Kg+~$pqA?}GtQZ! z+N|I{4kIgbU;i_%XbIi*G9Y6G=@#YKQZoSvleRJHtXr}yUbntN$zt1`oDcFq|9!rz zRHB_97Lh`s`EhS4H+a!Vtx_wy;&7qE45cRoH?~(M#i+uDxMSo!yV+VAr{k}2lB$G@ z;bAh3S~KMqNAs!`HIg|O>k!9{Tc;FGC)#ul=V6(%j$k6ivyaxlGzPz~wC3|yoMGEP zMBX&qUG-djn%ceV3f^?7SI8C1cf@6p&)=IiZX4f3ONbz0nasW%ISh3CmgI5e?)0!~q^+&oy`dsAg z(PZ!FZkH~izHsq)(e#?_`{O$qR=-U+%dC<2Y+3vPvOmikR&va92(S4T2bP<>QWw~) zX^*w<;gHO(8?!Df^5&b}Sn1WFUkLeJiE9LWgxdXt)x$*|H9vkPKS$^IfqtmOKz`M# zx;a=K^=mXSO2cI*tC}T>`kmK{&)=c?abVHctSD5nPM2})xAQSKXS@M0e}xc+$1gHVWR{j%V~ zzYMlst-$QF@n=DypKk5*nyA|>=6P<(HBB(%e~=2h0_H%HM;0dixwIYRsiOdG3Drhd zPBhhbMsJU>$WVVu&INLiiOUh|2fi$v{1f*)%|}8-!-w+(Qg=1l`0p^nnK!7UpT~if z@2}dQ0Ol^4NSp!YW+I7~Mhd&f>+DLp?UBx;u>C`3pEbGN?eN6m9p`X!+D<@(|3%6| znP3a*6tqI9BEt{RfsA9$9Q^V-C~z9eZ@(P_WC5kc&#P_1K$U9(axpC5s+_b_IIJ^5@e*kwtO$NNc=k*AA1`gIyFlx~u z@a<*LwF|+Yg`U{JwO$>;2Bisk?Y=dJwj7J46@|uPD6$j_G8l+f;zr+(lPL;Z;1XF2P^- z@S{kO08)6WJ-9pSbwaocDjbz;g{(goc^UQ~wS%c_PG_K>H-6S+1kMG;*HqBCug=X{ z$oBK~?oRkKrQNZTL1Jj&cMVRjgFl)6>(lHE3{5inqc?D;gMu8jW=Cu^Z&kf-QPxxV*P*<`vpo}k|eem^M3e|y%3so64g)%vL8yMyk4I_<(7jL%gRH55fF z&)3PU_B~C|<@)3m-`Vv#u_7VZTZF~<)+%a%_;Wzg`?7M`c6Jg}=bJnA;Z##fpk&>1 z%kN9O`Ow?ba7jdjRge&4>+#R|3|@zvu`_JjFa1R8e4Aw(qX<<7lPQ?U)PjCLb>)F! zL0~#G>VIDX{>`S=fqyYZqz%+3-Zx}yNTW;Nv@;IRD?f z^=A8js4K_F&o#^5@!6un=BHm;x6*g3Mt5I@*=_V)Q4d!SQGI*-8BF9Wtc*~Vd_?>M z=CK`lPl$2#gJ1)&u(smsj-~G%&@ko&{(Q4K=Jk6*{Jg@LenRP1AV}Pkidoc?-Hy2C zS4#^qfDK_J-c{%%Wde(tKIVOC`o8fs9<7B*@2hzR)H|OMMR0DS&Oy`sPbs1RGVD9r zOrYL#lc5=cErYi@-=OX+c1yPA3eT|We(H$oV=rvJdzt5Azk&$szNGr$BJeFg4nVOg z6h!R$8}9<(oX~IB-U{!sqmGF4ELQ6|G3gZLuiiQnQe#0OzRg)f6ucqHXY z9g||ffKH8&5ZryKQR_M{l?cxTG4jn%1(;1zL+pQUTkm`+%jmybyrlFW)JL@qfuwy( z=1Df^>`R9tpXjfB{{AzXyk0+MpbF}7+x?uA`NVmREzj$>C?af}I${E7tHw%bGD-2{ zorjzdbz?PF!Uyzxo1ssUV9tB3ByPi@)nxxw3d!#KaXUY8i1BQ>Iyg|gzbiFR4w0fa zGLyEPuf!gxL##BoVfj&Dtke)*5`vERfPJHp&VGn!4dCS*CCH!_wN&ypnbOh~l zgOGFcDG7asUOXyT~&hOVv`F_CK10EvHCx zGbANedLARx-RJl4b)SXMo9FKkbjFOlS8ax9shpIFBgVc?q-GS&2hmys5XGJajP)+G zh1dG=h%;t(K9l*MKHUTuR{-xHQl+8P*rSqjPLDrO&6|n>L7GJTC$VmsB;60^Q!#M+ z2F>u=$=RFy!|%kANQldzyKfU}@DPy$NAPCPICPPA@31zreIjt#LublYtzHlRjCu^@ zdmSFFN>_sz>~CE8+<>9heT0j(T*a0UV+!`v0kT%7+w;?<+k1Oem;yynl^Z5l_FD6A zTxm%}9UC>k?!3PNZ;h~`OXy>RLAg(kM>Bn*_n|jpz-wl)f6}oge5zS4d)jYKj2Hd+ zL#i-D$&-0dCP}lS#V05*LH*emXeS@^{iBtg2i?{AmQRuXg>3yea65$+=yOYUr`$^xhE zA@d?!gnAaFDNul|8K=K??Br1S`Xyw?FKaIav@L+>7ic7h95LfTfQYp-25J;Ax{QLp zc(zXp!gXWWH7_3;Y7Y9c2SD5MRG6&>u5vPvAVy8ZSXxzw26$D!-|Zu|ae%`*@%olw zSAlik@?v1i21VL!pkG<<^n7uV1KXN`)s*fKilk@AGoYZ+`b7G<)j&MColq2(zL27< z#MJk7ygdY8NE!!`~kMM%3EOs*-@n!dn$w|O>%$fJ|fqc(HVk>m+R9X-! zrDA75MIx9pSDpaN;6%2MxnhosHJi!l%|toCENgc;W)CB{6ZV^y3~>?ax?wd56!Y>N zad`6WT2BnSFsjCs-Ufk*-0VRr!bxy7o4|2;O-X(hM>sRhq$cDOO1C8$ibnC3sIq`}Z%p-_MNMk@A#<=CuHu|B=YCPfnx>|e^`{VIpLyrezeYmlj z<^6c0cXDP=+JYgKcf0onWjgAjNiUj%sORT^{CNQm$TR=zliQ2b9JzMJyL_kj*8*`} zZvW`7b4^FWkt3R{@emIPJ` zqC2+^bJ@-R<3lOw%M)A#mCMe&P2a;H(jd z5`={a=m+2>_==bRxH%suA@0#`f9g+w2|Z^-27b_*-kc(RSDd}uh5WdPIfesL5-rcy z-XukhZ0!8k&vPxPciiT$oa;>1sp|8`#{9urE3w*SR75g{JWf-Kkni;6r&Q)(Xdrz( zIzF!fu}ZP>XcQVJ4LTzDIhHeg0@e0^{ak-_gbgB+}n&Jp#dTKSDeQEbg{Pf=PsNtIULB8Ao&--gq#B;iYI}$8mVym z=ObyX|7g(yjsTXgsoc&@=4|(>(;tLrd9$50z{GQ{f+H2n4zcM&^*IS4ZM`uV+M2{i z`q$yHZ%O|@(6<5*@-OAMWTIg>Sf#yA2LEIS`@0Fm=-x~2emPN-4R7ob1AIxq-Rm`B{}4e)>Q8ME>6$FuqIz#4WpV za|UWSJ8W9;|MlYg7XQfu>7z$L9>iB%bc(CyRp-_Und7ZH| z8qjNG^vS%AoMK)y{|$X@jEJZZ$CKNZwXDOK$5_*q*RfX<``C5MD9cdtO$H~c!x7Im z`#*LZ{pT-n|8zdGoqqbLXAUOx|94s^pZOXAHK7J-FO>02{%-}c|3^#L|JUS2YzS1; z#*4KLyaIpiF|_@km#+E35r4DdqR=>y`Xbc@{(qE>{ZsFx(Vx=E$$s|vFZorjDF06F zwzjHVQUUa}k5^z(lKHb*<$V#9RLcAX@)CS!5itY7}mLwI* z&>0RE?)r#M`!ao`B3Z}c(m3tO;n_@L{>Qaw62xwZf!r|5F~XbuaT3n`2-aUq zHDj{Zfb$~epT4L6{c$2k@YVHs(!M0(ATBf=qTJpbjK@z_ZIVIMpIvKzN+5u_Sek7c zjWL*JI;l=RGZOnBe}*q`XtCdqfm&3XyUX2S?Ii97Bi@T2-NR-{S`-0LaZ?yf;fI;4^A z?(PO*q`RespV9_F}a(O`7&)BFs+q6Ufh*iqU)5kigsZHxbdUks29g1S_uhrbq@j?kMG zsUp-C+pxC=;<{3CeBSw6*h`~F_9Wv4!QTQvvOfYK-#kf>`Wpu~<|E@HdpU??^KZTQ z#Ur@#Sr+@>fZIOB3JC%}clEfj&VO#SpJM%dWPO@G`Aq${SfAAxk{X6{*+8ASt2@9^ z(I%RRhhmV3Cee|d`9HrgUB1a5cVttx;#K6oPt|-$;E?B&tNH6hV7n6E-`ZWe)wfs`0Y=LGy8+we)n7F#@`ybrJ z|8}Y|dI+7zYb>2~e)!8l%}z8kH4!M}2LSXc`p)G~jGpJ|qj^8}G2s6@rwy6>1gOqg zfp!0B-ElXPej))L#O2S4;leefGL}b$y7{W@yy?z}Vh(R7@;`QZu@O%aP467$n0a=% z9lM)%r5nmL<8lyzx2T6j3vERUJ;W24Wzg|BZArc~|Ml)aKcpYeFeJqzCsFWS59iyp z(H`K9X_eFVg;b$$nLSG0IsFX|v%O+D{Z*}&Dl4S8aQN{*{yYYV0d7FM@-{n+W7oS* zZwB&;*#E|7eTV;)aqtR0CL{ygs&`}0@!x&gQ2-d&OT>Kdi@fpml5x6{|AxWw+oF67 zRRK70Wfa@s-#B6bfZ1@)mq_Dn_~7rq@p6=pkHP+Gpvkkufekk9zwrnWFeE>uVRuF4pj>^{Gq~+>tJOB zg6WUnFuSlKQ=D)TWeZKls;pSo{~#;Q;}X+weXe=x#4$?zZ<}PlxWunc=8Z&fxC-X( zycGEBbDF1ElF$5PQ>efT{qIy7mMud1vQDZ0k&}kGy*BXt_iyTh1%lQzbgAIM`Hdzw z7MWiJdCYcNDeciPL7h^*w4wUvWO4B6YZl?Z9{dR@MVH{7kTZuS&j0m={qktOZsSKE zEqs*GmGY(0e1gR-jA+ozpl-nocK1J_0RWPD^?)YpSWok2)KOP1OixLpf>G<;KE3fE zlTJ~Vj%u6vv_tRnyZc~L(~(8=I`wod&aDUzdAERaoqDw+$c#2HvCh0X_$o=}eA@+{ zvvTb+rsuJ{5?`ZMuB_>pnaZ$O#1_0%55aGS9NE?w@5;ZKe^4s+Fs@IYl2t&cW}R<# zN^0wLtoK^j=ns?r-t&BA)7;gb)OWFZ>a`XuQ7e|yN)2}9e$>}Pp)80_%=@mgy;>59 zLAxM+3!FrwjB|+I@f%1=z7(rBMy-nDDm`a+AP1GH!@dl~QaRNr^?uWC+akW+jP|+F zY%TaTW1H+gvz?^xAsR&@K;K`{CQ?DG{8R41{h%*bfQRHd;__{goxb%deX8KK4$$eK z$JcfD!(jpC>*Qxlr16ZM%25}MKqE~&u4XV9klk?oawVdNz?E%BHBG<$&+{Dn&yF14 z^J#b9X$h|8$!a`o4(oeG>NuW$+ELtq>a?JEy7I0q$-pvgmfk0ktnhkxdk}h+Qbbe4Qr$$%=c{%IHmk9iGCLdWfGfAn<68c5%g1SJ z`}oU8j`3PLu?ZqLPYb{f+edAit)yjuZiU#dn@l9MpZ-BI&fImIT%TJ7jnd4Y^(|gX zc%{d2E_c9hP4^~YCbw>DUW|EMk*P+xL9@Eq=SXAmwi26*dc>Z*&TMHAuTaj5@mhnAZ6a!WTRy1cz~*=Jbd43&7(9guS^N z4dI>(opzb=m7k?*Ou?=-l-CLg52=+Mu}rC4GS^|a+i7V$*d5j>@*~-Ox~5oynGgN1 zl;;x$T0L{Vt_S}r1*=UUL^+y*f66!&h^L=dV;B7F>Q>l00rUX4A^AoQm7UERysk&Z zaT=|Tp*r3ZLAH)Ii!R?AS3&BeZV>Or<+a*y2>#*dw@6Z%l^1UNzt8p4m>V>x7T-@2 z+HT}oOxkrR5)+GLCh4aouI3qq;iCq)CAG7RXuCOA>Sudvrtmq{coA9bo~-vOm*}Wo z18BXKtke%3EY0$YW>~1GjZXymT76e2slYmgJ`((#l5b}_XJ}0R?6GZ94`tD0%+hq@ zMdUj80?|=?+m7co943@!61s6Vz)+hLibqheAK5sZr0+u+XKKgNBe1{5TND2l^q+@Anp#6ohU?Y+J5f zefA%djgicV3AiYnUsXcXmX+I0zMtcM(+kig5B4V5Vg2eR51GCNB2U(R_LQl%QZN?0n_ zanl!%QJr%$e;>Fh9{MJ9u3xkyeEdZiv>tlG_^*dD=BwO)d7{hWrURk-d1E8g*0=91 z-@PW}ijr9bg?15iiwbxKLPuQ}En**Pze$#uQ1Mhi7wyT$-}4UwRpo!=EEh(S{^DM{X6(W z^H`|f>_1(f!iDwy?wMPTv-I9`k4dcqgepCBSlpXhy=gE}h32VRNESUZEwv|`CbV;T zcld_$_$Qra4dV0YKltxc*!1O;FWrB$38e7f2JPncsamLiG0TggQC{!3|I)fVTa|L4 z5+~^6B}7}U^*$7bO7zay@OOIh0OCxT8@`xLqF$_N}+)e>cG(8DHrbEykQ zbE8b~=b$JFuUPttP4A7$PBeOkZ@>T+L(`Okkev3IUwwtZFz@2HY&+D6Qv%V5{@?v3Z=Utq=Y<=?7NGgOn*P-h2bH(KA#j_-oV_eU^$0ND4 zM7eL;Fj^kDRXfH7wrQMA1SVL-hFuAckg%(66_9V`M?5X<5w}|%4qF}fi@*1#1hSGq z=C1`88sOSa(sok}2U;CQ^+&a^Be+~40GzyL87V!#2QJ>rvY*|CjN{uGu8#V!(Y*%N z0_>aU2iUJ#?CnrPN*B5bmX*2OTD#CBMqTRirGl%!h?A$z)eZ2|U*J$#tDOT6ErLjX zYsF)>YK@qc`$M-$Wfc1fJVGQHtEyz!lmZ{3no%U*Zd&d3f|9~8G=U*mM8|de-r%Ajn5b0XAaxB5I z36qn!XrV-5pW7DUQ@-1#VwIV}OS7ikHa@$Ko01>eOSx*@L&buVfGmVioc>GaiNB!w zRv5$9539=8r$S0d!XB&VkjhEuQ7d#<>KYX%M!5G|dXaU(bxYw^1#N85!1Kvh%-@i4 zT=OwkLlGP91JLtV59#QagGrjykM`z27#*=WIQT$QUAP%XreW23n44lRF)Y}h#f2Br zeR_}hYR{oMN78+8`+ubZgym9C$E*c5$)-!LAlgM zY*`#LAqZo?VJDiXp}OT~xgYgYVEP^`VBIcbv}I1UQzOz=!W^DCDjDHiJ0U)UxiD~& zryox;Zfi?U!>iziS2kMQ_*Pz6%p;o~6@&2T;bo7?KQay0s~VoJoitk=qOun~znagk zvS{kbko_xh!FYkwg_1V5d1r_--z6OssC~8PO&3Sm^JRt^&Xx5LDXLlhFrldQeBvUy2BkBDg;Acx-SW$atPQs}zpJ-apHnc6l;8O>%r&#V_Dkrgy_F z(8J6~7{W#(=2Mk%)0ZWvynQ20u6chK&UBVlgVu1`8x8ef!F?T= zUeeFNyrg`N{MEL|BUKLoQm*N}7b@%TKT!`ovIb15v%1usqDn?4c{EfGb0mLIB!e#7 zXUraLOeabHP&Bbf;`aLNZf<*MJr}C(sPPRQVQq#$^G}4)wf!G|C z)?b|~K6K``mhh17mT_>ujo%#`cgW`wTf(*ahwxh1wD@qzX6cvDvE4OJ( zNGeg>m|d8NGv*Y8q$tEbsb;OBDX^>1Raw=j(ijzb;MZ(Cvw4g>$^1w`cT)5^ZgC_B z%bw?QXN!d4&h9T;m5sfm(U$f`2=en`GUJexS2R| zYgi&}CDTQ=U?j1C>`Q!l+Bvln@;_cvln4Iv&vWpRWAIp#3#U-sI;)nKUvonCAVAbD zp~fszO-8`XI2PJapqA_U%&))SKS`fDeAQj4#uOEPR&=_X8NvD@i>|!!4kH$Zv)U*Q?Wn4 zi8?pMHZ3D9o`1-1A%3^|@+`bBc#A) zTcG;1%sHQj!^TPVbfGK|VGvc36vvl5i|yxjp8QcPPraY432TzvLMl5E!7{eWv#kw2 z@OmAG?Sx6Mt|x2Z&qMQKRp+<0=fz+@(^TcJ_5TF#gl24;jFATETC@ewyiI1B)~Y!% zZO)O*?!xMdeOBgbXi@2gSEvzrD&7^iF$p8aY6esQ+dtUzx6Tl*U389pf%Q2?+i!ht z!YG2a{ZU7q0_H)${DM>Yk-?{(#WucFAiKX*KTo8WYbbSb0 zn{m?62*YQguhKxoFRy}xk7%1NHP}I&LZlQ!6ndHZ!*(44N*L^uL;OGSRGB;9+T1KZ z+$U*=>3RgMMmGou|GtKGBe+cm#DXna=p5##cpri3@(V8zSFRt%zNiS+_f?nz|tLrF9R=v}G&ny6SJG^tqfTYFKL?kJnn_gB9vGy$$b-A4|1g|6Oi!Irnxl?`cKYF3g4F0{&#u8HsWGXe zFypmEMd@cjRU$^A`h)fa1W-|>sVb@Os>j}KN&k@$hxEK^>CS8RtToCC)hUfYU`)9_ zwjLrlZd~-das>={Fs*b;`5p7jF z`_l_+UOm?X$sHYDR|$4#UkiM>ZuTo9j&NXYcs@lbaql|Vb|s8*IJ(aXasCr)`6}G> zX(a001uM0zBk6@(`)jnV|smL zUfiCeTY@+H-tZ6&P0h{_KHFpudjazZ-|rl^HPZ2OtIN7v%nP#7TI)YBqt0D&z7R0K zU1$AvE4h-yqS_gFpOfdQcfQjub{lj%b~NX|ii(ME!?WKPMJxo_ePZliDx-(XtO-lo z<6yVciNQ4XxoCIin%lIkC*oneUx&{uJ+uTxRK@*q;VU=peN6O)SPJSououTZx2>K} zO&tx^5?LH7q1ma>M748{T(;xxIw3oPaW#v45F^5;$~OJ3#wuxx z7p6m}(Nv`qOA~!_=+$Yh(Ww{wNo!Z%-jDDYZVJA*<7WIU+_mK*1i#Alu6HRF)UEkU zHiKe7JH_X@sW zt00IOvZ{ClMlD{eoYwzP|580hbpH5r+Hrh-S-Tf(xVhQMN*-U=+d;AJV<#tr`tVQR zMB>00$dp{_)^@WK^ms0JAX>q+dEeO74=_)lIDE5_$w$EDgwSmBnXYHCcdJh;^OOlL zeuaA|mh#>W1llp8oT>Pb5A8_xN}{pT>y8b*;mTm?j2{!am9@gN*Z*zSn#HeQvag|+ zV8yeli>B%rEb7STIy(9}qy)q_2NxU2yXm#aEg ziARzLfvkH=#l5;JrQ zhhNA*K09N}DqCY%^STFJL$RrH&e$@gR^8>9CBi=#m&3E09TN5KeOJcX=L43WJu3`X zC^y~nm#L?B?(LxCb^+4KxFgraq3~%l3|V6SFNKG%LHG!>bJhP8By+*G4IQKOHacr!$Wm-u(7vD%Y9iC zzs?ls;tTMgOiTqs5`&x@yrw^zFJIBIyWB9^On^JM`@Tc>XA^~^0a!J~szM#M+(O6g zy@bcfdPhabfDV4#2(V6bRwPTZr^58Od|U?45MdU=n!V5n3vl$=ep!*kwCt>)#rF@^ z|%oJ~E#aS2&_T{}U{z3{HD%Htd(rcWgVS+~VphxvF0+&0=WLleOZZ6Z*w~ zlc9VR&(%_3I~J{CoA1pELPY}BUlZ_SyNaow0;siF=YE7k}h3Zdpj0^;mV}d_CDbw9W}5H zyySOtFtYpMFqPK3xLNk4b%#z;2%qE!yX#58av1ps?eyk;CVdqE?dYRd6zJOc?sW-hWf|8&0`6n(imu%MnMqez*UT8;a@gsE7ZU5MP5y= zXOJ-CV;|8|nfDVs?WrPz-CZoO8%@uDwt0EBmbJ>nUiCofXBwK{`Kg0!wwHqsdY$6A zHp<}6f2KGV<3O-j1Y6ZNMM7QBZ*lLaOlXPuy8k;{;?~N6V1Nsv&hmN*<&&oI+uxy zLJONDx%(!n|CwI`T^MGBH`T9<<2<1`{xM2wtvSHn0iJjG<;-;W%tE}doyl5JxTslU{+$3oWp)STD}JX~^n&b+Ozvg}Ux1dPi1R%o-xiO; z{>;dPH{h{oWWib0-@9B|w?{vay(KK@v*82X3<(4iBhSa|&K@p--xvy>1!EkftoPOz z4M&Bnu~s?%UPeLGjx>jeugKC@BmX_`v-9vV7TFvK;irEaJCf*ZNPR;A2(PC-+@ z*D+{ar*|OIG(%O*E?Mvyk~AO?UYXi!KHgc)MpJEHf3a73G)TIiE3ny>`{i(P64;Uo zbGymn7j;&Dh~jg0j{A^^^IDx{hmLK@>UTEOUnDHSbe@j za~QS@kN7Xu<<-3U46vueY`l4R?j{-YhDjeiv65LI4H+OZH@K6hL8!WEY#hg9VPy*s zCFvynii8G|9*|GYA&lWr4?;zX+o^E3A1LFl4excJw1J(4+zN(V_I_J#0P;w&>O% z;gmwLqS5z(M=~G}WxyFV??+P&X&jAA4R*PfI|SI^N3nBttm_NaV-^fX5ViJ0tD!A; zLTSEKEshl;%SZ}6khm9TeC_4D3)B_=WpL)8c`xH4@j`7tVD;G0qFju8U58)801J86 zfz1DF?>p>$q!*v7E*vlZu=$Ws_yJ9wJ)iSTABFZfBx>1lyxEnvTvnw4{5*M??=7au ztWVZ_A4HjY-P@0xF<%%)_9`pX+cxWJpRw*5z{w;o4=SL1X`RL5E>z;)ya`PMvX;+L zczP0E$Aw15a~A8nK3FoB(i9tJBdeYM#z2p=Az7o!TF0I4@#sak+o3Zk`rI8uUlcmIE--?i>;vtn%_sitJDyn*Q1a>NHR`Z;UmB&Ex zRvpigqv_78`9Q-D%KKdEsVjK~G$0WDsc>Xpy+vZ%S^)O@AE6@(+G7y@QifrUB^_?4 z`c8;NcJsL%O&V#`?V5{DWc04zXplHHV|ibA zHf1&(_y$HjrAZ%3T8SFL%@@W1djMqQ(kQfxJ z%uNuH-0uzO9#hz55RS-Laa`uiY(m@IX?XMpQhkPVJke8b zWfvUbVaS9Ef7VAdMvv-y$Xjs1B;3OsLUyWSghg_=`Uzh*YXCV_x51qmSP#vVK(9yG zU|ze^R(5l~T0uJb7cVoSHyBfcV*0q9BJmy%Ve3nYKv2IR_jKiBTdvZwEd*WZch@fT znq$Ru6L>4NrAu11CVYF1j`vPGvYx&hO_+`r9c`XMe}txm>L&hBjtj}B@>A2ghmeX< z5J>o=oLQ%>=)qKaC^ed(R+Vf6?uXlNikk+`n7;^d5R{slq@o$BAEw$WsxhjF!3;pd zbebXqF$t;6fR6bua?`u58=Q;CTC#`>*vfKY7Bf^Z2=0X*jq^N zPd#ZOwgrS{U4=5XL#8te3Ux0(3zPhdsPP?&U#^!B&h7~S+Z^~|gKVz%1G~lZN!h3F z?%1R`)4-Vu^B~<)wXztq4C!lwI!(x|^6so}B%gOSBar#cDUA6F>o{xrA>43vR(9Me zUmnX`|Gp^Tb;&bbis$*LQUO%yrOD>m#Hq=?4X3hXdN=yC-n@o?y}7y(_i!7+BEAC( zixsLHj2hkke79-ws!s>xq%iBcCTEPL-#>>KVL0k~E||u00F>KZU7?4vO`N7X$Gy@% zwW~-lH}d8;{`)4-?tq5|SwMY-9Dv8{gmV-sHcR8~O_!&JV$I-*@G|xye@Kf*Y8;?@W#koMGVZCF)G@ZU#~^x2PL~hr9E! zxmygOHxP}z_G58c)%GuP3uXIgO}frQR4o;EEnbZsX!lmRh>l#d(%0#Nv|}^z++!j4 zvWx41x^``6VZYq(J-B={Xha2i+XH*y-oGa#wo78ER#@lREDESP$x*~SV(6^ohhoq5 zp9v_5gHH^SwiGPl?8%Yqa{h_jbO;3mfr#m`>)11Da054ymCH<#GZ1OjoZbg$?p)cTT`-nN^x>yJ)+&uCwGS z>W88WyehYkdlB73u_w3)K0WzXew_il-!wn!yZu^1liz-`wDK&Sp`vBB`7UhI_69O* z=M1mT?L%5T9*)ILvl>=+J2)#1WtWiSwRiw}+z!L4Z4=G)AQhxJPI*lfryn2eb_*&& z-kf@#rDDt#YrqHHhp5;5a*4v)&TB<1zZ6 zz4GI5G?YJ1Uu$wm{42G-jnKqz{z`+#q6OGn&3+TCEuz`Er-hh}HXN?C>}AO_$e2_- zY)~CHfxZ+Dziu>tAb|;4%J11IF2o#bH7>6glqUXu<&oHE_b^(lcmAR9<9B7WCJJB5 z%r~U59au`-&%+ZZK&BPtJG8=q0RVaPwK;J;e+jtYOZqo*hmofl#(Cw`s3XNOHj3H- zU#n+<SPOqFoDzCgu$w zn>ptoG8LFq@|X3A0QHW^WwH%M2J_}HayAXxF=`NDG5u+-JLVWJV#Le09NXu#J}M|@ z^bO35eAY`_BtWZLQ4(-lfXL6=fn{r%_&d5LcTKD{(iO~I)eV)^RocMY+_oJ#|J>pD z8m7}k%yIwmx@~le)T8aXff{Y`<7Rx*FC~ATbcu_9us1UR_GYCf%x~!^%SIkZofPcg z(pFj%-y||ok8Jpy!#{)(qQQeKf+qm>CNv7X-1U?%B)|dtOsQDf#;m)CD(c6{3|suq znUwPl?-kN6QkSW#ly?Cexucs4jGr&wrgzl*++XFuUIQ_YEg?ok zp6X=zu*ruK5!DG5tOPue-rf_dr`i|I?JU^LmIROlCtqUv@Z1>akoppg1vyn;HWo0N zYfobZM`2O5CMg*?O8A6iS4Ml;WmsT8@rfSFBp$nxSTt(0*a(IUQuL$m+6@OdpT6>3 zvWg$Ft=t{;qI#F2@z0vu1^8vsBot{|34yPXNZ+@XX)DEi0Z<8;xmwGwHlP?2wB`Df zS7bA?^Rrsr>vZ3Kx_1ZdHR3`KX!(53wA3}K zx?)d=hvwAl^RSL84<=-p1Vz`;?cUW~Q_UU2?DX7jT#60`Znx6SbUN2MJZCJl`7$qC z+j|y9Ynpx10)#s9;nhBu)i<{MRT{hyB`)T(l_F*NL!moAUh>`7sn1o8-8Ow24m7!5 zUA|uWH02p2fc~I_KIZ~j;a!7Dv>(_{J#~eEJIEJ92ZQs#XO5i`#`PLSo*WneJbTZ?v6X)nnbKdG~?+php zE&gPz8EsEp886jZpb!^$ZEo1|Z$^8Wamx{nzf17-Jm#C^f}t%vEzvuy=H3)cg*22W zaLu>`oLLthmg66HPOk|Ih%QFgF+=XIPGz|c%S)+wn=((V)SzuKx5kAj8Fz0plx$RQ zV?&OL3o~7Ou?OrNg`KW`3f*Cj-jRbSTExcb@`F%uQfa2C5LiR9Po8np&-t`_25)C4 z?l>^49W=892MXWOjg3tw;KbRi6H%+kZR>UzsY+s;Qi_*i zpo^Z_u2B1{HVe<_!%z!TVNtJA({AUCY4!}Qy6wmS(j0PY;f~W>ZXz**Z*jn^|B%F{ zoBD0V>edAtHHf_8Be(2l1B6M}Qv<(H{kZ5;?83#N*4ts_REE)FmBwl>5beT~h?^;W@2=B;vzg z_&u!#v9vj-Tn)Vc?KN68<$P|tux`EX`_sc{tm#-6*Nd1A0)#00O$M=$`jMB-=5Nse zJ>1?bAL~~G{>k>np!;>Pp1wP%3-KqUQxDJm>mxam%e z*K94-e9`KpUbpPnXs}ieiSL4GUChN?|3`!Q@fj2pW9)R+u%{phjUXJmX#PNXu=x!4 z9OMsUsO0b=IZMXeL~v@?7`Tz_Medpac?MOV3?`#(itu+@V7gKntwgz4o6p%C-^#@3 z^RJ;y-DZ$RKZfGS`wB|OMixn!nNj~+LJSP~-NDaWD+u1vXMe-01~#d!RQNL(u!-PV z2L$S#1e5(@8M#Pn4AO=3=^lTyH@$cDW?#BCA41*@J>pz$b8RWldd3#pJbR$zX4T1; zK(@0xw$rj3`LGKfY;hn@WZ#`@=b07|GPUP`wVe)fw!Ej^_{NjbQxTkSGRT}eG*ytF zyL?v$a#1Yd7rmO<_=f6|9Oc4-F12~2%{;ZML>-s0a9@4+dbf0Dj3*eR6+1i@c)BF$ zIaklSGLYcJgJGOvPl%81c7CHnV2IZPG@vO^S|`2M9CRrU5E!O8;tL=#oVNO{Xg_CJZoGMa)_b>NO!PT|Ai1 z(iY8YIIeI}uq^v}@Swd=9U%h)#&nTs&ScO0Yb$F1Hnj|YP#)~O7cm(#6j4*KyFU{7 zh0x?JO6y54x0ud)TJ;CDdhr$e)tj{H9t2dWe-vMy=?j@r1JKW^W4ShhE8V%YMw$X( zLyLZQsua_m!GX({9z5(G@mGe!e(rie8vl$NT`6R4Ox_EQMA|avMLpfm$K=+V?OR$ew4?nz$VU{Grjcr08drq}GT8&)C7j$Qu2nYEu(g=KsD(P<89GLH>=B?`m7 z)cRo{kB*9Z2I*FIzM)U6W{Gfp0(+Ji-8Rbl(h@zweU1=+w&Iu1<<3$dXWSGrl5o}G zug6+L66Vmto-wHpQpOvdhwrP;+E#A};x-D1EN~s>;e}F^=2qI3D5)v5^1xl}p(3jb z>SyU480BCb$vQk+Wr^cN#*=do!M`I~WJ3nE?THbR?)_*%MHOvI8d&hDR=+UqG4N_z zWW4<`AFN7LSN|<;t|pO-f6-jFh|>FfmbiVDd{CS_yh^yI!d~r6R>Z)gR}OjsC0ni| zzVVGaHU+~wGyKPYZoN4sXb6|^wJMX@Xs0q?Hq@Ux0M-tRv~#{AD;{|d6UElhYe?Ta z2fCC76{gY8Cpl&t%gMa@?RMVoEl@kHToMv4x-8w^V{PG&x%FJM?GBc%M4R#W->6JC zYAkMFOsfx&|L+Piw*w-@OAdrU9akM8E8Pj?)`%0ZJWHYvAmIvR0%?adS~I9NF_(Fh z^D}y015|eLe!yz?$R_hVAopOvf}_7Q?M4bUoDaG<^nVysnddxSkT*O(AHJ&K6&fkk z$DmlOIp@#zrQ{mp6H4DKm@4%71iOnTAO7J*fSHyq|Na(Z^$o-4fEG2k%Ah8U(3Sj0 zPH1ZCS<~$JfpdVE^7lYpATe(0z;yKsO((6j(-sWiWO5=|%Hr=`%dsPge|(^R;Gdv8l_RG477WY^)ie(KmIC&zkDkjTRkhl7gW zP8=(78~n*TGzBl%)Q*)djOCl$jozusc=QDbr0;uvEN1o2F9dV__YDZx=}R7c{e7|( zF@Gd`16uEV?J0OtDnz4pjkn;e5Rg^ zjA*BU$IdJ{Z<=@=I1@$|9QwBQjyjNbpg%ll2&wERB9BHtD|DfO8TY*~MtsKHn8rbO zU&-78cC5wqOVR_@lnAq7*@G30(hjccqC~y(S<~7rJh6D;QM__7PuoLhgUBa|M!%AN z8<5HyV1`$Lmv1j-g}3IAm13gF%o3;kIfFEaj_9b5*bH&xx1LWtqn-n_@ap7U?1`#m zeHYBfTan?$9H)D@x5@cdN*!du-l<;COXVk;>SpyBpS>AMaYSU&3`Y5KUPhR$n&%wp z%Wa5%7r{w^Y+}v!@{Tn$5+b~1S!UlSIuGsNDVG#1<4~(5Lq6l}{pnSC*f2VFmS)2? z)oN1KjFuS3;oXk+@~?WKI=7VxHM+QYD3wVskC#P>2TLa&bI37I>%;r)SBunPRbU)s zaB}zA>qRW%?SU;)8JiyIthsE#WY+b-wbhf=&APrp$DtjqOXXQtcwN#Hs2&656610G zfpM+ukD*&8=#LvMZZZB41zC*{Wq$w{N^Koe1?HQD6IoM%GFD}U*jH^>@}isY2 z%%%*P9$xF+BVg1hPYEDjWCF$U)B%z72}kv z96KB!bv%yb)r*6SLUE<7#iCl%_z-9Z2X8ACRmHuUY&O#>F4hdJg+wO%tE@@)RFX$`-s>SO2|Gu(6 z12CmucXqm#@OQ82iJzmIo{o&VDQ1XHHYEO8zVAL~MiQDJ*+7GM#GxUT>NWH?Xq;K? z3T%K7v=7*s$}^i77Q}RE6>|Q>zgm)9l0_z?8TV;54({;mKxN_gYsTPlyeQIfvf)$` zvX?&8;u;QQQ)sa;07EXP^ey653G zJvW$>NHA5X4|4bs&!jT^ItLLmhfzpsl^-8mvK_v1ekot8M(mtIAJ4)7EbVJ(U-qew zh_`;Ypgc+$yQSmz-A~j(MwVb*DZ<2^@v1gOo+o5!=8l#AJZGKT z=*cTP-isoIBcbw0@+l$Xy}_f zpG4*imLODIXHhyC#2UlRLFJ7SqcdQ9VdZ?4n3Ijxazia1SuWo6i4wL!PT6D-VsCq& zKz(|RGMOd2@Y@wVg^agvo0);exqeX|k$|V{VqJ^M4b0=GJhO?u99lS5*&inu?0^DD zg^4$!pHY4J{>hFBczz$_F=QKTZq-05wOp!Ay91@W9pZnSDjC0p{$Pyv;%!I(^weCc zM4gv?9k9LffF_OihbTR&m6{3-)^O!UBy;?YZTn1pFQ05$hs3)&<3jlcodtGwr$u*o z+7HM)(Kt!PtwHW3{pUZ99*_}_M|oj#mSH2vB~Pl>_3VCl=0>a8F$4QU1y5+li?O?O zIol+f7f`ms6`M9lWolLZraBr!ND)71;%m`N>u3~K2)(WliMXj8SP&A*XeFQe?M)9dhj2+AFU5WzhQQ!j?vZW$4+V$-r=Q1$z^Kw6Djt*i!neR31Zm?3)xSfN_ zVjqwP^=K)#C4M8Yfx;!|dUs}rM6Z}vN9z!LaR7>TKK?N|w{UL`0moMsC`1woYP_9p z?Y~4@ZoC~6j20x@u571f_5Mx~v=z^&iO?%{_F3j)|FUCNj_YS%H6_LK=tX@L-0efR z)i`xMAC*Hx(^SRy0rfBi;$|g#47?j zqlgz=#*2u}9gff2N#)>TUG5Evg&w$Fo}S!U{ci#(=fhjL$_ zKcG2+%;SZH{9IbLWqSRDeFADClorcn<$@XT9w+7Ejrxkr6>3Irp9&lnwQ1=2(fnpi z0vJ!a!`s6dBQMelhaMoCc|cU{>S8v975JHq2jnJqI~Hn`YQBt=fQcc==qNeueO_O& zVp`_^S2fqO)pv_Rx+A^3mU&`_S4u038F%+-{6}EFW2sjee5Z9AFvtZxeMQ4 zwuct93j&pVlinY0%2X+R6k~=BD^l?nwwqa_?lyU>P&+w+QjGjN7EOF0$^@V@Z3EPqc`|X_{w~j9Elm# z9a@W;@!YF&XLVCGQpKe*l8^Ijk%Xl$1lMfejmcBY*_dz&0P?h!!(FpZe3J{l>@_bC z{~~--6Ky$DW6`v)kn(@Wev%{z!0ZGTk0)!vtFcr8ZTJ`dL9WObk#8+JvsjT?t?s?o z-cN4$HHYmq7XGrrWBe?6)Woq3bI+@GIsf9}Fw2;}T#w5j-FSP1Ds}U_J7BU!DlW$9 zJ6b{KRZd2*hem$=A*6D9dAZs2qA%1i*XMM>Z1Q0u-eW3Cw{_S>^#RV*pxOJ)()Ot6 zbCHf_%f+AX^!Kr!!cESAm2lqwR2g`Zn(m(~4ikO)hsGeD2dVSiwI_Pnx6}K4M*Xt5 z3m$xP@x4?!(Ye>V)h6{LEVSxpdT*# zlBErFovjqMj4Oz4)$gXu>1_ zkC(G8Y?$%d~^m56I@%*$jC{A#)46|l1%Li zotaM!40kD6pOK77kaj5W)>-9`-N<3M=ZNtKR^@(PDABq?IK>uIv>(oiG?*lL_I>bH z@Ljp~JzvxNP!ZOt3Q;a4HPA0QH?L71Gl*2CK4^g{)BC0wZ?B*Jpr-kg(*rY;hH?@^ zT9wsJL^C^}UAk6zooJ18`q2*|Nh)+hJ?v&t&#PZ$K4F$Uur%Q3XRRPw49@|TdDCP+ z(mq}c_wLtk=Cyxtw)eUR%SLRx3)FFU@DoiJS$XLQ!pYdZ>09CS9-t%faqQK-s-qlhVe%lB23i=F+maHnU zzWZwpaoUtWJrzUPnJAba_XyLUKA6T=fk){P({zu06n$TvbUth+qR%lowa+Nav9CAp z)_0vlMhKVBCyw%{1pCTnET?CrmPF3XZ#UfQG$#^K*T*xK6-zs4!PZ&?(Al0OXfXD0 zO^R(W`D{mx`EWFYYYWD7JpeBDp`T1v*KEY8m(4^Ye0QJ)zc(6k8tT6^6}ddT{E6rH z=8I}YWvC$jG77A3K5kAU2k8`zd}Y&5~oaoHq< zSA9fu0q`(n(m?9e4h=z!Q9WrIwm|^DaE4O>gmNRQuPPMsFag@4Jle z`ilK`S0Q0I%(T6kL|Fq^>YnZ1(b)C2gX|w(l`Mfku6*053=tBZt0?e5-BWWIkhs2z zx@Z7X3;yfEpdsPke^$}^dD3o-@9MTXq}n|#P>u-k5URdK65ifOKnzh^r4#5VVM1pi zgt+Gp80hd5IP7NmYqn_q$8O(9w%;a=Xv8LSV?yA1TzbjcSvpel7L>pRzZg{8gCeDx zyA`*30*f%LXhYB{pn-Yl;OF2sMeAnx$!3&#CS5f>L#ZLc8@}qTlHq&AlWx)$sJkf< zooFhPBwgV&mT)in&Am2tpL=k@3|w58@niq)9+)UT-^~QivUJ4Mm#SNVkOQkkGeBC* zE%OhJGI8^609&>4Xl2B(k7=iLP%fRmFcZ{5_(Ad=?z;mK1R(znu{zRypVQ`!+@= zva}5=K?YHxAc{x^$&z!BARt*JNJb<{lpGw8AUP_MlOP#~48oA3l5-9M!T`g}kcTwz zF7`gp**@nyXJ3B5zr(M&SiO4nT2)n>K-JFd3^aqa6Y1Kg}HVJVMrPBqInQjB6rDI*i-ZJSBnxHe)S9L^a*GJt@*i7ncn!Uosr`BWpD3)uMI;9e1x^A%`FA zCiTq^t4aI8#8!b7@7Jn>up$hJMeJtJzYv<#z3{T{C&Y?t@Y?&jG(O@pvw594_BK0# zYRz8i*mbtlrHRj)U@vDyzst2)Z=vnwT_8~a7w$+L*VcM6n_k2f&+1Z5v2C_{Cr~85 z*m+!3wC^*1&gZqL@%^S_?nVkIu@n$P<0l2^P8+hg;x+8iP2kgNd6BA(u4x27ciQLa z+aNyU5nH0)sD>9dmUl8vQf-*ZVN$ODB;f(1^W~cNquG1Ym2h7yGv12x;NCzJk*%^j z0Nv@yE2S67;%>pdnx~%a&aFSI%Qcq4T&!EyDcoC~2}>e@VuDXyVnWIYbOV@}^f}k=@3JrxBQBu3TSMO{vkd-CEE8nIL*T5gIEA!M5USEuTL$?+Y$Z&8$Rm%Lx;& zw6q3rmqf2N^Jg|(AX9yZ501M(YEWg9g_hLVV>DunevkkGavw9j2qUONU;)$QDk2wE zNAZNaKT&tVnBT;-|Jzr+cSeeM=o={2GS}mQbJY1op4Ov?K-!eUv$c3c!vH zC>aOZ>J^r)@B<@`6f0)qHmP0%CX zwRz^8#Xro?D|C#mNZWW}RaNiO^4`6F!SpYa3fElNiX2+?+uLS? z^j$#Oc5f-vLwTdZVztSyq#ifdW_fmEsX>OHj5)=Ajc-qBt@64Y89!Z{XQ$D3#!aji zxe-=*H(6~aiBDX^6Oa-Bt^iwBE_IC`C+@d64%a@Rgb9LopX=$U$J~VnYfhB61-E|B zX6k`Eflr9=@%Mmqz`3JWbzck400cRnP;`pkjj$C7jSoQ%dIC#i%b1}YOtlGsc9$8L z#q+3XFbL7?{V|m%2F8^7I!tp1q$sQ|O}){)DzS9N6;z)~b4mX^}Z&GxYm#sD;%U zySIRC!8OXN@j0H*M}6|t2<2HWEg@}GJL=9&+6f>%qUpUAi4SC(dzCqr3NZ0WnweS& zVlr(hfdX(zrM=7=C1KH}a)z$Te-ss#WL$4mN@L!70e zaYY-eapyO}kuyur-(dRXqf?&6utl2{F|*0G%>opHf{1+P+BDutozb)HNUuycv8&CZ z0iuCEE^gCo?%M@uzDt7>bZKyW4!6cFWa=yj!PAG~UrQ5woL3};yu1}X#YgkFigWK! z3k+{PC=ZkrT0KstGyCW?ezQ2JhtBtreKa_;5nP=9EvFH0bH7*TYb$^q$57%LO)_RD zD7x%5S|Yq%L|>j(utAsJN_MzEIm^R9*tEY(roUPgN26VRXg_9}oFm=_LwQqeGu_c1 zTSwFVD-DSzE`th{=IG3xS2p~bF9^+NR%)p0b6Gi=ZVVv^W+>xD-dASocJL4$OwWy4 z^y;RJYC4XRF5X`RWBLWrcZbK+aiXjHow+GHqbsu3oi!<(WnYMkIlPDGm*@l^EnQh{ z6(%qQz`AT7o53%pwuc?`QKfqJJ|ga^#D!cQi1REp7DxZ(XkKgo)4M>foNih9X%tU7rm<)alp# z!3MTRfvZre{D|s`6H4cP05_5CW%w4W0HU}|JZQA>jM*^+@)AW7Ln;aoH%UFytlegj zwj+BEWY1~?&!Q*)*a0>GJ3#H|mM6L54As4H%cDTip)0MEBnTjzV44h%e`6Ip%Dt&l>nF~lIzt-S=i5*GWgRBe6j1YG8%7fLg7 zVLnSWP?jMi!Yw}Db$p47xKLlDMs)ee<@$Mc&q8RHG4*auqYX314oSI{_iyg_6E|!m zg&@?HJwGf5g3hDTR`*WoEKZIq>jqxBoyQ5Trge_esE|8i0$Ao5iQ zYiFVZ>DuS8tqXU5$25C*e!UIa#)OA)4$}{r@TGR(BzC?(IaP~M6b2qreI? zQEaPYHbj0|YDKO~&8C|eWOcKiq#&b;Mn>R>Z?@0P^fd(efV&{q{rm5km;IrIx4EN{ zl*=^AzH#nVe%2=M`y~$cD^F^L?LPD~rKquzN^#h@%NS)}r4lj=hrD?gZoFbzu`(YIX)0?AKxGKY)EHm1fzLZBsDV>O^NI;~U#SrWAkBzGi?vuYx}xGG%LbDDW2w3C z=k97?N5^%G_x=q-z01thD&4J530nZS39(lplgB=okYyX6)IHhz+M_T|<2_5I=VnfD z5@tX9+e8#S%kR}_KE(Q+d$p)ZzNDeQapq+-gySz{_u2g5UA;LDLC#}!C)(h4IkwGm zi7qG6lRIZ>z$ZD;g&&(BSk6nIdH0I?V=%A zx;?69I}-25Vt$*;clL1PG+5Ug@aGExEk}42AB0Ed=}uiH67_?~5J??3oKFu?9`9tL zkPYmN1t!pNC4=f`c}iZ)wHRzq) zZF`e@wy72-eAYAp8T6e!8x1;NG==0Iyz^X7q%*zBx4%ft>zTg8{j?!1CqV%14PU{= zJ9nlH;vbX)P90{sr|`2c9?C3Ui_)Jlrjv=+H8vn#wxZdZJFX0Dpy#G!m;w$UI7iw& zNh}CqL)}T*BFd2V&-z66M@iB3#F|U@dPa7; zBGPKDB(voo5!2WkiL~BvQYWvS#XQldve|f$uN+7Z zxdJk4XoGCD>qB3YxUaHfr#nulzSe0t`#Op^%vOVnCf1+tkh*_TVR$APmnp`Uu_8Tp z08dDCOZGjo4&FN(N@Bc(N;D1%lMk3X4wd-_eCD}oo zY!Qk}({4WnuBH!0?WmtzGPYWd$QphafVy~#HQV`qpDsq`?+7h_t*PD>Ex-=-`Ee>g zBL#R*Hr@2`;qg|@Rm}hv%SID}Wn?^d6SU=oYP#6Qh2`>~g`?PEBfdsd#!Fk6UGl_7 zCC36B3=Z1R_a$62F@9uU-h}08g?EIsaS_o~Kg$*)$xis;GWK5bM8WJ$BOJCpF5)X3 zOnimx&^QF2T~C@Y$@6&LL%r2xbZtR=REAvrDqn&!`+r%%I8XPIs|m`qnkT zFczR%F?9pk@hDk0m1T;Y<~~yb=8NRQezL(OWh+^YF+kb+-4-%A$>!_Eghv)r1Dpj^ zZ@o(;a+DC5lmYAX3W*$ZaWA!mr*dvzPoaIdGg=`a)n&drqj%aOS7y6DsUJeo2!UoVK0QW`?Y z+*_B4&oCuqII5to9Ky$=8Ce<|y-My?N_WEY3@eEcu?Og zge({>+8m~?QdXSn<96S6-Z=!-0T8zCx&epqX)!|JP3FwFgUQCwKLhd~;s65JE7*9U zcV|7eAZX~^Y;RM;g4FYOMm4dw>9d^(vxWCmfc=xqgT^+l$i>J9AU9`tqDhD>WbjiX zKZa-7^OVt?oD_Y{{JWFGdYLA&F?DbGES7e-1E7(CEK%n?p1s{YWBKmo4D5=%HD=ae zz2cIw)kwG&eCnarnr-1K9GQ0XBz@t4$7yttZ4+=E_sJt;F?0Cxn0Y<6zBo0Skl zkXgRT<@Y`9cl0i9F29WM$Kafo-M5i^|5je4w0HfrKiXSZaHZP1`y_M}$d@(=*-;iYMqo82>$Q~IF%`By{HMr)?~hna}hn_cDC*0{#}FA-)CmI4|~PJa^mV5*Gc zkdj2Tp$5?1cf5p~`t#(#iAS3(-3@o}dG=^Iaup(lBUi>~kyvRG(xcz^axXx*zy;Gr zx1@3porUuDk3l zZfmU31QXwV_m<9|<=@?4!Fr7C$A#)upfL09T6Fim!;d`dN2!S04m52{E=F6a56hY? z2T~F7bp1wtTd8}Nwgz^}C@`ALB>4^Q^R4j9oe^CqwuctQsqc?_I3{9}$%ur-pDYhl zgYbJwl^8}S5ZC8-{wQhh#X|p(;bnj{_J{jNSH-Qr-!K?^eGv%WCpAtQYmMpAYuW+! z-up3wYay0VT>a;H3#F0pfF9zE(AczR? zlMHw3Jr6#5aiU819!R6iSr)dpVtti0KC%jP)FwBmKx*swUc`E!6t%= z`omb)GASUKNPKx=mAmGJ_di>d=yd>6+c&7%Sa)vaYu;)k_pZSI(El~Dj8x}jrU>%2 zmwk)3P5h@lP%kd}c{s>IzwXr+>|Si<#UC_kKrtXLj?-{>&iPbAZ@>&$Jl%w7(l9VF z;{q+@v@RaT-KZ(Mwg*7a2|HZMq|GzPgRq54LA<1R3=?v1Lb)5bte45`AI@CqkjDh9 z-qFH~YX%tZpdX5ol4=uMd%m5(IkBSwj;S`Q&t(Lw0u=Ttr*+<{n@j}ifE&Pk)K4n2 z-g`>)=BfH_yZ7xAXONxdWTZAJP;BOvJ%Bt4wdalsYr^S0F?&jB590X#=$5}rUgo1l zdFX~4$QG##fYc8vO!K(qn}c^!p1&tt#;!7{AqbDL7^8i`volQOgIi#;5n*mqxg~xI zY?D*GaIF`F^(Y3aP?YLE>Gsv6D<+^v3X*)r1m%5TAq?@z&h^@@8X9wR4p~v#fnLH4 z*2r0&488DrhHZavI$g>DOT9wDmvnl&-aej8!+)c^sC=QQ9t(a_16Ed}HD= zA0uLxIU;eg&&tMp$reWL1&j8`INE&?4pK_M`>o*+2#?N3kJ#vzLyBAxnb$ zot=&UwK;i<^FFdAF0e3&CQgp3OKOIRB5Ip)ghHo9WoBdkv3WUxP5a#hk~V1#+z72h z0kiT~1^UQ=P*ar!GlUJ2k9z-ojJszg(bZdWCd}eB#9@zu_J-e+AiYX~bUs7A;91=-~mK zS}8^DJTKjSPd7(HLxX#j?I(!#7Ze;hbPjw>5BMmm&e~SQJD#j9xam%eZTq6bkiD$8i$WE!v}btEMM0R8 z2jeV}EqJ|DAF*#=J$myvB6kZk9 zE_)@g8D8O?H7N9Fs5q4^GG!>9PCa@ICS?D{#Cf$Z8Tijb3IY7J3@JpFT{biM>iV$x ziD3dqvc>Y(xTh<@MGllcEXEx1SDuXRayyJzrM<2P=sKRqD}I;o(B5IUI2n5#Q14(k zgb_3wsc(eL^zZi9ARMX^*Hk_NnS5Wn7t$|lRs!&D(p>&eCEVL?PfKZ_5ZGb)VIj87gPuz()z7VXb&(OjWtDCLq#H33T0@5|+351u z2qzBtV7dx@@(U()`YE-W57w!E0GVX+6|iqyPsB|F)?t1pW966V52VP}%V(IIb>F1$ z;(Gi|zq#x@{pfl_oXC?a+wDg|P=Tmt^8!(i(U;WR@FKZe*&NI3dzU4A5}NlO7d@mB z^xZlHU?evZMXO(ZH4B(YQ$$ZQ?&@Ve;SCE6%8jtrSmBl zJyavPr-7n<_1caE8D6uTV5_<+o5g7aww=_N*Sih0_o*&K!P1V|3eW_XgxGz)KT&|OyYH0B0J);!@9Dw_2Ha%SS@V~l}{w!Btd zjNcQvWAK+9a>7iIDI>ftCpg~|I8&*J#DJibm<)$~LVC1^$#WQ~#XE)CcNV?z@4bq+ z0S1vuA-!*}BLvgs@B{YiY(8`=Ru{3FE0V2$&(DYhU=^vd9ucssZeR63!aoBCCF^^p z2IVp$LXiOe!ART$sNc!-Wa#s(*Blko50~}~V3W^lJA=atpjyyrje9^jhU|okQ4HTe zj^T8L1vyiJd!X6Y#Wz{j6(98HYBFWN;sno0OyhiMn#aP$zFx-k( zm~yPH8#V1v!mbmW*H5l6TsWe9O*x7;#olHnPdAq#xuBI7*P&$uU$1@!>Dj&r*B%Wm zyqWjW)#geVxdzC3GTuJp6I zs3_aw>f1kqG;O4&`=R>N#rBW`Fy>0W(*+4w#%Y}u5{gPkrTOg9ET){FI3g))r;brx z9^sQ`Ws=S|l7b3Nw_h38zHW+M+3U!Oxgn{2*r>sN??Cpv|q?L{66TD$!8vU-jh*>A}!%@eRyqcqIlk!3HvV*w$! zJ1es51D!nCY4T|Aq)`(|64ldgC=K7UB^)J^q?n`@IzM;obesAS$ay}*3KmN-SetPw zX_A<5xtUkANEVHjl$z08Ymm}DC7ES)F?OgL^}xab5z|V)y>b-`XOn5D`C@wPV)r5j zRyBHJmEkKIGdnk_7H0;}aPPXaflVcN!>57lzHn@JSFPh>#`5XN!osp`!ys9aB`bg67#T?kC}ZiXpB*6 z{q2(UKDP1k0BM%B_^wwqrjsC19!l^bcZ0dp+x(BMTb`|_vPy~Y$^NYn5$+jW&ZF?W5jUzr^nrypLsyJzpW^Krj!a8akE z{v-%Q>uc3P5WH6Z;_bu5!+emu!2Otr$Ou1pn`=9ZP@ednYCWt;^~2q*aD-eI@8EVk z5_2LHd}Aoxbvfmhf2hl#NHIt;^43=PE|11XtB5BTDR_}9FUrd0l6Dfrg`wwTTGy~C z7;vF4FqE`)8*5U^$AT*IJCsuxRQ`Q-E){)h;dE;nyZcr5I3L`NVj~0kc^%yZr$MUv z9jwmcVf*ZOa*))7wk(JZ`1sVX6|wJj!nTPYu@eqTpk=%r3Gsd=yF3lDXg+9ZWDQU# zuj1Xl^C(QJ8PU5@j*>Kvw>cxbz2AI(ZlXGA3|!;& zWc1cPWlM2uN7<|B8XbRHW2TTA{AsFt<>$2jNK|H>h{W+vE%{K!fN)1JT!dAog9O>I zzTAroP4QNpHr0Jb?*ADyFP?_|I4V}foQxz`crS7^!qg$nwo9MBAl1rTg-CM$Tdb3D zTK@8p5hiX?tvBHyTEEkJb)_2JMr0TZne|i!_|BKgcFDXrxm*VBPP{#nesd+(QOO`^ zfkAKgM{P3?N0QXjI*?K`Tmr={lS}f<`$+5}DO${7Csy0(8 zcD5z;8hmto52vrP{-&4rcn|r8tcb>LTrW2j+S?+#w6?r;oyO#7*(6VVrQfuL<9^IJ zIe0%$s@|+a>T#%>J8~x7p=VMCEw%lnwL;|i=KF7x^4PKrl-5FxHcPuZEIQgqA> zyDrI^7o>);$76u6^V|6W*=4ki!QC4dNbwmN!)h9qe?tx=PxvqeYLBx zu=!)}remVCWsjqHt%+P)NE5x$wk?X<{sX7f0zO#muGp|`d-=(s-k<=s_HjnpM^oJ@ zi) zpXf8+pg34>-IBLvc*E@;=QSQ#U7RB0Z%b*QfDUQrJr##9_d>pxg`VHLnwUh*x|q$< zxkXRHyBYM;ll>Y=miI|A>UL##X9<PvgH*@?{t`%`Q4$edf;KDWH6 zHpon_RuJAezAXrwY=w;iz=ZI(J#FHr#!%(7m=ytuCK|u9BEVb-)HQvow2RWV$~l7= z`Hb{@?A7WMy|a|g24q~7u+5A*q{m1edY&tH_J+SmZF08?Pmbv_pua`B6eZlnQYvPn z7i-tB%LCOnWS937ku+gRcO6Nou>`T5dQ%}Tl#n?)-d`>IH-TWFd2l``0PTwYJbruv znYWE%uzz7vv({V`3GXsr8yqb4VGrrs^8R_EedZ^+7!g%1mGf?~)^0w*H`dAr(LWi! z)5K$$k{-!q856O+lE69ZoN4@Qzc(dvj`tDy-TsO3;= zMi?{&l`vlUb8bwa81p(L`J2G>sbOu}`#Vr(C|*I#<;wc!66>!-9!mT)h)SXUouRp` zp9f5L?}kV1tqbt}A`M5ta1`tf%foc699(E#rD~08)+lZoS_}iSq4{SASFz|GI+bxZ zHo<%zvXK&R6@hfBsp*C)k6>DpbHk|3VUj5Ea4Nc~@wL`3RC=jUE4Tl7G|K{t z)d&mZB5*9xD!T!NoE@c7m+;cFf9=z{xA(=!{MzRvGk%Bpd$pOB#OpX$TOn$~*Jm_X zuEH>))SUKAe>w@5uaX7wK!9iZjL2PaTOrpqv_&vjJX(|4Lilc~dvOH@YTqM`M(51z zGJbjWVchc-speF!`k(Er4j~k^&!M}!M0qZQzApN;$c|--#2BfamnS`o3(x2BI2>3! z;hDaIMn_b)BaJ^JH4dis$c%CP|8aIe#1B}v2Z;t_8D}^T31QzBtvH|9o;w->vn<3%gCfxh}F4XF3>C*GgPb{}m>YL1z|8lWFDjKGzhtwO$Q^8!nlSA2!&VGZDg~rp7 z!7uvB6QxoD=?TqTjW1LGj*e(N)3Sc}mh{ z#i~d3{?&CpIxN2k70dstfVsN52d+Tz`^(o9)ok%9Jb2#Ee+LQBghBby&X z$YqKs>{6N|n<3jQzY_Wk2I(vameFzL=DnYro5`8W2@ysv(Y+?hoAp8pU?@C6i`RTo zbCC%_`-?LF@>%|p%Ad~)FtR`N|F`acExcR&0IiT_k@>LD9`|G#Vb&)4*Sg^L57+5h+X_@_dxZ- zcfq6_-hpN;)(adE-S|AtuJaj*L=p-IrTm?E{t{TrFyqkTVrIr-4X2xjTkbf!?!=49 zJc+65+){38(Ko_h^y>NhQ~B~KtnVuQb=TiWM;eS{EaCdVezuOUKUqTT%O>KBe0x67 zq0eGd6>2qcw9Pz56=2xWBjAr>{H-x&Nn*M6Jy7*2gPW?g+xOdiOWE|8>HugcRCTXS{E(S`pGCZ;aX-7^0Y%!-FwprsCzCp9#I}{_q`7ZT>4`> zcm~+N)hTg7NXtN@&*-_XvC) zc?p>h5^}#Knk)XpxXXtTP?s!r%gyjykDX=Lt+V5t+v_b>qRdm|g8fSnBm}ZN5xEEJ zS9Y=NC*9p_8OTxk8&mSvB~f|WHp38OA5Ln|gK_$yajAzI{9mF!L99I$aG}|cUK?~J zoo@lukUJ&>zs|;A^#5O_H$#ym<>8>8hPtblKg%tMdfAN&y&EOwxvt^Oln*oRUo`1e zs(%~+*=a!iHH^RM4@7JY>Y&E5S{1_{NabUp{(vGC7V<0I8!2yvh~~T!`^{nXfI#ja zhqV&^W(xdzi5zII{kG`I?^zm9JPGj@TTVX5Zw`iy7;sdsXn*|85AcA2c1ruP`t#o$ z%rjb``PCw}-&0&#jmLlii_EC_&A}8%1I@9XX_Nk|;QwzKW<|a=)x9HO@wfAnnrYvKRX f>hu3E3x9z}9&RGdL8Iw)1^82vSCfOtmu+T{GfKW<#rpEo=ifBLHaAJUIDgrDo*-2d#f+mvq&iFY*tQ9Tt?=m8l8eUN97{Of zq<0;{0sA>GvmEMvbjeaDP_Q!^dxLTuQvUbOd7zB{`k{=6AmGv68n@tqC%N^3rPd?9QI8Y3l^Ve9M1dBD%2sW4Og*o(U@$JZ$rLwu7@qqKLfz2QD} z6r;p)Nrsy;A}J?tdg?T~RL}FEuT00;``0yeGdbE9369Tp9$kKZ;=EDPuN%*(W1M*} zpT1MNjCh!$6324{Blyis7=GWQvGeH2ujI#Yue|o^^zc1Z`$57}6Gob_dS-TSYq`8m zeRx<0=A69DDF=yf`?jZ_{qbD!Tai46z@#a1v0c4&k zb1a%zg-?+bdM=2S2gGkFaoQ$JcG*uM#L9y(a) z(E3C8R%lUl>t+soPp%MCouMFBh!>>t%pR14%Ris_wT(I8xV}C$eUCV!(_8;hB>@=( z=hjltpJm7LJUbxs2aDAEM~WwIAGW+YK5B+PM82na?Gd}vxdYE1X=}+}7RI3U_Mo#>W`YPyGBin<7^q|+KwgOj!5l9meK8*+!o@bBwgV6J)Z;l7QX34y& zyTcy0^@;WVan(-}bcUwZ=Wth**GI{&oCoAf=MN-)yXwFq_w3A#^5zrXLgUBSqrh4g z89quRxs$E1kgulK`F@`Ic{nS>+x1QL9i?yWnD+t;M>Yv!%dI<8=6CwIsI3iIYWG@F zs`q(+ww==8ALr>>Vzq=e;@gToMLuLmKM~5j(F+O@HB{=6E%q z=-}=n7%TOkt#!d_Svrf(8YvoO7}?(ogMWa)+d98zNM*=yxH`>YR90S|cUZl+zx9-q ztC%a8d9}he9pfZ#&$6$LzU7ehqynwIio?%(?i$XtZ!mIKr?UaAq=e_so z>iKrCPxadMM_z1d<6*SVXOa)_or#n8K4ABm{os>3Cy!2FJ)QhV1Mb{_SJq-Upag2&M|p38sL4Sjv}BL9nPZ(iTKcp&~9tCQ#zA@OVG!n5DGb9hIh_;rnhd0s31Vp}^& zykU0%dFQy9Ucj9T{j35nO7F7qMz&;ebRYL;wR$9THR*%%Ly_8J#oERnlDhdGs?+u_ ze__dsl#?^<*yreIBX!!KrC08(5}#y?K1RQY+x;1Cn|Cla;DL~jaMx$c>H{`k&If78 ze9oULTNMdXwTl?~nZN43rCra*(0zaE_)*G1cF=uhl#o&Onkd^Ycl8NY2-US3E|=k%fbR; zwju$qL9YkCi#oabo>WbdI?waQ;D`E;*6-jSxbH~uR8E6vF_AaIIk)qrvOTgDOk_;e zE+Gr7?ztB#E5@ekWavp}51QcakniF@78Iluz(4AsUJT3?CKN22siP@(ma}UsYctfq znmV>w*n447w4OXr&P8TANugC1U1;8zUdy&j{=%f%bO3s$u2 z756t#ivq>lU6suhdT*ZxgP153E7YWhc8;WSrA8@wK`0O9%5@<+ZnJLEJ);Ey*A{D( zZjI@y-?9GKa5V%)5npaS{^~*3)Rlv;wqJR^l7Gb}@FBK>%~;GJ|AuRo5z45lq_X5C z^a)8}f^{OhcKP-zNb@f_zvT90dWv!iGV<(+nd2NMtTl=PhX)>zHzOCfi7z&yH!1TI z-Siw>FMB_3&_2r^jli=gf0BQ$lzq7tIzgQ@9nm8!b(wMiNsJ z-KBAs*s9V>{c@{^W?q})R=8AG+xxbTVYo<_8F%^SyxzH5>7GQ-G|##(lxP7rf?E;< z`~Ws0SOF{J8T0z&#dqz@Pr2(?KKOo`9FjFb?>34TkcYW*^vA)?;-g7OR@Ctfnx2<_n=T~XZY^7vV zK+)p$A(tUElmxv}WmRPaJ4U@n-6T~Ft1VrkoTk9ppF}mb^2dqB#brV>;hA(z#+vU+ z8)cNZ8TxATOi!|>qhLrHe7&J6sEYn6)Fe!aUQSP?t%XU332mp)kM35|HD`u>OI7Mr zWOp-mocE%3@P3*#Wz}N)KRudyH2rAs>hkBU&ugELTywd0_L}0gi7zEzl1dF^)Z$EJtum1J8rA=U;q`I~ZU8 z-7KD(U6h@7L*Rv|uAUC~hE~j3E_wMm8>tI{CQDtzFrPi2_Gwbf*f+`PG>-YB;ztLL zk~t2X28&k4@u@-ws|{`)zBQc_>15)2eU0)fbM^7e&@1vyA475d8r3hsU(1j3i1Oa< zND1;k9d;Vg`}p2i zkgh?lMQ!sP8=X>}OQ-#sXQ#oI=PlDpCtdYi$lpnww>n1N8gv~WFuo4?iWFhE=;hJJZ28EQw%hla2s5h9`G8v zZRMBMpX=XewP4jeN-13_jSS#kOhnU7GA+ZizcxDHf}1v94mIM>je7@V<6A}zb<^~f z!hro18e*`LSgVg}>n;9Pylsty8Vm=Lcqp<1V<#$;JQ%gwrP&PIsMWyjI@Ujc|Xq0CEC@+})Na}HJ{n*OQzo=>FT~BznabxJmtG97! zVv2Z8MN4a9@5FL0@lJn(jCe@R9|S6w(!_ghS5+N!b_UcY7_+&YN9osq7N4sT~H zuTRua|DZ;b-6v(k?r*(cgN@;-<#iQ8VP{r8&E8$gu8TVqXRi^jwcU%E5?_W8<=?sv z;4+ZP0fz(ZwjFnb1Y~n$K`^e}f=vP~NlBJ{ZW4h7rAgP=JVrDa0gs_$@L#bLi=E5M32V+fSo?j)FU2Vl zS?VYEu}B|fIjo^<(u(Se3OjT+142K!Wo!`;xg-f?>HmtL&C6`T{Ej_p7-RYLgto!S z^5G%NH#m#`#{3sG@5DuQn8dMDnJ1L=!VKFD4>0mRcI9YoX_g<-H{-cnzRUAe=_$u0 zIXNC6yGV4lxaDGK$RYt8|G{!--)WWuz|lV7AIm;omP3CYv#{vz%=ZWWaeSB?xW;nLRQuK~ z;M3I6*V!5B_t48fsA6myIC0ebrllVX3%3aKci%0Oiz_TF`$r)6Ec`7DZ>u?adCEL+ z@^WyN3G(!2-iJjaNDVmjboPI6F38gZ>ZcZ@Dg0*(HQ<2 z@^wC^By(BjvM`AC+_`fazD_P`CO35d>JI#+Dg4mi-&;*q76yaKz~p7Td|hSbR8>`F zFJF; zu=uw<{`u^`_AvH^02A?mIW&;mUr+ewbAPqhkYz6ZKN-VsOZ)RIU}_*%4cY%OTM+Bs z_d{=4Sgx|%x^eAZ(7vUKBgvdPlSif+A05>()_HQ2<9?T`1}@$2b9z}aNAyvSxpY7C z4of*gQNAms#C~+mFo?;KQ-0|FtFXl>gHvn)Q!N zo`l}G|8w%s!TkLhE{StNc3EPR4>{#p9W3nevCH2>(Fk8Y2x-&sX|D%Qdp31KW58@=3 zo~hgX>BsK;*x$(53(=sDzg+#1f52hB8|~mtBvi#W>O`^0>$b8jXZsP1Fgv~mN%*Q8 z`9q9dm-^i<`fLsRSv3VL<7aE~0~#^gu~mU_wmqwrG*Mgmtw4_zie2fIl*@6-Tr&0c zOATD!bGJ`rXyn3WD}GJ)6q>b%pTkw*!+7k&RjaSfD0z;4t6%qb|7738hS=M7C5|_O zKZYcYxWqv6AmESWXTy*wf_)3S*3N6y%|krNvSnCR3B>m7(ANBn@fBD`iQ=+rGN`Or zE}?OkV!cCn8pm7qBN=OIr0Tpq+5=rEXzhpJHNF4sx&3sGDJ~X;ZjkGReG1;*9ExA< ztimVkXuWdH3^|Y8p0ob4zHrs%lFPk9g$MKo=I>2xg1HEC>|Tf zd=9WW^b5h|wnO4Jp@5uiLCvaT^ z;_@}Iv8>ZfpZ{3@G}B-mD=L%#8(ijLn)+Kd&4}tDd=RR7z895R9%ibnf33T?3a=R~ z1h>fBc>^DqaL5+#&&|3nLI>O{B4 zelP5%(e_*%*eLhWD@3_b%+pWsBG7`Q`Y~+;G2Xb`i?!!^M97*KOEfu zaO$Ca5C4k;=l^oS?+*65jX)B}!jhXta?QVb@&eMXXH9CeEpph_cJLa=V1|PNKr=MwVrEIK`JG63m>k}RF)f6P52l~RV|q~Di)6!Najl(Ox3=>Wvr!x3uB9o z5~|kBOCx!8%0)DWB8lRMaGDGgQn;)dbs=iL4 zYs5#@1ni#A59gt8UX8nnHAD_$kSQp?U8c;*CB&fQwqzGoll1-JSYVr|q2KTj{#zgJ zRoVbKO<^CO;ivs&uoUS*o3PoAZN{BiXpY$RP|C>XGbXirxnwW9vl>nhiQ>njS{1=I zC*R(_a?7sjMcy6?jg`TFeiv{Rm1wTSqlSBP_b)H?AE!0}NSf`$T}OR>Z*taC)oZDI zesTUyCk9`QUBd;)v|K-6xZ15gNaSTbK0^R9`9uzdEyO5CbB1 z8f!85hp2UmlyT2iHP4w3)4=X}X{^QfTJ;`loqhPKYJmA(X6V`v$4Pi?0-|y}v~O)( zYilaILrw;}wD6GlW&La(QU<$CspDN|z`s48XLM&Xz5KvOS=t;aMDY90?GCl}j@4}* zdu{>UMpn-GNl%i>+R9TtjkWXI*l18|20xXGP^lu%7;6XsTU7rggKbXm>IL4&;{vH; zA@Mrg9iDyv7t{Qk{U>qly_pDqj^oAOEhV(4I;$pMFu}&$mq6kb_w?{JpMdLx0rmU= z91I^2HIMmmcmU;Ik0mCmyVVPUy?Wj7yHKpAUo{g<)VOBc6Z5yL^J8^EdM}--K(*F4 z!ZkWIm!}Hd*AFoCD4>r<^PwF%dsK(Y-i37@BtrML>MUnPS{u2qkkAwvmN7DDB9UYr zSkO^09!S?r+0>EnYFwY}fVNAk|193igrAd0vwd%x&*sAnu3Sl(*5UhaOg@9wESJ>h zy(rwH;u+sHGl7`}RT&=mZz#MZK#{b@v&p%6|v6&$`t%6||w*kLXvpm3O zq(XXr^8vYNZB-2EY!(+Vp(ljkWjdNU6?Jm8p+9+0v8*_7)+G&k=B0G4_wvFT9d|f1 zvnDvug`jIbHzHd;wSyVpPUGe+wjQc+$QP*mu|y{z5iMfK@Wek~meE?8VhSfkBsr|S z-Z6V8uPkAfH-N_jzFV!c*t#k@H;ERK5zKY4MfZNpfO+h4<%kCCdAV?Ku)if;a)blc8+EzYquM#%0;Q(QdsyUBk^G~YWPbrLJI8szVc z{3{o@b93sY?2%cDb#Q@*@0gqN?SV>9|BR5XAo4D|n{xB#$M*dNd^}v6ona}td^}g| zoPHjft{B39lrQF#;a3RqW*Tq6XiJ_0;6$132*)qn8TRIu=ns$fLL2{|hsp}u0}kj@ z45Tf0H0Ib8GI_b}{j-?<O5^u=OYeN76lMNAoqq30SqAP3Xu#ZcQTx6HNH8 zcYaI`9ED8p_!-5V(u#W?LPmJs#&1lgd95Gl19q6SUC`<*?`~h(M)?cR%F!aDu;u90 zUCofK9kny`DKB{h}y zMBWYmoV1)?{JV_?<1t)cgFDf$%4u6P&p+&;+pWDg5Y0(rwZCk3aV#mW%0m4u-2}BUqX?3bB8=E7Lf=Yv_ecMw&PnP zTl$Rt4ohBbKXp>WOO^2T_U{e!7{htLkd$m3oOHyO8va@CbwqPh(s*SG4JOO6nr5IYxPcTRSo|p`vbUCEK`7t3WQo z%#XzxFDshoRLO7MhC%)fRB{kBaD>t>+K^qp^E>8_)G~`Edm%~g06Y4sd zaRO`;rmbPi+Lp()>7K;-?zWa}Bh6~uuJ-8tTH>BLn~PxR=gAqVbjuKA=Pu}wCy&+r zy`Z@oo-6rppt_e#_E^Wt_lL0F8Ilpdn)ZYsz7;xkUav(UHx1;L#2LK{*!}d=I8bZ| z33XNjJyAuYD;s>FZl(4HFnL5M+-C1q>i~IfY-YnWRucct#PRjv$s7cq4Q#j5XK@)eq)I9@ zM3tx#!)RD!=;%l;nO=*Tp;$KqSukoLkzNNvWyj4c&ZYJ+zUUd)<+0x);{ugeR&ijr3z${2eemj4JWQzTX4n&au7W6-a1GJws zIX(H*@0iwo<1yzS6P`)Siyjlh-EiJ>hd9dgt&06Z-zl2gt8!c1pX|u$i*An*hx-{j zzKYw@>LbGh`db;Mm6<2BtTD^Ese!0@t>Y@KZBgqVb{!(ti1fi{SlVxL)O0vD^Zxsar) zN24ijoaNyrmbhsk+sNx)Ja$2*@mIPqRPPt=hM{LXIyVHy{yfy zB}mBd(}xGlw|Hn}Z7h!lbnQ&RAp!F=eFKc)`fxU9iIK950t`|@wjA`y@7@knU$!I^ z9JdVPE2M7@*|%R?Kfnh1uyZ3SlGeCRqGS4?A*}*cHp(u!8=SZCx&9}DN!v-hqxIvN zB?9i~D$DfHX*o=@3m@00=)^(H5Vk>h86_uD{6{AA`zV*{X9KYG2!hts76=NUJ1*8Mx~Mh%7>&>L|xwkmmjCt5(F2A0phSb$NXK#4%6%hQ8L^1jn+rqOz_x zW}1Tm^0@w*u6k5#t{JLj_pNksaGIHDhJWw;uun@-9;VQJ2)Zy{qa3KSR9};Sv}^D? z%btb6Jq{(ABRH_#c~SA=&`dLE!)|8LLPu@+G}QbS>X58O?X{l(zT$K0STN0zDSlSB z>N2LR=5H6aR35!@IL*VB?SwfRP2L1D1&%Z*M*C+se858dS>r@(h%T#Rtl_PntJ0Mt zxjk9y3-sHl3hE_+CbL|wbg`D+8o&JUCVi9|AtBKzWX7UlTZFAP`Ob+;C@T*katN*82V^Tv!XY z51$Rd%#?&MM;KC%70WnPH693FUieDZON!l`qzgrBv#HktccXoVC7Lg3ZEpCXzyox( zv-#fPj@pEQBf9`7S1C2dR|BA-5KO=t5!U8=RaKd+50@Gmbl=KH%d?Rh-f#tZ^=<=t z=>M+Szr+*;^wk-Y+nws*5V5}y`Hskbh}l4Sp`W?sgKg)sLhIYMtOr2b{becZmcc(n z9M*n_;XK^M6CksFMZJ~F?=at{UpgsPB-i5gzjUz)xOCVK>;PH9^YYVv+9DZE8c}9o zI}E>v@%i4o#B$b+y(0pGklYcB^ab6W({<#4`x?7T$4;;*dJGR?z~c9kDMdiWYGtZ8D}1aUtaX2i;L~h2G%zGmwmfG|j`nJ#m+=2GLVZ!r z&AIPSz7U6o^#R`cR4YG1KZm?9oT$avkq;Ia6cdCm{SZB|IkWO}wkF3X)r5&R06LMK zZ(L!w&bdq)#_vklUdfJ)^IyNAEuxyAf9=&vUoesU3(jMsBole%@ZUoHnOpBWgLsEN z{k24okD5>Q_ezw&ojV7>UG8qDHo>-j+9bDL_Y#c$%VfvNi*^iZ+A6>pIV+*-r?nhQ ze~7j-*8;KQ#kDqIZYwt#w7#s?r~jb6)j%>Fja|?W7%c>AY^*r0_TjX^Xx{bU&EDYx z1;wDfO=WybR+Nyb-~}?mB?X>mu_x62E#0q65D_aVfDENsYHV(-*F0$9%+r{EePIR@ zMT!O>Qj)R(%Xm?oCNC6U>2{U~-P+XM( zet`2bPLfDGN)$zLfw%NBWFU6=$bj$+T4D4?mD9Q)UkfY4My7WY3Jg@r))f>7W^9sg zqH*x`gOKvbY&Z}_|w6!u5>~&C| zZZw3mhITY&3EM`ot86_VZxnzKh@&;1ch*~`UrO76Nc#?e;)V=r$cbti`wx z>3RQ=&R|++_Zm%L5Fd%`Bu zYMPj3jM+rn!g2EobVuywJ;A^&OuRW8PC{HGcngSJv149lV&fTII4-cQP=0#2(H!g7 zr%}gaF3>k({BUZj?(=IfiLk0Qeby(Lesi}lq)}87*dGmOS4yhz5VS7~Xh2!SANk}H zAsu#W^FeF95uG&Q&KP#i_jF&P3$;tzlFDE>TZhI2!L99RtddJwA&-5SqEsw7#ERqt z-b}j_&JOUULF=6`-~2Ino>9Q4RjTJE*JggETj4MaIpJ7i@~C2Q+)pRi_d4q;n@mL5 zn4xCC&aziuVpHn?`$7WP9(N)WUDcI}SsqL?BWG^c)?i&z-fr7F2PF9mj06ytKZFp; zmaK>K-IrF7;Sv$qQ3J|1Vk)VVjQ2a?DY*>6d}OGNKk>?9GKJHE1Y2LuCASb$ls(h^ z_C(@oId8gt#5b@r)A7&_3Q?!lC?j;iGS|oLa6bx0gAorSi>9`LLD*lIw+&0$A-Row@S!SzG^`1o>3EWx=3A<7<)abtd0pSPcH?VV+hhejCt*BJ zW#O6cAJ{=wM-rhf!F$9b*=CaLMAU(tC;z!P_~@$qu`P5^XXyT=%-^jZs_utyb#Vr^ zNkAfcwb{niQ3X2vxm=~NNl*M4&`y5GHvcEGpyCr~N=7QC6ijvmIL^t=^e?ofK4xDx#_x9E$z;U|I2lX0qyCQuY#zUL9Fbi#x?*s}# zRPpCpR2x49g&Q#x&ilqFZ!mjCJ{zQ?xN=ybOJn&N?=EWc9=n8nV9y9Z%@yE}_Zjo# zC+uH#FRpPLa0v)UY^>%7ESNN-bNQV2u-`t+a*=w)J8GHT0xw3sAfi7ZWgT zC~R32%M7h-p?~n|ZCj&G49K#LHxiN;g9a*z@+l=~2oZ_cKe5-0}+SdiwOuq+a zRceiW_->S}7NP_8J}|0Wc>BbVomKesjpfRXJ&HQ{g?;#Tg;C#T?p4}nTw{?bQi6ha zjZ>NFpOM!l6ljJC8LMK9iMsiz#{4%u3qn_C-v?2)llJH>j6ZqY{}jo)KQLWmKk@Qk za*+2m2Ti76`L@*91=Lq_Wc$9{PZm_DEec$bJOt6{4Wph3xaatEy`Kz;3K z0*{@qv-oggLmLdJv9|cM@=u=58=2zQjS6z zlTl$VRA#G2h>?aY)xQA=@j*Ws*{vtj%AYI)w4cgc*fI}nA`G!yxW%k%0(HLr0(9ZL zUw;Tk0u?+8f zE|(Mp@C{S~vX9o{dCL=`xCg*1F_=EHn3?g}!=Pe$yui`M_h(9gc}16Iwa;+34Zch` z%6t9`y(NziZJ*=*G_1GtlDLWumkn<8Tto{hv-kA?c5VUHZ|Q>#Y~+Y*$DWnir2l%m zG`0PhzWGYst3hTGu{ayFl%?LjZ>#(b9Pb!+34ir&h)@Nd;W(1(hgQ_qH-X%iMcy*1IVk##5%xI);Q%3gXSJ(P$=9s6v@A>u zFiEaKBQh%M$env>Z(2st8##}KbBDgrZ?Mu`6Re?ik|-;I|Fr6VqUBDhb3v0GVxLLB zbL7W!K-ssL30PsXUAakC!W3eS`?|4$<6+3`@Y7&{vvC>}C*4H^I7XAu@pj?F5f|td zj~S(IJ62`&m%KbRaeoXXV6uE|`Pm&^@UcvE<78@$X^HH38qa5AR8vWPuU6|TRT{44 z5Uy{$^vP4m=UZYK3X~Dl7D2t(Db0MoQv&9E@Du>vMmh$3`zU?J%C;W$51JlR`;{X& zlqF8a8i%2h%7x*BZUph$;olw{F(X@%7j<|nFPYo?6iNv*c^GnC8shwOIdI|Z5&qB{ z4Nty)#Oe>Rk!wqZAfkFE^o)UWd-@&L40QSQnwjC|p3zw6mG#>xVUgjA zQe{|doBx3UQXT1zeUh9fv48_U`_{NDj!FuKN{l`jlfEEe9AC;TiI?{oaTHrsL_=)} z5>Zq1#3u~)aX&wzS_3rZJc#e|WvaMH1H0zA^&e*y4{zbq%6r2n32R@X1Y>?`IcqhG zxJw26Z9cT4xPMbNWOJzFi`_qn=}ytt<|=q-zUB;2n){yK#N=_0;WUf+zqKuOcO- zU8e)A^mEQpl)&h1$-&0mt`J9>83HrNi)!ICy$#AMq+wg`4X#hyqZ>{!Ikpi;j1Eui zhN`SpFKg{F+4`$Fa_&h?0LVnx+&1^Ia)vk%bvk*d!rr4SAWjYSaEN5Oy9=%J;8Dg0 zyKWX9w#UO5+9_eZ+uLLKyiSjHluhuDW_hp|Vx|+erUa>^QMI-%$ExnV>)2)7WA)4H zCgtaHy^9LlHFQBPBtR0DB6zIc2b*mN7?c}Is>|YU55>&wWDcmRR!EpjOen7i+ps+pTeq4NxC91A*S=x(sjp4hSLkmqYWkXL7vy0njmmR)stn$z!1(pdt zQKDj}2nh51#gX|7%2&)P!hoV&VUlAwgEHo|3giK=;P$&ZwGSXKJC#c8%*c!7(Yq5> z_&ebypZiS>&AjmS9P&{;zVVjV{M5`g46B;BH4k)m;3jq)|ctJyJ{|T{6@7? zvfMYb)|X+Vl(X**O>arg*}*oOKUx*ZQhR+TtI~hqAbB7N?#J-ibOQnoVjwb+h zX^W)_FZ6fTq1+nJ2In~4;;$3qt#2zBf@vPF&uSQl2?cMhs!2|PQp?=J78;47i~073 zGD3(oOH!py6LR3lyc#vV2C!80!k6{)XCLu(g zL9pcv6`y2L)It~bO0;n3>41vwZ0}OF6x6G3AtLji5e$GePS z4qShb{bnA3fidG(49prld&yb$1GduDZtpq!Jv~xFNIPVm<%AcFC9XQlEk)&aimhF_ zcU%P?mx5`&s1SJ=I?QyabD$DitEw*ccghM$g%qtB2J-Y>Gi4AWzfqM)6!og# z4%{XHuIH7k!9FrT=2wOmpRos0!=jdVm8KQa<6$>7hbVbA_Ag0WD0C?r#Y|+A1G)~q zWRuUcpR5RYNUSaV`(04V+y#T%m-+s&3r;ds7LH>Al;?nq0tl{QuZNW(L80$_+|oA zjb(1-vwidY`snOHuPeD#C@Rh}CXiWzYuroNg^ve44ahb_Vb*VE&DZz7n}Sybe_g0} z`UHM{6)4}n%M|Sfd0EwL%HzgWC#8z@ZE`~C6Vw`Im7L}EudzBUB@?bUe^di`s5{vB z6M4wObCwC~{F*iZsQV^4XQU%-J!Ej@nljR$$&U}A+57g7lA67|pbboZoALXb#$sD2 zgX~!Yj*h%Y%=`>6VNTerXI)vq+thy zsrB&LoaOljM1jl^(0OwTAQtPGf?FNh1xoI*ieUTPPzz>R9eM`wHfD%#Q5!9A7!8^N z$Q!#|&#WQ8PE|c<`KHuAan`^7<}oPzWExuQBU2_9&=-R1FEc+eMf{&W$U>{btPHS$Q-uw z8gO)N8|y4rc3)v#MY|uaK|wTVZ^=(~5GvDXe}5029a2@2;u%wr?y$CVnX&RyOVQ<{ zFJ35$9XF9ji8)|enQ^7}giwWnDJ8Atj2#Z^PG)jyC|kwJ$+)&{kGLv<*r$*4GXhiR z0V2ph-KWLEDgKbd%!#3wA0(&BibM9Nu7{YzloG7h7tt3S@x6vd>>G%joIMW-Oc#MQSj3skX!y>@%nxGd!awTQ3H!eOcyjF{U9WQ0Eg)@BFLG zchq1iN+|r5bk&5&Wa#2V$BQb+gqjQp(EB)6B>bMh3Uw@nOWP7T@kJYZ@ZI#oMYOUayFWA_^>u7C>=lx+rknihS>Acc;1 z%Es6)Yn({`+6xNDPQ#R7;n5S}(531ki}#K%dFt1tsF#Ca_bApM@LWQe$N3BTOSt&}+L^$~x8jipmBT0v6J&Fe_h=DFm2~o%&A-Yo_I1&IJ5l zD@W0TzfgSQh1cf0y7K~>UpUL&H?G)Nu2R6nES z)Bp1bK~dQ~tTD|ijwwtmcL*`xrk{SgZX2=~*pLPC_pCFT#+$oJWKPJTEY=e$YyS@(3rRWbYFRDgT4e;$P2XP46HsOrI z!<&;O@{4_}fXYDWy^-~#VwRB}SIvG8_gd(DS3#fp6B28G_j9|?DQ3kjC9fGF%`lHDHI+|;9Xm6oz;Sa4giop5M;J@eq} zDr9-^t~$9MIg)M;u@A-QxRrM*dr^7GsKy#fWE!f`b7ix$YWyryr}uQX5M}`+ZKy_p zb($7q`w}(pxm-s5NAM$0!%QZaPro&br7x8|PC0_y8%z&tc@c4)0C^OrRS=SWXYO;S zjq1#+WoE|b+$B5_r2No(2HH=o!F|E>mC(Pb66{YKYp#dx0BAMEF`uusRG(JWjsSf} z2+Q%4;*gMj?>5-Ni>5a z7*R9pve*u{XvB_V%0sQD4aHGFLYTmuU-<06&o2EXsith{9hh2)$?n^4TQ%&fZihUfHfpxURb0U zAR74%f1>N(d|d;$&848D8Ts%@z@Pej{-XCyFhqcp$1YT7N5kR9HHk~uybIc5ga;BpS%WKXamL1W^tlfk7=zB4tt>;B}_ zs$1#z)POICjSt!3Weia+KthKjn^cg0N%szYD^?bwW*@xNhjz+Hnl*g~s zb~ZW{RZO}t#mTyWr^jhT>9nT5relgj038ef0Ynol-%_2TkO-MQ*b*?`neSH?=pD9z z4q>&Ywb|TD3n-wK>C(kNaz z?2;sRp4vcNb*x>1y5OfVjxiZumV%&T~tbo^f3|toM0Ylbe^=!uBK5r8d!BRYmL@MZe-Z8p_pGK z7>7cf`v65&8Xy}a=^oijF|Om*dVgZYFbca_Tkom&!u&+qKA#RanEtLCnv??yi*WpHV`N=rHyd8_%9a zwlSm6u25+CxO7+&Q_onPg)jE`{vfRF{Ex^=qH*)g^pY}NR*f<|-N5w}8lTT-3Ly#A7vm-VD>IYwW=}ePPx`~( z=WA;Rqcz=YXFJDU{lldO95ne<*qW%W%ZeZ*F!in%;xsMyek}Ig*|h@XEKHJlDe;Mk zJsYo0zI22;u3@TI#k(;xVAMp1b%#{jx_2Tn=g0}((_`dxP2IWA6%wwp#h{EQRLj}) zb(t50OhkcEAHhEK?2%20h#tDFS$=WhG*H}q$+=Tf_3pYeASA4tlsTbj2N3^qv+qp% zi4Ams#ELGB_4olTcgg~qj%K>L#Bxz$u6&oR^5gKe5XLG zQ`&LPk+OYQu#M-g%EF4a!_&<{UcYfO$DDyDvNcN!t(3VALSHP<$7#!B?^ppC%?W(lo8)G#xP+s-Iy&rjOl4OHx#l539=n<2UR?~5#h zS+xiIF7!6Wm|HqF4Tk`Q#dc|>v+D<+JFNi~i3G2A2FAyJNWvMDx0HE{fqMNcUxS-h z+MYl;P=3>`rBKKH2y*nI1?n;kuhWh`8YOxakHj@4W>Xu6ZBI3jV zP{&`w*)zu;1L~)|NkEmQR(vmgf2E^oZC8eX2T;xQn#l$4C8%%gvTMBI(?IO}aOFLo zk^k5}e6wPF^`(N;Wk!9?M6W&?{-+9yjHQF*>or5Sj88;$85}^Y{A`^?)`xKxFdXku z3Rl@Rwr-NXgAn~%2x^ii?+(8Y|+*xFUy#TF){^Frn+L2I^KR=bJ=6gH$Ye$6-Op$u^cNWogF3`rn<-FJ#8b5c)QotQ35-5hA zV}6qa8sc^nRmkL<>bH7Pn-tVUG6g|7E7k(2Joq7I6Qd%6v4gbi-Bvk3on^LP_9uRhWGit zEuzM5lQAj5ljc}HE?@8jD93Wj?QxqJ_0#;e1vG_8HQ+u1=_rRoA06mtrvFB16pep& z_ztPI?)~6*QK|)!9A#9Xw+euggMBZ>M2XSC{y!KS@ZQ`uxI)Rcq1n=?SU@5k54ptd z^gy82_lRL$t0Yvu9wv-TuRuX!qa)~->krq`p1zn*W<}2&d5<+HICEza>cXX!k64XZ zbyfcJ61hj^Z-AmHbaUYr#fr`u6%F~lxx019F`N0sgG)qEeZzVgQfBy0&v(dziX$9j z087!Hthb0=W2F|VRE$3(Yr-rjxiesk1ewG`T;OjEtLzz)&D=F`<3aglncuF9An1vp zCfTEABl31bySDjR&=@E>McT){)J2K4ub*?RaTVVeg_~enF8t{N9%IS>+a85lp9gXh zJihhZzjMZT7$D@lWB7$+g!+6;_d-rU4qe`wiXnl>@H!vH#P9XBq=;M95<$T7^ z0y*C64M8{2AyXw~Kd26`HmGmKEJY2zsK^T(&4-K&qq$ z_oLz?P<;dF9Ix6KUquT^ews{NbZ}f}Z0Mj)&VRU#N0;m|c276e-@nyF|1fZ?t$nss zX*;^|i<6QS5ZAT!DzltF^xt@Om-}~i6B*f~TWRI8U0oYLxbZ@~Az}zy zfIVps|F}6>mGuO_xkG5E&=4f+Kov1^kX82c%@8rrst*2KWPP}`?YL{Wmnw3py-Nar zz3%&M8aT&0AJ6=8y4YX18X+`B7WNzkU@x5@YGu|t}_sr@a=tQ z2M|ezEr9e_Z9d!e!x9Zms)xtk(jOwJ?FdQyz5$;=K$P|G{at*T=FQVB^Wkd%3v`}! z|Bm}t_P|}YhOO=T`G>1dqXh?Y&gywQI?amL0o=H>jV^H0wSPFD>0xRGvamVNd+MjBt?7nySA6-I92(h5lly+Q$=zxLbgk7cHcS` zNS}NQk>8ahr@gc}<$%t<+*_xP-o;EM9GdFcql>EtmtVA{$ zaL7e{>dNwy2aI_&b&J+?Eh_Ce{@f6;Jh$DvaG6@VXqHvxCtK>Gg%(*K-mQzWB4m8tjl6#r;3&i=iMKe^ifSDXG;)c?G- z|K5xLQ8s_W%`X%2fB*R#ZvKXw|8J(>Dax}i@^`PmmM!!|IE4fvW;O8-+=M@N1hnZ& zfB1==7W!}D>i;zXz$e+DXAB5*>gV6R4uN~w4^VIcYeAml?7;uk@qhayk>9)x+bOpb zlm4d%`*R>KfHDrX@Y0=s>ofYpVgAcCp`8psX>r^G^Zl<*_Thb?XmK`P;}1QBf1B37 zUh~-)NRRzJ#eZ3Ye^J7}SMfi^ihnEWe>4WbA%K7H#s7qtf5Xkcg73fK^51atH{ASf z`v1sB`WsRI*g*lnAKJ}M>dGxHsS`tdRoq>NoJKKU((71*Jh94ZizE&ZZqxC#j;CM$xpjOgYxH>R*JX7D;DMiCi?H9^dgN07sSE$Pdc0TJY3&pC`Qe}6 zJc{+C<4@!9zAfE3v9P-0SRUq_F8pw2o=2@8CrzfqZzfItx6zkuM(9jc;OBmd=;Qew z+9Vnpn#V4D>FCH7i1?eh_LMQyV&r|I`$?2jM(paJo~~lH6N&TX`#Zhj(aeg-+llx= zN4}J5?pt$?W@bWOJdp`GCvacypGj+6FmmG#Qr>lcm8neh+^D;Pw? z*V+A+G3_Ucq55{^cSKT5zR=U`X9P^0Ar61JpE5m7xx`&V>3?hGPdo};<2=jG$cxWa zk2VrdS+6DU+2ZdbNtT8SL=`afh&`LkU~0jx;q2W7f6in2TJ|7p>tx2{72edmTOLpn z;oZ(}wh)_RefR?y{C5d>%AF8UTe?s<@CmtIYaMr`rd%x{s zem+PZywv;8VUH-1Z!h5@2^-R}fyb*WPZgzT+tbo^>D&e%WS3mb&Hk#U&L$*;b zjO!b3in&iz4o}_#|HpZjdAPTe1wq_0B~3VW_NymvM&8#|b(H^9qtlb7Ht!)2S5TYl z3S-Tm!GiQTDw415Y4lwiFH%$Zead?J)A~<7RJ1=*iZGs$^I4r3OyQ_Dy?aq5+G@I? z?4vOs);*W+@TE56+ymIFS&e4dceCcFkEL~1BYOvDa1)o|bANgL+)RWaA&rkecfl4U z`kwm7FZujN@`2W*Gan)9~Dl>$o8ut z0!?&3g;1$yj*~m&3Su;6-V)tf5WKD2zWo#gpIr$WM_$|)Z&Sf!&I}gxP#K!u$)#`~0 zi_xAlUpwO6=WK(`cR!M0xtFNS{CX@W)6IzB!=(6p4Jkdj{juaGu6SOJ(=1wg`aKp1 z8TSEy;3aw2ZVrz>HJ)PB8%6p*l*{>4Sg7i~I>L!3Y8xA0xpu7sUuDI1v7rZ=7xhma>ey8v2>EVrbSMk+k`=%k~J&9TJZyEXzbNG&Y-IFY{b z3&_;SHrYe;A4k548<^qz0w2>hP|3xXs-FJkci$$%X$WszN?G9NtZbb%&liaJ4a;pb+UR;3;g8|nnjGN*$FzIvvJ1-|WO zb1QkhA|qd@)Ju}`YGdQE%@>rRaX;8@2DY$Zoqwuos-|dyv0}QG9EO5r@S`Q1+~EZf z2)rPn!8|jh&n$e(2;xF3$#rp#3=5)^#k_oxkU*beOVTYxK`Y&lNF8Z>x6A7GS{l{Q zvrAZZ0F&h%ZfUZgzFtqw^mPc^djkqF5RpdJ*0hghDhGk*SfsfQH{Q5(n)r-g#`qIf zS5rVnPW31lbhF=#$mgnIb}(*}vEyrgm!3NE40}?R?5ZHmQhoZy z<4f#Lz^(hie)$mk;Yt{D*OoAKCi<2@t+yM8F9;!#YTq#vhN4jaUR7MI7Ad11-oLc8 zwAOznU_ZocH_Ri@&d$yqmcHI3yHa{`Bc%q^sKKZ1b+aicHI+|cf6d3LKS|(b@c6KJ z(+e+oENjWs*YDx;5;^~T7Ar7 z$fq6C&W8Qr_R&OD@sQ=rZsnF`&&OI_ZO9jGc(PP{|HBhyppT3&is0cJ>?>bo@khYZMe=>p#}oZ(EKrbCHkqu zOt+_&t+er$2P{MvIjB-dnbqFbW>TV6wI)$sQRU1wDr_<8+0ogZuo*>_Z-H8q2dB%M!k1eC(J#qv(`qV9xCFFDB zPbW!n=X?)BE4T&gF_>S6r1TI!R`g~D`X0n%;0q|PQpFv;^`2pdUk=GzMFX$0hDwZz zK|I07K>9$uHu%fxIVY5^LaaonJ4|sTbwN}N@R6_56#L{|FuB6t{kFZmz2hSpwPjnO z4*Wh%`Py(**YgfWf9Hak8G$hf8D8|XJo0hS{_w`^V}1Rr+-e0w7fBp7}Qsh08F9*hMWh^qf z9Btg(aE4Pv{B~lQMf-hBaj`A|CgZQor{>%A`D*3O5s+L!dfCQ;x~Rpds7>vM54W~i zCC|QBigi*vJP1VzFv8QN_}gR5WMi~F=ItO7efvQQS0VQ0#W(_@erjqRo)tfZ)H8NMO2M{%zgn;LtTXbJN$uB5S zxg*Bv)QBLw$L!?gSxfou#$i)6E=o*pylLBj1k%1B1YZ|b(8C288$ae-%Ij=utmb*m zjM~HA4LLKIs&Kz@L^e7qYHltU`EV-BpW;=Lz+-gM*ak}|oFn@KHub?21!T52>k&Tm zfkOJ-QFuv!l19@$2?Lyy%!1{oPI^*>c8-TPw)grr4pe9TOUff8)HN#~j!fFsR1`Ya z-)GNsIL2)X_MaY31sU+>{lb-+FC;o%44a#c`Ih~?I_~{neLDAhCGL5W8a~wW&|s2B zrkCsWzpk>*eXPz?A@TxY!6b{MGb{If^bhoQo^+(iWEVg2l%}t474ZDHYwyMd0+779 zZ+3X$ysB5bO-_E-8>E?#&%{Qi?u@?A>(vcptn3vVxM}f$7i?T=GZzsI8GJ|sQ#c?f zy#LMj(fz4iHwyr3_td^ikHzm>w9jd3DsTdB3@@JNlrs~DnPf=jertwL#PhZ+@kHp( z$^^6QfWg5*Y{(B;`_i3WGh&~?(8G#)Q$1Tr0%X1`qjCDs?*4w+l-#-Vk9`P(ZhPfc zo0C3GT_WpDj$4^Njarj?+hg;|v~pZ&)uxTMiYQeHY~Y*{yC^2KZEZj}XnzbfOsM%56TBv?BX7~=GnH_I_b(@2q6glfJP`NiTKgiVII<{UsHh0+-rv@^mW&^Im1_v97o<9>VfIG8G z{hT1H)9<_C(|u6U6a7yIhx_bW#XxOmziC?8nPVHVSV>}dg1BcA+Zhw)Rqoiy3dBMK zrR{|;#QN1ASqt1BGpBMNO^rZQoz0IJlu?RqV6lC!Ec0aF_mZT!qB$;8V#>4dF@-Lv zfMXr~M^C+~;@1Z_4@3v>I)E28i^kOjqAk6Qrv55yyKO?nuR?s*Y}v&~37-h;whYop zKX-MpC6_6mtb(gSW%yDj;B==Dtjv_x?KuG@RGoUf-zUf6N`poMk^QBv``YKcM9AR| zg#zw$23me_LyAs+Flund(8D8s{IEZ(iYf!`%``_G0SB)dJDzpUglyLjR!*!I72ZC) zY!bep;6HuKp!$hLQBSS&rxVseh``yg!&z;6AJm3_vGA>BmqScY$!(Ohu8`sG%;AI( z;0)E$LQIHoh-ECA%VcugA9Iw-r*sW+E$^nk z>)?TYW1fyq;=;#QktaWR_+;TP7R}OR_B$9fv114`v~o%)X;VVTR@&6})>t{Va=-Nn zj&vDe7HRrEu;%t^Q{c|YpvFT*w0VH*ut|ungN`>7Nrnl>yPn#~M6SLbzlyyZ-us%G z-)9IRii-^uO}i}YKQU3CEER2E5;W@8f=bqEWbS+QPgOYusLF)P)oR;{3Yt|Xx zCy4~Liw8LkyPNz&MBC=QWZJ`f%iYl>0v*OqH+;4oJ!1ORdTMS#oTu1Mg*m+36W01* zR3z+Rdr{ta;Zj$H zO|&KOx;l-EMyP;|`DAW>5v@r#!dP*?iv&ZpSaxyC^tE_isnA1-oi%Do_~AhZ?>t}} z(mfy=1K(mAIF*$kD@d@*bZw)l!r_(@k7IyvU{PX4a%$8#5c1!+h2a^ogX=U^vWg`x zvInm`+UsuGe_&2XqSj35#3m*tel!)yQ>XD;^(u}CVfWjwzG?M?YiUeErnHtVl6-w< zZHiq#pCK28sQ*T3=Q05?iFzP78HSzy(CU7O2oW^fM;zHWwa(B#rFnuDL1V^TfKhxA(67P`mp4?sVQ@H>aIphEWsT zw2r@dV>TI;B4&|iX=SCdmmAGjNdnbnbQ!@p|K_~qw%X6P>>R!2Q0T+nLoLv!HcvRm z8Ii{S?0l|f)+BuGsawOE+Zfvx(Q}G6gubRqrgV+gX6sqT7ZbI7&D-IhTdkopys{0iZI#o)(@rC;C}Mv@i-dY3LzKexTa z9%rY7dLb8XNhcdCvZqM%ek^s&r;0zz=iz$bku&fcvxJK?gDg_fOQmD}2JVX=R*7UV zpNd+2*S~><|7zTwllV7>`>}DV%Uv!8JXr@{7)f+WXj7M)%BV7%)_0O~;Mz)FV|@or)8yX z*b7>AK@R48(2JLA46*!RkSHUUBPA|mqvCLxbb5lJ%ek?GbU!POp-5M%sB&~aU~&&0 zCpBJ%|9trg0Pp*s*lGM%(_jvV=w=5n<3(F@Gn0w!z9>DV7%j6MM94zFhOK^k;O<&| zM&X!)grlqYNUcNl{#F|!tgN9m9@pa2?MLOr$O1nWv313t1g9f_SL(eM7m%v61_YV#_scZ^cYir> zWJKfEAPEVU29k%Hv?K4E@)n?l;_Q1p6nB-TS_Ce%Gs|WnH^JkEx-FT z9!xFV4(sxHSu;qAZQ4+nqSPJk^O8En&h!)&QSy6r_7|V)nCjZd9#2p+Bq4&dR7a@5JGv*G|p?kVCq5da&!p! z@PdWhG0gsn5^s-bX!-KfJz^4t)yA;ZW=snE8JP^Aut3nGOE7b;eQRCdsg~nckTll* zqt@ruyf<9d^ZVIL7Ppbk{ZundKWLm=PCPXMKUtX0d(!2STDr&CwEtp%vGI_GG6OA1 z^9250T%c6NdGc+ud%%Ut)KXneA0tD<)IeNj$Z678FQ_id;iBA5B^9V{Z1GdXLKA*o zr~@2=&)#1-3_gEe+NWK>wDP-&Ctrj^?v}oIiF1PkqK#@w8mxQO#lTH^gmhbeYa`^$ z^N(KRqg&{$mI^z4J;Vf%EY5z7^X%o@X(?I9rA+5P5&#<#BV-9E1sOBLB?8bM7k{|F zATi%!wtt07v}lGny^L`PiwWorBfI>;W9K*LdN054wd5|dF>dTteA`g=1J`JzyoUTj z!*<77Y1~dU$hh9wBwKSTpIrpQPQ2%nCOCzip;eh$>C)*#7B@A#krCY8m<-u>E$Z=e z4GBo@G9*P1(xSHCLZ-{!ES#Ze8SF|IE=V`Z(ZSE#$mt^%#+ zM(b$0XrwmAcpb4b=|G%x!fe#FuWo%-IO(%C$*|}@#R`~0c$L!O^eNCo5I9=*xzAX7 zHr2fUBaN!s^%f_Bpbj31!Y;tQbRDYprm0XZ7PtVvO)dU;;YDNuv)qHa7=WxN1O^ydhEiQvR*p*MS3;M4KjD`6n z7il|WNkgij5dSL5jKj7){{wJHB68AKDTW0`o1(!l)`Z<2JBb#x1nUe<WsX@*l zw5WCoIDQP&aF`R<-j33*+WlwWxi>6$%m|CNEE4kFu z7zNDYBPoa5D81!2=Sv4{;+CIL!po{OnMGCNG?NZq_TV^wUEMs9y!k`?7Mi(y`OAwK zKP87n0%)h9y@0BOk-MfOr}C?bCmTCqg6^Z)jG>)fir332bkWS@wD+WT(prChYy#xe zI4!Ha1pClO@K+YV;W-(W@0Fwlqm7&9n=T>$YJS%{t7{PHB^{JSZuWY*I{Kr>-B9PC zAVany-E#`t?F!Ns?%2(onw&ehr*uv&w>;%`uEq06p|?VS#U(c*JwN;?`UQl|CECtV z&dE>Si~v4#7W?LKcRj%K>-v*EAh+s)M+fTwWOEQBObXepxeyGGC}^ z{sQVpIvuZ2a4fn;szT6gHydGh%JaerA~ye9;Eg_IqyDVujaz_O_?Ch1s&L0AGCEBF zCAMRCGmLU$eTsEtl4kxrJolVZm zPjcb=`}^+jtF@O84h%h4OMd&7ZBuqv#lH@!*jranihBSJw!GI$*NGrw+sWb==6*!odya zYOKD24#TEwAFq!5nAL7W*Mj^u&(I(f9&jN6oEZ3Rb9x}=;D|-Z4tqDYh zqhV;K{gb7fru|9n3tJ5(?PxK*^}5bZ2vpff+IA3oR_>XytuZ zr2FA>a+yg3>Amaf%<&zq-J3O7ri07K*i;}LgobJ+e%&LA6Hk-nt73LhPy>jP$k*Ed8A1 zL3ku>7mzdYGgot66fZ4jr8Y?UFxY`e<-?jXRCzJlM8V0o6W+(U4&FxjpEbt$aom=n)cBC$c9kr}m^W)#@a9()jrUW~?-r~0t@6MTfhRsyHW*Vkekb&_omPJ3 zC=Wk>SKt&PfwkUSD#*yk(I*Ur$7a{zf3%&$Pny+Ps(gnC$UWUijaX^k%qeMiND(Zp zo1Y*#SelHzUydLkCQ9dbighJ@!cqQ=FUjeq6rEa-3_sbbUA*h zNFnl`Dwq8lw~|I1PdS>G(B>U}VAelZO6&t++YAG_}RuJ^~y>zM_}tsBP@C=LlG`BZM^xdw|bXK6E=%$gco zH0S3Ev@~|*Ea!81c{MYsir0`}GxJ)J>kVFeMstaY{29NMupy`_0?|fEQVbarvknGw z4M&0C3tHz8BQ~Px)cyqwb;&xo#B9nj-@aORd&s+Ry}YR6%3=uc77)CBosG5icOVyO zg#tzUaDE5!`Vhx49_j@cwe*jkyK=zksea;?xu%Xk1_6Vs!r>b%DWS>7&u@iX>tf7S zOO-4e+)-9;$TMwUr~(N9+gNkGZn~ao(I4t-Jw9C4qIAV@C~Bc&f%ZJ~y|NODYytW{ zEg9ulq`+@dozib3+2XOyNo~$w{ z#GbHhkvxTg?CJkvGW=)q&oBSXSRi!#W6Lsrx9Ac3)MRvGRZ{Pwt1vROrv1^Al>^!zLo zQ&;gIfR#HSxqHQ#?7WSWfo<72a5I^Xq#L5Cso%vffTd*n1s7Xe9~X1Splb}}_!2jf#u5^H#349w(bZq9t+ z0z8X-xUitWAQ%_2^T6-!U2_uygOvTQIUNq4Mz=!W6yAtSBabq0pSP_INC zoz0@i5Kb{Mu~4s;F;d-oWj&kF^MQJy_w##uHn(n_STDz-=Ers>Yzk8qI0+u^IioC? z6cHP%*#+1p408Zpd~w>je+R~^aDZ6r*NvsjI~={69CE>NxS>NkT*&jA-PYO*ER?^f z9WUnOWuC;~Bxl;?3Acxywdn;SeP3U{wxW-Ww<;ZJ6}D@znS0fT`$8x=|L$2hm4}TP zNWam3i#ha~H6s0-{?PQnuG?%nDh=Z7WvUR=H#OdC{&jsO)a^P)&n41*K~hywy%lak zpl#o*4`0UNkKh6+zmFp6XWD;(!Joj3J6x`H`!)OCvzj>#5U;l@)0Co0xdwbj*e>%C zh+P-{ImD43CeCC%sM^j18N9$!Y0N@@hbB`?31uiYN|WvaWY+EN9WwZu3{U9u=Q~tw z_@2CIEGOq1vweN)V902GOU&iIN=F;x#%D^&Lo{O(7f@^!51oxp>=bJ#xn&EsAF9JF zY2|l>ria6y*luB6f_Lik@PzA28G}Zm^u%wYT_*x<;@S@lBb1te489~o?(C1*xv8d2 zhFAeZa2=|Z{awEeqVwP+MLL{;U#|o3K`a+=moyTw)9c;7Wz-FmHO zquwBe0#3Pv0vXKU1(s*Jk%z5XnlB{B0UuF|8(*_eOk{w+KdLlTGT{a~;xni(0nDt6 z?q#>5C}o8VeWgx~Z+44j7s+Q4HuWsK7av^)LuI^yqVMLAQF6UzMmzUHP|z3cqJu7K zA!&pMc5^O1i;%9CBHB@)2vtBS-sd6$u!K9x4*`WN2Th%;Ub-u#vd#BpqS5K5y$ z&{W>SR15l^^WFAa+MtIqY1tKxs4BZ%Q$(vABUlO-^a)e(Ht$ERnUVj*d4n?hL=nq& zxJl>-YEbRH$B!SAAi^1+5>v?$tj@Ip?aa>b*hy^c0%5I~Mb>k;*7dGX*n7uGM*njf ziF=o75v2R!n4}xlINM&;O@Xl$gNi#7TZHWWz8B~0CGCEEjr-uyjW8=&!^HH$r|Ta+ zKfORnqNk#$us#fe^cyuVBJ7)nECr)F2=n33;l#Mbzd**-;Lo1H@`@{PutT>GRaBoA zZb^zg7_{6RaMl&|&9*KTjZREYpl^^~=wdb9Fm9b`(FvXq>FJq!n$?4;pWN{7rD_?a zwBRhFS&PliJzPHYZ0CAAAKTeb#Opic0rs<_##49ICs8>`AsI~pJX;` zBP)q2Ibl*=(3%Min8=6EQcN8N13~V5)GBb=^gJ5e1=J`dQAM{*ocx8q>NV|iqk0Wy zHkyp4%GG$%IKa9Bw85(rA1+y zF|cwkleljq&w7!up(o$R<&uoQ=QyWZ_Q$j)8 zW*Hn_uG!kiEJ?pk2Reio6VHJJU3sI|NV(#B1>bC4cjRwdFp2TA4RuLJu%92z*~mionX^xjE4=fxzx^E@4+RSF8gnUnf0sIn-MybS(B%c50OIY z3R8YT?Qd8eGHw?cO*zHr6@9FBJR+fl++Y}2*eDCJ_iDM-RrXds1W5ILCt&Kd`nvDZ zw5-k*#J6a+3ZiR|3g*d&2cvo!(MspL+o%Q7bRX)?vIC`2fNBc0s@~e$dXV%SXQ$&w zG?+#{-Al!cFT+5K+rszsT!pO%f*95N# zw-D)h=~sPDJVUaWSYh;K-aX!XT`;){p$(aKu65dT zM-|`!CFs-UcX@X!ocvWpB&U-<3y`ytYJd88b3qa>wZ>$Q`JBhF0VMyp_(epxe8-cB zOSQ?Vdr@z!Hgk8>L{U>oL8FjY+hSF)euE*2_eIf_owbcS=g-sK3E%a96c({+cZx;C z$FWD8Z*lF?gvwoo=>a1k!1Nf!w7!*2|Cz9J5}Oc#18o0vnFUe+v+ycemDMS0Gj0eOqJkzqJ&3sz-TSr;>@3tNHlv#BuEm$8f`F>L4ygi97d z&%2LxcS5xp<1J{oWZa5^=$g2y#Sr19BmEFo^iNLspUIYmg08Huz^A@p>~89f`iViQpPAaBkW>fjHSfD{`w@R_FWH?a zu{=PT#NsoSbhck~O4KLHEtHUolroeiyYR5bGZjelXVU`?(%%3`HCI8uK%SlkVuG z-5N=i^6|~Ph}>F`hK4L5AMR`TSMuB(Js(Jf)`;P17bROQi{HuZd>Sdb7Z0a? z0n;5c7BNblh|=NoW#d;_xb;&nHBXdyPns8aXP#vj!dp2oP|)7*O+`BXwKQ)Xv^non z5f|9DX-k&quAe&5cUtc{&?u#U_2x0i=#w#xK3Z)6m_;SnzR+Kj{NIa-qYq7_$i&L& zG!>#!XQJ*tc-h9){p$w9q{~5t8SU*O;82Q#JEv!_tFXH3o=2Yq2rZP#HnF^x#`MJ2S8Th9m4DbvX4^%I|kLL>I@xE9bC3!=m{@BC> z6}aS5-q#n;Ehb=a{``uMA7<31yO;Aww7o*k0bP+gwDaBhjwIn?t#R;-Y;xEhaqv$(d7ATfN{Zq2HNpY=F zdi~3M!Gmx_v9HSVgD9xD)lmb9>pP2oy`*Ke zJV&|z<7zR>$)5wv&ZipC)&8D3MTC>WC8G5O?|VDpKq4<9X=TjNhr0nO4VDGh_nIf( zr6;YOnK^E)1Y~|cyF(iku(5@lYa0ZsScLCjW=^)kY|8rY-rf8Hu#S5+{A)H%`{}1l~GRY{J>Ty^_Sdw)gTibb+a92KUdVHw})%uatgsOT6=Q zOZ`sLJLCI-dw8P4Tc7d3=<(owf%A;WB6V!(xnlwJu#(xUy$mES2e~gVrif>lnpnMQ zIQZdxN$SbV>y5#wT=`rOJ2vs7j-D@+#jqfiq}TO>^anVQ0%)X|DR#Ox4YkRCk;!?= zZHBz(SDv%-ZnN&HvCD@{b5LPLRRtKz_a>16;yiSe4qH9nlf75>#>f)Wk}m!N9`0l6 z-l%PAP-Up9KJWRfOYm9q-B3AjBOmx+IoXLSgMwK%OF>9TZui+-u@PfKs?{%?P^ME8GJQPzxv(gX#*e?)$X5J;h@mMgX=#%QT?G%&1;72rGX6tV zKc2P4)!1uDXyPa{971}Ti*TV40C;-7Y+fjl$(f#e^&U3SGr&<~YQ{Uy(5r`iJ#y>D zKV_~$dIn^M(Uy)KnRy>gZbpE$x?7Gx-m7=q`R7;}Bp>NtRq&g8LT6eSJ94dftWGD> zf8)h_So^N65i)Dn&ckq_uIv z#`Qhp*ZG4tJg!Ok-5h@iY^UF?T&OeuLn9;fb9j-1{g+IOr%ykgURF>nzj?DYX7B|6 z5a_L#k+`yG(yd=o@nP_t9W?oPYooM8Pe`RhH8G)E`&LHVZ6HU1si+XC;!jFN<)ruS&s*>>du-%u8VC+hB+ z8ek(Z+lYq!cg7-4bvgO@I~TPyj!yzO>ql+UC@sVoFW1&uSBe(R z>&zHem_VWyGvmoha(?)oC6E5OCOyF?N}Nzzp?7T{CMx+sN1bmsh4dJY+MX4=$-Gy_ zIp4r08QT`&=|KM`=4)zTFhAIKWJ3R; z9oa_ID`rr`s7-7M>itlCHRYQSbr3OfC-OPmE4=u`$+^0EOp;H~mP4Ll0+bKckM5@A>`#pBQ%=<);v0MV1fA z5=5o2uyC$c&lKWS!S>5xfRL!G!^-_8PE28$6`XP!h|p-uf#F-1v#Ngl@b6f33!i88 z5lj-+iXgzt026{XnGKT2p#y$?iqo}tu9QAu7^@2bJlgs%Q%wT_C{hH}ME{tG?L=St z0zOf+=-8j+IBAQ1dX!k{5j--N9SUya(#rn!b--J-z22p#JFxw@5;_5}&;D+&oTA}-3NdmZ7 zLXKGZKI8*7Bl1_1M0P!;tP9BuOx`DHFLG>3@Bs?3Bez=It+w|=A;=j2o|?pI^^Jg0 z&gY!rA=r@v_j7Hi=2isaQaA09mJBMczjJ#=^8L0|CwzIE6A-+R}Z zoW@QbM?lD?$wf*Oc0E7eVK>E9qf57DfELen>&MBj z2Yl-!@fgVHCTtN!TWhY_o?EnVc(yFM=?3)H;K;Z;QW7Mne~NfpA27+RRrN)!MhA@- znNswh^WX@gF9TGvlwTT~^Vj_)81 zSG&eWhfrtIJCpvc(th;RKa9aP`_Ex$waVHxYx_52aut%v{4b_x>}{r{_NGB$&i8iB z2iGf2z2c=kiPv&=2 zwb98MJF_OM(Kvy2~4bb?{~xa*--o#gp{i$=Ia*=9hgmhoG33!O6GHw4<{6` z`PZq!HF+5Kz^hQHuoOntjbFzzouX6+FE4jlHVDe@9&RV4ub0l}X$-P5@02Om9c2Fq z84O{%_Y;LMRlPJGE6I(zDc$FM^@WbV6R8S``JWRcbA3~2yb6+%U*A)B*;tx_Q_FQf z8BTpZ{zuFo1l2I^f+j8=Zebw*`+Wn;G`+4*0ndwOjz>n`Y3l`uwH@ zTsF7UEMwZsxxn0zD!OE2MSTVb;>DXxGp1#2w`M^9Sd%I%6|l7i1JE0d_w?68DIo*xhYlRJtxbuI<;vx$ zv{F$cZmBiTZx|Pe0Nu~05s0)lr7-kssPNi{AX9vjCYwXoEGDE^8rZk~&`+ofXy!`e zX(S_~=#8%(t61mlVo)b;^tvI`P_`?SJ1qN)i(+Brley@yy{N%-2Np3Wgkjvy- zC4+WX(BO$RUcn}`$Im~Fadd~Ctc{F*9%cgzzIM9r-S|OmOl~FiIbr6p3tiRs zRVTNOmTN+-(r0%b6ey^Za^4}Q6Xu*TT%l~KSM%~(Mx*Q;o?nZGt`dIMdg{l$KBKHz zz(B$$Xd-e=O{+JZhRQEBt52s&{=<<_yq!nq-A##oZEt@uby`)qHQQK(m&&gOr4K?H z^|_z=BflNuw_e4qr>sLmQ>t%;D3PkId$oY{$QLMqX^zxcAjP0QcCTVQg zc&0a4eF>-OT^e8xXE}oipPu{Xo+;Bt7zy~aI0N-m#h5wnSNV}&KmA+1C10M!;n|armB?f* z%{T>2g~L;@-DQLk(U<$O94vSF2fZ_%?~N^oj8m5ZvpnTEUH2nK5T%~SzLA%**4wA% zdkWm+#srgk$KH|gSfg@k#M#dLj9Vx(Zn8q3h}N&7`@1Qq0%btjvC6{Hp1=BQ0hfJ4 zhb*rcRI>~y2L{Ya+)`fwh^v6gVji$PQc|PoVPQz4{Qowu0hF9Ycj9C%=~@T>6B}|%g@e-07Qwp_ggMB5l4GU@3wBk zB_Ds6#KieM>^HOwKWt1ucIPWi4HTqqJK3%G@EoexE-OiTdF2aU39PEBqLoDd2)Vs$ z>bCBeTY8U2i)_}VnUz)9%dpVU>XX60*a_b6O!RX5h-s?=Ubuy;Z8(?vkDSvY1b2 ztriJd>3Za+j!1TXwUn$|s--vARIps>TsbNt_YA$g%trFQR!Te|1{=s7`%-@OWo!TqI;Laae zRDEA|veFsUD3MdYs};4&62Bh21+e?H>`T7Ps2cUDUxJ19Cs3I0T}s^#L4){tcW3v@ z9lI55?Ff#v!^Qm8iGVpieC~hbB1-Z#(F!>g5QrEam;`Jfk=^4Bn0MzCxzE+VSw1Jb zxZ=ssGvgg9FFr*~r9|b8h>5g^<}BiVQPiFkptzMXV_?Hezxi6gvpZaOXXnJ!kEga( zJ&rewQ*xWmk}D(bNE(Q-V1nPb&fSgnWVgC;j^R5=LK4%1KAK_=%vbB3O))?{N9{;* zwHhxnwnv1K&u`iMz6yG(afI?pM^|H{xD6RwTVWruvD&iB>dS-ILwgL|b2YT~7e#Nj za+f|n4s%HFxh&r2j`NkUqUwaQ90pYinpg5g8-pAs%Jg)*#P_CU0`8G7lp$@KL8oe} zd~NyII9qYs|Bf0>umXmP|CTe?ab$E*+_=_if}tcBA=NYg z$NgbmC~EUQ`=LXg8_ki-5Nbj6;&)*Sv-_tc;8P9krio9B{ie7OLFMtuyoXB~zYwX$ zU&G231TK&kQmSq%l}^_wz75ptIPxLgxF9>~W_G;!j+1QSKW};@mzH?1@cd;Kql=!WpCD>@9OuK3eFt+nw z6R6S0nE{pQs&S~Qo*fO{Rr=e-iR%??H$_cDx2geUHct6VP^yPrx|55GEsp8w-`ME4 za_ESCyUkM^72!+q96vpbau)eDI6VHwG@P*VCC+^Lf5m_O_JQQdybKXU=X&4pg zn-Al*DbwgKSDbU8X}y)CmXzGDmV1QFS<`c>Dxk;v zi;~vCO|!7zBK~`TB6y0rwyWgcP**4`Psllt6U|rUSv`Dtl~Otv6lt9Z2z&6B#boo? zykm#s6_%z+^0}>_Mo*JMVXdv(^*ytl{{i}cW|AoACEQ?iR64h^(w(kk&7aMiX6<>} z$Lj8?wN>6%Z*<51MJ)O)C!KoPeGh#KOssW6Olb^K6gsNXVv8P$jIYcC$Vh z4LkyEJlm8byC@1_^k@FL~%B~zVfkqaJn@Q zeA@l`*1Ge{>4fR9r{NRm9hIS&zH#(dx|tdeU4wSXd0{2PFD3@KcbC$C zg%YH+eIEDqQ*vAVqC7S{PPw<%!;`pMdU8;}lqs(3(~5Q<8wdK}D4_HT9-p06VjY~i zT%hHIKvf>wquRFQAK68Sk_!wHlk_6L#57P#w?J05`s>rq&rtVxt z4{T3_xYJ}`ANDebLIPNmADE=a3rP%GH9&cg6#;C;#;$r=0R*!Hv&^(BGDyL>E3WOG zlCCv@fX&d`Kie_{Y2h+z&2s22D{n&uwM-20?)8EA-Qz<5=kJ-3!+u7>BWYwQE3BvH z_0+7|ohGO<5R#J0FPz)b6SKI$uo1HgcpxgN%NB(8KxJzeo^IBv0AhP&kok$7q2a;l zSC}RAyxK1EG=nDj%xDgS7DaH+3AJ+7_FtswWb9Wx<5qsW#}J?K40zGDd)FiyWttqE zo~8v}9+pSB(#~<)Y;{L6n{TQAhecw|<7(Rzn^_z1Ug6rVA8`Ws%ST9G z``t&`gnpgqHeu#F+)DBdd(_FO{C0S&yOEjWA9)?Q`RVRY7ngSd5g!T8KSOL!04meP zb9A}d!b3^E#>NT(@6f+kOyJ_-kwfO+L$^j&4zmqcZoT9s;y22BLBP$qmwZ=3{EaHU zu zk9JNaB?8f>ulN7PaV)=|g{%quV>cQg?G{Q)wXG<+P3UI)x?>afWbL8um!c?N5$VqLG(zQMK)1jpac{N6YF}UZrJ^ z?Z->J^40A)4FR}2nrJ{Clm=z3ABGACr|hh_-sspcLQ|o<;>TOvtNjS_a$nzd4y&51 z;~hGc)eNA);dxXLoiCj4u=k8p38au@YUbJ)q3>TQ@zkna; z5`DvrjIt!^5SJc;rwt|JaVOr&a0@j_FtSz*y6i>~~I*jMOY<8n0gXh6qixN90q%0i5oA8d){*>sIr>GXM3b;r>_;7eRGy{k0=DQO1> zS%~r9tG**15shobHE!ds$D$CGi$ussf7O~}(kqkeaQx%M;NGEcCm{SuTDWaKLUgI` zyQO}4upS!XNj&lzTEgNve{z6tnXvDal#=({=VrYp8$-`@_sjHmv=caetm+{KFb{*{ zL1^#j=+P25#2NhIMmsr-XXDhmHF(v^gLP?qmrDF`2j$PEUE1Uh_?1$4`O~jC*oinS zD!k6q1YWPNp9l(@I?vZUEOt1`E8fDA>8q;RsQA&F(1R$Jcgu9PX_UBTxp%!kHBDl^ zzk$`z_I6IP(d73=2Izkx9KCt0h0qLLh()qa;-iVouUs0?R3hQSjp3v<$Tso`JTeG+ z3P)-wASVw($gBU2HW{jICVR+^ic1eAY$wadNJ5P(di+%AvS(LeU4+rSQBir|>eFw2 z4oea{u?tGMVvTfoe{xmmLuYj6M>lQ1LhFUyhG2fW+VgY8v|?2b-(724^rCNsraPJ< z5A{ft8d^pxQQz97Co;yH7x-Kb^X)6)l7OFG-w}UF^fYv!lQTt{nPger<;mmEjB>`G z@Bah8Blmyx{bvT$CUFQiaX%z~O_iUE?r?LjydPBbyPN$EMtsj>wz%R+wNAIcN(&w>sP8QQLgFA|syW0ncH$D~O~9f>2_n2EL*X9RXWVH7HA zG?=BM#}D9HagfpC5pwjd-t#3qQQz{Fh-|){Y_t?BRoH0;b-i9 z*JsiQtpz2d<|`;ytTW+1x|2`DM?t|40zJO3!DH*B9h$weI{`2ueB_XC?wPPlFU6Y3eE2uw?_z6DVvg~Jtw#g(zW`9dD_`g@_ zkyRCMsPYaT9z+qZ`^CQ={2=-N@eh4@V(M<~Xycp}=3hQIm5a-6KINR{xGjRX_o8f` zM~J7ZzS9TI{bz{T`+h>hzlL~g#1)_Yqc0DN^MCb+AS%KhVut=4qY|!(COykGKvB$& zmm}jN>%ABrM<8@N(3r>+$M?lVKfOYW=Cy!N5V}{+|Ep)*U!08C4oAsDd6FA%UZcI< z^o830zcigfpZ&1UDkqtBJ}r5}p6AW~D6aqQ1^NoauUSXy!8u)a_hbBjnvh^KAB@z3c@9n=tF}oZLwU#-OaqhJ z$d`YYo&vTLN!X1rKF;PHUey1e)TYV5;-|h(9FcXi$vJ2Kj?n7J!Ym1YucmuJA0~*i zN~*^yQul>gmI3ib9{BM8ZybEzD`H2q()Cb;x0eEh1A?|zC`yc^p zG!e`6l)Ts>apDVC4SQ;M&Ie0{9Z2`(VnO=C?|q7Qr1 zm<0Gw{w>8Jnqqvgf39q%eNz)*rT@=bMm?zsT|uJeCh^UwF9{7c?cEFN`K z&setO5=KE4B>arCBSL>4`{=O(LY{`*%#6I4$N0W>oMq&S#upN!PTB$9&# zqsw^u-~~`>ZV-n@s+6p)@p|e%oe&Z(Mtm5eG&*BUjcw4J%8hdp z#SpvxE{nby>{GPl`hpp1!N9X=Op1SN3MI@F#Qv0LYiR7dvzaHv{9{5b6!(VNoJY*~ zb$)t8Y89dhyM~fhpYh2sGx9^4+9SEJG)oT)6B~#LH~p`W3*;{ZZ=QtvKbYqSO4y%q z`b>E3X3l|Nop08Q1ixjWkAaFZ=q}$cr37$vl~-wp@VIP|Y1C#%yBU+gE@4w*Y!P>|dZ{_7T+&4`IJ4n^Rl<#hUI zxFk^1Y-eLQvbd+s;oSx2JxCZ??9)Lz@#a-OS#{}Ymi$ptkwp=@9(({48nNT;KzDr` zv44$x+KNqkkVo(b9z`SynwnCka66{GyV zl!=4Fk3V{a*!E|+P>Dq12W~ebr=`IADt5mXa(7~%w768lPI{4x>!PM%1Y$@t3+ zXJsTeY;0E_@2{GEmezlvM*IYwG}7HwA%fdvSiJcY>oZxI+&GzY@xaKZFqu)!t8`QO z)zuOyNogevMVh5Co#AY$c$34^vh}BgKOEG7Hf&);5ep5N-3T^YtPJtrPA2Vfxs^xNu zmVC|3P_LGWqPp9+sq-*qpR*k%3v(N_z_{W>aYd#$2z_?Ph|_YjSS%qbrYvY|{v3cR zkjcv8LUJ~JGRtIPX{FTRb=^S4MBbYhlEf|dXIdY{f-wjeIHpxH`B7?Y5VAg*zIAY` zGxrt!Zj6z!jTSbh4T@r!TADUMPFEw@s61chT$0UIt8nj!L&L8jO^_yP@#R) zjMVROLOvpn(GLBo{A{2-)($Gk1GD$+q$)h&rwgP5!_&fjMSTUWgfm3O9;S@fCwrA% zhGWXg>RTs>X2K3sO5GW)e!>TRy`T5Ih~+tdqZbZtzmptskK6pYD1^_)gPM8@2h_bi z>5la-gVu9bacCKvhx(VRBm-|1TcXjZhoxLyoir7is=KPFYGRik^aCktv=qhS$bYyH~ z#Be(_f+H&+)y}1(xCFe)K0OsW5{O^Bx-!a5k+5@j*XT4Y(?kUx;`6K*FtFk@e(w1+ zvI=rekL!E^o78S0{1YFE5p7c{Il2 zVl{thZEUQC5#0vy@2A~LDN^NYsy6t4AYlxKb4n?)9#+?o`ZZPrhe`0*6>j+*85A#! zH1a0F1@0h#L5>XX#B7QylZ(x*KK>3E+J-QA)z+ESRSmXd7Vo_9&o9mnq-{E~Qf0&)%TUgrW_Djt_WVOqa}r@S%B)bRK(a_aLh?oJCh?!uU1*RX0Kq z)7-}Kgwu^IqL}DYrGrmv5rqt6hHX2~y0do|wTo*3rbgm5B}#E^jX)_#0NxyE#$3YVG z8+p9qgOzUpPEgaIu1L9W0#9Cb$R!#%&ll3rWV+s*no=t$sA20&*CV$c4j%3!PbGX= z8mTF>t+yWEz>~wQB_-tG)>F_PTJW$DMKQm_mUh;N6%|!O^*mJ?csjNEuf45i=rGVQ zvW7&}o7~4%> zjyaNk{Ef@}@V-0>eY|@aQ|q~n3hm3>5HoC@#?0J2yI=!1 zmh7B_v{kAed?v&D{@wGLm;t=hhQsh6&gi!*Ws=}J3nLbKtJa(VU#nqK&r&Y*=xc(WRE0L|O%OG;b__!G*d$@B& zwJYPUw8g^B#pOBm?16?G+M?)=RFLO+I6-~PVXaj?o*kbn0OQ~;ooWt~`j(s$RT*FD z^Mu^U*t~v&kN347)|{_^_gSDo-u1_&F0=JEc2{}t)sq|wYA)}@#PfR37fC{JV&aOu zKNFVQ3~{?qP(5!J+Nd?~M=ASLi#m~F&(EFCw{SuvWiyQEd338olqh}G00*VLABCR< z2df%ieH6j}UNvRAwA&EHV(K=ri2H1QM4(-fo0h(xK0<=0LEB`aBUIV*oVk*NLs8$Y zX2a6TXJ%DV`-QnmvH(t#I(PvI^i(^y9R@-A-(AN#Y?zsooNtjtN4-fEKqqus zw@Bk`7sbWZNnL>ExGFKh*#PPDdENR~b#seAuk5 z&-#jdhq1)N1(|%oplet}+naumvz>^k8N9vY{q*|LSz!r<6xGwL4%HMufLSMBZ)}=gn+ld~)GS0tl0wr)l71DxWhhPyXdwMRTf@h21Z9gb+!ci?5!Y zxeVIhs0rbn9p+lk>s=i@aK!@f=1kGlPC^nbRK!+41t)_YfNx4_f(}AM-tdyeFjLBY zlazPGlJlxlGM7zKwY$vpP4pzje8VT`yT?(NQ3p6Q0M2VJCpoc3 zmus#)UR`==chxT=zcQd-dww4;;`zBt?8fbpl*Z?6kg40*u#`&ntJg+{>X{K!^lzwg z9NpbxQ1E0}C`|J`wKeOly&@pmKJ=J6^x0-jfgGV-I4yoBr*22bz1J6TS2!#^pqF)9H=$>?ITEJ9%V%ztMJ011Rh*HYXp6cxVMI5gTxp=H`p09Au2Qdm!?-UmVLISq zhet*-pBn)yJ8$)|TzUOCW`*)=pz6qL{ zT0(E@=^C3dc5H0)R_xwL21ZxjB^(-}QBZg{UiK#hu<;?rXERvM4A@r6gGIxQhCp=4`mMniZa z-qagz_AdsDt;4lA#~@nkZ(!-(O6GE!?8>Xz;!e@WUh=`y5_$rrE`4&=%J$j1Y?Kfj z0!w#!4Yes*W3O;s-6hHPodXA)pV?Q^1S5`I3)m>jGVd3JMlZBuW2-)?L2Il4L}hLK z5+SvfmH0bJMfo1)@UH0YNNM!YFw~@a3gfw#xWYW<7pmag#`x=DTCEt?OBs2U{gg0 zvkE00RB}ub(=P&Ja?t#5u=fv=Fk>Z`VtnFABw3Q9`{smRi}`dp)baeyfIhPy&CcVd zn;1Sv?(7`0iILoGd=Qruwz@K<_gG*F0GUp!;Vbks+fh%>6w+~6PM0>po~Tpt2&r|z zfr!qZia8Yc47AP7X*%t}vSfB^H+*_mZi+`xmb2}7Zp?kXfVHs#6gpa1*%oo!Z3;D+ z^9kfns8iRj+)fZEs2GTUC7rY(?ApaED*g)(u{{R#CskKF z)$-n<(o)$xgucmHzB{5T*)pSGrBBP@t%owaJ^IMLK1IiIfle3PZR(O08;)tFrOrA0 z2`?KCkS+3le6f$tk;_)nsv1Zi(<`rkx(t0$9C4)Bd-3-kIu{oMU!1FlK6>sjVQ8=H z%qi6CUC}s~{g5Ma-~(TtbUoSATVXM1@O|&9vTs+>7|6(b%p-Nw zn-eDOM2|zcuP(~KE8w*O+jzE`B>&OnW|0&X-FM*z-mIamZP&Xie5hCiUZq~|Vp&8P zG&iMivjXnL7a~7EPH6T87vZhN8G|axHtT~RzSmcdWH2t?S%+t9mT4EPjR!jvD!KTc zi&Me0+TkOnEHm^=rDIAAyx`Bzv0elCGlLF@-bVrvro)6Mr|W#vVr%2Nra(zM+VycZ zTD8(L?{=-PYR%ImoQ%limhAg{57qN)e!A$%!mcZLUD}hQ-NE$6pK1QO`g@Zmz89@e zkcVWk@+{L!#=v6=!DGI+)!0R8X>khuOIp12XogQ*Gv*zFD}l^Z&5CY5N5fPe=1Wt< z&FG>8B0CA*C$pZhk92%xeN*K6y0fmm!fk!_nr65@AGlqFB6~%_t*u{Vj9Rno7)<_f zS!b@hjMH*QYkpP#gTQnI>S!KvBg@>HSDR02_FS?5i* zj48wS81hY%ch1I7wkKMnk1x_3Q#j*gG$zo^XBVZ0Cnn^QbOoAEKWy#hx95ugDmb3t zP57x=-89Q3Q=XUJ6mY`a)?_o)*icHln8wDClGB&?Z8LIHPAyGE%@lci#y(Nnoy2d{ zx2TF~y1JrJRI|b>grI(-SA5eniJdJN?!w0_96mnymXF#N~$qEwoCC z7~l65T}_Wvs7}@TCmmk*RQrNB$ln#^l{M_gvlcmO9y4jDGXMTk^tdF!om2>>%gC(f z+K==#?F|&alP3Sb6w!|24O`=}%_5%DS0Vk`+HxWU)-$Dn_;76pZ!O1WujN~9ii%E( zUrshc9tA^B{&H(rRRsoYCa!6N)RgpO&H>NXw`!luXE^r!>ZXr{f2q;FsfknX+BD&G z)@|R`G{RPPc(UVdq4Knz*?*kNZtYFIYD#rTe@D8u=&oi^`SnSP=R|4CSp*-dZyL8_ zzmyaGO28$*zaxH-r=kyV`JY6aUlRsZiaoNwR)=SO2sQM1tzk)nhJI!v zeguKX5*v4m@EeK2+GX^yHK>Ob&Q*V8gT=Q+TN@#^=mGs2b8xo_pqoKMo9B#?q7_~} zFe-54M|8lc(8=8mvl316bsak;RA0Sa7j5xF<2nHNwts`VC5&h>{s#Jjke_dz=sReh z(+^GwTvCn1#Eg$n$g_+IX8Zmv^9#Sv zF~Tqcfs7L=oPu{JY<#BwHcEI&hcw`wz=_QO7MEVBl3G>UD^G83j9(8 zTZJ31=u1H1^4wMX1e5#IY51X$739$*tb?`M91ho#-Ny~LL=3-5U1#m7O=I4+Eb+UN zCQ+Jt-K-1b0jxZ6Q!7@?XS)BC>lqpS+SR8*s1L4hA!EA6uI8SlVZ$V+_TKI!O$|SH z)?ZHJ4#x4-CyRnKYgrne^_25c4GfCu5CAUOC=~|EOBxFCt1FqqcNbsYz^AfO&>XC1 z1|p%{vZg0wDJrip6Rw#|dit^vLtolucEWA2S_LarLU6H*2U zmip*+f5F$g;)E-9s9ZV|0&NTx>okJaKLlMKLv@ZOV@+*#O0-ri0IGpp^wVkH=j0a5 z8s<=15q^(_K8_ax0_HO9mqWnxPF9DcZU(tn4&bdZ-y4?LV}+w%L&v*| z@wc6qCwKP{KACGg`?$ci6H?DP3#|9XyAJ4b4!6Hpgu6gjXdMuRFh50asr|I)cHHP{ zjt+39JRs;?F7CBbZ~EeM*tlEp-;#U;V*(pa)SLaB>u!ua3>n+yUG#%Z1g6O$fLP^U zv&afuq5b5P;2_%fiWkL2xrN{#$<*_(_M58Cr=k6oybuM9GLSRN_QhUSG*n{RR0SF6 z5(kxA^1GF6autD~4KYYmenh}nmxhsEw>O@}(aF(}I93Nh*C5mX<|g#5?~9VJl&?u{ zu2{$`Y;#$7imdHJ=W16zL2nk0t@9=mU4p?}qiC0|Z~LgFlk_umh{|B$o6|XdVQ03J z)9Qzf8oUjZaQ@-rnqt;Z|IaT2i?^v_K#Ey(b~|$JC%#L1*6hzP;!qpj+^Z z)?jQ-h}`!!bmD!Me{&uwF>Jc#0IU7%E_1^ov2%*3Db=o;aU9Vr;gG(boNvv7 zhC8e28BVRsI`O16Rn0k9M=1a_I>nRmGi7K*u>;W`W*E*4^TOAohN2&mzcBe!u*E|o;#P?-aFqNA{ViDjNz>ruy6H~)B^u}~weGwth- zhj5N2O3a~<<=b{>X(Y5JDzUPYi|@+FMsWYTBH(t_Y>w5#1Fp-^(IPD$a2}R~nUtM*_laR%wbGpD<$UzY<-M z{F7A(CU31n!%3 zBWb)YGqhZei6l``=A)mHFjC_s7sXTBTG0_v;UW83%{W^h@;W9`cnY_lN{E$`o&c!W zNkxJO7`A3a>TeIQH{K6|Jb@ROGq-I-2j@_-exx~`1Q3tFWr7MPoip9-aPAv2ia6<7 zTZy>|Ns{5XrY&PgZ~g6c{iuiND?N{$!bxmbn*~ycx|(ih>lvML^T%&CSWebg=L~C2 zL|$pa8QW?FkssUoBsR4%EB{!}>(VQIDK%hVeR zal4|2ROxAJUlOfKX_b%Ke)|@LfR=tLY8;nn*ijE|u~FglJldwp6V z|8OguAO<}4Q$@E}qc?T@z|Y`2=BOTkL{qc(7@Cer5ERkHc6ZwFi9CE83OOy(# zVwU6SD%0!68R5O&lEeZ#NE8Ha$|$Id+gh9}jcJrw}8GIYa$$}2vsPMVl61RxG;&X9ATG27dV zkH_h!XoQ|cnl(Wrt-|YP6XnuYbgi?blnU?c^ilYZO6!%mx}X)f8Q~sYagU=Gg?Ger zr`%toMe31>GF}o2NOtd2@lp&4YXcoG)xJmX&R*Kq`>)ET_2;~4B!gL#o0>*{VR_tB zU(uiqi^?{l)|9R=*{A0@` zX7TIxqD^x`p2pLYlYz%gw;71>xBV;2P4Pz~OrqM}%STdmr&UvCs-wzi?~gjIwPG1B z_JYyLU~b7woHh?uVw#s){ePMr784ZUhVA3tc;Sn*Uj-^G|2RB-cVmSW%`jIl(6ZD4 zJj~+6@yFbMwO+EN%20gdkA&(5tV-wtT^lm4cU`Q|yh-j_B4PF28S%|?nLKYAr@zte zEB42pbC1S4oP%>fl1iz6NwjP!(Qf7)(xytN<(oWVkq)6B^Bd*~RAaYT+E2W)F&p#K z28Nn~jdO2;3(OGb(yjtu_mD_>M%|&efUUy!ytEG1{+k^ z^{gN8?jj2|MDE8IAaiU?{gF@r7O14*Cb@rh{wk_F!Voqxto0*Cs5+Y z(%H(QMHE=nQ9z&z<_gDgs5=yH8%>8c`F1{qba=!7M$_t`{vqy~ zYa(5_#T%{E#Z`~ve6KClSVrB7w7@Z3Ust-?$*`81@)Ang^10QKi`@^-z+>N=f(5NQ z4<9R9Dn^$|ZP))AwFPokinN&zr4{vdX_X0$Uwb1hCvTuuBpj=(qh$Gp7tjNL@QBv3 zY(-p6Qt}7dH{ivpI5`ZUm85v=6C`|EjZRN^Vc~rNQrKnZ>BdB3Q|t`LD=$2yHbY~N zA~MPJ(%0pr5GR>1e`qD5|4RiTnau?=s$e_B9WH=1@UZ5Y1iY7;yQ71A6lY54Bfc0S zX6n}-k>Alyx_He&O=KtkG~8Wb77!7gHphrEt+&V~TIM#fqc$c_G_}2TqhDOmsGgRd(AFjS zLOE0ktvA`z0#JlbH+(NSLS(CCQ$vjWJP;!&Eg1AvyYMb3JMsx`>^(MBt)t_3&xk`I zbaRI!-_q(FFk^pYI=7S>N?ZBbjM>C?X*TO;exVF!JK_<#C?8lcyuBz-Knw37kHWbNsImT|1?77a!Ir}CB-5C(3d+k$;rtext%$=qt zo8XvxD7Uq+nkmrnvm?}eqAfx_8X6wuNF=e^DhwI!O(N*@G-=2YOpG+*FnfOwLH&PKjHO+KC24RZ4OkFewJC5=Pr%thA zf;RDM`kRX->cIsGuKMG8ygR-zGIMS2>s^?H#y|*y`pH3MVffu^#K5vAw zLs2^;X?Fo%z)`%>NPvz!K&-*hyl0DxR&+m#4D*Kf{Cl+8$biJ7680r(3JS_!K;L6a z`O~r!A^{6$kUZq_QCj0E|0Z8b>~pNBl4_3n2w+yhY}--{yn9(R!Y2aK0dN}O?IIxN zjb=Fcp<7oV$Yyz0Ab9AeG;(r*h!U(N%3bEm;Hy`xWrDibd(&mB0VX++pRBSi*{UPJ zrq{`kh6YI3!)Kpwuhuf})-I&0t9{_~b{DWg7#lP^HOV8XF1^L;B-jXwK%!9{vgUm{ zEc|IGjE=4%5N?cFe5MDTAnX%_q*%Aw2nSFd6BZht??ioESg8^g~N^w$n?0qnlM+G zT_P|4d*(^+6M?0MjmU|rXX0S z*k5J3SkzANTir{GGs0ORtZ$(UpW{E8wH`J~oY$>*rl5h`%D(BU^Cxuu#5+5T?tw#_ zn#k7m)h6e@>ZYM~kz|_lJ_rCt{>+~vEIJ*g`}Q~ktQ+l*eYUBZ zmQpPi6NRjp;C_n7;pfFGO+nR_@r`b1#&Qa(^*%ncte!#?Y#fzJ_B)uk;LAmlVnpzv zMvPA^0g#+T-Kx=Pgmv>w|6rJmJZ82ldc zh=Khnoojz4AZ@O=S7&=YlLlZJNX82-m%wyTiocU{&O{3fZPqcM6^QFgOcgd&z|_*| z`A&^XMKyBZrC_{q`?`3|lH;f9qMOt+BxL0&qNWZ*nU`F%&UczYJ8)U3F)3JF`jN?QU$nwj~<%X8ewJgRipZ=G`%gv)r{p-Y{{jP4N^1 zTlt-0ptQ965G*DlDY?hEE zCg=5E%`Rr$4yZ{9y=i|V+Qx?e{(R1b(?XcwAGiUHa{7XTgsMSqgKvJCjGMbJ?VQzA z{V%@WI;@SZ`xGIJ%FGiT1;XYIAu3Wmd47l@^o!y9RwOrZ!f&f22>%X!18$3npMWlA1R z6`kd@Ky%oeZ9-=3>!%VwT5=RH^Fs~X((?_KV{t98RZ21)ndJSKfJ^SE0@uOd-OCHB zSokh@w_n2lyN<-RHjkS*H%C7W`}FqoSr7O<#AoGdpy|w`qM;}54Btj2<<|u0WpY&0} ze~4UCPbL%7+>M7mD(36yiXaJ5qAnG!nN55ZjnZ10*j@2YQ{sW)vbX{kQJo4=YyKtZ zn0-Rm>uDodKG&22Cmra&$5KPDGUW?9n8y+h2^YUyHCy%u*UI7iT&DUS8fkRlp67a~ zdG;!4J@4gO((QBoIeW!riDH(t?Xa5t;!MYaEI*+-QJ33^%mj_Z525Eb&=jV>BS+4anlw~N ziD;E0s!?!QsaC=q?dgru%S^`KJTcE|#Eu47Kf1uj<(sywz7j1KW>6|%EtcZ}*0q1W ztoW#a93MEQgBeAL+=VJDTRge+iy>g=c(>;%Cco5aA0-xG5x_{sqA?#Klv|+!PL34Z zQm6{Y?6_%)4VcCnHk#G5L?Q4}LZy8Fw8^!2P6-xGsi_xj&qoEHM`M(BQie%IBsBMweCXFG5e~ zTzJctg0@@cm*}oTda*F+{2tyrJR3Mu@nH6N34pa6do9~Kh_0HD*`Jn+m2JNUaC1E7*gd zuSZna-MoQychnmc&(Qu$s&7?# z8KgQD8h|SKj4;8FmSs?4wjw26M(a(DCx_bc(vj=STeA0-$UVpH-*{%Ly^i&qBg|NI zvxBieSr9#D*uJTpuu3@+3F9^*lz55S*?Fj)`5jQ(-YyzmsWonCX~TukCMs6E3kzRa zA-gY~n+jc#zs=k?k`F%=`2neX;v@9?O`&>rm|xFnZjj3Jd*qBpxSv{vXM~`hV3Vj{ zXKzt0#1~;k#`)doK-(Q$A1^x|v|we9~7ErwRtw7$JTS zjK=esc7XzQ@KDV=aIX%)v6oIg9@)3Ej}YDB}}Z?BI1Whw5{0;+&Y!A z!mP-GG*VA5Yq-<)=Fe7pGtb7PSQOtB)6rE@o%;E5d7q2j8}7T5m6oWJtOrzDKhLuLbD!L&Z|La-sB=urVg;1C5r z`v6_iM|YNnF&{R+$S$dF?#(EFM1``W7E+x#w`e*ThLWirD+Sn1es- z;jsMd9uq^h|0d+HM}-rom-`t4R0nAE51n*T&}X@XmHKHeI*;^6q-yUp6z#XJe0gSn z26Teffoh!XE~*yeKS&guiE3@*gsXRO z!iP3>?j*C^o8W-0{+{=$|d@byxpG;mTs{WWnt;yfwtj0ic&9Zy5AeTz8DDFF! zNe`Q^(n$%iG}7Z4$;BSzGdWPi*gmt`6lb;}7haL1Ew^hUkPS`z>`st~*8k4Wye>?Q zRRACT)o%7*5V+hTMcb z@l6U4_Vi@Tp`tnS1AW?C=OgKMR2WOPHlrn7B^Nd5%?i+e9BZC~T$U>0b?5nt9_JpK zoRAF^d&)hw%kmt3At7^r6lrqIGOl>{1Gk1IUKRTNY z>#N#cM2JF9jk=vH%b`z zXb9wtG*QzNu|+Y$uFG;hdEbn>RVVRWj%!hdlL`w<2A{`&Opku`l<~k~2d|!8UE?jv za0$+9dogjSb`ZWBnY|;3aGe8F#%h&#h4I^T1P_F37lwuLnLk(#?kI>rU_Zi%b;36# zKy+)Q#cjgaB2GHBS4F9M2dj?a0`q+ zn9WJWkuAkgq%kY2cD~~kgn-a>EJbgIDPmfZJQrD8bzgmLe2nq<#p@Zf@*|GgnWN_G zQ&~4j@~dF+C#A<}%7R|nHMXxlQ=Lt1Pkyn=beBQpFeK|1MR3a`My;Y2|6xPLuYnYJ z6Bsldenf|yb-kn6IP*E7NpqQ-U1}H(zwUO{Es-@apOYJv`^V$cZz5+r`+f`Zb!PRO z*!|+LJdxk*e(&6G^d9JcD%Ox@K7sU~cFTnh#OvEDh+;{-;S=_&(6*fRCBY-1V=Jq} z?4m|wqllwff+s>L`HeEuaW0-z%fy{$+B$}-wI6JBLQM{GiJ|AsRav2EM?8w14Yj?A zaI3%v!2^WhB7_eN8mqrfq}9e3%stTd4&Rfvs`0r$tO+movuKRY?mi$YYFw<~zx?_& z;5Uc2Bn|dej>vQvy7`4~)b_PT{bS3$T}{xRmW)3nC0kyEv%mLMrEO_K1 zfSr~}2K5Y`9p~GA|EuZRpdvnA-dTaZ8($+2@L|`FD27I`*SqcE!;S=RD+PI~ zI-jIO)p2u%^@S^TE#45|DN+pi?1CQ!XZBwA@0LBS+8@g|ursD|t5O0a-$=yx9AC8z zqZ8T4i4evjzcn}@>HpkrwdwkMvHNpB1$hzHE{hkV<%2?8v=`+&x}atQnXDvIHgmP| z&gUIBP(Qp>!sNnwkK3xOYMKR{r!M@FEJRtdbuhikU$X2GH%N2Se7V!SAKAzIT%3W* z3_jiNSTg+TBYQ=KB^d=VyXvP@i##$sSvy8}eFTer-^ZqEP_zUj)a72G*|}U;Th|76 zbUyVeAMLtYN`08um(eLOanL3hO!^Q_lbfwXq@U*S`fi%V=g;*uv@hlxQ0dosACRyN zLC-wSNcBC!&Wk<}#0Ce4y_c=c&4F^?Y4fE18@`hwzv^|AqY3xD zL|#F2gWuZ6#1IS)<2<7y8v)g}q>1H#7?1YEETr@~lX>(kYJ-N82vD@i`7`_epvQhR=?$q@6+7b}jJs0smYZVVx=3Os3It}VmuBpx#08Q{y8VE+ zBZYruEaHQA67gm?xNhmO3UtD|jOdT?)Zn6Ne8_OhR%9sbMG~R%#uIdrZ1RGtnCHh> z&d!(2Kj@=;;5`x|$n_>}ix@7bpk+h4JP9u^KW%uX0|cvU=o@DFEeOK$NaSR zq0Fvt>Nh4RU-OBg1W&ix>&bwC?;OPQ z*Jw8tRUQK!!Ec6aSJ$ZCCdJtf>Zs^ibKB-KGL~-zk8ak{M4U*zMO_-hV=T9lsP9F5 z#+3O9Nk}XGx&ygrhd*(4;r`1dg(GGzIG?0R#T3VIpQ zSkHRLx1U&vJ7irrS2l{2;rk=;eGT}X;Q2b`_k5s!H#pK*Y2fyXx~r=KoXizG19(NT zM?jD~K<@9$DfxOm)}qao2*|YFmgC8RDYBd!bii;(^PFl2-J{xTuZ^c@#}gqk@^Mp6 zzk-2-wqJ$R>2z@7e5+DOkH>T}7 zs#s&<`Lf_|2+s=p3aXQb0duM&uzqG843$^>x;*GVYJGp?FZE(H{&ZcMsN7@D?K5d; zjl}zs$izZPdZyCc{CJFzFf?*f{V$O1U&EAcEtT1lCb9f*&ZC}-cv)FZi7_@Y-}N>S z_D~N7D630(qJFwi)8CRUu&KE`$q{ zUvv1j@_|T^1T60DEx5;qW#1eYnfclJj_j(ts#iv4OstQ~ z?HcAdglI1=w7ybB3ENy=UUlNra8A5L{2xu?2Jy)8uZ@kLWb+)yb0j!Dwr`{*Te{%y zlEH-{jLZAoRK>G->(1!s<&nChCqp9~6;f7_G%t%#lewMae7YIgj=ip*YA6|eHNsK2 zh4G@!{Az0Ka(O1f3Ifn+lrAAfl4M;8E=U3Zahww6Ok5;X;yqqRqN#IJ-a+8WUAFBS z0B+%YXlpT&q@5TdbTJ}KA*}{w8sQG1Cn>f#^@Pnk4VwOuw`riF=wQtAa$suPgp`BHOwmU^0%JuF1G! zQIIU(gIYA7bI3eX)fhV6%Z77`M%GGSB;I zY`sg`5(KbJUE}o~AgS-MhCByd(=XTzIgIK(`-?GdG53euGJqwU=430M1CreQj-#ZfgnOwE6lw)GhFN8|6dN z>4;TXw5JP}NR;x%y3k4dR55rhqD+E4w2_$dxb`w6nka(}Z zp1Tfx%rQ#h)Q*&@BFx7PTy|HPuZC#g;fxAFt_0>T57?Ysw$ZE_XCk4WqF>1GU%ZBRBfW{4}zm2?}PKc~dMwv_=E-Hq*CV4||K(V7keznJt4(|SV zS=I=nr%{55NOZDzQ5CplKII9E!!|`jGO&4_+IJsnt&u8sxad81u0l7!Tw2@s57S)B zN(fFCVwstZA$mT`-bBVcYp`v^6j!h1QzNHf+OyqtD%AMp>lH8ehYO9}K*js-bMs8t zbJ;=#!JMO)`pfQO`e7Z1cM^+NOJ$~9=Hf;b`Ayb z%tqln_deOzA9XCt6B7XaqXhsFg%{Huc#quOf+^BgRX#nnhB&CRGJM3TIU1OTXdNYSN-GL>xC}{0waU<9W9(NJ5eL zEiffVB@k*p-<9Qbrs;d(s#-1I<9V)B3V5U*KNWM69blec*DNM^<7pGn&UC>P1040P zG$ck{MRl}^rlHJyA*7|`e_=Ctzs`ab;V{G`f2gXW%#zej5XJk{XElJeu+Y4`=+QwL zCxt&=4hfi?3)Ej6P7B@SvCH-xx*zsssX_&p+l2|tD~{hD6J;3bLNaVug3;GpnOT6M zXH1$g(r**uTJWmx&p%#;4K3Jor$ETiU+T&sAZHXfZHFl~Sbt*IIJ*j0j<=`@J8Mib zO}gCpUwEg!B3LJ%KS*=&Ra?ke>=j{ZB%LyP{{U%G37x-fJQ=5TRj)WT=q)B16Iqd@ zDYsj)cE9o~nE*~wRDR$E^(O49s(zA%_qR?g%_@+VOS>ZQ?ZEdc) z#w_~evCo-V?}7mza*FSKD;_zF_Dxk6N z>C38vBXdn(55_9bSjf?&eU56Lc4Y|(dby62&*XGA!BhC9vc-*!XnSMep3(7(>H;A_ zyoWvV#krO(XGpoA;7~22!)2I7wav8Nrn`3x+`{YNRZ={DS)-bka%Y4w-Xw>TioVsh z`ElVg1y$Mb+iQO6v@BQM%T2W&1bU6tQtm}HV(u4zHwIo<7#c?h^R_Aoy1T0li{N*l zP9^)d2?q!QQH7x>yM3LZeFD0Ce#M*py|k^h#fOL=agA`6&daT?xfhklSXzczb%vQmym* zNis8Op?HZ6U z^cH$1;@`O8&I~2927GqKnxP^AH4s|4@!R_!n~KhNEZf@i353crP{}g0jE+%Nw=8SNO-(jn^_q*uM;RIJ1poGVmdxm;*U+NS)!8^Y{GsAJFiKSh|YRx(;O zSk?Yo_j@)!tmG^~Dbe{|Nkxnhz^#%wqyodhz!j#${`_Jw#eThuUg|W_0T%>?&O8GT zjh@e9Puyr|aE`5dZ5Nb*fg=}5IANjac^%c7^d_hE)!>U}qYw`#YkUA*Nk$w0yyLj> z15vD)fCde=iWyU1HukMsSkLB?$2!Jfa+2f+`;3XHIT;CEeHpz5%i55|HCp)!uPIZn zC@0LHq`U?mm)3C_XFMU@Ci;3e4W)qbMV!2(ns6tK>-?8J8>wJ3uOSoY8CR)!X_%-J zV&){l5&i=~AUH7pPsil%-MWD)=E*AIjSc_KhY&RHCm5~htn78h-TV)|bTMZI!z!n! zJ`Vi;UKsc_h@#*8EcP~JoDxYT69UkMUTT9K-Zr;52V$ji)w0uZIdtDaE6plg7q;GS zy6U#Dv;_9!vZNm}O5-BEv;WxA%A5!X%3Y=L2h)sLOw;)3{CIi6JFocogAE`*0zSfz z*$HEm7P0j*LZYLsOV@#}N*o@Kn(&N2sLnf%wTv88qvi4yjCDJFb zZJi0b^t@a0vrflQo%j8Rxx4*G6$GdXskii*Ls+kzmT{$dlu_k4By8IDW>3RnoJIR) zVMrNCk%X_)Hsfq8u)c-m@jUNBh8-HGlb=GnPVBKwyh)P=_7UopBQ-p5fmL)r1~lC+ zZeMgux_gXzU5!IP`kD8SuCoONZ}+E0t0#+6#sp-7j7FVYMZYz6ZOqNV0)=L9(tddo z@ywavZ$$7-=hRDqJue+xfWQe^S8bYA!4X*p^4r&mqd z0FY&UV{M$vSK#nKE15m7*(SGha>a?C-1ZO2?!hN>*sBdMok<_p zvKYA^OR)Ar3ViFo@qG^(lIRf7B`=+e^xATS*S8;K!rjEbdasUAt2MveYdIUCJfClQ z2q19#oDL@cb}vQh88kQdAijO`Amb>7sY(S0znif5v{>+YjH{(4I(S{5Un^S=RwyS=FLsUo&b z$!Wf2*07?Enum+OI`n}=?O^U9{-fQ2KZ)3NT5O=zmD9gs3V9V|e1oSCX|Tn`Lvy%FMs_zW@g1a@c}Nv6!nd|tz5c2 z_iYO4n>r$X*2YG9`XSJ<(goxeYl^%@#Ul`vHpFY6hg>l6o_7(d0yND*wBBRJ+OEgV zZ=&ODL4sp^YxEG`Tc?eZy5gfL?Ia5J)Xs0Q1q>S6pyJ7hgyu-2yq~X^?{JA-H%XoB zm-Y6JcB)M;O!GcsTF9j=c@Lqx{igrlz>r3InMjr7QU)=l_ma7R%uL=ZB3MJD5(7;a z>ID7RZ{S%ZuQOa9g4y*3R^ESQsXg#5J>z@wy1?=+%kQnN>Hq`e{8Dql;G3p7du&;O zh1GOK(1|YW@T}38{QA!0Cn7#OsY{|-0pQHZxyp-N(5yXX!>19e%LPj8JD6mc_nDKYcF66qZ*F?c*G z!^tq>1L{Tbs4LNw=y#?5u^Pdc`r}&Y@$4}Me9MCyQg_-%9$eu`>Mt=7LE!BZg!4kv3(|3ab%g~b}DJM_Ql78BvT|x-Nnc!^IY7Lr`k`nX?ru5X)Js*82ky1% zsYHPd2^>m2X@n&uHhYG#i~vUT_hmKxZXO;;%@F(Onj-XNU}Vj){2?wB zVRR7Pm~a-8E?z=sP1CtVba=75dp-*qy;-%%2JU-A&xN=)pCu zZ+anE?~*rp%f&uDp(N);92$2obFFIya=h_;xEuVUskon(`WR}`)a7NBU-+r?Vyx_x za%G5OTAer);_2#;>WMbZTD|XZP=D$Pkcy5>TEAq*eZ_fdHr3h2b5s26rV4j;#UPV{ z3$%1Q$?`&FqI%%B9$AMn@l%8>U*Va)n%umKyqQa2t_2Oz6W{$GUl*1OO3Sv7BR(+l zM!r1}EOtThx@%^P?ly4-pZritoe;XdeP`oYmYbUq(ReMWQeT|dwN3^+VzcmiTMFdK zEAu)Qnx*PoL%V8ud>}0pN}XWoPnbTTf^5~QqoAT?yd=p((E(;wG$dHwRN=SZP@3GC zlHCFGQh#klUA928Xp1}3Zqm}CxEuSg%R{=C9E9G7biaMJSX8IJ^6UEqpz8(OQ~rk- z8RThJ_9Kegkge0Sw4_Y?1~c{cGz}dysYf$n-klGcdiEa#;F5jcFRDpz&)caY_TW%* z)JsOWhW2Xkve!P_U#-+u|0fJjbaeQ7gj%;L-|FmoOfF5yz}x;Tb>FcfP&u$?pW;p&_N22dO?THl*LaOH8npk1y{hB_@%J*)bqZ__zAFUd(X|dieKX({dB#X zF3^!Q_4@q`hBBNjdKt}zqytj7{go*$268&&UJ_Yxr+&|1@nKPDyPp|M>l2ySkkWs@ z_Jr$k%6>MeSLEtJgNrWA_)og|mzKu(RBI@wd3>uYuT1KNB?#i1IS_ukvV@uBCHE z4*+wsyRD6F?CFY=F^itAVx@^W9ffd4{E_p+NXPu+K#+@EX-Q9{bUNG2{wJaX%GqWYZc)TFxIiEC zDUbMe`)NTs+V9JOs~xP3F|sI%A@;UYV_a^_9QAsirn>xEzdOTtc*AFv#_N-M)Nsg& zf>z#w9;2gafRhqxCu<5> zx5<3fj}_RT@z(OeS_Ao zFJ#FCEcd=}X{|JFk7!jbu*XL4{fg>KUF*Cv+s z2On~IS-GsKT25-|WuLqJ8Y#K00mu1*1?`uFWN;Zn_-77YLFAHvhZ%$5t5=1&VH0n( zTRiEKhma)D1O8UO;jg7q@H3a~q+V=pPJQyWT}*F$=WJ{&*}Chgbgs0MMe_rnGm0ef zapvoy9W^f=Q=ECFBr6#|sZ3-3ziN@K)KB4Ea6h$N;oC&Pgp}@x9f;soziPM*PW`tejzM=H=T^=L_tV0`h=0U5vShm!2=9@p;Jc+2U#wmygQKlcc}x?tdtC@gy%B%Kch>g=op|JJ&zb8%)A3X2R6$;3g$9h;+; zE7_g-tx5hhXm~SLL-*s21a*x5r_9Ev-@3aJ=FCy4$sczMdhNJ1*iWAQnV`QBm1=B^ z(}t!!%lU%ZTKVRFp8nrIx8^G7rO!S@wgVlb{@2T(oxAZ+i2N^zA+yI)F9H4Nr-aTL zj1;+fqf&=>=g-x78tNJh&OYYSJ+HkA&oK3v8M5#pJZtN1IG5iAWt>P5ze_)9p%p_6 zTANN5ws&i=WTR$2O02tAW}jIc7b}hPJvFg!#AA8BPTxpKxHEQ~mR-9hsxfQ#>YpF^ zTQk>pw#8%UTz3y>%OUXn+288v-`69|mpTno?CYs(@_*yT@9KH`Rr>xliG!F5_RS4U znMg>3Sg2KCd;gemg!0d_1wf3;Suk`L{cp7SLk58lxFoi3M|p7Ht|u8>$JM;(duWo? za}ZJqTp90^w}@|zxPkDK9K~ewMPLp2kxM6tMW@O?FhJSibEdP;d%_531zAN}A)%z_ z_k73AQr(KF*&lRo=G2PJd)op?xKz|s#57<5zMotz)u+B@az_%rO&4a-*?N4KH7kH4 z4oyu>J5g|$JSIE)sVqQ0uqeSOAtKA@*cexmn-8nU`NNN$kqXAkK*{tJ;vgt9;1gPm zyeH*Qk`QV_J|-R^-^TK2#>GHGbVEm?_F3cL5aa1nI(5XHt0j+Pgom`_ zXV|o}R7rzJm8hPn(+nS3p&vbljl+9^-fVX1_gk_JMRe)z-SS*H)oemx1SVNsVZp8` zo^8mpXf@25^&^7&S5h^>MJ&ziSp3UdZB|CzkyraF4OeMCH`FC7Hy`nOh}oUYjlqNy zMP>nVu^~-Mgo>ua3F~6p+Moj;XG545lWQp|?hCZHKj}csSow#d7tIlTLxpN3i55P* z*(~MW?JsUkH>8ueEq?CGCb*V3nd2X)ivR6^ z8_h_qIvyYRUM-!-%u9BJ!7!_Fb$pK0u|MO`0 znAw2SWUg8|?HgK#77bLKH}6Kgr$$%eBE8SO5N`ag3z)0&V*Bim-id@-58X=g%ZtjO zZ@2P!ldI8Rc8fK;V$DA)fxv6Z(sG$dyjW92>^5UKt2y_;QPg&u=6Nb%oHrx@9jInm zia%zayIzS_{0RIYJ*8Uk;0WCQ$^+e=?`?M9+&g@^?0N;zsr77FCTa#gZd|f3916Rq z8FvOiUT#SFem-ikNLmb}RuhVxJ{KtzulTH9F6`?XO}W0O=b3{;r#~8pcE`#ZLHmCl zn*aTPupWw&mFLe3^HjGN4tf2G)Tk+Vb~kh;C6_M!6)O##P0pdXQ7a94ipl&u<0}oK zrG%Wes=(zp$;?LxCJl>2e7Dv~+$JG7j22|VPYJl1o>|>XH6ZBpapKo+7y1yx|YOC7V zftE0v6C5npQfVAhY41K<3(8cNg-IFc<8kk#`xRfOo-{BkgihDixA4Gn*H}Cj?`O^9Xyj9ts2|K0m zdK%Cx}dpB&*fZfo|bxq6*+Tn{E%+B zyEOLkNZa7jhlBLVZ(f#m$vJhVuI(h@2mnQ=!{2g$)d=R=#v)}ix zf!-ghpa1s@S@(0xxFV~(+U5@Sf#O5E06*-_+;47bySi_Sn#_3!_@@=B&8G$=!{4r7 zNr+5YF#nhL{HW~Imw#H2H2N$gpU!yL23 zJdh%^?CprtIVl^d>Zl0yL`9={tMn zTuu_>1ytg+m8uk}_icZrcV^M8?x`k58Q7m8J)p1IrwA9fxe396wB#0P*F{7n7Pc^K zlsMKIfn6T%%g35oR15eQa2waLl9v-T`g@A$Cdy2nmni$aa2eXoGvnq$sfTLd@$TGt zt)~E+BDZebca*FaE8**eF1MzCz*Sbd$Kynq4i_uaoj6eNC4B-Oi{1j7%Q0I`LY0|E zl|f>cjeY7bkBn@O<1W4IbioRy%b6~SqQg4T@ZJp`c(6%wMHp|y#}S87@seD2{1b8- zC?C+Bxa6G1biG2X^TE2uSE2vmxAXJ=zUus+-fPvv;x6_E@)cY>rz>HK9dV;~D~M5# zGtE2yG%ur`8?q^KqPs|=kedA7we(5Fw`viqx%?;`v;cJu@39y<&Nc9{bpq;I7w2X`Fp?5tx*JBpdv_#cXP_!A5YODLa5%5SwT)l4<2_TjzZy7?8AxoJcDaw|Jo+_G zZ6RRD<6_6K#IeA#xwa$&bph#@V~NVIEwY%BX-gD3A0G1`Lcq33Fo*n8Yqd>5Sbi3( zYHxTA|NTa^SL@wTMPA)=66jo z=k{*%@1c5r;mk}8CKFk`L1MVcz~L`Afxvig2&AjSZgEqZ^R*ZNm3f zU;6ZuhSH(=y$1-DB}!V6VrFPa8+4NQmY-oUzkuoPT3{Xmi$t|)^Ky;&KTVbY%X`Ke z>AJxoQ{c_C5{Ym$IwsjpNc3iQHO6LkOhI3N+`99WRfFu%W`Atx!8oHvg{E4;dW~I; zqx_e*DsR1Zf-o;v&C)twrxdl&llok(M_xwlXZ3CH%yvbPemQPP|6;wPyRXEtI}OO$ zPS148yr8GL+tc}SgV8__=&nrxyUZfu$$D3P(N}TaSgiM(saE%et^x~cbQ&r$kipi|%>An-A zrGo6#>If&wNOxJ6)xcq9T@=jL%|ERi}dXU6Jt7j=uM)8rbh) z_H8MkZb;RahX}V!J11naOlLY$<-$R2?!kK^-@iz?^#;n)66QbdTuun?+k36sc|^=? z&i^~(!_Xx+Ejd0ISAh2aULEOLh2R6GUvI8#Obh_@XkacQXw_?r#HKF09}DQT7>XJ5 zs4-~N(hPmdfbFuG!QR)>&@bp_iX-!7I&MoZ&(NZbGoQfNRn3hF#|0q79)F z@J_G@aVeZg$i<14Lf2bz0?!1{)` zH{Ude&8xcUd7l*?M@wV0_(4Yn081>_BBwYm0bOWutwx>QYw_UN^HiR1$B}A3Pp0yWq3ASZ^G?oGV*7mZ4w0 zF1A%{+;&w+^U+$gg;ACZp&~vC$Xu>kf^Sc9TJRddqNXKZ_f_^6pk6P%WLgw$){7Go z2HTwe<4Pw?wv7Ddoi^hxC-DY^?|SjR<^Ah_Q!6ror(UVDQz*Fb*bX$&)1;YYjfXoO z{|zEU`3Nw^*kT+e)q&c>AZGFSVxjSWay8#O3Z*J}n1g zsuE?c_L>dAlT`b3JdxoiaI@eTRX7F3W9)&a#RxMf<dx zUIXGp_Gg>H&POvJTz9XaCUwLjY4H4QeQ}@aC5cswc0=F!*@(UJ_(Ss=eNdUhsu>5d z#&a-E$WSk*eczj;0}_$(&X=+DYAAYhib9Xp(W2wS0??0EYi$oA`!l5}=+}`1Y+aFK zrz+(Q4(4w~^k3ufepyFY$V;{p-IysNxXjBiDnotc1nJ>07qaAA2*4Dv+7f**0~@$0 z;oZ4TYitj5Ys3QNS0BIlT>tta_g>L8-?1VaOtvzhqFCd8)D~(tZa*ewXwf|UZEEWI zk+Y$2p^ey18{!}~#7e}pUul`l6=Jw9sq%zxW_n<2(Ar%C=sA(gL&>Bg{|jns(S&z5 zKzA2?#FIL{t&fr+xUMRRN*$q)ak=2z1pXu z)Hdsv&b^KQbK3cS6_lg4ij`HXlCXkjzHR?zzGV8yF$~J15w@O{&c*Xgn}?rpwF0aSAz!XUVtaLwrSU)7E#xeYIR5K}E6`c+RTx-h5rs1rqsX@~Nic+kS+xAwcO63X1 z;f+B!*6Jsm3!VW^8UflFfSQ0%yg=H#d)*8zwQMx)iqDFK1}ftDH!JGGI`Uqsv%LAAi08Rdrdc&&kJIUbJk8%Zm7YqV;MI;R z@Q)&O`=K1K(|kNJ+%kZil;@*DEtdU88gg-f-D=%mW_qHcN>MWM2>D5&`(>PqYA-d$ zIeSfm!A}hBKh`(`JTcI4Ex~?MnQ(^Sh}(O;jN3H8iEg)+Jk&`DwzJ6c`9W=RVM)MU zoY+H+QE2qA<=e>o6_G~@%j$}oE1y$p>P(S_DuzzRfu3!H+LxE96stp?RNTVanrXPcY2Cu9%l}zPQ#}s_QSzY7M9nVTnIO6!ldcv zBC@3Q-yTgnZHokhY?iPE6))&pWTl)q1ma%9`r=-b81QjE*n@t|Y(feCSx0(U5ON4O zWg6)M`r>F^Q&oe4|4BoI9sVpZeYi!!C;V?Q{=0rq{aITkOJ@96+*rwS|ARRQLonZI zRaXJp!yJGW1^*46*eQmDSzpO&I$uySM4rFIFZGD;Hd9ZaNp zjN*!?CE^7I6pAralawlzKzM4_x- zNO74V>{OyeeVNk_LaH*xFGTUdyvhx?kyv~PyaEDtl4q;ksscc>>E8GGP%h25SM8HC z8f9A81q3AL$lqb>qxm9!D<-9{j2_?L;}%}0&7!mek(8ECB#;vZlSuRO93!^7Za}0q zGc<^2eqSH-V(omyoXM3cbM4$Kzq&WM6Vd zsmqbxq?gHK!~Ex#3X?Smo2KQ_$L2?QKK&^1Z>k~v?gtz^jN{;=Nd_VRdzZa2lmei7 zOhXmmh(}5fEw8TOdFvLTE|ofrOB=+~?Kz{m!$$Z@>Fo-_QMzD_6)` zYpyx(IqosWea{;8vg|_lv#TwbjIxlX0B2&=wL~e%t7LN@syq4wuWfkKg8$Y^3mTdm z%au1Ju3M%VSb>b@JAsa5jXf)#ptr_JTrhXJpy1Mo0wb9B9P*`J!yeTO?|vz4^bM+f z*fm0;`xWc+s(Cs+Ggi4rvBJ#_d+F>mtkszjrobuN0NVlU=z8`$Q8Mo}N?1p74@Nt1 zaYX_*&xuH)9X#x+SlN~!DJ(XfhUS9RM1ih*fB*JpPu4@N_JXMM)|E1-JF|_Zg9^_S zRjX?Y>adpZAU4KKp>4eyaHavZ5I!Yj^v^$ z@|X|oYD+6mNo0zEf{J|m2Rk&9S!Wcx)8gwIgIAa84mkPG%`aO$`kkKM^X0|`xj+0; z-yTc(4UL}dhR0}U^-B%DANQBgU;BMj?`wwdX3Z#v2w;jnMVG7z?SuO;&y+p zc(NO#gk$p!^xlGvA$Ci5T^2W&H--`{?_HF<1g+E+b?W;ROx{7cvM4g<&p-&~^Ur#^ z)H_9Ah^_5Tg9TA$kImQ(6HlR1Zy@PjE>WltgSkdJdT`0fvAcIhij55PVLt@ajALzL zmJ36X+V3wNQC@H85LGgbp6}26))@Gh7$7QrQT? zBF9JyG$SCra)L4zm@?wgoIh9pcBv6)%yKI=_X2t~%}i)T(B+|SY^;j!Esve)HTe?d zy&I6-@UAD^-cbI;eD3~EAILtqdYiVSNn!BZV*nJpPz!CGsa4l1T(^x$b34g*e;*;e zQK3E>C0#kmGpPN%oO@Jr^g?n)k&c0QR$rR#ugJEGYer3N&dX4u!1m4cfvP8F$@sV0 zAM^aWd42`TE(dY=EA1?>j2WrJ85Ra#YeO2$>4JW(%`;luk;f(M+fkLJ$NNcUQKv+E zEGl>%CDc;1?>zx`fX#0ZD$Pgh?W09V35U+!pX>anLUIHJ6-k+wU79>vIK1U6EG9xh zh4B<h+5cSIYf-|kYmZcdSo zBqo@-enhfLct|l{#*TOKNQZTT*~(QfE{@4>qIsa8QLyIjHDagG@=Sy%gm#!g1~K%_ z)ci9F(yq72KT38sJ1+evwbkkO7&mbrQ!I3UmT*O?@_zZKu$`c4=TVWAf1FJ}zDuY} zGq8@5V|V}+uS|qp2uekrVE<#aj;jml&nj;&zDH@`hP5^^FFOr{GC}`XnYLQ$!g#sd zk&#nI&!0!5%UU}!oWt7`B;8Z5v>nS=4=Vr%88pDtryZBf!KES!j<)ZzA1xo}O&tW7 zKHewbZW|JIeIc`l4LzhH>w439xhLz5N$4jx+q+46hQ+2DjOoA-qJhKrqID5@W9K46 zq56h6k`1I%-f+%zg1Iu?+UQ%nwBclJ&njw*Bgwey&--)>U5A{I*}sEU48m2o+%R?o zH3)SAS@z0SUY>;i;Q5%V;5taJh$ld>_%a`eYs#1U0m22^UdOD6lJTm=ASbCKIdsxb z$0K; zy#}}G45Y5Ufo9w}T1x3CFOYtp>iE!s`~>QYWqVF-KZKbj3Iy)~aA%En6`_tdZ9%)v zRtGc-$OcNhsgf^JY2g*u+s%^19F^qrE2B&8hqJ3oT~@cNRsY1>Oue#AuKQ=^U>95> zMKb;wl&gFE7ejC-=Msv96l158vd!m|2kuUIg?+)Nt`zG)TT|?<;lIB) zci{aM(_vIXo{L;tf4Qm9DPt;cbiQHttaazy6_e^exI5y!bh3mzPFjX_FtXy&bzT~| zT%L!OE$2{b97R|2uQ~SX^G4}GTg?mO5%s9B)7cVlQSGS;ox>HbqK`Hj(#-o8h^XGj zq9w~2{-yR6K|k&j_J*0W5NmHzkR7+O^|8aRqD80xZ7I)PCs^JpU4wtV6QpKx)>cZZ z-#9covL_RISlNVcKzPlh$UVK^BYnou>2=-#9Ly8?r0#jkvcAWJ++JJlM)AQ8DvV)XJzVU@~vH4~9_)l#zhKwCP zXT#y_6xgD>@Hj-yn|o2gZOl0;g(?8#8A%9xsPePxap9?tx(mIZBnTGVWRQEKk3S6hS(m+8m zly;+qeigv3l9CqDCmsypjLLae-FUotATU`I>RM6$sKP zvugk10upw(Vv+bG^4bNNw}17ocY%KGQq6QOZoXqBlVRGXL)Jq1Pa;kkWk367u5ZN9 zs9WVzB3rK-brwZLAu6iT*+Z?-4_!zx1t!pA?0-CfmD+Lr^vMt&p^|#;DC6$>#(ckt zx3tO19wV&rElIPQkT{vgEfVkZt>Yd-Dz7(|a86!$^r8Pt^~zUGXx*df_>J=Tg}T*O z(%(^zUA!ce&IeT$>(AQb%SpE>sFHl`W=BDBof4|_XEQRlkMmY3e*&yInUA=uX7b!ebB-Is4bqDOc-G zDL!-=JSsfNpETy3lja%Oh*3yPO}IlXz5N3e#LOcf)Z8J6Day8L1aD1C8g0J7D#8Kg zTD~3ZRrr&C9PadLvYA^^lZd{QNbZ5!{*@aiX*R%8UY(4}j{MCp4rW?_6@3H*W2>1S-$vVVu7s0K|`c2?-{h_}Z#vj7Fr(-H`%F7*Bjs;3ga#46w z=xA!}(UIyh9^UXKqn87yV@5j1ln2+&6qDXLZ3>nKeY zHOLNChK7jWx@uV7SE{ShA1N$(+$1SvW)<4#p|WPKVZz-fCh^<7pKKHz-t6FHbputmLZf~%UN?i8RNVP-i$pgAhsZSs) zjN6n`NbAnjftm+`D(wjc)#xi2i7kzH4S;SHl7Jvi5pAf4A)+eJuvgW=6-aL(tWb%7 zMBIY)esJ;N)yE5c_}ap&h4aG`HlQ^Kw^xdI*wps!n-yl2xmT+@o^p4T*G_a*OZ0%q zwSh&=W#PPTFps#F2A#MTpE0R};EMu!u?%y{PL{CwINk8DTfpMYS}nj7u$T+yXFUFM z6_7**uZ|e%K!r1BKhmc+2;Or&6D=oG7iy0q)Eq5Dkzos~25l{?GHO&|Qpgv)#3zAg z7XXJZk)RQ=p~XoVos{+_{g8MgStPctaQ-p!+kZn(48HK`(VG&@))SYV-aF2kpS)=I zVPm?hI>sBr(^-Br%*iGhrJvwi>95;2-=AlbP`xW(LPzqmSlL_JKzM<9*4iqR2o^$n zq5D0B24!$dw?SVI6#DOat+P)W1 zMq^bK9lKEhVu$v%&V070eZ+1}X!CP=1ihRw1?#_%kSJ{dSVp9_xnLiE994a#NW&)H zrf30#&{64(O1>2krQ!~N@5}&oA&8q{5Xjh=SRl^L)u;}GU(#t@bZqX~krGeLGga!s z97daYfs#g-tZ(9TZZ-O*U=Ahsr^i%+P z=Ck9R$f%|?(>+&G@cBOznE&c8_i^4!cRpWvb6vCbcv#e7af8S94~8z-CDxmKqsN8+ z>IE=vyZVFjoTs{rM&I(^Q|^MjZk2~NDu`_2MboGi{m{6g0II3M`jWOQJ8$1$2(_;z zYpIZ{q#xR>lp?4lR2uqM!EI{s#H1DU13$O`9?iFau_TNgtCyy5eQlB6)$HuQ<&V{M z{!rk>LP=^L+mK$kJ+2tDJj~=5Sgd7RFO! z4^gs(k;`+k;7tB@-A8aq#5D50CVYnTAZo>kIY8zMF7`M|d zdj}@xBeuJ_uQLNP!=;-KSnhpS8TfVhS1kM${r)op%m4oDOb++^-G&>6!CopoEh0i5 zVK3tPN7IH?k0_JVz6>R7Oa67mFBMstBz0$3FdO`0+@B_D0>5Oq3o2Y8o z@nAD(+P9%Pxwto<4R)(kz3rCz0#TDG5`Td?9I#tugy;Tm#^}Gk_z%5cD1qaOJ0?qB zC1#;!hKPpT9pH(YqP9P%oh@fWC_^_NBntgw*fyZM;a?0kM|irR>t0~>L7tZUit!3X z?C`abkX_hUOsl-B*jJ{7uN>>T2;;x=>)*uW|LdSP$A6(q>tnLhQ=jMlRSm*A>mRi3RP~fu&>5^)wM}xaQ`h zW4C_MJpabargIxQm+04ps5)cDw}Sijr-JTlCBtzmW ze>vR6Qx~EcxtYRl=l%v;Jdh4L5!h+*-+0I8c89gc$w@7~;&D`J2Ct4l4R+LptO);m zdNzl1t;SzjHUsAbtOr&3@C;m>3wQ&+iI`KUTkW zHtaj!@k@tK(xO;zl1Y!2lofnThVt6ScynN_d4jTk5B3A#5e#DM(=C2xp_nrPOXhbt?rd2@zE?Zk zoZo=vzY47{)GPKLy9DXg@7#9RYaR)2k6Z69Q(*@=>^v|5F4$`5efQU$J`wEV=?w{R z7qWGud%g|0JsaTE+kN|NeX(XWY2=r&0>koc=pjgMZ%DO6{(? z)&IHrKL_OB|LcE7{(olO|HYF1pB48%vF`Uq`#;g<-x-SkpCauBhku*RATu~AwgUe? z;{UxH6T`WAsSTu5bTZy)P+H{XGyV91Zmf?J8Zhg}&)n)#;;<*eso{WvTH2}HD_y%e z36Hm#`W7wk!t_p7bV-sjDfv1-P~fDKEL;4F|CBLcU)dO9TQ~e)+{#~HG%&-m)7euW z`A8Kmr8co@svYmYX40w`q!;>29VQPwia466YV!&j&?tm3Prn`JJf2`_+MS`9$^5|( zHUnr*r0UjZekOCFD%tdt`fl9P_9{HrqVs3csgB}W9nrkv=!1gj`5b5dcWO!YpZUv) zYpY^%tH-Wwag;Gu8u9MpRtn)exIwk%VS;(+{fhB(nvurK$52MXzU*hn{XyM@X>7`L z0{hQS)g2mpC^iJ|MuWX-%VV^X<6m-cJ@N-T#9NJe{;EJDi!JL@5 z%Ld)#hSjZBug(l9!C6e+ z9wiv<^$EZ$DZLWprgn?^-eB?70)|9Fk5UGK-h;kol?Zq}`;u|dr`nBH$%>J%GPEJ) zW057|V-YYzaMK$kmx&BbRefZplA7+~sFH*^!LeT=uv!nqDnTMy6HOp0h{H@!S^KhQ z%L4|OKmPq&{WUA!@*cad=p&=Iy&Wa9$OReNM(^J9D1B8|REah%W#!|4wx}?y!%lXE z{Ful8Ds)9m}RlGCXUsiRIu<^Yc7jlM0F%y#*&t{`b*XR_c8J=F4kLh zJjxGmJ+18q>wBX$tSM}66@5y+EE5nSVwHe$l8p-xvl8#TcCUU?|9pRe6Hw$Djbrl( zOTXTmm@Mn+Jl&Si*+plNk(CeLiIYOO`J8)a+Kq5VQclEA!Qfo|vvlk+O9%8izLz2s)RNvbEzRwp^YxT{g%W zR)nM@FlP6<5U1o?d4-1I*z)J2!;!)=H1TK=wXq84A?rA3n7%ty)Jn4RcNxOwG&dy{RlE??y?MO@V>XC6*%5a@P<;6FgLlqt0|ghS z;-yTDBaU7;M-oCDb+&1Fd!Yc_7LVuw>Mp2(kVYuw{;GZZVwY}8eJSxF}{0e`yFKz?!P$l9G7)wy;JulUFG+wu)^V!(xf72yAqdisKpx;n_dm!o9p zlzaT(0&Ocrs{u=^d^f(1h_37STeMK$7aSy{4EeHUEG!GrH`CcY3@;awGVK!!SbGfm zV)SAsjjKKsuf!1nFX7U%*rrKC2uDOd)!heyawMu?}aSz4^q`rBwb zZI(f4a?*W4tW?UNxlddQ1JE62qTw04)?F~Ds^RrPi{!I@sZauOox4!T9@2XYHC}pq zC?ct2Az|d&>%o$#{OQZ%qQ)va&F5A75DTKP#E=OMmKw}LgtN)>?M2yIHP@OQW))21gp%vWQA#{o-gaC)uqma` zt#&m=VA{zXSS|3SptZzn8*c+asM_tGhpJcUmS49hQ5<%sMZlJw+B8_<4?i)Uh{yj~ zg3OA0fu|>{+QdsfyEY(Xm*`mTH0Vhdt|%FhGON0MuF*5d5^R|1K2d)QKSJd0uq!`?O0zDT;CLL0Mt)rH)!#a6< zHG{7Xg#rX1#t6Jp=<<_4P_YuSWNQ70#dD)_JrGdpB zy%%w#VI`rFtT? zT;AcjMxIdoj^YVX*oj9Fx%H`7C25qNFFWHna2M&g#52$6$jzrqvU=gX>10Yh8Q5Ib zc6+ACIHK-z^n{lNH=9N^_w%Mm-XFz7an035{a{tX6bAQ5W9P2UQpK&<9#?3A znL1|ponueDD#C16$o68tX%btdHwFEd8z1wyE@1B|L*~h=~&f6i_ z54{HdMRS9MSO$q zkhb2*szP;}Z4t z5D@NjEy-o@H289u6j0A>m{R-v-Ngv|mi1+n^Vn9`9qL6`ew46`>&n)0PTJxT2;BVD z;HY1wPU`hWnx{~=-r^|9Ef$}DTHg9)p+3e7IE5)nWU}0j_AzK-6h^%T!{)TbodEhSfyW(J1nfqHu2p_TH?vY)s7V02y~xX zSs0)25Yi$*@yV5QiK(sb4X>ho`^zhdf4U+6oX30iJF)_ruX6mBIZWaWJ+&F$*QW*h z#g$}ytBY8v$&ld=sEI&FXG-N4U>|*ox}PejL(XyZZ%Zir)70D@*5X}p%~6AAGs>Ul zSG!|mX(_iekX+dQ_MY!ddr#!;l)(Z^dS^HQ#iNdfVV92WmI1ILeaF0;kG=8koz(sm z2Zs-;K{T#e;WSuke*siUDazX7Z@`eqvv&pDxVk+r2mf}EwKNs`P8HAEVO1p~7upq) zxe9qVRoUNf;A8Q35i4^!Y~d}`E<^;{RE#7R^y7$GxnrCFN^Kq3h4lRnD?-Tc^b}h# z_(@*1GHWLlp=ZU8+-l3dIKZ6Y!W^kx(+1eA1mi#Bi!?&hE zU>Si57po8y1@!|XW8bg)`oMmNL75voVY3_r8E)jI_rg?TrwzKaJ+j(aNwM&&gM5HI z3vjCdyip%+b9htGE%ZMd1Y$p0{FzVeJXWR0CO?a8wTb069_%v%W*o;GDUS_obSGUJ85*58uNuR2swvel_eCU`*Zm zWw`x+s3R=w7c+X0l|Xgt4C)@n2p@C{H-VL0%o7)@P;$d{&-ZO*Ry8cWA(P|=$*I%O zu@XCUO27=asyBwt#qsVqxiJ^+Hbm?`zesw@rcMi|`NGtkY4q_t6uGl@iF~E-eD^^i z$@i6o1SXMLvz$OvGafg;HLweY$_S|jF|5`WZsw#JIJE*Hhl?`kg-Edt` zlEV{FP#%}A@w%Nn6eP|6+sV4w+;oXt4y#4HAc;=^f;N?NG_@e;uX}RGXkmE|goYn1 zMnnDZ>cO)Me}t$9gz*XmWznsYz3OF*jY^)5VtuIVs_+rvA$UFqVz~vncO2UaG@IN^ zN4~#e(gXQG_yYF3f9$1h#df}~K^8rP7M@Hzjl0BU26XGhDw+~#(iP18ewfRihch2^_d7%8=C4} zxeC@?0igKTH6r`T?hWGVA0vwyh8e92WEX$sFpTQn+cItCyFnr#` zJV;gE=*&4E7XU&miNScfO3&8<&KN78hOfdx1xHGgsp`3hV1(jW6JEZpE#g)5Q!uiM zY|pKB@%4p%AF}0uju)-=Mlu&id|eem&wpM1GyVncJ$9^fF5B+OmU0f*MC|{Z*jo<) z34^yO^F80;1dpfvwgn+*$JE@*?=g`*Z2WeQh2OF!sl*3O59sC=u9BIB@gN#jV|@Ed zLkDhY=o|M6U3Z-enLd9nrp$kJhP#yb_QFHw<;lh!pYRQ;l=)?qp68wK+}G!U!%f+y z;O|1pmU!zhft^aXLFtkQ$?`5GZqPq13kH{)w74;)3&JK3c=ZgS=HM@dp8XV`(}r7! zzS(Zt0D3`(l&ifGYW`(w@kVEvQz2e)uK4~_p9i>0UqlPTWSsTEYkpF>Z!yo!Dx4!k zlnc$PJzqur5n{*PKj%}Yo?%pPQs65HNXZ{_AHG^P8x6GU7^&Xa;Q`&vF7VjRxLcr` zq8M>;Ygt2RRX~l|oTfHBxP*@tyYY?uTsxNWSjTKF_JK2Be4<{A}*hqJ$` z?YNG{sjCuKK9Anw++xJ>Kt+%6_$&pwu& zn1`POME1PR?&IJ<4vm%O-18$3zofZfB$W@xRnP((tS1`9gH(Nn55i!LmVl`^&ZsSH zS`OKJG5Y96tisc^#9~@tmp`o|ho8L-fL2nh%FHFpPxk%uavKe?na(r$w+22TR3OSz z`ff+pViiWPKFR!B_Eqg4Zglcut#9g_k}J|p8BRtp?hkk)9wlBK%6^pWn)NWQ`abxR z$v)@3JG3L>u2hP@&vK()=Dp;KTIm;VC6c|O)B zq^u}wu<9M_QNnLl-Sx<_9DXmt^Q_^JGu1OE zXAqTb`FUwG>3@foJC}h1s<0{4u6h1JJG%Tw1@pjyXy4!;VNt%wH{>c08(bIoM(m*B z5vY!3h2=zjYs*hFDBE`Qkjyj5jqinnsJpd=@?duXlr1_W@TQ-t^Vj_(JRCd zI;|a-Z_pI;W?A`nUBtiAR#r$sVu^p0sLxjr?B-(Ma%??l0$%X=Jo8Yv$>l5Eg&cJb zJg@#UW5mbFulQst$h`Db-Z>@?uI33lm_oTdjx(=&GYYFa&2BRW2(-YUOG=NIwt=*~ z6rgcPe4VOCgp`q+%i{3;8_}vV_WM&8Tr*I!m~p2nIp2xJK}-L$$`TD;a5Z@QdL>Rt z@=*I#9gPJRt@?cLlJnIm76>tkzq9id<2w<=Js5i2nI)W7W22|Gij@A!!CbxsDvK0q z#eLZc9mr&L*X8C;jHfqH2T1hw;IXB>VFBgv;SOR20Jd4I20fyaTaBPn%Vvo)>uTh;EUI|o^T=7E?Vs}Me zWR+>#ajAUW@4}F}oty=V@W*K6u`bomE@qsY^bs#=aJX1Wi$OLKN3W$wk_RzhNs)(S zH3#-0z?+68Sggc7s}H>NP?Vgllgyl7o_Ic|>IP7cbM>)vy*W5g7(QliXXrRCo`iPj z{2ceV2dc38x2O7PPXEht?;InZ2GMwx?12h{D z3(UM+7O)kku+>$KN(TAd6&BWarAzW7xBAcEeng6xZ^E@kbh=@km17=-0x4)-Tu2pF z;5jw&`IQRbRM2`=RtABuuCr)2TlK!_Sw&j?)a#&y6oJLwPUg_boD^FrbD_oQHV6@z z@Ajw{hz1>G@sT^MYBCqPUc)~g_Cy*pmeT2 z$Kr`q5x43#{H($k$P@a|)QN?lXIjbgTx`ojo5L5e+5?OcMv)5VyISUe5M6VwCu@X} z-?T;D`LT-d2&G6&W;}PKeDpE3y=JmrV2~5fUYDsPDnm|kQt%+ABTc~qnPr(O;Un&p z*RfDJ*FgYAfwpTn%S+2Zh+QnLETtEgD}!A#Wfjq#j56*n+oHc;)e-hG4`^JcF+Rol zj-_>Y_Af3lC72Cb2nxb1ai*aeRKs$7<#l<~=Ht^n8?U=m$Hrp_8(Kp10%~QUYHr8w z4K5F+@JJJw7g>v$m2CeLSG0sB*HpJoo7LFn^`p(2yvJmHz#Tpq{u8oaRX1mESRS!p z9hapJK6il5@*VW?jB5q=6U?hiJg%l^wz<$Tb^0 zzkQ8Z);$XM9k;T)yz|;ix}?q$?x={-c<8|D`JlD%kdc^MzikH@8z=W-{U)ktqxl-4 zO!=`C2s6;INPR{V7Q)uCOVsXP@oQ_*j`K#PHVN9E{y(7`D;%(G)Ow#*V{dzjthUKy zYx85lW65p=mr<;5Io4<1mKbu~&KJVhoshusUv7NlmYK`_5)B191j(0t*TmEZ0G#r< zLN8_n{?G62*^V>3scyq%Eu*xxKqai1f`KLj$Q0JmU^ylcAQOnwb`+oJQ*w(lB5vMb zCCqZv$%P7~lyQGoB?kxK*mZgWn??G`u*zfuj#b~42ys(%ycB%A#)IGRc=_1;e0pI$E00tb23XJXs$8N`+C6qyniqzx~6&Tp##$U9xJ8$dwxS=h60aHDKWhl1aL$E>%uaLiUN8NCuRA~?LeisnCJ~B{9 zwz6w|ACsh62b}7jU-`>88}n_8fR?yX=f%Cat{4|TU$fGKe}Cn1{l1{B zbOB?YHu5ZjFzYipq#A@#=b$-o>BRRv?f`%@$_$V*uwyaT=b8Q(Z%X&uthU7}iyTsC zCbB$`j+8~s`(T@H4(*F(`oE6$09OhrA~~a~+3J2>v}%_801b71 zBn!&;5W=jx-aCZ%o_WAXyz~PBYughf^sTqHO6Ir z#f3BZF4X`BKAP|}oF&uulOgF9)-Ep4o^D0F&yKrfF>*@5@%>z8FnaQInFM%e^KIb9 z$6Ix?SNQp=Wsdg);?Y}gJXpFcIa#M41MW9>nN=WfPAnZczfCoX5|XrU5`&k4gm=Vo zhzYyKK#co}Lc^=Arq>^CO!-8d8vF_#91t#d{8_n`6mR)4Q{GXAa6;Cv$TG8o1RD$L z#Z=EraK!SaB^r2K%|;tf-5QEf+At+f{uJ z<+$onX`bNzZp`znX?cQWfTDyq#sj5{pry`_y5ZTW_v~2guJeOL(+Ogc0 zddthVPjMfD9bFiSk?5J|^6z?(?gRF>thJAZ4q2Ou|&F8yO&?W9r#ToYlEtqpYJiT5y9t)&byA4_{`(t#@e@_BOgSmjotpi z;`llk&YGVWszPkcHH|=B`G}wD$agzfwrx=0;i=dWiY20(F3BgH z`+#wRr!7W7xy*n^M8W>$-0j!yv&yr9-%@c$UW+JPJ24$jcr8i~*GZCn{jwwUkmzt_ zxkskf@D~}iNG54{A=Uqwm;l#goifDIKSF1F9S4k4cI`q;T}+A+wS(mLnm5HMsbdIT zSY@oo=HAb{$|IFRedv%1Ag20ltiER4N*aryCUshIP}N&48=c~-O}B!TsC-m%voLLI z)}B>1*1#_C&N%$D0mm-}YRkYXO(0crQyE4ZX||b0jau4Lgk8DKK!~-+mJh14(Fpj= zTn;IE5K1BJHE5h39P>NhknKB9yyi*jrX#nnHLNY|2Rp^M@LDJUd~oc}PB>`t6)Dtz zERX{?klO&d=#&YRDSD|@w zzkDeSA7%^eoG)SUW}pH*P%eWQGJN-ubr~9O{#PfI;#jVKc$EP7d}o2Y+YT40=xL^m zu15%;3Q%DbFpv3MtOL#!%7ughMhstJod~K|3}33`BOJi_GDKNeVv7l)0M)f@n)wQc zNuU zUw<^f6uv@)tA6Kq30IVN>LD^ft8F(e3(WheKsqb+wIW;R)YfDA8oP_R4w~IFt&zsX z&1;(*Q}`U{bFu;rSidPhdrGb^>`?ikF{t6iplkOONeLawb7JoRZJNb?2D>i;YZ}?q z^f;OVd%Wv1zFGT44e5v$aSx141qGz`$H)2D`3$Z7oocKRY~Cz*KIRf-xvC2(IFmd4 zsmr%ZR@Gcoc}y6uLEe^KWQ_&NKb@q>oOG+6yU5Y>`YxaE-8S-APz`iyTEe=;0_B+__#yu^{&!TY;b;D~JWc|A)`g zk1b(~{i)wXUm+jq_cH+O^p%=0bRR5-oT;V+-P|P6@ukax7dK7HVwX*IHyh1duj*O` z_&yu#z@Jndfp;sB_Z>QA+r%KUFTlIRPhdQRYz*Vno+%DY#0z#zOjDye7XZTpZ#KOJD znE0J+rSo+jU<<@VwRp)fMWb~Jkh8M>+VB%Tp_1>_bnqSuQ)|~YaBxb_day$c0s514 zR^+_+qY>!AObs>`ukBN#Ip_LX&e5>#F#Zrq#cialMfcWJjbAC#gUNq~m16~KIu!oL zVnXZ7p9Rd+Y+MLxj`7Z21ebHpCv9-7grT~5Q~C0Wa_&9Qp;;r8ah)q+J>cg?zi0r7 z-d>eqn&Ix40{$H8DE(MCj(lD5Zsg3}&{|&-+f8mH9>V=RY3pyk^yP z_+5HPJ@nARF%xZz&3i_v<1IuF9P(W*y(7Vvpab|7zOcn)fxdDn<7h*1WAj(tawtgj z2>XU_#+a6(Ue2QwTD(dlyW!=g z!ji!l!|pVmq{19Ci|N+5`cXFb>D3vjB?>vZa%b7h$WWYN-J9Kxcs~ZK-FWoqC(suk zwK=-CQ$)`BFiz>spwE0?q%t{Kub!~rqqK6LF#nXsIQDnV>NJsSOWF@n%jdwOL0wAU zY7*2-6w_dlbJ9$=_V&LCRlOx=zIeMAbvo4q4MoR`p5}FCbyf1kpf4w$J|vpPp{TLC z(9GO(%mp~}E*M@t2tHi3@k+!w20eoq0zKPnrx85d=C(<)F(m$aO zzI8M}f)zkl-9OcCw@MnvfSCZpLdRRVoGzdJ=pJJ_uDY3F6Mjh5PxTA;rii@r zGw5hS0=DZmxLN2vlKH`5BvxtFE2y(<4Cpdb4Z<5Sqi&Ri*u>uYHcIni#LC^m-#-5L zy%#V*d{I5KG^LkGt$$b;vA4c0LB_$2u#C)J@;5S*Q;5BtFKGTk&LZ$fT_81`hZghb zNmcpk5nNQzdugb|b=91-E55s*l}lWk7$B~{_kroYcG2j64)!q$qkJDOnQptW@OG*< zRD0SWr@h&GsF(y}mqUyJW77yM+ii%Uwqu*w^m;v3m{S<5#SlGGt$C12Dlg9AtpC|x2vN9RDDAn)MC&aM z30qWjj&L7pi*v5n z$7ROgT9dZO*w*)#sSGvoofd7Ol|kt86`*!_S1pFcZsX}LJ*?c>iuJT68}Iu+)8hS$ zqQODJ%48*6N_qKf@3Sr{*PRluR^?)js1A&JwM0@b7ei;zlHSC_xwHspSngOuEWDyw zCGN#46jV6bpGB>+1tg?0mAQ}&VBM$S!lpa}m)>Y?C%~r7?F7Nyhmufk(4VJEwgD%$r6oSXzDVW_;Q4IZ`(cAN`064q~H zrhBtV+RHK!4rJq4?MpaxliC7mZQ*g2wvgF$2NXR2h5}e+to4-F`)JUIo+G6t^Z@#- z^{!DLb-IQe)Li3tmz94_D}Zhrn}F`HIN4Sh^|GTaPSS~9%hn!=Y|J|$>SHTxxjP^P z?I%{4msBdI!N;TRw0K~7U+;Ah+UvafkNL=V-}u=-OiF=%arO@rq}#3nW`uhP^Pj{# z-K@|yQB3JJc{iOp@YQ8Xh7eP#U2t9XCN%sI40bhDu*>#@G!)B_u{t1sTc`t9 zEzFzxyD!_nXjBKbm!n-(m;GA8bP1yJS9-PdsoQ|1BziuOkpi2EQt_uoD~w3VIxtMm^PG$k#kc}L#j%7H0c2MRHcwHh+l2j_~OafevaIk0ChlX@7u(Q#htj}Y$Q?$~0 zYAjZN%LK!$=At-i%$i$XL~}6#7oIIt5f-CF_(oVkavTawD_$F3g2=f=BiAOv1u+W7 zi392G74@|fyp&~$VV>1LPNA^XsE_E&S4RM9^)B@M)K4)`KSVs-cit9M<+$XKiU{aS z_y}BpGMg@{G)$)BIx$s?B`-C$78S*bHE{bfR5FLsC`?_FhMubw>~A~{_zLEC-f^~b zd^~zRGs8k>ur|RRK7ZeDe`=&W8D9X~M^ThCPX4GOkjI&#d@z>cL=C@%j;V z?8%RWZ&?3e;Ps~N@sEh=wP`XFB+8ZYvBTW)g5{cg!s1Eqp9ZzXnGHoF2H&l$ON{RodaS>4^IA)| zW&vMfPBaryG-tgFT#w*1=pAi83JWTn!m(QWd$T*;Cx)x#-MbExCh#1UjE4?NRWmd? zJM`pgrkfe;`+8tgy=IhPs?6{Xpp>viZ+_N6Tfv^3npf@x>|ie5Fetx73J;}#Io)^d zP@z||!g=;Xa29fD2UlPde&<}emyR*s+O8_Fo}#FytCCe&yR;}cKrdy~q>KSXJ5mHD zHYERiq*+>-bDH&|L4m!M9kYiZ=7_zop+oTi28QilYBiDZsMARZs@+$Dg`4V1{0!)( z)UCkJgv$}bpIZzBby`<8YfqjbCoi;K-+SY424nSME<$Dn6=S8w%p^PXCG zV}hGOp&!NGrH+RRRPdOCsnI}N6e5^qQ5{eo=T!;KpWv!SAyQ2Y^tZD`T#L|{s(J_? zU~UdeqmL)a!NKd_XWT}rihmBOt0irwz@i{Y0Kv1$28;@!IUeD|-va4)5)C_b2qJ2q zmj-z{`Kgx5#mtk1FK^=c{2j`Hpj)~Q&NuFM#V_B(b%V&!Zc0_! zI~x*RzA#`?i}eBQ5?KpG^o8M0n;6yM(c$&kyHUD|77P7h2eB#+H(#cDU%dFctnpv{ zm_=y^4#N6#WNKHwaUAH<<5{^osxh3Z&)zUt<%=^%dwMQJQbU7VEkED6{rG*!m)n|$ z4|DG|eDC@hXMn|#91uMAABJ#&Hq{A%G7K*Z8RCZ<4)z^24lsW)4tc9!%SWrW%(mri zEAcVeH0+csPuZSY^MZKuoQU&82ip7mEG7U>yTwI7~2?T48M=Fea~~2 z=XAdR{+@q&F>{;dzV7RDt?%o7eg2RKdFLM1yOPZ+?_3-&VW-pAJwIKI*n0-PlHJYMB)hg2 z^>Wb(IZ?ARmew|_r71pxBwgR6 z_{NH6=_^C+HpStnFj?ELoGg2A%((?;3pNa_BI*QUtgFrG;`8%Y85zZMxfSoQFtP|6 zRCmm5JZj>rIRq1~R!}I}G|1mjee7m%i)OV=PN$VKpN5wav8=y?{HBI`5Ri0jimAY8 z6AoMWuKdiFDN`x1cM>462_?Raos{2SMAq zqmM=L8lyqYWUTmIQ4!#__20MtuYC=byJG)K%+P5b_>=&o{Jh9h+QOAeDfmM?`34v zog5`(aGtAT#Zmi~>+Fn08lnQ%5;=ko4pRx=e0xIN&h3WPi`5f?102G3hMGyRWO&rd z1SSHrS;M<#I_*T@lkobUx{gQAUJs_Vf5c`xl+?E6D1=ttE74FVVzi?|s^F9pFpF`Q zi=Qq(96LTGH>@?)OWWC7`Z&-+?1Z>&gfM2q*n>PPOzg2=UM7C(b3Or>kzs1k7O-x< z|5N_rqrm=?OYD&O^`x^+DCkRdK4(WZT5gI6s{q+ ze(i==!X>?#`6`cT^70JXN5X}+StX=0x4r{~4|lh1qr_C>mk}1?2sw>J5$$TaUFuXm z9=)<6Ps*#ppm8?^BgF?fW?c4Ei>}H(nP;2WegsORKSu2F;#fi!lD;W*4cX~;d#!&y zpYMwITGGpcjNdI#IUt7+TRu^Mf@8hXUzJq_%gQgV<=Qt}wV(1+OLKr5^y7@fu4Zbn zN7W=9gdxnGtGOyF1@ZLxaQCW*er>bIhSsLMd}dfCW!m3iATMw|nYV>;a2-5U#)BMA zHcq;E<=RKgoCWJ5%{^zE#JB$tU;nBiJDA^nci(0?^K9FW%h2J)@g~Ubw1o*pKpJ#G zzbf%JqMF}38+n@}(U+}rD*INrmI9+Z2KD8~h$g;i@;;v5(a+ol-j&JzGJuk z`Ay4<4GB^i@lvj>_b=M_l~$7j6ZqA-&AgZ9=j`eC5A+rzGAu(@Q7cAi7D-HLHZfu5 zZ#cf$J>**I%Ir+(9{$7GA&r{HV}dX%?4xC*hQ`BT2u~M<=i^hDN^xUu0oiusTa)MfSLTs+opVm^)*Sf)@LOtT5~2 z^p`hK+0d2di)%~c5!%*092Q)d)=lsJew5U_H69GvW-_@(!l$SQp zB*VVG0MEC;$}T;>;(Dn>l{sffA3sNn~0cF1YjyQR)dsUHC-emnx#4_733gmSEUPWa(chhpZ-`B^rz|(k4#NwS$-(PI-x?tHmgOeNBZVelTd-1&c!Io-LSY7EfMkjuBB! za;>PhP@-0>^r#Mw-tgwvcQ#fklyaGi%=E@yN?v-+-r3Y|U&Z$xWwpAoV>HKiP4}^T z#&nsTRd;@7@L|`A9c21qpu#IB@|^BYx-Ar;LNy?b$oDf7=_>5E+QA-QyUH;GWd5 zSa|4Xw)nZdr*@C~SYrf|Pb~-5F6Q=;){*Z%+JAbq6uHV)pQOq6n$}*5FBEXH-e-un zrp~K!6Ss%jS96|9kSeUGA|+tR$pIq(0zg-^_qw7{OQ3Rh50}bH&lARy(jtQ!snpvC z<@82ih!6#DqX$Fl%K49`_5&qkVt>8vrJEb##k+Li5l>a2Jn5xZ`i+B|pQQJtV0Iyy z@;)?(61_W*w_$?aKh*4YV-nQ&lg^H)cOE!^GPtdcbPE>4Qs0*#aRFjMxrzLRqq$sv z^7biF@|;Be#f?B?t%qV;qn~5@E*Rt~N;*xCNxBi&1>{WX}dt-F33iAF5GyN6r-+Bz*Kya-5t$tP*x6Z>&D? z;sB1N2ew;Mqlz*!Xd%#ybd9_aqm0xD=mIIQ&EWf;juXv#$VEqPtx&1d1(RMemw{bV z1E}Dmg#y#qSh4dLDx%>>UZD461nc_U>_}C)*bv6Y*jYBCkUR?qHQe3``y__343%(I z$BBCGC&%%!8T2fYH7XXJ#E6{s*V%37eqMl-qbt}99{tXGVPW(tD-NB<9hI9ah=EJ#x--b4*}h6Wte%lq>Z_~I+aY2?)@44 zKF%cD>_3rt$u}S4b>l7|@f6(L=JRAwN>9>JW&2?42%Vt*V*cvub4DAC?bWjnUs@!5Ol>G0Z4H?6WWf4mS3c^%ttslB?u zl5s1bm8AOWXoK0*C-(Gm=L@OBVTWIc?Jz-kaMwSlIT4T5+s*j?O85Dj_r83c*Z5!z zf9G$kD}OSrTlV$E$sHWmglYAojx*mJu&*nez2e%+Zu;)tm7*ap7=vjmFR~n8dSTdn?vU(fk@i<(a~e=cR-rAe>gXr3Jc?`j-uM;MY=u8BWd_xr=6kk4iTaaVT5xZ$NO zjgnW}3d7V&Z#VykfBq*=_uD^ixw}?o`uX82bF7GNbBtA+VO;e@bExlr!VO={uV3I- z3-toAV;9yGFGG0s({TL5r$E1sKM9XM^+)af&lCC6Pt1Y4o;_?Ya?JeGJAZzhe|5?B zeazBK+k%Y#_}%_w{C@q?zkkE$+ZllN6&tk8`IQ{&-+jzJc>vA-Ija4u@AeO`edgQ# zjF2~F2SfeCg}BfMe^%+kT?u5s4?**Wlg(fpOTe*fAI zW?Ecl(1H{^{2|6(Bg9@_s|Xf1xV>Q3&QOP21M?ZTIiL(LdesZ#w=$eM@aS z2PgFZ3r+`MIJfQEm$>)oSE2v$H2!ppfAcQLYJlM4=EQKXe)LcN$921SfyQpT>hhPU z?%!_l@1Eqta3HvEP9U(89KU#VpP5Cx_vY?=gTg+7LaD#y?#4%dyz=kN?C-7vLGHrC zdTE3PY&=UFM4}E1`}c>%idseMB<(~v_0hh*WVs1e1heNL5S$tyHNOBrmEoUXvfPfS zOCuMr*AQoR8K)h zo>v9}o@tnQ?>T3b?|l{CtHLFE<-p@Muqw(7Y%dKvuf?Y`?@4>ZB?5?8 z3Wy)t-=~qaG*v!5XHb%Si?x;8TnxMb%~NcQJP*VB5`;X?1BFyKKUo5`?ynxd4v_)jr1&Y6#15Aw}B&J znn>Y%4b7NMG2-Kn)KGvqP(fEC=>GAwL?c{QR?h4*)>p=rk5oQXSeJUVfB#}0-+D#) z)zRcRmcu(!tS6U;08d4d7!nBk=SG$8X9HWU(y7sfnxyEkfPlm zcvR>D=;p0R$QEi;K@y;n$nWkz)98ofz|L&2l5FmSSPF!J0aMqY9S> zuO`Z5`}?r2{TdUk5x4j2)h*bu2^oe2#fin(RT!wzt#jT0SiB@@3UuP0CmZCC!4j7RCHGL?X9bxvVf9J5Aac0zSG}{+vv;g zEokB6(#m|OjJw3BySavgE_P0~JI<^+;KUfQQ9|0PZd+JjpQ zxLeiJ?@?h`>C+eE#9hh#f*3vR1%CCIhl&KK-XW;xTH5*oAt`b1E~j|eu6Y|`PfA_; zG;yn)_AyD`?Gxd7EB$=OZ=0|F=_LIm=%)s}^;H~Y)?jpcIi6Q}aV&8hZu9}pzYmo> zTL$Q{WCnmyP~O?sm!+L~is8taCqZ<}VR?2NZOsxTf@IzIFRxL=ETDjNA@+B9kiCy}!vdcgKSMNy9kf zs8*@RVQ0*ZlqLo}R^EaoJJHVc+dFCcRfe0B?uBW)vpMvtO}L_2o~T2W-a&j=6mrEVAx)42rRv6$ca^ojU1GkKK#ZPn_3%_msiR`U^<{ zMrh;}eCstltiXQ7CA*UR^+XPRrxE)GosuF1gpRVx@*rTNu-VXZCly)ltGBM|<|sHo zZSfoQ#RfuBocdH}TDVYxq)Nk)vu8{`J)h9m$gVdi)!NcHs~-J4eS`K>%4(M+Y_tv{-a=+_ zj0)1(OI-#fS@KtkaLv61>LrkNp}Uo(*0;v15+z|vj>~an%y)qW9@r!npxrYI*MOPfYf9@a*x{inLXT2`?-dgLX zMWBUWL6Ti} zxOykY$CEL^x|4P<+w$&|t&uN89>?p!(*zNGN3*iPCPqX;o>Y-LH98HWLn6l!|#-(|{nKD1-fo6ethwcja)%+2+p;y@@^Cld}g zEKnNZXfED4(W;av>8#q5ZDd_;rFYVqv=sMbso%bd4Y+?$ig%cs)@uzHjMD4L8sJE& z#uYypQN%iw>qj)QoVmq36C#PVFeo|HWAAD?PKq*iQ>dm*oM?y=M-R?=w1i8Q=waNK z(lgJ^^|=KL>YEtH#ZHN?T6thAT&Et2J)TD-YP%0X;aO(4li~!lvnLlMEVUDUVrMd8 z-FHk@&hhh%{x_;K!{Cj-1SS4`9PZP$d+zqUq2P1xE1sn>Eho?zA-w`~MuhnYXR>zj z0oDtaE3ucYA-PqA<~Ra_d;dxD1K)M0csXz5tys9zy!q1ULh34*%v*nb2u>+yJkk6B zrIy3mLoWjMP36^C1KoC|AH6gHSTfe7?cyz(!LYDQ)8V|_jx4tvbtTgId#UsATPQ~~ z>p&4ozo?N=I@Ti!I$|9JF~(tw=8zg2DX;I(v6p{392>jhM_^n^lLJXd>HqFXtR8zo zsF3>ZL7Bx1CzQKHSmSb30w-G1RNmh>?D_$~gd+OZ|UY{EmcCCJ36EtC6PN4}Q8sv5DMWcT_{QzZXTe!I6_i#KClYi6+O_cwzZfmLqF$5;`7JA1331l8K^LyMdtNX2GMEg9t2|P009a-K{u2^$&HbAW6Yr4)@+P@<)&(468kR zMQ_y3NMzrED+tb}B&CEvNJG!7 zUJeaPx9s|=X$;_d?M*eYMI{=iJ~3$slPE~N9o2gO450LrBkjSS9@f97tKee0H+y<4 zxroJ8AP^oLN5w(;E}`iS*-t!bspK@>_7uJ+iS1co+;U!?Fi*{S=^Ycc*pt8jPOO3; zlMQu{M_!pBpDX5LA5T^tr{5XnmICD(M_nMO#sM`TfhdFJiwJ@#;b;u9_WI*YWy zyyg`sqYb4HZ=<*U7*;F0IH~bha70*ysc{zZRv+1ksow)}V zXeL!yS?c@Z`#5x2YY~2}&i%2@#N~090)Y>BYe9qb*AmbD^Vv3+g7LL2MOh_CB)ZtW z=~$-`Ij_#sztwp(w#b?^$FY5>8h$DCrojD6NYGJ#TN}V6C0-U8B?O(jTugCbeDGgY zxKmK_&-Y-99NmIl@ev>%yR~A|J67vOS9?JNLF#up0q~QCK1)ens6*QkG;O+qqaC@|1du1a zm4Jj7wpmD}xyx<$B#j=ePl8J7iS>!u44eaysw0f+#LLUE8?j>Qk}gvZ70Dd=Jge&0zB-Q4=`Joi+55B2gm)H1eqQOV4PG)y8bVl5i1#qxG47}LX)we;l^59@`NUJd1`zN)N(3wMx~1F zg+Y{CCc}@W7ZrfAP>j_ahZw0XGn`!~|2hLZwb!DA!Z1$SqbY5&gCzHc{z0U!MMF@Qqn51N;7QtPM!utZ1&>0r^$Prnl^l}zJXl7&1RUZx-h7@Nb*J3bZ@KM|+sc51 z1Y-Jas(O@07@x|KzWE3JMW5DueD|t9mGRLolPO^J!ab4xsmIRi`4dt_YAm<<=3n!( z{jHz6UIaQ=AF6l8|3X5+1A2LgcWr-ceE+XUgxPF|!d(}0jo1I{RexI`Vm`gq$9mEK z`OJUUKl}hd9^B3=@qdBq{i}%M53jvAyw$z@Z2K_mht&L!cOl=k)%TTzxBhoG|9Kzi z0ABtsc3I`$l`wy?;Jnj7bTJ;K`%l}kg#Yi5p1*r~DH=d@UA-{#zZT8k{n@Wta|3`p zf)AIxRQYc#+_oKe0gaXP^o+!n|4wWEzdr6hZWQZC98*H%?GsG(MOFOc6TP=jm26Z% zGbW^eR;BCJ@^?62#+}MLQy+Tc2wRK5!j+6y>oCVX|L$4Og?~gaBqk-~DF~qlU6X`f zcK)n@%)58=?9>+5guZQ})=|Yt$zi=`l~E7xUOKD+PQnQ5HK)}V01l_olRe10Rhnab1(xn;y|uLL6>b?$CX)5iTs975?aa_ zwB--k5a!D}LiHC61rw|%4NoL5y(lx`{3)B^r5^aUInleh=f)=c&4GhD*@hQ+ysmyG zY`-0Oj?Y_TqjyU>re55_Nlf$}f;r=^o4Db%VNL&rwF7L@OJ0Q8;!tjB+|!NJ25*m6 zy==>or_Tt!a$J0b)&vSCf8lOMXW4rrChi+Ckb7vmmNx&JakJufLvLLrS3>Zd=B*81 zf_b2geoG1rgpCWQnLm7dj`xase8rqLKyRM9(osc}k}X*yuFsF2Oc{Yew?>9|qKDT$ zw2vA_mAaC=V#K{e@1s|ukx8sSq~H4tc{`pFwhmB}HKj}-p{^-JF-*IWpTRLb;@ZOZ zJ+js-KbutEScY}i(R9Opm}0B&&`I$(iJ}F8V?3-c&37bgTFjO+=EHE9FmcVCw!H?Y zyGSJ}uD)F+eJ>R#M#G?QlE0c{HlbO-$C&DkF%G-7phQ7%kI0PhK35V?cGWAKHRWKDA$HnbJ1}fc~y6aTO4!jjK(ASq6{N$X%!mrVoSVem@v<~)r ze^kH~1LbVeH1&!ArCTTR&5{T@&0g-zV=NNv2k(IwBvdsD?r6X9kp{odGt6K;KS0wt z%{@lCOHcE&$E<^$>$0XY(qZ?%e+ewy)DiEi6p!BbBW{plz}vBbTaB|2F3UNq&{~+| z4Efm`yWQEnDo&b9P*0LgFvPeRX(Y(nN-}ZDp1K{G6Qau6gy$MFo!V-)5xUrOE-D8e zjdUi8ms53Ldinqhq5@RZGjS5&^cP3!l*R&h>Pm~fX@m8aJkoNom3Zq`R-W?&eT&^b zlc(-{c;K})N@F1jrfS1LQBAeujEXd(`WOUdY{AW(S2Es^p!J4J?j!*V&8^_B$AOGR zHU#%Y`A0gv%v`%pU33#F*?<}GbhjFDX8+`Dq(AOhk8`C=3{^TPWoaYs^XW&l&Zc*J zC>=<|MvH2KDV7s&x`xd_M`Bj#IgMPZolI*>u610kzpD}<;`gXEUPZ*B@r*uQJj7+O z>+u=|U**`dUzkncEnX7xRsIl*4G*Iu&o#GOhjg_++YgqTO~hQe>7w%XoJnq8d$JOt zyF^Dvufp#A7P|XNnm1)eln6+B=a<*?OQVR|G(cVNb_fYmc|2%ftIzJ;IdL(a1D^8y zWvb_z$dRM`w^|dVNbKIk33Y6~kYNFL$Ti2Y=nPVFz!4Py;Fci{9iTq>tU#1W^In$B z*Ck~FvwwSWtf4Z1YA9#7`)LB}y0aYWTeiB=mpC!|2-7+ZUEEtk0UeLe%S5+9Y_Ev9 z8cDCpX)51YSS*@y2dwx>y`Fc<3$1{!jVrNt zVf?qvIYrDGc^5|OD9ugyXC(?*_s@e)R|Sb|TK6WfI!ef}^b9aKc@r^ug;ZC!xNZr*505X_jY?AD8`CT&vT>4i2! z-?C}6CZPLS(hwD|4YL%?$jKFauB>ptAwB;d(IhGFAwz56X=Rcmpi)Fy`$n@-nIokj z%d|?)z9Ec1;>=5yxY|kb6rtaaDw7~II2aJ^Iy-b46bQvI>B*S;?~-I}hiH$XU_Ri} za$>u;{UOe4kA_aa%%;4PeF78SyfIk|0o>LlIZLrI(1{B%iikV%gxfz-p5!e*aNT%b zx*RI8^vmTwx7bRp{N}QB;DC=C_i=15iM;&W2BH4&lf9N9SVi$Mb&o@B=hcJRHBEQl zdoPsP>kHOz+(()(4qQQf;yi9t{4nAmeNk%5RWZo)Ew!7<^ix$HZ(?v~&z)O0SqX|O zj?NkZ1@op|th1xjNdHh`cjkEjFM54c81k*`y#57i=G8Pd{fcJ<$voSB*H@Tf_v?oC zSXrLNEuVXE8i%|HWjb*oa8r*jp@I?$CMe;>>IotmJ@(%&40==(rk0G4Uv_*w$x|Vk z4ZT^t{xGC!+P%B66OfQcG+;EHcjKL(>fg_~uaF2Yj_AtKlllD7deHx(k%B8-P1=4o zqkDAkp|ww15&AplrNq!z33%_}r_%aGi55cTxz3{^ESB^|B5AzI-s{qKGx*az$O>mW z`>ML6!iwfvHXg>+`|0ZoB?GzK2OWyP1|?o%I+CiA(Adjmv$i~SWwGgbgGlTC%cKFa zeeSIO%L?U!8;k2OO_d?(vS+p>Qk9JU-sH?Kl-3sdL5PQO{r*I=(Iho~$^rU#sj?8h zr2UdjJO^SB9PLe>}Nt-#{Ekj$L>(%S4T#who zVIdhgRWkj{^isQSMV9@Wc&WihQ~CW#d%4=)8|-zbN3bF0G7?q@wFl=Ha5EAIn#ohH z<+3xQHq`m+qyFSElz88ZY8^#u7~$eulGmzAu}!~;%vDF}Z}T@V3ML?Bx5@vIKBw=0 zF2~NQg?{V)OP>7^!PYi!S|ku$u9(6mY!rscz%EL$F^ab)VZId^)e>=4iupO)jRk3_ zS0jwE(q{o_2AN$FH^Q%x7$)q`7&R?=q9I#isp<{0r}j?2i~3dPIE2ov7K1~w7DBxX-U~X-Ss(IQLCCD%%+$j8WpbcSJ|d-)0(S>S)^Ao(<(6&=Mg?SC zG_1VtfuV%*J8?%M$lC*32dVXd`}6z$1Y$MtK_PRbYLffBX8UtlRd z4@lS;ZrAz@Xe{;_WE)dvrWv^}$gJK1N<9evw~Ecd(^~_&!7|2BfO28$7*QjyQ-hL3 z6s(}hut7Z@)yaPQr<2!;-h6Orp>m_v+_Z5ckr#|io_)*FsCz|~2wG;_&YgKj$Plkd zHO_$`L)V5D4kZvXuYQp69+6vA?O*93hnc<7CL8#Af+F>) zGU@dvkK0ZC(jGrue(8@=>mJK}E$2P*tg>t3BFo&0xUnE3#eLH>Y2-RVmSbhz^IJ)4 z4s#SannQOvr;upPFWNpoTIajImlrYXDo3@GoQpzKUsG_HQN~(7GDh!_z)!8`#7oZR zG|ot@$p~5U8rJrhl<9b6O-gm5cMSN%u2W38Pt-X9G zGqXbW<9y!3=BNvmF^jdmxhdeYT4I+@>j|tYW+-ex1wFzlo zd72t?F!OOsTSnrqTmWQh$g_`Di?3Pdo%f)XB!>WJ!&pNGbmhcY59uv6NMBET-SxTC zajY2QJ8|o^%1j~BRAsP%|s*V$$;g7s5WnUe6a zC9Xv6jt-K%QShkD!YeT9!jB0AsZOW7}tevmTw`g&?S2)Bvlykd99Pz!~ajH)#< z8QCHN5YScFu+kQkat-P*gR;K6V$nDmtMv$q-cvOzFtnANf9gS~LUHrBFZ#?29;imrp|X6+yOT;S zBhy_T10I`^k+-9S1O5F)*KJBVGqCkd=)$F48GzdGyqRjmrKIzWeneVN?=ssq5JdBv zsYX3LtQE7S*Ps}x z`E!OM{@bx#oZ^~uhK^&CcbD{Jp3H~9KD_3ZGETImrydoE#&>%!PhOttLX}Lq1{Y{& z=?=wOC^y83moP+cO53G0O#;IH^pus)P%o!BbJ?a@sg+AUo&lIt+c#-A8Lo}Le$-fs_(tbA3r%0D}(`5DrIDWyO*p~-=&V8QD-1aDN!feD#NK6R^hIK49(YPQzco!B|KxQ$p7 zJ!jMxCrhxf`&^%w@6dS+`%StGrV3hFE7?&3cbf}ls-Ee}x?I3LB-Tc4&NeEkDT?#bxoUSuQ#DFJNTCNK4$!v|m*s+8>HGa4{a__mwmZ6C zib!vprnaUh#r0?w<#HeUz^mZ!>``mw?w&lp`f~CKnc5p0%LSE--q_nc7h+@(7ykJm z?c-%;+3H0+ulUUQb5;38%?^&~94(r|6mS3pfkOL^F*&0erNL5Dg54Gmg+;tjSJ955 zK2Pm#u@ju$kLR^^`nEzUbq}x6OD-P~9va*%$k1-$OF5}%vAaJIAaP06bhFQ{CBBs_ zm#evDPwDQSScI(3-?TQ&f21;N(y`%s z>3&J0yeXc~$`f#lG@Hz$c1&u#v zbt&nuPx`92oy>QyQrD!S32?E>sZZZuTnIoI){!N?wP$W8paU2qsbnj-2i~UtMqhPz z`>wWG4Jl5knI)f&4+=&V216e2gO(?|HJuV$i_M5)I`f7+LZzPLuPzrW4U#GcA$=wA z!2+;&z`{srh@+>tf8_K{WuD=-y@HBXn{&N_XJ4^JkgC_z;+fK%h>q?19bDf6dIJv^ z98Co4;Z`fjj;_lg?(wNJjpWRj4DcpB&Vy8=!UrSY;TzdQmCiIa-sPHo7W1AI4Ezz` zq)EYZwWfF1C19A7lL7(F&el-~?nG}s&?$1GYs4-S&qdt^tRJbN-snj-o!JlSZMgJwGO)jW(9=Tkn~>13zJnn>3Ba9d5RBUF&u3e@!xX>;n$HF zQgPL|uNzkl7xkFVep8n_y;Vz0hxmD0oLvHGiNkIzn~+mLYo~tlI`r%M18$eRU@~9| zx>w@D=w7M8vh+at*)E2yd~w2r_1O;s3@;6k0K)(5C+ebH$c0Th1i6-Q&8xD1OZZ7cMgC zae*OGGUIS9<_W-X$ltjWQi(6AI4a2D_))h`$f)>UJZt=@K?Ew`;0yr-de>|l!ubn6 z)mPc02i6RbWUE_pS#kd43bH4b9A{1WW-jXC7+#DmTSDX#y!vyh|4WaEZdZOs}g=DbUU3EorPb~MGV^dy+}~U2YYMAP0fTkn%&-eAc44T z$-PLinc7xHL<{#36ZAL= z`o4|wwd#4Rvk3KLcu6h4uPl)Ktcsjr@!Lt6qPS|gV6YW=Kbc3yYhcL6n2ZnD#V}W$ zF%zVJ7cnqNdPMDF$VYTKt`RBFPs@VaRyKa9?ageo)p(4Os*(PsiY33>Zjg#KjDaAX z#Zd1my?mv{GLz*kkY?Ib>Tk~_Pvr-X4$K+$ylQOH$#HRMxT0CY%-Bq|;%J?<_2WyJ z^=A#+B$6s>x>5ZuyAE(!GvBQp-snSaQYta(iC&}!=4*~-o_ly^xl}<>yUcMSSEq$I zQ+B|R7+RDMXtUbqvzM?skT1g;M)2E39E@f&r3^J1^mNv3uEjKmDvoD}C9+s+=Z*S& z%7EydLBgPrS1_t(0mKt9Tb$`NBwvNtkX}37FwIdJ8mBlT2(5mNoz+*7(CesRU30z@ z({UE@1~Hj@Jxsk9YS2&oAFyaQY~VbBF8zlUqvJCa+8*`=VsHw0x6e#6 zXS6mG4}Z4@e{2>T%`yQsV^rY8#qV~_)P)0pqkFb0kuii{f$ha~( zWvIoR`rK(*MDdQiKzZHHOv&uDI0GD8qAY1Q=}r;F#ha)nVzF4#kj0!RpM9`2BTXZr z0|YH0S05iCLG|Wwa2?eQ6>@7R#s&SYGT)s^#0jT zet|0~?%jKpEBi;gv)o+EdWs|~FvTq%*S^wxe0J_V^b&KYn!Aq1#!~ofwG2j3^3?r; zUqg_^?R)N|w?xnOIst|cy?h(L`#DZDxCOOYBGI<|op$*N>AQ8^Y#pyCG#MrTWf9Y! zfe;sl;@SC;_cheHqSI~IfThZ<(U`%_IR_UNMl>(R`h5`?RpD^o=!F1KnKPZ$2ZY{3 zAWE*n)9RmNkvhl=pe!14uSe;ugkves%#+!VKlFGn-vUkOgXMZ^KM(u+L|=H9+&>Mm z>}P)%FAA@Anrmw{>=|$mG>nQ8GSYX?c08*Cy(I`{0+@;=phE*S)OapnNH0gcW#TD2 z@mAVndCoy2{@N6$U8K@}>h#mX$M+0;@XsB|Srv=XZSZjBl$yv{PY{1gc{`UVt~pOhfM5@ucjKO1@0K|g z>-teL)pXi-AI2b}7-#$K45-aAE>v-MI8>JSnKKI;*KeW&DHf-kv+k#K5A3w{z*ZVm ztQb{QC>qIGA;kTb1sRgLKVE)!gDj8&_^p>8Yofh^`?heY$2_Rc$rGd6wa#`V?DS9- zZM?gfRQiU!?-d2H&(>47n4e+3W*;GsEbCU%d{Vk}M5YH<4!MmJSYz>`4HoHnpZ^Z3 z_tA3Y3Sje4H&1xEv(+tU!I>OP>0fiw{sHj8%eQ}kMvAEWj5KfkC1qqjyF;I`zgNt@ zm#>cWyn^ilwXRGh_Ns^C_pgk!&CKsck9CsL-`>xzX>(x)nx0^vdE#Kyg6;2El!ksu_dpr>_o7X;G z(U2HSpC(>$Lz526ti0P!x*f1$|NJ-~j`DH!rPf%Qn7s=92I^5$PX#911RJdLBdx`U z&*baPcA9=-y>#=J%U@jHA=xEQ;9A-Vx@w3A9-1TBG%M146|%kJ-MHPY1H5R!r%Us z0yuTtkA8<){e1DmxTla>I~#2x-JuJccpJKo6={F+{z|G&uDtB|d!_Fz)O3%D=$^DB zr7JXpQ3W?wc#r7%EZ3RVkdZkDPwm?7g_A9_&$VmlR5CFUwyNLuvG_{7VX~**6doOg zPQE*o;j;?FKhP=tu!HOY{m?qN#@$Z2Z)qFAx@*M_uqaaIUfm@sVW>j7lZJ9l7h}aE zL5D6tcXrLo=);hpMTOf`7^ke`ebBXBIf2L*yzIqn?r0V&dbq_>D3$_5+zGGa8_I5n4%^mw;58E_TN7&JkUqxp z>95(xzzW^V-plx>>TxBp+o(L=8d2kdgAtQhXelqr`_ipyZn0}^oq0c0G`f3=`RG}T z@zh_9h4t=2eRRP^Xr5~1Mdi?B-=XL&SLc62$%^Beb8%PZ zJwJVSyo9Wzkg#6)CsEWM&92DgB*qX>CfRc!0qlcR5;CluGYXufG=3Y73w7-8NrBp zdQjqgc7XFq>&WJN<@XtI^LO9ZHGsCXA!4NwWl)j|-Beh~a_@%8-7SgC&_wRHXqrAG zpb@*65f4TPuY84@tWw@rRy;%DN)#(vi$T1nSK{@)ps6T(WYr#X;yeIh3(;1%p4bL5 z2wq17{&A||H*>-%mMk(jK+`Z&@>%p>o z^VvN(;p5jzlqVdw>O~6AcI9X-Z?0-#V7(=q@4#%XPC15F-DUml-Xhz13t5Ym?SK?e z;LEFH#UKo)$h~$!!O_{E7{4qfuk?i)xIeUeSxUripkfm}kn*MJ5dd2xc8<4%-(Ic| zt3+v_rgH?Qz(k58UC|FcN?*e{0oGf*TK(qNmbhyFUBsnBYkN;Ht8g(+vW{wM36>8PFd-LyzqaLT%q^lY{VePVKhF3Wial>}G5`hkReyl;?J zG4__#3TN8(x7Ou%Z@f_1;o7>_ zK7m%5PAZ~~TecPf^_EhEU85Q91NAxj{<5`vSi*b&Cu2cuJ=c<&8 zyG}grJIAnYCZG>(7x?DVId4)YR5iBxGTbGKNDjwh!YH!G&!I8B3IdSK)5^ECw~S}S z*Xx7`aE>AbwD+wf9$#8=13k#QmmCqAPBVQ{y539M1XlM--8$+DX3IZ`?(`-tf!Pn$ zU}L(WDG8?vDoSaPAgi3tDe1k8?95oTE0Bt+MJZs>qa1 zSzE9Qc7KCB0N_IA+RyQ0(*-)mE*cBi7ZT4Tl&^&8lVeqCQpUGX3E$#m(3x6zdUa2alw-s5wB?)!7!_c!m~e}_8GdA+Xd^}L?rF@PBF90qR)BZj;3E=A_7 z1MJGD{ciZzH>{yN`lW4UnBY$-@}r0QGZJhVbIu_bWNw?KIV2w`gl@8+Y!ga%YWb;=>Tp*VLqL`CW3F!#i(_>Z084iJ(EHsRFDW}BZyK)bEsj#f zsZO&ttt6T(zKbuw>znZu7F%SzcGa>>(EuXng~e}$I9`3xD!pb7QF_AtsNc-u*#%0| z3jsuFBo-gH6z^%_5_nMxwdXSG)Io^DN?_r9J$vih37F=0c<l^KcrG-+GI{*EaGNr&mz}Zxa83Om=H4(G?B{0=cc<-|J7Ir@77N z*YNl8JysSSUzHmIU5XLb;YuB6NO`o^Mg~hVimhD2EW_8bV-2TG@hq9S4R@%19_D6Z|I=AzN=gxJ44-WaJQYd}8#;G^Blaba|O=-Nr2Ag)o-M|VvL5tjRiPf6m zMP+2z_Ao_Ux=-*L@^&%xCebcUDPZV01&gQS3tdb1r5(A?h_5lbu>!!o8eZw==Z8aH zFB|a)z7+K4ISI@2_~y1!)~Iu1iO}Qt-YBQJ9(z<(c#@0pUOScb9$f4)zL0!p-2-%aENAxS_Pc(t82pmSHdjZ=q1); z<*@o@2`%+bjdw~F1DtKHPxwJ4X;+8^H>#W`h|Fy$lI$lPCM7i{%RX9~_~a!FEZ(?i z*XEZJ8vn0F%fA^uM9aOv;NXX)GijKFV{Z8NOp_m_Oua7~3#Cf*u{&ZJ^aByU$5-xj zI5KiEZO;ffru8n~wt+#K_>Nh?>~-d`;V~`hOZcA1N2kpK1ha~?7$W0suLyw6ZbN1sCZ4ntZaP z1U(vZ^lmWe6k2I1jO+a!Wb)@Eb;#4>Yh`9CWDl>m$iz248lV44ZpiB|lb+@7z1(?O zQ+1{M(={W5!kD+|(Ip_sZlEkg+YgLyGJwd$B^=wiQ8QYp*b!Gd!Gh#6eBlRp1s~_` zJX5f={9@p=-nT>z67Va88B`gGF&#}gb@y12V~KjjV=4}ale^ReMos%M$F z?Q4k&hf3^-*hZ#z@LQduyI|sN7Lj=^mv)8S=3CJDFwd$nt1LUWeSsGC#SQ^dG|^ko zYx}DyZJXea*yaILZE<3_#riX>tmQyb9yw>I3~IaUDL8kWKF*21txD>#?FjQ|X2H&djcR1JfFQDK!mF4!-abMRqVB?% zX0ldlt5;6K7ItM#)9~X9tyTy975QQFr`|}oH(Q9o&)nT(#yyAVW@;5wt0S?#Qo(DX zz{{^WJAIW1m~N6)JG)FcLAE;aZJB1Snfez`q&vb_W_BYGk8w^EWQlcggb#jw!tbk| zi<7!EPaFrtOYoZ>cy>`1Ab7t@d-mSCJI|K8qNa_fMe6I+ox{ifS5EsM%j;&gBTrN; z=l77(l65DkJxwFf?>G|d7D^IB<1Y{v7SG384>3CQhm>kkNdO2QC!3Di(!!Ww-uelN z46_b@(}sq|={7`dC~|k#C%lGba}4_Oit4%3CkJXN0T*{WQ)JPDk1-V$FJ88I@?;pG zp;w}_bV{7mxN6{E$L9TSQ*i=~eED!IlNCKG2h z*w~SuZqVPDZVOi>_&wUbHL`&dp9Z*9jT#SnLXzrGtrL~>6`BeBK!Nhc(#f#dl&+bH zymPL(96vV4pPszEP-8JelC);g=)%A7Mm#lDAIf(bz^G&=Sv8 zMi!RgPG306bKW#4C@9Q!hLpD_#4K{gina8*xfGs1O*wrI(`7odhn+*Z=$%{$++V&J~C@}`j9A5M>An*)>efNo!IPQl{5 zdZ%9^!=Lw)3CMHb9sK&nu1==_oZcMl?Dwx+y{g)5o?Q*R@QTalI4^GP5Hfdfmf{CY zu%K8?usktn^(YXYeOU~S-Oft}eka7QYAL=Zxt3$4E>Jql=4rJWAa|naV*fJ7phRh3 z-81xt;o26tYY;AZ1gM}>?&Ca! zj3B&O1`?^%6KD=}0W2ZD(>K#=7x(EpKlJ-40{7JLz=EZN*Nht`Z9eAC zvyM-B>4n7LB8%ZpVz{os$1Fk*|6{*Z)286EjELKP1Su9lB{8@8DKHf6XoI37ChV&( zyN$TUGFN$`Sh}du(FMel@fO9T=2K9W+$U0WwKZAKO)nBA0_~Rf^a*a?A>7tkb5ORh zfVyvNAsBxds7~fnd$~p|BfGj40U+}(iJkwyBGUgo#O=~GchaNuOKh%h#=da>a7lDm zzhJO6Q5q@HgGAsw`>4$TW0l5MW_(wUscZxv`S$%n{uud9}M9o;Pdm zmE*maePkQcD#tbHT!+kUrAG?>^0MEOf*6o^eHaquZWz)QC zJo@=3s4^=8kb+Bh?)EW@yA4^b?+Sj6*U~R=O#9l@q>sqb&fY%NSBpd=BKLL9Zfnlf z9rLo#)Z3RZ zW1ETwI^F+$rT+K7jz2x$+*wbc2HDU&0{OkVD_r+2C8}(fgowNC`%E$-H2v~0lS=1E zGz6uJVz_wmx=+p%Q@*a1nOHiSH(Um?NUnsCDwSJCjcq3xc#AK4p>ErsX64;iTPw}g zZ+l^pu(Ie_Cb&4VXM^eC%)ulZikGPi3+!aqliwuz@SY`Ydahdmr-U(-6uHDIrt>_;M7f^4E8-KYV}p5qkfDEF=Tc<|MGwX<+{A#%c)4a=ua zrQH^aVM&y7TPZ>2rFWCyAxv{h<=(#NV`Zj)-rZe5@F%MH%ywRena@7#A}I{qh+f_) zQkUcahOp`b;fuAD6UgluCh9lZhTNLcU5La>0+*+pY=N)w8Je3|QGFMEZrg?0+Ey<2 zW|{2go44xmazW>Z0bXgm@ALhYpRN}f-)@|`B@R{oAsFW*fr0janeY~#>4?P+P8Z?U zp!6hQ*)!FLv?4qM$=*r3SqX5cfS78Yy}saf_bA5A4Zo!0vneL}LVVlUw?{2rV$Oz> zlxIR@p_Ld9=s5Y?H8y}N5C3UFF0&iGaTMbH!HBTeejGANa8IA$UJmP zm>P>r|zN5jZoP*AYpowzux zn`bpc?nh*%el=t>j{8&x5szqqf&;0+{2PS#5uKFtQd+7=vco`gO~OcTFl{P^gnIZGe52O%^)^_KrW%@)@=pwHk(LFjN~x4 zY@kvqv0XXx1_wYIMdbD_0r*Z_rIgvd-64k9L%XZ=+>5SBdjqwmfnJiZ!uZm6nF+w# z@7r&N_f&oKDtkZlmGtLoHZ6E&({f2b8`icr^O9l4^_eq?y;BgC81}>uOcsI8b0be! zV6(}o0X=~-b;shCE7WGSqyX6dmJ#Q@f0-Wt*B{IMkMWy-Bg~pMnGGx;hHsSygfbbz z%ma3t1^jbzL*%s=Kk=3u`yN^9RG^dK)5(dd^2uW=QB8A)bAWyF#ki_$hY(D2LA-t~ zwL=sQ#6MRRGC)4gT~JZ%;RaMXrhuMm=@E}cN}6*_G4qo^=P)~J&!zzl#A8ANS={Sr zro_WUeb?^2qJE>3=BDTFUY~(aZII-D$!F%3f%&Y!Q4-RA5`B;!HBJg0##DEnho43&Yr!zuW%w2 zED^e6Xf%En^o7f~MPRFDS-R)w>8BIIBt-^B)$s2@xiqBo6R3@qM2bEC>HqcX*Vj7t zkRdoQTL3^OqbJ4g1vK#sdIHtVAhm24bZ05>LGBbhdoOS+ye2+9c`*%Gzq{#|*<1&T zZ!VQ)>v&|<@cS@#wtJ{d^Lu;JF7Fu9Jv9UoOQTNf28~3M7F&%DL9bUz+3rif8qrEl^+h^hEuD)ci$_OvJ&FOFt~pyaK!NButszOK z3zp=h1u~D9_>*^^Nbg!gIDYuCFe=zv-rM@2QbLeO=Z9<9es{yPKz#OW1I>WXv$i%m zEGAOwSoe*j3BvdF`LuQCj$h&R{NtG$4`pc9swV(8ee)-)9(LV4gVh-*LUFD;?asGA z-Bg9MRUOOA4qdGiqE0g^E0uyL@^}Wc#L@Y1wpblnp9Nf*?U%>0V+lO+A1-O6IhwbC zWOt|1CfWiG8|<%Jt&4<|(l^l=<-4!^)WG=5mt29Z4KBMj+{RWS=vzu+i06S?@J|lq z)WpnQoK{{UsKDZD`^_U2D+Cs@m(trHXW19u#%!m3kEL zs?5)Mm@aubE<489 zd|ma`Y>Q^b(2c=1ot%(`yo&G8uYlHVPKe;ycpR3U7a9_BrYhIOA$Kh|*EhP6YGvr? zb#mjek95#RBOk+=_IPNOpv-9Nv0RX)Ybr&z_R9q^l%QjYosw1_zxLFpzZX+@bfjBl5-|%c`ouu8E=QwKQUhj?h9)-Na zcj`jOh=4b#q+z#*OlqSGmga?`bBp+LWssn=ha0h`Gr33Li-r=R+iotMb-snui9Tka z|JM@mo@pBQ@#=JRM#8Faoaui#E&l_T=;C8G&0{Dav~m?6rPQMLYMWkD@QLien-ZX| zrfg65Y25}NlQqWJDH$o`y5p`-9E_ff0Wyrb!7Lqiq4@WRYK_&ImaAQybrOBYW)U^p z)(DwGo%=rXgBG^0%S`+zS2tqqhdMV?5lDs_R$J9rft8AYfY_36m!y|^u$M&6T%5D< zs7VxaY1LWn_hWQr-EQUf=6=)^Lyf~$S#db!orrjhN!m}TXN00d+ zAbfi9N?Ij0|3LLRjy#F)JkmM>pV)^ z+&6<)1n%|b%W=#C?&wz65U}T@k>3qe{R@@o?mR;)v$1Bl!}I?q{($@Xd2w+0Wke6np@w+ zm)ib3RyG`aCO@wWLTbzbU}r6Lx29vpva+p1z+N)Wa@If=1rz~FCeI<}?5a?6v>90y zI$6>{Nz-!KV7MgR8oo4A0iUT|eak)4x`qVOdp&sAcoI=9P1kd){y46eF?4S^C0GM~ zQPLjt<-AcnXGoy?T>*dd{-mfr*)Y2w@Vo%hY*F;;7_5o1Ym{o>pKs3pt{eXmMF06! z^YW34V(!4%&>GgF^|Sws&3@g!pfaG_2v|oenWyBSGKNRzI@*Y-=fs!RGSGBY7QKb%*)m8JVksUC+-lPbrW zdeeJl_kpE_3Pxu%16-)i)=gq}Apdy^Xic?FDi{j^<%RPMoV!wGptc=8$sF`%Wf*}8 zd;{^0o_ZQu^&phX)7e_vDc3HKejsvpH3+a@IyOq7J>N}xePZ>V>}7p_k2_pKOG~?l z=rLUg7KEmnM)G#IR6ipx`^l>t3oZ0#PUe}uBA)UgkvD%+-)Tc2m0~w=(=3{!UMqu1 z@P;r+?g`y&4Gxnsl(#qCJytRId^1$Ss5b&~LFD0SX0fR|ACCTwn4rlfdmNR4ikb*O zVBmZ3iZLy++G68=siA<;p9M7fLV-XWQ>xFf z7JSL%<;)%Ao>4z}J#xpiVG=}G`rLE5_0}}z&;`eZ-0pO3ZVQ+ZKKI9SE{WaGnNd=v zGb&Vf?qZvp6}psGHlyo2>R(cfQja8XEr?n7e=(tjZ>=u>EjdjrW5gSNxbQYs>}WQk5$tyIY)AYv8Ej&ZE9#y(GyMf9A6w(K~fXpqY{V zIkn;6Sg`wS`jk(aI0{bDta-21A75Q7LJb)cIq5!oPv(}LtJ!?|?QKg+jVIR-u#ljr zAZ!dzJ86>j{p#&wLh>zoI@h>4L&g@810`syh9&Vx2xay1OUpw$Fil@Q*;e2l@bUHQ z6XFkKmupngHo8o|@9f3hd@|yMGEd3PWRRMRwZDIK>8i>h2&BHgp5LPXrv6#|Y!?7^ zR`E8f?4=|K#zB%~4!?Q-o<8?y9xhLODSu*n)N|uCQk%o=eQypVI~J92M}AFjXsFnZ z=SgUOb0lBlDPh`M5@u53C0?K}B5s7QC)}l6_iptkack#9#Z0HV+oUOlPXTlE5K+#p zJYP6GYMhgBOU_X>J4_n10#A98T}T%tp)66ICG*?sRX>Z*eG6(UQkTXprq9;o7U&~Q z{NM`rvTJ=rJc`M-l-I?g?;iZNpI6y`P6BCI;;f@U0n^g`EhtP9FROW1QV^HPmG~PR z2xBV^&@}}|f;=fAvnM0SBF#0|$Y^X$SLp5(1;~6mS1stcTH&KrTpojgDt&9M>GNwB zG~WaJc6e!4NkG#AI-6^A#3j;FM6#eCbPJjMS;2UChlEF#elDjIL+F;}(Xc`lL^@TE zwqI4{AWU3+Fkp=r+XH?E_x-jQJ5Uikv8M#Jf5^vrZue`|NIwJWm>THx@JZI$K}|c{ zKo=zncURhn-Hk>%gWW?RI~E#FkW4>PMVVjH?d6yom1(q>2Rkz!_I%Q8C6wdb`SV!N z49)`L$gcD>gBcb*fhQSakhJsJ^ur|{RRKF;&|JOJPcLWi-4-C--7m`B8iBo{zrO;g z|E55FVf$)#YDQ@Qv)LtY4zg>Z7rWuGL&*V|Ti%W9Xs%1&rGABZ@J|aRLwTvFDr=VUoO<4z z!M>m_IE+G>=6ewIJ;Js_X2H5 zZqH7o@`>cp{7%^er{sLjRsm&yZY@*U>x#&X@UGK3$!C0Tm_t_9c76GUiZ=`An|JSu z>9>QU@sUy@$;lpIvck?czH>=s0fl6qvG6N(oWY#>rE$wYuKT>EQ4dA3hFyn|Vaq zp;w-@rEt0p_mx7WW!pYO<)h8a(bPZGz(OuktVusO2NhLWj{E&6`Rk7SzqX8n3B|qD z8PM6wu$^}L;YH7_b?w{9&uf<81I2;=f^4B<-wkoAe3%B?jT=h9j^g4#(5+#PtGCul z(5be8r@Qf)8NAV+GY`(nu$fYQO^~?WWNJ!hsAZk-R@wQS*nvVy$CD)E-5p@@V0H%* zLsqi7>YkvHc_=s*Y3v&(q<-=fFtLddhCZ{{d%m%F2=Z!cYN+G^ih3Ge`J0 zK@7U6!Z!2u4ZChCGW6BKZo~w|OVE$3u5cWD#U4psHQ=dtQh!eG{m(38iKD zfy;pYR;t_62@7uxx2@O)5XhF7azRrOr$=MU9|xLo47$}ewsk}C{k(WP744Sp#>$};sj^fBGq}bpY*tX1ECAFyUdG- zHnITK1I*!#vDjk9u(FNo=5vPT6a`9kPeH8%u6dx@c z_$};D8#kgHoE!S5va+X$7Kz~@AtC;Tt`LN@*ObDWH*Z$v(MhYh%;HM;Kvp#Ywl5P~ zXF|*4%1Hc5|*M$;Gj98yEF)6nyjnHYjMWx#pup5|)CJw*f;KR{{#3QuJvy_28I;Jy= z=qs=qOv}j(sl^!RCO+O;z$4RSe2=uaHDjZ8*Fhc9c6#41q zk5~EKcD+3HYARjm5_r!)9ks3m02BQdLvBClt!l3dM-6@ft<(1(3<=r^Yj>?b9+~S| zeMjZE0M|k2{1`(g7hQJDJ&kl((5YAnVDeg0ruBjz8DngAYxv37fR5A&L!TkN4d*t; zRl0PqrpJ2SBX=;CJS7Wh()bioU}x_UdzICJI4q29kWMD0AEM5+il= z(8muu#z24r+ena39^bR9fim0CA?7xKO^$;c#$t{2e0G_I_Pi~*jlG%fJ+tiaAm?gl z#|K6w(jREj<@@gH_*AV%jAqJDB5Ze2K;jY$iydMfOf<1E*@%d<_*&@E??tL|s|lN! z#M##GA6V^WzcVtj3OFQv5;$yO%XkTr?~r_oMM#BVZ!OGnzeTTCPO!WMU+v9wMFN=4 z>O+K)U7ca$4^t_!R=P?oXNuIeLZ0QoL*R5dnUKwt_>P&I@MDLOZIRq5+x^`?CjLb= z=c)QlZx6nT zF|HD#@rP0n=78D7?~tH7_;}T6Cm+>pv^^2a3fuBMKJ{VPJ`za%51B0NZnx=_?W^zY zPj&a3!{qiR@0|i>LN$@)=IumS+FN?_;tG}n6}~dm$0!{6XOxU&WT%Jr`|a=Z zs=EPlK}%e_LGV_db{SbPx4LFj>sx<+wmhxyiyYFRii4itjE8|=8^|uujdr9#ttQtS z;nMw8WlW6d!?1RbZHd*_#ddv4m7`9{(P(X8Js`AztXvW(GlEhl&JnCS-pR$$Hu*=&qzoI z59yITKpSJ@t=@!$h{(XgJda9=jji4r5~>7RvmF;dJ8Nm730y^H?L6(A@k83%Q4Zyr zgTr%99ev9cI{j(`0_D3qMX0w4iTOiB;1!5-pee#UAGW1ZFCl^H-6zkb+F?8Ukz&QC zrlzCMJ4)8;hz4avFOrhjtUUU(JMu_*iFdzI(=XPJ6$NzB3Bj+)K!*CcZtYE@A#&N@ zpgp&cMoo~Oz{ADN(%uF_wu3+IUrV}LTqy-plY+~x%*8;l`#vh;K+vl{H_+Od<-}0c zVD8%)H_6lIg;TvIX@s|48$cUo^)BlV2RqGdtt}j;Q4E|y*^_E_=DCSxi;wj)hwJ9EYqL%V+hk^MT-~}u*W@vvSZ7y*6QPP%P-q;_}`lF;H!Ei{CnQpt^*d27O*rEM7 zM%cLyb3te;&ako>9mds`P~$oM16aknNpID$pq*lcZEt92>#U3NNl6(2_$zLO;^g_( zLi0AS?;^*Yb9w}f4Z+~H%JXtRE=LwamwkUpl#^%pct=28V?$$FSb=JmX!1Jl%`wzE zK|M?KFid?i^?Og?L5in8@^hIaP&^-3;*W-X{`}eOTWh$h7C0WeKdUk!ExvyHHaNNO z*q9i}OYaljg{Ycgyl$Qa_2IXR+sE0i#Q{6v0rXU+rfuJ%yys@bJ_yT+Ml$d48lsl? z0Em^c0l?8J2oeM6a+FjU@K9+5z`KtCz*K<+)> z2T)5_2D$e{EIweyD-ynhw_x1YwMDI8qW1GLFhUIjP|MdkJ^dl1@nl)sFV(VDu2SfV z-O7dg5K=?Kh?5U-U02k3HpB*?4PfQ!ti-j+V8EcF-WVCFa8h<~$d3aCf?{KiLB;mN z{T6CI$3e3E(`m~qZy5V?&Vh~H$6FlwgHbSbFHAjsmKqk8X_N3-@q+)s7q1TX8=Jd# zLOvEurx9A8-JipDy5z&E=V<{v*BYCxgNOi=k?RKK4xc9p07t1SY;OmtVfB6t_}sjF zwBUfbeVc!#NBtXQc_jK~@qH>P+dieCs7M{D#HU$lK9)zYjJH^&O7b}M&0y#Dmw z6tK=Ny1NOo=<=qX%ICHH@%`&*Yt>KXf`*bO&YnGYBJz__u{%_Vx=*Gs#12KtRPgb zO(P6nyO8wd*@8uRw_nr&?!~!yF)67rgy(0jPTzws252gS#bA+oS3#Jm(Cr$BIpS7_ z03SI8N?Nf=bH-lEeL9w9C_Xoq5(rojPNzL<~w5iVZemdiR$3Zh6I0gu6ZB-vryP(BVv&K z7Re8W4;8Mtqv+1fZ5pPj#;%ma}|(YL!P$LbU% zZs5w!!55=0NqLEkqfPD|xc}y>tjRW<1{=oW^8_!jFtFJ1MTJ^pJ6iQd+*5Kxb4k)M zY(y|f8_omRPDN6w?Q|9JdK!iJDB_z~=xC%Z+9Fgxr=-O9>_fP_aS5Iy@X<~(W^b#J zGSvd$DQCF>nDyz?rz1*Ou{>pLNB38)RkZ=JD5ZsuA3qA+ud!@7*A;1`&z-z*NiEwh zD{b=*JxX%;l|dO+tlP8->>qxcWn+CsBi3`BsQr@(SRAm3+Ps`m(_Bb?W&pyK(=}tf z?Qro&MPLT_=+Pto1_`YszM76p1j&5@R9#-<{ewT!k?Du%Q9VQ_x2*UDb3n7&{tc4r zMO@V=QKA?39(zggS^HawmEc+iX)}_G#m7J?;-*!w)i*}~Mr35Xu@9yA*chnndxXLU zNsu3_hT&Gk?U9Dg`ms9S_EGaRN>Gla$m6F!HIdc|NFA^Q{H6<@S3L#_JgQq+`k{b< zQ8Px~9OK5nzclBBJ6U6g@>B@M1<)`Kv|?QrdJB3I0h}3l&>k}In=qByd00H}MgBpSJ(j$!cmLAlQk7no4pa{$>rj5F&bbhs;*R)81089#yg$wyyfxuaa8V? zy!s`eZBgaBI`wQHBC$AHr3)$$n?Ax@`WvZP{H5Nu8%x7n6X`LY00I2z+M|2-esFX2 zRr5_%hB;B)N9w%NjZH~E0x$S1m-Y!J!v?TnzENUwPP2`N^vtCbVdY5<6AjnJ0>NnV zjQgKO9q!iN2TN`XmTRr2emp@SOXg8u3TG^i9jK~cn>-;>f%r7jTWB7aRs_7!k5Nzz zW?|K)3B7TNlAf;D?X$8LxOr4%{TaXX>so{T=VpzLiCAIWs15PO$B8;A0XSWEp>xOI z7;n@eWI`}0m;Pp7U1mH7%r#s%T$y8FHiUOuKG$bp?F8~nWk9}v5b7ZHz>ap~b zrIQD^#&TzhZXWl98Uqu)z7(X**&B5pAM^e8Z??l(>RFT+Ia*$pE%yi@P=3_T-U~_T z>avRe`7+UMXN&4dfn&+n_!;3bgJoJjieK5XzhA-`HeZ{=F-;aRBPB|0D}v%2pr9-? zYq|1#B9LP>$UGEbMc-mn``WA}q7QrN&wvo}ghId}FT6w0IZv8UJcp9en>qLTt=P?z z=l5;`LyibZ&ke;q1JShmZ+<#zrvzw}so9b+0&_6==g*%zYA&elLrHNZ(t9?=0+pZd zX7N?P2-2(AO5`YEsLYlFv^5+yQ7)$I_E59o4QAJv@p0pdG+8FMR!}dt{)k^HttX!2 zc-!(Bc83EpR&6KE^y^Q0i0JdR^|hZgpW!oEtXQND&EEWil@)aY zpFyEkz$wo*!*0MWj~a(Pa@#Wyu1%0o<%}+|YK!DU004Pw??t;3R~Pd04X=t^0uR6Z?zfgC}r}<45V($8nk>9WWqGBB}8oP6)y!1Lm4FRW5(2Swuw%0djREWI_6ZV8HfW=s_R)@;ObO7A2*GRt=#WU! z&3Knzp8?C9J2I!UGE(s@nHR88Di*45B#OhU|8SLXxJ_-D=0whA-9E&?V|sFDeUoy| z5es*m@4fDpyiM$@%n1NI#|Ix+ByTVY?_HgkI9n>cv}SctWL6uDdmdce{cgM&Cu0jF z8*~|dBwokBrS<3gY7Fvj-Cb+%k1=pb+u0fwUTrH3@2>ntqrqQ%|5+79f6%W-GVfzx zCwNN;0hiQf6TZy)D)0a(w?nF?r#>i%!bKiTsZ=LUJN(f$@Kw2S=964uz*NNIH1%*I zte)fKvu>se=Llx#uKsiZSHe1VJamAB6An}Wo>qn@Xkyh9R&!6e@6vbK+CtP5c+n8F z!BDB)#D|l;4=msOT$MpB`Nk~0Y}c&PQSCYiC6$-O*ZE|E{3VHNukPJrhb)cgC?9pd z?7z&LV1L@}58R0i=lLz?o%4mTHl0mgrSMbX1*T(RpXI25B+FI({v1avuB^4guxh#n z2(Q8$G?=zDU2-3B@weYVxfJ?-Ys;Nz`a^H96rOQE2GEoA?psH9x=$<7YdBV- zE`XtgcO&QK7QJq7Bp~dRliGB1A|{)Hhso6Rt%^X57gy@d0NpjXF`%|A9zSCM-_2nyhL^Ocjc)1+$93T&4@_K8taB-&k>lj ztn8wx)TKF>o0;)oloqLrxo~~YZg}ZV&IhVOTA+f`i_mS#;LALUj9RYI33VHad!jj2 z7}FPTIJemzRA}17TLrLdfvKB~l4uc z6?Id;wv~_%XaiIUi=@{hV`(YstOrOR+18KDF-oYD9d{B7Ofgc!k9Y-o_YQsdpLmdY zrn9b)jt$)dlqzouLba3~13+w!mV&*!#DCZ+gPup@>!5@^9-~6k*_OWZup>@%(n&Sf zV2t>|bGly*2j4!&^0;`*0yMp(--yG-On|1W6AsdysVodzcwgULp?Mr>+OeKwM5-TC zZ*7p1CaSn>EuV$;9vkpe`9_Q#F58#`MLlAq4|lhVdSQ16$lax4diYwwL^K>7STkjR zqUk@UhX0yTve6%jzT%$nmSl46uOuQQr*3;aw)(rf&eX=dLVg(7fnDfV){Ldz6KTmt z$ppr}Y^9>~_xndAfSHa^&P%fpo`focIkFhjmg_w%@U06Q7QReQ zKA$E{nlQq~)T+)Kl{iH~Hy0=F^U-p83uq@OPClTSE~fgtSM40VD1Bxw;kMa;J{Zg` z0E>l7OSftEFg;{sCu~dtma2~yPxUCdagL<`lCBazC-rz4=#DwUa50XX@rEOW!GFEH z|C~Mg_MQYKNzFxzKXU{SFHYQ{Ts!N?Z7N~4crlL6^~K4Kp4qpQ{Lgpf0p*I*KUHMkq0UYs5A#s7gQm8m{D-~cZGy{VwUdU0wxrXZ(QuQA!X?DIqg~0U2*j~@o8yQD%k2e*6CI&k(loPC{ z*L4F=IfL2IM;?74Pn+GB9Q!>V@qQ=MuFy0i`fm`XjZj@+85D7fXJ0cYB(n!ulwobY zdR6`dW%W>%&JDDu!#%7=0f?oklRUQ5wsZ_`L!ez%RqvGRDh0ICGL#cMRv#fcD_jAZ z-VjX4GS+-%rm9B7=h?l zI#=X8UXMlcdgfS%w_-tf3`aILUL7d`9NuoOTY2%o@GZPQ*X~n`rbW4_q_FokQP)kl z`2O?jMmb+-!mEy>?27HttuzUV60mNq3wyf2axBbYKoqB&tuvS>Ji%_OkC~#Epe2;! zA(4nr0GuXKx{+HxW zfB%l>W5Hby!qH88{~3qap7;XYSlIi)3TPy{wums_>3fcu8MmnP__4plax>kMck(@o z5!~9Ch{qbzWwwOc-=E#1C3XAr=hIvsMjFhI0()GIG@R1Y_=*tHJe%#J|Lvsvr`*Tv z7^n&_PG}m_{P`9KBML@gyK6vJc^@o=a{*`#*(IE7BzumrQHc@2eRh5g_NP8S`&v%Y^UfydW|@7FAoWfzd22AO zFZ3y9k*e3}Hy$I6*52zvA`C>GI6cYy=1N@R0f&G2d$Wa_`iXTlXsU}gTmp9oAkuz7BikZh zyYSL4KWDsBzDYs&i1Mp{y@L(r3ly`K1ZTfR2DBDVXIFoKEvY1k$v*Rz9+q?lY%ztE zpWjBvS6-b|1gxeJGK|2`0vszw;TDrkgM@fhQo>_%XT;rS@>qmkNuj6v(#WfG3^V%c zUO-H*&|yV=sobAwUog${8s|)TJEZ*U=;lZ#tai`j!{Sg)6JePGV>gugajN-aqhTP{ z;wOs$6PTg#n-8fBmRkkINx{Dt1=;@fUH|hD@mIyF;0M-1AnEyE^6h5cz6`=f7TIud zH%E{ljyzYVy~w}F1LRiC9l=RGh}?TYlOny<9JYFY4;!Bx$^+GzBPgi5VFtfEPOdJx z-qrM>B_AzcK0z?OeS38z$j=-|8;;!09JNHJzw%T9U=~LRLh2lBImGkkm8ZxxujK-E zzy2RF0uhDg%2im2y`N_Wi(x@k2&T$@PtT2rK#$F^?3l7ooakald?R$6VxT2b7@=#v zt=^>i$=}4=u>TH(?d0UeF*>fkXfNr|z4McI} zshm0FaTJYznH4+?m={a9saOlK%nllI!Ug-<>Fw0@H_;vyg)Q$jGB(W5+3%LQj(X_N zRcU-%EnLUoyfZpt1QlwxY>J)dyHz(>Uw^70HKdvk^Eyt_pcO!l{R1{^Mts}nIQ#IE zr*}r|up(d|jFL~*tjve_7CHfIp^9xc=+MGMZTD>Jem%}ZsH-y}x(q6Hy~U|jRri}~ z6g?2Mh;;kq+j0hM4bY)~@%{Q=;pO=WfB&h-i<-zYG=3SW(;e~3V_vo)Ye3`-Gg0PJ zD;FbVHPz<^s=q<-WYQ==w<>@fAZ$J385bx;S0_k~a%fyJNE?{yj7qj?nZ)e-iwh(M z7K}26@t&_%r>BCbSe18oh&Dc;uic(dBs0b;_+5u~^HjXsQt#busoi`GjJLPmpu(}$ z!c@M+Trj&z`7vmTyc43O@ml}kjOc{rwplmN(#n@l9wcQb-cK1~nJoV73DwD6Nitat zmVX~&Uwergob;bS2An9){p?1Bv{}Ol!6%N_ZOl7c9qGgda0No0WM^bkDv(y|Fwy7P z$rBfAn^K${$uAp@Zha2~QCr)sAVc|H4?BJKu4kAmg2=&3r@|(F?HaVO$ohm4IwhH8 zGw9_*rH0cJoE|-kXn9>009;&KP?3uiM8R~Zi~Yf89e71Jvw@hJLWkVoud0AOLSm4#y^ab zvZ&d$6nJM?dY{V2@Z#zK0B*8NJVh=J7NID=%c{PpQ25yJdGq6tE08D@^0T8hIA zz(>*eVfx>|oq*uYD+G@C^TS-N|9?MyT;t5^;9%r$c43|eZ(oaWhtNx=V7A!O2KTbR1UWgwBEI(_QYueZJaI*gj%fo^lp za^~;Q4wYpgpIWa_X`fwz_PXo)06u_9u)9nRq= zoxYy%Zr8U!>epl3AE_pr%n^3D<_XZGd=@R~mUoj#w0QISbqL6iJfFR$N`Q=3R{#ND zS!((lpBe_e0^>!U9stqmo0yKc>-)`&*U<~_(MAO3sMf0{6BDaxR44Yd=iF>!B7FTX%H-=jFG)DCzK zTg_+p`Q}gZEk9OEJEm&o>h(bl9K|!#;tPE@fP`h}Xq8)4Bkg&r2 ztPYIKacXB>w(_gs$wD)HkB!l5(ibY zWY|;Xl(Zh+J9Y2={SeB-egL@4fA@^v-Yy=$H1>RKvAsH4nBa@ z*=Gug369gPaNG<3&-hvxA$@7lrz)ENS*jqnZnsOa`)TF% zn<_7;ZXM>VI~H|_&iKgR`N56yu#SY=N}RIP$}vIffsoY`=5tdG-J&@O__3rLYDpiNPFwBDBrGYSPAKF5m4z80g(N7<%ZT`7ZDKyPxm*j^q93y?=jm9K+0cUFY6w zuf6u#-v!U|V#cf22k62Yj!)aGQY8RAfk5e=8nDSqkGcVuED2#MOs}C2Ho{V3%XC<- zmeHS6P2|#h&k|Nif%XtwYzaLf?sOvlC%*ys-ApEEWu6BstIUd!FO=)fpXUJ}j|oB%i7@d-Eqah~b@bTW)VS>ksV>2X zlD3OL;~kEx!5@-=IvSSFuKD{AbOCIZ=b|IqZag1gfUvmPz|vTWfKgl6t395Mn_QJg zZ2I}w%&O^sCYQG-3b)swO5=2Qr@^Q-HPdI3hRv?}k4Gb4Nv+b8Q7e-3sLnD3yk~gZ zuGZp#EXe9RTh*zOdgy(z8jgNbgzop4cT9>`{O%Uzdb*|Ftfu_=XDTxuu%_MsbF4l9 z`;l?LNt9T>OJMr_k=z_mu`OUFr3*X%^lL=UcNiZ?z8DYnm{F{H+&N<$phES+D8E-) zj}B$j2znO0)|`+|iqC`dIXN(v$H8cOe>vbDij02pee|i}TrD4W%%AtV3@8E9E;=${ z)l3oDLD}@}u!LJ+^M$%^0M^y3V~PLr&HT^XOp5mR4LyV$G$As;*Gq*vR#+K;9m~& z%>r)7MFWm@PBjoYa;%(Iph#COGga8QM}^0b!hsS zh5AkLdmdnATYmf69H>?|*!7H72t)_lw=}3H_K(F?6PGoYi%GoxqIO9q;`vg$Qb#3o z<~}6%qLzk#E6;6r3)yi$Evlfj&7BpH`B=b;*wKitjP$0HL=jPA>=X6zSE?-FX7+5c z)l1XsJ5u@U<}l6?n7#V`YPa&0_kz$IqYl$cT=Gqdz<(aWe;r-MFb+Nv zFX1dSyD5ETl0Dv-^RJ>~%bcO0Y%hmchJ-aZ^B9#S7r5HPStUk#exUkQ(l($&5Ozk2 zFHWFhX-qFdod&=hmlaihXGf_sChE3d!k!^{{>NNPrx#P!{by2lun?qm_(T7%lvf{b zJd0fkWgXazsW|nu3eXcAw^Hbm+K!HxVW+On$nLk)G{( zSJ9~J%UU%_Xp8t0Zg$<>_tmF;*sI_TI| zC4VP9U!R)hgou*;imkW?dP^_hD+E=`GV%`MJ1$HURO&nCHfSK_O4BXX*T%aW;_3L^ z0=TuUi`3a_7M}03d;MYx+MTJKxWUBRYSS*Ynobw~tGh>ZU!_3=o6DnTG)>pe27PwmnNKT*TMDTN#_k2x8qzBdETB ztp;mO>%LB>x?Wwl8w&8fQBVAp0vBHq^#Z)0w!iMrLuFkZ*>vm6wwXunFBs|W?|gQ+ z#((EKnRl0<6FUahU2{s_GI4j$kQ|kINXcSA&w-_k#4JJI$nZ_n8E|)F8RcR$j7|4% zfG7G&g{_l_{-qS=I`=;edC7qUL00@fU(^2?%zVJb-*d&s1VBQK!D)%!j9px)Gj>^g zlJ6Q0!P^q8;v)Ig?qh1jArl)GV3Rj=Lh|Nm24MD0mF<|$rvALxD*6PZr88ZwEoS^_ zo+Tm6bgU)3D!BqOvPuZwwKr@A0h2t%@RQNfr>3*rxR=y9LIk^uiYAx7QXT%s#FT>* zg-UXMfVZL|LMY;uj90T#q3SphWHf*@az+PDX`MeTSa?kvt#rAnV!1+v)(dwv#|l~4 z8F^)v10o8N_VVbWQA7|g$OPS7TK86&j>AN`T4Tr+S3zF{;eNk+IYqU#~3q1Et_ha93%gmg$?m6-CuI~RH z_5Ax{5vM=k%*3+%uxsAugCybdVpWI>{Y*&zBzwb#_K?sel$#%u#hQT2zX0^+1Tl(r5ws@E z^+%?RI)PL3T|p&mLQN^`A`tuHpMpc>P6%*9X=YCD$MeYcJODl^jT(miI5CeA-{e}M z+p?WM-%pi3oJimtcn*5QmInkcaHesCsSKA#`p-lGPN;n5`{uT@5z*5tT8G~hv_s2R z^0)(Ft$`*0Fiu?C&WsPI37#3PeUNMmlNRC%UUnH3=ObVbb93fzy1%=wn-H{L_bwFB#Zt7`H8lWHB0Z+Ako_59b{^ z6rhez?pczx%k?ycF+bh^9VEc-Tj8mqdK!R7#Zxd)W0BW|j-_$%zR4_1MV>XT z4;N6oY@wX2G#~s#ni*_&ne4!TG6IN;@*{ZKm=Y3+ ze9tK8M0}G0zcSgvfQz6*-~Mm3tO>B)$r6fP5kP9dbOWr8ljqnw=jq5d4ud1fSk*KT z_KQ|stzHlnOA0X^CQ4Sb`K9_p>dgts+fy>AQrh$}2Vx{+Vj-?jK4;%&7Fdp1thIKB z$O^!wgoofjI-jxoHn$&nS9wZEDHCol))e|m9!BrYcVxATc!DZ4s`mC9aLDA};nUgR z!~Qw0WWfL*Enq}W6-$OPA4uF@PY_v-fO3K{trxb&AZ*_Sx&=tZZl9Z*et$eYg4$bD z`dj&mm(`5e;m306@p&u0?5ngzjMSPVR38g}#9st9@aea_))doqlHwILJloOIyfL}F z5jW-bs|N{%ubl$Z#8#(YNMN>s+MmGZ^`!ZNq!@~en8%0)Wf+iGq4SF#%PPg8!S%** zS7UO}@k_vk@l#&!2D&n7V7|Kv-Qw#XPot}>jLNLkFKs7EDEtD}FrsKAGgZIrv!=aP zFQw#WQ`-FX#>~|6|8Pi=`R4*Z`3K)d5l_pR93*87O+^ zR)Jtz=XXmu?gLmeofWRxF~F{M(4aTKc0_qZFeWZ&tv8~)wOQi+1F*RQUkI4u>EgQ? zS!{v+kQ6I=nV2c#_?xqyHpDXE#>r;!2h?4kz0;P}GyET3%>U#<95EcuKQf;Q@?^q^ zG=W8v);E8u$8paAFr(6+Wx#w04CzZ#pM{KbF)7|m3V)jN+TL1WAfG&2)g~@Ifh{8U z`_&}Iazod=UIy{K;>wem?gK08ngXnn)RgU1u1>ejP1skesqMghW3%7~7*6>>3U%3C zDy5n#F-W>1c*53XMaV0<@)dzU$jm~y#SCmdn_Hb9O0vDa%9uiSIFt}qqR4cwug{UJ zH@`m2eWP)`gi&m3HEBS>Mh6}D{$2-iw0%e5AT!lT;^}o$S-N=icQLer)Ph0>|*-p3`-#?dqrhvWBs$8hUkMe2zM@50-z&()Fsx8DkZxpXD1)$PK% zlTjS#c6*zT?Le1%^Vyz4c*mVcVnY(E%JUnDZLM+luaLNA z-lGIIqx34X&T~3ZtFu%AjM5W;qsvhsV*!olD}>1I!rG+nZcdYuiP2KTEO41gk?&T! z!?wGy=(AlABV+bQbfbW|+jb_SkNdbmO%m%R!#~DE>_7q&?CWN|ed-AGN}Aaq$ZAC~#3AvG6j3j?89{eGZgT0vCG%RPthrvG~DKog7hd=ih?mYCT3WpW%iRAjax^ z@4WV@MTM4tDR3>2@pTtgCw^|GJ{_HFQwqO)@S6wmk8Tn-x?#6c zyx+}%@Hp|e*bn)P|6S(~gZI|6;=_sxtgja#kMKd^trq_kEBn6+e{AWe^_$kQ4?J>D z*1I)O_1p&Um%Z;L{IDfOg9NMuoeBXu@%#O3W!p!OS+Kq8_+w5`EE#N0xV6iyQFAa3 zAOU}Um(gbpPdKQ^5LJSkf@K(Hg0N~wR||MTg#^6Nbkp z^)Uy3Py6QEkgdLu?+4;Qz9ZQjYJ#sjAoDgYB=UjL<(uR~gXf9QXAY3X&l+x^2&hd1Y0T_fNg z6AQv-s{Nq`o%zOdysd0Kkt%#7{NICuFRW;a$f4qPK?AXvLae@|vv(^8Ey#k??jM5w z_Lz#BdZDCz{yEs>Y!l4Hn2K!c0O9ruG7nMQu_vF3qUz#@L3s0SXcpIY0CJjvR8Nrx%`i#-M`u{os7Wep|E2^`2T56{8ul> z-UBVd9!0xxZI0%F7Nk|IpLR2|_jREox!? zTHyJ!Zi1{&1;AS3zZ_-J+oR{-WAo=l2!Nq~s;B{n?Y~#>YIN%S%xE-IR1FwLNpO*`0RCm8>u>1bSM@z-~?2m42*EjP`}@UE_0>5ycX4%eEvl=#fe(#T>SDY%*;zRwkB*T9=BDi*^?$jo7ka9wi zB9%{qt?Si%J7hTNu;QqgDMmA#As}17%(%ZKEZ&84b2ejKd){+4&Ulr|XE{=Y`NL&G zEUeMxXwBlv?3z$FXUsUrACLcXbDr^Z-$M?n2|)N_`~AlsmoDCmkqX_fWA6Nc68X6u zRwYIPdj7FF?V;NBc~jQ(Wfk8;C;iXanfJ#V)~+X;#`=6o+?h)7EDpw)v6#Bcn(|ZW zdCJK`OR8Zt=~;!XZ8f&^5aqOhN>kc;*$TA!|bcmEb#rwLA}^~CaGVyiLT_q zk7Z6{zd0~0%*f*hhV&6Z<=+%*N=gH3*7M&+lxY`!J9_CMKKSSQs7ndD_NN-Ty|U`C z@eG0oR6e23KNkDp$xc;2`d{blc@B^^PQVTUS|Y+3K`Y|AMceGXPs9-`NklH z^S=Q3({3E!i!tw10TjVN@vjT;=Pw#!0S;68`p28jKya1`))X61p4TlLO~jqe@@)=t zKUk0fhJc)JxJcs$aL5YBkC18_>z_xf*SwyHHL9%nrM_hMejc(Mi7=+SN%cn2K1ZTu zB)sc8!HF+`4L$`D^M^hctkJ6zp7wklI}tK2AF1E!p(BE|Htcgk73`uh1}T?Igc4$1qd(`g&5gAB!z>zDfAo()c{kJg}LmJN@AJQu9#e*6d*SD-L|I zS@c6=YrUoM1DKFi%B3e+uaAG=Vq;#hnBiOg4qPh0jq3MIZlM_5I&m6S z%1W&!E3hIHrB_XeJKw+h<*pHgORH<%-vxV|K)=NIP#id9e7X5n8N|}~&*aRYIXSP8 zAGzLBupLO{d}vv~L*j*wqvhS-WfpRCb)s|GPeyX>V|ab3-98+}NL>^JT_nAWj`&}( z+mtyF3L2~afC_P}D&Zs6cHzd5c8U%FWiMn3{_>!x#;{M~ zV>f%~(#;g4NRbtHrd)QYF<@Dm(6wWDgPn_WF&J?|;9vSL0C6Pq^gY~q-?BuZRF}H5 zLLX{&v+%_Dto~`>!%r>kvhV^(!jwwSvE5L;q1E2SVCs02^Sa#!l0UUxr9l6HV*cFs zl08GLISYW2^m;h($=>HH`S@OzeH=6gN1ybpX5+kE*R~$cUG}}xd)N7_q3TOIkAwYo zq0Vp#Nueu!S*6JM1W(toWn#eIOgMeb@3)u-26D>xVd-M??>N>kjyZZ+5dp(=Lwx7cC_rX zA=TV`Z*9C8#SN{ta`Fgs%U;n;h`CP^?@Tv|LlsVnKGOZ%VjYjUZ}c}@Vtw4irTicR zNnB^~{2ZhGPrIgfpc{2Z`NX_&ieZY>C$t8c6!Pne3uPvfz_sr7j5Hh_bW^)QH75<^ z@rgH%z>5xHr}polWein8AVuS(NHxMH5WT$@z-j)ziFKdL&p6ZWfpgTEr*t2Wz*6!Tk8Y>w6) zdz{-=f(yPj8RYW1VYBkAV=fzCSEmcb0un5aWXVe3Ok{;`R<`TbyKaGxd--Kf@v7%O z2;xREm?Il*1GxnbzK(@_R+V(Hj4XZ_R4TN(W>NOPl2M`qNd(T2&}zUL`N z3Q8PFfS-tx27gU!_dZKKHLR6?LzVM~lMI^yKyuv!iRRHP4@wRFIWzd@FPJ+o-`yM| zN&%GYXh{md3EZzqVi6Z3zJ+Vo6$l0L!ap<63GDa6^3Wk5o_R!BP$(RCH z`V@A!RtiZKr%Z)WTL~sky@(|fM}}>LM^*gq3FHbTlA4Et6B3VVUTZ^bPnWOg7>{P$ zf3NQ=0~u6H>nW*LJ#Mo(!y0d^0N2?qY~?&uL^#UFZAf_M*x}_UOMtGfkGf^A4%efc zC7<1Xt>4H@qU26isM;j#+TgJO5AGQyx6n}Rx~#5Qxc4QL&HxQIU5-i>zCWkih|_E2 zJ^bk7>e^0v+7FhB@!lB)For#X@8&GbK@sZ0Hp`ACs?tZ z?Ycb7>klUkAT5b9k8q0*^o=S}} zfX*<_Pg3YJ^v(K@^>JftWE2dKKLUl0q`_#dde49n!l(#Y{zk({sAIq7miLbjMpJ3jlp7ELq2j^QA;1()e_WWqZV6v1q^-E`p-t-YQM z50N#Vg8W~i5oWS@^Bi8+PNVkYrKmJkj~Bt+KT;Y885Sm6PG@lCXo}i%fB`4uuv?#c$A=9O+p&vjF?AJ{MS8#}^VwBPnro{3*tkuy#!t{ImR5B~;?!aXKBQ}P zl4$38G|p^u=1ZpM@g@~wT;to5ycv?Ke0SH&F6@|LG&%a+_~UTg?AGupbx7WYL`>u{ zp2dS5F*f8dOmPDHld3fD1ErZv3Om}?ctB}T>-*@+;ZbRVNs23fS6n3`j1PTbbyT(0 zQT=F59JkJ>eC_3Y^CjhQ_d1=d(KpL*F^Wr#LDIf_=g%ntf{1LcU=)#^T&+0BgPjp9 zYDXj$WL8;}mm%t5#PRwg>FRe&X{X+ZMgrtOOJ2Jm>vF*z{X$~+Rn;F4?@Ie?EUk<` zB>|nu(Ix{QeHZGnPDd4{FiphC>|`B7*WUS$JrazZcMsw9m7+PwR^6!;>vQY=KJaV_ zG*?rSi8u5Wn1XuqYP@^vVDTDfxpvkqoW2V_R;PvHH+Z{eFcaIna18vUDnEm}?jin(P%*az8yZs3l|`n#uOjnuL0F_}Hyfr1;3&eZc*_nZ$mQ_|e*#_A?$21B|>{icgB0F^_$| z7#h3DZxiKYlp~jAXiXcaE2Ruy#B2q~Hc?TneSS8c7a;druAt1vTsD%&V~_II^8U+f zO?MbizehGXmGbZNr5a=)wVVj|)x3lHn=w@MwTHd~5vAmUtyX|On~y^*S@LU#*~gP5 zCy}V>pX~wP9%d6%0R?P9(i@4;c0>6#OWU+7bynKfMM?~~vitY4w`!*~#=3=M8}*i` z=pR{V5prpRFvZs{9w^VAzM*7XaD|*C3&tqnev;2Y#KC%R%Tlfvqj{{5Fd?*TjP;k& zbG~0hV0BvA(d-FypVm<7O*(MlD9`7ErPC6sHJ3T);(sskN!E zv7yNxeS!DURJ)-PuYy(E9)qkQ`PnD_qbuLxH7&%=7lV60NiNvtCq4clUTX7>%CtK> zzw6}+A}Ul)hr}G8YX9z(54G+st^J&4B-sJM`k`WWVIGrGqH_Jd8a$?91Shm0XHCri zeDMDx)`I4vA3UPbuifTC?)JU$D^eP$Pee6(EbBU_yz!)#$PWly6BglLHfoRj_$$4T z8(#ceIuDH?c#tQ?UHtIXR-d}FmlPM?6|9$+0~WYnXB%LJdU*u>K3S#^@&E-IuRrf` zdu2La8|cZ>#wR06&%YyUGw~xu%&7L%?qAg@6M0w1Li3xG5IXoH=*blF1$1nk1y_|5 zkYsKV23)+svO~Sthab!949!X(IreTPS2Lrm z&#S@0^Y#WEzg67da;i}$gOKQtKZc|3X9CsztHRK9VvNwY@pwIBYT1M04z|AYkd}pm zaUPGz$bCz)?rS&t-AK#X%{fHk-C%NEHJLwt*MCEUVNtZnE&8*zq>7A1HG%)@SyA8^H_-=iJziDvk_?mkfAs;&|o=Rn2W{b z0@sbADf`EtRn?(ThqGW+60Sn=y*~d6ior=#@S72$nO2pW&Gt8*AFn*{Q95;svZ*ia zL(9#aijXOH;`+^w^{sUk;#gMvFk4AY6^=~W4m!{Gb(=DmnqsN}0WVDs`-gn;j-@xo z$_)ha-l@#PqRHRmp>uinxxlIDbo{*CzWe8VE5yK;BVdrX(-%l8?2pb;1`GjkR`9V} z_?|(0wGEBMEh{gduNkzS(xr;{y*9kuwxnoVSQGyQsw4!+ry?)xB?AqKn5C z@z~{5^SfV^DDIstwq@9-)$zxRjL={c9p-v_K>{S5w@@sD7F5%nOI?_D#j^ssQKt-} z+cp(Hi`JXzQxu8f-Acg{w1T3759B99J!x>i+5!+-FN6C8{PeU+KMJJs^LrQX#7x`e zOWb~jF?VUtRh8>NNddB}&;99=9sS1V_-3Nkb=xHaMLW^c?BOiCskjT}s1p6VqMKAW zEysf$rJi@ug+K?fRCA6plA4EbkIq&j{y6Z>>q*+F5A1IA25g+!Q6_>kZ;{K5F+qPu z7OzPYs|0W%N`=bwrR;ia9=C!uw%4rAYw6At4oQ($x82MRg9@3~?)KB7*4IXjMe?na zrP@zE&V$-z{lq{cQgo(cYYnet3E-HtJVTX;!VvhkdS4+9iUo-_batUY7(_+yKls1v`sN8V<)(U9In7`D2ac> zsYVjecq}A2$80FtAw@UekW1f&g(vgC_z{>zwxg5Ubh}x#e8*e4eUBA_D7B~DzZG_& zg0(;MlO$f?t=?_|`lcC^icL~|@t}GY+8L^NNu~MefSX3DN(;!$@%#%h7H>6bNsV)o z$4Kq_#rZSAt)8ZGo}Hf5eYJG442k=-p$DvN{Y7`Lgj#PmeK~YD6^3IB^@5RYJ~M<5 z{Ydqts`Ex?LZ%eXR`1OW=`HtGzh*hY%g<(X#0BCsf~OQu^;4@Yzwq>!_ebRjMR|0U z1mJSb)p;4+RXH#lWmamZ@3hV|+I2XXJ}3%6+B-X^G-v)h_mmLLoL{~9SKc|eivrDQ zeoO}yj?JKaTz93=ug*bMUc2ZIPclC+Jl|ydUu_}B7Shf@K`s@U$+!?5RI`nWt0k6* z@D2_pTiYqEP`c5R*Y7k%!pD!vy5|^ga-X*va{2r@y*#fLu{?7MwUva_n!bKPZhA`j zB2?F zGVsY70%h&e&q#6a#l4As;M*bsg&ZZfw8ErT!Hec2dogK`WW5b)57@9uG$IiAwmQ^S z!BS2I0O9j6Qsqj%GsmOj+3B<>a?t0EIc?%e&f70OLF2@aRWu8R$FW@6h=%Yon$Hsy z`9M!~bquqa#6ij*rQ{Sl-t<_=(Pg4@MFA9kN9#ydz*2sqohTVsP8tr2h}d#TycSy5 z?F#*Igx;>A9xXcBzKCmKWvJ{kP8t?+q+GOD3S?YZg!o9+WQAtNZn8E?>steE%n3-k z)6G#F`+x*pzREKu8ejQy5eORnaoEV)`9f{+>knUh{MR;nZ+B)btF%N0ZW$SZ+LCnD}2PJ=X5%zvlJF8j@gA}*oHlN#^D08Gwo0T7wq@E^&Wlswe_oK){%s% z2KFPZ$qnMO{S_Z|{_~trXgkBFv$X-Akc&hAdNNo@avRZkm15>;Q2v#Y_4Uf{fE8?O zJ<4V4`ZB$ITI%S4OYQ319xve(i|FLqU>w}!r}+)e(GLaiozoNmxe+gY{-{#>msW?L zht>K`XtMkkrqsx1$0=7)1|((0DTCt{oEsiRkP+S^Ov*G9=Cf3=xJhX8z2f#0FUB46pPbLAH>B>chC1ERP>JQDgAUY4EXeTD zqEu}7YjUr0efAy>ddan_A*D&T;U&wu=7&p`%K9N@RGAqWz1rUUeyt;5CUPE9 zFeJxq2EL2i>xTGo$1Ju<@iI+2=fJNA&Bj58r9iZ3$k5Fc*uFm_Ik1m-$Rpiiq!lM zjxIeyw`C*oqCK@?k8Qqhi-GqX&vkoAt(nGtXr@O#pxvrv%Jz?jYLSj20!#pcLx1)h zW_WBbNlG~_>&iPl(Ss|hg&A%y*wVo3Zt8I>=V(9+Qdu-*v$GNLucg_}WLs>z33OoE z9ewBudDo=B845;BOGio@f5Av=E1z1a>hu2 zn;^AzdO<50)UZ`qW>8pwAAMe`9Y8N;;-$pJ%%zPTfD6L=Yx zZ9J$5^aN;5PFB~T9gmOWr@A`{F#MiybOe!L{A|C)+Me8I*zkXUj|VL~D!pC(_@0S6 z$;6FR483yyZO7GU#zI5VziJN2UEY4n#NR`X+UHX2kL`ST*5H?qS}JB5hua-$J+#Hq z-BJh*y|KRWf1c~GbLaUu;~{Dk zq&$o&BdTDvPs&c~?zu(2WX>t|iX6U>?JBLH0~Iq@mh2*+JiHWc zQz?4ODQ5$kEbN`dupX1m=H5)++rli|$8|oVcSUAU4w&tSL3JTTHX~bm8#y*{YwNAw zO&*guJ0WUG_1$PK(Ae?&bgg= zc@(TJ{ax0@lP) z9~rZVxPZgOCjF?PR&xV#K}odIy*E23@zIz}OKo-AeN8aLqID!6kklX^(MzSt*f$pYhP578fe9VOy& zp1w(N=`t)5W#Mf*X-^&Ii#2+*IlR>v>w#Qo6D;YCPk=Ifidqve)*hWzYG-?tUF&dm zIh~OUXvOjpM03~0cc2lPb*Ox|mWqCZFVg-M?PmvVkKWxnZc4s0dB&n)y}ArO{aAL` zrCMZ3Y%>do1B}=(Obfml=GQpHv_Z>ed8k6cqs^|!by?B2dyy!yc++(5uj5njrc5qp zblDm1Bb|r^5bu*WIAOg-z?`WqoT8n~d~q;FXkmld9m((cCF1$qwUtgxqT_C^_=!ED$dc^c+L&}f!5;rLknOemHy_ELnU@;m*1UWZ;Na0W~(gftM3MY(@ zHu6&ItSZI%Ynn4=#D3zix-qZ`9+D-N=K0^jskk$)=-@M4%TnTRri5WSMKote&1$*5 z*%16_k&9ywU0b*4?!K(?1+5zn`&!0W-Q8VIGv|ou_AOD9BHT64(qnyyt@lWgZiIGV zt!1EJ+g1lnjd$#^sOyCg4c0MEo+mta`~YfXryPIxXSbnKpfT@0*GyK5sS6U=)oV9U8UW+Y`W&c>sb$+D=f`c#b^J!2ei ziF*TO78kl1?OvhY8mjx=aUZlyccy}I{kknNyb%dPy4737tk56v+Y&vs_q8li>r`ur z4!T8mB+@2l;Msq8>w;!kOH($-Y&1vEDA&R#2F{qQ5R+Zt$U5YYIjJ!W7Lx>bNHs-{ zTDiF{#52T!GIk5o%v(N2)tq?e51 zx^YUrZ9~t2s2dGAv>bZA*rL+(b20njTgInxzZWYxdnV{v&Ld*Vty3O5N^4P>08U`u z5|zsOIDfuq@x+sU_w}bpiYVAjNP4QsE-suM)=w z%$ZBdcz0V4yXRv>veMdZ8n!*=N8Y8d>d4@4q036XzT&2^Y?bqm)ct(rfw3*eurii6 zP{)+3=JY#OUH?k`BMs*M`_s}D`f7^ED)nxm@J#~0IWOGA@vEv;qRU{~u_RGIToP}s zrQC-bmZ%%M{) zASlYx@66^2x!by_HKa&Sk6xl4yehzQ>Qgej8KP^cvn&i4xwwzm>-}XEDmE^nUlzuia?9{<>W~rb|fm5>{7xgTMaA+iAR`d`S(` z@y_3U#zVxYa;eAh+`-dMEhwb~kv-hE79BW_a(*1ioq1jMb99XNf7DtyF%QaKw^m!F z>=Ws%oTBZ*ZKhijv};B~nPn`>3!U;1r(J!=)iHqNUIjZ5iWp7dC}-eQk_6n1@-|gy zHQFrmxbx@rf(XA%7KX?x&~gEcQ+?R`*AipmC@T*vys7m8{P+R0!%X+Rxx6M*#9H6b zt4|w3z;x@YLKQCsec<`Ow7r-bT*n^rHtKMa$)rrMgg;5-{@7_wqYF0pjhCopKT(n* zhH3lDax`N?G=A%4o^|DL0^)PqA2qI_MSXoa^~`aoiVWCB$Zl?O{CN`3Aba zY645^vb+P5J_8Onp~!XWu`IS;_ue@kk241-PIWRThS*yyLTh^)%eik&#}bY=Yn@BZ z3Nql~kWNx2F3%BvJ3iBtPhh{Dg|0%-qi0+_W$lu!v@3i?0cyin zisXI3z70pBQl2V|_D_Ri*324urm_ncn~#XLq6|rK&6vmq;q4Ovz#b%0-^_Ua!YI?X zmW8#6Xq<(N;RN`epR3-zGP4S>0e8GO+nO7ro;>2^>aZA2OY2GA!oO`{DC=6+IZLE=F7-Sf9vTmDbNST9ZtE7^z}FznH`)e)rA^^ z&vel^6UmGpvlE#PKdq^94P_28Rev1Z+VUa7?9~Uy(-zK0 zry?I!MT-E$ILGB%r{Oq(c06Ahz!vg5d%x&2k+E8XKne_sc0gM3oQRI03SR{LjaD48 z@+jq{Q|e7-9rl_D>%y{L(cXHlaOpEz00YnYbEUVRZ>PQ@n_8O+~ZJ+yCF>0c!aUW`AR*zWhWCh6W#|I`Z@ zzS5(5yHITFnz?_c3znVktm^iM{5Hs{ zSYoEspE*>5DL_fOrhOY)`m8!Vq1j83?ow69*!!QIEixgdfG?^XkL#X-tRf_dGHRym zHW{?p^c@&`7;(QLSa$)mW_>D<_N%@~v#NUhHw^xDB?(;XmIsMe%5>d>N`{W`ojPYK zPDveN&KUNfK7%qaE=jzhat9Gyw}Zgq7nosbs6A_+NO1kRyEUQCUk2;6@F*e*nvfkoR*Fo)S+6g!{&K zK^U*d*s`VA37?nwVd0lm_)B_wNwM9#dgdhPInL@W*PY{Pb|$$rYV4abR#uav#^h zssX8vt2}pIG!y}e(;Bs>1O9sW$6@LRd)w=4H&UGs-&8e%+z}OR$G)hFeYo>cwOFyx z$KfO;=)po8&$NdvW&faX0*mj!MJDg>A}ZJCCVZi~+ic`&XNsyE68J326i0YU3(mo4 ziM3OOU8AV7AW-%Uk7>d9iu>+a+n8~s)ScLzRK#8DtbdTN*P!w&p;FcA^#y90`>#e| zyCF9*K&};wkbW9c<+Nwn{#dcS@eLbqNyKX&JZA5@8@};^q@=!$_{3paJlO#=4O6`) z(-kbbv#jkuoqMC1P_8wWNROo(shqWsYo{ojH!qoYX{s!aT?Q1b3d~AosW_=AxEGKtQbD~`rmMK-Oj~0{ztqq>xJLWaw++b`SHc_GbzeM7JL<(yd}r z&m3}}l<2iZ?sc?3WzohP%F#FGB5#^0kIDQNS+5h~?WT-;NcR7b_7*^KZC%=MLP&6T zPeO2amq2ib0F7G+7Bsj9cXto&?(QBSxYM{poFa_@a--ui25YQCAOu2dIIHhs?C zd!4n`ene=Zq0#bQUgbtn-q5bN^0&01DjgEqe;9{`Hi~tJ*VfU6_5J~1w!)`UhS)-r7UE1 z2pKZDL7L?+nb`eU1;N-&Tjr1e@D@!41~MNY=dDG!J`pKea_Y&ihm&Y~D6uTxZoT|$ zXP{5-%l!5=xw2KY_OfeO|B2fj-8OrQs+&1@ZC6>wdC7EcIOV=4&Q68Y(z)EOf>Tq z!v<+^VAcM3SaQ9j%EFNBdhc9LYtST0E;h!&jcfV3>J0wtF zl0e+9Pv(61-^Ce}!I}vdv?cFtmYA}Z{ zSUE?tw$7}^nFSO>AyZq&UFB!-15XHCa};TU6vP+zN@$hn(P1Tb=zy=Yb5a1Hd2Sg{ zfxE8uX0R{!D;J$&D*{rOimioF5&J+cEHY^HXB&JzF@?HuIJBU~$z2WL@eqg^RQ)xN zG7SIG?~wcNRI^1^gal?*cxOznFstKyN@20xwm`hDAiSW~$Q@3R`G>HJPA1qQqc?)g zcHW~w9&CclYb1;Z$W${naoppbT&Czu?elQm`KqRc;4Sy)yvdh~rBr*$n#?=LDQSj% zD1!P`trE?-C~9UQ8*qG-Sc#k&#r^g9Us?dTY8r9#DD64&enaT}A2Hoc%+W7HkNFBA zKMd?9Q$nUjaIqAl#znu5dx)aW;~!4wZZser#)k@?C?3v0xtCA)?r>$0LbuRm{mf_T zLu=m9BW`J$wZ4bNVh2Hi2GnbgW;>y4{e@1h1NQ++?-qT z(xA1iT98-qV)xd1t!eH(zOricUnKplM;<8Wv1-xskp;C5Na6}?g@|3!z{i*ccX(=W z2#Fu2cabPi;xx$Mx4+ChfBgmN?xUaF@b^EGp2?_tEb_#Uzg8TfuGnsvI+fg2VZqwa zXIL#3>MsvRpSu$4n{8CR)2_6`!}4Z1L(1;P-#k}7+22Hw$gq$tdE!p{E8JY&pbxjZ zMy#(L=Z@h?r(7AN<%7-mmQLJ8snvY^dsVey=j=ul2|j%UH!Dq3k>Bs1pQKqF*^Wk~+Jibb#y~aJBKTx&i2znLva)PrHcK$fCqQGg@U8j7l zq}z9YSF1s#2KTtgWdzj_#-X+g(TMl}9}=;0OSG&U`gCYBJbxBjG+QHAr?&zfEiy_J zG^`8XU##gttOS`22QxcS#XVhMG3w$6s=iSkzc#x5IasX5=wL3uJGQInvQa`GDi2eUfu&1se9?vXf1OyhI-1-#dU4XF?!j~)Ok+S1fM2t(Gm5$CW?g zf{>MB=+qtSESg871yVoTN52|2{HYY|n!!Cr5vO_FTl~&*U-|Ow3*M2;qNQFJcM zOz9f2Q0(_0sJF2jD1(6g3wt>61?##yfFLV0j}KXlDXb5!vF-hlr{praM}jC;Nr`4r zE&bp*e$qDQhBXy=BfdgbFTJ4THD)eE4EBD~)V-Kdyyy#M0e|Mq;w~TmerZ;-qz?Yw zbb*(_f6s>$-!|N{kuBTphg;g~GeiOOfO1s2V*obQLoqtf z;@6<$Y2$5zyP%lN@&{S8FY?8xb!9-%*ru>rHOX~~r4|BgKIdM@e9<)QlW&|V=}1^< zpL^({wYqxj7bCzgT{VvH*WaBH=v5{Ca`W_Bro>ff9ZhXjG@6ARZt6RV=5Ga|XOR)G z4ucoI{m1l|vA&Ad@M0{avm8`mMDjXr$Wgx4pnS*vQkOUYCg_^|NvZh5BB!x`9xw9e zI|ff1B{Rf4-Lh`pg#ai+c6oztDRVWzv1%HAp+@*4n`_?C4*N|@o2;w9;Rq*7mBGmF z(q0hA1;&BbrcT9bZtlVkt|&CM>WpxeUY=<;%`-JCF3@wCjxtt%gMC=n=(#hH%-SBv zXZ#fkk6&(bU};S6UWJKRW)`=;o#riKDT8j>Y}tA?$fWJ+d3z8;?=PD(fzF}Yy!Xt! zbqAj`=ru}0jX@E|q_H1}H&9G+^*rIv>vLFuyC3V+~U1OMsFJ2OjlvdG1T7QO8O ztCHJ5@^RI<@JLsc#)IL543QNm06gH;D>R!8{2uE`b^MLGL1UI(Vy>rTqHugaL;za8%8g z>oPSZXpumAae$c-9S!sV0< zzyp%%TaQB%_&}uW+R!*3xmC;4^h{(J>=G2OYUn|X#*c5QBk9Gcd~Hy73J5~P4|#8Y zdPoa8_+qocwe0!(Rz2pON3FlB!?$G-rMK~w-3Vggz_00iSDRJVj~8)BvWN&s2QbXc zX~BXI43vWGng-s#gZGHm+{#4!_zt%$(#;hJjnpWzo{T;<1HqbO=vYWNh=~r&V9Ee#0>ftoQ1=Dg<>G0>h2+` zW;T_HcH@-akudexNE@5LsP(aOIhTI5`0^vht*t>*ZPcJv8t4ikAu%_6NCN$N#$snX zhRn9`GSzaWbsSm^H}W&PXOVhczT(Waf750c?chFqob|W!xK-qgz$aHyzk6i})d7PV z-3@HZbEyXFR6dG@>BiQ+oj$fhC!67Gf`?NQ@)O6hRF&uU6&<~ORSmNzb<)$w0$>asLdwC|B60dR=eCvUV5Fr~zXvp|89slQE7ZZJz~<7z38 zqqtN+%LHF1Pe@0x0JltZpN+&8NLf7xNoi%_ z`x%~`kS2m#nG^hm7SNA&8BhJpe4g&8T+>hzCME>EW_wwi%{sXSi_Q(T)iUQ&D9Du; z$)Hdmvi3F}2*F9Hnp{r)sXPeh&?E%J+qmQp+@UwLlfyQrRhX@iy38UV3w`uHjk&RJ|o$sQrh2hx4OXn{-|iyHxV{T?*VwD(kMe4O*?G z&Ymp-O6_F-L6&|ikIur@%lNV+u%;qAxszp!KD|l0ouo^!u~*?LdxTsb};Le zG3(XnW4CQ@{SHU%-EWPriRx**kG{P=-ft+PZ!`8Yy$T`W%^i{p{Z8#6QY7f7$YRQf z6ljv`BCm7nw5e?8HTZ2XSARw7EaW`|Uht{R+0=xGvLfJ?y+I>ki1kO{9ZhBbxzQIY z*~dwb5?T!l?oj4$7@^W{dSy`hX`9?;ZkNLK*|Rq*a=a0;;~@wfz9L7ssX?k(+JOeN zv%scp<3*q(&8zU6;~kJd%k7wNTCPT0W?_Ri<}B}2Zg9rs_J`6{H0*xLc6MdiZzS*4 zS>_oIkeB@HEI4(pPB;s%Z)I8xyVex22!Gn?aGQ zzI!{PsnfDs6jpt-nb)DAOP7sjkb#Dv=C6@fBwJ|14=s+6938joiF^g2#01wY;t17p z&N4lc>FyM1zJnk+tyba6jqssQC(FT8%~L3)Y#L?U1zn3KLrl}5*&Fap>f9olAf;@8kAB@NkQgc@ z1OaJ3y4bnA0_#JZEr!q{hV6t``6^Ppx*G`{a-LP$RG+FOG^Zf0~kvE7^Up&#hj!9to1O1d~Bw&R&>Gt-(`vGSNT zxuRXjc8^dLHWQbV^X*L8mF{%TsZgjpt4hDAe-oLXs^qV=y6PB|?eH2FL^DCEu7*9} zj7<(CPH+<~q(=(g2rmhmEKn%XIgpIX`*|fJcG$De(AycAzR?sfBG+s`F?>*YaJS|w zRVNS6U~Aj@DsLg?9*pyP9r)AOX7tMFk!YO*bdiifP-LAIHI@`XmpgHTyOSd z0XpGldI~;n&17Lv)!{-Jxxw9(BCNv~>dSR50FQpTZ=zMC75Q>FW$i_Xiuw^sQ*xMl8sdo?Ik92^@%ZIR9`1$j@{s?8@h|9D=SdCH7 zFj>N=`bY9xw9X$_6GjtvS;=HA-*tQ3Je-auzuI2_ZOAtDTsSPrDxV1eqVqeD%rfrr z6q)=0(3wpw6oPFb{c$q}5e%c(srZ`rB=X-xN{qiG^&rGS?V zJbu&J)sr%L_wz}hNGtK0*6ig{Dk^pe49w|AdKme?dH|`X5=IBdlBG{eq4Cb4SkcOQ z_L>R8JSX#KSm#zEO?pVkFAG##v#wvke#++y%5_!YJ{)e=a;!a!Zy!Gl7mE_%@r)m(X-$L0l+V{OPo4Uig!Ve0#oBK>2b@VQblUB01&s%8 zCE4L_0txQ~!ifccbH0O=Y15UZdsQ9FRJs1bkxpOqG=^Ck?A|IoH~f?twMeGB^w%CG zB{(og7aBlJ2NYuor)Mi%!(6S?rtjX@Aq!C<4|z~5`bKStg(_;Ya9UCI*>B=eyTxs| zaA#CV2hKIld>3)=YPWWf6Kb$3ujr4l_3+>pnli#s|4>&@Iyt!e3SpT95d8WVDC{rz zuN4Ay$xf!nlYg!7?HN>#6_NfTY#~|v?qtxm=r-^|E_Iqql{auHLdUH#8b+pHoXAxHONhcANuEl3<44)qTsRQ*mFl=B&p z>T9By1hZ!}r6T5dI^Ck{O};|I;23cUyU4p&Fcv%q?Y7$WvNYdvJEe2uh4)?Wuha9# z0PVnp68O%usZlMt$HG`eWW>M+=%Kap{vbXV_Ky|-Q_u7LYoh!>L?*kdt$S%HLF(%t zGjhn>!Wpz$yq%#NBs7p@GKgF!O8>=2VEe|?@>9_0+bo^%b$8C~*~Xe4)|0TJsqWFL zQPVk1&_p^7Nh_-PLs494T0XmOYyb~uyJq0wykAn0&1Y;eTWDB>Tuxv1e#OxNvBe!W zfe}#m&{}m3-VobwzyC$W46q+a^a>zsrE(A<8%1%|CPRA{b6o4jDGNtA2SKuPLX zGminlko?rswFEr^6bHmfNiQ|D?$U9Ho(|QUG7IP(4yU zW$FyypW^sa6bZ=g5M}&R1IsKqYd?7#K0F^?9TEVM&$s2$Lh`?Qd z`mmP0fjBT*VU2+$K1vDIKOZSv_pNp68U3!};@;lc!~k%8XB68c`uTOYlbtywGvoEI z$^pHFST(;ZTW1u=)k>9g#ez#8?ezKY!?t2}nv$K=hHn7al$NRTfJ|}yVSX{*%<=eT z!QAztm13A^S?n7blR}p?)z${Bmfi!wnY&xh_5-4R#2*YPTxvsz^BXH&(wOKTx;SgW zXtD^v0~TPqr3VCxtzHr3+`Cm`<|-4jv|)tC0=xS2pQheT4Z~7x`?@qv&e9=n1Maz5 zJZ=YY@(MV>d3LL+qY~7^@5MI^zXGtKE>e8mynjdm^>degn1byD-7iD{0hce6S&b-p z;%p!9fFtPjDCqcX0v+8t;okESdTfJXp3cn8(N?o{`xT?Z!D6B&04_&KpZEW$H?iuW zC&d(9IB1dRxIL?SIZC%FrcyAP5L*?W8q4tANn~G0$ny*e&>DH2_bdo(S9m${>Rdsk z4bVETth~!m*8y{uXdz6Bta_o=dh(6zXKexN4xpGW(jrG$FC17#HRguaDxR!yIrI}> zHPSZ#tkL|Y{SnfRj0)T)s{xB;t9A&`H=2myO2kqdzn1y?(OXeF-g{Rj4&JegaMwWiJg0C}MnoJDGm*ML3`3LDdT8ksFfoRK)KeZ1`+C0a&6)TKztcb7f~*~yiGiJtgjkj!VMWupbT zw#BQ!%5&>u7dMZb46_%Au|aw}WRj?z?@>5NkY0^Dndv}Kr}+jc2_1GD?Cvk4(nsqc z>2QA7W1v?tpC;6C;o8%yp0ASOllO(xlEWrzBcQG_j)rWK5ru-8;=7(+uH&ovq-nGK zX;cN>CRlB1WH{XpvFQ%QK>Eke9E3s!A!dlrn6ZXRyS~^qO5hQK*a5x!O|&UaxJ8)C zF%t?vZ8!cHfqLuuh)q()cU6dz@N!=$?li*wU{)R3?}=-F!NR!$5+Ru0bq(MIbs?&G zXe0GTcDPWupOQ@w;oA4|vEsAF(TIl*MCNXt`wngOL}>}@Y7WXR_O3&lM;mSFixiGG z2n|5=pzRQC58xk{{Rv-?iEq3UIQ|;*^q;WdWqIw+3pC}DProwKpC3S`Bca)X?TR<) zaTKPRS48aEx-c@U>SdTZPRLg&YU|zcyMLGK#&i+#<8Iq!=beg&?ve$G?dAphPK#rX z!=nmbp3Iji8N78RBh>mH4#Tw?`4eT@kKWr`2vaW@vRr^lBpRnQ@pFc-2J=1r$~8ts$|aHR7;+e_kp44Ov{hv zgW8JX#?@W1^RIf(-s7$Neo$Z1S?qqFfsyI(bZ4{Q+><5MTmw-B*XG7elcc%AoriUU zh5MhHT~MQ)=|5`rP8M&5OzQ%Hy^myk#R|MlT9Ny}wc zY=OK?9+v&Bp?*2IbN#X@DfdUjD$qjAMPa>7MrE?Cvs_90Bc&tG;&wtbnZjE7^C#BE zC7Oxp*n`+~-oOhN@A?>?mvYX%(aB@a{$(_VM-Aj) zctE$AXN(6xQZ!mHvMylmqX6duS%_Dq{I7fnc?URD#HG+C>w&g#*V5pOINH%2Ni-lb z2aq2eF>=Stv|eG%u&^G2!2?QDJIT48wMlTdt9iQX==Nm3BIf|wtmUHPUKYdpGQi^n z;aa^v6S&xHLqPa~4|5k?;VOq4mzmJYBC~_m@p#Qw_if346(9&J?yL!XB$%s{@DeG& z%+08b%2mhi*(#<$G8s|jM`VTnlSAQE^-1*jQJ%a+a?Qh4?n;Bbw>xn(L994?n-aw?g{fy~wA^jSl>miIYY z@_Bb9p4M_#aWaA8N1aJa==?2JT5GNnnIixv(9~EefrPSKjKP#W1vD?oU=^5$bXNwc zLz)&fPXRnekUNhn>*88gPG9cq_!kl^Qu*{JZ(IhK)PQherpF$_Y|RKzXy?}g=ZWsg zN#294q^LKcRf7XcR6Ta}ib=X>tAqr;Rf0aQ0fd8M89#>X8}UFNe3biwCa~aRRYGtW zz9hpiY15Nm4Rswh1Z^+Oz%tb$;WE0&q!#*C<6ixmeQ)jmQ@c>P$!FQ=yh2-+6fN)R zPqVBwnk$FlhiYlFJ%njfLT5pixyZJ=5n~cz9Q8`RRo#^8qYKtCN3I-=-SyEd%g^*Z zj#e^A#DYh1`rlPmJS9k`RJSgM-;b3&1E>9pN;>HVt<&hw$)fwV=I0J2I-KysG5|0) z73yNi1xwchP%4in_bDQ4Lb;R|DV!5eBi2kfTn0}B(;EA3E*Mm+{p#QAN9y{*WA&)I zy`p35Iwzr=NVziRX%gZSvhA8Xfvy7u-Ji7VxZB9#{j`v7xKg1o*nBs_QnY(r!h(kT%;9J9+~&UHB>Vv|!JVgE!lY0!ASd|Ub#E)MgH&fC9maZ%Vz9y48& zkoxp8(M@x_$D`s_-^qBQGvBP#!{j?L(g1fnpP?+%`=WRgfQ*BbpLA+A7ifuFk@_7_ zg^gQn@M@AN_YAJ0%R`5-lTt@#w;^a;U$=a|Z=$S5@#Y=uXs{ac>_ih<6?QljQV30K zCx7h9cCE=r*=k=9sx|Qm@9H8s*3G$9Qs?Y9uh|FY`SXGcG}H*k{Fv@{WX zyc3VN*ju9JMNK{yEkN|qj3cS@?fCa`LGrdPQ#>eZY42fDm1?W}n1u9wkCbLSS;cWs zNp6jLpUamT2XGag0sTvPFk1fQ^zH99wW04Yx9I!(%ruypQ&nplUIjXe7rx z8EC7$r>K8~7|;i>`4|r;;n_@uD?laHt}g9em$VHjZkB)BsleP_5J5D%1b=hpGilb#?qmSlSHMdMhX0J z`tQR1jpf)~OLMI&$}dF*GOBH-+rKmM*dzdl6AJ`E%$3aZ9p9absKbp~8rRD4sQny{ zev3O=>9$ccb5g4+!zx_*d7>L`1tCc)6r~Vn8`+i zhn$*d1L`PWc52VgAEJ~N=`h^=o z;2xlI?S$pw_ior|#@bk1Hcy=_i4C``NIfGgozP->(r{#!ez6aV7lS&BXik#CpAQyI2A-j)hSVg9@y-tgc-@wwdB~Z97M$gg# zvU04s=Z0!eGwslVjHp1mIF3LJOGx9UBf)${L9N=xU4Wb>yk}hFNwbAVX*RBkVdBfj zw566@wei+ME{vC9w)zdn#dv)qnGn)sPUBK`x8`(xfCfC=tqJi|FS$N&&2&{3*Wh5q z(mlGvRaA2*P#`G~irEZwfzPRQl|VGk>kpbiCv|Y^pFZ=sX2#sDhe(I*(#ITD1%^9h z`bBb7hJ{glmAi*$;h;=g_kEjDfKlwly!nzC&hGZCHXf7WdQU)c4x@I)F*Nqp)?;`i z$AJiIgdHZ&@}Z|}+;YZwujs3sEc-wi-rP}=O;U!{QlL%0PGU70GM4C|zSYuC8*YF4 zPf0KJ+mMV>2pvch0f8k1%9M?9ucKEoTKK6+@ljC^mJN9^O5|_MSYY~-6dH~S!C1-# z@?`_U`S3w}uDK9=a2ir=-{e5$w;KHkWq~Z%V{TZ5IUT2T4g_3VnFQ)$!Ed$FQONxS zAUOm}!Gwr`%oT%d17Hyu1Es&CwFMXjk1iVjq|pOcBGO$qva55Tl}I=&qKZy1q-Vu> z6U8;ntbvm3AV!+{9c2QdUfYhSi^n<<6zvhwZ}Y1m(_*u#xev>#So)&#@=cCJq{Rf~ zWEa^m%*TQi*LR~&3I$J$E{+4Et7m8Wl?h#L*% zwt`YOJUTU=ow@b-Rgu?Dc$-!%96Pftx<79EPbIE36i8u-%VhbS1xP(iogQ$jS)Sw( za`oAj9=-m~)!K%C(ezlxyL=vakwL;R4)K~VxaaG&zLHG{iaGN!_(Z@9+us|ePT8L8 z?}R?|(LF-w-g=%Oj)nLE_2oKo+7x6SRVwhUp9g^;bs&5-sVyL~G45^6}CIfmp;M3fovL`WfeaEFoLH}ut8KZsab zlz?#n7n2W0RI(Et>IR`gRDqu6-Sg4V8`kFwD@cWx>$OpXS>V0<=F;LN&1S8)fZLOm z7Z{c6Ss#3Oyd6B_sqC5=GQ&YxU|%6GkJn@I?j`2izhB-mVlQS#!yQ1#4+c#fNp6Lo zZ8e&{Wh}>XH9S?Jl0Cc&@TUU`tC&!+=q+^1k3=U+&Bz%hUEH=sEFH$=Tuwy? zeO%bwOAAFlg^?MyIieqVZ@(f8>sLOf zkZ|hxrc04_kq>>EkSHKBi_c6452u&TGgNS5X!zO_8*m5r{*}i2ZVOHt#Gr8rl`YEi z4EvOPO5UVnk80bKeOFUUO{f1co7)?$3`teTY^u6RR~l2Qa}wtOsrM^-)255Z(}JMj zMvY`mv1VTO*Eq%%9nDT3vZ3`;+Z>N2lgls2?k^CwVl)LVvR7f7XM3-2vURSqe*wa& zlVlOh$*)k=tmFi`F1*Q>b_;ir4cV#Gn?EDeW%pv&WlI=%5W>)q^A1EBm|B2!tAzEp zKM?NbL!?W0=~E6x(m>Zz_)M)$M-G-B@nDp>!)R>0^ifGMy~EKbZT8`cxJFNKnzmNyuUeSe1rm8Rj;hDN?N@U*TnA8#~Kfnue`m> z`3SulEET!avS|`kh-IHZ+}8O&tvM4kM(5$kBr;r@i(2a*91b75GQ`_>Ilh!@8h4le z^1Xo#N?d1d0uY3>)dyvL=930rU{qVN)_U%ucJDQe7%yLb-KMC;OXsbPdHvmlJA#I# z{nm6;+VYNg?nPibAz2vQ+^T--u(-nZ&01x+LS`H8Csac_(IpgK(*Wp1%3C z(dKn>r0ygCBMevC@I1Ue%xa_QIA{49+_jOzWbyRHD4FBYSRstNk%Xyv{aYkaGgI=< zaH{#uUUMN_AaoIP^_iC*;XIqVxUuY9Y&0AwZlm1@@M&fwDOwW`V+LRN7<{;UjTm~9 z6ECxSXtL~{*kwJBk7XFc7^D6z9+5%Yklm-b%^fFYE~7(Ggb||y*;5?! z2}u;`mUN-IFZOA$hqr%={9#8!iC~Y_qfp>|`#Yof4-bGc7>ajV>OB#)+$MzcSbv$}~bFqGic2tG_7-mcRF!q7Q%F-FT(y{OBebqFW2;%Mz`ledq4KL-ym zYj8OYdl zW8(Kz8#aATmZ}hMv{Nnd~$c19OkN+~7vM@31Qpsh8L6)(W<%o0f^6lf9lXj2eil{w-1=_R+%k7(owBhK#UjALUB- z~dP$D)m!ckJ1mTph#;Q1LchZ3KD9|-0-`{=%`l;y z*Is7%l116)x(Iy)C)pqUwD8`<^uLWb*c;1;geI~2U->(nNA}(SP!W{(Y|OL*E0hWH z53>}hG%eTL%9{~($eii=oH@4pocUY?Z%XV3Pv`SjE$bfKhpLJ?dOe1FDoBv+ZMOuZ zFXe4mFo_zTwEPteD4&wqI`*F6aF23GWc)UEzP94i9@~sB-D5sbB36YZ!{S;I>IQ_0 zrMl4!$W;;2x%KgDuIh%}wceI3(t0I97IbagtWG79dE7GUPw)06tR>QO@sM-ImT@s+ z3z!`v@AcBK6|n-1W)c^w!xk7RMK_%~$-JOa753XReDZ}bEhKb-(n3XQk?^12FnJ2yC@hUaa;^GjtB41*Lj;Mn&vf2w>M`}-NFsFvlPY&qtlz^@ zhJQL=%4-_8I;oH%{|z*HcJ2hT+ILs8YTs$pe8B~&q7AI`d@t!>zH7PDC6m`!@sl!# zs92S6+XY+L>#ZNO(eQU@pU{s;!F&Mh#BIbdf@(!Orfp<_wexYk2@3~G@#bYv>RAk~ zq)lLA^*t$lnV4|pSMTkbeaSUDa&$>kyCxmD5gC}FbYbO_SeW2S@+Kc zxow-W3cDGz{Q3Je$I`S`(+{gNQEnSmU*+)-K5T=>nG#~${#;;GD+2zCdSmxRNh}wt zE(t;uFwozx&Uo=PJ!1s-X&J0s_q-HV(?O({i6f~G!?Fy#aGCG@$r>k_&s?Q`z)Jpr z#q@r0pZn>@FM8nppV!-94E!6rR{!sy*b40;k>8V#l};Ukdgb4Wey%&84^!wLNP0iq zDB6?ynhUqH{22c5XPCFrFm6m?p~H6C*c4q`c%zq7#B@|AZ4@`X+a7tl{owJ6OTx}W zs+X9JX$YU%DId4p>*)FZ^=IiO)r|Vc^>GWVkea^P7zv+~%w~WCg>{Cu-3UX& zV3%*8@p_DLHM9+lBUGEGMrLJkCPn~H_w8SU2ybBNXU5(BN+$njOl?FMwy&e{zJU>c zWJqfBkrScXbSgIu_ulPUiU@JW7c zT^2Y+ z`yUhV0)ZR7GZI^cDl&d-1v;F9CYS)Y)gB53{&XF>z>`2pIIQSc-DiLN=bm@syu6kw zKV@P4*Vo5|^?W?$BH-Pi5V%{wZj>)GpRAR7^&Q{F?PQs!iO++VM)$Johk^6PFr^J( zDIxWJmqal#Q&@S1yAf={&|j;{4;Gb-ht_E|S`!hcRl%z1ZU^3o&DoBkv+6vr zsbHOw`?rQ9!8JNeISHPvj_ULSJ{O4MFfpCj18NH#F%wy@}-g?dE;xVDT zP^!D3 zUgT}jjg(VflV}5ZBN^ASjp*kFz^OKSn0r_$8gsE$XN6*z;HejQlz`O?goAEx%UG zy=`cV0T)R&H}q91bJgKoH-EdCS@!QO@n@5CZqME%pYvvQ1i~spvSec>|JRW|a~_Usv_-f9ikuKAS>#-KBO5Kl1-NP&qmbc81TqxIi+ z@PGC11m@3EeoebDiN7Y{e>KG46A0WV#PNzKqS28L|J6z|;Tt zlKy}ApZ~wnz3Kd3`0uI@hVK~vS*rh+mvAjCHGgCR@n-b@Wt2br1xMsZ_XXXS_yz0V zkC6Z2+W9Yj?X4!T3M6DA&i;$f{-0Lq8saM; z_l!4yOXt=EAxEwg;CBsx@x_S2a(+_<3Nj`2mU;4z(DMc8l~qiL#&X$Gy_9c%_;!&= zMl>Mr*-uYKDf@R`;tp@qt$f~;IJ7PeBJ%}{og<6|nn=B+!H`59x3Fsr$o)(O!o@s| z5pEpcRVWxAq{=K$l+*hbUt=3chwO|~TMjuNAT#`E1S83Csq5MkGP(8!y0-8+*W4@&Hw9$fx>RP8Yk5wy{Q z^3WwgHUtQulM6jv#?pfte_PtI1b2MdSLX5D@N<9eLDO6&iFUba1cQ9w{7Y-d%z?5XvNW;5ZOY ze0l&Lw)X9fXmDtx*HVU$yQ#T=UXk)$m8;#_+Plo}>d-Ere3HIejFgFvK|?z6%4NjA zk({!_5^Uo4wZi}Kb$QCfvp$0h9<9wj>tH&eY^ki3iR<0sI=MtyHQ)_Rretvc?ePtb zQgij^A~ii!5?)MI!bH}XCC**tgh%N?zF80MrGs3((EM?CxpO?)yf9SzD6d!e#@wz) zjh{x>3C$=eQd0$sns|8fglcPa?gOD!PIjKH7ne&hRdRoHy~@3dRiNaC_)8S>)5g&O*09zIMCBd2 zq_nETq4ddb{+>%#ipu*E>o0x7PM0%sN!)Uabyn}VR-Hvz+ZJy}T#Xh|&lg+8Vt7>E z6lvyd7`v5`ON6V${<@axf!5^#@Np%b0XVYq_~@L0k-9l;at90bNZN4I_vwI;@%~~? zM+EOl-a0^brKC$vuGvL^&9sqO16{X@>eAe}DE-v< zH~MF-f{EH0A1j)AY1;sESB>$dbBgcVV|>|OLRU;ig}nZ50+6e|ztWw@W{$@ks{PGl zK1Z!(=>%$8F0i{w?{+$iMR`h^r|h%0SOeqnd-+Xeahs(b?cr)Y1ebwkG!w1#S_=$M zFJeA}kTgJjeN}z4DV(T7vEv#0hUPL-=;=e`j4AckNeoW~-;#65hue2)?CzMY-Ds`& zZdALwv`v=;x?!oUlADNL%g%TI&DKn$@h)MSvZFGuGFUm@S1bE+Mr1t*$OF~w@n0P` z&{fTb+o@>?cpauY< z;LJ&3wN=-6#QyKd9s~ap-VuraG^$>6uU- z)^=KPhvW?pS@H*exhu&tQoPSEsty>$7>n{cLtlKM_`n{>vqr!i4a%7(FuUsvxo zfbrELb!qh{Cu6L}{d^*hA4JwcU|lozVj(rL1``^U{HPGLL>aq8GIM^6;!!C(yj~c8 zxgVAe^Mx_rgo6eTr;GDP5;|y|9~^$irq5zz`J>e-RKV#8}|@pp>{#zho8LGMO68v(W3XC8!l^{+c+?jkW##8a z>Umw3N3+In5>YL+q!(+=dO>UrdJD~0M3)93TGsKv5zS!TT>tZOD7ZEZDLw#UfC0hp zJ$muDB;w|@;#}v&f3vQEk51r$oa&8`du6hKQNY$PpasM(c#Tb`{K?e;!q|3yr7~hp z77rATh@Zggyg_5{O0AX)qo)hiVXV&l~Qx!&n3^fzwd-`D6M;o_-_H7s+;jy-T;T|l@=$uu62YaT@QEJOu@FV z=FjM6uiGn)@CziPMF29T`u!gh)br5{0bVdLxBG`}Y`=zH0#QE3dZssIb0E?pW1+?b zZ4lf+&rkVRs&X3eLTq={<_qmA`VUoWUK1Gnxs_<%)A=zTTo0+WN%Y43{#5&ZbX~oo z6Fcki`5o;dj>I{SCUnWflky*80&XWjbQF3PNZ4DZv8iNF>z;~)g~8Cz#0Wxu{O=OL zwc)owv4DSbM`-+RTF>4pQtfl_cttzZ-23%+5FMJjGwQnQxYf$mM6v2s_3|})&Pt@J zMh-*dlJiIcQ{pzAx_R_v%xfg~7`ENFml+%$&0+W)(rRV8We{HkP`fq{0jp`0;KOaK zLahT-=;^j6ZFUK;;hmLPM*No+z!17|*(%77mFnFK$V5wnAo1^p3P9^d1C^NH^e)+k ztU`zG)|XB_KMa|K9|TE#=CoX;2l@t+hL)Qq^+3xh8t`I1)#nZ+=(~$mYA_&VOe1jU6 zJUlHyVrW#vhmy%t$1K24ug<#yGY6liiz$ssY2V$f#tsx90~$0jb(y7!WL^(aCrwlh z*bEOpmL1dyGnu_nJ$XIE>phx8BZzq@8^8(DcDrLVF4LJ_&%U>RI!!kwgK#HlDg2jb zi{qDqi;IEMICJ;6-XrlECb8AUPAg1Q92N70 zQCZtPYGLqay%DAmseqO#nXG#x$rSeWiU^(IJAUY3AO(%b!}@QJN5YCxBm zjB$AJdjPW2h%XwpeXXSB#yaWA(*cnO20js1_GXa7b?>6b48?q!{pkWAmcw6 z`179C$8*3>zA)JLZhEV2nb+(4=wQjuVkF0Ps$mzvQIn%zk_HVsB(pQfb|P`fY3QV) zBlz}dsRC}czhk*w#Vsq{rdYI+_5Lnklo$Cx8t9B1NrCP&Y3TBvpDrq_WX@Mo)$eo4 z%d|R_AmCW8NY98d8l{C<%cafnXA+^8=WAU(YD+(M-*yy_cM)fd0_f#JmUW?nre8F7 z7hpON$~(*kqg^YW?>i<}t2J|g=u4f(oK9uAX6>G@-N-1H*6xGTGQY2Wvr?U0pz4yJ zQxTvcbWne*#(}}1X|mYYdw`@*;ThWeSKB&PJlfi}kA5Cj`Mz%5<<{^pvDyjpaz^37 zL;x*X&^P*0>iXf;F_FyfS0ZZShw%(mEJM|n+cH=2GxzLt*HiC%0ke|jbB=9sfyaYG zf$HybbBif_(8GI3Xy(zk?)ioS`NzVZ9wv}hCA=Fy=pMRe(EcK=F)Esio3i&LVa3?tIrV4K@K^7A42XnDuyDRG3TrrkMLIs>hQoJyMsfH zB`T|X(eB3&;vL>;YNu0{C#|BOGtmIlq>IBkiNFgkUU2b@KN9yG$xIR>h5-`A3hP2*&q zE_LW9?)|zxG+lQd8~A3}Es}tfFlTsAV%PCK8an~Ue$BC;bde>!=9!)QX)C;fpodrp zRr=!Zzxbvsb3~J)a2Bh>LsUXbGU{VtZPbf3?O%O-F9~jvt}n-gFc^Cj z6a$=OHHx%b>R0k5_r<@9UULG{wG@?A8J5tx5dDqKTu5~Lha~Fz9UNN?1 z(iidO8obyJZZSK%4SC59%TL5Zh2Vb>!+v^e&XR;PeaD8NLPxO)Oc`-@x{JVIu3%kV zJMe-$B-5|HL13mV4Dy!m?sGENCFcqxm31C+I0Z~RzKbV2zoMge5HpCG0auix6Df;P zlYLFONdp54b?$=3- z&>7>1W9?=Ek8nAe=i3weZJ={7TfyA@0_y%MO60>*y{LQoO2PH1b#?2&QhPeT{f}Ya zV^=1F)#mf}@Zu_n`EW`ppi?`EU!S_E=Q&U&KMZ=bsDFXuOh&%p)|4~R&h1x$h$)UO z)>%@PXIPp+v6{%Sh-GK@CU$w58f$1!C(tS|@P&qdRa-pB(y7)Yy|x)$M!rbdM4z|? zCN z9nD$jQ_t@XrUH2C*#LDo+PA+G1g{hkt{hg12x*?C?GyaF{RGnTE$lVz3hJ2;Y4e+q zH)5oIo|7N_s|HV$i{gX4Tn@7q{rhKhZ-LqTG{(u>$hsOUP2flpqc` zd+j$}>sgz2-X4!&mYbB1O6^?7EyYs$=LA8Kk6`4J_zlq)R+NS39NMz%`&kt)nH9DD zB<1=JrtD+3u8l<&#&_YNub#g+3{QjYIrQVp+hSFiuIv)eY#+1D$2X`mI<&fhmG8OT zAB;RVdT$=gX0*fcQO8yUx~t3ht%Wpca{Fbc-*xPQ?xY`&fFMn}kap2!(H++i(G8-GCl;i|w{(V2&HH)`ab6u0=x8U;`Ya#plYQ`zMH%D&mYRM>)p z3%w)4;vv;t0iIL&F2#7yr_RPKy8Bp6*0VNhcD40Es3?>iee@6b&i7Led|Y6rK3QG< zhS&;0nD!PS{=*X+3H0d7RA!`;()7N0fiqRXOetKG zRnC`Wr}fKr7hF5L3IBOar8q|udoTr^4<&6o-yIxQmXkv)yTd;$RSN!WB4gABJLR$Ip_RfR{StR1H4YCx*Gi@}tz6#q!+yt{Lz6r` zyZt0(j_}0imaTlw1rIH%VLIz2g0P~Uxoq9mM%0H;Eb@eE{7qyE_k5{bF~PxiaBcem z6I1XV6IC3!pi>^6%u0SR%@%WzXshIf@Qs08Gkqy;I(M1A{$D!(n}8-~)1Aru#Q@&2 zw(ib%F!Mv^??m@gK|nv);DwJi6{CJRV}^p?&TGvYhwS!*kM4ci$-1>BVeh*&4>HyR zaeEHJXTVxK_8Adh-aR&`dLhEEr!>p&uT)CY9y~9Tp584Zxo0`i8Eu(U3F?OT%-;27 zd@{1kZy&lec+1KrNA;`3^pZgmemDIRb}jMFs5KP8A{R;GuJ3b%?b)Aqad3s#(J1ke ze&Se&lCYcInPl!&iDTiCA;ml3Tx}5f0Sn^K(5$g2zFTA?5#gl2CyzV-tBs&d%5{FE z?X!IPc%&%fg7vAoIw%x-uraFnWvb<20K{V+&}^W2YHc)(6TD^bK+4svd6Kw(FUe^! ze1JUzkZ6M0^Gj*g)h2CyCcgfE%BLnQ$zCwOg8I1vfh)V5{34x?X9X;tyr3S}2{0m>Q#nxyY9J06W^#KH;t<=H#C7rWk>8&VyF2G-fr zKnKhT?;Gf!3o??x_)*6qN^cy13t?kj1ERRW%g!20;^F(VXw1^nB#mDRnis*Yk{S`A zjTl%6oN0ld`y5CaUEV9S#Ji)8$nhd*Ub|mnij05fvlH7*str7BJntjkZcS0VO~bPl zd@zwYOlMrSt_}@1)&J_#41HTfBa#%xvS&R^kJreoLr$HE?N1JRK-p)R>EI#$c$c`hXew zkwLZGINe=(s+a|RMnZxkqiR=IU(SoI^c|z_L;d}ifO}Jh)k$BRe3*in<0WQ)%?YDU zt4@aE*y}R>?(F+6D=xF604XviE!Vga?fU&o{|9gPYZR|m=r668h{Kc=eH`d0oivlb zFL$?%WKDVk9K6wuHW72 z?}VAzP9I1ex`2h35GIQEW7$?MXDY0JL(v}tlD#7y=E^<>)XMtVK%voxE)nrWti_|> zMknO6eawdao#t%EWh=9`UkQ;237|BJ%ch7f1)#Yn$)bja`NtNHPL2)vJ>IieI3@it z72DlYsPYO~;jI?e`|Fzv+GtvCH%=7P3IJCQrcxO0_=Av5g*DeHMEZi1t`0r0d$$Li z;;oQKbF+PtVuGY>lfL-7Yer0M{3bdzZ>%g@dMs8|W_>dS;yr7F3jLft#*@#OMiaZy zuics!Ubx!3F9S9_i7Ctq6Gc2ufO2hnT`M1bVht_|9!j)- zz|-tv=qjfmS)%a6IB_F*e5r;Yv*@gkBlatXN?3TGBk;(m{(iOtNq3B%wI~`>FV_=d zwyp?p1h$%e=di8Q`t-qW>o1ezXK8`5ZiB<^xFpTYu_PwHWriuUQa1=8Lz0#>Uxj?2 z8O$ev1loJNcwgr_dL%Ga@C&5h2HEOk=$r*~4-(|fleP}APiMSE2wxN#R=sfB$iyI|66p4`=8+OcgoezM#0QRwZ@_z?MpcDu zSiyW&?|QEw_t%)QcLmqJu`QplEZ~8E^6rjX_)e3Vj>AzuUUpZofBd z#ljlI&U-&Gil7h-4Oov<7S*q-86m;sa!45~P6M$p*he@4+_Ovg_qr!Qn~Ug#l{ip9wJSTv_m#Oh%T*|I)Z6Jl zxYfbOI=U_}*kE@&=7(Oq$u&q*Z_!f;57%m<`t-K?r2B`VlVxdcIiQV1Lh$eJbEt;O z*_dIX%P0cgMHm*L=Va@r%aG4SDps4{M-jm}c(41@VdpgIFJbJ{+#GhofB=3sScv); zSlsN})4xXY(Q;O+FtVOH1|#9cV!6oYq5Y3!^iYfCmeSQ|6%mK@>0}NeYcnG2T$`fP zzdBj1clUDo)-d)6?-4jUOoUU3y+SG0+y(XraZBCLd@$Ob?5-F{rx+{agNUrRo4iIs z9d{UJ#yzuj`5o3{!mMoNJV$O8wdl zEK1mCb5&9p_p&Ka`P=+F+3YUfz4R9Hzu@w2)(wEFcK;qBdTE)&hdaVE zG!3=*mr@c!`gDx_r=L|ER=z3@W z?}%yNbn#$~t$hELlgU`c{S3-0(`?q<Sa*v;y%ldi~8i3EI;?Cj?1Vgxiki_hRZ41{e$-@ zp#i3)Jd3EEegQE7jzlC;-q3Sv7tOf%bwkk?7BWd^Vrwd8e*z~v?A=Ml!@_HOHNg%W z9b?=!YNd9vGZMfJJ7$(C)f&T%S{CMFX?k)*_mg==_oDIuC8zF%-K2{1+Lh}^I}Hi# z=UfjPHim3O31&)I8gdzT7Z!ACwN<7&oYo1$B`o`lz#N_z+g;cBy~fcSN?OHu3^|yP z(pKf#HHfdUJ3g5Fe#J9X7jMv`4n z%^)@kL!b(E`v*XOH`LS&rjnZGU*@N(o!HpyS=cNA?VqgyX25&7)Q>TEz2SepWh3sK zi(l7%7RP@z+g&{@nCA8T2WiV@d0)!qgdI?JN>iln(5>2TkOnB5;D|>+c4Q3K%cSdg zJP+Loy-QzC&h$$#&#-%Y_0SDl50Pt!uF)~I=;wdmDHEl-tAkVr?uv%JtwWxKU0WRh zD-8~aUyLomu2Zjg%dl`5Le0-+l4I+T^mp%qzRViR zEq&Sf5o&rA-c99PfT>n(kWMt`@9Y3L?hR!*+WuY3VXFGi!qb5L!c+OeyqUz~GL>G> zv9G8sLaC$3^VOGxwdb;(E0N|{4JW@0;oXdP-sh{0^fQJJOQMY3->DaBjq%>yWnODQ zrS?RhgeoAQA_4(njr~2~B)9CKjyybk=+!U@rYW1tCH?extvIlWt62dJsBEixyt}Mq z)V?aIe`|ebC*v&O9Cj~8D)@ngDiun-I~qPgO20z$XZ$)K`QjQ5^K-pzA9w3 zT9-kWOal$--l?(WYh3qIAoYe6<5+7c>ipFCVSk-y%e-O7Rk}fC&STZ0W4D7#F1r8d z{vB+vy{-A6$O`XkQ5c*2z_*klIuD_)#(AEr{#*%ArTVcVzQ;~aSZuT1M*0*_SppjX zTl`g=Hes+9qB}BCD?R*Zy(rn?C+laLrkDQg8VTzZaa{$dcLg7r)U5)d1)=30g5FK# zx0+dyHEhU8ZKNjCnP1#_O1#-K+Fjd@+jHsGG4nsm6@XXy?QsS&xeZ>}g!$G}Z-JCY~magZcQylb`o;;>0a9T8a8@gfJl((5crV)rT7fbK;s!|D{6YZ{B zJPFLUTHurZOy3JM#0@|C*MujouH}GTs64a|3po$_jrb%9HzFaSZ?~wY^^9v|kZoc# z0;XC6rU^LT`N?ZfdBDEB_DZagc8I`bPA3i8F; z`+S*r2Oj*zLqRIkyJHcdo33x4yl(9aHCQmc*-7T1K z)|Y~DspYH6YYs{kL``Y{!UhlNv%V0d*Ko2}l%XH{mKp(RUVfDM*|Ohyj6#e<6Q+8X zIx7A~oT**rMfdUoCYjAS<6`(y&V_4`M-}tS+ zp&=FRXZUkya%Lm(J91R%4Ml9@9V><(PAMPVsU^t}G9*fS6sUPUFC^5D6#{xl1-hD0 z9>%h2@a);V)X?fHoBD3A>&&4Zj>-t-H7l0sFn(i?+zv~(_J4-kHa4xpUf83xx29qo@)39?>7|R`20ZaOxV-n z>zZVOPE`am#E!{=k-~7hTUep?FE7=BpH#6O4WPbls#<=B(=x5xhK;)hm(M;OipLX! z%3sjN(sTd}UOIm89G{lU4g#jZXkbB~mf+tZ4AbvLH?8}M>xgOZ?NhtJu*ju9h^X?M;$^9XR)D!90~);qJaqEW5HDOF)S_N1u8#>-3_Lx+t8xv`F_A$|&4mlVL;kwS(m;oQH<5Qj8%rXh}d`LH+!immgor*Ro zVQd7O=JKQ}SyP~1kGGidH7}Ey*_2EqHBIg#<8|q$+ zM&OuuDO8BvI+2H8%7c^KQCJooO~xS(6pK*ISk3(Wu392F#AO>7U*s|5r?(p!ja8_S zIXht1VsI^}NOofZ&_V=Yz8G2(wW&bf_h%CAZYbdgg3a!8YAwDhzCNtgp6=Xe?!`$_Mc~R*-(N&L`-Cc1 z`+(n0G0ZOVi%!LSAl#Um{?Mxsp#XmXtwpg-CfYYAK`n93uft-8FRj+Dg(;p26N(+} z%W52QwMb0x{FqsX8OL1o3a#NJ0NL7`uX^#(NapSXN1i@lEIPKyWci&c@#^_B%{X_SOz53i3}ZWdlwArn&N0=h zT@phEA?4@95{dKRpKUlrj~bs5Pkvn!=XqE9N8yJZId-jZm*wmnW-81KzP@iOGuA=r zad5z)Om!PWU$1iBJ5%EIII9BTpLndp$sw>jlM1#)&>~GWE!J6ux}z5-fBxmU(AR6_OFfxuMUVx8H1T^30J@rl*YKN=A^NS;Nsu3)1U$ z1HmkVhnWR}l^0g|!NOR!4}pZRG;(IykJs23?{OYZO{pW!J2-xOGj=7rYu@R(H#GwI zZ*jaA*Pls@Sn!12($MKYxapneU-XNL6Ss)J1(7cQauXoCL;uYa?xDe{EReqQVL2pG zxWCDDIi3(5%@8aFHGhebKi*1Qmj6!Z3K$D?7tY!RukG9EVhd!IhAv^~+hI?-&KdxE*jB17*DHix{d+^>J~d0uK=ypR>5L+$b?Ny1bMGp-((+s{^C8f5 zJcbbIp*K!=?}zjba)MD{Q=Bvr*|4tnF|hjf@s3*Ya~D+u-lR{~$>p2605oc9J)pRw zv*v1hxtQFqW^y1msG&1W8y8<}<}i*G&1hfHyM5}N)me3TdEgHDKohy`Ts*gZYh9o4 z`AD9>@a_ir3E;*H6Eey{(tdSLWfCO$oF-r`c ziZ0wqVB}5}?Joz5^k;RxH{ClUB<=3J1XgCv0!rGWDn`?3GeaaODvxCn*XT_D{fwPl z>UoJ*<-Lkp2Q+)SjY3R9%`89KW&h+SkoCIVKI!Ek32!!L`9S@vT<t9dJ7i$n%rY>ygJR=*E4J08>Tauy9uH_8!Jfn7i3`> zv{t}|7l5+Y{L0MUKTd)A&2|cfjZ+Qq>3r~-WwGUpq5BQ1o(c*`Fh~peRSQMg28#1P z#BX;;#N3Ob21s>8>e|a3+xNNPXlTyz#NXqwg#P~|IRL*;3=v+}KV*^f{OMECad#2! zqDqjaK+h(8Zq#cc+~NDF{I%{N25?{YqT}tQKXqRPt84m@=EjDu13`N$X`)1ZjM1rl znZj~BQeLO+oHK!MGo_M49=~dA(4TJ#J>CJi15%7P_w&K5#mf2UZ@5yPH97AlD7_cA zFoY*AH4KgFwoFfUrN$xm2L^cYnas0N)!xW(e5VHYJ(WKA=ZQa?6&^vNKz$KmLGgtb zF^o*4*V&saVQc-QoqzSh{o}qMkoOvB707M<==QVh-leVG7uE5c+j`W;*72^%zbzk# z#NpCBE`Qb2vH-clNEQHN;O=%;{{ld-VD?8rbOHIZ{Sjm(pF41A$UcaC&|#mSPgjMK zbR}!>f5YFfYupHlnWeFJp6It6hwO^#D3vr@p{3!^(Z8O5#mz{CtrtXN(To`w zQZ%vvRX!^yq84^RyEr+OTKE1@T9xu6NV9n(*DanQ#!BwPb-;oL4TTp~P<|d1Hghb~ zbcpXvL9TCK7nK@0S$GY+aK9G}QJ8w15BI`3xRtj~t+mgLJB7f8IU=ml3(GG$AR6SU z7ls0!5(i7hDY`}8Ux%kkq_}Z1Qq4d7L`b&&l%`ll=_OY#c2YQYSN1ZTX9qw0A_a!D zak9fx@3^^LN3KS?S4vIH;%nTbMK$4Wv>nO4nnh!^=DTF;HCgEgM3(^^zPm%mKkC@{ z4Y%<}Zd;R&r{{X{{3)%7)pIftI9OlX%WHL9h08CDygtqK8#59YPWhO@~1;k&w2ZeN-YYl@h`EbMt9f16b zTl`1H=&@v8sc>p;n>!IM`vq$9%6nOZeC(Vkd0jZ|#7*XfN3Iv{uF8G&U_eK=L{!Pc z7RIee{`~`rb=~!{;F>QB=OKu)bWM-Z(+Q;^hE%HWhXb$6%{TeCvl#Z+lsYc-fh!kn zQ=T+JsLbmTSuEffnDfOnUEv22n7{2-!B|l{44bJZu&V0IfYD$I*v`LhQ+wKkyAdT^ zeOyFYr$gJEFtvNqH9G)`EPJt6z4d&Ist@fOK0_P}6977?fB^K7uhzPlvAbWJpGBoP z$RM7$vBp$p;3iDP6u4Kxh!9*NI;*;#F=kBh;Y5NkP-V?m>Bc<6InMggU-VOEsPLGl@_y`@P4rtf$uK)2I z{o2_)44jE^)K)gijSiGg>}~~O(w|7UZP2dXmA#vAfn84xcEc{=5V-Qhxrn{F9Hnp4 zCCdjvaLulAY_-6+jMPR4zsRQgirW_R;GLSKWGa_i2gZWBze6+iw`y#%@242umd5u} z--0-OxqIPgX@VJoBN}mY0&pV4cA&V0Qw>wlEvmPK5ppPN1z+*TipjV7iBGb)?d58< zC-q0O#st`DVc>HQ>*8kIXles<`QUOdK1nkaj}B>x|9p}1fG<)4|72_T#vto|dB#-J z0DN~+1m(|znZjrvz%U#vF(b2Wdo%krplQl{^p8IxN*Tp1uqtXX+HF_J^tv~LdXkMm zRW3y;zpOw&{Qv^sz7vUh6|%`+$*T?Smz)j|Qxhx1&Nroz>SV+2ePaNaT?TeFs%FD& zw^|lV_J`Q9)$PW^6ClYft{+G@XMG=)PUq=Qwh@`BB-i-FyAQ--?e(Ds84#dt^{ z4z>v1R>d!-(g&bTgZLC_Db|YaRx*7BcJ6LC?qM{ag3V{(lXnAsi(Z10bZXn5{OeTxcCR^(Ojk79FKl&}TU7ZkM^nQ| zPH7mM`v(^?k2S70s_9$&n)TMqAFuAC0iTNX9qup~uIob_!RmHiw{*xc#yYuzpF=8< z)Z+%3d8G(~*LmwEAQFo0_)iL!t)>q)cM%hf{|}cFh0sWU)c(NnH=OB})V!+)T7Q{`Lp3Xy$=5x>!tdzAuk^ zFNa(o{2tz&DH39Tdbyb-go?FJ@rlcK{;eOF{f}QYqjii;p|*^*K8c*X7bq(Qpq=d{0OSO2gwuuXaNkTn9QFPUCx6Z!VB8(| z6f7&#U(yADz4C4_97_WZA$52J2y(Z|DhN5f>M}avQ9G$RLC_q7_3}jH@sj(##qG%m zv0?;T_1?m-T^F7R+Z8^ILvLXCcV!1^oEdk=_@nbFcsy-aMH&{GqVx-7rS9#%vdujO z5CF8k&Y|5VIee?cy0S@pP~eO5Zt!K*6{hpTC%mvlmxu+&4?qQzLvNTOY*n%ZjAcD)G-rOze2USF9Z zMSVGD9_$_*Hl$HFSq4ldiWodIIqT*8R+ZQPF_u4LYhYFee9?tv8*)`P3X$iim%z5~Kp30UZ51dZbvE*1hHETtSzt zV1ep@O3f0PzDH~3f_`{vjug8)BgWTK)Ey*Bn(@&zg!js3sbEG$3l(X?v$S3pA2p3t zGuL=-)d>dpe>-n%MKrA1O~P{cU;&?pTr^J^ViYK{8u+Ja8tXDE<`UAYv90$fc&4~W z&<%M%rqp&v8k2LAxg@#Z+3|ID3F@?$QK25sxW^b1f4#w{mWzEN7q~16Ykp?N=>YLw zSN&&(a=Uh5B-@0J!1YFnO!GSTgC+Y!3HsaFa3F-b=Ujye0t$t!Xq%Av-MSqF4+=92i zohji%cD%Bl&WeJBX5p1e@-&hA@i9dUXsM&WqNf7F0e|s7EVt7tl1KK)rf#?iOceS0 zkT+Ofs&!tkw^JS;MEA}qWd@WYHjFti9?AoRJi=&t%MF%2i%`apr`3f@nU2xb2CcV~ zMQK|JO5hcL>2!d zU~3n)o$D!oDbK@+#0b4E7|}LKX(r%7Jv|qA;iQ*HSo&Vj$1vc{B*<%cA*uHTr)QI7 zSm^Z{aCt_2%_0R$1<|B9TGiVmBja;zAk>ZVyPYH`2*Cx`>%5<6jL5IZSNt3QOO$*e za%bl6z79wdx41i$|K;&)^i*2FM$3UMx6VQ3bu!anKi&blhv{$CHZ0R^Xr^o!RKTBpx57}$2mel^TXk>l1!G1;mK>>6g1jkH9c>1qCP$EFr z67|I~BMI?F(*zZ?8V;%95c2V3vY^l>HIMq4AVX$R^xrB2Z@6uPW*4jT8}Ta)Yj@|r z%;TG8sM)PS00Eu`i2nytq! zllHm&q1&Fy7LC*A6t^Pwpv?;)Pc6BGCCJzby&Qwo2dz`vqOw@2MHABcR`19#W%aj( z20MI{%p?|2rZD?R&@R{N&@lFLFCXs2daztxIh@2fhgbC#F#nLdlD%O4s0uk>qciM+ReW#D%fWS#*M(&fhtEZ|)3%^7^ByUZHn_7_ZGo(l_m~P#fhd!Ur_6_{w`g46q%m?yq`nUFs~pLAhXWN=e6WZN zQG*A3q*P92x2;dnVUsfVhYV@NM{+4tZW)RU|1lifH<@^mH4O7neKf8NQ?xb1exOR?vC@;TFlMdbs$isWuC z7)=EDoX*bKtFh1ft!E8;+WKRXeAJ-{n|Td@LpW|1dS`X!A@wzy;gIGPF|detDEtN6 zO%Bz0SBF58^T5_1x^?3}6pUNB6`w~cWW^QS3l8l5PU{h(U3>6>Mm}xw-Ul$8u6Zx@ zi`r?wkVm=HBM;jshZFd(pY* zxu&P!J|d%%J`^bZLM3H zzR{!!(el?pK*c;wtbVd(TZ6&!ci(>5k3IAxu2$;NB^r$=heB%~6h7L&$JsCz2!ci; zhbNS{RppxQ!xkO{nHY|Kn7#Ub%}_FCGgq8A$`26#7Z;5`p8xt?*O$|}rT4bN)0iq! zqLe#khhcTL)GCW&kDRsMZmK;sgU3rG%o<}gQ^?>=>?Rub*7U1}Rr+A`l|mNl8w}&R z*S_zCC3H2<|F%pP4RF9nImiEXjtTEvLf!ut`(R8G2^EcC7*Z``Ci*q;Us`%TF#B8M zbGHm}G?4H*{b{bVe$VU!1gVkyI!VfmB!0RbOcw`+M4M^`>h{NbU>WUGjtQ~EQUi-) zPUS6-fFRy`0nn=gxm2}+5@*p6?8K2lBTxbr$d!IIzy@};@4E6QfhqM*65I3h;KBC> zVoPD>8NOP|1+v9Kcrvv!uBfPyL#Z^;K(sA6j$U)p!?U^fXK$O=RZJSM^TD0rH5tKL zBx0yU+~-FE1;L#rgOTk{rl;deM;alTyM%@&zSsHNyEV-^@yAbTuGgyS+M5V**t0)E zu}PD%ieARYe?dh*|3Ml+bocKTa|FtiL}tHsZPgwK(GH+{@9N8}yEn8&Ysa~* z-oomF{!$T5M$!JRYM84GL*yfZzuwA=k=pz}hZQZO{mP_B)}%`f>=Rv!@9w;)o(1l? zJOog|unGJrr*Gh=9x}U6QJag}#yo&MPaf^^o3c82PvD=Iq5OV3Y_`I3&M1g!6V04?Sd33Zet z9n$yVJ@K?^ViX9W@vBeykjE2Aa0|f;!<^?N72tS0e^H*`_B95Sjq&{{^IAzvg5v}_ zr6|Wgg%S-|((Ah7dZ`VP-rmU#12oIFy`e)TVYs5@e46dcqG&l?T~uOqNn+1Y!b>Ka z-K4)^R@3+V1&U;52VOB~X(TYG;M@Mh?H66?=6LB(P0dYYoffX2X{z}(nnx)V6*=zu za9^e0(vgeWIt%PiHT;TB1vP_iYn*vFiBGTLM;}YE%>;-{&`YPU?e4p8!kNIQ>XX>kA>>^Jfa)vD;-Pp?5TGj%QS)Qlm^v-K<*-kEEh3Qr1H>xK#sG%bwAN4{?bZjiW-e( zI%tWI_j*Edsq`GFBH!1&vDo|HT>xzl^Gl6rhyOQNV+GxOoxm7M@ATLyPn4&NSsg1VYrtaAIS>Fz+GHM#WP>P_RI*|jBg*Q@ zSj8U5-XhqpM5(j@&yum)ld^tPrX{xt?0Pdk^TCkT`cMrrw86%h>R zSAvRG%f;4k`ca%WSjF$=%*L>iFe`pl)KuxUn8{r~zYI#_@l@9LW~LHHA?N!C!Z<-{ zcC#uiPZq6u_Jg1-^(~6yr(Xq;`L3mZ(aO@_V~(;Uaf@PTQSOBV&U@_w+e0;GXK8dF z)PdD}$%4i?tsm`bek=2q;dEWl=m11(KEQK*@g%0gamy(WEjnBEzWn1-XCqi<{5~TL zw+?W zW!Vq%6Pc7adLoEIITaB)DFa~xfK?<)wdFy!(-VY*%Kw+%pBpI&C#tJ6#rrfPedd5Z zvhHk|?RjDb(2q1tu$4$Z!OQr*8^(&1lpkNU4hW6fYMHubTYP+C9XRpElP`K$abUbn z=sPmn&_5bd?&ySr*a4@-v3*5oQ7nT%AC~JGfIw*CQ;7CLHw7}_bkdlDCi51W)fc2p z(_+n4>)$Ut_W^N=(s@_*ha-N$K|CP^I%kgax2uC`Wd`&*`(s*ys~Qr!$u2Q8)WC7^ zu1RJw_CCW4rMXetKzVUm<;IvY9X@?=T1pB)o5AoHeJT^7(<`fzeJdnAY=RS8F*Pff zxgDq1=nGI^IN$F2Is{Y!kEGyph`0;Z#F}XS_kpTT#H&kuyA+6n-EB&3mC+180|C8t zqu_NA4Nz(Br&6x<+7zuReBCp8dcXcE-chTn&EcG#DS_RU$g-2(wbVQXBWTxO2e=Ob zQ`uE3i2?w5%JfBL<{VP=R2QXYr=^2k0 zm#c9sS@q-J7Myz8ttP<)X~Yw}C-`b7`;*m?WC}bxUnNgw|AZF6>zb1S}|eF)3Qf`AI9fuQ4YUxsN^pWa^{y0{_3$)<2jk_Y;XI9JoZW! zu9M?|d%xNQhF4@I9{vN41Q1M~=A}Lzzn1EhOAPV+SXSc+(N$3|Rh1F#c{^s)E;F^EeMF*X{0rHhL83p8W zw_J9M?|`7zoKhmQuj=>Lo~$V*%!fQUl@CB*Uyv6`O^hWtu-6QAy|M}Rb|7z&vld?M zR3+{MU`b5NK=n$w$-vJ6$4&XN3-?s!9+jjZ7qRmIR6LOu$>8C}H4m(<2SOTT^kw3i z>z3o#WryE_0%fo;{TZ=@L#1vdYq-~OUnV(wFN+z0JQt`t9a`wqh29px(h?jchk$}l_p`zA_trBkr;J9?qQ zW7&ZTmY{g}wr`TPb4#Qtc~lKmVm;qauV<9@on59^<((xsIObg&8Qmhx)KW%_gqHkA z1!V92w!_g}f#D*K|3s4MpI+xqd3r2kj4Rc7vp+FJq!^UEWr0!Py!h+P3F6fX4Ad*}K;54>fk7Si0$lj4g*+YrH3xu)C^ z;%kgp-~In0JTBw2Hdi`aufED?F&};PuNVGyM)={jmqAF;g#-jta+Ewy#DaZ+&uTWR zFopZkd|1{jt$7EZ{tK+P*|Cd_c98@aA<~JvPuYs!)(H4C3Udsi6f2ePHfJ_n@pC&S zULPFtaZx`AbWHN-92~mhW#}KUv-$mp$L|0CIQ9=mia24=*=T|mm@p><|5FnEf96tf z*!OYg$TD`q@@ z9~7vU`t;O}%clM3@9h6B>&*G!U>kQ%6#0L@-biE}$eXAJT#|2hJU$d6|9|YgcUV*H zvOX-LC?Ymgl#U80AQ5TOQB(veDotucdRIbk5m69OK#&fhC?F;D7OH?sCln#H5Rnpk z3nY*b@;(0c`<~-__WAAYyT9+x&&!KTueB1^dY+kiX70IXW)^8A84V~E??B6wgRwe4 z6J6>_+qucto~jk)-XIk$P0!~^XG`)+1%`m`r~59f_6@0)0CNm_M1DA;UFtpX%*^Zg z|NJHYx2^l%n;SlUi5vtTfk-LWB2W!Jj*?_MUjob1?K3`Blcn zGg6X|_dI@t_d72bns!3X!vp-4QpS`Pd|6ydNPr!|tBbA~G+@l!^XrHsD|P5G1scr* z8n|cgL7r#7{-`v0_I=y2VZZtwx_!r#{`}FzBwm-D-kAV;%X1Us)RVwf*RFuy+ZKi>lSNaUPxQ~P_mS!Fw|1Wj=ozMMobHMlg z2N4IHzaIb3b}i!sk0;`zlN4*%U*G8$DgIX%_CbKedWwbrCr?hd??en(xSMV@sE7ZV z(0BrgZA^N)??2kL$dCIDA_Q_ryT1IYvj6I3|0+I>OvmJoj=w(tH}>kk<>UWX_5bfw z-G6_9Y2CaB4e5S+bIa(>iZfgX%{eeh?Rw+)`#SN2vM&l(uMM;MQxbWW4|W zmsoeqFFXlZT&o*j5(qmDAxRhD!@)qB_8#Z~Pp6WOpDj`wm3PGn<2?2l$4Ee#Q+wOA zzpJ)GwLCXws#k?8YN{>j*0t{8W5bh1)q zh~mzRWH~RH&Vr}!{o~4P`-dFCcg51t!vKzTR+kKF1=(r zwc)izU6N@7yoSYHk`L*ftR1GPL*8_WFAtMX@=7k`L@Scsjp3?a7GfQK9@W1k;@>L{ z9g}N7?j&>k@qr(qpH|TgYN)lrBl@vyD6luZKQT7)g5L1*bUZFKELFeS&ma;^#F zh898YknyV751q5CRMx1=_OVy|T=@iBoQJD5x>FmDS#@6Ow^GfE?P_u$7#q}Dpk@Vp zYy2mxYvw+;Y11XmUe}k)?GGmTP9hc7@sZrv;%11$aJ5Nyvb4QaDl_9#kCpLK$VcgG z2j<6Ye_N71>jI0B|(R*mA9Nf>{I{ll6L!aN8Hl-N6|?t13H3Bp1;dT z-iCgr+d|CZ4n$W-|DA?WMH}c}xi~|A;Hc{80d9#?b@1pz(=_H(l}%54GFePJ>osu_ z)diE*DlkUmWtasw0!|G$e&7f8?K~1>qOKjC64#^aEj%&)X*oq1XHL&^!!upyVDijjBrpS`R?BwW56Od;l_`_{I_5DDl9_&+#FKJWYI9bHFOowZ6XxK?B zP6R2OEGB(}Tk@}m#lH}rV{!YVzD~D^O8vQ-Jd*pqzD}7%WHqL^Wtw6~+)e0*%v-jP z9E6zEH_0#0WphQ#czknJ##g_$X#3Ge*yM6bZPRwr)Af+Nx#?-0{0MGZ?W#JMZ?eKg z>^3SyrNgiv8@Ku2ti%S7;8utU;$GD?90QZsfE%py12xSeIiy)pmZp`81?_BfR4Fg7 zJIX1=(3El$|26cKlU|l0h-e5z5OgId_X6;2*ewCnWrkluN&H{m{#qn(}d<%;6yIPQUDfr*1+KR=Dr3aS%v zRP5m~p*RTiExVLcmC9BkC73$yhTpx8tHkc!mmlZ6=$h_t(*xC4*^%JiajMzHd@Npe z$A8)5p?p}l%zo6+Wv0_ig-i1BJ7L3GVTSIe$bYdJe|^8t+(4Jt@TQ{Lf9f(_KuA^; zDMmXax@}EM0vrYI(j#QKkKbsy5aQ{qpc!e;8qH z45*-d%ONXi<)Doyh%*-{(x2OppIy?)9h0&j>AT!1>o_9bZ51-Vwyc!Sq${ZZ4ipy; zNu;!V)i7#um3H5G(V0;C$?0y2Ow6}jgUg+}@efhno4FgFYX%;(QseWN!3w!!41U`Y z<0<~>aXWYl|A#zE-qO6AM%&m(~^~ov)48L}wtdHbP zgjYeZxXrf*3X8PNPH53q?Dgr7Tq~MA!VcJK6aIu_b%p4VXaV_}j=PW@_T2Hu;f{kP zb!CpU_#IO4^%8e%&=tQrH}t=g z|2WWi?jHiFOr6srhV!OA%{4S;MJZd~%6wc=+cLX^cWv#4uYP4}>Aez*r{I9WudnCT zO>KoFH99e0gp+XGO9H?a_vKoO%?+iefSUb5=}zj|pL*5V6#q1Ou?3cdUw@^@(8gJA zUzNn={}IU{?X6S4HQ$giP*(_T5V;#ySj4>q90oeCv@XBCwRHC=(CvbB&6+p2MRsx) zy?`lH?KrJrS;iP|&n{v0e8aac_`+Rx0%u#pZ7Tas=cmprsMBmWDpROCrGT6>x4w8+ z!3W5L^xRq>%#;MJ^;9kU;j`j_d#A$XGeX=R9K#GEmt!8^V)Bs>dh^>#aKVC35Fs#C z(^3A1_CJFL02F=bQ@MM4j8HacY~8z5NN1gG8}HjSve|7HpZbPfGSZ;GrG25W7R*y( zWiDOWm=jwc?V0FX-#xNUfG9UF!t^q{YsEAz{OgCPXA@-BAagtLh&x0FgXY7eQ-D0XE~PLTj_A6 zkbc3CnD|GiisAU^HxX#PuYu{xHo|A-f@)M%hwyJ3<;b4>A3YI0D_1bVe?$Zd`;!li zj!cYOkzQFxR)HRmVDgmcFo2Hacb&C2YahRKF>Ps2BUzGS2m_cO%} zMQ)VaW2RHax!bo1_)H4S-g2%}2JIj@?}7-xt5 zgkU~s#c66Ho05oEGf_N;dC=O4viSCj-8dwZCSXs#BJj! z^m$=D*^6m2xKrY~XLh;$XuHATS!?2=!F!7z$rr4sEyd?%E9z3UO8%Y#~9@ zzwgyuI88@u%lVw~$?cC|Q7MCEHQ&8}6%IU#-u5X(D!bG!GlE;*yHoF~M_Sp5ttd;{ z5PHOoJ-N%w|Fm$VRz0<(;lrJ`O-?wJedXd1szXF&bL)!0A(vZuC9*Kgp886?ay7m= ziYcf)eQf}9+;qnt1IFZf^W|CTu+x_uDs%hC{V0}gebespD&dO{8caEBd}Vj%z7^N$ z6zUBl$g30QuX@g2QB76AF2x_h-~7|j`ghLz_lkq})^;k2U=RK{f%4ZVjF z!ji3CWq2XC#iv_zXc{n_Yr<0gknit@eir3aI1JAekh}?&bC~Qn4B9+}%Y~wWxT-8-TxEmlcqhU) z*N@HjM2?PoB~A4ZWTqk7Bh zONJK>N~+{A<)!k}Uh|e)sptP+z1?drigP*Pr$=iYIkI%J`U$jSX5EYiSu$)z!74^P z7AT3M^P^}vDtvvI6*uh1HqHeOmv8x|x`Mib5c@%=NWQyq9bdxY#&~7rhHisF(-xR$ zzPp0=rN=qNUCqj*Nr@5*&Cc!^FeFL<3SWC`Qh2XE|68F?Fo>618M>i#n(_6--)ex|8t!5E?Tc3u0$!Tax9!Whuyb-)EMlJ*;W zamn38c?O1ulyKKHPNs@zDe%{g`+hOM`z?aR;V?=BgZZGB!5G+O5v1`h&RH>L@TlCm zQIExvQ7sVBbh}k?cHIE{sHy0qJ~fzp3OM}mTjQ~#iajF?=oG)DPH&4u)BbOBr8`_? zavo$0%k6KStCKVJ-hbszu%^~a+!vYO_i_w*Y@gT5h#aBY=aly>NcN|ZQ!-|GSnTp< z5nG8%(d+O@s_v*iot%bRPL=lvut_s80w}opGm6;R9Rdk-{mDE?NBrs8)>W5Jj~0?U zYU)f&b^DeA(rG|j(Z$bA2HO|<`Kgzw-gv_lK8v(=l!E4NtK5iM)jN` z+SiUVYz+0(+M27zNg}ZsF%zR@)fn@YCFnPwkpjoOQN`m{fAzNis<(eRjg*8BT@Y>T zIini1tl-&WdOKatE5ld1S6c6#K=YpKjK&TACV;Fr&lP=MoMwW+|Q#$qx{ z!wDk@`f4m2qsBIeOyjRQCK@-mYk&btykA03g$?$jjf#|R0Jg`@HsS8e_38I=L{J)O zG12tWJsD7w)Fy+N-^*p;_kgXu660Svlx7J=X~=oBr8Wn(6)FeanGQXb3B}bPZ;CgQ zDSf_Y@9}*-8biThZ|^V917k#hfcLZP%hJl=h@zS z``!~^!ffwsGnJtCt)qjI#c1sM>Q^o9a2EvA27s2pU5URBPAa)FT|*Tr5ysoz`|J%d ztL=z`bR1{7*b&bgXwexjstI+ubLUmExUPX4BH^lK=qAb(<#LPcxAE3zye!LmZm5A4 z`z=SiX4I*C6-(Rw30k?=Y%XCb6%1&_c^q`3x>|1AqU-?rz|Lym;z*s5p|~^GM+0&%{EjY~CI|bFD{bRf8YV2=kkG)bqX31q=}Fy+WOru>YVo-QF#} z6$0F<3l9T;qklL3#{NZttE|9vJ`d9b{c>&l@3cjINf}6k13WKHtCj9AqVO*P+GBYD zOQ)Y2%)jP0~RJ6}cu-Y=g?1sF(a7>-?J;r0sL9WiR_xyFSR)9Tl+D9t0gmIq-NV|cHuTM7|9f-S8cFuq9LFQDcmO==Q zwR!Px<^_0e13+PRmeFGMZk&iJK07rQaAevinu8r@B=_pbU~=(y;~;TI)d#+V>dp??I-d@RBGrO^ zZsN#hb#B0Xu#EdVz~mpHCWu*A@UDTWx!z0Mv$v7o$7XgZx9~|q)|%`xihkrW;j6km z!8>cgHec|S8kyG2`?58XBzw%>j}^Kj4J&eMg2vsBykue}%4U%@(8kE=#qcxxzO*J=-G5H19z1r*{+8~{&;K6%84`|F9j7EFu<%eqGDT{a30 z3Y73fd`62_j%?l_Eg!G8Xn&`gDB);VII$*>`pye;Srd!~1}-Es2QZw$d#ACfKHV!h-(0{vs|qb@wHy5rg#D>)346?lIRsiSjDk%^txvayFi_|nYur~V zV({o>dHcZ8OtsjP*^hnF%x6)cneJs!Tz=9|FZAT4b6-MFzlkYosQ-4QOGgT$SELMzOmSG+Omt|r#Z-~ z>4s7`2_^|t-|D6?=(7UafkX@&+pxtobcqRYdW_V}61tS@M&Bn%?BxF8QK-%lN< zZl>C2d6S!Nh21Qa{W<5b`0=g#Fe>$FUKYdvfC$%#iQ6e+x~)v7FFs1GS;c&p?$#u% z2kq`>s1ulOk1enH=I|cHpW=fULoIEW`Y_TUqB!L`r#ekX-me-owday`{L}((dv8Ul zq~gtlL7b5{6p-*{kwT+?Q!RJ`7{_FeCwvmQb~P?q*r4QKIt=|Z`B5&zBZ<|Vn6LD< zY`6W(RgR)6y$Dz629?6!IHW$#BD(>9W%|dLFi9Tvq(;Jr8SjOGJb~zJ^|z`s1a&|q zD)a#+gm_irh0&EyPZmD*zPIctY-k72PxCR`_zHt){=Uh~xE+KxeK}Q;q8^)|Mnq^t zz`u;;ot`n>-xn0YaVaqP@LTO{&6jfs5fd4&`4EArfC@*)cEY3)xE*uJGK$bnOx6Zp)U+W8oYvHpFcATe?0U?eX;J{gh3mB+0&| zVQcKOo!5XI^yyOo0&9`B&9q;Wf)0a>n7f)l&-64rN^=bLI${k5_dxxr<;`7mYQGMW@ILl;S&!M>ozeZ{l*_NYR-w^U9_ZfS?Vv_j)G>IY_jQXG0Zx=ljM- zM6|Dy&uYKKin$gHRKXGihrD+{`MZNMVLW_uOmgIfdxPZlM*`qt|j>RkaYsjCUS%o>StNl%$Q>T$*@JuHrK1r zq*4<>L&KZ)6+26qx~|Y>YF0k19N-@6K?RkSx1;pt2aC$b3l)Qp#x-<>9n|a(C6Rx; z3~EftO83NOaYJz?%qDOrgb=fyeva`lwSFFY?F=wN-dZ@%h07HEeH&iA_v*V#l!z-3#@UsV zbt;BS#&P9)tLtB_^`~7&1drv{I~N#0U-V=_-d{WNQe)Z!E!n@1SmVAh6VyGR1%3R) z)VN1*g?x_gCr{Z05%KTd z!8tk{Ng9D=61lq3`39vUJV=Jdl_Si`!Kp19XT`?u!TNoc-)c_?It8yZc($F`o+6-^ zW=6{GR6!MMW(JCtZO^!Q?tQIfan+!m-5O?Nc+&Y`Ip^nP2n{@OH<-IReN#X~ciksR zqO{R*sgm}IZq9aZllKv30@?0v>}3{a2m5|6#Q{_)atpb@wl(;yNY9dH{pApQsFSr| zs)|Q{CeQ&%(QY8yI1 zF|h{Pn6;C@MB(X=WUVpiQ{ z6SQHg!+yi*Ioo-Ae?M3E58=8KzgMbboxjsQEGDnduG#SZ!=xZhCRV|dLU`t>3x0Hri{U- z+}6`LuwXYg(7tEOk)>!r$<;}F!l=18Mo8BsPm@*no>FzTNm8dC{Pn<@YLnCh=#{%y@M)0WMhpHCTYWwX z5?%LewVE}i@w)syIdoW4ew+MLZzt7cA3gtK=w^_o?JTOBAi-f=u6ByObFdv#nOA-o zS%yrf2#=o7j~iG76{%{^?PM9{`KPXk};wV>y0l2PVZ zk|T%Ipr?RV&WvKQEOHp@#yQrkK!e`USrb%)L?qXsk`zVjT z=jTWr*f&roA>mGxrq*rx4DFk!E3bB&e^tMl=Gc%^%jEz|-n@pFJ?np1VlZ@zTyo%V z+vI|0BnyY8x^bJjtIW}M19pyB_0_BJ-JM<;f1LwDHo2GWr>&8-JI@%4>?DY%g!I*n zSRS}NAG|NGViAt<(#d_;%Ozvh_u=3`XNy++Ba&@T&4h^ucKu|BetwZ(*+?nFf}aF( zb$MTL`AD^O|Do?Ap-E{^62)!hJYE_M&SUCFz&f&}-z6R)E@|`s^uP$hDM$T8n}#RkN~=vr z@7iGJveB;FJ(TGMVENj-(wn5aw{u|q4SUm;K6*($cf1vLvKV4M`;>g5BSMkJk@NJq zJcRmpW8r1S3lTorq1<94!11rO6BPWJ5*n$r7jQJJ42Q8`KG`0KP-`{h^ajrB2)d@u z=k6}I{}Z9BEyO9Ka)(QR5Caf|@YX)djt?Nfb3deCV0=8A4W)-4R9mmKQ;FazH#rpl zsZy_+hzX77Spvbp)W%$Y243y$xtGlKsi5@(SJrhSco*ZE_rm*hXUAZ5iG_-#QJ;k$ zPO^YVVd5?rRwJ8H1pxNLQmwe5EiZcwHmO(;Tem4_(;2~KuDfQhtg#3r-|CI5ZHSfi zaDvyEJ3vt-meaiEat@yl0#yDPFe3Q%O%Gg%6Q;-)gkqV1vES?B4~Tk>xV(eo7T)RvF#=jQhEeQjgr*ey7l#rz8Yt!Q5P+)G0<`~lJ`OFL(zqN z-01v3ZV1EF?`KK=f#5|J?cd3EWg(WUkmtkW(+F^21cKtqb2pf<)wjI!hve?jm;Ed- z7dEKYDKo6myW(G|UpSQf+^3*MgO31d*uD+AcM5^-@2-J(&5Opey$yywJ62KW2l9@SQ$#Tt8qnK4))yk(g+KWRyw3FF z_t6oBkLXO029xav{6P25kYjR+9?Qw{!j|ptqIi`VoRzBp>V;c6L(haxT1jE(qH!pU zDWKy-OiE5s%X1s*AKlNlx$i~T#DL*$4@933KQiZ_>m->250LY(K_1>*>TKlp^9W>! zLhJa^p7!uiqFyObj4GVUQ;)-`mU+m9Iva1aeB;sXzf00Q<_tKB2fXq?|2ICG=&5H% zlX3?Qmn!wqhz#AZN$tFc>Cc(xopANk3JoLWuC7G!z9#mR zShlwc$$M|Hf5pb_o*u}>hSLc<;EI(iyvF(g8EAa{V?wD_8{E{h(X^po+I-gS8XY-V z+T~i!xg9z89nTkzy%oIA4q=&B>WwyFt3A91D7%<^gBrcl!p0X38%^aAOXDW~S7WHo zx@`(0GaTQQ!i)9JbFWfKbkrlNR;C;%6OFtU5^YtPlJRP?7WkaKp;U6p>}X-C%_@`@ z_4l3NEWkA7S+;+bvkyBtpl?Y+$Un$=dflIJ*5Yv4~^l42Cq#o^g~kV-3MplU%j z>}QLorv)9@+2SYPPha@BH%{uD}W{5Yl4&F=J%6VVngpBkO6YbVFH6n2sJRo7d z73W{gckFw2$t>$Zot>_SAE-ZAcG-Ffpu+a`Q$kJ7M{MLlf1e5*qRz*Uh%2!SL;F=v zZX}4Aa&B&WeK1$~*}zBMv_ar0av=Y{a#WDs6%W&jRf(2ae?8xn7K}Ud`odk#iKN|u zXEGjRyj_Wsf%FM#{=v0IoUQ zG0b!G@qf-3u#hA1BERJtH}kR4r5SGPq+QpF(G>sP1ncKbR7I+eq{G;mbc`9xM}Lft zt)k~@8Wyvvcm#Q>2}{PkX?BPj=9lkrQP0RULcq1fI&CxJP?1x zRi6U``SlY>!@>aE>8_;UxxOkh<(dNCQIg`;r@YFb56!`4@&|goEV`n;IQ99Yk-Ci7 z% ze~~P(v0mh1YsW`rus9JwOe$sd$M15)K}kZ_;0{xCE=(E0+`pM6(= zO&lTbQ6Qhga{H(&!zJykcUtNnjg`x+)xS*ZYgm%kSom8318?yz9TV&2C=YLvJAF6A zb!C|$F9ik&1{(7(^sYn8y~~+r?a!DtPs&to1EW`ptycYd`0Qu>Odn<#oV?Z&CFy9$ zei1&OU~RFPDB-lnFf|27+v#W4NtJ+I)+_rc_VOY%S~;ZSiA^_*|!E7NzDbl2dzwcZ{{EyNrY}D^Xtml)y^8ELU4uDWotBwe19glKukwTktw(=Q=|8tXw(Ppus<$ z*CpXu*|?A+IEGq!CO1G!rq3Hj1A zWN2F*mSf6cb8VO~6#sB=&#FA=#?$`7RAS2blg_(Boh%KJwWfrxK*x~htI@*_M*WjR9hBt55p|=gb%56+2*&)=Qg_- znU{V-XUJXuyr+4(NI^p{BjpyuwmQEW_hnuX#vNy#g>QCAxzjK>e<5=3LQn462;glM zI+@HlCr4jU{lUfFdZYK;t*MD%t|;CHno|n}z)sB^Srb$R{>br6d(4&5l$e&zyC)5& zO@zKia7uOif71yZ)y#w~mb^pf3;U6pW-x1bx7K}%Xsr| zT~id|H0zNK0s$9j*y*d?`2c3@RzQXyAkuKrOw2j)YRq_FNyWtQ;4v<-jlH3K05`~Buft`sRPE_QL^=@UmR z3U&+53sMZr9c%;fGPm!tym;e~r~5w0*C7`s8}L}-$ql?TY%0?IsNAFG?)tC#qbuq? z-rpk>0Yf#qNvCn}lYb#X{+5suF4uO`73sSRyo8K_{_>0}78%SvXBL(+_}B|q!c3P> zUi#jV!c$EtcGx#;l80x{EPV)=@nMsQGCO~Ji1lQdT@^!9N){c}*D9PIPHyuHxUP(6 zyHk(R5r((Dn3pKx8Xnx7%g*#%`!PuQ)EG$o5(s>mu%LKZx5YD7A`|Gh3Vz<(ef!4Y zz~NyJ9gGuBR>K;q))nPUQ9p`0EjCYrK))Qa?=W@pXk~j9L!Nof9BjLs5lg#f6WNIn zYtb80w5>1I?8hCI!L-)Co_Ia6k{vWQuvQAGFKTpI%4?*8siZ>DPy! z`V>IF?4d@P5pB>;jBgl93-E&u%$0L92W*`y@<;n8)c8l|W9#&ZoQk}F0Td3=wnqHE z3O`BO-}gc7(srq3%%PWuT2rRg?xLP=I}u)I&^A=nW9Ha<74uu2SS}3cweDkiKav5< z|JeOjO0LBaVoYY1rPiB{(H{ z+nc4=I_uGedffQe&=5!2Bb;wjtvh4*k}AJ^UVZ(O2JkO;doAN;1 zqx+iArslFxPb~@A|NbfQLBg56_4@(Q`w}_C;ck*7^4%p9nUdg|7TaeSqgR-vV_K)h zuN0;}N(_oG&lHQTw3|n!wu&<+u!6x_IwxH2L?J%?Y!HUc=ZIrKEU2ai7zbDsVpv1mo zuNVSKvmWfU>*g)q*(UJSDxNydeDg4<_I#@JNzB#1kTKcA))L%K zCKEkj<4^uFk{{30XLm=1@IPBV`BnYk1Prj>*V-()4}QKM0K<2;;h}zfFnm|^Pm-;H zq}{iR;oY&eyw8*#pYD-jN};7u6bX0-E>T*#%HK&tWhg5t&(drM)LpFFFE(_>+n1_+ zalyaKm|;mZ>f4g@nh6JQc=vnLOQx%q^E$n$BZ*S>{2o#^$13^T9DXkJ`PI_x2x&*d zIe@ihe&_!E^vEkKx-Nd(+p0;?2B9JN##srCrB#zL&hK*4U$dmj0(f}h_|iRZ`v@tyJJlMkSv zifYd33hN)c`RZu^Q=G^sY&Nwh{r8`8&tB2X-z{R27aypnLbARJcyei%{P)}crs4UA zfx2)EL|BYdw_VQuuEF}teM;(pgBWqi;Ol=@-+bZ~z%TZ>oJSr0ed+!sNyw80aHN)Z z!jr$xCi~U>sFB(R)#EjTwI1s{H!$$Fu<}%Umt?>E8rz|Dj6(`_2LHC}8Z42>oC6 z=r>gD=Pf`5p5sONI;wtsr2iR_{ra|lxld{DaZn(wYrd6;&mX~-jrOy3gmz=?y(o)&et;A9}~Ja(eIqcs(1nYCj`;>WBB6Q z_Q)h*Y;xGV$@Mo@-l-Hk>8LOLHTgd_`GQlxc>%4quksX&=Hwk7iV^_XwTe8N%Rgrg`jkkQ|G z(POXdfjmx*aGELnvDJ|m_Z{I-G(6IAb0n<-{+sLp}+9L$783{z-zG?ax*W*9+vCd@{PdN?{RVJ$1SfFjN+oLHx1g7 z7Kk?#^+{t)NtK6RpD8ea&>V(EB<{w&nzL<5Sfj;@>Th^taIl_jut|^`&GqVmjrwfF z`yBQ6-rTlfMo?-RdPPZTkXZWPu${-MG{9Rqv&d!tSIG3fkcU@0{#~X;;K zp_)-m{QRVtfN^@FjHg*6)())X3K;Zd$Jt{FzqEB=?rqmfuzzq!-@D$$q@_aPknqIS zT1lK$@To@BbitqJLna<*$A;Uq2FG5Z$zi?Rr7QU!(dO{F4wp3Ui0rRO%5&YXbC8wv9y?C z*55Eq0k8I|o_Bi8^j|gxi(qgxE9UErC7r~fi&o1Lv zL2Xk!c;O_Sz$cUHNB&Z-h1u<{?Yb`Pc&@ASlRKcW{DMgla2|Ngj!T)y*{IOSn z5kh}I;)0B1P!N|`=zdo*?E}MU@b%x zyuu+46gZBo)G#b99PGh^0o9eOD}(C#QZ^*rZLs@PqtO9(V&r5t>9TVIbFX+(u0 zv$T~^S~BdbQmzZpz40Jxm0F_-k6Tpm{apV^d$3ch#YJ#Nb`m?EZHkv6u<`+8z?;yQ z9rGA;>bXyFI3OF2Xu*x8_#`&Q>mHQQK0vKYBSEw(@-xsc^iW4X+`^MYBMH5~lMM)1 z-8)XB%?7XhQof*6cF-o$XjGJ?tkx+unw;ab^Q2k!mg4Tn36`lHhpmglPgkGBJIOeW z%C1D>4J#vG^Lw%5ZIS*rR&uI6DXwmr;6$`oRt;0uCe&=zRhe#&o-9p_Q#W<+NiF+mf`!38b&zXjODEo2eyhDj z==Cw`ZbhzkzbQ06C0&U8;5Doe zH@}fqmZLwqlm(Hhm3~rW!cA)&jgs9cl&|9Ke7>tdb{)Ft!-ZaA8`)hP?nWf}iY$b2 zrS5E3)*Ue+I*?Da*&i)0$JwKoDh-KJ9(aV^tZ`~1hyo$B`Q&kwqNdylW0Izb}O&;55Jt7B!)^v+?juFuY8C3@+M z+}7rO*j%PLD+eRHKFE1^-01DYzz=LnPPB2*%gMP29}-?nIm~5+|Nf$zX{LQb&kPgV zjC+INai(r1JcI1(+SjZVAGc?(qoG^9L6=e7= zHTM*_MMs!Z>@Vu)-L=?TYawoHi} z(sp@4c}b?(lFTEDgl*DwxPZjRolZP1Bq&1SSpjJxyd)FbuqO>&q@Zm$3~;B8x8~|; zeh79M^eID>aYHb9{i|$X&~2+B8(FLQb={xmb5%Nw**;DJa?)mV^L&3z81=M5cCIu)Qvi$WC3R+)}{96g&2|u_~YH9Sr>csHmIk%FVp2tq1Stc~_ zQGd9J--(QJ4!L3xMrGghYW)_nc}rXtju>r@8oh+VOXVD%FVMD^RZyPWB{rv~E6YP& z0(>P~KekXE1zGK{ZB;RT@ZV{HUe`So5nkNDEq0$@#J~xX{yxWHJj>saef8Ta9=N|# zc-4N(<;2aaT`yKTlI%W3h&w$hoQ$Z17PpA|qT-UJJxt3+Tu2n7A;-oYv8Qc{%iH#` zBD~nnGnVRm9yVuLYm~dluc&;rT)MdgO)tk9D{8&iKbh(5Q&s`3SC`sbY&;V?DTL6^y0;aQZlZp?W)_h z8b$U&Uh@L!Vc^2Z#;r8dtQ&QD5Sr}|aoc-2we0gGS(|6Fqy(Ip5pg*VKU(%qlQhQH zc7yid*|->TqX2WwmF!@&Uja&VFuVOGXT>}|x(L*Q{Uwh4=l^~drDKk;H>`f3_sIsV!3nL&6UGC$l zu6{S#6Am>4j$|yaBins$W?lP!9^We($|yD>M)K)g*AuU;Nv%{_>%f8PI+S#To{hVq zdUuz!iUmB#xay+w&Xu7c-_7wl>!x0>Qfs&1ekH=fQN`~;wGHg@y~xmI_t2pIaCGFj z3UU#*(x#cAyCLza0P@skzpL|7wb%7D7~$*sb4K<9EHWfXJ)Y4GQ}}RQ7_hMBWj9dO zb0nRlpv*wqQoVw5>}Jh-bOFd+yfI9vb;hfFcv`O3*DXgNUsX%fphr`KvA%6ec{L=j zjQ7aB%N&INEGv?rNbQA&S9>2@pJ)F zC{!NNjt}^_9~b=Brhhk8(?BSm5mUFBtye#gw#}y9Xdtw%Kv=dxjA)uZnY5(=`cy(K##{i6k9I9xxPhDkyZ|;!Y=TZGFCsRNBKyzwSf&KIk zNR2$7GR}M0Urw1|di0=+`IpBp9eea~j|RLv?Z0UB=#5%whdgXLvms>cJGd>}j zRnBwBIBwgayq6NO(uQ!Q3;$dR>eB%;%j) zY9Kqy7Yh%pAI{XYz0n&X?|{xc^<{HEle0!DQbi+XZpg07PAA5}v_6_G-q~jF*hqL{ z+-=wHfQm>aDLCt*oa+$tGas%b5b$n)?3g6VuAKcyT%2_DiWP++s2?`MRTr96y!+`8Er^Rp7Q%-==_j2eKYz7r$w*x|G@cl=NGmirUt>SHTz4JI z?L7GsH<+_#G9fj*NuDX>y>@E8{A#szm=I0{EnWQjpEdng@CID{9Q3LGa#B?`;|1U4 zvoaX^O;15Y&5Pb@5b7FApFR9%BzaJsCpD)POG}_dt8&)~K@>e3Y(DnoJoeq1x@07} zvunE2@y=4{+~F$MemxJP6qiiw^ePjZU7=1{5w@1_ePXqyu0W+_sQV0v>Ig<=)i&c} z8GaeEuFIQEa*}pwP$5k%T8VtUS84VCZ?mCAfV+J>6 z=pB1SE0xgGNL4=q7zz( zW}RZ<=KA$5V8>0bd1$6KPcb363L!}x@@`+P8&+kLY35t`UhYuxvPwdwdmwUys~%dk zkXN}<7HqR>Ahb=f5mBk?R-|uKIwXKw1(c!?g7hN2*U$n2A~rxpid0dGBoKNH5GjgC3B8BVJ4vX4 zge2d?xyOCa`M!JZ-oF2Sum72lA7nji%{AwkV~p8Yi@_%by2PK7!)At&aZw+M)dNJW zgyV~KI=Z2Z+19{~LRJUnK<_FM9Rj&j-r+?6IL(Xw_ue_NRjmC*J9p*Ql5E6g+*67uZB+X~t_2@{KzA z2}yYg`PzqG8ehO!upKK%M#aB9cVM1&Wnq7A zcp4=5==<1c!1e6A_5{s<_r=!b4!#5wnMMsQS^@i+4!GcLIh?tT&*4VTO3TQ(&$IfZ z7)}lNk(!THlRH2m9Nv5|<{g(^{&pMt-?6VlSZ*A~xXrt9&9L*1_&&XtF7Agf*Uq>S z?Y(pq94Cz4EeR)7gq|4#PdejwRI`K(uTV{!%)h@LB`v?V!p_cDcU1>gRFUADsT_apF9K*8LRa+R#epb#L zK3{yhQ||6$T;?hP0O7-{ZDELAg^= zE}Q9fB7OA3Zj9Lj>9U&IBi-eT0m%cO&s*&bC7C>lJ5n{-UcFHF<50^qzp+{9gc#12)c5dOAl+KQ1F-93Rj_YbIIqF5o;geP z`e71V#shzUxy$du=H*xUAcSX7Oq^^vgH&dZUod6|OeOe5@8rB9lvFa)Ktayw*f_&i zgStWzH5tpQo#*OxO?N;~Gq{%jBG~}CNG_~9Y#CgNu^mcl#OkBz4CjVZKT{M=`@>EN zA|kO#qAmDW=cQV-AyX<*v04qL3M+eMBNeg*Z)13uo{1D*+hM z+iL+{)xlec0?80$z}G(dIEF65l_*}>#M7CkZ`_Bv!2RO%+dYL6%}(+61;balCVT-Q zr-IUuA=*Wx^}RxzgDRwH&W$u?kUuy86zyw|5W zyM8|d3o|d5QaNn>=JJiK5YkSoi67W!Y5=}>;uI>W5`!Ytw`XFOeVWoQwPe-qvQwSz z=+aF1aiu_YaefTrwUu1vt_LL)LPxzTv9o-0XgJygSnqMSL!GgIMuq=$ar?7nK#3!l zj<*BB{#mb7JGqDBQmGW^2v5!6>(9?R{RQUB7#UOA#zNs(p0Z@5YHFr^{ZhTVn6x=~!i}h}_9} zueJ4AQcAe7E?}K=tPGl{7H9wcRlM^|Kf9|d2}Yj0wNpHRL*?D#o-P{?Vu=@ zK2qE}ezT^N26ho$OcZ^M#dAv6NOjpS)M#EO%KHy>UWO)!ODpD9-KZMpr74e*bg0%p z?3?bi_RS(60aeL0x5;acgCBHMdpi;(kgK^6DmF=L;#_aeT3{+jmn9~RRHG~BRAe%` zpJ3^_Gj;;!YPWnVY_WUp@f-XsGzMtc9)*MUa#nX*6t4M!2J7eMor6qwL6l^{zsB>! zhE92?&w^iNoS%I6r1R42=7pzO(u#@^|D8}(G1+6&66KE2oxot7XFGz8ZY^Drx0dCq zQ%dY9a*y-kE7C;gfe_|F^*8BqT5Kp=I141mCEM?eK=K* zg6%H=3ob|P5L#<3z2Nmt*?l~@{=lTOMVgoG%7yc%2B`DMJZKR%p|H5tm1Faqah2Pr zpu*(@Ac(k6Uw-0BFstinW2wfYhR$ISYAG1;%Z`n}$tr zog!u*Kha5L!v^sxU8(#UPqevbe#z#)SMdLp!2J2&Fjj{h)0=bY(OuQKw^ziulwt3#_}+(9<^^;{D3-4#id9H437CU+uv;Cg&*In90S zk;0?!W+2xRC&E8fj#2Fd&!I_kM&pzKx0f-+CgXs5%Ti{Kw*4W~3$R?m>JQCwT{(-R z4uWT?(p;{b;9!zcLEa1s(C4HIxQHcPNB_olVlrKZJEpGu1bX}XRCG$+Dyx%8L#jd+ zkVtGN?K;bSXX1+K&p870%Zl658O@txsjpG#-rqNe({ue2F=wT}nXzN^3f0fGI;Bgw z(w-hCpWE!`Ig(tzjZz*X>kk(Nx$jtgtQGiv-A8mLb(7|rqynAWWMIe59&18qp?r*R zA31THYgpJtS*_>Ru*cEhcuuu9n6PI;qR%c~2~T_Wk%{#gtF`q5Em%v>QO1UXSIaQ& zJ#N?O(sU7@LrNjTu5sCBVCnVsgP?T ztrQY>gSTAhE^CTHkonFPO#bg=4RWQZM8THyo!zEQBnK9*e8HM0qkT_$7?V5Cx08}} zjGL0em6sP7QuyZ6$|eE6IwzdodgkEf$~yej%vZ&~hGWDIH#W^b=`5XMPdcfLJrC>r znOs%tsjAHoc+!>&P1V}lsNlJ#Bvv(de0AIDw{%B3KE!NZ?I!$|BY*GKn4h6gX#HVp zvPDpJKgzUK5Y~T5cPJN6TSzRAL%8eiP1^G;2)Ht;y2vBiDP~n=f#K6})Ys$nypTdC zvS~;rI1=%g97y2S_w{D(mUADq6U6R}*Qn6KEe3SVM$)(u1AMQLex|oD^1YEw_InB~ zg6PXy&y?k__)8goel1e~X$lQqiim}CuM0g!X-#m|Gg{pO)*lYek59bzsbc|*tQc!N zDXZ#3HTDfP&HVP;Lov=q_;AGhClIwUJ=i@}!l?KMR6b*CP#x_$q#;-}Sjv^hEvonB zDb6o~Tg0G?;;5JKGE)iL^y>)XPa>vG#E9jM?PPqEc2Hx5IFKlz=u zYLc3Pbc>mV8bw)?o~esFf>N&r|vuP&A9z`{1FbBd}P3!OTa`itiSeY9(xo3V0f zz25>s)VCa0kS~h)PMgQaCx>TO`nW7Nl~efg6sreXdW_}3KLmo751G^x@JlN#G6;81 z$N(IUKc|b94z4<6nFxA4!hj}u>f~||ZEC6%ga>>nEMOfDBQI4LJkmY_LWWn(e%$s!f9(WS zyq{HC`CU-;1V@@6k*Nzi9`I7+OUg*XNjQF`5s#Bn#73$6hCq~|zE(1kN`mNJMM3(u z*eRO#y@;Ca&ul4%k|tNmW;vkeLJHYXEj>WYz?u_$(f!B=!@SG2Y73@}`RQ4-J8W8% z`z}<3@QJkaUNTsqEJm1*NBnj$cBMFNCLP&bM=ZOv)OQ?`Y3L7KwiRgG_&}-^%@A8D z%=xj4Ff7dZ;@iR%TL~RgFwj)IZX%BDYQ1x#wf(|QeT2vM8(%t-QWr!tPpNz4n3G{5 z{Mn+sq|Sekv662S!;aIB{LHO5xB>IQjS$|~H;q)vU(ufK{^NNu$G8pI!uvg@Qdri} zfhVn}hk#x>MHcrhb((F<9wNHVmy_cTAQGf_Dd?P5fn=9YMC%AA$b*$jBkW$Gty`Hv z{P#}X|DsaJ7QO5qnSKlR>D|bQRcwiQV!B{WIzQSkN_}3@JTxW4w{*&8Q7qfrc4iy6 zaN55+fBb2&{U6JE)dOLvBClH_xwq}Vb@3FME}{!g@+1-n*Gme0IG(cqS3xO%lt}?# zs)n!K{&-)bC@@>=jjU~J%>eBPWkV;h2$qlvUekmThn>X++)4crU{=0ddz#_B z_JGGoM&EYsn)G1<{h*jKLdLMoXen7xu&U6sqO6dTp5Q4uQEQDm0)>a59o56bzusa! zbMG$H5p-qm+yjyswi01#e@ovep#NL6;TYsGw0|uEXRu|`zXe$}W$zbDARk<57iiIr z1>1l-sckW?dRSe5O+qk2sJ<2M-Xt9JTV<$Ig{#M4oqvb%D8?NllYd`E8~c$iMe-uw zD;z26*0p=?jgS3d+h$A;j>@(J%n9B7c0hKmk$*Zw{^#|c;N<-7R(<)+k?KODjqzRf zFzT1ff2NBHh96Yv!G@&|m>F8wTUTG>iURvLeNJu-ZEmN@xtk;J7l()5QqjD|t93@g zDKNX-aSw4Vo!`VRR$I-O`D<_QX;kNn$G$sBz0~88Evb7L(z>1{WD z>$2qc0>qq}%vy7Vz8ft;BBp_F_s!N<%iH5IsSo(<#D)r`W&oosJ532CqJpc*C8(|{!K>j(S0%SsXTwEhj%oa zUBXrKF|Bu3-@LfwjYkD-uM|hIu-{v21coKJOX$is1XdL}FZN{XkAZUuq+!6RsUb?x z*wi!?$h)i1X9We_rRw)eq8Xd|fQL=c+ZgT~r;)7ZTi%&V@;tY(9omz76H3FCx*3NX zHn3(wH0*Gx3YOQsUX!FO2L?3R`v`_A>P%Ot(n$quf9wICzs5s=5BvczF;6~xi%JD# zobWkOSmU_3Y4SHXN2Zc&0xa_2y*YjA(rDL$Tu4m(4}oLqoNOmoBW%Xnw@`I*yKu1^rC5~4-TI}hPSq^&G&D;*z z3I=%n^?(Q$kR6%1)pipwM353;$mS-sgp~2+n1u1rg0zkJcw{f&qPoeDEqd~oV9=j2 z@E?b=dVf&U@z^hb>NUTVE&pldHIyiMn_tPLluak}Mo2AO+V;hX&~;6u1tw(2O+-@6 zqd=AF>|SftmLS2limA5#+U0dyymoiMWP4f{99c6mji?Y@@iT9U*~XyExKa(nb|6UT zD2q+Yh!P-QdwE8pWRy_Vl+G*N_<~z3!O{Zso74v_e4qgL`A!M#%ykoYbO#{)T#$MQ znJKJa*K+3ER_QC`0>*f?n)@$a26txynqiX+^SXT36xr%Xk-6o3z+^Mp7k?9SOA3N8 zXm2UL^b^ziPjJ}3ufS103R3_F?QVOOm@>U;I{5OUt|?^QQ!v;v4n*qTrFS3OEiE2izvM z*V&|u>S}zsR?>uZ1(fJRHU-dF0Y9<|d1&8+BheyMhPlti74gk5Yk{2DY zDw84?T0i@|JmT#3zCv~tkHNduPadIydD=TrtKl11^1RBU3qS0P4mU87gJdqf@Pio^{W zMAtjUT?wh&S{|fs6aeBm#WYxYy!idbdd4J+z6=#m-WY!ZELVN!2ba$$)==~O{CjiG zjIm%Yz7O3rO9A2me!Bg7Cuws1ksq17Q-dt5Xov+9qQAlSiVI@|R+Bb&7S8)fI3a$3 zXru0RCeyd0-|oS?Xvy1KgB6Vm20qyJlTm_hfUYi9UYw?`{=3L0tB4~YA=EMPyJH*5 zL7jY~vmvdj*aCu0mi=YxTr06AYN38^3}f_kg+ke9+FT7WmXL_qt-mLxaC;ODD-^7(hkC|~?mm}G=HtBEdv;&C4?ZHXyz4r1PYYe#+~cQ&e2Cez z2DaJ_nnd9&{po_c+MA2FtQAt>e=9}!d7pBA_hr)5&##UVnq@PD5SlYLNS|W7CpUlx zP0{+ns`u@@GIVGSMZf_b(-o~x;Z4~1ky-4OdqsjFfXR$SkZVb1SKoHZIci&xUG#a7 z4{9PHE9dos5Kk7gNWoM8OAR#*cKgJt6C;{e&X{pWp!hu=o;cwV=|ilg`6(6J+`*Vl zE;*ZN>{2?@Zc;_E5QuFjV%=5($FV5crxL-W$*Apgy9t3xpSiP5*~AJvzoV$R#@ zWA=J<&m0A-hmD9yqT40wcBV-E1Ebo3dL&r~)_!s=)s1og0%;AU9_p_@1Y6M89rlM7 zTTG6Uau8IgML@P$@MPp9@@5+mz1*`UqDp(U{7DyIFWVG}i11qy$=#XQH6hkcH!ZM! z4xNe7lx(M0A&pVg%;Fe0vF-!%G+r1^ug2ej<2^@RMF@rbn`Uu^>Xh@9C>hVTqk=TA zhPzm>d+hP^k1_y-EMt^2b-qkrw>*k~!H1?!5jGx&9dZTSaT@Fx;|SZkLElxJpEpgj z=CP2siveZgc>>i7mI12xqW$2gdvd#1T#7@_+)8N^G`EUxJk(;)eQJ6u6~5OhEIZc^ z)sC}kOrfVcdi_B(uM2*r9GsF6VSne~)=AXVo<53+c1LngkS=zzW*mvCYQ!hbq`&9* z7E_*xE0jOnPl)hdOsXIxyL0BS&oC(qY>@b)^`W{EGWb@%J9%?PA83Nk?~6@<%#j`9m+ zv47)#Vja;H&=AfHhb#gH|2t*|$0#F`W*49Klt)kqz`7q(pGUkyX|=KPg6Bl7#0=6P z+T5AwF`u}?sRGumK+yTVLZVoW^F7uM4ww~A4j-b_zzNT`gQh8dK5Y*8)#;Zz)j191 zzqho2>zy?oJZA(&8cG;kHdY(6l02P15v-*>xh3R-`cfnMkforG%^$g#uAL!$$tZ2c ztYLB@B;9x5p>!yZ!EDe{nPWZ1CRvUn8r*v9R6!R^JMIVwWuJWs|3hbpFEZ`f_IC3b1C-~J4ibyVy^eF5K z034%f3O>Ys1GqVt_$C|cx#1)10xmbx{V$$Y9Ke_GYb+Msab!llogg&1DMBYeO1#rv1( z8m{>%&ob?mkETIAAtMkK^HALi*HOEhxC%Fwpesq@oOyx+zoL%Eyf$|aF?8XAe)jquGkL#PNT-6|-MUWu!G|CT19cDdnf)OcltJ-FM*cLR zng|lqfN<_u%0+%*U?ElG<<5y5YX~o@NP+P~Kw(bPv%|i(!Gg%C9NQoj>nFYrgPjK2 zrmL4L&9y-mHNX3%jg-4)v2EXW!eVwlY6CvI?lm(XU#_IG^9RQVn3cMihfbf+UwD|g z4%kJQAmemi^hI~PR~p1fVm2RGM5>&%FWS&VS;LO;XpK*$V~)jL85R-F;B(O;(RX<) zp42O;8L2d|-v{NxxZ5HoJn}KWbNXbH0?Jwi&drtvcAV$kpe%2{W781yoq0p=6N474 zhdI$$!x}qw&5Y2f6uu8X<`8d>3NL^3w}|)k4exu1g#r z_!hYh$uz;nhVSXJ^V%WDt`{tKOC7wrlRln9J*%v=q|e>YJv0y?)1YcaX)(PDA6}o` zPDXDtd{@t}y=%kCi1>uQL~4aBUQSF3xv#E`pA12rs#V`WB;mVNQsT|7D%Zl5ymLxK(nkX3FkKy=jwZLK$TSDoI|an(=Sj+eICvtoTGcGk4?_6f#Aw`+S~o_Z*= z4~~O*H^Kup{&{BNj|5-A>^7jOi-+bi8d;-BO&J&KYgLCqRt0!Dh?_@`l)VEO%v`KnGojuS^g)T<*}kLGGMv= zvYwgiayWL9I9lmmIS#|wu8x$gVe2XushM4&n0h!(5tlm;80GC1-H?L%9bDC!Pn*x^ zK1dkWOESYoCtp$XQJ&{h8fBj7r>ClnV(w7x#E(e%#R;H1GS`5(NZZ_79ER$DpwC-` zthF_OlastM*m9Psc|^h?w~vv*5z+<)557iCMqv2Cz$l@NXQ%iCi80V3vzFyu%%B2G zruXV6O>vd6C$W|#U%$_1#8UQH;Ol@W_wVJx|M)poxBWSzooL!Nnm8iP^IF{C^J|V* zbFFd0X(~Ak4~U8(#4<7z$St-vVM%H#S6z^*PCL$&SLe0~RH6wFl(zF}ZJPETxCI<@ z(*m8sOOM1Sl+&MGh0JoBgLYLT-EeUv*?=Vj#(gV-5`&B(X=mEvn(mqj$%-W-sxt6- zafPqpF>K%NII&8yK*k^kRmxpIB8>_R`_k@)&1z^LAy={PIt9k$)UYnSNlxErGc8}5 zNf!Nq6!}AW_HpPA9igS-KCX;GG=;&NDupX+8z9OY#(=R?_*b7g`>fc;E}-VV(CWXL zmT!hku&}@}@A>L-$!Ujl-Ce~#B9yB47H~YFa*;EM-vj2;m!S;6iJdVU(M-+SF6DTN zwyZ!@EuVYQx5QT^3%LPZP%%cYOufA1#+sp5G6wmb!`%W-D_OGI;Fx&`vcI>pI(HD#R zd!1%cL-C#pfB(U~=4TGifAHGuM~uxY+k1eD^ZtvHSf)P23*^97f=RwpKjQwjauw3z zyvz^7d?Twwi@>}*Ru4bN4D^d>ImKKOR0`_5qX~61Rf7PcFp$G*jt-V-VD3P+1Z?nA z4lALx{E`{-qw=pWd4z{nEAb&48OwvndEsNf4V~RqYZzP^FvtvO?a8h;4&bcC6?xhq zqIM&Y{wTvjH{3JGj|g^1HAdsjMf!9sn0_{2hlCQvb%$dbVoKN%NxJf!fgBZ!Dn_}# z`;{H=-JwsmA$0i5aA+9I%f&#fUdH+?mvYnOp|(V+494dl^WDa8fIIvOL?(7X#|!IJ zMkb5tP8Qd@Y%A^Lh_mH_NmKSMrEx`$z7a#r@}-Vqbeup zpS1!@KgF`+-&s~?En@(?8qGZ!kc{9CER*I?kdU(TQ5tcXu5@bk?7+No7{Zp3Hdbue zpPVY^fBUalgMZe~3OYjHMVW`Q`CtcGj_kQJUCWIDv9%Nkk<99XQq&81%=|Nb74^*I z*`j823l5M|+%c!CfJtJJc*x3YGQwq8-eW}A4O}; z0p<~v7G|W-{-yp*z2a}z*EX;;tK2dtTVOB54%}+qswq?(O?t{1;^9I{?*_FpH%55p zkn60!zG(G38ImFQ1RnKP&#E!=fw6WQ{x{WgH?YcY1 z_mOo4;SoIW-kvhywezs?<~vpy;xKv;m@aDr6pfL@1jjahtU?P4L#R1S)6I}>0&>We zL7=9;VEw@dz2kjN_2FUkiO1Ab~75IfEAUt zm)eHqwPq;{-3e)Sz^DMQ>d6D$sxF!c*zrzViecw_4FNgw7sjw5yzVSt8q?sC?it;L z%@(%^R<31)oTq@8;b+8+x}ZISNtp7hka>8u2PWPly?im4 z=hA9RO{%bmSKQNW`YWRA9!KfQy1q`CbtLOt?Mut%|81@R{n~kb>F^Ycu>SD!3B!M` z{yM>Rq^z}OYhm(csYDEToHFhmo2TyM=cuOV-|k%a-n@IH=@1J=o5yGA5spxGX~gYZ zO}$5^I%q*)JWsua2q%T*J)l;_f=XUWRedq&mamOuaiNaED>khJopq~$z~0>O_2|hX2lHmU{hD`im|Wgmpp2{8q{22+)}ALc{oMUHCa39hRA7U zhRf6+_%J3y2hKOAB2PUKfRbq58`;Yy{p&Waz4eqIH(J%;4pYyJ4R@Or=QOCy+_jfn zduwdl5~F0}eZoka6$fKMPF(C`mFN4BCcNZVvW{azh^qP0;N5k4H9vW+g~5!yzC*Om zJA}0s$GT@ZpEwQ-D|w`XhQ*fMed`MhX3#E);f{K$8n*;l**W6tnv-&CXRldgpui^O z=5%8f>)*lu{|nss?#J}wsw?cAp>4uDvmdR|69J-xbi@%DW?9mt_R(|xfaa6iqzI1N z`Ca@zyFH)p16O+A#HX|#L_Dr5`|g6b@tM=F2+3Ekg4!V$3}?N{pPo2xbb*Hy(^pZw zh6NIQcc;FhOpehb7l`d=owKBWGtcd!cgW!$siY!BMUmD)q4Vd*MXU6jyXc+}G;q@Z zCETuxq-5tA3FAkAvB}--v&>&U$uL(&Q-jf^fn7A6>rv1q=;qV2e|RbImsKS zV{B|@8DMO(E#Opa@B#SIw~J<+X5qJuddD=x)YB$~a9CHbkSJ|v#I?1cGp52uEXvH`PQ1U;NwC;5v121H75DSY9L zxk`;Mn_nr~^sB|H_vG$8)%U&hKv(MnBl&$0bh!HtweGYY9SB*c_%abgz+u^3&IzTu zhPyroUrmzw)Vw+qr?ZaS&uOFHak+DVFlz%Q4c2dAk%JNYaJ{8;16qBQ^=58`qp1^O zieJKqPR56t=yii6ApN=y-HjPHr%ixV!y;xATZxCyN7Rj+;p1||7!8)o6XeAJh#612 z%L5Q`+N3IM+s-iV9$X!*h%Qh0jhfb9!(>r|_&pkn2TtfBqIbM7U(@1xEo{?@YP`77Ac`N!nJF5t_0HcF=ov6L* z@1Dl!lz4#i-pfpIm2AjpzE|>vw+bkd|aaO<2R9%tepI}*X}CRom7sUWJin; z>&Q5J41GHV0_#s^*(4Gp68D=~&4naR|KnyU>M@E}{6!4%a=zOY=@1(NF>9QA^K-oJ z$Ibe$V?n&e>}M3K5W`Nay$D7BtO~2?+vowvJ9m8XCl!#7L%3O|BYqs{PE8J;I4j?E z3MxAYKTBIe%3gYJ85I8`{wDl3+viXKB+Plyp66Q5et5ak6;H@`-rTZGj}4%0H6yu# zfk}KtjC{m+O?q|G;hJdcZ9Q3O@?tHyS**-{21&WfJ z8P@~V;)KGF9ghX?W&uFFWUI*L7E$~Bl6h?)m2@h6{X$=yK4G}okbIXtGmpFsb-6sWrTL$y005BB9c3F zlS7%)H)!evcyH&E@AB~VH7Aq0!1A=fx%Zh{X`L#AjbaH=vV0fWqIqRX^mReQ25Y7V zLm@ZVIaDmB+9BEgQ(2mRK&~+gftpT=8H#BjoHj94Hs*o5wkAPBuZ&9rw+>%pl}BpP z(o_PI!iVK>fk^(KcIo=YDug?OLM6%q1(p7e|Bt@BlSq59A^6;4?&y4+@Tj{RBOQi! zO)IP$KtKwdfY??Hq!$3X%&3=f?fFf!fBSI6bCQ}(KnG;n&;btkRr`hH>ina%A;f5d zT!k(KFC|(~Cj84ZkCr?2kyMbz8&sebHf*0}%c$}wV486PW(-%N$F{(V{zZa}IvVA@ zXw351wLQCU$EjcVbEnz3MIS|&ss5w17}})wc2?KVq157W5nnv0hvJU)^pcr77jrX0 zD;!a8DW))nWZk+}sCG-W=rU_Up(pv)vM@Rb@nX#hksfL?h|<(ai3e#O;c9vTprb=# zHs5x%W_Zjz!WI(`Ce&bb5u)I^^1)yZgkQ+U3AyGjlXdiI3FE#^{(f)pRQ3FxcX#c> z>i~k3w4c{(!d!}|%*@)&t@!|%;vnOmQ+QZ?DXiM!Ll+I)svXniO=;2v4+Hk4@LTX` zHEih4qmwGVvDJDMSJ?RCq1i#aPuwEUMISRD{g9)%4mKz7;P5BQ3kZA^_8xZlK^lu0 zzg!&YvRtXh+!$sh&E1YNm>D{2df{*H!=sOKzO0>CG6pUo&~Kt?EIPnIwqxU^rW3b$ zUZ>3=-isNj^hBCK;*I-D`qyV#ZOBz0D%DX4#NN7C;SN+2`qdQD*o5Q}>k3Y1TM~9- zeFC0QUnW;AZLu07y#oEWe3)@TReDQ_l6$Be?<;O*W|lOnU`SDi6`_RpB{ne&%(5w&dLP(EKs<03~!Nyx+FKKX>XXS&-FXW#|gd0 z{7ADa0SKL8Q4Mv4(^(e>2hw0GfCZ{Ns_O(jUS-M*7CE)qdz` z3t=2lQ)&J+)p4)FHM7ZwxRsX2XP5^rlRm*V2qU<_Dcf6ITsIumu$n+35L{(}TSWUu zAWwk+qj$!tN+VhLM6R}n1}$(apAo@fEXAXkLA>(lxKG^G^_u7B1M9ca*psj5Z>|hT z?3@iOBwn#N?!RyW>^LJOcT(+wAH*v$t-e>S_fd2dF63I)Ec-#)6K#f=WEnU?ywWkW zy%>K+s@!&2ZYaj=z1;hYu)+c5eRE!W;5z9v0WjZ?_Zg@a-u+UOEhZiE5Hq( zOaO*uwAA}?$LWs}md0_LS||~r_sR8b??mDdv?eU>vsKgQmT%VvNcGojs)3nf0}YY6 z+%@K9{>2>qe#-NyU3u%XZT7%r`JqnU>0LBqhHJFO*Ezh%K^xji0n+vDgDYKX&(G)p zSN}Nd0!~->9;u<09J@l*-mSxC*?61MOs5(!Yji z7cfX+2TNwpPK?~HCxu7TNH<64n3`E%BlCJF(SsI5q!O|OR+jUD5-0gYY38|bu&vW- zu4=Vth~W4pCc{DRan%#HW0OWq9gWx z`@QDX?t_O`E|X!Z^Sw`=3;CF51_z*}cyRX&YE@Cgzgj zoVZ-T%z3B+_!ATL{i+({oY)6Om(pp5jg(FN!Y|~JUP#q+8p`(DS1I%A9+=4`Nl6^e z{*lF^z;XUZYzLWNLUv6Wpv*~P5X2*HF#`>0gVdf0ZgrLEl~OpfFY{{%MA5hxdZ!uhcXze|SEw@T0H z+e*3m!Wg7zWu+$;kzCKQpXnZ40j~c?OYpwk(_R@$Fw~iXij=DRH9n4*;37shIW&@r zPX+)b6K)vJlR#7@i5)9`WlN)M(RT{mm|mpEi4v$e&jp6FXUiC?Y~OfWb>jjNXzI;h zaYy9*4q{Rxhz-=jmU{As=M~RsO7iI0P8p*btIhKqS$)L5^A`X+bi6KzX2(1n=gj&* zbRoH0n>O&-9Ea%{sbm_IaqiRD_F(q0eUo~pgg?t`IyVD87prGW0jNyb>kC7$L0=Ko zjB!p{!}fC;YaAhucr)%>be!AwEE%BolYyvWmkb-=ceb>gd$V3Sa9!^i5pMal7L&(k z))^H8saMPDLZ+4x$kT9lOo-pC%;3BbP8|riHbz#tH^@>{x z{EB?123v0ZiO2oj*jTLzkbc~1Myq#>0n;W|UG{UN22d-?60-J+>i}&`gTmm8cqnygvbL8ju?H;QF zXxvKVn@g@eJQh1$yKl4ax&l-u0J@b%u8L+1qRd7+M-Z$9n{*skJu#q{gT8N>!G=Qug4`_~S^&mH?v z5)gW~xgR|ApE!&E`t1JYEBx~hEvW=-`c~kfXZinrO7<^%_4O>^5?8u+=5O%g|G&Q* zU~c~3FaLj^{Lh{Iw-5XN{`>#=wtp?vc89a+fK&6=uSGoQe>sr9vY3ya9^l4azZnnu zFGrZ~w)-UQR$FNM?BDNI07Q1EC7w|?kQhL62J-j~qk`5Tj<3c|y z-1@a7MSuxSURgV&DM^j@&pM3ddP(|ey6`|k@3U&yqu5=0eEp-!(tl^R|3wU(m^ijS z#34gGdj-1l>}YzFbJ5REnd7Oe4?gRaw5$%J<%U;X{`jAw)qi?TULLqzh?P71N_(vj zeuPBEKAQ-rI`Y$?;PsWTaMV-)AhY}=z`G}Yw~v45S00j(hu&dXw_$7G{@QpiWs9+P z1=^1R3#-&Y&p9f>SSEsx6)q! z^QfzOG2H^DTc-@Vj=Zz1q!GZzvNa4#5tEWf;akCf@i71WX?*897}){HYHXDJ2xY*N zQ2%<9Taly0JWt_z!$zM&@y`*4Syw+WgjzGCGBVCp+6*TL8 zMY5eKd+_yTZcR_X@AAMzqr=3X3unbdIhMd``(p@&L%X6=Br1FBh&c4%jfh`cz}LI~ z5KRQ5o;hSPG3RS!svCCI<>sDL){PZt`R55$En%W?#Bb7Dv`U06e~C2xMeXGGAid5; z&Hp_=_>b>;<$P^_+kGzvvL5f10-K4S8~K;2i7yR+33P}@s=|MJxBklKU)wMP8k-@L zmM=e@yng9RghgCZCGAbfd!P9~S~{T?U_v~$xH2Z}uLb@8)9SF?c|5V(;5jG$KfU>v zuc^lDD{ySw%?Rgr=rn36jd}iwwA~eJHw?YKEZhB8zmX7t&AJi<=k`T)^Y)f~f`jq~U?x5oa!!Hg>^*LJM~Q zU@I?o9yBJq;qoDE3By)K($Q9g+5iAZ34cZv@Efu+1d4p5US@dzXLFS)P{WV*syhJB zujFB=-L8JB-%?(%Fq6ypuRhK&anA4e8#Yex%8kVEE9F1<`XWcq<@P=$W%NAHS;_mg zRqb)|8EbQE}8VvwOp=V*isT%(;GsCh1) z02{SlRJ=p@WVr^+VJZNa{IZWo6Tc1W0V)S1v=X&X+*R6or~mdzr@kUcGs@BEO^<>> zX2@oe^cjq6N?HJA(NX^9wtSXB-h%{T-SjTY8zMF-=qu&6z?L`wE3BwiCEP--2m9Y$ z6o@3YU1NGLQq)chY`#W@l_98G&k&xTj*5`mn{RTsyg7d@s1BP7nXq6jJgvq_PVGQp zUkP+*_WRANL=2;%o<`JO)w*pYY)dI!hh!J()d#;dd!PtvJ@}9HLO_vrHuxZH zRGF`Zw^Pt^`Cz_7_s3&T3!igh7;8}~w3X5dhwh8KiosR0-uNb+`E=e9sqCQBBX5`w zUwf#EF|`6no1|)Er>1#8#N0xuFH`-c%o5mpQm^IEO7HwKNxX%aadU606)Dr%jbdP6 z^E7y`3z+3P9|iL4`Hh#{eJ8K})pV>-H~Ojp2-wkgT)$QJ}`?Aep7^jKd!MneLP|D<7Khs(UI_t8k&*|abn zb2Spn{ab8#WK?H=6y@VDVg+A9nTBho0#Zt`6`h0T&dh9Q_S~D@^8R>!$`6K3uV2~) zMt`qP>w=17=2PP=GaLi~WPD!CVGwB|*C4r%kxa*`3*3z+t<_nx z%Ou5eXW^kU!CeZWz_evXamA>xvAjPvU8TN}0t`ZQ0YIc2QIhs~IUT=LHZ$gxek3%d zv2_HhJ)fdtAJdTn#!ms-Ns8o`hbU78jKp5Xd(=lN9Zq!tq?94PK=XBmoH20?g*|D@9u;c$87PH^uiw?0c}n^*rQI& zY{3^`poP>0&%85D*~P&sLKz|Mf_Wo!0+7)SNMVvF{nsA=ZGcex%hMunO)JlpXhU|T z>K88Z60Y1^s7I>9i;I>P-Vp9*aTPu+{We!)e{)bsSS9Xx#JK3-S$ficIzzx*vqeve6d|{2`=}>*A z(#>xd0^R+6ohMG1${Df)iikR144{f8)C$)EEDjds&MM*V-7uEUQnBS_MT_BAd{W`4 z29U_eEwvOx!9%ocgKT69_YVRNnaJ@*cxVAsOBDSYI)ShdGf=VK@d)}6`4&vsnLZpp z#a)J3dLg#5BvyEX`A5XSXe8B_@ImAQ|Etoh6rmdGOv)pCn5h?1IXBISqBO^q1$eE=Xb4U)1faw;hWi_t%*DdNAkq8x$D#H*O)`w?B#a5ZA|7 z$ci&FB?SRZbwJbQ*;)?d81?9A!iYgDwWAWxUPC74DclJ^#1ZTO;JVCl!dG;XPdg@d z)mWm#DfX3a{Kkcm#gb3J?aWj_)58swzyP-@J@G(A^^fU&jfvX>HPWlg^74nkPjpWY zO(P0hMmAT*$I6|r`k0qHL6q`r+w{I}E?&1w3SZ81M9@(dSM(N4xy}oOkkpj6lWZRJ zGE)VnnACPm6Z3w|vnC0b$P)eDebqzZnSvJ+te#>?srOyO*>36riV@f#&{jt;KIIMm z@G!m<^3^14@~6CFWVvvZVd^jcz@*+faz8MlXTR~T_S}_}hQvdAY4_uof%)9S`#LjZ z0n{(ku{<$7s1t(+R55S`*W^Mba8D%0#6lT;_ECHL^JFSdtsLx05U1_OOblajA(~g8Ook< z5Qd4cR~;f3RO$DecrM25C)BRoEqYsPp}LhjCm$- zW%nP#3#!jQFW>ozx%_3e|BtFjb)89}VrrKIOn2C)7h!7HS4V|ICIwoQ_~b1bl>>Tq;6q!(P#8?jyI}B1$jE@zXnoFWRm^AEc~ zLk6*)njeo$*SRT_tjY-*-2Skxkh$StoPXzk%C!IEYeF@u^$RH_#52rZEXtJ695!Ma z!F!%9w%;xBA!M_Z*~acjR_>B|$(a-&&po~fqlHZb0K6o9oW=MO18n5-fmprAh}2C{?9* zlwL#cAtGC`AV}{;kPe~OKq8{_-a~-UTL_`IK)4IfxA!^c`|dec_wV};9`b~@tTor1 zV~#N;VSTCrJNm}3(6Ya1=MF>}9wou|$xslY- zvR-tZRKL2bBNm!8}7 z>$#}sD-;qv$xQZiHxI~oPzM~4S6imXg)H2$c_0nqjd;~cMV~N2vmM&a_Zwa8pK&MT zQiDT{(`yxo!{=4NQq%$jHbSW`REd1Tcqr@rNY>@tKgAGtn4(wgvJ`W{8nwcvDaGFN zKi`ljRUC&o36w(Xz37A5C(#N>+Dl<1S0FN9u22ZO^*r?(<;ocL!bsjEhj#YL`k)K` zo3QD2A;&vjAa74ojGw=cW$1YGk&)k9FY5)uXbVhtCIIzz&Oxr^bW01tT%v<5YFB=8 zhYXl7OZaT$i3taUD6Ci`+f1ojVqiQgjn}pJM>m5wI>l?#B7F&e!5fWoGFQE@sikWx z?C|*uV=PA{nmipVS>=PsmNrD`RcIlXNONAe{a`tpCcvfY@PcTjqSm9$)3Rt?+lDE} z#+)9X5p5rCKdkvH#t(miC7(QfuUry`2zi$uL`qswT6%#7K!y(4$yRp$<6fuR`Xa$^ z>>mn-&7lWgUx>yPK!$M4l3I{gulSa;qk4N=K-&9-&~(LRnbo5wFYVkey*FJ@nvBSD zcjPui!68LJeZW?xn`WnKzg?f`QwUVc=5SsuqYsf>1ui_t^VXw>)qd#pwg{KMU#FAW z7nfP`&hm=zqoyVdmMYyo|4~Iry}lhin|uU&4t{*G|N15I_5h96*Eq-=4rkSu5uCNS zj;E0DT3ZTlx=;ua^eq19v#y@4n;v*TT}FLFaUc@90V=O;m;lwinMTKg zJMUk9o8boKc9Vw2Z{jcja~0umM|8}C3JwZty3(4%$IKq7JF0(P>?=yQ9jU^!iSIch ztF6J*+pgl=A>aA7Z2gy7;k6g)K9;Z_M?~P%#ey`{#rsg_i0SWfK@#M0;QP^ilVai3 z(UUrWP7Y%sfD`+sq<|dkfQ@?~ARxV;cz4X8Q{5Ok{VF5t5fcQG=| zfKPo&R3r4x8>!WRVGj~vEdBYPorQl(3jQmk#!PqAQE^GnclDBiXXCN_W0Jf1f>r~! zhq+>fYzyrW3`O@!vgC(16TG;%R6Bc+4adl|Ro2@mv``@CbNTQ5eo^ zIFbk*s)y*_a!;TZ%B+kKSl~j74NGW@8v2}giTCLNB7F`dzUL-Y$*Nwe7og*l7x%?1 z)cG}+^wiPlXvtw^v2`_FJV(rdZVdMs2s8ExNslz*Tuajj4)^^9Ng@gkvhBx+4;k3`{yeB3lgIbc%G=clZJ8D5u$#Z=!k z+65tB=HBaqR;@?meMg{l0ShZcHG{Hv5E!k5*qqeJ(M3yYDK4SR+nEeux||nW``5Fx zJjY9$kCr+)s1N7cXil>9RcjgJ7~b+?kE% zE*l7dqDx2+&xTSc&JGBfU-XP_f2BB_1Bya235R(-rcUA0VH~7=RoItVV&d`YYkTJ3 zDKQU%?x*G>T~F?jMo%b_<~dbO?bNTM>t7dL_(KVCK=%T;P4bbvYOhV!yANb04G?(% zS=6niqVZ;#29bevfNj}-ol)rKEWb{Y*s{Y;jB&a4^U?x zImV&7W7d-7Nmh))cEw-_^i7Nk=c{=l0F4XD$5CtYRw}@?zuzCcqj*It zD0Sb0G&Gx)a!h8qKGC~tWPjA}2r}#3W~=bXY(KmE2r0o!(O4bXsP2A1hc`UEpKcY- zK~8zqT_`^UpqbeN$AIx(a#5gqhdwu1dUg=>T^`=gg0ECe(N4p{%a9OW_a>?E?UTy0HcK#AOpC4 z9>oOGi1@`UA2}~>ovOR6sR!m=OF6C@sY|Q8mA9-CY0IUpWd&bHl7_wV4mec)CJ=~d zPi1-1c6pFgDb8nqnCv(P{IZ)#qq+BrWgcxNpXAARwH=@^}z#h9eDr*Dfws zr+QGcK_hhwuV++Ue3ECTHFizN`pSu=tdH)CB8D(ANy!oTNU>$;&m`A8h)HsisA<+E z^$+g(i%q~i*8x1=QjgW?sP>4i<$5Nk0AktYMWsKz02C;uYVN;rH$U28nq##TGqChy zvhnQ4S2ACm1C7vOBkH|+y{)|oUSmg9UC12aoY-KwD`%IVyzYTG^^#rUD5v_7O%v$d z9qU=K&oi6r!njjPJOlGtulqCg#a3bA;MZB7swbWb}fA&_YOC(0fa=bqK@fq66 z$4oU_Q+ZZvH)I2gRU}C*5KqT!jo`|+<%sTznKvA%?YoP}Pcq-*wSQww_y5$0pX_|S z=N5fWzqt5BebqeCXVCK?x0`a(U(ycGNhvvC!tYnQek{dU^=Fwz>N~!x1{noszQopY zJ>5d~*A+jWPd>G+!faX)?{bi<;=9Z8*w+qtgOXZ$075)y9^g_<037Ol zk}%qF$_q-&f}xi;uxa@uI-Y<$P2k{U`)z9(u;qQvw}*BuXY~!we+ga4&O>ipgCE^f z2b1smV%&%Foc5UNk$z+mCx+zC}tx%Ymjurj@^QlO!csE~~?i=DazTb`iFQuud4_Y&nWtq07O<@Ij>B=8HqbQ3TZG&8s?jbx=%X+i4 zyXyiNOcN_P+7GmkuQjbS*H>cfy!16oc0RDHdLP;a_6Q4=)`)kWgR%O(Yn3ke!Bxbo zjrZZ4Fcz!V^%o|qS5}%~cO?dET4M!QWwws#vCIR!dW&OJW_EWrC9n|bwXvq2nQ}Gp zEdBx|yi5H$WaT@ZWK@pi<^sm}#1v;n;xmOg_oHzTSM^po(ZmA8E3(j~4$TIACG}Zo z71n4{}9>wVfrDKf#4UzB7v5c8IiTlUIGkVHpAQHJ~G)B`E38tlm}K2 zCt?zND@*hz?qS3A(a{)zr_T#Xc5~#67s#x#^z+V|8K_9Fz_kDrU~8u}t%|uVo}%Fg zGQc)RiVdEA{YzOHOoLJ1?g2}iy=uZZM-6nivUi8}vQazIY5@!bR~UOF^qV<6IE=q& z{A4_OJ>eqpx?Zy=Vc^k{|G<+E3PIhu99QIR=+#SiX|uf1b6GEG;nVbxI@mY$hnkPX zQ`1oA9|dNsv8U>X+Gze*t8kjxS> z9PEF1(p`u}%BB-(>1nCXfjw*C5wQL|Ln7N{n~ykb+ck1^ricZU zbL)DStD^w(0gO%N`nn$|>-wDgch;PA-vQN>Ve@=@)+c8G!YU!AMH43rq2#3AddXkoc-Uij=Sd%=m#gO7MM- z#nK~JBs{5pt7iQt_vh0bU=i@gN74P2+x{9L_CMvXh%O)ix1rzxOuup8|GcRNXTYF{ zR&JsG^|I$*9-(g>H=v!G3vxb;f@ z(c9O~Za1#8uot{^cJqIQ^Pp|9pBX_m3+3##uMWjW^P9KDZMWv+IN?10^N3U5a&GBS zk0Hsw`!3%)P*AoT$3neY~%`*>}eLjh1y^tB*TAnDh*>*&xgXVSPJ32O#xkA%|d~?@j!}GI! zDgHf~hK4%Tq(%7h(p}cyUQoWVT+*Z4N}?1+GcCcu$wi?wM?2LqRo?Y{cbW?Namw>D z2mR5Z3Ma>v9(EB>TIxxiulxM~Xn*p=8F!yh<)*c0enJALS_X4?i(K}q>`8D_YrzH*g27&&sUz~F4a4EK2lkwWe6ZQ}UMN5nmfefI2~u82B6 z5`to`lQxNwr`$%^K`9ZgC*m|>PXhF(2EbZm`5rI(bO%n)_ z;(^-@Fq3Gnd=ki1RY(x`xT&%@6mt9d7Ek5?!-41Q%M{2v%Ci4_>lIv(8)8^4uUGl0V%p`_By^R!Mqg^E5T!xEqG;Q9|@R~OE8UkYSPzO5Jfv*I> z8fmPr_G~Pk>wb8c82^kwCZ&ET=;4FbP|l5{_{T}XM|i}fI5cx^>5ICZ=Fj(;obBY> z^@lt-jD>xmszz@#ajtxy2*Msa<=sD5#V<=Ex}Rn!HAht ze%V$n^sq@S#FTPCuw0LSjE#?C#xpmg%E2 z!XLg2m?`jP6?ESM+txKt88{dN^R|bQ`zs2JM&iRJH{n2EeE%xu)AL%JPX$yT-4+L) z7RDSh=VrJoX}+j8*Hw*dF)V)~P~-uGU_(9vw_!a|cU&ucsy$cCChF{7BB_LF8N{_@ zgRXKKZW4Jn!+E$xoF?tddZVHKr-CtQ@YXjY3$4PbrPNlgl#_x+#gCkp&SSN<iJcz&NQ3z z$6B8$zu3w=0)T35z_#MjsT!bK6B{Y@-7e_K=}vOuX3#>+@O+@Q=})E@i>dJ*+FC9I zPVFW+(UO*YiN`o&s(qUKa=Rwi`=>DG3BgCG&kdO2`|`>#1X_s5b#wp&%R0W#rG5Jo zFm8=KJuRwb)aHSz@-Tggf&63=Gu!xY$hyVAPXb9gh^~_ng`43a#~>N! ztVnkzSerVCKE?QS#aK(Gw)f^EFP6142se=9W0fk6LDr>e-C706fgzjhKbpFR$rHNVVyxFs|IzuCF|v!a`_cQwNOpY%7sem~QE+>#-DCNns^bP;0$yj#c6q*C*rwbp1I zWx7gYa7ulkKjuk4^fcxL)_WM9@9`)n*_kk8urk?pA%A#-pt*1BssZ#S*yxXab0AZZ znCOGUUlau}@z)f=&2h$TiVIQCAQ8v9s9C5I@NsydA~`8h+B(1{QsOw|6h~f z2Me2%rJw%EXL1EyGM%$WGo zZ|q7U(Mvs-@9^X}TdqwR6OSI_R2pg@a0pFtxh1tL&?J7ll2wNk5ZESUJ2K8x!~`#6 zs!mDK-TBY;$iEc=IiR#D08X2g?gflr^!93w9p3OpQ_H%&`jFllbvECCQ{a)P(tMBW zu*wbDhO+x_22M@LFp|_GZqxV1ca?a#MT732Ub%fu<_V`t?;j_SC_PN!^HR>IyH>If zY3%yBW|S+mCtNoYYfMZwY1iEPEWcJqIL{%dPKmCgx+fb$Fh87FZ&X1Y^lUmjBe+o6 z;;SBxGZ8)gZP+$zdb|Wp%PWLRtWYkzL_hEnZ-NObqQ-!~N%dz9ldPF!HFaY8BU4^Q zBG~uLwEgZ+_{`g8F~O2*y(`0;t+6}^_j#x323F36LqBQL=qr7aR1f;IYPRKEL_U(x$yPGSlECNL&whwnWRk^HsD^z}kHg9Dz2k1mw5 z8&KJY(}gM8QWvX*jWpSxeL2%=BLO9Q$I|b?p=jvoyRjg`Zi}s`D3-F zTeMP-$llwNOmu6xzGJBh;UOTlZ`qm65P6Jg+W3=(*#-O4dzHBuUFbm zy}ytO82l7iOQY7#6gk^H(4#N~e!>*CaVKo?=V|hJK1reVtLP_O>ppjP@CB1f-?i)p zTg!|BGG18yM_a8Q)H=Ac@}m_A`}yzbn2FX`G`#-dru>ItQW`ZMeEGFGrBFv%_}ZS; zCvWnU@M5K)b$@e1=fq@%d4fIOS$1DcKlRxtR#5!o_|mtncdEmk_*y+~oYi$=u@$lJ z<5j80?=NwVgKhs6u5aYjaB)t`J_7{WEwoR@709n8Zbk>?bLz>l@9*deoiTjKGCDRD z`HzDoev6pK8260@ZS6ID-Z?IQ&yp;4me@wb9--s&K>-$PdB>#rVrP5C9b=omg^R5* zk6ZnjP7L^ijCcEO@C{QyI!un%-PbB<@_FO489&(^t2tO=dE@oT3yx(HaB@{_>vBsp ze{$l|`Tdrm0*S??fIC^X%JGu6$d0eyft#a$HuA=p%_T&e#m{jVXHqD4SVJ_pV;imL zY$T!N>u!de`&tBQaVdxKFAF0@{rZVK`Z|pX+}X%0Hq*u1o6f7#xrMNL<<-8A#Ub3* z7gve)cAEsYno?@$@YoHxsjt3%3DeuFQ{jNj={f%mnQy&mpuJtsP_|voLJS7pcG1U% z$C26{Y>F|gkYad$UjA7B&_EMDOhUH0zFiTdLRVNBWn z%iZ<<)18)CVeBkk8#&X&v~Op9UUG}prO?Iz=AN98-H8HdMq`Lz(PUfneLm9_voW-! z=h=PylRTN~(qd4VX+?19KWAbUXRT0_Qs;=I-4Tic3J)wSK@ui=;$j0EFi{8w3WJ?R z))tRFtfNzd5C-X%7oTnQ<^G0eZk)JF6M6HyTfF4f z?3J>5ICI;%wiSC7>=Y&HHC8uy_}UrCI}hJ zc6@ajR^|`g+}Sb~v87A|Xf?NWe{Lu)lIKEeTCU8Zex*y1j3~&{Xa-sj_h!BG*?`;~ zwL^k4W%2+!AmpLHJYJpVb03v%jAaWN76SBkq1%uBGMK#9o@Q#MD@!NVaPSRzm`l6Z z1d18W+<6kr3fNcH!N|pszI~Em`NpRes_53*Uh0oO7-B`y}=zQ@Ti@h{Zzw6pCovsms@K`**vj!gtU=?)fPrYqZjB}R2R@mugzR+gu(H~EEVJE#aRx4++9FjWLw8Nx- z+G&P6Y*9igGM9CBH_;kyE>fQc{phi>ywjD_XdS{YCzLKa>s<5g$4ZF^?+3UpL%zY` zbU`F+H7##=>GR9k%k73(w@1iLWE8ryuL<-44%ev?$W2|z+Q#gvqDur3a{u%8MaW=* z8@^Nb%_-8;HK<6sMbcUl6qWD0u9J>IVvC7foEiw%n1M{Hcz6e@gurG{<(}YzItgfj zAyCB)VxkWdcYO;k4D}H5B;#2{@-C6U`mFGQ15SOhYJUI;TSHa#ENO2q&pchRB>E)N zs%}LTn#woNOuMy6jgsD_8=iwj)u1h;b~>MB7*Ry?8^^&z?YXU=Y>9D?gr#xoU^g!D z=n2f=Z)`hkeYFR}NN&*Z5v43LcEw*PeqEk-{qbR#JFj~12@_mkiAYk`_k7Q=6Ip^6 zdX(#_g6zeUw|ZfhYcd1oE1Y)THI?AkYR>->m~MA~8XC{-wnT|^FS0wODXzGFxcGe0 zWqxf=2)n&vPrp$K{pdR-48)9_2 z6M+j`-;nxsr|50E{rpcy=OjIKqbB-T`XAtCpb^91y`U=)>-3kQI6-xQ$kuiBc!xw&EJ*|TFg!>GNW#912gJiUS{ zeHgq~-fpx)BjBp=+#BDS&aot!X$ddb^HkvCH&2l#xaNFq;Acb}ip}Sn#5Y zn&Pbvc@ue25$)*Ry4=-743w{ab8)pqQF!kZ(^l$a1JmX2O!^Zb^?FZ3mb4}Zp3JqF z-UALP&5?2-yvqWM=^7ntaFloOHwW8y+O6(LWdpLIQ zs7b?DlyFNA!S0NSEdO)pp!XiGT)^@fQad|a>Pc_Wk2QLy798~@2^aE7u$<`g1(BuD zx-GG;^w7!@Yxzts;_?v;M8s3kzvjJm{)CG z>&hS_6zjhpM~Js{4v_PUj74|Ldi%A7vM6wGMdpVBXZ_#tv{zDeLFY6<&Zh zfLt4^9^*2Ews=xam1_OQrL{DdMN2HVf0PvOgFdDMAOHT?V+9i_Av3c)Q6D)};W&u@ zF6@M+nsp;>@kihFsI+4+>%^D$ryIV($mU9JUOq!M_*d{Y)45*Kb_TzbBmCh9UfmfxpbxP)DGXYgFT0$s( z=b}Vnn*(00{i7;aQ$N0!-~)nx4TBrfq3pjwYnSV=moCfi)W@~yF0iyKo9bThB00LV zL=hRmK8%qqijuM^j6(Nh1TRS0f6d%bql}1zLu;{ucI}6Z zjXrx%wwnXviTx^ZK4wE8YJ2|Zw5v^K3aOi1z zXP9NtE2vp&ae&Qa-}cZ)nK!eR(=uyyEcnu?MypH8Xq1POsWbpl|o;-LG(*{*(y6 zd=o3QcdeWCV_2l2cN^W4{piLmlxM4Gtf2SGR-!3(TAj~j{Lw+qkl?x_T?6urUAJpQ zw|O3aT8IN)A(G=7x-FIi-ImA^TDNx-9%F2=hn`M&X%5R6mz5cm3c75JC_2)$FJEL#^#-autvZd_~=280oSM^RaO)Ma3M~*QGBAt>9a6mcjb19o`v`t832*B6d+tm^p14#h z(-;#Cl4eQP>~U9Q4t+}`kD;Ak-R8ouwc7<7Y~MaI2IJ5(+sTTY^e^<PU_Q*6&>nwlZFMv6iBupTJ#ROAxX5ucaczN%!G8B~P>{_hdq9j7x4oyk_4UZURjEU| zg%N49cE{`0w|JMc@ybyGtwoI>0d|jo;pU!>Y?`JmG2cL;cj&6epYfr$6 zLaifaQHE<_Y)si~@nWUncW$6VMN6#j#tzISU8@~Fwpg4z>_u$I<`rIAQ_VN>p%=}_ zgYgFxEx!$(u<0l(x}raU84t}gAOW(>3=2Sz+UyoCi@Mp|XqViJs={Wy_Lf>>oa;b@ zBP9m{?4-U2pdEj#fR52U#sFN2)3NUQ2|K6OuslqUzYkVLen?gET#OMBnDA){@8iAk zrG!OS(G-e3I#y*De4$C3NggszqzLH}<3NhP2yJp|7a!}J+g?Q@hQj;J1jKgIB}8Kn zt<;6w@DdV}Z8eh`_lL5mrr~MDELK~mue}=JQr6g4h7>QyrJTJ{AA`=+l(o3!wMn3D zIJECW>f6v!Q>jL*SFP=ohqn3nrLZ&x&4V_$8u_bPn~+fD(z} zG_kI)ONWoq4yMqmlsfrmf)0m}LMe|5zwab~DT=b2!nrxu=X-)5tEM$nuH#cNy3lu2 zQr5RS0c{|zb!sGpMQL>KJbBp^~um@H6|+? zyz5G7F0JkenV!{)T=|gUGP8UAl5x)lOcbD|d67Hr{PZ zLt%JOkWjDc;rAM4NE1RtLB>kuV+7t&M=Y`(Jvp@XT-Yt;iJ3;O0Vg2OIrLjSSvlNW zMO)hIQ(n2dwR)?p)!WIYJ`m!DJ3fxM&EGjT8e5=ElJs0CZ;3@Wgu-<20R6uK^MvOV zQ9Oi%#EQa7b@P1Z^H2+W$VTl8lZ`%^Hthub3`4f{a1zS*$)X zPx4Skb4tB(*oJKtT!rOv$anFjzK5Fr*+Z(ki%I=8d+MB?g6zg1(wiWQsmQ# zq4{BCp5aLy5r-A$_E?RWs^u&;Yc%q6bmyr_RR24w1pNJrpTj}!&}2OAOSvB+?!o6V z|4!};uHSD+9d3&oYpbsdAaBYjpsFgS*82@Y@=)7-Jov-uwph1PDDQD1&WrrURPoKq z8=8dDz3u(2mN?Q0bx4FlBo9VCCevr8Z8(He^Ynh2u-*m^s@8U}K83{(G5ivBvMCAex+fefurGENaf`zKD%${RUlK%;nc@(bTbFC1SJr9}-(H`HIyP5YzK7njQjL8`Uap^y@h2_@gLKA&FA_T^9!lR0(qR?#1oTHY377bvm2FZMX^H0wjbo zOe1KVv&5X0s;X&GhKM+sxS2_G1Ng@yf}do1#!Vv)eWRM+Iz6(L(bc>8eR#I_B^HqU zOcOuh>qoUMFA6-uIp8X+BloS^9j}GjN*0H(xhc128BdS-4RC3as~3|nLu^V3(~<&7 z{e7C#LzdMM03>-WQXCG?*~oEYvRG0x&~1nlNuhDwgdMJJQn$W-!JR37hKCg~Y*erE zYo>ui-Urq4)|FylP1~tr()3BDpI>ChE>C7iMSLSle52upVfqa)=$`>EA4L0Bch-n`C7Z_}2tWs`v8J*Fh&MsvFvoCPL#67wmO?dr+} zS1aKgF7U*5&-InO6Ix8m6G0R%Ym2uSORJRt_Ps5|$mLS+Q3rI}I0(lPp!QivU2 ztH1inOfws94Q~#i>fj<}wp5@dPD5FT^GnTt-kkI-S|{|^I2|sezQ*+ zM1gRo#5D)u-kipmUL7eNfsA6SsrS#CP0pavRxF(hE>(@r*;8$@vSQ+^igUQ7O= zupn0-2KsvJnrX!}^Lgq;l{9brJku@u9C3x-KR7bgdEFx{ajtEP&?cjTq{S|?iQXo!by&vgkdeFH+55QDnzJBE)G6rs9Jj?cC)n(p^Xvz z;H?E`#8Vxk_00PLdNY-KM`GEH`bbm>kTqn4pDkSu&p%GhnmWZ_T_(Px!A!z%y#VGb zlnfau^uqPs;t;;w1&7A0N(g!I>^1+o>vt zZ{^ovHp~LC^!F8&1qXDC0yJ-FDhpf|*Dic}?3S#lH^|gVFnDUwp!e6Y*I#}mus-QK z!j9hh#=n@Z%ng0H4zPh#pJD$S7n~QhgrI|3l5?Hf0{30>!Q#BODirFl;+SFv$|G|L z`f9%tT8%OQF3_P4QyelcYwe7M;Bwf5W7 z{(Py#@H=R4b8_{n8On9TWoM0kwJvJAYrdNHOVeQ&lbryOHV(0mR&N#nYsPcu19+K5 z#g^R-J@3$u!sVwVU}7f$Pv)EpVPo0+S%uEk?PMr?^-VZFx-8zcK;+~538`tqZNN7C z*g94>fXdRyycN`!N7#UmeS0|9MNmb0r=Wv^AW~y?>HHv~ZUEY_#cd>1gsIVn*y*u2 zK3;;iZV>;ap1~i6%-)t~YH<+mKUr(OciEO6RqF`N4$+sJbs)579%J$yigLRS;lSLY zW;Z{yYs>?yogsdeh8ycv%|qKKP3v+z3Rh_9lB%s$qz&==L$}%mZMdAw10tR0x~OtC za~ncLLP3Svk;@>yb}Uq+`jr|IU8s90I+{8(jmqX?u(WxrC)srsubl)F`*7NyBp(@e z)QPpZaEjA(ehqVg?O?7HvM^sJHFO%Ucy>qdS$%)P?q$d>b_(A8>gcM%$N64w|I5?f zbI({P`{Yc;mCZeQ;5$B_0IVP0*lpG{{zJi(upeUts!P1@Bn z-RUYbcZQJWiElmfuNN2U_LIV#%}5PlvyyHOgu#yi^w?iIyI&snozXk`C(Gky;!6}J z1tjQ?YBgiZt*2Y`Z=rehuT3>YxZujhX0@shL< z$?Nyu?A^-M5pv3+OpD2cZFS_OjwwK!07Npd&LDs*bU79+2#Yf*Uk!X|`%QHj3`1!K zMfbOEfgYiWI)5g21mi>h`>&2Hr+QbA7UW1hhX_&{xxt{2zY&_=cAloA&vB~h!rZ#0 zVJHhMO1~|PJtbaL>RGDdHCr4Es~xZ=P&l!S`%Lw1UquQkT;jVHE9Bp>MXfr?=A@@= zRAjfh9f@mT{AdUow;@|@XE~dr`AnAZ6Rgqvsa`Q;1wld{5;nS87pbk2}>l*nXAO*kP}d7MnDyL#w&zya(`r z1k?T3?Vhlk>>+HCD43`Gd{5?3Z_E@4MA;h>eHE*h8Z%YXhsophLA-mYS@y$dvUrb5 z{b^z)!Nm;JJ**R}R4VaBXY@~l`Xut?B_7dXd7uYLeet@mD^G=YdkIE#7uwRAZjY>= zD6#CmDB`skhgC5#zb?jv+l#;J&?Ol#+GBZ}H2m;{g1>97$=ciJ$|dphhbhUv4GC0t zs_X%fChi4HuS(3fU3J|z=IS3ns2rLCw4F@b4WMrv&Exs2^VA-j_bFfE(Rl$0J4gfu zO1$!HNY+TCJmeF*E#JOe+V}4j9~AoSOj-u0{_B(cvn&4KxcCgmU(NG-XDx7PsphRQ zC3w@WF#3ebB`)vIGIS4YO@v)-O)}FNZVSkE${np7VZ0NWDj_!u%rWq5&%b5!<~2qL zegIY>jYXcI!NIqLC5pR@=||mm$V#`w@TE?6%FRC_=+M62u^gIx*cii7WR3*9Iug#4Vm6=1@?rBB5JI4|-C@Di`a2~;T^TsPlr~>8WWD&1I ziT^y-i2ND1cZhWlDM7-&zkJ+oqJera(H@$D!J!1O@9Hk1Q0 zYa+n0&(Zj}HocUoHQD#|YfCoIf}XnN3T5$Fd|kKVkjHBi0ax*}FnDCsRsa>TU;0FR z5zvzT5cFK$8yM=WT?@Y;e7sKcJ}keM9c0C!P=D_O%0%m69+b}>)Fbb^Qso~7E?|a) zL?;%ond0tC`GBJ#+h6e^yB~4P@8~b#(8tTZuHW*vhu*vY=f_7k&fL;YF5fc0M%=c9 zs4SVg-xN5cepo&ppu&=~*PAtS9Q{5BHn`8Ng!37V-d1r--7CE2rLn`^ldY;MFi{Ad zTYe1DE|YZcbHKTxRvG?~ux|6laU8@UbCH2pRKtDz(O6jXACJ2q6vc|Mh+gSRSF!es zh#w_BuZ2exOt5+JMOc@v2UYrvFZ3zzaSJ{f5q{*>kPk_{UypEHnPhiE+QJDS2ZJ$# zQ1_hwLlBeLQ`EeaRtTHQbm_amWs~*@bsybAMFHc=ixTF;l4Qlx5XteynU2r3UDa|> z5!c;v-E_>#>SWs-PyI%K1zj}5lTYef#ozT}c+maItQ*={>`?9n22Z4juYbRhup!ZU|IE&( zH)cis`$Am1F7Jxn(?4?QmwxN!8Sq(|TsEM^x-EnO00M)r+q6B|>9ROJUxN2F`86;0 zZ`}5RN@nG;-3Q?5&#hEmePpoMN7t+^22b^bI8>qgW`F5J)sCZ|{eAJxH99``Sz5a^ z8=6}_3z=^YLL|G|Ews(h<#Fn`zCfqx79N@eFODHn#~T1|8boj#vVijRc^L`RSOAb9 zHm)_Mh}U$vn%#1c=1b8M)Dcaorg?z;e(ssY{-MsF9tMacO1fABeR zv4uBMsn3Px)g~{o)arQ<7A+|hdC*kcwWCa?y;2_Y)VqJNx@+`DKMMdVLjW5MgMy`# zUcTl>czZZ0_?ag?TF(wC?vX`r)R#2dxD-+m;5#qxEf~n5ObF4IgxP`f#w2 zW-P+6Ln07Csd8Yvb+8KOdrRhdz36_|I1ye^kJ4xm(4P{~dhU6fOdr zRxAe;!M0oI-Md;{sS0?EEJ>@t(M7sr`#$XmuPWQ_i^FLWj6+5;e$u4%xhZPdKIS!A zQ5SSgM89T`Fhvn+3e~kK|K^n2j>c=zyc!h4+PrREw9TaQJ05??oApb>kBj(n( zE1?ng`gt=@KrsY7>~E-6f_G@YVep3SJ{$Epz5P(F!yUkOm$4&=Y7{4yjFT^|B=nCSvm-y^U*^sO5MG8oMB7cdS1)2Z?fd&A!ouZ?ZPiY zp__7$($LSi4X2`Wau#f94$Chdej#{)BbsE|bN8TUl+H8o4Rn7@YQEvbv*-IV#}4pa zGl@;^?zHxeBipk6epD@icWRpT#oRst+<>&$)@Ge%-FhCOJ(sS0g&pu_j$2Hq3BV@Q z>4oRLG7UGMKMuPUtV&Q84BeRs>I-M5lYLZ6RdcvH#aNu-))0O|8BiP|pMCbGciTtt z$Gocs%mO~+rpsdM`fmI)f@1sM3#1-{!~hWZ^cAbYY)DG9150ee~Zi{ z(l2H3|NEQ_t&Z}^E-Hfwc68{|K8Q#I^bs? z6&HH=zwhg>#{>RA-8}K*PcMLfi*En=A-iMXB|`m+TJHS4J9&BrAjl`&ZGL-F%g<*q zboyZBo;Hs*{wvP@$1DAR`$IBVLmuas27ZmL^S{5L{3>{f=&Nmf^>6=L)W5QH|NQU+ z8D`}p3B)0^;oALQI#&Mm!T;l8-okwEo^@v;aV`AT0E2(N_OAfJ9||`5dyk*_j2hSl zNnC1w@7*823*MgCImN%VhAED`0_r!Wu-CuujPj4iDEIQKliaIcqeT3*@&4TG(oev$ z_bYtz`wkR87YMM8cVB#T`@e3#|GnD&nQ0v70IT-NMUKDu+^&F2ze@f0qV6F|8Nl0z%`v2_@|FX$D zxtD8^AIHqZtNYl-H2 z4A99}rO2xsy_Ue}1@O&JW6I!YfD@~L@$1(CRW2NWG>^rLyt5;Lr)P6?cRFB&FFeb! zcDqvYB=}b`9%fKzdVRD!Us9YcMss;5pso3rZ`hIHKZp~ zRW8|oH1-N8TG|q&pnmyuQArE?_YOLzsk%W-!s20H9*4>U>0+21zCRK)Fo*N^;Hkc5 z6c7osfL%VTuq$`nZ_hrrWOnnLNeK0HnLnZr&sIBq`R)_I4rD*@^E{H!o*<+^_DVV^xnB;vHLUAx9NY_e#OZxMW;P8KlIQedESH%N6fwFG^@ zFDi+Wan<^^u8Tt=(P;b6iivk>siqBpG6wM|>h|Z!rf>xVk2#hKw0!vS^*NlDUo@Pr z1_gwm-3Y6-mF8y!_m1wH6%@wZb$IdEjIt%(vGd51Sba`6@Lp>O%T%3G=dS~-t2=dt z-cr~k-#DS@ibT(q$4H2|c!5N(F*!q5Sla@@o^El-MoGC9021)9 zMk)y;-34-F$F|TRC+M`haaGLbNlBn4{});&(H<)OGAOpMTd4A)a&*$F>U5Qe46(Bp`e3K? z*rh@Ek}sDoYRE1^%21Ng9$0ieJkXrycCQKnSq?p)P1M3*wQibnB2SDhI<$gk{8?l4 z4gUaP?cw&AudATPMRjUV<1U24FZuV!=Y73lFZiuj^~XI6O4H`B;C997xlB5?rtc0D zCVj=0uG@iPi)UI5Mhcu}jvwPs=ciZzMolrhcfHG7?nJlO16u-BR>e<%?}UABF}n5r zap*n9XB|7#X5KVPeM^?CPqh6an}^&CJ6y}|nF6l$mmGSWbsmZXclt8(w&J1Ei!_A^ z)~~krvrS)#qyrHmRqW>eJZG#3j#{*TLirQ3uNr(whJFBz4&!?I?O98LSif0gkTuWF zlZW|9tL3g&0xtdT!BDP$w85PzkUtSHsOWbHWuW5qQNg<}k7JVNHg!rugb3K?glaF- z*1?iwiuo~i+Ws8NB0V<1Y3T-njh?PF7wgz{@9ybVKu1x_G|Wp#SJFD^A9ST4J6jFG z#VcrIQLjJcd(0+2a9e{L#IF+ViA2vu5fP{90BrZ`$o}Pf?kJ{E%Accl@ZmXjYgtEOA23%4ZS+nMuBVM@<9f7qOr-Vx1#sy7r zY788^wPt&$rY)CO|lpcUO5R^)$geP$_~ zQ}d=tbAAP}9e#uZr7nl% zK$0&_i=@ZOm(~*o1@~UR9OKn5RYYCodjP25=ks}hc5X#ubvi+j2sj>F0m+GYW_WM@?k{*^tOk1RvKZEgTGIp(uxs>WAa!z2p!AMs5L;ajiGBW_%?+zR=h2#Sfa z|Il(%lA|1{^4~0}`LqYy14#-1O~}>W017C;-dTNv+yjakTqNgROer#{N2a=yJGlm_ z$D{Vt_xr5?R|R-swZ_-H1i9^<=-bSU8w0vMVLuWMnyW=T1$vFfZ~)0nNoVQ040F|k zx49_^(ef$2qdiuF|MvYAv^||9&ES0*^GIrZ;0XM?It5^qd;>%_7des>W&+se8h zdlRPUx$GeF^7V!=#`(y5%7Yvu$(nSIB1Ct9?XA*kgr|ojU`lg9R$Zk`HTU0_BRU|Z z-b|U0I)Y}of93%5R>mEC7J(^oV)>3her4Rs%x{x~8fw$!%c z`O>Xhr1}zfqEyYCfu8E+-Ig7o!qO?VqUNF=a7y9fF1bH6fYs0O42m|sCR0;9a`UJE zY{0|B{fg^JopGT1=!C`f#<(*_lHgL-0o|7o#<1s0hFPi z&U%rCa2Z}C$!hoMwSvgS=N$S$chg1ZQosT+%hZ zQ>G%b9O7lMKJ=v70L3)Z7df)? zffN`E6#Ms&5djGmM-;zoX0WpP07i;%<^d640QK1j>K!0ld}*EZ@7e-oG*i zGrz@`+GP0=QZgDF0IFMgWB-HY>k-IzRMTga)~JhPR|b3!2)x^#ZAr@NbP$s)X?+29a@}hs*7USoTKDUHd2>k_f;#dH3!euNj1HYDS`V z{9T(YkbJ~HP+rY-68Sn-x+y8vKy?XC=6&@C(EFuCy~%i-6lm`yDChKX4`8BWX%BQI zS^*|8Z7p6vXTj#+zGM*qDwf2bS(wpDyj@MG8f-K<=8wHq3PUuwd$$4F)>btK)wiGi zyi>UeFgeCHXXDy`r&0c z++LEi$F9gskx56T(}{~_jn_x6M^SQqt-E}2HihHLpTRDm33VDwm!=EJpJ8@t z9eZH~a(>t+dVa1qCpI_7;mxnfmLZ|pL^ZI*{f_r7wb3Ae0KQyYz@q5It2n28x)*)V z0@!eO18|L$vh0H~Rb}SuRyhIDLg(0iJ2`hr>nciD+~`;!ZUyKpZhHx!S3r?Ix#e$IX-gzFeUhAQO2GLt`m|a;IJJTFei{3Kd-Cr zSwwrE7y1SzsvW2QFi|iep_OGU`uqo|$Radqb5!akCDKu>%FGdDwiS`AH2rMz!yVah z(&EDwE14C`0JZh|Y;}#v3I<|IqTFGr3Vs#ISkvkf7TdISWHjrKQ^`dMaPlm_->j1$ zdQmT#(|6FW_tSfFs_#7+>u~=QB>wIHq5~T1%xGh@=Z$btRJmv-Q}RK4;Fs_ z68`{YEYh|$O}Hwy>OrX&NmZ#^>4lZZDK*y&e`+V#ZF|dy&}eJD4(Ku6n`{bxno#X? zGWuu~v)l4};+T?-zSqsS<3EY}h0Cm72T9vr=dJ&|Uu3s=eDABQ#x*bg%jdKrg`(ae z;@#Nk)u^0WvPz4NYtgSjI2#)31x@*V?Uj;!;^U+6m} z>%g?e;-xNc9i1Gt_4CrITn!6Io%$G8g@KTPc`2~xQjeonPQyi97J`A3baO_3wEaJB;)y7 zjtjg#=8d8217Lj7j)p$?Q_T_3!jhx+vO?=k9*#v_&=KJEST5s6_~mpKDkB|^#xWPS z=|7yKU0b3Ng$x#bM0=7krrn3<3q`R$&-n3^1&;!Kz@$`_-!gd(fk0N5av-T))`rlw zGme6KCJ3FNcsm&wqBZVX zraxksz2<>`qkqtm=%c;N-0g`MGH&a(eqmG1k+DF7`;Svj58b2iEf-T?TgwwWR;1zW ztCY@$Ag7wQeBxCU#VO{k@IWzd(kp>(;R1)N@s8IY!P^DH1#51MZ97VvS53hA^4rUG zZo6|U?}|eG{Pte+gp{dOi$36_0luSVwEt_@AD16MiM#&VedsM8?c4oCD?y}4fV}z* z*))%ZLawYD>gP(KD;_xC@;p7KzJV2uN0TcabQ1}BC+$c zd)Q;g71@_&o&yTEJF~Hk4_bcm+s)9ULCWQ}$hqfEP0 z=H_nf{Zu{*&$`#0<_@a_q}GznCDW}oJdO4zHm|F`9Yw|h{wMGXhcR)rt@S9C7=KlG z8PM3QZ4tNx>iNqbbw*tlw*tayuy#e@Mmfb6yq3+Yl{NJMioz^8%*SyFkhW7<@aNai z(D3IU+YBd~gK8L=)P)*qy)5Hu!*Dsm5n>zPiyCS9;$JQE^ygs;+L${H{wTofEuhcL z`hHKgh?obF5c9m7`sc(X@6AV3Go2ZyIze%j3Ds~xqt;k3_GMh9y1nZM+tXlYew8Q7 z6gx2wmhI-q=)$;L;Vyw&6aWAffDL>_&(Ev4I%|4Y3MgNvBV*5Z-Be%vRs2wnvx3LZ z*S1gib++E0Hb`Qwz+(P8aPQgABUD1l+(;!}RRT$jr8!L%-jcNt0Xv`Tr=*dUqVkBS3$*qotm7^= zveD5vSJ&khFM8=QA6!C($4-Z~S@hPuaD0|Qys1=!y%65&&>Y{Elv>oL>mClnd9;ct zwk0Z6I-*GuEaZ}3mctDF7U4|*)}wphlLb&3SuLt~03lPh&{N5?Qyjc&5Ae4TXgXGhAISV|8m5SDW*kRBCGVSJ}`t=xV4 zvvusyP8TSj_{O5tq}1vl9h5KT>ksG+_AJ~a(b9-EMKAB-0wudPdn$a*O+84y0euCv zn}@&PXF>Un>v$2x{vE0Lm81my-`96g;#mOK%lA=lJB9G5zAIlCxU1{#UhGH2&}OQ} zYbnwKRSTnwR5%%5J8Jb}v)f9Cn0VJ2l8Q^mx6?VspYLsVg=W)D3JVO>`fU^KOlj9E2W;x(< zpwAKGyoZc3EJYd_)WY77Ljw5CY}SW~!WhXcE&8%t!>Ao)tCTRj;P}fKe;1G~Sd86+ z)5X3xp_8cCsccZ!Q`PD}B-?4#9hmo6^De^4oN)qW`=q(Rq#m;D?=*$Jg^eyHcfF|> z@5Sfl2kbdEHhr7?*(;F2q!zV0Y-b)EBEM3+H|dx!5zm}y6QW{Q z!37`s_niUD(66A{oyB^Z*VMLG3qM={yj3HOR4HGM?Fk}tb+R5hPVEe6ant%}cxkxQ z{Z5n@cb8*-MsUu`5M}1>>$@Co(8lKvNwp@wY$!%;Fe008`D2Bs0~cg0*A_*;J?EfB z#9wXX_N6*z>}k_eW#n=P&g?*YSfS1Zb;>=y#yj_Gp8d-iO8|& zX*)LTS<|_%DnbC<<;5>I+ka47$_!U#Wxi-2GwEz{gZy<$eNFrAAUOzWah1Yz%A>KeP6hX*Rms_A#u*5Ny;dkgB*RL+%Xh9 zen9TD9?TmHqHekVyi>EZcSh-xc!0Aeb;)q=bO>wp1viGc|8i1;d*|zw^`0-Hbv>1b zwG-qdC+K%xU_`*%YEJ2!3o6$qP644x80cV>+%mvso0|N0XUStN_56 z#*KECuZ|#M`oO);n(D_%%4l3BC~sGx4m6$$QPdJ$9i?~vMea(i3EVwMA zacWmM)3H1!AgOw&;Kq_^{UVVNpv>)mq8=QN1)Wi>&DtDsZr?@L3G}U>$9f_S#Nq`z z=DugSpRn}65k}AKIJO@35=q#jA!{&8)^n#UzHaSHyY(Wa#$*2YyX0HMF&ige#KfkD zd!{75RluU$?P+58&Ar6(T?Npa1gPiU`vSv?HL{j)=3s(I}q`KK5AKao%!}e-Kmjc&Hb%M?pa6s&NNqX zq#*PIj#0Aig<5z1l!{7*J9kTg;LV4&XlB*z{s)W3{f8#t_u^1?`ZX4jBJukr1i22l9=Sp_W&3Vr;Zi+oJ4UOWqAir?LQccbEj{Q{g-f&T{UbaL` z6w{DRqOW@>Xr_Q;{r;XxV?~obtP_fwGaflW71&xh-Z@i1zRBF_9o;+dR7W)P0vsMw+h*l zumbDEs7eP|LVZ>$d&3`;Hiz$xXjCBK(sz7ZFvkjsAy?!ahOMwR+*VM_CHS!mk}@Qo z3*ez<){Fe5M;(#uq5Oa{Y|>LymBx#S$oNXGg}tX4-OmD2Gl5<3JYy`z{g9ou9OJRt z_9cPWDZ^WIbv3OHRgVIJ-loh zGUK)-fpj!H;R~%))cz<){L1_#u-)m@y5NV&i4K ztU@1khts>(ls}ULv{>t2cfjGZbvc&YeB0%!qwmrM9sAa|`~3_^9BIxR-R4yKq>9(x zMxHK(7WyRr@40_{LqLu{^j-pM&}Lavll2K91ZY3jELWX>N}VqyDEHnD7hb~O1HGg! zRP)>>R`H;yHM!d|` zz0Nd&Ey!sWe zE=2-MIWkse(wnV%9Q#+Ds>48_#(L*ezJI3Ws=%M?O*FlH?%>s|MdcnVAOv+4#M|1{D}Bv ztrx|d@)Cy%<$53k7)X8$%)gO+;5+0V^ckp3n|z#+NE z*P7ui!|)B-UX;bkHl+c2Xk}6y({847b_q)I{!CI&Pt@|oQIhJKu{69#;cUVK#Qe6% zkK@o!T6_(lSKAz6$aJb+5@k4PL3gL}Xfs`K`qzE1>Fn)&AoeM~D?542oewS%nKztK zej0u1;CIS9-|rVEUNLm6t!D1LUG>sRU&mKF)5c~rCVh9bUXUC0-QT%#(E}8QSgDb0 z27Ep#jCe$cs5+53ql~%capIw)%0Go+Ze^q*j zq(!!FPF#JlTr7pJnG@^PL(_JaYozw&$;ijuvt>+}38U+;HM7$~ zYIy1ZUnGj&!20A2T*R-#!V{jAXU}+CFw>bh(lb^w(l2EU^!A3^qdoJOzS^f$Cow|o zn~-aoYK%ve({yTrnit&ca_Ms?pX5j@W^egX@k)tzSgb?VpF4Badk%Uiuie=xd?`Cx zd3D+I){W!CkT}yl^K#2tsg6dT{oxe*vtjY#hFj%aUtV=@9<1}-{L5=AZn)*C8sojn zk9yccihgKB7J3$4Rgg^>!~(uuaD_qXJ8KeHg$qNXkKaGf;fz6F77dJHHaFXL(@xUk z3!jti(>Fo_Cj%}UqtShw$xvkavc#xOCF6=peYtCWoi8m6^4cic5kie{E}nXA7zs@+ z(oHQy|8Ork!RQNjSW$78)1QDeB0FX4sh;iz8%B-3kb;Pc%i?Jtjw!08dpR%gUzW$SIA!s-uGh;G3jrQIhtHr_6qVu*Wn6V#l?y;=HvAjc+&~4Pp%1Ld)y`enI&(`8oNQ z5%)-dL?cPUrig z+GOmbGQamb0(L=!b?FKQTiUn(qt2B0)6umMwZTExsYNHR(k3;ycVpm0vV%G5Wd5`2 zfOTrL{VzB;4?R8Sz^LXiqGs}_ldIk;f<}q`ZAK97J||bcAJnJy_JXV#zC6W|lZl$Xh~Bag%D_jv@4}f{6<_EUTqkO;OiB z)q}uy32I!O&Ge_9hH8H@{LqEA6*kSqowm?Jo}J>o4c$-_P~>3^*F2Xx54+%8K5NdG z(hqhNGs}7EBV>tNg3cH|B+BO``MY7dt7G!#+|AhunvkW%EGgp^HtJaqEae1ua!3GY z`GP>g{PS*m>bX4|-O~M*^0x~`W@kW%=&eWR&S4hr(zeX z86gdoUFA+<-<-MJ!&2}Yu0OCm5WYx`8l+?}Zy-q(^#tl$fZ-bFp5PPBue%I(S~BL>?K$sgkxx_*cR;ywJ1!Z~ZAi)41pPEK$d@Ti@?$?0 z;W-SQR~bv*hF?&V?0}CP^7!iT1GsXU)%ox%3dT%t3oy=k%23KVfi4a~PChv60+Xn;i1A>{oX66giVKXYnY1O;O~r)@Rj?1~tHYb~6CJ_p!F{pF3pF1hR4J z(v%stJK>=uK)fFvph|>@8bPet@A5Tmb+JzOW`-ooJyORsPJQldeg26P|4O}3TqXyB z)=dUBsLp zXk=uIH*;F#*#C-d%SPj+5~+|RzpXBF1@HT0ujJ0w%K5GZ z42d?1wk}u2-1Ej6NiI{ER{a&dj)H7!;_KgA`JUeGKfVuRUR{Fs(Fk_>Vdf7w$U49% z>2ryOc(3U0enz!3IGkfq6K@z9Z-L-8|I6IcruywRx>i_Z*+}2#%WO{ozoH)Oe19<{ zSlPJJjdb(fych>J?@r(aw*!rojcX5KT*LL62|L)Kc=uyB*G=_ZVEuLd5z7Oe`}0_# zqWGg2y8rk~UJN_L)p-t$&aub&+?HfW=A%6^NY15vU%#@}4Ox0UH&IU}bhW;RlxEfF z6`q%PAf`HztIrpQPwc7aWTE1(BhlCn43Z%?X?sT%$Bq4cu!5cg{!A9#8DSynXAjJN zrb4$j2^uG>H@Ne}>62;}U0(28cmI@Y24r0JN~BF#GCh~J0MG$w8NCiZ&le^H)W&#f z#f=3>8oUPW89{&wu(D%T!d*k5KS&zXI<`3^z4{Be_dQu5zm6+HH~i=5B<6&VF{Z^1Be3;0$$bp(Ofn}{b_GS~Nn#Vh z$L$XnzV#CcKB0EoBW3cWM{*YW64n(Md0+g?O=SDj%m)`q+uAK)m}f3)joH3K2fw9P zV@&8e<@Pw$xA!BFEcvsJV}fRUXLMY@{+f9!{h z-{=G%RN?D#oNXH3wGo8m^Wg>pewZDow>K%Ve6u1bZh|mDs~6RuDs>R`s1uV4J3S^9 zCK7Reh<5g)Fn3LeBQ-azK%A6+i0;RJMe2sH}UKj(59KsT2sXM(!buq}_0TFVCpJ3sFyxT7B<7O!}{`1~!Y zZL>2`uDs$dnt*LPn3W%BahOv-#7JqGb#f*T*+b!Vn!n?j{IYocBqIxV;y4HPL7nfG zSanLs0t|(?<&^!cU!mm8X-&l!9z=&hi?cht>pV5xolen8 zFN_fO%1ubQ;vk~YFf5OQoxIvlL&wyQLo9SF3kQPJoJs=D79|sZ<+0IAlG2c9ANwsU z$$%XYqVLYYg^DZV*Hm2!52I7CKH?t>t8+_yV)BjE#4oOnWkV!H_C;{VL|&XUy^CbE z@bIy%;#3XC;L+vyc;z9&{Yt4rvqb57zeOJx{VEmgBl=P31DHc4d535&t(j{k;Zc~4 zKsRH>0J%mw@>tMwa^ufLz&7+2GWZ#@#-7X^UUM(BJ8X}OK_1>UA5RSzB**3rduG1k zW>IFbyjD_EPUDx7XR`O#*C3dko=`;Kts@nRqpQ$AnNve0*P&E|b(e;5;@-V+6Ppa^=Mub28 zz3|r%#CvDQZX_t~2y>U~%)`|0|D}^C7w`1&&Omd` zYMprAg33S+6#G)@aoFtU_-r%Rjk#Ie#vY9gSbGLbqdRb_Iegpu!`OjzAC$72mBY8K zgpkdPBY_V3bkTm0S?|z-*xq0=UxpsPGSTm%DW3)NfV=)vdWoIW%UoD~%-o8#MWEU` z_ffRA9{c)+?6=<+*d_6xb&{TmVd*2a--r62Y!jDmu1oTH*$t~@r(_>?K^wn$y*f;d z{>FDjHQSH=vL3xC)YS9zeB;mVV!#V&JC7|rkCt_2>+SW)HQtSaS+rjc=h!ROQ<8okA>q^c%>`X&U+0tIItiab%ysz2LjpD)JNR7h z5R5Pp%2+6Kc)@ijKZ~;=dqnl+>)i-y*D%1d7!Z8z1ABAg%*%$S*FCp3ita8xAX7wk z7_qIF%e0$@(y_vyx2anTtx?o>`I7MBCJT}4-=v|&Rc7R_YKHBNT)Fzj;LIFClHCH> zVHJER$)oAuU8YTm3mgMtpXeW`Im4|7`9O@vGyMQVWZdtU+=>3m*AlR&`b#9CL`dVQHYQT0h4T z9#}_9kaXiCXohy3zoJJyNyDajMz=*9AVXoYJqRQ4>r1?t6ZKEe44pYvzdbArBv%)I z<)p~eX-%2MXl{EREbEQdtc5f8Tv80C$^T$|9*qc{|vt zqI(1P=A2180#)K@{OX^h`46Vbh5OImk}oXyyO;g_&@Yy%(=jmA0rUPqz5t_*GWV7U zqa0l!xTFad(#q&{opMOb?;xGX z*sBo*>lQqgU$BRo(YWMg#ZYXK2y1{Xm{D^Y`B6=*j+p8*$sRb4tXmlFtCX;+Uifi| zPJe9Q6)xoi^ZVuAN4ZL z{&#@(k4d_D%yxz?I@a7f9otuk?nJNUr&{Uol@nYVn=yCA?@AaC*V-9~a2j3Ra3-iO zglz0)$ESVMCtS7on~ad-;h1}s&zC=&KQre0?uA21gyp3z%0R14&n{X8_3^GAj{e5d zeYn+TrF&T}T+#^nQ&vx@)4~;!04ZjwIoCeU8j}hB!Kcil!{SKsFG0GUov;;j`F^+l zNr^@er(&d=8lU+iMZ?!n?~O16-CxY^ik_*%(YcU=W|Mukh!J$OB)}5_tS5muhd{yk z3zkN5B3BZenA&V*ie37J3X0$*^&2Z9$$|CU0y1A>YTi|M;R7@bjJKY<6F`tsC>JE{ zGow}*S$HWf-*foc9=WD)$}|+dDZ)Pdd9=)eyp_QvRdhZ9)Z6|E$eh2D^2Z2@H*p`X z^P_1GgL{LSZiT^Z3zM&2O!-mQz>hx-+f>oYEmEX+$ZE9ZqQ~1P!>+oJ=kEt3k7%J0h8^z z%!z86fkY0&dp3$_8D~?1xaN>Z1m#aKJt}uh+-Ei((LdVt*4BwPA(qoY_JiG09z@St zt*t>Hme6-U+RtNpWjrA~l`CU@>vQB(F#8a-(gV>It~?y8v7(rhe$;PpXK|&TOOM5U zqL^PnkLHE)P`3E3K~}x>_&3%Y%T|wxT#pBWxBj68@cTLapXvHDd-t$6VR2h-27GSp zaePIGbsoF0YeFvI02I2}QY6Lj#^ld1ySmfs-#Q!fi@fSDp>(N^*jJ%RQ!p3rST1ig zqXe)?w?&&eFT_YL4e`U*6SZDs)xd)1k(bfWLs+*&WaAAy5@!*RmHOozHR>KxHKv3n z4CPaJ*~Rz7X$74jLKYXy|L$FqJ;nOk4;>j5MS}~wtHtkEy0!Sx=5j{PK^0WW zW?jjAWe{7_ba#l_z-XVxu^YB|70_F2NC0I!^`KrkQcqS8DTb`$PwD{^OvkqMxov@m zai(NLqK+2-8e~vz--JPU`md<+qT~U)+ICnPgByyn=+6MMYMhA@TUdP`3!VOEK4qv< zPw+b5>v2*H389ld(|w$GWhhh2JAVCI(td~!b_FQVEz0<@3M;iqCwRwgIPX+?fxVLE zn@G_g45N80a|PN*)0FQ9uXamMs?0VmwI(=m=p&bbK1QbdCzO4BSF*V80_*A^ zb$F+P5QW*;KcKm-DCl6278%E3h?}z5%uH!w&yJpjCgs zx#|pmU|%Zz+tFlDWKS>}TGiG+4LezR1Asl()TS;QR=ZB!7IEl5*_!aYg!8eJS0r^z z)$mv4RWmX7Hme=ZcEmWRFBI80W-4W#JbyAOJZCDp5S=X%mQe8Ar({XSBxYKJEGXx2 zl|3pdI70SRW)Z`|h4|T0PjxOtjWCwhy_uUy^Mi5{7j7Q}K_HPFdA&;$I{fC)X%u@H z7C$2psQR5sAmDcQnYdjhTW{d}<+AY$JW<4P|6b^<(+{VYRg4z32%Wp&x8Kq;xT5mE zS9k3+j;DAt>;ii=q5L>ia+vJl4G$!09#wenC+=Vxj;qB+IE3KO;H*TbF_bb`xwzVT z$+IwnOWQ@F;3F>`*Hyb~4Z@>{jH`zue?t=bZtS_<7H$L;t%%>^|3Ch@A%F|;;rzN@ ziq+{a22|>=H8jd_1~ScF1CK!dfT6>N|*pI<5@)t0p%Q{V3la zg)3qLQD&>GPm8abDN{U{K zab(xrl4{ejZ+`he*Jie}q`Yy0llfgb-SXFm3&r=CXz_rn*G1&KZmc#r;kK#VoSSxY z8f1ZsSbLUFA%S2%8~WSRgk%FI@0yfaH^YBC*Khvw=IjqNB3vi`N!8ps>{5;@3bw=s|z|NXjue|Zvx^wYcg(GG7lm38Syy_TK z8VXr506N_tW_>3T$Lstm3t^RAge;3OT(C?IOK9DX-q|YRzxJlkq-FpE)Y9QSN{ZRy zp?HwhYxwA=2bt&UHboWre-Ea>En9K%DwFU1ftB=M5+av8f{=o8gN&q9ev5#ih`FB6 z^EI5~fNB!;#$29MoT}xF_Q=%paWu`}gT5TBnlpPC*ae?h0)p0_Ro$BzUqew>P5I~q zg{d%Gq6SLEuI;$leAAHGs6XIRyF|Z$)karfJs&#-Pxk}bQBxv36>zja-h1w$8o>tc z-=jM_HX_i6bp~1lsl;!in-s?%VA3F91nnQasaoeNgip#%>$N7M3>Qa!aXv>_(R1NB&H>q*nT zUOijR1~?;A!~_sn{hXl@IEr&jgAX?ArWcCz;#q(d%`aj+b0w{!= zZIj7BlLXcrgn-Og0Ak|N;;>=p42Fz$uay6BKnXC{+1p%-Trg4^4Nl^dhsrZ%oyHcg z(rYh`p!@TeRynw@APBoUt5?|y3gW?jF?+YJ*e98G!75MaNwyaU;($heq#U6r9I+LMRZq2SAly;^? zeB!f}{C|X3&Rfp{nViAVH~zGFAC?JPO*PCqT;g*rdGGnOor){dPWowo{H-Lmj|b%C z30$Qb|B7mN*k{kV=8&z+UBo>{6$$+O$-n<<&^i7~u%CWbA^&Pg|7n^3=ik2G0}yVp zB24?lKf%Ypx}*Pd(q8ff+6itscl}L;=)d`@peGlA32&S%CH8x;`oDRe<8?ql%K+Q_ z9@PH<&i~c7s23l-6w^qlt0{TQClmGO^xS{A}yUBSxal$9B6;aiT`aF_CHJ;+(+OM zJ<9w;U+BNMfqzV*dnLeVfAFU%c>a%`1HjZk=gXO%KaueM@e_Z47nfRrZnXPrL*&8V z*WiCQNq}z+Is|0%ZJyoa{~z6Yn&@tHZM8|Y{PzjXyF)C)d&WCu&r#UlKi`6#J+8U`?RwOgMaSZ8_eAfC86zP8%pf5n48QR=cq1^Hq^(&9jefQN!j=ioo{%p;tKYn<@P#{ zL9Sw1Z))}QGY&$?&N4Q8N^#jMmd|_DfAod)$neSuGbR_4tmLh_{fp@E_0v$$4b4{IJ7{1u5%ECwgG5Z)&kBQM&vd5^)=R zpXp1Uag_5Kluz6oIAFHj=`d2~7_x%x_9-0DS&W0j0qndI?HOhz!DWeQrIkvzdvv&u z3r@Z0k9WZPYzmmkGZwUQ-!E#Z6ZL}UAxD*eViI|QLJoq{mj}0TnAsS(lZx*;swn)7 zzmy|Ps=JH` z=8sl(wr(K&EZ|DhZxWxN)_fltyLTe+@deV*>J03xlr&_$>l0uW^%5HZtG{<)q}m4| zmiTc)wR}XbG&F>TZL3v3E>Eg;DDOO_I8g7sPO5P1TN;B~qm9?h|vZP>o0Su)@g2W;d- z_v=IdZm>CUM!y@tE3&+ju!=D38Ca!e=$G-VdnTpFYE1C^v6Ws}9BDw>*qRim@cB5S z>r*-NhME!$9BVxEcKR4jqAy@g6Tb!_Ky^2}La?&jDf|QG$Hb~4r3L8AR-E%Iqh-aa|&XoWvUo}a|EI!-R|1zVAu9d zMpdRmE7`zt=!=P63-sywQ+OaEF{WS@Y}WVvTGiWmT-OKCko2i2SYxe4j?uV3Vfl*L zHuVZKph-J_sN{XQGF2W}X9!tnFwD`}C?;^U$R%F#^k#G*2EL8@IhvufQVQuMZ*nd( zUOh{Y6fk>%am*MyYO&xuSpKvIf*5Wf+>ju4SXhOsCH`d2z539-O{}b67#WndP;<`f z6j5!2MmP`Wg!bD7-keRT8}GDxuaqsBKfqj=Ud2EB zNto;+w@#y>JrlJw+39KeQhN)W(qB8pjJXW&8fHz1F8GlM+5`ho@E5I7#wW?PgQ6h; z3FNyx<~B>xiy|koP)X}eyosxHuy*6ph7k}T7JMjK1DkJUw?N16m`kZdC3F@~3f6@f z`@E%xlBT{stF%L~u2*PUSFVRR4>T~wnX-r(bhp-sF0^hdSz`HDD8Xk=YG(OjsOL_7SqDS!Gm6K zOU~eni1SXUHV@ewba;|;GDg0a1MX3Se{d7^x$^MKB@GfrO{FEGZtjVo$K;Y!w+gsG zu8v2qYC&MZe>y%OzCL*(kvvXY=`p@YzZw0r9B-x5qWT#|!rk^TKiC<(gtS!O(Q+!X zU(SAGo4EE}KatUPRYY-T(u`-whF&+}{By7z=~s}yeRzFinrCSi4vSkDZC0$lI5#e> zWzkLW9doO`JerCYS|s|QeM(RNX{i6j-`1q{6?YPgZW*fgztX(V|NRh324(mTR;W?l zrn;Cg=Q+pV8!+!(k}c@{Ri;4Sx8pW494@z#`Rg$R(~_K6IqbC(No^lVanHzqo9y^L z-*S|Vyqmbe$|zU!mVSxVTn3cu}0` z=ebm(f@wo)jMNZ;0<~)*EtRJ(@%aE%v7*ehFkz#p7VR!!uDV?eJth}gZ%jtN{gZE&j$<9#86){W^?(>+d(JUU zhZn@lBT=gtMyjl`zzoJvV?D!r%0lacp@ImrK~1M|(Mo43qLwuWJZB)-@B%EzRfzP# z)nIYy{`Se5n5gjd0M2AR-(N5r+xmOeJZL4oIs$TU^M=+=l!t+&_I5yPeWT_IfAS7k zx~jjTz&zF??1xM8#bkM|fAcK*D0Sb%*u?Ri$$K-mCB!--*fRxb@+A$t^{Mi`k^Mzl zOg6z%V9~ekqgSnZ_fFLqXW>AO3uvV3fpJFI_+4-^t1G9%kEx5tEN!A-ul+2)2Rxo^ zdJiA^rlvd%M<4ZyT##D5VG$fk?1b6Nl1@4uOZDxYIV6ODEc^8^MtzWjeQM&`TRqOg zq3kHXagR4bo1deAQ4%^{5wMb=HuP;IrKF%rkozcytMrM8?O~vdu;%GBt`g{%2SGd} zd9@E0_p%4W*CE+_SB$3T^+6+Emzv<{?Lwm(n7ahD`bkrYn%Cp$Ax%|E(CU;k{yW4+ zZAg6J8^*VAepF<9HCIwdZA@%y2|w^;qWChNd!m#HFH4Xr7^7E@EUj9)Y4j`j;bvJa zPniBCRCcjMT@2^eHT?Di2OsZ@`H2!ZWl6)}(e}$43=PuDXGjE<%=S7vjA}|JHjy5@0GvuDfPLT~(w_J@VhuORIYR`1PLiW6WF3569xP$2WqI1U#AuI;;u{on^}TK_BYDODNL2HvdY&K#`wP{uu5}U_BmxA~N2Aq!%|CGUw9sj3p%H$J9tbWJitE*hFIE z3LCJ)KdwbN^nQ=N3xOZ6wd8H~}KqU)r0pl?+m>&@8xDi%y~A zsBXJ>Q|$|wfVZTC=-oM~+uy&7Q=}1rj{}uvnnRzYyGt6r+YEUjyq5kn*`3s6Ns};u zEI-D34-zt`FRM5kdz4wl4vir71Ul0`O zh=|e=0R;h(-a!OJ1XKteA|fEYh90UEk)qP13(`yIy@ZZ*5a}iKP(uk2ASB#}bI$MH zZ`^af^S!?J{&~j89}bDZ>}T(_=9+8mIUV%N#^tEsu6&UCb5|Rm*;j5|E;XluRjEnM zbMgwuZW0rl&h)ODV~52Aq*d-0om4k`AZRs&9InkFXBhuWAcvFu%LF!C86Cn;?pIeO z*_42=2cXp~))izMvE-cE&DBr)ht>9dYM^4m_M8lmDf*OFY@5>Awt(#qL75nrXRL4V zD{I_7dB*I&EualMyPtirZQr(PAPZ=`{%;(sxWr#DFeFfvB%fvhH!giGXUh; z^*d0UuncX-6?TBR>-_gjs|vBf2ug;y(gP{o>&C@E6$bMaKjW%b z#$Y8WQ}e-30OU7t*L`JTw;C%^)_W*XXhx-iU$_UJldXQkn(iazQHglXe#)<#mgpKg zZI^tSOkDXoU=QMwaBoMq#7xb`*$0`ml}R?HOF5#0{K@fF!67p4p*9$R-IEq0d$m!D zrblS-g;uW(C~kdBM?IY$D@qJ!J5Uq5de4UMjFp+J-N8|ep#x}L3*cC4=LfxyWEFK=oHc=i7O%2NK&sFS@emtd&Rr{n+O$x zg{~&a>}_zfHK)x2k4_;UiY3i{+Gr|w7CEfF%%!|x<&LgUw_DN?--`BC`sHXEaBLdS=d z+dz(Y%b7HxTWycLS$Fjzs25~CL^_ND0tM*1IO2CFVPmR>m;NzNoW(D`IEI^m%m30P}GjRcspi zU=DWoz!WQEZ#n?+;~L-p8gPtD7=^a@j;xmyPpM8RAd8F}c4k_CoDbU#$&6^L%9h*; ztRls87s@ijp?vgbX>vY+c7_cjoNvlay|r->%6_`P*WI{a{1-R3HtlR+K5s;XUb4 zx9$mLpISm6rDf;@_TctlaWx2pc2yS>{zhhGW4}Ms>7Z3_yXJTB8dMa`>fo1lNfxX% zigoEm0vYt~3{ZX*eoOa!W-RtP%cjf>hsZye3sUFWO?bD=`f$RtysnJ=zd^A7RdyoP zWJBMySRBZhjBNCrvi8Q7OF=#A2Bx6f9~yiO*t+*tX-j5S{2GzG_UX7GDYC3Uugis{ zsy({XWOryhl=$S=DEQ*uEotzcs@r{|W%pVSdF6(Y+twgPb6pROnpZyQ%#1a9*^(l* z6XQG8K+%K%5`e*mC`OF%{%o_QyAI4(YTR;2QH}bxAgRnz&s5X3`&cj{BbIH(ph0I3=k0Sj9RxIT%w`97+=$~pA zAAbE3yOTtAex(gB6kw@MO^&J#ijlG4| zknn+D{5sz~TFk_9c4BPJ8x?c=G~HV-0($CK4Ro>7VMM`-nw)6lZ(RSKTHiCFa zEkE3wp!!Wr50-Pvo+^GzdTRE+m_+^0C;ETonu59I_oP{?rNrzYaf7?%Kwg+!>~<{` zN^2U>#z}%0FD45w;a}MppXQc%_neQ_HpJsGx|l}v423+`%auKqvia5Eq8GR21e9CH z<5Ui6dcJrYz~{GKgf<>1$Y|Zl(*-0=9uHp|0G zo!Fnb|E&=K=UQXy+I9mqpUw+S1qN|k8S10R3W<|n%nS6aIo>xy_C$U9FNX2|*Jb)oi}=~(yX)Cl5d~pG%}y!1 zT-^ka8ZD0`fiZ@Q)9y+FOS>BLRY@QpIBi?@g^QThPA<<<$%=vGbzUvwqtB;9OU!s< zq^FYO=okeeQmc4*Gc1AqY06E!m+3ZRaS21^elidg~Hbj zHo#(H?SPnAZ*AN=cFkD{onoeUcB&1qrry2z+tcN9hgy^B?8xbzq|>>6R#|rQDo@CX z{bcR-My;4uMe`K$u6JuI+Iygg7{5j`@QiL1WF@C>c-C1txzvR6dhj!T1FIya@8gE} zRJ~rI8eiq2YfBp124Ru9o%W~crt6aZ`~Es^DPuzR4FJGRMGQsDh0?J>R1rM)m~m4T zQi4Qj_8!ENa<%xO%YX)Od1K>e5hitAWXpyji=%BxHucMs=@3#x0Ur~tYJladF>~|S zMpF^M6x^YjX#~3$8VNKkVe_QY1d01yM39($_2^`t)&zjtoCZ(IWvC@rXzrDOtj>Ng zvI+%H{;->y%@#!uERJ-RO{mm9@&1hhIi?usCMpkFWaFC z;a0)#z2gPtRKl&6(4k;eNzN&9+|-rh?N{_O-#^(49RB4M`l-C+i%7+-Y5M4Msg0EQ zsC0XTG^QFnvHby7#&e{#J|l~p&jq$e!h4X6fA+1thTq^xj{ZP0_7 z_Yf}!DO{rAYgjZ`@F>nS=X>pRE=GFSZ)u`i4t^K|H{20bU#91Tn&`Qx*H;kgO|BKm z;%n(hsOSn6hs8%1fz07x33p9s@@y@xhT^!uiOr*QmU^EEy7f+ZLZ>nXW9;GqnfUMU z>Fe&sW)$8`iI03@mZg#lz6;zI9J%=xzi<{>j8nQ&1kk7`3xDHkkJ+2LiTLY4@1_6| zO1L?qb6iU5cHHMa^_}cu;mT@Bs%Okm6ULwrQB@3or7$sg9^E}d?{^xFaK)Ytlv&S@ zlu`cP0&>BOs^3{~WbKh^^5IfLwdbk9))BVvMSvCYplZo9yCG7_;`2=eHP?f9Uk3CC47XupG&M5j4Rh*z!D9zr@P{l?ofb_Zzu;8D zp>C*|YXxF9)X9Sxj;Y<0A>4oVG}%6v0Ye56`RjVO2}#R%CM5w{Qynx%$7KJz!mQaH zRHH~ne*zzyZ#s}Q6oIBuT;LG49|~<@pLbApQg4?b*wfSTDdNPKD!d;)j-vTcL(62O!h($y*`DY3jma7U$>35FcB=)_JqC0m z=PRUm0RZCBwByM=Jp!*EE)c_k={B;iFt-GE&w6}$@R!U+FnhIc>*JX!kr@Ro$s#w% zmh0)p_|7z%v}xLt?LN?8N!_2dL(QJgSTyf{q)G`L_I%(T)JlVkE?f3>pHi@6h*AO_ z`iY;IV)sH+?5Y}i6vRY640|pwN0s|hne3JzhT)|wXfL>H(;Aas8f1XW1_=6T+qHiT zLwbmx??CuY3vh<@j#H22DUyNmd)V;yew{#_ZmY3+N&Dt$4Th-~I1$p$zYj|l} zy~;LGK4Wulx7})h)77uaIVyfpcg1}u+@4EM?!p@kX?`17pffjj)3H=nLKD*LZa;0! zo73;qiB{9!sUGmjlLH~zj9BIc^rhTSZM1jqfZh&O(qXrxN3ZZL`7R=w15ey5!*T~v z{}tN*-~Jt?0zgHCE~u`4{!8?$yYTz@l>{TFkNe`Bx9 z|6A_O$>jJ){~Q?mCIQ4f(Sxu4*^m`5b|yXaEj;QE!`{EAIN+iJpcC+TNJ{0O4V}FM zCWWNUcqIS4AAK1+IJJNP(IdUfB=V0|?q8AnpLqzzz5w?|9R;D^U(hf z3Gn{~BmcjjbYLv=blzD14(mT3|{BnMhhOL6^53<<5 zo-d&HBz`$MI%pYP_s{Z)5*sl5N%DD$_#b`wmnoOx2*`@=*?+V_1MUKYt0qK0xBuLS z>~aO}4_+!rqNmJI|7i2b(~tm5NBGX|Ki>=eJD+#c638f>r`v+=|Lnn$0ONM_1dkj3 zaL(gjyQbn2{P-Myllo^5j_7jf{z^3dvzYd;5Z9>>>{%yS@BV)_GJKy9SUM&De*XV0 z)cfy#5U_MPuq>V~v-ancVwYdwW&p5scjYbU|ISnXPwON9Kga$5kK<0N@@99w>6FC- zT-}4wnEZbN+TZ+>cn480%iHeuMW%OzfWv1QmE75V(ZYJjmO(UX&a4E`Pkqx$m@HwV zQ)Dv%GcH!Zos~@};a1wC<*H@$^~V8QKAWj>^8ggSqRh+=xUV-}>trTXVAc&~F+_d3 zB~a~PIaKWcXsYpO=e@C;sjjNFo7MyDT#h|w11BAkgQ=oE9VBwvxf-cOBa5L(?@1Ax z@ku-PS<^zD;__r)G~?E2gS%~n>$2a`(ZumWJC$J^4`{pD8&%d7$>3DP;3nXe*3g+#AlxvK;<5Ap`7u_ ztACOBxJPQ$Wdjh_v`b~wMHiVgdCAty16rWK%#EDP$s616&*ONeJ}U$jz3mJyGxwD7 zJV>ius0vON1FY)*t{3; z4lt^Q@woLGo{5DdNYOOC1aL`%N|*-1zkFCY&2UtfxcX-2#CQ#> zi$q*97I;f)+s@J)pKvL19^Mi5D2;Sc*x3-WnB>xxk*eg0<25WXY1&RIgq4R-Fotw0 zo#mOa$U0|H=~$L8gJ78lCj#*PEM91ee_M;D9JVNb$rBJVht?V)y}F~AtKW~@^=tnS zxHEBnj?5Xxd{l)A%-w%vKPzs78b8~0hy zTneT~EvQD^v0nO%v_>?lLDYR-8UbQ=8-P?XYOw?2n)pfOCpdJQkVk#=Ois3r0E;NDJWGY88+|kt9%{7ab z8NVqVtur^|Jdqe>Xy>k@(`EMSk8R-ZUnk5~Yg4%OTad-OU$B|*5UUt6Nv#VyUp8$& z-vX>mu%)O}Kqx-L+A{}RwDDwd!hc%yU8V=*wOR)^XX2w;E~xWCh@>NNhnLQ%-H zFxXhK1j>6~<`9R0)|A?r%*3u7mEbUN{2rs2y)Bj8p8S(_BVOQKn>R&C4YGKz&|DY= z;KhtQ8fr!7^1h@Gdx~p7)f9wXS;qJOyq*Tet;3N$#y8>%tZR7`YwX8(OzQ$0yIG)> zID8Gf$`HSI&9K}xUp?({=gpTxS*c#V58TEyec#gYXg^yw86Y-us5`X)ANh>eAf~3a%_VDk^B~!ad1JH7Lb-7%LY1B{wi9oH5=3GIDS|S;Tu5 zfF^wz-E$LAG!XCaK$SQNJ88nygd&mNzL9RM}$3g3v?pPWq5t>FuwyVbG21b7W zjNb2NbzrUSMYYck4Z-TVZobzTn`f5%U1GiYq!|#dQXk_pJzc0lRQoj^eg$rv^PAQk zyKJ@uxQ#}N<;>+iSpslCp87X&!HMI7yMA3(xtxH!(GncyjOzm=nW)ISG5LTXDDJSA zcjp>1eYYg>1cF5yjV(erK1^?l*c>oz9veCgj3IU0(I46dGx%H;_L3dyI^o)Jdl|4_ zem;f^PX-0f(dnYrhwn%6%gH>aV6Xj7w~=7_E&+`*qkvsQBN1QY=zVolXJZc19+l)L z`=<5xfx2F2T~{3UxTii(QIZ z&Wqs}O;lgFi(Cb+);WmY-fk9D$Ux^_;N7t^TAEW0yG;pu>g4t83*8AO)(K!Y{a#EP zaASS@dJV3WO7>YTS0x-hC^KPyK30<1u=b}M9n%saHvOG!N5Er=#-gGkJuzZevVvfz zLhbz5Z`fBO8G(_!JU}S$eqO$Xutq%D8r3=ifdO%@3z~K3%-Vk{4T3q;`zmN^vKobZJ4D% z^punMCJ_u#8w_8rc&E1yZqGJ|xg{7Mc>2NnQqo-TLhe%y7>ObIVX{}w zj^iu4MmjPdj`9gRPI+};u1>qv(chN0Mm}$67 zV!7cuZHn=>uxNIr^^DHi`EfFzWT?DcvE;ECnoQLFjmy;_qG$g9=(W&!SH2tl2yn>4 zY?mo@NNBS$=#&AzwnEy2+f(iDo{^ZtnMC^|=%n0iALy07`4+)AtH4(2%i&=1kzOF} z(kPvzRW1JhVj1?U9G<$CpQAUPpO5-JHXKl%>08K9RHvB|8 zv$0LvyNQTm6_|W8qRu?I_$+!W*ol1`|9p20aBjLj+l^l9F?+g^*5&cLyqAub)WoA) z9@=C=Qagj3;J#CPjcL~j_Xx?UmhiloL1Z@h+htPer%7fdy>)+~EHbw(_!E6!n}4&O z5EG)zioj?gtyz0+Dm3ptZY5vqvC^(320O_>i`fbYx|Z!3!}S`O>Q1C-C%<6%Jldow zBJU#ij!@R~?B`###gxUbH0}#g1Yo#lM}#bXzBvG(ar{sy`4`X4pHY$5$#1X`{G(R@ zedu4zL#+D-?fSP-tzp|!<*I-wI*U_B=$x3{{6i@>lyKs6o^T5F6lu@b%20Hq)1>`( z!rweR&LNIR5IVnTZ70`=zbvs zp@yzePio_%8DxzzRX(ka3F2}FuVk9^|9Jo9?t$tAta_cv>}mK-)#<`#jF2$F<*q5% zdd>pPMF69_G2LwY*ckvhx*$!jtE_+38H$~nk3W(lwRWdk9 zEnt}G5&>9X@TW>9YAxbSosL}XYi-;&CP%`~ogCQiszg0hBBhUK;W<>f1H(-^z>j|8 z$`TZaC>3PiehKf8=@fw$v>42h1hlMO2}1`lszPGgX1oSk-2CnD`d?}6BwoS+^@*B3 z-8u*0A-e_9lYaHxt=M4AxAZCUeTt%3*+LTg!V(;Ub&ILS*ST(d?LRrSj8QM<$E0sI znSC65bz6P+!UT}a>^Q7-BQY|&dj0cX|LE0%*7R>ft2B0vS`NnA@yvg~I=Is@C>~<- z2nf*rLvmt)8ev{&WFLJ2yJorshgPn7%pJ*Y92S$jf6#cGL}Q_-eU{aB%r`(xatk=_`Fv?l6b_wdz00+ zCQwd>EzDm{rUdjut-~+vi>Y1ZC|)Uuaj})cXg70qSte&q>y~tjUbZ$(Th5cTAITCM z0@fw@i%)|4xk$9rUL##Ci;SvzS!fvvOZD4i_Jk<;*Z=s~ z?GlOZ%H20Z@_)`zR7s-$##Z#5Pf{GmT^-(LNAcYr~t0lY%Y zi#tNojCYR!(V?^F(Ls zv1*-2l86THA(j~5Ul-!W{c~cE-q@4T^F7w;XbQ1m)^S?yqG*+=W_#l!9))Oz&MfcM z!i&^zp(xmxx?>2a0WnM0{CSqW*kD}~&raWxT49o(Z?ygrwR-fYdC};)yd(alt?cvQ zwNLNRc*V6~VUrRu8x?|j-p6bI{$XkCgf!_Hb+W)^z1nKC1b$~=kS!lnk<-Y$(0`s6 z!zB9lLy=zHuP|l!ZQ*A?_`kKu&dy)F>?jkG2W{*4xHeYd?w~5`R`l*=#!(E3c%1{H ztes?A46w2%K0u0%`YYCOuJIWc#5#S&h{4I|`M+^J(B5Mf%-l7UjI38G(QnX!88>i= z(-T!MICIJ~rHZzwPtQp3wZd}GJ`voV2hx=4kwKNTNLk;**1Zu&>NG>{fb~15O%H|h zCfifzMEIM4^D2jQr%n69sPP|1zH0Hp`D2_`~CVI=fb5$*&XoXhh$*^&$JVC%HK9f@|>s8RzV;_xzpcBAERK5AFzfisLoxl!G zW8Rcer{X;#f2~|iZ@NTphVf*>`mu4ZbZL*h7M{o5eF*}|9HPZ1)71XTAAgVGt(-E% zV%FsD$rJv;YwyE#X;%@y_Xc74D+AvK8;h(IJfcdY)Oq463Ow%pd-4mg4D@%1LkX3A z=$7k2^SIj2WCYO&z>n6Fe-)VbJxc7r#BlZWgM9^~W`eF!lG9#;AUoLZ*_|wgD&Hrs zInE!=Q5{0frK+n55Y2er10lDL8%ixn)!t$VAj8UjV-K`ONXoYD&F9dv=3&js3wdHGyEq*zW`E z;umh`9?LB+nv#!qib*dSY|xTVCwP6Jih1u8S?)N}o?4v$8`79kJ`gWb=#;JbnJr*Z#u7Cvzvy}UeA~&n z@^-qk=P&(jLFoPir|%eXiCbUgACXkrPk`b}PDX9iY>%I%h<~{^q8oB(m3=l7ch4^A zx5*5ggf9Jq$ z;nkTsUKMqLNxwPp1=MUi^%I;xJ*N1g;T_toL9<%;0cWHyH>*zBoAz6+?x|^a4Pt+O zeWYQldRr_?!pj=n7?)b_p<2x%ZYxlzH)wl<*GpJx=;=#B#v?4&C86YM@!qSividiA zm_isuDAOglS$*+Uv#;$~P*ZoBB-IY@zw+sCafw6YKJz#|>G{GHhYWeh8Ss+&`<^t> z@WRzun3;-%+j;FuZz7=JwqZ6~(nKMM$L^YM9IU!Q_A`~To!aF`B0*(u7N49N%)vRH zhOt4tj$<2nmX2-#jA2ZCv0;NYl~(AA$No}X>}jw3$_`6wYN0n=rKf#_sm(?FLy19? zK{(xmyEfy+LW(*tJ;6yuZgNM#Q0(DcgF7*B$;s)d(kRJZpi3QHi8a2+o=e<$B@DE( zzpgf$_AC8@~uTnP`SObIEIiwIw7c);3Zk$}Pp(Tu9*TIne+z%LS^_ zg?E%A5&#}H4J?x;h6Tg-I%%P)6CA$hC-@L=+ar&X__0V@Is9QvkzU!8>1u1vYY>i5 z&e#&YW}h%*U?f!^y_ zxl6sy^)!5lQ!mp{_{{o5?Cb84G~go2=yEu^IVejLN3N1_fIRQoV_mIiVsaNy69;lN;1sQpgFI zPSqp2G^%fsiuM;f;1VK1-?%5I+0Gea_m0g(DUJfUjKXVJB{$o=H@SBnNMjTMB~6`r z+xflGS?vh{ti?CdA>`&&>aal?WQj>kwxRp;)j)@DKF{zQa_053b>2g+`6iWY!ASOi z?rPub!=VU)%uoup%AMcYpSI;3Q2t#LfX1{gbA5tE2^?SFqbN;IP+#*6Ko2!M%l^^5 zYMVrn*bYPaMu`4yWN8GmlzVS(lUL%zK^!lRvdKiBgUvRIZs9koMoa!m?J9@kg$HeF zyZS{}BxBCp6?6&GPiNg|ceZZ1?zP$lQ1{hxH>UR4ARS?)9jlHqh$`=zH+*6DRKGq; zn=m2UpQ75zxu(|fQTc!|^i0fay)HnlAZc;W#GuI{av~#PBZR>Q9WbF> z#1?@3+!CU0nB|sAWj^a3lh!`wqb+byD>MSgLw@nJ+#Wwuw9F9G%dWd;3e%r^p8cDjOxfX&jI<@_4&z(BxQI&8;SAN%L6&jSm%W zleeeX{xII9LWHDa9(RrMW^L6HgMOxIA--4)5nG2|yR+y3C%E!%G69_ArXv&8t(o%q zbW>xO?@_de(=zFYpFttyUmTR94m|{3oGo6Xa!b9dSY>j3+lZZ$^{vd%dNi8zCHz9#VS^GlN79syT?^Tv>lVmG{fLT|I`iIs{>Okdmx2-p80g*ADCVg~)X|~V*7OEAv z8lXa#iUg3Xn~Wbrmtjp+uVky_5 ziCQ;0BWVkqCO;Cj#rzeFZ2j(ie&S@PO}g&{)D4ZtfgX(xyYnVnsdN>-I1rn)3OS=- zTVu-6;Ni+HoSwuN*J%fdP7`v6c`7ufYR@GscYWlb|Gk6l=!rQ#wQj^!%lBime#C|& zLbs;bNorE+*_EkOFIkzZg?WwUg*wi-Gss5X;JzLhu>a;XsC-|p$~@10J9LaqC> z9)9tBd$LSj;wehY_k1L*;+o%nu(s~F-Zfvrb_q5~aP!+>no7kFDBiE!n*6lfnxfJ+ zV|nZ!Z*YF4?X?P)T+pPX7Hi*W)L-su4!Is9nKE<3xP402aJi5IJx5-(th z=V^#$5rberGr!#rG?g9GU#61}un~Nczd445uDVg_Ll&X^MTpVmPPTyDIAG&yZpB?+ zZTwaES1myQl|yDk1CkH%SFRZ;2Nph_%}&2s{-G!^c?_4`ZGSfQbQf*C!8)d8p$nIwJmfDs6L z6KW>Kca#Jo9m|dlyV=OP@8)TD&AK8Ue|G?H4A)gKNxS|GTfD?~g4)_X>cML|@*Q*F z8PopK@MF6f2KGlp$JCKQ1cXtmgL7W@FDq{H*JZv1u@AmLK8Dlr{EQyWCF>Xj7bwb_ zd_z8=s|wxTqHzUw1BZYo>lpFj96OsI(pLC+W}KINc-U zwtZ4ch}_9}7_;fqF16Puv?gOB<|+STo=tumpAmk-Sa5qrE0zcJJS4vYmndZULgd2| zb!n(5gUEX};|6!JiCusskNC`t!Xby!#T2`u==vJ{x+54ddpBD>NS{1IzBUYA!@8mg zI}Rs1hHL!%TyU-3(~cugN*GbQ74;um=V;`zkY?UDJf^R=Klpu<)c%f%K zEJCYT%teeFv)q?eHD5GVw#C~L*k;a~zpcg@l%i!%jjnKbHtHCH@ROf_*_`Q$2?9USOF#Ld zvjk@tLcW|lJW4>#R}nJyq(mKB4yMKINV#pSmHHiBl;8Yn%z7GGy@r4Yep+eMrTto9XS_6Wz0O5a}Z>mE%x9k2)wes>l%j>QRj@1iIQAdmy2qr&~Y7 zt6zR^P4!{(;PO1T(!skM+zZhUP>{rIrAC6gMckBA85V$bY=cM;5I(vN@Juq$e2_kMcQ0fN@4-HOb^W+PZ!0!S$lDZvrP8x zqk93^+Ze8S7d@AQcdC)%h#Bm_#3W!t6W1OZO&TQ#QGDaP8qa6*bJrYo{LG`Tz-lUnWjIwDIu+e)?W z?yhsQH?8yD?uL8y)-ft(lwLq>_S*<&kgs374%{0W>pzN?lZ4Stjb?q-8&SNN`yrws zECTOOm5qF{{`j-L$xE6=%H4!AMVbQ%(xOuoG0&&v zjAc4w&S`L48=Uhl2f?{(?kdbqhJ1>1mqT^CXZB8nPaHK&Ju9j0mE6Z|@YD6Bi~A|S zW@|Z1y27nK^yDj^Kyx9Kn%y#zctPfigQH|GKC@@xANDyLBo`V=hBiOhzrk2L>Uo2hNWm8VMzcM>aBr^&d$Jt~|>`z?Y%7ralc0 zLRk{EODSY0wM*6EE?v^jIqG(>CK1e@c*q#U-LGU-BV8v`>EpIjFH)Pk95?)`E~~zy zQuT(~Eo1NZ^Ssf$v!^U`nHc;RA!L7*T1kEIiw9r(cxgC1QsCYh5Gs?A-KFJ19WKv6 z-2o_z2__1Ann)!}Fp>VUbrO}&7>UsW?mCwrEZaYK*4g>+t&p1pPJ!FxAb-p=E z?@6*X^Vcx|?(V7+x1X);!ZG1waFz=WF!of-XVbc`LS0CPXyUXFnWleUbTJ;B`V&+; z-X?!I;@L)YJu64j`eHZryM6WeX^X|m-r{4_jDz|7`DsoGXI97?IKaQ~*!Ee7V7xwf zOQsCL(}jC#a}g64@_Zu_M260NoV)6?t)lZQntAl*05Hq5u`-^w<{Dw5; z;vQ}`c_(F1T3_N`8c$br|u-WXwNG2UlVM{qa$E;@X87cL@oN$#^mL(d(+Pje*K{jl=W>)hx0 z7nlosqzvb#*!z6B+$+J;LD!DQFVNPt!W26`Mpv&}T(wn>!t03LU%VJ?;BP8Du9<(x z4%i`no>Ptu!j#dBA3eD&S1Wm9oZtK`x}Gi%P^Faqg2tgfLFc-WK1 z_X!Vs1uoDVfF@Kxi=vk`|U4WiR}Wp7ul?}%757MWDc^OQ%|y6?^7 z@{~9#7KdPO0{=!kd?BCxuNs#RTCP#&lXYa~6-i%3^0k%0t1nkX*j`iJEXl=8qnvbJ z;LNhm>(3V5^#p&H+MFg_*&cN`-||2F_Na5v;WjR~XsSy0EYDEJsQ-HV!nl}U@cggx%+!F^1V-w5`-AOxE!suq)dnj$&>&X8 zf48bQoR*!{Td@7Cr@ok^&BF^KU9I@U&af+Dl{CN^=ZiUa+2wa@^3H7kc3rtP`rw`Q z8?golu>>wPP{O0P$I=qMs6aJ|$NHe#^ZThaoT1@pMMgdBp1#ZK`c&w{l;x}!-{cLn zi{HS4DwsA3t;1)mpZa^|tYJm-cuVr1)h~rp^(z-yY;JEwzJlU=C`P9Z(D&Il791;O!6FflJlaqR~wo%~i{4BNaM1#fIT=_$JY(8=OS110kIi_i%F zt%gVMpi^fLxco2J7FnVuVj_OlYxSqUms31D^qrC4+i_qjs*uN@uMwA_tRNX799C~3 z-$NXk0uB5KTJ>kfiz6Rlfo6v@4rnl)FXcNT!u5qL4OB~1!iM(ivnW%)byMJ?Z!Nit zM=GZ5@cDidA29@#_OXBHB70IgnycAY=YqT!kS3K{zle3_fuylD*}u6h^Y#M`B7F?+ zxab2)RW5!6_s_r2ba!Q$TO*Hg<2px{WYW=a6bH|@{}6OZ`k}? zC;w-oORngz#-32OhUavB0O50WGH?elxntzEBC(et?x@ok@w_>=5iXdre;@4>MGJ^a zcB*AVW?=l9rH;q#(<2r=zgEv&vSNUwn^rMx!=N!%g~w#b8ZYm%*|m_oT~0k|S03Uw z35HX10zMfPv!s(oE3(x1-E*t1hjTA0jg-k5?{$XXbal9KFIwHVc-pm038}y!of8Pu zGW8@wNP%-j*)4?DUh6NdesF1rk}JMztK_4cm=nfnaD&x!6FS^)OX64(}6HcHBeHBas)YyRev)1VSaElJaisjae>`!rtze@EiU|1@4g<~ zG$xiGY-}-*@(>`v4<@hfmR*gMi*Hwn@+wkcPYW5&v$F9tB;E)63AX&WA}4x<5@#Lj zBVk)P;k58g?r`f>`I`~egSkmlLVB;iQ$@9mD*lz}qV~vMBGs@Wh?=?0<+t3;?+Hhn zzL%-;?vdpPJMqZ^TAGrLk?Rg+M^g|j8B*ICJ zsrjxWZA$K)TcAEQWEoYJ0kx#SOtm8M%BgWu`m2eZ=w=oU$Bg5fj;p92chSZwQK~id zRWE2GjAg+riRJHtdl+*MwQ^KTw~!D zF#0O`qd7*FN1i{|cJ|8oOf*Xg<)%U>?me05ewI?-yC8f2+cVACElu-=Mq1af0y#s~ zO33-?V2G!li96Wk3Q}1y*-i>O&0P){+Y19&7jt|LsN)c zw~yuV=3UjXrg@q3xgd{=H=|^zPK&r8`}OO6klLY0&s`#uW0OvX8y;@g@4fi;(iz#y z>8PL?_t+d5&lnt29yQj~e{3m=OU-n#OkT)hA0PZ8({PjLr99i(F{dQepa;9v6&h$4 z`JF7FF8wN@mGkD)U@9^!v!N3}1FLr>-NzNblQ#ncB%(|2)m%}ww6nrb_Z`hbw`v4N zJRQ5eW8L)VQS3zn33{w=PvzqIMRIun<9Odr2lhl-7^~ib`Nu)UUvdwpm%N@FMo};W zM@}zdKoB6-IbSaC^zv(gPmv4>(vNkI=r@uevW_&VX6=+lb&K{Rm1Z1{yGI8W$= zu<)nt@dMpsKB;J;*}A4q2Q(ozlj7`c`)9gnNmiSb1tu}u>AR81RoCxIMgv{7zLZC| zh}LDWBps zB4s*O)RSu<(bt+A25dEZ9ZB*suB`haeiytFdB*5R z%v$&t6aOJ9z?*6CI4rf{_$N}a+09`|kZKK;iT+R@#4>nt-PIBB zXJczM&3uP7;kR>F1NDW}29|<}?mAhvaQdu;j}6{U-;f(y&%dA&+7{i2;rx;@z}GhO)jMt&s1Fk?51E@C%Wf`5)r zrbS7C9wPLOTLJ9uO5yOX_f1FBX~nxrF^U$~qOF=3qGSr6&O9)ZaYO>>=Vym+6@DY* zh(J&MPINYb2SKiqH?lnhVD!RE{Zr`KywtB%KauRPBqD?T2>>);K7vrvZOkl$?`w~d z8tKXh=7VheW}VKW*K1t$jNJWV+Rsv8ORuz}1GujUk9%iorx_6k5dIB%fyI zRZN?r*l6Dc=n}e0W5Pq|F2@HJl3TVVTq4@}d>8X|vcq?~iZ}^wc<_^CJ)>Dak0wq% z+RMvFZR+?gwkdfxQ4dfuO1^x(DKJSw0t=01ut|yDZzUbcSJQl8Qu(z!+b8??q%CB5C4m$&pgqlM|^LMz? ziq~@Q{el^xa{C0nTHj$B3bDgsdsm+o4E#WPaM4o7GO_dHRwPLN+|jHRQlrx2xwVylo@Q@tkfCvczx#aU*;hI6LKNhj=DYF{ z&*Qhh#T$50$xEH2GoHKzCXQ#-g#4#*IO)wHYa3?vQj-d|pg0i~@ea)9h*J8km~q=1 zdS%X4mAmKpeMfE^3qRjn33QqJNXs2t5<)&aU`M>I<4E9u*wAt*ekrbh;Sp$KT)>t{p-MQ4@ry5d=Y^cM^#jBoe(v5JB|bg9y<}^cppK zHyDgA+K67qC}A*+ZZM4RvG;z@`Oe;_t5G=-PdKKQmB^~ z5ZYx-71HR1zH?8%&K8tf4zHgcsD6VpDRbrVQwRBiu=HXl5aJpoEQn%a_D_~;< z-a1(E>&vo)Or20|cw^Oivt8%)adDBnq}tn^yGwl)QK`2VD?rjFFB$o?p8LN?Z+oE! z>}g;IgBZ08UWKcIIF;kx){c{Pv4Xx?rcUW1VagqkJAfXCZxr^Gji>foS1AY9&Bv!%=C<2(8WSi=Q(Q;mvdqu+a_bat^#{GG8hwm8NUq*Pron9}?Xxn@&P4!HDV2SI?Y@J6|IT!1>M73V zBz#q0muiHXoY1aRw2bMs+Zx{@%3)!_L_arFJgq!rdDz%QhzY&#**9cd|Ma0c_X4}L zXD3bLqk;ej4~3So)Go8Rweo{P%IhWXKEr=8`Bv@EQ%*Y#VG6t}wt zS6%+ZwC#iC{4Y5l4P_%APlb05G<=wiZ6ktVPZ&D&*r)dWcyDuYJvd>=_&E5i*zch| z*-qOvLhd=ky;4Wxo^=5~m%{EDj!33Fb$um^c~<*8K4TP5vb0OnNk^!PDOvJMcsNL9 zm+{1Q;33Vh7*~HhC)+vTgD~Ri;=IhI!Rm^%!7xPs!R~$=Au#&OA!FrXX#Z6dlD6f`#pCQtUL;u!{ zyL%(9q7-W{AFOz;8-Z3%yv&dofTOm+6I*N5o*~1ESfi+(A>)U4bydztXtoziQxw}p zE1Qq5_lE)teX2t&2tod`Tw}3`3OLNW0fcqQSh_dkTomu|K4Srd6x0aRX6cY~Ow3O} zbek{R&Mkv`={iTWN;KoRLc&Yew8|60*63HhEOp0#_V7aOwQb=%BDV0##$(Ac081!a zb3RYd#`Bw5J99Fub$??9rMBiW z`M$6iO|AlnFw1E`ixA!VR9U*=5Wv`I-B-O5@S;C{PH&y6$25ysBWP_bbXE0^;=t`k zB_;j4z@}!fr?%MOQ0@y3RXI0sDR+%(fDm9~a9!-1n^6)?9A`V zhcFM)`Moh8bci&S`>66vHeU`LWgvXr-DqG=jgX25&6#gCBSR@ls ziB7$Ha_jKCey&=)#(bF;r=rlOT@+H<``33B2{fAQ>8d-wf5F;pb@A&5#`4lHY4(#x zTM}Ju^ri-xu@|sPTHI7BU^NVtxv}qYL{7c;LYw>n}PV%1_7g_}e0GFAqwJmF8F34oOR-zrbI;OcKS)g@Q*&(H4YWQ5h0; z=>2$|ZSo&Eh9_qpQo`j~q^nSSXuocxGpEp%u+@jtRU0`9UzVzb4ct_IMOoW5wC9{# z(~fu=SbTs*FE~%a{g!HkyrWIOZ6+H6Od`5ZPHj0r0{Uv!@j94BkF|A4+V)nM=#R^5#MsS%**)fJP`i<%Y0 zO#O0yK+}n|&p2{5%tjJuSwnsSwF{Q}l~Ig}}QToJJ4qIUt(04BVRci|^DKh+Nfi5fv{d+f$0WcjW2w zB9q;POUz?qk%VLjVkMD(T=U+X3x@aT zp;m6+wj}(qXeOO+xG?%P&}jrHY+Wmu78qvF@KspjOe0;w4)o+>E8ERO+NZ1CZ`FVW zi8^k4nxUazIzB!B<&)rYRaw?sON_B@&ve%|cVlqX**zhSRtT;&0-DbX56MKw{O zvbUT_uQu(c@J^#{78efXI7zk8iQ~7HKMiN@U50E_ij@f%td+wXZ8m%&GUv9u)DRyz z8PuOCh|SHl-LZD+2GT}CBkWSU&?IvC!$nc(LC4(+innISchbR*w&E3Z(tR$V?j6@A zZ`PmqObE4wdW&=)%APQ5VJT6_Bxs`}&|@y?QR$*#r2UFpw;vfUtS=qSJhz^{En~rC zx4xqj405WJypKaP6v%IK{c;T5?_Hr;#QL1Nmzc5uQ!?!F!mmOBUg+F#&a=4N!#T5I zGJe?Xbop#mj4iK+adH?9Z!!m`PUOnEORzi2K+J6RC^#xvyetYf(_)&qg-7xT*>X;> z$+EwJGO72IV10mCrk|ABQi{%G@zF!Sb7c&`;GMP|0$`sjZCZNs^&ytdh+Ob;^E3Y7 ztHLv~1~x&MxrIo>Z+#M4weV#z8Hx7vJ7j01)Zy104g)9cs`F26Xxl0U=BIDpJZ%sMu}#e*GB_*w1UEzo0!S4QMCvui!}*Keh5^ zfo?yUXFEi;3v?bgGr%(6TE*Bi-YjrK8>CM88l;IH2>N@?(%uoViQkL2mI(9CujCAT z{2R=ionylKFlef1h7K{uKG35njohXN$2i-N9s`^@;>+B6@-vz`|g(YKq{O zu${R{gXIkZU1JJ$H>CL0w@!rRELv`vsnv$|BialSbAhASm$M5U#Ud0g_T8U`5WrHwAAhFuF6M$JkfuhA4&$}rKurMyes4>`eg|6ERjNj(_a2Rv;Z#4 zKmjj4-cMi5l;ssNoAvmW^vF^U*%Id$`=L-n@rexl(J9lkUzo9Wxtrx{W@~l{CK358 z90nqpLT&x*r=DG9=mf>1qGBiam`z=_x|C@{uuH$Qe+6T^6Md=Tsg-1J97{TI!mfX( zkAvPw7;lnEpSB7%jT9_&*tz3Cj||*D0BS8XqM2TyuM%-dR?_wz?U4N>FxWRqwU2Ga zbGQAGo$lKSTcYx9LcqAnesXzzAdrq|GPfD-xPT8{Nuq+Fjb4WxaC4PZO6x3?|RC~QgxQJP3G zSsS2D=jzL>KZP*b4u81`CfZLwSIdxG83^Uu-NT2SY;n&qbs@CG7>0PP9!Gh6-3^IK?aO_p zvXTmCbPTROD!7^7w^lbOMPwBiN!b`%oml>(Ca-uCO5O-rbuR*qp3_J0Z;ZxN=c!qO z$ZT=nqN?ng*hdpDEE`K^D)m#eIr;ohbJk=pD_fC|Gahb9$~;A`%>THV8zCiopOU;< ze;4Q{>Amg~>NJ!4rSvYwT@7N|QmRe`y89(3Q}Z1I+1kxt&0W4e?O=hW;%{{AK7lxW z{fK_#htucza$c}=9%0{2Sys6<{>U^j?)9;QAt};y>E)%>?Git|B(FvSsZxBIn`K2` zOac0Y;rvw~TBLn2&J_$o@QsYG>M8xvY5YX>4LSqzD{P1njG>LRE&I#tr>^V!lIEc_+;8a-JZ)&8q?1Ax->-FMDS`i{nf!Y& zyOGOmo}aQz4Yn8@#p*fj<2|sYgo8pn-#q$tB?Jm~{;)N_S&4wbWq@97qH+4U%y9Bo zaEN-x>66;qu6O@OgxDXEbj6^Lr%kkTWgfQ-VLFpHSx5XFGl(&SC|ChJ&Gn5z64-1# zpPt7T1F?G}+!T3_Pe5kp^XFBE3nqbACKoaIeo1>fZ>m%nCcMowQ$XC6to-(HREC!d8e z%h4N%r_Elspc&76$F@Bv+)(pzni6Qm%CzaZ)ZX>5MDr+L;|w7eB7Y}O`<(aC68^&| zD%X9^6zoB>+h$XIQk7XlykJi@%fXF@%>|B1%8i4!8A5M5z<+Gou$`<42ayv(O7alOcU`wRhO^=eeg}6~yxC2|k3b z{bMpZ|IBpe8`)^k`u84Z!fXD-G<-^vC?7-DPx)g##w%V@Q7U$#iV~*PzOCM4-&fd4 zjT*hn}w#l{b6^S{i16B>RCjPK|)Z27#P0YVEzmexewZ zmv^03=3*muD1j$TWxf{&TnjXhE^1sfncsctu3MyCY{o~7-i8`_nXOj^`z6T+i6nZ1 zN{YL7+iGx>c!f*oETx{QPrgDlI1JToHI2Z#VS}O^SSMHO}{cH3X}w* z)~AP;QzJXTd#o|eTf(<$bB5tK_~uS%p5K+qrE&=AkJ4FC>4T`>X^~UEjzgG1K^`t= zfZ|(l1af0y8>agj!E1Ar*XEv$8$seU#jxGq{ne9<&YJ z8}d8TU3j|{KI^uvoca>3oak^fuaDzA*9wj0ed}qVj~1Lr>oXxg0<3MTfc_r+o4X5b zS9M!#3PRymYp;{n(H!jl+Q6I|dSix{SN0C?9IZ#&%2+j>6)fxSYB6dXKC8WpU&1x* zbWk6jB&7W?L149HnmN7On~&sh#=rWN_tpoRgmr?wR)klNr?H9Y#{$AZ7qnnxb;+B$ z0m{K$!}AhNYx4x(sI40zP?oVHD)p;QXfJzc$|1jB$2$yTr&RS@3~w?BwfG8V_nmh! zFVUDk+sWESy+gdDq3OMXcn0LXb{Z0>KK#8siclEGJ1ONHDS1+Zw&1m$KVFb9d&88- zFm(H}@($RV8_z%pqiE61XXR=nO0l9Zqn+A=8dZaBW2LOAdketBILVc#CJl~Vj+Ahl#G})QR6}Uk>dLs~JneYVYkLYLn&F|2 zZ>o?g^G^U@_B?-ASnX40FwLeNxv5e1fwOnPJvYS*Qy^qqEU(#iZh?utze7v`7eOcaO?uoUpVkb!|o) zvs0vUOC-`;qrA?UQDQg^w|yD4>=(@hb|VjIm7|+_`l5PL76M}?{k(Q=9z5GHeqC*( z-~h8_LlBj;eS^gt3gm+hG!S28B;bkCDjKQ3RsG$=u=|%%X-R*jhQEl}6dxp+Kg6>1 zyTSeMvtk=rO1I~Wz6jFsw-YuDt-tC*9x-t}b25t``*O3d(uro^8bL^A>HSQZ^051p zTxjP`&quyMwc*#3*3phO3^Eg;$VioE_X_QiHX2%gigyIS>U{toQ3qf@C@pox`)sTl z=|0=h)nLOJbiUOOWD(j8Ln*bO1C zQ;Gwr-N(ryTauK(ax-8*z%{=w>F3WYocM3u77NL(+r-p&+P-Ugq2@fq+}A?rr5#_R z=1yk0>Y`Ndj3++9-=Fs2j(K27$toSZa|_4cGGb=C-{Rbh&P0qcE~rY#T12t9&hS@v zMCv7DZa-id#IfknNl5X_#-*GQv(-U11t@vNtUb!sFA&VuSI4yyB#b(l;&4^@N=E*k ziO~IQ`W$niv>!F2vY$^9;Vz;4^#^XH)La&j&%8xPmNL1|m+SXHy;RafJ(P$FPUOfdGimM#Q>DI^5PWrdQTL3w&Xnw9DSNR~vJoB~5oDU}Zq&_#MFXpZ9Wga3~ zfaV5aRX&6I`;hMo-eV&0VvGH-hDY&NB-9mAzZlcvgNO`glwi@3GnKpg^f2RY zwe(wVg`<-0b^1ecspXcs_EeP?&FyWSZI~W!`@YYqLOehmh*oc`OO$MI>N7uo2sT?M zVx2n@q2ACJv|&yMAJ(PT+0JS(RC zJ5t_<+4y;?gIq>kGcWHxi!Ij^PGH}73o*tY58)voO-{sk^RK^Zk+D1V_o_@U&&uAR zs754j*O`0_ZaubJ+2zerD}?*VZAN$z+YTK=os%1*y+T9rjS{=5(Ae&rg63kB%a zw9SBBw^=g})(TKd6d#65t7TRRhMUVjE_=n^`(hqU6onM|uvlDpus;3l?HT}WK8>Jz z_IZ7-Xwp}mIYdGB%(&?_n}?>GVNW>isQI%0I8DIngJc1{=^h}FVOql!_q*EP^=p~D z90DeEH;}Onya4#E_;}+$cXL(MH}MwTs-(~smHzF0n%Q&oCd>lm0%m6RkGN?cgQgzL96sG>K*pRES)03 zjjY}PwRM$BK>kKGPW$;Uo>Y<8uN&_c!U2^G#RAgOP3tfjc0fu;mHm7t?w)$WYwUzW zkQin)%uA->z7&J@JD1l#)qeiYg?%HeU1V zLQLJnM+-D~WTVjV7=!hZW4iV+DV_lzSYVRrBOhh=IAwGER0+wh(JE4PC$EftBaPvX z4jw_1ran^(XcEOMHl@8EtidZfQIfQzf5=Wm`Cqx%CN};-Lg%-+G(n$M{aAD+$4i0J zXbkcOeY_htQenwGjhXa?$XQI1)y6?C2ul5LHs>lT|uttIGuV|t-qw{c&aAnHK9!a)s)Ge zJ%Wi1XlYy-d3QQ6JyK@s2EGAB$q}Hd_n~+h&v#xT)m$m&mKD6Een9oVyeb${>|ig& zo9r_l^DB>hBM=0-4N$ecrH{G2u6_c!9-HEHdjO!4ovM!7Z=kT`-IqE!-%%quc(@0T zxVlNy0D=^~@{mHR*XdJJ!t5cW+hTbW<;&Rby|kY;`GTM9Fworv9rO+q9GuyIip+_G ze>ptcecN?o@C75L)!}TcOT}^hl;lf6_E~O`Nl4ziH}kmkT+Umj+Mr@h)n_|h&#^L8 z6-*-qIv2&=_MDWRG=#^2ccQB^hB-dSy*M9|xkzzg_Z2GBwm4mg#7Nd%wWw&9iqRvv zSm0txalZsxAAi_T%ttc|GvJA2B)H(YwG25*h-b+3YezZ#U33FTI08~SPj{{m3gFKJl=D4{3$}f!*D9aF!2t}ZzZH`T(JPIaeaTRR+aL`t6 zi(*e}+r9vYSkOK-SCTq4G}V%Rh6T5!j7d42Ld?;uP@^zS(B{nF4EnWAcSRr_fa(Q! z@ph1tc8(4ae#?iG;WvE%)d!zQf7Po=v-R7TQYfqLx3Xx)3`Zj9nmEkoxoC?R7Nt3M zlDL<H;BoK&ppFus_V&5?y7k+-v57YYcI4ik2r&>tJYpeC~!1-HKw31^nW51=#yL zneuyoamw^onsNlBQKNk@!aR7XINtbAFYxzP;_c$6&w~8$JcwjFLnsC@ISw6YHie^w zi-BDiwTYPj8tMHvX)L@bS)+5?+c-V@K!*Ii-!`iuwt(WkIQr?z2~SuqeC6?6*koMD z+RND0$o1a{3Dwigae+69I-BHpQ%=Hr=1C$i2EViXx z_FDJ~T_hipSf1rX9g62e4+>lo54UDPY=zQ!96`RaQNx#|K^8Fq6Jb63W&haMlCtj&`P*{qxg=oLYrx|R>!t3vn=%Q zp6w2Nv|s_~cp7*>x~}od!{{GpXW|e+TCPu2B!aG6(H} z8k%J?;W~3Ke2z?{n(tiKl8U;w(?ppC6gZlz3}~(&bL{-u#62I1;Q6zHP=CmARZ_eM zmXlZ_n=~jAdL2+4QiJ{MSDp`FPw#pj_xH)y>U1flhQLxf{qD{G(=)cQUD*{gvX@cN z;YE~#!!|{fS3k~e#9P-YzM=zHVo)HH0*P}mL;NHscT5TVBcz5R#Yi@rQCK-7I z=W?(wA1`Icj1agBvwBFhMFhQAm4hgV`U1e8 z8hp&Ay!^_VjIi@PBENL*~n0WmP|9W2J5>iIvyQmfNb z=FigjD;=jUnK@Jrod5LR|L2c8n*g8YY#Q=;Xn?3lv+ynx=)*cc9E0BOpLRl+O>9o{ zFC#Kn)`oNTULq&>T@T)W0~Zo=#Wj>=JB1ux0FpFW@ehD?_b2AZKV9r^e|N`GG@{=` ziCG8lF3{)P8_OOarB@Rjl_DbI6PYQ>k_!58UCU4Se@4jv>6`!KM?yb<%1h?gEh+c# z`q+0P!*2Do`yylGUO@A^2B4HwE~TC!&PG5=AH7e*tGKCCdC$D>Ypjq8b@^7aW}1wj z(WROvEr-KWXDqh@>6g4^k1(M9_eZ*f+r7sDc(J@TdQpSebF}evjYoB%O0f;4mA?9A zHRbx#BVaHS?}*qb14>?lbW2@fh>^2=z+$d>_Urlnkj$$4#(||S$hTQ3=aSj~{B&=n zzsk6Qc@NRg!Pk5ctJI;y=oB30-RQi2&!y;f{v2F}j8QTg=5_SrdO89?ZrRL3Z^as6 zz(d2cqHTsb1z0o+78e*ynv6Uf>E}~fdiv#_HZ+YA*tXRNb>)8}CHkl7`fne}BZ1gT zrch5WZiZ*1$SCPp5kTh{P?I_b6v|XC z{I8l%@2&720!5SYOfv+X{AL}g78UaO(vy=I|27hc+V-Q!2SB?*; z(FYpvy?4j89rb6$m*cgK9wF=sQQLLz4TdZtENZ)&N*fYi`1?YneV&ikZO!H>Y|pbd zf7pGiyVt|S?>TV00nh>*AE#HL=HuO(sfhuWZsS)80iaYLTdojJ9x{>ck9=M0Xx?%E z-9QqQE74hTVr|p?0_#I7=2>$lX`R5l749w$ryfjEYN#^*_i+3h8Nt8qW}gK76FWS- zN!4*#?p{f>{*qfx+~Y-9mqwj`VY zBy=r5&5|dY-XKh&AkS4l*;QCS?d8A)p>_ai00m_@&eiZWdh3s`yyhi6?!D~{c?TQK zS5Gv4H&mw{aoCne&sU?+`Q>I0Z{RC%0YJCTqnI?-Ch(}QE_i7o4fWTldmCi3!aEP@ zo-W|sd}H|qYT6UYb1C9vyR*QLoHWj}JwL?_c2TDo5z;gLc+tZZk}2(!Pm$gMlf1ZN zhbP`x{LwedR`R2YID zY2+%7D5npGQSn&eExT>c$CGS8rQXenvVKL4)fR`-3Mt33sc!6stBmu|<64NOc1Yb> zdKXgYwmr=XS50D06}C$dLQeD^08+P5iEVd!)|w!4t06!Db+kFO@w15a+*C-z%a4aq z7?ecvNk1R|Myg%7lULs@&U?*7P!21x>?iSDml`95fHa<<^+@Me4~#`pV|=17lE-qk zDBg!uv>UDAc5o?J1hhGr567};)Hz!wvOqIs_}&o8i;ASdE(LWZDuzzBM3z5A*51&Z zbnUl+(cKfYt~o8!1y2l+;9^K^W|dDISMYdu2nUJ_pbWOhKCkO#0L5V^J{=#Af9MbX z$Jq9p(U(x>G6c%h(Tm~XF`qGhd+QLAPQeyR<+Lv4uxB=1t1GRGitpZw=@+#M2V+ z?ubUQi7S}2f4ZI*Y%6^q5K4Sqt~cJIy?OJmY|8&0n$h;XScL>mNHPG{AT@5p z9Ip#lm;v&PfZzgws^P{MK=+)QCWp=uhICj@d|xc78olh3BtuRDjjF!fn4@hG)Ow*t zehm^h?ZUUmORdcfBP&h2VkemV9%L3>x$G^k@?Dl^wQ{pOAWQ~~1XAY7XH1)Qq697h z&rIj_r3XYVr|EjN;aT%OR|t`SqBqMOXB)tA#<5V^4VCs`wgvXJxnRjnWo~;!QDA(W z$9L~5r{hT<)o2UQlIj|#pt~G#MOq<>Ar3%J6#*@wEr%GzSkcOZ!^*CO!Gg-U<9$>#VjFn4i}1hH>DS>C_WpBOkMqx=|96~W%w!{bplY6z%^@4z|OwP_#pGvA=tR z@1g&tfOHO-j5ABcyDdFhQTGZ?aPD0@H-M23Gz^%am3}D*k|OL04s=a?T_oj(BN9 za`ZEB<0jpz+y#7-!wH*NvmR)AmFFN`gX2o*rQ}bF-$9JGh-=BQlRMfiBYHwV+xx?n z+X7Z!9iq@8@_?M%(bT;f>&^xw1fAir1<3N%6(diM7$nSUZ8eS&nSN&D8{>|@)?Oxy z`Zhg{q@P}H&{h0@$?59wB)R1~j_Bzl@R}&y*Ygr{bw~5_vDkAA|GT+*W}BPr&)+99 zNIQ0_c`3_Ae|UCRL8kbIg34>u11F$`N{rR-;vB>ou-_o?K;dZC4J8_8I6}i`=-7+t zyQh?AoD$I0R)v0 zPP|c7@@Wu2VQQpYo$2JHATxI?^}=GCcOmyXTr^UKh-d@`i=e#wb}y0iaaTKO=Tp5~ z5AZ11p5o!)>py&9SxNXZ_ujo+xkiawXEH-%@9bRd(6-z4tTK>EYj9T zx`lP4&v_{sb9m}Q7U+2u<)3zsWL^g>{5aCtSlKkP^u~)c)~ovGCgSp+$H>s|-k|Gk z{|R9K4*kC4MU&K_kz?$t8Jk+VS?8fvpqx?e4ipVOor6y>TwhCM9%=gqtn8 z(G}L1wSr9stW!=cOjWn%s})=~rzor&G`DSi5$PKBFIYgfwR}&RikJ7@ff;hO6 zF2sF2)4Z;`k9}L})s;cw zGs#aT3D*bq25c$`;X(s!dO>qD*Gl^`=m**d}4|+u$(DLp35~h z?TH}Xh?{bntFuYHQ$Ee$_5)CK3Ftm}7r7=Rz-qnd-Q?T!=dEk+l>#1txrb&oPl>*A z8tRpSSLr~tnyi`PHhMo5L37^KV}+XiJQ(L=`*oQ&A0Hg=tJ6HedYyZoeCQlJe>ynO%t1Noj=M`iH`-2-RcRv`s~ec z?^jx7dMm4dnBYCgP#Wiu)J4*f5+If`P6NGgw0h}#JO4LKYK-0w= zijbGbTmmMKr|iBL%~LhCmkBGG%=tY-Bkz}yj8a_sLJy3ThkD_ZUXg%|-A!VQ{+*LS zk-hOr<9AI&C?{~Dg&`}bab ze2{wP7|0`&G8UyC^8s%eQ5_kRg-az03mkU~s;5`|xOMF0giaMOJj(ud> z9U{@Jbrq$xb+*q3`Rc{wE?W{>k6+)cJX2>wLUG8zZ4qd63UjaYcFRnMR(?*v1;~GY4R>!kXqvQ4pSS=NT6;JaVs7{k)>;w zYCJnVTxX4-6AY;Z$BFg4eHPaD5qrM!IYsCy!fyPR%>}RtCtM2KF_*d;gVfS>Fy=Ua z(y)s8qwL;wgDijOMXc$u&>fLR6XI2<&qxQyiVla*!yESUg!JOhidZ9vQQZgF2A46S zKUDKKSDmq>%N+7DGp!C#3aLWwp05}goEx@Qr zM#3MNe8vG~s)9WVOn6W;9PE}vTsdja=l{5R{%~UbPiIM*AK>mf1hIzF{>@P`4gd0E zKv*Zsp9zGB7}OZ~3fALlQm&Y>!HrL?CCKP}A z)&A|9WW#{~ew#S|zcZ5kZyu{1um&azY`^>G>?i;DXa2VVf7?gY&sqN4xAb2Q!6yo4 zcgAZ$&;H`TJp$gg>4T&vzv-m_7d(|1R|ZUFiSm#Q3il(*N%0 ze{o~_-yQug4&49e9X*==T`VVFtkJ?$rRg55^NYVfU}4Ow$)!y?pde7`d1$5EWTupb zGzU&r4UVff1~NCkY`0t(MKVZnv4zqk0`6e!>!<>?RCde}D&G5KPYEa~Wiw(5GW`+& z;6VWfhd&a7cF0JN{#I)C9~Pzs1W1*h8g#HKME(Hq05(7mGnPf}^V4v0-?&RC0T8@5 z{4NfF8H*C><|-E_2&VOk9EFANqq$0~Ir5QTff6x6LYK(yk;L{7yU_xIMi-y;yiq#I z?cwTzDG#Z$g+KsUX98^30r^#Vy0PNKtOXCa?|COGv$N)xXjCUX!R()m9za@^)&OKV zIZeVUiM;XG^%}hP#tN&<`>9X1As2HAv_kH_gX7v@a>s-Ry4mzw?x3p#q(5>wUzxSL z!F#j2!d3tqDBEX%8D8F-wgAVAW^|v+q(NNPBUTsbBC5vO8aqC~CEd}`(|MaWh?pvL zOlmq|R_x;Ycmt0PgX(OfP3oDnsl1@!5K?;mkcAN3{cS5+>oJ<^n|>xSUp;e87;yUM zn&7AY^N+>6K*Q{gnhbep(*`5Zt0B}&WpjEbmNnjcS@Z-->nT6T_k_XW`_QwjP-q^uJ<37MMGyd8k63lW5ZiUlHiO(#yAL&C(0&kozT?CH`3t}WUFphTzhCYxng2$zaUc59_|y%rCF4PqUAS> zj@+7QC_<=UPgT0#A2&MW$UpX>yl}dA~A%OEvNC zwv+)2D}mRL%L|AN0K%fIRBC2NHG6a91KbHQ7zg+@hr+mnh-nhCpM@%>2wGC|4{##E z%1qa$nmnp=RT7~d(~+g8*aNsCt&%gWEpsrtqjf@n zt3U&u+>*R2Brz0t`sKQ731E2W1{?4LOxJBsC-iN<)D|`e)9~uuLr#>&ou6if*;t_8 z0c-v3;UTsAq+oO7SDshhJXipgf;FV*q6`{xpYATj^5_fO#km^~_BL&fGCKbH!P0e* z0ZrsHF4@{N0C3U`=`NNsOq*~Gx*s)sa-JV_1O-5SLyIWM6)2fkl{R`mUg z5^L9H)KFQX50`9x-K+B5epq8yqt?V^0CK3tj*0sU#11ZPIa!%=2Nf}C6c^x`#vtM1 z_|4@)y2_e-7NzbbdQamLi}V)vuJ_Kf4JKtE+Hl?NqERWmV;k*Ssvq18-!vijs{eKZ zR1F4IQ~O@c3^5v1zC+xhlLTRQ0Wz1NC>+9y$*|9n{`xjE<;~*W&sD=$#OY*XTX4FE zvhq>1H=UcPDyA!+0OJ=7rj9&RuVLp-Y=qe^V$ZM?rtGyhLS1_r1n zN}BA?o`^`WVs1*R2lT(y^l7FpyITX|{Z>9-Pd)ei`Py7SrCsxRU9r^%9PMI4AMx3j zB&+(jCj$_!A0W?D*(Z(N45#E~$~A{MfI`*8v@rEXRC&a3mM0%ch36n{)B3n=sg_2a z@+7qx$O_2^*?E%HS2sCnRJv4$|MC2rOYsZaSskHmZK3aJcSCnBL~`VnWJpdtdCIS0 zd2W3eFW72&AbZwd4pGh&nO7WhIqX)2PKpS*qk^EN8cl9|o*4RKDx~q&^k^y-u^u3C z9Df@9$wT-1OaRg~lBzF}?0`5u3Fy6dMh3PhR#cP`QVY_Hnu99>s6g{*#HI--fkr$peXsUpIp3t1ieo8Oa zWCjJ_6)ud|o$a7IVzivj4ssKJT&^_?lOA9EsHF#b*Q4`->6&gC)O8S86NvVpkynFU zTHhB=r{w6gc;M|=2=*w|z-v&*)?>|Y^L$lm?WH~w=@;%S#zvm~-`lQa1#zjxWh5op zm#GGY4<7#=r0vDCot$H6+rPIgi6qdgfB%yCwxM)nw9ONB=eA70MpdAGoA^`J@WPi) z9A4TsW>M4b^gc=G<<*Q|^3*R%^%|;(dr(uA9r$gm_(P>cM~fZ7Fh=B95gSe5*EpY{M_L-6Pcg3L;D z@6-jU3{<-eA(xi1)pXceADtH>8W^9|=j;2cskXHYXgM-TWV&1Pp!E7ci>nNl0o5n)zqfb373 z#cM5^1v?m+&y!_GMqT5(!P%QNURJ7`boPwIddRqb3kW7JP|$i%bvs61o-`x9gBD;S86T{(Y|>UfW75 z%JBwN$E}~mhqVJwjMkc_c|i3=E5Lr{V)y~VDt%$lLNBI~kKGF+HZ<1E`RupfS`*Y= zO4OohArkx(T}bQtGjx+$oG)K(eHKO z3JC6+I#nR5Q@#C&`X?x>S@Zs-as4M>{s8=wQst%~=>AGSqPv0!0TfbP;l}yva<`gK z4;;ueorihh=4IeTUN@9w^_b6QGM zMfm#M1yqT|R1e%^SsCGT?O^Jr+KGzCfy=}_L7UrZsvsLd6L-k9GLSQQOO^`c%8TbQ!fpbHWoULo(a# zP4lO?C;MP!DD~@%?>IWdQ+i3qeH-J=yk(7z_z9)D$A^Hw2T4iw^4frX&`8lbHbFw0*F!l`^;IU2 z!SyY%0cJW_>G_(7an0>yu2Wh0`Xfh>ZIutkwep3RT`6PcGwy6mYC7aZkLp}WPimgA zwF2$^zIp>ScxZA1vj^HceW9XXx_2?)G}J(jW!hiuBFgaUd1G=&19o<3nZHea1|kq7 zWt3cFtgj$=y4mvGDK0rc7a$R`MdY0cjz5mHH6 z%bGn&iBQVEC+lGB+ZYp)B1?pjExR$sKE^IdcG<={$TApa?86K*_v!b&pYMGi&+j>| z-}Qa|<2ab*bDrnt^M1eHuSGHEMbLvx*g;;;NLjuCZM{Uz7P-0w1gsz2`nMXm{R zAf|z!Kr1dAs$A;v=tf-g5=6N37Mb|0O6Fr3GK#dFKn}mBZhrwin{H?cvPV!W~x)9UF zTj%O--639f8`0XO+Go2FspWV_Vsak$te#`N7Cq(@A8T2;u=D;@YB&z)Ko;`nDY4;y zP|g~Z%wXy`!zh9V^wQVdI59#u{7KcWK^j-sGRL4+y7V)n*QWJ+n>C6{z-ocS7mRCT zN3MGxNJ(PwVgu}T8J;%2g^h?l1d;&{5TaLQDY~kzSE~ZL*91~D*-pH5Yg77XRpq~- zov(Z+gfoN&q3d-%gVYoU^Tb(u)YLVdY6VaIs^B9^_rmQL=9*PT{TcB$by82VfJX8V zss6w@y+~NmehN#Z%f}!=7yczW_}z+=_wo-g9q3SVB*o z>VNUWb~!ycya!0B6Xu+-LVMnu}8172@z&BA8jvwYNYI;cYz#7mPrW}X5u~3dqgny?pVb>mWR52IK)yFe~?Co z_7d13V1J^(B<2Tyu)77ak-TaQv?79cEKYFs6Pf}4Q> zBr4H(;6Q+(UPwY44JC<`QL_p^qUB{pH*)|n3je--mD}l_a_UmJ8$d2T4ZIPF?3kJ+!YCC-mO-2C zgKJA^i`k5PKnHf!e7m?W!px!jko3Ljh!R$}J|Il!_gT5dOg)rxME8!>+I4^av~Ic* z1n^ACM<3U>V0lJGvv}8!7L?zfUsw5q$MJ>*JSzUeH*={PG!lOz>=gPwmBcMg;(K4t zs3!|LH*Fe5-ks-pxBJibs((*m-BTF=j^75|+C;UsDasX3mtdGzzg^iMX>dbj+&*O6 zPev0YFwQq!_r!^Xa5FzZ?RTuo!NWBXeSChaFGx<#RTyDb9XTvEo@5qR?Y2I#)ox_f zDUG=A#?c}+?ArPg1SDZ5#0zRYtfm37{kmYhjQ15Qvu2MR{40zo0IRx9?12b|uE)%G z4^|tDs`>JiY*%53EyzwQImy>`qIPAU%5Kz7+^~iEg!^Wl{w5Pax zhBp7JJKfFI7%53GYj9 za}}pLW9JkO*Xy|&K%k>Fm4K&k#H73_TpwLw+)juW)Yi(FLermEPqqwbXXt0A9JMn4 zoD5?PJ9YN`D&;8?3suOj+y76a9ovg9M!>mtL8&47c}}2DN^a(Yb*B9+L%-0|uggCx zy0#OmwUk-@7A^VzKuBDGQTdr#6~oz)&tS&JOj49M~BhDGZ)dpMlL7*zS#* z-P@gG zx|T$!&wJJyTFxp|3UW$6B895DH#%fa*egG*!c{y$=UYm|M zl5klndoqwHHtCkvBzMSq<}$p=P#QIRusQ;tPKM5;Nct4>e6vhfc$K*I56|)c^K95N z|F{9}Uz=B2f9xPI$l9-T{TS(+BV-_(xE)%?S^X(7yyCA{c!Pj^+lI~-c(}xISq8U1 zEQx_tzeCPD0|DbK+(Trth8JX)$Z!7A9o!fXL3;}z%Qw$_ zykZOJiX{zrOPK1y2gL4K)*B)UBT7mWaUCCdMLJ@5y-w`$IFFvKM;N$R0-`Z9qoYWX zQ1Z(n{q|^^Y&+&G+@NsJoM~31-T+F88{uSZ??5`qItJmYoF@d zzL=~%T4FK{Y5Q4DI?7OnOpzJ^aPFX9GWy?_=HEy9-yS>;g@4)E=hWWeEDkEw=1KwL z*DnriIFO7i&%-NR1XczlV5gOuvK{(T1tlCuyqSGeHpgr|kE$_0x%+uk%jZV+W9Gb( zsAAk`aHjl=VDgHPoNLIE_sSr8#gS?&;K^KVil0Uyja!teQmjyX%0G%n&EjOG*H*(v z5UH(_H{y%~SLxPNjJcDPb7 zC(A{-NC*E8MgTn3Q_#r!+3Ywl=@$^?``rzUYc>lT5gHQJBq(3q1X#-zCj9SDV?-Bb zs!XKNPnv(m@NkuCe@nI-b>8!(jyCbiXa#59y}o8lgzaY^bDl{@Jzm>>RU~r@gp1p{ z9Lw*?}XvsfSQ;5Ba}Lt5%}jr7Sm`=2@px?x%feN8NmI$PM9*cO@^KHbum05 z|L)*VK5N=JE^HkEI%ZtTRRuzUNq~9QVfJIbPXVU86ZJoRiO1Ng@%ym}OXX2uRPdH( z^`oVs0o>Xw?T*K{AnIar$o(p)aLZR!Y8=qHUV!x9FsfXgkJ!y-1A%u4>!s1k&#c#h z0bJ52E0`FRvg@@CoQF}DT>3s;j|kfM{df@8@}@<;WC--yW=BP+1_helnv0k=#=^t_ z97oyz)wTp){TAopUg;*8h3ChEvAw_Eh9mT9dc7Q^{Wg{{epn$I0>h9f%Fla_;;Vj>%T45XX zg}xVX3?t1j(XNwEQ>2w%^@cy3$EoA6n?)ooM~*O3{h z1!yPdzVA|*#x6_Qe|DSb|EV--YKzbZx6a9Mi=!BCu5~|{nnZ#9PAoHv*WRC~b`wK_ zZcWLg>vbgwsGnLX#1z%PJ6xX&b;yA9g*Qmqb=^?(T|2!d3`mP#VWjozfgN#*o~cch zX1>S-996;JXFMWuFo&5{x(77{ z9lSn#d*&722iKKdwFt{V3Dj@|A5b}5lkmeY{jBtw4tQlHTN4g^KG!r+Hp#I9ww=}S z7kt0w9an7`rjv@DdpBNQS;HUxvFvXjm`#*gkdH?Hy z)YXXMOtg}I|8-!rNr$H#=(Xv(&bXRXynZrVJ{GJ6l3CEiou4VcxpR+wa3-{eQwcS? zaYOhjU|VZoiN{*JYoay5f4JCqNO)%eTxawmQ1!-4x<&ar+Eez-z@2e^Gl}K%9rdn= z3A7A+Vd`Mq!NOhk&tF1@#=G9|Ah$#$BCyUN2S7$S?Mo4A6bP|S#4t;~$)o@CuDp!3 zk|zAEp8@`|DF?O~YIbYFlrV*!R<^2jw>dB2pK+Ga5~a)AS%SgS%VYriJLFCaz|1b3 zDOKZPdFwD77nti?ZW$E}C2O*=80e)+3f}Uz%nk}ThNs6swWEi9iqq((qHGBB)Fjm2 zic_+={P}gKqLI!S!<<}4U*g`2Zog{%CI%{ksae(AjboYENFh8_E<3-zIUdFU-Mw@Q z53ya9Xub!Q@OD!Rg0Rix-Q44ggpVMN`T@AOpXNOJaew86M|7*gKSPE7T@Bt-rsFy+ z5R><9w4Z+@=c2~)EyG2-t6X=_c#OJsuI%d0sYXUNe)(&LAI3xHKMCKBP`u7OVyZy| zv}js@SI-UYrJlLE_WGU&oxWEfq;+=YbGVV!aYKzGb0pz;IID8_m;U9>1;!UlIINGndxyRFGMQ(ZmE8xakMSiXR|Jr^| zp7{!=?LFI6T9%aZDm`r+Zcov2(N4G-4L_GTw%)j2Vc#pf=Tj;RpAD{SYkOkjf_q-` zH07a1{@QrO2WWa^5@2p{8`x4l-{Vk>Dr19R<2MpKIGu(DBIBRBBg3V$V0FZ_+L#2N z|Fhfuqo2;1&3+$?G7kN(y5;^Lld;{QRq)XWOW;tn+|BjbkN*kj{kYQU6bBR#uZ8Rx zfp>2?xLpRghg?<+Zn&c0|Lmqt$~D>%DU)&|^breBM&{M8GXL3j;7|^5UMlzhcq|{k zc#7x-K*^B|`tiO+AYUiVFtA}O(%k#EMu^?hzb(1``?9o*3*6s0T&vZ+B1m`oM#vk6 z3}m5UZ}ewn$li@hp9KC~sNMe)l?O~y0h)mDZ9wBZ<%>teE1T8S%G&V$XJ6s;*KA-% z>mOilvk`c6e~=j%r zTxU`D*neM{e^AY~$X^8(?@I0auXO+6QvB;G#eD@9B;iGJ`9FBX{yW`0#OppFl9P8o zFVFBFuG+s==c@cMxKJQ&`yXON|9fF@Z-6_}YxbHa)BmrvyYI<#PDxMK;^(7(@2daw zq8-P;1m^bjzm~LRT;ljqY<~CkKN1wRn21{2Y`5Ar4GkP@wxIol zv4^vF+4D5oCde9pO;^tSmwsk;nE9NN>h99o`G2rD|JR=g-SO`^$&#*j*3mf83pt1u z+Zhc&zU`-f`LBKE>nq?Y6kPf`E#R)gbO`jZo~;aAUxx}ZzIZ+Dk;O@06N>nY?)e!& z&SYG-yDVn=6DpK!RCJ@?+ffkcr#?EHDWuoaO>yA})r&oUe;xZ=Pr#O^zAuS!kNzGA zD9=i$$wFiLzvF6Ds`Lg<75Ljv5|k?pnzguo`9+d~2y3iuUN}?Fo2S3xkN2Eik)8t$ zu{&O)qT_&2^L2NUPO)Lpg)tN> z`h3@MytC&<_#EMgVz+vz5zD0cnqCrzF8dXk+~Sc}d%)r$zc`&(JVjo67!uPBF_XyYd*M8R;W) z1{=4Mu_4mrI?L*a_Mk204;=Y}Zfx|Y-j}Pubj^BLw{zN~+_KV5Ik0y6W!+lELxz6X z{z^ehHO*yz10ag#ZXyQJ(^9~Vkl@rz6+Y&(btMW5l2#;D=uQ{jU+PK8eF`K;1gs)Y z#3CQX=sw#BT9C`Jqr{|#tn-W_6JBR~@pxxaC71bW8*^S9Admb14_0qRg)B)lbFZWb{^IF5u2`70$!2mp#CMwHG$wA zjlbVwXU)%RYA#PHIxuJUTh{&gf!VrI6xd`#VGY;}>u<35H6$koe9Xh{&LIH*J@uu| z`qP+yK*ddq+40a#>O5$uF zmxgi}nOfuiVdr3L(m!@NlvzAZJBl-D4<(a$;h_KPtFleLd%S&tAM1xbwT-as_B!%YtRZ|i-{`FL?pNW7GZZd+7f`;rgm#07v+Xa zE|0{l^B+xAJENNnPc*rtNUBXnb4xwqmUdffiWkYcRHbfz?q2`NR9UA>-an=6dq3%< z)tnTIlCJcf4opEd-=df$@Zc#92~cuIJu$AtM6&+qSQ*O)3RqGaV4vyq!~ZrTG|_&y zV1W7d{N&r+=S*LFNgAIEH0Az>L%8p9$d9*u>UzO@gNxZwy4{C~cG?VkX19!3JE4Dw z<00Wse~7%XX?R#r(v|}~)kT}|^{{e1)!^5)m-dC%WOGP03HiC+a{YSS&BHHW1p3BZ zv6qzv+>;Os$4*g(b#jK^ywAwj-^-mPL!E+`5Dk*R& z;LWlXZRj(Dve$WE`7xFGxp_CKtdZfr~EyFQWRTik_# z!GV2O{gG^Pg<^Kh9C6X)&N-iA!^p(2GX$u8zul8e-L7Ty4~!(JIoAajZ=+xLHKu%z zPw7^@45Pj4OtyN#{JB@_mGMouq0C#}khQV9<8_eOlIaSWXGJqfEQh{dgXtbTFM5{y ztYp2vg8}H?Sj{`OHxCmiU(aL+q3+CIfAa=@I#^uFScCNTZJ%<~0uz7W{C4DDbaM=A z?3Y%Di*L`w@bNEO4?Sl%{{MbFrcq6Nlo6!sbF6(f_cGXh(P*~ErIwX}Wh71cn7yqo zaJx%tFRHZH`ycA38;@pRtK+Cp(_cGJ9l0$1v&{{fdQ#kP2|>?tW)Zjd)rFLGA28la zKFc0R*on2u+7*-2TA@1#Rp1+V6-ECjI$=&#$EQ2i>36%4>Vdureo(xSU+VOY2%vmh zcxKmAMt5vQepu~Y1h{~y`L0h00@f&_t66q{>!FejW%9cL+^Bo0H%$;q-YG?jQGiq4 z4#3JQyvHRw(?ya$-?B8X&@fT<1~Bl7*?v{K?j*5zIk4qzgW}JMtUM&|nqyMvcEw*_ z00-$iU!-!1^b5*S0n$lE+N8-GeZM(i8`rG{o#+xp(4k}$AA3yzHc$)t1(A#0+V9Q^ zS#kKaM{-oS9R>PX0QMY0K(>6+;OeEK6tsm1C{es#^j22NFvc$dD*`t}c#_66xVw7N z6;YD~44dCveKW<)n#`$Wyu_14h3bpqFK_j2S5hl+JsxdbBax)v?>Za3^bnNVs#hAQ zMhhh164UM6NNn7re7yXJ7uL?iv2Xzz=sN#1`gqKPX~(MXQQyOl0PC@O@S3vGw}&5U z>nd17wjX6VCA?E>-3C|@$v!Y-^5MbOBk6s}(Ww+gu-X9mW;D;qUW&sWVN$4j_jbVqjm^E;Xe$4OQ!9 z+-#E*9QsJuudh9dE3wLHaKFAim36l(Q7Ynfe{jR*rqqtr{W=U=me0o6UZfD`0#izg>f$*Js|SdTQPC=NC4Il zanerWA4cR#8R~b7HuJg-e+3>{W`VCBp@Ex9!t4>vRhJiIU!wl^XV=xMK!v*b%7K#` zs75?cK{eFI9@9m37{yD<;Wx~1dB$CiIJOvb&o1(!XTZuAmq-J9%NjT6tV^k=(S#2F z@8;jSR)@(hV>RlM1$rXwyI`!w4iPLgWKXT`Ia+ztw_hd)TAX@HfgrLHwj&+O?oWNc+c-Zcs%!@;wV+@SAtB` z#nGwzAz#NE+dh(HvT^5~T4p%Keg}=fDrnYE6StMgeydgh*eTojRRCJDfsz|7g;MN7 z>?PNN_x0#!(G{;b>uDvS!JgI9scDXrB_Biwn~VCb$|u(CQ>bE)Og>oMX;>S!+yQxr z4n!#ST_pP*POZAmMCm(O3Yb@tRz#sz0f<_A;R=VIt4X%z)lQY7Kq3oOIq)UZ>Tcar z;v(}dart-E5C4Oi>rSFzqmr`8#OdT_QR4+H8_N8JI&o$9)%+xh6SiGy%XGwN&`VP8 znu%ChX1GA*6(X0fe0alNW=;6wF~5)#tLiJ@sM^gy?)~}w7l1I&K3Jgo$SQB)k&6IN z!zX(?1fYMI(VzOYeAjj%j%X_?`WfQ4y=3aNBhF>W#M`#ShJP&7P&9cg41u-m3l(Yu zl7NL3q0FCQPehsbRlLY(QSb<#qUt&%521!n_s|YxAzW~ugBJIGkY_o7+3k1SGmq$#h`!6I{+Jl z{W}mUlX-!vi|HLdT;F{2Vd{_N@|jNCSfz94?OZggw*eE_`ZCYuQ_Q=`pVZW`Kw41~ z28LvfC`}Shlt=`qhA-S)UP}Cta4DDF68o~>)<2jJ#kItnft{{3TEPVG5Hr6>A8@UV z!!vs@{1r!H_3s)ed#mD@V)ZJqa$I##yH*kA)*)#&rk{^SYK@r-mRTVWR+h6m0In;Y zS*>k9EZ4cDVU?@9ax$*B0P)sdx@j@%J;vAjig*Bx9aKi!cfakJuL z?RtJokf+|-AY=5Q_@2~n;!u+x+pUKaen=H?cNv>1OGypmE$r%0TW>@LuBcAJ1(#6p zG~qGmkY7E2oIR!$W&14ru)<>eO?oid&hwZ%8UZUe}?yF>k$d`ZoP=APB%I2?c)MwBroE2W# z?=XasX3hDaWC6+Ths1~mM<(!WlM#)A};z( z=b`V>j;41*mON89qd>I2J0=Z#V6Z8(=B14*)Qy3R_#Sua^e_> zfNq2|6l71i_dZzFb8_?tTz^6a(BYW~eU!oMAvp~!{TU-Nm0#eJ5)K^Riqtl1ViVMs zArtX^H%p)crim_o6#ahA-fQ*nJF~dA+nguQP_en0RD_=N?9`L6p1tAc4qmwKh6v4Ro693|UjfX*8POv(>IXyKTu3YyYu~x006-k(S#r|e zZi;}Uv3?3l%aL6DK^uEd{7bhZV4UR5Y%14W5dJiS44UO)pp_FQo2^t1{of9Q{NabP z%KNRnu;zgkDt*Ms@IG}+W@G@o)}p%0F2rK|;GN3UIpZSsJut}_`lWToZ6tO`%Tc?F za?L%O0KMU|JQd@bqp~Q?^H|!TH zP0Q}7Ghx4k%ty{+%RSyTkh0DF)NNDbXnNowVG}k4Pa604;=R=e|o`2=YckF%{2J>KbR5UkO)=)Du=LE%< z+mwYReNx`!&Cc&8;`^iCI|8nsIe%?M^obPVx=C`eb!5*6dmIy}LQc)f;(1F-03q}} zx`ZFsvt5VAyc@CQ+CRHGn`TnX(MKeXOBgOs)wy8)qS-0KEB^SsvfN@t!1*jzxO9f;&s>_v#X&-;M}+QL`!5dLE&=o z2{(9y>qw4k32mgLNDb-nfmJ7hAaQq_^Ud1fT4m`g3hU8w9156ST0-VoxUQfEno=~J zXFhTZ4#8;C2C|ujIB7t(xtNA*x&Y_2%JZ~CjU$H&x^d1Nu4eUa={qZ!exGCd*74-| zC+Pi5X|_c-rn)W=AVr!R95e}$H56+w66F{SowbCcj1Sxq{El~`+KV&-31jR|4vbz` zDW~+Mro1zhB<+Eb_AVh_ezDFz0X(+Dzdj8DlV!OY1`;rK6sO2oCz$p$yygZ#*I|wyOHF z3=3m9Q*~oxk4PqDCwApmM-M=3xLnoE=enG)3Bw0=CP=;dpCh_^4QpzJI{?WC`7BJD z{1S;0uMR%@xp^03)kNOmxr9%CdzM^P164|E9&ZEn${ZeI2JNWrHU$p7WPs1ed(L;d z$Z~9LX?C{OU^;f#zAyEpDnn!zy2g&0=k-alYcOI^eDS_qIcu z*GCxp*03gg`W!9fo@ZW)K)N>TNm;a}>Z_6xmAjU8FF$1IO_W16K0P*je8_j<(wj4X z3x|Yk>$eD}npL*=J5~zgu%38ac(*!*mn{!6|HI<4viGEvT@&v8CjxQIFz+AMIo*@EL0MY983J+ zdPJ(U4aEiQzUV@DdQ10)?n?x3q^yneZ~JwxU6%S4EmPb{zv~uPEErWk>xJxZ{@_(R zof;5WF?018Qo$!bqCJ9v(^JKBX~R0vKOXE@^2qPXl_P6SUm+LYXBI>n6p~U6Ou)Xb zEJjX^52;^VJ~JdvvISa|=r=SA?gseVsv0Ac6yLcxuw_UjB{9R^&c?;e?X zC)@CMpG7PaW$JIDEd3Ivo^Nw^rpPCtjF-f%1pCiYueP$iJbCK9!>hT~YBgm~yAR{2 zquwbMkCu9&b?|=lPDq5f-^|VaDUT>)4nL$A(Em|pw6UHv{2OpbJ>G^`sYvrf1Y59s z&ls~Vj_5Ui7E|;)kkj@=MhEJh9qw8hWm*;bI!Q{X2xqE|&{;Ktua(*!d}j!YnVUPc z_u+!i0<})=q?p4^54zUQs2|a-hNmB1{FTG;HBHv(8r%q(M6jsWnqcRsa1Hb3e(F3` z_({~b_8B}zH%=pp>+ux_K(#Vx-d4M}2wv<9;#_ zh|!#6W+Ex?Nx~z`wLuM5$E;;Jme*#?EVt3&{7AP6O@>H|YY zD)!P9>B=U+67&-zvDpgyJ8Sb2OEf?wvg%3{!-uw2+V}gmRR@!uIstlZZuWG*7@$zC zX4)yYZkBJ%O&lOXXHOiDPR@=nZ30C)_v`axS6_es4)wzZ*B_E6!Gq;c%6&Uuy!{+} z)yQ0cZKZZv=8cEh4hjpOnd8}hs~ZVH24X7 zBvLImG>^j90rrPtT`r(MN7|9}tTBaNv3g+AKLU5YlvZTISY) zLN%x!a-0nDip^gspX0asB(Sr*9|z(LSF@pC9?u4G7V019-_B~TDzQMRv{-Yrm?>{+ zT4U47w1%lLe$kf@r}izZGid_xI5X>DLe5-5!kaKlS6lNvvi~&&jKQHP#%uw$h>`fS}3^aqxWAz5KGPh zi-y_#=JH-bt!km<0V9!R0QUG*@EKB!O{Z12d7=v#YYDQ1MJpZ7PEXK0$D6^;!9>av za)bPxXBv}yj{=K*gf1&OKP<7X1A>y8uh169e%3ddyf`|@dPAiIJo8-YArOWbGkU0+uC-|b8kd7NdIbIQkJ*Ntey0I6-|JNXMw3CH4tXyXN8`q<4~#BKpT(Wly7Cje$eA5a8*^)%JK=n zK((Lg`_d+CZ4; zNKtCEnPd$>9VO^zYi76D1gorYt;h$xwhpc^RB2u(!|As_-4u1F^u)2zhgdS-Awep_ zbL(&TNq)L0>eliZfuC(IdcAd2*1n+n!y)j0VAr1){Q#G z{(+w)u1?KH3WqeqPx+jcl6nF!l-OP5)>`Pkeq>htSyU*pZ~SKVJDYc(O8KGUa5+~G zxTNztVva6CP_O38m5Fs3WD;Zb>h2$QjqhtG-rwYQi{0^C`C+cjLt(MH-;`@Fiyigq z{49=86lXpaOLfg}$W?=Y3~wKTSXxxEH*T^g4ddo~b7=vZVV$ETThtnM`N*kN+m4G9 z?`YB*!5hk&?nilxdARE&hEk`29Xi(k#|t3MUapIo2+KG4>44TLSf#%})Z2f~S7IgN zU6_!Xc@I>2I1A6sG}yix?zMRKqrAB~)mJqTfyrU}CBxAbSV@>;gAm7$Y}e~kyr%AQ zcJfL+%Bs{gUY;RLQ!=7hCR2+m~5%DC#&% zRY*Jz&Cj?Fxw=T$4U+_1#${y)CQmlI82av_hjmt=B;JS09H)>(#$#`U=q1(Xf znuiP}LS+zbT)ktcr?Mi8$~q!SIsBA)X`Rk+z+X#L2{4My@h_hR4_gW`_nK9ntFq|$ zQT4-o#+$4j$FRIkWnxK(bi@IQ81zEC_S70MK5A$$w?u3;s(KTM1BWq+xyv7{NjF%~ znP+;7YfP7CCfHP|mumZuI_yKd0VUpdz%t~(=q^B55VoG7rM-PAhH?kh%K8TN4As;$ zJ;@=fO^be}MxHVgn{u5S=5e+Z0XH6)UOnzHpp=~#Vqw9z(~!;PV_`3zjN9WS00Jk1 zm3ymE4IF8JsTn|5KVV>9Zj(i7L!qzaN`344@QpV73~1v4S~|6HVU>HtqEIPE`F4;t z`$~bD4NU^g>5`$ai>cM^L_q$OM=Z8w(@z~F-?q@_*Q$F7WXJFjeLpFjrOdKB1?xrm zHwULm3p$>Lt5~6Ydz#mmly5#A@nVjSO?N+p)XCjZ6%T zP^Qvkc{K}P__sXG2Jkd_Hn%?zHp}Q8O2anVi`j2G@$fgn&TBrRN)oN zOiiR}UkZO!@jR~_BhU-w8BD~$L+cpc6*y;2fMR7!xh^W87P&xoN9-{p8+ElCyF2aL z^xN6&AJ9qj`Tb17-!OA~X>{3M;2`)f*Tjt|T<>+A(z*v)OR7|qrCH_`+}ygH0Mb_a z3Q%uR(^KV`$!2H^TSYnA(eljCN!bY_=9x*-NlKu@&Pl{mDtc<$nLl7}nk`IeJ%oiO z^ov3=(e^q|9Y6z)wB~Ld`0DZ@+rD12rN#+$USC>3wwIaPP`_-m#^65F$ONQi=CI?_ zOG{pGvpmoqHtXdl7|niP;LZ4b{=I0K5nS^aF}?XziEl|US1-l*s)BiUR<;vdg|~EE zuE|_JR@Kh76%0#5&F;(Hh4)zGdppsJvsRcEyQ3H?^=ih0MGJXx$%Y}Z!=g$U^a~-b z0n<9~BEw!62_P9FsfXLJs*A8${hj+yf3lpXd+DoirVxRpel}|Mjk0Se*a2IFo}=-N zxe!FjD&&FLD@51o?4_D&=yItd*k^7SK&KUNWv#k`5=*vMhB_=OS9_oE>Zjk?d<@`P z&WSn5`n}>TJ6YR~DRR+cTQQQvJzkuh+F`knuqLGARR~quDMt0-f>Vmr8H72wqrYNk zGku%~OAGkvGy*qB`qLh%kn=l`4F}EEJ9AksxT4-_{pIEWz3a3G@y)3wGYGXKIiur{ zfh*7llDy68C(TR0;~-4)ooRFpQs7=D`GDK0_er`Ptg-J^Tz@t@T|Md);=8Mt<|J~Y z8fcWSd&M<)v|&dz`{$uTpF^T}x`a1&JH_$QPO>jcJ7iKqO`*&`+2+#!MRWS^eb?+3 z$3T{UbcbTf>LUHCpm<#$J0ZtX9`|<%kY)4!JkJlz3|c;8PTUH}7fm0fgdx4`QR>yx zjBHOjg4Gwx0`zNYs+Pb}{J#EU1%qCNyYse|+AkO;oVa5w_ely4!%YylzENYHf9?RP z|9kLt)A4Di@ZN+5`pc__qxlK}WqY{PT@{;sEy2jHyW6kbX_)H8E=_V(vcd^}5d;OS zIN zS5Ie2$>dIFZG|_a#DFJJR#jR-E))G3vj+uBPAfU~{RswKnH%dadg_a8(yo6#PV{$h z>*WPC5)YB67zdej*#^C<#`WuiXf#j#J@@oZVSoCizQd0zgHIBilRbe>#i#gZM}!em zf?tuDfIsK3yf}7V@2V%$b?RPw64E;4*wSp+X=s5c1A59as#9T$>&p_@zW?;HjvfA4 z@0p8EdC$O#{u|%9@ZZCkDJK&fy6?3G=-91Ye;lijQywT`SXFqIvXEemz?Zx%d?pak z0EL*zKv&w`{gZ(q&M9L0?#}d?Sx7-Zgv_P<6A)#+j-QIE!)5P!9KJAZbwB+AJ09e^ zQheou_T38&z9~xp1~TIzhL0_Pamkqqu{f5cF~DzIn4%+eGM#UlGco-gIXzzu70LxK zR2;f6Zy}Naia6WI$__LytAKk~#9b)a`}ci#&63>og02Gq+yyacSy(&}{wyIPVp!_y z03?sr&(hCZeuyhbw&UPgZh+`&rKSQip}#JRfUqPq%I}- z(XQ6Z?ckfkfJD*pzbawj-n7NTj`{6|QSjIeK(6NHSJa9H>1f_Fh8y8b4_@tucK~U5 zQAzv07dEVr`r%tD%0Wjsp&fta7wyVmtkRMSdPB3aIY}G^)cU=QuHakaw!3k^UwwCH7QR^yfcbXMf2+cA!LJNLy9D*}j-e!sP;#Xc45 z;L)M3`H4d!Z#uTDHIMIrpvA>lr3ES5E5Lc!0^-=}k?CZlF^Bu+$0aTg@T5y$?B8>! z-R+nhmxGqam|-(s#mVrute`jMBAje17x#%atv}vavPED@u*HAic< zThvSz&%~mbHc|RS+4Q-v{95=B6Rvap?4!WgU+GeV(hlD<)SGOm5NzN~3da_7=F_Z6 z-I%#HxFGb~5^>?KEATwsRelEz=Di-7(rm*BqHgaTGB0$UciA-7!3>b*f$-;a7Cd?3 zI*Dws1OqShe?t~dG+l2ZMfx3_Zf-t;;0|We74kU~MGdxD7tWwll}p z$u?h+V~vg@D_xiKje!wC?MTD=3`z(8=Br$TOke}Je8>Z@oCkux6~jJI zOT~4&Ki^K!%Nk4JTec9(aRvfjDq6idZWb^4P?xQOKLILajp7|dVoTJQXRs7Wy9Ltb zd`@LinM5Uk-~}8-%RPh#Q&0{07lNyc4eWrRMbE>158AlOd`eE)HddYXuv{#YrDOiR zZ>^*6e6-{ig=k0qczF;IhzRz!KW5YUBHc+q1IOeZcl*3enbXPm#ukJeYU50XUaFEs z#`=CO#4@3POdSl2GEsRsYdG5{dKMbCN@zu7n&0@=1h=d;j+s$)kr`v`bmB1i{whP?d(pLCb_>|!6~`;7lw zruIU%LZo}7@Z$Pn~sFHdwXAD~%c5=d!VZX5- zX&w(D8k`UR7N6G~rK4tV?_~(l1Li9o% zL^TzUhQB=9aQ|$+eQ%sv@n=!aFd1>tVuq2_k&>I&3;HV9cl<4zNV-p)Dhv|1EQB-i zZN~9RiP#$D{2FD$x1a78QYjqd4U=9I=Idr<}FVBCR`q?QEO@-03p~&6utZ;HUlUzW-ou z+?URpDsg%N@MphX6n%Cd*hyR=-9A^MI0 zmUYo*D1T0v^$I@iozOE}}4N0ZB2ji2( z+RsSgqzwMNNb49WMDi%HZOiPa@)2td#er>sVvju7M9d#^k7C?`>0 z8ElymYTFdvMP?LP1beBprjq!&;$=W(?H7XT45cu$2deoA?O)Med4S_#zB7UE0JIX$ zZ(8J=F`N9B_ojM?nWnT#UW8<$nfQDZ78gcGooE$uo-bUav`gF?$;V%~|NXsdd;*;3MH61+mi8;o0R7bn?dIu)o8;0DKT{BMK7nv=3L4@h5f7D{$huuz zw=dpa=Gjg9j(XS)LVASZxW`7BNHX2#VnaHor=0+Hb^KX@NLTr8d17ir*H?i?9!V_! zH||~PxslGwH0L$!p4Oynyts}tG51XyL(cOxnc2t_H}mfVM3r2Y>iQHF{u)Kx7Axkm zPSi|@{~qwVe+^W`u^v{#jth zd~x{uzAx5BCkg1%kk2xip9)RuqBQjA#c)=QQ}rKVc{?skrXcOJ5=4)?S5yId_&kc|q$M)1Kw{7B1DiC0fN;eSroWAdgc)=WV{PhtC7)#P$|bh@ zrMmmatc~Z@7cETxep$|^Y7;sPH#6dvaXhzmF8x_RSvbI^vs0f?m?=L~Qz8ZZv(V{# z2O!mLt;7lJ)qV}=qujPlyQCwM0$Mho{ENL0b3Z66AEQ#7ztlR4@sV@Gy2xd#4y;r^ z5k>&kPWnJca{%e(PEml?v1I50n&+wnHHGb4iFj@~UEzo>kE?O78Mab8Fv!n^oIwG3oVdjk0}*q06p*sd0fW zruT@riP4Y6Vx#)SLmW!@Hm(L+FU2O6j!T*wD&Uhuo_fEH?Ful@#OJfKp z%RMo@_Z?9T<{=;co=(wec^{q*8gfSiiUE0fM4*cI+DRTw-v_JIIWGqVj(*^6{GQ!>j(6S$dQS-6aIEcb*SiP(m=3O=2yVp#$=PS}f?-Ojtl^oxl- z&d*dNLFc_ww;dCbuin@W($+WaiCti?im!v2q2|8_9~I#JEt^BV1I-tsem2#bWwa=Y5;Hn!g1Qc2#Dd-=noOz0f$}5?S(70G;zSqO?_OFH-1?SHD6gIjs^qCt zb$!w*vXyd2u=Mzr7XC$6V~};5Sa`QJ*)wbLl@#L9?rO~i z)G1sQw|Bx-f&azcdq*|3uG_;3HdI7KEFeu3qzFof&=e8rNR=+V6QzZY2!e_NBE2KM zgx*_FRHTL?1OlN*2|Yl75CY_TIp^$q?m73{dmr~dcl^e<|2X1UVXdsS-ts*2na`Y~ zw(KQ~N8-%OQ=h^G{6A4#!FTYvyM#-vztYMx)YGrDVb!N=B9L%OH)sn%d$fEH#@53N zc_gpn{FEUd5l|yWPT++aDki_nc9$9%m09%k zYzI>_OaoDA_nA}UD{N5toVN_p#tCB{=FRaOof+~6qNvzB$Aq(z^%PDlDI8U|wUI&| z`=I;*Anz;jrZ2GQiXG43v*Eq3u`Oo!6W6%|44_Vv$B>)NIf%bwRTcCXrDPC}b&oVofZsl%cuscD=;7&=Bj7lE-gs6Ix6W&h@BD}M+& zVwIAgZZh)1ci9NxmjQd$W-qRIqcuBV9S{_2LZm#~Pus%AjR1ZBG)%wA$1~Hf5xwG} zeC)HQ$+k&GXSVKEs8P!6w#hFYW&5p_UAtxs`5-t7>%9HsI7Rdv-BLb4)L-Mm*k!VAt*yDehgc*f*ys#y+4j9Ld;C=;iKKx4?I3(&mB4Y|ACcdnHRrcn?;-Rlc0JfPt(cMCGSu|(Swx>Fsi`;MKh?0DBK~#By`oN))MX39p9o4J)hVft^uqti1X{4j`ba&OOgSzT~4c>T;B@C{J#_NIMvAP<@f_mbx*orf!*XS!uaBS#3c@QJOZ|XeLn0mdt9=RBV~c7nv6T_S%4z(yN@0b^ILoG zRClj3KN_>?3Q^kEpA8fE6ZM1Vx-yhVwBYTc$@Sa>9WmyPLn>h2s$-WsA>J!K z&0Th{&{Wk z!;MABXtY|YB)1o}kZX}6F1;m^&4u0inZhgl_mug*eZlL+t<&R9Q1#a2O#Hyo>YD$k zt$qNxi%-5?n!RI25GUSUQPVJPSRJD_v`07^icv?DV+^-}Bv!DRzCSU#)`y zQ$O-%FhfOd!$7}!t7zd2_Kvo(ki?!3s%U*fQ|IPs7G*>EYbL9g5YoWDqp=iF~9yZ^dMPQ5gPW26mzj+m`D+{ ze6zbn!nqFs{B!u^MtT)(n5Sx*arGv=)`>=X^|{Suqk>pSvRDJD*WB?*m?y9ZSZ~Wo zMKkH6*VyWkPc@8~T@K*ZW+^Ai$p!vI>`C>WzbYhsnJi@*NtAk)m*5Vp!Cy!bC%F*v zZu46dmGSJ~n-+OWjGSF8wzo<7!9UEjzDo>MZaqHU+^8cL?s)0|NZ);fm=0s?7KpW$TsrNZTqcjr+_%G~QLUI+euQ5m%B%LIoWRmlw`!6xpzr?<){pmc@r_$!X;^Q9^WeN^%#T zwzk|+zX{{@toDU47!TUJEqzg7a7MXs+NB)YZ*51+1hw&t-ir0qa9b3wOnO9O8pl2= zd*at?w~1Cwkzx~R+|IVCT<<$~bS8epV0gaEPUizP1lESl{@heGlUVir9iQumg>M3K zZwjW7LW{l>V=xh0mTIqi6Yg1}`ycvT8$8x$XIC1F2DqUL!`IHSXK<9E zFPLn&-{*`F2pM%TqQu!nUN-h(Y=pPS2Pz-kHrs;k8?)i~Ps^Kt#?B4+WB!X2;lPKX zVPm17s&AXS7~w?O3=px%MsL0&U1bUCo4`#>_s??ClV$`KW7{fyks5r4+O=#eYR?ayWjoM577J`_m8x(HHrK4Cc# zQ+Kq``0CQO8Ke>PL1y>aX#>la*Si2v@tv>Q@!d{OX0oa8QgL&kPJz(%LG1>4R0+;$ zzPG6ZI*eo}J5*Nn0xX5uDK@{9b$*ZGvWgSS~z7uqMtmd3(l8}^r7iA5R z+mrmPCsl>hAFg}jSuq5RcGelT+Nec@8jVjTFc-zfn_M5m^_$G{8BX)k+OUjb zwF@7gW7u}v5;B_(*J$6p_klyr3BnmmJbHg5gJzshdSBw3t8?9|uNSrmvp$ib)+W2W z68-JnLWgO8yfP)4@|gduW|vWc*5Vxw*F@VQM^|pYYT~?siWy?;{i@ziI%f#Hn4HRtA@S7#q19TkLp*Nd`UCbw4%&5E3%Gt#ZWF8oLg?$K9|h1=F5amH|tFDbp8?av(h%#|I1?L}=p;1`(%KD|1vFlK3|aBZVx`Erj~lp>bCX;5)uY{E4Fs6iQxRP}xIbiw_6 zbjIEd&&s|umZ8gckYZH26q)Yt60?+YR5d!aC}KmgXJ?MJ)I8S_Ubf}>L~}>GNk06F zIa0S^-oC%bmjtxWMpWorsZ-RJ|3SpTcOJ zaZs*JfW!|U+jT?iY#gG_JZ>kXoI|;MHVyRvcZxct|Evs)6$MJj=?P`Id?!sDS{Pcp zAs7Pm^A!6T5|3b>M03&I_&_|Y-|`t9Q~dbGfSr)ATJ=pf(16LGoG9Ez-w$~#Jq~P6 zXsIv~ScTyR2P0jgw$JL-5oV`LxOx z?7z4&3o?nBZ?n}SYavWED4*r}vz@3XxDF#o+ak1SnYUol(l<*baZu{kSK`qhdMtsS zKKkQQjTC-R*O=GEgf-fnIF3NWIhl(CQe3(`z^th#-HUo0Co=G<(s8t6xM@+EUNPMM z6v{Wf^-)VbV*1SYO167LHlYk;=`*KN2jCak7R#xAZLHt5IpSHwqe3s zH5IVJ2)8NXYlLl-PwZZuTVgAoCyvh#pXtu^1zOa}8;kg`(ZN|Mf{P#_{Ak~5T})*E z#?Qf=cOHlbqTJ)p4D2}#vZDmrA%7?d(@e*rj?u~-T=i6@oD1r1fw@SG&+0G)xpDca zJ+2wjVM3ktxF-TvO&Q+Z(=YIZ+$w5)-*TiwdU-txeqY*a%cq*F#wy4zikA}_$3oE4 zyvdUv)3(d#+T0)B#(_|Y8o$J|F807kB;3>4E%CwT5_e#m!WZ5C+wNYPM&8>&c~$x~ zFlpN^+8Stqi2+*3@x0*GtEt%Zfay=kDN!^{%#N}|Rbf*?fUNUE=ap3TN*xovD2~`& zQS+juhHaQ1*06$}yiC0?gkSYXHM%3r|WiNPkm1SN3`t@gN9Zq*- z%$DmXHYQd1kt9^x+eo1}P-hB(`}XbQK)b9^*nLf(pwbt-ww)_ZxYj-x0Z&@5v7ld@}fnnokpx$ zYC-0ItdO3mLxp3(&T*ttU^3@c-)F5xzdUHh5{%M<|XdF?(`f$R34$VmSHvr5rb&L#N-{ z9WZ;h(6&9&Q7{O8<2iM~>i-?|T27woSiCC5|IOgB6m|LV?}>8LjnO z4UMO4)VPo?wfTlyosltUO69e1> zyhY?GNxa+7hGAMq4~$W>Ku?;2QmV{;XP$W%GFnbj$?)>qmBtL#b>;_trjRH1(XZmu z)VMs>=XssCSKrH&`)1qD?d}o09&+{DzWd{PbVS9GFkgQi)$oS163X7XifyCpZbN6( zf$`X&b(+l~-&a(H4|2|lhlic-IC5UpWh==+M6Ovi;jF6y{)-E>U2Asoo%rk2tT;*6 z#~L}RH;7&1s`$8)3fjkEUBc$&VLGW}miez!Z>!x$rV7+kOSeu#i`G1u=nDsBh}yFV zidyG~2`qX*pBd4?lcU`$^ez$k_~LEsM}H1AMHK-CyT(>SeEF+nwcf&OyBg$bp<%bQ%n7S?cR$uwW zcJEZ3LUqb*QwC316v2JvH%RfMvnXf-#+%@p+{u;Nfmn>~+F=}To{T{K5 zeR{7_&_P|K-8Tf^u(23-Y$Vm^=egdKGhh2&Ik4J$RWq|KaRc%JE=IR9idvs`s-*TxFX z54*&%@zi+Lt7wHUIy~BELGG5K9M7`KrEYgIVb{jSj8(LA&8Qdx`cwDIvi- zq4D=T7B284!LeyhM^$+5BQDw;-FY=C))G0;?-){UYb9Hdu(&yxKlf6FZbKbQXE&Z8 zS&NpSZgiM@b@^qtTO!qRR%<>g=YGI06}>pkOzbH3(V9f9$=xvK!)#}Be~R7KjcEo6 z7q5U-chqRDa~omYkKy(%%<}bha(UKI&-JEY;NcLXU4Z&-+~4uI;92Y~O|D1MvHEaN zd<@(fy)njPQuNlwx@4&0X-e!?CZ^mv)t6uXKq%*_DL(Vf+31hqj1b-fUJ5XGRH$>E zcaU4UHhU>fCSYSIX(YNvwPppl0L6KleXnwaSmewP6e4auh%@15yLR%yg2e<+syc4W zTOPM}P41*Zz%J2jH?&ELXxTBiytS*EBx+jq-sGg*h1-l=q@r)3`BUJ`XsyeNh6=Riu9GbW&3Hh5`eXU7e;j2oJ(9$QrI$j@ zHh2?^8?~rt#i3mES~-886cJYsuGL!b?GD=nhkBmI4b3~ksTcT+N1l*_GS!|}5%wQy19(+vL* z(K~!(do1?Q_a%xye*_b6&WdjB>4{0MH%Ou?U|CU}jv6NCuw4q*4&dzTDa|uGjT~}6 z-@c*v!0a~wXa5lKQhx@y|#xY0#(!4)>J$e5EO~bk#r^FA(2bHUzeIsq_lR1X8qDA zi+nGGcIK4(FJa(UiO=WT<^Dtz>i=hy{A0JS0b^xCxbyGslfV15-#c&{Fl%b5|H2w~ zaP@mY@0O;mMfopHl0#r;9;N)3XZ_E2^=F@0L0A+*Jf9FF{Mo(z&Lfd|2p$-#XWZ+*_V+TYN#4IV2>w2je)lL#CKzH75}kj{ z6#m_luy}M!{t3C^_ipK|1}KF+WU{*UTtz|HRAx7x!h=DD5;C;=i4 ztOHZ6;T^2y7#_W{hro&^WAF_Ayfrc}X_G z^A;my)ElZ!-ly_1vCYCG5xW{2coFLrL%{ac$TNoPm$^=-{nE+=mVtojYfp!dA=Q#Z zS^KTKqk!$$cffv|`H`o&m2jXM`sM=XHJU1Wj(7Beuh!glJb&Dbx95^9Hmpu%9}wGq_Y^&9{i4B`6IpAfh@_A z$cV%~Qkwr+-}QK5-?|Obsx0uuy5}l{n+sJZx1WPVw))wWgL;EvmSSI>c7YA|0dK{r zapq#4c6l}sA!~KQe}B)m9+LNBzmlrXzJ^pOG8O?1rXl3XFfB6%J^K{Jsi_FuKBn`| zFfNsDEsvX2IgPM~UgWD{Hyx!*yvOqCs57=bdV*vT^2_run(JU}B?>um(MYa7KNuHS zv{hTAT%8oT0q(U=ND$&5Jkn!+Uc!GTFUO}UhfQC>hyZhIi|kvO46!8d@09`n0YIzu z%Cw}l|GZp{VQ*>L)c^yJ9ayrQxFUgix>E<*7eO{3$p_>*jSJ)M$4&UTCh8Ozn{Jib z6Py-JSp5C+)SG-V?aw)Qk8SqHDQ*(GcC@*4FT(Yi3my3b$dcK0-rwF{!PwysmVw%A z+2^W!fCse9Ghsvqyf%K9Er;P+mz-%KtX`g(9H&N23G5tfdug~_xTKg?F_y~`BD71w zAP`5WAI^ zZi}dnGv+=!8@6p9Ak{Y5cQSMYUGiSF4aXTt{qxB#7^B~TpIhcl&o>o z)PYFF>R97#A3x85gOXAaX}T-jJE zzZ*DQ_Zb*hXX09!JD>Ge%2eLEAF8~aSI?7v@gyMRjMQkDzK}?UkeOA{XQ_{jHd|SI~5t@ z;w^Jd2KAr(7r;sDJCFE9ehnTI5D#0+%^%9wQNqiJ17&^sv{_ddhO0QZSaTu4sTB4LuwrZjk_fvuo$&o=3 z#Wk(pi?Id1w(n7OZY~cfn+gUN*_(?95y!q){%owGF!90Uya~d+Q7C- zhs-$d{sIxLTqAx}C2)W_VWAx@_{tE>(x(o>4kO#xRDTQ5Bkt}^RKAQJsrWmG32WUR z_sCd4JG=5)1?`45H}zZGaJl@Lkm+!7)Aqds-Y+J|rgDLbmIG}|uE(p|4wPlVONu6E z2mHymS`hpCR*CA>#}GoN{d5zBd-Y8M+komrBvkp8p6sOlONiRas0PAVGIcZgJ#Z#! z*B$p^WA@*84;cT$s$-U)TT1UJs?-@~&nMTbuwb$2~0i}3T0 zO45uJi@(Cg{Q;q9x4bh!(qMLfE#!xQ_H5;RQ^mXlT&rlgJYCi||KrH(N4pt>Je@>T z*db7@TE9=)+M+rReh_kgsn*4jBN#qX)fUR0G*+#~OfV*lZ{)NDE3lTo)9rHZ*1WKx z?XZy`oMl?JDhl)~s!LQ{Xz3IaPZh!m zU;Y!bg@1-JzkP~~Y^Kv&+eONSI^5La3!*(8HZM2aIN-cxL`jgM-x=E7<{| zeBIc?zN9y_KiK`oc#J#qzEx+yYC~l;}nk4MW|2S4fF~A zT|rsNJ+KzWZP~DKtLSD5Z}vOmtT~0b3j^&uQ!ixk;2rr<;?_3JjaH$<#B|XWo-N ziA&G!zhC11O@cA4ny4%aZbr4(l0&nr0g!&&D`mn2th>ta8)HyRel&wP>b2+k^m~-b zS5Kk3<=I4Eo@K?|z~4==e&@jTy8+xqKWlYvu57N<3-gSiwPuR6%dHA|JGzbPr@>mI2?gNNyM)ZJLTxEbR&RI3fnfkL%{^MRq z2P3cLy(((uT*Cr)G7x!(7#(L3h7m_c~5N)3))tC%4T?E{QXvOqIh*>|l5UI&UL)e~bzy{D(Yzi(+e z1H|?Vlv(t-gOA&}b#4j$SQ<}aDO06!H;LxQ7O6mZuR`3F`K`MhS~pW`jE}=)0p+v1 z;H7_uC0E0u&|v!t@bg=Zho|pfnccA~@az)m10AKQ#O_zYc?G%!RuFE{{v_D4O%HH0 zB{^pk5sS}kBfMj4MOqjoP+{kQ^(#={#1_?O0KI9O4xB)Thq3# zlq!qeB2mIEe~w!#^wp~fDO@9#Hez5E7nTfJwgby5ot;~rAz*^0K<}99LrMKf^}^Iq zAty231rJ)S_73z-jtv1l&TE>LQtGP`)p-lc)Q7wY8PRQ(wgZb#ZUNOp=bld2e81!R zss*o)+l}qnX>Z7MHjw&B0*_tt-`PJwj|jf<>p6n2)GpF$gK+mf z1-{jk$B=Bd=Hk-^dI*VU&UeJ}#exvc5^M8aI!Q%3M74imDLp~Q3Uz4TPxG)p*usQQ z;}8uxpJ@iK%J0#qVP}>*)!~YC2mY>CMGkN9@-^M0e}O{krgb0sRE2c?<;|3n%J9)A z{6V||q_mW8Zd1}W51w^#FlcJa!SF0~m14?LpZhcHD} z5k}cmx6xuu)Scq+aS#GG-wjOaeTd6JCYAPA*fr=pM#c-|z^@j!px?y5`FY2Y{z2Hv zj=oC0&HHOrWh5!ui)VhEzc!S}FLPr}ruF;>+h>hu_DSs)`N-q;dAvIIWF$0zGUcpC zi7G`mmls)nK-gi3MZl`7Y+$Q$9T0%(9W7UL2SYo2sNX|WxJN_m#z@+-+X3BTnNwfG zPP;SDN^MmMXaC&o94@z<$=BYZI=mBgXgMlc92gX$Ap3Pc-6i?i#@r(qT@Jbu!usVoh%G#dwCj<{5bM~^OQ9@|$|$5g1#$rx zY?PjrkKMKUx+xws^_F~x@nljAw=S>s?Fnzt;YXp0dhNyB2!Pd`^LzEx~%wF$9;)3ivB?$%P#(ljg=XvbJ`U{FUlwCE*m0e|4lXKH@6g*&yDMlH1sC1xC`KqIjzDAYy*x?enW8?6 z_)%$JQBZNIsfH0wUtDy6OabD;jsrxT?T(Q@_N%YW=3GbK!Gme}?0tHq3L5Kd*>Mb0 zSY1{FV{N1ncU`(U(zS!O0}OKNMN}+sy-)(Q1yBI4^?I95 zu>pZCJHr^YA7YGdAEN}#<``eXFbW70S>yo|A7IiEZ!R8`+kepV(m&g^o<&4Q2|J=M zJH*TVGTIe%sb(n<`?r9Q*`McN+gt(8^P6@9FSzXGesZLGezW#NNcmnnTIYLll|k&; z&!F)O?@dj7ST%FhU^(_@3B4iA?yXbyGO~@~(pzz^D}8bYfRHS(tRGt#!&t*Sow4q% z$i()UtfD3~WBAH*Cs|o6XdZjB!X?!q2ix-}@97Fdhs{u3Q+Ybe<)s$l=ff7yqBIG? z6(VxaZUwsW_o!Y{o!c#yX9fhaMv)2AWN|mHiH#c7jj>B!(q70m|B*4$z#MtC2mEV2AT15ddQ;ryZb#7Ngs1pbJGL#9YLI@>e12S0Vuy)~aT%mrTZ&TPYO7 z#_6b^Aa;L-f+)K&LOyG9sVkeJl;GAKRGKj&s9gWF_gein(D|q%dPTRL5-#>bK=p? zNq8-i!fHA~4}pTovj3$vKZzm~>(xUj zAzL$%a?m3}-0k3_baMXu<+am{C;4cS>gkkf57Rxf8~5I^KiP_llqt*s0DP zBy5@MxXR2nprtE7X`dh?SOd#R9kY@LZ{qAhAIx-nbcd5-%P@d~D(M#7UX0PQAxPa) zaxB-bU7J|R*DB%x?%`GVI>crS4{?!?;F8O_&KFD*E5+6}Z*Q0IHI)?DgfV|hpmR(Am5wj!NvVzs zN*7&W>p1tAFkj(Lz;JY_0R5?p$(ur%q;K}cuaC|1wO$14t!^xE_EMzjC#Ho`K^G+f zGVZv@{UN?6`Ao%3SE8=5(5Lyu zNxs6RlgmPHh&n?x=9cUjbxZR>zBZgIG?wQ*$Ysj;l2N!8nq^n(+9x<5T@*>9?6%w5 z`Qc1O`4<)!#G(>}J84zRwAe5&vcXDRYtEN$WpKdtq33KZ*1hiA@=|NmDdGTQSG=2e&90NXlo!H#O*JL4`6l2T8`6PeggC^==`Y_fshC#TEEw@g{;3P{kWcx z;(Z-Kl-LkbqZfTT@f@B|9F5;Be^y?LW%6ITbLIBaCnW>R&X{_)5Bq_~+74`L6tVg$ zCY1BNwg>ra_|C@r^i)4P$lW>XL8{~&5!UTQWo(Fi*Ioj-Lb(`c&b{Y%Wp>nu^c^~2ZLpDQXEQyj!l z%j}O4Fr2YLrG!b7PbBJyC*{ip%H;T9AGbZDdYFQR4ak^L zZ`wyMbHo1@Ybg~(K}V{(f{{Wz@bu+4LAYpBx5O0|Apfq>>1zCHen;QYzq&SDaHGg{ zYi=xX%>2ih3+PAFCG*!m5=>b&9gSMXfjE!zZg(XQ~5{O zej>$%=HF(^fBN4bU=Pg{84~~RH}fC*e1Xt6E zH`7RFlSHffzc^B;ZUUO!kuO%df9daKoDXcwuNKp{zWqyc^LiQ}EB(KF^@q_{X%ftT zv4j#3N5!LGZ_L{uJVaLz%Vjz~QtfcDoVrx-U@>o|<9s^cpKbKsx?ABN4G=*`tCrA~ z;YENWtFV1jOz=M%o)6DZ&18A|Y=0KeD71N>tE+rX$!CA&1aS#}E)oJ$^{kqj z>MOa6x$bOn_1K?pZZe`r8q!^;6RV-aKZ3fI9-u2=wXDhFM-18f6sZI!$I1I2W*9`2 z+FlTEnAf>dRiHA<6E&^7zvtf?$#i~g7RfL7Ce!7KpM0cYSU;}hirk@q%YY;q&U!iY zAJ6{mfs~mpF zeoLd$V$G(QIr(S~ko@FOo)mKgDz>roS!~Dg8Rf|`9Xy~Le(S2lRU4fi+kwtXE+BQ0 zjCh7(@?6C)09z-n`*oYiQ;ZOGjRbjK2#ArZ}{}8e#1Z&bS8#Mr<^Ay(l0Zo`rq{K&%Px! zGUVWOnIqaAR6SB%Buu|GI);krfPVOXnyaEfY?$%G)!}mQOXh#9ll`|r%5vqThwOAO zRq`Td6?9nnA{_H{ty0QC8|m$f4`f&>GAT8D`vQAPr?@iRMN$f$D{0leTXGyp+S@H~ z7>CKgSgE;nPuQ>Sj&KTC&Xz^HkM!>4px6lYpYWyrJQbky2E$FV@w>G-+iH^%=A5_+ zJKn`f$XfB4SXS$aLY-14JT=6-OHy2~IV^=o`*(y$hrM=2>KeI4Yu(Xat>4AzaBAm1 ziRvf^3ElaJbx5NVs4HcRzs+Nolp`w$oKu^v7sOsEQ5f#(u{-I3*4!2usd*%sr$M}j zGvBk!#Q4I_iPK0682mY}^SmL8aJmaQ(6GY2k(_z|O0i*CoXtz?oB>XgZ&w9Ow&{G>g9v-YaB7Q32{=)uzkTf#LKdk<_$TOsSjZO+$T zJPMB3Kz~M@v3z4n#p+2gwHs0#ek)N$rpxTrRc86r_B~?qd=tg#8^h_~^_Dk5T4lD| z;AS*;Ol+ZE`F5vTI~!*wU#7XdKR4j0MT2HN#mUd-6o^m7gnxS)|Gwz_69EmdIq?Ok zQ4RDa=VN<^j2ieN?-+>_OAU16D{S(#Zt6VdN<4{dtDGO7Z#=yG1pA)b-U_DkuJ6Xo zVl1TXv(RGZvuMRg5pQ9bu?GD^-qtJ8!*}q4L^krS`6Yx@4I%h0`X$LUiHNX zNbeS-7?%NI#U=GxT0f9~vrbW0zsE(#m%!2*^#kU-zF|ucXWuke32T)HO;5%`VORd~ zTR(F>)hu6}*{>pp4hU;zDc+!77zi7vQ!ve17~tY~*)nnYw-xoDH}^k2{k(Y=)l&;2 z(3`1n-3q5N829#V$#fPil?rps)9jE}lR7`O*%5coc}H>K7I=diunp-Z>`IMZ`;HNr zHb=Fi+2mW>_q_6@pX7uEEHr%kQIqeoCwbtCJj)VT(^^#9kNqR4olra{>a+dIVWg== z60*}?Vl+DY+@G6!G|++bOd$shfrBd*ny?v}>miVzpq zks^$@#k=ACfWVMU*Sts#uojW6ajFvzTx8KO^3up-tJ|l|t$BuD-z&JsDW`aW$Z;d< zW00bSE$zLV(YZX2qN1|CH(Ba_Z=wk3Ewt*?p(;ioK>dp0XqiTD?31Yu}+}nEA3sd*B$& z`KTe+;``H?sjelzrm^s^U%n&AMX^TpgojNf4H`v;SK8<=Shz_8!N5WTLaK|+1N@Qt zFkJEcrh_0eYnF>DwCINa{%yb;jIii%pHX<~-&#)dKVH?pzs-m96fNv()x39Qr%YTu z+VqXB{6g~KZ(4DI@OPXO;Q&x=(}*3fzjUU9jp!UKDqzxNTAoz5>%C0j>8unmT3hIA zaE;K07;%54BgJbL#;MI>SZ39vC4?3eWje*+z7j*3-B*DzL|Q`YFXHquUKwK57_Nlz zWb1B^85mx{caPr2*sjTH>*}kRo{1%mZ1ux!4CftE$FTJXPp`3-GPzFNdbvf_H~pq( z5PNT*Omv7wiE+t)tm<|i==dA>SIE-e2>B|J=tT&IRznxJwXd$8l9kZft?49N_>)Z|Gx7_f`pUY^S23~LX zuz6#qv$*&B2Jb&j-)#^=l5{@G__ZZVCfdB*gtHf~fSpHO--Q11E{tmtIdMH9-b(OYzN0w$V8hM-$ zeM2d>!vMG8dMe94XP5XHa2h4Ebh<{8%o}bwjFqWKZRaB9h?@*v^|q^|O%xg^4#SsJ zq{bz&*#X-xU2fB1?_9m>vii|Jt3Xl_~21Py)Kh3~Zm$NKG zts>izG_UO*&lY!gy-}HY{YrF*()K!z1ukPlphNckNvb(B@Hk8Y`y88 z@-;7Q3+moHPLIlFddDCX5o_$D3UrX3+x%&q-~5g%ZJ@idHn`i(cWBzStTU)Q)jv1P za{zE+5`%vf2ca;hVu=!S^f*z6HwlgF!?E3Z;&CO8Nsj5$q zrNv;OE-Ns^m6^8!JIkmWpJxc;3T}J!ar;&ar-|-6%G?c0Nj*G*xsQy*MZ7!RGz%^L z@f!^xvLlF_rmU>%vzNxd2OD{he&e!ZIzKm77Ek8^8<{NQcSb8;x#Jjo74odlvfwdc zYdMEG_D=T6Z6yAp9_h34;3SGrU$=>`>NEY0>fWm)K+%lzKFb6AUpFKQ2JE-@GVVyX zF)Qh|$Gc#cS_)y*NR)7@%=mm;>-o**z>zd2?u7E1MrBS8kD_{^G^6yPEcGhP-gcg5 z>};)rWubki*}c(ljAwm?Eg(}9Zo1BSj(bUwqO|yFq+EA*8T6{nV7E;Y8z+Sm-a{i2s{TH%o);%I{M}Ip520AyhsvK>5%Tk=iXXB4%`07lq>A14fE`i z5E`i@9bZQ;LWvSz_xB#H578pO>$V_VP9-lM-jQ93%JXk;p-yan7AsR%{7KMaj1<_k zu3QkyqlE0KWy)a!RiW5pR?6@o)jw~ia=L#Q_{j!5tM9AzO{Tw($|jl2i3|nx_>xz& zWEaqanJMSFGgCzn@+*Q47U1U@_dlZ5SP}P$QZ}i(r(0YHS}sJSm{*RtyTrdi1{!-B zJ0G#EUAR-E+xo@OABp?c7GdHSPA874Rnsqd?BxRyZsV2Z){q`2ls^U`V}MS+|H_BE zW!NPo4il zI6cU@<+&kIteU3@3U=X&8iahRicII2s+-=v|FNbim*PTzp$kZ&OUB-8~juN2z~ z?P5u14o+{rF&U0h_An_PHUKtdIP$iVRq+nCZqy!%aw&&q{3Lem(=~7YJz06)>_*%3 z)V;MEGaW^=8rAJT_yZiFUD zxM_5!B3OWlh|nh@!3gGUw4XU{&F?xS;zX+8A}v>IfB4?$ZTYobNcA@PIWb^kKE}~v ze~B})yH);vE%ld^zb&@^MSl2qWGmvuk&2JC!qjiQh3ngEftFfoa0g=UY<&VTOTA7q zjNCq#f_=!lSgx`;b<2B8u_=jFWBAC*CheC?@DY~YSpV1}{O(4M$xP8n%Ns4)yDN`d zkouf@l-1}8@r~wK1A{38hR9)Ns>c0zZXJV3XH-X?j`+Ryi@Z0Fhh6 z=xy3{tXq=yxvzNXbZp0_jt-~*I|j-ww?akpXVwX+$6llr&zCg}$XXI`Y&CbL$WprB2c;6c#p z(b<3?Z?5QA!92~-ZkN=Ji|6EnRMz7W_6w9`RiwX9Mfa9Xd#Y_TWDhpPSlRPDPDR`G(E-dcei?8?) zg-VTcRs=b{2B!?||F6CGj%xzVzK6$(sDO%yA_9&SDZ)sVZbOl(^d6B;B-GGC8AL&m zCcOwq5u}C|LV$pXfb=Fr2vs^EfCLC3l=sPZXJ+@eyR)wIdH-DhiXnvLsrR0H?m1Sb z)P*rL)1AG`v8tzIyT=7d%JxWQ64~Wl?&BhLMc?atwVd+KH?@)ph+=ORen@$-(6$G* zV6bXrA=AHHGA-y0E=^LM&F1_udR71eImw3qk(zwv`NeVcW>4!d?dD;HY^I$S*17MV z7-7w}O2Mr*Y`*9;U=!lJ>JsYaSy))6&MLln!FMhaU%{HTmnKK8%i$4nx&8Hm}8eWadZ) z^(3GWs6NvHnOf_4CJP-A9vLGFHFh-1r6~B`nYm)XeL<%75-=QfCCp+n6bnMnN)I`Z z-tTg%G~8VqR2UA9J_6;*sZ)_5olP)2iA+^<@njZrv%P^GZozQXeTJRlIw)X6m1k)) zyy(Rs2?g4emwFvOp4GPU$wevPOxm93ie>c&l8p)X)|=Pf3Ql)T>(9LEJ1r#isPvNi zc`D9?Vlk+K(A`Q}p{7hbIuD#5)4ip@A$y~#qbjB+VT0@f4OeO?3Y~D1JhtVLQA=M` z2i^6vdvAvKiXQVMC--DD+xHz^7~7L>J8RNn&5gp{ix#2pWaH+P!`;k_6N07Sb|>rG zl-S@LMfD_B4vj>nzAw1g;ZcztLej8xqtEX%6hOHAW@4VB0@^mL7~?+DlrEN#b?LX^ygyB z_~X4oNN-uwqSX>bTdILA?PJZ5Lkx(!F_(Fno5I}kf^1q|8xEDj?Bq`lVQ24l+DV%D zC5;X~JkwNlkN`)l+v{3Qtz7Ih09EA}Ds=3M;?6Hx3v%6`j?PtcELX8(JkmZUaM~_r z$9A5V--^WAW1*(Y1wbcqWmYXh>v_5r0$4n0v{t0mabsgMPAzMYT`s z1PS{S!!V;8!ysG7yC&L?A{D>AtNG&pZsf{?uvUrpY^>}$WM$cc(Y!fZ!7XC)4xAEXQu^Y40JuQ{? z*isU`Ma|c7 zJC=fP7SH#M4^5MYl*?3AjGfuJLS(C&6@Q`^p_2alveSgkdl{DytE7(V zbAM}!AJDiz3>T7zM2z|tM5y-?jT}t!D{Shv10#e*Ov*p5pYWQVp2dd0o!AI&@7plC zIHF(Hj3RS}DscOEBngs-iMtJGJuS{|5xWDM8`56?*?5I^ZtcJL=EgVk=p17}PLZiO zjpVmox2|5m_`bPdq&W}TR#LxGKHP37NiB4rUWL_@E^D9d)b9hnXCig6{l*pGoRWe) z)YbBk(Z~@}Oh<>jw7)SeM$D_PO zLpA+F73Mk1A=@ZH{mAxnqSu;-jv0zt#~?beWd~*u9k}OuVK}-oMbF~(>XCLKOuhU0 zK=BW-xPFIqV4?w!XDq@#m?#Wy6T6n5TfGykI0q_HzO=D%D{gLc5mnv#dG^$Vv zb?MauizWm2JgAS1O7 zErMnRNPaxN!=FykWAA?SZ1_^c*qw!1qRNet;5*T4etMZ+^>eAsf{ ze8rV?K_mL=>$bvm!8RSQTbBL&Wtm-e=z#s(`O=4}!!SaT-~`5%=iZ9kHK-oWJY`z0 zUi!6_X{FOQl}@e!IrYk6{bOAzq@+V#@|;GOzg{|!;Yu2wWy>Nr>7o ziKHPNPBeb^RxnYCurpc0d>i77dUvUov$dPLoNxDK45gHe6l6$+91|&AOyj@hjq5A% zNdZy<5DTZBBl~d?4=S5-UXoW@eEq)Faxwnxa^~8%KYix>IQz_bGyf~v*-!VJY$B57 zlb2mBLno8J0tdjM27~2^8H?;~)XMfQnJPPgQ|MBeEZPzGS;|-P>XZ-1G{POcQ^^Q; zY?Hc?jPAfXA_+t4XjV;(uPB(=yI)yoq?@acu8@29WRY#_H{qy14PwC$_;uflx$V|a zxb3PjdZM_H9$lZ)EM?T{paG&o4TAN#Xe=;hOAw-+kz~eiUy}HJy%Sf=;yu}&iPap6 z@F60%(LTI>+l);b(jv;9+S0c~I=&h`v7)!GkUQi2In&+~#(C(#x=%XCjjc@KG#Ab^ z3P_DRjymI;!4Wr>YnI9j7s!QR^Rm>Vn?@b5ocG2t;T(z+lHr7qF=SZ#5=821j)r0I z;SX)_P9ey7pj)6|B4Sd>)t9YFF2PY}jPvzkO)y1r{ol-xW~G2K>XqSa;ZFjdjEdpn$NO4eOKBZTUXG;-@r!Y)cr^l^wO17 zSd<)}hnL=6`BYKHtDNT3$U}4pWYE%b0u&|Es zfp;P!ZlZA7V&=7fdx8~~2C(>isZM6O8PWfW;qO`QpFFyL^u$?7idLNZt;tjnquIjr z<^9vgPI&F6%gn%e5@3+jpk6!YW44WpVzZquO(nb+A+FOEqC9V{2Ya14Uw`oS40@5t z$Qiqhaq-569N>M*`EJIr!*?wGRtCAn7*BXW#2cqQg&pfmY9w~c(`AkfnCiRV#ju`j zKRM-B0EPIsMU0uMf;_^b%2M>%KDxc_WE3!sn$X^h!Er#x7je%XM$^L`I_$^zf$*`@ z$P`>U)9cy?eNuth(ok2KmNz5m-p0`WiD?;;IAqv$(2&vm@#Ck9q=;LR+ULX(8wAeR zGKh_WwcV8l&s;LGuNip$e5rh}yGha{Y=T}}vsC*boo(IyVYnAxd%S?5R44rS&Y~E4 zoU)nb7X`Al!XDsKO_Lc8Y5wH)w;JA00R;XNi{D}vL3;0}S*N)0>(~y?)keAV1iDY@ z&XTJ!*<`TVqi&WLNkHora+let#qs!yM2~eaX#px-?4IP~7}6}#XE=_-^ZoW4vJZYk z)NfHrYiB^ftoaepJp1RQ@1HhOq{*T3A&jJBx+|j7Xx)c~Mm=t@Ci(GaEuB+9qqV;c z>%H7b?ej_6h}4I~e44ZMD%rb_d8puPtQb&D{d85d0(!S+M@p-0`rG`R$hlv$kyHzTTW-w+8o_hzi_*yRC6aVz|1@m;6M#@w`FlU@M;M=JuOT}+R z$o;&&fQ#qAB2+E2mPLlSlb+zrdy{jnl|p|~2#+T;vz!(*(os1r0t@p1IrkxMx25$* zCX%BF@<^1hsSLfSX8h+o!#*pfm%j~cVu38o#% zWFExwX*~+(BunQ2*>URdEasAW(q|W{16rkP$<%UdS+QkHLUk(m0i-RHfg`5e4k>4m zn`TY=#uq5ZWb^&fF)$+p-a(AOI|yJAd9z0JrZiKmpQ@mU^s$G~etiiyT&)zzW$@=` z+>oLZf}e*=%^ZTiMp8Z#Y1rSv@RSSuJ~!%MTFG311N>Tkj3A)_u$=HsihP~#Hp;sW zWl4W8ZAsDEh^X*;Q|92Pak)`4zOuUl8)99a_31>^ zc#rJ~wrpC}?VNb4*0_2<;z?jC&3t_#lpSm$e5zFYp35r>$^)~VAk&$w6YV0=*P zC!UqnTba*<^q1O`^rYc4p=|?SUb^b<33{8Rs^`~s?b)Q1Nz;(&IG!?8DtnFF&!bCk z#mvuqz2Y?O_^5_Sed~Z9WnCYXPF_kBAAWZ8Nuc_()lc6|#vM?uuUgO3!aQ*! z?gqHd4WS4rR%d=stp2nV5)i}nrry;7G_OWFnW*Q#Mr`zTV=mk#7d+v>dEUeBEmBC% zhZBJAyI`Wo9^H`>3*c^wtsdAdA`hsCy;-MFvL+^7F6)X zjej!EAU5WA@{JZ;mt)TOv6-ZcmG zUFZmJKKgN)9xmx?P@4K{ArW~aNUupPvdW_O3D3aJfE9MkQPPe(Y-A{2|O9ZebycT z#h9cidInT&WCXkP&&Pnn&2n?-D+hc<;>vxGb2~(Ilr-rr-_KD&6GLH9CbwocSfDH( zzf%sl{R0;I^Q)eryAnB@6n8D(tnFo?*RN_6>Hp%?&6+Vmn(AE(_&~QX5O-Bk%_&=a zyi6p^2V-1dKfQ6s_e1$CHBH%sr7K5mNBfSC)SX?48!ilGiuhIj`JTd;@}VTN$qoC? zQen$ak)sYNEZ;5cj^8=$#E@DY+(|gN?ZR5PqiJ6^+P^HfB;@~c4T^7dOe2(NAE?3_ z%U@AP`2NhNky4={y*WGfHZQlf(0um%TkJq(fE;NeM?-fGqF#xp$c6c;raw7lwVG_a zvB<)#7P&GkZKeOY{#R3Xgjm-zQ2z_$T`X{D-o0$6xG9LHoc<;46Gxob7w?Df9h!BE zPIqzg9ur|j$$D*t%5_uZd+euq8c1#V5%Q12w2($tgO5X|e8u2FsM;pEg!{7{Fs;Ue zn?Uo2M|BU+&DL9C+=p0>3F1AN8lzq(3Gtq~F)#i&OQPTm<@_%b`sY^d&(Df8*y}{A zS6@=fe-G`uzsv3lSsx2Np6y+u5Lyb@7digc(Yf!mQ5j`Rm=kAC39&W#ZxYS4s_s|j zhKiKJnl+?>?sK06r1ONG9#zA}ly(2aTXGKywd3|K_b15h-mH5>>#3pO>3mjYy?TVpuPl(LOhp#ZfEsogQnFfu5pG*>}_8H7fbQ(Hxb>^ ztkV*+eUb5XXD(cDcw&0KP#C!CM6G47Vn!;GBKLp=1vl_da#2kb9hh;QxpZ!z5aM`& zIZg_BJLd95ntw3UH>%%U_eE&9jJ1J5!W=u%LdtPe=)ufTS>g$w#NMTT6H_{Q_r(PG z(iLO0ili745NhrX8pB?&RJh6qS^aIT!A=L1&mf*;Yl=suH`F={&jZk~VEMe!&axf9 zX9AK8$fJ2opf9-YpDr69l&5X>x<$XA3|TbKl>W1g-6 zrm*aHT;R{c9Bz+&)mmORXXekzFu9>fPfc#UiD*d^%yEMa8I0~&17n>lr^p-nV=ZJy z50JD9`-*zJnN9dYfEa;Mc=lWi~B8Lp*#M z^QsE#9lOrhE4O%2HovN|(zMtq*35P>^qEc;CaB#j zof5Qep{8NcJ003sfu$-zQm+`twTq5V4`eLY8bu?PA_wz#OY*SD;c^K*DXii`S`SaK zy+_rknM7-xE-n%Xd<6d3lF|6* z5l0RN@f!|;P;Ux7&!?TE9u3-~+r|S9m{uFKu^BD*RHNrm-jX|2bHgQ>^!V7-;YOLi z1;J9UuT%B$tywMFR_HdS`*qtgjoTjXA$Zh6GpD>s9jt%|ODQ6$y!!C0AaBaieIAzT zEfIlD%4R~Tr${z<5BBxYP1Le)<$8@57h?GbXgK3rfRmR#nzo0g`PF?>p>_`3@?d*3 zhC)Q#R=VIME1XIBQn-PvovR4@{t?ZEIi~aCvb8R6xZ_VL-Cn&Th73$1qPoSj`W2gw zpac1w?2RhKl47)7WbBUC^QAZCk-pQXvywbyHm=8xQS z5cmvLwHh7s+H&TP;X`kW-1Ehet`W8<7k~A#%Yo`j>k2b{OY6%Y z?6e;mCd2B`TG9kU)X&#}*P^rEzwuz%UH?7o(}Q%) zRj+{vjZwfNDSl7O_JoOG!}4fZb6)CbCp&&OR{zlj)-(pT<9*$fO-}z905==%TW$zu zYBSTum^3^Z3uU)E)QXqetHGHoiZ$$kGTvq$i>%F3+^Pt{VIeD;&Rb^3!YS zXdkJsu6CoPVaA)oq%HA>7g7}=rAJcPNj+askZQvFUlUr~(HNW19wv%keddAH}5 zHFKJwsV!IB6lVLFzDoGxI@4!7+Eo>oXRBHqD6^@K)rAOS`qr4MJr--5W7-+W%?dvU z<3{oc^e&q>E*=?o8s`|{i%8VU5Gg!TxNmy6evn6KN(ENZyq9z?(AwPt?cq-fC2$NK z7|~5H7^7N}RV|)$l68b^u!{~w%gt)MJ+K?Rq_bmZPNr(s&UFtQ)q}o~i={wnO)f@_ zPKlW&A1*RZ(6sj7+C^~puJz^383l_VueQaDeK~4by-_!0RajMo%(p`)u|KV=N z)IP9sMXgeCJQZf67PUPO>z(dYfeluOQJ)QfP1CW8jw0)3%vZ-4;Z?7e92nN8<&8~yrvx{f%E z(uHeWoW83sfSJGn+yz;;Vv{GriY1x-_0{gnOO+Uo^J!KLC63}3ms!<|w_AACxz&@z z2PUAql;)#81UbL6ie_xU2hG-L6lK@A=FD4ML`*XSlT9&T-`}A-!v6Nfpwo+%*3TTiJBh=XWW&I zNb;%=n6DvqGpi^H3lgNAQ;OFr_mzD2PUpWluC-!PkT=`V&S$B(cVE7aqybBJt&6s@ zDLvh7-`#@xdF_REj#X1gzgF+lhD%4MBBni>P8S&0S;Uvgw;#-B9umz`(fya7F!qrUrCU~57(*|;|tk5r&)VD58nlYrZ$!!6>eLiU}_f?W|hT+vCG-5+fJknjl)mVIaAuTCQ%(WD$-8htb z)#5M|Ry*UdXb_6OI@qrLZq$gE^zm`q8TIMJ8O>Hd(gW*i^6ez@fm20_50ifN+gh8) zUTFG0f0av#-gbSOb&KJZd9=t5BhGl!XD~?JdQV*NO%W@3`xgA`@!Xut2YC^F$JWm8 zz1$YJ3DA8Z#|;?R1tTQb1PN4rKY@ZPcdO20G#(q z=h2g_|4JV9)i3`_S%(3YllZy;EL0zZba(JKty$`OUkRNXUdqxm9CpZi3SE(F)!9~n z?q8U1XBHn4e<+XWWQSsg3ehUqDtMt)mtt#-@D-r87-!WMk5~%9dH3<9t_;;#hq39s zA%s$nJQ=C>s9x`(%*&zRS0|hRT@~7VrXyLl-W5lzDySz-W`6}zF%t-=$L9k6t&G1l zzdDH>wOy%-p?u-)(LZXJ)85w%yo6$svQJ@<8_P!Iy`|vhZ!rZ5vB`Xz@77l(c=6xu zy$XmCcmpiDm5b_Ksc#AH#uese%-3)ZPdgpXjXH7%tSTLAn>!aDGJJE6mobX$L&6exST*UchLYGC*4uTH&@>k7lk})5i z9c@`<5zKl;e;wan61`gxaIPd0qX)SnB-B3icHhzgIwqMB@hKXQ0cDn6v&ZMMR5Qsq zTlrMT66hDYOnsVIzt7+`-QhZ)$zA>d12rafDGuc6v0XWLa-you0RwDYPqDErRCJj2 ztYCe8ft*k@&GA_>VS)U6$A_}lenGvWIK~TOuR?T}V|X=DQ2}WTB=3`3@)~KiODyw_ zfW-raNPfS4Q|jffdeR;Y$@qP4zPNErg}}+i+xeM@{O5jU*PK(Nir)L`3Vf;tJOwwq zzH6+MH2T+F#$}Rvb2UmS2+KLB1Ghtxpl?c`0Z^d54HjU?ESt4o@7vaj!^paG(!Jhr z1OWayCDp`>eXvP-JdQ(IFUIX0d*ytO&RW7k(T%%Q!5htXwb68}$_vux3! zo-qi4_mIGsZYHL2ICeeAkFKr=j1OFW*0C*qeF-_wPF@~cMyXQu&0ewMQPH<}pKjn| z_y+_*Kdv4ZHZOO^tW8HY0vc{cXZ<_2Y@!cR-At?D2u0c1?aJlrhNW(p&;D=1pH4Hx zW;Q>%Etul3=>3}AnervSyL)(b(<0(pbICuU(EsMY!v*`}Kg_PH^^!Myga(%Zw)G zl67aWw=j`RLIjnnLmbwDogt21`FDXu|ed8nod_p#+EjZcNppJ-$2S(YBZB)hefatu=AH zV32ehQfp*LY8WrRxWfZfm}{$wzYqUb&L4bd>(@8pBK$>(%Tc*FcL9R)p0~d8m|fr5 zB)-A0MkRp#hN0znTOqPkY6Ql+Z}089=)~I;m+69Y4x>KvnKz;a*NX6NCJj?SVP+LE z3W|4|BmRZ%hm(tID}1RNjb0&@TXP1$n>=V+#yR8o9IngzNcy8^M+Yi_Vh64K`iaWv zdwSMu?)M1{V3`}lzV!J*HEoF%hB*j)taKM5+4%T z%X29!hGI|v4ldqIL$U8Cd4F_XC||92Ww^WtOk3g205e`1#sv2Clse!gZFb^BtzJ)` zhhG$OGy|1Ah&yrg3Sg7Q&8GV5^=60p3W}K3Tn1_Hu_FyBDEpR=Y5sAvMgm%W$;P;1 zO(udqJ7UxrqON$@qPa80?7(#6OP1)kD@MvV*hoPG+F@)7m4+-C4F8Q!5pc45*_h*< zN80+)_T$$Cznd`pF@YcTx!^ubm3lj~oymNh{;My5Xoc_q{{|7yH{!h(`RS7~bPNj& z#q+y5D$+NwWTW^3-8$OQMVO{(jf#g}c%B{Gq8ck?m*V|7;i@PJS@%fYVj@FHbhUc< zkpo(a7LsuIY?^xw=}4tX*o0#Ti96-0wphy-r(o3of|>KL0D3^y$>*lLvUW{j>`9o3aD>a@y87_B$2Ku9El@~8w zyaM&8DpLo~-EAO>Etoz;;fWLC^%*Hcc+7vnlufVi%=Ki5FjzISE5hKd(R>j#eQ#fu zd(HDe<#qFPVA(ito?tJhH+8v>#x|zot^9RjfG_+9@4?wCI>m2l#`5Ejv&nrSY=h1g z-jq?#Tcc9@Hb7pjI*T4V!%$AB?CI&*_$3NwVlyi><~CsXebV2*e(!)R7G=qL(xb+Z z7CPRiJ`*1f4Pp^!T@)3Fd7oluf2^y>L{ERxv3@@Mg3R+;SW`Ajy`oe)u`8G%nsV&S zr^c78h25B-GC?3wGE(i%QpnBuDLZ|oB>QJ-Xrh?a^g_Shvke;Z*>Z}fAovLFu=nt$ z!MhJ#m+$&f3w2pc>l7Bdy@1)|Tl-jn-U$?ryP7^1IIjM&$N%d`<3mTXM9rsP;;e1& zU{m+g))Ww+C|vPsd48lNhQDuJ&Nrj8zw=--s@FHRHP-5Sou5LpjOWUm+e;&Yw3daR zpB+;L8~#6^yhnzA+TW7>mBR4wT(|0x ztFaU7vBN_}rq$g3yW3_!0oY_66MV(E!l18{nQg&iOWJkjwC@iqfY4Uv_}0;^+S;7S z$5Jof1G$=AqL7e1G|kI3}=v_*q@K#NM-t#yKG&(BKSt~2eWh%OEakvwJv#I@pX&{SPu zDKb(8bgaMM*xwnof2^@zS2?2$?GvH!2^rc29ZRz#_foT^VS%1jL`J zzKbuuk&&6Hl%EcLMtY?5Qe|2tHVfD`N`2~gIS0PT^b|NhSVhj)l{0mQ_d*N8^7ADXe>kB5L?j{v8b zBeHzIaX|iYR)2rZe|SzS985-xD>C7)efSe600U{C`E1q?EbG6I@%7e|fQ#<3#3J_B zKK%Zl0qKux@%8Ed@CE;T43wooZ>4r8_EGj<`|#JnywN>PiT-{D_wU}H|1MZGtr-3G zzxLseJO}gkYq6`=k8GDeh6${AuyLiyA&GzO!|wz0#&N=3%?&Hmr6l|Nqhzs>%?oBhx4;6Doz z|Npt!RZ7ILp5GpYq{uRVzpLWke8^YjQUV_a9PpMqTk8YU(=E|=1@!aWGR^GYBcVV6 zzSO+o@Vyin)aOF&W(NhuP)?CcwL5yiPK$4uyvPq2fzJ4@@3&6=u~Ytj`U1+=c}Y)& zDN>gefH4LxfJ8KSdwVh>N|MNc!0g*mF)nsW_xUy*ulH^kC~C)c0yIR61{dx=IYe_u z%GJWST%T=ip``_t2g8&6%SsO&YU~f&`FDZHpFcOOvp*Nv+@GrGZ)Qg9zL6vjwZ@jS zQBaSA881Zf`0qKS{A^m~=CZkBZdha4f|BF0K%7*L6^Q+e+4F;H0ziPq9WwJSp$Bn? zGFyP@t5>DAjYoUPPbKBuiA`ZI*e<2d&RqMk7$B=Xq{LVa|XW`68>z9{PTLP{Mo5PFqJ%y*B1u3o`VrO4kqtyB2AJo8mjQE-bR}T zN`T_6>S+;S8mvS^`i^T;wtEgU_@o+Azr+rzCaTN>{#4{KR^d;;_seo*Yu#l?$DRow zhcxw7yS}Y#W0W`QwVapMYGMfC(6Q6SooIhB)1f;L9z;nwn&sXvdJW1a9kV7T9`DBL zptx3=_~qsc=79tnau^wT0tbV&0l=ozeDV&|7$||i9w!5-+}pFPmN#P?+s{5b#^NH| ziEEwj%K^Q}v8By&%8~s#EW)N$<+=|PiUekiJ18_%5O8VBhke43_)gtqO-6t-F57(TE{ES28MDCgR0m)Pu)2g;HcZ~znq;E5HXT2=1%O3R?=r?A zma0@`YBrJ%($Q5rc$l6ePe#y?Qeew*sVJF7ZeI24N`>L&^m)}@UuO2FH0F6K`n8|! z8X2e>P1dAijOj40_&9`2r?5giZ;0D2=kt74Zk7d5=5Y=g9Y0ahXDi&9M{0nmH%U^c z-mn54Sy2(!eLl=S?Tqg7wc!2p^XBgd;2)1#jJeiPNvT0GbPOCq<}{)2V&vVvJY3)2 zcAtzcklRkln*t?NG>Vw!zZYv-?QYtyu#P z_)P^%pl_)B%#U#kGrL2J(yj$7)EofHtu1TeD`98p{_oYpsHM zvF?tu%Av1D1KoMukoDdSDFjUZn0G$hPd+|qO>wdBuB1uPn`Pqm<6FQ@lK)|F7u+3K zyd9WH4eQ3_93lMB&>q=5Q)$i8MVHh~0pjCt%W}Dj0!9fv<*j7E{4CJ6<8iDWfzPJ& zj_G_JDBjS~KDmRjNke&1d}yQ)myJb3uY6aVT>YGImZt;1qV#YB3&V&GuiEV`D7bz~ zSyaf9-CfbVEiah)3R-83q>V4QG42(a3PSuzVY&={^#GoSZF_o^-eX2z+*lqHaZIu& zu!RrUkC}#M9xeq!Ojhg3YXqboj~hb@S=*)?v0>kONe}6B%3oB~i-1R+rvH>WrAHC#8`UFzZ6}Y)u1)Ceux2A{61<;2Rh-Xfn zdUPloZaO=iUxgllj7c0H!C2h&>VEo4BB4M%a1zzNUIWdGBuFy6VxlexLetQSZlPOD z3YUox-+uZrf-{Z#K~{^D(JwL?FsI#IVm6(X3_Y11j?w@WemdJFvIFcERPNjcjz+{U z;}gMe)d>RwSIg9tNJ=kRt1m29MTtNZ=I!%B@ae82tVy!ywO?F|1Bbo41)cm6`j9z>sBGvr)$JRHS;cqb5YOwwh*9Htg;q2%*tq2 z(~j^d|FMLNKZ3zsUu9Na9mnV$TJA%pv3&LI(D_np23lc(+(3m!&OU40i4s_WD<;Wf z3IqxE>kHR%SgiBD+Q&LM7W!aZR^QQIf*O~&ywy#?8!Iw^Ww*fZSbaONA6WslA@=>c z^|fx#eanu=sc^KYkkY5As_ax;)&)<0&=AzAP+dpxLtt7aBV;GM5XrUB6;q@V9(cYK zdQc-kV?av<5ntGElOz>?^U$%O>fup1Dz zl1U+Q%#Wh9gq`Kc(bPJ7;pK~xrapok@KhA5z@7kXkQzUfs7Ao+4|^ z(vJ#}2by>=xqY}5jM;^B()nc$SEV1?p*TG*2Q9`r=RV$Zvar{Ht<`Mvb~#HfjbH#= zOwf8L8BZ3G8*?=4VRdHSJ?MZ?zNmMwJ|y<|InheSiHZW;IMVLjFsz4;Y3LmX+Tb^kr)D| zaB9HUrYnUO_|dCgw;gnNSoT}4@B&3yd38x^ww_MD+JG?IE{(FYTEp-%6|Dkto^WxH4SyjtG4NMCoO#%k44fGq|&CNuyYs2>3g~ zS@(zX4nyUujgc2Kbg`56c3L^%`KmP=c-L#LGodVgOdTos*>7|{IzD8tUkB4;8&UI{ z3t9WNH^mWtFC=;=)9glvy&XVXsa|5L~ar7GL)zsDv=>mvaM-#z&OdfjGwWBvoRy>C_S&9o=Jy&kZCU<>dpEN2P~JDoj7 zoP*0@{jG)(Q1hX+dA;}=uhmca#32fCm-;zZPl8YvFB}`l`%6mSTNex=z@2Lto=jLB zC^Sydg{03{Ha!mwb-TG*Wbx&qV_|oyv^H;Zi}VECz$qo-Fp?Gg|R1 zC`cw5eO47ml9t^T`a-&=lKK1;&uvJA6if(gD$((>F3TXVcQ-giyOGI35prK`?kxUL z{pxQC*?1F+a_=~%Zef}*fQ%n-7&3?g{M*Rq&yU<)e?w1!iA`8nk7#9wV$aS}fl<)y zX3(xj9HQ0D{UUl)27xh8nLy?q@Yhc(uP>A8B|AxL%Q}{70W0>fE4Rs&V-v-uYRQEbS<7>Jz1xSuNAQ*7ns~-8j&D z@!~~c<0$JA`{~xs>!eV9xEPQCzjbTg%Zn4m={~N884O}CP zeM_x7?BS{cml*cw=;)fgAKp}l_)@GWZE`2l$@S}mKD^xEU#MdOs#T=STgdR5;;xhM zjX)p)!4Lqoa&yoY)>k3BtEs7l`Y3vrqfJHOL*j`DX-_jimoP}T!Ufju&YbUADYwCw zxaQE{jsfe!Dl^(m@iHjYv5qJSHH{cUJEnn4)g0uDy({0E-@fwN*`O{LRiDBZu(S}b z?|=~Wh2C@WfSccJR~osZZCR25k)KilY2O4b&)69jH!?s&j;D2j1jEDI?5re zf@fxZ)qaCX^2cU6lC7e>yr66~9F!4{)xpS%8Y_q#`j^nN*uD1w0~d{mH6Z31WkgrZ zpKVigH!)F5)6DcP|c)U5kZsiI5GoEKh#v&a}P zzQV!5k@|*&9l?~r!@Ql4AjWvmq!+gC1H~1V38dgB_ExqjZ#KiV3l1RAA1s@|%`YmABptOD+Z zG+vGs2B_7g36tW*JMCp%D%om@1;rRsC1a}MXl03KGGd^>@U3H-LT^($%~q)wgup%B zKETCgDv~}GuYRlN0>s}mgyo98qe>|m?3Fe_kvgFRhZ&;g=4{Vx4%!T~^=Aa-wwvW! zt5Yf+AyB_>(q1%Gg`G_culj?_5^{=eUw$FPO1oO~St2Lxo_V8`0fh&@@Zp>5B*WQ`>#vw0m=8qt90c=90Xi| zF?^f1((T)o=b**7(VS(|QJ+4CIwVW=5C*6jk0N~GFHzXyTy9!IirR@dCsv>`m73C< z{E!0roeQ1&vegM^WJDlEuK{mQ(;m_&Em66%?b>CyZ!MZnQ)agJeU;g4zI7e)49EbE zi_i8vtm?ec&|Q>H0_PNJX47eBt2Rh^ZKElSt))4e4MTi;!2TyiV4A*a8+drt4`~+F z`gDvLBsJSPGi=LMB#oSfYSTe^i>A=kr}Q&S=Vf$OsYf>Xnv?~nn68?8?+lgPBGNH& zl)qjdW@(L&7g1}Zdcz64Ypd57|H3exzsVI#F9s1H`rIDfGCh{)zf40i z|1(bY|NQiz5tm(bLqKg3)NR)-dy8_#s+HS*?Q^z3rJfN^|E3YZtTJN1jzAmvEVG_+ zX5Mgu#5OZ4EW}=t=q^G&t@wEM(4j+ywe<2*%|GmpbcD<4(y^@CS0dri?<+|DGr0MG zckb`Rk^MiC5_19Vu`RfVI?Iewc_r+YL$59(0^ah+)zLi zXxZBMMf%4E;J4MowL&8b_%kK!`}ae!U1utry$?tioShky5dHFiF01C!^( z!G+;kNT}5mM59(3Nmb{UQ++1Kt})Xk=s(pbKzqx4WZqrbjsokksN2a4;srcEOr!(H z`2o9Yf-sZIcN%t>-%-RWX;lgEjweV`zmC#)!oB8zc{Le?WOwP{8A;Y$oK^fMl7F71#kFO3pOPJj5DsBl@APdQ37vn27~tIEcVf_Z=u+_ zOs)~AyiBecs3r$>QPCR`j|ece0=euXFk%Iy7Zv{Eh%@b1%FX?eFMsDt};obFmzvnQa zMX3*)=H$}E!L6#5bwqGu8znZC5R%Ek8#&su z2zw0T_=G{NlDaJw23dYx93A?VhXZJv&KFOJR726HP2*$g&^3m#xSba4<6Jz_oE7-Y zZ!)(Rs_~LZBbZb0d$68Mx;-#pw1$d{4afX#AMD*`81g4>s<;H3!HDbS!q=8=Zx9_U zKNDH85MYl;eK7?li|Ayqp?yAR0;Mxkej*o4WZIR!+A8-+ccuPrXr_wrDDq;)qSqwX z6Gb_RGc*{E)kpLC6~cG(!66!X|BBCp*M&aMoZoAjv)?m(2T?!>cMzC;xIS%d-_^id zTwbnkY8`Az|86hkOV0K7ArO79u!Ge}7zWRV48!s*h#Nx*@q^gPk~ZiSSW8$u2+NWR z{!@^oC>IA!O_eNMRn0e;N0db5Ur2*@kn7K4BHLYQ zwO_Dn#{ACc{$vrT)me#_wMf{*<4{W1h%6YC-K%0wSWRC+QC+>I2R|1= zuD#if4=Hr+#~3A;-c#k7A#Szq6#ZaDzoWvflC5H*72uO^=QBL^E1M;o4bR4Mw}$U5 zn%exF7H$xS#fHH~7v?L7a`xy%*#iW4W!C7wxBG$_1UX+@TYCd)ncRXoc8{zI>jm+X z@;ug;!n!Afj9n%g%F~Kjj}0*n}*ym5q4yV zSR9gWO2G(6XUeY!Kio}Ujjt8FYARZaq`XechsGAnypGt6rmKqN4cpo+ct$LajVmEZ z^hv}#M33g?6IYLne}Wh-S(Nyf1T9+lL@HEE`gfE};yN@tzX%FR79+`IRA>kjc(ySw zun(k+-{h*HYs-47y_rEJkE&J2B#9Wv!<$6+K-CXo6HCpvvZZT9`yr{CpE^luD|iog z(hXA_!O5vP3Tt6u$bzdl@SBsEhm?6hp+=#C;A7at9m!mrrx~Ry&G^p`&G3fZ)ZRjD zKTFo@&gs1Qjl<_RO3pvtO16`}$b7>=N4Y^zg^TBiOAt>}V5Iqq+lYsWGybzvHV)09B%Gc*|#)5-ki{oGZ@i% zCWs`={-U?Eal__MXQ3wiqBtZtH26z*XyR8cQ^sq>&kS^5XbP1|SPC2p-fOUF3NYrD z>1%!b&dVK_A)PJDS}>6p7N4RtQovfIJ*_fDF)3op@t%y^n?xd(p9G)8lSG%>_&1Y> zqClibqOllz5#=|NVt}2-`ycOHGW^FDGw?Gaxt;B9Y&q*??4<2?>{+Ll$~?u7T6pA_ z6$I4`N;|~-%5RuX2GL{8#@F8=#N5R=#&E@8lIO=YV5%}GmPpw&sg$cURW?>eJB6IS zUqxLlXgyJivFqRFbWa;edr0R^&y66)*h0s{Fc7NnT9`EhUq>9>9YkGyzP{OCoqQ;q z_>DC)F=vuzQUwIT@K$qu;z_fwo?4F)pLfaLFz$WT8yyLze9ok{x3wX%;OliMH;Xol zPpna|PMopLo?$zGQ#DIJ(5LRx#-izZl^Doqt(q)xW`~&Zj9ns$VZUoq(I5>w^Z%VoG|-N7jiC zolUik3ibN7+RoQ2`V$#8My5s<0TU6{TOYali+}I7vQDKqW;(WY-+U&w-?LA((=`Ju zQZxYS2$!9!ok_)sHaI2T<-7HHmd6HTQ#T2T09`doH2c{e>&`)dU7Jk&pA++PGR2TXtQ+K9E7s>!p^jqM;bBc_fjjYmjq|D~M>DX1tY zmLiX$mywm$l@g65#^>VNHDr0?rExsI0C0VF9oak`Sngxq%*5MAVM3Wj0pr2r>e4sH zlkz*wH7m*^%5N4%SZcUQT-QM;kNt8y@y*+RPaDl&r zwIuA#565TNALRn+x)@?p8%2H?{KT)oZ^ynz(M1!KUdvC;w>ellV4lvk@wTZL-XFah zV;MWfb?@KV)crvHA+vVPM%Ehq>vT+hbTLVBTzO2GWh%u|QAL#}N667FbH-So)-iXJ z-)iiu(}y&NHnK$31m5j~pOjAo;jxP!V9=fM9Rxm!21NoSzxiQ?+bcwge z%f_M40qEY!4FP^<`px(Se^r0K!DX7pKhazLXmcuhp7mLWg{4z(Q%ght!Yayis6SWs z$?(P<{QIit$H+>GqoIl++lN5+aPa5v$@vo&-lJRSz0D9|8+{y38khd#|JCyFgScE}jLTfhA>`Oa-0*==$>aQ#?y) zRVW|+{iE;4ok6aX;)2&H;SKxjGDc&n!%M4VZ9&J|;zN%HKhJBG-s}8TQfiLh*bAkY4BnR}M|;DleUbL+a##+md@|GlD9>Y=`&0p`=;{ZXDy(_HJwX)0-@ zx1pT_>9p7Jckpb|JX~dV$9P-dflOa)W9B)pGf1IjHqgNP;VwH%uqtrm<>Yd;1%eFu z4F0&r9$BT6utXhv(9v4cfgfoGdq!;%j=2g zX?$n#NJ-9K$SwlFe=57)yHDj|$9mp@=qf^*PvdlJbvvq3+oynaTR%pYcXLW}kic%B zfIyQ+@fXQi*_%Q4TwGD%2cc`U1I0z%<*HavPv`2<*vU!qh1FxH2`oBeJkJ44`x-0^ zD+&yv5U<8y`FLc&EA?!<2kcu_JNQNhEli&e zjK|e}w}4B^ks#nL8cq%dk8FVQh%)?3ap`!W=&f-5D+O&rj`}CUP@YgcMovyb=^0hR>j6*LA5y+vS2{}Y#iWrTtMM>-q~Or#wQ!oSL>K%alE1nB*T z=ATdaFA*?E&^KJ@?fnhz-=&c{zrp`Ij-Un2gAv!1l$VD-H7(q%teo6!ojn*gEDfLu zC@yjz++kn{=>FWW@*1>f(E1nbwDdgml$8W5oE_QBES=4**t{KG{?r2_K{csJ_^(5DXURQI=fj>@vw2QanOjMQc+O}xmj8Z zXh_NYiyZnUOk?Ze;Ud7!4gdhy09%zV0HI#@-Xveb#j06 z&r1GXkCc_Wg`1s=hn=$%)t`FJ%$+?wglTC0H1wbApL|+*+x<^VPVWEe7PN!xe{$G4 z**Mt$Q#X`U=ufJEnw__m!v`rlN9dSA+YsU8wlX3FH*h#BIW0M_ur)d zmG%E7)poaXlXP~3HtHerKNI#Z;{VS47oiaQpT7UsSp0LK|B(tEX%SQ*_W#V92x=P^ zs~mJ9$?T+5|3Cu@vp*NyA6PQ}^9hZ^{d~6SEcgZkBL*WcC9dTSdz_7w_PU=S9GqWD zB-SUUqdxQvSL=(V`q!8cT6Kii*y6Qv<6m%SP|=5kP`%y9GtSnx zFFnbg6r65z;DbDPl}LJeUYl^Q3VR-|eZJoV0ohSw5#)mYHhf_WW<&+a*YD~%;8MZ< zWl#?4;U$GbNr3sgAohh`Y(r39ml91Cj#lh1!$zj#Uw1-|N*?dqS$1K0d^p43_5rs! zzXvkGpi+(a_k*QD6+GIsy1S^Yd3%FS$3^qkZMi5?Euw&ojS~z$Daroj0Km~EoFOP# zy0Ck0N^P8+yqgLQ%W*Qv7`I_ua##gGB{J)#~dp{v{5e#hMTp{oCmJ zJ^r4rs4p<#V5twREU*6#Jy4JZ8Nj{nT>WA$@Yj3=OUVcI0P-sc2tNPqN|5`>3hlm= zJSX{I^Cgx~3k9~i4jP*DUr*|K3QV|jQ(;=%U-Lz!jtd1hV|jS^{|(@RdYTM0{%^p> z2*r!a*V!J4#J^nDFeuG*VjIh>OvP{hdLhVth2n)_*xoGf?_s$D?fyy1>*>Gei&g=O z7vYMOH}1S6-NfZP$K57Eqq9ajyt}_>`AKvVjvEwO1%t}|R~B&Nes-169)dv9&ydJp zKwz}tt!(y#c~5q(4Rk8x#NgCv$|`YM-p2l#91;?93LG{|$->2?);uEb2f?`CnWY zhR}s8ANI5V{i1Yb7jx8Co`cs(Jx#Kn4TYxe@ot}mGD~ja;)U>b|5ve%4S;(QWL5_T z7D!~|#y->tg&@dL?O-?G%d~h)s^FtO4SaPUJNGYwP7IvV#`2ql(TM+L1$=oM3=B*v z746h0+FMzbZ-Cm$aOL#i)ZNaD+iB)4HfRwhStXRI)bG$Bv%hcuJse-{mz{Gr!dH6e=Jz^Ea zL=HaR<7S7&+FroYaTD)y|A}?cb9us zL$b`CwKUQc34(QPAFB)+Pt_c^n+p~2>1My{=;W43WDhri$f9egXTOUu=^E`k>zUBc z^?aW_+8Iq3y#oA@+d%U7YquXkBM78@c>;)E?2OU1xR?8=;ZC+U=hB+GO6H)RID(#? zcP`mGzdSu)$rURB_8(jWJylhJ?yW(CNNszFLcf-9N@Mlpn@f?uWfydij-;J?F{wYA z)g2yQF2f^Pf!9Xfj8pnFf3h6g4WYK^7@&pzQV8S0xP4PJR?d$cSd(Hi9eDdHr$IZRf$t<7S+%3kz9^orRR&TqY zviYGO7ZS_#;WonbLhT$?f8Dp#)-3($ly)A5K`DzK9uZ^yphSs`)k-1NBcP%yjFQ!e z>4))2DZp{D+VGvt_`c%YRS{3E@0f4L+IMM9yZOc<=iKF29y7DP#fd5+drxKN(LPR> z$h@nCOk73IFVy#EP#AsX2P>oSmVb(i4(s-%^A&nAX9nFx3u|btWuM163(xe3zC6Cc zBCKNU$`v_LY_-|*mCt>)2Kh>`N%hcZ@dF=D3SavaW~uO8jfrT~$t0^q?munODDc>J)x*!^kTS#R91 zljReQI=oux$ZR*eDQ)tMQnTLwL`yw1{=sd+!%!J zL=7_EV_qKm9gIxs-(HjlTxd1yQ|{=R1T^?R4-vGS3Os+R#|}625D}H6_=bD@;%1YD z6!CODZ}PjruZswtm5MQroNH!`$)c!xJuTIsQPX^WzUIaZ{*+k~@j7wvZOV09E@>#< zPtz3phQkM3Phdf)(96n>Cn%1CFr;Jez1{ui3L{G4GyKpGOZ8s7->l~R+_y1i2YPDC zD_a#XkE1EpKX0{N@3V7(p2TXc#%M_AjVQ=otE@%crS+XvT!maOZ_y8}=Q)>kvvD4A za@N`Wk%s>YR4V`a$V5$3=iv%UJ10GN++B{DxBEXsK)~NS4fBt?0VkgGL#zC5zl0&H z43M2+y7W5B^Gr4-$CZ|VHgwAOa)d4VV;LGvX1#@4k3m))E*l|NhfCVeI$hUhp9m(K z?WThclD6VNs@Y$3Gq|Upav(dSOdagA?&e){(RQ`bhCp3&cI$jr%5dlS@APzNAhBED zCXtJ^2X{}R^Rc7Ue(bDJY8hERfj;LB?obC%y9OW(L6OqlgaZeWf|FbjGMRnCRT7sP z@@m!}#kJm)JjiwQ?f4KDg#b%Ku8+$2qpMD9*OWR5cPZYUP&*$LF5-BcUs$$DlW*cj z=iPqA6vd;)Yzsrga4->{hg1-5Q7(4-!i-OA1o|=PjdcjjN21qooP=}bel~V99QuW* z2`jU^V&n}_2;RQ#9MLMJv6-O?+Dir;GP7p(w;HpbGSs|NUdI5w1gQ8OZTP)XOxhLL z2}qZGXVosL@?DMss-5?~U0MN$D8&5!K^>VRY{&}} zs_6dB^w+c0@k=#OS1{s<_KvcLvaI z0)u-1k6gRN|H0oEzF@RHy2F_r##CcNjP`bF|A)HUoBIfNsos*@8^9sa@|jhn!wpfV~Hcr&&&aG$wss&HagTie-jHH@;y{R&~ zfy2_7jEFVC^(X{+Z((G>|JJyTQ?BTCAA3bDw;>;iY0jIsZsNEsO~Cqr{%l~StPhvH zeLplR!?^wrQYl$9^koTtcA!Jv0|^eE`b=_AT7F2DoL+3SOW%LRI1(+8W@uZZZtQ$` zj{FT=S7S}H9{#F6S04P<%MaCa0>e|ddr7X;Li2Z^!7UGDyeIrL70X7|J=|<`A?lsw z!afm1rQ|F&rv26!kr}(k$VLmE+18c8R8$m1%qxK%)vcH*;@)^g%AwT+>%9dzo|Bh} z%lTGJ!2exRZg6(%;o*(wK0xTPTIwqMIiOYkKvuDj@nZg=GqG~@9#w*|mm?PKRQf#t z;U@u3Ds~(hSmQRQSI7Sw&s58f~d zypz5OyomB@?~e1G5N6OOqbUOi(N@DC<)0Wd9jfgWK}Tpw{5ZH-e*K*}zUy(R`dfiH zftHh-<(+vv1_bKW%m!O(D`uUTO!^J*sp0tHkeTUxzrgLtvB{{+eBi0wu6C&Z2psp|DkK3_f@hF7 z6>oX+o0xVxIit6aIMK$wgC>}%{R^n9hIMf~Jj};;cNBMTSu;Caw+*%|`$}DX@!B=l zi*U=N-}ZzF0824{stqzkKK%n`%%bY*DQpIGfR(^Zhs9=^ZL(a^r>kfo{ZsiTiR+o9 zTQqHyrpkm21M9EFz+T>NX>!MCcyYy}Zm=z67P)1YMIfVG*2NUQ^f*>WR_e$oSB z-p*yYpV12Df({$YD}t^l%&B8Z4_{LQIeOzWYVp4x?$rf=uNeH1Mj{_7SzsUTh$(v@ zPvMFO$XK?j1+txG;nQENaZzu=ms5VAGPs8 zeq$N9>caeQ_`skL!Ks%tMj77DSI9gPk1j8d3w285EcXENXrv26QBzOqDLY?`eZQXo z%RruIv^noT>SX1|@n3GWHX*F9oPD`4d@;C&(D*+X?>q(Uw}EL@JN*1(CBFx zv^NCT`{{4U%QQy%uljNQnDDIbpp&k}dQR;Rx>sXk;^y}4wAUQ%WAR^>xT)=@lsjGF zSFU*q>{%u?1|es+nI9N-*-dWTk2QgYaOg@)I_Vp7da>LIdYNZ)U*ValWZh~-JDH%u^{)_G=+d-mI|IF9QlqNc)-e|uC$50U-g&&+Gx?f&q70`hMAYuW@j~5b zv1VD)G@Z@`{53p#G;{sL_~EPU>)q2%hR7FnA!Ir@c4MLi>-H_Lju_yLNYVFdFhwMo2z#pcAT-iNZUxiNWG)1lM2O+OWip1^Q zh^@DoD7n~*7yYO@3*Gqql9Wd4KhwwXGOp4SvklWF-4*P(+s*WBlU>G|N$JhrBAw3< zKb9T-7=N!2pPb}s@fi77g@M~)y+#m|udxJ@mec|&ib z0rLG8r{C5guv`mXS42yDL+0L_qjUYK{?*0adH9wqvp-nVI6SgbG`}-a*3-v*6;K5# z!#oRExg4^&A7wQF%sl?=2}|lcX*8fSSzGb#T7E#mjhsdtd!jXJt-n}@e|61&Ua&Pi z(W!P_ty*h6#&S0UA%>Uj;anj+{9>MLDyz032X&tl8Rc+7Y*I(4_)`_t{AfLyBW zI#RTfr_-el+2_Yg_Msf%^6lmawg8eDw0xDH*kyfb@@N~Ur)}Vw%x=e$Vfcv}KX8jd zB&B$Cgz3W07ROBLXxBI z_mg>Gt5{sh^Hs(VwR#m6R1!`L_1ZtN?ly>lZCe6aR><{EqM}=hx0PBoSyzWs{e<`o zxEpOM9rTBS4>f_2`tQN3VyQ(b}^AO_~|}pm$+7!b$4eVtLN(?*_Ls+ z_WEEM_(>__ou(*Zqm%xIfP;TCn14{aZQv4_;J9PSWD(x+QsP!-Z`{VOE1|0EnSQ(O zLSdGDX|VGNbO@p=%h!-kVM#BvO_YEY-g^>eJ|C}r*izkQxOT#uw^8B)(>aAZ3sp6O z7wBXIZ{=6JribAV0Vp{Iw}I|)U@_L*W%+;W@?V64fvWN|=@Y^oF*YA(T#`-veFEBW zI@N!^q6}oLOJ2t~TI}lD*4U`Zj(~tvMW6MQC%m%m&YIOyyv0t$pXj45C@d3)CYFl*hfTiAsClDJx;5-t!Fx4?INM# zf2sdI0qZ4m?HAZ;6K_1wzLWmWCa!_}(xTVkwuBP+Rk#f>!aqxSepNSd=&?%~XcBOD zw-+J~+kEOboAoMjj zr$f4t?pOV1ZqI9kaz!;M9!r2PMo*pos(z87KG}+EtAI7nRDa-Yb!YL=p`Kpw16X(Z zP#Utg%rfRV(6FYwYBiE$6iF)!ngoCXnW0=1%^9e%FW4AVzXKI5>3L~8r40$HwTY2T zBdRb?+|DQ2L<7K2WG(TZdANv3Cwx;@Xxt`72%ZmBfW2PF~g_r7I!xYUh{h^}@$ z3pf{|vx^^&(q(+-vmhl3i9J))#5LU>$a#34Lp_ip&D0!$7!If31629qdSR>voeE2U z!@Pd(I?7yp=CA4@eF=IYwj9(mYrKvZ%lfxYB|?E?kLmP4k9I7InPrBx^phJKTJ&T8 zN7iL>$VVNv1euAh3h6x5oe^Y#3wUB<5{2X+71ML&1_o~1&5D3CngFT|Qg+)jjef%~ z-jhoaYUjvk_Gr2{fpy2P&^&(L>)5w7zY&%vD=m0E$NTlg78)Ft)q+Xf{Vd{~KP@1;J7J;n7xe|7gO!qr-=ec2crGVY z>L)W4s#eWEq2fOa*t@o8e(Zg$s_L2=WO$cxX^vpmCE{$$De(2I!)%LynCHSVcyi;t z2QC0uQ1Dg5JekwaG^ho7HtjO$H=CUP|0*F?(^QN@0RZSY%8aDA+V&pXfL zv7>cd253713}W~jTtNr39d$pC1S+l3K*+=$1{k~D5Yo^rFNY8K_0Kp@u0nZdg5N@sMJG)~V7g68hcJ3@oQ)Q?F>BAnqv+W?t|0CiqaiNlAwQfw2-#mIiW zh&|5T#=!u`zG_&`oAe7_pU&WVyQf;A<#lOc#z9(zB2 zJFA`P$=6~hi#P3WF1bX)!1&Db1Gp}4a5{5+ZeBfg$l5-zlBTo}n8CYt$1`~+f_2UF zra6cTG$VB_Fode7UkOq`3ZDbo4!VbFcy&>!1ZWIdJ#Mbk2r zhxxKHzWW=>x8SiH(yb1x35AtR6T8Lta~<#UjtE2P@5}GTO2)iRWDRRtR(wAq9^{{5 zJ+DDVpO7G`Gk{F5s-y1e8)tMO3l3@6{APrA>}!f4tFlu!vRzMchouj~GB{^`D4`rS z6Y1=sk)1_?d|1a1d@>}J|6~0Y)1Y7XOuNJa0Zn$tbI9RQq;!ddy0Zx=jX_YEA0|jS z+1A9S(IFFC_*t?Vl-CofN+zhI{K1DuriL0pw@O>_dzmO#^O{Pv`-2y;pj%;sX&W*Fju?iNR*5=d}Db@Mr zB#FW(9|&r{y%@1StQ@>DySbSc@cf;B^SMazTI9-IS=-YJJ^vn??)l~7a>6?4ISh|v z_|d=4+O2~U2@`q88H2P~+U7ybMcnpu_lJJ}Zb?)gFNiczAkYM1G@d*n%gZ(iqlxDSM5xmDR{0SsDbyer-X?ca8h z*p1(Lfja)k>V`|N4zh`j@hOuK5jJEHcq}GIVoZ6_1wF2xbK}#l(>RVi2`VUYvM%mg z{}z*>uZ3wm8Mo8*xgMJ!1>!NuEeV?MZMxJ>ApfmpZ-HU4rr=n#NVw_@=>O&Tnq^r4A|6dIb?H zV#LGj3lUg>yRXn@_yvA#sAGh3gxmzDHcCczL88nZ^}A@RIxjS0%ojWlE5Jl3@D&TA zh&M-Uo;m024mlzLGJu@kBNV%Yxdjj~SM%~xsJ!E7wDEYc_OPXw)4PAc0)OOtH;JT0 zRJ@n0BURoAsKM+fVtb;aItE2P@bNtIW@*63)o`z};~GuAt?@_w?Z@HJ2`mxA^@mDx za`u8xYcNv?A+YG~U&hnksrZeUkzbBwV4s5m{lw<61im@ASBww-O~L*vqXximk8YM)=~m_S75DhcTVS*y+G```> zf*5$jtoF$({GEn2&1j5220E*eK7V!e85t#h>keMowf(h;ioH~z)aD*lnq+W8RAL!_mM2F1}b|9mG+d~@&2R%0oqAig+1 z4#e<9n$sD13v=n0GppC8-2|PSlL(;RJ!9KEO{CAigR;*A*!3&2tY>d#Q;Qr9G&Hn@ zP0Th>K?5JjCVu(Nz~FidQflQ)+^VC8?s4x@07PEkXB#e!dF@9L?u;5#(Px&ba-wC# zf@SY;xmi&WJ;hx{GU)1VXdbnEA~MDKZF{HBvl|hM@|1QTUMp3L3-I>vA}Wy!dl_O7 zEPVs-mCVL-cn9{n1cv2;>g_Vy*M7tocJXK5guSR>6o&#TYMd0Pu(;2L3%PUuA>N(c zDIU0RH%Q?-iRJ@MoKX`*(+qC)77*3wWb$R5cCHMGy3v<{Id*C z{cl^Y6TCgth!Os`-S3TTZlG$nREL=weGWBIm}fMJw}#RxkmFTYjoj%6dMw^tY;3_U z3gIAxjSbEO%=r)|mb2pu1mn~rXUBt?o|G9mh>MpIzRR&fjZsrtr-4W_LU>l z*PAQeUJsWDybnFXC0$!9s2~Gg=ai-h%-Ij`y`O!2YVEa6yiAx?S7AGR4>|=uRq>y; z$iAcqd~5ff_%0%{uw|dKzejc{MJePWMrGTRj*M8t*OuXY|7)YSwTtvpv&x`B0jwyu z@qj~T+#5}jq9T167MuA_8rY-Rnu(jdksSwFZFAdBxN2I-AMt-&;LP65@dGF4D8vl- zb#bDT5iOL&pKkSfc~|!?`+-YxhrtSH!Di#<1XZ2(n6B4R!HCH2nAz|Id)4`tRBg+hj}*jfOvxSf@cPt4*mwyUr@P#?qs0lHf=_ zhxYQ`AAmQssa0C?H)W0RPU&^lvyMAWChU;1_HT&TTK=0T0c)mKY$b^7Y2rBa&cGhX zmse<(bMHI`r$s^H|=(b$G0O&&!Ez;?#=_i><5ld+U}o9sv^DtT^}d z1#~f4x2DY|oQ^Rj+=5%>jpZ4Yuv}wQkjfOPbY^Z-)Df<+{01pS6n23rcp~HVo5CM0 zFrT>hm82KU!esmb1A^A6r$lA?b)5vku65w)OV5*FO9?=)yZdB9qQ0DX$$d#1@4%TQ z27M30a-_;JYO?*line!3cfspC;FKMFsI1b+9Qjo!i8(4u!dXBaYT}pO`}eJHX2Pzd zy^l8D4BF=rZnE_>_D1{1TpaUIkKnZ8Q4DADI8Wb6Y6Ul&jh<$?e-6J9L0&?p6O=*; zMJ~b!wGF{e+fN9e{pTTsBzq>%n1jRT)qTC-sV>y!$!H}NJM+CL31AEPd}DkKV3Ej| z6p~^5rF@iPk7XHlAAX0|8Hq{xLr8`G@ab&4R}8`Xc5?&0Y7GUiWEgtB+n~ndb)^>U z@-b7FB?6POK-gqbhZD3r-B6CQgmk<=WufpIXcQUJ5AU(j{()2I{6xTWuE}UlY@~Va zi`Cl^_AWAv`^j3t3C>cO?COEZvbd+}z1;Gu4r6dTSEjLSW=?54afvcewdw)STuIiE z150(UA#Y`NR;VZ!NFm&VZEkduG3%dW2`ab4%4qUd_F1D)heQLf6f-q%2VSFUe-|$A z=kO+#7zN(u(X=KJ!@caX3Ko8ZbY{dxO@3qVxrdl0{`UQ;NTtrd@Px1^wS*X}7U@UX z;}{Ar6g8z|ihTL_8Pz?y5f#?@`tu#9Xs;M@GhjJYm(`+R#LvWw^irVw}`LLAtF2U7BS?{LZ`Z>M_jj1h+%QocaXQ?DT5 z1Sin~i%E@siMT;9$9o5t(U$L{5T*pv0xpf06=V?O8$?J4w_5@M$dv`L$x}CU^(-#L zk7S18HCq-qm< z{aEZr(a)Tp7#7aG$7z=2!GV%j@%l=dp=w8^ToyNVM34|N0H~<9+hX{&yC{#{n+}_F z{}q4pdpK`z!CbRdzngi8G)VHgXL4k z9^L6vfS+fwK6m+T{?2-55Wq-inHrm2C)EGg%l0&PzWhMqjP%U{7vg-Ab-%={h=qtn zyD$4WfU+NgJmF7-$mDtQm=Op;a9>H~q%vF2d^5vsR8U;PSJEZ;x|5sj`9Gdia`XuK zDko}}ezJlt(!TNmt)7=hvyOd^z#iW5MdhfH0cN4!@7BGWdLWr^!)_Sf#_(4tLQXZ# zx{Cz)M|d3T zU7|rI-c4EWow6znI`kTILdYD;ES_PrPe6~`uL9I2rFSIntcyJaekH_lCs z*8QmIrFz}^jgHK8$$Nb0YtZd~<;=(vB324n>Tnsa;XUCH8E9ipP{lFKa>f&u?p>4G z$$GJ37H_3Ni4PU|($sFQFADua;2px+}^p#1osk)T!=9= znc<-?)8YcH!V=0+r}TxKxGT||qOaZV#7 z{{nZYViymIa{2f|@RIjHf|KI_E?|iqqBtvbcD5fX%h~>B<3|`rvUgbxhB;E2=~t6- z0_7s3<&;i-xWsX2PpGq~ZF@&%?DjqUYW2xaijd+-tAlK+IN9|U+!-IDRf62RA4>fD z?e*K+yG#Q=Z~%&#Bx1%F#I83K$xLB4XgPCf(rDn|`v^+9lPK0UY@?B39d5URyF3=( ze)ukNVX@}BjJL8K%%ffGk1q^<*W)M|)HvobgNWCfHE#`!%pOc{-{{wGG}q3RjygN>B<-8Q#P-Xw2D}#p z7me3H&rLmNl*QT4ba8r@rf6fGuB}d6Ze6e3%5U`5BnBT0-XNk# zm_F2rIU-DiY{u*SrWgr^`muNIvMvn|Yai9iwaPb`TngXRK5YuZ#=AXy*lH(mCtslW zRj@qT75)CUk{E~bC=Y72rbe}(LJb3=iv)jytHFW~&C>ay6Tmu zp9+q2SZh^J$_9B6`e#Q7`*2-0v>z>J1@7C=WQ=RBdiCamVGXgHOqaDk`F`4W8Vd|G zAI9~|Km_`%6ib=fkl!g1IvCLQ@Lm{2z@m`19jRbcY_(lM&jha45kJj;>M_gVUjR=i z{qk5T-h!yre~qFZW$Mb4Zi%>VncddsuZ z`-GD(d>X7m%4_yHO}zZJl47B0Fg@((e_V+ZEK#JM8s(FP1%BJm65{+?CiRKH?HqfS z8+WGshz7#0GcYjgebIT6ADr=DMZW7$45y70XCcdz(!L~huFOy**(kvI?VpD0$Z1mV zct*lw)y&Fk$|6AD94LJx0Gz6FK5du*}cnak-9p%AkZ z0YC~3mdV`PYK)AZiiZLmHvGG36lDl@NwVd|%CFW&(=`1T(m3T#xm-jl`D`@TPFl%zWXf`)DBoAoIMP_nf@ng&)Og7 zw5F+K1d|qgA~cPYy4offThj8rwiTKcyDEJR zR#ksc*c%}k6yEg5r_X{=HHH@zL3T8eB6M+7` zCC3E+wuO`g3leVURQl!ho?ty8{qMyHe$S~p1l41F#R6g0lPA51UudvoJ%|XHe`J^@ zOFe|U;w9!vC+FfBorbYbO^BuI*}tvIsoi`XoOQ#~wzY-~^3qfDetCl$Av=@*FMKK!xC~5f$`4rofm$_!GJ&k`No}YF@9Sh3HAECnuQ)nOS>q{u` zZ3@Q`y5azZj@UcZ(oyk%ic6GMFXHps+>)PfmL7=R=bC`5yFcLt9lu#^!&V(cB^4jR zqI(G-D`WdnBP?FJE!$;5W(7~1jYQ#7%3$$qr+c-KP5ofv16Wev8xmjH!@S-dig<+= z#VL)@?qN1f-}R&4w`21nH}eJ!%lK1G_oyL~&^7kqQto0tSJ;)9(i8cS>TP2e=!XIK zygKZ$g(Hs)Ox*}}!q>S7#YJ=>n&FgukkS?1x z0tEBB3l&p&d?W9jJ5`?%NKI%O)(5th(P^sVd*9`d^IsCozL_NDE-mN?$+p9e&409* zy|zXlc9S7>EpD1W*(>P~p1g?rQ4X&{iMvmyd4=V2akxEsy+PXb-|@$uW3cb zup}jfsI60pM9Dey-B11Y8L6}Jy$KgegQmbFdjMA7ufdb^U(@wQ=rTWNn2R;L0|^AB zUE5etJSTei2V=4sB_m?z$mu?XSYY|WMr&9xAC(1XsV6J>bofrlJM=VV+7xM$LHFa1 zD}_aIAt4~yWi3*)t3Zpxq38Dy)usue628+Nzo<)*&j)rCe)+|0m6w3@0E~UAUKmp) zL8*;p`d+JlJMRW>i%}gm#pW2wP&qqWI)$BxQGmQi*ikLeKzSFT%^30STRSmVXJ1$A z{F;RHnyWGKPT1oy-}*!1W#epXaLhRza&SlY7`w?3sh{FvYO=kLCug9_)V$-Q*aT-K zn$6RL?S@1;zvuTHMZYf>r#$(M$x_9o9GlFktkan?U6*l~HgU2 z<+vXcg4T#5fn7O#)IYTMT^~hQwCYPeyq{`Sfp#`LQ2PLgO9g zrsH_~)2Gc)%TqgmQpa4hU6neadwP|}d$KzT6OK&i(X2ZRj9WxGZ+3-KOPbdPPfX$T zih`xAce$4o-yC(6nV^{L`fzS8?-D-Q0@Ufeje?I!a5I&^C<3-@Q^~$2XJiL1#||<` zsIqAq-!xJy`9Qtdu8O&t3HPxpZ=S_QOp(Cj@!E3ZaYnv};*4i&ve`@40`~tDHU?sH z25p@uLSwb8hvD~Y85ULMe0~qzTz*6#qJn?0MGwYzaYhxcG=Y zMO92~d6*!cYSQ^gk6DZ3`B|KwS*KDxuc)aHx0{m@Iiw{0x!ckTq)V-ar5TDtz@q=% zbV)zh@I*7(DjZqW z)suGQ4Ru3)=QXMZl)nn1$D8U!o(jafU))!{m3g-{h6?Li*IGhagZZ#djiG9X4!6|e z8H@RH9DaroKX2G_+RJ2~e%=~^-Q*K+g(Z;miv!h56!~?_&p4jVtS{Z$@m}+dMpyCt zVK_~14Aa0#n$?_x8LGTJO`^_T?*6BbSVzH(X&Eq~I9U;Azxx$Lih19V3HH>saX;K0 z?56e6_-EFduO0MLvg*QJTa)K6kIqIsVR!gJj%l=WTgi41m-j}E7_pB{pAiQR$i<)g zcWwbQ>$%VHkq0Xr98*g87NXId(lIDCU?E3;=%M@GI}jZZUMYbfpRdo=qjBBOwhPUo zOBc)^C4cMnP~ovQfma*o7;mE#9>zb&*$^54l6ku2(%f^(IZs=7FRX^Qz z#sV}sUN&r_sTFR58|xT=hH&7w@iZe{yUxDMyVQeDH)!xI-iKH;ntFM`QrJ_(G>hZr zDj~bPT^7bTOFVLx6f#}1W8WNRIzQB0<;NZss{hitzAx~+t^QBfavmLL40){t{$e89 ztmq9MKUn1h`_>8mw)y?2NytXsEFFNFJvLkTqWqIRC6D~-XgPM|CTZjTNg$fQ;#B-o zrmK_k2r}<1RsT~Hvy7uvN_lX6rB7r)EJ!C=YROZZxW!rgt_muDwkdyBAG!-*A)^@l zXD?N?miQ_ggwb;J4bQ@)t|PXlUa+#Z`!Vh`@OlNcm7xiqXANuDt18v9ceBdVONX>w zRi&~YHV22Vrf%_cd(1)zpq_gdueQeC<<+`l9iKE+MKv**+0)UwOiec}7g~ukgn)|T z-kPC^Wc=Plt!;S%xRIPbSk1gMCejF%$GaW3zQ|UdJd3vEe34r9Y4c;JDgrc5OerSYa(etCMzCD77-1A->W zhR1G53qe)6IK6;W(4C&+B8V-xRWQQOagSBeCgi)hCCy0JxaJVWli)aHst*YoL%_wS z2CIHiJ8T#-M8H){_F20W9^?p}e>4q2BCFqy;l!zk+e`+oRZMsRxx=3#NEXzC&hJa;Dn^wCeAEbq*WKwK-$thbpoZ zsb5#*8$2p~*pCgL4XK-?6uDuTd$uQhgZBQR<;t3A3Q&i!G5SWx^a*+0y}f&U)pXL7 z)z>t>h$`n0(Fs`8iKjn~luSWfXkHa)&@a467OZMFC~n@<8`dm0tDLWPqySD zwji`0gC=`KM6A4^cG<~mx=L>1b=gfJz|BU!i7nmaZ{kC8$Kv{%}bg{hXNv zKFiiUGZ|@p_|*-sO7*VbPuVIYaqd=G>E#6_)f-X|5emrDKBqwO@cl3*^ZBwUtJ}XD zXJgi6YvRL6dko?dUuK#O0?SXZ>H06N=c?6JFMm+Ib<68Hw+e(<`n377u(gZ8K0>Hbm*Vwjoh4 zsb`wXF4|GvfF+iHoknXY#)|>Iik;jG3c8>dQDQl&%1rU!Lh7_GH1PyHPP_zrQe$y#Z z)KHsA0?m?ZH^yi7y(F%WKHKerEO=B`)*+*7Qkzs640om}#c!Y1QwJWt6gXWu8DPKm z_yr_k@R(RS@i19S3XxGbXN%pjH|^6QV1tm=t_M7qrE6@JigKM{Z%zu|9(AYQR{N=R zcJeWB;1|s05%3)W@8gZf7{u7YLVS<}!V{{)-1Mq*pc`p~<38xKSt-WyRdDUGam!S=nU z5}DHOX!kT%>gFTjJ z_T{4XyU9b*iKa^9G;O}ktOnnV<+AGwb#qSC?|AG=5DeJ*P|oCo)6G$vBe5+OmoAsp zrdyZ3jp1sH(HN&)9xRx_(P!>v)16hRC5n;8lI0L0D1euK942LN7M#+z&+I#EJy+u% zUZ<6PaRy!Rm@R)Q(q;#9WK4jYxWLFoeP`#nI#<1nIO@@7n*7LooQGUoOvoKK3K)x+>fp^ z_44~W1Xgx(V8}6&O@%+3i0{;@QJWjK=J^6K*;g=74?6GDT!SW5HaVT4y{Stbi3_ZViH(l+J zu@(-A^H@Z}6NFWgH4XaWJHo%{%+@oErlD@UqZiT(wbe-Mw*UAx1_31f!)7_%;XFMfU=JetOL|k@-Y{9lM2S7!Yb7(9wn`l&mHJNu^OwI*kfdCr)LfbSkyliMp zsIRN;p$YTxQ>{n)q9qXvivSHbe}X{ytlKv8(9?>#6CuV~irERWf`>g9>S0BLQ|Fct z=25+Yyy8tL?(JB0Q#33xclZT01A9*TevilViB^wKA1_nfx5UR3OR2FL+B(eilLOZ< zj2!?xIqddXs@fM6Mf!%H{Z5RBFl-`uA_8$-@nwsHm|5Yo>}Y|FX}f@h?I<*Yyxye$ z&ccYib^3s1c+X?P6~JztulzcVJCP47!WJ_PqFnO|JN+bXQ>M?c2*yLud0kI{S!T9b zUKvzz=cm&P%Z3@b@ik=a<{ot3yy0#b4t}u*y;L`7=JUP?`H*Pyc8D4y30cz$P#Jh@ zf)s@0K6m_5p$hxGoJ7Z7iCQOi$FVcJ062cbSupV7$_O3QgC!6MyG=6SVNNv|b(_|HmV0iahi&y{kwPe<|z*gW5L}+)<4cHKb)*&xpHY;y#l8+OR(4 zPS9MB2cmI?YM{-5JtGsZ@ardiK;}3hrjv%y8<;--8;I;s?Ys{7+Hn@D!iwpxfjcHo zCFrjt2%6?PCGV$qE4`a?j@Rnm`gR9Jk*s~T#S``rn9;SiG`*hKYcu#bU`-O`~%T6;t zG5Pi=Tt6-)gdjY3FXSeQPi!7<&Vg+N4WKKsr1A(C^qh>#*z2fjap>^NrNfrae&N_d z^F0>`HfqhD_+?*|0X@$c_>t7ypIqa=RqEUo2g=DgAB4LSIzn3J!$f+{TaXucEWXP$ z9*xcz2lGkK-_AY}KtHS>$T$}3PmP@h8dq*0(m|?)HT7Cqcc9c&7)|waTJq*M-7-im zm7s4H{5^ftdpdzxM5QtWG0y(7mq|cD7_HVa2Jvjt48h@&IY~(BT1aG|X0R=2|76@Q zY|qT!^TN%*mu*gbMmolaTj8b-yIKcxn|h*RJi}BQ4pm8EU!SF|=n>+>XjpKzXOeNL z;zRGFKFrDy0rM>yA5pCAf5ck7ovmzDhKJK7oTQ*Bp!N}8nzw~qY>;r>7X4%)uWX&Q zwnQWdcikn#!4Cj(*-Q{MoVw_M)qnOygzk~N=?CdEd11PME!)62hxV+FGW6U>7Lnu- z*7L*F;l7jBhHlk33&w>BAgvZxi~I~g8Ly<`6&}(#s!NZNSpYqKWaAUP!kWR0)irL# zo|9qPoHgT3;BRJg+A>(lk)%_>MZ73m=3vdnDJ7iMj@u!Qj2?aBrbpU-4X z`*>8u5@I!}f&~=5r(-YRneA0_cm?))2JlK_SoE{ob&6HK5;Hte64$87YG`#qoms1| z75}||;xT(U8X~H&@<5K~iCIORDvjibdPR!o?n3@Lez$VCN2MYFT*62)LNJ@Ky zI7}RkO6+CzR}$~yui5WrVm=4f6zIZdy?$E0$Cr77&d0wP+V1&`!BrAZu(C=rITdfY zw#972Bou4D65ah=#io85Hp!chrn1WVbj#$9uIwINZBHFU` zPtJjs;xzc7VfWPr_cMnto~>g$;5JvjxGaDYFaU<>X%MbtdUX8V_B^k%pYN9)?Fqd~ z_N|lAx6LuFR0;MTIx7{b0LoQJ+8hLPD$y6&qK&{ppbE$toBVW-y}=dUL~hvIt!BjL zE86Ci^(R7PllbiPlU&xh*w^i|tw`=8njs5Xn#D(6;hx*0tUpXWGh{Jgy$|U|kP%^R zj5OKs?qn`QCVnn(kH_AE$&jAI37pu~R6{d+gp%K8;?iY3I!3p_3@m!;d+na}Eajkk zZ@4u8XaAx!U+Y(z z%r;0S99-}@S3B{Re^k#DDeOp(>=0A31ZrF?-pfmXp1=sI9AJpgO?dW2=fEwyeTp3I zr=p34WttrwbtxyT%@ZmB7}b5suE@zUg?2#@`Q6#OWZHI-6bE6RI_f(2H#6=nQhmsr z?0*|fW< zQZ?rW!9V6b>Syn^VdS`M*VB!+*t=%$zR2oz1d}hda`H{}WFSDLx+P<754ZgeD`st- zdI??@#;h9;U7W+2hGFKQsiiVgf+W z4ad5rYM*N8C6HCR{f6bz8ok_Z!&O94DQOpyk(Ip)^Q-X*HZMghlKvX{8#^CKtHx>! zpN+W)E+OQ3A;Uw>XJ7Z@9m$(kBe-xxFL%T}Vh(z-Wi(v6lA_YKxw$pdDm$k_ej4k_e+kLqwG~ti?~n#nccU<+;?2Z4KPi|wYY+q$ob7Ncm3|z+%lVB z049(S0;=3P4W|!CFyeo$X@Oc9spn46dWXVj@a&(D@wBjV_UnEgia7J-K3`4A*S;F= zao1v)kgYqlp}^7C#(gHg@*qs)R3T#aAIw#ez+#^`wG zZ5@Zz6YVcYA-9&TjB~3sAmG;#Eu`GBo|6);H|2^^L-0}3>(nH$3cIWbQdc%soSU)_ zeg0XW`~2W=BM_CD&QJHLZfSvd4}NJ=D-3+;K>L7Z(_9^X>9^|K;Tulw<*SuL(&LKZR3B^(> za-2UpCp_LR?M8a-&V zI)BdNa@R#?rtU(>rtvt&mys|*Hc41>q4CDFG%L@fpmmk31OWe%Xl2j83fC7~tLb?} z5R#_g)3jfhp{eT-c<}5}eE&>6$&jm%`%Mkxan7vgPO<_WRIGUQBgx~aMC~&?l=N2F zrN_R@UI9xDVEqCni&qld1(uR{A=hc%iILl4Ops30I#c0K(=NTK)4vvAAl+lE(X z^rRq$1Jc}}?seM2P7+ew?d)wE#I@=rr>U-tVVy$sg3~4pavu4`q-O)2Sa3S4_LV22 zu7Kl+wxIo7-a1i(5aG5@;y3F-ulHv#mxw=|>_< zauO8i#}&}m29%}H&78s-(PBYvg$#27?WVJUFOlCkp!!C=5hEMY;2$vYdcN%qeJr?= zu$ww7YI79EkwJnTf-boBFtWv`^;7!Lv$0jDdL37~I-5GokWene73p2D1FBPlyWIiS zv2BP~zeQ}!&y%e0o9c%qvlC~u*EW-+AzA?sj`}?%fA7+*pD3TzGs=nq5J-s7AuUX$ zwy8Tf9B;dXgh%ng8VR&)((ZE$?z_paPlM4-=&IkwC|pmLoCcYz>!(V>o8>lp9v*&{ zG&tv_6z4O1B-FFaAGa9kXIdvMbV{7r*RNz;D^E&!QFu8Fq0k-4pS<^)VQ8Y9+0WY} z`dGY80y1~JcUg$1_VKvFWaeiiw=rE4RaEf2gtDZe3`^O0!vOxH6##LZ7JF_>&v{xv z_KlbR5`tfNNgMa8EJNb#^;iBp4jqO7IM^;Xxg#}U>#>J08 zb3U=RRmP2GmcJRZdyu`b@6O;Tn}iQfgNWcJM7y)Ljp&P}_d3MsPyMYNV}wo#X_kEz z{+HXa{zbiUxHB#JF>!O{Je+%Cg3|mvPF%c{T z0aSIjL zy*Dw4#;TwOp-zNie{_$~0sU98n$w++*?ScReaYJOG7n+43_Km|SDpLI-qZ`^yhYDw z*rVDy>RS$w-2iS}1Ev|Y1wf5>GkkLhLgC=ZxCMZ|hpT}+dFmzvDak@}u8&FY0p}6L zTbKe72)4`>0dtgp=~|GW$DQuOx$Am=>5t&=WkkpPZ+@-JhBdXZMp5R|idGRGf@NO| z((O*6X@I4$}0p}WK{lF*yM+=5Otf-Yl{n>6In ziB0`RsBk*a9xYoxg*w6Wu+$)C1Rq_d55_OhP=|lBJwYS32-=kmw+vMtKZd|JqHIqr zt!V9t;Ztz6=Vqa^Ax~n#-2<6DKh@3vgKic1)-Y{3Hqu4ki4_p_dv-&NAD<#(ShS(y zU|`qr0E@+!abO33O=-PiIEYmwBBw%6&Apv;lp8ly!MXeE^aBejIw6Yp_jsoSD_=n* zeX`xajNfTA#7wnBo>Z?^G}o|eY)2~7Tx-8+5mg{~OcZtvY9-x`k8lT)ZN=^G#h)+p zJ!A*!tEdR7d`?-t_x1y?^Ez}y4y9NTcRc8Di8U$EWZ{==?|aIsxST4%$TMoN6fadFUuze{QE}F5_Q1@rqd^7O;d{fL2bLs?z zLau93r8h&iu@e2#(Gd@j2f+ZWbf5)0I~ZValgxD>Egi)$gXcBkYziEBprMG?G5o-y z?Pp|OZ`!d|hS-62_vix`76D1tEia%oXMNRKa?u?03{bmJ2+wIi{}8fdj#j&P_L4Jw z>w9}mwRzULneX|xvGUuzAImhmW%{v$7tJ39BiiuGChdQlW_yvDs}JMaRed_oCCS;H z{4{@q<3Y<};)?|XS|r_;bRMp^z5x?V;IrKJwWACEFIcd%y_dd_!fgKs2nMxm(4%kT zuJDstP!bA^?vs{XI%(CamoFxuJ=6$*7Qj}drc}Z zpVQr2`Fi{L64*9!nP}HT3g&px0ZZnLQzO za5c{p_D^sVG!%m0H(8cVe@{0@b$NGg@BUTeuFAt0N82-0WltM}x zgrS}(CbN@D7a80Ng%R8Q-2imtYX96c2=TZur|s}ekaQCx_iUh<{IOj$jkxI=_p7|t zhBMk1aB2aWe*hRg{w8InBiat-2ocK>_3@bMS(fVmh8wHL-CvHp4nEk}D}V$skhEKQ zH4>Sr!Ek=vk6riFE8d7rdhQ6RtEOFOkJYRAD&BJa!ZdV49DL<<4ybX&8B>czGPo_b z)K+5U-~kYgsguhR9ljS}o`5`5a zGW<7xYrSU(O}eM3fp=PkT?)8`A!9Z}>C6GezYlmsKi(TW&i+N;&NYLHvpd)?8++`u zC9cgN{3=a$HQGba=mqwfI(UY<10>|NF{vLEK&D5oj-6CQ-hSG3gdbV(eFAJ((fkM~ z;XPy$QfeoJ&aY6br*|^>cJN=0*PHmJXzj%38NNMoaN<~eGhO43J|A!E(T0Mtl#Gcv z4`>R!?kfEz@~nyflQRz?%alYWcHX?2idnuS@$ zY?^vXe=@8pf$UG36YD;|o5;!mG?|87XEL-LDW2RLCJvT14;`N?CpcIiG}`g>4g#ca zaw~AN18{Q!3_R~;@U^`|LegtVj>E~j0VNnEKw4&Xqk36H2Ia8;MR1V^qq|tAgB69n zt1;LZ&P>hC<>%v>1Za9}&EKq@g^^3Ncc#BnkcohqqhY&CN~Z%f%#2ISK|-MP6Z1kY zbVgb80r%{W9;o9EyUzE5@3*+x7SUV^@;^>AX`B%96pk1>y`Jq;a9*e@0Ng$Zlg9bs zHwsZ)?vgD3&Mhk^7*>pXc0}qzuJs|>hNe( zk7}19aq%crM|ZyH`wiX8z@1@{o9wdlH>nf$_Ei}DAP0DeRBWsvsrBp5H?h`YxyAj-+Y`^^kLFE51>srS6!e=?V8XO><2h z`WuTSAm^A$aD3PT$sw2O8jxFaO&jiM7qTg@FrrQh?;_J-^czyLZ5KL|*KRc{D@orf z^RCM=OT}9c@h;+n5kksNwy(n;eT;Mfl9n@KXuO#(G(|PzCi_U6(6ayAqVEZ!*u`3T zbNzOE{$itz4gJI~is{)O&>27v(G>MY+amu~A^}qIGr#QIC^SZ*LacQ^4L)IXoWuY0 zbE}Sp#GX}8OiAExO5fZ@LNsxpfOQZkg@>@hI5?Zn(G0t>@oLJsx%N~lLiP8;nhn1H zAPHsCTjs;Ceu@7w$8=37Nd?O4@~Q{2!`u!>qzXN4nP$W4dUw+7c+q*RM6&ztpq~V| ze7Of*)f5EpVZR$nejl(ve2SNstZo5~L%r~yCj0U3AA9dK4#xo*LP^C4I*b8~A)=`a z6dzlJ(QQe7!&2@uuMMLho~E_uJJL30;U496dsj;S3^PS&fjv=5=C}6d8 zKE~uK^xX6DmjhpXn?~9`Qfv8C@yoy3{HbHJCHM~1^EoE$c)J(_CE61Ft>;_q{GiZp zzV`DDYN})bl{eCjJ^|d!C;J3~gj-ZP9mf*@RIvseO@?0QEfa9u=|i z&T)oiI*Nyh-Np!fME#%XRvH=@mP+HND$M0{DO?_V(g{rx!hrf+rOSGZLjicSyZU2# zLMZ6R`yQ6Z622GH_EveJNC&CD)epV#JDyZtQL zPalvn3ABwX;o^lT&H03vI7QouhqP$7=lys!0(hq_U{OHxD;@+`)K{ZcgI1d*9}o_q zH_x@y&|cP<)Xx_pkBLS4pZHSC{8O0j2v7jZz-70df=wnb^S;wHw0q_~eTFUjz?s46 zINF8RB)5^Ktqu>+I;E9rH~E63?C4t5!G25U~A%eAC5eIx-%FndGl3aGMBT>N0JL5Xx79#TSn(_N0!pAVdxFUd-#}v|*k7!2VZC9-k$UC+O$Y!cX^2j1X*!O;4{w$= zB>UnSB{!wz&JWgJ!q*%>@)OL(w%*=EWQw>Oh8F@ox{Sdg2l=(L#3KL6Q#U!zzzqs| zvY+DY62FK!hMX2$mt@cIUUqlZb0=JMEs8;ifwF8yx;&dokR%Amqfke`EP?^v)nvBl zq#x$VC1>?F!sPHj3$LL>nt7gMn;K(XX4#C-LlYtz`U;V>ptotxd5=Cnb}sbMl=}dy zLjtn8csV9fjBLI=KiWwD;V$pgwgX=)|A>JePWNQAQ}xEF4kK&+tOEfrv_<|dfU3Gl zva|~z?|P^W-(+H77(yaXW1syL=FZ9JpXbLjgL|5&aRV01jJig)->R=@ACS&oe4Cv7lc;7Z0u@TYo488 zFaa5n6d(1!{O%E;v21?UTYbUb)BvbO$LdEdp4L{d|SlCAJe zn0_6RmkW{N15}=(Q(R%ng}IJWKrENA|9ITz*K}~J*C(U1>A6(K#7_pl`LA%inEoM) z^uGx1XAGJEk3iV>EMg(JX~ssKo?EC$Wu28-=-^XjDe*`KKY2XuPM>9qjJu1>)S1e7b7jNRnHi`8x6Rj)&<>em3 zRe^tdQvVPR__u#3h4mOvw|=N6$er}Zp8rD<=HIs2pKlWw0SWD&)i1$n{{Zp-pOu5- zf-nHtY3ZZ^73M$66aPhqK5#`5koI=@s%J|4@0;U4-AchDz(?w6S|4ivDQffQCI9bV zJQBXs+@^V#G%5GrJnpd^m@;HUhVI#Hd0 z;$PjRfS*@e{OIk=YRAVx^H(h{PAUO#vZ6}5#MnxI)qt1??kq^^b>pAC&Hv_ZfVmF@ zfSEFtehB@uxcOha$9*CUIlz59;`;}7;SU4;k1i{Y>CQ(Hjd#lY7y0&oa-gV>onC|@ zax^kAbh#Y=gaQB8i&N}I<%4--*yt<3XZb$?{``OMp8n^9`N0|Zn$;KP-^b}#nf3o- z=0gHc;4-Vm)`-rTdV5lDUgAI9!hiCncb`is-;MNOo|LHk%Zrt8|L$z~Oa#XKj}Q9a zRw6FpZY06aZ$t0?>H`dXEd?wtU$LC&U!G1<@cQm-aM?Ya`O9Ym<=v*kc^aScZ`b>u z-?F6vJNuLICB);eTAd==-PsVVyu$o{T84jKt^d0u|6eSL&#MoTgR_srZOY?o(PO<+ z%bLGB_Qz}Om71B>OmM*W?iK!9O!z-dUXka4)@^oz?yV@N)@>GPlaqQL*H!+YA8*^> z5HaA&nV13rHs2Oiyzb=Bt}@3@*!o3@ktL(7ukUNGG~LDey96z*8~lCwbL~Sc0Sy-< ztsOl*jQ?f#{@{P0MLwufdf2L18W6)CeQ7pyU@Z?=qEeicu3eMUz^)5?Mpc=lRS?ru zYKi-<1+w+lwR}Opb9C0g*ULX8b9TNd?H|bH|763Mii@SL3QXfo^hC%gF+dboN}It6 z{@wln=02UM?3}1741gt;K0$JVLz)n1toa<`#W=7lYU?05S;+Ag~U6E%=|V@9&?-$zc{;oE8`#uS@88 z56)-FmBn?j1fsGKFyGF{8F3XrnFEJ1#54deobUYw_^XO3d7`DIweOpJHXtEvEPF*{xu+B=r48y+ zYPnMSl2w(7sq#o-*4=k4tLl@N;NRRuV1fGkp@IU_U=GQ@pTA7%4=_Of$MimWaLDn| z4FLP`CC?NJ@Q3k{0|!O|b>Ne8nA$_i6R~zXpe(>3-V9 zXJVf@CeRnnIE4Bz1wh3B`RL@uqPrOr%_kd78oE!pZAS|BnPK&QDQ=rHMk9qSN()rN zihKRp0}VsF?LQNXUSD@}ihwt@yO?FeLa@c1mpUsCT*^FLjy9Y(r&|(rpO|$!REw*T zV+I2?zrJJ#QNPunDMAe^zrdBC`GE&p`ziT7zDhZzSEUl#mENf7s$xKr~tYS@d*@6k|vD)1xUE{T;%uT-}pM)t8QM_8C{xwZk2ja^sY zsNc;U5I*cNxHe%o-tadgs+>7!f1$G4icQW``F(xmw%C0a7PNBF#_d-lH3cKKMq+j!8}$cze_Q2%DVyxVqQt(m@+? z)&X@F!bZoV`P5?OyQ*@5lnrH8SyP3&S3*;uIJDT6-w;o!#|~?q^Suw0%ihGa7~{V9 zaejr9M0Z)HUrtkG^FC|6TkX~cp}Oc;yzu-c?q-B#7gOmcnZqZ6PhS%J2DCkDUdYuT z-qk-g>P?OyVD>SnC3To#;(ThT7)f$_C_33Rn030)bhsdj=tk`C56-3W+YI`y_j~xS zg9qOT1LCaSgj<+fDq<(W-j2&RK!%>v%X@761km5T$kA|wvgywarF;EKdRaeTeds5Q z9BBC7_+k$jplH)7VK;nxb*Nzgw3XCTUd;FOC9{#bQ&(axwVe9Q4>hIlO;t<2p1;6# zS^Uwkm34~VbaV05#i;Fpw&r!~N#Pj!&W@jMA+*1~R{ELByFBgYwSHB-bGd|l?uIbE z8&_eR%&bzn9?08z+p?l+ejU5Ll# zT2v?qeX{lS@SM^eC228V+=ytN7st{T0hL9ijf~QJ5|9%W;23hAR?H^XSihT-5G&C= z<$;O&c%h2ny1=)I4^!VoDehiHZn`_&_^GWMk4c?*;i92SEnN#*@72RdccS&1h2*si72xMS}f ztK#KmEc1u&j?+dRyu5s0aC=E@=7}I_QCs6~JxNbfOV|V+6YG4AFQ5GaC3mBgV}Fv0 z7A$$3fxA6LMBkQ9oi|=xJ@5bW1qp7~+jdYoQO8jL}Y!*v*ka95* z&R{6DOvY(B#DnX2eup8xOy)d3b=xy?jK3)waOKM4l)w-^6Y)U>4H6LvHb5xLb0wOiB|y9Y3a>V$mSu zAkLGW0f_y%(MZW5KSR$8xF~A)M+{{CCn&y1hXlJ=|bT)6KyUtkB(tu~86HNp)&dpvOe1CS)i9j7jZRIMTl zX1@c|s=}t!vwWW1cMCmv6#Qx_oV+EYcvHbdT^Zht6t{MNeSfacNdwzve?g;tH7&F6 zyE}qC($vQUQ6qglb0Bv)!LUwixed`L`}G^1|IOTc-hFZ-=BY=Af1QtG-Li8T<8Zxr z%EWAYLzCwMxX0n(SKp|Z1y4|#AHod_*4rDqJyI^Uk?OkpZL*OW{v88#XMMdBGaogc z;;L@ib*(SXBHx|%y^wz;oyqDsX5#8%8$B}_KpRWnj|sxmd84-Ww@Lox>T8pf_`n04bC9ccd9Ym6szk|`X1I^9t|(mqek zpJXtUqUho($UVxYmKC^oR^Hn4LQr@)J4!j2vKq#?JMb6`M1n#bs4bpIN752{!OEYO*gN&80hM zSU45sE0bAQ(SN%9 zr8ms2n>5zA&5)H>xZQ-IGd~B`EGMg!lr}sIP#f?ky4t~(mRBQ6vl!RSN2Zmk8cc3fk z2{=|&dm);XOcYdwM=!?60L&C|&(T`V!x=y7du0uVwv`i-<>-J!(SpPgKYT^sI)ixlyGvJhH6Ew~l9g{MVvBIL8jr zC7f^41IKg6Y8qL}IkT_l#*|ts0P(29>!kNT`f###yICabviT*Yx0JLp)GcDQNhwm` zgEh+^)$!-l7?}aDGkQyip|l?}8X8&>3$VbB*f38rE)IZ^pa!S8#IcT*+LNPh(X3LZ`^R%EbPdj4h)ya%!f>%>^TN&&ztwG z1P~~p$Z53oVOH5O=vj#W&A9{+Eejn;l73_~s^30N?TYUfaC^5S>~>~D`cj;y`}MHT zN2a$=o;`f8cn|-i5<0_uTe4vK5HtWlWR4pPxU75wZn{zqu1_W3R#JH_wJo+)-|R+v z3wNt|c#B$aw2r@onufe+kz+%deOJ*WlG_Hv1e?zlZ*=_WS4;9FZc^v@`726PY!ZuZ$HwsAZ`zY(--a%3y(#oJ$!%ZiQ1M+{ zlzNr$7Er_x_GoD^g6b_twd)8pm&(4tg28`cQ?2etJ`9ofvh_QimzLi6HK%OcTOJas z=q^0L3)Ol9`Awp??7nPWAah=0XaDk)fXf5p+}SD&Gen@s!uU8=FV)g(pH|Y}IJ~3V zJw$#|kxXf9;_77TWn z{w$P1%daI6iRrX#81w#Zf*ll*KL}J$0{;+2iYb$d0Guh89H^ii1j?u%-w5W79pz* zMF8^ReHa2PqG2s>X9{xN5M&rOG;A@uq_q*0 zQmvmPsBjIhqx4Yz#9{RET~4#>U*M*I)5T^_G@3b1*O)lr>VEZVHFjSmeM2nX-#jQy)?!*d>)8W>PlFN=!tv!6No;Fu})%_^8&lS>RdLgYHe(UoNftt|`q~G;g zn^S2+n@{t#RueTOh%ss#Cw_l;JXbrh0?!OO_HZA%TFXGJoJsb$M}VSivQnQYREvRj zT^1pJr3;kEOtloQ7Agg5#G2=ZMZ%6-u|l!zy9p>G_mGcXWxKoW!X`3%lT76IAogjp z0u>sB34-@9@GVC~@;%uxCI~q-wx^?0Hm~nbEfDs+di$69u#D<`+xud8$3Il@Ax$WC2w9^S9?fpjc?Zv|$(bXZdx zzsaWD_?<|Cc-Thk{O6q1M|uQ`o=f^akv%=o7IQq;5bw1nR6QetGzX*pVJwwR$9F;%aIcYTy-?B7M|Ew*nzUE4?B z6Y~>DzmFZ`CIu!sF2mslnPQe^iM0ZrtVxK!naUJr_{j`XJ;Uuli+o!wNJ zc+)K>o23Z63tFY1r(S4>`CXRp%v8(Xyxgt)Mh})N;txRXT@RzvV~9LlYy`cuQnoU- zjf&iRqv#{$;3@?+ye;i6fXaN6UYc5>l=K65*?cE%uRs{{^n25ck!wSf4kF7eYRZe} z5F;>>{(M)+0C~_iBTCc1oqe(gWw22{_rCS_Qym?v2QcjOEX*qCdBf8j1I{cbkPO!p zaRofu1R4ifGye+cki>;zgD4zgp5@?ZJy^2!3V_i8lQ^_nq^NYz@z{pW4*6gIKHfTM z-qWVimj9}5&N9^bNl2Q7j_k@u%U{^tpA4VPO5*@XKzZHMW8IslRJQz7uwV-YO@*ph zaOylEfM{(h2P#kl(3H<~=k+Cc7T!j##VSp5(W(v;s}6g;-*2&ii&6hh#RDNpLFL2v z?4l^oJ~rr{=eNOZ;Z#1VR1NMdxb1H{Y6&ef90*cODQ>?y?tIm${9q~o+I!ZtJ^21d z^uxCya#l=KTJXS8ljy8Sp7us1>fi`83&Q%P^6`;!A4$%NwByv|=GG`OCe`IMgLRgW z8Q`J%0YppWnZUA5y4deHNqZ;QXD^%N*-JG2I@V1Xp>_4vv+0GV+1v%xMA@) z+_=_=%D)ZZ&=Gh(|0p;dZA5_xx*rtg+T=8>eXM!I%JlA8US9lfvI-fZ%o{j5_A}H| z`sc=`Mi669sjAc9?y3dxX5~#Em8r;Mh@U!uXJj3x&E+7rUwrjG4D{^1tkhJE^OrKI z=;{{UZXqfHAxayoU; z2d+``g!Tjg8L@L^c81fap)Wt$XxoZ%s5=e;XQf3w4-(kwp>&htN5)5gV@celr=0F_u1tPOJTX!QO>iP(Fy?z;LYkM zf#+HQ`(hobryEb+cY`RhWhvx;qzdDObb}@e;(;2f%J_WwR+{tLP3$FJg{PMZ>x4a= zny`lB*&kW}4EdgH!{n1fBElBN1`Ye!`y5gv7!jvd*|bEyCnCVq@a++aA&m8A)faOV zyp~__zSz_Ep^)||nK5Su#g5`Ix-qq#t#t2PF9|zkU6?zJKpc9{!#;}xWrBH3RF3op z`%2WNEEa6>E&>)EmRRh|JjaS6d*wRQ25;M5j}v?^I$4_1DJgIx$KiRUP8ln;fx%UyI0bsU_qK2U3X-@ZOVp3CUzBf1UnYs0on6Qlx%F zE1ZSJp~KN>X%|eJIrgfXouTN(Hk*JTYWzJ~5i7CrCcmHP`bL-|9?5Sd$ zLjc<63_3-%8g#AI6G!Y1M&X&)!~LUUxSm6u$hI3&{RWth#c7Z2;8$sU)>|reUys*i z_Zk{mQ<_`rju4lsuL{-GDad}<5`6bLqAIQw@Qs`bi=p+3t%KR_9IvIN=pr7TK%Z;; zS=l;3y~1<7*@A#;kV(JamJ*m)AW=OHwY~g&NTGy#zKXXu-xZJP0Scff>ZPaN8M1^M z@;U6Q7@4*`M%%dE#HZ4&)=gs&>r}ma8HC5<>sH)lL(8ioZlLn|gProkD;&=SG+op7 z4ENpm2;vB1VBy9seNkDU)~*W+82Zu5AiI(b3B;(&*_~(i`|dfZlp%T3_|cyFZF6;s zw#oK5H@9ESu>KP^xAF$T{oJc2t*tDR z4}<8=unnt#b}e_R3@KqX9#=EUk?sS6gD8 z7U5rCws;45Q~tcbp<)6buH#q4tOAEYxQK3jrk$Dg`trswKa6)#Yd<%P2BdoMs%a`O(Cyk!Wj~6EPuUQ-*uNL=>SGe zqdFL$&iSSEXA4iit3DE(BAYdz9eV=|A{Alu$A(phJQoU2&~6txxm65)Gh>Bi`W#eg zLU+;@>eitfe6Gt}n5$)Zoo*d;)cyOOe*R0wUE3h*#*(&L!Ut#&yr{Om0XF|;0n5g6 z*=&rg`SX(OHZ<-CdR9LV$vg^ws~v)>}qJ-FI#OiYN_3_fRTQ0@5&m zG)PGaf|N+N)C`Tl08*j|2n;RV-N+1~baxFcIdtRy=Xqb(dH=koV8sU8dcpq{9;eC)a^TgcV zHU)m=Pu0XVME8+UiEh|1hbajWo=B!|`C{s}Istb%KAjdJ>x(N-9=VU^={g%1#40ps zoUD70iB;DY&!IpQ?L5z#a*o)1>hQNc5)34sIWmS8xZe#+6z}cgStYdlJn+Cp=tq6! z4I7|iBMXns9UfHQ#P@QraR~hhof6l#=?DwkKEnwZYU=c~iIV@jAJ6VI)d8!S8rMIB zmMn|N>qEC?1ZAZ7qAVnCI`L)*$lK2LZn$Y!!DlR~OyOT`uEQwxm%|TAqdlc%E}>f) zq+3f`*a!6BI1ISyHQuRokuuq0kxynrUWS&?i0JAQCDCC0H*r~KzxC?dS|Mkh8Hv{Q z$8iY_uFu2Bp&#YmDFxq#Y@~mE_FYZ_CSS;A6@bYJequ(5P6FCMVgteMTm|MwHdc9#k1Xx=x&F8!>yE5e4)fs?51r1siPT9`1>LYEX@$@l zox{UJwn0d0ZteI7#le%XiqC68Tv2Ef!-MW4N~$7^V{_L*R5ORZ#v{R0GN{x_HqC>5 zi73gvYZsxYta*|y&7{H>*wK1_iGmH}${ZuxS^lKH)Woz_7WURkZ{d8Bt8AYncu)kA zmsiwi=Jf(qyYBs=oVE-TO{^n*d;aD0xG!;^YMkxd%kO%M|aB7pZYUF>-9oZWNTh?5R$8On<;kd{7s%9Y-u=sM!lf+i0IcI!U5+y*IvguRBL94ZU@YCw z)RzS$crZW^Y>+A!E#hbP$A2Ckb5Jl@f-lPUFb`v_8!?H=f?KxkJd;F-`ZcD>R_}e6 zw}j@&Z4P|nN_^LF{oZ&z-TzuEui^K4%J(v|>;r!%LPi`oF-^woc{zj00_Jq#==SO` zQOCqDINa*VgRPT}N&{7Uz`z4+9etDaUYIdH}` z&^<1}8{!95M-c+?tG@m-7Ppn$uXP{R{O}R(Fkdld+=hwB9^O!L^_b{+{ot{$b!aTT z_B~fid7(2NlyMzQvh}{<>xI-ftlj-1A1sW#HIc`J+s(vVWg|=)ATany?H}-nr1yQgvh}`X2C|U}hC3Z}g1NFkIK4H^znzRcta{;E;yXJSJO(hnLsh59kO@yOsiKC||nF~?C2g>aOO})w;qp1RRPe3(XOb&i z+%tDtxKz8M$cM#2k);vv0Hd^tC5i>ULZ{l`tAUrc?6J2_J^iCLgkq>+nF(JZ#_&9nNSU(S&=lQqHXcr#eo0%>Oo+kXS;HZwb#=n>-l@WRwszfiB_AshU@cgi;QR%}DsHKs{>o2P(@3)7Mydda@ee5?!=}1@J`h%|2#yz(9 zng*s6zfPaQXBSY!3@RQnlb_a?O1bH`s19QnaXz)_a9~cD1Yqt2z0i$NCGG_ni!h20 zBHKi4d?G-_bj2rvrf=ren$>uTDonx*)=w6{sd;5gjXLp&+I}#|vEf3JKUo6a&nYv0 zb5)|t*q};;7kDNB>7{P1)%E%$Pm7T{>l@X4E8MZ{w~)L+b2_a_^hiriVTwQP5E`~M z*3jg?jzoDKAjMct*cSL(4_aSWx@WXMo^#_~eJ**+*B!hHyh%n(22XG3x{u!_3Y=}5 zzew%kr$-$Lr8)CoC97=th?kB7Cv#1Ydd_@_pl`(0n^B?43hT*FjB?}E(nPx$veIT! zfp5Tq`A`EP3Tei)d8}=Mw^QG}WK!iiN(J*ghBCe;V0<_XyMxu_`2T#)JY$$=`K|Fl ze52ba8K7B(o@@y(dy1mzuXet7=&VqJooUNk^wB=`_?b6p_!h+53r%$tjch}-7h%CT z->oN0#ks{g=3?w_D@Wh*w+eI-GCgJ5StzNiY6&InYK}ZyhHpkX|Egt_utQTVnu1(S zQ$@8rY)nh)Q0c1_trgO^1w`=~Q7j=*=o`|7Wg2Pvw_eslBysd`&t842!B@lP_ja8< zwk8v|YV1?g*Di*$^&Yl)9g7Ts#)E!@cTeqwx~{xKU7lRFfLVnOPwng%yj62KUzOJ) z#8p(*!~p!jP4ZW+lj{spwuh6y-*4T|`_m!Nn)Z_@vH~bIHyOwOxtNI)EK7D z-#3xR<=f~U5J)(D>r>{*>8E&cHi9(y0{D1EzrU8uV=bqReC#iwuz%`k^a5wx2{97| z+(fK9c+nCQt}VMLWuV2hte}nIisX4``5bI?U1(Zjxq2PJ5A>*0)r8heRG@yr^-F`l zAQVg8Df=nao+<)>gEB|QKR5AB(RGzv<=tFKb5E36pgZp%jDM5k5058>u6g-DxWGGzW@P~<)X7!+~AH>EG5BL@y%4r5HJPP4st z6re&GaoxgA*8|t;jW&d8jmkPr=-ZP{-tF=n%oDiRL%OI}Z8ZjMV^g!)LAB6d9~%=* zQ@T@B^Vq?O5nu9pt2lF5fLOABfhd@yv&} zdO8YzOhlW@@&-(^G69Q0es?pXip50x4zpwyZB?6f=)6%4H@h1|enumY<{L~^UgKjK zZGYd+C2%Y}P>taZ3lrND$yOwuKzFOt@y^a7gb^!mfosKCqNS>-3dTS5pCZ1dzEcgT zNIW`Ns~E-0`d=}=m>+1tD~WbQdT6&Fw)CX5zNk?;Zo1>&CU_GHigmnNh#gP|)gX{e zwN;CgsfaNQH?Qmdy+;3~+xxW|<_s`qy3vuz#g1YsUY56XM=LOUasA~5;Q*5#MySdkXRgN_1bi+ci{FbA(6IyM_NQEX9G1lq%^w$S6^73K+rRlOzj^44!>@Q3a8TjdLQ451qMFVjak6LY#4zK*B>)Z~L5--vh|M#s??% z_L{INgrpGBC#Zox<~Db!;UF@orOf5X6V*c6x?7%sxmBXLGpiOXY=S#cYM9l@>c*hF zLD8D-pPLSzq>s+SxCjno?Lhcr|A)L%2Jr-29vGn9{gw|Ba=QFRoltt+iirSMa(v4 zemmb^6#U_-KRv3~zkN1xJ4TVF>h_n0V?XDh&R^yV_1D0UuH%luLm@-e7{ORU`1hk$ z?`ZF!ZUM7SqoNa4F?q_Sd$9=Vb+&O*_dulRG@nQu4bdP?BCW9F*3pY%-rS9r+XL+Y zl#mU8+UmX=`$Hu0e)AB*)?t`7bDE9c!zpYLP*+@CI!)Gw#>AL8+gAm>G?are&sv7- zoqH0LICZszZg3+JvsN?>^CW5;k9{QiEQ)P#llZI~PGnEIPn{Mk91+A3&J*9Zf03YP z-C6lyDI95xz-3}wa)@5|FKeY=Tp$$39{97F9zof#vyt~}(6JE^xRHjDxXuokPy?%% z<{?r%3-)X_Lhp@P_<+H9q3&m^e!9?9-Y*YOvl)bZ@*(XKrlKf6hnrdYH4UK&ez+{( zx`$7WhzB)-l<)YXt!{aGr-s(t4$kOmqxJMv2LuA&=P}}-c{TU@+pb#dhTnnqf~J@AJIbSu6vB7blFolB?<&od3M2} z<1XsvAn_9R%Oc9z;CrFoT4v8za(A*JF7ks7Os3q3Oena!?hv8y@u$upgnW)}(Z(@Z zXmlmQv=FX@z${1NiSIT={2UJ1HyD(_w+UIr0@7^;M_0OV2uiKiIVxtM10=6ZW<=I| zDBtY9vU>OoHVY(t2=L6(WRDF5kI+#_V8Y=9Li*C zaY{^bAMV(%2FNb;9KP)9J54z!oW5*P7n!?0YnWP1^3`=h*H*7u@hID( zeQst`j_XUGY(Obb%K@06dqKWPuKeLxEQsupw6YCv>huNf!U9UE*~wBWmlWy(x()Oj`Nc zXp$-wlV^y|6wl$^ny%vqcUq>H|Ci*lkbuF2HQ+6N*Sp6jn}XcN%WPr;NG@I!t@>ec za1qW9iX>BRHLx!RO4lJ6!l6(aW(T#iC}-SdT=-KveWr&ywi$J25YoI#5;|oUs##Mfa0E)p`_p+Jz! z0JKW>`I|PY*Ds4v;!tIj4jm#;xJbhWqkAfM4%fFzGji&BN|ppLTh7+K$6Y&xQ9pP6 zK^*Uc7#%yr9tKb2Y%xOQ%yHSAl#x zWBf$vh-ok)0ABb6pU+)Q9!DYRhPw;I|G1;o#P&v|$0wmC1-t6rgBV<5u@7G{`C-;v z;^uGx!lK|zKeJ8lPz{~9o$*lBV*&8?e2DW#Q>%dVEXwz+lt{*Wnoo|sh{OvA-s$3G z@1LUaVzV!tIJWB=x}cHvj7?Ngu(>g%D268dg_*c?l3!@b%)S=W?ljrXOS-3y^_JoF zhX$k1t5a7(gaz#_KO+NCXfji2VL?dA+(z+ze58w17I6E1Mq3YHwC z?cVp%?%SQA%tFQ;`^`E|It!;X%Y6I~TxnSO2{V^j zwq{O8#tJIm!kEb%MAi!Yo3>KTQc}Vh)X!s1nl1h)DnB);%&Fe@%^fY|m@TB_dN-4! za=8&`j#^zea^f#~(=rzKB+08ld*s&r_``nd&8x<068Wj9v4Xp^F?uZMI;VwV6hbpa zLd1YQ7}oXDWOaM~^TNt({G-q3hzWUT29i)`w^u8v-lVOL=01S*m}R-HWq{%l|B82t zBcHl0nm(HKk=~}P;Ct3N5hO4uoR4D)gVetXYD9C;wASH!17w&WoUXhVG;?R;{%hm! zpKK1lJv@>-@1p5Fp}$X;pDnat&aD!qfE)SHw@Ij&Jq)u;K}N)-M(Rny5^i2g{oRz% zz3XYf4%~;$mrblgN3i@YLyLNkYCqJ$@A@Fmu6RB0Q>e=?%yEzcSu2Z8`wB5Hfm7PP zIQa=q<~B?I6q@J8%slZPSl($tPMHAU!CX$uw8I0M&_lbaTSUgUQwSlx(o3F(d(jtD zIT|gG4RN`2%LBRJ@Cm$q3ff=^-E{{oXQm zULQxeVzTJ3&(R_d%#4XWK)L(v8YZn{vA6fkx*5 z?#SFZ3*mk}Kqv=#T883$f4B6)+r#c>;H@Ot!-J;V$5TOupx>f9)ikj*tdiWHY0UEK zHrVhMBNMq3^BV7qE2|?Q8j(U%#OBo?+4D)$fYD)Y&^Xq57{-wRTU=+G1gXA~MqYZ) z;Gv?)o<{-DHzxG9F^?KToTJ=)2j??~wRYZ{Nol5~kRQ)@=0=>(lEm)2Ts_Ch@z@sr zx!wE9M$=_Zgk@}*ap(4G!1vQ^oFuD4D{f-j>x87;gVqQy!K7Wko$vl{>81By{~U}& zBj{E@!A#=cmSH$}d8;nPmgc9y{1O%QV`uV+>a6#Y$mr))(>)WaDI9E1$S8Bj zic!~|a_~vlGG(jupJi3pCOXi-uI*m(wZ*rYK}+5y(a}_GS<2ul?NeB#RtvQQx`62|Fw)Wui>5Z zch#k^d%cYrg2LIYQe12u(GqoJG~21f&e7E%`)Ga7VX{Z75l@X}E6`z4)08kQU%@To_mWN z{t}aU6Z|cRHNfimkZfh&2IxiRv*C;WQNSBZ;yduh>Ws3)AI3_fRXi8K+W4yf<|(}% z8-Y`+h8)TlP?z?SNMmT)Dz!s=`JrjP%FCz4J9jR_M~s{B8RMX_gPjj0p!HB z#pHX}BPDk%3o8bnW>x=spUBrcu;irO;pvU`F#GN1-55`k^iKb|DhN~3i|{Lz@0=bU z$CS9;y}I($?%WD_T61C41p^L`KSUwQ7!4(6B=*xJnN<44W4HQ&<}UH!Qwy(v^J1(+ z+#Fw1ovy3ogGv=9r;jXRe7a-~y}5cH_HoVA?Bmi*E&C5t+X;cs!2V}GQt+;>2mHfH;56-TFTVz1B=z;>H=GU{f5Ns}VhWmqA#xvwrn z=V5FI2mj*TsY%Js-rpKuX`cOZe}+2D@#p;^W%IvHJnT8#g6TzLKy1gG3>zRcn#Mfg zi1K9&*HaU8?j8TX^qMn7C_%4|dBVzcG0#?;ueqZ#{3;%S>R{RaiGqdgZ`L>6kPL8I`94d}l zp`Fl#MO(}{(R5r(&EM0w2f%A`*1WSv10EOf_f3|)&X$Wm(+6)46S8!nV6IFOo@%H5 z*A>2}zt|7=5_poC4rOODvz1gQw##A@Y8a;4OdCD$MRyg{i^(!d{Fb9=KXd5hnjIR1 zI==&L!@6tN=cds=E9R0Gk!QTpf4ChrW3^0LJRJ*kqaliq`fLTQ6E1sXd1V}-h2OG2 z77%gj34J9`lApbWG`8E(cX#=ji@aJ1C0$rt<9aaAOR$pu_5E4mFzlQ3L5O&49dQ<6 zfx@}s>zC9Y6Q1zWSW{+(yZ%F|L8}WnZn4XI?!>#&W-b@iQVEdRolqvzm}CSFt_uoM@g&! zJ2+MX?ao&VOpXqASJ!WGvBSUe36y#d(7*p#ST~2_XC;(G4$Fnyo(Gp)&fkZfA2cgS z)w0YCoia9!{W-%vLCV)Ruqb)TY+?wWijHEafJMqpGK(LGsGcj{C?@&l;Y6AT6-3mAq7g$XD1(O@r*32@UNj8!YiEGkb3;4l z?65C;u4)C+&5}#xU+GK$7S)3M{Gd$3`DZB-Zn@3= z;&YN)GNmaf61_Tgylp_1R#atIg7iyl zVsrRcb_)Q-IYOh}PtJR(CXOQ5<_5V0LvZ zF|588`sfblCftLM3qQGpfDr9&R1f+6X!gf*V$I%tqMo(gI3slQ(&mLxwA3_{lO-%f z3#6FOF8En-bo`$Jay&3QPuv15jYD4X(5oi0-^+Ze{%z-X&M$#?tN9zpSyce|Ei$N| zbFMiB0O6P~KSJ-;^p4keUIm|Yg8zbWqNl^31b&L|+ADjh7kl41j8|lPhr@)~si`XG z$CnYRqwBXp2Y2~3G7xnwe{7CS=!mn|%qM0SUID)zD>d_nyU0}5xq?y8{0F8{OoZFl z?<8b!yIm%7jjdnTc(j2c(M9La!~}K`*#2qjwgM#FoP;#*xevX&ke@V)nt2QtgBURO_{z#HRkho zXXV7wAmW#v=dM-8)wXYI>lXJN4(4|o%g>nhC3^?BW*}K#-F+HkH72T6D|?L<2G}sp}jD z&?Z9{uIprfa#TRiQxBOv@|i-iIv#Ai)wk}o_;r&>LVq?mAJOfj{yJE%CzNa|jOPSRKU zp;u+Ew)@V^AZtX^D)CHzOyW&75KLR;RF>q6ujmgXN+K5S08x#3c1wEaU3+sK`v%=7 zwIpmcxQm#+z_16%ROiJ;MR-9u%7*zpKD+0%QSDxS3s% zX}hPOE4z%uH@oxOqzQuZ_+h(e7Y6g3rjNJn`US6e3XpyMiHVS~NLdG{%$eJ0G*M8V zS6S(do9$`p1)0Vj5As%ZWjS0Y=ch)@)0Ouj&k>e}ZT{KSHjFkPPw(%S?N}^gq{-yS z`Q)#WaEpZ*-s;uB zzp^6HFU)8WbR}~1-r$*UxNB>+0mNDl(=UrSbT@4uza+X0JHdI&6JkOGRFEWKOqAFn zfh+z1;W-K09Yg<`>Jc6Obx1eR{{8SfD+wff0yOt7Yh~+R{Q9gAd`4F`%l4=7Bn5ZR zNbLPPoxz<)?aSVNRDd@Oe~+(+5PL9eTmn;J5r7K`adlRB0=0lXl^)4x0B6bj1aq;} zxS%Ea>?BLBlyzFbTF-!^vQD1)&fC-gdHo=ukPK{mKv1VK)l18&bwH=xtPr+>6%uRF z{+=%3la&Y@RtO7;Jfo{(@WBc!(l0Z6nUZtNKbPQXxmU>si+>{LO<-vlkbyA1pZ_?H z2`MtA3(bIodFdARRjCJ~&j)Ny?SYO(Xj)}w`?E*yv15&1dkFiZ7$MvMFRUlTtgWd zs!h*A-Sn?|Jzs@E2CKN6XoZ zhL4mr8|P9&r)TkaEg6uqe(E8@r*aC;DT z`{23F>Gg|{uAy=o0S1#*)BVM=Uh!A`rhffG-cln>K08M)7G)B3a?0)@ieb312V=Z# z%O8KVD#pZ@qzM1M0Q5RNB`*lOBn~>aAV@N4U$G4x6qfTkCWcn-1xpR6sbK9$j`QRo>h zKI>eOn7??>dNlvIm*QJ%lVe^U@eGHyXd(g_b%VYn=|zwN8P1{yvn0SZ9AOx;FBe8! z1U@3!f@e|ZLGnU51*h7x>C(p0geIJrS3xg7z?J5pj&i{!8+qQpD%ZG<-vSfcu*Kt_ zArZV=klAXd#|ZJ-3a_FKK$|XXtm!jl7Bnv!xtL3eFT341t_oO+zPb3&dtIM#qRn-TP88emseZP|YMR*9j?7d)L%Ok!?%= zs%-*HhwBv$*O&#iRwUAw@^%IM9K)ZYF%xc9>f8V$joNGCAG*|_k!40Vjl?7l zBj@dJ2XUA<_zNqZhE=5{-(*N-YcO*QBIA)qGq_78g9X31X2x(2Xi%*{f32ybp=+vh zctbg2<8)Xt_?0nvthn2KueWT=I~LXkJlLPf$FX|#ICvnE60LS{PRI34w_Qp%AKrBn z_I|6T4W*yRH+c;w`du_S72tBaEn>&=cN57=%vb(`;)v>s#`_}={0tUbE`$|nuH-vyEO5nmbJMIQsvp@)=8(hn` zak^X(&ukq&xX~7v(`5DfagyQ4?|!ITPrNwF4ReI1VUW7ClBf576~R$7#e0d$1gbzK zjG0eU6epHY*9L%$F=5>tdrW50piY|FdkAj_3H?y#rtHy`lH@Jo=fc!|DHu!=0KdV{ z3e|MSRW9xpk-8jiz|Z2?df1iH`a75Y-Z$~vW&i0)K!*v7Vi8ay3@so}pz2FaGYbl^Vvt`WD*NMoO6+a++ z$&euo2YVrp|F#ZR_xv_|WTe)QdiNySTn#`4(d}>to}4d*B?`;h(z^i%pst1m2WtrW zVD>kU-}Sk)cI_L>_A}Y_zQi8-vO!JI0N4YMWtji?{=)tBfwaQKUA5Z-DQA|#$>O6V zs&#-cG#xh2(e=8muZ7xHpF2S4t)vJ{A$bPtvgH2Zuhj?(_ihzyZ}-C=b!6g4fkbdy znK8QAxwKx-M&$>@odNB*tRTn5 zR#iHwi5Fh1JbIN$bX#k%FEwe8Pkdy44x9)7`>N6dg|p8!B=WbCyM4l(!p|9FVF%}m zw`}oBr=@_8wbZcIv1s97P(C5}2VogTPWmQia;1e3HO?Rnwi!gpLZ>{nzze5*>L5)Y zPMH%xE_mX3$`p>oG%F}O1D@M9xJx~vlNILBDhOHgl$v(!7Fo~4g5Q5I*Yo?#>t>9 z2E2H+`b)>|dSme2$m;JxnSkilyP0qKi);oq zc*W^$W);VqlptTwX`NLxr*;-p>t-bIfR1HZCJAm$BF_Aru(qFmGj-8AR$#3DfC%^a znfrQ>;&s!<_+b*i@I%IQn{2s><@kHKbJ4Vpue3fLnQ=;SX8w935Qx5?Utsl96JGKm2k$&syKF zB#Suf2$fvpY|pC+A$Vye|B8?41@QjUa>{F93{y#oM~cMM$rpT_VH=r~M!&`*QJo#o zL(^m74v;pBAAmY&z6?>shvap~ zfU~y4r=|D@mjRcjRe+O@#KT&h3@;}U=Udj_;}SzJgU)QrYx-lnvD!g{Q*RpHZ~6Rr zb&?u71(3!ze_*@e&z`n0?F-?2rIuS<&@9TN>e+fi-n1~AEm4s5g_6)BMlRLedjQEs zJuJgq%QmBv{lH`O_Uc8^MxR0@thcFA;9>J`^A>NZM}Hly8@tj#$+Mq$g;*PtVA*0m z)iR0r100NReV6r7F)%X)qfy~8YS*w++1Eaw;>E@NJWLNtVG7lH++gsPiRirv zQ8eeSI0V=<2A(Y#Zp~AEumkp~I*c*_pJH{{Ny~p!(ZX>JzdVkb(8AxjnF3@dLlhdZ z`_Q(|<~IVubq#h=K*+Lqe`}a1>RXD}Ll<#VA^hc*cblJDwD94h!i2? zHR$7x9a-K#a*c{ROaVVXjwDa(diV(!VzULPIAQQ8;}iYY~m<0YnE%b zfK~%^rrCe!Obwj6#~P1f=y@Hv!|Gf@!@whH3j_Pu#bKM0jw25*%+8#y#~uv6#htgo zq02|+0gqFWgqzdtvS-FHzP3E%T8;_%DV4c0NITO%-G;dB07hv4yc1i@MAmk zi!yDKC%Ec+gqA!1_-rm@-5m?CMFc<71uJpOav_;hqqT^#a5)H4M7#{cCnIb_ekJfY zdi$;N`J z`2Y**s93z9DT7?KZ`B5b{DU;jx_KJlyzX?NUf>RuTy2tv`yEfy+aivRzW?%<4qxug zQM-xLU$43X6pbAEkpzUJeK=4YHrwr{S^P?zpH-M`RVn%bmHwqhvxd@{aIu06WZd*F!~jGlXIHM znLTtE{54;&LQ^`kI>ckB_Xhha9H=<6V90s)o(+~sZBVS=oPqx30@*XGc>LdP2MV4e z`2dKi^hlYUr<~2S9G-8^ZHVHdTS}?f0r|#tMN04Wg5shxD`+P*Ge6w#W0h1dyri%E zGAfy_9?bAGFVq@?HA z$BmN6ph2jsY-Vv$yk0lYQlW(mxi&=)Fq5?`kf>bop6?iiaPcAudn_|{BIV(gQxmwGbdl6(P-Ur zo&h3p5M6M&xW&vl$l0ihSSS1oPyA$?EjG&4Qg;yMzR6GJ)cn4Rcodpe=X2)&IC|`8 z!N1fjDH*k8+%aa{EO)wp6%|wZcAt9tpfu3y!R$~0*&-x@UvMG%<-!~;g0Y`7e}D-U zb~;&%u-$bdkHj$U*Ci#|cLA@w#W(X!9%Cqy0zBjYCN8Fo>E-KcC%5S~-D>t6(T{SKHf#ThJyj3R zvViOXA1E!4kXBZjkM!n4z>whf6RH$)#*Yp2Eqa)kc|&Wmr{sU7{){V#DGljq9jLta zqXMk`_87L6vU#~PosB;kcX;3?cHS<>gdk2u_#6o0okoYvc70CfVn2d`=aS&$wb+qQ(OIYypAHey#ON!55q$}c0(;mXif*OqU z;%c{#rwafrOUTBd1{Bfs-IOHv9QLhDLiTBS`@hP_VVToY52SO1;)i~G{c|xA6Yc(! z08Fc4P9iJNY_!ki!~0o5u+8Ad3ZA75qJ{_|PB&J@)1M;zVrhKuJ9=oKw-@pc5X$rB zzpcyjTxdU_i zzI!(`=iP4{6oAhIGa4+F1V8X-cA=AFx_)id&moTRZe~z`gCPY^t0i4 zZMWXu%=h4Gn_GhqS!umU5wjm`)TB&HzavQm-B$yjXxDjqFg`oyA4fkpiAH*5Y;|z? zvQgvKf*wt_weX*tQ2sPN_Ap`&kn$iBA)>FG%|THxe#1D`TkT$c8X&0G%>44EmnwRV zQ%KS6VPv)B0*}3JfPsP#3_>%E8emJXc7>dA`z%^55Kj}Li86Z+xFfAL3b8EGGF{9$4c=cSN zivKM?odVw*W$KgCz=3Qh_?eu0d?0n3Bc|EESWD72ci~G`OIgQdbLZ_#=`?!j*FL)R zcfq|{ZTJ*Fi`ZhA50yoOtFi@(rNLYi_HAO4GJNcGotjRq&@$MzRds+lDb&g*%y4OITP%Q8)0F4%y}};pc|c^1YfV5hzQfr8Ev0u>2C)vm z*xTmD&bYB#WY=nDd6!O4BkzD0aN?J)OUQUx;$@g4C+=)RCjHBw`@!$vl7dYh zVl;XacAiey{}Y&k|mx*s^~0j@Jfoco*-P-*?Gh%%!NN$NX@` z6~HGx$)c*YSNzc@r)ydHLRK(v0|+;IQxZC>{F{~MESp%dXLfK6qyoaf#Gp-eMVC`Q z3s}eGgG2rIqzz3reS<*RO0n0a$RU#wNr{$)6rI9bMTy6ynzUG(gl;ckq}lgg))O;w z=oD7@l!2UwBQ^uQiL=PH{5_=Gjz?6<{D$h80&Y%34vUhl`n^lQA0Ax5YL-shZgcm3 z>;JxD=;+TW2Zml1t+bBy2FEK6p*eZgmg!zQ2S<0Z2&+wZxNDuO#{2msfuJFZ==jR`?V?a> zcc{gM7A(T|&kf1@%))HwUMA|@d&3w0kqwO2IRin`*O+YI(~HC{;{*H#{z)QlWwtus z=F(uCi^(gJ{EDT3&%KrJsN9k6aN)CL>d}7@BdIso%-+`LezD?1I@W*MbB9Q^&ox*u zbB@nQmGk8HWw8!Ys3ZW=Lq!$A^k<|OF7fTZIgBF?&NZ$!J}H5| z{Rp65`|@Rjec*uWz0<}X-QNoC1N`{E6W~>3?|`T0(|_l&eKX_UN*Nzw(vUmfE9C4G zgV>N}T9x3l?^<3KRK_zDE=+K4>zL4j-g9U zp1TRQ0_&RxpGix7fJt9Fdj_SVO=a~0tK$D-?=8ck`rEy46(poZN*V+Nq@-IKr9@g_ z01=Sx96}nTrIC^@>1G%jq#GoM9%>l6<5~Rw`?~IXU;Ezs-p}*ud9mMc%nZk|)~w&G zb=L1ZzaIcHL5am{^H1IR=%=Xor`HhimbD}p^TUj*7rs2w4~z!yn)Vrk_|FbfRnJc` zH@Z`9*8f1MJ^R>Bt=Na0f5qwj5kAGPXZXiT{tb+1)lo1!NjVui*%AYkp?m14jPFZ9 zzn4*RfdGYI-0m@E4l3{yGE1ag{gbN_65Q(ULmhqW<3p%IJfDCnzx0a1?%%xU&{(!Y zLkGa|BD}ICvHp&SK5)QX_ou3%Mhm4^(;r8@f^aS2pKE#l!p?&Z zUPKdPki-rswt&&I?JV#*&gQ!Hvu9l!W+ zetK{$b-gHVru5CtyJK9~%>Rnn|ae|B>VJFQ3pT5dy4{JQZrJe=7a_=ai9ue8c4*&i`_l|KjccoZ-J%oPYl7 ze^%+AOYvX4`2XiB*-brxiPZe#;rNGO)_?vO!HIoKwDi~K9_C-cJ44l=>j}TZq2f+2 zV{|zM|Ggdk+nW9#|F+J&St=ob4hO}y8UKp|{4d_h>pWuKK1iz!9%}yYl9(P~rs4po zpbTKVx92~#QDD+^FM;c{^u;=HOf`zyTAy#^+AHA&!&$_~6N zOoA2($F&dD{+XZrXHUcnYG1;Zca4NCq1RVUzkPf{Q?KO^SL%m|3@||Kl9zq7B|Ubh zvIbG4BV&(IgBvh9cqe?GziVhMI>bO#>^?p)Md9`A^O3-^M;V74>$WW9wEZIu*PIy+1sLuUfK0h+YzRL>wRD z*?mS1x6t{iDYpL1hYt^#{LF`O;Qtzb(j1`PiZX89o+}3Oh$oy|H$29SZf%gF>F9FU z&eBzXc^o#JW~{tI51_FK4gZhn*1vx3Xid!PqpF;%8)FTR$b~db721BKN4*L(X{n7b zT7KV^$}cXY=Yq5M5#yov2mKF;)PHbo<}qK1o5kOSF_vf$4iLC;x>7*QS08%w>8yVrsuK6@Orr+#-Rge#(xJgbbZ)gR%vr1 zn+5?gG>0ytMDPO7o|8DG&?`U(=*4|54`_yl$8l&e?ui)p(oms^U!TWX%^us>)zuOI zW_3SY!@WIy?SJI9zZBJ$6Hi_G^6m{Qzl<_CDQ%Qp1Y& zdUe?QDeS>@V4Kc}5+XomH$I^1+k9&;i8wN!tN2<6aI@{XD7HIo zy882UGJDY!GgUHdmq52Jvu&w%_P9y!cbz>}uBkDfJCcst)t~@NK34b^HXVp z9z%ayu~b&nG0K_stE5E~kCUgm=cPJzr4vP3NIi*^S3RN1F~W7WiN-B=>|O5?;=;+e z8T-8(V%KQPkzpKb3LH`S;o^a~3*tM7WiC&8C z_IcRUg316|7gpZezhEFM8$n81Td{yhC+FktX_t5IjM+`SyEudOFc`FPOP~z!z}UU#@d3!a-eM_RL@EgTunqiU%@6CG zz?}A*(2l3K?7|y{^W$vm&^=PLifTEe1T7`o?eJo9>zsNGupd%Zx6FpaBacb+_L(YX za;PLNH#4pAizCGiozLma-??a|K0g^WA9;nt5_oBB|b9n5~9D3w^HplV1@$=3tO^*OkU7KN> zJz?JOF_Ypo;E_}ce4twVjfjnIwFh_nlUota#3K+nKTBbv$%c!bti?_~;|_G~^UFDJ zq^bALe5f7sUztJcTZN@Ml<(_j{y)>x)sFc>#cMtVUZgN)+#9JJIDJ%oo>zCW*}Fy0 zi?t%l(ilsC?$PhC?y$hl%{Jczt_vsUk~C}3M9hw5nvV~}c@0DEf5q43uC^EY%{v|V zY?w4$?x^7P+wevfc~>G|HACvKzgqw0PaMT@fXqWV&iSU{$!qpYypq?h~LK^^B&y@7ka%a>`E$%`P>jGT8oX(ZRTSJY$8(+}Fb z9t-c3aL1pSi1oT}F&z4TTRUh&r`~#~q@Bp5Vv~H@DbVXPsBuol)gr`S88b{$_IjnN zpShur*%RdvTCCywxA^{bc3YLMUxw|gn$<9{cAz|4a~W<%rS2;9?q|(sxf}YEpWEw= zPA$Cg-n4pqe_HBjsk#hHmvwqU>67;dxmr?83Q zg3F}*IYd)6@mhC%>xcvTOq%xVPqvX#C9JJ$ySV_-3ALf6wOeubK1DK7m+xY4MG~pD zCZ>)8_*?d~ME+bFHQ%N;A;ful*kXhA(;obGdyct&!xi?!wr}Ji3f<0# z9H;Z1p5WDP6@7oILAq|uyJny3Oo~DF=%ziWumvwgG%<%>L888Hd1hSw$yC7!({`{y zy3hG7c3z?8mSH-^cI&blt=O_l>*pRDnVtI*TR%!=MK;JZDo_$&>Rj)c#n1H3K**X2MG0j=CQAr6oTU|L{pGQgN1xUz$% z1HMHFAyKOp%3$O+dKH>t6t`ae3>rqvQB04O$u5<)VcI*zWMiHOuPj}->gDd&+L7Y3 z)>8>nK6MWMbGf!+_IY!k>m%o2^3jKxBDA7EYFM2`j>c(G24tZ%$iiws@I`6CU9EzA zEt%oy(LJ?G-xW~RfAOr_Yb%k~Y097oO><<0$xvL}Op%f!q!T}H(4byA2%9{+{{iFx z{TRVc$f%^H7UszoiUG?g#2@^!y4>}2)HMiqkJZGMgD8Ie?9_**`AUqjtmPm96lt#a zo@8_$)36Vvx~2^aMJ7>P6y}W2ggRE_I;HV*yJdQJy3DTzJbQ zve_}*_%L`GtC zS{fI0-wq5Q>chvTqGT>Uv&?y~)WRCI!|VT^247mEjb5))k@3D0nIJp_0AA>C7489E z9D~VfHmqvh6q;Z>Y9S~2MUNZjdw^HOUXzN2fb`gJ+~g%JiVzp%Y_3#VBA_5`j&5e+ zr`I#I5*JEnGMB*SFxo}Z?;4DUCD~YKF{jz5b<6Z9lRKWOZ~U%GMLhm~i(fX9n!4t> z*9F^i{;B?=qBh(?YEMp!o6yVUM6i#SsE z&(+Dq?TbLiyM6h`o>8Zl($IWoGnLm_wsi1Vg0r_lz6=8MOntQ+e^=Ub_Q0>(ipkM(^1y+B%Y6e5NVtTy=AX05PLD$ z6w;iRX#Xr3NKF8mw5-36Oji^0s#Nc%;m2WD5(;hFFl*?9ec%PCs*I83d&yM|M;URx z{&L$VtFm}8O_n#@-}q;Q_z~7ow64mZoZ!4#^2RZkzKVQ#9nzYgtvTrmsrf`cFIWSDTN*I62sR8vCSYP z%I}8($BrTHX?T4`ROx(FW%I*+#{SWMcNwTE9YA?aYOTlf78B>EfJ>c$sa2y@VuWbR zm)qkUxS6YehSdTx!PHcbIAMRG)L3hQ5zSv4CFHF6rnFjp@fzj{)~+-W)~dDAqvQod zOc}Woa?X!ZRc9|N9K2u0TZ`Mz+3LTN)%$z9!tN0Nx5FAkOv~L9M8Gwq8R4I!@!L`N zoLv)A@1(I@nRa?%meKp5=9>Qmgs!rp6Gq%c#3Vye)%dC@!V9a$ze*!V||-~n_VTowt@otCOT zqD1(ZBQbc;a&vmd@qFJ^s6+@iz&&dCeLBnrEoMn$z3sN5Zw}VilxMfRvo`x^`iqLT z_$bN6^>(pTU-5>XFB27Z_PoKLMtzPRNUyRj(|~KO(KxZQw}CwSIzeDVQ&}He>>K(| z>7_0#lsN^C25p}naWG}1F@$fF`E+bAu?Ai1;bf?n*H1s+_Fkqwn(wqx94C*s;5e_c%2Y-4Z^V_hzY0XRi!}>r06y-Cungt)tX^0zq1<0UFVKlA z%*!BgQvCK>3T(aTvNPB?KH2c=0>NN0yFRItFbLfsbPO=I`I9l>+*un~#zBhNL#Tef zSO$3_ezTj|h}SZ8!Dcr&SqgHfr#1lJD0ywVP&S$WjrMX<#btI{sPXU`xkM#r`hce7 z<*?{Jld~BibIf%xcF_I9ws@!9fW77VHFM6M88*e+&i5QV7PB9aLYCdDAAZh@)4Qf^ z(e`!TSa@$5TO6}WBN-!Z4}KpPJ1Pm`*Mhv)oo@)~O>G-47^~hv`s6?RuNvF}NaLBS zLrWnM2jX73=>l;6Qq44mq!@aIbRV+643f8@eWT<6hw2@!4`$=}@AwrEa!Rvsnbro{ zCGsQVJ-F%-8srKvGY}D8s;PQ zfs1Ig-2@8{fg#YFJ%uL7s$kH2Q_g^*?glfpVbS>v0TmAJaQt^IUp?}kS4CmT7OZlO zU7qlt_Le4|{&r`{v8^d0dVH2)9m3-;M8siNFUN`erRC0f7F%^av$cH1T9kU5)wjd2 zeerQ*G>Ge>hr3fGU;O5ndj?|i1IqudIJ#rV3xDyX3(K3mmX6f(2D^(A+LFVwp-kYQ{!HrJl4W`R;8)#Z zOqBSA`gF0NBZ%%X)*4BG3Z8|jk0Si5%d>5#mK(&OPL3Td8)thMSgLqcpb zLjB3)JawX-Xc?TqibXd2)d%PHO9v9>{CDSV2&2vnc&3$?w8d}YpguC z?q}>dvydVinS7${v;{C|>c^s;l##?AYhxQ{aUl)Pv^HoRY~17}AQ{&nW~uaiNTq!0 zHYqw z(|Yzg>y?K^At>-_ta@u?=opLgOMYNuE~o?jI4e_4g626^7DFpOFXm`$c!NH#Pz+pF7*Djdx*Uqukw%+@*j*%d+B}{m`;kZICjy@RBwJWRFD|5Uu z6Q90b#H!%fc2In5FYmsvD)t-Dc%3iU!xZNp4zIQ#$&t~e9$r+LaKu5_D4zL$p>!*A z67#tkngb<{_Pao=rCMCMTb%g_J)_RjMxgoa9>Xd*i+D#OD_-eng^6hwtG4&`DS6jN zK9n)9Q~L)a6I;#w$65; zs5xbM8QXW-+`#K1*Ymgy-Qysl>)TttN%#ZdqDxVj8-n+S@+big*{xLX=Mv|G1h!nS zV7VjWwbt+^sqJk$f;@ze@T`%prwBsqC#~)Vzc%8!A!A7-3&Z6IiJD=Fb#pK2X3Ni7 z)*n-KUmxWo7aPz;f{6o-p!D z^W*8qJ61>ij#L`m4#l1winE~gM{`N9?X=yy&2FY!oEjPr@_vZ$zKAsyx#+TAsWG0J zXp)=4T3KK#MYtl}8GcF}h3=7PZ9JV%YIpTpN^Qyw{A(_#&~ziB!tJ=!ap!r^)PK~l}S0qd53>1IKG)fsEgXdxn*y^E^)O< z*DPwL+cs?m&9D2-dM;|kAYX{Wcz$@|UD41G04Lwl07ekwSNIo5XRyp<`@IjD4K0MMXJ=xitW8$>Yy((j*-IZkh zz;a`l$2@b@Ya&E~;9%%gop`K~VgBmr!d^WLVwB6%)Dp4xzF55?HM3)+f8@(cHE82v zt*U=xW5-s-Pv#?gH%!=<%r*t*txAo{pH}%1?+-R#R0pzxn=h!)u$%xm7>`~;nRYfh zO;Yti=ur8?n?{9(%3MU(Ws{duPdLRZAU2P&Aa8lRc`l1CfEti!FuTk8-q^-=-~eZs zHbAfrJ&;0mrL~&O(Zu(Of0#1gT)F^UktN%%7evTaaC7H5hvj zpo3|*pM}l2^!>q9!4N7)L-Q>)$goj8GM+`zO`Tii7lAKJ;9-5pc%cS(pGOjS&N!0z z9PS?O6TENG!E?bM<*#b-x<#{GrWL@s_EnmfdO@O7DnJt^K*qh|AV!Z%-oUFKUwPSw zh1L+q?rgcn2{L(6>3jJayhkTdV1xal5{K^fM!oe^&QG(>5Y9f1bhEQUK_q#R@1xD@ zpk=`TbJXzUqc>kXLzL2}5x@HO6ojN9GiWD6t=^)jpy`7YHc$G`Je^qcFw(VZxk1*j ztM-2Un}_K}_((3M68C!+Sj*KIbO-}}wVMJrIWIU6imUXRw{TTYXhZR^5Ic47rpuZjwTKakvBoiQ}9#hAs~P#%56uRVWfD9rG?`P282jHLF@r{Ql)ahx5+jrQkj z*@k5hW_MrVT%EKNmOu@sN=xPtwsOljR%lFa{%1{s4L2zQ-=#hyrE(o=%FNL$B^Ms6 zQ$-@(4oyak@JS|ro8E<+eZ+tb+q%)lfBg}@k3)->g)TXV^t9I>b97zx=m@*^Vt6Nn zLoDQiUJmHxlv`p@hT~!(lbZ;BV8_1xw%ce5y8}3g))34Ema;w2-+xQ_xHjo z#pe}bNO0yat2wBGqyFi!2wS^XF`D+0nE!*2HiqIv4hJv3AJtrTIdWyb9$yP}f`%ys zD|a3`k{F~)*>L#eg90C7*2O~vE0gt^c0KKzseV5hR;HNa!^rqTKcCsjEv{R>X`U=# z{;^2tu{G&{s#t*!9?bFWhI^Q&5oZS=!ep}O2i&!3TKk)Yd%8`OK|&T+k%_y5S%8q& zJa@V(ZWf=q_~kKqR+r<%Du=btV$H47o`0>$mc!&p=5t6`W}4xEyR{2qkyR=va&){d z$$CHZ^0}V9w@t!kQF6ifp%Cs+YBkXfS_f7>(jT9Ns3xzYOA%oqiB+~N_rurFuuJ7BudvSu|c5zPhc3XVpbzGX}Pk4DQ|It4O`>_@iGKWzNJ(iCKgyp zVL+RdDGPIN=V!^luE+#tLdJrJ4L;I{V1z;f3&Z}qpqi~es?BwAQC8zQHu?*Ba$iny zd+I_<)D8Gb4LD@$DFv*O)>`Ic~iaXbW`svWk6LMK4e$P6RxTzr0Y|FZkHp$m6x-D^Kljkk0k;r%UrQIAv_D zR)d>t=uNR!$;eP(3`!BlogkdT{mmerx9--2*BRrW+o5GX}&pO5!NMVLu`=EzDw0$8DWPl<=C6eU1fa_ z86gM_i@vM=L*ChKtOcC_<&%f+s=sF996g1Wv1-2wm3Pj$Yc&!=WfKuw^xQ#&e2v1- z+Q%r-y%TFLS*m%wro}6cCNvlEGaF5^$@<=l2!+r|slzEHqcUUfAV*cbJ=9*$2?jWk zM_A;V4K7X${jEb~bFNLR-m8~HAM2A}Th~%8*C{xky1TcxjZ-?#v?YiLDlDpl$^eEK z2uSn6$FkehN!e1wCE*eQ7I~;_{8;{vg?dLMZg*2oF7Y=5q%PiMBgRoO3M&&{zsyqW z30_;q#gj4VfaVPi#TVd+djA;+!rNjx*=fM~AhWiy6jF{b zDwnT48j8o8bbR#aCHAKQzXmn>2vC)^_Bx4kC$NE5D5E^zAsDymmFFP=J_zL9cu%1E zY%!6e@F=kES*QI(8|gFCuzE)`C1s);`tQ*_B!3~rt^lOi0P8!HskR-6bnlZC4S4XJ z{6(cnPM!y}{V5SNVp%sh+YelT( zIf|ZJMcMQ2uVokSqyuQ%vGUaACzfQeX%{eF7c?!KHv^r}bi=CH0m;9zv|5Fy%|Tb$ zL;U`PT%2oU3^a3~LHQo`jCvc=xKJ|;={~H<*4Xbww{6ex|8_kRVg#f47AHb}W&&t> zrJ#cxr@b*<%`%<*BeYR+li_|&-ceEKAF_KK;~O`6GWSfePb7OMe25Qp@6It{q%Zug zp3*(uZtkBDelUv>#I_ekrp6@Ce1h#<#h{?|<>6w9)bss!vg$95Nd>;0R#kzxc4FNZ zvu@vbeG9?kj4S)0FrQP_Y`*h#_?sEAbe@S{-HD~eQY!&34c7Fdpsk1d;a`=~cw|c< zhfi)6kkq6eBeq6286{Ey{OV>PGqOy0hs!)i0}9114J%MgY3RyTq6&Rh7Qt$JjJg!voA@l z0mNW-AAlh1QT*n8mn7=dcq$;CDR5VKSvs2dq|7|YDAHvyxJPtjxMk@P5f}1lGu2A@ zL8pDWY3JoTnM26 zIfyxIV{m+HnHnh?@lr0U~PJreA;a4XkEKDPF~9%L1MQG znZ9-eNp#uLd;F04EPe2!y{uEYl>7S?)ShM0aEC!}yz}ht1XK-!T_(h6f3RI>HO zPX6VYgeq7yn&n)J+N-dsaxreOeDTok$zyLmbEi>)+RrXnr2BNtGIE6_mE;Wq>ZtgQ zz7c!S?)}9bV`3e)A*`e@VqE@dB_dB#^oREH8dQLA-k*Q-a%Qxs3iD^R5x&w$p;0^6 zRLdC^nOJoT$m}+-tu<%;#N2hhgIxBcdMi@b3Ra_lj%ePlx`aJWc+U#MPpt|VlKLq{tX>lo)q@l(3`0@FW#F^&Jtpxou4Hp4&z#7EcP52q5LMQ z@p25jGxanMj}#(L<(-p=`#21JlV$R)b^90NW-q0MP9v*TGXX6(6iT7v8e|AtQwN#< zlaU!q>~uqIyrw&7_w3H3(^{6|u=-+n671;)V`q|b3!tX7cD2{2^ce|5xb6~u0Eh&m`3_;1y_m%ts_}d1 znAg*(maMM+?S~wotV#nfzt7*<{ACNI7J}{OmL8~liX$Rz7jbo~|3SP&(`x16Eu-%B zlmFWRUi-XSSmM^N_tS=)s$Ea&ods1NqQR zm>ESM1)UCQM^#1CwTC?X^T4^ckhpm}^H zNe8qUkok|rLbfr^@c!nFRoz}Fd635uxdlg3#y@?eZJT+Ue8r5tOzz&f1npr8A6XHO z^lh`gnK0z9R;3}Od11C7#GBi4h}IcKw_T$ZpD4ENInn#aXS();5$r35famLak-v-) z;D0B3F6$Jq&AYwjL5iqhjM&o{^`HuwSeK~st6dS@PX6i(;Kp!=_keMG^ES*!hz$~u z`m}2ZJNnBXb09Z%3;Zb9t=kaYY;PuQ_K5;p9)KGQvosy>CeCL6oyZ@BiyrlUDo>1W zj#1Yv*~!m4OaF5ov-x64%s|>*&LWYs2B5pw(+^bYp>@f`38Ncb$CzJ^=(e%2KWY6q z)j#GY=D6Y%2bew7dBRDToqy!d%_ewv_^mYk-vsYmk)93a!L{;I_8zWZO7I%+?&;58 zS$~9;p_M+HJfTltAAY6FMoNPz1uj)7lRI&}GN>HZJ))3GyZ+O<2ryhDLR&y&@6l!H zR#fFa@0Y6-{B6xB$FNEQA5l%Ig}H8;&+e%^%6sbkZZzNpCbht@(K{6-9$>sMtt||IVvQf zxJw+MfQ5owDP}Qx>!$+af6r^i`#VNCF4mfaZTJF(*E|YpyRR_L77^p-w}7mrF-?bb z7bOHs5B#fIV;!-w`sp@5!;>yMj_E1}kdQ(wb3Z0{G;mTeuK1$mk9c{SOHc}Y&Wpx< ziN}+V0@&g+{H9}E(P^8%iDM=vaALxfy-zttgsE_@>;$sC;!)1oj~G4J*L#y6Iu4## zBcNp4#}1~QXBKLUFuEfuP9HaFMV9I;Oezrh>5?&Qwh<5gAYPaF1o~b+G!wd9H{|rq zSR5YqDxq!B&Cx)bvlW@!C-wBm`FYj&8-dROGPO8(ZQo6bEF=htR;C7rWrYR@T{b=I zFKXj%mL&Y`io5@EL!0#h1uLmhN?6~V<1)`rn;KWiJgusZGG3wxi`F4sdTu`Q;)#e{ zE~#Q8<()|HTS4NdF+me;sM*)Rm}VgaJl9L8W>$!s-*Z9>aB(Qx$^7^rifKV@Sgun= zWzEMUF(|~F^{0&vXMC9+_XVTSYkFGf_l>LsZcEywYX%ky7OKb3}k{j>^EmqRwSCxO!vg_ZpvD?AE3ck%Qx!B z8#q!2>v-F%v}Vuu>yC=yd7d3^)`}L{m5~upUv<6)GMSg*zAy1_xz173{P3?hIe(ki zZ1YTgdX=u>W;(NcO-KZ#6*FE+RI3l3HV%Bp>)edS@p68y)?MGS2^fat716RcxF?nQ z2j?gfuJGAm&9$QtN{V-mvXN`eH3hSHO0CURdz-M_W6E5oO3I?JY}eq^VzJvIRzDr; zo~YufzATs-R#DXPdx+=l60=R=3Zid8EKD%~@PKA#(&&+28G940w(|-ywy#PaD~C*n zv>T-hY9ou@L3aP41z^hfS*Wk`TQ1hI*brh#2N%zAf}gM(cY3*JkWb>Y4&7P^-c{Sr zw4HcjnUH5;u&muXrNh{A;O(+Gp)dF9HNbS9Rl_nn2Z0H_Nw4a00A|@uA33yvwNE282vnby)Le(q#u2|?8GwtEkihc#1PB#aiGQNKqD*>pDQBp z0K)6fs#?be_JF<>It)F-l`d+3(3rxz7lM7>hSvK8C|a6j*=06wQ-qN9N>do@!X2V7 zpCffwyRjoEXlGoxuF~J`2vH_?w`yWkhI}w2@3T1~MjX=!y>i99$CaIi&G_t|Gp*SZ zd7+x7H<+efNavJ3KjRrgGLh~Fm?Z+Z$Y2xmQFhLrcX_{}cm^H_PGnqxU_`uj7(1hR zJEP#tHB~Cj$zqKsuRP5Es;KULFHEcukGP-4Q%3cD z1+&!(Rx%f6a`fb|wap<@0BC2ig3#tNHczR6y)S_{clswxUfCNvvcHju8uAXLo$Uwj z)OnF^{S$g)7}OEb8^-ZwTpRx>?kvyM947WuSwu9@HbACFWMhLu3A(f_BJYmc(RjQ` z+hrR{%={BmBxeZ4F~K-;3-etxpLw6B)pq^onPoMOfFBx2@#Y zRWIw5L(?F~huHo&_Xy;cCigkxz>JV^2`o1CBqwc+^I`MYz=3}(Z-xtmC@ew@}3_uX9_Q*-w#+Fj{wU{DiEuMEi zGkYpLD1B%-hIg7Q6KTRhXf=*n_k_jUjChos^ra`0Hfd^8AOj#?Q&yLjK?u;?sPhPM zIZTsu|I84K+P1z&#E#hm=3_Iew^U%>k z!2o?*z|9s$McMdxZYD_3dGsy1$}W65&+X&v&H*JI2dW(g`KA@u#+jsbndNMKI`)V3 zLmN{enuBsrTiD+38q6@?IWB~WOV7W9oCWheN8Q&B(EdDxG6ztkY(y`2+?xrq@IFO{ z5*y}+BFl>Yf}Hu+sjUW`Z6l2;IH?|Eq-|1s&ct)N&JYyWi*;S}WTtDKot~I!_=H5B zQ94d$ui|X*mA&dCgUAxFj4RcjD$JEwjuvG@%1wUkG#d-uq&}(z7><>XP+c*~h2yks zpPO#DZ4jm(@27ti479C^fH9$colvJt;B<1zcKlQOIRN>&pXSgA59;PToDh;}2`7lS z@e8z_M2X-uyUC7{#uF*8JGqSHS>hFSobnX?+O~)*N!SU(kN`Pq3++rQus3Lnqf8Y( zBz!q;tbt(&eh`%=1_j5*#9KUgw$U4O5!$)w)(sip8~|gqiBVyF@LmwAPV3=>Ka6@4 z(CEUg>l75d*EL~nz`r~PZ4s`u{gwUb&(*9IP2AKkeeP$=a~#!;$!EmrRT!=|KaAG2 zsFT{1PQA|~8?tB(+f)6TK?S$TL*Q#!o}{Cpep5}Ey?jm)=pc=s{7O&D(oD=<<5Le* zp$(~f?I=e=gpqDV--3rbgf0;%-Ncvo`-wL3$v~*}HPJuDmF!3O8`6xGFRy;ieqP`| zdQoe{scF6bGfJo%3Nj^EwDlKhofVsqlndh8*;Sb^2GqI|e|%dYjaMGIiAi8?A`Smq zsHK?icLQF0fNAD#ztbZxJ|ch>KCN&N4Ydz`EC1RYav|GKR70j%ELB z^zHtlpI~kCEagVhsaa(^|E!7|ZrgjatUbl4zs=tA5J9&n=*WYl3YlD8IvxWj=s=Ti zZe%_wI->Y6+o0jR;J?Ly51L6D1Sp7!zw8oAo9sksl71+^ttz`n6*0*Rr~2`f>Z!jt z=jJ^-C|9GYm7>Q6>&4hqhLD%?3w4p9NkNSeL1jora)TAP1_x1z@Z>+zCD{-*qC;L7 zOO_AM1KX5^|T;ysS_eujs+Iu!{T4k>Vt2x3705;S%(pOh#gq5uxt@lndUyPMl#U> zEiZ)4djg{L$|Nj$+Lsg={EK3Uu6Mh9=|@t4#wpMzOjmq!Xq9h`1zQ@I)E|e%CmA$a z=}_v^e9t`Tic^zTe3b&mNj?uT==CH}qEBtOQ1yIr)me=i&rG<;4CpR;{ z;15JG9xq_P$L=lJi%!Yv^U9Ckd;n za&WF(uuHUi>?+$m4Q+UtJl=Oze87EPH8)E!S)|*VP&Bhw8bG2|?*y@90s2B#rVt%k zn!WeVjRZU8xKv~NaoseT%}sok9PGNDu1|h@J4bq@EgyV*Gw8T&+P(99@rQ729%LjL zflc3X*~$u#(d@ef8T{W;uj8iwMN#GCeTYUcFo=8e z&2AST+dt*z8oN?OmH?^*;;P@DflsmedIaO@-D=vj|OD5!5qKtz=IH=Bb>gCJ)NioC=ZZuvw{q;VJqZuK0JM;6p^I&oX5i=0qm9y#a;qex=*io< zGwDG`Gel|2bJ{2VWL!(5+N8s}y5}9wWH1P$%6DA6EF$68)pM%ImMZ?&nG+3Z)3)Hq z4Q+p7$e0~_R2{TC4#1)LYe)Yhu5)+6b#h2Y?-TA5rBf^c-9~BE4)zlN#X2H_9xA7W z&t<+A_5$NGpa~m|#B%*rH})086^k|BX(`<(>X|_2l(zmdc_=2@jNQJ*WSsoV3f-K1 z)@1JDcBO7;Ld^(5$Oxp~n)j*qKZ3Ye=DPXB`!Mc1Eq+kX88;v>UdeEW{mgj-y`6u+ zmb}ErqU(w03v1ws9sql`q&QxTr3z+8&NR9^3SN$@czEv}MXFqp*7#ofy8ne!vspZF zq~`Xlhg33k{lxB{mtNC4+lwD%+2wg98i)k8BxAl=EPlyXA{G5Oe1n6uaC2%RgVYVC zE<_b_BDM5m$I1@dZf}5?e+j4KqP^oP<2j#F_CA)xK@? zvcD%Uj^t_&`nmbO<>=>yI|)!VFZFXkqveF1M$FxCEL!_8SzcW4>4y+|zJ3Q#|RZ4@zHkV&7{1|krHaUHg;Y#Vt+dzg8}R?-Uv;NOP#wwLv{wJ?t;s+tu-({B_U+Ui2YZC z53b9F_agQqB1S8|;c5y!}|x=8cgZD9FY!Kh9^kWG7xr$1GNsn<%e?{n;Aun6QPME#34m z?}P0t^VbfKl6oZGClWRO%z>h(sWL;--CZ`mo0I(6`g5jDXYB=MAaLqiz>rMttW5uK;5@I_lG#!s(8$)0Yo+S8IQc@xjqPOjbMY?WGuun-N5e=Y~ByyOAkfnl%)CiV`fyOPYI#14ST;W zrR8fo&0p~o10CqWldcK$j*6Z`8KSy^v7*H!N*az~8LpnCm!E-_1?rzYdd?{rn1j*&MnPriSx~2HmQ^#gXKS3>d>psKU*Kx9lHh8if z`}Q0+%$J0}YF^>IR=CwF3UjqLpnQ5Wj@#&{IlxvIyDRl~68X%qqPBGaqjZR*HtDNf z&vQyxL#fu;(AWp}=WP(7_o-K%mpIE|&k-^fs6C=U1jj<{X@n+L%-Wb<-M6`uz71iD{iEWETVze?eqFj2VMKZj3p z>8N=reMH)ZZGnQ>wPY_3ex2$LG$G& zeU2?kP3)J#8{nfxsuLZ+@ZUG-=!)^04!j zx*dAm7yJW#i=cPIzE-uaSDkSbNi3K_=5hN}1XD-E<|_@n&X;;^p2(!}IXEvI|}SAuQUNa<=BXoKQ_otjcE7jh?dlx$=hotKXu=byu;WE6c?CotT3V z1pyfc4&?&~`jR&P!UqsuQD@v+(`d2;ba&GLpztt0B?mHE^8HOxk4y`C$&np|GhuB8 z7b=61)mSl&^BuF!fp&I1`oiMp4y6|{n*)%X+c(D|cpdbk9y7bcDt~p++CI{WERu-x zxk=hLGIj^k8wKdFYZts%jIy;N51QuRUdcPlQ^e9KakJl=6*hSzIwog8*Ez`{c1I!J zb!s?;HPI4G?I91n*&uH?sh#^UIBs2AAx`Vmk*AniusB!urNU)M%_7q7L~OHKf3B{7 z)mV+rMewl#y=;OysGd?E612R99cA|C%{p(Sdy^6V(bd^*Var_l0;vBasOL^e%`wS# z0iqN9k~&9Z%9u;B$b_LhF+?>%BVUp^d5`upG^WO3;YFr71ouW;UNlzQZ9OxB6? zv$n+Sv^5U>f7m;#u(-Bu&xai}I3!5$Bm{y>fB=OAcMZYa-Q9vagan5I0t9#W!rk2q zclSbjvCqAI&p!7%w|BqwLw{dAR25Qdt{QX9IfwjzfuA^Tjo|^5N*;E`Ya(xIvP(krP^*v~n)4kx-K+sqlbXMi+XftEa z^EQRPZO)BwlQd8cfjiOvsFc9PmbD+4`h}vk*5S<@_U&)z9_WA5n8MAtf`PZQ6w^lr z=ZNY6@i3PCGB1^N(}Xob-4_R(>g{@P!}j4{gQl(T-9CVKQVBVmR~6X!>@N55ceS-h zsLRG|)S>y(R6Hv0te0pv0?7E>E=Z(SU3VBqAHK$~)8zb!+)I~yTqcD*J=0CQRK z#*P1ZCz(B-z;QFsbmZWNy`|J_Ck?(k|FBji8s=GOAONOPM~sdN2kN&OsZl$vi^vn6 z7;f@8L=}5VE{eLT2XK9RSHBk-_oVeE^Rx`JOwWQ0x`!6+1f%ynb6zmPTBzDeGcDT2 z%7K*Urw3PXP?IVP$pUc37xM;qPb|(4KW_|=CwUl=xvnCDlI>+QUqG4ttv6gz7})^H zH%bm3K-{wz!!*?RWvUX>t!h-O^|4e$_v$PZ578oe`pDIAixQ~rm!a*sswS!LtG^&} z{B%eeX0+o3Bx{#cD||3&qP7`^^FK2hwWN!W<5f^e6MS&~TB7&$_l?laCyW6jU#Qa} znYoqO?Ty}xKM8InZmFO;!~U*Gx5Q~a|6f7WNZ2=i&e~I!_!dN6XH%|HLPP@RY4O@# zn(PHV@bB9OW0kci_NGCZO}c};K)hM3TWZHo>;rNI*~IdS_jSY3dZWcG{J1D#MTzvA zHs|sqz0Ze-?{QL*w2Z=hE|(Q7(xwjn zFiyoW-Qyl_u|~W57Q_UCa^#5_*QPQ#R2R#SlI-`$y9EqG!oODkv?RKTwS+isF1{~< zK6#Z=z^ia!*1UNo+q7TJ7ypzEnBq zs+}LjpUn1C+v`EqJ)Do% z(=ESx)q^ePqMps%;fF2?G&$8<%P5btZ5&v^pNaRNsQJr^nidYxB9sIspd&j=o*&2?X1MEd)->ssYCZLNC(uwTqB9W=N z%3!HBXBD)CgxUT(e{0-q%m{xlW3l2Ho_Drjke;`Dv-DjxcLnowCdrFB4PiFX`XtmP z4DCqIlt4uIHe3O@^3KLm6&)}B0bG-ug8@INH3Z`&_r zob$ohXRNl-K|n>Cj)>g~Mf#du09AmFL7NMD@ZBwuhw7K&wbepl?4?^fF{B`NI(b|W z6{Rr;QS}QSZ;jgwV2FlTFiCUZc(G~?xo1FbP^q6N`jk~dm_78p(y6K5Xj1Q~FKX^= zM}|!pF2jjjWApSksspu0&2tw=G<_}YfnoQu(`TI;ohNE)aoRPFz?PxO5?JjoJ+25$ zrAHU-_2=a|`Hv1ueAf%swrFV<;rnrWke*-vgmn^Ky0d3ANC=|(V z2j%y=aK?2BN?P?gt+3E`Pg!rhssvF>G%ivcygqgov|#WZ#>X`=q~0G~g6W%)Jn^X1 zjb~wz7bK$+qz9=iMSyV0$Jiwv#SCs82q0tGcO1Ruh?7uIpA-5{pJ*sYs5f*@F!LXM z4^-}s6IV}40UhGw=PZ0H+*2MG36XH)EIRyq+}6!6PiD49&C}^r>jlEDR3f(m;wiFj zs}C9!cyQZXmfT2i`!UHLWDwB_Da}*ewXE4)bQn#?aTI>v7}~C}dQF{3eO_GCy=2cO-jhgwAIAi1{l^Vd()f9FDGo6;jl=_qH~wSz{xY4fQ{K*bP%=u>cD>s#W~m9D8`Eqc^Q_wLK>Y1v^V%c&Jydo5Yz z)UkLir$jC11IOu|$s=0pEx0DPog;)84yVUof|(tAfK;sn)rNHBy{^nkgZ9YTRVc(o|M2j(Lms|GSsIRFmM^lVdj36e73 zrVcUheun*2 zOGPC}0wVxC6!F$!VtF{WT+G&gQhDX!s92q&oHy}K1PA4KLxuen_|qN&Ts4|E?prTo z_EDmDKyT+4QeU9TCnfC~yxq8TX=?Ctg(#Kqc%} z%(O&Msof8M6++K4W14t-RXHxBIR45aVxYmw+V4KB04oaXybakgvQk>~Q!zCAg7$5y z?xpQXlPrY7*JhI`=U8()aNZ%Ey(FyT>nwj&4A1ON=p`C1I+q<^;_b3l0!)J6{X+o4h~X{8m|at^>ExPlzsW zXc=E*o7ueXP(yBF0=5$lK%wwKE?>fHIK_9@v)g4q4NhZX1Twz0Qt+; z%>8;i)&Pd!#UqT83Jt>_oH#q5+ga7XUPP2{Pk#H@m^Cucd(KCo$^tSM~ zQ!+L0pdN*CsaKj9$!wPC^SMTufXhKHdX+yhb|&0@OEPNTI2XQIn;r4Z_&O{(EzFG8 zl_wr(kgb4(4*7o0ky+uMbuJ^nk^-Vb73d~Im}$p2O_Jyxew^H9cb^hl=aEm)8I5bM z1hPV$=LuQ2c4S!4<~qMg>YFYC8E)pZkjjg}W|Oo#<;lzhtF}QVOrDHiE%QXrP5jhE zXy<#Y(eL|!R{wK1s&kxX+orbMtMNUM6ilL0i~aWO=YM9lMYV8XwNAy5oLU^nKojk|+`Nho0v56Lh4>Lx>*lv`OgEBUEG#p_N z&Dy6m3^oarR2@XoOHC&gp)y12xO`({wXXoBjY%qy-~8?$JM@nQ>AiiY=$V zY{vbzSuF_yr1+h2A??SPB6Ebo*AHJEUoJW|W||xp_NwIy1d)L9S5yy-jum`rUijlU zfHGFA&`I3D4d-!_=2JVJz&=mEOZtbnY?ecPwe@W}&e+_1&?Gdz#N}(fgQP2yuFoPT zfR4ij&~ZQ)m##bw>&hjQEDR{d%~Fd(5OF9{zOo`@w%|7)g5aA$9HZeU@FDoCOawFC zcDILA9{TRb2*wT>ZELq9ze*fy%^e)(HUaj$0}Lo?-^P&euhD8WslH?>zv|X-L0*d( z0&sEG%G0PTJ`n+}t4}?`a~Daw9-sR2I5GNI`{3II;;FU1xN8x?fwtM$OgnErf?r`F zE*HUDGdnox7fUsaFJ$R!({g{z1AI$>EN@XiPje8Z$9yz}6@%?hpdCgeYczbVX)#kf^8r{V?Hp7WvZxOoPh)F(;$D0UJtbv^u0OL$Dwn>l)vp(KAOgqSw zz>rDI)3-61GsGOKDA*N_#+!uiCvk&G9Qdv>XbdPwR$Qb~3F=w|Hb%RjiLcRz1zZST zpRMV{V-BovsSiRdwZ08JXpyvIXvxB0THxL^+4vc?<(D)Mnq7OMe_&5QV zi>i}*CVxeQS`Nt*Esd&(czv##VgWWh-Mb257xe4Ge9?$E`l}S^9h-J-#JcS)za7TTy^^1a<5asoKZgz|tX4HmhZ?CaE($dAQpn9BEWL&e-MJb6 z%P(?d+~|Buv8RELVH|iHRZY4?HBH-!=O-Kvvn@xGbQOLs_~|ZCT5b1pR!w}JwCsi6 zSeLO{4L~$XJUXeqDAR6#kJ*>nRI*zS?-;7M-rbTOs${+cx9Vs%{eASUZw*_tmk7;Y zJf-`e!P=5Dv1%=my3BtRZNB*~ReMHm@6m0q^~96|ha&(?xtmOYDQC`@Bjr4uugMKa zawx2qVkNA+^1234m@Jk3Oq%F(W?jp=x4Eas)j6iVE_0$N`PyV$2JB%T0xtu%=mp%K zAV{EAGNoS)E7AuagDWm3K|`#u!DvrzPz`HUlvei~oejUTjCBa4_;9y|d8WAhc~L#& zCVq2(P&KfwHAgnJaQ;dq*qNuYH(ZoEu63M<6k3Mn_lMR@t-SS=s%taUW3Q;W0qCsP zNL4H4w;q>6$I5jD#oU;eG``6U6uP2XR4s`6Xu70f-C%R5;c7LRpId1e^(a*0&LUt} z2)g|p0r#B1!sm*fpzp(yaKvXU(eq~-eDil>9ZtWHp;XiZb)Kn-7eG?{d$C4i8wZAY z7*sc8NsJ>C8QK$u?M9*V%kLTZEBC~m*=4Ych4BDh6O*WY#e4MgyCv^OPpN?SUfNhJ zmWwyi^?wK;r7Oyv?{yrjBMm1?C>>2{Q68t7TNmX?@MLN!~_o1Fabi zt^Jg~#_;p0i_}VU3TmG-PLp}{K=7*u+jkzMLoRZ(SjC**j%Rv{0#$*m?`(`^I)^TY zltQki@_x5*#!o~b4ouhTjN-KRbA#NOp26vrTI!amG#b>M@BjP}_0%g{EMeipf!TKk zs0u_2%QD0o$K?!|hg=}@SPYar@3g=j&e<`5pHH6n_H1AgXl)9@+}TP(<2tdf%?k&V zCos+#z;ipW>CtC`IX7l(B^+((sZMHm^oM{Sd&~MKRIc4K*ax#FRw}W<6X+zaYC+On z7k!n4yjr7p&HfzUhVg25R2VB8|5v=N%-lm>AQu|C{acLd zNh&ec*{2QNmG!Rh^3u&9LTVzDL#49?R06e!Gxmz3_~hZ9tgkZB{-Vj(azr!SOhKwd=iQ*;s6?uHb?E*0 ze74TK#AyFfEl-t8 z;`wg!8@XRRJo{za#!Vy_@V>g@7vW7b+FG2syAPTTS1&rW(0;T(d!L{d}aX7I;2{!Y# ze?mvhWxO2`(1^NXGtY(0KmPDsn#vzn;oFi!v_-2VI+1E^L0{KBolte#eP8)OYpF?? z;nQPUFm|?Oz@1KuSxzM84*MSG)*V$o@f^(YHc}>*{h&bzkw~Ydr^(SA1%9%hu^;e1 zKIzoj!yoX1JrD%piwMo~UkiOG9c*R~@#m7g{4;27uYS5tx7bxa3L46isEls; zU_Rx&bIM|NO};48**sqj$?9ilTWiMxmX{vOvo-%cpH{L22Do7~A*^~q;V&QD^mSo( zXN*csXTAeT6><=c%|Z9ybm}aOW^eaUFdyh;IB!jIoYl?!3piGR-l;#Gm6t#DPV~>D zi9AW%QiZoby(Qw_y5sPI1X3B*e~fv_e8e88LXVncqRTzLuRGm7EN=8YSqSPr{zTcF z)J(YkoOD<=Uv9L7#&*HQRG+gkpx{eMF;5ft2t@^R)cvy3wESq0<3y1N6h`5YNw&@W zikAe?V=3G3ek*Dad8*^3iBjtV9mrmiFJ@E_+NEX}n+~qm1&)>spjwI^YiPMJWKD+< z4b&Gyo3o`A6ukQpjaa4GArHgVYU=Yc1ZcgwxCp1z(I-oxYV2k1o2KPA1 zeiggyw=b-eABxL0fg_`pb=NmhZWn}<;oRBhnXgo;mIMG_zvp7Uo1)MRE~Zy=903baBTKo`2Hl>K&)_l2ri8aI3~iLe2<5+rw(BMTd-ojtm&|l_n44D+g*A!e+NW zzXjR~ZAGrk$6(j4B_i7CORq)-sYCXr)uYD>X5<5XSiSxMgMu^Z6Ad0~5u2~GyGWI@edCQJ+|97kTJs!^9H4N_Q!m5J1l8t^7F!1I+@#_Z1 zE_t_i2N$MATpU2}<4D!Ij*N5}P~5KsPRxi%XxXM0Bkj?~-v$v4aWnQR%t*CkINhFh zhd|HJiv-jc!~DjcdKKNyApGtK!h8hLEv`EeO{Lozz=2k6t5-+;J==q8Z$3B0155hivC{$=`ojZn%fh z)Vojk6KE$MlTV&cJ-vLffC(nE%G|q80;;VGTJWBagwHzi36M=&PY7U&v^QfFF@GB& z@V|iB=ZLt@CpljGe}a8`eQdfA0ogytN#^SSv(bze{r|B=JIFdzI0*ZLNLc;uml$8tdR zsAb+`cKpmCi=`q0Cbj3}!x?$Tx>U`Jh{R^Pq;`4W&hg{roD$^WCi$?I4V8JOu|_Y# zNu@DPzK{JBEcKMuzb#^I9J>KC0VeP9PVs*`tM_UDA&?-l%<+o-=E{$1|CkWg0x9si zw~pQ7;H}u|`SqyRG`{S?L#OA_NXV!qkSJ8-A%9YyG77hgRT&u}j_K z`C61+D$e!F2MbKZ<36OJi&C6SS>+?4y5ib5*)|8UG70^ym_zUc4$Bs(t0afFN6q6$ z*|phL;XwATHwDlf(ElCg$xQz22oFDR-{VkZJV!hv@gfrj3C$ zz}opc7LP_4$FBK&eK^Ux*Yv7au zdYoFn{T!t!kxWx1ob*A7YA(wR6a&z`0Mm}97&v8%ijp=uj{J&5%wcx;q1UYHQlaCW zkPHy*w)tOkSrqDoA`);lUo|x=a(MYV(E$aDKv_~3=dZ{>QL^j=rsuCqi4mE1>-S!m zL{(H{^5nj*%@d9zC&uu42;c|7c5*V^m`3vr!z!7=^x=01L~_o6khbFl*1kR(Bf9f+ zA;ZB$I+FjHV?`x3#Ht&`j?|RiWB!chYI{mtt;Qmirr=3kmf$!+s3VAD$wNKN0!zT^ zYlfKLjUzmTU-H6qQd^CZ*$-g<^b%e@_=oyL%k#Hyk_4BV>9gMK=`7=tPgy{mq+e{Dc7k(?^6bxQ9)Qq&u0YMlqm)I37Wgr+ zwLe(x*d}Ny53y1wnLOo~^u;eMih~0mb{CF9l@%w~FPg?etTq*uR`|6*76C}?We$Q& z`=l0AyPfrGgG#!0-kpXA}o@3Js7_F_ z@AKrZF+!MuiO8?ko4$s&VzKfobZFEAcAE7f+QgD{R~eZfZF%KB@kPT~C%*R1s3W!H@lsTGn~u*Z8XVSA)$to2=hs}t`T^h&sx;F&dM;e3 z(jSN&ZvKs3{4wWD4j;YlPT&s25q060*iHzzB39Urn|2dKn(gzx|0?V7GQPn=mqTP| zTGyjr;F|8fArT=AcMNBm5_|{KSfu(iT~OTj*7Zz-a6FSKd<{|hxgR5z-g!=@D8kB2 z$rnXgc={l!{5wP*4y*WPcqRhGUw(|kvHgzxqBonH6Q5ppsH45%aygdd2}TCMt`=Ke zo~OA{e7F1Q#GI!;F{IPf16;e(v(k+&A6I&!>VeWMoZw_rU0|g$jZSkJ?`VXezpfc+ z;JKf$@Y%AU^D_x_t-)5qJaFQsb1C#JBVD>jQsRj^^#$r%Xzg=OX^thSX1C~1qpIgt z+MxwI5q6yy#6t5JmV@IHvtCr@UhEcapBh1`AFD5BC_x3B!?J+G&&nt0Z44clW zT7J)&m2Yp!JhJ9PVW*p;(V=%^DD@bm>quXk?8SaZu|(LRlx+VTw1KWX2qrch|1r5V z>>az8?qe{_{Wiujr9H%q$Y|ZU&E+Zg8)dM&VbD)=^@O4E@kgN^O8nGcL)UrqRC7Ihmvu4&Y362;bb>m~e@-MUq zQ-Nimz(I0+v*xk*Q@AAlU*F)4mcR~*xyvmVr9Brou0D?=2c_QacV}9DY5_h$ARd|Q>25w;IKJ5<8 zn)HnVa0RUkZdcD>8$Sgpe#g16Y_0cy*!YSe27)^j|0{mOW(;QrX#02AnG*GSfZ-3@ zg8ErWA={*4DMVUsve%daDtZC z-ya{W92xYE_ZtEs6}RT2G%qe1K$?*T;Z(cm5m;4+U7KVN-+VV6wt8rz1U7k2z0vmq zglu|p**us=Dez;-fSH`XA6k0~dwui!^Vr4Li~~C}2UJ?^3fBSW=A#(eVyY@URQn|o z?3&nPYMi2wld44p$Tka&VL|v*OLzSL1m|gwOT26nY)q93{fp(uAp8O7wnOc1c5=Wk za{BE5fe@4fzK{xnC;99aly}F|*uAIIQ8e&3!}$UankXxeIV|mUEwan6R(^?-qoWwe zuUT(q`8f+4@O32KoHC#Ni-r9UFM&@$ z{Gb2pKmYXqf5JcCjeqz2|KFv5cW3_j{r@-b5>?Uhprqtq9UVG8VMUvq8RGj&@9zJON%l>X2 z#^272@&5H6{&j8p=by50KDmgyKUi-6?U4Rus5bdeT)X`EXQ_OD(klR$_J4X9*a41< z6^3AN`4?yY@2^U1`3n`Ws(xXF19|_w(F3c@I=}%mK1J=>|JzW%Jeg(x7f@{yUXT2< zE!f}wvk&6SGRep=Vilw}Y^4 zje7uB;<3LC5S{t^p9z$yQh%`ezku~opa9#5@9_bN02%CWnI0StvIM{kmQ(n@8I>SF z^#fBdqOuDA``PrT>3#Bc4*E|mfd4X5te9s3N9kF$B!BlteCH1YHlW;sT6ji(U9kUo zI{fQ%a%y0Dd=|Im{<}AV+y_k@xD|nobOSH{{*}oA^TY4Qs?&cB?_Va}e|t&z0~kJ< zt7^iBzkg-vzyv{}>^%6_LH_%B$np*tKH+IOlj7gMvd@6^U`T|{|NUuUAok=Q)(v&D z|NfQv0@gzk^Rnjc-!{>n@4)bbTvWXt&lq6`SCDnc<1Kf`7eC62l5yFe6~lcr#^7kS z8H|O$Z5=W=z;z4Of5&QydN?%}%cD${scedAeIW8J>@8 z)>Iwi^NF4=UdDWn7F&7`@Bik=e166OrtqENfCKu7qae{sWiNhd|3lXQ!{2mD-}&ED zJXNxzZoF(}gsk6vSBL};;6v{Ss;~o&sEO}B{~xC3e~iTD2e2%Vj3Hz{9c$kIZ*2F! z8~A^|Z~OE;QZbbbe(-RQ?p5iJBH@0oT5iy*k{HU$LDkOL;o7%w_m1-WY`>k+@FURP z<)W7J%m?cAsJY~Is|njHd6ZZJDfb!1Yrb}zN<3H^uz*GQf?K`6TOvaEDU740xed~# zBH>x!;Fhium)|gl{l$>K4?Wog{M{G4<|tyAyw*&)=1)2mCWwY|DZftsWi#wdn=+sT z!f@PdMdWp|p$zcX3fz>c%wpH~D~3WXs`q|3jWumO>y10ma9uqawcChQ7Meo1KW)bp zHRzuraXYg!loSU_>=75)b`F|(?!c8+PBe{)G_sRfd~BY$9J((=!!sPcHhxNIM)oPw7mC^ZtDKRs^T4*8pTUC&(pzOcN}Zqoh~@_0`lP)tict(DC8!^_PBC6iY?$^(%Hv2Xt2iZ8}fLdxUu zKI)zEkfejBM-u=$`}7DJHn`i>7zP)<(lYFr^Vjx7ZoHV{KJ7}rT?Z&eG1^}DN({q} ziFm&K@w9J^!H(I(sfqTl`b>v;{R}TE%rBavDteOG3K}0yzru#SAKvOt(rY=$LDQFP zStsH-?F^=`WQzM4ezFdO+AS+`4Nbw&TtBccABzIdJx~PKw@E+0zT7&zQ6Q+j=YG*e z^g=SaEnt1Ye)yz`t9PoRIV4Sp)_72rI!ZQ`q0^d%W;~UkV2&xrIH^LtPHock@jfkI zkRhhLCO(uFY6L}PY~=5YrzM$gKCudCZ$l#UD0D7Rsz|!q(|!!)2$3q|S-KeMa%P=4 z$lW_z@1>8|@Q9v&uMsB$e~l(hH3CBWi>o&f_|q?#0D;bk$ZdeK&|d zAl+w2RnyQPPqOElaSW*iT+I)!t(?n@XAEo;7m0>fUgQ#XX2!vf9R59o8vcm?VfY8=uN=ilUTHV3ckOX)>gIG0*c}yW3o|7(VxMV2Z&O$9_xnEvWvya1Bzb+3wJIxIgJmQz$*AXE4ve z*B#M}XgK`dg?<#<7SM!r{@c+j)829Sz7P50BcgXiGLvu|&~pySZ~Y)6_%cIJIIQQo zti6)pP$LQon39!gszv|uiL3s}{6z!ukmX9Fslsmc%$1qQ%1EOyx2x^EjHWiE{1)K} zU-jG&-7n2-(zLL~a<*PTyOaE$|u*4Adhi3S4(f(gD#_% zTGsFp1Nh&4L*JhkOk%TKE;Bst)XFeYG32BVII@d1>yc+eUVhk&sB+l`?J@1*8Yk`C z-(Eo!1^y+?Txsi#=Z~}8sGsOV`NAIy`QGax5mZ(>(1?9_fg_O<^CH-v zoIJw+W3a!2>#{ZVx+#Je-}3 zoKi!~*O97zbc>Ra*a*Q7SR}F7hOhS}^3WxHHnd6*Oe6Dh?daxYr2S)9D(O9t4SI{^ zro*(xV86MyuE(}U$STO}(n(O5!Q$9gJFfS2j)G0GTHv=o9DgiLTs%JYGu>864;axvg9fmJ|ru7pRSYnQqEZVaTNnP%?^aV_BA0dwQJlTWM2Cy zYRZstod;A}oCPm)68XLQ$WjnX$h0?~Dxb77CTVy#qid&LH`bMBA6}K>_y^aZ>8{U(=`nnl3=oX+K%YdFJm_hotwWL%UaT&1z3Bti{z`Q8F3beo>GTKR9ha%)nb98<|Hg^G0gcjUq={j?am0AI9v`-A-Yt|l z#3?NcZjN3hjo>i?Wws;q6QyDDV4N~vd*L+AuAdM-`rYHR|s9wEC z_7uP#e@E%rpB=F$z9l*zz~db3ftspLBN!Ml)-R#$yjWE( zj{c?WOfidvL&44WRO}tZID<#%C_0mBJ1diPJJ-lzx(Q-5v){r2U3j}TX$P^kSjt}1 zRgXG<|A#A&`g$kT4JfZDRhw(tIZ#t>7H81?z|q=%5U*LCXGF;D%56Yp=WX>H*!G$% zyycsLP$3PS!xHGitc~+biw|1$ z8CGjhFLlC19(}!zZM|>$v-3wdS%mXaMQ*0@LVRUhpT_--}_5K$xVPr7)|)4tzDRPK@uPjUuWMI6wwW9DAC z{4JJvtb%#|6aUDK!Q-I;iuwm2Vd-&o^r*tcbLqca*D6K8p>bH(kbmI6mBZ|FNz$;x zj`4%Ql%M)i+UjgM^YnrPD$x}l-_v!$7=UgA%vhcX=CrTk?v2LYs*UFei=<1a^87I)wpi*G>Aaso8j%7#C4Y7m&JgpoC#}x zoS;d-@@oUBhx*mf{zqkVzNJ5zGD8}gHCU9F1PDN%vafvRa9;CfM+A#ND1O%s;^gi}24ufy1|X8s4z z(EgN3fHkr#5~gSIQhSMmhyB*T3MVbn(b^%;(YAWc(TwKVbJTr#4EsqX>8+ECx}6fNU9#y%8~U1$^&W# zIS2}ikNJ}^a1<)~OYh8nid+!ad^Rk$_4l-N&QtAS!p~Ker<8oj_H{L0lo;tTRY!Fc z+~CxY9|Vc1n1H10F810>%CrJX+zIRJP)+N@ukNR$@kG`GZ)VGv3a$ONe3htlN)$X_tym)@061lH>AvLUWDac04nLePBhZv^&UG<&*9Y38G#HdaY@@ODmQFUm9t(2pRm4^W&5@8f7t&SUO6D=I+c)YT`ynyk}q6&_D9Usod1q< z_{X*R1I+#*{T95{Z*H#LaTfp-ZHsfLadl5cjVfXEa_RQgL4DhOvbe>b+QS^~gy>qw zbu!0=*978my7O5x@?}rFa%>K#qsI5_18yf%=YR>`DwWWAf`$~q+s=WE`|FmRUGNe; zp^`h&2Fs~)C89gdThbXBxC4TiyiJ%eZkK$O&?gmg%y#d?14VXN|!{H5k-dg~3a&yGe{bLRz!6+Wljr4vVy;yCMi*k0v)|zZB9d2DSKV>%3M_baIy(faUe#v^hDa`%)LSu8U1lJ>;^7p=^jM{;7 zO}d_6tRB-cwxwoe|M6hR5qyof4A?3NTQV`!4fkkMl#Fb|a6TemGoZYEg74CE|GW+o zVd?xyHq6r2nHZM*m-wj6E1v`Ir5FXuFiO~E??wDN5oP=HfFlF`vr~Z;b>PH_Aqfke zWJ?94iN2FA;$EU&t_EPL?H{49vEd?u^Yy!Be@vRQf4WUbvKd*ql$%Ea8MC`MV!ZbL z?9_!09|7l=iW6*E?NfWcpH+4mze`tJqQz=iwO*&Hds}z0!y&XuuTscuTSiA}WTiu> z>%c|${0T5Mu*Xn@ti1~x7sy&|9k@zOX;b2L0(LMn0+S$^bQZBt~X-TCVn{}{sppJ)nU6fKr zr+2B&3@UZ?P-UZBA;R_)oY><3gHN6{sq*v0_N4pww=78p0v}(4a%=E10pT31J&~wn zYAm${L;-2&(qj9iVii(#?cy4T(U3kOkJQHvaXQzL7%D7OxhXVKE(tt(Ez(Og@^97d zxv>r=(_B(gl97U zk9SQKS3IA)dC_*wvPOp+t8Pi^AYS&C2W$mc8|y7pK+%VFsRe7Fd9aICTYc{umTnwF z5~U^C;X5J@OQ(!7ufel&tNQ}rtnB@!M&A@n{^)$L@LBtWM*YXBmU^pO>U*VY5prri z3cS`@Bv;o?W$wExOs)%jdRMKD#~ok8GRx=K&-4xz1#czss*Eese2!-_r=Rfzk2pO) zA@<8CZoZMIyN`6QoV?>4mPj0hX2Y54lw~88tHcH!Q6S(bx43L#hSpUO4Z4 zR+^zD*l1=x1d4u}6%jEBf7h%rS=n^h69BaCBdL}3>Y@XW(CXLKP5ifm3_mGMhQVy& zs0%K#q$Z16=(#TAk*P~YCS4OKsbu3$uE%ZVt~X7DymlTcM6V5B%=P#oU0xcPwY8AX zPIDB$XsChtT;+D-KHs?DsD6Lb+>`h=*$RKgG3F#lUn2eti|B-b^x-rZFQ>J2INeh| z!-0~16Gk5qIq7|KPL_{oLur9-(AMTeIhZsNufk4LvJuh7IfHl( zdX`J)k#p{l~3#Gb8M0@3sq049QRx z(akBmtFI-pWV#O68&2>CO=NvwJ*PV_%D2BZOPlpc7>rB|V+zBQig%tUoTi|`Yz4qF zD!W8rudZ*LMwmXa%virl;Y;H~kWfL|4opRWDj4(+7zu2xBLvs(xI&Pok8jlX-6d3x*>c?8_U$*L2J34(rsZ-!9fGkbD}l!)|+BS zyZfbxdlI;faMI7N!Z6WTP>V5I#qZOo<^a2x)(3iLwv4~mu~O~D;@PYSvT^s&VLz57 znfh)$FFW1!HdcyPB#EJR)_D_4WrHh8`jO;1;Wzw=`{q8Zo~}3gZNc4f6AkL*E>5-- z%3Kif$uUa;Od-p~)iU7J@uV$LE#}yc$bmliIbzCFj3EiTRuPq7@qJ`8H;q5$>2d$t z%~Y%?@;8Kjt#1e!n}m43Uijy|C7&1sv#Cx~b&*7#^g*cKI?he; z+%F)fomrXEdz3NvWM9>dp>tD+ykky5s!Bm#4CVHZn-r2-E@$)`>!F2LJATv(0rlr7fK} z9yBAoVg(Mpw62ueGDuG8rZ1Pme0|ulJop@mck~DgUbFI6 z0jNq|PRtwM+Rkk(@V9n1#P@KX$eSPNg}9V3Utjl(mLL=;x9IRwePuo&vcZ;o+H(%$ zE}o~`H6qn+ZR9sI6wS#xfnQ33Nr5e=6CdJu0-Tl!(o1>;F>mt(;I&JUYSlZ7Nn~zM zEhz5tUa{!UQy!DP-!W!kZ@oVLB~N6Nk76(ggI;mM@0)Zyfq`msgip9B{jFhZ!+e4{ zyN|}Ak^aYgRY#AV%6x(Qj}qU-*z}?JTX|roTQB=9#utXooM{h_U2zAgww$nTvR|u) zG^QN+&uIn*OS&Ti4s+EM=NDr##K)Q5B$BC z$NeHn7IHTTNE){A7_X{mzwBR*Xin}0F5T*+t!>6@M+MYh?H^q$gP%o$>%2h(38ias zAML<(OS2ne2ia}uJ@~U|)+-p6$58T2cdEK1Zt0c*;q?(9g^rQC`Fo(QC-E0!`rsIq}nwe zkr`PGk_}U`FH_{+WMqp6kP08c`; zH>!Lq$|VVP%JIVAcZY|uIe8Jc)X7&=b9uyU6nv!`B13LxjB2rJFDMB(wj%dBAUhyc zn3w&uBdR&N7UIu>f0;i%TgGY|ru@*v?CvACB=B3%MOliM<%&u4Cq@heA#4fH4uY9W z)Cy-y@7%pev2MS{k)%J}eQN@NS<`3sx0u`!aJ*++^^_+WAM&JHD1qo?T0cntu29X94 zBqXIfM7pG72q{5nkWK+fm1d}62m$Ht7`lfcXK4Nx_l}?U{_ST!_kQy~o)^zCFBpac zxUOrhbFFio@%g?ifvHcHd^Ksiptp)SV}D-HTo7ytAb zKlsUcHFj)P^TDJeRLCzDY&$=p6RQ8k8E-&v3RONZEf4_RN+1y^X|*$p-B;i}GmiIA=zs=2HFIf=>F(=kRs(QE$8$Z}MM zlUa9=U0t}9$1m;BXiB$P)&Q|1X;=&XOg@3=&SYI1i{=j-dXZ82leOoRPSGhY4}+s9 zFJCoCU8u?~>b5mKk9DT2;#aI)vpO(+m2g-a{V01O&*|qz0ZDty&%18)OP@SC8&x`# z@ghbBZXpZT;WY)*ZTc)ahzj7*a#oXN-+mo^w& zd(pTLD~-tD0W~e>A4_GtXn@Uc|6DFG_QQ7DiQ9vicPHqf1l)GuwdfgXyfdYdR_FZY zwDY&mh3D}O`@9tb;=YX83Ir!GT6Lx%OnqR<9^Lp1<5klL&Mlmk{fYU>M+Y|B?Ro`i z&QFBQi@mOg=9|G-O~<&H)msrG_^(Sg1t&}$3kELjaX}+jJg2Nhups3|wK5Ke;jZN@ z(}6_MUqN2)e|ym$A2jd1RuO^eSczb6VtI0w3Bc?HTmAKXj99;rsGn#px!9yuJ@%1e zmkA~2^-bFK73G7ZUamJ%9%ku3TEn<1=u0m!rgi<`4S7m9+uKvUU)^Ww&?_sjb15tY z+35}C*l83`CnBiUE^#rgl@i`LnKEsxb#_apn*B0@7l?I{GlIEUcDYQMp;^}HABL#$ z9qfeIV+K$!PWn0hxcibZjKR*esum7EVNMDVQA9cFpZKjCe_8DZ?M^s7aXB|MrK0NK z5=PKn(9ILZIZ@HqEv`Il8gg>nnSBw7O_C8zn6-dUL_?wfU~yC}};vNcHsKp+ERp`F84w>BZ1yFsi7pEyD+9zQXFt z-6kQNG`pClyp)FpLyGJmEn8(R*@QFkjH}bVCLwfpa)(3SUm@pFtviZjI1f2SuJ$>- zKc>W@^NM8dRxr(by&H^)&9r_$X+l=pqZZ$v&2W(Ab;U&KDx5Roxp~MH;+A+GvfO?< z1Jb8oeVybOFc{8egp81AJ?dd7#Bp9%A&Uq`ouEkOU-0Y@dbqxuK{$yXa?R6t7%s1w z4hPhlWQWX9Lx(;(3eM0sCF!I1!$kM=<*|Dm8swmKd&%e1tGBwz7&@d}k@qmEjyt&e z?FHPn7;$G=)vq{ZpI5-0#LcJKJ0-gPXXZIALnFyZEi4-Vij3}wUu~ONDjGU}JT+xd z_o=?@EjIvnFZ_%hnfExBQXsr6#BLJ(5N& zo>L8*x0Xj+<$k%6N*jHMZ!ci9E|>eUbfQ_9OCt8i)MnV4P!(d5gI%fo=1{EnRgX$# z$nIeDdfu(HB|**%NY*~3zXD^ao;40*}8)_L*>+KXsQKj2!Vp!_oFg_bHEfK5?X2HKM%zf={GH7CR z5603NL5BBbqSQd@ihsbSuU@+~6AdFV`L+ZZ3w+E^>JeR78aJ&XMa>X(?hUg;U>9LD z;uVa*MG547#FPeMwTH1`JV@Ra>|pqg1oX5zx6cP>`NoCC>dc?a+wutlls+bL;L6lh ze=N*Q{s_H^=Oqj)=+7!M52%;ABz@~j{mqEqd(ny#BqK-iW%A-M|y=4{oEROe}^V;9S?HWhj9aLyl>O zo2p;gOL`CH51Um|p}5ijw50dfwMg;owl#qS_(~)QpU)1?9VotQFS1vCIkhgsEyEms zX}8qg+>|s?m$YXe&Py0Rca`6lTyjVvydJ6Zb~vDQ5;%mix4VT|IEaYYh26{NmZgj^ zY}84f$~Owa?S;I1y4l88F0!{{xwYF{eu&rrk-tjr##Rim;fL*Q6as;(FKl!JX>OcQTpo z1Pe}~%Sp03Q$N-q9_Q_jSPz)`lj(L>nKtyEo=zy-atgH`4w%w#5ILkc?fuCgR^;SJ zLxg$q9afv*cv1s8_h~P#?)NRX`DAa2fp21t-oOVaJkI>6M235JWzu~(8_Q#~8oYav zd^QI04(wWHnYlzKBq5+2soaZ1ku=6=fe2+)b*&2K5R_~mX$`)~n1zPBgi)5+4llKo zS?=QZtlX}}BW^PG?DD@AuKOK>uLizH;D@Y>SGhpPTb>UNAa8d!s=}nI*G6&sT}JI5 zbh%q<vEBPNff{tG?8DX1c$?;#Q7GzkUt; z-7s8WvJ0!iBuFySoh)DU{aj;+LhFv52R1x0IN|x(5rHFf8h0ZAPqhgwi}Hzh_qB%)RzpC%`e!yq>*K3uoI>jTy!|B7BS=D zovjIXjly^BReauAfDPlopUDUsx(Cw{9||Uwzfs_r1<)aqFA6sAzvJ3>D4?obC-hTG zbg7iM2ud6`KwjKq{*H_|^yDcUS#$#^>)YhC*PIHGt0Hq(+znWwJN+@Vi!k{&kOn7_urT<1&}K<*_AR_%%)se4k*`T6*cg6vQa{9bsjxj1ohpo78#4< z-NxMECb^49+hplf>)aJg{$qc#7!SQ{oLgK1b9l;lJKA3-Uk3N9@$ucYYqVHxr}s&- z%$gMqJ4wa&`Pmi)J)dPakC@VTXvn+h;CpGwiP7)*ae?sbAUPa=gwusiH7!Mt@MN|J zUVpg+Hm2s$m!kDGy6WZsIU~x}EBJIGC`!H$s-yOdMRYR7waR8@Xo3Ts;_!!Adw`tj zG}2;fb%lNE)71G|lg)L1+4(eI{alYhuH7>Q#lZwOT1Ii|fC4p~LwQGkxIC({mYAst@No(Oc#uxU*!nrM=` z^x?O0_~bh9qI_*uPPip%0AD$1iJPUjMrG4^ezwPG6RW2oZF?=u8)LKOlcbTCd*_d9 z3|&c|_V;Mt_TF4{I!C-bd3G#}F=jgMr7GH+{nW10q=7G}RlBz({Q zKxvR@H4em`ma)DEH(y$D^Kn13yA&;_thc9$2As$rkjwYJH0d+uZqD0BI^7zFzPSA- z7VjLdSE5Q!?E$y(`)BHzB_>1BtBF$Shi_X%n}2)7)O$6nTqrx{2plNA!h%`SnN*g0 z+0dxJ9XVhj-akflwi=k%F{w`?K3_PUY-hCwLf1ZpxA_$yqoLPV`WglCKBzgA8oa-o z*Qr&G`5sUZ!qi9YsdJsl;}$sk8W75~%(?&~V>!X*Uhrd%3diftgRX+Ct_fp*hFXU- z%v6EICP;%&k`|;}{L z5`JD|W21_9`S;nhCY^>0V1y}Plk(XZ#>&Pfrv#TdaqFRCh_d>oTDy76Y*9^u%MT0O z?k`fta*%V|E+BUxG67sUCacyHI)}fdfNZ7B*9Fy>U#-NrQ&Ep35z792z+{B$YI58E z8Kg!Rh`~!XczU(J0aD1EzE_?DS2-4*uRh?Czr9JpZhj>{K1NAQG6T7@!F82r-&e&j zmjdxDL9hAh+Y zs?}z*>x;^Ygk$1}P?Pf+(E~i`FUI$!HPXi-jO-;GdR6Pd-u%AQM~#N3%Se1d-Ms zq?}qtqwLSUPFfO&FUzA$-A@*^$0rBG>%0g7IfKmRjOi!O#m_e01CAW$&z=RL8zSY*PFr4p`>d2I!J=QY zJBZ>%fGuxr(QiF!JjA1zWtPG?mS16?KNT3l8@&s4opkWO*tBF-Yiw)^M%_ldtjeDQ z5Prw0$8r4IN(Wo@q`AlWbA3Se3GGO>o#>%R zGkTW9sOGb{M{#Ti7={**>kjanAn^6mQN60`6LEyl6*`*KAUxF{J+uBkP@qK4obZHc zMZ~w|7!yH0FOck2Ss*vXi@n+A2rQA{A`?g4EwWL5E0>lG9ug*H&C?a~%fs(f6srIn zo=@vzlocX?q&Gb!ex=6wE0w^nb1ER3ixA?>bm^aRR@gOAYt$hsQf>P7nN9zdp-88b zqjKQV^?H{7Rt+o@jC!ZxgJVu~nb zCE%>zp!Y(+IUGt&p+iwGLC9&_(pVt1<^~+ZcHITH7^|1!+)=$JM9v=7Poij0T?lAa zj8Dj`l6+k`pk@D6y_@e(r?}zOMi&ORH^ZJl>G>~Y-?m$3=w-PX^doVj8^t&K?X}@U zAnh!$e$hq8V1Cx?)ei*n2WUeaISOGZPZHSVu6(>G8}ia|Da5t%6*Im<8XGLb4Se=r zz6J>#d-qdLy`9@y4Y*Pr&+QGJy1r6Agf23B2mSLXdT+@Yr^%QrXWe{-w%Ov_iE3a;9ACTv?m`aNgBAJH{Z1fuN~JIG zYfRg)=(3&R`%q&;t*+=6u)>cHXq+41-J5XBQEohQS;o`N^W1M$t8tYtQai;TXk1Hb z8u`lMptC3Ur@n+Dd6S)!E?%H8tqJUN6dRfclwA72M?SrLwej*u#=P*<%Q zAdTWD-HD1Akw1wdMQiIqY%u%GQ01eY9e=h^+S;Zat?=Om^}m?=60B@6vcC%a@Xf~Z z9CbqIG?T0ASv%(G&g$59bvBwQCKEOLj_1kT%(gezSiqEw+uPO?w>8gGoca&_RmA2` zY+G(EaZA5RW{_=)bAC{~**|yXOP#dV_lW5pM&ObZydiA5@@cGLz5*9e$&AC&vIBZu zS$a(SK_|p9P5^C z37ABzP<$T|4nY;x#hF>mKy&ErwyIlEwn3Q3i9q#%z_@8E8E1|QDg9I2gZl?sG5aLM zHlsF$hng|QU_(Iuyz8h48|{rrlG^Enuki&^Ar$@$aP6A27kO>_?}!napnfPhFWx1c zl*G(LZSl!H6bCEziy%ZfXrF(pi6r8sHEFE7)U-5_TDGu1zV#Gj60zP9FpeJm5m1X3i^(+K3hrXI!r28mhBk%5XmD$a^u9;c z)L_<=qY5oMBGMk0Ld(Zk%_6AKkHu{WUwltRB`chf$#cu-@}ExWR#@#j3tgKdgw8UbpTAQB(`ffkzeay->8g&<` z#BF4nIQi&~TK%dY9eup5TT_0R?khHYs?b9!IPjR-I&y99WxAp+6{}wjq(M2_qxoEKuFo&$-liQw_u6q%b|VuFbNsWb7p4bdJ6bR>KOPx*e+@@IX) zWf961Oq$&!(5@08S6>9F>(WCu-PZ#-t~lczC_ny&2IV=6z?4+N!mZI_I`daas(c{l z?nnU(C-fsV^OCS>sIFd(6}xZdhDBrONABGzc|WR9!sFf2EZS#LjRSLyX!ju=zYeJEO6gaKesqxo3HEuZ!y^({VsD1HkY#5uHvf>G8W9d}8CAfL;i@(1$FV!;p*0 z-eqB8`uB4l2-nkUyh((Bn5ZpWr8Gix(1LW|duV-seD`2|sFFM4T-gwiU}4j;CBoqI zv0t8$r~Le^?YrOHw9P{$36a_|7uQ$0E{{4me9Vq-;Z_8+Y~9AY?~kQS*jV|t&d4K` z&X6$IBs1fwRv%MQ_CQ~j9G}!>H_1sfKLI^1V9S$JOkYDBXK}o4mndDFZSz3rZ+E+z z+2JSq3V=GKH%l%29cG#oL_TWdpRLCJz3f`yOWxZ8Yi+UH)Vk<;Ga z%$XtxvPa#f=;Ix%(a_`Hk>jh~s<-c)c9T7+Y*mUkoj&D!i;BNVhUUo=R_}6ycuFNUq2gg>KN6RNfJ>71^??=E19 z+b3?~X(kGKJd5Vq>-+9VU$UiCF<>M5frCj$6!jB+TaA%*BY(S=UTStM@J;pJe!%-M z$#Akpt5SvqD@9!06=imXJLMbSp4e}6Um2`jv|e-rH|lWUsX@yQ zCi(wg4%C8`AY-A}eB9^>k`iPkBX9o*3}DVsUr zWNwn^i}*)Lc+#)AW@g!6y;FzBZlw=GU<<}tp?)sPbe?Pj$6$e&azXx z)kjqwGXrtHri?N*l5Wp$S0Bi)Iu_>IMeGz+&y>M6weoz?e@d9f=%LRpd3X~;T3CQA zX^Yx=!IuB4Z*$a2dbOP%OO{6a{d(R5ykqjYhd~g;;w~8Fy%zi0EXxVGI8(E%q$1!+ z?S+#tz6%uX5U+L2V-7*Gac@D>-PDS6DqiWW4O)K zR1C=YmBjUjqUld`JJioGE17eNm82%k9u5#8$aKy_wv%LIEEi}%P3Nb0^_g*}6WdLG zn`Lu;mv9eZCAlP+cpVViy8tptwjO6@IE3Cb($r4RVQGW<`>IWNbb(e zVx${CRu{JZ$1YR%8j^FY~QTFmyS3Mp@!mj+hJ5mT+x(xDVJ_7 z79BgMri=A+4gNC#WBIVA@FE8u0EvcV>Q-%X4}6jPc+V6SQEd-TQh~d6x$y`Fmbn^3 z(cwpZ4dI}UTxIa`u6XoU#g`Nn!-myVnKV1+rPO$@jJ>VG8-(^fj$k}*UuyND{r)P7 z$L!673FZr|{gLsy+LEoV*4QEkZ)9rmF7m|}Y)6j*f+wMa){esOW?Mf}8N2S~#5XHQ z^X($_0yPnS8m_1<4mJp=6 zBy1UDf0jYoh&~8J{8sp|9BOJ2Q6s6xrNS938|h&9ax9Hv=mTN+;BCuL$E;iil54oF zAi-9zQ$yqtMb1Ye=gx?%`)VbO!Grr@KDhfMJy0+I#r{$kq9Duzn4U~%kVhv~l$UWF z5a5e?EhC3WT}s~o53UR4_AOxgdTWe6AI+R38im))FK_RvZkDwafyp23eA)CCGZI)9s0dW znnI_grUeL2SBX{>M5QhS0_&-mU80=HsfyRYLium6_)%riH`WCBOAM5w9v`{@Wnbd2 zQ%$7(`3#=WJR+PM@9{8F4rGwO*187x21fSO)ns|OHdzNx=sTDS?Ebq z^*YtMxDlF5#C4)QjnnKGeEtciV3x4jH_gGY%Ini4V=2Nh#ndH0ZP1s+jL;9r9kni5 zb{?Zhn59socG+{1-^$?c@OOa%&g6}HC(CJ^T7r2#m$_0OUZfSn`+!q>npj7rSvr;( z_t^f|`HzOa+lAN!*xw>a^rCsE;xR|cO2veGqf=8;Z7j}@YUjyp1ve;z5kY6ofEa8j z*mQ&6q!+iNeW)}e`uxb>xfUOW+ggRlc(mH+E73sNY+}L}M&z|pCJV7VZwD=CBU>}{ z_GJo4?6GEy^A2?Y$s&JIsQEzY9t1&ZOzni>{YA;D6FcbmA`=RCXw9>ZYWPgpp%QZNDmiGaSq$b<`u%*s0 z=?x*coZt@|N^{*y7S5;>2^q6udfLgP zEa<7|u75o?(4k~D zX5iD^yOO`ea(SrPNB#DcMQ2Y(e`222Nyd+VNprq`&-B49{ab)3@m$3_KYbzkGbALO z*uwS&VZg7|2T_;bW;6CW#4K7;Jb(&Tu-FLHea^x=F}^NKX!m@>%rXXj7qR>Jrr1K* zDB68vILZWCmpIS%s0_1p6)5cf%7yLTq7eh}mzc zx2`=jOM#nxWAX^YI>j!6InhjTprzW7_VbW*qDAvUd){B38eC5l%n-rDEh%d1<;pd}2~l zk@PFty_|eus$5}ME1bs*d;htRIZR+!D3b;wsFe$Y&P&Y9=X}g@_Ekd*ZNc{$kB~Vu zC)u=HrY7(6wC8HvFFRCsFR)9VHyggfhOtPMccbqHAgjbSFxUr;0;#S*)qIN>CB#Mc zpelddg>rHriwneD)-uaSH_Ja4l}tFi!w{a#srg76g%qe2D->e!-TcDgP!PMmzZuUU z>-?75!0CLxQ_(IY?)b~*uzjZbWb+rzJv}zrpimAL(TBTFjsH--=lwC+`5K!ns!oCTok0QXQq`Oo z@O?bXVg}=zj@7KvN22tkhOV(Yv-R^+Gjr(X2#R{pI#bL6fB0^j2`@9T%2r-8Ar;t*kqv(8@T^csXSsiA63|*tqE@(J z@15nm6Fph24uV_@BZD=1V!9PYT5XBM~$q zu-UP_0NSiKK#D1XGBGN4X5E}SMB;wE+>+$(Km-nADL8vKfw>GgKc2xgKb6_X108gl zmbIhbVJ{F;;d}R12H2=Xs>VRv#s$c&4!(IX!tC{%06*;ojPtQK5|)ig9nwpK)gT(( ztBZT^S%~oy&a+jb$s+U9uae8-fnwURP>iGPFEGuO78L?NCJG}jrOZb>%7~1J%XVXs zd5DR>PBL)XJj3`QA-Tz|4ERbj!2!Nf`c*asJ+=mIF1A1=d{ZOny#tJvQ%gj;SM9jsZF5?1bWxP;2r4x4J2f9(|LCChRn03y zqO3p=>4lKkU_D(7rEw!G(PmGat#c9fh<+ALL9jo$EHVtT-(B^P_KLp=SQ`R8Jl5IW zs%Czg@Gzhf_4PF%tpP@DA<=hKKbRnnD$I#5$WekA2_c0zubGY9#u9kT{8d=hi-X_x zdVX9hLKpVR8Hle}=3AWv0@+{S@I8P*L5qfgH<46^aIw~mdIQ}X7(0i!ohzRn^yCib zy%xJk@3=3|>x&GZ+wYnq6WtN}c<07`r zNon>;Nq|$mom9pobYQt2|G7}$#iyD=J^mME0?)Q$4QB+p%~Lq}TV&j%eW-T_H46x+ zIUDz06oLGe>VOk}-0Io3G`3NFggW1urtWzxKA&Zj9ncytc)2?zfc+{I-Y<%YqJ{yL zEkOrGbx>=w1ys)65w{HMo!-~qSLiYtki!ySAojhZ-~1VK!4+ulD|TLa0ci4MJ9sp= zZ9CeHoE`y99X8iStEt$_K-=o-ab!uA$RXx{R3G)YIYZOS813Eq%8L6<)$e`w+mrri zw((sPy_~n`^NwfNe=72NhPS%gi^P&rsV)?yg{L}v>hkx(WP3}~b*ZdJ44%v;%|$Oq zKGa7&Kds*Go8$FgBuyDGQjM;<-HI2WI=om7jW zk@$gvIE+^X6bn4rSda86wg~L$*~)m|0-9>nm0BJXTB6bndbzsLB6X$*B<8hhvg~*=d2osS3CRm^f#HUKoow5;njvHYLX<-C%U#8Xo`AjZ=mv{@!*~$1+G(D zrdJ`C31B-#O`z>(A6n6F_j z7331`UJ4f|>1@2C5_fKVc>f(7h=X`Yb6ju&al_p~gBqWgIsGh?=WfKZuV#R>A+wFO z7Kbn`C7J7oEK#o);bVjX>wTH*R3gWO)dM-iO(Jrk?%zo$i=nbnK8$^z{Q)k&Nf{8f z@hNG4fJLQ(@ACpjwhXm(5@Iz0G{P~%-(l`k^!}scBO+h4MJ$fZrw#F%u`&CQtkDil ze50&dDdb>P;}rE7$kO=d2<$L8J)er?Jq#j1+3p_#v$%C$2rBw!H!|wo=d?l!6Iyq3 zja6Rhw`SaXcM&Qrfv3HUzKTT`6z=i<{?e~7)7zy^WCRHbc?8kvJz`p1d=&9&BrQQt zEjs(bqtAn`XX{~XrQA(`W_Xc$X8<}FYl3lF3$~RDz=Sm%RZq;NKF0+SJ!^OJf>|7 zFZ)0^Q?JQK&a-7rH9J3-=9bnTY##-nIC3%WFgkcF>bm(InYJ|e0s-vlQ=su}OH!u8 zF%LCU%NJOZw1+3aV!JhJpWkDEXv)vvzHy>HYnaK|Hg8y7{EEu@Io zpK?};MZR*{U;My)bWgHA%B=whSmnk=F~NIOwll&kS*hopfQnH=mG%JV@KvmXw<-y` z6xJeH?_1o1?ua54=Hh7j1lEQ3XXuDTo?RcT*UnQ*Q_es_D}rP_Id!$Mj&zkDrjbcJCk0ijImOQ`fGm8;~o-ZK+q^iBV zR>vmLPtEsR$sH^Mptf@p((_a2nJYb@@NYi*LdM`<>Iuc;#MO`h&UhnT)C(Oi0#M$A zH;p_j4Lqv%+&dQ19K+7$C4~^r%JcYBJtmrr+1d;o@AF`PTy2Ol30D6U%oH7EC` zl*Y@G$ZSgDujxJARM@t6ihoG`C&0*i^d$v=k(ci2+yNGlb0sF_G8{_Dc;+*COYD4n z#;sM;^?nWmDPIit;W(Kq(sg@Ua^Ozrds0V0I`$@|FBU{d#VB(k8LHSt{L(yqt5MwV z!Xtni&qWx}Le>Vz;S#^38GHu*$apI!pI!hRg=&Q+SJIL)3_fint*PQJ4)mr3FAj65 zs6W4Y(8GLhriqa2AploVb4n>V{6-}%ypi^II{(uWvDAJ_k#AQ0JUEXkznE6MZ891N z3p^ouJGdym*R^;MId+OJk!*hE36?eA17ff-ztWh&la<_r`NqsnPp#YDlT`cXu`r(V=BY;M5)iAnuT%{qQNn4ULiRtk-b&mA~!-tSHyga9a4Yy2YFD`aif>ZHu)zw zwGDiVjsg$U6-!E|&5j#u^F1< z`NRn=9E)ZdUY)ylOC%Z_Eqrxk#V-)dTs4?!T_vCKu?D$sh*yR;8-5hd9w z5Uhh0HDLs**5hC*&72E&=J_T41V;j>k`OJn`1NM3+&Z4N5oqz=lr(i@GwtZjJ|+Rq z&_Rr4_iklAXzfZuJQPEN?0vs`jNA37n%F`!|MOm?pb72puh%4_pU^3eZ!sAJNzUwE z3G?TL>uI7_E<PN%$_nVDmm{D=tOG`6U49QX}K7D@2A%|6c&oH2cRv{5W_H6HtxzX28;2tv@TEia( zLcJVqQALLRX=!GR>k8@5G9P>gi@f9uaEho*Bjl%D<%{ zSfmJ8I(<{a_tShPyWcC@i@`b<^P9uUlu6;XqNyVW%NNTj}Rm0yESkptZhS z$Ytg9Vc_H{eh?$v(L#3zXTAy!8FqY@g>=ZZnWW^_Jq>!TZ~aENFp>&LcoD za=XdVdD}IZ1n8DRo{iYVJhW(?!3cU*_NKe}3PSw7k;wk+#+y*)8u(XpmIBP#NAO+< zLNK5zP*RpTv9evwosj>FUYXK4Z$Hfm;@$PB=WfPD zNM-qRh7!;W_0TeKnQG~&DT?aQO60atSK9LQpz)Coni7;h79(>N_s>o&4Avpw!F%)4 zhO8IsVkLgysA7#UmYaf&6@wnl;)_pdVuy4iUkRfHgIj)mTd2uL(RXluZBT!qYlFvi zlq|JtXbLfur*aMFIFt7KR3(gE*RhG!u50mgt)*J)&jlQI`SU|AC|05wRNh{U$X;Aj z^HhL!Q{jhxBg4W$+N0zU=ew0!bW3?PEhZ1xW24ty>$VDIbZrNkl;9`Q^~T>2Zh3pl z)&w;mn6Gh?VO;`CC}Z2Q`A*7J@`I04v8>k3mxEshUcw2!^=PmMQ9cEYMdAF^|K|Fh zJcgz7HM5D(vJPIbexUT3M%ZWj?yor zZF`^3m14Zl-W^ChUJFyO#hh!aX~R7j&=A)Eq1*Dn*U}smtE%eY;*d-!?n{F)f?qC4 z#`)8Zbrm4|M}nH0Z8n5qG)o`Pa0cjD`38fQ=E~%4r<#_PAfN7H9H-j^AvY{-s(JIs z(MEZV{UKM>@-}IsY*ehV<2pB08X8q}dwuRZlRR?{(H6HdNl8%EN5c<2I@6&4GB@I6 zfw^%|%Xsi|QG|%t=y#s1;7fGwU-v-r=ikHLTlHmDsv_E>4*SxyZ1=I(pFWOBMF+Gu zMkCM=34T~6TD)4f4oD(-nsco_-t%S)I^UiJ^nf2F`w_2y+D!SZr9FK5hurJ$er}u+ z@UJU|LO=j&zdz_tJRN1c01w#tb0ElEs{Kxf+oRY--Vl5$`5ib@^R_3O{8%2XR{6#5fXDW=)_h7X%gU7~*yrC!!`Z!MLLQ#nz>`*+r^+%~Y^}E<)_OeBZ@cMQbKtg#fl%mL_=jn zi)e4C$*L;;sGshb09Y&j2Cnss6zF)|nkZqYb6JZC=WJ572S5Rd>|T=NKm0Yhw?$-; z-zv7gzVt51JxZMBWBtAfpW<%%Q#SAU9czFIh7c6XMfeMI`WHdYm8*qN>6NRW6pJcR zyG6=u2$kQeaTb(JEpbk-jI4op+=8z#4T@Rd>6) z62Lz7e9(S&H-B^2M)U2-;Fa^xOVYdb3O$LIYI-9FE+(_Tk0|hdnC6Co+18m>;L&c|#EZI`id^hF!g`47SiK zSe29S?zXR<=;l#In9$+Ft_~%~UkTa(AJm5o~>{`Y$8v{=N+V?*aP#lNb!F^kH(C zHSJ}@oxeU?g>Q^CL*lti)?b}_{jR{I7&xMAKKiRS^!)$3|D-_UL>wrbwiNcy|LYGx z?#_g z6Vf*)^Qyd!|3aQX30NmHm}qbq{`wL|$N|{!P+ZP`fz^L{<2I%Gut)qCi+&Zb>d7j6 z+WqS*#qbUA-uj@NTkU_AwcoAg-@PBY*(W$y>;F26UM~J-1@G!?mHX?%`u`8i|E+=P z+%LcIJ5Uk2I?pl!^Ve?Q*5%Pk; zI2y%{a6Z)wWpVMdzqsegh2Cu5n_qmdNA}fuM`oyedoP;kM?!XcE|h-ry8qqO{O2St zxG^{IxLFx@TV-@cV02U_rFw6X1E74 z6+8+0Yky|p>CN_`Oppv5fS&V`{Z&^?jvlypdGxjQ^j0uUN|l?~{d--ymKg~(^a(#F z)A6n4-o$NQJY|fTKXmqvzP3*Np3F`x2S(S)&zq6$47=HOo~JT3pD`*)WIlaRr=Fs% zyb2i^l`?tjgUua%lJn*6KbGY8;Q;xVG-l;WG(H4D<9{Jvvn!3+W0cGu zTIXz@bm#O?20qcVxX2a#D~ig|y2NJeNlyr=!GhpSy?ue`d?oZI83+{4=FI|?q)~=d z%n}!Sxd3@zSELl^G%;{$Umfy}oSG8=5(ot1v${9ob@EAQl17mM-5Pg1$#Ztxrn57O zsS5K@mzj%GhuE8hB{Wy4#YNQn$hOlua?7Lp8o=_ed)Q?DY^7HdRml`(Wz+sX(*Nl} zgQi10PyIzbkZDhlZHfgB5ehc*Ah2*I)uvVyWIw!|jvoP$?H6M9hAPvJcb(y-tpIV` zF@#=u4p{ives>R7W9_S}t8<#JBmwZOvjQjc#i-y%S+BT9`w>X$UykQUe|AI@DqsiZ z{~S0+Ju;+T*ET-@(uT))OohG2$8U8{R*8DBRUX`4aar-;3PK$(g>xm^{-}H9y!MOf zrt2$-&q$#SeMyrkv-@$5eXuZ(Uc870epeQkN>%?RXB7I*0lK*ikf>N96Aq#e$Rm30 z=W`kP{a`;pWo2)X%gtA-(8P~bg9T~;>Kh9PXrIQw#=Ur2p*Uc#b!Z>=u*o*CBKR^= zKi}19EUM|*FR#?B^&@>s z^VBtV2HU@YKH$h8@hd&k{67zA3&uB&?dJ#mZx<;7alIor_qTpLrhm?izx4tZ=eR!u)X$658)P1H zP5T{G@~svTEguO@20A@m8<4pEy!4f=3;0UE@y^F|s?neVk_>(41nZU+Bn&B zq8>8`aGt;zGKL`TCD{JrvLS3nv*~DKfX^_!aun^3mk!832!Z)a&_$!RxI9ZuIBOSt zy1rOk8+yR`4}$c%mUE-=H*PDHHJq+dn0No=@vcGhqae{(dn6z!3l$|^Sy!y86j>d_ z{y4?lVbTT78Z)8o)#1qpqV?vx-{ehqBdDSm(di-Ei$CH~Y{O!J{WEd7@9&!f-T3$8 z_Tv=jF}Ad8WR8+w;M0rc-mizVVvlpG0N&d*R)3x-u~@=No#@p$A~5Bv&T%^#?ri;H z2fOnHc=;eC_0%4>`?G_gNOzFEUa~?z&-k-2l0j>~xXhVvI-Bg@eSl#nIj0bS+=KaF zUxH`#8)AL-7X5DeP00l;tW@LT-;x$xh$g^k8k)&{(^2JWa3qYFo+Joo18 zd7~_f_RnLe9PiCuB(nxzR3vSJ`E1)4&18ps2qG6v?jq@0Kr2ui)0-ww9N9XQ2Fn@I zPs?uIx)Ix0y}%`}Y9`^~Ccp#`>C`QZr^FVbG$P0}1E5$%9!bGPP|jT@a<_~UvA^8; z^|I~zYB}e1-yMQamLm@dx&Tz%F|XP0gnYa(@_I)HfBxdH3pDUqbtAnxJ?>Wj^>I9> zl$SJ*Z8{jZ3i9?0YjwOJ70(C1{nY7+|Ea?1R_Y3BEM+&mifghS={U|i%!RjRiUkA_6Cs;ei){)CiCjGD)q+|lwt)~WVJE)$v2<9Q_@iLt)se|hNwsk zwxHe&A)zsg?YxO~qP_-d+v%J-S<&vFI?LqY{tQ0rf6FV@=>!Th`sI4b6g+;z`ffyS zRsP=WK^4}4VmAZpX*0JD4&19s-Bi`WL>{2vI(pQsw>%TFLZB!F6`Rg?LGXTCZcD(3 zsrJYZn8m=lBi$ZKs#@hh$xF$e{cvat#ze7X+5Xd^&shb)6B7fhL`Q=qiOV-}pp1Uimn%W}Po;ctS^~q~8CI zaKBtJVfM=7{ZINc8T2@1%Ac*xyZ37w^d5M59edZUZI!(3*|ufaWxiZZl3hi9LT#k< zjv?yQ;roUu-@!G>f~I*I4VBJsWXB)$HnJL3 zD8c24b*ve(>0$YY@p2i??guiFUNTIIijyzssrADqO|zobhg>^4Qg=sdWxqK^j7fwb zbg=U?uk$`L7m=1!`4{jliYZ4HaBF{*^*tvT0vGG8;*$$_CnIigRB2N0?j`SdyMWy# z9Fd>P%m>WugYsm&y`m?v^AZrL*`Ft2wVQ^3)|aI-eAxLk<+HF8K&QL`xIb#kvy|yF z)MR5M#lHOgh?B>C^T_WgF6#e$SfO32!w#N}V~gRuq1Sr=JFq#tr$=<|#m)twp|JAr z0QS$DE|DUTZ^9R*uw`7EHE>dkVla3V7pbB#d-YnO^C3%&wnJBj`am!qu`kg8gKSi> zZcUQh5PgX2;EfN~SigzqOLfka1clg5a`5Du&V7h{;v33XIUP;OcU19rj(vnUpU54% zQplQ(*;ka%Gxyr)sl&5v$EE}TWRESQy5654MZg(mA#}RJ`;LCa<&DR^=l_mm-!E$k zVC-`PJ^mjt_QD8cBSbrzY~V9%d_U^fxu2I%9mt1)L+=--ru;x56rcj6T=Z^aKi>NK zqp#A_;sUEB8y(!zv_<1nX1qQI(DOD@ZkfMIz@psT{exhS@-3G=0U`WZBudu9e2i)R zsatGra|6(ibENO%8;U(DrZ;(?xGRG%bX^P6lg0z#B^x@|WaiOFz+NUR)T;?J<61H; zicTM(JH*FhDtX-y0Pgl^0kq6rFw9FuEYvrSvFA9B%a~g1owPpFK0$YgjM9vi@koI< z#{;(6HHtf0+Md};nAkpY-T%YWS%yW`{a;^2L6Eo+Q5qEi5s(s)1_MO8LArB5YKCq_TDnWR zdx!ys?ob+LV34kvaj2nrj`#omJ=b~3E3P^F?7i1o-_LSMTGPZTXX(4hwk1GZ_IX4X z{52FRmx_~&dPm4<#2<~k3GV#J+DFqUyyy{Z0c5_~Ngc(sXo~w?EXGUs9gW&PeZj>i zn5&Df?*(!JbNsQDnTYcx5_jIET_?@Ul+&|aH;F(?^krlw)(Kt_nUxuuC9?N$J9_=pwajZ_c)5*loYY;!ijQ^O5CXW2>+t-#0FF5-33C5if|zrA3#6HV;~N!#j(2+5elP_d2=>tUW% zKaKc76Tw6jgPionvq#d+E(Kbr^h;}(NPs=ZJ~?{bvk9Dlf!ddN+Lnv7fG|e z_%@28QpIK$3w(J|TUWPW>)V#-lNALS-@vxqhzBnpG`)$_5l61mwu}7$J!)Eg2+}jX zn8$VEZ$SxyWU{&hIv?J$AL+*@U z&DEI4aU(|0yzhc?*XX8HSU>7$yYD)ArRNEjC7p#YZ7|soX;)~={gVxJp_PEd?-MF@ zFO`wqg3Oh!D@`B6HQVgvO&==gKjDV+fR{*Kz&jpb`Jqq5O06z`h%J~{mKaU;w8Q0ZpfBM`vf+_43@q`F%1FIDV~NVj#mN zDZ#HM=l_?2Kbpm-b`NDZ9clczo=lWQPF(OOC7XkoE7Ls!s;{VDm0z(^~ zYc{Ljkzo(a?(9c_AjpbEA-m6TOo*q;zy~xEpiRp?wJLe3cl-l$yP>;J9j9xy_^{H! z+m(sp;nqm6dem^XaS74mlxpW5Ij!sV81@~vfXqupIWp#F?sl^s>%~g57~d%g9UC>< zS*XsfJ9c~MD2etbB<88E-l_vJrQMc8Q+oW6rKjA*2+9ChnqcW*&ds*29?sQ1m8nQJ z&>X5p2P+e8V%gP0biFLo)*xzb9H5g&St+W9p}^+ZYyzFC-?sKfeLouH*hT zQPr`L`RYV|le>EFJ7a+MM5&TJkPhV~-Ff1`ViGA&l+@=mS(@tRw?~`^Y5anQMY&i@ zqnLuIIpgf&COAT?Yvfj8Sl`FM!1d4be_F7cc)w1=zYKF}I8&G?uc@^ZmIAeyJM=Ym zXI^pk2i2b1!<0ZuC<20aQ1hGp=&uET%H8)@?5JHi{oD7aT8M1xMje@`*Q2rY zheC71bMzRR41OTuoUxpG>4rD&6?N z*toute%|~0uEyuz&iuvz7JW`WT7L$iZ8qfojZ@*#XdILR5*L*&B<46%+-njJf%_Y|*f#KBCu@P6Rja18@Z&@rS*0#Omm{tjO4hhUMhe<_h?T&~hDp!5I z-h`u>69v?zdjF=A6#&(Xkw;R!g1GI#?*n_s@1b6SKHzsiN@-XfqCb97Q^^dzB$2`cu_B)^P1PkHjkz%j-$Nr6c~bTx5K(`ghDudp?LrPt+9oA+sM3L041=RqC%Y z@C6XbUY$3T{e~*cWG7uxXJOJ?~B{7tePFq2T53#wjCy3${fA1$jecX?V-iOxU*_4maL1$tGJUQxdhgw3@=6E z$&LOq#vd0^O4C2M^n2dUclB_lgW{q~<>|+RYOMSeBQ+{4&h<2JMU3D>a38m2;-~r0 zI+J8>!)1gmLpk9flfPLSCZ)A*jT*i5xVId)aFq5scr}G1oZ1-fh^nN{?qjr?Gb^kB z1<`fOfbS9n^Vu|+_tKIke6$=+igOZj`bF?ujyY4Gg9B#wMQHEx0m^bXLH%R?*#r!0 z(kXpYyU6blS2XsxE9KYQqi|`rG~Q~}LN|{2Jna!v_aHu+yIld_K|!J+(8Zi93Crd0 zl%H~=6JM}rq{s<5&aFM?cE%i}M;QHKxR6f@WHhew_ExZXY$X@cr#!kgjY{GdDgjgu0K zhv9au1dPrJF>T)Xp~5O*v5z${G3c*N`UPC|!7Iw%#!-iX4x@?U+(t7hfqFpO%%#@( z7t4?9H)_sRHkUvk#W~BZrt?Tm_rv+-N?rEyx&=6xl{SA#c--d%rBnw_4{R97wGx$y zK+p>XI-phI6q5jQ&f7qnW>q5G!*OKKIvcX{B1>jFe372_Zma#QBbtk=c44Pa zuC`PP_X=E7V1}=Zd1z>Is^?KkMBK`&K6WTMRBe-NSQ3(gU;eDk{Iq_F9pI68BaDrr8=BFppJh&AYlnYFO6ycYbjy zTnv5l+|*DRk7WW%-Lk!EKV(rwC-Y20x}K}@EeBJZU;-;O6O9MH3EBnRD+~G)tueLy z34?65+)lq9d=wu&Q;{|dd(rkeLgw{=Nk$up6J4b2#WjgxR=(xyxWbrV|Jn6kf#R7e zXJl5k7|RRIfle>7G?62^qZ?&XP`~TEJU3ZCC2g^rO+?M;i?2WCNi)pu-%4@AOlfnz z|MLcc9@(ZU=tZMn#Xq&XBSH9v_UHOx44q*9+lazmHXcFwL*eJ5^KT?v|Z= zReu!EYWgQr25~U2H_6ZosTFtYki0e@gJMS3E7m;25Vy2N5AAXT^i);O)swD2{9tBb zwcbYK#rNsG13h)&SsdeISq~eAW}_+CzSV1P^Q(%3D&8Sv<_w9;0-~C)TFU+=pp9U3h%dkBsR?vGF(mbJyOG!Pli$6QZFuuTJY0gYn`U&>Lqqj-!iYN+Lsu%e>Rsv zrKZPl@OKG$WVL@T7dkz z>L6R+jO}NFbkz)pI!k$Medur3L3PHC&sNTwViy~Ef_&_Mm1aVvR?GKela2Kfw%ba~ z_yA7RB~!w5f1gVD#o;i|7%YLG3xInGfQ*~wL}Do<&w6U!iU#c?@&&fS>who=IbF(* z=5;|Ym9j&75|y7yyjQ` zp~wo-!F6#yt!k@z79<+rQ6kS4t7ELcSKWNpa;D~|gj9>6JPzpWmOuwvkac~4lG1sq zV_NNEf3S{ET3!J=Qf5;QgHE|ibM$H|>sVCN`WJb*pGQlTZT0^7bvvxHTP-GVmVv68 zgVh^jN@_>mgc|$B z_(G=e&S_vI7sOHa^nF{yVkz~>m~g$@>5JGqRqaeS6@Y51A)o9gPKj~dz`G#^?_8fr z05l2ZtpGWH!l$1bNMkIkgMEcEiTbt zBLK`tFY@=zt)8up!(W@`D48ODOKKaAlX%{LMn&0AAd~f&LQV40qBI?LDK_ttO6_Tp zYz;Y|P*!Nnx)cmO;3>{3DW`3YZV86Bm-=`&n8vM}_V@2fCW-SMk=Qo1ACFS{V3l5D ztF9iaAM||{jk2*{IS+OUi)Hx^8eTH|?PLCnoX>A`+A2&)cquRX%V#8qrMJb-VG(00 zv3J8_mAO-wO!|Zm%akK!3z@kE=g{qTl#I({v>cb`S4g40_8cxyytS;mEp;KP!O3nX zU!G%LfKO)rofN@#rFNUuTJPyS7GNitz&yTZJ(gC6P^s=0i_A1FO?Vl7%~@4JuXoug z7mfUyZabOne9_oar%{Mc|9nyT$Pio+t^d@ot*bsYcX89|>*3r;Ru=W5|C^NmeVJW` zdFp3>GRvjDE}J@y{4KpUBH9r&FGCyemA#H`>Xvs<#nGN!`Vern&g}XL#Nze85`Y-#ulo(R& zLrb)uh?!Gwah9xdxWS1WujXJp`#lM?X9)AIn?x)EO0Q*sQSY)y#`c?FI01@FkXXK- z%yeZb#JaL(&H4xWalI$*(#FqQ7R7{>JD+G39ub@o{zXo(-V;f*2$l;i#y{hf^1rgw zUyB>Anq|O&G9>U{ifdj*a_2_RpEj;lSgJW!?|PMEKk=_VFj<-Mp5E-BxwYR_{YLK< z&`WYJ=>R#${wZniu;6V{sz*VYl=Z~;?v`idq%fZ8560r`QvE%-HfPqCZvcT3_B8~K)IDt;y0jYiA>&1>Z0DU-ms$5(yh z%iRmJoH3>kuYTvebzHi+4)ct*R}K9M4G#L3+aD9nO~^qh`GDThTkI-~|8wZ}6^Mew z=Wx(klzi5}KR$Z$eXGF^gl9QE zL1LtiCf}!9DP}dRRjmz0aesCLuA#0)d95K69nry=MSiCMyT3(;6JYn}JeIXa4l8xD zHQEYml^Rm_GH2>crOtq`#}&iRH69uHG#pd(MvD1Ndb&|$-7XldEOPW`^C1<@d26?? z2eM{5awP3r&bxQ#ER@-w&eFQ8W0~h7@1w&^?~hYr^X{B74j!JZ}4?WelXCKFw(a0W-%=ElP$w~-@l)j zKV}X0;KD;6`72n@5bWAAjb%%V4S9Yy%ovaXTvq4DXOtKKO>})F6vr3-aMIf=y z(o1h-KndAU2@Jb9-?V8sxD6q>)ffizR@eTX{TSwvnkzo{yP&q>d;D(1=%8WIlL$p$ z*`)=a5eYWV^mdxFxQ*uKh08*>20dJ*#hKur7-GQ>oC}2*dFE=lb83*uKzg~hPiP73 z_!fbCf=r}?GC9^Qq~mSipscdBl!Ta;*dd*jX?5QyN%36>r5 z{)txp`1E4bB-uOhv#R;rI60Yf-n;Jb-JFxTk53yXI^52FrGAY=a5B_=y4){wJgg7R zcD*(%`OWlzWapm+@bd1Fqqsouw-mJFrcx=*N!zSAUfsvsGLW)#D}=Uj!%lQ-YmL@2 zR-WjhnHsfKyeg6akp8LIDzTNi>#dKRdBG)SrrO^~1d9pL<9`)?B`=rrNf4d@{~F_h$j#e+)10 z{D7hfuj1$4j%LwW(I7^iiwl^D$&CuF^hoh{sbH05fcdW@%e!p8Vz$$U;PEr(ooVVO zapn=08ShFku97i=q_8pSQm;b=;A&%v2Ei$G_+!Dn90m__chRNbn;ea#J-1orJUJBi z@TP)H$m!C_*DoZ%ez3$3P=6%Cago5a5B*0iMnn?JjT8@ z+)0cJer~ZY|6NgBbuCQIC!L?kt@oo?x@GP91o<#la$dZ4xX!zJ+RvJ%S9n3dFZ@$T zq9aqyT3CsWE{A>5C^9cI1WT~rq#xs#<+v9J!jFjx4Ie|KmbaO)UEX%ZUNLwBtvaIw zG4G(5?E8eKCUJMPOP-P*swmQZZdj#M$m;Yh2M01jQNF4o^{WRL5m!`wBN$cPBU`xj zToCTn*rWW-7`DJ?~7CAW^r^xN2?$^r^FI{I}^ItF)ZC`YE1>3!3k*eTe2F z`1=ed2kH)C3FNn<_ND(_dEqj>cz=ePsOs$pt*@@9pSOeam>45G00Q700^2Qz)IWR8@xY4PwlG*qExl?N+~$ zbJh#O4GBLTZ`F?SlbJH`)nXEz{)3jwH08p@5)yTvbdXm~RFfFjxv*%jmACivok+t{ z-4qYxY(paz)HBq#cLb!7g`OS+XR(!@^m&gY<LE0r|NF-pJENG z`ya(WmgLl@&aZ;~ynDN~J^D%;q5kE6I%0`sQ@tk9K|u_@h%Xspfr-P2^wE&7YCh(> z{)iagY)2-KWNww#Xl34`7dAnWscv6ZKkS{Pd)HG&wmLg`+t<0_sUrPd8+1D6WR)K7%Ae{VzRtUx9X02%nMv=kf0t3X6AaT^3VSU& zJmq?>E8zW9gbhMeKtL6(NMlNk`fh5qYp3{9j`3>-g;xyd(Z6fR0?w1=eV%U9blcE5 zGQ^u*P>g~qHakVpY>C$1?4Rf>KF8n(C>fOCBvjEYEH1lEzD`Km4qZ{w)qjSa}EJdeWy_H$KveZI}YZ0q}#$SN8MT*IDAu zkp@PWX8mLBrvPRA9Y>Pow*IgMx`oGbRmXT3Nb^1Oa7<4cb&55Tz_>Dv`dh}JpQ^1@kV{SlWs$7incL)K1K#dC~ zzR)_;OV-1u%Dmq`7weC0QQK@DGDKsiWBzRp(E8gPaGk{EEB6U%S7q3`KjFZk(URZp zpQF#L#wosDoEIHOJ!}+Dr96Q z;5x$Rr}mGUvo@z!!{y(>!oE*}N4~0l=R`^K8-v^9AFzIJ>wC)%vJ>N2BgCS2gdN!I zB~Nq#DvpDvDMn+=6+gPruxt7Ihwyqt!|uDx18x*xmtO$ChuYZpzS=n+ zFyi#Tl5g+oo)(ZD!7S!K?gK!j4-r;M`H=C*mOYPp$K$foDkqm__vEfk&z9Dj{ z3yY}Cf4@@9vOsl!=nLtYDpbz;&9yhBr2D4@$&ZHW4vZ48@M{?&Z<+P*{k4RemaH`V z9K&M@9gZ2^zYLgDp0T*=ORRri$>wxb{R%unCLIs(D51+nqhp(e|h)l=F!vgc^D)+fy0zRFqfh8D$^(q`zq3ZMW>O zX(%I`amDy^G0f){$vG9P(Jb~a_9or|B z6Y-kfeL|*dUN1c7uuv?>)5UwZa3arJ7WT7=SflD=a@o`GRGDH8iP326E2B)#1wkz> z;~Q2z)i>1v@2!8#>U82jMWMNF_%#?q4LBX5+65O7^36i%PdN?H z4A7;xXICV!#E7e-5V&uwjm(7sAz1UX%^m89^|g2ve(l^! z^FGPkiguZ`O$Z(Z&on-%=7FCiSC>c6!%Z+($&gcf`%D1~$Tq3jW$lyi?%&uN8& zEH@~1Ev)^-`AxbKTDFi<{CI%3xpG5`P_A&7*?%vwj#E=>G3gs}dIQ{Cr9jlLL^Ds= zUt&|Y!2cxHZ!|!LVi+h+&S=^k^G(QrU_{iLCr(|a6Q>E*V^;@P`!K7@3{oE}%heXU z6Y+T-J1_fWd9!XDtR*TS>n6>@>}3mIvP?C6Oe_GtQ)aY-gnqn`gR)F6Jpk(L<2*qg2P z+3e?K8|0xUI6CK>Mj7FHD6`*)rXKGAK>^ObiieW4CI1Z$#95c^4Q|QM&-E$9rx6_o ztaav}C{zFK0)TAJL1)=>_ z^0P6!t*cz0Ni_M$D)Z1h=RKo%Gi&O^Sz}*i*oQknH40PJk?+)H;MhcqmMaJ=d}Dcc zzW>%x*}a!7>J;kIBBZQ*BQq6J1HIGsJ2%TEC78bh)d1EcH%WSh&=^X0WeYYKmC9?G zF|n%Fi;DvxwX&+CXJ0L-Q9_Npr$uAcliU01K2k@A479H-@|jW^SYl-8)*z?*mX?^- z2_VH*JeSG_eTmA6Frf{eANm%r|$4a@XHBPn37da4^slCQF(D%x%wFS zCkif8X?T!VwA0$}&}-puwhX{GQ~aK*uk(_kq4Dy}7*;p@2d^W=gI-9>dRTU=)c3zP zi)=$^ed|S+fN6LVI6mAZEMOv2&U8p%#WW$#hy=>(BB3Sz&m$@G;rKmmA9&juHHT(( zfU)LuK1?_t1bex#q#c*;)MJ^f9bXd9pU~zPsOWD8Jp}UF`tWNEeURi`#X0fm=xMla zIDL~sA;Z_G7!GsCUVKMq_j|Nr0efV{rQ4S<@Kza)c_G-vW8|0Iur)KBenDZBo{r?f zh${b>pO=e`zaROk<%Skd-H(dzFM1(Aj(TI_xt+|80%M7&%M2>nzb{QR1(I&zbK<#8 zE$A7F-rzD7;eGFD#HT8CylRs?ttwMqy>|F)T(oGO)!(wznBj zsQV2=y64}t9`OzS1A3YA?g0vnP$#;-pDm$m+WHtZHOAv5e#hJuD&y&xdls}+#O&Nd z2~BKQ$^UzpDu78P6W?O~Zw_6WoNc-2LQ*whNm`*75>7&+<0BV*aS`F}av4QCq)-P8 zICOGqHNUC);l^1Wg>|=chqMun-Yx^@(ZKv^L1LqN!8AS4&qn%Q@q>krjS!_)yMcH}KNR)f}w*_VODx{oWSAO^%?JO(` zZd@hqG{x)nCD%AZS%=eMe?vk~|D5c`$tjtapBw^qWI^d|NlqyDOFs0+{j$A9&fw}F z^aPqIqh!JJYqJK1Z-ci~1&SLSRxTIr((R!}` z#W@dfyD{#kUyA1rzRSW%M2TyE99$c)$Oc9)R6w2uC<^y&d_LPgwOClInUJgS@WHp? z8{&2Ofx`iE;YU?B#riy5oPTsZaZc-}s}#$3)HB5%`eiAQrowrLPo!ZY4E?mHyZ zDITeS;-fdMJ%Bk|=_igKtvRO}DW(iAI3^Q3YMYhP*RxW||AHuYtYe#~!hM`f!a_;3 z7t9*AcI|Y8eIiVA^1b!YvZmO$^>vh=72bJO%coxh0=iRDaaf-?U$*2uxvjF`{0hGz z_63D|(XK=Ta7;%`8qWx17HYba`NHZO)Thb$oxg3;?WTXDo|p?HU)9(Bb(OYYdE3_x z0jTdF%VQv|0nirt!HP8~Q%udLvr2N>RW&eeK%224fK;D8Y|I@{vB>BOF|VXK=peT* z)Aa19jN;U;6w>}wDcydO>M;~&e5TGi+*R)%w>^H!x9s}g#L7U)ZREc?09oO-s?R;B z=y)aEZEul@jaz(G%r(o{S&7m2>iYtnK{8Q>jKvtsr-R@9S(X+j#r+JzGMgHS6{=dW z-Fa(O&DVaLSC1(!_dd@wgz1#Kp3BChE;k*Cti=4J_$ALVt<`5{w?5%9hqr!9s*N>Hl8+XnQ6}mFdDco_V zkBq~LwMbL{rZxi8y;M|p z;nmy8!wA(sFSpIOV}s~zX&h%h6pa^7@}FBZP|?^ZQu1f3#Has3UqHu+1&@yfBM}zk zC++(SEuu;lh$VD7Q%0iwM8$)+RAg`IZsA^$M%L}Yp1-ynYLphGW2_@ZWeziF!fL3! zta}<`QTWYl$_6pHzb5?F=kXE` z`9eIPd6(D_HLp@1L&}|=cBxUZFj!eR43kUgorWEnliv26b6T!eyx48_!_VhaIdoI)|4m_=LfV4zxe?s+tvO$JE z;YcdV^xR2=gJ&|Ay>R6`6m4f_Geg@rAyM?88gR`gqW14wa>{|R^HkeQ_5oLD?Tk5G zTZ5%Ox=$g@3hC>>QSNo9Y=n#Y%+td4Rxez}tziLKXBz#r$(A(lrav!7yGC!vS}bas z>A`2tZom2Cvb=9!@3#UPY73VKhcqB|6lTc$w&KJuI)6G@NEh{4VkWw9{2lto^$g?BSIIBzqs=5A$=X`V`17tioK6#7d1^qMT^=Ia!WxWT_P+ocZzJhf$zQ$rsAi3f8BBSZLMX zE9^4?)%#YeV}8jfiFSA~{iu^ua#$*wh<^pJ@&%f4!%m2*W{TArQhxL5@e^0$zmB3t zG!ksJLPj`?1aIIc>#j65p8Yr(`|ERKDQQG8E^ol@FgUjwj5Seg&uI1J18Wsh{gIvh4Czy$*k+3 z2@58}n&_lVD}jKJQ7>e!v=m=WCZx~c+`lnfPfcSt{`5DA2PJV2d* z9Zu{tro7SBN+`zPL3RI0arMj7vn(vKF%c$#cg~28&x0|4Jtbo<170MMHV}Prp*b*4 zzbp9b_iB0rdKRtrJN@cJS#m>24~#D$)vM?$i3`RxK`@xSh*KZXh6lS z0EB)_7o-LcHysFD$D2u*HpmXX{-2rbO6#$k3CU|;Cn8~a(XQ3aNu9c`Eoq&(Aa>sw zE=IjA|DHyF>P9$i05B-Q8vA0paw5uYMDzF_vw*>`+>VZ1ozOdE4&^z5nbDdY?L-ue z+aY1zRVVWTp17-ECEx%u_r+(r1JJiMbFCzjN}}(_*FGyU_Nz?cS=PMmQ<)?F2{zC@(D{8j`IgZgP?a zrPUHJnX@ZKo6+@jzGR+5cX9Gu64LNiYvfK8h(^yY{p^77KJ<+7kinHe?}As7#`bq# z#+;y&$q^ua0#;L^^*y}jBFlBhJ%6KhCckQhAW>Oq&IsObwxLuBD5w}R)H%I;1wq=w zB4Gr3Bwd(h_-wgP0moQYY5hftL6z!7;4nFvQ*OBFSoqsal-iE&}f&T z{SiIaJfz8^$0xl*N>cCn2_;A#iOH#aOu-#yJip#xl+k$(%u-@y=*4Y9d3vherG8xC z){cGi+cuPuM*z|`QVM+oo=twNOqG?glRyL?%))Z{kEcEeWPGqE-r=h|pY)MYV~-?iVn#p+{`*y6DhjH|Z{Anuy(RN5Z=z<5LznRM~& z{h8HK()`>AoAq!0Wm`oD!T8l?U&x_o;tGq?atW#Ta;(?ryl#-lE=lP%+MuP_q0tS7~*(0u5)aa*)O)x+$cij63;245UMHH&Vl zU6y!5FKBb41N`Wi2`k|6ORrx-Tm9@bbmdq1 z+!t?Bx3ru6e-mY$HMa-Y^N=cuEA%+EO))V8)y91+%bW{6|nKdPqpXhKEV2Y7_40TQ(k(Qs65jl7X=Uh{PMdP=9`a|d)l5L_1q*JB! z`d`u)8khuyFlv`yM9+c%YnEZr|6@|1g7`iuW|Et;tB?HyE^0A;BwXux_!Dy_1QlBP&?#~xBxt;? zj%cV37o+v1ZnJ&a2al`h*zu}c_y^R%%L#28dc0s&SP|nD3>95q7|`(9&F!!O{89Gn z$Edakr#ODs+2L|KP78=H@k;)>Qqj4f{n{7Nw?-wz! zym=CNk}8UBSk(sB@Mp0P6@l4QGPPfiBtrA7qbaSpv9;{+A8R9wqZj^rO zsFwu5R=M}tBRkF6!{wRWzb8|`(#qlg{@2O!Fae)H;E|ZG$Q7-UV)D_+CDcf>wU;pJlXl0Y?30!2cjzxBO)v!B`frA}N22S!4$w z6EHupT-nC{j=j^+#wcnC#O)ZHMX|KmlzBb$)E2evcj-Mp@kKqSV>@?m`NODBA_BYh zSPX)3xnS_WaIUr2kzIOPE+q@2{=pcwSJII&n1MCMe=LP3UE^*}+=MCm{KjWycFUiOE(9sS~UMBCr-K2{i)lv(wS2ox$asZT!u& zVt#R?^jF1x?N;meMYi)tjzB%dHTzrUF=BqMo01=ou_zvZCJ^J7#5wz$|Eh-4#WHFq zn*N+kiwmf?mH~9Q&D*|zbwbln=9k{bd(K&jJjU9ElDL$da(ZCf>{Du;<*cmaP8R}! zTSwDGpN z`~?ER1&;=|e3Zl~Kt*@@Ti_3Cc&_?tk| zm&lv^=W$F7J*$qP-lB6vIt1ly>}U%PjR%VpaPkHd#|bV3o+T^NlCr4$$+EA6%m2^1 z65lw-`v21~84!>R>gZo<_B6VzjYz!k{tzvO(LMNJZ0;Pe_jTm28klyzx+*I>Edu8^ zGpcnsB?(0KG~Q&>sWga_8(c@j(Y(rTmvWaPn;CVP)=ly{r%_&ma4E9_g0X)#4>*kr zeRCvpj6wvLbqjr{0aHCmsXz`-D#A1aN0W0o*)PicZ6M#o0aE;97#QpPy89L?N9NCBx@OFmagkA!MP^SmHO>u!_8B@$WP-G7+qiAt1Vbxf0R{28QP*! z_&1*Rgz{v`06@(rT+~qM6dfO_@0I@myds)G$U@2fQthmRJu&bianLfU@f@81!IqF% z>w;k|HPo8#M6-te_T1H&qSMW?(nrP>*)cRi(mOLH>rFM51ZTHxl1sxtpU4@!_xNh{ z2cMdE;08-p3oOFSHq@q$DjJjRJ-<4Mjr=b4b)Qm`8v@)TcVsT=CpvI!%*oZfzkAV| z*KyCj#q#~!$@AWPagj^f#eSLAMMBguFe$RVZ@2gO9vW;7(0JBM*qM``ei`Rg&e#Q- z(4QaC2|WrS2o}1zQ1o4+ZyCm0ZA@W1#rr9X?wofvc@zEuFZoj;Iqxx^H9VWurYN^< zTL{O0P@1dG8ea*ZOHp?r>el8&TB3uiw{A8m=+z{oTp zz$Auop$cQs8d43av&pnAO+P-SU>@cirN%-ytYhrUO+Z938zWq)pSND&3b9S0l8UZgG&FqI5^TBb=`rtoV$B+i?%Pw#e8OY3+3LVhCaM)l{ zeruJuH-b9x%fY5>L%hZlsn8f7#lFZ)8@M*=?Zegd2+4UsRNk`wzWQ9DucvgkFSD6V zRQcvLFhVFlRYSKA5O<;~GJN(2gZx_gPG%zM1q=Ye*q&)y(&~pLhxS=q;@^*XSZm8) zoguW_`+vBsz(cAJp;;;6PKi#MKGqM6+@8JAY3rixArY91=@UpuQ+6YAchdRzQ{ufBLpIZD7ngXc+u^n;!D_fDH`eYOSHY1I z&zFl%vc!)#wxvF&kZ+te3*>i+tn=TYmBo>%+_j%b*iL=mt@;Y0PS7A&IiUK+(yJ3r zq2Bvuv{>6YHFvoD@+IlBB`*b*<~%I{nNz4z%FP=H-;{Gc^X#7l`=9lTO>Roa5M*vW z4kovP-+a5yUPg6KO3ATZo~fxSo|J+A4MkACIxgzOq$^BROmjp0+-K?rIQFUaz&BHZ z^^1eCU%@*@;Z7(0%Soc+8obY=i7EO_YWq8`B=SS(-U$cq1AlDle-|qgEH?lMw|3YR zaoix$@S6Yix$*M?CUMqJgmrfRQ1a8&g1dT3#zF`)b&n|GUkxR3vnCt4#3H8)wWd0v zovG|^+`>%Ju6PAl^+#t0$uvINkYs%y@$i>S2(3xS@*;~lQ*t7KLaV#yaH#P_BdfF{VJh76Hn!uyOYms5pUUuSzQV z(My>k3*X?P@NI|tZ&(pbla^$`b*hdge;Ux4jq-zWy#*fI;qAb`OHsO8tk&T5V4VMqFv=R_f!Ka3VJXfO%erP}0#dbDR4-HQ17`<#nfq@XXln86MQ!6Je z(;uw$|G!v0cf?;e7z6fgX`-6d1T~w;bPL9=`FcrNEnc^6A_U61M$2Y2w{bXmAV0jJ zl;03%Z3N2XYu^=2I<{$Bjlu!UCo^d;uq znczuQ8Dp>;>JGk7dgFl(IvWgfOs2Opv}X!!2P`z+tkwspKb!gxEh$`~m+*iDr15Vw zh3bT4K&?L0*xR?pVZK&NS#!u&@niQhF#Ev zo1q(gZ+7?D-S2*WyZh`v?>~G7$9tHo&g-n>I5M<87i%erjDW6Q%Y$(Ho~owKlt3&< z780)b4K63+p-^F>;xjIvD=xMK2E@y==_TcClQdummaFNd?x>OO6y9gQKje<$~0?oe5|RvEgJH* zuAUF!d`jDmsPlamwUBzXGYq0XFD|`Yde*a_18hU0Jt!t9Si1T5zPg#ZkN0PFZt(mn z+KspuyArS})Rs%vb=s)#n>;G1KlLk5SAq5!5zDrG)o@(;Xo&qYUTP@^8+mw{QRnX} zOYn^s)siBF=g6GXqs&6THP}A})#A}P<>T&dLcEmDNxubmc9?ei8SvvlAE(?E{(=La;7{bI#~kW5alVYWNqB-DyWqm_q=PZgpb3R z##DAYUGJBO@#Uy(0m;wVbpG&Pdx?!I-yj3JQ0mX%*I7EOV}>Hx#bGZZ?-nz8y{^>g zoGl(IEGww%M84=VH)i1fjWA%}htwBV7QIfDmK%TD?V&1fEcRV$S-IDG_YuK~_tSKP z(Rt4Lg#kJBUEBQ^<)sUhFdRA6wkIYe{D&G|*GTjTn#B8X-00;BKVE-LAyS5dWQYDy z0F^g53}i#%KzFP*jdpeV3an9`@NkcpriHr6lQl4O73`hu73J6U_8B~ z5~5Ql`ij%CEXunGsBshC9Jl*N|MX3Mci{4bfOFd}OP-`cjwG>5(V0dwnw@P`w~>N# z;&9y`zTq}=W_0$rN%;q7Wp&)sY03ry{Q~N_0KfjEs~-XP2u6eE=b_Jl{-80Fwk^vfcJU*U72;8GSIW(mzo!uz5pb{tKEjhSjE#u zBi|C*yc-87-Ev!buF``3@qItFrvP7gMRU!o>97H^$u#|G*y^lJGgxv3JF5T4v)=iBuY9E@UZ?tq;}qSFJdKi7w2p zksyi44|`vK!e+|;;jsKcxb+Fv7q~PNZL#xrqDj39Gd70Z?Ov^Pl8OKNMb>_6XS1QN zlD`Jn6vM5_(E-pPBre12Uj6c+QFer0L0BS@o^f+9A?B$!a;Y(DV1W8W%NFB$cU#9` z=D@XDcXj_33?&9RNj%;cNYW&DUpHSQit^5a(9#_w{`k%|J6M@(pg>pj%`2$)!DwOv zL~ZveOq1J9>%9MW%=k8BSnV_OjZ^%t-LG-Ff8E9Z_L+~D z*ud=Eo%yi*og(eOzWC$^Y((+)9Myka>wo*C?*SDs{LAg<`u~xz_W$hQXt?LJt#LkL ze;|JU#FO)~!0-=7VSE1b%>MJcWJrKjqIw?n2dl$BzxQumS7!jjpU*%DfBxrZ84K7k zGB6}9e_p!5fFBrs4L>>o?4O_IcTZ;%j@RvP9dUmg!GF6f&$0TvfZ-c^lQ6^opSk~! zb1jYm>=-1M^S__4|M|KY?KQ24?Wq<^oTCBEE#e29q;#54XV z8b?~d$wT!grthya%hitbnk^Ix%S%e&`R5nw^+z%dRgZ0-^Y_Sqe(n*tfL(!pjAC&9 zXXQD5$pJ_2DoYy%P|f)d8{wbd0>&eE;^|{d_UDxaf_a|4*9F#}C;V@J{^W?j^1>4S z={e(?LNX{>bKf83C*8YG{`13mhlIm*F%_Se5(CFBKI#+M_{%7nQQd_?dN<%dvm)dD(`GM+LWqlnDE8~P%dQ_Yn3SgRw{!Sw_n%0y+l6tPKbHyu z&ZJb4o_r)dhjEgu^OZ(-o%1e>>xAteIfeHl@JAjL|Lq8UyoYSf8Re7`7F3!|tog!?u2)@p^`OAfAQ>An;;U z`=V{xMj?q8ce**ybB?{CJLs%o*d02(AU^zEb$~^T^pH#E@kqiIKwb66ZMzWO?Q|*R z_7O&E&86sgFEfTEfpgMS+e*JbMObPT;>~wJn6xocpZ0`phzD{@gUYHZvOGh=FrR`t z&;iWZU}Bk#w#L!HghH_r(zf$AqK)+@?Rc)1tQmH@jTkYWwwhB)LcFTAMr+6iSJ%fk zhouU6_IaOHZ8ja}PjQRG@81ER{B5fBgvXJOcc$dE`#Qe8kQv0KWn%z$r2_k_%}wXou;RF>+tN25?J z%))*2L%$_8Z$mxXj)sECn2g)+*}})`%SP0J)es4ljZyC=9QtUThPovyuouKrmQx*k zPTW}5%`j0UM8rhpkx1sN*{EkZZL@{38=!VkMLc!vCQYfD5w7hV05m(AFd47dm>Kn zpzc#@Q>J!@b+ZHGiD;Db&SPzeBlnEXpOv%!7P!^4kPL)cqkPh`8Bh4!;Rt}sSk1TI z5lR1gSeKQG>p+i~vdZUW5iQnnvdz3xOFYw2e|HF4u1I6hXp%$5;aWbKXc}HVy(N-7 zhXN&%r>Y{ilb-dA$8pt^^_&za<7`N&D#2B+>(kUS07^1w&kU}+l1LbdRaaWXvPH)#o+{gq#F zM=ck<5+v}ov(ivW7j*WfTgVgfE%_f$Ximo}$8Sk2I=I5;+q~P9%0CMM3YA3&IT--` zS$M-oPE{=pRSt`#cc^=_iCIFoJXVUziIRzB^k1EmMh^2Hv;c;snl(z?(|2@hgj>Ye zRD3i-q+Y_D+AT{u>N!o}K;^i~uWVoPHF7m*LN@UGfJL=!&O8zDL406lSEt zp3NkQ->u7&tSof$jX3Yu9BkS#3KiRmJOCO#-;?ful0n9nPp{YusZ@0w8=*3w49`{- zv*}j#;ET5353(O_bdxtmJe73kYEfj#lpzbp)NRqPa7i1eHC{HL&gak5=V5x)7$rS)+JW3JlX-(tRwoPx=})4kp2(N}M}LPuwB zwvA>?x|p8-S~sE;;8^CcTzCfvl$Py5wi1E1eOjO@lJJV%tk1OIh3u^~fWsN7HC5aM zw7sVV04RbqfVxZt_%v#aqrwSTG+VFNUQ(6jMDj>b%6JHl6%>dw_zL#oCkeWS7I2IS zQL_Q5$oTo->NO(7=BooHS^oE`=vJ*Y1_JS&fSdw6@3<*wGah`p0t$;wN5ahwaDD7n z-FPf-xyalLFT##*=Lb=nr915^Ik>fsxm{1^S8V174I6HkFHe;8K1`IFz8&P)w~Eu# zMJFKdnRD_w*=GvU9=@L#e3_@eT`nI2Tm;C$E0~!JhU_o=ha+B1C6N2OiGhx90>#M* z*aCMKOG?<9*@XyK*LkP(`SXm52kdxoA=4FL0jFZ|XAIM{KhX}4`r=oV`r@Q~jZXae zALr=M9iVRuRhgei8bk5Em3buR%IC45x*AlPTt%_{J zT@AizDqZ*~WpW#IGKX3pAjWQAuctZuE^FnUTkW!}1PufAGd|}x?I_no`y4T&ZE94U zGsQGHVrWk8+t7E2xqBKZCOBTd*U>x7gbxn%P@KN-Qu7j9HH2R>FZ$Wq2q|ZBR(LKw8{oaqhvglba!NH{LX#CAti}Rk) zedWG#FdA{5$IYo;Hqj(1&c0+4*{(G4FN8rUK>*%>ZNR4eS_k5ON-;ZU!lbJ#!bPH*;UHlo)X7>c_A z`54SCA+3`Zvy0#-oa^_khbyOe0G^8%M-RVYrl}>?+=Z9KM+#=j&rn1ALZeszyI#nS zjvVoY>>EnLJnJ`5`@4=%V-yZuj5p0vtSAT1bDsJxEJIG4wP|die80O7lLInd2z(~7s$NRHoG1a*j_xg>c`NJR!JN2_2zOM^AF#IL{1{o zxR%5UKhEUvb6sa)?p^y;u(2MnP1cHck$KcgVn6p1nuJU2_-k4%H?W-(G&4A}On4Y< zpgoo=z8(NBki6kDoBE^<@TW&&_cRYC@r%*@Xt0JbW_X;2EZ}+gzf!%^xC(4?hL+>> zoG|s+QxQl&R=a}td(%M9i315aR2TX~4`mIBw8VnihN|3VtT~|4%C=_5ecBXG>zOc? zo=9RUvDdc3Y89rooSwIF)w&%p#Fy()vQu#{Cr-01P-@$4jm1~7t1n$fEu^d@bff`L zoDIodsI_jX_B#Ch{cywW&{iPRXo$EIPzN%cdb?iNoRseUc_1y#7G6v$>N#4hawi;I zSy`<7Z4dI@$12$%Up`$2t>Kc5gclX!Phe<6>y}q%yt~lWJ8j06HMf@7MgVtTcvkVg z-aKm!WZ&<4bd$81e&=QqR;5pp_uB4h)wNcQf%(mMQD6fQ%Dy{;GPiFvn7dcgEj zd-$6kS|B3MTee_$-(52-nTiko4qvL4A9gT>JQ}KkHBA>EwOzfZUmbe&)FR0*rjw%B zZIAU`Y$zlvE|d!tRo<}xm&PJ6q3bvB*{3PbFVy(e+k@v?T}<@ky@xV zjkCFr=v51~US~|0H5#^>HZ4@EignnXphys)!kNj+y#BCuoRjAb3`b7 zLai}1Vp@LX%_}rH99!U&UiK7Qm%_PS;6l;y;?wr0_p+?x1s}e?jf#Gcj5}qFcb7(l znHz9(v#oPpzt?A%>VDAh08rkPs?g1G7XNt2sgGpPX)TRcNpl6i|Fj$CVzqqR^4ZWG zti;>q$EXg_3}~%rp4NM511{>`(4d(-1zL#~UQ4T5o^=?G(ss=H!4UvAjt_!-W1Vk>d zY5YagQ*8uExS>n-7~xi$Uf1IVw4^fWY=5yO)P4UW52VP_Thrv0T>^4GQ>(Otx7cI#??Y}2yI{AyZe{1tG2d@ zT1U8ldre89v<$Fi2u~$KtoSdX!UIh4%3D*$wY|n|gAtdlI*COO#X>GobM*4a>CXPG`_?kNXqj z>XQ;_H*_Ns8S|MxQ=jvU@ z`m@#^4Yu?MC@S^jM{{dLr-(PNdVYaUJAyL`?OQoYv^1Os0g=9OWpu}7INkfV*6Hm>8wc=`WTV;%Ow%RYFXcp4 zt8@LH|7M5_xh*lD!JbM8E778!EHIJ^7dh6pLl}M1Qq!t`R61ODS@_ZKfj$<*cIYs* z&575R*A{`Vqp9o}mwm_Ax1VRtGF&F7ure^_9DfuN%NaBYO2D!xzD7KMuogkkSpSyOZak>c;FgH-POk$ zc4O9PI!gHN^7XVagPRd11H}9Kgo=SPrgbk#>$NOv_lNaG^X>-XIqbNCK>)b!nEYTf zt>M-{v$(RxW=yu!}GsuzK;B#x_=Vmi(Z2g4%)q@Kr@r7JIzA1vx*K}*Mr>Dj{8O?6tS7)i!PCH^jz9l3!n z-EPjF^`$Ltg;=Az zYHkOGZK<;xo?8z+WA6QMYcnMVKl%Xwxa2g~gc}NFH0!RI_d>JMWj^LO$nuIjWn4Es z|7}1Z92|d>#HV}Slf9j%K3M;m#sTP&_Iv)0-ya=9kp$2ja5Uaf^2zb)j&&_GW~>oO zV3#TYyO&PY=P$QD05GcEXNxW)?nj1vUrhlXZ;9#;EhNXUqmc)! zK`3o^K&OHd)I*$8bH)2A;^PbUk;@)J2mxCc$M$7z^x@d?m2w~Vt_Us;TEzJx@dMg|Ibb)-_MOF6~=vJ@U7TJ7`6Xsmks0DBV(N0<@ zfL64x1|eEB4Q$Oq8mJ^=l2%qqql4;0##LOpWX;`{hqZPuH($hQW==8pRD=__Yf{O2 zN{}cnQtH2nv8=q<;&X;j41ediTxNNLyNM5}+1eYr>oLoI24AFiOx(!=)AP+i*Zh8_ zffLY=`cOD<)a(}KvJP)oU|mt}M@+ZIenQ=fc_|(7eQ=MRbXOevY>PK}*U{wLr96bE z)XRxzV>I(Q--Yn|hP37*?MzCRB`+~CPZIDU-az18lS8zdqje=I=qd=GGU2JY+|>&5 zk;UF?bMX9a8*3=8AKx9y3^K&O@eJ zc2$T2W`)Dm49T^*t$J~$98tMpo>VUpNO*@ zx|7dvFM6BuozuLO6xGcnS4sg{pc(zy%kv2E2up+E2!2Lsv~mf5v=cE@5tgxei0&L0 zyRgz8Fm3}Pe7)$Pa6?J_J@JoLV%&{den0Yy4?`=T@}s*?vKT~!UW;3ygG>p*%HDla zIQ55qM~7Zgtf5eY&F5=e<|FpEDM7b-MDY!N1r_TIBA3ufz0>?4@4^%`w=Pl(Q* zRW(?Gigh8ZLxg!-Ny@S~97iMm~8FHvB6Zv26yp0N130s>1(jV*qm(#_-gZNY0$#}^EY7a+&RB|!S0{`H$v-8A*dum3c{ZtDIrO3CM83Yu9LfjkeBfs=K z_h$QW%lU+$ZK?$0m-Z2FmHl)KXtP#mtegwbsH=> z3%z9t3*iT<9$u^TW!QWnng5{|m&4rEIBL!4Cg4eby3nzIXFzCxJa*Ae!t7?bYX6x9 zWYuYqS35v42=SVBra)0?T0h|brv9F)U(9Tp4jxwzp!}HIBk~7Yym2YvF%F-<(HPfT zN;aoeBl89=PfAx5ivA!V5V>g|sFv2Tc8dQ3oVJ|NRyJtQ^oU~HbMwWPyA>l216?N5 zsWFYJkZJvdl9@5O=fB{sB0A{niT2oL=mdvrI*oBAO1PO_y9q4xc`nCSljEbj{8_3d zf{=*RA}mJkf4W!F;-D6mX|QO>!ASwC6v#n9n%jJ{qW<~{@>dxVk}N&HWunZT#X9o_ zrv%WIJget`OAGQ#^j0~#u3@|ro#N>jHSU*&6h@Stnx*jIp}l#vQ@Eoz;PL?YWPRkk z^zY32V%NIE1}YZg>6Mx#2{y&YxHrBACsiecMpx8iq-U$fyEdY_;r)D$y1V$<8H0MB zMNhADF2>IFA}q)6jP7mE_`m3g%x)tBoJS(c0L2Zge49mF@##y$omY9nJtMF#y)+WL z5$d^;kbBwIZw|Fq)0!*FyMY?iz2yEZG(##2vT5gP`OEwVF7C^{CD*N0R@i(yP)D@) zDi|CV9w|BWqI~VhqVe82xgE$?#>TRfbP9*pbB_3q&qIkzX@%y0UZItQb}SQ6x3O?{ zS)x>A`hIQ>ccW=#xiOuHC*(2N2ve7jn(mf6YM)~z!4}J>eZjf`1@d?~hcW!U#Yi~J zmQV{CAU#k)}z!*iJ%1?D(j>JQD!KB^Ic~8#xMBA9o_gH>Avq)J;|L#riVI{f$ zQ`4!l2Fs{q@v^m-Du`QmlOHAit4XkrWb_&+xgNKbn{#=o({wuv`~@Gj7FNnrfCxKH zgelM{`^I!uwZcgl?AXT~?r6DK9!9Sj&$4QD5!#(hX?J@gc6T?*e`RhS?_4wKmxgvQ z!Tfkq)M}|I^;M8)Rx3l{g0EeMG};wtELCdQb}CGGrgnqs&#MXW(}5o@nj~;i*NK)cDVlW8BnY0fK zqH2)D=vAuKLVcNw>#xbIR-^%*Q#T9tQI;VZOCKsE4!kzL(#KNvprS*^ke0$QU2F}T zL=$f8m3gC6{DEzNhjOOgWJOK(SjN2a=53P2=V4AtDGjuT zpHKQT4%#ZGh0D-r|Jcoe)DXFY*zMUXp4A;jyrN3e3R2}Yd>-WWGC+` zDm1+>A9|?5hu9A+8{?%k`T~0Y=xOvg8N933ht%hO{0W4_14lru%*84!eLxvP#Ay|u z&gZ`4AYi(bB_zfkq)zL=lchw=W|BEs50uR^7K+Ngex|r@)V>J!LB^wh_i%TmxDF$> zue>ywFRiH0(|kc|Fo#bq_p$69zjlN=F=dOJA+-g-uHkumF8?s;cwC&&|LY`ffgNvj z4BDQ{RVvX-YeZd@#Hdxg>~*v0Ml5^al)~@*`~gM6G&L=4bg#JFAC- z!(#4NE);;m#qUj*m~(DH(AYmrG&?il24|eVEtO%?{870lY3b0T4FD?G1i-R9yYHxE z)TAS{mUTn%naU?JB;+eoZ}kjP#ol8bgE(x(#RO>ocz#`=}ZRoPASeUy}OV@oVIY=cpJZWEWa-DsiT)ru((5V`rCu=mc{vrMzCmUtXnv1fFbyNLb5vt zC(D~0t7|NUI2|8jUg~XL(y?bCBOFuf#+Se<8|hCvY?YD3{K^5^0o=lKPA)OBGwI8e z2Hc%CokD~WUCdu;F$g2(QN2gtrQC_(ECWD9d4U1}6F~xO9lcz_@6^h~ii`AwVzmRR zKmLB{=i5JMZpKMqIg`b7fnQIn_^2m-L=mk^(P>km)gazk-BRo6??Q%7n#_h)mvYSK zcJPkfbb!7ee}#MJ>oas6@bRx{kp2vBo#P@JW>vn-t1_~s1w5Rvaeb{IRjm{4$$6i7GwH+0TZGqBFO}-aYmxdj_RaFx%91OYH$XrUR`{ z`|rOI-b7Nb6!lKURMh~psHVIBImv%C6aghKwT*STKEYyAP0)Zf$(XdmYT zME5Vu6N3b&-$G1!uIr)Znv;G__9Fa;n1+|Wr_B*wOj$XkC{XvMoO*IFqwtr?AgVne zI1$d|Kpk|sghwGEA6kMVut(A50-Uf5n+pl@DEf_vt2YDB9k|q*&1~l!krPOH;xB7y z^l>_~Qy#J{Ea!3qkIU3^ieD#rF2!uEW42mC1zb*FMb0J7wnUq7M{Z+;kAy@y;X)97 z(2!lU$v)5HK`&}ERE_!G#2EK|;Apt{LP00)fYNZcz~Gr{cd3QymMvj|d zh?T)M(bU~DM`wqlG%hcIM*@?8r{QuuP3t7W4~>f;F|(&DL0N}b8GiQl#rN6sXi(wV z>U{SlQ`?XJ4`Ak1A_(~|V~cRQ7B)O-TESRyruA>VQ<4r_i$$s($xYIM1v z+2WL`EaWsn_q{&K!tET!uV^XPGepcLKU%LCk=eKMa*r)r9c_+AvPOn@z}{aT&;-2% zQcD`~-%M}XSZ9`s8cTcpUT;COB^qT$fi5}`H}yBeLSq=*zHz)i$C{y1$p3E4nSRy6 z&EIf~b2OVD)ev_@)VZAFuTzcyO===ssTxeIh>6tU^73H1+8alzA%EW+_~K+HP%zo{ zkIpAH-`O?~fwcQOmnHfysP;fC@&!{50rk~4iwpr|NT0{)hPc^mc`mhn;oU}Gj8HJ| zXbG{6-JVQiK@hDY3F2NG0qN_kW^%KugL%aO_1vzeo3o6G^d1d^_bxbHHFIjYEUhg8 z)&Zo@v8)yY|H=czQ7aVkQTK~Inp*_kn*He#W0n4?@RKxtZ!gU2!vp~IjyY=P5r)t3 zS&H3iiEAq8Idn%RjYpyt@#P1XNk03mRJiM69kZBGf0zTNiLICQ!SdjwxR>6$8YFzv#N9|m%zetEDs*a)GQedPBVjt z{`wn{P->U6f>~lj%S9k3_ZCF~ulWDcVKUzKu=5&P-t5j%v9v*6cu<)$qXXJ%` zar})ck|RxG-vq!?`Z0Qpg!qY{o2F(vH*?5-N(q~PluA~(M%$5)1A6vDv=5F+0;yNe zN!Sa=4>zy6PLmBG=K&ad(||lj0CV@XbM6vAM&&fNuj&1oy+< zJbSP`RfdUhBQ)n$sI(a^#&v|WMAY+3FKp~|6zU>R>a24zz`8%=FdxcZY(M+q&1vgd z&tU)0A>jaMCR70B@R{#~DFzKWV^^wS(fhJR>m~2ui_0I6JW!b)sx+2e%dS<6HFBpuC}A7fbU%39taleP6~)T?ULTP5}MCY?W+ zr>~Pib$I!Y+l9VLGQHpSz0f9EuBXCCXMelqzG=!7rE=Z#em>`l3@I$?us4z<`aNG*q4#?*BN@7+#G8o8hmht* zt4d4creX`-rc;xEb9PzGQB0m3JK1mC){(oZ%hVAMUU5!OdEfE?B3UJR!_TwO{tNSE z)ylWO7=N^m#ehYp{e$3#97efn3uR2!R@gv`#wwHuh8vR_3pDOyU*zO z&-w4PLPx&r3~e1e^R0Dyn4RAp9q9r`Nk10lY5T29=N^R~lJ6u?YvIas?*%eWksVa))XjLmWx!Ue6{4#p@`$+{t42;@;@Bhz4gz`jfAleWG* zG?C*%)4$d(%LeV_ln0LOURGL##21~kLTxp-?o%&lJm@y39iI#P_{+7KE>~AZvTmg= z?3*~3%veNc)EG<25x}>9)5`y;fzRq{`v;hY?0B!LY%k;Y12Jc~T7b0+8UMWv>WlG3 z7l1kiMUR)dt6>R983^b0Tm@SmhmRyNRfLB`oD`%5oj)}+&bwkRUsai8O}0i$kpJk3w**HjCBX$vbKF!8&ZQF3FfkRV$L$qFpdO+d`x61$t6iY#%zX-D{3$j*Ku|$$daX zD2GA(#<2M<0<&mLtxCl2^mPIU`nr>eQ^=73X^WMI~Kg+fzFm(niclh(BVWy@HoVLglHac;i9k zhQ7viFoKVWYYwM%MKcF|=$n5%PwLl%GAUoC(!RnG%`)JwJamhlOtqfjdFRqTkRg}D0a8PoT> zSsLL+L*o%xKav1G&YIrn;OL|7DN7BF>vE;4QkSi{-D4@w#~b;;u`vu?T`oKqP>M4q zZzZPXi!F+;0}>CT0r^3y&Pv$Uk z;U+R0c9VJtfSSJFqQ$=GBJ}JY>9YKw5(VwY7BxFrdVuY+@pIIG%Ya~w?E7vIeXr*W;VGcNr#sONiV(Z5@AYwP-Og#m z2S-j;1s;4IDW+c9k7Mq z>_;e@W&t4E5oT=Y?^but)xMl2-U+AfFi@L=IuBI{Zw!+Wu?l)nDm*Q8(?OT^&9ay4 zA*wmE@q*3EcN5QcRBetcud(;h!-Y;gMuv!8%hd7$Na^EDtMXt~+xKkXH+$H7H>u!t(zIW*Z0WW~ZF%a~|s$`SL1 zK=p(?4!z)SkOE!AgAcEcd<+xCKo?k5t1XtnY@FH7@3MINY+Hx9U-P*M_Y4QM-yJrH zJF?WEt^8y+pHVU#bZdx7uKm{OdlbRB5-qoUtyL&T&QM+7S*ix^yaEcz1@I^YWUgew zZifqwQ?Tf%p;%Yem*sJvIve&y-Bzz4-WB=MDaAl{Ah9^B(*ajt^ik!-{_MVOf86Vi zgUxEzPJaH?yGdDoTx&Ojaj>hsa_O8w2U8cB1V<+3=V2VP|^mP#a4*^vLZaFDz z+KOx#>;4X1J(gJI=)0+Fm49_6KL=OcgeJ+*w!c9#5zYPir)En zzK1JXKHj!yubpiSyk~b^T%lyzt(unp){{x_?ZHL*5dQ6356LJ8hRZIEXE21k%a;BK zr~$8=pF?D?gXW%%KPPX-Y+4*7yWRx1MF%tlbMD25h-*6wR5AC;Z;!vi;(_dQjk&0o zD@dzHz)P_ibRL-RfmdS52E3I?%IQ`)Sw`LoVk`v7wIhn7xLXa5w1VV@7o(p>%jV$~U9R;|eyZDNphdJC zS@?)H0UOr-aJ>bg%KNQH*ZM1<_=ln`eQYI2up!^o#V6BJZ>te$9xB6~;z&JB%yb|@ z>6fDq`>JK=7@ZQmSu^k>pQy8_&o*K}a7-6`pKWrjm12cET!f=ZZSz^=;^vCOBHp0Q z68Z)srQYsxkF?_=Jo7Q5?b#7>RX<~$97^~ZL!IU1 zw#`;pyd1gj7BVeiEOE3stRLSHen+`jZMcM@P)O!VNpE;skBq8kuwIxreDVC9#@7!{ z5kx3AmMS21z!3_z(|1b7lGW!ixZgX7j|dGqJ12y*qSsVZXkCWY#@-e{N+< zjTv=feQ{)V$or(5g^zc^+>Li{`@$wk=X-at>1007#FvmU>!FVuD-G5eE|U89yA`ds z5ftjihqN zV(GgW4{WTIpZ&y+3M;tvyjm`3a@eKawOn01ber$}Qc%~Gyr?&prnh%^n3K#;Oaj^B z0{?_gQM#>rPj}T|?!5?mjh)y$?njAMJ)Qz0BF!tSF3|5%H1*BIB zj^e+O!oSrC+YU>h$5teowL!SQeWHY_r@ax5)U@R@T^qSC0aO!8qw8JzZV7 zsDjrrr{4xBNOod`>j3P7v^zUq{IWX|WJjt>g!D?A*y-jM0%@@4$iN452Ww|zOv+v7&UTIpP!D*$Qhcgn_p9nniWHdC7oS(v$;Y>43q!ZlTW~a}?+@VCzr~ z@I59?5vFhyRoGOx5e`Js&!2tz%)hz_0(RIEwuD$2rEy784?%2nP~M(mNgtNRwz5Z1kkUsV7GKaf{RRm(kOg z+xoNn7QPTK(#1^`v(luwZ35Y4TzTV#dMKkKEVDnCyK~~usTZ|MZKeaDctcP!4`Hc7 zh6uzXIK7S6L9|OvBGe%|qoMFvZwW-xAcRVU%K3VB9Jn*CKr=H(*e5{ny-?;&C)fQJ*cs9Mn?HZx?j20(=np)`xLXUi+<5 zIkrk5ERzK{c|TmtR`R`_;tDVc7m2u4j_%~F1xiLUCbFB`4$U3TdUxIE2m4jM-*=o6 zZMI?fSBYF@l37Ir-4_Y7S5QaoT)^HyPh0f>UaR{qOE>1;!-9E*@_MgqC1(T|l*)tY zbFEwyYAbLunDw&Dx3$*4>j0QEgK)>!>xJ^-k@WT7h{dl(^8rehDa(Rh1CL8zo?IS_ z>tnsWVNOVV<;gKzR8NQihr>tu$ZU6XAig?$tAC~J)&?ZyLB zM@S1dAhJzg$sax$drXW2C^bFRW#)lEEphD9!H<&_9ED&e z>nfsj3DS+Ebi<+>k(L&a?(UXGDGBKYX=&*N(n$BB7a=Tq(H&>{z3+F<-q-%l-hRK& zxh{Wr5sNvW`ONW*agTf4_v^0{BJSM&Yu>P-R7DoUYE8|jjJ|!|nD!Txm^exSwf|1VE4wpI1`92vD5u-U(cC46v@);a$CEf#UD_AbRKTgX$%wvtQb zasDEz?rAEQ=icpCo+0o9@UvoC1+D(@Ea4&4^K}olZoQa=)%tv|WYZSb%k2?>Fr~B>Mi@T{B+69D z(#e?WT7G3GHvaR+KeYh32lEB6ewi2|6&?BsE+YY}<1R(VCz9@JiXU9I zvc8%-O+CTu-3bFK>wHOkAy{+=tl0j}9F=POh_%DpH+#Vx>T-T8cXIwa1KGKV;pj6l zB&lUi+fI8C89&Xj#-hIN2ldI!E|Dc*QBxUC&*bJ#ga{b*)1XTub+Nk@Dy?c1@}J}$ z?!WB(n{#V`xP0SYdA)7w95?+6yx=So(;RJBW9SOo8?WH_dk@=35a z^XU`88kh}LQCFUH`R;^A%MA6s+eA1 z!>YOJGn_VU{WG4nhNb%^((MMbGb5i%?Ca|WsM!x13f$L0||tOiTl8O{+l} zK)$3Pn$WyouP2dJdv%(!i2QxAKvh5Vu)gVJ z9i6_iWa+&r8dT1K`1P=M$>HpPUR43N5U8>?`7qjdrRk}IsV9;I^AT6nQPg4Hc}lc+ z^#_=M_n~A+sRoz4*f9vdyhYsSnStM#Mm+O;*s`3LvEK>4@@os6g?wxT!fqnuFNZZA zF;o~~hjd{@MCiGFgInw)0xqT+hos!w*?VL9-wawf=Cl)Ld&@X_;fv^ep5dJeQ-kd9 z5Tp=N)GuXDAGy{9QhEQ#PYVFLM7SL{NqYA2ukl89bJf1kVzg?kP+99y=Z@e4X$Yys zr^9HI4)Et!hE7CC*f=OEg(S6u(| zUJb*neB*J*>mELHCkuv&W69d>D<#{0yJZPe48@p}TbC|&pXC9Q61%_po2|qvkjAIan z%n%`0;{o5@5%BYKjP2qyZPc*SZFm!c-C%b_E-@A9R|#f~b3Jdp)@4XtIHb<~Le2%8 z-TX?4-?wbPusN`P_v|RI$^ajPN(~;B3!Lo4LJ~?1GSjTzKad;U)CGvW=ep1t>*N!3 zrmjIn?`r(ekUS?5cf!ACXAEH5Cp+|uG!NQYR6wh36A6>NWOGQ5;X-(aKt{?*!G=vT z^>Ov{uqv?SWuUYVxk2Z}wl@VLt**VQyb>ce9lf)dbXu<(b3`~Ne=fn`;X7~n_LrQ{Ei&o-_7k@~i1KZ`)H@i@ zh(|XD8A=c-w8aL(I?uR8au?wnx@AHW$l{e| zpg`3cGU$WXY2RUs_0(tZV}eC7$9iEub$eL$jq?jt%YH}HJbwyHgWa)NtwwFHIr`2n z_wB@~mOc?yE>X8@QXb9WzO!vGsBEDYMg zDRDWUQ{7>AE?*fLGHq>Dc$-oXHy)IEd?VB zh;Wwt-EC%yOT)1CEqLX5zlSPBRog->JkNg-F>Afr%sA)yDv}&Sqyo<#vsc&JrcDPQ-11TzR;RQroIWJEHuWCsR?pU zhDtZk_VmMs^AnWF_QA+5s3Yfh2lS*1L1p_#(VXl7tDNx-T)s`rb@y{{ZuGei{g~q^ z(6gxX%gmdPr|1HcpnwI~^Zjs+9aGvtP4(=Z=IoO+AZ(nwRSCB#LvAP;Ru=^B8xkss zT8PL(uv)KP7YBd16LC+{12_z8LZhBL6(d3!?c?to^=0M630+-z#Vyp4`^VGn$Z*RJ z$)vpAd;G`(*~Bwx)_sQMniQkECIvlZ@ug9AecM+X6Qf`JAh7EN^s}D2RT#htGlDr> zhdNV!RcQ|x*7_A#QJYy8T!0~0RH$PU7Fq*m;sSx!jmfzO5rPG^Bl}u~vtC})p?lmJ z|I|mrzC7w~*tvtB)1r5c?t;>Wg3r&hcok)dypN>k+;(T|-UcE(jVJ2`fYb8p)9r6< z04021&h-nXkBH6$pu{UZoX)4Ud2_Z~1lqSiS(w2O&Fs9`t%O5N++q_?m4%K%nA8d^ zXY1?3&$h-*>Mp(W%fsnU@9!!#~v!gzWNJnA6>!>PAuzx2X;I+;e-g)}~g!Ybj;-mby>Pp>+|)#~~L;gHJq&Fm?6qO!wizZgK_ zfU4l`?gGk|`Runf^c59UYX6|jrQ{WBGu^oD+Waz*t@$tXeNVufkLdIed(Y@gkMvQ+ zc1t{}`jr8+MG|8iBO6(7A5UqIg~Dc8vtx2dQHpDOOdAdUK4sEn5&mS@LBBd)ms$9< z$+xajr-@-|a&8Y-ojWOqYT6UEfOeGxMu7O_*Sdv!a2${#Flc&u<~!vCWZQecyHM^? z?@)XMppA;2uD&Y{;|8&WG;_n)#Fy3PcBWq@U-2O0BNK&>Iax~0-Y@Egve(;(eJ*E* z@vi`dPxHq1yD(IltXo>zd4yJOoKj-)R!hrsVvi}IcFQpd7y{Qpb0%N?=vQzB)z(v- z9a69KFyeRZcIld-(NAsQEn3+fs78)PN~@q7@Y9#wO|>RbVBwJJ`zHzARg{vMCKTd} z!Hg`vB+5+O874#eo;E&2#NiZE_4`oz|N{ks&e$`kfo4n?a~lA8}7EL!)IMFHB?lg}K9j0`3? zicicQb6%1&A>uuG;yL0k&q!c{n@j)u!JDfA`yy(t$2F$w)Pc%dlchCo;_f$HyNpap!Zh%LG)Ggbz*# zI7e!6(wuz1DkN)06Z0fVhK6$^8fgQdsm94>?Cwaamck85h`Ogl`YK{(X9L)--gZs{ z(Abgg<~7j)kU&6QVuq?Oq0^aw^&MdPW}v7aaEC!10$;p}*R8XRBgLBeZaFToISjO# zNV0YTS>~jI{RI!~$Ar-K!`3Vwo1_pmzN^_zzN8M&bq|5z=jl*10?nStN6j&npokUCI06)f^F@!YU|7L1Gmo&|m46ZyOjuV#ZOA7sBK@*_ltAHEkbO z-H$zwj`70MmhWkdd}MF?Ihju>&o1vkUQP&n+?#YrUOk>zPOrhXG_-EUJUO>yzauo% zO7Rji2Jmh1XV%BJ}GAQMA8h-NRZW8|@;GNVnYyz?p1rO|!QQ_Y6Lq0e0dz z8`x!k_ysWwb}vciraCH~<#%vPdnb#qui%kav zqc$`jWEaM4jz4kpV5kswk!J;BCt?#5n=}ddTu`yy=!z49uQm&^0?+;NnZ0F%5~`Vj zV>bppj{YK8w<;x-`OP7HwtpdeA_6fl5W^Obxp+#oGzr6Jts3vY?{MQdh?TqGwyj=0 z#=KE$-xle!~QkYA*5ZZiw!+16d)|WInQ`c#x(onS@dTm(#w6)vboq`zkv5q z9Rl9Bb{72>mCsrLXswvK=T$^;^wF%AB@RT9J&oI}kXYEgCM1HKqrlLzya$MH9UWE% zFry+68G!gU_ugUEhC8zN86jHS_WBS(daS}2lR!|2dsE1nQZ7I@JHMIOC*q{0m=FYa zRHKsxy|!5;9;bM(wA$hhE<~kj8JEXgZXNM?vJvo+Q=>qVM53H#s+ohEwF?(xbAxf; z!!X^K{mf?1?%ER+>Lht&KUZVqRl}iDEi6Kgje?Ztc;n83*Sck78PV$&5}u3=}>?XuQl(-r!Wp{HuOFj^<6B1D9JDjy|e{_8EItCW7RaMRDiZL-XGa7JY z9GXCT;q5-;zsAx>yY_loaAI{gDcvFdqKu_2ZxVgyId*@v0Nmon-vc*h*u8#b=a;}m zNoX^=;!Y_1g$6Q$+P?R2L4Jt3*y~^3*JJ_Iqn`yF+8HI1)`^W%Xr^k-l}IYt6Qk2v z^r~gf_W9l(yzpGflf*!DG4Y2PO}w`@L|RzCaRp-qJ#TYjJ;%nqv>TF=>~OFEf{Xpv zFz=57vOdH%OD`7kFh%_~)t#~ImaE>ZUU{UgMl`eRG>)hFvGx4LD&ikjK}c~}uFKt- zZ_?LZLM4ky6CP)sQYG02f}Tv9dmj}{IDN|l`xtBO9eA0eAIi&xh?mslNAO-iP}g{G z59&(xRrd>OX#n&hI2DC!y`rjNAGq4mzcX}%!)DGZy>6N~z#U>Lr{o$q#@ z!0~>DtIX@GFN|~;h5l|-AhN$}IJrlNLC794zIjA_B-^&ZTl(PZE^7aU!QhnNB4bl~ zbG1EIBW&%)J>+VNY1}9_T^}s(mwn)&yupXj)$KZ%r>{XAVLo|I%!&elcXW&qjyO0%LHa_HMm2Is&sj|7iQrxln znTPJoaLFGs>+etzy{b{^xZSxtTIb~5nM6z8`PB9JC{OS7^vzR_+4>x9W}RjUH~1G< z(v5*6wao7gJiWSIcem#tq^VBnQ6}!o&MNL>d`V1T4Wz9VjWN;#mm@m%TNg3HtFW~oson5OZ^K+ahyaMR!NYa zDEoT;>N+INTbB&Hl{aZIrGD8fYp<+OC7bjTr`l4vDQ`Rp{B-TA$ zHv0?b>+M+)j-%&qNFbMh{mk1*i{S)|r&gR6%?h2P3z<^XPVAXFRbchEot%}=K;^-X z$CgWlYpY>5WdTt82DfOtcl`Be=+9oF|l%5;0uuyTQjPvU!g{BZr;}d-0`iYF9jVEYk69q?I zUBpxS&M0O?%i;Y`;Ko8G$m-~9G>PL3va9>ixl9=b9u$C*2va4fa%KG)nPan?m)Z-K zcPg0y9aqjFY2tDzrrA0Lhc!&+E|I`y#NjODwKma@(k^b3KB{0eRzuCNQifhKTEL%0yW zD(fbT%In@oeg5HC+hxK}Tu?0DRT6l#<@GFoq>j3ghH${PGKj^AA$1OL*4%0OGwtQZ zdt|oz3i4Re3qtfW5o9iGOQ3e2L&+oUOKAoypOS5&i(iGJZX}F6_ry1xJi|!j*dh!> z#`P5CA7r@k2MC6}NIVkiCdH4d#9qeN%y-@2-#C|!ii(!PCt=fjO=t!8AE}-uvH3CC zUM5LZ6hV*ZkI|d-WgE;0VLoG@RVTCqUD!(C4@_SM1{+T;Z660jiQ5PrsN%mN6z#?Sf}1remf-0PVM!j z3?47K9SvOLZa8X7#M6RE;TH(m3(@I!QSB0C&IFP1AlCbC9xS9qn1!&OVCXVDJ;8_s zRYX3CI6PpFSjO0y=1d>_Onx+@u;vFff%(J2t2=T0P&QbWFSg{gN~JjTv6`fIa0cBh zTv{LA7h7=Ks_1PFx67G*l80##a9Z*IQ|DcBCAM9;_|EWvT@*L1^MqvG+=TdN5T?UH zy(sce!Y-2KrD8q8maV)OjAUF`#-7R; zVH%r*fsn@yh(2a&;-oj z3wMxkOj`BrmGdj)XoF|)PE9*dIB|I#CPsthP_HZrdUR2`n(SyL#LYS* zGO^JL70L5Sj3ZND8cpVCze(ZNM>I|lAcl^Pv-yXVaX*Wl@v5a81luP``zR`2TNd^x zXyF)j7U(xy)+0A>Hn<4d^}ZvTTz;Fy+IE@u*)q89a5?BQ^mDg8!Y3`0Hw7-i^!|SP zfsa)C2!&REpT`iawauWS!jgJrcXdNRC`igutl*i+;ZzvXlWsc`OSyYpA5{rU$$Zl8 ziYzeFPwOJ~uu$x;?7igz9?Zd)80J-S8z>Vmov~!pv))rgynBRG?fv~VYUdl~un1J1 zI%HIX`oLJfW$kdMmsD~E3LZ?H2uZSQA8JUJ)geB&9tJh2IMxd(;*@6?Xnn{57q5fz(4QzGij3Ybj zwBBrDt0V;K9ZT(CO>o|oIW2-ID&k{w`n!)CA)0Zt4jG-_Ja0dW3#_(&rO6PS{nNI) z7-b@rWeyems4v*!+v~ID;~~LNht{H99`ASHyYmvPcZ*2vz+5eI8=W%pv2crr+_{r! zF9YW`yB)xUI%@9qEw~}JdLKvZvs$eWHkbTp5OnHUuTu0_*2I(Gx73Fqb*fp4@!n>$ zpexv~(qt>jq0Z7iOSu(N#gY}Uurj`+yFh7KnH?nC-=+LyuDsd*MAlTBhEpd`iR#!?F70v;}h8r->%? zx8gv5zT8u+&OZzn+N=8q>=?J{+WI>cKFSQie6w}>f%6IV3&H0^!H~^cugL}e1=8FGAJU_C|Zu2&U|`hL0v`l zbDYVTWSn^ZCso!%LA$VH z20*Mgt4qHwl=R1L<}&T4T!yuMBzxub#xUUS%N7Un0%q8F)epWq049A2dxYgulJ87BtQ|c~Y*NEi&K4U?VXHb=Hw_Q@D;o*>K`@cmq`GWy-4yWG|&+ z7q242iEU-jciEO34oVV|@+gS~yz?Yd-n8T*b!x$4zV)DyMpKHTL1#g|aB>mn))0Cg zjKNE>Zhq_av%Xt+J4JGJRU9u1D)b}gk8^)>K0+^GIt5$85=1C53vf(gOm$>vd`)P6 z@%S7}{tOXD9^S9(R|TdmOuZJ+o;OkK`L5)0Su7(oS?q6Mi|s(ZeFeF2BYP32odK9) z+UfWkoDm8GSN0HpAf=Q0g2e4jImIKfLDa=dNJD7g8yp@T*V{w+yF);xB(rVe=RQ|j z)tib=WIt^@|DLFYznxqXqIlVRZuuc7nT*(1#fmeG1Ikg)(JEOjJOq5hnL7`QNKhus z@`ppj8_P!_;Tm<(;z!HDMUt1j0=ZE#I+0X#E1}jTf>x%3Nz%RcwpUVPxdpT)DE)M~+G&vT7DItz%N^K|bKsJnhB24=pjzk-DMg+D1a zCpHEsE}wI!o$%{-^2MOV(XCiLCE=`U@RS~m4S2{^?=K!tn3{prM~d0ctP1N%`uuPo z!y$j;`lJ?9z2#jR0Mj1?-sbE8Zj_#5bJ3j<&&R);=%F;&JHRsxTkI1;O8bO^JXjIQ zB5;sVFxcv&JVdvcd}%0z))P(%M-GkadeB{C`irIOH0SjOZQdyOswicOXmzvGTrRBd z$ha7;6Xsd%g(malnb?l4TQ5Xa`Q-*(bq^A{q4RaQ5sNxCdzz-o*c5ctrt14Zs8A8| zs2Yg|+CF+SJn}4=3&TX*d&y%3F5wU1YBDlrvS1xX;{oIFa6ZapOrmTy9ql1%DVSb)~y;PjlVpWWF z;poqOANF}p+O+1+bM*g*Ip2D6_G$+rsw}{rvLXIv^)cpnx(q^5!lRvXKxu6v=z)WX z%Wo)!qr$D+GL~`aD!GglvJuNGFxiBibGVN{nV^X`ZQZQ&G&fNz^Bv!sLR!+QgxhiK_-d7yiunh-izW zmcGFRB;8n^3%N!EnBeDw9GM4o*eO)RV&$^^HeC2#uPKnEN@7Lt-w0(>R&>2$#h0v< z^;GjM>Iz$HbZFb|O?gF*&FqA==y>P0CXYLKAIA%zWix)g3FON;*}J{!8N2!(OBDx= z!je+sOj&o`Ef|7^l-z*g<3l-ijBqfyH}(9@VS97Hx-ew&JK4cLKwT+=edhD2IQHG8a!=H-JT9KE-A}G6-wn>xL(C3(;h6TO6SIK|2z#N6`dnq-+?SgmJyX@`UXZ zFsaYu^82LfT9huT+K+m-S}BIk#ovr}Uj!!q2(h48ZNeePfr>X;1XWF`vB{sl8&4&^ zSQyVYlv~0&>p$IQQ{Z0sQLkJlbO`!gW^<+C@=>|MPk2ruzgDY< z$mIA2HZ2m)6Z^Jpx{`%Plx`*RzfQr-zSY}k7jZw7zHqooYk*_G*xy1;EA5nr3)!M& zPoR6M54UWLOg{sh&==`Z5XB!EjR)N9YmwRCe^W_0VbyQm89Be`$-HZVWfavIrfcbY z`=2h~IKi1Q9OQtLClUaPDL)VWb13<*k4K>VfSLHa&b3e?RLci?>85wZt53{^97em} zi6byd{ZP16%3b1Ec&j8vO}}ON!k>&vw{EAH*Un2xFcDm5B3UZDUc3EHeP4Pl@NWl9 zu?$3i{-f`LmrM8j4V&gK523hX;hmeq>7mlZESh^%VCHJfGo1>5F@3Lbw}(5F!eg48 zL$a}_0BZsK~$ntKED_X4>J0b{F2bSh@}uxa+MJP-$}3AASjAM5VOtR$5B=JkzhVDa z5nVlM_Z}DS!_yJhX)qpmJrGud5$Zs|aK!A43O)LTGsO!r%w#PtPr>^8WB>BgC}2Xs zZhNKa_15p~jL&&-x>WNmEjbhQx(`kH6%vJS0s9O`r(BoyUyXG7lUrGYqVz|h@JU6X zvjc~iE27e11Yp5 zIkRYYm@LcotFF-UFHwpokX^zip?%z(!_4zaCl!l`>Y6^~j0#HBuG;Rw8q=Qzg^I+P zrTN+!im)9wvZjo`FLs0uLDv6Xj_hh9zjiy(tKuakqJ0x{elx9~%mi`@uqGcqkc(aw zU0*>G9&g|W@NQ4)7qhk<-d=L8B>g?h^p|PP=V!>IqW%)!bXvr}X@58HyPl&uTk?RG zohIuSa$1ZOAB@DSYLu(UY>b?V5%a5ONm$bz?kq#5rnX%+@MJJ3aw^)cOQ(DVh0Z6% zR=^jcDv|hX>L9~qpf$cj;&vRHuyc-1RA4)pN>gSHXm?;%d#+mX$0tI2=3FZnl@oma z_Ua1cO8-F-kgZ~P@2%~3a-(_%6MX=WIGazdjej*qfMxsSHlN3*Ph)J`s}p2+s9KAB z;X0pX^hNTtqV#|8za4$(J|W3kPJZVrA`O33NniVyC&gbME(b!qpBYbG*-)ebY|sMh zq*iR17DftP6csKQMnnHzP-EJ|^^XqIqx9`s?RQrRf=3XzF7iXLNnjsp-1aCk0%6O| ztqYq9!5x%VF!5@G=`K|3cT;>biod?j7aOl_@_2fq>3??b`O6{ZhmP2X&I=vKIImCt z>0$ggK0=?9Pyiy&pjI+W=1*OREFJ^zx-hEtNPZh$@Aa19VOt!%apqO5AMW&=V<@Z)(ZF{r3Mag`jxU0U?u$> zjQ@vU0Dk9({EFy{wphQD=o8O><4>(HQ#ww5hdgsn!fSgX`7`Q&{zLz~=seobn;~#Y zWd_}dH-GN?=6H>g(!V0U(`^yb=kb}|^_mUs;*aCxOKcV(fRD!;3;xfK_J1Fv@xTM) zI<7g)`FCy&2>QBe|ITg606WP~#sHrj?@!N(&ti4JhLu(LwEgE7HX;XXrP2Puf9Ji3 zNCFnXQl8__%^-ge$peQmk(Z9A%AdRvfBnojumEH%jD^25ia-0u4_E+RiqNp3Vf`DP zle)kHWM71&tG5qQM zQwCZB3lNipH~HrmmJtW+86;1gnCx5iJO5kdGr%?xF~!DJ=ML%xD^*|>G1T{6y&AXs z@%B#_=(7y63ZRB0A#r;DA5vqvBapxEV*H=oKIMzw&wfn$=AaMyfUE2PTkSfz*?Gn4 z#QG=si08>GOXIhAf zF28JJe9ug@sZpUu{FlLh_A!*vK-mG*l!gDIru=Ufl;ijP)38?plqYb1hR)!_Mf*dE z3*eJABv?!sOUIHS8C5j&%PLXKx;bIQV5>_BATMV=oTg(om_}<|dS0K(1^w#0*^hr#Z;uj7DqMKJ=cTCm<&~D2 zzVlH}t|?plap}%HIQw<>JtvRZ8p9W=u5Y<+<0HNgcUaapF`4nL+T%exJ7txh+$TdO z{qBYvc8>7(r;`o+zdk$N&3*laImt2W$n-OEFkAbzE}*$gU$;Oy2iMf8F-5(LF@FX* ziezg;rTzA{I;wdk2u-408TN*nnVJmtGl71(`AqMnPQ|;667&XXkM;v^uh=%(HceQT zPfV#=Q)^ZhiIrCFnYrF6Mu}QekRZHus{zBzc-VHSUh?ktgzdO|!~3vpEG_3?<2QdC zaB7e>8%>g4^7GhKxi2!Ii#Yz}N@`Je^W*OPz5{D|0p41MNq#H0X*V5OCuPsw7sqmVHWx(JB8jlv3IQQFC2}plx-I*;kVN zr+Q8S@;fhQ(c&c5G&0OS7$izZxg@u%@AW4@d#c1(8ttkerq%i`jHT4Ag*Dz0f|fp| z$d_sDYj@tAV7=f1GS%Nr8h+b%{*BsVORoCitYNB)hopbPB6&!?J_RO6d^tMjLQpI& zDI?$3@AZlp;eQP^47aC`Nf4PqJr9q7GY+YG2 zH2xsKExh@7;MM=xO@I?vrSrxs)}J!5gvF&=Wq4M&I-)>Nl*jj%XDX5&dra^B6vlGX zNwSLQ-Dz!vbXgKxggFvn8J6_0 zV^K0L@kIKfb^iRhzM;93=?GQaDWGSwk=6<@R;`En#dm53CusHHTAOqVdo@3H$(`iA zHQ5B6)(QsFxE!ROVJ9jGy_Le)gVL17=rAPoDQB?eHRx?78)g|yd)0+mPnLR3AzT6z z;(5JMndS8&4f=>a7HxpnMO*2+l5*nNN{cBk!IP>0fUYEJeV`sWmEi?UJ(w8Imr0!P zL4EUyMGCS3acs`6F&T^PH-o&H)4=j=>=nHsEgAU2vZ1h6JBnF${Kcjepc-x|$$7kL z1>M6R?ngb+wKZ#^B0@D!t+Js<@o72Pe#lV!FidW2D_{^CUay;UH1C;*@{RxS;HK30 zi|htpR%C-A^eb!gR_B=a$bU`bfv+Gc1E-?c8z35yAp4w>7vr<(ur5=9JZ#Z%Gw!hS zIgq0zAn#nh&x2he&)J1X*}S^l{HcdZc}U(S+xva(u?_-?Y?)zp&diH)SgI%sg0Z8U z-JZ=aToaR#Z2;Dv>g^Pwjfotn+y;+xYNq*DIFFqAA-@+!G);iJH z-U7y(uQg>F?Fa@rlW^qgsKE)wMuJuYZ+OfX6HP}&DpvcNYv%z|QnG;x<4{iP%Q1f= zoW!Bm1|*W5KB!4w%BEG}Z5}Eb71~PC6rd>(2jpMx6bWFbP!fI-TmTreO}@XzWh819 zit0;y))lGaLoYRe7K?oNVF1|2J*H8y05MLAkbBFZfaH7)V<|0Si)4E?Uk57|2E2!g z`fk%-#$PJOaerj53WWo%F;aJc(@d#GWMQd#xVpiLKazg?IIeXOW+s~QOUDnYz-U*w zL08`kK*vMU19||(@=Qb;I4<443t?5vInDBJKnB&>`0{nFn;cL?#u9Jl6OfS1Us;tCk zU3g15pfBV9vWij50*a3hgxUL@eC+tIROQJ4u42?&tJP91&d-=^e#Pi?cEyQr`roxT z`m=b}c;lYGWa^X}5XBkiQ(G=E^P@($U2XAgQnT*r{kA6FxZ5*~)0ZIPeTt@0;Q92( zGz@8?`7He@Q7HVE$H;)j?%VJYuxU@CeE!=ePTB(TYk1iAfIQ(p~gmp^K_qV&nL8~boh>#=0YP<#!3{}aDJ|nHLqK- z9A9knNge+tlk)fvv5#VRHbV944hqf>7TC54gm^NIMkTen{0e<4A6x&Ncv0g#br`s_ zx&wHDis5}bm-Azk#Fn_<_L`)QLD_GkfB4Ew6Mu> z@H`CSvH8ZMWLPk#=`ePgah)G}Dm_dDV2@*x4Rt-Qkq`(s@s=8hF zaW2l2Mh)A4zHF)IXzu&-pIQL*b`Y3{)KYmLP57e^y>fhA+P%HC7`D*_WE1ndRTNyC zXIg*E@z67htQ=WtBE8d)E#m+yt^$YP8=y>p|!o+;wQBHAvNZ=M=hl(Rcwq*zoMT z>}`tpoLL#-IhD+)|4uPos27e$JN3N@KY>2#?N;IncU0N^K?~A`fw0qYh$rid{^Y*t z^eNljX$E^}<%rLmtxX{z&28orcuDrDo78-UKuuVyTKP!^qO5wyB1`?{*G{?%BB)5)B2i&!pg%h~vPigw_73ZoJy^x+M=YQ1 zOrd{0d*z}?9Rgl11BV+C*i?1`lSABPi4fuEuMgovzcc(YpNq&U?-)T@vx!lqnHH|l z&|7zVlZP1CeW=`ZI%cw8I%YZs+={RQO(?>s-lI%DeYIXs`c1F5qmKqG@It${ zp7*0>pj)*rAa^5?C7RNZ62cS?aT`4#>vFfCmdM%DDFS7hjl&z{%V2~RgtM-1m~W}T8P;Z~?{u|cv~YFzNNWnIeaxV-k4;cQP9;|F- zI*U0gOD-O2HrTRgRF1@}EOsp48awW7BbWl*xhdwm_vcqqF2Lu_g5GMMcQ z_a4m$QBt4gD48#LfeC7MWDiTX5_<)7w-Dny8{&J%^O^(E!#KEG+~kSbW-&O3Njcj1hFG`- ziEVooQ%ZU5=kbYy$9_FoJ^8p9O&e9JV_xV_9R(Gv6-JD{e-f@>QlS&Z*i9R9eX9^2 z>(^-hMCWWT-m97Fx=p2K*)P2KX=GMh@2Fd|c(u)i|8p6_XwzQGJaO$1yTv*Ux1D*d zZk}3l^^g~toK9~!#ix{!ze9w1EIuJa$-eio@y~~GXvktxU_XcdX0zZtqr<8!Veu7h zF#BRX))*QgL92kL&Q;VDt|o1OosA~?)?i`=C=|HWuqLwP0PO@4jQT@0KldwZE3w4x zhkMC&T8Z+7&M)BQA*2q3=qBXa7UK=U?GNvdd+~x4+eP#>s`EnkdN>^nC&y`?mF`S; zg7@R?RDDbej5UYfy5?0CQ%N^gvc)T9;jF1v1dYln?h%&L0^s5PmDd} zw7#kd@Kgn0L1GZ?qt|9x(_C(D`&Fm&AwNF%N9@11bfplgN{$$h3xN|DEPm~c%BIGc zlGn?=j+I&YowRp`9*5)4lj%vIeIxQ5=?W&tSM-YS7q%S0jaX)|0Z4$~l0<5$_lR?H z!5g@-jGk+A#vyL~S+&@rltDkjFn99FziBLSeS?0IBc)W~wVrrQyyDcJlMuO&vgZ~% zdy~QCkQ=foBTK!&{!2WZUau4=5`1K$kk{^eZpw4dT`RtUP{r7TSA|225TgKzR?A!S zSRR#q(Am_*!K<}#+lidTzy@w=&2H;zWN1IhI%|9VgW@)K_1fH3?vgm^uwijs1Rr1x zuIlIGf%5Gny*ivuod1Y^K?EnM)il+b8bRV#5|MzWErizj)>An$bK@1QTb|F~|63wZg7q3`gx;=AKv)wLY$MwxTct zvzNS~NbZ0QeLPCZ$SuvCOyiBN8%4s+(@c{w`DeU9CCmd<#Hkc~=Ebc@zrqJarvd>J zffArZ%i1H0s56tVmszX&GK%{ZUj0$K=%Fe^8G?&Ne#<^*Z3{A0alkoG z_;%~j@f(AguiIt9iO;b5B>Gk=dcuu+(u*U>Bc6*5=5R9?YvxG>S_|h=3-Ys5fd$l< zU~T7EL3~sX&UTkHsA?jB1VaCM@3nsKXi^NiJ8>e-R`Dl}-X)>QVXkmz`mei7E*6Sv zL}$Xs+9&lp!B`?SSn7fyE6XRPMROSdA(Ep0+f^VkJ}#Xi7RJx_N%_9Tn&0z?bFwWO zYC8RqD1;765BeK*Fh~Sgbwqy8;5@l;OZG_qnXdjX@LDmD?7p|6^rb<0I8D?{6-Ipz z?@`3&_yPU0D*&X>!e(PSJ)VD-wdT{bWa!bZYB8AIn!6Nsz0~G0nNjD&gaLYvFTvud zSTRh~yA0f?U1Nu#>|4JPBjjv4vewy^Ve|B*M|HcQiaT`s(sn}*WD_}5WYO+~Brl$+ zGV0Yc16B|;oHJ~_PvJ$GV4F}l> zA;^BFrM4JaXYpm}^CoX$pCwVr=W~T|Uiq+){E&>T*8YvJAYdTn0La46Eg8-^1 z@>%p}XfO5zL6+QmGjwV^iHl|(Y-UHl1-C6$^6A;WCRkRPV?YMKlSJ#~N-9cl1R}>z z-kY>u8sCMXFpPtI>uFa4-hk_Ei!Y!6l#j$8xaDm2E{||a)u!jEEmO^adN=9cZ^SRQo z+D{wU`Z(an#y!AKFv+tuZ%Le;E?PaedW=Tdor--#;eQzykz`rduH@kx_+r91{wOQ@ zZ5E{;^+k@=Soj4^w?jy!j7~hqI@!(*+*PMIUvA|-=qg&073JN12ijfWqP!};N}EWn zr+vyr=^$N2=WUtrKb8 zh;@eGRSuiTb7Yo-sCy%dh6P_hM8=G4fD|DeC*)1mZ?Bg;GXF~5HZkhB6t}oSI%QX4 z@hdy@(YJG-V*5uA^cTuD)5Pn-PCET=10FmKQi@5eZt%YemMZGS3!b*tTlmz(7#l3C z-`)371PoXUJ;>z5zNJgO&3nxE&~is2p-%Wf;$BFA;F#C4#Azl2N(w4 zx6gU!+9-~R_9&QvoDME zI*xbUqJ5IlD4iS*+en}u zV?$p4D(w1G=8-;t8**Oe?UO@!-vJbQq4AgwA$;Efzz9F?TaudL>qkC|$1$m7xg_Sdw=^Ezw$%8}<_LnIE~`-N2-iCVM%*cEw|I@Q65#V}D3c?eKLH zQnsyLWz#j@18l1FM&p)_YRdyZUFdMkd5dq!RX010S}_o-zSyMcAxX=38*W(NxC)Gt z+cD}Hr{!l!#_!D}=Lj4p;U5bie)izy?XXRr4)k=s1%ErVyG8dj{LUXojuZk9C?dkL z)>xrb3dtr3^RQpXMnT0c2hH+su8HhcxOsqE2q%D_7XMK{jTt2)htfaW`TY68JuLD7 zVl22ZgMJ_M?07zSDR3&8*<00Tb+C1Q^Z-yl+AtK#(Wqs}sz)#e?;SO=+`lZMJriG$`V$P3xM{n@ zRbF~UuLw=Cm@&`h^h>)6i4N-a=MiJ=)Z!K$Kh4o`EP`cUx&qRAOcSVyq9#ddY00lU z3jbNybgmfxJo733bjc5(n>!Yt_$uxGn*2OpFe=xxHACBW_;zLNMDBAr;W$bwJ!L^! zF?+EN9-%_h!IXkJ+j$z=sR_9psfPH$TO{$nA(+C$aQ!KZR2ZH{5N4Pi2%h~kWHX2j zT%Tki3UphYpXLGZ6t;8VuE_$=Ao5H>=SIh z`4l>;rC0REW@q2`s2TSZFM)*ik^f10^F?8X*JE_Df|uqIqGPVb-X>2@WX~vho?J5{ zaBa!Rf5}<5km7MQC^oZ?q2c&i&2PpVZUhM6v{^(hJ?z3Huy{KqHo>9NB5O|NJ)h$I zp8Pv$G|DgM$SQIb)`b?wR-ifur$Pa33wH+hBgN7pZj2%Bh1_j|z?XxaC^5)AXQk%I!AS)EbPICA>73G9?mif4}u#Djt`N)7F%g5N!Og~r_;MXq)rL%k@Zg> zEUap4lo~S}L+hp5T~9dYFXQEaqUTYxNMAAY0X*b`Us}^37v${g%Kr zsaC9&UI!ShD*Dn!VKa57P@vHfSMVM4z=tD#n=46zo!;Q(;K#n4bBPsNEImQP?m!Z{ zq{T$rPrN*j#ETR1Rj=bFBH5nZUP@EuB+5lWS?v%nk`94Sq=ByY3YLUHML-Q|Xw(-o z%=7Z-946T~awm1MIFz&kn;xg_L(A($luC-v&i_K4V(mft6+Zt1bxJ9S}$sZq3e=8D?g*oR=-}#;Cr8F*uG)GBdXW@SJoLl*R(CKT|sUmG}p?Z zJ!Avl9`|B8Cz}jR%c82JHIO|-P=vviUoYa7W?o?PW!XPNaPx&f5C4mR ztJsGHe+w@E%B+GnY^-i<4`P0wD7wTN!|hqZhLLC5)lcz9Ve;Hs|z`q zk)&hrG1+{@#1q4#;Nm}{@d$LV|5GC6A7i-zO{(l#uIg;hN+an%j#%Qzh4tAC4nJIp z8l|=)d-nd6n4ONY2e-mSsGe7QTRLnqQHC$LBW*sxJevvj8$2|kzhevFdh@grRBL9Hjk=_L0cjyT|{))z8G`F{2tV+H(w8_g?NG&5P ziWM&hYRy3Y+(OYSS(`jO8JdNv@?OQlhSY(dA3bXhfApiBQeu`(@cuQtbAFBfe#S}-%(gQ^#3pFdI4Qh#Qpt?G-QeqgE4xk%#B zS1>BKihn+E4B}R6YnrfHgf1eXuvV|x8__NM=H8F$sJL89x6Cxj(K zjtKS`QSLAlxbx5a-B;#^j&a>`giMMT`HQ-j*gcB)VB6qUOpdBiAKqnl>C~?_u5|50 z7FW(>8P(ImKc!|7Ufzu#M)AiBV2cs!UI+EM{ZYwRvVli}IbMoWHc=DUQLHL{=DvE} z*=lU!B1nrP-{vn@=z56Tr!SD>9S$bTs^QlrA!% zWv#_3qf@zr+<5uhq*+I1bpTr~)&2$$2jmEMTCi#Z^m!nFFgAs^^-{ zRH)r$d+!T0GqroM5vHMl?20RMbXT1Ak*3QEEuv#honi{uz@ULX_c>MMbW*kC5x_RynzmV$m3NdQFd2L$6TS_BUcH1CUDjc+dDiRHw=DF%4?<|m{8(Mh zXB)~CK`aPxKwE6|pO3i>KExqC`Dr9MJZ_sE*kJ-LiThsvczrLXU^DK8crIUK%9Xc= z~o{M416NASx0%*&R&-XUTnUGLVqZp3Fntki^;3=D# zb#ms*+y3@;Gc~{jG92O-*b*3477Q#%WCdfRs@ddl z?5YQ5&j;G3A82DjJWHY1PdJ>t7#;UL)%yKE;o?3J7ehn4PeYCJuU}kIZuDaB(j~oN zEV+Den$>yHAq;jY&%1lnZhX-pO8o|M{x~Y`3ZG)=;+emEEPGR{W1G$VjDcL~o>X69 zu9pH9_c2OHw;}kYF0p$yvcR1Zf}=K?G_+yh&{Ys zgGHL1T*Kb?;yqm9X5%;p08E$o6dFFRFZcAy+7Mf#uW>a14 zm|4UcB);3$8GEVb(xTkQJJ*W4RBD>C2Vqmo@)==frdcGt#Ac)Fi<-!MDQlsF+UgGl0IrgcLd)wxS zkv^oZG9Ndxb|Q8A70X%_?jQZhf}Yd(osqn&VWgr@Z}hp`)|dJBvWRfHAKA`V2(eq2 z$uCl)i-(Y;3SwGe^zjBiiB&Q;3)(yJwe;)l9MNa=x6wx34GAXloAJ*=Tl56BHlKar zbTo>kBg4##37VaFre=ensk&R_GkB5T?-j8}QSqA>4V&TugRcRB1`#jPor z)puhL<1{rkW4*YJp`XdyX9f~G-IOzAaPY*8E{@{_Pivia%Uv!Wche1i{nj-dHF+*@XLHr8pTjeijWMBxTwfYuL|9^WO)HGdfraz5$0h= zr4-XcGVUJ-;!+Pu!C~g_sEU|g+NSMQIj0H0oDL5dj#Y_|$!%&sxa!XR zX1(ApwS;WAp3$&wOO%XYnjCStCb}k}y@j+?bqT0lop`4$P%JNcdX(39dXZEfR`@aO zgqsd$Y?0HwueqyVA?r>Bkexm0V9R?K_sg|nxKTh3C=wf=<4A!=qRX%`^w1tFgWtR6 zHNNn$XY)FR*r6WdzwgPPE)BILHPuxsNA%J{VleDzN8^KjCmiQj+48s9{_zCdm$6I{ z4yqg890K9dA52d2qZXoW`4f51293JdUwyWri_TH=Wsll?6-BH1$L3N9M6L zL-cD+BZ37PSE~aYWu6XHVEuxs=(JIzP@A^ic3PZN(_4`g9?K8C^WjOF=*1o;cxk+1EMH_|v5+!x%SgJDtxQWp(hGOzq^lr84f?_taPy{>c6W1as zGbinyK2j1oCue$|$x+qx6Pyp+8w%jw&o_(k(#L$J|T@)7Hicj)uiMH9OzVS&Y@ZW3PYZEd22_C>EcwuZ4hq zY?a@c0WXl;phSrv(VaH@GJ>q59lX{Oyo&46I>hZ$ps4sMAnAO2X)-U|9`y0nA6q4> zaGAM`8~*(7FF!*M&#!49%nxP2e2P$ip!*89dfv=9Qkl#pA`rA|nvD{r7{A#E1?K8_G~9 z8D#ZZQm4F3tN3ZG+}Z z{y=UW4wZHXOVC2Wk+mOLXrVseaXw$SoRT%7?V8^zo-pG(e^Sg8{Gq1>yP?9ynhI_T z@TU>o*R^Z2T0(Rvoieec_Fep}qK5WWF68T*Y1+gU;*4}%&pDInvx`i{2HRzeQ-4~u zhTtw+&|C0ZP)Vzp2E1Mn4w?t9%LTl4{dx2kg(y1@Z)e+>yK}xJt|4}-y<#!;WLaX? zkjucIpK*Oj$5#Zxx?zf$#|@;8b|HRm3u|7-(TzsT@XC)-Qp#1U3<_Ydd1!>Y zZ&vsqX^)eMbs{-5&3(usvi3ANIQdCjS}LZua}XC8FUnfm?LZs-pQN zfgOrBwLvB~f;)EGps`@P^|k^<(k+}2{S)uwjDX=6hVJP#!CEi;)5=xL@;x3K8bbkNK~13DW7_v^LeIV@SrV$x4O5zO*3ptkMYQ zVNa;s>n-N<_!e9*egsh^I#wN_(&H^#tQl)8FUmc)x~UBqc2)T- ziq0fDocBE7@>(W>(u`45FXr*DgX#LMr>X6n{JTecktPE6M`#`J4=J%?TJQJ%(+mUD z%h8%7^&4*)l^I>IyG(^zC**0g<1Nv#{xkl}fv01ueDm@r6{_t7X$mv8i!CpBxv&JZ z)+2x@)>+P0)`G&=6*DNopOJ`#08WKBhDKO_KMekQrL&dJ*4wvZv0!x#@E)yB9& zF0mQJI+#m;P1+{SS<_%q={9JVh}shpxJ zXkjPA_az1iz{G+(PQk<+#1wD7irVH3T1}Ql*Oz}G5@UE66xd+Cb26cC?vu)JtSHIi z#jPZ1I7f6M8>{>*6u)5QQhm^iU7LrAm&uvNSs{gMF|5h1h2;x)3XeD47eXYdLCPgC z+*0GRC|Ik%glORMp0Cw!(#;DQD(ZyL6cZ_?)NrH6-?Cb}|bH z8*gouRgfp#z|A2i78?$PVzjN>0E5{fu66IwW9V~0XnKdPD@Cq^`2DSK9oXD^Q=(fg z2Xd>R%r~gTbV78jZ(^8M)_WSom+(|$qF!Kl90$}69|xSD1YvEs3P$Ba_ z#JIRS_>LCJGKb+p!Li!R6J-csNOtw3OwVpjL1JoD+UKZ-{7phNDAM;|p}Y$nNNtuo zi&Yltck&wA3lL^K@RupD>SF{q0lKmygSdm+llS(?-B;Zl>7Lah)N|!ZP z!%nvH6rzudA&Qu7Ytj$E=X5xitY5TWD;w6P4?8i&&+PLIev`eQBalmDMEwpj$83SM zbKI@b-MDK}kYwcT=2U)DR)5Z|ELz>J_?BRWOlQp4Zm!jBAxeV5@{mElCIpE|kS>QHBUtBC8M+h5lA)MDpkgOu(pctzVL)j`Ob3HAH8 zf?BjY3iF*Ws*z0ulDG5#v$L|SrrsW)S!) z-5OtWr0rmErj=4tavBFY-1|OuO|SW3I`d_ulN24aXl%eY2ru(}3`z3!5HCZ>2%h4V zy{GuA?*-i-3{s-GK9TozJu7BOG;ph>Gna$h!;3sW1W|ACOS}7?vRr?;|OGB81oNVt-;{2IP~uO?I6e-943%6@`!{^aQ6e+X5tkAbUW> z#}z(QCD(PB5|^xwaZ%-Lq@sh7nVvQ|Rh_LzUp}R4SM=?=Gn|Macs9dM;iQL=Z2)n& z=HJv^Okr5#YZQ59a`^OiR+6Cw{AmBp3``2PS0{oTQVUINTIJ8JZOBd@kOSHw@UPlW zUu+HQ8egAtxFyJ51nux*kTO+&xjgjJg@DJy_1kq>YAFTnV_fJJPNV%}N1DSOQIK*e zG*=N}BG@vpmgA2~2Q=h@{rqpl@4Rn^NpgDkKNh;`2crU+6t-tC{ zb#Kgkf4H!Pz@l68bW1RtFeC{lNib}+s?RW;+&xdmTb9hC%k$nOEl8C*HIr6OL?HX^ z44jp?tb_5F4?s)5R2js`*|g7GS)WS}|CD#x&=^U%EH@`x&%j9jS)k`e2X!L`x&?v! zo{IXbikv0|qEPhsBX`FNwo+acWb9`%r0U=kwfh&g0$(S6dViHQ1h)Yh;Tzb})N4vN zu2=fb6cOH>kc&VY2t;`+l{>r^Y&R@{vh!-%8C-C-X6x!bCBYO3wo>O)9?VDxU5H3{ z;uCabwN{>f>gYWoFs1W=N$ys#6ns2BpfIw;fL9x-l=@+JGF;gUpK@)w1K#!%$L{(j zP_g_aY`*eQ(3dVuGsJG%XTYyT+_3D2y0}B0;0W>~d@B2v_6J~^cARm9funq0S-!G| zz16#Qj1#m3+)=}krJ|#E@g`w_DzdYpHl`udoZahae1rh1M)cLWv4hWO^gt21S$OGh z;w*WT-LwLoRaMh$(vzr5$Aj(Cba=~IXykI7j}RS^BD~IfeBfqkW||kA*I6mVJT7}# zo{jUkR|It>xXa(lq;rnw-X$G@qMaT%U^9;zQ6Eg}PChpNNwe(m$-`o=Nww?68_(0j zHP(yzHZP-L5ZoGmm?-$V9Vd={bI7Ove1k(Pk8$FC3t$Tcbn->`7Gyd@5syoJ3@NzX zR*}Qcho%n_BSkJ6NQG)7-+fTVUoO*qZ!3*Wwa!s8$mP_K`pU?$$l(<_QzY9kTgBbs zqacUmFA$p}FKZ_Stey{e%W~Tm&42Meh(o=v>`g2P*T?{?PjS*hxy3a)A==a-3e@7h zH{BIla+8L1J%Y2l%ntaJa&cIrT#}qpIWk@X>xv=R-PzG=76f7zh}~B;#yhA~0*zv1UYKz6r zQQS$rKG`I#4Cf`1Z8Xyg^M2p54Z^J?87X{`7$JEvwqwx~fUwv5;k)tuAr8;T%!PBzyFj^WfZY(D z5q_7Z7@L>E@q0a&ssrDh|EXFPq4RidzzI+t|p>VTe}4&pEOqf_~=K$*Yf$b zE2n1=b;07d*TUS+Cl6Q3RriGWWOs7YM)NRwZjO*7&8tTkU~K~QusMlKEEQ|9#{2RJ zoUx8^O74yWI1M1(szt9CyLjykvE9i9sUfXyG759Os(Y@rMJ3fSw9zs2|?!x2ga42MftD#nq z2r?6s~KpTN{;dtQVPhM@#dh1LRu~YUpu3Qf<;cJC7KAYiLI62q@S|4e%qPl zTDH6uL?-pu57^ZuCd4SWKX-aVF6!cMu~gbw)HS`6$e;AKXWHPdY!Hlwc!jr2$;lS* zKb^AFEGTbFFt@a-itDTP@Q%~*+5!FW;A(d3&cssxXRqSZVOQg9-$L4ygkVeBk~N@J zAL?LS5%)qaH9_DC{fmuc`oq~h_G;rD42Y{nCmd2zGXXO?Ba0Fhq;nc9e^T zX1I5sHF=5IsyRKV!_9ud($|HMc~Th@H3XhkWtMxk$>e2=tWwRjCn8yAXEtOAzud$< zdSur7iWZveBz2gAKK&DTfHA67n7h?lL3(vZygZxIFGZ<^7R1nSpCcM=a&QGwS&4$k z9c9ZEh_qU+clYw8_OK5HvkxXRgD(_P#fYxcsz-Q$!i)jMVxnsWFa4{8j9{i9JFh`QaE3dHcm{?C~AmI~V48 z1ur@Od!w1oyL`^ZELStc+W34Cr71^0)xmc&A)CfzDk7FMU-c;0`mNjaAQLf%bgcED zV7oct1FBFuyxE9q#n}9^G}=dAYl}#PV2I;|R5kt1A*U71#~-k}GRV-@Wf zN}zU1+n?Cjr-q+x>i5WMevW-p8E7m3|6(kNyqxG;!+-a&S4GtRr#-R00uLJsk-wXH zVx75}%f4$zbXU~e=owD0x4~3)D~PJc?IJeIYthNfNy+nY5}Elh6Zl68j?EFY&-&bk{QIxiR(0$Y@=)cZVt2 z#=^7|4S1=6cW;^PB33j5*0)`a*@KccRA;U8u2>W^Cone7tFG;`s&RLBZxLduKAZ{m za3+{)e0*I~cy$|mFlwg{Wyx3Y8q`Nr4-hBiL<(}bHypnq4yzAq&e|Z0ZLgBs#H(%)oGQkJCr#2#+ana+%y|z9p#}j|Ha1 zi+o{pSC@5<&S_kmw0UyXdWUGc>ydV8J)o!8EY({>bCB&|&PoD|wz^34hQM$Rh+9W8 zR!f#V6W?qoIf$TFSzAs+*6kz=k=BDCJ5GP}cyx;tNfDz0bf~NfyE&HUxSOK>5#I4I z)+=gszP@xg`Ue!svzp3vYP~cAI()YOg%8NkN*p-9nOncT87{5;@}QbB(=6#N!`4ok zwi?CCuRYUbLLx)kH;y|Om7`PZj=91x*8}I%%e1S;?bjEv;A!2dPQKU~&6@pR-^y$$ z9sqTngc66xc<`kIWo=9JV_6jX==V{0t+uUw(Qk0E0Zfl*?T@GKkraG=(DA!py~xsF zJND1Ap0p>Nvc%+6)Iw2~izn?aGT35MKF1r05jPR|%S#TvkgH*kdom+0pFnKu0s()N z-!rNE!KADaLML9tD*cilM}s)!fTE%CYl7WUbFxxvAk49%VI$a>_EXPpa$M<^aMSWH zv~>e50i%p#UAtB*AY#0-MA)C^+U}s_E#idV7vu*qUAg2RqYlCSD%5E@f1#gt#&NKh z;R|Vi#Pgz!)|wtjb32~#%c#@*Xc-lhrzr^oY)5CN;2XJS90Mo7ZJ1V~3txK7#KihM zCPr^GT@W&W7a`SR%dQo4_`!Y+yAiU9`oF1X=rTdQ3M%qA#Ei_l= z!-5~y20Uq0Vu&y?vDc+J_K^<>nFG02i!lJw`x;7?)(0N;P+X39N?5f>0>mw+F<)Iz zGA7nIElWh%M>%A*IBVf-(Vb9sk#~=8$Y(eirEzqCZ7q77);B6sw$j(GpQUNtRA&w0 zDGoRn;4L#hw5RO;++0?T(&J?~DXmeVZ3m_rar6@0BH?pskmqb~;tNV1Vp84+vnfivIm?g6xKp_n-L~@ zG%rXCP=fWkPtn(;o4{9s=6=)q2CM2JYLuP)!`PDRYph-MY@;y1Kc^!Hf2%Lvq%UTa$E7S0{?(IjI@;-!h4$N|#*0bJ(*q1;4&Xx6~1 zO>K}|ABwq?<4c6i`=W_Dlh)mjd{s94rzjR-S75WaFHOw; zg(n1w7W}Ts2j(I_cxZWSGMi`gea2{^&s9~h+QD6ATXax??tzA?h|`Azc`5@ko@z@n zqgO`BQU0Ch1cg!Af5fgFE1?PvgBYR51o`}*XQsW5UZW=ad?VegKol1Gb$ z>6H(S5KlNK(@&tIW)`&@ri=KgDClz15t*U_4~GyO$NA2DI_o4wDe=~L#kbRDKgwOX zgC4`^ak@wreMw>MKR$2Qwl~~CmHX@N7Jj#tLAAcQ-@}+nS-)f`f*dzf%DSKT*iO=t zfaX!-LAc!BNU0ukE(!_({z%697dq6CcBYC~^!|Lkb}u_l&Y9whWXV`_MOVl-04qg{ z@t7v>I|~oY72lTbj>~@~E9mxX*!Npc{-2O>GwAW=JT^RtDV*(mn4uOAAm!~2A1T>O z+3d_YZ&LhOG897C$WDRgaq2=<)tp^X?EZB$OoX|v8F+s^*p}JZix54WHr=QSU~l4f zU&M9gPyX5&V_Ji%N-y?pvl>X1?RRQ!tOvi9W5GNR~zta-KxO>+shzzA6`o z^02h*@l$n?4m*{xIYg?7rTH=#^B>?1e7VeL&(?{KXm{!^8tz?VUGkYIU+CE|vq+@nw=!)^MW-D4@r{CL|s8}TQu20UA*>s$p)*_aP%KSW%UKy#9G7xshcq4=O3z&Uajb20u zMeQzKib!hH6!S;@;c_9aGPXWgs0|9eQDPb4duf5WJKaoOmjr0*kjs4W45?|``i+eH z2WLkeI1nQOgvGeNAcQ9e<`b1^s=Rr*sq0kkTrBzSA?BUWce@NaY!)CgyJ_}(AZQ$% zTGs$ih7p9_;xVC+H?$o4WO{pfTxT|#05^{K>F?Y){IRvPlyk6(NsjcwTDIMqe4@3< zNwOTfv17>Bqi%Y$%=e4>Px(>a1A1_a#)l6`3jXfi9*SQ0hu2_je|@kOS<4d94872$Sr&)94OE;ec|wEoB}aX=55k_?0XZWU~SjcxH8bTUYEIGh3&wcm;XVZWwm$2Dd%*4_8yR_`Wbe z8y^Sz&s+Ob5%lv5xzcTcuCzml)?Ha6E=@rZv>kzweRzKS%KNSxZp0?q+s0tG`Co_B z&_(e5%#>^(ir@!5W9Z7A8*(-WmS4V7Z8ga(pt_Y zPDsyeJ#jR?@efM33|{WWM-3j}Xzg4uPN&SqbAjra)EF2*8d%pwT!@c{oX__}XTI|L z=53rXd?fCPPp;0eU&{o2n5L3#9AJe;g-9ZZEyGrqCro=H+DuiI;gRsSq(Xy^h@33q$xiNXF3;Meq=xhmOQ<~k+ zR<8u2W(7nX&Pc*@Mgh-C4xY68A^VJ#Bd|Bjx3<+g|v}N zSo&XKW)4S}>2WW+J=vY+Y_&;AB~TX1Pz(u_nDl}xzbu(9hZwhU^zvB_5DnQlohb^K z;I@r$T$QJp%s(I9I1A0<Drv4T>Oa8>CX?BGt>_#9rq@hf!tLlE3d1PUar|g z&CS~)-DhC*w7+6Yfj8*0W5rfj%c%!gz51NrZQN zfH(Q~dTSW1_vf&Ay!J57PTkgFROnmjljKtoM|{T9Mh3>O`@t@Y!5DH7!b(VAg2k+++=uq0949rbcVA)OB%$SqBS$)CaIS8yiPVeouS zl>?&6+|S|Mv^8875xMWt-bx8ss2Hs9akhS5v{X8xY3;>tKrMf>p=m(6J`-76#DrlJ zxGUx}fZtUW!x91J>r z&p1p7LmC!PD2rCwoXbvY5)w>Ca8C~7i11`;1)*nu@p0bgG|3%0)kj%tN3Dx!(~`mL zejJBlPxPX*!F!Gw_IMW!J!>N+T#1gIne;H!K4ZcR5@* z=lJgQG?PN7e9bB9&(nRv1!7mdZU$g|zUr&1vhw0xvv=G8IPy!Tbo&;8^`Cw=7%`DU zH&>xU)g{vc+ShB5h+6GB;|zH@)lu-CD^1iRfKJt7D$akJ_`g!>W{`gvcxb1hKF4KO zcf5&58FhE3k*G7Dn~&Ac@ontM`2?Aq5woUGD;`p*3p zJ`ws;i3azzSSJJp7y0zTU&IlEcWA!W zc$a`KstO8P|SPZA6A@xmrMD-uO0W3SOC+fBmI@?D4=QfZ;UUTL&1fXxZK9N{}ycsGw!uR z0!GB~|BitF-qlx20e1g?(%Z>D80j2e?E3fUYySs*g&v0d*7ALe|5!_L;=M7P zzqu^`b=IH%IO|`evS+`y7Uu^n>X+{?7a{M`MR$I0w+!gk(Bp15dA@eY(p!Ao-#dd` zhPONZFi%EiHpzw8zjGr#iOmumf&8}AnAnL~pr^@St44@2(0+#m_a0aP=TX2A0Q#-F zgv*==+`R#lI^5^KH(pl$9?)X=`R{I{5vWPjXRi*B{mpIb0Ni0Aw}Sl}QUl=Teg;4) zIVQ77V*K9tF?_L0C3y_UzZvFl&*CjRK*_`bYZ}$x8y}Sb$R!=s2EX4E9gcDUYwMX} zMt}aj@io9fjmNZ8{2t52$pe(!;L;6#`g`MZ89p7QCpNqy)zvST z9LIOvFa#hkF`-w_JMh{ekFlY@qh90F0Aglvk*t7}snH+*){5}}#wW(piUIq>=(#rb zykEce)wals@7K)zLXz4a6-diSx)mQMCdDy8D0~}VqZQ5iPvIffNGs-(rDuW~?fV}* z?fV80gxjjD)>NZSF5v|kc(dh0hn~5&UQMsH zK|!6xLPJ)O+4AIAVRhGbvRNiKmG@bt_J?tmGM!iBu0er^$*DF?a;iLyRJ2DWLx`P7 z;f=#mH5;4pA|phU&B+GPdB}EiW~<4P>lz#U z@UGE1c{a=NA$)jZ!FE24K&mg*>D!24b7b<>nKk&PQYHn+Yh*|-aQyY+hRjCl4$6WZ z5mWQJqE%$X^Q6yHEzbQLzk|X_Y?)nc_FBSIvqOphe)pM`0XcaZNA^zpGU3Bm%R5vi z`&V{ToeRP{`4j4PBt9jiJjS_TQNtUO@w%b%B4oEf+V%EY{xndCQX-(N%$I6+3;VSf zQjlhn;p&)By*GI(308uQuYT@zhfPjzCi^uxLu?tBokP(>-L!=_f1Ko_*Qefp7}OjE za7dp56f90D&gPEGVEA$X`-WmVNAu`sLNgC{EaMW1=ZNz$&=NF-sO*AA6nO*wbv8+C zmi&abkkXT5AT;J%FX0H#yJD1QF)tnLGWqj1f6rIn(aqsv%{J*nAQst1d6!1mc#?_B z_h#*Zi_NV}Vo>F_@ALYCnK-R>n zV*UcyCWMspM2CB=FInkZWMK20Wl1u!$%zYz&`%Q_G$&bs-659m2K{xJl2h#(7g_Z# zoo8M&EPV{UOQCs)NS-QlT>pSym353Th8-hcFPooY)S@F zX*-m}=avc&w45j3RbM~o7;N~ zNMKf1ub;ivU=g&HQp7y`57V*~4N$ah+3({6>}XmF?R>yn`f3ZiQM1+gCTPR6xK=M^ z{lxFCMx-|bW+Ueb^Gl4u8RP8pVm}_@2wM)f9dYsfaze08r&73>M{&0l`i(sm%o+LO zeYZY7GWcULf-}@d`y+eny5~5uP$EaEf89flL^8n61|)hK5~NyT3bZ=e*$i7EHq9ki z0X+IxPM1XI2oo`D2_G*74Wve2N&+Ni5*2JmQo?84jvKMf;3hm$9#|(0)e? zjAa{;6KCiF{in5k47Lz&mDtE@`e&xvK8bIge@G$2mRLy3_RHQ`flil)cxzdle7OEV zQcnFz-)vLAIHxD-lB=sy3R^NR0k@`FX9BV*HM7{9mHzy7^1Ch$IWV9C9 zoE}Ha+uE0WN49Qn)lP@=X=?lG+?vq$%Tkw{c1&4vD}FwfZmwK#MKV|KFp?p3D>IAN zJks5KNBa#)Cy!I-Ca3Nyn{xmOV76Jm?1OV-u>o(Wq#K;lwIN+#*$rqPw|rA3bS@2Q zwXQu+*sM430heeYy$(J&ED4w;#VQL;Ew6G1e$USI=Jh!6inmA!pBfW#JRZ#g|gn6`O=QdQ_s89?koyc-*EeI1@F zOFxa8hO~c5E?$oRw1z%N0Y=p@0kW9g(<*g90vj|nDs;ENJy^s&+I5JF_iv8=0hTm;%zEQ5@iW#tb zSA~PQ%tQ`}Yh2F;bH{KvNPRp%fGn(Oi0&a=Ue&#VFM~!YV9_fx-qG8_o}DBR-Ge?- z7snnNRG6JQhsWv3Kt-}udFu}$5W)>2-{U8AeI#7@Ac?%?CENrIY^bp+Th3bNx%)n+ z3`yHi1LKHg)1d6zQgw!MqvF2Q24&#iMPlg?Z^Hl>8c2lEG%X(a{^8s@EBHkdGy9_zz?1TQ*>xZ9PXVZaQKMvzdo*o{HYJ zr_Tx@qx%Hv;|FZym0n9`ZaPoA?*2G~VP_asWKSc=P2~K0Q`@JA5e}&I9BU zuV~;7yKPpcZOQpuZD^XpT@XOu6dm8Q-iaP^c7IrV-sUYTC8t>?B&3=IWZJOPq8Zd4 zAr7grtkMp3v3+HVaIqP5ZLZ>`%CAA~(yTd3$B|}vfn}4ooPC~QYhsY-1xmJ(##RdJ zXRBA_N`zDG?#p1ejgf!56Y-sonQF|_m;M~AaTKc84&iqrux;(oCF&IN3^l2#+Jz^UY#~P3CTvKbgU_ns0p~~)a~97A`rZl zn}w;$ux`cTtmu|Rz+vvW9&jsL&5i}f(^}I3Hr~jgdy_cGgJZ#Z#g>8DJ$^h27JS}y z#U$yCR(6~=nB(XWgsdjJYs1>Bgpi_M&&~Hg97w(=Vt-i7jh~Ke_64%;FlNX1i(i)E z_L6SqECdw@+t9329Nkh_se5cB*1&rO2GAMvVs%S>Ulungw^Hpr##ZuRqG$4zO zVlC|PwLvAzjA)$$B6Cr&P9^8s6wy?>$Xy)*{j${_U@-fxZZCxq#OLz*?sAPbCtj^- z88GCmFf&#EO=H%9i7bWoh}p)?oO68J62R>swaecbk~oq z`tR${H4~nDI9r=_>w{=@@fS6wML5b^O_N4*{R>AMcYzQjl%>Rn71>dyXIU$MIy0HO zWC!THh8F38D)^qLv-IU3MVsDrSIqUms{>ObFP{*~aONd=2Zr3j-1o|?cBje-JtlmY zdNd6@I)txlxQ8J(o#JM31J*!%f1y}azo1Ncy^i1H=QNWJUh{Yah?#^*12DREd8wAT zb!{6PEj}<|Zp^vQxrJR7xUU4~C_C};E%MLRST*a0m2A}4G`Bl2U2r{e=_;3mZ~jAy>t8S8&d|gc_`?-t|Gm1 z12Um}>EX~{AvtDn;4-T_Z&O|Ct|NVwP1wn`GFkS2xcU}3lL93I)-qh=+4tG(i?kiVOLRPS=gTQja z1pe{^guQRmc2A$ijR;V44U4-boCKwzZZH45*UL6M3YmZKc{WDa>x?bWuG&)I!M^;Y zOsKu8ccq!vuBpzW;Huo3_JzY>@1`V4>5}*FhC)d`3pDxjb;NsONU-B?>J#J(B#T@5 z#aH$1TirUN9+oxkHd+nqrDb-GXw!T)T~>kmo@aM)!+h1tIekIXhr$-zHXW>ENGcAEHh4RO#nxgZf!n<`pc`IXC=3ZtC{u*R5JsiZ zS-&^;x=pf&Ep2bC^Jb{Nr2(UpD-LLb-!E&WUE6pPamY@9jGTR_K^c?wf0}f8hE!j< z2%$djE(Le~?&B9HycGU$xXkH|?HxfI87_BU%f-Qd-?}`$Y&(1pjbP3i=Bt=IFj;Lh zoBM{j*@)c ze>(?`|0AHPr6k!={28|!`XKcQF?h&-1B@T6j!x)$W}G*Q7ipT^ zliiGbJ9A_Pd*SzN<3*^js@g;c_q$({Z;-car5mk)U4 zufB2RyW0lFOlgjAkjP{F&9fI&)Xjp&-pL-_kn(C26sU)GMf&{HfBS%pWRtIH z4|`U3iogVT^51tdYsZ_Y2M;EX9~ZCHmKfW@ZXNwQs?HGB)N_igxG262Vjt@#(C=b@ zCI7c(FE?OphkBH)np=YXhnkIoQ zH_9iWOA_R4A{gklPn&pg_{7Ln2RO(sx(p0SNnDG>X}@1YH7s=LT4iFX_3~GElzvaq zKJX=2&PKPtdGdizUIPB`_WY0%*t@_!ndvv~{6S=@PZQ-EwOQMlitt24| z8A`hWsYBH(znAt3`qMHF+EyYat84mqzJY#ltO0e7rS-)U;*Rq6I1h0zT{z+kRj*b) zh?5FZZNfK=r4z}-x2{zE?IcS>&(61>vp%0Bow6m%HE%8p`waYmrTTJ;N%>dLqeT%R zfnM}0JuWaj&uXQ8RI+40cou~4e-uOG4i{a#f2Ss5qD3MxWCP8wL0_wc^6E``3?#7M z$GLST58@lX6D0~t=c}uFUna}z9Ly-o&hAnWgF^$fTA2?zow+z80#n$=oMJ4g+VB>F?AuMn`&Q z=$a)-AFNQ{ar>2=oE#w&OQFg;c*^tG&fmPo#Vg2KI+L7`qN-3MJAAG2_LXyoKJG^c<}-jm_ClNw=41Jl z`MLfFq8@lZ9pcD|(kPSyx`tStyWLZT=U=uRLDwKW`C7NlV#pD=#Rv8}1e3q(ggC9c zso#&l5ao_PH<7}B7uVco^>9vd8FZd>4|CD28u0s& zM<>aNd;1_n;QQ=6!8Q8Vsg+Us(R_s7Y{HKztQ5(2g^@jZKc-xG_)1psSH+Gw%R(ju zeSMF}*O7dy#!K+_%ooKTU=i-Qq{8eopDQzlOgICD)ES$nn-Ai_<#nz-+b!=;AK;c_ zkMWF++BuQPbq$v%*_sb`;=z~aUb?QAmvTn`1rQ1&nzhlCO{XH((4Pz_3uk6u8ZayM zE_&|bEBMc}jQN~1=XU1GfZDfPSu>n&`!RFpO~#R9Wx#d;Er~gL)ZTLsnu!kOM?%AR zrvXxr5*yY$14Yobf~x8fA`~MvAAC>FvM&hwmdbxD=KP*$9kAa1Nz^GHES?#Wjkd;WOSxm00ka*3(`UQuE4L z{bMWzA;y5pqK=%+pC}Wmap_of)(x*pu5->+W*3AWjy_N5(@*RFSD&>P+)Jhj)K$TQ z32CdN?d2h&uBwL{`yt=n3ZXlh&xUVFC`lJBJ!#vRdS{gujM7T0<5J-7Xa5u(qinl)qkg7`dcw=VV>Q-zSRX~ zmY3twRCzyD2m9vyV&|NvW`gMcvR}7PbE%a8;-gM+AAY2zYuCg3DId?N%xK<7S{HRy zVRRYxej?6B8=-vMMk<>K%6EMJjAZ{skb-Vc`$@U&hTHnq6xg!J5R>09Z$(Y;mPPcB z&Mz%fcJ6~wTqyk=(B`tBjRX*x+gxWs*prjwy{Y7erdUbjME7dZwa+w`Hpur|U{xP} zW6Wnr^ett1qMam-FOb1DzqvU4;qeR}iprKRnD8yJL!!E4ga9J|`8|I*4;U2-5q4L?H zd%`^5rik>YE=z21(DrZt zP7!y#@p-s&S?#x#OT@?qYj;xz`Tb>KjJlQ2Rv*zTdfDzSdeT+V>Afwr+)~-I_4UAq zm5;}x38c4ES&g%ad$GL#38u}ch3%{@E`zq&D>+E)5F#GxOYq&ka@A+FrZlZxdeZn| zPP5Zw@m=rtxu~$i>9!=UJ)5wMnmto;Un%;Xsg8-x*wyqOg5La7l=V5H@v%w)2xTC@ zwk^JiFg!Lg8IDK!NQ5e#9?tL(P`YHSf8BFqp9MTG3g=_Am@1=55xQ{4gP2Yqq{~5y8|1@#0*2Q_1aEI%%Zj$bBwwF@Rbi^Zlyigz$!)g z(#VxACuEnYRneTBTUI0Z!`T(R6Eiv*Pt08<_wDR6f+yPW*&&_86SiK2?<_41a`af; zbruc(+7&ecjKF6U+Ir;;)6cE;`$vEwV1EtSvA+(rW)?)W*p;@{~f5+3?Uun zO_%3~9iNHV`c?D_F*$bQ_AxAWtzNz#?LSE5{C3iGeMEzL!sq;9<(KJwZPw94aNezQ z8XnJnOx$vyh&ZD4CXzqKP6QTOAGTF@*MD;+XJAEyY%*yMN7wJfT+H#?U0-A85)~YC z4BO3n4Vhwok7jia6rAY7ybRA-Sj9}BOa5 zy0n+SKE|`Owu(X5-)r&oZ4?~4COW=JC{iR84SWrd#<7crJ=Py{ir|wX&Q1O#u96i< z)@gi5oeVpb^6%W-CuWKtI|eQN`~9GQN%5$LI6QlYsUEKJ>1vyt?Zi2%u{}3OtEhIs z;|}cpOegYOHDTr7*I5Lr-9_|U(i_uJm#D)jcJcE==+U8Zv-;&(+pIuD=-(&=wun#y71$B|9#8)e{`n>`&@k$#(S3_@SGVya-R>bPV1da*ytg zZG9+nr80cBq=<9NM;cy)2!+4Wf?s~vJn@!O-ob&~1#odboU72uO$#^%Jzks#pG!Xx zW(B>yiC;%*i++uOxK-NfZo!@qq`BX&pu<<~^8(gbg}Xz6^HvdG2P=1X{rgRHZG-m> zia>Q*wrgzfF{OxBl1!jwkOR}+HN z(TJ}mB?sB9Zp7NamV0JJ0{e{&0kZM@18SW~C)_^y=TAQI7!yM>tHEVxyvs}$&gTBR zhCZhgSYt3{KR^Zy%gi;Y`fJNDq*x86ep#!j3ZPuAsn$+6A69Ly7g#xPF6(boeGwrU z;Tb8*T8DjUdd;@YTtg$q4dIW@JV+BA-VVc`d$4^iEk^nQ{@R25$ht|= zE|$;Ne$#6F;*3w^@WGZQYw|9vm&KzkbMjtK{)Qo~=}R+p>on5w>){9EZ8EEh!Y!fO z{LRc_o+qmTL+vd4Z_+Gi;0KMly6S6Gmk4JKvm2MWzPOd7=dew6B*`-w{ z6T!UGPuQf7eK^wfk4v7P;DxZ^#c9=ixUBQw*&il52avDY^Uaz8H@FZNW(aTbF792a z>o+Q%MIM*g5^|20-2!hrt_Mu|k2VSV{E66)A`m&h=%%Y3@=oq|E7O+71%3Y0ThesG z_9^y$6{(|@065s$y0KXVf2IB_%d(?y%CA0`qt&Xkm9-Q;# z_|+i)-j1{_z>foIDN2=Bk$D9k743VL+Zhzy!+)nIKKu%jH7$9E)Z6aM`v##oyC4`z!Vd ze2Rf_r)NU+*(GT9Y)l>zv1~T3x^TAshM~XZu?vZh5Km(JBB4(VKfEUxNj8}I;s5SU zU=K&PREFE{4ItDjF|oPzvCI_@&Se!lZ9xO7zxmCLjrWLIxt44DV&tm*Wma6R^1<%G zV9`DQ{`;H}I{VjJCIY$R8BVGOqwdkZO{osk165Z~Mzv|~)tT9gTy>727S*ReTbvU| zo1mL~pnf{)f;Wun+WqIGYTfP{`|m$sIhUrk`V5{Vy-*4nEB?0ev~k|}^`n%KRf0Y& zW^Ov-T`}eD&b>wrjvn5bt<%`vkeCQ+p)`{AZ5>^=w?C@!i3322e;-$9I^4|XNc7+S zwr_ZbKJzFqK1L9ZNjYBOi(}%RlNY@0)p(`b_>P>*-E7mD+uOjL3yYa_Q^=)yx@|sJ z662lckC`|f7i1a^YT;D;+sLfyJdVeE{olY^YrF7(9|!EB)<3l;tz;>1%%+^r-Pg0<=)W_9 zYfAcqLX7*h69-nM3?FKmyzrY4B%g5UTnl=C7GB3S5@jQw6+5@yp1~RU>))&>;abCR z`vDTfBbD{z?6%QdZYhU!R(T;w$A9S!Ek7K%yV2d2S=!IGR6PxtpJt$mJGWqyvr$wu zG2Nti_&%Rby;PS`a9&yb7WTRqwJjZ%74(3ayZe<+%S`g3_`9cf-?Tz=@}OIfOaAAA z{;P23KlioEN^t$9(E;Xe3d7#?zHeIDz!{v?)9>)iaO;TlF5aDUanDO>5to)GjlB5~ zK|qWpi%kgOQgq{Tu>P|8$~RgW`GXbspWoi|qD0G^gb_aj@I>ywTeuJQF@hdWh+5%& z`H50n6J6coj8*mPJ#)EncsHMi z^b!+V_OjJh&?GA)Uju(@l~o%DsaLG5QO>VDKh@=Ylw|K|H{g#JP^Y{OQ)a}ziZPba8Jpb?%AzZ=LUP`blUZ#9$n zs^R)p#^RhwV(r+6{@rZ#QZ+C0KAVH8G#^v^*!Ij~V|S6O_FP|A5>CT_5bkQ2Ejk@c z#&}={$9L+Ed*2WQbpWido${s)C>T(HtXh(#I&K>_7Tq{k6I|8vA>2oC;;51r*khu@ znGFPXtGetnkfoG&hRA&RQek#oRYnb>s%z`bjFqd_p=wEj=K8cb#JF}gU-A0(RHvZm zS*K`SQ1D9c%ic9?&D+5A5P75BVUz-5ZcIXF5m8`81?oTrXetSXM4HId zvK_$?={IRA*tca2+i1_ww=MP1a20HKS-uAWRqMgA0rN$(PtN66RRrtTJ`O4U^J>S9 z3p+dIUB;Zj$_q&Rd{IAqOKkbr&*Gla)r_IP#lfj)eA&)>xX}tET%&6s3B}bXN!{Kr zk5tHsS@A)55{bEv6eGaiKDql_-Q=0~y_A0IKX1#D9-Tzea{p+Lvf`QCH&fRdMsLRx zQZ$!!C?gXxWz1EKCT!WP#Ft;?fC&7QFKwzR!8shhm&CQ`O-5K4@%^z0OY6>|tKQpQ zCzKQcCRkvfFPSKw^ZsBy!!gsI00z@*(rWf{4fJhq>i}eMRf*Q+dk^ZQrxePXHwtgB zPtSMKeu(1VpUZ7VmKtrYrDz*&u1Q5N(ImZ_ta4t8nPZBR`c<}kaXWS$DmysD^>{dD zA}`ZKGiPJ}rhurgGo&QIwx!uQf>6vemq0y0`&zx%Yer8iP5`OBjDQlz9Rfn2^TQutcp3q}KoH=ztWyX9q`aAs_d9vp zw=c7_$F50{XiNd!wvSDdP*2dK5y)!_>JGba$4ghq>S)sWN$ zubrEKGDYb zi~&KHVrLwdHx%F@F@F_eK^lyI6rGbRowOelAj1&0yvO!#o!Re^E|1&61pUO!@ zfhgn4>V_tgj;HOOTOYJ@kPOrJJ#3y;kkGurN5_Y+u(?F#9r5Q%h-L2rhIXN?{yYPk ztI=fWwBV~~+{1Ox>Ou6{x4u5aPBqA!C$M(+#mt=l4=46)S*9azSir)3_>R}_zMV#M z!8mjE(x;L+MeY1J{v>+|m%j{;m#>uIQ6)dQ@kUB2Mj8F5BPhAPI62}6k8Rr;b!t%@4PRj`nvwmNp^F=Ql5C)|l}Q_f znV3rb{g^L%_njDx@;-v*7vSp^0M(Q|>HxRFyf9)WvDXFO;aQ$&KP{ee{dId7H_<%T z<7wZ^oykAeCirVtUwhthh+JMlXOH8eZa>TubCSBEx&Axf_?eIj<5Yf#i=u(c{TKh< zKaysU)_XNSggW^RmHp*6eU{d<(J7YsN2$(nXfyo#0rTT~%kOwssJeQvmv4|yU{3#i zJ|nLbL6twJ@)hF;v&z>I9m##%xj_Ujee$)~-b*_PFTV7t@tby!qV}o~ zhBSRo;Oz>#kHliDrswWE$CdFq0R(U`Z8p&4Dbelkd! zlU$5Gws9A==KeZgQoXB~;@-q`aP2qI!5W-hZWi-rw74p2SW74V8kU?tm4B5Ae>$dz zq5~?c9+9x0o4-N~oc5SHhmB*14Ovq&K0W8Dka=s>k`e3%#o2lbn@oX6)qkVz9`r|R zE6-TDI7sgKgczI75HNvX6-nWq=xj}M>FEd`hJlTGnKhqw69=YcCTyI*G|};QiNA5+ z>59)<@D*690)~fu--}x2*SBneQHc_NE>^Ac@61q{R?CxKh@76wEfg2(8?A1I)a7P* z5)zG8Z%-NMr;aIQ8j&+RKfhDeGlEo8Gn2z~kV_8-#WF1Gw;u0C{g+<%j^5>WSK(Gl z{kA=4aWwPIXB#Z}pX-k|`@(~7zU(~BbG_!2ekr%J56(YfJfDJoWzV>)ST51}@G6|j z;_2-~LS1pY3%2=8CqNp(1X$}^v10zOIaOqaVvz+aQpm80F~$B#3IOVs-o5%yyWrtm z!@OMY?0x77xF(aLI!cueiksjAOJVNoxg+CEY1ic5x z6kmA#`;-7G=sf^aOSl=Fd4Ds-dmhXA<6v-E7S%f7;fUHx`^6sbr&&$Y9r)@(zatF7 z>Q!Mylmp(4`;z8My9Q3(>U)_UG4o3(*{ErBsKiXx5B3DJreFF5GWO|}ZHwTu`*K(N zOfbissW>{W!Zu9^1!n^5+ZFBR?=eQ0_C;_rjbHN-lA{-CWq`)kil+wb|Bx4_oH^?+ zOwDkB_ck83#4ZftwIQiK3YUGF|Clkj$uR%!yje24c-62}K+$kt*70|q+eMS-+oh4p zrVCR{v4Gbf_8t54-pXmAAbw zLGFh)>>Gw|4bW5m#4pd^OV(^@BJf^NoqT>wYRuP5kDmiOlZHAN-_z{b%Vf^V7CJ?S zEdZ*e)3*E#wV-tDdKCwWLa4blS0--p)0Yu5CEo;cI{PR%r8VPM?Pbdy-Y-H*K1-Vy z8!I&$+a3-qo9t$oHe0~7G8bhWn!-U_Yjex8xU!^q_5XDQQQG zW=hR)QU8D*kjXH0h5oml>5kr|GyKi@8*)_HCK&WAPAiTVei;fnDst1WuHLZys1=Ze=1o>E4cry9uqxZm~$o}u=|qL7!ZMzrq!3qjP| z(Ge7{bqbi{U0$)prHS*}9@VE@ix5KN-n$Wgmmg|JQEy_Dc0u^v;H9^GxSHsL?=4q| z-yPJ}c!fTY5`~6U&DO%ltmn-bavnmnU3r%+23k7(vJ3-I-UadiF2MJlKM5@VTaIli ziX9Ye@}jgWE5S0b#0z*$U;pdKmEG$*WL#=eGtiekAk(b9zYcn;-s-a{0X3Ya!t@uu zT7*wk_HUe07?-qjJ5g?V)bnmjseG@(U)O@#TkA6UWddn3bo3l1xFJfQ**++Qb4z1Y zhGuyq8^k3go<1RQk56dU(ojHZAGl)L(Y`1*3*t9}`!g?X9l~}k1_3(3Ug2}AdmT*X zk-B(;Z%Qfc?QAEuX>BjN7f^54m1X4=@OQy0^J*>9KZ=0%M<%?%zTVaC!771v$qwM6 zxbZc{$1kY9d|A%9)+$h7Hl5=9tNfvuYkFyN(ar4Yu{NW30~=E6YAIUf~bj> z3$BV;N<$U20Do0YCp@zP6qo%zAFYVjWZFHk3eymE@27tbMEFsuyLg~t&p-}qGIGJT zvCQ&#ru9&mybRRvVH0COwztVl%#UxXb>p}H!-*qBVcW}VYW(Ah>zfhp z@+|a(oO@mO|1kCQPB?1pKT&6%B*&~?gBR0*iNUzq-3)I%A7vcS}i3s6DCOv*|nG{6H$iw|;)X=Z?kF z#7q3Vtih zm3DakFXf3BHr|bWpGW=_DCw_J8_3$KyM<_C+}+DyoBTAg%_}r5)gr=O!Colez>-(Y zrFs}0bw1N8I3;i$z#483NZ4=b?YCgV2!HLm5yB~-wNMLuPR9<)VOOy(`v?vKi1;X_zi z6OV%U*cItk?+Of!YKab9+U9nvskQW8xp0IdO(E-X3DL+oWp?>GrFs2F!TS9QKjgjC z^&mJTCDdafaPu$@nvr;Q;T3Kpu#SY^8`Dn!&*(0@)^aVeQV`fF| zMG+|v^N72q9jocaoYPIzmQlI;)*%aS@A%` z(@1S^8jj`~Te0vW0_a0wDVDp`|18vU_#a z6IIQ`rnOG`=>f4F4f1aNEAb6(LUdWOMny&-21G_jPgc7(msM%3{M?+@P}#T9}x1aEJI{a`wU^E`YzI0li7?Luxoh%hVamHtzxJ7?J^S7Q9r za`Z#>X_+MVWxjes{@j6FZV??<{zM3BGpD^RH&Io`x%z0v{6Nzz{M@}5=FQmAsgS|( zaTJp?YjvQn-H`fh%burGswAPfN7@(cF(Nq^>eM`rHmut3&{Dl7>VzRd=}m^I1Hqm4 z^z+y30f!7PqiJhXw*rn*cA&3a=;EWz=-5~-NYRkXL8$b@&hDPs9aFZ*?HP>W2Xl#qItPa zd}pPsZC<8F)y^^vGzFUk)gc+c=UjqUSSwmN2aMDc!9q$9f8wWKKGFO09;wi02FNh_ zii(}sfo`mrne_hE2+JI>|5N(|3Xo-1i_dlw7U!0=qmJJwi6kHF5qx#UR~P4uZ~Ug0 z>qNj<4BOZH%$d#}lA8NeJ#-tJ4+{|2ISgeYQX@n33By;7b19YtWtrd1g|@fMtzfTe zqtlp{ev;OV-`*C5Iy?H!g82>4!8B-)>!V`7l<&i~`6eLhH=i^5!o|;B%gr^xP2?%L z82Zy;<+(+NOpUvA7_y}9f8KYN%2u%bL_;Tyjx=xz4?kQQ6s3eO%JC!@e97f4UweYuSVNV)K0fx!=E);!5J7y}>84uIIh7EU=#i)mTa75O<%wy{e#>=eX?QYUX5g>lR+E*`)d6;pbM&vPk^w2nwtc=opiwwXgGQ~a|G4@Z=RT~6Bg`v|bM%v!*~8790- z;2#@;-%{8>j?mrV*WRAg#oS9RG9{-a($1zC&VM>+%o7t(9qx9N+n2dw9AK3-uPP@7 zJl$9>&O83I{$t)->WS+&bh6Cn>w#k}8(sG^al(k|+UlM9f*6G?+xra277)l7jT*T( zZ%qQ2>qDKdZ{{Dpqjo(4)zG$9H@#nsGm`xtez<3j$PX&cY1$ppxIK7QuC!m>amL`g zMF@Ch{zrMACf4L#AM_+?_M+2${P2%v5V>(0t$gJ1T_3fk0~?rhz0N|^;0sEV#yo)< zvHyNJ5h6@qTDheqn=}eY1WsJ%*7*A1ljaA8RP$c<&N26%;JIUv@hIPDJ_;myPHU|r z;=};RRR{UrM}BBYu_U>ue2?qG^6aeC^zQnwYY7@aJ+bSR3FS?~Fpt?*Z2*s$bU$i7 zW;_~}oVsbTgCpgW?@PV=&^gmL}h%xN8M zI~aUpzV&(QQzh}%Z5wOgn89vzQ8fzgZQHBpGFxH~z3IxPsy!D+gK>oa+@;N=FsnycWY@z1zN>{UP^DdHz1EA-@%(z09h796MlNYxW|juvSA zAGY{^lTja%70!shERo)Qz3v^c{^v^7=UK&8dgbdAJM4Ah(nJ505G9AX9Nd*3^U`F| zA`6obrQ?K8jwYL-^0tpbc6o7sl_pUoD|L)$P~OUJ|B*4J$P>Fl|8~>g`WC2)pPsFY zrn>Oq=8ca@zmWCsjXwg&M$Dvj{$jk3x95#Zvy{(>CPktpPOiG0aMv=tn-tYlo((N8 zC%3R~yh%4Zyq*T)U?U89tR632&+}06A7jdQ6$#8LOGl82-B!_V{3Ln9Su4S4nvB7} zIN)Txrb{}uLP81-U&RJNjy>%U!qHn|Tds1^JdBAM5BaxCsa+BBV!!Nj)=?Iw zl^P`=>e$Unu15>`rwdC=KBRwX>8uNir6wZ%!rwv24k7NV;g{*WanSLz$rYodoR%xX zv3|Qxh(()Xz!y(FeuRVF!HY69|L%UpqtLPt_kxc^HuaOuO5MInT;l_0!M5jH>4s&E z=l^)-yeVcB=Wb{`Ek>JLpX^kG-gxf=KrI}t=I~fb+bB4k{2ejX^o2iXS-3dK3b$nP z+C?p=dUWEp>q?eJ-A(gk5JAFqfM|BgW7Knb(nHzXZqmH(oH&QwM9oDZ9;5H;_*5)# zg8<@TR-Y(M=ZEmGZ?K^IuYl>YnJc+YummwOX6Tz5ZDlA`T21D z_?N{;p*2mpB|pB`Rno>1GFxPvwPevfd+K+0Z1e;(ATuhuslA^m@Q`7@0q4STeAAgX z`ZiiZE`K1ZrkX=Vi4!Z>vgALK6QdT*cQ02SzeTyuMlTZqVNa1)S34a`#pg-3 z?M`~9NMQ$gpR77N6!)1I(fD^i3|R7MR?(<9F5WuW1gi+nDH`^Nqc8D(FYtF=Ycr}< z8H@ddhoTvOpCk8Vt+Q|F2A{znU)IKz&ICo5(Z<>MYk1D{;|wQwdeh6sXAA9f%yX;5 z?+@HR$z@Y=W3W$iqU2)FP1`Xihye@_9cH?uB~?miIQiI-L6clbeqa8t z=pMttfXBf!d^DqLigJy&O^~$lZht(iH(U9ZJZ|f)hVNsL3&4>ISA*67U9fZ~EI#bnt*LDJD&9yb3_AzTI}!MDCDGH2mkTT<|@ zr&RLC;1TQ2M#f~qv=>C&lFf_%el%+-Y3Npp3s&4~MR&ZIJU;a)vsLz=7lsE0e>4ed@i7 z``dtbrgzNZ`NE}eH2tUP=r|-yBD|J*_*g4ke;VMJQ_m)T3A_SOv02keehz^!q&4_mt@#~UrbD@N>BRR0zvStW|R>_`yx#j87v+Xc8{%p|?9s>d*h?z4187S@dFoW*r26$_8 zkxD^Xx9oe|S|bRK8xX)>rV83p|Ap9QIEGtdHW`4YQOWXV0QK$U-_Qnxa0z`N+fJnK zc5_)h^&l2OThFYYig+sczNznp_UccuwMqN5vTk|o@1|0jNF;PGGj#S>QSUpad^_7b zT00WH=~T71T#C&D`s2O2Se>eZHY_JHxRjO|EazjBAK*~dzghKqu8+YTJ6JVZ-@(1m zVCzpF?Y;}r6A!;@VntoSZXN>m6qOopF)Le;chz)vv}?>f^IvbK`I-y2&gw-)!1@Id zrJGP|_$7ObkbRJmuGIYG)rHMf9bs=b?fpf*j3V%M`R+!iUSF?+C?WHG8~gc{efJZ2 zfjIWlB5Qpdqt;_UT5JW`i0b`Ut4aeBmg~t{07t7BIO0XuMDS?tEnoWUHYC#5n++gD z&mdfFLf?bY+slc<%6_M*8kG1<4mTF*GpTSy~!04KHefz>1^KW*+ zEA~~{mAm@hbK_(OuT}KPwd;EpqhATZ+=HHF^2tSgcBA7f?upC8#^K_t zd-v44!}jv&S<5SB;w$!e0@hGfG2%c#jz=7}C^p&=+Svuda_T;jv{@VINmEAXmeOrQ z^A#{tVJi(GAdNZO(68j;8+BW=w47!nd~dub%u*aqLS=h?FfR|F2>*%H|mF-*qVYfh3 z+*pMBgaO{j#e~(Q_EM^~zH*o{eXQr5F6*e&4DKx)1Kl1Tl_PBv?!itC2ZQ6o^0iC$ z5`ol(n9$Y!YpyG(+S9GS^;C>~n`yaAV8#X-_tmisy#(KKa>eg>dmlo)^GWx?5_O?L z6Q>0n{)Z^eJqt#wjA`I{cEWwKLo(tqeMjX>HtYruaAr*dpNDf7@H&S|!ajAc1Qpg5 z0NbG%#1#v-vc5HvYoax=zuC=uh1*QL0sGUm2PFKV(CgakgKsN~f+Pt_tZVnEmp^lt z++M3G3Nkl|t6F}q<&7MF=eZFx5`$%jXGYmp=--tHzGa>fpzzciUQ~UgHnCq3 z=aP{e*8FY2^!)tO$2b3*i2s}BCH)^g-NIaYw(BYXg3Ow~AwE|aMyg-aH(+}3(8c%e z3q4b?*V4|yAu0JlC`We=xFZyNCct(2fzG7zVwtX2CQ!*=id98^5c(&Po7+`b`loA= zpE3j7YEUH1K7IS(>%(d@;W`)4yjuExr?-u2Pgox?|6-(yp^G7TB&%}EHX`m6h+S^Igf28w@4 zK)YFQN;|XqN2kt&4FCHFz$9NEF^~H^hZc2;PN{vBy{?CQW~aCiCUu5cKI-6kqbf!8 z;Q>JCvTmrZL%SKC#EWUonJOktOeG{meNFu2nScoTm+YLp>_3o6F227}Bz3Ps>a9Id z%8@OKu#7Wumm<1gJe}O;l3Y6rj|dPF%y+edeF5;|553H>4yV+h{-MWq>%?VaNa}q# z{v**H=pKOblpibw@lc0=NRgT)AC}?dj(+3_5PX0T{yF-b1n7pi@1xcHpF~DJ0P9?P zh;8lvr|jdeKVMD$6kc_?X$1QWau$YufT`r`LgSv!@~g~V`6TT%Q9C}&5i|UiB^%bQ zEx@?gT9QBmT zZ}Grx_PV`(q;lmJik?}de#>YkMY?bo-|BU{=czBJ?33RPiqN7DX_cGW}l zA}*@&RV|lAc*Y1IQ`MAhl@ploX$t)|4#QYlH+fvXgK0|+p7$mrz&GiWp@tna-2SGl z6&^@@QH^dbski3}9ON4PcmUpPO*PT-&Qd0h?=|0ULquSVwKtxP#&}jnKCUxb7yF!g zi3Nkbjo&m5eyhqG2|%HeJdf+iBX?U1AZ6-RfOo?lJ4e-PdmhCo>PV{4V@gs-ttm?Y#zBuoxo9WaNB+2 z7XV@KL$H^}8*fo>&>8F0W5g%g9<5zC=IVUD0CZt?3vk8N(|3VjCNSJw$AkB^q+e)0 zNgF1&NNHw6)=xaA-Xyg`Oe09w#IlZql27@i4sY4-HqFEc3vKT7G-L+s1_WI*?D0${nn1q@DCvDNz^gG{DP zK6$>QH3o*&TTJt=Zvy;^;NQp~TixO#&*b;r{`m?x&RsuQW|uTn)?UKt z;^FvXp&8rA309fN!t2oIU6jHNZZ6Ifnfi=$sp#OrsJQ4TEtH>n)#$H67*-{^m7DZa zfD_wPl-Nx{_r$b~-WlqDAm$+z7$jf)RVzbHgjimC=Oy7K#uru8iXh`zi)=yVp4E4Z57R_9%aUue=TE zK5+K@zXR3LGwW$4?pDxlBynoOt&>9#H8A2qQOoDYX4M4{erVVe#lcrheImN8UVOG> z7AbqQA)(8HR^(pTMq1vDvAe!gg|OU7_kZoSzeSbk#Vbf1^BCyHg+r7; zx$w&Dy3`wX!~9>Rwv8DUOT0a{n2C0dZ2&y7_^Pmb4o8icK5ON8ZSo6;Vi9hllHM)ef2s{56D;FAYeIzPtW>*Dk_I{8pzQe5xe57_ ztIURI_veryQ`|$fHUG`bk|-)CUZ>kGoEVaCaKpxFC2=ZZ_*ngbtd3P;OzCWhO$pV9a73>)Xe$BbIj>@yeGb^WP}7BP+95V?*J=>Cg`9MiK)+g9(~-$NnPj^^k`23X%o=szb4fPC4f9O33Dwl|_8d;Rq}r^Ad!ZGu z;gr(a_#RiAwVWX5IBf}n2K=5LKBHC%jKI_;ZqL=dOu8vGLBEY|PA}xLe71WY7f)jv zZ#+SIZH);Nxq0^15+%!Pe1==m`^`U%Le}Z=w9MrH@R7yqsE~{=e|h`CA4a8%HL`>G zU|zGjZ!!+p+NpNq^IH9^+F^RR0blJTnHM)-cy$}M-~9oso~~2i)m0F2{{M|8yfc30 zZ#EDJ2r_(i`b7Sbn<%?M%hy6cgc{ZxF?}xJsYk{H_u3$S0ehoFSoxdP!w=VM*fa9~ zmE}B!3d=d%CA+q~j`k6DPLn`Q$-f1hlW8EXAW!}`Y@e#X>K!B6NSL2Zu(WJ`iTU&4 znf%$uk^3%A)k4&{WF5tcq$jftmVfx$-E4ayB-XLwy(0=D9RyUmC?KIo6Oi7UH0fX{ zA_+A}6Ok?*r1uWeTOuGJz1IXn?;-RS%8loF&e{7sW1Ktoc*p(YeaCx8@<#@w{K}eh ztu^O&e&%`F>Eebq+$hnOz3YTWzrIO5_TF!l!&FyisMYFoMo4Efr#!+HPWB?_vBmbX zx@0wKQ@^g%X?LqQyVPOIs(5J4&k6dSnzN?Z0}ia?tEu%YS1dG1a_*?K)orem1`yeO z*LaX~RuaC2ofV3JV<5D>suDbR!|fTXd0PPgoD~ibP%f!uc&hqr1~{MU3}A9ctppzI z&2y-ZjsXgIdvzmqKwoAH!Vf6?rpj?L;9dsVSz8vt6-2V}1BDX$FrS{U9OcfO{;+5L zU+e<2H#Hbu&r(EuN9B71(Z|5%UE5D&joDhcsd73#I9R1- zr#ju^k1sdxWX|)OK!eEt{5bYERyeT$%T&!Lw1mhafknbU6X%&Tn1bqgi-SNLB@i~qHp z?+@&M#pC@8v8YV*uzUUa^t#*$QsRHb#%AW?)});ZcAq}0#)f(W9KO3*y4ls2=R2W{ zY|xWEo--v-7naVF@Mq1Jj%fRz_&fKZt@n3 z2-KS-+@14O7y6(9(~`Q8aC}%9^fCxL591P=u|@}Ae~MU9^xvoH*J;0}Y04Yuhh)PR zEBkaAZiv9Pu5m|cZI?qQ#UKA>_`h!7N?VlJ-h8pP+jE*Mg7oV4Ih~_$>N}jRbx3J) zcS!!t7Br#Q`5MASg^PQjY`(u_*8jUE+W$oK+HR~#H|$dK<1mYc5|IX4e>9xK@UQ!l z!k52MiAjccV15|i8DhipviNzdaOFWZnCUP!^Dj5&K;dP;=W8tRa`h7LwcV6$f8%@0YYdg!|8i*m z`K1mS_J-@SXDD&}wZQ&Az2E;e7Hr%+YQszPxA46h&k3>A>j)pt_`jb1e*|nl`OUw5 zom%^l^iSjYcWmWfKl*Zl1xrnSevqsq^Pk`Jr(m=2|6cUJI>LXApZ`x^^g`@rHg7=i z4*tJR=Rb|yf4rdzR~Xx1U>eie*f^TemvHZHx25Ez2Zt211r`3^8UD8jbr^8sU!9eU z{tA8k_db9{`1}UTdj`w2{BIBKUj~x7gk8gwi+akd{11fYfA9}1pRnG6^sdui=Yaog z4=+1eu}0&1@FnSQ8O_I+q*#-@{|NCn>DymqvF>4M<(^>hpN98WoBbbu^}qj5w7j6^ z(%iOf|-|>R}uVwy$HGe$fz5g}Olep}xtyCXK_7lCd-0^Q* zpZ_pyG;aL^G#v>s98l`b>L*dO2w>2(R(OY;xp|CitX-}qb(Y@e63 z0@7F5t*K(bzprxt(;3TQS7q|J2`dbXfKZk5CNm8Ub3G&byn}m1Y+dTqzGtYWw^Ggha*|SB`QuX=!9P9mz1ZK2 z>w~fYyD(Q@XEH6e5bkDw$o>{z18O~*s+!^NQiZ&YK1mFvwKk1_o7oE7gI=xW|#8RD3 z;8R3@qQ7f~G(5e3*Xg7>w_48o-LM}bZ=ANamOs_0&1YU*gtlQ>&eRgYJYRbie-R}L z;(kdr^D;Y{|22-~fz4Qi;0tEaLVqK|lF?91TWPb?^D!(=l^F23ae15C@VOIWc)Zw< zy-Rnk%;oYaFg&fsMg^N^>d#Mg+F62PDZ~nCaT^$H^56~Tx~Cu`nIC#KTH3Vr#3oi% zSa={6^)6+n#d5IesCH^SSzR*Od9tXkZgafKH+1RF@f*Ym%L6qf811Pq*}9`=O!Dh*Gaw39^z=2Scq=!fIi%Zv}2yd0`kMM4knAh`L_Zz;3hd&N{h=D#RnIUzn) z7#Vbv0vc+pT)p3Qqr@0diJUAmPI|NY*!_nfc?Yb>sP>%=2AJB}@Z_f6DaECk&KE_N zZy%f1e{88Rjbti`Q_9u1q9dyh9r;;YQ9UTO-SA$&0HkUxRYvXJe){yG#tZ}<&nm|B zr*G5~{k^rdme!?#J|`Uie@3Kd`SCZ1^j9(06-kK`mP|a)1q93&TxhkhmE6faV6|Ve zh4K{H4@qq#B;X8F|HqN^_d_KIjOs*w*S^{ParS?VNPjal@A>0O=gRdGPI*j=l9#2U zdZ4Ejqj@)!O-u0s+gB?q8&098^m=h@F|ENGH6$GNb_LOe9@pz`>4T`tT{raA9MVgi zBID3YVyG47eS+CJl#s=jnlL7jn8Od)dQ)#$frNPe?BjvuQ7b-dJwd_w<(1>c$miTn z(?;_SbFwOhTdEsNO->`<7%bozes{6a0Th;u@W1rM!Lv9|irEDEMq2}Xq}M77$->}G z;LU|({=l84V{!&j7R#=P#C__5CyUoZPQWXA9K1Mx?!New@$T;OIZW@$81XgO74*1Odc7y;1Jnq+?D z%z+h>No%qA$ZHh%s3cAv-0*!CZFPJ|F5fbbrAQRH!A)kjC(Cvd>wrc<ny*bxOX)1J+)nJw(qrB$vP$FNe^|`SwgE-H%enovssn& z3+;hs{C0V~sRY1w>n$czfYIJ7WXuDts0`>z#8v$G-_J*9YFbR9U-6JHryCvllP^X; z_=cYquO71D4oZJ^oOEJg_WQ1+20-Z;Hal3fm$U>131BOXiDQ=hk7i!Se&+n27Vy zCSdnxQLQ~jf7uK|jjXq1ZMbOE@s{EZbdH_u?6N(Zl5{uA3$*cY2Jjr_r~|T*S`ih(C*FH zragug$BI|F5dluJ_!<_(c^iK2QXAtETD;xr{BTUUoe>ONY6H z*4WLcyzUlqS)t5&4;43KOXU9zBp~U{C*nMdQdxi4!o<@Y>zCj zMff>uANF`^?V&OWeU$X1_=jWfI_N>6b?@MHCQ<4M{K<^H1!=G=KI;h_ts1)LPK8g{ z6>|CY$gN+H%bp}XX(onJp>`!OJwJQ+x5u*92GeO=jf@!#7`UTA8&8xI%j52F@Wl`f zMSXy#Cdwu&po9fg%m}#hT0~vnsT5Eoc^o9)_wTT1NxXk+mh;vfmj8PC?kF2FUUeucEhAY%k54?QFO3KOcezm^{ zPkaP%Fsd%MvAt?qimVKC8cLU1iPKh#oNyTJ=W73vp4%o^>wUT#pTL{amlJhaPa^&( z6a3uwrCyn7NtEwxf=9$qp*zjhJy!vipImwN0gj2E#E7A)aJ8q~5uRHfW*nJX%{pJJ z!Xy<7lfzrSy>2^Q&EwvR#nVX4QajWNo#ZlgvRSJ?VQlIt5wfx8M?2`4p&Yyf*yaQU zMOO3l^PF#_jix`FRP;^A+`~Bd+qpZ{oZ+s<*ny&+U0LNOO3XY6`{Suz+vycWU3_p@ z!sZxLLdBahuvUnn1~spb=;C5Y1uVvONcC%C!N9QLA6%I#UKTZ`<21_Y!(oBDcAOiq zm+_A@RhXRSWXiRvLM*pyNIZ z9`b=|p|?Pb!owrvyf zSe!l)nwVhAM|IZA-J?NTP%({?+i&EN>oP@5a%Q9bOM*CIqN*8M{zy^R^;&(dO>TJ0 zQoC~zd-PnlcM-d6qS8NIjUtO9b>;#s$Ys-)sLZQu~;!SQ6qqC=r? zxIn?tt=d*Cekt*8gacQKzg)1u2m(7a=m$_c*%^ely}zz?J(#fCZISoNwmj~|(?rK! zpTN;8v%MmlY5mWgo22LbT3@o~EC4JAJ%MZV2Z|UzquT(r%EO0>7F;^X@8j`?p2Tza zR$Gony{7n1{fd7=W zf>x^@(AoozefEjsFkuHW&A_leTJ1tzt3Ag_;=-ZZ1=U{JhAr9Y%`69wY-wUe`w_D( zmL6f;){FBCNNLwM^ z#K(2gFfdZ((G#FJ*Tp|k>Fju5*NB}*A z8j-XzoeBB~)U^NIF!BkN}dNwE&xvYXCN9|)7 zf6^6!?h!%m3iHX$7ZzckE zxJxz|k44^?8NQMd5`FaLqR=sxW=Q-h&RjFVSUTyrhbz_Hbgten<6B;22Xw3VyFY;9 zGAIQEabLHthc&QY=0r)wBXz~v--Ry8;}zh&48G^%JX<4Q2fTTW+)U;g3f0QhkRB=_ z*I*LCQkOu@*4y3Bq$+$x^ChG5sB~bMZz_1Reu`^_x&tsU2K1@G6 zKg?iHEbNq-W~GjjX?4LC+c=aAMbPubbDJ!0ld9=A7|$6e=SL2O3}XfXTL7EDtv8k% zQBv0$GC|P*_MWh-~jvqkoi30>YD7%)OHnK65w(&?mktYUYPuf80w}V~NNAB=f*Y(AOxq zin>q8aPwZDRSBb{P!V?kHh%!oU+uQ3zvriGTc3oFEFgdTp*w@*78iodw{tZSKmFJ8J@Rip)m{T$ zGc}%JI(p#7rO2~NR5ak71zgN>2%C{Eu&H0Q&G5^@gL&%s(V+OYDk^(dfHmH>8NT^< zd9_3ZoZ-9Ku;X2C<}LR8)14W1T2l}IUB4Vugnm}~4;h9y&+(zd^_DH}BVkA;($zPj zsj1Y=Wbw6oTDx5ihp9DH3exwqwbU3UeXVuQmT(OcUljnvB^q3K8dgOGJ$sH~>SEm1 z<0o#3Bv~#iDQ}eKAM7L&Q$N}u<$C&PB{oWOJ{R+}5EFLy@y56(9jU^(c~sF?*0Wb> zUtAA}LpXwbgFVYlbbrMl1urI$XY42sm_a4Pcxr%*cFFq^NS2TUA-E`B8M~mvi&*O_tUSd86yV46oE6zci2sOv$ zi+E!F_%U?Vie@uZ0Ul#ktD9KNJ{}{oq&YGx2UmTA)+&C(!FaPau&bqCp4;6iY&kDj z5m5DjW_uN|{?3FJ@0d4S;!qBTd+xSAnAYo*^%=rF3PM7Upr26rbCF3W6LaNRQOkOv zC)*us1of~pM&%bZ%e2q@aKX!uWjSvO99lwLG-LkdA79?b*+82ZmVdaPLuIp6OS`D} z1){N+_OSAeYS?XOI*Kd^zs?oecC& z0%3{Y1_snF(=q^qdln1QecGw!XS~)VsPMbUU3}7o=tN98i!i4G{_nVxN=!;V9c|6bxUu?YL_btPR91g?OEis)9ytqB zDv@k$va|T`tnHN8{pZ1g1$(XeuW1o=l@;*zLDHKX+$XPQNR0bMF!soq`}vYe4=GYE zaAlmwpZi{rjW*k$o_+kd-A1q*qeme;1%3>-h&X|+l^u^3?)%Ofz~<+2+feQ~hWq1D zUAGp2v!!`7*(@;g7!q8fSLdr#IZ^9}%Vu%5uOzq)6s<}SjW=9T;+L6XUQJo(P^z9^ zFuS3YLPj&P37K0e4$@hS?ay&uC3`H`U3#PKd4SBmZ&v5(*>-bYwFa74idM1d#g`s^ zvHm_moNj4H1zKJmB$#SE@idX@!{C*ji#`iM}hcA{wZ%I>3 zbbq9rE7Rb+IN2@6%W#qary1(;GBvJ4C$Yhub*fjainoxNCv0@qIAQ+yM`*f5@Nksf zrb919B(mG}@Rzf}p4DB@3|M<=(Y)=J4w5V6m)nr~XgirDzWNbB({!%4yg>5GB~5Jp z+QhdWKm(W|4$~e^H!=F&aP^`C0HTzVcpL8auukw}CzHEV(PJ#OU}J`}sYB2PK~n7G ztwu7kh=0kacl5R(!yZwa0RmV-yJsiqyg?qqel4cF0mdv>84preO+~4)^rO`LWzsWS z$g;e!UGSdrm?}Wg)LkDgB2GNyCd>Bo7UOwM0LF!OXY1bK)6#a~ z227DsU)7M>@NJ}S-@3!$D++Ei$Ej9X^kuuq&dau+BNYzEg18y4EIO-=&lsPH5{(9? zCNXsUcyMYbZBeOTYCv`FfDySUM=nPV<>C7=*1n{_fCQl5%7;B-9W5JewUufMx&PjUI|!NGw^TQmJ&Pa`GsYtu2DU<5n_g#J2iCr1q~7yX z?`1cZLaq_=x}t?5?Iln(+Q9W=?`+biq8k_H>Z3d9pmhl*USIS(LjacIwP zbFbC=HqKFfkw9GB@G4I^iez8|JVS7P~9U zwhE2-0MLj_qiIVu^&emcMzQy2nWXIs?ENkzO4*%cB*K;SGl*&;sV{}vv>^0+8JEc( zy^4Tlh)caA{56T+{#>Db#o&kDCs65)FJts9F0-xPQNjt_`f9!+0w7`;-(pZs68n;r zHsEz-A5MF((Dj_IZ*7YBQvh%F1eXhcjJO@AOeGvYmQ{3%(p{-70B>c=Ovj{9Ek=Mk zPWm*OoxzkCP+I;-TN7IZktTeM*NSbA@ou=Dg?xHoWM@7ESH_sf;78TelD z#w+XbTvb2>_Z#XxsXdiDX>C{;6d1z+MZm19v-R{Imx~Zab~LH;KUn}y6{l8t`lYdD zoILqL^s0Li!MEaz^Z7?Tznf7M`N4T-Lles2+a?XYTV1}rHds~nMKeX`q*uu z0(Id*Io}k(uc~R=jwWk5nw0YwAF*u)&6~^@k}TN{Qy*7D0-SgF%8`}H^Pa?~nm6Wb z4~`=ZSqPvNYN^o{cjn54Gzp|P!WcgkQV$RPAQxQ$=dfIMvE8nJt(WXFwdv%CfaT-W+vWG|Fb>(V|LHYoCvb`UX}6`Gt< zs>xW`u9ont|DxpDI9~=&7JcjmIb|dSI-gjh&#HpGdb#4g;rPv-QHGVLVmTpqXq#s#hAjRQfzY!2!tu zwAisn=+V6y<;80k>hg0rL?Zu-{$9lO9!vYZyyoD^ps}c1iSd=nL`;6oMb{A0$!?ZvsVij7qFtG<~5%+1c$kN4KOVRiB zyw@&h%Tzzm(N_N&;wi5`7OX2XuBf8s9M>19Uv6eT8nj80XJ$CAn8Lw7nxLQetdNZF zZ!k7UyHl^9<`LxGl~wtOWZ7MS;*Jld*3qEFN{8+0!s8%tUu6EhT|L}104>)Sy$Q&7 zXt7-y;`-Ak<5E4+*nmZ(Sp*|((uW&T?*$`^D1DZb>T6_=Pgf%u?a%t&-N3FPwkcNy zoC-p!GuKNJ)yqy8k%vA&}(nEW22$Uoty27ny31@BR1pafcE}f3GdPJW@z0c z0Je~4Pt67UmLD1A*56XS*!qm)&D6y?$z65Kd7>GkM9?xe?_(zswAhC-3z@`hr8rse&u7Ll5P3su z=A`aj2vjKZ^GyC?Z_TZvTB2KE@!<0LrQ}=-M_F6i0OP6D&D+!*0->#q7YgV+dm*1= zK3-p-(Q(T0sN{uJtfkG=u7rJ*`_|&LKD9t(`o0t$X}B>&-MzG#8lU6-sKv8$>y5n; zx^^psXQnOyNP&pgC(tX*ShSOBYW?`dx~Tyh zDx#4@=q#0o<_e#5;+myGQs|jydSRccc0I?o;$ z8f9hY)tVM@a~JR$eHSOQ{$o)qOl6dY$7rpx5!>62eJ7vQD2c;e5g_wEkKX`U6VEgR zcjmblgyF||WY?+vsb=C(a-yK$4>FiIPHU&Kes0z~79U=?ctU=8hRae^UH0(S{-n9fe`;L7lA8*E|?{5!c_B+H?#Wijcc-&P}n#D7LxCBARTiZ@5vXUa8 z4$0UUQ$nfsk5c0i7FVwraR5_qz}lHU1{Ya-Yd=l2U%V_@#S!V`H3xyE&UdAOB0L}>!(d)yn+9_&)&dwep?PLv!rB4^j z$554CtqRKe+K}=`#*b0?vt*;g(EDJ@s0SibQ~NSVd{+fSfpCokJy`8v(b&i?UGtj# zl=jAQ8?!r^zoi-xe5^}7Wh+i|dT9``~$_S$%RoPtKgQ_pR%*!zX4S zQH=_OJ2>pww9syI#>i}z99#?_Uf9g!voS(?6*pTyxA{|vXhtaL`6K5X@|TkG-ejD6 z9V_0{Qz)}5ec@ugirBivxl?zkp$+w^W?Sj9Y@tRMS4k=Hr$`nzYaZ zEYeSsdgs;&Et#*Lg?CV$kc7R7Icfj-*d74In1Gu%=G(yLL{9r8|KqG?xaa z@B6yBO9b!Qb|5k^Cj}Cib}L=m1eaNnGwUstWjzhJ{n7BGWq1GxIUR`1r^r;BafQA+*~=k@qQhwA{6tK;lZk!c{7#Hnh&OM&9N+YsQyC{wz@3 zYE7PxPm>QgAJC}=ClnFT6L90(U@i=nqdI9jyiSM`-S=?V+p7VZK~|HbuOh|Bmnp|i zj{O{xaww#BKo5vzVK+Xa^)`=6J-c#TXVFD1a;mQXJ3{&Fwh@TsQ}%)xLrzX z=JmXtbjLC}949E$yh0|BNK?Bqq0Hc(1*e_Mom^XdMwcna0Q|)t4~zC`)qi%7XP~;a zo%_Rv17DCvvQYY)kTRh{O_n2PJ2%g;E%?qOZHNDun+*KOv^x#jj|{~pHC>d#Z*%em zT<$-0Y2LgBs*ko{X*gn}B8U&EIiym znJ$s;2l=YB$eiK2B}I4U*6&jcSygI)qr?{PzgRmbJdH3*k`rq0!?qHWXb(@zhiqCP z_PtY-=mGheTge zP1NR2!ZAs>?YK=!w(j3NgOEuEaycZoF1)cl!?J7GPj3@s&)J@Nlfv;m58}fxu6FNgic??MA}zCg}G(^SXz&yB~bbKyphY- zm58&=G(%!a=+b43KPr>=k(Rix;GJGUPL0;~KO}+}Wn`NfxP)%9FB{+wzy>Pflr%gw zF5Al@xE7zPd4HZh-9F}}+q71%dppv-Ja-XvA#|+stgW8vk>yArBf*&?O(Lt`jMqss zqa=S@7LevRdok33nGkd42GeF4p)=XklBv!<7D zDH{04FLILkoLbq?I(Lvq<^3_t7H5OmXvzgnaZS*X1CejPuX<0B1bbMn}d;7Wl=kqrGgOYJgRJqZujV$y69gLKc3yDRS}c@g<6d? zjQ2IEO_eydSpV(^4sPVy4B(7;C$V>YK#%Ot6{Qn zYL%e-*s$DO?58f38ZJ;1eO-q_q#Hu`r1E7up;8O(RM(qfYQW?|i2jUVJ86CN!$(e( zH60R!=5qk*9haogS4r+9P@o$GJ*{gqU1|bjy`73RPuK%x#panIK(V}JIk4A%>sPdR>{Lgj}NfD{^!`<+*Ws3%$SyIQ7%Mt-O8 zRSn*H;uAqpViP5UPYcpBuTJ1EY(w}J_ac0PP0k1boHBu1ULp$P+?<25>5<~R?ZPDf z%%D4%t47+S)?psfHm5rryAY^jT&S;gG@BzA z)oUNPv-i)C@Lu8?xZIA#OfwRR3^UhtID~ZY-mM=T#>pI;LN`7p&^O>nr=fMDzb|?E; zmy*7MM zeGQC9Dsp>_bx;NRu;3WTG6c3a17uTs-qFK(!n~#Y%_<1=t5QMI4P!||@5TH2`;{<` z*Xb^}gRnrkQGK}*1rLh~z(r!MF8k_2?m(7pb=Zk(?)UzN@_y=N)CaXyk0CX6D3|SZ zB12D%=f{6M7F_yoj|3rSlr|!cvKEY`6<@u|Tfj&~ioTjA9km%Uq->)NG!bI`>~G4m zBe)+Ns;a_p|8%Z>F|#c+V}6Y{1F$Pkok_>aad}GZLqH@pFaq0Up9hk!8N52J0!qg! z)}>Fyz`A_B72;Z##SKZrr6B6wU%LpQ$HfL9F{ik1m1_!c)a4Suk%&eS?I-e^O0Oa(|rBm z^UOcc%onGb6SX`L=+dDQ#KHhNn?qB|v5`wR+s%PmAndAwo*|(={H&t5%#<4qs96S< zOyOom!5OLJ)nD9n2+%&9Kk7cKeRSr%6Dp@zy_{}Ow847}z2kMK;I@+p0;Vc15V|ob z1Vd3vxGa4dwv|)C;B9t;kDQ&Pc(Ejhv!Xr?W}dNcwwQ|IM4V77AmMAOmvaHmgh$~L6`dw^^bL19 zl4>istF~2kM(a@xNEG9nLt0+4XO25vMw|%uCf9DXeL=}o0KPUPfD=8y#$1sTw8%gE z5bNJzJ6^qW{O%z$P?|1NAM34$gbOhth?5=UD?1{KfU7z35M;RlOYr27p)g=k{*(WS@lC;?ljF^(~Jhz6@ zWtAIGJ01ld*zp{AaDnmyR+bH{Q+!8cqsHP z_?wk^JMF_W^5SjMu}sLrOL+;qSxGQVa{e+}@kI?>$#@TV7*$Wo3e86+87@nJXAe&O z+>)%MY5H&f^RemqL;WmmE@)nHNQV#_70|Kv_OKt4zcPA28FRLnC_$Bfkm0HI6RmT7 zwibe9`zCcbj-Y}GVY>Zq6UeDv_F|zOZ2u)TBK;7sw^m%AyFbT!gkJBCh~9o;g4&qj z$ExA(au_7sjG;Xqr_w955*&`iyEKNrjYrJ2#0v|8Po6#Ya<#ac&(fyQ6i_x z%!|6d-seiqwL#%u_?``O<*liTY@EjAWW*NWlTD%Nr}z_N)z$c&w{r0wzbOsB#FFUW z>*rh?$VCm#h9hBO^>ZBP)|;$1uAWNClE7@?1X3E1=)uYhO*PR)To}veLV=RdXJu0F z6r~y)ZOE>e;^wa-t;dr^H#s)1<^0O3%m+)f8eF8ihs%!X`?|nV(O)_EW2OpV8$Mvz zJFKK&jsMM)45wM1EmOexZ@~vOMvZBTU0_#(&EGxy zY_mZHJr%Isesib(U##zaYq!~;W~NQqkFr8)8eqxdCwAL}5|?`=jyp9`A{6ac0$BAz zQ3;M!U&4^P??*3IL*mWf+Q5UYaKnQ+Fw)VsFX1rxys_wg3Xl*}%ZPIjJN(A$KUIO1FO48rKdHJC4# zEK=vr;pq4B$zLw0dvEv3f7h8lnyq(+-R0F(^Y$K^1HVoj+(b#&xIFtGx>5XPzniQ2 z-vXaRI&iZyY0jHO*qYb*WzLgTrJ^&Tuo2+0CtZbbDMzSXg50 z73NeKF3g+jUV_2qEdaA42gPWj=VwLpTI&fZ7mPPMUJfVyMOv3%vm?_nolZiD{K>rQ z8Q^l^w_CX~c>Hm}hc*pm@|TcR=t{bX5lXE6I}HF2WaEg*hwzLPM3cPg~HlwZLh;e9?U&aZ~2ZpYaS&y(J&&L=MeuoYp0F!!eQ&~Q{Z)SLB>~lN1el3%z6f9<*RaE z$>Qr2@v|QiqtH3o*&VC`t*>%1=NJ=$E>FN_bl11>XJ3xfYGgRi7Si4Cz%G7{U^-Wz zT7x&y8R25sl#Et>|EgA&&x<0xV=*&7DR2YZ2DJWIRkCGss=>Q(+?*i_SP4BQT&Ng-3FeSQ=^raorL8{Q~zB9npR zOCl9435rqfb*R4YY=N-WnFjq7=roX;hxoFcv(CTYpoLk*u?mp(-h2{F$L>>RAbV+l zihD_KHCH%8XNH6+Bv>fwAwRC;IFSDpkl;9u;W>MyhMogwil4Sy@zzW?K$6+dgmBbI zhp{>1jTgfqENZ9+aadvst87+VE`3l}7!5zIN&tuK9+0Hk%-j9z8<2e{>2(Ul3Cp3& z9UrbR|HYNl8ZP^H*eJ&krHt(H&b-oZ{VybB_*Q#Jta>lEWBf~m3KND13be4p=4w~` zF{51WHI_^_(L-y4#t&&t3|`Xkd#I)=;*$UvnC34aTZpVO1iU45k|HH~-~ET|ZE8mA zisE7}#7@r_tX?d!gnMF`+7PSuiRO%;x2YuaU%Ji9Oy$l6#Barh4*XXEEj zew7{Erd7GJ7e7Su>3%kPl9o4)A_!%{C>rRgz|E*lt?yGRlGt&mUWWUOslDx1pY<&;?wt5FNv&ikCd8_tOfrp%iKow+-uAnb(3vhws z!o`#qS&=y4pvj0!F_tfh-o(W{CLqsUx3)BoHwt$hYTb;zW)G3#cwT(N#k0pZg$#_ zmMNHBt)l7^?6Pnn)5q4|agm`l?iL-(TxXFS&c+_!^evI+luN5Y%Tb9@QXbqvJs;Sg zJ510LCI$9-*>%TBH}iCHpACk;_bBWOV=ma4RcL2>!_(D?BNcuVUaTX}t9QNSh!rW` zqr-|74_|apU#+lqyz1!{nG*GVXh|5B)l@8RS3eHz@gY9O>g214p_X*pECO5QUac-Q znjK!Fn9y$bvx^gEo~`lQJ@MQHJ{J})aA6yS$EK8dVI3`2%SjWRIIhI()$*wdR>mR+ zEWSv&Pq6YUrekvS*rO!?ZLozQRF}Q3gv;^#aD+F>2dTQtM1I&oX!{)+%Q+TU1g$=j zvKdIq?tZz!Uu@Df`yokCzv`<#F^%-JrIG(#rvXOXz)Jq4+&`Ke>U zi8i#{pnya^eM9x$CcMcmVb1LBNk^XPR-9Y9bH)xmx~qM~@%zCmEr$mHoFi^-9cuT2 zf@hAiicON~(~-`E$4<=>Cexx;<{3Xi2?&)~3j@Ty;Uat9nvag4?bS!81DumYVq^2l z8+1&4Jp)WbP0M9;C4RUURAbZKyXNv@Aa;?4*7Bq6IkF6&i2?UrN_Q2DpK^NN4b0bH zfTZ8HFP&U$29P2@UY6Wq9sVu4^gCC5@XD=2`36cuvE9jbqV|~Zt-vNJ>pjp+Gnh4i z6@SLSgd&(RMWisa^-gCPG06D-A%)ba5;k_H{^cHTHu!Az8bc%$uw>I-$oi?>V)reHM3XB$$#`8gN!9xLm^jQ#{9Sc2>zT0YsWQL!_X*`c>_es?QJiPOfRcBB zk@;3mqh&I)XE4Ntgg03h;An`J5f;!`tn-2P@E@~NwyR;cz%8hqpVixaR)kP$;CB0` zZuLDz<@uAPxBiX6bg{lC&ORw!;N;{@52i4JR3K!CrgI34=&GWE%~+|DNufW zx%h#ca5{76`~9c|Ky~k879YR#x9^tBM4p76MAldeD#=kc?+OLsVP*@Bh6C9Y4C#l= zQ&;!2?Pv}Vq*9ZAN^D6JZqok5TjgHK0)1G&>lXF+aNAWbsO$lBem3mhyQ~7C%PPDzZI=f>*xY*wyIktNcbACD6E0T@5ObX}`1$Yw?{fXY$^)L)H1j&%*>PkH zD|Ge3k4q!i1H1!#wSzH1W}Zx5n!_XV7%n>b>8uGY4z>Zai=6;Gq!ZhvD$c`~|9D2z z;Ue16I}eZWGq77GJcTsoVwldiIVMiNmSuotiavCa#+Nxx?7^FL0tB<;ZpiG}JVCmr z)$p(_gOK!Ta5Cmq^T`JNVkA;RMNw8t0^ts9zlUP#v|#TNwYl0o10m7&>&$Kn+y8GZ+pv5Pq1+Y{4NPtfiH7f<1#$y`DwxOt}epTW}OyOuAM~Dw2UcsEZG*F zEFK#U&DigN?80GEzq5tIHpVpdn33gJgi4YGGvN6>1L%O_3s zBBG0K$o;ZN=W(yRK7A$l{@f{DJDy1NjY?yNkh3Pf&>G|gOsuU!(jOkWCf#7M>qAnk zZob>J?YQlMJ9kXvwks9anOwb=zc_DQpFZW>p<~iXX)(WMzH9oEF)6RSw6r7m=^Ed5 z9R}iw>xsV^i1`LQv_NVn`?a<&&{Dv6D5S1-R$d^0j@W>^x}Mj}>1(gK+8tVQx5TKa zxo~W|8s%BMV7@U|ZR%8e=YEs>XJFmVBeB`E#07B6;vq>r^7OeQ5;V#-I~QI@pFc0> zL*;BdRMG!rF?V_Dt2H3!qD9wu$I|K5Ec( zcAjI>L2M}iI(O;^U;NqJ70pw7(O2d4qeafm3!A3g0HNn z5>HIB>P#>a*wyvcst3HOa&dFKHZhW7*QTz6R(ZY-l?*=8R|ULV2j^sj7tHDZFtN5k z$ldH=v!vq|&Dfq3*kk)0rSR^vHD(lfTEf7zb#UkBW57`i;v9FsjS=~);(ouw#Q*z3*b;Tp^l};RtHT2ps7-SfFU7_>b4x^ zMbXsSdskO#wHhLaamE|zx=*Ktax%sOb7eYvbAQL*YX-4BevqcIbDaI)xjC8C{X5xo zLd*s@C46(#v_2gsLR4nDWSJ9AhoAkc_i1ml3%v-FURr#3rbctSqxbg&*br<$ zsK3(c9?f4Iy8d*UsU%xJV+RvtwdEWR5tn$PnwAj#smJ>#BelNiiRyXqi~5h~ zoRG>^KZHjQLs-l!T7Z_h80{&akkeh*><-`aq-t&hGg7z7(p`dO_T=zPr zd9n&@CZ*?BU{N}u{cKBKq*M?fsl~M$dEI(gj~wlP=ye51kxx2Kr8K%3e5G*T=sJbo z=P=rgsftS{;^iE9tfxZjB`Oll){fh%vu3tje{~cV0_3D>1Aiao?#BxCZ4Q|po-s+; z7i%X7XgcnkNucnY=;wW<+I_J0bJ&Ll97H6`1yRE1m^^1vC2alg_~cHdo=Qv$vmpe~ z`Zi;DoSP@J;GOig^nH@{kEsqK!#u3_pCb2b;DK00$k98YaL-tMjgV`oWBsz4O(GFB zlX%W=bF$oGabGE;=7L96V0rI7Ax?{RChCpb0&uc%cbgHlxr+$)%PQJvRX-poC>_-` zg>B;6+#vUYk*3DBE}b!w<(7XHbs6L9nlci4CaE^?bW4Qf|KjZ{qvF`MZi565njpbF z1cyN5gy8NT+(U5pV8PvkTN2#eJ-EBOyEHWRid;G8zIX5a&iCgH#^}*>QFPU=wb$Bv ztvTmHz@8gOyb%%7{8B%EAuJN+LFyByr<>X-c zw^5)f{#K<8w7|aFGj4T&Y{%=0HnQ16C&j)BaoG0q5One|8(7IJMzq^A?qtSp19jtB zdRHG_?N>ZvH5=H}6eavFZavOUF^)|H;kshbY;k>l^aUz+LEQ}4eZF`PQ^m`XZ64&c zmbS9}D6YUpllkb@Zn+u77{Zs->vep#Oe3lw9D4OMmEsR4T50VDH;*+gIcRRt@3Z&_oBF7Zm}B4i)ogvNI0JFWy#c8KQLJ-ILa6-Lbpi886&Z zH1K|Dc3YdRH32IwIH*2NC91zor5HK^qU+4mvJR7_!SHtg-odDOJ*XNJc{T&y`yK2; zHQ&;xmA@Q{G08|*FLZTM24sxgW3XA zXYjnPZU%ABGB5d}FZj{Uv1%ln9<(&AY6Vyw&L0h|@If}d7EG(b9j$E8I&!{G}^ zcoP7O?wd79-!#*^{N`Y55_4Ncbi`@yZUW=c0&ep)JKOy}$nt66oW<%v-=k-)%@@AQ zL$wxILRvG4Spm%1oV^z{J8m5|HP;@4$$epL`QoRN!N0TfpGL^<;ZY{Rp~>74Xg|1{ z+ojOR5X9 zY7&c1_u=)_XigE}<8==-$P>9+Qc^4Xn1)dcVrvGrmh@o!6i&b?L@3NyG10Un_%%$y z2=c~**iGJHn`&%A#S>gD5;D&V=A+mP)INOKM|yPIobL`40^ho}c`K38| z$fKvGE&#`<&aTH%L2FUkbBNOu!`@ifOi`dhgF;$$iqMu5UGira0Xf? z#o7^JA2&65hOpr7Oo0Dl)NNi-Wdm;da=nuvnh|izivR8tL#+LfI0x#Z}E)pj5IBgM9SCH&AJf>*`72cK{LP;KduTS6YBt-AXfXb*_* zz#Dc24GvmE0}m2&Sik39yga=cD>TbSS%Qt5)oO6uua-uHy5Rjen7l{{QYt9C7f7wt zT~&~$PBrBp!4Ac}GVJu|0YG!?u=nV--hAkL)fV&BAzUn~BkWoCeL$LpE|Gh&?wrV5 z+9)**>g&GX-?pp{Cd`oZmWO)9w4G7$5&mDv)Dc4=70Zd+3j&%zH6!U9A}Q1SZuI6l zMp~goj}=G{lr+m$ooc>Ecq|1N*_mDr zH%3tgcs1O(w5!eL*(*P|;yxJXGc)wT-j#R?DjyC3QZ@yBM+z=S{=KfBLvM3`wD6F2 zI3wSes1y-nQN~UTWgXFJLXbGuL&purw}uTjlg&3)ynPj#T$O(?;$DSYU7Yr7DN8>< z_8GN6JhL?ov9y{Ra_pmU`OleItWT_G0pdV$#P)zhi{L`1)x`{sJuya$8z9s0#N&Cq zJCDE$0HArL&8h}b)5guDRs-r8O$}9+^mx~XjOz{)CY9`?jw@{{5nS#);jiCL=BwCN zI=tqC)czF4t8a2W@pjuN7P;K49G7`@fzLeE(G>c`ysaLuaHD3cv28FQW~?oM0Og)t zzIMbOVn(3%cj{pL}BfQ9IaDgc`P4JlxHL(%m%KdS;d4xhL_=w=YDMOcmNcX1PKGeZ3nSs5TCAcOu?M#SqtyXVP zikUOMwr`xBWWsC!1is5-RqfY|iJeSu2U@$N^x^ zV#4yeMN-Qq4|_o>oq67FtXDP>gWi(={-WY1Ii*p>w|A_2epoV*Kt*cz!q?jF%x`oI zvKpRs1i=3-H1mQIx4M5G%$zp8Q@J8tU`J{XC-%A%mu)sUYXoCO^sj5VnOlH`F!}V$ zo8;GTQ)<`oskquu-fDP2vj%On0R_W5F+;MAhT~=&hp{hP&uc=ViKVt?@)f!Ys?bTC zjvoD%MgX#&188qRmT$2Tr!tg~jUsCRVC$`e{KS8PM%mx%m2*VoCjk4fmsEE0o)Ynm zo$h19N}J0FvJdwd^h8YYfh+h|0P?s@s=8e^`wn*5@t)mc?_5Dd+BrZzMmvRS(e+|d zq$iAMZzIUL1E*V-pEZ%`2;g9*eFU@m9&_4l#rAMljmPw9Z)L@)=jbfDhB>4f9T;E) zEn zH3AlG1|fM=^U2zeB3}*EsHLMT@Xlo>?@1n@RcicP@%QnB^803K1b8*rWot&(--$mx z4x>xHXzs()2D@o(mdo7Ta222sIg!JQ6E9&{H~p|&-W=vgx9?4rNXt@~(xAao;&WLn~}=|^g} znl-NmJ(t_v=d#{^hawN{}!jOcq$Qg^!Pu&G6s1u-%&Gdo`f zHu}rPpI-*jS+y?xBOF6XUbGP-JB4!HvDAmTR+U zJ{g2&YSr$_i#jdlNyklY6L{rDR#ESX|}) zgS$Z>{##(-lU+{@NDSqmFSg`B>MiQdvf^>`C41__E8lqlMUc>|_?`1A1Ex*&*0Jb< z!5ZNtn*RDLak|ywchvw!_4;;VqvmuWs$o$y$}Grd0Iv~eWGqVBWBLF<&%?vN>W`gK}wQlsmE{4Hn9_f{nl&-#$rDWC4YNOrxyvC5*+q07O z+Dl9-q9CN^RjsVwx&^OzPqEx}R1b}t&XKr}rVCR_o~u;zwm6l((QFP*i!I7@ObjDn zmctJNxqCkN-J`YJm?K8eKNGlFfIt6o8s%>vS)Z5q5|T9O`S*ka)WB zn3-Z4-8D0t&=brJVVMy&!6wG_ zi1lqqULXZ8jn7oj`QQ?7pteboS!m6i0D>6g1kEm*Rpe@&*R38-{Q%YhE;|j0 zo5LnJl&hTcgBnZ+I-OnuDP%3^EyAq))FehPULP0sFH*EhmmSZhU&BDhtEWQV*M>ZX zr;@(kX6)px*b$Ge%y*V$RXUH3nn%5X_dD$3o16z}B3{2#c4DSf4*NTbh%XKdW{nq* z0ZKv9_T{|f$qPHRYgPayan^S+O^&Uvcc+igt((qTy>)77uA z&)%t8U;@ZRBp&Nez0`Y#SfOo9GWAlaWM9V3n2u;Y)%|l2nlZi#{RR^|_vZA=yE>@B*W7aMGg(;;!oGo4?LGXemIL(XhrhoRVb3a~bJhH1z5~{Z=yRN4e%YDvTrMjXV@F6BqOOQQrYA zgwgtL>$t%YG28%%lY6+2=sP?&Qe<%Y4I1hhBu5-gy9BQam?bG4V=ZySow#6B_4Z zl{e(!CFRVA^cVNxN^cNcOG{teSFkE^^4}h?8XOT*ye=+OLseAPzEd+w;EUF|?(pns zx7+|gLH9D-tQThI8=TCAm$Au$+-JF;5+)X>H6lYH10Hcojo!HI?v>aBS)O+X<4$XD zO#qfB`|)grD0sVUb90dwmcMfW+j{3HGi?gZ`=6Cbl9* z{=UsAS|J4}27BkQ{=FRa_o5U9w5Lxa1*fcu*h(o`>zS<@rAZ%owyd~3w~sG2Y89Yt z24KOmWRqR(ZWdw#f`~drp5eb>ptJTaq5R0J++@sK6wxqXvkm%_l>%Boh7QmGubajn z&;h~};}ZNw3GDZGij%z;(2jmTmPGFJtiK;4YCt?_qY$(u9zmU^!M3CzCSiWdve-l zW9Xj$35Un`{qBS`vdI9t&1rD%#YO6Q`^b?#>D}I8TLAJ#lbo-aN^7^~E-m&38vYQU6m~4hoZl1rg9w>9h|xaXU`; zB92a7p4s(iKjnOLA*1g)M)NN>_zyigYgp8uiYKL#iErI7bs_69%yiKYOCm<;%j zguT8kf6S?Ws+9eQH|2Q1$W^HOFxURebo<++`wy>Cz=IFpXw&)YXUv8IHa$z%hWqi~ z-PNDQffT?{P@p}K{kwyIJjy>E z{bdZ;&X6%;_sSpnihm|9{=*v~OkgOY;l2LF1OA7<{o5lpWCd);qypB!`p=W&zm50* zzo~<&+2&;U?&hzX!iqJu_0aAQYWiT&`%Cb$nYbOjNI)nBzlixeng7MpMI>McCPx}7 z3g2Hhb!-xAF(-Eu?`u|s|HWKSd7A4yP?!60dC79XZu-q`d^(O<(yA$%LSU{ z`mbJ?bq=7?cq=vx70Ult2f)zo$&Ain4Nv~68Lfy>TZ=tk=Jtf_ifs>H$NX0-{lgu# zVu-J#jLXVuG}2tlyq4e+`peH1J-$_s`dY@=ul+*~=U*Lf67K2T|J6~6e6I!!O8}*?-CrI0_t5H34p5y%9rw*&zs6F)Q=m!)Z~gaa`{$$OPtLr; zu$)QmuV0zoQ^bsaIrzVDF5*vC(QBxe@#pXU@9*26To@hAH}XHZ!hh=@|M}ss=qE=R z-P`{DuV0xTCE$+L6yAycW%G%od|HA~#`t^xbr+8CB#B#^V;cTf=c*WqBn94_J(z_$a|&H_P~^XcRAn4Hx{^uZ#|`n2zYI|Hagydy1?SRQ>6J=>A%c3TK0-rxVbK|CELLB_bYSRO7ne=IrcS#IfX_PXj@hJTFq zHG?mgL%@59i6FYNlN}yKAWBSK7tSm2kU(Ct(d!bN61!h%9dGv560pObForQQzomw& zgNQq41kcUh)R^{T!PZpm|1rFF%B+;Y+-}_R>h#LGwJ7@>2{M`l@xoOKK(8WfaB%%d z@%+HYNegzdU0|+jSI^2z1eOPcSPWhrEqB%6$b$>V00iz9S>9b(u*t(8g315MzP;&$ zSCX~T`#}x-!wP!(x?&`(2pjxm85_fjOstJuS$BYei1*%iFuFx=D69>oxO_V5B+Nwq zpXYHTG_==UlVtVbo!Zq;^1(b{6G<7VN$9P;vLQJkzvB|F1S=i^0yb11_ORWz`OkAG z>IfE-tURyj+tHt7hp&DACu1QAL7%BR7QH!`2D>;pVT+@|t&d=~YOVOo)G5gfIdo^f z_@%c7Kq$o}#DF~rZ1m6{e*HhVe}Z^el8dFg{_YMI%g~pI6dy8v<*CYDry{g^9V!4s zhd^terya0V8LOF@neBwUXzonH1eX^_Cux=6j;kYS;%{G^-o{s0e z!w)*mW(oWfr81us*F74&BC5|vJ?WnDx0;o402M<3D3hw^D;LFL5{eck2!|x}7i%>* zsMBigv4~QuySQxa+ikRZ``U_^k=dSh`4fU!*C_!Jc4I%*LgHUm$?RTK#GV&HpO^0H zQarAb>iHOo_cj0u$o{^RTQPyq43@QS)*EImYOk)gaeC zuit3kS6e<2C)H7D1HCcWwgx~(oLx?HFN`8x-VU(J0?dz(czvH$*!{SS`{Rq!1b9}x zditNF-nUcis#9zrW@*OP%mm)b1=D?K{C5-qZ(PT>G-0g*<14U_Y|GFeiBX?i6# zC1rFQ^{Mzt1|Q~;AaY!WTllTECMnf%=%S(wr+RULy@Qma0o&*4wkoS)(jFqW&xJCE z$mh5v1}R&)wwrxCk;P@iVn!cqbuU|j{mvpi~HDR3|FLWN&t4HRUlP3;;w)+9LSqJI+wi;iu zvIKJVlNa9NvjzE|pj2mBayK?CUd^n#?_Z1 zxBZ|aanG$3zHGandiDTM{MMG?^r0IFGX}{EltLQvN>FJ)p5)D4a4=V(IxL5{> zBpvztRR!2ojhJu65PT;uA{-8ish6oDh_m0`N&5I~wRg&?8)ZQ5G5i@;gMN1?0hvFt)=P}O3@%W6#xNSt*23T} z@a}FiK}USi)ybus@DNG`W8E;Iaqk@#<7K0U(pRmZq+Ed8MvX?gZ&JCDWG{dwFu6`kA|^TRl)WzIDkB71E7Sx9$k=`B~@pB-5+m|8VNbh_#n6{N^ zK5|rHX*v?RQy?_k>901+Ki)1s8s6Cr8}9iyu7jf&*MGs-jzIBMo_s#%dZKKCG_I&yHv$O0Tn`yoo-<7Gk znT=P)6D>W@_MU!@ApFAt#QFB8V-{W@mESBtZ;qB!_CY`C%*Mokbg!q+qdM{!l!j-Y zpvFn~c>|g-8xKoOZn;JwTW34ukB87`!P8pO#OMUeL5qkYV!*CIMBjHe^P%*H_?bTC8J+Ap4ewTc6nD3TOy%;-x=7vTi6`!6@z=h(Q)?Uaz z_|_V{?#*;s*RJsbKIZh+?r*R`l$u6h+wCrcMWEoHG0Llf{NV!vaz)ms(el&*omwG= zmcpC?XCwWO0w|dE8AEM|@aZ1a%8Y5Prz3Kv=Qe%P#yVygs<0J-xq#7iY{7gSw2SAe z=WBS2P-RuS%YJjStz-YvTIj}?{wSz69?`~KTC0J@c}L9S6B0J*i=Amj%OXu$vN<5V zx`J>Q)ptccvJT58?-fU*qaGn!+_aT17vjq#ps_l)=fklc8^T$kEyMJ3v0<=D!&!w5eQ1Pr3co&8WGJ)VK-w7e z?>$^ADf!9dv8G{98YAY_V>tKg?oE+=zh%Jy)Tqn->3t@><_4jlXpPn;yKR}9lMb}W zvXyl1WNe@dS|MzAnN_!)DDH=40bjV?VE}^{$70kcg4>pI3xM!qQC5|5RjFvsSyQ5U>^&$mdNM{0KTih#BEN?{S|jz^_jFQNNq6L(mFgO z_g}hr7;FT3E3ALb$9ti!bMi(yFzu3o;JnT z@;yNFpf51WVAA7s-JVzZVy*oQ5{iR%)5>KaR&&9FTn-)R4;J{<{v6#H^* zX~XT_pH^jzZ@QWbK5Km>9v zVE(o;pD4LiVYA*~PaqNZmi}E#m&&>OS;7#Qg7voeTc2KV|sLr~d^{%!dS&FKcKc?3(jdQzo zQ?v3Vqp#`aw|0;p`(ZV~2U=Lwnw)*ruP7if+SBj)D=SfN-QKO@S6nZ<9mCk{O(gUw zTvpFpzPdT*CD5zdt6t7fdvEG@IhO}0o-lxm6)%Y#mh%{T@YnITv-}Qxa^r+O>dJGo z>U-FAweIgDmg8QjuKL_eJ}Y4>c{rr_^zCv|BjK_dg2mj< z{N^&dF&u!+It^Jb2$i&hGX+IEPqC;sJ+7;`mI!7mW;N_rJl%nYfs@7aJoCz@Fvy*6 zMEep?MWFgIC`t|y&$<=i9o(Q*TT6?DKA;?-@!hFu2MCf$#t)DsG^Xn*DVu?+0CYy^QD7N@p(h*S9r4b9=EArL!Rm7&chlL#0K?l z2k&V_3g@?@;@g9+w-$Yn*ZpmW^y+YF2G-Ll)L;+qp0QBj>#3fdP4^3z)!Eb22f;l& zcNM;2^!SZT`18PdtNyGCit7o7Y1g7Cd1ArFl!YVirsa$yyvnu{h2%xj?*BBvPHOseaqIg44_gUoW1aZY@t0tCjAVic%Qr}nv z8*i5a^3%r}Om--7SQ`z3)`<#sd=R6ZAIOEVzMeFCKKe!&@>F4+J=_W}I7}1fKYp#A zH}~n{+~YCc+ODN`>-0cAMrwI{Xs9LMKZIT-wrAO=q~q1Q44J)b05?*i67rEr@M`ef z9gNJcu%Eve$SyMOzd4epC~0@1yFXCd^KPDI6%VnJ3UrXzLHk*K zf6f@Yzp809NU^^HgOOF{pcw;#Ox8Dh+FmG5cIPDATVLRe9CasK-?62mx|E#l6%!IE zxL)EFIwH!7oS5dsq@ILby|)inpUp{pJ%d#OADWta5kL7LAifIW^=~U zN7r4-wXkVWo~N<%rkBph5HZ(X(5C`Mal0VEVxS-aCG`5>V;_Eyo?OSXPEhFb%EK-*a}>)F&Os%fW&s%QnL= zjf!+e@w+;BL7m4Lm679D{ z8=yNAH>rG8=u6{m2u3SDUpWsT_CyCx!%FvX{P=mMRG{dsW_61DW0Pb&jZm3>*Cb2r zhk!Q`upw5VqiW+xtR}I~L1$}P&xjg`oLAZc`S|G?Pl?UCLwEBTZNZC9AwZzdHv;wA z^tv5~y)zc71CTaSU)S$6w#3O7Gfk9=r!dYJ=vPyTVqj&`c=DEZSq*<>4ABCq1B~Qd zgSiTm8h!7$u-9xV%#+MpmN>(AA&fFX9RX;_p_d=&={r_m^QjhpL@b8sg8-?7PJrg| zr6j+3;mxv9^Yt3CH|`MvzIwLEi_<}nPXAf>PoBV4O9!?Rm0$dTLQ?#;{hLpGsmn*_ zODHZ~4rWae<}+~^7I@vWPc0KNGKAFLuX2<8eA(>JadSf>pQR)$TxoPJ zlI1on*Qm?giEAqCZbiXOi-e*bmdvCKbcz%1_K5W+4uf5-P!8#?u2@?X!mV+rcSd*( z1NQ<4eSW~N?PyoAC+M*^U__A(|EIE~&NqAvWGS>yW(l7XFVR5l)&@VY;NnQ=$cmOr z)p@3i(q9uy?w_h}l|}3Kb;@I);n6u^eHe5IY(QLXsHnVtF3Q@b{|QLgRl}5Wr4z%| zqx_Inyk4p|i+i2ewX$fCD;bP8-py(8eeZplXu2Gb&~ez^_KrHdx0=v*E^>aorwTw4-{B%n zwpSLKAW5d=369V6C2mX3k`opqABx;Ot##xZRM_U-N3liLzi^HJ6vw9&$!OM%mK-J9 z)`3ph3#?n7wZjSSRWuRlROoW8F*$L+#p$; znoGr9BWwX*p*AR2_v^pXA|b&^@Q@^66#`MqLl5tJ`-)QRfZzLxv_O5Jpc3YwjE&f) zO2aq)z#Xz}zxcyEkc_gg08#01u2RkdK*>$@`w}(6{S z@f>7iWN=rPO1$_m6ITl;GAHCKzYX@jub;o7ms$z{w@_-UH^`&lb212UMkFwj!(AR{ z&Y5dfPY*A71O)nrB;=dy8izI1qDVbH)nNSMBR`VmCn+lyk8I^9$+f01y^{Zo?5XwWs;*=SS$6%~Ap=w4qb zXQGlGRiL{yyoF5jjIdHlE`y6W>i{&8U!aZv_U-6vGn)JrUn&O3jJg5D^{;!)-k zPWu^L=Hsq}iIuE6UaH!Q$m7|vX8ep$wRv5;mMyUJv5wZ)bavH*)hoP^6JCjWZc%CX!qmXKSGR!j*{cw2dZnxyHe#Wza6<0hvIG2ye1K39LMr$H%d*IYS-Zk| zm8{D7a!*;mH=1(qc4yQ;=(|)>`z&*>#Zb!Z*el4;3CsQSTW?|4qXn0VETBJia2&1b z&Lkoq(7B6juF|X^9$@PX>+g;wbOBr~%nQBo@6y!ys$LR2r$YX)M!Hab8&TwaYJH?zn+H9z!C zU}!ra;?=mxkfI86Lv>63nR!kr?$3`!GR{nJVeD|+Eio0TB++|@CN~DC2M0@K0*3J_ z6I^vgf@xLz?D+|7jVGjf&-R*cZ3APdWoQZrI-egi3L_)TiOIBs^eZQeFJpH=D=tyI z^ICmeudoRj0&#D&kwcjeWl^~v4Hy2CjC!E4axuig8D_hSB zEM0yS(@FpYDx_>ZJryBDpD@jWa<0bwl@0C|o5O2_z?s&w4ny|OVmk5#-IU}Er=w;o+;eq$x;Rt6xQeN+ zphB4KGjyxMV5!j4trt?ctCiPEG)|X^Yrjo^-(qJPKp(%reRB__RFsM<%}UT2hmyYY z(CfAdn04i8b$XG}r;uoTDC`^%%E_{|Mm|JUsJw!8xtkE+y+$@y01}K_^hy@u)t{Y$ zkB9oIUGVfPkIsfaM>sofO#8~c;E%crD(aJm%;_#j9(R(qzWctx zg)QGyJ+jTY8(?-6qA4LTa%URT0>M&>Ob=2qTMIbq&>aH!*aFQaX zp^T^)#;FX8P5(+7D+_!)4P;RKK8Q&!Dywum-@N%s89Wj3_|Q6NU&JLn!5bl*c;{48 zdBykkmajBl#Ag!Bg||I91i`nM9KZZVTSE7?;?6GOBKkb};`$jQy%>7Qs16i4XceE# zYWgS>$53iWKy$S0o>jtKl+$ybQi>{nI>V^5EUP30Z(nq?s;e{X+hjpt$%fx6>Q(Ng zzsDG0o={EMAG>Iyi@G6}z?A-+K#UO{HJQR*bgi33)y)#S4d+!#zqjNB+M|f`#bp&) z{?K)4{M_3JYqV2-z}JR;PgXhaY&@98+{=)B$byoeas0(VgjkSzkuOtAh?v*{MgMV@ zcU<$=wpfHE;UyrY;aq}`#FRwuYV*}}qB6`?#3+tPnrRoUC(IA=H6S92gZ^Svc95Z| z;3N4yGY|G^W8}zHiL*=d9&z6D$NSN3gti*@x-tFhUve`98!ZQ8{bZ16=lGUNIi#I< z+>?5ghHZF9yOk9^qaZX{7BO9vHjH}^@91?g_c40Eb{h4`0!z#+_~N|SFy0L8aWT`& zBGrLXg~V{2O6OB!u}pa@mI>8fmK_lMR+bNg0Tt}hK~*%#&4#_kbH42+)YP%_09sO1 zWpI;kA`bvvoKU8*&uA2;-{KhESlAWy7=XF4SMjk>j0e7z1h;jfS1xj`E|zJ8!L33< z%)P~rbUZJmo%InQwH3Kw3OoHDOH0QC^YB?7{vTj1^YnM|POgjJuMhD*HQ!PEE;XKruB_Bpg%Ce->=#6;`B|wsZVx1l zPtm(Zsx0?}6JmKlE>yOkkJf56m~7%2qYiH)$~)|gkghT}CtEeEyBhSrE%HNqQ(*Ga ze!ob)GeZEzsD&xDd@T)Rm3AR_9a8?G8#0i@+Rrz@W;^gUY-b*H!M8`{4nAJi*+UN~ z%fpOmTd%x$9b&YjYQVx$IHuQyg5lN^K4J)#<>Sl>$6B&vHA2Gs@IF@Lz&=#|LnOcT z%9pn|_peU)sc)-ABZyw4>i}_57>Iz%MEBdGH%m(;Iyx>WKPJi4ek=nD%Pj1WvXcg_ z?obdS0r+xShnSB_O|gTEh(@#e^VuDva=WcB+=~g8l>V7AgB+7(r%9_Fv_vuMy@|Y1 zQnKRNFHkU>xAMQ-UbB2_cLI7|bpg5a$*aWx%l{L+ykDq zlVRw;9Fs&g5w5L#v5B6aNvm;Y!{&@oz>Lzm;|E~BjZ)|XTb(nBCt7~kF6U}Fb=cZn z|2<Ys(meKRyU8HDrT#)W>LEEx#zURo)a_S7jSP4pFFm@NW%(l_cwg`2PATAW1>J zddoG7@V$X0)fI7)!GqbF%kfT$O%Bxt$8ugzm``29(Yye$mfzQ(1@3JWNq*wdvqf!` zK?iiw6%E_imvE~R_=wOLbUCLip|Ay2dV+#zAFAFzAo_SS^zV+pYz7r;QZ=RKU~Hy2 zOBfxG)uoneSIuKHjFJ*{G%fGXYb&czxV|5hxhCRuj3%)EmKXeKt~Tvk;4@?$o)e~4 zkjK`?GM9cwN`2|ncN&N_Tb{o9neZ)gs?34v*{(+35O3JgOphOFV_n+~dr(0$xpQIc zQ_0g4W+cyUQZR>J#fz3s*|?b|`Kl%=eeq^XWqK2KDHeu5rL0nd+bZ-SHI4IDm>CdD zxWvc4EnNV`jJMXz*R9qTD3wt2MHhpjNN|-(R5*EpHvLfCM#>$i&0O3RP*d*n=CBG* z!o1HV;?h-Oc$t18F}J*bX#BK-4n8Sfvk~Z!@?u0o&Wt1j zzyT%%u&A|Z{;K!{tW@Vac0QYYh1sKMEKQC((AF(k5#GI)Vu^uYO;92DV(9gnRl5|Z z{hl;)WOq(j%{E?fZ?u)Ff26jPVk!wd-k;7hWHN%xhoF^T`{_B~k9fde=#9en)alwn zLkemGq(cCH&*y^~N!}4~J&^j^N1SfgMT`IqSAM4BeZ!kGhs)EV*BdSlC+27oI*Xby zNKfXU{xG&ScA!%gGyM|FCW-`U6>p6#6HV=>se0Vv(d=3k0c(2mF_RhGJps>fO4_co z@fu#{t>&AaV+YoxOdY>SffM~Nx3Gs?RLE;d!Hz_Syj0VA<=-6D!#(GbZQIXKZio@V zcykD>N!x)z@Myi_@F%vp)K%@l?#&lO@0=dUs_;HM8~Dc8Mn|>~KF`Z<&8QN0FSxLe z9z!~k@J2;w(t&`O2j>L=i^L0EJ(l72n{jd;+$4`-=jAvu)Tz}Tll0V=5%DexMODc& z`5<2tTplZCC}<+Xf!+X!>#8dX(TF6Ea*@K!Ez}%OrTPxMz7JcYZicpt^l4?A@vP_Z zGRfS>bLW)i9y0g+pxYB9pR(tvauLi>P*@zyR9~uL*Q&R|%tx9Poi0;aSq)-P7QgQ2 zDWY$Gn$qOaIjTQLHPc%X^gkRshIRGwyakw>sm+hfs}%9g{w|)5#|^!!9unuf4SWA3 zeRoaF$qVp$DwjXiTsuAYLdE_?oxRIagA-$}S1U_{c9W}X9AtvI8Tr-Q99U1gO(5ds z(G`rszm?C*FUqU__~m@m!U8LL!S%xyD4Hx}AK!5{+GzZ5S%8V~*?Po}@XL znrzDL-li*ew^OjjFEpXi5=)YgwUeOQZXBajJ3@Gk8#DCgfZX=*uA>Y=CRd9$&&j^?Cq5ga$y)NQHVSBT8)zEwXRZ6T7ZE>kbenB@MQvX%kHAs@&my**-Cun_#h{< z1TJ0t!r2$lMGE(-N|<90cJHf9X{o>Y&2;r~Y|o;s`s}^~1WOMA=y6#fv=s4w*n97A zxcatjG!dQXC89^~LG(@p(FsN{S`dWMdx<^>g6J)Th~9f=5G@kDk6y!!UPjp~_x(Qa z^X+4Q@A2H4Iw5nQ@kdr@(zthm7FZSkj6z=AP@8Q>6GTN&yFzVgRs$Be~pLomLDBg8&_UF zuXT)uYq+D226y)k71{5vYu}r1c(?Oi;((4{Ppj&)Q=-{4Ib1$dU3(@M}z5rQcZpxp9N^%X{_B$Gtv_ z0dqK%0b;Y0BK~SVfCcub)Pt&$Z*Eo(OJ0Cb<>23dgb0f`0% zTXGAO7}1}!yNS_t)$A%+)o>L`bwTlZ367VHt=sF0$O-Q(=`~^#hSVfZBSUgAEe&X> z1SHFD+M#%LSDv}{SoXaiGL_%SY1@jl++qazgeY>}efw)vO$Hg1d|OabzdedX1Zi}x z2+VGkrlcaFI9aR>TID}snlht$*a3iMy`KAo3Q~T4x>ix*zP%1}339(7Cq|%c;(>pK zQO=rI0Odz_B2py=G6zu0aDB`UyUAh2>tjxjLYOu*Up%dqaoFNXaniZS*z4R0OU>c{ ziUqkZ^czCvXx|gZj>$-_K6+tZs#n!(Zo8%-WCJQP2Jpc$%ZK07UrUb5spIy#lwb%X z=Q;xX0CmImCz80``M~Y0q$X_AV6EcyS#?W+s!}*YHe&r*RHub14fmf(w!U(WKKg( zClDKB%}eP>UVMaY>cr)nf+KB|-R2n(uzgSr2Y&*#zocUhfLVHBjW1=FT2lJlBSJ)d zjbL}}@QNtoX4_xO!p?m&tOb$FEV=v-NL znqO*_A9z8P@1r7PF2<)quX6&fieBT}zY*-(7eTkoFV-t5ni=!)Z|yz@XpFrlJAB_P z=IR~T6gRt@l}~XFsrw_V%7Agt#}oO(Sai=mKr)A=fvEcHw;M7d|m`fJ||s9K!v zP8PR1E;g4O9rlp5Bbv78IEK#&<5D6#c{^n9>=v)6tR@5C{!Q!oK_53deh50RKKK6N zJ5i;XEy)t(RDP_8YGocNcoMiN@T>arbp@)unTYMsC{AJ5Pet#f?@;?<%3xrNEevsd zkN-WCQVO4wq5kQDWhxGZZsk~yDzkN?sr#hfrky6-RZ0u=Fx{{$bR>wDr+XVMYidQ2 zJj99U7oeBGyg8ayETV!eQj{-sB57wlVd!HjeOiXY?|w*vrTRO@MZeJ`_&eTx8j5;< zUr!txXJ7tj)8?{|9`tyq;PpLXki7V{gsj7f3R;vNvGV9{*hzR8(u+hh`+6x*Piyf; zp)`nG60_&Tn;}Qa1x1o=4Hlbbs3D1tNN*L;*p47 zNuPCcxsv|-4jEJV(*mr!C;NU$c#h3t(h^C9^ML=+M}E>>!`wT}U;Cl)Y3K* z+PMx0g><+R^_vkebf7ltD?gXS5$0^uYz^ybxz`fa(DA>sRA+y%f2g*zO3aTx5~{)S zoD`pKz6tUiY~SK@JR-tDjFXt6RoJh*_Xrl``BK9jRx9p78o5H7ZydH;>;?NB3Zhp% z(g`%bskh#EGHr3{E&E}vHXzIaEEp~s>c;l9o?3!?g7ybc74vPaRkk248qve-wLste zPo%{Uw|ccrJoiKKIxMrE* z-~k9cNw|v)Xl9CJ#wXjOqYXRTV?J$YmDuE|Gtpr|e73{9j(2CLC+Pi?1L1`Eb^2B$ z#V3l^?JdJLFCZB%ymvyaUZSdp3=C8|Tu{FV$0*?#OSkg6Ihf$hEqg)#c;KL^U`T{5 z^>r8|i*2q@SB84t)*8EK<;`UU!W}4Ni0(rRVnc6*uNFOUo@fm8b8>jDk#C^mt9>3D zc!y=CjX&hsf}V1V%tPNZZIn1q|5?~2B-`yVvjmB%^LI&aCEXXzEYBDi7PkeZh&dm* z0aoo~r{i-CQ{J^t!y`*lJMCDHY4*9h-IuiBHJQ~Hkg;#&?@`uSZd6lSSn?)aMvoH4 z^Jz`O+YgYXVNYNB=!DLmf7minNkMFk`!&e5x~e$}kzO@zBc9Bf5ULTp@26>1W$J6y zHJ#xe-#ho3Gc)lZI-CtL&n_3@8HtR%XTVl$2yFmV14QpMHZc2BZFbY zRn7cOl-XZ@v$Cb49C_G$%i)bIK_Z_ReoU4&ObyU%@?Y%Je=haf%gXtae@>k*uAfL- zf$;+Vo;h}6tvrube!#tzd1jya{9TEvE02#7>7eiV0h2=Jl1)dAXiC6gF9*2@H5#G; z9zd;)TN(*;Q8@-vJn5JOMLKd% z1Vj6_5eMEMM{h87!VWOODe~S7gp#HQ-d@`Ls$}+ncK-bvUs*f(xQHdPA#vPKv+fJZgju zZI$dbu#0rhbaE18Jdgmkobd$q^!HyCU;ds~x$qzrKgBDU?3EZC$9dq-rr}p^jGgZI z;xB-~(UEF}mnKACxngK$xAI64H!6s8cq1+gSyqabKSQ{ui~Xk3O@tI36Z_%ja#;p4 zM$QR8yvIt!*i(W9JHVTcv$&LJdCX%F_E89=#eL#DQDkD$>E5m@&b&q<4d!^$h6$|x zdrAToMeYopMcgeLNjb$_L{FB%V#-ZR0?M|yQbrANO7s}=Tfw{|2^ ziwxiX!U@Pfy5HA#*gUP5z|=*4@PMb!_qZSuHDe>wT*3JL;D(y*BPPE+dwK{h>~Ocs z6lC0S_k-Y=>`8kHuaRQ!^DydR?VdKHT1B!afy?O@IzL|Q^m(CT7jd3DK}m@oxm{qj zwasqIf2Lz%W9ocKNUHaOfco2l>u#$@4v^w?V`G94K=BmRyz|LvfpCkfXkp9}NfWaD zOQ-adlGmjy>qd)u)vro%@jX~CJqH;LzIk}J@s4NxSL~_{^jhxBCx$LT zi=s|eS}IDqyxU^RTl81Z_HJAR_A*YXD$c19QeW8iyTUkF8PhD|(i^uew{?2$zG=k)5OWePQ=pa@nkafgx=poaIm>xny(CTu$0y;`Vw+4{meb^Ys_ zZAX)h>P$iPQpzQc=LV4|k&F4&g!76XQU_g*hE)40V=qTnqH?4T`J^3yKQH+$cd<^b z7nsO%?+MvYm#9Qzo$b4qeObnE{FPm>>Fl#(!_<-D^Z8F8E8Ba@s!w}$F@i35FtYZG z^d7@UIq2k{5O$Nwq~y`Yxmsh8(`E5Xb?B4TW%>Rd(^S*VdrC4jH^tzV-`!)VAUK&l zicD2vf1^2VIFmI|JwsZlccbqZHO_>X&lWJ==_!7^qWzR=QtgVS3jZK)MR<;bv6W0- za$|srExAlorb;cv+&>`oym^t?&{f9Af+E>Gp^5j;D~@}QsWhoe&q-l7)Xy_Y;`;Y>@!*c$ zC*o?!FSfZ{<)5&IPz2Tn4+bCNelQYcX+$VB3Y^5Bw^1I|2ekq@n<0ewD}J`h?}T>Y zfUA*!U6cRL)(-s#59XYx{w9-pqeWUBmMtdtRQvWuZH(>e`+PHM9~Q1Zzx+b{L^&E# zV|;T{QqlftC1ok;@SADk3U9|$v|^3*NPc1O*b4jtEgqEx2<193??W2LGlLjS^W}yRFGJydcFX4Hf)Q3BY7k%D6`Eh`as=qL8+RetXS+1eENTlKX=tdsox3}g*(s6q6 z4kHX>Xqs#QL#P-;57Fp8ie&MaDcDs@r|0*ZVrm$JO+L;5-Ma3SDRRGs3cewRv~D zoze{>YcJ!A*}UbarmY!^9}cSc{3sPwjdEo3V=@)t$bQn!fF2>@2_Ya%-j{JwepPC0 zM2=xk54vFhSRs!@!}T5+^RyVcev@CPZ!C|%^BqqzBjO%jdtVn`QV=d_-8iH8Smw<8 z@J;W}VVQ=Ydr7lv^sd#ItHnKG6&iEoOx)DwE8D={2KgA=)uj-Tgt7!Jb;iV{1rxP9 z!{M%`;EldX$Im|(=q=JV11h1ffkU?x=w&QodE^I|h4wNm;RL&h4E8@yMY;&q(F%{B9;dHn zWw;kNsn;T!*8oV;fjlqsEC`2v54^~Jrqa;y&5}(cTZc5P^m2{3eKEExCeaWMG(j6| zp2jZjWdGu=A*O%Kue+EKi{+CWhG00rgXNTgTjJA=3r2~b=(qb34uA&6ZssiG=9;r# zfO)2j->%H#3x(iFr*4NlvmI0uA|{ciZB=h%f$@j?Ivj(w9rrf`u|~74)pv&)7Z@pB zJT^YZ48(JF4DPY|8SNFA7=kD?TV4S~$e^d39;CCSs9)x9GD8twsM(pg6rbe@)?M-6 zo3si!L&*3wjL=8sb z-cJYAjkhH5!uJZLm0+as7JFB(JY3dmX`N;5$++;(i(icT(G@M0NfBQWft7r%1pc{x z)j1gFn{g>z4%F@JnX?Ikzr)0 z{q|H&82RWq{O<{PbAa< z8o}A$u ztP$lAM?W6zV+e!175OvImON;*G1EBVUg~9 z8&Go3>P@1iw04Qw3N?T4%C3}U-v^893!Kfb+8r%8aG*Q%sj5pM(Va~#l^x|KGM7C4 zMah%glD1#e?$!5?y<{~21&Y1ab?b-A?-Nz>9RHu6yX<5D)*C6{OqlCJO{nT5lL#V} z=xZSVpm$wgu1Sz55}VfDa5~s~nbGaZEJ`~>#%GFXBaiBcqItcH{}An6Ldd<87mn`ZP`2KMfk1dc0Z zIQmJ1QlFH5EJ&1C)M7FEWh}HBcl;JRE3>XGqfY3EKubv5d2ZymJ^->%T6XfGaxb|5f?&11vuXKK z!0tCQxjvddhhQ#zN_nZ}tw9;1Q7d+~wV?+bfSdF?s=;pC6AI)u%@}k5e~sVuEZU>R z+1SIX+0EjvU1Y+i7UJ|UW_=^g6@G(}l;L%3809eH@iZN|UuZHm4OgX|d;-k^e#nVA zKGh-2oP&XWbh+)v{#5bulsy`h0IyI_H4|>~l%GwOtK;N@)&faF2 zp%y?UX;7-EoWAibQ9gjF!1vzfEOT`7n6H9o`%8+rH}efEZgxNa+58%!H_ODotv#LR z5&T-4(T{dg@Krshg!?z~_`F(bCK9wmVppBh&NqDO2w$@earXwG`OkkNJ+A@BXE+80 zgwAFi^>9jIyVu7HR!#6c9TDP)swxk`deCZ#?p_YwD{B{?T60D@?29l}0P9%uuXQ|! z`ZVvpEdEDyLRlT1AOKZ=aTpi+#GV^}0o{Qoz_;pb%G+|b0p6;b($1}4jvpfhz%lX* z9`g>X0FCt{Z>0pg=gjtAGP|~Xtl&DiVa*b8Y?VnzfD}0^J5WmG;Pr)SW{NWMj1Ufi zwowH(qnqa7{GWbE&W#VRR%P zh8yXzFV9$ELEb)HYRCfyNX70@BEf_H1U>o4&jm{@{Q{U_C^+-nv526)gRq$xPPJ*?# z&AemplY?%iim4vIca{8jr{B3@4oo3@Gu3!-|JPup5qvd`z^)oem&NV@S-D5|{%I%a|fN>V8EJT!` zu*>xL(V;YfdZ?hyQS(&>mwJGrDE>DfS1tIi3|0-wn%-rTH1}1cKQ41bpO@ONW&6&rKr-FKWmG?^%CWjz~P5rr4OD<{-)luUC!HY*jl}b zB~Q;ZasbMW!@Q@S1>5sI&!q*5*d8hGT{V8YVdhgBg;LuRCv#8y6tGKVb?Vff%O(ed zsmmSF>Fxv>Fp@kGP3mVsGTU>hTD)-x+)DE-{x{Bi-384?!APw8KRNwS`!<$2;N76{ zw(>PTxO4x$#r)>aP|m_*S!ImIvheSo>z(tv;!G7r5@yNd704I?w^Uj$zMz(O6*vbc zqu)UR(_b~|S00N^oelWhUA8%bQ!!ZlEE70kw#Fib`BG+#>m>4gJL|B4-{+Q3xL1La zS@1;*hiJ>G)JSzO04Fk7ta8i|j59ryjF5X(m+{n&F5;4wu5vkk7Z@C;G}_1)tYuLK zRx+z_sR?VDP5INBmVmgzhf^4s&H%u-xeEl5iEa^j?@|-m$Dz-T327vChSMmIJgh{h ze)-YEd}yNvxqA*!CZ0>g0WUC||Ad|ZX@+b+$s6ky*T{yY-mwNJiH=Wa(pLScLx1Uf z)DepU3E$xcU(S`dL!l_VisJDbKLh~NzWnl8a8TB4IJGJ^qS~U@rRv?RKwZF0<%-T$ zqe;_6cnO3Vc2}-bP<&N>*NIER|1G5cVrs)kW;9cb2Y^s`;qjv!Yl0V5lts2VPw zi}vMFeIc|;dte4dy=|7~HGJKtoKm%{e!k&zx5>7kxasrx<71?7Z|i3e124^LPnuN* zD(-C2fq`MJe(IF^&4M-lD<)gdEhjh@A!dPU5|N-OwtMyP{!+k}l=krstDSbrkNxL+ z`i$Cr!y&--=1Go2qzaeE>dzWz&npiW&RUv_JrT&KF?H3bBwQIq6{yC{i&@O zqz_5&di4okUz{0XIYa;C9b$jy9X2;HzK0F)A4N06)@lm&(lnD4&6Rv2Obrp?AB~@U< zgcN-o#_cxxj4O_>iZCpo7$wPkP7C;n-ph709AMx!XgM(!p(@ev;SB1G69ps=DOm}Y z?zD)7A}9We8Ee--JF5^dD;&LVF))^-J$fHyJCk2Gb9t4GK6?MS$ujF+$i5o@OJ&^L z^x|1Z1!ElTTF}b;3WOcgllmJ8LztUA9~z=8LVyHCAE*M{kGr|qSE#pIl==W5d6akhqy+7>HW#1tVe4sy{XP)1X42)-{a-`Ls z@Urp6TZ~W;R46OnTp5#qiesuA@8M&MZUWW=FQ8@CW!)sV&>>gbw8<~#AIIwVrQ&t# zD@F2d3{nw(;{JEKD^z&=U*m~hU?oyor~~N6v`@y&9|GHtGRTTY06-AnoD%Q4{!5x= z4(A$)YFPa_>*;6dRC9HvK+Z6R z+{WBF^Q#`Lu^ngS_jF`$)de`p1tRG*#PjKSm6NNen{#+#RkOmHK-ud$z{4X2Ix|&U zNhjs=;dhRW&Yn5W&5|#I)7}-jQ29(mH%Gdy7eFvRIo)*QjWe~LHUsk=+kc=~Tf;1+ zli#!B+&Wa@a*z5inbNw?2NKy7&kquPhY$Q2K3`F`C+*1~g0CU+R3k=MIBhTV_t4I0-&kUfJ(FA4&o zREUVvb1W(Aj%>0Trx@itFY=iBB2JI_#+>%Un>rd#AdDCp9Px7`XI~gc3fk-Q;u&GLlpxW3PiCC!L|Jg)qPYx@*q`zSE&#M-qp&a$?h4G4ecrWZddg!;?kaLwpyej@ zJ+v!Yea`?J=3A#);Sgx@%wb8Rton5UW|w?4JJCVC$`CiGc$THHTcNt|z9%(yC0D-# z6|D3+j81onB71K_zo|GJgAaB$K>8wzjgBwT%+>$S;yrC5-tXrO?+EW>2WMw(MEwQP z?#Kxlckoc_<1`M}sZ8m)7792*qu<2p9+({UJ&@4Zm&RuNo$+tuih=r528!+Hd}rlUfTMh!zJD4|j0k zDCjw5tyJaH&}bXUXfRlk%|f0Ta}|$*r{qVG=jhh+@xIec4=$VKz>LpXveuvvJuWl@ z>dWuOTFYkY{Zj@hD}c_oi9m0M%9;`s7*Ejax4d~ES85XRvook}L6+0 z`Obv$SKzO`pKOawJkOqfX%Lh-6)HZAS@aywMkW&Yjf-gBP40lWe&GjI~YM3()UdKs3h z60DsWDtOazNxU4^YScF6Q5OUywjO-$Yu|E61zBsYL5MWWamspdfc~88JnCSG z<*|rk>GOBALvtz@0M)3@O^ogcs#^+j)-HRk=X^ZUz|QD5T|V6lbb2NZ%@zemCBL-q z$v6Vt_UZG3#gb}|-!m#Wok#)Pp*^6-!)dVFOrv|zCje4HEHo*U@#>eRvg;MsHm=1= zz;16)y;r9d#hRslK<8=Dy>XEJ?%iy);c&K;(G9oht+RHnJjNv3IGV`dMDjF%y{61d z7l{JVOl1e`-}1qk?(sAysPeOmYdOZbS13XRzQASqDJ~_EV_d~#M~9_R1hOPAdA0rJ z%4zMY^*VtI#*}uQL_8fE6LZ#n8Cy88+{dc2$_22FcprozLK!rNQjWUs+CGhD2^q?M z3l<7pj4oJu-SI(zJ&0Oqae4dHt7vzoDFsyH(e|3`6H@ARUvuU&mS@P>dUrvN{JO~& z{J`{NMNsNkD1zZpsJWP*GiiGj4tGN*)q}c6sdu|26-h1j_&(_a>KAaurlK1=s3GI0+a zZTKr2g6;1wI55aU=!Xg^R*8G)k{K&$LVPVipFuX~xyHw@{L$z+m-T(_?d!GbDD91Ul%@FB%mj_zt+ynNU1c!aTO;C?^}Bx$~;oo&3Otc8=kK$&hgJf+q=*CCisy&l~;@i`!1GKLql=t`JU#OM>>e}N`u&YF)Nub6V&QHj5 zyI-Zb+rN0$gNKI;7ctkU+EtqKe`Jc~CUJnrh>jokF9M?pdDf?#D^EVzBy+k?z@9dm zgkyc85onFoNHq?T#zM!>S|xA4saH*904d|%G@k$T48o$^%cobP_y%rqp0R^-`q|$e zp4D`9s!BQ8k@5%O(m5kzofhk+qaIf9sxH?*9_C-UpQw!-235bU-yOBRbsVM5!pZ@t z2=8P72C-25L-_G}yXJ7FbeH}&7&@O*mLMcPojLj^N2=(Xum4LvrMyDzT*z-bvYS9N zOl-%VLv znRE_JiXB-daF@1giy$>>*XrC?ulC9xQWo^cn+Gknpe0)#vFx5yRF@VD7|*>Akg5a~ z{1ppN=Vg$(i0c!cg&%xkt(rUbBbtB9k0Q{6ojhaPyd&2fq&T5Vlf_ajNGdm<{kk%X zsZzsT|IV`C2`u#F>gmrSAHjp)1|XM#IeVgZVifh7fMOIGF)-e&JM}|p#45-zs>b7V}Vc6x5kHA8ULjE)d9dJxx4s*2xuCOM}EOEW87G9C; z(ddm$$!tq?yLqh(g6$}+i+Zdg%SO3Q`F8^NX8Q?M3z-Y>v^+|SXCmJWj2PO_W$v!I z20U@zud$5+1ET%Rlj+J=J>3tt7%qHh@b zXtfm`xlnd7cN7H&Gmj@w{CM&48b6UBKqZm)=26;UL%XbTqr#}L&JT~Qo>SU7JRD-2 z*ou>0;-lbw&H40Rc<8AvYq;^ma>Kmry{Z0mvm2c{AOGGnVll>LTB>RCs{V+FJN8Co zm|ApR`Ki@2VU6QW3e;A%SPlJ7EFCZUPQ#O6EGhZgs{sI>Nr0JrQTbgdSV){znv9n$ zYO#GQrU8B7GWWq@Pn--V;eGuTBTeJZwuKsB8*OkOYs8~qS2Qfx=WwO_*e1#zA%@NS z!AZc#$Io!{PE8}8mO3`}aR{v&^|j)qcrTZqf@qO;NxY~NvXT=wD{9Bq@+2H*$3*uy;_XIjWybTdJa2DJMCpop3?Ks#4 zE7HQ&BY@fyy28}YWv>Y;MSM&wSpNF>UwG5J_jc)@BCu-rDK^mr<{5$!mvXp*J8hZ5 zZ92Ox59lWs^Gu#=0!Xy;foT8k5tW;?)X^qtjs4%=ZA z?apce2Y9t<7`Q3gjvgyMo)4>U3bSzqShL3jeP^4zWy%gAQ91KWM*LzvzwIkHjTxV^{^l_GqSl#gRX>t@Lq+!;QBfa zh&qE~%BEQl*h>~$pN>zHpAQr&{U8)g+X`wIx)Mnhty-(A+bph>@N{avJ{p=hr4f5) zHly0}V;WABN`!TOs%*VrZ}O3vP3^k#c;`D?e?Ix}!$EbgCV_%i9gqDbycmPDBIN{Y zsS^7L<5J&k)tS=PBFmbV{HL?Vj)w?~c^*sneEW8^PI^^K>Zc!2T)ykO@_L!BvjzW3 z>}|gmCF{_DC^lO-V5sifpa`wQ!L8wTvJ2sv?Uwn&)93~9A${-iXgVCqr70RT`MO2w zr*rF(i@O}S4zg@whVLfR4JMbLbxTiH80m&-$;&*xxz`x*{DOrUy7X{&QU`uY-?6n@ zw`36*aD_D^z2vi12qE6TPPjsmHS+(hZVY>|JlYXRiSTces*j;@={fq-)IXBiey-}io|Cj)J)v>>|K@3sG|^OG|70OW<}`g zAB1Y=7g2Xvx5$w{mlW&bUFRRbwJDW`FsXIuQ#l!Ej7lDcZa zZOu5a9j{ObnWX1|ah2`(i@Lbg-soukvL`&7JrZ#%Lp=?j$nHFC-!3qRkEH66d`8l0 z34KO75RxR%?n|6uSHKtqLSGc3?{y20dT8^9xYOak>8>Vy@J6}vYPRuGXd`{$DtK48 za;=_VL*C6{E-x!H@K9(-IkQ+(OdlDIoL@k~P{_lCq)nvHrq9$$jRC3?BUqG@ z+ViYsaq27_o+-1+w{z?KXa{@0b+owEM0{olF(=2r<1kbXj^Tt%yV__ZJA=!oqm3&( zMRh-zF4ybgiO;wVCTB6;tsW~kkBMzXYi3%la+t>0{apwX2+HQP%M>7Dk@XP%hEf-fAzbA>cXE-%LQOYrgvn+H+~70O>{Q zR*XA~Dw$rDd)ZXlFVWBYFEVg~Xg4%+#D!P0U5lBnyqk62)TEhOgra(q%#-~C+O#dgvXLpIt`|ro9_hp)(aU8cbG34w>aw7EnZZO zMze8+b6;lS9=V!!=Cwp;N}}gd;Y?HV_x(KBMJJ@!WsA|z>i~yqLV2y;e5dr=4L$RM z`&N{syl5NVA15$mMjN^asU8UfSpsS6SxuEL-zmfE4D7?2$jtTdW4UhF3|IMuj8K&<&&R;b z((AUPGcF)Ybjra`BrScq!tk2Yp#IZQtAz%^_&n@NtpVsURAf+Ezy92Nv!-+ro&wF5 z5|0!Xc!PFfzDgA8q;~xi$APW26NDoB1r+2uK<{eueB3IiVYcwaEZ>i;^v~*1Q|A*K zioOP8m#qPL{+v3ihfW#M9{c-kMkXd(Osp=8S0hYU;@DYkFU~sL@9+^qZ^xGp*^v#nleA)qB-6cz~)&!s=9 z$$S>)F22&{|IN=v(dZrEy9-q1lcsD-x9E0$n0*D#+B(r_;HSiO=S;vjHNbezp337=i!St_)Ubwg z&$UHTW5`LZP&4q$$)iRGECoJpqy}NJUeOQyw$|Cv}E|)W?v8R~!g`ONh&+lBv z%5V2ayygsQ5TM9@fS9=%HNFP@WXODI4|@}Ey0 ze90-d(uXE<+Qt5&morDoGpM*`(*=Kcy2416;6PZ}Z1TX00njEG@@zbg5PC*p+}s`G zQ2XNnrs~S>V@DPDZeeKEOqr@L|C^adRyU;i$e0_m?g4i1Qq`2Qs7Kr*>8HZ{*6T!v zNQHB*lnM`E|LaFj2HJOALr>h&-MZg%TPSqfk@8x`{|<$4_b+e#-0BZC*|+#e?8wLu z-LoR1vVs|5eR+7B_l{Wx_^=XnO#N9Hm8)JM&5&C+a|{5Rwh%W%>$FOunds9Y;f;#aN;#;X}5PS-;UU5E92EKtsD0qk3c2kOv; z4hS}Vm$dN|U-l)=UnOdu)DR8fQG739W%92?SWe=Y0{o0{yb0PRn#Vs9PlV*Qo#fUlw`zr1$KF1kFisLdSw`e8pcAS(wq}3O*H6 zW3vxb=4(K?k(w-DT_vlDL=U^<>0G)LC}Qj!KJb^}!=czi${!Vf&~}U>o~eHuA3XSy z+Fz_8bc?q%hEO^BjbsQi(+k-piKbuRb+7?&w@7BLvEefvfD+97S~jc(!-9LaP-7YE z{6}Se?~N@$H>=1As9!Ru^;s@Zf1X1$l|Vmu22-ni=de&2n~pS_RysUdkn@(qhF>)j zp_!kxh7FP|ii@F@e`55ROi9l6=#Lh#@c(KQ2KltI)$DCf0bJi(mU0N8En{pzy2kbc zh6N3j2af|*TB}ocexLRiw#X5qhA_68Jil7<$pV{jjWLDzvmam1qc#7q8GueObuPtE zYbNi9>LJH8GKUy?TPNtm@8a62vXzj*4`_kMn1!|{-Aqd5Ui(FykC*DTvgkbzsx$bc(`9N+Wk>@ zBj$gamqHXs0IH?C`yER2=E~r;?QiszyH&7YZeWUW?+IWdg zbQ|UbCC$5N8>WTKT>sGJZK19au5248 z6oD!%KS9fexcL5-M+n)`;jl$%u`KQJAbs_0M(jB-c0X`TVcERW^6SYhZ->J6?btrDe&^!XM65*r&XB+4=9 z0QYd>d+-)bmaLtftL}Kyi;kmVx&LxA*~0O=L)kiC{Px|VL{j503!Nli%fIh#z))@lGzt1uW9R?vq|({Ta&K?;khE~% z(Xgu?b`)4je7^ZS=c@Z1zr_T)du31f87Ux_)z*mhqYd@s9w8;sZ()a7R@M)+tenc8 zXH)2uH*CSIgB^LGN^~nQp=8t38S{E9&2G5*-unE;Q73~<47S>n3`U`Xl znvlJ>dd>#-R~|x4I92H`i_wWzn&Q)nzX~gO$=0R`$SkW(l;6}UX9znO+u*4pev3%T z{*vJJ!4b(rW_?sKRxR!JG!8?r{WU^WPVkMH;(swKBoYDzf?wMfTT3vQbhltz`~2no47)5f?;lMS|cp2f!83TPG( zfhQ}V=P6(HwG=>KU1X}?R-&iZA|mQ{zc{|p#s?yFnbE9-@-E z8%%@m90JcG;IO+l3k}g(ye%gr@hi<-d}roPcmKYfK9*uLNV^sCw)G~L&6~ubBKwD+ z#5&ilyTv8O$Iezq3{F7z?ocg511(2mM`yMJ&|8b~`f5e0ByyoK?;?hikBLUsuG^zW zut}q8GRHmU_SIbMb8fNEd;*6oGr z^~n(Z{UnM&{+)_;`TUmEqYJ!cz-#7TCOaQr;xL4tpV7ATCKPaeNBgL3e0PJ1TxUoCmduQ@E$Npqu@W-CF z?#0>5K?P|jydw*>mj?>y8AA=arRbiVIc24W{o56#mFPWD>>YG}_>&LzHBXyp*I-8E zU%zEI{MV7JPSYF&{rwsX>BY)9q*pW0k1gP4Pw5bK7mz4_k@Eo~JAuy&S6T#i(0-k8 zM{3L0qSGwUp`YKeis@`JN3DybVvEuJg3NkW{va;+-(UXsmHHna88m>)yN2FB@;|=g z|6zNWfO#JmuCgYMWdG-%S!D(!95tL4KK%0ovnBfc|Nk@p|CRaw%54@z#^uNMuvHTG z?*__0cnrc*WN_{K5VX%Zk`s7y3Nb6_1%1>%vRwEu`%m5U-lHdo1(7lNu^sPq_WZBT zssHVg_P>3ijE2rz-0Mjq8c6rI^8Ei`!}NdiX_6IS1J-z&sVv<;_D9V3__Me8K3u;1 zr#^rGFa8ory^Gr;KuY*W&X(rCcDDcC;{JyxMG3N-AE+KkKycwfxo_nDy(Rt+@Bi;N zfN}CgS0uiZ9^uAc{_`lzD|vsPN>VvXR0ZURlo!POC!XZ-riLX*MAO`4(=Q*Z|jNJY|L$qLk|# zqLisPQF=pzI(Tlh^bXNTjCk#8b zQzj7ltRs*EMsKQ`i>g>2v{C{7yMee)mNBo>C?i;ht)Qy*b=11&^}sUp4=fq~;}$uu zSaniQC(Yd|yORg~$6ZGDH~K9we1E)0yfY@bvyEAmgWm9GB*hIa#f0a! zZI^BbH>(+}RSF#U^D$08n|S7(U1I8R4~UUc9Opq?ds|D_9@e|dldfXy8JwY~&Iu zaV`Eh71*wM1%d)A1{4n^-rEC<;vSSe7&s3Mq)n-TcQ*$4Dv7M|=LwCyg`uv>#`4@! zO+?BE?qW^!fi6XCk{W;hnEHDMt5p-3YIVtam8v&jM821vmm2fOMszPY8?R~dUp~f> z^7jr#0{u)WXiE#ugW_Ze=nwpxTq8P;@o+xMzGPzH>8S9BI*$5-^>N|S_jgBhYmhfL z=Q7ag`nGbd^|C zxonXol0}~rO2Bi3>8lzIOSN?yPwS7p6o6&{emCE)YQ<-m2mStW1+yaK>WAmma6{gP zAZ;&zq%#3cbdEtu{V~t9L$c+Fu4#BCMT=(#)sn-3@p%-DSan38?`81b#iRi|v_`)e zq;U=brJZzSWza%n)*}Z|aHg)}4wUUrkUQgAQs+Gs^Hq z)|ViC+ITj@UjU+qFklf@3-~t#4_%0}|H}gd_ zo`b2dHqgE@%q8M^HEMf?U3qLH*#J%xA+#qVE(+kE>+d@|&a#>YF4o1)x70PoQd4mO z!m{x!lY##J)6@W$7@{~=Yw(9t>AqoJuQF|t0{+CS*DZ9nRoG4hX*I)Tw&WEqGaCP6{uii`5jzd0Y-#$<>`&r(FkR%j)-)O|kf+uJ}n~r}18v`+R{-t)iB^J%NuU3w9H(7g5v7YLy z3KI$9Sm{%W$N_5vz_Ys8*ve7N%~R|pH!zO(G0*Q5ygS*}cE0$@sQpcu*m?1En(apL z`7pwsUgN4iEi?|8Ae5zK6T*WNsU1-UsNom`{wPn9xzjh&c2CldVaGlF#n-OZ8YC>L z7t}vN=ew-G>irG$mJVpCpKlH~gvJ$kmk?MCP3`Y`&QFafZpAuu@3jVaBX0axi_#yy zIxFBCA^)1XRrkT`cqF|>7N1W1r9-v|LcwBZRqT*r!(KuLzaB9I@GkACmjZKxFJS;> zCNxk9bp%)?MxlibLFogo`7=-38tS9>Dmr+tQSt$YJx}JGv!zg3-vQmFL17;lgLau+ z!P|^)r{iBoiF=e{4d7iAAKLOeZm(3=xn8}#zF5;-4MMp_>kUzQ%?=A`rXA_^{<7>> zj49-dChoxuJ;u~aK{WB@bUJ$;%;?uTEOHy)hFy?6_jO^69UqbgWMBDSd`M@wJ@2@i zwGJ$GQvup3W=4nIWg~Sw!__~FVnNjfyEzKk^gPx-0<`(PP%&k8vU51!0d2-;>6`H4 zNiP)?axtr%&mPbSCe!^ImOVS)@P9G&mQhi@-y82&DGBM65|joJ6p$7H>5%S}Zlqxb z2}uFzPC=xbA%`504uOFgVCWu(8amGSJLkXFc{{J3HS2lqd+&W+`}%;_{6pjZ-iH~C zi;-bNE{88Mhk@L<1Wu9bccpKs+-Wr4`PD2Z8y!mRz zBl@uz4x>;{>euv=Nai;4@ARh~JfX~HssuNr(^?iZ^( z4cGNA&%1O-_{R#ghe0}T{`8SVVaAOWO1wV)ExaE1M|4>&L55=gYK)i9LJ(3E4&%8* zjeZq7RfOgpo+bBLoS00iJJ8NlS+dK5ez6+QVaNU}58v|x*1&iM*3&8wY)eI*5CH=} z74`VPEP!_X?~GC=jGmj0vvSC5bn@R-ai)EYtLp{s*&f^Y-FfyfmfNGaED_()d~*HU zqq~bD(^zW9#Wy#S7f~S(rrX89>i@g-_98PsCnR8A@pN{~EB=1IVd8n+Wy$7n+~vO% z3T!)(ekvX)^$iJw7S>guuS*j3uk<#Xgtl>)BO+=n%s5h8PnmEF{QR)Xd_W~ifbG{L z9a2+D@;>voSwrFF?zXucV%bzaETE~#C5dzY?sY8zf>vjt*(T1eZZoxszQ2L7T_K0& z8-_eM{NjBgc>ZMdua&UzY{;cdZj^Mk zim?=Q7)eGYT(%cPEpP)E)@vme*mg>Yu)j7K(6q#?0)3}PN@ zU#2kO;34~!6cXW!+*O^HXcwatGhbwrRd0)Bk-^jlF{jJFs4p)~GzlI&>uNyE9ZqxA zG#I;xru=tCAvv8glOAo4j3dHh5{LCRo3*^adVkofWz{l_pE&Tso4Q8oX!)xWQp^;S z;AM4KIYi@C4_fwheAsyNwh8)Qo;2nXu77zn!OS{#vE6Bc>p@(55^>~`P7dFrA0R_J z&D6rZ`vp4)h3#QFDK#rq2{d&c+|+uv?Hp;q z2>%PLPd|NL$k*+EnTWA75{|~)D_=@RBA!jKYnfoIW*ELlkxi7a-}xPhbWEe?YnsOz z5<>|)^b_}$@a{D_EOPe?-?}7b++Rmeoe~fm&JDs7Zz(3c#oTZXea!xE>yw&x<*VGP zI7?l?Q)7Y_zR18&26KKMI0=uGh*W#4){O(3v907$>=rnm(kZkrft+smn}+|n(81oe zZ-6B0cuqgO+MlkfF~GQFVK#SdLh^&>RWGW=q4h;ersK=;tRbL{%ZMG+{g^?7G#g_KQGii_8mKHuIJprr5mZ%`w5jP`HwUM5> zx{p03*0ldR7Tt%;@KB8;VWNyAX83`kg+Suy-k4SPeFt>yx>7ut`P&`9JqTr1#S4^w zz`nUz@EW!yy<@SyVX-~M1Q`BIVwCpCq7|{s8on4Vhp_rq=X?T1r?J%GP25I-r{X1B zD=+%GG4+gJ$9F&PJ6kGg*NgGEPwPs-#}NG@j9)A@Df%wV zon1v!6m^;0XA=cttE?Fk(v)jBN!^3rVZV;9PHi-i<`@L5f&uxB_fo)vj6az?={w?n zwd*pi(SL??FMT8yiot71BakjA$5~F=42d`w1LC2+xf5HL_jCVv*g>7X(ipE2FeZxGZGk<>g7qWaOp>1D79eZ8P7AWN+A_ z#;T2hktErRp9M3oYz39JMnxlIFd4azsSqDBVh(0eyyPSj?p++F27!Hh^wnRW)kbR> z)WhlqGyoo|J^ed`?iwbJLq{OiGIj5zS(shv=7D#MZshYrfki(td|N4G`kboyaQz*+ zt4rs07_V56dHYQ3x+sT3Jtsx4S>$(y{U3U20nJwy;%n^b0hMy|Tx9efOA%xBz(Gi^ zAd{$-uieO=L@ZvloCSpN5gh(z$#&D`u+QIX)QL#&5Z0-0XtSAe_(-6&J3;))p3VTr z(@=m4q1osF+SF|Pu)4?A2Cy0RYiP6z(mMRPF?myM)n`0G!#LhBIP~_&8WGwpK-@eIh7Ml8*}P^M+y=u^YIH{NQJcG zCKytoDXIRy1Qu-QBF5^H z7s=GPczmfFBr9-=7C{W(9O3CM=ZMye8&$0UC60Pzy!V8W;ygf$5ZcWj#%`VR-D0?F zAZZzxJ*?nRL-4aA|A}N0%>rj^D+vkkWlbNM2h=oE-ql~)u8ptFFlL7)&y|7ZuiZeS zN0IYxm)&7`GF`9*-D0JUFdHBo+OM4mYg?DL&OqB|KOB&hT4Gb60H)I^} zcf6Pj#uyTZ4U0Q5W2hKIOcSHqLg8J2OaMo0p{2N%gqS66>j+pp?6yXWGz)_szskmt zW@XlIIZ8HpOy1j#WkH1WxUClMdDqol=38_-?msSW^f)A&H{_A|xr#y{ck&URyG5lQ z)_(u_A=T+z<6Dt2tX@6%B2_#I0=o0dp{JOh^j+&*2kd4shA|r5o}({mA9mJX?3tpz z2f4LvrhTtRjQlh?+VAK{^HX2HI~b_}>)n0MTWsApWsT$}+-tnK>)bnvWFG%n9X394 zFnA37l?m`T<5{p{B;DP-VaOA23BDt!elZ;4PUff4_Xeww^_Bp}U{FxgcP#qF))&D4 zb_9hs%bpBm(K77$ zq;WsoBr+T*o`HNjA=`?*{KgWAsXHSf^b?I_L!ip4_!E@Vl;UIz%HRe| zB5Z10UD1g<2y1&N86L2_x z$GQJG9gDv~upBLA{s8l5wPjtnnhoJHO|5{*Q73+=evziX#ZqVdEnQt{F5A7|2=^Ag zy8*h8>}gU$EECQRz(jC*(mE)@*m%|F-$YTd3Hwzp)$TYmabv(!$GNjqb1Or@&8MKB zUq_Y$RCrSe+!QfHVLA2Pd)O;KAh0F4>$dq`<~IMbACWp8hra{ zF-PEM3%|9q9E#dy2O=0mwe1rDC{>Avq`C97x`%4cG3xw2h+5}y zIP5qD%L_KD4*jVgO}zMJU~ZcN2xXQ122bbbKW{(2U$TnM2L}!k3vAUJyK%Nji+L&X znl>r|ZsKzV+443|WgL1QMq0mE9n-sU?z_LP@m*-US>`qD7a5E%6B|nz{BR<1Tgz$4 z?J9xzT(+M%xnEtgv~;of($=6lexR*=V{7)0o%8I6?OuL$&6}(@Rj>=x6=eMubGQ}|C3@XJK!a~T1+OlGJoIxt2 zysrxor4OVIfUD|FYQ22CuvkpCF&QRWVkWz?#9vK;nEts%PbTRuW<|9AGTeBtvt2lV z`!^Y!{?${PaIo<~f|ApcXEfml?Gmr%JbHP*GaZ;s;1ukI6V5u$5flxVk> z;9i?-nV|c7jgs&2#h1(1%fy&e7rBrSh4%M}k2VYF+-7p-tuzu4wKsz|C{2lsxA~0(5_C5n+D2Zct1MaE#nb{Nc)%Mc&fej zu#Pw_*1}E<`*4U{zDm2biHs-2BvbUpTflPqtk$n?o`emPjMzl2_Do0*qCsK@c)j4oXz zZ}3CK*>ei2eGD61f7a{Pb%(bRcdoTAdN@4=m_|e1S6E;jA^RU)dj0uXev{kg zjxM|3LpZh`1>q5bzB?=6wO4d+U}^<^2e8iktd|bIdYT4`Yv8W!@)CN#%3(GM{%+^y z?mQ3U?M{#%t2Ks)?cTtaQcN_#ufxF3F>wDNlgUgU`SPolxiB+i38nry;98`z2;Mm_ z@2@@$KdX1n9PEE`b@+=zXb)+7h1qfcnF0)8PMcaLv}ID?mkcD)Nq%!AvE1ZR#X$p* zxK2qbsrZf0gF`4$l;;$X-^nEIiz36qi_@X7^LobX!FYYKyXVNofCPmts85-*R^lBh zy!j|f2|Kmb&?0bMjzL%Fct>>iq$H?ueF020$=xWLv(k|X4N@QVBKK(@*nKzyb`Tf5VDoepc_ zRmNn~PF!YUId!u%OALx^dP#J-=k=r2Nj&is_?I_tKo{P{T%S6kb~sRmAf+Wm3Ki3_ zK;OBu<3H*hm!2X*W!U@wd~rt~tq8!#>n<_gWf8$c=^$$>D`NIsF(qDr8iII zo65;zkxJxW2nf8&GaOL<``gC_&4&#=K)x!i^BGP>JIy_*YC($l)wVo@yu(VLNMIlX zw0UbQ8Fw6o5=|I=s>+6&A5P12HDOBrl5@)(r{)elqXHdfmedChjhxBs8FE}a&Mysm zRsG(?xD1*p3G_nBT)AUdJvh~^`A%@*yu5DJ?PjSDb8cNZ@7$}1M8q)q$8l=o`61u_ ze<2@kl!k^=fYSlYhuyiwK6lrTk<)6|L=2ove^N(ajAPzVD6z|L)vJ@DnhvJziG0ur zc7n2QdDOsIHepAXOvmtdCH$tn0Xo!9#}W);j4}1CAwjxCU-rX2HxW>C{;ZM96^P0J zb34?Zy|}}684RUZSP})BKYQy~2(}?x`%|+*%Mfc_^eO8`QloMA_ho!q4)Ps0c>U{) z!=&m#@|#Kl`$p@1$3Z+AEzFjyk!r4#g0~hNRH?Nqq-7vK^Bdh5qY>k>Ghn)6%c>IYlM2kzmI>)nl)TYg*kVznWPU zFq}p1lNsQBZV~9bJlTP!^M zIR^QFm3HWkCNylZh3uM&s+Wi)nt!i_@sP$v)+2ny(a}f z7)-(+KX&+`Hv)+b{&OkuGzFctjO+BWkww*PN~`Vz9R{rUSz_}ilUx%|E$i&lFlg<^ zxjI=7rf|(m44y1?Ru~wDk9((i7|#gf-sEp~3K3TW)L7A*eLUS5DoG*lx(|?HlV#ZO zZ(!%qAe=ucYs6%su613m{ce-e)LZshC~qrf=$7rmik``TuEy)4Ns`nTgEx#a{9>jL zP&z-jh|Zt*1J(>koFovcdAGqCncOY-GPtnCFmuP8rc*#=fWK{kt881CpEzS-!K7_E zB8ib8iHkxZR`eh!JHN+XCKnp3G!r`NfW zXZFCP+Sq)G2)v{&I|F|N7Pu^(`92iwPt`--Sfy9;cTm0;jhDh87FBVE59lb?{2wN5 ze2mwO)}dq+J^ zM(G>a71AN%0@RRzUfbBr#y&#RX%9xI&Gg8{+GT=1Q=edF#0Ypv`9;%HW>UlQG1lIb32IW# zSZJlmfn!KH`NF;2&&vTod~f^NHHip`1 z2sT;iBRbXRAvleC-GxgsQ(<;7dACn$$lUm{C~Q5d_rrjcyzb1h7HpZ4#(KMZ{`qS$ zE?fOPw-J8muDV4+k5=&qd4PN8RB54fWD&Hc0O*H$B>Q^WYVN1+g!=GLlIoJb-ra9J zcGp^ut7Ul+#OFR@KD{~~>dlx;g*#Pn_Wt=dd^VJFtiCF*>HWB)cS4UWHjKwl*4=Fa z68KH$FuN#!w`H3-eQy0|{E^JfJv8-!AaJ7h%i$QblxITBmWiFCG@5woT=JTsjYPF_ zn|Az-=NYyDaP6(Pa!ZGA!;|F>oA#h!`-@r}Y zo6@Is{B+#t$W*sNoucAy4Lq>1N5P?i?^pA6<<9Nb8Sm%(gObyzlIpEac!dIokOR}j zqjCR+9BeaHFmo14CBs4owYxRfK${$0ucX1Nu<0u(G4tL6-KnK+0WuCAoG$%%Gb+5l zea>ee9M8THsDGgnjjk%J5lH{%H2IhTd37g8Vzo zAs6Eq7)ndxi`l}sETQD&^9tjIAZ|Wyyo>$Pfy08Ju`Isq06peZy$+fUTILN8I*#j1 z>fpfe-sYNkCS3xxyEtr{V@bi~GZ3baId<(YUG4x%Cb5^Twg1*J>x+4oBolBNu9_#L z!e_~~q{5wQnHygh&TH5ZlkEyNOuZV^_COpi)Z5VjN(7%F#Eo8iSp50=$T0{PZn^o? z2_?%6Byx-pk5hcKHF1}%#%uZp?*(O z_)37!;ez%C;N`8j#&+h9lX4=8>+N!Sc}~D`IMb&19YFNHnuc;7n1V3ct1` zH0g)ae9gkOO`T_=?-WHd2<|`3*13H^ih1eYoKkU9ahntc{WcoIWy#S$y*X=` zkj2O>wI|Zov(P!I;>2pLDuCN7qrEoNYG}eYrr;v=ou3-<1`gv70W34E#^(EeSBcKZ zzuP3CA>$)1TuW`nHa}umbTN1aGkOtGvrI3k2@vvkJeEi!vnZ#-rgbw(F7T*!YVLmd z4mVOt^rUt81~c~~T5S_@nfi$A5_ig*Y_FJuh`VX>;*tca<>I+`_ioS?G6!D|iKXjA zd~=+0C^|rQ+&2QzV9#-V zjiDuSzQl8_44VAaegT|?bvcd&IOUIb>9lgp07VO5I_K*zX&w$%%+{SS3Iqhrciv*` z6W@mQ0t8c#A-62<@>iJ110BQ!_fC&&85<8_KZrBoj)A(|RuF zO15F^)sx(JS7FML&0#nHZ4cZ_T&YA%<}aMSHy!}qij27ZYPgP6h^YY5Psn!$KZJx8 zMsD~p@Jo^yz|p`Ow&C9sadxHpMqkWRF-@$;PWBkm4A7aA^EsE$yPdqv5B+`pYT)-5KRu&h3mdgO@E*| z^!L&qF(w<~{YdW)r?pOE{}fJB(MGGyXB93*_bnIRbmh%-^c#an7xqB8@rr4`F9uEJ)A3KYW0F5wnjEf{9IyG|Af0dM>mD2NV}4l zURzo}e+UHeGL@V10y-YtI_9)CQH>;UVpm##O}qoQ(&RLBZ%J zJWIaJqa>2rEDbQ|PyfWp-4`mAfUY34`ftQ(g3%!3^4oEk@8QmgW-RwbIlw!{?um=x z49D5v#r_PCm*cG5hA}e0E$C;dH}9}n&4!xe!|!q3zUKXas;s3UvHNvCipsYJP-o~U za;=5&X3(dGOziXuJ_p>mx$$-Sko#;?dk|Ea#pd<-hSKRFSl#u>82$3haNtzSb`{{A zk5@X!47iQWl?c!p7}11*Q}61TBArZxwRZ+JsEu#DOgqw?w!E6hTMY$g95~QrE-nwF zoTbYNHiMV2?v~;*nL`UFT9)|1tmB&1&314->)c`B=pMs!orp#|9?1wA-uZ@Q9*#z?G z`##1ufMt%P4oVc>i`Fvlo&_`HXKcm_qxvT)N@Xn@o2ARrql< zALue;fx$y=uL|wM7x7}hKAp~jNog~_sb7H+H!Q6(k zd00O8yVR<^N9+-=g+M9W6KY!`8x9oqJDN(oVcp+_R%bl~PY(H;GN3cafSpEHmDi7X z4D@@juy{WnJXQ%Xi0l(P>KxD;qjE~aH$g1}3>PrX%_6=0K9ja4G6}*MX+5R3-fOiM zf6Dh#6HSMh>`_Ez?8|JFn&?S&JAef;VdRb^3rBaW$fN4X@)p zyW*@#xgS#(8#VH1`HlIRNAP^jfa0{B7!(5%b zVnjJok3jbTobQ?~KY?p9*%quppJ_y&P+w%MPBs^D}QfUQbAMa0#NCbIRy#jskk71}}V0|P*$EB$`$ zPFPc`DJhmdw;wNx)(rLC{!MI&O=~VyC2AzCraBMmo|v#*$5ngw(fM-+w6oJM?}XRt za8RsQr&|^uyO4Ax)R1s(Ev@3R_*&F=uDJGyUhFCFXG=rFYGaqS` zSzNS$Z-qBw@-w~BT3caAP>tO0)Kng7v1PbgzRS-KFINm275;iIZhY=qmN}EOWzH~| z5SSrk`AKRt5KVDj0R-C{Zp!rR=E*L))ehW1R!MvYPW8fkNks+JKPOVl=yIo%1nRK{ z^GDCFiO=2tC(y2v7GPrN-M2n?!d5dFC@p`5p@J5JQ|sd0J_RMS%iXq#MQ-zolF8Y) zJX8m}9DK%*<38(3{S8bqBo0E%{+`wJ2fwE!dqZSI>Meva?Rcp2_w|@>k2g%aewwe# z^r)jFs=Hn5L4W=5ON+XQ>?e76>A=bV?h|TNF2u7fp_Ttx0x>&AyW`Huy5CJ1*KCnb zfv|G;&>DO(0sXwc@p3$yKi}T1@!V(>my}~K zc5aDL=xIXI>C(y4+9rauQg_hrwr4!H+@XOImoXZ&A#KsL0II4g+Y5s9xPVtrB zcrLh@!Me(E7Li>2vQmzyuZU|x&o`*H2<*^%o$3>)(N;4`sw4kh9j&pOWOud!h zIePtn22^lN2xSe6_d~pVAimqx038!IqM|O4gH1<9hL{^sTi=qz#_YYMTpzCYOn@dY z=@-g6m0WhZ-xQ|zKldY@PL6v_=l)CgS3`y|ZXKKLN9iVns6N5Xe!cUV0A^UHz*GK6 zm;=(3Ip=b>podJou`CT#ASucs;N4i)6>*p?r##f`c1<4+xU;gAmZp5DWxt&90qskfltBx2kY+G zYj=R$wN9<4TRNpvbEc*BcjHAxrr;9$5s?8OhiZXZLJ|QN26ssT60TQO|N0@vIVwGg zR=HnUiwr+RkaGAsJSzWuzc{BcUlxkB$~RYK=7{-!vjsK^6@9Ju{Aw4U;4o0;{djcO z7W+|S5W>~#))F|d!KU1^{d7E7UsGNVF2gM^-|%x^rW(d_@)MMenW< z@H5xWV-FnP9E*n`&28|~qmV(&?lunUzM zQ@C%!JzYpZ>zYOj(h=Tb{VI^pIDaCV+#(^;amAR-QDvhzF8!(&>B7;c%0;>u`Z&#R z6ixmih+WGMjRp=AU9`D35ME4ay06|5Joa7q7JKsrRK+p4@{S+!&wN^v#Cv%E!ARTd z*4sqCMI0J>B+jNXh<0;LCiODPRbZzR!zwhL9Dlx5tXx{lc zO}6FHVi#%cf5`@r%1F$?3;b`DfSlX)U#UM!!He{__QXob`VqYzA+EkPO<-7j<&Frn z^avfZ6I+|J_KZvJ?@ZU#I((8X;=;KCHjEt_s=g#HLQMfItfVk$a;Kfl6jEb9Nanj7 z(E%a8GEsMVkmez6JHzU=sfny)tXgU6X!&DCxnPU)2hD)(v21vqn0u+2uKHd`f6--x zvGZT5cPD+}ed{Lx_tQC~6|zljzBsRhXBjAYVE8S%W;#afpi`Eu`ObllTi-w=t55+jDc1s-#!K8%J?3xXwZ$K8x*K?XH zSCRjAYNr1S@v>VcEKb7;Kzn{)jT?Qz1s;1VLph)mq%ce}j5M1z6*2L5qI+d6#XJwv z*x=E8-x@s!AG6`u1>anP5ZpE|QhI)rXjy^{xJOwui|yJZ!DI=$Sz=sN2sFKrmcW4S z>UU|Lbbgvsy9vFzM!V5`)H~|eDs5!upnxVZ$J<7S;apK*2do_wLm!VW7jlf^o-K!D zzx)coPs~@(r$(0Sdwjpzs6H+7fY?*=6PG(oUMDC&lO}ZCv5Zim|9J z{|t`>b`4rKZe3m{n8zS7)a05cV$oiI*}+#^;zFDeJpz1Y%w1bjj0jz3 zyH{jR&)&?jz&|^C8(e(|wJH!iyJ8~aMt|_kPEW7o?5&99`h#~!M`n3UF1l->*R!W!Fce7S`Hoe24=KX5n zf-O)}2B_`hVg|WPH`ToJ+OZCqPpK$L?c~}jNL~t^QMEJ26X#q#M?GjN(p)RE;Oqko zjxmON|38{-B*{&WfB ztYLZ?2|?rCG-v5PF^(*G&Wnp#KvDZnEB4&k`~&cRBu$t%y^Zra<8jJL#{7TZu-DqN zt@MiPCMXSwN-d_pNIE<=zRv}k$Ik54zgPG+XnwW5Z z{(N;78kkCzXV(2h*{y#lf?=_1tnJ(!h;HIE=W^AeVZXIlOUPEFnJJ15QkOYUe%tU6 zT4JG27rUg?$cDm@9$A{*qUnFXZ-c0$5n{w0b&89b_>tg7a%O3PV={{8EPm{%KbHAk zf5M>cJWo&|YK3KQXbc_>{uw-Bp)D5a$h>bht8Ta2 zx?wJ!f;GOHfCP;2eIsfc;OPk4aF{C9x2uRZ@Gn$TsQUXyd~x*W9%B7%lt<8OeJ|B zfhfDuu30o_WIcOX&33}{`SnxUU(3uSjPtH+ghd!3fRLZ)Xkg{#R#S*RgWsS>@;1Dq zlJY)dCSu6N$ZA*nC%`rv-4IN;chQt`R#j-|sq}9QdR zldz&q_<8YjJKo$NM66|};}2DWxb%AZ6^1g6-`G3QB&P3Pr;c^}wQD^=B)H+&q; zAs4v{?w%}U#Sk=%9WbJ_oZfPR6gSND-0bM?czrF^2dTs4n*6Nx%jnUn)rGnrVduwD zliF^_WWBskDf+oH7y135%Q<-uZD8r!5|KSw2nbv*@T%`La5EeVx*Z}rmW#XQ=4nv@erR+4*1`oYG-#F9XF=DmVED3viNTGj#k|;kT&I6fLRs7GbBcxjx>! zedd5Z-c7{Bd>I)ub)!4}J-tUTBEOFMAr6=S{q)#P9nKuGR7~J%r!UUiAr6i zlDfGiJVw{KK_;a0g&IEX%W#9ooS0qcC0q^O6LfFs;FBmu*;U-Ub?O%zNtzG#Kh4M0X5sWN32GH+I` z9M7AUwB{4mZZZx=KJ$7A$tcqt=Cz~9zY%Owu*Fa%>h0K=RIPajl-e}LxZSyi4SCsX z8{J9_z9zHYEN@!-ky8+&#E1wAG*MIh>%UV{b{VlB80016UgbW9Ii_aiefo-L&ppXuF!)p^(X z38r+MpRm496UdS;;r2+jxm`@&s&m~pL%&X#f$0BZix#B;gy4rlgaM{4q3LYR=I{|nL8uGq4t%E!9En*f!b*r8gS;!^SsYpXRa z&8w=YM-V$XsL?hI?IFv)KWOoeh-t z&4fbDfn~f!mUPIMznqRY3N8;Z%Y2u6JD*C6#9a5;mz=D7zf!kQnLnqRWo8yD?vUO^ z8Z7&*$E}I(!j`IYg#XS%#Z^?UQN?VM*odU_AuPZAulKlX#h=f$SOYQUr`AKx*b+tI zl)v-UZ1$LOpdMm5{!FwkNBqmLx9oumE?E*zX@8#A;J-}6NQbxgt!s{CZDmo28G+IU zo-QOEyWPc>tzp@vutR^XI>h&#gyi!_C6EwFk7NDB4?OSIamD5!UrIvWy5kYj_)6cp z6G-=AhsA3?vo2s}{)PFt)og@KIC{SoWoIy7mnj`%sGlOP0qxmqa0bn{Rfub1+VV=o z>y=O$b)+z+A$H-GC8MCLOYD?upB+y_OKrMRJ78#^X?^>*=YU#t6bpWrnT{1IYL^5NLU^YqJ|r-Xs# z>@kYlOO1k0US!#5-V>kqaO3Kf^HvLM2&(Wm9<6~zQ86GfN zj`B#DZwbR34}P8b+iKIuu|hOagF}{g5wmMmb~)ZgGJ8Y1SEPTTzXlNyxiN2FqaXYj z$iH(c#wS3wnHcKvY7Vn$$z}#DYtpXAc8S#?$<0h(6Hjx)tMMtHZ~>PNSRK2lf0*`% zr(Z8Yt6%!AQ|cgo-gFmh2D#0*$TtQ_x%Q^{n`a4>px=y-D?H$m_-S=Zy3$CHM3Nkq zE~5`_I)H;km_Aeb{Szj?c22Ybe#pChv*}8Cq({2b=yG1&)HYC^1rfiE;#41uqG52L6NiuB>LB6)zg{TUe_ z*5IDdBJG5(E@OeTC!2f6D6sa88li2I%)S5d)ixxdR6V)~5<3|Cx9V@k$am0U5m5b^ zxZX1-h}u~`n`z#9;K(a+kPF57$2hHQB!NoRcV6}?ceuF0)4pjs>WLq~X-5W+b{5U8 zi(P8Fmlj4o?JoLwruT2;2VNuUJ$`zAS~fSTR3LAm^}W{Xa$gO7cM_dLdB0n_c4yj7 zDZTNnQTAg3btbCgjn>>6+MFySLo*JcvnFp#UF7K19FwO4ml5rK$I z_brL=VtW&*wLti_iaMSxg3RqOKc5U`w%68EmYLV2(cU_`IeSt`?pPVAU_&0KZPUK( zRlN=HUdL@WCG_;iTNJa!Yz7*)I8|r(B;`%xW3=qjd$2JX=OaHF{3Xv zXfvH)F%6n6nDcYN5qQ6PbnhD(IK}@La$|aj#WVjpz@RJD&0ij!udndZ-2392@s&o;E7UD+ z*7Zo<7?P%W)>1}HTxp!PlU@;_jT!vf>CZoC0r`}TQSXc-b3KOIBbg#XWBk9u*)e%T$_BwUTNINRqU+5dgjHKELPVuYblnByf& zt~mrWPBB5}*EA1$z}n>6zLUH`zZyJZnNyEUwWmZ3qruK$watLCyVJ+aGDaQO*XiPD z%70!n5oq~Uc>R2Te+1mEOO4CuwG0-8&~x|13R%94;e%uAi0E?Nwe~D+ydh~@RUhr1 zh&$Vd&Ej`+{`TK0me$}wG(8&<2CVJTWtE`xJ_+0y8-vF6fVJG+opf{K_G9%hUwS^^ zNWvL@D~;E^s|sT%t{L4*^HR-ov!mU7AUWN1Dan_(0}#`-*gPND8dJ+SmbSTK?8O4V z!5Pd#Q-CTC5J1zTg7d1W2qge^*xs9mv`DhSg zMSsWn;d4j(?U-NLO_!L-Xcomb{_v1&+*MWhotUb2$9HG?`sH*;0Pz2oS;9+?Ncj4YEztyo|l_D-pskLk0Sqfz$HB? zzA*)Z%)e_tR?v2po?8Z*=C0q$7yvM3g$ME9rFWDkoh6==DRqLlr`1TfE`x2q^*PTk ztF|sLyQfZH+B2%#kTIw%N?KU{Edh8EPr8`nyBt^-_!5Q~@P|ey*D8&FAl) z9osJkE}EdGcKn{9!Rz#aV4eEC`Z>2n1B~JJ;-bJff^ch`3e~}Q@haNq$81!)D^r64 z_Z-L-RxB*ZzXV9WXbgZ}@+-dhL7^=)axArK&V@DMDxO9&9$ zgG+EHK!D)xG!h&Fgy8P(?hQeMg#f{&ad&7MXyiTo?wy(YzTcghJO6xtO;uMHMc1jb z&pvyvwf0)idYwiJWwjPLQw*^gGr7O+xRlnqkxZtixiX+dxvdR~`xIt79}!?fZz-qx$6 z1xQg@bt;Gt#H97il1s?J-$ECJF_Hy!TO;^K(Z{%nGK?&bUE~^UC*o5~X2WW%gt!^O zfhH?xM`u{#TKJGTf8aGYfE`pH6XS@pI!IX$~&2Ih`b+C>o%fp*iMM2^#^2siq} zg}K(-?}k5s44t1d?*R-V#fomhb#s1Fc~^#wJ_S1`a_LinOQ04VBIL@XxVG z(1muhG^fDmLEL2GCVThAE~IFvo-f>!Y6<=F;u2zRu$(MnfX}M>JCkPFzFlv1E=;!i z(t4|j+zx1=>m*y8FVP7HcF@mB@`nXM>}l5nbD5AX)w7kpvYn3S?{siO_-IvaWn$h3He}8U}FS4z?GJjAp9?EaTZMzJsfhbLXYy3jgaLd?4omj=l>_;a-#@yq6{LtCIK$AJ95S0_thDJmPbg_{x7hX$Hqium zK_)N|+xm+2dL-ErEv0qXEthe>cTk^#Z{)?Of88#| za^pjtUhP1}aG2Y-cc$s%bqWZ6%aZ%L8a33cIZ=aIGt4uo!nE!3sKP`8NU&0ELDel8 zLEc**Z;0oN~xv(?b{Ur-ejT4RCdQ)*FlHygtF` zzy{x{a8|Uip0a=eIa5)UfX%3{%pEYxPqn>EQ{ehDW39k6D1 z-tUg2zs2`lUsmG4v8w}FApPugK7^>OyKjiu9$jUkOM^S!Naen*FLw&YUQX3DYk7h> zV7h{dCPcqa+9al4)#`eHI^+30kg$DmN&Dsmjg3>8NcQ|R6qAL z@O%YrleIj@Sn6sO3}caJI&bGPAJsZlS-ok9Ykx5Q5Nfd)3U5Sfp5UgP_GPYVUgtOc zK6-_9^L*WK%Ay_RdSzYueHKGI@v0)E{6&}*!igS7f- z=!C9bplKE%Loi1Qnj9(6p=-q8wecvrm+u}tojyQ_x)nAa1T3h-0R^+47$UbU3Al~jSOTZJL{6S0|<8UoYshYg{xQk7y%d{EI_v4iU*FM z;1f-PAYh@Ri+@A7s-+tP- z)W;ViaqNW*V7Ox0=`a8>7^JvvPQFEaA3W-yKt9>g#i|+OsTdX7Fjh;sza9M2JuZn@ zbv}S#7xBi_nYM&P_uVWphSIuCZUnYjB!_=DViH`N-~TEN;sT{X_rZ6Nq}%RYTbNP^2P2YDVh#L1QDw?W%g{a zAZtZ^*;IMSG^dW6L#OZBdXFRCb2-7wWCo*>@3o{4a{c$2JdU|e2+=M$ZEJ8vOUDkM zLEMI8tP=H5T#{kDfta%94dT$?V0IdYKHh#7I`C!c?ENx;U!2PS(kchtd&)Mde0LH! z+iaI%7JMgjf14{`)hz3IA_o&c-&%9$J3`i4cNoo-XUaJ}3QAtBrh3?#<L) zNcq#cK*4p{qITd4cXSBHn%Pn5rp~+>bz*t;#=IBxrISmjuKoFh{MzdhjQ*>B=o!Cw zruDqKBDttXZ4$pbVbii(TC&Nb{^u+P7~;kkVL7ELx@r>w^(!4PGtxxggSbaPYe(RQvSE zs)3K84<_@H-ubq-`HI|DujyMv{X2BF^TjP2N1G<>ulr&`47!ko_6fm1x-&=UFN%DA zv|CQQIZH7cJi@au$QEbjH=(Ed4P!&1mF0$QV)b_Li20y5sdB57#L3%Ex-eBAY9aIW zK(2Ue;y)suXDFLTBz9Gbr>Z`tlojTO@2x0Fjw&ENph6N!u9yKz41gEUVZAWT4u63L zT@3*rDN-U-n6P1wM{~2fNzH(;-lG0_b1)mAxR)$!_O;*%%v;gbFg8&)Cz$xM`qrmF zp1pvUi#_zJ$cO@wje5i zuNjk2Q5ruVhVmRXtt8hAs40ZU=Fu~_$M+yZ8%(6>%;8bYo>xGbJGlgKz-bXLFqXw@ z4xtJG%<2pXUEmrQd#pYyfb5|!m)AVCPi-FnU4#$bbq7N>LLCqTIG$ah>T}qaic;h} zTUB;AmSD8dvf6)pLWdy5fW$U5F8vt@2|HhyJUeE+`pngL;LTC;eCQ>EWj(-QVn_aI z#j8UsFVX7BvaAkQRZ7}bdEyXs^l`Vv5zMNqKG%%JEh zR$W}^-LA=Nm!v%nLBqZn>Of^yl(Mqb;5Xqbpt(>S@-fTc6jPa6<*TVb!ZAj2ubEQi z4{0VANW&VbCsM`L`^@?M=hI8=Y&*VamIudO9+2cWw$WDB)><93N3C33--E+1{f%R= z52qjcz1V{?7B@rr#Ha!4hTe-3Q!;{t`DOm4QM%H>f_xjNGKX>vMS7k z*|QqFPQQ^FeIkE>j=JkFJ29L%5(BgI52xE?no_|_9dF8xIc08TeSHU@;(i8noD3Sr zo`pAvh|XD0Thd@r+)FAxk(xWqAzd-f1BJ3ZyD(9z{8KG2<)>f~8zdn8cq?QJhh@lno434>wNgWCk+TC>j(y-rFI``kl` zjBu_9LwhG4v2Vc9RPJiJ`h3&=FX{$#Boly5kiOqtoVa`b)zGnaZw!#M8nyR(>!q&}Vp7U!-1q?{JpC{E3y5 z{K>Fh^F_!k0i+tqPpb7OguB5YJH3K_{he}>+$R~82>Xo!uzi8ag~UsA-QlSjed_!8 z%&7XVFRh{Kdl26NGOxG@OkvkU0Um9yo;88*A62^wq5B%OSQBR* z4&%9*q2=1G>g}BNnR1h&N;t*7a|t&AAb9Ud0Af#cg>GKR=Rfr&^E`flZlTllS@m6E zAAp4xB9wB_I1NO{Ts9R5$YTIn;?2>X>GMFvg@dE2 zV6wFYUkCat*W)N=)$0{oGZYay5YOAKvMK%tzcM?Jm1EJbS9duaH=?c5&iUF!X*h@* zHt6-C&ocvdBOV|IwXW)TuPEyOW;Q-sJNZZx)?$v|45y3pcxakYLL;SxKTtiocdwWqBayjsvwkw6?7?I5G_;yoKBS_lT^#0Os z?XZ?U_2_f5@<(_5yM8N1t5`bnfeYlB)X?m>MEn;__AT9-wW%!1G3|wqCd}Yfx;mwf z!nHVM5eY}LgG>Ef(1NC1fKRB4b-1<9#!n)Xp^LdaLopJJuAaMgK3RYzK`7eE9he8USD%9}KPDc_keyU011XMWf( z_i`^^u=G@2V#;uQ40<>vmHi;S@PxVDNFiBJf|8u$CIsLe48>gAw?^H!R%jK^nrxG% zkCkYIJ5RJa+=Yyq%zyPPJ$^#p+Zx*DspeptZ4*4K*M8i!-@86B&kv#Y>;)*5*n7ln zRPhKn!mzYv5Nix;Z`<1*#e7`J=Ca`WWp!o#{`UqV$e{B7lJM-6d&>(#0rwT z;Dg=^?t|Ke>+vRw!`{)JN*#7g;GFDo_N*7Wz6pn0JLBw|4^_|f8}hJlfE+?bj!$Fc zjI%=Mg8vr;Hh3fvYkB!o$BM?|MOLp1FAbqOR}ot+pLPAhZ7hZtk%QJD9;=X{d+Y5Z zAtoDSAa=&ZqT$4coM;BNe*E%WKek86xd6@VSthhOQKel%EJzGTaC;!;W$Anf)3;kh zTEfLX2@V}>*qa+ZeQO4ulV9ljO7*O5M0i#Ayougo1a#dL@!P?TWEJ7hKIgew&8Wx? zx4nV9R{H6U3c4Da+GlNopIp6FR5}TqYGVW{qLVFuS!^JEcyqf2YH|4SwOW>!f3%jM zY-3~bXs>7~qs6x{a8g6-PF`5L3a=;hqDh>;qscbOs_58up(L7j-EjZ5M|KYXEBBlBDbGSXk7S~kPf)V5^AGsm8(kWcnQ4Z7u^+d9>RCLS`h<<)v|=- zzFHTpl7sC6vGuEkUl~Ho=CbHp*!fS1)>C&2y5TwPo_%WHG0&#tmM94z@7ViUbG|_u zJ?mebtfs1C=dZqN z`k|LWk4*Dok{r+j$9N*X8;(GQx~6R6;5Rto{xtG5%Evey(maPQCl(CB!61CN-lo{Q zSq*l)Pan*^lQ-zYF7j0;Cs`>GLuYA7`?;R8IQ`UYI!5(FV3JLHbNhT}sA~Dg&dDwY z?uE`ZFuoktN=nSpONjTGv21?_~m;hU%rC#JY}% z)S5r|6^IW1;HUdu{{Y;Vs8=u}+!wK9P=tq#ViVr+u%Fvf4eJ_Z6*Dv=4a? zXp;OSOHsZ*E4X@Z$lkH2$dVMNT;G)-h}^$~1Y};atO61idpwsc{G=D`UqAG2*@3k- z#*(zWpvlIdk=L5$J^idb^C&EJ%EQ9)O_qY_0Ofc|JI!Zu?bI2ZG2iC5p~;|_dCG@b z`XvA*kXw;EntNy^)@gZptuoK9^LAjofqN(%i~{<`c!C?~&Vb6&Dj-|t;Bz2xA_zd= z1@n&o02n*jb!o3g^bAdQ#lEKHtp+5&UOnE9f#lteOw*Nf{rx0Xw&EV!2_cdj+MkNN z<~XdGwp1Z8q(uOt8ax#;?6-1?Kse5Ecsq94qI2t{mc#xs--BG!p`#5-noTFaYkFGV_ayjQIKmoRigY==!6QKg zFtpngmf@GYlKrY73->arB_=wX3BBp>3Sv%6Lm%NYSjijJ-UD9NT0m2x;h}ogNym5n-(|C;qP^T{`pCZ6bC#%b`t0U{v(UKEMfh;haj@^v*QPqJi4 z(zG6|IDVp534x}EomK<)H!J2}b7iWMkr z6C}P(yce}VI;QauL8NH=_`M5b*Vq`~zkE8~Ye%ksFp$#OV3EJrLawrmG6LG-pI2`I zctq?w1uAdyqxKVnp|DwqzLI9}8$Net-u4`6J zCYOg&wz8D4uFCs%>W@y0(0sMg(&7*nhXPtN6Nh$~H=9p0+#$A6uFVC?5r-=%5|qtcO`UqCU!h+y zbb|K7toM4~&F_=TFSQSmG8&OaK1aCHY}05>oH<&eCQBlQOUx`&qZHKAyq~tDQf2}p zwvMR#2mlhXRyihStnN0%G6A#!Nh>))0ZQ6CU2hc8ktO(DZOY2Ka=OrUPovH$ePJqm zVW9y?LNd|6+GJLrntUZ)1gQQanFx6%3f}`|3k-+$q^l1;X{TyS@mv5Pf&t|$mDAc# zu#X&nl)Vw86wjHnRYp1QOwDlQxtaNGYV`xfJmg`@7 zut-Pht50Wk3wlS;5i1;Aqkccrb--9laofGfC4ToQo}}8nmc<3A_^R>S z%owp7w(Q+R%b$i#0w z;?c(aAu7o~pa4dK3R9F?h-RKA89%~49vk%bMz=R=8jqx!?29Z(i_@k$p-)AY{p!1q zyP=543OOEM8ZhxAhLhSW%MBc{ns~l|bh9a63HX+tk>oZRhsPjm2RFpVk>il}$l~VR z@n;Mu3w=>177fQV<3S1>fiFWy&84i^^DM(X^1gC5_{F$;ky0pj7L29MEk(qCoUzoI7qSC?`< zL!;C-f*mgijpQyVtTzH{GK7i{ZQ!xk>(w(@Q0_f`>@AxQCMUdQmqXXlUfp+R+X$Wg zkbH5uniIo>@2_ra`cYY}>cD?T=Ds2jh z)h2LdsiD!O>9+PFHCHI4_wX9dJ|w6Nrl+()xRF;o87ckHQCoj&PgaE;{pSI|vLtWA z>adzdxHj*?ys-$e3Po!|;LnT+7ngadrX;qv3$Oa`wZUXm4Qd+{`D?F^4LvB1=!EcA z$=;)9#+;~}0PJCFir+*@MAA)my$?XyY)LJE%_%$&wzd@d@Y8;luzfu$3G|l5aP&D9 zao8}M5PE&=^uSc6hkjN!>&4JFku94ECuH?U#pe=0LL@&H6Zz@tFG9jnVw+#7+>xs9UdrOy z5};*Q#ek?$Cd4On_kMWD4N3|)z8QMgz~LVZQT?i0eu$fhr|~So>4z;^KlrWz_(|~PRHo9=q$8&RK!7{ z3H>iiWW-yJ!|?}P!BZj3=+S$GlfC#Ll|4Q zOuXGxZ$wirH?;;nb%Inc=TbDtl?bivX3#X#2f){w0Mo5`a`v~)W$(2X9(Mu4h12q` zx_vaufROEAHApc0sU@3?yZO%g%Dhm)b2@%9{@a>TBPG7qPuO|_aeJRuLQZrN$kyg1 zWaE>+Yy;_y)CZr-bT9KRW;8m3s~7c6>yRuJwt2^j(~&@db#2-ln0WnmPu)?xk5<>6 z)1m83sqcKfN(>Jqh!m#mE8vkZY5V3Ci;74bp6H2P92Cs*!ob=)qi2!G8e+PYkUo>c zai<9;${t6v=ZS*=a}5R8!pL4R;vnDZH9Ktg@tWdL-t*UP*p-aabG(7W(aqb&emUX-b0 z^6ty?A;)DrH5;Mtjz3DSl=_K6rO^HTFM3m6XW^(`D~4x2A%Kk6p$mMEIOZU|{Wpr~KXv=nT*;mWW)1&?{qeWY_hewqvKWH> zlVtZhc;f)g$Y##d)3H_688(NPMh)*l1oT9Pn#&9IZ*)9ss*{pB)z}f3z-q@0n6>7~ zBV0g(=u30cMzr8Rd3pmLudy7!vTEiL#?<|#YS=ITNCTX8Tt7`{WDPA#DrKL>)BjDO zM~{tS+%!=vkq0tzU;!b)X0mnTQcFMwgT6klNNm;i-A9GSK?foz++Av(49tzuzl zq}cy1@+A|8RD7kRlLVSM>d1AR+*TJnEUNl;y~Y8-*E3dG!3a7TxZdyT&`cE66_>r*`O%zs%SYAvPt`w-W6){lf}%3OFmOy zl&=YMv7)8a{>jS^;H;xj#P({yv>=_2v_Sg&%xv2Hbnj``Ma|neElrN^CtWwM|3fs8 z5*0Z5pono=o@;j+NAAqF6p5B=^0shFAah#vihd|`k!_+`RC*e&O`&EQ9h9HfLmnSm+A!8+D ziqr}tp;XEgq?Jo#V9K+-qeH*Op1gDMO^-4ym^9=B6I{ zG50wu-G@5bOA0rfx7qOlpaNdgr*=N60aGPzBAOH}r0GSy#v>ebB6b4QeTSQimz!sT z(w8=+M$DS^--0mfD}#( zZnOB_aV6LwsB(~{mq3etMzwIq>BA}6%4Xmh92S{$|6MYr-jB!Ao1 z|H?N_km#14y`qfMSpDw3^RDVoE-9*`KdHBpg3s#c{&ENZsr=kO{3;WIq#Cs*@Uqq{ z_TL@xfBT0jae&{SHG$w76UYDX8~^6U|Ff&VzXrT@+^UrE|KLWqc-dQ@(tRZC)DU6aG!B`@ehk z$BVl}0{qlZL2DP5542`-|KZC1o8bkhOHcwbRLGh{z)@df@az9Ij{ow$x41H{5yzCw z8$?u-|GM}ojnH2*utJWt)mK=x;yxLB_WwZC^nYK>|6eZ#(1Gm(bS5F}hPV%R5F>=GJ@zs`1JB^* zqxrJl1}&q_<0pmx>LLtK7YAOoKO0T_@;3YJOBK6ui@5unvXa|qa*CLL(J0CU081Zh zCpUu+1{X2ewPk-4Y+d%q0CD~If~|k|Ljd0N5z(D=`6mS(VH4k*C*_SHg*V+uz`wdP zR8fINj@YW=)(104vBO~4{y9e~;>yPIf&gOt?g|K)|4T#m_g^U!0B<0WjjjL;g)#;6 z9|Gq8QUsdj8xVv(N7_nEkFe)`GidG2fRDR7Q6O&^Ecek;Re}sx+l)@R07=zLPic2k zxnMM6>hbLaNYZ!Ri;TZGNBZZ%F%ALnM0wqW583tQSHj+x>9h5oXQ=leJ@ z77X8a(7f52b&+ zxNOj(M9Zvw^1|_0=pF$5XB-dnoaHsIMY!a?hcY>NO69BRz@atCuT=BAG#ob*?qeF> zaB5E5&rMv2 zYkx{6sniPfq=2j%o;pW+)2b(EQt_qa+?$wd{>w?~tVm_yYlVUe-Gbcat<TG6R9e$`>hI5Ib7y>sxsfpmwwesED#k4Pe=pgM78Hd>mv1HJZZZ zz*nPl-RfJ{qtx>*0GO4i3X3EMi(Q)|OZ&FJ{BvEzZnHMA&0MR&TkG*@e!rUB;NCBu zljV?D&HMALhcK!><}l~ZTN7HLYKutH5%Bm;Mm%#(b(s1uE(>46XOW2-gYNhllV8$76G@<%=3+Q#+zCA%wiYxvLBR0wv%~GD~Wkb%CCT9`+jn} zPs@bKpA)S0oMO$kbG+c4z9^&r_XuFu+(zO3Hc%?_t&*(=gj-)irnLe4SxpKo~8=YLs= zZcz4ey+vH|tH`wodFP*mx0V2F7_E2lx|cLz->)+#C_ep9UM@ifUO z8UM=<9~l70Efv=fHBh~Q92C-^m$K3qeyIYOPuJShk$#t3N{m-3mU$!q&6MFumbmuK z9;evj1>Qyha5SgGhMBMfpaVp}Zj8n^F;h_Z;kG#ScMwRYcJWaxne}Ka7E`2$2Dhxb z_rtwRgtMeh{U>*AlVu`1n2U)7YH0d3SlxV${p6WmJJL7<=Y4PZ2pW!`K;Q7ua_mjNTcZW)Eh}2}Mp6CPXe= zgSYNi;sKJ$ZJ*D@(NBkl{qzF73F?8Y*t5k9L%+LD0+f6p=O8nhu?KDp(e z>zRx0Q9e4u8O{o3a6pYqi`811eWw~{+Dv=fJ1U*Ys11SPWgNf`bF6X3;!3LeNb?I% z+-t{!W6zWr;ikqPnZRZ0m6VzlhU{;a+{$K8m*>My^MPs0_b9i~FdMv;oRWKU57iC+ z3baqLZXcQ*M*!`6>H>5CJe^64i9#f{W)l`kB$Y#D_AGI8C=qwEs8=WbrEIRQq7AK% z3DcEoTP{K6MPz(2CBggy-edjaE16I)lw>TkP4wawhr%I+kV&>j?fa?fdf6XLjK5xJ zmf+t~a~?ww033141bDP2)s8@0Gz3)p}(<@Qw+1Xg%`Xne@tn+l|1int4#hM`8zKoZIV00ny5yR;OolBa zrd);y6G?o0DV48blwZ(tspybmziv?S$$qCnHK9TcLy+uEM>YS@8)VXZAjP@#P*pW; z<>h5zZsBgO?FJJPwkg`%+iPlT9k}01r^Z{jyIdNuck+zB;~^XH-uFE}^g2Id9*%0a zW+pfMfypLAo9-j_5(!g|1CJ0)TYp`>iK|Q7OvmNtci6H`J}rN%I|;kYsq`@HZfdVJ z^ghI_vs+5~<-OO1IGkEapYg4hUZYT3rrq-{s%fN^9sIWKuraOe*!7scH3>a+T~8(b z8%WM}`-7iShOT*c^NYdXttvD3x?f)j{rRjS{wH1wG*vovWVm!S@V+h!Gt}TNzO>L~ z*C1Caa8r^>r`DTq_8RhX6V-2f2{tlz$-f7#^0|n8KmPPn37XGw=R?5tr$dWI3g5$d zz0+zmM>Voy5%A?rTIEri{qDw9c(7CaaQwOkpNUzCu(En1B`n)7^Ptt*()reLpqL~n z5FW(~shL>?Id>tLjeKyc{)QZl-+Q6oyf(DISgA%A!`WapEu$TktDx^}zwVUje|4yH1;H z>#g&#P7c4@(@;h|H$Ug@(}6AK@6vQP!O@5FHAm=m33taFa6$gFupWgH)q-3_tVBHM zZgy8Emk|isu=xPKUYxFQ*XWBNzy&Yl!rv!Nc4*~6uQwe!7mY&?M*~pES1JRy)M`%d zBffkaYU_ho7sw|y4h-$iK;|KIUcA2NqfLdDasB47CAU#mk2}qYb1k#s#5qXESb{I5 z26|Hi<#s9rVzRouzRO*jR?)OPFQjbbW0~UV2Sq8IPANhen>gY7+?jCsvK(HtD29@1 z%0#iLx>`JvNaGVa#y)o}3$1}^RXKJFH7#3N&&(*7I(?CE!4p0P{OU|$($zSN=bnx; z1d$I>nqIk?`U;c{VY%-hT2?-vhvmhHIPNn3nYoA_8tce`kVMR1Y-$;`KxPwIogWOx z>u#*QjgHKZ2iK!3NS${!nGLyK@!XnY@$WowmbngG;#q=33{k7m6cz6_6=$N~C_Z%D zfy&=X4^U{$k$!?3Qk{d$orSXS8Ump>{!fr7pCF)$K6(1LKLgg}RuZ(dlGT(dzaLLB z@AzH|t01_EUa>u?_h_T^d38Rwy+87@fz8FkuKCDnvOrAS9N`2 z>O8P>YQCCF`@Y+Bmd0n<YTZ!Y{}=%s=yqNQLOMRw8of@mMKV)0)alNv**^f zgPPs)&xmd6-O5|F&EEAdCPH4A9I$1PVUUZ@#VU&8_JUU%M>noam;lX-c?fPt-5M}Zc9hPx*k_@~l#96oza)_K#bt=C)mCJ=N$&~DtT-@K(B^Yr{Ztm>JW?~Hx39WH4vWPO88 zYG?i{^_4T`+t0b1Q8THUta`+bgE1W!XKyC+b(D_I0-9iNleq1^Wi6fe4QZv1?ytt= zd`1D7GwQ#LAa@Qa(kk#+&(FFXPW-ZIn9}525VU^&e$K-!6q{!qtYdyL8@|=D9mMdYw$j>uDkFp?xCHnrH%nPFPe0&@Zr7(JVT?< zHSHJMO{CFw5w1Y!94v++d2?o)TM=d$HJw#%vY$ho*fGCk2Kmluzxyh}&J-Jd=BG~r8M*|*wCX5SmQ6P!nT?Q3AmzEfa{ zqej>QKJSji;e>z$;^%h83USE2#7KH5A(GiGPl55yR9;<6p%nA^EjoMKIoSP1vXif1 zw=d}}w>=oDbOL8d$hiF|kTB)VDd<2~wcSZ&iSbCQkae0JS^7davv+yn;CTH+U@7uu z)oJybo;=MIt0|DEr|_XgUHf86Q#)qfvMAF8d2zYt`P{gzuTG~T{F1_o+~Xuizf+OK zNEc#|$F_9S3VDIMZz04nb-Y)2uagadYZ-W^v@Bxf>$Y7wHz~prC(1o0KW3%WZMS%@ z>{oT3mMy7|nPQR?3syJrW!t1A+v=VGJ6$l#nJk^)|GU#oltdv!Z1(fr&mq{0-daNz zWuW@1OUAIznP&lL+BmCi{wuj{uYI<0S}xA4Moz4ZfAR&+0uB^nGPZ&%t zg>DjMdM+lw7B-eXRQ(`$Pa$r0!%)bva~2YNvRkoWd*~bv85Dj`w@}DWzz?6kGzg4| z$4y|;DE%y~*Y<1kX324YtXyWU@A*7amdsV9%O@#jDgzTQ=*!PoWe#EI6XPga>?<*3 z+`aMaM!(+L_$27LZ@eHFMNZql3B(pwZ}>&yEB_f=?@bLyyRB#^d12?>O6l#~**A!5 z)5KODw#aiNu1%0(>(y#kB$E%&Vje`rd}a1;TmbtKycC3BXn;}o!Bx0gMOST<=k7KA zi12hCPv&5nfoGW+*KwA~a=l>{;(-v2JdF?YpC(BTdlR-8!tm>kD$j(dsC&U zFDL0XaQ*P0mKyoOIo^hWpigJ~6%ISE6CWb7$AVE}n#CD!WdWl>PG#7su8!z=>eaS# z%fA)9is*zXI(g{96y!RfOtuLG={I?v)I@TDD3=j;edG3T@NU7Abiu)Puekj@miA z_;Aa|AlLIt;MhjE{`l$h`X~}Z8*uFx%MUnUS79w~`(0o3eIJ*l(oC0J@8fG&p|8=e zG%jkH)5(f~!!7PtFSeuwaHL|rKH;|BtwcUI4NY^}#B$k(T1HkD~Lmf8j@+&NlA=*4adtfAzI++~KzKRq}|$zXb*V z9v{e|1)PQ2IvvFOXGB~bO4*NnmdEdW0lu4(b&fd6KvW?Q$IbhMz0BK}t`F@^$IR)( zwN@(g%*3nS*mQjOI`~5~IJmu743GZle$PkXNFF85?TyVD>_RGN6CaC`92CH-DM++F zn5ytqu2C_Ebd0|U3=b>js5)cN*fDzU^TzWYI5JAp*UDnV)8c~N7hJXK)|CWEHoKwQ zRVUNb(oyadIJ8GhHf`-c2Xw~lC~Ea{m=dJ5DBR3$J`-a-BrBj|+D}UN+}RTE8EX=& zGuWo|VuXte2Nkw^B4hr%#yLcr6-oKHvM=`*6CHB)v6g5&-T9uj(qrJVA_HdC4k8GF z$4V2maO8}M+%=zzk_*12$95H5^}hO9V}6jqM;%|;QCRXSjFO!ngp;UsoZnWMLZ2kK zF3c3uL=jKB7{aQ?0^o)%7C7gY%9>p5tl4a}$9C?&4epmHwpH@{Igz8k6%mAiiIs0p zxTEN?LKj}y8*Y2)j-F?Tem^K6AY6Z2EK0G{A)7zxHpi{S?PzFlzutDV+gA9CV@Q@M zZU8I@>q(TMk$1DRg7smhr8YqfH%q^ZPE?Bt5mo9kPtW6&@YdT{2n`F@W#0PnpYk&K z$jg-L9Xrt%Gtjj4-o}+_Iz9~wIwLalnR40pY%^W&_}P6p?PJJS5C>=n`nAHcA7-m8 zvbqN&dR%9CoM91rTA$~qZTVY#nNO_;@y<;$iJ0P&To1>`JmDqjncTQg>luEbckd11 z@I%jeREnfz3*u!Z(FA;$(b2Iv!rgx}OR!fgjgo#S6R53H>C#?RDbLU21jQHLzs zZ7gL{4FBm!EjHS74@WK@WY!sH|8$BxLpv`r^%!I{&e08|-OgKAx-4(r^F|9)-%*?6 zDns?8J0&i>#n&2R`g86fg$6`00lAzrq6HVuah2Kd`P88n^RuSpU(m0U%5{PZeHQ|L zm8%HMBv&~MgkEgJ_$^|cjNCYih|U*q`OG7|+JPoNUKu2Chcd6TY+$PxK1#5|%<)SD z5zMbXpY9QVa8F$qeqYyP6xdsE&+lHgc97kUwjFuNt>o%+VK;TR(DcB;d-cK6{iw}C zb%fJ}6tnqv2ZO+|KBql-*a_~D_a*s9@>F6pBRQ2#Xm%?qp%@0Te|5w-3Bvv-VS0k2hzDmI3HF8NC~yRT=oh$3-tyDz`_L6x4wlj)ypx{K(1Shvr2s!HyEPBC<8@T?0Z+Bzr2r&+vc)wip8e&YpYs>`Mp0n3}z$P z=GvL*vp1&{_FblBXPoaBp3qQjMbx`Dem~ajpz*SK2IQZN=zKZ}>Wy~pbnh;lzs|=o z$OT5bOAI%fpT;Q*jPHBG6;MT90Ukaa*08UCkm^0Qv~H@3dgc6^lJdDhgLTFW47Brw zvsCjnL&X9kxz9qdFY-~o7CU$OVz|q2;e@;Mb304WIn-`Ntlb&*Fqs2Wd&eZzT)vzJ zJrETW3eDR!Q?<0NbspWY=fr&S=ZW-j2dOYhc+-^iAIDq^!PnYmP`1{~n`tqbD;FS4 z7$!3CFl>JQd>fjk{P5}ex$63O|J7V2|D%>CC7|Gi$KSC@m()W{x%WhXgkA|W5YIBJd0)ERq&yDiK# zKf+=^27bw_CXePtxqZz0lP?4My%B?W@B2WPt~-l8&kyA&^2G{_hcD0hd|0+#|Kq{^ zFCR@C>;HhYIVKc)WuPu;pMlcd)R}6j^a9L_U<0zY*v_Cv&}<1#(Brl_qluMH5#C`! z$rda>x0WW;TM9@0z`k1j4u>EQ*LG6cG#nSLtlX#b=5yB1cA4VC@F4r=>_e6>*|#g5 z_Vxnn1#7aZ$#qW^sH%`0JZCQtmN^-|KYCX+;Sun(nsBD)4{eHUh&Mc8W&TrySi5P+ zOdlS&$Jia}sQ=+fw_D6;YzgVyO*0d_hbR5bxk^j%NyWF)p?%}J^T%{7KUujl)fzlIQYxtyS zqk|M-It_kOp&s3?%{IB$)>WN-ETEf~X3&5iaSWR4;4j5k0-9!df6$Kz=m+-q5UlXe zXT?cU!S+UUbLxnqY^8p|%v6rzR8q5NxXyS&%Iea9-`hj=J21N5o8(-}YaFIQCDXaa z=lw0iw=Ym1N#^lQjp^EGDrfuDK<&s8k+{2nwp?c9o5Od8_}ZcEJ|=yO+-sMrfgoA; zR}{~vNTR66Iq$s2`1Qu@g3;_CC7}ZPR3Efo(eg|SmQ@b7fmWYwzpJu@^ebFMR&N&| zw|FSq`(cP^%ttx1ecmA~Qtp_<^gmg^h3=koRewf8?zC^6bUB_%&nIc4d_JnoQ{J_*}5=b|$ve5LTf5e|{Gm--xLfmY=ELn>;i@Dq?Q zZ6X?J6EP%UtV*+ihzy0ZRrzA-Pt?t(9dpfj3B%Vpg8!e46as4er>CCb5GmPtI1eB_D zQHoUQolpc+y7UgxJ4g+LAOa#PO_1IM=~Y_j2#C~x^iHINo&*RzlyA9Dd7gdtc3=DZ z`M&S@mkU<1)|z9?G464XGRK>?j0OCbm|E@+6zrI-%Wo|bCDcA>o4Uh6!bT3LckH-p ziTK{kHkRyrKxloM9LV7#+u<=;wwc3=h0Bwzkju~;$Gka0ivIF=*{WQYLVW7ecBg zgjWiqX9;K1y6$CZOP>i+KhgOpjiD15p@dTqAtE_>99QL5*c>{6S!*j`FWNfz4o}M)@tTlZoFx{ zi|k;m`NMDt?SJR19(YI|M} zL=pLs7LuHQHe(nsEA!PYT1akmLOK_@)wf?|qv~2f(-{^x7RrjLb3uDvA)$N;*Ggq& z7Zpi~e;!4qs(jc#vgbRy+^HjNm6WAq95}0^__5Jv%DvM~Yq8&dY|_tLc{l#BBnS`F zf_0c_9Bj6LGKI^TIch|~RCa+Ckndr18bOdJp+m{l3hklCN4CS+MoPmc>#VYa4+9Lx zx3xD^OCd8fo}YZMlnEl&pMvoyt;};X4;%_l`8CP4X3bD<4_q{x&P=rkuZCxBZ(_a7 z5&_lt;ymPf?^O4R)iAmak@HAX)RSOs?x}MTZ$^eaH1O^H`iNsINyp)R;mRL_p)NP( z3Sp+Y{7ceFBktF0avf$hXc_w4oK_GQEpYMRaA4O1&}yYWmp?L%Be0KE5|1 z8)wU0-9KphUFMZs5!bL&I819ip0#fM@z4x7C_pe@mdZ6VlQ*w`I##GGC|-Imf-$wQ z{D{Vhnkq?0*lAQ9G@gU^m=&@iR-|!qg->C3Fq18Ou1B56-vf!|yZ;W7(LIK(V?x<5 z{tr3(&p1rR{a*B=6CQ^wSSfLv&nvV>2u77@PgA>27b#ofh9IQhP(3O_hBBv!gNwx~ zm+hfiecGllGK(c4?Af}|GQchiAttuo88X(WTtoDHo;iHuXi}~DQ5wPp2UYJHB;@H} z0ecS~>`viHcP5#aNP5IE@BsX}yNw|*>DdiTYWoJD_fvC|Z}KXa`+UIcW&zueD>q7W zW?e9)eh6cI48KTO>gS!8KZ7(DQ99GLdkKs}Nh=LSU(YmG{lFZrY^8?N-11kWsqgy~P_`=T4b)CtRiDXic_vOnJWTs$y zYqA|o9j@Q_=oyyg-jAph_{K-e5RWU42AT?yAoQ(-jD3K9zAENa&FxfBuC_9Xx*N(K zhg}HuXb4b`ias3TvnZOewGFNlm+SKI$!#DuT zAV~rNMm<}n(}Y3D=H=k|R`o|&@k%@wo&7=u>v-}{E@wa5TV3u;$v0x_Zb7#f_*QTbP zrpmFYpeue+O58D9X`3dA{a7(SNc^8*mLwVNlgNsRryhPv)8Sk6EKk2!hhX&GOkaG* z-ye@hE)nVw@Vc#AqB(iFycyg+`%irzEo-r+7~2CV&v2LLbp=T7)1~*e#--l@b++L1 zpG&>HTx&`0e>oKNh?;F)LhJf-lk%=N4}mvPamg@+CE2KYv^nmhMJ58S=}v6u6*HmR zm)Xdg+n?n>XXX@n&X1iBCYAcsXt48DQcsYas}1$Y`((AIG;KP0qu3{rOREj`@Mh>h z!Y!o7ry?l0gVU-IFYTe`rF+m>Uk?dtDpOpDYI;{EquJyuOPj~24YnSCxAmBjB$zbz}WHXkN&4r_Q_@-{7xzD$? zykvW==WSS}>E(p>j~O309%6b<=Cs0(LtVo_x*WrXLXL^|V>|H_VgVKW)Hn=qCn5t< z=TDhl;L<*r)hUgW^phjXdY1$zo08354SFMe&OML44(TbJl`4=|JTN{7-q7jf{C=lX zK+Ogau?4%tAU8{8>SiyfA3`lr5Y9ec^V}~?)dD?G81RZ#BkPR5rZEI_b~!8sU@P2$ zqbQ`9lk8<4rD17mndgIopKRaZ-Ox@n8_xO=abZW!unl&#-Fssa(Fe9`rX8Vjsl3W| zzX$HpfLeZ|%#RT#iNI$V<_yLn7vd6^$ z+vCT1s-OYUp~yx>z|mpJUUn1%e=Y&oJCmnd~G z)6Q-EHt`lmmYQ8PV#P-RtRtfhx(+Lj-F$1D^mq~TVCNc%k?Av+ ztd;QfM+Xer^aZa!b0j#=0}M@*1~Sf&2Y*k2I(>zi+^z}MsP7yJXy!s5S&AvxD-^41 z{OqpRYI^M26v&~8u04kIDR2n+7Rm#~0NIjVUSspPdgIVrF^}+wbKZ1e0>3C8f37)c zm0Hg7B~JMxVFqDo9hJmf%;{*U3M%ydKZp5)Uoci?lL;POW>tKIvX|{k-+@ zM5;hzl`uTL@7}_bK5oJ`pF>HCI>i{Pc;3KS$-P|<*{V2zcolrhkhaLIb+`j2K+Geg zX|Cs=%!WG^NX0*8>q4hV54sOGQde@)8O}-8vD<#h80)O`=x_PTP6{~`dC!}$UYTjq ziME+LQJQt3@f^!M3*{jTD?NhZP>XV3K%yR{btV30dWv^&02qyblJR)uXfvvv;-8t= zf0M3E7lKsrvKD+a$(p>Vt|3xNtqnb{A?mi%{O5BN8Px-7vD9XMqg|Ftxm$0P7`?1Y z)4sh+u~M9I&okoP)?&7klql_4Q)C>=erdZ!&K$6y>wYR1#OP|z6%sf(zSD||{= zsUCiXk;II_B{)Ox+IV(kiH)$!NqZ>u_yuBvTzz(>qUY7$$O!{z->b%kw|DQqg*nH- zry;{|)GR9_+W4aEO_WdUzV5@Q% zPhBrp!1-`*X7b=6DK%G^hqzKGh814wQ$V(l2n1D{qXYUKg(tRw!U%GybHvPg5>@kX z6aMH|ZR!0YW7$yGy$%EvH;a5Kg&NV))_ICmSq$CiS3caw*}9qW z&hit#s?H|icuqbUkcMm|3g&UOB1Ri*_YAqke8MKx>uI49!f9gH&%)=p3q?Nos}73B zWgF!_z!HjNm?;C%*E;;CQu0Z7qUV`i+0VTynzsi)=K`H{frYdb1|uPut)o)deDiNv z!T>I|7XKD$6V+i(nkx!BWX&s}lSKRJk13o#g`s^Tnc&=X^o0m9BzA98>I-{C2=gj? zNP{{z`P3-f;erFDMD6AQLWD~HaO_8mrJ5-Gw7<}6c?~E;nRnUJ+|~C{ZyDHa3dlE# z*rLyA=l^|q0HBnaZr6Ob8}h^FvO>4gKKfssmk@abF6s7HJ&G_hJ9~Hgf$t|GBYJhw zZU0^g1yc^ffLnVEd)7;oTo`kM5f90j+LDUyno-H#H#v?V5yJ6C^SRTYgtUEZZq&h! z=uWz0$K%5>F`Jh`wjb=0RSwUP31$(JjVw?1)+Rfaz0SYRcwqW!FWK?WPlqWb)P920 z_YmD4h&Q~mJhfripsJ~Uu2R0mn83m>-@N07JO4t{+Pl4$>D}|%|&j+c* zs~nMXI=#Vqm*BO7^6;aTw6r4U=JO}f-+#RHW$dhOk$!i6I6P2_kv!`vPV^%Sx9aYP zzYVE}DpSEltb5UnT6t!xF7kOCkQy!y2&0#iSErEpgKwCkf{^dhr_a9*sJs%_n@y_4 z*Sgu|Sq#K-(u6OV9UHg$=q# z;WD@e6J3oc*HCS}-OC1z7j0aEPqX7V)Vca^UzgAKP!GNTN7dXSV9#2-&!%_(c?VP8Ok6x@Xa?ooU=e_8Q6Nmg99tYP2D<#>CIW+jqkj3#|hS{rM8cM%O~36vjETXF+)K0WacVhgfxx z&65utloZ`~Z-3SV3EFw)I4bQ*@8J)fa>=5Qdhsc-g>)FnvZf2^Ds=m}72EYtUu1bg zSdt`)LQNYS0z*i{@Er~fAK!{#lR@~ObFH-Ids{Jg!qy3tUgi4@g)ayBvq`8MIAiy> z*)o!ecpNrEgiGEMB21U;CVnsrc)F<_YPXc-)}KX`*`eE(9mLYNxTBQ8rix9$I;QbX zOscHtR_!Ylfd0~GGv9g28Z`uMYCx{DP30O_2e>jWNa9t`plbGiR`&?e>i>ftZnP)D zu2+0;`95+IhK39IK9MBVsqf^Je@H(SX%hB7}4ZWaa;O%s#qHd z9xZ>SkBZvv!!}rY5+FyPxm5KDEYSIV8?kG?F4e!sH7LNM^E0#aq-DfIZzW%rg6w)m zMS2Ad9ka6adE*>+!HlDLyt=~2Bi{kGlJOsS$ga690B|>}4i4pwI~$_7KGd*Ej60oC zVVy^`FV7XG>2`vjZ-(RxU( z;1<1h`spA}M8Qe}6m$Q2U0aM2T*+Y2@3cm@6rF!0UA#AlnDL_;YpE&#<4`(0$06yV zgBS4u2l6uF`}fI{gBDuj(JPeA)h*0|_eWzQvu#~Rc8wknxPlXVZ!~9z=)d{{QzpaI zf_I}CtIQ<}4#6CnH|jH8R2C!IlRXc$1)C&EOm5ctmYRxNkd&&F2`e%3{;4x-AoMROE!(e4o z0G^XDB9ND62!FMCGK-u@bw{XVN0s!1`2p!+#b<64{2k1xu$@Ma0b%+QSoO}yIuwda z^MMB(3p?n`tog&b=e@pD6Zu|?2!PWowb)C*uJps7z?C;nbk2EGaN=FOicE-Y3Zg9gf+swABncj>yX&iX{ZWoNo5L#3|2Gr$9` z?^V_tctDUihn1SdKJ?1rvVljht6R{k*6 zO}y{hMRFLUR8#B|)xTV5lhCfyrEqOJNEXA?%L@mPgtTb%^z z0^l;%%>-JJ%qUAeCyNba`^0(Tc1oMaX$``7|TMS#{pH^2n$UHlWTnjgz|>WJMD_cv8s9bEQ*F z0%fOLVu*SUDm(ijUx{hhugaEsIre!p zwwIAd%)ye@^R&G*2}fR0oBDNlZn!t#oR<-u*AQ8+WdBf3!4yq$F$Cke8oIA=NYZ)@ z-XqYUI<0E@sQck%6f33;sMo!9jlvMt$6^6M|?S*CwS}g0(^x9;>gIjpo673d9bxV(*{sKuzj6UU*!G0-0zcw zM<`I~dNwx1{q8%t?r)VUpKnQdXOX`?<)heva7tWw*cK}iu$kdKF6vW;^C~daGL!yb z53uYnNoP4g*t7d;-&LzHKnA2Cb)U?_iiVwOP%Bv0iY?u(7&meDeMV^^m+jHL_ed|7 zov$rJel4Y@P2cF10+$HRI~i$b_hN7OV0Hz{0*2FCo=KqMY!Q_UM%|7$nH9ADs{GNhU6m=r$hM3(MksN11;Am zs>VC$^(W$gaDAj_Ggw{&_uA?P{A#t!tu=@E=d|hLjzs(V(>6U;O|6ye zgA$bmPkUTxY(k~T=V5`hM9avXq8UK|s)2wI5X5#JN59(`omuL*sM>OoV#pA)cLD%4 z9eRN>XaRqTq3H_qV>@XFFe+nn_I*#&a#Y?%Hco;0lSpXT`6$T+F$H@0)$P!BJ(Eg?L>euw!AhpwXp`^Q2Ihfk*;Aupg-_A7q5Miy)Z99>^pCAv$ zd_3xKWu7eT`ubv{tH_5gDK>S1fD2l;tif}Bv0I~`12|W%@LW_Es}r~n*%0kPNcpa^ zAAB4L%#0Ite?CeM&%b#YsN}}*P}lTODa}imT%7_X@j%=esNc8Qx5GewM9;vN+f0* z@RaR0ew}xdD(L4-ZE1A1CR8YHB_$J!O~Nw`t=CPoah*~+QjqX*vOFfGng7ZI{w!T% zp3A$C8!r7W@3{+&8-HJo*M8~|Uz>T|bNu9+^0*Z8NYJ;?t_U{B#t*V-;4oH~d%_N1 zf<=h;HB7a^mpQ+da;0yT?9s6EoBz0$kR{to1{?`~s<1+t@CHzDLcgDmDCaJuD8Nr~ z!=Q+Q94Zq$)+_iZ`)SUbpT&N9 zyg56Nmtb-HnZh4){p{;J+LZ0xgEN|MYq~G=A`74Wjr{b!$y*>nP~-T0G@p;$gaV%c zH@B-I)uTxE^i!zaS6Va0^zVLU(oPD{ma-vC2)<82aU`G9p+0s=OTFQu1vKzf*Tqi$ z4Xjer5E4Sx!Y!^t5v#;#w3&W+0s1}2+62c_RX4`+30C`2b*c>(c@(ra;c8iyakcUY zF=kjsXF)S__$q*5{kZ2wGyT=Ge;3&Q53Lh@6)^sSKSF%~<3HTh*C-u%f45MMZ~gMg zoiI8&5k`BPNGhYZc$w<;Gf6xV&-YucX@ zFbj)=l}T-ijah0wPKJ`R&#Wp)7@uQ{UouVJsqTP(J z{-r5cSmIm+W8g`_(NKEcamdqP3hUtubWI22W#@BqX4cp}6-uI0m&qx=M;kD?EXNXC z^i*|55+n4ad$#2nYmgLq?#y&qV|`l=O&um(^`(P z|Bn>^rv`)Urp%$*L)MIqo6T5)cMHF>XJ2gCw$}18ZYyX!zL)R#r9E87u|66_Ol!RK ze9m=W@wv<&l6(NR*SK)$8sf?eDp%?djIXx-?{DfvfVcEm+My3mz+y-+!664IwO-6+ zyajSEyv^@a<+uZ!SFW^yU{VKu9L)TrL`M5i&ADtdZpvpMtx}7x zTJCo+MdAI6r#Fp98*_dVvc@=d$$9enxb4=?rfAoh_Jc4^0#@d(T7}vLpO6#R{!9op_kh>BT~E_;zMXJ{Yzef=uX(Ab z<+p-@M^3xyZ4pk8|BhZ|ufdCJ$0gaC2@2Q_2)VXY_7t0)HxYrM`l0fw9qpWn4tc3o zG$Un3P6HX_!GDamG7rgQ4*P#pe>gmy#spHnS%FD50qJjvhnVZ8>V+@=*B4~m*_iGa zpV!R!Z7e_wQyKq2UR=1;k$R!chp**JD#M@r^A9o%T2Q+%-%x92(NFrjSNhM6o`0^A z3?JKzEt5Xp2YJ8OqdzfBKUF%jmKLS zFQeN36Qbo$lqo1nF>~S?LV;Kk^~V~f|G3+4>K?=$CG*h_;tCeEyZLu^{Kab&Lcm5) zK)o3>{hKd!=fxI%aPXDCcQw1i&s{l1nceiawi(k6c>tbY#V7iES8I#_mY=$sQ_=sA zH2{D13rw#7Yj~Z6>Tg*Hnaq5^`|>#Gz1#o2|N66E{wX_|>wt~8MaT7La`~f8k~)B7nz z01K(v@T`jZ@A!!ec+UVib^b8F`1h`ULjx>imUo98@8475uQy~4BU@W9{k^OItMI?t zy8l)9fA63F+%*4H_+M?^|62H8`h)*k_%L^9|*I za5_E4e%4COJ0M;e<0mfCQ@u&r!4#eNB{d;FDg13bO~kFFsddA)bT-8fs$V&Q3ztHl zTu95**!FPyc}Yr|jIdXu9A$H!^j>Nbc7gw{I%b;2v}&^T`IiM{^B<{K`mqi31v;wO zsXS@~IkNB9*owm9(kh<9;liNj&mX>WFPfXaOd}MN!m1+R*mBx`-F5COoyS(tm51#v zH@Z}Rl&FJ5PZ`fu*m@^_ts?AMc;5-$tcGS!{?y+g zrhh&$7DwXoD1CF9g;;QtFz67kA1&79?AvG?c6dequi)127B*CoiB_DU&F73J}8XZ9rixSoX{SrL|y?+Q`$$KWw{glA^e__2e z`F;CFv&gn}Y2wW;#jyJ}uSMzqQk&)*{-sU-_rLTP(VXB57alM?ekiN`56J;i|9@!# z{;zmxRGZj11{hjZ3_FjMTz%1Ie?bX;Ehr-J%wD(TbcW8~4qOQzGqU}p#&O^HAKT&X z_-+DmOA_^*fpzmC8XpUf0wvL(Ri;fAdD?rlS%f9`gvALMIaljgky2&j0hjn=kCUT< zNAnmx;_jO`HZtD7d&!;N%_MzTU^WOT$hU& zp@N)xfw!CrpMHKh=^8Z~jD(9}v9E}ik6^l2!r~$|ZOVRD7(Kvl>md?#GN4qcPa-Ti zl1!985XNJZ!AnTw^8RfBz@_3c3N-89-v)cUZS?LQ-{)2!?JgQzo8j!$E8|0)Y-yQ> zy`D7vxOukx`Z|f=_W(Ee4^7}itq7ZpS#P0Ae;?r-?}|-c>1@zOJH@xf zR_1?b+ju%B@D8AguFhJ_5fiKs9s7t_@Au98cmXU6!3CDY$1A4xJ}YsF7nuo2ZdM)_ zdN?GF?eV@dNa$}{t&2+br{X0pT=g17=DJkJzC>!ov+oIOICUDr?Pt zcy7S-n=zj@%^x3wSkCEmX{mtt?nr5bZ&1=%uv9Z&y3glyp^CquV``W>~Mn@J5OzP8~Nhew8Cgnayp!csO;I9*(IZ%U};+VS8uQXXGHbC z5Mk0!Cd4g%?R8QCosPv17C*m~iP9Xg?JYR%O%aoMT%hpRBV|mpD@B&++*Xdy)L@31 zVy@8}hWF#$pFGM!LNO&JtH!7KcXz~73j*3cU)kwW)BVgW$i8`;)JYGZLjM6P1#%bp zL$%kr8p2lcot%Sh2j4*Ff_l4*F48!g05{nWKISrSZTP^Fak%-OC0}(Wdu>NtuG$_pMsRIRV$-Sj^p%Y^ z0h-4{4kG}hH1__qcTG<^{Na9UJuiI9wOI%#Dw^DETqLn+ne`Fr)Md{oKieBCg1}eI z@J4k&$B%%Hsl~EJ*c9Ywmel6JVSTU{#f*V%fm-H6I$23G0KpxeF%*)io;Rgtdt6abJS0ow} z_a&g13yKPkdY6i}!l6H>h*kBSxtKC^zL9A-%QI6+YTAFG!=d|t?55Ud(qfnLsdeH) zy%+~h292>3+XiP<=+{1Cp!7!sbUXsV0^LXq3QbZY>(`N+2WR;K+p`BYsBuKHkCQ)} z)bYb|1F;m+%Kl(}d-|WIXJXm*40C=qfI2vWZNQ^iTpswx&vjHASbQUTw&n*31je2% zr!YdFhXDtxJFDmvT?Y4tu*SY&*qLSwOcWV|Y?3-I36*#|HI3@Fii2X!q`qNR@{(%x zASTe{$gL8{RN1s=``rFqfL@;a5pq=PIp9Jbg%lRt9c+=S_ zZ;L+L_-SgqUj2v~TI1+ubKoIT&xbeA8CTj?gmK-gCMN|6e*DJ9?HW3Ivx(lMId!{J zrw6IteEe@>5XV4zU1oo6Y3) zZX&U_tjJMEUb>S@v48qT@jPk}f2m@-9j=UNv_0-mwD+`prJBAxnU2wr9Vv%F z?7u8{c1J~XDu6ck6g@8dg^znJ<4)epiylN28=W^byrjtgGCwxOb|V3HlZrqmK9@Qly7QGBcgyTCE1J?w&y!}XV`pQVTWJVeze^yYsdcS})Gl*C<-l$Sh3iXC7%#b>fwM<6utk$H#!|x9}X>REMWGn}r@(ZCxgv zl*Ya~+Q0-tgj@<+84^-W;WNH5qlY!K!Zy<|T{qBqO>=AgwzDfg4)Zx|O>o=guF_;V zeZ_fm?!l8C-=pLG4I}(<6pzSY}bzoDEbH{ZO@O#5;;xI-GBXeq?efG)$mxf~n0cfinn;$+W zSzoIA=Ag|gv7SWhEOof|41u#zvTiopat{dI>-cHnbauMoCNnfGL(SJJrf%bN2b2%3 zo&ER~Vq(U3V8`uFgtViyZ`>ezvej6oqR?=4cMhLI*2>L*@0}F=3vz({p-NF6{a+sB zKR!O83tHfToF7kPmJ7z`RD4Zn(=}&3>l!UeQDi+Xbl|)r>vq?=8vD?PFdjX@s!Xn9wCDbLwOl`2#!>K9A59wzvjf>rmHMW(#eOL0d z!(n3RdiZ?^^~2b1Wzsbv)LEh|sa}TBbXEP~%R~=Vr6^4Wz;HaHp)h!)bf~LycSX|4 zlfFsnd$zoB8O9DTn6?_t7|-hk{d$ntq>Vfmb(?<_IdyR!e&U7G=s9l3;#Vc0!S z7=yr|pUaB(H7|)%oD#!$>=c4-(Tkq48Q*lg?_KI?c|`7shwr$#>8XnvhKJ{=NBIuo zHaJOykzahixw$bFnBQEdz;jbN4ZG##8fh==*+|9-;We%q+D-H`bDO>Ar+lK;DY(8q z_0jSrhv52?OE5V;h=Y<1*XL~c#;`gv53+G_p4*S%8Y2C=*01ftjcerW{cQ)OIZti}xFY<=hx#w7 zCbvpt6btpvA%K}JTyZaTG)yYO)Ype=s%mGd?<@tCD-uayFTQW=6wRJwAg9qZjpgK(1oE@1DACTy+D)xbv z%15N*Uf0=f54$wol7jfden6GK8At%8wX|UXlC$$tx@0GYXRpk&ZUuE(8blp*832Etj5p?r5R`%?t=u9nbrpBs0SofIAYJ;KkqFPR- zLJdJfib{v`Y7Iu0w#8jebdNQY0Nt2Pq(nWMq2b=SnLqUx^ zw^Zs^<~L^Pv_de~pX#n?vhC!?+$LO$iM((8>a3)uZD)UfzT^Z!*zrSEWCLvChyU^B zy0%#Vb-TenSK-0PqDb&ciO9E*yW>Q1^b;wH9y9hz&%pkeN-xEas-Bj;98)(IX$q>R z`*R>Sv0c5rsu^pfc)jBjHx)&~6mNMYy1~H95wKbBrbOF`XqFza1=X}2Ly4h>o$pcB zW-d{ro40+-lm;%(Psdsk{sr2!xYS{AI0oh*UY#&m@$nhXXDe+~Z(lQKDSy1X zEBx7{H!*tEkXLr=WobaL#@G%GG!y8!zX|^V9Upmtszf$)kjVOB;IhQatO$ zLNP%1T=;JE>Ry2hKHa;j^g1%L_V`kU*l>TAnRH*Q#k>)?J4b@0^76qXry}m@BEyd2REI(V|ND-R81sHV%km&T;B-x(Ikh zq?)Vou(XSq_uXx}Sd3Jk5$-{PIPdvPQ=W5J&nJI|Gtm@Fpo3Vp z=dP3QMP$qV{^|<`#q`0X2S}2`$ybcw%V)Z?u*e3^TQ{>}G=UI`l;k$5mT*2(gT!L` za3y28m_#brn9iYc3ay&teC2urKvE=@TKltC_;tf`fnNUnOz2TKR zU)xs#PPg2btA!Y^6fMoJ)2}Ug(d<{L?WUdATUe&~)Rf)hg5bQ8d+w`0GuQTvcoiQn zQXyI01eA9w^|H4)9i2W8kui(PmBKV59Pl}3IKl&Vc=S?sD*oXV!t=t4D?cI|CW44e zXS8{C33R{eKsU>+<)XTGo5)-;GNK4SLj$txLo0A8`2L>nCg~_Gv+D;rXoWH-GbLUI zWF0({?5BCaoPs;)wfPcI861FAZxEj%+aWz#DBDxl<2VAL0u!T%JRHZo@hoHI3MDA zccstuYa};OwKP|}pX9u0^#%>wF9=ow5Nc*&%-i#Og@g!_!*#^g?S|TEW&kQbNxDu6 ze(1NLnw=et#>};DGTO>~!RWqd`8588$us=R$gDF@=flyqWw4S;F5@C&!Q=> zXXu|9YE;eU?DbJr4_u4S&hnOw!j$80N(0=3jUt}9*(B}QVoizH^<8hD6=5FFN;5UV z+0-^40OsX03@yK}w>fzk8sviu^2b=`de*aiGeTrIvhF52&2cuTCEaKatg>hx*cj?O zIUN9%PTAdCa7*-1Ao~90^_EN!`${=_Y151~>IXEElRk3=xeRV;aFDCSc({AfxB*n_Q z&Gz*wa}=mMC&pViwiKHM$@ilfYtBET0jEgJYJ^bXsMqpHc2r-NkXG}H9AS#^5SP%K z2Cmp=os~fIyjv$R2J~znvQd)0I9ft?xwn&)RCwHJTIkwnW-3z;@q0$Crb1B5*kK=y z;?XzjZU&j75|OkWfUI7t3Jds)q6?;Y8F#8?8O?;Yq}i+7o+Rv+dLSpVruijZEL|Rh4p-) zs^EwZ2vOeYe@TN(ijg2c!Xcf?1QjpvvD0eDG<^C|13wqUwPn_ur^zL&!jN}f+*EYe zO^W8mgEF)6%;cZKppWkKnaE0QALpR6H|T^yM>4lcTN@>}ieT#$uQ3+$xXUSRhlI2* zC0nTWU7`pP^V0OCdZ!Af-;6{lJ$BP;$8)u%7cwtH1|{>7M5820-?#oG0G|{CeQx$( zHGk#(t!?X_yH_F4?lzNBGp{tIv= zd8x0Qx`D)>e)jDV==hoTnvI z74qe0z?jCw=5gbjTw%7-HO$`G&GG=2n#Y;YN+0jvuIx+ZZNA^* zqu%JYsWA#IPR7UTT2>1Y<2`N!$1~%A} z8akmn`t(%IAe?uT{fOF83Bt5Z9s*c*WuI| zm?CBI65})TY>h3jfhI+B4nfm?t5R#hV~M(=0M6`@YG4Y^KR`&XFE)v+`TQ8inAdwB zR<}8Y9S{})YU~N}SmV2jAO~-}>1wR%weUG}L6!<$Npu=!G!kE(=qJslJx|t4ByJkH-5ClPILVPf za_WQryb<+M{O8!8Tm-doN~4@3`+XrUD`==N-8Sa>O$EEv(gqd=4BXH$z;N@c6Jw~QArH!^>!(_5*Gyh|B@edCk`lNM zH_SLh0%3kj;*j9vW%e9GerKy1PLwdJ$+(`d$E+8QdF1JC%l*7b7fJL+!WkltNl zGlrb$hYz)!-@a)B^nQ0n$=P!%hhtggrct)oTP)c+5buW#B)c-`AE!#ZYEuFEbbc|; zLbj-~f4}D!DNu}lZY3=TvNehgU%&I_*7y3eE6dA+~k&lQ*~XRkOekV*B4y@9hi{(V(NJbTK2bs!^P@4bRBhY4>i^Z4UJ= zGv+&DZ#+~LyF_7W7Qd#kQYb(}BY#JOf{?|Du)Q?KlxDj+iWsMn*Wz3x38%Zg80do{r^syp zsU?Mz7EiL1~aMKjSsOCzJRH$^8I8g{mE2+HY+jM{Y-4IZ|~Gq zBau>w4H-E*!ROlDbuBPx43Xg%?FZpRE<@Ayx)8wj@H;a z*>EGq!1Wrv3X+d+$mj<=j0@s~m=wT$$ZZriIe;&n2GHM!PqxR+N)6Xs4-6GG*`}48 zcGskZ39H8#U8V<1)^e81E39Xve`3lEI3N5sp4-)=ia`q_K$K8&y);8n)x=;$MXThg zg8wjsx$xOh9<31HX?PBR$W!FBHmlnAZ2(3?aT{k%_R_ItgXY?|l?+H0ydfqQJEg}! zY!pL0Z^YZvdrU$s-oXQOaU0_4^lMRm^Cb19 zNd;afe6nD8NN>KSI?D4c5wg@7u-a42bzHRcHXhY!i@&S^yTML?M`&~i1&`u#_f{YC zPiN7@h#Ui=F7LF)&|+$ops-v zdpsa&hiICuIAfpL6$)08z!z)b=YX`Fb{iVm#342%=`^Ci?vs!p~WBTj9Ho7%IrGfd(kTPHM$S*6s5Azx}Yb}2i!|J4(o3=`X z71DS+mV+4RRqczWeTf{ouMAr9yO$l-+)I8H>I&ZD?2$Fy&v1qbaL^9KCCnj+gq#!I z7;{e!3BwYBc&cNXr2zd2w=tvEWMXB5D#o4GuqGrQ3ws5L>>(_K;bl=%#d{V}uH4EL z38Ltdq)|fGnqc#`L8>-u&n9sp7jqhbq(i+Ym^Y%_Yw?LsvK{8}chtrTt~4HJpq?|U zY=A#OmS~Lc3=7x{X3l+cC535DWBnPXY&{Hq8W~>5`N2Vi;>X8hkZtRpam{BiQ*9cB zo5ILN4O5aN7vCnjy|!ZnkQi7+9%T_1l}2CoyWd@L2U0XgJ!fh?6_r7=#5*qDWEE$7 zVlDApmGq+P&hxe3hzT;ZPHiK2hBumM;0sXs4R>3cni4R9CHY2MlyvovJJ>48q{_G|nmUPT1r(HLkSj;YX4~0-9;sZCxv|+!$FYFhuUDSfHvZU0tND(R`~)W_I9KZMXC3RL!QPuy zr~Atk8|4lC;w@0~=^Hiid;KB+wZ_@VQG$J4z}`!A4PkKU1jiVdEPWt(Bjogv_pK(# zJ->}m%4%fMcp}LnriR|u(lgem)VMiSY;rDfr#9aDZYQj0;%rb}=5FW-QEwTkx)9+K z&jxX+8<*zriwhC#Sc}Q*>C*~1v3GUItCeAJodxbP{Ak_;lsyo^)=aCM{zU{!7Q`)k z7(c#~2#%UeogBc+(yef0axcXqiHmR5n3O&P!jX5DQ8jW{sLgsCOI?qBDBWtkF*F=2 z89^2O8iLIaCrI0;Z~(6ok0&n6E8#c1k^+W@m2M5pr3ka{V%~-9o^b;s##f|^pEoG; z3jfR`=dJ=bqfz)8VR)!N#ma}iT6F(}hHNhzXLt1S+!ZtTthjz%^g@}HA)|vUMM0*W z!bXYa$kj)d5v(_+FF6Oy%d!~C2S0I25&V{@8~nyzMKI#C>9$@{*TR=fuH!A*9F(spc)OB}_a~VXIX} z+Zj`?4l?R_P~OGt;L%!PVKy#lTQ6SHAuGl+4pTNmjBPx3=Q^!A*SenP`RDoT+<$$q z-+f>A=llJ9KHu;6{_9TR`dLlMNTiY<3;rR1L~U`O2u55P-FZH1zvRgAEcQOq9x>te zlN!OmF2P!hh*Kk>B{Ki?Uvpo_jXy1l&_G&QQ;JkeV$q2)zLVde>5O_uF~zT|zVpmR ztb64=zjp1$2MiM=rUVr|8UueA`BVBE@|#KvF+n1$0Z(ix&seyL=5Xr2>oln_Dtb~d z8R>wc9e3to-npa?FG(cEF%ajYolmx9G3K5Ce{-Z@7dab{1s8`q8b#0FYIj)wuud-z zw$;^~HfKvdD_bQQS%z;&V$U3U+u(3TXdW&s<5sF>m>i;PV!wjm^qAofryh^}7=`t2 zC|Vl7J(sz;B8xa|wtO%BxR63SAhj16_AJtsJOhXPg(qYcep@fS7E9b2EsEh8O_PR(UT=#Nlog8aJ5grw>F5yXCWLt5K{%&@Os(3mLc zVm`l2XpL>JwcI;oxpg3Jh;Ac0y}~PhpLP$+t|5Ds(56E9u{&+%j7%8FIt%Ug36?p1 zhG$eeFu2pjvEkvqC>JsWBrwbiE4a&;EiWf~#Q_ut&I}GpUZCTr^z*de&?l#Am$G!j zu6`?bNzB){CV>Q1YYVN(WxII?3 zcex*F>^BtNP4^ul4UYS(a*sX^g-b8*@%g7u2x2PCB*ZwQ3xOhmF(#a5=eWTZ`<#*Z z5JSr?k78%+X%R<*LzM+%6!rlgxQD9y)3PeVaR=f(?_rP?7dLwec7p6d&>xGWR%y_A z^d7Sa#|}%~tzHi^4_Cd-=wI{5q~F7?cqxwg%5H3Vk2}FkjF^3S$7Yjdu8BkZOkTs2eLIte%T(w~fSvMX3iu)Nez%?jlWPr8Hjf)p&gDY{hFxZotpP z=aJoE$E4@r=Q;}et}lo)?wuS$Y1}%AAUCc1Ld`$)Ek%IZb+CEyqP-$8vlxO(Nyy>T zo^JP?&ERHG3r0f2{BRBG^rw*`%sh|eO~tu$&-ex>;uiJsXROXkh5xc81~C%$2fa5Z zL!8UsMSB&;rVfb=`fv>_1Fs^7zyB73J%m&?Yc-XV+n$4E!`jZW_>Kge> z4>!0&66BiXI!nA+J7akxEC)U>U13a8^YY!ZjTh2{D$yA`^<)XS8x$vD%1TR2a~?($ zg>dXGH}k5qehN!pXy0CFPbJxD!uE3XM_D3e?o`E8%vC5ssTZu5si@xM)@-mwXk_*G{{G6n13N2j(HKJ(xnsj^Y)N4LD$+=hn2ma!S8-V4n=+vL%D6d%= zL-N~C(DUv#&8YBeOt9R#oEpb3Wa;&uis<}N(9uQuWB|CViujLEMPz_@22=w*CCF3A=d z)+aTqN*z#XFxq{~PWYxb_&4lRBPuB|FEqc!L*>gY{;^%5LJ%rt=y}L~V3h4qAAKf@ z+9f7jyFbfr=i}Yu>x;dgNyA!3J^A(O@X;dfB>~08XCYx@t1!#42)FbI8En&1u7)1O zx-{`Q9X39OZtE3JTPPJkK@bap4+{KcHIE>IJ+BSzSS&7^A1iU<;)%liT6NS|t1Gp7YXP(Da0T_|@H9QAom}Uu z4=A2ga>Z@eYhn8*PF7ib1(}!FBD$3jdZH!_s$}vh0f>MMAPF4XwC|n?N419I+ zaP!%SwekbS!OKI?>Cp0P#AB`?fpi&7gp(>S+Uv18;I&;XzaTgo;0f#Tro_om{zJ?*{$irZ4I`Y&o=dd=4P zkGXzD5NIZ_s%6XH3Y1zd;Syj<@d2v7MpKCs1# literal 0 HcmV?d00001 From 1d96685abd9f3f11e4516783f3534e8ae382c6d8 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 11:56:56 -0700 Subject: [PATCH 59/90] 114 Finishing up --- .../114-finalize_testing_reporting.ipynb | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 54c0ccd1c..22f9de0df 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1034,7 +1034,7 @@ "\n", "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", "\n", - "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**.\n", + "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**:\n", "\n", " \"Screenshot\n", "

\n", @@ -1057,9 +1057,9 @@ "\n", "7. Continue to work on your validation report by:\n", "\n", - " - **Generating drafts of risk assessment notes:**\n", - " - **Adding findings:**\n", - " - **Assessing compliance:**\n" + " - **Generating drafts of risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", + " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding***. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", + " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" ] }, { @@ -1102,7 +1102,7 @@ "\n", "### Work with your validation report\n", "\n", - "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to draft your validation report. Generate drafts of risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" + "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" ] }, { From 8515248382efdb4247176c05df66b4f12408267f Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 12:01:14 -0700 Subject: [PATCH 60/90] LLM custom context - session info --- notebooks/how_to/add_context_to_llm_descriptions.ipynb | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/notebooks/how_to/add_context_to_llm_descriptions.ipynb b/notebooks/how_to/add_context_to_llm_descriptions.ipynb index fc66635c2..45736d2c7 100644 --- a/notebooks/how_to/add_context_to_llm_descriptions.ipynb +++ b/notebooks/how_to/add_context_to_llm_descriptions.ipynb @@ -376,7 +376,7 @@ "\n", "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", "\n", - "This is a global setting that will affect all tests for your linked model:" + "This is a global setting that will affect all tests for your linked model for the duration of your ValidMind Library session:" ] }, { @@ -458,7 +458,7 @@ "\n", "To disable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `0`.\n", "\n", - "This is a global setting that will affect all tests for your linked model:" + "This is a global setting that will affect all tests for your linked model for the duration of your ValidMind Library session:" ] }, { From 2dd03e2dbb917d3bd83d69d586e483dc5b553beb Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 12:15:53 -0700 Subject: [PATCH 61/90] 114 quick edit to summary --- .../model_validation/114-finalize_testing_reporting.ipynb | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 22f9de0df..f02636df6 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1079,10 +1079,10 @@ "\n", "With our ValidMind for model validation series of notebooks, you learned how to validate a model end-to-end with the ValidMind Library by running through some common scenarios in a typical model validation setting:\n", "\n", - "- Running out-of-the-box tests and ensuring that validation is complete by running all tests provided by the model development team\n", - "- Leveraging ValidMind's flexible prompting to guide the text generation and insights from testing\n", - "- Extending the capabilities of the ValidMind Library by implementing custom tests\n", - "- Beginning the compliance assessment process by adding evidence to your validation report" + "- Verifying the data quality steps performed by the model development team\n", + "- Independently replicating the champion model's results and conducting additional tests to assess performance, stability, and robustness\n", + "- Setting up test inputs and challenger models for comparative analysis\n", + "- Running validation tests, analyzing results, and logging findings to ValidMind" ] }, { From d4b1f1b4a9796a71e6a94600d9a88ff1e7496d56 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 12:16:19 -0700 Subject: [PATCH 62/90] Removing the old notebook --- .../validation_of_application_scorecard.ipynb | 1448 ----------------- 1 file changed, 1448 deletions(-) delete mode 100644 notebooks/tutorials/model_validation/validation_of_application_scorecard.ipynb diff --git a/notebooks/tutorials/model_validation/validation_of_application_scorecard.ipynb b/notebooks/tutorials/model_validation/validation_of_application_scorecard.ipynb deleted file mode 100644 index 17a1007e7..000000000 --- a/notebooks/tutorials/model_validation/validation_of_application_scorecard.ipynb +++ /dev/null @@ -1,1448 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# Validation of an application scorecard model \n", - "\n", - "As a model validator, your task is to independently assess the application scorecard model developed using the ValidMind Library, which is based on Kaggle's [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data. Your role focuses on evaluating the model developer's work by conducting thorough testing and validation, potentially including the use of challenger models to benchmark performance.\n", - "\n", - "An application scorecard model is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", - " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", - " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", - "\n", - "This interactive notebook provides a step-by-step guide for:\n", - "\n", - " - Loading the developer model and verifying the data quality steps performed by the model developer.\n", - " - Independently replicating the model's results and conducting additional tests to assess performance, stability, and robustness.\n", - " - Setting up test inputs and challenger models for comparative analysis.\n", - " - Running validation tests, analyzing results, and logging findings to ValidMind." - ] - }, - { - "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 [Get started with the ValidMind Library](https://docs.validmind.ai/developer/get-started-validmind-library.html), we recommend you begin by exploring the available resources in this section. There, you can learn more about documenting models, find code samples, or read our developer reference.\n", - "\n", - "

For access to all features available in this notebook, create a free ValidMind account.\n", - "

\n", - "Signing up is FREE — Register with ValidMind
\n" - ] - }, - { - "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.\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", - "## 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": "code", - "execution_count": null, - "metadata": { - "jupyter": { - "source_hidden": true - } - }, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", - "\n", - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Initialize the ValidMind Library\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", - "\n", - "\n", - "### Get your code snippet\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 **Model Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Continue**. ([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:\n", - "\n", - " - Documentation template: `Credit Risk Scorecard`\n", - " - Use case: `Credit Risk - CECL`\n", - "\n", - " You can fill in other options according to your preference.\n", - "\n", - "4. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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", - "import validmind as vm\n", - "\n", - "vm.init(\n", - ",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Initialize the Python environment and import Model Developer Model\n", - "\n", - "Next, let's import the Model Developer model, used as the model developer as the champion model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Preview the Validation Report Template\n", - "\n", - "A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "You'll upload documentation and test results into this template later on. For now, take a look at the structure that the template provides with the `vm.preview_template()` function 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", - "## Load the sample dataset that Model Developer provided to the Validation Team, that was used to develop, train and test the model.\n", - "\n", - "The sample dataset used here is provided by the ValidMind library. To be able to use it, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.datasets.credit_risk import lending_club\n", - "\n", - "df = lending_club.load_data(source=\"offline\")\n", - "df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Obtain the Prepocessed dataset from Model Developer for data quality testing purposes" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "preprocess_df = lending_club.preprocess(df)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Obtain the final Feature engineered dataset that Model Developer uses to train and test the model\n", - "\n", - "In the feature engineering phase, we apply specific transformations to optimize the dataset for predictive modeling in our application scorecard. \n", - "\n", - "Using the `ending_club.feature_engineering()` function, the Model Developer conducted the following operations:\n", - "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", - "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "fe_df = lending_club.feature_engineering(preprocess_df)\n", - "fe_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## As a Model Validator we want to split the featured engineered dataset into train and test for Validation testing purposes. In addition, as a Validator we also want to include challenger/benchmark models.\n", - "\n", - "In this section, we will to a train and split randomly as the Validator want's to independently challenge the developer \n", - "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`). \n", - "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Split the data\n", - "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", - "\n", - "x_train = train_df.drop(lending_club.target_column, axis=1)\n", - "y_train = train_df[lending_club.target_column]\n", - "\n", - "x_test = test_df.drop(lending_club.target_column, axis=1)\n", - "y_test = test_df[lending_club.target_column]\n", - "\n", - "#now let's apply the order of features from the champion model construction\n", - "x_train = x_train[cols_when_model_builds]\n", - "x_test = x_test[cols_when_model_builds]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cols_use = ['annual_inc_woe',\n", - " 'verification_status_woe',\n", - " 'emp_length_woe',\n", - " 'installment_woe',\n", - " 'term_woe',\n", - " 'home_ownership_woe',\n", - " 'purpose_woe',\n", - " 'open_acc_woe',\n", - " 'total_acc_woe',\n", - " 'int_rate_woe',\n", - " 'sub_grade_woe',\n", - " 'grade_woe','loan_status']\n", - "\n", - "\n", - "train_df = train_df[cols_use]\n", - "test_df = test_df[cols_use]\n", - "test_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### As a Model Validator I also want to investigate potential challenger models - Let's train two challenger models as basis for the testing" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Define the Random Forest model\n", - "from sklearn.ensemble import RandomForestClassifier\n", - "rf_model = RandomForestClassifier(\n", - " n_estimators=50, \n", - " random_state=42,\n", - ")\n", - "\n", - "# Fit the model\n", - "rf_model.fit(x_train, y_train)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Second Challenger Model a Logistic Regression\n", - "from sklearn.linear_model import LogisticRegression\n", - "\n", - "# Logistic Regression grid params\n", - "log_reg_params = {\n", - " \"penalty\": [\"l1\", \"l2\"],\n", - " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", - " \"solver\": [\"liblinear\"],\n", - "}\n", - "\n", - "# Grid search for Logistic Regression\n", - "from sklearn.model_selection import GridSearchCV\n", - "\n", - "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", - "grid_log_reg.fit(x_train, y_train)\n", - "\n", - "# Logistic Regression best estimator\n", - "log_reg = grid_log_reg.best_estimator_\n", - "log_reg" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Compute probabilities as this is the raw probabilitistc output from the models of interest" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", - "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", - "\n", - "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", - "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", - "\n", - "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", - "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Compute binary predictions" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cut_off_threshold = 0.3 \n", - "\n", - "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", - "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", - "\n", - "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", - "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", - "\n", - "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", - "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## Document the model\n", - "\n", - "To document the model with the ValidMind Library, you'll need to:\n", - "1. Preprocess the raw dataset\n", - "2. Initialize some training and test datasets\n", - "3. Initialize a model object you can use for testing\n", - "4. Run the full suite of tests" - ] - }, - { - "cell_type": "markdown", - "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.\n", - "\n", - "This function takes a number of arguments:\n", - "\n", - "- `dataset`: The 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", - "With all datasets ready, you can now initialize the raw, processed, training and test datasets (`raw_df`, `preprocessed_df`, `fe_df`, `train_df` and `test_df`) created earlier into their own dataset objects using [`vm.init_dataset()`](https://docs.validmind.ai/validmind/validmind.html#init_dataset):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "vm_preprocess_dataset = vm.init_dataset(\n", - " dataset=preprocess_df,\n", - " input_id=\"preprocess_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "vm_fe_dataset = vm.init_dataset(\n", - " dataset=fe_df,\n", - " input_id=\"fe_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "vm_train_ds = vm.init_dataset(\n", - " dataset=train_df,\n", - " input_id=\"train_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "vm_test_ds = vm.init_dataset(\n", - " dataset=test_df,\n", - " input_id=\"test_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize a model object\n", - "\n", - "You will also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data. You simply intialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm_xgb_model = vm.init_model(\n", - " xgb_model,\n", - " input_id=\"xgb_model_developer_champion\",\n", - ")\n", - "\n", - "vm_rf_model = vm.init_model(\n", - " rf_model,\n", - " input_id=\"rf_model\",\n", - ")\n", - "\n", - "vm_log_model = vm.init_model(\n", - " log_reg,\n", - " input_id=\"log_model\",\n", - ")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Assign prediction values and probabilities to the datasets\n", - "\n", - "With our model now trained, we'll move on to assigning both the predictive probabilities coming directly from the model's predictions, and the binary prediction after applying the cutoff threshold described in the previous steps. \n", - "- These tasks are achieved through the use of the `assign_predictions()` method associated with the VM `dataset` object.\n", - "- This method links the model's class prediction values and probabilities to our VM train and test datasets." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# XGBoost Model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=train_xgb_binary_predictions,\n", - " prediction_probabilities=train_xgb_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=test_xgb_binary_predictions,\n", - " prediction_probabilities=test_xgb_prob,\n", - ")\n", - "\n", - "# Random Forest Model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=train_rf_binary_predictions,\n", - " prediction_probabilities=train_rf_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=test_rf_binary_predictions,\n", - " prediction_probabilities=test_rf_prob,\n", - ")\n", - "\n", - "\n", - "# Logistic Regression \n", - "vm_train_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=train_log_binary_predictions,\n", - " prediction_probabilities=train_log_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=test_log_binary_predictions,\n", - " prediction_probabilities=test_log_prob,\n", - ")\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Compute credit risk scores\n", - "\n", - "In this phase, we translate model predictions into actionable scores using probability estimates generated by our trained model." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", - "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", - "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", - "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", - "train_log_scores = lending_club.compute_scores(train_log_prob)\n", - "test_log_scores = lending_club.compute_scores(test_log_prob)\n", - "\n", - "# Assign scores to the datasets\n", - "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", - "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", - "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", - "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", - "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", - "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Model Validation Testing\n", - "\n", - "In the section below you (Model Validator) will select a series of tests from ValidMind in order to Independently challenge the Model Developer evidence and Assessment. In addition to this the Model Validator will also configure custom tets (tests not available out of the box). The focus will be on the following testing:\n", - "\n", - "- Ensuring Data used for training and testing the model is of appropriate data quality\n", - "- Ensuring that Raw Data has been pre-processed appropriately and the resulting feature engineered dataset reflects this\n", - "- Comprehensive testing around Model Performance of both the Developer Champion Model and challenger models developed by Validator\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "##first we start with Data Quality Testing, and we are going to leverage ValidMind tests for this purpose\n", - "#Explore tests\n", - "from validmind.tests import (\n", - " describe_test,\n", - " list_tests,\n", - " list_tasks,\n", - " list_tags,\n", - " list_tasks_and_tags,\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#let's find the data quality tests relevant for a classification use-case\n", - "list_tasks_and_tags()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#list the tests that we want to run as a validator for data_quality\n", - "list_tests(\n", - " tags=[\"data_quality\"], task=\"classification\"\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "dq = list_tests(tags=[\"data_quality\"], task=\"classification\",pretty=False)\n", - "dq " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#now let's run the list of tests, first focusing on dataquality of the datasets used for training, and then we will do a comparison test from raw to final train datasets, i.e. has the developer addressed potential concerns\n", - "for test in dq:\n", - " vm.tests.run_test(\n", - " test,\n", - " inputs={\n", - " \"dataset\": vm_preprocess_dataset\n", - " },\n", - " ).log() " - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Let's do a comparison of the raw dataset and the pre-processed dataset, i.e. has the Developer adjusted the processed the dataset from raw to pre-processed appropriately" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "######next let's do a comparison test between the pre-processed data and raw datasets\n", - "for test in dq:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Now let's do some independent testing with regards to performance of the champion model (xgboost). \n", - "#First we want to test independently on the champion model and then we will move forward to add challenger models that we have trained and defined before\n", - "list_tests(tags=[\"model_performance\"], task=\"classification\")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "mpt = ['validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion','validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion',\n", - " 'validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion', 'validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion','validmind.model_validation.sklearn.ROCCurve:xgboost_champion']" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in mpt:\n", - " vm.tests.run_test(\n", - " test,\n", - " inputs={\n", - " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#excellent- we have now conducted similar tests as developer in order to verify the results. Now let's provide some challenge by introducting challenger models\n", - "\n", - "#first let's provide some identifiers to the tests since they are now challenger model tests\n", - "mpt_chall = ['validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion_vs_challengers','validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers',\n", - " 'validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers', 'validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers','validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers']" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", - "\n", - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - "\n", - " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", - " - The metrics for the champion model compared agains the challenger models\n", - " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", - "\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#let's do same performance tests as above, but let's challenge the actual model itself and add two additional benchmark models\n", - "for test in mpt_chall:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model], \n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "### Now let's dig a little bit deeper into one of the tests that allows the Validator to custoimze parameters and thresholds for performance standards - and let's diregard RF model as we have\n", - "### learned that the RF model is not a viable candidate based on the performance metrics.\n", - "result = vm.tests.run_test(\n", - " 'validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold',\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], 'params':{'min_threshold': 0.35},\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "### Robustness and Stability Testing Comparison Between the Two Models\n", - "list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Let's see if models suffer from any overfit potentials and also where there are potential sub-segments of issues\n", - "overfit_testing = ['validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression','validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression'] \n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in overfit_testing:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "### Now finally let's conduct robustness and stability testing of the two models:\n", - "stab_robust = ['validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression'] # 'validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression'\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in stab_robust:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]], \"model\" : [vm_xgb_model,vm_log_model], \n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Let's verify the feature importance and inspect differences - different models might have more intuitive feature impacts that might lead to decisions in selection of a model" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "FI = list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", - "FI" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in FI:\n", - " vm.tests.run_test(\n", - " \"\".join((test,':Champion_vs_LogisticRegression')),\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model], \n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "### Let's finish off with a custom test example - scoring (customization of output to a FICO score type)\n", - "import numpy as np\n", - "import pandas as pd\n", - "import plotly.graph_objects as go\n", - "\n", - "\n", - "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", - "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", - " \"\"\"\n", - " Analyzes the relationship between score bands and odds (good:bad ratio).\n", - " Good odds = (1 - default_rate) / default_rate\n", - " \n", - " Higher scores should correspond to higher odds of being good.\n", - "\n", - " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", - "\n", - " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", - " \"\"\"\n", - " df = dataset.df\n", - " \n", - " # Create score bands\n", - " df['score_band'] = pd.cut(\n", - " df[score_column],\n", - " bins=[-np.inf] + score_bands + [np.inf],\n", - " labels=[f'<{score_bands[0]}'] + \n", - " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", - " [f'>{score_bands[-1]}']\n", - " )\n", - " \n", - " # Calculate metrics per band\n", - " results = df.groupby('score_band').agg({\n", - " dataset.target_column: ['mean', 'count']\n", - " })\n", - " \n", - " results.columns = ['Default Rate', 'Total']\n", - " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", - " results['Bad Count'] = results['Default Rate'] * results['Total']\n", - " results['Odds'] = results['Good Count'] / results['Bad Count']\n", - " \n", - " # Create visualization\n", - " fig = go.Figure()\n", - " \n", - " # Add odds bars\n", - " fig.add_trace(go.Bar(\n", - " name='Odds (Good:Bad)',\n", - " x=results.index,\n", - " y=results['Odds'],\n", - " marker_color='blue'\n", - " ))\n", - " \n", - " fig.update_layout(\n", - " title='Score-to-Odds Analysis',\n", - " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", - " showlegend=False\n", - " )\n", - " \n", - " return fig" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " param_grid={\n", - " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", - " \"score_bands\": [[500, 540, 570]],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Finally we got all of the tests from the Developer that was provided as evidence, now as a final task we will verify testing being appropriately recorded" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.utils import preview_test_config\n", - "\n", - "test_config = {'validmind.data_validation.DatasetDescription:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", - " 'validmind.data_validation.DescriptiveStatistics:raw_data': {'inputs': {'dataset': 'raw_dataset'}},\n", - " 'validmind.data_validation.MissingValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}},\n", - " 'validmind.data_validation.ClassImbalance:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 10}},\n", - " 'validmind.data_validation.Duplicates:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}},\n", - " 'validmind.data_validation.HighCardinality:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'num_threshold': 100,\n", - " 'percent_threshold': 0.1,\n", - " 'threshold_type': 'percent'}},\n", - " 'validmind.data_validation.Skewness:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_threshold': 1}},\n", - " 'validmind.data_validation.UniqueRows:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 1}},\n", - " 'validmind.data_validation.TooManyZeroValues:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_percent_threshold': 0.03}},\n", - " 'validmind.data_validation.IQROutliersTable:raw_data': {'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'threshold': 5}},\n", - " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.MissingValues:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'min_threshold': 1}},\n", - " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'}},\n", - " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'default_column': 'loan_status'}},\n", - " 'validmind.data_validation.DescriptiveStatistics:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.TabularDescriptionTables:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.ClassImbalance:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 10}},\n", - " 'validmind.data_validation.UniqueRows:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 1}},\n", - " 'validmind.data_validation.TabularNumericalHistograms:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.MutualInformation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'min_threshold': 0.01}},\n", - " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.data_validation.HighPearsonCorrelation:development_data': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}},\n", - " 'validmind.data_validation.WOEBinTable': {'input_grid': {'dataset': ['preprocess_dataset']},\n", - " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", - " 'int_rate': [10, 15, 20],\n", - " 'annual_inc': [50000, 100000, 150000]}}},\n", - " 'validmind.data_validation.WOEBinPlots': {'input_grid': {'dataset': ['preprocess_dataset']},\n", - " 'params': {'breaks_adj': {'loan_amnt': [5000, 10000, 15000, 20000, 25000],\n", - " 'int_rate': [10, 15, 20],\n", - " 'annual_inc': [50000, 100000, 150000]}}},\n", - " 'validmind.data_validation.DatasetSplit': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset']}},\n", - " 'validmind.model_validation.ModelMetadata': {'input_grid': {'model': ['xgb_model',\n", - " 'rf_model']}},\n", - " 'validmind.model_validation.sklearn.ModelParameters': {'input_grid': {'model': ['xgb_model',\n", - " 'rf_model']}},\n", - " 'validmind.model_validation.statsmodels.GINITable': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model', 'rf_model']}},\n", - " 'validmind.model_validation.sklearn.ClassifierPerformance': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model', 'rf_model']}},\n", - " 'validmind.model_validation.sklearn.TrainingTestDegradation:XGBoost': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'},\n", - " 'params': {'max_threshold': 0.1}},\n", - " 'validmind.model_validation.sklearn.TrainingTestDegradation:RandomForest': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'rf_model'},\n", - " 'params': {'max_threshold': 0.1}},\n", - " 'validmind.model_validation.sklearn.ROCCurve': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.5}},\n", - " 'validmind.model_validation.statsmodels.PredictionProbabilitiesHistogram': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.statsmodels.CumulativePredictionProbabilities': {'input_grid': {'model': ['xgb_model'],\n", - " 'dataset': ['train_dataset', 'test_dataset']}},\n", - " 'validmind.model_validation.sklearn.PopulationStabilityIndex': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'},\n", - " 'params': {'num_bins': 10, 'mode': 'fixed'}},\n", - " 'validmind.model_validation.sklearn.ConfusionMatrix': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.MinimumAccuracy': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.7}},\n", - " 'validmind.model_validation.sklearn.MinimumF1Score': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.5}},\n", - " 'validmind.model_validation.sklearn.PrecisionRecallCurve': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.CalibrationCurve': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.sklearn.ClassifierThresholdOptimization': {'inputs': {'dataset': 'train_dataset',\n", - " 'model': 'xgb_model'},\n", - " 'params': {'target_recall': 0.8}},\n", - " 'validmind.model_validation.statsmodels.ScorecardHistogram': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset']},\n", - " 'params': {'score_column': 'xgb_scores'}},\n", - " 'validmind.data_validation.ScoreBandDefaultRates': {'input_grid': {'dataset': ['train_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'score_column': 'xgb_scores', 'score_bands': [504, 537, 570]}},\n", - " 'validmind.model_validation.sklearn.ScoreProbabilityAlignment': {'input_grid': {'dataset': ['train_dataset'],\n", - " 'model': ['xgb_model']},\n", - " 'params': {'score_column': 'xgb_scores'}},\n", - " 'validmind.model_validation.sklearn.WeakspotsDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'}},\n", - " 'validmind.model_validation.sklearn.OverfitDiagnosis': {'inputs': {'model': 'xgb_model',\n", - " 'datasets': ['train_dataset', 'test_dataset']},\n", - " 'params': {'cut_off_threshold': 0.04}},\n", - " 'validmind.model_validation.sklearn.RobustnessDiagnosis': {'inputs': {'datasets': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': 'xgb_model'},\n", - " 'params': {'scaling_factor_std_dev_list': [0.1, 0.2, 0.3, 0.4, 0.5],\n", - " 'performance_decay_threshold': 0.05}},\n", - " 'validmind.model_validation.sklearn.PermutationFeatureImportance': {'input_grid': {'dataset': ['train_dataset',\n", - " 'test_dataset'],\n", - " 'model': ['xgb_model']}},\n", - " 'validmind.model_validation.FeaturesAUC': {'input_grid': {'model': ['xgb_model'],\n", - " 'dataset': ['train_dataset', 'test_dataset']}},\n", - " 'validmind.model_validation.sklearn.SHAPGlobalImportance': {'input_grid': {'model': ['xgb_model'],\n", - " 'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'kernel_explainer_samples': 10,\n", - " 'tree_or_linear_explainer_samples': 200}}}" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for t in test_config:\n", - " print(t)\n", - " try:\n", - " # Check if test has input_grid\n", - " if 'input_grid' in test_config[t]:\n", - " # For tests with input_grid, pass the input_grid configuration\n", - " if 'params' in test_config[t]:\n", - " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", - " else:\n", - " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", - " else:\n", - " # Original logic for regular inputs\n", - " if 'params' in test_config[t]:\n", - " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", - " else:\n", - " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", - " except Exception as e:\n", - " print(f\"Error running test {t}: {str(e)}\")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - } - ], - "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": "3.10.16" - } - }, - "nbformat": 4, - "nbformat_minor": 4 -} From 09a348bf243caceeb4c97a74656e19fc8b405319 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 14:28:38 -0700 Subject: [PATCH 63/90] Tweak --- .../model_validation/111-import_champion_model.ipynb | 7 +++++++ .../model_validation/114-finalize_testing_reporting.ipynb | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb index 5b31f7e9b..5976e6c08 100644 --- a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb +++ b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb @@ -249,6 +249,7 @@ }, { "cell_type": "markdown", + "id": "65dc004f", "metadata": {}, "source": [ "\n", @@ -300,6 +301,7 @@ }, { "cell_type": "markdown", + "id": "5e83409d", "metadata": {}, "source": [ "\n", @@ -347,6 +349,7 @@ }, { "cell_type": "markdown", + "id": "d04ae813", "metadata": {}, "source": [ "\n", @@ -384,6 +387,7 @@ }, { "cell_type": "markdown", + "id": "c5f2deb7", "metadata": {}, "source": [ "\n", @@ -410,6 +414,7 @@ }, { "cell_type": "markdown", + "id": "8461bb2e", "metadata": {}, "source": [ "\n", @@ -436,6 +441,7 @@ }, { "cell_type": "markdown", + "id": "e9e3df2c", "metadata": {}, "source": [ "\n", @@ -463,6 +469,7 @@ }, { "cell_type": "markdown", + "id": "eb8ef697", "metadata": {}, "source": [ "\n", diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index f02636df6..8c8f230ed 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1057,7 +1057,7 @@ "\n", "7. Continue to work on your validation report by:\n", "\n", - " - **Generating drafts of risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", + " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding***. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" ] From 0db6921db4929d0ffd620e573560e74fcaafac2f Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 15:01:17 -0700 Subject: [PATCH 64/90] Tweak2 --- .../112-develop_challenger_models.ipynb | 12 +++++++++--- .../114-finalize_testing_reporting.ipynb | 8 +++++++- 2 files changed, 16 insertions(+), 4 deletions(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 6e8fc3eb5..570db5fc9 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -355,9 +355,9 @@ "\n", "### Compute binary predictions\n", "\n", - "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of 0.3:\n", + "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", "\n", - "- If the probability is greater than 0.3, the prediction becomes `1` (positive).\n", + "- If the probability is greater than `0.3`, the prediction becomes `1` (positive).\n", "- Otherwise, it becomes `0` (negative)." ] }, @@ -792,8 +792,14 @@ } ], "metadata": { + "kernelspec": { + "display_name": "ValidMind Library", + "language": "python", + "name": "validmind" + }, "language_info": { - "name": "python" + "name": "python", + "version": "3.10.13" } }, "nbformat": 4, diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 8c8f230ed..3c6d92e93 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1146,8 +1146,14 @@ } ], "metadata": { + "kernelspec": { + "display_name": "ValidMind Library", + "language": "python", + "name": "validmind" + }, "language_info": { - "name": "python" + "name": "python", + "version": "3.10.13" } }, "nbformat": 4, From 584b1fcacba7e59b6a3aba702b53a38d2df38444 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 15:07:07 -0700 Subject: [PATCH 65/90] Typo --- .../model_validation/114-finalize_testing_reporting.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index 3c6d92e93..c362de8e0 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1058,7 +1058,7 @@ "7. Continue to work on your validation report by:\n", "\n", " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", - " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding***. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", + " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" ] }, From 6eb3eb10dc07d44c30ce265c0b1c3baa2f7af0a0 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 14 Mar 2025 15:13:35 -0700 Subject: [PATCH 66/90] Comment on .gitignore --- .gitignore | 1 + .../model_validation/111-import_champion_model.ipynb | 12 ++++-------- 2 files changed, 5 insertions(+), 8 deletions(-) diff --git a/.gitignore b/.gitignore index a3b44c3c4..e1f129048 100644 --- a/.gitignore +++ b/.gitignore @@ -191,6 +191,7 @@ lending_club_loan_data_*.csv # Pickle files *.pkl +# Sample application scorecard model for validation series — do not remove! !notebooks/tutorials/model_validation/xgb_model_champion.pkl notebooks/llm/datasets/*.jsonl diff --git a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb index 5976e6c08..099bd5131 100644 --- a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb +++ b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb @@ -485,11 +485,7 @@ "cell_type": "code", "execution_count": null, "id": "upgrade-show-752d8d8d-55f7-4465-aed0-34c5a3be8663", - "metadata": { - "vscode": { - "languageId": "plaintext" - } - }, + "metadata": {}, "outputs": [], "source": [ "%pip show validmind" @@ -550,13 +546,13 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "ValidMind Library", "language": "python", - "name": "python3" + "name": "validmind" }, "language_info": { "name": "python", - "version": "3.10" + "version": "3.10.13" } }, "nbformat": 4, From 6034dd3c1d54d30c67413874d6615b715c7d694f Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Tue, 25 Mar 2025 14:35:00 -0700 Subject: [PATCH 67/90] Updating the old intro notebook just in case someone finds it --- .../intro_for_model_developers.ipynb | 1785 +---------------- 1 file changed, 11 insertions(+), 1774 deletions(-) diff --git a/notebooks/tutorials/intro_for_model_developers.ipynb b/notebooks/tutorials/intro_for_model_developers.ipynb index a1412da7d..b3d62dae2 100644 --- a/notebooks/tutorials/intro_for_model_developers.ipynb +++ b/notebooks/tutorials/intro_for_model_developers.ipynb @@ -4,1789 +4,26 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "# ValidMind Introduction for Model Developers\n", + "# ValidMind introduction for model developers\n", "\n", - "As a model developer, learn how the end-to-end documentation process works based on common scenarios you encounter in model development settings.\n", + "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:\n", "\n", - "As a prerequisite, a model documentation template must be available on the ValidMind Platform. You can [view the available templates](https://docs.validmind.com/guide/model-documentation/swap-documentation-templates.html#view-current-templates) to see what templates been set up for your organization.\n", + "1. [101 Set up the ValidMind Library](/notebooks/tutorials/model_development/101-set_up_validmind.ipynb)\n", + "2. [102 Start the model development process](/notebooks/tutorials/model_development/102-start_development_process.ipynb)\n", + "3. [103 Integrate custom tests](/notebooks/tutorials/model_development/103-integrate_custom_tests.ipynb)\n", + "4. [104 Finalize testing and documentation](/notebooks/tutorials/model_development/104-finalize_testing_documentation.ipynb)\n", "\n", - "This notebook uses a binary classification model as an example, but the same principles shown here apply to other model types.\n", - "\n", - "\n", - "\n", - "## Overview of the notebook\n", - "\n", - "**1. Initializing the ValidMind Library**\n", - "\n", - "The ValidMind Library provides a rich collection of documentation tools and test suites, from documenting descriptions of datasets to validation and testing of models using a variety of open-source testing frameworks.\n", - "\n", - "**2. Start the model development process with raw data, run out-of-the box tests, and add evidence to model documentation**\n", - "\n", - "Learn how to access ValidMind's test repository of individual tests that you will use as building blocks to ensure a model is being built appropriately. The goal is to show how to run tests, investigate results, and add tests results or evidence to the documentation.\n", - "\n", - "For a full list of out-of-the-box tests, see [Test descriptions](https://docs.validmind.ai/developer/model-testing/test-descriptions.html) or try the interactive [Test sandbox](https://docs.validmind.ai/developer/model-testing/test-sandbox.html).\n", - "\n", - "**3. Implementing custom tests**\n", - "\n", - "Usually, model developers have their own custom tests and it is important to include this within the model documentation. We will show you how to include custom tests and then how they can be added to documentation as additional evidence.\n", - "\n", - "**4. Finalize testing and documentation**\n", - "\n", - "Learn how you can ensure that model documentation includes custom tests and how to make test configuration changes that apply to all tests in the model documentation template. At the end of this section you should have a fully documented model ready for review.\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Contents\n", - "\n", - "- [Overview of the notebook](#toc1_)\n", - "- [About ValidMind](#toc2_)\n", - " - [Before you begin](#toc2_1_)\n", - " - [New to ValidMind?](#toc2_2_)\n", - " - [Key concepts](#toc2_3_)\n", - "- [1. Initializing the ValidMind Library](#toc4_)\n", - " - [Install the ValidMind Library](#toc4_1_)\n", - " - [Initialize the ValidMind Library](#toc4_2_)\n", - " - [Get your code snippet](#toc4_2_1_)\n", - " - [Verify & preview the documentation template](#toc4_3_)\n", - "- [2. Start the model development process with raw data, run out-of-the box tests, and add evidence to model documentation](#toc5_)\n", - " - [Initialize the ValidMind datasets](#toc5_1_)\n", - " - [Run some tabular data tests](#toc5_2_)\n", - " - [Utilize test output](#toc5_3_)\n", - " - [Documenting the results based on two datasets](#toc5_4_)\n", - " - [Run `run_documentation_tests()` using `vm_raw_dataset_preprocessed` as input](#toc5_4_1_)\n", - " - [Log the individual result of the high correlation test that used `vm_balanced_raw_dataset` (that had a highly correlated `Age` column) as input](#toc5_4_2_)\n", - " - [Add individual test results to model documentation](#toc5_5_)\n", - " - [Model Testing](#toc5_6_)\n", - " - [Initialize model evaluation objects and assigning predictions](#toc5_7_)\n", - " - [Run the model evaluation tests](#toc5_8_)\n", - "- [3. Implementing custom tests](#toc6_)\n", - " - [Create a confusion matrix plot](#toc6_1_)\n", - " - [Add parameters to custom tests](#toc6_2_)\n", - " - [Pass parameters to custom tests](#toc6_3_)\n", - " - [Log the confusion matrix results](#toc6_4_)\n", - " - [Using external test providers](#toc6_5_)\n", - " - [Create a folder of custom tests from existing inline tests](#toc6_5_1_)\n", - " - [Save an inline test to a file](#toc6_5_2_)\n", - " - [Define and register a `LocalTestProvider` that points to that folder](#toc6_5_3_)\n", - " - [Initializing a local test provider](#toc6_6_)\n", - " - [Run test provider tests](#toc6_6_1_)\n", - " - [Add the test results to your documentation](#toc6_6_2_)\n", - "- [4. Finalize testing and documentation](#toc7_)\n", - " - [Use `run_documentation_tests()` to ensure custom test results are included in your documentation](#toc7_1_)\n", - " - [Viewing and updating the configuration for the entire model documentation template](#toc7_2_)\n", - " - [Update the config](#toc7_2_1_)\n", - "- [Where to go from here](#toc8_)\n", - " - [Use cases](#toc8_1_)\n", - " - [More how-to guides and code samples](#toc8_2_)\n", - " - [Discover more learning resources](#toc8_3_)\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. 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, create a free ValidMind account.\n", + "
Or, take our Developer Fundamentals course which walks you through the basics of ValidMind paired with this notebook series.\n", "

\n", - "Signing up is FREE — 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.\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## 1. Initializing the ValidMind Library\n" - ] - }, - { - "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 3.8 <= x <= 3.11\n", - "\n", - "To install the library:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind Library\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", - "\n", - "\n", - "### Get your code snippet\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 **Model Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Continue**. ([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:\n", - "\n", - " - Documentation template: `Binary classification`\n", - " - Use case: `Marketing/Sales - Attrition/Churn Management`\n", - "\n", - " You can fill in other options according to your preference.\n", - "\n", - "4. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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:\n" - ] - }, - { - "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": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Verify & preview the documentation template\n", - "\n", - "Let's verify that you have connected to ValidMind and that the appropriate template is selected. A template predefines sections for your model documentation and provides a general outline to follow, making the documentation process much easier.\n", - "\n", - "You will upload documentation and test results for this template later on. For now, take a look at the structure that the template provides with the `vm.preview_template()` function from the ValidMind library and note the empty sections:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Before learning how to run tests, let's explore the list of all available tests in the ValidMind Library. You can see that the documentation template for this model has references to some of the test IDs listed below.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm.tests.list_tests()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## 2. Start the model development process with raw data, run out-of-the box tests, and add evidence to model documentation\n", - "\n", - "In this section you learn how to explore the individual tests available in ValidMind and how to run them and change parameters as necessary. You will use a public dataset from Kaggle that models a bank customer churn prediction use case. The target column, `Exited` has a value of `1` when a customer has churned and `0` otherwise.\n", - "\n", - "You can find more information about this dataset [here](https://www.kaggle.com/datasets/shantanudhakadd/bank-customer-churn-prediction).\n", - "\n", - "The ValidMind Library provides a wrapper to automatically load the dataset as a Pandas DataFrame object.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "from validmind.datasets.classification import customer_churn as demo_dataset\n", - "\n", - "print(\n", - " f\"Loaded demo dataset with: \\n\\n\\t• Target column: '{demo_dataset.target_column}' \\n\\t• Class labels: {demo_dataset.class_labels}\"\n", - ")\n", - "\n", - "raw_df = demo_dataset.load_data()\n", - "raw_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's do some data quality assessments by running a few individual tests related to data assessment. You will use the `vm.tests.list_tests()` function introduced above in combination with `vm.tests.list_tags()` and `vm.tests.list_tasks()` to find which prebuilt tests are relevant for data quality assessment.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Get the list of available tags\n", - "sorted(vm.tests.list_tags())" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Get the list of available task types\n", - "sorted(vm.tests.list_tasks())" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can pass `tags` and `tasks` as parameters to the `vm.tests.list_tests()` function to filter the tests based on the tags and task types. For example, to find tests related to tabular data quality for classification models, you can call `list_tests()` like this:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm.tests.list_tests(task=\"classification\", tags=[\"tabular_data\", \"data_quality\"])" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize the ValidMind datasets\n", - "\n", - "Now, assume we have identified some tests we want to run with regards to the data we are intending to use. 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. You only need to do it one time per dataset.\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", - "This function takes a number of arguments:\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" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "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": [ - "\n", - "\n", - "### Run some tabular data tests\n", - "\n", - "Individual tests can be easily run by calling the `run_test` function provided by the `validmind.tests` module. The function takes the following arguments:\n", - "\n", - "- `test_id`: The ID of the test to run. To find a particular test and get its ID, refer to the [explore_tests](../how_to/explore_tests.ipynb) notebook. Look above for example after running 'vm.test_suites.describe_suite' as column 'Test ID' will contain the id.\n", - "- `params`: A dictionary of parameters for the test. These will override any `default_params` set in the test definition. Refer to the [explore_tests](../how_to/explore_tests.ipynb) notebook to find the default parameters for a test. See below for examples.\n", - "\n", - "The inputs expected by a test can also be found in the test definition. Let's take `validmind.data_validation.DescriptiveStatistics` as an example. Note that the output of the `describe_test()` function below shows that this test expects a `dataset` as input:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm.tests.describe_test(\"validmind.data_validation.DescriptiveStatistics\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Now, let's run a few tests to assess the quality of the dataset.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.DescriptiveStatistics\",\n", - " inputs={\"dataset\": vm_raw_dataset},\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "result2 = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.ClassImbalance\",\n", - " inputs={\"dataset\": vm_raw_dataset},\n", - " params={\"min_percent_threshold\": 30},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can see that the class imbalance test did not pass according to the value of `min_percent_threshold` we have set. Here is how you can re-run the test on some processed data to address this data quality issue. In this case we apply a very simple rebalancing technique to the dataset.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "import pandas as pd\n", - "\n", - "raw_copy_df = raw_df.sample(frac=1) # Create a copy of the raw dataset\n", - "\n", - "# Create a balanced dataset with the same number of exited and not exited customers\n", - "exited_df = raw_copy_df.loc[raw_copy_df[\"Exited\"] == 1]\n", - "not_exited_df = raw_copy_df.loc[raw_copy_df[\"Exited\"] == 0].sample(n=exited_df.shape[0])\n", - "\n", - "balanced_raw_df = pd.concat([exited_df, not_exited_df])\n", - "balanced_raw_df = balanced_raw_df.sample(frac=1, random_state=42)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "With this new raw dataset, you can re-run the individual test to see if it passes the class imbalance test requirement. Remember to register new VM Dataset object since that is the type of input required by `run_test()`:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Register new data and now 'balanced_raw_dataset' is the new dataset object of interest\n", - "vm_balanced_raw_dataset = vm.init_dataset(\n", - " dataset=balanced_raw_df,\n", - " input_id=\"balanced_raw_dataset\",\n", - " target_column=\"Exited\",\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.ClassImbalance\",\n", - " inputs={\"dataset\": vm_balanced_raw_dataset},\n", - " params={\"min_percent_threshold\": 30},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Utilize test output\n", - "\n", - "Here is an example for how you can utilize the output from a ValidMind test for futher use, for example, if you want to remove highly correlated features. The example below shows how you can get the list of features with the highest correlation coefficients and use them to reduce the final list of features for modeling.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "corr_result = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.HighPearsonCorrelation\",\n", - " params={\"max_threshold\": 0.3},\n", - " inputs={\"dataset\": vm_balanced_raw_dataset},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's assume we want to remove highly correlated features from the dataset. `corr_result` is an object of type `TestResult`. We can inspect the result object to see what the test has produced." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "print(type(corr_result))\n", - "print(\"Result ID: \", corr_result.result_id)\n", - "print(\"Params: \", corr_result.params)\n", - "print(\"Passed: \", corr_result.passed)\n", - "print(\"Tables: \", corr_result.tables)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's check out the table in the result and extract a list of features that failed the test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "features_df = corr_result.tables[0].data\n", - "features_df" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Remove the highly correlated features and create a new VM dataset object. Note the use of different `input_id`s. This allows tracking the inputs used when running each individual test.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "high_correlation_features = features_df[features_df[\"Pass/Fail\"] == \"Fail\"][\"Columns\"].tolist()\n", - "high_correlation_features" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Extract the feature names from the list of strings (e.g. '(Age, Exited)' -> 'Age')" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "high_correlation_features = [feature.split(\",\")[0].strip(\"()\") for feature in high_correlation_features]\n", - "high_correlation_features" - ] - }, - { - "cell_type": "code", - "execution_count": 42, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Remove the highly correlated features from the dataset\n", - "balanced_raw_no_age_df = balanced_raw_df.drop(columns=high_correlation_features)\n", - "\n", - "# Re-initialize the dataset object\n", - "vm_raw_dataset_preprocessed = vm.init_dataset(\n", - " dataset=balanced_raw_no_age_df,\n", - " input_id=\"raw_dataset_preprocessed\",\n", - " target_column=\"Exited\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Re-running the test with the reduced feature set should pass the test. You can also plot the correlation matrix to visualize the new correlation between features:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "corr_result = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.HighPearsonCorrelation\",\n", - " params={\"max_threshold\": 0.3},\n", - " inputs={\"dataset\": vm_raw_dataset_preprocessed},\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "corr_result = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.PearsonCorrelationMatrix\",\n", - " inputs={\"dataset\": vm_raw_dataset_preprocessed},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Documenting the results based on two datasets\n", - "\n", - "We have now done some analysis on two different datasets and we should able to document why certain things were done to the raw data with testing to support it. Every test result returned by the `run_test()` function has a `.log()` method that can be used to log the test results to ValidMind. When logging individual results to ValidMind you need to manually add those results in a specific section of the model documentation.\n", - "\n", - "When using `run_documentation_tests()`, it's possible to automatically populate a section with the results of all tests that were registered in the documentation template.\n", - "\n", - "To show how to add individual results to any documentation section, we're going to populate the entire `data_preparation` section of the documentation using the clean `vm_raw_dataset_preprocessed` dataset as input, and then we're going to document an additional result for the highly correlated dataset `vm_balanced_raw_dataset`. The following two steps will accomplish this:\n", - "\n", - "1. Run `run_documentation_tests()` using `vm_raw_dataset_preprocessed` as input. This populates the entire data preparation section for every test that is already part of the documentation template.\n", - "2. Log the individual result of the high correlation test that used `vm_balanced_raw_dataset` (that had a highly correlated `Age` column) as input\n", - "\n", - "After adding the result of step #2 to the documentation you will be able to explain the changes made to the raw data by editing the default description of the test result within the ValidMind Platform.\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Run `run_documentation_tests()` using `vm_raw_dataset_preprocessed` as input\n", - "\n", - "`run_documentation_tests()` allows you to run multiple tests at once and log the results to the documentation. The function takes the following arguments:\n", - "\n", - "- `inputs`: any inputs to be passed to the tests\n", - "- `config`: a dictionary `:` that allows configuring each test individually. Each test config has the following form:\n", - " - `params`: individual test parameters\n", - " - `inputs`: individual test inputs. When passed, this overrides any inputs passed from the `run_documentation_tests()` function\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "test_config = {\n", - " \"validmind.data_validation.ClassImbalance\": {\n", - " \"params\": {\"min_percent_threshold\": 30},\n", - " },\n", - " \"validmind.data_validation.HighPearsonCorrelation\": {\n", - " \"params\": {\"max_threshold\": 0.3},\n", - " },\n", - "}\n", - "\n", - "tests_suite = vm.run_documentation_tests(\n", - " inputs={\n", - " \"dataset\": vm_raw_dataset_preprocessed,\n", - " },\n", - " config=test_config,\n", - " section=[\"data_preparation\"],\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Log the individual result of the high correlation test that used `vm_balanced_raw_dataset` (that had a highly correlated `Age` column) as input\n", - "\n", - "Here you can use a custom `result_id` to tag the individual result with a unique identifier. This `result_id` can be appended to `test_id` with a `:` separator. The `balanced_raw_dataset` result identifier will correspond to the `balanced_raw_dataset` input, the dataset that still has the `Age` column.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " test_id=\"validmind.data_validation.HighPearsonCorrelation:balanced_raw_dataset\",\n", - " params={\"max_threshold\": 0.3},\n", - " inputs={\"dataset\": vm_balanced_raw_dataset},\n", - ")\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Add individual test results to model documentation\n", - "\n", - "You can now visit the documentation page for the model you connected to at the beginning of this notebook and add a new content block in the relevant section.\n", - "\n", - "To do this, go to the documentation page of your model and navigate to the `Data Preparation` -> `Correlations and Interactions` section. Then hover after the \"Pearson Correlation Matrix\" 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-correlations.png)\n", - "\n", - "Click on the `+` button and select `Test-Driven Block`. This will open a dialog where you can select `Threshold Test` as the type of the test-driven content block, and then select `High Pearson Correlation Vm Raw Dataset Test`. This will show a preview of the result and it should match the results shown above.\n", - "\n", - "![screenshot showing the selected test result in the dialog](../images/selecting-high-pearson-correlation-test.png)\n", - "\n", - "Finally, click on the `Insert block` button to add the test result to the documentation. You'll now see two individual results for the high correlation test in the `Correlations and Interactions` section of the documentation. To finalize the documentation, you can edit the test result's description block to explain the changes made to the raw data and the reasons behind them as we can see in the screenshot below.\n", - "\n", - "![screenshot showing the high pearson correlation block](../images/high-pearson-correlation-block.png)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Model Testing\n", - "\n", - "We have focused so far on the data assessment and pre-processing that usually occurs prior to any models being built. Now we are going to assume we have built a model and we want to incorporate some model results in our documentation.\n", - "\n", - "Let's train a simple logistic regression model on the dataset and evaluate its performance. You will use the `LogisticRegression` class from the `sklearn.linear_model` and use ValidMind tests to evaluate the model's performance.\n", - "\n", - "Before training the model, we need to encode the categorical features in the dataset. You will use the `OneHotEncoder` class from the `sklearn.preprocessing` module to encode the categorical features. The categorical features in the dataset are `Geography` and `Gender`.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "balanced_raw_no_age_df.head()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "balanced_raw_no_age_df = pd.get_dummies(\n", - " balanced_raw_no_age_df, columns=[\"Geography\", \"Gender\"], drop_first=True\n", - ")\n", - "balanced_raw_no_age_df.head()" - ] - }, - { - "cell_type": "code", - "execution_count": 49, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "from sklearn.linear_model import LogisticRegression\n", - "from sklearn.model_selection import train_test_split\n", - "\n", - "# Split the input and target variables\n", - "X = balanced_raw_no_age_df.drop(\"Exited\", axis=1)\n", - "y = balanced_raw_no_age_df[\"Exited\"]\n", - "X_train, X_test, y_train, y_test = train_test_split(\n", - " X,\n", - " y,\n", - " test_size=0.2,\n", - " random_state=42,\n", - ")\n", - "\n", - "# Logistic Regression grid params\n", - "log_reg_params = {\n", - " \"penalty\": [\"l1\", \"l2\"],\n", - " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", - " \"solver\": [\"liblinear\"],\n", - "}\n", - "\n", - "# Grid search for Logistic Regression\n", - "from sklearn.model_selection import GridSearchCV\n", - "\n", - "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", - "grid_log_reg.fit(X_train, y_train)\n", - "\n", - "# Logistic Regression best estimator\n", - "log_reg = grid_log_reg.best_estimator_" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Initialize model evaluation objects and assigning predictions\n", - "\n", - "The last step for evaluating the model's performance is to initialize the ValidMind `Dataset` and `Model` objects and assign model predictions to each dataset. You will use the `init_dataset`, `init_model` and `assign_predictions` functions to initialize these objects.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 50, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "train_df = X_train\n", - "train_df[\"Exited\"] = y_train\n", - "test_df = X_test\n", - "test_df[\"Exited\"] = y_test\n", - "\n", - "vm_train_ds = vm.init_dataset(\n", - " input_id=\"train_dataset_final\",\n", - " dataset=train_df,\n", - " target_column=\"Exited\",\n", - ")\n", - "\n", - "vm_test_ds = vm.init_dataset(\n", - " input_id=\"test_dataset_final\",\n", - " dataset=test_df,\n", - " target_column=\"Exited\",\n", - ")\n", - "\n", - "# Register the model\n", - "vm_model = vm.init_model(log_reg, input_id=\"log_reg_model_v1\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Once the model has been registered you can assign model predictions to the training and test datasets. The `assign_predictions()` method from the `Dataset` object can link existing predictions to any number of models. If no prediction values are passed, the method will compute predictions automatically:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm_train_ds.assign_predictions(model=vm_model)\n", - "vm_test_ds.assign_predictions(model=vm_model)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Run the model evaluation tests\n", - "\n", - "In this part, we focus on running the tests within the model development section of the model documentation. Only tests associated with this section will be executed, and the corresponding results will be updated in the model documentation. In the example below, you will focus on only running tests for the `model development` section of the document.\n", - "\n", - "Note the additional config that is passed to `run_documentation_tests()`. This allows you to override inputs or params in certain tests. In our case, we want to explicitly use the `vm_train_ds` for the `validmind.model_validation.sklearn.ClassifierPerformance:in_sample` test, since it's supposed to run on the training dataset and not the test dataset.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "test_config = {\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:in_sample\": {\n", - " \"inputs\": {\n", - " \"dataset\": vm_train_ds,\n", - " \"model\": vm_model,\n", - " },\n", - " }\n", - "}\n", - "results = vm.run_documentation_tests(\n", - " section=[\"model_development\"],\n", - " inputs={\n", - " \"dataset\": vm_test_ds, # Any test that requires a single dataset will use vm_test_ds\n", - " \"model\": vm_model,\n", - " \"datasets\": (\n", - " vm_train_ds,\n", - " vm_test_ds,\n", - " ), # Any test that requires multiple datasets will use vm_train_ds and vm_test_ds\n", - " },\n", - " config=test_config,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## 3. Implementing custom tests\n", - "\n", - "This section assumes that model developers already have a repository of custom made tests that they consider critical to include in the documentation. Here we provide details on how to easily integrate custom tests with ValidMind.\n", - "\n", - "For a more in-depth introduction to custom tests, refer to this [notebook](../code_samples/custom_tests/implement_custom_tests.ipynb).\n", - "\n", - "A custom test is any function that takes a set of inputs and parameters as arguments and returns one or more outputs. 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. 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", - "In the following example, you will learn how to implement a custom `inline` test that calculates the confusion matrix for a binary classification model. You will see that the custom test function is just a regular Python function that can include and require any Python library as you see fit.\n", - "\n", - "**NOTE**: in the context of Jupyter notebooks, we will use the word `inline` to refer to functions (or code) defined in the same notebook where they are used (this one) and not in a separate file, as we will see later with test providers.\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Create a confusion matrix plot\n", - "\n", - "To understand how to create a custom test from anything, let's first create a confusion matrix plot using the `confusion_matrix` function from the `sklearn.metrics` module.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "import matplotlib.pyplot as plt\n", - "from sklearn import metrics\n", - "\n", - "# Get the predicted classes\n", - "y_pred = log_reg.predict(vm_test_ds.x)\n", - "\n", - "confusion_matrix = metrics.confusion_matrix(y_test, y_pred)\n", - "\n", - "cm_display = metrics.ConfusionMatrixDisplay(\n", - " confusion_matrix=confusion_matrix, display_labels=[False, True]\n", - ")\n", - "cm_display.plot()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "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.metrics.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" - ] - }, - { - "cell_type": "code", - "execution_count": 54, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "@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=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": [ - "You can now run the newly created custom test on both the training and test datasets using the `run_test()` function:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Training dataset\n", - "result = vm.tests.run_test(\n", - " \"my_custom_tests.ConfusionMatrix:training_dataset\",\n", - " inputs={\"model\": vm_model, \"dataset\": vm_train_ds},\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Test dataset\n", - "result = vm.tests.run_test(\n", - " \"my_custom_tests.ConfusionMatrix:test_dataset\",\n", - " inputs={\"model\": vm_model, \"dataset\": vm_test_ds},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Add parameters to custom tests\n", - "\n", - "Custom tests can take parameters just like any other function. Let's modify the `confusion_matrix` function to take an additional parameter `normalize` that will allow you to normalize the confusion matrix.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 57, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "@vm.test(\"my_custom_tests.ConfusionMatrix\")\n", - "def confusion_matrix(dataset, model, normalize=False):\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=model)\n", - "\n", - " if normalize:\n", - " confusion_matrix = metrics.confusion_matrix(y_true, y_pred, normalize=\"all\")\n", - " else:\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": [ - "\n", - "\n", - "### Pass parameters to custom tests\n", - "\n", - "You can pass parameters to custom tests 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 `dataset` and `model` are still passed as `inputs`. Since these are `VMDataset` or `VMModel` inputs, they have a special meaning. When declaring a `dataset`, `model`, `datasets` or `models` argument in a custom test function, the ValidMind Library will expect these get passed as `inputs` to `run_test()` (or `run_documentation_tests()` instead).\n", - "\n", - "Re-running the confusion matrix with `normalize=True` looks like this:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "# Test dataset with normalize=True\n", - "result = vm.tests.run_test(\n", - " \"my_custom_tests.ConfusionMatrix:test_dataset_normalized\",\n", - " inputs={\"model\": vm_model, \"dataset\": vm_test_ds},\n", - " params={\"normalize\": True},\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Log the confusion matrix results\n", - "\n", - "As you saw in the pearson correlation example, you can log any 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.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 59, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Using external test providers\n", - "\n", - "Creating inline custom tests with a function is a great way to customize your model documentation. However, sometimes you may want to reuse the same set of tests across multiple models and share them with developers in your organization. In this case, you can create a custom test provider that will allow you to load custom tests from a local folder or a git repository.\n", - "\n", - "In this section you will learn how to declare a local filesystem test provider that allows loading tests from a local folder following these high level steps:\n", - "\n", - "1. Create a folder of custom tests from existing, inline tests (tests that exists in your active Jupyter notebook)\n", - "2. Save an inline test to a file\n", - "3. Define and register a `LocalTestProvider` that points to that folder\n", - "4. Run test provider tests\n", - "5. Add the test results to your documentation\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Create a folder of custom tests from existing inline tests\n", - "\n", - "Here you will create a new folder that will contain reusable, custom tests. The following code snippet will create a new `my_tests` directory in the current working directory if it doesn't exist.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 60, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "tests_folder = \"my_tests\"\n", - "\n", - "import os\n", - "\n", - "# create tests folder\n", - "os.makedirs(tests_folder, exist_ok=True)\n", - "\n", - "# remove existing tests\n", - "for f in os.listdir(tests_folder):\n", - " # remove files and pycache\n", - " if f.endswith(\".py\") or f == \"__pycache__\":\n", - " os.system(f\"rm -rf {tests_folder}/{f}\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "After running the command above, you should see a new directory next to this notebook file:\n", - "\n", - "![screenshot showing my_tests directory](../images/my_tests_directory.png)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Save an inline test to a file\n", - "\n", - "The `@vm.test` decorator that was used above to register these as one-off custom tests also adds a convenience method to the function object that allows you to simply call `.save()` to save it to a file. This will save the function to a Python file to a path you specify. In this case, you can pass the variable `tests_folder` to save it to the custom tests folder we created.\n", - "\n", - "Normally, this will get you started by creating the file and saving the function code with the correct name. But it won't automatically add any import or other functions/variables outside of the function that are needed for the test to run. The `save()` method allows you to pass an optional `imports` argument that will ensure the necessary imports are added to the file.\n", - "\n", - "For the `confusion_matrix` test, note the imports that are required for the function to run properly:\n", - "\n", - "```python\n", - "import matplotlib.pyplot as plt\n", - "from sklearn import metrics\n", - "```\n", - "\n", - "You can pass these imports to the `save()` method to ensure they are included in the file with the following command:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "confusion_matrix.save(\n", - " tests_folder,\n", - " imports=[\"import matplotlib.pyplot as plt\", \"from sklearn import metrics\"],\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### What happened?\n", - "\n", - "The `save()` method saved the `confusion_matrix` function to a file named `ConfusionMatrix.py` in the `my_tests` folder. Note that the new file provides some context on the origin of the test, which is useful for traceability.\n", - "\n", - "```\n", - "# Saved from __main__.confusion_matrix\n", - "# Original Test ID: my_custom_tests.ConfusionMatrix\n", - "# New Test ID: .ConfusionMatrix\n", - "```\n", - "\n", - "Additionally, the new test function has been stripped off its decorator, as it now resides in a file that will be loaded by the test provider:\n", - "\n", - "```python\n", - "def ConfusionMatrix(dataset, model, normalize=False):\n", - "```\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Define and register a `LocalTestProvider` that points to that folder\n", - "\n", - "With the `my_tests` folder now having a sample custom test, you can now initialize a test provider that will tell the ValidMind Library where to find these tests. ValidMind offers out-of-the-box test providers for local tests (i.e. tests in a folder) or a Github provider for tests in a Github repository. You can also create your own test provider by creating a class that has a `load_test` method that takes a test ID and returns the test function matching that ID.\n", - "\n", - "The most important attribute for a test provider is its `namespace`. This is a string that will be used to prefix test IDs in model documentation. This allows you to have multiple test providers with tests that can even share the same ID, but are distinguished by their namespace.\n", - "\n", - "An extended introduction to test providers can be found in [this](../code_samples/custom_tests/integrate_external_test_providers.ipynb) notebook.\n", - "\n", - "\n", - "\n", - "### Initializing a local test provider\n", - "\n", - "For most use-cases, the local test provider should be sufficient. This test provider allows you load custom tests from a designated directory. Let's go ahead and see how we can do this with our custom tests.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 62, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "from validmind.tests import LocalTestProvider\n", - "\n", - "# initialize the test provider with the tests folder we created earlier\n", - "my_test_provider = LocalTestProvider(tests_folder)\n", - "\n", - "vm.tests.register_test_provider(\n", - " namespace=\"my_test_provider\",\n", - " test_provider=my_test_provider,\n", - ")\n", - "# `my_test_provider.load_test()` will be called for any test ID that starts with `my_test_provider`\n", - "# e.g. `my_test_provider.ConfusionMatrix` will look for a function named `ConfusionMatrix` in `my_tests/ConfusionMatrix.py` file" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Run test provider tests\n", - "\n", - "Now that you have set up the test provider, you can run any test that's located in the tests folder by using the `run_test()` method as with any other test. For tests that reside in a test provider directory, the test ID will be the `namespace` specified when registering the provider, followed by the path to the test file relative to the tests folder. For example, the Confusion Matrix test we created earlier will have the test ID `my_test_provider.ConfusionMatrix`. You could organize the tests in subfolders, say `classification` and `regression`, and the test ID for the Confusion Matrix test would then be `my_test_provider.classification.ConfusionMatrix`.\n", - "\n", - "Let's go ahead and re-run the confusion matrix test by using the test ID `my_test_provider.ConfusionMatrix`. This should load the test from the test provider and run it as before.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " \"my_test_provider.ConfusionMatrix\",\n", - " inputs={\"model\": vm_model, \"dataset\": vm_test_ds},\n", - " params={\"normalize\": True},\n", - ")\n", - "\n", - "result.log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Add the test results to your documentation\n", - "\n", - "You have already seen how to add individual results to the model documentation using the ValidMind Platform. Let's repeat the process and add the confusion matrix to the `Model Development` -> `Model Evaluation` section of the documentation. The \"add test driven block\" dialog should now show the new test result coming from the test provider:\n", - "\n", - "![screenshot showing confusion matrix result](../images/insert-test-driven-block-custom-confusion-matrix.png)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "## 4. Finalize testing and documentation\n", - "\n", - "In this section we cover how to finalize the testing and documentation of your model by focusing on:\n", - "\n", - "1. Using `run_documentation_tests()` to ensure custom test results are included in your documentation\n", - "2. Viewing and updating the configuration for the entire model documentation template\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Use `run_documentation_tests()` to ensure custom test results are included in your documentation\n", - "\n", - "After adding test driven blocks to your model documentation, changes should persist and become available every time you call `vm.preview_template()`. However, you need to reload the connection to the ValidMind Platform if you have added test driven blocks when the connection was already established.\n" - ] - }, - { - "cell_type": "code", - "execution_count": 64, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm.reload()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Now, run `preview_template()` and verify that the new confusion matrix test you added is included in the proper section.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Since the test ID is now registered in the document you can now run tests for an entire section and all additional custom tests should be loaded without issues. Let's run all tests in the `model_evaluation` section of the documentation. Note that we have been running the sample custom confusion matrix with `normalize=True` to demonstrate the ability to provide custom parameters.\n", - "\n", - "In the `Run the model evaluation tests` section above you learned how to assign inputs to individual tests with `run_documentation_tests()`. Assigning parametesr is similar, you only need to provide assign a `params` dictionary to a given test ID, `my_test_provider.ConfusionMatrix` in this case.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "test_config = {\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:in_sample\": {\n", - " \"inputs\": {\n", - " \"dataset\": vm_train_ds,\n", - " \"model\": vm_model,\n", - " },\n", - " },\n", - " \"my_test_provider.ConfusionMatrix\": {\n", - " \"params\": {\"normalize\": True},\n", - " },\n", - "}\n", - "results = vm.run_documentation_tests(\n", - " section=[\"model_evaluation\"],\n", - " inputs={\n", - " \"dataset\": vm_test_ds, # Any test that requires a single dataset will use vm_test_ds\n", - " \"model\": vm_model,\n", - " \"datasets\": (\n", - " vm_train_ds,\n", - " vm_test_ds,\n", - " ), # Any test that requires multiple datasets will use vm_train_ds and vm_test_ds\n", - " },\n", - " config=test_config,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "### Viewing and updating the configuration for the entire model documentation template\n", - "\n", - "The ValidMind Library provides a utility function called `vm.get_test_suite().get_default_config()` that allows you to render the default configuration for the entire documentation template. This configuration will contain all the test IDs and their default parameters. You can then modify this configuration as needed and pass it to `run_documentation_tests()` to run all tests in the documentation template if needed. You also have the option to continue running tests for one section at a time, `get_default_config()` still provides a useful reference for providing default parametes to every test.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "import json\n", - "\n", - "model_test_suite = vm.get_test_suite()\n", - "config = model_test_suite.get_default_config()\n", - "print(\"Suite Config: \\n\", json.dumps(config, indent=2))" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Update the config\n", - "\n", - "Note that the default config does not assign any inputs to a test, this is expected. You can assign inputs to individual tests as needed, depending on the datasets and models you want to pass to individual tests. The `config` dictionary, as a mapping of test IDs to test configurations, allows you to do this.\n", - "\n", - "For this particular documentation template (binary classification), the ValidMind Library provides a sample configuration that can be used to populate the entire model documentation using the following inputs as placeholders:\n", - "\n", - "- A `raw_dataset` raw dataset\n", - "- A `train_dataset` training dataset\n", - "- A `test_dataset` test dataset\n", - "- A trained `model` instance\n", - "\n", - "As part of updating the `config` you will need to ensure the correct `input_id`s are used in the final config passed to `run_documentation_tests()`.\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "from validmind.datasets.classification import customer_churn\n", - "from validmind.utils import preview_test_config\n", - "\n", - "test_config = customer_churn.get_demo_test_config()\n", - "preview_test_config(test_config)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Using this sample configuration, let's finish populating model documentation by running all tests for the `model_development` section of the documentation. Recall that the training and test datasets in our exercise have the following `input_id` values:\n", - "\n", - "- `train_dataset_final` for the training dataset\n", - "- `test_dataset_final` for the test dataset\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "metadata": {} - }, - "outputs": [], - "source": [ - "config = {\n", - " \"validmind.model_validation.ModelMetadata\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\"},\n", - " },\n", - " \"validmind.data_validation.DatasetSplit\": {\n", - " \"inputs\": {\"datasets\": [\"train_dataset_final\", \"test_dataset_final\"]},\n", - " },\n", - " \"validmind.model_validation.sklearn.PopulationStabilityIndex\": {\n", - " \"inputs\": {\n", - " \"model\": \"log_reg_model_v1\",\n", - " \"datasets\": [\"train_dataset_final\", \"test_dataset_final\"],\n", - " },\n", - " \"params\": {\"num_bins\": 10, \"mode\": \"fixed\"},\n", - " },\n", - " \"validmind.model_validation.sklearn.ConfusionMatrix\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " },\n", - " \"my_test_provider.ConfusionMatrix\": {\n", - " \"inputs\": {\"dataset\": \"test_dataset_final\", \"model\": \"log_reg_model_v1\"},\n", - " },\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:in_sample\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"train_dataset_final\"}\n", - " },\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:out_of_sample\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"}\n", - " },\n", - " \"validmind.model_validation.sklearn.PrecisionRecallCurve\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " },\n", - " \"validmind.model_validation.sklearn.ROCCurve\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " },\n", - " \"validmind.model_validation.sklearn.TrainingTestDegradation\": {\n", - " \"inputs\": {\n", - " \"model\": \"log_reg_model_v1\",\n", - " \"datasets\": [\"train_dataset_final\", \"test_dataset_final\"],\n", - " },\n", - " \"params\": {\n", - " \"metrics\": [\"accuracy\", \"precision\", \"recall\", \"f1\"],\n", - " \"max_threshold\": 0.1,\n", - " },\n", - " },\n", - " \"validmind.model_validation.sklearn.MinimumAccuracy\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " \"params\": {\"min_threshold\": 0.7},\n", - " },\n", - " \"validmind.model_validation.sklearn.MinimumF1Score\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " \"params\": {\"min_threshold\": 0.5},\n", - " },\n", - " \"validmind.model_validation.sklearn.MinimumROCAUCScore\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " \"params\": {\"min_threshold\": 0.5},\n", - " },\n", - " \"validmind.model_validation.sklearn.PermutationFeatureImportance\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " },\n", - " \"validmind.model_validation.sklearn.SHAPGlobalImportance\": {\n", - " \"inputs\": {\"model\": \"log_reg_model_v1\", \"dataset\": \"test_dataset_final\"},\n", - " \"params\": {\"kernel_explainer_samples\": 10},\n", - " },\n", - " \"validmind.model_validation.sklearn.WeakspotsDiagnosis\": {\n", - " \"inputs\": {\n", - " \"model\": \"log_reg_model_v1\",\n", - " \"datasets\": [\"train_dataset_final\", \"test_dataset_final\"],\n", - " },\n", - " \"params\": {\n", - " \"thresholds\": {\"accuracy\": 0.75, \"precision\": 0.5, \"recall\": 0.5, \"f1\": 0.7}\n", - " },\n", - " },\n", - " \"validmind.model_validation.sklearn.OverfitDiagnosis\": {\n", - " \"inputs\": {\n", - " \"model\": \"log_reg_model_v1\",\n", - " \"datasets\": [\"train_dataset_final\", \"test_dataset_final\"],\n", - " },\n", - " \"params\": {\"cut_off_percentage\": 4},\n", - " },\n", - " \"validmind.model_validation.sklearn.RobustnessDiagnosis\": {\n", - " \"inputs\": {\n", - " \"model\": \"log_reg_model_v1\",\n", - " \"datasets\": [\"train_dataset_final\", \"test_dataset_final\"],\n", - " },\n", - " \"params\": {\n", - " \"scaling_factor_std_dev_list\": [0.0, 0.1, 0.2, 0.3, 0.4, 0.5],\n", - " \"accuracy_decay_threshold\": 4,\n", - " },\n", - " },\n", - "}\n", - "\n", - "\n", - "full_suite = vm.run_documentation_tests(\n", - " section=\"model_development\",\n", - " config=config,\n", - ")" - ] - }, - { - "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", - "- [Application scorecard demo](../code_samples/credit_risk/application_scorecard_demo.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 metrics](../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)\n" - ] - }, - { - "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." + "Training is FREE — Register now!
" ] } ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "ValidMind Library", "language": "python", - "name": "python3" + "name": "validmind" }, "language_info": { "codemirror_mode": { @@ -1798,7 +35,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.5" + "version": "3.10.13" } }, "nbformat": 4, From 5dcfa661ab2fdb1cfb5e3c03c16b6098bdcce459 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 26 Mar 2025 14:17:55 -0700 Subject: [PATCH 68/90] Edits to validation key concepts + dataset transformations --- .../model_validation/111-import_champion_model.ipynb | 11 +++++++---- .../112-develop_challenger_models.ipynb | 2 +- .../113-perform_validation_tests.ipynb | 10 ++++++++-- .../114-finalize_testing_reporting.ipynb | 2 +- 4 files changed, 17 insertions(+), 8 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb index 099bd5131..0095588cc 100644 --- a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb +++ b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb @@ -123,11 +123,11 @@ "\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", + "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\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", + "**Validation report template**: Serves as a standardized framework for conducting and documenting model validation activities. It outlines the required sections, recommended analyses, and expected validation tests, ensuring consistency and completeness across validation reports. The template helps guide validators through a systematic review process while promoting comparability and traceability of validation outcomes.\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", + "**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.\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", @@ -394,7 +394,10 @@ "\n", "### Load the sample dataset\n", "\n", - "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle that was used to develop our dummy champion model, and use it to replicate some steps that would have been conducted during the initial development and documentation of the model.\n", + "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", + "\n", + "- We'll use this dataset to review steps that should have been conducted during the initial development and documentation of the model to ensure that the model was built correctly.\n", + "- By independently performing steps such as preprocessing and feature engineering, we can confirm whether the model was built using appropriate and properly processed data.\n", "\n", "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" ] diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 570db5fc9..011bd531f 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -131,7 +131,7 @@ "\n", "### Import the champion model\n", "\n", - "Next, we'll import the champion model submitted by the model development team as we used in the last notebook (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" + "Next, we'll import the champion model submitted by the model development team as we used in the last notebook (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" ] }, { diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index b183c94c1..d28ef8ec2 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -149,7 +149,7 @@ "\n", "### Import the champion model\n", "\n", - "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" + "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" ] }, { @@ -1003,8 +1003,14 @@ } ], "metadata": { + "kernelspec": { + "display_name": "ValidMind Library", + "language": "python", + "name": "validmind" + }, "language_info": { - "name": "python" + "name": "python", + "version": "3.10.13" } }, "nbformat": 4, diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index c362de8e0..eedfb1138 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -146,7 +146,7 @@ "\n", "### Import the champion model\n", "\n", - "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the same sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset:" + "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" ] }, { From a424832a5de15979a6af0e770298e92556b6e320 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 26 Mar 2025 14:20:27 -0700 Subject: [PATCH 69/90] Removing test suites as well --- .../model_validation/111-import_champion_model.ipynb | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb index 0095588cc..facd731d5 100644 --- a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb +++ b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb @@ -142,11 +142,7 @@ "\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." + "**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." ] }, { From 2ed90a349a99306d4c5ded18806378ceaff878b4 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Wed, 26 Mar 2025 14:23:30 -0700 Subject: [PATCH 70/90] Tweak --- .../model_validation/112-develop_challenger_models.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index 011bd531f..ca118d3f9 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -179,7 +179,7 @@ "\n", "## Split the feature engineered dataset\n", "\n", - "With our dummy model imported and our preprocessed dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", + "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", "\n", "Splitting our dataset into training and testing is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", "\n", From 92749dfa4017cdb82edd57644e14e580d37587d5 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 11:52:10 -0700 Subject: [PATCH 71/90] Setup for advanced notebook --- .../111-import_champion_model.ipynb | 135 +++--------------- .../112-develop_challenger_models.ipynb | 81 ----------- .../113-perform_validation_tests.ipynb | 76 ---------- .../114-finalize_testing_reporting.ipynb | 82 ----------- .../validate_application_scorecard.ipynb | 32 +++++ 5 files changed, 54 insertions(+), 352 deletions(-) create mode 100644 notebooks/tutorials/validate_application_scorecard.ipynb diff --git a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb index facd731d5..730a25129 100644 --- a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb +++ b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "markdown", - "id": "7ecc4b20", + "id": "248883b1", "metadata": {}, "source": [ "# ValidMind for model validation — 111 Import the champion model\n", @@ -14,52 +14,9 @@ }, { "cell_type": "markdown", - "id": "42e1f410", - "metadata": {}, - "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [Introduction](#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", - " - [Register a sample model](#toc3_1_) \n", - " - [Assign validator credentials](#toc3_1_1_) \n", - " - [Install the ValidMind Library](#toc3_2_) \n", - " - [Initialize the ValidMind Library](#toc3_3_) \n", - " - [Get your code snippet](#toc3_3_1_) \n", - "- [Getting to know ValidMind](#toc4_) \n", - " - [Preview the validation report template](#toc4_1_) \n", - " - [View validation report in the ValidMind Platform](#toc4_2_) \n", - "- [Importing the champion model](#toc5_) \n", - " - [Load the sample dataset](#toc5_1_) \n", - " - [Preprocess the dataset](#toc5_1_1_) \n", - " - [Apply feature engineering to the dataset](#toc5_2_) \n", - "- [Upgrade ValidMind](#toc6_) \n", - "- [In summary](#toc7_) \n", - "- [Next steps](#toc8_) \n", - " - [Develop potential challenger models](#toc8_1_) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "id": "9e4a8f6f", + "id": "e1657d54", "metadata": {}, "source": [ - "\n", - "\n", "## Introduction\n", "\n", "Model validation aims to independently assess the compliance of *champion models* created by model developers with regulatory guidance by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance. Assessments, presented in the form of a validation report, typically include *model findings* and recommendations to address those issues.\n", @@ -72,11 +29,9 @@ }, { "cell_type": "markdown", - "id": "69c913b4", + "id": "434bea70", "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", @@ -86,11 +41,9 @@ }, { "cell_type": "markdown", - "id": "d302994e", + "id": "662d6027", "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", @@ -100,11 +53,9 @@ }, { "cell_type": "markdown", - "id": "9e7f591b", + "id": "5a1a2fa8", "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", @@ -116,11 +67,9 @@ }, { "cell_type": "markdown", - "id": "2d5a9c74", + "id": "aa9b1d8e", "metadata": {}, "source": [ - "\n", - "\n", "### Key concepts\n", "\n", "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", @@ -147,21 +96,17 @@ }, { "cell_type": "markdown", - "id": "29616ee3", + "id": "057c1417", "metadata": {}, "source": [ - "\n", - "\n", "## Setting up" ] }, { "cell_type": "markdown", - "id": "15a805a5", + "id": "ef1f5ce4", "metadata": {}, "source": [ - "\n", - "\n", "### Register a sample model\n", "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", @@ -184,11 +129,9 @@ }, { "cell_type": "markdown", - "id": "45c6d89b", + "id": "9743927c", "metadata": {}, "source": [ - "\n", - "\n", "#### Assign validator credentials\n", "\n", "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", @@ -208,10 +151,9 @@ }, { "cell_type": "markdown", + "id": "37a94e48", "metadata": {}, "source": [ - "\n", - "\n", "### Install the ValidMind Library\n", "\n", "
Recommended Python versions\n", @@ -224,7 +166,7 @@ { "cell_type": "code", "execution_count": null, - "id": "775b0796", + "id": "cef99db1", "metadata": {}, "outputs": [], "source": [ @@ -233,11 +175,8 @@ }, { "cell_type": "markdown", - "id": "2cecdcb7", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind Library\n", "\n", "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." @@ -245,11 +184,9 @@ }, { "cell_type": "markdown", - "id": "65dc004f", + "id": "30803184", "metadata": {}, "source": [ - "\n", - "\n", "#### Get your code snippet\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", @@ -264,7 +201,7 @@ { "cell_type": "code", "execution_count": null, - "id": "822f649c", + "id": "d5de450a", "metadata": {}, "outputs": [], "source": [ @@ -287,21 +224,16 @@ }, { "cell_type": "markdown", - "id": "3045cd0e", "metadata": {}, "source": [ - "\n", - "\n", "## Getting to know ValidMind" ] }, { "cell_type": "markdown", - "id": "5e83409d", + "id": "d22ef356", "metadata": {}, "source": [ - "\n", - "\n", "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", @@ -312,7 +244,7 @@ { "cell_type": "code", "execution_count": null, - "id": "b5a1f62c", + "id": "38592086", "metadata": {}, "outputs": [], "source": [ @@ -321,11 +253,8 @@ }, { "cell_type": "markdown", - "id": "7b996415", "metadata": {}, "source": [ - "\n", - "\n", "### View validation report in the ValidMind Platform\n", "\n", "Next, let's head to the ValidMind Platform to see the template in action:\n", @@ -345,11 +274,9 @@ }, { "cell_type": "markdown", - "id": "d04ae813", + "id": "943a45d6", "metadata": {}, "source": [ - "\n", - "\n", "## Importing the champion model\n", "\n", "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" @@ -358,7 +285,7 @@ { "cell_type": "code", "execution_count": null, - "id": "5b9e47f2", + "id": "3f9e197e", "metadata": {}, "outputs": [], "source": [ @@ -383,11 +310,8 @@ }, { "cell_type": "markdown", - "id": "c5f2deb7", "metadata": {}, "source": [ - "\n", - "\n", "### Load the sample dataset\n", "\n", "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", @@ -401,7 +325,7 @@ { "cell_type": "code", "execution_count": null, - "id": "332843f1", + "id": "40e73b7c", "metadata": {}, "outputs": [], "source": [ @@ -413,11 +337,8 @@ }, { "cell_type": "markdown", - "id": "8461bb2e", "metadata": {}, "source": [ - "\n", - "\n", "#### Preprocess the dataset\n", "\n", "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", @@ -431,7 +352,7 @@ { "cell_type": "code", "execution_count": null, - "id": "805cf3de", + "id": "e436e862", "metadata": {}, "outputs": [], "source": [ @@ -440,11 +361,8 @@ }, { "cell_type": "markdown", - "id": "e9e3df2c", "metadata": {}, "source": [ - "\n", - "\n", "### Apply feature engineering to the dataset\n", "\n", "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", @@ -458,7 +376,7 @@ { "cell_type": "code", "execution_count": null, - "id": "eea535a8", + "id": "9c6b641c", "metadata": {}, "outputs": [], "source": [ @@ -468,11 +386,8 @@ }, { "cell_type": "markdown", - "id": "eb8ef697", "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", @@ -483,7 +398,7 @@ { "cell_type": "code", "execution_count": null, - "id": "upgrade-show-752d8d8d-55f7-4465-aed0-34c5a3be8663", + "id": "10272aa9", "metadata": {}, "outputs": [], "source": [ @@ -512,11 +427,8 @@ }, { "cell_type": "markdown", - "id": "4849b3cc", "metadata": {}, "source": [ - "\n", - "\n", "## In summary\n", "\n", "In this first notebook, you learned how to:\n", @@ -529,14 +441,11 @@ }, { "cell_type": "markdown", + "id": "78f8161b", "metadata": {}, "source": [ - "\n", - "\n", "## Next steps\n", "\n", - "\n", - "\n", "### Develop potential challenger models\n", "\n", "Now that the ValidMind Library is connected to your model in the ValidMind Library with the correct template applied, we can go ahead and start the model validation process: **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**" diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb index ca118d3f9..68ac41403 100644 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb @@ -15,47 +15,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [Prerequisites](#toc1_) \n", - "- [Setting up](#toc2_) \n", - " - [Initialize the ValidMind Library](#toc2_1_) \n", - " - [Import the champion model](#toc2_2_) \n", - "- [Split the feature engineered dataset](#toc3_) \n", - "- [Investigate potential challenger models](#toc4_) \n", - " - [Random forest classification model](#toc4_1_) \n", - " - [Logistic regression model](#toc4_2_) \n", - "- [Extract predicted probabilities](#toc5_) \n", - " - [Compute binary predictions](#toc5_1_) \n", - "- [Initializing the ValidMind objects](#toc6_) \n", - " - [Initialize the ValidMind datasets](#toc6_1_) \n", - " - [Initialize the model objects](#toc6_2_) \n", - "- [Assign predictions](#toc7_) \n", - " - [Compute credit risk scores](#toc7_1_) \n", - "- [Enabling custom context for test descriptions](#toc8_) \n", - " - [Review default LLM-generated descriptions](#toc8_1_1_) \n", - " - [Enable use case context](#toc8_1_2_) \n", - "- [In summary](#toc9_) \n", - "- [Next steps](#toc10_) \n", - " - [Perform model validation tests](#toc10_1_) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", "## Prerequisites\n", "\n", "In order to continue on the model validation journey with notebook, you'll need to first have:\n", @@ -72,8 +31,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Setting up" ] }, @@ -81,8 +38,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind Library\n", "\n", "First, let's connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\n", @@ -127,8 +82,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Import the champion model\n", "\n", "Next, we'll import the champion model submitted by the model development team as we used in the last notebook (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" @@ -175,8 +128,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Split the feature engineered dataset\n", "\n", "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", @@ -236,8 +187,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Investigate potential challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", @@ -249,8 +198,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", @@ -281,8 +228,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Logistic regression model\n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", @@ -321,8 +266,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models:" @@ -351,8 +294,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Compute binary predictions\n", "\n", "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", @@ -386,8 +327,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Initializing the ValidMind objects" ] }, @@ -395,8 +334,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind datasets\n", "\n", "Before you can run tests, you'll need 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", @@ -461,8 +398,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the model objects\n", "\n", "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", @@ -499,8 +434,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", @@ -560,8 +493,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Compute credit risk scores\n", "\n", "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" @@ -594,8 +525,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Enabling custom context for test descriptions" ] }, @@ -617,8 +546,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Review default LLM-generated descriptions\n", "\n", "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", @@ -644,8 +571,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Enable use case context\n", "\n", "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", @@ -763,8 +688,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## In summary\n", "\n", "In this second notebook, you learned how to:\n", @@ -779,12 +702,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Next steps\n", "\n", - "\n", - "\n", "### Perform model validation tests\n", "\n", "Now that you're familiar with the basics of using the ValidMind Library, let's learn how to identify and run validation tests: **[113 Perform model validation tests](113-perform_validation_tests.ipynb)**" diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb index d28ef8ec2..eb96a5f74 100644 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb @@ -23,45 +23,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [Prerequisites](#toc1_) \n", - "- [Setting up](#toc2_) \n", - " - [Initialize the ValidMind Library](#toc2_1_) \n", - " - [Import the champion model](#toc2_2_) \n", - " - [Train challenger models](#toc2_3_) \n", - " - [Extract predicted probabilities](#toc2_4_) \n", - " - [Initialize the ValidMind objects](#toc2_5_) \n", - " - [Assign predictions](#toc2_6_) \n", - " - [Enable use case context](#toc2_7_) \n", - "- [Run data quality tests](#toc3_) \n", - " - [Identify qualitative tests](#toc3_1_) \n", - " - [Log data quality tests](#toc3_2_) \n", - " - [Run comparison tests](#toc3_3_) \n", - "- [Run performance tests](#toc4_) \n", - " - [Evaluate performance of the champion model](#toc4_1_) \n", - " - [Evaluate performance of challenger models](#toc4_2_) \n", - "- [In summary](#toc5_) \n", - "- [Next steps](#toc6_) \n", - " - [Finalize testing and reporting](#toc6_1_) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", "## Prerequisites\n", "\n", "In order to log tests results or evidence to your validation report with this notebook, you'll need to first have:\n", @@ -71,7 +32,6 @@ "- [ ] Learned how to import and initialize datasets for use with ValidMind\n", "- [ ] Understood the basics of how to enable custom context for test descriptions generated by ValidMind\n", "\n", - "\n", "
Need help with the above steps?\n", "

\n", "Refer to the first two notebooks in this series:\n", @@ -88,8 +48,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Setting up\n", "\n", "This section should be quite familiar to you — as we performed the same actions in the previous notebook, **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**." @@ -99,8 +57,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind Library\n", "\n", "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\n", @@ -145,8 +101,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Import the champion model\n", "\n", "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" @@ -245,8 +199,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Train challenger models\n", "\n", "Just like in the last notebook, we'll train two challenger models as the basis for our testing:" @@ -302,8 +254,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models and convert the probability predictions into a binary:" @@ -354,8 +304,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind objects\n", "\n", "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" @@ -432,8 +380,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" @@ -519,8 +465,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Enable use case context\n", "\n", "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned in preparation for running our tests:" @@ -598,8 +542,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run data quality tests\n", "\n", "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." @@ -609,8 +551,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Identify qualitative tests\n", "\n", "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." @@ -676,8 +616,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Log data quality tests\n", "\n", "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" @@ -732,8 +670,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Run comparison tests\n", "\n", "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" @@ -758,8 +694,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run performance tests\n", "\n", "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", @@ -811,8 +745,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Evaluate performance of the champion model\n", "\n", "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" @@ -837,8 +769,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Evaluate performance of challenger models\n", "\n", "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", @@ -975,8 +905,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## In summary\n", "\n", "In this third notebook, you learned how to:\n", @@ -990,12 +918,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Next steps\n", "\n", - "\n", - "\n", "### Finalize testing and reporting\n", "\n", "Now that you're familiar with the basics of using the ValidMind Library to run and log validation tests, let's learn how to implement some custom tests and wrap up our validation: **[104 Finalize testing and reporting](114-finalize_testing_reporting.ipynb)**" diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index eedfb1138..dc1bfeed2 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -15,47 +15,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "::: {.content-hidden when-format=\"html\"}\n", - "## Contents \n", - "- [Prerequisites](#toc1_) \n", - "- [Setting up](#toc2_) \n", - " - [Initialize the ValidMind Library](#toc2_1_) \n", - " - [Import the champion model](#toc2_2_) \n", - " - [Train the challenger model](#toc2_3_) \n", - " - [Extract predicted probabilities](#toc2_4_) \n", - " - [Initialize the ValidMind objects](#toc2_5_) \n", - " - [Assign predictions](#toc2_6_) \n", - " - [Enable use case context](#toc2_7_) \n", - "- [Adjust a ValidMind test](#toc3_) \n", - "- [Run diagnostic tests](#toc4_) \n", - "- [Run feature importance tests](#toc5_) \n", - "- [Implement a custom test](#toc6_) \n", - "- [Verify test runs](#toc7_) \n", - "- [Add test results to reporting](#toc8_) \n", - "- [In summary](#toc9_) \n", - "- [Next steps](#toc10_) \n", - " - [Work with your validation report](#toc10_1_) \n", - " - [Learn more](#toc10_2_) \n", - " - [More how-to guides and code samples](#toc10_2_1_) \n", - " - [Discover more learning resources](#toc10_2_2_) \n", - "\n", - ":::\n", - "\n", - "" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", "## Prerequisites\n", "\n", "In order to finalize the validation testing and reporting for your sample model, you'll need to first have:\n", @@ -67,7 +26,6 @@ "- [ ] Understood the basics of how to identify and run validation tests\n", "- [ ] Run data quality and model performance tests for your champion and challenger models, and logged the results of those tests to the ValidMind Platform\n", "\n", - "\n", "
Need help with the above steps?\n", "

\n", "Refer to the first three notebooks in this series:\n", @@ -85,8 +43,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Setting up\n", "\n", "This section should be very familiar to you now — as we performed the same actions in the previous two notebooks in this series." @@ -96,8 +52,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind Library\n", "\n", "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\n", @@ -142,8 +96,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Import the champion model\n", "\n", "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" @@ -235,8 +187,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Train the challenger model\n", "\n", "As we eliminated the random forest classification model as a challenger, we'll only train our logistic regression model here:" @@ -273,8 +223,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Extract predicted probabilities\n", "\n", "With our challenger model trained, let's extract the predicted probabilities from our two models and convert the probability predictions into a binary:" @@ -317,8 +265,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind objects\n", "\n", "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" @@ -389,8 +335,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" @@ -452,8 +396,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Enable use case context\n", "\n", "We'll also adjust the use case context to focus on comparison between our models for tests going forward:" @@ -535,8 +477,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Adjust a ValidMind test\n", "\n", "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", @@ -595,8 +535,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run diagnostic tests\n", "\n", "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", @@ -710,8 +648,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run feature importance tests\n", "\n", "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", @@ -750,8 +686,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Implement a custom test\n", "\n", "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", @@ -860,8 +794,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Verify test runs\n", "\n", "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", @@ -1022,8 +954,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Add test results to reporting\n", "\n", "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", @@ -1066,8 +996,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## In summary\n", "\n", "In this final notebook, you learned how to:\n", @@ -1089,8 +1017,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Next steps" ] }, @@ -1098,8 +1024,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Work with your validation report\n", "\n", "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" @@ -1109,8 +1033,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Learn more\n", "\n", "Now that you're familiar with the basics, you can explore the following notebooks to get a deeper understanding on how the ValidMind Library assists you in streamlining model validation:" @@ -1120,8 +1042,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### More how-to guides and code samples\n", "\n", "- [Explore available tests in detail](../../how_to/explore_tests.ipynb)\n", @@ -1134,8 +1054,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\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", diff --git a/notebooks/tutorials/validate_application_scorecard.ipynb b/notebooks/tutorials/validate_application_scorecard.ipynb new file mode 100644 index 000000000..e7d84efe9 --- /dev/null +++ b/notebooks/tutorials/validate_application_scorecard.ipynb @@ -0,0 +1,32 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Validate an application scorecard model\n", + "\n", + "Learn how to independently assess an application scorecard model developed using the ValidMind Library as a validator. You'll evaluate the development of the model by conducting thorough testing and analysis, including the use of challenger models to benchmark performance.\n", + "\n", + "An *application scorecard model* is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", + "\n", + " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", + " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", + "\n", + " This interactive notebook provides a step-by-step guide for:\n", + "\n", + "- Verifying the data quality steps performed by the model development team\n", + "- Independently replicating the champion model's results and conducting additional tests to assess performance, stability, and robustness\n", + "- Setting up test inputs and challenger models for comparative analysis\n", + "- Running validation tests, analyzing results, and logging findings to ValidMind" + ] + } + ], + "metadata": { + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} From 037faf4316e322d5b550060a7d9344ae158ea25c Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:18:26 -0700 Subject: [PATCH 72/90] Moving the application scorecard advanced notebook --- .../model_validation}/validate_application_scorecard.ipynb | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename notebooks/{tutorials => code_samples/model_validation}/validate_application_scorecard.ipynb (100%) diff --git a/notebooks/tutorials/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb similarity index 100% rename from notebooks/tutorials/validate_application_scorecard.ipynb rename to notebooks/code_samples/model_validation/validate_application_scorecard.ipynb From e41066c42335a2c345be88a933974bed4b2f1935 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:19:39 -0700 Subject: [PATCH 73/90] Moving 111 to new notebook --- .../validate_application_scorecard.ipynb | 359 +++++++++++++++++- 1 file changed, 358 insertions(+), 1 deletion(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index e7d84efe9..70d32c909 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -20,7 +20,364 @@ "- Setting up test inputs and challenger models for comparative analysis\n", "- Running validation tests, analyzing results, and logging findings to ValidMind" ] - } + }, + { + "cell_type": "markdown", + "id": "434bea70", + "metadata": {}, + "source": [ + "## 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 comparison and other validation tests, and then use the ValidMind Platform to submit compliance assessments of champion models via comprehensive validation reports. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model developers." + ] + }, + { + "cell_type": "markdown", + "id": "662d6027", + "metadata": {}, + "source": [ + "### 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": "5a1a2fa8", + "metadata": {}, + "source": [ + "### 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, create a free ValidMind account.\n", + "

\n", + "Signing up is FREE — Register with ValidMind
" + ] + }, + { + "cell_type": "markdown", + "id": "aa9b1d8e", + "metadata": {}, + "source": [ + "### Key concepts\n", + "\n", + "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", + "\n", + "**Validation report template**: Serves as a standardized framework for conducting and documenting model validation activities. It outlines the required sections, recommended analyses, and expected validation tests, ensuring consistency and completeness across validation reports. The template helps guide validators through a systematic review process while promoting comparability and traceability of validation outcomes.\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.\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/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." + ] + }, + { + "cell_type": "markdown", + "id": "057c1417", + "metadata": {}, + "source": [ + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "id": "ef1f5ce4", + "metadata": {}, + "source": [ + "### Register a sample model\n", + "\n", + "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", + "\n", + "For this series of notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\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 **Continue**. ([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:\n", + "\n", + " - Documentation template: `Credit Risk Scorecard`\n", + " - Use case: `Credit Risk — CECL`\n", + "\n", + " You can fill in other options according to your preference." + ] + }, + { + "cell_type": "markdown", + "id": "9743927c", + "metadata": {}, + "source": [ + "#### Assign validator credentials\n", + "\n", + "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", + "\n", + "1. Remove yourself as a developer: \n", + "\n", + " - Click on the **DEVELOPERS** tile.\n", + " - Click the **x** next to your name to remove yourself from that model's role.\n", + " - Click **Save** to apply your changes to that role.\n", + "\n", + "2. Add yourself as a validator: \n", + "\n", + " - Click on the **VALIDATORS** tile.\n", + " - Select your name from the drop-down menu.\n", + " - Click **Save** to apply your changes to that role." + ] + }, + { + "cell_type": "markdown", + "id": "37a94e48", + "metadata": {}, + "source": [ + "### 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, + "id": "cef99db1", + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the ValidMind Library\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." + ] + }, + { + "cell_type": "markdown", + "id": "30803184", + "metadata": {}, + "source": [ + "#### Get your code snippet\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "\n", + "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "\n", + "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": "d5de450a", + "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": [ + "## Getting to know ValidMind" + ] + }, + { + "cell_type": "markdown", + "id": "d22ef356", + "metadata": {}, + "source": [ + "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", + "\n", + "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "38592086", + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### View validation report in the ValidMind Platform\n", + "\n", + "Next, let's head to the ValidMind Platform to see the template in action:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "\n", + "3. Click on the **Validation Report** for your model and note:\n", + "\n", + " - [ ] The risk assessment compliance summary at the top of the report (screenshot below)\n", + " - [ ] How the structure of the validation report reflects the previewed template\n", + "\n", + " \"Screenshot\n", + "

" + ] + }, + { + "cell_type": "markdown", + "id": "943a45d6", + "metadata": {}, + "source": [ + "## Importing the champion model\n", + "\n", + "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3f9e197e", + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b6887d38", + "metadata": {}, + "outputs": [], + "source": [ + "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Load the sample dataset\n", + "\n", + "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", + "\n", + "- We'll use this dataset to review steps that should have been conducted during the initial development and documentation of the model to ensure that the model was built correctly.\n", + "- By independently performing steps such as preprocessing and feature engineering, we can confirm whether the model was built using appropriate and properly processed data.\n", + "\n", + "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "40e73b7c", + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.datasets.credit_risk import lending_club\n", + "\n", + "df = lending_club.load_data(source=\"offline\")\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Preprocess the dataset\n", + "\n", + "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", + "\n", + "- Filters the dataset to include only loans for debt consolidation or credit card purposes\n", + "- Removes loans classified under the riskier grades \"F\" and \"G\"\n", + "- Excludes uncommon home ownership types and standardizes employment length and loan terms into numerical formats\n", + "- Discards unnecessary fields and any entries with missing information to maintain a clean and robust dataset for modeling" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e436e862", + "metadata": {}, + "outputs": [], + "source": [ + "preprocess_df = lending_club.preprocess(df)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Apply feature engineering to the dataset\n", + "\n", + "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", + "\n", + "We'll apply the following transformations using the `ending_club.feature_engineering()` function to optimize the dataset for predictive modeling in our application scorecard:\n", + "\n", + "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", + "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9c6b641c", + "metadata": {}, + "outputs": [], + "source": [ + "fe_df = lending_club.feature_engineering(preprocess_df)\n", + "fe_df.head()" + ] + } ], "metadata": { "language_info": { From f7889ee41de438d5ecc11c3ea3fcd61036a96108 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:23:41 -0700 Subject: [PATCH 74/90] Moving 112 to new notebook --- .../validate_application_scorecard.ipynb | 560 ++++++++++++++++++ 1 file changed, 560 insertions(+) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index 70d32c909..268111c49 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -377,6 +377,566 @@ "fe_df = lending_club.feature_engineering(preprocess_df)\n", "fe_df.head()" ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Split the feature engineered dataset\n", + "\n", + "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", + "\n", + "Splitting our dataset into training and testing is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", + "\n", + "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", + "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Split the data\n", + "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", + "\n", + "x_train = train_df.drop(lending_club.target_column, axis=1)\n", + "y_train = train_df[lending_club.target_column]\n", + "\n", + "x_test = test_df.drop(lending_club.target_column, axis=1)\n", + "y_test = test_df[lending_club.target_column]\n", + "\n", + "# Now let's apply the order of features from the champion model construction\n", + "x_train = x_train[cols_when_model_builds]\n", + "x_test = x_test[cols_when_model_builds]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cols_use = ['annual_inc_woe',\n", + " 'verification_status_woe',\n", + " 'emp_length_woe',\n", + " 'installment_woe',\n", + " 'term_woe',\n", + " 'home_ownership_woe',\n", + " 'purpose_woe',\n", + " 'open_acc_woe',\n", + " 'total_acc_woe',\n", + " 'int_rate_woe',\n", + " 'sub_grade_woe',\n", + " 'grade_woe','loan_status']\n", + "\n", + "\n", + "train_df = train_df[cols_use]\n", + "test_df = test_df[cols_use]\n", + "test_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Investigate potential challenger models\n", + "\n", + "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", + "\n", + "Our selected options below offer decreased complexity in terms of implementation — such as lessened manual preprocessing — which can reduce the amount of risk for implementation. However, model risk is not calculated in isolation from a single factor, but rather in consideration with trade-offs in predictive performance, ease of interpretability, and overall alignment with business objectives." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Random forest classification model\n", + "\n", + "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", + "\n", + "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in their predictions." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Random Forest Classification model\n", + "from sklearn.ensemble import RandomForestClassifier\n", + "\n", + "# Create the model instance with 50 decision trees\n", + "rf_model = RandomForestClassifier(\n", + " n_estimators=50,\n", + " random_state=42,\n", + ")\n", + "\n", + "# Train the model\n", + "rf_model.fit(x_train, y_train)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Logistic regression model\n", + "\n", + "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", + "\n", + "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients (numerical value that represents the influence of a particular input feature on the model's prediction), but they may struggle with capturing complex, non-linear relationships in the data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Logistic Regression model\n", + "from sklearn.linear_model import LogisticRegression\n", + "\n", + "# Logistic Regression grid params\n", + "log_reg_params = {\n", + " \"penalty\": [\"l1\", \"l2\"],\n", + " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", + " \"solver\": [\"liblinear\"],\n", + "}\n", + "\n", + "# Grid search for Logistic Regression\n", + "from sklearn.model_selection import GridSearchCV\n", + "\n", + "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", + "grid_log_reg.fit(x_train, y_train)\n", + "\n", + "# Logistic Regression best estimator\n", + "log_reg = grid_log_reg.best_estimator_\n", + "log_reg" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Extract predicted probabilities\n", + "\n", + "With our challenger models trained, let's extract the predicted probabilities from our three models:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Champion — Application scorecard model\n", + "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", + "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", + "\n", + "# Challenger — Random forest classification model\n", + "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", + "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", + "\n", + "# Challenger — Logistic regression model\n", + "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", + "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Compute binary predictions\n", + "\n", + "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", + "\n", + "- If the probability is greater than `0.3`, the prediction becomes `1` (positive).\n", + "- Otherwise, it becomes `0` (negative)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cut_off_threshold = 0.3\n", + "\n", + "# Champion — Application scorecard model\n", + "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", + "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", + "\n", + "# Challenger — Random forest classification model\n", + "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", + "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", + "\n", + "# Challenger — Logistic regression model\n", + "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", + "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Initializing the ValidMind objects" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the ValidMind datasets\n", + "\n", + "Before you can run tests, you'll need 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", + "- **`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." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the raw dataset\n", + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the preprocessed dataset\n", + "vm_preprocess_dataset = vm.init_dataset(\n", + " dataset=preprocess_df,\n", + " input_id=\"preprocess_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the feature engineered dataset\n", + "vm_fe_dataset = vm.init_dataset(\n", + " dataset=fe_df,\n", + " input_id=\"fe_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the training dataset\n", + "vm_train_ds = vm.init_dataset(\n", + " dataset=train_df,\n", + " input_id=\"train_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the test dataset\n", + "vm_test_ds = vm.init_dataset(\n", + " dataset=test_df,\n", + " input_id=\"test_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "After initialization, you can pass the ValidMind `Dataset` objects `vm_raw_dataset`, `vm_preprocess_dataset`, `vm_fe_dataset`, `vm_train_ds`, and `vm_test_ds` into any ValidMind tests." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the model objects\n", + "\n", + "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", + "\n", + "You simply initialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the champion application scorecard model\n", + "vm_xgb_model = vm.init_model(\n", + " xgb_model,\n", + " input_id=\"xgb_model_developer_champion\",\n", + ")\n", + "\n", + "# Initialize the challenger random forest classification model\n", + "vm_rf_model = vm.init_model(\n", + " rf_model,\n", + " input_id=\"rf_model\",\n", + ")\n", + "\n", + "# Initialize the challenger logistic regression model\n", + "vm_log_model = vm.init_model(\n", + " log_reg,\n", + " input_id=\"log_model\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Assign predictions\n", + "\n", + "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", + "\n", + "- The [`assign_predictions()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset.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." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Champion — Application scorecard model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=train_xgb_binary_predictions,\n", + " prediction_probabilities=train_xgb_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=test_xgb_binary_predictions,\n", + " prediction_probabilities=test_xgb_prob,\n", + ")\n", + "\n", + "# Challenger — Random forest classification model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=train_rf_binary_predictions,\n", + " prediction_probabilities=train_rf_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=test_rf_binary_predictions,\n", + " prediction_probabilities=test_rf_prob,\n", + ")\n", + "\n", + "\n", + "# Challenger — Logistic regression model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=train_log_binary_predictions,\n", + " prediction_probabilities=train_log_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=test_log_binary_predictions,\n", + " prediction_probabilities=test_log_prob,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Compute credit risk scores\n", + "\n", + "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Compute the scores\n", + "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", + "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", + "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", + "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", + "train_log_scores = lending_club.compute_scores(train_log_prob)\n", + "test_log_scores = lending_club.compute_scores(test_log_prob)\n", + "\n", + "# Assign scores to the datasets\n", + "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", + "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", + "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", + "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", + "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", + "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Enabling custom context for test descriptions" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", + "\n", + "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", + "\n", + "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", + "

\n", + "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", + "
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Review default LLM-generated descriptions\n", + "\n", + "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", + "\n", + "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate. We'll go over running tests in more detail in a subsequent notebook:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " }\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Enable use case context\n", + "\n", + "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", + "\n", + "This is a global setting that will affect all tests for your linked model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" + ] + }, + { + "cell_type": "markdown", + "id": "0d1e90ba", + "metadata": {}, + "source": [ + "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "87f7fb2e", + "metadata": {}, + "outputs": [], + "source": [ + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", + " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "id": "d6e38bd2", + "metadata": {}, + "source": [ + "With the use case context set, generate an updated test description for the `HighPearsonCorrelation` test for comparision with default output:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " }\n", + ")" + ] } ], "metadata": { From 5b6e7efe818abff7cafed6ff97c874a17468b4a2 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:25:33 -0700 Subject: [PATCH 75/90] Moving 113 to new notebook --- .../validate_application_scorecard.ipynb | 363 ++++++++++++++++++ 1 file changed, 363 insertions(+) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index 268111c49..044951496 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -937,6 +937,369 @@ " }\n", ")" ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run data quality tests\n", + "\n", + "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Identify qualitative tests\n", + "\n", + "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", + "\n", + "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `classification` 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 `data_quality` tag." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tasks_and_tags()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then we'll call [`vm.tests.list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to list all the data quality tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests(\n", + " tags=[\"data_quality\"], task=\"classification\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "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
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Log data quality tests\n", + "\n", + "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "dq = vm.tests.list_tests(tags=[\"data_quality\"], task=\"classification\",pretty=False)\n", + "dq" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "With our data quality tests identified, let's run our first batch of tests using our previously initialized preprocessed dataset (`vm_preprocess_dataset`) and log their results.\n", + "\n", + "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", + "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in dq:\n", + " vm.tests.run_test(\n", + " test,\n", + " inputs={\n", + " \"dataset\": vm_preprocess_dataset\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "

\n", + "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Run comparison tests\n", + "\n", + "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in dq:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run performance tests\n", + "\n", + "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", + "\n", + "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "vm.tests.list_tests(tags=[\"model_performance\"], task=\"classification\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We'll isolate the specific tests we want to run in `mpt`:\n", + "\n", + "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", + "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", + "- [`MinimumAccuracy`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumAccuracy.html)\n", + "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", + "- [`ROCCurve`](https://docs.validmind.ai/tests/model_validation/sklearn/ROCCurve.html)\n", + "\n", + "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "mpt = [\n", + " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion\"\n", + "]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Evaluate performance of the champion model\n", + "\n", + "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in mpt:\n", + " vm.tests.run_test(\n", + " test,\n", + " inputs={\n", + " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Evaluate performance of challenger models\n", + "\n", + "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", + "\n", + "Next, let's see how our challenger models compare. We'll use the same batch of tests here as we did in `mpt`, but append a different `result_id` to indicate that these results should be associated with our challenger models:\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "mpt_chall = [\n", + " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers\"\n", + "]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We'll also adjust the use case context to focus on this comparison for tests going forward:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "\n", + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + "\n", + " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", + " - The metrics for the champion model compared against the challenger models\n", + " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", + "\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We'll run each test once for each model with the same `vm_test_ds` dataset to compare them:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in mpt_chall:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Based on the performance metrics, we can conclude that the random forest classification model is not a viable candidate for our use case and can be disregarded in our tests going forward.\n", + "

\n", + "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" + ] } ], "metadata": { From ebbc3a128d455601660c029ed1fc623cf077ac59 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:26:42 -0700 Subject: [PATCH 76/90] Moving 114 to new notebook --- .../validate_application_scorecard.ipynb | 519 ++++++++++++++++++ 1 file changed, 519 insertions(+) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index 044951496..e2790866c 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -1300,6 +1300,525 @@ "

\n", "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Adjust a ValidMind test\n", + "\n", + "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", + "\n", + "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_xgb_model, vm_log_model]\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As `MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_xgb_model, vm_log_model],\n", + " \"params\": {\"min_threshold\": 0.35}\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "

\n", + "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run diagnostic tests\n", + "\n", + "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", + "\n", + "Use [`list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to identify all the model diagnosis tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "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
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues. Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data.\n", + "\n", + "We'll select the following two tests:\n", + "\n", + "- **[`TrainingTestDegradation`](https://docs.validmind.ai/tests/model_validation/sklearn/TrainingTestDegradation.html):** Evaulates how much the models' performance degrade when moving from the training to the test datasets.\n", + "- **[`OverfitDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html):** Dives into detecting overfitting beyond performance degradation by examining learning curves, model complexity v.s. generalization, and cross-validation results." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Store our desired 2 tests\n", + "overfit_testing = [\n", + " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Run and log our 2 tests for both models for the training and testing datasets\n", + "for test in overfit_testing:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's also conduct *robustness* and *stability* testing of the two models. Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n", + "\n", + "We'll select the following two tests:\n", + "\n", + "- **[`RobustnessDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html):** Evaluates whether the models are resilient to small *perturbations* (small, controlled modifications) or *variations* (broader subgroup-level or distributional changes) in input data.\n", + "- **[`WeakspotsDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/WeakspotsDiagnosis.html):** Identifies specific conditions where the models consistently underperform." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Store our desired 2 tests\n", + "stab_robust = [\n", + " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", + " ]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Run and log our 2 tests for both models for the training and testing datasets\n", + "for test in stab_robust:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run feature importance tests\n", + "\n", + "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", + "\n", + "Use `list_tests()` to identify all the feature importance tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Store the feature importance tests\n", + "FI = vm.tests.list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", + "FI" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Run and log our feature importance tests for both models for the testing dataset\n", + "for test in FI:\n", + " vm.tests.run_test(\n", + " \"\".join((test,':Champion_vs_LogisticRegression')),\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model]\n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Implement a custom test\n", + "\n", + "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", + "\n", + "The [`@vm.test` wrapper](https://docs.validmind.ai/validmind/validmind.html#test) allows you to create a reusable test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "import plotly.graph_objects as go\n", + "\n", + "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", + "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", + " \"\"\"\n", + " Analyzes the relationship between score bands and odds (good:bad ratio).\n", + " Good odds = (1 - default_rate) / default_rate\n", + " \n", + " Higher scores should correspond to higher odds of being good.\n", + "\n", + " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", + "\n", + " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", + " \"\"\"\n", + " df = dataset.df\n", + " \n", + " # Create score bands\n", + " df['score_band'] = pd.cut(\n", + " df[score_column],\n", + " bins=[-np.inf] + score_bands + [np.inf],\n", + " labels=[f'<{score_bands[0]}'] + \n", + " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", + " [f'>{score_bands[-1]}']\n", + " )\n", + " \n", + " # Calculate metrics per band\n", + " results = df.groupby('score_band').agg({\n", + " dataset.target_column: ['mean', 'count']\n", + " })\n", + " \n", + " results.columns = ['Default Rate', 'Total']\n", + " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", + " results['Bad Count'] = results['Default Rate'] * results['Total']\n", + " results['Odds'] = results['Good Count'] / results['Bad Count']\n", + " \n", + " # Create visualization\n", + " fig = go.Figure()\n", + " \n", + " # Add odds bars\n", + " fig.add_trace(go.Bar(\n", + " name='Odds (Good:Bad)',\n", + " x=results.index,\n", + " y=results['Odds'],\n", + " marker_color='blue'\n", + " ))\n", + " \n", + " fig.update_layout(\n", + " title='Score-to-Odds Analysis',\n", + " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", + " showlegend=False\n", + " )\n", + " \n", + " return fig" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With the custom test available, run and log the test for our champion and challenger models with our testing dataset (`vm_test_ds`):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " param_grid={\n", + " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", + " \"score_bands\": [[500, 540, 570]],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Want to learn more about custom tests?\n", + "

\n", + "Refer to our in-depth introduction to custom tests: Implement custom tests
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Verify test runs\n", + "\n", + "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", + "\n", + "Here, we'll specify all the tests we'd like to independently rerun in a dictionary called `test_config`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "test_config = {\n", + " # Run with the raw dataset\n", + " 'validmind.data_validation.DatasetDescription:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'}\n", + " },\n", + " 'validmind.data_validation.DescriptiveStatistics:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'}\n", + " },\n", + " 'validmind.data_validation.MissingValues:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.ClassImbalance:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 10}\n", + " },\n", + " 'validmind.data_validation.Duplicates:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.HighCardinality:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {\n", + " 'num_threshold': 100,\n", + " 'percent_threshold': 0.1,\n", + " 'threshold_type': 'percent'\n", + " }\n", + " },\n", + " 'validmind.data_validation.Skewness:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.UniqueRows:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TooManyZeroValues:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_percent_threshold': 0.03}\n", + " },\n", + " 'validmind.data_validation.IQROutliersTable:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'threshold': 5}\n", + " },\n", + " # Run with the preprocessed dataset\n", + " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.MissingValues:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'default_column': 'loan_status'}\n", + " },\n", + " # Run with the training and test datasets\n", + " 'validmind.data_validation.DescriptiveStatistics:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.TabularDescriptionTables:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.ClassImbalance:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 10}\n", + " },\n", + " 'validmind.data_validation.UniqueRows:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TabularNumericalHistograms:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.MutualInformation:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_threshold': 0.01}\n", + " },\n", + " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.HighPearsonCorrelation:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", + " },\n", + " 'validmind.model_validation.ModelMetadata': {\n", + " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.ModelParameters': {\n", + " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.ROCCurve': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}\n", + " }\n", + "}" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then batch run and log our tests in `test_config`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for t in test_config:\n", + " print(t)\n", + " try:\n", + " # Check if test has input_grid\n", + " if 'input_grid' in test_config[t]:\n", + " # For tests with input_grid, pass the input_grid configuration\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", + " else:\n", + " # Original logic for regular inputs\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", + " except Exception as e:\n", + " print(f\"Error running test {t}: {str(e)}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Add test results to reporting\n", + "\n", + "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", + "\n", + "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", + "\n", + "2. In the left sidebar that appears for your model, click **Validation Report**.\n", + "\n", + "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", + "\n", + "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**:\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "5. Select the Minimum F1 Score test results we logged:\n", + "\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Adj Threshold**\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion**\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion Vs Challengers**\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "6. Finally, click **Update Linked Evidence** to add the test results to the validation report.\n", + "\n", + " Confirm that the results for the minimum F1 score tests have been correctly inserted into section **2.2.2. Model Performance** of the report:\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "7. Continue to work on your validation report by:\n", + "\n", + " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", + " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", + " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" + ] } ], "metadata": { From f81c9ae0f64b069fc9126a30533140833444c910 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:34:11 -0700 Subject: [PATCH 77/90] Moving the champion model --- .gitignore | 4 ++-- .../model_validation/xgb_model_champion.pkl | Bin 0 -> 117919 bytes 2 files changed, 2 insertions(+), 2 deletions(-) create mode 100644 notebooks/code_samples/model_validation/xgb_model_champion.pkl diff --git a/.gitignore b/.gitignore index e1f129048..eb447b41e 100644 --- a/.gitignore +++ b/.gitignore @@ -191,8 +191,8 @@ lending_club_loan_data_*.csv # Pickle files *.pkl -# Sample application scorecard model for validation series — do not remove! -!notebooks/tutorials/model_validation/xgb_model_champion.pkl +# Sample application scorecard model for validation notebook — do not remove! +!notebooks/code_samples/model_validation/xgb_model_champion.pkl notebooks/llm/datasets/*.jsonl diff --git a/notebooks/code_samples/model_validation/xgb_model_champion.pkl b/notebooks/code_samples/model_validation/xgb_model_champion.pkl new file mode 100644 index 0000000000000000000000000000000000000000..bfe7349b673d72f09dc01d5d1640bd3d62a5aa07 GIT binary patch literal 117919 zcmeFa2VB)k`u|U{fS^jcN*qAPISv~h=@@5rHk%X|=_=xgs zbGIffyqbG?d3!haXxYr&-P_GeE^eF<7CAUFF<8@j`2qH$9D->}k1#5<&<$Q4xW9}$w66dw^B6A~Sf(5s%DZd_hjwUC&Yq>!lK$e6I; zVX=SfmDSLQcv@SS=2U_c5@|IFzbjBJB07$CEM`FBz~7~tM8?q4QBlzmF^Rv+FrWhy z{ks(Nfw9pM!Lh^Wz$OfgjQd@lQCw1dTx>$b?-Gq;<04{$L&ARF)vAfHw88%)**KDR zfL@BO|^B>P6~DR*0+^St+t|q(P)%WR*yx zNaM(=k<}tiB26RBBF!T$A}u4UN7jh6inK0oRMY3knEtVUtInEM_fP#NS5{He`|m0$ z9{}Uf5SW+|kpl)MCiJQwUjAp5A`SjhL2a|jpB=Z?UftI|MtWvy!xT>&uk`$QTpfLX zo4U!3TWZsGyQOg-KNBW*4oh0xV@sz!RYu;hRwsG&R{N{EsD1jLQde5sP+jXxMJaIF z0BKgq8Fug6bcrcx%62;v$W9dPl)c*>V7ZohO2rCGWk5v><+=?=)#;=AsH4lUG~4m4 zx{dDvHCH`FZM)~2+M~o?+CJ;By570D(m8_->WUXzk-gK$NNY<+lA|vorK(kTDh=x$ zlddc@Bu`x%OD^Z$u&a(VWqK66V2eJCRC-?Ts+@K88MC*nGrM`Bcm%StGuY>GwbQeh6zORkFBE zmK$#-GhM@r77QX%wM8G9K9Ra?yBnvO{QM76o%_$pcHK&3+$k5ujn`{Q@r*P@VV$?W zPv3814V;@0=MGPqY5mVI!|YqJ)wrF^wIjJ~7UrEe%JDPQnYur|4by0iC$!QWcdJ9g+2y{4UCKmkB^A?ehi_rx)gm% zMM`CgA;pMNm106MqgYU?Q>-X8DYYrKl)4mqNS(v;#(X+~*IX-V;> z_)-{(lEQtXfyS63Qd&{kP})%fC>zhM z;y9c&dIN48H=JVz=%6=?Qu#$MQuzpTOWbf$CVy9q<8#dV@(V8_G&)>I+4jB;G6>f} zcRFt5zRl3#7Dd^=X0kyP% zIRk98fN2BjXsJtMz)p(;je&Yv>eCo#pv95KfRh$y8UqcrxX>7I)zXN@Kw~XUXbd#f z;znb@U5f{efo57fX$&;i(t^f7OD$eB2E4WS&=>$S3@EgKNd{Ozo)FdI;c@JTw-1;D zaNXfr!}W!03fB>?9b7NCMsQu=TEO_jn8P^3*xFHG9AWHWykLxATwpBV{NbG8JmK8n zeBd0Q-_T#^BckwN!f7l}{3xv{Z7IxV zM2VurP~s@@ltju<%5cg^N;0(jGX@w~2&M^fQFI9%78?^D3Cr@|-$K}R`n{wMoS^J6 zp*@?uWuuJJY4S*EZ65~!w!cR%L;BjiC+?#LvzEc4s+Vyb%dV^oX%I8pzI4&#(|x7h z&{M2_YP2$0oT?~kGJttFs6>-*AJmNLyzUD;utOTO!BqODfEenPk7s%YCO|o8Lq^jE zlr?j_u7S1zbwWPGa{HQNVISJ0bpV96s~i%S^|L>)SN&~f(Di3z*zcdlQ|P<_10frK zm`8E|BLaBO z0C`aEd!P*{3o!t7%l&{@ZeMQe`#R*~XtAF!$OfazZ_@%~pZ|ReCZhx@s|<{wN?!4d$>QoWE5y3_eqv7P z7tvyZuQ1NJop6vDB{nfyCDg85D)jN5k5g;x5a-mXBvjp6S+p8+8tYZ|7SXkL!pZ)X zg*WNz@qT4}!EyCryr*5BkiX+TE-?2OdL|nPPcH`WR;B9&vusB2&_lS<)gk-`9Dwz+ z4~eP!Z3SN43S0Pm#3zqT7v9|-i?2jwaSsMQ=MNS(7o1-#6>Ahs!AD4tu&~7@q59EK zVfDPvIAe7>|Fv@lzPaiH?%LfOd+X8~b6W`AdJ6blQ!C!K3BpsN#^L$9I||0N%7hH( zFG8n&i}CJ5<{0H27u@;;pi?6-A9F07zu6_9e-v98T@?m!(Mff9+O{xY$$YWZk}lX~ zZDYaOQo-Npy_?tbh`^kfhC{pL3$32s5N19x$DX!NaC)B^T$?z}>=d+Ww)q`jfqQW$`*;ph-5KF{L%0W1Nh?%?cE{MjhvCI+)`D z#;F`VmVa`17=J!18LvHXi7#?<=bw*8{4&3JJRJjmzN;ty^_v$q-O^X^XdJ~K(2c@N zV^ru#QXP(_^J{xO!hUNk%4e)b_WHd@X9W5iRrDefBf`rSI9)gS5KXpL94^n+bN7(* zHS(QCr6|wUclT&6XPbiJ0J3!d;Gq#=RE91urR$!O@;?h@W26>67f7F`e`8(y=BO3@ zhDbYiKV*_>4q-1hXsY&4yQ{7ias(T4&bEYISA0-fwcL}e`H)IJoC#$6OtNBvo7`Z>H}+w#+`OuEX(FoaCNyT( zsOOQT&m38+;M=UP<9zmZ->vNDd$~$xcTFiRJXV^#BV8JPXNDviY><{seJh17F_*fW zLXwA5wuJUiA}+HkOO>L}N`AZBOMx>o$-~SW64$aC$@Y0G`DSNH)puu;GmE;aaviLt z&biY_@r}x4u-ybj^@3yMm`flT>f4%h;HLXdSbd&sPc&gZuUn}&XY_`7>y$&@pK~Hv zOX`yO7T1f1M4cdy5+H35@xyX=hoXM~z z5k~CboFpkd{3uh$r=j9YEW=i)ev4!~FJpEe$WqjE3uZo?eXJ_ndzjgjf1g=BEQM`S zxheDT`c>9*;6v8-xWJBTe@vc4e=Wgm%zyZfD32OQX1emuk0mqcu6)~vTRc2Az@q>> z62O`j)~T>Ih4m<`L1EnqYfV^R!kQA+k@l2&lm-+hN<)e(1=fGC=7V(}tnFYu2WvQ3 zx4}9M)@DiyN5K@4(u&fC(vA{9=|Jg3=|bsB=}zfE=|$;738sWl!YC1x0hCC}U`jM4 zmNJBrKuMwuql}=8qNMzc>z(h2uIQuN|qzkSLN zo^8julzjaC-sa}`&lqwLHJ+}M>@xGm`=?9^#G$ktki z$+Oo{#&p+Fu4|a8oKU3%ndOx!-}ScREZ3HxITAwAP6#b--ogqJUkY~pqNoZrn(0TOT=zs|UKt2KVwaBX;Ff9PcDgc-n09F!! zp%!_?1m*`Y)>4(m0LU-^m?Z!t830TaV6Fws6#!BV0HzB7*#-bJ27rVEfJp;D&H=!@ z0U+%FVCn#nc>pkbfSne58Uqem>d_deucZNv0Y@!PGzOftG^8=$qQ#ZQKqD=UX$&;c z(v-%4n-+H(10GtM(HQX5(wxRX3oR{a40vhrrZM27#h1naNK*iqY5-)ae?&?fL(9QD zfa?#}8?G^2TezNZ{otCxwSsE{*8{Erj5~}qjISL9#uCO4#tg;@#sGF60s&%3lLO?*KwA(X27?7RC@@}d=r$ITCgJ78N{BOw6^bI-2%hJs zQ?c$0_xaO7{8?E?sNLv2u62atFK=+=-?bPkMhvB8E^oq>97FN3J{5!qc}MVtrG?z> zItBdYhu(rsMn}h3hV_Om{ z=ta8V%l;DtwnMbgVpId1{&@wyyLqHgq_*epJ!;GuHtNIeh?s|0%xr=ib}tiZ=GI1s zpECU2CwaVW*Ry>3nkBqPWpl1Wtq1(oE{;Om{!>MdE6auYncMjEMHhIJF8%qvv_D?= z^l{kfih_5=3}JE&L+qct6New!z^!b45(>*op>?? z>7@&NuRX!I!Yd`#xxAM59oL?};`xAQvmfA^Rfppp?dD)Fm(6%|Oh3FnrZzt2CSf0! zSfTRsiSp#A^6wJcTFNc{--~Tk9%o4xw!C6ioikR~VYf>QDvu#;bR*fY^p)x+a~nvv z9iB@ikFQIO*7PQ?9~&@BEjqE48s20V$8}J*$SzeIf7>9vTYNxT`t_!CVsHj|EFO~3 zhvzK2rIo~%`YC3Y&SDfU&g}TeWb&$03%21N59U*t9lQNxnY#61Z*`NfMQSXLQX5ut zSL+t7l}b&fN)LDGN}UfikRF7ukam9PrM6A+BkLTeNUEhtBDgzmnTrfgY3ig+8TVd#8v;#!pCeA-5;(kPECckU?} zO|h4(Ud&PKUe-pD<#3BEZy6&wyqzoQPT3(E+`{WEtKdDK8`)0208iERmV43j9sC?hDND8IxuItUPcEVc;+q4jawjE=bb>1^KI zpdTM(n$P#z>dtf55oaB$gU0Hnau2QZxxm?FyvhCsT&D-YT^KjmpwdPOG1) zTxynzx)yCjS1;)x)yP$x&JKSxIm?A>rW40yRvFA$Oz4a16mCSRk1ug^r~pYx4sPoejG?#>^$~qJhXOq6{?+FGPE%hN7I7 zb-8gl!_lPK2I&5d zLx6PH2Ehy3f_>Nq(M#U`UI>!~GWmEYC$|B@nGB>sfV$*1pez7sAjm--ve*Xu5FlT! zTa)lJ{!ca_e_63@aZIwGQPeuWmT_&xj&tqB^^@9)svZx;%^7P%>zWIMP(4#|z)*eR zkyuYm*q9?$)JLM{*a*>W!VBTetmXJ&d?%rieIMLC{jk_!&|+ND%S5<#eHOnz=%iTR z|BbNl04rQSVvkoo7>PIXtXOB1t)O=biza9L3DxiK7FHLC+^xfp@RjB>1cP@W*m%VY ze$S0W{!x8V@H+gEchKPtQmzk8I8pci8&yy|Cw`>Y$yqY1jWXy^>V<~ruT z3{(gfyG(>;K?!)%k|BJx`i^+&2$8Gmm5ok}yv|o2S%@7fY{Eu{HMo$}A>55+m-yKo zGle#;eTBo}k2t;eXZXONB>qvITX?PadtvtHFSyP{Z+_d_JwnIh`Q>x>uP3&VlXmx{ z`PQeHf)RVvdQtzJlaZstv#|}|9aql5r=@=#*uxq69AIwNaE0;>Or8H^rmQ=~l zue(%pQyH`Wv%R#BM36`CPLT_pIo96zn6yOQL~=MdT#DSVUBRtCEqQnI-1$}Q&fE-X z$-b_tN7Oz&nLD?#S;LD#Y)+t~RQ;U+TgmVd%Ukqf&)?fG#ZPD;wVilPazE9K#3^n` zIj#q#`qNKJerpDjuQw)5UPyVoz1c$|;WYxo<*NE?Rm*MS6<2%Z zGDXHa$cxo?nN(duj_x0%y60&`cCBKHPk5|REh;#|1P`7^#s|Cm$ZL<`d zg8?|t{M1379lyl3Ut-%YvF(@G_6IkRKTeVj$NwO<$uEn-3!UG;EJ{7kz8BjxLKX;E z;8{k?fuN-k-Ke0Z5y!MIi)utJS_aZI-eq)K6Kg!!)_u`#^fi)|pN$zEaK--#jZd$A3~DG;z^feQpH2#^lq7>HdU*vVxe4f15s4Yr{Sq<=4h z$@?@}Qx?k0Vi}~#0-1cgoF@xtPzJW49su$I*oI>uK)a9zu`J9%{8IaNFi{!~yHC3VQc;;mnK)!k3ikLWh)(LWRVZc+uWo_`>N+ z;#8yFV$Iz7xG?R#kiOd&f4p%{m@)4ar!oz|dvkjUHQO}eDqS+1ILnZA0&(mFbK4_B$)_)gK?>LswShL$am__RqSD zH5!y)hxVrU9d5?&j>UM^{fS&7DV4vKbC}P+x{A-ZW2;);$PjjZH|y)>vSe31P8Klbz?>LQ9oaYDVD6!(Y|WSL2aZLbyKDJ1ySVP z?Rrw$p@Sr+bEeeTY7dF`I4!k(zLc~wxIhwac{AaGpUIw+H<@v7?kG-EIS32%?Uz*!zw$$C&N-`p=n4>+tq$!Ui^7gW^ zq*Jw-&)g{nQdTXN| zTql`%xnPazvBd`F$kC3hzg>XT<=!o+!iCY&@SDpdkHooB{hg_jp5r>n{n2&O>+>jP z;(@`EkU^bD8Z9G>44owdQJ>v#J;e9T;UQ9o+on1Q7DOWF(b{NZ(yQtrz--Fs+n=7^O$AJhA9@>tWeD;j$^pCZ;Cc6 zcQ9eTrzML~nM}&W*5%Xom)Q1C+Mb{)|rM1_uFw^{f==~5hKyNOvG`M;*=ru zm)8Qs5@hI>iYC3WK&J~4nmoiCrHxKSb8Q--9aXMyRd&8W6Hmr*-N(#8^WA%KRaMu~ z*qx4?`MMzGn+qk%k_9EG$@NrJ!TvegdZ<8|qfABQb8S?S?S#e<#OYTvKwO_>G%ENd zimK>_Jc4vk{ViwEfXW{c+sqEx>1{?^nMvqNe+Se;w-O4pzl*$tZJbBId?bvrQGT## z%^8e-iyS7JBAbYXh&eou>+QUp^Y{dhz&~}6XU8wG?U&g0OKkfkw*A2k%pFEa2_0}%{f@ze@# z8c#R*byF(Rfe@w<)wEu48m}|D4fVse2G|B+4LsziD5fb-kAr?dnkJq03+mP!4|Rf& zrWN#L&%EEf+kWd4)Ti-kqh&PtR9yV^X#d-t_0PpN5XeAO``)un7R=yy5UfDB0-*}F z;f^hfSn@W+Ac{d55WZj=L^C)}4&NUGc~BMtl!tBk*zfn@7`ZRdhAf~#x?CTWhdekQ z(&T!ftSqd_b<(8r<3LROse^1l{<30QUi~NH6Yi3q_sS>2=Zik#dZS6=&c>I-qWU_* z))5bcYEpsNuVkL^cI^kz!wL2-Kx)%xt1(moxGbf6z9z4YR zw@wt;y-C9Cwrauk<{;r)*;!%yv0Gw=Y!<(%xk5lI#|Z8Cy<+^TkGM`*JRZL>Pl#?b zf}gW}6dxE~0l(dzjXvmK<;HiODUh9E!l_zA@btmdgDzEczVa957UPkZ zU-JRJZ}8|(5&WSQRrt3}Z{z6;rt|YF7U4d}p5iP09`VzjUJ@FwyDW5_?9J_%ohF(N zs#HFA|AJy$L`p5GdB@?5-I9vZ>RMmP)3&F*BNMWilAvAEzA64vcfHlpu*Ux6fOP}1 z?0r0gQ(iLerK{NLM%~n=Qzl7L=4WYqVSUMg875`zOD7(6)l&UyhQu%9y5znPkq&Il zBKwvhj5^Lx@?f)>h=iA9VBJNsBXGf1Z~Idf+`Ghjr2a-h%?s zdH30>m8*gjD@!k`UOVM0F6(qud}N+6n;P7e6rv$naQ+(Eu;3iCzpgJ+dU+A4RO@b0 z>!V*4&(<)ir=d%j?w1ZRdCL+Mob4p0{^Id$@$prRSv?i=*<)AvwEZQv{S&uddDK8` zt0?dM65D=>ZGW&ZlBdU?h;6;Te&!eQiQG11!VMkQkL&-Yj4N4qlJgF|jS~Hk@@e)o zl-@p$OYfYDj@_{2j9RJCf#`Q!;9G=xE!ROY2X#>YYXgzXGB@PCBo1x2P2(zG+lb~z zo1trW^Ej8uce&cZ#whuK7g|zr9*VO#g!X)xi!QaXQ?^@^idxSZggl4qprK!eDl1hj zL3N%UV$Vx%%DIm$k;{2IG;4TQWD{P1qVHco-RA0`E(Y(>)MpbpiwEB5+MZP8W1NXH zdyGS)FO{O*Zs|u2GTXc81S=g6EpH4UH!oh;&ef zigz0CG_7YFl+heV>xXSC`lx3fq=AU2DMz>ASdDO} z73F?C+W&UP{ByBQ_H6rJ7z2?C1S|-$hy?k~LEoS(oC9nFP(RcOG39^hK=XiOGXffYjR2>KrcVAd0PQ12DI9jEl;Mv&gC8n$c3-!^I|NCy+~mXFtQAxRV5zP96cJP8r29?ufJcPNBvZY9En{_lln zUMk`9>NNhHrL|aL=XzmqyE0+^^P&9mi|@GC`#K5hClAHC*I(m~5659EBOhLWaT;FQ zrW5{PpNTtdtIJ&-(SlE#YAN*iq$jLjJXQEQ=e`iR&JHhY^AyK4br4eT`w30HgyTam zN(EJNTf8)-wQxGyjo<#QnETvm7|zUoBHWwS4}TbLkFVBUBseJz1oOqW`Irxp97!@j z=dQKj)K&Ki%?t#=#e602z++J-ai9U!@y@-d%0TNj)3$l~*7f-J}YB7&${&`{ong zGG!MwI;FPhwRHr?YIRXwVrLxd9)#Deu_>Rse>t%&g*bcOk}OZZQOsL?NSbcj zgBjBTGi&dVw6W2xP`b*jB~y-Dps>CEk#R#J*yl8=5y z4>JDzH75JK3;W)9iPT7;uSiPi&o1lcmvH zn6`C>5~fZGxs&OmT3E)AH7`1_R?Wi{RlaqXj9bhnNeerXHkX>KZZ>z4>a42HI(!^Q zS{3wDESR;CBusj#n()#|b@Y;vZ|<75q_je9X4|$ds&_htOrBVklqR=i{D)1I^lZ+O zv`MeX$d4mc*~6czTIAGWOtRWbHC{AUz3rnSx}M_{s(bH=qgj#SblsVX)qB#IG4v~~ zV<#Uc%k$0`PszW_Fp*J=TeTXhJ1@5TF8olact&5VtXSy*ely&bQkP;+sYhu*aiTP&xKiMEyy3UI;rF_mQQ&vE;kUTq_qX9U zw_)9;q;M2W5h<-GZ7A*F>7B*_ln#_mlrEI6l$18;Y{#1F6{7_m;;U5Veyz}RNT;3@|&fsuIu11alDs0PgwQKc7UQ$1$ zj=@H>^HeJ8zUd>HbEG|*=syu%oY9u6YRD=_uP#B#OGsI-B2tdZor`+!EkSKl@{!^A z3(EKFwjrg%86*zkk?X^cXaK2!UQP@|2fVwX?#b!s{WVWzQ3a$tRS%)K)BTX?%Q$4b z;tEo3?8k-;>rJns4sy)why=UmsLGWnG-#y{^6q>axwKFzAM72gJYKOrnldC@SyJ~s z8gJ1JHSC;*{JTv-UC1Sb1A8eS9&v(3?Ts4GT!r|jJy98I z4Q>34L7o+VB&q%SlIZ`x&me#~1e~>iX#`xffSCjuX#tZ7z)$}JU_JqOlNkV03cx1; z05Gcnd>#M*6AQqn0st_#0DLw80MiS=Cj15CJt<&J`cnE)LMh>t{*-}~L6j&;3?+^d zPf4T;^KKrXBEugSP0t z#wU&*tMQxD_7l=HG8?>PO3P@D2e}N6(Y7f|aM0Gz{=oiY=?n~p5e1BdA?5o3=@7t} z$mr2+U5fl{YCyLuYxf~dp8^45C9$0UP|af}a@h*hhGLWU%nfb%2nU=_wvRyIuGgy%jHC zFiPmC9ECqtb4Ops2Z}9>MJ#<=EnIsV&0oEK56`~YQ@H4HLeL3y5Zu zwSqriGS>%>-~JQ_^u2(eBzDEm5*!2@vw>Xf$4_X-kt}XRVS7H;wmEmF&MjWiZwucb zwE+99OvfYI*$C}Qh6xqhSz%+_M%;z%ru@n&Cvo3x^Ki3-0&INqHGj2hBwjS|CN}7_ zLHX@%A^&;Q0ny>O3r@-N=UYzC!fT4sg_%9_M7__c%93N75Q)y^Q}PG!xjiTH=Zp-n zIzv~8diWWis&QAI6#ur58~n?@B$}dno%c)%=$WQ6KAtHh2302SO1hCi7eB?j)IuqL zO_0>R&JZcOjXk;0V;Y%jH$>$xrVvvDU52e=C2iF&lzJ6sNb?WemzsQzmfqU-lMJp{ zOLe^3kd#KP$<62O33FjEsae^EvGGnO+f8RMsmt7UUO61WWF0C|SL$Ie<%LZpGivvj z_AXFL*B?HU#@sz6`99vp*0LWe<;=s<)CbAZtqMyO^P4x3hE>@?%uo0#KAl=X&goy0 zID>Cw#px81^T^X@L992)J*lh0dzbkN>VvG~0#|86wasMQ=QKru_feLg;-J{wP=|!< zyvA5m9?6>6?~!7yu1N-y^oet|f#l7zEN0o_aAITYD%m6l6+1e0lsw!D$!_z{l8;js znK;3Y#NQsHD(W31RjT@mxiI%5`TD9hvo-9H)aLbi-&+f7Fkz2M$?m1~$eJ(y#67>3 z!mfG{>2C8;)opzmo7%91ytr~iHR(wj*;X=F^~LTg8TRsqYQcerq@8K5YI;F7u?@CT zOb@TAI(@mA88df*YWEf=MZ=f5s;$OeiglhR*}4Yz7@hkki2kIbQr$|+iR0Bz#kF%& zRT*(9iiNKSG4BqiDb6GfBhshIBXGTe&;hfvp*fx9$<%DtKLO= zi{2x}8B^ui_QO##Cnai8@4a%NzAu`8xeNtOk5k^-Sb{Q6uR@oru0-AYvB+bzf@^9q zR5|i&8S45qRXOL89kO0>M!BM}qw@5v_sa2}^xL0vpD8zXnht*s%1ud4&lhLd|y&}TbjD0M{l0*)c&qt7U!U@IEYdJ>w@`4bxBm50VMw#t2X z{7{d@H&N(C2_0G75)}_PgwC{g{)hk6{>uu=v*MRf_W#cJ?3YmXZxYJr-=;<+)80OQ z31$BeZXbW@YnB3^D**8be6%#86ZIMUO-!Pql-9osL?_u_OZFiHK}zFSrWL_J=zo~|@@f5WE|3Pt|LhO!e=d~C0+}3xumnLCrC=Wf zCRxyu1tz(ioCiV|2ySp32vrcvb;u&x_ktMI1N(41)CI@NA{gw;_0Y}VwGGF~eS%{l z$Z3!cbwc}c4E=&_$dl`bH2GM%`Ev#N$>J|6ludiNoPMp-&oA}PRbk!O$AaT@rTFUo z5WmXD(=nOl310){7ta46n3H@l@2Hp9qFcOhe)lHv;`$(*k<^#JGPVbvZtEj7TW`;; z9?_T|@TC`?*f0~%>$Foec)w3{Uc6R#J@lavH~5V(@$NRE=Jq#Y1wB7;_PXYLD%%PB zAD+p_n^x!j-Y(=0XxrK0PX%*3uvV1PbJ(#z>y%yK4lEy!rUW?z` z<~pskQHO~7|u@8Vin zRfO%_IKe){TlbFNYq34K8cE;uYY z?7z#S>|Z97C8##NN|$P#>{NX0K%g|L#ucgJ1}oLsfrA*Uf@M;4Ag8uAT`Eb{^SN6hTTq1}Q7@!^vU# zTV;ihhN&J|lqinY^!9Pvoyw#w`K-9OXdZLY?UK5l?rtgV(_5(^!$wk$NSD%PG$-q- zJ(h-t#Y#q9H%ftStH?6vm+GdzXBeB^Na}rEN9psyTk;KmPR2KWO!fxt^yy#Pj-1;) zPPO3PD^-uZrdL`;B=62}vfcX(Y15)JY17k2RWi7eWVn4H<2!nw@7O&V zq-MzwrfB~IY4VgBs{3|LnWKx+*`yH{Rr}rJB#V2WNe6dVQe)^+pZB((ijTW1h{w{0 zs)6&v6tylGk)g&pzRia(QRRMk@2%gXgy?>&?b~6_Ek#S?`HbuQ`LXN;U6gOxaEpdpH9U^MztvTxm{80p7L@7~D@sjD zZHg_W+&@gazGy(xoG7rq1IGt&b^r$laB2WY25?>ghXrs_0LKJ|Qtl%rKbq4tk>&M#-_+J zC|vn0F%@Oqe!!XDxq|pZiOBnqi}FmG4%@Y?pEBXADU)$Dm0l4YRO#F_<+wa+B(AG~ za)xz9=Z7bwJ8uJ()9>Cu-UXM?z#Uec!=sldzxx=}ae^_sr(v9Olc6rcbNef$Byjc6 z1m)y7H)Tdg19ZI64pe=cpYp3`vT~Wv3CL6i72QPe6_A6H)zzu;Tr%4D!|fBbh2; z44vEGoi+T41MsviKmEg#IXsO6@TU*RD+e$G06eJ!U=jdWDF9#|0Anp+Du8NQz-$1f zTEK+JW&}(L_zMAGQULHB0>Hcg)>^>S0JXG$*#W@62mliVfUgk%<_NIUVozhhK}!P~ z1CCn2Q~}Of8qyeW(b9;<0GKhLi54(vfSVRDZ-9puFm-^Z7BG823oT#*0r(>Tz#IZT zTEH{{;0L9UFLbr|$N>WrgTn?!Muo>m#C-pyodjAAt}9$ixQ1}e;5xzefolTS!JGnP z4&w~t3u6jn2;*i)fibENCXvQ4J}@RQ4siZ(-f*sPesE539?);-FZ7Y6APP?rpbiLh z+mF(k(v||J(Vo&#n+JcGfs}5PAWBb4Z%SWEKT0SioYJ2%kTQr8MTw!rQQ|d?{1^TY z1`5KT%1Q|!EB(Enl}1*g@{R2ORPM6`at-y@Qc~Y6jZ8z!Xrv|ZaiWj6koKEjRaYBX z>XD@?kanP~<`qIJ?Z`1mNg7`%ka@I^x0=@xL5c!szQ*s1Zo}hkxuit*HFZ$AO7j{d z)I((^c%894o%Rd*4*MFp367z@UjOZnw?CGP;13)OFa+6XLkvNW0>&B)I0RXKk&QIu zK^X|J4FO^>5@0Ok#n;eu;ZVfRoMr|LW!lxEi;ObW~572H6D=EouV|k=XEl4 zY;zKqpI^+Tw3HUd)+U8IW68$6evGr_dexJVG-i~Oi(+p#%<8>)Ev-6WCK(ydV|N^_ zBQZCfrRG&05TisVGNsKD_UapjlxSflg#|2RtBvR;*+$$YuX|P`ibMy+jA7eJ?BZr> z`*V&`^wZv?%GJ}-v^+zmQN78M#{~z~l2RMCalPJT^Zl8sgkhIS@yWW()*6LHEsmXG z>@A9jX>DuP-q%a2z2YsIJmsOPU;H-7`|&U$G%aI&RPC9K*+yhvy;tPQ40qCM!Wcz? zK~*w!>seCKBh9x@j{>R59Se5by->-#`WIry93l-y4`;4-IZDx~q7D>6vm*B6TQ)2-;;yP9H0Rt4ggR+qfU7)oj{yujEbc@iY( zOEz^+ky^9*GcDS=OP1X?5Nq;UQNciu=+qj;JTOc5O?60TjB1`HCY_tBF5h%1n)dPn z2~uol{BM^j()*oee3$oBB{{Y9ef(yGYTS?tiYd<-)jXvId#v<2llA%;qx(p=e7gRn z#GjOZw+a7Ud-9lp3}h(p{8$Ep&VGOEh1)SaZonf1tm|McC$HaNy#{MESeLjiK5`X1KSqn5K0wB=bWrQZS;(`r8#pC*W9E!$nE1M$w{Ef<0i+7=rJWplh#MVkaLO{6Z3+1kElTcL)f8@Gs zhca-N4(j-MIm$6CL5*YgBfO_bd9w_mx|4RH9o+|`%#KmYPhCo({+}_(v*M3rrC+ZJ z{<*va<`1Z^1xz6Tath$A1xzF0q6N$(&`1lIOaNpT0L&-wZ;_j5J7AIlkahmjuL;ur zx7Xf2I@0a`Ysrb$kw_Uz8BQ5V`SqIMKlGX)NKzmNX{0IeUZRqZM()wdNE*oqJd)to zv#9(d|B{u)O9{NE!25|xUQiCCDb2CqX{42ne)~17zsXk`&nRu(AaBVsmd3k@N>7jm z5A2$9Adf-2;Q0i7)8@;bRPbwBnmpM{iyYIoh@q_925if9 zLw!&e1jvVNNP__7pe_iI2itN!v2m?2-^Kk zIJN2)yu{+UVCB43oO?D?EWW-9x7@GhJ8u~+G*~f4&=09XJ(G^$p|%&qx;1VHUk#1; ziG8o()MRg*ld}-dQ5_MJolgn<4W5b1mEDELdNZ+Jye;3g^8_w)d?UWv^ZSDSoJ?%c z@|vJ~WenOaHs@?sjN{LJzRb_RJsvx)$;DS<97X*Jw!)5gu|k!4Ex8VB4)Fa>9mZ`< zUh%yaf8t-hbHGj4t-&3S8u9yj)D#4#V6k1DkGM)LCjqBN2;Y1oME|l+{F!bG1@mEQ z-kTZDuM|h~EtYpjZ$f6H%?5;PQSl>syYv)S<~5JE=o^4v9}xL%$=~>9cShk2cB^qi zw^gW#jtlBGQORBAAMp8&?D=Yj4!o=SJ|8+sm%nD_g7dAu@h^fF2>uo;g~bLD!f4$M z_}--uT(8Dcyu0W<=T$sFxjp?GU%lioH>*Bf9!8GgXZqjg&AR(?Ik{=PqERHDQ}&cA z-Xvk`mI=7hB7Humw+Y{1i@=9}eZn7bu@+o;8+_VR5Z!hR;ZN+l$yd5^5q~~C1W$DD zDYzQf7FKNy5)3!A7hRH^_>*7Fm9y$u;?x-|UfKuAlj7eclzIHigt7#c@90Fb+AqDh z?}aqUW=l^pso^7{pK7EqeHbK}9G{}r`OsI=Ej>tjcRxg)?1)oI6E+cNrja7&Vc{hiX^Zpjn0%EL~ z5O7#BKXy&g-_?>tTO=vgju|0M+!rBr8qkzYwi!&;1nnlVE=nfi)fiQaoky6_mh^YN zS~Mf6YxK$Wv!8r>ES;@t*vwY3OYB^nX?>BIVKGRmeCG+%%C&v*xG#+eo+lFf_ID)4 zxPtGJo5|9k%k>mb4`z|FWAmf|P3|bN_6}t_E$KsE4{~KTH_0Yd#s;Tmr;i~Y(BT3nVy084ii#>V zs;Oe1Tu_W(|5$Onz>DmAe}fb_%x4y74pg03^ig3vaT)P6Jx@?a(s^KSnid*=aGMY867P@-yM6?4Lf0T4t5vv9gwRLnW&oH1b5QNgd? z<7q$l?l61r?%kQ4aq4+~R(Ex!4!l+M_w7DtMtcfdE1IC|k9kbhypQhpN<^DSUB*pj zH$|`8xZr_rcH%1Cub~O|HltS?R$$%Yd*n|3hk$iq;BIDmx~@Wuf^Q3v2A1E8)0;H3l9(E=Vj z0I2Q&JPd%XmU<)xK#d3NwAhmvaM0pNVgMcr0Hi43pOvFz8OoEC0lc&{Co#}MOG^?1 zt+cczG2pGmhr|HZeI|tJ1Tz9$cevJYec_tI zb%bjN*UOFo*QOz%F#)a#TniTh$Zv4IaGr2(a6WJj&~NB3^br$S0#ES$M>9&&{0IRA zcxKxXIuL>hp@hy_^W9Y&cPD8*2)zh>2>l5C2?GcN34;jHgrS7tguk2@6kH?-|D5*| z$TjlIW*{wTq#^hPLXdvQF_3v=35g~jeFqsw_AzlhwW*@pf>8^p^>Q=V-9Rj0QV%SG@94iXOpT>|`}!yh@!*C2q@0`oUG z25DdfAb=65qy>%vP!94!88`;pa)5dv9Rk!L&k3O(xh$0b#~`->F_eQk1JJgd2I+D+IVMSejUfMlS0Dqyz5cy}+@*@dpHaTEzEtBM7%F`ex_EJ`27B^i z^l(+$)@Ht*<+j|?wF-A;R0vmXWHBd{Oyef4 zU&|*3*HsNTlC1LX?7~O(Ojqf;UT2P(N8&rDkFjqf1a{$#olLu?5zHO~JJz~Pft!Sx zGxmc{u*o%UG0r2MSw+WJc!MKiN*WumPU~y3JEzy@%#SB=LC>c$6Qq7z)h^y#%CG{q zJW$CVKjX>;L^WWZjcd!!Wy{z(mj-jQ)f4%V^EUJId&aZOo!VT$u#SAfkt4kM!$9sJ z$DY`Od6id8m; zEV)~IBe)m8T;-odd2?>-%H^B=dx<*#KJTXl!PsGiaIIdfQjxn;Trek7SaEQK(8j4) z`Fej1QP{s!)MNML?^rRYB#)lIx@h-;^Pkm~QlqHsI2|E9Xc)EF$4okuTqHi4 z+*v$XE*maD7C# z7#dn%k<{t7RLk>-(qe6{n0T@lwe|f8rP1)d!Ut)Q`a$Ygs#?@&^_nfi#0S$3tK}eV=bQokr4IVI2suSBv44Uvo3i3QE zR41a)*vgsIyxikrk#B;!xWxvkjmcc9)q?_xX_rMmDY6otPOGAbYoVu@TYm!;ls;FO zEbXPP>opZFEpk*GZuV7ubH1fAr*b9}@>K&tsH(hEu^63!ChTX;{L@3ISKZrf?L0q-#F*H|yA9nPfizhg|#!EY-Bf%vfcS({^$;U+8 z`dTwQ#&t5f_3SVXd9VfRJlujOz1xVZzn_i=mh{2>HYA~e)0*Jv@o&++#vRb99vg6d z$LqM&=w|rvzIc@KCJp;dxj~+gpE}4-$FB)a2?^wH<%g>yuQlX#0hkj2=J)`t1prtp z0AOZ*ocn{x0btz$fcXJTw174ZfVcsGP7Q$20f1%=fan2OXaNBP_`LK#bCeK19sYh(sV5X+L)k{zxE7X~Zrf zDrp2J5WE2JVS-odi0}lW64U|50T8V~MAFm;`9N?2NM4PY1tJ%u10asUF(N=|`T>HM z_DfJ8q=EmFrafqr^i3mR5#j2)sPr?Fp$^00RRe z4+t=jKmdaRh6MIyQ3_v;*Eema;$J@G2d3@d z_EkH?&!~Nv_pnIh^SsxyS;=*{nOm-~N!$YVtL=04!3rxb+1rTe7!ioyXTIduQEB|M zs*&uGwcfmq)f_%0-kiJNFrGg;?Fu*PXbE3!@NzEDumRKTLNSx$Ofk>v1+nw2j&XHs zWN=x^Ih@;>a`xC2Z4rMvQt%XJ(b@9`@Y2wru3YE6n=Kact~>-h8EYTT~5u z`fvv(jpg+({lbO?PvWxH#c*TZn)02urt-$Nj$HjK;e4J^06H+zgN+_Cn%V0;fwN4v z;VKm`W~(n~%Dj2>1;3v^hb`S+%&vLo&Kd@tW=E7TT+oL^y!e?PqhIQX`?qd}lcrzc zo*1v^JHGG9Kk{&8>$Wi=-ePyzR%_>RlNv4Nc2!N`;;irT!7T=JCmYOU-3?dr+nUzo zeBRq|=9?a|cTQd7#C7_d&%l0cqocN*{lXRU!|{8GZJw><7R^RQ#YXgwj_5m}_t=Q= z(ZfcD$VYTNCQgi#cTD>A9Tm|#CIUXKFerQwUld;z4wALkF+~?w9p$>|n0jrX5a%49rJf+Xr-aql z1?-t4K7DpmY-8S>dg-1ZjPKfn4ry0KN=)w}KI#2Y*=Js?n2R%&=c-oZ>?i^I}}6$z5hh&GZt*!S<>2kcS;;+v=;N z$1c;vkQAQMKS$F=r>&)F9=mDk{ZOf3`XlLvVH?ry!X6>j{RY*y#V5(()O2dc+nvdKy!03Yv?>PGhMaZ5N3B^)6AX-b@qT zCy!7cF;1h;)LTh)-%w3#etQ%Z_TZ+F7MLk1(jE%WA3RkojeRMk)f%kK3#dyEVO~=! zLuyd3hqjSwEbdI3c0VY&pKK$pe_caLzLhGa9GxLf>zGEZ>$*y~z#db+*zG9o&DJDwen2vPI6Z@36ZSRU-w!fvD z)U2r#7=JpaUuWU?rbSez{R`xWpvvE3+rKch$X5--HWPX0Z?WxfvF#t1L}XunJbpuL zi@l9-y%+ZA+JebA{dz9$qlm_>$D3n*fPfE}X5mv4KcZ8gvav(=nK*4k5!UP636%r} z;sIt4@kPT4ND#;4VeCgdsDCnE*sdcU`$@!u$_r3otzOu+N@u+0;zjJ+t0OA2-G>dk zvuK<8DDK|Y2Dhnt6|ZP@7w_*g8>30{P|1`Fc<$C^cxLSqTz#s5lk5BA)nV?~*-Zyq zy^6<9FGiuv(nM6RiGX%H_Cz_gV^Fv9acJZ0V!XVh5$-*@2R5Cw5glHXgY8!2qfaiK zkw7`1ds~iT>o;ZSka+~!x27-d(d95MPFjYWj%e^N{L%d-v&v7$-(uUp?%f3s5Ae6x z_W#$de~WFuVuJsQNwVSiFT^$?Mrl4@t`W+<32q>6$zqu7-33CKMnof?Wgxi80-8n? zBl%?!?z`6(5%InWa70KW0-DBSi-=$Hx3@s#gR(!?`8_{~Z$ub_I%RQ9Q$G>#K%mp) zBXw!SKG+A(Fc8v+7zg#s>EB+Q{8pEyJ*}skMxY~t+uu|BZzJY!i)|oAfgmRHt;{#U z4FoA!!~%gz7OWtTEP{a`_G6IqfQToHT5?_xzd*19;5d|(^OMc*`Jr7%hX8q@90XaQ zgL)wi`V4vHdSo$9PKP$+x?mgn0YKeQ$4?z(1M=q;+o~S)^_#z4waz%tw@$4lD(ba6 ze{Zt4Dm0i?-A}Vp)on7IFIUv%$mo3(7fTryzb`t^1ta(hsWU4m79dJB;g9{|X;a zJc3Uy8OmSx6I2lq@myAq6TEYYk~1FW%V#dmVs8{(=XNKy;DzCNtoKoCuA%2=#(LCi z?wR#`ZrjQM%=)iw@w1c7nGA=??An>Vn7t;w*}FBS@V$ad`9@c-u(#6huw0Y6TttMQ zs?o70?B=TJY>P_q%=;akIppuk<$U?TJ)K#?JvsM)uT}Ddc{_*YFON-NHoc#t>Tf)c z8**|ED_*qZ>>I4-E3XM=Ypxm29TV^W#DOy?8l*h8(3aJDT1_*VKWIRn$hyk6%w z>}ubKOj&pd)A&?lw$A8auEW_;Ovd6IF3o1T%KUIwyruRrcI_tz-1Xip)^W5iQ#2zH zU(W5u85ZiYy_7q?KfAxz$J+mGVq2Vgk>^*DsMJrdOnbGkH?gX_tgc zug=n@TegZPsZCM~i;v>i=uyZ`DiTuiCklZ=XXVX(orOjABb9{%FG_Pp-=p)oEZk!o z<1ZH8nJvC=v0gN+Jb-Ssxz3)t57pA3-ihMa>5oL$?M}+edl(@%(}FHYcu%$Cx9%}N z(OdeQanjqoldja}Ymrcxe_gEIY&%LR&68dyeki;@aGqFWf3To?B3PLm){j1SqfAV? z6Q=xn)k5fU?=)q*pj^29E?jDVXSo<1)?e~_8%Eb2)j)h#GmvV3qODkdq@Eb`?y17r zd#(6p%OHC2nC;@B=)qE)^9JftXo%^y6or#R+CU5@YpN=~h2LadwfpXmH6zc+_iD;m&!k>RtJTKKDis5T*sZ zP!3L{B!{@Av~Pb-cwRQ1;=6^brFusx!@Hu;Xui2}L46gIt=(QA%KA$26~t@fZGVTihz3vxP^c#2)t_q?-0o*RY)dJirz^wvYDd7Dcc&A4}fcJI~tSm^(5IBO0(1zeo2qXj% z+7mhwLI|A*6?(F1>xv+GA_-B1!Gs}%VT2Kc{~)kI{+}txk3)q*S{$~`4nbZz+p+o6 zAT+d9p9aT#+(_6R%Yj8{`FWiuR|$%qWFx5+;Boe zN9xdn-pxm^To^P{YKxSsbx>KH8M;xa2%mN*!rSY8Sig2TUU~I33Y^*!cL`XAeWTvs zpbT?VZ-8bhOr8sb)CpZu2JM_#&lKQ-LikN#EDk)-KnT-(tqWd>1c3=& zTO{6G8V@nz5hi{rKyK_zG?E1vYJ;T!848Y>ANQz*#@yo<7Eb41S9!1({T85mn`yuPzL=W z`%q3JvXOeA45a_8)9{LgFfb5c5Fmizhrt122rw{!!H|K0L^dLj2Mi0OK|Ux4z`l%J z4$|ei$fl+~h@lR-E!c;;<+2dVeSxxIz@R?ZhIRqihQSGr!8t&F2rxjwHUQ~zT~G#& zK|N3g(toNT8;A;oFL)OIs33Q(A`i1T-_#u&_>6sBeJ4jHbCKP8^3yIg<5v{A_>NE% zGsnILazR%{@e{^2XP546%vatM#<$pB?prw`PIYbLe5O`RCnj(l!D^6E2HhpZT$Mbnt|cGr2A zanpE*=i~XyVJWJZj<@mAj!)3N_;|E^r!7-+4~0%on1!RK7BGwTNgL%)*&4mBurD7s z#K!bPbnoynys*kVJlw1`-sEM>CcGWYMz21{Ts<|7x4Rk1Sl&IAOmczAe zZNWZYK)&%iyaY|k{K_O1O~E-;F7amWtJ!zwUUL^M5A%WM+f=;`Z>!vztmB4G%;pVN z+v1oz1x!lAOulx#N$iHx9^4^xpPSgyhR^;qj}xa{RasAQ;|o^yl^=%R&7;gqZt;KW zQ5L7}+h(G0vul92Q_L!)(9u8NOw3bAt@)n}u^F@UV z75-EmYd7`uC;g?V_Z4)PR;6N#%bv)w^){+tRuy5%!qyUU@)J(1*($EMS5)}kNKog5 zkk=C522e#yd=$~82I7=WQA*csabga$P|EPgp+8odA}xpvqn3~9Ax%BL81+BYPz*Vd zCe3T%LD%{`fJ$BeR`jlpXy0z{r4`lF#ADAAy%#bK#ZMobimj5*iG!0%sCqVWg6*2u z6q9y|Qme;FJ+t3Y(KbPJtKrsSt>+!6)^r`o&@D)~wmgIC-R+>d*1ZAZBx zy4hT6(ReTQKG#{|iNF}Kap4^G?C5EV^gFF7#~wOriz>7yerWz&9%TO)tUQQZ)#c0ge^}%qqaj1!2r5Dw zSWS@FUmFLK?I1#XLPx@XDRhxM5rjxW6k#x72w@mu1mSOyi(G_?MJ`l4_#tlA{v$dX zuY-GB^+MTp9_X~iS=`e9EcQND4|hyoh%GN2#Lg)hxMd4H^tyU8yn4byJTZ(znL1Y3 z*!mN8ZI^&PZJ3So+*Rnq`dC+Oj|MUn;rl4hR>N+3>Y%4og2;F!h}i{#a`0pA`Y zURxlrf%v2m#E57G^};k9%EL4q#4e2&8hC1H+k;QrYw9O$!?xyA_aK0Qcr`&aMCUtlTO2bTjU~tQ{P1)5P?Aa0l^7GA`pW>9D)G*5I~rM zJRlZ<$OOkB4a&kkhwdJWxNBf#dQv>_hn< z`vc{m4736HFRM-8h za#fGMW)1otbl<=OLXs+OW zy7c3=ZkoaC7TfbWb=vao_8(OFDf(RLx@UZ#Wg@ry z-G07WLKYv;{*uZ$xQN+3b^!hy?7$55jbK-24aYGifjHf08}7V40EwMnGkvEpTk&aj;1T;NO?Ho}mW_tUor?+S? z)CxJGoSccIGN(*>#_W+&&h?$5MfpHM|CJ;zR+rEwWk%9Lh3y`@oTg&Z!!bgbT1E9= zWhw4R{o>Q@R-7=y^QH3a$d%%u+r@jZSDDmxwLenWZk8tZTrEB^z9GKI@fZ5p+!VIG z((^G6D^fO3HaN3Tg6Qdj}@Ie zr_nmzf;dt7Ah{X~%7d}xlwHY4(V%r7$!uq#V%2LmDsA9E<^20Ll5yH-di32gk-w8J zO>1^cyf*01ZM0T(r z)qAroRbNa~3@cO!do%l}XN{X9=54$t3>LA78uV4H?08%eQ1cDtx!TSr@APG56#nJ= zWBI4@+=}^w#y9zDfuLn7?^G;kK{sJ=hRFjA(DHx{19dfmDWL|zf&gw2wF%(ya3@d@z-<5#7?@O(nA658vfYN@PY5Jb zC}_!(agryD(38-c(3cQSh#*7~q6mWt|4G=A%l}nl`Sw>RV8x)BJ5$ljM*1k&vNbwf zyCq6*UKbs`(HG@s^~Kf)$D*C}+MqRe6j)Jd2yUd?9&PLM0lS=Ahdy^-k1lU?$0`>S z^oXm1x-3dS)i?LUm%6V-`IDdFXR}|Tnw2bZ6`$w0aaKHbFPn~ghp6!QIf-~wXl?XB z#}c=kR}VQh00$2pyskDlN$8*__VL&%ayDL}%0@SDx}vj++IaE&DR^e&KHR=I0qsAi zpfmbmT-EG7u~RZIeT=6^VQFaZF}-2pHM0IWj*Fbx3A>j5wmz>o8P znD4`y008p=m}sd+VxYPfFe`wW7BDdYc>VxjZU9RyV0wUBTEGke@bCa&k^t}&0>C^0 z;57t*sRF=*2&k{cj>Le2mWCt-8fj@vV!%;L6A}Yp;s7Tt&Ljq0v@|0z;Hm{oAmFJ5 z%pvec2~;tpZMe^HZ{dEzeT4f4_Y3Y3+!weHaQ)$$!!?HM3fB^@pB(|N8C)y4K5$Lo zTEO|kIlB|!T;cp$5?T{{2ucD?zyub|E{S<<>`S)&2myq)gm#1ugkVA_p);W?p*!S< zK(>1jdJ*~%`VsmQ1`q}k1`(nOLkYtPBMGAkW51RD>;4ND55jK>QZX72DfoCjyy`~; zB#o#;ystp?(ReZu!Ac|g5K#w2CXGiH$*1v>BI1sw9J~?;o>c%5hBRp)RDl2mACo5{ zmFAU4t>^>-7yu%XMra~aW_XPdrpy}eC?YI@hm~Abe*I8W2WdmDOP(@o#3#}}%}4mj zxybv_M~w&tF{vN=4kDN)jr8qjnKD-_K!E`Oga4049E?7sK@0(m0_@8H3g zmiHkI3=06p1?mCdIJ774!(ahyxTgXLD2@y7PQ3 zKfPtL7uayGqgJSveYRuY-L>GBc->Oft6NhQ>a<<8=k{!7@}2#7@qxZfJe|*##AK=F zHD1T9Y;%ph9eJ2>KHC{TTQ`_nFsY-;*s_4j_PoQ1y*sLu@kjW)^@XYh?^olShbu9) zdQU~Uu{Uv-6QOwQmM#pQ-<>^bFoxUp-k85VC0ONIbp;pgOg^lCID`o(RI`W7M&P?= zH{z%aU7Q>2iYu*++jo(pWMhr&o>-LdrPll53hK}QWeV{=pBPC zZ|z~-b~D_Xue$uv%!ZuntFe6fjBb2jojsi6vkPp>#9>TCV+%I1!94s+<~F?iR59~) z>jZXMgR?lJ%2xEb{u4CY5n+e9eOc34Yq^eTyIBi|r%d4WA=q)}5$lJj^K#xCf+h~3b?lv|@B!e2}q?txnYKj-l?cFm&s zYcK@hI2l z$+<*!6%!3!C7Uwb~bLzC>@fS}qqqM8S(ejMaCeKQE#xcrsi&qL; zd%G)A+TWoZFOQL&m+TVgSto?kdLGKuq;9l$%TLnXR3x3+(hXJf8Guwy9cj0w
J zbfqJk??|(ne-*v)9ranJdq_ikV5f z3SV~zQ%S?x)5{Kic>S&wO%Hb{F(d|BrlJrk)5=MHKN?jj#(}uN=Qq$Zv(WBBV>FRIJ(i>`T zr`H~Ntb976xftncrK~e+hR>84NNV(Qt{8amgjBPdITh<{Bv!rQCq4&S z3Zk0drD{HjSJcn36z#JvQ2ULXm9Z0hDZAG0L8o4##q|q1tJj-17rb-MDwc(pQx;yW zh5SQZq?9IE!h)3A>PL+pDA&!MEI&MdE>Eh7e}q~8-llxbKzcHhcPf^iptJIn0ET!N z!eRJ_;TqP8zl-!l+TjQ)LK}iVA&?M6Xiw-!2qAPLbRl#jR46rR*CbIS&tSq3!Z5-J z!YINR0t~(532}ej)yHq2nfzE(C`nC3{(7I#i5*5LrNJC*l(7#p-BQri>dR5%OjF$2 z@hiTe7{wUct;Hc7&!FcQx})Pb5?#Fe4rThCMThujs1w%5C+tgc@u#iW!Y~adHX4b| zR7I%3HW@vy9FL9?S?j_ybDZKb8Y?Uspbdq6(8_RcY;o}%>bVqSJpMG+38{w4;&X6^ z+5&d+%EMOtYaEufnqJ%yC;1Y|b&`IdRgOfNo zd;wBNZ$RtgFOs{hgR&Ekqmq(+a$6H|cOPuPu4$3BL< zKVy)eieHneCMJ+4>W7O0CPSXZ!%SXYQNW4;08;|M%pZ_f2C&)yz{CKs8UVoD047?% z^Z+1(0APjykVJqQTFgldSZJ{%G4Q`FRoRm^00%7%NDMU8(ul-BV=ay(2AXJTN@BoC zi!+G<7cI?547h4>BQfBv1!OG1QwzvgKnpD(_W-T5fb;`+YXKPuP-p>32%xlpJoL}J zNC?9R+<&;|aG&Ac!u^E%2=@){72F@VC$$Jxgnzq)Me5`Uz63u)00Exmc7zUuU_vON zGeIL=k+?$13JfPXK8O%a7)lsU7)cmS7)uyOh$T#bygy@riwog5rLBpw{G<62705$G z;`-*Nw5bgIus~jdeUPF+8j@utke)PB7LmEYp9%6n3^JJJYgc3+WG=`r`!<1ert!hj z%21Fl%Tb#8z(7@_+&{KfYxdU%T#b?lpH!HCY+O@0{9>&&a#TpJU&uthzks_xGyI z)u&C^3wynJhh2mC{nId8X}1S^Gc*q0-{j4@`!~hc&Xlos-b=W+HHK_*s1G}LsSSJJ z)p*Xr)S3?+)sp*sFPS|O6vZ8E7QyCsyQrG%d==}()Z`=gPvQ0bt$F>S7kTdGMW$8l z4g8t&2ULdbj&dpf=IkwFiQh3KlC$(3!nlq9j4$c!Vh?wCj#nBR^Y#$nwG%lUb)Em*6PnL{E&d-*T=AzEGKir>x|-; z4Eiox{a!+uXAAi;{GSSC356H->Iu{RU3{9}&JhdFtAv)f{MACdTJfT&j$}?_>gqEe zrNdfhu~PRYf`@O4a%28MVf~io%6`u!sm0~7w85G6^pg9%B;V+4O7D(Rv^_XPs?tAQ zSfA@Fwy(O9eAjA{IJUGh)yHw2^62iDV*4(w)ZQo6$^+zkQ>E+oNLi^@(z!X~q|@#4sl$DTDyuxS61Obzpf)xS5$89y*<%*DUfjHIkov*VOX7lz z+ti4_V5J`wE^fWwf^MPq65rRqqujaHr!b9MwMU1Nuf@5jYamS=m>eB=l*Y(uH!F?c_9W%V?w-o$9hb=Ylch08;F zia|TNp=FS$dSWF_zgI5Wj&3bZ?LC`n7O6)yJbhoi*gRcKDCLEY#&^U*kH^#ii)mtG zD?_SpN{Fz*w4oHfW&riXDpIVub&`6Uej4?_sFUFMMiARQ3|6nI>r2f%wxV$Q?5|2o z@0DV$Ic0^pwpS>dK8bXiiHEqtiWah~@C9S4+@KV@T2ZIcrzlRCf$U% zGhO6|=+7mXCH!&5U(>RD)j%?rRwNY=4qU? zdkqpjkE2khJhX4)GF&--D_XR;2YU9YGg|+81!~#M1TBo}g{t(JhjJEl!fvC|(Wy#m zl%4SgPe>kvj+tFXi%M3L`dx6JI~(wb*JilU;SiKPZ9A^s_B{5vT7Y%SkK%@$K2BR_ zhX*G+W47rET&Kli?D1hG`lZYXjnyBH?ijdY>y4kjU7MfxU)gy4nuPZEFGYSceRkyW z0KfzS4qCt*0u8l*X#{?639UJ46ZpNQG#FN3g$OIgzl4-V&f}kw(MZ0(e<_k&1PH$= zsU?8a1TvP!J4=3jkw{P=If0a=k;sTF1|DDFK?Zpt)_9Y_Ylz@=MIM9nrxbVMSP%WLE}BGG+&CGy*^PikZ|QeN}RmZXoe zlqP$Bfm{Z%pC%vF4PI*6kJ*23leD4vwih|qzo+)!o{);AH83C$DiibxFdx+=fD|Pg z7Lc&uH~{ljIfiXGCQDSZdy|}rK(|ypKH9Q&5PkrOePtOrlrw3f-SN3SY@7!Kjwbx?> z`=~`K(?9DHleO>`6Kv+gx9(WV>#X9& zl*Y0}#Vjis8#C>rM{rFGQ~02VmHA2WZMeYM`l@*kD4upX!7k|7f?MEjz`KmR$@!an z;GR%N*=H-}a0BR6-agBQO)RRVG95BOWqm!8&u*N<+9y9_oiBZ0Zp_<@=R4%$?WyZ= zQm=)$^?^_5RM}NrbwD&WPMXSadq?1uu!h*4E)jQPt4jL*TPT)pb`c|XMzW=H#X zypKsdH}AF`S1s9?9oOKMs_xVpDkFX%uOC~&^*c11O?|4yZS$PjwdQtQr_NvaDwlV2 z1_RUBC5mwN)6E&&h0PsUo48Y~RsX8&>Os@lL~jGGv4b)D_RavlvT-SEH?JPE_ogk^ z$iJFO-|8v9Y5yC3+S_Y9?drmt*KaI89KV~`_V4?6ec{HE6N<{iuwZ}pqk8O!-pUD< zUxf5C>y<-mHV~FrpZAIBx0Z5Q8(O$`Qkc+Tk++z#I#ck+F;x3kuj#CfSA=c0>5@h7 z_X?X^OQ@{dIm#vPCkv-d##72G2ZhL5t?A@9FX*#}rid9~+k_`>QFMRT!PJpc_mm0V z%cNEvk0G;DO~npDDL#oO+Ddr^q13h1w~9+`y9-^nVe0MYuF97yzEG8xFO}#@!F1ze z>B5Im$0dICC~9K$+QRi59mU=unNn&-6Z+x)`P8w(6{vCpJ83~>9np6cB}Glo7Mk{2 zMBl&ifcm)ekoxI67dmM8Kx)8};o{Um{iX0RXOdWgHt%jn>_&qcixF9fFhEFqw1 znX=B6Ws?48LuycSXUUqSsVfdMl$piFK8t6Eiq2K<7oPEZrt}I3pt^Q(6Q?z^R&F1> zg|c0$P$+!gD4sJNrO+8q=vfDbD;_`67m{jCF5GsZrEv1_S!w+2+M@9y2jwpAvM{`D zUv-t`n-$4UdBWl)tLfsj7u4#KNYOoXG`&h^h}fK|BRLp9SL}9m5%t5Ti(5{8q&m&X zqwJiE1wU*i&O5$_9(Cz}Q27>waXNxRat{6T?g!2vJ_34eh@di^&0KkI1h(8l+btXS)tjRSL23OO0;{46GmOe zp`azZaBRc6cx6XD-0Iw9?4EZP&9C(syRT`37Ce24P7hv#ZcSDqJ6#ua@ZLxCi^~_h z)OsTde_RjkkaEyIyAHT%oxSKyxQL6d=HcYr=4eV}dAjKkHX+n9Sk1@it(L4F#3 zO*o57Ab%S_TnsP~0IU%JSPuXooB&`t09aedYYlk~^5Z%L))fG(9RL$8U}^yHUjjhS z2f&vJ0Bs)tza{{5e!x--X#N2BI{~$|fbSCk{}%v&9~4kmiw%hZ_&)*wUhxC!Yq29S zV6Vl2#6SZrzn5_4N!kOvwEW&e84MF}-~ZG?8JK%=4*dxOwA%y8_8>wup+dhi$on$| z*)aU3Fc$YstkQn73tl%Q-`0|S!$9x?;MWp0!kN}j3j`<-x-?=F5u(2Pc7g8~*_Zvt zWIr|dt`-PzkPfeHLOKXw8ox4<2I3b8X|kZD$qV^Gv?HRLMr70anQ6o@Z9g3_@QHx_$9*+vljPbndci0sIspeBp*6BF48!-JeOu^$_2f0uVNEMbN> zFJ#TbTe6=%Msd2;vbo6F=h;VT)v?k1R(SrwFjN#1z^KkF;_5}swh^$m7dz3&5CHmuRrI^RZa-wuz4hFl4{NcysOMLYPFpA za<9xb=-|L4wA+E#70zJ|{I>J{@q2jh4)58fZaMgn(GE7yY6kDxCyVdbSmI;#?(oyS zQ`wI#yP$mcP=@k|XH0aKqTQJ+Yni@@Sxr7h-@Nt^&hK7lw%aEKqqAxi`(Z;P-po7` zXMF0z*zTHv3Xb(>O%`5b+xVBVOF~AnA8y>o>n`TAjSe5;Xq$CJ4C}zQ6XWEE;dc|c z{%zm&N+_IuJ4u+)VWYQ~Yq~hOcd($>)*5^ zdxgg#A%$mNURE4%%n~!_-lP&g-j?1+RTXouoRJn&Zd7io%LTdyeMQrE3k7HUN|cHF zT5&=29mVR_m6UTzOoU4Liv;_G5~by^`I7Oh!F0`5o$2F!mI^ruH$>5J@19yOo=C?h zIxE*&XYZ-=fTQE;-4xlk#dO=0XraaWOzOzBlj6?a2b57;Z3^>R9hV$yhe&me{VC7M z`y|WR)kX7L2dUm44vFngT^EmcjaTleHd3g5-k8qI%Ty*k+%C?r>ZM4U@PMk+CrsSv zpjUY7{d?N$=v=W~j0L^0&_vl?Xh!S2l7x+=;Sw{k4qaJqn3QO*Cms%dC54U{K;`N` zrflp2q{`L#U*N` z@Rm}FM+)i-YmM}%$nMV-hIj~-Y_?Y@$vmVS%Iu&Gj?55!9gZmsJ^CpRWQ~xCHzq) zU)!#H)j;^NkasE;zM#7>L4b(^4C37!PopP``xZld#f}oI@H*$4CY3`wD*WIo3*|?ABa_>O2^sFZ} zt@CQSUn5(zrf4lT?lB#i?|qG(jc231Uc<1v>KvM1Z!>mVeH}Haeh&$a>!Pn7199jz zOB^@f9qXn75>dRY<^yibO=cf+x)A4J<*xw#y ze|wbaR38}?IdF9E@PSd${YOR&4Qm+PKz?RmSozZnW#k_}in5 zTm%TeDV!z9AEnoPPmBmpninA9*Axlhk)=tOJ-uL`2xcI70U)yd=s5<06r_VtCW}-W zu}o8s_N7nq5~s#%41_T7A|qlQpm}i;>LkLO#zPFsXuZO|_lfmjFWkVp1pgL-6f zP2*9f73M&Y(+Gj^5+?{}&>rcV+{WKi`)^N3#X=ew5D>4xh`|4($U+kcRRGLa<#fm+ z&tD-1U``8jS(wwxKsuz!b-}irpKN|_4~|0_s2_j;`5?%1WjU7n0A&D32O!Ax$ZbNJ zETF+TKt7mf1JEAigCLjxw)YeMD~}=-2y5_c{E@hCfx!LRn-lq%Zbt*_tYP_BO247SPE_U!Dg=lKb?$5fS+kN9&7tyMe9+pDe_bZ7UJ&tnsQ zF=E=z%VnO{J&di?dw$q!oKu)2~-p$B#|%{bB2vX4iA^gdUd}osyb( z?)W^GdXj(&KlDL2>=xm%`DJ)l@ijcZ%N11qQpu$0yu+h6A4R)@n&QegC!*w$W+>%g zAydzKJd-o-6cgp0!lXwBvv1BQxMvIQvpuKoU~J<*vPlbDvWN1vb5%1_*rtnjGX8$o z_?g{DvL1y`na!P&nS~=G81>*MoY$L;yrXG%j%_fCRgB)xU0!sSjaa{mD|TAQK3Lp| zTQOu3SIsm4AJ+3^N3GQ5r!HN>Et{-VO|f15|yn8FCJvp%=1W~tvVwoofB=qvpVKdPe( zPKYKCKML~#4;8p2J1N$?I8cs#8wwfm_mykwnA3U9JBcn9tE6-*M!+ZXseHdrh2iHG ziEj4imHw6D3NupXQRfspwjavk(-@bjL zPCNb!UH^<7UFH4`Av$uMxJi{QEa@{^TJ}T|;@i@~&eiv*PUkw)Ij2>U|IQ`Ug#oXm zROj)OHWJ|7CM=_1s=H#e; zb?znAPw%78p6n08pvOa$#pdzilVd^R%%D(ZS@v2 z?i;rYPYVJSJ9`<>gR`QAV@)P2H>WG8tXGtBtIY%IO71wR>gV@LJ@QRA^UoiIHT$~= zGnR--N6(3L`}_IgjVm35vM(2vOF!r-u55cE)y{rFIZfT5_;`AwIOj@N8ZF)_R`O^= zUH!CFylz&NvN`5R&?_EoNA-h`A%jSk-u&bKIXZB>^+Ij+4I&Fo=btEr7 z9?ZsWrYdxMizzAyJ&OFSU*PeGk_O^qyGn57Sw+}y(pB8`f+3#WIUDl_@8eFifO7X;!q2vD z#a?wfptIrd^y(nXv5nAKbrGIYeIU9w$pg~<%0a&Czb18!Ng&Vr4<`?Q@&K$lxCnb4KcozR2Oi_nMAkI&2t!*MzcP@uNFRTeiEzb| z6a2x0AqT)98);)O_CJP7WCxBx41jTm0EQpRfpM4fKv_5r$lFjB3<~6dARhzc0%;IK zUg!q|s1Ja2xnHmk?SXNE@MGIhKkP%0%Rw4!L-?^Dkp7nr@*jK!Qqmv0(5Og28sj^2 zMqABTbj(wS+JZjzJ8crkrBuqyEclA92bk5>^#KhSRGgGEvd?N zTlSRmc{_}sadH8FDDesoYu<}{c%(XeYCs|vY}A7b-+7#CHDwxKyZ#iu=&BVPGUYYL zOo`$0jui2WJ47?ORbo)NK`~>oA(U-hcAiUIu$xOFTEXO*Nqwt~NL~M1^o?p=V z6uWrWK4!s~O>9=V5wpu>p0f`<570Un|JGJVfDm)mrlOv!LkA9@5*lLlJV6zBJ~r6os0$*IdGD%MS3kv+kZRxR2k)-hPB?i-XRj(A^6n|9qU z^||o>+B+AxnzFu+OU0-ZDHVxGZlMgN&e?mdeK|+DBoYz2sU#_ePwvL0hFmi)<(4}S#x3bx|1)PBGw+z0=Xu_b=N)E!KHt{f>%P|B{r6hG&)WM$ zP#q&5s5aOAfvR6(D10^2NopuFR*gD(nwm)sQ_ep#MkuYmC*13=q%vAA7RrsnXalG5 zLjU5g=&aE#CBpYx>76%E2+y9ZrPjSDp&W|((;g9Rlt`~k9ldwCFmAh2oy1L1_uM~- zT2wGy$l7z4%Dr4mx2Bp3rB5~qDQ8bA+s#`;RdeP-A?B3ZswYdGH*sq7uC>&{X)BfO z7VZ)zZ5${$8t_bcbx^tBWVnQCVp&BMWFqD1?@TD;%w586yMxl(P4}v)P5lL<4Jt|T zpY#o8QU;^*CRRAQ z*F97@vlVXAbrf=qZHAVQxs84-G{vSmk8z25m!^?lz#sAI|CVI*SNU~CFgXCo7XX+a z;6Hs`5KI=(S__yj0HhEAOc?;O2+&STd*TCN-hhr;z|;W_TEOf9Afo_a0s)Xz05FFD z$SVMtMgXK10L&!dtOZOa&{GSTPvGB`j^asPSjoZa3|4xu%7WFK5y6(7*<1ijD2GD2dC-e>a1O0%uzn6}*=`jgo2|Phga3}O8z`g59@FEN(3?d9B_!9n< zbVSmPBSaGvgjm97|Er3k>4(7s_q!5OJjgB@iG|2LAjg1=1CowLTGH$;^0C%PJVXN0 z>|7$dk~HaIrx3_Ou=9!RL;`6?vm*+o%tTrezk8*TcZhTZa*igyR!Y*ugPcRA%$m9& zjdoX(raq9VK=J{ZibzG8G+NmRUQ+}KOCv9lFijk^1JXhna7@BAJF7^!;A(#DQUfc_gHVW3g<= z1Lpw5frI>T4rPE3^@(|*EJFgM5sx7r9E3xd*e*DRGN4>IhBBZ$@S)9O969{_ooE0) zA}NX2{!GEO%92P4LMo(H2kk;}$DD&bt6)cBI+P&7Ah4Vyf766bDk zo-ZA@RpvEq0=Ir%DQ7cRDGL%(WednIp}KAv{7zhh{pSVYV}mQ18M7q#mem$!K;!)^ z-ZY$j9QvG{{&o&t;CTz1EU01~<P zAIR>18p~GP64;($db5__hOU@RXzO z>@4ia-CR71U7D@rmuL3n_pjZ=*_?@Cv({f>Jxm+0Ze>@Qjh7g9S5#-VizSQgp1$T5 zyJhp8laBM-_YB|{Z{N)-mOteq5@yO;^`F6~z6xW5$_Dd=$KtuL1C6+%7@Du^{F1wF zznkmP_c*ukySw7W|5Uq<{_lQWF|o+?+&tB?%z=CB8cBuDO{P-v#gkO+&$M>yM42g9 z4^0xDhWXO%9O?@TnuQ3KFA9}gJGYcHy(3i?uR(OvtmlI9L{w}RI!MS~7B0Q>I+cFt zlCHLhHAK<@-zw+%ey?u*Vi)E7xU;IslyyaiPQ^;DOJ1syqZHB=X-Cwj8#~eqTg?^* zKmD5SVBVfu#CjEXx7bA)WW7;$xObcO=C7!850xO-!7f6Co3%RW@J*`qtpa-PIS=Ka z+FR;vTOLT6z&GmB{=)>HsswdWo5s}IJReE)xSc}%$mJ5(zWY>OFJhIZISZx91sQ_X z$R$$UF{_mA94=BxgVTkEb8o8KTUn_%<5Ifb@i^&~m>*R~rr4_;uoZO>uMn=E+OMA3 zd?&55Z>Z|Q5LWHHznLoYNfrII<1p1f9`}%*f0Clc5pSuZ@BT@vSc(o_6et{Ax>q=! z|CZX3FwtPyORY*Bm1R?zK-HCELWq)C&WY*qK@=PYp5oOIyP6T-ewPF0gwsdnFf zM%A$F2IXCDT6FSpKWXWVOv!|rN>wf1p{!iPOWN1HS@d}1MnUhvGs!CF*~+|=>C)c% z^_4vupHd}6-jd3{_ExPZ&y(s64-@R1S4yUZ^p}P{DW}RK`n|tB|574T;-@y@@9PuC z3uLCI;>pKlCTJ);hQ&uYJeFZv0FUv81QSAILK6b4_bdojgpbNhB(4Ji)?Bd8g0&T_ zr(g{Q>n0ZhtdBGj6Y+l`FKO2vy-2tRp%0-Sp+8{&!JFVi7$VN6$?;Iaza=%1eDPvA zpSv&K1|OB15@~LY9QoU>rj7bVp)3O*^wxVd-64Jt?U+9RmGo6m3rFhE@gjz~1)*vHbp4w087i$h zfL9fmT9k$5t5@U5 zZHd?_{Gn3+xE`Grx?O zNYcUehHDJh6|N;*Ke%pio#6Vw^?>UD;|^mDV+>W`AI7Wfz0&N3wBzmikw3ljT}Vk(8y3ANrAj2?vNsrYmk;;rxO5@ z5~SD2QBWuOd_1HTC)Z*e)GzLi(nwuI!h$x6yRI}}*MfGDwn92hc_4FrVUz2R%TZtu zz=06}M8jTJd;Z~r0*nO!4+3~tfP?f92Ir6#9vW~CK9m7zMMDGUVtJ4c&LIwfV*ngH zNFW~oh6?hCbwE90I&%1nI7km^Aq^ahY2g_924w(nF8bu~^LL^__=qee-UFY0AjF1! zOqfcPr)Bx@2GTTnGC!Ux+1rO}I5msU9{o=Kjo$$-bx1Z>@8Bpl(_tl>T9(1l_U8Pw zbrQMz)VH!eQI_o2zEUo%yfxFFYRbNv+LrZrQ_ke84zfw^-{HLFK3vndMy$;tGj?uu z7C(NYuIz{8MzWm=RWjpzUp~fAhcTG8f^D?>9Osk2fteqC39XJk!kBe=#|DpxVrQR< zWb0>|Ff-3(F+ryG_)+OW)^6NJW{>O?o_1^pTGo3y7G76yjb<@ScdJn5#y4?z*>aYh z!IbefQG2*tMV4%3`h5P`t!kOuR12A-VmH6EaT-6PmmS|)Kb^}lxQ&PGiNzc3YZlYWn_{(CMRg;79 zUEdvy*}Kz>(RLTSNw|y(eX8(*L=U!cxe;EJe+Mrl-?Uyh;VjPRB1gqtMziMrUhMf{ z<+$)t9Ny(SgxS0z4qr>q#RFDNLnpfFGtW+2vqsCB^2U{^Y{K;$T<^y_IGrC8_)yyj z*|cM$_?wo|?Awd^eC{J>zSQJ3ze8Qb=iBe(XWZgAry$G?9h}U2^~_~c+VtZaRoZb& zy`q_T&&%RX@TmlnGW4c3o(BBdDkZ&6$w+h34Fs1Vwoe>)9bd;W+vrCm&Z-!+228Gmm&wX`^qO)r2!j;JWWGb~P ze7<@@Qkd%4@T1a-Zv&`(wVhPe=Spd%MXp2=k*`X+&{|>>RU{Pd>nUB{dAAUg(vY^I zU8PGF8LMtJeI*FTt*N{l_o#v7)jGSZXm!&1MM9U9W`g&DXVkEi7u1|od)3#~2~_4h zbJgB$ro#0_Wh#Hq7eb$7snTy;rRwJ8>4ICc8uhG+70Lx$i)q9BL&A)e_f>WGrU~kJ ziFDSReXbL;a+IY__DBvbTPWF7U{F+Aubpb%?RC;tltNXu;H;2eryDiP&03f{q5(D8 zem@;~*@K#QXp%JBSuQ*&TqwPC-$G&>QzGbQ&yjxLZ?W?6i(>I+`D^`p;eX$zgFs~= z4&MJ$feKavf*_K-< zF8M8;TvP|^IWNFE0h3S<7lh1Dd*e`(PVyX%l*wuwoNtc;g95P0-AQQsp#W6y;0pfgdI#Dr*@fK4o1lx^XX2C= z!*N{H9kjIlx8zFHK}$9)#5#90$MoB>JX90tfiGpGV(Ydo=tD{0K^ly51L$KGr_kdy z2hg*nW3b}&5gfs}ptGaO(AiK2Y;mJIHr(QcQl^r2UB$>{Xaef7uL_5d6MCrFmrz_B;{s1_UJ$jnD(a6bL&+L;_I<#3KmT>_sABAW&)cDv@*`25H1D zcvX+=W6}VxxRE#zg*1Dqw6Eo9Ue6<6nS#l+MuZ~56HKn*bvP&=ehW|(r(kjo!j#BQ zb^r8*C`}tlJ49g(_F`$i0wwPAB5_bZlm~mLG=iG8J&^88n_T~~KqU%GAl`_G0uP8e zqR0f|5rl(q1YsaFLD;VaCJ=-G5R4!W%7lE7R?GuFI4BF^zOs4`=Xmj6As)cfMlG z7KgI>4HTT4$u{o3Y6usS)tWtX!Bb`!RwHkG^#?xPZ30^nlgFDM9L!g2-irqw%Vw^G zU&TL0&Bk^Y>hQfhv$^?;>amL0Dcr^E1Z=Cj317^6!Cbqz3FqCZViyFqXDih`IG6rm zT&CGtuIb%}>M%2 z+1jLhwpFKBT!M`U>#NLX_fODerukX16LsBjvzDFN`L^enlc%2J&3c`D)us`Mp+V8Wa^36YC$Uh>G+NoDiZ24Es5Q&ihmi#>GYY2agSmiFw}~9Y?WYh7cJV z5*QozL1lHFx{2`y5FZ<;2n&h*Ik(gMel&;*8WR#68$KcAed~YOag(6%ae<0Nx3N)S z;W4q{!D6|_ijXk>u~9KGzv_FVnBc&%A^y=(G5+x(@V(LxYn}YQ-uk4sG2v0;{Mv{G N>W0@1*9q5~{6Chflf(c3 literal 0 HcmV?d00001 From 649dff16ba5ba89f0f39e2e24a8df14c98b6e4b3 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 12:42:54 -0700 Subject: [PATCH 78/90] Shifting use case stuff around take 1 --- .../validate_application_scorecard.ipynb | 3330 ++++++++--------- 1 file changed, 1651 insertions(+), 1679 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index e2790866c..ce2565c22 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -22,1132 +22,1014 @@ ] }, { - "cell_type": "markdown", - "id": "434bea70", - "metadata": {}, - "source": [ - "## 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 comparison and other validation tests, and then use the ValidMind Platform to submit compliance assessments of champion models via comprehensive validation reports. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model developers." - ] - }, - { - "cell_type": "markdown", - "id": "662d6027", - "metadata": {}, - "source": [ - "### 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": "5a1a2fa8", - "metadata": {}, - "source": [ - "### 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, create a free ValidMind account.\n", - "

\n", - "Signing up is FREE — Register with ValidMind
" - ] - }, - { - "cell_type": "markdown", - "id": "aa9b1d8e", - "metadata": {}, - "source": [ - "### Key concepts\n", - "\n", - "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", - "\n", - "**Validation report template**: Serves as a standardized framework for conducting and documenting model validation activities. It outlines the required sections, recommended analyses, and expected validation tests, ensuring consistency and completeness across validation reports. The template helps guide validators through a systematic review process while promoting comparability and traceability of validation outcomes.\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.\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/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." - ] - }, - { - "cell_type": "markdown", - "id": "057c1417", - "metadata": {}, - "source": [ - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "id": "ef1f5ce4", - "metadata": {}, - "source": [ - "### Register a sample model\n", - "\n", - "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", - "\n", - "For this series of notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\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 **Continue**. ([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:\n", - "\n", - " - Documentation template: `Credit Risk Scorecard`\n", - " - Use case: `Credit Risk — CECL`\n", - "\n", - " You can fill in other options according to your preference." - ] - }, - { - "cell_type": "markdown", - "id": "9743927c", - "metadata": {}, - "source": [ - "#### Assign validator credentials\n", - "\n", - "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", - "\n", - "1. Remove yourself as a developer: \n", - "\n", - " - Click on the **DEVELOPERS** tile.\n", - " - Click the **x** next to your name to remove yourself from that model's role.\n", - " - Click **Save** to apply your changes to that role.\n", - "\n", - "2. Add yourself as a validator: \n", - "\n", - " - Click on the **VALIDATORS** tile.\n", - " - Select your name from the drop-down menu.\n", - " - Click **Save** to apply your changes to that role." - ] - }, - { - "cell_type": "markdown", - "id": "37a94e48", - "metadata": {}, - "source": [ - "### 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, - "id": "cef99db1", - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind Library\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." - ] - }, - { - "cell_type": "markdown", - "id": "30803184", - "metadata": {}, - "source": [ - "#### Get your code snippet\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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": "d5de450a", - "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": [ - "## Getting to know ValidMind" - ] - }, - { - "cell_type": "markdown", - "id": "d22ef356", - "metadata": {}, - "source": [ - "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", - "\n", - "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "38592086", - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### View validation report in the ValidMind Platform\n", - "\n", - "Next, let's head to the ValidMind Platform to see the template in action:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Click on the **Validation Report** for your model and note:\n", - "\n", - " - [ ] The risk assessment compliance summary at the top of the report (screenshot below)\n", - " - [ ] How the structure of the validation report reflects the previewed template\n", - "\n", - " \"Screenshot\n", - "

" - ] - }, - { - "cell_type": "markdown", - "id": "943a45d6", - "metadata": {}, - "source": [ - "## Importing the champion model\n", - "\n", - "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "3f9e197e", - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b6887d38", - "metadata": {}, - "outputs": [], - "source": [ - "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Load the sample dataset\n", - "\n", - "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", - "\n", - "- We'll use this dataset to review steps that should have been conducted during the initial development and documentation of the model to ensure that the model was built correctly.\n", - "- By independently performing steps such as preprocessing and feature engineering, we can confirm whether the model was built using appropriate and properly processed data.\n", - "\n", - "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "40e73b7c", - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.datasets.credit_risk import lending_club\n", - "\n", - "df = lending_club.load_data(source=\"offline\")\n", - "df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Preprocess the dataset\n", - "\n", - "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", - "\n", - "- Filters the dataset to include only loans for debt consolidation or credit card purposes\n", - "- Removes loans classified under the riskier grades \"F\" and \"G\"\n", - "- Excludes uncommon home ownership types and standardizes employment length and loan terms into numerical formats\n", - "- Discards unnecessary fields and any entries with missing information to maintain a clean and robust dataset for modeling" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e436e862", - "metadata": {}, - "outputs": [], - "source": [ - "preprocess_df = lending_club.preprocess(df)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Apply feature engineering to the dataset\n", - "\n", - "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", - "\n", - "We'll apply the following transformations using the `ending_club.feature_engineering()` function to optimize the dataset for predictive modeling in our application scorecard:\n", - "\n", - "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", - "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "9c6b641c", - "metadata": {}, - "outputs": [], - "source": [ - "fe_df = lending_club.feature_engineering(preprocess_df)\n", - "fe_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Split the feature engineered dataset\n", - "\n", - "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", - "\n", - "Splitting our dataset into training and testing is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", - "\n", - "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", - "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Split the data\n", - "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", - "\n", - "x_train = train_df.drop(lending_club.target_column, axis=1)\n", - "y_train = train_df[lending_club.target_column]\n", - "\n", - "x_test = test_df.drop(lending_club.target_column, axis=1)\n", - "y_test = test_df[lending_club.target_column]\n", - "\n", - "# Now let's apply the order of features from the champion model construction\n", - "x_train = x_train[cols_when_model_builds]\n", - "x_test = x_test[cols_when_model_builds]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cols_use = ['annual_inc_woe',\n", - " 'verification_status_woe',\n", - " 'emp_length_woe',\n", - " 'installment_woe',\n", - " 'term_woe',\n", - " 'home_ownership_woe',\n", - " 'purpose_woe',\n", - " 'open_acc_woe',\n", - " 'total_acc_woe',\n", - " 'int_rate_woe',\n", - " 'sub_grade_woe',\n", - " 'grade_woe','loan_status']\n", - "\n", - "\n", - "train_df = train_df[cols_use]\n", - "test_df = test_df[cols_use]\n", - "test_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Investigate potential challenger models\n", - "\n", - "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", - "\n", - "Our selected options below offer decreased complexity in terms of implementation — such as lessened manual preprocessing — which can reduce the amount of risk for implementation. However, model risk is not calculated in isolation from a single factor, but rather in consideration with trade-offs in predictive performance, ease of interpretability, and overall alignment with business objectives." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Random forest classification model\n", - "\n", - "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", - "\n", - "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in their predictions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Random Forest Classification model\n", - "from sklearn.ensemble import RandomForestClassifier\n", - "\n", - "# Create the model instance with 50 decision trees\n", - "rf_model = RandomForestClassifier(\n", - " n_estimators=50,\n", - " random_state=42,\n", - ")\n", - "\n", - "# Train the model\n", - "rf_model.fit(x_train, y_train)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Logistic regression model\n", - "\n", - "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", - "\n", - "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients (numerical value that represents the influence of a particular input feature on the model's prediction), but they may struggle with capturing complex, non-linear relationships in the data." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Logistic Regression model\n", - "from sklearn.linear_model import LogisticRegression\n", - "\n", - "# Logistic Regression grid params\n", - "log_reg_params = {\n", - " \"penalty\": [\"l1\", \"l2\"],\n", - " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", - " \"solver\": [\"liblinear\"],\n", - "}\n", - "\n", - "# Grid search for Logistic Regression\n", - "from sklearn.model_selection import GridSearchCV\n", - "\n", - "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", - "grid_log_reg.fit(x_train, y_train)\n", - "\n", - "# Logistic Regression best estimator\n", - "log_reg = grid_log_reg.best_estimator_\n", - "log_reg" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Extract predicted probabilities\n", - "\n", - "With our challenger models trained, let's extract the predicted probabilities from our three models:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", - "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Random forest classification model\n", - "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", - "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", - "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Compute binary predictions\n", - "\n", - "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", - "\n", - "- If the probability is greater than `0.3`, the prediction becomes `1` (positive).\n", - "- Otherwise, it becomes `0` (negative)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cut_off_threshold = 0.3\n", - "\n", - "# Champion — Application scorecard model\n", - "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", - "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Random forest classification model\n", - "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", - "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", - "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Initializing the ValidMind objects" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind datasets\n", - "\n", - "Before you can run tests, you'll need 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", - "- **`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." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the raw dataset\n", - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the preprocessed dataset\n", - "vm_preprocess_dataset = vm.init_dataset(\n", - " dataset=preprocess_df,\n", - " input_id=\"preprocess_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the feature engineered dataset\n", - "vm_fe_dataset = vm.init_dataset(\n", - " dataset=fe_df,\n", - " input_id=\"fe_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the training dataset\n", - "vm_train_ds = vm.init_dataset(\n", - " dataset=train_df,\n", - " input_id=\"train_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the test dataset\n", - "vm_test_ds = vm.init_dataset(\n", - " dataset=test_df,\n", - " input_id=\"test_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "After initialization, you can pass the ValidMind `Dataset` objects `vm_raw_dataset`, `vm_preprocess_dataset`, `vm_fe_dataset`, `vm_train_ds`, and `vm_test_ds` into any ValidMind tests." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the model objects\n", - "\n", - "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", - "\n", - "You simply initialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the champion application scorecard model\n", - "vm_xgb_model = vm.init_model(\n", - " xgb_model,\n", - " input_id=\"xgb_model_developer_champion\",\n", - ")\n", - "\n", - "# Initialize the challenger random forest classification model\n", - "vm_rf_model = vm.init_model(\n", - " rf_model,\n", - " input_id=\"rf_model\",\n", - ")\n", - "\n", - "# Initialize the challenger logistic regression model\n", - "vm_log_model = vm.init_model(\n", - " log_reg,\n", - " input_id=\"log_model\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Assign predictions\n", - "\n", - "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", - "\n", - "- The [`assign_predictions()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset.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." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=train_xgb_binary_predictions,\n", - " prediction_probabilities=train_xgb_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=test_xgb_binary_predictions,\n", - " prediction_probabilities=test_xgb_prob,\n", - ")\n", - "\n", - "# Challenger — Random forest classification model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=train_rf_binary_predictions,\n", - " prediction_probabilities=train_rf_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=test_rf_binary_predictions,\n", - " prediction_probabilities=test_rf_prob,\n", - ")\n", - "\n", - "\n", - "# Challenger — Logistic regression model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=train_log_binary_predictions,\n", - " prediction_probabilities=train_log_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=test_log_binary_predictions,\n", - " prediction_probabilities=test_log_prob,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Compute credit risk scores\n", - "\n", - "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Compute the scores\n", - "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", - "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", - "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", - "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", - "train_log_scores = lending_club.compute_scores(train_log_prob)\n", - "test_log_scores = lending_club.compute_scores(test_log_prob)\n", - "\n", - "# Assign scores to the datasets\n", - "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", - "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", - "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", - "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", - "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", - "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Enabling custom context for test descriptions" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", - "\n", - "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", - "\n", - "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", - "

\n", - "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Review default LLM-generated descriptions\n", - "\n", - "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", - "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate. We'll go over running tests in more detail in a subsequent notebook:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Enable use case context\n", - "\n", - "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", - "\n", - "This is a global setting that will affect all tests for your linked model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" - ] - }, - { - "cell_type": "markdown", - "id": "0d1e90ba", - "metadata": {}, - "source": [ - "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "87f7fb2e", - "metadata": {}, - "outputs": [], - "source": [ - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "id": "d6e38bd2", - "metadata": {}, - "source": [ - "With the use case context set, generate an updated test description for the `HighPearsonCorrelation` test for comparision with default output:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run data quality tests\n", - "\n", - "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Identify qualitative tests\n", - "\n", - "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", - "\n", - "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `classification` 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 `data_quality` tag." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tasks_and_tags()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Then we'll call [`vm.tests.list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to list all the data quality tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests(\n", - " tags=[\"data_quality\"], task=\"classification\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "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
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Log data quality tests\n", - "\n", - "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "dq = vm.tests.list_tests(tags=[\"data_quality\"], task=\"classification\",pretty=False)\n", - "dq" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "With our data quality tests identified, let's run our first batch of tests using our previously initialized preprocessed dataset (`vm_preprocess_dataset`) and log their results.\n", - "\n", - "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", - "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in dq:\n", - " vm.tests.run_test(\n", - " test,\n", - " inputs={\n", - " \"dataset\": vm_preprocess_dataset\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", - "

\n", - "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Run comparison tests\n", - "\n", - "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in dq:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run performance tests\n", - "\n", - "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", - "\n", - "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "vm.tests.list_tests(tags=[\"model_performance\"], task=\"classification\")" - ] - }, - { + "cell_type": "markdown", + "id": "434bea70", + "metadata": {}, + "source": [ + "## 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 comparison and other validation tests, and then use the ValidMind Platform to submit compliance assessments of champion models via comprehensive validation reports. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model developers." + ] + }, + { + "cell_type": "markdown", + "id": "662d6027", + "metadata": {}, + "source": [ + "### 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": "5a1a2fa8", + "metadata": {}, + "source": [ + "### 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, create a free ValidMind account.\n", + "

\n", + "Signing up is FREE — Register with ValidMind
" + ] + }, + { + "cell_type": "markdown", + "id": "aa9b1d8e", + "metadata": {}, + "source": [ + "### Key concepts\n", + "\n", + "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", + "\n", + "**Validation report template**: Serves as a standardized framework for conducting and documenting model validation activities. It outlines the required sections, recommended analyses, and expected validation tests, ensuring consistency and completeness across validation reports. The template helps guide validators through a systematic review process while promoting comparability and traceability of validation outcomes.\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.\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/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." + ] + }, + { + "cell_type": "markdown", + "id": "057c1417", + "metadata": {}, + "source": [ + "## Setting up" + ] + }, + { + "cell_type": "markdown", + "id": "ef1f5ce4", + "metadata": {}, + "source": [ + "### Register a sample model\n", + "\n", + "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", + "\n", + "For this series of notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\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 **Continue**. ([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:\n", + "\n", + " - Documentation template: `Credit Risk Scorecard`\n", + " - Use case: `Credit Risk — CECL`\n", + "\n", + " You can fill in other options according to your preference." + ] + }, + { + "cell_type": "markdown", + "id": "9743927c", + "metadata": {}, + "source": [ + "#### Assign validator credentials\n", + "\n", + "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", + "\n", + "1. Remove yourself as a developer: \n", + "\n", + " - Click on the **DEVELOPERS** tile.\n", + " - Click the **x** next to your name to remove yourself from that model's role.\n", + " - Click **Save** to apply your changes to that role.\n", + "\n", + "2. Add yourself as a validator: \n", + "\n", + " - Click on the **VALIDATORS** tile.\n", + " - Select your name from the drop-down menu.\n", + " - Click **Save** to apply your changes to that role." + ] + }, + { + "cell_type": "markdown", + "id": "37a94e48", + "metadata": {}, + "source": [ + "### 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, + "id": "cef99db1", + "metadata": {}, + "outputs": [], + "source": [ + "%pip install -q validmind" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the ValidMind Library\n", + "\n", + "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." + ] + }, + { + "cell_type": "markdown", + "id": "30803184", + "metadata": {}, + "source": [ + "#### Get your code snippet\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "\n", + "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", + "\n", + "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": "d5de450a", + "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": "d22ef356", + "metadata": {}, + "source": [ + "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", + "\n", + "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "38592086", + "metadata": {}, + "outputs": [], + "source": [ + "vm.preview_template()" + ] + }, + { + "cell_type": "markdown", + "id": "943a45d6", + "metadata": {}, + "source": [ + "## Importing the champion model\n", + "\n", + "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3f9e197e", + "metadata": {}, + "outputs": [], + "source": [ + "import xgboost as xgb\n", + "\n", + "#Load the saved model\n", + "xgb_model = xgb.XGBClassifier()\n", + "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", + "xgb_model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b6887d38", + "metadata": {}, + "outputs": [], + "source": [ + "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", + "cols_when_model_builds = xgb_model.get_booster().feature_names" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Load the sample dataset\n", + "\n", + "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", + "\n", + "- We'll use this dataset to review steps that should have been conducted during the initial development and documentation of the model to ensure that the model was built correctly.\n", + "- By independently performing steps such as preprocessing and feature engineering, we can confirm whether the model was built using appropriate and properly processed data.\n", + "\n", + "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "40e73b7c", + "metadata": {}, + "outputs": [], + "source": [ + "from validmind.datasets.credit_risk import lending_club\n", + "\n", + "df = lending_club.load_data(source=\"offline\")\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Preprocess the dataset\n", + "\n", + "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", + "\n", + "- Filters the dataset to include only loans for debt consolidation or credit card purposes\n", + "- Removes loans classified under the riskier grades \"F\" and \"G\"\n", + "- Excludes uncommon home ownership types and standardizes employment length and loan terms into numerical formats\n", + "- Discards unnecessary fields and any entries with missing information to maintain a clean and robust dataset for modeling" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e436e862", + "metadata": {}, + "outputs": [], + "source": [ + "preprocess_df = lending_club.preprocess(df)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Apply feature engineering to the dataset\n", + "\n", + "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", + "\n", + "We'll apply the following transformations using the `ending_club.feature_engineering()` function to optimize the dataset for predictive modeling in our application scorecard:\n", + "\n", + "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", + "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9c6b641c", + "metadata": {}, + "outputs": [], + "source": [ + "fe_df = lending_club.feature_engineering(preprocess_df)\n", + "fe_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Split the feature engineered dataset\n", + "\n", + "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", + "\n", + "Splitting our dataset into training and testing is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", + "\n", + "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", + "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Split the data\n", + "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", + "\n", + "x_train = train_df.drop(lending_club.target_column, axis=1)\n", + "y_train = train_df[lending_club.target_column]\n", + "\n", + "x_test = test_df.drop(lending_club.target_column, axis=1)\n", + "y_test = test_df[lending_club.target_column]\n", + "\n", + "# Now let's apply the order of features from the champion model construction\n", + "x_train = x_train[cols_when_model_builds]\n", + "x_test = x_test[cols_when_model_builds]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cols_use = ['annual_inc_woe',\n", + " 'verification_status_woe',\n", + " 'emp_length_woe',\n", + " 'installment_woe',\n", + " 'term_woe',\n", + " 'home_ownership_woe',\n", + " 'purpose_woe',\n", + " 'open_acc_woe',\n", + " 'total_acc_woe',\n", + " 'int_rate_woe',\n", + " 'sub_grade_woe',\n", + " 'grade_woe','loan_status']\n", + "\n", + "\n", + "train_df = train_df[cols_use]\n", + "test_df = test_df[cols_use]\n", + "test_df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Investigate potential challenger models\n", + "\n", + "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", + "\n", + "Our selected options below offer decreased complexity in terms of implementation — such as lessened manual preprocessing — which can reduce the amount of risk for implementation. However, model risk is not calculated in isolation from a single factor, but rather in consideration with trade-offs in predictive performance, ease of interpretability, and overall alignment with business objectives." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Random forest classification model\n", + "\n", + "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", + "\n", + "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in their predictions." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Random Forest Classification model\n", + "from sklearn.ensemble import RandomForestClassifier\n", + "\n", + "# Create the model instance with 50 decision trees\n", + "rf_model = RandomForestClassifier(\n", + " n_estimators=50,\n", + " random_state=42,\n", + ")\n", + "\n", + "# Train the model\n", + "rf_model.fit(x_train, y_train)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Logistic regression model\n", + "\n", + "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", + "\n", + "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients (numerical value that represents the influence of a particular input feature on the model's prediction), but they may struggle with capturing complex, non-linear relationships in the data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Import the Logistic Regression model\n", + "from sklearn.linear_model import LogisticRegression\n", + "\n", + "# Logistic Regression grid params\n", + "log_reg_params = {\n", + " \"penalty\": [\"l1\", \"l2\"],\n", + " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", + " \"solver\": [\"liblinear\"],\n", + "}\n", + "\n", + "# Grid search for Logistic Regression\n", + "from sklearn.model_selection import GridSearchCV\n", + "\n", + "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", + "grid_log_reg.fit(x_train, y_train)\n", + "\n", + "# Logistic Regression best estimator\n", + "log_reg = grid_log_reg.best_estimator_\n", + "log_reg" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Extract predicted probabilities\n", + "\n", + "With our challenger models trained, let's extract the predicted probabilities from our three models:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Champion — Application scorecard model\n", + "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", + "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", + "\n", + "# Challenger — Random forest classification model\n", + "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", + "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", + "\n", + "# Challenger — Logistic regression model\n", + "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", + "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Compute binary predictions\n", + "\n", + "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", + "\n", + "- If the probability is greater than `0.3`, the prediction becomes `1` (positive).\n", + "- Otherwise, it becomes `0` (negative)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "cut_off_threshold = 0.3\n", + "\n", + "# Champion — Application scorecard model\n", + "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", + "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", + "\n", + "# Challenger — Random forest classification model\n", + "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", + "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", + "\n", + "# Challenger — Logistic regression model\n", + "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", + "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Initializing the ValidMind objects" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the ValidMind datasets\n", + "\n", + "Before you can run tests, you'll need 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", + "- **`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." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the raw dataset\n", + "vm_raw_dataset = vm.init_dataset(\n", + " dataset=df,\n", + " input_id=\"raw_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the preprocessed dataset\n", + "vm_preprocess_dataset = vm.init_dataset(\n", + " dataset=preprocess_df,\n", + " input_id=\"preprocess_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the feature engineered dataset\n", + "vm_fe_dataset = vm.init_dataset(\n", + " dataset=fe_df,\n", + " input_id=\"fe_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the training dataset\n", + "vm_train_ds = vm.init_dataset(\n", + " dataset=train_df,\n", + " input_id=\"train_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")\n", + "\n", + "# Initialize the test dataset\n", + "vm_test_ds = vm.init_dataset(\n", + " dataset=test_df,\n", + " input_id=\"test_dataset\",\n", + " target_column=lending_club.target_column,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "After initialization, you can pass the ValidMind `Dataset` objects `vm_raw_dataset`, `vm_preprocess_dataset`, `vm_fe_dataset`, `vm_train_ds`, and `vm_test_ds` into any ValidMind tests." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Initialize the model objects\n", + "\n", + "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", + "\n", + "You simply initialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Initialize the champion application scorecard model\n", + "vm_xgb_model = vm.init_model(\n", + " xgb_model,\n", + " input_id=\"xgb_model_developer_champion\",\n", + ")\n", + "\n", + "# Initialize the challenger random forest classification model\n", + "vm_rf_model = vm.init_model(\n", + " rf_model,\n", + " input_id=\"rf_model\",\n", + ")\n", + "\n", + "# Initialize the challenger logistic regression model\n", + "vm_log_model = vm.init_model(\n", + " log_reg,\n", + " input_id=\"log_model\",\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Assign predictions\n", + "\n", + "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", + "\n", + "- The [`assign_predictions()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset.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." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Champion — Application scorecard model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=train_xgb_binary_predictions,\n", + " prediction_probabilities=train_xgb_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_xgb_model,\n", + " prediction_values=test_xgb_binary_predictions,\n", + " prediction_probabilities=test_xgb_prob,\n", + ")\n", + "\n", + "# Challenger — Random forest classification model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=train_rf_binary_predictions,\n", + " prediction_probabilities=train_rf_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_rf_model,\n", + " prediction_values=test_rf_binary_predictions,\n", + " prediction_probabilities=test_rf_prob,\n", + ")\n", + "\n", + "\n", + "# Challenger — Logistic regression model\n", + "vm_train_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=train_log_binary_predictions,\n", + " prediction_probabilities=train_log_prob,\n", + ")\n", + "\n", + "vm_test_ds.assign_predictions(\n", + " model=vm_log_model,\n", + " prediction_values=test_log_binary_predictions,\n", + " prediction_probabilities=test_log_prob,\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Compute credit risk scores\n", + "\n", + "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Compute the scores\n", + "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", + "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", + "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", + "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", + "train_log_scores = lending_club.compute_scores(train_log_prob)\n", + "test_log_scores = lending_club.compute_scores(test_log_prob)\n", + "\n", + "# Assign scores to the datasets\n", + "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", + "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", + "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", + "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", + "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", + "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run data quality tests\n", + "\n", + "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Identify qualitative tests\n", + "\n", + "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", + "\n", + "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `classification` 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 `data_quality` tag." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tasks_and_tags()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then we'll call [`vm.tests.list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to list all the data quality tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests(\n", + " tags=[\"data_quality\"], task=\"classification\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "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
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Log data quality tests\n", + "\n", + "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "dq = vm.tests.list_tests(tags=[\"data_quality\"], task=\"classification\",pretty=False)\n", + "dq" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "With our data quality tests identified, let's run our first batch of tests using our previously initialized preprocessed dataset (`vm_preprocess_dataset`) and log their results.\n", + "\n", + "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", + "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in dq:\n", + " vm.tests.run_test(\n", + " test,\n", + " inputs={\n", + " \"dataset\": vm_preprocess_dataset\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "

\n", + "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Run comparison tests\n", + "\n", + "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in dq:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run performance tests\n", + "\n", + "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", + "\n", + "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "\n", + "vm.tests.list_tests(tags=[\"model_performance\"], task=\"classification\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We'll isolate the specific tests we want to run in `mpt`:\n", + "\n", + "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", + "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", + "- [`MinimumAccuracy`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumAccuracy.html)\n", + "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", + "- [`ROCCurve`](https://docs.validmind.ai/tests/model_validation/sklearn/ROCCurve.html)\n", + "\n", + "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "mpt = [\n", + " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion\",\n", + " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion\"\n", + "]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Evaluate performance of the champion model\n", + "\n", + "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in mpt:\n", + " vm.tests.run_test(\n", + " test,\n", + " inputs={\n", + " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Evaluate performance of challenger models\n", + "\n", + "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", + "\n", + "Next, let's see how our challenger models compare. We'll use the same batch of tests here as we did in `mpt`, but append a different `result_id` to indicate that these results should be associated with our challenger models:\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "mpt_chall = [\n", + " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", + " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers\"\n", + "]" + ] + }, + { "cell_type": "markdown", "metadata": {}, "source": [ - "We'll isolate the specific tests we want to run in `mpt`:\n", - "\n", - "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", - "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", - "- [`MinimumAccuracy`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumAccuracy.html)\n", - "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", - "- [`ROCCurve`](https://docs.validmind.ai/tests/model_validation/sklearn/ROCCurve.html)\n", - "\n", - "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here:" + "## Enabling custom context for test descriptions" ] }, { - "cell_type": "code", - "execution_count": null, + "cell_type": "markdown", "metadata": {}, - "outputs": [], "source": [ - "mpt = [\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion\"\n", - "]" + "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", + "\n", + "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", + "\n", + "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", + "

\n", + "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", + "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### Evaluate performance of the champion model\n", + "#### Review default LLM-generated descriptions\n", + "\n", + "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", "\n", - "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" + "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate. We'll go over running tests in more detail in a subsequent notebook:" ] }, { @@ -1156,24 +1038,23 @@ "metadata": {}, "outputs": [], "source": [ - "for test in mpt:\n", - " vm.tests.run_test(\n", - " test,\n", - " inputs={\n", - " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", - " },\n", - " ).log()" + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " }\n", + ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### Evaluate performance of challenger models\n", + "#### Enable use case context\n", "\n", - "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", + "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", "\n", - "Next, let's see how our challenger models compare. We'll use the same batch of tests here as we did in `mpt`, but append a different `result_id` to indicate that these results should be associated with our challenger models:\n" + "This is a global setting that will affect all tests for your linked model:" ] }, { @@ -1182,31 +1063,25 @@ "metadata": {}, "outputs": [], "source": [ - "mpt_chall = [\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers\"\n", - "]" + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" ] }, { "cell_type": "markdown", + "id": "0d1e90ba", "metadata": {}, "source": [ - "We'll also adjust the use case context to focus on this comparison for tests going forward:" + "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" ] }, { "cell_type": "code", "execution_count": null, + "id": "87f7fb2e", "metadata": {}, "outputs": [], "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", - "\n", "context = \"\"\"\n", "FORMAT FOR THE LLM DESCRIPTIONS: \n", " **** is designed to \n", "\n", "ADDITIONAL INSTRUCTIONS:\n", - "\n", - " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", - " - The metrics for the champion model compared against the challenger models\n", - " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", - "\n", + " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", + " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", "\n", " For each metric in the test results, include in the test overview:\n", " - The metric's purpose and what it measures\n", @@ -1256,500 +1128,25 @@ " 6. Notable relationships or dependencies\n", "\n", " Remember:\n", - " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", " - Keep all insights at the same level (no sub-bullets or nested structures)\n", " - Make each insight complete and self-contained\n", " - Include specific numerical values and ranges\n", " - Cover all elements in the results comprehensively\n", " - Maintain clear, concise language\n", " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We'll run each test once for each model with the same `vm_test_ds` dataset to compare them:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in mpt_chall:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Based on the performance metrics, we can conclude that the random forest classification model is not a viable candidate for our use case and can be disregarded in our tests going forward.\n", - "

\n", - "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Adjust a ValidMind test\n", - "\n", - "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", - "\n", - "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_xgb_model, vm_log_model]\n", - " },\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "As `MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_xgb_model, vm_log_model],\n", - " \"params\": {\"min_threshold\": 0.35}\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", - "

\n", - "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run diagnostic tests\n", - "\n", - "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", - "\n", - "Use [`list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to identify all the model diagnosis tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "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
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues. Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data.\n", - "\n", - "We'll select the following two tests:\n", - "\n", - "- **[`TrainingTestDegradation`](https://docs.validmind.ai/tests/model_validation/sklearn/TrainingTestDegradation.html):** Evaulates how much the models' performance degrade when moving from the training to the test datasets.\n", - "- **[`OverfitDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html):** Dives into detecting overfitting beyond performance degradation by examining learning curves, model complexity v.s. generalization, and cross-validation results." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store our desired 2 tests\n", - "overfit_testing = [\n", - " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", - "]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Run and log our 2 tests for both models for the training and testing datasets\n", - "for test in overfit_testing:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", - " \"model\" : [vm_xgb_model,vm_log_model]\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's also conduct *robustness* and *stability* testing of the two models. Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n", - "\n", - "We'll select the following two tests:\n", - "\n", - "- **[`RobustnessDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html):** Evaluates whether the models are resilient to small *perturbations* (small, controlled modifications) or *variations* (broader subgroup-level or distributional changes) in input data.\n", - "- **[`WeakspotsDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/WeakspotsDiagnosis.html):** Identifies specific conditions where the models consistently underperform." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store our desired 2 tests\n", - "stab_robust = [\n", - " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", - " ]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Run and log our 2 tests for both models for the training and testing datasets\n", - "for test in stab_robust:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", - " \"model\" : [vm_xgb_model,vm_log_model]\n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run feature importance tests\n", - "\n", - "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", - "\n", - "Use `list_tests()` to identify all the feature importance tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store the feature importance tests\n", - "FI = vm.tests.list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", - "FI" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Run and log our feature importance tests for both models for the testing dataset\n", - "for test in FI:\n", - " vm.tests.run_test(\n", - " \"\".join((test,':Champion_vs_LogisticRegression')),\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model]\n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Implement a custom test\n", - "\n", - "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", - "\n", - "The [`@vm.test` wrapper](https://docs.validmind.ai/validmind/validmind.html#test) allows you to create a reusable test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import numpy as np\n", - "import pandas as pd\n", - "import plotly.graph_objects as go\n", - "\n", - "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", - "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", - " \"\"\"\n", - " Analyzes the relationship between score bands and odds (good:bad ratio).\n", - " Good odds = (1 - default_rate) / default_rate\n", - " \n", - " Higher scores should correspond to higher odds of being good.\n", - "\n", - " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", - "\n", - " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", - " \"\"\"\n", - " df = dataset.df\n", - " \n", - " # Create score bands\n", - " df['score_band'] = pd.cut(\n", - " df[score_column],\n", - " bins=[-np.inf] + score_bands + [np.inf],\n", - " labels=[f'<{score_bands[0]}'] + \n", - " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", - " [f'>{score_bands[-1]}']\n", - " )\n", - " \n", - " # Calculate metrics per band\n", - " results = df.groupby('score_band').agg({\n", - " dataset.target_column: ['mean', 'count']\n", - " })\n", - " \n", - " results.columns = ['Default Rate', 'Total']\n", - " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", - " results['Bad Count'] = results['Default Rate'] * results['Total']\n", - " results['Odds'] = results['Good Count'] / results['Bad Count']\n", - " \n", - " # Create visualization\n", - " fig = go.Figure()\n", - " \n", - " # Add odds bars\n", - " fig.add_trace(go.Bar(\n", - " name='Odds (Good:Bad)',\n", - " x=results.index,\n", - " y=results['Odds'],\n", - " marker_color='blue'\n", - " ))\n", - " \n", - " fig.update_layout(\n", - " title='Score-to-Odds Analysis',\n", - " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", - " showlegend=False\n", - " )\n", - " \n", - " return fig" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "With the custom test available, run and log the test for our champion and challenger models with our testing dataset (`vm_test_ds`):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " param_grid={\n", - " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", - " \"score_bands\": [[500, 540, 570]],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Want to learn more about custom tests?\n", - "

\n", - "Refer to our in-depth introduction to custom tests: Implement custom tests
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Verify test runs\n", - "\n", - "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", - "\n", - "Here, we'll specify all the tests we'd like to independently rerun in a dictionary called `test_config`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "test_config = {\n", - " # Run with the raw dataset\n", - " 'validmind.data_validation.DatasetDescription:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'}\n", - " },\n", - " 'validmind.data_validation.DescriptiveStatistics:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'}\n", - " },\n", - " 'validmind.data_validation.MissingValues:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.ClassImbalance:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 10}\n", - " },\n", - " 'validmind.data_validation.Duplicates:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.HighCardinality:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {\n", - " 'num_threshold': 100,\n", - " 'percent_threshold': 0.1,\n", - " 'threshold_type': 'percent'\n", - " }\n", - " },\n", - " 'validmind.data_validation.Skewness:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.UniqueRows:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.TooManyZeroValues:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_percent_threshold': 0.03}\n", - " },\n", - " 'validmind.data_validation.IQROutliersTable:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'threshold': 5}\n", - " },\n", - " # Run with the preprocessed dataset\n", - " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.MissingValues:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'min_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'default_column': 'loan_status'}\n", - " },\n", - " # Run with the training and test datasets\n", - " 'validmind.data_validation.DescriptiveStatistics:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.TabularDescriptionTables:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.ClassImbalance:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 10}\n", - " },\n", - " 'validmind.data_validation.UniqueRows:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.TabularNumericalHistograms:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.MutualInformation:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'min_threshold': 0.01}\n", - " },\n", - " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.HighPearsonCorrelation:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", - " },\n", - " 'validmind.model_validation.ModelMetadata': {\n", - " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", - " },\n", - " 'validmind.model_validation.sklearn.ModelParameters': {\n", - " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", - " },\n", - " 'validmind.model_validation.sklearn.ROCCurve': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']}\n", - " },\n", - " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.5}\n", - " }\n", - "}" + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" ] }, { "cell_type": "markdown", + "id": "d6e38bd2", "metadata": {}, "source": [ - "Then batch run and log our tests in `test_config`:" + "With the use case context set, generate an updated test description for the `HighPearsonCorrelation` test for comparision with default output:" ] }, { @@ -1758,68 +1155,643 @@ "metadata": {}, "outputs": [], "source": [ - "for t in test_config:\n", - " print(t)\n", - " try:\n", - " # Check if test has input_grid\n", - " if 'input_grid' in test_config[t]:\n", - " # For tests with input_grid, pass the input_grid configuration\n", - " if 'params' in test_config[t]:\n", - " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", - " else:\n", - " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", - " else:\n", - " # Original logic for regular inputs\n", - " if 'params' in test_config[t]:\n", - " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", - " else:\n", - " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", - " except Exception as e:\n", - " print(f\"Error running test {t}: {str(e)}\")" + "vm.tests.run_test(\n", + " \"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " }\n", + ")" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Add test results to reporting\n", - "\n", - "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", - "\n", - "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", - "\n", - "2. In the left sidebar that appears for your model, click **Validation Report**.\n", - "\n", - "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", - "\n", - "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**:\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "5. Select the Minimum F1 Score test results we logged:\n", - "\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Adj Threshold**\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion**\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion Vs Challengers**\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "6. Finally, click **Update Linked Evidence** to add the test results to the validation report.\n", - "\n", - " Confirm that the results for the minimum F1 score tests have been correctly inserted into section **2.2.2. Model Performance** of the report:\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "7. Continue to work on your validation report by:\n", - "\n", - " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", - " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", - " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" - ] - } + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We'll also adjust the use case context to focus on this comparison for tests going forward:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", + "\n", + "context = \"\"\"\n", + "FORMAT FOR THE LLM DESCRIPTIONS: \n", + " **** is designed to .\n", + "\n", + " The test operates by \n", + "\n", + " The primary advantages of this test include \n", + "\n", + " Users should be aware that \n", + "\n", + " **Key Insights:**\n", + "\n", + " The test results reveal:\n", + "\n", + " - ****: \n", + " - ****: \n", + " ...\n", + "\n", + " Based on these results, \n", + "\n", + "ADDITIONAL INSTRUCTIONS:\n", + "\n", + " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", + " - The metrics for the champion model compared against the challenger models\n", + " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", + "\n", + "\n", + " For each metric in the test results, include in the test overview:\n", + " - The metric's purpose and what it measures\n", + " - Its mathematical formula\n", + " - The range of possible values\n", + " - What constitutes good/bad performance\n", + " - How to interpret different values\n", + "\n", + " Each insight should progressively cover:\n", + " 1. Overall scope and distribution\n", + " 2. Complete breakdown of all elements with specific values\n", + " 3. Natural groupings and patterns\n", + " 4. Comparative analysis between datasets/categories\n", + " 5. Stability and variations\n", + " 6. Notable relationships or dependencies\n", + "\n", + " Remember:\n", + " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", + " - Keep all insights at the same level (no sub-bullets or nested structures)\n", + " - Make each insight complete and self-contained\n", + " - Include specific numerical values and ranges\n", + " - Cover all elements in the results comprehensively\n", + " - Maintain clear, concise language\n", + " - Use only \"- **Title**: Description\" format for insights\n", + " - Progress naturally from general to specific observations\n", + "\n", + "\"\"\".strip()\n", + "\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We'll run each test once for each model with the same `vm_test_ds` dataset to compare them:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for test in mpt_chall:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Based on the performance metrics, we can conclude that the random forest classification model is not a viable candidate for our use case and can be disregarded in our tests going forward.\n", + "

\n", + "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Adjust a ValidMind test\n", + "\n", + "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", + "\n", + "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_xgb_model, vm_log_model]\n", + " },\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "As `MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold\",\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds],\n", + " \"model\": [vm_xgb_model, vm_log_model],\n", + " \"params\": {\"min_threshold\": 0.35}\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "

\n", + "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run diagnostic tests\n", + "\n", + "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", + "\n", + "Use [`list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to identify all the model diagnosis tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "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
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues. Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data.\n", + "\n", + "We'll select the following two tests:\n", + "\n", + "- **[`TrainingTestDegradation`](https://docs.validmind.ai/tests/model_validation/sklearn/TrainingTestDegradation.html):** Evaulates how much the models' performance degrade when moving from the training to the test datasets.\n", + "- **[`OverfitDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html):** Dives into detecting overfitting beyond performance degradation by examining learning curves, model complexity v.s. generalization, and cross-validation results." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Store our desired 2 tests\n", + "overfit_testing = [\n", + " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Run and log our 2 tests for both models for the training and testing datasets\n", + "for test in overfit_testing:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " }\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's also conduct *robustness* and *stability* testing of the two models. Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n", + "\n", + "We'll select the following two tests:\n", + "\n", + "- **[`RobustnessDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html):** Evaluates whether the models are resilient to small *perturbations* (small, controlled modifications) or *variations* (broader subgroup-level or distributional changes) in input data.\n", + "- **[`WeakspotsDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/WeakspotsDiagnosis.html):** Identifies specific conditions where the models consistently underperform." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Store our desired 2 tests\n", + "stab_robust = [\n", + " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", + " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", + " ]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Run and log our 2 tests for both models for the training and testing datasets\n", + "for test in stab_robust:\n", + " vm.tests.run_test(\n", + " test,\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Run feature importance tests\n", + "\n", + "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", + "\n", + "Use `list_tests()` to identify all the feature importance tests for classification:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Store the feature importance tests\n", + "FI = vm.tests.list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", + "FI" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Run and log our feature importance tests for both models for the testing dataset\n", + "for test in FI:\n", + " vm.tests.run_test(\n", + " \"\".join((test,':Champion_vs_LogisticRegression')),\n", + " input_grid={\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model]\n", + " },\n", + " ).log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Implement a custom test\n", + "\n", + "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", + "\n", + "The [`@vm.test` wrapper](https://docs.validmind.ai/validmind/validmind.html#test) allows you to create a reusable test:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "import plotly.graph_objects as go\n", + "\n", + "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", + "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", + " \"\"\"\n", + " Analyzes the relationship between score bands and odds (good:bad ratio).\n", + " Good odds = (1 - default_rate) / default_rate\n", + " \n", + " Higher scores should correspond to higher odds of being good.\n", + "\n", + " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", + "\n", + " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", + " \"\"\"\n", + " df = dataset.df\n", + " \n", + " # Create score bands\n", + " df['score_band'] = pd.cut(\n", + " df[score_column],\n", + " bins=[-np.inf] + score_bands + [np.inf],\n", + " labels=[f'<{score_bands[0]}'] + \n", + " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", + " [f'>{score_bands[-1]}']\n", + " )\n", + " \n", + " # Calculate metrics per band\n", + " results = df.groupby('score_band').agg({\n", + " dataset.target_column: ['mean', 'count']\n", + " })\n", + " \n", + " results.columns = ['Default Rate', 'Total']\n", + " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", + " results['Bad Count'] = results['Default Rate'] * results['Total']\n", + " results['Odds'] = results['Good Count'] / results['Bad Count']\n", + " \n", + " # Create visualization\n", + " fig = go.Figure()\n", + " \n", + " # Add odds bars\n", + " fig.add_trace(go.Bar(\n", + " name='Odds (Good:Bad)',\n", + " x=results.index,\n", + " y=results['Odds'],\n", + " marker_color='blue'\n", + " ))\n", + " \n", + " fig.update_layout(\n", + " title='Score-to-Odds Analysis',\n", + " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", + " showlegend=False\n", + " )\n", + " \n", + " return fig" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With the custom test available, run and log the test for our champion and challenger models with our testing dataset (`vm_test_ds`):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "result = vm.tests.run_test(\n", + " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds,\n", + " },\n", + " param_grid={\n", + " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", + " \"score_bands\": [[500, 540, 570]],\n", + " },\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Want to learn more about custom tests?\n", + "

\n", + "Refer to our in-depth introduction to custom tests: Implement custom tests
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Verify test runs\n", + "\n", + "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", + "\n", + "Here, we'll specify all the tests we'd like to independently rerun in a dictionary called `test_config`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "test_config = {\n", + " # Run with the raw dataset\n", + " 'validmind.data_validation.DatasetDescription:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'}\n", + " },\n", + " 'validmind.data_validation.DescriptiveStatistics:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'}\n", + " },\n", + " 'validmind.data_validation.MissingValues:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.ClassImbalance:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 10}\n", + " },\n", + " 'validmind.data_validation.Duplicates:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.HighCardinality:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {\n", + " 'num_threshold': 100,\n", + " 'percent_threshold': 0.1,\n", + " 'threshold_type': 'percent'\n", + " }\n", + " },\n", + " 'validmind.data_validation.Skewness:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.UniqueRows:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'min_percent_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TooManyZeroValues:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'max_percent_threshold': 0.03}\n", + " },\n", + " 'validmind.data_validation.IQROutliersTable:raw_data': {\n", + " 'inputs': {'dataset': 'raw_dataset'},\n", + " 'params': {'threshold': 5}\n", + " },\n", + " # Run with the preprocessed dataset\n", + " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.MissingValues:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'min_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'}\n", + " },\n", + " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {\n", + " 'inputs': {'dataset': 'preprocess_dataset'},\n", + " 'params': {'default_column': 'loan_status'}\n", + " },\n", + " # Run with the training and test datasets\n", + " 'validmind.data_validation.DescriptiveStatistics:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.TabularDescriptionTables:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.ClassImbalance:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 10}\n", + " },\n", + " 'validmind.data_validation.UniqueRows:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_percent_threshold': 1}\n", + " },\n", + " 'validmind.data_validation.TabularNumericalHistograms:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.MutualInformation:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'min_threshold': 0.01}\n", + " },\n", + " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", + " },\n", + " 'validmind.data_validation.HighPearsonCorrelation:development_data': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", + " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", + " },\n", + " 'validmind.model_validation.ModelMetadata': {\n", + " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.ModelParameters': {\n", + " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.ROCCurve': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']}\n", + " },\n", + " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']},\n", + " 'params': {'min_threshold': 0.5}\n", + " }\n", + "}" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Then batch run and log our tests in `test_config`:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "for t in test_config:\n", + " print(t)\n", + " try:\n", + " # Check if test has input_grid\n", + " if 'input_grid' in test_config[t]:\n", + " # For tests with input_grid, pass the input_grid configuration\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", + " else:\n", + " # Original logic for regular inputs\n", + " if 'params' in test_config[t]:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", + " else:\n", + " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", + " except Exception as e:\n", + " print(f\"Error running test {t}: {str(e)}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Add test results to reporting\n", + "\n", + "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", + "\n", + "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", + "\n", + "2. In the left sidebar that appears for your model, click **Validation Report**.\n", + "\n", + "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", + "\n", + "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**:\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "5. Select the Minimum F1 Score test results we logged:\n", + "\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Adj Threshold**\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion**\n", + " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion Vs Challengers**\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "6. Finally, click **Update Linked Evidence** to add the test results to the validation report.\n", + "\n", + " Confirm that the results for the minimum F1 score tests have been correctly inserted into section **2.2.2. Model Performance** of the report:\n", + "\n", + " \"Screenshot\n", + "

\n", + "\n", + "7. Continue to work on your validation report by:\n", + "\n", + " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", + " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", + " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" + ] + } ], "metadata": { "language_info": { From 1b0a78a15428f5168d2906b606a4a2efc5cb736e Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 13:10:52 -0700 Subject: [PATCH 79/90] Moving more stuff around --- .../inserted-minimum-f1-scores.png | Bin 0 -> 110789 bytes .../link-validator-evidence.png | Bin 0 -> 398858 bytes .../selecting-minimum-f1-scores.png | Bin 0 -> 327835 bytes .../validate_application_scorecard.ipynb | 328 +++++++----------- .../114-finalize_testing_reporting.ipynb | 4 +- 5 files changed, 133 insertions(+), 199 deletions(-) create mode 100644 notebooks/code_samples/model_validation/inserted-minimum-f1-scores.png create mode 100644 notebooks/code_samples/model_validation/link-validator-evidence.png create mode 100644 notebooks/code_samples/model_validation/selecting-minimum-f1-scores.png diff --git a/notebooks/code_samples/model_validation/inserted-minimum-f1-scores.png b/notebooks/code_samples/model_validation/inserted-minimum-f1-scores.png new file mode 100644 index 0000000000000000000000000000000000000000..52ae43c719a2621e56360d2f2f357256253a4fe5 GIT binary patch literal 110789 zcmeFYcT^MI`ZkQBARt8$kX}@f-a%TVsPx`@@4bc)0!mf7^o~gHp#}&=1Vp;j(2?E( zks3N*JkR-+^}TC7=e+;Df1Io}naRxFvom|&``-6;-Pe3nSCu2gr^d&^!Xi|Vf31my zh3AWfbtm`%4(5z%4jvB{)Cq^h2&QIH`=s8f>$nSJg)=Eb`I z^BsX1W2)zuFW>HPV(%>tz%#Kl8WFTfhPTMu>NiR+5(u!QzA&ni_1|wMVYr9oMk;Ru z!Xhb*7s<_F)mfP&+5Pc3ghY?cBOG6Z%_9TfY`gAb)UwRgJ*-->T%G|eseGEliU4Vx ziM%fR_ld7xq(~ugYt&r#t(m&Lm>S;;yQ{Wn=4PLUsD57(9LEtlzlxTwh%Bs*$L{uY zAPTy;O^>}b@$`C(?WNyUmcvHyQRDl=qtj<#&ot6<*(>P;ibr@g!dNf3DNbiN_yn{b zkyh8rx!m*Q9;7mtk$9U!Hn6*C6?*lK_X8o5YUkU7dD|l?}`4q?5I`WAPCk4sC>jd+lWPol4H$mh!Y$=_C;RBucC$^0@ z$E)R_bPuL7V+*x=SAbg!PJNjCw~q`zANBQy5qC2`zJIUGvbT>}A)q3%_oM)H!&g?* zlr6|wwnUGnT))Qw68NyXc3y*Rb#(M=MSFYo%~)&6%j8@?5XCD&#dYF#%1~_X2Uwh+ z-w6>a;}}TI&+07QM`=q+zazG0#QyM3_LTrHb4|6}-RhdpSXX$T9{j-VJx8BJN&(h7 zh%Az`{NlAt=%oCEay6gRNQAwl41G`QSG}c`@X$XVD@W?>6>;1d6yN-zD3ty3R`S)? zK)aXmcqqHa*aEd%*eTsomRNkDPhQve5V|wW6A(w}z0%M25Z>n_=>>wnY01zmKU&7g z%XYU*Xi^vM9;o?Dw}X4K$-38j^-EWMoE+WTmM5nDJ-z9U`&W}+614M_!$^Ds*N&zB z3TU|fEmtr@j(P`O&&OcRaE+@M!m@IX(+*8UWiPz0tpc^%iwHL|#is?l1NVt%D8C6i zmppHLApS19OVyGR%2GiOg|6yjPX*KItaOs23*34L2Pwuk)r2RgT5Q{28Q8F_s|u-R ztD0#;Kwlg{{X;*pIkUOR?5sCxUjB|^vOIWpxAzgJ9jl!lR?-r~P;%|}3+As%a8-Ui zdr83kp5qrz6M>!@uHT)OPVwJV(j=rZvQIt$EQ9r#Pd^BJdm9ih^^EqTbV9uLGwe@{ z_|`1^3~bW6%Q5Sb@r%2AomfTTe0*AicdV?8IZ2hikMU6n({OxO ztWm6>a0(qcr=E)Sv0(6E9{y_3gl*i(h*VfCkSEPel4EDCe7eU>$7=H2Cms+#`GdvU&XYtmos)?&L8(u?ulI*u z-^h<#wv5M0k*v=Xm7?B)02E-lGm;ciMYT!5#MpM7tA0Zic3)*$mc55 z8CM;p{|T_?6{HpNqn3#gr6#BLq1F>J9b@xW5{m$QGL_;fWcX}W1a#08tQ2g{2pF2l zAkT;pa&tJf=c|9~@Wyf7k!y6e)JJ-^Sy*9CQC!`qq)iH3e#*Aj`!L#K_!s}Z=<{fo zXn|-Vx-YQ}L~5)`#jovPs^zM%s>Z4)*Wi7@1^k8lmObTYhn_V)ue5=*%XE?S+;A$w z)rVw+MiLdi(~}mclkna1?T^QiC#PErKQ9YL#)v0Irp)rps)LrWM4$p6gwq_Mqral1 z5$@T`rd{{Dq9RZXHcTSQsJeX7xbewwh)9%aktwkoR0^+-8zT($b1 z&GL|pfvl#_F>#e82t4mZK?UOO-;%HB8S;o6yIqP^aGdRn+Sn<<2ZU*92CORjv z5YPas1Hr;R_UK2YQwULX7-(|Nq@<@fag7+X!)hB9>kaL7+)m~VM>6b8%uTEyBjL8I zP69neV;e18qbV+#F7VFNNIJ(&$5aPB3*ZcW1F(*A&JF5DBmHEVPlo@CXEztF4{oe% zyKGFCtXz>*lU1Ho+v>n4il>QkRZl>!Ya*?oL2WNtg+F*XLOK5;5MEJ7) zGUMV9!U178Prt;wZoHISo%XDK34h6Roqc6<6LGcVCD~lp#CjL{?$^7OcT-Y(U(dcC ze#Moxm8Oywl3wVn>J$i83nh{HCZiiVze06zuhr2MSOVUv_&IK|V^L$VZV~rw@0-nX zjMZpOo|Rrq8>Iym)5nsyd!Jf==zK!wSL7GHru)e9R_=}7>sK*UWY#X)$=RZQ><=jn z&Lk=gs|-nUj;2^Esj3O*NVs@rPM8`#bIDy1wHZosHAr)Y(|%Hm7g^i>%5Y5)7PG?I z(-n|=6?sLhMOH+gRwSgtUun~_aNIMVOMZ|1iXS_dDQ#3@+|D(a4k}|CdyzoCV7S%bKF%EQ z$xmZtW3HglnuS|CJMdB4R;KCW>0H>*M|F0?50#-YUUUO=6~z39P#fu zi9&QNG0waD`3t8%w<<&a^&UI_~1zP_7 z=CtQR$u)Jy(AbFRdgo#{PZu`TGO(XYQ{`vu;7l{_i!4J;CL-=uWw#B(1GnFH$24Xh zh`22&Hcti``CXo8XNgw_4&3e?Ei|Japd(REi#!nXv(JaXxg~TxynzAo|7Ba93Rtjt1AXNo&p{Kko>yzWb-0bn1}dg9j&K? zZrYF4tJQ0(PHi0xs#~obnA^xH$-zZ=EdxkNnT2v5Ns>W=*?Bf zEPZQV*d5y2OFp!@YB#&{@I{>PHdgE69V{+9EF1|D&EE3i2*`bnY==vdGc|pma3qH- z*2K46H_mo|6AUnaM9716^h{}Kp_H9wg zaKWoH$@=?>I+VQi*OVc`A!IN3_$V>xBGN`*!B#~Diyd?P0PEfzYOMR1qdS-laEIo< zkKf*Tfrb5#^LMeZA{?;p{j-cJX8-3CkJ4z5RQe5`9+G^{663PuhMwFpRxbz z_)ak98kV${tbzh&uVv+FW8>;&@8(@Roo04%+(O z`YOs|R&Flb7S?W-Hr#$L?tkusCE+KAIdrk{wqW#gad!0*^OI!$M+q^^@t@N?%#8mi z;_W2KtgoWZDC_2F!zj$n%gxISz-MG+l<>5+71MnE_Mg=;za*LMy}jMVczA$7AU9Be z+s)IChfh>gl!up}ho7GdQ-aG2Cljj1Z}=d75zgP)DF!D|N> zOrK%e00_JkmiR~c|KrO4wD`Yj>iWaVm`5$L79Sy*j;Q8-C1K=k%KRCyXBdx=0HEqluV`cyNbYi|= zV75Pd%+?s{x1z{}g(Zch@LF2i?+$VSuT|NHq4yvc5`~;et+gKDR#GS|vF2b?NQrNI zFe96UOmOZad!rpn+ONrOQ~E6Vc|iij42*N+7U(@J>9s$f-i#))ts1!n*$fBXY(j?T zo8!(et`oCGRO}xxzQelv?|wac8Xs-0*F8@l7X0s5dZ2&z9^8+FL<;N9y?^iP{R1+_ zX>tpZ(Eb0LKZj`q=ij^kKmNym8u?Fm{a={=f2`|2lkwkiVE=I||8Xn-kGPfBDPO<4 zh_8JM60eNa;q=WunXNKi__6b%P5O@^%WGnm<%Eg7HOqfv(fs!Ovqe>@QcAiyB1|f^ zW3{mySWNQ!khP2d&yxPW8=nNJSiyMt^j|;uGtu{@N&w!xs`X(Xhh%QJW-7OL_0#!v zUcUE#LJE?%cY;kUzDj@&lOBQtqd7rA>Dr~*(mDH+D)^M_Pr{#ieXyta)%;-tF1PgP zZzGB->pQ>_r7q*Wok_mDD&`vb`8k#C#Zi=Sjc?0qf*Ub!{TspCmT#8&(Bo7L5B|K zXRB)23oOwzOt5+5g>*aSMk&}n#QG$op++whx;7+Y`{Sf1DjaHaCA{4HA(ztVUvw%7 z-rap@N_b0=H|jCeXn#*cnV!sjPF;^4;A;$6yvnWv^~u~{o1I6zJnCej2x>RY$v!E{ zk_4){a%*XR_x0zT<~}eCIy`nbcV8751{C5`#(!&;LFLw2(ULI)G%E}@@!#`s-l%I* z{AigVa5M=RAQd1Z>XPk^?PJY^i7k|XuA^7Rp>x5jOn+~>B!qYO<8@!;abIT4#(@-sUFRHR%$G}NPr61r=)g&5xox23 zeX^BnM?SmRRl~VQW!mM2GwE?DALSqVzB%%G#D&fdWd0Xj#{E7tyyf`H7g}!8CO!m> z@gZ9%?8bo?ao$K3wZZ0h%mEbO_q6|O_S%aR&`CM+RX8E7~pb% zO_-?0YgGJ*XyfujsgbqB=@KB%ZS`Aj2OW`s8S=~N(F?JeE8~m~G27#oxVOYHQ%cPR z*RdbW5gJ;7P~*Z3&#~kth^>?xsN&?QG@ohCMcG3zcafN0kuuf#?upYgpPdP5G@(z3 zc_+8zaJqoqngCfmyHQb+F~p}xop=u@9&8e5aF_cS2ku&>p`w_~N$<`8Ubm_=3VqC6 z4~rXdss~UWG{0?p^Y^g<vYjy}7&=ZZxmwEhBcX4$fGls|=>l=>$Ya(~e2w(6 zVgm0LziN0=XH^ANH7Q*rzrItDn+cl+qZxEa#*~3bHrRy83wK_<9I*|SF}g1xu`|Kg zm5(E#e^<{x#AG%udYiko`q<$7=%>Z`dE6@q`_OX{i~em%_zyw^h2k%R#3@(917*5B zimE)Pz18QFhz}2Z*<6dOB-a;ul<4f}NZk*DMY7JDSayU0B=hx0bf6 z2b-LpcO57~Ic_2s%@M!zunXQC6&<`%zlZW*%Z++q``PJVK*&&?uS7Q6(qz4c*EY>l`KNun(i8X_$E^bWADu@%@D1Cfe`5)C(p}oG2@Vwt*eZB~8kkSceMxC5 z=tiF|@QcmvMlpwRIvEOQ;Ond+)~k9%Sk*MQnai1Yb-qIHb)0GzlvpORu)X+ZV>pLf zWU^$ZZZJpG^SA^#eUNz7hyscRUWKmpWh>#Qb-RA{+G_6%Ho+yPPiX77a?E{OYu(3r zKA0(>KG*0HaaHdWmvw_SYwipnA?J)=VuMgEVW|aFD{r?{q9$~F z0rA^vY)%01sdIujLTFeF)qQgvq>uQ$>vs9fWma%Xh?2`c+gd(%5nejdir_#9i9VN$ zMNSF7a-HKgQue%B3L23=anvc0`d~F2{&aeMPp?e7bnRm04dh}!`(QuuyLKdpZpDX& zbl^f^Aa_04MaAv$xa+Ras*avQdjLrJdEM@t0RWEUML>=iWOlZD7up+*eI3%Q)*DAh z*|zj1%GilentID`(P!UIE;lv!{*h8%@cm~HiQB@jOo{%BjgaCi$3TiSk@&T~M5X>D zcDC)Y!jqf?(7yoeSNPigt?-6n{=vWd_#QJXX*Au$F+}Ct@R)ywS4nb!0kqL7&X%HcGY(m zZg%;kkbp8tnd)v(ZpGYGz?V3!^B3)%RZA+#wYhTSitM-tkn*qT2>M!Z$e(ex; z>vkm`FBfT6rC*Y6FZT=4X~jeKHfE^l=N$|$8GDr*8PysqG^f{a@Qo9@;0{?w2W_!U zM91e-3~N`-NBoCF_iNE39Th{7YeQTnG?R5jb!m?lR?5bFe{b=fPg(OHgI)&Rw667M zyg5D}$9KnbSd28e2{!RJmekAC;OXDUhG&mg(2LK{tKBYHFuV+S-H5!I*|Ebvk_>cR zt$YWXGB*6H%SZZ~V6YttLflH|BQbj|LTV^c{@p{VpB z^-?kc;9|(8R{-^4hj1eT;gkNa=FRHDK}A92k~hH(yVjEkzOYOdZ&xRm+;3OWEu$v_ z>`6DK3UjZOp-K@I9A&4Iu|~NzRJvmg%b+Hrlfg?zat~h`Y@@t3YmCN&AwQa}kNS`))6+-U4<9 z;jFsV1{^%mIp*KAjTWMwN8oA1&`27)@)7%#bXJ9$B2rN8sVyfNXIV>2${bew-tJ5K z~*Y&dRUH5)!KMhx(7 zq&dGL0hKsd*#UmOei>{6M_+M@Ot#F4#>Dbn_Y`wfV12IpDZpp`(;alukk;=RLp_y9 z*$Z%US#KLwIx*hrLj+uvDrfM^0aIyOI;ffbIE%nwJrrlV1yPSQ~5v7~Gef zuU-eQIknY)Ntdtc@Z1_};n3B&R{17w>0%$$nc@n-c z^D`Ax`fcemJ@Hi--efqAC#1 z-<@8=Q*@ikd%Uc&eY$1^Vd3AsDXv8FAb}S9caH{59z1SAI5C!*w26Iz95uSnE1~)9 zIIq9f)(a6)m>n2aJ%7V&LJwj?$o&gk*WMB+at%(wO1`W z;gQ#33OP+X1z~BkQCSk(+9|$GAvy%D@ZFDs4o_&ha5K|3=bEj`?v`AD)@o*#J2VG* zLn{tClzAl*NViKZkQP4y5i0LnJwYW*uxZW0^{W@ri$AA`l|Wk2g_~=7>zB9T<@LZH zsZTSH`O^u+TOD|S9W37muIWcfw&w3hRt5&No{O_8)p`!{Zi`EI@pP$m z)>WQ0&H!CrE4)853_K3}st$5F%^a>|GbsxVfLga>{8!1+nZ?ZUq{#4!aAn58ulH)= zM`ATUlbUx&o!({fyDgem%EEQWs@3B3Gs8HhPj+riQ>A6!>MQi=v+@ztM~-zL@ca-b zpor4tt(zOvGfkLui4R^Y>(C8gO$hh@yv6+AF3o!vW(pkhV1 zdfBQaW;$ut*|s65^q?GBBVmxZ@2Y+?`J_Q}HmVt>L-sq5?Utu>a1idh zalZls9?au*RFAgt2M&7ezZtGys8-LRtK<|$pKNQMhK~*3?0;Nms7l~o@E@pI?Olkn zQrQ)F>G6#~^Qedb{d5Y4^KT&^Nz<46NcB`^=QN4$z2DkYWzDE7K9>9I1<05KSn+th z*g{MabCEByS(R(buFiJHe2PBz(p-Z-=#tfDiWrZzDabnbf>po`oc#A+2>!N);eU6x zM4Z;hu%G&)oGdX;n-l#LVxSyWWb-uVqYmR+SZ0G|a&!>5M%B>8F-}#rhZ|_$LlZk5 z153v@iHpmA7BM6*cawqRnd9o);gPmmaY~7twPROopyxBK@7MUe!Si{8(D(V8>FnkN zNly5|T_M7^F@dcOOP|#(tDaVQeHWQ7WP-Yq75AsU>ejA51Ugr8o@l~)S1Tgd-Ezg@PMi5 zwy}m0kbCpg7WsT=rg55ICgU}%b&pNrF#0}o8pR{>1EOLic61s_beS^#@VYn;T4y&h z6IF>=mc^Bi+E9U{ggu>Z*pDI{Tg%f0Tu&z2hvk=7^)t%F(l!i(8rVJ+%VL~pmvAxh zqPd=BwNl(dQ(qW#?0~En^3eN5>*GEx{Y}@!ijQhP_b|}A?}Nr_cHj(Iy7H`k#QL~r z;C{;NL!~C~gG|jJ5ixaN1X!MZ47Rv5J8)5a|hs=6KQc$TBk$Is)aIwDnDW7+k z2rY%vsvl=lQYSZ|HF6QQ^e{}<_;BL~`>XVAl_JA*_=)qZe8!@GUEL2FZVz0z_k%`y zo{{*vTS#%S?50XknNpVQs3@+F$}{0cYIzvJTF%@2vnA9^Ke3){Z*9IaZHQ{ku? znF{nzeanMnYqZbjk-I^as`nmgeN>7tug@fkC{F%VZ-{`s+IvK|qQSxAV8a3jg_uOd z(j^K%HA`BsJ$~l0Jzn&Q1vWuKIDTytY>KL!GbhAzKvgIA3jVSYaf5O&p*7=9HpRQ%sJSU5Q}_&j5`MQbULT0P+#e<79I#hJn-~~SiRoYT0UEW~SXxQ~+_HwG!So*lR zCKKHTY@ixdygq^Gu3uRCzefpPYeTtm@YLXQQX zdT*2DM5xqI5dMU;f}*YtH0kP6PL%iucWF2md^)1ATHP}mnlc7q#v<=28aNAcey zuMX!wc<^ry!74%5vJo~DQXk>Zg{KSOCmJ^F4yeI>#UmyehmN{%6tl%$$q3(W^S@aRT^`_aRnTj$Qbj(ncw$npcWAw=O{u*)CSv985Q}m(E6Sg+)R(^WQi$I&%5h+b%rMpZA#QJ zS%BPIs9wx_VkCJ1ooHWx^{0cu;1<78=08jXcnW_yxjZ0BYgr}egQNby5;S4a;tAZ_ zk)thlXM=VCoQ?4TO}SAnMVxkY0^LP)&ST~#L0a#tCi91~%mCh*L?-_IadVcW{U+it zfi4PPpNmSeee+ZnzmvAoyuKSqi_-Sk^K~d&=f(a?;grBGJ$}tjO!8V~vj8M^9W~S3 zFg3*Q?EZ?)ob-FdAkQ#_&e_Y|8X2lu4H!X# zNx!+C;ZfU40V_S4%F=Cdr@_|#D==%{@r7cShVN0h64D$2WO-KDrI{mvL37&JG6McS z$yr5E*?>F{e3hl{;iDl9XE}0?ruA{Y#3r1CR3UjG6#}-<e@2Q-pNtl{6`&_>XEGIQ0zlSZ4*DE})qRhscRbyoyKDjhi#ydF4ZfrAhARMS ziaAJwJz0HOyhe~kut2w|FPXOgZS#B8nn!=l-3Zk9GV|VDrkxo!g5 z9^Nd`f^`ii+f2lST6@#gkUlSQBdxxfyjLmjP&py1onU&_I~d2N>*>_Ye<*XAG7Mbz zhlJP;e?qHh(8GOrW$7B5FNwgyznm~YVBPk?kHG3dPYIzdbWXFKudU^9f1J&6YuIq} zcW>vXVr!-FZLnG{1AK^P!l-Hp5+k$=L2Z8nEAuKxkTjIR!Au)c83HE zHyhG#m~^dL)y5j*9gH(-C#k$^4a0C=S31m@l3ukFE$b|okEhB{n<2&ffV{^h%S66m zM=i(Gc{p%?faK46zbpYwLbR}zYv5~+faXloU5E;UkA%-cg`OD&;oQ5 zG5CAv7o>|UK=DZ0^O1*id5OB?_{NzckUz2IV&H;)VM41zxS{ek6~VO89hIdh8boCh=WSv5 zl0nn!766sNL>Z2do$?V~yOvrB+b?lcyP9CPElh`lLO~kqE=eK04^|0gu9CQsQWl_sqQEh?Hvn zuJ0yK!Zip2TgB^h)k^y53=c5vtc-acEi^40iw*d{eLCjjBBzp6?R#bIwt)dg>`K5> z+BI!q@I~I``i)VU_}nbnq6HVLw*%{>l8Ia8DGr>}lSjz9z~U*q$A@e0d+Rf~*z9=C zFamiUt%W}*Yl?fhiRg0Pqb-~i>3E8K;pC?*86G|+_;8(QcKa}&Rka^GMD7OzF7b8+ z`4eOFvih*G)=uuqtRiV~7eg+jn^x$WoL0Aed3oB$s@>^duDOod#Kbf%F2%e1ykQYJ zPGF0Ap=?;DF3eHTy@`wnTtDnGR;VHA)PQa)9@)4YAwHz4(h!+Y^Z=gFb7HVm?df=Y z;iV)k(VxIj?J!_YE_?HvSm`ElunB!UiQQA~=?0fe)#jm4)p?Z=6YAf!bEu++7VA4O zapD?ov*s%kjsQzHvRuTD;q$>z6J+DnHLS4UT>2mL07D@>*D4ySCuUtASYV8zZ2f~P zywkeH_Z^X@Yh=hngGbgZ(>~k#7${nQS9?ZLsFGdflx=e`heZVOf$hMa2Z6W=3GW+3 zSdO*C&`c%bS>*zSF)?4g&I;TQ*x{dLLR;GeT z^8S9fTgU$5v#Iz*%sTTjmw>Kc9ra1KVyx&+dG%EF0iIK{_$H)ivyD%-Os{;!d4u_vc}w)`sk7`@VjFbUUj3M%i0l!{sK2E-3HqWz$F-@38iU7~6UbNHp;?w?1P> z;^)A66}j^tCj>o_g(!Kv@}z<`;9xnL3S8@iyM16?oJN=?^S6X?)0cZl#uV`^a!2TZ zNy9)DuT?}F_uCRBx52`PzADIB)3TK4^T7E)UVoKzHnD(gVY>Cr0fl^|&I)0wpg=xb zC7T$`3`5o-)|5zR-Ki~3h8WN;^ zDp{nO?*+t+l*7i7;g+ee9R$sFZf!W#vFN*WQ41MgG@-w$YF3N-{mQ!1q{%Vu!7&Ve z0acIA{lz~k#;rvhFA6FSbnz@~TDK|dKuV|P+gZV-Y)cvq-6qA9pB zCj6`!Pp*mzowXGo?I9Fw;?Z}{l-{G0d(YKGAZs2&Hv#=`w9&(duT#@%V{bEs^ezYk zrVOf?I0>!7#_CrlF+{Fm*Zuw3HaWn}nt(ym79`h1$h&Zv0}hhqwUQ%DWI^|Yj74 ziORRD7xY(&M*D%jUV7X|r%KpnM(r09cuq5yp@)WL#y+e~DO1s0*Tb!d6^Y>yaL}hy zI-V;60?9_8ey{p)ukhca6$RO^L_XV-O(FWeF9+yO4}m=xlJlmKV6#?<3hAG3M$Bd# zzI4>ccSPAw!(eijW5i)t83d}Y`zT_(QgPrY>Xgjh|7ZZD&-^Dje-aUWk31&G9p@+!tSL)8#1^q$X$d=~0an+wY7@u!rT-+OQ z+F*xyWZk`5%quX3jgGWSQ^!-8JyoiIf$b^(+G?FX#mhYDXLQuO04`d@BV+)^Oy`sf zD{PBMQx@(L)|_GRiS=400aC^0Ld&GU_|v!H0AsQPJYO66OkMhFGrf#?&936FF|b^Z z>;ZVI{Ng-6P)g0I2IEp?Eq8vV-ONT8GV~V!tqyrlkCyl3WZLew$z7d9n5Vu<{x!hN zKV3LkS)cJ~p?^``a1jiwlH`c1-(f7*?R|)zLT~Sn)j*u=+JGx_Tt=0Q10tEQv^=Vg z#HTEkR3%~J(>}m!+(8gbtT+n*YMrQDy^SfP5ds%f2DF}f*YnsXo3O8(48QG3ias?@osAYuj&v2uKYmuw9AYAp`D0uY!ER6`|#5VyWxeY+a?pSXvj?4i-PT}^y4| zkwoEu~WnwcwGF;q9dy^sD9?&v~X~RTFSOJStUe4*VBz{9{=62~y=#S#y0a zytHD<_<^V1c8lV9=o0%rML-lVphKg+KcUh5J!>wVY!)Vd3MZyxz4Uu3NS2vJkvgjLehQir4OS4vHY!Fw4Fm@jHG)bH5O-#BP*00mR%JxI*f!C+CCr*_8%vHa{rX_m`s z#*a*{OnY(x7fi-?o6i-+Cq+ajUX-Z2&R%Iwo=XwC`@pvylPbx!4J|E_eJd6LLgt9C zAAonoj`ly!aOe&+!6-$eY>#)?MU0Jpq{lg7CThkc44(8Neq`oqk{FPXO3}VKDAJLi zBH>{Mu`^}$y=qaTr==br!H5i!NbetAI@*%8$_?9i6HgAhd{!!!L!0{g)hB zvcojfN>xLNO5fJWw^t z&Gp*CiTe;s0lT!f7%9y9B6#QJ(ITGPbUz&uf753h)V7)NYE!3FcM{fJ0HfB%Bqy!+ zcG5f{i#K3djnsTgJ796v_3g=Y6-dy(psZ9QU)18 z{}I)>*;H8eM1COhpmHqNuF0Bc-=nZQ`eR5Vs?tY!YxF7Oq89OZZaq~PWUo}7LaDhW z?c1QcIrz!Mu-EERJ+|*(EOwu`NE8{B$~{y}YL)*eK}@1M+*9d@j$r$icz)Ol)Fh*y zD2TA{swo{op!-J5FgkaX~{yTZ@% zTLL#I>$GYViRE;hu3-Wu$supNANueC+C$4V1*ij!%eS)YvQ@JSUB}uapT=6R6E@fz(Y6m z#1450r&cU|dg6h#4uw`B@WK<5hU|0oKY?V%KeH4* zvoaf(3%(~iic8y+k-hd?wxiISDS3;t-NTubeMl--N5f&!ynnn{h~)=d7U-i zu_Fd}f^POWfw=J4Yza1C+Pj(Upb`CDhAK#u1SkmG&Il}nsGdG+cs!D;_lJrZ0gf*} z@C8L^PFOtm@lybbd!1HD@SffOqe-d6clNLJ6)JN#j<__wDZ;$8=lKRds-w&j1F5}p z#0ftS@lCG$NwjaecQBmw;$JxAHki1vyABB_!(<#VF`9og5O8iB-T_z=onH{q73}i9 zpy2e-i{fwy4iEqXk7xpn&nIm;UKgWHIDG#^B8nF6;z8({z=Q#E#&H*+4+!WjFY8zR zLE6ojK)x>5@Gg#@;O_nE%SjsKh3D9d+|8Lbu5rG*k7$r$o@*sZ7^+fW?Tha0h3|MV zi}{3l<1Lygp8pO0OnqR3z5T%4_qLY?Z)4K8IHxQ&yqiRWDm@$KY?gu?3{4m?E5WJk zr4@|1=-zr6dDK-g-O~SnE*x-%fNHK0- zmdKK?uB3aAES12afgM2qSKkaTJsIWGX`kP}n08xkuhC}Ztp05w7wCix*rb>LRB4yh zSTTZZxE0Ji?Gt{~{1cE;sjWrC0I>9t<}lm_%bJmmeIrLkqUy%;yok@ct}M0MogLWh z5_%e3n-!|~OWj@mibKtw+!}6P>!gma3;qiU2=cRTz+D4sobe)^qWYOU?(9WgP;|R- zEhg69R(B$vD0}q#YFo4T(?qGJ;oLoZ;1tmYF%Vkr*thcKeTg|2B%?lo-6m2$^V+3dwrzL2bC~3l;j(jwAYhKd z)99SWJ9&J5&i<=h1QBNnPh10l7VU}34eARiDw`?2ZKoVb17s`jjv8QO9WL3jscJ!P z&0FdrzDK$lo@hQMgaw006@r_rsDINa2&!AGlxPT`6`Clc-&ni};So`|rK-z*V}nnb zRFh4gIH+fG)_(rGW_1j_b8Fqk=4>TGcTEN#1W#DUcJd;}FIBg7`(nhxE{2^6~S2DpgC&(OkpdfEwBE(I!Vw`U5y{!uN2OKkp!Fq#GAcLtR%&$ zr@TcVL?@tnquF`~>v78R%4IE|nK=RqC5$o`Iq4~zp1=|+s#hPJ1tVnO)1}}vX0}n0 zs&Gyp)5PR#G?R$$@sJ(;9X^&yAeWbi4#X@I`Sh^TZUEj|7Gc&>_1MD6ocf)_MydBv z?xF8nmq~$8rIA|mVDQooY`i$W~zO!knV{9!0w;A7iSzWB1!oKPsfe2i!utOW0Wuv-) z{>}RY5;KsGBjqB~Iv9m-p(QQD?BeQ_NTKji2J!PCmmazD!bvMWhG z_Ibjox)rxqCKUZX;k(U0kFMLR z70X()a{~OPG8D+_)-_Q>{07L<37fMvBH}g9hL4 z{q1{+Qeza*5Cc3so0-cpb(vG*2%Dt4jwDNTZ)Y3P zdUoPLWV$*gjSBbEw+w8+A?*dXlcHSSlJ2x&Juen7hC>$|{f>=5CsPzFV zH`;t@b~6MGyE;mEPe~XY&rqH+cU134wtE?k?pXa{@x0}g37@n?ylTJ%5S1(T5_IoQRC?CZ|GK!H4lx(?2M+|08%++U% zh|s+H=v5-h<)Yu>>?7)rOsgwzXH!aUHN~g+!6WXwUmMKZqRjE&8l#rAgGW6=UJNJQ zZ^el7Jx`sLzfFqfImv|B^(b2&YawW+|AKE04P_dstX6tHV?LEe08Uw)o-2|pz9S#1 zmuI3T>*Heu3Qvnw5~eS)0c~ufO~^J_fqPyW1AmLv3k*j<7RtLy=oH-Hq}gXXx!#$^E_>glD!x7+!h0Jccuv4M`-JNl2ES)hmoG9EnCu4A zMp@4sP@9AwW(Giiuiln#x$E0g71>jj$hIIJ7LfOeWM^!cCCnW0QLDJ{{5KzfygUbV z%v7pZ>G|{W(Y`stL7+?{+caeDM^KlnBQB1_m^Sr96W?%qD5;^S?_q(m!PR?Cqw9^vYZ)+FaDhKrahV*CVW $v+SmSk%&vPpl+B=`J?|ZXPgwB#x{#xSB$=prtJ*=z@%secAtv#!d)Bx3oenu>^GE5B5?+%bF`+n>E)oON zsZiHlgk1XM_Ke#N@=wOU{f<9%=4@>%$cgTspsZXzY~7W?lK`RdgoA1@bA8b+JeH)* z6{zI$o%?2IU{dutCt3_FHu0As0x{euzvWufXS94!JTcAKlG{tJc}!-(sAMEIynbUO z+u=U#o|9*XV`1AL60I{NVm|6xb!n-oJ=))*q(J!*nA4?097^){^p(N`=lMq2*|>D; za1jE@X>zu_s%IY9RnVxP7N|hJH^QoE!7ttnelQW^G?mG~`6;-TznraYs}G>Fq5hYhCgX_@cM(Y*68@5zy|{_W!Z>mQisv>(*!>kU-Eta0n#02M_M< z!KKmQ?ry5`;6$$5-rA{eWX9ljD<_CQam2Z+#;$o z>Sg#h8W;e?6p&u3W{?!txU5@_Bnd0e950$J67jfV4&^VNlOVT`gd zYEM@Ugf8T)U+-T8Z4NUs@=%xi@+fYm^IM zvv*t~KU$5;P0l;{c-A$rad=pTq3ws-? zE=wxXb$xmb@&3cHhFNg#>T-}JgeMoJpf=t(HpqMNPc48a?boo9iX;lJ&wf>vJ>uQ|<%V4TNc%)i+``mqWTaJn#${1? zdmx6p7O#aV(v&JH=x^l$Y>kj+RT`DwmD(m)vpDrJ&(&Gw2V%`QtAWYXBA>xVy%#8_ z{dClod(ex832AW5Nk#nIE%T4cDhUQ$9>Yx=Tx~ozXlylpiFMHEWrHJLCWBUx8XS6v zGRzlHl)DLoPAh^lh({~p+uwHZA4DhHB>1K^1{)U|?8f1MhJA@{1JS0UL~-ET-b?xF zsfjE_NRQ)v0VM<^#r%P6fwmkBimO>wk)zoUfTRf5z_0^aTLk!z#D4zbk$r7RHXj>V zTYeT!(x~+S9Kk(AfMDr<6-)c~W#hl1WBsUb01<^Pj_*f*pIZ8BUj@bi;f1*KBWksP z|2!Z6WPd3#0I#RthGgmYRQv}()c@_EUk)YEW^YNU6yZ;sz`$i`1G5~vBG3RViN6cB z{`#q;JkVymF(24Bf7%2=r2biZDOIbVBk`a2Ejk@&^Ir_oKbryIUx(NOYf#bq`5xb& z_l=AfXd?+tw-^26;gFzj0Yu-d`xoP0fB1HQzDcI`ZhigZ;RulQ!y`D0-k{{% z{dwQMpTi+IlQ(nVZ6)!&mP{@Qbr|J^|UyMcZitpD9WB)$Uw z|K30$vwAZ>C-S$h`(o_Jo6%OGtSQecK%aDPyRG@&V~NDt+K^~AJ3H}O=2;Q1(MmI9v!m#$=6sA zF6Z@td#B?cPHf3P8fm=TB~8X1e#%BO0Cge6R;f#ozwPJ09N1*^wnvMcoK|8lc=W{( zdED>oScLR02b0K!8ypo~4qszJ?1LLvN8`G4wh$adm(E0j|JVT5L16JuZ+pMt7r-!D zjo}`B4be+LZLNcrwbF^u`#4{ysA){x|3iS`pQrkNm=`kT)zPtR`%$>2>aF2)-ohZG zZvg3<_$BHuASf=*#Hj*#ISE0Z-~G8+e+J73>qE@-S>;G7fg*f!FnP)ZL^h&*`Cp6S zs`V*;voBh5Ih7O7n<+X2`}Xf$%D)UlAQo}A{9EheNpLE+kB+_30EOA7{D`l#b&&Q_ zA5`M_&fW;gJY+mF={y-lKQYQbo>Ku5cepI3Y{xl&(B0#$wA|BOnYQ7$tywl13Jz~1 z8SmdG75?ST#~3TOKL9_#x{^GC%w@xF0y+tokB^Ni+jLKCGarx*m3f#QG5*va@!Q21 zUBkUPT4C&qA&;?A7fT6>mo0yPiex?J|iWRi5Z}A)cGF^XlKmXk(!cqG@K(A~Ry7YwJ?SJR&zkbS43b>}Ow_*h9JdBY;f3UhsKlq*Hqy4&>c=~nY zgEnweMJY0zhRx}7^h#;f{&k%YTr@e3WV&W_t^{S@C!^@kKdTa-Ds|rB5PK!yS#grn zIE${Zh(-Rv7rmEVg-~<4tmB-mv>*wVN*oYB2tMJlj{aEy;iUYgfPk>*rz&3``c~U8Z5fNY8K)ydUziTR>Eg$7M*Gb z+tpFPl?~k9w8XN~lj}|TgB4h306&~@D?R*n{=NOO)c#jrv5qKx|C30`?`W;c6o|=RV z6M-m!y37dU(&WwA4!K6<=Y*pGoHZHAHGBqGq~LEtex^S0+09Ri*S|)$q)BX{3bd8p92#Qs%7M=2Q zp&j-&Ht&(|II>vVX-(6n#Hs4@NIcsl2(x*K*yo!xjH#j6A=KoK~KrWPOXPB9Kx>Y>sR_yjz;9 zZf|yx+@JiJb|cLhg&U^#z~i(ZOQ(72SECUpg_Wb%e&6m~FyBb3 zUKQ0`7DPEeUYW+zC2jD?{K%-cJKEHI zJ;Sm-A0SC;dkNumGg&j}eXmxU`nyIt>`8^W0EWieu%}>5Kh>D4&`P&7xj{4#XYR&1 zf0$&QVK$eHjPz>qzh7!}IxjY2PxQSyW0}2dcx(~7>N#L(^BM#|qWu{@4_5T!J0ckg zUDaN#P!4|GmYAWG)|gK52q{2iXAtU{n`0VvfbpOaljuzz=!KIw(!sJ_rLX|7(&CEB zhTnh(UY&bHJPwamU6J&q-k2wGnH9gG`p%eh_E>FEYq@wXE3x3tF}`ATUJqW#D{F~L z<3Ei5&E#*~p9bSl*3U>>7b6E>pz&KL*wk9qcr(;P4Wk?NwbB2_QtQ%Mh4`V^KdyA1 zEHlocYSnh#r@8ix=lWX>-@3lq<-cF7^KK^OwHGN|a*WB5jG;&u*E^HIzBcPcy*UXc z&e06VkLHw)mpCe%s{Qpg5_&h(~Qx$U!m=a#R6P3; z-G8XYSXb9P{DX!gO`b;&@b?{e7dyGSa5752H@Vq!2U{$G!|0xDUiRJ$h67@|#hr zM5$-L%VXQkOr1$+>~^>3W2YsIsa{-|Fb@54&sx%6;QC4%~lTr<}d9W-MFvI6ZxULe9N)si5D9 zVS3Cryt1z~e4`_a>I8Qh=)ej!nV@wa>x;?HkqG@7;TMFj35R2zg3MMN{ltoVNfiV zCsqYbirSxN!~=YK?z7M|gmaJKY%4CVr*h%WL3%Bw7r|)o1^hRIbRC3ZH`t07A?*YspP zn$kw6bwPBZgK?%Lrh-&V9b`RPSHW}ri*#)A;p3e<5ES4omSv=V@X;-m751ZF0Eau= z<&uOI7JR5zflP!ZBCIa^+}-3k5%|Z3pPmja&g41 zu9LyCg1>FMG{42?S19n~R<=^ zgAPqv!{zxN2f!sqRT{HuUHnZw(B*n-KCdpuM?9a9kMBNN9u3a)zmqD$9`Zy>^{uaK z2|=Ez@?-+g+*1u&s6>30UV@Jz3B86OuaDm8U0l25I=1Uwl1kmSzPoo)S`Ch7CHtJ3 zB?rDg<*qxyhpKvNGgu7D3OY&5I$yD9v#R8f&RJ&WsMSt`?%-A%uh-D)i1m*~888R$B}5$Zp(N<^hweY~I4 z(=D9*2@BPmw!Srp%aOwJj=165HFaoNM-?cn21hX1tP>CS+K zcl?|3+ZGpSS(eIX^&S)!#8nfQpAJb~ePuZ9SK7w^g6Ax?>9%#I`$}GKd^ZS~2XWmN<48zneVWi4!eH%#VCkH5mH-4xlIJe86gKSw60) z;vv*0-O+o^^BOTDZmpbR7ONc%4iK^@t zZI{^0evO#(C70i0nrRrjRn8Ib72G`C`ixhLbBPSDSEla(FABcISv);y%Sfi<@q$fD z)Y=HohGn782%p=w%vt22;Zs^}^Oy&;#iB)vTTiUL@BIZndo_+$*{w^hgUJXuuZ&Sq zh(@hszv|I`2yXw(4+!W=4;QrbyE|q;l%3+?4cbfJuH2N2%icwQS%*CdlI}uo(^k5Mf!_m$=k#`m2k zbh?9jYI~<_#r?LoRWd$#Hx{7}ZtKv*1bdCE^arD5Eu<19zT02^;~RZ)d{>}O5dil@ zDI6SG4J(L$t=V|%m9axTJF=)ZgSm@`2!_U0FB3s3eVl(Q>0q}wtvw;YRk@zznlT$p zx@Kem%;RLYjqVx&{t{)MKw#`O|fohyo* z0R~6C{Z9No#|p$zxQPF7p;n&V=ORueRZxx9<6NJ#)oT)ktwBu1%@R zz-?m()vog>Lko}o7tbd1xXoWlN{_c^@?-BP7n~qewaa$msds!>QvAJf8X@SlC zD>?2%4bzO=7Y9>uqSAcjV9Hsnl*2>#uw1PxzUi+&V|x0RT)H2^f~*~XbM27>wpeN`Bs`s!a zzdqz_ylyQ}pp-!ZG6IyOzje7ODuMy}2e9_`$&0lgOW)AqUneL%@Dp+@xz}FY0E$$5 zjI&rqbZW{jOBope09B>ZlC4N6gHN7Op3n&1zi` zz?MvwPW?<`lmLnuz9WqP@Ks zkfqbxgt;5Zw_W&lNKO(-X19oGfpvxGeACK7_dk*L)_ELrkD`g-&xR zph4RN?pc-0Pvm=ML1$;+W{=BXEXFr>XD->Ai2HFfjLMr=*1h*zuSb2JJ=Oi>abE92 z99tvjNZAK9Iv>SiR5kpQLyM9cgX=tuCO|Ost4nRAg=0}ej}X19#@x9#y7QA3-3INwnj4K&6SG` zRIpC1i_ZLnUiduTB(OoUt<{zYNoDb$$gj_(k1q&Xqtm!uQqadVSgK{H4>YTz6@M_+ zRcEw5g5{Zw2dOkAMN?FP(9H8BZ(tyS9zR<4(z8Z%qIXK}NYpj#p( ze2c77@@Xu8U*X4D(14IXt7hD0g06QA9YADCzgyp)o7w5toj4p&DKsm==WwrE1n>{7 zL-!05KU8n}Yt^yF_NVJXdGc?eF5DiO9w7sLjN#ZQ5v~gWT%uTi^1Qj=9cD>`Qub|l zG_{f#G!Nk6>54RDs6neBko%RcOCtWhczUfZ01>zCrLGM`Pt=+f5?X+U0MuueqS5)V zVmWwT7bWAC;H%0?B*9Qs*5Qd9FCBl4>3mLR4IQAu_d77N;F`>!Y_Jnb7rfi)n^H50 z2E<}CcP5{NkEdo<`r}Jh!cfgrd(K7uqY%w*9oGDeOG-dbSN&x54`8H`qfNsc;watHM(}qh{ANZ?HiZ<` zk{3#)^xD5Pi>-CN>-%06zoEA;xqz70vDtr8>LflA? ze`U?AxmWk9`vu^S#!nj%!73J=nE{|O*k;D1kw#U}IP=)NBUSTLl?#O}mx>Lc4_&5@ zk$jkf#3yI!EIQF#A?LY)WOC(Kb650Q(~vt1t5>Bco;}UbIRHxGa0FbncfY(-tda^yqwQS{~YY*f(7(d5=E z<`p|50T|A_Iw!C9CaS^?rdatpO3iEV`;u!8F1ROyp+iV|dk-JX`eiO2!OB-fZdDY4 zsTr*c_x_OCdg;ibU{;(Vd&;{BZ_vTE#%cQb`u2!6i_z^v{2I^XN2zjpay%`0T zgqqb}7B%XslLdgTI|pi4ycmxWa3-|fa_4a#oMd?-Lt#jNpjdC7s7U|5mcZ zdF_>vM5FlDTsDsp5wE*4v(a!d=Tb78MSDKYJNXLO-bVs@K&(X*X{R-14_ ze!k?LLTKzu@CG1h{Dsq@+J*5$nNuIPTEv1)g02g6`ifO!D{zZZT|tWXzMQXCWB=KO zuI0i~H3$rugV&-k@G{!DJUpF<IMs)?slga{0bg0R^x9h*Pi3BG`dO)Ort#?Kr1F49zvVKVu6e!SJ(v5#(`j{7H@eEz`+IRnO{F9owr{-a?38kUEWt zo`VMuUC4E^j$`dWkI+SVyO*}zS0a<)jn1~%HJ56fyjJxSKc0mjN!Yx~t~teene!tb zG83pkiuomKwRzYYdN7#Vl#q@A%qHX^t>cJvxWmed4@h>2^n>LaLvCd1dVUoaSKgJj zV;=5dv`Xv|=ziyl3!pgqvocy$Oq*IW%Rms-Mt4No1zQEmMxK*yz#Ba_vw5+2pVK6{ zv%~q3=M)R1`=X|c*VN2TJ%op21CwatA*4mB-a@w*-n_d)p5=iOn@Bs$myidL&f#Jw z)))yf@`?`bXl)8MY&RL?_CBoa^Ktpumq5=l{nypvwQ5^yFueiZnu-l~Y?NA=4~r3HyHXW#=#2^{TQkw`T@KgDB`?1pdA@m){UI+u*S; z2x=8N{0ejhWbrL4c)U($e=v4#yY;8j7ZvN&4}K1HrYqwSw^{;&3YAOAvV+l(LpUyM zZndB99UIUt5x1yiDH)=hoI|E1@j1S^i|PyiY>!D9s&E z6~UzHmY!9q%Je>Z2dh)HzfmqK=B+)tTz0ifsRg!_z z=6k2-eMZa32jzq|HHv7q&Pw8hv-iZOoYY(+G^%=EW}bh7232j-A-h65wdXX$hYsgi zBiU7TjxGbK&<#5$&8lhyMz@^|w`JxhG(Ws+#q;^#oz=WRbEI3y=rOy2w4JE4pN66H z@qrElN}k!Hc>WNTO^L9rvv7PjE^&O|6?LRDFz^#$`4(U5Um&X{Sz` z%;i$b*Xx5Gj-t2@_VT%33tEGNWuL6Kt1>=qCDA<%Mz@?S*Vj}wGU>uWu5i*jn=bK6 zvzXm$-VvUBuu|x#vxtic4V}mzVC>gY<8VUzfB-I}QUzZqFVB>N^gk5oG=k}>kET-c zJXTtwIaXS^le=@5WEV2g;=DFr?`6(!bcJFb<+f$7W6DcBP{=N}P6QXSh`5YQ7XLg; zsjSE~E%&Za2%THjYEnOj*IX!7I&hIq=17Rn&GHyxHtNZ1ah_XCr<^{-tx#3WmNRUQ zuhAQhwY3r75!BfX4WpuiAaygn3j|fbCE)t? zrMhe?Jon)xiiaLGohF>V^6&m^i0&f$Nr-s?(oI=5+8fcC|eD4^l2h(S9Vzf#>ayO@w_z2m!s{XJxRAI9RpI+177?433!>Hy(@3hZ0 z)!oc%Q+V2S?5~3>?3+2`a(b=AR_*czGoJU(Dq7Vo*Z%|SQp!+P|jgEb)JWRc~$%= zg;AK;>e@@Bxnu8_&tnpF4@-5ib@pae!%}3tY=N?W#a;0_i_RM)8XEzAo29$~LG@(T zpKlu(~GuaR{^I3CO_tSz015bDiRZx@UH<*M)7%yELD7<)tbRl6*8B&508b$y-8< zB!0Km&-X^5Ntc)@)OawNvCJv_cI$;cx^SM-a^@%>`gPkXr*(w6IFBw^wD8H`o9_&N zh^uY*pgBZjMdP9S)0O=;pC+pnSy3zpYfj)b(sJ4*6{8DS!fboeZmfx;0sba+wu1qI z+M!_QcrIwj$(*0jBP3wvgDX_{G2&47`{~ywO!=Adn{^rpvrg*=@eT*9CY6WN0c&zk zzOsS9onja4;cx6u1Y(&1AlK61NJ2{Ssah=(k>v)bU|MuvoYBZwQ~WFy1gX`~mm6cX zDeR4B1lxOa6O=DPN)@TkcmgrVNI)Nf9(=ii@)vZMTJO^eRkN)1eyh#?hUV4rY4+)2gHPWt)1!o%b3@-}ePVSkgOW`@)T&-;dHn?N5qAab|Xlw|>O^Xib@ z@jAsFwLLLjaR}}BeHguYcsBs^Hwth_WwD7$e%`p3Guwh;NP3i8FmHh4V4_lEbVOC8 z)879L>LF)<>8zpv9Y)8#8H#HM7c`|FB4s-8vRQWyU?V?VKdyIgmb){uL6vdUR&;r| z=P%EV?~7G0s4uo+(kgE}o;ZzfH<`w^H76dUTp3RY#1Bp0&U8WvT(5#~F`>_Aw=Y6_ z-^Q(F_bAR9;RlFgkih)+r>-AS`{QlN#~s*K5^Q2J^wB0>M%$}ru3=x=V-jXCI)0n@ z(kWdp^6VfnFa2trjh;_?d-+TI55+w67tlGSG9N+hSQa&g=AKH!J7-QO=6$66NR>h| zl#EEt=d5@>a}~lQvwoe3xLpc!4x8V3`wicKVd?I6vw=bf@j=}HHN;&Yv}zR|DW1V= zaz(dSHWZ_}ZDQ&`iWMGefNsB(&XWK6SEAy~&8pvXF8R{M`rKd)tu-5y9+Gbaqm8vX zl)@i*zPD@5Y|Ex>&gL=_#WN_0Y1UYnR0cQ&d`DgQAha3eUhA<(?iH>o^~OnN3M;)W zDB>-V!t^F0W_54*((z#Kv`@ub#4@!HorFG(OD#ZDAYIl{?lU3FvF8wy@y(^U4&3-! zDy~dj5nRyn+`4wzIpv5p7>#Irrrk$2R^Vu?H@oExe&O&|?8lLiN13_y01$E0S;Sy6 z`w%_xa^L94X#9uj9a>PMl%G)GBI`${--aFrKH%n^Mj>+a`_5KW_m08%)xN=m6uWl_ z)g7;01QA{+CRB7$9Z4Z`k#|xSXKQ|L%>v&=@)O=`OQ!I;WX4=vdYnh@>j@TRyP&}2Flet3;ke34a(WUJt?)3Is)Izm&yX~Q-P`==w} z{*M4O{oca`m$v$Y^mQ)Ah_IELHpOM!#d}@P`uY0Cw&!w3jbRO2WuV%z@ZL}i;@vur zS8s=e3sS+dAFSEL7y zUZiF384fwCl^rUkh$^Q8XO3`;j~~mN_x85cmb(@|I$$8a$fIlB$j>NoJSaChUmnT#3d!$g*q?mRxo}tXAk7; zG%rK@w%?L=8;pLBy5-s}+MtlJi%sll?jgRrZ-aSpH+~1_!^bK5mtLA~?;SQ0yzYMzn8@|> zoW@2#kL}ekb0Ra3P%gahe-);|gDHvJoE>W<$K6uY3@M=~Xa9!RV<)OFkj< zS^E@mDwB72&%ZO$=p?|;a4Vnz>Ro`+ZMya%ap;G{sOcPJo^9|+^C&CwBVDoyeRj&e z|L9wV@pkW&DZYr_WQpf8RF_x~f6H#Dy~^*l#;qbI8|-9XWl*YMKveq}lRxaZF({dH zYjdf6ljaF8y=&5aPU?W-z zsO*r9a;K>%#QW&j+q5>E zSIqz!yke;ExMiu!5@CRx!s*eH-6eNKPczn#-Ey3%krkO#(Mk4U11T;p1^}8!maFEp3oATgG%cL6k-&+iNgoyyouv6 zEhp^+iYsm(Yqy1P?#F>Uxe@H#CC?QfuKsXt@CIn-hSGkw4xfC2Gh+mLlzQV;AJKa< zG?QU>=GXm|6SUM+;8OAl&+)!ojRrn|fSp**Yi{UiW5=e!%+`7qkZPbd{kSgfTnZsv z2G53z*ZV+0`%T(q<-;r0O3t7TqBxvAnp!$8QRS88i6A<_o-~`E z8HR4NoS1TVni*tpjjeCYA5pDT6GeC_w{jWSiE*bK@IAR+n$Do8m^512v|gVc>Y(1@ z^&gGb)t%kBR4rB3mZb0g;C-!DW*P5$SuyZU+6!f*Ji^>l@*DekILA_J?z&B3=Jc_m zPghE7^LogA=op5-95uW4kJ+ZIcwApWIvfm zxv$+|y`XO82|qRO8ZMGh4OYRX-;Q~z2CfianQ6avn-01-*J7}hGoCb`wzc@40|HSw zu78vIXgN-xtije;XEO&~;+!!fpWA_Hub^*`eG@i$;?K;1;PPP$XFyRbrJ9N9!`^+w zWk>btkIa~#R_%^ylkDl5)muYFO}S?JNOgi#@cszM87CW9&eZZM){729^X_3285PG! zcJQadBso0na5-UcNObqfToD-`*?xGQ3A&j(baTeh%&@oLn%4f*^?>_~aGLq_W32&I zUAZ?)|AW%P$sTJzi}W)egx%voR-38zD-E%X4p`zNOr#(k7Nn@m0K#RDsthF?KBdRd z7|ILSmbIcL{zNLT0AF}a0yu+yBr$>1n`cbVPB%1=DW*ds%W22G%AvmcdGj#h9`B~O z<6n@S-}pYN-OcT=z3e<)?m*+}CylPUULuF-Aj)|E<*Qjm;==XKbb~g+yC&H8jdZPi zU+PJq%~uD^$ArSa-VCA#HwY){^Y=zUT3x-H9KvfEw)qrr5D@+GMDl{F=szLFPZJm; zHO7kL6P19or?&JfJfSwN9#y*|VwP}YSe6e#veraRhZyFA6&cbYMPCpnE}{IZuiQvJ zUnc8sdX*j?sJdop`cieqGTDtg)-JaP^hxCNeBNV4e{voAw$~1a=sa&5079p5U13ys z(StwHf|lUJyzA9;wj6AYfBZFk?u@dBaTDzVvpC$qfj1{SvK;B=gyFM@lAgZ_;=@jO zlpztq#Yt+<=YGs$hwpK|K-4F}m9p>e zS`1a&L=CTfA794TnII%tR;^Tw$=?t1U{ZB*14T!s<}-2%eVilG)cWze#z5XjGRM1Q zM(R~O1mo-GXHe~H-d~UJpN%JEJfq&=dwBr9RM&b!sn$evsc$!ehAlEVM2H7Y-`l8%d=cCE!u+EAd3)T=+|`4& z$wY5wn%^O9os*TymfTANGKZ3S(v*WblQ@#?w#c(3Tg<14{t9cIkX)JTd2;{7W9o;4 z`ZbETGp0)PWhvYCuNt{%*CM?l^W;+0zBwM8T{J#qP2qKvy@*`Qny4;L<^J()TYhVD zt&UPDoily*!mn!MpfVvX2}jj3W23PXHzn9uzVN$-gQ1++Y?cmvNi<8&R-L+_043}Ar^y-xdh>8KP$5C;479(EW z5gO^eo+4elp*Bp;DpbbA2kKb6^uNsw{ckE>JS+_w#=*%|B7%>c!X-|P!9qRgsV$s_Tn*G zVC|ci{Q{M>}Ndim}Iki9j~47qEge) zb`rUd2Y;l48ZJ)64n~eCOpiztYk&PnVt}7N%cmxtJZbau&-{u+%h1xnpbaOa<$BL- z2}lk);Po)rcqH`;mYa?X7eGvKd--%u+^X$6h9d1{R|3V^t8_0RQwjeFC91)GM27=h}(qsNdT?R<;Nt1}_iy^#ANShha zQUfAv()-GRu#tLt2y+K^D`}S&IDtkRN2lfc+qcd4`zHn^9!~fn9Wu#o32^$q?iTVi ztm{I1?M~|0OJ{#?LF|?}8dI-jK!0o|-}@%zxrml?8s&N6*0kFb+S#Ig1Bb^bN(S=} zZ-v*XAi0k_=*xJF$`AALE@gpD=%Kv%JUV_hF};=b%cBr`;s0mfc>((lvypFm@zV2! zjV1+QuY!|U6ZHbNvtUjz`}451GCnMYbP#wiA00HmmN!^8Im1Aide+6$&_i}4C6&cs zD`(zO`nkYS-GiM%A*-v>C0Elk_JyqGUZbbQB7$47f!kFCqjnvY=vRI^AT<%!FVD+h z6F~^+>2|c&IW#H4Mi6!L{VpghzP~xbR!Yg+*t~D-<1{G zobP)y`PYH_xfp-)*IQO1&QA^E<}<}K!Y?kJCko(c6=7K)+^$ExKRqP*Txo$f+^<|^ z@i{!FhfFe>?jFTYg%%Q=MrU*xVk~QorgowX^mJ)w>orZIZ^O0r3$}<`=x>f%q+{#k zzb@;l9yWSt{xe7Zs}PKbK@J-;ju!u&lZA4>Zp@IP-?yFg{AN2`S4ymgxPqzX2X#Mp z+@fw%;=Hs+A!3Ac*M_^x<4X~GR>|C6+F#d9osP*86(24NfKm(c?gIV%g>2i17WgM` zc*YoU0=)*+?*!w>;cv^)M=-4D28+41C2};J0*D-&Dhk8{?oi(W@9jZ3qpBxKz5ViA zbm$cf1eNewY1-;qRmXkH=|=i6=^I!iWj+s^MAXgtyCz#<|ES>Fg;7lbgOn{r?psfR z?P1RW28LD71MuOe76TO`>gLbH?o)B-1>6Uqs~zh?`r!){aVn$^A0Q2A~lAX_D-=-O-RD$zUV%n zSmoFZ%k0W`eersWO$LB=$zMsH%km31pW#sFE)@12q;1|%Ts%ZKtW_Mgjwri*5^G^< z8zP2uE%t6i^1jFHMMaAjdAvXSUO>O)NH{xLtV3?vpyy{YoQk!DC@QW$+`0I&NVO`H zDyrhQD$ePU+Q4t2KrX^3YBc?mB)chWI9zC9_I+Lfyl0=}Bxud#!0Lr^q*0>(8FCf0PsvRQ7^MbW%;a`r5&s=qPCs z^((0y?CX6ya0NF7dC%l|P2RQ5Pi2@=P=vdsSDFma?B;t1F~J*3m)HNABz9oz+0=`t zwNX^eGTWPdgg0AZH!rQ4I9%~j_!>u*?=OaYz&KUNXCxlXQ@bNbDzE8 zM@g>kkyDI9-8ZY5J~1uvkY$7D8EToFsF#HF`**{DCUotZA`5pi-kCDD@L z+B>G}B(njKv6gv7Gb`%auVnz^6OmE~2u}2~E z>1Md*z#`Zv%+cqNjp)&HApTQ}&}6c|ZXGpR?`z?pUmQiMk^pvQtA|blLUnb2zBgeH z*;Ra~ANe5|eK1^5gBTm>{L%*Zhy$oy*zbT`Bs1TO=IE_;BnL(ypS8N+&_~y^81GIp zHeF}Mx?ivPrcNxoOUE;4WOZ^m9nu7T6)Txb6e**zZatdMJzz3*UFCW`51>@Cr;TY; z3(3OMeU?8@^wxp7nH#Ot3S+DHaC4)|k!lx8uiE^gx#iz-a4Dxf?+4#@Lt7sx7;DV3 z2MR!C>6#9O+wmqD-*9=8JN2?5252av3~g{c+(KG@$qyD?cDxAFVa-_aE7uqDCc#$KQrdf^~E~=9_KQA?@8h-0JSn%%j zJSK9_fnOg7j=*J;x|s4b=)&S4&>*dShaUl+1eZ!P=SS1sC0!Z~1|7sbN7ln%8{kmh zObRqt75b<{g2Uf3{Q}K`aR`KKe%iDHccw9|e>Do*D&={4#Y}SSJDRQM=V8>>=-?Ih z0*Imx`^nJ(xKxLa_=4m^oo3nQ8Aj z5?hanZVXqv(})a`=O%g&b|zy@ki1D8L5<$o5nreb>M>~619@VTy3NaG^W~cJPPYon zE^3&KC5Q_VcxP6(mgEu8_Z|s~nZ2*{^HsDg%wFSn+{4K@yWcI!&*nD4h0b3pP^K{% z700T)he@1ARYEi#=zD#EB>!MyA@#(! z?2buF>yu=;p8r8ejP()JcyCvO>0?*z;57563IA9a(G=J7^xy@n;gRBAoH$@J_PkpL;`AeB*yQAdE*>gegB*B=?Oh9=<5mChqud18 zQWrq&)J!{H4p{KS#&qgGX7ss=DxD7(sgsyZ;>R#4UkqKIMe<8Al1qFcf1$6Ikf0l{ z>kL8AoYC_ubAmFUcj5Vmn1eA49_rkxuyd1a2`BR98HcoLBPd!bCtOE#&1U2-kLI%d zl?qx*Ji6_!xe2;#%A*BEBr`=m0oAP&GndJ3EN6{%+5m#E>hgmH6oStR8GH?{HvG&p zU3RydeWzo0(1Ra#z?HRJQxxUblWI8VM?Tc!68(dUK09a}191{NST1e{-`|LU3RnU` z%kg1r4hFE(RU;%xAI%zcVuQuWNiCUR(+OBiqMcSe@g2urBaP7O?2_DF5{EeVs?!Da z1{o3i&;W9HlRDDt2NeU@7R{}$;5Q7)B=b(PzVj{eh0`IY2xqz&&Z1xa);3+vF`eha zF1_5Hqr_0MbhKAxU+LsBQ|b4oh0Y()_(er#jU15jAFNxezPFU=b(G>&ZA%kx zdBHDtS#}y2lxlP4>OFdhSUvxl{hk`~^^h0tz`tNJ4%->ec=crkDxA&-*9?CgE<01Ieu_)tOfD@8ta#Z!7}fp7wW1>}KCm$TM%tJYWvGpnSz-28<2jQnT<- zg=2e{`|HqWBmU?H>{d5fE%yLiF9G)EY|r6U<2tm0mI-v5peX_t~aAGsZ!S4|W zUQ)>$UGw&VXNQ@<<+PbA(gtU#KdnS+W{h4nLZ-L=9fU)>8IFPQK{eGuaXE2a>Rl{l z>1-?>br}d;UL;XraB82xV9i$k_T|Z_jmOpghlS^Hl9Q#PNSv;#zB~$*Ig=KPvv7BB z-)(*gC*u4B;8K#XNo5t+neGoqEs9$2KZ&6b0O!mXLQ*hwCY-dDzpK}md>Yr{FQWa1 zGuLC;TW46bWQKV%{_5;S7zJJIp4`o?Y|aQ7?PT;Z!(Jy`{+^sSTj9NJ@1R3%019#{w~6K4en%jZ^TE}}L-iuip!<~t%>NLIQn?j9sKDsucG#XX!2OP;fqwRus0%gnFE3M8>Ik-mz}1vGqT z#!xo82Ech$;#{G-JqLD|SkoX=S$t+4kBjJOy|QUZLp-I&JBg(`S>sJLJ%AqkR>a!a ze9y;d9^d+d*ZDibm~W4>ExBCRjmYQ`M#3jSdalO7r!=()A*GRxeX_nE!0)2Ox6P zxr~H*x#!$}j=H8#9*7iM)_Ny(1Yl3@3eIhZv2-sL7Y^IC5@rudpqe{Ma}n|S3Zkj2 zd&?BisYK6J*}WzxZg>3hFS{jR8~HVAWT?XM^UB!Gwq)wxbsdQ>dz{{fruvxBp+**1 z9Nt~#oUx7GP`Y%Q%BM3O>z&QNj?^1jtY6~1|5$<+?sdAG$DJQAAAYiG|6I!0BR&VT zOcNJdBaeYo+5TW$>wd(=<-QSwp}L^ZV&4y+^2L5vAP8q6zvMkKF*k)+2zGotlhrJ4 zp-i+uOdxACRf^HszP-fRzC%FtTZQb?;Krz^eE1j9NnrRio6`Klr*&j#fv?>IYQ<|G zn;(h9-4MrBG;gQPA8`!;mxc8c`?NEWHQpv6{Uv$cd)wc14b7n^x$G zlaPhVYYqj3IllR}6+bHvJ{(NW; zB%U?nZkb>w(RNF>D=Xne3?MY6);?RcuMO3^e%N{sSV*n!qy4`!)@5Z~Acz7g`Q)5+ z>&A6makpsugQp6o{j1MaS<5Lu;~Z}KB0twfAq(WL#|W~JEZz&%4?5a!GGgU^Wg9QBv@{N^cyn^VOLqP;uXE?d$c3dRQ!$NLz6LwOLKQVw@_Ew$H39`DQf6k+LUe;ZB{yuh!6w+xJ(5w)=HJF|V~N zC-v1sK~vn9Xv$@?A=d}b5a)6_X*aTob@(2fMF$xZfwDd$gvTFMyW8|?i9f*+<^49T zD~Aup@g8BechqRTE<^UrIBOrf^YrR??~rh|JgW}SSJJzWnd~Yyhsq}DGi(J5BHc$d z9!8+8y@bQyA+RG>{mhPmOg*Z|K>p6$NbIsHPkGgGyB~;~QOm{AjqX&b?JeEW;&4eV9Q=N8X3xvO*s}I-0XQi`SFQE4_oD_-YdP%uFu7F!7$IeK6&vi{gxwQ z5U9j}mE$Onxd*C?aN)Fca--bR`XM5{b+>1I5v;@AtQ5NXZso9uy1~s9!Fpq>dQxUr zZ!~eND6Z1pK>j)n>49@4vACo0ZT*VIyY=Z$w=-UxbBr!Pw0L~&@XQWLbvUg_WZSJ~ zC~TxaPvBy>(qc)u_TXrk2tnf7zPW?Dw0p81vcoPcx-q`)_Pui{0V6+P{wpkE2fhyy zT}kD9+9Z4GEMRV&eRA8J`}{UzkD%}X$RO>!v51k`w$y&d@Lgu(rOPBUR;V>D3UT^f z=7BAUrG+((Tzs>1uB4YRa!i)v*cmRla@FUPXGf4lF+MGzRgzrYJFUYwR|$F)mQG#0ICcH4@-9 zVY)k5e_GJvWvz~98tq%E?H3wO5vyU=Q&@y{C z$9N#wrJPyj9HWnh&kmiV;Ty!CRpktd+zS!>H2qa}Q{n@k+C zuFKZ6Ms^S^N*sD<=mb=_%O=%~Uq^IL3tSyuFMGFbN9W1a3ukDK@?)cHbCe1PWz`H! z&C;BH5L@Bjm=zym9{+N~>RbL9UhnR+9WMNuu5{nielr&T{Ze4+jILuuHW+^P^Eb?5 zlI$<~@Y}9mSW4%etKk|+zOX$?y2({W)@|e1$l#UTy!_f>W(3!mX-g+B$}}jK>d_GT zDEAtH=qR^#rGy39FjTnFpAOnVJ7W1|s>{kVjeapK$Mfwub_#XT)cCB=i=G&4EKsq1UPI-StY|VXo~cYkcAp zQe?5!YN)^!kkbU1(LG@Ot4t2M8bi`ob;!j_A-Nd?bi8A5R6+VUZB;;Hdk|XXWr^uGPm3PDT zxrpXFy6D$doX^$UL);=oGrF}BxBNS>;tWgAm&n+Uh|0v+zGa0{(R>+PK@7wy-lC1nP<>k47? zlG9V+5|R~r7yvKn4>Ww1YL(1J!Hk-p4n#DVo9!^Vs1LWPc3E2 zXxhDV6`2TZW6Q!r*23}6SUC6^T|aQam>2T?**P}TreK|+7KOUy-H}Hv(XzLBX~0u) z&9$BV2GdSkNHXIIoD1uZ7|h9Wm|?IhA83{5Zuxwxb%o1eOJ=OtS?dCf47*v>B_7lr0!kzC-U8VaGx>gtV2WTloof-&AQQ94dT}!|>ejwr$oSG{( z_V8NX>Xy6Oyu6*JJE#D%Y^bl;g4TxEgbdU$GV0t9I_~kCvJ(00t?ozk`U6a(p1SB$~FhWjji5YL1^Cp|$1XbjRb&q_wp74tvW}jhZ@2{k@r4b@f-BW=GL! zcu}*M*8)43Cii%w{xvoE0w`6jCl|ax&C>5Yn==x98_BK91XL@O>n)s>!hE_NnuWw6 zcrf#stgyG{Ac7=wjjA}VG!2>19Qt_f%b9=4=JG0=&5dNt2W80f3a;@d?2v288?&Ku zyY7XQoJn1~^Y>D~nMEXYI?qX?)>8ZYM}-C?f9FdqD3PmXZ{DKHmrS|gPDBG3KO+%V zm7w3j^cii}j`lhhZNE*R`OcbmM7EA=o@ilh{>Xcc zUa9N@QKMm@Lw*Hm&`TEwW@ir)1m7~Rl~JeTl__fFL<#i5mfX;Ua2&Ny@$rI&leFEA z*ZtpqN?CU#;#`*&&*lf@TXAgwHNhVNe=v&I&Hi?)Xw!b5kpo)n5{OFX0MSYAQtQLL ziF_f#dLA1h1q=VP$x*G1;qvH@{o;d{-i_v~AJQze<=veBG(-;v(6#9@+(abMfA zb=~~3Ow5=Nh={)`T^Y2s3tn3k-4nIyrv|bgvqJggd*%#i2r?Sp;?*?t0MKV@d{Dqs z>ukwOO;EA=F;}@nL9dpXC~v>4QSOlAF;RQ_hnfas)2~FOkpIAWm|2 zRFovw~j&IfOt~~nnwNUdM|>*K5Aj}l4v>St@EUb!t&_bh5KNvH~<&= z3rO(sIS)7~Fx z%r2!0uU=-wqW3De+%!!sp$~}Z7L@e-It3x_XJ3O0(?C2?| zrV2#e(4yH)N8t8mN}U|8C;YyIV{9hM2n?3Fcxw=W4p8~9ex|i$SVVYNIBYXFEcac^ z+@Sh;8z#G{PUc9{8@)%GF-kb&rMW?XXP+VAZ~2GHq17udv;-hi5zsi;9qiQ-CF$&& zT=J&0-vrW|=4x_?Y-|+P+2(Jy8MTZLZfNf`gfamE9P|h=+1;Mdrlvw-c0OP*ALL3U zu=1~|7%qz-2|JsjWH;x)qI8{lEOHqgQ55AX+VSH7{a_pvxRtp>_7vfh$KH$s0vIZK zo)iU&ITPK$0it@}W<+EuP%?{GO!A2==R%wp*_Ek-{o;Jl1$(-mb=%;m67gDV)v z_8O>Kaa&Jr58ae5&p)jVDSVa5mc7rv_NfM-v96+kV*RMG3REu8mIRKopMqz^TkjUW0tY&}I@R7QW(ki=UN@`fCqKnv zQ7%GWLv>IAP^)i)wlaG4z~#0^Q*a`ou~HQo{K3~#LF-_b)p2cZwVZCo8kx5eV6b40 zhykhFZ#IZ96>_~?2XM(VeLDak#*GT-mi8tegv`hM16whzAUoKdS+McrNQ^=f-^vG~ zB6Z!-94SFY>39&`I<-tBV8;3mxhe*?>6hbUXcgB^o1<6*Sid(GX z=h9ZUV zWi~m`{0dH@dcOMACeqP@+Vo*!%R zLoU-Mj0YsDHLEE^dP0AVBY%!T6D;U#y<+VDg@3_scdzBTt9bnx;^XsIJ!w?$5X z3f)1*I4gZPnM=v}Xl^y`YbywgPEjP02#@T8}^4TT;~56r>Ll5*koi)TnS0#DodtqVgidn%^010BMfy#LN*Q^EeTN2F6>p7XA^x4g6re?>7e}rch4B?}N>KiaNxkj2@ z{6)UQnSc|3Z0J{cVgys(c6I=$C9GS`7^PgIt9RaEYV>n`nl_$MNiz)OxA|qijHCa8 zl?n-kLP!D%0(an*h1R#VS@9t;lB<%H{G9tJ`1T$|GaD2JeSAN#L(f#XB3SSzvo`EB zXxQn{Naot7>MGMC8;I8HS?Lvt6>*Nb^&VCLJeIb2`f5dBcoY&WtBK-j`kkuw?>2^| zZpV)~zWm#mAQCH1D;a}$S(T4-C#=3)V?1e4xoGW8oM&MW<4`qJggtcCOZh~{vN&60 zLHuD$ahIx>@aGi$k>ts+_7STat+)c^B)>U$g!=aw=C+V>%7mxOG0R@6Cm%lG8*|_uw-~*6O!7fDS zp+9w;1XqqA^p@M4{?K7m31Ik5AoSZGV^*21H41Um{L}sFSYdOJFMAtQT~dgpceH$Z zOpF~F_x?8rNDA9av^rPwcT68L>w>B3W7fE8wCxi~xxrh}IGmvP^mGMk`+KD9RCVq1Eu1Dy|6PRu$ z5^$E8poP%d!nZ*{X}q0&d3~3Mrj_$SyTW7;D~n~YEJfmrA8`v5k^*6uNsUmx%gq?+ z<13>oGnO7ZE0=rjaz130?SuhgU6^H(cTK0c$s6vYnuapBbN2K7XW^3~dAF>)o^jzDk48Uu7^c>iEo&i#u>1+aK7ZKD2jX&vV!f4;c#_M{E&Y zv3;QvBGf#Om}K;Q{b!>YVf21%NHAF%^;>dJ0d1fM=&g@gBjz+txdF^xPeaTA+3 z9Oe*lNgq%*-RECHC6`1bTL($1a_`KzIeuoG8EXYC8r(g#>)@y}W@_d4t;pb!d!k7r z>yvK3VPk@2{Jskkv+9Y-t3hXc-a$shyAJ*BK-)ZzG6+o!SR0)!WJX9ZcB)f=X70wCQEqIj_8^1e&HU z)k1PzmJ$~(Q_yAR0xnK$XB$^g6c)&L57cd!&K~g|4bh^%_iHnkdY#w)LL^l^YbLzP-q^WXpRYvA!#l9f_`iX1n!fU>cvzQz+2w|lM*am58`<;#5B`^=DSSxlny7|yuDzdu8M1{&V(ci&PE45EIe zK=82*(1CSuij#tbl_&YZZbm8AV+ac()g{IA>B?6MR+XgxOw`DL6Uo_KJ6pbl#fz}{ zAp``}CZbN2e55GdWPC~V)mBH`3L;U@O5saDYtir%J;Kinc;CX67*1$m6b6a~EypjD zrIGqRPY6EnO_GmUyp!#XFrx2S>n{-UyyM9cp_Bz4E-J6+zA)Z0&9Xs6Fz`$Z)70n; z~@P$!4k&8?3aTkXyBIhW1Xi;QNm*UfWyJw`*joF*z-rduCMOGz6L zXJwwq#l!oDIZVfOsYweZVM0a|Vo@$(RZO&(9ngg5Fbl)KGBWWW!k^xj?P;_%CF@w7 zd)IF-0Ehi{=aJTBb*klo z_XQT~n%GgRwN!&c#V9of=E`_b^d4Y8%wbA@kwjEyu!-pZQ=SbC0%E~qPkzOVkUB$` ztPWAt=XWmya&sh+D7GU+QSiO*sH5f|t_y_>WIKW4RDM z`=m9ZeTStzkb05Na5jMcIj_|5T5-xLK+CtIRQ;Wx;!CW-1ta@Ey^CH#e+S=YVDUEO z_#QUcQN%{)zR>XdsJ<@a1Id0yzPUJaf1nOjUqVjp@C4Xn$ zfC9#EjxgH1)zVB))%3f{3G9?W>cSE#O3x*bqV*|!5NM1<^A<24!slPHlBYh4HA1}p zQ@bDvO~_R4>llktN!sh>X!$}l4^72fiO5g;1HqBMixL4pr?7*>9vSGKY%NIlBEue~ zJWV#Z1jJ%ZjKqyj()*BcZ~I*&M!~uY2Ag=CpyDL$xRKh&h4h?%_qaa|2)NYV{MEmx zi_1mXW_(ARlaTVlo=JC8+4Y{{gqZRhLwyjMJ*Bi=iT+?W(C#-sJ@k1zS@y10%s`D{ zfu#TWuoZc01*E4gMYcC zfP|Ek%@d8#`TBhl4EHyt04x0&*i>h-=52;<9x}fR>i>s_euIIbU{6KkbEfo_>!kai zX8SLL|HnU5FaYvy*w69*zaTWF~Q4Y@!`ukqq}^>MxD;PxJkcmp-EMFBVvyyidMsI}+4u9MUCle?6dc z?GDiS^utd)yl=PCbo1`{Q68Z4jo-mf#{5m)<9~XG1RfL;u>e!v)00-IY^r9($DxF` zbS37AH2yl+MOJ9n$_J;Cq?a{z%#&}n_DXdq-wVWw@fzry<(1xWTYC|0dFzdUsgU8 zSdb_VsQ?>`Tsl|QD;;b?aWzKT;D!&Uq5uRj0L)0fn4Ze-fkkZgANK!mmkh-MayD0v zad*1ZN7C$G?4Ale6+EbXbWrbbR;xIgs?d8bC25M`2sDQOi?Z~GAh1AzMf9br+4Pf? z3pPOJ@oiIk2{DZ0aE(!KlxI@^&XTW*09x>7%t-g|3`gPxB!)BP3uUZ-XYzl$ME`yB zlvL0_&2xv9`7eJ#Kz&lY0CXqX3%ZcMv1dqpZcqU0!+((w|96%Q?eh!Zu<#;Q?6`nYiUVkD}fFU)>d7hBz=c({Y`mR+u@Z3 z>a{dAox!YLp8Q~jQSZf)ppeVs^Aq(f-K92NPPR0#^%sw$xc<&OCjT%U+fyT-^QM`6 z0Cl#oM9SiDcVWL&Jxf*ZA^*06No(qrklO2g~>hQ6pVet7&TRKLS%?~p4K#SOIWv4d9Ao^!hB zg5^4$)r-1n*ryPguabns`?tRR|L_XH?WYh6n|OwJ@#6UzA5D$vzxiYjQeQ1%bmR9D z3s1EM$G}Z2g*mzeIZC;>NVsrJNVxb-NJ!K|N|6`5r6CE1yiZ?g8sV-zK3rb%xbC>` zVDr>3okl?9NFA_E;Xt$$uaJEeP1MGSiEOuL6)ZZHaXIOs$ z$W__Au8z`^+1|RIzxj|GiO&H#Wt=R;$-U!=e`Wpt?>u2%LaEPfuFJyMw0T+*iJSh- zr9wb~vEim%4rWDg#4CmT3t#g8jzk|aLmHS4{~6-{7elF;!+gz+sOIxGgbNsw&lqJ} z2NUt_7Y=>?cM(}^e_25Pm=?-w7=MmbG_3=;m#JTh{-TKg_reJ7py?aLM|gJP|Mk!O zlbu}u!7C!P^*H);^yQnhN6Cu^8?yCVi}yasgHXEYkf>y6VxhvQWX-au%Ry48nWESr zR^8~Hz*dtl{}?#PE7W4>FYQf(u(nDdAvS8UcI^J`8-4L~#C|7KLR$zqiN7xDN;vt` z8mgH4d3P*;d8Gi>I3h@!+@C|gw0fKVFDU`nC0JgNV`SR1-KkS=+aVygZHPWW#}v6i z$5hVf9lFeZg07Ra<=v(67||e! z$YptP)aB4k5wdW+Pk*QgtVYPCmz)BBX$r16uo4QhHobbg@9ftB%?d*Z8JTIUZ>P5N z{gO@%%h@epw^d>laSp@|vNQc+@WCGjRae_5WeqGc!$-qg*)9MQ?YoB3+miyA#o=VKVHJ-BO^DEu}kT{H;w zqu|KcT4PVio6cmp-RkN&r{gED`39@JgI}bJ*;QtWZcD}&RkTVJc?i3cuBkIRT}89r z8IRCNgWuoe1W!IcUDx1RjV`~ODzLB##=GQ8;x>CrDV5NMaDKIM)O=rM% z+_2~5UBSw!^-_Z@VscH+K8N+#_b0vr(ee)OZQLk?2Ls#n9>)1<({v5XJ4Z7bl>wjx zmV>%3qE)xm;60;S4TjiPwC*PFNa`o>iSj7G>NPfN!7=dZN2PIwKkT}P6j^6Gb_)6X zPzlU|*huj*v|tH3#;yp=0WYaYGO>$y1q-gpHI`Y>$V3CChtZaa!<;^QgY;@;Ii{m& z3Jp~Sv`&Q>CQr<36e#QcA^CZ=#@}00xxM0mu04`@qbWHN>s`4(9$WGCY3Tb1qTzS= zd91VVBC+&m#j9vM>0o$0u2NJ>pL#x`@>~VRFv~NY&nHJ$ZZnOQuXp%gwlb-W$kXU9 zMod$_+N*!dR+Um^KC99q01%PM63*9v{m*i!A!n}h@@`? zF<+oxV1pJf@j1G0&d>=3p3wPRZ-OTBq#3t7&1l^!vb|d^9tzZ4B|T0~6dF8l3yZD8 zKA0v7r%pSimz9H&4%?yBdm07of0`+>^47%y8^Lbtb)mT4J@WU&w;zM)#i z4t#}xI(n^yvh5;uAuQ)xS2aB>78uAkrpz?22!NWYEKF$=51sS=Nt5yKq*)&yfF|$g zU5Zb2HZxY4TP55B3`hq*gzW^vjZ9|0+gZ#%96Gx#)u<@#EtXf~ZXNbVX7%Q4=ZqR( zs0?mJGO3=vi}==)%x}bWiO;D`6*y+PSy4|qJW%HN#4^`nggsrlvdAFdQhHG3xGHl# zKiR(mdY09&$H-jqzF!J;f7oFPK8@0)dIGcH>pR$QAMp*uFt#EP@|d%zl-?xD%BUSn zDw&M^G%(7hp7Sl|I7$aSWyqvsd|v^91QXxn=2aJAyG0MN%4Ha)kJq0IC{@#8ewsRg zFj*A5-0Y%#gTA+vV}x%$S@<>aIfN6o%k_+GZ&wyqDQ`5Fw6k{tKu<2cXV$Gk3B%{> zIy9`9OxSr!R=+HYQwKp%*M*1e}FCTYl`8z=T4pXM9Y+Zs& z?ibj?9n~hNIZ{|TyV1Q#vn?rtA1AWiJh)HSC2RXh6mi+?QDPZ%yR0}pUZb|LU6|eC z*{Ad%)6A4;$hgL(Xm2y$jb&2Of5g~7KZ+SiX89$71FnSYf(QckLr|Ixmh60Y>k;}e zBu%49l;mDq&IuSEuR=(8Zq+oOZr@ByQWZTw+nmMfpcj$;7DM4;V8-y_aJgmK!a~`i<%b~Sf}^t;HhiPDElh|t z2d#LTp}(Yfm^r3d*Vt1js%0sOdlwn@;%D!r0`JtMcdX6VHRy*>h0KAuV0xSOiHmzb z=qa!_qxo!u2KYU+Aunv)F8tNa`Dm*@W~Z+IHr&x+?Ml>)gMP_=8d%Qt?tqJSPE37G z6xOU~$}WaX?2UHPh}jT3IakWuC2Al=nejK3xmPeK(!y_d!~ClA2;4%X2KDG#RCZAF zw-q3Jk02m%SuUB8dvNI?%8%Yv?%#J6z` z&wJD1S?pFM=Xg3{yh#~BiU|Oao8?<7k5`^$w_X;Bd_|}H=G5wbiS3KY#X0m!v({23 zb-os^Jb*QEoc?Jg9zAao9J&6H{#18wDkIh^q}Fmt=o0@`K&L%^SG|wpQ~0A#@!;T) z#p>FoB2E}L0cyI@Lph1Z?ipqwJyab|NUUEDyZX@m0YdSoiUsHCKX4!j=?w-c(~A<+!F!!=BHw=HdmoanQlT^B?l(2-2o2b(ZXzK@bQnl zp)^Ks%V&YfQTGb#VZ8DhXZZtPbe#u@+9qk1P-TWR0lm8%snjI?{=`Yj2V))1>D%R{ z#;II@-OfRAEnCv;q%V@WD2YEFpG%b}Q^Hca&D7SL;+IQ%qDuu932e=SQ}I=`Fbusg%NdDY^Gnnx0)qD zV-gx~dNfA-`mW=GXKfRDkVJjxaK50Vy1=^3V#ZoVoyPlk$>mj`_!}hnM&GPLg`)CB zRn)i-;e^Y()51nYH9DVwwa$6PAYee&940lDOFOdEiEDruP~f1$<2JTQL{AY?CqAjd4v-2 zY0Q-q+-!i3`N2>eJ9>#O(S=YOIy@!Jc&{8#OydT%R_|B{@1JL92-?61=(L2{IUY@% zY^6}2fX@cn$D&>mzZJfMP*=hqtK|v9<mnVEzhm6VaZ8m3s zT0S*{+20AAwL%>IG-rZmgvxUb1*W9mn*SLo*v*g7GX;rpv2{6mQ{R;v``1xaS=ywhP6fqcP#>vaps88jl!K0S>oJX5 zd|!3n-7lMohtokn89sGmpNm$*y(esq7TR!32 zmk7drwBz!+YeWI$dDLQml=Eqe4Gra|7f^(p<{MaGUM zVt7E3Tt99PCZjGc=fmFc1Hx&fMT+$_@4Q3xGFKK)BuDy({9BaW({+LFccHhNjFq}D2WakG|S6GSxXDJak2V}oa{~H#)09K=Fjfe@S1f*0wy#n z48#PS@tC~Gp_WR<- z&~H;ih;JXy5ud;{XCu6%JTRZB$Vx%a!cFrgad_yGEL^F>TxxO)$$ak>;S4)idxy+_ zv)q(D($@W3<);rxN_KK_Wu3VV3ZyB~A>*;oq4WL}+z1%zILjaTm2)Gbd!@Q9GF}1C zMyE?O3_`=|;I_O}?Bd%~Cw-u%Z-HAd&i=aUv1neC6!7SU<9w<524x!uAZ%t z6;d@Sh6R6jy*hAf@Z2o)Ea;9PjyT&GU@C#rSYMkrV_m;G_$7a$-Bg-^hn>{l#)*=%Rmyye)nmWfWR~Z02W^r&fWT4(po29GNi_sKTFBS_- zH^-W zbXgH{Kf4wXwv7*+fIUs&SkU7VO#X(2eDKhb+PpQtmm?oYu@jxP9QOGoDp8d;iNMStnA9c=$YQao3V?wR+v`=GY zi(Fl!zg$0Bzf^YR@xm_yb@J78H{-GQSFL% z{6hY!hJPQFs>|8N2q`vZiQRN{>L>BzvYD3RiNbZjrv-R*(y~{)pLKWxX=%Og-a>_l z*)GF1%BAv4V>9cJQLF%&8L>Z=}y=li z@q@)^YWYvk01D!AkGMTu76EMlu8 z`upWbZYpcTC`D1oQ8c$7{%p@u1Vci+zGGrA3FrdCk)$s@CVv`a?XGr;Ph~6U>@9UV zd1zWZ-6Ip^4ovS$+F1I)?0ed(8pKaJ6kNNg>-Z*q;YIV9=fjV$ zP;N3>O|Xw|kOdHalxbdEn|^|Al!~R1dU=1zQj&B>eo&}fLPW^vsy>o218VLH$hkD^ zDZVv)s!_P2hlxz7spN5f5a$N1W)i3wh@ohgs}$GD<@7pNmUnwF=*%mJ?_NG73R$iq zLLfqwyQ=AHOUJ#1UFEZIdJ5N=dAQiGKac-#tt+%{T-kG(Z+b0fa&i`{r_>m}wCSXT zIwK9wMs7x(u&!?npDUmKWHjP-8i3Z5>UEDiFqhaxfVBWgw{3mUi%m3VjoCr2-qTIc zyVR(H&ChJ%NzjQJGH>yRO&z2Z5agN6>b9BVVu4ma!AN z&XY)lcyYIm1Rk5WF6vRtSmUzfwOhTF_@NWf0cb+IPEBsQyQ%+7QyW8&{BHyB{zi~J zE7v23E3bJ?hYsTxo|76v_12mwzRB!g=Vnp?2s%!CsR+1{`OP^kCEF=ym$iU~Y z{(y1nMYu*GlcJ6@5cDx@&mOUTZq*NZ_o_q-gWN<$D(wwrcL=eb=k;R{aCWCyCo79E z?0q+RIA1PO8qq0?2_>U?z0ce_c3WwhU~kVTnb6h^4QTw+OBK0wPV?G`1oj%MS8Mf zOat2er^kB?HdTiH-Ss-V)F`emhbxtVJ&%otB>vh0L;`a3@l3HRdJ*Z|>3k!M&mB%@Ao6Cw}W`(A3-Oy0gw z{*375E%vL6T->*l@`{z8?|oU8xTd)xcZ^^~n^%>WogX3Xl!@yY8Io7;6)uWoG%Dx! zSd`ORrm#0}_Q=FWMNu=mPS}6Ie{BU&l_-xSJlVUtqFK5%a{aUKTCT7fN#q7}zGrXD zWSYI)dlG}z`f_96w2`Osv93`s*Qgw4rfJ4%jzN#Y{k{5y)p&7+_Snp$-H|n`NVgrF zk{wuoMO;1CXH3uq;QivVYEr*Io3WyMl_;fMZJCB*X#6hAAQCae3dOt4S2(YNc{>Jsz$uarxiPDS<#B1I`ob-h%Dx>&OUgiKXk;}Yjb0BHw^ntyQ1Roq%}-n0 zCAqnPe)M7lMvWTH7O2anKX|{GhQ>Ol-AL5+Gb&Bf+x@#$Q}&TU_+EB9s|CP4nFgx_ z>|al*8DO6QA=!a0$Xn#V76*Xw>(inplbrCy#bY}m6Z3+HA=z8wJ%%Jco5-r2oV`Hu zu1Fw!mkTh=)awM*q93|W*3hegGMF2#*~Q)#YBRh;CY;iO^WlYUboQ*(vexi_rC!wO zYqGMe_rb@rewak3&dlAM8tnNR{*&!;DD~5Y5c@y4zJLbXRiFCd;ynPaFQ<^6sazHG zbkK0Q=`jRrxh2iQ8&3(nLXGAM-s?*Nc>cA|@-4Bh*xA=4Q8Ie0r7nyym4^W-VK&ndPIsC`NcfbSGr#P?%y4U2|kPU&mB! zwi4!ixJ+LQR<>g~LO{vHRUbZ^-Uaxd%mTaJcvo1vbPvUmEFGGj!OPI}-s(YLzNWw+__Rs# zQ;X9k3B%S$?C@^b|0l)H%lzP%r2UJD(Mxh)rq5$YytU^W9ipYs4%9`}!?_)<2x6Jl zzEFh{ZlRU#t<)J1?v0fO93WHs?gz?m^EFxRuI8Y2oR! zR#Gd6i{$mC8mxN3GQZ-+uL9JuocvtTe7UJ33H&2dTXx7nf*JAGLd6MhbznM3g@;lF zBz6$n3A+{tc8ws9n7#AIi;bb9LxSUlU^H39KeG(=4=2_p^n|AqSjq@J(yP!&0~Bwz zmLUSiT2t_be7u#F+dxYt;LBB!KSC!3iVoiQ*GZ-lcO4V$jlRB3N?zyAF_iqgUA{& zwf)w;?hkVy$Q&~I+kB$G8e~Q}qEQ$eW>W{MrfOporn3Q059yGsAB9M%Ta^jG|LAc&P|)bob*JN(D@*IL{8P2^m)chw zU065=HxQD~&=pSQGtXAUF(z$VIef&+7E2nN%jw(?o)i6O5RZt%a}0h_Y#7TuvGxk! z6&uScc30o${nSU3#A_8tAb54W%z$mLLZCiFqf?qxh4DgQBa&ao{B-bUB^xZ>?H7xt zPCuSH8Gj*X(Dzd7zS?YdT6@gr6PZ<#@`NT1mI(M0?{^zy+BsoQ9s}$NL5gvpf1MOl z%SA8q6^PThLmQ|0M;utzDpM>rH` zFI2I|f4p8H4zj|5?~7+zBPq3?+bdB1K{pWR%J{$FY-Pu?y3j zvPZM8v%tqfFM_pDtUX>-=^hX>SEun+;pNQkG{mQZ*Ws>-T%%&E;M8bxjJ+-HAV%*F z?Z6u=U^C6hA-`$9oJkgDc%!2^-hSDyWIDviFq+20e2m=1InW`;Uw%_oOzws!|1Dj> z>jSKUJ}Q~?76MerI~?gKSRuS(|2E$*njgC|NuerQIMvxD(>ql#?kt<+=~v@dt-@LBUYcH0l?sNQQma0Rc#h#4w&eN#3bb9FeCn zFTMJMQEVDu6#MX=p&TZ>rSiHl%*3vc5S&&it}Y^?9n`A7v>Obz&kKp!8cjnRa?eVP z9RA95;449v;fh3h@VT^AWy-Y3RvA?y0mvAjR_g4egAG{Ij1anAkpj=k`L^UpZ+7OJ z)2j}A-8=EnL^%d752vTI!0O#9;(wSBCY5mS3-P&WX_zF_VI7)EzwanoEnlYE;V>BR z+V6-7(gf58xSj4oZ-0@9T@8JmXo#vNXuQBVKi#L~gdrZcaVkS2ft$f)LfVngq9n*e7CJklvas;#` zKR19yT6?P`frCMhS1VWslbpaw?qJCf z+b#1xo5}g9H!!p()Mb&_&{@$6bD2ua_48)ZDE0QDJ)1W*xsXMYmKMIfh~3u(VN*z^ zD0^?aJT7s%pEK6(m^~r#&RTucR_a%S64#k6s#2WnA(9Z(>j(6| zx%j+T91G431c^RS?!A`Q%cYdD9kAT0HJ>z?qv*H^g*^4Rlb7jq&}>~3{$84|nPg14 z?sya@@**E2@m8Fv5^`USmoJeBMRU|RC$PQa(BYX^J6dTX8|v_R6Ch9lDbD)y=+T0N zoW@sTH zxekKJf9SF$?dD)XaRPNBRCaaG;E4^^&KqmstCj0mB8DTFxSy1Ik~syXiB~B##w7O%RN?U1w}276oCa&vgCO7!L)RC#Z*mQ$n^}1DCsBexeo6Wq zcFn5{sMZYQs^`MLF&Clp-7FD-;cERHAq`Jo_JNYKoTDhHw~1DZQ=!Q4%K+>PGaUtgr15`DyyT|iazvuaZXeC zkmNAM6b?{!Rd0f=5CsE#O`plin7yguEg2{fNgiY{pN7vc{6&w29G^HeVO}OE;&YdU zCpT$2YKb^xR6X`|c3!NGaJ~WP1Or@@QW*o#NNf|4;8=#`+re`)(~5hFUo}~YYu|pL zRBt>R6>=p%>w_@Vr+t{(p>pyH^KMOur)nh->Xf%n*9 z(NnJ9D8xx#UVTDzHF76eiUW5LS@(IMZ+z^f-~9e+v$&14o1V6xr(~N(t!=#^6Ndib zN#L$bcACiPzSi^2OT#RCAuSWcj+#n}LpD=H0(_|2N@qN9aTj32uKE*b$OXRs=e(qq%gN>sb zhrUkuaw7`w3iSAM&F9Pl^WB36FRX6wL+lyo1K+xKqzltvAJLY>QCj=4<}_9Wy+OFJ za7fIEK|s&_Va(U|%${>o%G$nja+;d{NB>Ji-}L!NKy7w(qul#wrXN&{o*wID)rtQqQ=tQ7 zj2J%LTfXRGq}bxXR4C{p;2H#Wdo%p0T(*@}$Wlj-KC1Gc=MwQ3PfII?_YQ`Z0*{ei z+@nQlbQaKdQN#B`_vxK4R1kihZ3FB@f>#X;R*Ut5<*$=ZBCT-DdttW^p1Gf^qKqL? zuQ($X3)PbJ+n?^gPE@M%u`Y*N>!#?pr|TNB(Wc7>sSG@{RT-KSZB*)h9J0lyx-2m( zco|+E?w(?rzAKj;!B=($))MR}L?t=Y7mNJVBv);f5)?uc+ z_p^trxT>(<;=SduL&HqdScMzCoZ7=I4ah;XN5A;uKRvN(@Tzb`5?GqV*)6Ea*%T

>5?nALj(>}qhDS0f>Axo_tKn*WoVi!gl0#AE`LkuB2b=Yq8E7)8 zQ&oK}#IRPMGI);GiDm0A|N1Q{lh(!@k4A z(BGwd%ucsfydGy-5DV5GPUArp_6NJ1RV85c81sFk=i8d0F#)tFGLXkb)~b(c191q< za8Hue#X40him+&k>-Aa)sTLi%RJ9PRE-e6mrha#<_(S?2QFs@GTFJA|n^P2ifD8`o zS|vzKS!UfBP>dY%{2Sp|4tG@~e`kfjRR_>lQ)&UjCpnTwDKjjV7D+`uZ{N#gXE_tl z0Yqu$fK5!p=@ci1PJ&eMQkXeOZ}DnT{4llsKf zKEtz4nGS#0Ka@?1s167O<07mRl`0v!|9M}pf`~q&erNc#xg9>tbFZGD8HCW{X?J^v zzJWc-{m=vxZ;K0Vd_;tOBXl6q#t#5`(@cl22VMo-y2|o$>LaPz_dw|Dr=8vt7>Gmr zT>aZhxNRo)`K3UbGrI%YD?K3F##CJr-*&?U>1;kNpvG?Ez(wy_M*bWm0GX z?vY`I(Z?0bcUl(tQAi-0CiHq@Uq0QMFz5+`+QUn)k@VllZnbEgIu};f^UW`H3Xr?+ zR3BWQc}=Bvc@j8~c=Aoa--hf1wOd{qbtB8~-YiY`#b$cjwBS~DsJT3PV|hdya?PzR zugfY2pX8gQQ&Q^gzq5CgoniLsE@B*U+)FQI?d0^%%Hhit7`JrVOAqD>>TUV1L8$wW z-eSZkxW}fZWek5;iXR|SESPKS`|XeLV5xq;+*l~`8Yp0X6Bo@#dv3a2^!ZCax(kZ@wTtLjCpRHq@hfheMfUO`m)J37oN;0AyvvPxv65QP7i}$nUGJ?L=-au zv`Jczsnw8R-(P!unpP*!RFN^Cd1#|pDjs!TQPo#lEpRi=0Da!gVNjv$4%_2Qql0rF z)E#3bznDHdab@UDgzmLqx&&e?U~>e+r^hf}r3Pw0O)7@VfN+}ba}Vc$k&Z--~#DTQQGnv(4F=5QArtGWYyo(DYvA-^mYr&0D5Q4 zpF2t^j@Y}Gy4;0WbHs4t$Dgur=Hg55F9OXe0HO2ebPiUxJa9EzRS}nUgGvY_Ofj7W z*`E1?2$ku(uLQsKu}KK$qQ<7i8r1m4jhar2EI!S*gs!cZx2m-NFf+xS-ldpqz->1e zp{GzK@bxi_<0wGA1FiAo3npBi%hf@VC_NQPaGf=TN7NgHLsQE8IqU=RmZEjdYuHjx z!z}^#V{9^^0e0z`_}nJI0M?EJ!CU`;mv-l^GAj$VJM#fd-4J1LB{exR?Yf|jgzXhc?yrata}?wHYt zv_Js6&T(ohZ+8FCc6aGyc(@_tbQgjealJ?M{dvw6pgeam!lkwrp~KYseCZv8ubQi; z-<(H&UE5Lv0#m%2*Ra7(%@~{?}0wB=xj#E6W5M^#p zOuL<9x%l8Ig$A9CdKMkmTe<@wRM0}_dZIz@EN;ROle7A9VhJ8xI4xhmp!3@e960_N z&z4V7o+_(>fh0@Po98Z7D(Cs@Kd*Co!L2fBUF+z*8a)FR>O*)$hHP*aA-XVc2TSNs z9oi^qMJ#kEJ)S2X6@o>tE|`F@R&ezDQSD--YU*)T34~Y7wZR&6^x~vVWrpIUsmFXJ zJF#^10n<5qmG)Iq;;YWxV`H0MO9(MaXw4nCfS)y+x4=JrZg8spHfg)0O593v7fo*4 z2w<*<(VX_eP&UHT-}A4v6Qz7Vtv>qk*O2S3?m7C8#+Vkg-_*mlAp2eO!!jo~vIE6V zMqYlOQ77QI)z1--tW1_GXKO)jgghbV)urMvi(Mo9kWS?^il^6|-!2#4A^8lZZNVr37RD}7$|S^k2(2dY3AN0OG+cxS@CVfC9X0+TQ6r;Jbgc5 zOz)HQ&Bl(hX+UXd{|(4-Q+$exmg$y&l+=c84)p&1RpT zk%JUS@wO|1Toj|QT+7jtq`3zwz-(I# zO-1(7*Fb{lWeZx*{`!1J2Zr-w2P4yLlOHH9dUjd_#&!vPUGPR1!q|M#BQ*U|)hG?W z%%JG8P&0AzxcI`Gd(-rd5|#%PYT&Ig@&CCxl2gfIs~hg<`v?}{(a3v{6(gz=+mp!l#>&v=L^*dr{^ z15tQ6w1wFH`+D>A?glL!WU%^mK+?=%dYQSM^{n`hnL4w~SmD;`RzDRgo3zRY>_q{0 z`w{|K%^7`Pb>4j`vrQUZ%p*V#V>14V_ySiu!jB1TkjoRIHGS#m3vvcL<5=@Br$GxW zI6>7y9L}!@&I#L&meDixhANzMva%<28q~^vEzFL}qKmnxvnlGoIc=X?&n8sfFKAR zw~VJ??--wO{^NS-w-m8F7VMzRNzgSXK$D~4PRXtXHOw_^qa0f5b9Vo#ezj%`VFI*= zWGMy}+dQfqZO=(bfj3Cr4X&r7oSN<@no=dNd)g9e1e zy)Hha==gPBL1W@k1y4R%?D1MqH5d;!V!*2u=T3xM|4^E7DD`YJRwL!cT6dBg;tttU z+$2E5nk;kU3e5wAMqsm}fItz4x7_3iDL9Y2rRPAzBftoCdFDYzgl@o z4u+2JPMJ!E$nNxxXu|m{e<;%pzrA?uJij$rrbKJsN}NTww90NIdg|+BIQZVtrChAR zo4i#@vrz=tdQ7}$&`*Zk(KI6@kK>276fdpRhpB}?f}N-66d^>3mq$xRxBFCo22bn3 zbUJQ&Z4RY(EMq+%6&?TBVRhLjr;oJ8LNZreF}0|BQAxf?RgSW=;sO z{Uz!EAkx>^)UQ}rgP*!p$;rVY{Wz(!m04`XN`nsi%VAiI+M+-lfRcb0n!e)0Ujq2c zsn9e(44z5qM&^z2(bJ4k$)v~v57F4|zNn)doQd2a2Od9zj-(7eFDal4sZj92L&^vC z)cu0m>#56b5HO9Gt5*q#A;TfJ5-Q}ji67q|`#7r*3*W=+HWE-WnLcyGKb6(;xHPZU zCpRG_eQ+VU*kgqRSuNLw0Wg%$y=V!DJN*-eam>P88Vx2@yEeS2$o}M5pCv=aB_K$o zRnNpaY15_ImZX=ImXYY-$+4n~HR4~hSu}39MQn51g~^;B%sW`fA3N-=7N%yT4>dvX zDM=vF-s#-+1@+}A%>f%iA?juuTjk{St0W|Fx#8H1Q^3;Azvs@P#Pui~#&^|efD_sy zWHH}@>a;(}0KtQuh_qjk9Qsoar=Gx_gvAvJn2u(|ZanV|6yRLiu56zd5K0pTf-xjD zBhkF?>C?DX_zFw#(_i*>`3FZ0W?|68Ol&y~(b)g$r!85`)g~%dC}loesF6#+By(qL zFpsBN`QW<6U#Son^yVnHWyCP+t}?mG*>COBPJpB)JUaXgK2$g`}h9cDXwKN-3acGvU#7q&O*h%J8QtqnhK5TkeAWg5VBgx-|aea3*GFqvLmbceDV%C-}VVLq*jk$>*{2mok6OCr&q>cBAqaL?_1WXjqZv^_q@E~>KKrVeBUK~u~%xAOFCg>6`y8B z=cMJ*4vBKHkUH<&iI$JgFm^rmJp=kwOZeTUp_b zPB3ydsTb^!Tcg#4!1~dd8@!^JCxL3-oaB$msL?-cx)e}MUM~_m+ZTC|lf>+|a$y4Z zj;bxM&lBB&tr6a#Pyw(TY&Nk!prgLzG#T(~y?*Dgr{R<;<(&lkyi}`Fus2!trp_Fb zZMNQ|#^#9xji2G$N=;jeKu;J(tr&A<;CM*MtOa=`_SjPxpeUFJ?HYCl*(4|e$){na z{ZPJl^EHMqDP(YT#RONV$wW`{YOT-kELjrwJGy$u^JrM}R>w>u72~KCj9?GxXC%TA ziP!kVaCs)lNopofGO*{ZV~rl_uaLO(Xk9(xj^Z-#EEmx ztU^cGwcFMcla|Mw+;_YEkJcU4D7&W2HW|qRUw4n&&fca!o(Mpj_X)3;{KxnKPqawz zB|I+~V#7v&TLkFS5V$}FUo^0F;j(ar1-IeL)s}9Qi z=jcWsT*k`0^!Wi}y9U(r#NZ_T1Z|OrR;84(D{Y;)Ge@P=>g+egGfT|6Qn+$UN+57i z;!Sw&#dFHxamz-pSu0_Z$SOx@0c`E1X8_PY_H?B}SI^i~PmA%;FD@CE=-D1*Y-*D- z2%G+`W5a;#!#VoJF>dYWg_ck62X%Maj}0tUD!?;a845?9H$_gtZ#yP!<{P6;WYcLM z?%v=E(J1D}Y?~=%eipRE_?p6ANZR8+gMV;;5#ix@B*UM{U+KY_;j%v!{MWv|3Gg@5 zr?E$Qb9rmcETe#^CUQh`3(7SV3d#icY2A~S-4g4qclYAySU6&?Kl2I&ATlVN@P5SU z*c?qw)NnzWBrB58Mf_Z(bRU3ZehryDo+9G+mU;7Zh)gjUm+k)5ugyhN!{>;5^3MMF z70P?D0g-m)(c)2OHy@v0h;N8E`JBZTq zz&NRi!Kpfab-zOChI=9wl`hf=`C>f*M9;F`9%{@51;>AZ<|v&eARxoo6dzu_rPAjW zWf&B8%?+TJ`hEtuD;y=jq>Y-6-JK3K2?RU0UsE+A1q*?0G-`{f1Z(oLzWyF-pFr(q z$bPB{GL1t4O2?oIp+PD}iF|&2i)Va*wyyzT93TLOH|@E6#@{Fhv;KS2NGXkaRVd*m z6M@i|;j)l^{~C1XU4zUc?M0<cQ(-wvD|s_>GY5e z=3CJ}05I-b_ZlV>$)iiO&Peb)vg8_aw;a%nh7|rWK-5D<ccU+=d1lFO~U<@bOa+_DM?7gdE*!exC^X^hhlTT=jY0RNL)j>*w9?ctko|s2a zEXA@Q&o?%6w811kdL&DI7V2-E47!UyE8gWOw!Ky$K1a!eN;ALY4{;*??f0Imh<&<5 zT?HNnj_5=eV(_%}70wh|wYPUh&g`rXI#81}FV`1+yYcS)Ig%Jh6&*nBQA z64_?LInS_S=hl_oeSCJfJGVmwT5Y0fRy1P^hVw+FRlm)3PUX9L`o8Q~lPAF!Yid3N z3O1@VA?b1I^_I&KTuz9$Cm}jPmD|pK6qbBaA^Wm(n`3|OPq=lbTl$k z(4hme&Op?)?%^k#?D;yU=U<(Vy{|5Fx(zv*o#*x?U7DC~;8By#S~j2=_`5J3xEzMoAYC z&HE4)a+LCRl3PP>CpYFoVV&VXt(5xQrXl-l>F?biJ(P1~jOSlmAUi@%{d#C19aS0& zil2DBw(z}Fs|spsUp^Px4BMoZJHn6CfMiH~BhH`!4bg)A^qET){1<)zO~dg^qDFcT z@y*NF*P!yaN-gOvX76G763C+xEB`){hycjod#e<%cV&j2R_iPx!9a3Wo8r*2Sb9rJ zjaJXL+J(3e;)3R)z3!QZ0L@g~Dv+XFa=J_>Iq2OkGNY#2ZB2{?Y*1@&yn^{-(16j@UI%AASGPoUG@YBpOvt`r^!B5d_?q!ydpP z9uUt~N`V3S<=4NqoiAe+fFB--~b^8w2&g&s&S~dYhsqsxG0rhh^CN> zdviHQU%1)ln9vvtYohty5?*#24%Kj<^3%>yI(OU$59ht}N)`4(9u@axt~8a9f^}Di zh-+D)umiW$ldwbk=P+{ixX)CNKENSh9pTGe#A zj;}*Am#Xnty{mD>fl2*6LWViZR?NxmInf^4_I};_k#hqWM`dS)R;x(?Dsni3c^S4r z^mqgvmZz2URgf~>w{qE&nJ41=0t!|`Bpsz(=+$vlLW5Y+N^ zimXDjeSY{ajaxnoCkb~U&1$i23akq@XzEVkH1$OZPvl4^DU-x?_W}B7(xjIFlP~L8 zlgDv}H00<7g>e!HvH4n)F`!9?%)A85@_FcE`p8d#Dgk~0-PyB`$S3RFyWSqrOB7*q ze!aga>dE8ug@&mg0yf_!Htrcbb#$5F{jPTqVx!!^WYngC#QiY|K;Ndf9uEP&h*78 zs@_$;5GhyI3}T(Nb-|6YxyW~kI@~qN`y%BbMM zVv8s7+OQ; zJ>9z5L_k((uR>`M&Q`MmgkBU^82aLK4#QrqkTj#Ov*RX^Nb4+f&o#LpY1SX2fro+V zPNYXWX=J)}7ZJd(RFW|BoK_ zw`XB&gW3G09?gc^gqw#K6N%@y_eY&$*dr`0yO|*e$4SOmJ3XP$EHislEPHK5nKYIz z6m(TO?M_?YOT7&E3jd&p*H{%E5%|0v!o55Bu6k2XG$3A%G*kb&13gCWT0{nTTM-;<5iVT@)9ao+Cc4e2wTE`_uUSf3-}7(M$N$bIB2Ghdef(8%GWNBcej6*#c!vSw}fCY^2iT;lL#ArDbqaw>oQiIj*y7mqa4d zucx2u8iq!FF}?Q+Zz9ka+t|8%miRhkW`%z}U0v9WFy+BKoZ}S$z1t)W(AS#-TuM_% zP`GcF$rGWgqm|5faS=8_e+lf74n58rOBI9KHr|#dN0qA>RpV%j=WouM#VVKE#%ZegECn(XG`w zB|O+9)$PhWrjJ5rTO-*RQZp9Qe4QH+oH&!b+v~!J4d%(aB5+e1dau33+bo@|cSGv= za4>2zU+66B9J--wTG$d?$H;J27IWvdu~RU?;6_T&Sn6G6CFGE3>$4Z$()!SKjkdB> z#Eib_63Y?9jy>Zjm2s^vU6Z)ii3g>bG8q_6Bpso8g~jcDi3_eQ?|ep}vHXhO2;YSi z74!I4F)PbkU^!q=ess{5CyrmpL{WgInv0@uEIBsGgH5X|3@mG_bt&|*@o*7eQG)+u z`4(s|L5mYfLUbW5-3i>Md@%HD4kVZ|zkikpcBOhLc^OmE17di4GqBatx0Bq867A=I zbS*PIwttKI>alDzHIUg74`!*~+Ar@Hc!FMl3+mD*!F+(|ip)-xZeyPW#Pjm@^8|Cq zLh55~e$H2MU9?=29w;KWHu``N=9(DY&smQ%RsPi30iqWFMT^PstqrqtSAh*6@>jS9 zip*h#@D*Xvrw8N|nMnj{zrkxrh=7hqipo%3{hlCGqEU;-uyPh-+@{k+ z=%Fy+D|$e?0{KWWZ7Ke4$Hqyl`u*7{d)}wqcx|DAz+KA+%QhW&BEA$-h5CKN^P^AO z!0l`-fz1NA0Oqtkxla3($ZnoK*W7g|lnzlE_`jX&(F4_%SKgF{u-I&}-jtJzCd?Io zoLpPow9^jmisLW(y(u;y8zVlf6!_epCzI6U$88Ew=W*Mt_5jf`_hi{^(0QIzqVP_RYsHh>Ox^WcGkR_h*C&~;0>14f+oCzEuOhn#x%yCN z+vHm~T@VXnr&m6@w@N2@+W`{$OD9onCh>7?;+2N}cW0ut^XXmOVi&?8YEIaQ`q(QS z1B#~cheMjs+Kn`Y`0_2pqC}?(qbnU%^JxVnJ0R2H!yoQ;HVmgYXPo}B+8v-G-!1ArW}~+I+xO^VJ6X@>>pZNYv<2vI&Zn z`nc8A;6VAXqt8S2{FiYFW}K-DC7OG8ayZexhDde|H7VoQUqF4d2Py2;ZOK@6S7n4o z_nx*1RUVtqa;@Xp!~nz0@NWSm*myxzPDYX3G|vP|2<#|Boo3PTFc`i!ZOv(fW#V(s zO%&bx2rki$ex{J=x8?6Ye}DH^cOIF4@=>B8@`xjC>jffT*kZqIK2PuQZ<|NZ2SKBP zQ(=wxu}g~9uYd^%=c?yHc})s`wn+KKORs0s_4;$Rt%NIZYta`k$=b34EtRHuo&cqibs0*DnZ#o`L5y39!oao=bX35>e~V0z8Rf z9=oH_4ygoMY;ms>)RSq_@K_L8nFi3G_p;;J4F@L`;MDKf1I-yk%Rh8_udQQtR6C+M z|B;U8kM;i4=cyt99jX#r`t;TmTzvIN>bY6=^$!b6z~`n_Ua`@DMf0WVlA~`T zJF+1ykxitOSx}Fq@!?D4{N@W3UZMvbb?i4PJdg_cF_A3beCOAPsq6#(ca;pTu)_iw zRN}z~hP`+iHSL*FuR=oF&>_?-q8CakE`XK{!54A`b~##;CCG*F9RlZqHofRSDDr?m z*ef0$PnK6MZo}Oe3MyR?bYpAkcPu)};-5eIR=s;z-aEkJtNPARe-!ToR;e#J?4&*7 z-BYPGg5Q`sAsRUlu6arcG@}CECvtGIX%770nQXrlRY{&*p0+r_D%?*_@11~ki;zx8 z$P#DB`VZEQ&I)270O_IsSRf*}6KguAZ5yn0tKXu0Y?xQo78|D`vGjhJV!3IJtS|NI z^FXxn_Ayh3+M;N3jL@Qpa1?O+V z73@ZwXm0ianQNv`d8kKOQXQOL)g-Zx$q)x>C($+TbMNR{R3PY>f@gfvId%>7xPAG{ z#7l@IzvrY)2VVX+#*<0_Ai{G@di$GtZXA3`jru7ok`d_J`}h{ni!q*CXRM9{b^tE4 zh^@_-nRI$^1`wpNzIj^UDH8>6V9ldHS;+g?OwjNU4d((HN8!=*X zm6>vT+5^J~ejj;>2X zr?Ej?|E^sp+D`P_|L6W~XphOAidWjH2!7#h2@ekea|P6+3a&F1O84ToYn`bfwdm_` z54D++*$5!&muhb`CWi2qupk7LZ1QnAb$5fKYFTlv&Xj1nXc`jqXZYWnE*30S$|tU> z5C-5I1);H~e-kv^8EOov(g1Oc`3!pHnvh zg^3?&&71FbCcA%gWAHAQ1DuxGdn$C}{rWFue5zV;Lwk_jc%6Ufm1v!?R(1|ctcyUz zvnc5oq#1vpI;k*#P?U9?#jB{Lfl|E0j=%M?W)aLq8^u7{y3L24jdOtN6!K8eFvXqV z72nfI_lAI3wt&~$5xXaZaG4?#*r&+@s3nQ^n4WO?Y1&O1pOvNa>#;s^tZPXyLk!4w ze9@+z;#hb~23KnmA*=^*h)af;7_AsfVHQ@TFUkWrA7dXr?{)D%>rbTUcw9cGNwXS4 z*{j8fUtI%N{PHvc1_oK+tjlNSRQP=A!DSk(}TtJh@JmWX}W- zr(+p_OBHYUd;#t)@7>!1WNhz*3$dKaf=IG+^yOmH3jiN5nRi0=3x^W?4y|G>trsug zZV{msS596w8Vz%2c;3OD=X>MNa&%I-n=+&!Hx|jDbMqqY)Ghm#uVlTZOB(=sYy4Mn z&953i2^vN6Tp)SzZ7iR*4?P%&)W*&gNQ?FHcD7EYYZD3L>zn(RC|Ak}84=&j0;naL za97m{1jc*M(2qW_^dOY?WEq9q7*vAui4lX@e@e)7OFf06=YMTOqJ`1871l{tuC@d6@P0 zLC!r8uZ=$Y*865(zHV@Hx9c5$R?Rl!c;kr@c`5+O0;0WRN;me#W}S2Tw3cqQrR{_? z<|SnT)c|Qn^04R3$cg>)QjS&AYw2UV++g|q*T&@vR+Sf5+rzuMiU4!YYU35H0OF~n z)h1=%I6x1(jVt%&|2u2o8@guhNro z7al2GQCVjg*;dz}CqnQO-gJHpMOeQ}F-U%@cVy9QuU7s1e4bS*j_LW~-0hn`K&|0U z8s6LK-cOq53u+EV?mX{esUqUMM2vradQ!G~_>(J@?F;pbpTm`H{(}j)0IOLo46yj_@GaR{lT?W%iI-#o!hA$OCAq{MCXLjrF%=X9A3h zNwbKhN^C_wawiP*;MG_m5h=?ybvc0h7h?`6%fKFpJ{RDtrMmqT37INVoNhkv+vtn? z_=%SJxKgqw924z0`Dvnp_s;w#tWpv}5@-G|$>P;lU3bdIHk&{kt+wrPgu8GSCZg}$ zn*dEQC+djm#P9`|*H&7Z(!H%2l_1nUr`GT9uDSLCPFw4EJsGdzsT=sEatR)P`- zA48~L#_djIwdnY^)404;w_-@$6jMHUy#TbvRSLDg7G=Ar z)2?6de2ktxviARLR^@8=76DAFaI(>>A>eI_Tyd1RGi5drt7Y1|BbRl=E4Qz-Y*x18 z+b?!W9v#*L?nu0?gy?!a;4`)V5DxRDv2wx5tRYwZA2^R>GMUx9Kg+~$pqA?}GtQZ! z+N|I{4kIgbU;i_%XbIi*G9Y6G=@#YKQZoSvleRJHtXr}yUbntN$zt1`oDcFq|9!rz zRHB_97Lh`s`EhS4H+a!Vtx_wy;&7qE45cRoH?~(M#i+uDxMSo!yV+VAr{k}2lB$G@ z;bAh3S~KMqNAs!`HIg|O>k!9{Tc;FGC)#ul=V6(%j$k6ivyaxlGzPz~wC3|yoMGEP zMBX&qUG-djn%ceV3f^?7SI8C1cf@6p&)=IiZX4f3ONbz0nasW%ISh3CmgI5e?)0!~q^+&oy`dsAg z(PZ!FZkH~izHsq)(e#?_`{O$qR=-U+%dC<2Y+3vPvOmikR&va92(S4T2bP<>QWw~) zX^*w<;gHO(8?!Df^5&b}Sn1WFUkLeJiE9LWgxdXt)x$*|H9vkPKS$^IfqtmOKz`M# zx;a=K^=mXSO2cI*tC}T>`kmK{&)=c?abVHctSD5nPM2})xAQSKXS@M0e}xc+$1gHVWR{j%V~ zzYMlst-$QF@n=DypKk5*nyA|>=6P<(HBB(%e~=2h0_H%HM;0dixwIYRsiOdG3Drhd zPBhhbMsJU>$WVVu&INLiiOUh|2fi$v{1f*)%|}8-!-w+(Qg=1l`0p^nnK!7UpT~if z@2}dQ0Ol^4NSp!YW+I7~Mhd&f>+DLp?UBx;u>C`3pEbGN?eN6m9p`X!+D<@(|3%6| znP3a*6tqI9BEt{RfsA9$9Q^V-C~z9eZ@(P_WC5kc&#P_1K$U9(axpC5s+_b_IIJ^5@e*kwtO$NNc=k*AA1`gIyFlx~u z@a<*LwF|+Yg`U{JwO$>;2Bisk?Y=dJwj7J46@|uPD6$j_G8l+f;zr+(lPL;Z;1XF2P^- z@S{kO08)6WJ-9pSbwaocDjbz;g{(goc^UQ~wS%c_PG_K>H-6S+1kMG;*HqBCug=X{ z$oBK~?oRkKrQNZTL1Jj&cMVRjgFl)6>(lHE3{5inqc?D;gMu8jW=Cu^Z&kf-QPxxV*P*<`vpo}k|eem^M3e|y%3so64g)%vL8yMyk4I_<(7jL%gRH55fF z&)3PU_B~C|<@)3m-`Vv#u_7VZTZF~<)+%a%_;Wzg`?7M`c6Jg}=bJnA;Z##fpk&>1 z%kN9O`Ow?ba7jdjRge&4>+#R|3|@zvu`_JjFa1R8e4Aw(qX<<7lPQ?U)PjCLb>)F! zL0~#G>VIDX{>`S=fqyYZqz%+3-Zx}yNTW;Nv@;IRD?f z^=A8js4K_F&o#^5@!6un=BHm;x6*g3Mt5I@*=_V)Q4d!SQGI*-8BF9Wtc*~Vd_?>M z=CK`lPl$2#gJ1)&u(smsj-~G%&@ko&{(Q4K=Jk6*{Jg@LenRP1AV}Pkidoc?-Hy2C zS4#^qfDK_J-c{%%Wde(tKIVOC`o8fs9<7B*@2hzR)H|OMMR0DS&Oy`sPbs1RGVD9r zOrYL#lc5=cErYi@-=OX+c1yPA3eT|We(H$oV=rvJdzt5Azk&$szNGr$BJeFg4nVOg z6h!R$8}9<(oX~IB-U{!sqmGF4ELQ6|G3gZLuiiQnQe#0OzRg)f6ucqHXY z9g||ffKH8&5ZryKQR_M{l?cxTG4jn%1(;1zL+pQUTkm`+%jmybyrlFW)JL@qfuwy( z=1Df^>`R9tpXjfB{{AzXyk0+MpbF}7+x?uA`NVmREzj$>C?af}I${E7tHw%bGD-2{ zorjzdbz?PF!Uyzxo1ssUV9tB3ByPi@)nxxw3d!#KaXUY8i1BQ>Iyg|gzbiFR4w0fa zGLyEPuf!gxL##BoVfj&Dtke)*5`vERfPJHp&VGn!4dCS*CCH!_wN&ypnbOh~l zgOGFcDG7asUOXyT~&hOVv`F_CK10EvHCx zGbANedLARx-RJl4b)SXMo9FKkbjFOlS8ax9shpIFBgVc?q-GS&2hmys5XGJajP)+G zh1dG=h%;t(K9l*MKHUTuR{-xHQl+8P*rSqjPLDrO&6|n>L7GJTC$VmsB;60^Q!#M+ z2F>u=$=RFy!|%kANQldzyKfU}@DPy$NAPCPICPPA@31zreIjt#LublYtzHlRjCu^@ zdmSFFN>_sz>~CE8+<>9heT0j(T*a0UV+!`v0kT%7+w;?<+k1Oem;yynl^Z5l_FD6A zTxm%}9UC>k?!3PNZ;h~`OXy>RLAg(kM>Bn*_n|jpz-wl)f6}oge5zS4d)jYKj2Hd+ zL#i-D$&-0dCP}lS#V05*LH*emXeS@^{iBtg2i?{AmQRuXg>3yea65$+=yOYUr`$^xhE zA@d?!gnAaFDNul|8K=K??Br1S`Xyw?FKaIav@L+>7ic7h95LfTfQYp-25J;Ax{QLp zc(zXp!gXWWH7_3;Y7Y9c2SD5MRG6&>u5vPvAVy8ZSXxzw26$D!-|Zu|ae%`*@%olw zSAlik@?v1i21VL!pkG<<^n7uV1KXN`)s*fKilk@AGoYZ+`b7G<)j&MColq2(zL27< z#MJk7ygdY8NE!!`~kMM%3EOs*-@n!dn$w|O>%$fJ|fqc(HVk>m+R9X-! zrDA75MIx9pSDpaN;6%2MxnhosHJi!l%|toCENgc;W)CB{6ZV^y3~>?ax?wd56!Y>N zad`6WT2BnSFsjCs-Ufk*-0VRr!bxy7o4|2;O-X(hM>sRhq$cDOO1C8$ibnC3sIq`}Z%p-_MNMk@A#<=CuHu|B=YCPfnx>|e^`{VIpLyrezeYmlj z<^6c0cXDP=+JYgKcf0onWjgAjNiUj%sORT^{CNQm$TR=zliQ2b9JzMJyL_kj*8*`} zZvW`7b4^FWkt3R{@emIPJ` zqC2+^bJ@-R<3lOw%M)A#mCMe&P2a;H(jd z5`={a=m+2>_==bRxH%suA@0#`f9g+w2|Z^-27b_*-kc(RSDd}uh5WdPIfesL5-rcy z-XukhZ0!8k&vPxPciiT$oa;>1sp|8`#{9urE3w*SR75g{JWf-Kkni;6r&Q)(Xdrz( zIzF!fu}ZP>XcQVJ4LTzDIhHeg0@e0^{ak-_gbgB+}n&Jp#dTKSDeQEbg{Pf=PsNtIULB8Ao&--gq#B;iYI}$8mVym z=ObyX|7g(yjsTXgsoc&@=4|(>(;tLrd9$50z{GQ{f+H2n4zcM&^*IS4ZM`uV+M2{i z`q$yHZ%O|@(6<5*@-OAMWTIg>Sf#yA2LEIS`@0Fm=-x~2emPN-4R7ob1AIxq-Rm`B{}4e)>Q8ME>6$FuqIz#4WpV za|UWSJ8W9;|MlYg7XQfu>7z$L9>iB%bc(CyRp-_Und7ZH| z8qjNG^vS%AoMK)y{|$X@jEJZZ$CKNZwXDOK$5_*q*RfX<``C5MD9cdtO$H~c!x7Im z`#*LZ{pT-n|8zdGoqqbLXAUOx|94s^pZOXAHK7J-FO>02{%-}c|3^#L|JUS2YzS1; z#*4KLyaIpiF|_@km#+E35r4DdqR=>y`Xbc@{(qE>{ZsFx(Vx=E$$s|vFZorjDF06F zwzjHVQUUa}k5^z(lKHb*<$V#9RLcAX@)CS!5itY7}mLwI* z&>0RE?)r#M`!ao`B3Z}c(m3tO;n_@L{>Qaw62xwZf!r|5F~XbuaT3n`2-aUq zHDj{Zfb$~epT4L6{c$2k@YVHs(!M0(ATBf=qTJpbjK@z_ZIVIMpIvKzN+5u_Sek7c zjWL*JI;l=RGZOnBe}*q`XtCdqfm&3XyUX2S?Ii97Bi@T2-NR-{S`-0LaZ?yf;fI;4^A z?(PO*q`RespV9_F}a(O`7&)BFs+q6Ufh*iqU)5kigsZHxbdUks29g1S_uhrbq@j?kMG zsUp-C+pxC=;<{3CeBSw6*h`~F_9Wv4!QTQvvOfYK-#kf>`Wpu~<|E@HdpU??^KZTQ z#Ur@#Sr+@>fZIOB3JC%}clEfj&VO#SpJM%dWPO@G`Aq${SfAAxk{X6{*+8ASt2@9^ z(I%RRhhmV3Cee|d`9HrgUB1a5cVttx;#K6oPt|-$;E?B&tNH6hV7n6E-`ZWe)wfs`0Y=LGy8+we)n7F#@`ybrJ z|8}Y|dI+7zYb>2~e)!8l%}z8kH4!M}2LSXc`p)G~jGpJ|qj^8}G2s6@rwy6>1gOqg zfp!0B-ElXPej))L#O2S4;leefGL}b$y7{W@yy?z}Vh(R7@;`QZu@O%aP467$n0a=% z9lM)%r5nmL<8lyzx2T6j3vERUJ;W24Wzg|BZArc~|Ml)aKcpYeFeJqzCsFWS59iyp z(H`K9X_eFVg;b$$nLSG0IsFX|v%O+D{Z*}&Dl4S8aQN{*{yYYV0d7FM@-{n+W7oS* zZwB&;*#E|7eTV;)aqtR0CL{ygs&`}0@!x&gQ2-d&OT>Kdi@fpml5x6{|AxWw+oF67 zRRK70Wfa@s-#B6bfZ1@)mq_Dn_~7rq@p6=pkHP+Gpvkkufekk9zwrnWFeE>uVRuF4pj>^{Gq~+>tJOB zg6WUnFuSlKQ=D)TWeZKls;pSo{~#;Q;}X+weXe=x#4$?zZ<}PlxWunc=8Z&fxC-X( zycGEBbDF1ElF$5PQ>efT{qIy7mMud1vQDZ0k&}kGy*BXt_iyTh1%lQzbgAIM`Hdzw z7MWiJdCYcNDeciPL7h^*w4wUvWO4B6YZl?Z9{dR@MVH{7kTZuS&j0m={qktOZsSKE zEqs*GmGY(0e1gR-jA+ozpl-nocK1J_0RWPD^?)YpSWok2)KOP1OixLpf>G<;KE3fE zlTJ~Vj%u6vv_tRnyZc~L(~(8=I`wod&aDUzdAERaoqDw+$c#2HvCh0X_$o=}eA@+{ zvvTb+rsuJ{5?`ZMuB_>pnaZ$O#1_0%55aGS9NE?w@5;ZKe^4s+Fs@IYl2t&cW}R<# zN^0wLtoK^j=ns?r-t&BA)7;gb)OWFZ>a`XuQ7e|yN)2}9e$>}Pp)80_%=@mgy;>59 zLAxM+3!FrwjB|+I@f%1=z7(rBMy-nDDm`a+AP1GH!@dl~QaRNr^?uWC+akW+jP|+F zY%TaTW1H+gvz?^xAsR&@K;K`{CQ?DG{8R41{h%*bfQRHd;__{goxb%deX8KK4$$eK z$JcfD!(jpC>*Qxlr16ZM%25}MKqE~&u4XV9klk?oawVdNz?E%BHBG<$&+{Dn&yF14 z^J#b9X$h|8$!a`o4(oeG>NuW$+ELtq>a?JEy7I0q$-pvgmfk0ktnhkxdk}h+Qbbe4Qr$$%=c{%IHmk9iGCLdWfGfAn<68c5%g1SJ z`}oU8j`3PLu?ZqLPYb{f+edAit)yjuZiU#dn@l9MpZ-BI&fImIT%TJ7jnd4Y^(|gX zc%{d2E_c9hP4^~YCbw>DUW|EMk*P+xL9@Eq=SXAmwi26*dc>Z*&TMHAuTaj5@mhnAZ6a!WTRy1cz~*=Jbd43&7(9guS^N z4dI>(opzb=m7k?*Ou?=-l-CLg52=+Mu}rC4GS^|a+i7V$*d5j>@*~-Ox~5oynGgN1 zl;;x$T0L{Vt_S}r1*=UUL^+y*f66!&h^L=dV;B7F>Q>l00rUX4A^AoQm7UERysk&Z zaT=|Tp*r3ZLAH)Ii!R?AS3&BeZV>Or<+a*y2>#*dw@6Z%l^1UNzt8p4m>V>x7T-@2 z+HT}oOxkrR5)+GLCh4aouI3qq;iCq)CAG7RXuCOA>Sudvrtmq{coA9bo~-vOm*}Wo z18BXKtke%3EY0$YW>~1GjZXymT76e2slYmgJ`((#l5b}_XJ}0R?6GZ94`tD0%+hq@ zMdUj80?|=?+m7co943@!61s6Vz)+hLibqheAK5sZr0+u+XKKgNBe1{5TND2l^q+@Anp#6ohU?Y+J5f zefA%djgicV3AiYnUsXcXmX+I0zMtcM(+kig5B4V5Vg2eR51GCNB2U(R_LQl%QZN?0n_ zanl!%QJr%$e;>Fh9{MJ9u3xkyeEdZiv>tlG_^*dD=BwO)d7{hWrURk-d1E8g*0=91 z-@PW}ijr9bg?15iiwbxKLPuQ}En**Pze$#uQ1Mhi7wyT$-}4UwRpo!=EEh(S{^DM{X6(W z^H`|f>_1(f!iDwy?wMPTv-I9`k4dcqgepCBSlpXhy=gE}h32VRNESUZEwv|`CbV;T zcld_$_$Qra4dV0YKltxc*!1O;FWrB$38e7f2JPncsamLiG0TggQC{!3|I)fVTa|L4 z5+~^6B}7}U^*$7bO7zay@OOIh0OCxT8@`xLqF$_N}+)e>cG(8DHrbEykQ zbE8b~=b$JFuUPttP4A7$PBeOkZ@>T+L(`Okkev3IUwwtZFz@2HY&+D6Qv%V5{@?v3Z=Utq=Y<=?7NGgOn*P-h2bH(KA#j_-oV_eU^$0ND4 zM7eL;Fj^kDRXfH7wrQMA1SVL-hFuAckg%(66_9V`M?5X<5w}|%4qF}fi@*1#1hSGq z=C1`88sOSa(sok}2U;CQ^+&a^Be+~40GzyL87V!#2QJ>rvY*|CjN{uGu8#V!(Y*%N z0_>aU2iUJ#?CnrPN*B5bmX*2OTD#CBMqTRirGl%!h?A$z)eZ2|U*J$#tDOT6ErLjX zYsF)>YK@qc`$M-$Wfc1fJVGQHtEyz!lmZ{3no%U*Zd&d3f|9~8G=U*mM8|de-r%Ajn5b0XAaxB5I z36qn!XrV-5pW7DUQ@-1#VwIV}OS7ikHa@$Ko01>eOSx*@L&buVfGmVioc>GaiNB!w zRv5$9539=8r$S0d!XB&VkjhEuQ7d#<>KYX%M!5G|dXaU(bxYw^1#N85!1Kvh%-@i4 zT=OwkLlGP91JLtV59#QagGrjykM`z27#*=WIQT$QUAP%XreW23n44lRF)Y}h#f2Br zeR_}hYR{oMN78+8`+ubZgym9C$E*c5$)-!LAlgM zY*`#LAqZo?VJDiXp}OT~xgYgYVEP^`VBIcbv}I1UQzOz=!W^DCDjDHiJ0U)UxiD~& zryox;Zfi?U!>iziS2kMQ_*Pz6%p;o~6@&2T;bo7?KQay0s~VoJoitk=qOun~znagk zvS{kbko_xh!FYkwg_1V5d1r_--z6OssC~8PO&3Sm^JRt^&Xx5LDXLlhFrldQeBvUy2BkBDg;Acx-SW$atPQs}zpJ-apHnc6l;8O>%r&#V_Dkrgy_F z(8J6~7{W#(=2Mk%)0ZWvynQ20u6chK&UBVlgVu1`8x8ef!F?T= zUeeFNyrg`N{MEL|BUKLoQm*N}7b@%TKT!`ovIb15v%1usqDn?4c{EfGb0mLIB!e#7 zXUraLOeabHP&Bbf;`aLNZf<*MJr}C(sPPRQVQq#$^G}4)wf!G|C z)?b|~K6K``mhh17mT_>ujo%#`cgW`wTf(*ahwxh1wD@qzX6cvDvE4OJ( zNGeg>m|d8NGv*Y8q$tEbsb;OBDX^>1Raw=j(ijzb;MZ(Cvw4g>$^1w`cT)5^ZgC_B z%bw?QXN!d4&h9T;m5sfm(U$f`2=en`GUJexS2R| zYgi&}CDTQ=U?j1C>`Q!l+Bvln@;_cvln4Iv&vWpRWAIp#3#U-sI;)nKUvonCAVAbD zp~fszO-8`XI2PJapqA_U%&))SKS`fDeAQj4#uOEPR&=_X8NvD@i>|!!4kH$Zv)U*Q?Wn4 zi8?pMHZ3D9o`1-1A%3^|@+`bBc#A) zTcG;1%sHQj!^TPVbfGK|VGvc36vvl5i|yxjp8QcPPraY432TzvLMl5E!7{eWv#kw2 z@OmAG?Sx6Mt|x2Z&qMQKRp+<0=fz+@(^TcJ_5TF#gl24;jFATETC@ewyiI1B)~Y!% zZO)O*?!xMdeOBgbXi@2gSEvzrD&7^iF$p8aY6esQ+dtUzx6Tl*U389pf%Q2?+i!ht z!YG2a{ZU7q0_H)${DM>Yk-?{(#WucFAiKX*KTo8WYbbSb0 zn{m?62*YQguhKxoFRy}xk7%1NHP}I&LZlQ!6ndHZ!*(44N*L^uL;OGSRGB;9+T1KZ z+$U*=>3RgMMmGou|GtKGBe+cm#DXna=p5##cpri3@(V8zSFRt%zNiS+_f?nz|tLrF9R=v}G&ny6SJG^tqfTYFKL?kJnn_gB9vGy$$b-A4|1g|6Oi!Irnxl?`cKYF3g4F0{&#u8HsWGXe zFypmEMd@cjRU$^A`h)fa1W-|>sVb@Os>j}KN&k@$hxEK^>CS8RtToCC)hUfYU`)9_ zwjLrlZd~-das>={Fs*b;`5p7jF z`_l_+UOm?X$sHYDR|$4#UkiM>ZuTo9j&NXYcs@lbaql|Vb|s8*IJ(aXasCr)`6}G> zX(a001uM0zBk6@(`)jnV|smL zUfiCeTY@+H-tZ6&P0h{_KHFpudjazZ-|rl^HPZ2OtIN7v%nP#7TI)YBqt0D&z7R0K zU1$AvE4h-yqS_gFpOfdQcfQjub{lj%b~NX|ii(ME!?WKPMJxo_ePZliDx-(XtO-lo z<6yVciNQ4XxoCIin%lIkC*oneUx&{uJ+uTxRK@*q;VU=peN6O)SPJSououTZx2>K} zO&tx^5?LH7q1ma>M748{T(;xxIw3oPaW#v45F^5;$~OJ3#wuxx z7p6m}(Nv`qOA~!_=+$Yh(Ww{wNo!Z%-jDDYZVJA*<7WIU+_mK*1i#Alu6HRF)UEkU zHiKe7JH_X@sW zt00IOvZ{ClMlD{eoYwzP|580hbpH5r+Hrh-S-Tf(xVhQMN*-U=+d;AJV<#tr`tVQR zMB>00$dp{_)^@WK^ms0JAX>q+dEeO74=_)lIDE5_$w$EDgwSmBnXYHCcdJh;^OOlL zeuaA|mh#>W1llp8oT>Pb5A8_xN}{pT>y8b*;mTm?j2{!am9@gN*Z*zSn#HeQvag|+ zV8yeli>B%rEb7STIy(9}qy)q_2NxU2yXm#aEg ziARzLfvkH=#l5;JrQ zhhNA*K09N}DqCY%^STFJL$RrH&e$@gR^8>9CBi=#m&3E09TN5KeOJcX=L43WJu3`X zC^y~nm#L?B?(LxCb^+4KxFgraq3~%l3|V6SFNKG%LHG!>bJhP8By+*G4IQKOHacr!$Wm-u(7vD%Y9iC zzs?ls;tTMgOiTqs5`&x@yrw^zFJIBIyWB9^On^JM`@Tc>XA^~^0a!J~szM#M+(O6g zy@bcfdPhabfDV4#2(V6bRwPTZr^58Od|U?45MdU=n!V5n3vl$=ep!*kwCt>)#rF@^ z|%oJ~E#aS2&_T{}U{z3{HD%Htd(rcWgVS+~VphxvF0+&0=WLleOZZ6Z*w~ zlc9VR&(%_3I~J{CoA1pELPY}BUlZ_SyNaow0;siF=YE7k}h3Zdpj0^;mV}d_CDbw9W}5H zyySOtFtYpMFqPK3xLNk4b%#z;2%qE!yX#58av1ps?eyk;CVdqE?dYRd6zJOc?sW-hWf|8&0`6n(imu%MnMqez*UT8;a@gsE7ZU5MP5y= zXOJ-CV;|8|nfDVs?WrPz-CZoO8%@uDwt0EBmbJ>nUiCofXBwK{`Kg0!wwHqsdY$6A zHp<}6f2KGV<3O-j1Y6ZNMM7QBZ*lLaOlXPuy8k;{;?~N6V1Nsv&hmN*<&&oI+uxy zLJONDx%(!n|CwI`T^MGBH`T9<<2<1`{xM2wtvSHn0iJjG<;-;W%tE}doyl5JxTslU{+$3oWp)STD}JX~^n&b+Ozvg}Ux1dPi1R%o-xiO; z{>;dPH{h{oWWib0-@9B|w?{vay(KK@v*82X3<(4iBhSa|&K@p--xvy>1!EkftoPOz z4M&Bnu~s?%UPeLGjx>jeugKC@BmX_`v-9vV7TFvK;irEaJCf*ZNPR;A2(PC-+@ z*D+{ar*|OIG(%O*E?Mvyk~AO?UYXi!KHgc)MpJEHf3a73G)TIiE3ny>`{i(P64;Uo zbGymn7j;&Dh~jg0j{A^^^IDx{hmLK@>UTEOUnDHSbe@j za~QS@kN7Xu<<-3U46vueY`l4R?j{-YhDjeiv65LI4H+OZH@K6hL8!WEY#hg9VPy*s zCFvynii8G|9*|GYA&lWr4?;zX+o^E3A1LFl4excJw1J(4+zN(V_I_J#0P;w&>O% z;gmwLqS5z(M=~G}WxyFV??+P&X&jAA4R*PfI|SI^N3nBttm_NaV-^fX5ViJ0tD!A; zLTSEKEshl;%SZ}6khm9TeC_4D3)B_=WpL)8c`xH4@j`7tVD;G0qFju8U58)801J86 zfz1DF?>p>$q!*v7E*vlZu=$Ws_yJ9wJ)iSTABFZfBx>1lyxEnvTvnw4{5*M??=7au ztWVZ_A4HjY-P@0xF<%%)_9`pX+cxWJpRw*5z{w;o4=SL1X`RL5E>z;)ya`PMvX;+L zczP0E$Aw15a~A8nK3FoB(i9tJBdeYM#z2p=Az7o!TF0I4@#sak+o3Zk`rI8uUlcmIE--?i>;vtn%_sitJDyn*Q1a>NHR`Z;UmB&Ex zRvpigqv_78`9Q-D%KKdEsVjK~G$0WDsc>Xpy+vZ%S^)O@AE6@(+G7y@QifrUB^_?4 z`c8;NcJsL%O&V#`?V5{DWc04zXplHHV|ibA zHf1&(_y$HjrAZ%3T8SFL%@@W1djMqQ(kQfxJ z%uNuH-0uzO9#hz55RS-Laa`uiY(m@IX?XMpQhkPVJke8b zWfvUbVaS9Ef7VAdMvv-y$Xjs1B;3OsLUyWSghg_=`Uzh*YXCV_x51qmSP#vVK(9yG zU|ze^R(5l~T0uJb7cVoSHyBfcV*0q9BJmy%Ve3nYKv2IR_jKiBTdvZwEd*WZch@fT znq$Ru6L>4NrAu11CVYF1j`vPGvYx&hO_+`r9c`XMe}txm>L&hBjtj}B@>A2ghmeX< z5J>o=oLQ%>=)qKaC^ed(R+Vf6?uXlNikk+`n7;^d5R{slq@o$BAEw$WsxhjF!3;pd zbebXqF$t;6fR6bua?`u58=Q;CTC#`>*vfKY7Bf^Z2=0X*jq^N zPd#ZOwgrS{U4=5XL#8te3Ux0(3zPhdsPP?&U#^!B&h7~S+Z^~|gKVz%1G~lZN!h3F z?%1R`)4-Vu^B~<)wXztq4C!lwI!(x|^6so}B%gOSBar#cDUA6F>o{xrA>43vR(9Me zUmnX`|Gp^Tb;&bbis$*LQUO%yrOD>m#Hq=?4X3hXdN=yC-n@o?y}7y(_i!7+BEAC( zixsLHj2hkke79-ws!s>xq%iBcCTEPL-#>>KVL0k~E||u00F>KZU7?4vO`N7X$Gy@% zwW~-lH}d8;{`)4-?tq5|SwMY-9Dv8{gmV-sHcR8~O_!&JV$I-*@G|xye@Kf*Y8;?@W#koMGVZCF)G@ZU#~^x2PL~hr9E! zxmygOHxP}z_G58c)%GuP3uXIgO}frQR4o;EEnbZsX!lmRh>l#d(%0#Nv|}^z++!j4 zvWx41x^``6VZYq(J-B={Xha2i+XH*y-oGa#wo78ER#@lREDESP$x*~SV(6^ohhoq5 zp9v_5gHH^SwiGPl?8%Yqa{h_jbO;3mfr#m`>)11Da054ymCH<#GZ1OjoZbg$?p)cTT`-nN^x>yJ)+&uCwGS z>W88WyehYkdlB73u_w3)K0WzXew_il-!wn!yZu^1liz-`wDK&Sp`vBB`7UhI_69O* z=M1mT?L%5T9*)ILvl>=+J2)#1WtWiSwRiw}+z!L4Z4=G)AQhxJPI*lfryn2eb_*&& z-kf@#rDDt#YrqHHhp5;5a*4v)&TB<1zZ6 zz4GI5G?YJ1Uu$wm{42G-jnKqz{z`+#q6OGn&3+TCEuz`Er-hh}HXN?C>}AO_$e2_- zY)~CHfxZ+Dziu>tAb|;4%J11IF2o#bH7>6glqUXu<&oHE_b^(lcmAR9<9B7WCJJB5 z%r~U59au`-&%+ZZK&BPtJG8=q0RVaPwK;J;e+jtYOZqo*hmofl#(Cw`s3XNOHj3H- zU#n+<SPOqFoDzCgu$w zn>ptoG8LFq@|X3A0QHW^WwH%M2J_}HayAXxF=`NDG5u+-JLVWJV#Le09NXu#J}M|@ z^bO35eAY`_BtWZLQ4(-lfXL6=fn{r%_&d5LcTKD{(iO~I)eV)^RocMY+_oJ#|J>pD z8m7}k%yIwmx@~le)T8aXff{Y`<7Rx*FC~ATbcu_9us1UR_GYCf%x~!^%SIkZofPcg z(pFj%-y||ok8Jpy!#{)(qQQeKf+qm>CNv7X-1U?%B)|dtOsQDf#;m)CD(c6{3|suq znUwPl?-kN6QkSW#ly?Cexucs4jGr&wrgzl*++XFuUIQ_YEg?ok zp6X=zu*ruK5!DG5tOPue-rf_dr`i|I?JU^LmIROlCtqUv@Z1>akoppg1vyn;HWo0N zYfobZM`2O5CMg*?O8A6iS4Ml;WmsT8@rfSFBp$nxSTt(0*a(IUQuL$m+6@OdpT6>3 zvWg$Ft=t{;qI#F2@z0vu1^8vsBot{|34yPXNZ+@XX)DEi0Z<8;xmwGwHlP?2wB`Df zS7bA?^Rrsr>vZ3Kx_1ZdHR3`KX!(53wA3}K zx?)d=hvwAl^RSL84<=-p1Vz`;?cUW~Q_UU2?DX7jT#60`Znx6SbUN2MJZCJl`7$qC z+j|y9Ynpx10)#s9;nhBu)i<{MRT{hyB`)T(l_F*NL!moAUh>`7sn1o8-8Ow24m7!5 zUA|uWH02p2fc~I_KIZ~j;a!7Dv>(_{J#~eEJIEJ92ZQs#XO5i`#`PLSo*WneJbTZ?v6X)nnbKdG~?+php zE&gPz8EsEp886jZpb!^$ZEo1|Z$^8Wamx{nzf17-Jm#C^f}t%vEzvuy=H3)cg*22W zaLu>`oLLthmg66HPOk|Ih%QFgF+=XIPGz|c%S)+wn=((V)SzuKx5kAj8Fz0plx$RQ zV?&OL3o~7Ou?OrNg`KW`3f*Cj-jRbSTExcb@`F%uQfa2C5LiR9Po8np&-t`_25)C4 z?l>^49W=892MXWOjg3tw;KbRi6H%+kZR>UzsY+s;Qi_*i zpo^Z_u2B1{HVe<_!%z!TVNtJA({AUCY4!}Qy6wmS(j0PY;f~W>ZXz**Z*jn^|B%F{ zoBD0V>edAtHHf_8Be(2l1B6M}Qv<(H{kZ5;?83#N*4ts_REE)FmBwl>5beT~h?^;W@2=B;vzg z_&u!#v9vj-Tn)Vc?KN68<$P|tux`EX`_sc{tm#-6*Nd1A0)#00O$M=$`jMB-=5Nse zJ>1?bAL~~G{>k>np!;>Pp1wP%3-KqUQxDJm>mxam%e z*K94-e9`KpUbpPnXs}ieiSL4GUChN?|3`!Q@fj2pW9)R+u%{phjUXJmX#PNXu=x!4 z9OMsUsO0b=IZMXeL~v@?7`Tz_Medpac?MOV3?`#(itu+@V7gKntwgz4o6p%C-^#@3 z^RJ;y-DZ$RKZfGS`wB|OMixn!nNj~+LJSP~-NDaWD+u1vXMe-01~#d!RQNL(u!-PV z2L$S#1e5(@8M#Pn4AO=3=^lTyH@$cDW?#BCA41*@J>pz$b8RWldd3#pJbR$zX4T1; zK(@0xw$rj3`LGKfY;hn@WZ#`@=b07|GPUP`wVe)fw!Ej^_{NjbQxTkSGRT}eG*ytF zyL?v$a#1Yd7rmO<_=f6|9Oc4-F12~2%{;ZML>-s0a9@4+dbf0Dj3*eR6+1i@c)BF$ zIaklSGLYcJgJGOvPl%81c7CHnV2IZPG@vO^S|`2M9CRrU5E!O8;tL=#oVNO{Xg_CJZoGMa)_b>NO!PT|Ai1 z(iY8YIIeI}uq^v}@Swd=9U%h)#&nTs&ScO0Yb$F1Hnj|YP#)~O7cm(#6j4*KyFU{7 zh0x?JO6y54x0ud)TJ;CDdhr$e)tj{H9t2dWe-vMy=?j@r1JKW^W4ShhE8V%YMw$X( zLyLZQsua_m!GX({9z5(G@mGe!e(rie8vl$NT`6R4Ox_EQMA|avMLpfm$K=+V?OR$ew4?nz$VU{Grjcr08drq}GT8&)C7j$Qu2nYEu(g=KsD(P<89GLH>=B?`m7 z)cRo{kB*9Z2I*FIzM)U6W{Gfp0(+Ji-8Rbl(h@zweU1=+w&Iu1<<3$dXWSGrl5o}G zug6+L66Vmto-wHpQpOvdhwrP;+E#A};x-D1EN~s>;e}F^=2qI3D5)v5^1xl}p(3jb z>SyU480BCb$vQk+Wr^cN#*=do!M`I~WJ3nE?THbR?)_*%MHOvI8d&hDR=+UqG4N_z zWW4<`AFN7LSN|<;t|pO-f6-jFh|>FfmbiVDd{CS_yh^yI!d~r6R>Z)gR}OjsC0ni| zzVVGaHU+~wGyKPYZoN4sXb6|^wJMX@Xs0q?Hq@Ux0M-tRv~#{AD;{|d6UElhYe?Ta z2fCC76{gY8Cpl&t%gMa@?RMVoEl@kHToMv4x-8w^V{PG&x%FJM?GBc%M4R#W->6JC zYAkMFOsfx&|L+Piw*w-@OAdrU9akM8E8Pj?)`%0ZJWHYvAmIvR0%?adS~I9NF_(Fh z^D}y015|eLe!yz?$R_hVAopOvf}_7Q?M4bUoDaG<^nVysnddxSkT*O(AHJ&K6&fkk z$DmlOIp@#zrQ{mp6H4DKm@4%71iOnTAO7J*fSHyq|Na(Z^$o-4fEG2k%Ah8U(3Sj0 zPH1ZCS<~$JfpdVE^7lYpATe(0z;yKsO((6j(-sWiWO5=|%Hr=`%dsPge|(^R;Gdv8l_RG477WY^)ie(KmIC&zkDkjTRkhl7gW zP8=(78~n*TGzBl%)Q*)djOCl$jozusc=QDbr0;uvEN1o2F9dV__YDZx=}R7c{e7|( zF@Gd`16uEV?J0OtDnz4pjkn;e5Rg^ zjA*BU$IdJ{Z<=@=I1@$|9QwBQjyjNbpg%ll2&wERB9BHtD|DfO8TY*~MtsKHn8rbO zU&-78cC5wqOVR_@lnAq7*@G30(hjccqC~y(S<~7rJh6D;QM__7PuoLhgUBa|M!%AN z8<5HyV1`$Lmv1j-g}3IAm13gF%o3;kIfFEaj_9b5*bH&xx1LWtqn-n_@ap7U?1`#m zeHYBfTan?$9H)D@x5@cdN*!du-l<;COXVk;>SpyBpS>AMaYSU&3`Y5KUPhR$n&%wp z%Wa5%7r{w^Y+}v!@{Tn$5+b~1S!UlSIuGsNDVG#1<4~(5Lq6l}{pnSC*f2VFmS)2? z)oN1KjFuS3;oXk+@~?WKI=7VxHM+QYD3wVskC#P>2TLa&bI37I>%;r)SBunPRbU)s zaB}zA>qRW%?SU;)8JiyIthsE#WY+b-wbhf=&APrp$DtjqOXXQtcwN#Hs2&656610G zfpM+ukD*&8=#LvMZZZB41zC*{Wq$w{N^Koe1?HQD6IoM%GFD}U*jH^>@}isY2 z%%%*P9$xF+BVg1hPYEDjWCF$U)B%z72}kv z96KB!bv%yb)r*6SLUE<7#iCl%_z-9Z2X8ACRmHuUY&O#>F4hdJg+wO%tE@@)RFX$`-s>SO2|Gu(6 z12CmucXqm#@OQ82iJzmIo{o&VDQ1XHHYEO8zVAL~MiQDJ*+7GM#GxUT>NWH?Xq;K? z3T%K7v=7*s$}^i77Q}RE6>|Q>zgm)9l0_z?8TV;54({;mKxN_gYsTPlyeQIfvf)$` zvX?&8;u;QQQ)sa;07EXP^ey653G zJvW$>NHA5X4|4bs&!jT^ItLLmhfzpsl^-8mvK_v1ekot8M(mtIAJ4)7EbVJ(U-qew zh_`;Ypgc+$yQSmz-A~j(MwVb*DZ<2^@v1gOo+o5!=8l#AJZGKT z=*cTP-isoIBcbw0@+l$Xy}_f zpG4*imLODIXHhyC#2UlRLFJ7SqcdQ9VdZ?4n3Ijxazia1SuWo6i4wL!PT6D-VsCq& zKz(|RGMOd2@Y@wVg^agvo0);exqeX|k$|V{VqJ^M4b0=GJhO?u99lS5*&inu?0^DD zg^4$!pHY4J{>hFBczz$_F=QKTZq-05wOp!Ay91@W9pZnSDjC0p{$Pyv;%!I(^weCc zM4gv?9k9LffF_OihbTR&m6{3-)^O!UBy;?YZTn1pFQ05$hs3)&<3jlcodtGwr$u*o z+7HM)(Kt!PtwHW3{pUZ99*_}_M|oj#mSH2vB~Pl>_3VCl=0>a8F$4QU1y5+li?O?O zIol+f7f`ms6`M9lWolLZraBr!ND)71;%m`N>u3~K2)(WliMXj8SP&A*XeFQe?M)9dhj2+AFU5WzhQQ!j?vZW$4+V$-r=Q1$z^Kw6Djt*i!neR31Zm?3)xSfN_ zVjqwP^=K)#C4M8Yfx;!|dUs}rM6Z}vN9z!LaR7>TKK?N|w{UL`0moMsC`1woYP_9p z?Y~4@ZoC~6j20x@u571f_5Mx~v=z^&iO?%{_F3j)|FUCNj_YS%H6_LK=tX@L-0efR z)i`xMAC*Hx(^SRy0rfBi;$|g#47?j zqlgz=#*2u}9gff2N#)>TUG5Evg&w$Fo}S!U{ci#(=fhjL$_ zKcG2+%;SZH{9IbLWqSRDeFADClorcn<$@XT9w+7Ejrxkr6>3Irp9&lnwQ1=2(fnpi z0vJ!a!`s6dBQMelhaMoCc|cU{>S8v975JHq2jnJqI~Hn`YQBt=fQcc==qNeueO_O& zVp`_^S2fqO)pv_Rx+A^3mU&`_S4u038F%+-{6}EFW2sjee5Z9AFvtZxeMQ4 zwuct93j&pVlinY0%2X+R6k~=BD^l?nwwqa_?lyU>P&+w+QjGjN7EOF0$^@V@Z3EPqc`|X_{w~j9Elm# z9a@W;@!YF&XLVCGQpKe*l8^Ijk%Xl$1lMfejmcBY*_dz&0P?h!!(FpZe3J{l>@_bC z{~~--6Ky$DW6`v)kn(@Wev%{z!0ZGTk0)!vtFcr8ZTJ`dL9WObk#8+JvsjT?t?s?o z-cN4$HHYmq7XGrrWBe?6)Woq3bI+@GIsf9}Fw2;}T#w5j-FSP1Ds}U_J7BU!DlW$9 zJ6b{KRZd2*hem$=A*6D9dAZs2qA%1i*XMM>Z1Q0u-eW3Cw{_S>^#RV*pxOJ)()Ot6 zbCHf_%f+AX^!Kr!!cESAm2lqwR2g`Zn(m(~4ikO)hsGeD2dVSiwI_Pnx6}K4M*Xt5 z3m$xP@x4?!(Ye>V)h6{LEVSxpdT*# zlBErFovjqMj4Oz4)$gXu>1_ zkC(G8Y?$%d~^m56I@%*$jC{A#)46|l1%Li zotaM!40kD6pOK77kaj5W)>-9`-N<3M=ZNtKR^@(PDABq?IK>uIv>(oiG?*lL_I>bH z@Ljp~JzvxNP!ZOt3Q;a4HPA0QH?L71Gl*2CK4^g{)BC0wZ?B*Jpr-kg(*rY;hH?@^ zT9wsJL^C^}UAk6zooJ18`q2*|Nh)+hJ?v&t&#PZ$K4F$Uur%Q3XRRPw49@|TdDCP+ z(mq}c_wLtk=Cyxtw)eUR%SLRx3)FFU@DoiJS$XLQ!pYdZ>09CS9-t%faqQK-s-qlhVe%lB23i=F+maHnU zzWZwpaoUtWJrzUPnJAba_XyLUKA6T=fk){P({zu06n$TvbUth+qR%lowa+Nav9CAp z)_0vlMhKVBCyw%{1pCTnET?CrmPF3XZ#UfQG$#^K*T*xK6-zs4!PZ&?(Al0OXfXD0 zO^R(W`D{mx`EWFYYYWD7JpeBDp`T1v*KEY8m(4^Ye0QJ)zc(6k8tT6^6}ddT{E6rH z=8I}YWvC$jG77A3K5kAU2k8`zd}Y&5~oaoHq< zSA9fu0q`(n(m?9e4h=z!Q9WrIwm|^DaE4O>gmNRQuPPMsFag@4Jle z`ilK`S0Q0I%(T6kL|Fq^>YnZ1(b)C2gX|w(l`Mfku6*053=tBZt0?e5-BWWIkhs2z zx@Z7X3;yfEpdsPke^$}^dD3o-@9MTXq}n|#P>u-k5URdK65ifOKnzh^r4#5VVM1pi zgt+Gp80hd5IP7NmYqn_q$8O(9w%;a=Xv8LSV?yA1TzbjcSvpel7L>pRzZg{8gCeDx zyA`*30*f%LXhYB{pn-Yl;OF2sMeAnx$!3&#CS5f>L#ZLc8@}qTlHq&AlWx)$sJkf< zooFhPBwgV&mT)in&Am2tpL=k@3|w58@niq)9+)UT-^~QivUJ4Mm#SNVkOQkkGeBC* zE%OhJGI8^609&>4Xl2B(k7=iLP%fRmFcZ{5_(Ad=?z;mK1R(znu{zRypVQ`!+@= zva}5=K?YHxAc{x^$&z!BARt*JNJb<{lpGw8AUP_MlOP#~48oA3l5-9M!T`g}kcTwz zF7`gp**@nyXJ3B5zr(M&SiO4nT2)n>K-JFd3^aqa6Y1Kg}HVJVMrPBqInQjB6rDI*i-ZJSBnxHe)S9L^a*GJt@*i7ncn!Uosr`BWpD3)uMI;9e1x^A%`FA zCiTq^t4aI8#8!b7@7Jn>up$hJMeJtJzYv<#z3{T{C&Y?t@Y?&jG(O@pvw594_BK0# zYRz8i*mbtlrHRj)U@vDyzst2)Z=vnwT_8~a7w$+L*VcM6n_k2f&+1Z5v2C_{Cr~85 z*m+!3wC^*1&gZqL@%^S_?nVkIu@n$P<0l2^P8+hg;x+8iP2kgNd6BA(u4x27ciQLa z+aNyU5nH0)sD>9dmUl8vQf-*ZVN$ODB;f(1^W~cNquG1Ym2h7yGv12x;NCzJk*%^j z0Nv@yE2S67;%>pdnx~%a&aFSI%Qcq4T&!EyDcoC~2}>e@VuDXyVnWIYbOV@}^f}k=@3JrxBQBu3TSMO{vkd-CEE8nIL*T5gIEA!M5USEuTL$?+Y$Z&8$Rm%Lx;& zw6q3rmqf2N^Jg|(AX9yZ501M(YEWg9g_hLVV>DunevkkGavw9j2qUONU;)$QDk2wE zNAZNaKT&tVnBT;-|Jzr+cSeeM=o={2GS}mQbJY1op4Ov?K-!eUv$c3c!vH zC>aOZ>J^r)@B<@`6f0)qHmP0%CX zwRz^8#Xro?D|C#mNZWW}RaNiO^4`6F!SpYa3fElNiX2+?+uLS? z^j$#Oc5f-vLwTdZVztSyq#ifdW_fmEsX>OHj5)=Ajc-qBt@64Y89!Z{XQ$D3#!aji zxe-=*H(6~aiBDX^6Oa-Bt^iwBE_IC`C+@d64%a@Rgb9LopX=$U$J~VnYfhB61-E|B zX6k`Eflr9=@%Mmqz`3JWbzck400cRnP;`pkjj$C7jSoQ%dIC#i%b1}YOtlGsc9$8L z#q+3XFbL7?{V|m%2F8^7I!tp1q$sQ|O}){)DzS9N6;z)~b4mX^}Z&GxYm#sD;%U zySIRC!8OXN@j0H*M}6|t2<2HWEg@}GJL=9&+6f>%qUpUAi4SC(dzCqr3NZ0WnweS& zVlr(hfdX(zrM=7=C1KH}a)z$Te-ss#WL$4mN@L!70e zaYY-eapyO}kuyur-(dRXqf?&6utl2{F|*0G%>opHf{1+P+BDutozb)HNUuycv8&CZ z0iuCEE^gCo?%M@uzDt7>bZKyW4!6cFWa=yj!PAG~UrQ5woL3};yu1}X#YgkFigWK! z3k+{PC=ZkrT0KstGyCW?ezQ2JhtBtreKa_;5nP=9EvFH0bH7*TYb$^q$57%LO)_RD zD7x%5S|Yq%L|>j(utAsJN_MzEIm^R9*tEY(roUPgN26VRXg_9}oFm=_LwQqeGu_c1 zTSwFVD-DSzE`th{=IG3xS2p~bF9^+NR%)p0b6Gi=ZVVv^W+>xD-dASocJL4$OwWy4 z^y;RJYC4XRF5X`RWBLWrcZbK+aiXjHow+GHqbsu3oi!<(WnYMkIlPDGm*@l^EnQh{ z6(%qQz`AT7o53%pwuc?`QKfqJJ|ga^#D!cQi1REp7DxZ(XkKgo)4M>foNih9X%tU7rm<)alp# z!3MTRfvZre{D|s`6H4cP05_5CW%w4W0HU}|JZQA>jM*^+@)AW7Ln;aoH%UFytlegj zwj+BEWY1~?&!Q*)*a0>GJ3#H|mM6L54As4H%cDTip)0MEBnTjzV44h%e`6Ip%Dt&l>nF~lIzt-S=i5*GWgRBe6j1YG8%7fLg7 zVLnSWP?jMi!Yw}Db$p47xKLlDMs)ee<@$Mc&q8RHG4*auqYX314oSI{_iyg_6E|!m zg&@?HJwGf5g3hDTR`*WoEKZIq>jqxBoyQ5Trge_esE|8i0$Ao5iQ zYiFVZ>DuS8tqXU5$25C*e!UIa#)OA)4$}{r@TGR(BzC?(IaP~M6b2qreI? zQEaPYHbj0|YDKO~&8C|eWOcKiq#&b;Mn>R>Z?@0P^fd(efV&{q{rm5km;IrIx4EN{ zl*=^AzH#nVe%2=M`y~$cD^F^L?LPD~rKquzN^#h@%NS)}r4lj=hrD?gZoFbzu`(YIX)0?AKxGKY)EHm1fzLZBsDV>O^NI;~U#SrWAkBzGi?vuYx}xGG%LbDDW2w3C z=k97?N5^%G_x=q-z01thD&4J530nZS39(lplgB=okYyX6)IHhz+M_T|<2_5I=VnfD z5@tX9+e8#S%kR}_KE(Q+d$p)ZzNDeQapq+-gySz{_u2g5UA;LDLC#}!C)(h4IkwGm zi7qG6lRIZ>z$ZD;g&&(BSk6nIdH0I?V=%A zx;?69I}-25Vt$*;clL1PG+5Ug@aGExEk}42AB0Ed=}uiH67_?~5J??3oKFu?9`9tL zkPYmN1t!pNC4=f`c}iZ)wHRzq) zZF`e@wy72-eAYAp8T6e!8x1;NG==0Iyz^X7q%*zBx4%ft>zTg8{j?!1CqV%14PU{= zJ9nlH;vbX)P90{sr|`2c9?C3Ui_)Jlrjv=+H8vn#wxZdZJFX0Dpy#G!m;w$UI7iw& zNh}CqL)}T*BFd2V&-z66M@iB3#F|U@dPa7; zBGPKDB(voo5!2WkiL~BvQYWvS#XQldve|f$uN+7Z zxdJk4XoGCD>qB3YxUaHfr#nulzSe0t`#Op^%vOVnCf1+tkh*_TVR$APmnp`Uu_8Tp z08dDCOZGjo4&FN(N@Bc(N;D1%lMk3X4wd-_eCD}oo zY!Qk}({4WnuBH!0?WmtzGPYWd$QphafVy~#HQV`qpDsq`?+7h_t*PD>Ex-=-`Ee>g zBL#R*Hr@2`;qg|@Rm}hv%SID}Wn?^d6SU=oYP#6Qh2`>~g`?PEBfdsd#!Fk6UGl_7 zCC36B3=Z1R_a$62F@9uU-h}08g?EIsaS_o~Kg$*)$xis;GWK5bM8WJ$BOJCpF5)X3 zOnimx&^QF2T~C@Y$@6&LL%r2xbZtR=REAvrDqn&!`+r%%I8XPIs|m`qnkT zFczR%F?9pk@hDk0m1T;Y<~~yb=8NRQezL(OWh+^YF+kb+-4-%A$>!_Eghv)r1Dpj^ zZ@o(;a+DC5lmYAX3W*$ZaWA!mr*dvzPoaIdGg=`a)n&drqj%aOS7y6DsUJeo2!UoVK0QW`?Y z+*_B4&oCuqII5to9Ky$=8Ce<|y-My?N_WEY3@eEcu?Og zge({>+8m~?QdXSn<96S6-Z=!-0T8zCx&epqX)!|JP3FwFgUQCwKLhd~;s65JE7*9U zcV|7eAZX~^Y;RM;g4FYOMm4dw>9d^(vxWCmfc=xqgT^+l$i>J9AU9`tqDhD>WbjiX zKZa-7^OVt?oD_Y{{JWFGdYLA&F?DbGES7e-1E7(CEK%n?p1s{YWBKmo4D5=%HD=ae zz2cIw)kwG&eCnarnr-1K9GQ0XBz@t4$7yttZ4+=E_sJt;F?0Cxn0Y<6zBo0Skl zkXgRT<@Y`9cl0i9F29WM$Kafo-M5i^|5je4w0HfrKiXSZaHZP1`y_M}$d@(=*-;iYMqo82>$Q~IF%`By{HMr)?~hna}hn_cDC*0{#}FA-)CmI4|~PJa^mV5*Gc zkdj2Tp$5?1cf5p~`t#(#iAS3(-3@o}dG=^Iaup(lBUi>~kyvRG(xcz^axXx*zy;Gr zx1@3porUuDk3l zZfmU31QXwV_m<9|<=@?4!Fr7C$A#)upfL09T6Fim!;d`dN2!S04m52{E=F6a56hY? z2T~F7bp1wtTd8}Nwgz^}C@`ALB>4^Q^R4j9oe^CqwuctQsqc?_I3{9}$%ur-pDYhl zgYbJwl^8}S5ZC8-{wQhh#X|p(;bnj{_J{jNSH-Qr-!K?^eGv%WCpAtQYmMpAYuW+! z-up3wYay0VT>a;H3#F0pfF9zE(AczR? zlMHw3Jr6#5aiU819!R6iSr)dpVtti0KC%jP)FwBmKx*swUc`E!6t%= z`omb)GASUKNPKx=mAmGJ_di>d=yd>6+c&7%Sa)vaYu;)k_pZSI(El~Dj8x}jrU>%2 zmwk)3P5h@lP%kd}c{s>IzwXr+>|Si<#UC_kKrtXLj?-{>&iPbAZ@>&$Jl%w7(l9VF z;{q+@v@RaT-KZ(Mwg*7a2|HZMq|GzPgRq54LA<1R3=?v1Lb)5bte45`AI@CqkjDh9 z-qFH~YX%tZpdX5ol4=uMd%m5(IkBSwj;S`Q&t(Lw0u=Ttr*+<{n@j}ifE&Pk)K4n2 z-g`>)=BfH_yZ7xAXONxdWTZAJP;BOvJ%Bt4wdalsYr^S0F?&jB590X#=$5}rUgo1l zdFX~4$QG##fYc8vO!K(qn}c^!p1&tt#;!7{AqbDL7^8i`volQOgIi#;5n*mqxg~xI zY?D*GaIF`F^(Y3aP?YLE>Gsv6D<+^v3X*)r1m%5TAq?@z&h^@@8X9wR4p~v#fnLH4 z*2r0&488DrhHZavI$g>DOT9wDmvnl&-aej8!+)c^sC=QQ9t(a_16Ed}HD= zA0uLxIU;eg&&tMp$reWL1&j8`INE&?4pK_M`>o*+2#?N3kJ#vzLyBAxnb$ zot=&UwK;i<^FFdAF0e3&CQgp3OKOIRB5Ip)ghHo9WoBdkv3WUxP5a#hk~V1#+z72h z0kiT~1^UQ=P*ar!GlUJ2k9z-ojJszg(bZdWCd}eB#9@zu_J-e+AiYX~bUs7A;91=-~mK zS}8^DJTKjSPd7(HLxX#j?I(!#7Ze;hbPjw>5BMmm&e~SQJD#j9xam%eZTq6bkiD$8i$WE!v}btEMM0R8 z2jeV}EqJ|DAF*#=J$myvB6kZk9 zE_)@g8D8O?H7N9Fs5q4^GG!>9PCa@ICS?D{#Cf$Z8Tijb3IY7J3@JpFT{biM>iV$x ziD3dqvc>Y(xTh<@MGllcEXEx1SDuXRayyJzrM<2P=sKRqD}I;o(B5IUI2n5#Q14(k zgb_3wsc(eL^zZi9ARMX^*Hk_NnS5Wn7t$|lRs!&D(p>&eCEVL?PfKZ_5ZGb)VIj87gPuz()z7VXb&(OjWtDCLq#H33T0@5|+351u z2qzBtV7dx@@(U()`YE-W57w!E0GVX+6|iqyPsB|F)?t1pW966V52VP}%V(IIb>F1$ z;(Gi|zq#x@{pfl_oXC?a+wDg|P=Tmt^8!(i(U;WR@FKZe*&NI3dzU4A5}NlO7d@mB z^xZlHU?evZMXO(ZH4B(YQ$$ZQ?&@Ve;SCE6%8jtrSmBl zJyavPr-7n<_1caE8D6uTV5_<+o5g7aww=_N*Sih0_o*&K!P1V|3eW_XgxGz)KT&|OyYH0B0J);!@9Dw_2Ha%SS@V~l}{w!Btd zjNcQvWAK+9a>7iIDI>ftCpg~|I8&*J#DJibm<)$~LVC1^$#WQ~#XE)CcNV?z@4bq+ z0S1vuA-!*}BLvgs@B{YiY(8`=Ru{3FE0V2$&(DYhU=^vd9ucssZeR63!aoBCCF^^p z2IVp$LXiOe!ART$sNc!-Wa#s(*Blko50~}~V3W^lJA=atpjyyrje9^jhU|okQ4HTe zj^T8L1vyiJd!X6Y#Wz{j6(98HYBFWN;sno0OyhiMn#aP$zFx-k( zm~yPH8#V1v!mbmW*H5l6TsWe9O*x7;#olHnPdAq#xuBI7*P&$uU$1@!>Dj&r*B%Wm zyqWjW)#geVxdzC3GTuJp6I zs3_aw>f1kqG;O4&`=R>N#rBW`Fy>0W(*+4w#%Y}u5{gPkrTOg9ET){FI3g))r;brx z9^sQ`Ws=S|l7b3Nw_h38zHW+M+3U!Oxgn{2*r>sN??Cpv|q?L{66TD$!8vU-jh*>A}!%@eRyqcqIlk!3HvV*w$! zJ1es51D!nCY4T|Aq)`(|64ldgC=K7UB^)J^q?n`@IzM;obesAS$ay}*3KmN-SetPw zX_A<5xtUkANEVHjl$z08Ymm}DC7ES)F?OgL^}xab5z|V)y>b-`XOn5D`C@wPV)r5j zRyBHJmEkKIGdnk_7H0;}aPPXaflVcN!>57lzHn@JSFPh>#`5XN!osp`!ys9aB`bg67#T?kC}ZiXpB*6 z{q2(UKDP1k0BM%B_^wwqrjsC19!l^bcZ0dp+x(BMTb`|_vPy~Y$^NYn5$+jW&ZF?W5jUzr^nrypLsyJzpW^Krj!a8akE z{v-%Q>uc3P5WH6Z;_bu5!+emu!2Otr$Ou1pn`=9ZP@ednYCWt;^~2q*aD-eI@8EVk z5_2LHd}Aoxbvfmhf2hl#NHIt;^43=PE|11XtB5BTDR_}9FUrd0l6Dfrg`wwTTGy~C z7;vF4FqE`)8*5U^$AT*IJCsuxRQ`Q-E){)h;dE;nyZcr5I3L`NVj~0kc^%yZr$MUv z9jwmcVf*ZOa*))7wk(JZ`1sVX6|wJj!nTPYu@eqTpk=%r3Gsd=yF3lDXg+9ZWDQU# zuj1Xl^C(QJ8PU5@j*>Kvw>cxbz2AI(ZlXGA3|!;& zWc1cPWlM2uN7<|B8XbRHW2TTA{AsFt<>$2jNK|H>h{W+vE%{K!fN)1JT!dAog9O>I zzTAroP4QNpHr0Jb?*ADyFP?_|I4V}foQxz`crS7^!qg$nwo9MBAl1rTg-CM$Tdb3D zTK@8p5hiX?tvBHyTEEkJb)_2JMr0TZne|i!_|BKgcFDXrxm*VBPP{#nesd+(QOO`^ zfkAKgM{P3?N0QXjI*?K`Tmr={lS}f<`$+5}DO${7Csy0(8 zcD5z;8hmto52vrP{-&4rcn|r8tcb>LTrW2j+S?+#w6?r;oyO#7*(6VVrQfuL<9^IJ zIe0%$s@|+a>T#%>J8~x7p=VMCEw%lnwL;|i=KF7x^4PKrl-5FxHcPuZEIQgqA> zyDrI^7o>);$76u6^V|6W*=4ki!QC4dNbwmN!)h9qe?tx=PxvqeYLBx zu=!)}remVCWsjqHt%+P)NE5x$wk?X<{sX7f0zO#muGp|`d-=(s-k<=s_HjnpM^oJ@ zi) zpXf8+pg34>-IBLvc*E@;=QSQ#U7RB0Z%b*QfDUQrJr##9_d>pxg`VHLnwUh*x|q$< zxkXRHyBYM;ll>Y=miI|A>UL##X9<PvgH*@?{t`%`Q4$edf;KDWH6 zHpon_RuJAezAXrwY=w;iz=ZI(J#FHr#!%(7m=ytuCK|u9BEVb-)HQvow2RWV$~l7= z`Hb{@?A7WMy|a|g24q~7u+5A*q{m1edY&tH_J+SmZF08?Pmbv_pua`B6eZlnQYvPn z7i-tB%LCOnWS937ku+gRcO6Nou>`T5dQ%}Tl#n?)-d`>IH-TWFd2l``0PTwYJbruv znYWE%uzz7vv({V`3GXsr8yqb4VGrrs^8R_EedZ^+7!g%1mGf?~)^0w*H`dAr(LWi! z)5K$$k{-!q856O+lE69ZoN4@Qzc(dvj`tDy-TsO3;= zMi?{&l`vlUb8bwa81p(L`J2G>sbOu}`#Vr(C|*I#<;wc!66>!-9!mT)h)SXUouRp` zp9f5L?}kV1tqbt}A`M5ta1`tf%foc699(E#rD~08)+lZoS_}iSq4{SASFz|GI+bxZ zHo<%zvXK&R6@hfBsp*C)k6>DpbHk|3VUj5Ea4Nc~@wL`3RC=jUE4Tl7G|K{t z)d&mZB5*9xD!T!NoE@c7m+;cFf9=z{xA(=!{MzRvGk%Bpd$pOB#OpX$TOn$~*Jm_X zuEH>))SUKAe>w@5uaX7wK!9iZjL2PaTOrpqv_&vjJX(|4Lilc~dvOH@YTqM`M(51z zGJbjWVchc-speF!`k(Er4j~k^&!M}!M0qZQzApN;$c|--#2BfamnS`o3(x2BI2>3! z;hDaIMn_b)BaJ^JH4dis$c%CP|8aIe#1B}v2Z;t_8D}^T31QzBtvH|9o;w->vn<3%gCfxh}F4XF3>C*GgPb{}m>YL1z|8lWFDjKGzhtwO$Q^8!nlSA2!&VGZDg~rp7 z!7uvB6QxoD=?TqTjW1LGj*e(N)3Sc}mh{ z#i~d3{?&CpIxN2k70dstfVsN52d+Tz`^(o9)ok%9Jb2#Ee+LQBghBby&X z$YqKs>{6N|n<3jQzY_Wk2I(vameFzL=DnYro5`8W2@ysv(Y+?hoAp8pU?@C6i`RTo zbCC%_`-?LF@>%|p%Ad~)FtR`N|F`acExcR&0IiT_k@>LD9`|G#Vb&)4*Sg^L57+5h+X_@_dxZ- zcfq6_-hpN;)(adE-S|AtuJaj*L=p-IrTm?E{t{TrFyqkTVrIr-4X2xjTkbf!?!=49 zJc+65+){38(Ko_h^y>NhQ~B~KtnVuQb=TiWM;eS{EaCdVezuOUKUqTT%O>KBe0x67 zq0eGd6>2qcw9Pz56=2xWBjAr>{H-x&Nn*M6Jy7*2gPW?g+xOdiOWE|8>HugcRCTXS{E(S`pGCZ;aX-7^0Y%!-FwprsCzCp9#I}{_q`7ZT>4`> zcm~+N)hTg7NXtN@&*-_XvC) zc?p>h5^}#Knk)XpxXXtTP?s!r%gyjykDX=Lt+V5t+v_b>qRdm|g8fSnBm}ZN5xEEJ zS9Y=NC*9p_8OTxk8&mSvB~f|WHp38OA5Ln|gK_$yajAzI{9mF!L99I$aG}|cUK?~J zoo@lukUJ&>zs|;A^#5O_H$#ym<>8>8hPtblKg%tMdfAN&y&EOwxvt^Oln*oRUo`1e zs(%~+*=a!iHH^RM4@7JY>Y&E5S{1_{NabUp{(vGC7V<0I8!2yvh~~T!`^{nXfI#ja zhqV&^W(xdzi5zII{kG`I?^zm9JPGj@TTVX5Zw`iy7;sdsXn*|85AcA2c1ruP`t#o$ z%rjb``PCw}-&0&#jmLlii_EC_&A}8%1I@9XX_Nk|;QwzKW<|a=)x9HO@wfAnnrYvKRX f>hu3E3x9z}9&RGdL8Iw)1^82vSCfOtmu+T{GfKW<#rpEo=ifBLHaAJUIDgrDo*-2d#f+mvq&iFY*tQ9Tt?=m8l8eUN97{Of zq<0;{0sA>GvmEMvbjeaDP_Q!^dxLTuQvUbOd7zB{`k{=6AmGv68n@tqC%N^3rPd?9QI8Y3l^Ve9M1dBD%2sW4Og*o(U@$JZ$rLwu7@qqKLfz2QD} z6r;p)Nrsy;A}J?tdg?T~RL}FEuT00;``0yeGdbE9369Tp9$kKZ;=EDPuN%*(W1M*} zpT1MNjCh!$6324{Blyis7=GWQvGeH2ujI#Yue|o^^zc1Z`$57}6Gob_dS-TSYq`8m zeRx<0=A69DDF=yf`?jZ_{qbD!Tai46z@#a1v0c4&k zb1a%zg-?+bdM=2S2gGkFaoQ$JcG*uM#L9y(a) z(E3C8R%lUl>t+soPp%MCouMFBh!>>t%pR14%Ris_wT(I8xV}C$eUCV!(_8;hB>@=( z=hjltpJm7LJUbxs2aDAEM~WwIAGW+YK5B+PM82na?Gd}vxdYE1X=}+}7RI3U_Mo#>W`YPyGBin<7^q|+KwgOj!5l9meK8*+!o@bBwgV6J)Z;l7QX34y& zyTcy0^@;WVan(-}bcUwZ=Wth**GI{&oCoAf=MN-)yXwFq_w3A#^5zrXLgUBSqrh4g z89quRxs$E1kgulK`F@`Ic{nS>+x1QL9i?yWnD+t;M>Yv!%dI<8=6CwIsI3iIYWG@F zs`q(+ww==8ALr>>Vzq=e;@gToMLuLmKM~5j(F+O@HB{=6E%q z=-}=n7%TOkt#!d_Svrf(8YvoO7}?(ogMWa)+d98zNM*=yxH`>YR90S|cUZl+zx9-q ztC%a8d9}he9pfZ#&$6$LzU7ehqynwIio?%(?i$XtZ!mIKr?UaAq=e_so z>iKrCPxadMM_z1d<6*SVXOa)_or#n8K4ABm{os>3Cy!2FJ)QhV1Mb{_SJq-Upag2&M|p38sL4Sjv}BL9nPZ(iTKcp&~9tCQ#zA@OVG!n5DGb9hIh_;rnhd0s31Vp}^& zykU0%dFQy9Ucj9T{j35nO7F7qMz&;ebRYL;wR$9THR*%%Ly_8J#oERnlDhdGs?+u_ ze__dsl#?^<*yreIBX!!KrC08(5}#y?K1RQY+x;1Cn|Cla;DL~jaMx$c>H{`k&If78 ze9oULTNMdXwTl?~nZN43rCra*(0zaE_)*G1cF=uhl#o&Onkd^Ycl8NY2-US3E|=k%fbR; zwju$qL9YkCi#oabo>WbdI?waQ;D`E;*6-jSxbH~uR8E6vF_AaIIk)qrvOTgDOk_;e zE+Gr7?ztB#E5@ekWavp}51QcakniF@78Iluz(4AsUJT3?CKN22siP@(ma}UsYctfq znmV>w*n447w4OXr&P8TANugC1U1;8zUdy&j{=%f%bO3s$u2 z756t#ivq>lU6suhdT*ZxgP153E7YWhc8;WSrA8@wK`0O9%5@<+ZnJLEJ);Ey*A{D( zZjI@y-?9GKa5V%)5npaS{^~*3)Rlv;wqJR^l7Gb}@FBK>%~;GJ|AuRo5z45lq_X5C z^a)8}f^{OhcKP-zNb@f_zvT90dWv!iGV<(+nd2NMtTl=PhX)>zHzOCfi7z&yH!1TI z-Siw>FMB_3&_2r^jli=gf0BQ$lzq7tIzgQ@9nm8!b(wMiNsJ z-KBAs*s9V>{c@{^W?q})R=8AG+xxbTVYo<_8F%^SyxzH5>7GQ-G|##(lxP7rf?E;< z`~Ws0SOF{J8T0z&#dqz@Pr2(?KKOo`9FjFb?>34TkcYW*^vA)?;-g7OR@Ctfnx2<_n=T~XZY^7vV zK+)p$A(tUElmxv}WmRPaJ4U@n-6T~Ft1VrkoTk9ppF}mb^2dqB#brV>;hA(z#+vU+ z8)cNZ8TxATOi!|>qhLrHe7&J6sEYn6)Fe!aUQSP?t%XU332mp)kM35|HD`u>OI7Mr zWOp-mocE%3@P3*#Wz}N)KRudyH2rAs>hkBU&ugELTywd0_L}0gi7zEzl1dF^)Z$EJtum1J8rA=U;q`I~ZU8 z-7KD(U6h@7L*Rv|uAUC~hE~j3E_wMm8>tI{CQDtzFrPi2_Gwbf*f+`PG>-YB;ztLL zk~t2X28&k4@u@-ws|{`)zBQc_>15)2eU0)fbM^7e&@1vyA475d8r3hsU(1j3i1Oa< zND1;k9d;Vg`}p2i zkgh?lMQ!sP8=X>}OQ-#sXQ#oI=PlDpCtdYi$lpnww>n1N8gv~WFuo4?iWFhE=;hJJZ28EQw%hla2s5h9`G8v zZRMBMpX=XewP4jeN-13_jSS#kOhnU7GA+ZizcxDHf}1v94mIM>je7@V<6A}zb<^~f z!hro18e*`LSgVg}>n;9Pylsty8Vm=Lcqp<1V<#$;JQ%gwrP&PIsMWyjI@Ujc|Xq0CEC@+})Na}HJ{n*OQzo=>FT~BznabxJmtG97! zVv2Z8MN4a9@5FL0@lJn(jCe@R9|S6w(!_ghS5+N!b_UcY7_+&YN9osq7N4sT~H zuTRua|DZ;b-6v(k?r*(cgN@;-<#iQ8VP{r8&E8$gu8TVqXRi^jwcU%E5?_W8<=?sv z;4+ZP0fz(ZwjFnb1Y~n$K`^e}f=vP~NlBJ{ZW4h7rAgP=JVrDa0gs_$@L#bLi=E5M32V+fSo?j)FU2Vl zS?VYEu}B|fIjo^<(u(Se3OjT+142K!Wo!`;xg-f?>HmtL&C6`T{Ej_p7-RYLgto!S z^5G%NH#m#`#{3sG@5DuQn8dMDnJ1L=!VKFD4>0mRcI9YoX_g<-H{-cnzRUAe=_$u0 zIXNC6yGV4lxaDGK$RYt8|G{!--)WWuz|lV7AIm;omP3CYv#{vz%=ZWWaeSB?xW;nLRQuK~ z;M3I6*V!5B_t48fsA6myIC0ebrllVX3%3aKci%0Oiz_TF`$r)6Ec`7DZ>u?adCEL+ z@^WyN3G(!2-iJjaNDVmjboPI6F38gZ>ZcZ@Dg0*(HQ<2 z@^wC^By(BjvM`AC+_`fazD_P`CO35d>JI#+Dg4mi-&;*q76yaKz~p7Td|hSbR8>`F zFJF; zu=uw<{`u^`_AvH^02A?mIW&;mUr+ewbAPqhkYz6ZKN-VsOZ)RIU}_*%4cY%OTM+Bs z_d{=4Sgx|%x^eAZ(7vUKBgvdPlSif+A05>()_HQ2<9?T`1}@$2b9z}aNAyvSxpY7C z4of*gQNAms#C~+mFo?;KQ-0|FtFXl>gHvn)Q!N zo`l}G|8w%s!TkLhE{StNc3EPR4>{#p9W3nevCH2>(Fk8Y2x-&sX|D%Qdp31KW58@=3 zo~hgX>BsK;*x$(53(=sDzg+#1f52hB8|~mtBvi#W>O`^0>$b8jXZsP1Fgv~mN%*Q8 z`9q9dm-^i<`fLsRSv3VL<7aE~0~#^gu~mU_wmqwrG*Mgmtw4_zie2fIl*@6-Tr&0c zOATD!bGJ`rXyn3WD}GJ)6q>b%pTkw*!+7k&RjaSfD0z;4t6%qb|7738hS=M7C5|_O zKZYcYxWqv6AmESWXTy*wf_)3S*3N6y%|krNvSnCR3B>m7(ANBn@fBD`iQ=+rGN`Or zE}?OkV!cCn8pm7qBN=OIr0Tpq+5=rEXzhpJHNF4sx&3sGDJ~X;ZjkGReG1;*9ExA< ztimVkXuWdH3^|Y8p0ob4zHrs%lFPk9g$MKo=I>2xg1HEC>|Tf zd=9WW^b5h|wnO4Jp@5uiLCvaT^ z;_@}Iv8>ZfpZ{3@G}B-mD=L%#8(ijLn)+Kd&4}tDd=RR7z895R9%ibnf33T?3a=R~ z1h>fBc>^DqaL5+#&&|3nLI>O{B4 zelP5%(e_*%*eLhWD@3_b%+pWsBG7`Q`Y~+;G2Xb`i?!!^M97*KOEfu zaO$Ca5C4k;=l^oS?+*65jX)B}!jhXta?QVb@&eMXXH9CeEpph_cJLa=V1|PNKr=MwVrEIK`JG63m>k}RF)f6P52l~RV|q~Di)6!Najl(Ox3=>Wvr!x3uB9o z5~|kBOCx!8%0)DWB8lRMaGDGgQn;)dbs=iL4 zYs5#@1ni#A59gt8UX8nnHAD_$kSQp?U8c;*CB&fQwqzGoll1-JSYVr|q2KTj{#zgJ zRoVbKO<^CO;ivs&uoUS*o3PoAZN{BiXpY$RP|C>XGbXirxnwW9vl>nhiQ>njS{1=I zC*R(_a?7sjMcy6?jg`TFeiv{Rm1wTSqlSBP_b)H?AE!0}NSf`$T}OR>Z*taC)oZDI zesTUyCk9`QUBd;)v|K-6xZ15gNaSTbK0^R9`9uzdEyO5CbB1 z8f!85hp2UmlyT2iHP4w3)4=X}X{^QfTJ;`loqhPKYJmA(X6V`v$4Pi?0-|y}v~O)( zYilaILrw;}wD6GlW&La(QU<$CspDN|z`s48XLM&Xz5KvOS=t;aMDY90?GCl}j@4}* zdu{>UMpn-GNl%i>+R9TtjkWXI*l18|20xXGP^lu%7;6XsTU7rggKbXm>IL4&;{vH; zA@Mrg9iDyv7t{Qk{U>qly_pDqj^oAOEhV(4I;$pMFu}&$mq6kb_w?{JpMdLx0rmU= z91I^2HIMmmcmU;Ik0mCmyVVPUy?Wj7yHKpAUo{g<)VOBc6Z5yL^J8^EdM}--K(*F4 z!ZkWIm!}Hd*AFoCD4>r<^PwF%dsK(Y-i37@BtrML>MUnPS{u2qkkAwvmN7DDB9UYr zSkO^09!S?r+0>EnYFwY}fVNAk|193igrAd0vwd%x&*sAnu3Sl(*5UhaOg@9wESJ>h zy(rwH;u+sHGl7`}RT&=mZz#MZK#{b@v&p%6|v6&$`t%6||w*kLXvpm3O zq(XXr^8vYNZB-2EY!(+Vp(ljkWjdNU6?Jm8p+9+0v8*_7)+G&k=B0G4_wvFT9d|f1 zvnDvug`jIbHzHd;wSyVpPUGe+wjQc+$QP*mu|y{z5iMfK@Wek~meE?8VhSfkBsr|S z-Z6V8uPkAfH-N_jzFV!c*t#k@H;ERK5zKY4MfZNpfO+h4<%kCCdAV?Ku)if;a)blc8+EzYquM#%0;Q(QdsyUBk^G~YWPbrLJI8szVc z{3{o@b93sY?2%cDb#Q@*@0gqN?SV>9|BR5XAo4D|n{xB#$M*dNd^}v6ona}td^}g| zoPHjft{B39lrQF#;a3RqW*Tq6XiJ_0;6$132*)qn8TRIu=ns$fLL2{|hsp}u0}kj@ z45Tf0H0Ib8GI_b}{j-?<O5^u=OYeN76lMNAoqq30SqAP3Xu#ZcQTx6HNH8 zcYaI`9ED8p_!-5V(u#W?LPmJs#&1lgd95Gl19q6SUC`<*?`~h(M)?cR%F!aDu;u90 zUCofK9kny`DKB{h}y zMBWYmoV1)?{JV_?<1t)cgFDf$%4u6P&p+&;+pWDg5Y0(rwZCk3aV#mW%0m4u-2}BUqX?3bB8=E7Lf=Yv_ecMw&PnP zTl$Rt4ohBbKXp>WOO^2T_U{e!7{htLkd$m3oOHyO8va@CbwqPh(s*SG4JOO6nr5IYxPcTRSo|p`vbUCEK`7t3WQo z%#XzxFDshoRLO7MhC%)fRB{kBaD>t>+K^qp^E>8_)G~`Edm%~g06Y4sd zaRO`;rmbPi+Lp()>7K;-?zWa}Bh6~uuJ-8tTH>BLn~PxR=gAqVbjuKA=Pu}wCy&+r zy`Z@oo-6rppt_e#_E^Wt_lL0F8Ilpdn)ZYsz7;xkUav(UHx1;L#2LK{*!}d=I8bZ| z33XNjJyAuYD;s>FZl(4HFnL5M+-C1q>i~IfY-YnWRucct#PRjv$s7cq4Q#j5XK@)eq)I9@ zM3tx#!)RD!=;%l;nO=*Tp;$KqSukoLkzNNvWyj4c&ZYJ+zUUd)<+0x);{ugeR&ijr3z${2eemj4JWQzTX4n&au7W6-a1GJws zIX(H*@0iwo<1yzS6P`)Siyjlh-EiJ>hd9dgt&06Z-zl2gt8!c1pX|u$i*An*hx-{j zzKYw@>LbGh`db;Mm6<2BtTD^Ese!0@t>Y@KZBgqVb{!(ti1fi{SlVxL)O0vD^Zxsar) zN24ijoaNyrmbhsk+sNx)Ja$2*@mIPqRPPt=hM{LXIyVHy{yfy zB}mBd(}xGlw|Hn}Z7h!lbnQ&RAp!F=eFKc)`fxU9iIK950t`|@wjA`y@7@knU$!I^ z9JdVPE2M7@*|%R?Kfnh1uyZ3SlGeCRqGS4?A*}*cHp(u!8=SZCx&9}DN!v-hqxIvN zB?9i~D$DfHX*o=@3m@00=)^(H5Vk>h86_uD{6{AA`zV*{X9KYG2!hts76=NUJ1*8Mx~Mh%7>&>L|xwkmmjCt5(F2A0phSb$NXK#4%6%hQ8L^1jn+rqOz_x zW}1Tm^0@w*u6k5#t{JLj_pNksaGIHDhJWw;uun@-9;VQJ2)Zy{qa3KSR9};Sv}^D? z%btb6Jq{(ABRH_#c~SA=&`dLE!)|8LLPu@+G}QbS>X58O?X{l(zT$K0STN0zDSlSB z>N2LR=5H6aR35!@IL*VB?SwfRP2L1D1&%Z*M*C+se858dS>r@(h%T#Rtl_PntJ0Mt zxjk9y3-sHl3hE_+CbL|wbg`D+8o&JUCVi9|AtBKzWX7UlTZFAP`Ob+;C@T*katN*82V^Tv!XY z51$Rd%#?&MM;KC%70WnPH693FUieDZON!l`qzgrBv#HktccXoVC7Lg3ZEpCXzyox( zv-#fPj@pEQBf9`7S1C2dR|BA-5KO=t5!U8=RaKd+50@Gmbl=KH%d?Rh-f#tZ^=<=t z=>M+Szr+*;^wk-Y+nws*5V5}y`Hskbh}l4Sp`W?sgKg)sLhIYMtOr2b{becZmcc(n z9M*n_;XK^M6CksFMZJ~F?=at{UpgsPB-i5gzjUz)xOCVK>;PH9^YYVv+9DZE8c}9o zI}E>v@%i4o#B$b+y(0pGklYcB^ab6W({<#4`x?7T$4;;*dJGR?z~c9kDMdiWYGtZ8D}1aUtaX2i;L~h2G%zGmwmfG|j`nJ#m+=2GLVZ!r z&AIPSz7U6o^#R`cR4YG1KZm?9oT$avkq;Ia6cdCm{SZB|IkWO}wkF3X)r5&R06LMK zZ(L!w&bdq)#_vklUdfJ)^IyNAEuxyAf9=&vUoesU3(jMsBole%@ZUoHnOpBWgLsEN z{k24okD5>Q_ezw&ojV7>UG8qDHo>-j+9bDL_Y#c$%VfvNi*^iZ+A6>pIV+*-r?nhQ ze~7j-*8;KQ#kDqIZYwt#w7#s?r~jb6)j%>Fja|?W7%c>AY^*r0_TjX^Xx{bU&EDYx z1;wDfO=WybR+Nyb-~}?mB?X>mu_x62E#0q65D_aVfDENsYHV(-*F0$9%+r{EePIR@ zMT!O>Qj)R(%Xm?oCNC6U>2{U~-P+XM( zet`2bPLfDGN)$zLfw%NBWFU6=$bj$+T4D4?mD9Q)UkfY4My7WY3Jg@r))f>7W^9sg zqH*x`gOKvbY&Z}_|w6!u5>~&C| zZZw3mhITY&3EM`ot86_VZxnzKh@&;1ch*~`UrO76Nc#?e;)V=r$cbti`wx z>3RQ=&R|++_Zm%L5Fd%`Bu zYMPj3jM+rn!g2EobVuywJ;A^&OuRW8PC{HGcngSJv149lV&fTII4-cQP=0#2(H!g7 zr%}gaF3>k({BUZj?(=IfiLk0Qeby(Lesi}lq)}87*dGmOS4yhz5VS7~Xh2!SANk}H zAsu#W^FeF95uG&Q&KP#i_jF&P3$;tzlFDE>TZhI2!L99RtddJwA&-5SqEsw7#ERqt z-b}j_&JOUULF=6`-~2Ino>9Q4RjTJE*JggETj4MaIpJ7i@~C2Q+)pRi_d4q;n@mL5 zn4xCC&aziuVpHn?`$7WP9(N)WUDcI}SsqL?BWG^c)?i&z-fr7F2PF9mj06ytKZFp; zmaK>K-IrF7;Sv$qQ3J|1Vk)VVjQ2a?DY*>6d}OGNKk>?9GKJHE1Y2LuCASb$ls(h^ z_C(@oId8gt#5b@r)A7&_3Q?!lC?j;iGS|oLa6bx0gAorSi>9`LLD*lIw+&0$A-Row@S!SzG^`1o>3EWx=3A<7<)abtd0pSPcH?VV+hhejCt*BJ zW#O6cAJ{=wM-rhf!F$9b*=CaLMAU(tC;z!P_~@$qu`P5^XXyT=%-^jZs_utyb#Vr^ zNkAfcwb{niQ3X2vxm=~NNl*M4&`y5GHvcEGpyCr~N=7QC6ijvmIL^t=^e?ofK4xDx#_x9E$z;U|I2lX0qyCQuY#zUL9Fbi#x?*s}# zRPpCpR2x49g&Q#x&ilqFZ!mjCJ{zQ?xN=ybOJn&N?=EWc9=n8nV9y9Z%@yE}_Zjo# zC+uH#FRpPLa0v)UY^>%7ESNN-bNQV2u-`t+a*=w)J8GHT0xw3sAfi7ZWgT zC~R32%M7h-p?~n|ZCj&G49K#LHxiN;g9a*z@+l=~2oZ_cKe5-0}+SdiwOuq+a zRceiW_->S}7NP_8J}|0Wc>BbVomKesjpfRXJ&HQ{g?;#Tg;C#T?p4}nTw{?bQi6ha zjZ>NFpOM!l6ljJC8LMK9iMsiz#{4%u3qn_C-v?2)llJH>j6ZqY{}jo)KQLWmKk@Qk za*+2m2Ti76`L@*91=Lq_Wc$9{PZm_DEec$bJOt6{4Wph3xaatEy`Kz;3K z0*{@qv-oggLmLdJv9|cM@=u=58=2zQjS6z zlTl$VRA#G2h>?aY)xQA=@j*Ws*{vtj%AYI)w4cgc*fI}nA`G!yxW%k%0(HLr0(9ZL zUw;Tk0u?+8f zE|(Mp@C{S~vX9o{dCL=`xCg*1F_=EHn3?g}!=Pe$yui`M_h(9gc}16Iwa;+34Zch` z%6t9`y(NziZJ*=*G_1GtlDLWumkn<8Tto{hv-kA?c5VUHZ|Q>#Y~+Y*$DWnir2l%m zG`0PhzWGYst3hTGu{ayFl%?LjZ>#(b9Pb!+34ir&h)@Nd;W(1(hgQ_qH-X%iMcy*1IVk##5%xI);Q%3gXSJ(P$=9s6v@A>u zFiEaKBQh%M$env>Z(2st8##}KbBDgrZ?Mu`6Re?ik|-;I|Fr6VqUBDhb3v0GVxLLB zbL7W!K-ssL30PsXUAakC!W3eS`?|4$<6+3`@Y7&{vvC>}C*4H^I7XAu@pj?F5f|td zj~S(IJ62`&m%KbRaeoXXV6uE|`Pm&^@UcvE<78@$X^HH38qa5AR8vWPuU6|TRT{44 z5Uy{$^vP4m=UZYK3X~Dl7D2t(Db0MoQv&9E@Du>vMmh$3`zU?J%C;W$51JlR`;{X& zlqF8a8i%2h%7x*BZUph$;olw{F(X@%7j<|nFPYo?6iNv*c^GnC8shwOIdI|Z5&qB{ z4Nty)#Oe>Rk!wqZAfkFE^o)UWd-@&L40QSQnwjC|p3zw6mG#>xVUgjA zQe{|doBx3UQXT1zeUh9fv48_U`_{NDj!FuKN{l`jlfEEe9AC;TiI?{oaTHrsL_=)} z5>Zq1#3u~)aX&wzS_3rZJc#e|WvaMH1H0zA^&e*y4{zbq%6r2n32R@X1Y>?`IcqhG zxJw26Z9cT4xPMbNWOJzFi`_qn=}ytt<|=q-zUB;2n){yK#N=_0;WUf+zqKuOcO- zU8e)A^mEQpl)&h1$-&0mt`J9>83HrNi)!ICy$#AMq+wg`4X#hyqZ>{!Ikpi;j1Eui zhN`SpFKg{F+4`$Fa_&h?0LVnx+&1^Ia)vk%bvk*d!rr4SAWjYSaEN5Oy9=%J;8Dg0 zyKWX9w#UO5+9_eZ+uLLKyiSjHluhuDW_hp|Vx|+erUa>^QMI-%$ExnV>)2)7WA)4H zCgtaHy^9LlHFQBPBtR0DB6zIc2b*mN7?c}Is>|YU55>&wWDcmRR!EpjOen7i+ps+pTeq4NxC91A*S=x(sjp4hSLkmqYWkXL7vy0njmmR)stn$z!1(pdt zQKDj}2nh51#gX|7%2&)P!hoV&VUlAwgEHo|3giK=;P$&ZwGSXKJC#c8%*c!7(Yq5> z_&ebypZiS>&AjmS9P&{;zVVjV{M5`g46B;BH4k)m;3jq)|ctJyJ{|T{6@7? zvfMYb)|X+Vl(X**O>arg*}*oOKUx*ZQhR+TtI~hqAbB7N?#J-ibOQnoVjwb+h zX^W)_FZ6fTq1+nJ2In~4;;$3qt#2zBf@vPF&uSQl2?cMhs!2|PQp?=J78;47i~073 zGD3(oOH!py6LR3lyc#vV2C!80!k6{)XCLu(g zL9pcv6`y2L)It~bO0;n3>41vwZ0}OF6x6G3AtLji5e$GePS z4qShb{bnA3fidG(49prld&yb$1GduDZtpq!Jv~xFNIPVm<%AcFC9XQlEk)&aimhF_ zcU%P?mx5`&s1SJ=I?QyabD$DitEw*ccghM$g%qtB2J-Y>Gi4AWzfqM)6!og# z4%{XHuIH7k!9FrT=2wOmpRos0!=jdVm8KQa<6$>7hbVbA_Ag0WD0C?r#Y|+A1G)~q zWRuUcpR5RYNUSaV`(04V+y#T%m-+s&3r;ds7LH>Al;?nq0tl{QuZNW(L80$_+|oA zjb(1-vwidY`snOHuPeD#C@Rh}CXiWzYuroNg^ve44ahb_Vb*VE&DZz7n}Sybe_g0} z`UHM{6)4}n%M|Sfd0EwL%HzgWC#8z@ZE`~C6Vw`Im7L}EudzBUB@?bUe^di`s5{vB z6M4wObCwC~{F*iZsQV^4XQU%-J!Ej@nljR$$&U}A+57g7lA67|pbboZoALXb#$sD2 zgX~!Yj*h%Y%=`>6VNTerXI)vq+thy zsrB&LoaOljM1jl^(0OwTAQtPGf?FNh1xoI*ieUTPPzz>R9eM`wHfD%#Q5!9A7!8^N z$Q!#|&#WQ8PE|c<`KHuAan`^7<}oPzWExuQBU2_9&=-R1FEc+eMf{&W$U>{btPHS$Q-uw z8gO)N8|y4rc3)v#MY|uaK|wTVZ^=(~5GvDXe}5029a2@2;u%wr?y$CVnX&RyOVQ<{ zFJ35$9XF9ji8)|enQ^7}giwWnDJ8Atj2#Z^PG)jyC|kwJ$+)&{kGLv<*r$*4GXhiR z0V2ph-KWLEDgKbd%!#3wA0(&BibM9Nu7{YzloG7h7tt3S@x6vd>>G%joIMW-Oc#MQSj3skX!y>@%nxGd!awTQ3H!eOcyjF{U9WQ0Eg)@BFLG zchq1iN+|r5bk&5&Wa#2V$BQb+gqjQp(EB)6B>bMh3Uw@nOWP7T@kJYZ@ZI#oMYOUayFWA_^>u7C>=lx+rknihS>Acc;1 z%Es6)Yn({`+6xNDPQ#R7;n5S}(531ki}#K%dFt1tsF#Ca_bApM@LWQe$N3BTOSt&}+L^$~x8jipmBT0v6J&Fe_h=DFm2~o%&A-Yo_I1&IJ5l zD@W0TzfgSQh1cf0y7K~>UpUL&H?G)Nu2R6nES z)Bp1bK~dQ~tTD|ijwwtmcL*`xrk{SgZX2=~*pLPC_pCFT#+$oJWKPJTEY=e$YyS@(3rRWbYFRDgT4e;$P2XP46HsOrI z!<&;O@{4_}fXYDWy^-~#VwRB}SIvG8_gd(DS3#fp6B28G_j9|?DQ3kjC9fGF%`lHDHI+|;9Xm6oz;Sa4giop5M;J@eq} zDr9-^t~$9MIg)M;u@A-QxRrM*dr^7GsKy#fWE!f`b7ix$YWyryr}uQX5M}`+ZKy_p zb($7q`w}(pxm-s5NAM$0!%QZaPro&br7x8|PC0_y8%z&tc@c4)0C^OrRS=SWXYO;S zjq1#+WoE|b+$B5_r2No(2HH=o!F|E>mC(Pb66{YKYp#dx0BAMEF`uusRG(JWjsSf} z2+Q%4;*gMj?>5-Ni>5a z7*R9pve*u{XvB_V%0sQD4aHGFLYTmuU-<06&o2EXsith{9hh2)$?n^4TQ%&fZihUfHfpxURb0U zAR74%f1>N(d|d;$&848D8Ts%@z@Pej{-XCyFhqcp$1YT7N5kR9HHk~uybIc5ga;BpS%WKXamL1W^tlfk7=zB4tt>;B}_ zs$1#z)POICjSt!3Weia+KthKjn^cg0N%szYD^?bwW*@xNhjz+Hnl*g~s zb~ZW{RZO}t#mTyWr^jhT>9nT5relgj038ef0Ynol-%_2TkO-MQ*b*?`neSH?=pD9z z4q>&Ywb|TD3n-wK>C(kNaz z?2;sRp4vcNb*x>1y5OfVjxiZumV%&T~tbo^f3|toM0Ylbe^=!uBK5r8d!BRYmL@MZe-Z8p_pGK z7>7cf`v65&8Xy}a=^oijF|Om*dVgZYFbca_Tkom&!u&+qKA#RanEtLCnv??yi*WpHV`N=rHyd8_%9a zwlSm6u25+CxO7+&Q_onPg)jE`{vfRF{Ex^=qH*)g^pY}NR*f<|-N5w}8lTT-3Ly#A7vm-VD>IYwW=}ePPx`~( z=WA;Rqcz=YXFJDU{lldO95ne<*qW%W%ZeZ*F!in%;xsMyek}Ig*|h@XEKHJlDe;Mk zJsYo0zI22;u3@TI#k(;xVAMp1b%#{jx_2Tn=g0}((_`dxP2IWA6%wwp#h{EQRLj}) zb(t50OhkcEAHhEK?2%20h#tDFS$=WhG*H}q$+=Tf_3pYeASA4tlsTbj2N3^qv+qp% zi4Ams#ELGB_4olTcgg~qj%K>L#Bxz$u6&oR^5gKe5XLG zQ`&LPk+OYQu#M-g%EF4a!_&<{UcYfO$DDyDvNcN!t(3VALSHP<$7#!B?^ppC%?W(lo8)G#xP+s-Iy&rjOl4OHx#l539=n<2UR?~5#h zS+xiIF7!6Wm|HqF4Tk`Q#dc|>v+D<+JFNi~i3G2A2FAyJNWvMDx0HE{fqMNcUxS-h z+MYl;P=3>`rBKKH2y*nI1?n;kuhWh`8YOxakHj@4W>Xu6ZBI3jV zP{&`w*)zu;1L~)|NkEmQR(vmgf2E^oZC8eX2T;xQn#l$4C8%%gvTMBI(?IO}aOFLo zk^k5}e6wPF^`(N;Wk!9?M6W&?{-+9yjHQF*>or5Sj88;$85}^Y{A`^?)`xKxFdXku z3Rl@Rwr-NXgAn~%2x^ii?+(8Y|+*xFUy#TF){^Frn+L2I^KR=bJ=6gH$Ye$6-Op$u^cNWogF3`rn<-FJ#8b5c)QotQ35-5hA zV}6qa8sc^nRmkL<>bH7Pn-tVUG6g|7E7k(2Joq7I6Qd%6v4gbi-Bvk3on^LP_9uRhWGit zEuzM5lQAj5ljc}HE?@8jD93Wj?QxqJ_0#;e1vG_8HQ+u1=_rRoA06mtrvFB16pep& z_ztPI?)~6*QK|)!9A#9Xw+euggMBZ>M2XSC{y!KS@ZQ`uxI)Rcq1n=?SU@5k54ptd z^gy82_lRL$t0Yvu9wv-TuRuX!qa)~->krq`p1zn*W<}2&d5<+HICEza>cXX!k64XZ zbyfcJ61hj^Z-AmHbaUYr#fr`u6%F~lxx019F`N0sgG)qEeZzVgQfBy0&v(dziX$9j z087!Hthb0=W2F|VRE$3(Yr-rjxiesk1ewG`T;OjEtLzz)&D=F`<3aglncuF9An1vp zCfTEABl31bySDjR&=@E>McT){)J2K4ub*?RaTVVeg_~enF8t{N9%IS>+a85lp9gXh zJihhZzjMZT7$D@lWB7$+g!+6;_d-rU4qe`wiXnl>@H!vH#P9XBq=;M95<$T7^ z0y*C64M8{2AyXw~Kd26`HmGmKEJY2zsK^T(&4-K&qq$ z_oLz?P<;dF9Ix6KUquT^ews{NbZ}f}Z0Mj)&VRU#N0;m|c276e-@nyF|1fZ?t$nss zX*;^|i<6QS5ZAT!DzltF^xt@Om-}~i6B*f~TWRI8U0oYLxbZ@~Az}zy zfIVps|F}6>mGuO_xkG5E&=4f+Kov1^kX82c%@8rrst*2KWPP}`?YL{Wmnw3py-Nar zz3%&M8aT&0AJ6=8y4YX18X+`B7WNzkU@x5@YGu|t}_sr@a=tQ z2M|ezEr9e_Z9d!e!x9Zms)xtk(jOwJ?FdQyz5$;=K$P|G{at*T=FQVB^Wkd%3v`}! z|Bm}t_P|}YhOO=T`G>1dqXh?Y&gywQI?amL0o=H>jV^H0wSPFD>0xRGvamVNd+MjBt?7nySA6-I92(h5lly+Q$=zxLbgk7cHcS` zNS}NQk>8ahr@gc}<$%t<+*_xP-o;EM9GdFcql>EtmtVA{$ zaL7e{>dNwy2aI_&b&J+?Eh_Ce{@f6;Jh$DvaG6@VXqHvxCtK>Gg%(*K-mQzWB4m8tjl6#r;3&i=iMKe^ifSDXG;)c?G- z|K5xLQ8s_W%`X%2fB*R#ZvKXw|8J(>Dax}i@^`PmmM!!|IE4fvW;O8-+=M@N1hnZ& zfB1==7W!}D>i;zXz$e+DXAB5*>gV6R4uN~w4^VIcYeAml?7;uk@qhayk>9)x+bOpb zlm4d%`*R>KfHDrX@Y0=s>ofYpVgAcCp`8psX>r^G^Zl<*_Thb?XmK`P;}1QBf1B37 zUh~-)NRRzJ#eZ3Ye^J7}SMfi^ihnEWe>4WbA%K7H#s7qtf5Xkcg73fK^51atH{ASf z`v1sB`WsRI*g*lnAKJ}M>dGxHsS`tdRoq>NoJKKU((71*Jh94ZizE&ZZqxC#j;CM$xpjOgYxH>R*JX7D;DMiCi?H9^dgN07sSE$Pdc0TJY3&pC`Qe}6 zJc{+C<4@!9zAfE3v9P-0SRUq_F8pw2o=2@8CrzfqZzfItx6zkuM(9jc;OBmd=;Qew z+9Vnpn#V4D>FCH7i1?eh_LMQyV&r|I`$?2jM(paJo~~lH6N&TX`#Zhj(aeg-+llx= zN4}J5?pt$?W@bWOJdp`GCvacypGj+6FmmG#Qr>lcm8neh+^D;Pw? z*V+A+G3_Ucq55{^cSKT5zR=U`X9P^0Ar61JpE5m7xx`&V>3?hGPdo};<2=jG$cxWa zk2VrdS+6DU+2ZdbNtT8SL=`afh&`LkU~0jx;q2W7f6in2TJ|7p>tx2{72edmTOLpn z;oZ(}wh)_RefR?y{C5d>%AF8UTe?s<@CmtIYaMr`rd%x{s zem+PZywv;8VUH-1Z!h5@2^-R}fyb*WPZgzT+tbo^>D&e%WS3mb&Hk#U&L$*;b zjO!b3in&iz4o}_#|HpZjdAPTe1wq_0B~3VW_NymvM&8#|b(H^9qtlb7Ht!)2S5TYl z3S-Tm!GiQTDw415Y4lwiFH%$Zead?J)A~<7RJ1=*iZGs$^I4r3OyQ_Dy?aq5+G@I? z?4vOs);*W+@TE56+ymIFS&e4dceCcFkEL~1BYOvDa1)o|bANgL+)RWaA&rkecfl4U z`kwm7FZujN@`2W*Gan)9~Dl>$o8ut z0!?&3g;1$yj*~m&3Su;6-V)tf5WKD2zWo#gpIr$WM_$|)Z&Sf!&I}gxP#K!u$)#`~0 zi_xAlUpwO6=WK(`cR!M0xtFNS{CX@W)6IzB!=(6p4Jkdj{juaGu6SOJ(=1wg`aKp1 z8TSEy;3aw2ZVrz>HJ)PB8%6p*l*{>4Sg7i~I>L!3Y8xA0xpu7sUuDI1v7rZ=7xhma>ey8v2>EVrbSMk+k`=%k~J&9TJZyEXzbNG&Y-IFY{b z3&_;SHrYe;A4k548<^qz0w2>hP|3xXs-FJkci$$%X$WszN?G9NtZbb%&liaJ4a;pb+UR;3;g8|nnjGN*$FzIvvJ1-|WO zb1QkhA|qd@)Ju}`YGdQE%@>rRaX;8@2DY$Zoqwuos-|dyv0}QG9EO5r@S`Q1+~EZf z2)rPn!8|jh&n$e(2;xF3$#rp#3=5)^#k_oxkU*beOVTYxK`Y&lNF8Z>x6A7GS{l{Q zvrAZZ0F&h%ZfUZgzFtqw^mPc^djkqF5RpdJ*0hghDhGk*SfsfQH{Q5(n)r-g#`qIf zS5rVnPW31lbhF=#$mgnIb}(*}vEyrgm!3NE40}?R?5ZHmQhoZy z<4f#Lz^(hie)$mk;Yt{D*OoAKCi<2@t+yM8F9;!#YTq#vhN4jaUR7MI7Ad11-oLc8 zwAOznU_ZocH_Ri@&d$yqmcHI3yHa{`Bc%q^sKKZ1b+aicHI+|cf6d3LKS|(b@c6KJ z(+e+oENjWs*YDx;5;^~T7Ar7 z$fq6C&W8Qr_R&OD@sQ=rZsnF`&&OI_ZO9jGc(PP{|HBhyppT3&is0cJ>?>bo@khYZMe=>p#}oZ(EKrbCHkqu zOt+_&t+er$2P{MvIjB-dnbqFbW>TV6wI)$sQRU1wDr_<8+0ogZuo*>_Z-H8q2dB%M!k1eC(J#qv(`qV9xCFFDB zPbW!n=X?)BE4T&gF_>S6r1TI!R`g~D`X0n%;0q|PQpFv;^`2pdUk=GzMFX$0hDwZz zK|I07K>9$uHu%fxIVY5^LaaonJ4|sTbwN}N@R6_56#L{|FuB6t{kFZmz2hSpwPjnO z4*Wh%`Py(**YgfWf9Hak8G$hf8D8|XJo0hS{_w`^V}1Rr+-e0w7fBp7}Qsh08F9*hMWh^qf z9Btg(aE4Pv{B~lQMf-hBaj`A|CgZQor{>%A`D*3O5s+L!dfCQ;x~Rpds7>vM54W~i zCC|QBigi*vJP1VzFv8QN_}gR5WMi~F=ItO7efvQQS0VQ0#W(_@erjqRo)tfZ)H8NMO2M{%zgn;LtTXbJN$uB5S zxg*Bv)QBLw$L!?gSxfou#$i)6E=o*pylLBj1k%1B1YZ|b(8C288$ae-%Ij=utmb*m zjM~HA4LLKIs&Kz@L^e7qYHltU`EV-BpW;=Lz+-gM*ak}|oFn@KHub?21!T52>k&Tm zfkOJ-QFuv!l19@$2?Lyy%!1{oPI^*>c8-TPw)grr4pe9TOUff8)HN#~j!fFsR1`Ya z-)GNsIL2)X_MaY31sU+>{lb-+FC;o%44a#c`Ih~?I_~{neLDAhCGL5W8a~wW&|s2B zrkCsWzpk>*eXPz?A@TxY!6b{MGb{If^bhoQo^+(iWEVg2l%}t474ZDHYwyMd0+779 zZ+3X$ysB5bO-_E-8>E?#&%{Qi?u@?A>(vcptn3vVxM}f$7i?T=GZzsI8GJ|sQ#c?f zy#LMj(fz4iHwyr3_td^ikHzm>w9jd3DsTdB3@@JNlrs~DnPf=jertwL#PhZ+@kHp( z$^^6QfWg5*Y{(B;`_i3WGh&~?(8G#)Q$1Tr0%X1`qjCDs?*4w+l-#-Vk9`P(ZhPfc zo0C3GT_WpDj$4^Njarj?+hg;|v~pZ&)uxTMiYQeHY~Y*{yC^2KZEZj}XnzbfOsM%56TBv?BX7~=GnH_I_b(@2q6glfJP`NiTKgiVII<{UsHh0+-rv@^mW&^Im1_v97o<9>VfIG8G z{hT1H)9<_C(|u6U6a7yIhx_bW#XxOmziC?8nPVHVSV>}dg1BcA+Zhw)Rqoiy3dBMK zrR{|;#QN1ASqt1BGpBMNO^rZQoz0IJlu?RqV6lC!Ec0aF_mZT!qB$;8V#>4dF@-Lv zfMXr~M^C+~;@1Z_4@3v>I)E28i^kOjqAk6Qrv55yyKO?nuR?s*Y}v&~37-h;whYop zKX-MpC6_6mtb(gSW%yDj;B==Dtjv_x?KuG@RGoUf-zUf6N`poMk^QBv``YKcM9AR| zg#zw$23me_LyAs+Flund(8D8s{IEZ(iYf!`%``_G0SB)dJDzpUglyLjR!*!I72ZC) zY!bep;6HuKp!$hLQBSS&rxVseh``yg!&z;6AJm3_vGA>BmqScY$!(Ohu8`sG%;AI( z;0)E$LQIHoh-ECA%VcugA9Iw-r*sW+E$^nk z>)?TYW1fyq;=;#QktaWR_+;TP7R}OR_B$9fv114`v~o%)X;VVTR@&6})>t{Va=-Nn zj&vDe7HRrEu;%t^Q{c|YpvFT*w0VH*ut|ungN`>7Nrnl>yPn#~M6SLbzlyyZ-us%G z-)9IRii-^uO}i}YKQU3CEER2E5;W@8f=bqEWbS+QPgOYusLF)P)oR;{3Yt|Xx zCy4~Liw8LkyPNz&MBC=QWZJ`f%iYl>0v*OqH+;4oJ!1ORdTMS#oTu1Mg*m+36W01* zR3z+Rdr{ta;Zj$H zO|&KOx;l-EMyP;|`DAW>5v@r#!dP*?iv&ZpSaxyC^tE_isnA1-oi%Do_~AhZ?>t}} z(mfy=1K(mAIF*$kD@d@*bZw)l!r_(@k7IyvU{PX4a%$8#5c1!+h2a^ogX=U^vWg`x zvInm`+UsuGe_&2XqSj35#3m*tel!)yQ>XD;^(u}CVfWjwzG?M?YiUeErnHtVl6-w< zZHiq#pCK28sQ*T3=Q05?iFzP78HSzy(CU7O2oW^fM;zHWwa(B#rFnuDL1V^TfKhxA(67P`mp4?sVQ@H>aIphEWsT zw2r@dV>TI;B4&|iX=SCdmmAGjNdnbnbQ!@p|K_~qw%X6P>>R!2Q0T+nLoLv!HcvRm z8Ii{S?0l|f)+BuGsawOE+Zfvx(Q}G6gubRqrgV+gX6sqT7ZbI7&D-IhTdkopys{0iZI#o)(@rC;C}Mv@i-dY3LzKexTa z9%rY7dLb8XNhcdCvZqM%ek^s&r;0zz=iz$bku&fcvxJK?gDg_fOQmD}2JVX=R*7UV zpNd+2*S~><|7zTwllV7>`>}DV%Uv!8JXr@{7)f+WXj7M)%BV7%)_0O~;Mz)FV|@or)8yX z*b7>AK@R48(2JLA46*!RkSHUUBPA|mqvCLxbb5lJ%ek?GbU!POp-5M%sB&~aU~&&0 zCpBJ%|9trg0Pp*s*lGM%(_jvV=w=5n<3(F@Gn0w!z9>DV7%j6MM94zFhOK^k;O<&| zM&X!)grlqYNUcNl{#F|!tgN9m9@pa2?MLOr$O1nWv313t1g9f_SL(eM7m%v61_YV#_scZ^cYir> zWJKfEAPEVU29k%Hv?K4E@)n?l;_Q1p6nB-TS_Ce%Gs|WnH^JkEx-FT z9!xFV4(sxHSu;qAZQ4+nqSPJk^O8En&h!)&QSy6r_7|V)nCjZd9#2p+Bq4&dR7a@5JGv*G|p?kVCq5da&!p! z@PdWhG0gsn5^s-bX!-KfJz^4t)yA;ZW=snE8JP^Aut3nGOE7b;eQRCdsg~nckTll* zqt@ruyf<9d^ZVIL7Ppbk{ZundKWLm=PCPXMKUtX0d(!2STDr&CwEtp%vGI_GG6OA1 z^9250T%c6NdGc+ud%%Ut)KXneA0tD<)IeNj$Z678FQ_id;iBA5B^9V{Z1GdXLKA*o zr~@2=&)#1-3_gEe+NWK>wDP-&Ctrj^?v}oIiF1PkqK#@w8mxQO#lTH^gmhbeYa`^$ z^N(KRqg&{$mI^z4J;Vf%EY5z7^X%o@X(?I9rA+5P5&#<#BV-9E1sOBLB?8bM7k{|F zATi%!wtt07v}lGny^L`PiwWorBfI>;W9K*LdN054wd5|dF>dTteA`g=1J`JzyoUTj z!*<77Y1~dU$hh9wBwKSTpIrpQPQ2%nCOCzip;eh$>C)*#7B@A#krCY8m<-u>E$Z=e z4GBo@G9*P1(xSHCLZ-{!ES#Ze8SF|IE=V`Z(ZSE#$mt^%#+ zM(b$0XrwmAcpb4b=|G%x!fe#FuWo%-IO(%C$*|}@#R`~0c$L!O^eNCo5I9=*xzAX7 zHr2fUBaN!s^%f_Bpbj31!Y;tQbRDYprm0XZ7PtVvO)dU;;YDNuv)qHa7=WxN1O^ydhEiQvR*p*MS3;M4KjD`6n z7il|WNkgij5dSL5jKj7){{wJHB68AKDTW0`o1(!l)`Z<2JBb#x1nUe<WsX@*l zw5WCoIDQP&aF`R<-j33*+WlwWxi>6$%m|CNEE4kFu z7zNDYBPoa5D81!2=Sv4{;+CIL!po{OnMGCNG?NZq_TV^wUEMs9y!k`?7Mi(y`OAwK zKP87n0%)h9y@0BOk-MfOr}C?bCmTCqg6^Z)jG>)fir332bkWS@wD+WT(prChYy#xe zI4!Ha1pClO@K+YV;W-(W@0Fwlqm7&9n=T>$YJS%{t7{PHB^{JSZuWY*I{Kr>-B9PC zAVany-E#`t?F!Ns?%2(onw&ehr*uv&w>;%`uEq06p|?VS#U(c*JwN;?`UQl|CECtV z&dE>Si~v4#7W?LKcRj%K>-v*EAh+s)M+fTwWOEQBObXepxeyGGC}^ z{sQVpIvuZ2a4fn;szT6gHydGh%JaerA~ye9;Eg_IqyDVujaz_O_?Ch1s&L0AGCEBF zCAMRCGmLU$eTsEtl4kxrJolVZm zPjcb=`}^+jtF@O84h%h4OMd&7ZBuqv#lH@!*jranihBSJw!GI$*NGrw+sWb==6*!odya zYOKD24#TEwAFq!5nAL7W*Mj^u&(I(f9&jN6oEZ3Rb9x}=;D|-Z4tqDYh zqhV;K{gb7fru|9n3tJ5(?PxK*^}5bZ2vpff+IA3oR_>XytuZ zr2FA>a+yg3>Amaf%<&zq-J3O7ri07K*i;}LgobJ+e%&LA6Hk-nt73LhPy>jP$k*Ed8A1 zL3ku>7mzdYGgot66fZ4jr8Y?UFxY`e<-?jXRCzJlM8V0o6W+(U4&FxjpEbt$aom=n)cBC$c9kr}m^W)#@a9()jrUW~?-r~0t@6MTfhRsyHW*Vkekb&_omPJ3 zC=Wk>SKt&PfwkUSD#*yk(I*Ur$7a{zf3%&$Pny+Ps(gnC$UWUijaX^k%qeMiND(Zp zo1Y*#SelHzUydLkCQ9dbighJ@!cqQ=FUjeq6rEa-3_sbbUA*h zNFnl`Dwq8lw~|I1PdS>G(B>U}VAelZO6&t++YAG_}RuJ^~y>zM_}tsBP@C=LlG`BZM^xdw|bXK6E=%$gco zH0S3Ev@~|*Ea!81c{MYsir0`}GxJ)J>kVFeMstaY{29NMupy`_0?|fEQVbarvknGw z4M&0C3tHz8BQ~Px)cyqwb;&xo#B9nj-@aORd&s+Ry}YR6%3=uc77)CBosG5icOVyO zg#tzUaDE5!`Vhx49_j@cwe*jkyK=zksea;?xu%Xk1_6Vs!r>b%DWS>7&u@iX>tf7S zOO-4e+)-9;$TMwUr~(N9+gNkGZn~ao(I4t-Jw9C4qIAV@C~Bc&f%ZJ~y|NODYytW{ zEg9ulq`+@dozib3+2XOyNo~$w{ z#GbHhkvxTg?CJkvGW=)q&oBSXSRi!#W6Lsrx9Ac3)MRvGRZ{Pwt1vROrv1^Al>^!zLo zQ&;gIfR#HSxqHQ#?7WSWfo<72a5I^Xq#L5Cso%vffTd*n1s7Xe9~X1Splb}}_!2jf#u5^H#349w(bZq9t+ z0z8X-xUitWAQ%_2^T6-!U2_uygOvTQIUNq4Mz=!W6yAtSBabq0pSP_INC zoz0@i5Kb{Mu~4s;F;d-oWj&kF^MQJy_w##uHn(n_STDz-=Ers>Yzk8qI0+u^IioC? z6cHP%*#+1p408Zpd~w>je+R~^aDZ6r*NvsjI~={69CE>NxS>NkT*&jA-PYO*ER?^f z9WUnOWuC;~Bxl;?3Acxywdn;SeP3U{wxW-Ww<;ZJ6}D@znS0fT`$8x=|L$2hm4}TP zNWam3i#ha~H6s0-{?PQnuG?%nDh=Z7WvUR=H#OdC{&jsO)a^P)&n41*K~hywy%lak zpl#o*4`0UNkKh6+zmFp6XWD;(!Joj3J6x`H`!)OCvzj>#5U;l@)0Co0xdwbj*e>%C zh+P-{ImD43CeCC%sM^j18N9$!Y0N@@hbB`?31uiYN|WvaWY+EN9WwZu3{U9u=Q~tw z_@2CIEGOq1vweN)V902GOU&iIN=F;x#%D^&Lo{O(7f@^!51oxp>=bJ#xn&EsAF9JF zY2|l>ria6y*luB6f_Lik@PzA28G}Zm^u%wYT_*x<;@S@lBb1te489~o?(C1*xv8d2 zhFAeZa2=|Z{awEeqVwP+MLL{;U#|o3K`a+=moyTw)9c;7Wz-FmHO zquwBe0#3Pv0vXKU1(s*Jk%z5XnlB{B0UuF|8(*_eOk{w+KdLlTGT{a~;xni(0nDt6 z?q#>5C}o8VeWgx~Z+44j7s+Q4HuWsK7av^)LuI^yqVMLAQF6UzMmzUHP|z3cqJu7K zA!&pMc5^O1i;%9CBHB@)2vtBS-sd6$u!K9x4*`WN2Th%;Ub-u#vd#BpqS5K5y$ z&{W>SR15l^^WFAa+MtIqY1tKxs4BZ%Q$(vABUlO-^a)e(Ht$ERnUVj*d4n?hL=nq& zxJl>-YEbRH$B!SAAi^1+5>v?$tj@Ip?aa>b*hy^c0%5I~Mb>k;*7dGX*n7uGM*njf ziF=o75v2R!n4}xlINM&;O@Xl$gNi#7TZHWWz8B~0CGCEEjr-uyjW8=&!^HH$r|Ta+ zKfORnqNk#$us#fe^cyuVBJ7)nECr)F2=n33;l#Mbzd**-;Lo1H@`@{PutT>GRaBoA zZb^zg7_{6RaMl&|&9*KTjZREYpl^^~=wdb9Fm9b`(FvXq>FJq!n$?4;pWN{7rD_?a zwBRhFS&PliJzPHYZ0CAAAKTeb#Opic0rs<_##49ICs8>`AsI~pJX;` zBP)q2Ibl*=(3%Min8=6EQcN8N13~V5)GBb=^gJ5e1=J`dQAM{*ocx8q>NV|iqk0Wy zHkyp4%GG$%IKa9Bw85(rA1+y zF|cwkleljq&w7!up(o$R<&uoQ=QyWZ_Q$j)8 zW*Hn_uG!kiEJ?pk2Reio6VHJJU3sI|NV(#B1>bC4cjRwdFp2TA4RuLJu%92z*~mionX^xjE4=fxzx^E@4+RSF8gnUnf0sIn-MybS(B%c50OIY z3R8YT?Qd8eGHw?cO*zHr6@9FBJR+fl++Y}2*eDCJ_iDM-RrXds1W5ILCt&Kd`nvDZ zw5-k*#J6a+3ZiR|3g*d&2cvo!(MspL+o%Q7bRX)?vIC`2fNBc0s@~e$dXV%SXQ$&w zG?+#{-Al!cFT+5K+rszsT!pO%f*95N# zw-D)h=~sPDJVUaWSYh;K-aX!XT`;){p$(aKu65dT zM-|`!CFs-UcX@X!ocvWpB&U-<3y`ytYJd88b3qa>wZ>$Q`JBhF0VMyp_(epxe8-cB zOSQ?Vdr@z!Hgk8>L{U>oL8FjY+hSF)euE*2_eIf_owbcS=g-sK3E%a96c({+cZx;C z$FWD8Z*lF?gvwoo=>a1k!1Nf!w7!*2|Cz9J5}Oc#18o0vnFUe+v+ycemDMS0Gj0eOqJkzqJ&3sz-TSr;>@3tNHlv#BuEm$8f`F>L4ygi97d z&%2LxcS5xp<1J{oWZa5^=$g2y#Sr19BmEFo^iNLspUIYmg08Huz^A@p>~89f`iViQpPAaBkW>fjHSfD{`w@R_FWH?a zu{=PT#NsoSbhck~O4KLHEtHUolroeiyYR5bGZjelXVU`?(%%3`HCI8uK%SlkVuG z-5N=i^6|~Ph}>F`hK4L5AMR`TSMuB(Js(Jf)`;P17bROQi{HuZd>Sdb7Z0a? z0n;5c7BNblh|=NoW#d;_xb;&nHBXdyPns8aXP#vj!dp2oP|)7*O+`BXwKQ)Xv^non z5f|9DX-k&quAe&5cUtc{&?u#U_2x0i=#w#xK3Z)6m_;SnzR+Kj{NIa-qYq7_$i&L& zG!>#!XQJ*tc-h9){p$w9q{~5t8SU*O;82Q#JEv!_tFXH3o=2Yq2rZP#HnF^x#`MJ2S8Th9m4DbvX4^%I|kLL>I@xE9bC3!=m{@BC> z6}aS5-q#n;Ehb=a{``uMA7<31yO;Aww7o*k0bP+gwDaBhjwIn?t#R;-Y;xEhaqv$(d7ATfN{Zq2HNpY=F zdi~3M!Gmx_v9HSVgD9xD)lmb9>pP2oy`*Ke zJV&|z<7zR>$)5wv&ZipC)&8D3MTC>WC8G5O?|VDpKq4<9X=TjNhr0nO4VDGh_nIf( zr6;YOnK^E)1Y~|cyF(iku(5@lYa0ZsScLCjW=^)kY|8rY-rf8Hu#S5+{A)H%`{}1l~GRY{J>Ty^_Sdw)gTibb+a92KUdVHw})%uatgsOT6=Q zOZ`sLJLCI-dw8P4Tc7d3=<(owf%A;WB6V!(xnlwJu#(xUy$mES2e~gVrif>lnpnMQ zIQZdxN$SbV>y5#wT=`rOJ2vs7j-D@+#jqfiq}TO>^anVQ0%)X|DR#Ox4YkRCk;!?= zZHBz(SDv%-ZnN&HvCD@{b5LPLRRtKz_a>16;yiSe4qH9nlf75>#>f)Wk}m!N9`0l6 z-l%PAP-Up9KJWRfOYm9q-B3AjBOmx+IoXLSgMwK%OF>9TZui+-u@PfKs?{%?P^ME8GJQPzxv(gX#*e?)$X5J;h@mMgX=#%QT?G%&1;72rGX6tV zKc2P4)!1uDXyPa{971}Ti*TV40C;-7Y+fjl$(f#e^&U3SGr&<~YQ{Uy(5r`iJ#y>D zKV_~$dIn^M(Uy)KnRy>gZbpE$x?7Gx-m7=q`R7;}Bp>NtRq&g8LT6eSJ94dftWGD> zf8)h_So^N65i)Dn&ckq_uIv z#`Qhp*ZG4tJg!Ok-5h@iY^UF?T&OeuLn9;fb9j-1{g+IOr%ykgURF>nzj?DYX7B|6 z5a_L#k+`yG(yd=o@nP_t9W?oPYooM8Pe`RhH8G)E`&LHVZ6HU1si+XC;!jFN<)ruS&s*>>du-%u8VC+hB+ z8ek(Z+lYq!cg7-4bvgO@I~TPyj!yzO>ql+UC@sVoFW1&uSBe(R z>&zHem_VWyGvmoha(?)oC6E5OCOyF?N}Nzzp?7T{CMx+sN1bmsh4dJY+MX4=$-Gy_ zIp4r08QT`&=|KM`=4)zTFhAIKWJ3R; z9oa_ID`rr`s7-7M>itlCHRYQSbr3OfC-OPmE4=u`$+^0EOp;H~mP4Ll0+bKckM5@A>`#pBQ%=<);v0MV1fA z5=5o2uyC$c&lKWS!S>5xfRL!G!^-_8PE28$6`XP!h|p-uf#F-1v#Ngl@b6f33!i88 z5lj-+iXgzt026{XnGKT2p#y$?iqo}tu9QAu7^@2bJlgs%Q%wT_C{hH}ME{tG?L=St z0zOf+=-8j+IBAQ1dX!k{5j--N9SUya(#rn!b--J-z22p#JFxw@5;_5}&;D+&oTA}-3NdmZ7 zLXKGZKI8*7Bl1_1M0P!;tP9BuOx`DHFLG>3@Bs?3Bez=It+w|=A;=j2o|?pI^^Jg0 z&gY!rA=r@v_j7Hi=2isaQaA09mJBMczjJ#=^8L0|CwzIE6A-+R}Z zoW@QbM?lD?$wf*Oc0E7eVK>E9qf57DfELen>&MBj z2Yl-!@fgVHCTtN!TWhY_o?EnVc(yFM=?3)H;K;Z;QW7Mne~NfpA27+RRrN)!MhA@- znNswh^WX@gF9TGvlwTT~^Vj_)81 zSG&eWhfrtIJCpvc(th;RKa9aP`_Ex$waVHxYx_52aut%v{4b_x>}{r{_NGB$&i8iB z2iGf2z2c=kiPv&=2 zwb98MJF_OM(Kvy2~4bb?{~xa*--o#gp{i$=Ia*=9hgmhoG33!O6GHw4<{6` z`PZq!HF+5Kz^hQHuoOntjbFzzouX6+FE4jlHVDe@9&RV4ub0l}X$-P5@02Om9c2Fq z84O{%_Y;LMRlPJGE6I(zDc$FM^@WbV6R8S``JWRcbA3~2yb6+%U*A)B*;tx_Q_FQf z8BTpZ{zuFo1l2I^f+j8=Zebw*`+Wn;G`+4*0ndwOjz>n`Y3l`uwH@ zTsF7UEMwZsxxn0zD!OE2MSTVb;>DXxGp1#2w`M^9Sd%I%6|l7i1JE0d_w?68DIo*xhYlRJtxbuI<;vx$ zv{F$cZmBiTZx|Pe0Nu~05s0)lr7-kssPNi{AX9vjCYwXoEGDE^8rZk~&`+ofXy!`e zX(S_~=#8%(t61mlVo)b;^tvI`P_`?SJ1qN)i(+Brley@yy{N%-2Np3Wgkjvy- zC4+WX(BO$RUcn}`$Im~Fadd~Ctc{F*9%cgzzIM9r-S|OmOl~FiIbr6p3tiRs zRVTNOmTN+-(r0%b6ey^Za^4}Q6Xu*TT%l~KSM%~(Mx*Q;o?nZGt`dIMdg{l$KBKHz zz(B$$Xd-e=O{+JZhRQEBt52s&{=<<_yq!nq-A##oZEt@uby`)qHQQK(m&&gOr4K?H z^|_z=BflNuw_e4qr>sLmQ>t%;D3PkId$oY{$QLMqX^zxcAjP0QcCTVQg zc&0a4eF>-OT^e8xXE}oipPu{Xo+;Bt7zy~aI0N-m#h5wnSNV}&KmA+1C10M!;n|armB?f* z%{T>2g~L;@-DQLk(U<$O94vSF2fZ_%?~N^oj8m5ZvpnTEUH2nK5T%~SzLA%**4wA% zdkWm+#srgk$KH|gSfg@k#M#dLj9Vx(Zn8q3h}N&7`@1Qq0%btjvC6{Hp1=BQ0hfJ4 zhb*rcRI>~y2L{Ya+)`fwh^v6gVji$PQc|PoVPQz4{Qowu0hF9Ycj9C%=~@T>6B}|%g@e-07Qwp_ggMB5l4GU@3wBk zB_Ds6#KieM>^HOwKWt1ucIPWi4HTqqJK3%G@EoexE-OiTdF2aU39PEBqLoDd2)Vs$ z>bCBeTY8U2i)_}VnUz)9%dpVU>XX60*a_b6O!RX5h-s?=Ubuy;Z8(?vkDSvY1b2 ztriJd>3Za+j!1TXwUn$|s--vARIps>TsbNt_YA$g%trFQR!Te|1{=s7`%-@OWo!TqI;Laae zRDEA|veFsUD3MdYs};4&62Bh21+e?H>`T7Ps2cUDUxJ19Cs3I0T}s^#L4){tcW3v@ z9lI55?Ff#v!^Qm8iGVpieC~hbB1-Z#(F!>g5QrEam;`Jfk=^4Bn0MzCxzE+VSw1Jb zxZ=ssGvgg9FFr*~r9|b8h>5g^<}BiVQPiFkptzMXV_?Hezxi6gvpZaOXXnJ!kEga( zJ&rewQ*xWmk}D(bNE(Q-V1nPb&fSgnWVgC;j^R5=LK4%1KAK_=%vbB3O))?{N9{;* zwHhxnwnv1K&u`iMz6yG(afI?pM^|H{xD6RwTVWruvD&iB>dS-ILwgL|b2YT~7e#Nj za+f|n4s%HFxh&r2j`NkUqUwaQ90pYinpg5g8-pAs%Jg)*#P_CU0`8G7lp$@KL8oe} zd~NyII9qYs|Bf0>umXmP|CTe?ab$E*+_=_if}tcBA=NYg z$NgbmC~EUQ`=LXg8_ki-5Nbj6;&)*Sv-_tc;8P9krio9B{ie7OLFMtuyoXB~zYwX$ zU&G231TK&kQmSq%l}^_wz75ptIPxLgxF9>~W_G;!j+1QSKW};@mzH?1@cd;Kql=!WpCD>@9OuK3eFt+nw z6R6S0nE{pQs&S~Qo*fO{Rr=e-iR%??H$_cDx2geUHct6VP^yPrx|55GEsp8w-`ME4 za_ESCyUkM^72!+q96vpbau)eDI6VHwG@P*VCC+^Lf5m_O_JQQdybKXU=X&4pg zn-Al*DbwgKSDbU8X}y)CmXzGDmV1QFS<`c>Dxk;v zi;~vCO|!7zBK~`TB6y0rwyWgcP**4`Psllt6U|rUSv`Dtl~Otv6lt9Z2z&6B#boo? zykm#s6_%z+^0}>_Mo*JMVXdv(^*ytl{{i}cW|AoACEQ?iR64h^(w(kk&7aMiX6<>} z$Lj8?wN>6%Z*<51MJ)O)C!KoPeGh#KOssW6Olb^K6gsNXVv8P$jIYcC$Vh z4LkyEJlm8byC@1_^k@FL~%B~zVfkqaJn@Q zeA@l`*1Ge{>4fR9r{NRm9hIS&zH#(dx|tdeU4wSXd0{2PFD3@KcbC$C zg%YH+eIEDqQ*vAVqC7S{PPw<%!;`pMdU8;}lqs(3(~5Q<8wdK}D4_HT9-p06VjY~i zT%hHIKvf>wquRFQAK68Sk_!wHlk_6L#57P#w?J05`s>rq&rtVxt z4{T3_xYJ}`ANDebLIPNmADE=a3rP%GH9&cg6#;C;#;$r=0R*!Hv&^(BGDyL>E3WOG zlCCv@fX&d`Kie_{Y2h+z&2s22D{n&uwM-20?)8EA-Qz<5=kJ-3!+u7>BWYwQE3BvH z_0+7|ohGO<5R#J0FPz)b6SKI$uo1HgcpxgN%NB(8KxJzeo^IBv0AhP&kok$7q2a;l zSC}RAyxK1EG=nDj%xDgS7DaH+3AJ+7_FtswWb9Wx<5qsW#}J?K40zGDd)FiyWttqE zo~8v}9+pSB(#~<)Y;{L6n{TQAhecw|<7(Rzn^_z1Ug6rVA8`Ws%ST9G z``t&`gnpgqHeu#F+)DBdd(_FO{C0S&yOEjWA9)?Q`RVRY7ngSd5g!T8KSOL!04meP zb9A}d!b3^E#>NT(@6f+kOyJ_-kwfO+L$^j&4zmqcZoT9s;y22BLBP$qmwZ=3{EaHU zu zk9JNaB?8f>ulN7PaV)=|g{%quV>cQg?G{Q)wXG<+P3UI)x?>afWbL8um!c?N5$VqLG(zQMK)1jpac{N6YF}UZrJ^ z?Z->J^40A)4FR}2nrJ{Clm=z3ABGACr|hh_-sspcLQ|o<;>TOvtNjS_a$nzd4y&51 z;~hGc)eNA);dxXLoiCj4u=k8p38au@YUbJ)q3>TQ@zkna; z5`DvrjIt!^5SJc;rwt|JaVOr&a0@j_FtSz*y6i>~~I*jMOY<8n0gXh6qixN90q%0i5oA8d){*>sIr>GXM3b;r>_;7eRGy{k0=DQO1> zS%~r9tG**15shobHE!ds$D$CGi$ussf7O~}(kqkeaQx%M;NGEcCm{SuTDWaKLUgI` zyQO}4upS!XNj&lzTEgNve{z6tnXvDal#=({=VrYp8$-`@_sjHmv=caetm+{KFb{*{ zL1^#j=+P25#2NhIMmsr-XXDhmHF(v^gLP?qmrDF`2j$PEUE1Uh_?1$4`O~jC*oinS zD!k6q1YWPNp9l(@I?vZUEOt1`E8fDA>8q;RsQA&F(1R$Jcgu9PX_UBTxp%!kHBDl^ zzk$`z_I6IP(d73=2Izkx9KCt0h0qLLh()qa;-iVouUs0?R3hQSjp3v<$Tso`JTeG+ z3P)-wASVw($gBU2HW{jICVR+^ic1eAY$wadNJ5P(di+%AvS(LeU4+rSQBir|>eFw2 z4oea{u?tGMVvTfoe{xmmLuYj6M>lQ1LhFUyhG2fW+VgY8v|?2b-(724^rCNsraPJ< z5A{ft8d^pxQQz97Co;yH7x-Kb^X)6)l7OFG-w}UF^fYv!lQTt{nPger<;mmEjB>`G z@Bah8Blmyx{bvT$CUFQiaX%z~O_iUE?r?LjydPBbyPN$EMtsj>wz%R+wNAIcN(&w>sP8QQLgFA|syW0ncH$D~O~9f>2_n2EL*X9RXWVH7HA zG?=BM#}D9HagfpC5pwjd-t#3qQQz{Fh-|){Y_t?BRoH0;b-i9 z*JsiQtpz2d<|`;ytTW+1x|2`DM?t|40zJO3!DH*B9h$weI{`2ueB_XC?wPPlFU6Y3eE2uw?_z6DVvg~Jtw#g(zW`9dD_`g@_ zkyRCMsPYaT9z+qZ`^CQ={2=-N@eh4@V(M<~Xycp}=3hQIm5a-6KINR{xGjRX_o8f` zM~J7ZzS9TI{bz{T`+h>hzlL~g#1)_Yqc0DN^MCb+AS%KhVut=4qY|!(COykGKvB$& zmm}jN>%ABrM<8@N(3r>+$M?lVKfOYW=Cy!N5V}{+|Ep)*U!08C4oAsDd6FA%UZcI< z^o830zcigfpZ&1UDkqtBJ}r5}p6AW~D6aqQ1^NoauUSXy!8u)a_hbBjnvh^KAB@z3c@9n=tF}oZLwU#-OaqhJ z$d`YYo&vTLN!X1rKF;PHUey1e)TYV5;-|h(9FcXi$vJ2Kj?n7J!Ym1YucmuJA0~*i zN~*^yQul>gmI3ib9{BM8ZybEzD`H2q()Cb;x0eEh1A?|zC`yc^p zG!e`6l)Ts>apDVC4SQ;M&Ie0{9Z2`(VnO=C?|q7Qr1 zm<0Gw{w>8Jnqqvgf39q%eNz)*rT@=bMm?zsT|uJeCh^UwF9{7c?cEFN`K z&setO5=KE4B>arCBSL>4`{=O(LY{`*%#6I4$N0W>oMq&S#upN!PTB$9&# zqsw^u-~~`>ZV-n@s+6p)@p|e%oe&Z(Mtm5eG&*BUjcw4J%8hdp z#SpvxE{nby>{GPl`hpp1!N9X=Op1SN3MI@F#Qv0LYiR7dvzaHv{9{5b6!(VNoJY*~ zb$)t8Y89dhyM~fhpYh2sGx9^4+9SEJG)oT)6B~#LH~p`W3*;{ZZ=QtvKbYqSO4y%q z`b>E3X3l|Nop08Q1ixjWkAaFZ=q}$cr37$vl~-wp@VIP|Y1C#%yBU+gE@4w*Y!P>|dZ{_7T+&4`IJ4n^Rl<#hUI zxFk^1Y-eLQvbd+s;oSx2JxCZ??9)Lz@#a-OS#{}Ymi$ptkwp=@9(({48nNT;KzDr` zv44$x+KNqkkVo(b9z`SynwnCka66{GyV zl!=4Fk3V{a*!E|+P>Dq12W~ebr=`IADt5mXa(7~%w768lPI{4x>!PM%1Y$@t3+ zXJsTeY;0E_@2{GEmezlvM*IYwG}7HwA%fdvSiJcY>oZxI+&GzY@xaKZFqu)!t8`QO z)zuOyNogevMVh5Co#AY$c$34^vh}BgKOEG7Hf&);5ep5N-3T^YtPJtrPA2Vfxs^xNu zmVC|3P_LGWqPp9+sq-*qpR*k%3v(N_z_{W>aYd#$2z_?Ph|_YjSS%qbrYvY|{v3cR zkjcv8LUJ~JGRtIPX{FTRb=^S4MBbYhlEf|dXIdY{f-wjeIHpxH`B7?Y5VAg*zIAY` zGxrt!Zj6z!jTSbh4T@r!TADUMPFEw@s61chT$0UIt8nj!L&L8jO^_yP@#R) zjMVROLOvpn(GLBo{A{2-)($Gk1GD$+q$)h&rwgP5!_&fjMSTUWgfm3O9;S@fCwrA% zhGWXg>RTs>X2K3sO5GW)e!>TRy`T5Ih~+tdqZbZtzmptskK6pYD1^_)gPM8@2h_bi z>5la-gVu9bacCKvhx(VRBm-|1TcXjZhoxLyoir7is=KPFYGRik^aCktv=qhS$bYyH~ z#Be(_f+H&+)y}1(xCFe)K0OsW5{O^Bx-!a5k+5@j*XT4Y(?kUx;`6K*FtFk@e(w1+ zvI=rekL!E^o78S0{1YFE5p7c{Il2 zVl{thZEUQC5#0vy@2A~LDN^NYsy6t4AYlxKb4n?)9#+?o`ZZPrhe`0*6>j+*85A#! zH1a0F1@0h#L5>XX#B7QylZ(x*KK>3E+J-QA)z+ESRSmXd7Vo_9&o9mnq-{E~Qf0&)%TUgrW_Djt_WVOqa}r@S%B)bRK(a_aLh?oJCh?!uU1*RX0Kq z)7-}Kgwu^IqL}DYrGrmv5rqt6hHX2~y0do|wTo*3rbgm5B}#E^jX)_#0NxyE#$3YVG z8+p9qgOzUpPEgaIu1L9W0#9Cb$R!#%&ll3rWV+s*no=t$sA20&*CV$c4j%3!PbGX= z8mTF>t+yWEz>~wQB_-tG)>F_PTJW$DMKQm_mUh;N6%|!O^*mJ?csjNEuf45i=rGVQ zvW7&}o7~4%> zjyaNk{Ef@}@V-0>eY|@aQ|q~n3hm3>5HoC@#?0J2yI=!1 zmh7B_v{kAed?v&D{@wGLm;t=hhQsh6&gi!*Ws=}J3nLbKtJa(VU#nqK&r&Y*=xc(WRE0L|O%OG;b__!G*d$@B& zwJYPUw8g^B#pOBm?16?G+M?)=RFLO+I6-~PVXaj?o*kbn0OQ~;ooWt~`j(s$RT*FD z^Mu^U*t~v&kN347)|{_^_gSDo-u1_&F0=JEc2{}t)sq|wYA)}@#PfR37fC{JV&aOu zKNFVQ3~{?qP(5!J+Nd?~M=ASLi#m~F&(EFCw{SuvWiyQEd338olqh}G00*VLABCR< z2df%ieH6j}UNvRAwA&EHV(K=ri2H1QM4(-fo0h(xK0<=0LEB`aBUIV*oVk*NLs8$Y zX2a6TXJ%DV`-QnmvH(t#I(PvI^i(^y9R@-A-(AN#Y?zsooNtjtN4-fEKqqus zw@Bk`7sbWZNnL>ExGFKh*#PPDdENR~b#seAuk5 z&-#jdhq1)N1(|%oplet}+naumvz>^k8N9vY{q*|LSz!r<6xGwL4%HMufLSMBZ)}=gn+ld~)GS0tl0wr)l71DxWhhPyXdwMRTf@h21Z9gb+!ci?5!Y zxeVIhs0rbn9p+lk>s=i@aK!@f=1kGlPC^nbRK!+41t)_YfNx4_f(}AM-tdyeFjLBY zlazPGlJlxlGM7zKwY$vpP4pzje8VT`yT?(NQ3p6Q0M2VJCpoc3 zmus#)UR`==chxT=zcQd-dww4;;`zBt?8fbpl*Z?6kg40*u#`&ntJg+{>X{K!^lzwg z9NpbxQ1E0}C`|J`wKeOly&@pmKJ=J6^x0-jfgGV-I4yoBr*22bz1J6TS2!#^pqF)9H=$>?ITEJ9%V%ztMJ011Rh*HYXp6cxVMI5gTxp=H`p09Au2Qdm!?-UmVLISq zhet*-pBn)yJ8$)|TzUOCW`*)=pz6qL{ zT0(E@=^C3dc5H0)R_xwL21ZxjB^(-}QBZg{UiK#hu<;?rXERvM4A@r6gGIxQhCp=4`mMniZa z-qagz_AdsDt;4lA#~@nkZ(!-(O6GE!?8>Xz;!e@WUh=`y5_$rrE`4&=%J$j1Y?Kfj z0!w#!4Yes*W3O;s-6hHPodXA)pV?Q^1S5`I3)m>jGVd3JMlZBuW2-)?L2Il4L}hLK z5+SvfmH0bJMfo1)@UH0YNNM!YFw~@a3gfw#xWYW<7pmag#`x=DTCEt?OBs2U{gg0 zvkE00RB}ub(=P&Ja?t#5u=fv=Fk>Z`VtnFABw3Q9`{smRi}`dp)baeyfIhPy&CcVd zn;1Sv?(7`0iILoGd=Qruwz@K<_gG*F0GUp!;Vbks+fh%>6w+~6PM0>po~Tpt2&r|z zfr!qZia8Yc47AP7X*%t}vSfB^H+*_mZi+`xmb2}7Zp?kXfVHs#6gpa1*%oo!Z3;D+ z^9kfns8iRj+)fZEs2GTUC7rY(?ApaED*g)(u{{R#CskKF z)$-n<(o)$xgucmHzB{5T*)pSGrBBP@t%owaJ^IMLK1IiIfle3PZR(O08;)tFrOrA0 z2`?KCkS+3le6f$tk;_)nsv1Zi(<`rkx(t0$9C4)Bd-3-kIu{oMU!1FlK6>sjVQ8=H z%qi6CUC}s~{g5Ma-~(TtbUoSATVXM1@O|&9vTs+>7|6(b%p-Nw zn-eDOM2|zcuP(~KE8w*O+jzE`B>&OnW|0&X-FM*z-mIamZP&Xie5hCiUZq~|Vp&8P zG&iMivjXnL7a~7EPH6T87vZhN8G|axHtT~RzSmcdWH2t?S%+t9mT4EPjR!jvD!KTc zi&Me0+TkOnEHm^=rDIAAyx`Bzv0elCGlLF@-bVrvro)6Mr|W#vVr%2Nra(zM+VycZ zTD8(L?{=-PYR%ImoQ%limhAg{57qN)e!A$%!mcZLUD}hQ-NE$6pK1QO`g@Zmz89@e zkcVWk@+{L!#=v6=!DGI+)!0R8X>khuOIp12XogQ*Gv*zFD}l^Z&5CY5N5fPe=1Wt< z&FG>8B0CA*C$pZhk92%xeN*K6y0fmm!fk!_nr65@AGlqFB6~%_t*u{Vj9Rno7)<_f zS!b@hjMH*QYkpP#gTQnI>S!KvBg@>HSDR02_FS?5i* zj48wS81hY%ch1I7wkKMnk1x_3Q#j*gG$zo^XBVZ0Cnn^QbOoAEKWy#hx95ugDmb3t zP57x=-89Q3Q=XUJ6mY`a)?_o)*icHln8wDClGB&?Z8LIHPAyGE%@lci#y(Nnoy2d{ zx2TF~y1JrJRI|b>grI(-SA5eniJdJN?!w0_96mnymXF#N~$qEwoCC z7~l65T}_Wvs7}@TCmmk*RQrNB$ln#^l{M_gvlcmO9y4jDGXMTk^tdF!om2>>%gC(f z+K==#?F|&alP3Sb6w!|24O`=}%_5%DS0Vk`+HxWU)-$Dn_;76pZ!O1WujN~9ii%E( zUrshc9tA^B{&H(rRRsoYCa!6N)RgpO&H>NXw`!luXE^r!>ZXr{f2q;FsfknX+BD&G z)@|R`G{RPPc(UVdq4Knz*?*kNZtYFIYD#rTe@D8u=&oi^`SnSP=R|4CSp*-dZyL8_ zzmyaGO28$*zaxH-r=kyV`JY6aUlRsZiaoNwR)=SO2sQM1tzk)nhJI!v zeguKX5*v4m@EeK2+GX^yHK>Ob&Q*V8gT=Q+TN@#^=mGs2b8xo_pqoKMo9B#?q7_~} zFe-54M|8lc(8=8mvl316bsak;RA0Sa7j5xF<2nHNwts`VC5&h>{s#Jjke_dz=sReh z(+^GwTvCn1#Eg$n$g_+IX8Zmv^9#Sv zF~Tqcfs7L=oPu{JY<#BwHcEI&hcw`wz=_QO7MEVBl3G>UD^G83j9(8 zTZJ31=u1H1^4wMX1e5#IY51X$739$*tb?`M91ho#-Ny~LL=3-5U1#m7O=I4+Eb+UN zCQ+Jt-K-1b0jxZ6Q!7@?XS)BC>lqpS+SR8*s1L4hA!EA6uI8SlVZ$V+_TKI!O$|SH z)?ZHJ4#x4-CyRnKYgrne^_25c4GfCu5CAUOC=~|EOBxFCt1FqqcNbsYz^AfO&>XC1 z1|p%{vZg0wDJrip6Rw#|dit^vLtolucEWA2S_LarLU6H*2U zmip*+f5F$g;)E-9s9ZV|0&NTx>okJaKLlMKLv@ZOV@+*#O0-ri0IGpp^wVkH=j0a5 z8s<=15q^(_K8_ax0_HO9mqWnxPF9DcZU(tn4&bdZ-y4?LV}+w%L&v*| z@wc6qCwKP{KACGg`?$ci6H?DP3#|9XyAJ4b4!6Hpgu6gjXdMuRFh50asr|I)cHHP{ zjt+39JRs;?F7CBbZ~EeM*tlEp-;#U;V*(pa)SLaB>u!ua3>n+yUG#%Z1g6O$fLP^U zv&afuq5b5P;2_%fiWkL2xrN{#$<*_(_M58Cr=k6oybuM9GLSRN_QhUSG*n{RR0SF6 z5(kxA^1GF6autD~4KYYmenh}nmxhsEw>O@}(aF(}I93Nh*C5mX<|g#5?~9VJl&?u{ zu2{$`Y;#$7imdHJ=W16zL2nk0t@9=mU4p?}qiC0|Z~LgFlk_umh{|B$o6|XdVQ03J z)9Qzf8oUjZaQ@-rnqt;Z|IaT2i?^v_K#Ey(b~|$JC%#L1*6hzP;!qpj+^Z z)?jQ-h}`!!bmD!Me{&uwF>Jc#0IU7%E_1^ov2%*3Db=o;aU9Vr;gG(boNvv7 zhC8e28BVRsI`O16Rn0k9M=1a_I>nRmGi7K*u>;W`W*E*4^TOAohN2&mzcBe!u*E|o;#P?-aFqNA{ViDjNz>ruy6H~)B^u}~weGwth- zhj5N2O3a~<<=b{>X(Y5JDzUPYi|@+FMsWYTBH(t_Y>w5#1Fp-^(IPD$a2}R~nUtM*_laR%wbGpD<$UzY<-M z{F7A(CU31n!%3 zBWb)YGqhZei6l``=A)mHFjC_s7sXTBTG0_v;UW83%{W^h@;W9`cnY_lN{E$`o&c!W zNkxJO7`A3a>TeIQH{K6|Jb@ROGq-I-2j@_-exx~`1Q3tFWr7MPoip9-aPAv2ia6<7 zTZy>|Ns{5XrY&PgZ~g6c{iuiND?N{$!bxmbn*~ycx|(ih>lvML^T%&CSWebg=L~C2 zL|$pa8QW?FkssUoBsR4%EB{!}>(VQIDK%hVeR zal4|2ROxAJUlOfKX_b%Ke)|@LfR=tLY8;nn*ijE|u~FglJldwp6V z|8OguAO<}4Q$@E}qc?T@z|Y`2=BOTkL{qc(7@Cer5ERkHc6ZwFi9CE83OOy(# zVwU6SD%0!68R5O&lEeZ#NE8Ha$|$Id+gh9}jcJrw}8GIYa$$}2vsPMVl61RxG;&X9ATG27dV zkH_h!XoQ|cnl(Wrt-|YP6XnuYbgi?blnU?c^ilYZO6!%mx}X)f8Q~sYagU=Gg?Ger zr`%toMe31>GF}o2NOtd2@lp&4YXcoG)xJmX&R*Kq`>)ET_2;~4B!gL#o0>*{VR_tB zU(uiqi^?{l)|9R=*{A0@` zX7TIxqD^x`p2pLYlYz%gw;71>xBV;2P4Pz~OrqM}%STdmr&UvCs-wzi?~gjIwPG1B z_JYyLU~b7woHh?uVw#s){ePMr784ZUhVA3tc;Sn*Uj-^G|2RB-cVmSW%`jIl(6ZD4 zJj~+6@yFbMwO+EN%20gdkA&(5tV-wtT^lm4cU`Q|yh-j_B4PF28S%|?nLKYAr@zte zEB42pbC1S4oP%>fl1iz6NwjP!(Qf7)(xytN<(oWVkq)6B^Bd*~RAaYT+E2W)F&p#K z28Nn~jdO2;3(OGb(yjtu_mD_>M%|&efUUy!ytEG1{+k^ z^{gN8?jj2|MDE8IAaiU?{gF@r7O14*Cb@rh{wk_F!Voqxto0*Cs5+Y z(%H(QMHE=nQ9z&z<_gDgs5=yH8%>8c`F1{qba=!7M$_t`{vqy~ zYa(5_#T%{E#Z`~ve6KClSVrB7w7@Z3Ust-?$*`81@)Ang^10QKi`@^-z+>N=f(5NQ z4<9R9Dn^$|ZP))AwFPokinN&zr4{vdX_X0$Uwb1hCvTuuBpj=(qh$Gp7tjNL@QBv3 zY(-p6Qt}7dH{ivpI5`ZUm85v=6C`|EjZRN^Vc~rNQrKnZ>BdB3Q|t`LD=$2yHbY~N zA~MPJ(%0pr5GR>1e`qD5|4RiTnau?=s$e_B9WH=1@UZ5Y1iY7;yQ71A6lY54Bfc0S zX6n}-k>Alyx_He&O=KtkG~8Wb77!7gHphrEt+&V~TIM#fqc$c_G_}2TqhDOmsGgRd(AFjS zLOE0ktvA`z0#JlbH+(NSLS(CCQ$vjWJP;!&Eg1AvyYMb3JMsx`>^(MBt)t_3&xk`I zbaRI!-_q(FFk^pYI=7S>N?ZBbjM>C?X*TO;exVF!JK_<#C?8lcyuBz-Knw37kHWbNsImT|1?77a!Ir}CB-5C(3d+k$;rtext%$=qt zo8XvxD7Uq+nkmrnvm?}eqAfx_8X6wuNF=e^DhwI!O(N*@G-=2YOpG+*FnfOwLH&PKjHO+KC24RZ4OkFewJC5=Pr%thA zf;RDM`kRX->cIsGuKMG8ygR-zGIMS2>s^?H#y|*y`pH3MVffu^#K5vAw zLs2^;X?Fo%z)`%>NPvz!K&-*hyl0DxR&+m#4D*Kf{Cl+8$biJ7680r(3JS_!K;L6a z`O~r!A^{6$kUZq_QCj0E|0Z8b>~pNBl4_3n2w+yhY}--{yn9(R!Y2aK0dN}O?IIxN zjb=Fcp<7oV$Yyz0Ab9AeG;(r*h!U(N%3bEm;Hy`xWrDibd(&mB0VX++pRBSi*{UPJ zrq{`kh6YI3!)Kpwuhuf})-I&0t9{_~b{DWg7#lP^HOV8XF1^L;B-jXwK%!9{vgUm{ zEc|IGjE=4%5N?cFe5MDTAnX%_q*%Aw2nSFd6BZht??ioESg8^g~N^w$n?0qnlM+G zT_P|4d*(^+6M?0MjmU|rXX0S z*k5J3SkzANTir{GGs0ORtZ$(UpW{E8wH`J~oY$>*rl5h`%D(BU^Cxuu#5+5T?tw#_ zn#k7m)h6e@>ZYM~kz|_lJ_rCt{>+~vEIJ*g`}Q~ktQ+l*eYUBZ zmQpPi6NRjp;C_n7;pfFGO+nR_@r`b1#&Qa(^*%ncte!#?Y#fzJ_B)uk;LAmlVnpzv zMvPA^0g#+T-Kx=Pgmv>w|6rJmJZ82ldc zh=Khnoojz4AZ@O=S7&=YlLlZJNX82-m%wyTiocU{&O{3fZPqcM6^QFgOcgd&z|_*| z`A&^XMKyBZrC_{q`?`3|lH;f9qMOt+BxL0&qNWZ*nU`F%&UczYJ8)U3F)3JF`jN?QU$nwj~<%X8ewJgRipZ=G`%gv)r{p-Y{{jP4N^1 zTlt-0ptQ965G*DlDY?hEE zCg=5E%`Rr$4yZ{9y=i|V+Qx?e{(R1b(?XcwAGiUHa{7XTgsMSqgKvJCjGMbJ?VQzA z{V%@WI;@SZ`xGIJ%FGiT1;XYIAu3Wmd47l@^o!y9RwOrZ!f&f22>%X!18$3npMWlA1R z6`kd@Ky%oeZ9-=3>!%VwT5=RH^Fs~X((?_KV{t98RZ21)ndJSKfJ^SE0@uOd-OCHB zSokh@w_n2lyN<-RHjkS*H%C7W`}FqoSr7O<#AoGdpy|w`qM;}54Btj2<<|u0WpY&0} ze~4UCPbL%7+>M7mD(36yiXaJ5qAnG!nN55ZjnZ10*j@2YQ{sW)vbX{kQJo4=YyKtZ zn0-Rm>uDodKG&22Cmra&$5KPDGUW?9n8y+h2^YUyHCy%u*UI7iT&DUS8fkRlp67a~ zdG;!4J@4gO((QBoIeW!riDH(t?Xa5t;!MYaEI*+-QJ33^%mj_Z525Eb&=jV>BS+4anlw~N ziD;E0s!?!QsaC=q?dgru%S^`KJTcE|#Eu47Kf1uj<(sywz7j1KW>6|%EtcZ}*0q1W ztoW#a93MEQgBeAL+=VJDTRge+iy>g=c(>;%Cco5aA0-xG5x_{sqA?#Klv|+!PL34Z zQm6{Y?6_%)4VcCnHk#G5L?Q4}LZy8Fw8^!2P6-xGsi_xj&qoEHM`M(BQie%IBsBMweCXFG5e~ zTzJctg0@@cm*}oTda*F+{2tyrJR3Mu@nH6N34pa6do9~Kh_0HD*`Jn+m2JNUaC1E7*gd zuSZna-MoQychnmc&(Qu$s&7?# z8KgQD8h|SKj4;8FmSs?4wjw26M(a(DCx_bc(vj=STeA0-$UVpH-*{%Ly^i&qBg|NI zvxBieSr9#D*uJTpuu3@+3F9^*lz55S*?Fj)`5jQ(-YyzmsWonCX~TukCMs6E3kzRa zA-gY~n+jc#zs=k?k`F%=`2neX;v@9?O`&>rm|xFnZjj3Jd*qBpxSv{vXM~`hV3Vj{ zXKzt0#1~;k#`)doK-(Q$A1^x|v|we9~7ErwRtw7$JTS zjK=esc7XzQ@KDV=aIX%)v6oIg9@)3Ej}YDB}}Z?BI1Whw5{0;+&Y!A z!mP-GG*VA5Yq-<)=Fe7pGtb7PSQOtB)6rE@o%;E5d7q2j8}7T5m6oWJtOrzDKhLuLbD!L&Z|La-sB=urVg;1C5r z`v6_iM|YNnF&{R+$S$dF?#(EFM1``W7E+x#w`e*ThLWirD+Sn1es- z;jsMd9uq^h|0d+HM}-rom-`t4R0nAE51n*T&}X@XmHKHeI*;^6q-yUp6z#XJe0gSn z26Teffoh!XE~*yeKS&guiE3@*gsXRO z!iP3>?j*C^o8W-0{+{=$|d@byxpG;mTs{WWnt;yfwtj0ic&9Zy5AeTz8DDFF! zNe`Q^(n$%iG}7Z4$;BSzGdWPi*gmt`6lb;}7haL1Ew^hUkPS`z>`st~*8k4Wye>?Q zRRACT)o%7*5V+hTMcb z@l6U4_Vi@Tp`tnS1AW?C=OgKMR2WOPHlrn7B^Nd5%?i+e9BZC~T$U>0b?5nt9_JpK zoRAF^d&)hw%kmt3At7^r6lrqIGOl>{1Gk1IUKRTNY z>#N#cM2JF9jk=vH%b`z zXb9wtG*QzNu|+Y$uFG;hdEbn>RVVRWj%!hdlL`w<2A{`&Opku`l<~k~2d|!8UE?jv za0$+9dogjSb`ZWBnY|;3aGe8F#%h&#h4I^T1P_F37lwuLnLk(#?kI>rU_Zi%b;36# zKy+)Q#cjgaB2GHBS4F9M2dj?a0`q+ zn9WJWkuAkgq%kY2cD~~kgn-a>EJbgIDPmfZJQrD8bzgmLe2nq<#p@Zf@*|GgnWN_G zQ&~4j@~dF+C#A<}%7R|nHMXxlQ=Lt1Pkyn=beBQpFeK|1MR3a`My;Y2|6xPLuYnYJ z6Bsldenf|yb-kn6IP*E7NpqQ-U1}H(zwUO{Es-@apOYJv`^V$cZz5+r`+f`Zb!PRO z*!|+LJdxk*e(&6G^d9JcD%Ox@K7sU~cFTnh#OvEDh+;{-;S=_&(6*fRCBY-1V=Jq} z?4m|wqllwff+s>L`HeEuaW0-z%fy{$+B$}-wI6JBLQM{GiJ|AsRav2EM?8w14Yj?A zaI3%v!2^WhB7_eN8mqrfq}9e3%stTd4&Rfvs`0r$tO+movuKRY?mi$YYFw<~zx?_& z;5Uc2Bn|dej>vQvy7`4~)b_PT{bS3$T}{xRmW)3nC0kyEv%mLMrEO_K1 zfSr~}2K5Y`9p~GA|EuZRpdvnA-dTaZ8($+2@L|`FD27I`*SqcE!;S=RD+PI~ zI-jIO)p2u%^@S^TE#45|DN+pi?1CQ!XZBwA@0LBS+8@g|ursD|t5O0a-$=yx9AC8z zqZ8T4i4evjzcn}@>HpkrwdwkMvHNpB1$hzHE{hkV<%2?8v=`+&x}atQnXDvIHgmP| z&gUIBP(Qp>!sNnwkK3xOYMKR{r!M@FEJRtdbuhikU$X2GH%N2Se7V!SAKAzIT%3W* z3_jiNSTg+TBYQ=KB^d=VyXvP@i##$sSvy8}eFTer-^ZqEP_zUj)a72G*|}U;Th|76 zbUyVeAMLtYN`08um(eLOanL3hO!^Q_lbfwXq@U*S`fi%V=g;*uv@hlxQ0dosACRyN zLC-wSNcBC!&Wk<}#0Ce4y_c=c&4F^?Y4fE18@`hwzv^|AqY3xD zL|#F2gWuZ6#1IS)<2<7y8v)g}q>1H#7?1YEETr@~lX>(kYJ-N82vD@i`7`_epvQhR=?$q@6+7b}jJs0smYZVVx=3Os3It}VmuBpx#08Q{y8VE+ zBZYruEaHQA67gm?xNhmO3UtD|jOdT?)Zn6Ne8_OhR%9sbMG~R%#uIdrZ1RGtnCHh> z&d!(2Kj@=;;5`x|$n_>}ix@7bpk+h4JP9u^KW%uX0|cvU=o@DFEeOK$NaSR zq0Fvt>Nh4RU-OBg1W&ix>&bwC?;OPQ z*Jw8tRUQK!!Ec6aSJ$ZCCdJtf>Zs^ibKB-KGL~-zk8ak{M4U*zMO_-hV=T9lsP9F5 z#+3O9Nk}XGx&ygrhd*(4;r`1dg(GGzIG?0R#T3VIpQ zSkHRLx1U&vJ7irrS2l{2;rk=;eGT}X;Q2b`_k5s!H#pK*Y2fyXx~r=KoXizG19(NT zM?jD~K<@9$DfxOm)}qao2*|YFmgC8RDYBd!bii;(^PFl2-J{xTuZ^c@#}gqk@^Mp6 zzk-2-wqJ$R>2z@7e5+DOkH>T}7 zs#s&<`Lf_|2+s=p3aXQb0duM&uzqG843$^>x;*GVYJGp?FZE(H{&ZcMsN7@D?K5d; zjl}zs$izZPdZyCc{CJFzFf?*f{V$O1U&EAcEtT1lCb9f*&ZC}-cv)FZi7_@Y-}N>S z_D~N7D630(qJFwi)8CRUu&KE`$q{ zUvv1j@_|T^1T60DEx5;qW#1eYnfclJj_j(ts#iv4OstQ~ z?HcAdglI1=w7ybB3ENy=UUlNra8A5L{2xu?2Jy)8uZ@kLWb+)yb0j!Dwr`{*Te{%y zlEH-{jLZAoRK>G->(1!s<&nChCqp9~6;f7_G%t%#lewMae7YIgj=ip*YA6|eHNsK2 zh4G@!{Az0Ka(O1f3Ifn+lrAAfl4M;8E=U3Zahww6Ok5;X;yqqRqN#IJ-a+8WUAFBS z0B+%YXlpT&q@5TdbTJ}KA*}{w8sQG1Cn>f#^@Pnk4VwOuw`riF=wQtAa$suPgp`BHOwmU^0%JuF1G! zQIIU(gIYA7bI3eX)fhV6%Z77`M%GGSB;I zY`sg`5(KbJUE}o~AgS-MhCByd(=XTzIgIK(`-?GdG53euGJqwU=430M1CreQj-#ZfgnOwE6lw)GhFN8|6dN z>4;TXw5JP}NR;x%y3k4dR55rhqD+E4w2_$dxb`w6nka(}Z zp1Tfx%rQ#h)Q*&@BFx7PTy|HPuZC#g;fxAFt_0>T57?Ysw$ZE_XCk4WqF>1GU%ZBRBfW{4}zm2?}PKc~dMwv_=E-Hq*CV4||K(V7keznJt4(|SV zS=I=nr%{55NOZDzQ5CplKII9E!!|`jGO&4_+IJsnt&u8sxad81u0l7!Tw2@s57S)B zN(fFCVwstZA$mT`-bBVcYp`v^6j!h1QzNHf+OyqtD%AMp>lH8ehYO9}K*js-bMs8t zbJ;=#!JMO)`pfQO`e7Z1cM^+NOJ$~9=Hf;b`Ayb z%tqln_deOzA9XCt6B7XaqXhsFg%{Huc#quOf+^BgRX#nnhB&CRGJM3TIU1OTXdNYSN-GL>xC}{0waU<9W9(NJ5eL zEiffVB@k*p-<9Qbrs;d(s#-1I<9V)B3V5U*KNWM69blec*DNM^<7pGn&UC>P1040P zG$ck{MRl}^rlHJyA*7|`e_=Ctzs`ab;V{G`f2gXW%#zej5XJk{XElJeu+Y4`=+QwL zCxt&=4hfi?3)Ej6P7B@SvCH-xx*zsssX_&p+l2|tD~{hD6J;3bLNaVug3;GpnOT6M zXH1$g(r**uTJWmx&p%#;4K3Jor$ETiU+T&sAZHXfZHFl~Sbt*IIJ*j0j<=`@J8Mib zO}gCpUwEg!B3LJ%KS*=&Ra?ke>=j{ZB%LyP{{U%G37x-fJQ=5TRj)WT=q)B16Iqd@ zDYsj)cE9o~nE*~wRDR$E^(O49s(zA%_qR?g%_@+VOS>ZQ?ZEdc) z#w_~evCo-V?}7mza*FSKD;_zF_Dxk6N z>C38vBXdn(55_9bSjf?&eU56Lc4Y|(dby62&*XGA!BhC9vc-*!XnSMep3(7(>H;A_ zyoWvV#krO(XGpoA;7~22!)2I7wav8Nrn`3x+`{YNRZ={DS)-bka%Y4w-Xw>TioVsh z`ElVg1y$Mb+iQO6v@BQM%T2W&1bU6tQtm}HV(u4zHwIo<7#c?h^R_Aoy1T0li{N*l zP9^)d2?q!QQH7x>yM3LZeFD0Ce#M*py|k^h#fOL=agA`6&daT?xfhklSXzczb%vQmym* zNis8Op?HZ6U z^cH$1;@`O8&I~2927GqKnxP^AH4s|4@!R_!n~KhNEZf@i353crP{}g0jE+%Nw=8SNO-(jn^_q*uM;RIJ1poGVmdxm;*U+NS)!8^Y{GsAJFiKSh|YRx(;O zSk?Yo_j@)!tmG^~Dbe{|Nkxnhz^#%wqyodhz!j#${`_Jw#eThuUg|W_0T%>?&O8GT zjh@e9Puyr|aE`5dZ5Nb*fg=}5IANjac^%c7^d_hE)!>U}qYw`#YkUA*Nk$w0yyLj> z15vD)fCde=iWyU1HukMsSkLB?$2!Jfa+2f+`;3XHIT;CEeHpz5%i55|HCp)!uPIZn zC@0LHq`U?mm)3C_XFMU@Ci;3e4W)qbMV!2(ns6tK>-?8J8>wJ3uOSoY8CR)!X_%-J zV&){l5&i=~AUH7pPsil%-MWD)=E*AIjSc_KhY&RHCm5~htn78h-TV)|bTMZI!z!n! zJ`Vi;UKsc_h@#*8EcP~JoDxYT69UkMUTT9K-Zr;52V$ji)w0uZIdtDaE6plg7q;GS zy6U#Dv;_9!vZNm}O5-BEv;WxA%A5!X%3Y=L2h)sLOw;)3{CIi6JFocogAE`*0zSfz z*$HEm7P0j*LZYLsOV@#}N*o@Kn(&N2sLnf%wTv88qvi4yjCDJFb zZJi0b^t@a0vrflQo%j8Rxx4*G6$GdXskii*Ls+kzmT{$dlu_k4By8IDW>3RnoJIR) zVMrNCk%X_)Hsfq8u)c-m@jUNBh8-HGlb=GnPVBKwyh)P=_7UopBQ-p5fmL)r1~lC+ zZeMgux_gXzU5!IP`kD8SuCoONZ}+E0t0#+6#sp-7j7FVYMZYz6ZOqNV0)=L9(tddo z@ywavZ$$7-=hRDqJue+xfWQe^S8bYA!4X*p^4r&mqd z0FY&UV{M$vSK#nKE15m7*(SGha>a?C-1ZO2?!hN>*sBdMok<_p zvKYA^OR)Ar3ViFo@qG^(lIRf7B`=+e^xATS*S8;K!rjEbdasUAt2MveYdIUCJfClQ z2q19#oDL@cb}vQh88kQdAijO`Amb>7sY(S0znif5v{>+YjH{(4I(S{5Un^S=RwyS=FLsUo&b z$!Wf2*07?Enum+OI`n}=?O^U9{-fQ2KZ)3NT5O=zmD9gs3V9V|e1oSCX|Tn`Lvy%FMs_zW@g1a@c}Nv6!nd|tz5c2 z_iYO4n>r$X*2YG9`XSJ<(goxeYl^%@#Ul`vHpFY6hg>l6o_7(d0yND*wBBRJ+OEgV zZ=&ODL4sp^YxEG`Tc?eZy5gfL?Ia5J)Xs0Q1q>S6pyJ7hgyu-2yq~X^?{JA-H%XoB zm-Y6JcB)M;O!GcsTF9j=c@Lqx{igrlz>r3InMjr7QU)=l_ma7R%uL=ZB3MJD5(7;a z>ID7RZ{S%ZuQOa9g4y*3R^ESQsXg#5J>z@wy1?=+%kQnN>Hq`e{8Dql;G3p7du&;O zh1GOK(1|YW@T}38{QA!0Cn7#OsY{|-0pQHZxyp-N(5yXX!>19e%LPj8JD6mc_nDKYcF66qZ*F?c*G z!^tq>1L{Tbs4LNw=y#?5u^Pdc`r}&Y@$4}Me9MCyQg_-%9$eu`>Mt=7LE!BZg!4kv3(|3ab%g~b}DJM_Ql78BvT|x-Nnc!^IY7Lr`k`nX?ru5X)Js*82ky1% zsYHPd2^>m2X@n&uHhYG#i~vUT_hmKxZXO;;%@F(Onj-XNU}Vj){2?wB zVRR7Pm~a-8E?z=sP1CtVba=75dp-*qy;-%%2JU-A&xN=)pCu zZ+anE?~*rp%f&uDp(N);92$2obFFIya=h_;xEuVUskon(`WR}`)a7NBU-+r?Vyx_x za%G5OTAer);_2#;>WMbZTD|XZP=D$Pkcy5>TEAq*eZ_fdHr3h2b5s26rV4j;#UPV{ z3$%1Q$?`&FqI%%B9$AMn@l%8>U*Va)n%umKyqQa2t_2Oz6W{$GUl*1OO3Sv7BR(+l zM!r1}EOtThx@%^P?ly4-pZritoe;XdeP`oYmYbUq(ReMWQeT|dwN3^+VzcmiTMFdK zEAu)Qnx*PoL%V8ud>}0pN}XWoPnbTTf^5~QqoAT?yd=p((E(;wG$dHwRN=SZP@3GC zlHCFGQh#klUA928Xp1}3Zqm}CxEuSg%R{=C9E9G7biaMJSX8IJ^6UEqpz8(OQ~rk- z8RThJ_9Kegkge0Sw4_Y?1~c{cGz}dysYf$n-klGcdiEa#;F5jcFRDpz&)caY_TW%* z)JsOWhW2Xkve!P_U#-+u|0fJjbaeQ7gj%;L-|FmoOfF5yz}x;Tb>FcfP&u$?pW;p&_N22dO?THl*LaOH8npk1y{hB_@%J*)bqZ__zAFUd(X|dieKX({dB#X zF3^!Q_4@q`hBBNjdKt}zqytj7{go*$268&&UJ_Yxr+&|1@nKPDyPp|M>l2ySkkWs@ z_Jr$k%6>MeSLEtJgNrWA_)og|mzKu(RBI@wd3>uYuT1KNB?#i1IS_ukvV@uBCHE z4*+wsyRD6F?CFY=F^itAVx@^W9ffd4{E_p+NXPu+K#+@EX-Q9{bUNG2{wJaX%GqWYZc)TFxIiEC zDUbMe`)NTs+V9JOs~xP3F|sI%A@;UYV_a^_9QAsirn>xEzdOTtc*AFv#_N-M)Nsg& zf>z#w9;2gafRhqxCu<5> zx5<3fj}_RT@z(OeS_Ao zFJ#FCEcd=}X{|JFk7!jbu*XL4{fg>KUF*Cv+s z2On~IS-GsKT25-|WuLqJ8Y#K00mu1*1?`uFWN;Zn_-77YLFAHvhZ%$5t5=1&VH0n( zTRiEKhma)D1O8UO;jg7q@H3a~q+V=pPJQyWT}*F$=WJ{&*}Chgbgs0MMe_rnGm0ef zapvoy9W^f=Q=ECFBr6#|sZ3-3ziN@K)KB4Ea6h$N;oC&Pgp}@x9f;soziPM*PW`tejzM=H=T^=L_tV0`h=0U5vShm!2=9@p;Jc+2U#wmygQKlcc}x?tdtC@gy%B%Kch>g=op|JJ&zb8%)A3X2R6$;3g$9h;+; zE7_g-tx5hhXm~SLL-*s21a*x5r_9Ev-@3aJ=FCy4$sczMdhNJ1*iWAQnV`QBm1=B^ z(}t!!%lU%ZTKVRFp8nrIx8^G7rO!S@wgVlb{@2T(oxAZ+i2N^zA+yI)F9H4Nr-aTL zj1;+fqf&=>=g-x78tNJh&OYYSJ+HkA&oK3v8M5#pJZtN1IG5iAWt>P5ze_)9p%p_6 zTANN5ws&i=WTR$2O02tAW}jIc7b}hPJvFg!#AA8BPTxpKxHEQ~mR-9hsxfQ#>YpF^ zTQk>pw#8%UTz3y>%OUXn+288v-`69|mpTno?CYs(@_*yT@9KH`Rr>xliG!F5_RS4U znMg>3Sg2KCd;gemg!0d_1wf3;Suk`L{cp7SLk58lxFoi3M|p7Ht|u8>$JM;(duWo? za}ZJqTp90^w}@|zxPkDK9K~ewMPLp2kxM6tMW@O?FhJSibEdP;d%_531zAN}A)%z_ z_k73AQr(KF*&lRo=G2PJd)op?xKz|s#57<5zMotz)u+B@az_%rO&4a-*?N4KH7kH4 z4oyu>J5g|$JSIE)sVqQ0uqeSOAtKA@*cexmn-8nU`NNN$kqXAkK*{tJ;vgt9;1gPm zyeH*Qk`QV_J|-R^-^TK2#>GHGbVEm?_F3cL5aa1nI(5XHt0j+Pgom`_ zXV|o}R7rzJm8hPn(+nS3p&vbljl+9^-fVX1_gk_JMRe)z-SS*H)oemx1SVNsVZp8` zo^8mpXf@25^&^7&S5h^>MJ&ziSp3UdZB|CzkyraF4OeMCH`FC7Hy`nOh}oUYjlqNy zMP>nVu^~-Mgo>ua3F~6p+Moj;XG545lWQp|?hCZHKj}csSow#d7tIlTLxpN3i55P* z*(~MW?JsUkH>8ueEq?CGCb*V3nd2X)ivR6^ z8_h_qIvyYRUM-!-%u9BJ!7!_Fb$pK0u|MO`0 znAw2SWUg8|?HgK#77bLKH}6Kgr$$%eBE8SO5N`ag3z)0&V*Bim-id@-58X=g%ZtjO zZ@2P!ldI8Rc8fK;V$DA)fxv6Z(sG$dyjW92>^5UKt2y_;QPg&u=6Nb%oHrx@9jInm zia%zayIzS_{0RIYJ*8Uk;0WCQ$^+e=?`?M9+&g@^?0N;zsr77FCTa#gZd|f3916Rq z8FvOiUT#SFem-ikNLmb}RuhVxJ{KtzulTH9F6`?XO}W0O=b3{;r#~8pcE`#ZLHmCl zn*aTPupWw&mFLe3^HjGN4tf2G)Tk+Vb~kh;C6_M!6)O##P0pdXQ7a94ipl&u<0}oK zrG%Wes=(zp$;?LxCJl>2e7Dv~+$JG7j22|VPYJl1o>|>XH6ZBpapKo+7y1yx|YOC7V zftE0v6C5npQfVAhY41K<3(8cNg-IFc<8kk#`xRfOo-{BkgihDixA4Gn*H}Cj?`O^9Xyj9ts2|K0m zdK%Cx}dpB&*fZfo|bxq6*+Tn{E%+B zyEOLkNZa7jhlBLVZ(f#m$vJhVuI(h@2mnQ=!{2g$)d=R=#v)}ix zf!-ghpa1s@S@(0xxFV~(+U5@Sf#O5E06*-_+;47bySi_Sn#_3!_@@=B&8G$=!{4r7 zNr+5YF#nhL{HW~Imw#H2H2N$gpU!yL23 zJdh%^?CprtIVl^d>Zl0yL`9={tMn zTuu_>1ytg+m8uk}_icZrcV^M8?x`k58Q7m8J)p1IrwA9fxe396wB#0P*F{7n7Pc^K zlsMKIfn6T%%g35oR15eQa2waLl9v-T`g@A$Cdy2nmni$aa2eXoGvnq$sfTLd@$TGt zt)~E+BDZebca*FaE8**eF1MzCz*Sbd$Kynq4i_uaoj6eNC4B-Oi{1j7%Q0I`LY0|E zl|f>cjeY7bkBn@O<1W4IbioRy%b6~SqQg4T@ZJp`c(6%wMHp|y#}S87@seD2{1b8- zC?C+Bxa6G1biG2X^TE2uSE2vmxAXJ=zUus+-fPvv;x6_E@)cY>rz>HK9dV;~D~M5# zGtE2yG%ur`8?q^KqPs|=kedA7we(5Fw`viqx%?;`v;cJu@39y<&Nc9{bpq;I7w2X`Fp?5tx*JBpdv_#cXP_!A5YODLa5%5SwT)l4<2_TjzZy7?8AxoJcDaw|Jo+_G zZ6RRD<6_6K#IeA#xwa$&bph#@V~NVIEwY%BX-gD3A0G1`Lcq33Fo*n8Yqd>5Sbi3( zYHxTA|NTa^SL@wTMPA)=66jo z=k{*%@1c5r;mk}8CKFk`L1MVcz~L`Afxvig2&AjSZgEqZ^R*ZNm3f zU;6ZuhSH(=y$1-DB}!V6VrFPa8+4NQmY-oUzkuoPT3{Xmi$t|)^Ky;&KTVbY%X`Ke z>AJxoQ{c_C5{Ym$IwsjpNc3iQHO6LkOhI3N+`99WRfFu%W`Atx!8oHvg{E4;dW~I; zqx_e*DsR1Zf-o;v&C)twrxdl&llok(M_xwlXZ3CH%yvbPemQPP|6;wPyRXEtI}OO$ zPS148yr8GL+tc}SgV8__=&nrxyUZfu$$D3P(N}TaSgiM(saE%et^x~cbQ&r$kipi|%>An-A zrGo6#>If&wNOxJ6)xcq9T@=jL%|ERi}dXU6Jt7j=uM)8rbh) z_H8MkZb;RahX}V!J11naOlLY$<-$R2?!kK^-@iz?^#;n)66QbdTuun?+k36sc|^=? z&i^~(!_Xx+Ejd0ISAh2aULEOLh2R6GUvI8#Obh_@XkacQXw_?r#HKF09}DQT7>XJ5 zs4-~N(hPmdfbFuG!QR)>&@bp_iX-!7I&MoZ&(NZbGoQfNRn3hF#|0q79)F z@J_G@aVeZg$i<14Lf2bz0?!1{)` zH{Ude&8xcUd7l*?M@wV0_(4Yn081>_BBwYm0bOWutwx>QYw_UN^HiR1$B}A3Pp0yWq3ASZ^G?oGV*7mZ4w0 zF1A%{+;&w+^U+$gg;ACZp&~vC$Xu>kf^Sc9TJRddqNXKZ_f_^6pk6P%WLgw$){7Go z2HTwe<4Pw?wv7Ddoi^hxC-DY^?|SjR<^Ah_Q!6ror(UVDQz*Fb*bX$&)1;YYjfXoO z{|zEU`3Nw^*kT+e)q&c>AZGFSVxjSWay8#O3Z*J}n1g zsuE?c_L>dAlT`b3JdxoiaI@eTRX7F3W9)&a#RxMf<dx zUIXGp_Gg>H&POvJTz9XaCUwLjY4H4QeQ}@aC5cswc0=F!*@(UJ_(Ss=eNdUhsu>5d z#&a-E$WSk*eczj;0}_$(&X=+DYAAYhib9Xp(W2wS0??0EYi$oA`!l5}=+}`1Y+aFK zrz+(Q4(4w~^k3ufepyFY$V;{p-IysNxXjBiDnotc1nJ>07qaAA2*4Dv+7f**0~@$0 z;oZ4TYitj5Ys3QNS0BIlT>tta_g>L8-?1VaOtvzhqFCd8)D~(tZa*ewXwf|UZEEWI zk+Y$2p^ey18{!}~#7e}pUul`l6=Jw9sq%zxW_n<2(Ar%C=sA(gL&>Bg{|jns(S&z5 zKzA2?#FIL{t&fr+xUMRRN*$q)ak=2z1pXu z)Hdsv&b^KQbK3cS6_lg4ij`HXlCXkjzHR?zzGV8yF$~J15w@O{&c*Xgn}?rpwF0aSAz!XUVtaLwrSU)7E#xeYIR5K}E6`c+RTx-h5rs1rqsX@~Nic+kS+xAwcO63X1 z;f+B!*6Jsm3!VW^8UflFfSQ0%yg=H#d)*8zwQMx)iqDFK1}ftDH!JGGI`Uqsv%LAAi08Rdrdc&&kJIUbJk8%Zm7YqV;MI;R z@Q)&O`=K1K(|kNJ+%kZil;@*DEtdU88gg-f-D=%mW_qHcN>MWM2>D5&`(>PqYA-d$ zIeSfm!A}hBKh`(`JTcI4Ex~?MnQ(^Sh}(O;jN3H8iEg)+Jk&`DwzJ6c`9W=RVM)MU zoY+H+QE2qA<=e>o6_G~@%j$}oE1y$p>P(S_DuzzRfu3!H+LxE96stp?RNTVanrXPcY2Cu9%l}zPQ#}s_QSzY7M9nVTnIO6!ldcv zBC@3Q-yTgnZHokhY?iPE6))&pWTl)q1ma%9`r=-b81QjE*n@t|Y(feCSx0(U5ON4O zWg6)M`r>F^Q&oe4|4BoI9sVpZeYi!!C;V?Q{=0rq{aITkOJ@96+*rwS|ARRQLonZI zRaXJp!yJGW1^*46*eQmDSzpO&I$uySM4rFIFZGD;Hd9ZaNp zjN*!?CE^7I6pAralawlzKzM4_x- zNO74V>{OyeeVNk_LaH*xFGTUdyvhx?kyv~PyaEDtl4q;ksscc>>E8GGP%h25SM8HC z8f9A81q3AL$lqb>qxm9!D<-9{j2_?L;}%}0&7!mek(8ECB#;vZlSuRO93!^7Za}0q zGc<^2eqSH-V(omyoXM3cbM4$Kzq&WM6Vd zsmqbxq?gHK!~Ex#3X?Smo2KQ_$L2?QKK&^1Z>k~v?gtz^jN{;=Nd_VRdzZa2lmei7 zOhXmmh(}5fEw8TOdFvLTE|ofrOB=+~?Kz{m!$$Z@>Fo-_QMzD_6)` zYpyx(IqosWea{;8vg|_lv#TwbjIxlX0B2&=wL~e%t7LN@syq4wuWfkKg8$Y^3mTdm z%au1Ju3M%VSb>b@JAsa5jXf)#ptr_JTrhXJpy1Mo0wb9B9P*`J!yeTO?|vz4^bM+f z*fm0;`xWc+s(Cs+Ggi4rvBJ#_d+F>mtkszjrobuN0NVlU=z8`$Q8Mo}N?1p74@Nt1 zaYX_*&xuH)9X#x+SlN~!DJ(XfhUS9RM1ih*fB*JpPu4@N_JXMM)|E1-JF|_Zg9^_S zRjX?Y>adpZAU4KKp>4eyaHavZ5I!Yj^v^$ z@|X|oYD+6mNo0zEf{J|m2Rk&9S!Wcx)8gwIgIAa84mkPG%`aO$`kkKM^X0|`xj+0; z-yTc(4UL}dhR0}U^-B%DANQBgU;BMj?`wwdX3Z#v2w;jnMVG7z?SuO;&y+p zc(NO#gk$p!^xlGvA$Ci5T^2W&H--`{?_HF<1g+E+b?W;ROx{7cvM4g<&p-&~^Ur#^ z)H_9Ah^_5Tg9TA$kImQ(6HlR1Zy@PjE>WltgSkdJdT`0fvAcIhij55PVLt@ajALzL zmJ36X+V3wNQC@H85LGgbp6}26))@Gh7$7QrQT? zBF9JyG$SCra)L4zm@?wgoIh9pcBv6)%yKI=_X2t~%}i)T(B+|SY^;j!Esve)HTe?d zy&I6-@UAD^-cbI;eD3~EAILtqdYiVSNn!BZV*nJpPz!CGsa4l1T(^x$b34g*e;*;e zQK3E>C0#kmGpPN%oO@Jr^g?n)k&c0QR$rR#ugJEGYer3N&dX4u!1m4cfvP8F$@sV0 zAM^aWd42`TE(dY=EA1?>j2WrJ85Ra#YeO2$>4JW(%`;luk;f(M+fkLJ$NNcUQKv+E zEGl>%CDc;1?>zx`fX#0ZD$Pgh?W09V35U+!pX>anLUIHJ6-k+wU79>vIK1U6EG9xh zh4B<h+5cSIYf-|kYmZcdSo zBqo@-enhfLct|l{#*TOKNQZTT*~(QfE{@4>qIsa8QLyIjHDagG@=Sy%gm#!g1~K%_ z)ci9F(yq72KT38sJ1+evwbkkO7&mbrQ!I3UmT*O?@_zZKu$`c4=TVWAf1FJ}zDuY} zGq8@5V|V}+uS|qp2uekrVE<#aj;jml&nj;&zDH@`hP5^^FFOr{GC}`XnYLQ$!g#sd zk&#nI&!0!5%UU}!oWt7`B;8Z5v>nS=4=Vr%88pDtryZBf!KES!j<)ZzA1xo}O&tW7 zKHewbZW|JIeIc`l4LzhH>w439xhLz5N$4jx+q+46hQ+2DjOoA-qJhKrqID5@W9K46 zq56h6k`1I%-f+%zg1Iu?+UQ%nwBclJ&njw*Bgwey&--)>U5A{I*}sEU48m2o+%R?o zH3)SAS@z0SUY>;i;Q5%V;5taJh$ld>_%a`eYs#1U0m22^UdOD6lJTm=ASbCKIdsxb z$0K; zy#}}G45Y5Ufo9w}T1x3CFOYtp>iE!s`~>QYWqVF-KZKbj3Iy)~aA%En6`_tdZ9%)v zRtGc-$OcNhsgf^JY2g*u+s%^19F^qrE2B&8hqJ3oT~@cNRsY1>Oue#AuKQ=^U>95> zMKb;wl&gFE7ejC-=Msv96l158vd!m|2kuUIg?+)Nt`zG)TT|?<;lIB) zci{aM(_vIXo{L;tf4Qm9DPt;cbiQHttaazy6_e^exI5y!bh3mzPFjX_FtXy&bzT~| zT%L!OE$2{b97R|2uQ~SX^G4}GTg?mO5%s9B)7cVlQSGS;ox>HbqK`Hj(#-o8h^XGj zq9w~2{-yR6K|k&j_J*0W5NmHzkR7+O^|8aRqD80xZ7I)PCs^JpU4wtV6QpKx)>cZZ z-#9covL_RISlNVcKzPlh$UVK^BYnou>2=-#9Ly8?r0#jkvcAWJ++JJlM)AQ8DvV)XJzVU@~vH4~9_)l#zhKwCP zXT#y_6xgD>@Hj-yn|o2gZOl0;g(?8#8A%9xsPePxap9?tx(mIZBnTGVWRQEKk3S6hS(m+8m zly;+qeigv3l9CqDCmsypjLLae-FUotATU`I>RM6$sKP zvugk10upw(Vv+bG^4bNNw}17ocY%KGQq6QOZoXqBlVRGXL)Jq1Pa;kkWk367u5ZN9 zs9WVzB3rK-brwZLAu6iT*+Z?-4_!zx1t!pA?0-CfmD+Lr^vMt&p^|#;DC6$>#(ckt zx3tO19wV&rElIPQkT{vgEfVkZt>Yd-Dz7(|a86!$^r8Pt^~zUGXx*df_>J=Tg}T*O z(%(^zUA!ce&IeT$>(AQb%SpE>sFHl`W=BDBof4|_XEQRlkMmY3e*&yInUA=uX7b!ebB-Is4bqDOc-G zDL!-=JSsfNpETy3lja%Oh*3yPO}IlXz5N3e#LOcf)Z8J6Day8L1aD1C8g0J7D#8Kg zTD~3ZRrr&C9PadLvYA^^lZd{QNbZ5!{*@aiX*R%8UY(4}j{MCp4rW?_6@3H*W2>1S-$vVVu7s0K|`c2?-{h_}Z#vj7Fr(-H`%F7*Bjs;3ga#46w z=xA!}(UIyh9^UXKqn87yV@5j1ln2+&6qDXLZ3>nKeY zHOLNChK7jWx@uV7SE{ShA1N$(+$1SvW)<4#p|WPKVZz-fCh^<7pKKHz-t6FHbputmLZf~%UN?i8RNVP-i$pgAhsZSs) zjN6n`NbAnjftm+`D(wjc)#xi2i7kzH4S;SHl7Jvi5pAf4A)+eJuvgW=6-aL(tWb%7 zMBIY)esJ;N)yE5c_}ap&h4aG`HlQ^Kw^xdI*wps!n-yl2xmT+@o^p4T*G_a*OZ0%q zwSh&=W#PPTFps#F2A#MTpE0R};EMu!u?%y{PL{CwINk8DTfpMYS}nj7u$T+yXFUFM z6_7**uZ|e%K!r1BKhmc+2;Or&6D=oG7iy0q)Eq5Dkzos~25l{?GHO&|Qpgv)#3zAg z7XXJZk)RQ=p~XoVos{+_{g8MgStPctaQ-p!+kZn(48HK`(VG&@))SYV-aF2kpS)=I zVPm?hI>sBr(^-Br%*iGhrJvwi>95;2-=AlbP`xW(LPzqmSlL_JKzM<9*4iqR2o^$n zq5D0B24!$dw?SVI6#DOat+P)W1 zMq^bK9lKEhVu$v%&V070eZ+1}X!CP=1ihRw1?#_%kSJ{dSVp9_xnLiE994a#NW&)H zrf30#&{64(O1>2krQ!~N@5}&oA&8q{5Xjh=SRl^L)u;}GU(#t@bZqX~krGeLGga!s z97daYfs#g-tZ(9TZZ-O*U=Ahsr^i%+P z=Ck9R$f%|?(>+&G@cBOznE&c8_i^4!cRpWvb6vCbcv#e7af8S94~8z-CDxmKqsN8+ z>IE=vyZVFjoTs{rM&I(^Q|^MjZk2~NDu`_2MboGi{m{6g0II3M`jWOQJ8$1$2(_;z zYpIZ{q#xR>lp?4lR2uqM!EI{s#H1DU13$O`9?iFau_TNgtCyy5eQlB6)$HuQ<&V{M z{!rk>LP=^L+mK$kJ+2tDJj~=5Sgd7RFO! z4^gs(k;`+k;7tB@-A8aq#5D50CVYnTAZo>kIY8zMF7`M|d zdj}@xBeuJ_uQLNP!=;-KSnhpS8TfVhS1kM${r)op%m4oDOb++^-G&>6!CopoEh0i5 zVK3tPN7IH?k0_JVz6>R7Oa67mFBMstBz0$3FdO`0+@B_D0>5Oq3o2Y8o z@nAD(+P9%Pxwto<4R)(kz3rCz0#TDG5`Td?9I#tugy;Tm#^}Gk_z%5cD1qaOJ0?qB zC1#;!hKPpT9pH(YqP9P%oh@fWC_^_NBntgw*fyZM;a?0kM|irR>t0~>L7tZUit!3X z?C`abkX_hUOsl-B*jJ{7uN>>T2;;x=>)*uW|LdSP$A6(q>tnLhQ=jMlRSm*A>mRi3RP~fu&>5^)wM}xaQ`h zW4C_MJpabargIxQm+04ps5)cDw}Sijr-JTlCBtzmW ze>vR6Qx~EcxtYRl=l%v;Jdh4L5!h+*-+0I8c89gc$w@7~;&D`J2Ct4l4R+LptO);m zdNzl1t;SzjHUsAbtOr&3@C;m>3wQ&+iI`KUTkW zHtaj!@k@tK(xO;zl1Y!2lofnThVt6ScynN_d4jTk5B3A#5e#DM(=C2xp_nrPOXhbt?rd2@zE?Zk zoZo=vzY47{)GPKLy9DXg@7#9RYaR)2k6Z69Q(*@=>^v|5F4$`5efQU$J`wEV=?w{R z7qWGud%g|0JsaTE+kN|NeX(XWY2=r&0>koc=pjgMZ%DO6{(? z)&IHrKL_OB|LcE7{(olO|HYF1pB48%vF`Uq`#;g<-x-SkpCauBhku*RATu~AwgUe? z;{UxH6T`WAsSTu5bTZy)P+H{XGyV91Zmf?J8Zhg}&)n)#;;<*eso{WvTH2}HD_y%e z36Hm#`W7wk!t_p7bV-sjDfv1-P~fDKEL;4F|CBLcU)dO9TQ~e)+{#~HG%&-m)7euW z`A8Kmr8co@svYmYX40w`q!;>29VQPwia466YV!&j&?tm3Prn`JJf2`_+MS`9$^5|( zHUnr*r0UjZekOCFD%tdt`fl9P_9{HrqVs3csgB}W9nrkv=!1gj`5b5dcWO!YpZUv) zYpY^%tH-Wwag;Gu8u9MpRtn)exIwk%VS;(+{fhB(nvurK$52MXzU*hn{XyM@X>7`L z0{hQS)g2mpC^iJ|MuWX-%VV^X<6m-cJ@N-T#9NJe{;EJDi!JL@5 z%Ld)#hSjZBug(l9!C6e+ z9wiv<^$EZ$DZLWprgn?^-eB?70)|9Fk5UGK-h;kol?Zq}`;u|dr`nBH$%>J%GPEJ) zW057|V-YYzaMK$kmx&BbRefZplA7+~sFH*^!LeT=uv!nqDnTMy6HOp0h{H@!S^KhQ z%L4|OKmPq&{WUA!@*cad=p&=Iy&Wa9$OReNM(^J9D1B8|REah%W#!|4wx}?y!%lXE z{Ful8Ds)9m}RlGCXUsiRIu<^Yc7jlM0F%y#*&t{`b*XR_c8J=F4kLh zJjxGmJ+18q>wBX$tSM}66@5y+EE5nSVwHe$l8p-xvl8#TcCUU?|9pRe6Hw$Djbrl( zOTXTmm@Mn+Jl&Si*+plNk(CeLiIYOO`J8)a+Kq5VQclEA!Qfo|vvlk+O9%8izLz2s)RNvbEzRwp^YxT{g%W zR)nM@FlP6<5U1o?d4-1I*z)J2!;!)=H1TK=wXq84A?rA3n7%ty)Jn4RcNxOwG&dy{RlE??y?MO@V>XC6*%5a@P<;6FgLlqt0|ghS z;-yTDBaU7;M-oCDb+&1Fd!Yc_7LVuw>Mp2(kVYuw{;GZZVwY}8eJSxF}{0e`yFKz?!P$l9G7)wy;JulUFG+wu)^V!(xf72yAqdisKpx;n_dm!o9p zlzaT(0&Ocrs{u=^d^f(1h_37STeMK$7aSy{4EeHUEG!GrH`CcY3@;awGVK!!SbGfm zV)SAsjjKKsuf!1nFX7U%*rrKC2uDOd)!heyawMu?}aSz4^q`rBwb zZI(f4a?*W4tW?UNxlddQ1JE62qTw04)?F~Ds^RrPi{!I@sZauOox4!T9@2XYHC}pq zC?ct2Az|d&>%o$#{OQZ%qQ)va&F5A75DTKP#E=OMmKw}LgtN)>?M2yIHP@OQW))21gp%vWQA#{o-gaC)uqma` zt#&m=VA{zXSS|3SptZzn8*c+asM_tGhpJcUmS49hQ5<%sMZlJw+B8_<4?i)Uh{yj~ zg3OA0fu|>{+QdsfyEY(Xm*`mTH0Vhdt|%FhGON0MuF*5d5^R|1K2d)QKSJd0uq!`?O0zDT;CLL0Mt)rH)!#a6< zHG{7Xg#rX1#t6Jp=<<_4P_YuSWNQ70#dD)_JrGdpB zy%%w#VI`rFtT? zT;AcjMxIdoj^YVX*oj9Fx%H`7C25qNFFWHna2M&g#52$6$jzrqvU=gX>10Yh8Q5Ib zc6+ACIHK-z^n{lNH=9N^_w%Mm-XFz7an035{a{tX6bAQ5W9P2UQpK&<9#?3A znL1|ponueDD#C16$o68tX%btdHwFEd8z1wyE@1B|L*~h=~&f6i_ z54{HdMRS9MSO$q zkhb2*szP;}Z4t z5D@NjEy-o@H289u6j0A>m{R-v-Ngv|mi1+n^Vn9`9qL6`ew46`>&n)0PTJxT2;BVD z;HY1wPU`hWnx{~=-r^|9Ef$}DTHg9)p+3e7IE5)nWU}0j_AzK-6h^%T!{)TbodEhSfyW(J1nfqHu2p_TH?vY)s7V02y~xX zSs0)25Yi$*@yV5QiK(sb4X>ho`^zhdf4U+6oX30iJF)_ruX6mBIZWaWJ+&F$*QW*h z#g$}ytBY8v$&ld=sEI&FXG-N4U>|*ox}PejL(XyZZ%Zir)70D@*5X}p%~6AAGs>Ul zSG!|mX(_iekX+dQ_MY!ddr#!;l)(Z^dS^HQ#iNdfVV92WmI1ILeaF0;kG=8koz(sm z2Zs-;K{T#e;WSuke*siUDazX7Z@`eqvv&pDxVk+r2mf}EwKNs`P8HAEVO1p~7upq) zxe9qVRoUNf;A8Q35i4^!Y~d}`E<^;{RE#7R^y7$GxnrCFN^Kq3h4lRnD?-Tc^b}h# z_(@*1GHWLlp=ZU8+-l3dIKZ6Y!W^kx(+1eA1mi#Bi!?&hE zU>Si57po8y1@!|XW8bg)`oMmNL75voVY3_r8E)jI_rg?TrwzKaJ+j(aNwM&&gM5HI z3vjCdyip%+b9htGE%ZMd1Y$p0{FzVeJXWR0CO?a8wTb069_%v%W*o;GDUS_obSGUJ85*58uNuR2swvel_eCU`*Zm zWw`x+s3R=w7c+X0l|Xgt4C)@n2p@C{H-VL0%o7)@P;$d{&-ZO*Ry8cWA(P|=$*I%O zu@XCUO27=asyBwt#qsVqxiJ^+Hbm?`zesw@rcMi|`NGtkY4q_t6uGl@iF~E-eD^^i z$@i6o1SXMLvz$OvGafg;HLweY$_S|jF|5`WZsw#JIJE*Hhl?`kg-Edt` zlEV{FP#%}A@w%Nn6eP|6+sV4w+;oXt4y#4HAc;=^f;N?NG_@e;uX}RGXkmE|goYn1 zMnnDZ>cO)Me}t$9gz*XmWznsYz3OF*jY^)5VtuIVs_+rvA$UFqVz~vncO2UaG@IN^ zN4~#e(gXQG_yYF3f9$1h#df}~K^8rP7M@Hzjl0BU26XGhDw+~#(iP18ewfRihch2^_d7%8=C4} zxeC@?0igKTH6r`T?hWGVA0vwyh8e92WEX$sFpTQn+cItCyFnr#` zJV;gE=*&4E7XU&miNScfO3&8<&KN78hOfdx1xHGgsp`3hV1(jW6JEZpE#g)5Q!uiM zY|pKB@%4p%AF}0uju)-=Mlu&id|eem&wpM1GyVncJ$9^fF5B+OmU0f*MC|{Z*jo<) z34^yO^F80;1dpfvwgn+*$JE@*?=g`*Z2WeQh2OF!sl*3O59sC=u9BIB@gN#jV|@Ed zLkDhY=o|M6U3Z-enLd9nrp$kJhP#yb_QFHw<;lh!pYRQ;l=)?qp68wK+}G!U!%f+y z;O|1pmU!zhft^aXLFtkQ$?`5GZqPq13kH{)w74;)3&JK3c=ZgS=HM@dp8XV`(}r7! zzS(Zt0D3`(l&ifGYW`(w@kVEvQz2e)uK4~_p9i>0UqlPTWSsTEYkpF>Z!yo!Dx4!k zlnc$PJzqur5n{*PKj%}Yo?%pPQs65HNXZ{_AHG^P8x6GU7^&Xa;Q`&vF7VjRxLcr` zq8M>;Ygt2RRX~l|oTfHBxP*@tyYY?uTsxNWSjTKF_JK2Be4<{A}*hqJ$` z?YNG{sjCuKK9Anw++xJ>Kt+%6_$&pwu& zn1`POME1PR?&IJ<4vm%O-18$3zofZfB$W@xRnP((tS1`9gH(Nn55i!LmVl`^&ZsSH zS`OKJG5Y96tisc^#9~@tmp`o|ho8L-fL2nh%FHFpPxk%uavKe?na(r$w+22TR3OSz z`ff+pViiWPKFR!B_Eqg4Zglcut#9g_k}J|p8BRtp?hkk)9wlBK%6^pWn)NWQ`abxR z$v)@3JG3L>u2hP@&vK()=Dp;KTIm;VC6c|O)B zq^u}wu<9M_QNnLl-Sx<_9DXmt^Q_^JGu1OE zXAqTb`FUwG>3@foJC}h1s<0{4u6h1JJG%Tw1@pjyXy4!;VNt%wH{>c08(bIoM(m*B z5vY!3h2=zjYs*hFDBE`Qkjyj5jqinnsJpd=@?duXlr1_W@TQ-t^Vj_(JRCd zI;|a-Z_pI;W?A`nUBtiAR#r$sVu^p0sLxjr?B-(Ma%??l0$%X=Jo8Yv$>l5Eg&cJb zJg@#UW5mbFulQst$h`Db-Z>@?uI33lm_oTdjx(=&GYYFa&2BRW2(-YUOG=NIwt=*~ z6rgcPe4VOCgp`q+%i{3;8_}vV_WM&8Tr*I!m~p2nIp2xJK}-L$$`TD;a5Z@QdL>Rt z@=*I#9gPJRt@?cLlJnIm76>tkzq9id<2w<=Js5i2nI)W7W22|Gij@A!!CbxsDvK0q z#eLZc9mr&L*X8C;jHfqH2T1hw;IXB>VFBgv;SOR20Jd4I20fyaTaBPn%Vvo)>uTh;EUI|o^T=7E?Vs}Me zWR+>#ajAUW@4}F}oty=V@W*K6u`bomE@qsY^bs#=aJX1Wi$OLKN3W$wk_RzhNs)(S zH3#-0z?+68Sggc7s}H>NP?Vgllgyl7o_Ic|>IP7cbM>)vy*W5g7(QliXXrRCo`iPj z{2ceV2dc38x2O7PPXEht?;InZ2GMwx?12h{D z3(UM+7O)kku+>$KN(TAd6&BWarAzW7xBAcEeng6xZ^E@kbh=@km17=-0x4)-Tu2pF z;5jw&`IQRbRM2`=RtABuuCr)2TlK!_Sw&j?)a#&y6oJLwPUg_boD^FrbD_oQHV6@z z@Ajw{hz1>G@sT^MYBCqPUc)~g_Cy*pmeT2 z$Kr`q5x43#{H($k$P@a|)QN?lXIjbgTx`ojo5L5e+5?OcMv)5VyISUe5M6VwCu@X} z-?T;D`LT-d2&G6&W;}PKeDpE3y=JmrV2~5fUYDsPDnm|kQt%+ABTc~qnPr(O;Un&p z*RfDJ*FgYAfwpTn%S+2Zh+QnLETtEgD}!A#Wfjq#j56*n+oHc;)e-hG4`^JcF+Rol zj-_>Y_Af3lC72Cb2nxb1ai*aeRKs$7<#l<~=Ht^n8?U=m$Hrp_8(Kp10%~QUYHr8w z4K5F+@JJJw7g>v$m2CeLSG0sB*HpJoo7LFn^`p(2yvJmHz#Tpq{u8oaRX1mESRS!p z9hapJK6il5@*VW?jB5q=6U?hiJg%l^wz<$Tb^0 zzkQ8Z);$XM9k;T)yz|;ix}?q$?x={-c<8|D`JlD%kdc^MzikH@8z=W-{U)ktqxl-4 zO!=`C2s6;INPR{V7Q)uCOVsXP@oQ_*j`K#PHVN9E{y(7`D;%(G)Ow#*V{dzjthUKy zYx85lW65p=mr<;5Io4<1mKbu~&KJVhoshusUv7NlmYK`_5)B191j(0t*TmEZ0G#r< zLN8_n{?G62*^V>3scyq%Eu*xxKqai1f`KLj$Q0JmU^ylcAQOnwb`+oJQ*w(lB5vMb zCCqZv$%P7~lyQGoB?kxK*mZgWn??G`u*zfuj#b~42ys(%ycB%A#)IGRc=_1;e0pI$E00tb23XJXs$8N`+C6qyniqzx~6&Tp##$U9xJ8$dwxS=h60aHDKWhl1aL$E>%uaLiUN8NCuRA~?LeisnCJ~B{9 zwz6w|ACsh62b}7jU-`>88}n_8fR?yX=f%Cat{4|TU$fGKe}Cn1{l1{B zbOB?YHu5ZjFzYipq#A@#=b$-o>BRRv?f`%@$_$V*uwyaT=b8Q(Z%X&uthU7}iyTsC zCbB$`j+8~s`(T@H4(*F(`oE6$09OhrA~~a~+3J2>v}%_801b71 zBn!&;5W=jx-aCZ%o_WAXyz~PBYughf^sTqHO6Ir z#f3BZF4X`BKAP|}oF&uulOgF9)-Ep4o^D0F&yKrfF>*@5@%>z8FnaQInFM%e^KIb9 z$6Ix?SNQp=Wsdg);?Y}gJXpFcIa#M41MW9>nN=WfPAnZczfCoX5|XrU5`&k4gm=Vo zhzYyKK#co}Lc^=Arq>^CO!-8d8vF_#91t#d{8_n`6mR)4Q{GXAa6;Cv$TG8o1RD$L z#Z=EraK!SaB^r2K%|;tf-5QEf+At+f{uJ z<+$onX`bNzZp`znX?cQWfTDyq#sj5{pry`_y5ZTW_v~2guJeOL(+Ogc0 zddthVPjMfD9bFiSk?5J|^6z?(?gRF>thJAZ4q2Ou|&F8yO&?W9r#ToYlEtqpYJiT5y9t)&byA4_{`(t#@e@_BOgSmjotpi z;`llk&YGVWszPkcHH|=B`G}wD$agzfwrx=0;i=dWiY20(F3BgH z`+#wRr!7W7xy*n^M8W>$-0j!yv&yr9-%@c$UW+JPJ24$jcr8i~*GZCn{jwwUkmzt_ zxkskf@D~}iNG54{A=Uqwm;l#goifDIKSF1F9S4k4cI`q;T}+A+wS(mLnm5HMsbdIT zSY@oo=HAb{$|IFRedv%1Ag20ltiER4N*aryCUshIP}N&48=c~-O}B!TsC-m%voLLI z)}B>1*1#_C&N%$D0mm-}YRkYXO(0crQyE4ZX||b0jau4Lgk8DKK!~-+mJh14(Fpj= zTn;IE5K1BJHE5h39P>NhknKB9yyi*jrX#nnHLNY|2Rp^M@LDJUd~oc}PB>`t6)Dtz zERX{?klO&d=#&YRDSD|@w zzkDeSA7%^eoG)SUW}pH*P%eWQGJN-ubr~9O{#PfI;#jVKc$EP7d}o2Y+YT40=xL^m zu15%;3Q%DbFpv3MtOL#!%7ughMhstJod~K|3}33`BOJi_GDKNeVv7l)0M)f@n)wQc zNuU zUw<^f6uv@)tA6Kq30IVN>LD^ft8F(e3(WheKsqb+wIW;R)YfDA8oP_R4w~IFt&zsX z&1;(*Q}`U{bFu;rSidPhdrGb^>`?ikF{t6iplkOONeLawb7JoRZJNb?2D>i;YZ}?q z^f;OVd%Wv1zFGT44e5v$aSx141qGz`$H)2D`3$Z7oocKRY~Cz*KIRf-xvC2(IFmd4 zsmr%ZR@Gcoc}y6uLEe^KWQ_&NKb@q>oOG+6yU5Y>`YxaE-8S-APz`iyTEe=;0_B+__#yu^{&!TY;b;D~JWc|A)`g zk1b(~{i)wXUm+jq_cH+O^p%=0bRR5-oT;V+-P|P6@ukax7dK7HVwX*IHyh1duj*O` z_&yu#z@Jndfp;sB_Z>QA+r%KUFTlIRPhdQRYz*Vno+%DY#0z#zOjDye7XZTpZ#KOJD znE0J+rSo+jU<<@VwRp)fMWb~Jkh8M>+VB%Tp_1>_bnqSuQ)|~YaBxb_day$c0s514 zR^+_+qY>!AObs>`ukBN#Ip_LX&e5>#F#Zrq#cialMfcWJjbAC#gUNq~m16~KIu!oL zVnXZ7p9Rd+Y+MLxj`7Z21ebHpCv9-7grT~5Q~C0Wa_&9Qp;;r8ah)q+J>cg?zi0r7 z-d>eqn&Ix40{$H8DE(MCj(lD5Zsg3}&{|&-+f8mH9>V=RY3pyk^yP z_+5HPJ@nARF%xZz&3i_v<1IuF9P(W*y(7Vvpab|7zOcn)fxdDn<7h*1WAj(tawtgj z2>XU_#+a6(Ue2QwTD(dlyW!=g z!ji!l!|pVmq{19Ci|N+5`cXFb>D3vjB?>vZa%b7h$WWYN-J9Kxcs~ZK-FWoqC(suk zwK=-CQ$)`BFiz>spwE0?q%t{Kub!~rqqK6LF#nXsIQDnV>NJsSOWF@n%jdwOL0wAU zY7*2-6w_dlbJ9$=_V&LCRlOx=zIeMAbvo4q4MoR`p5}FCbyf1kpf4w$J|vpPp{TLC z(9GO(%mp~}E*M@t2tHi3@k+!w20eoq0zKPnrx85d=C(<)F(m$aO zzI8M}f)zkl-9OcCw@MnvfSCZpLdRRVoGzdJ=pJJ_uDY3F6Mjh5PxTA;rii@r zGw5hS0=DZmxLN2vlKH`5BvxtFE2y(<4Cpdb4Z<5Sqi&Ri*u>uYHcIni#LC^m-#-5L zy%#V*d{I5KG^LkGt$$b;vA4c0LB_$2u#C)J@;5S*Q;5BtFKGTk&LZ$fT_81`hZghb zNmcpk5nNQzdugb|b=91-E55s*l}lWk7$B~{_kroYcG2j64)!q$qkJDOnQptW@OG*< zRD0SWr@h&GsF(y}mqUyJW77yM+ii%Uwqu*w^m;v3m{S<5#SlGGt$C12Dlg9AtpC|x2vN9RDDAn)MC&aM z30qWjj&L7pi*v5n z$7ROgT9dZO*w*)#sSGvoofd7Ol|kt86`*!_S1pFcZsX}LJ*?c>iuJT68}Iu+)8hS$ zqQODJ%48*6N_qKf@3Sr{*PRluR^?)js1A&JwM0@b7ei;zlHSC_xwHspSngOuEWDyw zCGN#46jV6bpGB>+1tg?0mAQ}&VBM$S!lpa}m)>Y?C%~r7?F7Nyhmufk(4VJEwgD%$r6oSXzDVW_;Q4IZ`(cAN`064q~H zrhBtV+RHK!4rJq4?MpaxliC7mZQ*g2wvgF$2NXR2h5}e+to4-F`)JUIo+G6t^Z@#- z^{!DLb-IQe)Li3tmz94_D}Zhrn}F`HIN4Sh^|GTaPSS~9%hn!=Y|J|$>SHTxxjP^P z?I%{4msBdI!N;TRw0K~7U+;Ah+UvafkNL=V-}u=-OiF=%arO@rq}#3nW`uhP^Pj{# z-K@|yQB3JJc{iOp@YQ8Xh7eP#U2t9XCN%sI40bhDu*>#@G!)B_u{t1sTc`t9 zEzFzxyD!_nXjBKbm!n-(m;GA8bP1yJS9-PdsoQ|1BziuOkpi2EQt_uoD~w3VIxtMm^PG$k#kc}L#j%7H0c2MRHcwHh+l2j_~OafevaIk0ChlX@7u(Q#htj}Y$Q?$~0 zYAjZN%LK!$=At-i%$i$XL~}6#7oIIt5f-CF_(oVkavTawD_$F3g2=f=BiAOv1u+W7 zi392G74@|fyp&~$VV>1LPNA^XsE_E&S4RM9^)B@M)K4)`KSVs-cit9M<+$XKiU{aS z_y}BpGMg@{G)$)BIx$s?B`-C$78S*bHE{bfR5FLsC`?_FhMubw>~A~{_zLEC-f^~b zd^~zRGs8k>ur|RRK7ZeDe`=&W8D9X~M^ThCPX4GOkjI&#d@z>cL=C@%j;V z?8%RWZ&?3e;Ps~N@sEh=wP`XFB+8ZYvBTW)g5{cg!s1Eqp9ZzXnGHoF2H&l$ON{RodaS>4^IA)| zW&vMfPBaryG-tgFT#w*1=pAi83JWTn!m(QWd$T*;Cx)x#-MbExCh#1UjE4?NRWmd? zJM`pgrkfe;`+8tgy=IhPs?6{Xpp>viZ+_N6Tfv^3npf@x>|ie5Fetx73J;}#Io)^d zP@z||!g=;Xa29fD2UlPde&<}emyR*s+O8_Fo}#FytCCe&yR;}cKrdy~q>KSXJ5mHD zHYERiq*+>-bDH&|L4m!M9kYiZ=7_zop+oTi28QilYBiDZsMARZs@+$Dg`4V1{0!)( z)UCkJgv$}bpIZzBby`<8YfqjbCoi;K-+SY424nSME<$Dn6=S8w%p^PXCG zV}hGOp&!NGrH+RRRPdOCsnI}N6e5^qQ5{eo=T!;KpWv!SAyQ2Y^tZD`T#L|{s(J_? zU~UdeqmL)a!NKd_XWT}rihmBOt0irwz@i{Y0Kv1$28;@!IUeD|-va4)5)C_b2qJ2q zmj-z{`Kgx5#mtk1FK^=c{2j`Hpj)~Q&NuFM#V_B(b%V&!Zc0_! zI~x*RzA#`?i}eBQ5?KpG^o8M0n;6yM(c$&kyHUD|77P7h2eB#+H(#cDU%dFctnpv{ zm_=y^4#N6#WNKHwaUAH<<5{^osxh3Z&)zUt<%=^%dwMQJQbU7VEkED6{rG*!m)n|$ z4|DG|eDC@hXMn|#91uMAABJ#&Hq{A%G7K*Z8RCZ<4)z^24lsW)4tc9!%SWrW%(mri zEAcVeH0+csPuZSY^MZKuoQU&82ip7mEG7U>yTwI7~2?T48M=Fea~~2 z=XAdR{+@q&F>{;dzV7RDt?%o7eg2RKdFLM1yOPZ+?_3-&VW-pAJwIKI*n0-PlHJYMB)hg2 z^>Wb(IZ?ARmew|_r71pxBwgR6 z_{NH6=_^C+HpStnFj?ELoGg2A%((?;3pNa_BI*QUtgFrG;`8%Y85zZMxfSoQFtP|6 zRCmm5JZj>rIRq1~R!}I}G|1mjee7m%i)OV=PN$VKpN5wav8=y?{HBI`5Ri0jimAY8 z6AoMWuKdiFDN`x1cM>462_?Raos{2SMAq zqmM=L8lyqYWUTmIQ4!#__20MtuYC=byJG)K%+P5b_>=&o{Jh9h+QOAeDfmM?`34v zog5`(aGtAT#Zmi~>+Fn08lnQ%5;=ko4pRx=e0xIN&h3WPi`5f?102G3hMGyRWO&rd z1SSHrS;M<#I_*T@lkobUx{gQAUJs_Vf5c`xl+?E6D1=ttE74FVVzi?|s^F9pFpF`Q zi=Qq(96LTGH>@?)OWWC7`Z&-+?1Z>&gfM2q*n>PPOzg2=UM7C(b3Or>kzs1k7O-x< z|5N_rqrm=?OYD&O^`x^+DCkRdK4(WZT5gI6s{q+ ze(i==!X>?#`6`cT^70JXN5X}+StX=0x4r{~4|lh1qr_C>mk}1?2sw>J5$$TaUFuXm z9=)<6Ps*#ppm8?^BgF?fW?c4Ei>}H(nP;2WegsORKSu2F;#fi!lD;W*4cX~;d#!&y zpYMwITGGpcjNdI#IUt7+TRu^Mf@8hXUzJq_%gQgV<=Qt}wV(1+OLKr5^y7@fu4Zbn zN7W=9gdxnGtGOyF1@ZLxaQCW*er>bIhSsLMd}dfCW!m3iATMw|nYV>;a2-5U#)BMA zHcq;E<=RKgoCWJ5%{^zE#JB$tU;nBiJDA^nci(0?^K9FW%h2J)@g~Ubw1o*pKpJ#G zzbf%JqMF}38+n@}(U+}rD*INrmI9+Z2KD8~h$g;i@;;v5(a+ol-j&JzGJuk z`Ay4<4GB^i@lvj>_b=M_l~$7j6ZqA-&AgZ9=j`eC5A+rzGAu(@Q7cAi7D-HLHZfu5 zZ#cf$J>**I%Ir+(9{$7GA&r{HV}dX%?4xC*hQ`BT2u~M<=i^hDN^xUu0oiusTa)MfSLTs+opVm^)*Sf)@LOtT5~2 z^p`hK+0d2di)%~c5!%*092Q)d)=lsJew5U_H69GvW-_@(!l$SQp zB*VVG0MEC;$}T;>;(Dn>l{sffA3sNn~0cF1YjyQR)dsUHC-emnx#4_733gmSEUPWa(chhpZ-`B^rz|(k4#NwS$-(PI-x?tHmgOeNBZVelTd-1&c!Io-LSY7EfMkjuBB! za;>PhP@-0>^r#Mw-tgwvcQ#fklyaGi%=E@yN?v-+-r3Y|U&Z$xWwpAoV>HKiP4}^T z#&nsTRd;@7@L|`A9c21qpu#IB@|^BYx-Ar;LNy?b$oDf7=_>5E+QA-QyUH;GWd5 zSa|4Xw)nZdr*@C~SYrf|Pb~-5F6Q=;){*Z%+JAbq6uHV)pQOq6n$}*5FBEXH-e-un zrp~K!6Ss%jS96|9kSeUGA|+tR$pIq(0zg-^_qw7{OQ3Rh50}bH&lARy(jtQ!snpvC z<@82ih!6#DqX$Fl%K49`_5&qkVt>8vrJEb##k+Li5l>a2Jn5xZ`i+B|pQQJtV0Iyy z@;)?(61_W*w_$?aKh*4YV-nQ&lg^H)cOE!^GPtdcbPE>4Qs0*#aRFjMxrzLRqq$sv z^7biF@|;Be#f?B?t%qV;qn~5@E*Rt~N;*xCNxBi&1>{WX}dt-F33iAF5GyN6r-+Bz*Kya-5t$tP*x6Z>&D? z;sB1N2ew;Mqlz*!Xd%#ybd9_aqm0xD=mIIQ&EWf;juXv#$VEqPtx&1d1(RMemw{bV z1E}Dmg#y#qSh4dLDx%>>UZD461nc_U>_}C)*bv6Y*jYBCkUR?qHQe3``y__343%(I z$BBCGC&%%!8T2fYH7XXJ#E6{s*V%37eqMl-qbt}99{tXGVPW(tD-NB<9hI9ah=EJ#x--b4*}h6Wte%lq>Z_~I+aY2?)@44 zKF%cD>_3rt$u}S4b>l7|@f6(L=JRAwN>9>JW&2?42%Vt*V*cvub4DAC?bWjnUs@!5Ol>G0Z4H?6WWf4mS3c^%ttslB?u zl5s1bm8AOWXoK0*C-(Gm=L@OBVTWIc?Jz-kaMwSlIT4T5+s*j?O85Dj_r83c*Z5!z zf9G$kD}OSrTlV$E$sHWmglYAojx*mJu&*nez2e%+Zu;)tm7*ap7=vjmFR~n8dSTdn?vU(fk@i<(a~e=cR-rAe>gXr3Jc?`j-uM;MY=u8BWd_xr=6kk4iTaaVT5xZ$NO zjgnW}3d7V&Z#VykfBq*=_uD^ixw}?o`uX82bF7GNbBtA+VO;e@bExlr!VO={uV3I- z3-toAV;9yGFGG0s({TL5r$E1sKM9XM^+)af&lCC6Pt1Y4o;_?Ya?JeGJAZzhe|5?B zeazBK+k%Y#_}%_w{C@q?zkkE$+ZllN6&tk8`IQ{&-+jzJc>vA-Ija4u@AeO`edgQ# zjF2~F2SfeCg}BfMe^%+kT?u5s4?**Wlg(fpOTe*fAI zW?Ecl(1H{^{2|6(Bg9@_s|Xf1xV>Q3&QOP21M?ZTIiL(LdesZ#w=$eM@aS z2PgFZ3r+`MIJfQEm$>)oSE2v$H2!ppfAcQLYJlM4=EQKXe)LcN$921SfyQpT>hhPU z?%!_l@1Eqta3HvEP9U(89KU#VpP5Cx_vY?=gTg+7LaD#y?#4%dyz=kN?C-7vLGHrC zdTE3PY&=UFM4}E1`}c>%idseMB<(~v_0hh*WVs1e1heNL5S$tyHNOBrmEoUXvfPfS zOCuMr*AQoR8K)h zo>v9}o@tnQ?>T3b?|l{CtHLFE<-p@Muqw(7Y%dKvuf?Y`?@4>ZB?5?8 z3Wy)t-=~qaG*v!5XHb%Si?x;8TnxMb%~NcQJP*VB5`;X?1BFyKKUo5`?ynxd4v_)jr1&Y6#15Aw}B&J znn>Y%4b7NMG2-Kn)KGvqP(fEC=>GAwL?c{QR?h4*)>p=rk5oQXSeJUVfB#}0-+D#) z)zRcRmcu(!tS6U;08d4d7!nBk=SG$8X9HWU(y7sfnxyEkfPlm zcvR>D=;p0R$QEi;K@y;n$nWkz)98ofz|L&2l5FmSSPF!J0aMqY9S> zuO`Z5`}?r2{TdUk5x4j2)h*bu2^oe2#fin(RT!wzt#jT0SiB@@3UuP0CmZCC!4j7RCHGL?X9bxvVf9J5Aac0zSG}{+vv;g zEokB6(#m|OjJw3BySavgE_P0~JI<^+;KUfQQ9|0PZd+JjpQ zxLeiJ?@?h`>C+eE#9hh#f*3vR1%CCIhl&KK-XW;xTH5*oAt`b1E~j|eu6Y|`PfA_; zG;yn)_AyD`?Gxd7EB$=OZ=0|F=_LIm=%)s}^;H~Y)?jpcIi6Q}aV&8hZu9}pzYmo> zTL$Q{WCnmyP~O?sm!+L~is8taCqZ<}VR?2NZOsxTf@IzIFRxL=ETDjNA@+B9kiCy}!vdcgKSMNy9kf zs8*@RVQ0*ZlqLo}R^EaoJJHVc+dFCcRfe0B?uBW)vpMvtO}L_2o~T2W-a&j=6mrEVAx)42rRv6$ca^ojU1GkKK#ZPn_3%_msiR`U^<{ zMrh;}eCstltiXQ7CA*UR^+XPRrxE)GosuF1gpRVx@*rTNu-VXZCly)ltGBM|<|sHo zZSfoQ#RfuBocdH}TDVYxq)Nk)vu8{`J)h9m$gVdi)!NcHs~-J4eS`K>%4(M+Y_tv{-a=+_ zj0)1(OI-#fS@KtkaLv61>LrkNp}Uo(*0;v15+z|vj>~an%y)qW9@r!npxrYI*MOPfYf9@a*x{inLXT2`?-dgLX zMWBUWL6Ti} zxOykY$CEL^x|4P<+w$&|t&uN89>?p!(*zNGN3*iPCPqX;o>Y-LH98HWLn6l!|#-(|{nKD1-fo6ethwcja)%+2+p;y@@^Cld}g zEKnNZXfED4(W;av>8#q5ZDd_;rFYVqv=sMbso%bd4Y+?$ig%cs)@uzHjMD4L8sJE& z#uYypQN%iw>qj)QoVmq36C#PVFeo|HWAAD?PKq*iQ>dm*oM?y=M-R?=w1i8Q=waNK z(lgJ^^|=KL>YEtH#ZHN?T6thAT&Et2J)TD-YP%0X;aO(4li~!lvnLlMEVUDUVrMd8 z-FHk@&hhh%{x_;K!{Cj-1SS4`9PZP$d+zqUq2P1xE1sn>Eho?zA-w`~MuhnYXR>zj z0oDtaE3ucYA-PqA<~Ra_d;dxD1K)M0csXz5tys9zy!q1ULh34*%v*nb2u>+yJkk6B zrIy3mLoWjMP36^C1KoC|AH6gHSTfe7?cyz(!LYDQ)8V|_jx4tvbtTgId#UsATPQ~~ z>p&4ozo?N=I@Ti!I$|9JF~(tw=8zg2DX;I(v6p{392>jhM_^n^lLJXd>HqFXtR8zo zsF3>ZL7Bx1CzQKHSmSb30w-G1RNmh>?D_$~gd+OZ|UY{EmcCCJ36EtC6PN4}Q8sv5DMWcT_{QzZXTe!I6_i#KClYi6+O_cwzZfmLqF$5;`7JA1331l8K^LyMdtNX2GMEg9t2|P009a-K{u2^$&HbAW6Yr4)@+P@<)&(468kR zMQ_y3NMzrED+tb}B&CEvNJG!7 zUJeaPx9s|=X$;_d?M*eYMI{=iJ~3$slPE~N9o2gO450LrBkjSS9@f97tKee0H+y<4 zxroJ8AP^oLN5w(;E}`iS*-t!bspK@>_7uJ+iS1co+;U!?Fi*{S=^Ycc*pt8jPOO3; zlMQu{M_!pBpDX5LA5T^tr{5XnmICD(M_nMO#sM`TfhdFJiwJ@#;b;u9_WI*YWy zyyg`sqYb4HZ=<*U7*;F0IH~bha70*ysc{zZRv+1ksow)}V zXeL!yS?c@Z`#5x2YY~2}&i%2@#N~090)Y>BYe9qb*AmbD^Vv3+g7LL2MOh_CB)ZtW z=~$-`Ij_#sztwp(w#b?^$FY5>8h$DCrojD6NYGJ#TN}V6C0-U8B?O(jTugCbeDGgY zxKmK_&-Y-99NmIl@ev>%yR~A|J67vOS9?JNLF#up0q~QCK1)ens6*QkG;O+qqaC@|1du1a zm4Jj7wpmD}xyx<$B#j=ePl8J7iS>!u44eaysw0f+#LLUE8?j>Qk}gvZ70Dd=Jge&0zB-Q4=`Joi+55B2gm)H1eqQOV4PG)y8bVl5i1#qxG47}LX)we;l^59@`NUJd1`zN)N(3wMx~1F zg+Y{CCc}@W7ZrfAP>j_ahZw0XGn`!~|2hLZwb!DA!Z1$SqbY5&gCzHc{z0U!MMF@Qqn51N;7QtPM!utZ1&>0r^$Prnl^l}zJXl7&1RUZx-h7@Nb*J3bZ@KM|+sc51 z1Y-Jas(O@07@x|KzWE3JMW5DueD|t9mGRLolPO^J!ab4xsmIRi`4dt_YAm<<=3n!( z{jHz6UIaQ=AF6l8|3X5+1A2LgcWr-ceE+XUgxPF|!d(}0jo1I{RexI`Vm`gq$9mEK z`OJUUKl}hd9^B3=@qdBq{i}%M53jvAyw$z@Z2K_mht&L!cOl=k)%TTzxBhoG|9Kzi z0ABtsc3I`$l`wy?;Jnj7bTJ;K`%l}kg#Yi5p1*r~DH=d@UA-{#zZT8k{n@Wta|3`p zf)AIxRQYc#+_oKe0gaXP^o+!n|4wWEzdr6hZWQZC98*H%?GsG(MOFOc6TP=jm26Z% zGbW^eR;BCJ@^?62#+}MLQy+Tc2wRK5!j+6y>oCVX|L$4Og?~gaBqk-~DF~qlU6X`f zcK)n@%)58=?9>+5guZQ})=|Yt$zi=`l~E7xUOKD+PQnQ5HK)}V01l_olRe10Rhnab1(xn;y|uLL6>b?$CX)5iTs975?aa_ zwB--k5a!D}LiHC61rw|%4NoL5y(lx`{3)B^r5^aUInleh=f)=c&4GhD*@hQ+ysmyG zY`-0Oj?Y_TqjyU>re55_Nlf$}f;r=^o4Db%VNL&rwF7L@OJ0Q8;!tjB+|!NJ25*m6 zy==>or_Tt!a$J0b)&vSCf8lOMXW4rrChi+Ckb7vmmNx&JakJufLvLLrS3>Zd=B*81 zf_b2geoG1rgpCWQnLm7dj`xase8rqLKyRM9(osc}k}X*yuFsF2Oc{Yew?>9|qKDT$ zw2vA_mAaC=V#K{e@1s|ukx8sSq~H4tc{`pFwhmB}HKj}-p{^-JF-*IWpTRLb;@ZOZ zJ+js-KbutEScY}i(R9Opm}0B&&`I$(iJ}F8V?3-c&37bgTFjO+=EHE9FmcVCw!H?Y zyGSJ}uD)F+eJ>R#M#G?QlE0c{HlbO-$C&DkF%G-7phQ7%kI0PhK35V?cGWAKHRWKDA$HnbJ1}fc~y6aTO4!jjK(ASq6{N$X%!mrVoSVem@v<~)r ze^kH~1LbVeH1&!ArCTTR&5{T@&0g-zV=NNv2k(IwBvdsD?r6X9kp{odGt6K;KS0wt z%{@lCOHcE&$E<^$>$0XY(qZ?%e+ewy)DiEi6p!BbBW{plz}vBbTaB|2F3UNq&{~+| z4Efm`yWQEnDo&b9P*0LgFvPeRX(Y(nN-}ZDp1K{G6Qau6gy$MFo!V-)5xUrOE-D8e zjdUi8ms53Ldinqhq5@RZGjS5&^cP3!l*R&h>Pm~fX@m8aJkoNom3Zq`R-W?&eT&^b zlc(-{c;K})N@F1jrfS1LQBAeujEXd(`WOUdY{AW(S2Es^p!J4J?j!*V&8^_B$AOGR zHU#%Y`A0gv%v`%pU33#F*?<}GbhjFDX8+`Dq(AOhk8`C=3{^TPWoaYs^XW&l&Zc*J zC>=<|MvH2KDV7s&x`xd_M`Bj#IgMPZolI*>u610kzpD}<;`gXEUPZ*B@r*uQJj7+O z>+u=|U**`dUzkncEnX7xRsIl*4G*Iu&o#GOhjg_++YgqTO~hQe>7w%XoJnq8d$JOt zyF^Dvufp#A7P|XNnm1)eln6+B=a<*?OQVR|G(cVNb_fYmc|2%ftIzJ;IdL(a1D^8y zWvb_z$dRM`w^|dVNbKIk33Y6~kYNFL$Ti2Y=nPVFz!4Py;Fci{9iTq>tU#1W^In$B z*Ck~FvwwSWtf4Z1YA9#7`)LB}y0aYWTeiB=mpC!|2-7+ZUEEtk0UeLe%S5+9Y_Ev9 z8cDCpX)51YSS*@y2dwx>y`Fc<3$1{!jVrNt zVf?qvIYrDGc^5|OD9ugyXC(?*_s@e)R|Sb|TK6WfI!ef}^b9aKc@r^ug;ZC!xNZr*505X_jY?AD8`CT&vT>4i2! z-?C}6CZPLS(hwD|4YL%?$jKFauB>ptAwB;d(IhGFAwz56X=Rcmpi)Fy`$n@-nIokj z%d|?)z9Ec1;>=5yxY|kb6rtaaDw7~II2aJ^Iy-b46bQvI>B*S;?~-I}hiH$XU_Ri} za$>u;{UOe4kA_aa%%;4PeF78SyfIk|0o>LlIZLrI(1{B%iikV%gxfz-p5!e*aNT%b zx*RI8^vmTwx7bRp{N}QB;DC=C_i=15iM;&W2BH4&lf9N9SVi$Mb&o@B=hcJRHBEQl zdoPsP>kHOz+(()(4qQQf;yi9t{4nAmeNk%5RWZo)Ew!7<^ix$HZ(?v~&z)O0SqX|O zj?NkZ1@op|th1xjNdHh`cjkEjFM54c81k*`y#57i=G8Pd{fcJ<$voSB*H@Tf_v?oC zSXrLNEuVXE8i%|HWjb*oa8r*jp@I?$CMe;>>IotmJ@(%&40==(rk0G4Uv_*w$x|Vk z4ZT^t{xGC!+P%B66OfQcG+;EHcjKL(>fg_~uaF2Yj_AtKlllD7deHx(k%B8-P1=4o zqkDAkp|ww15&AplrNq!z33%_}r_%aGi55cTxz3{^ESB^|B5AzI-s{qKGx*az$O>mW z`>ML6!iwfvHXg>+`|0ZoB?GzK2OWyP1|?o%I+CiA(Adjmv$i~SWwGgbgGlTC%cKFa zeeSIO%L?U!8;k2OO_d?(vS+p>Qk9JU-sH?Kl-3sdL5PQO{r*I=(Iho~$^rU#sj?8h zr2UdjJO^SB9PLe>}Nt-#{Ekj$L>(%S4T#who zVIdhgRWkj{^isQSMV9@Wc&WihQ~CW#d%4=)8|-zbN3bF0G7?q@wFl=Ha5EAIn#ohH z<+3xQHq`m+qyFSElz88ZY8^#u7~$eulGmzAu}!~;%vDF}Z}T@V3ML?Bx5@vIKBw=0 zF2~NQg?{V)OP>7^!PYi!S|ku$u9(6mY!rscz%EL$F^ab)VZId^)e>=4iupO)jRk3_ zS0jwE(q{o_2AN$FH^Q%x7$)q`7&R?=q9I#isp<{0r}j?2i~3dPIE2ov7K1~w7DBxX-U~X-Ss(IQLCCD%%+$j8WpbcSJ|d-)0(S>S)^Ao(<(6&=Mg?SC zG_1VtfuV%*J8?%M$lC*32dVXd`}6z$1Y$MtK_PRbYLffBX8UtlRd z4@lS;ZrAz@Xe{;_WE)dvrWv^}$gJK1N<9evw~Ecd(^~_&!7|2BfO28$7*QjyQ-hL3 z6s(}hut7Z@)yaPQr<2!;-h6Orp>m_v+_Z5ckr#|io_)*FsCz|~2wG;_&YgKj$Plkd zHO_$`L)V5D4kZvXuYQp69+6vA?O*93hnc<7CL8#Af+F>) zGU@dvkK0ZC(jGrue(8@=>mJK}E$2P*tg>t3BFo&0xUnE3#eLH>Y2-RVmSbhz^IJ)4 z4s#SannQOvr;upPFWNpoTIajImlrYXDo3@GoQpzKUsG_HQN~(7GDh!_z)!8`#7oZR zG|ot@$p~5U8rJrhl<9b6O-gm5cMSN%u2W38Pt-X9G zGqXbW<9y!3=BNvmF^jdmxhdeYT4I+@>j|tYW+-ex1wFzlo zd72t?F!OOsTSnrqTmWQh$g_`Di?3Pdo%f)XB!>WJ!&pNGbmhcY59uv6NMBET-SxTC zajY2QJ8|o^%1j~BRAsP%|s*V$$;g7s5WnUe6a zC9Xv6jt-K%QShkD!YeT9!jB0AsZOW7}tevmTw`g&?S2)Bvlykd99Pz!~ajH)#< z8QCHN5YScFu+kQkat-P*gR;K6V$nDmtMv$q-cvOzFtnANf9gS~LUHrBFZ#?29;imrp|X6+yOT;S zBhy_T10I`^k+-9S1O5F)*KJBVGqCkd=)$F48GzdGyqRjmrKIzWeneVN?=ssq5JdBv zsYX3LtQE7S*Ps}x z`E!OM{@bx#oZ^~uhK^&CcbD{Jp3H~9KD_3ZGETImrydoE#&>%!PhOttLX}Lq1{Y{& z=?=wOC^y83moP+cO53G0O#;IH^pus)P%o!BbJ?a@sg+AUo&lIt+c#-A8Lo}Le$-fs_(tbA3r%0D}(`5DrIDWyO*p~-=&V8QD-1aDN!feD#NK6R^hIK49(YPQzco!B|KxQ$p7 zJ!jMxCrhxf`&^%w@6dS+`%StGrV3hFE7?&3cbf}ls-Ee}x?I3LB-Tc4&NeEkDT?#bxoUSuQ#DFJNTCNK4$!v|m*s+8>HGa4{a__mwmZ6C zib!vprnaUh#r0?w<#HeUz^mZ!>``mw?w&lp`f~CKnc5p0%LSE--q_nc7h+@(7ykJm z?c-%;+3H0+ulUUQb5;38%?^&~94(r|6mS3pfkOL^F*&0erNL5Dg54Gmg+;tjSJ955 zK2Pm#u@ju$kLR^^`nEzUbq}x6OD-P~9va*%$k1-$OF5}%vAaJIAaP06bhFQ{CBBs_ zm#evDPwDQSScI(3-?TQ&f21;N(y`%s z>3&J0yeXc~$`f#lG@Hz$c1&u#v zbt&nuPx`92oy>QyQrD!S32?E>sZZZuTnIoI){!N?wP$W8paU2qsbnj-2i~UtMqhPz z`>wWG4Jl5knI)f&4+=&V216e2gO(?|HJuV$i_M5)I`f7+LZzPLuPzrW4U#GcA$=wA z!2+;&z`{srh@+>tf8_K{WuD=-y@HBXn{&N_XJ4^JkgC_z;+fK%h>q?19bDf6dIJv^ z98Co4;Z`fjj;_lg?(wNJjpWRj4DcpB&Vy8=!UrSY;TzdQmCiIa-sPHo7W1AI4Ezz` zq)EYZwWfF1C19A7lL7(F&el-~?nG}s&?$1GYs4-S&qdt^tRJbN-snj-o!JlSZMgJwGO)jW(9=Tkn~>13zJnn>3Ba9d5RBUF&u3e@!xX>;n$HF zQgPL|uNzkl7xkFVep8n_y;Vz0hxmD0oLvHGiNkIzn~+mLYo~tlI`r%M18$eRU@~9| zx>w@D=w7M8vh+at*)E2yd~w2r_1O;s3@;6k0K)(5C+ebH$c0Th1i6-Q&8xD1OZZ7cMgC zae*OGGUIS9<_W-X$ltjWQi(6AI4a2D_))h`$f)>UJZt=@K?Ew`;0yr-de>|l!ubn6 z)mPc02i6RbWUE_pS#kd43bH4b9A{1WW-jXC7+#DmTSDX#y!vyh|4WaEZdZOs}g=DbUU3EorPb~MGV^dy+}~U2YYMAP0fTkn%&-eAc44T z$-PLinc7xHL<{#36ZAL= z`o4|wwd#4Rvk3KLcu6h4uPl)Ktcsjr@!Lt6qPS|gV6YW=Kbc3yYhcL6n2ZnD#V}W$ zF%zVJ7cnqNdPMDF$VYTKt`RBFPs@VaRyKa9?ageo)p(4Os*(PsiY33>Zjg#KjDaAX z#Zd1my?mv{GLz*kkY?Ib>Tk~_Pvr-X4$K+$ylQOH$#HRMxT0CY%-Bq|;%J?<_2WyJ z^=A#+B$6s>x>5ZuyAE(!GvBQp-snSaQYta(iC&}!=4*~-o_ly^xl}<>yUcMSSEq$I zQ+B|R7+RDMXtUbqvzM?skT1g;M)2E39E@f&r3^J1^mNv3uEjKmDvoD}C9+s+=Z*S& z%7EydLBgPrS1_t(0mKt9Tb$`NBwvNtkX}37FwIdJ8mBlT2(5mNoz+*7(CesRU30z@ z({UE@1~Hj@Jxsk9YS2&oAFyaQY~VbBF8zlUqvJCa+8*`=VsHw0x6e#6 zXS6mG4}Z4@e{2>T%`yQsV^rY8#qV~_)P)0pqkFb0kuii{f$ha~( zWvIoR`rK(*MDdQiKzZHHOv&uDI0GD8qAY1Q=}r;F#ha)nVzF4#kj0!RpM9`2BTXZr z0|YH0S05iCLG|Wwa2?eQ6>@7R#s&SYGT)s^#0jT zet|0~?%jKpEBi;gv)o+EdWs|~FvTq%*S^wxe0J_V^b&KYn!Aq1#!~ofwG2j3^3?r; zUqg_^?R)N|w?xnOIst|cy?h(L`#DZDxCOOYBGI<|op$*N>AQ8^Y#pyCG#MrTWf9Y! zfe;sl;@SC;_cheHqSI~IfThZ<(U`%_IR_UNMl>(R`h5`?RpD^o=!F1KnKPZ$2ZY{3 zAWE*n)9RmNkvhl=pe!14uSe;ugkves%#+!VKlFGn-vUkOgXMZ^KM(u+L|=H9+&>Mm z>}P)%FAA@Anrmw{>=|$mG>nQ8GSYX?c08*Cy(I`{0+@;=phE*S)OapnNH0gcW#TD2 z@mAVndCoy2{@N6$U8K@}>h#mX$M+0;@XsB|Srv=XZSZjBl$yv{PY{1gc{`UVt~pOhfM5@ucjKO1@0K|g z>-teL)pXi-AI2b}7-#$K45-aAE>v-MI8>JSnKKI;*KeW&DHf-kv+k#K5A3w{z*ZVm ztQb{QC>qIGA;kTb1sRgLKVE)!gDj8&_^p>8Yofh^`?heY$2_Rc$rGd6wa#`V?DS9- zZM?gfRQiU!?-d2H&(>47n4e+3W*;GsEbCU%d{Vk}M5YH<4!MmJSYz>`4HoHnpZ^Z3 z_tA3Y3Sje4H&1xEv(+tU!I>OP>0fiw{sHj8%eQ}kMvAEWj5KfkC1qqjyF;I`zgNt@ zm#>cWyn^ilwXRGh_Ns^C_pgk!&CKsck9CsL-`>xzX>(x)nx0^vdE#Kyg6;2El!ksu_dpr>_o7X;G z(U2HSpC(>$Lz526ti0P!x*f1$|NJ-~j`DH!rPf%Qn7s=92I^5$PX#911RJdLBdx`U z&*baPcA9=-y>#=J%U@jHA=xEQ;9A-Vx@w3A9-1TBG%M146|%kJ-MHPY1H5R!r%Us z0yuTtkA8<){e1DmxTla>I~#2x-JuJccpJKo6={F+{z|G&uDtB|d!_Fz)O3%D=$^DB zr7JXpQ3W?wc#r7%EZ3RVkdZkDPwm?7g_A9_&$VmlR5CFUwyNLuvG_{7VX~**6doOg zPQE*o;j;?FKhP=tu!HOY{m?qN#@$Z2Z)qFAx@*M_uqaaIUfm@sVW>j7lZJ9l7h}aE zL5D6tcXrLo=);hpMTOf`7^ke`ebBXBIf2L*yzIqn?r0V&dbq_>D3$_5+zGGa8_I5n4%^mw;58E_TN7&JkUqxp z>95(xzzW^V-plx>>TxBp+o(L=8d2kdgAtQhXelqr`_ipyZn0}^oq0c0G`f3=`RG}T z@zh_9h4t=2eRRP^Xr5~1Mdi?B-=XL&SLc62$%^Beb8%PZ zJwJVSyo9Wzkg#6)CsEWM&92DgB*qX>CfRc!0qlcR5;CluGYXufG=3Y73w7-8NrBp zdQjqgc7XFq>&WJN<@XtI^LO9ZHGsCXA!4NwWl)j|-Beh~a_@%8-7SgC&_wRHXqrAG zpb@*65f4TPuY84@tWw@rRy;%DN)#(vi$T1nSK{@)ps6T(WYr#X;yeIh3(;1%p4bL5 z2wq17{&A||H*>-%mMk(jK+`Z&@>%p>o z^VvN(;p5jzlqVdw>O~6AcI9X-Z?0-#V7(=q@4#%XPC15F-DUml-Xhz13t5Ym?SK?e z;LEFH#UKo)$h~$!!O_{E7{4qfuk?i)xIeUeSxUripkfm}kn*MJ5dd2xc8<4%-(Ic| zt3+v_rgH?Qz(k58UC|FcN?*e{0oGf*TK(qNmbhyFUBsnBYkN;Ht8g(+vW{wM36>8PFd-LyzqaLT%q^lY{VePVKhF3Wial>}G5`hkReyl;?J zG4__#3TN8(x7Ou%Z@f_1;o7>_ zK7m%5PAZ~~TecPf^_EhEU85Q91NAxj{<5`vSi*b&Cu2cuJ=c<&8 zyG}grJIAnYCZG>(7x?DVId4)YR5iBxGTbGKNDjwh!YH!G&!I8B3IdSK)5^ECw~S}S z*Xx7`aE>AbwD+wf9$#8=13k#QmmCqAPBVQ{y539M1XlM--8$+DX3IZ`?(`-tf!Pn$ zU}L(WDG8?vDoSaPAgi3tDe1k8?95oTE0Bt+MJZs>qa1 zSzE9Qc7KCB0N_IA+RyQ0(*-)mE*cBi7ZT4Tl&^&8lVeqCQpUGX3E$#m(3x6zdUa2alw-s5wB?)!7!_c!m~e}_8GdA+Xd^}L?rF@PBF90qR)BZj;3E=A_7 z1MJGD{ciZzH>{yN`lW4UnBY$-@}r0QGZJhVbIu_bWNw?KIV2w`gl@8+Y!ga%YWb;=>Tp*VLqL`CW3F!#i(_>Z084iJ(EHsRFDW}BZyK)bEsj#f zsZO&ttt6T(zKbuw>znZu7F%SzcGa>>(EuXng~e}$I9`3xD!pb7QF_AtsNc-u*#%0| z3jsuFBo-gH6z^%_5_nMxwdXSG)Io^DN?_r9J$vih37F=0c<l^KcrG-+GI{*EaGNr&mz}Zxa83Om=H4(G?B{0=cc<-|J7Ir@77N z*YNl8JysSSUzHmIU5XLb;YuB6NO`o^Mg~hVimhD2EW_8bV-2TG@hq9S4R@%19_D6Z|I=AzN=gxJ44-WaJQYd}8#;G^Blaba|O=-Nr2Ag)o-M|VvL5tjRiPf6m zMP+2z_Ao_Ux=-*L@^&%xCebcUDPZV01&gQS3tdb1r5(A?h_5lbu>!!o8eZw==Z8aH zFB|a)z7+K4ISI@2_~y1!)~Iu1iO}Qt-YBQJ9(z<(c#@0pUOScb9$f4)zL0!p-2-%aENAxS_Pc(t82pmSHdjZ=q1); z<*@o@2`%+bjdw~F1DtKHPxwJ4X;+8^H>#W`h|Fy$lI$lPCM7i{%RX9~_~a!FEZ(?i z*XEZJ8vn0F%fA^uM9aOv;NXX)GijKFV{Z8NOp_m_Oua7~3#Cf*u{&ZJ^aByU$5-xj zI5KiEZO;ffru8n~wt+#K_>Nh?>~-d`;V~`hOZcA1N2kpK1ha~?7$W0suLyw6ZbN1sCZ4ntZaP z1U(vZ^lmWe6k2I1jO+a!Wb)@Eb;#4>Yh`9CWDl>m$iz248lV44ZpiB|lb+@7z1(?O zQ+1{M(={W5!kD+|(Ip_sZlEkg+YgLyGJwd$B^=wiQ8QYp*b!Gd!Gh#6eBlRp1s~_` zJX5f={9@p=-nT>z67Va88B`gGF&#}gb@y12V~KjjV=4}ale^ReMos%M$F z?Q4k&hf3^-*hZ#z@LQduyI|sN7Lj=^mv)8S=3CJDFwd$nt1LUWeSsGC#SQ^dG|^ko zYx}DyZJXea*yaILZE<3_#riX>tmQyb9yw>I3~IaUDL8kWKF*21txD>#?FjQ|X2H&djcR1JfFQDK!mF4!-abMRqVB?% zX0ldlt5;6K7ItM#)9~X9tyTy975QQFr`|}oH(Q9o&)nT(#yyAVW@;5wt0S?#Qo(DX zz{{^WJAIW1m~N6)JG)FcLAE;aZJB1Snfez`q&vb_W_BYGk8w^EWQlcggb#jw!tbk| zi<7!EPaFrtOYoZ>cy>`1Ab7t@d-mSCJI|K8qNa_fMe6I+ox{ifS5EsM%j;&gBTrN; z=l77(l65DkJxwFf?>G|d7D^IB<1Y{v7SG384>3CQhm>kkNdO2QC!3Di(!!Ww-uelN z46_b@(}sq|={7`dC~|k#C%lGba}4_Oit4%3CkJXN0T*{WQ)JPDk1-V$FJ88I@?;pG zp;w}_bV{7mxN6{E$L9TSQ*i=~eED!IlNCKG2h z*w~SuZqVPDZVOi>_&wUbHL`&dp9Z*9jT#SnLXzrGtrL~>6`BeBK!Nhc(#f#dl&+bH zymPL(96vV4pPszEP-8JelC);g=)%A7Mm#lDAIf(bz^G&=Sv8 zMi!RgPG306bKW#4C@9Q!hLpD_#4K{gina8*xfGs1O*wrI(`7odhn+*Z=$%{$++V&J~C@}`j9A5M>An*)>efNo!IPQl{5 zdZ%9^!=Lw)3CMHb9sK&nu1==_oZcMl?Dwx+y{g)5o?Q*R@QTalI4^GP5Hfdfmf{CY zu%K8?usktn^(YXYeOU~S-Oft}eka7QYAL=Zxt3$4E>Jql=4rJWAa|naV*fJ7phRh3 z-81xt;o26tYY;AZ1gM}>?&Ca! zj3B&O1`?^%6KD=}0W2ZD(>K#=7x(EpKlJ-40{7JLz=EZN*Nht`Z9eAC zvyM-B>4n7LB8%ZpVz{os$1Fk*|6{*Z)286EjELKP1Su9lB{8@8DKHf6XoI37ChV&( zyN$TUGFN$`Sh}du(FMel@fO9T=2K9W+$U0WwKZAKO)nBA0_~Rf^a*a?A>7tkb5ORh zfVyvNAsBxds7~fnd$~p|BfGj40U+}(iJkwyBGUgo#O=~GchaNuOKh%h#=da>a7lDm zzhJO6Q5q@HgGAsw`>4$TW0l5MW_(wUscZxv`S$%n{uud9}M9o;Pdm zmE*maePkQcD#tbHT!+kUrAG?>^0MEOf*6o^eHaquZWz)QC zJo@=3s4^=8kb+Bh?)EW@yA4^b?+Sj6*U~R=O#9l@q>sqb&fY%NSBpd=BKLL9Zfnlf z9rLo#)Z3RZ zW1ETwI^F+$rT+K7jz2x$+*wbc2HDU&0{OkVD_r+2C8}(fgowNC`%E$-H2v~0lS=1E zGz6uJVz_wmx=+p%Q@*a1nOHiSH(Um?NUnsCDwSJCjcq3xc#AK4p>ErsX64;iTPw}g zZ+l^pu(Ie_Cb&4VXM^eC%)ulZikGPi3+!aqliwuz@SY`Ydahdmr-U(-6uHDIrt>_;M7f^4E8-KYV}p5qkfDEF=Tc<|MGwX<+{A#%c)4a=ua zrQH^aVM&y7TPZ>2rFWCyAxv{h<=(#NV`Zj)-rZe5@F%MH%ywRena@7#A}I{qh+f_) zQkUcahOp`b;fuAD6UgluCh9lZhTNLcU5La>0+*+pY=N)w8Je3|QGFMEZrg?0+Ey<2 zW|{2go44xmazW>Z0bXgm@ALhYpRN}f-)@|`B@R{oAsFW*fr0janeY~#>4?P+P8Z?U zp!6hQ*)!FLv?4qM$=*r3SqX5cfS78Yy}saf_bA5A4Zo!0vneL}LVVlUw?{2rV$Oz> zlxIR@p_Ld9=s5Y?H8y}N5C3UFF0&iGaTMbH!HBTeejGANa8IA$UJmP zm>P>r|zN5jZoP*AYpowzux zn`bpc?nh*%el=t>j{8&x5szqqf&;0+{2PS#5uKFtQd+7=vco`gO~OcTFl{P^gnIZGe52O%^)^_KrW%@)@=pwHk(LFjN~x4 zY@kvqv0XXx1_wYIMdbD_0r*Z_rIgvd-64k9L%XZ=+>5SBdjqwmfnJiZ!uZm6nF+w# z@7r&N_f&oKDtkZlmGtLoHZ6E&({f2b8`icr^O9l4^_eq?y;BgC81}>uOcsI8b0be! zV6(}o0X=~-b;shCE7WGSqyX6dmJ#Q@f0-Wt*B{IMkMWy-Bg~pMnGGx;hHsSygfbbz z%ma3t1^jbzL*%s=Kk=3u`yN^9RG^dK)5(dd^2uW=QB8A)bAWyF#ki_$hY(D2LA-t~ zwL=sQ#6MRRGC)4gT~JZ%;RaMXrhuMm=@E}cN}6*_G4qo^=P)~J&!zzl#A8ANS={Sr zro_WUeb?^2qJE>3=BDTFUY~(aZII-D$!F%3f%&Y!Q4-RA5`B;!HBJg0##DEnho43&Yr!zuW%w2 zED^e6Xf%En^o7f~MPRFDS-R)w>8BIIBt-^B)$s2@xiqBo6R3@qM2bEC>HqcX*Vj7t zkRdoQTL3^OqbJ4g1vK#sdIHtVAhm24bZ05>LGBbhdoOS+ye2+9c`*%Gzq{#|*<1&T zZ!VQ)>v&|<@cS@#wtJ{d^Lu;JF7Fu9Jv9UoOQTNf28~3M7F&%DL9bUz+3rif8qrEl^+h^hEuD)ci$_OvJ&FOFt~pyaK!NButszOK z3zp=h1u~D9_>*^^Nbg!gIDYuCFe=zv-rM@2QbLeO=Z9<9es{yPKz#OW1I>WXv$i%m zEGAOwSoe*j3BvdF`LuQCj$h&R{NtG$4`pc9swV(8ee)-)9(LV4gVh-*LUFD;?asGA z-Bg9MRUOOA4qdGiqE0g^E0uyL@^}Wc#L@Y1wpblnp9Nf*?U%>0V+lO+A1-O6IhwbC zWOt|1CfWiG8|<%Jt&4<|(l^l=<-4!^)WG=5mt29Z4KBMj+{RWS=vzu+i06S?@J|lq z)WpnQoK{{UsKDZD`^_U2D+Cs@m(trHXW19u#%!m3kEL zs?5)Mm@aubE<489 zd|ma`Y>Q^b(2c=1ot%(`yo&G8uYlHVPKe;ycpR3U7a9_BrYhIOA$Kh|*EhP6YGvr? zb#mjek95#RBOk+=_IPNOpv-9Nv0RX)Ybr&z_R9q^l%QjYosw1_zxLFpzZX+@bfjBl5-|%c`ouu8E=QwKQUhj?h9)-Na zcj`jOh=4b#q+z#*OlqSGmga?`bBp+LWssn=ha0h`Gr33Li-r=R+iotMb-snui9Tka z|JM@mo@pBQ@#=JRM#8Faoaui#E&l_T=;C8G&0{Dav~m?6rPQMLYMWkD@QLien-ZX| zrfg65Y25}NlQqWJDH$o`y5p`-9E_ff0Wyrb!7Lqiq4@WRYK_&ImaAQybrOBYW)U^p z)(DwGo%=rXgBG^0%S`+zS2tqqhdMV?5lDs_R$J9rft8AYfY_36m!y|^u$M&6T%5D< zs7VxaY1LWn_hWQr-EQUf=6=)^Lyf~$S#db!orrjhN!m}TXN00d+ zAbfi9N?Ij0|3LLRjy#F)JkmM>pV)^ z+&6<)1n%|b%W=#C?&wz65U}T@k>3qe{R@@o?mR;)v$1Bl!}I?q{($@Xd2w+0Wke6np@w+ zm)ib3RyG`aCO@wWLTbzbU}r6Lx29vpva+p1z+N)Wa@If=1rz~FCeI<}?5a?6v>90y zI$6>{Nz-!KV7MgR8oo4A0iUT|eak)4x`qVOdp&sAcoI=9P1kd){y46eF?4S^C0GM~ zQPLjt<-AcnXGoy?T>*dd{-mfr*)Y2w@Vo%hY*F;;7_5o1Ym{o>pKs3pt{eXmMF06! z^YW34V(!4%&>GgF^|Sws&3@g!pfaG_2v|oenWyBSGKNRzI@*Y-=fs!RGSGBY7QKb%*)m8JVksUC+-lPbrW zdeeJl_kpE_3Pxu%16-)i)=gq}Apdy^Xic?FDi{j^<%RPMoV!wGptc=8$sF`%Wf*}8 zd;{^0o_ZQu^&phX)7e_vDc3HKejsvpH3+a@IyOq7J>N}xePZ>V>}7p_k2_pKOG~?l z=rLUg7KEmnM)G#IR6ipx`^l>t3oZ0#PUe}uBA)UgkvD%+-)Tc2m0~w=(=3{!UMqu1 z@P;r+?g`y&4Gxnsl(#qCJytRId^1$Ss5b&~LFD0SX0fR|ACCTwn4rlfdmNR4ikb*O zVBmZ3iZLy++G68=siA<;p9M7fLV-XWQ>xFf z7JSL%<;)%Ao>4z}J#xpiVG=}G`rLE5_0}}z&;`eZ-0pO3ZVQ+ZKKI9SE{WaGnNd=v zGb&Vf?qZvp6}psGHlyo2>R(cfQja8XEr?n7e=(tjZ>=u>EjdjrW5gSNxbQYs>}WQk5$tyIY)AYv8Ej&ZE9#y(GyMf9A6w(K~fXpqY{V zIkn;6Sg`wS`jk(aI0{bDta-21A75Q7LJb)cIq5!oPv(}LtJ!?|?QKg+jVIR-u#ljr zAZ!dzJ86>j{p#&wLh>zoI@h>4L&g@810`syh9&Vx2xay1OUpw$Fil@Q*;e2l@bUHQ z6XFkKmupngHo8o|@9f3hd@|yMGEd3PWRRMRwZDIK>8i>h2&BHgp5LPXrv6#|Y!?7^ zR`E8f?4=|K#zB%~4!?Q-o<8?y9xhLODSu*n)N|uCQk%o=eQypVI~J92M}AFjXsFnZ z=SgUOb0lBlDPh`M5@u53C0?K}B5s7QC)}l6_iptkack#9#Z0HV+oUOlPXTlE5K+#p zJYP6GYMhgBOU_X>J4_n10#A98T}T%tp)66ICG*?sRX>Z*eG6(UQkTXprq9;o7U&~Q z{NM`rvTJ=rJc`M-l-I?g?;iZNpI6y`P6BCI;;f@U0n^g`EhtP9FROW1QV^HPmG~PR z2xBV^&@}}|f;=fAvnM0SBF#0|$Y^X$SLp5(1;~6mS1stcTH&KrTpojgDt&9M>GNwB zG~WaJc6e!4NkG#AI-6^A#3j;FM6#eCbPJjMS;2UChlEF#elDjIL+F;}(Xc`lL^@TE zwqI4{AWU3+Fkp=r+XH?E_x-jQJ5Uikv8M#Jf5^vrZue`|NIwJWm>THx@JZI$K}|c{ zKo=zncURhn-Hk>%gWW?RI~E#FkW4>PMVVjH?d6yom1(q>2Rkz!_I%Q8C6wdb`SV!N z49)`L$gcD>gBcb*fhQSakhJsJ^ur|{RRKF;&|JOJPcLWi-4-C--7m`B8iBo{zrO;g z|E55FVf$)#YDQ@Qv)LtY4zg>Z7rWuGL&*V|Ti%W9Xs%1&rGABZ@J|aRLwTvFDr=VUoO<4z z!M>m_IE+G>=6ewIJ;Js_X2H5 zZqH7o@`>cp{7%^er{sLjRsm&yZY@*U>x#&X@UGK3$!C0Tm_t_9c76GUiZ=`An|JSu z>9>QU@sUy@$;lpIvck?czH>=s0fl6qvG6N(oWY#>rE$wYuKT>EQ4dA3hFyn|Vaq zp;w-@rEt0p_mx7WW!pYO<)h8a(bPZGz(OuktVusO2NhLWj{E&6`Rk7SzqX8n3B|qD z8PM6wu$^}L;YH7_b?w{9&uf<81I2;=f^4B<-wkoAe3%B?jT=h9j^g4#(5+#PtGCul z(5be8r@Qf)8NAV+GY`(nu$fYQO^~?WWNJ!hsAZk-R@wQS*nvVy$CD)E-5p@@V0H%* zLsqi7>YkvHc_=s*Y3v&(q<-=fFtLddhCZ{{d%m%F2=Z!cYN+G^ih3Ge`J0 zK@7U6!Z!2u4ZChCGW6BKZo~w|OVE$3u5cWD#U4psHQ=dtQh!eG{m(38iKD zfy;pYR;t_62@7uxx2@O)5XhF7azRrOr$=MU9|xLo47$}ewsk}C{k(WP744Sp#>$};sj^fBGq}bpY*tX1ECAFyUdG- zHnITK1I*!#vDjk9u(FNo=5vPT6a`9kPeH8%u6dx@c z_$};D8#kgHoE!S5va+X$7Kz~@AtC;Tt`LN@*ObDWH*Z$v(MhYh%;HM;Kvp#Ywl5P~ zXF|*4%1Hc5|*M$;Gj98yEF)6nyjnHYjMWx#pup5|)CJw*f;KR{{#3QuJvy_28I;Jy= z=qs=qOv}j(sl^!RCO+O;z$4RSe2=uaHDjZ8*Fhc9c6#41q zk5~EKcD+3HYARjm5_r!)9ks3m02BQdLvBClt!l3dM-6@ft<(1(3<=r^Yj>?b9+~S| zeMjZE0M|k2{1`(g7hQJDJ&kl((5YAnVDeg0ruBjz8DngAYxv37fR5A&L!TkN4d*t; zRl0PqrpJ2SBX=;CJS7Wh()bioU}x_UdzICJI4q29kWMD0AEM5+il= z(8muu#z24r+ena39^bR9fim0CA?7xKO^$;c#$t{2e0G_I_Pi~*jlG%fJ+tiaAm?gl z#|K6w(jREj<@@gH_*AV%jAqJDB5Ze2K;jY$iydMfOf<1E*@%d<_*&@E??tL|s|lN! z#M##GA6V^WzcVtj3OFQv5;$yO%XkTr?~r_oMM#BVZ!OGnzeTTCPO!WMU+v9wMFN=4 z>O+K)U7ca$4^t_!R=P?oXNuIeLZ0QoL*R5dnUKwt_>P&I@MDLOZIRq5+x^`?CjLb= z=c)QlZx6nT zF|HD#@rP0n=78D7?~tH7_;}T6Cm+>pv^^2a3fuBMKJ{VPJ`za%51B0NZnx=_?W^zY zPj&a3!{qiR@0|i>LN$@)=IumS+FN?_;tG}n6}~dm$0!{6XOxU&WT%Jr`|a=Z zs=EPlK}%e_LGV_db{SbPx4LFj>sx<+wmhxyiyYFRii4itjE8|=8^|uujdr9#ttQtS z;nMw8WlW6d!?1RbZHd*_#ddv4m7`9{(P(X8Js`AztXvW(GlEhl&JnCS-pR$$Hu*=&qzoI z59yITKpSJ@t=@!$h{(XgJda9=jji4r5~>7RvmF;dJ8Nm730y^H?L6(A@k83%Q4Zyr zgTr%99ev9cI{j(`0_D3qMX0w4iTOiB;1!5-pee#UAGW1ZFCl^H-6zkb+F?8Ukz&QC zrlzCMJ4)8;hz4avFOrhjtUUU(JMu_*iFdzI(=XPJ6$NzB3Bj+)K!*CcZtYE@A#&N@ zpgp&cMoo~Oz{ADN(%uF_wu3+IUrV}LTqy-plY+~x%*8;l`#vh;K+vl{H_+Od<-}0c zVD8%)H_6lIg;TvIX@s|48$cUo^)BlV2RqGdtt}j;Q4E|y*^_E_=DCSxi;wj)hwJ9EYqL%V+hk^MT-~}u*W@vvSZ7y*6QPP%P-q;_}`lF;H!Ei{CnQpt^*d27O*rEM7 zM%cLyb3te;&ako>9mds`P~$oM16aknNpID$pq*lcZEt92>#U3NNl6(2_$zLO;^g_( zLi0AS?;^*Yb9w}f4Z+~H%JXtRE=LwamwkUpl#^%pct=28V?$$FSb=JmX!1Jl%`wzE zK|M?KFid?i^?Og?L5in8@^hIaP&^-3;*W-X{`}eOTWh$h7C0WeKdUk!ExvyHHaNNO z*q9i}OYaljg{Ycgyl$Qa_2IXR+sE0i#Q{6v0rXU+rfuJ%yys@bJ_yT+Ml$d48lsl? z0Em^c0l?8J2oeM6a+FjU@K9+5z`KtCz*K<+)> z2T)5_2D$e{EIweyD-ynhw_x1YwMDI8qW1GLFhUIjP|MdkJ^dl1@nl)sFV(VDu2SfV z-O7dg5K=?Kh?5U-U02k3HpB*?4PfQ!ti-j+V8EcF-WVCFa8h<~$d3aCf?{KiLB;mN z{T6CI$3e3E(`m~qZy5V?&Vh~H$6FlwgHbSbFHAjsmKqk8X_N3-@q+)s7q1TX8=Jd# zLOvEurx9A8-JipDy5z&E=V<{v*BYCxgNOi=k?RKK4xc9p07t1SY;OmtVfB6t_}sjF zwBUfbeVc!#NBtXQc_jK~@qH>P+dieCs7M{D#HU$lK9)zYjJH^&O7b}M&0y#Dmw z6tK=Ny1NOo=<=qX%ICHH@%`&*Yt>KXf`*bO&YnGYBJz__u{%_Vx=*Gs#12KtRPgb zO(P6nyO8wd*@8uRw_nr&?!~!yF)67rgy(0jPTzws252gS#bA+oS3#Jm(Cr$BIpS7_ z03SI8N?Nf=bH-lEeL9w9C_Xoq5(rojPNzL<~w5iVZemdiR$3Zh6I0gu6ZB-vryP(BVv&K z7Re8W4;8Mtqv+1fZ5pPj#;%ma}|(YL!P$LbU% zZs5w!!55=0NqLEkqfPD|xc}y>tjRW<1{=oW^8_!jFtFJ1MTJ^pJ6iQd+*5Kxb4k)M zY(y|f8_omRPDN6w?Q|9JdK!iJDB_z~=xC%Z+9Fgxr=-O9>_fP_aS5Iy@X<~(W^b#J zGSvd$DQCF>nDyz?rz1*Ou{>pLNB38)RkZ=JD5ZsuA3qA+ud!@7*A;1`&z-z*NiEwh zD{b=*JxX%;l|dO+tlP8->>qxcWn+CsBi3`BsQr@(SRAm3+Ps`m(_Bb?W&pyK(=}tf z?Qro&MPLT_=+Pto1_`YszM76p1j&5@R9#-<{ewT!k?Du%Q9VQ_x2*UDb3n7&{tc4r zMO@V=QKA?39(zggS^HawmEc+iX)}_G#m7J?;-*!w)i*}~Mr35Xu@9yA*chnndxXLU zNsu3_hT&Gk?U9Dg`ms9S_EGaRN>Gla$m6F!HIdc|NFA^Q{H6<@S3L#_JgQq+`k{b< zQ8Px~9OK5nzclBBJ6U6g@>B@M1<)`Kv|?QrdJB3I0h}3l&>k}In=qByd00H}MgBpSJ(j$!cmLAlQk7no4pa{$>rj5F&bbhs;*R)81089#yg$wyyfxuaa8V? zy!s`eZBgaBI`wQHBC$AHr3)$$n?Ax@`WvZP{H5Nu8%x7n6X`LY00I2z+M|2-esFX2 zRr5_%hB;B)N9w%NjZH~E0x$S1m-Y!J!v?TnzENUwPP2`N^vtCbVdY5<6AjnJ0>NnV zjQgKO9q!iN2TN`XmTRr2emp@SOXg8u3TG^i9jK~cn>-;>f%r7jTWB7aRs_7!k5Nzz zW?|K)3B7TNlAf;D?X$8LxOr4%{TaXX>so{T=VpzLiCAIWs15PO$B8;A0XSWEp>xOI z7;n@eWI`}0m;Pp7U1mH7%r#s%T$y8FHiUOuKG$bp?F8~nWk9}v5b7ZHz>ap~b zrIQD^#&TzhZXWl98Uqu)z7(X**&B5pAM^e8Z??l(>RFT+Ia*$pE%yi@P=3_T-U~_T z>avRe`7+UMXN&4dfn&+n_!;3bgJoJjieK5XzhA-`HeZ{=F-;aRBPB|0D}v%2pr9-? zYq|1#B9LP>$UGEbMc-mn``WA}q7QrN&wvo}ghId}FT6w0IZv8UJcp9en>qLTt=P?z z=l5;`LyibZ&ke;q1JShmZ+<#zrvzw}so9b+0&_6==g*%zYA&elLrHNZ(t9?=0+pZd zX7N?P2-2(AO5`YEsLYlFv^5+yQ7)$I_E59o4QAJv@p0pdG+8FMR!}dt{)k^HttX!2 zc-!(Bc83EpR&6KE^y^Q0i0JdR^|hZgpW!oEtXQND&EEWil@)aY zpFyEkz$wo*!*0MWj~a(Pa@#Wyu1%0o<%}+|YK!DU004Pw??t;3R~Pd04X=t^0uR6Z?zfgC}r}<45V($8nk>9WWqGBB}8oP6)y!1Lm4FRW5(2Swuw%0djREWI_6ZV8HfW=s_R)@;ObO7A2*GRt=#WU! z&3Knzp8?C9J2I!UGE(s@nHR88Di*45B#OhU|8SLXxJ_-D=0whA-9E&?V|sFDeUoy| z5es*m@4fDpyiM$@%n1NI#|Ix+ByTVY?_HgkI9n>cv}SctWL6uDdmdce{cgM&Cu0jF z8*~|dBwokBrS<3gY7Fvj-Cb+%k1=pb+u0fwUTrH3@2>ntqrqQ%|5+79f6%W-GVfzx zCwNN;0hiQf6TZy)D)0a(w?nF?r#>i%!bKiTsZ=LUJN(f$@Kw2S=964uz*NNIH1%*I zte)fKvu>se=Llx#uKsiZSHe1VJamAB6An}Wo>qn@Xkyh9R&!6e@6vbK+CtP5c+n8F z!BDB)#D|l;4=msOT$MpB`Nk~0Y}c&PQSCYiC6$-O*ZE|E{3VHNukPJrhb)cgC?9pd z?7z&LV1L@}58R0i=lLz?o%4mTHl0mgrSMbX1*T(RpXI25B+FI({v1avuB^4guxh#n z2(Q8$G?=zDU2-3B@weYVxfJ?-Ys;Nz`a^H96rOQE2GEoA?psH9x=$<7YdBV- zE`XtgcO&QK7QJq7Bp~dRliGB1A|{)Hhso6Rt%^X57gy@d0NpjXF`%|A9zSCM-_2nyhL^Ocjc)1+$93T&4@_K8taB-&k>lj ztn8wx)TKF>o0;)oloqLrxo~~YZg}ZV&IhVOTA+f`i_mS#;LALUj9RYI33VHad!jj2 z7}FPTIJemzRA}17TLrLdfvKB~l4uc z6?Id;wv~_%XaiIUi=@{hV`(YstOrOR+18KDF-oYD9d{B7Ofgc!k9Y-o_YQsdpLmdY zrn9b)jt$)dlqzouLba3~13+w!mV&*!#DCZ+gPup@>!5@^9-~6k*_OWZup>@%(n&Sf zV2t>|bGly*2j4!&^0;`*0yMp(--yG-On|1W6AsdysVodzcwgULp?Mr>+OeKwM5-TC zZ*7p1CaSn>EuV$;9vkpe`9_Q#F58#`MLlAq4|lhVdSQ16$lax4diYwwL^K>7STkjR zqUk@UhX0yTve6%jzT%$nmSl46uOuQQr*3;aw)(rf&eX=dLVg(7fnDfV){Ldz6KTmt z$ppr}Y^9>~_xndAfSHa^&P%fpo`focIkFhjmg_w%@U06Q7QReQ zKA$E{nlQq~)T+)Kl{iH~Hy0=F^U-p83uq@OPClTSE~fgtSM40VD1Bxw;kMa;J{Zg` z0E>l7OSftEFg;{sCu~dtma2~yPxUCdagL<`lCBazC-rz4=#DwUa50XX@rEOW!GFEH z|C~Mg_MQYKNzFxzKXU{SFHYQ{Ts!N?Z7N~4crlL6^~K4Kp4qpQ{Lgpf0p*I*KUHMkq0UYs5A#s7gQm8m{D-~cZGy{VwUdU0wxrXZ(QuQA!X?DIqg~0U2*j~@o8yQD%k2e*6CI&k(loPC{ z*L4F=IfL2IM;?74Pn+GB9Q!>V@qQ=MuFy0i`fm`XjZj@+85D7fXJ0cYB(n!ulwobY zdR6`dW%W>%&JDDu!#%7=0f?oklRUQ5wsZ_`L!ez%RqvGRDh0ICGL#cMRv#fcD_jAZ z-VjX4GS+-%rm9B7=h?l zI#=X8UXMlcdgfS%w_-tf3`aILUL7d`9NuoOTY2%o@GZPQ*X~n`rbW4_q_FokQP)kl z`2O?jMmb+-!mEy>?27HttuzUV60mNq3wyf2axBbYKoqB&tuvS>Ji%_OkC~#Epe2;! zA(4nr0GuXKx{+HxW zfB%l>W5Hby!qH88{~3qap7;XYSlIi)3TPy{wums_>3fcu8MmnP__4plax>kMck(@o z5!~9Ch{qbzWwwOc-=E#1C3XAr=hIvsMjFhI0()GIG@R1Y_=*tHJe%#J|Lvsvr`*Tv z7^n&_PG}m_{P`9KBML@gyK6vJc^@o=a{*`#*(IE7BzumrQHc@2eRh5g_NP8S`&v%Y^UfydW|@7FAoWfzd22AO zFZ3y9k*e3}Hy$I6*52zvA`C>GI6cYy=1N@R0f&G2d$Wa_`iXTlXsU}gTmp9oAkuz7BikZh zyYSL4KWDsBzDYs&i1Mp{y@L(r3ly`K1ZTfR2DBDVXIFoKEvY1k$v*Rz9+q?lY%ztE zpWjBvS6-b|1gxeJGK|2`0vszw;TDrkgM@fhQo>_%XT;rS@>qmkNuj6v(#WfG3^V%c zUO-H*&|yV=sobAwUog${8s|)TJEZ*U=;lZ#tai`j!{Sg)6JePGV>gugajN-aqhTP{ z;wOs$6PTg#n-8fBmRkkINx{Dt1=;@fUH|hD@mIyF;0M-1AnEyE^6h5cz6`=f7TIud zH%E{ljyzYVy~w}F1LRiC9l=RGh}?TYlOny<9JYFY4;!Bx$^+GzBPgi5VFtfEPOdJx z-qrM>B_AzcK0z?OeS38z$j=-|8;;!09JNHJzw%T9U=~LRLh2lBImGkkm8ZxxujK-E zzy2RF0uhDg%2im2y`N_Wi(x@k2&T$@PtT2rK#$F^?3l7ooakald?R$6VxT2b7@=#v zt=^>i$=}4=u>TH(?d0UeF*>fkXfNr|z4McI} zshm0FaTJYznH4+?m={a9saOlK%nllI!Ug-<>Fw0@H_;vyg)Q$jGB(W5+3%LQj(X_N zRcU-%EnLUoyfZpt1QlwxY>J)dyHz(>Uw^70HKdvk^Eyt_pcO!l{R1{^Mts}nIQ#IE zr*}r|up(d|jFL~*tjve_7CHfIp^9xc=+MGMZTD>Jem%}ZsH-y}x(q6Hy~U|jRri}~ z6g?2Mh;;kq+j0hM4bY)~@%{Q=;pO=WfB&h-i<-zYG=3SW(;e~3V_vo)Ye3`-Gg0PJ zD;FbVHPz<^s=q<-WYQ==w<>@fAZ$J385bx;S0_k~a%fyJNE?{yj7qj?nZ)e-iwh(M z7K}26@t&_%r>BCbSe18oh&Dc;uic(dBs0b;_+5u~^HjXsQt#busoi`GjJLPmpu(}$ z!c@M+Trj&z`7vmTyc43O@ml}kjOc{rwplmN(#n@l9wcQb-cK1~nJoV73DwD6Nitat zmVX~&Uwergob;bS2An9){p?1Bv{}Ol!6%N_ZOl7c9qGgda0No0WM^bkDv(y|Fwy7P z$rBfAn^K${$uAp@Zha2~QCr)sAVc|H4?BJKu4kAmg2=&3r@|(F?HaVO$ohm4IwhH8 zGw9_*rH0cJoE|-kXn9>009;&KP?3uiM8R~Zi~Yf89e71Jvw@hJLWkVoud0AOLSm4#y^ab zvZ&d$6nJM?dY{V2@Z#zK0B*8NJVh=J7NID=%c{PpQ25yJdGq6tE08D@^0T8hIA zz(>*eVfx>|oq*uYD+G@C^TS-N|9?MyT;t5^;9%r$c43|eZ(oaWhtNx=V7A!O2KTbR1UWgwBEI(_QYueZJaI*gj%fo^lp za^~;Q4wYpgpIWa_X`fwz_PXo)06u_9u)9nRq= zoxYy%Zr8U!>epl3AE_pr%n^3D<_XZGd=@R~mUoj#w0QISbqL6iJfFR$N`Q=3R{#ND zS!((lpBe_e0^>!U9stqmo0yKc>-)`&*U<~_(MAO3sMf0{6BDaxR44Yd=iF>!B7FTX%H-=jFG)DCzK zTg_+p`Q}gZEk9OEJEm&o>h(bl9K|!#;tPE@fP`h}Xq8)4Bkg&r2 ztPYIKacXB>w(_gs$wD)HkB!l5(ibY zWY|;Xl(Zh+J9Y2={SeB-egL@4fA@^v-Yy=$H1>RKvAsH4nBa@ z*=Gug369gPaNG<3&-hvxA$@7lrz)ENS*jqnZnsOa`)TF% zn<_7;ZXM>VI~H|_&iKgR`N56yu#SY=N}RIP$}vIffsoY`=5tdG-J&@O__3rLYDpiNPFwBDBrGYSPAKF5m4z80g(N7<%ZT`7ZDKyPxm*j^q93y?=jm9K+0cUFY6w zuf6u#-v!U|V#cf22k62Yj!)aGQY8RAfk5e=8nDSqkGcVuED2#MOs}C2Ho{V3%XC<- zmeHS6P2|#h&k|Nif%XtwYzaLf?sOvlC%*ys-ApEEWu6BstIUd!FO=)fpXUJ}j|oB%i7@d-Eqah~b@bTW)VS>ksV>2X zlD3OL;~kEx!5@-=IvSSFuKD{AbOCIZ=b|IqZag1gfUvmPz|vTWfKgl6t395Mn_QJg zZ2I}w%&O^sCYQG-3b)swO5=2Qr@^Q-HPdI3hRv?}k4Gb4Nv+b8Q7e-3sLnD3yk~gZ zuGZp#EXe9RTh*zOdgy(z8jgNbgzop4cT9>`{O%Uzdb*|Ftfu_=XDTxuu%_MsbF4l9 z`;l?LNt9T>OJMr_k=z_mu`OUFr3*X%^lL=UcNiZ?z8DYnm{F{H+&N<$phES+D8E-) zj}B$j2znO0)|`+|iqC`dIXN(v$H8cOe>vbDij02pee|i}TrD4W%%AtV3@8E9E;=${ z)l3oDLD}@}u!LJ+^M$%^0M^y3V~PLr&HT^XOp5mR4LyV$G$As;*Gq*vR#+K;9m~& z%>r)7MFWm@PBjoYa;%(Iph#COGga8QM}^0b!hsS zh5AkLdmdnATYmf69H>?|*!7H72t)_lw=}3H_K(F?6PGoYi%GoxqIO9q;`vg$Qb#3o z<~}6%qLzk#E6;6r3)yi$Evlfj&7BpH`B=b;*wKitjP$0HL=jPA>=X6zSE?-FX7+5c z)l1XsJ5u@U<}l6?n7#V`YPa&0_kz$IqYl$cT=Gqdz<(aWe;r-MFb+Nv zFX1dSyD5ETl0Dv-^RJ>~%bcO0Y%hmchJ-aZ^B9#S7r5HPStUk#exUkQ(l($&5Ozk2 zFHWFhX-qFdod&=hmlaihXGf_sChE3d!k!^{{>NNPrx#P!{by2lun?qm_(T7%lvf{b zJd0fkWgXazsW|nu3eXcAw^Hbm+K!HxVW+On$nLk)G{( zSJ9~J%UU%_Xp8t0Zg$<>_tmF;*sI_TI| zC4VP9U!R)hgou*;imkW?dP^_hD+E=`GV%`MJ1$HURO&nCHfSK_O4BXX*T%aW;_3L^ z0=TuUi`3a_7M}03d;MYx+MTJKxWUBRYSS*Ynobw~tGh>ZU!_3=o6DnTG)>pe27PwmnNKT*TMDTN#_k2x8qzBdETB ztp;mO>%LB>x?Wwl8w&8fQBVAp0vBHq^#Z)0w!iMrLuFkZ*>vm6wwXunFBs|W?|gQ+ z#((EKnRl0<6FUahU2{s_GI4j$kQ|kINXcSA&w-_k#4JJI$nZ_n8E|)F8RcR$j7|4% zfG7G&g{_l_{-qS=I`=;edC7qUL00@fU(^2?%zVJb-*d&s1VBQK!D)%!j9px)Gj>^g zlJ6Q0!P^q8;v)Ig?qh1jArl)GV3Rj=Lh|Nm24MD0mF<|$rvALxD*6PZr88ZwEoS^_ zo+Tm6bgU)3D!BqOvPuZwwKr@A0h2t%@RQNfr>3*rxR=y9LIk^uiYAx7QXT%s#FT>* zg-UXMfVZL|LMY;uj90T#q3SphWHf*@az+PDX`MeTSa?kvt#rAnV!1+v)(dwv#|l~4 z8F^)v10o8N_VVbWQA7|g$OPS7TK86&j>AN`T4Tr+S3zF{;eNk+IYqU#~3q1Et_ha93%gmg$?m6-CuI~RH z_5Ax{5vM=k%*3+%uxsAugCybdVpWI>{Y*&zBzwb#_K?sel$#%u#hQT2zX0^+1Tl(r5ws@E z^+%?RI)PL3T|p&mLQN^`A`tuHpMpc>P6%*9X=YCD$MeYcJODl^jT(miI5CeA-{e}M z+p?WM-%pi3oJimtcn*5QmInkcaHesCsSKA#`p-lGPN;n5`{uT@5z*5tT8G~hv_s2R z^0)(Ft$`*0Fiu?C&WsPI37#3PeUNMmlNRC%UUnH3=ObVbb93fzy1%=wn-H{L_bwFB#Zt7`H8lWHB0Z+Ako_59b{^ z6rhez?pczx%k?ycF+bh^9VEc-Tj8mqdK!R7#Zxd)W0BW|j-_$%zR4_1MV>XT z4;N6oY@wX2G#~s#ni*_&ne4!TG6IN;@*{ZKm=Y3+ ze9tK8M0}G0zcSgvfQz6*-~Mm3tO>B)$r6fP5kP9dbOWr8ljqnw=jq5d4ud1fSk*KT z_KQ|stzHlnOA0X^CQ4Sb`K9_p>dgts+fy>AQrh$}2Vx{+Vj-?jK4;%&7Fdp1thIKB z$O^!wgoofjI-jxoHn$&nS9wZEDHCol))e|m9!BrYcVxATc!DZ4s`mC9aLDA};nUgR z!~Qw0WWfL*Enq}W6-$OPA4uF@PY_v-fO3K{trxb&AZ*_Sx&=tZZl9Z*et$eYg4$bD z`dj&mm(`5e;m306@p&u0?5ngzjMSPVR38g}#9st9@aea_))doqlHwILJloOIyfL}F z5jW-bs|N{%ubl$Z#8#(YNMN>s+MmGZ^`!ZNq!@~en8%0)Wf+iGq4SF#%PPg8!S%** zS7UO}@k_vk@l#&!2D&n7V7|Kv-Qw#XPot}>jLNLkFKs7EDEtD}FrsKAGgZIrv!=aP zFQw#WQ`-FX#>~|6|8Pi=`R4*Z`3K)d5l_pR93*87O+^ zR)Jtz=XXmu?gLmeofWRxF~F{M(4aTKc0_qZFeWZ&tv8~)wOQi+1F*RQUkI4u>EgQ? zS!{v+kQ6I=nV2c#_?xqyHpDXE#>r;!2h?4kz0;P}GyET3%>U#<95EcuKQf;Q@?^q^ zG=W8v);E8u$8paAFr(6+Wx#w04CzZ#pM{KbF)7|m3V)jN+TL1WAfG&2)g~@Ifh{8U z`_&}Iazod=UIy{K;>wem?gK08ngXnn)RgU1u1>ejP1skesqMghW3%7~7*6>>3U%3C zDy5n#F-W>1c*53XMaV0<@)dzU$jm~y#SCmdn_Hb9O0vDa%9uiSIFt}qqR4cwug{UJ zH@`m2eWP)`gi&m3HEBS>Mh6}D{$2-iw0%e5AT!lT;^}o$S-N=icQLer)Ph0>|*-p3`-#?dqrhvWBs$8hUkMe2zM@50-z&()Fsx8DkZxpXD1)$PK% zlTjS#c6*zT?Le1%^Vyz4c*mVcVnY(E%JUnDZLM+luaLNA z-lGIIqx34X&T~3ZtFu%AjM5W;qsvhsV*!olD}>1I!rG+nZcdYuiP2KTEO41gk?&T! z!?wGy=(AlABV+bQbfbW|+jb_SkNdbmO%m%R!#~DE>_7q&?CWN|ed-AGN}Aaq$ZAC~#3AvG6j3j?89{eGZgT0vCG%RPthrvG~DKog7hd=ih?mYCT3WpW%iRAjax^ z@4WV@MTM4tDR3>2@pTtgCw^|GJ{_HFQwqO)@S6wmk8Tn-x?#6c zyx+}%@Hp|e*bn)P|6S(~gZI|6;=_sxtgja#kMKd^trq_kEBn6+e{AWe^_$kQ4?J>D z*1I)O_1p&Um%Z;L{IDfOg9NMuoeBXu@%#O3W!p!OS+Kq8_+w5`EE#N0xV6iyQFAa3 zAOU}Um(gbpPdKQ^5LJSkf@K(Hg0N~wR||MTg#^6Nbkp z^)Uy3Py6QEkgdLu?+4;Qz9ZQjYJ#sjAoDgYB=UjL<(uR~gXf9QXAY3X&l+x^2&hd1Y0T_fNg z6AQv-s{Nq`o%zOdysd0Kkt%#7{NICuFRW;a$f4qPK?AXvLae@|vv(^8Ey#k??jM5w z_Lz#BdZDCz{yEs>Y!l4Hn2K!c0O9ruG7nMQu_vF3qUz#@L3s0SXcpIY0CJjvR8Nrx%`i#-M`u{os7Wep|E2^`2T56{8ul> z-UBVd9!0xxZI0%F7Nk|IpLR2|_jREox!? zTHyJ!Zi1{&1;AS3zZ_-J+oR{-WAo=l2!Nq~s;B{n?Y~#>YIN%S%xE-IR1FwLNpO*`0RCm8>u>1bSM@z-~?2m42*EjP`}@UE_0>5ycX4%eEvl=#fe(#T>SDY%*;zRwkB*T9=BDi*^?$jo7ka9wi zB9%{qt?Si%J7hTNu;QqgDMmA#As}17%(%ZKEZ&84b2ejKd){+4&Ulr|XE{=Y`NL&G zEUeMxXwBlv?3z$FXUsUrACLcXbDr^Z-$M?n2|)N_`~AlsmoDCmkqX_fWA6Nc68X6u zRwYIPdj7FF?V;NBc~jQ(Wfk8;C;iXanfJ#V)~+X;#`=6o+?h)7EDpw)v6#Bcn(|ZW zdCJK`OR8Zt=~;!XZ8f&^5aqOhN>kc;*$TA!|bcmEb#rwLA}^~CaGVyiLT_q zk7Z6{zd0~0%*f*hhV&6Z<=+%*N=gH3*7M&+lxY`!J9_CMKKSSQs7ndD_NN-Ty|U`C z@eG0oR6e23KNkDp$xc;2`d{blc@B^^PQVTUS|Y+3K`Y|AMceGXPs9-`NklH z^S=Q3({3E!i!tw10TjVN@vjT;=Pw#!0S;68`p28jKya1`))X61p4TlLO~jqe@@)=t zKUk0fhJc)JxJcs$aL5YBkC18_>z_xf*SwyHHL9%nrM_hMejc(Mi7=+SN%cn2K1ZTu zB)sc8!HF+`4L$`D^M^hctkJ6zp7wklI}tK2AF1E!p(BE|Htcgk73`uh1}T?Igc4$1qd(`g&5gAB!z>zDfAo()c{kJg}LmJN@AJQu9#e*6d*SD-L|I zS@c6=YrUoM1DKFi%B3e+uaAG=Vq;#hnBiOg4qPh0jq3MIZlM_5I&m6S z%1W&!E3hIHrB_XeJKw+h<*pHgORH<%-vxV|K)=NIP#id9e7X5n8N|}~&*aRYIXSP8 zAGzLBupLO{d}vv~L*j*wqvhS-WfpRCb)s|GPeyX>V|ab3-98+}NL>^JT_nAWj`&}( z+mtyF3L2~afC_P}D&Zs6cHzd5c8U%FWiMn3{_>!x#;{M~ zV>f%~(#;g4NRbtHrd)QYF<@Dm(6wWDgPn_WF&J?|;9vSL0C6Pq^gY~q-?BuZRF}H5 zLLX{&v+%_Dto~`>!%r>kvhV^(!jwwSvE5L;q1E2SVCs02^Sa#!l0UUxr9l6HV*cFs zl08GLISYW2^m;h($=>HH`S@OzeH=6gN1ybpX5+kE*R~$cUG}}xd)N7_q3TOIkAwYo zq0Vp#Nueu!S*6JM1W(toWn#eIOgMeb@3)u-26D>xVd-M??>N>kjyZZ+5dp(=Lwx7cC_rX zA=TV`Z*9C8#SN{ta`Fgs%U;n;h`CP^?@Tv|LlsVnKGOZ%VjYjUZ}c}@Vtw4irTicR zNnB^~{2ZhGPrIgfpc{2Z`NX_&ieZY>C$t8c6!Pne3uPvfz_sr7j5Hh_bW^)QH75<^ z@rgH%z>5xHr}polWein8AVuS(NHxMH5WT$@z-j)ziFKdL&p6ZWfpgTEr*t2Wz*6!Tk8Y>w6) zdz{-=f(yPj8RYW1VYBkAV=fzCSEmcb0un5aWXVe3Ok{;`R<`TbyKaGxd--Kf@v7%O z2;xREm?Il*1GxnbzK(@_R+V(Hj4XZ_R4TN(W>NOPl2M`qNd(T2&}zUL`N z3Q8PFfS-tx27gU!_dZKKHLR6?LzVM~lMI^yKyuv!iRRHP4@wRFIWzd@FPJ+o-`yM| zN&%GYXh{md3EZzqVi6Z3zJ+Vo6$l0L!ap<63GDa6^3Wk5o_R!BP$(RCH z`V@A!RtiZKr%Z)WTL~sky@(|fM}}>LM^*gq3FHbTlA4Et6B3VVUTZ^bPnWOg7>{P$ zf3NQ=0~u6H>nW*LJ#Mo(!y0d^0N2?qY~?&uL^#UFZAf_M*x}_UOMtGfkGf^A4%efc zC7<1Xt>4H@qU26isM;j#+TgJO5AGQyx6n}Rx~#5Qxc4QL&HxQIU5-i>zCWkih|_E2 zJ^bk7>e^0v+7FhB@!lB)For#X@8&GbK@sZ0Hp`ACs?tZ z?Ycb7>klUkAT5b9k8q0*^o=S}} zfX*<_Pg3YJ^v(K@^>JftWE2dKKLUl0q`_#dde49n!l(#Y{zk({sAIq7miLbjMpJ3jlp7ELq2j^QA;1()e_WWqZV6v1q^-E`p-t-YQM z50N#Vg8W~i5oWS@^Bi8+PNVkYrKmJkj~Bt+KT;Y885Sm6PG@lCXo}i%fB`4uuv?#c$A=9O+p&vjF?AJ{MS8#}^VwBPnro{3*tkuy#!t{ImR5B~;?!aXKBQ}P zl4$38G|p^u=1ZpM@g@~wT;to5ycv?Ke0SH&F6@|LG&%a+_~UTg?AGupbx7WYL`>u{ zp2dS5F*f8dOmPDHld3fD1ErZv3Om}?ctB}T>-*@+;ZbRVNs23fS6n3`j1PTbbyT(0 zQT=F59JkJ>eC_3Y^CjhQ_d1=d(KpL*F^Wr#LDIf_=g%ntf{1LcU=)#^T&+0BgPjp9 zYDXj$WL8;}mm%t5#PRwg>FRe&X{X+ZMgrtOOJ2Jm>vF*z{X$~+Rn;F4?@Ie?EUk<` zB>|nu(Ix{QeHZGnPDd4{FiphC>|`B7*WUS$JrazZcMsw9m7+PwR^6!;>vQY=KJaV_ zG*?rSi8u5Wn1XuqYP@^vVDTDfxpvkqoW2V_R;PvHH+Z{eFcaIna18vUDnEm}?jin(P%*az8yZs3l|`n#uOjnuL0F_}Hyfr1;3&eZc*_nZ$mQ_|e*#_A?$21B|>{icgB0F^_$| z7#h3DZxiKYlp~jAXiXcaE2Ruy#B2q~Hc?TneSS8c7a;druAt1vTsD%&V~_II^8U+f zO?MbizehGXmGbZNr5a=)wVVj|)x3lHn=w@MwTHd~5vAmUtyX|On~y^*S@LU#*~gP5 zCy}V>pX~wP9%d6%0R?P9(i@4;c0>6#OWU+7bynKfMM?~~vitY4w`!*~#=3=M8}*i` z=pR{V5prpRFvZs{9w^VAzM*7XaD|*C3&tqnev;2Y#KC%R%Tlfvqj{{5Fd?*TjP;k& zbG~0hV0BvA(d-FypVm<7O*(MlD9`7ErPC6sHJ3T);(sskN!E zv7yNxeS!DURJ)-PuYy(E9)qkQ`PnD_qbuLxH7&%=7lV60NiNvtCq4clUTX7>%CtK> zzw6}+A}Ul)hr}G8YX9z(54G+st^J&4B-sJM`k`WWVIGrGqH_Jd8a$?91Shm0XHCri zeDMDx)`I4vA3UPbuifTC?)JU$D^eP$Pee6(EbBU_yz!)#$PWly6BglLHfoRj_$$4T z8(#ceIuDH?c#tQ?UHtIXR-d}FmlPM?6|9$+0~WYnXB%LJdU*u>K3S#^@&E-IuRrf` zdu2La8|cZ>#wR06&%YyUGw~xu%&7L%?qAg@6M0w1Li3xG5IXoH=*blF1$1nk1y_|5 zkYsKV23)+svO~Sthab!949!X(IreTPS2Lrm z&#S@0^Y#WEzg67da;i}$gOKQtKZc|3X9CsztHRK9VvNwY@pwIBYT1M04z|AYkd}pm zaUPGz$bCz)?rS&t-AK#X%{fHk-C%NEHJLwt*MCEUVNtZnE&8*zq>7A1HG%)@SyA8^H_-=iJziDvk_?mkfAs;&|o=Rn2W{b z0@sbADf`EtRn?(ThqGW+60Sn=y*~d6ior=#@S72$nO2pW&Gt8*AFn*{Q95;svZ*ia zL(9#aijXOH;`+^w^{sUk;#gMvFk4AY6^=~W4m!{Gb(=DmnqsN}0WVDs`-gn;j-@xo z$_)ha-l@#PqRHRmp>uinxxlIDbo{*CzWe8VE5yK;BVdrX(-%l8?2pb;1`GjkR`9V} z_?|(0wGEBMEh{gduNkzS(xr;{y*9kuwxnoVSQGyQsw4!+ry?)xB?AqKn5C z@z~{5^SfV^DDIstwq@9-)$zxRjL={c9p-v_K>{S5w@@sD7F5%nOI?_D#j^ssQKt-} z+cp(Hi`JXzQxu8f-Acg{w1T3759B99J!x>i+5!+-FN6C8{PeU+KMJJs^LrQX#7x`e zOWb~jF?VUtRh8>NNddB}&;99=9sS1V_-3Nkb=xHaMLW^c?BOiCskjT}s1p6VqMKAW zEysf$rJi@ug+K?fRCA6plA4EbkIq&j{y6Z>>q*+F5A1IA25g+!Q6_>kZ;{K5F+qPu z7OzPYs|0W%N`=bwrR;ia9=C!uw%4rAYw6At4oQ($x82MRg9@3~?)KB7*4IXjMe?na zrP@zE&V$-z{lq{cQgo(cYYnet3E-HtJVTX;!VvhkdS4+9iUo-_batUY7(_+yKls1v`sN8V<)(U9In7`D2ac> zsYVjecq}A2$80FtAw@UekW1f&g(vgC_z{>zwxg5Ubh}x#e8*e4eUBA_D7B~DzZG_& zg0(;MlO$f?t=?_|`lcC^icL~|@t}GY+8L^NNu~MefSX3DN(;!$@%#%h7H>6bNsV)o z$4Kq_#rZSAt)8ZGo}Hf5eYJG442k=-p$DvN{Y7`Lgj#PmeK~YD6^3IB^@5RYJ~M<5 z{Ydqts`Ex?LZ%eXR`1OW=`HtGzh*hY%g<(X#0BCsf~OQu^;4@Yzwq>!_ebRjMR|0U z1mJSb)p;4+RXH#lWmamZ@3hV|+I2XXJ}3%6+B-X^G-v)h_mmLLoL{~9SKc|eivrDQ zeoO}yj?JKaTz93=ug*bMUc2ZIPclC+Jl|ydUu_}B7Shf@K`s@U$+!?5RI`nWt0k6* z@D2_pTiYqEP`c5R*Y7k%!pD!vy5|^ga-X*va{2r@y*#fLu{?7MwUva_n!bKPZhA`j zB2?F zGVsY70%h&e&q#6a#l4As;M*bsg&ZZfw8ErT!Hec2dogK`WW5b)57@9uG$IiAwmQ^S z!BS2I0O9j6Qsqj%GsmOj+3B<>a?t0EIc?%e&f70OLF2@aRWu8R$FW@6h=%Yon$Hsy z`9M!~bquqa#6ij*rQ{Sl-t<_=(Pg4@MFA9kN9#ydz*2sqohTVsP8tr2h}d#TycSy5 z?F#*Igx;>A9xXcBzKCmKWvJ{kP8t?+q+GOD3S?YZg!o9+WQAtNZn8E?>steE%n3-k z)6G#F`+x*pzREKu8ejQy5eORnaoEV)`9f{+>knUh{MR;nZ+B)btF%N0ZW$SZ+LCnD}2PJ=X5%zvlJF8j@gA}*oHlN#^D08Gwo0T7wq@E^&Wlswe_oK){%s% z2KFPZ$qnMO{S_Z|{_~trXgkBFv$X-Akc&hAdNNo@avRZkm15>;Q2v#Y_4Uf{fE8?O zJ<4V4`ZB$ITI%S4OYQ319xve(i|FLqU>w}!r}+)e(GLaiozoNmxe+gY{-{#>msW?L zht>K`XtMkkrqsx1$0=7)1|((0DTCt{oEsiRkP+S^Ov*G9=Cf3=xJhX8z2f#0FUB46pPbLAH>B>chC1ERP>JQDgAUY4EXeTD zqEu}7YjUr0efAy>ddan_A*D&T;U&wu=7&p`%K9N@RGAqWz1rUUeyt;5CUPE9 zFeJxq2EL2i>xTGo$1Ju<@iI+2=fJNA&Bj58r9iZ3$k5Fc*uFm_Ik1m-$Rpiiq!lM zjxIeyw`C*oqCK@?k8Qqhi-GqX&vkoAt(nGtXr@O#pxvrv%Jz?jYLSj20!#pcLx1)h zW_WBbNlG~_>&iPl(Ss|hg&A%y*wVo3Zt8I>=V(9+Qdu-*v$GNLucg_}WLs>z33OoE z9ewBudDo=B845;BOGio@f5Av=E1z1a>hu2 zn;^AzdO<50)UZ`qW>8pwAAMe`9Y8N;;-$pJ%%zPTfD6L=Yx zZ9J$5^aN;5PFB~T9gmOWr@A`{F#MiybOe!L{A|C)+Me8I*zkXUj|VL~D!pC(_@0S6 z$;6FR483yyZO7GU#zI5VziJN2UEY4n#NR`X+UHX2kL`ST*5H?qS}JB5hua-$J+#Hq z-BJh*y|KRWf1c~GbLaUu;~{Dk zq&$o&BdTDvPs&c~?zu(2WX>t|iX6U>?JBLH0~Iq@mh2*+JiHWc zQz?4ODQ5$kEbN`dupX1m=H5)++rli|$8|oVcSUAU4w&tSL3JTTHX~bm8#y*{YwNAw zO&*guJ0WUG_1$PK(Ae?&bgg= zc@(TJ{ax0@lP) z9~rZVxPZgOCjF?PR&xV#K}odIy*E23@zIz}OKo-AeN8aLqID!6kklX^(MzSt*f$pYhP578fe9VOy& zp1w(N=`t)5W#Mf*X-^&Ii#2+*IlR>v>w#Qo6D;YCPk=Ifidqve)*hWzYG-?tUF&dm zIh~OUXvOjpM03~0cc2lPb*Ox|mWqCZFVg-M?PmvVkKWxnZc4s0dB&n)y}ArO{aAL` zrCMZ3Y%>do1B}=(Obfml=GQpHv_Z>ed8k6cqs^|!by?B2dyy!yc++(5uj5njrc5qp zblDm1Bb|r^5bu*WIAOg-z?`WqoT8n~d~q;FXkmld9m((cCF1$qwUtgxqT_C^_=!ED$dc^c+L&}f!5;rLknOemHy_ELnU@;m*1UWZ;Na0W~(gftM3MY(@ zHu6&ItSZI%Ynn4=#D3zix-qZ`9+D-N=K0^jskk$)=-@M4%TnTRri5WSMKote&1$*5 z*%16_k&9ywU0b*4?!K(?1+5zn`&!0W-Q8VIGv|ou_AOD9BHT64(qnyyt@lWgZiIGV zt!1EJ+g1lnjd$#^sOyCg4c0MEo+mta`~YfXryPIxXSbnKpfT@0*GyK5sS6U=)oV9U8UW+Y`W&c>sb$+D=f`c#b^J!2ei ziF*TO78kl1?OvhY8mjx=aUZlyccy}I{kknNyb%dPy4737tk56v+Y&vs_q8li>r`ur z4!T8mB+@2l;Msq8>w;!kOH($-Y&1vEDA&R#2F{qQ5R+Zt$U5YYIjJ!W7Lx>bNHs-{ zTDiF{#52T!GIk5o%v(N2)tq?e51 zx^YUrZ9~t2s2dGAv>bZA*rL+(b20njTgInxzZWYxdnV{v&Ld*Vty3O5N^4P>08U`u z5|zsOIDfuq@x+sU_w}bpiYVAjNP4QsE-suM)=w z%$ZBdcz0V4yXRv>veMdZ8n!*=N8Y8d>d4@4q036XzT&2^Y?bqm)ct(rfw3*eurii6 zP{)+3=JY#OUH?k`BMs*M`_s}D`f7^ED)nxm@J#~0IWOGA@vEv;qRU{~u_RGIToP}s zrQC-bmZ%%M{) zASlYx@66^2x!by_HKa&Sk6xl4yehzQ>Qgej8KP^cvn&i4xwwzm>-}XEDmE^nUlzuia?9{<>W~rb|fm5>{7xgTMaA+iAR`d`S(` z@y_3U#zVxYa;eAh+`-dMEhwb~kv-hE79BW_a(*1ioq1jMb99XNf7DtyF%QaKw^m!F z>=Ws%oTBZ*ZKhijv};B~nPn`>3!U;1r(J!=)iHqNUIjZ5iWp7dC}-eQk_6n1@-|gy zHQFrmxbx@rf(XA%7KX?x&~gEcQ+?R`*AipmC@T*vys7m8{P+R0!%X+Rxx6M*#9H6b zt4|w3z;x@YLKQCsec<`Ow7r-bT*n^rHtKMa$)rrMgg;5-{@7_wqYF0pjhCopKT(n* zhH3lDax`N?G=A%4o^|DL0^)PqA2qI_MSXoa^~`aoiVWCB$Zl?O{CN`3Aba zY645^vb+P5J_8Onp~!XWu`IS;_ue@kk241-PIWRThS*yyLTh^)%eik&#}bY=Yn@BZ z3Nql~kWNx2F3%BvJ3iBtPhh{Dg|0%-qi0+_W$lu!v@3i?0cyin zisXI3z70pBQl2V|_D_Ri*324urm_ncn~#XLq6|rK&6vmq;q4Ovz#b%0-^_Ua!YI?X zmW8#6Xq<(N;RN`epR3-zGP4S>0e8GO+nO7ro;>2^>aZA2OY2GA!oO`{DC=6+IZLE=F7-Sf9vTmDbNST9ZtE7^z}FznH`)e)rA^^ z&vel^6UmGpvlE#PKdq^94P_28Rev1Z+VUa7?9~Uy(-zK0 zry?I!MT-E$ILGB%r{Oq(c06Ahz!vg5d%x&2k+E8XKne_sc0gM3oQRI03SR{LjaD48 z@+jq{Q|e7-9rl_D>%y{L(cXHlaOpEz00YnYbEUVRZ>PQ@n_8O+~ZJ+yCF>0c!aUW`AR*zWhWCh6W#|I`Z@ zzS5(5yHITFnz?_c3znVktm^iM{5Hs{ zSYoEspE*>5DL_fOrhOY)`m8!Vq1j83?ow69*!!QIEixgdfG?^XkL#X-tRf_dGHRym zHW{?p^c@&`7;(QLSa$)mW_>D<_N%@~v#NUhHw^xDB?(;XmIsMe%5>d>N`{W`ojPYK zPDveN&KUNfK7%qaE=jzhat9Gyw}Zgq7nosbs6A_+NO1kRyEUQCUk2;6@F*e*nvfkoR*Fo)S+6g!{&K zK^U*d*s`VA37?nwVd0lm_)B_wNwM9#dgdhPInL@W*PY{Pb|$$rYV4abR#uav#^h zssX8vt2}pIG!y}e(;Bs>1O9sW$6@LRd)w=4H&UGs-&8e%+z}OR$G)hFeYo>cwOFyx z$KfO;=)po8&$NdvW&faX0*mj!MJDg>A}ZJCCVZi~+ic`&XNsyE68J326i0YU3(mo4 ziM3OOU8AV7AW-%Uk7>d9iu>+a+n8~s)ScLzRK#8DtbdTN*P!w&p;FcA^#y90`>#e| zyCF9*K&};wkbW9c<+Nwn{#dcS@eLbqNyKX&JZA5@8@};^q@=!$_{3paJlO#=4O6`) z(-kbbv#jkuoqMC1P_8wWNROo(shqWsYo{ojH!qoYX{s!aT?Q1b3d~AosW_=AxEGKtQbD~`rmMK-Oj~0{ztqq>xJLWaw++b`SHc_GbzeM7JL<(yd}r z&m3}}l<2iZ?sc?3WzohP%F#FGB5#^0kIDQNS+5h~?WT-;NcR7b_7*^KZC%=MLP&6T zPeO2amq2ib0F7G+7Bsj9cXto&?(QBSxYM{poFa_@a--ui25YQCAOu2dIIHhs?C zd!4n`ene=Zq0#bQUgbtn-q5bN^0&01DjgEqe;9{`Hi~tJ*VfU6_5J~1w!)`UhS)-r7UE1 z2pKZDL7L?+nb`eU1;N-&Tjr1e@D@!41~MNY=dDG!J`pKea_Y&ihm&Y~D6uTxZoT|$ zXP{5-%l!5=xw2KY_OfeO|B2fj-8OrQs+&1@ZC6>wdC7EcIOV=4&Q68Y(z)EOf>Tq z!v<+^VAcM3SaQ9j%EFNBdhc9LYtST0E;h!&jcfV3>J0wtF zl0e+9Pv(61-^Ce}!I}vdv?cFtmYA}Z{ zSUE?tw$7}^nFSO>AyZq&UFB!-15XHCa};TU6vP+zN@$hn(P1Tb=zy=Yb5a1Hd2Sg{ zfxE8uX0R{!D;J$&D*{rOimioF5&J+cEHY^HXB&JzF@?HuIJBU~$z2WL@eqg^RQ)xN zG7SIG?~wcNRI^1^gal?*cxOznFstKyN@20xwm`hDAiSW~$Q@3R`G>HJPA1qQqc?)g zcHW~w9&CclYb1;Z$W${naoppbT&Czu?elQm`KqRc;4Sy)yvdh~rBr*$n#?=LDQSj% zD1!P`trE?-C~9UQ8*qG-Sc#k&#r^g9Us?dTY8r9#DD64&enaT}A2Hoc%+W7HkNFBA zKMd?9Q$nUjaIqAl#znu5dx)aW;~!4wZZser#)k@?C?3v0xtCA)?r>$0LbuRm{mf_T zLu=m9BW`J$wZ4bNVh2Hi2GnbgW;>y4{e@1h1NQ++?-qT z(xA1iT98-qV)xd1t!eH(zOricUnKplM;<8Wv1-xskp;C5Na6}?g@|3!z{i*ccX(=W z2#Fu2cabPi;xx$Mx4+ChfBgmN?xUaF@b^EGp2?_tEb_#Uzg8TfuGnsvI+fg2VZqwa zXIL#3>MsvRpSu$4n{8CR)2_6`!}4Z1L(1;P-#k}7+22Hw$gq$tdE!p{E8JY&pbxjZ zMy#(L=Z@h?r(7AN<%7-mmQLJ8snvY^dsVey=j=ul2|j%UH!Dq3k>Bs1pQKqF*^Wk~+Jibb#y~aJBKTx&i2znLva)PrHcK$fCqQGg@U8j7l zq}z9YSF1s#2KTtgWdzj_#-X+g(TMl}9}=;0OSG&U`gCYBJbxBjG+QHAr?&zfEiy_J zG^`8XU##gttOS`22QxcS#XVhMG3w$6s=iSkzc#x5IasX5=wL3uJGQInvQa`GDi2eUfu&1se9?vXf1OyhI-1-#dU4XF?!j~)Ok+S1fM2t(Gm5$CW?g zf{>MB=+qtSESg871yVoTN52|2{HYY|n!!Cr5vO_FTl~&*U-|Ow3*M2;qNQFJcM zOz9f2Q0(_0sJF2jD1(6g3wt>61?##yfFLV0j}KXlDXb5!vF-hlr{praM}jC;Nr`4r zE&bp*e$qDQhBXy=BfdgbFTJ4THD)eE4EBD~)V-Kdyyy#M0e|Mq;w~TmerZ;-qz?Yw zbb*(_f6s>$-!|N{kuBTphg;g~GeiOOfO1s2V*obQLoqtf z;@6<$Y2$5zyP%lN@&{S8FY?8xb!9-%*ru>rHOX~~r4|BgKIdM@e9<)QlW&|V=}1^< zpL^({wYqxj7bCzgT{VvH*WaBH=v5{Ca`W_Bro>ff9ZhXjG@6ARZt6RV=5Ga|XOR)G z4ucoI{m1l|vA&Ad@M0{avm8`mMDjXr$Wgx4pnS*vQkOUYCg_^|NvZh5BB!x`9xw9e zI|ff1B{Rf4-Lh`pg#ai+c6oztDRVWzv1%HAp+@*4n`_?C4*N|@o2;w9;Rq*7mBGmF z(q0hA1;&BbrcT9bZtlVkt|&CM>WpxeUY=<;%`-JCF3@wCjxtt%gMC=n=(#hH%-SBv zXZ#fkk6&(bU};S6UWJKRW)`=;o#riKDT8j>Y}tA?$fWJ+d3z8;?=PD(fzF}Yy!Xt! zbqAj`=ru}0jX@E|q_H1}H&9G+^*rIv>vLFuyC3V+~U1OMsFJ2OjlvdG1T7QO8O ztCHJ5@^RI<@JLsc#)IL543QNm06gH;D>R!8{2uE`b^MLGL1UI(Vy>rTqHugaL;za8%8g z>oPSZXpumAae$c-9S!sV0< zzyp%%TaQB%_&}uW+R!*3xmC;4^h{(J>=G2OYUn|X#*c5QBk9Gcd~Hy73J5~P4|#8Y zdPoa8_+qocwe0!(Rz2pON3FlB!?$G-rMK~w-3Vggz_00iSDRJVj~8)BvWN&s2QbXc zX~BXI43vWGng-s#gZGHm+{#4!_zt%$(#;hJjnpWzo{T;<1HqbO=vYWNh=~r&V9Ee#0>ftoQ1=Dg<>G0>h2+` zW;T_HcH@-akudexNE@5LsP(aOIhTI5`0^vht*t>*ZPcJv8t4ikAu%_6NCN$N#$snX zhRn9`GSzaWbsSm^H}W&PXOVhczT(Waf750c?chFqob|W!xK-qgz$aHyzk6i})d7PV z-3@HZbEyXFR6dG@>BiQ+oj$fhC!67Gf`?NQ@)O6hRF&uU6&<~ORSmNzb<)$w0$>asLdwC|B60dR=eCvUV5Fr~zXvp|89slQE7ZZJz~<7z38 zqqtN+%LHF1Pe@0x0JltZpN+&8NLf7xNoi%_ z`x%~`kS2m#nG^hm7SNA&8BhJpe4g&8T+>hzCME>EW_wwi%{sXSi_Q(T)iUQ&D9Du; z$)Hdmvi3F}2*F9Hnp{r)sXPeh&?E%J+qmQp+@UwLlfyQrRhX@iy38UV3w`uHjk&RJ|o$sQrh2hx4OXn{-|iyHxV{T?*VwD(kMe4O*?G z&Ymp-O6_F-L6&|ikIur@%lNV+u%;qAxszp!KD|l0ouo^!u~*?LdxTsb};Le zG3(XnW4CQ@{SHU%-EWPriRx**kG{P=-ft+PZ!`8Yy$T`W%^i{p{Z8#6QY7f7$YRQf z6ljv`BCm7nw5e?8HTZ2XSARw7EaW`|Uht{R+0=xGvLfJ?y+I>ki1kO{9ZhBbxzQIY z*~dwb5?T!l?oj4$7@^W{dSy`hX`9?;ZkNLK*|Rq*a=a0;;~@wfz9L7ssX?k(+JOeN zv%scp<3*q(&8zU6;~kJd%k7wNTCPT0W?_Ri<}B}2Zg9rs_J`6{H0*xLc6MdiZzS*4 zS>_oIkeB@HEI4(pPB;s%Z)I8xyVex22!Gn?aGQ zzI!{PsnfDs6jpt-nb)DAOP7sjkb#Dv=C6@fBwJ|14=s+6938joiF^g2#01wY;t17p z&N4lc>FyM1zJnk+tyba6jqssQC(FT8%~L3)Y#L?U1zn3KLrl}5*&Fap>f9olAf;@8kAB@NkQgc@ z1OaJ3y4bnA0_#JZEr!q{hV6t``6^Ppx*G`{a-LP$RG+FOG^Zf0~kvE7^Up&#hj!9to1O1d~Bw&R&>Gt-(`vGSNT zxuRXjc8^dLHWQbV^X*L8mF{%TsZgjpt4hDAe-oLXs^qV=y6PB|?eH2FL^DCEu7*9} zj7<(CPH+<~q(=(g2rmhmEKn%XIgpIX`*|fJcG$De(AycAzR?sfBG+s`F?>*YaJS|w zRVNS6U~Aj@DsLg?9*pyP9r)AOX7tMFk!YO*bdiifP-LAIHI@`XmpgHTyOSd z0XpGldI~;n&17Lv)!{-Jxxw9(BCNv~>dSR50FQpTZ=zMC75Q>FW$i_Xiuw^sQ*xMl8sdo?Ik92^@%ZIR9`1$j@{s?8@h|9D=SdCH7 zFj>N=`bY9xw9X$_6GjtvS;=HA-*tQ3Je-auzuI2_ZOAtDTsSPrDxV1eqVqeD%rfrr z6q)=0(3wpw6oPFb{c$q}5e%c(srZ`rB=X-xN{qiG^&rGS?V zJbu&J)sr%L_wz}hNGtK0*6ig{Dk^pe49w|AdKme?dH|`X5=IBdlBG{eq4Cb4SkcOQ z_L>R8JSX#KSm#zEO?pVkFAG##v#wvke#++y%5_!YJ{)e=a;!a!Zy!Gl7mE_%@r)m(X-$L0l+V{OPo4Uig!Ve0#oBK>2b@VQblUB01&s%8 zCE4L_0txQ~!ifccbH0O=Y15UZdsQ9FRJs1bkxpOqG=^Ck?A|IoH~f?twMeGB^w%CG zB{(og7aBlJ2NYuor)Mi%!(6S?rtjX@Aq!C<4|z~5`bKStg(_;Ya9UCI*>B=eyTxs| zaA#CV2hKIld>3)=YPWWf6Kb$3ujr4l_3+>pnli#s|4>&@Iyt!e3SpT95d8WVDC{rz zuN4Ay$xf!nlYg!7?HN>#6_NfTY#~|v?qtxm=r-^|E_Iqql{auHLdUH#8b+pHoXAxHONhcANuEl3<44)qTsRQ*mFl=B&p z>T9By1hZ!}r6T5dI^Ck{O};|I;23cUyU4p&Fcv%q?Y7$WvNYdvJEe2uh4)?Wuha9# z0PVnp68O%usZlMt$HG`eWW>M+=%Kap{vbXV_Ky|-Q_u7LYoh!>L?*kdt$S%HLF(%t zGjhn>!Wpz$yq%#NBs7p@GKgF!O8>=2VEe|?@>9_0+bo^%b$8C~*~Xe4)|0TJsqWFL zQPVk1&_p^7Nh_-PLs494T0XmOYyb~uyJq0wykAn0&1Y;eTWDB>Tuxv1e#OxNvBe!W zfe}#m&{}m3-VobwzyC$W46q+a^a>zsrE(A<8%1%|CPRA{b6o4jDGNtA2SKuPLX zGminlko?rswFEr^6bHmfNiQ|D?$U9Ho(|QUG7IP(4yU zW$FyypW^sa6bZ=g5M}&R1IsKqYd?7#K0F^?9TEVM&$s2$Lh`?Qd z`mmP0fjBT*VU2+$K1vDIKOZSv_pNp68U3!};@;lc!~k%8XB68c`uTOYlbtywGvoEI z$^pHFST(;ZTW1u=)k>9g#ez#8?ezKY!?t2}nv$K=hHn7al$NRTfJ|}yVSX{*%<=eT z!QAztm13A^S?n7blR}p?)z${Bmfi!wnY&xh_5-4R#2*YPTxvsz^BXH&(wOKTx;SgW zXtD^v0~TPqr3VCxtzHr3+`Cm`<|-4jv|)tC0=xS2pQheT4Z~7x`?@qv&e9=n1Maz5 zJZ=YY@(MV>d3LL+qY~7^@5MI^zXGtKE>e8mynjdm^>degn1byD-7iD{0hce6S&b-p z;%p!9fFtPjDCqcX0v+8t;okESdTfJXp3cn8(N?o{`xT?Z!D6B&04_&KpZEW$H?iuW zC&d(9IB1dRxIL?SIZC%FrcyAP5L*?W8q4tANn~G0$ny*e&>DH2_bdo(S9m${>Rdsk z4bVETth~!m*8y{uXdz6Bta_o=dh(6zXKexN4xpGW(jrG$FC17#HRguaDxR!yIrI}> zHPSZ#tkL|Y{SnfRj0)T)s{xB;t9A&`H=2myO2kqdzn1y?(OXeF-g{Rj4&JegaMwWiJg0C}MnoJDGm*ML3`3LDdT8ksFfoRK)KeZ1`+C0a&6)TKztcb7f~*~yiGiJtgjkj!VMWupbT zw#BQ!%5&>u7dMZb46_%Au|aw}WRj?z?@>5NkY0^Dndv}Kr}+jc2_1GD?Cvk4(nsqc z>2QA7W1v?tpC;6C;o8%yp0ASOllO(xlEWrzBcQG_j)rWK5ru-8;=7(+uH&ovq-nGK zX;cN>CRlB1WH{XpvFQ%QK>Eke9E3s!A!dlrn6ZXRyS~^qO5hQK*a5x!O|&UaxJ8)C zF%t?vZ8!cHfqLuuh)q()cU6dz@N!=$?li*wU{)R3?}=-F!NR!$5+Ru0bq(MIbs?&G zXe0GTcDPWupOQ@w;oA4|vEsAF(TIl*MCNXt`wngOL}>}@Y7WXR_O3&lM;mSFixiGG z2n|5=pzRQC58xk{{Rv-?iEq3UIQ|;*^q;WdWqIw+3pC}DProwKpC3S`Bca)X?TR<) zaTKPRS48aEx-c@U>SdTZPRLg&YU|zcyMLGK#&i+#<8Iq!=beg&?ve$G?dAphPK#rX z!=nmbp3Iji8N78RBh>mH4#Tw?`4eT@kKWr`2vaW@vRr^lBpRnQ@pFc-2J=1r$~8ts$|aHR7;+e_kp44Ov{hv zgW8JX#?@W1^RIf(-s7$Neo$Z1S?qqFfsyI(bZ4{Q+><5MTmw-B*XG7elcc%AoriUU zh5MhHT~MQ)=|5`rP8M&5OzQ%Hy^myk#R|MlT9Ny}wc zY=OK?9+v&Bp?*2IbN#X@DfdUjD$qjAMPa>7MrE?Cvs_90Bc&tG;&wtbnZjE7^C#BE zC7Oxp*n`+~-oOhN@A?>?mvYX%(aB@a{$(_VM-Aj) zctE$AXN(6xQZ!mHvMylmqX6duS%_Dq{I7fnc?URD#HG+C>w&g#*V5pOINH%2Ni-lb z2aq2eF>=Stv|eG%u&^G2!2?QDJIT48wMlTdt9iQX==Nm3BIf|wtmUHPUKYdpGQi^n z;aa^v6S&xHLqPa~4|5k?;VOq4mzmJYBC~_m@p#Qw_if346(9&J?yL!XB$%s{@DeG& z%+08b%2mhi*(#<$G8s|jM`VTnlSAQE^-1*jQJ%a+a?Qh4?n;Bbw>xn(L994?n-aw?g{fy~wA^jSl>miIYY z@_Bb9p4M_#aWaA8N1aJa==?2JT5GNnnIixv(9~EefrPSKjKP#W1vD?oU=^5$bXNwc zLz)&fPXRnekUNhn>*88gPG9cq_!kl^Qu*{JZ(IhK)PQherpF$_Y|RKzXy?}g=ZWsg zN#294q^LKcRf7XcR6Ta}ib=X>tAqr;Rf0aQ0fd8M89#>X8}UFNe3biwCa~aRRYGtW zz9hpiY15Nm4Rswh1Z^+Oz%tb$;WE0&q!#*C<6ixmeQ)jmQ@c>P$!FQ=yh2-+6fN)R zPqVBwnk$FlhiYlFJ%njfLT5pixyZJ=5n~cz9Q8`RRo#^8qYKtCN3I-=-SyEd%g^*Z zj#e^A#DYh1`rlPmJS9k`RJSgM-;b3&1E>9pN;>HVt<&hw$)fwV=I0J2I-KysG5|0) z73yNi1xwchP%4in_bDQ4Lb;R|DV!5eBi2kfTn0}B(;EA3E*Mm+{p#QAN9y{*WA&)I zy`p35Iwzr=NVziRX%gZSvhA8Xfvy7u-Ji7VxZB9#{j`v7xKg1o*nBs_QnY(r!h(kT%;9J9+~&UHB>Vv|!JVgE!lY0!ASd|Ub#E)MgH&fC9maZ%Vz9y48& zkoxp8(M@x_$D`s_-^qBQGvBP#!{j?L(g1fnpP?+%`=WRgfQ*BbpLA+A7ifuFk@_7_ zg^gQn@M@AN_YAJ0%R`5-lTt@#w;^a;U$=a|Z=$S5@#Y=uXs{ac>_ih<6?QljQV30K zCx7h9cCE=r*=k=9sx|Qm@9H8s*3G$9Qs?Y9uh|FY`SXGcG}H*k{Fv@{WX zyc3VN*ju9JMNK{yEkN|qj3cS@?fCa`LGrdPQ#>eZY42fDm1?W}n1u9wkCbLSS;cWs zNp6jLpUamT2XGag0sTvPFk1fQ^zH99wW04Yx9I!(%ruypQ&nplUIjXe7rx z8EC7$r>K8~7|;i>`4|r;;n_@uD?laHt}g9em$VHjZkB)BsleP_5J5D%1b=hpGilb#?qmSlSHMdMhX0J z`tQR1jpf)~OLMI&$}dF*GOBH-+rKmM*dzdl6AJ`E%$3aZ9p9absKbp~8rRD4sQny{ zev3O=>9$ccb5g4+!zx_*d7>L`1tCc)6r~Vn8`+i zhn$*d1L`PWc52VgAEJ~N=`h^=o z;2xlI?S$pw_ior|#@bk1Hcy=_i4C``NIfGgozP->(r{#!ez6aV7lS&BXik#CpAQyI2A-j)hSVg9@y-tgc-@wwdB~Z97M$gg# zvU04s=Z0!eGwslVjHp1mIF3LJOGx9UBf)${L9N=xU4Wb>yk}hFNwbAVX*RBkVdBfj zw566@wei+ME{vC9w)zdn#dv)qnGn)sPUBK`x8`(xfCfC=tqJi|FS$N&&2&{3*Wh5q z(mlGvRaA2*P#`G~irEZwfzPRQl|VGk>kpbiCv|Y^pFZ=sX2#sDhe(I*(#ITD1%^9h z`bBb7hJ{glmAi*$;h;=g_kEjDfKlwly!nzC&hGZCHXf7WdQU)c4x@I)F*Nqp)?;`i z$AJiIgdHZ&@}Z|}+;YZwujs3sEc-wi-rP}=O;U!{QlL%0PGU70GM4C|zSYuC8*YF4 zPf0KJ+mMV>2pvch0f8k1%9M?9ucKEoTKK6+@ljC^mJN9^O5|_MSYY~-6dH~S!C1-# z@?`_U`S3w}uDK9=a2ir=-{e5$w;KHkWq~Z%V{TZ5IUT2T4g_3VnFQ)$!Ed$FQONxS zAUOm}!Gwr`%oT%d17Hyu1Es&CwFMXjk1iVjq|pOcBGO$qva55Tl}I=&qKZy1q-Vu> z6U8;ntbvm3AV!+{9c2QdUfYhSi^n<<6zvhwZ}Y1m(_*u#xev>#So)&#@=cCJq{Rf~ zWEa^m%*TQi*LR~&3I$J$E{+4Et7m8Wl?h#L*% zwt`YOJUTU=ow@b-Rgu?Dc$-!%96Pftx<79EPbIE36i8u-%VhbS1xP(iogQ$jS)Sw( za`oAj9=-m~)!K%C(ezlxyL=vakwL;R4)K~VxaaG&zLHG{iaGN!_(Z@9+us|ePT8L8 z?}R?|(LF-w-g=%Oj)nLE_2oKo+7x6SRVwhUp9g^;bs&5-sVyL~G45^6}CIfmp;M3fovL`WfeaEFoLH}ut8KZsab zlz?#n7n2W0RI(Et>IR`gRDqu6-Sg4V8`kFwD@cWx>$OpXS>V0<=F;LN&1S8)fZLOm z7Z{c6Ss#3Oyd6B_sqC5=GQ&YxU|%6GkJn@I?j`2izhB-mVlQS#!yQ1#4+c#fNp6Lo zZ8e&{Wh}>XH9S?Jl0Cc&@TUU`tC&!+=q+^1k3=U+&Bz%hUEH=sEFH$=Tuwy? zeO%bwOAAFlg^?MyIieqVZ@(f8>sLOf zkZ|hxrc04_kq>>EkSHKBi_c6452u&TGgNS5X!zO_8*m5r{*}i2ZVOHt#Gr8rl`YEi z4EvOPO5UVnk80bKeOFUUO{f1co7)?$3`teTY^u6RR~l2Qa}wtOsrM^-)255Z(}JMj zMvY`mv1VTO*Eq%%9nDT3vZ3`;+Z>N2lgls2?k^CwVl)LVvR7f7XM3-2vURSqe*wa& zlVlOh$*)k=tmFi`F1*Q>b_;ir4cV#Gn?EDeW%pv&WlI=%5W>)q^A1EBm|B2!tAzEp zKM?NbL!?W0=~E6x(m>Zz_)M)$M-G-B@nDp>!)R>0^ifGMy~EKbZT8`cxJFNKnzmNyuUeSe1rm8Rj;hDN?N@U*TnA8#~Kfnue`m> z`3SulEET!avS|`kh-IHZ+}8O&tvM4kM(5$kBr;r@i(2a*91b75GQ`_>Ilh!@8h4le z^1Xo#N?d1d0uY3>)dyvL=930rU{qVN)_U%ucJDQe7%yLb-KMC;OXsbPdHvmlJA#I# z{nm6;+VYNg?nPibAz2vQ+^T--u(-nZ&01x+LS`H8Csac_(IpgK(*Wp1%3C z(dKn>r0ygCBMevC@I1Ue%xa_QIA{49+_jOzWbyRHD4FBYSRstNk%Xyv{aYkaGgI=< zaH{#uUUMN_AaoIP^_iC*;XIqVxUuY9Y&0AwZlm1@@M&fwDOwW`V+LRN7<{;UjTm~9 z6ECxSXtL~{*kwJBk7XFc7^D6z9+5%Yklm-b%^fFYE~7(Ggb||y*;5?! z2}u;`mUN-IFZOA$hqr%={9#8!iC~Y_qfp>|`#Yof4-bGc7>ajV>OB#)+$MzcSbv$}~bFqGic2tG_7-mcRF!q7Q%F-FT(y{OBebqFW2;%Mz`ledq4KL-ym zYj8OYdl zW8(Kz8#aATmZ}hMv{Nnd~$c19OkN+~7vM@31Qpsh8L6)(W<%o0f^6lf9lXj2eil{w-1=_R+%k7(owBhK#UjALUB- z~dP$D)m!ckJ1mTph#;Q1LchZ3KD9|-0-`{=%`l;y z*Is7%l116)x(Iy)C)pqUwD8`<^uLWb*c;1;geI~2U->(nNA}(SP!W{(Y|OL*E0hWH z53>}hG%eTL%9{~($eii=oH@4pocUY?Z%XV3Pv`SjE$bfKhpLJ?dOe1FDoBv+ZMOuZ zFXe4mFo_zTwEPteD4&wqI`*F6aF23GWc)UEzP94i9@~sB-D5sbB36YZ!{S;I>IQ_0 zrMl4!$W;;2x%KgDuIh%}wceI3(t0I97IbagtWG79dE7GUPw)06tR>QO@sM-ImT@s+ z3z!`v@AcBK6|n-1W)c^w!xk7RMK_%~$-JOa753XReDZ}bEhKb-(n3XQk?^12FnJ2yC@hUaa;^GjtB41*Lj;Mn&vf2w>M`}-NFsFvlPY&qtlz^@ zhJQL=%4-_8I;oH%{|z*HcJ2hT+ILs8YTs$pe8B~&q7AI`d@t!>zH7PDC6m`!@sl!# zs92S6+XY+L>#ZNO(eQU@pU{s;!F&Mh#BIbdf@(!Orfp<_wexYk2@3~G@#bYv>RAk~ zq)lLA^*t$lnV4|pSMTkbeaSUDa&$>kyCxmD5gC}FbYbO_SeW2S@+Kc zxow-W3cDGz{Q3Je$I`S`(+{gNQEnSmU*+)-K5T=>nG#~${#;;GD+2zCdSmxRNh}wt zE(t;uFwozx&Uo=PJ!1s-X&J0s_q-HV(?O({i6f~G!?Fy#aGCG@$r>k_&s?Q`z)Jpr z#q@r0pZn>@FM8nppV!-94E!6rR{!sy*b40;k>8V#l};Ukdgb4Wey%&84^!wLNP0iq zDB6?ynhUqH{22c5XPCFrFm6m?p~H6C*c4q`c%zq7#B@|AZ4@`X+a7tl{owJ6OTx}W zs+X9JX$YU%DId4p>*)FZ^=IiO)r|Vc^>GWVkea^P7zv+~%w~WCg>{Cu-3UX& zV3%*8@p_DLHM9+lBUGEGMrLJkCPn~H_w8SU2ybBNXU5(BN+$njOl?FMwy&e{zJU>c zWJqfBkrScXbSgIu_ulPUiU@JW7c zT^2Y+ z`yUhV0)ZR7GZI^cDl&d-1v;F9CYS)Y)gB53{&XF>z>`2pIIQSc-DiLN=bm@syu6kw zKV@P4*Vo5|^?W?$BH-Pi5V%{wZj>)GpRAR7^&Q{F?PQs!iO++VM)$Johk^6PFr^J( zDIxWJmqal#Q&@S1yAf={&|j;{4;Gb-ht_E|S`!hcRl%z1ZU^3o&DoBkv+6vr zsbHOw`?rQ9!8JNeISHPvj_ULSJ{O4MFfpCj18NH#F%wy@}-g?dE;xVDT zP^!D3 zUgT}jjg(VflV}5ZBN^ASjp*kFz^OKSn0r_$8gsE$XN6*z;HejQlz`O?goAEx%UG zy=`cV0T)R&H}q91bJgKoH-EdCS@!QO@n@5CZqME%pYvvQ1i~spvSec>|JRW|a~_Usv_-f9ikuKAS>#-KBO5Kl1-NP&qmbc81TqxIi+ z@PGC11m@3EeoebDiN7Y{e>KG46A0WV#PNzKqS28L|J6z|;Tt zlKy}ApZ~wnz3Kd3`0uI@hVK~vS*rh+mvAjCHGgCR@n-b@Wt2br1xMsZ_XXXS_yz0V zkC6Z2+W9Yj?X4!T3M6DA&i;$f{-0Lq8saM; z_l!4yOXt=EAxEwg;CBsx@x_S2a(+_<3Nj`2mU;4z(DMc8l~qiL#&X$Gy_9c%_;!&= zMl>Mr*-uYKDf@R`;tp@qt$f~;IJ7PeBJ%}{og<6|nn=B+!H`59x3Fsr$o)(O!o@s| z5pEpcRVWxAq{=K$l+*hbUt=3chwO|~TMjuNAT#`E1S83Csq5MkGP(8!y0-8+*W4@&Hw9$fx>RP8Yk5wy{Q z^3WwgHUtQulM6jv#?pfte_PtI1b2MdSLX5D@N<9eLDO6&iFUba1cQ9w{7Y-d%z?5XvNW;5ZOY ze0l&Lw)X9fXmDtx*HVU$yQ#T=UXk)$m8;#_+Plo}>d-Ere3HIejFgFvK|?z6%4NjA zk({!_5^Uo4wZi}Kb$QCfvp$0h9<9wj>tH&eY^ki3iR<0sI=MtyHQ)_Rretvc?ePtb zQgij^A~ii!5?)MI!bH}XCC**tgh%N?zF80MrGs3((EM?CxpO?)yf9SzD6d!e#@wz) zjh{x>3C$=eQd0$sns|8fglcPa?gOD!PIjKH7ne&hRdRoHy~@3dRiNaC_)8S>)5g&O*09zIMCBd2 zq_nETq4ddb{+>%#ipu*E>o0x7PM0%sN!)Uabyn}VR-Hvz+ZJy}T#Xh|&lg+8Vt7>E z6lvyd7`v5`ON6V${<@axf!5^#@Np%b0XVYq_~@L0k-9l;at90bNZN4I_vwI;@%~~? zM+EOl-a0^brKC$vuGvL^&9sqO16{X@>eAe}DE-v< zH~MF-f{EH0A1j)AY1;sESB>$dbBgcVV|>|OLRU;ig}nZ50+6e|ztWw@W{$@ks{PGl zK1Z!(=>%$8F0i{w?{+$iMR`h^r|h%0SOeqnd-+Xeahs(b?cr)Y1ebwkG!w1#S_=$M zFJeA}kTgJjeN}z4DV(T7vEv#0hUPL-=;=e`j4AckNeoW~-;#65hue2)?CzMY-Ds`& zZdALwv`v=;x?!oUlADNL%g%TI&DKn$@h)MSvZFGuGFUm@S1bE+Mr1t*$OF~w@n0P` z&{fTb+o@>?cpauY< z;LJ&3wN=-6#QyKd9s~ap-VuraG^$>6uU- z)^=KPhvW?pS@H*exhu&tQoPSEsty>$7>n{cLtlKM_`n{>vqr!i4a%7(FuUsvxo zfbrELb!qh{Cu6L}{d^*hA4JwcU|lozVj(rL1``^U{HPGLL>aq8GIM^6;!!C(yj~c8 zxgVAe^Mx_rgo6eTr;GDP5;|y|9~^$irq5zz`J>e-RKV#8}|@pp>{#zho8LGMO68v(W3XC8!l^{+c+?jkW##8a z>Umw3N3+In5>YL+q!(+=dO>UrdJD~0M3)93TGsKv5zS!TT>tZOD7ZEZDLw#UfC0hp zJ$muDB;w|@;#}v&f3vQEk51r$oa&8`du6hKQNY$PpasM(c#Tb`{K?e;!q|3yr7~hp z77rATh@Zggyg_5{O0AX)qo)hiVXV&l~Qx!&n3^fzwd-`D6M;o_-_H7s+;jy-T;T|l@=$uu62YaT@QEJOu@FV z=FjM6uiGn)@CziPMF29T`u!gh)br5{0bVdLxBG`}Y`=zH0#QE3dZssIb0E?pW1+?b zZ4lf+&rkVRs&X3eLTq={<_qmA`VUoWUK1Gnxs_<%)A=zTTo0+WN%Y43{#5&ZbX~oo z6Fcki`5o;dj>I{SCUnWflky*80&XWjbQF3PNZ4DZv8iNF>z;~)g~8Cz#0Wxu{O=OL zwc)owv4DSbM`-+RTF>4pQtfl_cttzZ-23%+5FMJjGwQnQxYf$mM6v2s_3|})&Pt@J zMh-*dlJiIcQ{pzAx_R_v%xfg~7`ENFml+%$&0+W)(rRV8We{HkP`fq{0jp`0;KOaK zLahT-=;^j6ZFUK;;hmLPM*No+z!17|*(%77mFnFK$V5wnAo1^p3P9^d1C^NH^e)+k ztU`zG)|XB_KMa|K9|TE#=CoX;2l@t+hL)Qq^+3xh8t`I1)#nZ+=(~$mYA_&VOe1jU6 zJUlHyVrW#vhmy%t$1K24ug<#yGY6liiz$ssY2V$f#tsx90~$0jb(y7!WL^(aCrwlh z*bEOpmL1dyGnu_nJ$XIE>phx8BZzq@8^8(DcDrLVF4LJ_&%U>RI!!kwgK#HlDg2jb zi{qDqi;IEMICJ;6-XrlECb8AUPAg1Q92N70 zQCZtPYGLqay%DAmseqO#nXG#x$rSeWiU^(IJAUY3AO(%b!}@QJN5YCxBm zjB$AJdjPW2h%XwpeXXSB#yaWA(*cnO20js1_GXa7b?>6b48?q!{pkWAmcw6 z`179C$8*3>zA)JLZhEV2nb+(4=wQjuVkF0Ps$mzvQIn%zk_HVsB(pQfb|P`fY3QV) zBlz}dsRC}czhk*w#Vsq{rdYI+_5Lnklo$Cx8t9B1NrCP&Y3TBvpDrq_WX@Mo)$eo4 z%d|R_AmCW8NY98d8l{C<%cafnXA+^8=WAU(YD+(M-*yy_cM)fd0_f#JmUW?nre8F7 z7hpON$~(*kqg^YW?>i<}t2J|g=u4f(oK9uAX6>G@-N-1H*6xGTGQY2Wvr?U0pz4yJ zQxTvcbWne*#(}}1X|mYYdw`@*;ThWeSKB&PJlfi}kA5Cj`Mz%5<<{^pvDyjpaz^37 zL;x*X&^P*0>iXf;F_FyfS0ZZShw%(mEJM|n+cH=2GxzLt*HiC%0ke|jbB=9sfyaYG zf$HybbBif_(8GI3Xy(zk?)ioS`NzVZ9wv}hCA=Fy=pMRe(EcK=F)Esio3i&LVa3?tIrV4K@K^7A42XnDuyDRG3TrrkMLIs>hQoJyMsfH zB`T|X(eB3&;vL>;YNu0{C#|BOGtmIlq>IBkiNFgkUU2b@KN9yG$xIR>h5-`A3hP2*&q zE_LW9?)|zxG+lQd8~A3}Es}tfFlTsAV%PCK8an~Ue$BC;bde>!=9!)QX)C;fpodrp zRr=!Zzxbvsb3~J)a2Bh>LsUXbGU{VtZPbf3?O%O-F9~jvt}n-gFc^Cj z6a$=OHHx%b>R0k5_r<@9UULG{wG@?A8J5tx5dDqKTu5~Lha~Fz9UNN?1 z(iidO8obyJZZSK%4SC59%TL5Zh2Vb>!+v^e&XR;PeaD8NLPxO)Oc`-@x{JVIu3%kV zJMe-$B-5|HL13mV4Dy!m?sGENCFcqxm31C+I0Z~RzKbV2zoMge5HpCG0auix6Df;P zlYLFONdp54b?$=3- z&>7>1W9?=Ek8nAe=i3weZJ={7TfyA@0_y%MO60>*y{LQoO2PH1b#?2&QhPeT{f}Ya zV^=1F)#mf}@Zu_n`EW`ppi?`EU!S_E=Q&U&KMZ=bsDFXuOh&%p)|4~R&h1x$h$)UO z)>%@PXIPp+v6{%Sh-GK@CU$w58f$1!C(tS|@P&qdRa-pB(y7)Yy|x)$M!rbdM4z|? zCN z9nD$jQ_t@XrUH2C*#LDo+PA+G1g{hkt{hg12x*?C?GyaF{RGnTE$lVz3hJ2;Y4e+q zH)5oIo|7N_s|HV$i{gX4Tn@7q{rhKhZ-LqTG{(u>$hsOUP2flpqc` zd+j$}>sgz2-X4!&mYbB1O6^?7EyYs$=LA8Kk6`4J_zlq)R+NS39NMz%`&kt)nH9DD zB<1=JrtD+3u8l<&#&_YNub#g+3{QjYIrQVp+hSFiuIv)eY#+1D$2X`mI<&fhmG8OT zAB;RVdT$=gX0*fcQO8yUx~t3ht%Wpca{Fbc-*xPQ?xY`&fFMn}kap2!(H++i(G8-GCl;i|w{(V2&HH)`ab6u0=x8U;`Ya#plYQ`zMH%D&mYRM>)p z3%w)4;vv;t0iIL&F2#7yr_RPKy8Bp6*0VNhcD40Es3?>iee@6b&i7Led|Y6rK3QG< zhS&;0nD!PS{=*X+3H0d7RA!`;()7N0fiqRXOetKG zRnC`Wr}fKr7hF5L3IBOar8q|udoTr^4<&6o-yIxQmXkv)yTd;$RSN!WB4gABJLR$Ip_RfR{StR1H4YCx*Gi@}tz6#q!+yt{Lz6r` zyZt0(j_}0imaTlw1rIH%VLIz2g0P~Uxoq9mM%0H;Eb@eE{7qyE_k5{bF~PxiaBcem z6I1XV6IC3!pi>^6%u0SR%@%WzXshIf@Qs08Gkqy;I(M1A{$D!(n}8-~)1Aru#Q@&2 zw(ib%F!Mv^??m@gK|nv);DwJi6{CJRV}^p?&TGvYhwS!*kM4ci$-1>BVeh*&4>HyR zaeEHJXTVxK_8Adh-aR&`dLhEEr!>p&uT)CY9y~9Tp584Zxo0`i8Eu(U3F?OT%-;27 zd@{1kZy&lec+1KrNA;`3^pZgmemDIRb}jMFs5KP8A{R;GuJ3b%?b)Aqad3s#(J1ke ze&Se&lCYcInPl!&iDTiCA;ml3Tx}5f0Sn^K(5$g2zFTA?5#gl2CyzV-tBs&d%5{FE z?X!IPc%&%fg7vAoIw%x-uraFnWvb<20K{V+&}^W2YHc)(6TD^bK+4svd6Kw(FUe^! ze1JUzkZ6M0^Gj*g)h2CyCcgfE%BLnQ$zCwOg8I1vfh)V5{34x?X9X;tyr3S}2{0m>Q#nxyY9J06W^#KH;t<=H#C7rWk>8&VyF2G-fr zKnKhT?;Gf!3o??x_)*6qN^cy13t?kj1ERRW%g!20;^F(VXw1^nB#mDRnis*Yk{S`A zjTl%6oN0ld`y5CaUEV9S#Ji)8$nhd*Ub|mnij05fvlH7*str7BJntjkZcS0VO~bPl zd@zwYOlMrSt_}@1)&J_#41HTfBa#%xvS&R^kJreoLr$HE?N1JRK-p)R>EI#$c$c`hXew zkwLZGINe=(s+a|RMnZxkqiR=IU(SoI^c|z_L;d}ifO}Jh)k$BRe3*in<0WQ)%?YDU zt4@aE*y}R>?(F+6D=xF604XviE!Vga?fU&o{|9gPYZR|m=r668h{Kc=eH`d0oivlb zFL$?%WKDVk9K6wuHW72 z?}VAzP9I1ex`2h35GIQEW7$?MXDY0JL(v}tlD#7y=E^<>)XMtVK%voxE)nrWti_|> zMknO6eawdao#t%EWh=9`UkQ;237|BJ%ch7f1)#Yn$)bja`NtNHPL2)vJ>IieI3@it z72DlYsPYO~;jI?e`|Fzv+GtvCH%=7P3IJCQrcxO0_=Av5g*DeHMEZi1t`0r0d$$Li z;;oQKbF+PtVuGY>lfL-7Yer0M{3bdzZ>%g@dMs8|W_>dS;yr7F3jLft#*@#OMiaZy zuics!Ubx!3F9S9_i7Ctq6Gc2ufO2hnT`M1bVht_|9!j)- zz|-tv=qjfmS)%a6IB_F*e5r;Yv*@gkBlatXN?3TGBk;(m{(iOtNq3B%wI~`>FV_=d zwyp?p1h$%e=di8Q`t-qW>o1ezXK8`5ZiB<^xFpTYu_PwHWriuUQa1=8Lz0#>Uxj?2 z8O$ev1loJNcwgr_dL%Ga@C&5h2HEOk=$r*~4-(|fleP}APiMSE2wxN#R=sfB$iyI|66p4`=8+OcgoezM#0QRwZ@_z?MpcDu zSiyW&?|QEw_t%)QcLmqJu`QplEZ~8E^6rjX_)e3Vj>AzuUUpZofBd z#ljlI&U-&Gil7h-4Oov<7S*q-86m;sa!45~P6M$p*he@4+_Ovg_qr!Qn~Ug#l{ip9wJSTv_m#Oh%T*|I)Z6Jl zxYfbOI=U_}*kE@&=7(Oq$u&q*Z_!f;57%m<`t-K?r2B`VlVxdcIiQV1Lh$eJbEt;O z*_dIX%P0cgMHm*L=Va@r%aG4SDps4{M-jm}c(41@VdpgIFJbJ{+#GhofB=3sScv); zSlsN})4xXY(Q;O+FtVOH1|#9cV!6oYq5Y3!^iYfCmeSQ|6%mK@>0}NeYcnG2T$`fP zzdBj1clUDo)-d)6?-4jUOoUU3y+SG0+y(XraZBCLd@$Ob?5-F{rx+{agNUrRo4iIs z9d{UJ#yzuj`5o3{!mMoNJV$O8wdl zEK1mCb5&9p_p&Ka`P=+F+3YUfz4R9Hzu@w2)(wEFcK;qBdTE)&hdaVE zG!3=*mr@c!`gDx_r=L|ER=z3@W z?}%yNbn#$~t$hELlgU`c{S3-0(`?q<Sa*v;y%ldi~8i3EI;?Cj?1Vgxiki_hRZ41{e$-@ zp#i3)Jd3EEegQE7jzlC;-q3Sv7tOf%bwkk?7BWd^Vrwd8e*z~v?A=Ml!@_HOHNg%W z9b?=!YNd9vGZMfJJ7$(C)f&T%S{CMFX?k)*_mg==_oDIuC8zF%-K2{1+Lh}^I}Hi# z=UfjPHim3O31&)I8gdzT7Z!ACwN<7&oYo1$B`o`lz#N_z+g;cBy~fcSN?OHu3^|yP z(pKf#HHfdUJ3g5Fe#J9X7jMv`4n z%^)@kL!b(E`v*XOH`LS&rjnZGU*@N(o!HpyS=cNA?VqgyX25&7)Q>TEz2SepWh3sK zi(l7%7RP@z+g&{@nCA8T2WiV@d0)!qgdI?JN>iln(5>2TkOnB5;D|>+c4Q3K%cSdg zJP+Loy-QzC&h$$#&#-%Y_0SDl50Pt!uF)~I=;wdmDHEl-tAkVr?uv%JtwWxKU0WRh zD-8~aUyLomu2Zjg%dl`5Le0-+l4I+T^mp%qzRViR zEq&Sf5o&rA-c99PfT>n(kWMt`@9Y3L?hR!*+WuY3VXFGi!qb5L!c+OeyqUz~GL>G> zv9G8sLaC$3^VOGxwdb;(E0N|{4JW@0;oXdP-sh{0^fQJJOQMY3->DaBjq%>yWnODQ zrS?RhgeoAQA_4(njr~2~B)9CKjyybk=+!U@rYW1tCH?extvIlWt62dJsBEixyt}Mq z)V?aIe`|ebC*v&O9Cj~8D)@ngDiun-I~qPgO20z$XZ$)K`QjQ5^K-pzA9w3 zT9-kWOal$--l?(WYh3qIAoYe6<5+7c>ipFCVSk-y%e-O7Rk}fC&STZ0W4D7#F1r8d z{vB+vy{-A6$O`XkQ5c*2z_*klIuD_)#(AEr{#*%ArTVcVzQ;~aSZuT1M*0*_SppjX zTl`g=Hes+9qB}BCD?R*Zy(rn?C+laLrkDQg8VTzZaa{$dcLg7r)U5)d1)=30g5FK# zx0+dyHEhU8ZKNjCnP1#_O1#-K+Fjd@+jHsGG4nsm6@XXy?QsS&xeZ>}g!$G}Z-JCY~magZcQylb`o;;>0a9T8a8@gfJl((5crV)rT7fbK;s!|D{6YZ{B zJPFLUTHurZOy3JM#0@|C*MujouH}GTs64a|3po$_jrb%9HzFaSZ?~wY^^9v|kZoc# z0;XC6rU^LT`N?ZfdBDEB_DZagc8I`bPA3i8F; z`+S*r2Oj*zLqRIkyJHcdo33x4yl(9aHCQmc*-7T1K z)|Y~DspYH6YYs{kL``Y{!UhlNv%V0d*Ko2}l%XH{mKp(RUVfDM*|Ohyj6#e<6Q+8X zIx7A~oT**rMfdUoCYjAS<6`(y&V_4`M-}tS+ zp&=FRXZUkya%Lm(J91R%4Ml9@9V><(PAMPVsU^t}G9*fS6sUPUFC^5D6#{xl1-hD0 z9>%h2@a);V)X?fHoBD3A>&&4Zj>-t-H7l0sFn(i?+zv~(_J4-kHa4xpUf83xx29qo@)39?>7|R`20ZaOxV-n z>zZVOPE`am#E!{=k-~7hTUep?FE7=BpH#6O4WPbls#<=B(=x5xhK;)hm(M;OipLX! z%3sjN(sTd}UOIm89G{lU4g#jZXkbB~mf+tZ4AbvLH?8}M>xgOZ?NhtJu*ju9h^X?M;$^9XR)D!90~);qJaqEW5HDOF)S_N1u8#>-3_Lx+t8xv`F_A$|&4mlVL;kwS(m;oQH<5Qj8%rXh}d`LH+!immgor*Ro zVQd7O=JKQ}SyP~1kGGidH7}Ey*_2EqHBIg#<8|q$+ zM&OuuDO8BvI+2H8%7c^KQCJooO~xS(6pK*ISk3(Wu392F#AO>7U*s|5r?(p!ja8_S zIXht1VsI^}NOofZ&_V=Yz8G2(wW&bf_h%CAZYbdgg3a!8YAwDhzCNtgp6=Xe?!`$_Mc~R*-(N&L`-Cc1 z`+(n0G0ZOVi%!LSAl#Um{?Mxsp#XmXtwpg-CfYYAK`n93uft-8FRj+Dg(;p26N(+} z%W52QwMb0x{FqsX8OL1o3a#NJ0NL7`uX^#(NapSXN1i@lEIPKyWci&c@#^_B%{X_SOz53i3}ZWdlwArn&N0=h zT@phEA?4@95{dKRpKUlrj~bs5Pkvn!=XqE9N8yJZId-jZm*wmnW-81KzP@iOGuA=r zad5z)Om!PWU$1iBJ5%EIII9BTpLndp$sw>jlM1#)&>~GWE!J6ux}z5-fBxmU(AR6_OFfxuMUVx8H1T^30J@rl*YKN=A^NS;Nsu3)1U$ z1HmkVhnWR}l^0g|!NOR!4}pZRG;(IykJs23?{OYZO{pW!J2-xOGj=7rYu@R(H#GwI zZ*jaA*Pls@Sn!12($MKYxapneU-XNL6Ss)J1(7cQauXoCL;uYa?xDe{EReqQVL2pG zxWCDDIi3(5%@8aFHGhebKi*1Qmj6!Z3K$D?7tY!RukG9EVhd!IhAv^~+hI?-&KdxE*jB17*DHix{d+^>J~d0uK=ypR>5L+$b?Ny1bMGp-((+s{^C8f5 zJcbbIp*K!=?}zjba)MD{Q=Bvr*|4tnF|hjf@s3*Ya~D+u-lR{~$>p2605oc9J)pRw zv*v1hxtQFqW^y1msG&1W8y8<}<}i*G&1hfHyM5}N)me3TdEgHDKohy`Ts*gZYh9o4 z`AD9>@a_ir3E;*H6Eey{(tdSLWfCO$oF-r`c ziZ0wqVB}5}?Joz5^k;RxH{ClUB<=3J1XgCv0!rGWDn`?3GeaaODvxCn*XT_D{fwPl z>UoJ*<-Lkp2Q+)SjY3R9%`89KW&h+SkoCIVKI!Ek32!!L`9S@vT<t9dJ7i$n%rY>ygJR=*E4J08>Tauy9uH_8!Jfn7i3`> zv{t}|7l5+Y{L0MUKTd)A&2|cfjZ+Qq>3r~-WwGUpq5BQ1o(c*`Fh~peRSQMg28#1P z#BX;;#N3Ob21s>8>e|a3+xNNPXlTyz#NXqwg#P~|IRL*;3=v+}KV*^f{OMECad#2! zqDqjaK+h(8Zq#cc+~NDF{I%{N25?{YqT}tQKXqRPt84m@=EjDu13`N$X`)1ZjM1rl znZj~BQeLO+oHK!MGo_M49=~dA(4TJ#J>CJi15%7P_w&K5#mf2UZ@5yPH97AlD7_cA zFoY*AH4KgFwoFfUrN$xm2L^cYnas0N)!xW(e5VHYJ(WKA=ZQa?6&^vNKz$KmLGgtb zF^o*4*V&saVQc-QoqzSh{o}qMkoOvB707M<==QVh-leVG7uE5c+j`W;*72^%zbzk# z#NpCBE`Qb2vH-clNEQHN;O=%;{{ld-VD?8rbOHIZ{Sjm(pF41A$UcaC&|#mSPgjMK zbR}!>f5YFfYupHlnWeFJp6It6hwO^#D3vr@p{3!^(Z8O5#mz{CtrtXN(To`w zQZ%vvRX!^yq84^RyEr+OTKE1@T9xu6NV9n(*DanQ#!BwPb-;oL4TTp~P<|d1Hghb~ zbcpXvL9TCK7nK@0S$GY+aK9G}QJ8w15BI`3xRtj~t+mgLJB7f8IU=ml3(GG$AR6SU z7ls0!5(i7hDY`}8Ux%kkq_}Z1Qq4d7L`b&&l%`ll=_OY#c2YQYSN1ZTX9qw0A_a!D zak9fx@3^^LN3KS?S4vIH;%nTbMK$4Wv>nO4nnh!^=DTF;HCgEgM3(^^zPm%mKkC@{ z4Y%<}Zd;R&r{{X{{3)%7)pIftI9OlX%WHL9h08CDygtqK8#59YPWhO@~1;k&w2ZeN-YYl@h`EbMt9f16b zTl`1H=&@v8sc>p;n>!IM`vq$9%6nOZeC(Vkd0jZ|#7*XfN3Iv{uF8G&U_eK=L{!Pc z7RIee{`~`rb=~!{;F>QB=OKu)bWM-Z(+Q;^hE%HWhXb$6%{TeCvl#Z+lsYc-fh!kn zQ=T+JsLbmTSuEffnDfOnUEv22n7{2-!B|l{44bJZu&V0IfYD$I*v`LhQ+wKkyAdT^ zeOyFYr$gJEFtvNqH9G)`EPJt6z4d&Ist@fOK0_P}6977?fB^K7uhzPlvAbWJpGBoP z$RM7$vBp$p;3iDP6u4Kxh!9*NI;*;#F=kBh;Y5NkP-V?m>Bc<6InMggU-VOEsPLGl@_y`@P4rtf$uK)2I z{o2_)44jE^)K)gijSiGg>}~~O(w|7UZP2dXmA#vAfn84xcEc{=5V-Qhxrn{F9Hnp4 zCCdjvaLulAY_-6+jMPR4zsRQgirW_R;GLSKWGa_i2gZWBze6+iw`y#%@242umd5u} z--0-OxqIPgX@VJoBN}mY0&pV4cA&V0Qw>wlEvmPK5ppPN1z+*TipjV7iBGb)?d58< zC-q0O#st`DVc>HQ>*8kIXles<`QUOdK1nkaj}B>x|9p}1fG<)4|72_T#vto|dB#-J z0DN~+1m(|znZjrvz%U#vF(b2Wdo%krplQl{^p8IxN*Tp1uqtXX+HF_J^tv~LdXkMm zRW3y;zpOw&{Qv^sz7vUh6|%`+$*T?Smz)j|Qxhx1&Nroz>SV+2ePaNaT?TeFs%FD& zw^|lV_J`Q9)$PW^6ClYft{+G@XMG=)PUq=Qwh@`BB-i-FyAQ--?e(Ds84#dt^{ z4z>v1R>d!-(g&bTgZLC_Db|YaRx*7BcJ6LC?qM{ag3V{(lXnAsi(Z10bZXn5{OeTxcCR^(Ojk79FKl&}TU7ZkM^nQ| zPH7mM`v(^?k2S70s_9$&n)TMqAFuAC0iTNX9qup~uIob_!RmHiw{*xc#yYuzpF=8< z)Z+%3d8G(~*LmwEAQFo0_)iL!t)>q)cM%hf{|}cFh0sWU)c(NnH=OB})V!+)T7Q{`Lp3Xy$=5x>!tdzAuk^ zFNa(o{2tz&DH39Tdbyb-go?FJ@rlcK{;eOF{f}QYqjii;p|*^*K8c*X7bq(Qpq=d{0OSO2gwuuXaNkTn9QFPUCx6Z!VB8(| z6f7&#U(yADz4C4_97_WZA$52J2y(Z|DhN5f>M}avQ9G$RLC_q7_3}jH@sj(##qG%m zv0?;T_1?m-T^F7R+Z8^ILvLXCcV!1^oEdk=_@nbFcsy-aMH&{GqVx-7rS9#%vdujO z5CF8k&Y|5VIee?cy0S@pP~eO5Zt!K*6{hpTC%mvlmxu+&4?qQzLvNTOY*n%ZjAcD)G-rOze2USF9Z zMSVGD9_$_*Hl$HFSq4ldiWodIIqT*8R+ZQPF_u4LYhYFee9?tv8*)`P3X$iim%z5~Kp30UZ51dZbvE*1hHETtSzt zV1ep@O3f0PzDH~3f_`{vjug8)BgWTK)Ey*Bn(@&zg!js3sbEG$3l(X?v$S3pA2p3t zGuL=-)d>dpe>-n%MKrA1O~P{cU;&?pTr^J^ViYK{8u+Ja8tXDE<`UAYv90$fc&4~W z&<%M%rqp&v8k2LAxg@#Z+3|ID3F@?$QK25sxW^b1f4#w{mWzEN7q~16Ykp?N=>YLw zSN&&(a=Uh5B-@0J!1YFnO!GSTgC+Y!3HsaFa3F-b=Ujye0t$t!Xq%Av-MSqF4+=92i zohji%cD%Bl&WeJBX5p1e@-&hA@i9dUXsM&WqNf7F0e|s7EVt7tl1KK)rf#?iOceS0 zkT+Ofs&!tkw^JS;MEA}qWd@WYHjFti9?AoRJi=&t%MF%2i%`apr`3f@nU2xb2CcV~ zMQK|JO5hcL>2!d zU~3n)o$D!oDbK@+#0b4E7|}LKX(r%7Jv|qA;iQ*HSo&Vj$1vc{B*<%cA*uHTr)QI7 zSm^Z{aCt_2%_0R$1<|B9TGiVmBja;zAk>ZVyPYH`2*Cx`>%5<6jL5IZSNt3QOO$*e za%bl6z79wdx41i$|K;&)^i*2FM$3UMx6VQ3bu!anKi&blhv{$CHZ0R^Xr^o!RKTBpx57}$2mel^TXk>l1!G1;mK>>6g1jkH9c>1qCP$EFr z67|I~BMI?F(*zZ?8V;%95c2V3vY^l>HIMq4AVX$R^xrB2Z@6uPW*4jT8}Ta)Yj@|r z%;TG8sM)PS00Eu`i2nytq! zllHm&q1&Fy7LC*A6t^Pwpv?;)Pc6BGCCJzby&Qwo2dz`vqOw@2MHABcR`19#W%aj( z20MI{%p?|2rZD?R&@R{N&@lFLFCXs2daztxIh@2fhgbC#F#nLdlD%O4s0uk>qciM+ReW#D%fWS#*M(&fhtEZ|)3%^7^ByUZHn_7_ZGo(l_m~P#fhd!Ur_6_{w`g46q%m?yq`nUFs~pLAhXWN=e6WZN zQG*A3q*P92x2;dnVUsfVhYV@NM{+4tZW)RU|1lifH<@^mH4O7neKf8NQ?xb1exOR?vC@;TFlMdbs$isWuC z7)=EDoX*bKtFh1ft!E8;+WKRXeAJ-{n|Td@LpW|1dS`X!A@wzy;gIGPF|detDEtN6 zO%Bz0SBF58^T5_1x^?3}6pUNB6`w~cWW^QS3l8l5PU{h(U3>6>Mm}xw-Ul$8u6Zx@ zi`r?wkVm=HBM;jshZFd(pY* zxu&P!J|d%%J`^bZLM3H zzR{!!(el?pK*c;wtbVd(TZ6&!ci(>5k3IAxu2$;NB^r$=heB%~6h7L&$JsCz2!ci; zhbNS{RppxQ!xkO{nHY|Kn7#Ub%}_FCGgq8A$`26#7Z;5`p8xt?*O$|}rT4bN)0iq! zqLe#khhcTL)GCW&kDRsMZmK;sgU3rG%o<}gQ^?>=>?Rub*7U1}Rr+A`l|mNl8w}&R z*S_zCC3H2<|F%pP4RF9nImiEXjtTEvLf!ut`(R8G2^EcC7*Z``Ci*q;Us`%TF#B8M zbGHm}G?4H*{b{bVe$VU!1gVkyI!VfmB!0RbOcw`+M4M^`>h{NbU>WUGjtQ~EQUi-) zPUS6-fFRy`0nn=gxm2}+5@*p6?8K2lBTxbr$d!IIzy@};@4E6QfhqM*65I3h;KBC> zVoPD>8NOP|1+v9Kcrvv!uBfPyL#Z^;K(sA6j$U)p!?U^fXK$O=RZJSM^TD0rH5tKL zBx0yU+~-FE1;L#rgOTk{rl;deM;alTyM%@&zSsHNyEV-^@yAbTuGgyS+M5V**t0)E zu}PD%ieARYe?dh*|3Ml+bocKTa|FtiL}tHsZPgwK(GH+{@9N8}yEn8&Ysa~* z-oomF{!$T5M$!JRYM84GL*yfZzuwA=k=pz}hZQZO{mP_B)}%`f>=Rv!@9w;)o(1l? zJOog|unGJrr*Gh=9x}U6QJag}#yo&MPaf^^o3c82PvD=Iq5OV3Y_`I3&M1g!6V04?Sd33Zet z9n$yVJ@K?^ViX9W@vBeykjE2Aa0|f;!<^?N72tS0e^H*`_B95Sjq&{{^IAzvg5v}_ zr6|Wgg%S-|((Ah7dZ`VP-rmU#12oIFy`e)TVYs5@e46dcqG&l?T~uOqNn+1Y!b>Ka z-K4)^R@3+V1&U;52VOB~X(TYG;M@Mh?H66?=6LB(P0dYYoffX2X{z}(nnx)V6*=zu za9^e0(vgeWIt%PiHT;TB1vP_iYn*vFiBGTLM;}YE%>;-{&`YPU?e4p8!kNIQ>XX>kA>>^Jfa)vD;-Pp?5TGj%QS)Qlm^v-K<*-kEEh3Qr1H>xK#sG%bwAN4{?bZjiW-e( zI%tWI_j*Edsq`GFBH!1&vDo|HT>xzl^Gl6rhyOQNV+GxOoxm7M@ATLyPn4&NSsg1VYrtaAIS>Fz+GHM#WP>P_RI*|jBg*Q@ zSj8U5-XhqpM5(j@&yum)ld^tPrX{xt?0Pdk^TCkT`cMrrw86%h>R zSAvRG%f;4k`ca%WSjF$=%*L>iFe`pl)KuxUn8{r~zYI#_@l@9LW~LHHA?N!C!Z<-{ zcC#uiPZq6u_Jg1-^(~6yr(Xq;`L3mZ(aO@_V~(;Uaf@PTQSOBV&U@_w+e0;GXK8dF z)PdD}$%4i?tsm`bek=2q;dEWl=m11(KEQK*@g%0gamy(WEjnBEzWn1-XCqi<{5~TL zw+?W zW!Vq%6Pc7adLoEIITaB)DFa~xfK?<)wdFy!(-VY*%Kw+%pBpI&C#tJ6#rrfPedd5Z zvhHk|?RjDb(2q1tu$4$Z!OQr*8^(&1lpkNU4hW6fYMHubTYP+C9XRpElP`K$abUbn z=sPmn&_5bd?&ySr*a4@-v3*5oQ7nT%AC~JGfIw*CQ;7CLHw7}_bkdlDCi51W)fc2p z(_+n4>)$Ut_W^N=(s@_*ha-N$K|CP^I%kgax2uC`Wd`&*`(s*ys~Qr!$u2Q8)WC7^ zu1RJw_CCW4rMXetKzVUm<;IvY9X@?=T1pB)o5AoHeJT^7(<`fzeJdnAY=RS8F*Pff zxgDq1=nGI^IN$F2Is{Y!kEGyph`0;Z#F}XS_kpTT#H&kuyA+6n-EB&3mC+180|C8t zqu_NA4Nz(Br&6x<+7zuReBCp8dcXcE-chTn&EcG#DS_RU$g-2(wbVQXBWTxO2e=Ob zQ`uE3i2?w5%JfBL<{VP=R2QXYr=^2k0 zm#c9sS@q-J7Myz8ttP<)X~Yw}C-`b7`;*m?WC}bxUnNgw|AZF6>zb1S}|eF)3Qf`AI9fuQ4YUxsN^pWa^{y0{_3$)<2jk_Y;XI9JoZW! zu9M?|d%xNQhF4@I9{vN41Q1M~=A}Lzzn1EhOAPV+SXSc+(N$3|Rh1F#c{^s)E;F^EeMF*X{0rHhL83p8W zw_J9M?|`7zoKhmQuj=>Lo~$V*%!fQUl@CB*Uyv6`O^hWtu-6QAy|M}Rb|7z&vld?M zR3+{MU`b5NK=n$w$-vJ6$4&XN3-?s!9+jjZ7qRmIR6LOu$>8C}H4m(<2SOTT^kw3i z>z3o#WryE_0%fo;{TZ=@L#1vdYq-~OUnV(wFN+z0JQt`t9a`wqh29px(h?jchk$}l_p`zA_trBkr;J9?qQ zW7&ZTmY{g}wr`TPb4#Qtc~lKmVm;qauV<9@on59^<((xsIObg&8Qmhx)KW%_gqHkA z1!V92w!_g}f#D*K|3s4MpI+xqd3r2kj4Rc7vp+FJq!^UEWr0!Py!h+P3F6fX4Ad*}K;54>fk7Si0$lj4g*+YrH3xu)C^ z;%kgp-~In0JTBw2Hdi`aufED?F&};PuNVGyM)={jmqAF;g#-jta+Ewy#DaZ+&uTWR zFopZkd|1{jt$7EZ{tK+P*|Cd_c98@aA<~JvPuYs!)(H4C3Udsi6f2ePHfJ_n@pC&S zULPFtaZx`AbWHN-92~mhW#}KUv-$mp$L|0CIQ9=mia24=*=T|mm@p><|5FnEf96tf z*!OYg$TD`q@@ z9~7vU`t;O}%clM3@9h6B>&*G!U>kQ%6#0L@-biE}$eXAJT#|2hJU$d6|9|YgcUV*H zvOX-LC?Ymgl#U80AQ5TOQB(veDotucdRIbk5m69OK#&fhC?F;D7OH?sCln#H5Rnpk z3nY*b@;(0c`<~-__WAAYyT9+x&&!KTueB1^dY+kiX70IXW)^8A84V~E??B6wgRwe4 z6J6>_+qucto~jk)-XIk$P0!~^XG`)+1%`m`r~59f_6@0)0CNm_M1DA;UFtpX%*^Zg z|NJHYx2^l%n;SlUi5vtTfk-LWB2W!Jj*?_MUjob1?K3`Blcn zGg6X|_dI@t_d72bns!3X!vp-4QpS`Pd|6ydNPr!|tBbA~G+@l!^XrHsD|P5G1scr* z8n|cgL7r#7{-`v0_I=y2VZZtwx_!r#{`}FzBwm-D-kAV;%X1Us)RVwf*RFuy+ZKi>lSNaUPxQ~P_mS!Fw|1Wj=ozMMobHMlg z2N4IHzaIb3b}i!sk0;`zlN4*%U*G8$DgIX%_CbKedWwbrCr?hd??en(xSMV@sE7ZV z(0BrgZA^N)??2kL$dCIDA_Q_ryT1IYvj6I3|0+I>OvmJoj=w(tH}>kk<>UWX_5bfw z-G6_9Y2CaB4e5S+bIa(>iZfgX%{eeh?Rw+)`#SN2vM&l(uMM;MQxbWW4|W zmsoeqFFXlZT&o*j5(qmDAxRhD!@)qB_8#Z~Pp6WOpDj`wm3PGn<2?2l$4Ee#Q+wOA zzpJ)GwLCXws#k?8YN{>j*0t{8W5bh1)q zh~mzRWH~RH&Vr}!{o~4P`-dFCcg51t!vKzTR+kKF1=(r zwc)izU6N@7yoSYHk`L*ftR1GPL*8_WFAtMX@=7k`L@Scsjp3?a7GfQK9@W1k;@>L{ z9g}N7?j&>k@qr(qpH|TgYN)lrBl@vyD6luZKQT7)g5L1*bUZFKELFeS&ma;^#F zh898YknyV751q5CRMx1=_OVy|T=@iBoQJD5x>FmDS#@6Ow^GfE?P_u$7#q}Dpk@Vp zYy2mxYvw+;Y11XmUe}k)?GGmTP9hc7@sZrv;%11$aJ5Nyvb4QaDl_9#kCpLK$VcgG z2j<6Ye_N71>jI0B|(R*mA9Nf>{I{ll6L!aN8Hl-N6|?t13H3Bp1;dT z-iCgr+d|CZ4n$W-|DA?WMH}c}xi~|A;Hc{80d9#?b@1pz(=_H(l}%54GFePJ>osu_ z)diE*DlkUmWtasw0!|G$e&7f8?K~1>qOKjC64#^aEj%&)X*oq1XHL&^!!upyVDijjBrpS`R?BwW56Od;l_`_{I_5DDl9_&+#FKJWYI9bHFOowZ6XxK?B zP6R2OEGB(}Tk@}m#lH}rV{!YVzD~D^O8vQ-Jd*pqzD}7%WHqL^Wtw6~+)e0*%v-jP z9E6zEH_0#0WphQ#czknJ##g_$X#3Ge*yM6bZPRwr)Af+Nx#?-0{0MGZ?W#JMZ?eKg z>^3SyrNgiv8@Ku2ti%S7;8utU;$GD?90QZsfE%py12xSeIiy)pmZp`81?_BfR4Fg7 zJIX1=(3El$|26cKlU|l0h-e5z5OgId_X6;2*ewCnWrkluN&H{m{#qn(}d<%;6yIPQUDfr*1+KR=Dr3aS%v zRP5m~p*RTiExVLcmC9BkC73$yhTpx8tHkc!mmlZ6=$h_t(*xC4*^%JiajMzHd@Npe z$A8)5p?p}l%zo6+Wv0_ig-i1BJ7L3GVTSIe$bYdJe|^8t+(4Jt@TQ{Lf9f(_KuA^; zDMmXax@}EM0vrYI(j#QKkKbsy5aQ{qpc!e;8qH z45*-d%ONXi<)Doyh%*-{(x2OppIy?)9h0&j>AT!1>o_9bZ51-Vwyc!Sq${ZZ4ipy; zNu;!V)i7#um3H5G(V0;C$?0y2Ow6}jgUg+}@efhno4FgFYX%;(QseWN!3w!!41U`Y z<0<~>aXWYl|A#zE-qO6AM%&m(~^~ov)48L}wtdHbP zgjYeZxXrf*3X8PNPH53q?Dgr7Tq~MA!VcJK6aIu_b%p4VXaV_}j=PW@_T2Hu;f{kP zb!CpU_#IO4^%8e%&=tQrH}t=g z|2WWi?jHiFOr6srhV!OA%{4S;MJZd~%6wc=+cLX^cWv#4uYP4}>Aez*r{I9WudnCT zO>KoFH99e0gp+XGO9H?a_vKoO%?+iefSUb5=}zj|pL*5V6#q1Ou?3cdUw@^@(8gJA zUzNn={}IU{?X6S4HQ$giP*(_T5V;#ySj4>q90oeCv@XBCwRHC=(CvbB&6+p2MRsx) zy?`lH?KrJrS;iP|&n{v0e8aac_`+Rx0%u#pZ7Tas=cmprsMBmWDpROCrGT6>x4w8+ z!3W5L^xRq>%#;MJ^;9kU;j`j_d#A$XGeX=R9K#GEmt!8^V)Bs>dh^>#aKVC35Fs#C z(^3A1_CJFL02F=bQ@MM4j8HacY~8z5NN1gG8}HjSve|7HpZbPfGSZ;GrG25W7R*y( zWiDOWm=jwc?V0FX-#xNUfG9UF!t^q{YsEAz{OgCPXA@-BAagtLh&x0FgXY7eQ-D0XE~PLTj_A6 zkbc3CnD|GiisAU^HxX#PuYu{xHo|A-f@)M%hwyJ3<;b4>A3YI0D_1bVe?$Zd`;!li zj!cYOkzQFxR)HRmVDgmcFo2Hacb&C2YahRKF>Ps2BUzGS2m_cO%} zMQ)VaW2RHax!bo1_)H4S-g2%}2JIj@?}7-xt5 zgkU~s#c66Ho05oEGf_N;dC=O4viSCj-8dwZCSXs#BJj! z^m$=D*^6m2xKrY~XLh;$XuHATS!?2=!F!7z$rr4sEyd?%E9z3UO8%Y#~9@ zzwgyuI88@u%lVw~$?cC|Q7MCEHQ&8}6%IU#-u5X(D!bG!GlE;*yHoF~M_Sp5ttd;{ z5PHOoJ-N%w|Fm$VRz0<(;lrJ`O-?wJedXd1szXF&bL)!0A(vZuC9*Kgp886?ay7m= ziYcf)eQf}9+;qnt1IFZf^W|CTu+x_uDs%hC{V0}gebespD&dO{8caEBd}Vj%z7^N$ z6zUBl$g30QuX@g2QB76AF2x_h-~7|j`ghLz_lkq})^;k2U=RK{f%4ZVjF z!ji3CWq2XC#iv_zXc{n_Yr<0gknit@eir3aI1JAekh}?&bC~Qn4B9+}%Y~wWxT-8-TxEmlcqhU) z*N@HjM2?PoB~A4ZWTqk7Bh zONJK>N~+{A<)!k}Uh|e)sptP+z1?drigP*Pr$=iYIkI%J`U$jSX5EYiSu$)z!74^P z7AT3M^P^}vDtvvI6*uh1HqHeOmv8x|x`Mib5c@%=NWQyq9bdxY#&~7rhHisF(-xR$ zzPp0=rN=qNUCqj*Nr@5*&Cc!^FeFL<3SWC`Qh2XE|68F?Fo>618M>i#n(_6--)ex|8t!5E?Tc3u0$!Tax9!Whuyb-)EMlJ*;W zamn38c?O1ulyKKHPNs@zDe%{g`+hOM`z?aR;V?=BgZZGB!5G+O5v1`h&RH>L@TlCm zQIExvQ7sVBbh}k?cHIE{sHy0qJ~fzp3OM}mTjQ~#iajF?=oG)DPH&4u)BbOBr8`_? zavo$0%k6KStCKVJ-hbszu%^~a+!vYO_i_w*Y@gT5h#aBY=aly>NcN|ZQ!-|GSnTp< z5nG8%(d+O@s_v*iot%bRPL=lvut_s80w}opGm6;R9Rdk-{mDE?NBrs8)>W5Jj~0?U zYU)f&b^DeA(rG|j(Z$bA2HO|<`Kgzw-gv_lK8v(=l!E4NtK5iM)jN` z+SiUVYz+0(+M27zNg}ZsF%zR@)fn@YCFnPwkpjoOQN`m{fAzNis<(eRjg*8BT@Y>T zIini1tl-&WdOKatE5ld1S6c6#K=YpKjK&TACV;Fr&lP=MoMwW+|Q#$qx{ z!wDk@`f4m2qsBIeOyjRQCK@-mYk&btykA03g$?$jjf#|R0Jg`@HsS8e_38I=L{J)O zG12tWJsD7w)Fy+N-^*p;_kgXu660Svlx7J=X~=oBr8Wn(6)FeanGQXb3B}bPZ;CgQ zDSf_Y@9}*-8biThZ|^V917k#hfcLZP%hJl=h@zS z``!~^!ffwsGnJtCt)qjI#c1sM>Q^o9a2EvA27s2pU5URBPAa)FT|*Tr5ysoz`|J%d ztL=z`bR1{7*b&bgXwexjstI+ubLUmExUPX4BH^lK=qAb(<#LPcxAE3zye!LmZm5A4 z`z=SiX4I*C6-(Rw30k?=Y%XCb6%1&_c^q`3x>|1AqU-?rz|Lym;z*s5p|~^GM+0&%{EjY~CI|bFD{bRf8YV2=kkG)bqX31q=}Fy+WOru>YVo-QF#} z6$0F<3l9T;qklL3#{NZttE|9vJ`d9b{c>&l@3cjINf}6k13WKHtCj9AqVO*P+GBYD zOQ)Y2%)jP0~RJ6}cu-Y=g?1sF(a7>-?J;r0sL9WiR_xyFSR)9Tl+D9t0gmIq-NV|cHuTM7|9f-S8cFuq9LFQDcmO==Q zwR!Px<^_0e13+PRmeFGMZk&iJK07rQaAevinu8r@B=_pbU~=(y;~;TI)d#+V>dp??I-d@RBGrO^ zZsN#hb#B0Xu#EdVz~mpHCWu*A@UDTWx!z0Mv$v7o$7XgZx9~|q)|%`xihkrW;j6km z!8>cgHec|S8kyG2`?58XBzw%>j}^Kj4J&eMg2vsBykue}%4U%@(8kE=#qcxxzO*J=-G5H19z1r*{+8~{&;K6%84`|F9j7EFu<%eqGDT{a30 z3Y73fd`62_j%?l_Eg!G8Xn&`gDB);VII$*>`pye;Srd!~1}-Es2QZw$d#ACfKHV!h-(0{vs|qb@wHy5rg#D>)346?lIRsiSjDk%^txvayFi_|nYur~V zV({o>dHcZ8OtsjP*^hnF%x6)cneJs!Tz=9|FZAT4b6-MFzlkYosQ-4QOGgT$SELMzOmSG+Omt|r#Z-~ z>4s7`2_^|t-|D6?=(7UafkX@&+pxtobcqRYdW_V}61tS@M&Bn%?BxF8QK-%lN< zZl>C2d6S!Nh21Qa{W<5b`0=g#Fe>$FUKYdvfC$%#iQ6e+x~)v7FFs1GS;c&p?$#u% z2kq`>s1ulOk1enH=I|cHpW=fULoIEW`Y_TUqB!L`r#ekX-me-owday`{L}((dv8Ul zq~gtlL7b5{6p-*{kwT+?Q!RJ`7{_FeCwvmQb~P?q*r4QKIt=|Z`B5&zBZ<|Vn6LD< zY`6W(RgR)6y$Dz629?6!IHW$#BD(>9W%|dLFi9Tvq(;Jr8SjOGJb~zJ^|z`s1a&|q zD)a#+gm_irh0&EyPZmD*zPIctY-k72PxCR`_zHt){=Uh~xE+KxeK}Q;q8^)|Mnq^t zz`u;;ot`n>-xn0YaVaqP@LTO{&6jfs5fd4&`4EArfC@*)cEY3)xE*uJGK$bnOx6Zp)U+W8oYvHpFcATe?0U?eX;J{gh3mB+0&| zVQcKOo!5XI^yyOo0&9`B&9q;Wf)0a>n7f)l&-64rN^=bLI${k5_dxxr<;`7mYQGMW@ILl;S&!M>ozeZ{l*_NYR-w^U9_ZfS?Vv_j)G>IY_jQXG0Zx=ljM- zM6|Dy&uYKKin$gHRKXGihrD+{`MZNMVLW_uOmgIfdxPZlM*`qt|j>RkaYsjCUS%o>StNl%$Q>T$*@JuHrK1r zq*4<>L&KZ)6+26qx~|Y>YF0k19N-@6K?RkSx1;pt2aC$b3l)Qp#x-<>9n|a(C6Rx; z3~EftO83NOaYJz?%qDOrgb=fyeva`lwSFFY?F=wN-dZ@%h07HEeH&iA_v*V#l!z-3#@UsV zbt;BS#&P9)tLtB_^`~7&1drv{I~N#0U-V=_-d{WNQe)Z!E!n@1SmVAh6VyGR1%3R) z)VN1*g?x_gCr{Z05%KTd z!8tk{Ng9D=61lq3`39vUJV=Jdl_Si`!Kp19XT`?u!TNoc-)c_?It8yZc($F`o+6-^ zW=6{GR6!MMW(JCtZO^!Q?tQIfan+!m-5O?Nc+&Y`Ip^nP2n{@OH<-IReN#X~ciksR zqO{R*sgm}IZq9aZllKv30@?0v>}3{a2m5|6#Q{_)atpb@wl(;yNY9dH{pApQsFSr| zs)|Q{CeQ&%(QY8yI1 zF|h{Pn6;C@MB(X=WUVpiQ{ z6SQHg!+yi*Ioo-Ae?M3E58=8KzgMbboxjsQEGDnduG#SZ!=xZhCRV|dLU`t>3x0Hri{U- z+}6`LuwXYg(7tEOk)>!r$<;}F!l=18Mo8BsPm@*no>FzTNm8dC{Pn<@YLnCh=#{%y@M)0WMhpHCTYWwX z5?%LewVE}i@w)syIdoW4ew+MLZzt7cA3gtK=w^_o?JTOBAi-f=u6ByObFdv#nOA-o zS%yrf2#=o7j~iG76{%{^?PM9{`KPXk};wV>y0l2PVZ zk|T%Ipr?RV&WvKQEOHp@#yQrkK!e`USrb%)L?qXsk`zVjT z=jTWr*f&roA>mGxrq*rx4DFk!E3bB&e^tMl=Gc%^%jEz|-n@pFJ?np1VlZ@zTyo%V z+vI|0BnyY8x^bJjtIW}M19pyB_0_BJ-JM<;f1LwDHo2GWr>&8-JI@%4>?DY%g!I*n zSRS}NAG|NGViAt<(#d_;%Ozvh_u=3`XNy++Ba&@T&4h^ucKu|BetwZ(*+?nFf}aF( zb$MTL`AD^O|Do?Ap-E{^62)!hJYE_M&SUCFz&f&}-z6R)E@|`s^uP$hDM$T8n}#RkN~=vr z@7iGJveB;FJ(TGMVENj-(wn5aw{u|q4SUm;K6*($cf1vLvKV4M`;>g5BSMkJk@NJq zJcRmpW8r1S3lTorq1<94!11rO6BPWJ5*n$r7jQJJ42Q8`KG`0KP-`{h^ajrB2)d@u z=k6}I{}Z9BEyO9Ka)(QR5Caf|@YX)djt?Nfb3deCV0=8A4W)-4R9mmKQ;FazH#rpl zsZy_+hzX77Spvbp)W%$Y243y$xtGlKsi5@(SJrhSco*ZE_rm*hXUAZ5iG_-#QJ;k$ zPO^YVVd5?rRwJ8H1pxNLQmwe5EiZcwHmO(;Tem4_(;2~KuDfQhtg#3r-|CI5ZHSfi zaDvyEJ3vt-meaiEat@yl0#yDPFe3Q%O%Gg%6Q;-)gkqV1vES?B4~Tk>xV(eo7T)RvF#=jQhEeQjgr*ey7l#rz8Yt!Q5P+)G0<`~lJ`OFL(zqN z-01v3ZV1EF?`KK=f#5|J?cd3EWg(WUkmtkW(+F^21cKtqb2pf<)wjI!hve?jm;Ed- z7dEKYDKo6myW(G|UpSQf+^3*MgO31d*uD+AcM5^-@2-J(&5Opey$yywJ62KW2l9@SQ$#Tt8qnK4))yk(g+KWRyw3FF z_t6oBkLXO029xav{6P25kYjR+9?Qw{!j|ptqIi`VoRzBp>V;c6L(haxT1jE(qH!pU zDWKy-OiE5s%X1s*AKlNlx$i~T#DL*$4@933KQiZ_>m->250LY(K_1>*>TKlp^9W>! zLhJa^p7!uiqFyObj4GVUQ;)-`mU+m9Iva1aeB;sXzf00Q<_tKB2fXq?|2ICG=&5H% zlX3?Qmn!wqhz#AZN$tFc>Cc(xopANk3JoLWuC7G!z9#mR zShlwc$$M|Hf5pb_o*u}>hSLc<;EI(iyvF(g8EAa{V?wD_8{E{h(X^po+I-gS8XY-V z+T~i!xg9z89nTkzy%oIA4q=&B>WwyFt3A91D7%<^gBrcl!p0X38%^aAOXDW~S7WHo zx@`(0GaTQQ!i)9JbFWfKbkrlNR;C;%6OFtU5^YtPlJRP?7WkaKp;U6p>}X-C%_@`@ z_4l3NEWkA7S+;+bvkyBtpl?Y+$Un$=dflIJ*5Yv4~^l42Cq#o^g~kV-3MplU%j z>}QLorv)9@+2SYPPha@BH%{uD}W{5Yl4&F=J%6VVngpBkO6YbVFH6n2sJRo7d z73W{gckFw2$t>$Zot>_SAE-ZAcG-Ffpu+a`Q$kJ7M{MLlf1e5*qRz*Uh%2!SL;F=v zZX}4Aa&B&WeK1$~*}zBMv_ar0av=Y{a#WDs6%W&jRf(2ae?8xn7K}Ud`odk#iKN|u zXEGjRyj_Wsf%FM#{=v0IoUQ zG0b!G@qf-3u#hA1BERJtH}kR4r5SGPq+QpF(G>sP1ncKbR7I+eq{G;mbc`9xM}Lft zt)k~@8Wyvvcm#Q>2}{PkX?BPj=9lkrQP0RULcq1fI&CxJP?1x zRi6U``SlY>!@>aE>8_;UxxOkh<(dNCQIg`;r@YFb56!`4@&|goEV`n;IQ99Yk-Ci7 z% ze~~P(v0mh1YsW`rus9JwOe$sd$M15)K}kZ_;0{xCE=(E0+`pM6(= zO&lTbQ6Qhga{H(&!zJykcUtNnjg`x+)xS*ZYgm%kSom8318?yz9TV&2C=YLvJAF6A zb!C|$F9ik&1{(7(^sYn8y~~+r?a!DtPs&to1EW`ptycYd`0Qu>Odn<#oV?Z&CFy9$ zei1&OU~RFPDB-lnFf|27+v#W4NtJ+I)+_rc_VOY%S~;ZSiA^_*|!E7NzDbl2dzwcZ{{EyNrY}D^Xtml)y^8ELU4uDWotBwe19glKukwTktw(=Q=|8tXw(Ppus<$ z*CpXu*|?A+IEGq!CO1G!rq3Hj1A zWN2F*mSf6cb8VO~6#sB=&#FA=#?$`7RAS2blg_(Boh%KJwWfrxK*x~htI@*_M*WjR9hBt55p|=gb%56+2*&)=Qg_- znU{V-XUJXuyr+4(NI^p{BjpyuwmQEW_hnuX#vNy#g>QCAxzjK>e<5=3LQn462;glM zI+@HlCr4jU{lUfFdZYK;t*MD%t|;CHno|n}z)sB^Srb$R{>br6d(4&5l$e&zyC)5& zO@zKia7uOif71yZ)y#w~mb^pf3;U6pW-x1bx7K}%Xsr| zT~id|H0zNK0s$9j*y*d?`2c3@RzQXyAkuKrOw2j)YRq_FNyWtQ;4v<-jlH3K05`~Buft`sRPE_QL^=@UmR z3U&+53sMZr9c%;fGPm!tym;e~r~5w0*C7`s8}L}-$ql?TY%0?IsNAFG?)tC#qbuq? z-rpk>0Yf#qNvCn}lYb#X{+5suF4uO`73sSRyo8K_{_>0}78%SvXBL(+_}B|q!c3P> zUi#jV!c$EtcGx#;l80x{EPV)=@nMsQGCO~Ji1lQdT@^!9N){c}*D9PIPHyuHxUP(6 zyHk(R5r((Dn3pKx8Xnx7%g*#%`!PuQ)EG$o5(s>mu%LKZx5YD7A`|Gh3Vz<(ef!4Y zz~NyJ9gGuBR>K;q))nPUQ9p`0EjCYrK))Qa?=W@pXk~j9L!Nof9BjLs5lg#f6WNIn zYtb80w5>1I?8hCI!L-)Co_Ia6k{vWQuvQAGFKTpI%4?*8siZ>DPy! z`V>IF?4d@P5pB>;jBgl93-E&u%$0L92W*`y@<;n8)c8l|W9#&ZoQk}F0Td3=wnqHE z3O`BO-}gc7(srq3%%PWuT2rRg?xLP=I}u)I&^A=nW9Ha<74uu2SS}3cweDkiKav5< z|JeOjO0LBaVoYY1rPiB{(H{ z+nc4=I_uGedffQe&=5!2Bb;wjtvh4*k}AJ^UVZ(O2JkO;doAN;1 zqx+iArslFxPb~@A|NbfQLBg56_4@(Q`w}_C;ck*7^4%p9nUdg|7TaeSqgR-vV_K)h zuN0;}N(_oG&lHQTw3|n!wu&<+u!6x_IwxH2L?J%?Y!HUc=ZIrKEU2ai7zbDsVpv1mo zuNVSKvmWfU>*g)q*(UJSDxNydeDg4<_I#@JNzB#1kTKcA))L%K zCKEkj<4^uFk{{30XLm=1@IPBV`BnYk1Prj>*V-()4}QKM0K<2;;h}zfFnm|^Pm-;H zq}{iR;oY&eyw8*#pYD-jN};7u6bX0-E>T*#%HK&tWhg5t&(drM)LpFFFE(_>+n1_+ zalyaKm|;mZ>f4g@nh6JQc=vnLOQx%q^E$n$BZ*S>{2o#^$13^T9DXkJ`PI_x2x&*d zIe@ihe&_!E^vEkKx-Nd(+p0;?2B9JN##srCrB#zL&hK*4U$dmj0(f}h_|iRZ`v@tyJJlMkSv zifYd33hN)c`RZu^Q=G^sY&Nwh{r8`8&tB2X-z{R27aypnLbARJcyei%{P)}crs4UA zfx2)EL|BYdw_VQuuEF}teM;(pgBWqi;Ol=@-+bZ~z%TZ>oJSr0ed+!sNyw80aHN)Z z!jr$xCi~U>sFB(R)#EjTwI1s{H!$$Fu<}%Umt?>E8rz|Dj6(`_2LHC}8Z42>oC6 z=r>gD=Pf`5p5sONI;wtsr2iR_{ra|lxld{DaZn(wYrd6;&mX~-jrOy3gmz=?y(o)&et;A9}~Ja(eIqcs(1nYCj`;>WBB6Q z_Q)h*Y;xGV$@Mo@-l-Hk>8LOLHTgd_`GQlxc>%4quksX&=Hwk7iV^_XwTe8N%Rgrg`jkkQ|G z(POXdfjmx*aGELnvDJ|m_Z{I-G(6IAb0n<-{+sLp}+9L$783{z-zG?ax*W*9+vCd@{PdN?{RVJ$1SfFjN+oLHx1g7 z7Kk?#^+{t)NtK6RpD8ea&>V(EB<{w&nzL<5Sfj;@>Th^taIl_jut|^`&GqVmjrwfF z`yBQ6-rTlfMo?-RdPPZTkXZWPu${-MG{9Rqv&d!tSIG3fkcU@0{#~X;;K zp_)-m{QRVtfN^@FjHg*6)())X3K;Zd$Jt{FzqEB=?rqmfuzzq!-@D$$q@_aPknqIS zT1lK$@To@BbitqJLna<*$A;Uq2FG5Z$zi?Rr7QU!(dO{F4wp3Ui0rRO%5&YXbC8wv9y?C z*55Eq0k8I|o_Bi8^j|gxi(qgxE9UErC7r~fi&o1Lv zL2Xk!c;O_Sz$cUHNB&Z-h1u<{?Yb`Pc&@ASlRKcW{DMgla2|Ngj!T)y*{IOSn z5kh}I;)0B1P!N|`=zdo*?E}MU@b%x zyuu+46gZBo)G#b99PGh^0o9eOD}(C#QZ^*rZLs@PqtO9(V&r5t>9TVIbFX+(u0 zv$T~^S~BdbQmzZpz40Jxm0F_-k6Tpm{apV^d$3ch#YJ#Nb`m?EZHkv6u<`+8z?;yQ z9rGA;>bXyFI3OF2Xu*x8_#`&Q>mHQQK0vKYBSEw(@-xsc^iW4X+`^MYBMH5~lMM)1 z-8)XB%?7XhQof*6cF-o$XjGJ?tkx+unw;ab^Q2k!mg4Tn36`lHhpmglPgkGBJIOeW z%C1D>4J#vG^Lw%5ZIS*rR&uI6DXwmr;6$`oRt;0uCe&=zRhe#&o-9p_Q#W<+NiF+mf`!38b&zXjODEo2eyhDj z==Cw`ZbhzkzbQ06C0&U8;5Doe zH@}fqmZLwqlm(Hhm3~rW!cA)&jgs9cl&|9Ke7>tdb{)Ft!-ZaA8`)hP?nWf}iY$b2 zrS5E3)*Ue+I*?Da*&i)0$JwKoDh-KJ9(aV^tZ`~1hyo$B`Q&kwqNdylW0Izb}O&;55Jt7B!)^v+?juFuY8C3@+M z+}7rO*j%PLD+eRHKFE1^-01DYzz=LnPPB2*%gMP29}-?nIm~5+|Nf$zX{LQb&kPgV zjC+INai(r1JcI1(+SjZVAGc?(qoG^9L6=e7= zHTM*_MMs!Z>@Vu)-L=?TYawoHi} z(sp@4c}b?(lFTEDgl*DwxPZjRolZP1Bq&1SSpjJxyd)FbuqO>&q@Zm$3~;B8x8~|; zeh79M^eID>aYHb9{i|$X&~2+B8(FLQb={xmb5%Nw**;DJa?)mV^L&3z81=M5cCIu)Qvi$WC3R+)}{96g&2|u_~YH9Sr>csHmIk%FVp2tq1Stc~_ zQGd9J--(QJ4!L3xMrGghYW)_nc}rXtju>r@8oh+VOXVD%FVMD^RZyPWB{rv~E6YP& z0(>P~KekXE1zGK{ZB;RT@ZV{HUe`So5nkNDEq0$@#J~xX{yxWHJj>saef8Ta9=N|# zc-4N(<;2aaT`yKTlI%W3h&w$hoQ$Z17PpA|qT-UJJxt3+Tu2n7A;-oYv8Qc{%iH#` zBD~nnGnVRm9yVuLYm~dluc&;rT)MdgO)tk9D{8&iKbh(5Q&s`3SC`sbY&;V?DTL6^y0;aQZlZp?W)_h z8b$U&Uh@L!Vc^2Z#;r8dtQ&QD5Sr}|aoc-2we0gGS(|6Fqy(Ip5pg*VKU(%qlQhQH zc7yid*|->TqX2WwmF!@&Uja&VFuVOGXT>}|x(L*Q{Uwh4=l^~drDKk;H>`f3_sIsV!3nL&6UGC$l zu6{S#6Am>4j$|yaBins$W?lP!9^We($|yD>M)K)g*AuU;Nv%{_>%f8PI+S#To{hVq zdUuz!iUmB#xay+w&Xu7c-_7wl>!x0>Qfs&1ekH=fQN`~;wGHg@y~xmI_t2pIaCGFj z3UU#*(x#cAyCLza0P@skzpL|7wb%7D7~$*sb4K<9EHWfXJ)Y4GQ}}RQ7_hMBWj9dO zb0nRlpv*wqQoVw5>}Jh-bOFd+yfI9vb;hfFcv`O3*DXgNUsX%fphr`KvA%6ec{L=j zjQ7aB%N&INEGv?rNbQA&S9>2@pJ)F zC{!NNjt}^_9~b=Brhhk8(?BSm5mUFBtye#gw#}y9Xdtw%Kv=dxjA)uZnY5(=`cy(K##{i6k9I9xxPhDkyZ|;!Y=TZGFCsRNBKyzwSf&KIk zNR2$7GR}M0Urw1|di0=+`IpBp9eea~j|RLv?Z0UB=#5%whdgXLvms>cJGd>}j zRnBwBIBwgayq6NO(uQ!Q3;$dR>eB%;%j) zY9Kqy7Yh%pAI{XYz0n&X?|{xc^<{HEle0!DQbi+XZpg07PAA5}v_6_G-q~jF*hqL{ z+-=wHfQm>aDLCt*oa+$tGas%b5b$n)?3g6VuAKcyT%2_DiWP++s2?`MRTr96y!+`8Er^Rp7Q%-==_j2eKYz7r$w*x|G@cl=NGmirUt>SHTz4JI z?L7GsH<+_#G9fj*NuDX>y>@E8{A#szm=I0{EnWQjpEdng@CID{9Q3LGa#B?`;|1U4 zvoaX^O;15Y&5Pb@5b7FApFR9%BzaJsCpD)POG}_dt8&)~K@>e3Y(DnoJoeq1x@07} zvunE2@y=4{+~F$MemxJP6qiiw^ePjZU7=1{5w@1_ePXqyu0W+_sQV0v>Ig<=)i&c} z8GaeEuFIQEa*}pwP$5k%T8VtUS84VCZ?mCAfV+J>6 z=pB1SE0xgGNL4=q7zz( zW}RZ<=KA$5V8>0bd1$6KPcb363L!}x@@`+P8&+kLY35t`UhYuxvPwdwdmwUys~%dk zkXN}<7HqR>Ahb=f5mBk?R-|uKIwXKw1(c!?g7hN2*U$n2A~rxpid0dGBoKNH5GjgC3B8BVJ4vX4 zge2d?xyOCa`M!JZ-oF2Sum72lA7nji%{AwkV~p8Yi@_%by2PK7!)At&aZw+M)dNJW zgyV~KI=Z2Z+19{~LRJUnK<_FM9Rj&j-r+?6IL(Xw_ue_NRjmC*J9p*Ql5E6g+*67uZB+X~t_2@{KzA z2}yYg`PzqG8ehO!upKK%M#aB9cVM1&Wnq7A zcp4=5==<1c!1e6A_5{s<_r=!b4!#5wnMMsQS^@i+4!GcLIh?tT&*4VTO3TQ(&$IfZ z7)}lNk(!THlRH2m9Nv5|<{g(^{&pMt-?6VlSZ*A~xXrt9&9L*1_&&XtF7Agf*Uq>S z?Y(pq94Cz4EeR)7gq|4#PdejwRI`K(uTV{!%)h@LB`v?V!p_cDcU1>gRFUADsT_apF9K*8LRa+R#epb#L zK3{yhQ||6$T;?hP0O7-{ZDELAg^= zE}Q9fB7OA3Zj9Lj>9U&IBi-eT0m%cO&s*&bC7C>lJ5n{-UcFHF<50^qzp+{9gc#12)c5dOAl+KQ1F-93Rj_YbIIqF5o;geP z`e71V#shzUxy$du=H*xUAcSX7Oq^^vgH&dZUod6|OeOe5@8rB9lvFa)Ktayw*f_&i zgStWzH5tpQo#*OxO?N;~Gq{%jBG~}CNG_~9Y#CgNu^mcl#OkBz4CjVZKT{M=`@>EN zA|kO#qAmDW=cQV-AyX<*v04qL3M+eMBNeg*Z)13uo{1D*+hM z+iL+{)xlec0?80$z}G(dIEF65l_*}>#M7CkZ`_Bv!2RO%+dYL6%}(+61;balCVT-Q zr-IUuA=*Wx^}RxzgDRwH&W$u?kUuy86zyw|5W zyM8|d3o|d5QaNn>=JJiK5YkSoi67W!Y5=}>;uI>W5`!Ytw`XFOeVWoQwPe-qvQwSz z=+aF1aiu_YaefTrwUu1vt_LL)LPxzTv9o-0XgJygSnqMSL!GgIMuq=$ar?7nK#3!l zj<*BB{#mb7JGqDBQmGW^2v5!6>(9?R{RQUB7#UOA#zNs(p0Z@5YHFr^{ZhTVn6x=~!i}h}_9} zueJ4AQcAe7E?}K=tPGl{7H9wcRlM^|Kf9|d2}Yj0wNpHRL*?D#o-P{?Vu=@ zK2qE}ezT^N26ho$OcZ^M#dAv6NOjpS)M#EO%KHy>UWO)!ODpD9-KZMpr74e*bg0%p z?3?bi_RS(60aeL0x5;acgCBHMdpi;(kgK^6DmF=L;#_aeT3{+jmn9~RRHG~BRAe%` zpJ3^_Gj;;!YPWnVY_WUp@f-XsGzMtc9)*MUa#nX*6t4M!2J7eMor6qwL6l^{zsB>! zhE92?&w^iNoS%I6r1R42=7pzO(u#@^|D8}(G1+6&66KE2oxot7XFGz8ZY^Drx0dCq zQ%dY9a*y-kE7C;gfe_|F^*8BqT5Kp=I141mCEM?eK=K* zg6%H=3ob|P5L#<3z2Nmt*?l~@{=lTOMVgoG%7yc%2B`DMJZKR%p|H5tm1Faqah2Pr zpu*(@Ac(k6Uw-0BFstinW2wfYhR$ISYAG1;%Z`n}$tr zog!u*Kha5L!v^sxU8(#UPqevbe#z#)SMdLp!2J2&Fjj{h)0=bY(OuQKw^ziulwt3#_}+(9<^^;{D3-4#id9H437CU+uv;Cg&*In90S zk;0?!W+2xRC&E8fj#2Fd&!I_kM&pzKx0f-+CgXs5%Ti{Kw*4W~3$R?m>JQCwT{(-R z4uWT?(p;{b;9!zcLEa1s(C4HIxQHcPNB_olVlrKZJEpGu1bX}XRCG$+Dyx%8L#jd+ zkVtGN?K;bSXX1+K&p870%Zl658O@txsjpG#-rqNe({ue2F=wT}nXzN^3f0fGI;Bgw z(w-hCpWE!`Ig(tzjZz*X>kk(Nx$jtgtQGiv-A8mLb(7|rqynAWWMIe59&18qp?r*R zA31THYgpJtS*_>Ru*cEhcuuu9n6PI;qR%c~2~T_Wk%{#gtF`q5Em%v>QO1UXSIaQ& zJ#N?O(sU7@LrNjTu5sCBVCnVsgP?T ztrQY>gSTAhE^CTHkonFPO#bg=4RWQZM8THyo!zEQBnK9*e8HM0qkT_$7?V5Cx08}} zjGL0em6sP7QuyZ6$|eE6IwzdodgkEf$~yej%vZ&~hGWDIH#W^b=`5XMPdcfLJrC>r znOs%tsjAHoc+!>&P1V}lsNlJ#Bvv(de0AIDw{%B3KE!NZ?I!$|BY*GKn4h6gX#HVp zvPDpJKgzUK5Y~T5cPJN6TSzRAL%8eiP1^G;2)Ht;y2vBiDP~n=f#K6})Ys$nypTdC zvS~;rI1=%g97y2S_w{D(mUADq6U6R}*Qn6KEe3SVM$)(u1AMQLex|oD^1YEw_InB~ zg6PXy&y?k__)8goel1e~X$lQqiim}CuM0g!X-#m|Gg{pO)*lYek59bzsbc|*tQc!N zDXZ#3HTDfP&HVP;Lov=q_;AGhClIwUJ=i@}!l?KMR6b*CP#x_$q#;-}Sjv^hEvonB zDb6o~Tg0G?;;5JKGE)iL^y>)XPa>vG#E9jM?PPqEc2Hx5IFKlz=u zYLc3Pbc>mV8bw)?o~esFf>N&r|vuP&A9z`{1FbBd}P3!OTa`itiSeY9(xo3V0f zz25>s)VCa0kS~h)PMgQaCx>TO`nW7Nl~efg6sreXdW_}3KLmo751G^x@JlN#G6;81 z$N(IUKc|b94z4<6nFxA4!hj}u>f~||ZEC6%ga>>nEMOfDBQI4LJkmY_LWWn(e%$s!f9(WS zyq{HC`CU-;1V@@6k*Nzi9`I7+OUg*XNjQF`5s#Bn#73$6hCq~|zE(1kN`mNJMM3(u z*eRO#y@;Ca&ul4%k|tNmW;vkeLJHYXEj>WYz?u_$(f!B=!@SG2Y73@}`RQ4-J8W8% z`z}<3@QJkaUNTsqEJm1*NBnj$cBMFNCLP&bM=ZOv)OQ?`Y3L7KwiRgG_&}-^%@A8D z%=xj4Ff7dZ;@iR%TL~RgFwj)IZX%BDYQ1x#wf(|QeT2vM8(%t-QWr!tPpNz4n3G{5 z{Mn+sq|Sekv662S!;aIB{LHO5xB>IQjS$|~H;q)vU(ufK{^NNu$G8pI!uvg@Qdri} zfhVn}hk#x>MHcrhb((F<9wNHVmy_cTAQGf_Dd?P5fn=9YMC%AA$b*$jBkW$Gty`Hv z{P#}X|DsaJ7QO5qnSKlR>D|bQRcwiQV!B{WIzQSkN_}3@JTxW4w{*&8Q7qfrc4iy6 zaN55+fBb2&{U6JE)dOLvBClH_xwq}Vb@3FME}{!g@+1-n*Gme0IG(cqS3xO%lt}?# zs)n!K{&-)bC@@>=jjU~J%>eBPWkV;h2$qlvUekmThn>X++)4crU{=0ddz#_B z_JGGoM&EYsn)G1<{h*jKLdLMoXen7xu&U6sqO6dTp5Q4uQEQDm0)>a59o56bzusa! zbMG$H5p-qm+yjyswi01#e@ovep#NL6;TYsGw0|uEXRu|`zXe$}W$zbDARk<57iiIr z1>1l-sckW?dRSe5O+qk2sJ<2M-Xt9JTV<$Ig{#M4oqvb%D8?NllYd`E8~c$iMe-uw zD;z26*0p=?jgS3d+h$A;j>@(J%n9B7c0hKmk$*Zw{^#|c;N<-7R(<)+k?KODjqzRf zFzT1ff2NBHh96Yv!G@&|m>F8wTUTG>iURvLeNJu-ZEmN@xtk;J7l()5QqjD|t93@g zDKNX-aSw4Vo!`VRR$I-O`D<_QX;kNn$G$sBz0~88Evb7L(z>1{WD z>$2qc0>qq}%vy7Vz8ft;BBp_F_s!N<%iH5IsSo(<#D)r`W&oosJ532CqJpc*C8(|{!K>j(S0%SsXTwEhj%oa zUBXrKF|Bu3-@LfwjYkD-uM|hIu-{v21coKJOX$is1XdL}FZN{XkAZUuq+!6RsUb?x z*wi!?$h)i1X9We_rRw)eq8Xd|fQL=c+ZgT~r;)7ZTi%&V@;tY(9omz76H3FCx*3NX zHn3(wH0*Gx3YOQsUX!FO2L?3R`v`_A>P%Ot(n$quf9wICzs5s=5BvczF;6~xi%JD# zobWkOSmU_3Y4SHXN2Zc&0xa_2y*YjA(rDL$Tu4m(4}oLqoNOmoBW%Xnw@`I*yKu1^rC5~4-TI}hPSq^&G&D;*z z3I=%n^?(Q$kR6%1)pipwM353;$mS-sgp~2+n1u1rg0zkJcw{f&qPoeDEqd~oV9=j2 z@E?b=dVf&U@z^hb>NUTVE&pldHIyiMn_tPLluak}Mo2AO+V;hX&~;6u1tw(2O+-@6 zqd=AF>|SftmLS2limA5#+U0dyymoiMWP4f{99c6mji?Y@@iT9U*~XyExKa(nb|6UT zD2q+Yh!P-QdwE8pWRy_Vl+G*N_<~z3!O{Zso74v_e4qgL`A!M#%ykoYbO#{)T#$MQ znJKJa*K+3ER_QC`0>*f?n)@$a26txynqiX+^SXT36xr%Xk-6o3z+^Mp7k?9SOA3N8 zXm2UL^b^ziPjJ}3ufS103R3_F?QVOOm@>U;I{5OUt|?^QQ!v;v4n*qTrFS3OEiE2izvM z*V&|u>S}zsR?>uZ1(fJRHU-dF0Y9<|d1&8+BheyMhPlti74gk5Yk{2DY zDw84?T0i@|JmT#3zCv~tkHNduPadIydD=TrtKl11^1RBU3qS0P4mU87gJdqf@Pio^{W zMAtjUT?wh&S{|fs6aeBm#WYxYy!idbdd4J+z6=#m-WY!ZELVN!2ba$$)==~O{CjiG zjIm%Yz7O3rO9A2me!Bg7Cuws1ksq17Q-dt5Xov+9qQAlSiVI@|R+Bb&7S8)fI3a$3 zXru0RCeyd0-|oS?Xvy1KgB6Vm20qyJlTm_hfUYi9UYw?`{=3L0tB4~YA=EMPyJH*5 zL7jY~vmvdj*aCu0mi=YxTr06AYN38^3}f_kg+ke9+FT7WmXL_qt-mLxaC;ODD-^7(hkC|~?mm}G=HtBEdv;&C4?ZHXyz4r1PYYe#+~cQ&e2Cez z2DaJ_nnd9&{po_c+MA2FtQAt>e=9}!d7pBA_hr)5&##UVnq@PD5SlYLNS|W7CpUlx zP0{+ns`u@@GIVGSMZf_b(-o~x;Z4~1ky-4OdqsjFfXR$SkZVb1SKoHZIci&xUG#a7 z4{9PHE9dos5Kk7gNWoM8OAR#*cKgJt6C;{e&X{pWp!hu=o;cwV=|ilg`6(6J+`*Vl zE;*ZN>{2?@Zc;_E5QuFjV%=5($FV5crxL-W$*Apgy9t3xpSiP5*~AJvzoV$R#@ zWA=J<&m0A-hmD9yqT40wcBV-E1Ebo3dL&r~)_!s=)s1og0%;AU9_p_@1Y6M89rlM7 zTTG6Uau8IgML@P$@MPp9@@5+mz1*`UqDp(U{7DyIFWVG}i11qy$=#XQH6hkcH!ZM! z4xNe7lx(M0A&pVg%;Fe0vF-!%G+r1^ug2ej<2^@RMF@rbn`Uu^>Xh@9C>hVTqk=TA zhPzm>d+hP^k1_y-EMt^2b-qkrw>*k~!H1?!5jGx&9dZTSaT@Fx;|SZkLElxJpEpgj z=CP2siveZgc>>i7mI12xqW$2gdvd#1T#7@_+)8N^G`EUxJk(;)eQJ6u6~5OhEIZc^ z)sC}kOrfVcdi_B(uM2*r9GsF6VSne~)=AXVo<53+c1LngkS=zzW*mvCYQ!hbq`&9* z7E_*xE0jOnPl)hdOsXIxyL0BS&oC(qY>@b)^`W{EGWb@%J9%?PA83Nk?~6@<%#j`9m+ zv47)#Vja;H&=AfHhb#gH|2t*|$0#F`W*49Klt)kqz`7q(pGUkyX|=KPg6Bl7#0=6P z+T5AwF`u}?sRGumK+yTVLZVoW^F7uM4ww~A4j-b_zzNT`gQh8dK5Y*8)#;Zz)j191 zzqho2>zy?oJZA(&8cG;kHdY(6l02P15v-*>xh3R-`cfnMkforG%^$g#uAL!$$tZ2c ztYLB@B;9x5p>!yZ!EDe{nPWZ1CRvUn8r*v9R6!R^JMIVwWuJWs|3hbpFEZ`f_IC3b1C-~J4ibyVy^eF5K z034%f3O>Ys1GqVt_$C|cx#1)10xmbx{V$$Y9Ke_GYb+Msab!llogg&1DMBYeO1#rv1( z8m{>%&ob?mkETIAAtMkK^HALi*HOEhxC%Fwpesq@oOyx+zoL%Eyf$|aF?8XAe)jquGkL#PNT-6|-MUWu!G|CT19cDdnf)OcltJ-FM*cLR zng|lqfN<_u%0+%*U?ElG<<5y5YX~o@NP+P~Kw(bPv%|i(!Gg%C9NQoj>nFYrgPjK2 zrmL4L&9y-mHNX3%jg-4)v2EXW!eVwlY6CvI?lm(XU#_IG^9RQVn3cMihfbf+UwD|g z4%kJQAmemi^hI~PR~p1fVm2RGM5>&%FWS&VS;LO;XpK*$V~)jL85R-F;B(O;(RX<) zp42O;8L2d|-v{NxxZ5HoJn}KWbNXbH0?Jwi&drtvcAV$kpe%2{W781yoq0p=6N474 zhdI$$!x}qw&5Y2f6uu8X<`8d>3NL^3w}|)k4exu1g#r z_!hYh$uz;nhVSXJ^V%WDt`{tKOC7wrlRln9J*%v=q|e>YJv0y?)1YcaX)(PDA6}o` zPDXDtd{@t}y=%kCi1>uQL~4aBUQSF3xv#E`pA12rs#V`WB;mVNQsT|7D%Zl5ymLxK(nkX3FkKy=jwZLK$TSDoI|an(=Sj+eICvtoTGcGk4?_6f#Aw`+S~o_Z*= z4~~O*H^Kup{&{BNj|5-A>^7jOi-+bi8d;-BO&J&KYgLCqRt0!Dh?_@`l)VEO%v`KnGojuS^g)T<*}kLGGMv= zvYwgiayWL9I9lmmIS#|wu8x$gVe2XushM4&n0h!(5tlm;80GC1-H?L%9bDC!Pn*x^ zK1dkWOESYoCtp$XQJ&{h8fBj7r>ClnV(w7x#E(e%#R;H1GS`5(NZZ_79ER$DpwC-` zthF_OlastM*m9Psc|^h?w~vv*5z+<)557iCMqv2Cz$l@NXQ%iCi80V3vzFyu%%B2G zruXV6O>vd6C$W|#U%$_1#8UQH;Ol@W_wVJx|M)poxBWSzooL!Nnm8iP^IF{C^J|V* zbFFd0X(~Ak4~U8(#4<7z$St-vVM%H#S6z^*PCL$&SLe0~RH6wFl(zF}ZJPETxCI<@ z(*m8sOOM1Sl+&MGh0JoBgLYLT-EeUv*?=Vj#(gV-5`&B(X=mEvn(mqj$%-W-sxt6- zafPqpF>K%NII&8yK*k^kRmxpIB8>_R`_k@)&1z^LAy={PIt9k$)UYnSNlxErGc8}5 zNf!Nq6!}AW_HpPA9igS-KCX;GG=;&NDupX+8z9OY#(=R?_*b7g`>fc;E}-VV(CWXL zmT!hku&}@}@A>L-$!Ujl-Ce~#B9yB47H~YFa*;EM-vj2;m!S;6iJdVU(M-+SF6DTN zwyZ!@EuVYQx5QT^3%LPZP%%cYOufA1#+sp5G6wmb!`%W-D_OGI;Fx&`vcI>pI(HD#R zd!1%cL-C#pfB(U~=4TGifAHGuM~uxY+k1eD^ZtvHSf)P23*^97f=RwpKjQwjauw3z zyvz^7d?Twwi@>}*Ru4bN4D^d>ImKKOR0`_5qX~61Rf7PcFp$G*jt-V-VD3P+1Z?nA z4lALx{E`{-qw=pWd4z{nEAb&48OwvndEsNf4V~RqYZzP^FvtvO?a8h;4&bcC6?xhq zqIM&Y{wTvjH{3JGj|g^1HAdsjMf!9sn0_{2hlCQvb%$dbVoKN%NxJf!fgBZ!Dn_}# z`;{H=-JwsmA$0i5aA+9I%f&#fUdH+?mvYnOp|(V+494dl^WDa8fIIvOL?(7X#|!IJ zMkb5tP8Qd@Y%A^Lh_mH_NmKSMrEx`$z7a#r@}-Vqbeup zpS1!@KgF`+-&s~?En@(?8qGZ!kc{9CER*I?kdU(TQ5tcXu5@bk?7+No7{Zp3Hdbue zpPVY^fBUalgMZe~3OYjHMVW`Q`CtcGj_kQJUCWIDv9%Nkk<99XQq&81%=|Nb74^*I z*`j823l5M|+%c!CfJtJJc*x3YGQwq8-eW}A4O}; z0p<~v7G|W-{-yp*z2a}z*EX;;tK2dtTVOB54%}+qswq?(O?t{1;^9I{?*_FpH%55p zkn60!zG(G38ImFQ1RnKP&#E!=fw6WQ{x{WgH?YcY1 z_mOo4;SoIW-kvhywezs?<~vpy;xKv;m@aDr6pfL@1jjahtU?P4L#R1S)6I}>0&>We zL7=9;VEw@dz2kjN_2FUkiO1Ab~75IfEAUt zm)eHqwPq;{-3e)Sz^DMQ>d6D$sxF!c*zrzViecw_4FNgw7sjw5yzVSt8q?sC?it;L z%@(%^R<31)oTq@8;b+8+x}ZISNtp7hka>8u2PWPly?im4 z=hA9RO{%bmSKQNW`YWRA9!KfQy1q`CbtLOt?Mut%|81@R{n~kb>F^Ycu>SD!3B!M` z{yM>Rq^z}OYhm(csYDEToHFhmo2TyM=cuOV-|k%a-n@IH=@1J=o5yGA5spxGX~gYZ zO}$5^I%q*)JWsua2q%T*J)l;_f=XUWRedq&mamOuaiNaED>khJopq~$z~0>O_2|hX2lHmU{hD`im|Wgmpp2{8q{22+)}ALc{oMUHCa39hRA7U zhRf6+_%J3y2hKOAB2PUKfRbq58`;Yy{p&Waz4eqIH(J%;4pYyJ4R@Or=QOCy+_jfn zduwdl5~F0}eZoka6$fKMPF(C`mFN4BCcNZVvW{azh^qP0;N5k4H9vW+g~5!yzC*Om zJA}0s$GT@ZpEwQ-D|w`XhQ*fMed`MhX3#E);f{K$8n*;l**W6tnv-&CXRldgpui^O z=5%8f>)*lu{|nss?#J}wsw?cAp>4uDvmdR|69J-xbi@%DW?9mt_R(|xfaa6iqzI1N z`Ca@zyFH)p16O+A#HX|#L_Dr5`|g6b@tM=F2+3Ekg4!V$3}?N{pPo2xbb*Hy(^pZw zh6NIQcc;FhOpehb7l`d=owKBWGtcd!cgW!$siY!BMUmD)q4Vd*MXU6jyXc+}G;q@Z zCETuxq-5tA3FAkAvB}--v&>&U$uL(&Q-jf^fn7A6>rv1q=;qV2e|RbImsKS zV{B|@8DMO(E#Opa@B#SIw~J<+X5qJuddD=x)YB$~a9CHbkSJ|v#I?1cGp52uEXvH`PQ1U;NwC;5v121H75DSY9L zxk`;Mn_nr~^sB|H_vG$8)%U&hKv(MnBl&$0bh!HtweGYY9SB*c_%abgz+u^3&IzTu zhPyroUrmzw)Vw+qr?ZaS&uOFHak+DVFlz%Q4c2dAk%JNYaJ{8;16qBQ^=58`qp1^O zieJKqPR56t=yii6ApN=y-HjPHr%ixV!y;xATZxCyN7Rj+;p1||7!8)o6XeAJh#612 z%L5Q`+N3IM+s-iV9$X!*h%Qh0jhfb9!(>r|_&pkn2TtfBqIbM7U(@1xEo{?@YP`77Ac`N!nJF5t_0HcF=ov6L* z@1Dl!lz4#i-pfpIm2AjpzE|>vw+bkd|aaO<2R9%tepI}*X}CRom7sUWJin; z>&Q5J41GHV0_#s^*(4Gp68D=~&4naR|KnyU>M@E}{6!4%a=zOY=@1(NF>9QA^K-oJ z$Ibe$V?n&e>}M3K5W`Nay$D7BtO~2?+vowvJ9m8XCl!#7L%3O|BYqs{PE8J;I4j?E z3MxAYKTBIe%3gYJ85I8`{wDl3+viXKB+Plyp66Q5et5ak6;H@`-rTZGj}4%0H6yu# zfk}KtjC{m+O?q|G;hJdcZ9Q3O@?tHyS**-{21&WfJ z8P@~V;)KGF9ghX?W&uFFWUI*L7E$~Bl6h?)m2@h6{X$=yK4G}okbIXtGmpFsb-6sWrTL$y005BB9c3F zlS7%)H)!evcyH&E@AB~VH7Aq0!1A=fx%Zh{X`L#AjbaH=vV0fWqIqRX^mReQ25Y7V zLm@ZVIaDmB+9BEgQ(2mRK&~+gftpT=8H#BjoHj94Hs*o5wkAPBuZ&9rw+>%pl}BpP z(o_PI!iVK>fk^(KcIo=YDug?OLM6%q1(p7e|Bt@BlSq59A^6;4?&y4+@Tj{RBOQi! zO)IP$KtKwdfY??Hq!$3X%&3=f?fFf!fBSI6bCQ}(KnG;n&;btkRr`hH>ina%A;f5d zT!k(KFC|(~Cj84ZkCr?2kyMbz8&sebHf*0}%c$}wV486PW(-%N$F{(V{zZa}IvVA@ zXw351wLQCU$EjcVbEnz3MIS|&ss5w17}})wc2?KVq157W5nnv0hvJU)^pcr77jrX0 zD;!a8DW))nWZk+}sCG-W=rU_Up(pv)vM@Rb@nX#hksfL?h|<(ai3e#O;c9vTprb=# zHs5x%W_Zjz!WI(`Ce&bb5u)I^^1)yZgkQ+U3AyGjlXdiI3FE#^{(f)pRQ3FxcX#c> z>i~k3w4c{(!d!}|%*@)&t@!|%;vnOmQ+QZ?DXiM!Ll+I)svXniO=;2v4+Hk4@LTX` zHEih4qmwGVvDJDMSJ?RCq1i#aPuwEUMISRD{g9)%4mKz7;P5BQ3kZA^_8xZlK^lu0 zzg!&YvRtXh+!$sh&E1YNm>D{2df{*H!=sOKzO0>CG6pUo&~Kt?EIPnIwqxU^rW3b$ zUZ>3=-isNj^hBCK;*I-D`qyV#ZOBz0D%DX4#NN7C;SN+2`qdQD*o5Q}>k3Y1TM~9- zeFC0QUnW;AZLu07y#oEWe3)@TReDQ_l6$Be?<;O*W|lOnU`SDi6`_RpB{ne&%(5w&dLP(EKs<03~!Nyx+FKKX>XXS&-FXW#|gd0 z{7ADa0SKL8Q4Mv4(^(e>2hw0GfCZ{Ns_O(jUS-M*7CE)qdz` z3t=2lQ)&J+)p4)FHM7ZwxRsX2XP5^rlRm*V2qU<_Dcf6ITsIumu$n+35L{(}TSWUu zAWwk+qj$!tN+VhLM6R}n1}$(apAo@fEXAXkLA>(lxKG^G^_u7B1M9ca*psj5Z>|hT z?3@iOBwn#N?!RyW>^LJOcT(+wAH*v$t-e>S_fd2dF63I)Ec-#)6K#f=WEnU?ywWkW zy%>K+s@!&2ZYaj=z1;hYu)+c5eRE!W;5z9v0WjZ?_Zg@a-u+UOEhZiE5Hq( zOaO*uwAA}?$LWs}md0_LS||~r_sR8b??mDdv?eU>vsKgQmT%VvNcGojs)3nf0}YY6 z+%@K9{>2>qe#-NyU3u%XZT7%r`JqnU>0LBqhHJFO*Ezh%K^xji0n+vDgDYKX&(G)p zSN}Nd0!~->9;u<09J@l*-mSxC*?61MOs5(!Yji z7cfX+2TNwpPK?~HCxu7TNH<64n3`E%BlCJF(SsI5q!O|OR+jUD5-0gYY38|bu&vW- zu4=Vth~W4pCc{DRan%#HW0OWq9gWx z`@QDX?t_O`E|X!Z^Sw`=3;CF51_z*}cyRX&YE@Cgzgj zoVZ-T%z3B+_!ATL{i+({oY)6Om(pp5jg(FN!Y|~JUP#q+8p`(DS1I%A9+=4`Nl6^e z{*lF^z;XUZYzLWNLUv6Wpv*~P5X2*HF#`>0gVdf0ZgrLEl~OpfFY{{%MA5hxdZ!uhcXze|SEw@T0H z+e*3m!Wg7zWu+$;kzCKQpXnZ40j~c?OYpwk(_R@$Fw~iXij=DRH9n4*;37shIW&@r zPX+)b6K)vJlR#7@i5)9`WlN)M(RT{mm|mpEi4v$e&jp6FXUiC?Y~OfWb>jjNXzI;h zaYy9*4q{Rxhz-=jmU{As=M~RsO7iI0P8p*btIhKqS$)L5^A`X+bi6KzX2(1n=gj&* zbRoH0n>O&-9Ea%{sbm_IaqiRD_F(q0eUo~pgg?t`IyVD87prGW0jNyb>kC7$L0=Ko zjB!p{!}fC;YaAhucr)%>be!AwEE%BolYyvWmkb-=ceb>gd$V3Sa9!^i5pMal7L&(k z))^H8saMPDLZ+4x$kT9lOo-pC%;3BbP8|riHbz#tH^@>{x z{EB?123v0ZiO2oj*jTLzkbc~1Myq#>0n;W|UG{UN22d-?60-J+>i}&`gTmm8cqnygvbL8ju?H;QF zXxvKVn@g@eJQh1$yKl4ax&l-u0J@b%u8L+1qRd7+M-Z$9n{*skJu#q{gT8N>!G=Qug4`_~S^&mH?v z5)gW~xgR|ApE!&E`t1JYEBx~hEvW=-`c~kfXZinrO7<^%_4O>^5?8u+=5O%g|G&Q* zU~c~3FaLj^{Lh{Iw-5XN{`>#=wtp?vc89a+fK&6=uSGoQe>sr9vY3ya9^l4azZnnu zFGrZ~w)-UQR$FNM?BDNI07Q1EC7w|?kQhL62J-j~qk`5Tj<3c|y z-1@a7MSuxSURgV&DM^j@&pM3ddP(|ey6`|k@3U&yqu5=0eEp-!(tl^R|3wU(m^ijS z#34gGdj-1l>}YzFbJ5REnd7Oe4?gRaw5$%J<%U;X{`jAw)qi?TULLqzh?P71N_(vj zeuPBEKAQ-rI`Y$?;PsWTaMV-)AhY}=z`G}Yw~v45S00j(hu&dXw_$7G{@QpiWs9+P z1=^1R3#-&Y&p9f>SSEsx6)q! z^QfzOG2H^DTc-@Vj=Zz1q!GZzvNa4#5tEWf;akCf@i71WX?*897}){HYHXDJ2xY*N zQ2%<9Taly0JWt_z!$zM&@y`*4Syw+WgjzGCGBVCp+6*TL8 zMY5eKd+_yTZcR_X@AAMzqr=3X3unbdIhMd``(p@&L%X6=Br1FBh&c4%jfh`cz}LI~ z5KRQ5o;hSPG3RS!svCCI<>sDL){PZt`R55$En%W?#Bb7Dv`U06e~C2xMeXGGAid5; z&Hp_=_>b>;<$P^_+kGzvvL5f10-K4S8~K;2i7yR+33P}@s=|MJxBklKU)wMP8k-@L zmM=e@yng9RghgCZCGAbfd!P9~S~{T?U_v~$xH2Z}uLb@8)9SF?c|5V(;5jG$KfU>v zuc^lDD{ySw%?Rgr=rn36jd}iwwA~eJHw?YKEZhB8zmX7t&AJi<=k`T)^Y)f~f`jq~U?x5oa!!Hg>^*LJM~Q zU@I?o9yBJq;qoDE3By)K($Q9g+5iAZ34cZv@Efu+1d4p5US@dzXLFS)P{WV*syhJB zujFB=-L8JB-%?(%Fq6ypuRhK&anA4e8#Yex%8kVEE9F1<`XWcq<@P=$W%NAHS;_mg zRqb)|8EbQE}8VvwOp=V*isT%(;GsCh1) z02{SlRJ=p@WVr^+VJZNa{IZWo6Tc1W0V)S1v=X&X+*R6or~mdzr@kUcGs@BEO^<>> zX2@oe^cjq6N?HJA(NX^9wtSXB-h%{T-SjTY8zMF-=qu&6z?L`wE3BwiCEP--2m9Y$ z6o@3YU1NGLQq)chY`#W@l_98G&k&xTj*5`mn{RTsyg7d@s1BP7nXq6jJgvq_PVGQp zUkP+*_WRANL=2;%o<`JO)w*pYY)dI!hh!J()d#;dd!PtvJ@}9HLO_vrHuxZH zRGF`Zw^Pt^`Cz_7_s3&T3!igh7;8}~w3X5dhwh8KiosR0-uNb+`E=e9sqCQBBX5`w zUwf#EF|`6no1|)Er>1#8#N0xuFH`-c%o5mpQm^IEO7HwKNxX%aadU606)Dr%jbdP6 z^E7y`3z+3P9|iL4`Hh#{eJ8K})pV>-H~Ojp2-wkgT)$QJ}`?Aep7^jKd!MneLP|D<7Khs(UI_t8k&*|abn zb2Spn{ab8#WK?H=6y@VDVg+A9nTBho0#Zt`6`h0T&dh9Q_S~D@^8R>!$`6K3uV2~) zMt`qP>w=17=2PP=GaLi~WPD!CVGwB|*C4r%kxa*`3*3z+t<_nx z%Ou5eXW^kU!CeZWz_evXamA>xvAjPvU8TN}0t`ZQ0YIc2QIhs~IUT=LHZ$gxek3%d zv2_HhJ)fdtAJdTn#!ms-Ns8o`hbU78jKp5Xd(=lN9Zq!tq?94PK=XBmoH20?g*|D@9u;c$87PH^uiw?0c}n^*rQI& zY{3^`poP>0&%85D*~P&sLKz|Mf_Wo!0+7)SNMVvF{nsA=ZGcex%hMunO)JlpXhU|T z>K88Z60Y1^s7I>9i;I>P-Vp9*aTPu+{We!)e{)bsSS9Xx#JK3-S$ficIzzx*vqeve6d|{2`=}>*A z(#>xd0^R+6ohMG1${Df)iikR144{f8)C$)EEDjds&MM*V-7uEUQnBS_MT_BAd{W`4 z29U_eEwvOx!9%ocgKT69_YVRNnaJ@*cxVAsOBDSYI)ShdGf=VK@d)}6`4&vsnLZpp z#a)J3dLg#5BvyEX`A5XSXe8B_@ImAQ|Etoh6rmdGOv)pCn5h?1IXBISqBO^q1$eE=Xb4U)1faw;hWi_t%*DdNAkq8x$D#H*O)`w?B#a5ZA|7 z$ci&FB?SRZbwJbQ*;)?d81?9A!iYgDwWAWxUPC74DclJ^#1ZTO;JVCl!dG;XPdg@d z)mWm#DfX3a{Kkcm#gb3J?aWj_)58swzyP-@J@G(A^^fU&jfvX>HPWlg^74nkPjpWY zO(P0hMmAT*$I6|r`k0qHL6q`r+w{I}E?&1w3SZ81M9@(dSM(N4xy}oOkkpj6lWZRJ zGE)VnnACPm6Z3w|vnC0b$P)eDebqzZnSvJ+te#>?srOyO*>36riV@f#&{jt;KIIMm z@G!m<^3^14@~6CFWVvvZVd^jcz@*+faz8MlXTR~T_S}_}hQvdAY4_uof%)9S`#LjZ z0n{(ku{<$7s1t(+R55S`*W^Mba8D%0#6lT;_ECHL^JFSdtsLx05U1_OOblajA(~g8Ook< z5Qd4cR~;f3RO$DecrM25C)BRoEqYsPp}LhjCm$- zW%nP#3#!jQFW>ozx%_3e|BtFjb)89}VrrKIOn2C)7h!7HS4V|ICIwoQ_~b1bl>>Tq;6q!(P#8?jyI}B1$jE@zXnoFWRm^AEc~ zLk6*)njeo$*SRT_tjY-*-2Skxkh$StoPXzk%C!IEYeF@u^$RH_#52rZEXtJ695!Ma z!F!%9w%;xBA!M_Z*~acjR_>B|$(a-&&po~fqlHZb0K6o9oW=MO18n5-fmprAh}2C{?9* zlwL#cAtGC`AV}{;kPe~OKq8{_-a~-UTL_`IK)4IfxA!^c`|dec_wV};9`b~@tTor1 zV~#N;VSTCrJNm}3(6Ya1=MF>}9wou|$xslY- zvR-tZRKL2bBNm!8}7 z>$#}sD-;qv$xQZiHxI~oPzM~4S6imXg)H2$c_0nqjd;~cMV~N2vmM&a_Zwa8pK&MT zQiDT{(`yxo!{=4NQq%$jHbSW`REd1Tcqr@rNY>@tKgAGtn4(wgvJ`W{8nwcvDaGFN zKi`ljRUC&o36w(Xz37A5C(#N>+Dl<1S0FN9u22ZO^*r?(<;ocL!bsjEhj#YL`k)K` zo3QD2A;&vjAa74ojGw=cW$1YGk&)k9FY5)uXbVhtCIIzz&Oxr^bW01tT%v<5YFB=8 zhYXl7OZaT$i3taUD6Ci`+f1ojVqiQgjn}pJM>m5wI>l?#B7F&e!5fWoGFQE@sikWx z?C|*uV=PA{nmipVS>=PsmNrD`RcIlXNONAe{a`tpCcvfY@PcTjqSm9$)3Rt?+lDE} z#+)9X5p5rCKdkvH#t(miC7(QfuUry`2zi$uL`qswT6%#7K!y(4$yRp$<6fuR`Xa$^ z>>mn-&7lWgUx>yPK!$M4l3I{gulSa;qk4N=K-&9-&~(LRnbo5wFYVkey*FJ@nvBSD zcjPui!68LJeZW?xn`WnKzg?f`QwUVc=5SsuqYsf>1ui_t^VXw>)qd#pwg{KMU#FAW z7nfP`&hm=zqoyVdmMYyo|4~Iry}lhin|uU&4t{*G|N15I_5h96*Eq-=4rkSu5uCNS zj;E0DT3ZTlx=;ua^eq19v#y@4n;v*TT}FLFaUc@90V=O;m;lwinMTKg zJMUk9o8boKc9Vw2Z{jcja~0umM|8}C3JwZty3(4%$IKq7JF0(P>?=yQ9jU^!iSIch ztF6J*+pgl=A>aA7Z2gy7;k6g)K9;Z_M?~P%#ey`{#rsg_i0SWfK@#M0;QP^ilVai3 z(UUrWP7Y%sfD`+sq<|dkfQ@?~ARxV;cz4X8Q{5Ok{VF5t5fcQG=| zfKPo&R3r4x8>!WRVGj~vEdBYPorQl(3jQmk#!PqAQE^GnclDBiXXCN_W0Jf1f>r~! zhq+>fYzyrW3`O@!vgC(16TG;%R6Bc+4adl|Ro2@mv``@CbNTQ5eo^ zIFbk*s)y*_a!;TZ%B+kKSl~j74NGW@8v2}giTCLNB7F`dzUL-Y$*Nwe7og*l7x%?1 z)cG}+^wiPlXvtw^v2`_FJV(rdZVdMs2s8ExNslz*Tuajj4)^^9Ng@gkvhBx+4;k3`{yeB3lgIbc%G=clZJ8D5u$#Z=!k z+65tB=HBaqR;@?meMg{l0ShZcHG{Hv5E!k5*qqeJ(M3yYDK4SR+nEeux||nW``5Fx zJjY9$kCr+)s1N7cXil>9RcjgJ7~b+?kE% zE*l7dqDx2+&xTSc&JGBfU-XP_f2BB_1Bya235R(-rcUA0VH~7=RoItVV&d`YYkTJ3 zDKQU%?x*G>T~F?jMo%b_<~dbO?bNTM>t7dL_(KVCK=%T;P4bbvYOhV!yANb04G?(% zS=6niqVZ;#29bevfNj}-ol)rKEWb{Y*s{Y;jB&a4^U?x zImV&7W7d-7Nmh))cEw-_^i7Nk=c{=l0F4XD$5CtYRw}@?zuzCcqj*It zD0Sb0G&Gx)a!h8qKGC~tWPjA}2r}#3W~=bXY(KmE2r0o!(O4bXsP2A1hc`UEpKcY- zK~8zqT_`^UpqbeN$AIx(a#5gqhdwu1dUg=>T^`=gg0ECe(N4p{%a9OW_a>?E?UTy0HcK#AOpC4 z9>oOGi1@`UA2}~>ovOR6sR!m=OF6C@sY|Q8mA9-CY0IUpWd&bHl7_wV4mec)CJ=~d zPi1-1c6pFgDb8nqnCv(P{IZ)#qq+BrWgcxNpXAARwH=@^}z#h9eDr*Dfws zr+QGcK_hhwuV++Ue3ECTHFizN`pSu=tdH)CB8D(ANy!oTNU>$;&m`A8h)HsisA<+E z^$+g(i%q~i*8x1=QjgW?sP>4i<$5Nk0AktYMWsKz02C;uYVN;rH$U28nq##TGqChy zvhnQ4S2ACm1C7vOBkH|+y{)|oUSmg9UC12aoY-KwD`%IVyzYTG^^#rUD5v_7O%v$d z9qU=K&oi6r!njjPJOlGtulqCg#a3bA;MZB7swbWb}fA&_YOC(0fa=bqK@fq66 z$4oU_Q+ZZvH)I2gRU}C*5KqT!jo`|+<%sTznKvA%?YoP}Pcq-*wSQww_y5$0pX_|S z=N5fWzqt5BebqeCXVCK?x0`a(U(ycGNhvvC!tYnQek{dU^=Fwz>N~!x1{noszQopY zJ>5d~*A+jWPd>G+!faX)?{bi<;=9Z8*w+qtgOXZ$075)y9^g_<037Ol zk}%qF$_q-&f}xi;uxa@uI-Y<$P2k{U`)z9(u;qQvw}*BuXY~!we+ga4&O>ipgCE^f z2b1smV%&%Foc5UNk$z+mCx+zC}tx%Ymjurj@^QlO!csE~~?i=DazTb`iFQuud4_Y&nWtq07O<@Ij>B=8HqbQ3TZG&8s?jbx=%X+i4 zyXyiNOcN_P+7GmkuQjbS*H>cfy!16oc0RDHdLP;a_6Q4=)`)kWgR%O(Yn3ke!Bxbo zjrZZ4Fcz!V^%o|qS5}%~cO?dET4M!QWwws#vCIR!dW&OJW_EWrC9n|bwXvq2nQ}Gp zEdBx|yi5H$WaT@ZWK@pi<^sm}#1v;n;xmOg_oHzTSM^po(ZmA8E3(j~4$TIACG}Zo z71n4{}9>wVfrDKf#4UzB7v5c8IiTlUIGkVHpAQHJ~G)B`E38tlm}K2 zCt?zND@*hz?qS3A(a{)zr_T#Xc5~#67s#x#^z+V|8K_9Fz_kDrU~8u}t%|uVo}%Fg zGQc)RiVdEA{YzOHOoLJ1?g2}iy=uZZM-6nivUi8}vQazIY5@!bR~UOF^qV<6IE=q& z{A4_OJ>eqpx?Zy=Vc^k{|G<+E3PIhu99QIR=+#SiX|uf1b6GEG;nVbxI@mY$hnkPX zQ`1oA9|dNsv8U>X+Gze*t8kjxS> z9PEF1(p`u}%BB-(>1nCXfjw*C5wQL|Ln7N{n~ykb+ck1^ricZU zbL)DStD^w(0gO%N`nn$|>-wDgch;PA-vQN>Ve@=@)+c8G!YU!AMH43rq2#3AddXkoc-Uij=Sd%=m#gO7MM- z#nK~JBs{5pt7iQt_vh0bU=i@gN74P2+x{9L_CMvXh%O)ix1rzxOuup8|GcRNXTYF{ zR&JsG^|I$*9-(g>H=v!G3vxb;f@ z(c9O~Za1#8uot{^cJqIQ^Pp|9pBX_m3+3##uMWjW^P9KDZMWv+IN?10^N3U5a&GBS zk0Hsw`!3%)P*AoT$3neY~%`*>}eLjh1y^tB*TAnDh*>*&xgXVSPJ32O#xkA%|d~?@j!}GI! zDgHf~hK4%Tq(%7h(p}cyUQoWVT+*Z4N}?1+GcCcu$wi?wM?2LqRo?Y{cbW?Namw>D z2mR5Z3Ma>v9(EB>TIxxiulxM~Xn*p=8F!yh<)*c0enJALS_X4?i(K}q>`8D_YrzH*g27&&sUz~F4a4EK2lkwWe6ZQ}UMN5nmfefI2~u82B6 z5`to`lQxNwr`$%^K`9ZgC*m|>PXhF(2EbZm`5rI(bO%n)_ z;(^-@Fq3Gnd=ki1RY(x`xT&%@6mt9d7Ek5?!-41Q%M{2v%Ci4_>lIv(8)8^4uUGl0V%p`_By^R!Mqg^E5T!xEqG;Q9|@R~OE8UkYSPzO5Jfv*I> z8fmPr_G~Pk>wb8c82^kwCZ&ET=;4FbP|l5{_{T}XM|i}fI5cx^>5ICZ=Fj(;obBY> z^@lt-jD>xmszz@#ajtxy2*Msa<=sD5#V<=Ex}Rn!HAht ze%V$n^sq@S#FTPCuw0LSjE#?C#xpmg%E2 z!XLg2m?`jP6?ESM+txKt88{dN^R|bQ`zs2JM&iRJH{n2EeE%xu)AL%JPX$yT-4+L) z7RDSh=VrJoX}+j8*Hw*dF)V)~P~-uGU_(9vw_!a|cU&ucsy$cCChF{7BB_LF8N{_@ zgRXKKZW4Jn!+E$xoF?tddZVHKr-CtQ@YXjY3$4PbrPNlgl#_x+#gCkp&SSN<iJcz&NQ3z z$6B8$zu3w=0)T35z_#MjsT!bK6B{Y@-7e_K=}vOuX3#>+@O+@Q=})E@i>dJ*+FC9I zPVFW+(UO*YiN`o&s(qUKa=Rwi`=>DG3BgCG&kdO2`|`>#1X_s5b#wp&%R0W#rG5Jo zFm8=KJuRwb)aHSz@-Tggf&63=Gu!xY$hyVAPXb9gh^~_ng`43a#~>N! ztVnkzSerVCKE?QS#aK(Gw)f^EFP6142se=9W0fk6LDr>e-C706fgzjhKbpFR$rHNVVyxFs|IzuCF|v!a`_cQwNOpY%7sem~QE+>#-DCNns^bP;0$yj#c6q*C*rwbp1I zWx7gYa7ulkKjuk4^fcxL)_WM9@9`)n*_kk8urk?pA%A#-pt*1BssZ#S*yxXab0AZZ znCOGUUlau}@z)f=&2h$TiVIQCAQ8v9s9C5I@NsydA~`8h+B(1{QsOw|6h~f z2Me2%rJw%EXL1EyGM%$WGo zZ|q7U(Mvs-@9^X}TdqwR6OSI_R2pg@a0pFtxh1tL&?J7ll2wNk5ZESUJ2K8x!~`#6 zs!mDK-TBY;$iEc=IiR#D08X2g?gflr^!93w9p3OpQ_H%&`jFllbvECCQ{a)P(tMBW zu*wbDhO+x_22M@LFp|_GZqxV1ca?a#MT732Ub%fu<_V`t?;j_SC_PN!^HR>IyH>If zY3%yBW|S+mCtNoYYfMZwY1iEPEWcJqIL{%dPKmCgx+fb$Fh87FZ&X1Y^lUmjBe+o6 z;;SBxGZ8)gZP+$zdb|Wp%PWLRtWYkzL_hEnZ-NObqQ-!~N%dz9ldPF!HFaY8BU4^Q zBG~uLwEgZ+_{`g8F~O2*y(`0;t+6}^_j#x323F36LqBQL=qr7aR1f;IYPRKEL_U(x$yPGSlECNL&whwnWRk^HsD^z}kHg9Dz2k1mw5 z8&KJY(}gM8QWvX*jWpSxeL2%=BLO9Q$I|b?p=jvoyRjg`Zi}s`D3-F zTeMP-$llwNOmu6xzGJBh;UOTlZ`qm65P6Jg+W3=(*#-O4dzHBuUFbm zy}ytO82l7iOQY7#6gk^H(4#N~e!>*CaVKo?=V|hJK1reVtLP_O>ppjP@CB1f-?i)p zTg!|BGG18yM_a8Q)H=Ac@}m_A`}yzbn2FX`G`#-dru>ItQW`ZMeEGFGrBFv%_}ZS; zCvWnU@M5K)b$@e1=fq@%d4fIOS$1DcKlRxtR#5!o_|mtncdEmk_*y+~oYi$=u@$lJ z<5j80?=NwVgKhs6u5aYjaB)t`J_7{WEwoR@709n8Zbk>?bLz>l@9*deoiTjKGCDRD z`HzDoev6pK8260@ZS6ID-Z?IQ&yp;4me@wb9--s&K>-$PdB>#rVrP5C9b=omg^R5* zk6ZnjP7L^ijCcEO@C{QyI!un%-PbB<@_FO489&(^t2tO=dE@oT3yx(HaB@{_>vBsp ze{$l|`Tdrm0*S??fIC^X%JGu6$d0eyft#a$HuA=p%_T&e#m{jVXHqD4SVJ_pV;imL zY$T!N>u!de`&tBQaVdxKFAF0@{rZVK`Z|pX+}X%0Hq*u1o6f7#xrMNL<<-8A#Ub3* z7gve)cAEsYno?@$@YoHxsjt3%3DeuFQ{jNj={f%mnQy&mpuJtsP_|voLJS7pcG1U% z$C26{Y>F|gkYad$UjA7B&_EMDOhUH0zFiTdLRVNBWn z%iZ<<)18)CVeBkk8#&X&v~Op9UUG}prO?Iz=AN98-H8HdMq`Lz(PUfneLm9_voW-! z=h=PylRTN~(qd4VX+?19KWAbUXRT0_Qs;=I-4Tic3J)wSK@ui=;$j0EFi{8w3WJ?R z))tRFtfNzd5C-X%7oTnQ<^G0eZk)JF6M6HyTfF4f z?3J>5ICI;%wiSC7>=Y&HHC8uy_}UrCI}hJ zc6@ajR^|`g+}Sb~v87A|Xf?NWe{Lu)lIKEeTCU8Zex*y1j3~&{Xa-sj_h!BG*?`;~ zwL^k4W%2+!AmpLHJYJpVb03v%jAaWN76SBkq1%uBGMK#9o@Q#MD@!NVaPSRzm`l6Z z1d18W+<6kr3fNcH!N|pszI~Em`NpRes_53*Uh0oO7-B`y}=zQ@Ti@h{Zzw6pCovsms@K`**vj!gtU=?)fPrYqZjB}R2R@mugzR+gu(H~EEVJE#aRx4++9FjWLw8Nx- z+G&P6Y*9igGM9CBH_;kyE>fQc{phi>ywjD_XdS{YCzLKa>s<5g$4ZF^?+3UpL%zY` zbU`F+H7##=>GR9k%k73(w@1iLWE8ryuL<-44%ev?$W2|z+Q#gvqDur3a{u%8MaW=* z8@^Nb%_-8;HK<6sMbcUl6qWD0u9J>IVvC7foEiw%n1M{Hcz6e@gurG{<(}YzItgfj zAyCB)VxkWdcYO;k4D}H5B;#2{@-C6U`mFGQ15SOhYJUI;TSHa#ENO2q&pchRB>E)N zs%}LTn#woNOuMy6jgsD_8=iwj)u1h;b~>MB7*Ry?8^^&z?YXU=Y>9D?gr#xoU^g!D z=n2f=Z)`hkeYFR}NN&*Z5v43LcEw*PeqEk-{qbR#JFj~12@_mkiAYk`_k7Q=6Ip^6 zdX(#_g6zeUw|ZfhYcd1oE1Y)THI?AkYR>->m~MA~8XC{-wnT|^FS0wODXzGFxcGe0 zWqxf=2)n&vPrp$K{pdR-48)9_2 z6M+j`-;nxsr|50E{rpcy=OjIKqbB-T`XAtCpb^91y`U=)>-3kQI6-xQ$kuiBc!xw&EJ*|TFg!>GNW#912gJiUS{ zeHgq~-fpx)BjBp=+#BDS&aot!X$ddb^HkvCH&2l#xaNFq;Acb}ip}Sn#5Y zn&Pbvc@ue25$)*Ry4=-743w{ab8)pqQF!kZ(^l$a1JmX2O!^Zb^?FZ3mb4}Zp3JqF z-UALP&5?2-yvqWM=^7ntaFloOHwW8y+O6(LWdpLIQ zs7b?DlyFNA!S0NSEdO)pp!XiGT)^@fQad|a>Pc_Wk2QLy798~@2^aE7u$<`g1(BuD zx-GG;^w7!@Yxzts;_?v;M8s3kzvjJm{)CG z>&hS_6zjhpM~Js{4v_PUj74|Ldi%A7vM6wGMdpVBXZ_#tv{zDeLFY6<&Zh zfLt4^9^*2Ews=xam1_OQrL{DdMN2HVf0PvOgFdDMAOHT?V+9i_Av3c)Q6D)};W&u@ zF6@M+nsp;>@kihFsI+4+>%^D$ryIV($mU9JUOq!M_*d{Y)45*Kb_TzbBmCh9UfmfxpbxP)DGXYgFT0$s( z=b}Vnn*(00{i7;aQ$N0!-~)nx4TBrfq3pjwYnSV=moCfi)W@~yF0iyKo9bThB00LV zL=hRmK8%qqijuM^j6(Nh1TRS0f6d%bql}1zLu;{ucI}6Z zjXrx%wwnXviTx^ZK4wE8YJ2|Zw5v^K3aOi1z zXP9NtE2vp&ae&Qa-}cZ)nK!eR(=uyyEcnu?MypH8Xq1POsWbpl|o;-LG(*{*(y6 zd=o3QcdeWCV_2l2cN^W4{piLmlxM4Gtf2SGR-!3(TAj~j{Lw+qkl?x_T?6urUAJpQ zw|O3aT8IN)A(G=7x-FIi-ImA^TDNx-9%F2=hn`M&X%5R6mz5cm3c75JC_2)$FJEL#^#-autvZd_~=280oSM^RaO)Ma3M~*QGBAt>9a6mcjb19o`v`t832*B6d+tm^p14#h z(-;#Cl4eQP>~U9Q4t+}`kD;Ak-R8ouwc7<7Y~MaI2IJ5(+sTTY^e^<PU_Q*6&>nwlZFMv6iBupTJ#ROAxX5ucaczN%!G8B~P>{_hdq9j7x4oyk_4UZURjEU| zg%N49cE{`0w|JMc@ybyGtwoI>0d|jo;pU!>Y?`JmG2cL;cj&6epYfr$6 zLaifaQHE<_Y)si~@nWUncW$6VMN6#j#tzISU8@~Fwpg4z>_u$I<`rIAQ_VN>p%=}_ zgYgFxEx!$(u<0l(x}raU84t}gAOW(>3=2Sz+UyoCi@Mp|XqViJs={Wy_Lf>>oa;b@ zBP9m{?4-U2pdEj#fR52U#sFN2)3NUQ2|K6OuslqUzYkVLen?gET#OMBnDA){@8iAk zrG!OS(G-e3I#y*De4$C3NggszqzLH}<3NhP2yJp|7a!}J+g?Q@hQj;J1jKgIB}8Kn zt<;6w@DdV}Z8eh`_lL5mrr~MDELK~mue}=JQr6g4h7>QyrJTJ{AA`=+l(o3!wMn3D zIJECW>f6v!Q>jL*SFP=ohqn3nrLZ&x&4V_$8u_bPn~+fD(z} zG_kI)ONWoq4yMqmlsfrmf)0m}LMe|5zwab~DT=b2!nrxu=X-)5tEM$nuH#cNy3lu2 zQr5RS0c{|zb!sGpMQL>KJbBp^~um@H6|+? zyz5G7F0JkenV!{)T=|gUGP8UAl5x)lOcbD|d67Hr{PZ zLt%JOkWjDc;rAM4NE1RtLB>kuV+7t&M=Y`(Jvp@XT-Yt;iJ3;O0Vg2OIrLjSSvlNW zMO)hIQ(n2dwR)?p)!WIYJ`m!DJ3fxM&EGjT8e5=ElJs0CZ;3@Wgu-<20R6uK^MvOV zQ9Oi%#EQa7b@P1Z^H2+W$VTl8lZ`%^Hthub3`4f{a1zS*$)X zPx4Skb4tB(*oJKtT!rOv$anFjzK5Fr*+Z(ki%I=8d+MB?g6zg1(wiWQsmQ# zq4{BCp5aLy5r-A$_E?RWs^u&;Yc%q6bmyr_RR24w1pNJrpTj}!&}2OAOSvB+?!o6V z|4!};uHSD+9d3&oYpbsdAaBYjpsFgS*82@Y@=)7-Jov-uwph1PDDQD1&WrrURPoKq z8=8dDz3u(2mN?Q0bx4FlBo9VCCevr8Z8(He^Ynh2u-*m^s@8U}K83{(G5ivBvMCAex+fefurGENaf`zKD%${RUlK%;nc@(bTbFC1SJr9}-(H`HIyP5YzK7njQjL8`Uap^y@h2_@gLKA&FA_T^9!lR0(qR?#1oTHY377bvm2FZMX^H0wjbo zOe1KVv&5X0s;X&GhKM+sxS2_G1Ng@yf}do1#!Vv)eWRM+Iz6(L(bc>8eR#I_B^HqU zOcOuh>qoUMFA6-uIp8X+BloS^9j}GjN*0H(xhc128BdS-4RC3as~3|nLu^V3(~<&7 z{e7C#LzdMM03>-WQXCG?*~oEYvRG0x&~1nlNuhDwgdMJJQn$W-!JR37hKCg~Y*erE zYo>ui-Urq4)|FylP1~tr()3BDpI>ChE>C7iMSLSle52upVfqa)=$`>EA4L0Bch-n`C7Z_}2tWs`v8J*Fh&MsvFvoCPL#67wmO?dr+} zS1aKgF7U*5&-InO6Ix8m6G0R%Ym2uSORJRt_Ps5|$mLS+Q3rI}I0(lPp!QivU2 ztH1inOfws94Q~#i>fj<}wp5@dPD5FT^GnTt-kkI-S|{|^I2|sezQ*+ zM1gRo#5D)u-kipmUL7eNfsA6SsrS#CP0pavRxF(hE>(@r*;8$@vSQ+^igUQ7O= zupn0-2KsvJnrX!}^Lgq;l{9brJku@u9C3x-KR7bgdEFx{ajtEP&?cjTq{S|?iQXo!by&vgkdeFH+55QDnzJBE)G6rs9Jj?cC)n(p^Xvz z;H?E`#8Vxk_00PLdNY-KM`GEH`bbm>kTqn4pDkSu&p%GhnmWZ_T_(Px!A!z%y#VGb zlnfau^uqPs;t;;w1&7A0N(g!I>^1+o>vt zZ{^ovHp~LC^!F8&1qXDC0yJ-FDhpf|*Dic}?3S#lH^|gVFnDUwp!e6Y*I#}mus-QK z!j9hh#=n@Z%ng0H4zPh#pJD$S7n~QhgrI|3l5?Hf0{30>!Q#BODirFl;+SFv$|G|L z`f9%tT8%OQF3_P4QyelcYwe7M;Bwf5W7 z{(Py#@H=R4b8_{n8On9TWoM0kwJvJAYrdNHOVeQ&lbryOHV(0mR&N#nYsPcu19+K5 z#g^R-J@3$u!sVwVU}7f$Pv)EpVPo0+S%uEk?PMr?^-VZFx-8zcK;+~538`tqZNN7C z*g94>fXdRyycN`!N7#UmeS0|9MNmb0r=Wv^AW~y?>HHv~ZUEY_#cd>1gsIVn*y*u2 zK3;;iZV>;ap1~i6%-)t~YH<+mKUr(OciEO6RqF`N4$+sJbs)579%J$yigLRS;lSLY zW;Z{yYs>?yogsdeh8ycv%|qKKP3v+z3Rh_9lB%s$qz&==L$}%mZMdAw10tR0x~OtC za~ncLLP3Svk;@>yb}Uq+`jr|IU8s90I+{8(jmqX?u(WxrC)srsubl)F`*7NyBp(@e z)QPpZaEjA(ehqVg?O?7HvM^sJHFO%Ucy>qdS$%)P?q$d>b_(A8>gcM%$N64w|I5?f zbI({P`{Yc;mCZeQ;5$B_0IVP0*lpG{{zJi(upeUts!P1@Bn z-RUYbcZQJWiElmfuNN2U_LIV#%}5PlvyyHOgu#yi^w?iIyI&snozXk`C(Gky;!6}J z1tjQ?YBgiZt*2Y`Z=rehuT3>YxZujhX0@shL< z$?Nyu?A^-M5pv3+OpD2cZFS_OjwwK!07Npd&LDs*bU79+2#Yf*Uk!X|`%QHj3`1!K zMfbOEfgYiWI)5g21mi>h`>&2Hr+QbA7UW1hhX_&{xxt{2zY&_=cAloA&vB~h!rZ#0 zVJHhMO1~|PJtbaL>RGDdHCr4Es~xZ=P&l!S`%Lw1UquQkT;jVHE9Bp>MXfr?=A@@= zRAjfh9f@mT{AdUow;@|@XE~dr`AnAZ6Rgqvsa`Q;1wld{5;nS87pbk2}>l*nXAO*kP}d7MnDyL#w&zya(`r z1k?T3?Vhlk>>+HCD43`Gd{5?3Z_E@4MA;h>eHE*h8Z%YXhsophLA-mYS@y$dvUrb5 z{b^z)!Nm;JJ**R}R4VaBXY@~l`Xut?B_7dXd7uYLeet@mD^G=YdkIE#7uwRAZjY>= zD6#CmDB`skhgC5#zb?jv+l#;J&?Ol#+GBZ}H2m;{g1>97$=ciJ$|dphhbhUv4GC0t zs_X%fChi4HuS(3fU3J|z=IS3ns2rLCw4F@b4WMrv&Exs2^VA-j_bFfE(Rl$0J4gfu zO1$!HNY+TCJmeF*E#JOe+V}4j9~AoSOj-u0{_B(cvn&4KxcCgmU(NG-XDx7PsphRQ zC3w@WF#3ebB`)vIGIS4YO@v)-O)}FNZVSkE${np7VZ0NWDj_!u%rWq5&%b5!<~2qL zegIY>jYXcI!NIqLC5pR@=||mm$V#`w@TE?6%FRC_=+M62u^gIx*cii7WR3*9Iug#4Vm6=1@?rBB5JI4|-C@Di`a2~;T^TsPlr~>8WWD&1I ziT^y-i2ND1cZhWlDM7-&zkJ+oqJera(H@$D!J!1O@9Hk1Q0 zYa+n0&(Zj}HocUoHQD#|YfCoIf}XnN3T5$Fd|kKVkjHBi0ax*}FnDCsRsa>TU;0FR z5zvzT5cFK$8yM=WT?@Y;e7sKcJ}keM9c0C!P=D_O%0%m69+b}>)Fbb^Qso~7E?|a) zL?;%ond0tC`GBJ#+h6e^yB~4P@8~b#(8tTZuHW*vhu*vY=f_7k&fL;YF5fc0M%=c9 zs4SVg-xN5cepo&ppu&=~*PAtS9Q{5BHn`8Ng!37V-d1r--7CE2rLn`^ldY;MFi{Ad zTYe1DE|YZcbHKTxRvG?~ux|6laU8@UbCH2pRKtDz(O6jXACJ2q6vc|Mh+gSRSF!es zh#w_BuZ2exOt5+JMOc@v2UYrvFZ3zzaSJ{f5q{*>kPk_{UypEHnPhiE+QJDS2ZJ$# zQ1_hwLlBeLQ`EeaRtTHQbm_amWs~*@bsybAMFHc=ixTF;l4Qlx5XteynU2r3UDa|> z5!c;v-E_>#>SWs-PyI%K1zj}5lTYef#ozT}c+maItQ*={>`?9n22Z4juYbRhup!ZU|IE&( zH)cis`$Am1F7Jxn(?4?QmwxN!8Sq(|TsEM^x-EnO00M)r+q6B|>9ROJUxN2F`86;0 zZ`}5RN@nG;-3Q?5&#hEmePpoMN7t+^22b^bI8>qgW`F5J)sCZ|{eAJxH99``Sz5a^ z8=6}_3z=^YLL|G|Ews(h<#Fn`zCfqx79N@eFODHn#~T1|8boj#vVijRc^L`RSOAb9 zHm)_Mh}U$vn%#1c=1b8M)Dcaorg?z;e(ssY{-MsF9tMacO1fABeR zv4uBMsn3Px)g~{o)arQ<7A+|hdC*kcwWCa?y;2_Y)VqJNx@+`DKMMdVLjW5MgMy`# zUcTl>czZZ0_?ag?TF(wC?vX`r)R#2dxD-+m;5#qxEf~n5ObF4IgxP`f#w2 zW-P+6Ln07Csd8Yvb+8KOdrRhdz36_|I1ye^kJ4xm(4P{~dhU6fOdr zRxAe;!M0oI-Md;{sS0?EEJ>@t(M7sr`#$XmuPWQ_i^FLWj6+5;e$u4%xhZPdKIS!A zQ5SSgM89T`Fhvn+3e~kK|K^n2j>c=zyc!h4+PrREw9TaQJ05??oApb>kBj(n( zE1?ng`gt=@KrsY7>~E-6f_G@YVep3SJ{$Epz5P(F!yUkOm$4&=Y7{4yjFT^|B=nCSvm-y^U*^sO5MG8oMB7cdS1)2Z?fd&A!ouZ?ZPiY zp__7$($LSi4X2`Wau#f94$Chdej#{)BbsE|bN8TUl+H8o4Rn7@YQEvbv*-IV#}4pa zGl@;^?zHxeBipk6epD@icWRpT#oRst+<>&$)@Ge%-FhCOJ(sS0g&pu_j$2Hq3BV@Q z>4oRLG7UGMKMuPUtV&Q84BeRs>I-M5lYLZ6RdcvH#aNu-))0O|8BiP|pMCbGciTtt z$Gocs%mO~+rpsdM`fmI)f@1sM3#1-{!~hWZ^cAbYY)DG9150ee~Zi{ z(l2H3|NEQ_t&Z}^E-Hfwc68{|K8Q#I^bs? z6&HH=zwhg>#{>RA-8}K*PcMLfi*En=A-iMXB|`m+TJHS4J9&BrAjl`&ZGL-F%g<*q zboyZBo;Hs*{wvP@$1DAR`$IBVLmuas27ZmL^S{5L{3>{f=&Nmf^>6=L)W5QH|NQU+ z8D`}p3B)0^;oALQI#&Mm!T;l8-okwEo^@v;aV`AT0E2(N_OAfJ9||`5dyk*_j2hSl zNnC1w@7*823*MgCImN%VhAED`0_r!Wu-CuujPj4iDEIQKliaIcqeT3*@&4TG(oev$ z_bYtz`wkR87YMM8cVB#T`@e3#|GnD&nQ0v70IT-NMUKDu+^&F2ze@f0qV6F|8Nl0z%`v2_@|FX$D zxtD8^AIHqZtNYl-H2 z4A99}rO2xsy_Ue}1@O&JW6I!YfD@~L@$1(CRW2NWG>^rLyt5;Lr)P6?cRFB&FFeb! zcDqvYB=}b`9%fKzdVRD!Us9YcMss;5pso3rZ`hIHKZp~ zRW8|oH1-N8TG|q&pnmyuQArE?_YOLzsk%W-!s20H9*4>U>0+21zCRK)Fo*N^;Hkc5 z6c7osfL%VTuq$`nZ_hrrWOnnLNeK0HnLnZr&sIBq`R)_I4rD*@^E{H!o*<+^_DVV^xnB;vHLUAx9NY_e#OZxMW;P8KlIQedESH%N6fwFG^@ zFDi+Wan<^^u8Tt=(P;b6iivk>siqBpG6wM|>h|Z!rf>xVk2#hKw0!vS^*NlDUo@Pr z1_gwm-3Y6-mF8y!_m1wH6%@wZb$IdEjIt%(vGd51Sba`6@Lp>O%T%3G=dS~-t2=dt z-cr~k-#DS@ibT(q$4H2|c!5N(F*!q5Sla@@o^El-MoGC9021)9 zMk)y;-34-F$F|TRC+M`haaGLbNlBn4{});&(H<)OGAOpMTd4A)a&*$F>U5Qe46(Bp`e3K? z*rh@Ek}sDoYRE1^%21Ng9$0ieJkXrycCQKnSq?p)P1M3*wQibnB2SDhI<$gk{8?l4 z4gUaP?cw&AudATPMRjUV<1U24FZuV!=Y73lFZiuj^~XI6O4H`B;C997xlB5?rtc0D zCVj=0uG@iPi)UI5Mhcu}jvwPs=ciZzMolrhcfHG7?nJlO16u-BR>e<%?}UABF}n5r zap*n9XB|7#X5KVPeM^?CPqh6an}^&CJ6y}|nF6l$mmGSWbsmZXclt8(w&J1Ei!_A^ z)~~krvrS)#qyrHmRqW>eJZG#3j#{*TLirQ3uNr(whJFBz4&!?I?O98LSif0gkTuWF zlZW|9tL3g&0xtdT!BDP$w85PzkUtSHsOWbHWuW5qQNg<}k7JVNHg!rugb3K?glaF- z*1?iwiuo~i+Ws8NB0V<1Y3T-njh?PF7wgz{@9ybVKu1x_G|Wp#SJFD^A9ST4J6jFG z#VcrIQLjJcd(0+2a9e{L#IF+ViA2vu5fP{90BrZ`$o}Pf?kJ{E%Accl@ZmXjYgtEOA23%4ZS+nMuBVM@<9f7qOr-Vx1#sy7r zY788^wPt&$rY)CO|lpcUO5R^)$geP$_~ zQ}d=tbAAP}9e#uZr7nl% zK$0&_i=@ZOm(~*o1@~UR9OKn5RYYCodjP25=ks}hc5X#ubvi+j2sj>F0m+GYW_WM@?k{*^tOk1RvKZEgTGIp(uxs>WAa!z2p!AMs5L;ajiGBW_%?+zR=h2#Sfa z|Il(%lA|1{^4~0}`LqYy14#-1O~}>W017C;-dTNv+yjakTqNgROer#{N2a=yJGlm_ z$D{Vt_xr5?R|R-swZ_-H1i9^<=-bSU8w0vMVLuWMnyW=T1$vFfZ~)0nNoVQ040F|k zx49_^(ef$2qdiuF|MvYAv^||9&ES0*^GIrZ;0XM?It5^qd;>%_7des>W&+se8h zdlRPUx$GeF^7V!=#`(y5%7Yvu$(nSIB1Ct9?XA*kgr|ojU`lg9R$Zk`HTU0_BRU|Z z-b|U0I)Y}of93%5R>mEC7J(^oV)>3her4Rs%x{x~8fw$!%c z`O>Xhr1}zfqEyYCfu8E+-Ig7o!qO?VqUNF=a7y9fF1bH6fYs0O42m|sCR0;9a`UJE zY{0|B{fg^JopGT1=!C`f#<(*_lHgL-0o|7o#<1s0hFPi z&U%rCa2Z}C$!hoMwSvgS=N$S$chg1ZQosT+%hZ zQ>G%b9O7lMKJ=v70L3)Z7df)? zffN`E6#Ms&5djGmM-;zoX0WpP07i;%<^d640QK1j>K!0ld}*EZ@7e-oG*i zGrz@`+GP0=QZgDF0IFMgWB-HY>k-IzRMTga)~JhPR|b3!2)x^#ZAr@NbP$s)X?+29a@}hs*7USoTKDUHd2>k_f;#dH3!euNj1HYDS`V z{9T(YkbJ~HP+rY-68Sn-x+y8vKy?XC=6&@C(EFuCy~%i-6lm`yDChKX4`8BWX%BQI zS^*|8Z7p6vXTj#+zGM*qDwf2bS(wpDyj@MG8f-K<=8wHq3PUuwd$$4F)>btK)wiGi zyi>UeFgeCHXXDy`r&0c z++LEi$F9gskx56T(}{~_jn_x6M^SQqt-E}2HihHLpTRDm33VDwm!=EJpJ8@t z9eZH~a(>t+dVa1qCpI_7;mxnfmLZ|pL^ZI*{f_r7wb3Ae0KQyYz@q5It2n28x)*)V z0@!eO18|L$vh0H~Rb}SuRyhIDLg(0iJ2`hr>nciD+~`;!ZUyKpZhHx!S3r?Ix#e$IX-gzFeUhAQO2GLt`m|a;IJJTFei{3Kd-Cr zSwwrE7y1SzsvW2QFi|iep_OGU`uqo|$Radqb5!akCDKu>%FGdDwiS`AH2rMz!yVah z(&EDwE14C`0JZh|Y;}#v3I<|IqTFGr3Vs#ISkvkf7TdISWHjrKQ^`dMaPlm_->j1$ zdQmT#(|6FW_tSfFs_#7+>u~=QB>wIHq5~T1%xGh@=Z$btRJmv-Q}RK4;Fs_ z68`{YEYh|$O}Hwy>OrX&NmZ#^>4lZZDK*y&e`+V#ZF|dy&}eJD4(Ku6n`{bxno#X? zGWuu~v)l4};+T?-zSqsS<3EY}h0Cm72T9vr=dJ&|Uu3s=eDABQ#x*bg%jdKrg`(ae z;@#Nk)u^0WvPz4NYtgSjI2#)31x@*V?Uj;!;^U+6m} z>%g?e;-xNc9i1Gt_4CrITn!6Io%$G8g@KTPc`2~xQjeonPQyi97J`A3baO_3wEaJB;)y7 zjtjg#=8d8217Lj7j)p$?Q_T_3!jhx+vO?=k9*#v_&=KJEST5s6_~mpKDkB|^#xWPS z=|7yKU0b3Ng$x#bM0=7krrn3<3q`R$&-n3^1&;!Kz@$`_-!gd(fk0N5av-T))`rlw zGme6KCJ3FNcsm&wqBZVX zraxksz2<>`qkqtm=%c;N-0g`MGH&a(eqmG1k+DF7`;Svj58b2iEf-T?TgwwWR;1zW ztCY@$Ag7wQeBxCU#VO{k@IWzd(kp>(;R1)N@s8IY!P^DH1#51MZ97VvS53hA^4rUG zZo6|U?}|eG{Pte+gp{dOi$36_0luSVwEt_@AD16MiM#&VedsM8?c4oCD?y}4fV}z* z*))%ZLawYD>gP(KD;_xC@;p7KzJV2uN0TcabQ1}BC+$c zd)Q;g71@_&o&yTEJF~Hk4_bcm+s)9ULCWQ}$hqfEP0 z=H_nf{Zu{*&$`#0<_@a_q}GznCDW}oJdO4zHm|F`9Yw|h{wMGXhcR)rt@S9C7=KlG z8PM3QZ4tNx>iNqbbw*tlw*tayuy#e@Mmfb6yq3+Yl{NJMioz^8%*SyFkhW7<@aNai z(D3IU+YBd~gK8L=)P)*qy)5Hu!*Dsm5n>zPiyCS9;$JQE^ygs;+L${H{wTofEuhcL z`hHKgh?obF5c9m7`sc(X@6AV3Go2ZyIze%j3Ds~xqt;k3_GMh9y1nZM+tXlYew8Q7 z6gx2wmhI-q=)$;L;Vyw&6aWAffDL>_&(Ev4I%|4Y3MgNvBV*5Z-Be%vRs2wnvx3LZ z*S1gib++E0Hb`Qwz+(P8aPQgABUD1l+(;!}RRT$jr8!L%-jcNt0Xv`Tr=*dUqVkBS3$*qotm7^= zveD5vSJ&khFM8=QA6!C($4-Z~S@hPuaD0|Qys1=!y%65&&>Y{Elv>oL>mClnd9;ct zwk0Z6I-*GuEaZ}3mctDF7U4|*)}wphlLb&3SuLt~03lPh&{N5?Qyjc&5Ae4TXgXGhAISV|8m5SDW*kRBCGVSJ}`t=xV4 zvvusyP8TSj_{O5tq}1vl9h5KT>ksG+_AJ~a(b9-EMKAB-0wudPdn$a*O+84y0euCv zn}@&PXF>Un>v$2x{vE0Lm81my-`96g;#mOK%lA=lJB9G5zAIlCxU1{#UhGH2&}OQ} zYbnwKRSTnwR5%%5J8Jb}v)f9Cn0VJ2l8Q^mx6?VspYLsVg=W)D3JVO>`fU^KOlj9E2W;x(< zpwAKGyoZc3EJYd_)WY77Ljw5CY}SW~!WhXcE&8%t!>Ao)tCTRj;P}fKe;1G~Sd86+ z)5X3xp_8cCsccZ!Q`PD}B-?4#9hmo6^De^4oN)qW`=q(Rq#m;D?=*$Jg^eyHcfF|> z@5Sfl2kbdEHhr7?*(;F2q!zV0Y-b)EBEM3+H|dx!5zm}y6QW{Q z!37`s_niUD(66A{oyB^Z*VMLG3qM={yj3HOR4HGM?Fk}tb+R5hPVEe6ant%}cxkxQ z{Z5n@cb8*-MsUu`5M}1>>$@Co(8lKvNwp@wY$!%;Fe008`D2Bs0~cg0*A_*;J?EfB z#9wXX_N6*z>}k_eW#n=P&g?*YSfS1Zb;>=y#yj_Gp8d-iO8|& zX*)LTS<|_%DnbC<<;5>I+ka47$_!U#Wxi-2GwEz{gZy<$eNFrAAUOzWah1Yz%A>KeP6hX*Rms_A#u*5Ny;dkgB*RL+%Xh9 zen9TD9?TmHqHekVyi>EZcSh-xc!0Aeb;)q=bO>wp1viGc|8i1;d*|zw^`0-Hbv>1b zwG-qdC+K%xU_`*%YEJ2!3o6$qP644x80cV>+%mvso0|N0XUStN_56 z#*KECuZ|#M`oO);n(D_%%4l3BC~sGx4m6$$QPdJ$9i?~vMea(i3EVwMA zacWmM)3H1!AgOw&;Kq_^{UVVNpv>)mq8=QN1)Wi>&DtDsZr?@L3G}U>$9f_S#Nq`z z=DugSpRn}65k}AKIJO@35=q#jA!{&8)^n#UzHaSHyY(Wa#$*2YyX0HMF&ige#KfkD zd!{75RluU$?P+58&Ar6(T?Npa1gPiU`vSv?HL{j)=3s(I}q`KK5AKao%!}e-Kmjc&Hb%M?pa6s&NNqX zq#*PIj#0Aig<5z1l!{7*J9kTg;LV4&XlB*z{s)W3{f8#t_u^1?`ZX4jBJukr1i22l9=Sp_W&3Vr;Zi+oJ4UOWqAir?LQccbEj{Q{g-f&T{UbaL` z6w{DRqOW@>Xr_Q;{r;XxV?~obtP_fwGaflW71&xh-Z@i1zRBF_9o;+dR7W)P0vsMw+h*l zumbDEs7eP|LVZ>$d&3`;Hiz$xXjCBK(sz7ZFvkjsAy?!ahOMwR+*VM_CHS!mk}@Qo z3*ez<){Fe5M;(#uq5Oa{Y|>LymBx#S$oNXGg}tX4-OmD2Gl5<3JYy`z{g9ou9OJRt z_9cPWDZ^WIbv3OHRgVIJ-loh zGUK)-fpj!H;R~%))cz<){L1_#u-)m@y5NV&i4K ztU@1khts>(ls}ULv{>t2cfjGZbvc&YeB0%!qwmrM9sAa|`~3_^9BIxR-R4yKq>9(x zMxHK(7WyRr@40_{LqLu{^j-pM&}Lavll2K91ZY3jELWX>N}VqyDEHnD7hb~O1HGg! zRP)>>R`H;yHM!d|` zz0Nd&Ey!sWe zE=2-MIWkse(wnV%9Q#+Ds>48_#(L*ezJI3Ws=%M?O*FlH?%>s|MdcnVAOv+4#M|1{D}Bv ztrx|d@)Cy%<$53k7)X8$%)gO+;5+0V^ckp3n|z#+NE z*P7ui!|)B-UX;bkHl+c2Xk}6y({847b_q)I{!CI&Pt@|oQIhJKu{69#;cUVK#Qe6% zkK@o!T6_(lSKAz6$aJb+5@k4PL3gL}Xfs`K`qzE1>Fn)&AoeM~D?542oewS%nKztK zej0u1;CIS9-|rVEUNLm6t!D1LUG>sRU&mKF)5c~rCVh9bUXUC0-QT%#(E}8QSgDb0 z27Ep#jCe$cs5+53ql~%capIw)%0Go+Ze^q*j zq(!!FPF#JlTr7pJnG@^PL(_JaYozw&$;ijuvt>+}38U+;HM7$~ zYIy1ZUnGj&!20A2T*R-#!V{jAXU}+CFw>bh(lb^w(l2EU^!A3^qdoJOzS^f$Cow|o zn~-aoYK%ve({yTrnit&ca_Ms?pX5j@W^egX@k)tzSgb?VpF4Badk%Uiuie=xd?`Cx zd3D+I){W!CkT}yl^K#2tsg6dT{oxe*vtjY#hFj%aUtV=@9<1}-{L5=AZn)*C8sojn zk9yccihgKB7J3$4Rgg^>!~(uuaD_qXJ8KeHg$qNXkKaGf;fz6F77dJHHaFXL(@xUk z3!jti(>Fo_Cj%}UqtShw$xvkavc#xOCF6=peYtCWoi8m6^4cic5kie{E}nXA7zs@+ z(oHQy|8Ork!RQNjSW$78)1QDeB0FX4sh;iz8%B-3kb;Pc%i?Jtjw!08dpR%gUzW$SIA!s-uGh;G3jrQIhtHr_6qVu*Wn6V#l?y;=HvAjc+&~4Pp%1Ld)y`enI&(`8oNQ z5%)-dL?cPUrig z+GOmbGQamb0(L=!b?FKQTiUn(qt2B0)6umMwZTExsYNHR(k3;ycVpm0vV%G5Wd5`2 zfOTrL{VzB;4?R8Sz^LXiqGs}_ldIk;f<}q`ZAK97J||bcAJnJy_JXV#zC6W|lZl$Xh~Bag%D_jv@4}f{6<_EUTqkO;OiB z)q}uy32I!O&Ge_9hH8H@{LqEA6*kSqowm?Jo}J>o4c$-_P~>3^*F2Xx54+%8K5NdG z(hqhNGs}7EBV>tNg3cH|B+BO``MY7dt7G!#+|AhunvkW%EGgp^HtJaqEae1ua!3GY z`GP>g{PS*m>bX4|-O~M*^0x~`W@kW%=&eWR&S4hr(zeX z86gdoUFA+<-<-MJ!&2}Yu0OCm5WYx`8l+?}Zy-q(^#tl$fZ-bFp5PPBue%I(S~BL>?K$sgkxx_*cR;ywJ1!Z~ZAi)41pPEK$d@Ti@?$?0 z;W-SQR~bv*hF?&V?0}CP^7!iT1GsXU)%ox%3dT%t3oy=k%23KVfi4a~PChv60+Xn;i1A>{oX66giVKXYnY1O;O~r)@Rj?1~tHYb~6CJ_p!F{pF3pF1hR4J z(v%stJK>=uK)fFvph|>@8bPet@A5Tmb+JzOW`-ooJyORsPJQldeg26P|4O}3TqXyB z)=dUBsLp zXk=uIH*;F#*#C-d%SPj+5~+|RzpXBF1@HT0ujJ0w%K5GZ z42d?1wk}u2-1Ej6NiI{ER{a&dj)H7!;_KgA`JUeGKfVuRUR{Fs(Fk_>Vdf7w$U49% z>2ryOc(3U0enz!3IGkfq6K@z9Z-L-8|I6IcruywRx>i_Z*+}2#%WO{ozoH)Oe19<{ zSlPJJjdb(fych>J?@r(aw*!rojcX5KT*LL62|L)Kc=uyB*G=_ZVEuLd5z7Oe`}0_# zqWGg2y8rk~UJN_L)p-t$&aub&+?HfW=A%6^NY15vU%#@}4Ox0UH&IU}bhW;RlxEfF z6`q%PAf`HztIrpQPwc7aWTE1(BhlCn43Z%?X?sT%$Bq4cu!5cg{!A9#8DSynXAjJN zrb4$j2^uG>H@Ne}>62;}U0(28cmI@Y24r0JN~BF#GCh~J0MG$w8NCiZ&le^H)W&#f z#f=3>8oUPW89{&wu(D%T!d*k5KS&zXI<`3^z4{Be_dQu5zm6+HH~i=5B<6&VF{Z^1Be3;0$$bp(Ofn}{b_GS~Nn#Vh z$L$XnzV#CcKB0EoBW3cWM{*YW64n(Md0+g?O=SDj%m)`q+uAK)m}f3)joH3K2fw9P zV@&8e<@Pw$xA!BFEcvsJV}fRUXLMY@{+f9!{h z-{=G%RN?D#oNXH3wGo8m^Wg>pewZDow>K%Ve6u1bZh|mDs~6RuDs>R`s1uV4J3S^9 zCK7Reh<5g)Fn3LeBQ-azK%A6+i0;RJMe2sH}UKj(59KsT2sXM(!buq}_0TFVCpJ3sFyxT7B<7O!}{`1~!Y zZL>2`uDs$dnt*LPn3W%BahOv-#7JqGb#f*T*+b!Vn!n?j{IYocBqIxV;y4HPL7nfG zSanLs0t|(?<&^!cU!mm8X-&l!9z=&hi?cht>pV5xolen8 zFN_fO%1ubQ;vk~YFf5OQoxIvlL&wyQLo9SF3kQPJoJs=D79|sZ<+0IAlG2c9ANwsU z$$%XYqVLYYg^DZV*Hm2!52I7CKH?t>t8+_yV)BjE#4oOnWkV!H_C;{VL|&XUy^CbE z@bIy%;#3XC;L+vyc;z9&{Yt4rvqb57zeOJx{VEmgBl=P31DHc4d535&t(j{k;Zc~4 zKsRH>0J%mw@>tMwa^ufLz&7+2GWZ#@#-7X^UUM(BJ8X}OK_1>UA5RSzB**3rduG1k zW>IFbyjD_EPUDx7XR`O#*C3dko=`;Kts@nRqpQ$AnNve0*P&E|b(e;5;@-V+6Ppa^=Mub28 zz3|r%#CvDQZX_t~2y>U~%)`|0|D}^C7w`1&&Omd` zYMprAg33S+6#G)@aoFtU_-r%Rjk#Ie#vY9gSbGLbqdRb_Iegpu!`OjzAC$72mBY8K zgpkdPBY_V3bkTm0S?|z-*xq0=UxpsPGSTm%DW3)NfV=)vdWoIW%UoD~%-o8#MWEU` z_ffRA9{c)+?6=<+*d_6xb&{TmVd*2a--r62Y!jDmu1oTH*$t~@r(_>?K^wn$y*f;d z{>FDjHQSH=vL3xC)YS9zeB;mVV!#V&JC7|rkCt_2>+SW)HQtSaS+rjc=h!ROQ<8okA>q^c%>`X&U+0tIItiab%ysz2LjpD)JNR7h z5R5Pp%2+6Kc)@ijKZ~;=dqnl+>)i-y*D%1d7!Z8z1ABAg%*%$S*FCp3ita8xAX7wk z7_qIF%e0$@(y_vyx2anTtx?o>`I7MBCJT}4-=v|&Rc7R_YKHBNT)Fzj;LIFClHCH> zVHJER$)oAuU8YTm3mgMtpXeW`Im4|7`9O@vGyMQVWZdtU+=>3m*AlR&`b#9CL`dVQHYQT0h4T z9#}_9kaXiCXohy3zoJJyNyDajMz=*9AVXoYJqRQ4>r1?t6ZKEe44pYvzdbArBv%)I z<)p~eX-%2MXl{EREbEQdtc5f8Tv80C$^T$|9*qc{|vt zqI(1P=A2180#)K@{OX^h`46Vbh5OImk}oXyyO;g_&@Yy%(=jmA0rUPqz5t_*GWV7U zqa0l!xTFad(#q&{opMOb?;xGX z*sBo*>lQqgU$BRo(YWMg#ZYXK2y1{Xm{D^Y`B6=*j+p8*$sRb4tXmlFtCX;+Uifi| zPJe9Q6)xoi^ZVuAN4ZL z{&#@(k4d_D%yxz?I@a7f9otuk?nJNUr&{Uol@nYVn=yCA?@AaC*V-9~a2j3Ra3-iO zglz0)$ESVMCtS7on~ad-;h1}s&zC=&KQre0?uA21gyp3z%0R14&n{X8_3^GAj{e5d zeYn+TrF&T}T+#^nQ&vx@)4~;!04ZjwIoCeU8j}hB!Kcil!{SKsFG0GUov;;j`F^+l zNr^@er(&d=8lU+iMZ?!n?~O16-CxY^ik_*%(YcU=W|Mukh!J$OB)}5_tS5muhd{yk z3zkN5B3BZenA&V*ie37J3X0$*^&2Z9$$|CU0y1A>YTi|M;R7@bjJKY<6F`tsC>JE{ zGow}*S$HWf-*foc9=WD)$}|+dDZ)Pdd9=)eyp_QvRdhZ9)Z6|E$eh2D^2Z2@H*p`X z^P_1GgL{LSZiT^Z3zM&2O!-mQz>hx-+f>oYEmEX+$ZE9ZqQ~1P!>+oJ=kEt3k7%J0h8^z z%!z86fkY0&dp3$_8D~?1xaN>Z1m#aKJt}uh+-Ei((LdVt*4BwPA(qoY_JiG09z@St zt*t>Hme6-U+RtNpWjrA~l`CU@>vQB(F#8a-(gV>It~?y8v7(rhe$;PpXK|&TOOM5U zqL^PnkLHE)P`3E3K~}x>_&3%Y%T|wxT#pBWxBj68@cTLapXvHDd-t$6VR2h-27GSp zaePIGbsoF0YeFvI02I2}QY6Lj#^ld1ySmfs-#Q!fi@fSDp>(N^*jJ%RQ!p3rST1ig zqXe)?w?&&eFT_YL4e`U*6SZDs)xd)1k(bfWLs+*&WaAAy5@!*RmHOozHR>KxHKv3n z4CPaJ*~Rz7X$74jLKYXy|L$FqJ;nOk4;>j5MS}~wtHtkEy0!Sx=5j{PK^0WW zW?jjAWe{7_ba#l_z-XVxu^YB|70_F2NC0I!^`KrkQcqS8DTb`$PwD{^OvkqMxov@m zai(NLqK+2-8e~vz--JPU`md<+qT~U)+ICnPgByyn=+6MMYMhA@TUdP`3!VOEK4qv< zPw+b5>v2*H389ld(|w$GWhhh2JAVCI(td~!b_FQVEz0<@3M;iqCwRwgIPX+?fxVLE zn@G_g45N80a|PN*)0FQ9uXamMs?0VmwI(=m=p&bbK1QbdCzO4BSF*V80_*A^ zb$F+P5QW*;KcKm-DCl6278%E3h?}z5%uH!w&yJpjCgs zx#|pmU|%Zz+tFlDWKS>}TGiG+4LezR1Asl()TS;QR=ZB!7IEl5*_!aYg!8eJS0r^z z)$mv4RWmX7Hme=ZcEmWRFBI80W-4W#JbyAOJZCDp5S=X%mQe8Ar({XSBxYKJEGXx2 zl|3pdI70SRW)Z`|h4|T0PjxOtjWCwhy_uUy^Mi5{7j7Q}K_HPFdA&;$I{fC)X%u@H z7C$2psQR5sAmDcQnYdjhTW{d}<+AY$JW<4P|6b^<(+{VYRg4z32%Wp&x8Kq;xT5mE zS9k3+j;DAt>;ii=q5L>ia+vJl4G$!09#wenC+=Vxj;qB+IE3KO;H*TbF_bb`xwzVT z$+IwnOWQ@F;3F>`*Hyb~4Z@>{jH`zue?t=bZtS_<7H$L;t%%>^|3Ch@A%F|;;rzN@ ziq+{a22|>=H8jd_1~ScF1CK!dfT6>N|*pI<5@)t0p%Q{V3la zg)3qLQD&>GPm8abDN{U{K zab(xrl4{ejZ+`he*Jie}q`Yy0llfgb-SXFm3&r=CXz_rn*G1&KZmc#r;kK#VoSSxY z8f1ZsSbLUFA%S2%8~WSRgk%FI@0yfaH^YBC*Khvw=IjqNB3vi`N!8ps>{5;@3bw=s|z|NXjue|Zvx^wYcg(GG7lm38Syy_TK z8VXr506N_tW_>3T$Lstm3t^RAge;3OT(C?IOK9DX-q|YRzxJlkq-FpE)Y9QSN{ZRy zp?HwhYxwA=2bt&UHboWre-Ea>En9K%DwFU1ftB=M5+av8f{=o8gN&q9ev5#ih`FB6 z^EI5~fNB!;#$29MoT}xF_Q=%paWu`}gT5TBnlpPC*ae?h0)p0_Ro$BzUqew>P5I~q zg{d%Gq6SLEuI;$leAAHGs6XIRyF|Z$)karfJs&#-Pxk}bQBxv36>zja-h1w$8o>tc z-=jM_HX_i6bp~1lsl;!in-s?%VA3F91nnQasaoeNgip#%>$N7M3>Qa!aXv>_(R1NB&H>q*nT zUOijR1~?;A!~_sn{hXl@IEr&jgAX?ArWcCz;#q(d%`aj+b0w{!= zZIj7BlLXcrgn-Og0Ak|N;;>=p42Fz$uay6BKnXC{+1p%-Trg4^4Nl^dhsrZ%oyHcg z(rYh`p!@TeRynw@APBoUt5?|y3gW?jF?+YJ*e98G!75MaNwyaU;($heq#U6r9I+LMRZq2SAly;^? zeB!f}{C|X3&Rfp{nViAVH~zGFAC?JPO*PCqT;g*rdGGnOor){dPWowo{H-Lmj|b%C z30$Qb|B7mN*k{kV=8&z+UBo>{6$$+O$-n<<&^i7~u%CWbA^&Pg|7n^3=ik2G0}yVp zB24?lKf%Ypx}*Pd(q8ff+6itscl}L;=)d`@peGlA32&S%CH8x;`oDRe<8?ql%K+Q_ z9@PH<&i~c7s23l-6w^qlt0{TQClmGO^xS{A}yUBSxal$9B6;aiT`aF_CHJ;+(+OM zJ<9w;U+BNMfqzV*dnLeVfAFU%c>a%`1HjZk=gXO%KaueM@e_Z47nfRrZnXPrL*&8V z*WiCQNq}z+Is|0%ZJyoa{~z6Yn&@tHZM8|Y{PzjXyF)C)d&WCu&r#UlKi`6#J+8U`?RwOgMaSZ8_eAfC86zP8%pf5n48QR=cq1^Hq^(&9jefQN!j=ioo{%p;tKYn<@P#{ zL9Sw1Z))}QGY&$?&N4Q8N^#jMmd|_DfAod)$neSuGbR_4tmLh_{fp@E_0v$$4b4{IJ7{1u5%ECwgG5Z)&kBQM&vd5^)=R zpXp1Uag_5Kluz6oIAFHj=`d2~7_x%x_9-0DS&W0j0qndI?HOhz!DWeQrIkvzdvv&u z3r@Z0k9WZPYzmmkGZwUQ-!E#Z6ZL}UAxD*eViI|QLJoq{mj}0TnAsS(lZx*;swn)7 zzmy|Ps=JH` z=8sl(wr(K&EZ|DhZxWxN)_fltyLTe+@deV*>J03xlr&_$>l0uW^%5HZtG{<)q}m4| zmiTc)wR}XbG&F>TZL3v3E>Eg;DDOO_I8g7sPO5P1TN;B~qm9?h|vZP>o0Su)@g2W;d- z_v=IdZm>CUM!y@tE3&+ju!=D38Ca!e=$G-VdnTpFYE1C^v6Ws}9BDw>*qRim@cB5S z>r*-NhME!$9BVxEcKR4jqAy@g6Tb!_Ky^2}La?&jDf|QG$Hb~4r3L8AR-E%Iqh-aa|&XoWvUo}a|EI!-R|1zVAu9d zMpdRmE7`zt=!=P63-sywQ+OaEF{WS@Y}WVvTGiWmT-OKCko2i2SYxe4j?uV3Vfl*L zHuVZKph-J_sN{XQGF2W}X9!tnFwD`}C?;^U$R%F#^k#G*2EL8@IhvufQVQuMZ*nd( zUOh{Y6fk>%am*MyYO&xuSpKvIf*5Wf+>ju4SXhOsCH`d2z539-O{}b67#WndP;<`f z6j5!2MmP`Wg!bD7-keRT8}GDxuaqsBKfqj=Ud2EB zNto;+w@#y>JrlJw+39KeQhN)W(qB8pjJXW&8fHz1F8GlM+5`ho@E5I7#wW?PgQ6h; z3FNyx<~B>xiy|koP)X}eyosxHuy*6ph7k}T7JMjK1DkJUw?N16m`kZdC3F@~3f6@f z`@E%xlBT{stF%L~u2*PUSFVRR4>T~wnX-r(bhp-sF0^hdSz`HDD8Xk=YG(OjsOL_7SqDS!Gm6K zOU~eni1SXUHV@ewba;|;GDg0a1MX3Se{d7^x$^MKB@GfrO{FEGZtjVo$K;Y!w+gsG zu8v2qYC&MZe>y%OzCL*(kvvXY=`p@YzZw0r9B-x5qWT#|!rk^TKiC<(gtS!O(Q+!X zU(SAGo4EE}KatUPRYY-T(u`-whF&+}{By7z=~s}yeRzFinrCSi4vSkDZC0$lI5#e> zWzkLW9doO`JerCYS|s|QeM(RNX{i6j-`1q{6?YPgZW*fgztX(V|NRh324(mTR;W?l zrn;Cg=Q+pV8!+!(k}c@{Ri;4Sx8pW494@z#`Rg$R(~_K6IqbC(No^lVanHzqo9y^L z-*S|Vyqmbe$|zU!mVSxVTn3cu}0` z=ebm(f@wo)jMNZ;0<~)*EtRJ(@%aE%v7*ehFkz#p7VR!!uDV?eJth}gZ%jtN{gZE&j$<9#86){W^?(>+d(JUU zhZn@lBT=gtMyjl`zzoJvV?D!r%0lacp@ImrK~1M|(Mo43qLwuWJZB)-@B%EzRfzP# z)nIYy{`Se5n5gjd0M2AR-(N5r+xmOeJZL4oIs$TU^M=+=l!t+&_I5yPeWT_IfAS7k zx~jjTz&zF??1xM8#bkM|fAcK*D0Sb%*u?Ri$$K-mCB!--*fRxb@+A$t^{Mi`k^Mzl zOg6z%V9~ekqgSnZ_fFLqXW>AO3uvV3fpJFI_+4-^t1G9%kEx5tEN!A-ul+2)2Rxo^ zdJiA^rlvd%M<4ZyT##D5VG$fk?1b6Nl1@4uOZDxYIV6ODEc^8^MtzWjeQM&`TRqOg zq3kHXagR4bo1deAQ4%^{5wMb=HuP;IrKF%rkozcytMrM8?O~vdu;%GBt`g{%2SGd} zd9@E0_p%4W*CE+_SB$3T^+6+Emzv<{?Lwm(n7ahD`bkrYn%Cp$Ax%|E(CU;k{yW4+ zZAg6J8^*VAepF<9HCIwdZA@%y2|w^;qWChNd!m#HFH4Xr7^7E@EUj9)Y4j`j;bvJa zPniBCRCcjMT@2^eHT?Di2OsZ@`H2!ZWl6)}(e}$43=PuDXGjE<%=S7vjA}|JHjy5@0GvuDfPLT~(w_J@VhuORIYR`1PLiW6WF3569xP$2WqI1U#AuI;;u{on^}TK_BYDODNL2HvdY&K#`wP{uu5}U_BmxA~N2Aq!%|CGUw9sj3p%H$J9tbWJitE*hFIE z3LCJ)KdwbN^nQ=N3xOZ6wd8H~}KqU)r0pl?+m>&@8xDi%y~A zsBXJ>Q|$|wfVZTC=-oM~+uy&7Q=}1rj{}uvnnRzYyGt6r+YEUjyq5kn*`3s6Ns};u zEI-D34-zt`FRM5kdz4wl4vir71Ul0`O zh=|e=0R;h(-a!OJ1XKteA|fEYh90UEk)qP13(`yIy@ZZ*5a}iKP(uk2ASB#}bI$MH zZ`^af^S!?J{&~j89}bDZ>}T(_=9+8mIUV%N#^tEsu6&UCb5|Rm*;j5|E;XluRjEnM zbMgwuZW0rl&h)ODV~52Aq*d-0om4k`AZRs&9InkFXBhuWAcvFu%LF!C86Cn;?pIeO z*_42=2cXp~))izMvE-cE&DBr)ht>9dYM^4m_M8lmDf*OFY@5>Awt(#qL75nrXRL4V zD{I_7dB*I&EualMyPtirZQr(PAPZ=`{%;(sxWr#DFeFfvB%fvhH!giGXUh; z^*d0UuncX-6?TBR>-_gjs|vBf2ug;y(gP{o>&C@E6$bMaKjW%b z#$Y8WQ}e-30OU7t*L`JTw;C%^)_W*XXhx-iU$_UJldXQkn(iazQHglXe#)<#mgpKg zZI^tSOkDXoU=QMwaBoMq#7xb`*$0`ml}R?HOF5#0{K@fF!67p4p*9$R-IEq0d$m!D zrblS-g;uW(C~kdBM?IY$D@qJ!J5Uq5de4UMjFp+J-N8|ep#x}L3*cC4=LfxyWEFK=oHc=i7O%2NK&sFS@emtd&Rr{n+O$x zg{~&a>}_zfHK)x2k4_;UiY3i{+Gr|w7CEfF%%!|x<&LgUw_DN?--`BC`sHXEaBLdS=d z+dz(Y%b7HxTWycLS$Fjzs25~CL^_ND0tM*1IO2CFVPmR>m;NzNoW(D`IEI^m%m30P}GjRcspi zU=DWoz!WQEZ#n?+;~L-p8gPtD7=^a@j;xmyPpM8RAd8F}c4k_CoDbU#$&6^L%9h*; ztRls87s@ijp?vgbX>vY+c7_cjoNvlay|r->%6_`P*WI{a{1-R3HtlR+K5s;XUb4 zx9$mLpISm6rDf;@_TctlaWx2pc2yS>{zhhGW4}Ms>7Z3_yXJTB8dMa`>fo1lNfxX% zigoEm0vYt~3{ZX*eoOa!W-RtP%cjf>hsZye3sUFWO?bD=`f$RtysnJ=zd^A7RdyoP zWJBMySRBZhjBNCrvi8Q7OF=#A2Bx6f9~yiO*t+*tX-j5S{2GzG_UX7GDYC3Uugis{ zsy({XWOryhl=$S=DEQ*uEotzcs@r{|W%pVSdF6(Y+twgPb6pROnpZyQ%#1a9*^(l* z6XQG8K+%K%5`e*mC`OF%{%o_QyAI4(YTR;2QH}bxAgRnz&s5X3`&cj{BbIH(ph0I3=k0Sj9RxIT%w`97+=$~pA zAAbE3yOTtAex(gB6kw@MO^&J#ijlG4| zknn+D{5sz~TFk_9c4BPJ8x?c=G~HV-0($CK4Ro>7VMM`-nw)6lZ(RSKTHiCFa zEkE3wp!!Wr50-Pvo+^GzdTRE+m_+^0C;ETonu59I_oP{?rNrzYaf7?%Kwg+!>~<{` zN^2U>#z}%0FD45w;a}MppXQc%_neQ_HpJsGx|l}v423+`%auKqvia5Eq8GR21e9CH z<5Ui6dcJrYz~{GKgf<>1$Y|Zl(*-0=9uHp|0G zo!Fnb|E&=K=UQXy+I9mqpUw+S1qN|k8S10R3W<|n%nS6aIo>xy_C$U9FNX2|*Jb)oi}=~(yX)Cl5d~pG%}y!1 zT-^ka8ZD0`fiZ@Q)9y+FOS>BLRY@QpIBi?@g^QThPA<<<$%=vGbzUvwqtB;9OU!s< zq^FYO=okeeQmc4*Gc1AqY06E!m+3ZRaS21^elidg~Hbj zHo#(H?SPnAZ*AN=cFkD{onoeUcB&1qrry2z+tcN9hgy^B?8xbzq|>>6R#|rQDo@CX z{bcR-My;4uMe`K$u6JuI+Iygg7{5j`@QiL1WF@C>c-C1txzvR6dhj!T1FIya@8gE} zRJ~rI8eiq2YfBp124Ru9o%W~crt6aZ`~Es^DPuzR4FJGRMGQsDh0?J>R1rM)m~m4T zQi4Qj_8!ENa<%xO%YX)Od1K>e5hitAWXpyji=%BxHucMs=@3#x0Ur~tYJladF>~|S zMpF^M6x^YjX#~3$8VNKkVe_QY1d01yM39($_2^`t)&zjtoCZ(IWvC@rXzrDOtj>Ng zvI+%H{;->y%@#!uERJ-RO{mm9@&1hhIi?usCMpkFWaFC z;a0)#z2gPtRKl&6(4k;eNzN&9+|-rh?N{_O-#^(49RB4M`l-C+i%7+-Y5M4Msg0EQ zsC0XTG^QFnvHby7#&e{#J|l~p&jq$e!h4X6fA+1thTq^xj{ZP0_7 z_Yf}!DO{rAYgjZ`@F>nS=X>pRE=GFSZ)u`i4t^K|H{20bU#91Tn&`Qx*H;kgO|BKm z;%n(hsOSn6hs8%1fz07x33p9s@@y@xhT^!uiOr*QmU^EEy7f+ZLZ>nXW9;GqnfUMU z>Fe&sW)$8`iI03@mZg#lz6;zI9J%=xzi<{>j8nQ&1kk7`3xDHkkJ+2LiTLY4@1_6| zO1L?qb6iU5cHHMa^_}cu;mT@Bs%Okm6ULwrQB@3or7$sg9^E}d?{^xFaK)Ytlv&S@ zlu`cP0&>BOs^3{~WbKh^^5IfLwdbk9))BVvMSvCYplZo9yCG7_;`2=eHP?f9Uk3CC47XupG&M5j4Rh*z!D9zr@P{l?ofb_Zzu;8D zp>C*|YXxF9)X9Sxj;Y<0A>4oVG}%6v0Ye56`RjVO2}#R%CM5w{Qynx%$7KJz!mQaH zRHH~ne*zzyZ#s}Q6oIBuT;LG49|~<@pLbApQg4?b*wfSTDdNPKD!d;)j-vTcL(62O!h($y*`DY3jma7U$>35FcB=)_JqC0m z=PRUm0RZCBwByM=Jp!*EE)c_k={B;iFt-GE&w6}$@R!U+FnhIc>*JX!kr@Ro$s#w% zmh0)p_|7z%v}xLt?LN?8N!_2dL(QJgSTyf{q)G`L_I%(T)JlVkE?f3>pHi@6h*AO_ z`iY;IV)sH+?5Y}i6vRY640|pwN0s|hne3JzhT)|wXfL>H(;Aas8f1XW1_=6T+qHiT zLwbmx??CuY3vh<@j#H22DUyNmd)V;yew{#_ZmY3+N&Dt$4Th-~I1$p$zYj|l} zy~;LGK4Wulx7})h)77uaIVyfpcg1}u+@4EM?!p@kX?`17pffjj)3H=nLKD*LZa;0! zo73;qiB{9!sUGmjlLH~zj9BIc^rhTSZM1jqfZh&O(qXrxN3ZZL`7R=w15ey5!*T~v z{}tN*-~Jt?0zgHCE~u`4{!8?$yYTz@l>{TFkNe`Bx9 z|6A_O$>jJ){~Q?mCIQ4f(Sxu4*^m`5b|yXaEj;QE!`{EAIN+iJpcC+TNJ{0O4V}FM zCWWNUcqIS4AAK1+IJJNP(IdUfB=V0|?q8AnpLqzzz5w?|9R;D^U(hf z3Gn{~BmcjjbYLv=blzD14(mT3|{BnMhhOL6^53<<5 zo-d&HBz`$MI%pYP_s{Z)5*sl5N%DD$_#b`wmnoOx2*`@=*?+V_1MUKYt0qK0xBuLS z>~aO}4_+!rqNmJI|7i2b(~tm5NBGX|Ki>=eJD+#c638f>r`v+=|Lnn$0ONM_1dkj3 zaL(gjyQbn2{P-Myllo^5j_7jf{z^3dvzYd;5Z9>>>{%yS@BV)_GJKy9SUM&De*XV0 z)cfy#5U_MPuq>V~v-ancVwYdwW&p5scjYbU|ISnXPwON9Kga$5kK<0N@@99w>6FC- zT-}4wnEZbN+TZ+>cn480%iHeuMW%OzfWv1QmE75V(ZYJjmO(UX&a4E`Pkqx$m@HwV zQ)Dv%GcH!Zos~@};a1wC<*H@$^~V8QKAWj>^8ggSqRh+=xUV-}>trTXVAc&~F+_d3 zB~a~PIaKWcXsYpO=e@C;sjjNFo7MyDT#h|w11BAkgQ=oE9VBwvxf-cOBa5L(?@1Ax z@ku-PS<^zD;__r)G~?E2gS%~n>$2a`(ZumWJC$J^4`{pD8&%d7$>3DP;3nXe*3g+#AlxvK;<5Ap`7u_ ztACOBxJPQ$Wdjh_v`b~wMHiVgdCAty16rWK%#EDP$s616&*ONeJ}U$jz3mJyGxwD7 zJV>ius0vON1FY)*t{3; z4lt^Q@woLGo{5DdNYOOC1aL`%N|*-1zkFCY&2UtfxcX-2#CQ#> zi$q*97I;f)+s@J)pKvL19^Mi5D2;Sc*x3-WnB>xxk*eg0<25WXY1&RIgq4R-Fotw0 zo#mOa$U0|H=~$L8gJ78lCj#*PEM91ee_M;D9JVNb$rBJVht?V)y}F~AtKW~@^=tnS zxHEBnj?5Xxd{l)A%-w%vKPzs78b8~0hy zTneT~EvQD^v0nO%v_>?lLDYR-8UbQ=8-P?XYOw?2n)pfOCpdJQkVk#=Ois3r0E;NDJWGY88+|kt9%{7ab z8NVqVtur^|Jdqe>Xy>k@(`EMSk8R-ZUnk5~Yg4%OTad-OU$B|*5UUt6Nv#VyUp8$& z-vX>mu%)O}Kqx-L+A{}RwDDwd!hc%yU8V=*wOR)^XX2w;E~xWCh@>NNhnLQ%-H zFxXhK1j>6~<`9R0)|A?r%*3u7mEbUN{2rs2y)Bj8p8S(_BVOQKn>R&C4YGKz&|DY= z;KhtQ8fr!7^1h@Gdx~p7)f9wXS;qJOyq*Tet;3N$#y8>%tZR7`YwX8(OzQ$0yIG)> zID8Gf$`HSI&9K}xUp?({=gpTxS*c#V58TEyec#gYXg^yw86Y-us5`X)ANh>eAf~3a%_VDk^B~!ad1JH7Lb-7%LY1B{wi9oH5=3GIDS|S;Tu5 zfF^wz-E$LAG!XCaK$SQNJ88nygd&mNzL9RM}$3g3v?pPWq5t>FuwyVbG21b7W zjNb2NbzrUSMYYck4Z-TVZobzTn`f5%U1GiYq!|#dQXk_pJzc0lRQoj^eg$rv^PAQk zyKJ@uxQ#}N<;>+iSpslCp87X&!HMI7yMA3(xtxH!(GncyjOzm=nW)ISG5LTXDDJSA zcjp>1eYYg>1cF5yjV(erK1^?l*c>oz9veCgj3IU0(I46dGx%H;_L3dyI^o)Jdl|4_ zem;f^PX-0f(dnYrhwn%6%gH>aV6Xj7w~=7_E&+`*qkvsQBN1QY=zVolXJZc19+l)L z`=<5xfx2F2T~{3UxTii(QIZ z&Wqs}O;lgFi(Cb+);WmY-fk9D$Ux^_;N7t^TAEW0yG;pu>g4t83*8AO)(K!Y{a#EP zaASS@dJV3WO7>YTS0x-hC^KPyK30<1u=b}M9n%saHvOG!N5Er=#-gGkJuzZevVvfz zLhbz5Z`fBO8G(_!JU}S$eqO$Xutq%D8r3=ifdO%@3z~K3%-Vk{4T3q;`zmN^vKobZJ4D% z^punMCJ_u#8w_8rc&E1yZqGJ|xg{7Mc>2NnQqo-TLhe%y7>ObIVX{}w zj^iu4MmjPdj`9gRPI+};u1>qv(chN0Mm}$67 zV!7cuZHn=>uxNIr^^DHi`EfFzWT?DcvE;ECnoQLFjmy;_qG$g9=(W&!SH2tl2yn>4 zY?mo@NNBS$=#&AzwnEy2+f(iDo{^ZtnMC^|=%n0iALy07`4+)AtH4(2%i&=1kzOF} z(kPvzRW1JhVj1?U9G<$CpQAUPpO5-JHXKl%>08K9RHvB|8 zv$0LvyNQTm6_|W8qRu?I_$+!W*ol1`|9p20aBjLj+l^l9F?+g^*5&cLyqAub)WoA) z9@=C=Qagj3;J#CPjcL~j_Xx?UmhiloL1Z@h+htPer%7fdy>)+~EHbw(_!E6!n}4&O z5EG)zioj?gtyz0+Dm3ptZY5vqvC^(320O_>i`fbYx|Z!3!}S`O>Q1C-C%<6%Jldow zBJU#ij!@R~?B`###gxUbH0}#g1Yo#lM}#bXzBvG(ar{sy`4`X4pHY$5$#1X`{G(R@ zedu4zL#+D-?fSP-tzp|!<*I-wI*U_B=$x3{{6i@>lyKs6o^T5F6lu@b%20Hq)1>`( z!rweR&LNIR5IVnTZ70`=zbvs zp@yzePio_%8DxzzRX(ka3F2}FuVk9^|9Jo9?t$tAta_cv>}mK-)#<`#jF2$F<*q5% zdd>pPMF69_G2LwY*ckvhx*$!jtE_+38H$~nk3W(lwRWdk9 zEnt}G5&>9X@TW>9YAxbSosL}XYi-;&CP%`~ogCQiszg0hBBhUK;W<>f1H(-^z>j|8 z$`TZaC>3PiehKf8=@fw$v>42h1hlMO2}1`lszPGgX1oSk-2CnD`d?}6BwoS+^@*B3 z-8u*0A-e_9lYaHxt=M4AxAZCUeTt%3*+LTg!V(;Ub&ILS*ST(d?LRrSj8QM<$E0sI znSC65bz6P+!UT}a>^Q7-BQY|&dj0cX|LE0%*7R>ft2B0vS`NnA@yvg~I=Is@C>~<- z2nf*rLvmt)8ev{&WFLJ2yJorshgPn7%pJ*Y92S$jf6#cGL}Q_-eU{aB%r`(xatk=_`Fv?l6b_wdz00+ zCQwd>EzDm{rUdjut-~+vi>Y1ZC|)Uuaj})cXg70qSte&q>y~tjUbZ$(Th5cTAITCM z0@fw@i%)|4xk$9rUL##Ci;SvzS!fvvOZD4i_Jk<;*Z=s~ z?GlOZ%H20Z@_)`zR7s-$##Z#5Pf{GmT^-(LNAcYr~t0lY%Y zi#tNojCYR!(V?^F(Ls zv1*-2l86THA(j~5Ul-!W{c~cE-q@4T^F7w;XbQ1m)^S?yqG*+=W_#l!9))Oz&MfcM z!i&^zp(xmxx?>2a0WnM0{CSqW*kD}~&raWxT49o(Z?ygrwR-fYdC};)yd(alt?cvQ zwNLNRc*V6~VUrRu8x?|j-p6bI{$XkCgf!_Hb+W)^z1nKC1b$~=kS!lnk<-Y$(0`s6 z!zB9lLy=zHuP|l!ZQ*A?_`kKu&dy)F>?jkG2W{*4xHeYd?w~5`R`l*=#!(E3c%1{H ztes?A46w2%K0u0%`YYCOuJIWc#5#S&h{4I|`M+^J(B5Mf%-l7UjI38G(QnX!88>i= z(-T!MICIJ~rHZzwPtQp3wZd}GJ`voV2hx=4kwKNTNLk;**1Zu&>NG>{fb~15O%H|h zCfifzMEIM4^D2jQr%n69sPP|1zH0Hp`D2_`~CVI=fb5$*&XoXhh$*^&$JVC%HK9f@|>s8RzV;_xzpcBAERK5AFzfisLoxl!G zW8Rcer{X;#f2~|iZ@NTphVf*>`mu4ZbZL*h7M{o5eF*}|9HPZ1)71XTAAgVGt(-E% zV%FsD$rJv;YwyE#X;%@y_Xc74D+AvK8;h(IJfcdY)Oq463Ow%pd-4mg4D@%1LkX3A z=$7k2^SIj2WCYO&z>n6Fe-)VbJxc7r#BlZWgM9^~W`eF!lG9#;AUoLZ*_|wgD&Hrs zInE!=Q5{0frK+n55Y2er10lDL8%ixn)!t$VAj8UjV-K`ONXoYD&F9dv=3&js3wdHGyEq*zW`E z;umh`9?LB+nv#!qib*dSY|xTVCwP6Jih1u8S?)N}o?4v$8`79kJ`gWb=#;JbnJr*Z#u7Cvzvy}UeA~&n z@^-qk=P&(jLFoPir|%eXiCbUgACXkrPk`b}PDX9iY>%I%h<~{^q8oB(m3=l7ch4^A zx5*5ggf9Jq$ z;nkTsUKMqLNxwPp1=MUi^%I;xJ*N1g;T_toL9<%;0cWHyH>*zBoAz6+?x|^a4Pt+O zeWYQldRr_?!pj=n7?)b_p<2x%ZYxlzH)wl<*GpJx=;=#B#v?4&C86YM@!qSividiA zm_isuDAOglS$*+Uv#;$~P*ZoBB-IY@zw+sCafw6YKJz#|>G{GHhYWeh8Ss+&`<^t> z@WRzun3;-%+j;FuZz7=JwqZ6~(nKMM$L^YM9IU!Q_A`~To!aF`B0*(u7N49N%)vRH zhOt4tj$<2nmX2-#jA2ZCv0;NYl~(AA$No}X>}jw3$_`6wYN0n=rKf#_sm(?FLy19? zK{(xmyEfy+LW(*tJ;6yuZgNM#Q0(DcgF7*B$;s)d(kRJZpi3QHi8a2+o=e<$B@DE( zzpgf$_AC8@~uTnP`SObIEIiwIw7c);3Zk$}Pp(Tu9*TIne+z%LS^_ zg?E%A5&#}H4J?x;h6Tg-I%%P)6CA$hC-@L=+ar&X__0V@Is9QvkzU!8>1u1vYY>i5 z&e#&YW}h%*U?f!^y_ zxl6sy^)!5lQ!mp{_{{o5?Cb84G~go2=yEu^IVejLN3N1_fIRQoV_mIiVsaNy69;lN;1sQpgFI zPSqp2G^%fsiuM;f;1VK1-?%5I+0Gea_m0g(DUJfUjKXVJB{$o=H@SBnNMjTMB~6`r z+xflGS?vh{ti?CdA>`&&>aal?WQj>kwxRp;)j)@DKF{zQa_053b>2g+`6iWY!ASOi z?rPub!=VU)%uoup%AMcYpSI;3Q2t#LfX1{gbA5tE2^?SFqbN;IP+#*6Ko2!M%l^^5 zYMVrn*bYPaMu`4yWN8GmlzVS(lUL%zK^!lRvdKiBgUvRIZs9koMoa!m?J9@kg$HeF zyZS{}BxBCp6?6&GPiNg|ceZZ1?zP$lQ1{hxH>UR4ARS?)9jlHqh$`=zH+*6DRKGq; zn=m2UpQ75zxu(|fQTc!|^i0fay)HnlAZc;W#GuI{av~#PBZR>Q9WbF> z#1?@3+!CU0nB|sAWj^a3lh!`wqb+byD>MSgLw@nJ+#Wwuw9F9G%dWd;3e%r^p8cDjOxfX&jI<@_4&z(BxQI&8;SAN%L6&jSm%W zleeeX{xII9LWHDa9(RrMW^L6HgMOxIA--4)5nG2|yR+y3C%E!%G69_ArXv&8t(o%q zbW>xO?@_de(=zFYpFttyUmTR94m|{3oGo6Xa!b9dSY>j3+lZZ$^{vd%dNi8zCHz9#VS^GlN79syT?^Tv>lVmG{fLT|I`iIs{>Okdmx2-p80g*ADCVg~)X|~V*7OEAv z8lXa#iUg3Xn~Wbrmtjp+uVky_5 ziCQ;0BWVkqCO;Cj#rzeFZ2j(ie&S@PO}g&{)D4ZtfgX(xyYnVnsdN>-I1rn)3OS=- zTVu-6;Ni+HoSwuN*J%fdP7`v6c`7ufYR@GscYWlb|Gk6l=!rQ#wQj^!%lBime#C|& zLbs;bNorE+*_EkOFIkzZg?WwUg*wi-Gss5X;JzLhu>a;XsC-|p$~@10J9LaqC> z9)9tBd$LSj;wehY_k1L*;+o%nu(s~F-Zfvrb_q5~aP!+>no7kFDBiE!n*6lfnxfJ+ zV|nZ!Z*YF4?X?P)T+pPX7Hi*W)L-su4!Is9nKE<3xP402aJi5IJx5-(th z=V^#$5rberGr!#rG?g9GU#61}un~Nczd445uDVg_Ll&X^MTpVmPPTyDIAG&yZpB?+ zZTwaES1myQl|yDk1CkH%SFRZ;2Nph_%}&2s{-G!^c?_4`ZGSfQbQf*C!8)d8p$nIwJmfDs6L z6KW>Kca#Jo9m|dlyV=OP@8)TD&AK8Ue|G?H4A)gKNxS|GTfD?~g4)_X>cML|@*Q*F z8PopK@MF6f2KGlp$JCKQ1cXtmgL7W@FDq{H*JZv1u@AmLK8Dlr{EQyWCF>Xj7bwb_ zd_z8=s|wxTqHzUw1BZYo>lpFj96OsI(pLC+W}KINc-U zwtZ4ch}_9}7_;fqF16Puv?gOB<|+STo=tumpAmk-Sa5qrE0zcJJS4vYmndZULgd2| zb!n(5gUEX};|6!JiCusskNC`t!Xby!#T2`u==vJ{x+54ddpBD>NS{1IzBUYA!@8mg zI}Rs1hHL!%TyU-3(~cugN*GbQ74;um=V;`zkY?UDJf^R=Klpu<)c%f%K zEJCYT%teeFv)q?eHD5GVw#C~L*k;a~zpcg@l%i!%jjnKbHtHCH@ROf_*_`Q$2?9USOF#Ld zvjk@tLcW|lJW4>#R}nJyq(mKB4yMKINV#pSmHHiBl;8Yn%z7GGy@r4Yep+eMrTto9XS_6Wz0O5a}Z>mE%x9k2)wes>l%j>QRj@1iIQAdmy2qr&~Y7 zt6zR^P4!{(;PO1T(!skM+zZhUP>{rIrAC6gMckBA85V$bY=cM;5I(vN@Juq$e2_kMcQ0fN@4-HOb^W+PZ!0!S$lDZvrP8x zqk93^+Ze8S7d@AQcdC)%h#Bm_#3W!t6W1OZO&TQ#QGDaP8qa6*bJrYo{LG`Tz-lUnWjIwDIu+e)?W z?yhsQH?8yD?uL8y)-ft(lwLq>_S*<&kgs374%{0W>pzN?lZ4Stjb?q-8&SNN`yrws zECTOOm5qF{{`j-L$xE6=%H4!AMVbQ%(xOuoG0&&v zjAc4w&S`L48=Uhl2f?{(?kdbqhJ1>1mqT^CXZB8nPaHK&Ju9j0mE6Z|@YD6Bi~A|S zW@|Z1y27nK^yDj^Kyx9Kn%y#zctPfigQH|GKC@@xANDyLBo`V=hBiOhzrk2L>Uo2hNWm8VMzcM>aBr^&d$Jt~|>`z?Y%7ralc0 zLRk{EODSY0wM*6EE?v^jIqG(>CK1e@c*q#U-LGU-BV8v`>EpIjFH)Pk95?)`E~~zy zQuT(~Eo1NZ^Ssf$v!^U`nHc;RA!L7*T1kEIiw9r(cxgC1QsCYh5Gs?A-KFJ19WKv6 z-2o_z2__1Ann)!}Fp>VUbrO}&7>UsW?mCwrEZaYK*4g>+t&p1pPJ!FxAb-p=E z?@6*X^Vcx|?(V7+x1X);!ZG1waFz=WF!of-XVbc`LS0CPXyUXFnWleUbTJ;B`V&+; z-X?!I;@L)YJu64j`eHZryM6WeX^X|m-r{4_jDz|7`DsoGXI97?IKaQ~*!Ee7V7xwf zOQsCL(}jC#a}g64@_Zu_M260NoV)6?t)lZQntAl*05Hq5u`-^w<{Dw5; z;vQ}`c_(F1T3_N`8c$br|u-WXwNG2UlVM{qa$E;@X87cL@oN$#^mL(d(+Pje*K{jl=W>)hx0 z7nlosqzvb#*!z6B+$+J;LD!DQFVNPt!W26`Mpv&}T(wn>!t03LU%VJ?;BP8Du9<(x z4%i`no>Ptu!j#dBA3eD&S1Wm9oZtK`x}Gi%P^Faqg2tgfLFc-WK1 z_X!Vs1uoDVfF@Kxi=vk`|U4WiR}Wp7ul?}%757MWDc^OQ%|y6?^7 z@{~9#7KdPO0{=!kd?BCxuNs#RTCP#&lXYa~6-i%3^0k%0t1nkX*j`iJEXl=8qnvbJ z;LNhm>(3V5^#p&H+MFg_*&cN`-||2F_Na5v;WjR~XsSy0EYDEJsQ-HV!nl}U@cggx%+!F^1V-w5`-AOxE!suq)dnj$&>&X8 zf48bQoR*!{Td@7Cr@ok^&BF^KU9I@U&af+Dl{CN^=ZiUa+2wa@^3H7kc3rtP`rw`Q z8?golu>>wPP{O0P$I=qMs6aJ|$NHe#^ZThaoT1@pMMgdBp1#ZK`c&w{l;x}!-{cLn zi{HS4DwsA3t;1)mpZa^|tYJm-cuVr1)h~rp^(z-yY;JEwzJlU=C`P9Z(D&Il791;O!6FflJlaqR~wo%~i{4BNaM1#fIT=_$JY(8=OS110kIi_i%F zt%gVMpi^fLxco2J7FnVuVj_OlYxSqUms31D^qrC4+i_qjs*uN@uMwA_tRNX799C~3 z-$NXk0uB5KTJ>kfiz6Rlfo6v@4rnl)FXcNT!u5qL4OB~1!iM(ivnW%)byMJ?Z!Nit zM=GZ5@cDidA29@#_OXBHB70IgnycAY=YqT!kS3K{zle3_fuylD*}u6h^Y#M`B7F?+ zxab2)RW5!6_s_r2ba!Q$TO*Hg<2px{WYW=a6bH|@{}6OZ`k}? zC;w-oORngz#-32OhUavB0O50WGH?elxntzEBC(et?x@ok@w_>=5iXdre;@4>MGJ^a zcB*AVW?=l9rH;q#(<2r=zgEv&vSNUwn^rMx!=N!%g~w#b8ZYm%*|m_oT~0k|S03Uw z35HX10zMfPv!s(oE3(x1-E*t1hjTA0jg-k5?{$XXbal9KFIwHVc-pm038}y!of8Pu zGW8@wNP%-j*)4?DUh6NdesF1rk}JMztK_4cm=nfnaD&x!6FS^)OX64(}6HcHBeHBas)YyRev)1VSaElJaisjae>`!rtze@EiU|1@4g<~ zG$xiGY-}-*@(>`v4<@hfmR*gMi*Hwn@+wkcPYW5&v$F9tB;E)63AX&WA}4x<5@#Lj zBVk)P;k58g?r`f>`I`~egSkmlLVB;iQ$@9mD*lz}qV~vMBGs@Wh?=?0<+t3;?+Hhn zzL%-;?vdpPJMqZ^TAGrLk?Rg+M^g|j8B*ICJ zsrjxWZA$K)TcAEQWEoYJ0kx#SOtm8M%BgWu`m2eZ=w=oU$Bg5fj;p92chSZwQK~id zRWE2GjAg+riRJHtdl+*MwQ^KTw~!D zF#0O`qd7*FN1i{|cJ|8oOf*Xg<)%U>?me05ewI?-yC8f2+cVACElu-=Mq1af0y#s~ zO33-?V2G!li96Wk3Q}1y*-i>O&0P){+Y19&7jt|LsN)c zw~yuV=3UjXrg@q3xgd{=H=|^zPK&r8`}OO6klLY0&s`#uW0OvX8y;@g@4fi;(iz#y z>8PL?_t+d5&lnt29yQj~e{3m=OU-n#OkT)hA0PZ8({PjLr99i(F{dQepa;9v6&h$4 z`JF7FF8wN@mGkD)U@9^!v!N3}1FLr>-NzNblQ#ncB%(|2)m%}ww6nrb_Z`hbw`v4N zJRQ5eW8L)VQS3zn33{w=PvzqIMRIun<9Odr2lhl-7^~ib`Nu)UUvdwpm%N@FMo};W zM@}zdKoB6-IbSaC^zv(gPmv4>(vNkI=r@uevW_&VX6=+lb&K{Rm1Z1{yGI8W$= zu<)nt@dMpsKB;J;*}A4q2Q(ozlj7`c`)9gnNmiSb1tu}u>AR81RoCxIMgv{7zLZC| zh}LDWBps zB4s*O)RSu<(bt+A25dEZ9ZB*suB`haeiytFdB*5R z%v$&t6aOJ9z?*6CI4rf{_$N}a+09`|kZKK;iT+R@#4>nt-PIBB zXJczM&3uP7;kR>F1NDW}29|<}?mAhvaQdu;j}6{U-;f(y&%dA&+7{i2;rx;@z}GhO)jMt&s1Fk?51E@C%Wf`5)r zrbS7C9wPLOTLJ9uO5yOX_f1FBX~nxrF^U$~qOF=3qGSr6&O9)ZaYO>>=Vym+6@DY* zh(J&MPINYb2SKiqH?lnhVD!RE{Zr`KywtB%KauRPBqD?T2>>);K7vrvZOkl$?`w~d z8tKXh=7VheW}VKW*K1t$jNJWV+Rsv8ORuz}1GujUk9%iorx_6k5dIB%fyI zRZN?r*l6Dc=n}e0W5Pq|F2@HJl3TVVTq4@}d>8X|vcq?~iZ}^wc<_^CJ)>Dak0wq% z+RMvFZR+?gwkdfxQ4dfuO1^x(DKJSw0t=01ut|yDZzUbcSJQl8Qu(z!+b8??q%CB5C4m$&pgqlM|^LMz? ziq~@Q{el^xa{C0nTHj$B3bDgsdsm+o4E#WPaM4o7GO_dHRwPLN+|jHRQlrx2xwVylo@Q@tkfCvczx#aU*;hI6LKNhj=DYF{ z&*Qhh#T$50$xEH2GoHKzCXQ#-g#4#*IO)wHYa3?vQj-d|pg0i~@ea)9h*J8km~q=1 zdS%X4mAmKpeMfE^3qRjn33QqJNXs2t5<)&aU`M>I<4E9u*wAt*ekrbh;Sp$KT)>t{p-MQ4@ry5d=Y^cM^#jBoe(v5JB|bg9y<}^cppK zHyDgA+K67qC}A*+ZZM4RvG;z@`Oe;_t5G=-PdKKQmB^~ z5ZYx-71HR1zH?8%&K8tf4zHgcsD6VpDRbrVQwRBiu=HXl5aJpoEQn%a_D_~;< z-a1(E>&vo)Or20|cw^Oivt8%)adDBnq}tn^yGwl)QK`2VD?rjFFB$o?p8LN?Z+oE! z>}g;IgBZ08UWKcIIF;kx){c{Pv4Xx?rcUW1VagqkJAfXCZxr^Gji>foS1AY9&Bv!%=C<2(8WSi=Q(Q;mvdqu+a_bat^#{GG8hwm8NUq*Pron9}?Xxn@&P4!HDV2SI?Y@J6|IT!1>M73V zBz#q0muiHXoY1aRw2bMs+Zx{@%3)!_L_arFJgq!rdDz%QhzY&#**9cd|Ma0c_X4}L zXD3bLqk;ej4~3So)Go8Rweo{P%IhWXKEr=8`Bv@EQ%*Y#VG6t}wt zS6%+ZwC#iC{4Y5l4P_%APlb05G<=wiZ6ktVPZ&D&*r)dWcyDuYJvd>=_&E5i*zch| z*-qOvLhd=ky;4Wxo^=5~m%{EDj!33Fb$um^c~<*8K4TP5vb0OnNk^!PDOvJMcsNL9 zm+{1Q;33Vh7*~HhC)+vTgD~Ri;=IhI!Rm^%!7xPs!R~$=Au#&OA!FrXX#Z6dlD6f`#pCQtUL;u!{ zyL%(9q7-W{AFOz;8-Z3%yv&dofTOm+6I*N5o*~1ESfi+(A>)U4bydztXtoziQxw}p zE1Qq5_lE)teX2t&2tod`Tw}3`3OLNW0fcqQSh_dkTomu|K4Srd6x0aRX6cY~Ow3O} zbek{R&Mkv`={iTWN;KoRLc&Yew8|60*63HhEOp0#_V7aOwQb=%BDV0##$(Ac081!a zb3RYd#`Bw5J99Fub$??9rMBiW z`M$6iO|AlnFw1E`ixA!VR9U*=5Wv`I-B-O5@S;C{PH&y6$25ysBWP_bbXE0^;=t`k zB_;j4z@}!fr?%MOQ0@y3RXI0sDR+%(fDm9~a9!-1n^6)?9A`V zhcFM)`Moh8bci&S`>66vHeU`LWgvXr-DqG=jgX25&6#gCBSR@ls ziB7$Ha_jKCey&=)#(bF;r=rlOT@+H<``33B2{fAQ>8d-wf5F;pb@A&5#`4lHY4(#x zTM}Ju^ri-xu@|sPTHI7BU^NVtxv}qYL{7c;LYw>n}PV%1_7g_}e0GFAqwJmF8F34oOR-zrbI;OcKS)g@Q*&(H4YWQ5h0; z=>2$|ZSo&Eh9_qpQo`j~q^nSSXuocxGpEp%u+@jtRU0`9UzVzb4ct_IMOoW5wC9{# z(~fu=SbTs*FE~%a{g!HkyrWIOZ6+H6Od`5ZPHj0r0{Uv!@j94BkF|A4+V)nM=#R^5#MsS%**)fJP`i<%Y0 zO#O0yK+}n|&p2{5%tjJuSwnsSwF{Q}l~Ig}}QToJJ4qIUt(04BVRci|^DKh+Nfi5fv{d+f$0WcjW2w zB9q;POUz?qk%VLjVkMD(T=U+X3x@aT zp;m6+wj}(qXeOO+xG?%P&}jrHY+Wmu78qvF@KspjOe0;w4)o+>E8ERO+NZ1CZ`FVW zi8^k4nxUazIzB!B<&)rYRaw?sON_B@&ve%|cVlqX**zhSRtT;&0-DbX56MKw{O zvbUT_uQu(c@J^#{78efXI7zk8iQ~7HKMiN@U50E_ij@f%td+wXZ8m%&GUv9u)DRyz z8PuOCh|SHl-LZD+2GT}CBkWSU&?IvC!$nc(LC4(+innISchbR*w&E3Z(tR$V?j6@A zZ`PmqObE4wdW&=)%APQ5VJT6_Bxs`}&|@y?QR$*#r2UFpw;vfUtS=qSJhz^{En~rC zx4xqj405WJypKaP6v%IK{c;T5?_Hr;#QL1Nmzc5uQ!?!F!mmOBUg+F#&a=4N!#T5I zGJe?Xbop#mj4iK+adH?9Z!!m`PUOnEORzi2K+J6RC^#xvyetYf(_)&qg-7xT*>X;> z$+EwJGO72IV10mCrk|ABQi{%G@zF!Sb7c&`;GMP|0$`sjZCZNs^&ytdh+Ob;^E3Y7 ztHLv~1~x&MxrIo>Z+#M4weV#z8Hx7vJ7j01)Zy104g)9cs`F26Xxl0U=BIDpJZ%sMu}#e*GB_*w1UEzo0!S4QMCvui!}*Keh5^ zfo?yUXFEi;3v?bgGr%(6TE*Bi-YjrK8>CM88l;IH2>N@?(%uoViQkL2mI(9CujCAT z{2R=ionylKFlef1h7K{uKG35njohXN$2i-N9s`^@;>+B6@-vz`|g(YKq{O zu${R{gXIkZU1JJ$H>CL0w@!rRELv`vsnv$|BialSbAhASm$M5U#Ud0g_T8U`5WrHwAAhFuF6M$JkfuhA4&$}rKurMyes4>`eg|6ERjNj(_a2Rv;Z#4 zKmjj4-cMi5l;ssNoAvmW^vF^U*%Id$`=L-n@rexl(J9lkUzo9Wxtrx{W@~l{CK358 z90nqpLT&x*r=DG9=mf>1qGBiam`z=_x|C@{uuH$Qe+6T^6Md=Tsg-1J97{TI!mfX( zkAvPw7;lnEpSB7%jT9_&*tz3Cj||*D0BS8XqM2TyuM%-dR?_wz?U4N>FxWRqwU2Ga zbGQAGo$lKSTcYx9LcqAnesXzzAdrq|GPfD-xPT8{Nuq+Fjb4WxaC4PZO6x3?|RC~QgxQJP3G zSsS2D=jzL>KZP*b4u81`CfZLwSIdxG83^Uu-NT2SY;n&qbs@CG7>0PP9!Gh6-3^IK?aO_p zvXTmCbPTROD!7^7w^lbOMPwBiN!b`%oml>(Ca-uCO5O-rbuR*qp3_J0Z;ZxN=c!qO z$ZT=nqN?ng*hdpDEE`K^D)m#eIr;ohbJk=pD_fC|Gahb9$~;A`%>THV8zCiopOU;< ze;4Q{>Amg~>NJ!4rSvYwT@7N|QmRe`y89(3Q}Z1I+1kxt&0W4e?O=hW;%{{AK7lxW z{fK_#htucza$c}=9%0{2Sys6<{>U^j?)9;QAt};y>E)%>?Git|B(FvSsZxBIn`K2` zOac0Y;rvw~TBLn2&J_$o@QsYG>M8xvY5YX>4LSqzD{P1njG>LRE&I#tr>^V!lIEc_+;8a-JZ)&8q?1Ax->-FMDS`i{nf!Y& zyOGOmo}aQz4Yn8@#p*fj<2|sYgo8pn-#q$tB?Jm~{;)N_S&4wbWq@97qH+4U%y9Bo zaEN-x>66;qu6O@OgxDXEbj6^Lr%kkTWgfQ-VLFpHSx5XFGl(&SC|ChJ&Gn5z64-1# zpPt7T1F?G}+!T3_Pe5kp^XFBE3nqbACKoaIeo1>fZ>m%nCcMowQ$XC6to-(HREC!d8e z%h4N%r_Elspc&76$F@Bv+)(pzni6Qm%CzaZ)ZX>5MDr+L;|w7eB7Y}O`<(aC68^&| zD%X9^6zoB>+h$XIQk7XlykJi@%fXF@%>|B1%8i4!8A5M5z<+Gou$`<42ayv(O7alOcU`wRhO^=eeg}6~yxC2|k3b z{bMpZ|IBpe8`)^k`u84Z!fXD-G<-^vC?7-DPx)g##w%V@Q7U$#iV~*PzOCM4-&fd4 zjT*hn}w#l{b6^S{i16B>RCjPK|)Z27#P0YVEzmexewZ zmv^03=3*muD1j$TWxf{&TnjXhE^1sfncsctu3MyCY{o~7-i8`_nXOj^`z6T+i6nZ1 zN{YL7+iGx>c!f*oETx{QPrgDlI1JToHI2Z#VS}O^SSMHO}{cH3X}w* z)~AP;QzJXTd#o|eTf(<$bB5tK_~uS%p5K+qrE&=AkJ4FC>4T`>X^~UEjzgG1K^`t= zfZ|(l1af0y8>agj!E1Ar*XEv$8$seU#jxGq{ne9<&YJ z8}d8TU3j|{KI^uvoca>3oak^fuaDzA*9wj0ed}qVj~1Lr>oXxg0<3MTfc_r+o4X5b zS9M!#3PRymYp;{n(H!jl+Q6I|dSix{SN0C?9IZ#&%2+j>6)fxSYB6dXKC8WpU&1x* zbWk6jB&7W?L149HnmN7On~&sh#=rWN_tpoRgmr?wR)klNr?H9Y#{$AZ7qnnxb;+B$ z0m{K$!}AhNYx4x(sI40zP?oVHD)p;QXfJzc$|1jB$2$yTr&RS@3~w?BwfG8V_nmh! zFVUDk+sWESy+gdDq3OMXcn0LXb{Z0>KK#8siclEGJ1ONHDS1+Zw&1m$KVFb9d&88- zFm(H}@($RV8_z%pqiE61XXR=nO0l9Zqn+A=8dZaBW2LOAdketBILVc#CJl~Vj+Ahl#G})QR6}Uk>dLs~JneYVYkLYLn&F|2 zZ>o?g^G^U@_B?-ASnX40FwLeNxv5e1fwOnPJvYS*Qy^qqEU(#iZh?utze7v`7eOcaO?uoUpVkb!|o) zvs0vUOC-`;qrA?UQDQg^w|yD4>=(@hb|VjIm7|+_`l5PL76M}?{k(Q=9z5GHeqC*( z-~h8_LlBj;eS^gt3gm+hG!S28B;bkCDjKQ3RsG$=u=|%%X-R*jhQEl}6dxp+Kg6>1 zyTSeMvtk=rO1I~Wz6jFsw-YuDt-tC*9x-t}b25t``*O3d(uro^8bL^A>HSQZ^051p zTxjP`&quyMwc*#3*3phO3^Eg;$VioE_X_QiHX2%gigyIS>U{toQ3qf@C@pox`)sTl z=|0=h)nLOJbiUOOWD(j8Ln*bO1C zQ;Gwr-N(ryTauK(ax-8*z%{=w>F3WYocM3u77NL(+r-p&+P-Ugq2@fq+}A?rr5#_R z=1yk0>Y`Ndj3++9-=Fs2j(K27$toSZa|_4cGGb=C-{Rbh&P0qcE~rY#T12t9&hS@v zMCv7DZa-id#IfknNl5X_#-*GQv(-U11t@vNtUb!sFA&VuSI4yyB#b(l;&4^@N=E*k ziO~IQ`W$niv>!F2vY$^9;Vz;4^#^XH)La&j&%8xPmNL1|m+SXHy;RafJ(P$FPUOfdGimM#Q>DI^5PWrdQTL3w&Xnw9DSNR~vJoB~5oDU}Zq&_#MFXpZ9Wga3~ zfaV5aRX&6I`;hMo-eV&0VvGH-hDY&NB-9mAzZlcvgNO`glwi@3GnKpg^f2RY zwe(wVg`<-0b^1ecspXcs_EeP?&FyWSZI~W!`@YYqLOehmh*oc`OO$MI>N7uo2sT?M zVx2n@q2ACJv|&yMAJ(PT+0JS(RC zJ5t_<+4y;?gIq>kGcWHxi!Ij^PGH}73o*tY58)voO-{sk^RK^Zk+D1V_o_@U&&uAR zs754j*O`0_ZaubJ+2zerD}?*VZAN$z+YTK=os%1*y+T9rjS{=5(Ae&rg63kB%a zw9SBBw^=g})(TKd6d#65t7TRRhMUVjE_=n^`(hqU6onM|uvlDpus;3l?HT}WK8>Jz z_IZ7-Xwp}mIYdGB%(&?_n}?>GVNW>isQI%0I8DIngJc1{=^h}FVOql!_q*EP^=p~D z90DeEH;}Onya4#E_;}+$cXL(MH}MwTs-(~smHzF0n%Q&oCd>lm0%m6RkGN?cgQgzL96sG>K*pRES)03 zjjY}PwRM$BK>kKGPW$;Uo>Y<8uN&_c!U2^G#RAgOP3tfjc0fu;mHm7t?w)$WYwUzW zkQin)%uA->z7&J@JD1l#)qeiYg?%HeU1V zLQLJnM+-D~WTVjV7=!hZW4iV+DV_lzSYVRrBOhh=IAwGER0+wh(JE4PC$EftBaPvX z4jw_1ran^(XcEOMHl@8EtidZfQIfQzf5=Wm`Cqx%CN};-Lg%-+G(n$M{aAD+$4i0J zXbkcOeY_htQenwGjhXa?$XQI1)y6?C2ul5LHs>lT|uttIGuV|t-qw{c&aAnHK9!a)s)Ge zJ%Wi1XlYy-d3QQ6JyK@s2EGAB$q}Hd_n~+h&v#xT)m$m&mKD6Een9oVyeb${>|ig& zo9r_l^DB>hBM=0-4N$ecrH{G2u6_c!9-HEHdjO!4ovM!7Z=kT`-IqE!-%%quc(@0T zxVlNy0D=^~@{mHR*XdJJ!t5cW+hTbW<;&Rby|kY;`GTM9Fworv9rO+q9GuyIip+_G ze>ptcecN?o@C75L)!}TcOT}^hl;lf6_E~O`Nl4ziH}kmkT+Umj+Mr@h)n_|h&#^L8 z6-*-qIv2&=_MDWRG=#^2ccQB^hB-dSy*M9|xkzzg_Z2GBwm4mg#7Nd%wWw&9iqRvv zSm0txalZsxAAi_T%ttc|GvJA2B)H(YwG25*h-b+3YezZ#U33FTI08~SPj{{m3gFKJl=D4{3$}f!*D9aF!2t}ZzZH`T(JPIaeaTRR+aL`t6 zi(*e}+r9vYSkOK-SCTq4G}V%Rh6T5!j7d42Ld?;uP@^zS(B{nF4EnWAcSRr_fa(Q! z@ph1tc8(4ae#?iG;WvE%)d!zQf7Po=v-R7TQYfqLx3Xx)3`Zj9nmEkoxoC?R7Nt3M zlDL<H;BoK&ppFus_V&5?y7k+-v57YYcI4ik2r&>tJYpeC~!1-HKw31^nW51=#yL zneuyoamw^onsNlBQKNk@!aR7XINtbAFYxzP;_c$6&w~8$JcwjFLnsC@ISw6YHie^w zi-BDiwTYPj8tMHvX)L@bS)+5?+c-V@K!*Ii-!`iuwt(WkIQr?z2~SuqeC6?6*koMD z+RND0$o1a{3Dwigae+69I-BHpQ%=Hr=1C$i2EViXx z_FDJ~T_hipSf1rX9g62e4+>lo54UDPY=zQ!96`RaQNx#|K^8Fq6Jb63W&haMlCtj&`P*{qxg=oLYrx|R>!t3vn=%Q zp6w2Nv|s_~cp7*>x~}od!{{GpXW|e+TCPu2B!aG6(H} z8k%J?;W~3Ke2z?{n(tiKl8U;w(?ppC6gZlz3}~(&bL{-u#62I1;Q6zHP=CmARZ_eM zmXlZ_n=~jAdL2+4QiJ{MSDp`FPw#pj_xH)y>U1flhQLxf{qD{G(=)cQUD*{gvX@cN z;YE~#!!|{fS3k~e#9P-YzM=zHVo)HH0*P}mL;NHscT5TVBcz5R#Yi@rQCK-7I z=W?(wA1`Icj1agBvwBFhMFhQAm4hgV`U1e8 z8hp&Ay!^_VjIi@PBENL*~n0WmP|9W2J5>iIvyQmfNb z=FigjD;=jUnK@Jrod5LR|L2c8n*g8YY#Q=;Xn?3lv+ynx=)*cc9E0BOpLRl+O>9o{ zFC#Kn)`oNTULq&>T@T)W0~Zo=#Wj>=JB1ux0FpFW@ehD?_b2AZKV9r^e|N`GG@{=` ziCG8lF3{)P8_OOarB@Rjl_DbI6PYQ>k_!58UCU4Se@4jv>6`!KM?yb<%1h?gEh+c# z`q+0P!*2Do`yylGUO@A^2B4HwE~TC!&PG5=AH7e*tGKCCdC$D>Ypjq8b@^7aW}1wj z(WROvEr-KWXDqh@>6g4^k1(M9_eZ*f+r7sDc(J@TdQpSebF}evjYoB%O0f;4mA?9A zHRbx#BVaHS?}*qb14>?lbW2@fh>^2=z+$d>_Urlnkj$$4#(||S$hTQ3=aSj~{B&=n zzsk6Qc@NRg!Pk5ctJI;y=oB30-RQi2&!y;f{v2F}j8QTg=5_SrdO89?ZrRL3Z^as6 zz(d2cqHTsb1z0o+78e*ynv6Uf>E}~fdiv#_HZ+YA*tXRNb>)8}CHkl7`fne}BZ1gT zrch5WZiZ*1$SCPp5kTh{P?I_b6v|XC z{I8l%@2&720!5SYOfv+X{AL}g78UaO(vy=I|27hc+V-Q!2SB?*; z(FYpvy?4j89rb6$m*cgK9wF=sQQLLz4TdZtENZ)&N*fYi`1?YneV&ikZO!H>Y|pbd zf7pGiyVt|S?>TV00nh>*AE#HL=HuO(sfhuWZsS)80iaYLTdojJ9x{>ck9=M0Xx?%E z-9QqQE74hTVr|p?0_#I7=2>$lX`R5l749w$ryfjEYN#^*_i+3h8Nt8qW}gK76FWS- zN!4*#?p{f>{*qfx+~Y-9mqwj`VY zBy=r5&5|dY-XKh&AkS4l*;QCS?d8A)p>_ai00m_@&eiZWdh3s`yyhi6?!D~{c?TQK zS5Gv4H&mw{aoCne&sU?+`Q>I0Z{RC%0YJCTqnI?-Ch(}QE_i7o4fWTldmCi3!aEP@ zo-W|sd}H|qYT6UYb1C9vyR*QLoHWj}JwL?_c2TDo5z;gLc+tZZk}2(!Pm$gMlf1ZN zhbP`x{LwedR`R2YID zY2+%7D5npGQSn&eExT>c$CGS8rQXenvVKL4)fR`-3Mt33sc!6stBmu|<64NOc1Yb> zdKXgYwmr=XS50D06}C$dLQeD^08+P5iEVd!)|w!4t06!Db+kFO@w15a+*C-z%a4aq z7?ecvNk1R|Myg%7lULs@&U?*7P!21x>?iSDml`95fHa<<^+@Me4~#`pV|=17lE-qk zDBg!uv>UDAc5o?J1hhGr567};)Hz!wvOqIs_}&o8i;ASdE(LWZDuzzBM3z5A*51&Z zbnUl+(cKfYt~o8!1y2l+;9^K^W|dDISMYdu2nUJ_pbWOhKCkO#0L5V^J{=#Af9MbX z$Jq9p(U(x>G6c%h(Tm~XF`qGhd+QLAPQeyR<+Lv4uxB=1t1GRGitpZw=@+#M2V+ z?ubUQi7S}2f4ZI*Y%6^q5K4Sqt~cJIy?OJmY|8&0n$h;XScL>mNHPG{AT@5p z9Ip#lm;v&PfZzgws^P{MK=+)QCWp=uhICj@d|xc78olh3BtuRDjjF!fn4@hG)Ow*t zehm^h?ZUUmORdcfBP&h2VkemV9%L3>x$G^k@?Dl^wQ{pOAWQ~~1XAY7XH1)Qq697h z&rIj_r3XYVr|EjN;aT%OR|t`SqBqMOXB)tA#<5V^4VCs`wgvXJxnRjnWo~;!QDA(W z$9L~5r{hT<)o2UQlIj|#pt~G#MOq<>Ar3%J6#*@wEr%GzSkcOZ!^*CO!Gg-U<9$>#VjFn4i}1hH>DS>C_WpBOkMqx=|96~W%w!{bplY6z%^@4z|OwP_#pGvA=tR z@1g&tfOHO-j5ABcyDdFhQTGZ?aPD0@H-M23Gz^%am3}D*k|OL04s=a?T_oj(BN9 za`ZEB<0jpz+y#7-!wH*NvmR)AmFFN`gX2o*rQ}bF-$9JGh-=BQlRMfiBYHwV+xx?n z+X7Z!9iq@8@_?M%(bT;f>&^xw1fAir1<3N%6(diM7$nSUZ8eS&nSN&D8{>|@)?Oxy z`Zhg{q@P}H&{h0@$?59wB)R1~j_Bzl@R}&y*Ygr{bw~5_vDkAA|GT+*W}BPr&)+99 zNIQ0_c`3_Ae|UCRL8kbIg34>u11F$`N{rR-;vB>ou-_o?K;dZC4J8_8I6}i`=-7+t zyQh?AoD$I0R)v0 zPP|c7@@Wu2VQQpYo$2JHATxI?^}=GCcOmyXTr^UKh-d@`i=e#wb}y0iaaTKO=Tp5~ z5AZ11p5o!)>py&9SxNXZ_ujo+xkiawXEH-%@9bRd(6-z4tTK>EYj9T zx`lP4&v_{sb9m}Q7U+2u<)3zsWL^g>{5aCtSlKkP^u~)c)~ovGCgSp+$H>s|-k|Gk z{|R9K4*kC4MU&K_kz?$t8Jk+VS?8fvpqx?e4ipVOor6y>TwhCM9%=gqtn8 z(G}L1wSr9stW!=cOjWn%s})=~rzor&G`DSi5$PKBFIYgfwR}&RikJ7@ff;hO6 zF2sF2)4Z;`k9}L})s;cw zGs#aT3D*bq25c$`;X(s!dO>qD*Gl^`=m**d}4|+u$(DLp35~h z?TH}Xh?{bntFuYHQ$Ee$_5)CK3Ftm}7r7=Rz-qnd-Q?T!=dEk+l>#1txrb&oPl>*A z8tRpSSLr~tnyi`PHhMo5L37^KV}+XiJQ(L=`*oQ&A0Hg=tJ6HedYyZoeCQlJe>ynO%t1Noj=M`iH`-2-RcRv`s~ec z?^jx7dMm4dnBYCgP#Wiu)J4*f5+If`P6NGgw0h}#JO4LKYK-0w= zijbGbTmmMKr|iBL%~LhCmkBGG%=tY-Bkz}yj8a_sLJy3ThkD_ZUXg%|-A!VQ{+*LS zk-hOr<9AI&C?{~Dg&`}bab ze2{wP7|0`&G8UyC^8s%eQ5_kRg-az03mkU~s;5`|xOMF0giaMOJj(ud> z9U{@Jbrq$xb+*q3`Rc{wE?W{>k6+)cJX2>wLUG8zZ4qd63UjaYcFRnMR(?*v1;~GY4R>!kXqvQ4pSS=NT6;JaVs7{k)>;w zYCJnVTxX4-6AY;Z$BFg4eHPaD5qrM!IYsCy!fyPR%>}RtCtM2KF_*d;gVfS>Fy=Ua z(y)s8qwL;wgDijOMXc$u&>fLR6XI2<&qxQyiVla*!yESUg!JOhidZ9vQQZgF2A46S zKUDKKSDmq>%N+7DGp!C#3aLWwp05}goEx@Qr zM#3MNe8vG~s)9WVOn6W;9PE}vTsdja=l{5R{%~UbPiIM*AK>mf1hIzF{>@P`4gd0E zKv*Zsp9zGB7}OZ~3fALlQm&Y>!HrL?CCKP}A z)&A|9WW#{~ew#S|zcZ5kZyu{1um&azY`^>G>?i;DXa2VVf7?gY&sqN4xAb2Q!6yo4 zcgAZ$&;H`TJp$gg>4T&vzv-m_7d(|1R|ZUFiSm#Q3il(*N%0 ze{o~_-yQug4&49e9X*==T`VVFtkJ?$rRg55^NYVfU}4Ow$)!y?pde7`d1$5EWTupb zGzU&r4UVff1~NCkY`0t(MKVZnv4zqk0`6e!>!<>?RCde}D&G5KPYEa~Wiw(5GW`+& z;6VWfhd&a7cF0JN{#I)C9~Pzs1W1*h8g#HKME(Hq05(7mGnPf}^V4v0-?&RC0T8@5 z{4NfF8H*C><|-E_2&VOk9EFANqq$0~Ir5QTff6x6LYK(yk;L{7yU_xIMi-y;yiq#I z?cwTzDG#Z$g+KsUX98^30r^#Vy0PNKtOXCa?|COGv$N)xXjCUX!R()m9za@^)&OKV zIZeVUiM;XG^%}hP#tN&<`>9X1As2HAv_kH_gX7v@a>s-Ry4mzw?x3p#q(5>wUzxSL z!F#j2!d3tqDBEX%8D8F-wgAVAW^|v+q(NNPBUTsbBC5vO8aqC~CEd}`(|MaWh?pvL zOlmq|R_x;Ycmt0PgX(OfP3oDnsl1@!5K?;mkcAN3{cS5+>oJ<^n|>xSUp;e87;yUM zn&7AY^N+>6K*Q{gnhbep(*`5Zt0B}&WpjEbmNnjcS@Z-->nT6T_k_XW`_QwjP-q^uJ<37MMGyd8k63lW5ZiUlHiO(#yAL&C(0&kozT?CH`3t}WUFphTzhCYxng2$zaUc59_|y%rCF4PqUAS> zj@+7QC_<=UPgT0#A2&MW$UpX>yl}dA~A%OEvNC zwv+)2D}mRL%L|AN0K%fIRBC2NHG6a91KbHQ7zg+@hr+mnh-nhCpM@%>2wGC|4{##E z%1qa$nmnp=RT7~d(~+g8*aNsCt&%gWEpsrtqjf@n zt3U&u+>*R2Brz0t`sKQ731E2W1{?4LOxJBsC-iN<)D|`e)9~uuLr#>&ou6if*;t_8 z0c-v3;UTsAq+oO7SDshhJXipgf;FV*q6`{xpYATj^5_fO#km^~_BL&fGCKbH!P0e* z0ZrsHF4@{N0C3U`=`NNsOq*~Gx*s)sa-JV_1O-5SLyIWM6)2fkl{R`mUg z5^L9H)KFQX50`9x-K+B5epq8yqt?V^0CK3tj*0sU#11ZPIa!%=2Nf}C6c^x`#vtM1 z_|4@)y2_e-7NzbbdQamLi}V)vuJ_Kf4JKtE+Hl?NqERWmV;k*Ssvq18-!vijs{eKZ zR1F4IQ~O@c3^5v1zC+xhlLTRQ0Wz1NC>+9y$*|9n{`xjE<;~*W&sD=$#OY*XTX4FE zvhq>1H=UcPDyA!+0OJ=7rj9&RuVLp-Y=qe^V$ZM?rtGyhLS1_r1n zN}BA?o`^`WVs1*R2lT(y^l7FpyITX|{Z>9-Pd)ei`Py7SrCsxRU9r^%9PMI4AMx3j zB&+(jCj$_!A0W?D*(Z(N45#E~$~A{MfI`*8v@rEXRC&a3mM0%ch36n{)B3n=sg_2a z@+7qx$O_2^*?E%HS2sCnRJv4$|MC2rOYsZaSskHmZK3aJcSCnBL~`VnWJpdtdCIS0 zd2W3eFW72&AbZwd4pGh&nO7WhIqX)2PKpS*qk^EN8cl9|o*4RKDx~q&^k^y-u^u3C z9Df@9$wT-1OaRg~lBzF}?0`5u3Fy6dMh3PhR#cP`QVY_Hnu99>s6g{*#HI--fkr$peXsUpIp3t1ieo8Oa zWCjJ_6)ud|o$a7IVzivj4ssKJT&^_?lOA9EsHF#b*Q4`->6&gC)O8S86NvVpkynFU zTHhB=r{w6gc;M|=2=*w|z-v&*)?>|Y^L$lm?WH~w=@;%S#zvm~-`lQa1#zjxWh5op zm#GGY4<7#=r0vDCot$H6+rPIgi6qdgfB%yCwxM)nw9ONB=eA70MpdAGoA^`J@WPi) z9A4TsW>M4b^gc=G<<*Q|^3*R%^%|;(dr(uA9r$gm_(P>cM~fZ7Fh=B95gSe5*EpY{M_L-6Pcg3L;D z@6-jU3{<-eA(xi1)pXceADtH>8W^9|=j;2cskXHYXgM-TWV&1Pp!E7ci>nNl0o5n)zqfb373 z#cM5^1v?m+&y!_GMqT5(!P%QNURJ7`boPwIddRqb3kW7JP|$i%bvs61o-`x9gBD;S86T{(Y|>UfW75 z%JBwN$E}~mhqVJwjMkc_c|i3=E5Lr{V)y~VDt%$lLNBI~kKGF+HZ<1E`RupfS`*Y= zO4OohArkx(T}bQtGjx+$oG)K(eHKO z3JC6+I#nR5Q@#C&`X?x>S@Zs-as4M>{s8=wQst%~=>AGSqPv0!0TfbP;l}yva<`gK z4;;ueorihh=4IeTUN@9w^_b6QGM zMfm#M1yqT|R1e%^SsCGT?O^Jr+KGzCfy=}_L7UrZsvsLd6L-k9GLSQQOO^`c%8TbQ!fpbHWoULo(a# zP4lO?C;MP!DD~@%?>IWdQ+i3qeH-J=yk(7z_z9)D$A^Hw2T4iw^4frX&`8lbHbFw0*F!l`^;IU2 z!SyY%0cJW_>G_(7an0>yu2Wh0`Xfh>ZIutkwep3RT`6PcGwy6mYC7aZkLp}WPimgA zwF2$^zIp>ScxZA1vj^HceW9XXx_2?)G}J(jW!hiuBFgaUd1G=&19o<3nZHea1|kq7 zWt3cFtgj$=y4mvGDK0rc7a$R`MdY0cjz5mHH6 z%bGn&iBQVEC+lGB+ZYp)B1?pjExR$sKE^IdcG<={$TApa?86K*_v!b&pYMGi&+j>| z-}Qa|<2ab*bDrnt^M1eHuSGHEMbLvx*g;;;NLjuCZM{Uz7P-0w1gsz2`nMXm{R zAf|z!Kr1dAs$A;v=tf-g5=6N37Mb|0O6Fr3GK#dFKn}mBZhrwin{H?cvPV!W~x)9UF zTj%O--639f8`0XO+Go2FspWV_Vsak$te#`N7Cq(@A8T2;u=D;@YB&z)Ko;`nDY4;y zP|g~Z%wXy`!zh9V^wQVdI59#u{7KcWK^j-sGRL4+y7V)n*QWJ+n>C6{z-ocS7mRCT zN3MGxNJ(PwVgu}T8J;%2g^h?l1d;&{5TaLQDY~kzSE~ZL*91~D*-pH5Yg77XRpq~- zov(Z+gfoN&q3d-%gVYoU^Tb(u)YLVdY6VaIs^B9^_rmQL=9*PT{TcB$by82VfJX8V zss6w@y+~NmehN#Z%f}!=7yczW_}z+=_wo-g9q3SVB*o z>VNUWb~!ycya!0B6Xu+-LVMnu}8172@z&BA8jvwYNYI;cYz#7mPrW}X5u~3dqgny?pVb>mWR52IK)yFe~?Co z_7d13V1J^(B<2Tyu)77ak-TaQv?79cEKYFs6Pf}4Q> zBr4H(;6Q+(UPwY44JC<`QL_p^qUB{pH*)|n3je--mD}l_a_UmJ8$d2T4ZIPF?3kJ+!YCC-mO-2C zgKJA^i`k5PKnHf!e7m?W!px!jko3Ljh!R$}J|Il!_gT5dOg)rxME8!>+I4^av~Ic* z1n^ACM<3U>V0lJGvv}8!7L?zfUsw5q$MJ>*JSzUeH*={PG!lOz>=gPwmBcMg;(K4t zs3!|LH*Fe5-ks-pxBJibs((*m-BTF=j^75|+C;UsDasX3mtdGzzg^iMX>dbj+&*O6 zPev0YFwQq!_r!^Xa5FzZ?RTuo!NWBXeSChaFGx<#RTyDb9XTvEo@5qR?Y2I#)ox_f zDUG=A#?c}+?ArPg1SDZ5#0zRYtfm37{kmYhjQ15Qvu2MR{40zo0IRx9?12b|uE)%G z4^|tDs`>JiY*%53EyzwQImy>`qIPAU%5Kz7+^~iEg!^Wl{w5Pax zhBp7JJKfFI7%53GYj9 za}}pLW9JkO*Xy|&K%k>Fm4K&k#H73_TpwLw+)juW)Yi(FLermEPqqwbXXt0A9JMn4 zoD5?PJ9YN`D&;8?3suOj+y76a9ovg9M!>mtL8&47c}}2DN^a(Yb*B9+L%-0|uggCx zy0#OmwUk-@7A^VzKuBDGQTdr#6~oz)&tS&JOj49M~BhDGZ)dpMlL7*zS#* z-P@gG zx|T$!&wJJyTFxp|3UW$6B895DH#%fa*egG*!c{y$=UYm|M zl5klndoqwHHtCkvBzMSq<}$p=P#QIRusQ;tPKM5;Nct4>e6vhfc$K*I56|)c^K95N z|F{9}Uz=B2f9xPI$l9-T{TS(+BV-_(xE)%?S^X(7yyCA{c!Pj^+lI~-c(}xISq8U1 zEQx_tzeCPD0|DbK+(Trth8JX)$Z!7A9o!fXL3;}z%Qw$_ zykZOJiX{zrOPK1y2gL4K)*B)UBT7mWaUCCdMLJ@5y-w`$IFFvKM;N$R0-`Z9qoYWX zQ1Z(n{q|^^Y&+&G+@NsJoM~31-T+F88{uSZ??5`qItJmYoF@d zzL=~%T4FK{Y5Q4DI?7OnOpzJ^aPFX9GWy?_=HEy9-yS>;g@4)E=hWWeEDkEw=1KwL z*DnriIFO7i&%-NR1XczlV5gOuvK{(T1tlCuyqSGeHpgr|kE$_0x%+uk%jZV+W9Gb( zsAAk`aHjl=VDgHPoNLIE_sSr8#gS?&;K^KVil0Uyja!teQmjyX%0G%n&EjOG*H*(v z5UH(_H{y%~SLxPNjJcDPb7 zC(A{-NC*E8MgTn3Q_#r!+3Ywl=@$^?``rzUYc>lT5gHQJBq(3q1X#-zCj9SDV?-Bb zs!XKNPnv(m@NkuCe@nI-b>8!(jyCbiXa#59y}o8lgzaY^bDl{@Jzm>>RU~r@gp1p{ z9Lw*?}XvsfSQ;5Ba}Lt5%}jr7Sm`=2@px?x%feN8NmI$PM9*cO@^KHbum05 z|L)*VK5N=JE^HkEI%ZtTRRuzUNq~9QVfJIbPXVU86ZJoRiO1Ng@%ym}OXX2uRPdH( z^`oVs0o>Xw?T*K{AnIar$o(p)aLZR!Y8=qHUV!x9FsfXgkJ!y-1A%u4>!s1k&#c#h z0bJ52E0`FRvg@@CoQF}DT>3s;j|kfM{df@8@}@<;WC--yW=BP+1_helnv0k=#=^t_ z97oyz)wTp){TAopUg;*8h3ChEvAw_Eh9mT9dc7Q^{Wg{{epn$I0>h9f%Fla_;;Vj>%T45XX zg}xVX3?t1j(XNwEQ>2w%^@cy3$EoA6n?)ooM~*O3{h z1!yPdzVA|*#x6_Qe|DSb|EV--YKzbZx6a9Mi=!BCu5~|{nnZ#9PAoHv*WRC~b`wK_ zZcWLg>vbgwsGnLX#1z%PJ6xX&b;yA9g*Qmqb=^?(T|2!d3`mP#VWjozfgN#*o~cch zX1>S-996;JXFMWuFo&5{x(77{ z9lSn#d*&722iKKdwFt{V3Dj@|A5b}5lkmeY{jBtw4tQlHTN4g^KG!r+Hp#I9ww=}S z7kt0w9an7`rjv@DdpBNQS;HUxvFvXjm`#*gkdH?Hy z)YXXMOtg}I|8-!rNr$H#=(Xv(&bXRXynZrVJ{GJ6l3CEiou4VcxpR+wa3-{eQwcS? zaYOhjU|VZoiN{*JYoay5f4JCqNO)%eTxawmQ1!-4x<&ar+Eez-z@2e^Gl}K%9rdn= z3A7A+Vd`Mq!NOhk&tF1@#=G9|Ah$#$BCyUN2S7$S?Mo4A6bP|S#4t;~$)o@CuDp!3 zk|zAEp8@`|DF?O~YIbYFlrV*!R<^2jw>dB2pK+Ga5~a)AS%SgS%VYriJLFCaz|1b3 zDOKZPdFwD77nti?ZW$E}C2O*=80e)+3f}Uz%nk}ThNs6swWEi9iqq((qHGBB)Fjm2 zic_+={P}gKqLI!S!<<}4U*g`2Zog{%CI%{ksae(AjboYENFh8_E<3-zIUdFU-Mw@Q z53ya9Xub!Q@OD!Rg0Rix-Q44ggpVMN`T@AOpXNOJaew86M|7*gKSPE7T@Bt-rsFy+ z5R><9w4Z+@=c2~)EyG2-t6X=_c#OJsuI%d0sYXUNe)(&LAI3xHKMCKBP`u7OVyZy| zv}js@SI-UYrJlLE_WGU&oxWEfq;+=YbGVV!aYKzGb0pz;IID8_m;U9>1;!UlIINGndxyRFGMQ(ZmE8xakMSiXR|Jr^| zp7{!=?LFI6T9%aZDm`r+Zcov2(N4G-4L_GTw%)j2Vc#pf=Tj;RpAD{SYkOkjf_q-` zH07a1{@QrO2WWa^5@2p{8`x4l-{Vk>Dr19R<2MpKIGu(DBIBRBBg3V$V0FZ_+L#2N z|Fhfuqo2;1&3+$?G7kN(y5;^Lld;{QRq)XWOW;tn+|BjbkN*kj{kYQU6bBR#uZ8Rx zfp>2?xLpRghg?<+Zn&c0|Lmqt$~D>%DU)&|^breBM&{M8GXL3j;7|^5UMlzhcq|{k zc#7x-K*^B|`tiO+AYUiVFtA}O(%k#EMu^?hzb(1``?9o*3*6s0T&vZ+B1m`oM#vk6 z3}m5UZ}ewn$li@hp9KC~sNMe)l?O~y0h)mDZ9wBZ<%>teE1T8S%G&V$XJ6s;*KA-% z>mOilvk`c6e~=j%r zTxU`D*neM{e^AY~$X^8(?@I0auXO+6QvB;G#eD@9B;iGJ`9FBX{yW`0#OppFl9P8o zFVFBFuG+s==c@cMxKJQ&`yXON|9fF@Z-6_}YxbHa)BmrvyYI<#PDxMK;^(7(@2daw zq8-P;1m^bjzm~LRT;ljqY<~CkKN1wRn21{2Y`5Ar4GkP@wxIol zv4^vF+4D5oCde9pO;^tSmwsk;nE9NN>h99o`G2rD|JR=g-SO`^$&#*j*3mf83pt1u z+Zhc&zU`-f`LBKE>nq?Y6kPf`E#R)gbO`jZo~;aAUxx}ZzIZ+Dk;O@06N>nY?)e!& z&SYG-yDVn=6DpK!RCJ@?+ffkcr#?EHDWuoaO>yA})r&oUe;xZ=Pr#O^zAuS!kNzGA zD9=i$$wFiLzvF6Ds`Lg<75Ljv5|k?pnzguo`9+d~2y3iuUN}?Fo2S3xkN2Eik)8t$ zu{&O)qT_&2^L2NUPO)Lpg)tN> z`h3@MytC&<_#EMgVz+vz5zD0cnqCrzF8dXk+~Sc}d%)r$zc`&(JVjo67!uPBF_XyYd*M8R;W) z1{=4Mu_4mrI?L*a_Mk204;=Y}Zfx|Y-j}Pubj^BLw{zN~+_KV5Ik0y6W!+lELxz6X z{z^ehHO*yz10ag#ZXyQJ(^9~Vkl@rz6+Y&(btMW5l2#;D=uQ{jU+PK8eF`K;1gs)Y z#3CQX=sw#BT9C`Jqr{|#tn-W_6JBR~@pxxaC71bW8*^S9Admb14_0qRg)B)lbFZWb{^IF5u2`70$!2mp#CMwHG$wA zjlbVwXU)%RYA#PHIxuJUTh{&gf!VrI6xd`#VGY;}>u<35H6$koe9Xh{&LIH*J@uu| z`qP+yK*ddq+40a#>O5$uF zmxgi}nOfuiVdr3L(m!@NlvzAZJBl-D4<(a$;h_KPtFleLd%S&tAM1xbwT-as_B!%YtRZ|i-{`FL?pNW7GZZd+7f`;rgm#07v+Xa zE|0{l^B+xAJENNnPc*rtNUBXnb4xwqmUdffiWkYcRHbfz?q2`NR9UA>-an=6dq3%< z)tnTIlCJcf4opEd-=df$@Zc#92~cuIJu$AtM6&+qSQ*O)3RqGaV4vyq!~ZrTG|_&y zV1W7d{N&r+=S*LFNgAIEH0Az>L%8p9$d9*u>UzO@gNxZwy4{C~cG?VkX19!3JE4Dw z<00Wse~7%XX?R#r(v|}~)kT}|^{{e1)!^5)m-dC%WOGP03HiC+a{YSS&BHHW1p3BZ zv6qzv+>;Os$4*g(b#jK^ywAwj-^-mPL!E+`5Dk*R& z;LWlXZRj(Dve$WE`7xFGxp_CKtdZfr~EyFQWRTik_# z!GV2O{gG^Pg<^Kh9C6X)&N-iA!^p(2GX$u8zul8e-L7Ty4~!(JIoAajZ=+xLHKu%z zPw7^@45Pj4OtyN#{JB@_mGMouq0C#}khQV9<8_eOlIaSWXGJqfEQh{dgXtbTFM5{y ztYp2vg8}H?Sj{`OHxCmiU(aL+q3+CIfAa=@I#^uFScCNTZJ%<~0uz7W{C4DDbaM=A z?3Y%Di*L`w@bNEO4?Sl%{{MbFrcq6Nlo6!sbF6(f_cGXh(P*~ErIwX}Wh71cn7yqo zaJx%tFRHZH`ycA38;@pRtK+Cp(_cGJ9l0$1v&{{fdQ#kP2|>?tW)Zjd)rFLGA28la zKFc0R*on2u+7*-2TA@1#Rp1+V6-ECjI$=&#$EQ2i>36%4>Vdureo(xSU+VOY2%vmh zcxKmAMt5vQepu~Y1h{~y`L0h00@f&_t66q{>!FejW%9cL+^Bo0H%$;q-YG?jQGiq4 z4#3JQyvHRw(?ya$-?B8X&@fT<1~Bl7*?v{K?j*5zIk4qzgW}JMtUM&|nqyMvcEw*_ z00-$iU!-!1^b5*S0n$lE+N8-GeZM(i8`rG{o#+xp(4k}$AA3yzHc$)t1(A#0+V9Q^ zS#kKaM{-oS9R>PX0QMY0K(>6+;OeEK6tsm1C{es#^j22NFvc$dD*`t}c#_66xVw7N z6;YD~44dCveKW<)n#`$Wyu_14h3bpqFK_j2S5hl+JsxdbBax)v?>Za3^bnNVs#hAQ zMhhh164UM6NNn7re7yXJ7uL?iv2Xzz=sN#1`gqKPX~(MXQQyOl0PC@O@S3vGw}&5U z>nd17wjX6VCA?E>-3C|@$v!Y-^5MbOBk6s}(Ww+gu-X9mW;D;qUW&sWVN$4j_jbVqjm^E;Xe$4OQ!9 z+-#E*9QsJuudh9dE3wLHaKFAim36l(Q7Ynfe{jR*rqqtr{W=U=me0o6UZfD`0#izg>f$*Js|SdTQPC=NC4Il zanerWA4cR#8R~b7HuJg-e+3>{W`VCBp@Ex9!t4>vRhJiIU!wl^XV=xMK!v*b%7K#` zs75?cK{eFI9@9m37{yD<;Wx~1dB$CiIJOvb&o1(!XTZuAmq-J9%NjT6tV^k=(S#2F z@8;jSR)@(hV>RlM1$rXwyI`!w4iPLgWKXT`Ia+ztw_hd)TAX@HfgrLHwj&+O?oWNc+c-Zcs%!@;wV+@SAtB` z#nGwzAz#NE+dh(HvT^5~T4p%Keg}=fDrnYE6StMgeydgh*eTojRRCJDfsz|7g;MN7 z>?PNN_x0#!(G{;b>uDvS!JgI9scDXrB_Biwn~VCb$|u(CQ>bE)Og>oMX;>S!+yQxr z4n!#ST_pP*POZAmMCm(O3Yb@tRz#sz0f<_A;R=VIt4X%z)lQY7Kq3oOIq)UZ>Tcar z;v(}dart-E5C4Oi>rSFzqmr`8#OdT_QR4+H8_N8JI&o$9)%+xh6SiGy%XGwN&`VP8 znu%ChX1GA*6(X0fe0alNW=;6wF~5)#tLiJ@sM^gy?)~}w7l1I&K3Jgo$SQB)k&6IN z!zX(?1fYMI(VzOYeAjj%j%X_?`WfQ4y=3aNBhF>W#M`#ShJP&7P&9cg41u-m3l(Yu zl7NL3q0FCQPehsbRlLY(QSb<#qUt&%521!n_s|YxAzW~ugBJIGkY_o7+3k1SGmq$#h`!6I{+Jl z{W}mUlX-!vi|HLdT;F{2Vd{_N@|jNCSfz94?OZggw*eE_`ZCYuQ_Q=`pVZW`Kw41~ z28LvfC`}Shlt=`qhA-S)UP}Cta4DDF68o~>)<2jJ#kItnft{{3TEPVG5Hr6>A8@UV z!!vs@{1r!H_3s)ed#mD@V)ZJqa$I##yH*kA)*)#&rk{^SYK@r-mRTVWR+h6m0In;Y zS*>k9EZ4cDVU?@9ax$*B0P)sdx@j@%J;vAjig*Bx9aKi!cfakJuL z?RtJokf+|-AY=5Q_@2~n;!u+x+pUKaen=H?cNv>1OGypmE$r%0TW>@LuBcAJ1(#6p zG~qGmkY7E2oIR!$W&14ru)<>eO?oid&hwZ%8UZUe}?yF>k$d`ZoP=APB%I2?c)MwBroE2W# z?=XasX3hDaWC6+Ths1~mM<(!WlM#)A};z( z=b`V>j;41*mON89qd>I2J0=Z#V6Z8(=B14*)Qy3R_#Sua^e_> zfNq2|6l71i_dZzFb8_?tTz^6a(BYW~eU!oMAvp~!{TU-Nm0#eJ5)K^Riqtl1ViVMs zArtX^H%p)crim_o6#ahA-fQ*nJF~dA+nguQP_en0RD_=N?9`L6p1tAc4qmwKh6v4Ro693|UjfX*8POv(>IXyKTu3YyYu~x006-k(S#r|e zZi;}Uv3?3l%aL6DK^uEd{7bhZV4UR5Y%14W5dJiS44UO)pp_FQo2^t1{of9Q{NabP z%KNRnu;zgkDt*Ms@IG}+W@G@o)}p%0F2rK|;GN3UIpZSsJut}_`lWToZ6tO`%Tc?F za?L%O0KMU|JQd@bqp~Q?^H|!TH zP0Q}7Ghx4k%ty{+%RSyTkh0DF)NNDbXnNowVG}k4Pa604;=R=e|o`2=YckF%{2J>KbR5UkO)=)Du=LE%< z+mwYReNx`!&Cc&8;`^iCI|8nsIe%?M^obPVx=C`eb!5*6dmIy}LQc)f;(1F-03q}} zx`ZFsvt5VAyc@CQ+CRHGn`TnX(MKeXOBgOs)wy8)qS-0KEB^SsvfN@t!1*jzxO9f;&s>_v#X&-;M}+QL`!5dLE&=o z2{(9y>qw4k32mgLNDb-nfmJ7hAaQq_^Ud1fT4m`g3hU8w9156ST0-VoxUQfEno=~J zXFhTZ4#8;C2C|ujIB7t(xtNA*x&Y_2%JZ~CjU$H&x^d1Nu4eUa={qZ!exGCd*74-| zC+Pi5X|_c-rn)W=AVr!R95e}$H56+w66F{SowbCcj1Sxq{El~`+KV&-31jR|4vbz` zDW~+Mro1zhB<+Eb_AVh_ezDFz0X(+Dzdj8DlV!OY1`;rK6sO2oCz$p$yygZ#*I|wyOHF z3=3m9Q*~oxk4PqDCwApmM-M=3xLnoE=enG)3Bw0=CP=;dpCh_^4QpzJI{?WC`7BJD z{1S;0uMR%@xp^03)kNOmxr9%CdzM^P164|E9&ZEn${ZeI2JNWrHU$p7WPs1ed(L;d z$Z~9LX?C{OU^;f#zAyEpDnn!zy2g&0=k-alYcOI^eDS_qIcu z*GCxp*03gg`W!9fo@ZW)K)N>TNm;a}>Z_6xmAjU8FF$1IO_W16K0P*je8_j<(wj4X z3x|Yk>$eD}npL*=J5~zgu%38ac(*!*mn{!6|HI<4viGEvT@&v8CjxQIFz+AMIo*@EL0MY983J+ zdPJ(U4aEiQzUV@DdQ10)?n?x3q^yneZ~JwxU6%S4EmPb{zv~uPEErWk>xJxZ{@_(R zof;5WF?018Qo$!bqCJ9v(^JKBX~R0vKOXE@^2qPXl_P6SUm+LYXBI>n6p~U6Ou)Xb zEJjX^52;^VJ~JdvvISa|=r=SA?gseVsv0Ac6yLcxuw_UjB{9R^&c?;e?X zC)@CMpG7PaW$JIDEd3Ivo^Nw^rpPCtjF-f%1pCiYueP$iJbCK9!>hT~YBgm~yAR{2 zquwbMkCu9&b?|=lPDq5f-^|VaDUT>)4nL$A(Em|pw6UHv{2OpbJ>G^`sYvrf1Y59s z&ls~Vj_5Ui7E|;)kkj@=MhEJh9qw8hWm*;bI!Q{X2xqE|&{;Ktua(*!d}j!YnVUPc z_u+!i0<})=q?p4^54zUQs2|a-hNmB1{FTG;HBHv(8r%q(M6jsWnqcRsa1Hb3e(F3` z_({~b_8B}zH%=pp>+ux_K(#Vx-d4M}2wv<9;#_ zh|!#6W+Ex?Nx~z`wLuM5$E;;Jme*#?EVt3&{7AP6O@>H|YY zD)!P9>B=U+67&-zvDpgyJ8Sb2OEf?wvg%3{!-uw2+V}gmRR@!uIstlZZuWG*7@$zC zX4)yYZkBJ%O&lOXXHOiDPR@=nZ30C)_v`axS6_es4)wzZ*B_E6!Gq;c%6&Uuy!{+} z)yQ0cZKZZv=8cEh4hjpOnd8}hs~ZVH24X7 zBvLImG>^j90rrPtT`r(MN7|9}tTBaNv3g+AKLU5YlvZTISY) zLN%x!a-0nDip^gspX0asB(Sr*9|z(LSF@pC9?u4G7V019-_B~TDzQMRv{-Yrm?>{+ zT4U47w1%lLe$kf@r}izZGid_xI5X>DLe5-5!kaKlS6lNvvi~&&jKQHP#%uw$h>`fS}3^aqxWAz5KGPh zi-y_#=JH-bt!km<0V9!R0QUG*@EKB!O{Z12d7=v#YYDQ1MJpZ7PEXK0$D6^;!9>av za)bPxXBv}yj{=K*gf1&OKP<7X1A>y8uh169e%3ddyf`|@dPAiIJo8-YArOWbGkU0+uC-|b8kd7NdIbIQkJ*Ntey0I6-|JNXMw3CH4tXyXN8`q<4~#BKpT(Wly7Cje$eA5a8*^)%JK=n zK((Lg`_d+CZ4; zNKtCEnPd$>9VO^zYi76D1gorYt;h$xwhpc^RB2u(!|As_-4u1F^u)2zhgdS-Awep_ zbL(&TNq)L0>eliZfuC(IdcAd2*1n+n!y)j0VAr1){Q#G z{(+w)u1?KH3WqeqPx+jcl6nF!l-OP5)>`Pkeq>htSyU*pZ~SKVJDYc(O8KGUa5+~G zxTNztVva6CP_O38m5Fs3WD;Zb>h2$QjqhtG-rwYQi{0^C`C+cjLt(MH-;`@Fiyigq z{49=86lXpaOLfg}$W?=Y3~wKTSXxxEH*T^g4ddo~b7=vZVV$ETThtnM`N*kN+m4G9 z?`YB*!5hk&?nilxdARE&hEk`29Xi(k#|t3MUapIo2+KG4>44TLSf#%})Z2f~S7IgN zU6_!Xc@I>2I1A6sG}yix?zMRKqrAB~)mJqTfyrU}CBxAbSV@>;gAm7$Y}e~kyr%AQ zcJfL+%Bs{gUY;RLQ!=7hCR2+m~5%DC#&% zRY*Jz&Cj?Fxw=T$4U+_1#${y)CQmlI82av_hjmt=B;JS09H)>(#$#`U=q1(Xf znuiP}LS+zbT)ktcr?Mi8$~q!SIsBA)X`Rk+z+X#L2{4My@h_hR4_gW`_nK9ntFq|$ zQT4-o#+$4j$FRIkWnxK(bi@IQ81zEC_S70MK5A$$w?u3;s(KTM1BWq+xyv7{NjF%~ znP+;7YfP7CCfHP|mumZuI_yKd0VUpdz%t~(=q^B55VoG7rM-PAhH?kh%K8TN4As;$ zJ;@=fO^be}MxHVgn{u5S=5e+Z0XH6)UOnzHpp=~#Vqw9z(~!;PV_`3zjN9WS00Jk1 zm3ymE4IF8JsTn|5KVV>9Zj(i7L!qzaN`344@QpV73~1v4S~|6HVU>HtqEIPE`F4;t z`$~bD4NU^g>5`$ai>cM^L_q$OM=Z8w(@z~F-?q@_*Q$F7WXJFjeLpFjrOdKB1?xrm zHwULm3p$>Lt5~6Ydz#mmly5#A@nVjSO?N+p)XCjZ6%T zP^Qvkc{K}P__sXG2Jkd_Hn%?zHp}Q8O2anVi`j2G@$fgn&TBrRN)oN zOiiR}UkZO!@jR~_BhU-w8BD~$L+cpc6*y;2fMR7!xh^W87P&xoN9-{p8+ElCyF2aL z^xN6&AJ9qj`Tb17-!OA~X>{3M;2`)f*Tjt|T<>+A(z*v)OR7|qrCH_`+}ygH0Mb_a z3Q%uR(^KV`$!2H^TSYnA(eljCN!bY_=9x*-NlKu@&Pl{mDtc<$nLl7}nk`IeJ%oiO z^ov3=(e^q|9Y6z)wB~Ld`0DZ@+rD12rN#+$USC>3wwIaPP`_-m#^65F$ONQi=CI?_ zOG{pGvpmoqHtXdl7|niP;LZ4b{=I0K5nS^aF}?XziEl|US1-l*s)BiUR<;vdg|~EE zuE|_JR@Kh76%0#5&F;(Hh4)zGdppsJvsRcEyQ3H?^=ih0MGJXx$%Y}Z!=g$U^a~-b z0n<9~BEw!62_P9FsfXLJs*A8${hj+yf3lpXd+DoirVxRpel}|Mjk0Se*a2IFo}=-N zxe!FjD&&FLD@51o?4_D&=yItd*k^7SK&KUNWv#k`5=*vMhB_=OS9_oE>Zjk?d<@`P z&WSn5`n}>TJ6YR~DRR+cTQQQvJzkuh+F`knuqLGARR~quDMt0-f>Vmr8H72wqrYNk zGku%~OAGkvGy*qB`qLh%kn=l`4F}EEJ9AksxT4-_{pIEWz3a3G@y)3wGYGXKIiur{ zfh*7llDy68C(TR0;~-4)ooRFpQs7=D`GDK0_er`Ptg-J^Tz@t@T|Md);=8Mt<|J~Y z8fcWSd&M<)v|&dz`{$uTpF^T}x`a1&JH_$QPO>jcJ7iKqO`*&`+2+#!MRWS^eb?+3 z$3T{UbcbTf>LUHCpm<#$J0ZtX9`|<%kY)4!JkJlz3|c;8PTUH}7fm0fgdx4`QR>yx zjBHOjg4Gwx0`zNYs+Pb}{J#EU1%qCNyYse|+AkO;oVa5w_ely4!%YylzENYHf9?RP z|9kLt)A4Di@ZN+5`pc__qxlK}WqY{PT@{;sEy2jHyW6kbX_)H8E=_V(vcd^}5d;OS zIN zS5Ie2$>dIFZG|_a#DFJJR#jR-E))G3vj+uBPAfU~{RswKnH%dadg_a8(yo6#PV{$h z>*WPC5)YB67zdej*#^C<#`WuiXf#j#J@@oZVSoCizQd0zgHIBilRbe>#i#gZM}!em zf?tuDfIsK3yf}7V@2V%$b?RPw64E;4*wSp+X=s5c1A59as#9T$>&p_@zW?;HjvfA4 z@0p8EdC$O#{u|%9@ZZCkDJK&fy6?3G=-91Ye;lijQywT`SXFqIvXEemz?Zx%d?pak z0EL*zKv&w`{gZ(q&M9L0?#}d?Sx7-Zgv_P<6A)#+j-QIE!)5P!9KJAZbwB+AJ09e^ zQheou_T38&z9~xp1~TIzhL0_Pamkqqu{f5cF~DzIn4%+eGM#UlGco-gIXzzu70LxK zR2;f6Zy}Naia6WI$__LytAKk~#9b)a`}ci#&63>og02Gq+yyacSy(&}{wyIPVp!_y z03?sr&(hCZeuyhbw&UPgZh+`&rKSQip}#JRfUqPq%I}- z(XQ6Z?ckfkfJD*pzbawj-n7NTj`{6|QSjIeK(6NHSJa9H>1f_Fh8y8b4_@tucK~U5 zQAzv07dEVr`r%tD%0Wjsp&fta7wyVmtkRMSdPB3aIY}G^)cU=QuHakaw!3k^UwwCH7QR^yfcbXMf2+cA!LJNLy9D*}j-e!sP;#Xc45 z;L)M3`H4d!Z#uTDHIMIrpvA>lr3ES5E5Lc!0^-=}k?CZlF^Bu+$0aTg@T5y$?B8>! z-R+nhmxGqam|-(s#mVrute`jMBAje17x#%atv}vavPED@u*HAic< zThvSz&%~mbHc|RS+4Q-v{95=B6Rvap?4!WgU+GeV(hlD<)SGOm5NzN~3da_7=F_Z6 z-I%#HxFGb~5^>?KEATwsRelEz=Di-7(rm*BqHgaTGB0$UciA-7!3>b*f$-;a7Cd?3 zI*Dws1OqShe?t~dG+l2ZMfx3_Zf-t;;0|We74kU~MGdxD7tWwll}p z$u?h+V~vg@D_xiKje!wC?MTD=3`z(8=Br$TOke}Je8>Z@oCkux6~jJI zOT~4&Ki^K!%Nk4JTec9(aRvfjDq6idZWb^4P?xQOKLILajp7|dVoTJQXRs7Wy9Ltb zd`@LinM5Uk-~}8-%RPh#Q&0{07lNyc4eWrRMbE>158AlOd`eE)HddYXuv{#YrDOiR zZ>^*6e6-{ig=k0qczF;IhzRz!KW5YUBHc+q1IOeZcl*3enbXPm#ukJeYU50XUaFEs z#`=CO#4@3POdSl2GEsRsYdG5{dKMbCN@zu7n&0@=1h=d;j+s$)kr`v`bmB1i{whP?d(pLCb_>|!6~`;7lw zruIU%LZo}7@Z$Pn~sFHdwXAD~%c5=d!VZX5- zX&w(D8k`UR7N6G~rK4tV?_~(l1Li9o% zL^TzUhQB=9aQ|$+eQ%sv@n=!aFd1>tVuq2_k&>I&3;HV9cl<4zNV-p)Dhv|1EQB-i zZN~9RiP#$D{2FD$x1a78QYjqd4U=9I=Idr<}FVBCR`q?QEO@-03p~&6utZ;HUlUzW-ou z+?URpDsg%N@MphX6n%Cd*hyR=-9A^MI0 zmUYo*D1T0v^$I@iozOE}}4N0ZB2ji2( z+RsSgqzwMNNb49WMDi%HZOiPa@)2td#er>sVvju7M9d#^k7C?`>0 z8ElymYTFdvMP?LP1beBprjq!&;$=W(?H7XT45cu$2deoA?O)Med4S_#zB7UE0JIX$ zZ(8J=F`N9B_ojM?nWnT#UW8<$nfQDZ78gcGooE$uo-bUav`gF?$;V%~|NXsdd;*;3MH61+mi8;o0R7bn?dIu)o8;0DKT{BMK7nv=3L4@h5f7D{$huuz zw=dpa=Gjg9j(XS)LVASZxW`7BNHX2#VnaHor=0+Hb^KX@NLTr8d17ir*H?i?9!V_! zH||~PxslGwH0L$!p4Oynyts}tG51XyL(cOxnc2t_H}mfVM3r2Y>iQHF{u)Kx7Axkm zPSi|@{~qwVe+^W`u^v{#jth zd~x{uzAx5BCkg1%kk2xip9)RuqBQjA#c)=QQ}rKVc{?skrXcOJ5=4)?S5yId_&kc|q$M)1Kw{7B1DiC0fN;eSroWAdgc)=WV{PhtC7)#P$|bh@ zrMmmatc~Z@7cETxep$|^Y7;sPH#6dvaXhzmF8x_RSvbI^vs0f?m?=L~Qz8ZZv(V{# z2O!mLt;7lJ)qV}=qujPlyQCwM0$Mho{ENL0b3Z66AEQ#7ztlR4@sV@Gy2xd#4y;r^ z5k>&kPWnJca{%e(PEml?v1I50n&+wnHHGb4iFj@~UEzo>kE?O78Mab8Fv!n^oIwG3oVdjk0}*q06p*sd0fW zruT@riP4Y6Vx#)SLmW!@Hm(L+FU2O6j!T*wD&Uhuo_fEH?Ful@#OJfKp z%RMo@_Z?9T<{=;co=(wec^{q*8gfSiiUE0fM4*cI+DRTw-v_JIIWGqVj(*^6{GQ!>j(6S$dQS-6aIEcb*SiP(m=3O=2yVp#$=PS}f?-Ojtl^oxl- z&d*dNLFc_ww;dCbuin@W($+WaiCti?im!v2q2|8_9~I#JEt^BV1I-tsem2#bWwa=Y5;Hn!g1Qc2#Dd-=noOz0f$}5?S(70G;zSqO?_OFH-1?SHD6gIjs^qCt zb$!w*vXyd2u=Mzr7XC$6V~};5Sa`QJ*)wbLl@#L9?rO~i z)G1sQw|Bx-f&azcdq*|3uG_;3HdI7KEFeu3qzFof&=e8rNR=+V6QzZY2!e_NBE2KM zgx*_FRHTL?1OlN*2|Yl75CY_TIp^$q?m73{dmr~dcl^e<|2X1UVXdsS-ts*2na`Y~ zw(KQ~N8-%OQ=h^G{6A4#!FTYvyM#-vztYMx)YGrDVb!N=B9L%OH)sn%d$fEH#@53N zc_gpn{FEUd5l|yWPT++aDki_nc9$9%m09%k zYzI>_OaoDA_nA}UD{N5toVN_p#tCB{=FRaOof+~6qNvzB$Aq(z^%PDlDI8U|wUI&| z`=I;*Anz;jrZ2GQiXG43v*Eq3u`Oo!6W6%|44_Vv$B>)NIf%bwRTcCXrDPC}b&oVofZsl%cuscD=;7&=Bj7lE-gs6Ix6W&h@BD}M+& zVwIAgZZh)1ci9NxmjQd$W-qRIqcuBV9S{_2LZm#~Pus%AjR1ZBG)%wA$1~Hf5xwG} zeC)HQ$+k&GXSVKEs8P!6w#hFYW&5p_UAtxs`5-t7>%9HsI7Rdv-BLb4)L-Mm*k!VAt*yDehgc*f*ys#y+4j9Ld;C=;iKKx4?I3(&mB4Y|ACcdnHRrcn?;-Rlc0JfPt(cMCGSu|(Swx>Fsi`;MKh?0DBK~#By`oN))MX39p9o4J)hVft^uqti1X{4j`ba&OOgSzT~4c>T;B@C{J#_NIMvAP<@f_mbx*orf!*XS!uaBS#3c@QJOZ|XeLn0mdt9=RBV~c7nv6T_S%4z(yN@0b^ILoG zRClj3KN_>?3Q^kEpA8fE6ZM1Vx-yhVwBYTc$@Sa>9WmyPLn>h2s$-WsA>J!K z&0Th{&{Wk z!;MABXtY|YB)1o}kZX}6F1;m^&4u0inZhgl_mug*eZlL+t<&R9Q1#a2O#Hyo>YD$k zt$qNxi%-5?n!RI25GUSUQPVJPSRJD_v`07^icv?DV+^-}Bv!DRzCSU#)`y zQ$O-%FhfOd!$7}!t7zd2_Kvo(ki?!3s%U*fQ|IPs7G*>EYbL9g5YoWDqp=iF~9yZ^dMPQ5gPW26mzj+m`D+{ ze6zbn!nqFs{B!u^MtT)(n5Sx*arGv=)`>=X^|{Suqk>pSvRDJD*WB?*m?y9ZSZ~Wo zMKkH6*VyWkPc@8~T@K*ZW+^Ai$p!vI>`C>WzbYhsnJi@*NtAk)m*5Vp!Cy!bC%F*v zZu46dmGSJ~n-+OWjGSF8wzo<7!9UEjzDo>MZaqHU+^8cL?s)0|NZ);fm=0s?7KpW$TsrNZTqcjr+_%G~QLUI+euQ5m%B%LIoWRmlw`!6xpzr?<){pmc@r_$!X;^Q9^WeN^%#T zwzk|+zX{{@toDU47!TUJEqzg7a7MXs+NB)YZ*51+1hw&t-ir0qa9b3wOnO9O8pl2= zd*at?w~1Cwkzx~R+|IVCT<<$~bS8epV0gaEPUizP1lESl{@heGlUVir9iQumg>M3K zZwjW7LW{l>V=xh0mTIqi6Yg1}`ycvT8$8x$XIC1F2DqUL!`IHSXK<9E zFPLn&-{*`F2pM%TqQu!nUN-h(Y=pPS2Pz-kHrs;k8?)i~Ps^Kt#?B4+WB!X2;lPKX zVPm17s&AXS7~w?O3=px%MsL0&U1bUCo4`#>_s??ClV$`KW7{fyks5r4+O=#eYR?ayWjoM577J`_m8x(HHrK4Cc# zQ+Kq``0CQO8Ke>PL1y>aX#>la*Si2v@tv>Q@!d{OX0oa8QgL&kPJz(%LG1>4R0+;$ zzPG6ZI*eo}J5*Nn0xX5uDK@{9b$*ZGvWgSS~z7uqMtmd3(l8}^r7iA5R z+mrmPCsl>hAFg}jSuq5RcGelT+Nec@8jVjTFc-zfn_M5m^_$G{8BX)k+OUjb zwF@7gW7u}v5;B_(*J$6p_klyr3BnmmJbHg5gJzshdSBw3t8?9|uNSrmvp$ib)+W2W z68-JnLWgO8yfP)4@|gduW|vWc*5Vxw*F@VQM^|pYYT~?siWy?;{i@ziI%f#Hn4HRtA@S7#q19TkLp*Nd`UCbw4%&5E3%Gt#ZWF8oLg?$K9|h1=F5amH|tFDbp8?av(h%#|I1?L}=p;1`(%KD|1vFlK3|aBZVx`Erj~lp>bCX;5)uY{E4Fs6iQxRP}xIbiw_6 zbjIEd&&s|umZ8gckYZH26q)Yt60?+YR5d!aC}KmgXJ?MJ)I8S_Ubf}>L~}>GNk06F zIa0S^-oC%bmjtxWMpWorsZ-RJ|3SpTcOJ zaZs*JfW!|U+jT?iY#gG_JZ>kXoI|;MHVyRvcZxct|Evs)6$MJj=?P`Id?!sDS{Pcp zAs7Pm^A!6T5|3b>M03&I_&_|Y-|`t9Q~dbGfSr)ATJ=pf(16LGoG9Ez-w$~#Jq~P6 zXsIv~ScTyR2P0jgw$JL-5oV`LxOx z?7z4&3o?nBZ?n}SYavWED4*r}vz@3XxDF#o+ak1SnYUol(l<*baZu{kSK`qhdMtsS zKKkQQjTC-R*O=GEgf-fnIF3NWIhl(CQe3(`z^th#-HUo0Co=G<(s8t6xM@+EUNPMM z6v{Wf^-)VbV*1SYO167LHlYk;=`*KN2jCak7R#xAZLHt5IpSHwqe3s zH5IVJ2)8NXYlLl-PwZZuTVgAoCyvh#pXtu^1zOa}8;kg`(ZN|Mf{P#_{Ak~5T})*E z#?Qf=cOHlbqTJ)p4D2}#vZDmrA%7?d(@e*rj?u~-T=i6@oD1r1fw@SG&+0G)xpDca zJ+2wjVM3ktxF-TvO&Q+Z(=YIZ+$w5)-*TiwdU-txeqY*a%cq*F#wy4zikA}_$3oE4 zyvdUv)3(d#+T0)B#(_|Y8o$J|F807kB;3>4E%CwT5_e#m!WZ5C+wNYPM&8>&c~$x~ zFlpN^+8Stqi2+*3@x0*GtEt%Zfay=kDN!^{%#N}|Rbf*?fUNUE=ap3TN*xovD2~`& zQS+juhHaQ1*06$}yiC0?gkSYXHM%3r|WiNPkm1SN3`t@gN9Zq*- z%$DmXHYQd1kt9^x+eo1}P-hB(`}XbQK)b9^*nLf(pwbt-ww)_ZxYj-x0Z&@5v7ld@}fnnokpx$ zYC-0ItdO3mLxp3(&T*ttU^3@c-)F5xzdUHh5{%M<|XdF?(`f$R34$VmSHvr5rb&L#N-{ z9WZ;h(6&9&Q7{O8<2iM~>i-?|T27woSiCC5|IOgB6m|LV?}>8LjnO z4UMO4)VPo?wfTlyosltUO69e1> zyhY?GNxa+7hGAMq4~$W>Ku?;2QmV{;XP$W%GFnbj$?)>qmBtL#b>;_trjRH1(XZmu z)VMs>=XssCSKrH&`)1qD?d}o09&+{DzWd{PbVS9GFkgQi)$oS163X7XifyCpZbN6( zf$`X&b(+l~-&a(H4|2|lhlic-IC5UpWh==+M6Ovi;jF6y{)-E>U2Asoo%rk2tT;*6 z#~L}RH;7&1s`$8)3fjkEUBc$&VLGW}miez!Z>!x$rV7+kOSeu#i`G1u=nDsBh}yFV zidyG~2`qX*pBd4?lcU`$^ez$k_~LEsM}H1AMHK-CyT(>SeEF+nwcf&OyBg$bp<%bQ%n7S?cR$uwW zcJEZ3LUqb*QwC316v2JvH%RfMvnXf-#+%@p+{u;Nfmn>~+F=}To{T{K5 zeR{7_&_P|K-8Tf^u(23-Y$Vm^=egdKGhh2&Ik4J$RWq|KaRc%JE=IR9idvs`s-*TxFX z54*&%@zi+Lt7wHUIy~BELGG5K9M7`KrEYgIVb{jSj8(LA&8Qdx`cwDIvi- zq4D=T7B284!LeyhM^$+5BQDw;-FY=C))G0;?-){UYb9Hdu(&yxKlf6FZbKbQXE&Z8 zS&NpSZgiM@b@^qtTO!qRR%<>g=YGI06}>pkOzbH3(V9f9$=xvK!)#}Be~R7KjcEo6 z7q5U-chqRDa~omYkKy(%%<}bha(UKI&-JEY;NcLXU4Z&-+~4uI;92Y~O|D1MvHEaN zd<@(fy)njPQuNlwx@4&0X-e!?CZ^mv)t6uXKq%*_DL(Vf+31hqj1b-fUJ5XGRH$>E zcaU4UHhU>fCSYSIX(YNvwPppl0L6KleXnwaSmewP6e4auh%@15yLR%yg2e<+syc4W zTOPM}P41*Zz%J2jH?&ELXxTBiytS*EBx+jq-sGg*h1-l=q@r)3`BUJ`XsyeNh6=Riu9GbW&3Hh5`eXU7e;j2oJ(9$QrI$j@ zHh2?^8?~rt#i3mES~-886cJYsuGL!b?GD=nhkBmI4b3~ksTcT+N1l*_GS!|}5%wQy19(+vL* z(K~!(do1?Q_a%xye*_b6&WdjB>4{0MH%Ou?U|CU}jv6NCuw4q*4&dzTDa|uGjT~}6 z-@c*v!0a~wXa5lKQhx@y|#xY0#(!4)>J$e5EO~bk#r^FA(2bHUzeIsq_lR1X8qDA zi+nGGcIK4(FJa(UiO=WT<^Dtz>i=hy{A0JS0b^xCxbyGslfV15-#c&{Fl%b5|H2w~ zaP@mY@0O;mMfopHl0#r;9;N)3XZ_E2^=F@0L0A+*Jf9FF{Mo(z&Lfd|2p$-#XWZ+*_V+TYN#4IV2>w2je)lL#CKzH75}kj{ z6#m_luy}M!{t3C^_ipK|1}KF+WU{*UTtz|HRAx7x!h=DD5;C;=i4 ztOHZ6;T^2y7#_W{hro&^WAF_Ayfrc}X_G z^A;my)ElZ!-ly_1vCYCG5xW{2coFLrL%{ac$TNoPm$^=-{nE+=mVtojYfp!dA=Q#Z zS^KTKqk!$$cffv|`H`o&m2jXM`sM=XHJU1Wj(7Beuh!glJb&Dbx95^9Hmpu%9}wGq_Y^&9{i4B`6IpAfh@_A z$cV%~Qkwr+-}QK5-?|Obsx0uuy5}l{n+sJZx1WPVw))wWgL;EvmSSI>c7YA|0dK{r zapq#4c6l}sA!~KQe}B)m9+LNBzmlrXzJ^pOG8O?1rXl3XFfB6%J^K{Jsi_FuKBn`| zFfNsDEsvX2IgPM~UgWD{Hyx!*yvOqCs57=bdV*vT^2_run(JU}B?>um(MYa7KNuHS zv{hTAT%8oT0q(U=ND$&5Jkn!+Uc!GTFUO}UhfQC>hyZhIi|kvO46!8d@09`n0YIzu z%Cw}l|GZp{VQ*>L)c^yJ9ayrQxFUgix>E<*7eO{3$p_>*jSJ)M$4&UTCh8Ozn{Jib z6Py-JSp5C+)SG-V?aw)Qk8SqHDQ*(GcC@*4FT(Yi3my3b$dcK0-rwF{!PwysmVw%A z+2^W!fCse9Ghsvqyf%K9Er;P+mz-%KtX`g(9H&N23G5tfdug~_xTKg?F_y~`BD71w zAP`5WAI^ zZi}dnGv+=!8@6p9Ak{Y5cQSMYUGiSF4aXTt{qxB#7^B~TpIhcl&o>o z)PYFF>R97#A3x85gOXAaX}T-jJE zzZ*DQ_Zb*hXX09!JD>Ge%2eLEAF8~aSI?7v@gyMRjMQkDzK}?UkeOA{XQ_{jHd|SI~5t@ z;w^Jd2KAr(7r;sDJCFE9ehnTI5D#0+%^%9wQNqiJ17&^sv{_ddhO0QZSaTu4sTB4LuwrZjk_fvuo$&o=3 z#Wk(pi?Id1w(n7OZY~cfn+gUN*_(?95y!q){%owGF!90Uya~d+Q7C- zhs-$d{sIxLTqAx}C2)W_VWAx@_{tE>(x(o>4kO#xRDTQ5Bkt}^RKAQJsrWmG32WUR z_sCd4JG=5)1?`45H}zZGaJl@Lkm+!7)Aqds-Y+J|rgDLbmIG}|uE(p|4wPlVONu6E z2mHymS`hpCR*CA>#}GoN{d5zBd-Y8M+komrBvkp8p6sOlONiRas0PAVGIcZgJ#Z#! z*B$p^WA@*84;cT$s$-U)TT1UJs?-@~&nMTbuwb$2~0i}3T0 zO45uJi@(Cg{Q;q9x4bh!(qMLfE#!xQ_H5;RQ^mXlT&rlgJYCi||KrH(N4pt>Je@>T z*db7@TE9=)+M+rReh_kgsn*4jBN#qX)fUR0G*+#~OfV*lZ{)NDE3lTo)9rHZ*1WKx z?XZy`oMl?JDhl)~s!LQ{Xz3IaPZh!m zU;Y!bg@1-JzkP~~Y^Kv&+eONSI^5La3!*(8HZM2aIN-cxL`jgM-x=E7<{| zeBIc?zN9y_KiK`oc#J#qzEx+yYC~l;}nk4MW|2S4fF~A zT|rsNJ+KzWZP~DKtLSD5Z}vOmtT~0b3j^&uQ!ixk;2rr<;?_3JjaH$<#B|XWo-N ziA&G!zhC11O@cA4ny4%aZbr4(l0&nr0g!&&D`mn2th>ta8)HyRel&wP>b2+k^m~-b zS5Kk3<=I4Eo@K?|z~4==e&@jTy8+xqKWlYvu57N<3-gSiwPuR6%dHA|JGzbPr@>mI2?gNNyM)ZJLTxEbR&RI3fnfkL%{^MRq z2P3cLy(((uT*Cr)G7x!(7#(L3h7m_c~5N)3))tC%4T?E{QXvOqIh*>|l5UI&UL)e~bzy{D(Yzi(+e z1H|?Vlv(t-gOA&}b#4j$SQ<}aDO06!H;LxQ7O6mZuR`3F`K`MhS~pW`jE}=)0p+v1 z;H7_uC0E0u&|v!t@bg=Zho|pfnccA~@az)m10AKQ#O_zYc?G%!RuFE{{v_D4O%HH0 zB{^pk5sS}kBfMj4MOqjoP+{kQ^(#={#1_?O0KI9O4xB)Thq3# zlq!qeB2mIEe~w!#^wp~fDO@9#Hez5E7nTfJwgby5ot;~rAz*^0K<}99LrMKf^}^Iq zAty231rJ)S_73z-jtv1l&TE>LQtGP`)p-lc)Q7wY8PRQ(wgZb#ZUNOp=bld2e81!R zss*o)+l}qnX>Z7MHjw&B0*_tt-`PJwj|jf<>p6n2)GpF$gK+mf z1-{jk$B=Bd=Hk-^dI*VU&UeJ}#exvc5^M8aI!Q%3M74imDLp~Q3Uz4TPxG)p*usQQ z;}8uxpJ@iK%J0#qVP}>*)!~YC2mY>CMGkN9@-^M0e}O{krgb0sRE2c?<;|3n%J9)A z{6V||q_mW8Zd1}W51w^#FlcJa!SF0~m14?LpZhcHD} z5k}cmx6xuu)Scq+aS#GG-wjOaeTd6JCYAPA*fr=pM#c-|z^@j!px?y5`FY2Y{z2Hv zj=oC0&HHOrWh5!ui)VhEzc!S}FLPr}ruF;>+h>hu_DSs)`N-q;dAvIIWF$0zGUcpC zi7G`mmls)nK-gi3MZl`7Y+$Q$9T0%(9W7UL2SYo2sNX|WxJN_m#z@+-+X3BTnNwfG zPP;SDN^MmMXaC&o94@z<$=BYZI=mBgXgMlc92gX$Ap3Pc-6i?i#@r(qT@Jbu!usVoh%G#dwCj<{5bM~^OQ9@|$|$5g1#$rx zY?PjrkKMKUx+xws^_F~x@nljAw=S>s?Fnzt;YXp0dhNyB2!Pd`^LzEx~%wF$9;)3ivB?$%P#(ljg=XvbJ`U{FUlwCE*m0e|4lXKH@6g*&yDMlH1sC1xC`KqIjzDAYy*x?enW8?6 z_)%$JQBZNIsfH0wUtDy6OabD;jsrxT?T(Q@_N%YW=3GbK!Gme}?0tHq3L5Kd*>Mb0 zSY1{FV{N1ncU`(U(zS!O0}OKNMN}+sy-)(Q1yBI4^?I95 zu>pZCJHr^YA7YGdAEN}#<``eXFbW70S>yo|A7IiEZ!R8`+kepV(m&g^o<&4Q2|J=M zJH*TVGTIe%sb(n<`?r9Q*`McN+gt(8^P6@9FSzXGesZLGezW#NNcmnnTIYLll|k&; z&!F)O?@dj7ST%FhU^(_@3B4iA?yXbyGO~@~(pzz^D}8bYfRHS(tRGt#!&t*Sow4q% z$i()UtfD3~WBAH*Cs|o6XdZjB!X?!q2ix-}@97Fdhs{u3Q+Ybe<)s$l=ff7yqBIG? z6(VxaZUwsW_o!Y{o!c#yX9fhaMv)2AWN|mHiH#c7jj>B!(q70m|B*4$z#MtC2mEV2AT15ddQ;ryZb#7Ngs1pbJGL#9YLI@>e12S0Vuy)~aT%mrTZ&TPYO7 z#_6b^Aa;L-f+)K&LOyG9sVkeJl;GAKRGKj&s9gWF_gein(D|q%dPTRL5-#>bK=p? zNq8-i!fHA~4}pTovj3$vKZzm~>(xUj zAzL$%a?m3}-0k3_baMXu<+am{C;4cS>gkkf57Rxf8~5I^KiP_llqt*s0DP zBy5@MxXR2nprtE7X`dh?SOd#R9kY@LZ{qAhAIx-nbcd5-%P@d~D(M#7UX0PQAxPa) zaxB-bU7J|R*DB%x?%`GVI>crS4{?!?;F8O_&KFD*E5+6}Z*Q0IHI)?DgfV|hpmR(Am5wj!NvVzs zN*7&W>p1tAFkj(Lz;JY_0R5?p$(ur%q;K}cuaC|1wO$14t!^xE_EMzjC#Ho`K^G+f zGVZv@{UN?6`Ao%3SE8=5(5Lyu zNxs6RlgmPHh&n?x=9cUjbxZR>zBZgIG?wQ*$Ysj;l2N!8nq^n(+9x<5T@*>9?6%w5 z`Qc1O`4<)!#G(>}J84zRwAe5&vcXDRYtEN$WpKdtq33KZ*1hiA@=|NmDdGTQSG=2e&90NXlo!H#O*JL4`6l2T8`6PeggC^==`Y_fshC#TEEw@g{;3P{kWcx z;(Z-Kl-LkbqZfTT@f@B|9F5;Be^y?LW%6ITbLIBaCnW>R&X{_)5Bq_~+74`L6tVg$ zCY1BNwg>ra_|C@r^i)4P$lW>XL8{~&5!UTQWo(Fi*Ioj-Lb(`c&b{Y%Wp>nu^c^~2ZLpDQXEQyj!l z%j}O4Fr2YLrG!b7PbBJyC*{ip%H;T9AGbZDdYFQR4ak^L zZ`wyMbHo1@Ybg~(K}V{(f{{Wz@bu+4LAYpBx5O0|Apfq>>1zCHen;QYzq&SDaHGg{ zYi=xX%>2ih3+PAFCG*!m5=>b&9gSMXfjE!zZg(XQ~5{O zej>$%=HF(^fBN4bU=Pg{84~~RH}fC*e1Xt6E zH`7RFlSHffzc^B;ZUUO!kuO%df9daKoDXcwuNKp{zWqyc^LiQ}EB(KF^@q_{X%ftT zv4j#3N5!LGZ_L{uJVaLz%Vjz~QtfcDoVrx-U@>o|<9s^cpKbKsx?ABN4G=*`tCrA~ z;YENWtFV1jOz=M%o)6DZ&18A|Y=0KeD71N>tE+rX$!CA&1aS#}E)oJ$^{kqj z>MOa6x$bOn_1K?pZZe`r8q!^;6RV-aKZ3fI9-u2=wXDhFM-18f6sZI!$I1I2W*9`2 z+FlTEnAf>dRiHA<6E&^7zvtf?$#i~g7RfL7Ce!7KpM0cYSU;}hirk@q%YY;q&U!iY zAJ6{mfs~mpF zeoLd$V$G(QIr(S~ko@FOo)mKgDz>roS!~Dg8Rf|`9Xy~Le(S2lRU4fi+kwtXE+BQ0 zjCh7(@?6C)09z-n`*oYiQ;ZOGjRbjK2#ArZ}{}8e#1Z&bS8#Mr<^Ay(l0Zo`rq{K&%Px! zGUVWOnIqaAR6SB%Buu|GI);krfPVOXnyaEfY?$%G)!}mQOXh#9ll`|r%5vqThwOAO zRq`Td6?9nnA{_H{ty0QC8|m$f4`f&>GAT8D`vQAPr?@iRMN$f$D{0leTXGyp+S@H~ z7>CKgSgE;nPuQ>Sj&KTC&Xz^HkM!>4px6lYpYWyrJQbky2E$FV@w>G-+iH^%=A5_+ zJKn`f$XfB4SXS$aLY-14JT=6-OHy2~IV^=o`*(y$hrM=2>KeI4Yu(Xat>4AzaBAm1 ziRvf^3ElaJbx5NVs4HcRzs+Nolp`w$oKu^v7sOsEQ5f#(u{-I3*4!2usd*%sr$M}j zGvBk!#Q4I_iPK0682mY}^SmL8aJmaQ(6GY2k(_z|O0i*CoXtz?oB>XgZ&w9Ow&{G>g9v-YaB7Q32{=)uzkTf#LKdk<_$TOsSjZO+$T zJPMB3Kz~M@v3z4n#p+2gwHs0#ek)N$rpxTrRc86r_B~?qd=tg#8^h_~^_Dk5T4lD| z;AS*;Ol+ZE`F5vTI~!*wU#7XdKR4j0MT2HN#mUd-6o^m7gnxS)|Gwz_69EmdIq?Ok zQ4RDa=VN<^j2ieN?-+>_OAU16D{S(#Zt6VdN<4{dtDGO7Z#=yG1pA)b-U_DkuJ6Xo zVl1TXv(RGZvuMRg5pQ9bu?GD^-qtJ8!*}q4L^krS`6Yx@4I%h0`X$LUiHNX zNbeS-7?%NI#U=GxT0f9~vrbW0zsE(#m%!2*^#kU-zF|ucXWuke32T)HO;5%`VORd~ zTR(F>)hu6}*{>pp4hU;zDc+!77zi7vQ!ve17~tY~*)nnYw-xoDH}^k2{k(Y=)l&;2 z(3`1n-3q5N829#V$#fPil?rps)9jE}lR7`O*%5coc}H>K7I=diunp-Z>`IMZ`;HNr zHb=Fi+2mW>_q_6@pX7uEEHr%kQIqeoCwbtCJj)VT(^^#9kNqR4olra{>a+dIVWg== z60*}?Vl+DY+@G6!G|++bOd$shfrBd*ny?v}>miVzpq zks^$@#k=ACfWVMU*Sts#uojW6ajFvzTx8KO^3up-tJ|l|t$BuD-z&JsDW`aW$Z;d< zW00bSE$zLV(YZX2qN1|CH(Ba_Z=wk3Ewt*?p(;ioK>dp0XqiTD?31Yu}+}nEA3sd*B$& z`KTe+;``H?sjelzrm^s^U%n&AMX^TpgojNf4H`v;SK8<=Shz_8!N5WTLaK|+1N@Qt zFkJEcrh_0eYnF>DwCINa{%yb;jIii%pHX<~-&#)dKVH?pzs-m96fNv()x39Qr%YTu z+VqXB{6g~KZ(4DI@OPXO;Q&x=(}*3fzjUU9jp!UKDqzxNTAoz5>%C0j>8unmT3hIA zaE;K07;%54BgJbL#;MI>SZ39vC4?3eWje*+z7j*3-B*DzL|Q`YFXHquUKwK57_Nlz zWb1B^85mx{caPr2*sjTH>*}kRo{1%mZ1ux!4CftE$FTJXPp`3-GPzFNdbvf_H~pq( z5PNT*Omv7wiE+t)tm<|i==dA>SIE-e2>B|J=tT&IRznxJwXd$8l9kZft?49N_>)Z|Gx7_f`pUY^S23~LX zuz6#qv$*&B2Jb&j-)#^=l5{@G__ZZVCfdB*gtHf~fSpHO--Q11E{tmtIdMH9-b(OYzN0w$V8hM-$ zeM2d>!vMG8dMe94XP5XHa2h4Ebh<{8%o}bwjFqWKZRaB9h?@*v^|q^|O%xg^4#SsJ zq{bz&*#X-xU2fB1?_9m>vii|Jt3Xl_~21Py)Kh3~Zm$NKG zts>izG_UO*&lY!gy-}HY{YrF*()K!z1ukPlphNckNvb(B@Hk8Y`y88 z@-;7Q3+moHPLIlFddDCX5o_$D3UrX3+x%&q-~5g%ZJ@idHn`i(cWBzStTU)Q)jv1P za{zE+5`%vf2ca;hVu=!S^f*z6HwlgF!?E3Z;&CO8Nsj5$q zrNv;OE-Ns^m6^8!JIkmWpJxc;3T}J!ar;&ar-|-6%G?c0Nj*G*xsQy*MZ7!RGz%^L z@f!^xvLlF_rmU>%vzNxd2OD{he&e!ZIzKm77Ek8^8<{NQcSb8;x#Jjo74odlvfwdc zYdMEG_D=T6Z6yAp9_h34;3SGrU$=>`>NEY0>fWm)K+%lzKFb6AUpFKQ2JE-@GVVyX zF)Qh|$Gc#cS_)y*NR)7@%=mm;>-o**z>zd2?u7E1MrBS8kD_{^G^6yPEcGhP-gcg5 z>};)rWubki*}c(ljAwm?Eg(}9Zo1BSj(bUwqO|yFq+EA*8T6{nV7E;Y8z+Sm-a{i2s{TH%o);%I{M}Ip520AyhsvK>5%Tk=iXXB4%`07lq>A14fE`i z5E`i@9bZQ;LWvSz_xB#H578pO>$V_VP9-lM-jQ93%JXk;p-yan7AsR%{7KMaj1<_k zu3QkyqlE0KWy)a!RiW5pR?6@o)jw~ia=L#Q_{j!5tM9AzO{Tw($|jl2i3|nx_>xz& zWEaqanJMSFGgCzn@+*Q47U1U@_dlZ5SP}P$QZ}i(r(0YHS}sJSm{*RtyTrdi1{!-B zJ0G#EUAR-E+xo@OABp?c7GdHSPA874Rnsqd?BxRyZsV2Z){q`2ls^U`V}MS+|H_BE zW!NPo4il zI6cU@<+&kIteU3@3U=X&8iahRicII2s+-=v|FNbim*PTzp$kZ&OUB-8~juN2z~ z?P5u14o+{rF&U0h_An_PHUKtdIP$iVRq+nCZqy!%aw&&q{3Lem(=~7YJz06)>_*%3 z)V;MEGaW^=8rAJT_yZiFUD zxM_5!B3OWlh|nh@!3gGUw4XU{&F?xS;zX+8A}v>IfB4?$ZTYobNcA@PIWb^kKE}~v ze~B})yH);vE%ld^zb&@^MSl2qWGmvuk&2JC!qjiQh3ngEftFfoa0g=UY<&VTOTA7q zjNCq#f_=!lSgx`;b<2B8u_=jFWBAC*CheC?@DY~YSpV1}{O(4M$xP8n%Ns4)yDN`d zkouf@l-1}8@r~wK1A{38hR9)Ns>c0zZXJV3XH-X?j`+Ryi@Z0Fhh6 z=xy3{tXq=yxvzNXbZp0_jt-~*I|j-ww?akpXVwX+$6llr&zCg}$XXI`Y&CbL$WprB2c;6c#p z(b<3?Z?5QA!92~-ZkN=Ji|6EnRMz7W_6w9`RiwX9Mfa9Xd#Y_TWDhpPSlRPDPDR`G(E-dcei?8?) zg-VTcRs=b{2B!?||F6CGj%xzVzK6$(sDO%yA_9&SDZ)sVZbOl(^d6B;B-GGC8AL&m zCcOwq5u}C|LV$pXfb=Fr2vs^EfCLC3l=sPZXJ+@eyR)wIdH-DhiXnvLsrR0H?m1Sb z)P*rL)1AG`v8tzIyT=7d%JxWQ64~Wl?&BhLMc?atwVd+KH?@)ph+=ORen@$-(6$G* zV6bXrA=AHHGA-y0E=^LM&F1_udR71eImw3qk(zwv`NeVcW>4!d?dD;HY^I$S*17MV z7-7w}O2Mr*Y`*9;U=!lJ>JsYaSy))6&MLln!FMhaU%{HTmnKK8%i$4nx&8Hm}8eWadZ) z^(3GWs6NvHnOf_4CJP-A9vLGFHFh-1r6~B`nYm)XeL<%75-=QfCCp+n6bnMnN)I`Z z-tTg%G~8VqR2UA9J_6;*sZ)_5olP)2iA+^<@njZrv%P^GZozQXeTJRlIw)X6m1k)) zyy(Rs2?g4emwFvOp4GPU$wevPOxm93ie>c&l8p)X)|=Pf3Ql)T>(9LEJ1r#isPvNi zc`D9?Vlk+K(A`Q}p{7hbIuD#5)4ip@A$y~#qbjB+VT0@f4OeO?3Y~D1JhtVLQA=M` z2i^6vdvAvKiXQVMC--DD+xHz^7~7L>J8RNn&5gp{ix#2pWaH+P!`;k_6N07Sb|>rG zl-S@LMfD_B4vj>nzAw1g;ZcztLej8xqtEX%6hOHAW@4VB0@^mL7~?+DlrEN#b?LX^ygyB z_~X4oNN-uwqSX>bTdILA?PJZ5Lkx(!F_(Fno5I}kf^1q|8xEDj?Bq`lVQ24l+DV%D zC5;X~JkwNlkN`)l+v{3Qtz7Ih09EA}Ds=3M;?6Hx3v%6`j?PtcELX8(JkmZUaM~_r z$9A5V--^WAW1*(Y1wbcqWmYXh>v_5r0$4n0v{t0mabsgMPAzMYT`s z1PS{S!!V;8!ysG7yC&L?A{D>AtNG&pZsf{?uvUrpY^>}$WM$cc(Y!fZ!7XC)4xAEXQu^Y40JuQ{? z*isU`Ma|c7 zJC=fP7SH#M4^5MYl*?3AjGfuJLS(C&6@Q`^p_2alveSgkdl{DytE7(V zbAM}!AJDiz3>T7zM2z|tM5y-?jT}t!D{Shv10#e*Ov*p5pYWQVp2dd0o!AI&@7plC zIHF(Hj3RS}DscOEBngs-iMtJGJuS{|5xWDM8`56?*?5I^ZtcJL=EgVk=p17}PLZiO zjpVmox2|5m_`bPdq&W}TR#LxGKHP37NiB4rUWL_@E^D9d)b9hnXCig6{l*pGoRWe) z)YbBk(Z~@}Oh<>jw7)SeM$D_PO zLpA+F73Mk1A=@ZH{mAxnqSu;-jv0zt#~?beWd~*u9k}OuVK}-oMbF~(>XCLKOuhU0 zK=BW-xPFIqV4?w!XDq@#m?#Wy6T6n5TfGykI0q_HzO=D%D{gLc5mnv#dG^$Vv zb?MauizWm2JgAS1O7 zErMnRNPaxN!=FykWAA?SZ1_^c*qw!1qRNet;5*T4etMZ+^>eAsf{ ze8rV?K_mL=>$bvm!8RSQTbBL&Wtm-e=z#s(`O=4}!!SaT-~`5%=iZ9kHK-oWJY`z0 zUi!6_X{FOQl}@e!IrYk6{bOAzq@+V#@|;GOzg{|!;Yu2wWy>Nr>7o ziKHPNPBeb^RxnYCurpc0d>i77dUvUov$dPLoNxDK45gHe6l6$+91|&AOyj@hjq5A% zNdZy<5DTZBBl~d?4=S5-UXoW@eEq)Faxwnxa^~8%KYix>IQz_bGyf~v*-!VJY$B57 zlb2mBLno8J0tdjM27~2^8H?;~)XMfQnJPPgQ|MBeEZPzGS;|-P>XZ-1G{POcQ^^Q; zY?Hc?jPAfXA_+t4XjV;(uPB(=yI)yoq?@acu8@29WRY#_H{qy14PwC$_;uflx$V|a zxb3PjdZM_H9$lZ)EM?T{paG&o4TAN#Xe=;hOAw-+kz~eiUy}HJy%Sf=;yu}&iPap6 z@F60%(LTI>+l);b(jv;9+S0c~I=&h`v7)!GkUQi2In&+~#(C(#x=%XCjjc@KG#Ab^ z3P_DRjymI;!4Wr>YnI9j7s!QR^Rm>Vn?@b5ocG2t;T(z+lHr7qF=SZ#5=821j)r0I z;SX)_P9ey7pj)6|B4Sd>)t9YFF2PY}jPvzkO)y1r{ol-xW~G2K>XqSa;ZFjdjEdpn$NO4eOKBZTUXG;-@r!Y)cr^l^wO17 zSd<)}hnL=6`BYKHtDNT3$U}4pWYE%b0u&|Es zfp;P!ZlZA7V&=7fdx8~~2C(>isZM6O8PWfW;qO`QpFFyL^u$?7idLNZt;tjnquIjr z<^9vgPI&F6%gn%e5@3+jpk6!YW44WpVzZquO(nb+A+FOEqC9V{2Ya14Uw`oS40@5t z$Qiqhaq-569N>M*`EJIr!*?wGRtCAn7*BXW#2cqQg&pfmY9w~c(`AkfnCiRV#ju`j zKRM-B0EPIsMU0uMf;_^b%2M>%KDxc_WE3!sn$X^h!Er#x7je%XM$^L`I_$^zf$*`@ z$P`>U)9cy?eNuth(ok2KmNz5m-p0`WiD?;;IAqv$(2&vm@#Ck9q=;LR+ULX(8wAeR zGKh_WwcV8l&s;LGuNip$e5rh}yGha{Y=T}}vsC*boo(IyVYnAxd%S?5R44rS&Y~E4 zoU)nb7X`Al!XDsKO_Lc8Y5wH)w;JA00R;XNi{D}vL3;0}S*N)0>(~y?)keAV1iDY@ z&XTJ!*<`TVqi&WLNkHora+let#qs!yM2~eaX#px-?4IP~7}6}#XE=_-^ZoW4vJZYk z)NfHrYiB^ftoaepJp1RQ@1HhOq{*T3A&jJBx+|j7Xx)c~Mm=t@Ci(GaEuB+9qqV;c z>%H7b?ej_6h}4I~e44ZMD%rb_d8puPtQb&D{d85d0(!S+M@p-0`rG`R$hlv$kyHzTTW-w+8o_hzi_*yRC6aVz|1@m;6M#@w`FlU@M;M=JuOT}+R z$o;&&fQ#qAB2+E2mPLlSlb+zrdy{jnl|p|~2#+T;vz!(*(os1r0t@p1IrkxMx25$* zCX%BF@<^1hsSLfSX8h+o!#*pfm%j~cVu38o#% zWFExwX*~+(BunQ2*>URdEasAW(q|W{16rkP$<%UdS+QkHLUk(m0i-RHfg`5e4k>4m zn`TY=#uq5ZWb^&fF)$+p-a(AOI|yJAd9z0JrZiKmpQ@mU^s$G~etiiyT&)zzW$@=` z+>oLZf}e*=%^ZTiMp8Z#Y1rSv@RSSuJ~!%MTFG311N>Tkj3A)_u$=HsihP~#Hp;sW zWl4W8ZAsDEh^X*;Q|92Pak)`4zOuUl8)99a_31>^ zc#rJ~wrpC}?VNb4*0_2<;z?jC&3t_#lpSm$e5zFYp35r>$^)~VAk&$w6YV0=*P zC!UqnTba*<^q1O`^rYc4p=|?SUb^b<33{8Rs^`~s?b)Q1Nz;(&IG!?8DtnFF&!bCk z#mvuqz2Y?O_^5_Sed~Z9WnCYXPF_kBAAWZ8Nuc_()lc6|#vM?uuUgO3!aQ*! z?gqHd4WS4rR%d=stp2nV5)i}nrry;7G_OWFnW*Q#Mr`zTV=mk#7d+v>dEUeBEmBC% zhZBJAyI`Wo9^H`>3*c^wtsdAdA`hsCy;-MFvL+^7F6)X zjej!EAU5WA@{JZ;mt)TOv6-ZcmG zUFZmJKKgN)9xmx?P@4K{ArW~aNUupPvdW_O3D3aJfE9MkQPPe(Y-A{2|O9ZebycT z#h9cidInT&WCXkP&&Pnn&2n?-D+hc<;>vxGb2~(Ilr-rr-_KD&6GLH9CbwocSfDH( zzf%sl{R0;I^Q)eryAnB@6n8D(tnFo?*RN_6>Hp%?&6+Vmn(AE(_&~QX5O-Bk%_&=a zyi6p^2V-1dKfQ6s_e1$CHBH%sr7K5mNBfSC)SX?48!ilGiuhIj`JTd;@}VTN$qoC? zQen$ak)sYNEZ;5cj^8=$#E@DY+(|gN?ZR5PqiJ6^+P^HfB;@~c4T^7dOe2(NAE?3_ z%U@AP`2NhNky4={y*WGfHZQlf(0um%TkJq(fE;NeM?-fGqF#xp$c6c;raw7lwVG_a zvB<)#7P&GkZKeOY{#R3Xgjm-zQ2z_$T`X{D-o0$6xG9LHoc<;46Gxob7w?Df9h!BE zPIqzg9ur|j$$D*t%5_uZd+euq8c1#V5%Q12w2($tgO5X|e8u2FsM;pEg!{7{Fs;Ue zn?Uo2M|BU+&DL9C+=p0>3F1AN8lzq(3Gtq~F)#i&OQPTm<@_%b`sY^d&(Df8*y}{A zS6@=fe-G`uzsv3lSsx2Np6y+u5Lyb@7digc(Yf!mQ5j`Rm=kAC39&W#ZxYS4s_s|j zhKiKJnl+?>?sK06r1ONG9#zA}ly(2aTXGKywd3|K_b15h-mH5>>#3pO>3mjYy?TVpuPl(LOhp#ZfEsogQnFfu5pG*>}_8H7fbQ(Hxb>^ ztkV*+eUb5XXD(cDcw&0KP#C!CM6G47Vn!;GBKLp=1vl_da#2kb9hh;QxpZ!z5aM`& zIZg_BJLd95ntw3UH>%%U_eE&9jJ1J5!W=u%LdtPe=)ufTS>g$w#NMTT6H_{Q_r(PG z(iLO0ili745NhrX8pB?&RJh6qS^aIT!A=L1&mf*;Yl=suH`F={&jZk~VEMe!&axf9 zX9AK8$fJ2opf9-YpDr69l&5X>x<$XA3|TbKl>W1g-6 zrm*aHT;R{c9Bz+&)mmORXXekzFu9>fPfc#UiD*d^%yEMa8I0~&17n>lr^p-nV=ZJy z50JD9`-*zJnN9dYfEa;Mc=lWi~B8Lp*#M z^QsE#9lOrhE4O%2HovN|(zMtq*35P>^qEc;CaB#j zof5Qep{8NcJ003sfu$-zQm+`twTq5V4`eLY8bu?PA_wz#OY*SD;c^K*DXii`S`SaK zy+_rknM7-xE-n%Xd<6d3lF|6* z5l0RN@f!|;P;Ux7&!?TE9u3-~+r|S9m{uFKu^BD*RHNrm-jX|2bHgQ>^!V7-;YOLi z1;J9UuT%B$tywMFR_HdS`*qtgjoTjXA$Zh6GpD>s9jt%|ODQ6$y!!C0AaBaieIAzT zEfIlD%4R~Tr${z<5BBxYP1Le)<$8@57h?GbXgK3rfRmR#nzo0g`PF?>p>_`3@?d*3 zhC)Q#R=VIME1XIBQn-PvovR4@{t?ZEIi~aCvb8R6xZ_VL-Cn&Th73$1qPoSj`W2gw zpac1w?2RhKl47)7WbBUC^QAZCk-pQXvywbyHm=8xQS z5cmvLwHh7s+H&TP;X`kW-1Ehet`W8<7k~A#%Yo`j>k2b{OY6%Y z?6e;mCd2B`TG9kU)X&#}*P^rEzwuz%UH?7o(}Q%) zRj+{vjZwfNDSl7O_JoOG!}4fZb6)CbCp&&OR{zlj)-(pT<9*$fO-}z905==%TW$zu zYBSTum^3^Z3uU)E)QXqetHGHoiZ$$kGTvq$i>%F3+^Pt{VIeD;&Rb^3!YS zXdkJsu6CoPVaA)oq%HA>7g7}=rAJcPNj+askZQvFUlUr~(HNW19wv%keddAH}5 zHFKJwsV!IB6lVLFzDoGxI@4!7+Eo>oXRBHqD6^@K)rAOS`qr4MJr--5W7-+W%?dvU z<3{oc^e&q>E*=?o8s`|{i%8VU5Gg!TxNmy6evn6KN(ENZyq9z?(AwPt?cq-fC2$NK z7|~5H7^7N}RV|)$l68b^u!{~w%gt)MJ+K?Rq_bmZPNr(s&UFtQ)q}o~i={wnO)f@_ zPKlW&A1*RZ(6sj7+C^~puJz^383l_VueQaDeK~4by-_!0RajMo%(p`)u|KV=N z)IP9sMXgeCJQZf67PUPO>z(dYfeluOQJ)QfP1CW8jw0)3%vZ-4;Z?7e92nN8<&8~yrvx{f%E z(uHeWoW83sfSJGn+yz;;Vv{GriY1x-_0{gnOO+Uo^J!KLC63}3ms!<|w_AACxz&@z z2PUAql;)#81UbL6ie_xU2hG-L6lK@A=FD4ML`*XSlT9&T-`}A-!v6Nfpwo+%*3TTiJBh=XWW&I zNb;%=n6DvqGpi^H3lgNAQ;OFr_mzD2PUpWluC-!PkT=`V&S$B(cVE7aqybBJt&6s@ zDLvh7-`#@xdF_REj#X1gzgF+lhD%4MBBni>P8S&0S;Uvgw;#-B9umz`(fya7F!qrUrCU~57(*|;|tk5r&)VD58nlYrZ$!!6>eLiU}_f?W|hT+vCG-5+fJknjl)mVIaAuTCQ%(WD$-8htb z)#5M|Ry*UdXb_6OI@qrLZq$gE^zm`q8TIMJ8O>Hd(gW*i^6ez@fm20_50ifN+gh8) zUTFG0f0av#-gbSOb&KJZd9=t5BhGl!XD~?JdQV*NO%W@3`xgA`@!Xut2YC^F$JWm8 zz1$YJ3DA8Z#|;?R1tTQb1PN4rKY@ZPcdO20G#(q z=h2g_|4JV9)i3`_S%(3YllZy;EL0zZba(JKty$`OUkRNXUdqxm9CpZi3SE(F)!9~n z?q8U1XBHn4e<+XWWQSsg3ehUqDtMt)mtt#-@D-r87-!WMk5~%9dH3<9t_;;#hq39s zA%s$nJQ=C>s9x`(%*&zRS0|hRT@~7VrXyLl-W5lzDySz-W`6}zF%t-=$L9k6t&G1l zzdDH>wOy%-p?u-)(LZXJ)85w%yo6$svQJ@<8_P!Iy`|vhZ!rZ5vB`Xz@77l(c=6xu zy$XmCcmpiDm5b_Ksc#AH#uese%-3)ZPdgpXjXH7%tSTLAn>!aDGJJE6mobX$L&6exST*UchLYGC*4uTH&@>k7lk})5i z9c@`<5zKl;e;wan61`gxaIPd0qX)SnB-B3icHhzgIwqMB@hKXQ0cDn6v&ZMMR5Qsq zTlrMT66hDYOnsVIzt7+`-QhZ)$zA>d12rafDGuc6v0XWLa-you0RwDYPqDErRCJj2 ztYCe8ft*k@&GA_>VS)U6$A_}lenGvWIK~TOuR?T}V|X=DQ2}WTB=3`3@)~KiODyw_ zfW-raNPfS4Q|jffdeR;Y$@qP4zPNErg}}+i+xeM@{O5jU*PK(Nir)L`3Vf;tJOwwq zzH6+MH2T+F#$}Rvb2UmS2+KLB1Ghtxpl?c`0Z^d54HjU?ESt4o@7vaj!^paG(!Jhr z1OWayCDp`>eXvP-JdQ(IFUIX0d*ytO&RW7k(T%%Q!5htXwb68}$_vux3! zo-qi4_mIGsZYHL2ICeeAkFKr=j1OFW*0C*qeF-_wPF@~cMyXQu&0ewMQPH<}pKjn| z_y+_*Kdv4ZHZOO^tW8HY0vc{cXZ<_2Y@!cR-At?D2u0c1?aJlrhNW(p&;D=1pH4Hx zW;Q>%Etul3=>3}AnervSyL)(b(<0(pbICuU(EsMY!v*`}Kg_PH^^!Myga(%Zw)G zl67aWw=j`RLIjnnLmbwDogt21`FDXu|ed8nod_p#+EjZcNppJ-$2S(YBZB)hefatu=AH zV32ehQfp*LY8WrRxWfZfm}{$wzYqUb&L4bd>(@8pBK$>(%Tc*FcL9R)p0~d8m|fr5 zB)-A0MkRp#hN0znTOqPkY6Ql+Z}089=)~I;m+69Y4x>KvnKz;a*NX6NCJj?SVP+LE z3W|4|BmRZ%hm(tID}1RNjb0&@TXP1$n>=V+#yR8o9IngzNcy8^M+Yi_Vh64K`iaWv zdwSMu?)M1{V3`}lzV!J*HEoF%hB*j)taKM5+4%T z%X29!hGI|v4ldqIL$U8Cd4F_XC||92Ww^WtOk3g205e`1#sv2Clse!gZFb^BtzJ)` zhhG$OGy|1Ah&yrg3Sg7Q&8GV5^=60p3W}K3Tn1_Hu_FyBDEpR=Y5sAvMgm%W$;P;1 zO(udqJ7UxrqON$@qPa80?7(#6OP1)kD@MvV*hoPG+F@)7m4+-C4F8Q!5pc45*_h*< zN80+)_T$$Cznd`pF@YcTx!^ubm3lj~oymNh{;My5Xoc_q{{|7yH{!h(`RS7~bPNj& z#q+y5D$+NwWTW^3-8$OQMVO{(jf#g}c%B{Gq8ck?m*V|7;i@PJS@%fYVj@FHbhUc< zkpo(a7LsuIY?^xw=}4tX*o0#Ti96-0wphy-r(o3of|>KL0D3^y$>*lLvUW{j>`9o3aD>a@y87_B$2Ku9El@~8w zyaM&8DpLo~-EAO>Etoz;;fWLC^%*Hcc+7vnlufVi%=Ki5FjzISE5hKd(R>j#eQ#fu zd(HDe<#qFPVA(ito?tJhH+8v>#x|zot^9RjfG_+9@4?wCI>m2l#`5Ejv&nrSY=h1g z-jq?#Tcc9@Hb7pjI*T4V!%$AB?CI&*_$3NwVlyi><~CsXebV2*e(!)R7G=qL(xb+Z z7CPRiJ`*1f4Pp^!T@)3Fd7oluf2^y>L{ERxv3@@Mg3R+;SW`Ajy`oe)u`8G%nsV&S zr^c78h25B-GC?3wGE(i%QpnBuDLZ|oB>QJ-Xrh?a^g_Shvke;Z*>Z}fAovLFu=nt$ z!MhJ#m+$&f3w2pc>l7Bdy@1)|Tl-jn-U$?ryP7^1IIjM&$N%d`<3mTXM9rsP;;e1& zU{m+g))Ww+C|vPsd48lNhQDuJ&Nrj8zw=--s@FHRHP-5Sou5LpjOWUm+e;&Yw3daR zpB+;L8~#6^yhnzA+TW7>mBR4wT(|0x ztFaU7vBN_}rq$g3yW3_!0oY_66MV(E!l18{nQg&iOWJkjwC@iqfY4Uv_}0;^+S;7S z$5Jof1G$=AqL7e1G|kI3}=v_*q@K#NM-t#yKG&(BKSt~2eWh%OEakvwJv#I@pX&{SPu zDKb(8bgaMM*xwnof2^@zS2?2$?GvH!2^rc29ZRz#_foT^VS%1jL`J zzKbuuk&&6Hl%EcLMtY?5Qe|2tHVfD`N`2~gIS0PT^b|NhSVhj)l{0mQ_d*N8^7ADXe>kB5L?j{v8b zBeHzIaX|iYR)2rZe|SzS985-xD>C7)efSe600U{C`E1q?EbG6I@%7e|fQ#<3#3J_B zKK%Zl0qKux@%8Ed@CE;T43wooZ>4r8_EGj<`|#JnywN>PiT-{D_wU}H|1MZGtr-3G zzxLseJO}gkYq6`=k8GDeh6${AuyLiyA&GzO!|wz0#&N=3%?&Hmr6l|Nqhzs>%?oBhx4;6Doz z|Npt!RZ7ILp5GpYq{uRVzpLWke8^YjQUV_a9PpMqTk8YU(=E|=1@!aWGR^GYBcVV6 zzSO+o@Vyin)aOF&W(NhuP)?CcwL5yiPK$4uyvPq2fzJ4@@3&6=u~Ytj`U1+=c}Y)& zDN>gefH4LxfJ8KSdwVh>N|MNc!0g*mF)nsW_xUy*ulH^kC~C)c0yIR61{dx=IYe_u z%GJWST%T=ip``_t2g8&6%SsO&YU~f&`FDZHpFcOOvp*Nv+@GrGZ)Qg9zL6vjwZ@jS zQBaSA881Zf`0qKS{A^m~=CZkBZdha4f|BF0K%7*L6^Q+e+4F;H0ziPq9WwJSp$Bn? zGFyP@t5>DAjYoUPPbKBuiA`ZI*e<2d&RqMk7$B=Xq{LVa|XW`68>z9{PTLP{Mo5PFqJ%y*B1u3o`VrO4kqtyB2AJo8mjQE-bR}T zN`T_6>S+;S8mvS^`i^T;wtEgU_@o+Azr+rzCaTN>{#4{KR^d;;_seo*Yu#l?$DRow zhcxw7yS}Y#W0W`QwVapMYGMfC(6Q6SooIhB)1f;L9z;nwn&sXvdJW1a9kV7T9`DBL zptx3=_~qsc=79tnau^wT0tbV&0l=ozeDV&|7$||i9w!5-+}pFPmN#P?+s{5b#^NH| ziEEwj%K^Q}v8By&%8~s#EW)N$<+=|PiUekiJ18_%5O8VBhke43_)gtqO-6t-F57(TE{ES28MDCgR0m)Pu)2g;HcZ~znq;E5HXT2=1%O3R?=r?A zma0@`YBrJ%($Q5rc$l6ePe#y?Qeew*sVJF7ZeI24N`>L&^m)}@UuO2FH0F6K`n8|! z8X2e>P1dAijOj40_&9`2r?5giZ;0D2=kt74Zk7d5=5Y=g9Y0ahXDi&9M{0nmH%U^c z-mn54Sy2(!eLl=S?Tqg7wc!2p^XBgd;2)1#jJeiPNvT0GbPOCq<}{)2V&vVvJY3)2 zcAtzcklRkln*t?NG>Vw!zZYv-?QYtyu#P z_)P^%pl_)B%#U#kGrL2J(yj$7)EofHtu1TeD`98p{_oYpsHM zvF?tu%Av1D1KoMukoDdSDFjUZn0G$hPd+|qO>wdBuB1uPn`Pqm<6FQ@lK)|F7u+3K zyd9WH4eQ3_93lMB&>q=5Q)$i8MVHh~0pjCt%W}Dj0!9fv<*j7E{4CJ6<8iDWfzPJ& zj_G_JDBjS~KDmRjNke&1d}yQ)myJb3uY6aVT>YGImZt;1qV#YB3&V&GuiEV`D7bz~ zSyaf9-CfbVEiah)3R-83q>V4QG42(a3PSuzVY&={^#GoSZF_o^-eX2z+*lqHaZIu& zu!RrUkC}#M9xeq!Ojhg3YXqboj~hb@S=*)?v0>kONe}6B%3oB~i-1R+rvH>WrAHC#8`UFzZ6}Y)u1)Ceux2A{61<;2Rh-Xfn zdUPloZaO=iUxgllj7c0H!C2h&>VEo4BB4M%a1zzNUIWdGBuFy6VxlexLetQSZlPOD z3YUox-+uZrf-{Z#K~{^D(JwL?FsI#IVm6(X3_Y11j?w@WemdJFvIFcERPNjcjz+{U z;}gMe)d>RwSIg9tNJ=kRt1m29MTtNZ=I!%B@ae82tVy!ywO?F|1Bbo41)cm6`j9z>sBGvr)$JRHS;cqb5YOwwh*9Htg;q2%*tq2 z(~j^d|FMLNKZ3zsUu9Na9mnV$TJA%pv3&LI(D_np23lc(+(3m!&OU40i4s_WD<;Wf z3IqxE>kHR%SgiBD+Q&LM7W!aZR^QQIf*O~&ywy#?8!Iw^Ww*fZSbaONA6WslA@=>c z^|fx#eanu=sc^KYkkY5As_ax;)&)<0&=AzAP+dpxLtt7aBV;GM5XrUB6;q@V9(cYK zdQc-kV?av<5ntGElOz>?^U$%O>fup1Dz zl1U+Q%#Wh9gq`Kc(bPJ7;pK~xrapok@KhA5z@7kXkQzUfs7Ao+4|^ z(vJ#}2by>=xqY}5jM;^B()nc$SEV1?p*TG*2Q9`r=RV$Zvar{Ht<`Mvb~#HfjbH#= zOwf8L8BZ3G8*?=4VRdHSJ?MZ?zNmMwJ|y<|InheSiHZW;IMVLjFsz4;Y3LmX+Tb^kr)D| zaB9HUrYnUO_|dCgw;gnNSoT}4@B&3yd38x^ww_MD+JG?IE{(FYTEp-%6|Dkto^WxH4SyjtG4NMCoO#%k44fGq|&CNuyYs2>3g~ zS@(zX4nyUujgc2Kbg`56c3L^%`KmP=c-L#LGodVgOdTos*>7|{IzD8tUkB4;8&UI{ z3t9WNH^mWtFC=;=)9glvy&XVXsa|5L~ar7GL)zsDv=>mvaM-#z&OdfjGwWBvoRy>C_S&9o=Jy&kZCU<>dpEN2P~JDoj7 zoP*0@{jG)(Q1hX+dA;}=uhmca#32fCm-;zZPl8YvFB}`l`%6mSTNex=z@2Lto=jLB zC^Sydg{03{Ha!mwb-TG*Wbx&qV_|oyv^H;Zi}VECz$qo-Fp?Gg|R1 zC`cw5eO47ml9t^T`a-&=lKK1;&uvJA6if(gD$((>F3TXVcQ-giyOGI35prK`?kxUL z{pxQC*?1F+a_=~%Zef}*fQ%n-7&3?g{M*Rq&yU<)e?w1!iA`8nk7#9wV$aS}fl<)y zX3(xj9HQ0D{UUl)27xh8nLy?q@Yhc(uP>A8B|AxL%Q}{70W0>fE4Rs&V-v-uYRQEbS<7>Jz1xSuNAQ*7ns~-8j&D z@!~~c<0$JA`{~xs>!eV9xEPQCzjbTg%Zn4m={~N884O}CP zeM_x7?BS{cml*cw=;)fgAKp}l_)@GWZE`2l$@S}mKD^xEU#MdOs#T=STgdR5;;xhM zjX)p)!4Lqoa&yoY)>k3BtEs7l`Y3vrqfJHOL*j`DX-_jimoP}T!Ufju&YbUADYwCw zxaQE{jsfe!Dl^(m@iHjYv5qJSHH{cUJEnn4)g0uDy({0E-@fwN*`O{LRiDBZu(S}b z?|=~Wh2C@WfSccJR~osZZCR25k)KilY2O4b&)69jH!?s&j;D2j1jEDI?5re zf@fxZ)qaCX^2cU6lC7e>yr66~9F!4{)xpS%8Y_q#`j^nN*uD1w0~d{mH6Z31WkgrZ zpKVigH!)F5)6DcP|c)U5kZsiI5GoEKh#v&a}P zzQV!5k@|*&9l?~r!@Ql4AjWvmq!+gC1H~1V38dgB_ExqjZ#KiV3l1RAA1s@|%`YmABptOD+Z zG+vGs2B_7g36tW*JMCp%D%om@1;rRsC1a}MXl03KGGd^>@U3H-LT^($%~q)wgup%B zKETCgDv~}GuYRlN0>s}mgyo98qe>|m?3Fe_kvgFRhZ&;g=4{Vx4%!T~^=Aa-wwvW! zt5Yf+AyB_>(q1%Gg`G_culj?_5^{=eUw$FPO1oO~St2Lxo_V8`0fh&@@Zp>5B*WQ`>#vw0m=8qt90c=90Xi| zF?^f1((T)o=b**7(VS(|QJ+4CIwVW=5C*6jk0N~GFHzXyTy9!IirR@dCsv>`m73C< z{E!0roeQ1&vegM^WJDlEuK{mQ(;m_&Em66%?b>CyZ!MZnQ)agJeU;g4zI7e)49EbE zi_i8vtm?ec&|Q>H0_PNJX47eBt2Rh^ZKElSt))4e4MTi;!2TyiV4A*a8+drt4`~+F z`gDvLBsJSPGi=LMB#oSfYSTe^i>A=kr}Q&S=Vf$OsYf>Xnv?~nn68?8?+lgPBGNH& zl)qjdW@(L&7g1}Zdcz64Ypd57|H3exzsVI#F9s1H`rIDfGCh{)zf40i z|1(bY|NQiz5tm(bLqKg3)NR)-dy8_#s+HS*?Q^z3rJfN^|E3YZtTJN1jzAmvEVG_+ zX5Mgu#5OZ4EW}=t=q^G&t@wEM(4j+ywe<2*%|GmpbcD<4(y^@CS0dri?<+|DGr0MG zckb`Rk^MiC5_19Vu`RfVI?Iewc_r+YL$59(0^ah+)zLi zXxZBMMf%4E;J4MowL&8b_%kK!`}ae!U1utry$?tioShky5dHFiF01C!^( z!G+;kNT}5mM59(3Nmb{UQ++1Kt})Xk=s(pbKzqx4WZqrbjsokksN2a4;srcEOr!(H z`2o9Yf-sZIcN%t>-%-RWX;lgEjweV`zmC#)!oB8zc{Le?WOwP{8A;Y$oK^fMl7F71#kFO3pOPJj5DsBl@APdQ37vn27~tIEcVf_Z=u+_ zOs)~AyiBecs3r$>QPCR`j|ece0=euXFk%Iy7Zv{Eh%@b1%FX?eFMsDt};obFmzvnQa zMX3*)=H$}E!L6#5bwqGu8znZC5R%Ek8#&su z2zw0T_=G{NlDaJw23dYx93A?VhXZJv&KFOJR726HP2*$g&^3m#xSba4<6Jz_oE7-Y zZ!)(Rs_~LZBbZb0d$68Mx;-#pw1$d{4afX#AMD*`81g4>s<;H3!HDbS!q=8=Zx9_U zKNDH85MYl;eK7?li|Ayqp?yAR0;Mxkej*o4WZIR!+A8-+ccuPrXr_wrDDq;)qSqwX z6Gb_RGc*{E)kpLC6~cG(!66!X|BBCp*M&aMoZoAjv)?m(2T?!>cMzC;xIS%d-_^id zTwbnkY8`Az|86hkOV0K7ArO79u!Ge}7zWRV48!s*h#Nx*@q^gPk~ZiSSW8$u2+NWR z{!@^oC>IA!O_eNMRn0e;N0db5Ur2*@kn7K4BHLYQ zwO_Dn#{ACc{$vrT)me#_wMf{*<4{W1h%6YC-K%0wSWRC+QC+>I2R|1= zuD#if4=Hr+#~3A;-c#k7A#Szq6#ZaDzoWvflC5H*72uO^=QBL^E1M;o4bR4Mw}$U5 zn%exF7H$xS#fHH~7v?L7a`xy%*#iW4W!C7wxBG$_1UX+@TYCd)ncRXoc8{zI>jm+X z@;ug;!n!Afj9n%g%F~Kjj}0*n}*ym5q4yV zSR9gWO2G(6XUeY!Kio}Ujjt8FYARZaq`XechsGAnypGt6rmKqN4cpo+ct$LajVmEZ z^hv}#M33g?6IYLne}Wh-S(Nyf1T9+lL@HEE`gfE};yN@tzX%FR79+`IRA>kjc(ySw zun(k+-{h*HYs-47y_rEJkE&J2B#9Wv!<$6+K-CXo6HCpvvZZT9`yr{CpE^luD|iog z(hXA_!O5vP3Tt6u$bzdl@SBsEhm?6hp+=#C;A7at9m!mrrx~Ry&G^p`&G3fZ)ZRjD zKTFo@&gs1Qjl<_RO3pvtO16`}$b7>=N4Y^zg^TBiOAt>}V5Iqq+lYsWGybzvHV)09B%Gc*|#)5-ki{oGZ@i% zCWs`={-U?Eal__MXQ3wiqBtZtH26z*XyR8cQ^sq>&kS^5XbP1|SPC2p-fOUF3NYrD z>1%!b&dVK_A)PJDS}>6p7N4RtQovfIJ*_fDF)3op@t%y^n?xd(p9G)8lSG%>_&1Y> zqClibqOllz5#=|NVt}2-`ycOHGW^FDGw?Gaxt;B9Y&q*??4<2?>{+Ll$~?u7T6pA_ z6$I4`N;|~-%5RuX2GL{8#@F8=#N5R=#&E@8lIO=YV5%}GmPpw&sg$cURW?>eJB6IS zUqxLlXgyJivFqRFbWa;edr0R^&y66)*h0s{Fc7NnT9`EhUq>9>9YkGyzP{OCoqQ;q z_>DC)F=vuzQUwIT@K$qu;z_fwo?4F)pLfaLFz$WT8yyLze9ok{x3wX%;OliMH;Xol zPpna|PMopLo?$zGQ#DIJ(5LRx#-izZl^Doqt(q)xW`~&Zj9ns$VZUoq(I5>w^Z%VoG|-N7jiC zolUik3ibN7+RoQ2`V$#8My5s<0TU6{TOYali+}I7vQDKqW;(WY-+U&w-?LA((=`Ju zQZxYS2$!9!ok_)sHaI2T<-7HHmd6HTQ#T2T09`doH2c{e>&`)dU7Jk&pA++PGR2TXtQ+K9E7s>!p^jqM;bBc_fjjYmjq|D~M>DX1tY zmLiX$mywm$l@g65#^>VNHDr0?rExsI0C0VF9oak`Sngxq%*5MAVM3Wj0pr2r>e4sH zlkz*wH7m*^%5N4%SZcUQT-QM;kNt8y@y*+RPaDl&r zwIuA#565TNALRn+x)@?p8%2H?{KT)oZ^ynz(M1!KUdvC;w>ellV4lvk@wTZL-XFah zV;MWfb?@KV)crvHA+vVPM%Ehq>vT+hbTLVBTzO2GWh%u|QAL#}N667FbH-So)-iXJ z-)iiu(}y&NHnK$31m5j~pOjAo;jxP!V9=fM9Rxm!21NoSzxiQ?+bcwge z%f_M40qEY!4FP^<`px(Se^r0K!DX7pKhazLXmcuhp7mLWg{4z(Q%ght!Yayis6SWs z$?(P<{QIit$H+>GqoIl++lN5+aPa5v$@vo&-lJRSz0D9|8+{y38khd#|JCyFgScE}jLTfhA>`Oa-0*==$>aQ#?y) zRVW|+{iE;4ok6aX;)2&H;SKxjGDc&n!%M4VZ9&J|;zN%HKhJBG-s}8TQfiLh*bAkY4BnR}M|;DleUbL+a##+md@|GlD9>Y=`&0p`=;{ZXDy(_HJwX)0-@ zx1pT_>9p7Jckpb|JX~dV$9P-dflOa)W9B)pGf1IjHqgNP;VwH%uqtrm<>Yd;1%eFu z4F0&r9$BT6utXhv(9v4cfgfoGdq!;%j=2g zX?$n#NJ-9K$SwlFe=57)yHDj|$9mp@=qf^*PvdlJbvvq3+oynaTR%pYcXLW}kic%B zfIyQ+@fXQi*_%Q4TwGD%2cc`U1I0z%<*HavPv`2<*vU!qh1FxH2`oBeJkJ44`x-0^ zD+&yv5U<8y`FLc&EA?!<2kcu_JNQNhEli&e zjK|e}w}4B^ks#nL8cq%dk8FVQh%)?3ap`!W=&f-5D+O&rj`}CUP@YgcMovyb=^0hR>j6*LA5y+vS2{}Y#iWrTtMM>-q~Or#wQ!oSL>K%alE1nB*T z=ATdaFA*?E&^KJ@?fnhz-=&c{zrp`Ij-Un2gAv!1l$VD-H7(q%teo6!ojn*gEDfLu zC@yjz++kn{=>FWW@*1>f(E1nbwDdgml$8W5oE_QBES=4**t{KG{?r2_K{csJ_^(5DXURQI=fj>@vw2QanOjMQc+O}xmj8Z zXh_NYiyZnUOk?Ze;Ud7!4gdhy09%zV0HI#@-Xveb#j06 z&r1GXkCc_Wg`1s=hn=$%)t`FJ%$+?wglTC0H1wbApL|+*+x<^VPVWEe7PN!xe{$G4 z**Mt$Q#X`U=ufJEnw__m!v`rlN9dSA+YsU8wlX3FH*h#BIW0M_ur)d zmG%E7)poaXlXP~3HtHerKNI#Z;{VS47oiaQpT7UsSp0LK|B(tEX%SQ*_W#V92x=P^ zs~mJ9$?T+5|3Cu@vp*NyA6PQ}^9hZ^{d~6SEcgZkBL*WcC9dTSdz_7w_PU=S9GqWD zB-SUUqdxQvSL=(V`q!8cT6Kii*y6Qv<6m%SP|=5kP`%y9GtSnx zFFnbg6r65z;DbDPl}LJeUYl^Q3VR-|eZJoV0ohSw5#)mYHhf_WW<&+a*YD~%;8MZ< zWl#?4;U$GbNr3sgAohh`Y(r39ml91Cj#lh1!$zj#Uw1-|N*?dqS$1K0d^p43_5rs! zzXvkGpi+(a_k*QD6+GIsy1S^Yd3%FS$3^qkZMi5?Euw&ojS~z$Daroj0Km~EoFOP# zy0Ck0N^P8+yqgLQ%W*Qv7`I_ua##gGB{J)#~dp{v{5e#hMTp{oCmJ zJ^r4rs4p<#V5twREU*6#Jy4JZ8Nj{nT>WA$@Yj3=OUVcI0P-sc2tNPqN|5`>3hlm= zJSX{I^Cgx~3k9~i4jP*DUr*|K3QV|jQ(;=%U-Lz!jtd1hV|jS^{|(@RdYTM0{%^p> z2*r!a*V!J4#J^nDFeuG*VjIh>OvP{hdLhVth2n)_*xoGf?_s$D?fyy1>*>Gei&g=O z7vYMOH}1S6-NfZP$K57Eqq9ajyt}_>`AKvVjvEwO1%t}|R~B&Nes-169)dv9&ydJp zKwz}tt!(y#c~5q(4Rk8x#NgCv$|`YM-p2l#91;?93LG{|$->2?);uEb2f?`CnWY zhR}s8ANI5V{i1Yb7jx8Co`cs(Jx#Kn4TYxe@ot}mGD~ja;)U>b|5ve%4S;(QWL5_T z7D!~|#y->tg&@dL?O-?G%d~h)s^FtO4SaPUJNGYwP7IvV#`2ql(TM+L1$=oM3=B*v z746h0+FMzbZ-Cm$aOL#i)ZNaD+iB)4HfRwhStXRI)bG$Bv%hcuJse-{mz{Gr!dH6e=Jz^Ea zL=HaR<7S7&+FroYaTD)y|A}?cb9us zL$b`CwKUQc34(QPAFB)+Pt_c^n+p~2>1My{=;W43WDhri$f9egXTOUu=^E`k>zUBc z^?aW_+8Iq3y#oA@+d%U7YquXkBM78@c>;)E?2OU1xR?8=;ZC+U=hB+GO6H)RID(#? zcP`mGzdSu)$rURB_8(jWJylhJ?yW(CNNszFLcf-9N@Mlpn@f?uWfydij-;J?F{wYA z)g2yQF2f^Pf!9Xfj8pnFf3h6g4WYK^7@&pzQV8S0xP4PJR?d$cSd(Hi9eDdHr$IZRf$t<7S+%3kz9^orRR&TqY zviYGO7ZS_#;WonbLhT$?f8Dp#)-3($ly)A5K`DzK9uZ^yphSs`)k-1NBcP%yjFQ!e z>4))2DZp{D+VGvt_`c%YRS{3E@0f4L+IMM9yZOc<=iKF29y7DP#fd5+drxKN(LPR> z$h@nCOk73IFVy#EP#AsX2P>oSmVb(i4(s-%^A&nAX9nFx3u|btWuM163(xe3zC6Cc zBCKNU$`v_LY_-|*mCt>)2Kh>`N%hcZ@dF=D3SavaW~uO8jfrT~$t0^q?munODDc>J)x*!^kTS#R91 zljReQI=oux$ZR*eDQ)tMQnTLwL`yw1{=sd+!%!J zL=7_EV_qKm9gIxs-(HjlTxd1yQ|{=R1T^?R4-vGS3Os+R#|}625D}H6_=bD@;%1YD z6!CODZ}PjruZswtm5MQroNH!`$)c!xJuTIsQPX^WzUIaZ{*+k~@j7wvZOV09E@>#< zPtz3phQkM3Phdf)(96n>Cn%1CFr;Jez1{ui3L{G4GyKpGOZ8s7->l~R+_y1i2YPDC zD_a#XkE1EpKX0{N@3V7(p2TXc#%M_AjVQ=otE@%crS+XvT!maOZ_y8}=Q)>kvvD4A za@N`Wk%s>YR4V`a$V5$3=iv%UJ10GN++B{DxBEXsK)~NS4fBt?0VkgGL#zC5zl0&H z43M2+y7W5B^Gr4-$CZ|VHgwAOa)d4VV;LGvX1#@4k3m))E*l|NhfCVeI$hUhp9m(K z?WThclD6VNs@Y$3Gq|Upav(dSOdagA?&e){(RQ`bhCp3&cI$jr%5dlS@APzNAhBED zCXtJ^2X{}R^Rc7Ue(bDJY8hERfj;LB?obC%y9OW(L6OqlgaZeWf|FbjGMRnCRT7sP z@@m!}#kJm)JjiwQ?f4KDg#b%Ku8+$2qpMD9*OWR5cPZYUP&*$LF5-BcUs$$DlW*cj z=iPqA6vd;)Yzsrga4->{hg1-5Q7(4-!i-OA1o|=PjdcjjN21qooP=}bel~V99QuW* z2`jU^V&n}_2;RQ#9MLMJv6-O?+Dir;GP7p(w;HpbGSs|NUdI5w1gQ8OZTP)XOxhLL z2}qZGXVosL@?DMss-5?~U0MN$D8&5!K^>VRY{&}} zs_6dB^w+c0@k=#OS1{s<_KvcLvaI z0)u-1k6gRN|H0oEzF@RHy2F_r##CcNjP`bF|A)HUoBIfNsos*@8^9sa@|jhn!wpfV~Hcr&&&aG$wss&HagTie-jHH@;y{R&~ zfy2_7jEFVC^(X{+Z((G>|JJyTQ?BTCAA3bDw;>;iY0jIsZsNEsO~Cqr{%l~StPhvH zeLplR!?^wrQYl$9^koTtcA!Jv0|^eE`b=_AT7F2DoL+3SOW%LRI1(+8W@uZZZtQ$` zj{FT=S7S}H9{#F6S04P<%MaCa0>e|ddr7X;Li2Z^!7UGDyeIrL70X7|J=|<`A?lsw z!afm1rQ|F&rv26!kr}(k$VLmE+18c8R8$m1%qxK%)vcH*;@)^g%AwT+>%9dzo|Bh} z%lTGJ!2exRZg6(%;o*(wK0xTPTIwqMIiOYkKvuDj@nZg=GqG~@9#w*|mm?PKRQf#t z;U@u3Ds~(hSmQRQSI7Sw&s58f~d zypz5OyomB@?~e1G5N6OOqbUOi(N@DC<)0Wd9jfgWK}Tpw{5ZH-e*K*}zUy(R`dfiH zftHh-<(+vv1_bKW%m!O(D`uUTO!^J*sp0tHkeTUxzrgLtvB{{+eBi0wu6C&Z2psp|DkK3_f@hF7 z6>oX+o0xVxIit6aIMK$wgC>}%{R^n9hIMf~Jj};;cNBMTSu;Caw+*%|`$}DX@!B=l zi*U=N-}ZzF0824{stqzkKK%n`%%bY*DQpIGfR(^Zhs9=^ZL(a^r>kfo{ZsiTiR+o9 zTQqHyrpkm21M9EFz+T>NX>!MCcyYy}Zm=z67P)1YMIfVG*2NUQ^f*>WR_e$oSB z-p*yYpV12Df({$YD}t^l%&B8Z4_{LQIeOzWYVp4x?$rf=uNeH1Mj{_7SzsUTh$(v@ zPvMFO$XK?j1+txG;nQENaZzu=ms5VAGPs8 zeq$N9>caeQ_`skL!Ks%tMj77DSI9gPk1j8d3w285EcXENXrv26QBzOqDLY?`eZQXo z%RruIv^noT>SX1|@n3GWHX*F9oPD`4d@;C&(D*+X?>q(Uw}EL@JN*1(CBFx zv^NCT`{{4U%QQy%uljNQnDDIbpp&k}dQR;Rx>sXk;^y}4wAUQ%WAR^>xT)=@lsjGF zSFU*q>{%u?1|es+nI9N-*-dWTk2QgYaOg@)I_Vp7da>LIdYNZ)U*ValWZh~-JDH%u^{)_G=+d-mI|IF9QlqNc)-e|uC$50U-g&&+Gx?f&q70`hMAYuW@j~5b zv1VD)G@Z@`{53p#G;{sL_~EPU>)q2%hR7FnA!Ir@c4MLi>-H_Lju_yLNYVFdFhwMo2z#pcAT-iNZUxiNWG)1lM2O+OWip1^Q zh^@DoD7n~*7yYO@3*Gqql9Wd4KhwwXGOp4SvklWF-4*P(+s*WBlU>G|N$JhrBAw3< zKb9T-7=N!2pPb}s@fi77g@M~)y+#m|udxJ@mec|&ib z0rLG8r{C5guv`mXS42yDL+0L_qjUYK{?*0adH9wqvp-nVI6SgbG`}-a*3-v*6;K5# z!#oRExg4^&A7wQF%sl?=2}|lcX*8fSSzGb#T7E#mjhsdtd!jXJt-n}@e|61&Ua&Pi z(W!P_ty*h6#&S0UA%>Uj;anj+{9>MLDyz032X&tl8Rc+7Y*I(4_)`_t{AfLyBW zI#RTfr_-el+2_Yg_Msf%^6lmawg8eDw0xDH*kyfb@@N~Ur)}Vw%x=e$Vfcv}KX8jd zB&B$Cgz3W07ROBLXxBI z_mg>Gt5{sh^Hs(VwR#m6R1!`L_1ZtN?ly>lZCe6aR><{EqM}=hx0PBoSyzWs{e<`o zxEpOM9rTBS4>f_2`tQN3VyQ(b}^AO_~|}pm$+7!b$4eVtLN(?*_Ls+ z_WEEM_(>__ou(*Zqm%xIfP;TCn14{aZQv4_;J9PSWD(x+QsP!-Z`{VOE1|0EnSQ(O zLSdGDX|VGNbO@p=%h!-kVM#BvO_YEY-g^>eJ|C}r*izkQxOT#uw^8B)(>aAZ3sp6O z7wBXIZ{=6JribAV0Vp{Iw}I|)U@_L*W%+;W@?V64fvWN|=@Y^oF*YA(T#`-veFEBW zI@N!^q6}oLOJ2t~TI}lD*4U`Zj(~tvMW6MQC%m%m&YIOyyv0t$pXj45C@d3)CYFl*hfTiAsClDJx;5-t!Fx4?INM# zf2sdI0qZ4m?HAZ;6K_1wzLWmWCa!_}(xTVkwuBP+Rk#f>!aqxSepNSd=&?%~XcBOD zw-+J~+kEOboAoMjj zr$f4t?pOV1ZqI9kaz!;M9!r2PMo*pos(z87KG}+EtAI7nRDa-Yb!YL=p`Kpw16X(Z zP#Utg%rfRV(6FYwYBiE$6iF)!ngoCXnW0=1%^9e%FW4AVzXKI5>3L~8r40$HwTY2T zBdRb?+|DQ2L<7K2WG(TZdANv3Cwx;@Xxt`72%ZmBfW2PF~g_r7I!xYUh{h^}@$ z3pf{|vx^^&(q(+-vmhl3i9J))#5LU>$a#34Lp_ip&D0!$7!If31629qdSR>voeE2U z!@Pd(I?7yp=CA4@eF=IYwj9(mYrKvZ%lfxYB|?E?kLmP4k9I7InPrBx^phJKTJ&T8 zN7iL>$VVNv1euAh3h6x5oe^Y#3wUB<5{2X+71ML&1_o~1&5D3CngFT|Qg+)jjef%~ z-jhoaYUjvk_Gr2{fpy2P&^&(L>)5w7zY&%vD=m0E$NTlg78)Ft)q+Xf{Vd{~KP@1;J7J;n7xe|7gO!qr-=ec2crGVY z>L)W4s#eWEq2fOa*t@o8e(Zg$s_L2=WO$cxX^vpmCE{$$De(2I!)%LynCHSVcyi;t z2QC0uQ1Dg5JekwaG^ho7HtjO$H=CUP|0*F?(^QN@0RZSY%8aDA+V&pXfL zv7>cd253713}W~jTtNr39d$pC1S+l3K*+=$1{k~D5Yo^rFNY8K_0Kp@u0nZdg5N@sMJG)~V7g68hcJ3@oQ)Q?F>BAnqv+W?t|0CiqaiNlAwQfw2-#mIiW zh&|5T#=!u`zG_&`oAe7_pU&WVyQf;A<#lOc#z9(zB2 zJFA`P$=6~hi#P3WF1bX)!1&Db1Gp}4a5{5+ZeBfg$l5-zlBTo}n8CYt$1`~+f_2UF zra6cTG$VB_Fode7UkOq`3ZDbo4!VbFcy&>!1ZWIdJ#Mbk2r zhxxKHzWW=>x8SiH(yb1x35AtR6T8Lta~<#UjtE2P@5}GTO2)iRWDRRtR(wAq9^{{5 zJ+DDVpO7G`Gk{F5s-y1e8)tMO3l3@6{APrA>}!f4tFlu!vRzMchouj~GB{^`D4`rS z6Y1=sk)1_?d|1a1d@>}J|6~0Y)1Y7XOuNJa0Zn$tbI9RQq;!ddy0Zx=jX_YEA0|jS z+1A9S(IFFC_*t?Vl-CofN+zhI{K1DuriL0pw@O>_dzmO#^O{Pv`-2y;pj%;sX&W*Fju?iNR*5=d}Db@Mr zB#FW(9|&r{y%@1StQ@>DySbSc@cf;B^SMazTI9-IS=-YJJ^vn??)l~7a>6?4ISh|v z_|d=4+O2~U2@`q88H2P~+U7ybMcnpu_lJJ}Zb?)gFNiczAkYM1G@d*n%gZ(iqlxDSM5xmDR{0SsDbyer-X?ca8h z*p1(Lfja)k>V`|N4zh`j@hOuK5jJEHcq}GIVoZ6_1wF2xbK}#l(>RVi2`VUYvM%mg z{}z*>uZ3wm8Mo8*xgMJ!1>!NuEeV?MZMxJ>ApfmpZ-HU4rr=n#NVw_@=>O&Tnq^r4A|6dIb?H zV#LGj3lUg>yRXn@_yvA#sAGh3gxmzDHcCczL88nZ^}A@RIxjS0%ojWlE5Jl3@D&TA zh&M-Uo;m024mlzLGJu@kBNV%Yxdjj~SM%~xsJ!E7wDEYc_OPXw)4PAc0)OOtH;JT0 zRJ@n0BURoAsKM+fVtb;aItE2P@bNtIW@*63)o`z};~GuAt?@_w?Z@HJ2`mxA^@mDx za`u8xYcNv?A+YG~U&hnksrZeUkzbBwV4s5m{lw<61im@ASBww-O~L*vqXximk8YM)=~m_S75DhcTVS*y+G```> zf*5$jtoF$({GEn2&1j5220E*eK7V!e85t#h>keMowf(h;ioH~z)aD*lnq+W8RAL!_mM2F1}b|9mG+d~@&2R%0oqAig+1 z4#e<9n$sD13v=n0GppC8-2|PSlL(;RJ!9KEO{CAigR;*A*!3&2tY>d#Q;Qr9G&Hn@ zP0Th>K?5JjCVu(Nz~FidQflQ)+^VC8?s4x@07PEkXB#e!dF@9L?u;5#(Px&ba-wC# zf@SY;xmi&WJ;hx{GU)1VXdbnEA~MDKZF{HBvl|hM@|1QTUMp3L3-I>vA}Wy!dl_O7 zEPVs-mCVL-cn9{n1cv2;>g_Vy*M7tocJXK5guSR>6o&#TYMd0Pu(;2L3%PUuA>N(c zDIU0RH%Q?-iRJ@MoKX`*(+qC)77*3wWb$R5cCHMGy3v<{Id*C z{cl^Y6TCgth!Os`-S3TTZlG$nREL=weGWBIm}fMJw}#RxkmFTYjoj%6dMw^tY;3_U z3gIAxjSbEO%=r)|mb2pu1mn~rXUBt?o|G9mh>MpIzRR&fjZsrtr-4W_LU>l z*PAQeUJsWDybnFXC0$!9s2~Gg=ai-h%-Ij`y`O!2YVEa6yiAx?S7AGR4>|=uRq>y; z$iAcqd~5ff_%0%{uw|dKzejc{MJePWMrGTRj*M8t*OuXY|7)YSwTtvpv&x`B0jwyu z@qj~T+#5}jq9T167MuA_8rY-Rnu(jdksSwFZFAdBxN2I-AMt-&;LP65@dGF4D8vl- zb#bDT5iOL&pKkSfc~|!?`+-YxhrtSH!Di#<1XZ2(n6B4R!HCH2nAz|Id)4`tRBg+hj}*jfOvxSf@cPt4*mwyUr@P#?qs0lHf=_ zhxYQ`AAmQssa0C?H)W0RPU&^lvyMAWChU;1_HT&TTK=0T0c)mKY$b^7Y2rBa&cGhX zmse<(bMHI`r$s^H|=(b$G0O&&!Ez;?#=_i><5ld+U}o9sv^DtT^}d z1#~f4x2DY|oQ^Rj+=5%>jpZ4Yuv}wQkjfOPbY^Z-)Df<+{01pS6n23rcp~HVo5CM0 zFrT>hm82KU!esmb1A^A6r$lA?b)5vku65w)OV5*FO9?=)yZdB9qQ0DX$$d#1@4%TQ z27M30a-_;JYO?*line!3cfspC;FKMFsI1b+9Qjo!i8(4u!dXBaYT}pO`}eJHX2Pzd zy^l8D4BF=rZnE_>_D1{1TpaUIkKnZ8Q4DADI8Wb6Y6Ul&jh<$?e-6J9L0&?p6O=*; zMJ~b!wGF{e+fN9e{pTTsBzq>%n1jRT)qTC-sV>y!$!H}NJM+CL31AEPd}DkKV3Ej| z6p~^5rF@iPk7XHlAAX0|8Hq{xLr8`G@ab&4R}8`Xc5?&0Y7GUiWEgtB+n~ndb)^>U z@-b7FB?6POK-gqbhZD3r-B6CQgmk<=WufpIXcQUJ5AU(j{()2I{6xTWuE}UlY@~Va zi`Cl^_AWAv`^j3t3C>cO?COEZvbd+}z1;Gu4r6dTSEjLSW=?54afvcewdw)STuIiE z150(UA#Y`NR;VZ!NFm&VZEkduG3%dW2`ab4%4qUd_F1D)heQLf6f-q%2VSFUe-|$A z=kO+#7zN(u(X=KJ!@caX3Ko8ZbY{dxO@3qVxrdl0{`UQ;NTtrd@Px1^wS*X}7U@UX z;}{Ar6g8z|ihTL_8Pz?y5f#?@`tu#9Xs;M@GhjJYm(`+R#LvWw^irVw}`LLAtF2U7BS?{LZ`Z>M_jj1h+%QocaXQ?DT5 z1Sin~i%E@siMT;9$9o5t(U$L{5T*pv0xpf06=V?O8$?J4w_5@M$dv`L$x}CU^(-#L zk7S18HCq-qm< z{aEZr(a)Tp7#7aG$7z=2!GV%j@%l=dp=w8^ToyNVM34|N0H~<9+hX{&yC{#{n+}_F z{}q4pdpK`z!CbRdzngi8G)VHgXL4k z9^L6vfS+fwK6m+T{?2-55Wq-inHrm2C)EGg%l0&PzWhMqjP%U{7vg-Ab-%={h=qtn zyD$4WfU+NgJmF7-$mDtQm=Op;a9>H~q%vF2d^5vsR8U;PSJEZ;x|5sj`9Gdia`XuK zDko}}ezJlt(!TNmt)7=hvyOd^z#iW5MdhfH0cN4!@7BGWdLWr^!)_Sf#_(4tLQXZ# zx{Cz)M|d3T zU7|rI-c4EWow6znI`kTILdYD;ES_PrPe6~`uL9I2rFSIntcyJaekH_lCs z*8QmIrFz}^jgHK8$$Nb0YtZd~<;=(vB324n>Tnsa;XUCH8E9ipP{lFKa>f&u?p>4G z$$GJ37H_3Ni4PU|($sFQFADua;2px+}^p#1osk)T!=9= znc<-?)8YcH!V=0+r}TxKxGT||qOaZV#7 z{{nZYViymIa{2f|@RIjHf|KI_E?|iqqBtvbcD5fX%h~>B<3|`rvUgbxhB;E2=~t6- z0_7s3<&;i-xWsX2PpGq~ZF@&%?DjqUYW2xaijd+-tAlK+IN9|U+!-IDRf62RA4>fD z?e*K+yG#Q=Z~%&#Bx1%F#I83K$xLB4XgPCf(rDn|`v^+9lPK0UY@?B39d5URyF3=( ze)ukNVX@}BjJL8K%%ffGk1q^<*W)M|)HvobgNWCfHE#`!%pOc{-{{wGG}q3RjygN>B<-8Q#P-Xw2D}#p z7me3H&rLmNl*QT4ba8r@rf6fGuB}d6Ze6e3%5U`5BnBT0-XNk# zm_F2rIU-DiY{u*SrWgr^`muNIvMvn|Yai9iwaPb`TngXRK5YuZ#=AXy*lH(mCtslW zRj@qT75)CUk{E~bC=Y72rbe}(LJb3=iv)jytHFW~&C>ay6Tmu zp9+q2SZh^J$_9B6`e#Q7`*2-0v>z>J1@7C=WQ=RBdiCamVGXgHOqaDk`F`4W8Vd|G zAI9~|Km_`%6ib=fkl!g1IvCLQ@Lm{2z@m`19jRbcY_(lM&jha45kJj;>M_gVUjR=i z{qk5T-h!yre~qFZW$Mb4Zi%>VncddsuZ z`-GD(d>X7m%4_yHO}zZJl47B0Fg@((e_V+ZEK#JM8s(FP1%BJm65{+?CiRKH?HqfS z8+WGshz7#0GcYjgebIT6ADr=DMZW7$45y70XCcdz(!L~huFOy**(kvI?VpD0$Z1mV zct*lw)y&Fk$|6AD94LJx0Gz6FK5du*}cnak-9p%AkZ z0YC~3mdV`PYK)AZiiZLmHvGG36lDl@NwVd|%CFW&(=`1T(m3T#xm-jl`D`@TPFl%zWXf`)DBoAoIMP_nf@ng&)Og7 zw5F+K1d|qgA~cPYy4offThj8rwiTKcyDEJR zR#ksc*c%}k6yEg5r_X{=HHH@zL3T8eB6M+7` zCC3E+wuO`g3leVURQl!ho?ty8{qMyHe$S~p1l41F#R6g0lPA51UudvoJ%|XHe`J^@ zOFe|U;w9!vC+FfBorbYbO^BuI*}tvIsoi`XoOQ#~wzY-~^3qfDetCl$Av=@*FMKK!xC~5f$`4rofm$_!GJ&k`No}YF@9Sh3HAECnuQ)nOS>q{u` zZ3@Q`y5azZj@UcZ(oyk%ic6GMFXHps+>)PfmL7=R=bC`5yFcLt9lu#^!&V(cB^4jR zqI(G-D`WdnBP?FJE!$;5W(7~1jYQ#7%3$$qr+c-KP5ofv16Wev8xmjH!@S-dig<+= z#VL)@?qN1f-}R&4w`21nH}eJ!%lK1G_oyL~&^7kqQto0tSJ;)9(i8cS>TP2e=!XIK zygKZ$g(Hs)Ox*}}!q>S7#YJ=>n&FgukkS?1x z0tEBB3l&p&d?W9jJ5`?%NKI%O)(5th(P^sVd*9`d^IsCozL_NDE-mN?$+p9e&409* zy|zXlc9S7>EpD1W*(>P~p1g?rQ4X&{iMvmyd4=V2akxEsy+PXb-|@$uW3cb zup}jfsI60pM9Dey-B11Y8L6}Jy$KgegQmbFdjMA7ufdb^U(@wQ=rTWNn2R;L0|^AB zUE5etJSTei2V=4sB_m?z$mu?XSYY|WMr&9xAC(1XsV6J>bofrlJM=VV+7xM$LHFa1 zD}_aIAt4~yWi3*)t3Zpxq38Dy)usue628+Nzo<)*&j)rCe)+|0m6w3@0E~UAUKmp) zL8*;p`d+JlJMRW>i%}gm#pW2wP&qqWI)$BxQGmQi*ikLeKzSFT%^30STRSmVXJ1$A z{F;RHnyWGKPT1oy-}*!1W#epXaLhRza&SlY7`w?3sh{FvYO=kLCug9_)V$-Q*aT-K zn$6RL?S@1;zvuTHMZYf>r#$(M$x_9o9GlFktkan?U6*l~HgU2 z<+vXcg4T#5fn7O#)IYTMT^~hQwCYPeyq{`Sfp#`LQ2PLgO9g zrsH_~)2Gc)%TqgmQpa4hU6neadwP|}d$KzT6OK&i(X2ZRj9WxGZ+3-KOPbdPPfX$T zih`xAce$4o-yC(6nV^{L`fzS8?-D-Q0@Ufeje?I!a5I&^C<3-@Q^~$2XJiL1#||<` zsIqAq-!xJy`9Qtdu8O&t3HPxpZ=S_QOp(Cj@!E3ZaYnv};*4i&ve`@40`~tDHU?sH z25p@uLSwb8hvD~Y85ULMe0~qzTz*6#qJn?0MGwYzaYhxcG=Y zMO92~d6*!cYSQ^gk6DZ3`B|KwS*KDxuc)aHx0{m@Iiw{0x!ckTq)V-ar5TDtz@q=% zbV)zh@I*7(DjZqW z)suGQ4Ru3)=QXMZl)nn1$D8U!o(jafU))!{m3g-{h6?Li*IGhagZZ#djiG9X4!6|e z8H@RH9DaroKX2G_+RJ2~e%=~^-Q*K+g(Z;miv!h56!~?_&p4jVtS{Z$@m}+dMpyCt zVK_~14Aa0#n$?_x8LGTJO`^_T?*6BbSVzH(X&Eq~I9U;Azxx$Lih19V3HH>saX;K0 z?56e6_-EFduO0MLvg*QJTa)K6kIqIsVR!gJj%l=WTgi41m-j}E7_pB{pAiQR$i<)g zcWwbQ>$%VHkq0Xr98*g87NXId(lIDCU?E3;=%M@GI}jZZUMYbfpRdo=qjBBOwhPUo zOBc)^C4cMnP~ovQfma*o7;mE#9>zb&*$^54l6ku2(%f^(IZs=7FRX^Qz z#sV}sUN&r_sTFR58|xT=hH&7w@iZe{yUxDMyVQeDH)!xI-iKH;ntFM`QrJ_(G>hZr zDj~bPT^7bTOFVLx6f#}1W8WNRIzQB0<;NZss{hitzAx~+t^QBfavmLL40){t{$e89 ztmq9MKUn1h`_>8mw)y?2NytXsEFFNFJvLkTqWqIRC6D~-XgPM|CTZjTNg$fQ;#B-o zrmK_k2r}<1RsT~Hvy7uvN_lX6rB7r)EJ!C=YROZZxW!rgt_muDwkdyBAG!-*A)^@l zXD?N?miQ_ggwb;J4bQ@)t|PXlUa+#Z`!Vh`@OlNcm7xiqXANuDt18v9ceBdVONX>w zRi&~YHV22Vrf%_cd(1)zpq_gdueQeC<<+`l9iKE+MKv**+0)UwOiec}7g~ukgn)|T z-kPC^Wc=Plt!;S%xRIPbSk1gMCejF%$GaW3zQ|UdJd3vEe34r9Y4c;JDgrc5OerSYa(etCMzCD77-1A->W zhR1G53qe)6IK6;W(4C&+B8V-xRWQQOagSBeCgi)hCCy0JxaJVWli)aHst*YoL%_wS z2CIHiJ8T#-M8H){_F20W9^?p}e>4q2BCFqy;l!zk+e`+oRZMsRxx=3#NEXzC&hJa;Dn^wCeAEbq*WKwK-$thbpoZ zsb5#*8$2p~*pCgL4XK-?6uDuTd$uQhgZBQR<;t3A3Q&i!G5SWx^a*+0y}f&U)pXL7 z)z>t>h$`n0(Fs`8iKjn~luSWfXkHa)&@a467OZMFC~n@<8`dm0tDLWPqySD zwji`0gC=`KM6A4^cG<~mx=L>1b=gfJz|BU!i7nmaZ{kC8$Kv{%}bg{hXNv zKFiiUGZ|@p_|*-sO7*VbPuVIYaqd=G>E#6_)f-X|5emrDKBqwO@cl3*^ZBwUtJ}XD zXJgi6YvRL6dko?dUuK#O0?SXZ>H06N=c?6JFMm+Ib<68Hw+e(<`n377u(gZ8K0>Hbm*Vwjoh4 zsb`wXF4|GvfF+iHoknXY#)|>Iik;jG3c8>dQDQl&%1rU!Lh7_GH1PyHPP_zrQe$y#Z z)KHsA0?m?ZH^yi7y(F%WKHKerEO=B`)*+*7Qkzs640om}#c!Y1QwJWt6gXWu8DPKm z_yr_k@R(RS@i19S3XxGbXN%pjH|^6QV1tm=t_M7qrE6@JigKM{Z%zu|9(AYQR{N=R zcJeWB;1|s05%3)W@8gZf7{u7YLVS<}!V{{)-1Mq*pc`p~<38xKSt-WyRdDUGam!S=nU z5}DHOX!kT%>gFTjJ z_T{4XyU9b*iKa^9G;O}ktOnnV<+AGwb#qSC?|AG=5DeJ*P|oCo)6G$vBe5+OmoAsp zrdyZ3jp1sH(HN&)9xRx_(P!>v)16hRC5n;8lI0L0D1euK942LN7M#+z&+I#EJy+u% zUZ<6PaRy!Rm@R)Q(q;#9WK4jYxWLFoeP`#nI#<1nIO@@7n*7LooQGUoOvoKK3K)x+>fp^ z_44~W1Xgx(V8}6&O@%+3i0{;@QJWjK=J^6K*;g=74?6GDT!SW5HaVT4y{Stbi3_ZViH(l+J zu@(-A^H@Z}6NFWgH4XaWJHo%{%+@oErlD@UqZiT(wbe-Mw*UAx1_31f!)7_%;XFMfU=JetOL|k@-Y{9lM2S7!Yb7(9wn`l&mHJNu^OwI*kfdCr)LfbSkyliMp zsIRN;p$YTxQ>{n)q9qXvivSHbe}X{ytlKv8(9?>#6CuV~irERWf`>g9>S0BLQ|Fct z=25+Yyy8tL?(JB0Q#33xclZT01A9*TevilViB^wKA1_nfx5UR3OR2FL+B(eilLOZ< zj2!?xIqddXs@fM6Mf!%H{Z5RBFl-`uA_8$-@nwsHm|5Yo>}Y|FX}f@h?I<*Yyxye$ z&ccYib^3s1c+X?P6~JztulzcVJCP47!WJ_PqFnO|JN+bXQ>M?c2*yLud0kI{S!T9b zUKvzz=cm&P%Z3@b@ik=a<{ot3yy0#b4t}u*y;L`7=JUP?`H*Pyc8D4y30cz$P#Jh@ zf)s@0K6m_5p$hxGoJ7Z7iCQOi$FVcJ062cbSupV7$_O3QgC!6MyG=6SVNNv|b(_|HmV0iahi&y{kwPe<|z*gW5L}+)<4cHKb)*&xpHY;y#l8+OR(4 zPS9MB2cmI?YM{-5JtGsZ@ardiK;}3hrjv%y8<;--8;I;s?Ys{7+Hn@D!iwpxfjcHo zCFrjt2%6?PCGV$qE4`a?j@Rnm`gR9Jk*s~T#S``rn9;SiG`*hKYcu#bU`-O`~%T6;t zG5Pi=Tt6-)gdjY3FXSeQPi!7<&Vg+N4WKKsr1A(C^qh>#*z2fjap>^NrNfrae&N_d z^F0>`HfqhD_+?*|0X@$c_>t7ypIqa=RqEUo2g=DgAB4LSIzn3J!$f+{TaXucEWXP$ z9*xcz2lGkK-_AY}KtHS>$T$}3PmP@h8dq*0(m|?)HT7Cqcc9c&7)|waTJq*M-7-im zm7s4H{5^ftdpdzxM5QtWG0y(7mq|cD7_HVa2Jvjt48h@&IY~(BT1aG|X0R=2|76@Q zY|qT!^TN%*mu*gbMmolaTj8b-yIKcxn|h*RJi}BQ4pm8EU!SF|=n>+>XjpKzXOeNL z;zRGFKFrDy0rM>yA5pCAf5ck7ovmzDhKJK7oTQ*Bp!N}8nzw~qY>;r>7X4%)uWX&Q zwnQWdcikn#!4Cj(*-Q{MoVw_M)qnOygzk~N=?CdEd11PME!)62hxV+FGW6U>7Lnu- z*7L*F;l7jBhHlk33&w>BAgvZxi~I~g8Ly<`6&}(#s!NZNSpYqKWaAUP!kWR0)irL# zo|9qPoHgT3;BRJg+A>(lk)%_>MZ73m=3vdnDJ7iMj@u!Qj2?aBrbpU-4X z`*>8u5@I!}f&~=5r(-YRneA0_cm?))2JlK_SoE{ob&6HK5;Hte64$87YG`#qoms1| z75}||;xT(U8X~H&@<5K~iCIORDvjibdPR!o?n3@Lez$VCN2MYFT*62)LNJ@Ky zI7}RkO6+CzR}$~yui5WrVm=4f6zIZdy?$E0$Cr77&d0wP+V1&`!BrAZu(C=rITdfY zw#972Bou4D65ah=#io85Hp!chrn1WVbj#$9uIwINZBHFU` zPtJjs;xzc7VfWPr_cMnto~>g$;5JvjxGaDYFaU<>X%MbtdUX8V_B^k%pYN9)?Fqd~ z_N|lAx6LuFR0;MTIx7{b0LoQJ+8hLPD$y6&qK&{ppbE$toBVW-y}=dUL~hvIt!BjL zE86Ci^(R7PllbiPlU&xh*w^i|tw`=8njs5Xn#D(6;hx*0tUpXWGh{Jgy$|U|kP%^R zj5OKs?qn`QCVnn(kH_AE$&jAI37pu~R6{d+gp%K8;?iY3I!3p_3@m!;d+na}Eajkk zZ@4u8XaAx!U+Y(z z%r;0S99-}@S3B{Re^k#DDeOp(>=0A31ZrF?-pfmXp1=sI9AJpgO?dW2=fEwyeTp3I zr=p34WttrwbtxyT%@ZmB7}b5suE@zUg?2#@`Q6#OWZHI-6bE6RI_f(2H#6=nQhmsr z?0*|fW< zQZ?rW!9V6b>Syn^VdS`M*VB!+*t=%$zR2oz1d}hda`H{}WFSDLx+P<754ZgeD`st- zdI??@#;h9;U7W+2hGFKQsiiVgf+W z4ad5rYM*N8C6HCR{f6bz8ok_Z!&O94DQOpyk(Ip)^Q-X*HZMghlKvX{8#^CKtHx>! zpN+W)E+OQ3A;Uw>XJ7Z@9m$(kBe-xxFL%T}Vh(z-Wi(v6lA_YKxw$pdDm$k_ej4k_e+kLqwG~ti?~n#nccU<+;?2Z4KPi|wYY+q$ob7Ncm3|z+%lVB z049(S0;=3P4W|!CFyeo$X@Oc9spn46dWXVj@a&(D@wBjV_UnEgia7J-K3`4A*S;F= zao1v)kgYqlp}^7C#(gHg@*qs)R3T#aAIw#ez+#^`wG zZ5@Zz6YVcYA-9&TjB~3sAmG;#Eu`GBo|6);H|2^^L-0}3>(nH$3cIWbQdc%soSU)_ zeg0XW`~2W=BM_CD&QJHLZfSvd4}NJ=D-3+;K>L7Z(_9^X>9^|K;Tulw<*SuL(&LKZR3B^(> za-2UpCp_LR?M8a-&V zI)BdNa@R#?rtU(>rtvt&mys|*Hc41>q4CDFG%L@fpmmk31OWe%Xl2j83fC7~tLb?} z5R#_g)3jfhp{eT-c<}5}eE&>6$&jm%`%Mkxan7vgPO<_WRIGUQBgx~aMC~&?l=N2F zrN_R@UI9xDVEqCni&qld1(uR{A=hc%iILl4Ops30I#c0K(=NTK)4vvAAl+lE(X z^rRq$1Jc}}?seM2P7+ew?d)wE#I@=rr>U-tVVy$sg3~4pavu4`q-O)2Sa3S4_LV22 zu7Kl+wxIo7-a1i(5aG5@;y3F-ulHv#mxw=|>_< zauO8i#}&}m29%}H&78s-(PBYvg$#27?WVJUFOlCkp!!C=5hEMY;2$vYdcN%qeJr?= zu$ww7YI79EkwJnTf-boBFtWv`^;7!Lv$0jDdL37~I-5GokWene73p2D1FBPlyWIiS zv2BP~zeQ}!&y%e0o9c%qvlC~u*EW-+AzA?sj`}?%fA7+*pD3TzGs=nq5J-s7AuUX$ zwy8Tf9B;dXgh%ng8VR&)((ZE$?z_paPlM4-=&IkwC|pmLoCcYz>!(V>o8>lp9v*&{ zG&tv_6z4O1B-FFaAGa9kXIdvMbV{7r*RNz;D^E&!QFu8Fq0k-4pS<^)VQ8Y9+0WY} z`dGY80y1~JcUg$1_VKvFWaeiiw=rE4RaEf2gtDZe3`^O0!vOxH6##LZ7JF_>&v{xv z_KlbR5`tfNNgMa8EJNb#^;iBp4jqO7IM^;Xxg#}U>#>J08 zb3U=RRmP2GmcJRZdyu`b@6O;Tn}iQfgNWcJM7y)Ljp&P}_d3MsPyMYNV}wo#X_kEz z{+HXa{zbiUxHB#JF>!O{Je+%Cg3|mvPF%c{T z0aSIjL zy*Dw4#;TwOp-zNie{_$~0sU98n$w++*?ScReaYJOG7n+43_Km|SDpLI-qZ`^yhYDw z*rVDy>RS$w-2iS}1Ev|Y1wf5>GkkLhLgC=ZxCMZ|hpT}+dFmzvDak@}u8&FY0p}6L zTbKe72)4`>0dtgp=~|GW$DQuOx$Am=>5t&=WkkpPZ+@-JhBdXZMp5R|idGRGf@NO| z((O*6X@I4$}0p}WK{lF*yM+=5Otf-Yl{n>6In ziB0`RsBk*a9xYoxg*w6Wu+$)C1Rq_d55_OhP=|lBJwYS32-=kmw+vMtKZd|JqHIqr zt!V9t;Ztz6=Vqa^Ax~n#-2<6DKh@3vgKic1)-Y{3Hqu4ki4_p_dv-&NAD<#(ShS(y zU|`qr0E@+!abO33O=-PiIEYmwBBw%6&Apv;lp8ly!MXeE^aBejIw6Yp_jsoSD_=n* zeX`xajNfTA#7wnBo>Z?^G}o|eY)2~7Tx-8+5mg{~OcZtvY9-x`k8lT)ZN=^G#h)+p zJ!A*!tEdR7d`?-t_x1y?^Ez}y4y9NTcRc8Di8U$EWZ{==?|aIsxST4%$TMoN6fadFUuze{QE}F5_Q1@rqd^7O;d{fL2bLs?z zLau93r8h&iu@e2#(Gd@j2f+ZWbf5)0I~ZValgxD>Egi)$gXcBkYziEBprMG?G5o-y z?Pp|OZ`!d|hS-62_vix`76D1tEia%oXMNRKa?u?03{bmJ2+wIi{}8fdj#j&P_L4Jw z>w9}mwRzULneX|xvGUuzAImhmW%{v$7tJ39BiiuGChdQlW_yvDs}JMaRed_oCCS;H z{4{@q<3Y<};)?|XS|r_;bRMp^z5x?V;IrKJwWACEFIcd%y_dd_!fgKs2nMxm(4%kT zuJDstP!bA^?vs{XI%(CamoFxuJ=6$*7Qj}drc}Z zpVQr2`Fi{L64*9!nP}HT3g&px0ZZnLQzO za5c{p_D^sVG!%m0H(8cVe@{0@b$NGg@BUTeuFAt0N82-0WltM}x zgrS}(CbN@D7a80Ng%R8Q-2imtYX96c2=TZur|s}ekaQCx_iUh<{IOj$jkxI=_p7|t zhBMk1aB2aWe*hRg{w8InBiat-2ocK>_3@bMS(fVmh8wHL-CvHp4nEk}D}V$skhEKQ zH4>Sr!Ek=vk6riFE8d7rdhQ6RtEOFOkJYRAD&BJa!ZdV49DL<<4ybX&8B>czGPo_b z)K+5U-~kYgsguhR9ljS}o`5`5a zGW<7xYrSU(O}eM3fp=PkT?)8`A!9Z}>C6GezYlmsKi(TW&i+N;&NYLHvpd)?8++`u zC9cgN{3=a$HQGba=mqwfI(UY<10>|NF{vLEK&D5oj-6CQ-hSG3gdbV(eFAJ((fkM~ z;XPy$QfeoJ&aY6br*|^>cJN=0*PHmJXzj%38NNMoaN<~eGhO43J|A!E(T0Mtl#Gcv z4`>R!?kfEz@~nyflQRz?%alYWcHX?2idnuS@$ zY?^vXe=@8pf$UG36YD;|o5;!mG?|87XEL-LDW2RLCJvT14;`N?CpcIiG}`g>4g#ca zaw~AN18{Q!3_R~;@U^`|LegtVj>E~j0VNnEKw4&Xqk36H2Ia8;MR1V^qq|tAgB69n zt1;LZ&P>hC<>%v>1Za9}&EKq@g^^3Ncc#BnkcohqqhY&CN~Z%f%#2ISK|-MP6Z1kY zbVgb80r%{W9;o9EyUzE5@3*+x7SUV^@;^>AX`B%96pk1>y`Jq;a9*e@0Ng$Zlg9bs zHwsZ)?vgD3&Mhk^7*>pXc0}qzuJs|>hNe( zk7}19aq%crM|ZyH`wiX8z@1@{o9wdlH>nf$_Ei}DAP0DeRBWsvsrBp5H?h`YxyAj-+Y`^^kLFE51>srS6!e=?V8XO><2h z`WuTSAm^A$aD3PT$sw2O8jxFaO&jiM7qTg@FrrQh?;_J-^czyLZ5KL|*KRc{D@orf z^RCM=OT}9c@h;+n5kksNwy(n;eT;Mfl9n@KXuO#(G(|PzCi_U6(6ayAqVEZ!*u`3T zbNzOE{$itz4gJI~is{)O&>27v(G>MY+amu~A^}qIGr#QIC^SZ*LacQ^4L)IXoWuY0 zbE}Sp#GX}8OiAExO5fZ@LNsxpfOQZkg@>@hI5?Zn(G0t>@oLJsx%N~lLiP8;nhn1H zAPHsCTjs;Ceu@7w$8=37Nd?O4@~Q{2!`u!>qzXN4nP$W4dUw+7c+q*RM6&ztpq~V| ze7Of*)f5EpVZR$nejl(ve2SNstZo5~L%r~yCj0U3AA9dK4#xo*LP^C4I*b8~A)=`a z6dzlJ(QQe7!&2@uuMMLho~E_uJJL30;U496dsj;S3^PS&fjv=5=C}6d8 zKE~uK^xX6DmjhpXn?~9`Qfv8C@yoy3{HbHJCHM~1^EoE$c)J(_CE61Ft>;_q{GiZp zzV`DDYN})bl{eCjJ^|d!C;J3~gj-ZP9mf*@RIvseO@?0QEfa9u=|i z&T)oiI*Nyh-Np!fME#%XRvH=@mP+HND$M0{DO?_V(g{rx!hrf+rOSGZLjicSyZU2# zLMZ6R`yQ6Z622GH_EveJNC&CD)epV#JDyZtQL zPalvn3ABwX;o^lT&H03vI7QouhqP$7=lys!0(hq_U{OHxD;@+`)K{ZcgI1d*9}o_q zH_x@y&|cP<)Xx_pkBLS4pZHSC{8O0j2v7jZz-70df=wnb^S;wHw0q_~eTFUjz?s46 zINF8RB)5^Ktqu>+I;E9rH~E63?C4t5!G25U~A%eAC5eIx-%FndGl3aGMBT>N0JL5Xx79#TSn(_N0!pAVdxFUd-#}v|*k7!2VZC9-k$UC+O$Y!cX^2j1X*!O;4{w$= zB>UnSB{!wz&JWgJ!q*%>@)OL(w%*=EWQw>Oh8F@ox{Sdg2l=(L#3KL6Q#U!zzzqs| zvY+DY62FK!hMX2$mt@cIUUqlZb0=JMEs8;ifwF8yx;&dokR%Amqfke`EP?^v)nvBl zq#x$VC1>?F!sPHj3$LL>nt7gMn;K(XX4#C-LlYtz`U;V>ptotxd5=Cnb}sbMl=}dy zLjtn8csV9fjBLI=KiWwD;V$pgwgX=)|A>JePWNQAQ}xEF4kK&+tOEfrv_<|dfU3Gl zva|~z?|P^W-(+H77(yaXW1syL=FZ9JpXbLjgL|5&aRV01jJig)->R=@ACS&oe4Cv7lc;7Z0u@TYo488 zFaa5n6d(1!{O%E;v21?UTYbUb)BvbO$LdEdp4L{d|SlCAJe zn0_6RmkW{N15}=(Q(R%ng}IJWKrENA|9ITz*K}~J*C(U1>A6(K#7_pl`LA%inEoM) z^uGx1XAGJEk3iV>EMg(JX~ssKo?EC$Wu28-=-^XjDe*`KKY2XuPM>9qjJu1>)S1e7b7jNRnHi`8x6Rj)&<>em3 zRe^tdQvVPR__u#3h4mOvw|=N6$er}Zp8rD<=HIs2pKlWw0SWD&)i1$n{{Zp-pOu5- zf-nHtY3ZZ^73M$66aPhqK5#`5koI=@s%J|4@0;U4-AchDz(?w6S|4ivDQffQCI9bV zJQBXs+@^V#G%5GrJnpd^m@;HUhVI#Hd0 z;$PjRfS*@e{OIk=YRAVx^H(h{PAUO#vZ6}5#MnxI)qt1??kq^^b>pAC&Hv_ZfVmF@ zfSEFtehB@uxcOha$9*CUIlz59;`;}7;SU4;k1i{Y>CQ(Hjd#lY7y0&oa-gV>onC|@ zax^kAbh#Y=gaQB8i&N}I<%4--*yt<3XZb$?{``OMp8n^9`N0|Zn$;KP-^b}#nf3o- z=0gHc;4-Vm)`-rTdV5lDUgAI9!hiCncb`is-;MNOo|LHk%Zrt8|L$z~Oa#XKj}Q9a zRw6FpZY06aZ$t0?>H`dXEd?wtU$LC&U!G1<@cQm-aM?Ya`O9Ym<=v*kc^aScZ`b>u z-?F6vJNuLICB);eTAd==-PsVVyu$o{T84jKt^d0u|6eSL&#MoTgR_srZOY?o(PO<+ z%bLGB_Qz}Om71B>OmM*W?iK!9O!z-dUXka4)@^oz?yV@N)@>GPlaqQL*H!+YA8*^> z5HaA&nV13rHs2Oiyzb=Bt}@3@*!o3@ktL(7ukUNGG~LDey96z*8~lCwbL~Sc0Sy-< ztsOl*jQ?f#{@{P0MLwufdf2L18W6)CeQ7pyU@Z?=qEeicu3eMUz^)5?Mpc=lRS?ru zYKi-<1+w+lwR}Opb9C0g*ULX8b9TNd?H|bH|763Mii@SL3QXfo^hC%gF+dboN}It6 z{@wln=02UM?3}1741gt;K0$JVLz)n1toa<`#W=7lYU?05S;+Ag~U6E%=|V@9&?-$zc{;oE8`#uS@88 z56)-FmBn?j1fsGKFyGF{8F3XrnFEJ1#54deobUYw_^XO3d7`DIweOpJHXtEvEPF*{xu+B=r48y+ zYPnMSl2w(7sq#o-*4=k4tLl@N;NRRuV1fGkp@IU_U=GQ@pTA7%4=_Of$MimWaLDn| z4FLP`CC?NJ@Q3k{0|!O|b>Ne8nA$_i6R~zXpe(>3-V9 zXJVf@CeRnnIE4Bz1wh3B`RL@uqPrOr%_kd78oE!pZAS|BnPK&QDQ=rHMk9qSN()rN zihKRp0}VsF?LQNXUSD@}ihwt@yO?FeLa@c1mpUsCT*^FLjy9Y(r&|(rpO|$!REw*T zV+I2?zrJJ#QNPunDMAe^zrdBC`GE&p`ziT7zDhZzSEUl#mENf7s$xKr~tYS@d*@6k|vD)1xUE{T;%uT-}pM)t8QM_8C{xwZk2ja^sY zsNc;U5I*cNxHe%o-tadgs+>7!f1$G4icQW``F(xmw%C0a7PNBF#_d-lH3cKKMq+j!8}$cze_Q2%DVyxVqQt(m@+? z)&X@F!bZoV`P5?OyQ*@5lnrH8SyP3&S3*;uIJDT6-w;o!#|~?q^Suw0%ihGa7~{V9 zaejr9M0Z)HUrtkG^FC|6TkX~cp}Oc;yzu-c?q-B#7gOmcnZqZ6PhS%J2DCkDUdYuT z-qk-g>P?OyVD>SnC3To#;(ThT7)f$_C_33Rn030)bhsdj=tk`C56-3W+YI`y_j~xS zg9qOT1LCaSgj<+fDq<(W-j2&RK!%>v%X@761km5T$kA|wvgywarF;EKdRaeTeds5Q z9BBC7_+k$jplH)7VK;nxb*Nzgw3XCTUd;FOC9{#bQ&(axwVe9Q4>hIlO;t<2p1;6# zS^Uwkm34~VbaV05#i;Fpw&r!~N#Pj!&W@jMA+*1~R{ELByFBgYwSHB-bGd|l?uIbE z8&_eR%&bzn9?08z+p?l+ejU5Ll# zT2v?qeX{lS@SM^eC228V+=ytN7st{T0hL9ijf~QJ5|9%W;23hAR?H^XSihT-5G&C= z<$;O&c%h2ny1=)I4^!VoDehiHZn`_&_^GWMk4c?*;i92SEnN#*@72RdccS&1h2*si72xMS}f ztK#KmEc1u&j?+dRyu5s0aC=E@=7}I_QCs6~JxNbfOV|V+6YG4AFQ5GaC3mBgV}Fv0 z7A$$3fxA6LMBkQ9oi|=xJ@5bW1qp7~+jdYoQO8jL}Y!*v*ka95* z&R{6DOvY(B#DnX2eup8xOy)d3b=xy?jK3)waOKM4l)w-^6Y)U>4H6LvHb5xLb0wOiB|y9Y3a>V$mSu zAkLGW0f_y%(MZW5KSR$8xF~A)M+{{CCn&y1hXlJ=|bT)6KyUtkB(tu~86HNp)&dpvOe1CS)i9j7jZRIMTl zX1@c|s=}t!vwWW1cMCmv6#Qx_oV+EYcvHbdT^Zht6t{MNeSfacNdwzve?g;tH7&F6 zyE}qC($vQUQ6qglb0Bv)!LUwixed`L`}G^1|IOTc-hFZ-=BY=Af1QtG-Li8T<8Zxr z%EWAYLzCwMxX0n(SKp|Z1y4|#AHod_*4rDqJyI^Uk?OkpZL*OW{v88#XMMdBGaogc z;;L@ib*(SXBHx|%y^wz;oyqDsX5#8%8$B}_KpRWnj|sxmd84-Ww@Lox>T8pf_`n04bC9ccd9Ym6szk|`X1I^9t|(mqek zpJXtUqUho($UVxYmKC^oR^Hn4LQr@)J4!j2vKq#?JMb6`M1n#bs4bpIN752{!OEYO*gN&80hM zSU45sE0bAQ(SN%9 zr8ms2n>5zA&5)H>xZQ-IGd~B`EGMg!lr}sIP#f?ky4t~(mRBQ6vl!RSN2Zmk8cc3fk z2{=|&dm);XOcYdwM=!?60L&C|&(T`V!x=y7du0uVwv`i-<>-J!(SpPgKYT^sI)ixlyGvJhH6Ew~l9g{MVvBIL8jr zC7f^41IKg6Y8qL}IkT_l#*|ts0P(29>!kNT`f###yICabviT*Yx0JLp)GcDQNhwm` zgEh+^)$!-l7?}aDGkQyip|l?}8X8&>3$VbB*f38rE)IZ^pa!S8#IcT*+LNPh(X3LZ`^R%EbPdj4h)ya%!f>%>^TN&&ztwG z1P~~p$Z53oVOH5O=vj#W&A9{+Eejn;l73_~s^30N?TYUfaC^5S>~>~D`cj;y`}MHT zN2a$=o;`f8cn|-i5<0_uTe4vK5HtWlWR4pPxU75wZn{zqu1_W3R#JH_wJo+)-|R+v z3wNt|c#B$aw2r@onufe+kz+%deOJ*WlG_Hv1e?zlZ*=_WS4;9FZc^v@`726PY!ZuZ$HwsAZ`zY(--a%3y(#oJ$!%ZiQ1M+{ zlzNr$7Er_x_GoD^g6b_twd)8pm&(4tg28`cQ?2etJ`9ofvh_QimzLi6HK%OcTOJas z=q^0L3)Ol9`Awp??7nPWAah=0XaDk)fXf5p+}SD&Gen@s!uU8=FV)g(pH|Y}IJ~3V zJw$#|kxXf9;_77TWn z{w$P1%daI6iRrX#81w#Zf*ll*KL}J$0{;+2iYb$d0Guh89H^ii1j?u%-w5W79pz* zMF8^ReHa2PqG2s>X9{xN5M&rOG;A@uq_q*0 zQmvmPsBjIhqx4Yz#9{RET~4#>U*M*I)5T^_G@3b1*O)lr>VEZVHFjSmeM2nX-#jQy)?!*d>)8W>PlFN=!tv!6No;Fu})%_^8&lS>RdLgYHe(UoNftt|`q~G;g zn^S2+n@{t#RueTOh%ss#Cw_l;JXbrh0?!OO_HZA%TFXGJoJsb$M}VSivQnQYREvRj zT^1pJr3;kEOtloQ7Agg5#G2=ZMZ%6-u|l!zy9p>G_mGcXWxKoW!X`3%lT76IAogjp z0u>sB34-@9@GVC~@;%uxCI~q-wx^?0Hm~nbEfDs+di$69u#D<`+xud8$3Il@Ax$WC2w9^S9?fpjc?Zv|$(bXZdx zzsaWD_?<|Cc-Thk{O6q1M|uQ`o=f^akv%=o7IQq;5bw1nR6QetGzX*pVJwwR$9F;%aIcYTy-?B7M|Ew*nzUE4?B z6Y~>DzmFZ`CIu!sF2mslnPQe^iM0ZrtVxK!naUJr_{j`XJ;Uuli+o!wNJ zc+)K>o23Z63tFY1r(S4>`CXRp%v8(Xyxgt)Mh})N;txRXT@RzvV~9LlYy`cuQnoU- zjf&iRqv#{$;3@?+ye;i6fXaN6UYc5>l=K65*?cE%uRs{{^n25ck!wSf4kF7eYRZe} z5F;>>{(M)+0C~_iBTCc1oqe(gWw22{_rCS_Qym?v2QcjOEX*qCdBf8j1I{cbkPO!p zaRofu1R4ifGye+cki>;zgD4zgp5@?ZJy^2!3V_i8lQ^_nq^NYz@z{pW4*6gIKHfTM z-qWVimj9}5&N9^bNl2Q7j_k@u%U{^tpA4VPO5*@XKzZHMW8IslRJQz7uwV-YO@*ph zaOylEfM{(h2P#kl(3H<~=k+Cc7T!j##VSp5(W(v;s}6g;-*2&ii&6hh#RDNpLFL2v z?4l^oJ~rr{=eNOZ;Z#1VR1NMdxb1H{Y6&ef90*cODQ>?y?tIm${9q~o+I!ZtJ^21d z^uxCya#l=KTJXS8ljy8Sp7us1>fi`83&Q%P^6`;!A4$%NwByv|=GG`OCe`IMgLRgW z8Q`J%0YppWnZUA5y4deHNqZ;QXD^%N*-JG2I@V1Xp>_4vv+0GV+1v%xMA@) z+_=_=%D)ZZ&=Gh(|0p;dZA5_xx*rtg+T=8>eXM!I%JlA8US9lfvI-fZ%o{j5_A}H| z`sc=`Mi669sjAc9?y3dxX5~#Em8r;Mh@U!uXJj3x&E+7rUwrjG4D{^1tkhJE^OrKI z=;{{UZXqfHAxayoU; z2d+``g!Tjg8L@L^c81fap)Wt$XxoZ%s5=e;XQf3w4-(kwp>&htN5)5gV@celr=0F_u1tPOJTX!QO>iP(Fy?z;LYkM zf#+HQ`(hobryEb+cY`RhWhvx;qzdDObb}@e;(;2f%J_WwR+{tLP3$FJg{PMZ>x4a= zny`lB*&kW}4EdgH!{n1fBElBN1`Ye!`y5gv7!jvd*|bEyCnCVq@a++aA&m8A)faOV zyp~__zSz_Ep^)||nK5Su#g5`Ix-qq#t#t2PF9|zkU6?zJKpc9{!#;}xWrBH3RF3op z`%2WNEEa6>E&>)EmRRh|JjaS6d*wRQ25;M5j}v?^I$4_1DJgIx$KiRUP8ln;fx%UyI0bsU_qK2U3X-@ZOVp3CUzBf1UnYs0on6Qlx%F zE1ZSJp~KN>X%|eJIrgfXouTN(Hk*JTYWzJ~5i7CrCcmHP`bL-|9?5Sd$ zLjc<63_3-%8g#AI6G!Y1M&X&)!~LUUxSm6u$hI3&{RWth#c7Z2;8$sU)>|reUys*i z_Zk{mQ<_`rju4lsuL{-GDad}<5`6bLqAIQw@Qs`bi=p+3t%KR_9IvIN=pr7TK%Z;; zS=l;3y~1<7*@A#;kV(JamJ*m)AW=OHwY~g&NTGy#zKXXu-xZJP0Scff>ZPaN8M1^M z@;U6Q7@4*`M%%dE#HZ4&)=gs&>r}ma8HC5<>sH)lL(8ioZlLn|gProkD;&=SG+op7 z4ENpm2;vB1VBy9seNkDU)~*W+82Zu5AiI(b3B;(&*_~(i`|dfZlp%T3_|cyFZF6;s zw#oK5H@9ESu>KP^xAF$T{oJc2t*tDR z4}<8=unnt#b}e_R3@KqX9#=EUk?sS6gD8 z7U5rCws;45Q~tcbp<)6buH#q4tOAEYxQK3jrk$Dg`trswKa6)#Yd<%P2BdoMs%a`O(Cyk!Wj~6EPuUQ-*uNL=>SGe zqdFL$&iSSEXA4iit3DE(BAYdz9eV=|A{Alu$A(phJQoU2&~6txxm65)Gh>Bi`W#eg zLU+;@>eitfe6Gt}n5$)Zoo*d;)cyOOe*R0wUE3h*#*(&L!Ut#&yr{Om0XF|;0n5g6 z*=&rg`SX(OHZ<-CdR9LV$vg^ws~v)>}qJ-FI#OiYN_3_fRTQ0@5&m zG)PGaf|N+N)C`Tl08*j|2n;RV-N+1~baxFcIdtRy=Xqb(dH=koV8sU8dcpq{9;eC)a^TgcV zHU)m=Pu0XVME8+UiEh|1hbajWo=B!|`C{s}Istb%KAjdJ>x(N-9=VU^={g%1#40ps zoUD70iB;DY&!IpQ?L5z#a*o)1>hQNc5)34sIWmS8xZe#+6z}cgStYdlJn+Cp=tq6! z4I7|iBMXns9UfHQ#P@QraR~hhof6l#=?DwkKEnwZYU=c~iIV@jAJ6VI)d8!S8rMIB zmMn|N>qEC?1ZAZ7qAVnCI`L)*$lK2LZn$Y!!DlR~OyOT`uEQwxm%|TAqdlc%E}>f) zq+3f`*a!6BI1ISyHQuRokuuq0kxynrUWS&?i0JAQCDCC0H*r~KzxC?dS|Mkh8Hv{Q z$8iY_uFu2Bp&#YmDFxq#Y@~mE_FYZ_CSS;A6@bYJequ(5P6FCMVgteMTm|MwHdc9#k1Xx=x&F8!>yE5e4)fs?51r1siPT9`1>LYEX@$@l zox{UJwn0d0ZteI7#le%XiqC68Tv2Ef!-MW4N~$7^V{_L*R5ORZ#v{R0GN{x_HqC>5 zi73gvYZsxYta*|y&7{H>*wK1_iGmH}${ZuxS^lKH)Woz_7WURkZ{d8Bt8AYncu)kA zmsiwi=Jf(qyYBs=oVE-TO{^n*d;aD0xG!;^YMkxd%kO%M|aB7pZYUF>-9oZWNTh?5R$8On<;kd{7s%9Y-u=sM!lf+i0IcI!U5+y*IvguRBL94ZU@YCw z)RzS$crZW^Y>+A!E#hbP$A2Ckb5Jl@f-lPUFb`v_8!?H=f?KxkJd;F-`ZcD>R_}e6 zw}j@&Z4P|nN_^LF{oZ&z-TzuEui^K4%J(v|>;r!%LPi`oF-^woc{zj00_Jq#==SO` zQOCqDINa*VgRPT}N&{7Uz`z4+9etDaUYIdH}` z&^<1}8{!95M-c+?tG@m-7Ppn$uXP{R{O}R(Fkdld+=hwB9^O!L^_b{+{ot{$b!aTT z_B~fid7(2NlyMzQvh}{<>xI-ftlj-1A1sW#HIc`J+s(vVWg|=)ATany?H}-nr1yQgvh}`X2C|U}hC3Z}g1NFkIK4H^znzRcta{;E;yXJSJO(hnLsh59kO@yOsiKC||nF~?C2g>aOO})w;qp1RRPe3(XOb&i z+%tDtxKz8M$cM#2k);vv0Hd^tC5i>ULZ{l`tAUrc?6J2_J^iCLgkq>+nF(JZ#_&9nNSU(S&=lQqHXcr#eo0%>Oo+kXS;HZwb#=n>-l@WRwszfiB_AshU@cgi;QR%}DsHKs{>o2P(@3)7Mydda@ee5?!=}1@J`h%|2#yz(9 zng*s6zfPaQXBSY!3@RQnlb_a?O1bH`s19QnaXz)_a9~cD1Yqt2z0i$NCGG_ni!h20 zBHKi4d?G-_bj2rvrf=ren$>uTDonx*)=w6{sd;5gjXLp&+I}#|vEf3JKUo6a&nYv0 zb5)|t*q};;7kDNB>7{P1)%E%$Pm7T{>l@X4E8MZ{w~)L+b2_a_^hiriVTwQP5E`~M z*3jg?jzoDKAjMct*cSL(4_aSWx@WXMo^#_~eJ**+*B!hHyh%n(22XG3x{u!_3Y=}5 zzew%kr$-$Lr8)CoC97=th?kB7Cv#1Ydd_@_pl`(0n^B?43hT*FjB?}E(nPx$veIT! zfp5Tq`A`EP3Tei)d8}=Mw^QG}WK!iiN(J*ghBCe;V0<_XyMxu_`2T#)JY$$=`K|Fl ze52ba8K7B(o@@y(dy1mzuXet7=&VqJooUNk^wB=`_?b6p_!h+53r%$tjch}-7h%CT z->oN0#ks{g=3?w_D@Wh*w+eI-GCgJ5StzNiY6&InYK}ZyhHpkX|Egt_utQTVnu1(S zQ$@8rY)nh)Q0c1_trgO^1w`=~Q7j=*=o`|7Wg2Pvw_eslBysd`&t842!B@lP_ja8< zwk8v|YV1?g*Di*$^&Yl)9g7Ts#)E!@cTeqwx~{xKU7lRFfLVnOPwng%yj62KUzOJ) z#8p(*!~p!jP4ZW+lj{spwuh6y-*4T|`_m!Nn)Z_@vH~bIHyOwOxtNI)EK7D z-#3xR<=f~U5J)(D>r>{*>8E&cHi9(y0{D1EzrU8uV=bqReC#iwuz%`k^a5wx2{97| z+(fK9c+nCQt}VMLWuV2hte}nIisX4``5bI?U1(Zjxq2PJ5A>*0)r8heRG@yr^-F`l zAQVg8Df=nao+<)>gEB|QKR5AB(RGzv<=tFKb5E36pgZp%jDM5k5058>u6g-DxWGGzW@P~<)X7!+~AH>EG5BL@y%4r5HJPP4st z6re&GaoxgA*8|t;jW&d8jmkPr=-ZP{-tF=n%oDiRL%OI}Z8ZjMV^g!)LAB6d9~%=* zQ@T@B^Vq?O5nu9pt2lF5fLOABfhd@yv&} zdO8YzOhlW@@&-(^G69Q0es?pXip50x4zpwyZB?6f=)6%4H@h1|enumY<{L~^UgKjK zZGYd+C2%Y}P>taZ3lrND$yOwuKzFOt@y^a7gb^!mfosKCqNS>-3dTS5pCZ1dzEcgT zNIW`Ns~E-0`d=}=m>+1tD~WbQdT6&Fw)CX5zNk?;Zo1>&CU_GHigmnNh#gP|)gX{e zwN;CgsfaNQH?Qmdy+;3~+xxW|<_s`qy3vuz#g1YsUY56XM=LOUasA~5;Q*5#MySdkXRgN_1bi+ci{FbA(6IyM_NQEX9G1lq%^w$S6^73K+rRlOzj^44!>@Q3a8TjdLQ451qMFVjak6LY#4zK*B>)Z~L5--vh|M#s??% z_L{INgrpGBC#Zox<~Db!;UF@orOf5X6V*c6x?7%sxmBXLGpiOXY=S#cYM9l@>c*hF zLD8D-pPLSzq>s+SxCjno?Lhcr|A)L%2Jr-29vGn9{gw|Ba=QFRoltt+iirSMa(v4 zemmb^6#U_-KRv3~zkN1xJ4TVF>h_n0V?XDh&R^yV_1D0UuH%luLm@-e7{ORU`1hk$ z?`ZF!ZUM7SqoNa4F?q_Sd$9=Vb+&O*_dulRG@nQu4bdP?BCW9F*3pY%-rS9r+XL+Y zl#mU8+UmX=`$Hu0e)AB*)?t`7bDE9c!zpYLP*+@CI!)Gw#>AL8+gAm>G?are&sv7- zoqH0LICZszZg3+JvsN?>^CW5;k9{QiEQ)P#llZI~PGnEIPn{Mk91+A3&J*9Zf03YP z-C6lyDI95xz-3}wa)@5|FKeY=Tp$$39{97F9zof#vyt~}(6JE^xRHjDxXuokPy?%% z<{?r%3-)X_Lhp@P_<+H9q3&m^e!9?9-Y*YOvl)bZ@*(XKrlKf6hnrdYH4UK&ez+{( zx`$7WhzB)-l<)YXt!{aGr-s(t4$kOmqxJMv2LuA&=P}}-c{TU@+pb#dhTnnqf~J@AJIbSu6vB7blFolB?<&od3M2} z<1XsvAn_9R%Oc9z;CrFoT4v8za(A*JF7ks7Os3q3Oena!?hv8y@u$upgnW)}(Z(@Z zXmlmQv=FX@z${1NiSIT={2UJ1HyD(_w+UIr0@7^;M_0OV2uiKiIVxtM10=6ZW<=I| zDBtY9vU>OoHVY(t2=L6(WRDF5kI+#_V8Y=9Li*C zaY{^bAMV(%2FNb;9KP)9J54z!oW5*P7n!?0YnWP1^3`=h*H*7u@hID( zeQst`j_XUGY(Obb%K@06dqKWPuKeLxEQsupw6YCv>huNf!U9UE*~wBWmlWy(x()Oj`Nc zXp$-wlV^y|6wl$^ny%vqcUq>H|Ci*lkbuF2HQ+6N*Sp6jn}XcN%WPr;NG@I!t@>ec za1qW9iX>BRHLx!RO4lJ6!l6(aW(T#iC}-SdT=-KveWr&ywi$J25YoI#5;|oUs##Mfa0E)p`_p+Jz! z0JKW>`I|PY*Ds4v;!tIj4jm#;xJbhWqkAfM4%fFzGji&BN|ppLTh7+K$6Y&xQ9pP6 zK^*Uc7#%yr9tKb2Y%xOQ%yHSAl#x zWBf$vh-ok)0ABb6pU+)Q9!DYRhPw;I|G1;o#P&v|$0wmC1-t6rgBV<5u@7G{`C-;v z;^uGx!lK|zKeJ8lPz{~9o$*lBV*&8?e2DW#Q>%dVEXwz+lt{*Wnoo|sh{OvA-s$3G z@1LUaVzV!tIJWB=x}cHvj7?Ngu(>g%D268dg_*c?l3!@b%)S=W?ljrXOS-3y^_JoF zhX$k1t5a7(gaz#_KO+NCXfji2VL?dA+(z+ze58w17I6E1Mq3YHwC z?cVp%?%SQA%tFQ;`^`E|It!;X%Y6I~TxnSO2{V^j zwq{O8#tJIm!kEb%MAi!Yo3>KTQc}Vh)X!s1nl1h)DnB);%&Fe@%^fY|m@TB_dN-4! za=8&`j#^zea^f#~(=rzKB+08ld*s&r_``nd&8x<068Wj9v4Xp^F?uZMI;VwV6hbpa zLd1YQ7}oXDWOaM~^TNt({G-q3hzWUT29i)`w^u8v-lVOL=01S*m}R-HWq{%l|B82t zBcHl0nm(HKk=~}P;Ct3N5hO4uoR4D)gVetXYD9C;wASH!17w&WoUXhVG;?R;{%hm! zpKK1lJv@>-@1p5Fp}$X;pDnat&aD!qfE)SHw@Ij&Jq)u;K}N)-M(Rny5^i2g{oRz% zz3XYf4%~;$mrblgN3i@YLyLNkYCqJ$@A@Fmu6RB0Q>e=?%yEzcSu2Z8`wB5Hfm7PP zIQa=q<~B?I6q@J8%slZPSl($tPMHAU!CX$uw8I0M&_lbaTSUgUQwSlx(o3F(d(jtD zIT|gG4RN`2%LBRJ@Cm$q3ff=^-E{{oXQm zULQxeVzTJ3&(R_d%#4XWK)L(v8YZn{vA6fkx*5 z?#SFZ3*mk}Kqv=#T883$f4B6)+r#c>;H@Ot!-J;V$5TOupx>f9)ikj*tdiWHY0UEK zHrVhMBNMq3^BV7qE2|?Q8j(U%#OBo?+4D)$fYD)Y&^Xq57{-wRTU=+G1gXA~MqYZ) z;Gv?)o<{-DHzxG9F^?KToTJ=)2j??~wRYZ{Nol5~kRQ)@=0=>(lEm)2Ts_Ch@z@sr zx!wE9M$=_Zgk@}*ap(4G!1vQ^oFuD4D{f-j>x87;gVqQy!K7Wko$vl{>81By{~U}& zBj{E@!A#=cmSH$}d8;nPmgc9y{1O%QV`uV+>a6#Y$mr))(>)WaDI9E1$S8Bj zic!~|a_~vlGG(jupJi3pCOXi-uI*m(wZ*rYK}+5y(a}_GS<2ul?NeB#RtvQQx`62|Fw)Wui>5Z zch#k^d%cYrg2LIYQe12u(GqoJG~21f&e7E%`)Ga7VX{Z75l@X}E6`z4)08kQU%@To_mWN z{t}aU6Z|cRHNfimkZfh&2IxiRv*C;WQNSBZ;yduh>Ws3)AI3_fRXi8K+W4yf<|(}% z8-Y`+h8)TlP?z?SNMmT)Dz!s=`JrjP%FCz4J9jR_M~s{B8RMX_gPjj0p!HB z#pHX}BPDk%3o8bnW>x=spUBrcu;irO;pvU`F#GN1-55`k^iKb|DhN~3i|{Lz@0=bU z$CS9;y}I($?%WD_T61C41p^L`KSUwQ7!4(6B=*xJnN<44W4HQ&<}UH!Qwy(v^J1(+ z+#Fw1ovy3ogGv=9r;jXRe7a-~y}5cH_HoVA?Bmi*E&C5t+X;cs!2V}GQt+;>2mHfH;56-TFTVz1B=z;>H=GU{f5Ns}VhWmqA#xvwrn z=V5FI2mj*TsY%Js-rpKuX`cOZe}+2D@#p;^W%IvHJnT8#g6TzLKy1gG3>zRcn#Mfg zi1K9&*HaU8?j8TX^qMn7C_%4|dBVzcG0#?;ueqZ#{3;%S>R{RaiGqdgZ`L>6kPL8I`94d}l zp`Fl#MO(}{(R5r(&EM0w2f%A`*1WSv10EOf_f3|)&X$Wm(+6)46S8!nV6IFOo@%H5 z*A>2}zt|7=5_poC4rOODvz1gQw##A@Y8a;4OdCD$MRyg{i^(!d{Fb9=KXd5hnjIR1 zI==&L!@6tN=cds=E9R0Gk!QTpf4ChrW3^0LJRJ*kqaliq`fLTQ6E1sXd1V}-h2OG2 z77%gj34J9`lApbWG`8E(cX#=ji@aJ1C0$rt<9aaAOR$pu_5E4mFzlQ3L5O&49dQ<6 zfx@}s>zC9Y6Q1zWSW{+(yZ%F|L8}WnZn4XI?!>#&W-b@iQVEdRolqvzm}CSFt_uoM@g&! zJ2+MX?ao&VOpXqASJ!WGvBSUe36y#d(7*p#ST~2_XC;(G4$Fnyo(Gp)&fkZfA2cgS z)w0YCoia9!{W-%vLCV)Ruqb)TY+?wWijHEafJMqpGK(LGsGcj{C?@&l;Y6AT6-3mAq7g$XD1(O@r*32@UNj8!YiEGkb3;4l z?65C;u4)C+&5}#xU+GK$7S)3M{Gd$3`DZB-Zn@3= z;&YN)GNmaf61_Tgylp_1R#atIg7iyl zVsrRcb_)Q-IYOh}PtJR(CXOQ5<_5V0LvZ zF|588`sfblCftLM3qQGpfDr9&R1f+6X!gf*V$I%tqMo(gI3slQ(&mLxwA3_{lO-%f z3#6FOF8En-bo`$Jay&3QPuv15jYD4X(5oi0-^+Ze{%z-X&M$#?tN9zpSyce|Ei$N| zbFMiB0O6P~KSJ-;^p4keUIm|Yg8zbWqNl^31b&L|+ADjh7kl41j8|lPhr@)~si`XG z$CnYRqwBXp2Y2~3G7xnwe{7CS=!mn|%qM0SUID)zD>d_nyU0}5xq?y8{0F8{OoZFl z?<8b!yIm%7jjdnTc(j2c(M9La!~}K`*#2qjwgM#FoP;#*xevX&ke@V)nt2QtgBURO_{z#HRkho zXXV7wAmW#v=dM-8)wXYI>lXJN4(4|o%g>nhC3^?BW*}K#-F+HkH72T6D|?L<2G}sp}jD z&?Z9{uIprfa#TRiQxBOv@|i-iIv#Ai)wk}o_;r&>LVq?mAJOfj{yJE%CzNa|jOPSRKU zp;u+Ew)@V^AZtX^D)CHzOyW&75KLR;RF>q6ujmgXN+K5S08x#3c1wEaU3+sK`v%=7 zwIpmcxQm#+z_16%ROiJ;MR-9u%7*zpKD+0%QSDxS3s% zX}hPOE4z%uH@oxOqzQuZ_+h(e7Y6g3rjNJn`US6e3XpyMiHVS~NLdG{%$eJ0G*M8V zS6S(do9$`p1)0Vj5As%ZWjS0Y=ch)@)0Ouj&k>e}ZT{KSHjFkPPw(%S?N}^gq{-yS z`Q)#WaEpZ*-s;uB zzp^6HFU)8WbR}~1-r$*UxNB>+0mNDl(=UrSbT@4uza+X0JHdI&6JkOGRFEWKOqAFn zfh+z1;W-K09Yg<`>Jc6Obx1eR{{8SfD+wff0yOt7Yh~+R{Q9gAd`4F`%l4=7Bn5ZR zNbLPPoxz<)?aSVNRDd@Oe~+(+5PL9eTmn;J5r7K`adlRB0=0lXl^)4x0B6bj1aq;} zxS%Ea>?BLBlyzFbTF-!^vQD1)&fC-gdHo=ukPK{mKv1VK)l18&bwH=xtPr+>6%uRF z{+=%3la&Y@RtO7;Jfo{(@WBc!(l0Z6nUZtNKbPQXxmU>si+>{LO<-vlkbyA1pZ_?H z2`MtA3(bIodFdARRjCJ~&j)Ny?SYO(Xj)}w`?E*yv15&1dkFiZ7$MvMFRUlTtgWd zs!h*A-Sn?|Jzs@E2CKN6XoZ zhL4mr8|P9&r)TkaEg6uqe(E8@r*aC;DT z`{23F>Gg|{uAy=o0S1#*)BVM=Uh!A`rhffG-cln>K08M)7G)B3a?0)@ieb312V=Z# z%O8KVD#pZ@qzM1M0Q5RNB`*lOBn~>aAV@N4U$G4x6qfTkCWcn-1xpR6sbK9$j`QRo>h zKI>eOn7??>dNlvIm*QJ%lVe^U@eGHyXd(g_b%VYn=|zwN8P1{yvn0SZ9AOx;FBe8! z1U@3!f@e|ZLGnU51*h7x>C(p0geIJrS3xg7z?J5pj&i{!8+qQpD%ZG<-vSfcu*Kt_ zArZV=klAXd#|ZJ-3a_FKK$|XXtm!jl7Bnv!xtL3eFT341t_oO+zPb3&dtIM#qRn-TP88emseZP|YMR*9j?7d)L%Ok!?%= zs%-*HhwBv$*O&#iRwUAw@^%IM9K)ZYF%xc9>f8V$joNGCAG*|_k!40Vjl?7l zBj@dJ2XUA<_zNqZhE=5{-(*N-YcO*QBIA)qGq_78g9X31X2x(2Xi%*{f32ybp=+vh zctbg2<8)Xt_?0nvthn2KueWT=I~LXkJlLPf$FX|#ICvnE60LS{PRI34w_Qp%AKrBn z_I|6T4W*yRH+c;w`du_S72tBaEn>&=cN57=%vb(`;)v>s#`_}={0tUbE`$|nuH-vyEO5nmbJMIQsvp@)=8(hn` zak^X(&ukq&xX~7v(`5DfagyQ4?|!ITPrNwF4ReI1VUW7ClBf576~R$7#e0d$1gbzK zjG0eU6epHY*9L%$F=5>tdrW50piY|FdkAj_3H?y#rtHy`lH@Jo=fc!|DHu!=0KdV{ z3e|MSRW9xpk-8jiz|Z2?df1iH`a75Y-Z$~vW&i0)K!*v7Vi8ay3@so}pz2FaGYbl^Vvt`WD*NMoO6+a++ z$&euo2YVrp|F#ZR_xv_|WTe)QdiNySTn#`4(d}>to}4d*B?`;h(z^i%pst1m2WtrW zVD>kU-}Sk)cI_L>_A}Y_zQi8-vO!JI0N4YMWtji?{=)tBfwaQKUA5Z-DQA|#$>O6V zs&#-cG#xh2(e=8muZ7xHpF2S4t)vJ{A$bPtvgH2Zuhj?(_ihzyZ}-C=b!6g4fkbdy znK8QAxwKx-M&$>@odNB*tRTn5 zR#iHwi5Fh1JbIN$bX#k%FEwe8Pkdy44x9)7`>N6dg|p8!B=WbCyM4l(!p|9FVF%}m zw`}oBr=@_8wbZcIv1s97P(C5}2VogTPWmQia;1e3HO?Rnwi!gpLZ>{nzze5*>L5)Y zPMH%xE_mX3$`p>oG%F}O1D@M9xJx~vlNILBDhOHgl$v(!7Fo~4g5Q5I*Yo?#>t>9 z2E2H+`b)>|dSme2$m;JxnSkilyP0qKi);oq zc*W^$W);VqlptTwX`NLxr*;-p>t-bIfR1HZCJAm$BF_Aru(qFmGj-8AR$#3DfC%^a znfrQ>;&s!<_+b*i@I%IQn{2s><@kHKbJ4Vpue3fLnQ=;SX8w935Qx5?Utsl96JGKm2k$&syKF zB#Suf2$fvpY|pC+A$Vye|B8?41@QjUa>{F93{y#oM~cMM$rpT_VH=r~M!&`*QJo#o zL(^m74v;pBAAmY&z6?>shvap~ zfU~y4r=|D@mjRcjRe+O@#KT&h3@;}U=Udj_;}SzJgU)QrYx-lnvD!g{Q*RpHZ~6Rr zb&?u71(3!ze_*@e&z`n0?F-?2rIuS<&@9TN>e+fi-n1~AEm4s5g_6)BMlRLedjQEs zJuJgq%QmBv{lH`O_Uc8^MxR0@thcFA;9>J`^A>NZM}Hly8@tj#$+Mq$g;*PtVA*0m z)iR0r100NReV6r7F)%X)qfy~8YS*w++1Eaw;>E@NJWLNtVG7lH++gsPiRirv zQ8eeSI0V=<2A(Y#Zp~AEumkp~I*c*_pJH{{Ny~p!(ZX>JzdVkb(8AxjnF3@dLlhdZ z`_Q(|<~IVubq#h=K*+Lqe`}a1>RXD}Ll<#VA^hc*cblJDwD94h!i2? zHR$7x9a-K#a*c{ROaVVXjwDa(diV(!VzULPIAQQ8;}iYY~m<0YnE%b zfK~%^rrCe!Obwj6#~P1f=y@Hv!|Gf@!@whH3j_Pu#bKM0jw25*%+8#y#~uv6#htgo zq02|+0gqFWgqzdtvS-FHzP3E%T8;_%DV4c0NITO%-G;dB07hv4yc1i@MAmk zi!yDKC%Ec+gqA!1_-rm@-5m?CMFc<71uJpOav_;hqqT^#a5)H4M7#{cCnIb_ekJfY zdi$;N`J z`2Y**s93z9DT7?KZ`B5b{DU;jx_KJlyzX?NUf>RuTy2tv`yEfy+aivRzW?%<4qxug zQM-xLU$43X6pbAEkpzUJeK=4YHrwr{S^P?zpH-M`RVn%bmHwqhvxd@{aIu06WZd*F!~jGlXIHM znLTtE{54;&LQ^`kI>ckB_Xhha9H=<6V90s)o(+~sZBVS=oPqx30@*XGc>LdP2MV4e z`2dKi^hlYUr<~2S9G-8^ZHVHdTS}?f0r|#tMN04Wg5shxD`+P*Ge6w#W0h1dyri%E zGAfy_9?bAGFVq@?HA z$BmN6ph2jsY-Vv$yk0lYQlW(mxi&=)Fq5?`kf>bop6?iiaPcAudn_|{BIV(gQxmwGbdl6(P-Ur zo&h3p5M6M&xW&vl$l0ihSSS1oPyA$?EjG&4Qg;yMzR6GJ)cn4Rcodpe=X2)&IC|`8 z!N1fjDH*k8+%aa{EO)wp6%|wZcAt9tpfu3y!R$~0*&-x@UvMG%<-!~;g0Y`7e}D-U zb~;&%u-$bdkHj$U*Ci#|cLA@w#W(X!9%Cqy0zBjYCN8Fo>E-KcC%5S~-D>t6(T{SKHf#ThJyj3R zvViOXA1E!4kXBZjkM!n4z>whf6RH$)#*Yp2Eqa)kc|&Wmr{sU7{){V#DGljq9jLta zqXMk`_87L6vU#~PosB;kcX;3?cHS<>gdk2u_#6o0okoYvc70CfVn2d`=aS&$wb+qQ(OIYypAHey#ON!55q$}c0(;mXif*OqU z;%c{#rwafrOUTBd1{Bfs-IOHv9QLhDLiTBS`@hP_VVToY52SO1;)i~G{c|xA6Yc(! z08Fc4P9iJNY_!ki!~0o5u+8Ad3ZA75qJ{_|PB&J@)1M;zVrhKuJ9=oKw-@pc5X$rB zzpcyjTxdU_i zzI!(`=iP4{6oAhIGa4+F1V8X-cA=AFx_)id&moTRZe~z`gCPY^t0i4 zZMWXu%=h4Gn_GhqS!umU5wjm`)TB&HzavQm-B$yjXxDjqFg`oyA4fkpiAH*5Y;|z? zvQgvKf*wt_weX*tQ2sPN_Ap`&kn$iBA)>FG%|THxe#1D`TkT$c8X&0G%>44EmnwRV zQ%KS6VPv)B0*}3JfPsP#3_>%E8emJXc7>dA`z%^55Kj}Li86Z+xFfAL3b8EGGF{9$4c=cSN zivKM?odVw*W$KgCz=3Qh_?eu0d?0n3Bc|EESWD72ci~G`OIgQdbLZ_#=`?!j*FL)R zcfq|{ZTJ*Fi`ZhA50yoOtFi@(rNLYi_HAO4GJNcGotjRq&@$MzRds+lDb&g*%y4OITP%Q8)0F4%y}};pc|c^1YfV5hzQfr8Ev0u>2C)vm z*xTmD&bYB#WY=nDd6!O4BkzD0aN?J)OUQUx;$@g4C+=)RCjHBw`@!$vl7dYh zVl;XacAiey{}Y&k|mx*s^~0j@Jfoco*-P-*?Gh%%!NN$NX@` z6~HGx$)c*YSNzc@r)ydHLRK(v0|+;IQxZC>{F{~MESp%dXLfK6qyoaf#Gp-eMVC`Q z3s}eGgG2rIqzz3reS<*RO0n0a$RU#wNr{$)6rI9bMTy6ynzUG(gl;ckq}lgg))O;w z=oD7@l!2UwBQ^uQiL=PH{5_=Gjz?6<{D$h80&Y%34vUhl`n^lQA0Ax5YL-shZgcm3 z>;JxD=;+TW2Zml1t+bBy2FEK6p*eZgmg!zQ2S<0Z2&+wZxNDuO#{2msfuJFZ==jR`?V?a> zcc{gM7A(T|&kf1@%))HwUMA|@d&3w0kqwO2IRin`*O+YI(~HC{;{*H#{z)QlWwtus z=F(uCi^(gJ{EDT3&%KrJsN9k6aN)CL>d}7@BdIso%-+`LezD?1I@W*MbB9Q^&ox*u zbB@nQmGk8HWw8!Ys3ZW=Lq!$A^k<|OF7fTZIgBF?&NZ$!J}H5| z{Rp65`|@Rjec*uWz0<}X-QNoC1N`{E6W~>3?|`T0(|_l&eKX_UN*Nzw(vUmfE9C4G zgV>N}T9x3l?^<3KRK_zDE=+K4>zL4j-g9U zp1TRQ0_&RxpGix7fJt9Fdj_SVO=a~0tK$D-?=8ck`rEy46(poZN*V+Nq@-IKr9@g_ z01=Sx96}nTrIC^@>1G%jq#GoM9%>l6<5~Rw`?~IXU;Ezs-p}*ud9mMc%nZk|)~w&G zb=L1ZzaIcHL5am{^H1IR=%=Xor`HhimbD}p^TUj*7rs2w4~z!yn)Vrk_|FbfRnJc` zH@Z`9*8f1MJ^R>Bt=Na0f5qwj5kAGPXZXiT{tb+1)lo1!NjVui*%AYkp?m14jPFZ9 zzn4*RfdGYI-0m@E4l3{yGE1ag{gbN_65Q(ULmhqW<3p%IJfDCnzx0a1?%%xU&{(!Y zLkGa|BD}ICvHp&SK5)QX_ou3%Mhm4^(;r8@f^aS2pKE#l!p?&Z zUPKdPki-rswt&&I?JV#*&gQ!Hvu9l!W+ zetK{$b-gHVru5CtyJK9~%>Rnn|ae|B>VJFQ3pT5dy4{JQZrJe=7a_=ai9ue8c4*&i`_l|KjccoZ-J%oPYl7 ze^%+AOYvX4`2XiB*-brxiPZe#;rNGO)_?vO!HIoKwDi~K9_C-cJ44l=>j}TZq2f+2 zV{|zM|Ggdk+nW9#|F+J&St=ob4hO}y8UKp|{4d_h>pWuKK1iz!9%}yYl9(P~rs4po zpbTKVx92~#QDD+^FM;c{^u;=HOf`zyTAy#^+AHA&!&$_~6N zOoA2($F&dD{+XZrXHUcnYG1;Zca4NCq1RVUzkPf{Q?KO^SL%m|3@||Kl9zq7B|Ubh zvIbG4BV&(IgBvh9cqe?GziVhMI>bO#>^?p)Md9`A^O3-^M;V74>$WW9wEZIu*PIy+1sLuUfK0h+YzRL>wRD z*?mS1x6t{iDYpL1hYt^#{LF`O;Qtzb(j1`PiZX89o+}3Oh$oy|H$29SZf%gF>F9FU z&eBzXc^o#JW~{tI51_FK4gZhn*1vx3Xid!PqpF;%8)FTR$b~db721BKN4*L(X{n7b zT7KV^$}cXY=Yq5M5#yov2mKF;)PHbo<}qK1o5kOSF_vf$4iLC;x>7*QS08%w>8yVrsuK6@Orr+#-Rge#(xJgbbZ)gR%vr1 zn+5?gG>0ytMDPO7o|8DG&?`U(=*4|54`_yl$8l&e?ui)p(oms^U!TWX%^us>)zuOI zW_3SY!@WIy?SJI9zZBJ$6Hi_G^6m{Qzl<_CDQ%Qp1Y& zdUe?QDeS>@V4Kc}5+XomH$I^1+k9&;i8wN!tN2<6aI@{XD7HIo zy882UGJDY!GgUHdmq52Jvu&w%_P9y!cbz>}uBkDfJCcst)t~@NK34b^HXVp z9z%ayu~b&nG0K_stE5E~kCUgm=cPJzr4vP3NIi*^S3RN1F~W7WiN-B=>|O5?;=;+e z8T-8(V%KQPkzpKb3LH`S;o^a~3*tM7WiC&8C z_IcRUg316|7gpZezhEFM8$n81Td{yhC+FktX_t5IjM+`SyEudOFc`FPOP~z!z}UU#@d3!a-eM_RL@EgTunqiU%@6CG zz?}A*(2l3K?7|y{^W$vm&^=PLifTEe1T7`o?eJo9>zsNGupd%Zx6FpaBacb+_L(YX za;PLNH#4pAizCGiozLma-??a|K0g^WA9;nt5_oBB|b9n5~9D3w^HplV1@$=3tO^*OkU7KN> zJz?JOF_Ypo;E_}ce4twVjfjnIwFh_nlUota#3K+nKTBbv$%c!bti?_~;|_G~^UFDJ zq^bALe5f7sUztJcTZN@Ml<(_j{y)>x)sFc>#cMtVUZgN)+#9JJIDJ%oo>zCW*}Fy0 zi?t%l(ilsC?$PhC?y$hl%{Jczt_vsUk~C}3M9hw5nvV~}c@0DEf5q43uC^EY%{v|V zY?w4$?x^7P+wevfc~>G|HACvKzgqw0PaMT@fXqWV&iSU{$!qpYypq?h~LK^^B&y@7ka%a>`E$%`P>jGT8oX(ZRTSJY$8(+}Fb z9t-c3aL1pSi1oT}F&z4TTRUh&r`~#~q@Bp5Vv~H@DbVXPsBuol)gr`S88b{$_IjnN zpShur*%RdvTCCywxA^{bc3YLMUxw|gn$<9{cAz|4a~W<%rS2;9?q|(sxf}YEpWEw= zPA$Cg-n4pqe_HBjsk#hHmvwqU>67;dxmr?83Q zg3F}*IYd)6@mhC%>xcvTOq%xVPqvX#C9JJ$ySV_-3ALf6wOeubK1DK7m+xY4MG~pD zCZ>)8_*?d~ME+bFHQ%N;A;ful*kXhA(;obGdyct&!xi?!wr}Ji3f<0# z9H;Z1p5WDP6@7oILAq|uyJny3Oo~DF=%ziWumvwgG%<%>L888Hd1hSw$yC7!({`{y zy3hG7c3z?8mSH-^cI&blt=O_l>*pRDnVtI*TR%!=MK;JZDo_$&>Rj)c#n1H3K**X2MG0j=CQAr6oTU|L{pGQgN1xUz$% z1HMHFAyKOp%3$O+dKH>t6t`ae3>rqvQB04O$u5<)VcI*zWMiHOuPj}->gDd&+L7Y3 z)>8>nK6MWMbGf!+_IY!k>m%o2^3jKxBDA7EYFM2`j>c(G24tZ%$iiws@I`6CU9EzA zEt%oy(LJ?G-xW~RfAOr_Yb%k~Y097oO><<0$xvL}Op%f!q!T}H(4byA2%9{+{{iFx z{TRVc$f%^H7UszoiUG?g#2@^!y4>}2)HMiqkJZGMgD8Ie?9_**`AUqjtmPm96lt#a zo@8_$)36Vvx~2^aMJ7>P6y}W2ggRE_I;HV*yJdQJy3DTzJbQ zve_}*_%L`GtC zS{fI0-wq5Q>chvTqGT>Uv&?y~)WRCI!|VT^247mEjb5))k@3D0nIJp_0AA>C7489E z9D~VfHmqvh6q;Z>Y9S~2MUNZjdw^HOUXzN2fb`gJ+~g%JiVzp%Y_3#VBA_5`j&5e+ zr`I#I5*JEnGMB*SFxo}Z?;4DUCD~YKF{jz5b<6Z9lRKWOZ~U%GMLhm~i(fX9n!4t> z*9F^i{;B?=qBh(?YEMp!o6yVUM6i#SsE z&(+Dq?TbLiyM6h`o>8Zl($IWoGnLm_wsi1Vg0r_lz6=8MOntQ+e^=Ub_Q0>(ipkM(^1y+B%Y6e5NVtTy=AX05PLD$ z6w;iRX#Xr3NKF8mw5-36Oji^0s#Nc%;m2WD5(;hFFl*?9ec%PCs*I83d&yM|M;URx z{&L$VtFm}8O_n#@-}q;Q_z~7ow64mZoZ!4#^2RZkzKVQ#9nzYgtvTrmsrf`cFIWSDTN*I62sR8vCSYP z%I}8($BrTHX?T4`ROx(FW%I*+#{SWMcNwTE9YA?aYOTlf78B>EfJ>c$sa2y@VuWbR zm)qkUxS6YehSdTx!PHcbIAMRG)L3hQ5zSv4CFHF6rnFjp@fzj{)~+-W)~dDAqvQod zOc}Woa?X!ZRc9|N9K2u0TZ`Mz+3LTN)%$z9!tN0Nx5FAkOv~L9M8Gwq8R4I!@!L`N zoLv)A@1(I@nRa?%meKp5=9>Qmgs!rp6Gq%c#3Vye)%dC@!V9a$ze*!V||-~n_VTowt@otCOT zqD1(ZBQbc;a&vmd@qFJ^s6+@iz&&dCeLBnrEoMn$z3sN5Zw}VilxMfRvo`x^`iqLT z_$bN6^>(pTU-5>XFB27Z_PoKLMtzPRNUyRj(|~KO(KxZQw}CwSIzeDVQ&}He>>K(| z>7_0#lsN^C25p}naWG}1F@$fF`E+bAu?Ai1;bf?n*H1s+_Fkqwn(wqx94C*s;5e_c%2Y-4Z^V_hzY0XRi!}>r06y-Cungt)tX^0zq1<0UFVKlA z%*!BgQvCK>3T(aTvNPB?KH2c=0>NN0yFRItFbLfsbPO=I`I9l>+*un~#zBhNL#Tef zSO$3_ezTj|h}SZ8!Dcr&SqgHfr#1lJD0ywVP&S$WjrMX<#btI{sPXU`xkM#r`hce7 z<*?{Jld~BibIf%xcF_I9ws@!9fW77VHFM6M88*e+&i5QV7PB9aLYCdDAAZh@)4Qf^ z(e`!TSa@$5TO6}WBN-!Z4}KpPJ1Pm`*Mhv)oo@)~O>G-47^~hv`s6?RuNvF}NaLBS zLrWnM2jX73=>l;6Qq44mq!@aIbRV+643f8@eWT<6hw2@!4`$=}@AwrEa!Rvsnbro{ zCGsQVJ-F%-8srKvGY}D8s;PQ zfs1Ig-2@8{fg#YFJ%uL7s$kH2Q_g^*?glfpVbS>v0TmAJaQt^IUp?}kS4CmT7OZlO zU7qlt_Le4|{&r`{v8^d0dVH2)9m3-;M8siNFUN`erRC0f7F%^av$cH1T9kU5)wjd2 zeerQ*G>Ge>hr3fGU;O5ndj?|i1IqudIJ#rV3xDyX3(K3mmX6f(2D^(A+LFVwp-kYQ{!HrJl4W`R;8)#Z zOqBSA`gF0NBZ%%X)*4BG3Z8|jk0Si5%d>5#mK(&OPL3Td8)thMSgLqcpb zLjB3)JawX-Xc?TqibXd2)d%PHO9v9>{CDSV2&2vnc&3$?w8d}YpguC z?q}>dvydVinS7${v;{C|>c^s;l##?AYhxQ{aUl)Pv^HoRY~17}AQ{&nW~uaiNTq!0 zHYqw z(|Yzg>y?K^At>-_ta@u?=opLgOMYNuE~o?jI4e_4g626^7DFpOFXm`$c!NH#Pz+pF7*Djdx*Uqukw%+@*j*%d+B}{m`;kZICjy@RBwJWRFD|5Uu z6Q90b#H!%fc2In5FYmsvD)t-Dc%3iU!xZNp4zIQ#$&t~e9$r+LaKu5_D4zL$p>!*A z67#tkngb<{_Pao=rCMCMTb%g_J)_RjMxgoa9>Xd*i+D#OD_-eng^6hwtG4&`DS6jN zK9n)9Q~L)a6I;#w$65; zs5xbM8QXW-+`#K1*Ymgy-Qysl>)TttN%#ZdqDxVj8-n+S@+big*{xLX=Mv|G1h!nS zV7VjWwbt+^sqJk$f;@ze@T`%prwBsqC#~)Vzc%8!A!A7-3&Z6IiJD=Fb#pK2X3Ni7 z)*n-KUmxWo7aPz;f{6o-p!D z^W*8qJ61>ij#L`m4#l1winE~gM{`N9?X=yy&2FY!oEjPr@_vZ$zKAsyx#+TAsWG0J zXp)=4T3KK#MYtl}8GcF}h3=7PZ9JV%YIpTpN^Qyw{A(_#&~ziB!tJ=!ap!r^)PK~l}S0qd53>1IKG)fsEgXdxn*y^E^)O< z*DPwL+cs?m&9D2-dM;|kAYX{Wcz$@|UD41G04Lwl07ekwSNIo5XRyp<`@IjD4K0MMXJ=xitW8$>Yy((j*-IZkh zz;a`l$2@b@Ya&E~;9%%gop`K~VgBmr!d^WLVwB6%)Dp4xzF55?HM3)+f8@(cHE82v zt*U=xW5-s-Pv#?gH%!=<%r*t*txAo{pH}%1?+-R#R0pzxn=h!)u$%xm7>`~;nRYfh zO;Yti=ur8?n?{9(%3MU(Ws{duPdLRZAU2P&Aa8lRc`l1CfEti!FuTk8-q^-=-~eZs zHbAfrJ&;0mrL~&O(Zu(Of0#1gT)F^UktN%%7evTaaC7H5hvj zpo3|*pM}l2^!>q9!4N7)L-Q>)$goj8GM+`zO`Tii7lAKJ;9-5pc%cS(pGOjS&N!0z z9PS?O6TENG!E?bM<*#b-x<#{GrWL@s_EnmfdO@O7DnJt^K*qh|AV!Z%-oUFKUwPSw zh1L+q?rgcn2{L(6>3jJayhkTdV1xal5{K^fM!oe^&QG(>5Y9f1bhEQUK_q#R@1xD@ zpk=`TbJXzUqc>kXLzL2}5x@HO6ojN9GiWD6t=^)jpy`7YHc$G`Je^qcFw(VZxk1*j ztM-2Un}_K}_((3M68C!+Sj*KIbO-}}wVMJrIWIU6imUXRw{TTYXhZR^5Ic47rpuZjwTKakvBoiQ}9#hAs~P#%56uRVWfD9rG?`P282jHLF@r{Ql)ahx5+jrQkj z*@k5hW_MrVT%EKNmOu@sN=xPtwsOljR%lFa{%1{s4L2zQ-=#hyrE(o=%FNL$B^Ms6 zQ$-@(4oyak@JS|ro8E<+eZ+tb+q%)lfBg}@k3)->g)TXV^t9I>b97zx=m@*^Vt6Nn zLoDQiUJmHxlv`p@hT~!(lbZ;BV8_1xw%ce5y8}3g))34Ema;w2-+xQ_xHjo z#pe}bNO0yat2wBGqyFi!2wS^XF`D+0nE!*2HiqIv4hJv3AJtrTIdWyb9$yP}f`%ys zD|a3`k{F~)*>L#eg90C7*2O~vE0gt^c0KKzseV5hR;HNa!^rqTKcCsjEv{R>X`U=# z{;^2tu{G&{s#t*!9?bFWhI^Q&5oZS=!ep}O2i&!3TKk)Yd%8`OK|&T+k%_y5S%8q& zJa@V(ZWf=q_~kKqR+r<%Du=btV$H47o`0>$mc!&p=5t6`W}4xEyR{2qkyR=va&){d z$$CHZ^0}V9w@t!kQF6ifp%Cs+YBkXfS_f7>(jT9Ns3xzYOA%oqiB+~N_rurFuuJ7BudvSu|c5zPhc3XVpbzGX}Pk4DQ|It4O`>_@iGKWzNJ(iCKgyp zVL+RdDGPIN=V!^luE+#tLdJrJ4L;I{V1z;f3&Z}qpqi~es?BwAQC8zQHu?*Ba$iny zd+I_<)D8Gb4LD@$DFv*O)>`Ic~iaXbW`svWk6LMK4e$P6RxTzr0Y|FZkHp$m6x-D^Kljkk0k;r%UrQIAv_D zR)d>t=uNR!$;eP(3`!BlogkdT{mmerx9--2*BRrW+o5GX}&pO5!NMVLu`=EzDw0$8DWPl<=C6eU1fa_ z86gM_i@vM=L*ChKtOcC_<&%f+s=sF996g1Wv1-2wm3Pj$Yc&!=WfKuw^xQ#&e2v1- z+Q%r-y%TFLS*m%wro}6cCNvlEGaF5^$@<=l2!+r|slzEHqcUUfAV*cbJ=9*$2?jWk zM_A;V4K7X${jEb~bFNLR-m8~HAM2A}Th~%8*C{xky1TcxjZ-?#v?YiLDlDpl$^eEK z2uSn6$FkehN!e1wCE*eQ7I~;_{8;{vg?dLMZg*2oF7Y=5q%PiMBgRoO3M&&{zsyqW z30_;q#gj4VfaVPi#TVd+djA;+!rNjx*=fM~AhWiy6jF{b zDwnT48j8o8bbR#aCHAKQzXmn>2vC)^_Bx4kC$NE5D5E^zAsDymmFFP=J_zL9cu%1E zY%!6e@F=kES*QI(8|gFCuzE)`C1s);`tQ*_B!3~rt^lOi0P8!HskR-6bnlZC4S4XJ z{6(cnPM!y}{V5SNVp%sh+YelT( zIf|ZJMcMQ2uVokSqyuQ%vGUaACzfQeX%{eF7c?!KHv^r}bi=CH0m;9zv|5Fy%|Tb$ zL;U`PT%2oU3^a3~LHQo`jCvc=xKJ|;={~H<*4Xbww{6ex|8_kRVg#f47AHb}W&&t> zrJ#cxr@b*<%`%<*BeYR+li_|&-ceEKAF_KK;~O`6GWSfePb7OMe25Qp@6It{q%Zug zp3*(uZtkBDelUv>#I_ekrp6@Ce1h#<#h{?|<>6w9)bss!vg$95Nd>;0R#kzxc4FNZ zvu@vbeG9?kj4S)0FrQP_Y`*h#_?sEAbe@S{-HD~eQY!&34c7Fdpsk1d;a`=~cw|c< zhfi)6kkq6eBeq6286{Ey{OV>PGqOy0hs!)i0}9114J%MgY3RyTq6&Rh7Qt$JjJg!voA@l z0mNW-AAlh1QT*n8mn7=dcq$;CDR5VKSvs2dq|7|YDAHvyxJPtjxMk@P5f}1lGu2A@ zL8pDWY3JoTnM26 zIfyxIV{m+HnHnh?@lr0U~PJreA;a4XkEKDPF~9%L1MQG znZ9-eNp#uLd;F04EPe2!y{uEYl>7S?)ShM0aEC!}yz}ht1XK-!T_(h6f3RI>HO zPX6VYgeq7yn&n)J+N-dsaxreOeDTok$zyLmbEi>)+RrXnr2BNtGIE6_mE;Wq>ZtgQ zz7c!S?)}9bV`3e)A*`e@VqE@dB_dB#^oREH8dQLA-k*Q-a%Qxs3iD^R5x&w$p;0^6 zRLdC^nOJoT$m}+-tu<%;#N2hhgIxBcdMi@b3Ra_lj%ePlx`aJWc+U#MPpt|VlKLq{tX>lo)q@l(3`0@FW#F^&Jtpxou4Hp4&z#7EcP52q5LMQ z@p25jGxanMj}#(L<(-p=`#21JlV$R)b^90NW-q0MP9v*TGXX6(6iT7v8e|AtQwN#< zlaU!q>~uqIyrw&7_w3H3(^{6|u=-+n671;)V`q|b3!tX7cD2{2^ce|5xb6~u0Eh&m`3_;1y_m%ts_}d1 znAg*(maMM+?S~wotV#nfzt7*<{ACNI7J}{OmL8~liX$Rz7jbo~|3SP&(`x16Eu-%B zlmFWRUi-XSSmM^N_tS=)s$Ea&ods1NqQR zm>ESM1)UCQM^#1CwTC?X^T4^ckhpm}^H zNe8qUkok|rLbfr^@c!nFRoz}Fd635uxdlg3#y@?eZJT+Ue8r5tOzz&f1npr8A6XHO z^lh`gnK0z9R;3}Od11C7#GBi4h}IcKw_T$ZpD4ENInn#aXS();5$r35famLak-v-) z;D0B3F6$Jq&AYwjL5iqhjM&o{^`HuwSeK~st6dS@PX6i(;Kp!=_keMG^ES*!hz$~u z`m}2ZJNnBXb09Z%3;Zb9t=kaYY;PuQ_K5;p9)KGQvosy>CeCL6oyZ@BiyrlUDo>1W zj#1Yv*~!m4OaF5ov-x64%s|>*&LWYs2B5pw(+^bYp>@f`38Ncb$CzJ^=(e%2KWY6q z)j#GY=D6Y%2bew7dBRDToqy!d%_ewv_^mYk-vsYmk)93a!L{;I_8zWZO7I%+?&;58 zS$~9;p_M+HJfTltAAY6FMoNPz1uj)7lRI&}GN>HZJ))3GyZ+O<2ryhDLR&y&@6l!H zR#fFa@0Y6-{B6xB$FNEQA5l%Ig}H8;&+e%^%6sbkZZzNpCbht@(K{6-9$>sMtt||IVvQf zxJw+MfQ5owDP}Qx>!$+af6r^i`#VNCF4mfaZTJF(*E|YpyRR_L77^p-w}7mrF-?bb z7bOHs5B#fIV;!-w`sp@5!;>yMj_E1}kdQ(wb3Z0{G;mTeuK1$mk9c{SOHc}Y&Wpx< ziN}+V0@&g+{H9}E(P^8%iDM=vaALxfy-zttgsE_@>;$sC;!)1oj~G4J*L#y6Iu4## zBcNp4#}1~QXBKLUFuEfuP9HaFMV9I;Oezrh>5?&Qwh<5gAYPaF1o~b+G!wd9H{|rq zSR5YqDxq!B&Cx)bvlW@!C-wBm`FYj&8-dROGPO8(ZQo6bEF=htR;C7rWrYR@T{b=I zFKXj%mL&Y`io5@EL!0#h1uLmhN?6~V<1)`rn;KWiJgusZGG3wxi`F4sdTu`Q;)#e{ zE~#Q8<()|HTS4NdF+me;sM*)Rm}VgaJl9L8W>$!s-*Z9>aB(Qx$^7^rifKV@Sgun= zWzEMUF(|~F^{0&vXMC9+_XVTSYkFGf_l>LsZcEywYX%ky7OKb3}k{j>^EmqRwSCxO!vg_ZpvD?AE3ck%Qx!B z8#q!2>v-F%v}Vuu>yC=yd7d3^)`}L{m5~upUv<6)GMSg*zAy1_xz173{P3?hIe(ki zZ1YTgdX=u>W;(NcO-KZ#6*FE+RI3l3HV%Bp>)edS@p68y)?MGS2^fat716RcxF?nQ z2j?gfuJGAm&9$QtN{V-mvXN`eH3hSHO0CURdz-M_W6E5oO3I?JY}eq^VzJvIRzDr; zo~YufzATs-R#DXPdx+=l60=R=3Zid8EKD%~@PKA#(&&+28G940w(|-ywy#PaD~C*n zv>T-hY9ou@L3aP41z^hfS*Wk`TQ1hI*brh#2N%zAf}gM(cY3*JkWb>Y4&7P^-c{Sr zw4HcjnUH5;u&muXrNh{A;O(+Gp)dF9HNbS9Rl_nn2Z0H_Nw4a00A|@uA33yvwNE282vnby)Le(q#u2|?8GwtEkihc#1PB#aiGQNKqD*>pDQBp z0K)6fs#?be_JF<>It)F-l`d+3(3rxz7lM7>hSvK8C|a6j*=06wQ-qN9N>do@!X2V7 zpCffwyRjoEXlGoxuF~J`2vH_?w`yWkhI}w2@3T1~MjX=!y>i99$CaIi&G_t|Gp*SZ zd7+x7H<+efNavJ3KjRrgGLh~Fm?Z+Z$Y2xmQFhLrcX_{}cm^H_PGnqxU_`uj7(1hR zJEP#tHB~Cj$zqKsuRP5Es;KULFHEcukGP-4Q%3cD z1+&!(Rx%f6a`fb|wap<@0BC2ig3#tNHczR6y)S_{clswxUfCNvvcHju8uAXLo$Uwj z)OnF^{S$g)7}OEb8^-ZwTpRx>?kvyM947WuSwu9@HbACFWMhLu3A(f_BJYmc(RjQ` z+hrR{%={BmBxeZ4F~K-;3-etxpLw6B)pq^onPoMOfFBx2@#Y zRWIw5L(?F~huHo&_Xy;cCigkxz>JV^2`o1CBqwc+^I`MYz=3}(Z-xtmC@ew@}3_uX9_Q*-w#+Fj{wU{DiEuMEi zGkYpLD1B%-hIg7Q6KTRhXf=*n_k_jUjChos^ra`0Hfd^8AOj#?Q&yLjK?u;?sPhPM zIZTsu|I84K+P1z&#E#hm=3_Iew^U%>k z!2o?*z|9s$McMdxZYD_3dGsy1$}W65&+X&v&H*JI2dW(g`KA@u#+jsbndNMKI`)V3 zLmN{enuBsrTiD+38q6@?IWB~WOV7W9oCWheN8Q&B(EdDxG6ztkY(y`2+?xrq@IFO{ z5*y}+BFl>Yf}Hu+sjUW`Z6l2;IH?|Eq-|1s&ct)N&JYyWi*;S}WTtDKot~I!_=H5B zQ94d$ui|X*mA&dCgUAxFj4RcjD$JEwjuvG@%1wUkG#d-uq&}(z7><>XP+c*~h2yks zpPO#DZ4jm(@27ti479C^fH9$colvJt;B<1zcKlQOIRN>&pXSgA59;PToDh;}2`7lS z@e8z_M2X-uyUC7{#uF*8JGqSHS>hFSobnX?+O~)*N!SU(kN`Pq3++rQus3Lnqf8Y( zBz!q;tbt(&eh`%=1_j5*#9KUgw$U4O5!$)w)(sip8~|gqiBVyF@LmwAPV3=>Ka6@4 z(CEUg>l75d*EL~nz`r~PZ4s`u{gwUb&(*9IP2AKkeeP$=a~#!;$!EmrRT!=|KaAG2 zsFT{1PQA|~8?tB(+f)6TK?S$TL*Q#!o}{Cpep5}Ey?jm)=pc=s{7O&D(oD=<<5Le* zp$(~f?I=e=gpqDV--3rbgf0;%-Ncvo`-wL3$v~*}HPJuDmF!3O8`6xGFRy;ieqP`| zdQoe{scF6bGfJo%3Nj^EwDlKhofVsqlndh8*;Sb^2GqI|e|%dYjaMGIiAi8?A`Smq zsHK?icLQF0fNAD#ztbZxJ|ch>KCN&N4Ydz`EC1RYav|GKR70j%ELB z^zHtlpI~kCEagVhsaa(^|E!7|ZrgjatUbl4zs=tA5J9&n=*WYl3YlD8IvxWj=s=Ti zZe%_wI->Y6+o0jR;J?Ly51L6D1Sp7!zw8oAo9sksl71+^ttz`n6*0*Rr~2`f>Z!jt z=jJ^-C|9GYm7>Q6>&4hqhLD%?3w4p9NkNSeL1jora)TAP1_x1z@Z>+zCD{-*qC;L7 zOO_AM1KX5^|T;ysS_eujs+Iu!{T4k>Vt2x3705;S%(pOh#gq5uxt@lndUyPMl#U> zEiZ)4djg{L$|Nj$+Lsg={EK3Uu6Mh9=|@t4#wpMzOjmq!Xq9h`1zQ@I)E|e%CmA$a z=}_v^e9t`Tic^zTe3b&mNj?uT==CH}qEBtOQ1yIr)me=i&rG<;4CpR;{ z;15JG9xq_P$L=lJi%!Yv^U9Ckd;n za&WF(uuHUi>?+$m4Q+UtJl=Oze87EPH8)E!S)|*VP&Bhw8bG2|?*y@90s2B#rVt%k zn!WeVjRZU8xKv~NaoseT%}sok9PGNDu1|h@J4bq@EgyV*Gw8T&+P(99@rQ729%LjL zflc3X*~$u#(d@ef8T{W;uj8iwMN#GCeTYUcFo=8e z&2AST+dt*z8oN?OmH?^*;;P@DflsmedIaO@-D=vj|OD5!5qKtz=IH=Bb>gCJ)NioC=ZZuvw{q;VJqZuK0JM;6p^I&oX5i=0qm9y#a;qex=*io< zGwDG`Gel|2bJ{2VWL!(5+N8s}y5}9wWH1P$%6DA6EF$68)pM%ImMZ?&nG+3Z)3)Hq z4Q+p7$e0~_R2{TC4#1)LYe)Yhu5)+6b#h2Y?-TA5rBf^c-9~BE4)zlN#X2H_9xA7W z&t<+A_5$NGpa~m|#B%*rH})086^k|BX(`<(>X|_2l(zmdc_=2@jNQJ*WSsoV3f-K1 z)@1JDcBO7;Ld^(5$Oxp~n)j*qKZ3Ye=DPXB`!Mc1Eq+kX88;v>UdeEW{mgj-y`6u+ zmb}ErqU(w03v1ws9sql`q&QxTr3z+8&NR9^3SN$@czEv}MXFqp*7#ofy8ne!vspZF zq~`Xlhg33k{lxB{mtNC4+lwD%+2wg98i)k8BxAl=EPlyXA{G5Oe1n6uaC2%RgVYVC zE<_b_BDM5m$I1@dZf}5?e+j4KqP^oP<2j#F_CA)xK@? zvcD%Uj^t_&`nmbO<>=>yI|)!VFZFXkqveF1M$FxCEL!_8SzcW4>4y+|zJ3Q#|RZ4@zHkV&7{1|krHaUHg;Y#Vt+dzg8}R?-Uv;NOP#wwLv{wJ?t;s+tu-({B_U+Ui2YZC z53b9F_agQqB1S8|;c5y!}|x=8cgZD9FY!Kh9^kWG7xr$1GNsn<%e?{n;Aun6QPME#34m z?}P0t^VbfKl6oZGClWRO%z>h(sWL;--CZ`mo0I(6`g5jDXYB=MAaLqiz>rMttW5uK;5@I_lG#!s(8$)0Yo+S8IQc@xjqPOjbMY?WGuun-N5e=Y~ByyOAkfnl%)CiV`fyOPYI#14ST;W zrR8fo&0p~o10CqWldcK$j*6Z`8KSy^v7*H!N*az~8LpnCm!E-_1?rzYdd?{rn1j*&MnPriSx~2HmQ^#gXKS3>d>psKU*Kx9lHh8if z`}Q0+%$J0}YF^>IR=CwF3UjqLpnQ5Wj@#&{IlxvIyDRl~68X%qqPBGaqjZR*HtDNf z&vQyxL#fu;(AWp}=WP(7_o-K%mpIE|&k-^fs6C=U1jj<{X@n+L%-Wb<-M6`uz71iD{iEWETVze?eqFj2VMKZj3p z>8N=reMH)ZZGnQ>wPY_3ex2$LG$G& zeU2?kP3)J#8{nfxsuLZ+@ZUG-=!)^04!j zx*dAm7yJW#i=cPIzE-uaSDkSbNi3K_=5hN}1XD-E<|_@n&X;;^p2(!}IXEvI|}SAuQUNa<=BXoKQ_otjcE7jh?dlx$=hotKXu=byu;WE6c?CotT3V z1pyfc4&?&~`jR&P!UqsuQD@v+(`d2;ba&GLpztt0B?mHE^8HOxk4y`C$&np|GhuB8 z7b=61)mSl&^BuF!fp&I1`oiMp4y6|{n*)%X+c(D|cpdbk9y7bcDt~p++CI{WERu-x zxk=hLGIj^k8wKdFYZts%jIy;N51QuRUdcPlQ^e9KakJl=6*hSzIwog8*Ez`{c1I!J zb!s?;HPI4G?I91n*&uH?sh#^UIBs2AAx`Vmk*AniusB!urNU)M%_7q7L~OHKf3B{7 z)mV+rMewl#y=;OysGd?E612R99cA|C%{p(Sdy^6V(bd^*Var_l0;vBasOL^e%`wS# z0iqN9k~&9Z%9u;B$b_LhF+?>%BVUp^d5`upG^WO3;YFr71ouW;UNlzQZ9OxB6? zv$n+Sv^5U>f7m;#u(-Bu&xai}I3!5$Bm{y>fB=OAcMZYa-Q9vagan5I0t9#W!rk2q zclSbjvCqAI&p!7%w|BqwLw{dAR25Qdt{QX9IfwjzfuA^Tjo|^5N*;E`Ya(xIvP(krP^*v~n)4kx-K+sqlbXMi+XftEa z^EQRPZO)BwlQd8cfjiOvsFc9PmbD+4`h}vk*5S<@_U&)z9_WA5n8MAtf`PZQ6w^lr z=ZNY6@i3PCGB1^N(}Xob-4_R(>g{@P!}j4{gQl(T-9CVKQVBVmR~6X!>@N55ceS-h zsLRG|)S>y(R6Hv0te0pv0?7E>E=Z(SU3VBqAHK$~)8zb!+)I~yTqcD*J=0CQRK z#*P1ZCz(B-z;QFsbmZWNy`|J_Ck?(k|FBji8s=GOAONOPM~sdN2kN&OsZl$vi^vn6 z7;f@8L=}5VE{eLT2XK9RSHBk-_oVeE^Rx`JOwWQ0x`!6+1f%ynb6zmPTBzDeGcDT2 z%7K*Urw3PXP?IVP$pUc37xM;qPb|(4KW_|=CwUl=xvnCDlI>+QUqG4ttv6gz7})^H zH%bm3K-{wz!!*?RWvUX>t!h-O^|4e$_v$PZ578oe`pDIAixQ~rm!a*sswS!LtG^&} z{B%eeX0+o3Bx{#cD||3&qP7`^^FK2hwWN!W<5f^e6MS&~TB7&$_l?laCyW6jU#Qa} znYoqO?Ty}xKM8InZmFO;!~U*Gx5Q~a|6f7WNZ2=i&e~I!_!dN6XH%|HLPP@RY4O@# zn(PHV@bB9OW0kci_NGCZO}c};K)hM3TWZHo>;rNI*~IdS_jSY3dZWcG{J1D#MTzvA zHs|sqz0Ze-?{QL*w2Z=hE|(Q7(xwjn zFiyoW-Qyl_u|~W57Q_UCa^#5_*QPQ#R2R#SlI-`$y9EqG!oODkv?RKTwS+isF1{~< zK6#Z=z^ia!*1UNo+q7TJ7ypzEnBq zs+}LjpUn1C+v`EqJ)Do% z(=ESx)q^ePqMps%;fF2?G&$8<%P5btZ5&v^pNaRNsQJr^nidYxB9sIspd&j=o*&2?X1MEd)->ssYCZLNC(uwTqB9W=N z%3!HBXBD)CgxUT(e{0-q%m{xlW3l2Ho_Drjke;`Dv-DjxcLnowCdrFB4PiFX`XtmP z4DCqIlt4uIHe3O@^3KLm6&)}B0bG-ug8@INH3Z`&_r zob$ohXRNl-K|n>Cj)>g~Mf#du09AmFL7NMD@ZBwuhw7K&wbepl?4?^fF{B`NI(b|W z6{Rr;QS}QSZ;jgwV2FlTFiCUZc(G~?xo1FbP^q6N`jk~dm_78p(y6K5Xj1Q~FKX^= zM}|!pF2jjjWApSksspu0&2tw=G<_}YfnoQu(`TI;ohNE)aoRPFz?PxO5?JjoJ+25$ zrAHU-_2=a|`Hv1ueAf%swrFV<;rnrWke*-vgmn^Ky0d3ANC=|(V z2j%y=aK?2BN?P?gt+3E`Pg!rhssvF>G%ivcygqgov|#WZ#>X`=q~0G~g6W%)Jn^X1 zjb~wz7bK$+qz9=iMSyV0$Jiwv#SCs82q0tGcO1Ruh?7uIpA-5{pJ*sYs5f*@F!LXM z4^-}s6IV}40UhGw=PZ0H+*2MG36XH)EIRyq+}6!6PiD49&C}^r>jlEDR3f(m;wiFj zs}C9!cyQZXmfT2i`!UHLWDwB_Da}*ewXE4)bQn#?aTI>v7}~C}dQF{3eO_GCy=2cO-jhgwAIAi1{l^Vd()f9FDGo6;jl=_qH~wSz{xY4fQ{K*bP%=u>cD>s#W~m9D8`Eqc^Q_wLK>Y1v^V%c&Jydo5Yz z)UkLir$jC11IOu|$s=0pEx0DPog;)84yVUof|(tAfK;sn)rNHBy{^nkgZ9YTRVc(o|M2j(Lms|GSsIRFmM^lVdj36e73 zrVcUheun*2 zOGPC}0wVxC6!F$!VtF{WT+G&gQhDX!s92q&oHy}K1PA4KLxuen_|qN&Ts4|E?prTo z_EDmDKyT+4QeU9TCnfC~yxq8TX=?Ctg(#Kqc%} z%(O&Msof8M6++K4W14t-RXHxBIR45aVxYmw+V4KB04oaXybakgvQk>~Q!zCAg7$5y z?xpQXlPrY7*JhI`=U8()aNZ%Ey(FyT>nwj&4A1ON=p`C1I+q<^;_b3l0!)J6{X+o4h~X{8m|at^>ExPlzsW zXc=E*o7ueXP(yBF0=5$lK%wwKE?>fHIK_9@v)g4q4NhZX1Twz0Qt+; z%>8;i)&Pd!#UqT83Jt>_oH#q5+ga7XUPP2{Pk#H@m^Cucd(KCo$^tSM~ zQ!+L0pdN*CsaKj9$!wPC^SMTufXhKHdX+yhb|&0@OEPNTI2XQIn;r4Z_&O{(EzFG8 zl_wr(kgb4(4*7o0ky+uMbuJ^nk^-Vb73d~Im}$p2O_Jyxew^H9cb^hl=aEm)8I5bM z1hPV$=LuQ2c4S!4<~qMg>YFYC8E)pZkjjg}W|Oo#<;lzhtF}QVOrDHiE%QXrP5jhE zXy<#Y(eL|!R{wK1s&kxX+orbMtMNUM6ilL0i~aWO=YM9lMYV8XwNAy5oLU^nKojk|+`Nho0v56Lh4>Lx>*lv`OgEBUEG#p_N z&Dy6m3^oarR2@XoOHC&gp)y12xO`({wXXoBjY%qy-~8?$JM@nQ>AiiY=$V zY{vbzSuF_yr1+h2A??SPB6Ebo*AHJEUoJW|W||xp_NwIy1d)L9S5yy-jum`rUijlU zfHGFA&`I3D4d-!_=2JVJz&=mEOZtbnY?ecPwe@W}&e+_1&?Gdz#N}(fgQP2yuFoPT zfR4ij&~ZQ)m##bw>&hjQEDR{d%~Fd(5OF9{zOo`@w%|7)g5aA$9HZeU@FDoCOawFC zcDILA9{TRb2*wT>ZELq9ze*fy%^e)(HUaj$0}Lo?-^P&euhD8WslH?>zv|X-L0*d( z0&sEG%G0PTJ`n+}t4}?`a~Daw9-sR2I5GNI`{3II;;FU1xN8x?fwtM$OgnErf?r`F zE*HUDGdnox7fUsaFJ$R!({g{z1AI$>EN@XiPje8Z$9yz}6@%?hpdCgeYczbVX)#kf^8r{V?Hp7WvZxOoPh)F(;$D0UJtbv^u0OL$Dwn>l)vp(KAOgqSw zz>rDI)3-61GsGOKDA*N_#+!uiCvk&G9Qdv>XbdPwR$Qb~3F=w|Hb%RjiLcRz1zZST zpRMV{V-BovsSiRdwZ08JXpyvIXvxB0THxL^+4vc?<(D)Mnq7OMe_&5QV zi>i}*CVxeQS`Nt*Esd&(czv##VgWWh-Mb257xe4Ge9?$E`l}S^9h-J-#JcS)za7TTy^^1a<5asoKZgz|tX4HmhZ?CaE($dAQpn9BEWL&e-MJb6 z%P(?d+~|Buv8RELVH|iHRZY4?HBH-!=O-Kvvn@xGbQOLs_~|ZCT5b1pR!w}JwCsi6 zSeLO{4L~$XJUXeqDAR6#kJ*>nRI*zS?-;7M-rbTOs${+cx9Vs%{eASUZw*_tmk7;Y zJf-`e!P=5Dv1%=my3BtRZNB*~ReMHm@6m0q^~96|ha&(?xtmOYDQC`@Bjr4uugMKa zawx2qVkNA+^1234m@Jk3Oq%F(W?jp=x4Eas)j6iVE_0$N`PyV$2JB%T0xtu%=mp%K zAV{EAGNoS)E7AuagDWm3K|`#u!DvrzPz`HUlvei~oejUTjCBa4_;9y|d8WAhc~L#& zCVq2(P&KfwHAgnJaQ;dq*qNuYH(ZoEu63M<6k3Mn_lMR@t-SS=s%taUW3Q;W0qCsP zNL4H4w;q>6$I5jD#oU;eG``6U6uP2XR4s`6Xu70f-C%R5;c7LRpId1e^(a*0&LUt} z2)g|p0r#B1!sm*fpzp(yaKvXU(eq~-eDil>9ZtWHp;XiZb)Kn-7eG?{d$C4i8wZAY z7*sc8NsJ>C8QK$u?M9*V%kLTZEBC~m*=4Ych4BDh6O*WY#e4MgyCv^OPpN?SUfNhJ zmWwyi^?wK;r7Oyv?{yrjBMm1?C>>2{Q68t7TNmX?@MLN!~_o1Fabi zt^Jg~#_;p0i_}VU3TmG-PLp}{K=7*u+jkzMLoRZ(SjC**j%Rv{0#$*m?`(`^I)^TY zltQki@_x5*#!o~b4ouhTjN-KRbA#NOp26vrTI!amG#b>M@BjP}_0%g{EMeipf!TKk zs0u_2%QD0o$K?!|hg=}@SPYar@3g=j&e<`5pHH6n_H1AgXl)9@+}TP(<2tdf%?k&V zCos+#z;ipW>CtC`IX7l(B^+((sZMHm^oM{Sd&~MKRIc4K*ax#FRw}W<6X+zaYC+On z7k!n4yjr7p&HfzUhVg25R2VB8|5v=N%-lm>AQu|C{acLd zNh&ec*{2QNmG!Rh^3u&9LTVzDL#49?R06e!Gxmz3_~hZ9tgkZB{-Vj(azr!SOhKwd=iQ*;s6?uHb?E*0 ze74TK#AyFfEl-t8 z;`wg!8@XRRJo{za#!Vy_@V>g@7vW7b+FG2syAPTTS1&rW(0;T(d!L{d}aX7I;2{!Y# ze?mvhWxO2`(1^NXGtY(0KmPDsn#vzn;oFi!v_-2VI+1E^L0{KBolte#eP8)OYpF?? z;nQPUFm|?Oz@1KuSxzM84*MSG)*V$o@f^(YHc}>*{h&bzkw~Ydr^(SA1%9%hu^;e1 zKIzoj!yoX1JrD%piwMo~UkiOG9c*R~@#m7g{4;27uYS5tx7bxa3L46isEls; zU_Rx&bIM|NO};48**sqj$?9ilTWiMxmX{vOvo-%cpH{L22Do7~A*^~q;V&QD^mSo( zXN*csXTAeT6><=c%|Z9ybm}aOW^eaUFdyh;IB!jIoYl?!3piGR-l;#Gm6t#DPV~>D zi9AW%QiZoby(Qw_y5sPI1X3B*e~fv_e8e88LXVncqRTzLuRGm7EN=8YSqSPr{zTcF z)J(YkoOD<=Uv9L7#&*HQRG+gkpx{eMF;5ft2t@^R)cvy3wESq0<3y1N6h`5YNw&@W zikAe?V=3G3ek*Dad8*^3iBjtV9mrmiFJ@E_+NEX}n+~qm1&)>spjwI^YiPMJWKD+< z4b&Gyo3o`A6ukQpjaa4GArHgVYU=Yc1ZcgwxCp1z(I-oxYV2k1o2KPA1 zeiggyw=b-eABxL0fg_`pb=NmhZWn}<;oRBhnXgo;mIMG_zvp7Uo1)MRE~Zy=903baBTKo`2Hl>K&)_l2ri8aI3~iLe2<5+rw(BMTd-ojtm&|l_n44D+g*A!e+NW zzXjR~ZAGrk$6(j4B_i7CORq)-sYCXr)uYD>X5<5XSiSxMgMu^Z6Ad0~5u2~GyGWI@edCQJ+|97kTJs!^9H4N_Q!m5J1l8t^7F!1I+@#_Z1 zE_t_i2N$MATpU2}<4D!Ij*N5}P~5KsPRxi%XxXM0Bkj?~-v$v4aWnQR%t*CkINhFh zhd|HJiv-jc!~DjcdKKNyApGtK!h8hLEv`EeO{Lozz=2k6t5-+;J==q8Z$3B0155hivC{$=`ojZn%fh z)Vojk6KE$MlTV&cJ-vLffC(nE%G|q80;;VGTJWBagwHzi36M=&PY7U&v^QfFF@GB& z@V|iB=ZLt@CpljGe}a8`eQdfA0ogytN#^SSv(bze{r|B=JIFdzI0*ZLNLc;uml$8tdR zsAb+`cKpmCi=`q0Cbj3}!x?$Tx>U`Jh{R^Pq;`4W&hg{roD$^WCi$?I4V8JOu|_Y# zNu@DPzK{JBEcKMuzb#^I9J>KC0VeP9PVs*`tM_UDA&?-l%<+o-=E{$1|CkWg0x9si zw~pQ7;H}u|`SqyRG`{S?L#OA_NXV!qkSJ8-A%9YyG77hgRT&u}j_K z`C61+D$e!F2MbKZ<36OJi&C6SS>+?4y5ib5*)|8UG70^ym_zUc4$Bs(t0afFN6q6$ z*|phL;XwATHwDlf(ElCg$xQz22oFDR-{VkZJV!hv@gfrj3C$ zz}opc7LP_4$FBK&eK^Ux*Yv7au zdYoFn{T!t!kxWx1ob*A7YA(wR6a&z`0Mm}97&v8%ijp=uj{J&5%wcx;q1UYHQlaCW zkPHy*w)tOkSrqDoA`);lUo|x=a(MYV(E$aDKv_~3=dZ{>QL^j=rsuCqi4mE1>-S!m zL{(H{^5nj*%@d9zC&uu42;c|7c5*V^m`3vr!z!7=^x=01L~_o6khbFl*1kR(Bf9f+ zA;ZB$I+FjHV?`x3#Ht&`j?|RiWB!chYI{mtt;Qmirr=3kmf$!+s3VAD$wNKN0!zT^ zYlfKLjUzmTU-H6qQd^CZ*$-g<^b%e@_=oyL%k#Hyk_4BV>9gMK=`7=tPgy{mq+e{Dc7k(?^6bxQ9)Qq&u0YMlqm)I37Wgr+ zwLe(x*d}Ny53y1wnLOo~^u;eMih~0mb{CF9l@%w~FPg?etTq*uR`|6*76C}?We$Q& z`=l0AyPfrGgG#!0-kpXA}o@3Js7_F_ z@AKrZF+!MuiO8?ko4$s&VzKfobZFEAcAE7f+QgD{R~eZfZF%KB@kPT~C%*R1s3W!H@lsTGn~u*Z8XVSA)$to2=hs}t`T^h&sx;F&dM;e3 z(jSN&ZvKs3{4wWD4j;YlPT&s25q060*iHzzB39Urn|2dKn(gzx|0?V7GQPn=mqTP| zTGyjr;F|8fArT=AcMNBm5_|{KSfu(iT~OTj*7Zz-a6FSKd<{|hxgR5z-g!=@D8kB2 z$rnXgc={l!{5wP*4y*WPcqRhGUw(|kvHgzxqBonH6Q5ppsH45%aygdd2}TCMt`=Ke zo~OA{e7F1Q#GI!;F{IPf16;e(v(k+&A6I&!>VeWMoZw_rU0|g$jZSkJ?`VXezpfc+ z;JKf$@Y%AU^D_x_t-)5qJaFQsb1C#JBVD>jQsRj^^#$r%Xzg=OX^thSX1C~1qpIgt z+MxwI5q6yy#6t5JmV@IHvtCr@UhEcapBh1`AFD5BC_x3B!?J+G&&nt0Z44clW zT7J)&m2Yp!JhJ9PVW*p;(V=%^DD@bm>quXk?8SaZu|(LRlx+VTw1KWX2qrch|1r5V z>>az8?qe{_{Wiujr9H%q$Y|ZU&E+Zg8)dM&VbD)=^@O4E@kgN^O8nGcL)UrqRC7Ihmvu4&Y362;bb>m~e@-MUq zQ-Nimz(I0+v*xk*Q@AAlU*F)4mcR~*xyvmVr9Brou0D?=2c_QacV}9DY5_h$ARd|Q>25w;IKJ5<8 zn)HnVa0RUkZdcD>8$Sgpe#g16Y_0cy*!YSe27)^j|0{mOW(;QrX#02AnG*GSfZ-3@ zg8ErWA={*4DMVUsve%daDtZC z-ya{W92xYE_ZtEs6}RT2G%qe1K$?*T;Z(cm5m;4+U7KVN-+VV6wt8rz1U7k2z0vmq zglu|p**us=Dez;-fSH`XA6k0~dwui!^Vr4Li~~C}2UJ?^3fBSW=A#(eVyY@URQn|o z?3&nPYMi2wld44p$Tka&VL|v*OLzSL1m|gwOT26nY)q93{fp(uAp8O7wnOc1c5=Wk za{BE5fe@4fzK{xnC;99aly}F|*uAIIQ8e&3!}$UankXxeIV|mUEwan6R(^?-qoWwe zuUT(q`8f+4@O32KoHC#Ni-r9UFM&@$ z{Gb2pKmYXqf5JcCjeqz2|KFv5cW3_j{r@-b5>?Uhprqtq9UVG8VMUvq8RGj&@9zJON%l>X2 z#^272@&5H6{&j8p=by50KDmgyKUi-6?U4Rus5bdeT)X`EXQ_OD(klR$_J4X9*a41< z6^3AN`4?yY@2^U1`3n`Ws(xXF19|_w(F3c@I=}%mK1J=>|JzW%Jeg(x7f@{yUXT2< zE!f}wvk&6SGRep=Vilw}Y^4 zje7uB;<3LC5S{t^p9z$yQh%`ezku~opa9#5@9_bN02%CWnI0StvIM{kmQ(n@8I>SF z^#fBdqOuDA``PrT>3#Bc4*E|mfd4X5te9s3N9kF$B!BlteCH1YHlW;sT6ji(U9kUo zI{fQ%a%y0Dd=|Im{<}AV+y_k@xD|nobOSH{{*}oA^TY4Qs?&cB?_Va}e|t&z0~kJ< zt7^iBzkg-vzyv{}>^%6_LH_%B$np*tKH+IOlj7gMvd@6^U`T|{|NUuUAok=Q)(v&D z|NfQv0@gzk^Rnjc-!{>n@4)bbTvWXt&lq6`SCDnc<1Kf`7eC62l5yFe6~lcr#^7kS z8H|O$Z5=W=z;z4Of5&QydN?%}%cD${scedAeIW8J>@8 z)>Iwi^NF4=UdDWn7F&7`@Bik=e166OrtqENfCKu7qae{sWiNhd|3lXQ!{2mD-}&ED zJXNxzZoF(}gsk6vSBL};;6v{Ss;~o&sEO}B{~xC3e~iTD2e2%Vj3Hz{9c$kIZ*2F! z8~A^|Z~OE;QZbbbe(-RQ?p5iJBH@0oT5iy*k{HU$LDkOL;o7%w_m1-WY`>k+@FURP z<)W7J%m?cAsJY~Is|njHd6ZZJDfb!1Yrb}zN<3H^uz*GQf?K`6TOvaEDU740xed~# zBH>x!;Fhium)|gl{l$>K4?Wog{M{G4<|tyAyw*&)=1)2mCWwY|DZftsWi#wdn=+sT z!f@PdMdWp|p$zcX3fz>c%wpH~D~3WXs`q|3jWumO>y10ma9uqawcChQ7Meo1KW)bp zHRzuraXYg!loSU_>=75)b`F|(?!c8+PBe{)G_sRfd~BY$9J((=!!sPcHhxNIM)oPw7mC^ZtDKRs^T4*8pTUC&(pzOcN}Zqoh~@_0`lP)tict(DC8!^_PBC6iY?$^(%Hv2Xt2iZ8}fLdxUu zKI)zEkfejBM-u=$`}7DJHn`i>7zP)<(lYFr^Vjx7ZoHV{KJ7}rT?Z&eG1^}DN({q} ziFm&K@w9J^!H(I(sfqTl`b>v;{R}TE%rBavDteOG3K}0yzru#SAKvOt(rY=$LDQFP zStsH-?F^=`WQzM4ezFdO+AS+`4Nbw&TtBccABzIdJx~PKw@E+0zT7&zQ6Q+j=YG*e z^g=SaEnt1Ye)yz`t9PoRIV4Sp)_72rI!ZQ`q0^d%W;~UkV2&xrIH^LtPHock@jfkI zkRhhLCO(uFY6L}PY~=5YrzM$gKCudCZ$l#UD0D7Rsz|!q(|!!)2$3q|S-KeMa%P=4 z$lW_z@1>8|@Q9v&uMsB$e~l(hH3CBWi>o&f_|q?#0D;bk$ZdeK&|d zAl+w2RnyQPPqOElaSW*iT+I)!t(?n@XAEo;7m0>fUgQ#XX2!vf9R59o8vcm?VfY8=uN=ilUTHV3ckOX)>gIG0*c}yW3o|7(VxMV2Z&O$9_xnEvWvya1Bzb+3wJIxIgJmQz$*AXE4ve z*B#M}XgK`dg?<#<7SM!r{@c+j)829Sz7P50BcgXiGLvu|&~pySZ~Y)6_%cIJIIQQo zti6)pP$LQon39!gszv|uiL3s}{6z!ukmX9Fslsmc%$1qQ%1EOyx2x^EjHWiE{1)K} zU-jG&-7n2-(zLL~a<*PTyOaE$|u*4Adhi3S4(f(gD#_% zTGsFp1Nh&4L*JhkOk%TKE;Bst)XFeYG32BVII@d1>yc+eUVhk&sB+l`?J@1*8Yk`C z-(Eo!1^y+?Txsi#=Z~}8sGsOV`NAIy`QGax5mZ(>(1?9_fg_O<^CH-v zoIJw+W3a!2>#{ZVx+#Je-}3 zoKi!~*O97zbc>Ra*a*Q7SR}F7hOhS}^3WxHHnd6*Oe6Dh?daxYr2S)9D(O9t4SI{^ zro*(xV86MyuE(}U$STO}(n(O5!Q$9gJFfS2j)G0GTHv=o9DgiLTs%JYGu>864;axvg9fmJ|ru7pRSYnQqEZVaTNnP%?^aV_BA0dwQJlTWM2Cy zYRZstod;A}oCPm)68XLQ$WjnX$h0?~Dxb77CTVy#qid&LH`bMBA6}K>_y^aZ>8{U(=`nnl3=oX+K%YdFJm_hotwWL%UaT&1z3Bti{z`Q8F3beo>GTKR9ha%)nb98<|Hg^G0gcjUq={j?am0AI9v`-A-Yt|l z#3?NcZjN3hjo>i?Wws;q6QyDDV4N~vd*L+AuAdM-`rYHR|s9wEC z_7uP#e@E%rpB=F$z9l*zz~db3ftspLBN!Ml)-R#$yjWE( zj{c?WOfidvL&44WRO}tZID<#%C_0mBJ1diPJJ-lzx(Q-5v){r2U3j}TX$P^kSjt}1 zRgXG<|A#A&`g$kT4JfZDRhw(tIZ#t>7H81?z|q=%5U*LCXGF;D%56Yp=WX>H*!G$% zyycsLP$3PS!xHGitc~+biw|1$ z8CGjhFLlC19(}!zZM|>$v-3wdS%mXaMQ*0@LVRUhpT_--}_5K$xVPr7)|)4tzDRPK@uPjUuWMI6wwW9DAC z{4JJvtb%#|6aUDK!Q-I;iuwm2Vd-&o^r*tcbLqca*D6K8p>bH(kbmI6mBZ|FNz$;x zj`4%Ql%M)i+UjgM^YnrPD$x}l-_v!$7=UgA%vhcX=CrTk?v2LYs*UFei=<1a^87I)wpi*G>Aaso8j%7#C4Y7m&JgpoC#}x zoS;d-@@oUBhx*mf{zqkVzNJ5zGD8}gHCU9F1PDN%vafvRa9;CfM+A#ND1O%s;^gi}24ufy1|X8s4z z(EgN3fHkr#5~gSIQhSMmhyB*T3MVbn(b^%;(YAWc(TwKVbJTr#4EsqX>8+ECx}6fNU9#y%8~U1$^&W# zIS2}ikNJ}^a1<)~OYh8nid+!ad^Rk$_4l-N&QtAS!p~Ker<8oj_H{L0lo;tTRY!Fc z+~CxY9|Vc1n1H10F810>%CrJX+zIRJP)+N@ukNR$@kG`GZ)VGv3a$ONe3htlN)$X_tym)@061lH>AvLUWDac04nLePBhZv^&UG<&*9Y38G#HdaY@@ODmQFUm9t(2pRm4^W&5@8f7t&SUO6D=I+c)YT`ynyk}q6&_D9Usod1q< z_{X*R1I+#*{T95{Z*H#LaTfp-ZHsfLadl5cjVfXEa_RQgL4DhOvbe>b+QS^~gy>qw zbu!0=*978my7O5x@?}rFa%>K#qsI5_18yf%=YR>`DwWWAf`$~q+s=WE`|FmRUGNe; zp^`h&2Fs~)C89gdThbXBxC4TiyiJ%eZkK$O&?gmg%y#d?14VXN|!{H5k-dg~3a&yGe{bLRz!6+Wljr4vVy;yCMi*k0v)|zZB9d2DSKV>%3M_baIy(faUe#v^hDa`%)LSu8U1lJ>;^7p=^jM{;7 zO}d_6tRB-cwxwoe|M6hR5qyof4A?3NTQV`!4fkkMl#Fb|a6TemGoZYEg74CE|GW+o zVd?xyHq6r2nHZM*m-wj6E1v`Ir5FXuFiO~E??wDN5oP=HfFlF`vr~Z;b>PH_Aqfke zWJ?94iN2FA;$EU&t_EPL?H{49vEd?u^Yy!Be@vRQf4WUbvKd*ql$%Ea8MC`MV!ZbL z?9_!09|7l=iW6*E?NfWcpH+4mze`tJqQz=iwO*&Hds}z0!y&XuuTscuTSiA}WTiu> z>%c|${0T5Mu*Xn@ti1~x7sy&|9k@zOX;b2L0(LMn0+S$^bQZBt~X-TCVn{}{sppJ)nU6fKr zr+2B&3@UZ?P-UZBA;R_)oY><3gHN6{sq*v0_N4pww=78p0v}(4a%=E10pT31J&~wn zYAm${L;-2&(qj9iVii(#?cy4T(U3kOkJQHvaXQzL7%D7OxhXVKE(tt(Ez(Og@^97d zxv>r=(_B(gl97U zk9SQKS3IA)dC_*wvPOp+t8Pi^AYS&C2W$mc8|y7pK+%VFsRe7Fd9aICTYc{umTnwF z5~U^C;X5J@OQ(!7ufel&tNQ}rtnB@!M&A@n{^)$L@LBtWM*YXBmU^pO>U*VY5prri z3cS`@Bv;o?W$wExOs)%jdRMKD#~ok8GRx=K&-4xz1#czss*Eese2!-_r=Rfzk2pO) zA@<8CZoZMIyN`6QoV?>4mPj0hX2Y54lw~88tHcH!Q6S(bx43L#hSpUO4Z4 zR+^zD*l1=x1d4u}6%jEBf7h%rS=n^h69BaCBdL}3>Y@XW(CXLKP5ifm3_mGMhQVy& zs0%K#q$Z16=(#TAk*P~YCS4OKsbu3$uE%ZVt~X7DymlTcM6V5B%=P#oU0xcPwY8AX zPIDB$XsChtT;+D-KHs?DsD6Lb+>`h=*$RKgG3F#lUn2eti|B-b^x-rZFQ>J2INeh| z!-0~16Gk5qIq7|KPL_{oLur9-(AMTeIhZsNufk4LvJuh7IfHl( zdX`J)k#p{l~3#Gb8M0@3sq049QRx z(akBmtFI-pWV#O68&2>CO=NvwJ*PV_%D2BZOPlpc7>rB|V+zBQig%tUoTi|`Yz4qF zD!W8rudZ*LMwmXa%virl;Y;H~kWfL|4opRWDj4(+7zu2xBLvs(xI&Pok8jlX-6d3x*>c?8_U$*L2J34(rsZ-!9fGkbD}l!)|+BS zyZfbxdlI;faMI7N!Z6WTP>V5I#qZOo<^a2x)(3iLwv4~mu~O~D;@PYSvT^s&VLz57 znfh)$FFW1!HdcyPB#EJR)_D_4WrHh8`jO;1;Wzw=`{q8Zo~}3gZNc4f6AkL*E>5-- z%3Kif$uUa;Od-p~)iU7J@uV$LE#}yc$bmliIbzCFj3EiTRuPq7@qJ`8H;q5$>2d$t z%~Y%?@;8Kjt#1e!n}m43Uijy|C7&1sv#Cx~b&*7#^g*cKI?he; z+%F)fomrXEdz3NvWM9>dp>tD+ykky5s!Bm#4CVHZn-r2-E@$)`>!F2LJATv(0rlr7fK} z9yBAoVg(Mpw62ueGDuG8rZ1Pme0|ulJop@mck~DgUbFI6 z0jNq|PRtwM+Rkk(@V9n1#P@KX$eSPNg}9V3Utjl(mLL=;x9IRwePuo&vcZ;o+H(%$ zE}o~`H6qn+ZR9sI6wS#xfnQ33Nr5e=6CdJu0-Tl!(o1>;F>mt(;I&JUYSlZ7Nn~zM zEhz5tUa{!UQy!DP-!W!kZ@oVLB~N6Nk76(ggI;mM@0)Zyfq`msgip9B{jFhZ!+e4{ zyN|}Ak^aYgRY#AV%6x(Qj}qU-*z}?JTX|roTQB=9#utXooM{h_U2zAgww$nTvR|u) zG^QN+&uIn*OS&Ti4s+EM=NDr##K)Q5B$BC z$NeHn7IHTTNE){A7_X{mzwBR*Xin}0F5T*+t!>6@M+MYh?H^q$gP%o$>%2h(38ias zAML<(OS2ne2ia}uJ@~U|)+-p6$58T2cdEK1Zt0c*;q?(9g^rQC`Fo(QC-E0!`rsIq}nwe zkr`PGk_}U`FH_{+WMqp6kP08c`; zH>!Lq$|VVP%JIVAcZY|uIe8Jc)X7&=b9uyU6nv!`B13LxjB2rJFDMB(wj%dBAUhyc zn3w&uBdR&N7UIu>f0;i%TgGY|ru@*v?CvACB=B3%MOliM<%&u4Cq@heA#4fH4uY9W z)Cy-y@7%pev2MS{k)%J}eQN@NS<`3sx0u`!aJ*++^^_+WAM&JHD1qo?T0cntu29X94 zBqXIfM7pG72q{5nkWK+fm1d}62m$Ht7`lfcXK4Nx_l}?U{_ST!_kQy~o)^zCFBpac zxUOrhbFFio@%g?ifvHcHd^Ksiptp)SV}D-HTo7ytAb zKlsUcHFj)P^TDJeRLCzDY&$=p6RQ8k8E-&v3RONZEf4_RN+1y^X|*$p-B;i}GmiIA=zs=2HFIf=>F(=kRs(QE$8$Z}MM zlUa9=U0t}9$1m;BXiB$P)&Q|1X;=&XOg@3=&SYI1i{=j-dXZ82leOoRPSGhY4}+s9 zFJCoCU8u?~>b5mKk9DT2;#aI)vpO(+m2g-a{V01O&*|qz0ZDty&%18)OP@SC8&x`# z@ghbBZXpZT;WY)*ZTc)ahzj7*a#oXN-+mo^w& zd(pTLD~-tD0W~e>A4_GtXn@Uc|6DFG_QQ7DiQ9vicPHqf1l)GuwdfgXyfdYdR_FZY zwDY&mh3D}O`@9tb;=YX83Ir!GT6Lx%OnqR<9^Lp1<5klL&Mlmk{fYU>M+Y|B?Ro`i z&QFBQi@mOg=9|G-O~<&H)msrG_^(Sg1t&}$3kELjaX}+jJg2Nhups3|wK5Ke;jZN@ z(}6_MUqN2)e|ym$A2jd1RuO^eSczb6VtI0w3Bc?HTmAKXj99;rsGn#px!9yuJ@%1e zmkA~2^-bFK73G7ZUamJ%9%ku3TEn<1=u0m!rgi<`4S7m9+uKvUU)^Ww&?_sjb15tY z+35}C*l83`CnBiUE^#rgl@i`LnKEsxb#_apn*B0@7l?I{GlIEUcDYQMp;^}HABL#$ z9qfeIV+K$!PWn0hxcibZjKR*esum7EVNMDVQA9cFpZKjCe_8DZ?M^s7aXB|MrK0NK z5=PKn(9ILZIZ@HqEv`Il8gg>nnSBw7O_C8zn6-dUL_?wfU~yC}};vNcHsKp+ERp`F84w>BZ1yFsi7pEyD+9zQXFt z-6kQNG`pClyp)FpLyGJmEn8(R*@QFkjH}bVCLwfpa)(3SUm@pFtviZjI1f2SuJ$>- zKc>W@^NM8dRxr(by&H^)&9r_$X+l=pqZZ$v&2W(Ab;U&KDx5Roxp~MH;+A+GvfO?< z1Jb8oeVybOFc{8egp81AJ?dd7#Bp9%A&Uq`ouEkOU-0Y@dbqxuK{$yXa?R6t7%s1w z4hPhlWQWX9Lx(;(3eM0sCF!I1!$kM=<*|Dm8swmKd&%e1tGBwz7&@d}k@qmEjyt&e z?FHPn7;$G=)vq{ZpI5-0#LcJKJ0-gPXXZIALnFyZEi4-Vij3}wUu~ONDjGU}JT+xd z_o=?@EjIvnFZ_%hnfExBQXsr6#BLJ(5N& zo>L8*x0Xj+<$k%6N*jHMZ!ci9E|>eUbfQ_9OCt8i)MnV4P!(d5gI%fo=1{EnRgX$# z$nIeDdfu(HB|**%NY*~3zXD^ao;40*}8)_L*>+KXsQKj2!Vp!_oFg_bHEfK5?X2HKM%zf={GH7CR z5603NL5BBbqSQd@ihsbSuU@+~6AdFV`L+ZZ3w+E^>JeR78aJ&XMa>X(?hUg;U>9LD z;uVa*MG547#FPeMwTH1`JV@Ra>|pqg1oX5zx6cP>`NoCC>dc?a+wutlls+bL;L6lh ze=N*Q{s_H^=Oqj)=+7!M52%;ABz@~j{mqEqd(ny#BqK-iW%A-M|y=4{oEROe}^V;9S?HWhj9aLyl>O zo2p;gOL`CH51Um|p}5ijw50dfwMg;owl#qS_(~)QpU)1?9VotQFS1vCIkhgsEyEms zX}8qg+>|s?m$YXe&Py0Rca`6lTyjVvydJ6Zb~vDQ5;%mix4VT|IEaYYh26{NmZgj^ zY}84f$~Owa?S;I1y4l88F0!{{xwYF{eu&rrk-tjr##Rim;fL*Q6as;(FKl!JX>OcQTpo z1Pe}~%Sp03Q$N-q9_Q_jSPz)`lj(L>nKtyEo=zy-atgH`4w%w#5ILkc?fuCgR^;SJ zLxg$q9afv*cv1s8_h~P#?)NRX`DAa2fp21t-oOVaJkI>6M235JWzu~(8_Q#~8oYav zd^QI04(wWHnYlzKBq5+2soaZ1ku=6=fe2+)b*&2K5R_~mX$`)~n1zPBgi)5+4llKo zS?=QZtlX}}BW^PG?DD@AuKOK>uLizH;D@Y>SGhpPTb>UNAa8d!s=}nI*G6&sT}JI5 zbh%q<vEBPNff{tG?8DX1c$?;#Q7GzkUt; z-7s8WvJ0!iBuFySoh)DU{aj;+LhFv52R1x0IN|x(5rHFf8h0ZAPqhgwi}Hzh_qB%)RzpC%`e!yq>*K3uoI>jTy!|B7BS=D zovjIXjly^BReauAfDPlopUDUsx(Cw{9||Uwzfs_r1<)aqFA6sAzvJ3>D4?obC-hTG zbg7iM2ud6`KwjKq{*H_|^yDcUS#$#^>)YhC*PIHGt0Hq(+znWwJN+@Vi!k{&kOn7_urT<1&}K<*_AR_%%)se4k*`T6*cg6vQa{9bsjxj1ohpo78#4< z-NxMECb^49+hplf>)aJg{$qc#7!SQ{oLgK1b9l;lJKA3-Uk3N9@$ucYYqVHxr}s&- z%$gMqJ4wa&`Pmi)J)dPakC@VTXvn+h;CpGwiP7)*ae?sbAUPa=gwusiH7!Mt@MN|J zUVpg+Hm2s$m!kDGy6WZsIU~x}EBJIGC`!H$s-yOdMRYR7waR8@Xo3Ts;_!!Adw`tj zG}2;fb%lNE)71G|lg)L1+4(eI{alYhuH7>Q#lZwOT1Ii|fC4p~LwQGkxIC({mYAst@No(Oc#uxU*!nrM=` z^x?O0_~bh9qI_*uPPip%0AD$1iJPUjMrG4^ezwPG6RW2oZF?=u8)LKOlcbTCd*_d9 z3|&c|_V;Mt_TF4{I!C-bd3G#}F=jgMr7GH+{nW10q=7G}RlBz({Q zKxvR@H4em`ma)DEH(y$D^Kn13yA&;_thc9$2As$rkjwYJH0d+uZqD0BI^7zFzPSA- z7VjLdSE5Q!?E$y(`)BHzB_>1BtBF$Shi_X%n}2)7)O$6nTqrx{2plNA!h%`SnN*g0 z+0dxJ9XVhj-akflwi=k%F{w`?K3_PUY-hCwLf1ZpxA_$yqoLPV`WglCKBzgA8oa-o z*Qr&G`5sUZ!qi9YsdJsl;}$sk8W75~%(?&~V>!X*Uhrd%3diftgRX+Ct_fp*hFXU- z%v6EICP;%&k`|;}{L z5`JD|W21_9`S;nhCY^>0V1y}Plk(XZ#>&Pfrv#TdaqFRCh_d>oTDy76Y*9^u%MT0O z?k`fta*%V|E+BUxG67sUCacyHI)}fdfNZ7B*9Fy>U#-NrQ&Ep35z792z+{B$YI58E z8Kg!Rh`~!XczU(J0aD1EzE_?DS2-4*uRh?Czr9JpZhj>{K1NAQG6T7@!F82r-&e&j zmjdxDL9hAh+Y zs?}z*>x;^Ygk$1}P?Pf+(E~i`FUI$!HPXi-jO-;GdR6Pd-u%AQM~#N3%Se1d-Ms zq?}qtqwLSUPFfO&FUzA$-A@*^$0rBG>%0g7IfKmRjOi!O#m_e01CAW$&z=RL8zSY*PFr4p`>d2I!J=QY zJBZ>%fGuxr(QiF!JjA1zWtPG?mS16?KNT3l8@&s4opkWO*tBF-Yiw)^M%_ldtjeDQ z5Prw0$8r4IN(Wo@q`AlWbA3Se3GGO>o#>%R zGkTW9sOGb{M{#Ti7={**>kjanAn^6mQN60`6LEyl6*`*KAUxF{J+uBkP@qK4obZHc zMZ~w|7!yH0FOck2Ss*vXi@n+A2rQA{A`?g4EwWL5E0>lG9ug*H&C?a~%fs(f6srIn zo=@vzlocX?q&Gb!ex=6wE0w^nb1ER3ixA?>bm^aRR@gOAYt$hsQf>P7nN9zdp-88b zqjKQV^?H{7Rt+o@jC!ZxgJVu~nb zCE%>zp!Y(+IUGt&p+iwGLC9&_(pVt1<^~+ZcHITH7^|1!+)=$JM9v=7Poij0T?lAa zj8Dj`l6+k`pk@D6y_@e(r?}zOMi&ORH^ZJl>G>~Y-?m$3=w-PX^doVj8^t&K?X}@U zAnh!$e$hq8V1Cx?)ei*n2WUeaISOGZPZHSVu6(>G8}ia|Da5t%6*Im<8XGLb4Se=r zz6J>#d-qdLy`9@y4Y*Pr&+QGJy1r6Agf23B2mSLXdT+@Yr^%QrXWe{-w%Ov_iE3a;9ACTv?m`aNgBAJH{Z1fuN~JIG zYfRg)=(3&R`%q&;t*+=6u)>cHXq+41-J5XBQEohQS;o`N^W1M$t8tYtQai;TXk1Hb z8u`lMptC3Ur@n+Dd6S)!E?%H8tqJUN6dRfclwA72M?SrLwej*u#=P*<%Q zAdTWD-HD1Akw1wdMQiIqY%u%GQ01eY9e=h^+S;Zat?=Om^}m?=60B@6vcC%a@Xf~Z z9CbqIG?T0ASv%(G&g$59bvBwQCKEOLj_1kT%(gezSiqEw+uPO?w>8gGoca&_RmA2` zY+G(EaZA5RW{_=)bAC{~**|yXOP#dV_lW5pM&ObZydiA5@@cGLz5*9e$&AC&vIBZu zS$a(SK_|p9P5^C z37ABzP<$T|4nY;x#hF>mKy&ErwyIlEwn3Q3i9q#%z_@8E8E1|QDg9I2gZl?sG5aLM zHlsF$hng|QU_(Iuyz8h48|{rrlG^Enuki&^Ar$@$aP6A27kO>_?}!napnfPhFWx1c zl*G(LZSl!H6bCEziy%ZfXrF(pi6r8sHEFE7)U-5_TDGu1zV#Gj60zP9FpeJm5m1X3i^(+K3hrXI!r28mhBk%5XmD$a^u9;c z)L_<=qY5oMBGMk0Ld(Zk%_6AKkHu{WUwltRB`chf$#cu-@}ExWR#@#j3tgKdgw8UbpTAQB(`ffkzeay->8g&<` z#BF4nIQi&~TK%dY9eup5TT_0R?khHYs?b9!IPjR-I&y99WxAp+6{}wjq(M2_qxoEKuFo&$-liQw_u6q%b|VuFbNsWb7p4bdJ6bR>KOPx*e+@@IX) zWf961Oq$&!(5@08S6>9F>(WCu-PZ#-t~lczC_ny&2IV=6z?4+N!mZI_I`daas(c{l z?nnU(C-fsV^OCS>sIFd(6}xZdhDBrONABGzc|WR9!sFf2EZS#LjRSLyX!ju=zYeJEO6gaKesqxo3HEuZ!y^({VsD1HkY#5uHvf>G8W9d}8CAfL;i@(1$FV!;p*0 z-eqB8`uB4l2-nkUyh((Bn5ZpWr8Gix(1LW|duV-seD`2|sFFM4T-gwiU}4j;CBoqI zv0t8$r~Le^?YrOHw9P{$36a_|7uQ$0E{{4me9Vq-;Z_8+Y~9AY?~kQS*jV|t&d4K` z&X6$IBs1fwRv%MQ_CQ~j9G}!>H_1sfKLI^1V9S$JOkYDBXK}o4mndDFZSz3rZ+E+z z+2JSq3V=GKH%l%29cG#oL_TWdpRLCJz3f`yOWxZ8Yi+UH)Vk<;Ga z%$XtxvPa#f=;Ix%(a_`Hk>jh~s<-c)c9T7+Y*mUkoj&D!i;BNVhUUo=R_}6ycuFNUq2gg>KN6RNfJ>71^??=E19 z+b3?~X(kGKJd5Vq>-+9VU$UiCF<>M5frCj$6!jB+TaA%*BY(S=UTStM@J;pJe!%-M z$#Akpt5SvqD@9!06=imXJLMbSp4e}6Um2`jv|e-rH|lWUsX@yQ zCi(wg4%C8`AY-A}eB9^>k`iPkBX9o*3}DVsUr zWNwn^i}*)Lc+#)AW@g!6y;FzBZlw=GU<<}tp?)sPbe?Pj$6$e&azXx z)kjqwGXrtHri?N*l5Wp$S0Bi)Iu_>IMeGz+&y>M6weoz?e@d9f=%LRpd3X~;T3CQA zX^Yx=!IuB4Z*$a2dbOP%OO{6a{d(R5ykqjYhd~g;;w~8Fy%zi0EXxVGI8(E%q$1!+ z?S+#tz6%uX5U+L2V-7*Gac@D>-PDS6DqiWW4O)K zR1C=YmBjUjqUld`JJioGE17eNm82%k9u5#8$aKy_wv%LIEEi}%P3Nb0^_g*}6WdLG zn`Lu;mv9eZCAlP+cpVViy8tptwjO6@IE3Cb($r4RVQGW<`>IWNbb(e zVx${CRu{JZ$1YR%8j^FY~QTFmyS3Mp@!mj+hJ5mT+x(xDVJ_7 z79BgMri=A+4gNC#WBIVA@FE8u0EvcV>Q-%X4}6jPc+V6SQEd-TQh~d6x$y`Fmbn^3 z(cwpZ4dI}UTxIa`u6XoU#g`Nn!-myVnKV1+rPO$@jJ>VG8-(^fj$k}*UuyND{r)P7 z$L!673FZr|{gLsy+LEoV*4QEkZ)9rmF7m|}Y)6j*f+wMa){esOW?Mf}8N2S~#5XHQ z^X($_0yPnS8m_1<4mJp=6 zBy1UDf0jYoh&~8J{8sp|9BOJ2Q6s6xrNS938|h&9ax9Hv=mTN+;BCuL$E;iil54oF zAi-9zQ$yqtMb1Ye=gx?%`)VbO!Grr@KDhfMJy0+I#r{$kq9Duzn4U~%kVhv~l$UWF z5a5e?EhC3WT}s~o53UR4_AOxgdTWe6AI+R38im))FK_RvZkDwafyp23eA)CCGZI)9s0dW znnI_grUeL2SBX{>M5QhS0_&-mU80=HsfyRYLium6_)%riH`WCBOAM5w9v`{@Wnbd2 zQ%$7(`3#=WJR+PM@9{8F4rGwO*187x21fSO)ns|OHdzNx=sTDS?Ebq z^*YtMxDlF5#C4)QjnnKGeEtciV3x4jH_gGY%Ini4V=2Nh#ndH0ZP1s+jL;9r9kni5 zb{?Zhn59socG+{1-^$?c@OOa%&g6}HC(CJ^T7r2#m$_0OUZfSn`+!q>npj7rSvr;( z_t^f|`HzOa+lAN!*xw>a^rCsE;xR|cO2veGqf=8;Z7j}@YUjyp1ve;z5kY6ofEa8j z*mQ&6q!+iNeW)}e`uxb>xfUOW+ggRlc(mH+E73sNY+}L}M&z|pCJV7VZwD=CBU>}{ z_GJo4?6GEy^A2?Y$s&JIsQEzY9t1&ZOzni>{YA;D6FcbmA`=RCXw9>ZYWPgpp%QZNDmiGaSq$b<`u%*s0 z=?x*coZt@|N^{*y7S5;>2^q6udfLgP zEa<7|u75o?(4k~D zX5iD^yOO`ea(SrPNB#DcMQ2Y(e`222Nyd+VNprq`&-B49{ab)3@m$3_KYbzkGbALO z*uwS&VZg7|2T_;bW;6CW#4K7;Jb(&Tu-FLHea^x=F}^NKX!m@>%rXXj7qR>Jrr1K* zDB68vILZWCmpIS%s0_1p6)5cf%7yLTq7eh}mzc zx2`=jOM#nxWAX^YI>j!6InhjTprzW7_VbW*qDAvUd){B38eC5l%n-rDEh%d1<;pd}2~l zk@PFty_|eus$5}ME1bs*d;htRIZR+!D3b;wsFe$Y&P&Y9=X}g@_Ekd*ZNc{$kB~Vu zC)u=HrY7(6wC8HvFFRCsFR)9VHyggfhOtPMccbqHAgjbSFxUr;0;#S*)qIN>CB#Mc zpelddg>rHriwneD)-uaSH_Ja4l}tFi!w{a#srg76g%qe2D->e!-TcDgP!PMmzZuUU z>-?75!0CLxQ_(IY?)b~*uzjZbWb+rzJv}zrpimAL(TBTFjsH--=lwC+`5K!ns!oCTok0QXQq`Oo z@O?bXVg}=zj@7KvN22tkhOV(Yv-R^+Gjr(X2#R{pI#bL6fB0^j2`@9T%2r-8Ar;t*kqv(8@T^csXSsiA63|*tqE@(J z@15nm6Fph24uV_@BZD=1V!9PYT5XBM~$q zu-UP_0NSiKK#D1XGBGN4X5E}SMB;wE+>+$(Km-nADL8vKfw>GgKc2xgKb6_X108gl zmbIhbVJ{F;;d}R12H2=Xs>VRv#s$c&4!(IX!tC{%06*;ojPtQK5|)ig9nwpK)gT(( ztBZT^S%~oy&a+jb$s+U9uae8-fnwURP>iGPFEGuO78L?NCJG}jrOZb>%7~1J%XVXs zd5DR>PBL)XJj3`QA-Tz|4ERbj!2!Nf`c*asJ+=mIF1A1=d{ZOny#tJvQ%gj;SM9jsZF5?1bWxP;2r4x4J2f9(|LCChRn03y zqO3p=>4lKkU_D(7rEw!G(PmGat#c9fh<+ALL9jo$EHVtT-(B^P_KLp=SQ`R8Jl5IW zs%Czg@Gzhf_4PF%tpP@DA<=hKKbRnnD$I#5$WekA2_c0zubGY9#u9kT{8d=hi-X_x zdVX9hLKpVR8Hle}=3AWv0@+{S@I8P*L5qfgH<46^aIw~mdIQ}X7(0i!ohzRn^yCib zy%xJk@3=3|>x&GZ+wYnq6WtN}c<07`r zNon>;Nq|$mom9pobYQt2|G7}$#iyD=J^mME0?)Q$4QB+p%~Lq}TV&j%eW-T_H46x+ zIUDz06oLGe>VOk}-0Io3G`3NFggW1urtWzxKA&Zj9ncytc)2?zfc+{I-Y<%YqJ{yL zEkOrGbx>=w1ys)65w{HMo!-~qSLiYtki!ySAojhZ-~1VK!4+ulD|TLa0ci4MJ9sp= zZ9CeHoE`y99X8iStEt$_K-=o-ab!uA$RXx{R3G)YIYZOS813Eq%8L6<)$e`w+mrri zw((sPy_~n`^NwfNe=72NhPS%gi^P&rsV)?yg{L}v>hkx(WP3}~b*ZdJ44%v;%|$Oq zKGa7&Kds*Go8$FgBuyDGQjM;<-HI2WI=om7jW zk@$gvIE+^X6bn4rSda86wg~L$*~)m|0-9>nm0BJXTB6bndbzsLB6X$*B<8hhvg~*=d2osS3CRm^f#HUKoow5;njvHYLX<-C%U#8Xo`AjZ=mv{@!*~$1+G(D zrdJ`C31B-#O`z>(A6n6F_j z7331`UJ4f|>1@2C5_fKVc>f(7h=X`Yb6ju&al_p~gBqWgIsGh?=WfKZuV#R>A+wFO z7Kbn`C7J7oEK#o);bVjX>wTH*R3gWO)dM-iO(Jrk?%zo$i=nbnK8$^z{Q)k&Nf{8f z@hNG4fJLQ(@ACpjwhXm(5@Iz0G{P~%-(l`k^!}scBO+h4MJ$fZrw#F%u`&CQtkDil ze50&dDdb>P;}rE7$kO=d2<$L8J)er?Jq#j1+3p_#v$%C$2rBw!H!|wo=d?l!6Iyq3 zja6Rhw`SaXcM&Qrfv3HUzKTT`6z=i<{?e~7)7zy^WCRHbc?8kvJz`p1d=&9&BrQQt zEjs(bqtAn`XX{~XrQA(`W_Xc$X8<}FYl3lF3$~RDz=Sm%RZq;NKF0+SJ!^OJf>|7 zFZ)0^Q?JQK&a-7rH9J3-=9bnTY##-nIC3%WFgkcF>bm(InYJ|e0s-vlQ=su}OH!u8 zF%LCU%NJOZw1+3aV!JhJpWkDEXv)vvzHy>HYnaK|Hg8y7{EEu@Io zpK?};MZR*{U;My)bWgHA%B=whSmnk=F~NIOwll&kS*hopfQnH=mG%JV@KvmXw<-y` z6xJeH?_1o1?ua54=Hh7j1lEQ3XXuDTo?RcT*UnQ*Q_es_D}rP_Id!$Mj&zkDrjbcJCk0ijImOQ`fGm8;~o-ZK+q^iBV zR>vmLPtEsR$sH^Mptf@p((_a2nJYb@@NYi*LdM`<>Iuc;#MO`h&UhnT)C(Oi0#M$A zH;p_j4Lqv%+&dQ19K+7$C4~^r%JcYBJtmrr+1d;o@AF`PTy2Ol30D6U%oH7EC` zl*Y@G$ZSgDujxJARM@t6ihoG`C&0*i^d$v=k(ci2+yNGlb0sF_G8{_Dc;+*COYD4n z#;sM;^?nWmDPIit;W(Kq(sg@Ua^Ozrds0V0I`$@|FBU{d#VB(k8LHSt{L(yqt5MwV z!Xtni&qWx}Le>Vz;S#^38GHu*$apI!pI!hRg=&Q+SJIL)3_fint*PQJ4)mr3FAj65 zs6W4Y(8GLhriqa2AploVb4n>V{6-}%ypi^II{(uWvDAJ_k#AQ0JUEXkznE6MZ891N z3p^ouJGdym*R^;MId+OJk!*hE36?eA17ff-ztWh&la<_r`NqsnPp#YDlT`cXu`r(V=BY;M5)iAnuT%{qQNn4ULiRtk-b&mA~!-tSHyga9a4Yy2YFD`aif>ZHu)zw zwGDiVjsg$U6-!E|&5j#u^F1< z`NRn=9E)ZdUY)ylOC%Z_Eqrxk#V-)dTs4?!T_vCKu?D$sh*yR;8-5hd9w z5Uhh0HDLs**5hC*&72E&=J_T41V;j>k`OJn`1NM3+&Z4N5oqz=lr(i@GwtZjJ|+Rq z&_Rr4_iklAXzfZuJQPEN?0vs`jNA37n%F`!|MOm?pb72puh%4_pU^3eZ!sAJNzUwE z3G?TL>uI7_E<PN%$_nVDmm{D=tOG`6U49QX}K7D@2A%|6c&oH2cRv{5W_H6HtxzX28;2tv@TEia( zLcJVqQALLRX=!GR>k8@5G9P>gi@f9uaEho*Bjl%D<%{ zSfmJ8I(<{a_tShPyWcC@i@`b<^P9uUlu6;XqNyVW%NNTj}Rm0yESkptZhS z$Ytg9Vc_H{eh?$v(L#3zXTAy!8FqY@g>=ZZnWW^_Jq>!TZ~aENFp>&LcoD za=XdVdD}IZ1n8DRo{iYVJhW(?!3cU*_NKe}3PSw7k;wk+#+y*)8u(XpmIBP#NAO+< zLNK5zP*RpTv9evwosj>FUYXK4Z$Hfm;@$PB=WfPD zNM-qRh7!;W_0TeKnQG~&DT?aQO60atSK9LQpz)Coni7;h79(>N_s>o&4Avpw!F%)4 zhO8IsVkLgysA7#UmYaf&6@wnl;)_pdVuy4iUkRfHgIj)mTd2uL(RXluZBT!qYlFvi zlq|JtXbLfur*aMFIFt7KR3(gE*RhG!u50mgt)*J)&jlQI`SU|AC|05wRNh{U$X;Aj z^HhL!Q{jhxBg4W$+N0zU=ew0!bW3?PEhZ1xW24ty>$VDIbZrNkl;9`Q^~T>2Zh3pl z)&w;mn6Gh?VO;`CC}Z2Q`A*7J@`I04v8>k3mxEshUcw2!^=PmMQ9cEYMdAF^|K|Fh zJcgz7HM5D(vJPIbexUT3M%ZWj?yor zZF`^3m14Zl-W^ChUJFyO#hh!aX~R7j&=A)Eq1*Dn*U}smtE%eY;*d-!?n{F)f?qC4 z#`)8Zbrm4|M}nH0Z8n5qG)o`Pa0cjD`38fQ=E~%4r<#_PAfN7H9H-j^AvY{-s(JIs z(MEZV{UKM>@-}IsY*ehV<2pB08X8q}dwuRZlRR?{(H6HdNl8%EN5c<2I@6&4GB@I6 zfw^%|%Xsi|QG|%t=y#s1;7fGwU-v-r=ikHLTlHmDsv_E>4*SxyZ1=I(pFWOBMF+Gu zMkCM=34T~6TD)4f4oD(-nsco_-t%S)I^UiJ^nf2F`w_2y+D!SZr9FK5hurJ$er}u+ z@UJU|LO=j&zdz_tJRN1c01w#tb0ElEs{Kxf+oRY--Vl5$`5ib@^R_3O{8%2XR{6#5fXDW=)_h7X%gU7~*yrC!!`Z!MLLQ#nz>`*+r^+%~Y^}E<)_OeBZ@cMQbKtg#fl%mL_=jn zi)e4C$*L;;sGshb09Y&j2Cnss6zF)|nkZqYb6JZC=WJ572S5Rd>|T=NKm0Yhw?$-; z-zv7gzVt51JxZMBWBtAfpW<%%Q#SAU9czFIh7c6XMfeMI`WHdYm8*qN>6NRW6pJcR zyG6=u2$kQeaTb(JEpbk-jI4op+=8z#4T@Rd>6) z62Lz7e9(S&H-B^2M)U2-;Fa^xOVYdb3O$LIYI-9FE+(_Tk0|hdnC6Co+18m>;L&c|#EZI`id^hF!g`47SiK zSe29S?zXR<=;l#In9$+Ft_~%~UkTa(AJm5o~>{`Y$8v{=N+V?*aP#lNb!F^kH(C zHSJ}@oxeU?g>Q^CL*lti)?b}_{jR{I7&xMAKKiRS^!)$3|D-_UL>wrbwiNcy|LYGx z?#_g z6Vf*)^Qyd!|3aQX30NmHm}qbq{`wL|$N|{!P+ZP`fz^L{<2I%Gut)qCi+&Zb>d7j6 z+WqS*#qbUA-uj@NTkU_AwcoAg-@PBY*(W$y>;F26UM~J-1@G!?mHX?%`u`8i|E+=P z+%LcIJ5Uk2I?pl!^Ve?Q*5%Pk; zI2y%{a6Z)wWpVMdzqsegh2Cu5n_qmdNA}fuM`oyedoP;kM?!XcE|h-ry8qqO{O2St zxG^{IxLFx@TV-@cV02U_rFw6X1E74 z6+8+0Yky|p>CN_`Oppv5fS&V`{Z&^?jvlypdGxjQ^j0uUN|l?~{d--ymKg~(^a(#F z)A6n4-o$NQJY|fTKXmqvzP3*Np3F`x2S(S)&zq6$47=HOo~JT3pD`*)WIlaRr=Fs% zyb2i^l`?tjgUua%lJn*6KbGY8;Q;xVG-l;WG(H4D<9{Jvvn!3+W0cGu zTIXz@bm#O?20qcVxX2a#D~ig|y2NJeNlyr=!GhpSy?ue`d?oZI83+{4=FI|?q)~=d z%n}!Sxd3@zSELl^G%;{$Umfy}oSG8=5(ot1v${9ob@EAQl17mM-5Pg1$#Ztxrn57O zsS5K@mzj%GhuE8hB{Wy4#YNQn$hOlua?7Lp8o=_ed)Q?DY^7HdRml`(Wz+sX(*Nl} zgQi10PyIzbkZDhlZHfgB5ehc*Ah2*I)uvVyWIw!|jvoP$?H6M9hAPvJcb(y-tpIV` zF@#=u4p{ives>R7W9_S}t8<#JBmwZOvjQjc#i-y%S+BT9`w>X$UykQUe|AI@DqsiZ z{~S0+Ju;+T*ET-@(uT))OohG2$8U8{R*8DBRUX`4aar-;3PK$(g>xm^{-}H9y!MOf zrt2$-&q$#SeMyrkv-@$5eXuZ(Uc870epeQkN>%?RXB7I*0lK*ikf>N96Aq#e$Rm30 z=W`kP{a`;pWo2)X%gtA-(8P~bg9T~;>Kh9PXrIQw#=Ur2p*Uc#b!Z>=u*o*CBKR^= zKi}19EUM|*FR#?B^&@>s z^VBtV2HU@YKH$h8@hd&k{67zA3&uB&?dJ#mZx<;7alIor_qTpLrhm?izx4tZ=eR!u)X$658)P1H zP5T{G@~svTEguO@20A@m8<4pEy!4f=3;0UE@y^F|s?neVk_>(41nZU+Bn&B zq8>8`aGt;zGKL`TCD{JrvLS3nv*~DKfX^_!aun^3mk!832!Z)a&_$!RxI9ZuIBOSt zy1rOk8+yR`4}$c%mUE-=H*PDHHJq+dn0No=@vcGhqae{(dn6z!3l$|^Sy!y86j>d_ z{y4?lVbTT78Z)8o)#1qpqV?vx-{ehqBdDSm(di-Ei$CH~Y{O!J{WEd7@9&!f-T3$8 z_Tv=jF}Ad8WR8+w;M0rc-mizVVvlpG0N&d*R)3x-u~@=No#@p$A~5Bv&T%^#?ri;H z2fOnHc=;eC_0%4>`?G_gNOzFEUa~?z&-k-2l0j>~xXhVvI-Bg@eSl#nIj0bS+=KaF zUxH`#8)AL-7X5DeP00l;tW@LT-;x$xh$g^k8k)&{(^2JWa3qYFo+Joo18 zd7~_f_RnLe9PiCuB(nxzR3vSJ`E1)4&18ps2qG6v?jq@0Kr2ui)0-ww9N9XQ2Fn@I zPs?uIx)Ix0y}%`}Y9`^~Ccp#`>C`QZr^FVbG$P0}1E5$%9!bGPP|jT@a<_~UvA^8; z^|I~zYB}e1-yMQamLm@dx&Tz%F|XP0gnYa(@_I)HfBxdH3pDUqbtAnxJ?>Wj^>I9> zl$SJ*Z8{jZ3i9?0YjwOJ70(C1{nY7+|Ea?1R_Y3BEM+&mifghS={U|i%!RjRiUkA_6Cs;ei){)CiCjGD)q+|lwt)~WVJE)$v2<9Q_@iLt)se|hNwsk zwxHe&A)zsg?YxO~qP_-d+v%J-S<&vFI?LqY{tQ0rf6FV@=>!Th`sI4b6g+;z`ffyS zRsP=WK^4}4VmAZpX*0JD4&19s-Bi`WL>{2vI(pQsw>%TFLZB!F6`Rg?LGXTCZcD(3 zsrJYZn8m=lBi$ZKs#@hh$xF$e{cvat#ze7X+5Xd^&shb)6B7fhL`Q=qiOV-}pp1Uimn%W}Po;ctS^~q~8CI zaKBtJVfM=7{ZINc8T2@1%Ac*xyZ37w^d5M59edZUZI!(3*|ufaWxiZZl3hi9LT#k< zjv?yQ;roUu-@!G>f~I*I4VBJsWXB)$HnJL3 zD8c24b*ve(>0$YY@p2i??guiFUNTIIijyzssrADqO|zobhg>^4Qg=sdWxqK^j7fwb zbg=U?uk$`L7m=1!`4{jliYZ4HaBF{*^*tvT0vGG8;*$$_CnIigRB2N0?j`SdyMWy# z9Fd>P%m>WugYsm&y`m?v^AZrL*`Ft2wVQ^3)|aI-eAxLk<+HF8K&QL`xIb#kvy|yF z)MR5M#lHOgh?B>C^T_WgF6#e$SfO32!w#N}V~gRuq1Sr=JFq#tr$=<|#m)twp|JAr z0QS$DE|DUTZ^9R*uw`7EHE>dkVla3V7pbB#d-YnO^C3%&wnJBj`am!qu`kg8gKSi> zZcUQh5PgX2;EfN~SigzqOLfka1clg5a`5Du&V7h{;v33XIUP;OcU19rj(vnUpU54% zQplQ(*;ka%Gxyr)sl&5v$EE}TWRESQy5654MZg(mA#}RJ`;LCa<&DR^=l_mm-!E$k zVC-`PJ^mjt_QD8cBSbrzY~V9%d_U^fxu2I%9mt1)L+=--ru;x56rcj6T=Z^aKi>NK zqp#A_;sUEB8y(!zv_<1nX1qQI(DOD@ZkfMIz@psT{exhS@-3G=0U`WZBudu9e2i)R zsatGra|6(ibENO%8;U(DrZ;(?xGRG%bX^P6lg0z#B^x@|WaiOFz+NUR)T;?J<61H; zicTM(JH*FhDtX-y0Pgl^0kq6rFw9FuEYvrSvFA9B%a~g1owPpFK0$YgjM9vi@koI< z#{;(6HHtf0+Md};nAkpY-T%YWS%yW`{a;^2L6Eo+Q5qEi5s(s)1_MO8LArB5YKCq_TDnWR zdx!ys?ob+LV34kvaj2nrj`#omJ=b~3E3P^F?7i1o-_LSMTGPZTXX(4hwk1GZ_IX4X z{52FRmx_~&dPm4<#2<~k3GV#J+DFqUyyy{Z0c5_~Ngc(sXo~w?EXGUs9gW&PeZj>i zn5&Df?*(!JbNsQDnTYcx5_jIET_?@Ul+&|aH;F(?^krlw)(Kt_nUxuuC9?N$J9_=pwajZ_c)5*loYY;!ijQ^O5CXW2>+t-#0FF5-33C5if|zrA3#6HV;~N!#j(2+5elP_d2=>tUW% zKaKc76Tw6jgPionvq#d+E(Kbr^h;}(NPs=ZJ~?{bvk9Dlf!ddN+Lnv7fG|e z_%@28QpIK$3w(J|TUWPW>)V#-lNALS-@vxqhzBnpG`)$_5l61mwu}7$J!)Eg2+}jX zn8$VEZ$SxyWU{&hIv?J$AL+*@U z&DEI4aU(|0yzhc?*XX8HSU>7$yYD)ArRNEjC7p#YZ7|soX;)~={gVxJp_PEd?-MF@ zFO`wqg3Oh!D@`B6HQVgvO&==gKjDV+fR{*Kz&jpb`Jqq5O06z`h%J~{mKaU;w8Q0ZpfBM`vf+_43@q`F%1FIDV~NVj#mN zDZ#HM=l_?2Kbpm-b`NDZ9clczo=lWQPF(OOC7XkoE7Ls!s;{VDm0z(^~ zYc{Ljkzo(a?(9c_AjpbEA-m6TOo*q;zy~xEpiRp?wJLe3cl-l$yP>;J9j9xy_^{H! z+m(sp;nqm6dem^XaS74mlxpW5Ij!sV81@~vfXqupIWp#F?sl^s>%~g57~d%g9UC>< zS*XsfJ9c~MD2etbB<88E-l_vJrQMc8Q+oW6rKjA*2+9ChnqcW*&ds*29?sQ1m8nQJ z&>X5p2P+e8V%gP0biFLo)*xzb9H5g&St+W9p}^+ZYyzFC-?sKfeLouH*hT zQPr`L`RYV|le>EFJ7a+MM5&TJkPhV~-Ff1`ViGA&l+@=mS(@tRw?~`^Y5anQMY&i@ zqnLuIIpgf&COAT?Yvfj8Sl`FM!1d4be_F7cc)w1=zYKF}I8&G?uc@^ZmIAeyJM=Ym zXI^pk2i2b1!<0ZuC<20aQ1hGp=&uET%H8)@?5JHi{oD7aT8M1xMje@`*Q2rY zheC71bMzRR41OTuoUxpG>4rD&6?N z*toute%|~0uEyuz&iuvz7JW`WT7L$iZ8qfojZ@*#XdILR5*L*&B<46%+-njJf%_Y|*f#KBCu@P6Rja18@Z&@rS*0#Omm{tjO4hhUMhe<_h?T&~hDp!5I z-h`u>69v?zdjF=A6#&(Xkw;R!g1GI#?*n_s@1b6SKHzsiN@-XfqCb97Q^^dzB$2`cu_B)^P1PkHjkz%j-$Nr6c~bTx5K(`ghDudp?LrPt+9oA+sM3L041=RqC%Y z@C6XbUY$3T{e~*cWG7uxXJOJ?~B{7tePFq2T53#wjCy3${fA1$jecX?V-iOxU*_4maL1$tGJUQxdhgw3@=6E z$&LOq#vd0^O4C2M^n2dUclB_lgW{q~<>|+RYOMSeBQ+{4&h<2JMU3D>a38m2;-~r0 zI+J8>!)1gmLpk9flfPLSCZ)A*jT*i5xVId)aFq5scr}G1oZ1-fh^nN{?qjr?Gb^kB z1<`fOfbS9n^Vu|+_tKIke6$=+igOZj`bF?ujyY4Gg9B#wMQHEx0m^bXLH%R?*#r!0 z(kXpYyU6blS2XsxE9KYQqi|`rG~Q~}LN|{2Jna!v_aHu+yIld_K|!J+(8Zi93Crd0 zl%H~=6JM}rq{s<5&aFM?cE%i}M;QHKxR6f@WHhew_ExZXY$X@cr#!kgjY{GdDgjgu0K zhv9au1dPrJF>T)Xp~5O*v5z${G3c*N`UPC|!7Iw%#!-iX4x@?U+(t7hfqFpO%%#@( z7t4?9H)_sRHkUvk#W~BZrt?Tm_rv+-N?rEyx&=6xl{SA#c--d%rBnw_4{R97wGx$y zK+p>XI-phI6q5jQ&f7qnW>q5G!*OKKIvcX{B1>jFe372_Zma#QBbtk=c44Pa zuC`PP_X=E7V1}=Zd1z>Is^?KkMBK`&K6WTMRBe-NSQ3(gU;eDk{Iq_F9pI68BaDrr8=BFppJh&AYlnYFO6ycYbjy zTnv5l+|*DRk7WW%-Lk!EKV(rwC-Y20x}K}@EeBJZU;-;O6O9MH3EBnRD+~G)tueLy z34?65+)lq9d=wu&Q;{|dd(rkeLgw{=Nk$up6J4b2#WjgxR=(xyxWbrV|Jn6kf#R7e zXJl5k7|RRIfle>7G?62^qZ?&XP`~TEJU3ZCC2g^rO+?M;i?2WCNi)pu-%4@AOlfnz z|MLcc9@(ZU=tZMn#Xq&XBSH9v_UHOx44q*9+lazmHXcFwL*eJ5^KT?v|Z= zReu!EYWgQr25~U2H_6ZosTFtYki0e@gJMS3E7m;25Vy2N5AAXT^i);O)swD2{9tBb zwcbYK#rNsG13h)&SsdeISq~eAW}_+CzSV1P^Q(%3D&8Sv<_w9;0-~C)TFU+=pp9U3h%dkBsR?vGF(mbJyOG!Pli$6QZFuuTJY0gYn`U&>Lqqj-!iYN+Lsu%e>Rsv zrKZPl@OKG$WVL@T7dkz z>L6R+jO}NFbkz)pI!k$Medur3L3PHC&sNTwViy~Ef_&_Mm1aVvR?GKela2Kfw%ba~ z_yA7RB~!w5f1gVD#o;i|7%YLG3xInGfQ*~wL}Do<&w6U!iU#c?@&&fS>who=IbF(* z=5;|Ym9j&75|y7yyjQ` zp~wo-!F6#yt!k@z79<+rQ6kS4t7ELcSKWNpa;D~|gj9>6JPzpWmOuwvkac~4lG1sq zV_NNEf3S{ET3!J=Qf5;QgHE|ibM$H|>sVCN`WJb*pGQlTZT0^7bvvxHTP-GVmVv68 zgVh^jN@_>mgc|$B z_(G=e&S_vI7sOHa^nF{yVkz~>m~g$@>5JGqRqaeS6@Y51A)o9gPKj~dz`G#^?_8fr z05l2ZtpGWH!l$1bNMkIkgMEcEiTbt zBLK`tFY@=zt)8up!(W@`D48ODOKKaAlX%{LMn&0AAd~f&LQV40qBI?LDK_ttO6_Tp zYz;Y|P*!Nnx)cmO;3>{3DW`3YZV86Bm-=`&n8vM}_V@2fCW-SMk=Qo1ACFS{V3l5D ztF9iaAM||{jk2*{IS+OUi)Hx^8eTH|?PLCnoX>A`+A2&)cquRX%V#8qrMJb-VG(00 zv3J8_mAO-wO!|Zm%akK!3z@kE=g{qTl#I({v>cb`S4g40_8cxyytS;mEp;KP!O3nX zU!G%LfKO)rofN@#rFNUuTJPyS7GNitz&yTZJ(gC6P^s=0i_A1FO?Vl7%~@4JuXoug z7mfUyZabOne9_oar%{Mc|9nyT$Pio+t^d@ot*bsYcX89|>*3r;Ru=W5|C^NmeVJW` zdFp3>GRvjDE}J@y{4KpUBH9r&FGCyemA#H`>Xvs<#nGN!`Vern&g}XL#Nze85`Y-#ulo(R& zLrb)uh?!Gwah9xdxWS1WujXJp`#lM?X9)AIn?x)EO0Q*sQSY)y#`c?FI01@FkXXK- z%yeZb#JaL(&H4xWalI$*(#FqQ7R7{>JD+G39ub@o{zXo(-V;f*2$l;i#y{hf^1rgw zUyB>Anq|O&G9>U{ifdj*a_2_RpEj;lSgJW!?|PMEKk=_VFj<-Mp5E-BxwYR_{YLK< z&`WYJ=>R#${wZniu;6V{sz*VYl=Z~;?v`idq%fZ8560r`QvE%-HfPqCZvcT3_B8~K)IDt;y0jYiA>&1>Z0DU-ms$5(yh z%iRmJoH3>kuYTvebzHi+4)ct*R}K9M4G#L3+aD9nO~^qh`GDThTkI-~|8wZ}6^Mew z=Wx(klzi5}KR$Z$eXGF^gl9QE zL1LtiCf}!9DP}dRRjmz0aesCLuA#0)d95K69nry=MSiCMyT3(;6JYn}JeIXa4l8xD zHQEYml^Rm_GH2>crOtq`#}&iRH69uHG#pd(MvD1Ndb&|$-7XldEOPW`^C1<@d26?? z2eM{5awP3r&bxQ#ER@-w&eFQ8W0~h7@1w&^?~hYr^X{B74j!JZ}4?WelXCKFw(a0W-%=ElP$w~-@l)j zKV}X0;KD;6`72n@5bWAAjb%%V4S9Yy%ovaXTvq4DXOtKKO>})F6vr3-aMIf=y z(o1h-KndAU2@Jb9-?V8sxD6q>)ffizR@eTX{TSwvnkzo{yP&q>d;D(1=%8WIlL$p$ z*`)=a5eYWV^mdxFxQ*uKh08*>20dJ*#hKur7-GQ>oC}2*dFE=lb83*uKzg~hPiP73 z_!fbCf=r}?GC9^Qq~mSipscdBl!Ta;*dd*jX?5QyN%36>r5 z{)txp`1E4bB-uOhv#R;rI60Yf-n;Jb-JFxTk53yXI^52FrGAY=a5B_=y4){wJgg7R zcD*(%`OWlzWapm+@bd1Fqqsouw-mJFrcx=*N!zSAUfsvsGLW)#D}=Uj!%lQ-YmL@2 zR-WjhnHsfKyeg6akp8LIDzTNi>#dKRdBG)SrrO^~1d9pL<9`)?B`=rrNf4d@{~F_h$j#e+)10 z{D7hfuj1$4j%LwW(I7^iiwl^D$&CuF^hoh{sbH05fcdW@%e!p8Vz$$U;PEr(ooVVO zapn=08ShFku97i=q_8pSQm;b=;A&%v2Ei$G_+!Dn90m__chRNbn;ea#J-1orJUJBi z@TP)H$m!C_*DoZ%ez3$3P=6%Cago5a5B*0iMnn?JjT8@ z+)0cJer~ZY|6NgBbuCQIC!L?kt@oo?x@GP91o<#la$dZ4xX!zJ+RvJ%S9n3dFZ@$T zq9aqyT3CsWE{A>5C^9cI1WT~rq#xs#<+v9J!jFjx4Ie|KmbaO)UEX%ZUNLwBtvaIw zG4G(5?E8eKCUJMPOP-P*swmQZZdj#M$m;Yh2M01jQNF4o^{WRL5m!`wBN$cPBU`xj zToCTn*rWW-7`DJ?~7CAW^r^xN2?$^r^FI{I}^ItF)ZC`YE1>3!3k*eTe2F z`1=ed2kH)C3FNn<_ND(_dEqj>cz=ePsOs$pt*@@9pSOeam>45G00Q700^2Qz)IWR8@xY4PwlG*qExl?N+~$ zbJh#O4GBLTZ`F?SlbJH`)nXEz{)3jwH08p@5)yTvbdXm~RFfFjxv*%jmACivok+t{ z-4qYxY(paz)HBq#cLb!7g`OS+XR(!@^m&gY<LE0r|NF-pJENG z`ya(WmgLl@&aZ;~ynDN~J^D%;q5kE6I%0`sQ@tk9K|u_@h%Xspfr-P2^wE&7YCh(> z{)iagY)2-KWNww#Xl34`7dAnWscv6ZKkS{Pd)HG&wmLg`+t<0_sUrPd8+1D6WR)K7%Ae{VzRtUx9X02%nMv=kf0t3X6AaT^3VSU& zJmq?>E8zW9gbhMeKtL6(NMlNk`fh5qYp3{9j`3>-g;xyd(Z6fR0?w1=eV%U9blcE5 zGQ^u*P>g~qHakVpY>C$1?4Rf>KF8n(C>fOCBvjEYEH1lEzD`Km4qZ{w)qjSa}EJdeWy_H$KveZI}YZ0q}#$SN8MT*IDAu zkp@PWX8mLBrvPRA9Y>Pow*IgMx`oGbRmXT3Nb^1Oa7<4cb&55Tz_>Dv`dh}JpQ^1@kV{SlWs$7incL)K1K#dC~ zzR)_;OV-1u%Dmq`7weC0QQK@DGDKsiWBzRp(E8gPaGk{EEB6U%S7q3`KjFZk(URZp zpQF#L#wosDoEIHOJ!}+Dr96Q z;5x$Rr}mGUvo@z!!{y(>!oE*}N4~0l=R`^K8-v^9AFzIJ>wC)%vJ>N2BgCS2gdN!I zB~Nq#DvpDvDMn+=6+gPruxt7Ihwyqt!|uDx18x*xmtO$ChuYZpzS=n+ zFyi#Tl5g+oo)(ZD!7S!K?gK!j4-r;M`H=C*mOYPp$K$foDkqm__vEfk&z9Dj{ z3yY}Cf4@@9vOsl!=nLtYDpbz;&9yhBr2D4@$&ZHW4vZ48@M{?&Z<+P*{k4RemaH`V z9K&M@9gZ2^zYLgDp0T*=ORRri$>wxb{R%unCLIs(D51+nqhp(e|h)l=F!vgc^D)+fy0zRFqfh8D$^(q`zq3ZMW>O zX(%I`amDy^G0f){$vG9P(Jb~a_9or|B z6Y-kfeL|*dUN1c7uuv?>)5UwZa3arJ7WT7=SflD=a@o`GRGDH8iP326E2B)#1wkz> z;~Q2z)i>1v@2!8#>U82jMWMNF_%#?q4LBX5+65O7^36i%PdN?H z4A7;xXICV!#E7e-5V&uwjm(7sAz1UX%^m89^|g2ve(l^! z^FGPkiguZ`O$Z(Z&on-%=7FCiSC>c6!%Z+($&gcf`%D1~$Tq3jW$lyi?%&uN8& zEH@~1Ev)^-`AxbKTDFi<{CI%3xpG5`P_A&7*?%vwj#E=>G3gs}dIQ{Cr9jlLL^Ds= zUt&|Y!2cxHZ!|!LVi+h+&S=^k^G(QrU_{iLCr(|a6Q>E*V^;@P`!K7@3{oE}%heXU z6Y+T-J1_fWd9!XDtR*TS>n6>@>}3mIvP?C6Oe_GtQ)aY-gnqn`gR)F6Jpk(L<2*qg2P z+3e?K8|0xUI6CK>Mj7FHD6`*)rXKGAK>^ObiieW4CI1Z$#95c^4Q|QM&-E$9rx6_o ztaav}C{zFK0)TAJL1)=>_ z^0P6!t*cz0Ni_M$D)Z1h=RKo%Gi&O^Sz}*i*oQknH40PJk?+)H;MhcqmMaJ=d}Dcc zzW>%x*}a!7>J;kIBBZQ*BQq6J1HIGsJ2%TEC78bh)d1EcH%WSh&=^X0WeYYKmC9?G zF|n%Fi;DvxwX&+CXJ0L-Q9_Npr$uAcliU01K2k@A479H-@|jW^SYl-8)*z?*mX?^- z2_VH*JeSG_eTmA6Frf{eANm%r|$4a@XHBPn37da4^slCQF(D%x%wFS zCkif8X?T!VwA0$}&}-puwhX{GQ~aK*uk(_kq4Dy}7*;p@2d^W=gI-9>dRTU=)c3zP zi)=$^ed|S+fN6LVI6mAZEMOv2&U8p%#WW$#hy=>(BB3Sz&m$@G;rKmmA9&juHHT(( zfU)LuK1?_t1bex#q#c*;)MJ^f9bXd9pU~zPsOWD8Jp}UF`tWNEeURi`#X0fm=xMla zIDL~sA;Z_G7!GsCUVKMq_j|Nr0efV{rQ4S<@Kza)c_G-vW8|0Iur)KBenDZBo{r?f zh${b>pO=e`zaROk<%Skd-H(dzFM1(Aj(TI_xt+|80%M7&%M2>nzb{QR1(I&zbK<#8 zE$A7F-rzD7;eGFD#HT8CylRs?ttwMqy>|F)T(oGO)!(wznBj zsQV2=y64}t9`OzS1A3YA?g0vnP$#;-pDm$m+WHtZHOAv5e#hJuD&y&xdls}+#O&Nd z2~BKQ$^UzpDu78P6W?O~Zw_6WoNc-2LQ*whNm`*75>7&+<0BV*aS`F}av4QCq)-P8 zICOGqHNUC);l^1Wg>|=chqMun-Yx^@(ZKv^L1LqN!8AS4&qn%Q@q>krjS!_)yMcH}KNR)f}w*_VODx{oWSAO^%?JO(` zZd@hqG{x)nCD%AZS%=eMe?vk~|D5c`$tjtapBw^qWI^d|NlqyDOFs0+{j$A9&fw}F z^aPqIqh!JJYqJK1Z-ci~1&SLSRxTIr((R!}` z#W@dfyD{#kUyA1rzRSW%M2TyE99$c)$Oc9)R6w2uC<^y&d_LPgwOClInUJgS@WHp? z8{&2Ofx`iE;YU?B#riy5oPTsZaZc-}s}#$3)HB5%`eiAQrowrLPo!ZY4E?mHyZ zDITeS;-fdMJ%Bk|=_igKtvRO}DW(iAI3^Q3YMYhP*RxW||AHuYtYe#~!hM`f!a_;3 z7t9*AcI|Y8eIiVA^1b!YvZmO$^>vh=72bJO%coxh0=iRDaaf-?U$*2uxvjF`{0hGz z_63D|(XK=Ta7;%`8qWx17HYba`NHZO)Thb$oxg3;?WTXDo|p?HU)9(Bb(OYYdE3_x z0jTdF%VQv|0nirt!HP8~Q%udLvr2N>RW&eeK%224fK;D8Y|I@{vB>BOF|VXK=peT* z)Aa19jN;U;6w>}wDcydO>M;~&e5TGi+*R)%w>^H!x9s}g#L7U)ZREc?09oO-s?R;B z=y)aEZEul@jaz(G%r(o{S&7m2>iYtnK{8Q>jKvtsr-R@9S(X+j#r+JzGMgHS6{=dW z-Fa(O&DVaLSC1(!_dd@wgz1#Kp3BChE;k*Cti=4J_$ALVt<`5{w?5%9hqr!9s*N>Hl8+XnQ6}mFdDco_V zkBq~LwMbL{rZxi8y;M|p z;nmy8!wA(sFSpIOV}s~zX&h%h6pa^7@}FBZP|?^ZQu1f3#Has3UqHu+1&@yfBM}zk zC++(SEuu;lh$VD7Q%0iwM8$)+RAg`IZsA^$M%L}Yp1-ynYLphGW2_@ZWeziF!fL3! zta}<`QTWYl$_6pHzb5?F=kXE` z`9eIPd6(D_HLp@1L&}|=cBxUZFj!eR43kUgorWEnliv26b6T!eyx48_!_VhaIdoI)|4m_=LfV4zxe?s+tvO$JE z;YcdV^xR2=gJ&|Ay>R6`6m4f_Geg@rAyM?88gR`gqW14wa>{|R^HkeQ_5oLD?Tk5G zTZ5%Ox=$g@3hC>>QSNo9Y=n#Y%+td4Rxez}tziLKXBz#r$(A(lrav!7yGC!vS}bas z>A`2tZom2Cvb=9!@3#UPY73VKhcqB|6lTc$w&KJuI)6G@NEh{4VkWw9{2lto^$g?BSIIBzqs=5A$=X`V`17tioK6#7d1^qMT^=Ia!WxWT_P+ocZzJhf$zQ$rsAi3f8BBSZLMX zE9^4?)%#YeV}8jfiFSA~{iu^ua#$*wh<^pJ@&%f4!%m2*W{TArQhxL5@e^0$zmB3t zG!ksJLPj`?1aIIc>#j65p8Yr(`|ERKDQQG8E^ol@FgUjwj5Seg&uI1J18Wsh{gIvh4Czy$*k+3 z2@58}n&_lVD}jKJQ7>e!v=m=WCZx~c+`lnfPfcSt{`5DA2PJV2d* z9Zu{tro7SBN+`zPL3RI0arMj7vn(vKF%c$#cg~28&x0|4Jtbo<170MMHV}Prp*b*4 zzbp9b_iB0rdKRtrJN@cJS#m>24~#D$)vM?$i3`RxK`@xSh*KZXh6lS z0EB)_7o-LcHysFD$D2u*HpmXX{-2rbO6#$k3CU|;Cn8~a(XQ3aNu9c`Eoq&(Aa>sw zE=IjA|DHyF>P9$i05B-Q8vA0paw5uYMDzF_vw*>`+>VZ1ozOdE4&^z5nbDdY?L-ue z+aY1zRVVWTp17-ECEx%u_r+(r1JJiMbFCzjN}}(_*FGyU_Nz?cS=PMmQ<)?F2{zC@(D{8j`IgZgP?a zrPUHJnX@ZKo6+@jzGR+5cX9Gu64LNiYvfK8h(^yY{p^77KJ<+7kinHe?}As7#`bq# z#+;y&$q^ua0#;L^^*y}jBFlBhJ%6KhCckQhAW>Oq&IsObwxLuBD5w}R)H%I;1wq=w zB4Gr3Bwd(h_-wgP0moQYY5hftL6z!7;4nFvQ*OBFSoqsal-iE&}f&T z{SiIaJfz8^$0xl*N>cCn2_;A#iOH#aOu-#yJip#xl+k$(%u-@y=*4Y9d3vherG8xC z){cGi+cuPuM*z|`QVM+oo=twNOqG?glRyL?%))Z{kEcEeWPGqE-r=h|pY)MYV~-?iVn#p+{`*y6DhjH|Z{Anuy(RN5Z=z<5LznRM~& z{h8HK()`>AoAq!0Wm`oD!T8l?U&x_o;tGq?atW#Ta;(?ryl#-lE=lP%+MuP_q0tS7~*(0u5)aa*)O)x+$cij63;245UMHH&Vl zU6y!5FKBb41N`Wi2`k|6ORrx-Tm9@bbmdq1 z+!t?Bx3ru6e-mY$HMa-Y^N=cuEA%+EO))V8)y91+%bW{6|nKdPqpXhKEV2Y7_40TQ(k(Qs65jl7X=Uh{PMdP=9`a|d)l5L_1q*JB! z`d`u)8khuyFlv`yM9+c%YnEZr|6@|1g7`iuW|Et;tB?HyE^0A;BwXux_!Dy_1QlBP&?#~xBxt;? zj%cV37o+v1ZnJ&a2al`h*zu}c_y^R%%L#28dc0s&SP|nD3>95q7|`(9&F!!O{89Gn z$Edakr#ODs+2L|KP78=H@k;)>Qqj4f{n{7Nw?-wz! zym=CNk}8UBSk(sB@Mp0P6@l4QGPPfiBtrA7qbaSpv9;{+A8R9wqZj^rO zsFwu5R=M}tBRkF6!{wRWzb8|`(#qlg{@2O!Fae)H;E|ZG$Q7-UV)D_+CDcf>wU;pJlXl0Y?30!2cjzxBO)v!B`frA}N22S!4$w z6EHupT-nC{j=j^+#wcnC#O)ZHMX|KmlzBb$)E2evcj-Mp@kKqSV>@?m`NODBA_BYh zSPX)3xnS_WaIUr2kzIOPE+q@2{=pcwSJII&n1MCMe=LP3UE^*}+=MCm{KjWycFUiOE(9sS~UMBCr-K2{i)lv(wS2ox$asZT!u& zVt#R?^jF1x?N;meMYi)tjzB%dHTzrUF=BqMo01=ou_zvZCJ^J7#5wz$|Eh-4#WHFq zn*N+kiwmf?mH~9Q&D*|zbwbln=9k{bd(K&jJjU9ElDL$da(ZCf>{Du;<*cmaP8R}! zTSwDGpN z`~?ER1&;=|e3Zl~Kt*@@Ti_3Cc&_?tk| zm&lv^=W$F7J*$qP-lB6vIt1ly>}U%PjR%VpaPkHd#|bV3o+T^NlCr4$$+EA6%m2^1 z65lw-`v21~84!>R>gZo<_B6VzjYz!k{tzvO(LMNJZ0;Pe_jTm28klyzx+*I>Edu8^ zGpcnsB?(0KG~Q&>sWga_8(c@j(Y(rTmvWaPn;CVP)=ly{r%_&ma4E9_g0X)#4>*kr zeRCvpj6wvLbqjr{0aHCmsXz`-D#A1aN0W0o*)PicZ6M#o0aE;97#QpPy89L?N9NCBx@OFmagkA!MP^SmHO>u!_8B@$WP-G7+qiAt1Vbxf0R{28QP*! z_&1*Rgz{v`06@(rT+~qM6dfO_@0I@myds)G$U@2fQthmRJu&bianLfU@f@81!IqF% z>w;k|HPo8#M6-te_T1H&qSMW?(nrP>*)cRi(mOLH>rFM51ZTHxl1sxtpU4@!_xNh{ z2cMdE;08-p3oOFSHq@q$DjJjRJ-<4Mjr=b4b)Qm`8v@)TcVsT=CpvI!%*oZfzkAV| z*KyCj#q#~!$@AWPagj^f#eSLAMMBguFe$RVZ@2gO9vW;7(0JBM*qM``ei`Rg&e#Q- z(4QaC2|WrS2o}1zQ1o4+ZyCm0ZA@W1#rr9X?wofvc@zEuFZoj;Iqxx^H9VWurYN^< zTL{O0P@1dG8ea*ZOHp?r>el8&TB3uiw{A8m=+z{oTp zz$Auop$cQs8d43av&pnAO+P-SU>@cirN%-ytYhrUO+Z938zWq)pSND&3b9S0l8UZgG&FqI5^TBb=`rtoV$B+i?%Pw#e8OY3+3LVhCaM)l{ zeruJuH-b9x%fY5>L%hZlsn8f7#lFZ)8@M*=?Zegd2+4UsRNk`wzWQ9DucvgkFSD6V zRQcvLFhVFlRYSKA5O<;~GJN(2gZx_gPG%zM1q=Ye*q&)y(&~pLhxS=q;@^*XSZm8) zoguW_`+vBsz(cAJp;;;6PKi#MKGqM6+@8JAY3rixArY91=@UpuQ+6YAchdRzQ{ufBLpIZD7ngXc+u^n;!D_fDH`eYOSHY1I z&zFl%vc!)#wxvF&kZ+te3*>i+tn=TYmBo>%+_j%b*iL=mt@;Y0PS7A&IiUK+(yJ3r zq2Bvuv{>6YHFvoD@+IlBB`*b*<~%I{nNz4z%FP=H-;{Gc^X#7l`=9lTO>Roa5M*vW z4kovP-+a5yUPg6KO3ATZo~fxSo|J+A4MkACIxgzOq$^BROmjp0+-K?rIQFUaz&BHZ z^^1eCU%@*@;Z7(0%Soc+8obY=i7EO_YWq8`B=SS(-U$cq1AlDle-|qgEH?lMw|3YR zaoix$@S6Yix$*M?CUMqJgmrfRQ1a8&g1dT3#zF`)b&n|GUkxR3vnCt4#3H8)wWd0v zovG|^+`>%Ju6PAl^+#t0$uvINkYs%y@$i>S2(3xS@*;~lQ*t7KLaV#yaH#P_BdfF{VJh76Hn!uyOYms5pUUuSzQV z(My>k3*X?P@NI|tZ&(pbla^$`b*hdge;Ux4jq-zWy#*fI;qAb`OHsO8tk&T5V4VMqFv=R_f!Ka3VJXfO%erP}0#dbDR4-HQ17`<#nfq@XXln86MQ!6Je z(;uw$|G!v0cf?;e7z6fgX`-6d1T~w;bPL9=`FcrNEnc^6A_U61M$2Y2w{bXmAV0jJ zl;03%Z3N2XYu^=2I<{$Bjlu!UCo^d;uq znczuQ8Dp>;>JGk7dgFl(IvWgfOs2Opv}X!!2P`z+tkwspKb!gxEh$`~m+*iDr15Vw zh3bT4K&?L0*xR?pVZK&NS#!u&@niQhF#Ev zo1q(gZ+7?D-S2*WyZh`v?>~G7$9tHo&g-n>I5M<87i%erjDW6Q%Y$(Ho~owKlt3&< z780)b4K63+p-^F>;xjIvD=xMK2E@y==_TcClQdummaFNd?x>OO6y9gQKje<$~0?oe5|RvEgJH* zuAUF!d`jDmsPlamwUBzXGYq0XFD|`Yde*a_18hU0Jt!t9Si1T5zPg#ZkN0PFZt(mn z+KspuyArS})Rs%vb=s)#n>;G1KlLk5SAq5!5zDrG)o@(;Xo&qYUTP@^8+mw{QRnX} zOYn^s)siBF=g6GXqs&6THP}A})#A}P<>T&dLcEmDNxubmc9?ei8SvvlAE(?E{(=La;7{bI#~kW5alVYWNqB-DyWqm_q=PZgpb3R z##DAYUGJBO@#Uy(0m;wVbpG&Pdx?!I-yj3JQ0mX%*I7EOV}>Hx#bGZZ?-nz8y{^>g zoGl(IEGww%M84=VH)i1fjWA%}htwBV7QIfDmK%TD?V&1fEcRV$S-IDG_YuK~_tSKP z(Rt4Lg#kJBUEBQ^<)sUhFdRA6wkIYe{D&G|*GTjTn#B8X-00;BKVE-LAyS5dWQYDy z0F^g53}i#%KzFP*jdpeV3an9`@NkcpriHr6lQl4O73`hu73J6U_8B~ z5~5Ql`ij%CEXunGsBshC9Jl*N|MX3Mci{4bfOFd}OP-`cjwG>5(V0dwnw@P`w~>N# z;&9y`zTq}=W_0$rN%;q7Wp&)sY03ry{Q~N_0KfjEs~-XP2u6eE=b_Jl{-80Fwk^vfcJU*U72;8GSIW(mzo!uz5pb{tKEjhSjE#u zBi|C*yc-87-Ev!buF``3@qItFrvP7gMRU!o>97H^$u#|G*y^lJGgxv3JF5T4v)=iBuY9E@UZ?tq;}qSFJdKi7w2p zksyi44|`vK!e+|;;jsKcxb+Fv7q~PNZL#xrqDj39Gd70Z?Ov^Pl8OKNMb>_6XS1QN zlD`Jn6vM5_(E-pPBre12Uj6c+QFer0L0BS@o^f+9A?B$!a;Y(DV1W8W%NFB$cU#9` z=D@XDcXj_33?&9RNj%;cNYW&DUpHSQit^5a(9#_w{`k%|J6M@(pg>pj%`2$)!DwOv zL~ZveOq1J9>%9MW%=k8BSnV_OjZ^%t-LG-Ff8E9Z_L+~D z*ud=Eo%yi*og(eOzWC$^Y((+)9Myka>wo*C?*SDs{LAg<`u~xz_W$hQXt?LJt#LkL ze;|JU#FO)~!0-=7VSE1b%>MJcWJrKjqIw?n2dl$BzxQumS7!jjpU*%DfBxrZ84K7k zGB6}9e_p!5fFBrs4L>>o?4O_IcTZ;%j@RvP9dUmg!GF6f&$0TvfZ-c^lQ6^opSk~! zb1jYm>=-1M^S__4|M|KY?KQ24?Wq<^oTCBEE#e29q;#54XV z8b?~d$wT!grthya%hitbnk^Ix%S%e&`R5nw^+z%dRgZ0-^Y_Sqe(n*tfL(!pjAC&9 zXXQD5$pJ_2DoYy%P|f)d8{wbd0>&eE;^|{d_UDxaf_a|4*9F#}C;V@J{^W?j^1>4S z={e(?LNX{>bKf83C*8YG{`13mhlIm*F%_Se5(CFBKI#+M_{%7nQQd_?dN<%dvm)dD(`GM+LWqlnDE8~P%dQ_Yn3SgRw{!Sw_n%0y+l6tPKbHyu z&ZJb4o_r)dhjEgu^OZ(-o%1e>>xAteIfeHl@JAjL|Lq8UyoYSf8Re7`7F3!|tog!?u2)@p^`OAfAQ>An;;U z`=V{xMj?q8ce**ybB?{CJLs%o*d02(AU^zEb$~^T^pH#E@kqiIKwb66ZMzWO?Q|*R z_7O&E&86sgFEfTEfpgMS+e*JbMObPT;>~wJn6xocpZ0`phzD{@gUYHZvOGh=FrR`t z&;iWZU}Bk#w#L!HghH_r(zf$AqK)+@?Rc)1tQmH@jTkYWwwhB)LcFTAMr+6iSJ%fk zhouU6_IaOHZ8ja}PjQRG@81ER{B5fBgvXJOcc$dE`#Qe8kQv0KWn%z$r2_k_%}wXou;RF>+tN25?J z%))*2L%$_8Z$mxXj)sECn2g)+*}})`%SP0J)es4ljZyC=9QtUThPovyuouKrmQx*k zPTW}5%`j0UM8rhpkx1sN*{EkZZL@{38=!VkMLc!vCQYfD5w7hV05m(AFd47dm>Kn zpzc#@Q>J!@b+ZHGiD;Db&SPzeBlnEXpOv%!7P!^4kPL)cqkPh`8Bh4!;Rt}sSk1TI z5lR1gSeKQG>p+i~vdZUW5iQnnvdz3xOFYw2e|HF4u1I6hXp%$5;aWbKXc}HVy(N-7 zhXN&%r>Y{ilb-dA$8pt^^_&za<7`N&D#2B+>(kUS07^1w&kU}+l1LbdRaaWXvPH)#o+{gq#F zM=ck<5+v}ov(ivW7j*WfTgVgfE%_f$Ximo}$8Sk2I=I5;+q~P9%0CMM3YA3&IT--` zS$M-oPE{=pRSt`#cc^=_iCIFoJXVUziIRzB^k1EmMh^2Hv;c;snl(z?(|2@hgj>Ye zRD3i-q+Y_D+AT{u>N!o}K;^i~uWVoPHF7m*LN@UGfJL=!&O8zDL406lSEt zp3NkQ->u7&tSof$jX3Yu9BkS#3KiRmJOCO#-;?ful0n9nPp{YusZ@0w8=*3w49`{- zv*}j#;ET5353(O_bdxtmJe73kYEfj#lpzbp)NRqPa7i1eHC{HL&gak5=V5x)7$rS)+JW3JlX-(tRwoPx=})4kp2(N}M}LPuwB zwvA>?x|p8-S~sE;;8^CcTzCfvl$Py5wi1E1eOjO@lJJV%tk1OIh3u^~fWsN7HC5aM zw7sVV04RbqfVxZt_%v#aqrwSTG+VFNUQ(6jMDj>b%6JHl6%>dw_zL#oCkeWS7I2IS zQL_Q5$oTo->NO(7=BooHS^oE`=vJ*Y1_JS&fSdw6@3<*wGah`p0t$;wN5ahwaDD7n z-FPf-xyalLFT##*=Lb=nr915^Ik>fsxm{1^S8V174I6HkFHe;8K1`IFz8&P)w~Eu# zMJFKdnRD_w*=GvU9=@L#e3_@eT`nI2Tm;C$E0~!JhU_o=ha+B1C6N2OiGhx90>#M* z*aCMKOG?<9*@XyK*LkP(`SXm52kdxoA=4FL0jFZ|XAIM{KhX}4`r=oV`r@Q~jZXae zALr=M9iVRuRhgei8bk5Em3buR%IC45x*AlPTt%_{J zT@AizDqZ*~WpW#IGKX3pAjWQAuctZuE^FnUTkW!}1PufAGd|}x?I_no`y4T&ZE94U zGsQGHVrWk8+t7E2xqBKZCOBTd*U>x7gbxn%P@KN-Qu7j9HH2R>FZ$Wq2q|ZBR(LKw8{oaqhvglba!NH{LX#CAti}Rk) zedWG#FdA{5$IYo;Hqj(1&c0+4*{(G4FN8rUK>*%>ZNR4eS_k5ON-;ZU!lbJ#!bPH*;UHlo)X7>c_A z`54SCA+3`Zvy0#-oa^_khbyOe0G^8%M-RVYrl}>?+=Z9KM+#=j&rn1ALZeszyI#nS zjvVoY>>EnLJnJ`5`@4=%V-yZuj5p0vtSAT1bDsJxEJIG4wP|die80O7lLInd2z(~7s$NRHoG1a*j_xg>c`NJR!JN2_2zOM^AF#IL{1{o zxR%5UKhEUvb6sa)?p^y;u(2MnP1cHck$KcgVn6p1nuJU2_-k4%H?W-(G&4A}On4Y< zpgoo=z8(NBki6kDoBE^<@TW&&_cRYC@r%*@Xt0JbW_X;2EZ}+gzf!%^xC(4?hL+>> zoG|s+QxQl&R=a}td(%M9i315aR2TX~4`mIBw8VnihN|3VtT~|4%C=_5ecBXG>zOc? zo=9RUvDdc3Y89rooSwIF)w&%p#Fy()vQu#{Cr-01P-@$4jm1~7t1n$fEu^d@bff`L zoDIodsI_jX_B#Ch{cywW&{iPRXo$EIPzN%cdb?iNoRseUc_1y#7G6v$>N#4hawi;I zSy`<7Z4dI@$12$%Up`$2t>Kc5gclX!Phe<6>y}q%yt~lWJ8j06HMf@7MgVtTcvkVg z-aKm!WZ&<4bd$81e&=QqR;5pp_uB4h)wNcQf%(mMQD6fQ%Dy{;GPiFvn7dcgEj zd-$6kS|B3MTee_$-(52-nTiko4qvL4A9gT>JQ}KkHBA>EwOzfZUmbe&)FR0*rjw%B zZIAU`Y$zlvE|d!tRo<}xm&PJ6q3bvB*{3PbFVy(e+k@v?T}<@ky@xV zjkCFr=v51~US~|0H5#^>HZ4@EignnXphys)!kNj+y#BCuoRjAb3`b7 zLai}1Vp@LX%_}rH99!U&UiK7Qm%_PS;6l;y;?wr0_p+?x1s}e?jf#Gcj5}qFcb7(l znHz9(v#oPpzt?A%>VDAh08rkPs?g1G7XNt2sgGpPX)TRcNpl6i|Fj$CVzqqR^4ZWG zti;>q$EXg_3}~%rp4NM511{>`(4d(-1zL#~UQ4T5o^=?G(ss=H!4UvAjt_!-W1Vk>d zY5YagQ*8uExS>n-7~xi$Uf1IVw4^fWY=5yO)P4UW52VP_Thrv0T>^4GQ>(Otx7cI#??Y}2yI{AyZe{1tG2d@ zT1U8ldre89v<$Fi2u~$KtoSdX!UIh4%3D*$wY|n|gAtdlI*COO#X>GobM*4a>CXPG`_?kNXqj z>XQ;_H*_Ns8S|MxQ=jvU@ z`m@#^4Yu?MC@S^jM{{dLr-(PNdVYaUJAyL`?OQoYv^1Os0g=9OWpu}7INkfV*6Hm>8wc=`WTV;%Ow%RYFXcp4 zt8@LH|7M5_xh*lD!JbM8E778!EHIJ^7dh6pLl}M1Qq!t`R61ODS@_ZKfj$<*cIYs* z&575R*A{`Vqp9o}mwm_Ax1VRtGF&F7ure^_9DfuN%NaBYO2D!xzD7KMuogkkSpSyOZak>c;FgH-POk$ zc4O9PI!gHN^7XVagPRd11H}9Kgo=SPrgbk#>$NOv_lNaG^X>-XIqbNCK>)b!nEYTf zt>M-{v$(RxW=yu!}GsuzK;B#x_=Vmi(Z2g4%)q@Kr@r7JIzA1vx*K}*Mr>Dj{8O?6tS7)i!PCH^jz9l3!n z-EPjF^`$Ltg;=Az zYHkOGZK<;xo?8z+WA6QMYcnMVKl%Xwxa2g~gc}NFH0!RI_d>JMWj^LO$nuIjWn4Es z|7}1Z92|d>#HV}Slf9j%K3M;m#sTP&_Iv)0-ya=9kp$2ja5Uaf^2zb)j&&_GW~>oO zV3#TYyO&PY=P$QD05GcEXNxW)?nj1vUrhlXZ;9#;EhNXUqmc)! zK`3o^K&OHd)I*$8bH)2A;^PbUk;@)J2mxCc$M$7z^x@d?m2w~Vt_Us;TEzJx@dMg|Ibb)-_MOF6~=vJ@U7TJ7`6Xsmks0DBV(N0<@ zfL64x1|eEB4Q$Oq8mJ^=l2%qqql4;0##LOpWX;`{hqZPuH($hQW==8pRD=__Yf{O2 zN{}cnQtH2nv8=q<;&X;j41ediTxNNLyNM5}+1eYr>oLoI24AFiOx(!=)AP+i*Zh8_ zffLY=`cOD<)a(}KvJP)oU|mt}M@+ZIenQ=fc_|(7eQ=MRbXOevY>PK}*U{wLr96bE z)XRxzV>I(Q--Yn|hP37*?MzCRB`+~CPZIDU-az18lS8zdqje=I=qd=GGU2JY+|>&5 zk;UF?bMX9a8*3=8AKx9y3^K&O@eJ zc2$T2W`)Dm49T^*t$J~$98tMpo>VUpNO*@ zx|7dvFM6BuozuLO6xGcnS4sg{pc(zy%kv2E2up+E2!2Lsv~mf5v=cE@5tgxei0&L0 zyRgz8Fm3}Pe7)$Pa6?J_J@JoLV%&{den0Yy4?`=T@}s*?vKT~!UW;3ygG>p*%HDla zIQ55qM~7Zgtf5eY&F5=e<|FpEDM7b-MDY!N1r_TIBA3ufz0>?4@4^%`w=Pl(Q* zRW(?Gigh8ZLxg!-Ny@S~97iMm~8FHvB6Zv26yp0N130s>1(jV*qm(#_-gZNY0$#}^EY7a+&RB|!S0{`H$v-8A*dum3c{ZtDIrO3CM83Yu9LfjkeBfs=K z_h$QW%lU+$ZK?$0m-Z2FmHl)KXtP#mtegwbsH=> z3%z9t3*iT<9$u^TW!QWnng5{|m&4rEIBL!4Cg4eby3nzIXFzCxJa*Ae!t7?bYX6x9 zWYuYqS35v42=SVBra)0?T0h|brv9F)U(9Tp4jxwzp!}HIBk~7Yym2YvF%F-<(HPfT zN;aoeBl89=PfAx5ivA!V5V>g|sFv2Tc8dQ3oVJ|NRyJtQ^oU~HbMwWPyA>l216?N5 zsWFYJkZJvdl9@5O=fB{sB0A{niT2oL=mdvrI*oBAO1PO_y9q4xc`nCSljEbj{8_3d zf{=*RA}mJkf4W!F;-D6mX|QO>!ASwC6v#n9n%jJ{qW<~{@>dxVk}N&HWunZT#X9o_ zrv%WIJget`OAGQ#^j0~#u3@|ro#N>jHSU*&6h@Stnx*jIp}l#vQ@Eoz;PL?YWPRkk z^zY32V%NIE1}YZg>6Mx#2{y&YxHrBACsiecMpx8iq-U$fyEdY_;r)D$y1V$<8H0MB zMNhADF2>IFA}q)6jP7mE_`m3g%x)tBoJS(c0L2Zge49mF@##y$omY9nJtMF#y)+WL z5$d^;kbBwIZw|Fq)0!*FyMY?iz2yEZG(##2vT5gP`OEwVF7C^{CD*N0R@i(yP)D@) zDi|CV9w|BWqI~VhqVe82xgE$?#>TRfbP9*pbB_3q&qIkzX@%y0UZItQb}SQ6x3O?{ zS)x>A`hIQ>ccW=#xiOuHC*(2N2ve7jn(mf6YM)~z!4}J>eZjf`1@d?~hcW!U#Yi~J zmQV{CAU#k)}z!*iJ%1?D(j>JQD!KB^Ic~8#xMBA9o_gH>Avq)J;|L#riVI{f$ zQ`4!l2Fs{q@v^m-Du`QmlOHAit4XkrWb_&+xgNKbn{#=o({wuv`~@Gj7FNnrfCxKH zgelM{`^I!uwZcgl?AXT~?r6DK9!9Sj&$4QD5!#(hX?J@gc6T?*e`RhS?_4wKmxgvQ z!Tfkq)M}|I^;M8)Rx3l{g0EeMG};wtELCdQb}CGGrgnqs&#MXW(}5o@nj~;i*NK)cDVlW8BnY0fK zqH2)D=vAuKLVcNw>#xbIR-^%*Q#T9tQI;VZOCKsE4!kzL(#KNvprS*^ke0$QU2F}T zL=$f8m3gC6{DEzNhjOOgWJOK(SjN2a=53P2=V4AtDGjuT zpHKQT4%#ZGh0D-r|Jcoe)DXFY*zMUXp4A;jyrN3e3R2}Yd>-WWGC+` zDm1+>A9|?5hu9A+8{?%k`T~0Y=xOvg8N933ht%hO{0W4_14lru%*84!eLxvP#Ay|u z&gZ`4AYi(bB_zfkq)zL=lchw=W|BEs50uR^7K+Ngex|r@)V>J!LB^wh_i%TmxDF$> zue>ywFRiH0(|kc|Fo#bq_p$69zjlN=F=dOJA+-g-uHkumF8?s;cwC&&|LY`ffgNvj z4BDQ{RVvX-YeZd@#Hdxg>~*v0Ml5^al)~@*`~gM6G&L=4bg#JFAC- z!(#4NE);;m#qUj*m~(DH(AYmrG&?il24|eVEtO%?{870lY3b0T4FD?G1i-R9yYHxE z)TAS{mUTn%naU?JB;+eoZ}kjP#ol8bgE(x(#RO>ocz#`=}ZRoPASeUy}OV@oVIY=cpJZWEWa-DsiT)ru((5V`rCu=mc{vrMzCmUtXnv1fFbyNLb5vt zC(D~0t7|NUI2|8jUg~XL(y?bCBOFuf#+Se<8|hCvY?YD3{K^5^0o=lKPA)OBGwI8e z2Hc%CokD~WUCdu;F$g2(QN2gtrQC_(ECWD9d4U1}6F~xO9lcz_@6^h~ii`AwVzmRR zKmLB{=i5JMZpKMqIg`b7fnQIn_^2m-L=mk^(P>km)gazk-BRo6??Q%7n#_h)mvYSK zcJPkfbb!7ee}#MJ>oas6@bRx{kp2vBo#P@JW>vn-t1_~s1w5Rvaeb{IRjm{4$$6i7GwH+0TZGqBFO}-aYmxdj_RaFx%91OYH$XrUR`{ z`|rOI-b7Nb6!lKURMh~psHVIBImv%C6aghKwT*STKEYyAP0)Zf$(XdmYT zME5Vu6N3b&-$G1!uIr)Znv;G__9Fa;n1+|Wr_B*wOj$XkC{XvMoO*IFqwtr?AgVne zI1$d|Kpk|sghwGEA6kMVut(A50-Uf5n+pl@DEf_vt2YDB9k|q*&1~l!krPOH;xB7y z^l>_~Qy#J{Ea!3qkIU3^ieD#rF2!uEW42mC1zb*FMb0J7wnUq7M{Z+;kAy@y;X)97 z(2!lU$v)5HK`&}ERE_!G#2EK|;Apt{LP00)fYNZcz~Gr{cd3QymMvj|d zh?T)M(bU~DM`wqlG%hcIM*@?8r{QuuP3t7W4~>f;F|(&DL0N}b8GiQl#rN6sXi(wV z>U{SlQ`?XJ4`Ak1A_(~|V~cRQ7B)O-TESRyruA>VQ<4r_i$$s($xYIM1v z+2WL`EaWsn_q{&K!tET!uV^XPGepcLKU%LCk=eKMa*r)r9c_+AvPOn@z}{aT&;-2% zQcD`~-%M}XSZ9`s8cTcpUT;COB^qT$fi5}`H}yBeLSq=*zHz)i$C{y1$p3E4nSRy6 z&EIf~b2OVD)ev_@)VZAFuTzcyO===ssTxeIh>6tU^73H1+8alzA%EW+_~K+HP%zo{ zkIpAH-`O?~fwcQOmnHfysP;fC@&!{50rk~4iwpr|NT0{)hPc^mc`mhn;oU}Gj8HJ| zXbG{6-JVQiK@hDY3F2NG0qN_kW^%KugL%aO_1vzeo3o6G^d1d^_bxbHHFIjYEUhg8 z)&Zo@v8)yY|H=czQ7aVkQTK~Inp*_kn*He#W0n4?@RKxtZ!gU2!vp~IjyY=P5r)t3 zS&H3iiEAq8Idn%RjYpyt@#P1XNk03mRJiM69kZBGf0zTNiLICQ!SdjwxR>6$8YFzv#N9|m%zetEDs*a)GQedPBVjt z{`wn{P->U6f>~lj%S9k3_ZCF~ulWDcVKUzKu=5&P-t5j%v9v*6cu<)$qXXJ%` zar})ck|RxG-vq!?`Z0Qpg!qY{o2F(vH*?5-N(q~PluA~(M%$5)1A6vDv=5F+0;yNe zN!Sa=4>zy6PLmBG=K&ad(||lj0CV@XbM6vAM&&fNuj&1oy+< zJbSP`RfdUhBQ)n$sI(a^#&v|WMAY+3FKp~|6zU>R>a24zz`8%=FdxcZY(M+q&1vgd z&tU)0A>jaMCR70B@R{#~DFzKWV^^wS(fhJR>m~2ui_0I6JW!b)sx+2e%dS<6HFBpuC}A7fbU%39taleP6~)T?ULTP5}MCY?W+ zr>~Pib$I!Y+l9VLGQHpSz0f9EuBXCCXMelqzG=!7rE=Z#em>`l3@I$?us4z<`aNG*q4#?*BN@7+#G8o8hmht* zt4d4creX`-rc;xEb9PzGQB0m3JK1mC){(oZ%hVAMUU5!OdEfE?B3UJR!_TwO{tNSE z)ylWO7=N^m#ehYp{e$3#97efn3uR2!R@gv`#wwHuh8vR_3pDOyU*zO z&-w4PLPx&r3~e1e^R0Dyn4RAp9q9r`Nk10lY5T29=N^R~lJ6u?YvIas?*%eWksVa))XjLmWx!Ue6{4#p@`$+{t42;@;@Bhz4gz`jfAleWG* zG?C*%)4$d(%LeV_ln0LOURGL##21~kLTxp-?o%&lJm@y39iI#P_{+7KE>~AZvTmg= z?3*~3%veNc)EG<25x}>9)5`y;fzRq{`v;hY?0B!LY%k;Y12Jc~T7b0+8UMWv>WlG3 z7l1kiMUR)dt6>R983^b0Tm@SmhmRyNRfLB`oD`%5oj)}+&bwkRUsai8O}0i$kpJk3w**HjCBX$vbKF!8&ZQF3FfkRV$L$qFpdO+d`x61$t6iY#%zX-D{3$j*Ku|$$daX zD2GA(#<2M<0<&mLtxCl2^mPIU`nr>eQ^=73X^WMI~Kg+fzFm(niclh(BVWy@HoVLglHac;i9k zhQ7viFoKVWYYwM%MKcF|=$n5%PwLl%GAUoC(!RnG%`)JwJamhlOtqfjdFRqTkRg}D0a8PoT> zSsLL+L*o%xKav1G&YIrn;OL|7DN7BF>vE;4QkSi{-D4@w#~b;;u`vu?T`oKqP>M4q zZzZPXi!F+;0}>CT0r^3y&Pv$Uk z;U+R0c9VJtfSSJFqQ$=GBJ}JY>9YKw5(VwY7BxFrdVuY+@pIIG%Ya~w?E7vIeXr*W;VGcNr#sONiV(Z5@AYwP-Og#m z2S-j;1s;4IDW+c9k7Mq z>_;e@W&t4E5oT=Y?^but)xMl2-U+AfFi@L=IuBI{Zw!+Wu?l)nDm*Q8(?OT^&9ay4 zA*wmE@q*3EcN5QcRBetcud(;h!-Y;gMuv!8%hd7$Na^EDtMXt~+xKkXH+$H7H>u!t(zIW*Z0WW~ZF%a~|s$`SL1 zK=p(?4!z)SkOE!AgAcEcd<+xCKo?k5t1XtnY@FH7@3MINY+Hx9U-P*M_Y4QM-yJrH zJF?WEt^8y+pHVU#bZdx7uKm{OdlbRB5-qoUtyL&T&QM+7S*ix^yaEcz1@I^YWUgew zZifqwQ?Tf%p;%Yem*sJvIve&y-Bzz4-WB=MDaAl{Ah9^B(*ajt^ik!-{_MVOf86Vi zgUxEzPJaH?yGdDoTx&Ojaj>hsa_O8w2U8cB1V<+3=V2VP|^mP#a4*^vLZaFDz z+KOx#>;4X1J(gJI=)0+Fm49_6KL=OcgeJ+*w!c9#5zYPir)En zzK1JXKHj!yubpiSyk~b^T%lyzt(unp){{x_?ZHL*5dQ6356LJ8hRZIEXE21k%a;BK zr~$8=pF?D?gXW%%KPPX-Y+4*7yWRx1MF%tlbMD25h-*6wR5AC;Z;!vi;(_dQjk&0o zD@dzHz)P_ibRL-RfmdS52E3I?%IQ`)Sw`LoVk`v7wIhn7xLXa5w1VV@7o(p>%jV$~U9R;|eyZDNphdJC zS@?)H0UOr-aJ>bg%KNQH*ZM1<_=ln`eQYI2up!^o#V6BJZ>te$9xB6~;z&JB%yb|@ z>6fDq`>JK=7@ZQmSu^k>pQy8_&o*K}a7-6`pKWrjm12cET!f=ZZSz^=;^vCOBHp0Q z68Z)srQYsxkF?_=Jo7Q5?b#7>RX<~$97^~ZL!IU1 zw#`;pyd1gj7BVeiEOE3stRLSHen+`jZMcM@P)O!VNpE;skBq8kuwIxreDVC9#@7!{ z5kx3AmMS21z!3_z(|1b7lGW!ixZgX7j|dGqJ12y*qSsVZXkCWY#@-e{N+< zjTv=feQ{)V$or(5g^zc^+>Li{`@$wk=X-at>1007#FvmU>!FVuD-G5eE|U89yA`ds z5ftjihqN zV(GgW4{WTIpZ&y+3M;tvyjm`3a@eKawOn01ber$}Qc%~Gyr?&prnh%^n3K#;Oaj^B z0{?_gQM#>rPj}T|?!5?mjh)y$?njAMJ)Qz0BF!tSF3|5%H1*BIB zj^e+O!oSrC+YU>h$5teowL!SQeWHY_r@ax5)U@R@T^qSC0aO!8qw8JzZV7 zsDjrrr{4xBNOod`>j3P7v^zUq{IWX|WJjt>g!D?A*y-jM0%@@4$iN452Ww|zOv+v7&UTIpP!D*$Qhcgn_p9nniWHdC7oS(v$;Y>43q!ZlTW~a}?+@VCzr~ z@I59?5vFhyRoGOx5e`Js&!2tz%)hz_0(RIEwuD$2rEy784?%2nP~M(mNgtNRwz5Z1kkUsV7GKaf{RRm(kOg z+xoNn7QPTK(#1^`v(luwZ35Y4TzTV#dMKkKEVDnCyK~~usTZ|MZKeaDctcP!4`Hc7 zh6uzXIK7S6L9|OvBGe%|qoMFvZwW-xAcRVU%K3VB9Jn*CKr=H(*e5{ny-?;&C)fQJ*cs9Mn?HZx?j20(=np)`xLXUi+<5 zIkrk5ERzK{c|TmtR`R`_;tDVc7m2u4j_%~F1xiLUCbFB`4$U3TdUxIE2m4jM-*=o6 zZMI?fSBYF@l37Ir-4_Y7S5QaoT)^HyPh0f>UaR{qOE>1;!-9E*@_MgqC1(T|l*)tY zbFEwyYAbLunDw&Dx3$*4>j0QEgK)>!>xJ^-k@WT7h{dl(^8rehDa(Rh1CL8zo?IS_ z>tnsWVNOVV<;gKzR8NQihr>tu$ZU6XAig?$tAC~J)&?ZyLB zM@S1dAhJzg$sax$drXW2C^bFRW#)lEEphD9!H<&_9ED&e z>nfsj3DS+Ebi<+>k(L&a?(UXGDGBKYX=&*N(n$BB7a=Tq(H&>{z3+F<-q-%l-hRK& zxh{Wr5sNvW`ONW*agTf4_v^0{BJSM&Yu>P-R7DoUYE8|jjJ|!|nD!Txm^exSwf|1VE4wpI1`92vD5u-U(cC46v@);a$CEf#UD_AbRKTgX$%wvtQb zasDEz?rAEQ=icpCo+0o9@UvoC1+D(@Ea4&4^K}olZoQa=)%tv|WYZSb%k2?>Fr~B>Mi@T{B+69D z(#e?WT7G3GHvaR+KeYh32lEB6ewi2|6&?BsE+YY}<1R(VCz9@JiXU9I zvc8%-O+CTu-3bFK>wHOkAy{+=tl0j}9F=POh_%DpH+#Vx>T-T8cXIwa1KGKV;pj6l zB&lUi+fI8C89&Xj#-hIN2ldI!E|Dc*QBxUC&*bJ#ga{b*)1XTub+Nk@Dy?c1@}J}$ z?!WB(n{#V`xP0SYdA)7w95?+6yx=So(;RJBW9SOo8?WH_dk@=35a z^XU`88kh}LQCFUH`R;^A%MA6s+eA1 z!>YOJGn_VU{WG4nhNb%^((MMbGb5i%?Ca|WsM!x13f$L0||tOiTl8O{+l} zK)$3Pn$WyouP2dJdv%(!i2QxAKvh5Vu)gVJ z9i6_iWa+&r8dT1K`1P=M$>HpPUR43N5U8>?`7qjdrRk}IsV9;I^AT6nQPg4Hc}lc+ z^#_=M_n~A+sRoz4*f9vdyhYsSnStM#Mm+O;*s`3LvEK>4@@os6g?wxT!fqnuFNZZA zF;o~~hjd{@MCiGFgInw)0xqT+hos!w*?VL9-wawf=Cl)Ld&@X_;fv^ep5dJeQ-kd9 z5Tp=N)GuXDAGy{9QhEQ#PYVFLM7SL{NqYA2ukl89bJf1kVzg?kP+99y=Z@e4X$Yys zr^9HI4)Et!hE7CC*f=OEg(S6u(| zUJb*neB*J*>mELHCkuv&W69d>D<#{0yJZPe48@p}TbC|&pXC9Q61%_po2|qvkjAIan z%n%`0;{o5@5%BYKjP2qyZPc*SZFm!c-C%b_E-@A9R|#f~b3Jdp)@4XtIHb<~Le2%8 z-TX?4-?wbPusN`P_v|RI$^ajPN(~;B3!Lo4LJ~?1GSjTzKad;U)CGvW=ep1t>*N!3 zrmjIn?`r(ekUS?5cf!ACXAEH5Cp+|uG!NQYR6wh36A6>NWOGQ5;X-(aKt{?*!G=vT z^>Ov{uqv?SWuUYVxk2Z}wl@VLt**VQyb>ce9lf)dbXu<(b3`~Ne=fn`;X7~n_LrQ{Ei&o-_7k@~i1KZ`)H@i@ zh(|XD8A=c-w8aL(I?uR8au?wnx@AHW$l{e| zpg`3cGU$WXY2RUs_0(tZV}eC7$9iEub$eL$jq?jt%YH}HJbwyHgWa)NtwwFHIr`2n z_wB@~mOc?yE>X8@QXb9WzO!vGsBEDYMg zDRDWUQ{7>AE?*fLGHq>Dc$-oXHy)IEd?VB zh;Wwt-EC%yOT)1CEqLX5zlSPBRog->JkNg-F>Afr%sA)yDv}&Sqyo<#vsc&JrcDPQ-11TzR;RQroIWJEHuWCsR?pU zhDtZk_VmMs^AnWF_QA+5s3Yfh2lS*1L1p_#(VXl7tDNx-T)s`rb@y{{ZuGei{g~q^ z(6gxX%gmdPr|1HcpnwI~^Zjs+9aGvtP4(=Z=IoO+AZ(nwRSCB#LvAP;Ru=^B8xkss zT8PL(uv)KP7YBd16LC+{12_z8LZhBL6(d3!?c?to^=0M630+-z#Vyp4`^VGn$Z*RJ z$)vpAd;G`(*~Bwx)_sQMniQkECIvlZ@ug9AecM+X6Qf`JAh7EN^s}D2RT#htGlDr> zhdNV!RcQ|x*7_A#QJYy8T!0~0RH$PU7Fq*m;sSx!jmfzO5rPG^Bl}u~vtC})p?lmJ z|I|mrzC7w~*tvtB)1r5c?t;>Wg3r&hcok)dypN>k+;(T|-UcE(jVJ2`fYb8p)9r6< z04021&h-nXkBH6$pu{UZoX)4Ud2_Z~1lqSiS(w2O&Fs9`t%O5N++q_?m4%K%nA8d^ zXY1?3&$h-*>Mp(W%fsnU@9!!#~v!gzWNJnA6>!>PAuzx2X;I+;e-g)}~g!Ybj;-mby>Pp>+|)#~~L;gHJq&Fm?6qO!wizZgK_ zfU4l`?gGk|`Runf^c59UYX6|jrQ{WBGu^oD+Waz*t@$tXeNVufkLdIed(Y@gkMvQ+ zc1t{}`jr8+MG|8iBO6(7A5UqIg~Dc8vtx2dQHpDOOdAdUK4sEn5&mS@LBBd)ms$9< z$+xajr-@-|a&8Y-ojWOqYT6UEfOeGxMu7O_*Sdv!a2${#Flc&u<~!vCWZQecyHM^? z?@)XMppA;2uD&Y{;|8&WG;_n)#Fy3PcBWq@U-2O0BNK&>Iax~0-Y@Egve(;(eJ*E* z@vi`dPxHq1yD(IltXo>zd4yJOoKj-)R!hrsVvi}IcFQpd7y{Qpb0%N?=vQzB)z(v- z9a69KFyeRZcIld-(NAsQEn3+fs78)PN~@q7@Y9#wO|>RbVBwJJ`zHzARg{vMCKTd} z!Hg`vB+5+O874#eo;E&2#NiZE_4`oz|N{ks&e$`kfo4n?a~lA8}7EL!)IMFHB?lg}K9j0`3? zicicQb6%1&A>uuG;yL0k&q!c{n@j)u!JDfA`yy(t$2F$w)Pc%dlchCo;_f$HyNpap!Zh%LG)Ggbz*# zI7e!6(wuz1DkN)06Z0fVhK6$^8fgQdsm94>?Cwaamck85h`Ogl`YK{(X9L)--gZs{ z(Abgg<~7j)kU&6QVuq?Oq0^aw^&MdPW}v7aaEC!10$;p}*R8XRBgLBeZaFToISjO# zNV0YTS>~jI{RI!~$Ar-K!`3Vwo1_pmzN^_zzN8M&bq|5z=jl*10?nStN6j&npokUCI06)f^F@!YU|7L1Gmo&|m46ZyOjuV#ZOA7sBK@*_ltAHEkbO z-H$zwj`70MmhWkdd}MF?Ihju>&o1vkUQP&n+?#YrUOk>zPOrhXG_-EUJUO>yzauo% zO7Rji2Jmh1XV%BJ}GAQMA8h-NRZW8|@;GNVnYyz?p1rO|!QQ_Y6Lq0e0dz z8`x!k_ysWwb}vciraCH~<#%vPdnb#qui%kav zqc$`jWEaM4jz4kpV5kswk!J;BCt?#5n=}ddTu`yy=!z49uQm&^0?+;NnZ0F%5~`Vj zV>bppj{YK8w<;x-`OP7HwtpdeA_6fl5W^Obxp+#oGzr6Jts3vY?{MQdh?TqGwyj=0 z#=KE$-xle!~QkYA*5ZZiw!+16d)|WInQ`c#x(onS@dTm(#w6)vboq`zkv5q z9Rl9Bb{72>mCsrLXswvK=T$^;^wF%AB@RT9J&oI}kXYEgCM1HKqrlLzya$MH9UWE% zFry+68G!gU_ugUEhC8zN86jHS_WBS(daS}2lR!|2dsE1nQZ7I@JHMIOC*q{0m=FYa zRHKsxy|!5;9;bM(wA$hhE<~kj8JEXgZXNM?vJvo+Q=>qVM53H#s+ohEwF?(xbAxf; z!!X^K{mf?1?%ER+>Lht&KUZVqRl}iDEi6Kgje?Ztc;n83*Sck78PV$&5}u3=}>?XuQl(-r!Wp{HuOFj^<6B1D9JDjy|e{_8EItCW7RaMRDiZL-XGa7JY z9GXCT;q5-;zsAx>yY_loaAI{gDcvFdqKu_2ZxVgyId*@v0Nmon-vc*h*u8#b=a;}m zNoX^=;!Y_1g$6Q$+P?R2L4Jt3*y~^3*JJ_Iqn`yF+8HI1)`^W%Xr^k-l}IYt6Qk2v z^r~gf_W9l(yzpGflf*!DG4Y2PO}w`@L|RzCaRp-qJ#TYjJ;%nqv>TF=>~OFEf{Xpv zFz=57vOdH%OD`7kFh%_~)t#~ImaE>ZUU{UgMl`eRG>)hFvGx4LD&ikjK}c~}uFKt- zZ_?LZLM4ky6CP)sQYG02f}Tv9dmj}{IDN|l`xtBO9eA0eAIi&xh?mslNAO-iP}g{G z59&(xRrd>OX#n&hI2DC!y`rjNAGq4mzcX}%!)DGZy>6N~z#U>Lr{o$q#@ z!0~>DtIX@GFN|~;h5l|-AhN$}IJrlNLC794zIjA_B-^&ZTl(PZE^7aU!QhnNB4bl~ zbG1EIBW&%)J>+VNY1}9_T^}s(mwn)&yupXj)$KZ%r>{XAVLo|I%!&elcXW&qjyO0%LHa_HMm2Is&sj|7iQrxln znTPJoaLFGs>+etzy{b{^xZSxtTIb~5nM6z8`PB9JC{OS7^vzR_+4>x9W}RjUH~1G< z(v5*6wao7gJiWSIcem#tq^VBnQ6}!o&MNL>d`V1T4Wz9VjWN;#mm@m%TNg3HtFW~oson5OZ^K+ahyaMR!NYa zDEoT;>N+INTbB&Hl{aZIrGD8fYp<+OC7bjTr`l4vDQ`Rp{B-TA$ zHv0?b>+M+)j-%&qNFbMh{mk1*i{S)|r&gR6%?h2P3z<^XPVAXFRbchEot%}=K;^-X z$CgWlYpY>5WdTt82DfOtcl`Be=+9oF|l%5;0uuyTQjPvU!g{BZr;}d-0`iYF9jVEYk69q?I zUBpxS&M0O?%i;Y`;Ko8G$m-~9G>PL3va9>ixl9=b9u$C*2va4fa%KG)nPan?m)Z-K zcPg0y9aqjFY2tDzrrA0Lhc!&+E|I`y#NjODwKma@(k^b3KB{0eRzuCNQifhKTEL%0yW zD(fbT%In@oeg5HC+hxK}Tu?0DRT6l#<@GFoq>j3ghH${PGKj^AA$1OL*4%0OGwtQZ zdt|oz3i4Re3qtfW5o9iGOQ3e2L&+oUOKAoypOS5&i(iGJZX}F6_ry1xJi|!j*dh!> z#`P5CA7r@k2MC6}NIVkiCdH4d#9qeN%y-@2-#C|!ii(!PCt=fjO=t!8AE}-uvH3CC zUM5LZ6hV*ZkI|d-WgE;0VLoG@RVTCqUD!(C4@_SM1{+T;Z660jiQ5PrsN%mN6z#?Sf}1remf-0PVM!j z3?47K9SvOLZa8X7#M6RE;TH(m3(@I!QSB0C&IFP1AlCbC9xS9qn1!&OVCXVDJ;8_s zRYX3CI6PpFSjO0y=1d>_Onx+@u;vFff%(J2t2=T0P&QbWFSg{gN~JjTv6`fIa0cBh zTv{LA7h7=Ks_1PFx67G*l80##a9Z*IQ|DcBCAM9;_|EWvT@*L1^MqvG+=TdN5T?UH zy(sce!Y-2KrD8q8maV)OjAUF`#-7R; zVH%r*fsn@yh(2a&;-oj z3wMxkOj`BrmGdj)XoF|)PE9*dIB|I#CPsthP_HZrdUR2`n(SyL#LYS* zGO^JL70L5Sj3ZND8cpVCze(ZNM>I|lAcl^Pv-yXVaX*Wl@v5a81luP``zR`2TNd^x zXyF)j7U(xy)+0A>Hn<4d^}ZvTTz;Fy+IE@u*)q89a5?BQ^mDg8!Y3`0Hw7-i^!|SP zfsa)C2!&REpT`iawauWS!jgJrcXdNRC`igutl*i+;ZzvXlWsc`OSyYpA5{rU$$Zl8 ziYzeFPwOJ~uu$x;?7igz9?Zd)80J-S8z>Vmov~!pv))rgynBRG?fv~VYUdl~un1J1 zI%HIX`oLJfW$kdMmsD~E3LZ?H2uZSQA8JUJ)geB&9tJh2IMxd(;*@6?Xnn{57q5fz(4QzGij3Ybj zwBBrDt0V;K9ZT(CO>o|oIW2-ID&k{w`n!)CA)0Zt4jG-_Ja0dW3#_(&rO6PS{nNI) z7-b@rWeyems4v*!+v~ID;~~LNht{H99`ASHyYmvPcZ*2vz+5eI8=W%pv2crr+_{r! zF9YW`yB)xUI%@9qEw~}JdLKvZvs$eWHkbTp5OnHUuTu0_*2I(Gx73Fqb*fp4@!n>$ zpexv~(qt>jq0Z7iOSu(N#gY}Uurj`+yFh7KnH?nC-=+LyuDsd*MAlTBhEpd`iR#!?F70v;}h8r->%? zx8gv5zT8u+&OZzn+N=8q>=?J{+WI>cKFSQie6w}>f%6IV3&H0^!H~^cugL}e1=8FGAJU_C|Zu2&U|`hL0v`l zbDYVTWSn^ZCso!%LA$VH z20*Mgt4qHwl=R1L<}&T4T!yuMBzxub#xUUS%N7Un0%q8F)epWq049A2dxYgulJ87BtQ|c~Y*NEi&K4U?VXHb=Hw_Q@D;o*>K`@cmq`GWy-4yWG|&+ z7q242iEU-jciEO34oVV|@+gS~yz?Yd-n8T*b!x$4zV)DyMpKHTL1#g|aB>mn))0Cg zjKNE>Zhq_av%Xt+J4JGJRU9u1D)b}gk8^)>K0+^GIt5$85=1C53vf(gOm$>vd`)P6 z@%S7}{tOXD9^S9(R|TdmOuZJ+o;OkK`L5)0Su7(oS?q6Mi|s(ZeFeF2BYP32odK9) z+UfWkoDm8GSN0HpAf=Q0g2e4jImIKfLDa=dNJD7g8yp@T*V{w+yF);xB(rVe=RQ|j z)tib=WIt^@|DLFYznxqXqIlVRZuuc7nT*(1#fmeG1Ikg)(JEOjJOq5hnL7`QNKhus z@`ppj8_P!_;Tm<(;z!HDMUt1j0=ZE#I+0X#E1}jTf>x%3Nz%RcwpUVPxdpT)DE)M~+G&vT7DItz%N^K|bKsJnhB24=pjzk-DMg+D1a zCpHEsE}wI!o$%{-^2MOV(XCiLCE=`U@RS~m4S2{^?=K!tn3{prM~d0ctP1N%`uuPo z!y$j;`lJ?9z2#jR0Mj1?-sbE8Zj_#5bJ3j<&&R);=%F;&JHRsxTkI1;O8bO^JXjIQ zB5;sVFxcv&JVdvcd}%0z))P(%M-GkadeB{C`irIOH0SjOZQdyOswicOXmzvGTrRBd z$ha7;6Xsd%g(malnb?l4TQ5Xa`Q-*(bq^A{q4RaQ5sNxCdzz-o*c5ctrt14Zs8A8| zs2Yg|+CF+SJn}4=3&TX*d&y%3F5wU1YBDlrvS1xX;{oIFa6ZapOrmTy9ql1%DVSb)~y;PjlVpWWF z;poqOANF}p+O+1+bM*g*Ip2D6_G$+rsw}{rvLXIv^)cpnx(q^5!lRvXKxu6v=z)WX z%Wo)!qr$D+GL~`aD!GglvJuNGFxiBibGVN{nV^X`ZQZQ&G&fNz^Bv!sLR!+QgxhiK_-d7yiunh-izW zmcGFRB;8n^3%N!EnBeDw9GM4o*eO)RV&$^^HeC2#uPKnEN@7Lt-w0(>R&>2$#h0v< z^;GjM>Iz$HbZFb|O?gF*&FqA==y>P0CXYLKAIA%zWix)g3FON;*}J{!8N2!(OBDx= z!je+sOj&o`Ef|7^l-z*g<3l-ijBqfyH}(9@VS97Hx-ew&JK4cLKwT+=edhD2IQHG8a!=H-JT9KE-A}G6-wn>xL(C3(;h6TO6SIK|2z#N6`dnq-+?SgmJyX@`UXZ zFsaYu^82LfT9huT+K+m-S}BIk#ovr}Uj!!q2(h48ZNeePfr>X;1XWF`vB{sl8&4&^ zSQyVYlv~0&>p$IQQ{Z0sQLkJlbO`!gW^<+C@=>|MPk2ruzgDY< z$mIA2HZ2m)6Z^Jpx{`%Plx`*RzfQr-zSY}k7jZw7zHqooYk*_G*xy1;EA5nr3)!M& zPoR6M54UWLOg{sh&==`Z5XB!EjR)N9YmwRCe^W_0VbyQm89Be`$-HZVWfavIrfcbY z`=2h~IKi1Q9OQtLClUaPDL)VWb13<*k4K>VfSLHa&b3e?RLci?>85wZt53{^97em} zi6byd{ZP16%3b1Ec&j8vO}}ON!k>&vw{EAH*Un2xFcDm5B3UZDUc3EHeP4Pl@NWl9 zu?$3i{-f`LmrM8j4V&gK523hX;hmeq>7mlZESh^%VCHJfGo1>5F@3Lbw}(5F!eg48 zL$a}_0BZsK~$ntKED_X4>J0b{F2bSh@}uxa+MJP-$}3AASjAM5VOtR$5B=JkzhVDa z5nVlM_Z}DS!_yJhX)qpmJrGud5$Zs|aK!A43O)LTGsO!r%w#PtPr>^8WB>BgC}2Xs zZhNKa_15p~jL&&-x>WNmEjbhQx(`kH6%vJS0s9O`r(BoyUyXG7lUrGYqVz|h@JU6X zvjc~iE27e11Yp5 zIkRYYm@LcotFF-UFHwpokX^zip?%z(!_4zaCl!l`>Y6^~j0#HBuG;Rw8q=Qzg^I+P zrTN+!im)9wvZjo`FLs0uLDv6Xj_hh9zjiy(tKuakqJ0x{elx9~%mi`@uqGcqkc(aw zU0*>G9&g|W@NQ4)7qhk<-d=L8B>g?h^p|PP=V!>IqW%)!bXvr}X@58HyPl&uTk?RG zohIuSa$1ZOAB@DSYLu(UY>b?V5%a5ONm$bz?kq#5rnX%+@MJJ3aw^)cOQ(DVh0Z6% zR=^jcDv|hX>L9~qpf$cj;&vRHuyc-1RA4)pN>gSHXm?;%d#+mX$0tI2=3FZnl@oma z_Ua1cO8-F-kgZ~P@2%~3a-(_%6MX=WIGazdjej*qfMxsSHlN3*Ph)J`s}p2+s9KAB z;X0pX^hNTtqV#|8za4$(J|W3kPJZVrA`O33NniVyC&gbME(b!qpBYbG*-)ebY|sMh zq*iR17DftP6csKQMnnHzP-EJ|^^XqIqx9`s?RQrRf=3XzF7iXLNnjsp-1aCk0%6O| ztqYq9!5x%VF!5@G=`K|3cT;>biod?j7aOl_@_2fq>3??b`O6{ZhmP2X&I=vKIImCt z>0$ggK0=?9Pyiy&pjI+W=1*OREFJ^zx-hEtNPZh$@Aa19VOt!%apqO5AMW&=V<@Z)(ZF{r3Mag`jxU0U?u$> zjQ@vU0Dk9({EFy{wphQD=o8O><4>(HQ#ww5hdgsn!fSgX`7`Q&{zLz~=seobn;~#Y zWd_}dH-GN?=6H>g(!V0U(`^yb=kb}|^_mUs;*aCxOKcV(fRD!;3;xfK_J1Fv@xTM) zI<7g)`FCy&2>QBe|ITg606WP~#sHrj?@!N(&ti4JhLu(LwEgE7HX;XXrP2Puf9Ji3 zNCFnXQl8__%^-ge$peQmk(Z9A%AdRvfBnojumEH%jD^25ia-0u4_E+RiqNp3Vf`DP zle)kHWM71&tG5qQM zQwCZB3lNipH~HrmmJtW+86;1gnCx5iJO5kdGr%?xF~!DJ=ML%xD^*|>G1T{6y&AXs z@%B#_=(7y63ZRB0A#r;DA5vqvBapxEV*H=oKIMzw&wfn$=AaMyfUE2PTkSfz*?Gn4 z#QG=si08>GOXIhAf zF28JJe9ug@sZpUu{FlLh_A!*vK-mG*l!gDIru=Ufl;ijP)38?plqYb1hR)!_Mf*dE z3*eJABv?!sOUIHS8C5j&%PLXKx;bIQV5>_BATMV=oTg(om_}<|dS0K(1^w#0*^hr#Z;uj7DqMKJ=cTCm<&~D2 zzVlH}t|?plap}%HIQw<>JtvRZ8p9W=u5Y<+<0HNgcUaapF`4nL+T%exJ7txh+$TdO z{qBYvc8>7(r;`o+zdk$N&3*laImt2W$n-OEFkAbzE}*$gU$;Oy2iMf8F-5(LF@FX* ziezg;rTzA{I;wdk2u-408TN*nnVJmtGl71(`AqMnPQ|;667&XXkM;v^uh=%(HceQT zPfV#=Q)^ZhiIrCFnYrF6Mu}QekRZHus{zBzc-VHSUh?ktgzdO|!~3vpEG_3?<2QdC zaB7e>8%>g4^7GhKxi2!Ii#Yz}N@`Je^W*OPz5{D|0p41MNq#H0X*V5OCuPsw7sqmVHWx(JB8jlv3IQQFC2}plx-I*;kVN zr+Q8S@;fhQ(c&c5G&0OS7$izZxg@u%@AW4@d#c1(8ttkerq%i`jHT4Ag*Dz0f|fp| z$d_sDYj@tAV7=f1GS%Nr8h+b%{*BsVORoCitYNB)hopbPB6&!?J_RO6d^tMjLQpI& zDI?$3@AZlp;eQP^47aC`Nf4PqJr9q7GY+YG2 zH2xsKExh@7;MM=xO@I?vrSrxs)}J!5gvF&=Wq4M&I-)>Nl*jj%XDX5&dra^B6vlGX zNwSLQ-Dz!vbXgKxggFvn8J6_0 zV^K0L@kIKfb^iRhzM;93=?GQaDWGSwk=6<@R;`En#dm53CusHHTAOqVdo@3H$(`iA zHQ5B6)(QsFxE!ROVJ9jGy_Le)gVL17=rAPoDQB?eHRx?78)g|yd)0+mPnLR3AzT6z z;(5JMndS8&4f=>a7HxpnMO*2+l5*nNN{cBk!IP>0fUYEJeV`sWmEi?UJ(w8Imr0!P zL4EUyMGCS3acs`6F&T^PH-o&H)4=j=>=nHsEgAU2vZ1h6JBnF${Kcjepc-x|$$7kL z1>M6R?ngb+wKZ#^B0@D!t+Js<@o72Pe#lV!FidW2D_{^CUay;UH1C;*@{RxS;HK30 zi|htpR%C-A^eb!gR_B=a$bU`bfv+Gc1E-?c8z35yAp4w>7vr<(ur5=9JZ#Z%Gw!hS zIgq0zAn#nh&x2he&)J1X*}S^l{HcdZc}U(S+xva(u?_-?Y?)zp&diH)SgI%sg0Z8U z-JZ=aToaR#Z2;Dv>g^Pwjfotn+y;+xYNq*DIFFqAA-@+!G);iJH z-U7y(uQg>F?Fa@rlW^qgsKE)wMuJuYZ+OfX6HP}&DpvcNYv%z|QnG;x<4{iP%Q1f= zoW!Bm1|*W5KB!4w%BEG}Z5}Eb71~PC6rd>(2jpMx6bWFbP!fI-TmTreO}@XzWh819 zit0;y))lGaLoYRe7K?oNVF1|2J*H8y05MLAkbBFZfaH7)V<|0Si)4E?Uk57|2E2!g z`fk%-#$PJOaerj53WWo%F;aJc(@d#GWMQd#xVpiLKazg?IIeXOW+s~QOUDnYz-U*w zL08`kK*vMU19||(@=Qb;I4<443t?5vInDBJKnB&>`0{nFn;cL?#u9Jl6OfS1Us;tCk zU3g15pfBV9vWij50*a3hgxUL@eC+tIROQJ4u42?&tJP91&d-=^e#Pi?cEyQr`roxT z`m=b}c;lYGWa^X}5XBkiQ(G=E^P@($U2XAgQnT*r{kA6FxZ5*~)0ZIPeTt@0;Q92( zGz@8?`7He@Q7HVE$H;)j?%VJYuxU@CeE!=ePTB(TYk1iAfIQ(p~gmp^K_qV&nL8~boh>#=0YP<#!3{}aDJ|nHLqK- z9A9knNge+tlk)fvv5#VRHbV944hqf>7TC54gm^NIMkTen{0e<4A6x&Ncv0g#br`s_ zx&wHDis5}bm-Azk#Fn_<_L`)QLD_GkfB4Ew6Mu> z@H`CSvH8ZMWLPk#=`ePgah)G}Dm_dDV2@*x4Rt-Qkq`(s@s=8hF zaW2l2Mh)A4zHF)IXzu&-pIQL*b`Y3{)KYmLP57e^y>fhA+P%HC7`D*_WE1ndRTNyC zXIg*E@z67htQ=WtBE8d)E#m+yt^$YP8=y>p|!o+;wQBHAvNZ=M=hl(Rcwq*zoMT z>}`tpoLL#-IhD+)|4uPos27e$JN3N@KY>2#?N;IncU0N^K?~A`fw0qYh$rid{^Y*t z^eNljX$E^}<%rLmtxX{z&28orcuDrDo78-UKuuVyTKP!^qO5wyB1`?{*G{?%BB)5)B2i&!pg%h~vPigw_73ZoJy^x+M=YQ1 zOrd{0d*z}?9Rgl11BV+C*i?1`lSABPi4fuEuMgovzcc(YpNq&U?-)T@vx!lqnHH|l z&|7zVlZP1CeW=`ZI%cw8I%YZs+={RQO(?>s-lI%DeYIXs`c1F5qmKqG@It${ zp7*0>pj)*rAa^5?C7RNZ62cS?aT`4#>vFfCmdM%DDFS7hjl&z{%V2~RgtM-1m~W}T8P;Z~?{u|cv~YFzNNWnIeaxV-k4;cQP9;|F- zI*U0gOD-O2HrTRgRF1@}EOsp48awW7BbWl*xhdwm_vcqqF2Lu_g5GMMcQ z_a4m$QBt4gD48#LfeC7MWDiTX5_<)7w-Dny8{&J%^O^(E!#KEG+~kSbW-&O3Njcj1hFG`- ziEVooQ%ZU5=kbYy$9_FoJ^8p9O&e9JV_xV_9R(Gv6-JD{e-f@>QlS&Z*i9R9eX9^2 z>(^-hMCWWT-m97Fx=p2K*)P2KX=GMh@2Fd|c(u)i|8p6_XwzQGJaO$1yTv*Ux1D*d zZk}3l^^g~toK9~!#ix{!ze9w1EIuJa$-eio@y~~GXvktxU_XcdX0zZtqr<8!Veu7h zF#BRX))*QgL92kL&Q;VDt|o1OosA~?)?i`=C=|HWuqLwP0PO@4jQT@0KldwZE3w4x zhkMC&T8Z+7&M)BQA*2q3=qBXa7UK=U?GNvdd+~x4+eP#>s`EnkdN>^nC&y`?mF`S; zg7@R?RDDbej5UYfy5?0CQ%N^gvc)T9;jF1v1dYln?h%&L0^s5PmDd} zw7#kd@Kgn0L1GZ?qt|9x(_C(D`&Fm&AwNF%N9@11bfplgN{$$h3xN|DEPm~c%BIGc zlGn?=j+I&YowRp`9*5)4lj%vIeIxQ5=?W&tSM-YS7q%S0jaX)|0Z4$~l0<5$_lR?H z!5g@-jGk+A#vyL~S+&@rltDkjFn99FziBLSeS?0IBc)W~wVrrQyyDcJlMuO&vgZ~% zdy~QCkQ=foBTK!&{!2WZUau4=5`1K$kk{^eZpw4dT`RtUP{r7TSA|225TgKzR?A!S zSRR#q(Am_*!K<}#+lidTzy@w=&2H;zWN1IhI%|9VgW@)K_1fH3?vgm^uwijs1Rr1x zuIlIGf%5Gny*ivuod1Y^K?EnM)il+b8bRV#5|MzWErizj)>An$bK@1QTb|F~|63wZg7q3`gx;=AKv)wLY$MwxTct zvzNS~NbZ0QeLPCZ$SuvCOyiBN8%4s+(@c{w`DeU9CCmd<#Hkc~=Ebc@zrqJarvd>J zffArZ%i1H0s56tVmszX&GK%{ZUj0$K=%Fe^8G?&Ne#<^*Z3{A0alkoG z_;%~j@f(AguiIt9iO;b5B>Gk=dcuu+(u*U>Bc6*5=5R9?YvxG>S_|h=3-Ys5fd$l< zU~T7EL3~sX&UTkHsA?jB1VaCM@3nsKXi^NiJ8>e-R`Dl}-X)>QVXkmz`mei7E*6Sv zL}$Xs+9&lp!B`?SSn7fyE6XRPMROSdA(Ep0+f^VkJ}#Xi7RJx_N%_9Tn&0z?bFwWO zYC8RqD1;765BeK*Fh~Sgbwqy8;5@l;OZG_qnXdjX@LDmD?7p|6^rb<0I8D?{6-Ipz z?@`3&_yPU0D*&X>!e(PSJ)VD-wdT{bWa!bZYB8AIn!6Nsz0~G0nNjD&gaLYvFTvud zSTRh~yA0f?U1Nu#>|4JPBjjv4vewy^Ve|B*M|HcQiaT`s(sn}*WD_}5WYO+~Brl$+ zGV0Yc16B|;oHJ~_PvJ$GV4F}l> zA;^BFrM4JaXYpm}^CoX$pCwVr=W~T|Uiq+){E&>T*8YvJAYdTn0La46Eg8-^1 z@>%p}XfO5zL6+QmGjwV^iHl|(Y-UHl1-C6$^6A;WCRkRPV?YMKlSJ#~N-9cl1R}>z z-kY>u8sCMXFpPtI>uFa4-hk_Ei!Y!6l#j$8xaDm2E{||a)u!jEEmO^adN=9cZ^SRQo z+D{wU`Z(an#y!AKFv+tuZ%Le;E?PaedW=Tdor--#;eQzykz`rduH@kx_+r91{wOQ@ zZ5E{;^+k@=Soj4^w?jy!j7~hqI@!(*+*PMIUvA|-=qg&073JN12ijfWqP!};N}EWn zr+vyr=^$N2=WUtrKb8 zh;@eGRSuiTb7Yo-sCy%dh6P_hM8=G4fD|DeC*)1mZ?Bg;GXF~5HZkhB6t}oSI%QX4 z@hdy@(YJG-V*5uA^cTuD)5Pn-PCET=10FmKQi@5eZt%YemMZGS3!b*tTlmz(7#l3C z-`)371PoXUJ;>z5zNJgO&3nxE&~is2p-%Wf;$BFA;F#C4#Azl2N(w4 zx6gU!+9-~R_9&QvoDME zI*xbUqJ5IlD4iS*+en}u zV?$p4D(w1G=8-;t8**Oe?UO@!-vJbQq4AgwA$;Efzz9F?TaudL>qkC|$1$m7xg_Sdw=^Ezw$%8}<_LnIE~`-N2-iCVM%*cEw|I@Q65#V}D3c?eKLH zQnsyLWz#j@18l1FM&p)_YRdyZUFdMkd5dq!RX010S}_o-zSyMcAxX=38*W(NxC)Gt z+cD}Hr{!l!#_!D}=Lj4p;U5bie)izy?XXRr4)k=s1%ErVyG8dj{LUXojuZk9C?dkL z)>xrb3dtr3^RQpXMnT0c2hH+su8HhcxOsqE2q%D_7XMK{jTt2)htfaW`TY68JuLD7 zVl22ZgMJ_M?07zSDR3&8*<00Tb+C1Q^Z-yl+AtK#(Wqs}sz)#e?;SO=+`lZMJriG$`V$P3xM{n@ zRbF~UuLw=Cm@&`h^h>)6i4N-a=MiJ=)Z!K$Kh4o`EP`cUx&qRAOcSVyq9#ddY00lU z3jbNybgmfxJo733bjc5(n>!Yt_$uxGn*2OpFe=xxHACBW_;zLNMDBAr;W$bwJ!L^! zF?+EN9-%_h!IXkJ+j$z=sR_9psfPH$TO{$nA(+C$aQ!KZR2ZH{5N4Pi2%h~kWHX2j zT%Tki3UphYpXLGZ6t;8VuE_$=Ao5H>=SIh z`4l>;rC0REW@q2`s2TSZFM)*ik^f10^F?8X*JE_Df|uqIqGPVb-X>2@WX~vho?J5{ zaBa!Rf5}<5km7MQC^oZ?q2c&i&2PpVZUhM6v{^(hJ?z3Huy{KqHo>9NB5O|NJ)h$I zp8Pv$G|DgM$SQIb)`b?wR-ifur$Pa33wH+hBgN7pZj2%Bh1_j|z?XxaC^5)AXQk%I!AS)EbPICA>73G9?mif4}u#Djt`N)7F%g5N!Og~r_;MXq)rL%k@Zg> zEUap4lo~S}L+hp5T~9dYFXQEaqUTYxNMAAY0X*b`Us}^37v${g%Kr zsaC9&UI!ShD*Dn!VKa57P@vHfSMVM4z=tD#n=46zo!;Q(;K#n4bBPsNEImQP?m!Z{ zq{T$rPrN*j#ETR1Rj=bFBH5nZUP@EuB+5lWS?v%nk`94Sq=ByY3YLUHML-Q|Xw(-o z%=7Z-946T~awm1MIFz&kn;xg_L(A($luC-v&i_K4V(mft6+Zt1bxJ9S}$sZq3e=8D?g*oR=-}#;Cr8F*uG)GBdXW@SJoLl*R(CKT|sUmG}p?Z zJ!Avl9`|B8Cz}jR%c82JHIO|-P=vviUoYa7W?o?PW!XPNaPx&f5C4mR ztJsGHe+w@E%B+GnY^-i<4`P0wD7wTN!|hqZhLLC5)lcz9Ve;Hs|z`q zk)&hrG1+{@#1q4#;Nm}{@d$LV|5GC6A7i-zO{(l#uIg;hN+an%j#%Qzh4tAC4nJIp z8l|=)d-nd6n4ONY2e-mSsGe7QTRLnqQHC$LBW*sxJevvj8$2|kzhevFdh@grRBL9Hjk=_L0cjyT|{))z8G`F{2tV+H(w8_g?NG&5P ziWM&hYRy3Y+(OYSS(`jO8JdNv@?OQlhSY(dA3bXhfApiBQeu`(@cuQtbAFBfe#S}-%(gQ^#3pFdI4Qh#Qpt?G-QeqgE4xk%#B zS1>BKihn+E4B}R6YnrfHgf1eXuvV|x8__NM=H8F$sJL89x6Cxj(K zjtKS`QSLAlxbx5a-B;#^j&a>`giMMT`HQ-j*gcB)VB6qUOpdBiAKqnl>C~?_u5|50 z7FW(>8P(ImKc!|7Ufzu#M)AiBV2cs!UI+EM{ZYwRvVli}IbMoWHc=DUQLHL{=DvE} z*=lU!B1nrP-{vn@=z56Tr!SD>9S$bTs^QlrA!% zWv#_3qf@zr+<5uhq*+I1bpTr~)&2$$2jmEMTCi#Z^m!nFFgAs^^-{ zRH)r$d+!T0GqroM5vHMl?20RMbXT1Ak*3QEEuv#honi{uz@ULX_c>MMbW*kC5x_RynzmV$m3NdQFd2L$6TS_BUcH1CUDjc+dDiRHw=DF%4?<|m{8(Mh zXB)~CK`aPxKwE6|pO3i>KExqC`Dr9MJZ_sE*kJ-LiThsvczrLXU^DK8crIUK%9Xc= z~o{M416NASx0%*&R&-XUTnUGLVqZp3Fntki^;3=D# zb#ms*+y3@;Gc~{jG92O-*b*3477Q#%WCdfRs@ddl z?5YQ5&j;G3A82DjJWHY1PdJ>t7#;UL)%yKE;o?3J7ehn4PeYCJuU}kIZuDaB(j~oN zEV+Den$>yHAq;jY&%1lnZhX-pO8o|M{x~Y`3ZG)=;+emEEPGR{W1G$VjDcL~o>X69 zu9pH9_c2OHw;}kYF0p$yvcR1Zf}=K?G_+yh&{Ys zgGHL1T*Kb?;yqm9X5%;p08E$o6dFFRFZcAy+7Mf#uW>a14 zm|4UcB);3$8GEVb(xTkQJJ*W4RBD>C2Vqmo@)==frdcGt#Ac)Fi<-!MDQlsF+UgGl0IrgcLd)wxS zkv^oZG9Ndxb|Q8A70X%_?jQZhf}Yd(osqn&VWgr@Z}hp`)|dJBvWRfHAKA`V2(eq2 z$uCl)i-(Y;3SwGe^zjBiiB&Q;3)(yJwe;)l9MNa=x6wx34GAXloAJ*=Tl56BHlKar zbTo>kBg4##37VaFre=ensk&R_GkB5T?-j8}QSqA>4V&TugRcRB1`#jPor z)puhL<1{rkW4*YJp`XdyX9f~G-IOzAaPY*8E{@{_Pivia%Uv!Wche1i{nj-dHF+*@XLHr8pTjeijWMBxTwfYuL|9^WO)HGdfraz5$0h= zr4-XcGVUJ-;!+Pu!C~g_sEU|g+NSMQIj0H0oDL5dj#Y_|$!%&sxa!XR zX1(ApwS;WAp3$&wOO%XYnjCStCb}k}y@j+?bqT0lop`4$P%JNcdX(39dXZEfR`@aO zgqsd$Y?0HwueqyVA?r>Bkexm0V9R?K_sg|nxKTh3C=wf=<4A!=qRX%`^w1tFgWtR6 zHNNn$XY)FR*r6WdzwgPPE)BILHPuxsNA%J{VleDzN8^KjCmiQj+48s9{_zCdm$6I{ z4yqg890K9dA52d2qZXoW`4f51293JdUwyWri_TH=Wsll?6-BH1$L3N9M6L zL-cD+BZ37PSE~aYWu6XHVEuxs=(JIzP@A^ic3PZN(_4`g9?K8C^WjOF=*1o;cxk+1EMH_|v5+!x%SgJDtxQWp(hGOzq^lr84f?_taPy{>c6W1as zGbinyK2j1oCue$|$x+qx6Pyp+8w%jw&o_(k(#L$J|T@)7Hicj)uiMH9OzVS&Y@ZW3PYZEd22_C>EcwuZ4hq zY?a@c0WXl;phSrv(VaH@GJ>q59lX{Oyo&46I>hZ$ps4sMAnAO2X)-U|9`y0nA6q4> zaGAM`8~*(7FF!*M&#!49%nxP2e2P$ip!*89dfv=9Qkl#pA`rA|nvD{r7{A#E1?K8_G~9 z8D#ZZQm4F3tN3ZG+}Z z{y=UW4wZHXOVC2Wk+mOLXrVseaXw$SoRT%7?V8^zo-pG(e^Sg8{Gq1>yP?9ynhI_T z@TU>o*R^Z2T0(Rvoieec_Fep}qK5WWF68T*Y1+gU;*4}%&pDInvx`i{2HRzeQ-4~u zhTtw+&|C0ZP)Vzp2E1Mn4w?t9%LTl4{dx2kg(y1@Z)e+>yK}xJt|4}-y<#!;WLaX? zkjucIpK*Oj$5#Zxx?zf$#|@;8b|HRm3u|7-(TzsT@XC)-Qp#1U3<_Ydd1!>Y zZ&vsqX^)eMbs{-5&3(usvi3ANIQdCjS}LZua}XC8FUnfm?LZs-pQN zfgOrBwLvB~f;)EGps`@P^|k^<(k+}2{S)uwjDX=6hVJP#!CEi;)5=xL@;x3K8bbkNK~13DW7_v^LeIV@SrV$x4O5zO*3ptkMYQ zVNa;s>n-N<_!e9*egsh^I#wN_(&H^#tQl)8FUmc)x~UBqc2)T- ziq0fDocBE7@>(W>(u`45FXr*DgX#LMr>X6n{JTecktPE6M`#`J4=J%?TJQJ%(+mUD z%h8%7^&4*)l^I>IyG(^zC**0g<1Nv#{xkl}fv01ueDm@r6{_t7X$mv8i!CpBxv&JZ z)+2x@)>+P0)`G&=6*DNopOJ`#08WKBhDKO_KMekQrL&dJ*4wvZv0!x#@E)yB9& zF0mQJI+#m;P1+{SS<_%q={9JVh}shpxJ zXkjPA_az1iz{G+(PQk<+#1wD7irVH3T1}Ql*Oz}G5@UE66xd+Cb26cC?vu)JtSHIi z#jPZ1I7f6M8>{>*6u)5QQhm^iU7LrAm&uvNSs{gMF|5h1h2;x)3XeD47eXYdLCPgC z+*0GRC|Ik%glORMp0Cw!(#;DQD(ZyL6cZ_?)NrH6-?Cb}|bH z8*gouRgfp#z|A2i78?$PVzjN>0E5{fu66IwW9V~0XnKdPD@Cq^`2DSK9oXD^Q=(fg z2Xd>R%r~gTbV78jZ(^8M)_WSom+(|$qF!Kl90$}69|xSD1YvEs3P$Ba_ z#JIRS_>LCJGKb+p!Li!R6J-csNOtw3OwVpjL1JoD+UKZ-{7phNDAM;|p}Y$nNNtuo zi&Yltck&wA3lL^K@RupD>SF{q0lKmygSdm+llS(?-B;Zl>7Lah)N|!ZP z!%nvH6rzudA&Qu7Ytj$E=X5xitY5TWD;w6P4?8i&&+PLIev`eQBalmDMEwpj$83SM zbKI@b-MDK}kYwcT=2U)DR)5Z|ELz>J_?BRWOlQp4Zm!jBAxeV5@{mElCIpE|kS>QHBUtBC8M+h5lA)MDpkgOu(pctzVL)j`Ob3HAH8 zf?BjY3iF*Ws*z0ulDG5#v$L|SrrsW)S!) z-5OtWr0rmErj=4tavBFY-1|OuO|SW3I`d_ulN24aXl%eY2ru(}3`z3!5HCZ>2%h4V zy{GuA?*-i-3{s-GK9TozJu7BOG;ph>Gna$h!;3sW1W|ACOS}7?vRr?;|OGB81oNVt-;{2IP~uO?I6e-943%6@`!{^aQ6e+X5tkAbUW> z#}z(QCD(PB5|^xwaZ%-Lq@sh7nVvQ|Rh_LzUp}R4SM=?=Gn|Macs9dM;iQL=Z2)n& z=HJv^Okr5#YZQ59a`^OiR+6Cw{AmBp3``2PS0{oTQVUINTIJ8JZOBd@kOSHw@UPlW zUu+HQ8egAtxFyJ51nux*kTO+&xjgjJg@DJy_1kq>YAFTnV_fJJPNV%}N1DSOQIK*e zG*=N}BG@vpmgA2~2Q=h@{rqpl@4Rn^NpgDkKNh;`2crU+6t-tC{ zb#Kgkf4H!Pz@l68bW1RtFeC{lNib}+s?RW;+&xdmTb9hC%k$nOEl8C*HIr6OL?HX^ z44jp?tb_5F4?s)5R2js`*|g7GS)WS}|CD#x&=^U%EH@`x&%j9jS)k`e2X!L`x&?v! zo{IXbikv0|qEPhsBX`FNwo+acWb9`%r0U=kwfh&g0$(S6dViHQ1h)Yh;Tzb})N4vN zu2=fb6cOH>kc&VY2t;`+l{>r^Y&R@{vh!-%8C-C-X6x!bCBYO3wo>O)9?VDxU5H3{ z;uCabwN{>f>gYWoFs1W=N$ys#6ns2BpfIw;fL9x-l=@+JGF;gUpK@)w1K#!%$L{(j zP_g_aY`*eQ(3dVuGsJG%XTYyT+_3D2y0}B0;0W>~d@B2v_6J~^cARm9funq0S-!G| zz16#Qj1#m3+)=}krJ|#E@g`w_DzdYpHl`udoZahae1rh1M)cLWv4hWO^gt21S$OGh z;w*WT-LwLoRaMh$(vzr5$Aj(Cba=~IXykI7j}RS^BD~IfeBfqkW||kA*I6mVJT7}# zo{jUkR|It>xXa(lq;rnw-X$G@qMaT%U^9;zQ6Eg}PChpNNwe(m$-`o=Nww?68_(0j zHP(yzHZP-L5ZoGmm?-$V9Vd={bI7Ove1k(Pk8$FC3t$Tcbn->`7Gyd@5syoJ3@NzX zR*}Qcho%n_BSkJ6NQG)7-+fTVUoO*qZ!3*Wwa!s8$mP_K`pU?$$l(<_QzY9kTgBbs zqacUmFA$p}FKZ_Stey{e%W~Tm&42Meh(o=v>`g2P*T?{?PjS*hxy3a)A==a-3e@7h zH{BIla+8L1J%Y2l%ntaJa&cIrT#}qpIWk@X>xv=R-PzG=76f7zh}~B;#yhA~0*zv1UYKz6r zQQS$rKG`I#4Cf`1Z8Xyg^M2p54Z^J?87X{`7$JEvwqwx~fUwv5;k)tuAr8;T%!PBzyFj^WfZY(D z5q_7Z7@L>E@q0a&ssrDh|EXFPq4RidzzI+t|p>VTe}4&pEOqf_~=K$*Yf$b zE2n1=b;07d*TUS+Cl6Q3RriGWWOs7YM)NRwZjO*7&8tTkU~K~QusMlKEEQ|9#{2RJ zoUx8^O74yWI1M1(szt9CyLjykvE9i9sUfXyG759Os(Y@rMJ3fSw9zs2|?!x2ga42MftD#nq z2r?6s~KpTN{;dtQVPhM@#dh1LRu~YUpu3Qf<;cJC7KAYiLI62q@S|4e%qPl zTDH6uL?-pu57^ZuCd4SWKX-aVF6!cMu~gbw)HS`6$e;AKXWHPdY!Hlwc!jr2$;lS* zKb^AFEGTbFFt@a-itDTP@Q%~*+5!FW;A(d3&cssxXRqSZVOQg9-$L4ygkVeBk~N@J zAL?LS5%)qaH9_DC{fmuc`oq~h_G;rD42Y{nCmd2zGXXO?Ba0Fhq;nc9e^T zX1I5sHF=5IsyRKV!_9ud($|HMc~Th@H3XhkWtMxk$>e2=tWwRjCn8yAXEtOAzud$< zdSur7iWZveBz2gAKK&DTfHA67n7h?lL3(vZygZxIFGZ<^7R1nSpCcM=a&QGwS&4$k z9c9ZEh_qU+clYw8_OK5HvkxXRgD(_P#fYxcsz-Q$!i)jMVxnsWFa4{8j9{i9JFh`QaE3dHcm{?C~AmI~V48 z1ur@Od!w1oyL`^ZELStc+W34Cr71^0)xmc&A)CfzDk7FMU-c;0`mNjaAQLf%bgcED zV7oct1FBFuyxE9q#n}9^G}=dAYl}#PV2I;|R5kt1A*U71#~-k}GRV-@Wf zN}zU1+n?Cjr-q+x>i5WMevW-p8E7m3|6(kNyqxG;!+-a&S4GtRr#-R00uLJsk-wXH zVx75}%f4$zbXU~e=owD0x4~3)D~PJc?IJeIYthNfNy+nY5}Elh6Zl68j?EFY&-&bk{QIxiR(0$Y@=)cZVt2 z#=^7|4S1=6cW;^PB33j5*0)`a*@KccRA;U8u2>W^Cone7tFG;`s&RLBZxLduKAZ{m za3+{)e0*I~cy$|mFlwg{Wyx3Y8q`Nr4-hBiL<(}bHypnq4yzAq&e|Z0ZLgBs#H(%)oGQkJCr#2#+ana+%y|z9p#}j|Ha1 zi+o{pSC@5<&S_kmw0UyXdWUGc>ydV8J)o!8EY({>bCB&|&PoD|wz^34hQM$Rh+9W8 zR!f#V6W?qoIf$TFSzAs+*6kz=k=BDCJ5GP}cyx;tNfDz0bf~NfyE&HUxSOK>5#I4I z)+=gszP@xg`Ue!svzp3vYP~cAI()YOg%8NkN*p-9nOncT87{5;@}QbB(=6#N!`4ok zwi?CCuRYUbLLx)kH;y|Om7`PZj=91x*8}I%%e1S;?bjEv;A!2dPQKU~&6@pR-^y$$ z9sqTngc66xc<`kIWo=9JV_6jX==V{0t+uUw(Qk0E0Zfl*?T@GKkraG=(DA!py~xsF zJND1Ap0p>Nvc%+6)Iw2~izn?aGT35MKF1r05jPR|%S#TvkgH*kdom+0pFnKu0s()N z-!rNE!KADaLML9tD*cilM}s)!fTE%CYl7WUbFxxvAk49%VI$a>_EXPpa$M<^aMSWH zv~>e50i%p#UAtB*AY#0-MA)C^+U}s_E#idV7vu*qUAg2RqYlCSD%5E@f1#gt#&NKh z;R|Vi#Pgz!)|wtjb32~#%c#@*Xc-lhrzr^oY)5CN;2XJS90Mo7ZJ1V~3txK7#KihM zCPr^GT@W&W7a`SR%dQo4_`!Y+yAiU9`oF1X=rTdQ3M%qA#Ei_l= z!-5~y20Uq0Vu&y?vDc+J_K^<>nFG02i!lJw`x;7?)(0N;P+X39N?5f>0>mw+F<)Iz zGA7nIElWh%M>%A*IBVf-(Vb9sk#~=8$Y(eirEzqCZ7q77);B6sw$j(GpQUNtRA&w0 zDGoRn;4L#hw5RO;++0?T(&J?~DXmeVZ3m_rar6@0BH?pskmqb~;tNV1Vp84+vnfivIm?g6xKp_n-L~@ zG%rXCP=fWkPtn(;o4{9s=6=)q2CM2JYLuP)!`PDRYph-MY@;y1Kc^!Hf2%Lvq%UTa$E7S0{?(IjI@;-!h4$N|#*0bJ(*q1;4&Xx6~1 zO>K}|ABwq?<4c6i`=W_Dlh)mjd{s94rzjR-S75WaFHOw; zg(n1w7W}Ts2j(I_cxZWSGMi`gea2{^&s9~h+QD6ATXax??tzA?h|`Azc`5@ko@z@n zqgO`BQU0Ch1cg!Af5fgFE1?PvgBYR51o`}*XQsW5UZW=ad?VegKol1Gb$ z>6H(S5KlNK(@&tIW)`&@ri=KgDClz15t*U_4~GyO$NA2DI_o4wDe=~L#kbRDKgwOX zgC4`^ak@wreMw>MKR$2Qwl~~CmHX@N7Jj#tLAAcQ-@}+nS-)f`f*dzf%DSKT*iO=t zfaX!-LAc!BNU0ukE(!_({z%697dq6CcBYC~^!|Lkb}u_l&Y9whWXV`_MOVl-04qg{ z@t7v>I|~oY72lTbj>~@~E9mxX*!Npc{-2O>GwAW=JT^RtDV*(mn4uOAAm!~2A1T>O z+3d_YZ&LhOG897C$WDRgaq2=<)tp^X?EZB$OoX|v8F+s^*p}JZix54WHr=QSU~l4f zU&M9gPyX5&V_Ji%N-y?pvl>X1?RRQ!tOvi9W5GNR~zta-KxO>+shzzA6`o z^02h*@l$n?4m*{xIYg?7rTH=#^B>?1e7VeL&(?{KXm{!^8tz?VUGkYIU+CE|vq+@nw=!)^MW-D4@r{CL|s8}TQu20UA*>s$p)*_aP%KSW%UKy#9G7xshcq4=O3z&Uajb20u zMeQzKib!hH6!S;@;c_9aGPXWgs0|9eQDPb4duf5WJKaoOmjr0*kjs4W45?|``i+eH z2WLkeI1nQOgvGeNAcQ9e<`b1^s=Rr*sq0kkTrBzSA?BUWce@NaY!)CgyJ_}(AZQ$% zTGs$ih7p9_;xVC+H?$o4WO{pfTxT|#05^{K>F?Y){IRvPlyk6(NsjcwTDIMqe4@3< zNwOTfv17>Bqi%Y$%=e4>Px(>a1A1_a#)l6`3jXfi9*SQ0hu2_je|@kOS<4d94872$Sr&)94OE;ec|wEoB}aX=55k_?0XZWU~SjcxH8bTUYEIGh3&wcm;XVZWwm$2Dd%*4_8yR_`Wbe z8y^Sz&s+Ob5%lv5xzcTcuCzml)?Ha6E=@rZv>kzweRzKS%KNSxZp0?q+s0tG`Co_B z&_(e5%#>^(ir@!5W9Z7A8*(-WmS4V7Z8ga(pt_Y zPDsyeJ#jR?@efM33|{WWM-3j}Xzg4uPN&SqbAjra)EF2*8d%pwT!@c{oX__}XTI|L z=53rXd?fCPPp;0eU&{o2n5L3#9AJe;g-9ZZEyGrqCro=H+DuiI;gRsSq(Xy^h@33q$xiNXF3;Meq=xhmOQ<~k+ zR<8u2W(7nX&Pc*@Mgh-C4xY68A^VJ#Bd|Bjx3<+g|v}N zSo&XKW)4S}>2WW+J=vY+Y_&;AB~TX1Pz(u_nDl}xzbu(9hZwhU^zvB_5DnQlohb^K z;I@r$T$QJp%s(I9I1A0<Drv4T>Oa8>CX?BGt>_#9rq@hf!tLlE3d1PUar|g z&CS~)-DhC*w7+6Yfj8*0W5rfj%c%!gz51NrZQN zfH(Q~dTSW1_vf&Ay!J57PTkgFROnmjljKtoM|{T9Mh3>O`@t@Y!5DH7!b(VAg2k+++=uq0949rbcVA)OB%$SqBS$)CaIS8yiPVeouS zl>?&6+|S|Mv^8875xMWt-bx8ss2Hs9akhS5v{X8xY3;>tKrMf>p=m(6J`-76#DrlJ zxGUx}fZtUW!x91J>r z&p1p7LmC!PD2rCwoXbvY5)w>Ca8C~7i11`;1)*nu@p0bgG|3%0)kj%tN3Dx!(~`mL zejJBlPxPX*!F!Gw_IMW!J!>N+T#1gIne;H!K4ZcR5@* z=lJgQG?PN7e9bB9&(nRv1!7mdZU$g|zUr&1vhw0xvv=G8IPy!Tbo&;8^`Cw=7%`DU zH&>xU)g{vc+ShB5h+6GB;|zH@)lu-CD^1iRfKJt7D$akJ_`g!>W{`gvcxb1hKF4KO zcf5&58FhE3k*G7Dn~&Ac@ontM`2?Aq5woUGD;`p*3p zJ`ws;i3azzSSJJp7y0zTU&IlEcWA!W zc$a`KstO8P|SPZA6A@xmrMD-uO0W3SOC+fBmI@?D4=QfZ;UUTL&1fXxZK9N{}ycsGw!uR z0!GB~|BitF-qlx20e1g?(%Z>D80j2e?E3fUYySs*g&v0d*7ALe|5!_L;=M7P zzqu^`b=IH%IO|`evS+`y7Uu^n>X+{?7a{M`MR$I0w+!gk(Bp15dA@eY(p!Ao-#dd` zhPONZFi%EiHpzw8zjGr#iOmumf&8}AnAnL~pr^@St44@2(0+#m_a0aP=TX2A0Q#-F zgv*==+`R#lI^5^KH(pl$9?)X=`R{I{5vWPjXRi*B{mpIb0Ni0Aw}Sl}QUl=Teg;4) zIVQ77V*K9tF?_L0C3y_UzZvFl&*CjRK*_`bYZ}$x8y}Sb$R!=s2EX4E9gcDUYwMX} zMt}aj@io9fjmNZ8{2t52$pe(!;L;6#`g`MZ89p7QCpNqy)zvST z9LIOvFa#hkF`-w_JMh{ekFlY@qh90F0Aglvk*t7}snH+*){5}}#wW(piUIq>=(#rb zykEce)wals@7K)zLXz4a6-diSx)mQMCdDy8D0~}VqZQ5iPvIffNGs-(rDuW~?fV}* z?fV80gxjjD)>NZSF5v|kc(dh0hn~5&UQMsH zK|!6xLPJ)O+4AIAVRhGbvRNiKmG@bt_J?tmGM!iBu0er^$*DF?a;iLyRJ2DWLx`P7 z;f=#mH5;4pA|phU&B+GPdB}EiW~<4P>lz#U z@UGE1c{a=NA$)jZ!FE24K&mg*>D!24b7b<>nKk&PQYHn+Yh*|-aQyY+hRjCl4$6WZ z5mWQJqE%$X^Q6yHEzbQLzk|X_Y?)nc_FBSIvqOphe)pM`0XcaZNA^zpGU3Bm%R5vi z`&V{ToeRP{`4j4PBt9jiJjS_TQNtUO@w%b%B4oEf+V%EY{xndCQX-(N%$I6+3;VSf zQjlhn;p&)By*GI(308uQuYT@zhfPjzCi^uxLu?tBokP(>-L!=_f1Ko_*Qefp7}OjE za7dp56f90D&gPEGVEA$X`-WmVNAu`sLNgC{EaMW1=ZNz$&=NF-sO*AA6nO*wbv8+C zmi&abkkXT5AT;J%FX0H#yJD1QF)tnLGWqj1f6rIn(aqsv%{J*nAQst1d6!1mc#?_B z_h#*Zi_NV}Vo>F_@ALYCnK-R>n zV*UcyCWMspM2CB=FInkZWMK20Wl1u!$%zYz&`%Q_G$&bs-659m2K{xJl2h#(7g_Z# zoo8M&EPV{UOQCs)NS-QlT>pSym353Th8-hcFPooY)S@F zX*-m}=avc&w45j3RbM~o7;N~ zNMKf1ub;ivU=g&HQp7y`57V*~4N$ah+3({6>}XmF?R>yn`f3ZiQM1+gCTPR6xK=M^ z{lxFCMx-|bW+Ueb^Gl4u8RP8pVm}_@2wM)f9dYsfaze08r&73>M{&0l`i(sm%o+LO zeYZY7GWcULf-}@d`y+eny5~5uP$EaEf89flL^8n61|)hK5~NyT3bZ=e*$i7EHq9ki z0X+IxPM1XI2oo`D2_G*74Wve2N&+Ni5*2JmQo?84jvKMf;3hm$9#|(0)e? zjAa{;6KCiF{in5k47Lz&mDtE@`e&xvK8bIge@G$2mRLy3_RHQ`flil)cxzdle7OEV zQcnFz-)vLAIHxD-lB=sy3R^NR0k@`FX9BV*HM7{9mHzy7^1Ch$IWV9C9 zoE}Ha+uE0WN49Qn)lP@=X=?lG+?vq$%Tkw{c1&4vD}FwfZmwK#MKV|KFp?p3D>IAN zJks5KNBa#)Cy!I-Ca3Nyn{xmOV76Jm?1OV-u>o(Wq#K;lwIN+#*$rqPw|rA3bS@2Q zwXQu+*sM430heeYy$(J&ED4w;#VQL;Ew6G1e$USI=Jh!6inmA!pBfW#JRZ#g|gn6`O=QdQ_s89?koyc-*EeI1@F zOFxa8hO~c5E?$oRw1z%N0Y=p@0kW9g(<*g90vj|nDs;ENJy^s&+I5JF_iv8=0hTm;%zEQ5@iW#tb zSA~PQ%tQ`}Yh2F;bH{KvNPRp%fGn(Oi0&a=Ue&#VFM~!YV9_fx-qG8_o}DBR-Ge?- z7snnNRG6JQhsWv3Kt-}udFu}$5W)>2-{U8AeI#7@Ac?%?CENrIY^bp+Th3bNx%)n+ z3`yHi1LKHg)1d6zQgw!MqvF2Q24&#iMPlg?Z^Hl>8c2lEG%X(a{^8s@EBHkdGy9_zz?1TQ*>xZ9PXVZaQKMvzdo*o{HYJ zr_Tx@qx%Hv;|FZym0n9`ZaPoA?*2G~VP_asWKSc=P2~K0Q`@JA5e}&I9BU zuV~;7yKPpcZOQpuZD^XpT@XOu6dm8Q-iaP^c7IrV-sUYTC8t>?B&3=IWZJOPq8Zd4 zAr7grtkMp3v3+HVaIqP5ZLZ>`%CAA~(yTd3$B|}vfn}4ooPC~QYhsY-1xmJ(##RdJ zXRBA_N`zDG?#p1ejgf!56Y-sonQF|_m;M~AaTKc84&iqrux;(oCF&IN3^l2#+Jz^UY#~P3CTvKbgU_ns0p~~)a~97A`rZl zn}w;$ux`cTtmu|Rz+vvW9&jsL&5i}f(^}I3Hr~jgdy_cGgJZ#Z#g>8DJ$^h27JS}y z#U$yCR(6~=nB(XWgsdjJYs1>Bgpi_M&&~Hg97w(=Vt-i7jh~Ke_64%;FlNX1i(i)E z_L6SqECdw@+t9329Nkh_se5cB*1&rO2GAMvVs%S>Ulungw^Hpr##ZuRqG$4zO zVlC|PwLvAzjA)$$B6Cr&P9^8s6wy?>$Xy)*{j${_U@-fxZZCxq#OLz*?sAPbCtj^- z88GCmFf&#EO=H%9i7bWoh}p)?oO68J62R>swaecbk~oq z`tR${H4~nDI9r=_>w{=@@fS6wML5b^O_N4*{R>AMcYzQjl%>Rn71>dyXIU$MIy0HO zWC!THh8F38D)^qLv-IU3MVsDrSIqUms{>ObFP{*~aONd=2Zr3j-1o|?cBje-JtlmY zdNd6@I)txlxQ8J(o#JM31J*!%f1y}azo1Ncy^i1H=QNWJUh{Yah?#^*12DREd8wAT zb!{6PEj}<|Zp^vQxrJR7xUU4~C_C};E%MLRST*a0m2A}4G`Bl2U2r{e=_;3mZ~jAy>t8S8&d|gc_`?-t|Gm1 z12Um}>EX~{AvtDn;4-T_Z&O|Ct|NVwP1wn`GFkS2xcU}3lL93I)-qh=+4tG(i?kiVOLRPS=gTQja z1pe{^guQRmc2A$ijR;V44U4-boCKwzZZH45*UL6M3YmZKc{WDa>x?bWuG&)I!M^;Y zOsKu8ccq!vuBpzW;Huo3_JzY>@1`V4>5}*FhC)d`3pDxjb;NsONU-B?>J#J(B#T@5 z#aH$1TirUN9+oxkHd+nqrDb-GXw!T)T~>kmo@aM)!+h1tIekIXhr$-zHXW>ENGcAEHh4RO#nxgZf!n<`pc`IXC=3ZtC{u*R5JsiZ zS-&^;x=pf&Ep2bC^Jb{Nr2(UpD-LLb-!E&WUE6pPamY@9jGTR_K^c?wf0}f8hE!j< z2%$djE(Le~?&B9HycGU$xXkH|?HxfI87_BU%f-Qd-?}`$Y&(1pjbP3i=Bt=IFj;Lh zoBM{j*@)c ze>(?`|0AHPr6k!={28|!`XKcQF?h&-1B@T6j!x)$W}G*Q7ipT^ zliiGbJ9A_Pd*SzN<3*^js@g;c_q$({Z;-car5mk)U4 zufB2RyW0lFOlgjAkjP{F&9fI&)Xjp&-pL-_kn(C26sU)GMf&{HfBS%pWRtIH z4|`U3iogVT^51tdYsZ_Y2M;EX9~ZCHmKfW@ZXNwQs?HGB)N_igxG262Vjt@#(C=b@ zCI7c(FE?OphkBH)np=YXhnkIoQ zH_9iWOA_R4A{gklPn&pg_{7Ln2RO(sx(p0SNnDG>X}@1YH7s=LT4iFX_3~GElzvaq zKJX=2&PKPtdGdizUIPB`_WY0%*t@_!ndvv~{6S=@PZQ-EwOQMlitt24| z8A`hWsYBH(znAt3`qMHF+EyYat84mqzJY#ltO0e7rS-)U;*Rq6I1h0zT{z+kRj*b) zh?5FZZNfK=r4z}-x2{zE?IcS>&(61>vp%0Bow6m%HE%8p`waYmrTTJ;N%>dLqeT%R zfnM}0JuWaj&uXQ8RI+40cou~4e-uOG4i{a#f2Ss5qD3MxWCP8wL0_wc^6E``3?#7M z$GLST58@lX6D0~t=c}uFUna}z9Ly-o&hAnWgF^$fTA2?zow+z80#n$=oMJ4g+VB>F?AuMn`&Q z=$a)-AFNQ{ar>2=oE#w&OQFg;c*^tG&fmPo#Vg2KI+L7`qN-3MJAAG2_LXyoKJG^c<}-jm_ClNw=41Jl z`MLfFq8@lZ9pcD|(kPSyx`tStyWLZT=U=uRLDwKW`C7NlV#pD=#Rv8}1e3q(ggC9c zso#&l5ao_PH<7}B7uVco^>9vd8FZd>4|CD28u0s& zM<>aNd;1_n;QQ=6!8Q8Vsg+Us(R_s7Y{HKztQ5(2g^@jZKc-xG_)1psSH+Gw%R(ju zeSMF}*O7dy#!K+_%ooKTU=i-Qq{8eopDQzlOgICD)ES$nn-Ai_<#nz-+b!=;AK;c_ zkMWF++BuQPbq$v%*_sb`;=z~aUb?QAmvTn`1rQ1&nzhlCO{XH((4Pz_3uk6u8ZayM zE_&|bEBMc}jQN~1=XU1GfZDfPSu>n&`!RFpO~#R9Wx#d;Er~gL)ZTLsnu!kOM?%AR zrvXxr5*yY$14Yobf~x8fA`~MvAAC>FvM&hwmdbxD=KP*$9kAa1Nz^GHES?#Wjkd;WOSxm00ka*3(`UQuE4L z{bMWzA;y5pqK=%+pC}Wmap_of)(x*pu5->+W*3AWjy_N5(@*RFSD&>P+)Jhj)K$TQ z32CdN?d2h&uBwL{`yt=n3ZXlh&xUVFC`lJBJ!#vRdS{gujM7T0<5J-7Xa5u(qinl)qkg7`dcw=VV>Q-zSRX~ zmY3twRCzyD2m9vyV&|NvW`gMcvR}7PbE%a8;-gM+AAY2zYuCg3DId?N%xK<7S{HRy zVRRYxej?6B8=-vMMk<>K%6EMJjAZ{skb-Vc`$@U&hTHnq6xg!J5R>09Z$(Y;mPPcB z&Mz%fcJ6~wTqyk=(B`tBjRX*x+gxWs*prjwy{Y7erdUbjME7dZwa+w`Hpur|U{xP} zW6Wnr^ett1qMam-FOb1DzqvU4;qeR}iprKRnD8yJL!!E4ga9J|`8|I*4;U2-5q4L?H zd%`^5rik>YE=z21(DrZt zP7!y#@p-s&S?#x#OT@?qYj;xz`Tb>KjJlQ2Rv*zTdfDzSdeT+V>Afwr+)~-I_4UAq zm5;}x38c4ES&g%ad$GL#38u}ch3%{@E`zq&D>+E)5F#GxOYq&ka@A+FrZlZxdeZn| zPP5Zw@m=rtxu~$i>9!=UJ)5wMnmto;Un%;Xsg8-x*wyqOg5La7l=V5H@v%w)2xTC@ zwk^JiFg!Lg8IDK!NQ5e#9?tL(P`YHSf8BFqp9MTG3g=_Am@1=55xQ{4gP2Yqq{~5y8|1@#0*2Q_1aEI%%Zj$bBwwF@Rbi^Zlyigz$!)g z(#VxACuEnYRneTBTUI0Z!`T(R6Eiv*Pt08<_wDR6f+yPW*&&_86SiK2?<_41a`af; zbruc(+7&ecjKF6U+Ir;;)6cE;`$vEwV1EtSvA+(rW)?)W*p;@{~f5+3?Uun zO_%3~9iNHV`c?D_F*$bQ_AxAWtzNz#?LSE5{C3iGeMEzL!sq;9<(KJwZPw94aNezQ z8XnJnOx$vyh&ZD4CXzqKP6QTOAGTF@*MD;+XJAEyY%*yMN7wJfT+H#?U0-A85)~YC z4BO3n4Vhwok7jia6rAY7ybRA-Sj9}BOa5 zy0n+SKE|`Owu(X5-)r&oZ4?~4COW=JC{iR84SWrd#<7crJ=Py{ir|wX&Q1O#u96i< z)@gi5oeVpb^6%W-CuWKtI|eQN`~9GQN%5$LI6QlYsUEKJ>1vyt?Zi2%u{}3OtEhIs z;|}cpOegYOHDTr7*I5Lr-9_|U(i_uJm#D)jcJcE==+U8Zv-;&(+pIuD=-(&=wun#y71$B|9#8)e{`n>`&@k$#(S3_@SGVya-R>bPV1da*ytg zZG9+nr80cBq=<9NM;cy)2!+4Wf?s~vJn@!O-ob&~1#odboU72uO$#^%Jzks#pG!Xx zW(B>yiC;%*i++uOxK-NfZo!@qq`BX&pu<<~^8(gbg}Xz6^HvdG2P=1X{rgRHZG-m> zia>Q*wrgzfF{OxBl1!jwkOR}+HN z(TJ}mB?sB9Zp7NamV0JJ0{e{&0kZM@18SW~C)_^y=TAQI7!yM>tHEVxyvs}$&gTBR zhCZhgSYt3{KR^Zy%gi;Y`fJNDq*x86ep#!j3ZPuAsn$+6A69Ly7g#xPF6(boeGwrU z;Tb8*T8DjUdd;@YTtg$q4dIW@JV+BA-VVc`d$4^iEk^nQ{@R25$ht|= zE|$;Ne$#6F;*3w^@WGZQYw|9vm&KzkbMjtK{)Qo~=}R+p>on5w>){9EZ8EEh!Y!fO z{LRc_o+qmTL+vd4Z_+Gi;0KMly6S6Gmk4JKvm2MWzPOd7=dew6B*`-w{ z6T!UGPuQf7eK^wfk4v7P;DxZ^#c9=ixUBQw*&il52avDY^Uaz8H@FZNW(aTbF792a z>o+Q%MIM*g5^|20-2!hrt_Mu|k2VSV{E66)A`m&h=%%Y3@=oq|E7O+71%3Y0ThesG z_9^y$6{(|@065s$y0KXVf2IB_%d(?y%CA0`qt&Xkm9-Q;# z_|+i)-j1{_z>foIDN2=Bk$D9k743VL+Zhzy!+)nIKKu%jH7$9E)Z6aM`v##oyC4`z!Vd ze2Rf_r)NU+*(GT9Y)l>zv1~T3x^TAshM~XZu?vZh5Km(JBB4(VKfEUxNj8}I;s5SU zU=K&PREFE{4ItDjF|oPzvCI_@&Se!lZ9xO7zxmCLjrWLIxt44DV&tm*Wma6R^1<%G zV9`DQ{`;H}I{VjJCIY$R8BVGOqwdkZO{osk165Z~Mzv|~)tT9gTy>727S*ReTbvU| zo1mL~pnf{)f;Wun+WqIGYTfP{`|m$sIhUrk`V5{Vy-*4nEB?0ev~k|}^`n%KRf0Y& zW^Ov-T`}eD&b>wrjvn5bt<%`vkeCQ+p)`{AZ5>^=w?C@!i3322e;-$9I^4|XNc7+S zwr_ZbKJzFqK1L9ZNjYBOi(}%RlNY@0)p(`b_>P>*-E7mD+uOjL3yYa_Q^=)yx@|sJ z662lckC`|f7i1a^YT;D;+sLfyJdVeE{olY^YrF7(9|!EB)<3l;tz;>1%%+^r-Pg0<=)W_9 zYfAcqLX7*h69-nM3?FKmyzrY4B%g5UTnl=C7GB3S5@jQw6+5@yp1~RU>))&>;abCR z`vDTfBbD{z?6%QdZYhU!R(T;w$A9S!Ek7K%yV2d2S=!IGR6PxtpJt$mJGWqyvr$wu zG2Nti_&%Rby;PS`a9&yb7WTRqwJjZ%74(3ayZe<+%S`g3_`9cf-?Tz=@}OIfOaAAA z{;P23KlioEN^t$9(E;Xe3d7#?zHeIDz!{v?)9>)iaO;TlF5aDUanDO>5to)GjlB5~ zK|qWpi%kgOQgq{Tu>P|8$~RgW`GXbspWoi|qD0G^gb_aj@I>ywTeuJQF@hdWh+5%& z`H50n6J6coj8*mPJ#)EncsHMi z^b!+V_OjJh&?GA)Uju(@l~o%DsaLG5QO>VDKh@=Ylw|K|H{g#JP^Y{OQ)a}ziZPba8Jpb?%AzZ=LUP`blUZ#9$n zs^R)p#^RhwV(r+6{@rZ#QZ+C0KAVH8G#^v^*!Ij~V|S6O_FP|A5>CT_5bkQ2Ejk@c z#&}={$9L+Ed*2WQbpWido${s)C>T(HtXh(#I&K>_7Tq{k6I|8vA>2oC;;51r*khu@ znGFPXtGetnkfoG&hRA&RQek#oRYnb>s%z`bjFqd_p=wEj=K8cb#JF}gU-A0(RHvZm zS*K`SQ1D9c%ic9?&D+5A5P75BVUz-5ZcIXF5m8`81?oTrXetSXM4HId zvK_$?={IRA*tca2+i1_ww=MP1a20HKS-uAWRqMgA0rN$(PtN66RRrtTJ`O4U^J>S9 z3p+dIUB;Zj$_q&Rd{IAqOKkbr&*Gla)r_IP#lfj)eA&)>xX}tET%&6s3B}bXN!{Kr zk5tHsS@A)55{bEv6eGaiKDql_-Q=0~y_A0IKX1#D9-Tzea{p+Lvf`QCH&fRdMsLRx zQZ$!!C?gXxWz1EKCT!WP#Ft;?fC&7QFKwzR!8shhm&CQ`O-5K4@%^z0OY6>|tKQpQ zCzKQcCRkvfFPSKw^ZsBy!!gsI00z@*(rWf{4fJhq>i}eMRf*Q+dk^ZQrxePXHwtgB zPtSMKeu(1VpUZ7VmKtrYrDz*&u1Q5N(ImZ_ta4t8nPZBR`c<}kaXWS$DmysD^>{dD zA}`ZKGiPJ}rhurgGo&QIwx!uQf>6vemq0y0`&zx%Yer8iP5`OBjDQlz9Rfn2^TQutcp3q}KoH=ztWyX9q`aAs_d9vp zw=c7_$F50{XiNd!wvSDdP*2dK5y)!_>JGba$4ghq>S)sWN$ zubrEKGDYb zi~&KHVrLwdHx%F@F@F_eK^lyI6rGbRowOelAj1&0yvO!#o!Re^E|1&61pUO!@ zfhgn4>V_tgj;HOOTOYJ@kPOrJJ#3y;kkGurN5_Y+u(?F#9r5Q%h-L2rhIXN?{yYPk ztI=fWwBV~~+{1Ox>Ou6{x4u5aPBqA!C$M(+#mt=l4=46)S*9azSir)3_>R}_zMV#M z!8mjE(x;L+MeY1J{v>+|m%j{;m#>uIQ6)dQ@kUB2Mj8F5BPhAPI62}6k8Rr;b!t%@4PRj`nvwmNp^F=Ql5C)|l}Q_f znV3rb{g^L%_njDx@;-v*7vSp^0M(Q|>HxRFyf9)WvDXFO;aQ$&KP{ee{dId7H_<%T z<7wZ^oykAeCirVtUwhthh+JMlXOH8eZa>TubCSBEx&Axf_?eIj<5Yf#i=u(c{TKh< zKaysU)_XNSggW^RmHp*6eU{d<(J7YsN2$(nXfyo#0rTT~%kOwssJeQvmv4|yU{3#i zJ|nLbL6twJ@)hF;v&z>I9m##%xj_Ujee$)~-b*_PFTV7t@tby!qV}o~ zhBSRo;Oz>#kHliDrswWE$CdFq0R(U`Z8p&4Dbelkd! zlU$5Gws9A==KeZgQoXB~;@-q`aP2qI!5W-hZWi-rw74p2SW74V8kU?tm4B5Ae>$dz zq5~?c9+9x0o4-N~oc5SHhmB*14Ovq&K0W8Dka=s>k`e3%#o2lbn@oX6)qkVz9`r|R zE6-TDI7sgKgczI75HNvX6-nWq=xj}M>FEd`hJlTGnKhqw69=YcCTyI*G|};QiNA5+ z>59)<@D*690)~fu--}x2*SBneQHc_NE>^Ac@61q{R?CxKh@76wEfg2(8?A1I)a7P* z5)zG8Z%-NMr;aIQ8j&+RKfhDeGlEo8Gn2z~kV_8-#WF1Gw;u0C{g+<%j^5>WSK(Gl z{kA=4aWwPIXB#Z}pX-k|`@(~7zU(~BbG_!2ekr%J56(YfJfDJoWzV>)ST51}@G6|j z;_2-~LS1pY3%2=8CqNp(1X$}^v10zOIaOqaVvz+aQpm80F~$B#3IOVs-o5%yyWrtm z!@OMY?0x77xF(aLI!cueiksjAOJVNoxg+CEY1ic5x z6kmA#`;-7G=sf^aOSl=Fd4Ds-dmhXA<6v-E7S%f7;fUHx`^6sbr&&$Y9r)@(zatF7 z>Q!Mylmp(4`;z8My9Q3(>U)_UG4o3(*{ErBsKiXx5B3DJreFF5GWO|}ZHwTu`*K(N zOfbissW>{W!Zu9^1!n^5+ZFBR?=eQ0_C;_rjbHN-lA{-CWq`)kil+wb|Bx4_oH^?+ zOwDkB_ck83#4ZftwIQiK3YUGF|Clkj$uR%!yje24c-62}K+$kt*70|q+eMS-+oh4p zrVCR{v4Gbf_8t54-pXmAAbw zLGFh)>>Gw|4bW5m#4pd^OV(^@BJf^NoqT>wYRuP5kDmiOlZHAN-_z{b%Vf^V7CJ?S zEdZ*e)3*E#wV-tDdKCwWLa4blS0--p)0Yu5CEo;cI{PR%r8VPM?Pbdy-Y-H*K1-Vy z8!I&$+a3-qo9t$oHe0~7G8bhWn!-U_Yjex8xU!^q_5XDQQQG zW=hR)QU8D*kjXH0h5oml>5kr|GyKi@8*)_HCK&WAPAiTVei;fnDst1WuHLZys1=Ze=1o>E4cry9uqxZm~$o}u=|qL7!ZMzrq!3qjP| z(Ge7{bqbi{U0$)prHS*}9@VE@ix5KN-n$Wgmmg|JQEy_Dc0u^v;H9^GxSHsL?=4q| z-yPJ}c!fTY5`~6U&DO%ltmn-bavnmnU3r%+23k7(vJ3-I-UadiF2MJlKM5@VTaIli ziX9Ye@}jgWE5S0b#0z*$U;pdKmEG$*WL#=eGtiekAk(b9zYcn;-s-a{0X3Ya!t@uu zT7*wk_HUe07?-qjJ5g?V)bnmjseG@(U)O@#TkA6UWddn3bo3l1xFJfQ**++Qb4z1Y zhGuyq8^k3go<1RQk56dU(ojHZAGl)L(Y`1*3*t9}`!g?X9l~}k1_3(3Ug2}AdmT*X zk-B(;Z%Qfc?QAEuX>BjN7f^54m1X4=@OQy0^J*>9KZ=0%M<%?%zTVaC!771v$qwM6 zxbZc{$1kY9d|A%9)+$h7Hl5=9tNfvuYkFyN(ar4Yu{NW30~=E6YAIUf~bj> z3$BV;N<$U20Do0YCp@zP6qo%zAFYVjWZFHk3eymE@27tbMEFsuyLg~t&p-}qGIGJT zvCQ&#ru9&mybRRvVH0COwztVl%#UxXb>p}H!-*qBVcW}VYW(Ah>zfhp z@+|a(oO@mO|1kCQPB?1pKT&6%B*&~?gBR0*iNUzq-3)I%A7vcS}i3s6DCOv*|nG{6H$iw|;)X=Z?kF z#7q3Vtih zm3DakFXf3BHr|bWpGW=_DCw_J8_3$KyM<_C+}+DyoBTAg%_}r5)gr=O!Colez>-(Y zrFs}0bw1N8I3;i$z#483NZ4=b?YCgV2!HLm5yB~-wNMLuPR9<)VOOy(`v?vKi1;X_zi z6OV%U*cItk?+Of!YKab9+U9nvskQW8xp0IdO(E-X3DL+oWp?>GrFs2F!TS9QKjgjC z^&mJTCDdafaPu$@nvr;Q;T3Kpu#SY^8`Dn!&*(0@)^aVeQV`fF| zMG+|v^N72q9jocaoYPIzmQlI;)*%aS@A%` z(@1S^8jj`~Te0vW0_a0wDVDp`|18vU_#a z6IIQ`rnOG`=>f4F4f1aNEAb6(LUdWOMny&-21G_jPgc7(msM%3{M?+@P}#T9}x1aEJI{a`wU^E`YzI0li7?Luxoh%hVamHtzxJ7?J^S7Q9r za`Z#>X_+MVWxjes{@j6FZV??<{zM3BGpD^RH&Io`x%z0v{6Nzz{M@}5=FQmAsgS|( zaTJp?YjvQn-H`fh%burGswAPfN7@(cF(Nq^>eM`rHmut3&{Dl7>VzRd=}m^I1Hqm4 z^z+y30f!7PqiJhXw*rn*cA&3a=;EWz=-5~-NYRkXL8$b@&hDPs9aFZ*?HP>W2Xl#qItPa zd}pPsZC<8F)y^^vGzFUk)gc+c=UjqUSSwmN2aMDc!9q$9f8wWKKGFO09;wi02FNh_ zii(}sfo`mrne_hE2+JI>|5N(|3Xo-1i_dlw7U!0=qmJJwi6kHF5qx#UR~P4uZ~Ug0 z>qNj<4BOZH%$d#}lA8NeJ#-tJ4+{|2ISgeYQX@n33By;7b19YtWtrd1g|@fMtzfTe zqtlp{ev;OV-`*C5Iy?H!g82>4!8B-)>!V`7l<&i~`6eLhH=i^5!o|;B%gr^xP2?%L z82Zy;<+(+NOpUvA7_y}9f8KYN%2u%bL_;Tyjx=xz4?kQQ6s3eO%JC!@e97f4UweYuSVNV)K0fx!=E);!5J7y}>84uIIh7EU=#i)mTa75O<%wy{e#>=eX?QYUX5g>lR+E*`)d6;pbM&vPk^w2nwtc=opiwwXgGQ~a|G4@Z=RT~6Bg`v|bM%v!*~8790- z;2#@;-%{8>j?mrV*WRAg#oS9RG9{-a($1zC&VM>+%o7t(9qx9N+n2dw9AK3-uPP@7 zJl$9>&O83I{$t)->WS+&bh6Cn>w#k}8(sG^al(k|+UlM9f*6G?+xra277)l7jT*T( zZ%qQ2>qDKdZ{{Dpqjo(4)zG$9H@#nsGm`xtez<3j$PX&cY1$ppxIK7QuC!m>amL`g zMF@Ch{zrMACf4L#AM_+?_M+2${P2%v5V>(0t$gJ1T_3fk0~?rhz0N|^;0sEV#yo)< zvHyNJ5h6@qTDheqn=}eY1WsJ%*7*A1ljaA8RP$c<&N26%;JIUv@hIPDJ_;myPHU|r z;=};RRR{UrM}BBYu_U>ue2?qG^6aeC^zQnwYY7@aJ+bSR3FS?~Fpt?*Z2*s$bU$i7 zW;_~}oVsbTgCpgW?@PV=&^gmL}h%xN8M zI~aUpzV&(QQzh}%Z5wOgn89vzQ8fzgZQHBpGFxH~z3IxPsy!D+gK>oa+@;N=FsnycWY@z1zN>{UP^DdHz1EA-@%(z09h796MlNYxW|juvSA zAGY{^lTja%70!shERo)Qz3v^c{^v^7=UK&8dgbdAJM4Ah(nJ505G9AX9Nd*3^U`F| zA`6obrQ?K8jwYL-^0tpbc6o7sl_pUoD|L)$P~OUJ|B*4J$P>Fl|8~>g`WC2)pPsFY zrn>Oq=8ca@zmWCsjXwg&M$Dvj{$jk3x95#Zvy{(>CPktpPOiG0aMv=tn-tYlo((N8 zC%3R~yh%4Zyq*T)U?U89tR632&+}06A7jdQ6$#8LOGl82-B!_V{3Ln9Su4S4nvB7} zIN)Txrb{}uLP81-U&RJNjy>%U!qHn|Tds1^JdBAM5BaxCsa+BBV!!Nj)=?Iw zl^P`=>e$Unu15>`rwdC=KBRwX>8uNir6wZ%!rwv24k7NV;g{*WanSLz$rYodoR%xX zv3|Qxh(()Xz!y(FeuRVF!HY69|L%UpqtLPt_kxc^HuaOuO5MInT;l_0!M5jH>4s&E z=l^)-yeVcB=Wb{`Ek>JLpX^kG-gxf=KrI}t=I~fb+bB4k{2ejX^o2iXS-3dK3b$nP z+C?p=dUWEp>q?eJ-A(gk5JAFqfM|BgW7Knb(nHzXZqmH(oH&QwM9oDZ9;5H;_*5)# zg8<@TR-Y(M=ZEmGZ?K^IuYl>YnJc+YummwOX6Tz5ZDlA`T21D z_?N{;p*2mpB|pB`Rno>1GFxPvwPevfd+K+0Z1e;(ATuhuslA^m@Q`7@0q4STeAAgX z`ZiiZE`K1ZrkX=Vi4!Z>vgALK6QdT*cQ02SzeTyuMlTZqVNa1)S34a`#pg-3 z?M`~9NMQ$gpR77N6!)1I(fD^i3|R7MR?(<9F5WuW1gi+nDH`^Nqc8D(FYtF=Ycr}< z8H@ddhoTvOpCk8Vt+Q|F2A{znU)IKz&ICo5(Z<>MYk1D{;|wQwdeh6sXAA9f%yX;5 z?+@HR$z@Y=W3W$iqU2)FP1`Xihye@_9cH?uB~?miIQiI-L6clbeqa8t z=pMttfXBf!d^DqLigJy&O^~$lZht(iH(U9ZJZ|f)hVNsL3&4>ISA*67U9fZ~EI#bnt*LDJD&9yb3_AzTI}!MDCDGH2mkTT<|@ zr&RLC;1TQ2M#f~qv=>C&lFf_%el%+-Y3Npp3s&4~MR&ZIJU;a)vsLz=7lsE0e>4ed@i7 z``dtbrgzNZ`NE}eH2tUP=r|-yBD|J*_*g4ke;VMJQ_m)T3A_SOv02keehz^!q&4_mt@#~UrbD@N>BRR0zvStW|R>_`yx#j87v+Xc8{%p|?9s>d*h?z4187S@dFoW*r26$_8 zkxD^Xx9oe|S|bRK8xX)>rV83p|Ap9QIEGtdHW`4YQOWXV0QK$U-_Qnxa0z`N+fJnK zc5_)h^&l2OThFYYig+sczNznp_UccuwMqN5vTk|o@1|0jNF;PGGj#S>QSUpad^_7b zT00WH=~T71T#C&D`s2O2Se>eZHY_JHxRjO|EazjBAK*~dzghKqu8+YTJ6JVZ-@(1m zVCzpF?Y;}r6A!;@VntoSZXN>m6qOopF)Le;chz)vv}?>f^IvbK`I-y2&gw-)!1@Id zrJGP|_$7ObkbRJmuGIYG)rHMf9bs=b?fpf*j3V%M`R+!iUSF?+C?WHG8~gc{efJZ2 zfjIWlB5Qpdqt;_UT5JW`i0b`Ut4aeBmg~t{07t7BIO0XuMDS?tEnoWUHYC#5n++gD z&mdfFLf?bY+slc<%6_M*8kG1<4mTF*GpTSy~!04KHefz>1^KW*+ zEA~~{mAm@hbK_(OuT}KPwd;EpqhATZ+=HHF^2tSgcBA7f?upC8#^K_t zd-v44!}jv&S<5SB;w$!e0@hGfG2%c#jz=7}C^p&=+Svuda_T;jv{@VINmEAXmeOrQ z^A#{tVJi(GAdNZO(68j;8+BW=w47!nd~dub%u*aqLS=h?FfR|F2>*%H|mF-*qVYfh3 z+*pMBgaO{j#e~(Q_EM^~zH*o{eXQr5F6*e&4DKx)1Kl1Tl_PBv?!itC2ZQ6o^0iC$ z5`ol(n9$Y!YpyG(+S9GS^;C>~n`yaAV8#X-_tmisy#(KKa>eg>dmlo)^GWx?5_O?L z6Q>0n{)Z^eJqt#wjA`I{cEWwKLo(tqeMjX>HtYruaAr*dpNDf7@H&S|!ajAc1Qpg5 z0NbG%#1#v-vc5HvYoax=zuC=uh1*QL0sGUm2PFKV(CgakgKsN~f+Pt_tZVnEmp^lt z++M3G3Nkl|t6F}q<&7MF=eZFx5`$%jXGYmp=--tHzGa>fpzzciUQ~UgHnCq3 z=aP{e*8FY2^!)tO$2b3*i2s}BCH)^g-NIaYw(BYXg3Ow~AwE|aMyg-aH(+}3(8c%e z3q4b?*V4|yAu0JlC`We=xFZyNCct(2fzG7zVwtX2CQ!*=id98^5c(&Po7+`b`loA= zpE3j7YEUH1K7IS(>%(d@;W`)4yjuExr?-u2Pgox?|6-(yp^G7TB&%}EHX`m6h+S^Igf28w@4 zK)YFQN;|XqN2kt&4FCHFz$9NEF^~H^hZc2;PN{vBy{?CQW~aCiCUu5cKI-6kqbf!8 z;Q>JCvTmrZL%SKC#EWUonJOktOeG{meNFu2nScoTm+YLp>_3o6F227}Bz3Ps>a9Id z%8@OKu#7Wumm<1gJe}O;l3Y6rj|dPF%y+edeF5;|553H>4yV+h{-MWq>%?VaNa}q# z{v**H=pKOblpibw@lc0=NRgT)AC}?dj(+3_5PX0T{yF-b1n7pi@1xcHpF~DJ0P9?P zh;8lvr|jdeKVMD$6kc_?X$1QWau$YufT`r`LgSv!@~g~V`6TT%Q9C}&5i|UiB^%bQ zEx@?gT9QBmT zZ}Grx_PV`(q;lmJik?}de#>YkMY?bo-|BU{=czBJ?33RPiqN7DX_cGW}l zA}*@&RV|lAc*Y1IQ`MAhl@ploX$t)|4#QYlH+fvXgK0|+p7$mrz&GiWp@tna-2SGl z6&^@@QH^dbski3}9ON4PcmUpPO*PT-&Qd0h?=|0ULquSVwKtxP#&}jnKCUxb7yF!g zi3Nkbjo&m5eyhqG2|%HeJdf+iBX?U1AZ6-RfOo?lJ4e-PdmhCo>PV{4V@gs-ttm?Y#zBuoxo9WaNB+2 z7XV@KL$H^}8*fo>&>8F0W5g%g9<5zC=IVUD0CZt?3vk8N(|3VjCNSJw$AkB^q+e)0 zNgF1&NNHw6)=xaA-Xyg`Oe09w#IlZql27@i4sY4-HqFEc3vKT7G-L+s1_WI*?D0${nn1q@DCvDNz^gG{DP zK6$>QH3o*&TTJt=Zvy;^;NQp~TixO#&*b;r{`m?x&RsuQW|uTn)?UKt z;^FvXp&8rA309fN!t2oIU6jHNZZ6Ifnfi=$sp#OrsJQ4TEtH>n)#$H67*-{^m7DZa zfD_wPl-Nx{_r$b~-WlqDAm$+z7$jf)RVzbHgjimC=Oy7K#uru8iXh`zi)=yVp4E4Z57R_9%aUue=TE zK5+K@zXR3LGwW$4?pDxlBynoOt&>9#H8A2qQOoDYX4M4{erVVe#lcrheImN8UVOG> z7AbqQA)(8HR^(pTMq1vDvAe!gg|OU7_kZoSzeSbk#Vbf1^BCyHg+r7; zx$w&Dy3`wX!~9>Rwv8DUOT0a{n2C0dZ2&y7_^Pmb4o8icK5ON8ZSo6;Vi9hllHM)ef2s{56D;FAYeIzPtW>*Dk_I{8pzQe5xe57_ ztIURI_veryQ`|$fHUG`bk|-)CUZ>kGoEVaCaKpxFC2=ZZ_*ngbtd3P;OzCWhO$pV9a73>)Xe$BbIj>@yeGb^WP}7BP+95V?*J=>Cg`9MiK)+g9(~-$NnPj^^k`23X%o=szb4fPC4f9O33Dwl|_8d;Rq}r^Ad!ZGu z;gr(a_#RiAwVWX5IBf}n2K=5LKBHC%jKI_;ZqL=dOu8vGLBEY|PA}xLe71WY7f)jv zZ#+SIZH);Nxq0^15+%!Pe1==m`^`U%Le}Z=w9MrH@R7yqsE~{=e|h`CA4a8%HL`>G zU|zGjZ!!+p+NpNq^IH9^+F^RR0blJTnHM)-cy$}M-~9oso~~2i)m0F2{{M|8yfc30 zZ#EDJ2r_(i`b7Sbn<%?M%hy6cgc{ZxF?}xJsYk{H_u3$S0ehoFSoxdP!w=VM*fa9~ zmE}B!3d=d%CA+q~j`k6DPLn`Q$-f1hlW8EXAW!}`Y@e#X>K!B6NSL2Zu(WJ`iTU&4 znf%$uk^3%A)k4&{WF5tcq$jftmVfx$-E4ayB-XLwy(0=D9RyUmC?KIo6Oi7UH0fX{ zA_+A}6Ok?*r1uWeTOuGJz1IXn?;-RS%8loF&e{7sW1Ktoc*p(YeaCx8@<#@w{K}eh ztu^O&e&%`F>Eebq+$hnOz3YTWzrIO5_TF!l!&FyisMYFoMo4Efr#!+HPWB?_vBmbX zx@0wKQ@^g%X?LqQyVPOIs(5J4&k6dSnzN?Z0}ia?tEu%YS1dG1a_*?K)orem1`yeO z*LaX~RuaC2ofV3JV<5D>suDbR!|fTXd0PPgoD~ibP%f!uc&hqr1~{MU3}A9ctppzI z&2y-ZjsXgIdvzmqKwoAH!Vf6?rpj?L;9dsVSz8vt6-2V}1BDX$FrS{U9OcfO{;+5L zU+e<2H#Hbu&r(EuN9B71(Z|5%UE5D&joDhcsd73#I9R1- zr#ju^k1sdxWX|)OK!eEt{5bYERyeT$%T&!Lw1mhafknbU6X%&Tn1bqgi-SNLB@i~qHp z?+@&M#pC@8v8YV*uzUUa^t#*$QsRHb#%AW?)});ZcAq}0#)f(W9KO3*y4ls2=R2W{ zY|xWEo--v-7naVF@Mq1Jj%fRz_&fKZt@n3 z2-KS-+@14O7y6(9(~`Q8aC}%9^fCxL591P=u|@}Ae~MU9^xvoH*J;0}Y04Yuhh)PR zEBkaAZiv9Pu5m|cZI?qQ#UKA>_`h!7N?VlJ-h8pP+jE*Mg7oV4Ih~_$>N}jRbx3J) zcS!!t7Br#Q`5MASg^PQjY`(u_*8jUE+W$oK+HR~#H|$dK<1mYc5|IX4e>9xK@UQ!l z!k52MiAjccV15|i8DhipviNzdaOFWZnCUP!^Dj5&K;dP;=W8tRa`h7LwcV6$f8%@0YYdg!|8i*m z`K1mS_J-@SXDD&}wZQ&Az2E;e7Hr%+YQszPxA46h&k3>A>j)pt_`jb1e*|nl`OUw5 zom%^l^iSjYcWmWfKl*Zl1xrnSevqsq^Pk`Jr(m=2|6cUJI>LXApZ`x^^g`@rHg7=i z4*tJR=Rb|yf4rdzR~Xx1U>eie*f^TemvHZHx25Ez2Zt211r`3^8UD8jbr^8sU!9eU z{tA8k_db9{`1}UTdj`w2{BIBKUj~x7gk8gwi+akd{11fYfA9}1pRnG6^sdui=Yaog z4=+1eu}0&1@FnSQ8O_I+q*#-@{|NCn>DymqvF>4M<(^>hpN98WoBbbu^}qj5w7j6^ z(%iOf|-|>R}uVwy$HGe$fz5g}Olep}xtyCXK_7lCd-0^Q* zpZ_pyG;aL^G#v>s98l`b>L*dO2w>2(R(OY;xp|CitX-}qb(Y@e63 z0@7F5t*K(bzprxt(;3TQS7q|J2`dbXfKZk5CNm8Ub3G&byn}m1Y+dTqzGtYWw^Ggha*|SB`QuX=!9P9mz1ZK2 z>w~fYyD(Q@XEH6e5bkDw$o>{z18O~*s+!^NQiZ&YK1mFvwKk1_o7oE7gI=xW|#8RD3 z;8R3@qQ7f~G(5e3*Xg7>w_48o-LM}bZ=ANamOs_0&1YU*gtlQ>&eRgYJYRbie-R}L z;(kdr^D;Y{|22-~fz4Qi;0tEaLVqK|lF?91TWPb?^D!(=l^F23ae15C@VOIWc)Zw< zy-Rnk%;oYaFg&fsMg^N^>d#Mg+F62PDZ~nCaT^$H^56~Tx~Cu`nIC#KTH3Vr#3oi% zSa={6^)6+n#d5IesCH^SSzR*Od9tXkZgafKH+1RF@f*Ym%L6qf811Pq*}9`=O!Dh*Gaw39^z=2Scq=!fIi%Zv}2yd0`kMM4knAh`L_Zz;3hd&N{h=D#RnIUzn) z7#Vbv0vc+pT)p3Qqr@0diJUAmPI|NY*!_nfc?Yb>sP>%=2AJB}@Z_f6DaECk&KE_N zZy%f1e{88Rjbti`Q_9u1q9dyh9r;;YQ9UTO-SA$&0HkUxRYvXJe){yG#tZ}<&nm|B zr*G5~{k^rdme!?#J|`Uie@3Kd`SCZ1^j9(06-kK`mP|a)1q93&TxhkhmE6faV6|Ve zh4K{H4@qq#B;X8F|HqN^_d_KIjOs*w*S^{ParS?VNPjal@A>0O=gRdGPI*j=l9#2U zdZ4Ejqj@)!O-u0s+gB?q8&098^m=h@F|ENGH6$GNb_LOe9@pz`>4T`tT{raA9MVgi zBID3YVyG47eS+CJl#s=jnlL7jn8Od)dQ)#$frNPe?BjvuQ7b-dJwd_w<(1>c$miTn z(?;_SbFwOhTdEsNO->`<7%bozes{6a0Th;u@W1rM!Lv9|irEDEMq2}Xq}M77$->}G z;LU|({=l84V{!&j7R#=P#C__5CyUoZPQWXA9K1Mx?!New@$T;OIZW@$81XgO74*1Odc7y;1Jnq+?D z%z+h>No%qA$ZHh%s3cAv-0*!CZFPJ|F5fbbrAQRH!A)kjC(Cvd>wrc<ny*bxOX)1J+)nJw(qrB$vP$FNe^|`SwgE-H%enovssn& z3+;hs{C0V~sRY1w>n$czfYIJ7WXuDts0`>z#8v$G-_J*9YFbR9U-6JHryCvllP^X; z_=cYquO71D4oZJ^oOEJg_WQ1+20-Z;Hal3fm$U>131BOXiDQ=hk7i!Se&+n27Vy zCSdnxQLQ~jf7uK|jjXq1ZMbOE@s{EZbdH_u?6N(Zl5{uA3$*cY2Jjr_r~|T*S`ih(C*FH zragug$BI|F5dluJ_!<_(c^iK2QXAtETD;xr{BTUUoe>ONY6H z*4WLcyzUlqS)t5&4;43KOXU9zBp~U{C*nMdQdxi4!o<@Y>zCj zMff>uANF`^?V&OWeU$X1_=jWfI_N>6b?@MHCQ<4M{K<^H1!=G=KI;h_ts1)LPK8g{ z6>|CY$gN+H%bp}XX(onJp>`!OJwJQ+x5u*92GeO=jf@!#7`UTA8&8xI%j52F@Wl`f zMSXy#Cdwu&po9fg%m}#hT0~vnsT5Eoc^o9)_wTT1NxXk+mh;vfmj8PC?kF2FUUeucEhAY%k54?QFO3KOcezm^{ zPkaP%Fsd%MvAt?qimVKC8cLU1iPKh#oNyTJ=W73vp4%o^>wUT#pTL{amlJhaPa^&( z6a3uwrCyn7NtEwxf=9$qp*zjhJy!vipImwN0gj2E#E7A)aJ8q~5uRHfW*nJX%{pJJ z!Xy<7lfzrSy>2^Q&EwvR#nVX4QajWNo#ZlgvRSJ?VQlIt5wfx8M?2`4p&Yyf*yaQU zMOO3l^PF#_jix`FRP;^A+`~Bd+qpZ{oZ+s<*ny&+U0LNOO3XY6`{Suz+vycWU3_p@ z!sZxLLdBahuvUnn1~spb=;C5Y1uVvONcC%C!N9QLA6%I#UKTZ`<21_Y!(oBDcAOiq zm+_A@RhXRSWXiRvLM*pyNIZ z9`b=|p|?Pb!owrvyf zSe!l)nwVhAM|IZA-J?NTP%({?+i&EN>oP@5a%Q9bOM*CIqN*8M{zy^R^;&(dO>TJ0 zQoC~zd-PnlcM-d6qS8NIjUtO9b>;#s$Ys-)sLZQu~;!SQ6qqC=r? zxIn?tt=d*Cekt*8gacQKzg)1u2m(7a=m$_c*%^ely}zz?J(#fCZISoNwmj~|(?rK! zpTN;8v%MmlY5mWgo22LbT3@o~EC4JAJ%MZV2Z|UzquT(r%EO0>7F;^X@8j`?p2Tza zR$Gony{7n1{fd7=W zf>x^@(AoozefEjsFkuHW&A_leTJ1tzt3Ag_;=-ZZ1=U{JhAr9Y%`69wY-wUe`w_D( zmL6f;){FBCNNLwM^ z#K(2gFfdZ((G#FJ*Tp|k>Fju5*NB}*A z8j-XzoeBB~)U^NIF!BkN}dNwE&xvYXCN9|)7 zf6^6!?h!%m3iHX$7ZzckE zxJxz|k44^?8NQMd5`FaLqR=sxW=Q-h&RjFVSUTyrhbz_Hbgten<6B;22Xw3VyFY;9 zGAIQEabLHthc&QY=0r)wBXz~v--Ry8;}zh&48G^%JX<4Q2fTTW+)U;g3f0QhkRB=_ z*I*LCQkOu@*4y3Bq$+$x^ChG5sB~bMZz_1Reu`^_x&tsU2K1@G6 zKg?iHEbNq-W~GjjX?4LC+c=aAMbPubbDJ!0ld9=A7|$6e=SL2O3}XfXTL7EDtv8k% zQBv0$GC|P*_MWh-~jvqkoi30>YD7%)OHnK65w(&?mktYUYPuf80w}V~NNAB=f*Y(AOxq zin>q8aPwZDRSBb{P!V?kHh%!oU+uQ3zvriGTc3oFEFgdTp*w@*78iodw{tZSKmFJ8J@Rip)m{T$ zGc}%JI(p#7rO2~NR5ak71zgN>2%C{Eu&H0Q&G5^@gL&%s(V+OYDk^(dfHmH>8NT^< zd9_3ZoZ-9Ku;X2C<}LR8)14W1T2l}IUB4Vugnm}~4;h9y&+(zd^_DH}BVkA;($zPj zsj1Y=Wbw6oTDx5ihp9DH3exwqwbU3UeXVuQmT(OcUljnvB^q3K8dgOGJ$sH~>SEm1 z<0o#3Bv~#iDQ}eKAM7L&Q$N}u<$C&PB{oWOJ{R+}5EFLy@y56(9jU^(c~sF?*0Wb> zUtAA}LpXwbgFVYlbbrMl1urI$XY42sm_a4Pcxr%*cFFq^NS2TUA-E`B8M~mvi&*O_tUSd86yV46oE6zci2sOv$ zi+E!F_%U?Vie@uZ0Ul#ktD9KNJ{}{oq&YGx2UmTA)+&C(!FaPau&bqCp4;6iY&kDj z5m5DjW_uN|{?3FJ@0d4S;!qBTd+xSAnAYo*^%=rF3PM7Upr26rbCF3W6LaNRQOkOv zC)*us1of~pM&%bZ%e2q@aKX!uWjSvO99lwLG-LkdA79?b*+82ZmVdaPLuIp6OS`D} z1){N+_OSAeYS?XOI*Kd^zs?oecC& z0%3{Y1_snF(=q^qdln1QecGw!XS~)VsPMbUU3}7o=tN98i!i4G{_nVxN=!;V9c|6bxUu?YL_btPR91g?OEis)9ytqB zDv@k$va|T`tnHN8{pZ1g1$(XeuW1o=l@;*zLDHKX+$XPQNR0bMF!soq`}vYe4=GYE zaAlmwpZi{rjW*k$o_+kd-A1q*qeme;1%3>-h&X|+l^u^3?)%Ofz~<+2+feQ~hWq1D zUAGp2v!!`7*(@;g7!q8fSLdr#IZ^9}%Vu%5uOzq)6s<}SjW=9T;+L6XUQJo(P^z9^ zFuS3YLPj&P37K0e4$@hS?ay&uC3`H`U3#PKd4SBmZ&v5(*>-bYwFa74idM1d#g`s^ zvHm_moNj4H1zKJmB$#SE@idX@!{C*ji#`iM}hcA{wZ%I>3 zbbq9rE7Rb+IN2@6%W#qary1(;GBvJ4C$Yhub*fjainoxNCv0@qIAQ+yM`*f5@Nksf zrb919B(mG}@Rzf}p4DB@3|M<=(Y)=J4w5V6m)nr~XgirDzWNbB({!%4yg>5GB~5Jp z+QhdWKm(W|4$~e^H!=F&aP^`C0HTzVcpL8auukw}CzHEV(PJ#OU}J`}sYB2PK~n7G ztwu7kh=0kacl5R(!yZwa0RmV-yJsiqyg?qqel4cF0mdv>84preO+~4)^rO`LWzsWS z$g;e!UGSdrm?}Wg)LkDgB2GNyCd>Bo7UOwM0LF!OXY1bK)6#a~ z227DsU)7M>@NJ}S-@3!$D++Ei$Ej9X^kuuq&dau+BNYzEg18y4EIO-=&lsPH5{(9? zCNXsUcyMYbZBeOTYCv`FfDySUM=nPV<>C7=*1n{_fCQl5%7;B-9W5JewUufMx&PjUI|!NGw^TQmJ&Pa`GsYtu2DU<5n_g#J2iCr1q~7yX z?`1cZLaq_=x}t?5?Iln(+Q9W=?`+biq8k_H>Z3d9pmhl*USIS(LjacIwP zbFbC=HqKFfkw9GB@G4I^iez8|JVS7P~9U zwhE2-0MLj_qiIVu^&emcMzQy2nWXIs?ENkzO4*%cB*K;SGl*&;sV{}vv>^0+8JEc( zy^4Tlh)caA{56T+{#>Db#o&kDCs65)FJts9F0-xPQNjt_`f9!+0w7`;-(pZs68n;r zHsEz-A5MF((Dj_IZ*7YBQvh%F1eXhcjJO@AOeGvYmQ{3%(p{-70B>c=Ovj{9Ek=Mk zPWm*OoxzkCP+I;-TN7IZktTeM*NSbA@ou=Dg?xHoWM@7ESH_sf;78TelD z#w+XbTvb2>_Z#XxsXdiDX>C{;6d1z+MZm19v-R{Imx~Zab~LH;KUn}y6{l8t`lYdD zoILqL^s0Li!MEaz^Z7?Tznf7M`N4T-Lles2+a?XYTV1}rHds~nMKeX`q*uu z0(Id*Io}k(uc~R=jwWk5nw0YwAF*u)&6~^@k}TN{Qy*7D0-SgF%8`}H^Pa?~nm6Wb z4~`=ZSqPvNYN^o{cjn54Gzp|P!WcgkQV$RPAQxQ$=dfIMvE8nJt(WXFwdv%CfaT-W+vWG|Fb>(V|LHYoCvb`UX}6`Gt< zs>xW`u9ont|DxpDI9~=&7JcjmIb|dSI-gjh&#HpGdb#4g;rPv-QHGVLVmTpqXq#s#hAjRQfzY!2!tu zwAisn=+V6y<;80k>hg0rL?Zu-{$9lO9!vYZyyoD^ps}c1iSd=nL`;6oMb{A0$!?ZvsVij7qFtG<~5%+1c$kN4KOVRiB zyw@&h%Tzzm(N_N&;wi5`7OX2XuBf8s9M>19Uv6eT8nj80XJ$CAn8Lw7nxLQetdNZF zZ!k7UyHl^9<`LxGl~wtOWZ7MS;*Jld*3qEFN{8+0!s8%tUu6EhT|L}104>)Sy$Q&7 zXt7-y;`-Ak<5E4+*nmZ(Sp*|((uW&T?*$`^D1DZb>T6_=Pgf%u?a%t&-N3FPwkcNy zoC-p!GuKNJ)yqy8k%vA&}(nEW22$Uoty27ny31@BR1pafcE}f3GdPJW@z0c z0Je~4Pt67UmLD1A*56XS*!qm)&D6y?$z65Kd7>GkM9?xe?_(zswAhC-3z@`hr8rse&u7Ll5P3su z=A`aj2vjKZ^GyC?Z_TZvTB2KE@!<0LrQ}=-M_F6i0OP6D&D+!*0->#q7YgV+dm*1= zK3-p-(Q(T0sN{uJtfkG=u7rJ*`_|&LKD9t(`o0t$X}B>&-MzG#8lU6-sKv8$>y5n; zx^^psXQnOyNP&pgC(tX*ShSOBYW?`dx~Tyh zDx#4@=q#0o<_e#5;+myGQs|jydSRccc0I?o;$ z8f9hY)tVM@a~JR$eHSOQ{$o)qOl6dY$7rpx5!>62eJ7vQD2c;e5g_wEkKX`U6VEgR zcjmblgyF||WY?+vsb=C(a-yK$4>FiIPHU&Kes0z~79U=?ctU=8hRae^UH0(S{-n9fe`;L7lA8*E|?{5!c_B+H?#Wijcc-&P}n#D7LxCBARTiZ@5vXUa8 z4$0UUQ$nfsk5c0i7FVwraR5_qz}lHU1{Ya-Yd=l2U%V_@#S!V`H3xyE&UdAOB0L}>!(d)yn+9_&)&dwep?PLv!rB4^j z$554CtqRKe+K}=`#*b0?vt*;g(EDJ@s0SibQ~NSVd{+fSfpCokJy`8v(b&i?UGtj# zl=jAQ8?!r^zoi-xe5^}7Wh+i|dT9``~$_S$%RoPtKgQ_pR%*!zX4S zQH=_OJ2>pww9syI#>i}z99#?_Uf9g!voS(?6*pTyxA{|vXhtaL`6K5X@|TkG-ejD6 z9V_0{Qz)}5ec@ugirBivxl?zkp$+w^W?Sj9Y@tRMS4k=Hr$`nzYaZ zEYeSsdgs;&Et#*Lg?CV$kc7R7Icfj-*d74In1Gu%=G(yLL{9r8|KqG?xaa z@B6yBO9b!Qb|5k^Cj}Cib}L=m1eaNnGwUstWjzhJ{n7BGWq1GxIUR`1r^r;BafQA+*~=k@qQhwA{6tK;lZk!c{7#Hnh&OM&9N+YsQyC{wz@3 zYE7PxPm>QgAJC}=ClnFT6L90(U@i=nqdI9jyiSM`-S=?V+p7VZK~|HbuOh|Bmnp|i zj{O{xaww#BKo5vzVK+Xa^)`=6J-c#TXVFD1a;mQXJ3{&Fwh@TsQ}%)xLrzX z=JmXtbjLC}949E$yh0|BNK?Bqq0Hc(1*e_Mom^XdMwcna0Q|)t4~zC`)qi%7XP~;a zo%_Rv17DCvvQYY)kTRh{O_n2PJ2%g;E%?qOZHNDun+*KOv^x#jj|{~pHC>d#Z*%em zT<$-0Y2LgBs*ko{X*gn}B8U&EIiym znJ$s;2l=YB$eiK2B}I4U*6&jcSygI)qr?{PzgRmbJdH3*k`rq0!?qHWXb(@zhiqCP z_PtY-=mGheTge zP1NR2!ZAs>?YK=!w(j3NgOEuEaycZoF1)cl!?J7GPj3@s&)J@Nlfv;m58}fxu6FNgic??MA}zCg}G(^SXz&yB~bbKyphY- zm58&=G(%!a=+b43KPr>=k(Rix;GJGUPL0;~KO}+}Wn`NfxP)%9FB{+wzy>Pflr%gw zF5Al@xE7zPd4HZh-9F}}+q71%dppv-Ja-XvA#|+stgW8vk>yArBf*&?O(Lt`jMqss zqa=S@7LevRdok33nGkd42GeF4p)=XklBv!<7D zDH{04FLILkoLbq?I(Lvq<^3_t7H5OmXvzgnaZS*X1CejPuX<0B1bbMn}d;7Wl=kqrGgOYJgRJqZujV$y69gLKc3yDRS}c@g<6d? zjQ2IEO_eydSpV(^4sPVy4B(7;C$V>YK#%Ot6{Qn zYL%e-*s$DO?58f38ZJ;1eO-q_q#Hu`r1E7up;8O(RM(qfYQW?|i2jUVJ86CN!$(e( zH60R!=5qk*9haogS4r+9P@o$GJ*{gqU1|bjy`73RPuK%x#panIK(V}JIk4A%>sPdR>{Lgj}NfD{^!`<+*Ws3%$SyIQ7%Mt-O8 zRSn*H;uAqpViP5UPYcpBuTJ1EY(w}J_ac0PP0k1boHBu1ULp$P+?<25>5<~R?ZPDf z%%D4%t47+S)?psfHm5rryAY^jT&S;gG@BzA z)oUNPv-i)C@Lu8?xZIA#OfwRR3^UhtID~ZY-mM=T#>pI;LN`7p&^O>nr=fMDzb|?E; zmy*7MM zeGQC9Dsp>_bx;NRu;3WTG6c3a17uTs-qFK(!n~#Y%_<1=t5QMI4P!||@5TH2`;{<` z*Xb^}gRnrkQGK}*1rLh~z(r!MF8k_2?m(7pb=Zk(?)UzN@_y=N)CaXyk0CX6D3|SZ zB12D%=f{6M7F_yoj|3rSlr|!cvKEY`6<@u|Tfj&~ioTjA9km%Uq->)NG!bI`>~G4m zBe)+Ns;a_p|8%Z>F|#c+V}6Y{1F$Pkok_>aad}GZLqH@pFaq0Up9hk!8N52J0!qg! z)}>Fyz`A_B72;Z##SKZrr6B6wU%LpQ$HfL9F{ik1m1_!c)a4Suk%&eS?I-e^O0Oa(|rBm z^UOcc%onGb6SX`L=+dDQ#KHhNn?qB|v5`wR+s%PmAndAwo*|(={H&t5%#<4qs96S< zOyOom!5OLJ)nD9n2+%&9Kk7cKeRSr%6Dp@zy_{}Ow847}z2kMK;I@+p0;Vc15V|ob z1Vd3vxGa4dwv|)C;B9t;kDQ&Pc(Ejhv!Xr?W}dNcwwQ|IM4V77AmMAOmvaHmgh$~L6`dw^^bL19 zl4>istF~2kM(a@xNEG9nLt0+4XO25vMw|%uCf9DXeL=}o0KPUPfD=8y#$1sTw8%gE z5bNJzJ6^qW{O%z$P?|1NAM34$gbOhth?5=UD?1{KfU7z35M;RlOYr27p)g=k{*(WS@lC;?ljF^(~Jhz6@ zWtAIGJ01ld*zp{AaDnmyR+bH{Q+!8cqsHP z_?wk^JMF_W^5SjMu}sLrOL+;qSxGQVa{e+}@kI?>$#@TV7*$Wo3e86+87@nJXAe&O z+>)%MY5H&f^RemqL;WmmE@)nHNQV#_70|Kv_OKt4zcPA28FRLnC_$Bfkm0HI6RmT7 zwibe9`zCcbj-Y}GVY>Zq6UeDv_F|zOZ2u)TBK;7sw^m%AyFbT!gkJBCh~9o;g4&qj z$ExA(au_7sjG;Xqr_w955*&`iyEKNrjYrJ2#0v|8Po6#Ya<#ac&(fyQ6i_x z%!|6d-seiqwL#%u_?``O<*liTY@EjAWW*NWlTD%Nr}z_N)z$c&w{r0wzbOsB#FFUW z>*rh?$VCm#h9hBO^>ZBP)|;$1uAWNClE7@?1X3E1=)uYhO*PR)To}veLV=RdXJu0F z6r~y)ZOE>e;^wa-t;dr^H#s)1<^0O3%m+)f8eF8ihs%!X`?|nV(O)_EW2OpV8$Mvz zJFKK&jsMM)45wM1EmOexZ@~vOMvZBTU0_#(&EGxy zY_mZHJr%Isesib(U##zaYq!~;W~NQqkFr8)8eqxdCwAL}5|?`=jyp9`A{6ac0$BAz zQ3;M!U&4^P??*3IL*mWf+Q5UYaKnQ+Fw)VsFX1rxys_wg3Xl*}%ZPIjJN(A$KUIO1FO48rKdHJC4# zEK=vr;pq4B$zLw0dvEv3f7h8lnyq(+-R0F(^Y$K^1HVoj+(b#&xIFtGx>5XPzniQ2 z-vXaRI&iZyY0jHO*qYb*WzLgTrJ^&Tuo2+0CtZbbDMzSXg50 z73NeKF3g+jUV_2qEdaA42gPWj=VwLpTI&fZ7mPPMUJfVyMOv3%vm?_nolZiD{K>rQ z8Q^l^w_CX~c>Hm}hc*pm@|TcR=t{bX5lXE6I}HF2WaEg*hwzLPM3cPg~HlwZLh;e9?U&aZ~2ZpYaS&y(J&&L=MeuoYp0F!!eQ&~Q{Z)SLB>~lN1el3%z6f9<*RaE z$>Qr2@v|QiqtH3o*&VC`t*>%1=NJ=$E>FN_bl11>XJ3xfYGgRi7Si4Cz%G7{U^-Wz zT7x&y8R25sl#Et>|EgA&&x<0xV=*&7DR2YZ2DJWIRkCGss=>Q(+?*i_SP4BQT&Ng-3FeSQ=^raorL8{Q~zB9npR zOCl9435rqfb*R4YY=N-WnFjq7=roX;hxoFcv(CTYpoLk*u?mp(-h2{F$L>>RAbV+l zihD_KHCH%8XNH6+Bv>fwAwRC;IFSDpkl;9u;W>MyhMogwil4Sy@zzW?K$6+dgmBbI zhp{>1jTgfqENZ9+aadvst87+VE`3l}7!5zIN&tuK9+0Hk%-j9z8<2e{>2(Ul3Cp3& z9UrbR|HYNl8ZP^H*eJ&krHt(H&b-oZ{VybB_*Q#Jta>lEWBf~m3KND13be4p=4w~` zF{51WHI_^_(L-y4#t&&t3|`Xkd#I)=;*$UvnC34aTZpVO1iU45k|HH~-~ET|ZE8mA zisE7}#7@r_tX?d!gnMF`+7PSuiRO%;x2YuaU%Ji9Oy$l6#Barh4*XXEEj zew7{Erd7GJ7e7Su>3%kPl9o4)A_!%{C>rRgz|E*lt?yGRlGt&mUWWUOslDx1pY<&;?wt5FNv&ikCd8_tOfrp%iKow+-uAnb(3vhws z!o`#qS&=y4pvj0!F_tfh-o(W{CLqsUx3)BoHwt$hYTb;zW)G3#cwT(N#k0pZg$#_ zmMNHBt)l7^?6Pnn)5q4|agm`l?iL-(TxXFS&c+_!^evI+luN5Y%Tb9@QXbqvJs;Sg zJ510LCI$9-*>%TBH}iCHpACk;_bBWOV=ma4RcL2>!_(D?BNcuVUaTX}t9QNSh!rW` zqr-|74_|apU#+lqyz1!{nG*GVXh|5B)l@8RS3eHz@gY9O>g214p_X*pECO5QUac-Q znjK!Fn9y$bvx^gEo~`lQJ@MQHJ{J})aA6yS$EK8dVI3`2%SjWRIIhI()$*wdR>mR+ zEWSv&Pq6YUrekvS*rO!?ZLozQRF}Q3gv;^#aD+F>2dTQtM1I&oX!{)+%Q+TU1g$=j zvKdIq?tZz!Uu@Df`yokCzv`<#F^%-JrIG(#rvXOXz)Jq4+&`Ke>U zi8i#{pnya^eM9x$CcMcmVb1LBNk^XPR-9Y9bH)xmx~qM~@%zCmEr$mHoFi^-9cuT2 zf@hAiicON~(~-`E$4<=>Cexx;<{3Xi2?&)~3j@Ty;Uat9nvag4?bS!81DumYVq^2l z8+1&4Jp)WbP0M9;C4RUURAbZKyXNv@Aa;?4*7Bq6IkF6&i2?UrN_Q2DpK^NN4b0bH zfTZ8HFP&U$29P2@UY6Wq9sVu4^gCC5@XD=2`36cuvE9jbqV|~Zt-vNJ>pjp+Gnh4i z6@SLSgd&(RMWisa^-gCPG06D-A%)ba5;k_H{^cHTHu!Az8bc%$uw>I-$oi?>V)reHM3XB$$#`8gN!9xLm^jQ#{9Sc2>zT0YsWQL!_X*`c>_es?QJiPOfRcBB zk@;3mqh&I)XE4Ntgg03h;An`J5f;!`tn-2P@E@~NwyR;cz%8hqpVixaR)kP$;CB0` zZuLDz<@uAPxBiX6bg{lC&ORw!;N;{@52i4JR3K!CrgI34=&GWE%~+|DNufW zx%h#ca5{76`~9c|Ky~k879YR#x9^tBM4p76MAldeD#=kc?+OLsVP*@Bh6C9Y4C#l= zQ&;!2?Pv}Vq*9ZAN^D6JZqok5TjgHK0)1G&>lXF+aNAWbsO$lBem3mhyQ~7C%PPDzZI=f>*xY*wyIktNcbACD6E0T@5ObX}`1$Yw?{fXY$^)L)H1j&%*>PkH zD|Ge3k4q!i1H1!#wSzH1W}Zx5n!_XV7%n>b>8uGY4z>Zai=6;Gq!ZhvD$c`~|9D2z z;Ue16I}eZWGq77GJcTsoVwldiIVMiNmSuotiavCa#+Nxx?7^FL0tB<;ZpiG}JVCmr z)$p(_gOK!Ta5Cmq^T`JNVkA;RMNw8t0^ts9zlUP#v|#TNwYl0o10m7&>&$Kn+y8GZ+pv5Pq1+Y{4NPtfiH7f<1#$y`DwxOt}epTW}OyOuAM~Dw2UcsEZG*F zEFK#U&DigN?80GEzq5tIHpVpdn33gJgi4YGGvN6>1L%O_3s zBBG0K$o;ZN=W(yRK7A$l{@f{DJDy1NjY?yNkh3Pf&>G|gOsuU!(jOkWCf#7M>qAnk zZob>J?YQlMJ9kXvwks9anOwb=zc_DQpFZW>p<~iXX)(WMzH9oEF)6RSw6r7m=^Ed5 z9R}iw>xsV^i1`LQv_NVn`?a<&&{Dv6D5S1-R$d^0j@W>^x}Mj}>1(gK+8tVQx5TKa zxo~W|8s%BMV7@U|ZR%8e=YEs>XJFmVBeB`E#07B6;vq>r^7OeQ5;V#-I~QI@pFc0> zL*;BdRMG!rF?V_Dt2H3!qD9wu$I|K5Ec( zcAjI>L2M}iI(O;^U;NqJ70pw7(O2d4qeafm3!A3g0HNn z5>HIB>P#>a*wyvcst3HOa&dFKHZhW7*QTz6R(ZY-l?*=8R|ULV2j^sj7tHDZFtN5k z$ldH=v!vq|&Dfq3*kk)0rSR^vHD(lfTEf7zb#UkBW57`i;v9FsjS=~);(ouw#Q*z3*b;Tp^l};RtHT2ps7-SfFU7_>b4x^ zMbXsSdskO#wHhLaamE|zx=*Ktax%sOb7eYvbAQL*YX-4BevqcIbDaI)xjC8C{X5xo zLd*s@C46(#v_2gsLR4nDWSJ9AhoAkc_i1ml3%v-FURr#3rbctSqxbg&*br<$ zsK3(c9?f4Iy8d*UsU%xJV+RvtwdEWR5tn$PnwAj#smJ>#BelNiiRyXqi~5h~ zoRG>^KZHjQLs-l!T7Z_h80{&akkeh*><-`aq-t&hGg7z7(p`dO_T=zPr zd9n&@CZ*?BU{N}u{cKBKq*M?fsl~M$dEI(gj~wlP=ye51kxx2Kr8K%3e5G*T=sJbo z=P=rgsftS{;^iE9tfxZjB`Oll){fh%vu3tje{~cV0_3D>1Aiao?#BxCZ4Q|po-s+; z7i%X7XgcnkNucnY=;wW<+I_J0bJ&Ll97H6`1yRE1m^^1vC2alg_~cHdo=Qv$vmpe~ z`Zi;DoSP@J;GOig^nH@{kEsqK!#u3_pCb2b;DK00$k98YaL-tMjgV`oWBsz4O(GFB zlX%W=bF$oGabGE;=7L96V0rI7Ax?{RChCpb0&uc%cbgHlxr+$)%PQJvRX-poC>_-` zg>B;6+#vUYk*3DBE}b!w<(7XHbs6L9nlci4CaE^?bW4Qf|KjZ{qvF`MZi565njpbF z1cyN5gy8NT+(U5pV8PvkTN2#eJ-EBOyEHWRid;G8zIX5a&iCgH#^}*>QFPU=wb$Bv ztvTmHz@8gOyb%%7{8B%EAuJN+LFyByr<>X-c zw^5)f{#K<8w7|aFGj4T&Y{%=0HnQ16C&j)BaoG0q5One|8(7IJMzq^A?qtSp19jtB zdRHG_?N>ZvH5=H}6eavFZavOUF^)|H;kshbY;k>l^aUz+LEQ}4eZF`PQ^m`XZ64&c zmbS9}D6YUpllkb@Zn+u77{Zs->vep#Oe3lw9D4OMmEsR4T50VDH;*+gIcRRt@3Z&_oBF7Zm}B4i)ogvNI0JFWy#c8KQLJ-ILa6-Lbpi886&Z zH1K|Dc3YdRH32IwIH*2NC91zor5HK^qU+4mvJR7_!SHtg-odDOJ*XNJc{T&y`yK2; zHQ&;xmA@Q{G08|*FLZTM24sxgW3XA zXYjnPZU%ABGB5d}FZj{Uv1%ln9<(&AY6Vyw&L0h|@If}d7EG(b9j$E8I&!{G}^ zcoP7O?wd79-!#*^{N`Y55_4Ncbi`@yZUW=c0&ep)JKOy}$nt66oW<%v-=k-)%@@AQ zL$wxILRvG4Spm%1oV^z{J8m5|HP;@4$$epL`QoRN!N0TfpGL^<;ZY{Rp~>74Xg|1{ z+ojOR5X9 zY7&c1_u=)_XigE}<8==-$P>9+Qc^4Xn1)dcVrvGrmh@o!6i&b?L@3NyG10Un_%%$y z2=c~**iGJHn`&%A#S>gD5;D&V=A+mP)INOKM|yPIobL`40^ho}c`K38| z$fKvGE&#`<&aTH%L2FUkbBNOu!`@ifOi`dhgF;$$iqMu5UGira0Xf? z#o7^JA2&65hOpr7Oo0Dl)NNi-Wdm;da=nuvnh|izivR8tL#+LfI0x#Z}E)pj5IBgM9SCH&AJf>*`72cK{LP;KduTS6YBt-AXfXb*_* zz#Dc24GvmE0}m2&Sik39yga=cD>TbSS%Qt5)oO6uua-uHy5Rjen7l{{QYt9C7f7wt zT~&~$PBrBp!4Ac}GVJu|0YG!?u=nV--hAkL)fV&BAzUn~BkWoCeL$LpE|Gh&?wrV5 z+9)**>g&GX-?pp{Cd`oZmWO)9w4G7$5&mDv)Dc4=70Zd+3j&%zH6!U9A}Q1SZuI6l zMp~goj}=G{lr+m$ooc>Ecq|1N*_mDr zH%3tgcs1O(w5!eL*(*P|;yxJXGc)wT-j#R?DjyC3QZ@yBM+z=S{=KfBLvM3`wD6F2 zI3wSes1y-nQN~UTWgXFJLXbGuL&purw}uTjlg&3)ynPj#T$O(?;$DSYU7Yr7DN8>< z_8GN6JhL?ov9y{Ra_pmU`OleItWT_G0pdV$#P)zhi{L`1)x`{sJuya$8z9s0#N&Cq zJCDE$0HArL&8h}b)5guDRs-r8O$}9+^mx~XjOz{)CY9`?jw@{{5nS#);jiCL=BwCN zI=tqC)czF4t8a2W@pjuN7P;K49G7`@fzLeE(G>c`ysaLuaHD3cv28FQW~?oM0Og)t zzIMbOVn(3%cj{pL}BfQ9IaDgc`P4JlxHL(%m%KdS;d4xhL_=w=YDMOcmNcX1PKGeZ3nSs5TCAcOu?M#SqtyXVP zikUOMwr`xBWWsC!1is5-RqfY|iJeSu2U@$N^x^ zV#4yeMN-Qq4|_o>oq67FtXDP>gWi(={-WY1Ii*p>w|A_2epoV*Kt*cz!q?jF%x`oI zvKpRs1i=3-H1mQIx4M5G%$zp8Q@J8tU`J{XC-%A%mu)sUYXoCO^sj5VnOlH`F!}V$ zo8;GTQ)<`oskquu-fDP2vj%On0R_W5F+;MAhT~=&hp{hP&uc=ViKVt?@)f!Ys?bTC zjvoD%MgX#&188qRmT$2Tr!tg~jUsCRVC$`e{KS8PM%mx%m2*VoCjk4fmsEE0o)Ynm zo$h19N}J0FvJdwd^h8YYfh+h|0P?s@s=8e^`wn*5@t)mc?_5Dd+BrZzMmvRS(e+|d zq$iAMZzIUL1E*V-pEZ%`2;g9*eFU@m9&_4l#rAMljmPw9Z)L@)=jbfDhB>4f9T;E) zEn zH3AlG1|fM=^U2zeB3}*EsHLMT@Xlo>?@1n@RcicP@%QnB^803K1b8*rWot&(--$mx z4x>xHXzs()2D@o(mdo7Ta222sIg!JQ6E9&{H~p|&-W=vgx9?4rNXt@~(xAao;&WLn~}=|^g} znl-NmJ(t_v=d#{^hawN{}!jOcq$Qg^!Pu&G6s1u-%&Gdo`f zHu}rPpI-*jS+y?xBOF6XUbGP-JB4!HvDAmTR+U zJ{g2&YSr$_i#jdlNyklY6L{rDR#ESX|}) zgS$Z>{##(-lU+{@NDSqmFSg`B>MiQdvf^>`C41__E8lqlMUc>|_?`1A1Ex*&*0Jb< z!5ZNtn*RDLak|ywchvw!_4;;VqvmuWs$o$y$}Grd0Iv~eWGqVBWBLF<&%?vN>W`gK}wQlsmE{4Hn9_f{nl&-#$rDWC4YNOrxyvC5*+q07O z+Dl9-q9CN^RjsVwx&^OzPqEx}R1b}t&XKr}rVCR_o~u;zwm6l((QFP*i!I7@ObjDn zmctJNxqCkN-J`YJm?K8eKNGlFfIt6o8s%>vS)Z5q5|T9O`S*ka)WB zn3-Z4-8D0t&=brJVVMy&!6wG_ zi1lqqULXZ8jn7oj`QQ?7pteboS!m6i0D>6g1kEm*Rpe@&*R38-{Q%YhE;|j0 zo5LnJl&hTcgBnZ+I-OnuDP%3^EyAq))FehPULP0sFH*EhmmSZhU&BDhtEWQV*M>ZX zr;@(kX6)px*b$Ge%y*V$RXUH3nn%5X_dD$3o16z}B3{2#c4DSf4*NTbh%XKdW{nq* z0ZKv9_T{|f$qPHRYgPayan^S+O^&Uvcc+igt((qTy>)77uA z&)%t8U;@ZRBp&Nez0`Y#SfOo9GWAlaWM9V3n2u;Y)%|l2nlZi#{RR^|_vZA=yE>@B*W7aMGg(;;!oGo4?LGXemIL(XhrhoRVb3a~bJhH1z5~{Z=yRN4e%YDvTrMjXV@F6BqOOQQrYA zgwgtL>$t%YG28%%lY6+2=sP?&Qe<%Y4I1hhBu5-gy9BQam?bG4V=ZySow#6B_4Z zl{e(!CFRVA^cVNxN^cNcOG{teSFkE^^4}h?8XOT*ye=+OLseAPzEd+w;EUF|?(pns zx7+|gLH9D-tQThI8=TCAm$Au$+-JF;5+)X>H6lYH10Hcojo!HI?v>aBS)O+X<4$XD zO#qfB`|)grD0sVUb90dwmcMfW+j{3HGi?gZ`=6Cbl9* z{=UsAS|J4}27BkQ{=FRa_o5U9w5Lxa1*fcu*h(o`>zS<@rAZ%owyd~3w~sG2Y89Yt z24KOmWRqR(ZWdw#f`~drp5eb>ptJTaq5R0J++@sK6wxqXvkm%_l>%Boh7QmGubajn z&;h~};}ZNw3GDZGij%z;(2jmTmPGFJtiK;4YCt?_qY$(u9zmU^!M3CzCSiWdve-l zW9Xj$35Un`{qBS`vdI9t&1rD%#YO6Q`^b?#>D}I8TLAJ#lbo-aN^7^~E-m&38vYQU6m~4hoZl1rg9w>9h|xaXU`; zB92a7p4s(iKjnOLA*1g)M)NN>_zyigYgp8uiYKL#iErI7bs_69%yiKYOCm<;%j zguT8kf6S?Ws+9eQH|2Q1$W^HOFxURebo<++`wy>Cz=IFpXw&)YXUv8IHa$z%hWqi~ z-PNDQffT?{P@p}K{kwyIJjy>E z{bdZ;&X6%;_sSpnihm|9{=*v~OkgOY;l2LF1OA7<{o5lpWCd);qypB!`p=W&zm50* zzo~<&+2&;U?&hzX!iqJu_0aAQYWiT&`%Cb$nYbOjNI)nBzlixeng7MpMI>McCPx}7 z3g2Hhb!-xAF(-Eu?`u|s|HWKSd7A4yP?!60dC79XZu-q`d^(O<(yA$%LSU{ z`mbJ?bq=7?cq=vx70Ult2f)zo$&Ain4Nv~68Lfy>TZ=tk=Jtf_ifs>H$NX0-{lgu# zVu-J#jLXVuG}2tlyq4e+`peH1J-$_s`dY@=ul+*~=U*Lf67K2T|J6~6e6I!!O8}*?-CrI0_t5H34p5y%9rw*&zs6F)Q=m!)Z~gaa`{$$OPtLr; zu$)QmuV0zoQ^bsaIrzVDF5*vC(QBxe@#pXU@9*26To@hAH}XHZ!hh=@|M}ss=qE=R z-P`{DuV0xTCE$+L6yAycW%G%od|HA~#`t^xbr+8CB#B#^V;cTf=c*WqBn94_J(z_$a|&H_P~^XcRAn4Hx{^uZ#|`n2zYI|Hagydy1?SRQ>6J=>A%c3TK0-rxVbK|CELLB_bYSRO7ne=IrcS#IfX_PXj@hJTFq zHG?mgL%@59i6FYNlN}yKAWBSK7tSm2kU(Ct(d!bN61!h%9dGv560pObForQQzomw& zgNQq41kcUh)R^{T!PZpm|1rFF%B+;Y+-}_R>h#LGwJ7@>2{M`l@xoOKK(8WfaB%%d z@%+HYNegzdU0|+jSI^2z1eOPcSPWhrEqB%6$b$>V00iz9S>9b(u*t(8g315MzP;&$ zSCX~T`#}x-!wP!(x?&`(2pjxm85_fjOstJuS$BYei1*%iFuFx=D69>oxO_V5B+Nwq zpXYHTG_==UlVtVbo!Zq;^1(b{6G<7VN$9P;vLQJkzvB|F1S=i^0yb11_ORWz`OkAG z>IfE-tURyj+tHt7hp&DACu1QAL7%BR7QH!`2D>;pVT+@|t&d=~YOVOo)G5gfIdo^f z_@%c7Kq$o}#DF~rZ1m6{e*HhVe}Z^el8dFg{_YMI%g~pI6dy8v<*CYDry{g^9V!4s zhd^terya0V8LOF@neBwUXzonH1eX^_Cux=6j;kYS;%{G^-o{s0e z!w)*mW(oWfr81us*F74&BC5|vJ?WnDx0;o402M<3D3hw^D;LFL5{eck2!|x}7i%>* zsMBigv4~QuySQxa+ikRZ``U_^k=dSh`4fU!*C_!Jc4I%*LgHUm$?RTK#GV&HpO^0H zQarAb>iHOo_cj0u$o{^RTQPyq43@QS)*EImYOk)gaeC zuit3kS6e<2C)H7D1HCcWwgx~(oLx?HFN`8x-VU(J0?dz(czvH$*!{SS`{Rq!1b9}x zditNF-nUcis#9zrW@*OP%mm)b1=D?K{C5-qZ(PT>G-0g*<14U_Y|GFeiBX?i6# zC1rFQ^{Mzt1|Q~;AaY!WTllTECMnf%=%S(wr+RULy@Qma0o&*4wkoS)(jFqW&xJCE z$mh5v1}R&)wwrxCk;P@iVn!cqbuU|j{mvpi~HDR3|FLWN&t4HRUlP3;;w)+9LSqJI+wi;iu zvIKJVlNa9NvjzE|pj2mBayK?CUd^n#?_Z1 zxBZ|aanG$3zHGandiDTM{MMG?^r0IFGX}{EltLQvN>FJ)p5)D4a4=V(IxL5{> zBpvztRR!2ojhJu65PT;uA{-8ish6oDh_m0`N&5I~wRg&?8)ZQ5G5i@;gMN1?0hvFt)=P}O3@%W6#xNSt*23T} z@a}FiK}USi)ybus@DNG`W8E;Iaqk@#<7K0U(pRmZq+Ed8MvX?gZ&JCDWG{dwFu6`kA|^TRl)WzIDkB71E7Sx9$k=`B~@pB-5+m|8VNbh_#n6{N^ zK5|rHX*v?RQy?_k>901+Ki)1s8s6Cr8}9iyu7jf&*MGs-jzIBMo_s#%dZKKCG_I&yHv$O0Tn`yoo-<7Gk znT=P)6D>W@_MU!@ApFAt#QFB8V-{W@mESBtZ;qB!_CY`C%*Mokbg!q+qdM{!l!j-Y zpvFn~c>|g-8xKoOZn;JwTW34ukB87`!P8pO#OMUeL5qkYV!*CIMBjHe^P%*H_?bTC8J+Ap4ewTc6nD3TOy%;-x=7vTi6`!6@z=h(Q)?Uaz z_|_V{?#*;s*RJsbKIZh+?r*R`l$u6h+wCrcMWEoHG0Llf{NV!vaz)ms(el&*omwG= zmcpC?XCwWO0w|dE8AEM|@aZ1a%8Y5Prz3Kv=Qe%P#yVygs<0J-xq#7iY{7gSw2SAe z=WBS2P-RuS%YJjStz-YvTIj}?{wSz69?`~KTC0J@c}L9S6B0J*i=Amj%OXu$vN<5V zx`J>Q)ptccvJT58?-fU*qaGn!+_aT17vjq#ps_l)=fklc8^T$kEyMJ3v0<=D!&!w5eQ1Pr3co&8WGJ)VK-w7e z?>$^ADf!9dv8G{98YAY_V>tKg?oE+=zh%Jy)Tqn->3t@><_4jlXpPn;yKR}9lMb}W zvXyl1WNe@dS|MzAnN_!)DDH=40bjV?VE}^{$70kcg4>pI3xM!qQC5|5RjFvsSyQ5U>^&$mdNM{0KTih#BEN?{S|jz^_jFQNNq6L(mFgO z_g}hr7;FT3E3ALb$9ti!bMi(yFzu3o;JnT z@;yNFpf51WVAA7s-JVzZVy*oQ5{iR%)5>KaR&&9FTn-)R4;J{<{v6#H^* zX~XT_pH^jzZ@QWbK5Km>9v zVE(o;pD4LiVYA*~PaqNZmi}E#m&&>OS;7#Qg7voeTc2KV|sLr~d^{%!dS&FKcKc?3(jdQzo zQ?v3Vqp#`aw|0;p`(ZV~2U=Lwnw)*ruP7if+SBj)D=SfN-QKO@S6nZ<9mCk{O(gUw zTvpFpzPdT*CD5zdt6t7fdvEG@IhO}0o-lxm6)%Y#mh%{T@YnITv-}Qxa^r+O>dJGo z>U-FAweIgDmg8QjuKL_eJ}Y4>c{rr_^zCv|BjK_dg2mj< z{N^&dF&u!+It^Jb2$i&hGX+IEPqC;sJ+7;`mI!7mW;N_rJl%nYfs@7aJoCz@Fvy*6 zMEep?MWFgIC`t|y&$<=i9o(Q*TT6?DKA;?-@!hFu2MCf$#t)DsG^Xn*DVu?+0CYy^QD7N@p(h*S9r4b9=EArL!Rm7&chlL#0K?l z2k&V_3g@?@;@g9+w-$Yn*ZpmW^y+YF2G-Ll)L;+qp0QBj>#3fdP4^3z)!Eb22f;l& zcNM;2^!SZT`18PdtNyGCit7o7Y1g7Cd1ArFl!YVirsa$yyvnu{h2%xj?*BBvPHOseaqIg44_gUoW1aZY@t0tCjAVic%Qr}nv z8*i5a^3%r}Om--7SQ`z3)`<#sd=R6ZAIOEVzMeFCKKe!&@>F4+J=_W}I7}1fKYp#A zH}~n{+~YCc+ODN`>-0cAMrwI{Xs9LMKZIT-wrAO=q~q1Q44J)b05?*i67rEr@M`ef z9gNJcu%Eve$SyMOzd4epC~0@1yFXCd^KPDI6%VnJ3UrXzLHk*K zf6f@Yzp809NU^^HgOOF{pcw;#Ox8Dh+FmG5cIPDATVLRe9CasK-?62mx|E#l6%!IE zxL)EFIwH!7oS5dsq@ILby|)inpUp{pJ%d#OADWta5kL7LAifIW^=~U zN7r4-wXkVWo~N<%rkBph5HZ(X(5C`Mal0VEVxS-aCG`5>V;_Eyo?OSXPEhFb%EK-*a}>)F&Os%fW&s%QnL= zjf!+e@w+;BL7m4Lm679D{ z8=yNAH>rG8=u6{m2u3SDUpWsT_CyCx!%FvX{P=mMRG{dsW_61DW0Pb&jZm3>*Cb2r zhk!Q`upw5VqiW+xtR}I~L1$}P&xjg`oLAZc`S|G?Pl?UCLwEBTZNZC9AwZzdHv;wA z^tv5~y)zc71CTaSU)S$6w#3O7Gfk9=r!dYJ=vPyTVqj&`c=DEZSq*<>4ABCq1B~Qd zgSiTm8h!7$u-9xV%#+MpmN>(AA&fFX9RX;_p_d=&={r_m^QjhpL@b8sg8-?7PJrg| zr6j+3;mxv9^Yt3CH|`MvzIwLEi_<}nPXAf>PoBV4O9!?Rm0$dTLQ?#;{hLpGsmn*_ zODHZ~4rWae<}+~^7I@vWPc0KNGKAFLuX2<8eA(>JadSf>pQR)$TxoPJ zlI1on*Qm?giEAqCZbiXOi-e*bmdvCKbcz%1_K5W+4uf5-P!8#?u2@?X!mV+rcSd*( z1NQ<4eSW~N?PyoAC+M*^U__A(|EIE~&NqAvWGS>yW(l7XFVR5l)&@VY;NnQ=$cmOr z)p@3i(q9uy?w_h}l|}3Kb;@I);n6u^eHe5IY(QLXsHnVtF3Q@b{|QLgRl}5Wr4z%| zqx_Inyk4p|i+i2ewX$fCD;bP8-py(8eeZplXu2Gb&~ez^_KrHdx0=v*E^>aorwTw4-{B%n zwpSLKAW5d=369V6C2mX3k`opqABx;Ot##xZRM_U-N3liLzi^HJ6vw9&$!OM%mK-J9 z)`3ph3#?n7wZjSSRWuRlROoW8F*$L+#p$; znoGr9BWwX*p*AR2_v^pXA|b&^@Q@^66#`MqLl5tJ`-)QRfZzLxv_O5Jpc3YwjE&f) zO2aq)z#Xz}zxcyEkc_gg08#01u2RkdK*>$@`w}(6{S z@f>7iWN=rPO1$_m6ITl;GAHCKzYX@jub;o7ms$z{w@_-UH^`&lb212UMkFwj!(AR{ z&Y5dfPY*A71O)nrB;=dy8izI1qDVbH)nNSMBR`VmCn+lyk8I^9$+f01y^{Zo?5XwWs;*=SS$6%~Ap=w4qb zXQGlGRiL{yyoF5jjIdHlE`y6W>i{&8U!aZv_U-6vGn)JrUn&O3jJg5D^{;!)-k zPWu^L=Hsq}iIuE6UaH!Q$m7|vX8ep$wRv5;mMyUJv5wZ)bavH*)hoP^6JCjWZc%CX!qmXKSGR!j*{cw2dZnxyHe#Wza6<0hvIG2ye1K39LMr$H%d*IYS-Zk| zm8{D7a!*;mH=1(qc4yQ;=(|)>`z&*>#Zb!Z*el4;3CsQSTW?|4qXn0VETBJia2&1b z&Lkoq(7B6juF|X^9$@PX>+g;wbOBr~%nQBo@6y!ys$LR2r$YX)M!Hab8&TwaYJH?zn+H9z!C zU}!ra;?=mxkfI86Lv>63nR!kr?$3`!GR{nJVeD|+Eio0TB++|@CN~DC2M0@K0*3J_ z6I^vgf@xLz?D+|7jVGjf&-R*cZ3APdWoQZrI-egi3L_)TiOIBs^eZQeFJpH=D=tyI z^ICmeudoRj0&#D&kwcjeWl^~v4Hy2CjC!E4axuig8D_hSB zEM0yS(@FpYDx_>ZJryBDpD@jWa<0bwl@0C|o5O2_z?s&w4ny|OVmk5#-IU}Er=w;o+;eq$x;Rt6xQeN+ zphB4KGjyxMV5!j4trt?ctCiPEG)|X^Yrjo^-(qJPKp(%reRB__RFsM<%}UT2hmyYY z(CfAdn04i8b$XG}r;uoTDC`^%%E_{|Mm|JUsJw!8xtkE+y+$@y01}K_^hy@u)t{Y$ zkB9oIUGVfPkIsfaM>sofO#8~c;E%crD(aJm%;_#j9(R(qzWctx zg)QGyJ+jTY8(?-6qA4LTa%URT0>M&>Ob=2qTMIbq&>aH!*aFQaX zp^T^)#;FX8P5(+7D+_!)4P;RKK8Q&!Dywum-@N%s89Wj3_|Q6NU&JLn!5bl*c;{48 zdBykkmajBl#Ag!Bg||I91i`nM9KZZVTSE7?;?6GOBKkb};`$jQy%>7Qs16i4XceE# zYWgS>$53iWKy$S0o>jtKl+$ybQi>{nI>V^5EUP30Z(nq?s;e{X+hjpt$%fx6>Q(Ng zzsDG0o={EMAG>Iyi@G6}z?A-+K#UO{HJQR*bgi33)y)#S4d+!#zqjNB+M|f`#bp&) z{?K)4{M_3JYqV2-z}JR;PgXhaY&@98+{=)B$byoeas0(VgjkSzkuOtAh?v*{MgMV@ zcU<$=wpfHE;UyrY;aq}`#FRwuYV*}}qB6`?#3+tPnrRoUC(IA=H6S92gZ^Svc95Z| z;3N4yGY|G^W8}zHiL*=d9&z6D$NSN3gti*@x-tFhUve`98!ZQ8{bZ16=lGUNIi#I< z+>?5ghHZF9yOk9^qaZX{7BO9vHjH}^@91?g_c40Eb{h4`0!z#+_~N|SFy0L8aWT`& zBGrLXg~V{2O6OB!u}pa@mI>8fmK_lMR+bNg0Tt}hK~*%#&4#_kbH42+)YP%_09sO1 zWpI;kA`bvvoKU8*&uA2;-{KhESlAWy7=XF4SMjk>j0e7z1h;jfS1xj`E|zJ8!L33< z%)P~rbUZJmo%InQwH3Kw3OoHDOH0QC^YB?7{vTj1^YnM|POgjJuMhD*HQ!PEE;XKruB_Bpg%Ce->=#6;`B|wsZVx1l zPtm(Zsx0?}6JmKlE>yOkkJf56m~7%2qYiH)$~)|gkghT}CtEeEyBhSrE%HNqQ(*Ga ze!ob)GeZEzsD&xDd@T)Rm3AR_9a8?G8#0i@+Rrz@W;^gUY-b*H!M8`{4nAJi*+UN~ z%fpOmTd%x$9b&YjYQVx$IHuQyg5lN^K4J)#<>Sl>$6B&vHA2Gs@IF@Lz&=#|LnOcT z%9pn|_peU)sc)-ABZyw4>i}_57>Iz%MEBdGH%m(;Iyx>WKPJi4ek=nD%Pj1WvXcg_ z?obdS0r+xShnSB_O|gTEh(@#e^VuDva=WcB+=~g8l>V7AgB+7(r%9_Fv_vuMy@|Y1 zQnKRNFHkU>xAMQ-UbB2_cLI7|bpg5a$*aWx%l{L+ykDq zlVRw;9Fs&g5w5L#v5B6aNvm;Y!{&@oz>Lzm;|E~BjZ)|XTb(nBCt7~kF6U}Fb=cZn z|2<Ys(meKRyU8HDrT#)W>LEEx#zURo)a_S7jSP4pFFm@NW%(l_cwg`2PATAW1>J zddoG7@V$X0)fI7)!GqbF%kfT$O%Bxt$8ugzm``29(Yye$mfzQ(1@3JWNq*wdvqf!` zK?iiw6%E_imvE~R_=wOLbUCLip|Ay2dV+#zAFAFzAo_SS^zV+pYz7r;QZ=RKU~Hy2 zOBfxG)uoneSIuKHjFJ*{G%fGXYb&czxV|5hxhCRuj3%)EmKXeKt~Tvk;4@?$o)e~4 zkjK`?GM9cwN`2|ncN&N_Tb{o9neZ)gs?34v*{(+35O3JgOphOFV_n+~dr(0$xpQIc zQ_0g4W+cyUQZR>J#fz3s*|?b|`Kl%=eeq^XWqK2KDHeu5rL0nd+bZ-SHI4IDm>CdD zxWvc4EnNV`jJMXz*R9qTD3wt2MHhpjNN|-(R5*EpHvLfCM#>$i&0O3RP*d*n=CBG* z!o1HV;?h-Oc$t18F}J*bX#BK-4n8Sfvk~Z!@?u0o&Wt1j zzyT%%u&A|Z{;K!{tW@Vac0QYYh1sKMEKQC((AF(k5#GI)Vu^uYO;92DV(9gnRl5|Z z{hl;)WOq(j%{E?fZ?u)Ff26jPVk!wd-k;7hWHN%xhoF^T`{_B~k9fde=#9en)alwn zLkemGq(cCH&*y^~N!}4~J&^j^N1SfgMT`IqSAM4BeZ!kGhs)EV*BdSlC+27oI*Xby zNKfXU{xG&ScA!%gGyM|FCW-`U6>p6#6HV=>se0Vv(d=3k0c(2mF_RhGJps>fO4_co z@fu#{t>&AaV+YoxOdY>SffM~Nx3Gs?RLE;d!Hz_Syj0VA<=-6D!#(GbZQIXKZio@V zcykD>N!x)z@Myi_@F%vp)K%@l?#&lO@0=dUs_;HM8~Dc8Mn|>~KF`Z<&8QN0FSxLe z9z!~k@J2;w(t&`O2j>L=i^L0EJ(l72n{jd;+$4`-=jAvu)Tz}Tll0V=5%DexMODc& z`5<2tTplZCC}<+Xf!+X!>#8dX(TF6Ea*@K!Ez}%OrTPxMz7JcYZicpt^l4?A@vP_Z zGRfS>bLW)i9y0g+pxYB9pR(tvauLi>P*@zyR9~uL*Q&R|%tx9Poi0;aSq)-P7QgQ2 zDWY$Gn$qOaIjTQLHPc%X^gkRshIRGwyakw>sm+hfs}%9g{w|)5#|^!!9unuf4SWA3 zeRoaF$qVp$DwjXiTsuAYLdE_?oxRIagA-$}S1U_{c9W}X9AtvI8Tr-Q99U1gO(5ds z(G`rszm?C*FUqU__~m@m!U8LL!S%xyD4Hx}AK!5{+GzZ5S%8V~*?Po}@XL znrzDL-li*ew^OjjFEpXi5=)YgwUeOQZXBajJ3@Gk8#DCgfZX=*uA>Y=CRd9$&&j^?Cq5ga$y)NQHVSBT8)zEwXRZ6T7ZE>kbenB@MQvX%kHAs@&my**-Cun_#h{< z1TJ0t!r2$lMGE(-N|<90cJHf9X{o>Y&2;r~Y|o;s`s}^~1WOMA=y6#fv=s4w*n97A zxcatjG!dQXC89^~LG(@p(FsN{S`dWMdx<^>g6J)Th~9f=5G@kDk6y!!UPjp~_x(Qa z^X+4Q@A2H4Iw5nQ@kdr@(zthm7FZSkj6z=AP@8Q>6GTN&yFzVgRs$Be~pLomLDBg8&_UF zuXT)uYq+D226y)k71{5vYu}r1c(?Oi;((4{Ppj&)Q=-{4Ib1$dU3(@M}z5rQcZpxp9N^%X{_B$Gtv_ z0dqK%0b;Y0BK~SVfCcub)Pt&$Z*Eo(OJ0Cb<>23dgb0f`0% zTXGAO7}1}!yNS_t)$A%+)o>L`bwTlZ367VHt=sF0$O-Q(=`~^#hSVfZBSUgAEe&X> z1SHFD+M#%LSDv}{SoXaiGL_%SY1@jl++qazgeY>}efw)vO$Hg1d|OabzdedX1Zi}x z2+VGkrlcaFI9aR>TID}snlht$*a3iMy`KAo3Q~T4x>ix*zP%1}339(7Cq|%c;(>pK zQO=rI0Odz_B2py=G6zu0aDB`UyUAh2>tjxjLYOu*Up%dqaoFNXaniZS*z4R0OU>c{ ziUqkZ^czCvXx|gZj>$-_K6+tZs#n!(Zo8%-WCJQP2Jpc$%ZK07UrUb5spIy#lwb%X z=Q;xX0CmImCz80``M~Y0q$X_AV6EcyS#?W+s!}*YHe&r*RHub14fmf(w!U(WKKg( zClDKB%}eP>UVMaY>cr)nf+KB|-R2n(uzgSr2Y&*#zocUhfLVHBjW1=FT2lJlBSJ)d zjbL}}@QNtoX4_xO!p?m&tOb$FEV=v-NL znqO*_A9z8P@1r7PF2<)quX6&fieBT}zY*-(7eTkoFV-t5ni=!)Z|yz@XpFrlJAB_P z=IR~T6gRt@l}~XFsrw_V%7Agt#}oO(Sai=mKr)A=fvEcHw;M7d|m`fJ||s9K!v zP8PR1E;g4O9rlp5Bbv78IEK#&<5D6#c{^n9>=v)6tR@5C{!Q!oK_53deh50RKKK6N zJ5i;XEy)t(RDP_8YGocNcoMiN@T>arbp@)unTYMsC{AJ5Pet#f?@;?<%3xrNEevsd zkN-WCQVO4wq5kQDWhxGZZsk~yDzkN?sr#hfrky6-RZ0u=Fx{{$bR>wDr+XVMYidQ2 zJj99U7oeBGyg8ayETV!eQj{-sB57wlVd!HjeOiXY?|w*vrTRO@MZeJ`_&eTx8j5;< zUr!txXJ7tj)8?{|9`tyq;PpLXki7V{gsj7f3R;vNvGV9{*hzR8(u+hh`+6x*Piyf; zp)`nG60_&Tn;}Qa1x1o=4Hlbbs3D1tNN*L;*p47 zNuPCcxsv|-4jEJV(*mr!C;NU$c#h3t(h^C9^ML=+M}E>>!`wT}U;Cl)Y3K* z+PMx0g><+R^_vkebf7ltD?gXS5$0^uYz^ybxz`fa(DA>sRA+y%f2g*zO3aTx5~{)S zoD`pKz6tUiY~SK@JR-tDjFXt6RoJh*_Xrl``BK9jRx9p78o5H7ZydH;>;?NB3Zhp% z(g`%bskh#EGHr3{E&E}vHXzIaEEp~s>c;l9o?3!?g7ybc74vPaRkk248qve-wLste zPo%{Uw|ccrJoiKKIxMrE* z-~k9cNw|v)Xl9CJ#wXjOqYXRTV?J$YmDuE|Gtpr|e73{9j(2CLC+Pi?1L1`Eb^2B$ z#V3l^?JdJLFCZB%ymvyaUZSdp3=C8|Tu{FV$0*?#OSkg6Ihf$hEqg)#c;KL^U`T{5 z^>r8|i*2q@SB84t)*8EK<;`UU!W}4Ni0(rRVnc6*uNFOUo@fm8b8>jDk#C^mt9>3D zc!y=CjX&hsf}V1V%tPNZZIn1q|5?~2B-`yVvjmB%^LI&aCEXXzEYBDi7PkeZh&dm* z0aoo~r{i-CQ{J^t!y`*lJMCDHY4*9h-IuiBHJQ~Hkg;#&?@`uSZd6lSSn?)aMvoH4 z^Jz`O+YgYXVNYNB=!DLmf7minNkMFk`!&e5x~e$}kzO@zBc9Bf5ULTp@26>1W$J6y zHJ#xe-#ho3Gc)lZI-CtL&n_3@8HtR%XTVl$2yFmV14QpMHZc2BZFbY zRn7cOl-XZ@v$Cb49C_G$%i)bIK_Z_ReoU4&ObyU%@?Y%Je=haf%gXtae@>k*uAfL- zf$;+Vo;h}6tvrube!#tzd1jya{9TEvE02#7>7eiV0h2=Jl1)dAXiC6gF9*2@H5#G; z9zd;)TN(*;Q8@-vJn5JOMLKd% z1Vj6_5eMEMM{h87!VWOODe~S7gp#HQ-d@`Ls$}+ncK-bvUs*f(xQHdPA#vPKv+fJZgju zZI$dbu#0rhbaE18Jdgmkobd$q^!HyCU;ds~x$qzrKgBDU?3EZC$9dq-rr}p^jGgZI z;xB-~(UEF}mnKACxngK$xAI64H!6s8cq1+gSyqabKSQ{ui~Xk3O@tI36Z_%ja#;p4 zM$QR8yvIt!*i(W9JHVTcv$&LJdCX%F_E89=#eL#DQDkD$>E5m@&b&q<4d!^$h6$|x zdrAToMeYopMcgeLNjb$_L{FB%V#-ZR0?M|yQbrANO7s}=Tfw{|2^ ziwxiX!U@Pfy5HA#*gUP5z|=*4@PMb!_qZSuHDe>wT*3JL;D(y*BPPE+dwK{h>~Ocs z6lC0S_k-Y=>`8kHuaRQ!^DydR?VdKHT1B!afy?O@IzL|Q^m(CT7jd3DK}m@oxm{qj zwasqIf2Lz%W9ocKNUHaOfco2l>u#$@4v^w?V`G94K=BmRyz|LvfpCkfXkp9}NfWaD zOQ-adlGmjy>qd)u)vro%@jX~CJqH;LzIk}J@s4NxSL~_{^jhxBCx$LT zi=s|eS}IDqyxU^RTl81Z_HJAR_A*YXD$c19QeW8iyTUkF8PhD|(i^uew{?2$zG=k)5OWePQ=pa@nkafgx=poaIm>xny(CTu$0y;`Vw+4{meb^Ys_ zZAX)h>P$iPQpzQc=LV4|k&F4&g!76XQU_g*hE)40V=qTnqH?4T`J^3yKQH+$cd<^b z7nsO%?+MvYm#9Qzo$b4qeObnE{FPm>>Fl#(!_<-D^Z8F8E8Ba@s!w}$F@i35FtYZG z^d7@UIq2k{5O$Nwq~y`Yxmsh8(`E5Xb?B4TW%>Rd(^S*VdrC4jH^tzV-`!)VAUK&l zicD2vf1^2VIFmI|JwsZlccbqZHO_>X&lWJ==_!7^qWzR=QtgVS3jZK)MR<;bv6W0- za$|srExAlorb;cv+&>`oym^t?&{f9Af+E>Gp^5j;D~@}QsWhoe&q-l7)Xy_Y;`;Y>@!*c$ zC*o?!FSfZ{<)5&IPz2Tn4+bCNelQYcX+$VB3Y^5Bw^1I|2ekq@n<0ewD}J`h?}T>Y zfUA*!U6cRL)(-s#59XYx{w9-pqeWUBmMtdtRQvWuZH(>e`+PHM9~Q1Zzx+b{L^&E# zV|;T{QqlftC1ok;@SADk3U9|$v|^3*NPc1O*b4jtEgqEx2<193??W2LGlLjS^W}yRFGJydcFX4Hf)Q3BY7k%D6`Eh`as=qL8+RetXS+1eENTlKX=tdsox3}g*(s6q6 z4kHX>Xqs#QL#P-;57Fp8ie&MaDcDs@r|0*ZVrm$JO+L;5-Ma3SDRRGs3cewRv~D zoze{>YcJ!A*}UbarmY!^9}cSc{3sPwjdEo3V=@)t$bQn!fF2>@2_Ya%-j{JwepPC0 zM2=xk54vFhSRs!@!}T5+^RyVcev@CPZ!C|%^BqqzBjO%jdtVn`QV=d_-8iH8Smw<8 z@J;W}VVQ=Ydr7lv^sd#ItHnKG6&iEoOx)DwE8D={2KgA=)uj-Tgt7!Jb;iV{1rxP9 z!{M%`;EldX$Im|(=q=JV11h1ffkU?x=w&QodE^I|h4wNm;RL&h4E8@yMY;&q(F%{B9;dHn zWw;kNsn;T!*8oV;fjlqsEC`2v54^~Jrqa;y&5}(cTZc5P^m2{3eKEExCeaWMG(j6| zp2jZjWdGu=A*O%Kue+EKi{+CWhG00rgXNTgTjJA=3r2~b=(qb34uA&6ZssiG=9;r# zfO)2j->%H#3x(iFr*4NlvmI0uA|{ciZB=h%f$@j?Ivj(w9rrf`u|~74)pv&)7Z@pB zJT^YZ48(JF4DPY|8SNFA7=kD?TV4S~$e^d39;CCSs9)x9GD8twsM(pg6rbe@)?M-6 zo3si!L&*3wjL=8sb z-cJYAjkhH5!uJZLm0+as7JFB(JY3dmX`N;5$++;(i(icT(G@M0NfBQWft7r%1pc{x z)j1gFn{g>z4%F@JnX?Ikzr)0 z{q|H&82RWq{O<{PbAa< z8o}A$u ztP$lAM?W6zV+e!175OvImON;*G1EBVUg~9 z8&Go3>P@1iw04Qw3N?T4%C3}U-v^893!Kfb+8r%8aG*Q%sj5pM(Va~#l^x|KGM7C4 zMah%glD1#e?$!5?y<{~21&Y1ab?b-A?-Nz>9RHu6yX<5D)*C6{OqlCJO{nT5lL#V} z=xZSVpm$wgu1Sz55}VfDa5~s~nbGaZEJ`~>#%GFXBaiBcqItcH{}An6Ldd<87mn`ZP`2KMfk1dc0Z zIQmJ1QlFH5EJ&1C)M7FEWh}HBcl;JRE3>XGqfY3EKubv5d2ZymJ^->%T6XfGaxb|5f?&11vuXKK z!0tCQxjvddhhQ#zN_nZ}tw9;1Q7d+~wV?+bfSdF?s=;pC6AI)u%@}k5e~sVuEZU>R z+1SIX+0EjvU1Y+i7UJ|UW_=^g6@G(}l;L%3809eH@iZN|UuZHm4OgX|d;-k^e#nVA zKGh-2oP&XWbh+)v{#5bulsy`h0IyI_H4|>~l%GwOtK;N@)&faF2 zp%y?UX;7-EoWAibQ9gjF!1vzfEOT`7n6H9o`%8+rH}efEZgxNa+58%!H_ODotv#LR z5&T-4(T{dg@Krshg!?z~_`F(bCK9wmVppBh&NqDO2w$@earXwG`OkkNJ+A@BXE+80 zgwAFi^>9jIyVu7HR!#6c9TDP)swxk`deCZ#?p_YwD{B{?T60D@?29l}0P9%uuXQ|! z`ZVvpEdEDyLRlT1AOKZ=aTpi+#GV^}0o{Qoz_;pb%G+|b0p6;b($1}4jvpfhz%lX* z9`g>X0FCt{Z>0pg=gjtAGP|~Xtl&DiVa*b8Y?VnzfD}0^J5WmG;Pr)SW{NWMj1Ufi zwowH(qnqa7{GWbE&W#VRR%P zh8yXzFV9$ELEb)HYRCfyNX70@BEf_H1U>o4&jm{@{Q{U_C^+-nv526)gRq$xPPJ*?# z&AemplY?%iim4vIca{8jr{B3@4oo3@Gu3!-|JPup5qvd`z^)oem&NV@S-D5|{%I%a|fN>V8EJT!` zu*>xL(V;YfdZ?hyQS(&>mwJGrDE>DfS1tIi3|0-wn%-rTH1}1cKQ41bpO@ONW&6&rKr-FKWmG?^%CWjz~P5rr4OD<{-)luUC!HY*jl}b zB~Q;ZasbMW!@Q@S1>5sI&!q*5*d8hGT{V8YVdhgBg;LuRCv#8y6tGKVb?Vff%O(ed zsmmSF>Fxv>Fp@kGP3mVsGTU>hTD)-x+)DE-{x{Bi-384?!APw8KRNwS`!<$2;N76{ zw(>PTxO4x$#r)>aP|m_*S!ImIvheSo>z(tv;!G7r5@yNd704I?w^Uj$zMz(O6*vbc zqu)UR(_b~|S00N^oelWhUA8%bQ!!ZlEE70kw#Fib`BG+#>m>4gJL|B4-{+Q3xL1La zS@1;*hiJ>G)JSzO04Fk7ta8i|j59ryjF5X(m+{n&F5;4wu5vkk7Z@C;G}_1)tYuLK zRx+z_sR?VDP5INBmVmgzhf^4s&H%u-xeEl5iEa^j?@|-m$Dz-T327vChSMmIJgh{h ze)-YEd}yNvxqA*!CZ0>g0WUC||Ad|ZX@+b+$s6ky*T{yY-mwNJiH=Wa(pLScLx1Uf z)DepU3E$xcU(S`dL!l_VisJDbKLh~NzWnl8a8TB4IJGJ^qS~U@rRv?RKwZF0<%-T$ zqe;_6cnO3Vc2}-bP<&N>*NIER|1G5cVrs)kW;9cb2Y^s`;qjv!Yl0V5lts2VPw zi}vMFeIc|;dte4dy=|7~HGJKtoKm%{e!k&zx5>7kxasrx<71?7Z|i3e124^LPnuN* zD(-C2fq`MJe(IF^&4M-lD<)gdEhjh@A!dPU5|N-OwtMyP{!+k}l=krstDSbrkNxL+ z`i$Cr!y&--=1Go2qzaeE>dzWz&npiW&RUv_JrT&KF?H3bBwQIq6{yC{i&@O zqz_5&di4okUz{0XIYa;C9b$jy9X2;HzK0F)A4N06)@lm&(lnD4&6Rv2Obrp?AB~@U< zgcN-o#_cxxj4O_>iZCpo7$wPkP7C;n-ph709AMx!XgM(!p(@ev;SB1G69ps=DOm}Y z?zD)7A}9We8Ee--JF5^dD;&LVF))^-J$fHyJCk2Gb9t4GK6?MS$ujF+$i5o@OJ&^L z^x|1Z1!ElTTF}b;3WOcgllmJ8LztUA9~z=8LVyHCAE*M{kGr|qSE#pIl==W5d6akhqy+7>HW#1tVe4sy{XP)1X42)-{a-`Ls z@Urp6TZ~W;R46OnTp5#qiesuA@8M&MZUWW=FQ8@CW!)sV&>>gbw8<~#AIIwVrQ&t# zD@F2d3{nw(;{JEKD^z&=U*m~hU?oyor~~N6v`@y&9|GHtGRTTY06-AnoD%Q4{!5x= z4(A$)YFPa_>*;6dRC9HvK+Z6R z+{WBF^Q#`Lu^ngS_jF`$)de`p1tRG*#PjKSm6NNen{#+#RkOmHK-ud$z{4X2Ix|&U zNhjs=;dhRW&Yn5W&5|#I)7}-jQ29(mH%Gdy7eFvRIo)*QjWe~LHUsk=+kc=~Tf;1+ zli#!B+&Wa@a*z5inbNw?2NKy7&kquPhY$Q2K3`F`C+*1~g0CU+R3k=MIBhTV_t4I0-&kUfJ(FA4&o zREUVvb1W(Aj%>0Trx@itFY=iBB2JI_#+>%Un>rd#AdDCp9Px7`XI~gc3fk-Q;u&GLlpxW3PiCC!L|Jg)qPYx@*q`zSE&#M-qp&a$?h4G4ecrWZddg!;?kaLwpyej@ zJ+v!Yea`?J=3A#);Sgx@%wb8Rton5UW|w?4JJCVC$`CiGc$THHTcNt|z9%(yC0D-# z6|D3+j81onB71K_zo|GJgAaB$K>8wzjgBwT%+>$S;yrC5-tXrO?+EW>2WMw(MEwQP z?#Kxlckoc_<1`M}sZ8m)7792*qu<2p9+({UJ&@4Zm&RuNo$+tuih=r528!+Hd}rlUfTMh!zJD4|j0k zDCjw5tyJaH&}bXUXfRlk%|f0Ta}|$*r{qVG=jhh+@xIec4=$VKz>LpXveuvvJuWl@ z>dWuOTFYkY{Zj@hD}c_oi9m0M%9;`s7*Ejax4d~ES85XRvook}L6+0 z`Obv$SKzO`pKOawJkOqfX%Lh-6)HZAS@aywMkW&Yjf-gBP40lWe&GjI~YM3()UdKs3h z60DsWDtOazNxU4^YScF6Q5OUywjO-$Yu|E61zBsYL5MWWamspdfc~88JnCSG z<*|rk>GOBALvtz@0M)3@O^ogcs#^+j)-HRk=X^ZUz|QD5T|V6lbb2NZ%@zemCBL-q z$v6Vt_UZG3#gb}|-!m#Wok#)Pp*^6-!)dVFOrv|zCje4HEHo*U@#>eRvg;MsHm=1= zz;16)y;r9d#hRslK<8=Dy>XEJ?%iy);c&K;(G9oht+RHnJjNv3IGV`dMDjF%y{61d z7l{JVOl1e`-}1qk?(sAysPeOmYdOZbS13XRzQASqDJ~_EV_d~#M~9_R1hOPAdA0rJ z%4zMY^*VtI#*}uQL_8fE6LZ#n8Cy88+{dc2$_22FcprozLK!rNQjWUs+CGhD2^q?M z3l<7pj4oJu-SI(zJ&0Oqae4dHt7vzoDFsyH(e|3`6H@ARUvuU&mS@P>dUrvN{JO~& z{J`{NMNsNkD1zZpsJWP*GiiGj4tGN*)q}c6sdu|26-h1j_&(_a>KAaurlK1=s3GI0+a zZTKr2g6;1wI55aU=!Xg^R*8G)k{K&$LVPVipFuX~xyHw@{L$z+m-T(_?d!GbDD91Ul%@FB%mj_zt+ynNU1c!aTO;C?^}Bx$~;oo&3Otc8=kK$&hgJf+q=*CCisy&l~;@i`!1GKLql=t`JU#OM>>e}N`u&YF)Nub6V&QHj5 zyI-Zb+rN0$gNKI;7ctkU+EtqKe`Jc~CUJnrh>jokF9M?pdDf?#D^EVzBy+k?z@9dm zgkyc85onFoNHq?T#zM!>S|xA4saH*904d|%G@k$T48o$^%cobP_y%rqp0R^-`q|$e zp4D`9s!BQ8k@5%O(m5kzofhk+qaIf9sxH?*9_C-UpQw!-235bU-yOBRbsVM5!pZ@t z2=8P72C-25L-_G}yXJ7FbeH}&7&@O*mLMcPojLj^N2=(Xum4LvrMyDzT*z-bvYS9N zOl-%VLv znRE_JiXB-daF@1giy$>>*XrC?ulC9xQWo^cn+Gknpe0)#vFx5yRF@VD7|*>Akg5a~ z{1ppN=Vg$(i0c!cg&%xkt(rUbBbtB9k0Q{6ojhaPyd&2fq&T5Vlf_ajNGdm<{kk%X zsZzsT|IV`C2`u#F>gmrSAHjp)1|XM#IeVgZVifh7fMOIGF)-e&JM}|p#45-zs>b7V}Vc6x5kHA8ULjE)d9dJxx4s*2xuCOM}EOEW87G9C; z(ddm$$!tq?yLqh(g6$}+i+Zdg%SO3Q`F8^NX8Q?M3z-Y>v^+|SXCmJWj2PO_W$v!I z20U@zud$5+1ET%Rlj+J=J>3tt7%qHh@b zXtfm`xlnd7cN7H&Gmj@w{CM&48b6UBKqZm)=26;UL%XbTqr#}L&JT~Qo>SU7JRD-2 z*ou>0;-lbw&H40Rc<8AvYq;^ma>Kmry{Z0mvm2c{AOGGnVll>LTB>RCs{V+FJN8Co zm|ApR`Ki@2VU6QW3e;A%SPlJ7EFCZUPQ#O6EGhZgs{sI>Nr0JrQTbgdSV){znv9n$ zYO#GQrU8B7GWWq@Pn--V;eGuTBTeJZwuKsB8*OkOYs8~qS2Qfx=WwO_*e1#zA%@NS z!AZc#$Io!{PE8}8mO3`}aR{v&^|j)qcrTZqf@qO;NxY~NvXT=wD{9Bq@+2H*$3*uy;_XIjWybTdJa2DJMCpop3?Ks#4 zE7HQ&BY@fyy28}YWv>Y;MSM&wSpNF>UwG5J_jc)@BCu-rDK^mr<{5$!mvXp*J8hZ5 zZ92Ox59lWs^Gu#=0!Xy;foT8k5tW;?)X^qtjs4%=ZA z?apce2Y9t<7`Q3gjvgyMo)4>U3bSzqShL3jeP^4zWy%gAQ91KWM*LzvzwIkHjTxV^{^l_GqSl#gRX>t@Lq+!;QBfa zh&qE~%BEQl*h>~$pN>zHpAQr&{U8)g+X`wIx)Mnhty-(A+bph>@N{avJ{p=hr4f5) zHly0}V;WABN`!TOs%*VrZ}O3vP3^k#c;`D?e?Ix}!$EbgCV_%i9gqDbycmPDBIN{Y zsS^7L<5J&k)tS=PBFmbV{HL?Vj)w?~c^*sneEW8^PI^^K>Zc!2T)ykO@_L!BvjzW3 z>}|gmCF{_DC^lO-V5sifpa`wQ!L8wTvJ2sv?Uwn&)93~9A${-iXgVCqr70RT`MO2w zr*rF(i@O}S4zg@whVLfR4JMbLbxTiH80m&-$;&*xxz`x*{DOrUy7X{&QU`uY-?6n@ zw`36*aD_D^z2vi12qE6TPPjsmHS+(hZVY>|JlYXRiSTces*j;@={fq-)IXBiey-}io|Cj)J)v>>|K@3sG|^OG|70OW<}`g zAB1Y=7g2Xvx5$w{mlW&bUFRRbwJDW`FsXIuQ#l!Ej7lDcZa zZOu5a9j{ObnWX1|ah2`(i@Lbg-soukvL`&7JrZ#%Lp=?j$nHFC-!3qRkEH66d`8l0 z34KO75RxR%?n|6uSHKtqLSGc3?{y20dT8^9xYOak>8>Vy@J6}vYPRuGXd`{$DtK48 za;=_VL*C6{E-x!H@K9(-IkQ+(OdlDIoL@k~P{_lCq)nvHrq9$$jRC3?BUqG@ z+ViYsaq27_o+-1+w{z?KXa{@0b+owEM0{olF(=2r<1kbXj^Tt%yV__ZJA=!oqm3&( zMRh-zF4ybgiO;wVCTB6;tsW~kkBMzXYi3%la+t>0{apwX2+HQP%M>7Dk@XP%hEf-fAzbA>cXE-%LQOYrgvn+H+~70O>{Q zR*XA~Dw$rDd)ZXlFVWBYFEVg~Xg4%+#D!P0U5lBnyqk62)TEhOgra(q%#-~C+O#dgvXLpIt`|ro9_hp)(aU8cbG34w>aw7EnZZO zMze8+b6;lS9=V!!=Cwp;N}}gd;Y?HV_x(KBMJJ@!WsA|z>i~yqLV2y;e5dr=4L$RM z`&N{syl5NVA15$mMjN^asU8UfSpsS6SxuEL-zmfE4D7?2$jtTdW4UhF3|IMuj8K&<&&R;b z((AUPGcF)Ybjra`BrScq!tk2Yp#IZQtAz%^_&n@NtpVsURAf+Ezy92Nv!-+ro&wF5 z5|0!Xc!PFfzDgA8q;~xi$APW26NDoB1r+2uK<{eueB3IiVYcwaEZ>i;^v~*1Q|A*K zioOP8m#qPL{+v3ihfW#M9{c-kMkXd(Osp=8S0hYU;@DYkFU~sL@9+^qZ^xGp*^v#nleA)qB-6cz~)&!s=9 z$$S>)F22&{|IN=v(dZrEy9-q1lcsD-x9E0$n0*D#+B(r_;HSiO=S;vjHNbezp337=i!St_)Ubwg z&$UHTW5`LZP&4q$$)iRGECoJpqy}NJUeOQyw$|Cv}E|)W?v8R~!g`ONh&+lBv z%5V2ayygsQ5TM9@fS9=%HNFP@WXODI4|@}Ey0 ze90-d(uXE<+Qt5&morDoGpM*`(*=Kcy2416;6PZ}Z1TX00njEG@@zbg5PC*p+}s`G zQ2XNnrs~S>V@DPDZeeKEOqr@L|C^adRyU;i$e0_m?g4i1Qq`2Qs7Kr*>8HZ{*6T!v zNQHB*lnM`E|LaFj2HJOALr>h&-MZg%TPSqfk@8x`{|<$4_b+e#-0BZC*|+#e?8wLu z-LoR1vVs|5eR+7B_l{Wx_^=XnO#N9Hm8)JM&5&C+a|{5Rwh%W%>$FOunds9Y;f;#aN;#;X}5PS-;UU5E92EKtsD0qk3c2kOv; z4hS}Vm$dN|U-l)=UnOdu)DR8fQG739W%92?SWe=Y0{o0{yb0PRn#Vs9PlV*Qo#fUlw`zr1$KF1kFisLdSw`e8pcAS(wq}3O*H6 zW3vxb=4(K?k(w-DT_vlDL=U^<>0G)LC}Qj!KJb^}!=czi${!Vf&~}U>o~eHuA3XSy z+Fz_8bc?q%hEO^BjbsQi(+k-piKbuRb+7?&w@7BLvEefvfD+97S~jc(!-9LaP-7YE z{6}Se?~N@$H>=1As9!Ru^;s@Zf1X1$l|Vmu22-ni=de&2n~pS_RysUdkn@(qhF>)j zp_!kxh7FP|ii@F@e`55ROi9l6=#Lh#@c(KQ2KltI)$DCf0bJi(mU0N8En{pzy2kbc zh6N3j2af|*TB}ocexLRiw#X5qhA_68Jil7<$pV{jjWLDzvmam1qc#7q8GueObuPtE zYbNi9>LJH8GKUy?TPNtm@8a62vXzj*4`_kMn1!|{-Aqd5Ui(FykC*DTvgkbzsx$bc(`9N+Wk>@ zBj$gamqHXs0IH?C`yER2=E~r;?QiszyH&7YZeWUW?+IWdg zbQ|UbCC$5N8>WTKT>sGJZK19au5248 z6oD!%KS9fexcL5-M+n)`;jl$%u`KQJAbs_0M(jB-c0X`TVcERW^6SYhZ->J6?btrDe&^!XM65*r&XB+4=9 z0QYd>d+-)bmaLtftL}Kyi;kmVx&LxA*~0O=L)kiC{Px|VL{j503!Nli%fIh#z))@lGzt1uW9R?vq|({Ta&K?;khE~% z(Xgu?b`)4je7^ZS=c@Z1zr_T)du31f87Ux_)z*mhqYd@s9w8;sZ()a7R@M)+tenc8 zXH)2uH*CSIgB^LGN^~nQp=8t38S{E9&2G5*-unE;Q73~<47S>n3`U`Xl znvlJ>dd>#-R~|x4I92H`i_wWzn&Q)nzX~gO$=0R`$SkW(l;6}UX9znO+u*4pev3%T z{*vJJ!4b(rW_?sKRxR!JG!8?r{WU^WPVkMH;(swKBoYDzf?wMfTT3vQbhltz`~2no47)5f?;lMS|cp2f!83TPG( zfhQ}V=P6(HwG=>KU1X}?R-&iZA|mQ{zc{|p#s?yFnbE9-@-E z8%%@m90JcG;IO+l3k}g(ye%gr@hi<-d}roPcmKYfK9*uLNV^sCw)G~L&6~ubBKwD+ z#5&ilyTv8O$Iezq3{F7z?ocg511(2mM`yMJ&|8b~`f5e0ByyoK?;?hikBLUsuG^zW zut}q8GRHmU_SIbMb8fNEd;*6oGr z^~n(Z{UnM&{+)_;`TUmEqYJ!cz-#7TCOaQr;xL4tpV7ATCKPaeNBgL3e0PJ1TxUoCmduQ@E$Npqu@W-CF z?#0>5K?P|jydw*>mj?>y8AA=arRbiVIc24W{o56#mFPWD>>YG}_>&LzHBXyp*I-8E zU%zEI{MV7JPSYF&{rwsX>BY)9q*pW0k1gP4Pw5bK7mz4_k@Eo~JAuy&S6T#i(0-k8 zM{3L0qSGwUp`YKeis@`JN3DybVvEuJg3NkW{va;+-(UXsmHHna88m>)yN2FB@;|=g z|6zNWfO#JmuCgYMWdG-%S!D(!95tL4KK%0ovnBfc|Nk@p|CRaw%54@z#^uNMuvHTG z?*__0cnrc*WN_{K5VX%Zk`s7y3Nb6_1%1>%vRwEu`%m5U-lHdo1(7lNu^sPq_WZBT zssHVg_P>3ijE2rz-0Mjq8c6rI^8Ei`!}NdiX_6IS1J-z&sVv<;_D9V3__Me8K3u;1 zr#^rGFa8ory^Gr;KuY*W&X(rCcDDcC;{JyxMG3N-AE+KkKycwfxo_nDy(Rt+@Bi;N zfN}CgS0uiZ9^uAc{_`lzD|vsPN>VvXR0ZURlo!POC!XZ-riLX*MAO`4(=Q*Z|jNJY|L$qLk|# zqLisPQF=pzI(Tlh^bXNTjCk#8b zQzj7ltRs*EMsKQ`i>g>2v{C{7yMee)mNBo>C?i;ht)Qy*b=11&^}sUp4=fq~;}$uu zSaniQC(Yd|yORg~$6ZGDH~K9we1E)0yfY@bvyEAmgWm9GB*hIa#f0a! zZI^BbH>(+}RSF#U^D$08n|S7(U1I8R4~UUc9Opq?ds|D_9@e|dldfXy8JwY~&Iu zaV`Eh71*wM1%d)A1{4n^-rEC<;vSSe7&s3Mq)n-TcQ*$4Dv7M|=LwCyg`uv>#`4@! zO+?BE?qW^!fi6XCk{W;hnEHDMt5p-3YIVtam8v&jM821vmm2fOMszPY8?R~dUp~f> z^7jr#0{u)WXiE#ugW_Ze=nwpxTq8P;@o+xMzGPzH>8S9BI*$5-^>N|S_jgBhYmhfL z=Q7ag`nGbd^|C zxonXol0}~rO2Bi3>8lzIOSN?yPwS7p6o6&{emCE)YQ<-m2mStW1+yaK>WAmma6{gP zAZ;&zq%#3cbdEtu{V~t9L$c+Fu4#BCMT=(#)sn-3@p%-DSan38?`81b#iRi|v_`)e zq;U=brJZzSWza%n)*}Z|aHg)}4wUUrkUQgAQs+Gs^Hq z)|ViC+ITj@UjU+qFklf@3-~t#4_%0}|H}gd_ zo`b2dHqgE@%q8M^HEMf?U3qLH*#J%xA+#qVE(+kE>+d@|&a#>YF4o1)x70PoQd4mO z!m{x!lY##J)6@W$7@{~=Yw(9t>AqoJuQF|t0{+CS*DZ9nRoG4hX*I)Tw&WEqGaCP6{uii`5jzd0Y-#$<>`&r(FkR%j)-)O|kf+uJ}n~r}18v`+R{-t)iB^J%NuU3w9H(7g5v7YLy z3KI$9Sm{%W$N_5vz_Ys8*ve7N%~R|pH!zO(G0*Q5ygS*}cE0$@sQpcu*m?1En(apL z`7pwsUgN4iEi?|8Ae5zK6T*WNsU1-UsNom`{wPn9xzjh&c2CldVaGlF#n-OZ8YC>L z7t}vN=ew-G>irG$mJVpCpKlH~gvJ$kmk?MCP3`Y`&QFafZpAuu@3jVaBX0axi_#yy zIxFBCA^)1XRrkT`cqF|>7N1W1r9-v|LcwBZRqT*r!(KuLzaB9I@GkACmjZKxFJS;> zCNxk9bp%)?MxlibLFogo`7=-38tS9>Dmr+tQSt$YJx}JGv!zg3-vQmFL17;lgLau+ z!P|^)r{iBoiF=e{4d7iAAKLOeZm(3=xn8}#zF5;-4MMp_>kUzQ%?=A`rXA_^{<7>> zj49-dChoxuJ;u~aK{WB@bUJ$;%;?uTEOHy)hFy?6_jO^69UqbgWMBDSd`M@wJ@2@i zwGJ$GQvup3W=4nIWg~Sw!__~FVnNjfyEzKk^gPx-0<`(PP%&k8vU51!0d2-;>6`H4 zNiP)?axtr%&mPbSCe!^ImOVS)@P9G&mQhi@-y82&DGBM65|joJ6p$7H>5%S}Zlqxb z2}uFzPC=xbA%`504uOFgVCWu(8amGSJLkXFc{{J3HS2lqd+&W+`}%;_{6pjZ-iH~C zi;-bNE{88Mhk@L<1Wu9bccpKs+-Wr4`PD2Z8y!mRz zBl@uz4x>;{>euv=Nai;4@ARh~JfX~HssuNr(^?iZ^( z4cGNA&%1O-_{R#ghe0}T{`8SVVaAOWO1wV)ExaE1M|4>&L55=gYK)i9LJ(3E4&%8* zjeZq7RfOgpo+bBLoS00iJJ8NlS+dK5ez6+QVaNU}58v|x*1&iM*3&8wY)eI*5CH=} z74`VPEP!_X?~GC=jGmj0vvSC5bn@R-ai)EYtLp{s*&f^Y-FfyfmfNGaED_()d~*HU zqq~bD(^zW9#Wy#S7f~S(rrX89>i@g-_98PsCnR8A@pN{~EB=1IVd8n+Wy$7n+~vO% z3T!)(ekvX)^$iJw7S>guuS*j3uk<#Xgtl>)BO+=n%s5h8PnmEF{QR)Xd_W~ifbG{L z9a2+D@;>voSwrFF?zXucV%bzaETE~#C5dzY?sY8zf>vjt*(T1eZZoxszQ2L7T_K0& z8-_eM{NjBgc>ZMdua&UzY{;cdZj^Mk zim?=Q7)eGYT(%cPEpP)E)@vme*mg>Yu)j7K(6q#?0)3}PN@ zU#2kO;34~!6cXW!+*O^HXcwatGhbwrRd0)Bk-^jlF{jJFs4p)~GzlI&>uNyE9ZqxA zG#I;xru=tCAvv8glOAo4j3dHh5{LCRo3*^adVkofWz{l_pE&Tso4Q8oX!)xWQp^;S z;AM4KIYi@C4_fwheAsyNwh8)Qo;2nXu77zn!OS{#vE6Bc>p@(55^>~`P7dFrA0R_J z&D6rZ`vp4)h3#QFDK#rq2{d&c+|+uv?Hp;q z2>%PLPd|NL$k*+EnTWA75{|~)D_=@RBA!jKYnfoIW*ELlkxi7a-}xPhbWEe?YnsOz z5<>|)^b_}$@a{D_EOPe?-?}7b++Rmeoe~fm&JDs7Zz(3c#oTZXea!xE>yw&x<*VGP zI7?l?Q)7Y_zR18&26KKMI0=uGh*W#4){O(3v907$>=rnm(kZkrft+smn}+|n(81oe zZ-6B0cuqgO+MlkfF~GQFVK#SdLh^&>RWGW=q4h;ersK=;tRbL{%ZMG+{g^?7G#g_KQGii_8mKHuIJprr5mZ%`w5jP`HwUM5> zx{p03*0ldR7Tt%;@KB8;VWNyAX83`kg+Suy-k4SPeFt>yx>7ut`P&`9JqTr1#S4^w zz`nUz@EW!yy<@SyVX-~M1Q`BIVwCpCq7|{s8on4Vhp_rq=X?T1r?J%GP25I-r{X1B zD=+%GG4+gJ$9F&PJ6kGg*NgGEPwPs-#}NG@j9)A@Df%wV zon1v!6m^;0XA=cttE?Fk(v)jBN!^3rVZV;9PHi-i<`@L5f&uxB_fo)vj6az?={w?n zwd*pi(SL??FMT8yiot71BakjA$5~F=42d`w1LC2+xf5HL_jCVv*g>7X(ipE2FeZxGZGk<>g7qWaOp>1D79eZ8P7AWN+A_ z#;T2hktErRp9M3oYz39JMnxlIFd4azsSqDBVh(0eyyPSj?p++F27!Hh^wnRW)kbR> z)WhlqGyoo|J^ed`?iwbJLq{OiGIj5zS(shv=7D#MZshYrfki(td|N4G`kboyaQz*+ zt4rs07_V56dHYQ3x+sT3Jtsx4S>$(y{U3U20nJwy;%n^b0hMy|Tx9efOA%xBz(Gi^ zAd{$-uieO=L@ZvloCSpN5gh(z$#&D`u+QIX)QL#&5Z0-0XtSAe_(-6&J3;))p3VTr z(@=m4q1osF+SF|Pu)4?A2Cy0RYiP6z(mMRPF?myM)n`0G!#LhBIP~_&8WGwpK-@eIh7Ml8*}P^M+y=u^YIH{NQJcG zCKytoDXIRy1Qu-QBF5^H z7s=GPczmfFBr9-=7C{W(9O3CM=ZMye8&$0UC60Pzy!V8W;ygf$5ZcWj#%`VR-D0?F zAZZzxJ*?nRL-4aA|A}N0%>rj^D+vkkWlbNM2h=oE-ql~)u8ptFFlL7)&y|7ZuiZeS zN0IYxm)&7`GF`9*-D0JUFdHBo+OM4mYg?DL&OqB|KOB&hT4Gb60H)I^} zcf6Pj#uyTZ4U0Q5W2hKIOcSHqLg8J2OaMo0p{2N%gqS66>j+pp?6yXWGz)_szskmt zW@XlIIZ8HpOy1j#WkH1WxUClMdDqol=38_-?msSW^f)A&H{_A|xr#y{ck&URyG5lQ z)_(u_A=T+z<6Dt2tX@6%B2_#I0=o0dp{JOh^j+&*2kd4shA|r5o}({mA9mJX?3tpz z2f4LvrhTtRjQlh?+VAK{^HX2HI~b_}>)n0MTWsApWsT$}+-tnK>)bnvWFG%n9X394 zFnA37l?m`T<5{p{B;DP-VaOA23BDt!elZ;4PUff4_Xeww^_Bp}U{FxgcP#qF))&D4 zb_9hs%bpBm(K77$ zq;WsoBr+T*o`HNjA=`?*{KgWAsXHSf^b?I_L!ip4_!E@Vl;UIz%HRe| zB5Z10UD1g<2y1&N86L2_x z$GQJG9gDv~upBLA{s8l5wPjtnnhoJHO|5{*Q73+=evziX#ZqVdEnQt{F5A7|2=^Ag zy8*h8>}gU$EECQRz(jC*(mE)@*m%|F-$YTd3Hwzp)$TYmabv(!$GNjqb1Or@&8MKB zUq_Y$RCrSe+!QfHVLA2Pd)O;KAh0F4>$dq`<~IMbACWp8hra{ zF-PEM3%|9q9E#dy2O=0mwe1rDC{>Avq`C97x`%4cG3xw2h+5}y zIP5qD%L_KD4*jVgO}zMJU~ZcN2xXQ122bbbKW{(2U$TnM2L}!k3vAUJyK%Nji+L&X znl>r|ZsKzV+443|WgL1QMq0mE9n-sU?z_LP@m*-US>`qD7a5E%6B|nz{BR<1Tgz$4 z?J9xzT(+M%xnEtgv~;of($=6lexR*=V{7)0o%8I6?OuL$&6}(@Rj>=x6=eMubGQ}|C3@XJK!a~T1+OlGJoIxt2 zysrxor4OVIfUD|FYQ22CuvkpCF&QRWVkWz?#9vK;nEts%PbTRuW<|9AGTeBtvt2lV z`!^Y!{?${PaIo<~f|ApcXEfml?Gmr%JbHP*GaZ;s;1ukI6V5u$5flxVk> z;9i?-nV|c7jgs&2#h1(1%fy&e7rBrSh4%M}k2VYF+-7p-tuzu4wKsz|C{2lsxA~0(5_C5n+D2Zct1MaE#nb{Nc)%Mc&fej zu#Pw_*1}E<`*4U{zDm2biHs-2BvbUpTflPqtk$n?o`emPjMzl2_Do0*qCsK@c)j4oXz zZ}3CK*>ei2eGD61f7a{Pb%(bRcdoTAdN@4=m_|e1S6E;jA^RU)dj0uXev{kg zjxM|3LpZh`1>q5bzB?=6wO4d+U}^<^2e8iktd|bIdYT4`Yv8W!@)CN#%3(GM{%+^y z?mQ3U?M{#%t2Ks)?cTtaQcN_#ufxF3F>wDNlgUgU`SPolxiB+i38nry;98`z2;Mm_ z@2@@$KdX1n9PEE`b@+=zXb)+7h1qfcnF0)8PMcaLv}ID?mkcD)Nq%!AvE1ZR#X$p* zxK2qbsrZf0gF`4$l;;$X-^nEIiz36qi_@X7^LobX!FYYKyXVNofCPmts85-*R^lBh zy!j|f2|Kmb&?0bMjzL%Fct>>iq$H?ueF020$=xWLv(k|X4N@QVBKK(@*nKzyb`Tf5VDoepc_ zRmNn~PF!YUId!u%OALx^dP#J-=k=r2Nj&is_?I_tKo{P{T%S6kb~sRmAf+Wm3Ki3_ zK;OBu<3H*hm!2X*W!U@wd~rt~tq8!#>n<_gWf8$c=^$$>D`NIsF(qDr8iII zo65;zkxJxW2nf8&GaOL<``gC_&4&#=K)x!i^BGP>JIy_*YC($l)wVo@yu(VLNMIlX zw0UbQ8Fw6o5=|I=s>+6&A5P12HDOBrl5@)(r{)elqXHdfmedChjhxBs8FE}a&Mysm zRsG(?xD1*p3G_nBT)AUdJvh~^`A%@*yu5DJ?PjSDb8cNZ@7$}1M8q)q$8l=o`61u_ ze<2@kl!k^=fYSlYhuyiwK6lrTk<)6|L=2ove^N(ajAPzVD6z|L)vJ@DnhvJziG0ur zc7n2QdDOsIHepAXOvmtdCH$tn0Xo!9#}W);j4}1CAwjxCU-rX2HxW>C{;ZM96^P0J zb34?Zy|}}684RUZSP})BKYQy~2(}?x`%|+*%Mfc_^eO8`QloMA_ho!q4)Ps0c>U{) z!=&m#@|#Kl`$p@1$3Z+AEzFjyk!r4#g0~hNRH?Nqq-7vK^Bdh5qY>k>Ghn)6%c>IYlM2kzmI>)nl)TYg*kVznWPU zFq}p1lNsQBZV~9bJlTP!^M zIR^QFm3HWkCNylZh3uM&s+Wi)nt!i_@sP$v)+2ny(a}f z7)-(+KX&+`Hv)+b{&OkuGzFctjO+BWkww*PN~`Vz9R{rUSz_}ilUx%|E$i&lFlg<^ zxjI=7rf|(m44y1?Ru~wDk9((i7|#gf-sEp~3K3TW)L7A*eLUS5DoG*lx(|?HlV#ZO zZ(!%qAe=ucYs6%su613m{ce-e)LZshC~qrf=$7rmik``TuEy)4Ns`nTgEx#a{9>jL zP&z-jh|Zt*1J(>koFovcdAGqCncOY-GPtnCFmuP8rc*#=fWK{kt881CpEzS-!K7_E zB8ib8iHkxZR`eh!JHN+XCKnp3G!r`NfW zXZFCP+Sq)G2)v{&I|F|N7Pu^(`92iwPt`--Sfy9;cTm0;jhDh87FBVE59lb?{2wN5 ze2mwO)}dq+J^ zM(G>a71AN%0@RRzUfbBr#y&#RX%9xI&Gg8{+GT=1Q=edF#0Ypv`9;%HW>UlQG1lIb32IW# zSZJlmfn!KH`NF;2&&vTod~f^NHHip`1 z2sT;iBRbXRAvleC-GxgsQ(<;7dACn$$lUm{C~Q5d_rrjcyzb1h7HpZ4#(KMZ{`qS$ zE?fOPw-J8muDV4+k5=&qd4PN8RB54fWD&Hc0O*H$B>Q^WYVN1+g!=GLlIoJb-ra9J zcGp^ut7Ul+#OFR@KD{~~>dlx;g*#Pn_Wt=dd^VJFtiCF*>HWB)cS4UWHjKwl*4=Fa z68KH$FuN#!w`H3-eQy0|{E^JfJv8-!AaJ7h%i$QblxITBmWiFCG@5woT=JTsjYPF_ zn|Az-=NYyDaP6(Pa!ZGA!;|F>oA#h!`-@r}Y zo6@Is{B+#t$W*sNoucAy4Lq>1N5P?i?^pA6<<9Nb8Sm%(gObyzlIpEac!dIokOR}j zqjCR+9BeaHFmo14CBs4owYxRfK${$0ucX1Nu<0u(G4tL6-KnK+0WuCAoG$%%Gb+5l zea>ee9M8THsDGgnjjk%J5lH{%H2IhTd37g8Vzo zAs6Eq7)ndxi`l}sETQD&^9tjIAZ|Wyyo>$Pfy08Ju`Isq06peZy$+fUTILN8I*#j1 z>fpfe-sYNkCS3xxyEtr{V@bi~GZ3baId<(YUG4x%Cb5^Twg1*J>x+4oBolBNu9_#L z!e_~~q{5wQnHygh&TH5ZlkEyNOuZV^_COpi)Z5VjN(7%F#Eo8iSp50=$T0{PZn^o? z2_?%6Byx-pk5hcKHF1}%#%uZp?*(O z_)37!;ez%C;N`8j#&+h9lX4=8>+N!Sc}~D`IMb&19YFNHnuc;7n1V3ct1` zH0g)ae9gkOO`T_=?-WHd2<|`3*13H^ih1eYoKkU9ahntc{WcoIWy#S$y*X=` zkj2O>wI|Zov(P!I;>2pLDuCN7qrEoNYG}eYrr;v=ou3-<1`gv70W34E#^(EeSBcKZ zzuP3CA>$)1TuW`nHa}umbTN1aGkOtGvrI3k2@vvkJeEi!vnZ#-rgbw(F7T*!YVLmd z4mVOt^rUt81~c~~T5S_@nfi$A5_ig*Y_FJuh`VX>;*tca<>I+`_ioS?G6!D|iKXjA zd~=+0C^|rQ+&2QzV9#-V zjiDuSzQl8_44VAaegT|?bvcd&IOUIb>9lgp07VO5I_K*zX&w$%%+{SS3Iqhrciv*` z6W@mQ0t8c#A-62<@>iJ110BQ!_fC&&85<8_KZrBoj)A(|RuF zO15F^)sx(JS7FML&0#nHZ4cZ_T&YA%<}aMSHy!}qij27ZYPgP6h^YY5Psn!$KZJx8 zMsD~p@Jo^yz|p`Ow&C9sadxHpMqkWRF-@$;PWBkm4A7aA^EsE$yPdqv5B+`pYT)-5KRu&h3mdgO@E*| z^!L&qF(w<~{YdW)r?pOE{}fJB(MGGyXB93*_bnIRbmh%-^c#an7xqB8@rr4`F9uEJ)A3KYW0F5wnjEf{9IyG|Af0dM>mD2NV}4l zURzo}e+UHeGL@V10y-YtI_9)CQH>;UVpm##O}qoQ(&RLBZ%J zJWIaJqa>2rEDbQ|PyfWp-4`mAfUY34`ftQ(g3%!3^4oEk@8QmgW-RwbIlw!{?um=x z49D5v#r_PCm*cG5hA}e0E$C;dH}9}n&4!xe!|!q3zUKXas;s3UvHNvCipsYJP-o~U za;=5&X3(dGOziXuJ_p>mx$$-Sko#;?dk|Ea#pd<-hSKRFSl#u>82$3haNtzSb`{{A zk5@X!47iQWl?c!p7}11*Q}61TBArZxwRZ+JsEu#DOgqw?w!E6hTMY$g95~QrE-nwF zoTbYNHiMV2?v~;*nL`UFT9)|1tmB&1&314->)c`B=pMs!orp#|9?1wA-uZ@Q9*#z?G z`##1ufMt%P4oVc>i`Fvlo&_`HXKcm_qxvT)N@Xn@o2ARrql< zALue;fx$y=uL|wM7x7}hKAp~jNog~_sb7H+H!Q6(k zd00O8yVR<^N9+-=g+M9W6KY!`8x9oqJDN(oVcp+_R%bl~PY(H;GN3cafSpEHmDi7X z4D@@juy{WnJXQ%Xi0l(P>KxD;qjE~aH$g1}3>PrX%_6=0K9ja4G6}*MX+5R3-fOiM zf6Dh#6HSMh>`_Ez?8|JFn&?S&JAef;VdRb^3rBaW$fN4X@)p zyW*@#xgS#(8#VH1`HlIRNAP^jfa0{B7!(5%b zVnjJok3jbTobQ?~KY?p9*%quppJ_y&P+w%MPBs^D}QfUQbAMa0#NCbIRy#jskk71}}V0|P*$EB$`$ zPFPc`DJhmdw;wNx)(rLC{!MI&O=~VyC2AzCraBMmo|v#*$5ngw(fM-+w6oJM?}XRt za8RsQr&|^uyO4Ax)R1s(Ev@3R_*&F=uDJGyUhFCFXG=rFYGaqS` zSzNS$Z-qBw@-w~BT3caAP>tO0)Kng7v1PbgzRS-KFINm275;iIZhY=qmN}EOWzH~| z5SSrk`AKRt5KVDj0R-C{Zp!rR=E*L))ehW1R!MvYPW8fkNks+JKPOVl=yIo%1nRK{ z^GDCFiO=2tC(y2v7GPrN-M2n?!d5dFC@p`5p@J5JQ|sd0J_RMS%iXq#MQ-zolF8Y) zJX8m}9DK%*<38(3{S8bqBo0E%{+`wJ2fwE!dqZSI>Meva?Rcp2_w|@>k2g%aewwe# z^r)jFs=Hn5L4W=5ON+XQ>?e76>A=bV?h|TNF2u7fp_Ttx0x>&AyW`Huy5CJ1*KCnb zfv|G;&>DO(0sXwc@p3$yKi}T1@!V(>my}~K zc5aDL=xIXI>C(y4+9rauQg_hrwr4!H+@XOImoXZ&A#KsL0II4g+Y5s9xPVtrB zcrLh@!Me(E7Li>2vQmzyuZU|x&o`*H2<*^%o$3>)(N;4`sw4kh9j&pOWOud!h zIePtn22^lN2xSe6_d~pVAimqx038!IqM|O4gH1<9hL{^sTi=qz#_YYMTpzCYOn@dY z=@-g6m0WhZ-xQ|zKldY@PL6v_=l)CgS3`y|ZXKKLN9iVns6N5Xe!cUV0A^UHz*GK6 zm;=(3Ip=b>podJou`CT#ASucs;N4i)6>*p?r##f`c1<4+xU;gAmZp5DWxt&90qskfltBx2kY+G zYj=R$wN9<4TRNpvbEc*BcjHAxrr;9$5s?8OhiZXZLJ|QN26ssT60TQO|N0@vIVwGg zR=HnUiwr+RkaGAsJSzWuzc{BcUlxkB$~RYK=7{-!vjsK^6@9Ju{Aw4U;4o0;{djcO z7W+|S5W>~#))F|d!KU1^{d7E7UsGNVF2gM^-|%x^rW(d_@)MMenW< z@H5xWV-FnP9E*n`&28|~qmV(&?lunUzM zQ@C%!JzYpZ>zYOj(h=Tb{VI^pIDaCV+#(^;amAR-QDvhzF8!(&>B7;c%0;>u`Z&#R z6ixmih+WGMjRp=AU9`D35ME4ay06|5Joa7q7JKsrRK+p4@{S+!&wN^v#Cv%E!ARTd z*4sqCMI0J>B+jNXh<0;LCiODPRbZzR!zwhL9Dlx5tXx{lc zO}6FHVi#%cf5`@r%1F$?3;b`DfSlX)U#UM!!He{__QXob`VqYzA+EkPO<-7j<&Frn z^avfZ6I+|J_KZvJ?@ZU#I((8X;=;KCHjEt_s=g#HLQMfItfVk$a;Kfl6jEb9Nanj7 z(E%a8GEsMVkmez6JHzU=sfny)tXgU6X!&DCxnPU)2hD)(v21vqn0u+2uKHd`f6--x zvGZT5cPD+}ed{Lx_tQC~6|zljzBsRhXBjAYVE8S%W;#afpi`Eu`ObllTi-w=t55+jDc1s-#!K8%J?3xXwZ$K8x*K?XH zSCRjAYNr1S@v>VcEKb7;Kzn{)jT?Qz1s;1VLph)mq%ce}j5M1z6*2L5qI+d6#XJwv z*x=E8-x@s!AG6`u1>anP5ZpE|QhI)rXjy^{xJOwui|yJZ!DI=$Sz=sN2sFKrmcW4S z>UU|Lbbgvsy9vFzM!V5`)H~|eDs5!upnxVZ$J<7S;apK*2do_wLm!VW7jlf^o-K!D zzx)coPs~@(r$(0Sdwjpzs6H+7fY?*=6PG(oUMDC&lO}ZCv5Zim|9J z{|t`>b`4rKZe3m{n8zS7)a05cV$oiI*}+#^;zFDeJpz1Y%w1bjj0jz3 zyH{jR&)&?jz&|^C8(e(|wJH!iyJ8~aMt|_kPEW7o?5&99`h#~!M`n3UF1l->*R!W!Fce7S`Hoe24=KX5n zf-O)}2B_`hVg|WPH`ToJ+OZCqPpK$L?c~}jNL~t^QMEJ26X#q#M?GjN(p)RE;Oqko zjxmON|38{-B*{&WfB ztYLZ?2|?rCG-v5PF^(*G&Wnp#KvDZnEB4&k`~&cRBu$t%y^Zra<8jJL#{7TZu-DqN zt@MiPCMXSwN-d_pNIE<=zRv}k$Ik54zgPG+XnwW5Z z{(N;78kkCzXV(2h*{y#lf?=_1tnJ(!h;HIE=W^AeVZXIlOUPEFnJJ15QkOYUe%tU6 zT4JG27rUg?$cDm@9$A{*qUnFXZ-c0$5n{w0b&89b_>tg7a%O3PV={{8EPm{%KbHAk zf5M>cJWo&|YK3KQXbc_>{uw-Bp)D5a$h>bht8Ta2 zx?wJ!f;GOHfCP;2eIsfc;OPk4aF{C9x2uRZ@Gn$TsQUXyd~x*W9%B7%lt<8OeJ|B zfhfDuu30o_WIcOX&33}{`SnxUU(3uSjPtH+ghd!3fRLZ)Xkg{#R#S*RgWsS>@;1Dq zlJY)dCSu6N$ZA*nC%`rv-4IN;chQt`R#j-|sq}9QdR zldz&q_<8YjJKo$NM66|};}2DWxb%AZ6^1g6-`G3QB&P3Pr;c^}wQD^=B)H+&q; zAs4v{?w%}U#Sk=%9WbJ_oZfPR6gSND-0bM?czrF^2dTs4n*6Nx%jnUn)rGnrVduwD zliF^_WWBskDf+oH7y135%Q<-uZD8r!5|KSw2nbv*@T%`La5EeVx*Z}rmW#XQ=4nv@erR+4*1`oYG-#F9XF=DmVED3viNTGj#k|;kT&I6fLRs7GbBcxjx>! zedd5Z-c7{Bd>I)ub)!4}J-tUTBEOFMAr6=S{q)#P9nKuGR7~J%r!UUiAr6i zlDfGiJVw{KK_;a0g&IEX%W#9ooS0qcC0q^O6LfFs;FBmu*;U-Ub?O%zNtzG#Kh4M0X5sWN32GH+I` z9M7AUwB{4mZZZx=KJ$7A$tcqt=Cz~9zY%Owu*Fa%>h0K=RIPajl-e}LxZSyi4SCsX z8{J9_z9zHYEN@!-ky8+&#E1wAG*MIh>%UV{b{VlB80016UgbW9Ii_aiefo-L&ppXuF!)p^(X z38r+MpRm496UdS;;r2+jxm`@&s&m~pL%&X#f$0BZix#B;gy4rlgaM{4q3LYR=I{|nL8uGq4t%E!9En*f!b*r8gS;!^SsYpXRa z&8w=YM-V$XsL?hI?IFv)KWOoeh-t z&4fbDfn~f!mUPIMznqRY3N8;Z%Y2u6JD*C6#9a5;mz=D7zf!kQnLnqRWo8yD?vUO^ z8Z7&*$E}I(!j`IYg#XS%#Z^?UQN?VM*odU_AuPZAulKlX#h=f$SOYQUr`AKx*b+tI zl)v-UZ1$LOpdMm5{!FwkNBqmLx9oumE?E*zX@8#A;J-}6NQbxgt!s{CZDmo28G+IU zo-QOEyWPc>tzp@vutR^XI>h&#gyi!_C6EwFk7NDB4?OSIamD5!UrIvWy5kYj_)6cp z6G-=AhsA3?vo2s}{)PFt)og@KIC{SoWoIy7mnj`%sGlOP0qxmqa0bn{Rfub1+VV=o z>y=O$b)+z+A$H-GC8MCLOYD?upB+y_OKrMRJ78#^X?^>*=YU#t6bpWrnT{1IYL^5NLU^YqJ|r-Xs# z>@kYlOO1k0US!#5-V>kqaO3Kf^HvLM2&(Wm9<6~zQ86GfN zj`B#DZwbR34}P8b+iKIuu|hOagF}{g5wmMmb~)ZgGJ8Y1SEPTTzXlNyxiN2FqaXYj z$iH(c#wS3wnHcKvY7Vn$$z}#DYtpXAc8S#?$<0h(6Hjx)tMMtHZ~>PNSRK2lf0*`% zr(Z8Yt6%!AQ|cgo-gFmh2D#0*$TtQ_x%Q^{n`a4>px=y-D?H$m_-S=Zy3$CHM3Nkq zE~5`_I)H;km_Aeb{Szj?c22Ybe#pChv*}8Cq({2b=yG1&)HYC^1rfiE;#41uqG52L6NiuB>LB6)zg{TUe_ z*5IDdBJG5(E@OeTC!2f6D6sa88li2I%)S5d)ixxdR6V)~5<3|Cx9V@k$am0U5m5b^ zxZX1-h}u~`n`z#9;K(a+kPF57$2hHQB!NoRcV6}?ceuF0)4pjs>WLq~X-5W+b{5U8 zi(P8Fmlj4o?JoLwruT2;2VNuUJ$`zAS~fSTR3LAm^}W{Xa$gO7cM_dLdB0n_c4yj7 zDZTNnQTAg3btbCgjn>>6+MFySLo*JcvnFp#UF7K19FwO4ml5rK$I z_brL=VtW&*wLti_iaMSxg3RqOKc5U`w%68EmYLV2(cU_`IeSt`?pPVAU_&0KZPUK( zRlN=HUdL@WCG_;iTNJa!Yz7*)I8|r(B;`%xW3=qjd$2JX=OaHF{3Xv zXfvH)F%6n6nDcYN5qQ6PbnhD(IK}@La$|aj#WVjpz@RJD&0ij!udndZ-2392@s&o;E7UD+ z*7Zo<7?P%W)>1}HTxp!PlU@;_jT!vf>CZoC0r`}TQSXc-b3KOIBbg#XWBk9u*)e%T$_BwUTNINRqU+5dgjHKELPVuYblnByf& zt~mrWPBB5}*EA1$z}n>6zLUH`zZyJZnNyEUwWmZ3qruK$watLCyVJ+aGDaQO*XiPD z%70!n5oq~Uc>R2Te+1mEOO4CuwG0-8&~x|13R%94;e%uAi0E?Nwe~D+ydh~@RUhr1 zh&$Vd&Ej`+{`TK0me$}wG(8&<2CVJTWtE`xJ_+0y8-vF6fVJG+opf{K_G9%hUwS^^ zNWvL@D~;E^s|sT%t{L4*^HR-ov!mU7AUWN1Dan_(0}#`-*gPND8dJ+SmbSTK?8O4V z!5Pd#Q-CTC5J1zTg7d1W2qge^*xs9mv`DhSg zMSsWn;d4j(?U-NLO_!L-Xcomb{_v1&+*MWhotUb2$9HG?`sH*;0Pz2oS;9+?Ncj4YEztyo|l_D-pskLk0Sqfz$HB? zzA*)Z%)e_tR?v2po?8Z*=C0q$7yvM3g$ME9rFWDkoh6==DRqLlr`1TfE`x2q^*PTk ztF|sLyQfZH+B2%#kTIw%N?KU{Edh8EPr8`nyBt^-_!5Q~@P|ey*D8&FAl) z9osJkE}EdGcKn{9!Rz#aV4eEC`Z>2n1B~JJ;-bJff^ch`3e~}Q@haNq$81!)D^r64 z_Z-L-RxB*ZzXV9WXbgZ}@+-dhL7^=)axArK&V@DMDxO9&9$ zgG+EHK!D)xG!h&Fgy8P(?hQeMg#f{&ad&7MXyiTo?wy(YzTcghJO6xtO;uMHMc1jb z&pvyvwf0)idYwiJWwjPLQw*^gGr7O+xRlnqkxZtixiX+dxvdR~`xIt79}!?fZz-qx$6 z1xQg@bt;Gt#H97il1s?J-$ECJF_Hy!TO;^K(Z{%nGK?&bUE~^UC*o5~X2WW%gt!^O zfhH?xM`u{#TKJGTf8aGYfE`pH6XS@pI!IX$~&2Ih`b+C>o%fp*iMM2^#^2siq} zg}K(-?}k5s44t1d?*R-V#fomhb#s1Fc~^#wJ_S1`a_LinOQ04VBIL@XxVG z(1muhG^fDmLEL2GCVThAE~IFvo-f>!Y6<=F;u2zRu$(MnfX}M>JCkPFzFlv1E=;!i z(t4|j+zx1=>m*y8FVP7HcF@mB@`nXM>}l5nbD5AX)w7kpvYn3S?{siO_-IvaWn$h3He}8U}FS4z?GJjAp9?EaTZMzJsfhbLXYy3jgaLd?4omj=l>_;a-#@yq6{LtCIK$AJ95S0_thDJmPbg_{x7hX$Hqium zK_)N|+xm+2dL-ErEv0qXEthe>cTk^#Z{)?Of88#| za^pjtUhP1}aG2Y-cc$s%bqWZ6%aZ%L8a33cIZ=aIGt4uo!nE!3sKP`8NU&0ELDel8 zLEc**Z;0oN~xv(?b{Ur-ejT4RCdQ)*FlHygtF` zzy{x{a8|Uip0a=eIa5)UfX%3{%pEYxPqn>EQ{ehDW39k6D1 z-tUg2zs2`lUsmG4v8w}FApPugK7^>OyKjiu9$jUkOM^S!Naen*FLw&YUQX3DYk7h> zV7h{dCPcqa+9al4)#`eHI^+30kg$DmN&Dsmjg3>8NcQ|R6qAL z@O%YrleIj@Sn6sO3}caJI&bGPAJsZlS-ok9Ykx5Q5Nfd)3U5Sfp5UgP_GPYVUgtOc zK6-_9^L*WK%Ay_RdSzYueHKGI@v0)E{6&}*!igS7f- z=!C9bplKE%Loi1Qnj9(6p=-q8wecvrm+u}tojyQ_x)nAa1T3h-0R^+47$UbU3Al~jSOTZJL{6S0|<8UoYshYg{xQk7y%d{EI_v4iU*FM z;1f-PAYh@Ri+@A7s-+tP- z)W;ViaqNW*V7Ox0=`a8>7^JvvPQFEaA3W-yKt9>g#i|+OsTdX7Fjh;sza9M2JuZn@ zbv}S#7xBi_nYM&P_uVWphSIuCZUnYjB!_=DViH`N-~TEN;sT{X_rZ6Nq}%RYTbNP^2P2YDVh#L1QDw?W%g{a zAZtZ^*;IMSG^dW6L#OZBdXFRCb2-7wWCo*>@3o{4a{c$2JdU|e2+=M$ZEJ8vOUDkM zLEMI8tP=H5T#{kDfta%94dT$?V0IdYKHh#7I`C!c?ENx;U!2PS(kchtd&)Mde0LH! z+iaI%7JMgjf14{`)hz3IA_o&c-&%9$J3`i4cNoo-XUaJ}3QAtBrh3?#<L) zNcq#cK*4p{qITd4cXSBHn%Pn5rp~+>bz*t;#=IBxrISmjuKoFh{MzdhjQ*>B=o!Cw zruDqKBDttXZ4$pbVbii(TC&Nb{^u+P7~;kkVL7ELx@r>w^(!4PGtxxggSbaPYe(RQvSE zs)3K84<_@H-ubq-`HI|DujyMv{X2BF^TjP2N1G<>ulr&`47!ko_6fm1x-&=UFN%DA zv|CQQIZH7cJi@au$QEbjH=(Ed4P!&1mF0$QV)b_Li20y5sdB57#L3%Ex-eBAY9aIW zK(2Ue;y)suXDFLTBz9Gbr>Z`tlojTO@2x0Fjw&ENph6N!u9yKz41gEUVZAWT4u63L zT@3*rDN-U-n6P1wM{~2fNzH(;-lG0_b1)mAxR)$!_O;*%%v;gbFg8&)Cz$xM`qrmF zp1pvUi#_zJ$cO@wje5i zuNjk2Q5ruVhVmRXtt8hAs40ZU=Fu~_$M+yZ8%(6>%;8bYo>xGbJGlgKz-bXLFqXw@ z4xtJG%<2pXUEmrQd#pYyfb5|!m)AVCPi-FnU4#$bbq7N>LLCqTIG$ah>T}qaic;h} zTUB;AmSD8dvf6)pLWdy5fW$U5F8vt@2|HhyJUeE+`pngL;LTC;eCQ>EWj(-QVn_aI z#j8UsFVX7BvaAkQRZ7}bdEyXs^l`Vv5zMNqKG%%JEh zR$W}^-LA=Nm!v%nLBqZn>Of^yl(Mqb;5Xqbpt(>S@-fTc6jPa6<*TVb!ZAj2ubEQi z4{0VANW&VbCsM`L`^@?M=hI8=Y&*VamIudO9+2cWw$WDB)><93N3C33--E+1{f%R= z52qjcz1V{?7B@rr#Ha!4hTe-3Q!;{t`DOm4QM%H>f_xjNGKX>vMS7k z*|QqFPQQ^FeIkE>j=JkFJ29L%5(BgI52xE?no_|_9dF8xIc08TeSHU@;(i8noD3Sr zo`pAvh|XD0Thd@r+)FAxk(xWqAzd-f1BJ3ZyD(9z{8KG2<)>f~8zdn8cq?QJhh@lno434>wNgWCk+TC>j(y-rFI``kl` zjBu_9LwhG4v2Vc9RPJiJ`h3&=FX{$#Boly5kiOqtoVa`b)zGnaZw!#M8nyR(>!q&}Vp7U!-1q?{JpC{E3y5 z{K>Fh^F_!k0i+tqPpb7OguB5YJH3K_{he}>+$R~82>Xo!uzi8ag~UsA-QlSjed_!8 z%&7XVFRh{Kdl26NGOxG@OkvkU0Um9yo;88*A62^wq5B%OSQBR* z4&%9*q2=1G>g}BNnR1h&N;t*7a|t&AAb9Ud0Af#cg>GKR=Rfr&^E`flZlTllS@m6E zAAp4xB9wB_I1NO{Ts9R5$YTIn;?2>X>GMFvg@dE2 zV6wFYUkCat*W)N=)$0{oGZYay5YOAKvMK%tzcM?Jm1EJbS9duaH=?c5&iUF!X*h@* zHt6-C&ocvdBOV|IwXW)TuPEyOW;Q-sJNZZx)?$v|45y3pcxakYLL;SxKTtiocdwWqBayjsvwkw6?7?I5G_;yoKBS_lT^#0Os z?XZ?U_2_f5@<(_5yM8N1t5`bnfeYlB)X?m>MEn;__AT9-wW%!1G3|wqCd}Yfx;mwf z!nHVM5eY}LgG>Ef(1NC1fKRB4b-1<9#!n)Xp^LdaLopJJuAaMgK3RYzK`7eE9he8USD%9}KPDc_keyU011XMWf( z_i`^^u=G@2V#;uQ40<>vmHi;S@PxVDNFiBJf|8u$CIsLe48>gAw?^H!R%jK^nrxG% zkCkYIJ5RJa+=Yyq%zyPPJ$^#p+Zx*DspeptZ4*4K*M8i!-@86B&kv#Y>;)*5*n7ln zRPhKn!mzYv5Nix;Z`<1*#e7`J=Ca`WWp!o#{`UqV$e{B7lJM-6d&>(#0rwT z;Dg=^?t|Ke>+vRw!`{)JN*#7g;GFDo_N*7Wz6pn0JLBw|4^_|f8}hJlfE+?bj!$Fc zjI%=Mg8vr;Hh3fvYkB!o$BM?|MOLp1FAbqOR}ot+pLPAhZ7hZtk%QJD9;=X{d+Y5Z zAtoDSAa=&ZqT$4coM;BNe*E%WKek86xd6@VSthhOQKel%EJzGTaC;!;W$Anf)3;kh zTEfLX2@V}>*qa+ZeQO4ulV9ljO7*O5M0i#Ayougo1a#dL@!P?TWEJ7hKIgew&8Wx? zx4nV9R{H6U3c4Da+GlNopIp6FR5}TqYGVW{qLVFuS!^JEcyqf2YH|4SwOW>!f3%jM zY-3~bXs>7~qs6x{a8g6-PF`5L3a=;hqDh>;qscbOs_58up(L7j-EjZ5M|KYXEBBlBDbGSXk7S~kPf)V5^AGsm8(kWcnQ4Z7u^+d9>RCLS`h<<)v|=- zzFHTpl7sC6vGuEkUl~Ho=CbHp*!fS1)>C&2y5TwPo_%WHG0&#tmM94z@7ViUbG|_u zJ?mebtfs1C=dZqN z`k|LWk4*Dok{r+j$9N*X8;(GQx~6R6;5Rto{xtG5%Evey(maPQCl(CB!61CN-lo{Q zSq*l)Pan*^lQ-zYF7j0;Cs`>GLuYA7`?;R8IQ`UYI!5(FV3JLHbNhT}sA~Dg&dDwY z?uE`ZFuoktN=nSpONjTGv21?_~m;hU%rC#JY}% z)S5r|6^IW1;HUdu{{Y;Vs8=u}+!wK9P=tq#ViVr+u%Fvf4eJ_Z6*Dv=4a? zXp;OSOHsZ*E4X@Z$lkH2$dVMNT;G)-h}^$~1Y};atO61idpwsc{G=D`UqAG2*@3k- z#*(zWpvlIdk=L5$J^idb^C&EJ%EQ9)O_qY_0Ofc|JI!Zu?bI2ZG2iC5p~;|_dCG@b z`XvA*kXw;EntNy^)@gZptuoK9^LAjofqN(%i~{<`c!C?~&Vb6&Dj-|t;Bz2xA_zd= z1@n&o02n*jb!o3g^bAdQ#lEKHtp+5&UOnE9f#lteOw*Nf{rx0Xw&EV!2_cdj+MkNN z<~XdGwp1Z8q(uOt8ax#;?6-1?Kse5Ecsq94qI2t{mc#xs--BG!p`#5-noTFaYkFGV_ayjQIKmoRigY==!6QKg zFtpngmf@GYlKrY73->arB_=wX3BBp>3Sv%6Lm%NYSjijJ-UD9NT0m2x;h}ogNym5n-(|C;qP^T{`pCZ6bC#%b`t0U{v(UKEMfh;haj@^v*QPqJi4 z(zG6|IDVp534x}EomK<)H!J2}b7iWMkr z6C}P(yce}VI;QauL8NH=_`M5b*Vq`~zkE8~Ye%ksFp$#OV3EJrLawrmG6LG-pI2`I zctq?w1uAdyqxKVnp|DwqzLI9}8$Net-u4`6J zCYOg&wz8D4uFCs%>W@y0(0sMg(&7*nhXPtN6Nh$~H=9p0+#$A6uFVC?5r-=%5|qtcO`UqCU!h+y zbb|K7toM4~&F_=TFSQSmG8&OaK1aCHY}05>oH<&eCQBlQOUx`&qZHKAyq~tDQf2}p zwvMR#2mlhXRyihStnN0%G6A#!Nh>))0ZQ6CU2hc8ktO(DZOY2Ka=OrUPovH$ePJqm zVW9y?LNd|6+GJLrntUZ)1gQQanFx6%3f}`|3k-+$q^l1;X{TyS@mv5Pf&t|$mDAc# zu#X&nl)Vw86wjHnRYp1QOwDlQxtaNGYV`xfJmg`@7 zut-Pht50Wk3wlS;5i1;Aqkccrb--9laofGfC4ToQo}}8nmc<3A_^R>S z%owp7w(Q+R%b$i#0w z;?c(aAu7o~pa4dK3R9F?h-RKA89%~49vk%bMz=R=8jqx!?29Z(i_@k$p-)AY{p!1q zyP=543OOEM8ZhxAhLhSW%MBc{ns~l|bh9a63HX+tk>oZRhsPjm2RFpVk>il}$l~VR z@n;Mu3w=>177fQV<3S1>fiFWy&84i^^DM(X^1gC5_{F$;ky0pj7L29MEk(qCoUzoI7qSC?`< zL!;C-f*mgijpQyVtTzH{GK7i{ZQ!xk>(w(@Q0_f`>@AxQCMUdQmqXXlUfp+R+X$Wg zkbH5uniIo>@2_ra`cYY}>cD?T=Ds2jh z)h2LdsiD!O>9+PFHCHI4_wX9dJ|w6Nrl+()xRF;o87ckHQCoj&PgaE;{pSI|vLtWA z>adzdxHj*?ys-$e3Po!|;LnT+7ngadrX;qv3$Oa`wZUXm4Qd+{`D?F^4LvB1=!EcA z$=;)9#+;~}0PJCFir+*@MAA)my$?XyY)LJE%_%$&wzd@d@Y8;luzfu$3G|l5aP&D9 zao8}M5PE&=^uSc6hkjN!>&4JFku94ECuH?U#pe=0LL@&H6Zz@tFG9jnVw+#7+>xs9UdrOy z5};*Q#ek?$Cd4On_kMWD4N3|)z8QMgz~LVZQT?i0eu$fhr|~So>4z;^KlrWz_(|~PRHo9=q$8&RK!7{ z3H>iiWW-yJ!|?}P!BZj3=+S$GlfC#Ll|4Q zOuXGxZ$wirH?;;nb%Inc=TbDtl?bivX3#X#2f){w0Mo5`a`v~)W$(2X9(Mu4h12q` zx_vaufROEAHApc0sU@3?yZO%g%Dhm)b2@%9{@a>TBPG7qPuO|_aeJRuLQZrN$kyg1 zWaE>+Yy;_y)CZr-bT9KRW;8m3s~7c6>yRuJwt2^j(~&@db#2-ln0WnmPu)?xk5<>6 z)1m83sqcKfN(>Jqh!m#mE8vkZY5V3Ci;74bp6H2P92Cs*!ob=)qi2!G8e+PYkUo>c zai<9;${t6v=ZS*=a}5R8!pL4R;vnDZH9Ktg@tWdL-t*UP*p-aabG(7W(aqb&emUX-b0 z^6ty?A;)DrH5;Mtjz3DSl=_K6rO^HTFM3m6XW^(`D~4x2A%Kk6p$mMEIOZU|{Wpr~KXv=nT*;mWW)1&?{qeWY_hewqvKWH> zlVtZhc;f)g$Y##d)3H_688(NPMh)*l1oT9Pn#&9IZ*)9ss*{pB)z}f3z-q@0n6>7~ zBV0g(=u30cMzr8Rd3pmLudy7!vTEiL#?<|#YS=ITNCTX8Tt7`{WDPA#DrKL>)BjDO zM~{tS+%!=vkq0tzU;!b)X0mnTQcFMwgT6klNNm;i-A9GSK?foz++Av(49tzuzl zq}cy1@+A|8RD7kRlLVSM>d1AR+*TJnEUNl;y~Y8-*E3dG!3a7TxZdyT&`cE66_>r*`O%zs%SYAvPt`w-W6){lf}%3OFmOy zl&=YMv7)8a{>jS^;H;xj#P({yv>=_2v_Sg&%xv2Hbnj``Ma|neElrN^CtWwM|3fs8 z5*0Z5pono=o@;j+NAAqF6p5B=^0shFAah#vihd|`k!_+`RC*e&O`&EQ9h9HfLmnSm+A!8+D ziqr}tp;XEgq?Jo#V9K+-qeH*Op1gDMO^-4ym^9=B6I{ zG50wu-G@5bOA0rfx7qOlpaNdgr*=N60aGPzBAOH}r0GSy#v>ebB6b4QeTSQimz!sT z(w8=+M$DS^--0mfD}#( zZnOB_aV6LwsB(~{mq3etMzwIq>BA}6%4Xmh92S{$|6MYr-jB!Ao1 z|H?N_km#14y`qfMSpDw3^RDVoE-9*`KdHBpg3s#c{&ENZsr=kO{3;WIq#Cs*@Uqq{ z_TL@xfBT0jae&{SHG$w76UYDX8~^6U|Ff&VzXrT@+^UrE|KLWqc-dQ@(tRZC)DU6aG!B`@ehk z$BVl}0{qlZL2DP5542`-|KZC1o8bkhOHcwbRLGh{z)@df@az9Ij{ow$x41H{5yzCw z8$?u-|GM}ojnH2*utJWt)mK=x;yxLB_WwZC^nYK>|6eZ#(1Gm(bS5F}hPV%R5F>=GJ@zs`1JB^* zqxrJl1}&q_<0pmx>LLtK7YAOoKO0T_@;3YJOBK6ui@5unvXa|qa*CLL(J0CU081Zh zCpUu+1{X2ewPk-4Y+d%q0CD~If~|k|Ljd0N5z(D=`6mS(VH4k*C*_SHg*V+uz`wdP zR8fINj@YW=)(104vBO~4{y9e~;>yPIf&gOt?g|K)|4T#m_g^U!0B<0WjjjL;g)#;6 z9|Gq8QUsdj8xVv(N7_nEkFe)`GidG2fRDR7Q6O&^Ecek;Re}sx+l)@R07=zLPic2k zxnMM6>hbLaNYZ!Ri;TZGNBZZ%F%ALnM0wqW583tQSHj+x>9h5oXQ=leJ@ z77X8a(7f52b&+ zxNOj(M9Zvw^1|_0=pF$5XB-dnoaHsIMY!a?hcY>NO69BRz@atCuT=BAG#ob*?qeF> zaB5E5&rMv2 zYkx{6sniPfq=2j%o;pW+)2b(EQt_qa+?$wd{>w?~tVm_yYlVUe-Gbcat<TG6R9e$`>hI5Ib7y>sxsfpmwwesED#k4Pe=pgM78Hd>mv1HJZZZ zz*nPl-RfJ{qtx>*0GO4i3X3EMi(Q)|OZ&FJ{BvEzZnHMA&0MR&TkG*@e!rUB;NCBu zljV?D&HMALhcK!><}l~ZTN7HLYKutH5%Bm;Mm%#(b(s1uE(>46XOW2-gYNhllV8$76G@<%=3+Q#+zCA%wiYxvLBR0wv%~GD~Wkb%CCT9`+jn} zPs@bKpA)S0oMO$kbG+c4z9^&r_XuFu+(zO3Hc%?_t&*(=gj-)irnLe4SxpKo~8=YLs= zZcz4ey+vH|tH`wodFP*mx0V2F7_E2lx|cLz->)+#C_ep9UM@ifUO z8UM=<9~l70Efv=fHBh~Q92C-^m$K3qeyIYOPuJShk$#t3N{m-3mU$!q&6MFumbmuK z9;evj1>Qyha5SgGhMBMfpaVp}Zj8n^F;h_Z;kG#ScMwRYcJWaxne}Ka7E`2$2Dhxb z_rtwRgtMeh{U>*AlVu`1n2U)7YH0d3SlxV${p6WmJJL7<=Y4PZ2pW!`K;Q7ua_mjNTcZW)Eh}2}Mp6CPXe= zgSYNi;sKJ$ZJ*D@(NBkl{qzF73F?8Y*t5k9L%+LD0+f6p=O8nhu?KDp(e z>zRx0Q9e4u8O{o3a6pYqi`811eWw~{+Dv=fJ1U*Ys11SPWgNf`bF6X3;!3LeNb?I% z+-t{!W6zWr;ikqPnZRZ0m6VzlhU{;a+{$K8m*>My^MPs0_b9i~FdMv;oRWKU57iC+ z3baqLZXcQ*M*!`6>H>5CJe^64i9#f{W)l`kB$Y#D_AGI8C=qwEs8=WbrEIRQq7AK% z3DcEoTP{K6MPz(2CBggy-edjaE16I)lw>TkP4wawhr%I+kV&>j?fa?fdf6XLjK5xJ zmf+t~a~?ww033141bDP2)s8@0Gz3)p}(<@Qw+1Xg%`Xne@tn+l|1int4#hM`8zKoZIV00ny5yR;OolBa zrd);y6G?o0DV48blwZ(tspybmziv?S$$qCnHK9TcLy+uEM>YS@8)VXZAjP@#P*pW; z<>h5zZsBgO?FJJPwkg`%+iPlT9k}01r^Z{jyIdNuck+zB;~^XH-uFE}^g2Id9*%0a zW+pfMfypLAo9-j_5(!g|1CJ0)TYp`>iK|Q7OvmNtci6H`J}rN%I|;kYsq`@HZfdVJ z^ghI_vs+5~<-OO1IGkEapYg4hUZYT3rrq-{s%fN^9sIWKuraOe*!7scH3>a+T~8(b z8%WM}`-7iShOT*c^NYdXttvD3x?f)j{rRjS{wH1wG*vovWVm!S@V+h!Gt}TNzO>L~ z*C1Caa8r^>r`DTq_8RhX6V-2f2{tlz$-f7#^0|n8KmPPn37XGw=R?5tr$dWI3g5$d zz0+zmM>Voy5%A?rTIEri{qDw9c(7CaaQwOkpNUzCu(En1B`n)7^Ptt*()reLpqL~n z5FW(~shL>?Id>tLjeKyc{)QZl-+Q6oyf(DISgA%A!`WapEu$TktDx^}zwVUje|4yH1;H z>#g&#P7c4@(@;h|H$Ug@(}6AK@6vQP!O@5FHAm=m33taFa6$gFupWgH)q-3_tVBHM zZgy8Emk|isu=xPKUYxFQ*XWBNzy&Yl!rv!Nc4*~6uQwe!7mY&?M*~pES1JRy)M`%d zBffkaYU_ho7sw|y4h-$iK;|KIUcA2NqfLdDasB47CAU#mk2}qYb1k#s#5qXESb{I5 z26|Hi<#s9rVzRouzRO*jR?)OPFQjbbW0~UV2Sq8IPANhen>gY7+?jCsvK(HtD29@1 z%0#iLx>`JvNaGVa#y)o}3$1}^RXKJFH7#3N&&(*7I(?CE!4p0P{OU|$($zSN=bnx; z1d$I>nqIk?`U;c{VY%-hT2?-vhvmhHIPNn3nYoA_8tce`kVMR1Y-$;`KxPwIogWOx z>u#*QjgHKZ2iK!3NS${!nGLyK@!XnY@$WowmbngG;#q=33{k7m6cz6_6=$N~C_Z%D zfy&=X4^U{$k$!?3Qk{d$orSXS8Ump>{!fr7pCF)$K6(1LKLgg}RuZ(dlGT(dzaLLB z@AzH|t01_EUa>u?_h_T^d38Rwy+87@fz8FkuKCDnvOrAS9N`2 z>O8P>YQCCF`@Y+Bmd0n<YTZ!Y{}=%s=yqNQLOMRw8of@mMKV)0)alNv**^f zgPPs)&xmd6-O5|F&EEAdCPH4A9I$1PVUUZ@#VU&8_JUU%M>noam;lX-c?fPt-5M}Zc9hPx*k_@~l#96oza)_K#bt=C)mCJ=N$&~DtT-@K(B^Yr{Ztm>JW?~Hx39WH4vWPO88 zYG?i{^_4T`+t0b1Q8THUta`+bgE1W!XKyC+b(D_I0-9iNleq1^Wi6fe4QZv1?ytt= zd`1D7GwQ#LAa@Qa(kk#+&(FFXPW-ZIn9}525VU^&e$K-!6q{!qtYdyL8@|=D9mMdYw$j>uDkFp?xCHnrH%nPFPe0&@Zr7(JVT?< zHSHJMO{CFw5w1Y!94v++d2?o)TM=d$HJw#%vY$ho*fGCk2Kmluzxyh}&J-Jd=BG~r8M*|*wCX5SmQ6P!nT?Q3AmzEfa{ zqej>QKJSji;e>z$;^%h83USE2#7KH5A(GiGPl55yR9;<6p%nA^EjoMKIoSP1vXif1 zw=d}}w>=oDbOL8d$hiF|kTB)VDd<2~wcSZ&iSbCQkae0JS^7davv+yn;CTH+U@7uu z)oJybo;=MIt0|DEr|_XgUHf86Q#)qfvMAF8d2zYt`P{gzuTG~T{F1_o+~Xuizf+OK zNEc#|$F_9S3VDIMZz04nb-Y)2uagadYZ-W^v@Bxf>$Y7wHz~prC(1o0KW3%WZMS%@ z>{oT3mMy7|nPQR?3syJrW!t1A+v=VGJ6$l#nJk^)|GU#oltdv!Z1(fr&mq{0-daNz zWuW@1OUAIznP&lL+BmCi{wuj{uYI<0S}xA4Moz4ZfAR&+0uB^nGPZ&%t zg>DjMdM+lw7B-eXRQ(`$Pa$r0!%)bva~2YNvRkoWd*~bv85Dj`w@}DWzz?6kGzg4| z$4y|;DE%y~*Y<1kX324YtXyWU@A*7amdsV9%O@#jDgzTQ=*!PoWe#EI6XPga>?<*3 z+`aMaM!(+L_$27LZ@eHFMNZql3B(pwZ}>&yEB_f=?@bLyyRB#^d12?>O6l#~**A!5 z)5KODw#aiNu1%0(>(y#kB$E%&Vje`rd}a1;TmbtKycC3BXn;}o!Bx0gMOST<=k7KA zi12hCPv&5nfoGW+*KwA~a=l>{;(-v2JdF?YpC(BTdlR-8!tm>kD$j(dsC&U zFDL0XaQ*P0mKyoOIo^hWpigJ~6%ISE6CWb7$AVE}n#CD!WdWl>PG#7su8!z=>eaS# z%fA)9is*zXI(g{96y!RfOtuLG={I?v)I@TDD3=j;edG3T@NU7Abiu)Puekj@miA z_;Aa|AlLIt;MhjE{`l$h`X~}Z8*uFx%MUnUS79w~`(0o3eIJ*l(oC0J@8fG&p|8=e zG%jkH)5(f~!!7PtFSeuwaHL|rKH;|BtwcUI4NY^}#B$k(T1HkD~Lmf8j@+&NlA=*4adtfAzI++~KzKRq}|$zXb*V z9v{e|1)PQ2IvvFOXGB~bO4*NnmdEdW0lu4(b&fd6KvW?Q$IbhMz0BK}t`F@^$IR)( zwN@(g%*3nS*mQjOI`~5~IJmu743GZle$PkXNFF85?TyVD>_RGN6CaC`92CH-DM++F zn5ytqu2C_Ebd0|U3=b>js5)cN*fDzU^TzWYI5JAp*UDnV)8c~N7hJXK)|CWEHoKwQ zRVUNb(oyadIJ8GhHf`-c2Xw~lC~Ea{m=dJ5DBR3$J`-a-BrBj|+D}UN+}RTE8EX=& zGuWo|VuXte2Nkw^B4hr%#yLcr6-oKHvM=`*6CHB)v6g5&-T9uj(qrJVA_HdC4k8GF z$4V2maO8}M+%=zzk_*12$95H5^}hO9V}6jqM;%|;QCRXSjFO!ngp;UsoZnWMLZ2kK zF3c3uL=jKB7{aQ?0^o)%7C7gY%9>p5tl4a}$9C?&4epmHwpH@{Igz8k6%mAiiIs0p zxTEN?LKj}y8*Y2)j-F?Tem^K6AY6Z2EK0G{A)7zxHpi{S?PzFlzutDV+gA9CV@Q@M zZU8I@>q(TMk$1DRg7smhr8YqfH%q^ZPE?Bt5mo9kPtW6&@YdT{2n`F@W#0PnpYk&K z$jg-L9Xrt%Gtjj4-o}+_Iz9~wIwLalnR40pY%^W&_}P6p?PJJS5C>=n`nAHcA7-m8 zvbqN&dR%9CoM91rTA$~qZTVY#nNO_;@y<;$iJ0P&To1>`JmDqjncTQg>luEbckd11 z@I%jeREnfz3*u!Z(FA;$(b2Iv!rgx}OR!fgjgo#S6R53H>C#?RDbLU21jQHLzs zZ7gL{4FBm!EjHS74@WK@WY!sH|8$BxLpv`r^%!I{&e08|-OgKAx-4(r^F|9)-%*?6 zDns?8J0&i>#n&2R`g86fg$6`00lAzrq6HVuah2Kd`P88n^RuSpU(m0U%5{PZeHQ|L zm8%HMBv&~MgkEgJ_$^|cjNCYih|U*q`OG7|+JPoNUKu2Chcd6TY+$PxK1#5|%<)SD z5zMbXpY9QVa8F$qeqYyP6xdsE&+lHgc97kUwjFuNt>o%+VK;TR(DcB;d-cK6{iw}C zb%fJ}6tnqv2ZO+|KBql-*a_~D_a*s9@>F6pBRQ2#Xm%?qp%@0Te|5w-3Bvv-VS0k2hzDmI3HF8NC~yRT=oh$3-tyDz`_L6x4wlj)ypx{K(1Shvr2s!HyEPBC<8@T?0Z+Bzr2r&+vc)wip8e&YpYs>`Mp0n3}z$P z=GvL*vp1&{_FblBXPoaBp3qQjMbx`Dem~ajpz*SK2IQZN=zKZ}>Wy~pbnh;lzs|=o z$OT5bOAI%fpT;Q*jPHBG6;MT90Ukaa*08UCkm^0Qv~H@3dgc6^lJdDhgLTFW47Brw zvsCjnL&X9kxz9qdFY-~o7CU$OVz|q2;e@;Mb304WIn-`Ntlb&*Fqs2Wd&eZzT)vzJ zJrETW3eDR!Q?<0NbspWY=fr&S=ZW-j2dOYhc+-^iAIDq^!PnYmP`1{~n`tqbD;FS4 z7$!3CFl>JQd>fjk{P5}ex$63O|J7V2|D%>CC7|Gi$KSC@m()W{x%WhXgkA|W5YIBJd0)ERq&yDiK# zKf+=^27bw_CXePtxqZz0lP?4My%B?W@B2WPt~-l8&kyA&^2G{_hcD0hd|0+#|Kq{^ zFCR@C>;HhYIVKc)WuPu;pMlcd)R}6j^a9L_U<0zY*v_Cv&}<1#(Brl_qluMH5#C`! z$rda>x0WW;TM9@0z`k1j4u>EQ*LG6cG#nSLtlX#b=5yB1cA4VC@F4r=>_e6>*|#g5 z_Vxnn1#7aZ$#qW^sH%`0JZCQtmN^-|KYCX+;Sun(nsBD)4{eHUh&Mc8W&TrySi5P+ zOdlS&$Jia}sQ=+fw_D6;YzgVyO*0d_hbR5bxk^j%NyWF)p?%}J^T%{7KUujl)fzlIQYxtyS zqk|M-It_kOp&s3?%{IB$)>WN-ETEf~X3&5iaSWR4;4j5k0-9!df6$Kz=m+-q5UlXe zXT?cU!S+UUbLxnqY^8p|%v6rzR8q5NxXyS&%Iea9-`hj=J21N5o8(-}YaFIQCDXaa z=lw0iw=Ym1N#^lQjp^EGDrfuDK<&s8k+{2nwp?c9o5Od8_}ZcEJ|=yO+-sMrfgoA; zR}{~vNTR66Iq$s2`1Qu@g3;_CC7}ZPR3Efo(eg|SmQ@b7fmWYwzpJu@^ebFMR&N&| zw|FSq`(cP^%ttx1ecmA~Qtp_<^gmg^h3=koRewf8?zC^6bUB_%&nIc4d_JnoQ{J_*}5=b|$ve5LTf5e|{Gm--xLfmY=ELn>;i@Dq?Q zZ6X?J6EP%UtV*+ihzy0ZRrzA-Pt?t(9dpfj3B%Vpg8!e46as4er>CCb5GmPtI1eB_D zQHoUQolpc+y7UgxJ4g+LAOa#PO_1IM=~Y_j2#C~x^iHINo&*RzlyA9Dd7gdtc3=DZ z`M&S@mkU<1)|z9?G464XGRK>?j0OCbm|E@+6zrI-%Wo|bCDcA>o4Uh6!bT3LckH-p ziTK{kHkRyrKxloM9LV7#+u<=;wwc3=h0Bwzkju~;$Gka0ivIF=*{WQYLVW7ecBg zgjWiqX9;K1y6$CZOP>i+KhgOpjiD15p@dTqAtE_>99QL5*c>{6S!*j`FWNfz4o}M)@tTlZoFx{ zi|k;m`NMDt?SJR19(YI|M} zL=pLs7LuHQHe(nsEA!PYT1akmLOK_@)wf?|qv~2f(-{^x7RrjLb3uDvA)$N;*Ggq& z7Zpi~e;!4qs(jc#vgbRy+^HjNm6WAq95}0^__5Jv%DvM~Yq8&dY|_tLc{l#BBnS`F zf_0c_9Bj6LGKI^TIch|~RCa+Ckndr18bOdJp+m{l3hklCN4CS+MoPmc>#VYa4+9Lx zx3xD^OCd8fo}YZMlnEl&pMvoyt;};X4;%_l`8CP4X3bD<4_q{x&P=rkuZCxBZ(_a7 z5&_lt;ymPf?^O4R)iAmak@HAX)RSOs?x}MTZ$^eaH1O^H`iNsINyp)R;mRL_p)NP( z3Sp+Y{7ceFBktF0avf$hXc_w4oK_GQEpYMRaA4O1&}yYWmp?L%Be0KE5|1 z8)wU0-9KphUFMZs5!bL&I819ip0#fM@z4x7C_pe@mdZ6VlQ*w`I##GGC|-Imf-$wQ z{D{Vhnkq?0*lAQ9G@gU^m=&@iR-|!qg->C3Fq18Ou1B56-vf!|yZ;W7(LIK(V?x<5 z{tr3(&p1rR{a*B=6CQ^wSSfLv&nvV>2u77@PgA>27b#ofh9IQhP(3O_hBBv!gNwx~ zm+hfiecGllGK(c4?Af}|GQchiAttuo88X(WTtoDHo;iHuXi}~DQ5wPp2UYJHB;@H} z0ecS~>`viHcP5#aNP5IE@BsX}yNw|*>DdiTYWoJD_fvC|Z}KXa`+UIcW&zueD>q7W zW?e9)eh6cI48KTO>gS!8KZ7(DQ99GLdkKs}Nh=LSU(YmG{lFZrY^8?N-11kWsqgy~P_`=T4b)CtRiDXic_vOnJWTs$y zYqA|o9j@Q_=oyyg-jAph_{K-e5RWU42AT?yAoQ(-jD3K9zAENa&FxfBuC_9Xx*N(K zhg}HuXb4b`ias3TvnZOewGFNlm+SKI$!#DuT zAV~rNMm<}n(}Y3D=H=k|R`o|&@k%@wo&7=u>v-}{E@wa5TV3u;$v0x_Zb7#f_*QTbP zrpmFYpeue+O58D9X`3dA{a7(SNc^8*mLwVNlgNsRryhPv)8Sk6EKk2!hhX&GOkaG* z-ye@hE)nVw@Vc#AqB(iFycyg+`%irzEo-r+7~2CV&v2LLbp=T7)1~*e#--l@b++L1 zpG&>HTx&`0e>oKNh?;F)LhJf-lk%=N4}mvPamg@+CE2KYv^nmhMJ58S=}v6u6*HmR zm)Xdg+n?n>XXX@n&X1iBCYAcsXt48DQcsYas}1$Y`((AIG;KP0qu3{rOREj`@Mh>h z!Y!o7ry?l0gVU-IFYTe`rF+m>Uk?dtDpOpDYI;{EquJyuOPj~24YnSCxAmBjB$zbz}WHXkN&4r_Q_@-{7xzD$? zykvW==WSS}>E(p>j~O309%6b<=Cs0(LtVo_x*WrXLXL^|V>|H_VgVKW)Hn=qCn5t< z=TDhl;L<*r)hUgW^phjXdY1$zo08354SFMe&OML44(TbJl`4=|JTN{7-q7jf{C=lX zK+Ogau?4%tAU8{8>SiyfA3`lr5Y9ec^V}~?)dD?G81RZ#BkPR5rZEI_b~!8sU@P2$ zqbQ`9lk8<4rD17mndgIopKRaZ-Ox@n8_xO=abZW!unl&#-Fssa(Fe9`rX8Vjsl3W| zzX$HpfLeZ|%#RT#iNI$V<_yLn7vd6^$ z+vCT1s-OYUp~yx>z|mpJUUn1%e=Y&oJCmnd~G z)6Q-EHt`lmmYQ8PV#P-RtRtfhx(+Lj-F$1D^mq~TVCNc%k?Av+ ztd;QfM+Xer^aZa!b0j#=0}M@*1~Sf&2Y*k2I(>zi+^z}MsP7yJXy!s5S&AvxD-^41 z{OqpRYI^M26v&~8u04kIDR2n+7Rm#~0NIjVUSspPdgIVrF^}+wbKZ1e0>3C8f37)c zm0Hg7B~JMxVFqDo9hJmf%;{*U3M%ydKZp5)Uoci?lL;POW>tKIvX|{k-+@ zM5;hzl`uTL@7}_bK5oJ`pF>HCI>i{Pc;3KS$-P|<*{V2zcolrhkhaLIb+`j2K+Geg zX|Cs=%!WG^NX0*8>q4hV54sOGQde@)8O}-8vD<#h80)O`=x_PTP6{~`dC!}$UYTjq ziME+LQJQt3@f^!M3*{jTD?NhZP>XV3K%yR{btV30dWv^&02qyblJR)uXfvvv;-8t= zf0M3E7lKsrvKD+a$(p>Vt|3xNtqnb{A?mi%{O5BN8Px-7vD9XMqg|Ftxm$0P7`?1Y z)4sh+u~M9I&okoP)?&7klql_4Q)C>=erdZ!&K$6y>wYR1#OP|z6%sf(zSD||{= zsUCiXk;II_B{)Ox+IV(kiH)$!NqZ>u_yuBvTzz(>qUY7$$O!{z->b%kw|DQqg*nH- zry;{|)GR9_+W4aEO_WdUzV5@Q% zPhBrp!1-`*X7b=6DK%G^hqzKGh814wQ$V(l2n1D{qXYUKg(tRw!U%GybHvPg5>@kX z6aMH|ZR!0YW7$yGy$%EvH;a5Kg&NV))_ICmSq$CiS3caw*}9qW z&hit#s?H|icuqbUkcMm|3g&UOB1Ri*_YAqke8MKx>uI49!f9gH&%)=p3q?Nos}73B zWgF!_z!HjNm?;C%*E;;CQu0Z7qUV`i+0VTynzsi)=K`H{frYdb1|uPut)o)deDiNv z!T>I|7XKD$6V+i(nkx!BWX&s}lSKRJk13o#g`s^Tnc&=X^o0m9BzA98>I-{C2=gj? zNP{{z`P3-f;erFDMD6AQLWD~HaO_8mrJ5-Gw7<}6c?~E;nRnUJ+|~C{ZyDHa3dlE# z*rLyA=l^|q0HBnaZr6Ob8}h^FvO>4gKKfssmk@abF6s7HJ&G_hJ9~Hgf$t|GBYJhw zZU0^g1yc^ffLnVEd)7;oTo`kM5f90j+LDUyno-H#H#v?V5yJ6C^SRTYgtUEZZq&h! z=uWz0$K%5>F`Jh`wjb=0RSwUP31$(JjVw?1)+Rfaz0SYRcwqW!FWK?WPlqWb)P920 z_YmD4h&Q~mJhfripsJ~Uu2R0mn83m>-@N07JO4t{+Pl4$>D}|%|&j+c* zs~nMXI=#Vqm*BO7^6;aTw6r4U=JO}f-+#RHW$dhOk$!i6I6P2_kv!`vPV^%Sx9aYP zzYVE}DpSEltb5UnT6t!xF7kOCkQy!y2&0#iSErEpgKwCkf{^dhr_a9*sJs%_n@y_4 z*Sgu|Sq#K-(u6OV9UHg$=q# z;WD@e6J3oc*HCS}-OC1z7j0aEPqX7V)Vca^UzgAKP!GNTN7dXSV9#2-&!%_(c?VP8Ok6x@Xa?ooU=e_8Q6Nmg99tYP2D<#>CIW+jqkj3#|hS{rM8cM%O~36vjETXF+)K0WacVhgfxx z&65utloZ`~Z-3SV3EFw)I4bQ*@8J)fa>=5Qdhsc-g>)FnvZf2^Ds=m}72EYtUu1bg zSdt`)LQNYS0z*i{@Er~fAK!{#lR@~ObFH-Ids{Jg!qy3tUgi4@g)ayBvq`8MIAiy> z*)o!ecpNrEgiGEMB21U;CVnsrc)F<_YPXc-)}KX`*`eE(9mLYNxTBQ8rix9$I;QbX zOscHtR_!Ylfd0~GGv9g28Z`uMYCx{DP30O_2e>jWNa9t`plbGiR`&?e>i>ftZnP)D zu2+0;`95+IhK39IK9MBVsqf^Je@H(SX%hB7}4ZWaa;O%s#qHd z9xZ>SkBZvv!!}rY5+FyPxm5KDEYSIV8?kG?F4e!sH7LNM^E0#aq-DfIZzW%rg6w)m zMS2Ad9ka6adE*>+!HlDLyt=~2Bi{kGlJOsS$ga690B|>}4i4pwI~$_7KGd*Ej60oC zVVy^`FV7XG>2`vjZ-(RxU( z;1<1h`spA}M8Qe}6m$Q2U0aM2T*+Y2@3cm@6rF!0UA#AlnDL_;YpE&#<4`(0$06yV zgBS4u2l6uF`}fI{gBDuj(JPeA)h*0|_eWzQvu#~Rc8wknxPlXVZ!~9z=)d{{QzpaI zf_I}CtIQ<}4#6CnH|jH8R2C!IlRXc$1)C&EOm5ctmYRxNkd&&F2`e%3{;4x-AoMROE!(e4o z0G^XDB9ND62!FMCGK-u@bw{XVN0s!1`2p!+#b<64{2k1xu$@Ma0b%+QSoO}yIuwda z^MMB(3p?n`tog&b=e@pD6Zu|?2!PWowb)C*uJps7z?C;nbk2EGaN=FOicE-Y3Zg9gf+swABncj>yX&iX{ZWoNo5L#3|2Gr$9` z?^V_tctDUihn1SdKJ?1rvVljht6R{k*6 zO}y{hMRFLUR8#B|)xTV5lhCfyrEqOJNEXA?%L@mPgtTb%^z z0^l;%%>-JJ%qUAeCyNba`^0(Tc1oMaX$``7|TMS#{pH^2n$UHlWTnjgz|>WJMD_cv8s9bEQ*F z0%fOLVu*SUDm(ijUx{hhugaEsIre!p zwwIAd%)ye@^R&G*2}fR0oBDNlZn!t#oR<-u*AQ8+WdBf3!4yq$F$Cke8oIA=NYZ)@ z-XqYUI<0E@sQck%6f33;sMo!9jlvMt$6^6M|?S*CwS}g0(^x9;>gIjpo673d9bxV(*{sKuzj6UU*!G0-0zcw zM<`I~dNwx1{q8%t?r)VUpKnQdXOX`?<)heva7tWw*cK}iu$kdKF6vW;^C~daGL!yb z53uYnNoP4g*t7d;-&LzHKnA2Cb)U?_iiVwOP%Bv0iY?u(7&meDeMV^^m+jHL_ed|7 zov$rJel4Y@P2cF10+$HRI~i$b_hN7OV0Hz{0*2FCo=KqMY!Q_UM%|7$nH9ADs{GNhU6m=r$hM3(MksN11;Am zs>VC$^(W$gaDAj_Ggw{&_uA?P{A#t!tu=@E=d|hLjzs(V(>6U;O|6ye zgA$bmPkUTxY(k~T=V5`hM9avXq8UK|s)2wI5X5#JN59(`omuL*sM>OoV#pA)cLD%4 z9eRN>XaRqTq3H_qV>@XFFe+nn_I*#&a#Y?%Hco;0lSpXT`6$T+F$H@0)$P!BJ(Eg?L>euw!AhpwXp`^Q2Ihfk*;Aupg-_A7q5Miy)Z99>^pCAv$ zd_3xKWu7eT`ubv{tH_5gDK>S1fD2l;tif}Bv0I~`12|W%@LW_Es}r~n*%0kPNcpa^ zAAB4L%#0Ite?CeM&%b#YsN}}*P}lTODa}imT%7_X@j%=esNc8Qx5GewM9;vN+f0* z@RaR0ew}xdD(L4-ZE1A1CR8YHB_$J!O~Nw`t=CPoah*~+QjqX*vOFfGng7ZI{w!T% zp3A$C8!r7W@3{+&8-HJo*M8~|Uz>T|bNu9+^0*Z8NYJ;?t_U{B#t*V-;4oH~d%_N1 zf<=h;HB7a^mpQ+da;0yT?9s6EoBz0$kR{to1{?`~s<1+t@CHzDLcgDmDCaJuD8Nr~ z!=Q+Q94Zq$)+_iZ`)SUbpT&N9 zyg56Nmtb-HnZh4){p{;J+LZ0xgEN|MYq~G=A`74Wjr{b!$y*>nP~-T0G@p;$gaV%c zH@B-I)uTxE^i!zaS6Va0^zVLU(oPD{ma-vC2)<82aU`G9p+0s=OTFQu1vKzf*Tqi$ z4Xjer5E4Sx!Y!^t5v#;#w3&W+0s1}2+62c_RX4`+30C`2b*c>(c@(ra;c8iyakcUY zF=kjsXF)S__$q*5{kZ2wGyT=Ge;3&Q53Lh@6)^sSKSF%~<3HTh*C-u%f45MMZ~gMg zoiI8&5k`BPNGhYZc$w<;Gf6xV&-YucX@ zFbj)=l}T-ijah0wPKJ`R&#Wp)7@uQ{UouVJsqTP(J z{-r5cSmIm+W8g`_(NKEcamdqP3hUtubWI22W#@BqX4cp}6-uI0m&qx=M;kD?EXNXC z^i*|55+n4ad$#2nYmgLq?#y&qV|`l=O&um(^`(P z|Bn>^rv`)Urp%$*L)MIqo6T5)cMHF>XJ2gCw$}18ZYyX!zL)R#r9E87u|66_Ol!RK ze9m=W@wv<&l6(NR*SK)$8sf?eDp%?djIXx-?{DfvfVcEm+My3mz+y-+!664IwO-6+ zyajSEyv^@a<+uZ!SFW^yU{VKu9L)TrL`M5i&ADtdZpvpMtx}7x zTJCo+MdAI6r#Fp98*_dVvc@=d$$9enxb4=?rfAoh_Jc4^0#@d(T7}vLpO6#R{!9op_kh>BT~E_;zMXJ{Yzef=uX(Ab z<+p-@M^3xyZ4pk8|BhZ|ufdCJ$0gaC2@2Q_2)VXY_7t0)HxYrM`l0fw9qpWn4tc3o zG$Un3P6HX_!GDamG7rgQ4*P#pe>gmy#spHnS%FD50qJjvhnVZ8>V+@=*B4~m*_iGa zpV!R!Z7e_wQyKq2UR=1;k$R!chp**JD#M@r^A9o%T2Q+%-%x92(NFrjSNhM6o`0^A z3?JKzEt5Xp2YJ8OqdzfBKUF%jmKLS zFQeN36Qbo$lqo1nF>~S?LV;Kk^~V~f|G3+4>K?=$CG*h_;tCeEyZLu^{Kab&Lcm5) zK)o3>{hKd!=fxI%aPXDCcQw1i&s{l1nceiawi(k6c>tbY#V7iES8I#_mY=$sQ_=sA zH2{D13rw#7Yj~Z6>Tg*Hnaq5^`|>#Gz1#o2|N66E{wX_|>wt~8MaT7La`~f8k~)B7nz z01K(v@T`jZ@A!!ec+UVib^b8F`1h`ULjx>imUo98@8475uQy~4BU@W9{k^OItMI?t zy8l)9fA63F+%*4H_+M?^|62H8`h)*k_%L^9|*I za5_E4e%4COJ0M;e<0mfCQ@u&r!4#eNB{d;FDg13bO~kFFsddA)bT-8fs$V&Q3ztHl zTu95**!FPyc}Yr|jIdXu9A$H!^j>Nbc7gw{I%b;2v}&^T`IiM{^B<{K`mqi31v;wO zsXS@~IkNB9*owm9(kh<9;liNj&mX>WFPfXaOd}MN!m1+R*mBx`-F5COoyS(tm51#v zH@Z}Rl&FJ5PZ`fu*m@^_ts?AMc;5-$tcGS!{?y+g zrhh&$7DwXoD1CF9g;;QtFz67kA1&79?AvG?c6dequi)127B*CoiB_DU&F73J}8XZ9rixSoX{SrL|y?+Q`$$KWw{glA^e__2e z`F;CFv&gn}Y2wW;#jyJ}uSMzqQk&)*{-sU-_rLTP(VXB57alM?ekiN`56J;i|9@!# z{;zmxRGZj11{hjZ3_FjMTz%1Ie?bX;Ehr-J%wD(TbcW8~4qOQzGqU}p#&O^HAKT&X z_-+DmOA_^*fpzmC8XpUf0wvL(Ri;fAdD?rlS%f9`gvALMIaljgky2&j0hjn=kCUT< zNAnmx;_jO`HZtD7d&!;N%_MzTU^WOT$hU& zp@N)xfw!CrpMHKh=^8Z~jD(9}v9E}ik6^l2!r~$|ZOVRD7(Kvl>md?#GN4qcPa-Ti zl1!985XNJZ!AnTw^8RfBz@_3c3N-89-v)cUZS?LQ-{)2!?JgQzo8j!$E8|0)Y-yQ> zy`D7vxOukx`Z|f=_W(Ee4^7}itq7ZpS#P0Ae;?r-?}|-c>1@zOJH@xf zR_1?b+ju%B@D8AguFhJ_5fiKs9s7t_@Au98cmXU6!3CDY$1A4xJ}YsF7nuo2ZdM)_ zdN?GF?eV@dNa$}{t&2+br{X0pT=g17=DJkJzC>!ov+oIOICUDr?Pt zcy7S-n=zj@%^x3wSkCEmX{mtt?nr5bZ&1=%uv9Z&y3glyp^CquV``W>~Mn@J5OzP8~Nhew8Cgnayp!csO;I9*(IZ%U};+VS8uQXXGHbC z5Mk0!Cd4g%?R8QCosPv17C*m~iP9Xg?JYR%O%aoMT%hpRBV|mpD@B&++*Xdy)L@31 zVy@8}hWF#$pFGM!LNO&JtH!7KcXz~73j*3cU)kwW)BVgW$i8`;)JYGZLjM6P1#%bp zL$%kr8p2lcot%Sh2j4*Ff_l4*F48!g05{nWKISrSZTP^Fak%-OC0}(Wdu>NtuG$_pMsRIRV$-Sj^p%Y^ z0h-4{4kG}hH1__qcTG<^{Na9UJuiI9wOI%#Dw^DETqLn+ne`Fr)Md{oKieBCg1}eI z@J4k&$B%%Hsl~EJ*c9Ywmel6JVSTU{#f*V%fm-H6I$23G0KpxeF%*)io;Rgtdt6abJS0ow} z_a&g13yKPkdY6i}!l6H>h*kBSxtKC^zL9A-%QI6+YTAFG!=d|t?55Ud(qfnLsdeH) zy%+~h292>3+XiP<=+{1Cp!7!sbUXsV0^LXq3QbZY>(`N+2WR;K+p`BYsBuKHkCQ)} z)bYb|1F;m+%Kl(}d-|WIXJXm*40C=qfI2vWZNQ^iTpswx&vjHASbQUTw&n*31je2% zr!YdFhXDtxJFDmvT?Y4tu*SY&*qLSwOcWV|Y?3-I36*#|HI3@Fii2X!q`qNR@{(%x zASTe{$gL8{RN1s=``rFqfL@;a5pq=PIp9Jbg%lRt9c+=S_ zZ;L+L_-SgqUj2v~TI1+ubKoIT&xbeA8CTj?gmK-gCMN|6e*DJ9?HW3Ivx(lMId!{J zrw6IteEe@>5XV4zU1oo6Y3) zZX&U_tjJMEUb>S@v48qT@jPk}f2m@-9j=UNv_0-mwD+`prJBAxnU2wr9Vv%F z?7u8{c1J~XDu6ck6g@8dg^znJ<4)epiylN28=W^byrjtgGCwxOb|V3HlZrqmK9@Qly7QGBcgyTCE1J?w&y!}XV`pQVTWJVeze^yYsdcS})Gl*C<-l$Sh3iXC7%#b>fwM<6utk$H#!|x9}X>REMWGn}r@(ZCxgv zl*Ya~+Q0-tgj@<+84^-W;WNH5qlY!K!Zy<|T{qBqO>=AgwzDfg4)Zx|O>o=guF_;V zeZ_fm?!l8C-=pLG4I}(<6pzSY}bzoDEbH{ZO@O#5;;xI-GBXeq?efG)$mxf~n0cfinn;$+W zSzoIA=Ag|gv7SWhEOof|41u#zvTiopat{dI>-cHnbauMoCNnfGL(SJJrf%bN2b2%3 zo&ER~Vq(U3V8`uFgtViyZ`>ezvej6oqR?=4cMhLI*2>L*@0}F=3vz({p-NF6{a+sB zKR!O83tHfToF7kPmJ7z`RD4Zn(=}&3>l!UeQDi+Xbl|)r>vq?=8vD?PFdjX@s!Xn9wCDbLwOl`2#!>K9A59wzvjf>rmHMW(#eOL0d z!(n3RdiZ?^^~2b1Wzsbv)LEh|sa}TBbXEP~%R~=Vr6^4Wz;HaHp)h!)bf~LycSX|4 zlfFsnd$zoB8O9DTn6?_t7|-hk{d$ntq>Vfmb(?<_IdyR!e&U7G=s9l3;#Vc0!S z7=yr|pUaB(H7|)%oD#!$>=c4-(Tkq48Q*lg?_KI?c|`7shwr$#>8XnvhKJ{=NBIuo zHaJOykzahixw$bFnBQEdz;jbN4ZG##8fh==*+|9-;We%q+D-H`bDO>Ar+lK;DY(8q z_0jSrhv52?OE5V;h=Y<1*XL~c#;`gv53+G_p4*S%8Y2C=*01ftjcerW{cQ)OIZti}xFY<=hx#w7 zCbvpt6btpvA%K}JTyZaTG)yYO)Ype=s%mGd?<@tCD-uayFTQW=6wRJwAg9qZjpgK(1oE@1DACTy+D)xbv z%15N*Uf0=f54$wol7jfden6GK8At%8wX|UXlC$$tx@0GYXRpk&ZUuE(8blp*832Etj5p?r5R`%?t=u9nbrpBs0SofIAYJ;KkqFPR- zLJdJfib{v`Y7Iu0w#8jebdNQY0Nt2Pq(nWMq2b=SnLqUx^ zw^Zs^<~L^Pv_de~pX#n?vhC!?+$LO$iM((8>a3)uZD)UfzT^Z!*zrSEWCLvChyU^B zy0%#Vb-TenSK-0PqDb&ciO9E*yW>Q1^b;wH9y9hz&%pkeN-xEas-Bj;98)(IX$q>R z`*R>Sv0c5rsu^pfc)jBjHx)&~6mNMYy1~H95wKbBrbOF`XqFza1=X}2Ly4h>o$pcB zW-d{ro40+-lm;%(Psdsk{sr2!xYS{AI0oh*UY#&m@$nhXXDe+~Z(lQKDSy1X zEBx7{H!*tEkXLr=WobaL#@G%GG!y8!zX|^V9Upmtszf$)kjVOB;IhQatO$ zLNP%1T=;JE>Ry2hKHa;j^g1%L_V`kU*l>TAnRH*Q#k>)?J4b@0^76qXry}m@BEyd2REI(V|ND-R81sHV%km&T;B-x(Ikh zq?)Vou(XSq_uXx}Sd3Jk5$-{PIPdvPQ=W5J&nJI|Gtm@Fpo3Vp z=dP3QMP$qV{^|<`#q`0X2S}2`$ybcw%V)Z?u*e3^TQ{>}G=UI`l;k$5mT*2(gT!L` za3y28m_#brn9iYc3ay&teC2urKvE=@TKltC_;tf`fnNUnOz2TKR zU)xs#PPg2btA!Y^6fMoJ)2}Ug(d<{L?WUdATUe&~)Rf)hg5bQ8d+w`0GuQTvcoiQn zQXyI01eA9w^|H4)9i2W8kui(PmBKV59Pl}3IKl&Vc=S?sD*oXV!t=t4D?cI|CW44e zXS8{C33R{eKsU>+<)XTGo5)-;GNK4SLj$txLo0A8`2L>nCg~_Gv+D;rXoWH-GbLUI zWF0({?5BCaoPs;)wfPcI861FAZxEj%+aWz#DBDxl<2VAL0u!T%JRHZo@hoHI3MDA zccstuYa};OwKP|}pX9u0^#%>wF9=ow5Nc*&%-i#Og@g!_!*#^g?S|TEW&kQbNxDu6 ze(1NLnw=et#>};DGTO>~!RWqd`8588$us=R$gDF@=flyqWw4S;F5@C&!Q=> zXXu|9YE;eU?DbJr4_u4S&hnOw!j$80N(0=3jUt}9*(B}QVoizH^<8hD6=5FFN;5UV z+0-^40OsX03@yK}w>fzk8sviu^2b=`de*aiGeTrIvhF52&2cuTCEaKatg>hx*cj?O zIUN9%PTAdCa7*-1Ao~90^_EN!`${=_Y151~>IXEElRk3=xeRV;aFDCSc({AfxB*n_Q z&Gz*wa}=mMC&pViwiKHM$@ilfYtBET0jEgJYJ^bXsMqpHc2r-NkXG}H9AS#^5SP%K z2Cmp=os~fIyjv$R2J~znvQd)0I9ft?xwn&)RCwHJTIkwnW-3z;@q0$Crb1B5*kK=y z;?XzjZU&j75|OkWfUI7t3Jds)q6?;Y8F#8?8O?;Yq}i+7o+Rv+dLSpVruijZEL|Rh4p-) zs^EwZ2vOeYe@TN(ijg2c!Xcf?1QjpvvD0eDG<^C|13wqUwPn_ur^zL&!jN}f+*EYe zO^W8mgEF)6%;cZKppWkKnaE0QALpR6H|T^yM>4lcTN@>}ieT#$uQ3+$xXUSRhlI2* zC0nTWU7`pP^V0OCdZ!Af-;6{lJ$BP;$8)u%7cwtH1|{>7M5820-?#oG0G|{CeQx$( zHGk#(t!?X_yH_F4?lzNBGp{tIv= zd8x0Qx`D)>e)jDV==hoTnvI z74qe0z?jCw=5gbjTw%7-HO$`G&GG=2n#Y;YN+0jvuIx+ZZNA^* zqu%JYsWA#IPR7UTT2>1Y<2`N!$1~%A} z8akmn`t(%IAe?uT{fOF83Bt5Z9s*c*WuI| zm?CBI65})TY>h3jfhI+B4nfm?t5R#hV~M(=0M6`@YG4Y^KR`&XFE)v+`TQ8inAdwB zR<}8Y9S{})YU~N}SmV2jAO~-}>1wR%weUG}L6!<$Npu=!G!kE(=qJslJx|t4ByJkH-5ClPILVPf za_WQryb<+M{O8!8Tm-doN~4@3`+XrUD`==N-8Sa>O$EEv(gqd=4BXH$z;N@c6Jw~QArH!^>!(_5*Gyh|B@edCk`lNM zH_SLh0%3kj;*j9vW%e9GerKy1PLwdJ$+(`d$E+8QdF1JC%l*7b7fJL+!WkltNl zGlrb$hYz)!-@a)B^nQ0n$=P!%hhtggrct)oTP)c+5buW#B)c-`AE!#ZYEuFEbbc|; zLbj-~f4}D!DNu}lZY3=TvNehgU%&I_*7y3eE6dA+~k&lQ*~XRkOekV*B4y@9hi{(V(NJbTK2bs!^P@4bRBhY4>i^Z4UJ= zGv+&DZ#+~LyF_7W7Qd#kQYb(}BY#JOf{?|Du)Q?KlxDj+iWsMn*Wz3x38%Zg80do{r^syp zsU?Mz7EiL1~aMKjSsOCzJRH$^8I8g{mE2+HY+jM{Y-4IZ|~Gq zBau>w4H-E*!ROlDbuBPx43Xg%?FZpRE<@Ayx)8wj@H;a z*>EGq!1Wrv3X+d+$mj<=j0@s~m=wT$$ZZriIe;&n2GHM!PqxR+N)6Xs4-6GG*`}48 zcGskZ39H8#U8V<1)^e81E39Xve`3lEI3N5sp4-)=ia`q_K$K8&y);8n)x=;$MXThg zg8wjsx$xOh9<31HX?PBR$W!FBHmlnAZ2(3?aT{k%_R_ItgXY?|l?+H0ydfqQJEg}! zY!pL0Z^YZvdrU$s-oXQOaU0_4^lMRm^Cb19 zNd;afe6nD8NN>KSI?D4c5wg@7u-a42bzHRcHXhY!i@&S^yTML?M`&~i1&`u#_f{YC zPiN7@h#Ui=F7LF)&|+$ops-v zdpsa&hiICuIAfpL6$)08z!z)b=YX`Fb{iVm#342%=`^Ci?vs!p~WBTj9Ho7%IrGfd(kTPHM$S*6s5Azx}Yb}2i!|J4(o3=`X z71DS+mV+4RRqczWeTf{ouMAr9yO$l-+)I8H>I&ZD?2$Fy&v1qbaL^9KCCnj+gq#!I z7;{e!3BwYBc&cNXr2zd2w=tvEWMXB5D#o4GuqGrQ3ws5L>>(_K;bl=%#d{V}uH4EL z38Ltdq)|fGnqc#`L8>-u&n9sp7jqhbq(i+Ym^Y%_Yw?LsvK{8}chtrTt~4HJpq?|U zY=A#OmS~Lc3=7x{X3l+cC535DWBnPXY&{Hq8W~>5`N2Vi;>X8hkZtRpam{BiQ*9cB zo5ILN4O5aN7vCnjy|!ZnkQi7+9%T_1l}2CoyWd@L2U0XgJ!fh?6_r7=#5*qDWEE$7 zVlDApmGq+P&hxe3hzT;ZPHiK2hBumM;0sXs4R>3cni4R9CHY2MlyvovJJ>48q{_G|nmUPT1r(HLkSj;YX4~0-9;sZCxv|+!$FYFhuUDSfHvZU0tND(R`~)W_I9KZMXC3RL!QPuy zr~Atk8|4lC;w@0~=^Hiid;KB+wZ_@VQG$J4z}`!A4PkKU1jiVdEPWt(Bjogv_pK(# zJ->}m%4%fMcp}LnriR|u(lgem)VMiSY;rDfr#9aDZYQj0;%rb}=5FW-QEwTkx)9+K z&jxX+8<*zriwhC#Sc}Q*>C*~1v3GUItCeAJodxbP{Ak_;lsyo^)=aCM{zU{!7Q`)k z7(c#~2#%UeogBc+(yef0axcXqiHmR5n3O&P!jX5DQ8jW{sLgsCOI?qBDBWtkF*F=2 z89^2O8iLIaCrI0;Z~(6ok0&n6E8#c1k^+W@m2M5pr3ka{V%~-9o^b;s##f|^pEoG; z3jfR`=dJ=bqfz)8VR)!N#ma}iT6F(}hHNhzXLt1S+!ZtTthjz%^g@}HA)|vUMM0*W z!bXYa$kj)d5v(_+FF6Oy%d!~C2S0I25&V{@8~nyzMKI#C>9$@{*TR=fuH!A*9F(spc)OB}_a~VXIX} z+Zj`?4l?R_P~OGt;L%!PVKy#lTQ6SHAuGl+4pTNmjBPx3=Q^!A*SenP`RDoT+<$$q z-+f>A=llJ9KHu;6{_9TR`dLlMNTiY<3;rR1L~U`O2u55P-FZH1zvRgAEcQOq9x>te zlN!OmF2P!hh*Kk>B{Ki?Uvpo_jXy1l&_G&QQ;JkeV$q2)zLVde>5O_uF~zT|zVpmR ztb64=zjp1$2MiM=rUVr|8UueA`BVBE@|#KvF+n1$0Z(ix&seyL=5Xr2>oln_Dtb~d z8R>wc9e3to-npa?FG(cEF%ajYolmx9G3K5Ce{-Z@7dab{1s8`q8b#0FYIj)wuud-z zw$;^~HfKvdD_bQQS%z;&V$U3U+u(3TXdW&s<5sF>m>i;PV!wjm^qAofryh^}7=`t2 zC|Vl7J(sz;B8xa|wtO%BxR63SAhj16_AJtsJOhXPg(qYcep@fS7E9b2EsEh8O_PR(UT=#Nlog8aJ5grw>F5yXCWLt5K{%&@Os(3mLc zVm`l2XpL>JwcI;oxpg3Jh;Ac0y}~PhpLP$+t|5Ds(56E9u{&+%j7%8FIt%Ug36?p1 zhG$eeFu2pjvEkvqC>JsWBrwbiE4a&;EiWf~#Q_ut&I}GpUZCTr^z*de&?l#Am$G!j zu6`?bNzB){CV>Q1YYVN(WxII?3 zcex*F>^BtNP4^ul4UYS(a*sX^g-b8*@%g7u2x2PCB*ZwQ3xOhmF(#a5=eWTZ`<#*Z z5JSr?k78%+X%R<*LzM+%6!rlgxQD9y)3PeVaR=f(?_rP?7dLwec7p6d&>xGWR%y_A z^d7Sa#|}%~tzHi^4_Cd-=wI{5q~F7?cqxwg%5H3Vk2}FkjF^3S$7Yjdu8BkZOkTs2eLIte%T(w~fSvMX3iu)Nez%?jlWPr8Hjf)p&gDY{hFxZotpP z=aJoE$E4@r=Q;}et}lo)?wuS$Y1}%AAUCc1Ld`$)Ek%IZb+CEyqP-$8vlxO(Nyy>T zo^JP?&ERHG3r0f2{BRBG^rw*`%sh|eO~tu$&-ex>;uiJsXROXkh5xc81~C%$2fa5Z zL!8UsMSB&;rVfb=`fv>_1Fs^7zyB73J%m&?Yc-XV+n$4E!`jZW_>Kge> z4>!0&66BiXI!nA+J7akxEC)U>U13a8^YY!ZjTh2{D$yA`^<)XS8x$vD%1TR2a~?($ zg>dXGH}k5qehN!pXy0CFPbJxD!uE3XM_D3e?o`E8%vC5ssTZu5si@xM)@-mwXk_*G{{G6n13N2j(HKJ(xnsj^Y)N4LD$+=hn2ma!S8-V4n=+vL%D6d%= zL-N~C(DUv#&8YBeOt9R#oEpb3Wa;&uis<}N(9uQuWB|CViujLEMPz_@22=w*CCF3A=d z)+aTqN*z#XFxq{~PWYxb_&4lRBPuB|FEqc!L*>gY{;^%5LJ%rt=y}L~V3h4qAAKf@ z+9f7jyFbfr=i}Yu>x;dgNyA!3J^A(O@X;dfB>~08XCYx@t1!#42)FbI8En&1u7)1O zx-{`Q9X39OZtE3JTPPJkK@bap4+{KcHIE>IJ+BSzSS&7^A1iU<;)%liT6NS|t1Gp7YXP(Da0T_|@H9QAom}Uu z4=A2ga>Z@eYhn8*PF7ib1(}!FBD$3jdZH!_s$}vh0f>MMAPF4XwC|n?N419I+ zaP!%SwekbS!OKI?>Cp0P#AB`?fpi&7gp(>S+Uv18;I&;XzaTgo;0f#Tro_om{zJ?*{$irZ4I`Y&o=dd=4P zkGXzD5NIZ_s%6XH3Y1zd;Syj<@d2v7MpKCs1# literal 0 HcmV?d00001 diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index ce2565c22..237747ad3 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -216,34 +216,12 @@ ")" ] }, - { - "cell_type": "markdown", - "id": "d22ef356", - "metadata": {}, - "source": [ - "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", - "\n", - "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "38592086", - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, { "cell_type": "markdown", "id": "943a45d6", "metadata": {}, "source": [ - "## Importing the champion model\n", + "### Importing the champion model\n", "\n", "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" ] @@ -413,7 +391,14 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Investigate potential challenger models\n", + "## Developing potential challenger models" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Train potential challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", "\n", @@ -424,7 +409,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Random forest classification model\n", + "#### Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", "\n", @@ -454,7 +439,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Logistic regression model\n", + "#### Logistic regression model\n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", "\n", @@ -492,7 +477,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Extract predicted probabilities\n", + "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models:" ] @@ -520,7 +505,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Compute binary predictions\n", + "#### Compute binary predictions\n", "\n", "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", "\n", @@ -660,7 +645,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Assign predictions\n", + "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", "\n", @@ -1001,173 +986,39 @@ ] }, { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Enabling custom context for test descriptions" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", - "\n", - "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", - "\n", - "

Want to learn more about setting custom context for LLM-generated test descriptions?\n", - "

\n", - "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Review default LLM-generated descriptions\n", - "\n", - "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", - "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate. We'll go over running tests in more detail in a subsequent notebook:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Enable use case context\n", - "\n", - "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", - "\n", - "This is a global setting that will affect all tests for your linked model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" - ] - }, - { - "cell_type": "markdown", - "id": "0d1e90ba", - "metadata": {}, - "source": [ - "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "87f7fb2e", - "metadata": {}, - "outputs": [], - "source": [ - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "id": "d6e38bd2", - "metadata": {}, - "source": [ - "With the use case context set, generate an updated test description for the `HighPearsonCorrelation` test for comparision with default output:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We'll also adjust the use case context to focus on this comparison for tests going forward:" + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Enable custom context for test descriptions" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", + "\n", + "Before we run our next batch of tests, we'll include some custom use case context to focus on comparison testing going forward, improving the relevancy, insight, and format of the test descriptions returned. By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context. To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", + "\n", + "This is a global setting that will affect all tests for your linked model:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" + ] + }, + { + "cell_type": "markdown", + "id": "0d1e90ba", + "metadata": {}, + "source": [ + "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" ] }, { @@ -1244,9 +1095,15 @@ }, { "cell_type": "markdown", + "id": "d6e38bd2", "metadata": {}, "source": [ - "We'll run each test once for each model with the same `vm_test_ds` dataset to compare them:" + "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", + "

\n", + "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", + "
\n", + "\n", + "With the use case context set, we'll run each test in `mpt_chall` once for each model with the same `vm_test_ds` dataset to compare them:" ] }, { @@ -1754,7 +1611,21 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Add test results to reporting\n", + "## Next steps" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Working with your validation report" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Add test results to reporting\n", "\n", "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", "\n", @@ -1791,6 +1662,67 @@ " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Finalize your validation report\n", + "\n", + "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### 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)\n", + "\n", + "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 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." + ] } ], "metadata": { diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb index dc1bfeed2..5446026dc 100644 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb @@ -1059,7 +1059,9 @@ "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)" + "- [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." ] } ], From 0224c69fd726639dac5d22cf1d3009e90f1d1bf8 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 13:41:55 -0700 Subject: [PATCH 80/90] Adding individual tests --- .../validate_application_scorecard.ipynb | 135 +++++++++++++----- 1 file changed, 99 insertions(+), 36 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index 237747ad3..8f821f847 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -105,7 +105,7 @@ "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", "\n", - "For this series of notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", + "For this notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", "\n", @@ -185,7 +185,7 @@ "\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", + "2. In the left sidebar, navigate to **Inventory** and select the model you registered for this notebook.\n", "\n", "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", "\n", @@ -806,13 +806,52 @@ "Refer to our notebook outlining the utilities available for viewing and understanding available ValidMind tests: Explore tests
" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Run and log an individual data quality test\n", + "\n", + "Next, we'll use our previously initialized preprocessed dataset (`vm_preprocess_dataset`) as input to run an individual test, then log the result to the ValidMind Platform.\n", + "\n", + "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", + "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform.\n", + "\n", + "Here, we'll use the `HighPearsonCorrelation` test as an example:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " test_id=\"validmind.data_validation.HighPearsonCorrelation\",\n", + " inputs={\n", + " \"dataset\": vm_preprocess_dataset\n", + " }\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "

\n", + "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform. You'll continue to see this message throughout this notebook as we run and log more tests.
" + ] + }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Log data quality tests\n", "\n", - "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" + "Now that we understand how to run a test with ValidMind, we want to run all the tests that were returned for our `classification` tasks focusing on `data_quality`.\n", + "\n", + "We'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" ] }, { @@ -830,10 +869,7 @@ "metadata": {}, "source": [ "\n", - "With our data quality tests identified, let's run our first batch of tests using our previously initialized preprocessed dataset (`vm_preprocess_dataset`) and log their results.\n", - "\n", - "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", - "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" + "With our data quality tests stored, let's run our first batch of tests using the same preprocessed dataset (`vm_preprocess_dataset`) and log their results." ] }, { @@ -855,16 +891,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", - "

\n", - "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Run comparison tests\n", + "### Run data quality comparison tests\n", "\n", "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" ] @@ -890,7 +917,14 @@ "source": [ "## Run performance tests\n", "\n", - "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", + "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Identify performance tests\n", "\n", "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" ] @@ -909,7 +943,36 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We'll isolate the specific tests we want to run in `mpt`:\n", + "### Run and log an individual performance test\n", + "\n", + "Before we run our batch of performance tests, we'll use our previously initialized testing dataset (`vm_test_ds`) as input to run an individual test, then log the result to the ValidMind Platform.\n", + "\n", + "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here (`xgboost_champion`):\n", + "\n", + "Here, we'll use the `ClassifierPerformance` test as an example:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "vm.tests.run_test(\n", + " test_id=\"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion\",\n", + " inputs={\n", + " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model\n", + " }\n", + ").log()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Log performance tests\n", + "\n", + "We only want to run a few of the tests that were returned for our `classification` tasks focusing on `model_performance`, so we'll isolate the specific tests we want to run in `mpt`:\n", "\n", "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", @@ -917,7 +980,7 @@ "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", "- [`ROCCurve`](https://docs.validmind.ai/tests/model_validation/sklearn/ROCCurve.html)\n", "\n", - "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here:" + "Note the custom `result_id`s appended to the `test_id`s for our champion model (`xgboost_champion`):" ] }, { @@ -954,7 +1017,7 @@ " vm.tests.run_test(\n", " test,\n", " inputs={\n", - " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", + " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model\n", " },\n", " ).log()" ] @@ -1101,7 +1164,14 @@ "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", "

\n", "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", - "
\n", + "
" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Run performance comparison tests\n", "\n", "With the use case context set, we'll run each test in `mpt_chall` once for each model with the same `vm_test_ds` dataset to compare them:" ] @@ -1127,7 +1197,7 @@ "source": [ "
Based on the performance metrics, we can conclude that the random forest classification model is not a viable candidate for our use case and can be disregarded in our tests going forward.\n", "

\n", - "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" + "In the next section, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" ] }, { @@ -1136,7 +1206,7 @@ "source": [ "## Adjust a ValidMind test\n", "\n", - "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", + "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in Run performance tests to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", "\n", "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" ] @@ -1183,7 +1253,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", + "
Again, note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", "

\n", "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" ] @@ -1208,15 +1278,6 @@ "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "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
" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -1627,7 +1688,9 @@ "source": [ "#### Add test results to reporting\n", "\n", - "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", + "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)).\n", + "\n", + "We'll insert a specific set of test results in this example:\n", "\n", "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", "\n", @@ -1669,7 +1732,7 @@ "source": [ "#### Finalize your validation report\n", "\n", - "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" + "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include more test results, create risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" ] }, { From cd4e7661d7a34aa84c562a836ee9287c34b4d87f Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 13:49:33 -0700 Subject: [PATCH 81/90] Editing --- .../validate_application_scorecard.ipynb | 25 ++++++------------- 1 file changed, 8 insertions(+), 17 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index 8f821f847..b5afd5933 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -783,7 +783,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Then we'll call [`vm.tests.list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to list all the data quality tests for classification:" + "Then we'll call `vm.tests.list_tests()` to list all the data quality tests for classification:" ] }, { @@ -817,7 +817,7 @@ "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform.\n", "\n", - "Here, we'll use the `HighPearsonCorrelation` test as an example:" + "Here, we'll use the [`HighPearsonCorrelation` test](https://docs.validmind.ai/tests/data_validation/HighPearsonCorrelation.html) as an example:" ] }, { @@ -949,7 +949,7 @@ "\n", "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here (`xgboost_champion`):\n", "\n", - "Here, we'll use the `ClassifierPerformance` test as an example:" + "Here, we'll use the [`ClassifierPerformance` test](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html) as an example:" ] }, { @@ -972,9 +972,9 @@ "source": [ "### Log performance tests\n", "\n", - "We only want to run a few of the tests that were returned for our `classification` tasks focusing on `model_performance`, so we'll isolate the specific tests we want to run in `mpt`:\n", + "We only want to run a few other tests that were returned for our `classification` tasks focusing on `model_performance`, so we'll isolate the specific tests we want to batch run in `mpt`:\n", "\n", - "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", + "- `ClassifierPerformance`\n", "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", "- [`MinimumAccuracy`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumAccuracy.html)\n", "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", @@ -1186,7 +1186,7 @@ " vm.tests.run_test(\n", " test,\n", " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", + " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model]\n", " }\n", " ).log()" ] @@ -1208,7 +1208,7 @@ "\n", "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in Run performance tests to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", "\n", - "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" + "Use `run_test()` with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" ] }, { @@ -1249,15 +1249,6 @@ ").log()" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Again, note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", - "

\n", - "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -1266,7 +1257,7 @@ "\n", "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", "\n", - "Use [`list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to identify all the model diagnosis tests for classification:" + "Use `list_tests()` to identify all the model diagnosis tests for classification:" ] }, { From 53fad6b9adbb022b37ead2595e56a3da664493c7 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 14:08:03 -0700 Subject: [PATCH 82/90] More editing --- .../validate_application_scorecard.ipynb | 30 ++----------------- 1 file changed, 3 insertions(+), 27 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index b5afd5933..ae87198ca 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -738,33 +738,9 @@ "source": [ "## Run data quality tests\n", "\n", - "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Identify qualitative tests\n", - "\n", - "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ + "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately.\n", "\n", - "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", + "We want to narrow down the tests we want to run from the selection provided by ValidMind, so we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", "\n", "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `classification` 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 `data_quality` tag." @@ -783,7 +759,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Then we'll call `vm.tests.list_tests()` to list all the data quality tests for classification:" + "Then we'll call [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to list all the data quality tests for classification:" ] }, { From a5bc2c821b6b093aaa3d423f610bcb4e0c5ba588 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 14:19:25 -0700 Subject: [PATCH 83/90] Adding ToC --- .../validate_application_scorecard.ipynb | 207 ++++++++++++++++-- 1 file changed, 187 insertions(+), 20 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index ae87198ca..f60f9880e 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -23,9 +23,77 @@ }, { "cell_type": "markdown", - "id": "434bea70", "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", + " - [Register a sample model](#toc2_1_) \n", + " - [Assign validator credentials](#toc2_1_1_) \n", + " - [Install the ValidMind Library](#toc2_2_) \n", + " - [Initialize the ValidMind Library](#toc2_3_) \n", + " - [Get your code snippet](#toc2_3_1_) \n", + " - [Importing the champion model](#toc2_4_) \n", + " - [Load the sample dataset](#toc2_5_) \n", + " - [Preprocess the dataset](#toc2_5_1_) \n", + " - [Apply feature engineering to the dataset](#toc2_5_2_) \n", + " - [Split the feature engineered dataset](#toc2_6_) \n", + "- [Developing potential challenger models](#toc3_) \n", + " - [Train potential challenger models](#toc3_1_) \n", + " - [Random forest classification model](#toc3_1_1_) \n", + " - [Logistic regression model](#toc3_1_2_) \n", + " - [Extract predicted probabilities](#toc3_2_) \n", + " - [Compute binary predictions](#toc3_2_1_) \n", + "- [Initializing the ValidMind objects](#toc4_) \n", + " - [Initialize the ValidMind datasets](#toc4_1_) \n", + " - [Initialize the model objects](#toc4_2_) \n", + " - [Assign predictions](#toc4_3_) \n", + " - [Compute credit risk scores](#toc4_4_) \n", + "- [Run data quality tests](#toc5_) \n", + " - [Run and log an individual data quality test](#toc5_1_) \n", + " - [Log data quality tests](#toc5_2_) \n", + " - [Run data quality comparison tests](#toc5_3_) \n", + "- [Run performance tests](#toc6_) \n", + " - [Identify performance tests](#toc6_1_) \n", + " - [Run and log an individual performance test](#toc6_2_) \n", + " - [Log performance tests](#toc6_3_) \n", + " - [Evaluate performance of the champion model](#toc6_4_) \n", + " - [Evaluate performance of challenger models](#toc6_5_) \n", + " - [Enable custom context for test descriptions](#toc6_5_1_) \n", + " - [Run performance comparison tests](#toc6_5_2_) \n", + "- [Adjust a ValidMind test](#toc7_) \n", + "- [Run diagnostic tests](#toc8_) \n", + "- [Run feature importance tests](#toc9_) \n", + "- [Implement a custom test](#toc10_) \n", + "- [Verify test runs](#toc11_) \n", + "- [Next steps](#toc12_) \n", + " - [Working with your validation report](#toc12_1_) \n", + " - [Add test results to reporting](#toc12_1_1_) \n", + " - [Finalize your validation report](#toc12_1_2_) \n", + " - [Discover more learning resources](#toc12_2_) \n", + "- [Upgrade ValidMind](#toc13_) \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", @@ -35,9 +103,10 @@ }, { "cell_type": "markdown", - "id": "662d6027", "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", @@ -47,9 +116,10 @@ }, { "cell_type": "markdown", - "id": "5a1a2fa8", "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", @@ -61,9 +131,10 @@ }, { "cell_type": "markdown", - "id": "aa9b1d8e", "metadata": {}, "source": [ + "\n", + "\n", "### Key concepts\n", "\n", "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", @@ -90,22 +161,44 @@ }, { "cell_type": "markdown", - "id": "057c1417", "metadata": {}, "source": [ + "\n", + "\n", "## Setting up" ] }, { "cell_type": "markdown", - "id": "ef1f5ce4", "metadata": {}, "source": [ + "\n", + "\n", "### Register a sample model\n", "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", "\n", - "For this notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", + "For this notebook, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\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 **Continue**. ([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:\n", + "\n", + " - Documentation template: `Credit Risk Scorecard`\n", + " - Use case: `Credit Risk — CECL`\n", + "\n", + " You can fill in other options according to your preference." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + ", we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", "\n", @@ -123,9 +216,10 @@ }, { "cell_type": "markdown", - "id": "9743927c", "metadata": {}, "source": [ + "\n", + "\n", "#### Assign validator credentials\n", "\n", "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", @@ -145,9 +239,10 @@ }, { "cell_type": "markdown", - "id": "37a94e48", "metadata": {}, "source": [ + "\n", + "\n", "### Install the ValidMind Library\n", "\n", "
Recommended Python versions\n", @@ -160,7 +255,6 @@ { "cell_type": "code", "execution_count": null, - "id": "cef99db1", "metadata": {}, "outputs": [], "source": [ @@ -171,6 +265,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind Library\n", "\n", "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." @@ -178,9 +274,10 @@ }, { "cell_type": "markdown", - "id": "30803184", "metadata": {}, "source": [ + "\n", + "\n", "#### Get your code snippet\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", @@ -195,7 +292,6 @@ { "cell_type": "code", "execution_count": null, - "id": "d5de450a", "metadata": {}, "outputs": [], "source": [ @@ -218,9 +314,10 @@ }, { "cell_type": "markdown", - "id": "943a45d6", "metadata": {}, "source": [ + "\n", + "\n", "### Importing the champion model\n", "\n", "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" @@ -229,7 +326,6 @@ { "cell_type": "code", "execution_count": null, - "id": "3f9e197e", "metadata": {}, "outputs": [], "source": [ @@ -244,7 +340,6 @@ { "cell_type": "code", "execution_count": null, - "id": "b6887d38", "metadata": {}, "outputs": [], "source": [ @@ -256,6 +351,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Load the sample dataset\n", "\n", "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", @@ -269,7 +366,6 @@ { "cell_type": "code", "execution_count": null, - "id": "40e73b7c", "metadata": {}, "outputs": [], "source": [ @@ -283,6 +379,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Preprocess the dataset\n", "\n", "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", @@ -296,7 +394,6 @@ { "cell_type": "code", "execution_count": null, - "id": "e436e862", "metadata": {}, "outputs": [], "source": [ @@ -307,6 +404,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Apply feature engineering to the dataset\n", "\n", "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", @@ -320,7 +419,6 @@ { "cell_type": "code", "execution_count": null, - "id": "9c6b641c", "metadata": {}, "outputs": [], "source": [ @@ -332,6 +430,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Split the feature engineered dataset\n", "\n", "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", @@ -391,6 +491,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Developing potential challenger models" ] }, @@ -398,6 +500,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Train potential challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", @@ -409,6 +513,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", @@ -439,6 +545,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Logistic regression model\n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", @@ -477,6 +585,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models:" @@ -505,6 +615,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Compute binary predictions\n", "\n", "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", @@ -538,6 +650,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Initializing the ValidMind objects" ] }, @@ -545,6 +659,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind datasets\n", "\n", "Before you can run tests, you'll need 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", @@ -609,6 +725,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the model objects\n", "\n", "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", @@ -645,6 +763,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", @@ -704,6 +824,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Compute credit risk scores\n", "\n", "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" @@ -736,6 +858,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run data quality tests\n", "\n", "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately.\n", @@ -786,6 +910,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run and log an individual data quality test\n", "\n", "Next, we'll use our previously initialized preprocessed dataset (`vm_preprocess_dataset`) as input to run an individual test, then log the result to the ValidMind Platform.\n", @@ -823,6 +949,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Log data quality tests\n", "\n", "Now that we understand how to run a test with ValidMind, we want to run all the tests that were returned for our `classification` tasks focusing on `data_quality`.\n", @@ -867,6 +995,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run data quality comparison tests\n", "\n", "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" @@ -891,6 +1021,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run performance tests\n", "\n", "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models." @@ -900,6 +1032,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Identify performance tests\n", "\n", "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" @@ -919,6 +1053,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run and log an individual performance test\n", "\n", "Before we run our batch of performance tests, we'll use our previously initialized testing dataset (`vm_test_ds`) as input to run an individual test, then log the result to the ValidMind Platform.\n", @@ -946,6 +1082,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Log performance tests\n", "\n", "We only want to run a few other tests that were returned for our `classification` tasks focusing on `model_performance`, so we'll isolate the specific tests we want to batch run in `mpt`:\n", @@ -978,6 +1116,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Evaluate performance of the champion model\n", "\n", "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" @@ -1002,6 +1142,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Evaluate performance of challenger models\n", "\n", "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", @@ -1028,6 +1170,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Enable custom context for test descriptions" ] }, @@ -1134,7 +1278,6 @@ }, { "cell_type": "markdown", - "id": "d6e38bd2", "metadata": {}, "source": [ "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", @@ -1147,6 +1290,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Run performance comparison tests\n", "\n", "With the use case context set, we'll run each test in `mpt_chall` once for each model with the same `vm_test_ds` dataset to compare them:" @@ -1180,9 +1325,11 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Adjust a ValidMind test\n", "\n", - "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in Run performance tests to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", + "Let's dig deeper into the `MinimumF1Score` test we ran previously in Run performance tests to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", "\n", "Use `run_test()` with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" ] @@ -1229,6 +1376,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run diagnostic tests\n", "\n", "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", @@ -1333,6 +1482,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run feature importance tests\n", "\n", "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", @@ -1371,6 +1522,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Implement a custom test\n", "\n", "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", @@ -1479,6 +1632,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Verify test runs\n", "\n", "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", @@ -1639,6 +1794,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Next steps" ] }, @@ -1646,6 +1803,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Working with your validation report" ] }, @@ -1653,6 +1812,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Add test results to reporting\n", "\n", "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)).\n", @@ -1697,6 +1858,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Finalize your validation report\n", "\n", "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include more test results, create risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" @@ -1706,6 +1869,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\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", @@ -1720,6 +1885,8 @@ "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", From c9ad497a4a4ea928a2b9ed1788be28a3740ef647 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 14:20:10 -0700 Subject: [PATCH 84/90] Ugh --- .../model_validation/validate_application_scorecard.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index f60f9880e..2884450dc 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -198,7 +198,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - ", we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", + "For this notebook, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", "\n", From 9295b73bfe49886372d15866b712e762d21deb83 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 17:18:28 -0700 Subject: [PATCH 85/90] Cleaning up the What's next section --- .../inserted-minimum-f1-scores.png | Bin 110789 -> 0 bytes .../link-validator-evidence.png | Bin 398858 -> 0 bytes .../selecting-minimum-f1-scores.png | Bin 327835 -> 0 bytes .../validate_application_scorecard.ipynb | 218 +----------------- 4 files changed, 4 insertions(+), 214 deletions(-) delete mode 100644 notebooks/code_samples/model_validation/inserted-minimum-f1-scores.png delete mode 100644 notebooks/code_samples/model_validation/link-validator-evidence.png delete mode 100644 notebooks/code_samples/model_validation/selecting-minimum-f1-scores.png diff --git a/notebooks/code_samples/model_validation/inserted-minimum-f1-scores.png b/notebooks/code_samples/model_validation/inserted-minimum-f1-scores.png deleted file mode 100644 index 52ae43c719a2621e56360d2f2f357256253a4fe5..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 110789 zcmeFYcT^MI`ZkQBARt8$kX}@f-a%TVsPx`@@4bc)0!mf7^o~gHp#}&=1Vp;j(2?E( zks3N*JkR-+^}TC7=e+;Df1Io}naRxFvom|&``-6;-Pe3nSCu2gr^d&^!Xi|Vf31my zh3AWfbtm`%4(5z%4jvB{)Cq^h2&QIH`=s8f>$nSJg)=Eb`I z^BsX1W2)zuFW>HPV(%>tz%#Kl8WFTfhPTMu>NiR+5(u!QzA&ni_1|wMVYr9oMk;Ru z!Xhb*7s<_F)mfP&+5Pc3ghY?cBOG6Z%_9TfY`gAb)UwRgJ*-->T%G|eseGEliU4Vx ziM%fR_ld7xq(~ugYt&r#t(m&Lm>S;;yQ{Wn=4PLUsD57(9LEtlzlxTwh%Bs*$L{uY zAPTy;O^>}b@$`C(?WNyUmcvHyQRDl=qtj<#&ot6<*(>P;ibr@g!dNf3DNbiN_yn{b zkyh8rx!m*Q9;7mtk$9U!Hn6*C6?*lK_X8o5YUkU7dD|l?}`4q?5I`WAPCk4sC>jd+lWPol4H$mh!Y$=_C;RBucC$^0@ z$E)R_bPuL7V+*x=SAbg!PJNjCw~q`zANBQy5qC2`zJIUGvbT>}A)q3%_oM)H!&g?* zlr6|wwnUGnT))Qw68NyXc3y*Rb#(M=MSFYo%~)&6%j8@?5XCD&#dYF#%1~_X2Uwh+ z-w6>a;}}TI&+07QM`=q+zazG0#QyM3_LTrHb4|6}-RhdpSXX$T9{j-VJx8BJN&(h7 zh%Az`{NlAt=%oCEay6gRNQAwl41G`QSG}c`@X$XVD@W?>6>;1d6yN-zD3ty3R`S)? zK)aXmcqqHa*aEd%*eTsomRNkDPhQve5V|wW6A(w}z0%M25Z>n_=>>wnY01zmKU&7g z%XYU*Xi^vM9;o?Dw}X4K$-38j^-EWMoE+WTmM5nDJ-z9U`&W}+614M_!$^Ds*N&zB z3TU|fEmtr@j(P`O&&OcRaE+@M!m@IX(+*8UWiPz0tpc^%iwHL|#is?l1NVt%D8C6i zmppHLApS19OVyGR%2GiOg|6yjPX*KItaOs23*34L2Pwuk)r2RgT5Q{28Q8F_s|u-R ztD0#;Kwlg{{X;*pIkUOR?5sCxUjB|^vOIWpxAzgJ9jl!lR?-r~P;%|}3+As%a8-Ui zdr83kp5qrz6M>!@uHT)OPVwJV(j=rZvQIt$EQ9r#Pd^BJdm9ih^^EqTbV9uLGwe@{ z_|`1^3~bW6%Q5Sb@r%2AomfTTe0*AicdV?8IZ2hikMU6n({OxO ztWm6>a0(qcr=E)Sv0(6E9{y_3gl*i(h*VfCkSEPel4EDCe7eU>$7=H2Cms+#`GdvU&XYtmos)?&L8(u?ulI*u z-^h<#wv5M0k*v=Xm7?B)02E-lGm;ciMYT!5#MpM7tA0Zic3)*$mc55 z8CM;p{|T_?6{HpNqn3#gr6#BLq1F>J9b@xW5{m$QGL_;fWcX}W1a#08tQ2g{2pF2l zAkT;pa&tJf=c|9~@Wyf7k!y6e)JJ-^Sy*9CQC!`qq)iH3e#*Aj`!L#K_!s}Z=<{fo zXn|-Vx-YQ}L~5)`#jovPs^zM%s>Z4)*Wi7@1^k8lmObTYhn_V)ue5=*%XE?S+;A$w z)rVw+MiLdi(~}mclkna1?T^QiC#PErKQ9YL#)v0Irp)rps)LrWM4$p6gwq_Mqral1 z5$@T`rd{{Dq9RZXHcTSQsJeX7xbewwh)9%aktwkoR0^+-8zT($b1 z&GL|pfvl#_F>#e82t4mZK?UOO-;%HB8S;o6yIqP^aGdRn+Sn<<2ZU*92CORjv z5YPas1Hr;R_UK2YQwULX7-(|Nq@<@fag7+X!)hB9>kaL7+)m~VM>6b8%uTEyBjL8I zP69neV;e18qbV+#F7VFNNIJ(&$5aPB3*ZcW1F(*A&JF5DBmHEVPlo@CXEztF4{oe% zyKGFCtXz>*lU1Ho+v>n4il>QkRZl>!Ya*?oL2WNtg+F*XLOK5;5MEJ7) zGUMV9!U178Prt;wZoHISo%XDK34h6Roqc6<6LGcVCD~lp#CjL{?$^7OcT-Y(U(dcC ze#Moxm8Oywl3wVn>J$i83nh{HCZiiVze06zuhr2MSOVUv_&IK|V^L$VZV~rw@0-nX zjMZpOo|Rrq8>Iym)5nsyd!Jf==zK!wSL7GHru)e9R_=}7>sK*UWY#X)$=RZQ><=jn z&Lk=gs|-nUj;2^Esj3O*NVs@rPM8`#bIDy1wHZosHAr)Y(|%Hm7g^i>%5Y5)7PG?I z(-n|=6?sLhMOH+gRwSgtUun~_aNIMVOMZ|1iXS_dDQ#3@+|D(a4k}|CdyzoCV7S%bKF%EQ z$xmZtW3HglnuS|CJMdB4R;KCW>0H>*M|F0?50#-YUUUO=6~z39P#fu zi9&QNG0waD`3t8%w<<&a^&UI_~1zP_7 z=CtQR$u)Jy(AbFRdgo#{PZu`TGO(XYQ{`vu;7l{_i!4J;CL-=uWw#B(1GnFH$24Xh zh`22&Hcti``CXo8XNgw_4&3e?Ei|Japd(REi#!nXv(JaXxg~TxynzAo|7Ba93Rtjt1AXNo&p{Kko>yzWb-0bn1}dg9j&K? zZrYF4tJQ0(PHi0xs#~obnA^xH$-zZ=EdxkNnT2v5Ns>W=*?Bf zEPZQV*d5y2OFp!@YB#&{@I{>PHdgE69V{+9EF1|D&EE3i2*`bnY==vdGc|pma3qH- z*2K46H_mo|6AUnaM9716^h{}Kp_H9wg zaKWoH$@=?>I+VQi*OVc`A!IN3_$V>xBGN`*!B#~Diyd?P0PEfzYOMR1qdS-laEIo< zkKf*Tfrb5#^LMeZA{?;p{j-cJX8-3CkJ4z5RQe5`9+G^{663PuhMwFpRxbz z_)ak98kV${tbzh&uVv+FW8>;&@8(@Roo04%+(O z`YOs|R&Flb7S?W-Hr#$L?tkusCE+KAIdrk{wqW#gad!0*^OI!$M+q^^@t@N?%#8mi z;_W2KtgoWZDC_2F!zj$n%gxISz-MG+l<>5+71MnE_Mg=;za*LMy}jMVczA$7AU9Be z+s)IChfh>gl!up}ho7GdQ-aG2Cljj1Z}=d75zgP)DF!D|N> zOrK%e00_JkmiR~c|KrO4wD`Yj>iWaVm`5$L79Sy*j;Q8-C1K=k%KRCyXBdx=0HEqluV`cyNbYi|= zV75Pd%+?s{x1z{}g(Zch@LF2i?+$VSuT|NHq4yvc5`~;et+gKDR#GS|vF2b?NQrNI zFe96UOmOZad!rpn+ONrOQ~E6Vc|iij42*N+7U(@J>9s$f-i#))ts1!n*$fBXY(j?T zo8!(et`oCGRO}xxzQelv?|wac8Xs-0*F8@l7X0s5dZ2&z9^8+FL<;N9y?^iP{R1+_ zX>tpZ(Eb0LKZj`q=ij^kKmNym8u?Fm{a={=f2`|2lkwkiVE=I||8Xn-kGPfBDPO<4 zh_8JM60eNa;q=WunXNKi__6b%P5O@^%WGnm<%Eg7HOqfv(fs!Ovqe>@QcAiyB1|f^ zW3{mySWNQ!khP2d&yxPW8=nNJSiyMt^j|;uGtu{@N&w!xs`X(Xhh%QJW-7OL_0#!v zUcUE#LJE?%cY;kUzDj@&lOBQtqd7rA>Dr~*(mDH+D)^M_Pr{#ieXyta)%;-tF1PgP zZzGB->pQ>_r7q*Wok_mDD&`vb`8k#C#Zi=Sjc?0qf*Ub!{TspCmT#8&(Bo7L5B|K zXRB)23oOwzOt5+5g>*aSMk&}n#QG$op++whx;7+Y`{Sf1DjaHaCA{4HA(ztVUvw%7 z-rap@N_b0=H|jCeXn#*cnV!sjPF;^4;A;$6yvnWv^~u~{o1I6zJnCej2x>RY$v!E{ zk_4){a%*XR_x0zT<~}eCIy`nbcV8751{C5`#(!&;LFLw2(ULI)G%E}@@!#`s-l%I* z{AigVa5M=RAQd1Z>XPk^?PJY^i7k|XuA^7Rp>x5jOn+~>B!qYO<8@!;abIT4#(@-sUFRHR%$G}NPr61r=)g&5xox23 zeX^BnM?SmRRl~VQW!mM2GwE?DALSqVzB%%G#D&fdWd0Xj#{E7tyyf`H7g}!8CO!m> z@gZ9%?8bo?ao$K3wZZ0h%mEbO_q6|O_S%aR&`CM+RX8E7~pb% zO_-?0YgGJ*XyfujsgbqB=@KB%ZS`Aj2OW`s8S=~N(F?JeE8~m~G27#oxVOYHQ%cPR z*RdbW5gJ;7P~*Z3&#~kth^>?xsN&?QG@ohCMcG3zcafN0kuuf#?upYgpPdP5G@(z3 zc_+8zaJqoqngCfmyHQb+F~p}xop=u@9&8e5aF_cS2ku&>p`w_~N$<`8Ubm_=3VqC6 z4~rXdss~UWG{0?p^Y^g<vYjy}7&=ZZxmwEhBcX4$fGls|=>l=>$Ya(~e2w(6 zVgm0LziN0=XH^ANH7Q*rzrItDn+cl+qZxEa#*~3bHrRy83wK_<9I*|SF}g1xu`|Kg zm5(E#e^<{x#AG%udYiko`q<$7=%>Z`dE6@q`_OX{i~em%_zyw^h2k%R#3@(917*5B zimE)Pz18QFhz}2Z*<6dOB-a;ul<4f}NZk*DMY7JDSayU0B=hx0bf6 z2b-LpcO57~Ic_2s%@M!zunXQC6&<`%zlZW*%Z++q``PJVK*&&?uS7Q6(qz4c*EY>l`KNun(i8X_$E^bWADu@%@D1Cfe`5)C(p}oG2@Vwt*eZB~8kkSceMxC5 z=tiF|@QcmvMlpwRIvEOQ;Ond+)~k9%Sk*MQnai1Yb-qIHb)0GzlvpORu)X+ZV>pLf zWU^$ZZZJpG^SA^#eUNz7hyscRUWKmpWh>#Qb-RA{+G_6%Ho+yPPiX77a?E{OYu(3r zKA0(>KG*0HaaHdWmvw_SYwipnA?J)=VuMgEVW|aFD{r?{q9$~F z0rA^vY)%01sdIujLTFeF)qQgvq>uQ$>vs9fWma%Xh?2`c+gd(%5nejdir_#9i9VN$ zMNSF7a-HKgQue%B3L23=anvc0`d~F2{&aeMPp?e7bnRm04dh}!`(QuuyLKdpZpDX& zbl^f^Aa_04MaAv$xa+Ras*avQdjLrJdEM@t0RWEUML>=iWOlZD7up+*eI3%Q)*DAh z*|zj1%GilentID`(P!UIE;lv!{*h8%@cm~HiQB@jOo{%BjgaCi$3TiSk@&T~M5X>D zcDC)Y!jqf?(7yoeSNPigt?-6n{=vWd_#QJXX*Au$F+}Ct@R)ywS4nb!0kqL7&X%HcGY(m zZg%;kkbp8tnd)v(ZpGYGz?V3!^B3)%RZA+#wYhTSitM-tkn*qT2>M!Z$e(ex; z>vkm`FBfT6rC*Y6FZT=4X~jeKHfE^l=N$|$8GDr*8PysqG^f{a@Qo9@;0{?w2W_!U zM91e-3~N`-NBoCF_iNE39Th{7YeQTnG?R5jb!m?lR?5bFe{b=fPg(OHgI)&Rw667M zyg5D}$9KnbSd28e2{!RJmekAC;OXDUhG&mg(2LK{tKBYHFuV+S-H5!I*|Ebvk_>cR zt$YWXGB*6H%SZZ~V6YttLflH|BQbj|LTV^c{@p{VpB z^-?kc;9|(8R{-^4hj1eT;gkNa=FRHDK}A92k~hH(yVjEkzOYOdZ&xRm+;3OWEu$v_ z>`6DK3UjZOp-K@I9A&4Iu|~NzRJvmg%b+Hrlfg?zat~h`Y@@t3YmCN&AwQa}kNS`))6+-U4<9 z;jFsV1{^%mIp*KAjTWMwN8oA1&`27)@)7%#bXJ9$B2rN8sVyfNXIV>2${bew-tJ5K z~*Y&dRUH5)!KMhx(7 zq&dGL0hKsd*#UmOei>{6M_+M@Ot#F4#>Dbn_Y`wfV12IpDZpp`(;alukk;=RLp_y9 z*$Z%US#KLwIx*hrLj+uvDrfM^0aIyOI;ffbIE%nwJrrlV1yPSQ~5v7~Gef zuU-eQIknY)Ntdtc@Z1_};n3B&R{17w>0%$$nc@n-c z^D`Ax`fcemJ@Hi--efqAC#1 z-<@8=Q*@ikd%Uc&eY$1^Vd3AsDXv8FAb}S9caH{59z1SAI5C!*w26Iz95uSnE1~)9 zIIq9f)(a6)m>n2aJ%7V&LJwj?$o&gk*WMB+at%(wO1`W z;gQ#33OP+X1z~BkQCSk(+9|$GAvy%D@ZFDs4o_&ha5K|3=bEj`?v`AD)@o*#J2VG* zLn{tClzAl*NViKZkQP4y5i0LnJwYW*uxZW0^{W@ri$AA`l|Wk2g_~=7>zB9T<@LZH zsZTSH`O^u+TOD|S9W37muIWcfw&w3hRt5&No{O_8)p`!{Zi`EI@pP$m z)>WQ0&H!CrE4)853_K3}st$5F%^a>|GbsxVfLga>{8!1+nZ?ZUq{#4!aAn58ulH)= zM`ATUlbUx&o!({fyDgem%EEQWs@3B3Gs8HhPj+riQ>A6!>MQi=v+@ztM~-zL@ca-b zpor4tt(zOvGfkLui4R^Y>(C8gO$hh@yv6+AF3o!vW(pkhV1 zdfBQaW;$ut*|s65^q?GBBVmxZ@2Y+?`J_Q}HmVt>L-sq5?Utu>a1idh zalZls9?au*RFAgt2M&7ezZtGys8-LRtK<|$pKNQMhK~*3?0;Nms7l~o@E@pI?Olkn zQrQ)F>G6#~^Qedb{d5Y4^KT&^Nz<46NcB`^=QN4$z2DkYWzDE7K9>9I1<05KSn+th z*g{MabCEByS(R(buFiJHe2PBz(p-Z-=#tfDiWrZzDabnbf>po`oc#A+2>!N);eU6x zM4Z;hu%G&)oGdX;n-l#LVxSyWWb-uVqYmR+SZ0G|a&!>5M%B>8F-}#rhZ|_$LlZk5 z153v@iHpmA7BM6*cawqRnd9o);gPmmaY~7twPROopyxBK@7MUe!Si{8(D(V8>FnkN zNly5|T_M7^F@dcOOP|#(tDaVQeHWQ7WP-Yq75AsU>ejA51Ugr8o@l~)S1Tgd-Ezg@PMi5 zwy}m0kbCpg7WsT=rg55ICgU}%b&pNrF#0}o8pR{>1EOLic61s_beS^#@VYn;T4y&h z6IF>=mc^Bi+E9U{ggu>Z*pDI{Tg%f0Tu&z2hvk=7^)t%F(l!i(8rVJ+%VL~pmvAxh zqPd=BwNl(dQ(qW#?0~En^3eN5>*GEx{Y}@!ijQhP_b|}A?}Nr_cHj(Iy7H`k#QL~r z;C{;NL!~C~gG|jJ5ixaN1X!MZ47Rv5J8)5a|hs=6KQc$TBk$Is)aIwDnDW7+k z2rY%vsvl=lQYSZ|HF6QQ^e{}<_;BL~`>XVAl_JA*_=)qZe8!@GUEL2FZVz0z_k%`y zo{{*vTS#%S?50XknNpVQs3@+F$}{0cYIzvJTF%@2vnA9^Ke3){Z*9IaZHQ{ku? znF{nzeanMnYqZbjk-I^as`nmgeN>7tug@fkC{F%VZ-{`s+IvK|qQSxAV8a3jg_uOd z(j^K%HA`BsJ$~l0Jzn&Q1vWuKIDTytY>KL!GbhAzKvgIA3jVSYaf5O&p*7=9HpRQ%sJSU5Q}_&j5`MQbULT0P+#e<79I#hJn-~~SiRoYT0UEW~SXxQ~+_HwG!So*lR zCKKHTY@ixdygq^Gu3uRCzefpPYeTtm@YLXQQX zdT*2DM5xqI5dMU;f}*YtH0kP6PL%iucWF2md^)1ATHP}mnlc7q#v<=28aNAcey zuMX!wc<^ry!74%5vJo~DQXk>Zg{KSOCmJ^F4yeI>#UmyehmN{%6tl%$$q3(W^S@aRT^`_aRnTj$Qbj(ncw$npcWAw=O{u*)CSv985Q}m(E6Sg+)R(^WQi$I&%5h+b%rMpZA#QJ zS%BPIs9wx_VkCJ1ooHWx^{0cu;1<78=08jXcnW_yxjZ0BYgr}egQNby5;S4a;tAZ_ zk)thlXM=VCoQ?4TO}SAnMVxkY0^LP)&ST~#L0a#tCi91~%mCh*L?-_IadVcW{U+it zfi4PPpNmSeee+ZnzmvAoyuKSqi_-Sk^K~d&=f(a?;grBGJ$}tjO!8V~vj8M^9W~S3 zFg3*Q?EZ?)ob-FdAkQ#_&e_Y|8X2lu4H!X# zNx!+C;ZfU40V_S4%F=Cdr@_|#D==%{@r7cShVN0h64D$2WO-KDrI{mvL37&JG6McS z$yr5E*?>F{e3hl{;iDl9XE}0?ruA{Y#3r1CR3UjG6#}-<e@2Q-pNtl{6`&_>XEGIQ0zlSZ4*DE})qRhscRbyoyKDjhi#ydF4ZfrAhARMS ziaAJwJz0HOyhe~kut2w|FPXOgZS#B8nn!=l-3Zk9GV|VDrkxo!g5 z9^Nd`f^`ii+f2lST6@#gkUlSQBdxxfyjLmjP&py1onU&_I~d2N>*>_Ye<*XAG7Mbz zhlJP;e?qHh(8GOrW$7B5FNwgyznm~YVBPk?kHG3dPYIzdbWXFKudU^9f1J&6YuIq} zcW>vXVr!-FZLnG{1AK^P!l-Hp5+k$=L2Z8nEAuKxkTjIR!Au)c83HE zHyhG#m~^dL)y5j*9gH(-C#k$^4a0C=S31m@l3ukFE$b|okEhB{n<2&ffV{^h%S66m zM=i(Gc{p%?faK46zbpYwLbR}zYv5~+faXloU5E;UkA%-cg`OD&;oQ5 zG5CAv7o>|UK=DZ0^O1*id5OB?_{NzckUz2IV&H;)VM41zxS{ek6~VO89hIdh8boCh=WSv5 zl0nn!766sNL>Z2do$?V~yOvrB+b?lcyP9CPElh`lLO~kqE=eK04^|0gu9CQsQWl_sqQEh?Hvn zuJ0yK!Zip2TgB^h)k^y53=c5vtc-acEi^40iw*d{eLCjjBBzp6?R#bIwt)dg>`K5> z+BI!q@I~I``i)VU_}nbnq6HVLw*%{>l8Ia8DGr>}lSjz9z~U*q$A@e0d+Rf~*z9=C zFamiUt%W}*Yl?fhiRg0Pqb-~i>3E8K;pC?*86G|+_;8(QcKa}&Rka^GMD7OzF7b8+ z`4eOFvih*G)=uuqtRiV~7eg+jn^x$WoL0Aed3oB$s@>^duDOod#Kbf%F2%e1ykQYJ zPGF0Ap=?;DF3eHTy@`wnTtDnGR;VHA)PQa)9@)4YAwHz4(h!+Y^Z=gFb7HVm?df=Y z;iV)k(VxIj?J!_YE_?HvSm`ElunB!UiQQA~=?0fe)#jm4)p?Z=6YAf!bEu++7VA4O zapD?ov*s%kjsQzHvRuTD;q$>z6J+DnHLS4UT>2mL07D@>*D4ySCuUtASYV8zZ2f~P zywkeH_Z^X@Yh=hngGbgZ(>~k#7${nQS9?ZLsFGdflx=e`heZVOf$hMa2Z6W=3GW+3 zSdO*C&`c%bS>*zSF)?4g&I;TQ*x{dLLR;GeT z^8S9fTgU$5v#Iz*%sTTjmw>Kc9ra1KVyx&+dG%EF0iIK{_$H)ivyD%-Os{;!d4u_vc}w)`sk7`@VjFbUUj3M%i0l!{sK2E-3HqWz$F-@38iU7~6UbNHp;?w?1P> z;^)A66}j^tCj>o_g(!Kv@}z<`;9xnL3S8@iyM16?oJN=?^S6X?)0cZl#uV`^a!2TZ zNy9)DuT?}F_uCRBx52`PzADIB)3TK4^T7E)UVoKzHnD(gVY>Cr0fl^|&I)0wpg=xb zC7T$`3`5o-)|5zR-Ki~3h8WN;^ zDp{nO?*+t+l*7i7;g+ee9R$sFZf!W#vFN*WQ41MgG@-w$YF3N-{mQ!1q{%Vu!7&Ve z0acIA{lz~k#;rvhFA6FSbnz@~TDK|dKuV|P+gZV-Y)cvq-6qA9pB zCj6`!Pp*mzowXGo?I9Fw;?Z}{l-{G0d(YKGAZs2&Hv#=`w9&(duT#@%V{bEs^ezYk zrVOf?I0>!7#_CrlF+{Fm*Zuw3HaWn}nt(ym79`h1$h&Zv0}hhqwUQ%DWI^|Yj74 ziORRD7xY(&M*D%jUV7X|r%KpnM(r09cuq5yp@)WL#y+e~DO1s0*Tb!d6^Y>yaL}hy zI-V;60?9_8ey{p)ukhca6$RO^L_XV-O(FWeF9+yO4}m=xlJlmKV6#?<3hAG3M$Bd# zzI4>ccSPAw!(eijW5i)t83d}Y`zT_(QgPrY>Xgjh|7ZZD&-^Dje-aUWk31&G9p@+!tSL)8#1^q$X$d=~0an+wY7@u!rT-+OQ z+F*xyWZk`5%quX3jgGWSQ^!-8JyoiIf$b^(+G?FX#mhYDXLQuO04`d@BV+)^Oy`sf zD{PBMQx@(L)|_GRiS=400aC^0Ld&GU_|v!H0AsQPJYO66OkMhFGrf#?&936FF|b^Z z>;ZVI{Ng-6P)g0I2IEp?Eq8vV-ONT8GV~V!tqyrlkCyl3WZLew$z7d9n5Vu<{x!hN zKV3LkS)cJ~p?^``a1jiwlH`c1-(f7*?R|)zLT~Sn)j*u=+JGx_Tt=0Q10tEQv^=Vg z#HTEkR3%~J(>}m!+(8gbtT+n*YMrQDy^SfP5ds%f2DF}f*YnsXo3O8(48QG3ias?@osAYuj&v2uKYmuw9AYAp`D0uY!ER6`|#5VyWxeY+a?pSXvj?4i-PT}^y4| zkwoEu~WnwcwGF;q9dy^sD9?&v~X~RTFSOJStUe4*VBz{9{=62~y=#S#y0a zytHD<_<^V1c8lV9=o0%rML-lVphKg+KcUh5J!>wVY!)Vd3MZyxz4Uu3NS2vJkvgjLehQir4OS4vHY!Fw4Fm@jHG)bH5O-#BP*00mR%JxI*f!C+CCr*_8%vHa{rX_m`s z#*a*{OnY(x7fi-?o6i-+Cq+ajUX-Z2&R%Iwo=XwC`@pvylPbx!4J|E_eJd6LLgt9C zAAonoj`ly!aOe&+!6-$eY>#)?MU0Jpq{lg7CThkc44(8Neq`oqk{FPXO3}VKDAJLi zBH>{Mu`^}$y=qaTr==br!H5i!NbetAI@*%8$_?9i6HgAhd{!!!L!0{g)hB zvcojfN>xLNO5fJWw^t z&Gp*CiTe;s0lT!f7%9y9B6#QJ(ITGPbUz&uf753h)V7)NYE!3FcM{fJ0HfB%Bqy!+ zcG5f{i#K3djnsTgJ796v_3g=Y6-dy(psZ9QU)18 z{}I)>*;H8eM1COhpmHqNuF0Bc-=nZQ`eR5Vs?tY!YxF7Oq89OZZaq~PWUo}7LaDhW z?c1QcIrz!Mu-EERJ+|*(EOwu`NE8{B$~{y}YL)*eK}@1M+*9d@j$r$icz)Ol)Fh*y zD2TA{swo{op!-J5FgkaX~{yTZ@% zTLL#I>$GYViRE;hu3-Wu$supNANueC+C$4V1*ij!%eS)YvQ@JSUB}uapT=6R6E@fz(Y6m z#1450r&cU|dg6h#4uw`B@WK<5hU|0oKY?V%KeH4* zvoaf(3%(~iic8y+k-hd?wxiISDS3;t-NTubeMl--N5f&!ynnn{h~)=d7U-i zu_Fd}f^POWfw=J4Yza1C+Pj(Upb`CDhAK#u1SkmG&Il}nsGdG+cs!D;_lJrZ0gf*} z@C8L^PFOtm@lybbd!1HD@SffOqe-d6clNLJ6)JN#j<__wDZ;$8=lKRds-w&j1F5}p z#0ftS@lCG$NwjaecQBmw;$JxAHki1vyABB_!(<#VF`9og5O8iB-T_z=onH{q73}i9 zpy2e-i{fwy4iEqXk7xpn&nIm;UKgWHIDG#^B8nF6;z8({z=Q#E#&H*+4+!WjFY8zR zLE6ojK)x>5@Gg#@;O_nE%SjsKh3D9d+|8Lbu5rG*k7$r$o@*sZ7^+fW?Tha0h3|MV zi}{3l<1Lygp8pO0OnqR3z5T%4_qLY?Z)4K8IHxQ&yqiRWDm@$KY?gu?3{4m?E5WJk zr4@|1=-zr6dDK-g-O~SnE*x-%fNHK0- zmdKK?uB3aAES12afgM2qSKkaTJsIWGX`kP}n08xkuhC}Ztp05w7wCix*rb>LRB4yh zSTTZZxE0Ji?Gt{~{1cE;sjWrC0I>9t<}lm_%bJmmeIrLkqUy%;yok@ct}M0MogLWh z5_%e3n-!|~OWj@mibKtw+!}6P>!gma3;qiU2=cRTz+D4sobe)^qWYOU?(9WgP;|R- zEhg69R(B$vD0}q#YFo4T(?qGJ;oLoZ;1tmYF%Vkr*thcKeTg|2B%?lo-6m2$^V+3dwrzL2bC~3l;j(jwAYhKd z)99SWJ9&J5&i<=h1QBNnPh10l7VU}34eARiDw`?2ZKoVb17s`jjv8QO9WL3jscJ!P z&0FdrzDK$lo@hQMgaw006@r_rsDINa2&!AGlxPT`6`Clc-&ni};So`|rK-z*V}nnb zRFh4gIH+fG)_(rGW_1j_b8Fqk=4>TGcTEN#1W#DUcJd;}FIBg7`(nhxE{2^6~S2DpgC&(OkpdfEwBE(I!Vw`U5y{!uN2OKkp!Fq#GAcLtR%&$ zr@TcVL?@tnquF`~>v78R%4IE|nK=RqC5$o`Iq4~zp1=|+s#hPJ1tVnO)1}}vX0}n0 zs&Gyp)5PR#G?R$$@sJ(;9X^&yAeWbi4#X@I`Sh^TZUEj|7Gc&>_1MD6ocf)_MydBv z?xF8nmq~$8rIA|mVDQooY`i$W~zO!knV{9!0w;A7iSzWB1!oKPsfe2i!utOW0Wuv-) z{>}RY5;KsGBjqB~Iv9m-p(QQD?BeQ_NTKji2J!PCmmazD!bvMWhG z_Ibjox)rxqCKUZX;k(U0kFMLR z70X()a{~OPG8D+_)-_Q>{07L<37fMvBH}g9hL4 z{q1{+Qeza*5Cc3so0-cpb(vG*2%Dt4jwDNTZ)Y3P zdUoPLWV$*gjSBbEw+w8+A?*dXlcHSSlJ2x&Juen7hC>$|{f>=5CsPzFV zH`;t@b~6MGyE;mEPe~XY&rqH+cU134wtE?k?pXa{@x0}g37@n?ylTJ%5S1(T5_IoQRC?CZ|GK!H4lx(?2M+|08%++U% zh|s+H=v5-h<)Yu>>?7)rOsgwzXH!aUHN~g+!6WXwUmMKZqRjE&8l#rAgGW6=UJNJQ zZ^el7Jx`sLzfFqfImv|B^(b2&YawW+|AKE04P_dstX6tHV?LEe08Uw)o-2|pz9S#1 zmuI3T>*Heu3Qvnw5~eS)0c~ufO~^J_fqPyW1AmLv3k*j<7RtLy=oH-Hq}gXXx!#$^E_>glD!x7+!h0Jccuv4M`-JNl2ES)hmoG9EnCu4A zMp@4sP@9AwW(Giiuiln#x$E0g71>jj$hIIJ7LfOeWM^!cCCnW0QLDJ{{5KzfygUbV z%v7pZ>G|{W(Y`stL7+?{+caeDM^KlnBQB1_m^Sr96W?%qD5;^S?_q(m!PR?Cqw9^vYZ)+FaDhKrahV*CVW $v+SmSk%&vPpl+B=`J?|ZXPgwB#x{#xSB$=prtJ*=z@%secAtv#!d)Bx3oenu>^GE5B5?+%bF`+n>E)oON zsZiHlgk1XM_Ke#N@=wOU{f<9%=4@>%$cgTspsZXzY~7W?lK`RdgoA1@bA8b+JeH)* z6{zI$o%?2IU{dutCt3_FHu0As0x{euzvWufXS94!JTcAKlG{tJc}!-(sAMEIynbUO z+u=U#o|9*XV`1AL60I{NVm|6xb!n-oJ=))*q(J!*nA4?097^){^p(N`=lMq2*|>D; za1jE@X>zu_s%IY9RnVxP7N|hJH^QoE!7ttnelQW^G?mG~`6;-TznraYs}G>Fq5hYhCgX_@cM(Y*68@5zy|{_W!Z>mQisv>(*!>kU-Eta0n#02M_M< z!KKmQ?ry5`;6$$5-rA{eWX9ljD<_CQam2Z+#;$o z>Sg#h8W;e?6p&u3W{?!txU5@_Bnd0e950$J67jfV4&^VNlOVT`gd zYEM@Ugf8T)U+-T8Z4NUs@=%xi@+fYm^IM zvv*t~KU$5;P0l;{c-A$rad=pTq3ws-? zE=wxXb$xmb@&3cHhFNg#>T-}JgeMoJpf=t(HpqMNPc48a?boo9iX;lJ&wf>vJ>uQ|<%V4TNc%)i+``mqWTaJn#${1? zdmx6p7O#aV(v&JH=x^l$Y>kj+RT`DwmD(m)vpDrJ&(&Gw2V%`QtAWYXBA>xVy%#8_ z{dClod(ex832AW5Nk#nIE%T4cDhUQ$9>Yx=Tx~ozXlylpiFMHEWrHJLCWBUx8XS6v zGRzlHl)DLoPAh^lh({~p+uwHZA4DhHB>1K^1{)U|?8f1MhJA@{1JS0UL~-ET-b?xF zsfjE_NRQ)v0VM<^#r%P6fwmkBimO>wk)zoUfTRf5z_0^aTLk!z#D4zbk$r7RHXj>V zTYeT!(x~+S9Kk(AfMDr<6-)c~W#hl1WBsUb01<^Pj_*f*pIZ8BUj@bi;f1*KBWksP z|2!Z6WPd3#0I#RthGgmYRQv}()c@_EUk)YEW^YNU6yZ;sz`$i`1G5~vBG3RViN6cB z{`#q;JkVymF(24Bf7%2=r2biZDOIbVBk`a2Ejk@&^Ir_oKbryIUx(NOYf#bq`5xb& z_l=AfXd?+tw-^26;gFzj0Yu-d`xoP0fB1HQzDcI`ZhigZ;RulQ!y`D0-k{{% z{dwQMpTi+IlQ(nVZ6)!&mP{@Qbr|J^|UyMcZitpD9WB)$Uw z|K30$vwAZ>C-S$h`(o_Jo6%OGtSQecK%aDPyRG@&V~NDt+K^~AJ3H}O=2;Q1(MmI9v!m#$=6sA zF6Z@td#B?cPHf3P8fm=TB~8X1e#%BO0Cge6R;f#ozwPJ09N1*^wnvMcoK|8lc=W{( zdED>oScLR02b0K!8ypo~4qszJ?1LLvN8`G4wh$adm(E0j|JVT5L16JuZ+pMt7r-!D zjo}`B4be+LZLNcrwbF^u`#4{ysA){x|3iS`pQrkNm=`kT)zPtR`%$>2>aF2)-ohZG zZvg3<_$BHuASf=*#Hj*#ISE0Z-~G8+e+J73>qE@-S>;G7fg*f!FnP)ZL^h&*`Cp6S zs`V*;voBh5Ih7O7n<+X2`}Xf$%D)UlAQo}A{9EheNpLE+kB+_30EOA7{D`l#b&&Q_ zA5`M_&fW;gJY+mF={y-lKQYQbo>Ku5cepI3Y{xl&(B0#$wA|BOnYQ7$tywl13Jz~1 z8SmdG75?ST#~3TOKL9_#x{^GC%w@xF0y+tokB^Ni+jLKCGarx*m3f#QG5*va@!Q21 zUBkUPT4C&qA&;?A7fT6>mo0yPiex?J|iWRi5Z}A)cGF^XlKmXk(!cqG@K(A~Ry7YwJ?SJR&zkbS43b>}Ow_*h9JdBY;f3UhsKlq*Hqy4&>c=~nY zgEnweMJY0zhRx}7^h#;f{&k%YTr@e3WV&W_t^{S@C!^@kKdTa-Ds|rB5PK!yS#grn zIE${Zh(-Rv7rmEVg-~<4tmB-mv>*wVN*oYB2tMJlj{aEy;iUYgfPk>*rz&3``c~U8Z5fNY8K)ydUziTR>Eg$7M*Gb z+tpFPl?~k9w8XN~lj}|TgB4h306&~@D?R*n{=NOO)c#jrv5qKx|C30`?`W;c6o|=RV z6M-m!y37dU(&WwA4!K6<=Y*pGoHZHAHGBqGq~LEtex^S0+09Ri*S|)$q)BX{3bd8p92#Qs%7M=2Q zp&j-&Ht&(|II>vVX-(6n#Hs4@NIcsl2(x*K*yo!xjH#j6A=KoK~KrWPOXPB9Kx>Y>sR_yjz;9 zZf|yx+@JiJb|cLhg&U^#z~i(ZOQ(72SECUpg_Wb%e&6m~FyBb3 zUKQ0`7DPEeUYW+zC2jD?{K%-cJKEHI zJ;Sm-A0SC;dkNumGg&j}eXmxU`nyIt>`8^W0EWieu%}>5Kh>D4&`P&7xj{4#XYR&1 zf0$&QVK$eHjPz>qzh7!}IxjY2PxQSyW0}2dcx(~7>N#L(^BM#|qWu{@4_5T!J0ckg zUDaN#P!4|GmYAWG)|gK52q{2iXAtU{n`0VvfbpOaljuzz=!KIw(!sJ_rLX|7(&CEB zhTnh(UY&bHJPwamU6J&q-k2wGnH9gG`p%eh_E>FEYq@wXE3x3tF}`ATUJqW#D{F~L z<3Ei5&E#*~p9bSl*3U>>7b6E>pz&KL*wk9qcr(;P4Wk?NwbB2_QtQ%Mh4`V^KdyA1 zEHlocYSnh#r@8ix=lWX>-@3lq<-cF7^KK^OwHGN|a*WB5jG;&u*E^HIzBcPcy*UXc z&e06VkLHw)mpCe%s{Qpg5_&h(~Qx$U!m=a#R6P3; z-G8XYSXb9P{DX!gO`b;&@b?{e7dyGSa5752H@Vq!2U{$G!|0xDUiRJ$h67@|#hr zM5$-L%VXQkOr1$+>~^>3W2YsIsa{-|Fb@54&sx%6;QC4%~lTr<}d9W-MFvI6ZxULe9N)si5D9 zVS3Cryt1z~e4`_a>I8Qh=)ej!nV@wa>x;?HkqG@7;TMFj35R2zg3MMN{ltoVNfiV zCsqYbirSxN!~=YK?z7M|gmaJKY%4CVr*h%WL3%Bw7r|)o1^hRIbRC3ZH`t07A?*YspP zn$kw6bwPBZgK?%Lrh-&V9b`RPSHW}ri*#)A;p3e<5ES4omSv=V@X;-m751ZF0Eau= z<&uOI7JR5zflP!ZBCIa^+}-3k5%|Z3pPmja&g41 zu9LyCg1>FMG{42?S19n~R<=^ zgAPqv!{zxN2f!sqRT{HuUHnZw(B*n-KCdpuM?9a9kMBNN9u3a)zmqD$9`Zy>^{uaK z2|=Ez@?-+g+*1u&s6>30UV@Jz3B86OuaDm8U0l25I=1Uwl1kmSzPoo)S`Ch7CHtJ3 zB?rDg<*qxyhpKvNGgu7D3OY&5I$yD9v#R8f&RJ&WsMSt`?%-A%uh-D)i1m*~888R$B}5$Zp(N<^hweY~I4 z(=D9*2@BPmw!Srp%aOwJj=165HFaoNM-?cn21hX1tP>CS+K zcl?|3+ZGpSS(eIX^&S)!#8nfQpAJb~ePuZ9SK7w^g6Ax?>9%#I`$}GKd^ZS~2XWmN<48zneVWi4!eH%#VCkH5mH-4xlIJe86gKSw60) z;vv*0-O+o^^BOTDZmpbR7ONc%4iK^@t zZI{^0evO#(C70i0nrRrjRn8Ib72G`C`ixhLbBPSDSEla(FABcISv);y%Sfi<@q$fD z)Y=HohGn782%p=w%vt22;Zs^}^Oy&;#iB)vTTiUL@BIZndo_+$*{w^hgUJXuuZ&Sq zh(@hszv|I`2yXw(4+!W=4;QrbyE|q;l%3+?4cbfJuH2N2%icwQS%*CdlI}uo(^k5Mf!_m$=k#`m2k zbh?9jYI~<_#r?LoRWd$#Hx{7}ZtKv*1bdCE^arD5Eu<19zT02^;~RZ)d{>}O5dil@ zDI6SG4J(L$t=V|%m9axTJF=)ZgSm@`2!_U0FB3s3eVl(Q>0q}wtvw;YRk@zznlT$p zx@Kem%;RLYjqVx&{t{)MKw#`O|fohyo* z0R~6C{Z9No#|p$zxQPF7p;n&V=ORueRZxx9<6NJ#)oT)ktwBu1%@R zz-?m()vog>Lko}o7tbd1xXoWlN{_c^@?-BP7n~qewaa$msds!>QvAJf8X@SlC zD>?2%4bzO=7Y9>uqSAcjV9Hsnl*2>#uw1PxzUi+&V|x0RT)H2^f~*~XbM27>wpeN`Bs`s!a zzdqz_ylyQ}pp-!ZG6IyOzje7ODuMy}2e9_`$&0lgOW)AqUneL%@Dp+@xz}FY0E$$5 zjI&rqbZW{jOBope09B>ZlC4N6gHN7Op3n&1zi` zz?MvwPW?<`lmLnuz9WqP@Ks zkfqbxgt;5Zw_W&lNKO(-X19oGfpvxGeACK7_dk*L)_ELrkD`g-&xR zph4RN?pc-0Pvm=ML1$;+W{=BXEXFr>XD->Ai2HFfjLMr=*1h*zuSb2JJ=Oi>abE92 z99tvjNZAK9Iv>SiR5kpQLyM9cgX=tuCO|Ost4nRAg=0}ej}X19#@x9#y7QA3-3INwnj4K&6SG` zRIpC1i_ZLnUiduTB(OoUt<{zYNoDb$$gj_(k1q&Xqtm!uQqadVSgK{H4>YTz6@M_+ zRcEw5g5{Zw2dOkAMN?FP(9H8BZ(tyS9zR<4(z8Z%qIXK}NYpj#p( ze2c77@@Xu8U*X4D(14IXt7hD0g06QA9YADCzgyp)o7w5toj4p&DKsm==WwrE1n>{7 zL-!05KU8n}Yt^yF_NVJXdGc?eF5DiO9w7sLjN#ZQ5v~gWT%uTi^1Qj=9cD>`Qub|l zG_{f#G!Nk6>54RDs6neBko%RcOCtWhczUfZ01>zCrLGM`Pt=+f5?X+U0MuueqS5)V zVmWwT7bWAC;H%0?B*9Qs*5Qd9FCBl4>3mLR4IQAu_d77N;F`>!Y_Jnb7rfi)n^H50 z2E<}CcP5{NkEdo<`r}Jh!cfgrd(K7uqY%w*9oGDeOG-dbSN&x54`8H`qfNsc;watHM(}qh{ANZ?HiZ<` zk{3#)^xD5Pi>-CN>-%06zoEA;xqz70vDtr8>LflA? ze`U?AxmWk9`vu^S#!nj%!73J=nE{|O*k;D1kw#U}IP=)NBUSTLl?#O}mx>Lc4_&5@ zk$jkf#3yI!EIQF#A?LY)WOC(Kb650Q(~vt1t5>Bco;}UbIRHxGa0FbncfY(-tda^yqwQS{~YY*f(7(d5=E z<`p|50T|A_Iw!C9CaS^?rdatpO3iEV`;u!8F1ROyp+iV|dk-JX`eiO2!OB-fZdDY4 zsTr*c_x_OCdg;ibU{;(Vd&;{BZ_vTE#%cQb`u2!6i_z^v{2I^XN2zjpay%`0T zgqqb}7B%XslLdgTI|pi4ycmxWa3-|fa_4a#oMd?-Lt#jNpjdC7s7U|5mcZ zdF_>vM5FlDTsDsp5wE*4v(a!d=Tb78MSDKYJNXLO-bVs@K&(X*X{R-14_ ze!k?LLTKzu@CG1h{Dsq@+J*5$nNuIPTEv1)g02g6`ifO!D{zZZT|tWXzMQXCWB=KO zuI0i~H3$rugV&-k@G{!DJUpF<IMs)?slga{0bg0R^x9h*Pi3BG`dO)Ort#?Kr1F49zvVKVu6e!SJ(v5#(`j{7H@eEz`+IRnO{F9owr{-a?38kUEWt zo`VMuUC4E^j$`dWkI+SVyO*}zS0a<)jn1~%HJ56fyjJxSKc0mjN!Yx~t~teene!tb zG83pkiuomKwRzYYdN7#Vl#q@A%qHX^t>cJvxWmed4@h>2^n>LaLvCd1dVUoaSKgJj zV;=5dv`Xv|=ziyl3!pgqvocy$Oq*IW%Rms-Mt4No1zQEmMxK*yz#Ba_vw5+2pVK6{ zv%~q3=M)R1`=X|c*VN2TJ%op21CwatA*4mB-a@w*-n_d)p5=iOn@Bs$myidL&f#Jw z)))yf@`?`bXl)8MY&RL?_CBoa^Ktpumq5=l{nypvwQ5^yFueiZnu-l~Y?NA=4~r3HyHXW#=#2^{TQkw`T@KgDB`?1pdA@m){UI+u*S; z2x=8N{0ejhWbrL4c)U($e=v4#yY;8j7ZvN&4}K1HrYqwSw^{;&3YAOAvV+l(LpUyM zZndB99UIUt5x1yiDH)=hoI|E1@j1S^i|PyiY>!D9s&E z6~UzHmY!9q%Je>Z2dh)HzfmqK=B+)tTz0ifsRg!_z z=6k2-eMZa32jzq|HHv7q&Pw8hv-iZOoYY(+G^%=EW}bh7232j-A-h65wdXX$hYsgi zBiU7TjxGbK&<#5$&8lhyMz@^|w`JxhG(Ws+#q;^#oz=WRbEI3y=rOy2w4JE4pN66H z@qrElN}k!Hc>WNTO^L9rvv7PjE^&O|6?LRDFz^#$`4(U5Um&X{Sz` z%;i$b*Xx5Gj-t2@_VT%33tEGNWuL6Kt1>=qCDA<%Mz@?S*Vj}wGU>uWu5i*jn=bK6 zvzXm$-VvUBuu|x#vxtic4V}mzVC>gY<8VUzfB-I}QUzZqFVB>N^gk5oG=k}>kET-c zJXTtwIaXS^le=@5WEV2g;=DFr?`6(!bcJFb<+f$7W6DcBP{=N}P6QXSh`5YQ7XLg; zsjSE~E%&Za2%THjYEnOj*IX!7I&hIq=17Rn&GHyxHtNZ1ah_XCr<^{-tx#3WmNRUQ zuhAQhwY3r75!BfX4WpuiAaygn3j|fbCE)t? zrMhe?Jon)xiiaLGohF>V^6&m^i0&f$Nr-s?(oI=5+8fcC|eD4^l2h(S9Vzf#>ayO@w_z2m!s{XJxRAI9RpI+177?433!>Hy(@3hZ0 z)!oc%Q+V2S?5~3>?3+2`a(b=AR_*czGoJU(Dq7Vo*Z%|SQp!+P|jgEb)JWRc~$%= zg;AK;>e@@Bxnu8_&tnpF4@-5ib@pae!%}3tY=N?W#a;0_i_RM)8XEzAo29$~LG@(T zpKlu(~GuaR{^I3CO_tSz015bDiRZx@UH<*M)7%yELD7<)tbRl6*8B&508b$y-8< zB!0Km&-X^5Ntc)@)OawNvCJv_cI$;cx^SM-a^@%>`gPkXr*(w6IFBw^wD8H`o9_&N zh^uY*pgBZjMdP9S)0O=;pC+pnSy3zpYfj)b(sJ4*6{8DS!fboeZmfx;0sba+wu1qI z+M!_QcrIwj$(*0jBP3wvgDX_{G2&47`{~ywO!=Adn{^rpvrg*=@eT*9CY6WN0c&zk zzOsS9onja4;cx6u1Y(&1AlK61NJ2{Ssah=(k>v)bU|MuvoYBZwQ~WFy1gX`~mm6cX zDeR4B1lxOa6O=DPN)@TkcmgrVNI)Nf9(=ii@)vZMTJO^eRkN)1eyh#?hUV4rY4+)2gHPWt)1!o%b3@-}ePVSkgOW`@)T&-;dHn?N5qAab|Xlw|>O^Xib@ z@jAsFwLLLjaR}}BeHguYcsBs^Hwth_WwD7$e%`p3Guwh;NP3i8FmHh4V4_lEbVOC8 z)879L>LF)<>8zpv9Y)8#8H#HM7c`|FB4s-8vRQWyU?V?VKdyIgmb){uL6vdUR&;r| z=P%EV?~7G0s4uo+(kgE}o;ZzfH<`w^H76dUTp3RY#1Bp0&U8WvT(5#~F`>_Aw=Y6_ z-^Q(F_bAR9;RlFgkih)+r>-AS`{QlN#~s*K5^Q2J^wB0>M%$}ru3=x=V-jXCI)0n@ z(kWdp^6VfnFa2trjh;_?d-+TI55+w67tlGSG9N+hSQa&g=AKH!J7-QO=6$66NR>h| zl#EEt=d5@>a}~lQvwoe3xLpc!4x8V3`wicKVd?I6vw=bf@j=}HHN;&Yv}zR|DW1V= zaz(dSHWZ_}ZDQ&`iWMGefNsB(&XWK6SEAy~&8pvXF8R{M`rKd)tu-5y9+Gbaqm8vX zl)@i*zPD@5Y|Ex>&gL=_#WN_0Y1UYnR0cQ&d`DgQAha3eUhA<(?iH>o^~OnN3M;)W zDB>-V!t^F0W_54*((z#Kv`@ub#4@!HorFG(OD#ZDAYIl{?lU3FvF8wy@y(^U4&3-! zDy~dj5nRyn+`4wzIpv5p7>#Irrrk$2R^Vu?H@oExe&O&|?8lLiN13_y01$E0S;Sy6 z`w%_xa^L94X#9uj9a>PMl%G)GBI`${--aFrKH%n^Mj>+a`_5KW_m08%)xN=m6uWl_ z)g7;01QA{+CRB7$9Z4Z`k#|xSXKQ|L%>v&=@)O=`OQ!I;WX4=vdYnh@>j@TRyP&}2Flet3;ke34a(WUJt?)3Is)Izm&yX~Q-P`==w} z{*M4O{oca`m$v$Y^mQ)Ah_IELHpOM!#d}@P`uY0Cw&!w3jbRO2WuV%z@ZL}i;@vur zS8s=e3sS+dAFSEL7y zUZiF384fwCl^rUkh$^Q8XO3`;j~~mN_x85cmb(@|I$$8a$fIlB$j>NoJSaChUmnT#3d!$g*q?mRxo}tXAk7; zG%rK@w%?L=8;pLBy5-s}+MtlJi%sll?jgRrZ-aSpH+~1_!^bK5mtLA~?;SQ0yzYMzn8@|> zoW@2#kL}ekb0Ra3P%gahe-);|gDHvJoE>W<$K6uY3@M=~Xa9!RV<)OFkj< zS^E@mDwB72&%ZO$=p?|;a4Vnz>Ro`+ZMya%ap;G{sOcPJo^9|+^C&CwBVDoyeRj&e z|L9wV@pkW&DZYr_WQpf8RF_x~f6H#Dy~^*l#;qbI8|-9XWl*YMKveq}lRxaZF({dH zYjdf6ljaF8y=&5aPU?W-z zsO*r9a;K>%#QW&j+q5>E zSIqz!yke;ExMiu!5@CRx!s*eH-6eNKPczn#-Ey3%krkO#(Mk4U11T;p1^}8!maFEp3oATgG%cL6k-&+iNgoyyouv6 zEhp^+iYsm(Yqy1P?#F>Uxe@H#CC?QfuKsXt@CIn-hSGkw4xfC2Gh+mLlzQV;AJKa< zG?QU>=GXm|6SUM+;8OAl&+)!ojRrn|fSp**Yi{UiW5=e!%+`7qkZPbd{kSgfTnZsv z2G53z*ZV+0`%T(q<-;r0O3t7TqBxvAnp!$8QRS88i6A<_o-~`E z8HR4NoS1TVni*tpjjeCYA5pDT6GeC_w{jWSiE*bK@IAR+n$Do8m^512v|gVc>Y(1@ z^&gGb)t%kBR4rB3mZb0g;C-!DW*P5$SuyZU+6!f*Ji^>l@*DekILA_J?z&B3=Jc_m zPghE7^LogA=op5-95uW4kJ+ZIcwApWIvfm zxv$+|y`XO82|qRO8ZMGh4OYRX-;Q~z2CfianQ6avn-01-*J7}hGoCb`wzc@40|HSw zu78vIXgN-xtije;XEO&~;+!!fpWA_Hub^*`eG@i$;?K;1;PPP$XFyRbrJ9N9!`^+w zWk>btkIa~#R_%^ylkDl5)muYFO}S?JNOgi#@cszM87CW9&eZZM){729^X_3285PG! zcJQadBso0na5-UcNObqfToD-`*?xGQ3A&j(baTeh%&@oLn%4f*^?>_~aGLq_W32&I zUAZ?)|AW%P$sTJzi}W)egx%voR-38zD-E%X4p`zNOr#(k7Nn@m0K#RDsthF?KBdRd z7|ILSmbIcL{zNLT0AF}a0yu+yBr$>1n`cbVPB%1=DW*ds%W22G%AvmcdGj#h9`B~O z<6n@S-}pYN-OcT=z3e<)?m*+}CylPUULuF-Aj)|E<*Qjm;==XKbb~g+yC&H8jdZPi zU+PJq%~uD^$ArSa-VCA#HwY){^Y=zUT3x-H9KvfEw)qrr5D@+GMDl{F=szLFPZJm; zHO7kL6P19or?&JfJfSwN9#y*|VwP}YSe6e#veraRhZyFA6&cbYMPCpnE}{IZuiQvJ zUnc8sdX*j?sJdop`cieqGTDtg)-JaP^hxCNeBNV4e{voAw$~1a=sa&5079p5U13ys z(StwHf|lUJyzA9;wj6AYfBZFk?u@dBaTDzVvpC$qfj1{SvK;B=gyFM@lAgZ_;=@jO zlpztq#Yt+<=YGs$hwpK|K-4F}m9p>e zS`1a&L=CTfA794TnII%tR;^Tw$=?t1U{ZB*14T!s<}-2%eVilG)cWze#z5XjGRM1Q zM(R~O1mo-GXHe~H-d~UJpN%JEJfq&=dwBr9RM&b!sn$evsc$!ehAlEVM2H7Y-`l8%d=cCE!u+EAd3)T=+|`4& z$wY5wn%^O9os*TymfTANGKZ3S(v*WblQ@#?w#c(3Tg<14{t9cIkX)JTd2;{7W9o;4 z`ZbETGp0)PWhvYCuNt{%*CM?l^W;+0zBwM8T{J#qP2qKvy@*`Qny4;L<^J()TYhVD zt&UPDoily*!mn!MpfVvX2}jj3W23PXHzn9uzVN$-gQ1++Y?cmvNi<8&R-L+_043}Ar^y-xdh>8KP$5C;479(EW z5gO^eo+4elp*Bp;DpbbA2kKb6^uNsw{ckE>JS+_w#=*%|B7%>c!X-|P!9qRgsV$s_Tn*G zVC|ci{Q{M>}Ndim}Iki9j~47qEge) zb`rUd2Y;l48ZJ)64n~eCOpiztYk&PnVt}7N%cmxtJZbau&-{u+%h1xnpbaOa<$BL- z2}lk);Po)rcqH`;mYa?X7eGvKd--%u+^X$6h9d1{R|3V^t8_0RQwjeFC91)GM27=h}(qsNdT?R<;Nt1}_iy^#ANShha zQUfAv()-GRu#tLt2y+K^D`}S&IDtkRN2lfc+qcd4`zHn^9!~fn9Wu#o32^$q?iTVi ztm{I1?M~|0OJ{#?LF|?}8dI-jK!0o|-}@%zxrml?8s&N6*0kFb+S#Ig1Bb^bN(S=} zZ-v*XAi0k_=*xJF$`AALE@gpD=%Kv%JUV_hF};=b%cBr`;s0mfc>((lvypFm@zV2! zjV1+QuY!|U6ZHbNvtUjz`}451GCnMYbP#wiA00HmmN!^8Im1Aide+6$&_i}4C6&cs zD`(zO`nkYS-GiM%A*-v>C0Elk_JyqGUZbbQB7$47f!kFCqjnvY=vRI^AT<%!FVD+h z6F~^+>2|c&IW#H4Mi6!L{VpghzP~xbR!Yg+*t~D-<1{G zobP)y`PYH_xfp-)*IQO1&QA^E<}<}K!Y?kJCko(c6=7K)+^$ExKRqP*Txo$f+^<|^ z@i{!FhfFe>?jFTYg%%Q=MrU*xVk~QorgowX^mJ)w>orZIZ^O0r3$}<`=x>f%q+{#k zzb@;l9yWSt{xe7Zs}PKbK@J-;ju!u&lZA4>Zp@IP-?yFg{AN2`S4ymgxPqzX2X#Mp z+@fw%;=Hs+A!3Ac*M_^x<4X~GR>|C6+F#d9osP*86(24NfKm(c?gIV%g>2i17WgM` zc*YoU0=)*+?*!w>;cv^)M=-4D28+41C2};J0*D-&Dhk8{?oi(W@9jZ3qpBxKz5ViA zbm$cf1eNewY1-;qRmXkH=|=i6=^I!iWj+s^MAXgtyCz#<|ES>Fg;7lbgOn{r?psfR z?P1RW28LD71MuOe76TO`>gLbH?o)B-1>6Uqs~zh?`r!){aVn$^A0Q2A~lAX_D-=-O-RD$zUV%n zSmoFZ%k0W`eersWO$LB=$zMsH%km31pW#sFE)@12q;1|%Ts%ZKtW_Mgjwri*5^G^< z8zP2uE%t6i^1jFHMMaAjdAvXSUO>O)NH{xLtV3?vpyy{YoQk!DC@QW$+`0I&NVO`H zDyrhQD$ePU+Q4t2KrX^3YBc?mB)chWI9zC9_I+Lfyl0=}Bxud#!0Lr^q*0>(8FCf0PsvRQ7^MbW%;a`r5&s=qPCs z^((0y?CX6ya0NF7dC%l|P2RQ5Pi2@=P=vdsSDFma?B;t1F~J*3m)HNABz9oz+0=`t zwNX^eGTWPdgg0AZH!rQ4I9%~j_!>u*?=OaYz&KUNXCxlXQ@bNbDzE8 zM@g>kkyDI9-8ZY5J~1uvkY$7D8EToFsF#HF`**{DCUotZA`5pi-kCDD@L z+B>G}B(njKv6gv7Gb`%auVnz^6OmE~2u}2~E z>1Md*z#`Zv%+cqNjp)&HApTQ}&}6c|ZXGpR?`z?pUmQiMk^pvQtA|blLUnb2zBgeH z*;Ra~ANe5|eK1^5gBTm>{L%*Zhy$oy*zbT`Bs1TO=IE_;BnL(ypS8N+&_~y^81GIp zHeF}Mx?ivPrcNxoOUE;4WOZ^m9nu7T6)Txb6e**zZatdMJzz3*UFCW`51>@Cr;TY; z3(3OMeU?8@^wxp7nH#Ot3S+DHaC4)|k!lx8uiE^gx#iz-a4Dxf?+4#@Lt7sx7;DV3 z2MR!C>6#9O+wmqD-*9=8JN2?5252av3~g{c+(KG@$qyD?cDxAFVa-_aE7uqDCc#$KQrdf^~E~=9_KQA?@8h-0JSn%%j zJSK9_fnOg7j=*J;x|s4b=)&S4&>*dShaUl+1eZ!P=SS1sC0!Z~1|7sbN7ln%8{kmh zObRqt75b<{g2Uf3{Q}K`aR`KKe%iDHccw9|e>Do*D&={4#Y}SSJDRQM=V8>>=-?Ih z0*Imx`^nJ(xKxLa_=4m^oo3nQ8Aj z5?hanZVXqv(})a`=O%g&b|zy@ki1D8L5<$o5nreb>M>~619@VTy3NaG^W~cJPPYon zE^3&KC5Q_VcxP6(mgEu8_Z|s~nZ2*{^HsDg%wFSn+{4K@yWcI!&*nD4h0b3pP^K{% z700T)he@1ARYEi#=zD#EB>!MyA@#(! z?2buF>yu=;p8r8ejP()JcyCvO>0?*z;57563IA9a(G=J7^xy@n;gRBAoH$@J_PkpL;`AeB*yQAdE*>gegB*B=?Oh9=<5mChqud18 zQWrq&)J!{H4p{KS#&qgGX7ss=DxD7(sgsyZ;>R#4UkqKIMe<8Al1qFcf1$6Ikf0l{ z>kL8AoYC_ubAmFUcj5Vmn1eA49_rkxuyd1a2`BR98HcoLBPd!bCtOE#&1U2-kLI%d zl?qx*Ji6_!xe2;#%A*BEBr`=m0oAP&GndJ3EN6{%+5m#E>hgmH6oStR8GH?{HvG&p zU3RydeWzo0(1Ra#z?HRJQxxUblWI8VM?Tc!68(dUK09a}191{NST1e{-`|LU3RnU` z%kg1r4hFE(RU;%xAI%zcVuQuWNiCUR(+OBiqMcSe@g2urBaP7O?2_DF5{EeVs?!Da z1{o3i&;W9HlRDDt2NeU@7R{}$;5Q7)B=b(PzVj{eh0`IY2xqz&&Z1xa);3+vF`eha zF1_5Hqr_0MbhKAxU+LsBQ|b4oh0Y()_(er#jU15jAFNxezPFU=b(G>&ZA%kx zdBHDtS#}y2lxlP4>OFdhSUvxl{hk`~^^h0tz`tNJ4%->ec=crkDxA&-*9?CgE<01Ieu_)tOfD@8ta#Z!7}fp7wW1>}KCm$TM%tJYWvGpnSz-28<2jQnT<- zg=2e{`|HqWBmU?H>{d5fE%yLiF9G)EY|r6U<2tm0mI-v5peX_t~aAGsZ!S4|W zUQ)>$UGw&VXNQ@<<+PbA(gtU#KdnS+W{h4nLZ-L=9fU)>8IFPQK{eGuaXE2a>Rl{l z>1-?>br}d;UL;XraB82xV9i$k_T|Z_jmOpghlS^Hl9Q#PNSv;#zB~$*Ig=KPvv7BB z-)(*gC*u4B;8K#XNo5t+neGoqEs9$2KZ&6b0O!mXLQ*hwCY-dDzpK}md>Yr{FQWa1 zGuLC;TW46bWQKV%{_5;S7zJJIp4`o?Y|aQ7?PT;Z!(Jy`{+^sSTj9NJ@1R3%019#{w~6K4en%jZ^TE}}L-iuip!<~t%>NLIQn?j9sKDsucG#XX!2OP;fqwRus0%gnFE3M8>Ik-mz}1vGqT z#!xo82Ech$;#{G-JqLD|SkoX=S$t+4kBjJOy|QUZLp-I&JBg(`S>sJLJ%AqkR>a!a ze9y;d9^d+d*ZDibm~W4>ExBCRjmYQ`M#3jSdalO7r!=()A*GRxeX_nE!0)2Ox6P zxr~H*x#!$}j=H8#9*7iM)_Ny(1Yl3@3eIhZv2-sL7Y^IC5@rudpqe{Ma}n|S3Zkj2 zd&?BisYK6J*}WzxZg>3hFS{jR8~HVAWT?XM^UB!Gwq)wxbsdQ>dz{{fruvxBp+**1 z9Nt~#oUx7GP`Y%Q%BM3O>z&QNj?^1jtY6~1|5$<+?sdAG$DJQAAAYiG|6I!0BR&VT zOcNJdBaeYo+5TW$>wd(=<-QSwp}L^ZV&4y+^2L5vAP8q6zvMkKF*k)+2zGotlhrJ4 zp-i+uOdxACRf^HszP-fRzC%FtTZQb?;Krz^eE1j9NnrRio6`Klr*&j#fv?>IYQ<|G zn;(h9-4MrBG;gQPA8`!;mxc8c`?NEWHQpv6{Uv$cd)wc14b7n^x$G zlaPhVYYqj3IllR}6+bHvJ{(NW; zB%U?nZkb>w(RNF>D=Xne3?MY6);?RcuMO3^e%N{sSV*n!qy4`!)@5Z~Acz7g`Q)5+ z>&A6makpsugQp6o{j1MaS<5Lu;~Z}KB0twfAq(WL#|W~JEZz&%4?5a!GGgU^Wg9QBv@{N^cyn^VOLqP;uXE?d$c3dRQ!$NLz6LwOLKQVw@_Ew$H39`DQf6k+LUe;ZB{yuh!6w+xJ(5w)=HJF|V~N zC-v1sK~vn9Xv$@?A=d}b5a)6_X*aTob@(2fMF$xZfwDd$gvTFMyW8|?i9f*+<^49T zD~Aup@g8BechqRTE<^UrIBOrf^YrR??~rh|JgW}SSJJzWnd~Yyhsq}DGi(J5BHc$d z9!8+8y@bQyA+RG>{mhPmOg*Z|K>p6$NbIsHPkGgGyB~;~QOm{AjqX&b?JeEW;&4eV9Q=N8X3xvO*s}I-0XQi`SFQE4_oD_-YdP%uFu7F!7$IeK6&vi{gxwQ z5U9j}mE$Onxd*C?aN)Fca--bR`XM5{b+>1I5v;@AtQ5NXZso9uy1~s9!Fpq>dQxUr zZ!~eND6Z1pK>j)n>49@4vACo0ZT*VIyY=Z$w=-UxbBr!Pw0L~&@XQWLbvUg_WZSJ~ zC~TxaPvBy>(qc)u_TXrk2tnf7zPW?Dw0p81vcoPcx-q`)_Pui{0V6+P{wpkE2fhyy zT}kD9+9Z4GEMRV&eRA8J`}{UzkD%}X$RO>!v51k`w$y&d@Lgu(rOPBUR;V>D3UT^f z=7BAUrG+((Tzs>1uB4YRa!i)v*cmRla@FUPXGf4lF+MGzRgzrYJFUYwR|$F)mQG#0ICcH4@-9 zVY)k5e_GJvWvz~98tq%E?H3wO5vyU=Q&@y{C z$9N#wrJPyj9HWnh&kmiV;Ty!CRpktd+zS!>H2qa}Q{n@k+C zuFKZ6Ms^S^N*sD<=mb=_%O=%~Uq^IL3tSyuFMGFbN9W1a3ukDK@?)cHbCe1PWz`H! z&C;BH5L@Bjm=zym9{+N~>RbL9UhnR+9WMNuu5{nielr&T{Ze4+jILuuHW+^P^Eb?5 zlI$<~@Y}9mSW4%etKk|+zOX$?y2({W)@|e1$l#UTy!_f>W(3!mX-g+B$}}jK>d_GT zDEAtH=qR^#rGy39FjTnFpAOnVJ7W1|s>{kVjeapK$Mfwub_#XT)cCB=i=G&4EKsq1UPI-StY|VXo~cYkcAp zQe?5!YN)^!kkbU1(LG@Ot4t2M8bi`ob;!j_A-Nd?bi8A5R6+VUZB;;Hdk|XXWr^uGPm3PDT zxrpXFy6D$doX^$UL);=oGrF}BxBNS>;tWgAm&n+Uh|0v+zGa0{(R>+PK@7wy-lC1nP<>k47? zlG9V+5|R~r7yvKn4>Ww1YL(1J!Hk-p4n#DVo9!^Vs1LWPc3E2 zXxhDV6`2TZW6Q!r*23}6SUC6^T|aQam>2T?**P}TreK|+7KOUy-H}Hv(XzLBX~0u) z&9$BV2GdSkNHXIIoD1uZ7|h9Wm|?IhA83{5Zuxwxb%o1eOJ=OtS?dCf47*v>B_7lr0!kzC-U8VaGx>gtV2WTloof-&AQQ94dT}!|>ejwr$oSG{( z_V8NX>Xy6Oyu6*JJE#D%Y^bl;g4TxEgbdU$GV0t9I_~kCvJ(00t?ozk`U6a(p1SB$~FhWjji5YL1^Cp|$1XbjRb&q_wp74tvW}jhZ@2{k@r4b@f-BW=GL! zcu}*M*8)43Cii%w{xvoE0w`6jCl|ax&C>5Yn==x98_BK91XL@O>n)s>!hE_NnuWw6 zcrf#stgyG{Ac7=wjjA}VG!2>19Qt_f%b9=4=JG0=&5dNt2W80f3a;@d?2v288?&Ku zyY7XQoJn1~^Y>D~nMEXYI?qX?)>8ZYM}-C?f9FdqD3PmXZ{DKHmrS|gPDBG3KO+%V zm7w3j^cii}j`lhhZNE*R`OcbmM7EA=o@ilh{>Xcc zUa9N@QKMm@Lw*Hm&`TEwW@ir)1m7~Rl~JeTl__fFL<#i5mfX;Ua2&Ny@$rI&leFEA z*ZtpqN?CU#;#`*&&*lf@TXAgwHNhVNe=v&I&Hi?)Xw!b5kpo)n5{OFX0MSYAQtQLL ziF_f#dLA1h1q=VP$x*G1;qvH@{o;d{-i_v~AJQze<=veBG(-;v(6#9@+(abMfA zb=~~3Ow5=Nh={)`T^Y2s3tn3k-4nIyrv|bgvqJggd*%#i2r?Sp;?*?t0MKV@d{Dqs z>ukwOO;EA=F;}@nL9dpXC~v>4QSOlAF;RQ_hnfas)2~FOkpIAWm|2 zRFovw~j&IfOt~~nnwNUdM|>*K5Aj}l4v>St@EUb!t&_bh5KNvH~<&= z3rO(sIS)7~Fx z%r2!0uU=-wqW3De+%!!sp$~}Z7L@e-It3x_XJ3O0(?C2?| zrV2#e(4yH)N8t8mN}U|8C;YyIV{9hM2n?3Fcxw=W4p8~9ex|i$SVVYNIBYXFEcac^ z+@Sh;8z#G{PUc9{8@)%GF-kb&rMW?XXP+VAZ~2GHq17udv;-hi5zsi;9qiQ-CF$&& zT=J&0-vrW|=4x_?Y-|+P+2(Jy8MTZLZfNf`gfamE9P|h=+1;Mdrlvw-c0OP*ALL3U zu=1~|7%qz-2|JsjWH;x)qI8{lEOHqgQ55AX+VSH7{a_pvxRtp>_7vfh$KH$s0vIZK zo)iU&ITPK$0it@}W<+EuP%?{GO!A2==R%wp*_Ek-{o;Jl1$(-mb=%;m67gDV)v z_8O>Kaa&Jr58ae5&p)jVDSVa5mc7rv_NfM-v96+kV*RMG3REu8mIRKopMqz^TkjUW0tY&}I@R7QW(ki=UN@`fCqKnv zQ7%GWLv>IAP^)i)wlaG4z~#0^Q*a`ou~HQo{K3~#LF-_b)p2cZwVZCo8kx5eV6b40 zhykhFZ#IZ96>_~?2XM(VeLDak#*GT-mi8tegv`hM16whzAUoKdS+McrNQ^=f-^vG~ zB6Z!-94SFY>39&`I<-tBV8;3mxhe*?>6hbUXcgB^o1<6*Sid(GX z=h9ZUV zWi~m`{0dH@dcOMACeqP@+Vo*!%R zLoU-Mj0YsDHLEE^dP0AVBY%!T6D;U#y<+VDg@3_scdzBTt9bnx;^XsIJ!w?$5X z3f)1*I4gZPnM=v}Xl^y`YbywgPEjP02#@T8}^4TT;~56r>Ll5*koi)TnS0#DodtqVgidn%^010BMfy#LN*Q^EeTN2F6>p7XA^x4g6re?>7e}rch4B?}N>KiaNxkj2@ z{6)UQnSc|3Z0J{cVgys(c6I=$C9GS`7^PgIt9RaEYV>n`nl_$MNiz)OxA|qijHCa8 zl?n-kLP!D%0(an*h1R#VS@9t;lB<%H{G9tJ`1T$|GaD2JeSAN#L(f#XB3SSzvo`EB zXxQn{Naot7>MGMC8;I8HS?Lvt6>*Nb^&VCLJeIb2`f5dBcoY&WtBK-j`kkuw?>2^| zZpV)~zWm#mAQCH1D;a}$S(T4-C#=3)V?1e4xoGW8oM&MW<4`qJggtcCOZh~{vN&60 zLHuD$ahIx>@aGi$k>ts+_7STat+)c^B)>U$g!=aw=C+V>%7mxOG0R@6Cm%lG8*|_uw-~*6O!7fDS zp+9w;1XqqA^p@M4{?K7m31Ik5AoSZGV^*21H41Um{L}sFSYdOJFMAtQT~dgpceH$Z zOpF~F_x?8rNDA9av^rPwcT68L>w>B3W7fE8wCxi~xxrh}IGmvP^mGMk`+KD9RCVq1Eu1Dy|6PRu$ z5^$E8poP%d!nZ*{X}q0&d3~3Mrj_$SyTW7;D~n~YEJfmrA8`v5k^*6uNsUmx%gq?+ z<13>oGnO7ZE0=rjaz130?SuhgU6^H(cTK0c$s6vYnuapBbN2K7XW^3~dAF>)o^jzDk48Uu7^c>iEo&i#u>1+aK7ZKD2jX&vV!f4;c#_M{E&Y zv3;QvBGf#Om}K;Q{b!>YVf21%NHAF%^;>dJ0d1fM=&g@gBjz+txdF^xPeaTA+3 z9Oe*lNgq%*-RECHC6`1bTL($1a_`KzIeuoG8EXYC8r(g#>)@y}W@_d4t;pb!d!k7r z>yvK3VPk@2{Jskkv+9Y-t3hXc-a$shyAJ*BK-)ZzG6+o!SR0)!WJX9ZcB)f=X70wCQEqIj_8^1e&HU z)k1PzmJ$~(Q_yAR0xnK$XB$^g6c)&L57cd!&K~g|4bh^%_iHnkdY#w)LL^l^YbLzP-q^WXpRYvA!#l9f_`iX1n!fU>cvzQz+2w|lM*am58`<;#5B`^=DSSxlny7|yuDzdu8M1{&V(ci&PE45EIe zK=82*(1CSuij#tbl_&YZZbm8AV+ac()g{IA>B?6MR+XgxOw`DL6Uo_KJ6pbl#fz}{ zAp``}CZbN2e55GdWPC~V)mBH`3L;U@O5saDYtir%J;Kinc;CX67*1$m6b6a~EypjD zrIGqRPY6EnO_GmUyp!#XFrx2S>n{-UyyM9cp_Bz4E-J6+zA)Z0&9Xs6Fz`$Z)70n; z~@P$!4k&8?3aTkXyBIhW1Xi;QNm*UfWyJw`*joF*z-rduCMOGz6L zXJwwq#l!oDIZVfOsYweZVM0a|Vo@$(RZO&(9ngg5Fbl)KGBWWW!k^xj?P;_%CF@w7 zd)IF-0Ehi{=aJTBb*klo z_XQT~n%GgRwN!&c#V9of=E`_b^d4Y8%wbA@kwjEyu!-pZQ=SbC0%E~qPkzOVkUB$` ztPWAt=XWmya&sh+D7GU+QSiO*sH5f|t_y_>WIKW4RDM z`=m9ZeTStzkb05Na5jMcIj_|5T5-xLK+CtIRQ;Wx;!CW-1ta@Ey^CH#e+S=YVDUEO z_#QUcQN%{)zR>XdsJ<@a1Id0yzPUJaf1nOjUqVjp@C4Xn$ zfC9#EjxgH1)zVB))%3f{3G9?W>cSE#O3x*bqV*|!5NM1<^A<24!slPHlBYh4HA1}p zQ@bDvO~_R4>llktN!sh>X!$}l4^72fiO5g;1HqBMixL4pr?7*>9vSGKY%NIlBEue~ zJWV#Z1jJ%ZjKqyj()*BcZ~I*&M!~uY2Ag=CpyDL$xRKh&h4h?%_qaa|2)NYV{MEmx zi_1mXW_(ARlaTVlo=JC8+4Y{{gqZRhLwyjMJ*Bi=iT+?W(C#-sJ@k1zS@y10%s`D{ zfu#TWuoZc01*E4gMYcC zfP|Ek%@d8#`TBhl4EHyt04x0&*i>h-=52;<9x}fR>i>s_euIIbU{6KkbEfo_>!kai zX8SLL|HnU5FaYvy*w69*zaTWF~Q4Y@!`ukqq}^>MxD;PxJkcmp-EMFBVvyyidMsI}+4u9MUCle?6dc z?GDiS^utd)yl=PCbo1`{Q68Z4jo-mf#{5m)<9~XG1RfL;u>e!v)00-IY^r9($DxF` zbS37AH2yl+MOJ9n$_J;Cq?a{z%#&}n_DXdq-wVWw@fzry<(1xWTYC|0dFzdUsgU8 zSdb_VsQ?>`Tsl|QD;;b?aWzKT;D!&Uq5uRj0L)0fn4Ze-fkkZgANK!mmkh-MayD0v zad*1ZN7C$G?4Ale6+EbXbWrbbR;xIgs?d8bC25M`2sDQOi?Z~GAh1AzMf9br+4Pf? z3pPOJ@oiIk2{DZ0aE(!KlxI@^&XTW*09x>7%t-g|3`gPxB!)BP3uUZ-XYzl$ME`yB zlvL0_&2xv9`7eJ#Kz&lY0CXqX3%ZcMv1dqpZcqU0!+((w|96%Q?eh!Zu<#;Q?6`nYiUVkD}fFU)>d7hBz=c({Y`mR+u@Z3 z>a{dAox!YLp8Q~jQSZf)ppeVs^Aq(f-K92NPPR0#^%sw$xc<&OCjT%U+fyT-^QM`6 z0Cl#oM9SiDcVWL&Jxf*ZA^*06No(qrklO2g~>hQ6pVet7&TRKLS%?~p4K#SOIWv4d9Ao^!hB zg5^4$)r-1n*ryPguabns`?tRR|L_XH?WYh6n|OwJ@#6UzA5D$vzxiYjQeQ1%bmR9D z3s1EM$G}Z2g*mzeIZC;>NVsrJNVxb-NJ!K|N|6`5r6CE1yiZ?g8sV-zK3rb%xbC>` zVDr>3okl?9NFA_E;Xt$$uaJEeP1MGSiEOuL6)ZZHaXIOs$ z$W__Au8z`^+1|RIzxj|GiO&H#Wt=R;$-U!=e`Wpt?>u2%LaEPfuFJyMw0T+*iJSh- zr9wb~vEim%4rWDg#4CmT3t#g8jzk|aLmHS4{~6-{7elF;!+gz+sOIxGgbNsw&lqJ} z2NUt_7Y=>?cM(}^e_25Pm=?-w7=MmbG_3=;m#JTh{-TKg_reJ7py?aLM|gJP|Mk!O zlbu}u!7C!P^*H);^yQnhN6Cu^8?yCVi}yasgHXEYkf>y6VxhvQWX-au%Ry48nWESr zR^8~Hz*dtl{}?#PE7W4>FYQf(u(nDdAvS8UcI^J`8-4L~#C|7KLR$zqiN7xDN;vt` z8mgH4d3P*;d8Gi>I3h@!+@C|gw0fKVFDU`nC0JgNV`SR1-KkS=+aVygZHPWW#}v6i z$5hVf9lFeZg07Ra<=v(67||e! z$YptP)aB4k5wdW+Pk*QgtVYPCmz)BBX$r16uo4QhHobbg@9ftB%?d*Z8JTIUZ>P5N z{gO@%%h@epw^d>laSp@|vNQc+@WCGjRae_5WeqGc!$-qg*)9MQ?YoB3+miyA#o=VKVHJ-BO^DEu}kT{H;w zqu|KcT4PVio6cmp-RkN&r{gED`39@JgI}bJ*;QtWZcD}&RkTVJc?i3cuBkIRT}89r z8IRCNgWuoe1W!IcUDx1RjV`~ODzLB##=GQ8;x>CrDV5NMaDKIM)O=rM% z+_2~5UBSw!^-_Z@VscH+K8N+#_b0vr(ee)OZQLk?2Ls#n9>)1<({v5XJ4Z7bl>wjx zmV>%3qE)xm;60;S4TjiPwC*PFNa`o>iSj7G>NPfN!7=dZN2PIwKkT}P6j^6Gb_)6X zPzlU|*huj*v|tH3#;yp=0WYaYGO>$y1q-gpHI`Y>$V3CChtZaa!<;^QgY;@;Ii{m& z3Jp~Sv`&Q>CQr<36e#QcA^CZ=#@}00xxM0mu04`@qbWHN>s`4(9$WGCY3Tb1qTzS= zd91VVBC+&m#j9vM>0o$0u2NJ>pL#x`@>~VRFv~NY&nHJ$ZZnOQuXp%gwlb-W$kXU9 zMod$_+N*!dR+Um^KC99q01%PM63*9v{m*i!A!n}h@@`? zF<+oxV1pJf@j1G0&d>=3p3wPRZ-OTBq#3t7&1l^!vb|d^9tzZ4B|T0~6dF8l3yZD8 zKA0v7r%pSimz9H&4%?yBdm07of0`+>^47%y8^Lbtb)mT4J@WU&w;zM)#i z4t#}xI(n^yvh5;uAuQ)xS2aB>78uAkrpz?22!NWYEKF$=51sS=Nt5yKq*)&yfF|$g zU5Zb2HZxY4TP55B3`hq*gzW^vjZ9|0+gZ#%96Gx#)u<@#EtXf~ZXNbVX7%Q4=ZqR( zs0?mJGO3=vi}==)%x}bWiO;D`6*y+PSy4|qJW%HN#4^`nggsrlvdAFdQhHG3xGHl# zKiR(mdY09&$H-jqzF!J;f7oFPK8@0)dIGcH>pR$QAMp*uFt#EP@|d%zl-?xD%BUSn zDw&M^G%(7hp7Sl|I7$aSWyqvsd|v^91QXxn=2aJAyG0MN%4Ha)kJq0IC{@#8ewsRg zFj*A5-0Y%#gTA+vV}x%$S@<>aIfN6o%k_+GZ&wyqDQ`5Fw6k{tKu<2cXV$Gk3B%{> zIy9`9OxSr!R=+HYQwKp%*M*1e}FCTYl`8z=T4pXM9Y+Zs& z?ibj?9n~hNIZ{|TyV1Q#vn?rtA1AWiJh)HSC2RXh6mi+?QDPZ%yR0}pUZb|LU6|eC z*{Ad%)6A4;$hgL(Xm2y$jb&2Of5g~7KZ+SiX89$71FnSYf(QckLr|Ixmh60Y>k;}e zBu%49l;mDq&IuSEuR=(8Zq+oOZr@ByQWZTw+nmMfpcj$;7DM4;V8-y_aJgmK!a~`i<%b~Sf}^t;HhiPDElh|t z2d#LTp}(Yfm^r3d*Vt1js%0sOdlwn@;%D!r0`JtMcdX6VHRy*>h0KAuV0xSOiHmzb z=qa!_qxo!u2KYU+Aunv)F8tNa`Dm*@W~Z+IHr&x+?Ml>)gMP_=8d%Qt?tqJSPE37G z6xOU~$}WaX?2UHPh}jT3IakWuC2Al=nejK3xmPeK(!y_d!~ClA2;4%X2KDG#RCZAF zw-q3Jk02m%SuUB8dvNI?%8%Yv?%#J6z` z&wJD1S?pFM=Xg3{yh#~BiU|Oao8?<7k5`^$w_X;Bd_|}H=G5wbiS3KY#X0m!v({23 zb-os^Jb*QEoc?Jg9zAao9J&6H{#18wDkIh^q}Fmt=o0@`K&L%^SG|wpQ~0A#@!;T) z#p>FoB2E}L0cyI@Lph1Z?ipqwJyab|NUUEDyZX@m0YdSoiUsHCKX4!j=?w-c(~A<+!F!!=BHw=HdmoanQlT^B?l(2-2o2b(ZXzK@bQnl zp)^Ks%V&YfQTGb#VZ8DhXZZtPbe#u@+9qk1P-TWR0lm8%snjI?{=`Yj2V))1>D%R{ z#;II@-OfRAEnCv;q%V@WD2YEFpG%b}Q^Hca&D7SL;+IQ%qDuu932e=SQ}I=`Fbusg%NdDY^Gnnx0)qD zV-gx~dNfA-`mW=GXKfRDkVJjxaK50Vy1=^3V#ZoVoyPlk$>mj`_!}hnM&GPLg`)CB zRn)i-;e^Y()51nYH9DVwwa$6PAYee&940lDOFOdEiEDruP~f1$<2JTQL{AY?CqAjd4v-2 zY0Q-q+-!i3`N2>eJ9>#O(S=YOIy@!Jc&{8#OydT%R_|B{@1JL92-?61=(L2{IUY@% zY^6}2fX@cn$D&>mzZJfMP*=hqtK|v9<mnVEzhm6VaZ8m3s zT0S*{+20AAwL%>IG-rZmgvxUb1*W9mn*SLo*v*g7GX;rpv2{6mQ{R;v``1xaS=ywhP6fqcP#>vaps88jl!K0S>oJX5 zd|!3n-7lMohtokn89sGmpNm$*y(esq7TR!32 zmk7drwBz!+YeWI$dDLQml=Eqe4Gra|7f^(p<{MaGUM zVt7E3Tt99PCZjGc=fmFc1Hx&fMT+$_@4Q3xGFKK)BuDy({9BaW({+LFccHhNjFq}D2WakG|S6GSxXDJak2V}oa{~H#)09K=Fjfe@S1f*0wy#n z48#PS@tC~Gp_WR<- z&~H;ih;JXy5ud;{XCu6%JTRZB$Vx%a!cFrgad_yGEL^F>TxxO)$$ak>;S4)idxy+_ zv)q(D($@W3<);rxN_KK_Wu3VV3ZyB~A>*;oq4WL}+z1%zILjaTm2)Gbd!@Q9GF}1C zMyE?O3_`=|;I_O}?Bd%~Cw-u%Z-HAd&i=aUv1neC6!7SU<9w<524x!uAZ%t z6;d@Sh6R6jy*hAf@Z2o)Ea;9PjyT&GU@C#rSYMkrV_m;G_$7a$-Bg-^hn>{l#)*=%Rmyye)nmWfWR~Z02W^r&fWT4(po29GNi_sKTFBS_- zH^-W zbXgH{Kf4wXwv7*+fIUs&SkU7VO#X(2eDKhb+PpQtmm?oYu@jxP9QOGoDp8d;iNMStnA9c=$YQao3V?wR+v`=GY zi(Fl!zg$0Bzf^YR@xm_yb@J78H{-GQSFL% z{6hY!hJPQFs>|8N2q`vZiQRN{>L>BzvYD3RiNbZjrv-R*(y~{)pLKWxX=%Og-a>_l z*)GF1%BAv4V>9cJQLF%&8L>Z=}y=li z@q@)^YWYvk01D!AkGMTu76EMlu8 z`upWbZYpcTC`D1oQ8c$7{%p@u1Vci+zGGrA3FrdCk)$s@CVv`a?XGr;Ph~6U>@9UV zd1zWZ-6Ip^4ovS$+F1I)?0ed(8pKaJ6kNNg>-Z*q;YIV9=fjV$ zP;N3>O|Xw|kOdHalxbdEn|^|Al!~R1dU=1zQj&B>eo&}fLPW^vsy>o218VLH$hkD^ zDZVv)s!_P2hlxz7spN5f5a$N1W)i3wh@ohgs}$GD<@7pNmUnwF=*%mJ?_NG73R$iq zLLfqwyQ=AHOUJ#1UFEZIdJ5N=dAQiGKac-#tt+%{T-kG(Z+b0fa&i`{r_>m}wCSXT zIwK9wMs7x(u&!?npDUmKWHjP-8i3Z5>UEDiFqhaxfVBWgw{3mUi%m3VjoCr2-qTIc zyVR(H&ChJ%NzjQJGH>yRO&z2Z5agN6>b9BVVu4ma!AN z&XY)lcyYIm1Rk5WF6vRtSmUzfwOhTF_@NWf0cb+IPEBsQyQ%+7QyW8&{BHyB{zi~J zE7v23E3bJ?hYsTxo|76v_12mwzRB!g=Vnp?2s%!CsR+1{`OP^kCEF=ym$iU~Y z{(y1nMYu*GlcJ6@5cDx@&mOUTZq*NZ_o_q-gWN<$D(wwrcL=eb=k;R{aCWCyCo79E z?0q+RIA1PO8qq0?2_>U?z0ce_c3WwhU~kVTnb6h^4QTw+OBK0wPV?G`1oj%MS8Mf zOat2er^kB?HdTiH-Ss-V)F`emhbxtVJ&%otB>vh0L;`a3@l3HRdJ*Z|>3k!M&mB%@Ao6Cw}W`(A3-Oy0gw z{*375E%vL6T->*l@`{z8?|oU8xTd)xcZ^^~n^%>WogX3Xl!@yY8Io7;6)uWoG%Dx! zSd`ORrm#0}_Q=FWMNu=mPS}6Ie{BU&l_-xSJlVUtqFK5%a{aUKTCT7fN#q7}zGrXD zWSYI)dlG}z`f_96w2`Osv93`s*Qgw4rfJ4%jzN#Y{k{5y)p&7+_Snp$-H|n`NVgrF zk{wuoMO;1CXH3uq;QivVYEr*Io3WyMl_;fMZJCB*X#6hAAQCae3dOt4S2(YNc{>Jsz$uarxiPDS<#B1I`ob-h%Dx>&OUgiKXk;}Yjb0BHw^ntyQ1Roq%}-n0 zCAqnPe)M7lMvWTH7O2anKX|{GhQ>Ol-AL5+Gb&Bf+x@#$Q}&TU_+EB9s|CP4nFgx_ z>|al*8DO6QA=!a0$Xn#V76*Xw>(inplbrCy#bY}m6Z3+HA=z8wJ%%Jco5-r2oV`Hu zu1Fw!mkTh=)awM*q93|W*3hegGMF2#*~Q)#YBRh;CY;iO^WlYUboQ*(vexi_rC!wO zYqGMe_rb@rewak3&dlAM8tnNR{*&!;DD~5Y5c@y4zJLbXRiFCd;ynPaFQ<^6sazHG zbkK0Q=`jRrxh2iQ8&3(nLXGAM-s?*Nc>cA|@-4Bh*xA=4Q8Ie0r7nyym4^W-VK&ndPIsC`NcfbSGr#P?%y4U2|kPU&mB! zwi4!ixJ+LQR<>g~LO{vHRUbZ^-Uaxd%mTaJcvo1vbPvUmEFGGj!OPI}-s(YLzNWw+__Rs# zQ;X9k3B%S$?C@^b|0l)H%lzP%r2UJD(Mxh)rq5$YytU^W9ipYs4%9`}!?_)<2x6Jl zzEFh{ZlRU#t<)J1?v0fO93WHs?gz?m^EFxRuI8Y2oR! zR#Gd6i{$mC8mxN3GQZ-+uL9JuocvtTe7UJ33H&2dTXx7nf*JAGLd6MhbznM3g@;lF zBz6$n3A+{tc8ws9n7#AIi;bb9LxSUlU^H39KeG(=4=2_p^n|AqSjq@J(yP!&0~Bwz zmLUSiT2t_be7u#F+dxYt;LBB!KSC!3iVoiQ*GZ-lcO4V$jlRB3N?zyAF_iqgUA{& zwf)w;?hkVy$Q&~I+kB$G8e~Q}qEQ$eW>W{MrfOporn3Q059yGsAB9M%Ta^jG|LAc&P|)bob*JN(D@*IL{8P2^m)chw zU065=HxQD~&=pSQGtXAUF(z$VIef&+7E2nN%jw(?o)i6O5RZt%a}0h_Y#7TuvGxk! z6&uScc30o${nSU3#A_8tAb54W%z$mLLZCiFqf?qxh4DgQBa&ao{B-bUB^xZ>?H7xt zPCuSH8Gj*X(Dzd7zS?YdT6@gr6PZ<#@`NT1mI(M0?{^zy+BsoQ9s}$NL5gvpf1MOl z%SA8q6^PThLmQ|0M;utzDpM>rH` zFI2I|f4p8H4zj|5?~7+zBPq3?+bdB1K{pWR%J{$FY-Pu?y3j zvPZM8v%tqfFM_pDtUX>-=^hX>SEun+;pNQkG{mQZ*Ws>-T%%&E;M8bxjJ+-HAV%*F z?Z6u=U^C6hA-`$9oJkgDc%!2^-hSDyWIDviFq+20e2m=1InW`;Uw%_oOzws!|1Dj> z>jSKUJ}Q~?76MerI~?gKSRuS(|2E$*njgC|NuerQIMvxD(>ql#?kt<+=~v@dt-@LBUYcH0l?sNQQma0Rc#h#4w&eN#3bb9FeCn zFTMJMQEVDu6#MX=p&TZ>rSiHl%*3vc5S&&it}Y^?9n`A7v>Obz&kKp!8cjnRa?eVP z9RA95;449v;fh3h@VT^AWy-Y3RvA?y0mvAjR_g4egAG{Ij1anAkpj=k`L^UpZ+7OJ z)2j}A-8=EnL^%d752vTI!0O#9;(wSBCY5mS3-P&WX_zF_VI7)EzwanoEnlYE;V>BR z+V6-7(gf58xSj4oZ-0@9T@8JmXo#vNXuQBVKi#L~gdrZcaVkS2ft$f)LfVngq9n*e7CJklvas;#` zKR19yT6?P`frCMhS1VWslbpaw?qJCf z+b#1xo5}g9H!!p()Mb&_&{@$6bD2ua_48)ZDE0QDJ)1W*xsXMYmKMIfh~3u(VN*z^ zD0^?aJT7s%pEK6(m^~r#&RTucR_a%S64#k6s#2WnA(9Z(>j(6| zx%j+T91G431c^RS?!A`Q%cYdD9kAT0HJ>z?qv*H^g*^4Rlb7jq&}>~3{$84|nPg14 z?sya@@**E2@m8Fv5^`USmoJeBMRU|RC$PQa(BYX^J6dTX8|v_R6Ch9lDbD)y=+T0N zoW@sTH zxekKJf9SF$?dD)XaRPNBRCaaG;E4^^&KqmstCj0mB8DTFxSy1Ik~syXiB~B##w7O%RN?U1w}276oCa&vgCO7!L)RC#Z*mQ$n^}1DCsBexeo6Wq zcFn5{sMZYQs^`MLF&Clp-7FD-;cERHAq`Jo_JNYKoTDhHw~1DZQ=!Q4%K+>PGaUtgr15`DyyT|iazvuaZXeC zkmNAM6b?{!Rd0f=5CsE#O`plin7yguEg2{fNgiY{pN7vc{6&w29G^HeVO}OE;&YdU zCpT$2YKb^xR6X`|c3!NGaJ~WP1Or@@QW*o#NNf|4;8=#`+re`)(~5hFUo}~YYu|pL zRBt>R6>=p%>w_@Vr+t{(p>pyH^KMOur)nh->Xf%n*9 z(NnJ9D8xx#UVTDzHF76eiUW5LS@(IMZ+z^f-~9e+v$&14o1V6xr(~N(t!=#^6Ndib zN#L$bcACiPzSi^2OT#RCAuSWcj+#n}LpD=H0(_|2N@qN9aTj32uKE*b$OXRs=e(qq%gN>sb zhrUkuaw7`w3iSAM&F9Pl^WB36FRX6wL+lyo1K+xKqzltvAJLY>QCj=4<}_9Wy+OFJ za7fIEK|s&_Va(U|%${>o%G$nja+;d{NB>Ji-}L!NKy7w(qul#wrXN&{o*wID)rtQqQ=tQ7 zj2J%LTfXRGq}bxXR4C{p;2H#Wdo%p0T(*@}$Wlj-KC1Gc=MwQ3PfII?_YQ`Z0*{ei z+@nQlbQaKdQN#B`_vxK4R1kihZ3FB@f>#X;R*Ut5<*$=ZBCT-DdttW^p1Gf^qKqL? zuQ($X3)PbJ+n?^gPE@M%u`Y*N>!#?pr|TNB(Wc7>sSG@{RT-KSZB*)h9J0lyx-2m( zco|+E?w(?rzAKj;!B=($))MR}L?t=Y7mNJVBv);f5)?uc+ z_p^trxT>(<;=SduL&HqdScMzCoZ7=I4ah;XN5A;uKRvN(@Tzb`5?GqV*)6Ea*%T

>5?nALj(>}qhDS0f>Axo_tKn*WoVi!gl0#AE`LkuB2b=Yq8E7)8 zQ&oK}#IRPMGI);GiDm0A|N1Q{lh(!@k4A z(BGwd%ucsfydGy-5DV5GPUArp_6NJ1RV85c81sFk=i8d0F#)tFGLXkb)~b(c191q< za8Hue#X40him+&k>-Aa)sTLi%RJ9PRE-e6mrha#<_(S?2QFs@GTFJA|n^P2ifD8`o zS|vzKS!UfBP>dY%{2Sp|4tG@~e`kfjRR_>lQ)&UjCpnTwDKjjV7D+`uZ{N#gXE_tl z0Yqu$fK5!p=@ci1PJ&eMQkXeOZ}DnT{4llsKf zKEtz4nGS#0Ka@?1s167O<07mRl`0v!|9M}pf`~q&erNc#xg9>tbFZGD8HCW{X?J^v zzJWc-{m=vxZ;K0Vd_;tOBXl6q#t#5`(@cl22VMo-y2|o$>LaPz_dw|Dr=8vt7>Gmr zT>aZhxNRo)`K3UbGrI%YD?K3F##CJr-*&?U>1;kNpvG?Ez(wy_M*bWm0GX z?vY`I(Z?0bcUl(tQAi-0CiHq@Uq0QMFz5+`+QUn)k@VllZnbEgIu};f^UW`H3Xr?+ zR3BWQc}=Bvc@j8~c=Aoa--hf1wOd{qbtB8~-YiY`#b$cjwBS~DsJT3PV|hdya?PzR zugfY2pX8gQQ&Q^gzq5CgoniLsE@B*U+)FQI?d0^%%Hhit7`JrVOAqD>>TUV1L8$wW z-eSZkxW}fZWek5;iXR|SESPKS`|XeLV5xq;+*l~`8Yp0X6Bo@#dv3a2^!ZCax(kZ@wTtLjCpRHq@hfheMfUO`m)J37oN;0AyvvPxv65QP7i}$nUGJ?L=-au zv`Jczsnw8R-(P!unpP*!RFN^Cd1#|pDjs!TQPo#lEpRi=0Da!gVNjv$4%_2Qql0rF z)E#3bznDHdab@UDgzmLqx&&e?U~>e+r^hf}r3Pw0O)7@VfN+}ba}Vc$k&Z--~#DTQQGnv(4F=5QArtGWYyo(DYvA-^mYr&0D5Q4 zpF2t^j@Y}Gy4;0WbHs4t$Dgur=Hg55F9OXe0HO2ebPiUxJa9EzRS}nUgGvY_Ofj7W z*`E1?2$ku(uLQsKu}KK$qQ<7i8r1m4jhar2EI!S*gs!cZx2m-NFf+xS-ldpqz->1e zp{GzK@bxi_<0wGA1FiAo3npBi%hf@VC_NQPaGf=TN7NgHLsQE8IqU=RmZEjdYuHjx z!z}^#V{9^^0e0z`_}nJI0M?EJ!CU`;mv-l^GAj$VJM#fd-4J1LB{exR?Yf|jgzXhc?yrata}?wHYt zv_Js6&T(ohZ+8FCc6aGyc(@_tbQgjealJ?M{dvw6pgeam!lkwrp~KYseCZv8ubQi; z-<(H&UE5Lv0#m%2*Ra7(%@~{?}0wB=xj#E6W5M^#p zOuL<9x%l8Ig$A9CdKMkmTe<@wRM0}_dZIz@EN;ROle7A9VhJ8xI4xhmp!3@e960_N z&z4V7o+_(>fh0@Po98Z7D(Cs@Kd*Co!L2fBUF+z*8a)FR>O*)$hHP*aA-XVc2TSNs z9oi^qMJ#kEJ)S2X6@o>tE|`F@R&ezDQSD--YU*)T34~Y7wZR&6^x~vVWrpIUsmFXJ zJF#^10n<5qmG)Iq;;YWxV`H0MO9(MaXw4nCfS)y+x4=JrZg8spHfg)0O593v7fo*4 z2w<*<(VX_eP&UHT-}A4v6Qz7Vtv>qk*O2S3?m7C8#+Vkg-_*mlAp2eO!!jo~vIE6V zMqYlOQ77QI)z1--tW1_GXKO)jgghbV)urMvi(Mo9kWS?^il^6|-!2#4A^8lZZNVr37RD}7$|S^k2(2dY3AN0OG+cxS@CVfC9X0+TQ6r;Jbgc5 zOz)HQ&Bl(hX+UXd{|(4-Q+$exmg$y&l+=c84)p&1RpT zk%JUS@wO|1Toj|QT+7jtq`3zwz-(I# zO-1(7*Fb{lWeZx*{`!1J2Zr-w2P4yLlOHH9dUjd_#&!vPUGPR1!q|M#BQ*U|)hG?W z%%JG8P&0AzxcI`Gd(-rd5|#%PYT&Ig@&CCxl2gfIs~hg<`v?}{(a3v{6(gz=+mp!l#>&v=L^*dr{^ z15tQ6w1wFH`+D>A?glL!WU%^mK+?=%dYQSM^{n`hnL4w~SmD;`RzDRgo3zRY>_q{0 z`w{|K%^7`Pb>4j`vrQUZ%p*V#V>14V_ySiu!jB1TkjoRIHGS#m3vvcL<5=@Br$GxW zI6>7y9L}!@&I#L&meDixhANzMva%<28q~^vEzFL}qKmnxvnlGoIc=X?&n8sfFKAR zw~VJ??--wO{^NS-w-m8F7VMzRNzgSXK$D~4PRXtXHOw_^qa0f5b9Vo#ezj%`VFI*= zWGMy}+dQfqZO=(bfj3Cr4X&r7oSN<@no=dNd)g9e1e zy)Hha==gPBL1W@k1y4R%?D1MqH5d;!V!*2u=T3xM|4^E7DD`YJRwL!cT6dBg;tttU z+$2E5nk;kU3e5wAMqsm}fItz4x7_3iDL9Y2rRPAzBftoCdFDYzgl@o z4u+2JPMJ!E$nNxxXu|m{e<;%pzrA?uJij$rrbKJsN}NTww90NIdg|+BIQZVtrChAR zo4i#@vrz=tdQ7}$&`*Zk(KI6@kK>276fdpRhpB}?f}N-66d^>3mq$xRxBFCo22bn3 zbUJQ&Z4RY(EMq+%6&?TBVRhLjr;oJ8LNZreF}0|BQAxf?RgSW=;sO z{Uz!EAkx>^)UQ}rgP*!p$;rVY{Wz(!m04`XN`nsi%VAiI+M+-lfRcb0n!e)0Ujq2c zsn9e(44z5qM&^z2(bJ4k$)v~v57F4|zNn)doQd2a2Od9zj-(7eFDal4sZj92L&^vC z)cu0m>#56b5HO9Gt5*q#A;TfJ5-Q}ji67q|`#7r*3*W=+HWE-WnLcyGKb6(;xHPZU zCpRG_eQ+VU*kgqRSuNLw0Wg%$y=V!DJN*-eam>P88Vx2@yEeS2$o}M5pCv=aB_K$o zRnNpaY15_ImZX=ImXYY-$+4n~HR4~hSu}39MQn51g~^;B%sW`fA3N-=7N%yT4>dvX zDM=vF-s#-+1@+}A%>f%iA?juuTjk{St0W|Fx#8H1Q^3;Azvs@P#Pui~#&^|efD_sy zWHH}@>a;(}0KtQuh_qjk9Qsoar=Gx_gvAvJn2u(|ZanV|6yRLiu56zd5K0pTf-xjD zBhkF?>C?DX_zFw#(_i*>`3FZ0W?|68Ol&y~(b)g$r!85`)g~%dC}loesF6#+By(qL zFpsBN`QW<6U#Son^yVnHWyCP+t}?mG*>COBPJpB)JUaXgK2$g`}h9cDXwKN-3acGvU#7q&O*h%J8QtqnhK5TkeAWg5VBgx-|aea3*GFqvLmbceDV%C-}VVLq*jk$>*{2mok6OCr&q>cBAqaL?_1WXjqZv^_q@E~>KKrVeBUK~u~%xAOFCg>6`y8B z=cMJ*4vBKHkUH<&iI$JgFm^rmJp=kwOZeTUp_b zPB3ydsTb^!Tcg#4!1~dd8@!^JCxL3-oaB$msL?-cx)e}MUM~_m+ZTC|lf>+|a$y4Z zj;bxM&lBB&tr6a#Pyw(TY&Nk!prgLzG#T(~y?*Dgr{R<;<(&lkyi}`Fus2!trp_Fb zZMNQ|#^#9xji2G$N=;jeKu;J(tr&A<;CM*MtOa=`_SjPxpeUFJ?HYCl*(4|e$){na z{ZPJl^EHMqDP(YT#RONV$wW`{YOT-kELjrwJGy$u^JrM}R>w>u72~KCj9?GxXC%TA ziP!kVaCs)lNopofGO*{ZV~rl_uaLO(Xk9(xj^Z-#EEmx ztU^cGwcFMcla|Mw+;_YEkJcU4D7&W2HW|qRUw4n&&fca!o(Mpj_X)3;{KxnKPqawz zB|I+~V#7v&TLkFS5V$}FUo^0F;j(ar1-IeL)s}9Qi z=jcWsT*k`0^!Wi}y9U(r#NZ_T1Z|OrR;84(D{Y;)Ge@P=>g+egGfT|6Qn+$UN+57i z;!Sw&#dFHxamz-pSu0_Z$SOx@0c`E1X8_PY_H?B}SI^i~PmA%;FD@CE=-D1*Y-*D- z2%G+`W5a;#!#VoJF>dYWg_ck62X%Maj}0tUD!?;a845?9H$_gtZ#yP!<{P6;WYcLM z?%v=E(J1D}Y?~=%eipRE_?p6ANZR8+gMV;;5#ix@B*UM{U+KY_;j%v!{MWv|3Gg@5 zr?E$Qb9rmcETe#^CUQh`3(7SV3d#icY2A~S-4g4qclYAySU6&?Kl2I&ATlVN@P5SU z*c?qw)NnzWBrB58Mf_Z(bRU3ZehryDo+9G+mU;7Zh)gjUm+k)5ugyhN!{>;5^3MMF z70P?D0g-m)(c)2OHy@v0h;N8E`JBZTq zz&NRi!Kpfab-zOChI=9wl`hf=`C>f*M9;F`9%{@51;>AZ<|v&eARxoo6dzu_rPAjW zWf&B8%?+TJ`hEtuD;y=jq>Y-6-JK3K2?RU0UsE+A1q*?0G-`{f1Z(oLzWyF-pFr(q z$bPB{GL1t4O2?oIp+PD}iF|&2i)Va*wyyzT93TLOH|@E6#@{Fhv;KS2NGXkaRVd*m z6M@i|;j)l^{~C1XU4zUc?M0<cQ(-wvD|s_>GY5e z=3CJ}05I-b_ZlV>$)iiO&Peb)vg8_aw;a%nh7|rWK-5D<ccU+=d1lFO~U<@bOa+_DM?7gdE*!exC^X^hhlTT=jY0RNL)j>*w9?ctko|s2a zEXA@Q&o?%6w811kdL&DI7V2-E47!UyE8gWOw!Ky$K1a!eN;ALY4{;*??f0Imh<&<5 zT?HNnj_5=eV(_%}70wh|wYPUh&g`rXI#81}FV`1+yYcS)Ig%Jh6&*nBQA z64_?LInS_S=hl_oeSCJfJGVmwT5Y0fRy1P^hVw+FRlm)3PUX9L`o8Q~lPAF!Yid3N z3O1@VA?b1I^_I&KTuz9$Cm}jPmD|pK6qbBaA^Wm(n`3|OPq=lbTl$k z(4hme&Op?)?%^k#?D;yU=U<(Vy{|5Fx(zv*o#*x?U7DC~;8By#S~j2=_`5J3xEzMoAYC z&HE4)a+LCRl3PP>CpYFoVV&VXt(5xQrXl-l>F?biJ(P1~jOSlmAUi@%{d#C19aS0& zil2DBw(z}Fs|spsUp^Px4BMoZJHn6CfMiH~BhH`!4bg)A^qET){1<)zO~dg^qDFcT z@y*NF*P!yaN-gOvX76G763C+xEB`){hycjod#e<%cV&j2R_iPx!9a3Wo8r*2Sb9rJ zjaJXL+J(3e;)3R)z3!QZ0L@g~Dv+XFa=J_>Iq2OkGNY#2ZB2{?Y*1@&yn^{-(16j@UI%AASGPoUG@YBpOvt`r^!B5d_?q!ydpP z9uUt~N`V3S<=4NqoiAe+fFB--~b^8w2&g&s&S~dYhsqsxG0rhh^CN> zdviHQU%1)ln9vvtYohty5?*#24%Kj<^3%>yI(OU$59ht}N)`4(9u@axt~8a9f^}Di zh-+D)umiW$ldwbk=P+{ixX)CNKENSh9pTGe#A zj;}*Am#Xnty{mD>fl2*6LWViZR?NxmInf^4_I};_k#hqWM`dS)R;x(?Dsni3c^S4r z^mqgvmZz2URgf~>w{qE&nJ41=0t!|`Bpsz(=+$vlLW5Y+N^ zimXDjeSY{ajaxnoCkb~U&1$i23akq@XzEVkH1$OZPvl4^DU-x?_W}B7(xjIFlP~L8 zlgDv}H00<7g>e!HvH4n)F`!9?%)A85@_FcE`p8d#Dgk~0-PyB`$S3RFyWSqrOB7*q ze!aga>dE8ug@&mg0yf_!Htrcbb#$5F{jPTqVx!!^WYngC#QiY|K;Ndf9uEP&h*78 zs@_$;5GhyI3}T(Nb-|6YxyW~kI@~qN`y%BbMM zVv8s7+OQ; zJ>9z5L_k((uR>`M&Q`MmgkBU^82aLK4#QrqkTj#Ov*RX^Nb4+f&o#LpY1SX2fro+V zPNYXWX=J)}7ZJd(RFW|BoK_ zw`XB&gW3G09?gc^gqw#K6N%@y_eY&$*dr`0yO|*e$4SOmJ3XP$EHislEPHK5nKYIz z6m(TO?M_?YOT7&E3jd&p*H{%E5%|0v!o55Bu6k2XG$3A%G*kb&13gCWT0{nTTM-;<5iVT@)9ao+Cc4e2wTE`_uUSf3-}7(M$N$bIB2Ghdef(8%GWNBcej6*#c!vSw}fCY^2iT;lL#ArDbqaw>oQiIj*y7mqa4d zucx2u8iq!FF}?Q+Zz9ka+t|8%miRhkW`%z}U0v9WFy+BKoZ}S$z1t)W(AS#-TuM_% zP`GcF$rGWgqm|5faS=8_e+lf74n58rOBI9KHr|#dN0qA>RpV%j=WouM#VVKE#%ZegECn(XG`w zB|O+9)$PhWrjJ5rTO-*RQZp9Qe4QH+oH&!b+v~!J4d%(aB5+e1dau33+bo@|cSGv= za4>2zU+66B9J--wTG$d?$H;J27IWvdu~RU?;6_T&Sn6G6CFGE3>$4Z$()!SKjkdB> z#Eib_63Y?9jy>Zjm2s^vU6Z)ii3g>bG8q_6Bpso8g~jcDi3_eQ?|ep}vHXhO2;YSi z74!I4F)PbkU^!q=ess{5CyrmpL{WgInv0@uEIBsGgH5X|3@mG_bt&|*@o*7eQG)+u z`4(s|L5mYfLUbW5-3i>Md@%HD4kVZ|zkikpcBOhLc^OmE17di4GqBatx0Bq867A=I zbS*PIwttKI>alDzHIUg74`!*~+Ar@Hc!FMl3+mD*!F+(|ip)-xZeyPW#Pjm@^8|Cq zLh55~e$H2MU9?=29w;KWHu``N=9(DY&smQ%RsPi30iqWFMT^PstqrqtSAh*6@>jS9 zip*h#@D*Xvrw8N|nMnj{zrkxrh=7hqipo%3{hlCGqEU;-uyPh-+@{k+ z=%Fy+D|$e?0{KWWZ7Ke4$Hqyl`u*7{d)}wqcx|DAz+KA+%QhW&BEA$-h5CKN^P^AO z!0l`-fz1NA0Oqtkxla3($ZnoK*W7g|lnzlE_`jX&(F4_%SKgF{u-I&}-jtJzCd?Io zoLpPow9^jmisLW(y(u;y8zVlf6!_epCzI6U$88Ew=W*Mt_5jf`_hi{^(0QIzqVP_RYsHh>Ox^WcGkR_h*C&~;0>14f+oCzEuOhn#x%yCN z+vHm~T@VXnr&m6@w@N2@+W`{$OD9onCh>7?;+2N}cW0ut^XXmOVi&?8YEIaQ`q(QS z1B#~cheMjs+Kn`Y`0_2pqC}?(qbnU%^JxVnJ0R2H!yoQ;HVmgYXPo}B+8v-G-!1ArW}~+I+xO^VJ6X@>>pZNYv<2vI&Zn z`nc8A;6VAXqt8S2{FiYFW}K-DC7OG8ayZexhDde|H7VoQUqF4d2Py2;ZOK@6S7n4o z_nx*1RUVtqa;@Xp!~nz0@NWSm*myxzPDYX3G|vP|2<#|Boo3PTFc`i!ZOv(fW#V(s zO%&bx2rki$ex{J=x8?6Ye}DH^cOIF4@=>B8@`xjC>jffT*kZqIK2PuQZ<|NZ2SKBP zQ(=wxu}g~9uYd^%=c?yHc})s`wn+KKORs0s_4;$Rt%NIZYta`k$=b34EtRHuo&cqibs0*DnZ#o`L5y39!oao=bX35>e~V0z8Rf z9=oH_4ygoMY;ms>)RSq_@K_L8nFi3G_p;;J4F@L`;MDKf1I-yk%Rh8_udQQtR6C+M z|B;U8kM;i4=cyt99jX#r`t;TmTzvIN>bY6=^$!b6z~`n_Ua`@DMf0WVlA~`T zJF+1ykxitOSx}Fq@!?D4{N@W3UZMvbb?i4PJdg_cF_A3beCOAPsq6#(ca;pTu)_iw zRN}z~hP`+iHSL*FuR=oF&>_?-q8CakE`XK{!54A`b~##;CCG*F9RlZqHofRSDDr?m z*ef0$PnK6MZo}Oe3MyR?bYpAkcPu)};-5eIR=s;z-aEkJtNPARe-!ToR;e#J?4&*7 z-BYPGg5Q`sAsRUlu6arcG@}CECvtGIX%770nQXrlRY{&*p0+r_D%?*_@11~ki;zx8 z$P#DB`VZEQ&I)270O_IsSRf*}6KguAZ5yn0tKXu0Y?xQo78|D`vGjhJV!3IJtS|NI z^FXxn_Ayh3+M;N3jL@Qpa1?O+V z73@ZwXm0ianQNv`d8kKOQXQOL)g-Zx$q)x>C($+TbMNR{R3PY>f@gfvId%>7xPAG{ z#7l@IzvrY)2VVX+#*<0_Ai{G@di$GtZXA3`jru7ok`d_J`}h{ni!q*CXRM9{b^tE4 zh^@_-nRI$^1`wpNzIj^UDH8>6V9ldHS;+g?OwjNU4d((HN8!=*X zm6>vT+5^J~ejj;>2X zr?Ej?|E^sp+D`P_|L6W~XphOAidWjH2!7#h2@ekea|P6+3a&F1O84ToYn`bfwdm_` z54D++*$5!&muhb`CWi2qupk7LZ1QnAb$5fKYFTlv&Xj1nXc`jqXZYWnE*30S$|tU> z5C-5I1);H~e-kv^8EOov(g1Oc`3!pHnvh zg^3?&&71FbCcA%gWAHAQ1DuxGdn$C}{rWFue5zV;Lwk_jc%6Ufm1v!?R(1|ctcyUz zvnc5oq#1vpI;k*#P?U9?#jB{Lfl|E0j=%M?W)aLq8^u7{y3L24jdOtN6!K8eFvXqV z72nfI_lAI3wt&~$5xXaZaG4?#*r&+@s3nQ^n4WO?Y1&O1pOvNa>#;s^tZPXyLk!4w ze9@+z;#hb~23KnmA*=^*h)af;7_AsfVHQ@TFUkWrA7dXr?{)D%>rbTUcw9cGNwXS4 z*{j8fUtI%N{PHvc1_oK+tjlNSRQP=A!DSk(}TtJh@JmWX}W- zr(+p_OBHYUd;#t)@7>!1WNhz*3$dKaf=IG+^yOmH3jiN5nRi0=3x^W?4y|G>trsug zZV{msS596w8Vz%2c;3OD=X>MNa&%I-n=+&!Hx|jDbMqqY)Ghm#uVlTZOB(=sYy4Mn z&953i2^vN6Tp)SzZ7iR*4?P%&)W*&gNQ?FHcD7EYYZD3L>zn(RC|Ak}84=&j0;naL za97m{1jc*M(2qW_^dOY?WEq9q7*vAui4lX@e@e)7OFf06=YMTOqJ`1871l{tuC@d6@P0 zLC!r8uZ=$Y*865(zHV@Hx9c5$R?Rl!c;kr@c`5+O0;0WRN;me#W}S2Tw3cqQrR{_? z<|SnT)c|Qn^04R3$cg>)QjS&AYw2UV++g|q*T&@vR+Sf5+rzuMiU4!YYU35H0OF~n z)h1=%I6x1(jVt%&|2u2o8@guhNro z7al2GQCVjg*;dz}CqnQO-gJHpMOeQ}F-U%@cVy9QuU7s1e4bS*j_LW~-0hn`K&|0U z8s6LK-cOq53u+EV?mX{esUqUMM2vradQ!G~_>(J@?F;pbpTm`H{(}j)0IOLo46yj_@GaR{lT?W%iI-#o!hA$OCAq{MCXLjrF%=X9A3h zNwbKhN^C_wawiP*;MG_m5h=?ybvc0h7h?`6%fKFpJ{RDtrMmqT37INVoNhkv+vtn? z_=%SJxKgqw924z0`Dvnp_s;w#tWpv}5@-G|$>P;lU3bdIHk&{kt+wrPgu8GSCZg}$ zn*dEQC+djm#P9`|*H&7Z(!H%2l_1nUr`GT9uDSLCPFw4EJsGdzsT=sEatR)P`- zA48~L#_djIwdnY^)404;w_-@$6jMHUy#TbvRSLDg7G=Ar z)2?6de2ktxviARLR^@8=76DAFaI(>>A>eI_Tyd1RGi5drt7Y1|BbRl=E4Qz-Y*x18 z+b?!W9v#*L?nu0?gy?!a;4`)V5DxRDv2wx5tRYwZA2^R>GMUx9Kg+~$pqA?}GtQZ! z+N|I{4kIgbU;i_%XbIi*G9Y6G=@#YKQZoSvleRJHtXr}yUbntN$zt1`oDcFq|9!rz zRHB_97Lh`s`EhS4H+a!Vtx_wy;&7qE45cRoH?~(M#i+uDxMSo!yV+VAr{k}2lB$G@ z;bAh3S~KMqNAs!`HIg|O>k!9{Tc;FGC)#ul=V6(%j$k6ivyaxlGzPz~wC3|yoMGEP zMBX&qUG-djn%ceV3f^?7SI8C1cf@6p&)=IiZX4f3ONbz0nasW%ISh3CmgI5e?)0!~q^+&oy`dsAg z(PZ!FZkH~izHsq)(e#?_`{O$qR=-U+%dC<2Y+3vPvOmikR&va92(S4T2bP<>QWw~) zX^*w<;gHO(8?!Df^5&b}Sn1WFUkLeJiE9LWgxdXt)x$*|H9vkPKS$^IfqtmOKz`M# zx;a=K^=mXSO2cI*tC}T>`kmK{&)=c?abVHctSD5nPM2})xAQSKXS@M0e}xc+$1gHVWR{j%V~ zzYMlst-$QF@n=DypKk5*nyA|>=6P<(HBB(%e~=2h0_H%HM;0dixwIYRsiOdG3Drhd zPBhhbMsJU>$WVVu&INLiiOUh|2fi$v{1f*)%|}8-!-w+(Qg=1l`0p^nnK!7UpT~if z@2}dQ0Ol^4NSp!YW+I7~Mhd&f>+DLp?UBx;u>C`3pEbGN?eN6m9p`X!+D<@(|3%6| znP3a*6tqI9BEt{RfsA9$9Q^V-C~z9eZ@(P_WC5kc&#P_1K$U9(axpC5s+_b_IIJ^5@e*kwtO$NNc=k*AA1`gIyFlx~u z@a<*LwF|+Yg`U{JwO$>;2Bisk?Y=dJwj7J46@|uPD6$j_G8l+f;zr+(lPL;Z;1XF2P^- z@S{kO08)6WJ-9pSbwaocDjbz;g{(goc^UQ~wS%c_PG_K>H-6S+1kMG;*HqBCug=X{ z$oBK~?oRkKrQNZTL1Jj&cMVRjgFl)6>(lHE3{5inqc?D;gMu8jW=Cu^Z&kf-QPxxV*P*<`vpo}k|eem^M3e|y%3so64g)%vL8yMyk4I_<(7jL%gRH55fF z&)3PU_B~C|<@)3m-`Vv#u_7VZTZF~<)+%a%_;Wzg`?7M`c6Jg}=bJnA;Z##fpk&>1 z%kN9O`Ow?ba7jdjRge&4>+#R|3|@zvu`_JjFa1R8e4Aw(qX<<7lPQ?U)PjCLb>)F! zL0~#G>VIDX{>`S=fqyYZqz%+3-Zx}yNTW;Nv@;IRD?f z^=A8js4K_F&o#^5@!6un=BHm;x6*g3Mt5I@*=_V)Q4d!SQGI*-8BF9Wtc*~Vd_?>M z=CK`lPl$2#gJ1)&u(smsj-~G%&@ko&{(Q4K=Jk6*{Jg@LenRP1AV}Pkidoc?-Hy2C zS4#^qfDK_J-c{%%Wde(tKIVOC`o8fs9<7B*@2hzR)H|OMMR0DS&Oy`sPbs1RGVD9r zOrYL#lc5=cErYi@-=OX+c1yPA3eT|We(H$oV=rvJdzt5Azk&$szNGr$BJeFg4nVOg z6h!R$8}9<(oX~IB-U{!sqmGF4ELQ6|G3gZLuiiQnQe#0OzRg)f6ucqHXY z9g||ffKH8&5ZryKQR_M{l?cxTG4jn%1(;1zL+pQUTkm`+%jmybyrlFW)JL@qfuwy( z=1Df^>`R9tpXjfB{{AzXyk0+MpbF}7+x?uA`NVmREzj$>C?af}I${E7tHw%bGD-2{ zorjzdbz?PF!Uyzxo1ssUV9tB3ByPi@)nxxw3d!#KaXUY8i1BQ>Iyg|gzbiFR4w0fa zGLyEPuf!gxL##BoVfj&Dtke)*5`vERfPJHp&VGn!4dCS*CCH!_wN&ypnbOh~l zgOGFcDG7asUOXyT~&hOVv`F_CK10EvHCx zGbANedLARx-RJl4b)SXMo9FKkbjFOlS8ax9shpIFBgVc?q-GS&2hmys5XGJajP)+G zh1dG=h%;t(K9l*MKHUTuR{-xHQl+8P*rSqjPLDrO&6|n>L7GJTC$VmsB;60^Q!#M+ z2F>u=$=RFy!|%kANQldzyKfU}@DPy$NAPCPICPPA@31zreIjt#LublYtzHlRjCu^@ zdmSFFN>_sz>~CE8+<>9heT0j(T*a0UV+!`v0kT%7+w;?<+k1Oem;yynl^Z5l_FD6A zTxm%}9UC>k?!3PNZ;h~`OXy>RLAg(kM>Bn*_n|jpz-wl)f6}oge5zS4d)jYKj2Hd+ zL#i-D$&-0dCP}lS#V05*LH*emXeS@^{iBtg2i?{AmQRuXg>3yea65$+=yOYUr`$^xhE zA@d?!gnAaFDNul|8K=K??Br1S`Xyw?FKaIav@L+>7ic7h95LfTfQYp-25J;Ax{QLp zc(zXp!gXWWH7_3;Y7Y9c2SD5MRG6&>u5vPvAVy8ZSXxzw26$D!-|Zu|ae%`*@%olw zSAlik@?v1i21VL!pkG<<^n7uV1KXN`)s*fKilk@AGoYZ+`b7G<)j&MColq2(zL27< z#MJk7ygdY8NE!!`~kMM%3EOs*-@n!dn$w|O>%$fJ|fqc(HVk>m+R9X-! zrDA75MIx9pSDpaN;6%2MxnhosHJi!l%|toCENgc;W)CB{6ZV^y3~>?ax?wd56!Y>N zad`6WT2BnSFsjCs-Ufk*-0VRr!bxy7o4|2;O-X(hM>sRhq$cDOO1C8$ibnC3sIq`}Z%p-_MNMk@A#<=CuHu|B=YCPfnx>|e^`{VIpLyrezeYmlj z<^6c0cXDP=+JYgKcf0onWjgAjNiUj%sORT^{CNQm$TR=zliQ2b9JzMJyL_kj*8*`} zZvW`7b4^FWkt3R{@emIPJ` zqC2+^bJ@-R<3lOw%M)A#mCMe&P2a;H(jd z5`={a=m+2>_==bRxH%suA@0#`f9g+w2|Z^-27b_*-kc(RSDd}uh5WdPIfesL5-rcy z-XukhZ0!8k&vPxPciiT$oa;>1sp|8`#{9urE3w*SR75g{JWf-Kkni;6r&Q)(Xdrz( zIzF!fu}ZP>XcQVJ4LTzDIhHeg0@e0^{ak-_gbgB+}n&Jp#dTKSDeQEbg{Pf=PsNtIULB8Ao&--gq#B;iYI}$8mVym z=ObyX|7g(yjsTXgsoc&@=4|(>(;tLrd9$50z{GQ{f+H2n4zcM&^*IS4ZM`uV+M2{i z`q$yHZ%O|@(6<5*@-OAMWTIg>Sf#yA2LEIS`@0Fm=-x~2emPN-4R7ob1AIxq-Rm`B{}4e)>Q8ME>6$FuqIz#4WpV za|UWSJ8W9;|MlYg7XQfu>7z$L9>iB%bc(CyRp-_Und7ZH| z8qjNG^vS%AoMK)y{|$X@jEJZZ$CKNZwXDOK$5_*q*RfX<``C5MD9cdtO$H~c!x7Im z`#*LZ{pT-n|8zdGoqqbLXAUOx|94s^pZOXAHK7J-FO>02{%-}c|3^#L|JUS2YzS1; z#*4KLyaIpiF|_@km#+E35r4DdqR=>y`Xbc@{(qE>{ZsFx(Vx=E$$s|vFZorjDF06F zwzjHVQUUa}k5^z(lKHb*<$V#9RLcAX@)CS!5itY7}mLwI* z&>0RE?)r#M`!ao`B3Z}c(m3tO;n_@L{>Qaw62xwZf!r|5F~XbuaT3n`2-aUq zHDj{Zfb$~epT4L6{c$2k@YVHs(!M0(ATBf=qTJpbjK@z_ZIVIMpIvKzN+5u_Sek7c zjWL*JI;l=RGZOnBe}*q`XtCdqfm&3XyUX2S?Ii97Bi@T2-NR-{S`-0LaZ?yf;fI;4^A z?(PO*q`RespV9_F}a(O`7&)BFs+q6Ufh*iqU)5kigsZHxbdUks29g1S_uhrbq@j?kMG zsUp-C+pxC=;<{3CeBSw6*h`~F_9Wv4!QTQvvOfYK-#kf>`Wpu~<|E@HdpU??^KZTQ z#Ur@#Sr+@>fZIOB3JC%}clEfj&VO#SpJM%dWPO@G`Aq${SfAAxk{X6{*+8ASt2@9^ z(I%RRhhmV3Cee|d`9HrgUB1a5cVttx;#K6oPt|-$;E?B&tNH6hV7n6E-`ZWe)wfs`0Y=LGy8+we)n7F#@`ybrJ z|8}Y|dI+7zYb>2~e)!8l%}z8kH4!M}2LSXc`p)G~jGpJ|qj^8}G2s6@rwy6>1gOqg zfp!0B-ElXPej))L#O2S4;leefGL}b$y7{W@yy?z}Vh(R7@;`QZu@O%aP467$n0a=% z9lM)%r5nmL<8lyzx2T6j3vERUJ;W24Wzg|BZArc~|Ml)aKcpYeFeJqzCsFWS59iyp z(H`K9X_eFVg;b$$nLSG0IsFX|v%O+D{Z*}&Dl4S8aQN{*{yYYV0d7FM@-{n+W7oS* zZwB&;*#E|7eTV;)aqtR0CL{ygs&`}0@!x&gQ2-d&OT>Kdi@fpml5x6{|AxWw+oF67 zRRK70Wfa@s-#B6bfZ1@)mq_Dn_~7rq@p6=pkHP+Gpvkkufekk9zwrnWFeE>uVRuF4pj>^{Gq~+>tJOB zg6WUnFuSlKQ=D)TWeZKls;pSo{~#;Q;}X+weXe=x#4$?zZ<}PlxWunc=8Z&fxC-X( zycGEBbDF1ElF$5PQ>efT{qIy7mMud1vQDZ0k&}kGy*BXt_iyTh1%lQzbgAIM`Hdzw z7MWiJdCYcNDeciPL7h^*w4wUvWO4B6YZl?Z9{dR@MVH{7kTZuS&j0m={qktOZsSKE zEqs*GmGY(0e1gR-jA+ozpl-nocK1J_0RWPD^?)YpSWok2)KOP1OixLpf>G<;KE3fE zlTJ~Vj%u6vv_tRnyZc~L(~(8=I`wod&aDUzdAERaoqDw+$c#2HvCh0X_$o=}eA@+{ zvvTb+rsuJ{5?`ZMuB_>pnaZ$O#1_0%55aGS9NE?w@5;ZKe^4s+Fs@IYl2t&cW}R<# zN^0wLtoK^j=ns?r-t&BA)7;gb)OWFZ>a`XuQ7e|yN)2}9e$>}Pp)80_%=@mgy;>59 zLAxM+3!FrwjB|+I@f%1=z7(rBMy-nDDm`a+AP1GH!@dl~QaRNr^?uWC+akW+jP|+F zY%TaTW1H+gvz?^xAsR&@K;K`{CQ?DG{8R41{h%*bfQRHd;__{goxb%deX8KK4$$eK z$JcfD!(jpC>*Qxlr16ZM%25}MKqE~&u4XV9klk?oawVdNz?E%BHBG<$&+{Dn&yF14 z^J#b9X$h|8$!a`o4(oeG>NuW$+ELtq>a?JEy7I0q$-pvgmfk0ktnhkxdk}h+Qbbe4Qr$$%=c{%IHmk9iGCLdWfGfAn<68c5%g1SJ z`}oU8j`3PLu?ZqLPYb{f+edAit)yjuZiU#dn@l9MpZ-BI&fImIT%TJ7jnd4Y^(|gX zc%{d2E_c9hP4^~YCbw>DUW|EMk*P+xL9@Eq=SXAmwi26*dc>Z*&TMHAuTaj5@mhnAZ6a!WTRy1cz~*=Jbd43&7(9guS^N z4dI>(opzb=m7k?*Ou?=-l-CLg52=+Mu}rC4GS^|a+i7V$*d5j>@*~-Ox~5oynGgN1 zl;;x$T0L{Vt_S}r1*=UUL^+y*f66!&h^L=dV;B7F>Q>l00rUX4A^AoQm7UERysk&Z zaT=|Tp*r3ZLAH)Ii!R?AS3&BeZV>Or<+a*y2>#*dw@6Z%l^1UNzt8p4m>V>x7T-@2 z+HT}oOxkrR5)+GLCh4aouI3qq;iCq)CAG7RXuCOA>Sudvrtmq{coA9bo~-vOm*}Wo z18BXKtke%3EY0$YW>~1GjZXymT76e2slYmgJ`((#l5b}_XJ}0R?6GZ94`tD0%+hq@ zMdUj80?|=?+m7co943@!61s6Vz)+hLibqheAK5sZr0+u+XKKgNBe1{5TND2l^q+@Anp#6ohU?Y+J5f zefA%djgicV3AiYnUsXcXmX+I0zMtcM(+kig5B4V5Vg2eR51GCNB2U(R_LQl%QZN?0n_ zanl!%QJr%$e;>Fh9{MJ9u3xkyeEdZiv>tlG_^*dD=BwO)d7{hWrURk-d1E8g*0=91 z-@PW}ijr9bg?15iiwbxKLPuQ}En**Pze$#uQ1Mhi7wyT$-}4UwRpo!=EEh(S{^DM{X6(W z^H`|f>_1(f!iDwy?wMPTv-I9`k4dcqgepCBSlpXhy=gE}h32VRNESUZEwv|`CbV;T zcld_$_$Qra4dV0YKltxc*!1O;FWrB$38e7f2JPncsamLiG0TggQC{!3|I)fVTa|L4 z5+~^6B}7}U^*$7bO7zay@OOIh0OCxT8@`xLqF$_N}+)e>cG(8DHrbEykQ zbE8b~=b$JFuUPttP4A7$PBeOkZ@>T+L(`Okkev3IUwwtZFz@2HY&+D6Qv%V5{@?v3Z=Utq=Y<=?7NGgOn*P-h2bH(KA#j_-oV_eU^$0ND4 zM7eL;Fj^kDRXfH7wrQMA1SVL-hFuAckg%(66_9V`M?5X<5w}|%4qF}fi@*1#1hSGq z=C1`88sOSa(sok}2U;CQ^+&a^Be+~40GzyL87V!#2QJ>rvY*|CjN{uGu8#V!(Y*%N z0_>aU2iUJ#?CnrPN*B5bmX*2OTD#CBMqTRirGl%!h?A$z)eZ2|U*J$#tDOT6ErLjX zYsF)>YK@qc`$M-$Wfc1fJVGQHtEyz!lmZ{3no%U*Zd&d3f|9~8G=U*mM8|de-r%Ajn5b0XAaxB5I z36qn!XrV-5pW7DUQ@-1#VwIV}OS7ikHa@$Ko01>eOSx*@L&buVfGmVioc>GaiNB!w zRv5$9539=8r$S0d!XB&VkjhEuQ7d#<>KYX%M!5G|dXaU(bxYw^1#N85!1Kvh%-@i4 zT=OwkLlGP91JLtV59#QagGrjykM`z27#*=WIQT$QUAP%XreW23n44lRF)Y}h#f2Br zeR_}hYR{oMN78+8`+ubZgym9C$E*c5$)-!LAlgM zY*`#LAqZo?VJDiXp}OT~xgYgYVEP^`VBIcbv}I1UQzOz=!W^DCDjDHiJ0U)UxiD~& zryox;Zfi?U!>iziS2kMQ_*Pz6%p;o~6@&2T;bo7?KQay0s~VoJoitk=qOun~znagk zvS{kbko_xh!FYkwg_1V5d1r_--z6OssC~8PO&3Sm^JRt^&Xx5LDXLlhFrldQeBvUy2BkBDg;Acx-SW$atPQs}zpJ-apHnc6l;8O>%r&#V_Dkrgy_F z(8J6~7{W#(=2Mk%)0ZWvynQ20u6chK&UBVlgVu1`8x8ef!F?T= zUeeFNyrg`N{MEL|BUKLoQm*N}7b@%TKT!`ovIb15v%1usqDn?4c{EfGb0mLIB!e#7 zXUraLOeabHP&Bbf;`aLNZf<*MJr}C(sPPRQVQq#$^G}4)wf!G|C z)?b|~K6K``mhh17mT_>ujo%#`cgW`wTf(*ahwxh1wD@qzX6cvDvE4OJ( zNGeg>m|d8NGv*Y8q$tEbsb;OBDX^>1Raw=j(ijzb;MZ(Cvw4g>$^1w`cT)5^ZgC_B z%bw?QXN!d4&h9T;m5sfm(U$f`2=en`GUJexS2R| zYgi&}CDTQ=U?j1C>`Q!l+Bvln@;_cvln4Iv&vWpRWAIp#3#U-sI;)nKUvonCAVAbD zp~fszO-8`XI2PJapqA_U%&))SKS`fDeAQj4#uOEPR&=_X8NvD@i>|!!4kH$Zv)U*Q?Wn4 zi8?pMHZ3D9o`1-1A%3^|@+`bBc#A) zTcG;1%sHQj!^TPVbfGK|VGvc36vvl5i|yxjp8QcPPraY432TzvLMl5E!7{eWv#kw2 z@OmAG?Sx6Mt|x2Z&qMQKRp+<0=fz+@(^TcJ_5TF#gl24;jFATETC@ewyiI1B)~Y!% zZO)O*?!xMdeOBgbXi@2gSEvzrD&7^iF$p8aY6esQ+dtUzx6Tl*U389pf%Q2?+i!ht z!YG2a{ZU7q0_H)${DM>Yk-?{(#WucFAiKX*KTo8WYbbSb0 zn{m?62*YQguhKxoFRy}xk7%1NHP}I&LZlQ!6ndHZ!*(44N*L^uL;OGSRGB;9+T1KZ z+$U*=>3RgMMmGou|GtKGBe+cm#DXna=p5##cpri3@(V8zSFRt%zNiS+_f?nz|tLrF9R=v}G&ny6SJG^tqfTYFKL?kJnn_gB9vGy$$b-A4|1g|6Oi!Irnxl?`cKYF3g4F0{&#u8HsWGXe zFypmEMd@cjRU$^A`h)fa1W-|>sVb@Os>j}KN&k@$hxEK^>CS8RtToCC)hUfYU`)9_ zwjLrlZd~-das>={Fs*b;`5p7jF z`_l_+UOm?X$sHYDR|$4#UkiM>ZuTo9j&NXYcs@lbaql|Vb|s8*IJ(aXasCr)`6}G> zX(a001uM0zBk6@(`)jnV|smL zUfiCeTY@+H-tZ6&P0h{_KHFpudjazZ-|rl^HPZ2OtIN7v%nP#7TI)YBqt0D&z7R0K zU1$AvE4h-yqS_gFpOfdQcfQjub{lj%b~NX|ii(ME!?WKPMJxo_ePZliDx-(XtO-lo z<6yVciNQ4XxoCIin%lIkC*oneUx&{uJ+uTxRK@*q;VU=peN6O)SPJSououTZx2>K} zO&tx^5?LH7q1ma>M748{T(;xxIw3oPaW#v45F^5;$~OJ3#wuxx z7p6m}(Nv`qOA~!_=+$Yh(Ww{wNo!Z%-jDDYZVJA*<7WIU+_mK*1i#Alu6HRF)UEkU zHiKe7JH_X@sW zt00IOvZ{ClMlD{eoYwzP|580hbpH5r+Hrh-S-Tf(xVhQMN*-U=+d;AJV<#tr`tVQR zMB>00$dp{_)^@WK^ms0JAX>q+dEeO74=_)lIDE5_$w$EDgwSmBnXYHCcdJh;^OOlL zeuaA|mh#>W1llp8oT>Pb5A8_xN}{pT>y8b*;mTm?j2{!am9@gN*Z*zSn#HeQvag|+ zV8yeli>B%rEb7STIy(9}qy)q_2NxU2yXm#aEg ziARzLfvkH=#l5;JrQ zhhNA*K09N}DqCY%^STFJL$RrH&e$@gR^8>9CBi=#m&3E09TN5KeOJcX=L43WJu3`X zC^y~nm#L?B?(LxCb^+4KxFgraq3~%l3|V6SFNKG%LHG!>bJhP8By+*G4IQKOHacr!$Wm-u(7vD%Y9iC zzs?ls;tTMgOiTqs5`&x@yrw^zFJIBIyWB9^On^JM`@Tc>XA^~^0a!J~szM#M+(O6g zy@bcfdPhabfDV4#2(V6bRwPTZr^58Od|U?45MdU=n!V5n3vl$=ep!*kwCt>)#rF@^ z|%oJ~E#aS2&_T{}U{z3{HD%Htd(rcWgVS+~VphxvF0+&0=WLleOZZ6Z*w~ zlc9VR&(%_3I~J{CoA1pELPY}BUlZ_SyNaow0;siF=YE7k}h3Zdpj0^;mV}d_CDbw9W}5H zyySOtFtYpMFqPK3xLNk4b%#z;2%qE!yX#58av1ps?eyk;CVdqE?dYRd6zJOc?sW-hWf|8&0`6n(imu%MnMqez*UT8;a@gsE7ZU5MP5y= zXOJ-CV;|8|nfDVs?WrPz-CZoO8%@uDwt0EBmbJ>nUiCofXBwK{`Kg0!wwHqsdY$6A zHp<}6f2KGV<3O-j1Y6ZNMM7QBZ*lLaOlXPuy8k;{;?~N6V1Nsv&hmN*<&&oI+uxy zLJONDx%(!n|CwI`T^MGBH`T9<<2<1`{xM2wtvSHn0iJjG<;-;W%tE}doyl5JxTslU{+$3oWp)STD}JX~^n&b+Ozvg}Ux1dPi1R%o-xiO; z{>;dPH{h{oWWib0-@9B|w?{vay(KK@v*82X3<(4iBhSa|&K@p--xvy>1!EkftoPOz z4M&Bnu~s?%UPeLGjx>jeugKC@BmX_`v-9vV7TFvK;irEaJCf*ZNPR;A2(PC-+@ z*D+{ar*|OIG(%O*E?Mvyk~AO?UYXi!KHgc)MpJEHf3a73G)TIiE3ny>`{i(P64;Uo zbGymn7j;&Dh~jg0j{A^^^IDx{hmLK@>UTEOUnDHSbe@j za~QS@kN7Xu<<-3U46vueY`l4R?j{-YhDjeiv65LI4H+OZH@K6hL8!WEY#hg9VPy*s zCFvynii8G|9*|GYA&lWr4?;zX+o^E3A1LFl4excJw1J(4+zN(V_I_J#0P;w&>O% z;gmwLqS5z(M=~G}WxyFV??+P&X&jAA4R*PfI|SI^N3nBttm_NaV-^fX5ViJ0tD!A; zLTSEKEshl;%SZ}6khm9TeC_4D3)B_=WpL)8c`xH4@j`7tVD;G0qFju8U58)801J86 zfz1DF?>p>$q!*v7E*vlZu=$Ws_yJ9wJ)iSTABFZfBx>1lyxEnvTvnw4{5*M??=7au ztWVZ_A4HjY-P@0xF<%%)_9`pX+cxWJpRw*5z{w;o4=SL1X`RL5E>z;)ya`PMvX;+L zczP0E$Aw15a~A8nK3FoB(i9tJBdeYM#z2p=Az7o!TF0I4@#sak+o3Zk`rI8uUlcmIE--?i>;vtn%_sitJDyn*Q1a>NHR`Z;UmB&Ex zRvpigqv_78`9Q-D%KKdEsVjK~G$0WDsc>Xpy+vZ%S^)O@AE6@(+G7y@QifrUB^_?4 z`c8;NcJsL%O&V#`?V5{DWc04zXplHHV|ibA zHf1&(_y$HjrAZ%3T8SFL%@@W1djMqQ(kQfxJ z%uNuH-0uzO9#hz55RS-Laa`uiY(m@IX?XMpQhkPVJke8b zWfvUbVaS9Ef7VAdMvv-y$Xjs1B;3OsLUyWSghg_=`Uzh*YXCV_x51qmSP#vVK(9yG zU|ze^R(5l~T0uJb7cVoSHyBfcV*0q9BJmy%Ve3nYKv2IR_jKiBTdvZwEd*WZch@fT znq$Ru6L>4NrAu11CVYF1j`vPGvYx&hO_+`r9c`XMe}txm>L&hBjtj}B@>A2ghmeX< z5J>o=oLQ%>=)qKaC^ed(R+Vf6?uXlNikk+`n7;^d5R{slq@o$BAEw$WsxhjF!3;pd zbebXqF$t;6fR6bua?`u58=Q;CTC#`>*vfKY7Bf^Z2=0X*jq^N zPd#ZOwgrS{U4=5XL#8te3Ux0(3zPhdsPP?&U#^!B&h7~S+Z^~|gKVz%1G~lZN!h3F z?%1R`)4-Vu^B~<)wXztq4C!lwI!(x|^6so}B%gOSBar#cDUA6F>o{xrA>43vR(9Me zUmnX`|Gp^Tb;&bbis$*LQUO%yrOD>m#Hq=?4X3hXdN=yC-n@o?y}7y(_i!7+BEAC( zixsLHj2hkke79-ws!s>xq%iBcCTEPL-#>>KVL0k~E||u00F>KZU7?4vO`N7X$Gy@% zwW~-lH}d8;{`)4-?tq5|SwMY-9Dv8{gmV-sHcR8~O_!&JV$I-*@G|xye@Kf*Y8;?@W#koMGVZCF)G@ZU#~^x2PL~hr9E! zxmygOHxP}z_G58c)%GuP3uXIgO}frQR4o;EEnbZsX!lmRh>l#d(%0#Nv|}^z++!j4 zvWx41x^``6VZYq(J-B={Xha2i+XH*y-oGa#wo78ER#@lREDESP$x*~SV(6^ohhoq5 zp9v_5gHH^SwiGPl?8%Yqa{h_jbO;3mfr#m`>)11Da054ymCH<#GZ1OjoZbg$?p)cTT`-nN^x>yJ)+&uCwGS z>W88WyehYkdlB73u_w3)K0WzXew_il-!wn!yZu^1liz-`wDK&Sp`vBB`7UhI_69O* z=M1mT?L%5T9*)ILvl>=+J2)#1WtWiSwRiw}+z!L4Z4=G)AQhxJPI*lfryn2eb_*&& z-kf@#rDDt#YrqHHhp5;5a*4v)&TB<1zZ6 zz4GI5G?YJ1Uu$wm{42G-jnKqz{z`+#q6OGn&3+TCEuz`Er-hh}HXN?C>}AO_$e2_- zY)~CHfxZ+Dziu>tAb|;4%J11IF2o#bH7>6glqUXu<&oHE_b^(lcmAR9<9B7WCJJB5 z%r~U59au`-&%+ZZK&BPtJG8=q0RVaPwK;J;e+jtYOZqo*hmofl#(Cw`s3XNOHj3H- zU#n+<SPOqFoDzCgu$w zn>ptoG8LFq@|X3A0QHW^WwH%M2J_}HayAXxF=`NDG5u+-JLVWJV#Le09NXu#J}M|@ z^bO35eAY`_BtWZLQ4(-lfXL6=fn{r%_&d5LcTKD{(iO~I)eV)^RocMY+_oJ#|J>pD z8m7}k%yIwmx@~le)T8aXff{Y`<7Rx*FC~ATbcu_9us1UR_GYCf%x~!^%SIkZofPcg z(pFj%-y||ok8Jpy!#{)(qQQeKf+qm>CNv7X-1U?%B)|dtOsQDf#;m)CD(c6{3|suq znUwPl?-kN6QkSW#ly?Cexucs4jGr&wrgzl*++XFuUIQ_YEg?ok zp6X=zu*ruK5!DG5tOPue-rf_dr`i|I?JU^LmIROlCtqUv@Z1>akoppg1vyn;HWo0N zYfobZM`2O5CMg*?O8A6iS4Ml;WmsT8@rfSFBp$nxSTt(0*a(IUQuL$m+6@OdpT6>3 zvWg$Ft=t{;qI#F2@z0vu1^8vsBot{|34yPXNZ+@XX)DEi0Z<8;xmwGwHlP?2wB`Df zS7bA?^Rrsr>vZ3Kx_1ZdHR3`KX!(53wA3}K zx?)d=hvwAl^RSL84<=-p1Vz`;?cUW~Q_UU2?DX7jT#60`Znx6SbUN2MJZCJl`7$qC z+j|y9Ynpx10)#s9;nhBu)i<{MRT{hyB`)T(l_F*NL!moAUh>`7sn1o8-8Ow24m7!5 zUA|uWH02p2fc~I_KIZ~j;a!7Dv>(_{J#~eEJIEJ92ZQs#XO5i`#`PLSo*WneJbTZ?v6X)nnbKdG~?+php zE&gPz8EsEp886jZpb!^$ZEo1|Z$^8Wamx{nzf17-Jm#C^f}t%vEzvuy=H3)cg*22W zaLu>`oLLthmg66HPOk|Ih%QFgF+=XIPGz|c%S)+wn=((V)SzuKx5kAj8Fz0plx$RQ zV?&OL3o~7Ou?OrNg`KW`3f*Cj-jRbSTExcb@`F%uQfa2C5LiR9Po8np&-t`_25)C4 z?l>^49W=892MXWOjg3tw;KbRi6H%+kZR>UzsY+s;Qi_*i zpo^Z_u2B1{HVe<_!%z!TVNtJA({AUCY4!}Qy6wmS(j0PY;f~W>ZXz**Z*jn^|B%F{ zoBD0V>edAtHHf_8Be(2l1B6M}Qv<(H{kZ5;?83#N*4ts_REE)FmBwl>5beT~h?^;W@2=B;vzg z_&u!#v9vj-Tn)Vc?KN68<$P|tux`EX`_sc{tm#-6*Nd1A0)#00O$M=$`jMB-=5Nse zJ>1?bAL~~G{>k>np!;>Pp1wP%3-KqUQxDJm>mxam%e z*K94-e9`KpUbpPnXs}ieiSL4GUChN?|3`!Q@fj2pW9)R+u%{phjUXJmX#PNXu=x!4 z9OMsUsO0b=IZMXeL~v@?7`Tz_Medpac?MOV3?`#(itu+@V7gKntwgz4o6p%C-^#@3 z^RJ;y-DZ$RKZfGS`wB|OMixn!nNj~+LJSP~-NDaWD+u1vXMe-01~#d!RQNL(u!-PV z2L$S#1e5(@8M#Pn4AO=3=^lTyH@$cDW?#BCA41*@J>pz$b8RWldd3#pJbR$zX4T1; zK(@0xw$rj3`LGKfY;hn@WZ#`@=b07|GPUP`wVe)fw!Ej^_{NjbQxTkSGRT}eG*ytF zyL?v$a#1Yd7rmO<_=f6|9Oc4-F12~2%{;ZML>-s0a9@4+dbf0Dj3*eR6+1i@c)BF$ zIaklSGLYcJgJGOvPl%81c7CHnV2IZPG@vO^S|`2M9CRrU5E!O8;tL=#oVNO{Xg_CJZoGMa)_b>NO!PT|Ai1 z(iY8YIIeI}uq^v}@Swd=9U%h)#&nTs&ScO0Yb$F1Hnj|YP#)~O7cm(#6j4*KyFU{7 zh0x?JO6y54x0ud)TJ;CDdhr$e)tj{H9t2dWe-vMy=?j@r1JKW^W4ShhE8V%YMw$X( zLyLZQsua_m!GX({9z5(G@mGe!e(rie8vl$NT`6R4Ox_EQMA|avMLpfm$K=+V?OR$ew4?nz$VU{Grjcr08drq}GT8&)C7j$Qu2nYEu(g=KsD(P<89GLH>=B?`m7 z)cRo{kB*9Z2I*FIzM)U6W{Gfp0(+Ji-8Rbl(h@zweU1=+w&Iu1<<3$dXWSGrl5o}G zug6+L66Vmto-wHpQpOvdhwrP;+E#A};x-D1EN~s>;e}F^=2qI3D5)v5^1xl}p(3jb z>SyU480BCb$vQk+Wr^cN#*=do!M`I~WJ3nE?THbR?)_*%MHOvI8d&hDR=+UqG4N_z zWW4<`AFN7LSN|<;t|pO-f6-jFh|>FfmbiVDd{CS_yh^yI!d~r6R>Z)gR}OjsC0ni| zzVVGaHU+~wGyKPYZoN4sXb6|^wJMX@Xs0q?Hq@Ux0M-tRv~#{AD;{|d6UElhYe?Ta z2fCC76{gY8Cpl&t%gMa@?RMVoEl@kHToMv4x-8w^V{PG&x%FJM?GBc%M4R#W->6JC zYAkMFOsfx&|L+Piw*w-@OAdrU9akM8E8Pj?)`%0ZJWHYvAmIvR0%?adS~I9NF_(Fh z^D}y015|eLe!yz?$R_hVAopOvf}_7Q?M4bUoDaG<^nVysnddxSkT*O(AHJ&K6&fkk z$DmlOIp@#zrQ{mp6H4DKm@4%71iOnTAO7J*fSHyq|Na(Z^$o-4fEG2k%Ah8U(3Sj0 zPH1ZCS<~$JfpdVE^7lYpATe(0z;yKsO((6j(-sWiWO5=|%Hr=`%dsPge|(^R;Gdv8l_RG477WY^)ie(KmIC&zkDkjTRkhl7gW zP8=(78~n*TGzBl%)Q*)djOCl$jozusc=QDbr0;uvEN1o2F9dV__YDZx=}R7c{e7|( zF@Gd`16uEV?J0OtDnz4pjkn;e5Rg^ zjA*BU$IdJ{Z<=@=I1@$|9QwBQjyjNbpg%ll2&wERB9BHtD|DfO8TY*~MtsKHn8rbO zU&-78cC5wqOVR_@lnAq7*@G30(hjccqC~y(S<~7rJh6D;QM__7PuoLhgUBa|M!%AN z8<5HyV1`$Lmv1j-g}3IAm13gF%o3;kIfFEaj_9b5*bH&xx1LWtqn-n_@ap7U?1`#m zeHYBfTan?$9H)D@x5@cdN*!du-l<;COXVk;>SpyBpS>AMaYSU&3`Y5KUPhR$n&%wp z%Wa5%7r{w^Y+}v!@{Tn$5+b~1S!UlSIuGsNDVG#1<4~(5Lq6l}{pnSC*f2VFmS)2? z)oN1KjFuS3;oXk+@~?WKI=7VxHM+QYD3wVskC#P>2TLa&bI37I>%;r)SBunPRbU)s zaB}zA>qRW%?SU;)8JiyIthsE#WY+b-wbhf=&APrp$DtjqOXXQtcwN#Hs2&656610G zfpM+ukD*&8=#LvMZZZB41zC*{Wq$w{N^Koe1?HQD6IoM%GFD}U*jH^>@}isY2 z%%%*P9$xF+BVg1hPYEDjWCF$U)B%z72}kv z96KB!bv%yb)r*6SLUE<7#iCl%_z-9Z2X8ACRmHuUY&O#>F4hdJg+wO%tE@@)RFX$`-s>SO2|Gu(6 z12CmucXqm#@OQ82iJzmIo{o&VDQ1XHHYEO8zVAL~MiQDJ*+7GM#GxUT>NWH?Xq;K? z3T%K7v=7*s$}^i77Q}RE6>|Q>zgm)9l0_z?8TV;54({;mKxN_gYsTPlyeQIfvf)$` zvX?&8;u;QQQ)sa;07EXP^ey653G zJvW$>NHA5X4|4bs&!jT^ItLLmhfzpsl^-8mvK_v1ekot8M(mtIAJ4)7EbVJ(U-qew zh_`;Ypgc+$yQSmz-A~j(MwVb*DZ<2^@v1gOo+o5!=8l#AJZGKT z=*cTP-isoIBcbw0@+l$Xy}_f zpG4*imLODIXHhyC#2UlRLFJ7SqcdQ9VdZ?4n3Ijxazia1SuWo6i4wL!PT6D-VsCq& zKz(|RGMOd2@Y@wVg^agvo0);exqeX|k$|V{VqJ^M4b0=GJhO?u99lS5*&inu?0^DD zg^4$!pHY4J{>hFBczz$_F=QKTZq-05wOp!Ay91@W9pZnSDjC0p{$Pyv;%!I(^weCc zM4gv?9k9LffF_OihbTR&m6{3-)^O!UBy;?YZTn1pFQ05$hs3)&<3jlcodtGwr$u*o z+7HM)(Kt!PtwHW3{pUZ99*_}_M|oj#mSH2vB~Pl>_3VCl=0>a8F$4QU1y5+li?O?O zIol+f7f`ms6`M9lWolLZraBr!ND)71;%m`N>u3~K2)(WliMXj8SP&A*XeFQe?M)9dhj2+AFU5WzhQQ!j?vZW$4+V$-r=Q1$z^Kw6Djt*i!neR31Zm?3)xSfN_ zVjqwP^=K)#C4M8Yfx;!|dUs}rM6Z}vN9z!LaR7>TKK?N|w{UL`0moMsC`1woYP_9p z?Y~4@ZoC~6j20x@u571f_5Mx~v=z^&iO?%{_F3j)|FUCNj_YS%H6_LK=tX@L-0efR z)i`xMAC*Hx(^SRy0rfBi;$|g#47?j zqlgz=#*2u}9gff2N#)>TUG5Evg&w$Fo}S!U{ci#(=fhjL$_ zKcG2+%;SZH{9IbLWqSRDeFADClorcn<$@XT9w+7Ejrxkr6>3Irp9&lnwQ1=2(fnpi z0vJ!a!`s6dBQMelhaMoCc|cU{>S8v975JHq2jnJqI~Hn`YQBt=fQcc==qNeueO_O& zVp`_^S2fqO)pv_Rx+A^3mU&`_S4u038F%+-{6}EFW2sjee5Z9AFvtZxeMQ4 zwuct93j&pVlinY0%2X+R6k~=BD^l?nwwqa_?lyU>P&+w+QjGjN7EOF0$^@V@Z3EPqc`|X_{w~j9Elm# z9a@W;@!YF&XLVCGQpKe*l8^Ijk%Xl$1lMfejmcBY*_dz&0P?h!!(FpZe3J{l>@_bC z{~~--6Ky$DW6`v)kn(@Wev%{z!0ZGTk0)!vtFcr8ZTJ`dL9WObk#8+JvsjT?t?s?o z-cN4$HHYmq7XGrrWBe?6)Woq3bI+@GIsf9}Fw2;}T#w5j-FSP1Ds}U_J7BU!DlW$9 zJ6b{KRZd2*hem$=A*6D9dAZs2qA%1i*XMM>Z1Q0u-eW3Cw{_S>^#RV*pxOJ)()Ot6 zbCHf_%f+AX^!Kr!!cESAm2lqwR2g`Zn(m(~4ikO)hsGeD2dVSiwI_Pnx6}K4M*Xt5 z3m$xP@x4?!(Ye>V)h6{LEVSxpdT*# zlBErFovjqMj4Oz4)$gXu>1_ zkC(G8Y?$%d~^m56I@%*$jC{A#)46|l1%Li zotaM!40kD6pOK77kaj5W)>-9`-N<3M=ZNtKR^@(PDABq?IK>uIv>(oiG?*lL_I>bH z@Ljp~JzvxNP!ZOt3Q;a4HPA0QH?L71Gl*2CK4^g{)BC0wZ?B*Jpr-kg(*rY;hH?@^ zT9wsJL^C^}UAk6zooJ18`q2*|Nh)+hJ?v&t&#PZ$K4F$Uur%Q3XRRPw49@|TdDCP+ z(mq}c_wLtk=Cyxtw)eUR%SLRx3)FFU@DoiJS$XLQ!pYdZ>09CS9-t%faqQK-s-qlhVe%lB23i=F+maHnU zzWZwpaoUtWJrzUPnJAba_XyLUKA6T=fk){P({zu06n$TvbUth+qR%lowa+Nav9CAp z)_0vlMhKVBCyw%{1pCTnET?CrmPF3XZ#UfQG$#^K*T*xK6-zs4!PZ&?(Al0OXfXD0 zO^R(W`D{mx`EWFYYYWD7JpeBDp`T1v*KEY8m(4^Ye0QJ)zc(6k8tT6^6}ddT{E6rH z=8I}YWvC$jG77A3K5kAU2k8`zd}Y&5~oaoHq< zSA9fu0q`(n(m?9e4h=z!Q9WrIwm|^DaE4O>gmNRQuPPMsFag@4Jle z`ilK`S0Q0I%(T6kL|Fq^>YnZ1(b)C2gX|w(l`Mfku6*053=tBZt0?e5-BWWIkhs2z zx@Z7X3;yfEpdsPke^$}^dD3o-@9MTXq}n|#P>u-k5URdK65ifOKnzh^r4#5VVM1pi zgt+Gp80hd5IP7NmYqn_q$8O(9w%;a=Xv8LSV?yA1TzbjcSvpel7L>pRzZg{8gCeDx zyA`*30*f%LXhYB{pn-Yl;OF2sMeAnx$!3&#CS5f>L#ZLc8@}qTlHq&AlWx)$sJkf< zooFhPBwgV&mT)in&Am2tpL=k@3|w58@niq)9+)UT-^~QivUJ4Mm#SNVkOQkkGeBC* zE%OhJGI8^609&>4Xl2B(k7=iLP%fRmFcZ{5_(Ad=?z;mK1R(znu{zRypVQ`!+@= zva}5=K?YHxAc{x^$&z!BARt*JNJb<{lpGw8AUP_MlOP#~48oA3l5-9M!T`g}kcTwz zF7`gp**@nyXJ3B5zr(M&SiO4nT2)n>K-JFd3^aqa6Y1Kg}HVJVMrPBqInQjB6rDI*i-ZJSBnxHe)S9L^a*GJt@*i7ncn!Uosr`BWpD3)uMI;9e1x^A%`FA zCiTq^t4aI8#8!b7@7Jn>up$hJMeJtJzYv<#z3{T{C&Y?t@Y?&jG(O@pvw594_BK0# zYRz8i*mbtlrHRj)U@vDyzst2)Z=vnwT_8~a7w$+L*VcM6n_k2f&+1Z5v2C_{Cr~85 z*m+!3wC^*1&gZqL@%^S_?nVkIu@n$P<0l2^P8+hg;x+8iP2kgNd6BA(u4x27ciQLa z+aNyU5nH0)sD>9dmUl8vQf-*ZVN$ODB;f(1^W~cNquG1Ym2h7yGv12x;NCzJk*%^j z0Nv@yE2S67;%>pdnx~%a&aFSI%Qcq4T&!EyDcoC~2}>e@VuDXyVnWIYbOV@}^f}k=@3JrxBQBu3TSMO{vkd-CEE8nIL*T5gIEA!M5USEuTL$?+Y$Z&8$Rm%Lx;& zw6q3rmqf2N^Jg|(AX9yZ501M(YEWg9g_hLVV>DunevkkGavw9j2qUONU;)$QDk2wE zNAZNaKT&tVnBT;-|Jzr+cSeeM=o={2GS}mQbJY1op4Ov?K-!eUv$c3c!vH zC>aOZ>J^r)@B<@`6f0)qHmP0%CX zwRz^8#Xro?D|C#mNZWW}RaNiO^4`6F!SpYa3fElNiX2+?+uLS? z^j$#Oc5f-vLwTdZVztSyq#ifdW_fmEsX>OHj5)=Ajc-qBt@64Y89!Z{XQ$D3#!aji zxe-=*H(6~aiBDX^6Oa-Bt^iwBE_IC`C+@d64%a@Rgb9LopX=$U$J~VnYfhB61-E|B zX6k`Eflr9=@%Mmqz`3JWbzck400cRnP;`pkjj$C7jSoQ%dIC#i%b1}YOtlGsc9$8L z#q+3XFbL7?{V|m%2F8^7I!tp1q$sQ|O}){)DzS9N6;z)~b4mX^}Z&GxYm#sD;%U zySIRC!8OXN@j0H*M}6|t2<2HWEg@}GJL=9&+6f>%qUpUAi4SC(dzCqr3NZ0WnweS& zVlr(hfdX(zrM=7=C1KH}a)z$Te-ss#WL$4mN@L!70e zaYY-eapyO}kuyur-(dRXqf?&6utl2{F|*0G%>opHf{1+P+BDutozb)HNUuycv8&CZ z0iuCEE^gCo?%M@uzDt7>bZKyW4!6cFWa=yj!PAG~UrQ5woL3};yu1}X#YgkFigWK! z3k+{PC=ZkrT0KstGyCW?ezQ2JhtBtreKa_;5nP=9EvFH0bH7*TYb$^q$57%LO)_RD zD7x%5S|Yq%L|>j(utAsJN_MzEIm^R9*tEY(roUPgN26VRXg_9}oFm=_LwQqeGu_c1 zTSwFVD-DSzE`th{=IG3xS2p~bF9^+NR%)p0b6Gi=ZVVv^W+>xD-dASocJL4$OwWy4 z^y;RJYC4XRF5X`RWBLWrcZbK+aiXjHow+GHqbsu3oi!<(WnYMkIlPDGm*@l^EnQh{ z6(%qQz`AT7o53%pwuc?`QKfqJJ|ga^#D!cQi1REp7DxZ(XkKgo)4M>foNih9X%tU7rm<)alp# z!3MTRfvZre{D|s`6H4cP05_5CW%w4W0HU}|JZQA>jM*^+@)AW7Ln;aoH%UFytlegj zwj+BEWY1~?&!Q*)*a0>GJ3#H|mM6L54As4H%cDTip)0MEBnTjzV44h%e`6Ip%Dt&l>nF~lIzt-S=i5*GWgRBe6j1YG8%7fLg7 zVLnSWP?jMi!Yw}Db$p47xKLlDMs)ee<@$Mc&q8RHG4*auqYX314oSI{_iyg_6E|!m zg&@?HJwGf5g3hDTR`*WoEKZIq>jqxBoyQ5Trge_esE|8i0$Ao5iQ zYiFVZ>DuS8tqXU5$25C*e!UIa#)OA)4$}{r@TGR(BzC?(IaP~M6b2qreI? zQEaPYHbj0|YDKO~&8C|eWOcKiq#&b;Mn>R>Z?@0P^fd(efV&{q{rm5km;IrIx4EN{ zl*=^AzH#nVe%2=M`y~$cD^F^L?LPD~rKquzN^#h@%NS)}r4lj=hrD?gZoFbzu`(YIX)0?AKxGKY)EHm1fzLZBsDV>O^NI;~U#SrWAkBzGi?vuYx}xGG%LbDDW2w3C z=k97?N5^%G_x=q-z01thD&4J530nZS39(lplgB=okYyX6)IHhz+M_T|<2_5I=VnfD z5@tX9+e8#S%kR}_KE(Q+d$p)ZzNDeQapq+-gySz{_u2g5UA;LDLC#}!C)(h4IkwGm zi7qG6lRIZ>z$ZD;g&&(BSk6nIdH0I?V=%A zx;?69I}-25Vt$*;clL1PG+5Ug@aGExEk}42AB0Ed=}uiH67_?~5J??3oKFu?9`9tL zkPYmN1t!pNC4=f`c}iZ)wHRzq) zZF`e@wy72-eAYAp8T6e!8x1;NG==0Iyz^X7q%*zBx4%ft>zTg8{j?!1CqV%14PU{= zJ9nlH;vbX)P90{sr|`2c9?C3Ui_)Jlrjv=+H8vn#wxZdZJFX0Dpy#G!m;w$UI7iw& zNh}CqL)}T*BFd2V&-z66M@iB3#F|U@dPa7; zBGPKDB(voo5!2WkiL~BvQYWvS#XQldve|f$uN+7Z zxdJk4XoGCD>qB3YxUaHfr#nulzSe0t`#Op^%vOVnCf1+tkh*_TVR$APmnp`Uu_8Tp z08dDCOZGjo4&FN(N@Bc(N;D1%lMk3X4wd-_eCD}oo zY!Qk}({4WnuBH!0?WmtzGPYWd$QphafVy~#HQV`qpDsq`?+7h_t*PD>Ex-=-`Ee>g zBL#R*Hr@2`;qg|@Rm}hv%SID}Wn?^d6SU=oYP#6Qh2`>~g`?PEBfdsd#!Fk6UGl_7 zCC36B3=Z1R_a$62F@9uU-h}08g?EIsaS_o~Kg$*)$xis;GWK5bM8WJ$BOJCpF5)X3 zOnimx&^QF2T~C@Y$@6&LL%r2xbZtR=REAvrDqn&!`+r%%I8XPIs|m`qnkT zFczR%F?9pk@hDk0m1T;Y<~~yb=8NRQezL(OWh+^YF+kb+-4-%A$>!_Eghv)r1Dpj^ zZ@o(;a+DC5lmYAX3W*$ZaWA!mr*dvzPoaIdGg=`a)n&drqj%aOS7y6DsUJeo2!UoVK0QW`?Y z+*_B4&oCuqII5to9Ky$=8Ce<|y-My?N_WEY3@eEcu?Og zge({>+8m~?QdXSn<96S6-Z=!-0T8zCx&epqX)!|JP3FwFgUQCwKLhd~;s65JE7*9U zcV|7eAZX~^Y;RM;g4FYOMm4dw>9d^(vxWCmfc=xqgT^+l$i>J9AU9`tqDhD>WbjiX zKZa-7^OVt?oD_Y{{JWFGdYLA&F?DbGES7e-1E7(CEK%n?p1s{YWBKmo4D5=%HD=ae zz2cIw)kwG&eCnarnr-1K9GQ0XBz@t4$7yttZ4+=E_sJt;F?0Cxn0Y<6zBo0Skl zkXgRT<@Y`9cl0i9F29WM$Kafo-M5i^|5je4w0HfrKiXSZaHZP1`y_M}$d@(=*-;iYMqo82>$Q~IF%`By{HMr)?~hna}hn_cDC*0{#}FA-)CmI4|~PJa^mV5*Gc zkdj2Tp$5?1cf5p~`t#(#iAS3(-3@o}dG=^Iaup(lBUi>~kyvRG(xcz^axXx*zy;Gr zx1@3porUuDk3l zZfmU31QXwV_m<9|<=@?4!Fr7C$A#)upfL09T6Fim!;d`dN2!S04m52{E=F6a56hY? z2T~F7bp1wtTd8}Nwgz^}C@`ALB>4^Q^R4j9oe^CqwuctQsqc?_I3{9}$%ur-pDYhl zgYbJwl^8}S5ZC8-{wQhh#X|p(;bnj{_J{jNSH-Qr-!K?^eGv%WCpAtQYmMpAYuW+! z-up3wYay0VT>a;H3#F0pfF9zE(AczR? zlMHw3Jr6#5aiU819!R6iSr)dpVtti0KC%jP)FwBmKx*swUc`E!6t%= z`omb)GASUKNPKx=mAmGJ_di>d=yd>6+c&7%Sa)vaYu;)k_pZSI(El~Dj8x}jrU>%2 zmwk)3P5h@lP%kd}c{s>IzwXr+>|Si<#UC_kKrtXLj?-{>&iPbAZ@>&$Jl%w7(l9VF z;{q+@v@RaT-KZ(Mwg*7a2|HZMq|GzPgRq54LA<1R3=?v1Lb)5bte45`AI@CqkjDh9 z-qFH~YX%tZpdX5ol4=uMd%m5(IkBSwj;S`Q&t(Lw0u=Ttr*+<{n@j}ifE&Pk)K4n2 z-g`>)=BfH_yZ7xAXONxdWTZAJP;BOvJ%Bt4wdalsYr^S0F?&jB590X#=$5}rUgo1l zdFX~4$QG##fYc8vO!K(qn}c^!p1&tt#;!7{AqbDL7^8i`volQOgIi#;5n*mqxg~xI zY?D*GaIF`F^(Y3aP?YLE>Gsv6D<+^v3X*)r1m%5TAq?@z&h^@@8X9wR4p~v#fnLH4 z*2r0&488DrhHZavI$g>DOT9wDmvnl&-aej8!+)c^sC=QQ9t(a_16Ed}HD= zA0uLxIU;eg&&tMp$reWL1&j8`INE&?4pK_M`>o*+2#?N3kJ#vzLyBAxnb$ zot=&UwK;i<^FFdAF0e3&CQgp3OKOIRB5Ip)ghHo9WoBdkv3WUxP5a#hk~V1#+z72h z0kiT~1^UQ=P*ar!GlUJ2k9z-ojJszg(bZdWCd}eB#9@zu_J-e+AiYX~bUs7A;91=-~mK zS}8^DJTKjSPd7(HLxX#j?I(!#7Ze;hbPjw>5BMmm&e~SQJD#j9xam%eZTq6bkiD$8i$WE!v}btEMM0R8 z2jeV}EqJ|DAF*#=J$myvB6kZk9 zE_)@g8D8O?H7N9Fs5q4^GG!>9PCa@ICS?D{#Cf$Z8Tijb3IY7J3@JpFT{biM>iV$x ziD3dqvc>Y(xTh<@MGllcEXEx1SDuXRayyJzrM<2P=sKRqD}I;o(B5IUI2n5#Q14(k zgb_3wsc(eL^zZi9ARMX^*Hk_NnS5Wn7t$|lRs!&D(p>&eCEVL?PfKZ_5ZGb)VIj87gPuz()z7VXb&(OjWtDCLq#H33T0@5|+351u z2qzBtV7dx@@(U()`YE-W57w!E0GVX+6|iqyPsB|F)?t1pW966V52VP}%V(IIb>F1$ z;(Gi|zq#x@{pfl_oXC?a+wDg|P=Tmt^8!(i(U;WR@FKZe*&NI3dzU4A5}NlO7d@mB z^xZlHU?evZMXO(ZH4B(YQ$$ZQ?&@Ve;SCE6%8jtrSmBl zJyavPr-7n<_1caE8D6uTV5_<+o5g7aww=_N*Sih0_o*&K!P1V|3eW_XgxGz)KT&|OyYH0B0J);!@9Dw_2Ha%SS@V~l}{w!Btd zjNcQvWAK+9a>7iIDI>ftCpg~|I8&*J#DJibm<)$~LVC1^$#WQ~#XE)CcNV?z@4bq+ z0S1vuA-!*}BLvgs@B{YiY(8`=Ru{3FE0V2$&(DYhU=^vd9ucssZeR63!aoBCCF^^p z2IVp$LXiOe!ART$sNc!-Wa#s(*Blko50~}~V3W^lJA=atpjyyrje9^jhU|okQ4HTe zj^T8L1vyiJd!X6Y#Wz{j6(98HYBFWN;sno0OyhiMn#aP$zFx-k( zm~yPH8#V1v!mbmW*H5l6TsWe9O*x7;#olHnPdAq#xuBI7*P&$uU$1@!>Dj&r*B%Wm zyqWjW)#geVxdzC3GTuJp6I zs3_aw>f1kqG;O4&`=R>N#rBW`Fy>0W(*+4w#%Y}u5{gPkrTOg9ET){FI3g))r;brx z9^sQ`Ws=S|l7b3Nw_h38zHW+M+3U!Oxgn{2*r>sN??Cpv|q?L{66TD$!8vU-jh*>A}!%@eRyqcqIlk!3HvV*w$! zJ1es51D!nCY4T|Aq)`(|64ldgC=K7UB^)J^q?n`@IzM;obesAS$ay}*3KmN-SetPw zX_A<5xtUkANEVHjl$z08Ymm}DC7ES)F?OgL^}xab5z|V)y>b-`XOn5D`C@wPV)r5j zRyBHJmEkKIGdnk_7H0;}aPPXaflVcN!>57lzHn@JSFPh>#`5XN!osp`!ys9aB`bg67#T?kC}ZiXpB*6 z{q2(UKDP1k0BM%B_^wwqrjsC19!l^bcZ0dp+x(BMTb`|_vPy~Y$^NYn5$+jW&ZF?W5jUzr^nrypLsyJzpW^Krj!a8akE z{v-%Q>uc3P5WH6Z;_bu5!+emu!2Otr$Ou1pn`=9ZP@ednYCWt;^~2q*aD-eI@8EVk z5_2LHd}Aoxbvfmhf2hl#NHIt;^43=PE|11XtB5BTDR_}9FUrd0l6Dfrg`wwTTGy~C z7;vF4FqE`)8*5U^$AT*IJCsuxRQ`Q-E){)h;dE;nyZcr5I3L`NVj~0kc^%yZr$MUv z9jwmcVf*ZOa*))7wk(JZ`1sVX6|wJj!nTPYu@eqTpk=%r3Gsd=yF3lDXg+9ZWDQU# zuj1Xl^C(QJ8PU5@j*>Kvw>cxbz2AI(ZlXGA3|!;& zWc1cPWlM2uN7<|B8XbRHW2TTA{AsFt<>$2jNK|H>h{W+vE%{K!fN)1JT!dAog9O>I zzTAroP4QNpHr0Jb?*ADyFP?_|I4V}foQxz`crS7^!qg$nwo9MBAl1rTg-CM$Tdb3D zTK@8p5hiX?tvBHyTEEkJb)_2JMr0TZne|i!_|BKgcFDXrxm*VBPP{#nesd+(QOO`^ zfkAKgM{P3?N0QXjI*?K`Tmr={lS}f<`$+5}DO${7Csy0(8 zcD5z;8hmto52vrP{-&4rcn|r8tcb>LTrW2j+S?+#w6?r;oyO#7*(6VVrQfuL<9^IJ zIe0%$s@|+a>T#%>J8~x7p=VMCEw%lnwL;|i=KF7x^4PKrl-5FxHcPuZEIQgqA> zyDrI^7o>);$76u6^V|6W*=4ki!QC4dNbwmN!)h9qe?tx=PxvqeYLBx zu=!)}remVCWsjqHt%+P)NE5x$wk?X<{sX7f0zO#muGp|`d-=(s-k<=s_HjnpM^oJ@ zi) zpXf8+pg34>-IBLvc*E@;=QSQ#U7RB0Z%b*QfDUQrJr##9_d>pxg`VHLnwUh*x|q$< zxkXRHyBYM;ll>Y=miI|A>UL##X9<PvgH*@?{t`%`Q4$edf;KDWH6 zHpon_RuJAezAXrwY=w;iz=ZI(J#FHr#!%(7m=ytuCK|u9BEVb-)HQvow2RWV$~l7= z`Hb{@?A7WMy|a|g24q~7u+5A*q{m1edY&tH_J+SmZF08?Pmbv_pua`B6eZlnQYvPn z7i-tB%LCOnWS937ku+gRcO6Nou>`T5dQ%}Tl#n?)-d`>IH-TWFd2l``0PTwYJbruv znYWE%uzz7vv({V`3GXsr8yqb4VGrrs^8R_EedZ^+7!g%1mGf?~)^0w*H`dAr(LWi! z)5K$$k{-!q856O+lE69ZoN4@Qzc(dvj`tDy-TsO3;= zMi?{&l`vlUb8bwa81p(L`J2G>sbOu}`#Vr(C|*I#<;wc!66>!-9!mT)h)SXUouRp` zp9f5L?}kV1tqbt}A`M5ta1`tf%foc699(E#rD~08)+lZoS_}iSq4{SASFz|GI+bxZ zHo<%zvXK&R6@hfBsp*C)k6>DpbHk|3VUj5Ea4Nc~@wL`3RC=jUE4Tl7G|K{t z)d&mZB5*9xD!T!NoE@c7m+;cFf9=z{xA(=!{MzRvGk%Bpd$pOB#OpX$TOn$~*Jm_X zuEH>))SUKAe>w@5uaX7wK!9iZjL2PaTOrpqv_&vjJX(|4Lilc~dvOH@YTqM`M(51z zGJbjWVchc-speF!`k(Er4j~k^&!M}!M0qZQzApN;$c|--#2BfamnS`o3(x2BI2>3! z;hDaIMn_b)BaJ^JH4dis$c%CP|8aIe#1B}v2Z;t_8D}^T31QzBtvH|9o;w->vn<3%gCfxh}F4XF3>C*GgPb{}m>YL1z|8lWFDjKGzhtwO$Q^8!nlSA2!&VGZDg~rp7 z!7uvB6QxoD=?TqTjW1LGj*e(N)3Sc}mh{ z#i~d3{?&CpIxN2k70dstfVsN52d+Tz`^(o9)ok%9Jb2#Ee+LQBghBby&X z$YqKs>{6N|n<3jQzY_Wk2I(vameFzL=DnYro5`8W2@ysv(Y+?hoAp8pU?@C6i`RTo zbCC%_`-?LF@>%|p%Ad~)FtR`N|F`acExcR&0IiT_k@>LD9`|G#Vb&)4*Sg^L57+5h+X_@_dxZ- zcfq6_-hpN;)(adE-S|AtuJaj*L=p-IrTm?E{t{TrFyqkTVrIr-4X2xjTkbf!?!=49 zJc+65+){38(Ko_h^y>NhQ~B~KtnVuQb=TiWM;eS{EaCdVezuOUKUqTT%O>KBe0x67 zq0eGd6>2qcw9Pz56=2xWBjAr>{H-x&Nn*M6Jy7*2gPW?g+xOdiOWE|8>HugcRCTXS{E(S`pGCZ;aX-7^0Y%!-FwprsCzCp9#I}{_q`7ZT>4`> zcm~+N)hTg7NXtN@&*-_XvC) zc?p>h5^}#Knk)XpxXXtTP?s!r%gyjykDX=Lt+V5t+v_b>qRdm|g8fSnBm}ZN5xEEJ zS9Y=NC*9p_8OTxk8&mSvB~f|WHp38OA5Ln|gK_$yajAzI{9mF!L99I$aG}|cUK?~J zoo@lukUJ&>zs|;A^#5O_H$#ym<>8>8hPtblKg%tMdfAN&y&EOwxvt^Oln*oRUo`1e zs(%~+*=a!iHH^RM4@7JY>Y&E5S{1_{NabUp{(vGC7V<0I8!2yvh~~T!`^{nXfI#ja zhqV&^W(xdzi5zII{kG`I?^zm9JPGj@TTVX5Zw`iy7;sdsXn*|85AcA2c1ruP`t#o$ z%rjb``PCw}-&0&#jmLlii_EC_&A}8%1I@9XX_Nk|;QwzKW<|a=)x9HO@wfAnnrYvKRX f>hu3E3x9z}9&RGdL8Iw)1^82vSCfOtmu+T{GfKW<#rpEo=ifBLHaAJUIDgrDo*-2d#f+mvq&iFY*tQ9Tt?=m8l8eUN97{Of zq<0;{0sA>GvmEMvbjeaDP_Q!^dxLTuQvUbOd7zB{`k{=6AmGv68n@tqC%N^3rPd?9QI8Y3l^Ve9M1dBD%2sW4Og*o(U@$JZ$rLwu7@qqKLfz2QD} z6r;p)Nrsy;A}J?tdg?T~RL}FEuT00;``0yeGdbE9369Tp9$kKZ;=EDPuN%*(W1M*} zpT1MNjCh!$6324{Blyis7=GWQvGeH2ujI#Yue|o^^zc1Z`$57}6Gob_dS-TSYq`8m zeRx<0=A69DDF=yf`?jZ_{qbD!Tai46z@#a1v0c4&k zb1a%zg-?+bdM=2S2gGkFaoQ$JcG*uM#L9y(a) z(E3C8R%lUl>t+soPp%MCouMFBh!>>t%pR14%Ris_wT(I8xV}C$eUCV!(_8;hB>@=( z=hjltpJm7LJUbxs2aDAEM~WwIAGW+YK5B+PM82na?Gd}vxdYE1X=}+}7RI3U_Mo#>W`YPyGBin<7^q|+KwgOj!5l9meK8*+!o@bBwgV6J)Z;l7QX34y& zyTcy0^@;WVan(-}bcUwZ=Wth**GI{&oCoAf=MN-)yXwFq_w3A#^5zrXLgUBSqrh4g z89quRxs$E1kgulK`F@`Ic{nS>+x1QL9i?yWnD+t;M>Yv!%dI<8=6CwIsI3iIYWG@F zs`q(+ww==8ALr>>Vzq=e;@gToMLuLmKM~5j(F+O@HB{=6E%q z=-}=n7%TOkt#!d_Svrf(8YvoO7}?(ogMWa)+d98zNM*=yxH`>YR90S|cUZl+zx9-q ztC%a8d9}he9pfZ#&$6$LzU7ehqynwIio?%(?i$XtZ!mIKr?UaAq=e_so z>iKrCPxadMM_z1d<6*SVXOa)_or#n8K4ABm{os>3Cy!2FJ)QhV1Mb{_SJq-Upag2&M|p38sL4Sjv}BL9nPZ(iTKcp&~9tCQ#zA@OVG!n5DGb9hIh_;rnhd0s31Vp}^& zykU0%dFQy9Ucj9T{j35nO7F7qMz&;ebRYL;wR$9THR*%%Ly_8J#oERnlDhdGs?+u_ ze__dsl#?^<*yreIBX!!KrC08(5}#y?K1RQY+x;1Cn|Cla;DL~jaMx$c>H{`k&If78 ze9oULTNMdXwTl?~nZN43rCra*(0zaE_)*G1cF=uhl#o&Onkd^Ycl8NY2-US3E|=k%fbR; zwju$qL9YkCi#oabo>WbdI?waQ;D`E;*6-jSxbH~uR8E6vF_AaIIk)qrvOTgDOk_;e zE+Gr7?ztB#E5@ekWavp}51QcakniF@78Iluz(4AsUJT3?CKN22siP@(ma}UsYctfq znmV>w*n447w4OXr&P8TANugC1U1;8zUdy&j{=%f%bO3s$u2 z756t#ivq>lU6suhdT*ZxgP153E7YWhc8;WSrA8@wK`0O9%5@<+ZnJLEJ);Ey*A{D( zZjI@y-?9GKa5V%)5npaS{^~*3)Rlv;wqJR^l7Gb}@FBK>%~;GJ|AuRo5z45lq_X5C z^a)8}f^{OhcKP-zNb@f_zvT90dWv!iGV<(+nd2NMtTl=PhX)>zHzOCfi7z&yH!1TI z-Siw>FMB_3&_2r^jli=gf0BQ$lzq7tIzgQ@9nm8!b(wMiNsJ z-KBAs*s9V>{c@{^W?q})R=8AG+xxbTVYo<_8F%^SyxzH5>7GQ-G|##(lxP7rf?E;< z`~Ws0SOF{J8T0z&#dqz@Pr2(?KKOo`9FjFb?>34TkcYW*^vA)?;-g7OR@Ctfnx2<_n=T~XZY^7vV zK+)p$A(tUElmxv}WmRPaJ4U@n-6T~Ft1VrkoTk9ppF}mb^2dqB#brV>;hA(z#+vU+ z8)cNZ8TxATOi!|>qhLrHe7&J6sEYn6)Fe!aUQSP?t%XU332mp)kM35|HD`u>OI7Mr zWOp-mocE%3@P3*#Wz}N)KRudyH2rAs>hkBU&ugELTywd0_L}0gi7zEzl1dF^)Z$EJtum1J8rA=U;q`I~ZU8 z-7KD(U6h@7L*Rv|uAUC~hE~j3E_wMm8>tI{CQDtzFrPi2_Gwbf*f+`PG>-YB;ztLL zk~t2X28&k4@u@-ws|{`)zBQc_>15)2eU0)fbM^7e&@1vyA475d8r3hsU(1j3i1Oa< zND1;k9d;Vg`}p2i zkgh?lMQ!sP8=X>}OQ-#sXQ#oI=PlDpCtdYi$lpnww>n1N8gv~WFuo4?iWFhE=;hJJZ28EQw%hla2s5h9`G8v zZRMBMpX=XewP4jeN-13_jSS#kOhnU7GA+ZizcxDHf}1v94mIM>je7@V<6A}zb<^~f z!hro18e*`LSgVg}>n;9Pylsty8Vm=Lcqp<1V<#$;JQ%gwrP&PIsMWyjI@Ujc|Xq0CEC@+})Na}HJ{n*OQzo=>FT~BznabxJmtG97! zVv2Z8MN4a9@5FL0@lJn(jCe@R9|S6w(!_ghS5+N!b_UcY7_+&YN9osq7N4sT~H zuTRua|DZ;b-6v(k?r*(cgN@;-<#iQ8VP{r8&E8$gu8TVqXRi^jwcU%E5?_W8<=?sv z;4+ZP0fz(ZwjFnb1Y~n$K`^e}f=vP~NlBJ{ZW4h7rAgP=JVrDa0gs_$@L#bLi=E5M32V+fSo?j)FU2Vl zS?VYEu}B|fIjo^<(u(Se3OjT+142K!Wo!`;xg-f?>HmtL&C6`T{Ej_p7-RYLgto!S z^5G%NH#m#`#{3sG@5DuQn8dMDnJ1L=!VKFD4>0mRcI9YoX_g<-H{-cnzRUAe=_$u0 zIXNC6yGV4lxaDGK$RYt8|G{!--)WWuz|lV7AIm;omP3CYv#{vz%=ZWWaeSB?xW;nLRQuK~ z;M3I6*V!5B_t48fsA6myIC0ebrllVX3%3aKci%0Oiz_TF`$r)6Ec`7DZ>u?adCEL+ z@^WyN3G(!2-iJjaNDVmjboPI6F38gZ>ZcZ@Dg0*(HQ<2 z@^wC^By(BjvM`AC+_`fazD_P`CO35d>JI#+Dg4mi-&;*q76yaKz~p7Td|hSbR8>`F zFJF; zu=uw<{`u^`_AvH^02A?mIW&;mUr+ewbAPqhkYz6ZKN-VsOZ)RIU}_*%4cY%OTM+Bs z_d{=4Sgx|%x^eAZ(7vUKBgvdPlSif+A05>()_HQ2<9?T`1}@$2b9z}aNAyvSxpY7C z4of*gQNAms#C~+mFo?;KQ-0|FtFXl>gHvn)Q!N zo`l}G|8w%s!TkLhE{StNc3EPR4>{#p9W3nevCH2>(Fk8Y2x-&sX|D%Qdp31KW58@=3 zo~hgX>BsK;*x$(53(=sDzg+#1f52hB8|~mtBvi#W>O`^0>$b8jXZsP1Fgv~mN%*Q8 z`9q9dm-^i<`fLsRSv3VL<7aE~0~#^gu~mU_wmqwrG*Mgmtw4_zie2fIl*@6-Tr&0c zOATD!bGJ`rXyn3WD}GJ)6q>b%pTkw*!+7k&RjaSfD0z;4t6%qb|7738hS=M7C5|_O zKZYcYxWqv6AmESWXTy*wf_)3S*3N6y%|krNvSnCR3B>m7(ANBn@fBD`iQ=+rGN`Or zE}?OkV!cCn8pm7qBN=OIr0Tpq+5=rEXzhpJHNF4sx&3sGDJ~X;ZjkGReG1;*9ExA< ztimVkXuWdH3^|Y8p0ob4zHrs%lFPk9g$MKo=I>2xg1HEC>|Tf zd=9WW^b5h|wnO4Jp@5uiLCvaT^ z;_@}Iv8>ZfpZ{3@G}B-mD=L%#8(ijLn)+Kd&4}tDd=RR7z895R9%ibnf33T?3a=R~ z1h>fBc>^DqaL5+#&&|3nLI>O{B4 zelP5%(e_*%*eLhWD@3_b%+pWsBG7`Q`Y~+;G2Xb`i?!!^M97*KOEfu zaO$Ca5C4k;=l^oS?+*65jX)B}!jhXta?QVb@&eMXXH9CeEpph_cJLa=V1|PNKr=MwVrEIK`JG63m>k}RF)f6P52l~RV|q~Di)6!Najl(Ox3=>Wvr!x3uB9o z5~|kBOCx!8%0)DWB8lRMaGDGgQn;)dbs=iL4 zYs5#@1ni#A59gt8UX8nnHAD_$kSQp?U8c;*CB&fQwqzGoll1-JSYVr|q2KTj{#zgJ zRoVbKO<^CO;ivs&uoUS*o3PoAZN{BiXpY$RP|C>XGbXirxnwW9vl>nhiQ>njS{1=I zC*R(_a?7sjMcy6?jg`TFeiv{Rm1wTSqlSBP_b)H?AE!0}NSf`$T}OR>Z*taC)oZDI zesTUyCk9`QUBd;)v|K-6xZ15gNaSTbK0^R9`9uzdEyO5CbB1 z8f!85hp2UmlyT2iHP4w3)4=X}X{^QfTJ;`loqhPKYJmA(X6V`v$4Pi?0-|y}v~O)( zYilaILrw;}wD6GlW&La(QU<$CspDN|z`s48XLM&Xz5KvOS=t;aMDY90?GCl}j@4}* zdu{>UMpn-GNl%i>+R9TtjkWXI*l18|20xXGP^lu%7;6XsTU7rggKbXm>IL4&;{vH; zA@Mrg9iDyv7t{Qk{U>qly_pDqj^oAOEhV(4I;$pMFu}&$mq6kb_w?{JpMdLx0rmU= z91I^2HIMmmcmU;Ik0mCmyVVPUy?Wj7yHKpAUo{g<)VOBc6Z5yL^J8^EdM}--K(*F4 z!ZkWIm!}Hd*AFoCD4>r<^PwF%dsK(Y-i37@BtrML>MUnPS{u2qkkAwvmN7DDB9UYr zSkO^09!S?r+0>EnYFwY}fVNAk|193igrAd0vwd%x&*sAnu3Sl(*5UhaOg@9wESJ>h zy(rwH;u+sHGl7`}RT&=mZz#MZK#{b@v&p%6|v6&$`t%6||w*kLXvpm3O zq(XXr^8vYNZB-2EY!(+Vp(ljkWjdNU6?Jm8p+9+0v8*_7)+G&k=B0G4_wvFT9d|f1 zvnDvug`jIbHzHd;wSyVpPUGe+wjQc+$QP*mu|y{z5iMfK@Wek~meE?8VhSfkBsr|S z-Z6V8uPkAfH-N_jzFV!c*t#k@H;ERK5zKY4MfZNpfO+h4<%kCCdAV?Ku)if;a)blc8+EzYquM#%0;Q(QdsyUBk^G~YWPbrLJI8szVc z{3{o@b93sY?2%cDb#Q@*@0gqN?SV>9|BR5XAo4D|n{xB#$M*dNd^}v6ona}td^}g| zoPHjft{B39lrQF#;a3RqW*Tq6XiJ_0;6$132*)qn8TRIu=ns$fLL2{|hsp}u0}kj@ z45Tf0H0Ib8GI_b}{j-?<O5^u=OYeN76lMNAoqq30SqAP3Xu#ZcQTx6HNH8 zcYaI`9ED8p_!-5V(u#W?LPmJs#&1lgd95Gl19q6SUC`<*?`~h(M)?cR%F!aDu;u90 zUCofK9kny`DKB{h}y zMBWYmoV1)?{JV_?<1t)cgFDf$%4u6P&p+&;+pWDg5Y0(rwZCk3aV#mW%0m4u-2}BUqX?3bB8=E7Lf=Yv_ecMw&PnP zTl$Rt4ohBbKXp>WOO^2T_U{e!7{htLkd$m3oOHyO8va@CbwqPh(s*SG4JOO6nr5IYxPcTRSo|p`vbUCEK`7t3WQo z%#XzxFDshoRLO7MhC%)fRB{kBaD>t>+K^qp^E>8_)G~`Edm%~g06Y4sd zaRO`;rmbPi+Lp()>7K;-?zWa}Bh6~uuJ-8tTH>BLn~PxR=gAqVbjuKA=Pu}wCy&+r zy`Z@oo-6rppt_e#_E^Wt_lL0F8Ilpdn)ZYsz7;xkUav(UHx1;L#2LK{*!}d=I8bZ| z33XNjJyAuYD;s>FZl(4HFnL5M+-C1q>i~IfY-YnWRucct#PRjv$s7cq4Q#j5XK@)eq)I9@ zM3tx#!)RD!=;%l;nO=*Tp;$KqSukoLkzNNvWyj4c&ZYJ+zUUd)<+0x);{ugeR&ijr3z${2eemj4JWQzTX4n&au7W6-a1GJws zIX(H*@0iwo<1yzS6P`)Siyjlh-EiJ>hd9dgt&06Z-zl2gt8!c1pX|u$i*An*hx-{j zzKYw@>LbGh`db;Mm6<2BtTD^Ese!0@t>Y@KZBgqVb{!(ti1fi{SlVxL)O0vD^Zxsar) zN24ijoaNyrmbhsk+sNx)Ja$2*@mIPqRPPt=hM{LXIyVHy{yfy zB}mBd(}xGlw|Hn}Z7h!lbnQ&RAp!F=eFKc)`fxU9iIK950t`|@wjA`y@7@knU$!I^ z9JdVPE2M7@*|%R?Kfnh1uyZ3SlGeCRqGS4?A*}*cHp(u!8=SZCx&9}DN!v-hqxIvN zB?9i~D$DfHX*o=@3m@00=)^(H5Vk>h86_uD{6{AA`zV*{X9KYG2!hts76=NUJ1*8Mx~Mh%7>&>L|xwkmmjCt5(F2A0phSb$NXK#4%6%hQ8L^1jn+rqOz_x zW}1Tm^0@w*u6k5#t{JLj_pNksaGIHDhJWw;uun@-9;VQJ2)Zy{qa3KSR9};Sv}^D? z%btb6Jq{(ABRH_#c~SA=&`dLE!)|8LLPu@+G}QbS>X58O?X{l(zT$K0STN0zDSlSB z>N2LR=5H6aR35!@IL*VB?SwfRP2L1D1&%Z*M*C+se858dS>r@(h%T#Rtl_PntJ0Mt zxjk9y3-sHl3hE_+CbL|wbg`D+8o&JUCVi9|AtBKzWX7UlTZFAP`Ob+;C@T*katN*82V^Tv!XY z51$Rd%#?&MM;KC%70WnPH693FUieDZON!l`qzgrBv#HktccXoVC7Lg3ZEpCXzyox( zv-#fPj@pEQBf9`7S1C2dR|BA-5KO=t5!U8=RaKd+50@Gmbl=KH%d?Rh-f#tZ^=<=t z=>M+Szr+*;^wk-Y+nws*5V5}y`Hskbh}l4Sp`W?sgKg)sLhIYMtOr2b{becZmcc(n z9M*n_;XK^M6CksFMZJ~F?=at{UpgsPB-i5gzjUz)xOCVK>;PH9^YYVv+9DZE8c}9o zI}E>v@%i4o#B$b+y(0pGklYcB^ab6W({<#4`x?7T$4;;*dJGR?z~c9kDMdiWYGtZ8D}1aUtaX2i;L~h2G%zGmwmfG|j`nJ#m+=2GLVZ!r z&AIPSz7U6o^#R`cR4YG1KZm?9oT$avkq;Ia6cdCm{SZB|IkWO}wkF3X)r5&R06LMK zZ(L!w&bdq)#_vklUdfJ)^IyNAEuxyAf9=&vUoesU3(jMsBole%@ZUoHnOpBWgLsEN z{k24okD5>Q_ezw&ojV7>UG8qDHo>-j+9bDL_Y#c$%VfvNi*^iZ+A6>pIV+*-r?nhQ ze~7j-*8;KQ#kDqIZYwt#w7#s?r~jb6)j%>Fja|?W7%c>AY^*r0_TjX^Xx{bU&EDYx z1;wDfO=WybR+Nyb-~}?mB?X>mu_x62E#0q65D_aVfDENsYHV(-*F0$9%+r{EePIR@ zMT!O>Qj)R(%Xm?oCNC6U>2{U~-P+XM( zet`2bPLfDGN)$zLfw%NBWFU6=$bj$+T4D4?mD9Q)UkfY4My7WY3Jg@r))f>7W^9sg zqH*x`gOKvbY&Z}_|w6!u5>~&C| zZZw3mhITY&3EM`ot86_VZxnzKh@&;1ch*~`UrO76Nc#?e;)V=r$cbti`wx z>3RQ=&R|++_Zm%L5Fd%`Bu zYMPj3jM+rn!g2EobVuywJ;A^&OuRW8PC{HGcngSJv149lV&fTII4-cQP=0#2(H!g7 zr%}gaF3>k({BUZj?(=IfiLk0Qeby(Lesi}lq)}87*dGmOS4yhz5VS7~Xh2!SANk}H zAsu#W^FeF95uG&Q&KP#i_jF&P3$;tzlFDE>TZhI2!L99RtddJwA&-5SqEsw7#ERqt z-b}j_&JOUULF=6`-~2Ino>9Q4RjTJE*JggETj4MaIpJ7i@~C2Q+)pRi_d4q;n@mL5 zn4xCC&aziuVpHn?`$7WP9(N)WUDcI}SsqL?BWG^c)?i&z-fr7F2PF9mj06ytKZFp; zmaK>K-IrF7;Sv$qQ3J|1Vk)VVjQ2a?DY*>6d}OGNKk>?9GKJHE1Y2LuCASb$ls(h^ z_C(@oId8gt#5b@r)A7&_3Q?!lC?j;iGS|oLa6bx0gAorSi>9`LLD*lIw+&0$A-Row@S!SzG^`1o>3EWx=3A<7<)abtd0pSPcH?VV+hhejCt*BJ zW#O6cAJ{=wM-rhf!F$9b*=CaLMAU(tC;z!P_~@$qu`P5^XXyT=%-^jZs_utyb#Vr^ zNkAfcwb{niQ3X2vxm=~NNl*M4&`y5GHvcEGpyCr~N=7QC6ijvmIL^t=^e?ofK4xDx#_x9E$z;U|I2lX0qyCQuY#zUL9Fbi#x?*s}# zRPpCpR2x49g&Q#x&ilqFZ!mjCJ{zQ?xN=ybOJn&N?=EWc9=n8nV9y9Z%@yE}_Zjo# zC+uH#FRpPLa0v)UY^>%7ESNN-bNQV2u-`t+a*=w)J8GHT0xw3sAfi7ZWgT zC~R32%M7h-p?~n|ZCj&G49K#LHxiN;g9a*z@+l=~2oZ_cKe5-0}+SdiwOuq+a zRceiW_->S}7NP_8J}|0Wc>BbVomKesjpfRXJ&HQ{g?;#Tg;C#T?p4}nTw{?bQi6ha zjZ>NFpOM!l6ljJC8LMK9iMsiz#{4%u3qn_C-v?2)llJH>j6ZqY{}jo)KQLWmKk@Qk za*+2m2Ti76`L@*91=Lq_Wc$9{PZm_DEec$bJOt6{4Wph3xaatEy`Kz;3K z0*{@qv-oggLmLdJv9|cM@=u=58=2zQjS6z zlTl$VRA#G2h>?aY)xQA=@j*Ws*{vtj%AYI)w4cgc*fI}nA`G!yxW%k%0(HLr0(9ZL zUw;Tk0u?+8f zE|(Mp@C{S~vX9o{dCL=`xCg*1F_=EHn3?g}!=Pe$yui`M_h(9gc}16Iwa;+34Zch` z%6t9`y(NziZJ*=*G_1GtlDLWumkn<8Tto{hv-kA?c5VUHZ|Q>#Y~+Y*$DWnir2l%m zG`0PhzWGYst3hTGu{ayFl%?LjZ>#(b9Pb!+34ir&h)@Nd;W(1(hgQ_qH-X%iMcy*1IVk##5%xI);Q%3gXSJ(P$=9s6v@A>u zFiEaKBQh%M$env>Z(2st8##}KbBDgrZ?Mu`6Re?ik|-;I|Fr6VqUBDhb3v0GVxLLB zbL7W!K-ssL30PsXUAakC!W3eS`?|4$<6+3`@Y7&{vvC>}C*4H^I7XAu@pj?F5f|td zj~S(IJ62`&m%KbRaeoXXV6uE|`Pm&^@UcvE<78@$X^HH38qa5AR8vWPuU6|TRT{44 z5Uy{$^vP4m=UZYK3X~Dl7D2t(Db0MoQv&9E@Du>vMmh$3`zU?J%C;W$51JlR`;{X& zlqF8a8i%2h%7x*BZUph$;olw{F(X@%7j<|nFPYo?6iNv*c^GnC8shwOIdI|Z5&qB{ z4Nty)#Oe>Rk!wqZAfkFE^o)UWd-@&L40QSQnwjC|p3zw6mG#>xVUgjA zQe{|doBx3UQXT1zeUh9fv48_U`_{NDj!FuKN{l`jlfEEe9AC;TiI?{oaTHrsL_=)} z5>Zq1#3u~)aX&wzS_3rZJc#e|WvaMH1H0zA^&e*y4{zbq%6r2n32R@X1Y>?`IcqhG zxJw26Z9cT4xPMbNWOJzFi`_qn=}ytt<|=q-zUB;2n){yK#N=_0;WUf+zqKuOcO- zU8e)A^mEQpl)&h1$-&0mt`J9>83HrNi)!ICy$#AMq+wg`4X#hyqZ>{!Ikpi;j1Eui zhN`SpFKg{F+4`$Fa_&h?0LVnx+&1^Ia)vk%bvk*d!rr4SAWjYSaEN5Oy9=%J;8Dg0 zyKWX9w#UO5+9_eZ+uLLKyiSjHluhuDW_hp|Vx|+erUa>^QMI-%$ExnV>)2)7WA)4H zCgtaHy^9LlHFQBPBtR0DB6zIc2b*mN7?c}Is>|YU55>&wWDcmRR!EpjOen7i+ps+pTeq4NxC91A*S=x(sjp4hSLkmqYWkXL7vy0njmmR)stn$z!1(pdt zQKDj}2nh51#gX|7%2&)P!hoV&VUlAwgEHo|3giK=;P$&ZwGSXKJC#c8%*c!7(Yq5> z_&ebypZiS>&AjmS9P&{;zVVjV{M5`g46B;BH4k)m;3jq)|ctJyJ{|T{6@7? zvfMYb)|X+Vl(X**O>arg*}*oOKUx*ZQhR+TtI~hqAbB7N?#J-ibOQnoVjwb+h zX^W)_FZ6fTq1+nJ2In~4;;$3qt#2zBf@vPF&uSQl2?cMhs!2|PQp?=J78;47i~073 zGD3(oOH!py6LR3lyc#vV2C!80!k6{)XCLu(g zL9pcv6`y2L)It~bO0;n3>41vwZ0}OF6x6G3AtLji5e$GePS z4qShb{bnA3fidG(49prld&yb$1GduDZtpq!Jv~xFNIPVm<%AcFC9XQlEk)&aimhF_ zcU%P?mx5`&s1SJ=I?QyabD$DitEw*ccghM$g%qtB2J-Y>Gi4AWzfqM)6!og# z4%{XHuIH7k!9FrT=2wOmpRos0!=jdVm8KQa<6$>7hbVbA_Ag0WD0C?r#Y|+A1G)~q zWRuUcpR5RYNUSaV`(04V+y#T%m-+s&3r;ds7LH>Al;?nq0tl{QuZNW(L80$_+|oA zjb(1-vwidY`snOHuPeD#C@Rh}CXiWzYuroNg^ve44ahb_Vb*VE&DZz7n}Sybe_g0} z`UHM{6)4}n%M|Sfd0EwL%HzgWC#8z@ZE`~C6Vw`Im7L}EudzBUB@?bUe^di`s5{vB z6M4wObCwC~{F*iZsQV^4XQU%-J!Ej@nljR$$&U}A+57g7lA67|pbboZoALXb#$sD2 zgX~!Yj*h%Y%=`>6VNTerXI)vq+thy zsrB&LoaOljM1jl^(0OwTAQtPGf?FNh1xoI*ieUTPPzz>R9eM`wHfD%#Q5!9A7!8^N z$Q!#|&#WQ8PE|c<`KHuAan`^7<}oPzWExuQBU2_9&=-R1FEc+eMf{&W$U>{btPHS$Q-uw z8gO)N8|y4rc3)v#MY|uaK|wTVZ^=(~5GvDXe}5029a2@2;u%wr?y$CVnX&RyOVQ<{ zFJ35$9XF9ji8)|enQ^7}giwWnDJ8Atj2#Z^PG)jyC|kwJ$+)&{kGLv<*r$*4GXhiR z0V2ph-KWLEDgKbd%!#3wA0(&BibM9Nu7{YzloG7h7tt3S@x6vd>>G%joIMW-Oc#MQSj3skX!y>@%nxGd!awTQ3H!eOcyjF{U9WQ0Eg)@BFLG zchq1iN+|r5bk&5&Wa#2V$BQb+gqjQp(EB)6B>bMh3Uw@nOWP7T@kJYZ@ZI#oMYOUayFWA_^>u7C>=lx+rknihS>Acc;1 z%Es6)Yn({`+6xNDPQ#R7;n5S}(531ki}#K%dFt1tsF#Ca_bApM@LWQe$N3BTOSt&}+L^$~x8jipmBT0v6J&Fe_h=DFm2~o%&A-Yo_I1&IJ5l zD@W0TzfgSQh1cf0y7K~>UpUL&H?G)Nu2R6nES z)Bp1bK~dQ~tTD|ijwwtmcL*`xrk{SgZX2=~*pLPC_pCFT#+$oJWKPJTEY=e$YyS@(3rRWbYFRDgT4e;$P2XP46HsOrI z!<&;O@{4_}fXYDWy^-~#VwRB}SIvG8_gd(DS3#fp6B28G_j9|?DQ3kjC9fGF%`lHDHI+|;9Xm6oz;Sa4giop5M;J@eq} zDr9-^t~$9MIg)M;u@A-QxRrM*dr^7GsKy#fWE!f`b7ix$YWyryr}uQX5M}`+ZKy_p zb($7q`w}(pxm-s5NAM$0!%QZaPro&br7x8|PC0_y8%z&tc@c4)0C^OrRS=SWXYO;S zjq1#+WoE|b+$B5_r2No(2HH=o!F|E>mC(Pb66{YKYp#dx0BAMEF`uusRG(JWjsSf} z2+Q%4;*gMj?>5-Ni>5a z7*R9pve*u{XvB_V%0sQD4aHGFLYTmuU-<06&o2EXsith{9hh2)$?n^4TQ%&fZihUfHfpxURb0U zAR74%f1>N(d|d;$&848D8Ts%@z@Pej{-XCyFhqcp$1YT7N5kR9HHk~uybIc5ga;BpS%WKXamL1W^tlfk7=zB4tt>;B}_ zs$1#z)POICjSt!3Weia+KthKjn^cg0N%szYD^?bwW*@xNhjz+Hnl*g~s zb~ZW{RZO}t#mTyWr^jhT>9nT5relgj038ef0Ynol-%_2TkO-MQ*b*?`neSH?=pD9z z4q>&Ywb|TD3n-wK>C(kNaz z?2;sRp4vcNb*x>1y5OfVjxiZumV%&T~tbo^f3|toM0Ylbe^=!uBK5r8d!BRYmL@MZe-Z8p_pGK z7>7cf`v65&8Xy}a=^oijF|Om*dVgZYFbca_Tkom&!u&+qKA#RanEtLCnv??yi*WpHV`N=rHyd8_%9a zwlSm6u25+CxO7+&Q_onPg)jE`{vfRF{Ex^=qH*)g^pY}NR*f<|-N5w}8lTT-3Ly#A7vm-VD>IYwW=}ePPx`~( z=WA;Rqcz=YXFJDU{lldO95ne<*qW%W%ZeZ*F!in%;xsMyek}Ig*|h@XEKHJlDe;Mk zJsYo0zI22;u3@TI#k(;xVAMp1b%#{jx_2Tn=g0}((_`dxP2IWA6%wwp#h{EQRLj}) zb(t50OhkcEAHhEK?2%20h#tDFS$=WhG*H}q$+=Tf_3pYeASA4tlsTbj2N3^qv+qp% zi4Ams#ELGB_4olTcgg~qj%K>L#Bxz$u6&oR^5gKe5XLG zQ`&LPk+OYQu#M-g%EF4a!_&<{UcYfO$DDyDvNcN!t(3VALSHP<$7#!B?^ppC%?W(lo8)G#xP+s-Iy&rjOl4OHx#l539=n<2UR?~5#h zS+xiIF7!6Wm|HqF4Tk`Q#dc|>v+D<+JFNi~i3G2A2FAyJNWvMDx0HE{fqMNcUxS-h z+MYl;P=3>`rBKKH2y*nI1?n;kuhWh`8YOxakHj@4W>Xu6ZBI3jV zP{&`w*)zu;1L~)|NkEmQR(vmgf2E^oZC8eX2T;xQn#l$4C8%%gvTMBI(?IO}aOFLo zk^k5}e6wPF^`(N;Wk!9?M6W&?{-+9yjHQF*>or5Sj88;$85}^Y{A`^?)`xKxFdXku z3Rl@Rwr-NXgAn~%2x^ii?+(8Y|+*xFUy#TF){^Frn+L2I^KR=bJ=6gH$Ye$6-Op$u^cNWogF3`rn<-FJ#8b5c)QotQ35-5hA zV}6qa8sc^nRmkL<>bH7Pn-tVUG6g|7E7k(2Joq7I6Qd%6v4gbi-Bvk3on^LP_9uRhWGit zEuzM5lQAj5ljc}HE?@8jD93Wj?QxqJ_0#;e1vG_8HQ+u1=_rRoA06mtrvFB16pep& z_ztPI?)~6*QK|)!9A#9Xw+euggMBZ>M2XSC{y!KS@ZQ`uxI)Rcq1n=?SU@5k54ptd z^gy82_lRL$t0Yvu9wv-TuRuX!qa)~->krq`p1zn*W<}2&d5<+HICEza>cXX!k64XZ zbyfcJ61hj^Z-AmHbaUYr#fr`u6%F~lxx019F`N0sgG)qEeZzVgQfBy0&v(dziX$9j z087!Hthb0=W2F|VRE$3(Yr-rjxiesk1ewG`T;OjEtLzz)&D=F`<3aglncuF9An1vp zCfTEABl31bySDjR&=@E>McT){)J2K4ub*?RaTVVeg_~enF8t{N9%IS>+a85lp9gXh zJihhZzjMZT7$D@lWB7$+g!+6;_d-rU4qe`wiXnl>@H!vH#P9XBq=;M95<$T7^ z0y*C64M8{2AyXw~Kd26`HmGmKEJY2zsK^T(&4-K&qq$ z_oLz?P<;dF9Ix6KUquT^ews{NbZ}f}Z0Mj)&VRU#N0;m|c276e-@nyF|1fZ?t$nss zX*;^|i<6QS5ZAT!DzltF^xt@Om-}~i6B*f~TWRI8U0oYLxbZ@~Az}zy zfIVps|F}6>mGuO_xkG5E&=4f+Kov1^kX82c%@8rrst*2KWPP}`?YL{Wmnw3py-Nar zz3%&M8aT&0AJ6=8y4YX18X+`B7WNzkU@x5@YGu|t}_sr@a=tQ z2M|ezEr9e_Z9d!e!x9Zms)xtk(jOwJ?FdQyz5$;=K$P|G{at*T=FQVB^Wkd%3v`}! z|Bm}t_P|}YhOO=T`G>1dqXh?Y&gywQI?amL0o=H>jV^H0wSPFD>0xRGvamVNd+MjBt?7nySA6-I92(h5lly+Q$=zxLbgk7cHcS` zNS}NQk>8ahr@gc}<$%t<+*_xP-o;EM9GdFcql>EtmtVA{$ zaL7e{>dNwy2aI_&b&J+?Eh_Ce{@f6;Jh$DvaG6@VXqHvxCtK>Gg%(*K-mQzWB4m8tjl6#r;3&i=iMKe^ifSDXG;)c?G- z|K5xLQ8s_W%`X%2fB*R#ZvKXw|8J(>Dax}i@^`PmmM!!|IE4fvW;O8-+=M@N1hnZ& zfB1==7W!}D>i;zXz$e+DXAB5*>gV6R4uN~w4^VIcYeAml?7;uk@qhayk>9)x+bOpb zlm4d%`*R>KfHDrX@Y0=s>ofYpVgAcCp`8psX>r^G^Zl<*_Thb?XmK`P;}1QBf1B37 zUh~-)NRRzJ#eZ3Ye^J7}SMfi^ihnEWe>4WbA%K7H#s7qtf5Xkcg73fK^51atH{ASf z`v1sB`WsRI*g*lnAKJ}M>dGxHsS`tdRoq>NoJKKU((71*Jh94ZizE&ZZqxC#j;CM$xpjOgYxH>R*JX7D;DMiCi?H9^dgN07sSE$Pdc0TJY3&pC`Qe}6 zJc{+C<4@!9zAfE3v9P-0SRUq_F8pw2o=2@8CrzfqZzfItx6zkuM(9jc;OBmd=;Qew z+9Vnpn#V4D>FCH7i1?eh_LMQyV&r|I`$?2jM(paJo~~lH6N&TX`#Zhj(aeg-+llx= zN4}J5?pt$?W@bWOJdp`GCvacypGj+6FmmG#Qr>lcm8neh+^D;Pw? z*V+A+G3_Ucq55{^cSKT5zR=U`X9P^0Ar61JpE5m7xx`&V>3?hGPdo};<2=jG$cxWa zk2VrdS+6DU+2ZdbNtT8SL=`afh&`LkU~0jx;q2W7f6in2TJ|7p>tx2{72edmTOLpn z;oZ(}wh)_RefR?y{C5d>%AF8UTe?s<@CmtIYaMr`rd%x{s zem+PZywv;8VUH-1Z!h5@2^-R}fyb*WPZgzT+tbo^>D&e%WS3mb&Hk#U&L$*;b zjO!b3in&iz4o}_#|HpZjdAPTe1wq_0B~3VW_NymvM&8#|b(H^9qtlb7Ht!)2S5TYl z3S-Tm!GiQTDw415Y4lwiFH%$Zead?J)A~<7RJ1=*iZGs$^I4r3OyQ_Dy?aq5+G@I? z?4vOs);*W+@TE56+ymIFS&e4dceCcFkEL~1BYOvDa1)o|bANgL+)RWaA&rkecfl4U z`kwm7FZujN@`2W*Gan)9~Dl>$o8ut z0!?&3g;1$yj*~m&3Su;6-V)tf5WKD2zWo#gpIr$WM_$|)Z&Sf!&I}gxP#K!u$)#`~0 zi_xAlUpwO6=WK(`cR!M0xtFNS{CX@W)6IzB!=(6p4Jkdj{juaGu6SOJ(=1wg`aKp1 z8TSEy;3aw2ZVrz>HJ)PB8%6p*l*{>4Sg7i~I>L!3Y8xA0xpu7sUuDI1v7rZ=7xhma>ey8v2>EVrbSMk+k`=%k~J&9TJZyEXzbNG&Y-IFY{b z3&_;SHrYe;A4k548<^qz0w2>hP|3xXs-FJkci$$%X$WszN?G9NtZbb%&liaJ4a;pb+UR;3;g8|nnjGN*$FzIvvJ1-|WO zb1QkhA|qd@)Ju}`YGdQE%@>rRaX;8@2DY$Zoqwuos-|dyv0}QG9EO5r@S`Q1+~EZf z2)rPn!8|jh&n$e(2;xF3$#rp#3=5)^#k_oxkU*beOVTYxK`Y&lNF8Z>x6A7GS{l{Q zvrAZZ0F&h%ZfUZgzFtqw^mPc^djkqF5RpdJ*0hghDhGk*SfsfQH{Q5(n)r-g#`qIf zS5rVnPW31lbhF=#$mgnIb}(*}vEyrgm!3NE40}?R?5ZHmQhoZy z<4f#Lz^(hie)$mk;Yt{D*OoAKCi<2@t+yM8F9;!#YTq#vhN4jaUR7MI7Ad11-oLc8 zwAOznU_ZocH_Ri@&d$yqmcHI3yHa{`Bc%q^sKKZ1b+aicHI+|cf6d3LKS|(b@c6KJ z(+e+oENjWs*YDx;5;^~T7Ar7 z$fq6C&W8Qr_R&OD@sQ=rZsnF`&&OI_ZO9jGc(PP{|HBhyppT3&is0cJ>?>bo@khYZMe=>p#}oZ(EKrbCHkqu zOt+_&t+er$2P{MvIjB-dnbqFbW>TV6wI)$sQRU1wDr_<8+0ogZuo*>_Z-H8q2dB%M!k1eC(J#qv(`qV9xCFFDB zPbW!n=X?)BE4T&gF_>S6r1TI!R`g~D`X0n%;0q|PQpFv;^`2pdUk=GzMFX$0hDwZz zK|I07K>9$uHu%fxIVY5^LaaonJ4|sTbwN}N@R6_56#L{|FuB6t{kFZmz2hSpwPjnO z4*Wh%`Py(**YgfWf9Hak8G$hf8D8|XJo0hS{_w`^V}1Rr+-e0w7fBp7}Qsh08F9*hMWh^qf z9Btg(aE4Pv{B~lQMf-hBaj`A|CgZQor{>%A`D*3O5s+L!dfCQ;x~Rpds7>vM54W~i zCC|QBigi*vJP1VzFv8QN_}gR5WMi~F=ItO7efvQQS0VQ0#W(_@erjqRo)tfZ)H8NMO2M{%zgn;LtTXbJN$uB5S zxg*Bv)QBLw$L!?gSxfou#$i)6E=o*pylLBj1k%1B1YZ|b(8C288$ae-%Ij=utmb*m zjM~HA4LLKIs&Kz@L^e7qYHltU`EV-BpW;=Lz+-gM*ak}|oFn@KHub?21!T52>k&Tm zfkOJ-QFuv!l19@$2?Lyy%!1{oPI^*>c8-TPw)grr4pe9TOUff8)HN#~j!fFsR1`Ya z-)GNsIL2)X_MaY31sU+>{lb-+FC;o%44a#c`Ih~?I_~{neLDAhCGL5W8a~wW&|s2B zrkCsWzpk>*eXPz?A@TxY!6b{MGb{If^bhoQo^+(iWEVg2l%}t474ZDHYwyMd0+779 zZ+3X$ysB5bO-_E-8>E?#&%{Qi?u@?A>(vcptn3vVxM}f$7i?T=GZzsI8GJ|sQ#c?f zy#LMj(fz4iHwyr3_td^ikHzm>w9jd3DsTdB3@@JNlrs~DnPf=jertwL#PhZ+@kHp( z$^^6QfWg5*Y{(B;`_i3WGh&~?(8G#)Q$1Tr0%X1`qjCDs?*4w+l-#-Vk9`P(ZhPfc zo0C3GT_WpDj$4^Njarj?+hg;|v~pZ&)uxTMiYQeHY~Y*{yC^2KZEZj}XnzbfOsM%56TBv?BX7~=GnH_I_b(@2q6glfJP`NiTKgiVII<{UsHh0+-rv@^mW&^Im1_v97o<9>VfIG8G z{hT1H)9<_C(|u6U6a7yIhx_bW#XxOmziC?8nPVHVSV>}dg1BcA+Zhw)Rqoiy3dBMK zrR{|;#QN1ASqt1BGpBMNO^rZQoz0IJlu?RqV6lC!Ec0aF_mZT!qB$;8V#>4dF@-Lv zfMXr~M^C+~;@1Z_4@3v>I)E28i^kOjqAk6Qrv55yyKO?nuR?s*Y}v&~37-h;whYop zKX-MpC6_6mtb(gSW%yDj;B==Dtjv_x?KuG@RGoUf-zUf6N`poMk^QBv``YKcM9AR| zg#zw$23me_LyAs+Flund(8D8s{IEZ(iYf!`%``_G0SB)dJDzpUglyLjR!*!I72ZC) zY!bep;6HuKp!$hLQBSS&rxVseh``yg!&z;6AJm3_vGA>BmqScY$!(Ohu8`sG%;AI( z;0)E$LQIHoh-ECA%VcugA9Iw-r*sW+E$^nk z>)?TYW1fyq;=;#QktaWR_+;TP7R}OR_B$9fv114`v~o%)X;VVTR@&6})>t{Va=-Nn zj&vDe7HRrEu;%t^Q{c|YpvFT*w0VH*ut|ungN`>7Nrnl>yPn#~M6SLbzlyyZ-us%G z-)9IRii-^uO}i}YKQU3CEER2E5;W@8f=bqEWbS+QPgOYusLF)P)oR;{3Yt|Xx zCy4~Liw8LkyPNz&MBC=QWZJ`f%iYl>0v*OqH+;4oJ!1ORdTMS#oTu1Mg*m+36W01* zR3z+Rdr{ta;Zj$H zO|&KOx;l-EMyP;|`DAW>5v@r#!dP*?iv&ZpSaxyC^tE_isnA1-oi%Do_~AhZ?>t}} z(mfy=1K(mAIF*$kD@d@*bZw)l!r_(@k7IyvU{PX4a%$8#5c1!+h2a^ogX=U^vWg`x zvInm`+UsuGe_&2XqSj35#3m*tel!)yQ>XD;^(u}CVfWjwzG?M?YiUeErnHtVl6-w< zZHiq#pCK28sQ*T3=Q05?iFzP78HSzy(CU7O2oW^fM;zHWwa(B#rFnuDL1V^TfKhxA(67P`mp4?sVQ@H>aIphEWsT zw2r@dV>TI;B4&|iX=SCdmmAGjNdnbnbQ!@p|K_~qw%X6P>>R!2Q0T+nLoLv!HcvRm z8Ii{S?0l|f)+BuGsawOE+Zfvx(Q}G6gubRqrgV+gX6sqT7ZbI7&D-IhTdkopys{0iZI#o)(@rC;C}Mv@i-dY3LzKexTa z9%rY7dLb8XNhcdCvZqM%ek^s&r;0zz=iz$bku&fcvxJK?gDg_fOQmD}2JVX=R*7UV zpNd+2*S~><|7zTwllV7>`>}DV%Uv!8JXr@{7)f+WXj7M)%BV7%)_0O~;Mz)FV|@or)8yX z*b7>AK@R48(2JLA46*!RkSHUUBPA|mqvCLxbb5lJ%ek?GbU!POp-5M%sB&~aU~&&0 zCpBJ%|9trg0Pp*s*lGM%(_jvV=w=5n<3(F@Gn0w!z9>DV7%j6MM94zFhOK^k;O<&| zM&X!)grlqYNUcNl{#F|!tgN9m9@pa2?MLOr$O1nWv313t1g9f_SL(eM7m%v61_YV#_scZ^cYir> zWJKfEAPEVU29k%Hv?K4E@)n?l;_Q1p6nB-TS_Ce%Gs|WnH^JkEx-FT z9!xFV4(sxHSu;qAZQ4+nqSPJk^O8En&h!)&QSy6r_7|V)nCjZd9#2p+Bq4&dR7a@5JGv*G|p?kVCq5da&!p! z@PdWhG0gsn5^s-bX!-KfJz^4t)yA;ZW=snE8JP^Aut3nGOE7b;eQRCdsg~nckTll* zqt@ruyf<9d^ZVIL7Ppbk{ZundKWLm=PCPXMKUtX0d(!2STDr&CwEtp%vGI_GG6OA1 z^9250T%c6NdGc+ud%%Ut)KXneA0tD<)IeNj$Z678FQ_id;iBA5B^9V{Z1GdXLKA*o zr~@2=&)#1-3_gEe+NWK>wDP-&Ctrj^?v}oIiF1PkqK#@w8mxQO#lTH^gmhbeYa`^$ z^N(KRqg&{$mI^z4J;Vf%EY5z7^X%o@X(?I9rA+5P5&#<#BV-9E1sOBLB?8bM7k{|F zATi%!wtt07v}lGny^L`PiwWorBfI>;W9K*LdN054wd5|dF>dTteA`g=1J`JzyoUTj z!*<77Y1~dU$hh9wBwKSTpIrpQPQ2%nCOCzip;eh$>C)*#7B@A#krCY8m<-u>E$Z=e z4GBo@G9*P1(xSHCLZ-{!ES#Ze8SF|IE=V`Z(ZSE#$mt^%#+ zM(b$0XrwmAcpb4b=|G%x!fe#FuWo%-IO(%C$*|}@#R`~0c$L!O^eNCo5I9=*xzAX7 zHr2fUBaN!s^%f_Bpbj31!Y;tQbRDYprm0XZ7PtVvO)dU;;YDNuv)qHa7=WxN1O^ydhEiQvR*p*MS3;M4KjD`6n z7il|WNkgij5dSL5jKj7){{wJHB68AKDTW0`o1(!l)`Z<2JBb#x1nUe<WsX@*l zw5WCoIDQP&aF`R<-j33*+WlwWxi>6$%m|CNEE4kFu z7zNDYBPoa5D81!2=Sv4{;+CIL!po{OnMGCNG?NZq_TV^wUEMs9y!k`?7Mi(y`OAwK zKP87n0%)h9y@0BOk-MfOr}C?bCmTCqg6^Z)jG>)fir332bkWS@wD+WT(prChYy#xe zI4!Ha1pClO@K+YV;W-(W@0Fwlqm7&9n=T>$YJS%{t7{PHB^{JSZuWY*I{Kr>-B9PC zAVany-E#`t?F!Ns?%2(onw&ehr*uv&w>;%`uEq06p|?VS#U(c*JwN;?`UQl|CECtV z&dE>Si~v4#7W?LKcRj%K>-v*EAh+s)M+fTwWOEQBObXepxeyGGC}^ z{sQVpIvuZ2a4fn;szT6gHydGh%JaerA~ye9;Eg_IqyDVujaz_O_?Ch1s&L0AGCEBF zCAMRCGmLU$eTsEtl4kxrJolVZm zPjcb=`}^+jtF@O84h%h4OMd&7ZBuqv#lH@!*jranihBSJw!GI$*NGrw+sWb==6*!odya zYOKD24#TEwAFq!5nAL7W*Mj^u&(I(f9&jN6oEZ3Rb9x}=;D|-Z4tqDYh zqhV;K{gb7fru|9n3tJ5(?PxK*^}5bZ2vpff+IA3oR_>XytuZ zr2FA>a+yg3>Amaf%<&zq-J3O7ri07K*i;}LgobJ+e%&LA6Hk-nt73LhPy>jP$k*Ed8A1 zL3ku>7mzdYGgot66fZ4jr8Y?UFxY`e<-?jXRCzJlM8V0o6W+(U4&FxjpEbt$aom=n)cBC$c9kr}m^W)#@a9()jrUW~?-r~0t@6MTfhRsyHW*Vkekb&_omPJ3 zC=Wk>SKt&PfwkUSD#*yk(I*Ur$7a{zf3%&$Pny+Ps(gnC$UWUijaX^k%qeMiND(Zp zo1Y*#SelHzUydLkCQ9dbighJ@!cqQ=FUjeq6rEa-3_sbbUA*h zNFnl`Dwq8lw~|I1PdS>G(B>U}VAelZO6&t++YAG_}RuJ^~y>zM_}tsBP@C=LlG`BZM^xdw|bXK6E=%$gco zH0S3Ev@~|*Ea!81c{MYsir0`}GxJ)J>kVFeMstaY{29NMupy`_0?|fEQVbarvknGw z4M&0C3tHz8BQ~Px)cyqwb;&xo#B9nj-@aORd&s+Ry}YR6%3=uc77)CBosG5icOVyO zg#tzUaDE5!`Vhx49_j@cwe*jkyK=zksea;?xu%Xk1_6Vs!r>b%DWS>7&u@iX>tf7S zOO-4e+)-9;$TMwUr~(N9+gNkGZn~ao(I4t-Jw9C4qIAV@C~Bc&f%ZJ~y|NODYytW{ zEg9ulq`+@dozib3+2XOyNo~$w{ z#GbHhkvxTg?CJkvGW=)q&oBSXSRi!#W6Lsrx9Ac3)MRvGRZ{Pwt1vROrv1^Al>^!zLo zQ&;gIfR#HSxqHQ#?7WSWfo<72a5I^Xq#L5Cso%vffTd*n1s7Xe9~X1Splb}}_!2jf#u5^H#349w(bZq9t+ z0z8X-xUitWAQ%_2^T6-!U2_uygOvTQIUNq4Mz=!W6yAtSBabq0pSP_INC zoz0@i5Kb{Mu~4s;F;d-oWj&kF^MQJy_w##uHn(n_STDz-=Ers>Yzk8qI0+u^IioC? z6cHP%*#+1p408Zpd~w>je+R~^aDZ6r*NvsjI~={69CE>NxS>NkT*&jA-PYO*ER?^f z9WUnOWuC;~Bxl;?3Acxywdn;SeP3U{wxW-Ww<;ZJ6}D@znS0fT`$8x=|L$2hm4}TP zNWam3i#ha~H6s0-{?PQnuG?%nDh=Z7WvUR=H#OdC{&jsO)a^P)&n41*K~hywy%lak zpl#o*4`0UNkKh6+zmFp6XWD;(!Joj3J6x`H`!)OCvzj>#5U;l@)0Co0xdwbj*e>%C zh+P-{ImD43CeCC%sM^j18N9$!Y0N@@hbB`?31uiYN|WvaWY+EN9WwZu3{U9u=Q~tw z_@2CIEGOq1vweN)V902GOU&iIN=F;x#%D^&Lo{O(7f@^!51oxp>=bJ#xn&EsAF9JF zY2|l>ria6y*luB6f_Lik@PzA28G}Zm^u%wYT_*x<;@S@lBb1te489~o?(C1*xv8d2 zhFAeZa2=|Z{awEeqVwP+MLL{;U#|o3K`a+=moyTw)9c;7Wz-FmHO zquwBe0#3Pv0vXKU1(s*Jk%z5XnlB{B0UuF|8(*_eOk{w+KdLlTGT{a~;xni(0nDt6 z?q#>5C}o8VeWgx~Z+44j7s+Q4HuWsK7av^)LuI^yqVMLAQF6UzMmzUHP|z3cqJu7K zA!&pMc5^O1i;%9CBHB@)2vtBS-sd6$u!K9x4*`WN2Th%;Ub-u#vd#BpqS5K5y$ z&{W>SR15l^^WFAa+MtIqY1tKxs4BZ%Q$(vABUlO-^a)e(Ht$ERnUVj*d4n?hL=nq& zxJl>-YEbRH$B!SAAi^1+5>v?$tj@Ip?aa>b*hy^c0%5I~Mb>k;*7dGX*n7uGM*njf ziF=o75v2R!n4}xlINM&;O@Xl$gNi#7TZHWWz8B~0CGCEEjr-uyjW8=&!^HH$r|Ta+ zKfORnqNk#$us#fe^cyuVBJ7)nECr)F2=n33;l#Mbzd**-;Lo1H@`@{PutT>GRaBoA zZb^zg7_{6RaMl&|&9*KTjZREYpl^^~=wdb9Fm9b`(FvXq>FJq!n$?4;pWN{7rD_?a zwBRhFS&PliJzPHYZ0CAAAKTeb#Opic0rs<_##49ICs8>`AsI~pJX;` zBP)q2Ibl*=(3%Min8=6EQcN8N13~V5)GBb=^gJ5e1=J`dQAM{*ocx8q>NV|iqk0Wy zHkyp4%GG$%IKa9Bw85(rA1+y zF|cwkleljq&w7!up(o$R<&uoQ=QyWZ_Q$j)8 zW*Hn_uG!kiEJ?pk2Reio6VHJJU3sI|NV(#B1>bC4cjRwdFp2TA4RuLJu%92z*~mionX^xjE4=fxzx^E@4+RSF8gnUnf0sIn-MybS(B%c50OIY z3R8YT?Qd8eGHw?cO*zHr6@9FBJR+fl++Y}2*eDCJ_iDM-RrXds1W5ILCt&Kd`nvDZ zw5-k*#J6a+3ZiR|3g*d&2cvo!(MspL+o%Q7bRX)?vIC`2fNBc0s@~e$dXV%SXQ$&w zG?+#{-Al!cFT+5K+rszsT!pO%f*95N# zw-D)h=~sPDJVUaWSYh;K-aX!XT`;){p$(aKu65dT zM-|`!CFs-UcX@X!ocvWpB&U-<3y`ytYJd88b3qa>wZ>$Q`JBhF0VMyp_(epxe8-cB zOSQ?Vdr@z!Hgk8>L{U>oL8FjY+hSF)euE*2_eIf_owbcS=g-sK3E%a96c({+cZx;C z$FWD8Z*lF?gvwoo=>a1k!1Nf!w7!*2|Cz9J5}Oc#18o0vnFUe+v+ycemDMS0Gj0eOqJkzqJ&3sz-TSr;>@3tNHlv#BuEm$8f`F>L4ygi97d z&%2LxcS5xp<1J{oWZa5^=$g2y#Sr19BmEFo^iNLspUIYmg08Huz^A@p>~89f`iViQpPAaBkW>fjHSfD{`w@R_FWH?a zu{=PT#NsoSbhck~O4KLHEtHUolroeiyYR5bGZjelXVU`?(%%3`HCI8uK%SlkVuG z-5N=i^6|~Ph}>F`hK4L5AMR`TSMuB(Js(Jf)`;P17bROQi{HuZd>Sdb7Z0a? z0n;5c7BNblh|=NoW#d;_xb;&nHBXdyPns8aXP#vj!dp2oP|)7*O+`BXwKQ)Xv^non z5f|9DX-k&quAe&5cUtc{&?u#U_2x0i=#w#xK3Z)6m_;SnzR+Kj{NIa-qYq7_$i&L& zG!>#!XQJ*tc-h9){p$w9q{~5t8SU*O;82Q#JEv!_tFXH3o=2Yq2rZP#HnF^x#`MJ2S8Th9m4DbvX4^%I|kLL>I@xE9bC3!=m{@BC> z6}aS5-q#n;Ehb=a{``uMA7<31yO;Aww7o*k0bP+gwDaBhjwIn?t#R;-Y;xEhaqv$(d7ATfN{Zq2HNpY=F zdi~3M!Gmx_v9HSVgD9xD)lmb9>pP2oy`*Ke zJV&|z<7zR>$)5wv&ZipC)&8D3MTC>WC8G5O?|VDpKq4<9X=TjNhr0nO4VDGh_nIf( zr6;YOnK^E)1Y~|cyF(iku(5@lYa0ZsScLCjW=^)kY|8rY-rf8Hu#S5+{A)H%`{}1l~GRY{J>Ty^_Sdw)gTibb+a92KUdVHw})%uatgsOT6=Q zOZ`sLJLCI-dw8P4Tc7d3=<(owf%A;WB6V!(xnlwJu#(xUy$mES2e~gVrif>lnpnMQ zIQZdxN$SbV>y5#wT=`rOJ2vs7j-D@+#jqfiq}TO>^anVQ0%)X|DR#Ox4YkRCk;!?= zZHBz(SDv%-ZnN&HvCD@{b5LPLRRtKz_a>16;yiSe4qH9nlf75>#>f)Wk}m!N9`0l6 z-l%PAP-Up9KJWRfOYm9q-B3AjBOmx+IoXLSgMwK%OF>9TZui+-u@PfKs?{%?P^ME8GJQPzxv(gX#*e?)$X5J;h@mMgX=#%QT?G%&1;72rGX6tV zKc2P4)!1uDXyPa{971}Ti*TV40C;-7Y+fjl$(f#e^&U3SGr&<~YQ{Uy(5r`iJ#y>D zKV_~$dIn^M(Uy)KnRy>gZbpE$x?7Gx-m7=q`R7;}Bp>NtRq&g8LT6eSJ94dftWGD> zf8)h_So^N65i)Dn&ckq_uIv z#`Qhp*ZG4tJg!Ok-5h@iY^UF?T&OeuLn9;fb9j-1{g+IOr%ykgURF>nzj?DYX7B|6 z5a_L#k+`yG(yd=o@nP_t9W?oPYooM8Pe`RhH8G)E`&LHVZ6HU1si+XC;!jFN<)ruS&s*>>du-%u8VC+hB+ z8ek(Z+lYq!cg7-4bvgO@I~TPyj!yzO>ql+UC@sVoFW1&uSBe(R z>&zHem_VWyGvmoha(?)oC6E5OCOyF?N}Nzzp?7T{CMx+sN1bmsh4dJY+MX4=$-Gy_ zIp4r08QT`&=|KM`=4)zTFhAIKWJ3R; z9oa_ID`rr`s7-7M>itlCHRYQSbr3OfC-OPmE4=u`$+^0EOp;H~mP4Ll0+bKckM5@A>`#pBQ%=<);v0MV1fA z5=5o2uyC$c&lKWS!S>5xfRL!G!^-_8PE28$6`XP!h|p-uf#F-1v#Ngl@b6f33!i88 z5lj-+iXgzt026{XnGKT2p#y$?iqo}tu9QAu7^@2bJlgs%Q%wT_C{hH}ME{tG?L=St z0zOf+=-8j+IBAQ1dX!k{5j--N9SUya(#rn!b--J-z22p#JFxw@5;_5}&;D+&oTA}-3NdmZ7 zLXKGZKI8*7Bl1_1M0P!;tP9BuOx`DHFLG>3@Bs?3Bez=It+w|=A;=j2o|?pI^^Jg0 z&gY!rA=r@v_j7Hi=2isaQaA09mJBMczjJ#=^8L0|CwzIE6A-+R}Z zoW@QbM?lD?$wf*Oc0E7eVK>E9qf57DfELen>&MBj z2Yl-!@fgVHCTtN!TWhY_o?EnVc(yFM=?3)H;K;Z;QW7Mne~NfpA27+RRrN)!MhA@- znNswh^WX@gF9TGvlwTT~^Vj_)81 zSG&eWhfrtIJCpvc(th;RKa9aP`_Ex$waVHxYx_52aut%v{4b_x>}{r{_NGB$&i8iB z2iGf2z2c=kiPv&=2 zwb98MJF_OM(Kvy2~4bb?{~xa*--o#gp{i$=Ia*=9hgmhoG33!O6GHw4<{6` z`PZq!HF+5Kz^hQHuoOntjbFzzouX6+FE4jlHVDe@9&RV4ub0l}X$-P5@02Om9c2Fq z84O{%_Y;LMRlPJGE6I(zDc$FM^@WbV6R8S``JWRcbA3~2yb6+%U*A)B*;tx_Q_FQf z8BTpZ{zuFo1l2I^f+j8=Zebw*`+Wn;G`+4*0ndwOjz>n`Y3l`uwH@ zTsF7UEMwZsxxn0zD!OE2MSTVb;>DXxGp1#2w`M^9Sd%I%6|l7i1JE0d_w?68DIo*xhYlRJtxbuI<;vx$ zv{F$cZmBiTZx|Pe0Nu~05s0)lr7-kssPNi{AX9vjCYwXoEGDE^8rZk~&`+ofXy!`e zX(S_~=#8%(t61mlVo)b;^tvI`P_`?SJ1qN)i(+Brley@yy{N%-2Np3Wgkjvy- zC4+WX(BO$RUcn}`$Im~Fadd~Ctc{F*9%cgzzIM9r-S|OmOl~FiIbr6p3tiRs zRVTNOmTN+-(r0%b6ey^Za^4}Q6Xu*TT%l~KSM%~(Mx*Q;o?nZGt`dIMdg{l$KBKHz zz(B$$Xd-e=O{+JZhRQEBt52s&{=<<_yq!nq-A##oZEt@uby`)qHQQK(m&&gOr4K?H z^|_z=BflNuw_e4qr>sLmQ>t%;D3PkId$oY{$QLMqX^zxcAjP0QcCTVQg zc&0a4eF>-OT^e8xXE}oipPu{Xo+;Bt7zy~aI0N-m#h5wnSNV}&KmA+1C10M!;n|armB?f* z%{T>2g~L;@-DQLk(U<$O94vSF2fZ_%?~N^oj8m5ZvpnTEUH2nK5T%~SzLA%**4wA% zdkWm+#srgk$KH|gSfg@k#M#dLj9Vx(Zn8q3h}N&7`@1Qq0%btjvC6{Hp1=BQ0hfJ4 zhb*rcRI>~y2L{Ya+)`fwh^v6gVji$PQc|PoVPQz4{Qowu0hF9Ycj9C%=~@T>6B}|%g@e-07Qwp_ggMB5l4GU@3wBk zB_Ds6#KieM>^HOwKWt1ucIPWi4HTqqJK3%G@EoexE-OiTdF2aU39PEBqLoDd2)Vs$ z>bCBeTY8U2i)_}VnUz)9%dpVU>XX60*a_b6O!RX5h-s?=Ubuy;Z8(?vkDSvY1b2 ztriJd>3Za+j!1TXwUn$|s--vARIps>TsbNt_YA$g%trFQR!Te|1{=s7`%-@OWo!TqI;Laae zRDEA|veFsUD3MdYs};4&62Bh21+e?H>`T7Ps2cUDUxJ19Cs3I0T}s^#L4){tcW3v@ z9lI55?Ff#v!^Qm8iGVpieC~hbB1-Z#(F!>g5QrEam;`Jfk=^4Bn0MzCxzE+VSw1Jb zxZ=ssGvgg9FFr*~r9|b8h>5g^<}BiVQPiFkptzMXV_?Hezxi6gvpZaOXXnJ!kEga( zJ&rewQ*xWmk}D(bNE(Q-V1nPb&fSgnWVgC;j^R5=LK4%1KAK_=%vbB3O))?{N9{;* zwHhxnwnv1K&u`iMz6yG(afI?pM^|H{xD6RwTVWruvD&iB>dS-ILwgL|b2YT~7e#Nj za+f|n4s%HFxh&r2j`NkUqUwaQ90pYinpg5g8-pAs%Jg)*#P_CU0`8G7lp$@KL8oe} zd~NyII9qYs|Bf0>umXmP|CTe?ab$E*+_=_if}tcBA=NYg z$NgbmC~EUQ`=LXg8_ki-5Nbj6;&)*Sv-_tc;8P9krio9B{ie7OLFMtuyoXB~zYwX$ zU&G231TK&kQmSq%l}^_wz75ptIPxLgxF9>~W_G;!j+1QSKW};@mzH?1@cd;Kql=!WpCD>@9OuK3eFt+nw z6R6S0nE{pQs&S~Qo*fO{Rr=e-iR%??H$_cDx2geUHct6VP^yPrx|55GEsp8w-`ME4 za_ESCyUkM^72!+q96vpbau)eDI6VHwG@P*VCC+^Lf5m_O_JQQdybKXU=X&4pg zn-Al*DbwgKSDbU8X}y)CmXzGDmV1QFS<`c>Dxk;v zi;~vCO|!7zBK~`TB6y0rwyWgcP**4`Psllt6U|rUSv`Dtl~Otv6lt9Z2z&6B#boo? zykm#s6_%z+^0}>_Mo*JMVXdv(^*ytl{{i}cW|AoACEQ?iR64h^(w(kk&7aMiX6<>} z$Lj8?wN>6%Z*<51MJ)O)C!KoPeGh#KOssW6Olb^K6gsNXVv8P$jIYcC$Vh z4LkyEJlm8byC@1_^k@FL~%B~zVfkqaJn@Q zeA@l`*1Ge{>4fR9r{NRm9hIS&zH#(dx|tdeU4wSXd0{2PFD3@KcbC$C zg%YH+eIEDqQ*vAVqC7S{PPw<%!;`pMdU8;}lqs(3(~5Q<8wdK}D4_HT9-p06VjY~i zT%hHIKvf>wquRFQAK68Sk_!wHlk_6L#57P#w?J05`s>rq&rtVxt z4{T3_xYJ}`ANDebLIPNmADE=a3rP%GH9&cg6#;C;#;$r=0R*!Hv&^(BGDyL>E3WOG zlCCv@fX&d`Kie_{Y2h+z&2s22D{n&uwM-20?)8EA-Qz<5=kJ-3!+u7>BWYwQE3BvH z_0+7|ohGO<5R#J0FPz)b6SKI$uo1HgcpxgN%NB(8KxJzeo^IBv0AhP&kok$7q2a;l zSC}RAyxK1EG=nDj%xDgS7DaH+3AJ+7_FtswWb9Wx<5qsW#}J?K40zGDd)FiyWttqE zo~8v}9+pSB(#~<)Y;{L6n{TQAhecw|<7(Rzn^_z1Ug6rVA8`Ws%ST9G z``t&`gnpgqHeu#F+)DBdd(_FO{C0S&yOEjWA9)?Q`RVRY7ngSd5g!T8KSOL!04meP zb9A}d!b3^E#>NT(@6f+kOyJ_-kwfO+L$^j&4zmqcZoT9s;y22BLBP$qmwZ=3{EaHU zu zk9JNaB?8f>ulN7PaV)=|g{%quV>cQg?G{Q)wXG<+P3UI)x?>afWbL8um!c?N5$VqLG(zQMK)1jpac{N6YF}UZrJ^ z?Z->J^40A)4FR}2nrJ{Clm=z3ABGACr|hh_-sspcLQ|o<;>TOvtNjS_a$nzd4y&51 z;~hGc)eNA);dxXLoiCj4u=k8p38au@YUbJ)q3>TQ@zkna; z5`DvrjIt!^5SJc;rwt|JaVOr&a0@j_FtSz*y6i>~~I*jMOY<8n0gXh6qixN90q%0i5oA8d){*>sIr>GXM3b;r>_;7eRGy{k0=DQO1> zS%~r9tG**15shobHE!ds$D$CGi$ussf7O~}(kqkeaQx%M;NGEcCm{SuTDWaKLUgI` zyQO}4upS!XNj&lzTEgNve{z6tnXvDal#=({=VrYp8$-`@_sjHmv=caetm+{KFb{*{ zL1^#j=+P25#2NhIMmsr-XXDhmHF(v^gLP?qmrDF`2j$PEUE1Uh_?1$4`O~jC*oinS zD!k6q1YWPNp9l(@I?vZUEOt1`E8fDA>8q;RsQA&F(1R$Jcgu9PX_UBTxp%!kHBDl^ zzk$`z_I6IP(d73=2Izkx9KCt0h0qLLh()qa;-iVouUs0?R3hQSjp3v<$Tso`JTeG+ z3P)-wASVw($gBU2HW{jICVR+^ic1eAY$wadNJ5P(di+%AvS(LeU4+rSQBir|>eFw2 z4oea{u?tGMVvTfoe{xmmLuYj6M>lQ1LhFUyhG2fW+VgY8v|?2b-(724^rCNsraPJ< z5A{ft8d^pxQQz97Co;yH7x-Kb^X)6)l7OFG-w}UF^fYv!lQTt{nPger<;mmEjB>`G z@Bah8Blmyx{bvT$CUFQiaX%z~O_iUE?r?LjydPBbyPN$EMtsj>wz%R+wNAIcN(&w>sP8QQLgFA|syW0ncH$D~O~9f>2_n2EL*X9RXWVH7HA zG?=BM#}D9HagfpC5pwjd-t#3qQQz{Fh-|){Y_t?BRoH0;b-i9 z*JsiQtpz2d<|`;ytTW+1x|2`DM?t|40zJO3!DH*B9h$weI{`2ueB_XC?wPPlFU6Y3eE2uw?_z6DVvg~Jtw#g(zW`9dD_`g@_ zkyRCMsPYaT9z+qZ`^CQ={2=-N@eh4@V(M<~Xycp}=3hQIm5a-6KINR{xGjRX_o8f` zM~J7ZzS9TI{bz{T`+h>hzlL~g#1)_Yqc0DN^MCb+AS%KhVut=4qY|!(COykGKvB$& zmm}jN>%ABrM<8@N(3r>+$M?lVKfOYW=Cy!N5V}{+|Ep)*U!08C4oAsDd6FA%UZcI< z^o830zcigfpZ&1UDkqtBJ}r5}p6AW~D6aqQ1^NoauUSXy!8u)a_hbBjnvh^KAB@z3c@9n=tF}oZLwU#-OaqhJ z$d`YYo&vTLN!X1rKF;PHUey1e)TYV5;-|h(9FcXi$vJ2Kj?n7J!Ym1YucmuJA0~*i zN~*^yQul>gmI3ib9{BM8ZybEzD`H2q()Cb;x0eEh1A?|zC`yc^p zG!e`6l)Ts>apDVC4SQ;M&Ie0{9Z2`(VnO=C?|q7Qr1 zm<0Gw{w>8Jnqqvgf39q%eNz)*rT@=bMm?zsT|uJeCh^UwF9{7c?cEFN`K z&setO5=KE4B>arCBSL>4`{=O(LY{`*%#6I4$N0W>oMq&S#upN!PTB$9&# zqsw^u-~~`>ZV-n@s+6p)@p|e%oe&Z(Mtm5eG&*BUjcw4J%8hdp z#SpvxE{nby>{GPl`hpp1!N9X=Op1SN3MI@F#Qv0LYiR7dvzaHv{9{5b6!(VNoJY*~ zb$)t8Y89dhyM~fhpYh2sGx9^4+9SEJG)oT)6B~#LH~p`W3*;{ZZ=QtvKbYqSO4y%q z`b>E3X3l|Nop08Q1ixjWkAaFZ=q}$cr37$vl~-wp@VIP|Y1C#%yBU+gE@4w*Y!P>|dZ{_7T+&4`IJ4n^Rl<#hUI zxFk^1Y-eLQvbd+s;oSx2JxCZ??9)Lz@#a-OS#{}Ymi$ptkwp=@9(({48nNT;KzDr` zv44$x+KNqkkVo(b9z`SynwnCka66{GyV zl!=4Fk3V{a*!E|+P>Dq12W~ebr=`IADt5mXa(7~%w768lPI{4x>!PM%1Y$@t3+ zXJsTeY;0E_@2{GEmezlvM*IYwG}7HwA%fdvSiJcY>oZxI+&GzY@xaKZFqu)!t8`QO z)zuOyNogevMVh5Co#AY$c$34^vh}BgKOEG7Hf&);5ep5N-3T^YtPJtrPA2Vfxs^xNu zmVC|3P_LGWqPp9+sq-*qpR*k%3v(N_z_{W>aYd#$2z_?Ph|_YjSS%qbrYvY|{v3cR zkjcv8LUJ~JGRtIPX{FTRb=^S4MBbYhlEf|dXIdY{f-wjeIHpxH`B7?Y5VAg*zIAY` zGxrt!Zj6z!jTSbh4T@r!TADUMPFEw@s61chT$0UIt8nj!L&L8jO^_yP@#R) zjMVROLOvpn(GLBo{A{2-)($Gk1GD$+q$)h&rwgP5!_&fjMSTUWgfm3O9;S@fCwrA% zhGWXg>RTs>X2K3sO5GW)e!>TRy`T5Ih~+tdqZbZtzmptskK6pYD1^_)gPM8@2h_bi z>5la-gVu9bacCKvhx(VRBm-|1TcXjZhoxLyoir7is=KPFYGRik^aCktv=qhS$bYyH~ z#Be(_f+H&+)y}1(xCFe)K0OsW5{O^Bx-!a5k+5@j*XT4Y(?kUx;`6K*FtFk@e(w1+ zvI=rekL!E^o78S0{1YFE5p7c{Il2 zVl{thZEUQC5#0vy@2A~LDN^NYsy6t4AYlxKb4n?)9#+?o`ZZPrhe`0*6>j+*85A#! zH1a0F1@0h#L5>XX#B7QylZ(x*KK>3E+J-QA)z+ESRSmXd7Vo_9&o9mnq-{E~Qf0&)%TUgrW_Djt_WVOqa}r@S%B)bRK(a_aLh?oJCh?!uU1*RX0Kq z)7-}Kgwu^IqL}DYrGrmv5rqt6hHX2~y0do|wTo*3rbgm5B}#E^jX)_#0NxyE#$3YVG z8+p9qgOzUpPEgaIu1L9W0#9Cb$R!#%&ll3rWV+s*no=t$sA20&*CV$c4j%3!PbGX= z8mTF>t+yWEz>~wQB_-tG)>F_PTJW$DMKQm_mUh;N6%|!O^*mJ?csjNEuf45i=rGVQ zvW7&}o7~4%> zjyaNk{Ef@}@V-0>eY|@aQ|q~n3hm3>5HoC@#?0J2yI=!1 zmh7B_v{kAed?v&D{@wGLm;t=hhQsh6&gi!*Ws=}J3nLbKtJa(VU#nqK&r&Y*=xc(WRE0L|O%OG;b__!G*d$@B& zwJYPUw8g^B#pOBm?16?G+M?)=RFLO+I6-~PVXaj?o*kbn0OQ~;ooWt~`j(s$RT*FD z^Mu^U*t~v&kN347)|{_^_gSDo-u1_&F0=JEc2{}t)sq|wYA)}@#PfR37fC{JV&aOu zKNFVQ3~{?qP(5!J+Nd?~M=ASLi#m~F&(EFCw{SuvWiyQEd338olqh}G00*VLABCR< z2df%ieH6j}UNvRAwA&EHV(K=ri2H1QM4(-fo0h(xK0<=0LEB`aBUIV*oVk*NLs8$Y zX2a6TXJ%DV`-QnmvH(t#I(PvI^i(^y9R@-A-(AN#Y?zsooNtjtN4-fEKqqus zw@Bk`7sbWZNnL>ExGFKh*#PPDdENR~b#seAuk5 z&-#jdhq1)N1(|%oplet}+naumvz>^k8N9vY{q*|LSz!r<6xGwL4%HMufLSMBZ)}=gn+ld~)GS0tl0wr)l71DxWhhPyXdwMRTf@h21Z9gb+!ci?5!Y zxeVIhs0rbn9p+lk>s=i@aK!@f=1kGlPC^nbRK!+41t)_YfNx4_f(}AM-tdyeFjLBY zlazPGlJlxlGM7zKwY$vpP4pzje8VT`yT?(NQ3p6Q0M2VJCpoc3 zmus#)UR`==chxT=zcQd-dww4;;`zBt?8fbpl*Z?6kg40*u#`&ntJg+{>X{K!^lzwg z9NpbxQ1E0}C`|J`wKeOly&@pmKJ=J6^x0-jfgGV-I4yoBr*22bz1J6TS2!#^pqF)9H=$>?ITEJ9%V%ztMJ011Rh*HYXp6cxVMI5gTxp=H`p09Au2Qdm!?-UmVLISq zhet*-pBn)yJ8$)|TzUOCW`*)=pz6qL{ zT0(E@=^C3dc5H0)R_xwL21ZxjB^(-}QBZg{UiK#hu<;?rXERvM4A@r6gGIxQhCp=4`mMniZa z-qagz_AdsDt;4lA#~@nkZ(!-(O6GE!?8>Xz;!e@WUh=`y5_$rrE`4&=%J$j1Y?Kfj z0!w#!4Yes*W3O;s-6hHPodXA)pV?Q^1S5`I3)m>jGVd3JMlZBuW2-)?L2Il4L}hLK z5+SvfmH0bJMfo1)@UH0YNNM!YFw~@a3gfw#xWYW<7pmag#`x=DTCEt?OBs2U{gg0 zvkE00RB}ub(=P&Ja?t#5u=fv=Fk>Z`VtnFABw3Q9`{smRi}`dp)baeyfIhPy&CcVd zn;1Sv?(7`0iILoGd=Qruwz@K<_gG*F0GUp!;Vbks+fh%>6w+~6PM0>po~Tpt2&r|z zfr!qZia8Yc47AP7X*%t}vSfB^H+*_mZi+`xmb2}7Zp?kXfVHs#6gpa1*%oo!Z3;D+ z^9kfns8iRj+)fZEs2GTUC7rY(?ApaED*g)(u{{R#CskKF z)$-n<(o)$xgucmHzB{5T*)pSGrBBP@t%owaJ^IMLK1IiIfle3PZR(O08;)tFrOrA0 z2`?KCkS+3le6f$tk;_)nsv1Zi(<`rkx(t0$9C4)Bd-3-kIu{oMU!1FlK6>sjVQ8=H z%qi6CUC}s~{g5Ma-~(TtbUoSATVXM1@O|&9vTs+>7|6(b%p-Nw zn-eDOM2|zcuP(~KE8w*O+jzE`B>&OnW|0&X-FM*z-mIamZP&Xie5hCiUZq~|Vp&8P zG&iMivjXnL7a~7EPH6T87vZhN8G|axHtT~RzSmcdWH2t?S%+t9mT4EPjR!jvD!KTc zi&Me0+TkOnEHm^=rDIAAyx`Bzv0elCGlLF@-bVrvro)6Mr|W#vVr%2Nra(zM+VycZ zTD8(L?{=-PYR%ImoQ%limhAg{57qN)e!A$%!mcZLUD}hQ-NE$6pK1QO`g@Zmz89@e zkcVWk@+{L!#=v6=!DGI+)!0R8X>khuOIp12XogQ*Gv*zFD}l^Z&5CY5N5fPe=1Wt< z&FG>8B0CA*C$pZhk92%xeN*K6y0fmm!fk!_nr65@AGlqFB6~%_t*u{Vj9Rno7)<_f zS!b@hjMH*QYkpP#gTQnI>S!KvBg@>HSDR02_FS?5i* zj48wS81hY%ch1I7wkKMnk1x_3Q#j*gG$zo^XBVZ0Cnn^QbOoAEKWy#hx95ugDmb3t zP57x=-89Q3Q=XUJ6mY`a)?_o)*icHln8wDClGB&?Z8LIHPAyGE%@lci#y(Nnoy2d{ zx2TF~y1JrJRI|b>grI(-SA5eniJdJN?!w0_96mnymXF#N~$qEwoCC z7~l65T}_Wvs7}@TCmmk*RQrNB$ln#^l{M_gvlcmO9y4jDGXMTk^tdF!om2>>%gC(f z+K==#?F|&alP3Sb6w!|24O`=}%_5%DS0Vk`+HxWU)-$Dn_;76pZ!O1WujN~9ii%E( zUrshc9tA^B{&H(rRRsoYCa!6N)RgpO&H>NXw`!luXE^r!>ZXr{f2q;FsfknX+BD&G z)@|R`G{RPPc(UVdq4Knz*?*kNZtYFIYD#rTe@D8u=&oi^`SnSP=R|4CSp*-dZyL8_ zzmyaGO28$*zaxH-r=kyV`JY6aUlRsZiaoNwR)=SO2sQM1tzk)nhJI!v zeguKX5*v4m@EeK2+GX^yHK>Ob&Q*V8gT=Q+TN@#^=mGs2b8xo_pqoKMo9B#?q7_~} zFe-54M|8lc(8=8mvl316bsak;RA0Sa7j5xF<2nHNwts`VC5&h>{s#Jjke_dz=sReh z(+^GwTvCn1#Eg$n$g_+IX8Zmv^9#Sv zF~Tqcfs7L=oPu{JY<#BwHcEI&hcw`wz=_QO7MEVBl3G>UD^G83j9(8 zTZJ31=u1H1^4wMX1e5#IY51X$739$*tb?`M91ho#-Ny~LL=3-5U1#m7O=I4+Eb+UN zCQ+Jt-K-1b0jxZ6Q!7@?XS)BC>lqpS+SR8*s1L4hA!EA6uI8SlVZ$V+_TKI!O$|SH z)?ZHJ4#x4-CyRnKYgrne^_25c4GfCu5CAUOC=~|EOBxFCt1FqqcNbsYz^AfO&>XC1 z1|p%{vZg0wDJrip6Rw#|dit^vLtolucEWA2S_LarLU6H*2U zmip*+f5F$g;)E-9s9ZV|0&NTx>okJaKLlMKLv@ZOV@+*#O0-ri0IGpp^wVkH=j0a5 z8s<=15q^(_K8_ax0_HO9mqWnxPF9DcZU(tn4&bdZ-y4?LV}+w%L&v*| z@wc6qCwKP{KACGg`?$ci6H?DP3#|9XyAJ4b4!6Hpgu6gjXdMuRFh50asr|I)cHHP{ zjt+39JRs;?F7CBbZ~EeM*tlEp-;#U;V*(pa)SLaB>u!ua3>n+yUG#%Z1g6O$fLP^U zv&afuq5b5P;2_%fiWkL2xrN{#$<*_(_M58Cr=k6oybuM9GLSRN_QhUSG*n{RR0SF6 z5(kxA^1GF6autD~4KYYmenh}nmxhsEw>O@}(aF(}I93Nh*C5mX<|g#5?~9VJl&?u{ zu2{$`Y;#$7imdHJ=W16zL2nk0t@9=mU4p?}qiC0|Z~LgFlk_umh{|B$o6|XdVQ03J z)9Qzf8oUjZaQ@-rnqt;Z|IaT2i?^v_K#Ey(b~|$JC%#L1*6hzP;!qpj+^Z z)?jQ-h}`!!bmD!Me{&uwF>Jc#0IU7%E_1^ov2%*3Db=o;aU9Vr;gG(boNvv7 zhC8e28BVRsI`O16Rn0k9M=1a_I>nRmGi7K*u>;W`W*E*4^TOAohN2&mzcBe!u*E|o;#P?-aFqNA{ViDjNz>ruy6H~)B^u}~weGwth- zhj5N2O3a~<<=b{>X(Y5JDzUPYi|@+FMsWYTBH(t_Y>w5#1Fp-^(IPD$a2}R~nUtM*_laR%wbGpD<$UzY<-M z{F7A(CU31n!%3 zBWb)YGqhZei6l``=A)mHFjC_s7sXTBTG0_v;UW83%{W^h@;W9`cnY_lN{E$`o&c!W zNkxJO7`A3a>TeIQH{K6|Jb@ROGq-I-2j@_-exx~`1Q3tFWr7MPoip9-aPAv2ia6<7 zTZy>|Ns{5XrY&PgZ~g6c{iuiND?N{$!bxmbn*~ycx|(ih>lvML^T%&CSWebg=L~C2 zL|$pa8QW?FkssUoBsR4%EB{!}>(VQIDK%hVeR zal4|2ROxAJUlOfKX_b%Ke)|@LfR=tLY8;nn*ijE|u~FglJldwp6V z|8OguAO<}4Q$@E}qc?T@z|Y`2=BOTkL{qc(7@Cer5ERkHc6ZwFi9CE83OOy(# zVwU6SD%0!68R5O&lEeZ#NE8Ha$|$Id+gh9}jcJrw}8GIYa$$}2vsPMVl61RxG;&X9ATG27dV zkH_h!XoQ|cnl(Wrt-|YP6XnuYbgi?blnU?c^ilYZO6!%mx}X)f8Q~sYagU=Gg?Ger zr`%toMe31>GF}o2NOtd2@lp&4YXcoG)xJmX&R*Kq`>)ET_2;~4B!gL#o0>*{VR_tB zU(uiqi^?{l)|9R=*{A0@` zX7TIxqD^x`p2pLYlYz%gw;71>xBV;2P4Pz~OrqM}%STdmr&UvCs-wzi?~gjIwPG1B z_JYyLU~b7woHh?uVw#s){ePMr784ZUhVA3tc;Sn*Uj-^G|2RB-cVmSW%`jIl(6ZD4 zJj~+6@yFbMwO+EN%20gdkA&(5tV-wtT^lm4cU`Q|yh-j_B4PF28S%|?nLKYAr@zte zEB42pbC1S4oP%>fl1iz6NwjP!(Qf7)(xytN<(oWVkq)6B^Bd*~RAaYT+E2W)F&p#K z28Nn~jdO2;3(OGb(yjtu_mD_>M%|&efUUy!ytEG1{+k^ z^{gN8?jj2|MDE8IAaiU?{gF@r7O14*Cb@rh{wk_F!Voqxto0*Cs5+Y z(%H(QMHE=nQ9z&z<_gDgs5=yH8%>8c`F1{qba=!7M$_t`{vqy~ zYa(5_#T%{E#Z`~ve6KClSVrB7w7@Z3Ust-?$*`81@)Ang^10QKi`@^-z+>N=f(5NQ z4<9R9Dn^$|ZP))AwFPokinN&zr4{vdX_X0$Uwb1hCvTuuBpj=(qh$Gp7tjNL@QBv3 zY(-p6Qt}7dH{ivpI5`ZUm85v=6C`|EjZRN^Vc~rNQrKnZ>BdB3Q|t`LD=$2yHbY~N zA~MPJ(%0pr5GR>1e`qD5|4RiTnau?=s$e_B9WH=1@UZ5Y1iY7;yQ71A6lY54Bfc0S zX6n}-k>Alyx_He&O=KtkG~8Wb77!7gHphrEt+&V~TIM#fqc$c_G_}2TqhDOmsGgRd(AFjS zLOE0ktvA`z0#JlbH+(NSLS(CCQ$vjWJP;!&Eg1AvyYMb3JMsx`>^(MBt)t_3&xk`I zbaRI!-_q(FFk^pYI=7S>N?ZBbjM>C?X*TO;exVF!JK_<#C?8lcyuBz-Knw37kHWbNsImT|1?77a!Ir}CB-5C(3d+k$;rtext%$=qt zo8XvxD7Uq+nkmrnvm?}eqAfx_8X6wuNF=e^DhwI!O(N*@G-=2YOpG+*FnfOwLH&PKjHO+KC24RZ4OkFewJC5=Pr%thA zf;RDM`kRX->cIsGuKMG8ygR-zGIMS2>s^?H#y|*y`pH3MVffu^#K5vAw zLs2^;X?Fo%z)`%>NPvz!K&-*hyl0DxR&+m#4D*Kf{Cl+8$biJ7680r(3JS_!K;L6a z`O~r!A^{6$kUZq_QCj0E|0Z8b>~pNBl4_3n2w+yhY}--{yn9(R!Y2aK0dN}O?IIxN zjb=Fcp<7oV$Yyz0Ab9AeG;(r*h!U(N%3bEm;Hy`xWrDibd(&mB0VX++pRBSi*{UPJ zrq{`kh6YI3!)Kpwuhuf})-I&0t9{_~b{DWg7#lP^HOV8XF1^L;B-jXwK%!9{vgUm{ zEc|IGjE=4%5N?cFe5MDTAnX%_q*%Aw2nSFd6BZht??ioESg8^g~N^w$n?0qnlM+G zT_P|4d*(^+6M?0MjmU|rXX0S z*k5J3SkzANTir{GGs0ORtZ$(UpW{E8wH`J~oY$>*rl5h`%D(BU^Cxuu#5+5T?tw#_ zn#k7m)h6e@>ZYM~kz|_lJ_rCt{>+~vEIJ*g`}Q~ktQ+l*eYUBZ zmQpPi6NRjp;C_n7;pfFGO+nR_@r`b1#&Qa(^*%ncte!#?Y#fzJ_B)uk;LAmlVnpzv zMvPA^0g#+T-Kx=Pgmv>w|6rJmJZ82ldc zh=Khnoojz4AZ@O=S7&=YlLlZJNX82-m%wyTiocU{&O{3fZPqcM6^QFgOcgd&z|_*| z`A&^XMKyBZrC_{q`?`3|lH;f9qMOt+BxL0&qNWZ*nU`F%&UczYJ8)U3F)3JF`jN?QU$nwj~<%X8ewJgRipZ=G`%gv)r{p-Y{{jP4N^1 zTlt-0ptQ965G*DlDY?hEE zCg=5E%`Rr$4yZ{9y=i|V+Qx?e{(R1b(?XcwAGiUHa{7XTgsMSqgKvJCjGMbJ?VQzA z{V%@WI;@SZ`xGIJ%FGiT1;XYIAu3Wmd47l@^o!y9RwOrZ!f&f22>%X!18$3npMWlA1R z6`kd@Ky%oeZ9-=3>!%VwT5=RH^Fs~X((?_KV{t98RZ21)ndJSKfJ^SE0@uOd-OCHB zSokh@w_n2lyN<-RHjkS*H%C7W`}FqoSr7O<#AoGdpy|w`qM;}54Btj2<<|u0WpY&0} ze~4UCPbL%7+>M7mD(36yiXaJ5qAnG!nN55ZjnZ10*j@2YQ{sW)vbX{kQJo4=YyKtZ zn0-Rm>uDodKG&22Cmra&$5KPDGUW?9n8y+h2^YUyHCy%u*UI7iT&DUS8fkRlp67a~ zdG;!4J@4gO((QBoIeW!riDH(t?Xa5t;!MYaEI*+-QJ33^%mj_Z525Eb&=jV>BS+4anlw~N ziD;E0s!?!QsaC=q?dgru%S^`KJTcE|#Eu47Kf1uj<(sywz7j1KW>6|%EtcZ}*0q1W ztoW#a93MEQgBeAL+=VJDTRge+iy>g=c(>;%Cco5aA0-xG5x_{sqA?#Klv|+!PL34Z zQm6{Y?6_%)4VcCnHk#G5L?Q4}LZy8Fw8^!2P6-xGsi_xj&qoEHM`M(BQie%IBsBMweCXFG5e~ zTzJctg0@@cm*}oTda*F+{2tyrJR3Mu@nH6N34pa6do9~Kh_0HD*`Jn+m2JNUaC1E7*gd zuSZna-MoQychnmc&(Qu$s&7?# z8KgQD8h|SKj4;8FmSs?4wjw26M(a(DCx_bc(vj=STeA0-$UVpH-*{%Ly^i&qBg|NI zvxBieSr9#D*uJTpuu3@+3F9^*lz55S*?Fj)`5jQ(-YyzmsWonCX~TukCMs6E3kzRa zA-gY~n+jc#zs=k?k`F%=`2neX;v@9?O`&>rm|xFnZjj3Jd*qBpxSv{vXM~`hV3Vj{ zXKzt0#1~;k#`)doK-(Q$A1^x|v|we9~7ErwRtw7$JTS zjK=esc7XzQ@KDV=aIX%)v6oIg9@)3Ej}YDB}}Z?BI1Whw5{0;+&Y!A z!mP-GG*VA5Yq-<)=Fe7pGtb7PSQOtB)6rE@o%;E5d7q2j8}7T5m6oWJtOrzDKhLuLbD!L&Z|La-sB=urVg;1C5r z`v6_iM|YNnF&{R+$S$dF?#(EFM1``W7E+x#w`e*ThLWirD+Sn1es- z;jsMd9uq^h|0d+HM}-rom-`t4R0nAE51n*T&}X@XmHKHeI*;^6q-yUp6z#XJe0gSn z26Teffoh!XE~*yeKS&guiE3@*gsXRO z!iP3>?j*C^o8W-0{+{=$|d@byxpG;mTs{WWnt;yfwtj0ic&9Zy5AeTz8DDFF! zNe`Q^(n$%iG}7Z4$;BSzGdWPi*gmt`6lb;}7haL1Ew^hUkPS`z>`st~*8k4Wye>?Q zRRACT)o%7*5V+hTMcb z@l6U4_Vi@Tp`tnS1AW?C=OgKMR2WOPHlrn7B^Nd5%?i+e9BZC~T$U>0b?5nt9_JpK zoRAF^d&)hw%kmt3At7^r6lrqIGOl>{1Gk1IUKRTNY z>#N#cM2JF9jk=vH%b`z zXb9wtG*QzNu|+Y$uFG;hdEbn>RVVRWj%!hdlL`w<2A{`&Opku`l<~k~2d|!8UE?jv za0$+9dogjSb`ZWBnY|;3aGe8F#%h&#h4I^T1P_F37lwuLnLk(#?kI>rU_Zi%b;36# zKy+)Q#cjgaB2GHBS4F9M2dj?a0`q+ zn9WJWkuAkgq%kY2cD~~kgn-a>EJbgIDPmfZJQrD8bzgmLe2nq<#p@Zf@*|GgnWN_G zQ&~4j@~dF+C#A<}%7R|nHMXxlQ=Lt1Pkyn=beBQpFeK|1MR3a`My;Y2|6xPLuYnYJ z6Bsldenf|yb-kn6IP*E7NpqQ-U1}H(zwUO{Es-@apOYJv`^V$cZz5+r`+f`Zb!PRO z*!|+LJdxk*e(&6G^d9JcD%Ox@K7sU~cFTnh#OvEDh+;{-;S=_&(6*fRCBY-1V=Jq} z?4m|wqllwff+s>L`HeEuaW0-z%fy{$+B$}-wI6JBLQM{GiJ|AsRav2EM?8w14Yj?A zaI3%v!2^WhB7_eN8mqrfq}9e3%stTd4&Rfvs`0r$tO+movuKRY?mi$YYFw<~zx?_& z;5Uc2Bn|dej>vQvy7`4~)b_PT{bS3$T}{xRmW)3nC0kyEv%mLMrEO_K1 zfSr~}2K5Y`9p~GA|EuZRpdvnA-dTaZ8($+2@L|`FD27I`*SqcE!;S=RD+PI~ zI-jIO)p2u%^@S^TE#45|DN+pi?1CQ!XZBwA@0LBS+8@g|ursD|t5O0a-$=yx9AC8z zqZ8T4i4evjzcn}@>HpkrwdwkMvHNpB1$hzHE{hkV<%2?8v=`+&x}atQnXDvIHgmP| z&gUIBP(Qp>!sNnwkK3xOYMKR{r!M@FEJRtdbuhikU$X2GH%N2Se7V!SAKAzIT%3W* z3_jiNSTg+TBYQ=KB^d=VyXvP@i##$sSvy8}eFTer-^ZqEP_zUj)a72G*|}U;Th|76 zbUyVeAMLtYN`08um(eLOanL3hO!^Q_lbfwXq@U*S`fi%V=g;*uv@hlxQ0dosACRyN zLC-wSNcBC!&Wk<}#0Ce4y_c=c&4F^?Y4fE18@`hwzv^|AqY3xD zL|#F2gWuZ6#1IS)<2<7y8v)g}q>1H#7?1YEETr@~lX>(kYJ-N82vD@i`7`_epvQhR=?$q@6+7b}jJs0smYZVVx=3Os3It}VmuBpx#08Q{y8VE+ zBZYruEaHQA67gm?xNhmO3UtD|jOdT?)Zn6Ne8_OhR%9sbMG~R%#uIdrZ1RGtnCHh> z&d!(2Kj@=;;5`x|$n_>}ix@7bpk+h4JP9u^KW%uX0|cvU=o@DFEeOK$NaSR zq0Fvt>Nh4RU-OBg1W&ix>&bwC?;OPQ z*Jw8tRUQK!!Ec6aSJ$ZCCdJtf>Zs^ibKB-KGL~-zk8ak{M4U*zMO_-hV=T9lsP9F5 z#+3O9Nk}XGx&ygrhd*(4;r`1dg(GGzIG?0R#T3VIpQ zSkHRLx1U&vJ7irrS2l{2;rk=;eGT}X;Q2b`_k5s!H#pK*Y2fyXx~r=KoXizG19(NT zM?jD~K<@9$DfxOm)}qao2*|YFmgC8RDYBd!bii;(^PFl2-J{xTuZ^c@#}gqk@^Mp6 zzk-2-wqJ$R>2z@7e5+DOkH>T}7 zs#s&<`Lf_|2+s=p3aXQb0duM&uzqG843$^>x;*GVYJGp?FZE(H{&ZcMsN7@D?K5d; zjl}zs$izZPdZyCc{CJFzFf?*f{V$O1U&EAcEtT1lCb9f*&ZC}-cv)FZi7_@Y-}N>S z_D~N7D630(qJFwi)8CRUu&KE`$q{ zUvv1j@_|T^1T60DEx5;qW#1eYnfclJj_j(ts#iv4OstQ~ z?HcAdglI1=w7ybB3ENy=UUlNra8A5L{2xu?2Jy)8uZ@kLWb+)yb0j!Dwr`{*Te{%y zlEH-{jLZAoRK>G->(1!s<&nChCqp9~6;f7_G%t%#lewMae7YIgj=ip*YA6|eHNsK2 zh4G@!{Az0Ka(O1f3Ifn+lrAAfl4M;8E=U3Zahww6Ok5;X;yqqRqN#IJ-a+8WUAFBS z0B+%YXlpT&q@5TdbTJ}KA*}{w8sQG1Cn>f#^@Pnk4VwOuw`riF=wQtAa$suPgp`BHOwmU^0%JuF1G! zQIIU(gIYA7bI3eX)fhV6%Z77`M%GGSB;I zY`sg`5(KbJUE}o~AgS-MhCByd(=XTzIgIK(`-?GdG53euGJqwU=430M1CreQj-#ZfgnOwE6lw)GhFN8|6dN z>4;TXw5JP}NR;x%y3k4dR55rhqD+E4w2_$dxb`w6nka(}Z zp1Tfx%rQ#h)Q*&@BFx7PTy|HPuZC#g;fxAFt_0>T57?Ysw$ZE_XCk4WqF>1GU%ZBRBfW{4}zm2?}PKc~dMwv_=E-Hq*CV4||K(V7keznJt4(|SV zS=I=nr%{55NOZDzQ5CplKII9E!!|`jGO&4_+IJsnt&u8sxad81u0l7!Tw2@s57S)B zN(fFCVwstZA$mT`-bBVcYp`v^6j!h1QzNHf+OyqtD%AMp>lH8ehYO9}K*js-bMs8t zbJ;=#!JMO)`pfQO`e7Z1cM^+NOJ$~9=Hf;b`Ayb z%tqln_deOzA9XCt6B7XaqXhsFg%{Huc#quOf+^BgRX#nnhB&CRGJM3TIU1OTXdNYSN-GL>xC}{0waU<9W9(NJ5eL zEiffVB@k*p-<9Qbrs;d(s#-1I<9V)B3V5U*KNWM69blec*DNM^<7pGn&UC>P1040P zG$ck{MRl}^rlHJyA*7|`e_=Ctzs`ab;V{G`f2gXW%#zej5XJk{XElJeu+Y4`=+QwL zCxt&=4hfi?3)Ej6P7B@SvCH-xx*zsssX_&p+l2|tD~{hD6J;3bLNaVug3;GpnOT6M zXH1$g(r**uTJWmx&p%#;4K3Jor$ETiU+T&sAZHXfZHFl~Sbt*IIJ*j0j<=`@J8Mib zO}gCpUwEg!B3LJ%KS*=&Ra?ke>=j{ZB%LyP{{U%G37x-fJQ=5TRj)WT=q)B16Iqd@ zDYsj)cE9o~nE*~wRDR$E^(O49s(zA%_qR?g%_@+VOS>ZQ?ZEdc) z#w_~evCo-V?}7mza*FSKD;_zF_Dxk6N z>C38vBXdn(55_9bSjf?&eU56Lc4Y|(dby62&*XGA!BhC9vc-*!XnSMep3(7(>H;A_ zyoWvV#krO(XGpoA;7~22!)2I7wav8Nrn`3x+`{YNRZ={DS)-bka%Y4w-Xw>TioVsh z`ElVg1y$Mb+iQO6v@BQM%T2W&1bU6tQtm}HV(u4zHwIo<7#c?h^R_Aoy1T0li{N*l zP9^)d2?q!QQH7x>yM3LZeFD0Ce#M*py|k^h#fOL=agA`6&daT?xfhklSXzczb%vQmym* zNis8Op?HZ6U z^cH$1;@`O8&I~2927GqKnxP^AH4s|4@!R_!n~KhNEZf@i353crP{}g0jE+%Nw=8SNO-(jn^_q*uM;RIJ1poGVmdxm;*U+NS)!8^Y{GsAJFiKSh|YRx(;O zSk?Yo_j@)!tmG^~Dbe{|Nkxnhz^#%wqyodhz!j#${`_Jw#eThuUg|W_0T%>?&O8GT zjh@e9Puyr|aE`5dZ5Nb*fg=}5IANjac^%c7^d_hE)!>U}qYw`#YkUA*Nk$w0yyLj> z15vD)fCde=iWyU1HukMsSkLB?$2!Jfa+2f+`;3XHIT;CEeHpz5%i55|HCp)!uPIZn zC@0LHq`U?mm)3C_XFMU@Ci;3e4W)qbMV!2(ns6tK>-?8J8>wJ3uOSoY8CR)!X_%-J zV&){l5&i=~AUH7pPsil%-MWD)=E*AIjSc_KhY&RHCm5~htn78h-TV)|bTMZI!z!n! zJ`Vi;UKsc_h@#*8EcP~JoDxYT69UkMUTT9K-Zr;52V$ji)w0uZIdtDaE6plg7q;GS zy6U#Dv;_9!vZNm}O5-BEv;WxA%A5!X%3Y=L2h)sLOw;)3{CIi6JFocogAE`*0zSfz z*$HEm7P0j*LZYLsOV@#}N*o@Kn(&N2sLnf%wTv88qvi4yjCDJFb zZJi0b^t@a0vrflQo%j8Rxx4*G6$GdXskii*Ls+kzmT{$dlu_k4By8IDW>3RnoJIR) zVMrNCk%X_)Hsfq8u)c-m@jUNBh8-HGlb=GnPVBKwyh)P=_7UopBQ-p5fmL)r1~lC+ zZeMgux_gXzU5!IP`kD8SuCoONZ}+E0t0#+6#sp-7j7FVYMZYz6ZOqNV0)=L9(tddo z@ywavZ$$7-=hRDqJue+xfWQe^S8bYA!4X*p^4r&mqd z0FY&UV{M$vSK#nKE15m7*(SGha>a?C-1ZO2?!hN>*sBdMok<_p zvKYA^OR)Ar3ViFo@qG^(lIRf7B`=+e^xATS*S8;K!rjEbdasUAt2MveYdIUCJfClQ z2q19#oDL@cb}vQh88kQdAijO`Amb>7sY(S0znif5v{>+YjH{(4I(S{5Un^S=RwyS=FLsUo&b z$!Wf2*07?Enum+OI`n}=?O^U9{-fQ2KZ)3NT5O=zmD9gs3V9V|e1oSCX|Tn`Lvy%FMs_zW@g1a@c}Nv6!nd|tz5c2 z_iYO4n>r$X*2YG9`XSJ<(goxeYl^%@#Ul`vHpFY6hg>l6o_7(d0yND*wBBRJ+OEgV zZ=&ODL4sp^YxEG`Tc?eZy5gfL?Ia5J)Xs0Q1q>S6pyJ7hgyu-2yq~X^?{JA-H%XoB zm-Y6JcB)M;O!GcsTF9j=c@Lqx{igrlz>r3InMjr7QU)=l_ma7R%uL=ZB3MJD5(7;a z>ID7RZ{S%ZuQOa9g4y*3R^ESQsXg#5J>z@wy1?=+%kQnN>Hq`e{8Dql;G3p7du&;O zh1GOK(1|YW@T}38{QA!0Cn7#OsY{|-0pQHZxyp-N(5yXX!>19e%LPj8JD6mc_nDKYcF66qZ*F?c*G z!^tq>1L{Tbs4LNw=y#?5u^Pdc`r}&Y@$4}Me9MCyQg_-%9$eu`>Mt=7LE!BZg!4kv3(|3ab%g~b}DJM_Ql78BvT|x-Nnc!^IY7Lr`k`nX?ru5X)Js*82ky1% zsYHPd2^>m2X@n&uHhYG#i~vUT_hmKxZXO;;%@F(Onj-XNU}Vj){2?wB zVRR7Pm~a-8E?z=sP1CtVba=75dp-*qy;-%%2JU-A&xN=)pCu zZ+anE?~*rp%f&uDp(N);92$2obFFIya=h_;xEuVUskon(`WR}`)a7NBU-+r?Vyx_x za%G5OTAer);_2#;>WMbZTD|XZP=D$Pkcy5>TEAq*eZ_fdHr3h2b5s26rV4j;#UPV{ z3$%1Q$?`&FqI%%B9$AMn@l%8>U*Va)n%umKyqQa2t_2Oz6W{$GUl*1OO3Sv7BR(+l zM!r1}EOtThx@%^P?ly4-pZritoe;XdeP`oYmYbUq(ReMWQeT|dwN3^+VzcmiTMFdK zEAu)Qnx*PoL%V8ud>}0pN}XWoPnbTTf^5~QqoAT?yd=p((E(;wG$dHwRN=SZP@3GC zlHCFGQh#klUA928Xp1}3Zqm}CxEuSg%R{=C9E9G7biaMJSX8IJ^6UEqpz8(OQ~rk- z8RThJ_9Kegkge0Sw4_Y?1~c{cGz}dysYf$n-klGcdiEa#;F5jcFRDpz&)caY_TW%* z)JsOWhW2Xkve!P_U#-+u|0fJjbaeQ7gj%;L-|FmoOfF5yz}x;Tb>FcfP&u$?pW;p&_N22dO?THl*LaOH8npk1y{hB_@%J*)bqZ__zAFUd(X|dieKX({dB#X zF3^!Q_4@q`hBBNjdKt}zqytj7{go*$268&&UJ_Yxr+&|1@nKPDyPp|M>l2ySkkWs@ z_Jr$k%6>MeSLEtJgNrWA_)og|mzKu(RBI@wd3>uYuT1KNB?#i1IS_ukvV@uBCHE z4*+wsyRD6F?CFY=F^itAVx@^W9ffd4{E_p+NXPu+K#+@EX-Q9{bUNG2{wJaX%GqWYZc)TFxIiEC zDUbMe`)NTs+V9JOs~xP3F|sI%A@;UYV_a^_9QAsirn>xEzdOTtc*AFv#_N-M)Nsg& zf>z#w9;2gafRhqxCu<5> zx5<3fj}_RT@z(OeS_Ao zFJ#FCEcd=}X{|JFk7!jbu*XL4{fg>KUF*Cv+s z2On~IS-GsKT25-|WuLqJ8Y#K00mu1*1?`uFWN;Zn_-77YLFAHvhZ%$5t5=1&VH0n( zTRiEKhma)D1O8UO;jg7q@H3a~q+V=pPJQyWT}*F$=WJ{&*}Chgbgs0MMe_rnGm0ef zapvoy9W^f=Q=ECFBr6#|sZ3-3ziN@K)KB4Ea6h$N;oC&Pgp}@x9f;soziPM*PW`tejzM=H=T^=L_tV0`h=0U5vShm!2=9@p;Jc+2U#wmygQKlcc}x?tdtC@gy%B%Kch>g=op|JJ&zb8%)A3X2R6$;3g$9h;+; zE7_g-tx5hhXm~SLL-*s21a*x5r_9Ev-@3aJ=FCy4$sczMdhNJ1*iWAQnV`QBm1=B^ z(}t!!%lU%ZTKVRFp8nrIx8^G7rO!S@wgVlb{@2T(oxAZ+i2N^zA+yI)F9H4Nr-aTL zj1;+fqf&=>=g-x78tNJh&OYYSJ+HkA&oK3v8M5#pJZtN1IG5iAWt>P5ze_)9p%p_6 zTANN5ws&i=WTR$2O02tAW}jIc7b}hPJvFg!#AA8BPTxpKxHEQ~mR-9hsxfQ#>YpF^ zTQk>pw#8%UTz3y>%OUXn+288v-`69|mpTno?CYs(@_*yT@9KH`Rr>xliG!F5_RS4U znMg>3Sg2KCd;gemg!0d_1wf3;Suk`L{cp7SLk58lxFoi3M|p7Ht|u8>$JM;(duWo? za}ZJqTp90^w}@|zxPkDK9K~ewMPLp2kxM6tMW@O?FhJSibEdP;d%_531zAN}A)%z_ z_k73AQr(KF*&lRo=G2PJd)op?xKz|s#57<5zMotz)u+B@az_%rO&4a-*?N4KH7kH4 z4oyu>J5g|$JSIE)sVqQ0uqeSOAtKA@*cexmn-8nU`NNN$kqXAkK*{tJ;vgt9;1gPm zyeH*Qk`QV_J|-R^-^TK2#>GHGbVEm?_F3cL5aa1nI(5XHt0j+Pgom`_ zXV|o}R7rzJm8hPn(+nS3p&vbljl+9^-fVX1_gk_JMRe)z-SS*H)oemx1SVNsVZp8` zo^8mpXf@25^&^7&S5h^>MJ&ziSp3UdZB|CzkyraF4OeMCH`FC7Hy`nOh}oUYjlqNy zMP>nVu^~-Mgo>ua3F~6p+Moj;XG545lWQp|?hCZHKj}csSow#d7tIlTLxpN3i55P* z*(~MW?JsUkH>8ueEq?CGCb*V3nd2X)ivR6^ z8_h_qIvyYRUM-!-%u9BJ!7!_Fb$pK0u|MO`0 znAw2SWUg8|?HgK#77bLKH}6Kgr$$%eBE8SO5N`ag3z)0&V*Bim-id@-58X=g%ZtjO zZ@2P!ldI8Rc8fK;V$DA)fxv6Z(sG$dyjW92>^5UKt2y_;QPg&u=6Nb%oHrx@9jInm zia%zayIzS_{0RIYJ*8Uk;0WCQ$^+e=?`?M9+&g@^?0N;zsr77FCTa#gZd|f3916Rq z8FvOiUT#SFem-ikNLmb}RuhVxJ{KtzulTH9F6`?XO}W0O=b3{;r#~8pcE`#ZLHmCl zn*aTPupWw&mFLe3^HjGN4tf2G)Tk+Vb~kh;C6_M!6)O##P0pdXQ7a94ipl&u<0}oK zrG%Wes=(zp$;?LxCJl>2e7Dv~+$JG7j22|VPYJl1o>|>XH6ZBpapKo+7y1yx|YOC7V zftE0v6C5npQfVAhY41K<3(8cNg-IFc<8kk#`xRfOo-{BkgihDixA4Gn*H}Cj?`O^9Xyj9ts2|K0m zdK%Cx}dpB&*fZfo|bxq6*+Tn{E%+B zyEOLkNZa7jhlBLVZ(f#m$vJhVuI(h@2mnQ=!{2g$)d=R=#v)}ix zf!-ghpa1s@S@(0xxFV~(+U5@Sf#O5E06*-_+;47bySi_Sn#_3!_@@=B&8G$=!{4r7 zNr+5YF#nhL{HW~Imw#H2H2N$gpU!yL23 zJdh%^?CprtIVl^d>Zl0yL`9={tMn zTuu_>1ytg+m8uk}_icZrcV^M8?x`k58Q7m8J)p1IrwA9fxe396wB#0P*F{7n7Pc^K zlsMKIfn6T%%g35oR15eQa2waLl9v-T`g@A$Cdy2nmni$aa2eXoGvnq$sfTLd@$TGt zt)~E+BDZebca*FaE8**eF1MzCz*Sbd$Kynq4i_uaoj6eNC4B-Oi{1j7%Q0I`LY0|E zl|f>cjeY7bkBn@O<1W4IbioRy%b6~SqQg4T@ZJp`c(6%wMHp|y#}S87@seD2{1b8- zC?C+Bxa6G1biG2X^TE2uSE2vmxAXJ=zUus+-fPvv;x6_E@)cY>rz>HK9dV;~D~M5# zGtE2yG%ur`8?q^KqPs|=kedA7we(5Fw`viqx%?;`v;cJu@39y<&Nc9{bpq;I7w2X`Fp?5tx*JBpdv_#cXP_!A5YODLa5%5SwT)l4<2_TjzZy7?8AxoJcDaw|Jo+_G zZ6RRD<6_6K#IeA#xwa$&bph#@V~NVIEwY%BX-gD3A0G1`Lcq33Fo*n8Yqd>5Sbi3( zYHxTA|NTa^SL@wTMPA)=66jo z=k{*%@1c5r;mk}8CKFk`L1MVcz~L`Afxvig2&AjSZgEqZ^R*ZNm3f zU;6ZuhSH(=y$1-DB}!V6VrFPa8+4NQmY-oUzkuoPT3{Xmi$t|)^Ky;&KTVbY%X`Ke z>AJxoQ{c_C5{Ym$IwsjpNc3iQHO6LkOhI3N+`99WRfFu%W`Atx!8oHvg{E4;dW~I; zqx_e*DsR1Zf-o;v&C)twrxdl&llok(M_xwlXZ3CH%yvbPemQPP|6;wPyRXEtI}OO$ zPS148yr8GL+tc}SgV8__=&nrxyUZfu$$D3P(N}TaSgiM(saE%et^x~cbQ&r$kipi|%>An-A zrGo6#>If&wNOxJ6)xcq9T@=jL%|ERi}dXU6Jt7j=uM)8rbh) z_H8MkZb;RahX}V!J11naOlLY$<-$R2?!kK^-@iz?^#;n)66QbdTuun?+k36sc|^=? z&i^~(!_Xx+Ejd0ISAh2aULEOLh2R6GUvI8#Obh_@XkacQXw_?r#HKF09}DQT7>XJ5 zs4-~N(hPmdfbFuG!QR)>&@bp_iX-!7I&MoZ&(NZbGoQfNRn3hF#|0q79)F z@J_G@aVeZg$i<14Lf2bz0?!1{)` zH{Ude&8xcUd7l*?M@wV0_(4Yn081>_BBwYm0bOWutwx>QYw_UN^HiR1$B}A3Pp0yWq3ASZ^G?oGV*7mZ4w0 zF1A%{+;&w+^U+$gg;ACZp&~vC$Xu>kf^Sc9TJRddqNXKZ_f_^6pk6P%WLgw$){7Go z2HTwe<4Pw?wv7Ddoi^hxC-DY^?|SjR<^Ah_Q!6ror(UVDQz*Fb*bX$&)1;YYjfXoO z{|zEU`3Nw^*kT+e)q&c>AZGFSVxjSWay8#O3Z*J}n1g zsuE?c_L>dAlT`b3JdxoiaI@eTRX7F3W9)&a#RxMf<dx zUIXGp_Gg>H&POvJTz9XaCUwLjY4H4QeQ}@aC5cswc0=F!*@(UJ_(Ss=eNdUhsu>5d z#&a-E$WSk*eczj;0}_$(&X=+DYAAYhib9Xp(W2wS0??0EYi$oA`!l5}=+}`1Y+aFK zrz+(Q4(4w~^k3ufepyFY$V;{p-IysNxXjBiDnotc1nJ>07qaAA2*4Dv+7f**0~@$0 z;oZ4TYitj5Ys3QNS0BIlT>tta_g>L8-?1VaOtvzhqFCd8)D~(tZa*ewXwf|UZEEWI zk+Y$2p^ey18{!}~#7e}pUul`l6=Jw9sq%zxW_n<2(Ar%C=sA(gL&>Bg{|jns(S&z5 zKzA2?#FIL{t&fr+xUMRRN*$q)ak=2z1pXu z)Hdsv&b^KQbK3cS6_lg4ij`HXlCXkjzHR?zzGV8yF$~J15w@O{&c*Xgn}?rpwF0aSAz!XUVtaLwrSU)7E#xeYIR5K}E6`c+RTx-h5rs1rqsX@~Nic+kS+xAwcO63X1 z;f+B!*6Jsm3!VW^8UflFfSQ0%yg=H#d)*8zwQMx)iqDFK1}ftDH!JGGI`Uqsv%LAAi08Rdrdc&&kJIUbJk8%Zm7YqV;MI;R z@Q)&O`=K1K(|kNJ+%kZil;@*DEtdU88gg-f-D=%mW_qHcN>MWM2>D5&`(>PqYA-d$ zIeSfm!A}hBKh`(`JTcI4Ex~?MnQ(^Sh}(O;jN3H8iEg)+Jk&`DwzJ6c`9W=RVM)MU zoY+H+QE2qA<=e>o6_G~@%j$}oE1y$p>P(S_DuzzRfu3!H+LxE96stp?RNTVanrXPcY2Cu9%l}zPQ#}s_QSzY7M9nVTnIO6!ldcv zBC@3Q-yTgnZHokhY?iPE6))&pWTl)q1ma%9`r=-b81QjE*n@t|Y(feCSx0(U5ON4O zWg6)M`r>F^Q&oe4|4BoI9sVpZeYi!!C;V?Q{=0rq{aITkOJ@96+*rwS|ARRQLonZI zRaXJp!yJGW1^*46*eQmDSzpO&I$uySM4rFIFZGD;Hd9ZaNp zjN*!?CE^7I6pAralawlzKzM4_x- zNO74V>{OyeeVNk_LaH*xFGTUdyvhx?kyv~PyaEDtl4q;ksscc>>E8GGP%h25SM8HC z8f9A81q3AL$lqb>qxm9!D<-9{j2_?L;}%}0&7!mek(8ECB#;vZlSuRO93!^7Za}0q zGc<^2eqSH-V(omyoXM3cbM4$Kzq&WM6Vd zsmqbxq?gHK!~Ex#3X?Smo2KQ_$L2?QKK&^1Z>k~v?gtz^jN{;=Nd_VRdzZa2lmei7 zOhXmmh(}5fEw8TOdFvLTE|ofrOB=+~?Kz{m!$$Z@>Fo-_QMzD_6)` zYpyx(IqosWea{;8vg|_lv#TwbjIxlX0B2&=wL~e%t7LN@syq4wuWfkKg8$Y^3mTdm z%au1Ju3M%VSb>b@JAsa5jXf)#ptr_JTrhXJpy1Mo0wb9B9P*`J!yeTO?|vz4^bM+f z*fm0;`xWc+s(Cs+Ggi4rvBJ#_d+F>mtkszjrobuN0NVlU=z8`$Q8Mo}N?1p74@Nt1 zaYX_*&xuH)9X#x+SlN~!DJ(XfhUS9RM1ih*fB*JpPu4@N_JXMM)|E1-JF|_Zg9^_S zRjX?Y>adpZAU4KKp>4eyaHavZ5I!Yj^v^$ z@|X|oYD+6mNo0zEf{J|m2Rk&9S!Wcx)8gwIgIAa84mkPG%`aO$`kkKM^X0|`xj+0; z-yTc(4UL}dhR0}U^-B%DANQBgU;BMj?`wwdX3Z#v2w;jnMVG7z?SuO;&y+p zc(NO#gk$p!^xlGvA$Ci5T^2W&H--`{?_HF<1g+E+b?W;ROx{7cvM4g<&p-&~^Ur#^ z)H_9Ah^_5Tg9TA$kImQ(6HlR1Zy@PjE>WltgSkdJdT`0fvAcIhij55PVLt@ajALzL zmJ36X+V3wNQC@H85LGgbp6}26))@Gh7$7QrQT? zBF9JyG$SCra)L4zm@?wgoIh9pcBv6)%yKI=_X2t~%}i)T(B+|SY^;j!Esve)HTe?d zy&I6-@UAD^-cbI;eD3~EAILtqdYiVSNn!BZV*nJpPz!CGsa4l1T(^x$b34g*e;*;e zQK3E>C0#kmGpPN%oO@Jr^g?n)k&c0QR$rR#ugJEGYer3N&dX4u!1m4cfvP8F$@sV0 zAM^aWd42`TE(dY=EA1?>j2WrJ85Ra#YeO2$>4JW(%`;luk;f(M+fkLJ$NNcUQKv+E zEGl>%CDc;1?>zx`fX#0ZD$Pgh?W09V35U+!pX>anLUIHJ6-k+wU79>vIK1U6EG9xh zh4B<h+5cSIYf-|kYmZcdSo zBqo@-enhfLct|l{#*TOKNQZTT*~(QfE{@4>qIsa8QLyIjHDagG@=Sy%gm#!g1~K%_ z)ci9F(yq72KT38sJ1+evwbkkO7&mbrQ!I3UmT*O?@_zZKu$`c4=TVWAf1FJ}zDuY} zGq8@5V|V}+uS|qp2uekrVE<#aj;jml&nj;&zDH@`hP5^^FFOr{GC}`XnYLQ$!g#sd zk&#nI&!0!5%UU}!oWt7`B;8Z5v>nS=4=Vr%88pDtryZBf!KES!j<)ZzA1xo}O&tW7 zKHewbZW|JIeIc`l4LzhH>w439xhLz5N$4jx+q+46hQ+2DjOoA-qJhKrqID5@W9K46 zq56h6k`1I%-f+%zg1Iu?+UQ%nwBclJ&njw*Bgwey&--)>U5A{I*}sEU48m2o+%R?o zH3)SAS@z0SUY>;i;Q5%V;5taJh$ld>_%a`eYs#1U0m22^UdOD6lJTm=ASbCKIdsxb z$0K; zy#}}G45Y5Ufo9w}T1x3CFOYtp>iE!s`~>QYWqVF-KZKbj3Iy)~aA%En6`_tdZ9%)v zRtGc-$OcNhsgf^JY2g*u+s%^19F^qrE2B&8hqJ3oT~@cNRsY1>Oue#AuKQ=^U>95> zMKb;wl&gFE7ejC-=Msv96l158vd!m|2kuUIg?+)Nt`zG)TT|?<;lIB) zci{aM(_vIXo{L;tf4Qm9DPt;cbiQHttaazy6_e^exI5y!bh3mzPFjX_FtXy&bzT~| zT%L!OE$2{b97R|2uQ~SX^G4}GTg?mO5%s9B)7cVlQSGS;ox>HbqK`Hj(#-o8h^XGj zq9w~2{-yR6K|k&j_J*0W5NmHzkR7+O^|8aRqD80xZ7I)PCs^JpU4wtV6QpKx)>cZZ z-#9covL_RISlNVcKzPlh$UVK^BYnou>2=-#9Ly8?r0#jkvcAWJ++JJlM)AQ8DvV)XJzVU@~vH4~9_)l#zhKwCP zXT#y_6xgD>@Hj-yn|o2gZOl0;g(?8#8A%9xsPePxap9?tx(mIZBnTGVWRQEKk3S6hS(m+8m zly;+qeigv3l9CqDCmsypjLLae-FUotATU`I>RM6$sKP zvugk10upw(Vv+bG^4bNNw}17ocY%KGQq6QOZoXqBlVRGXL)Jq1Pa;kkWk367u5ZN9 zs9WVzB3rK-brwZLAu6iT*+Z?-4_!zx1t!pA?0-CfmD+Lr^vMt&p^|#;DC6$>#(ckt zx3tO19wV&rElIPQkT{vgEfVkZt>Yd-Dz7(|a86!$^r8Pt^~zUGXx*df_>J=Tg}T*O z(%(^zUA!ce&IeT$>(AQb%SpE>sFHl`W=BDBof4|_XEQRlkMmY3e*&yInUA=uX7b!ebB-Is4bqDOc-G zDL!-=JSsfNpETy3lja%Oh*3yPO}IlXz5N3e#LOcf)Z8J6Day8L1aD1C8g0J7D#8Kg zTD~3ZRrr&C9PadLvYA^^lZd{QNbZ5!{*@aiX*R%8UY(4}j{MCp4rW?_6@3H*W2>1S-$vVVu7s0K|`c2?-{h_}Z#vj7Fr(-H`%F7*Bjs;3ga#46w z=xA!}(UIyh9^UXKqn87yV@5j1ln2+&6qDXLZ3>nKeY zHOLNChK7jWx@uV7SE{ShA1N$(+$1SvW)<4#p|WPKVZz-fCh^<7pKKHz-t6FHbputmLZf~%UN?i8RNVP-i$pgAhsZSs) zjN6n`NbAnjftm+`D(wjc)#xi2i7kzH4S;SHl7Jvi5pAf4A)+eJuvgW=6-aL(tWb%7 zMBIY)esJ;N)yE5c_}ap&h4aG`HlQ^Kw^xdI*wps!n-yl2xmT+@o^p4T*G_a*OZ0%q zwSh&=W#PPTFps#F2A#MTpE0R};EMu!u?%y{PL{CwINk8DTfpMYS}nj7u$T+yXFUFM z6_7**uZ|e%K!r1BKhmc+2;Or&6D=oG7iy0q)Eq5Dkzos~25l{?GHO&|Qpgv)#3zAg z7XXJZk)RQ=p~XoVos{+_{g8MgStPctaQ-p!+kZn(48HK`(VG&@))SYV-aF2kpS)=I zVPm?hI>sBr(^-Br%*iGhrJvwi>95;2-=AlbP`xW(LPzqmSlL_JKzM<9*4iqR2o^$n zq5D0B24!$dw?SVI6#DOat+P)W1 zMq^bK9lKEhVu$v%&V070eZ+1}X!CP=1ihRw1?#_%kSJ{dSVp9_xnLiE994a#NW&)H zrf30#&{64(O1>2krQ!~N@5}&oA&8q{5Xjh=SRl^L)u;}GU(#t@bZqX~krGeLGga!s z97daYfs#g-tZ(9TZZ-O*U=Ahsr^i%+P z=Ck9R$f%|?(>+&G@cBOznE&c8_i^4!cRpWvb6vCbcv#e7af8S94~8z-CDxmKqsN8+ z>IE=vyZVFjoTs{rM&I(^Q|^MjZk2~NDu`_2MboGi{m{6g0II3M`jWOQJ8$1$2(_;z zYpIZ{q#xR>lp?4lR2uqM!EI{s#H1DU13$O`9?iFau_TNgtCyy5eQlB6)$HuQ<&V{M z{!rk>LP=^L+mK$kJ+2tDJj~=5Sgd7RFO! z4^gs(k;`+k;7tB@-A8aq#5D50CVYnTAZo>kIY8zMF7`M|d zdj}@xBeuJ_uQLNP!=;-KSnhpS8TfVhS1kM${r)op%m4oDOb++^-G&>6!CopoEh0i5 zVK3tPN7IH?k0_JVz6>R7Oa67mFBMstBz0$3FdO`0+@B_D0>5Oq3o2Y8o z@nAD(+P9%Pxwto<4R)(kz3rCz0#TDG5`Td?9I#tugy;Tm#^}Gk_z%5cD1qaOJ0?qB zC1#;!hKPpT9pH(YqP9P%oh@fWC_^_NBntgw*fyZM;a?0kM|irR>t0~>L7tZUit!3X z?C`abkX_hUOsl-B*jJ{7uN>>T2;;x=>)*uW|LdSP$A6(q>tnLhQ=jMlRSm*A>mRi3RP~fu&>5^)wM}xaQ`h zW4C_MJpabargIxQm+04ps5)cDw}Sijr-JTlCBtzmW ze>vR6Qx~EcxtYRl=l%v;Jdh4L5!h+*-+0I8c89gc$w@7~;&D`J2Ct4l4R+LptO);m zdNzl1t;SzjHUsAbtOr&3@C;m>3wQ&+iI`KUTkW zHtaj!@k@tK(xO;zl1Y!2lofnThVt6ScynN_d4jTk5B3A#5e#DM(=C2xp_nrPOXhbt?rd2@zE?Zk zoZo=vzY47{)GPKLy9DXg@7#9RYaR)2k6Z69Q(*@=>^v|5F4$`5efQU$J`wEV=?w{R z7qWGud%g|0JsaTE+kN|NeX(XWY2=r&0>koc=pjgMZ%DO6{(? z)&IHrKL_OB|LcE7{(olO|HYF1pB48%vF`Uq`#;g<-x-SkpCauBhku*RATu~AwgUe? z;{UxH6T`WAsSTu5bTZy)P+H{XGyV91Zmf?J8Zhg}&)n)#;;<*eso{WvTH2}HD_y%e z36Hm#`W7wk!t_p7bV-sjDfv1-P~fDKEL;4F|CBLcU)dO9TQ~e)+{#~HG%&-m)7euW z`A8Kmr8co@svYmYX40w`q!;>29VQPwia466YV!&j&?tm3Prn`JJf2`_+MS`9$^5|( zHUnr*r0UjZekOCFD%tdt`fl9P_9{HrqVs3csgB}W9nrkv=!1gj`5b5dcWO!YpZUv) zYpY^%tH-Wwag;Gu8u9MpRtn)exIwk%VS;(+{fhB(nvurK$52MXzU*hn{XyM@X>7`L z0{hQS)g2mpC^iJ|MuWX-%VV^X<6m-cJ@N-T#9NJe{;EJDi!JL@5 z%Ld)#hSjZBug(l9!C6e+ z9wiv<^$EZ$DZLWprgn?^-eB?70)|9Fk5UGK-h;kol?Zq}`;u|dr`nBH$%>J%GPEJ) zW057|V-YYzaMK$kmx&BbRefZplA7+~sFH*^!LeT=uv!nqDnTMy6HOp0h{H@!S^KhQ z%L4|OKmPq&{WUA!@*cad=p&=Iy&Wa9$OReNM(^J9D1B8|REah%W#!|4wx}?y!%lXE z{Ful8Ds)9m}RlGCXUsiRIu<^Yc7jlM0F%y#*&t{`b*XR_c8J=F4kLh zJjxGmJ+18q>wBX$tSM}66@5y+EE5nSVwHe$l8p-xvl8#TcCUU?|9pRe6Hw$Djbrl( zOTXTmm@Mn+Jl&Si*+plNk(CeLiIYOO`J8)a+Kq5VQclEA!Qfo|vvlk+O9%8izLz2s)RNvbEzRwp^YxT{g%W zR)nM@FlP6<5U1o?d4-1I*z)J2!;!)=H1TK=wXq84A?rA3n7%ty)Jn4RcNxOwG&dy{RlE??y?MO@V>XC6*%5a@P<;6FgLlqt0|ghS z;-yTDBaU7;M-oCDb+&1Fd!Yc_7LVuw>Mp2(kVYuw{;GZZVwY}8eJSxF}{0e`yFKz?!P$l9G7)wy;JulUFG+wu)^V!(xf72yAqdisKpx;n_dm!o9p zlzaT(0&Ocrs{u=^d^f(1h_37STeMK$7aSy{4EeHUEG!GrH`CcY3@;awGVK!!SbGfm zV)SAsjjKKsuf!1nFX7U%*rrKC2uDOd)!heyawMu?}aSz4^q`rBwb zZI(f4a?*W4tW?UNxlddQ1JE62qTw04)?F~Ds^RrPi{!I@sZauOox4!T9@2XYHC}pq zC?ct2Az|d&>%o$#{OQZ%qQ)va&F5A75DTKP#E=OMmKw}LgtN)>?M2yIHP@OQW))21gp%vWQA#{o-gaC)uqma` zt#&m=VA{zXSS|3SptZzn8*c+asM_tGhpJcUmS49hQ5<%sMZlJw+B8_<4?i)Uh{yj~ zg3OA0fu|>{+QdsfyEY(Xm*`mTH0Vhdt|%FhGON0MuF*5d5^R|1K2d)QKSJd0uq!`?O0zDT;CLL0Mt)rH)!#a6< zHG{7Xg#rX1#t6Jp=<<_4P_YuSWNQ70#dD)_JrGdpB zy%%w#VI`rFtT? zT;AcjMxIdoj^YVX*oj9Fx%H`7C25qNFFWHna2M&g#52$6$jzrqvU=gX>10Yh8Q5Ib zc6+ACIHK-z^n{lNH=9N^_w%Mm-XFz7an035{a{tX6bAQ5W9P2UQpK&<9#?3A znL1|ponueDD#C16$o68tX%btdHwFEd8z1wyE@1B|L*~h=~&f6i_ z54{HdMRS9MSO$q zkhb2*szP;}Z4t z5D@NjEy-o@H289u6j0A>m{R-v-Ngv|mi1+n^Vn9`9qL6`ew46`>&n)0PTJxT2;BVD z;HY1wPU`hWnx{~=-r^|9Ef$}DTHg9)p+3e7IE5)nWU}0j_AzK-6h^%T!{)TbodEhSfyW(J1nfqHu2p_TH?vY)s7V02y~xX zSs0)25Yi$*@yV5QiK(sb4X>ho`^zhdf4U+6oX30iJF)_ruX6mBIZWaWJ+&F$*QW*h z#g$}ytBY8v$&ld=sEI&FXG-N4U>|*ox}PejL(XyZZ%Zir)70D@*5X}p%~6AAGs>Ul zSG!|mX(_iekX+dQ_MY!ddr#!;l)(Z^dS^HQ#iNdfVV92WmI1ILeaF0;kG=8koz(sm z2Zs-;K{T#e;WSuke*siUDazX7Z@`eqvv&pDxVk+r2mf}EwKNs`P8HAEVO1p~7upq) zxe9qVRoUNf;A8Q35i4^!Y~d}`E<^;{RE#7R^y7$GxnrCFN^Kq3h4lRnD?-Tc^b}h# z_(@*1GHWLlp=ZU8+-l3dIKZ6Y!W^kx(+1eA1mi#Bi!?&hE zU>Si57po8y1@!|XW8bg)`oMmNL75voVY3_r8E)jI_rg?TrwzKaJ+j(aNwM&&gM5HI z3vjCdyip%+b9htGE%ZMd1Y$p0{FzVeJXWR0CO?a8wTb069_%v%W*o;GDUS_obSGUJ85*58uNuR2swvel_eCU`*Zm zWw`x+s3R=w7c+X0l|Xgt4C)@n2p@C{H-VL0%o7)@P;$d{&-ZO*Ry8cWA(P|=$*I%O zu@XCUO27=asyBwt#qsVqxiJ^+Hbm?`zesw@rcMi|`NGtkY4q_t6uGl@iF~E-eD^^i z$@i6o1SXMLvz$OvGafg;HLweY$_S|jF|5`WZsw#JIJE*Hhl?`kg-Edt` zlEV{FP#%}A@w%Nn6eP|6+sV4w+;oXt4y#4HAc;=^f;N?NG_@e;uX}RGXkmE|goYn1 zMnnDZ>cO)Me}t$9gz*XmWznsYz3OF*jY^)5VtuIVs_+rvA$UFqVz~vncO2UaG@IN^ zN4~#e(gXQG_yYF3f9$1h#df}~K^8rP7M@Hzjl0BU26XGhDw+~#(iP18ewfRihch2^_d7%8=C4} zxeC@?0igKTH6r`T?hWGVA0vwyh8e92WEX$sFpTQn+cItCyFnr#` zJV;gE=*&4E7XU&miNScfO3&8<&KN78hOfdx1xHGgsp`3hV1(jW6JEZpE#g)5Q!uiM zY|pKB@%4p%AF}0uju)-=Mlu&id|eem&wpM1GyVncJ$9^fF5B+OmU0f*MC|{Z*jo<) z34^yO^F80;1dpfvwgn+*$JE@*?=g`*Z2WeQh2OF!sl*3O59sC=u9BIB@gN#jV|@Ed zLkDhY=o|M6U3Z-enLd9nrp$kJhP#yb_QFHw<;lh!pYRQ;l=)?qp68wK+}G!U!%f+y z;O|1pmU!zhft^aXLFtkQ$?`5GZqPq13kH{)w74;)3&JK3c=ZgS=HM@dp8XV`(}r7! zzS(Zt0D3`(l&ifGYW`(w@kVEvQz2e)uK4~_p9i>0UqlPTWSsTEYkpF>Z!yo!Dx4!k zlnc$PJzqur5n{*PKj%}Yo?%pPQs65HNXZ{_AHG^P8x6GU7^&Xa;Q`&vF7VjRxLcr` zq8M>;Ygt2RRX~l|oTfHBxP*@tyYY?uTsxNWSjTKF_JK2Be4<{A}*hqJ$` z?YNG{sjCuKK9Anw++xJ>Kt+%6_$&pwu& zn1`POME1PR?&IJ<4vm%O-18$3zofZfB$W@xRnP((tS1`9gH(Nn55i!LmVl`^&ZsSH zS`OKJG5Y96tisc^#9~@tmp`o|ho8L-fL2nh%FHFpPxk%uavKe?na(r$w+22TR3OSz z`ff+pViiWPKFR!B_Eqg4Zglcut#9g_k}J|p8BRtp?hkk)9wlBK%6^pWn)NWQ`abxR z$v)@3JG3L>u2hP@&vK()=Dp;KTIm;VC6c|O)B zq^u}wu<9M_QNnLl-Sx<_9DXmt^Q_^JGu1OE zXAqTb`FUwG>3@foJC}h1s<0{4u6h1JJG%Tw1@pjyXy4!;VNt%wH{>c08(bIoM(m*B z5vY!3h2=zjYs*hFDBE`Qkjyj5jqinnsJpd=@?duXlr1_W@TQ-t^Vj_(JRCd zI;|a-Z_pI;W?A`nUBtiAR#r$sVu^p0sLxjr?B-(Ma%??l0$%X=Jo8Yv$>l5Eg&cJb zJg@#UW5mbFulQst$h`Db-Z>@?uI33lm_oTdjx(=&GYYFa&2BRW2(-YUOG=NIwt=*~ z6rgcPe4VOCgp`q+%i{3;8_}vV_WM&8Tr*I!m~p2nIp2xJK}-L$$`TD;a5Z@QdL>Rt z@=*I#9gPJRt@?cLlJnIm76>tkzq9id<2w<=Js5i2nI)W7W22|Gij@A!!CbxsDvK0q z#eLZc9mr&L*X8C;jHfqH2T1hw;IXB>VFBgv;SOR20Jd4I20fyaTaBPn%Vvo)>uTh;EUI|o^T=7E?Vs}Me zWR+>#ajAUW@4}F}oty=V@W*K6u`bomE@qsY^bs#=aJX1Wi$OLKN3W$wk_RzhNs)(S zH3#-0z?+68Sggc7s}H>NP?Vgllgyl7o_Ic|>IP7cbM>)vy*W5g7(QliXXrRCo`iPj z{2ceV2dc38x2O7PPXEht?;InZ2GMwx?12h{D z3(UM+7O)kku+>$KN(TAd6&BWarAzW7xBAcEeng6xZ^E@kbh=@km17=-0x4)-Tu2pF z;5jw&`IQRbRM2`=RtABuuCr)2TlK!_Sw&j?)a#&y6oJLwPUg_boD^FrbD_oQHV6@z z@Ajw{hz1>G@sT^MYBCqPUc)~g_Cy*pmeT2 z$Kr`q5x43#{H($k$P@a|)QN?lXIjbgTx`ojo5L5e+5?OcMv)5VyISUe5M6VwCu@X} z-?T;D`LT-d2&G6&W;}PKeDpE3y=JmrV2~5fUYDsPDnm|kQt%+ABTc~qnPr(O;Un&p z*RfDJ*FgYAfwpTn%S+2Zh+QnLETtEgD}!A#Wfjq#j56*n+oHc;)e-hG4`^JcF+Rol zj-_>Y_Af3lC72Cb2nxb1ai*aeRKs$7<#l<~=Ht^n8?U=m$Hrp_8(Kp10%~QUYHr8w z4K5F+@JJJw7g>v$m2CeLSG0sB*HpJoo7LFn^`p(2yvJmHz#Tpq{u8oaRX1mESRS!p z9hapJK6il5@*VW?jB5q=6U?hiJg%l^wz<$Tb^0 zzkQ8Z);$XM9k;T)yz|;ix}?q$?x={-c<8|D`JlD%kdc^MzikH@8z=W-{U)ktqxl-4 zO!=`C2s6;INPR{V7Q)uCOVsXP@oQ_*j`K#PHVN9E{y(7`D;%(G)Ow#*V{dzjthUKy zYx85lW65p=mr<;5Io4<1mKbu~&KJVhoshusUv7NlmYK`_5)B191j(0t*TmEZ0G#r< zLN8_n{?G62*^V>3scyq%Eu*xxKqai1f`KLj$Q0JmU^ylcAQOnwb`+oJQ*w(lB5vMb zCCqZv$%P7~lyQGoB?kxK*mZgWn??G`u*zfuj#b~42ys(%ycB%A#)IGRc=_1;e0pI$E00tb23XJXs$8N`+C6qyniqzx~6&Tp##$U9xJ8$dwxS=h60aHDKWhl1aL$E>%uaLiUN8NCuRA~?LeisnCJ~B{9 zwz6w|ACsh62b}7jU-`>88}n_8fR?yX=f%Cat{4|TU$fGKe}Cn1{l1{B zbOB?YHu5ZjFzYipq#A@#=b$-o>BRRv?f`%@$_$V*uwyaT=b8Q(Z%X&uthU7}iyTsC zCbB$`j+8~s`(T@H4(*F(`oE6$09OhrA~~a~+3J2>v}%_801b71 zBn!&;5W=jx-aCZ%o_WAXyz~PBYughf^sTqHO6Ir z#f3BZF4X`BKAP|}oF&uulOgF9)-Ep4o^D0F&yKrfF>*@5@%>z8FnaQInFM%e^KIb9 z$6Ix?SNQp=Wsdg);?Y}gJXpFcIa#M41MW9>nN=WfPAnZczfCoX5|XrU5`&k4gm=Vo zhzYyKK#co}Lc^=Arq>^CO!-8d8vF_#91t#d{8_n`6mR)4Q{GXAa6;Cv$TG8o1RD$L z#Z=EraK!SaB^r2K%|;tf-5QEf+At+f{uJ z<+$onX`bNzZp`znX?cQWfTDyq#sj5{pry`_y5ZTW_v~2guJeOL(+Ogc0 zddthVPjMfD9bFiSk?5J|^6z?(?gRF>thJAZ4q2Ou|&F8yO&?W9r#ToYlEtqpYJiT5y9t)&byA4_{`(t#@e@_BOgSmjotpi z;`llk&YGVWszPkcHH|=B`G}wD$agzfwrx=0;i=dWiY20(F3BgH z`+#wRr!7W7xy*n^M8W>$-0j!yv&yr9-%@c$UW+JPJ24$jcr8i~*GZCn{jwwUkmzt_ zxkskf@D~}iNG54{A=Uqwm;l#goifDIKSF1F9S4k4cI`q;T}+A+wS(mLnm5HMsbdIT zSY@oo=HAb{$|IFRedv%1Ag20ltiER4N*aryCUshIP}N&48=c~-O}B!TsC-m%voLLI z)}B>1*1#_C&N%$D0mm-}YRkYXO(0crQyE4ZX||b0jau4Lgk8DKK!~-+mJh14(Fpj= zTn;IE5K1BJHE5h39P>NhknKB9yyi*jrX#nnHLNY|2Rp^M@LDJUd~oc}PB>`t6)Dtz zERX{?klO&d=#&YRDSD|@w zzkDeSA7%^eoG)SUW}pH*P%eWQGJN-ubr~9O{#PfI;#jVKc$EP7d}o2Y+YT40=xL^m zu15%;3Q%DbFpv3MtOL#!%7ughMhstJod~K|3}33`BOJi_GDKNeVv7l)0M)f@n)wQc zNuU zUw<^f6uv@)tA6Kq30IVN>LD^ft8F(e3(WheKsqb+wIW;R)YfDA8oP_R4w~IFt&zsX z&1;(*Q}`U{bFu;rSidPhdrGb^>`?ikF{t6iplkOONeLawb7JoRZJNb?2D>i;YZ}?q z^f;OVd%Wv1zFGT44e5v$aSx141qGz`$H)2D`3$Z7oocKRY~Cz*KIRf-xvC2(IFmd4 zsmr%ZR@Gcoc}y6uLEe^KWQ_&NKb@q>oOG+6yU5Y>`YxaE-8S-APz`iyTEe=;0_B+__#yu^{&!TY;b;D~JWc|A)`g zk1b(~{i)wXUm+jq_cH+O^p%=0bRR5-oT;V+-P|P6@ukax7dK7HVwX*IHyh1duj*O` z_&yu#z@Jndfp;sB_Z>QA+r%KUFTlIRPhdQRYz*Vno+%DY#0z#zOjDye7XZTpZ#KOJD znE0J+rSo+jU<<@VwRp)fMWb~Jkh8M>+VB%Tp_1>_bnqSuQ)|~YaBxb_day$c0s514 zR^+_+qY>!AObs>`ukBN#Ip_LX&e5>#F#Zrq#cialMfcWJjbAC#gUNq~m16~KIu!oL zVnXZ7p9Rd+Y+MLxj`7Z21ebHpCv9-7grT~5Q~C0Wa_&9Qp;;r8ah)q+J>cg?zi0r7 z-d>eqn&Ix40{$H8DE(MCj(lD5Zsg3}&{|&-+f8mH9>V=RY3pyk^yP z_+5HPJ@nARF%xZz&3i_v<1IuF9P(W*y(7Vvpab|7zOcn)fxdDn<7h*1WAj(tawtgj z2>XU_#+a6(Ue2QwTD(dlyW!=g z!ji!l!|pVmq{19Ci|N+5`cXFb>D3vjB?>vZa%b7h$WWYN-J9Kxcs~ZK-FWoqC(suk zwK=-CQ$)`BFiz>spwE0?q%t{Kub!~rqqK6LF#nXsIQDnV>NJsSOWF@n%jdwOL0wAU zY7*2-6w_dlbJ9$=_V&LCRlOx=zIeMAbvo4q4MoR`p5}FCbyf1kpf4w$J|vpPp{TLC z(9GO(%mp~}E*M@t2tHi3@k+!w20eoq0zKPnrx85d=C(<)F(m$aO zzI8M}f)zkl-9OcCw@MnvfSCZpLdRRVoGzdJ=pJJ_uDY3F6Mjh5PxTA;rii@r zGw5hS0=DZmxLN2vlKH`5BvxtFE2y(<4Cpdb4Z<5Sqi&Ri*u>uYHcIni#LC^m-#-5L zy%#V*d{I5KG^LkGt$$b;vA4c0LB_$2u#C)J@;5S*Q;5BtFKGTk&LZ$fT_81`hZghb zNmcpk5nNQzdugb|b=91-E55s*l}lWk7$B~{_kroYcG2j64)!q$qkJDOnQptW@OG*< zRD0SWr@h&GsF(y}mqUyJW77yM+ii%Uwqu*w^m;v3m{S<5#SlGGt$C12Dlg9AtpC|x2vN9RDDAn)MC&aM z30qWjj&L7pi*v5n z$7ROgT9dZO*w*)#sSGvoofd7Ol|kt86`*!_S1pFcZsX}LJ*?c>iuJT68}Iu+)8hS$ zqQODJ%48*6N_qKf@3Sr{*PRluR^?)js1A&JwM0@b7ei;zlHSC_xwHspSngOuEWDyw zCGN#46jV6bpGB>+1tg?0mAQ}&VBM$S!lpa}m)>Y?C%~r7?F7Nyhmufk(4VJEwgD%$r6oSXzDVW_;Q4IZ`(cAN`064q~H zrhBtV+RHK!4rJq4?MpaxliC7mZQ*g2wvgF$2NXR2h5}e+to4-F`)JUIo+G6t^Z@#- z^{!DLb-IQe)Li3tmz94_D}Zhrn}F`HIN4Sh^|GTaPSS~9%hn!=Y|J|$>SHTxxjP^P z?I%{4msBdI!N;TRw0K~7U+;Ah+UvafkNL=V-}u=-OiF=%arO@rq}#3nW`uhP^Pj{# z-K@|yQB3JJc{iOp@YQ8Xh7eP#U2t9XCN%sI40bhDu*>#@G!)B_u{t1sTc`t9 zEzFzxyD!_nXjBKbm!n-(m;GA8bP1yJS9-PdsoQ|1BziuOkpi2EQt_uoD~w3VIxtMm^PG$k#kc}L#j%7H0c2MRHcwHh+l2j_~OafevaIk0ChlX@7u(Q#htj}Y$Q?$~0 zYAjZN%LK!$=At-i%$i$XL~}6#7oIIt5f-CF_(oVkavTawD_$F3g2=f=BiAOv1u+W7 zi392G74@|fyp&~$VV>1LPNA^XsE_E&S4RM9^)B@M)K4)`KSVs-cit9M<+$XKiU{aS z_y}BpGMg@{G)$)BIx$s?B`-C$78S*bHE{bfR5FLsC`?_FhMubw>~A~{_zLEC-f^~b zd^~zRGs8k>ur|RRK7ZeDe`=&W8D9X~M^ThCPX4GOkjI&#d@z>cL=C@%j;V z?8%RWZ&?3e;Ps~N@sEh=wP`XFB+8ZYvBTW)g5{cg!s1Eqp9ZzXnGHoF2H&l$ON{RodaS>4^IA)| zW&vMfPBaryG-tgFT#w*1=pAi83JWTn!m(QWd$T*;Cx)x#-MbExCh#1UjE4?NRWmd? zJM`pgrkfe;`+8tgy=IhPs?6{Xpp>viZ+_N6Tfv^3npf@x>|ie5Fetx73J;}#Io)^d zP@z||!g=;Xa29fD2UlPde&<}emyR*s+O8_Fo}#FytCCe&yR;}cKrdy~q>KSXJ5mHD zHYERiq*+>-bDH&|L4m!M9kYiZ=7_zop+oTi28QilYBiDZsMARZs@+$Dg`4V1{0!)( z)UCkJgv$}bpIZzBby`<8YfqjbCoi;K-+SY424nSME<$Dn6=S8w%p^PXCG zV}hGOp&!NGrH+RRRPdOCsnI}N6e5^qQ5{eo=T!;KpWv!SAyQ2Y^tZD`T#L|{s(J_? zU~UdeqmL)a!NKd_XWT}rihmBOt0irwz@i{Y0Kv1$28;@!IUeD|-va4)5)C_b2qJ2q zmj-z{`Kgx5#mtk1FK^=c{2j`Hpj)~Q&NuFM#V_B(b%V&!Zc0_! zI~x*RzA#`?i}eBQ5?KpG^o8M0n;6yM(c$&kyHUD|77P7h2eB#+H(#cDU%dFctnpv{ zm_=y^4#N6#WNKHwaUAH<<5{^osxh3Z&)zUt<%=^%dwMQJQbU7VEkED6{rG*!m)n|$ z4|DG|eDC@hXMn|#91uMAABJ#&Hq{A%G7K*Z8RCZ<4)z^24lsW)4tc9!%SWrW%(mri zEAcVeH0+csPuZSY^MZKuoQU&82ip7mEG7U>yTwI7~2?T48M=Fea~~2 z=XAdR{+@q&F>{;dzV7RDt?%o7eg2RKdFLM1yOPZ+?_3-&VW-pAJwIKI*n0-PlHJYMB)hg2 z^>Wb(IZ?ARmew|_r71pxBwgR6 z_{NH6=_^C+HpStnFj?ELoGg2A%((?;3pNa_BI*QUtgFrG;`8%Y85zZMxfSoQFtP|6 zRCmm5JZj>rIRq1~R!}I}G|1mjee7m%i)OV=PN$VKpN5wav8=y?{HBI`5Ri0jimAY8 z6AoMWuKdiFDN`x1cM>462_?Raos{2SMAq zqmM=L8lyqYWUTmIQ4!#__20MtuYC=byJG)K%+P5b_>=&o{Jh9h+QOAeDfmM?`34v zog5`(aGtAT#Zmi~>+Fn08lnQ%5;=ko4pRx=e0xIN&h3WPi`5f?102G3hMGyRWO&rd z1SSHrS;M<#I_*T@lkobUx{gQAUJs_Vf5c`xl+?E6D1=ttE74FVVzi?|s^F9pFpF`Q zi=Qq(96LTGH>@?)OWWC7`Z&-+?1Z>&gfM2q*n>PPOzg2=UM7C(b3Or>kzs1k7O-x< z|5N_rqrm=?OYD&O^`x^+DCkRdK4(WZT5gI6s{q+ ze(i==!X>?#`6`cT^70JXN5X}+StX=0x4r{~4|lh1qr_C>mk}1?2sw>J5$$TaUFuXm z9=)<6Ps*#ppm8?^BgF?fW?c4Ei>}H(nP;2WegsORKSu2F;#fi!lD;W*4cX~;d#!&y zpYMwITGGpcjNdI#IUt7+TRu^Mf@8hXUzJq_%gQgV<=Qt}wV(1+OLKr5^y7@fu4Zbn zN7W=9gdxnGtGOyF1@ZLxaQCW*er>bIhSsLMd}dfCW!m3iATMw|nYV>;a2-5U#)BMA zHcq;E<=RKgoCWJ5%{^zE#JB$tU;nBiJDA^nci(0?^K9FW%h2J)@g~Ubw1o*pKpJ#G zzbf%JqMF}38+n@}(U+}rD*INrmI9+Z2KD8~h$g;i@;;v5(a+ol-j&JzGJuk z`Ay4<4GB^i@lvj>_b=M_l~$7j6ZqA-&AgZ9=j`eC5A+rzGAu(@Q7cAi7D-HLHZfu5 zZ#cf$J>**I%Ir+(9{$7GA&r{HV}dX%?4xC*hQ`BT2u~M<=i^hDN^xUu0oiusTa)MfSLTs+opVm^)*Sf)@LOtT5~2 z^p`hK+0d2di)%~c5!%*092Q)d)=lsJew5U_H69GvW-_@(!l$SQp zB*VVG0MEC;$}T;>;(Dn>l{sffA3sNn~0cF1YjyQR)dsUHC-emnx#4_733gmSEUPWa(chhpZ-`B^rz|(k4#NwS$-(PI-x?tHmgOeNBZVelTd-1&c!Io-LSY7EfMkjuBB! za;>PhP@-0>^r#Mw-tgwvcQ#fklyaGi%=E@yN?v-+-r3Y|U&Z$xWwpAoV>HKiP4}^T z#&nsTRd;@7@L|`A9c21qpu#IB@|^BYx-Ar;LNy?b$oDf7=_>5E+QA-QyUH;GWd5 zSa|4Xw)nZdr*@C~SYrf|Pb~-5F6Q=;){*Z%+JAbq6uHV)pQOq6n$}*5FBEXH-e-un zrp~K!6Ss%jS96|9kSeUGA|+tR$pIq(0zg-^_qw7{OQ3Rh50}bH&lARy(jtQ!snpvC z<@82ih!6#DqX$Fl%K49`_5&qkVt>8vrJEb##k+Li5l>a2Jn5xZ`i+B|pQQJtV0Iyy z@;)?(61_W*w_$?aKh*4YV-nQ&lg^H)cOE!^GPtdcbPE>4Qs0*#aRFjMxrzLRqq$sv z^7biF@|;Be#f?B?t%qV;qn~5@E*Rt~N;*xCNxBi&1>{WX}dt-F33iAF5GyN6r-+Bz*Kya-5t$tP*x6Z>&D? z;sB1N2ew;Mqlz*!Xd%#ybd9_aqm0xD=mIIQ&EWf;juXv#$VEqPtx&1d1(RMemw{bV z1E}Dmg#y#qSh4dLDx%>>UZD461nc_U>_}C)*bv6Y*jYBCkUR?qHQe3``y__343%(I z$BBCGC&%%!8T2fYH7XXJ#E6{s*V%37eqMl-qbt}99{tXGVPW(tD-NB<9hI9ah=EJ#x--b4*}h6Wte%lq>Z_~I+aY2?)@44 zKF%cD>_3rt$u}S4b>l7|@f6(L=JRAwN>9>JW&2?42%Vt*V*cvub4DAC?bWjnUs@!5Ol>G0Z4H?6WWf4mS3c^%ttslB?u zl5s1bm8AOWXoK0*C-(Gm=L@OBVTWIc?Jz-kaMwSlIT4T5+s*j?O85Dj_r83c*Z5!z zf9G$kD}OSrTlV$E$sHWmglYAojx*mJu&*nez2e%+Zu;)tm7*ap7=vjmFR~n8dSTdn?vU(fk@i<(a~e=cR-rAe>gXr3Jc?`j-uM;MY=u8BWd_xr=6kk4iTaaVT5xZ$NO zjgnW}3d7V&Z#VykfBq*=_uD^ixw}?o`uX82bF7GNbBtA+VO;e@bExlr!VO={uV3I- z3-toAV;9yGFGG0s({TL5r$E1sKM9XM^+)af&lCC6Pt1Y4o;_?Ya?JeGJAZzhe|5?B zeazBK+k%Y#_}%_w{C@q?zkkE$+ZllN6&tk8`IQ{&-+jzJc>vA-Ija4u@AeO`edgQ# zjF2~F2SfeCg}BfMe^%+kT?u5s4?**Wlg(fpOTe*fAI zW?Ecl(1H{^{2|6(Bg9@_s|Xf1xV>Q3&QOP21M?ZTIiL(LdesZ#w=$eM@aS z2PgFZ3r+`MIJfQEm$>)oSE2v$H2!ppfAcQLYJlM4=EQKXe)LcN$921SfyQpT>hhPU z?%!_l@1Eqta3HvEP9U(89KU#VpP5Cx_vY?=gTg+7LaD#y?#4%dyz=kN?C-7vLGHrC zdTE3PY&=UFM4}E1`}c>%idseMB<(~v_0hh*WVs1e1heNL5S$tyHNOBrmEoUXvfPfS zOCuMr*AQoR8K)h zo>v9}o@tnQ?>T3b?|l{CtHLFE<-p@Muqw(7Y%dKvuf?Y`?@4>ZB?5?8 z3Wy)t-=~qaG*v!5XHb%Si?x;8TnxMb%~NcQJP*VB5`;X?1BFyKKUo5`?ynxd4v_)jr1&Y6#15Aw}B&J znn>Y%4b7NMG2-Kn)KGvqP(fEC=>GAwL?c{QR?h4*)>p=rk5oQXSeJUVfB#}0-+D#) z)zRcRmcu(!tS6U;08d4d7!nBk=SG$8X9HWU(y7sfnxyEkfPlm zcvR>D=;p0R$QEi;K@y;n$nWkz)98ofz|L&2l5FmSSPF!J0aMqY9S> zuO`Z5`}?r2{TdUk5x4j2)h*bu2^oe2#fin(RT!wzt#jT0SiB@@3UuP0CmZCC!4j7RCHGL?X9bxvVf9J5Aac0zSG}{+vv;g zEokB6(#m|OjJw3BySavgE_P0~JI<^+;KUfQQ9|0PZd+JjpQ zxLeiJ?@?h`>C+eE#9hh#f*3vR1%CCIhl&KK-XW;xTH5*oAt`b1E~j|eu6Y|`PfA_; zG;yn)_AyD`?Gxd7EB$=OZ=0|F=_LIm=%)s}^;H~Y)?jpcIi6Q}aV&8hZu9}pzYmo> zTL$Q{WCnmyP~O?sm!+L~is8taCqZ<}VR?2NZOsxTf@IzIFRxL=ETDjNA@+B9kiCy}!vdcgKSMNy9kf zs8*@RVQ0*ZlqLo}R^EaoJJHVc+dFCcRfe0B?uBW)vpMvtO}L_2o~T2W-a&j=6mrEVAx)42rRv6$ca^ojU1GkKK#ZPn_3%_msiR`U^<{ zMrh;}eCstltiXQ7CA*UR^+XPRrxE)GosuF1gpRVx@*rTNu-VXZCly)ltGBM|<|sHo zZSfoQ#RfuBocdH}TDVYxq)Nk)vu8{`J)h9m$gVdi)!NcHs~-J4eS`K>%4(M+Y_tv{-a=+_ zj0)1(OI-#fS@KtkaLv61>LrkNp}Uo(*0;v15+z|vj>~an%y)qW9@r!npxrYI*MOPfYf9@a*x{inLXT2`?-dgLX zMWBUWL6Ti} zxOykY$CEL^x|4P<+w$&|t&uN89>?p!(*zNGN3*iPCPqX;o>Y-LH98HWLn6l!|#-(|{nKD1-fo6ethwcja)%+2+p;y@@^Cld}g zEKnNZXfED4(W;av>8#q5ZDd_;rFYVqv=sMbso%bd4Y+?$ig%cs)@uzHjMD4L8sJE& z#uYypQN%iw>qj)QoVmq36C#PVFeo|HWAAD?PKq*iQ>dm*oM?y=M-R?=w1i8Q=waNK z(lgJ^^|=KL>YEtH#ZHN?T6thAT&Et2J)TD-YP%0X;aO(4li~!lvnLlMEVUDUVrMd8 z-FHk@&hhh%{x_;K!{Cj-1SS4`9PZP$d+zqUq2P1xE1sn>Eho?zA-w`~MuhnYXR>zj z0oDtaE3ucYA-PqA<~Ra_d;dxD1K)M0csXz5tys9zy!q1ULh34*%v*nb2u>+yJkk6B zrIy3mLoWjMP36^C1KoC|AH6gHSTfe7?cyz(!LYDQ)8V|_jx4tvbtTgId#UsATPQ~~ z>p&4ozo?N=I@Ti!I$|9JF~(tw=8zg2DX;I(v6p{392>jhM_^n^lLJXd>HqFXtR8zo zsF3>ZL7Bx1CzQKHSmSb30w-G1RNmh>?D_$~gd+OZ|UY{EmcCCJ36EtC6PN4}Q8sv5DMWcT_{QzZXTe!I6_i#KClYi6+O_cwzZfmLqF$5;`7JA1331l8K^LyMdtNX2GMEg9t2|P009a-K{u2^$&HbAW6Yr4)@+P@<)&(468kR zMQ_y3NMzrED+tb}B&CEvNJG!7 zUJeaPx9s|=X$;_d?M*eYMI{=iJ~3$slPE~N9o2gO450LrBkjSS9@f97tKee0H+y<4 zxroJ8AP^oLN5w(;E}`iS*-t!bspK@>_7uJ+iS1co+;U!?Fi*{S=^Ycc*pt8jPOO3; zlMQu{M_!pBpDX5LA5T^tr{5XnmICD(M_nMO#sM`TfhdFJiwJ@#;b;u9_WI*YWy zyyg`sqYb4HZ=<*U7*;F0IH~bha70*ysc{zZRv+1ksow)}V zXeL!yS?c@Z`#5x2YY~2}&i%2@#N~090)Y>BYe9qb*AmbD^Vv3+g7LL2MOh_CB)ZtW z=~$-`Ij_#sztwp(w#b?^$FY5>8h$DCrojD6NYGJ#TN}V6C0-U8B?O(jTugCbeDGgY zxKmK_&-Y-99NmIl@ev>%yR~A|J67vOS9?JNLF#up0q~QCK1)ens6*QkG;O+qqaC@|1du1a zm4Jj7wpmD}xyx<$B#j=ePl8J7iS>!u44eaysw0f+#LLUE8?j>Qk}gvZ70Dd=Jge&0zB-Q4=`Joi+55B2gm)H1eqQOV4PG)y8bVl5i1#qxG47}LX)we;l^59@`NUJd1`zN)N(3wMx~1F zg+Y{CCc}@W7ZrfAP>j_ahZw0XGn`!~|2hLZwb!DA!Z1$SqbY5&gCzHc{z0U!MMF@Qqn51N;7QtPM!utZ1&>0r^$Prnl^l}zJXl7&1RUZx-h7@Nb*J3bZ@KM|+sc51 z1Y-Jas(O@07@x|KzWE3JMW5DueD|t9mGRLolPO^J!ab4xsmIRi`4dt_YAm<<=3n!( z{jHz6UIaQ=AF6l8|3X5+1A2LgcWr-ceE+XUgxPF|!d(}0jo1I{RexI`Vm`gq$9mEK z`OJUUKl}hd9^B3=@qdBq{i}%M53jvAyw$z@Z2K_mht&L!cOl=k)%TTzxBhoG|9Kzi z0ABtsc3I`$l`wy?;Jnj7bTJ;K`%l}kg#Yi5p1*r~DH=d@UA-{#zZT8k{n@Wta|3`p zf)AIxRQYc#+_oKe0gaXP^o+!n|4wWEzdr6hZWQZC98*H%?GsG(MOFOc6TP=jm26Z% zGbW^eR;BCJ@^?62#+}MLQy+Tc2wRK5!j+6y>oCVX|L$4Og?~gaBqk-~DF~qlU6X`f zcK)n@%)58=?9>+5guZQ})=|Yt$zi=`l~E7xUOKD+PQnQ5HK)}V01l_olRe10Rhnab1(xn;y|uLL6>b?$CX)5iTs975?aa_ zwB--k5a!D}LiHC61rw|%4NoL5y(lx`{3)B^r5^aUInleh=f)=c&4GhD*@hQ+ysmyG zY`-0Oj?Y_TqjyU>re55_Nlf$}f;r=^o4Db%VNL&rwF7L@OJ0Q8;!tjB+|!NJ25*m6 zy==>or_Tt!a$J0b)&vSCf8lOMXW4rrChi+Ckb7vmmNx&JakJufLvLLrS3>Zd=B*81 zf_b2geoG1rgpCWQnLm7dj`xase8rqLKyRM9(osc}k}X*yuFsF2Oc{Yew?>9|qKDT$ zw2vA_mAaC=V#K{e@1s|ukx8sSq~H4tc{`pFwhmB}HKj}-p{^-JF-*IWpTRLb;@ZOZ zJ+js-KbutEScY}i(R9Opm}0B&&`I$(iJ}F8V?3-c&37bgTFjO+=EHE9FmcVCw!H?Y zyGSJ}uD)F+eJ>R#M#G?QlE0c{HlbO-$C&DkF%G-7phQ7%kI0PhK35V?cGWAKHRWKDA$HnbJ1}fc~y6aTO4!jjK(ASq6{N$X%!mrVoSVem@v<~)r ze^kH~1LbVeH1&!ArCTTR&5{T@&0g-zV=NNv2k(IwBvdsD?r6X9kp{odGt6K;KS0wt z%{@lCOHcE&$E<^$>$0XY(qZ?%e+ewy)DiEi6p!BbBW{plz}vBbTaB|2F3UNq&{~+| z4Efm`yWQEnDo&b9P*0LgFvPeRX(Y(nN-}ZDp1K{G6Qau6gy$MFo!V-)5xUrOE-D8e zjdUi8ms53Ldinqhq5@RZGjS5&^cP3!l*R&h>Pm~fX@m8aJkoNom3Zq`R-W?&eT&^b zlc(-{c;K})N@F1jrfS1LQBAeujEXd(`WOUdY{AW(S2Es^p!J4J?j!*V&8^_B$AOGR zHU#%Y`A0gv%v`%pU33#F*?<}GbhjFDX8+`Dq(AOhk8`C=3{^TPWoaYs^XW&l&Zc*J zC>=<|MvH2KDV7s&x`xd_M`Bj#IgMPZolI*>u610kzpD}<;`gXEUPZ*B@r*uQJj7+O z>+u=|U**`dUzkncEnX7xRsIl*4G*Iu&o#GOhjg_++YgqTO~hQe>7w%XoJnq8d$JOt zyF^Dvufp#A7P|XNnm1)eln6+B=a<*?OQVR|G(cVNb_fYmc|2%ftIzJ;IdL(a1D^8y zWvb_z$dRM`w^|dVNbKIk33Y6~kYNFL$Ti2Y=nPVFz!4Py;Fci{9iTq>tU#1W^In$B z*Ck~FvwwSWtf4Z1YA9#7`)LB}y0aYWTeiB=mpC!|2-7+ZUEEtk0UeLe%S5+9Y_Ev9 z8cDCpX)51YSS*@y2dwx>y`Fc<3$1{!jVrNt zVf?qvIYrDGc^5|OD9ugyXC(?*_s@e)R|Sb|TK6WfI!ef}^b9aKc@r^ug;ZC!xNZr*505X_jY?AD8`CT&vT>4i2! z-?C}6CZPLS(hwD|4YL%?$jKFauB>ptAwB;d(IhGFAwz56X=Rcmpi)Fy`$n@-nIokj z%d|?)z9Ec1;>=5yxY|kb6rtaaDw7~II2aJ^Iy-b46bQvI>B*S;?~-I}hiH$XU_Ri} za$>u;{UOe4kA_aa%%;4PeF78SyfIk|0o>LlIZLrI(1{B%iikV%gxfz-p5!e*aNT%b zx*RI8^vmTwx7bRp{N}QB;DC=C_i=15iM;&W2BH4&lf9N9SVi$Mb&o@B=hcJRHBEQl zdoPsP>kHOz+(()(4qQQf;yi9t{4nAmeNk%5RWZo)Ew!7<^ix$HZ(?v~&z)O0SqX|O zj?NkZ1@op|th1xjNdHh`cjkEjFM54c81k*`y#57i=G8Pd{fcJ<$voSB*H@Tf_v?oC zSXrLNEuVXE8i%|HWjb*oa8r*jp@I?$CMe;>>IotmJ@(%&40==(rk0G4Uv_*w$x|Vk z4ZT^t{xGC!+P%B66OfQcG+;EHcjKL(>fg_~uaF2Yj_AtKlllD7deHx(k%B8-P1=4o zqkDAkp|ww15&AplrNq!z33%_}r_%aGi55cTxz3{^ESB^|B5AzI-s{qKGx*az$O>mW z`>ML6!iwfvHXg>+`|0ZoB?GzK2OWyP1|?o%I+CiA(Adjmv$i~SWwGgbgGlTC%cKFa zeeSIO%L?U!8;k2OO_d?(vS+p>Qk9JU-sH?Kl-3sdL5PQO{r*I=(Iho~$^rU#sj?8h zr2UdjJO^SB9PLe>}Nt-#{Ekj$L>(%S4T#who zVIdhgRWkj{^isQSMV9@Wc&WihQ~CW#d%4=)8|-zbN3bF0G7?q@wFl=Ha5EAIn#ohH z<+3xQHq`m+qyFSElz88ZY8^#u7~$eulGmzAu}!~;%vDF}Z}T@V3ML?Bx5@vIKBw=0 zF2~NQg?{V)OP>7^!PYi!S|ku$u9(6mY!rscz%EL$F^ab)VZId^)e>=4iupO)jRk3_ zS0jwE(q{o_2AN$FH^Q%x7$)q`7&R?=q9I#isp<{0r}j?2i~3dPIE2ov7K1~w7DBxX-U~X-Ss(IQLCCD%%+$j8WpbcSJ|d-)0(S>S)^Ao(<(6&=Mg?SC zG_1VtfuV%*J8?%M$lC*32dVXd`}6z$1Y$MtK_PRbYLffBX8UtlRd z4@lS;ZrAz@Xe{;_WE)dvrWv^}$gJK1N<9evw~Ecd(^~_&!7|2BfO28$7*QjyQ-hL3 z6s(}hut7Z@)yaPQr<2!;-h6Orp>m_v+_Z5ckr#|io_)*FsCz|~2wG;_&YgKj$Plkd zHO_$`L)V5D4kZvXuYQp69+6vA?O*93hnc<7CL8#Af+F>) zGU@dvkK0ZC(jGrue(8@=>mJK}E$2P*tg>t3BFo&0xUnE3#eLH>Y2-RVmSbhz^IJ)4 z4s#SannQOvr;upPFWNpoTIajImlrYXDo3@GoQpzKUsG_HQN~(7GDh!_z)!8`#7oZR zG|ot@$p~5U8rJrhl<9b6O-gm5cMSN%u2W38Pt-X9G zGqXbW<9y!3=BNvmF^jdmxhdeYT4I+@>j|tYW+-ex1wFzlo zd72t?F!OOsTSnrqTmWQh$g_`Di?3Pdo%f)XB!>WJ!&pNGbmhcY59uv6NMBET-SxTC zajY2QJ8|o^%1j~BRAsP%|s*V$$;g7s5WnUe6a zC9Xv6jt-K%QShkD!YeT9!jB0AsZOW7}tevmTw`g&?S2)Bvlykd99Pz!~ajH)#< z8QCHN5YScFu+kQkat-P*gR;K6V$nDmtMv$q-cvOzFtnANf9gS~LUHrBFZ#?29;imrp|X6+yOT;S zBhy_T10I`^k+-9S1O5F)*KJBVGqCkd=)$F48GzdGyqRjmrKIzWeneVN?=ssq5JdBv zsYX3LtQE7S*Ps}x z`E!OM{@bx#oZ^~uhK^&CcbD{Jp3H~9KD_3ZGETImrydoE#&>%!PhOttLX}Lq1{Y{& z=?=wOC^y83moP+cO53G0O#;IH^pus)P%o!BbJ?a@sg+AUo&lIt+c#-A8Lo}Le$-fs_(tbA3r%0D}(`5DrIDWyO*p~-=&V8QD-1aDN!feD#NK6R^hIK49(YPQzco!B|KxQ$p7 zJ!jMxCrhxf`&^%w@6dS+`%StGrV3hFE7?&3cbf}ls-Ee}x?I3LB-Tc4&NeEkDT?#bxoUSuQ#DFJNTCNK4$!v|m*s+8>HGa4{a__mwmZ6C zib!vprnaUh#r0?w<#HeUz^mZ!>``mw?w&lp`f~CKnc5p0%LSE--q_nc7h+@(7ykJm z?c-%;+3H0+ulUUQb5;38%?^&~94(r|6mS3pfkOL^F*&0erNL5Dg54Gmg+;tjSJ955 zK2Pm#u@ju$kLR^^`nEzUbq}x6OD-P~9va*%$k1-$OF5}%vAaJIAaP06bhFQ{CBBs_ zm#evDPwDQSScI(3-?TQ&f21;N(y`%s z>3&J0yeXc~$`f#lG@Hz$c1&u#v zbt&nuPx`92oy>QyQrD!S32?E>sZZZuTnIoI){!N?wP$W8paU2qsbnj-2i~UtMqhPz z`>wWG4Jl5knI)f&4+=&V216e2gO(?|HJuV$i_M5)I`f7+LZzPLuPzrW4U#GcA$=wA z!2+;&z`{srh@+>tf8_K{WuD=-y@HBXn{&N_XJ4^JkgC_z;+fK%h>q?19bDf6dIJv^ z98Co4;Z`fjj;_lg?(wNJjpWRj4DcpB&Vy8=!UrSY;TzdQmCiIa-sPHo7W1AI4Ezz` zq)EYZwWfF1C19A7lL7(F&el-~?nG}s&?$1GYs4-S&qdt^tRJbN-snj-o!JlSZMgJwGO)jW(9=Tkn~>13zJnn>3Ba9d5RBUF&u3e@!xX>;n$HF zQgPL|uNzkl7xkFVep8n_y;Vz0hxmD0oLvHGiNkIzn~+mLYo~tlI`r%M18$eRU@~9| zx>w@D=w7M8vh+at*)E2yd~w2r_1O;s3@;6k0K)(5C+ebH$c0Th1i6-Q&8xD1OZZ7cMgC zae*OGGUIS9<_W-X$ltjWQi(6AI4a2D_))h`$f)>UJZt=@K?Ew`;0yr-de>|l!ubn6 z)mPc02i6RbWUE_pS#kd43bH4b9A{1WW-jXC7+#DmTSDX#y!vyh|4WaEZdZOs}g=DbUU3EorPb~MGV^dy+}~U2YYMAP0fTkn%&-eAc44T z$-PLinc7xHL<{#36ZAL= z`o4|wwd#4Rvk3KLcu6h4uPl)Ktcsjr@!Lt6qPS|gV6YW=Kbc3yYhcL6n2ZnD#V}W$ zF%zVJ7cnqNdPMDF$VYTKt`RBFPs@VaRyKa9?ageo)p(4Os*(PsiY33>Zjg#KjDaAX z#Zd1my?mv{GLz*kkY?Ib>Tk~_Pvr-X4$K+$ylQOH$#HRMxT0CY%-Bq|;%J?<_2WyJ z^=A#+B$6s>x>5ZuyAE(!GvBQp-snSaQYta(iC&}!=4*~-o_ly^xl}<>yUcMSSEq$I zQ+B|R7+RDMXtUbqvzM?skT1g;M)2E39E@f&r3^J1^mNv3uEjKmDvoD}C9+s+=Z*S& z%7EydLBgPrS1_t(0mKt9Tb$`NBwvNtkX}37FwIdJ8mBlT2(5mNoz+*7(CesRU30z@ z({UE@1~Hj@Jxsk9YS2&oAFyaQY~VbBF8zlUqvJCa+8*`=VsHw0x6e#6 zXS6mG4}Z4@e{2>T%`yQsV^rY8#qV~_)P)0pqkFb0kuii{f$ha~( zWvIoR`rK(*MDdQiKzZHHOv&uDI0GD8qAY1Q=}r;F#ha)nVzF4#kj0!RpM9`2BTXZr z0|YH0S05iCLG|Wwa2?eQ6>@7R#s&SYGT)s^#0jT zet|0~?%jKpEBi;gv)o+EdWs|~FvTq%*S^wxe0J_V^b&KYn!Aq1#!~ofwG2j3^3?r; zUqg_^?R)N|w?xnOIst|cy?h(L`#DZDxCOOYBGI<|op$*N>AQ8^Y#pyCG#MrTWf9Y! zfe;sl;@SC;_cheHqSI~IfThZ<(U`%_IR_UNMl>(R`h5`?RpD^o=!F1KnKPZ$2ZY{3 zAWE*n)9RmNkvhl=pe!14uSe;ugkves%#+!VKlFGn-vUkOgXMZ^KM(u+L|=H9+&>Mm z>}P)%FAA@Anrmw{>=|$mG>nQ8GSYX?c08*Cy(I`{0+@;=phE*S)OapnNH0gcW#TD2 z@mAVndCoy2{@N6$U8K@}>h#mX$M+0;@XsB|Srv=XZSZjBl$yv{PY{1gc{`UVt~pOhfM5@ucjKO1@0K|g z>-teL)pXi-AI2b}7-#$K45-aAE>v-MI8>JSnKKI;*KeW&DHf-kv+k#K5A3w{z*ZVm ztQb{QC>qIGA;kTb1sRgLKVE)!gDj8&_^p>8Yofh^`?heY$2_Rc$rGd6wa#`V?DS9- zZM?gfRQiU!?-d2H&(>47n4e+3W*;GsEbCU%d{Vk}M5YH<4!MmJSYz>`4HoHnpZ^Z3 z_tA3Y3Sje4H&1xEv(+tU!I>OP>0fiw{sHj8%eQ}kMvAEWj5KfkC1qqjyF;I`zgNt@ zm#>cWyn^ilwXRGh_Ns^C_pgk!&CKsck9CsL-`>xzX>(x)nx0^vdE#Kyg6;2El!ksu_dpr>_o7X;G z(U2HSpC(>$Lz526ti0P!x*f1$|NJ-~j`DH!rPf%Qn7s=92I^5$PX#911RJdLBdx`U z&*baPcA9=-y>#=J%U@jHA=xEQ;9A-Vx@w3A9-1TBG%M146|%kJ-MHPY1H5R!r%Us z0yuTtkA8<){e1DmxTla>I~#2x-JuJccpJKo6={F+{z|G&uDtB|d!_Fz)O3%D=$^DB zr7JXpQ3W?wc#r7%EZ3RVkdZkDPwm?7g_A9_&$VmlR5CFUwyNLuvG_{7VX~**6doOg zPQE*o;j;?FKhP=tu!HOY{m?qN#@$Z2Z)qFAx@*M_uqaaIUfm@sVW>j7lZJ9l7h}aE zL5D6tcXrLo=);hpMTOf`7^ke`ebBXBIf2L*yzIqn?r0V&dbq_>D3$_5+zGGa8_I5n4%^mw;58E_TN7&JkUqxp z>95(xzzW^V-plx>>TxBp+o(L=8d2kdgAtQhXelqr`_ipyZn0}^oq0c0G`f3=`RG}T z@zh_9h4t=2eRRP^Xr5~1Mdi?B-=XL&SLc62$%^Beb8%PZ zJwJVSyo9Wzkg#6)CsEWM&92DgB*qX>CfRc!0qlcR5;CluGYXufG=3Y73w7-8NrBp zdQjqgc7XFq>&WJN<@XtI^LO9ZHGsCXA!4NwWl)j|-Beh~a_@%8-7SgC&_wRHXqrAG zpb@*65f4TPuY84@tWw@rRy;%DN)#(vi$T1nSK{@)ps6T(WYr#X;yeIh3(;1%p4bL5 z2wq17{&A||H*>-%mMk(jK+`Z&@>%p>o z^VvN(;p5jzlqVdw>O~6AcI9X-Z?0-#V7(=q@4#%XPC15F-DUml-Xhz13t5Ym?SK?e z;LEFH#UKo)$h~$!!O_{E7{4qfuk?i)xIeUeSxUripkfm}kn*MJ5dd2xc8<4%-(Ic| zt3+v_rgH?Qz(k58UC|FcN?*e{0oGf*TK(qNmbhyFUBsnBYkN;Ht8g(+vW{wM36>8PFd-LyzqaLT%q^lY{VePVKhF3Wial>}G5`hkReyl;?J zG4__#3TN8(x7Ou%Z@f_1;o7>_ zK7m%5PAZ~~TecPf^_EhEU85Q91NAxj{<5`vSi*b&Cu2cuJ=c<&8 zyG}grJIAnYCZG>(7x?DVId4)YR5iBxGTbGKNDjwh!YH!G&!I8B3IdSK)5^ECw~S}S z*Xx7`aE>AbwD+wf9$#8=13k#QmmCqAPBVQ{y539M1XlM--8$+DX3IZ`?(`-tf!Pn$ zU}L(WDG8?vDoSaPAgi3tDe1k8?95oTE0Bt+MJZs>qa1 zSzE9Qc7KCB0N_IA+RyQ0(*-)mE*cBi7ZT4Tl&^&8lVeqCQpUGX3E$#m(3x6zdUa2alw-s5wB?)!7!_c!m~e}_8GdA+Xd^}L?rF@PBF90qR)BZj;3E=A_7 z1MJGD{ciZzH>{yN`lW4UnBY$-@}r0QGZJhVbIu_bWNw?KIV2w`gl@8+Y!ga%YWb;=>Tp*VLqL`CW3F!#i(_>Z084iJ(EHsRFDW}BZyK)bEsj#f zsZO&ttt6T(zKbuw>znZu7F%SzcGa>>(EuXng~e}$I9`3xD!pb7QF_AtsNc-u*#%0| z3jsuFBo-gH6z^%_5_nMxwdXSG)Io^DN?_r9J$vih37F=0c<l^KcrG-+GI{*EaGNr&mz}Zxa83Om=H4(G?B{0=cc<-|J7Ir@77N z*YNl8JysSSUzHmIU5XLb;YuB6NO`o^Mg~hVimhD2EW_8bV-2TG@hq9S4R@%19_D6Z|I=AzN=gxJ44-WaJQYd}8#;G^Blaba|O=-Nr2Ag)o-M|VvL5tjRiPf6m zMP+2z_Ao_Ux=-*L@^&%xCebcUDPZV01&gQS3tdb1r5(A?h_5lbu>!!o8eZw==Z8aH zFB|a)z7+K4ISI@2_~y1!)~Iu1iO}Qt-YBQJ9(z<(c#@0pUOScb9$f4)zL0!p-2-%aENAxS_Pc(t82pmSHdjZ=q1); z<*@o@2`%+bjdw~F1DtKHPxwJ4X;+8^H>#W`h|Fy$lI$lPCM7i{%RX9~_~a!FEZ(?i z*XEZJ8vn0F%fA^uM9aOv;NXX)GijKFV{Z8NOp_m_Oua7~3#Cf*u{&ZJ^aByU$5-xj zI5KiEZO;ffru8n~wt+#K_>Nh?>~-d`;V~`hOZcA1N2kpK1ha~?7$W0suLyw6ZbN1sCZ4ntZaP z1U(vZ^lmWe6k2I1jO+a!Wb)@Eb;#4>Yh`9CWDl>m$iz248lV44ZpiB|lb+@7z1(?O zQ+1{M(={W5!kD+|(Ip_sZlEkg+YgLyGJwd$B^=wiQ8QYp*b!Gd!Gh#6eBlRp1s~_` zJX5f={9@p=-nT>z67Va88B`gGF&#}gb@y12V~KjjV=4}ale^ReMos%M$F z?Q4k&hf3^-*hZ#z@LQduyI|sN7Lj=^mv)8S=3CJDFwd$nt1LUWeSsGC#SQ^dG|^ko zYx}DyZJXea*yaILZE<3_#riX>tmQyb9yw>I3~IaUDL8kWKF*21txD>#?FjQ|X2H&djcR1JfFQDK!mF4!-abMRqVB?% zX0ldlt5;6K7ItM#)9~X9tyTy975QQFr`|}oH(Q9o&)nT(#yyAVW@;5wt0S?#Qo(DX zz{{^WJAIW1m~N6)JG)FcLAE;aZJB1Snfez`q&vb_W_BYGk8w^EWQlcggb#jw!tbk| zi<7!EPaFrtOYoZ>cy>`1Ab7t@d-mSCJI|K8qNa_fMe6I+ox{ifS5EsM%j;&gBTrN; z=l77(l65DkJxwFf?>G|d7D^IB<1Y{v7SG384>3CQhm>kkNdO2QC!3Di(!!Ww-uelN z46_b@(}sq|={7`dC~|k#C%lGba}4_Oit4%3CkJXN0T*{WQ)JPDk1-V$FJ88I@?;pG zp;w}_bV{7mxN6{E$L9TSQ*i=~eED!IlNCKG2h z*w~SuZqVPDZVOi>_&wUbHL`&dp9Z*9jT#SnLXzrGtrL~>6`BeBK!Nhc(#f#dl&+bH zymPL(96vV4pPszEP-8JelC);g=)%A7Mm#lDAIf(bz^G&=Sv8 zMi!RgPG306bKW#4C@9Q!hLpD_#4K{gina8*xfGs1O*wrI(`7odhn+*Z=$%{$++V&J~C@}`j9A5M>An*)>efNo!IPQl{5 zdZ%9^!=Lw)3CMHb9sK&nu1==_oZcMl?Dwx+y{g)5o?Q*R@QTalI4^GP5Hfdfmf{CY zu%K8?usktn^(YXYeOU~S-Oft}eka7QYAL=Zxt3$4E>Jql=4rJWAa|naV*fJ7phRh3 z-81xt;o26tYY;AZ1gM}>?&Ca! zj3B&O1`?^%6KD=}0W2ZD(>K#=7x(EpKlJ-40{7JLz=EZN*Nht`Z9eAC zvyM-B>4n7LB8%ZpVz{os$1Fk*|6{*Z)286EjELKP1Su9lB{8@8DKHf6XoI37ChV&( zyN$TUGFN$`Sh}du(FMel@fO9T=2K9W+$U0WwKZAKO)nBA0_~Rf^a*a?A>7tkb5ORh zfVyvNAsBxds7~fnd$~p|BfGj40U+}(iJkwyBGUgo#O=~GchaNuOKh%h#=da>a7lDm zzhJO6Q5q@HgGAsw`>4$TW0l5MW_(wUscZxv`S$%n{uud9}M9o;Pdm zmE*maePkQcD#tbHT!+kUrAG?>^0MEOf*6o^eHaquZWz)QC zJo@=3s4^=8kb+Bh?)EW@yA4^b?+Sj6*U~R=O#9l@q>sqb&fY%NSBpd=BKLL9Zfnlf z9rLo#)Z3RZ zW1ETwI^F+$rT+K7jz2x$+*wbc2HDU&0{OkVD_r+2C8}(fgowNC`%E$-H2v~0lS=1E zGz6uJVz_wmx=+p%Q@*a1nOHiSH(Um?NUnsCDwSJCjcq3xc#AK4p>ErsX64;iTPw}g zZ+l^pu(Ie_Cb&4VXM^eC%)ulZikGPi3+!aqliwuz@SY`Ydahdmr-U(-6uHDIrt>_;M7f^4E8-KYV}p5qkfDEF=Tc<|MGwX<+{A#%c)4a=ua zrQH^aVM&y7TPZ>2rFWCyAxv{h<=(#NV`Zj)-rZe5@F%MH%ywRena@7#A}I{qh+f_) zQkUcahOp`b;fuAD6UgluCh9lZhTNLcU5La>0+*+pY=N)w8Je3|QGFMEZrg?0+Ey<2 zW|{2go44xmazW>Z0bXgm@ALhYpRN}f-)@|`B@R{oAsFW*fr0janeY~#>4?P+P8Z?U zp!6hQ*)!FLv?4qM$=*r3SqX5cfS78Yy}saf_bA5A4Zo!0vneL}LVVlUw?{2rV$Oz> zlxIR@p_Ld9=s5Y?H8y}N5C3UFF0&iGaTMbH!HBTeejGANa8IA$UJmP zm>P>r|zN5jZoP*AYpowzux zn`bpc?nh*%el=t>j{8&x5szqqf&;0+{2PS#5uKFtQd+7=vco`gO~OcTFl{P^gnIZGe52O%^)^_KrW%@)@=pwHk(LFjN~x4 zY@kvqv0XXx1_wYIMdbD_0r*Z_rIgvd-64k9L%XZ=+>5SBdjqwmfnJiZ!uZm6nF+w# z@7r&N_f&oKDtkZlmGtLoHZ6E&({f2b8`icr^O9l4^_eq?y;BgC81}>uOcsI8b0be! zV6(}o0X=~-b;shCE7WGSqyX6dmJ#Q@f0-Wt*B{IMkMWy-Bg~pMnGGx;hHsSygfbbz z%ma3t1^jbzL*%s=Kk=3u`yN^9RG^dK)5(dd^2uW=QB8A)bAWyF#ki_$hY(D2LA-t~ zwL=sQ#6MRRGC)4gT~JZ%;RaMXrhuMm=@E}cN}6*_G4qo^=P)~J&!zzl#A8ANS={Sr zro_WUeb?^2qJE>3=BDTFUY~(aZII-D$!F%3f%&Y!Q4-RA5`B;!HBJg0##DEnho43&Yr!zuW%w2 zED^e6Xf%En^o7f~MPRFDS-R)w>8BIIBt-^B)$s2@xiqBo6R3@qM2bEC>HqcX*Vj7t zkRdoQTL3^OqbJ4g1vK#sdIHtVAhm24bZ05>LGBbhdoOS+ye2+9c`*%Gzq{#|*<1&T zZ!VQ)>v&|<@cS@#wtJ{d^Lu;JF7Fu9Jv9UoOQTNf28~3M7F&%DL9bUz+3rif8qrEl^+h^hEuD)ci$_OvJ&FOFt~pyaK!NButszOK z3zp=h1u~D9_>*^^Nbg!gIDYuCFe=zv-rM@2QbLeO=Z9<9es{yPKz#OW1I>WXv$i%m zEGAOwSoe*j3BvdF`LuQCj$h&R{NtG$4`pc9swV(8ee)-)9(LV4gVh-*LUFD;?asGA z-Bg9MRUOOA4qdGiqE0g^E0uyL@^}Wc#L@Y1wpblnp9Nf*?U%>0V+lO+A1-O6IhwbC zWOt|1CfWiG8|<%Jt&4<|(l^l=<-4!^)WG=5mt29Z4KBMj+{RWS=vzu+i06S?@J|lq z)WpnQoK{{UsKDZD`^_U2D+Cs@m(trHXW19u#%!m3kEL zs?5)Mm@aubE<489 zd|ma`Y>Q^b(2c=1ot%(`yo&G8uYlHVPKe;ycpR3U7a9_BrYhIOA$Kh|*EhP6YGvr? zb#mjek95#RBOk+=_IPNOpv-9Nv0RX)Ybr&z_R9q^l%QjYosw1_zxLFpzZX+@bfjBl5-|%c`ouu8E=QwKQUhj?h9)-Na zcj`jOh=4b#q+z#*OlqSGmga?`bBp+LWssn=ha0h`Gr33Li-r=R+iotMb-snui9Tka z|JM@mo@pBQ@#=JRM#8Faoaui#E&l_T=;C8G&0{Dav~m?6rPQMLYMWkD@QLien-ZX| zrfg65Y25}NlQqWJDH$o`y5p`-9E_ff0Wyrb!7Lqiq4@WRYK_&ImaAQybrOBYW)U^p z)(DwGo%=rXgBG^0%S`+zS2tqqhdMV?5lDs_R$J9rft8AYfY_36m!y|^u$M&6T%5D< zs7VxaY1LWn_hWQr-EQUf=6=)^Lyf~$S#db!orrjhN!m}TXN00d+ zAbfi9N?Ij0|3LLRjy#F)JkmM>pV)^ z+&6<)1n%|b%W=#C?&wz65U}T@k>3qe{R@@o?mR;)v$1Bl!}I?q{($@Xd2w+0Wke6np@w+ zm)ib3RyG`aCO@wWLTbzbU}r6Lx29vpva+p1z+N)Wa@If=1rz~FCeI<}?5a?6v>90y zI$6>{Nz-!KV7MgR8oo4A0iUT|eak)4x`qVOdp&sAcoI=9P1kd){y46eF?4S^C0GM~ zQPLjt<-AcnXGoy?T>*dd{-mfr*)Y2w@Vo%hY*F;;7_5o1Ym{o>pKs3pt{eXmMF06! z^YW34V(!4%&>GgF^|Sws&3@g!pfaG_2v|oenWyBSGKNRzI@*Y-=fs!RGSGBY7QKb%*)m8JVksUC+-lPbrW zdeeJl_kpE_3Pxu%16-)i)=gq}Apdy^Xic?FDi{j^<%RPMoV!wGptc=8$sF`%Wf*}8 zd;{^0o_ZQu^&phX)7e_vDc3HKejsvpH3+a@IyOq7J>N}xePZ>V>}7p_k2_pKOG~?l z=rLUg7KEmnM)G#IR6ipx`^l>t3oZ0#PUe}uBA)UgkvD%+-)Tc2m0~w=(=3{!UMqu1 z@P;r+?g`y&4Gxnsl(#qCJytRId^1$Ss5b&~LFD0SX0fR|ACCTwn4rlfdmNR4ikb*O zVBmZ3iZLy++G68=siA<;p9M7fLV-XWQ>xFf z7JSL%<;)%Ao>4z}J#xpiVG=}G`rLE5_0}}z&;`eZ-0pO3ZVQ+ZKKI9SE{WaGnNd=v zGb&Vf?qZvp6}psGHlyo2>R(cfQja8XEr?n7e=(tjZ>=u>EjdjrW5gSNxbQYs>}WQk5$tyIY)AYv8Ej&ZE9#y(GyMf9A6w(K~fXpqY{V zIkn;6Sg`wS`jk(aI0{bDta-21A75Q7LJb)cIq5!oPv(}LtJ!?|?QKg+jVIR-u#ljr zAZ!dzJ86>j{p#&wLh>zoI@h>4L&g@810`syh9&Vx2xay1OUpw$Fil@Q*;e2l@bUHQ z6XFkKmupngHo8o|@9f3hd@|yMGEd3PWRRMRwZDIK>8i>h2&BHgp5LPXrv6#|Y!?7^ zR`E8f?4=|K#zB%~4!?Q-o<8?y9xhLODSu*n)N|uCQk%o=eQypVI~J92M}AFjXsFnZ z=SgUOb0lBlDPh`M5@u53C0?K}B5s7QC)}l6_iptkack#9#Z0HV+oUOlPXTlE5K+#p zJYP6GYMhgBOU_X>J4_n10#A98T}T%tp)66ICG*?sRX>Z*eG6(UQkTXprq9;o7U&~Q z{NM`rvTJ=rJc`M-l-I?g?;iZNpI6y`P6BCI;;f@U0n^g`EhtP9FROW1QV^HPmG~PR z2xBV^&@}}|f;=fAvnM0SBF#0|$Y^X$SLp5(1;~6mS1stcTH&KrTpojgDt&9M>GNwB zG~WaJc6e!4NkG#AI-6^A#3j;FM6#eCbPJjMS;2UChlEF#elDjIL+F;}(Xc`lL^@TE zwqI4{AWU3+Fkp=r+XH?E_x-jQJ5Uikv8M#Jf5^vrZue`|NIwJWm>THx@JZI$K}|c{ zKo=zncURhn-Hk>%gWW?RI~E#FkW4>PMVVjH?d6yom1(q>2Rkz!_I%Q8C6wdb`SV!N z49)`L$gcD>gBcb*fhQSakhJsJ^ur|{RRKF;&|JOJPcLWi-4-C--7m`B8iBo{zrO;g z|E55FVf$)#YDQ@Qv)LtY4zg>Z7rWuGL&*V|Ti%W9Xs%1&rGABZ@J|aRLwTvFDr=VUoO<4z z!M>m_IE+G>=6ewIJ;Js_X2H5 zZqH7o@`>cp{7%^er{sLjRsm&yZY@*U>x#&X@UGK3$!C0Tm_t_9c76GUiZ=`An|JSu z>9>QU@sUy@$;lpIvck?czH>=s0fl6qvG6N(oWY#>rE$wYuKT>EQ4dA3hFyn|Vaq zp;w-@rEt0p_mx7WW!pYO<)h8a(bPZGz(OuktVusO2NhLWj{E&6`Rk7SzqX8n3B|qD z8PM6wu$^}L;YH7_b?w{9&uf<81I2;=f^4B<-wkoAe3%B?jT=h9j^g4#(5+#PtGCul z(5be8r@Qf)8NAV+GY`(nu$fYQO^~?WWNJ!hsAZk-R@wQS*nvVy$CD)E-5p@@V0H%* zLsqi7>YkvHc_=s*Y3v&(q<-=fFtLddhCZ{{d%m%F2=Z!cYN+G^ih3Ge`J0 zK@7U6!Z!2u4ZChCGW6BKZo~w|OVE$3u5cWD#U4psHQ=dtQh!eG{m(38iKD zfy;pYR;t_62@7uxx2@O)5XhF7azRrOr$=MU9|xLo47$}ewsk}C{k(WP744Sp#>$};sj^fBGq}bpY*tX1ECAFyUdG- zHnITK1I*!#vDjk9u(FNo=5vPT6a`9kPeH8%u6dx@c z_$};D8#kgHoE!S5va+X$7Kz~@AtC;Tt`LN@*ObDWH*Z$v(MhYh%;HM;Kvp#Ywl5P~ zXF|*4%1Hc5|*M$;Gj98yEF)6nyjnHYjMWx#pup5|)CJw*f;KR{{#3QuJvy_28I;Jy= z=qs=qOv}j(sl^!RCO+O;z$4RSe2=uaHDjZ8*Fhc9c6#41q zk5~EKcD+3HYARjm5_r!)9ks3m02BQdLvBClt!l3dM-6@ft<(1(3<=r^Yj>?b9+~S| zeMjZE0M|k2{1`(g7hQJDJ&kl((5YAnVDeg0ruBjz8DngAYxv37fR5A&L!TkN4d*t; zRl0PqrpJ2SBX=;CJS7Wh()bioU}x_UdzICJI4q29kWMD0AEM5+il= z(8muu#z24r+ena39^bR9fim0CA?7xKO^$;c#$t{2e0G_I_Pi~*jlG%fJ+tiaAm?gl z#|K6w(jREj<@@gH_*AV%jAqJDB5Ze2K;jY$iydMfOf<1E*@%d<_*&@E??tL|s|lN! z#M##GA6V^WzcVtj3OFQv5;$yO%XkTr?~r_oMM#BVZ!OGnzeTTCPO!WMU+v9wMFN=4 z>O+K)U7ca$4^t_!R=P?oXNuIeLZ0QoL*R5dnUKwt_>P&I@MDLOZIRq5+x^`?CjLb= z=c)QlZx6nT zF|HD#@rP0n=78D7?~tH7_;}T6Cm+>pv^^2a3fuBMKJ{VPJ`za%51B0NZnx=_?W^zY zPj&a3!{qiR@0|i>LN$@)=IumS+FN?_;tG}n6}~dm$0!{6XOxU&WT%Jr`|a=Z zs=EPlK}%e_LGV_db{SbPx4LFj>sx<+wmhxyiyYFRii4itjE8|=8^|uujdr9#ttQtS z;nMw8WlW6d!?1RbZHd*_#ddv4m7`9{(P(X8Js`AztXvW(GlEhl&JnCS-pR$$Hu*=&qzoI z59yITKpSJ@t=@!$h{(XgJda9=jji4r5~>7RvmF;dJ8Nm730y^H?L6(A@k83%Q4Zyr zgTr%99ev9cI{j(`0_D3qMX0w4iTOiB;1!5-pee#UAGW1ZFCl^H-6zkb+F?8Ukz&QC zrlzCMJ4)8;hz4avFOrhjtUUU(JMu_*iFdzI(=XPJ6$NzB3Bj+)K!*CcZtYE@A#&N@ zpgp&cMoo~Oz{ADN(%uF_wu3+IUrV}LTqy-plY+~x%*8;l`#vh;K+vl{H_+Od<-}0c zVD8%)H_6lIg;TvIX@s|48$cUo^)BlV2RqGdtt}j;Q4E|y*^_E_=DCSxi;wj)hwJ9EYqL%V+hk^MT-~}u*W@vvSZ7y*6QPP%P-q;_}`lF;H!Ei{CnQpt^*d27O*rEM7 zM%cLyb3te;&ako>9mds`P~$oM16aknNpID$pq*lcZEt92>#U3NNl6(2_$zLO;^g_( zLi0AS?;^*Yb9w}f4Z+~H%JXtRE=LwamwkUpl#^%pct=28V?$$FSb=JmX!1Jl%`wzE zK|M?KFid?i^?Og?L5in8@^hIaP&^-3;*W-X{`}eOTWh$h7C0WeKdUk!ExvyHHaNNO z*q9i}OYaljg{Ycgyl$Qa_2IXR+sE0i#Q{6v0rXU+rfuJ%yys@bJ_yT+Ml$d48lsl? z0Em^c0l?8J2oeM6a+FjU@K9+5z`KtCz*K<+)> z2T)5_2D$e{EIweyD-ynhw_x1YwMDI8qW1GLFhUIjP|MdkJ^dl1@nl)sFV(VDu2SfV z-O7dg5K=?Kh?5U-U02k3HpB*?4PfQ!ti-j+V8EcF-WVCFa8h<~$d3aCf?{KiLB;mN z{T6CI$3e3E(`m~qZy5V?&Vh~H$6FlwgHbSbFHAjsmKqk8X_N3-@q+)s7q1TX8=Jd# zLOvEurx9A8-JipDy5z&E=V<{v*BYCxgNOi=k?RKK4xc9p07t1SY;OmtVfB6t_}sjF zwBUfbeVc!#NBtXQc_jK~@qH>P+dieCs7M{D#HU$lK9)zYjJH^&O7b}M&0y#Dmw z6tK=Ny1NOo=<=qX%ICHH@%`&*Yt>KXf`*bO&YnGYBJz__u{%_Vx=*Gs#12KtRPgb zO(P6nyO8wd*@8uRw_nr&?!~!yF)67rgy(0jPTzws252gS#bA+oS3#Jm(Cr$BIpS7_ z03SI8N?Nf=bH-lEeL9w9C_Xoq5(rojPNzL<~w5iVZemdiR$3Zh6I0gu6ZB-vryP(BVv&K z7Re8W4;8Mtqv+1fZ5pPj#;%ma}|(YL!P$LbU% zZs5w!!55=0NqLEkqfPD|xc}y>tjRW<1{=oW^8_!jFtFJ1MTJ^pJ6iQd+*5Kxb4k)M zY(y|f8_omRPDN6w?Q|9JdK!iJDB_z~=xC%Z+9Fgxr=-O9>_fP_aS5Iy@X<~(W^b#J zGSvd$DQCF>nDyz?rz1*Ou{>pLNB38)RkZ=JD5ZsuA3qA+ud!@7*A;1`&z-z*NiEwh zD{b=*JxX%;l|dO+tlP8->>qxcWn+CsBi3`BsQr@(SRAm3+Ps`m(_Bb?W&pyK(=}tf z?Qro&MPLT_=+Pto1_`YszM76p1j&5@R9#-<{ewT!k?Du%Q9VQ_x2*UDb3n7&{tc4r zMO@V=QKA?39(zggS^HawmEc+iX)}_G#m7J?;-*!w)i*}~Mr35Xu@9yA*chnndxXLU zNsu3_hT&Gk?U9Dg`ms9S_EGaRN>Gla$m6F!HIdc|NFA^Q{H6<@S3L#_JgQq+`k{b< zQ8Px~9OK5nzclBBJ6U6g@>B@M1<)`Kv|?QrdJB3I0h}3l&>k}In=qByd00H}MgBpSJ(j$!cmLAlQk7no4pa{$>rj5F&bbhs;*R)81089#yg$wyyfxuaa8V? zy!s`eZBgaBI`wQHBC$AHr3)$$n?Ax@`WvZP{H5Nu8%x7n6X`LY00I2z+M|2-esFX2 zRr5_%hB;B)N9w%NjZH~E0x$S1m-Y!J!v?TnzENUwPP2`N^vtCbVdY5<6AjnJ0>NnV zjQgKO9q!iN2TN`XmTRr2emp@SOXg8u3TG^i9jK~cn>-;>f%r7jTWB7aRs_7!k5Nzz zW?|K)3B7TNlAf;D?X$8LxOr4%{TaXX>so{T=VpzLiCAIWs15PO$B8;A0XSWEp>xOI z7;n@eWI`}0m;Pp7U1mH7%r#s%T$y8FHiUOuKG$bp?F8~nWk9}v5b7ZHz>ap~b zrIQD^#&TzhZXWl98Uqu)z7(X**&B5pAM^e8Z??l(>RFT+Ia*$pE%yi@P=3_T-U~_T z>avRe`7+UMXN&4dfn&+n_!;3bgJoJjieK5XzhA-`HeZ{=F-;aRBPB|0D}v%2pr9-? zYq|1#B9LP>$UGEbMc-mn``WA}q7QrN&wvo}ghId}FT6w0IZv8UJcp9en>qLTt=P?z z=l5;`LyibZ&ke;q1JShmZ+<#zrvzw}so9b+0&_6==g*%zYA&elLrHNZ(t9?=0+pZd zX7N?P2-2(AO5`YEsLYlFv^5+yQ7)$I_E59o4QAJv@p0pdG+8FMR!}dt{)k^HttX!2 zc-!(Bc83EpR&6KE^y^Q0i0JdR^|hZgpW!oEtXQND&EEWil@)aY zpFyEkz$wo*!*0MWj~a(Pa@#Wyu1%0o<%}+|YK!DU004Pw??t;3R~Pd04X=t^0uR6Z?zfgC}r}<45V($8nk>9WWqGBB}8oP6)y!1Lm4FRW5(2Swuw%0djREWI_6ZV8HfW=s_R)@;ObO7A2*GRt=#WU! z&3Knzp8?C9J2I!UGE(s@nHR88Di*45B#OhU|8SLXxJ_-D=0whA-9E&?V|sFDeUoy| z5es*m@4fDpyiM$@%n1NI#|Ix+ByTVY?_HgkI9n>cv}SctWL6uDdmdce{cgM&Cu0jF z8*~|dBwokBrS<3gY7Fvj-Cb+%k1=pb+u0fwUTrH3@2>ntqrqQ%|5+79f6%W-GVfzx zCwNN;0hiQf6TZy)D)0a(w?nF?r#>i%!bKiTsZ=LUJN(f$@Kw2S=964uz*NNIH1%*I zte)fKvu>se=Llx#uKsiZSHe1VJamAB6An}Wo>qn@Xkyh9R&!6e@6vbK+CtP5c+n8F z!BDB)#D|l;4=msOT$MpB`Nk~0Y}c&PQSCYiC6$-O*ZE|E{3VHNukPJrhb)cgC?9pd z?7z&LV1L@}58R0i=lLz?o%4mTHl0mgrSMbX1*T(RpXI25B+FI({v1avuB^4guxh#n z2(Q8$G?=zDU2-3B@weYVxfJ?-Ys;Nz`a^H96rOQE2GEoA?psH9x=$<7YdBV- zE`XtgcO&QK7QJq7Bp~dRliGB1A|{)Hhso6Rt%^X57gy@d0NpjXF`%|A9zSCM-_2nyhL^Ocjc)1+$93T&4@_K8taB-&k>lj ztn8wx)TKF>o0;)oloqLrxo~~YZg}ZV&IhVOTA+f`i_mS#;LALUj9RYI33VHad!jj2 z7}FPTIJemzRA}17TLrLdfvKB~l4uc z6?Id;wv~_%XaiIUi=@{hV`(YstOrOR+18KDF-oYD9d{B7Ofgc!k9Y-o_YQsdpLmdY zrn9b)jt$)dlqzouLba3~13+w!mV&*!#DCZ+gPup@>!5@^9-~6k*_OWZup>@%(n&Sf zV2t>|bGly*2j4!&^0;`*0yMp(--yG-On|1W6AsdysVodzcwgULp?Mr>+OeKwM5-TC zZ*7p1CaSn>EuV$;9vkpe`9_Q#F58#`MLlAq4|lhVdSQ16$lax4diYwwL^K>7STkjR zqUk@UhX0yTve6%jzT%$nmSl46uOuQQr*3;aw)(rf&eX=dLVg(7fnDfV){Ldz6KTmt z$ppr}Y^9>~_xndAfSHa^&P%fpo`focIkFhjmg_w%@U06Q7QReQ zKA$E{nlQq~)T+)Kl{iH~Hy0=F^U-p83uq@OPClTSE~fgtSM40VD1Bxw;kMa;J{Zg` z0E>l7OSftEFg;{sCu~dtma2~yPxUCdagL<`lCBazC-rz4=#DwUa50XX@rEOW!GFEH z|C~Mg_MQYKNzFxzKXU{SFHYQ{Ts!N?Z7N~4crlL6^~K4Kp4qpQ{Lgpf0p*I*KUHMkq0UYs5A#s7gQm8m{D-~cZGy{VwUdU0wxrXZ(QuQA!X?DIqg~0U2*j~@o8yQD%k2e*6CI&k(loPC{ z*L4F=IfL2IM;?74Pn+GB9Q!>V@qQ=MuFy0i`fm`XjZj@+85D7fXJ0cYB(n!ulwobY zdR6`dW%W>%&JDDu!#%7=0f?oklRUQ5wsZ_`L!ez%RqvGRDh0ICGL#cMRv#fcD_jAZ z-VjX4GS+-%rm9B7=h?l zI#=X8UXMlcdgfS%w_-tf3`aILUL7d`9NuoOTY2%o@GZPQ*X~n`rbW4_q_FokQP)kl z`2O?jMmb+-!mEy>?27HttuzUV60mNq3wyf2axBbYKoqB&tuvS>Ji%_OkC~#Epe2;! zA(4nr0GuXKx{+HxW zfB%l>W5Hby!qH88{~3qap7;XYSlIi)3TPy{wums_>3fcu8MmnP__4plax>kMck(@o z5!~9Ch{qbzWwwOc-=E#1C3XAr=hIvsMjFhI0()GIG@R1Y_=*tHJe%#J|Lvsvr`*Tv z7^n&_PG}m_{P`9KBML@gyK6vJc^@o=a{*`#*(IE7BzumrQHc@2eRh5g_NP8S`&v%Y^UfydW|@7FAoWfzd22AO zFZ3y9k*e3}Hy$I6*52zvA`C>GI6cYy=1N@R0f&G2d$Wa_`iXTlXsU}gTmp9oAkuz7BikZh zyYSL4KWDsBzDYs&i1Mp{y@L(r3ly`K1ZTfR2DBDVXIFoKEvY1k$v*Rz9+q?lY%ztE zpWjBvS6-b|1gxeJGK|2`0vszw;TDrkgM@fhQo>_%XT;rS@>qmkNuj6v(#WfG3^V%c zUO-H*&|yV=sobAwUog${8s|)TJEZ*U=;lZ#tai`j!{Sg)6JePGV>gugajN-aqhTP{ z;wOs$6PTg#n-8fBmRkkINx{Dt1=;@fUH|hD@mIyF;0M-1AnEyE^6h5cz6`=f7TIud zH%E{ljyzYVy~w}F1LRiC9l=RGh}?TYlOny<9JYFY4;!Bx$^+GzBPgi5VFtfEPOdJx z-qrM>B_AzcK0z?OeS38z$j=-|8;;!09JNHJzw%T9U=~LRLh2lBImGkkm8ZxxujK-E zzy2RF0uhDg%2im2y`N_Wi(x@k2&T$@PtT2rK#$F^?3l7ooakald?R$6VxT2b7@=#v zt=^>i$=}4=u>TH(?d0UeF*>fkXfNr|z4McI} zshm0FaTJYznH4+?m={a9saOlK%nllI!Ug-<>Fw0@H_;vyg)Q$jGB(W5+3%LQj(X_N zRcU-%EnLUoyfZpt1QlwxY>J)dyHz(>Uw^70HKdvk^Eyt_pcO!l{R1{^Mts}nIQ#IE zr*}r|up(d|jFL~*tjve_7CHfIp^9xc=+MGMZTD>Jem%}ZsH-y}x(q6Hy~U|jRri}~ z6g?2Mh;;kq+j0hM4bY)~@%{Q=;pO=WfB&h-i<-zYG=3SW(;e~3V_vo)Ye3`-Gg0PJ zD;FbVHPz<^s=q<-WYQ==w<>@fAZ$J385bx;S0_k~a%fyJNE?{yj7qj?nZ)e-iwh(M z7K}26@t&_%r>BCbSe18oh&Dc;uic(dBs0b;_+5u~^HjXsQt#busoi`GjJLPmpu(}$ z!c@M+Trj&z`7vmTyc43O@ml}kjOc{rwplmN(#n@l9wcQb-cK1~nJoV73DwD6Nitat zmVX~&Uwergob;bS2An9){p?1Bv{}Ol!6%N_ZOl7c9qGgda0No0WM^bkDv(y|Fwy7P z$rBfAn^K${$uAp@Zha2~QCr)sAVc|H4?BJKu4kAmg2=&3r@|(F?HaVO$ohm4IwhH8 zGw9_*rH0cJoE|-kXn9>009;&KP?3uiM8R~Zi~Yf89e71Jvw@hJLWkVoud0AOLSm4#y^ab zvZ&d$6nJM?dY{V2@Z#zK0B*8NJVh=J7NID=%c{PpQ25yJdGq6tE08D@^0T8hIA zz(>*eVfx>|oq*uYD+G@C^TS-N|9?MyT;t5^;9%r$c43|eZ(oaWhtNx=V7A!O2KTbR1UWgwBEI(_QYueZJaI*gj%fo^lp za^~;Q4wYpgpIWa_X`fwz_PXo)06u_9u)9nRq= zoxYy%Zr8U!>epl3AE_pr%n^3D<_XZGd=@R~mUoj#w0QISbqL6iJfFR$N`Q=3R{#ND zS!((lpBe_e0^>!U9stqmo0yKc>-)`&*U<~_(MAO3sMf0{6BDaxR44Yd=iF>!B7FTX%H-=jFG)DCzK zTg_+p`Q}gZEk9OEJEm&o>h(bl9K|!#;tPE@fP`h}Xq8)4Bkg&r2 ztPYIKacXB>w(_gs$wD)HkB!l5(ibY zWY|;Xl(Zh+J9Y2={SeB-egL@4fA@^v-Yy=$H1>RKvAsH4nBa@ z*=Gug369gPaNG<3&-hvxA$@7lrz)ENS*jqnZnsOa`)TF% zn<_7;ZXM>VI~H|_&iKgR`N56yu#SY=N}RIP$}vIffsoY`=5tdG-J&@O__3rLYDpiNPFwBDBrGYSPAKF5m4z80g(N7<%ZT`7ZDKyPxm*j^q93y?=jm9K+0cUFY6w zuf6u#-v!U|V#cf22k62Yj!)aGQY8RAfk5e=8nDSqkGcVuED2#MOs}C2Ho{V3%XC<- zmeHS6P2|#h&k|Nif%XtwYzaLf?sOvlC%*ys-ApEEWu6BstIUd!FO=)fpXUJ}j|oB%i7@d-Eqah~b@bTW)VS>ksV>2X zlD3OL;~kEx!5@-=IvSSFuKD{AbOCIZ=b|IqZag1gfUvmPz|vTWfKgl6t395Mn_QJg zZ2I}w%&O^sCYQG-3b)swO5=2Qr@^Q-HPdI3hRv?}k4Gb4Nv+b8Q7e-3sLnD3yk~gZ zuGZp#EXe9RTh*zOdgy(z8jgNbgzop4cT9>`{O%Uzdb*|Ftfu_=XDTxuu%_MsbF4l9 z`;l?LNt9T>OJMr_k=z_mu`OUFr3*X%^lL=UcNiZ?z8DYnm{F{H+&N<$phES+D8E-) zj}B$j2znO0)|`+|iqC`dIXN(v$H8cOe>vbDij02pee|i}TrD4W%%AtV3@8E9E;=${ z)l3oDLD}@}u!LJ+^M$%^0M^y3V~PLr&HT^XOp5mR4LyV$G$As;*Gq*vR#+K;9m~& z%>r)7MFWm@PBjoYa;%(Iph#COGga8QM}^0b!hsS zh5AkLdmdnATYmf69H>?|*!7H72t)_lw=}3H_K(F?6PGoYi%GoxqIO9q;`vg$Qb#3o z<~}6%qLzk#E6;6r3)yi$Evlfj&7BpH`B=b;*wKitjP$0HL=jPA>=X6zSE?-FX7+5c z)l1XsJ5u@U<}l6?n7#V`YPa&0_kz$IqYl$cT=Gqdz<(aWe;r-MFb+Nv zFX1dSyD5ETl0Dv-^RJ>~%bcO0Y%hmchJ-aZ^B9#S7r5HPStUk#exUkQ(l($&5Ozk2 zFHWFhX-qFdod&=hmlaihXGf_sChE3d!k!^{{>NNPrx#P!{by2lun?qm_(T7%lvf{b zJd0fkWgXazsW|nu3eXcAw^Hbm+K!HxVW+On$nLk)G{( zSJ9~J%UU%_Xp8t0Zg$<>_tmF;*sI_TI| zC4VP9U!R)hgou*;imkW?dP^_hD+E=`GV%`MJ1$HURO&nCHfSK_O4BXX*T%aW;_3L^ z0=TuUi`3a_7M}03d;MYx+MTJKxWUBRYSS*Ynobw~tGh>ZU!_3=o6DnTG)>pe27PwmnNKT*TMDTN#_k2x8qzBdETB ztp;mO>%LB>x?Wwl8w&8fQBVAp0vBHq^#Z)0w!iMrLuFkZ*>vm6wwXunFBs|W?|gQ+ z#((EKnRl0<6FUahU2{s_GI4j$kQ|kINXcSA&w-_k#4JJI$nZ_n8E|)F8RcR$j7|4% zfG7G&g{_l_{-qS=I`=;edC7qUL00@fU(^2?%zVJb-*d&s1VBQK!D)%!j9px)Gj>^g zlJ6Q0!P^q8;v)Ig?qh1jArl)GV3Rj=Lh|Nm24MD0mF<|$rvALxD*6PZr88ZwEoS^_ zo+Tm6bgU)3D!BqOvPuZwwKr@A0h2t%@RQNfr>3*rxR=y9LIk^uiYAx7QXT%s#FT>* zg-UXMfVZL|LMY;uj90T#q3SphWHf*@az+PDX`MeTSa?kvt#rAnV!1+v)(dwv#|l~4 z8F^)v10o8N_VVbWQA7|g$OPS7TK86&j>AN`T4Tr+S3zF{;eNk+IYqU#~3q1Et_ha93%gmg$?m6-CuI~RH z_5Ax{5vM=k%*3+%uxsAugCybdVpWI>{Y*&zBzwb#_K?sel$#%u#hQT2zX0^+1Tl(r5ws@E z^+%?RI)PL3T|p&mLQN^`A`tuHpMpc>P6%*9X=YCD$MeYcJODl^jT(miI5CeA-{e}M z+p?WM-%pi3oJimtcn*5QmInkcaHesCsSKA#`p-lGPN;n5`{uT@5z*5tT8G~hv_s2R z^0)(Ft$`*0Fiu?C&WsPI37#3PeUNMmlNRC%UUnH3=ObVbb93fzy1%=wn-H{L_bwFB#Zt7`H8lWHB0Z+Ako_59b{^ z6rhez?pczx%k?ycF+bh^9VEc-Tj8mqdK!R7#Zxd)W0BW|j-_$%zR4_1MV>XT z4;N6oY@wX2G#~s#ni*_&ne4!TG6IN;@*{ZKm=Y3+ ze9tK8M0}G0zcSgvfQz6*-~Mm3tO>B)$r6fP5kP9dbOWr8ljqnw=jq5d4ud1fSk*KT z_KQ|stzHlnOA0X^CQ4Sb`K9_p>dgts+fy>AQrh$}2Vx{+Vj-?jK4;%&7Fdp1thIKB z$O^!wgoofjI-jxoHn$&nS9wZEDHCol))e|m9!BrYcVxATc!DZ4s`mC9aLDA};nUgR z!~Qw0WWfL*Enq}W6-$OPA4uF@PY_v-fO3K{trxb&AZ*_Sx&=tZZl9Z*et$eYg4$bD z`dj&mm(`5e;m306@p&u0?5ngzjMSPVR38g}#9st9@aea_))doqlHwILJloOIyfL}F z5jW-bs|N{%ubl$Z#8#(YNMN>s+MmGZ^`!ZNq!@~en8%0)Wf+iGq4SF#%PPg8!S%** zS7UO}@k_vk@l#&!2D&n7V7|Kv-Qw#XPot}>jLNLkFKs7EDEtD}FrsKAGgZIrv!=aP zFQw#WQ`-FX#>~|6|8Pi=`R4*Z`3K)d5l_pR93*87O+^ zR)Jtz=XXmu?gLmeofWRxF~F{M(4aTKc0_qZFeWZ&tv8~)wOQi+1F*RQUkI4u>EgQ? zS!{v+kQ6I=nV2c#_?xqyHpDXE#>r;!2h?4kz0;P}GyET3%>U#<95EcuKQf;Q@?^q^ zG=W8v);E8u$8paAFr(6+Wx#w04CzZ#pM{KbF)7|m3V)jN+TL1WAfG&2)g~@Ifh{8U z`_&}Iazod=UIy{K;>wem?gK08ngXnn)RgU1u1>ejP1skesqMghW3%7~7*6>>3U%3C zDy5n#F-W>1c*53XMaV0<@)dzU$jm~y#SCmdn_Hb9O0vDa%9uiSIFt}qqR4cwug{UJ zH@`m2eWP)`gi&m3HEBS>Mh6}D{$2-iw0%e5AT!lT;^}o$S-N=icQLer)Ph0>|*-p3`-#?dqrhvWBs$8hUkMe2zM@50-z&()Fsx8DkZxpXD1)$PK% zlTjS#c6*zT?Le1%^Vyz4c*mVcVnY(E%JUnDZLM+luaLNA z-lGIIqx34X&T~3ZtFu%AjM5W;qsvhsV*!olD}>1I!rG+nZcdYuiP2KTEO41gk?&T! z!?wGy=(AlABV+bQbfbW|+jb_SkNdbmO%m%R!#~DE>_7q&?CWN|ed-AGN}Aaq$ZAC~#3AvG6j3j?89{eGZgT0vCG%RPthrvG~DKog7hd=ih?mYCT3WpW%iRAjax^ z@4WV@MTM4tDR3>2@pTtgCw^|GJ{_HFQwqO)@S6wmk8Tn-x?#6c zyx+}%@Hp|e*bn)P|6S(~gZI|6;=_sxtgja#kMKd^trq_kEBn6+e{AWe^_$kQ4?J>D z*1I)O_1p&Um%Z;L{IDfOg9NMuoeBXu@%#O3W!p!OS+Kq8_+w5`EE#N0xV6iyQFAa3 zAOU}Um(gbpPdKQ^5LJSkf@K(Hg0N~wR||MTg#^6Nbkp z^)Uy3Py6QEkgdLu?+4;Qz9ZQjYJ#sjAoDgYB=UjL<(uR~gXf9QXAY3X&l+x^2&hd1Y0T_fNg z6AQv-s{Nq`o%zOdysd0Kkt%#7{NICuFRW;a$f4qPK?AXvLae@|vv(^8Ey#k??jM5w z_Lz#BdZDCz{yEs>Y!l4Hn2K!c0O9ruG7nMQu_vF3qUz#@L3s0SXcpIY0CJjvR8Nrx%`i#-M`u{os7Wep|E2^`2T56{8ul> z-UBVd9!0xxZI0%F7Nk|IpLR2|_jREox!? zTHyJ!Zi1{&1;AS3zZ_-J+oR{-WAo=l2!Nq~s;B{n?Y~#>YIN%S%xE-IR1FwLNpO*`0RCm8>u>1bSM@z-~?2m42*EjP`}@UE_0>5ycX4%eEvl=#fe(#T>SDY%*;zRwkB*T9=BDi*^?$jo7ka9wi zB9%{qt?Si%J7hTNu;QqgDMmA#As}17%(%ZKEZ&84b2ejKd){+4&Ulr|XE{=Y`NL&G zEUeMxXwBlv?3z$FXUsUrACLcXbDr^Z-$M?n2|)N_`~AlsmoDCmkqX_fWA6Nc68X6u zRwYIPdj7FF?V;NBc~jQ(Wfk8;C;iXanfJ#V)~+X;#`=6o+?h)7EDpw)v6#Bcn(|ZW zdCJK`OR8Zt=~;!XZ8f&^5aqOhN>kc;*$TA!|bcmEb#rwLA}^~CaGVyiLT_q zk7Z6{zd0~0%*f*hhV&6Z<=+%*N=gH3*7M&+lxY`!J9_CMKKSSQs7ndD_NN-Ty|U`C z@eG0oR6e23KNkDp$xc;2`d{blc@B^^PQVTUS|Y+3K`Y|AMceGXPs9-`NklH z^S=Q3({3E!i!tw10TjVN@vjT;=Pw#!0S;68`p28jKya1`))X61p4TlLO~jqe@@)=t zKUk0fhJc)JxJcs$aL5YBkC18_>z_xf*SwyHHL9%nrM_hMejc(Mi7=+SN%cn2K1ZTu zB)sc8!HF+`4L$`D^M^hctkJ6zp7wklI}tK2AF1E!p(BE|Htcgk73`uh1}T?Igc4$1qd(`g&5gAB!z>zDfAo()c{kJg}LmJN@AJQu9#e*6d*SD-L|I zS@c6=YrUoM1DKFi%B3e+uaAG=Vq;#hnBiOg4qPh0jq3MIZlM_5I&m6S z%1W&!E3hIHrB_XeJKw+h<*pHgORH<%-vxV|K)=NIP#id9e7X5n8N|}~&*aRYIXSP8 zAGzLBupLO{d}vv~L*j*wqvhS-WfpRCb)s|GPeyX>V|ab3-98+}NL>^JT_nAWj`&}( z+mtyF3L2~afC_P}D&Zs6cHzd5c8U%FWiMn3{_>!x#;{M~ zV>f%~(#;g4NRbtHrd)QYF<@Dm(6wWDgPn_WF&J?|;9vSL0C6Pq^gY~q-?BuZRF}H5 zLLX{&v+%_Dto~`>!%r>kvhV^(!jwwSvE5L;q1E2SVCs02^Sa#!l0UUxr9l6HV*cFs zl08GLISYW2^m;h($=>HH`S@OzeH=6gN1ybpX5+kE*R~$cUG}}xd)N7_q3TOIkAwYo zq0Vp#Nueu!S*6JM1W(toWn#eIOgMeb@3)u-26D>xVd-M??>N>kjyZZ+5dp(=Lwx7cC_rX zA=TV`Z*9C8#SN{ta`Fgs%U;n;h`CP^?@Tv|LlsVnKGOZ%VjYjUZ}c}@Vtw4irTicR zNnB^~{2ZhGPrIgfpc{2Z`NX_&ieZY>C$t8c6!Pne3uPvfz_sr7j5Hh_bW^)QH75<^ z@rgH%z>5xHr}polWein8AVuS(NHxMH5WT$@z-j)ziFKdL&p6ZWfpgTEr*t2Wz*6!Tk8Y>w6) zdz{-=f(yPj8RYW1VYBkAV=fzCSEmcb0un5aWXVe3Ok{;`R<`TbyKaGxd--Kf@v7%O z2;xREm?Il*1GxnbzK(@_R+V(Hj4XZ_R4TN(W>NOPl2M`qNd(T2&}zUL`N z3Q8PFfS-tx27gU!_dZKKHLR6?LzVM~lMI^yKyuv!iRRHP4@wRFIWzd@FPJ+o-`yM| zN&%GYXh{md3EZzqVi6Z3zJ+Vo6$l0L!ap<63GDa6^3Wk5o_R!BP$(RCH z`V@A!RtiZKr%Z)WTL~sky@(|fM}}>LM^*gq3FHbTlA4Et6B3VVUTZ^bPnWOg7>{P$ zf3NQ=0~u6H>nW*LJ#Mo(!y0d^0N2?qY~?&uL^#UFZAf_M*x}_UOMtGfkGf^A4%efc zC7<1Xt>4H@qU26isM;j#+TgJO5AGQyx6n}Rx~#5Qxc4QL&HxQIU5-i>zCWkih|_E2 zJ^bk7>e^0v+7FhB@!lB)For#X@8&GbK@sZ0Hp`ACs?tZ z?Ycb7>klUkAT5b9k8q0*^o=S}} zfX*<_Pg3YJ^v(K@^>JftWE2dKKLUl0q`_#dde49n!l(#Y{zk({sAIq7miLbjMpJ3jlp7ELq2j^QA;1()e_WWqZV6v1q^-E`p-t-YQM z50N#Vg8W~i5oWS@^Bi8+PNVkYrKmJkj~Bt+KT;Y885Sm6PG@lCXo}i%fB`4uuv?#c$A=9O+p&vjF?AJ{MS8#}^VwBPnro{3*tkuy#!t{ImR5B~;?!aXKBQ}P zl4$38G|p^u=1ZpM@g@~wT;to5ycv?Ke0SH&F6@|LG&%a+_~UTg?AGupbx7WYL`>u{ zp2dS5F*f8dOmPDHld3fD1ErZv3Om}?ctB}T>-*@+;ZbRVNs23fS6n3`j1PTbbyT(0 zQT=F59JkJ>eC_3Y^CjhQ_d1=d(KpL*F^Wr#LDIf_=g%ntf{1LcU=)#^T&+0BgPjp9 zYDXj$WL8;}mm%t5#PRwg>FRe&X{X+ZMgrtOOJ2Jm>vF*z{X$~+Rn;F4?@Ie?EUk<` zB>|nu(Ix{QeHZGnPDd4{FiphC>|`B7*WUS$JrazZcMsw9m7+PwR^6!;>vQY=KJaV_ zG*?rSi8u5Wn1XuqYP@^vVDTDfxpvkqoW2V_R;PvHH+Z{eFcaIna18vUDnEm}?jin(P%*az8yZs3l|`n#uOjnuL0F_}Hyfr1;3&eZc*_nZ$mQ_|e*#_A?$21B|>{icgB0F^_$| z7#h3DZxiKYlp~jAXiXcaE2Ruy#B2q~Hc?TneSS8c7a;druAt1vTsD%&V~_II^8U+f zO?MbizehGXmGbZNr5a=)wVVj|)x3lHn=w@MwTHd~5vAmUtyX|On~y^*S@LU#*~gP5 zCy}V>pX~wP9%d6%0R?P9(i@4;c0>6#OWU+7bynKfMM?~~vitY4w`!*~#=3=M8}*i` z=pR{V5prpRFvZs{9w^VAzM*7XaD|*C3&tqnev;2Y#KC%R%Tlfvqj{{5Fd?*TjP;k& zbG~0hV0BvA(d-FypVm<7O*(MlD9`7ErPC6sHJ3T);(sskN!E zv7yNxeS!DURJ)-PuYy(E9)qkQ`PnD_qbuLxH7&%=7lV60NiNvtCq4clUTX7>%CtK> zzw6}+A}Ul)hr}G8YX9z(54G+st^J&4B-sJM`k`WWVIGrGqH_Jd8a$?91Shm0XHCri zeDMDx)`I4vA3UPbuifTC?)JU$D^eP$Pee6(EbBU_yz!)#$PWly6BglLHfoRj_$$4T z8(#ceIuDH?c#tQ?UHtIXR-d}FmlPM?6|9$+0~WYnXB%LJdU*u>K3S#^@&E-IuRrf` zdu2La8|cZ>#wR06&%YyUGw~xu%&7L%?qAg@6M0w1Li3xG5IXoH=*blF1$1nk1y_|5 zkYsKV23)+svO~Sthab!949!X(IreTPS2Lrm z&#S@0^Y#WEzg67da;i}$gOKQtKZc|3X9CsztHRK9VvNwY@pwIBYT1M04z|AYkd}pm zaUPGz$bCz)?rS&t-AK#X%{fHk-C%NEHJLwt*MCEUVNtZnE&8*zq>7A1HG%)@SyA8^H_-=iJziDvk_?mkfAs;&|o=Rn2W{b z0@sbADf`EtRn?(ThqGW+60Sn=y*~d6ior=#@S72$nO2pW&Gt8*AFn*{Q95;svZ*ia zL(9#aijXOH;`+^w^{sUk;#gMvFk4AY6^=~W4m!{Gb(=DmnqsN}0WVDs`-gn;j-@xo z$_)ha-l@#PqRHRmp>uinxxlIDbo{*CzWe8VE5yK;BVdrX(-%l8?2pb;1`GjkR`9V} z_?|(0wGEBMEh{gduNkzS(xr;{y*9kuwxnoVSQGyQsw4!+ry?)xB?AqKn5C z@z~{5^SfV^DDIstwq@9-)$zxRjL={c9p-v_K>{S5w@@sD7F5%nOI?_D#j^ssQKt-} z+cp(Hi`JXzQxu8f-Acg{w1T3759B99J!x>i+5!+-FN6C8{PeU+KMJJs^LrQX#7x`e zOWb~jF?VUtRh8>NNddB}&;99=9sS1V_-3Nkb=xHaMLW^c?BOiCskjT}s1p6VqMKAW zEysf$rJi@ug+K?fRCA6plA4EbkIq&j{y6Z>>q*+F5A1IA25g+!Q6_>kZ;{K5F+qPu z7OzPYs|0W%N`=bwrR;ia9=C!uw%4rAYw6At4oQ($x82MRg9@3~?)KB7*4IXjMe?na zrP@zE&V$-z{lq{cQgo(cYYnet3E-HtJVTX;!VvhkdS4+9iUo-_batUY7(_+yKls1v`sN8V<)(U9In7`D2ac> zsYVjecq}A2$80FtAw@UekW1f&g(vgC_z{>zwxg5Ubh}x#e8*e4eUBA_D7B~DzZG_& zg0(;MlO$f?t=?_|`lcC^icL~|@t}GY+8L^NNu~MefSX3DN(;!$@%#%h7H>6bNsV)o z$4Kq_#rZSAt)8ZGo}Hf5eYJG442k=-p$DvN{Y7`Lgj#PmeK~YD6^3IB^@5RYJ~M<5 z{Ydqts`Ex?LZ%eXR`1OW=`HtGzh*hY%g<(X#0BCsf~OQu^;4@Yzwq>!_ebRjMR|0U z1mJSb)p;4+RXH#lWmamZ@3hV|+I2XXJ}3%6+B-X^G-v)h_mmLLoL{~9SKc|eivrDQ zeoO}yj?JKaTz93=ug*bMUc2ZIPclC+Jl|ydUu_}B7Shf@K`s@U$+!?5RI`nWt0k6* z@D2_pTiYqEP`c5R*Y7k%!pD!vy5|^ga-X*va{2r@y*#fLu{?7MwUva_n!bKPZhA`j zB2?F zGVsY70%h&e&q#6a#l4As;M*bsg&ZZfw8ErT!Hec2dogK`WW5b)57@9uG$IiAwmQ^S z!BS2I0O9j6Qsqj%GsmOj+3B<>a?t0EIc?%e&f70OLF2@aRWu8R$FW@6h=%Yon$Hsy z`9M!~bquqa#6ij*rQ{Sl-t<_=(Pg4@MFA9kN9#ydz*2sqohTVsP8tr2h}d#TycSy5 z?F#*Igx;>A9xXcBzKCmKWvJ{kP8t?+q+GOD3S?YZg!o9+WQAtNZn8E?>steE%n3-k z)6G#F`+x*pzREKu8ejQy5eORnaoEV)`9f{+>knUh{MR;nZ+B)btF%N0ZW$SZ+LCnD}2PJ=X5%zvlJF8j@gA}*oHlN#^D08Gwo0T7wq@E^&Wlswe_oK){%s% z2KFPZ$qnMO{S_Z|{_~trXgkBFv$X-Akc&hAdNNo@avRZkm15>;Q2v#Y_4Uf{fE8?O zJ<4V4`ZB$ITI%S4OYQ319xve(i|FLqU>w}!r}+)e(GLaiozoNmxe+gY{-{#>msW?L zht>K`XtMkkrqsx1$0=7)1|((0DTCt{oEsiRkP+S^Ov*G9=Cf3=xJhX8z2f#0FUB46pPbLAH>B>chC1ERP>JQDgAUY4EXeTD zqEu}7YjUr0efAy>ddan_A*D&T;U&wu=7&p`%K9N@RGAqWz1rUUeyt;5CUPE9 zFeJxq2EL2i>xTGo$1Ju<@iI+2=fJNA&Bj58r9iZ3$k5Fc*uFm_Ik1m-$Rpiiq!lM zjxIeyw`C*oqCK@?k8Qqhi-GqX&vkoAt(nGtXr@O#pxvrv%Jz?jYLSj20!#pcLx1)h zW_WBbNlG~_>&iPl(Ss|hg&A%y*wVo3Zt8I>=V(9+Qdu-*v$GNLucg_}WLs>z33OoE z9ewBudDo=B845;BOGio@f5Av=E1z1a>hu2 zn;^AzdO<50)UZ`qW>8pwAAMe`9Y8N;;-$pJ%%zPTfD6L=Yx zZ9J$5^aN;5PFB~T9gmOWr@A`{F#MiybOe!L{A|C)+Me8I*zkXUj|VL~D!pC(_@0S6 z$;6FR483yyZO7GU#zI5VziJN2UEY4n#NR`X+UHX2kL`ST*5H?qS}JB5hua-$J+#Hq z-BJh*y|KRWf1c~GbLaUu;~{Dk zq&$o&BdTDvPs&c~?zu(2WX>t|iX6U>?JBLH0~Iq@mh2*+JiHWc zQz?4ODQ5$kEbN`dupX1m=H5)++rli|$8|oVcSUAU4w&tSL3JTTHX~bm8#y*{YwNAw zO&*guJ0WUG_1$PK(Ae?&bgg= zc@(TJ{ax0@lP) z9~rZVxPZgOCjF?PR&xV#K}odIy*E23@zIz}OKo-AeN8aLqID!6kklX^(MzSt*f$pYhP578fe9VOy& zp1w(N=`t)5W#Mf*X-^&Ii#2+*IlR>v>w#Qo6D;YCPk=Ifidqve)*hWzYG-?tUF&dm zIh~OUXvOjpM03~0cc2lPb*Ox|mWqCZFVg-M?PmvVkKWxnZc4s0dB&n)y}ArO{aAL` zrCMZ3Y%>do1B}=(Obfml=GQpHv_Z>ed8k6cqs^|!by?B2dyy!yc++(5uj5njrc5qp zblDm1Bb|r^5bu*WIAOg-z?`WqoT8n~d~q;FXkmld9m((cCF1$qwUtgxqT_C^_=!ED$dc^c+L&}f!5;rLknOemHy_ELnU@;m*1UWZ;Na0W~(gftM3MY(@ zHu6&ItSZI%Ynn4=#D3zix-qZ`9+D-N=K0^jskk$)=-@M4%TnTRri5WSMKote&1$*5 z*%16_k&9ywU0b*4?!K(?1+5zn`&!0W-Q8VIGv|ou_AOD9BHT64(qnyyt@lWgZiIGV zt!1EJ+g1lnjd$#^sOyCg4c0MEo+mta`~YfXryPIxXSbnKpfT@0*GyK5sS6U=)oV9U8UW+Y`W&c>sb$+D=f`c#b^J!2ei ziF*TO78kl1?OvhY8mjx=aUZlyccy}I{kknNyb%dPy4737tk56v+Y&vs_q8li>r`ur z4!T8mB+@2l;Msq8>w;!kOH($-Y&1vEDA&R#2F{qQ5R+Zt$U5YYIjJ!W7Lx>bNHs-{ zTDiF{#52T!GIk5o%v(N2)tq?e51 zx^YUrZ9~t2s2dGAv>bZA*rL+(b20njTgInxzZWYxdnV{v&Ld*Vty3O5N^4P>08U`u z5|zsOIDfuq@x+sU_w}bpiYVAjNP4QsE-suM)=w z%$ZBdcz0V4yXRv>veMdZ8n!*=N8Y8d>d4@4q036XzT&2^Y?bqm)ct(rfw3*eurii6 zP{)+3=JY#OUH?k`BMs*M`_s}D`f7^ED)nxm@J#~0IWOGA@vEv;qRU{~u_RGIToP}s zrQC-bmZ%%M{) zASlYx@66^2x!by_HKa&Sk6xl4yehzQ>Qgej8KP^cvn&i4xwwzm>-}XEDmE^nUlzuia?9{<>W~rb|fm5>{7xgTMaA+iAR`d`S(` z@y_3U#zVxYa;eAh+`-dMEhwb~kv-hE79BW_a(*1ioq1jMb99XNf7DtyF%QaKw^m!F z>=Ws%oTBZ*ZKhijv};B~nPn`>3!U;1r(J!=)iHqNUIjZ5iWp7dC}-eQk_6n1@-|gy zHQFrmxbx@rf(XA%7KX?x&~gEcQ+?R`*AipmC@T*vys7m8{P+R0!%X+Rxx6M*#9H6b zt4|w3z;x@YLKQCsec<`Ow7r-bT*n^rHtKMa$)rrMgg;5-{@7_wqYF0pjhCopKT(n* zhH3lDax`N?G=A%4o^|DL0^)PqA2qI_MSXoa^~`aoiVWCB$Zl?O{CN`3Aba zY645^vb+P5J_8Onp~!XWu`IS;_ue@kk241-PIWRThS*yyLTh^)%eik&#}bY=Yn@BZ z3Nql~kWNx2F3%BvJ3iBtPhh{Dg|0%-qi0+_W$lu!v@3i?0cyin zisXI3z70pBQl2V|_D_Ri*324urm_ncn~#XLq6|rK&6vmq;q4Ovz#b%0-^_Ua!YI?X zmW8#6Xq<(N;RN`epR3-zGP4S>0e8GO+nO7ro;>2^>aZA2OY2GA!oO`{DC=6+IZLE=F7-Sf9vTmDbNST9ZtE7^z}FznH`)e)rA^^ z&vel^6UmGpvlE#PKdq^94P_28Rev1Z+VUa7?9~Uy(-zK0 zry?I!MT-E$ILGB%r{Oq(c06Ahz!vg5d%x&2k+E8XKne_sc0gM3oQRI03SR{LjaD48 z@+jq{Q|e7-9rl_D>%y{L(cXHlaOpEz00YnYbEUVRZ>PQ@n_8O+~ZJ+yCF>0c!aUW`AR*zWhWCh6W#|I`Z@ zzS5(5yHITFnz?_c3znVktm^iM{5Hs{ zSYoEspE*>5DL_fOrhOY)`m8!Vq1j83?ow69*!!QIEixgdfG?^XkL#X-tRf_dGHRym zHW{?p^c@&`7;(QLSa$)mW_>D<_N%@~v#NUhHw^xDB?(;XmIsMe%5>d>N`{W`ojPYK zPDveN&KUNfK7%qaE=jzhat9Gyw}Zgq7nosbs6A_+NO1kRyEUQCUk2;6@F*e*nvfkoR*Fo)S+6g!{&K zK^U*d*s`VA37?nwVd0lm_)B_wNwM9#dgdhPInL@W*PY{Pb|$$rYV4abR#uav#^h zssX8vt2}pIG!y}e(;Bs>1O9sW$6@LRd)w=4H&UGs-&8e%+z}OR$G)hFeYo>cwOFyx z$KfO;=)po8&$NdvW&faX0*mj!MJDg>A}ZJCCVZi~+ic`&XNsyE68J326i0YU3(mo4 ziM3OOU8AV7AW-%Uk7>d9iu>+a+n8~s)ScLzRK#8DtbdTN*P!w&p;FcA^#y90`>#e| zyCF9*K&};wkbW9c<+Nwn{#dcS@eLbqNyKX&JZA5@8@};^q@=!$_{3paJlO#=4O6`) z(-kbbv#jkuoqMC1P_8wWNROo(shqWsYo{ojH!qoYX{s!aT?Q1b3d~AosW_=AxEGKtQbD~`rmMK-Oj~0{ztqq>xJLWaw++b`SHc_GbzeM7JL<(yd}r z&m3}}l<2iZ?sc?3WzohP%F#FGB5#^0kIDQNS+5h~?WT-;NcR7b_7*^KZC%=MLP&6T zPeO2amq2ib0F7G+7Bsj9cXto&?(QBSxYM{poFa_@a--ui25YQCAOu2dIIHhs?C zd!4n`ene=Zq0#bQUgbtn-q5bN^0&01DjgEqe;9{`Hi~tJ*VfU6_5J~1w!)`UhS)-r7UE1 z2pKZDL7L?+nb`eU1;N-&Tjr1e@D@!41~MNY=dDG!J`pKea_Y&ihm&Y~D6uTxZoT|$ zXP{5-%l!5=xw2KY_OfeO|B2fj-8OrQs+&1@ZC6>wdC7EcIOV=4&Q68Y(z)EOf>Tq z!v<+^VAcM3SaQ9j%EFNBdhc9LYtST0E;h!&jcfV3>J0wtF zl0e+9Pv(61-^Ce}!I}vdv?cFtmYA}Z{ zSUE?tw$7}^nFSO>AyZq&UFB!-15XHCa};TU6vP+zN@$hn(P1Tb=zy=Yb5a1Hd2Sg{ zfxE8uX0R{!D;J$&D*{rOimioF5&J+cEHY^HXB&JzF@?HuIJBU~$z2WL@eqg^RQ)xN zG7SIG?~wcNRI^1^gal?*cxOznFstKyN@20xwm`hDAiSW~$Q@3R`G>HJPA1qQqc?)g zcHW~w9&CclYb1;Z$W${naoppbT&Czu?elQm`KqRc;4Sy)yvdh~rBr*$n#?=LDQSj% zD1!P`trE?-C~9UQ8*qG-Sc#k&#r^g9Us?dTY8r9#DD64&enaT}A2Hoc%+W7HkNFBA zKMd?9Q$nUjaIqAl#znu5dx)aW;~!4wZZser#)k@?C?3v0xtCA)?r>$0LbuRm{mf_T zLu=m9BW`J$wZ4bNVh2Hi2GnbgW;>y4{e@1h1NQ++?-qT z(xA1iT98-qV)xd1t!eH(zOricUnKplM;<8Wv1-xskp;C5Na6}?g@|3!z{i*ccX(=W z2#Fu2cabPi;xx$Mx4+ChfBgmN?xUaF@b^EGp2?_tEb_#Uzg8TfuGnsvI+fg2VZqwa zXIL#3>MsvRpSu$4n{8CR)2_6`!}4Z1L(1;P-#k}7+22Hw$gq$tdE!p{E8JY&pbxjZ zMy#(L=Z@h?r(7AN<%7-mmQLJ8snvY^dsVey=j=ul2|j%UH!Dq3k>Bs1pQKqF*^Wk~+Jibb#y~aJBKTx&i2znLva)PrHcK$fCqQGg@U8j7l zq}z9YSF1s#2KTtgWdzj_#-X+g(TMl}9}=;0OSG&U`gCYBJbxBjG+QHAr?&zfEiy_J zG^`8XU##gttOS`22QxcS#XVhMG3w$6s=iSkzc#x5IasX5=wL3uJGQInvQa`GDi2eUfu&1se9?vXf1OyhI-1-#dU4XF?!j~)Ok+S1fM2t(Gm5$CW?g zf{>MB=+qtSESg871yVoTN52|2{HYY|n!!Cr5vO_FTl~&*U-|Ow3*M2;qNQFJcM zOz9f2Q0(_0sJF2jD1(6g3wt>61?##yfFLV0j}KXlDXb5!vF-hlr{praM}jC;Nr`4r zE&bp*e$qDQhBXy=BfdgbFTJ4THD)eE4EBD~)V-Kdyyy#M0e|Mq;w~TmerZ;-qz?Yw zbb*(_f6s>$-!|N{kuBTphg;g~GeiOOfO1s2V*obQLoqtf z;@6<$Y2$5zyP%lN@&{S8FY?8xb!9-%*ru>rHOX~~r4|BgKIdM@e9<)QlW&|V=}1^< zpL^({wYqxj7bCzgT{VvH*WaBH=v5{Ca`W_Bro>ff9ZhXjG@6ARZt6RV=5Ga|XOR)G z4ucoI{m1l|vA&Ad@M0{avm8`mMDjXr$Wgx4pnS*vQkOUYCg_^|NvZh5BB!x`9xw9e zI|ff1B{Rf4-Lh`pg#ai+c6oztDRVWzv1%HAp+@*4n`_?C4*N|@o2;w9;Rq*7mBGmF z(q0hA1;&BbrcT9bZtlVkt|&CM>WpxeUY=<;%`-JCF3@wCjxtt%gMC=n=(#hH%-SBv zXZ#fkk6&(bU};S6UWJKRW)`=;o#riKDT8j>Y}tA?$fWJ+d3z8;?=PD(fzF}Yy!Xt! zbqAj`=ru}0jX@E|q_H1}H&9G+^*rIv>vLFuyC3V+~U1OMsFJ2OjlvdG1T7QO8O ztCHJ5@^RI<@JLsc#)IL543QNm06gH;D>R!8{2uE`b^MLGL1UI(Vy>rTqHugaL;za8%8g z>oPSZXpumAae$c-9S!sV0< zzyp%%TaQB%_&}uW+R!*3xmC;4^h{(J>=G2OYUn|X#*c5QBk9Gcd~Hy73J5~P4|#8Y zdPoa8_+qocwe0!(Rz2pON3FlB!?$G-rMK~w-3Vggz_00iSDRJVj~8)BvWN&s2QbXc zX~BXI43vWGng-s#gZGHm+{#4!_zt%$(#;hJjnpWzo{T;<1HqbO=vYWNh=~r&V9Ee#0>ftoQ1=Dg<>G0>h2+` zW;T_HcH@-akudexNE@5LsP(aOIhTI5`0^vht*t>*ZPcJv8t4ikAu%_6NCN$N#$snX zhRn9`GSzaWbsSm^H}W&PXOVhczT(Waf750c?chFqob|W!xK-qgz$aHyzk6i})d7PV z-3@HZbEyXFR6dG@>BiQ+oj$fhC!67Gf`?NQ@)O6hRF&uU6&<~ORSmNzb<)$w0$>asLdwC|B60dR=eCvUV5Fr~zXvp|89slQE7ZZJz~<7z38 zqqtN+%LHF1Pe@0x0JltZpN+&8NLf7xNoi%_ z`x%~`kS2m#nG^hm7SNA&8BhJpe4g&8T+>hzCME>EW_wwi%{sXSi_Q(T)iUQ&D9Du; z$)Hdmvi3F}2*F9Hnp{r)sXPeh&?E%J+qmQp+@UwLlfyQrRhX@iy38UV3w`uHjk&RJ|o$sQrh2hx4OXn{-|iyHxV{T?*VwD(kMe4O*?G z&Ymp-O6_F-L6&|ikIur@%lNV+u%;qAxszp!KD|l0ouo^!u~*?LdxTsb};Le zG3(XnW4CQ@{SHU%-EWPriRx**kG{P=-ft+PZ!`8Yy$T`W%^i{p{Z8#6QY7f7$YRQf z6ljv`BCm7nw5e?8HTZ2XSARw7EaW`|Uht{R+0=xGvLfJ?y+I>ki1kO{9ZhBbxzQIY z*~dwb5?T!l?oj4$7@^W{dSy`hX`9?;ZkNLK*|Rq*a=a0;;~@wfz9L7ssX?k(+JOeN zv%scp<3*q(&8zU6;~kJd%k7wNTCPT0W?_Ri<}B}2Zg9rs_J`6{H0*xLc6MdiZzS*4 zS>_oIkeB@HEI4(pPB;s%Z)I8xyVex22!Gn?aGQ zzI!{PsnfDs6jpt-nb)DAOP7sjkb#Dv=C6@fBwJ|14=s+6938joiF^g2#01wY;t17p z&N4lc>FyM1zJnk+tyba6jqssQC(FT8%~L3)Y#L?U1zn3KLrl}5*&Fap>f9olAf;@8kAB@NkQgc@ z1OaJ3y4bnA0_#JZEr!q{hV6t``6^Ppx*G`{a-LP$RG+FOG^Zf0~kvE7^Up&#hj!9to1O1d~Bw&R&>Gt-(`vGSNT zxuRXjc8^dLHWQbV^X*L8mF{%TsZgjpt4hDAe-oLXs^qV=y6PB|?eH2FL^DCEu7*9} zj7<(CPH+<~q(=(g2rmhmEKn%XIgpIX`*|fJcG$De(AycAzR?sfBG+s`F?>*YaJS|w zRVNS6U~Aj@DsLg?9*pyP9r)AOX7tMFk!YO*bdiifP-LAIHI@`XmpgHTyOSd z0XpGldI~;n&17Lv)!{-Jxxw9(BCNv~>dSR50FQpTZ=zMC75Q>FW$i_Xiuw^sQ*xMl8sdo?Ik92^@%ZIR9`1$j@{s?8@h|9D=SdCH7 zFj>N=`bY9xw9X$_6GjtvS;=HA-*tQ3Je-auzuI2_ZOAtDTsSPrDxV1eqVqeD%rfrr z6q)=0(3wpw6oPFb{c$q}5e%c(srZ`rB=X-xN{qiG^&rGS?V zJbu&J)sr%L_wz}hNGtK0*6ig{Dk^pe49w|AdKme?dH|`X5=IBdlBG{eq4Cb4SkcOQ z_L>R8JSX#KSm#zEO?pVkFAG##v#wvke#++y%5_!YJ{)e=a;!a!Zy!Gl7mE_%@r)m(X-$L0l+V{OPo4Uig!Ve0#oBK>2b@VQblUB01&s%8 zCE4L_0txQ~!ifccbH0O=Y15UZdsQ9FRJs1bkxpOqG=^Ck?A|IoH~f?twMeGB^w%CG zB{(og7aBlJ2NYuor)Mi%!(6S?rtjX@Aq!C<4|z~5`bKStg(_;Ya9UCI*>B=eyTxs| zaA#CV2hKIld>3)=YPWWf6Kb$3ujr4l_3+>pnli#s|4>&@Iyt!e3SpT95d8WVDC{rz zuN4Ay$xf!nlYg!7?HN>#6_NfTY#~|v?qtxm=r-^|E_Iqql{auHLdUH#8b+pHoXAxHONhcANuEl3<44)qTsRQ*mFl=B&p z>T9By1hZ!}r6T5dI^Ck{O};|I;23cUyU4p&Fcv%q?Y7$WvNYdvJEe2uh4)?Wuha9# z0PVnp68O%usZlMt$HG`eWW>M+=%Kap{vbXV_Ky|-Q_u7LYoh!>L?*kdt$S%HLF(%t zGjhn>!Wpz$yq%#NBs7p@GKgF!O8>=2VEe|?@>9_0+bo^%b$8C~*~Xe4)|0TJsqWFL zQPVk1&_p^7Nh_-PLs494T0XmOYyb~uyJq0wykAn0&1Y;eTWDB>Tuxv1e#OxNvBe!W zfe}#m&{}m3-VobwzyC$W46q+a^a>zsrE(A<8%1%|CPRA{b6o4jDGNtA2SKuPLX zGminlko?rswFEr^6bHmfNiQ|D?$U9Ho(|QUG7IP(4yU zW$FyypW^sa6bZ=g5M}&R1IsKqYd?7#K0F^?9TEVM&$s2$Lh`?Qd z`mmP0fjBT*VU2+$K1vDIKOZSv_pNp68U3!};@;lc!~k%8XB68c`uTOYlbtywGvoEI z$^pHFST(;ZTW1u=)k>9g#ez#8?ezKY!?t2}nv$K=hHn7al$NRTfJ|}yVSX{*%<=eT z!QAztm13A^S?n7blR}p?)z${Bmfi!wnY&xh_5-4R#2*YPTxvsz^BXH&(wOKTx;SgW zXtD^v0~TPqr3VCxtzHr3+`Cm`<|-4jv|)tC0=xS2pQheT4Z~7x`?@qv&e9=n1Maz5 zJZ=YY@(MV>d3LL+qY~7^@5MI^zXGtKE>e8mynjdm^>degn1byD-7iD{0hce6S&b-p z;%p!9fFtPjDCqcX0v+8t;okESdTfJXp3cn8(N?o{`xT?Z!D6B&04_&KpZEW$H?iuW zC&d(9IB1dRxIL?SIZC%FrcyAP5L*?W8q4tANn~G0$ny*e&>DH2_bdo(S9m${>Rdsk z4bVETth~!m*8y{uXdz6Bta_o=dh(6zXKexN4xpGW(jrG$FC17#HRguaDxR!yIrI}> zHPSZ#tkL|Y{SnfRj0)T)s{xB;t9A&`H=2myO2kqdzn1y?(OXeF-g{Rj4&JegaMwWiJg0C}MnoJDGm*ML3`3LDdT8ksFfoRK)KeZ1`+C0a&6)TKztcb7f~*~yiGiJtgjkj!VMWupbT zw#BQ!%5&>u7dMZb46_%Au|aw}WRj?z?@>5NkY0^Dndv}Kr}+jc2_1GD?Cvk4(nsqc z>2QA7W1v?tpC;6C;o8%yp0ASOllO(xlEWrzBcQG_j)rWK5ru-8;=7(+uH&ovq-nGK zX;cN>CRlB1WH{XpvFQ%QK>Eke9E3s!A!dlrn6ZXRyS~^qO5hQK*a5x!O|&UaxJ8)C zF%t?vZ8!cHfqLuuh)q()cU6dz@N!=$?li*wU{)R3?}=-F!NR!$5+Ru0bq(MIbs?&G zXe0GTcDPWupOQ@w;oA4|vEsAF(TIl*MCNXt`wngOL}>}@Y7WXR_O3&lM;mSFixiGG z2n|5=pzRQC58xk{{Rv-?iEq3UIQ|;*^q;WdWqIw+3pC}DProwKpC3S`Bca)X?TR<) zaTKPRS48aEx-c@U>SdTZPRLg&YU|zcyMLGK#&i+#<8Iq!=beg&?ve$G?dAphPK#rX z!=nmbp3Iji8N78RBh>mH4#Tw?`4eT@kKWr`2vaW@vRr^lBpRnQ@pFc-2J=1r$~8ts$|aHR7;+e_kp44Ov{hv zgW8JX#?@W1^RIf(-s7$Neo$Z1S?qqFfsyI(bZ4{Q+><5MTmw-B*XG7elcc%AoriUU zh5MhHT~MQ)=|5`rP8M&5OzQ%Hy^myk#R|MlT9Ny}wc zY=OK?9+v&Bp?*2IbN#X@DfdUjD$qjAMPa>7MrE?Cvs_90Bc&tG;&wtbnZjE7^C#BE zC7Oxp*n`+~-oOhN@A?>?mvYX%(aB@a{$(_VM-Aj) zctE$AXN(6xQZ!mHvMylmqX6duS%_Dq{I7fnc?URD#HG+C>w&g#*V5pOINH%2Ni-lb z2aq2eF>=Stv|eG%u&^G2!2?QDJIT48wMlTdt9iQX==Nm3BIf|wtmUHPUKYdpGQi^n z;aa^v6S&xHLqPa~4|5k?;VOq4mzmJYBC~_m@p#Qw_if346(9&J?yL!XB$%s{@DeG& z%+08b%2mhi*(#<$G8s|jM`VTnlSAQE^-1*jQJ%a+a?Qh4?n;Bbw>xn(L994?n-aw?g{fy~wA^jSl>miIYY z@_Bb9p4M_#aWaA8N1aJa==?2JT5GNnnIixv(9~EefrPSKjKP#W1vD?oU=^5$bXNwc zLz)&fPXRnekUNhn>*88gPG9cq_!kl^Qu*{JZ(IhK)PQherpF$_Y|RKzXy?}g=ZWsg zN#294q^LKcRf7XcR6Ta}ib=X>tAqr;Rf0aQ0fd8M89#>X8}UFNe3biwCa~aRRYGtW zz9hpiY15Nm4Rswh1Z^+Oz%tb$;WE0&q!#*C<6ixmeQ)jmQ@c>P$!FQ=yh2-+6fN)R zPqVBwnk$FlhiYlFJ%njfLT5pixyZJ=5n~cz9Q8`RRo#^8qYKtCN3I-=-SyEd%g^*Z zj#e^A#DYh1`rlPmJS9k`RJSgM-;b3&1E>9pN;>HVt<&hw$)fwV=I0J2I-KysG5|0) z73yNi1xwchP%4in_bDQ4Lb;R|DV!5eBi2kfTn0}B(;EA3E*Mm+{p#QAN9y{*WA&)I zy`p35Iwzr=NVziRX%gZSvhA8Xfvy7u-Ji7VxZB9#{j`v7xKg1o*nBs_QnY(r!h(kT%;9J9+~&UHB>Vv|!JVgE!lY0!ASd|Ub#E)MgH&fC9maZ%Vz9y48& zkoxp8(M@x_$D`s_-^qBQGvBP#!{j?L(g1fnpP?+%`=WRgfQ*BbpLA+A7ifuFk@_7_ zg^gQn@M@AN_YAJ0%R`5-lTt@#w;^a;U$=a|Z=$S5@#Y=uXs{ac>_ih<6?QljQV30K zCx7h9cCE=r*=k=9sx|Qm@9H8s*3G$9Qs?Y9uh|FY`SXGcG}H*k{Fv@{WX zyc3VN*ju9JMNK{yEkN|qj3cS@?fCa`LGrdPQ#>eZY42fDm1?W}n1u9wkCbLSS;cWs zNp6jLpUamT2XGag0sTvPFk1fQ^zH99wW04Yx9I!(%ruypQ&nplUIjXe7rx z8EC7$r>K8~7|;i>`4|r;;n_@uD?laHt}g9em$VHjZkB)BsleP_5J5D%1b=hpGilb#?qmSlSHMdMhX0J z`tQR1jpf)~OLMI&$}dF*GOBH-+rKmM*dzdl6AJ`E%$3aZ9p9absKbp~8rRD4sQny{ zev3O=>9$ccb5g4+!zx_*d7>L`1tCc)6r~Vn8`+i zhn$*d1L`PWc52VgAEJ~N=`h^=o z;2xlI?S$pw_ior|#@bk1Hcy=_i4C``NIfGgozP->(r{#!ez6aV7lS&BXik#CpAQyI2A-j)hSVg9@y-tgc-@wwdB~Z97M$gg# zvU04s=Z0!eGwslVjHp1mIF3LJOGx9UBf)${L9N=xU4Wb>yk}hFNwbAVX*RBkVdBfj zw566@wei+ME{vC9w)zdn#dv)qnGn)sPUBK`x8`(xfCfC=tqJi|FS$N&&2&{3*Wh5q z(mlGvRaA2*P#`G~irEZwfzPRQl|VGk>kpbiCv|Y^pFZ=sX2#sDhe(I*(#ITD1%^9h z`bBb7hJ{glmAi*$;h;=g_kEjDfKlwly!nzC&hGZCHXf7WdQU)c4x@I)F*Nqp)?;`i z$AJiIgdHZ&@}Z|}+;YZwujs3sEc-wi-rP}=O;U!{QlL%0PGU70GM4C|zSYuC8*YF4 zPf0KJ+mMV>2pvch0f8k1%9M?9ucKEoTKK6+@ljC^mJN9^O5|_MSYY~-6dH~S!C1-# z@?`_U`S3w}uDK9=a2ir=-{e5$w;KHkWq~Z%V{TZ5IUT2T4g_3VnFQ)$!Ed$FQONxS zAUOm}!Gwr`%oT%d17Hyu1Es&CwFMXjk1iVjq|pOcBGO$qva55Tl}I=&qKZy1q-Vu> z6U8;ntbvm3AV!+{9c2QdUfYhSi^n<<6zvhwZ}Y1m(_*u#xev>#So)&#@=cCJq{Rf~ zWEa^m%*TQi*LR~&3I$J$E{+4Et7m8Wl?h#L*% zwt`YOJUTU=ow@b-Rgu?Dc$-!%96Pftx<79EPbIE36i8u-%VhbS1xP(iogQ$jS)Sw( za`oAj9=-m~)!K%C(ezlxyL=vakwL;R4)K~VxaaG&zLHG{iaGN!_(Z@9+us|ePT8L8 z?}R?|(LF-w-g=%Oj)nLE_2oKo+7x6SRVwhUp9g^;bs&5-sVyL~G45^6}CIfmp;M3fovL`WfeaEFoLH}ut8KZsab zlz?#n7n2W0RI(Et>IR`gRDqu6-Sg4V8`kFwD@cWx>$OpXS>V0<=F;LN&1S8)fZLOm z7Z{c6Ss#3Oyd6B_sqC5=GQ&YxU|%6GkJn@I?j`2izhB-mVlQS#!yQ1#4+c#fNp6Lo zZ8e&{Wh}>XH9S?Jl0Cc&@TUU`tC&!+=q+^1k3=U+&Bz%hUEH=sEFH$=Tuwy? zeO%bwOAAFlg^?MyIieqVZ@(f8>sLOf zkZ|hxrc04_kq>>EkSHKBi_c6452u&TGgNS5X!zO_8*m5r{*}i2ZVOHt#Gr8rl`YEi z4EvOPO5UVnk80bKeOFUUO{f1co7)?$3`teTY^u6RR~l2Qa}wtOsrM^-)255Z(}JMj zMvY`mv1VTO*Eq%%9nDT3vZ3`;+Z>N2lgls2?k^CwVl)LVvR7f7XM3-2vURSqe*wa& zlVlOh$*)k=tmFi`F1*Q>b_;ir4cV#Gn?EDeW%pv&WlI=%5W>)q^A1EBm|B2!tAzEp zKM?NbL!?W0=~E6x(m>Zz_)M)$M-G-B@nDp>!)R>0^ifGMy~EKbZT8`cxJFNKnzmNyuUeSe1rm8Rj;hDN?N@U*TnA8#~Kfnue`m> z`3SulEET!avS|`kh-IHZ+}8O&tvM4kM(5$kBr;r@i(2a*91b75GQ`_>Ilh!@8h4le z^1Xo#N?d1d0uY3>)dyvL=930rU{qVN)_U%ucJDQe7%yLb-KMC;OXsbPdHvmlJA#I# z{nm6;+VYNg?nPibAz2vQ+^T--u(-nZ&01x+LS`H8Csac_(IpgK(*Wp1%3C z(dKn>r0ygCBMevC@I1Ue%xa_QIA{49+_jOzWbyRHD4FBYSRstNk%Xyv{aYkaGgI=< zaH{#uUUMN_AaoIP^_iC*;XIqVxUuY9Y&0AwZlm1@@M&fwDOwW`V+LRN7<{;UjTm~9 z6ECxSXtL~{*kwJBk7XFc7^D6z9+5%Yklm-b%^fFYE~7(Ggb||y*;5?! z2}u;`mUN-IFZOA$hqr%={9#8!iC~Y_qfp>|`#Yof4-bGc7>ajV>OB#)+$MzcSbv$}~bFqGic2tG_7-mcRF!q7Q%F-FT(y{OBebqFW2;%Mz`ledq4KL-ym zYj8OYdl zW8(Kz8#aATmZ}hMv{Nnd~$c19OkN+~7vM@31Qpsh8L6)(W<%o0f^6lf9lXj2eil{w-1=_R+%k7(owBhK#UjALUB- z~dP$D)m!ckJ1mTph#;Q1LchZ3KD9|-0-`{=%`l;y z*Is7%l116)x(Iy)C)pqUwD8`<^uLWb*c;1;geI~2U->(nNA}(SP!W{(Y|OL*E0hWH z53>}hG%eTL%9{~($eii=oH@4pocUY?Z%XV3Pv`SjE$bfKhpLJ?dOe1FDoBv+ZMOuZ zFXe4mFo_zTwEPteD4&wqI`*F6aF23GWc)UEzP94i9@~sB-D5sbB36YZ!{S;I>IQ_0 zrMl4!$W;;2x%KgDuIh%}wceI3(t0I97IbagtWG79dE7GUPw)06tR>QO@sM-ImT@s+ z3z!`v@AcBK6|n-1W)c^w!xk7RMK_%~$-JOa753XReDZ}bEhKb-(n3XQk?^12FnJ2yC@hUaa;^GjtB41*Lj;Mn&vf2w>M`}-NFsFvlPY&qtlz^@ zhJQL=%4-_8I;oH%{|z*HcJ2hT+ILs8YTs$pe8B~&q7AI`d@t!>zH7PDC6m`!@sl!# zs92S6+XY+L>#ZNO(eQU@pU{s;!F&Mh#BIbdf@(!Orfp<_wexYk2@3~G@#bYv>RAk~ zq)lLA^*t$lnV4|pSMTkbeaSUDa&$>kyCxmD5gC}FbYbO_SeW2S@+Kc zxow-W3cDGz{Q3Je$I`S`(+{gNQEnSmU*+)-K5T=>nG#~${#;;GD+2zCdSmxRNh}wt zE(t;uFwozx&Uo=PJ!1s-X&J0s_q-HV(?O({i6f~G!?Fy#aGCG@$r>k_&s?Q`z)Jpr z#q@r0pZn>@FM8nppV!-94E!6rR{!sy*b40;k>8V#l};Ukdgb4Wey%&84^!wLNP0iq zDB6?ynhUqH{22c5XPCFrFm6m?p~H6C*c4q`c%zq7#B@|AZ4@`X+a7tl{owJ6OTx}W zs+X9JX$YU%DId4p>*)FZ^=IiO)r|Vc^>GWVkea^P7zv+~%w~WCg>{Cu-3UX& zV3%*8@p_DLHM9+lBUGEGMrLJkCPn~H_w8SU2ybBNXU5(BN+$njOl?FMwy&e{zJU>c zWJqfBkrScXbSgIu_ulPUiU@JW7c zT^2Y+ z`yUhV0)ZR7GZI^cDl&d-1v;F9CYS)Y)gB53{&XF>z>`2pIIQSc-DiLN=bm@syu6kw zKV@P4*Vo5|^?W?$BH-Pi5V%{wZj>)GpRAR7^&Q{F?PQs!iO++VM)$Johk^6PFr^J( zDIxWJmqal#Q&@S1yAf={&|j;{4;Gb-ht_E|S`!hcRl%z1ZU^3o&DoBkv+6vr zsbHOw`?rQ9!8JNeISHPvj_ULSJ{O4MFfpCj18NH#F%wy@}-g?dE;xVDT zP^!D3 zUgT}jjg(VflV}5ZBN^ASjp*kFz^OKSn0r_$8gsE$XN6*z;HejQlz`O?goAEx%UG zy=`cV0T)R&H}q91bJgKoH-EdCS@!QO@n@5CZqME%pYvvQ1i~spvSec>|JRW|a~_Usv_-f9ikuKAS>#-KBO5Kl1-NP&qmbc81TqxIi+ z@PGC11m@3EeoebDiN7Y{e>KG46A0WV#PNzKqS28L|J6z|;Tt zlKy}ApZ~wnz3Kd3`0uI@hVK~vS*rh+mvAjCHGgCR@n-b@Wt2br1xMsZ_XXXS_yz0V zkC6Z2+W9Yj?X4!T3M6DA&i;$f{-0Lq8saM; z_l!4yOXt=EAxEwg;CBsx@x_S2a(+_<3Nj`2mU;4z(DMc8l~qiL#&X$Gy_9c%_;!&= zMl>Mr*-uYKDf@R`;tp@qt$f~;IJ7PeBJ%}{og<6|nn=B+!H`59x3Fsr$o)(O!o@s| z5pEpcRVWxAq{=K$l+*hbUt=3chwO|~TMjuNAT#`E1S83Csq5MkGP(8!y0-8+*W4@&Hw9$fx>RP8Yk5wy{Q z^3WwgHUtQulM6jv#?pfte_PtI1b2MdSLX5D@N<9eLDO6&iFUba1cQ9w{7Y-d%z?5XvNW;5ZOY ze0l&Lw)X9fXmDtx*HVU$yQ#T=UXk)$m8;#_+Plo}>d-Ere3HIejFgFvK|?z6%4NjA zk({!_5^Uo4wZi}Kb$QCfvp$0h9<9wj>tH&eY^ki3iR<0sI=MtyHQ)_Rretvc?ePtb zQgij^A~ii!5?)MI!bH}XCC**tgh%N?zF80MrGs3((EM?CxpO?)yf9SzD6d!e#@wz) zjh{x>3C$=eQd0$sns|8fglcPa?gOD!PIjKH7ne&hRdRoHy~@3dRiNaC_)8S>)5g&O*09zIMCBd2 zq_nETq4ddb{+>%#ipu*E>o0x7PM0%sN!)Uabyn}VR-Hvz+ZJy}T#Xh|&lg+8Vt7>E z6lvyd7`v5`ON6V${<@axf!5^#@Np%b0XVYq_~@L0k-9l;at90bNZN4I_vwI;@%~~? zM+EOl-a0^brKC$vuGvL^&9sqO16{X@>eAe}DE-v< zH~MF-f{EH0A1j)AY1;sESB>$dbBgcVV|>|OLRU;ig}nZ50+6e|ztWw@W{$@ks{PGl zK1Z!(=>%$8F0i{w?{+$iMR`h^r|h%0SOeqnd-+Xeahs(b?cr)Y1ebwkG!w1#S_=$M zFJeA}kTgJjeN}z4DV(T7vEv#0hUPL-=;=e`j4AckNeoW~-;#65hue2)?CzMY-Ds`& zZdALwv`v=;x?!oUlADNL%g%TI&DKn$@h)MSvZFGuGFUm@S1bE+Mr1t*$OF~w@n0P` z&{fTb+o@>?cpauY< z;LJ&3wN=-6#QyKd9s~ap-VuraG^$>6uU- z)^=KPhvW?pS@H*exhu&tQoPSEsty>$7>n{cLtlKM_`n{>vqr!i4a%7(FuUsvxo zfbrELb!qh{Cu6L}{d^*hA4JwcU|lozVj(rL1``^U{HPGLL>aq8GIM^6;!!C(yj~c8 zxgVAe^Mx_rgo6eTr;GDP5;|y|9~^$irq5zz`J>e-RKV#8}|@pp>{#zho8LGMO68v(W3XC8!l^{+c+?jkW##8a z>Umw3N3+In5>YL+q!(+=dO>UrdJD~0M3)93TGsKv5zS!TT>tZOD7ZEZDLw#UfC0hp zJ$muDB;w|@;#}v&f3vQEk51r$oa&8`du6hKQNY$PpasM(c#Tb`{K?e;!q|3yr7~hp z77rATh@Zggyg_5{O0AX)qo)hiVXV&l~Qx!&n3^fzwd-`D6M;o_-_H7s+;jy-T;T|l@=$uu62YaT@QEJOu@FV z=FjM6uiGn)@CziPMF29T`u!gh)br5{0bVdLxBG`}Y`=zH0#QE3dZssIb0E?pW1+?b zZ4lf+&rkVRs&X3eLTq={<_qmA`VUoWUK1Gnxs_<%)A=zTTo0+WN%Y43{#5&ZbX~oo z6Fcki`5o;dj>I{SCUnWflky*80&XWjbQF3PNZ4DZv8iNF>z;~)g~8Cz#0Wxu{O=OL zwc)owv4DSbM`-+RTF>4pQtfl_cttzZ-23%+5FMJjGwQnQxYf$mM6v2s_3|})&Pt@J zMh-*dlJiIcQ{pzAx_R_v%xfg~7`ENFml+%$&0+W)(rRV8We{HkP`fq{0jp`0;KOaK zLahT-=;^j6ZFUK;;hmLPM*No+z!17|*(%77mFnFK$V5wnAo1^p3P9^d1C^NH^e)+k ztU`zG)|XB_KMa|K9|TE#=CoX;2l@t+hL)Qq^+3xh8t`I1)#nZ+=(~$mYA_&VOe1jU6 zJUlHyVrW#vhmy%t$1K24ug<#yGY6liiz$ssY2V$f#tsx90~$0jb(y7!WL^(aCrwlh z*bEOpmL1dyGnu_nJ$XIE>phx8BZzq@8^8(DcDrLVF4LJ_&%U>RI!!kwgK#HlDg2jb zi{qDqi;IEMICJ;6-XrlECb8AUPAg1Q92N70 zQCZtPYGLqay%DAmseqO#nXG#x$rSeWiU^(IJAUY3AO(%b!}@QJN5YCxBm zjB$AJdjPW2h%XwpeXXSB#yaWA(*cnO20js1_GXa7b?>6b48?q!{pkWAmcw6 z`179C$8*3>zA)JLZhEV2nb+(4=wQjuVkF0Ps$mzvQIn%zk_HVsB(pQfb|P`fY3QV) zBlz}dsRC}czhk*w#Vsq{rdYI+_5Lnklo$Cx8t9B1NrCP&Y3TBvpDrq_WX@Mo)$eo4 z%d|R_AmCW8NY98d8l{C<%cafnXA+^8=WAU(YD+(M-*yy_cM)fd0_f#JmUW?nre8F7 z7hpON$~(*kqg^YW?>i<}t2J|g=u4f(oK9uAX6>G@-N-1H*6xGTGQY2Wvr?U0pz4yJ zQxTvcbWne*#(}}1X|mYYdw`@*;ThWeSKB&PJlfi}kA5Cj`Mz%5<<{^pvDyjpaz^37 zL;x*X&^P*0>iXf;F_FyfS0ZZShw%(mEJM|n+cH=2GxzLt*HiC%0ke|jbB=9sfyaYG zf$HybbBif_(8GI3Xy(zk?)ioS`NzVZ9wv}hCA=Fy=pMRe(EcK=F)Esio3i&LVa3?tIrV4K@K^7A42XnDuyDRG3TrrkMLIs>hQoJyMsfH zB`T|X(eB3&;vL>;YNu0{C#|BOGtmIlq>IBkiNFgkUU2b@KN9yG$xIR>h5-`A3hP2*&q zE_LW9?)|zxG+lQd8~A3}Es}tfFlTsAV%PCK8an~Ue$BC;bde>!=9!)QX)C;fpodrp zRr=!Zzxbvsb3~J)a2Bh>LsUXbGU{VtZPbf3?O%O-F9~jvt}n-gFc^Cj z6a$=OHHx%b>R0k5_r<@9UULG{wG@?A8J5tx5dDqKTu5~Lha~Fz9UNN?1 z(iidO8obyJZZSK%4SC59%TL5Zh2Vb>!+v^e&XR;PeaD8NLPxO)Oc`-@x{JVIu3%kV zJMe-$B-5|HL13mV4Dy!m?sGENCFcqxm31C+I0Z~RzKbV2zoMge5HpCG0auix6Df;P zlYLFONdp54b?$=3- z&>7>1W9?=Ek8nAe=i3weZJ={7TfyA@0_y%MO60>*y{LQoO2PH1b#?2&QhPeT{f}Ya zV^=1F)#mf}@Zu_n`EW`ppi?`EU!S_E=Q&U&KMZ=bsDFXuOh&%p)|4~R&h1x$h$)UO z)>%@PXIPp+v6{%Sh-GK@CU$w58f$1!C(tS|@P&qdRa-pB(y7)Yy|x)$M!rbdM4z|? zCN z9nD$jQ_t@XrUH2C*#LDo+PA+G1g{hkt{hg12x*?C?GyaF{RGnTE$lVz3hJ2;Y4e+q zH)5oIo|7N_s|HV$i{gX4Tn@7q{rhKhZ-LqTG{(u>$hsOUP2flpqc` zd+j$}>sgz2-X4!&mYbB1O6^?7EyYs$=LA8Kk6`4J_zlq)R+NS39NMz%`&kt)nH9DD zB<1=JrtD+3u8l<&#&_YNub#g+3{QjYIrQVp+hSFiuIv)eY#+1D$2X`mI<&fhmG8OT zAB;RVdT$=gX0*fcQO8yUx~t3ht%Wpca{Fbc-*xPQ?xY`&fFMn}kap2!(H++i(G8-GCl;i|w{(V2&HH)`ab6u0=x8U;`Ya#plYQ`zMH%D&mYRM>)p z3%w)4;vv;t0iIL&F2#7yr_RPKy8Bp6*0VNhcD40Es3?>iee@6b&i7Led|Y6rK3QG< zhS&;0nD!PS{=*X+3H0d7RA!`;()7N0fiqRXOetKG zRnC`Wr}fKr7hF5L3IBOar8q|udoTr^4<&6o-yIxQmXkv)yTd;$RSN!WB4gABJLR$Ip_RfR{StR1H4YCx*Gi@}tz6#q!+yt{Lz6r` zyZt0(j_}0imaTlw1rIH%VLIz2g0P~Uxoq9mM%0H;Eb@eE{7qyE_k5{bF~PxiaBcem z6I1XV6IC3!pi>^6%u0SR%@%WzXshIf@Qs08Gkqy;I(M1A{$D!(n}8-~)1Aru#Q@&2 zw(ib%F!Mv^??m@gK|nv);DwJi6{CJRV}^p?&TGvYhwS!*kM4ci$-1>BVeh*&4>HyR zaeEHJXTVxK_8Adh-aR&`dLhEEr!>p&uT)CY9y~9Tp584Zxo0`i8Eu(U3F?OT%-;27 zd@{1kZy&lec+1KrNA;`3^pZgmemDIRb}jMFs5KP8A{R;GuJ3b%?b)Aqad3s#(J1ke ze&Se&lCYcInPl!&iDTiCA;ml3Tx}5f0Sn^K(5$g2zFTA?5#gl2CyzV-tBs&d%5{FE z?X!IPc%&%fg7vAoIw%x-uraFnWvb<20K{V+&}^W2YHc)(6TD^bK+4svd6Kw(FUe^! ze1JUzkZ6M0^Gj*g)h2CyCcgfE%BLnQ$zCwOg8I1vfh)V5{34x?X9X;tyr3S}2{0m>Q#nxyY9J06W^#KH;t<=H#C7rWk>8&VyF2G-fr zKnKhT?;Gf!3o??x_)*6qN^cy13t?kj1ERRW%g!20;^F(VXw1^nB#mDRnis*Yk{S`A zjTl%6oN0ld`y5CaUEV9S#Ji)8$nhd*Ub|mnij05fvlH7*str7BJntjkZcS0VO~bPl zd@zwYOlMrSt_}@1)&J_#41HTfBa#%xvS&R^kJreoLr$HE?N1JRK-p)R>EI#$c$c`hXew zkwLZGINe=(s+a|RMnZxkqiR=IU(SoI^c|z_L;d}ifO}Jh)k$BRe3*in<0WQ)%?YDU zt4@aE*y}R>?(F+6D=xF604XviE!Vga?fU&o{|9gPYZR|m=r668h{Kc=eH`d0oivlb zFL$?%WKDVk9K6wuHW72 z?}VAzP9I1ex`2h35GIQEW7$?MXDY0JL(v}tlD#7y=E^<>)XMtVK%voxE)nrWti_|> zMknO6eawdao#t%EWh=9`UkQ;237|BJ%ch7f1)#Yn$)bja`NtNHPL2)vJ>IieI3@it z72DlYsPYO~;jI?e`|Fzv+GtvCH%=7P3IJCQrcxO0_=Av5g*DeHMEZi1t`0r0d$$Li z;;oQKbF+PtVuGY>lfL-7Yer0M{3bdzZ>%g@dMs8|W_>dS;yr7F3jLft#*@#OMiaZy zuics!Ubx!3F9S9_i7Ctq6Gc2ufO2hnT`M1bVht_|9!j)- zz|-tv=qjfmS)%a6IB_F*e5r;Yv*@gkBlatXN?3TGBk;(m{(iOtNq3B%wI~`>FV_=d zwyp?p1h$%e=di8Q`t-qW>o1ezXK8`5ZiB<^xFpTYu_PwHWriuUQa1=8Lz0#>Uxj?2 z8O$ev1loJNcwgr_dL%Ga@C&5h2HEOk=$r*~4-(|fleP}APiMSE2wxN#R=sfB$iyI|66p4`=8+OcgoezM#0QRwZ@_z?MpcDu zSiyW&?|QEw_t%)QcLmqJu`QplEZ~8E^6rjX_)e3Vj>AzuUUpZofBd z#ljlI&U-&Gil7h-4Oov<7S*q-86m;sa!45~P6M$p*he@4+_Ovg_qr!Qn~Ug#l{ip9wJSTv_m#Oh%T*|I)Z6Jl zxYfbOI=U_}*kE@&=7(Oq$u&q*Z_!f;57%m<`t-K?r2B`VlVxdcIiQV1Lh$eJbEt;O z*_dIX%P0cgMHm*L=Va@r%aG4SDps4{M-jm}c(41@VdpgIFJbJ{+#GhofB=3sScv); zSlsN})4xXY(Q;O+FtVOH1|#9cV!6oYq5Y3!^iYfCmeSQ|6%mK@>0}NeYcnG2T$`fP zzdBj1clUDo)-d)6?-4jUOoUU3y+SG0+y(XraZBCLd@$Ob?5-F{rx+{agNUrRo4iIs z9d{UJ#yzuj`5o3{!mMoNJV$O8wdl zEK1mCb5&9p_p&Ka`P=+F+3YUfz4R9Hzu@w2)(wEFcK;qBdTE)&hdaVE zG!3=*mr@c!`gDx_r=L|ER=z3@W z?}%yNbn#$~t$hELlgU`c{S3-0(`?q<Sa*v;y%ldi~8i3EI;?Cj?1Vgxiki_hRZ41{e$-@ zp#i3)Jd3EEegQE7jzlC;-q3Sv7tOf%bwkk?7BWd^Vrwd8e*z~v?A=Ml!@_HOHNg%W z9b?=!YNd9vGZMfJJ7$(C)f&T%S{CMFX?k)*_mg==_oDIuC8zF%-K2{1+Lh}^I}Hi# z=UfjPHim3O31&)I8gdzT7Z!ACwN<7&oYo1$B`o`lz#N_z+g;cBy~fcSN?OHu3^|yP z(pKf#HHfdUJ3g5Fe#J9X7jMv`4n z%^)@kL!b(E`v*XOH`LS&rjnZGU*@N(o!HpyS=cNA?VqgyX25&7)Q>TEz2SepWh3sK zi(l7%7RP@z+g&{@nCA8T2WiV@d0)!qgdI?JN>iln(5>2TkOnB5;D|>+c4Q3K%cSdg zJP+Loy-QzC&h$$#&#-%Y_0SDl50Pt!uF)~I=;wdmDHEl-tAkVr?uv%JtwWxKU0WRh zD-8~aUyLomu2Zjg%dl`5Le0-+l4I+T^mp%qzRViR zEq&Sf5o&rA-c99PfT>n(kWMt`@9Y3L?hR!*+WuY3VXFGi!qb5L!c+OeyqUz~GL>G> zv9G8sLaC$3^VOGxwdb;(E0N|{4JW@0;oXdP-sh{0^fQJJOQMY3->DaBjq%>yWnODQ zrS?RhgeoAQA_4(njr~2~B)9CKjyybk=+!U@rYW1tCH?extvIlWt62dJsBEixyt}Mq z)V?aIe`|ebC*v&O9Cj~8D)@ngDiun-I~qPgO20z$XZ$)K`QjQ5^K-pzA9w3 zT9-kWOal$--l?(WYh3qIAoYe6<5+7c>ipFCVSk-y%e-O7Rk}fC&STZ0W4D7#F1r8d z{vB+vy{-A6$O`XkQ5c*2z_*klIuD_)#(AEr{#*%ArTVcVzQ;~aSZuT1M*0*_SppjX zTl`g=Hes+9qB}BCD?R*Zy(rn?C+laLrkDQg8VTzZaa{$dcLg7r)U5)d1)=30g5FK# zx0+dyHEhU8ZKNjCnP1#_O1#-K+Fjd@+jHsGG4nsm6@XXy?QsS&xeZ>}g!$G}Z-JCY~magZcQylb`o;;>0a9T8a8@gfJl((5crV)rT7fbK;s!|D{6YZ{B zJPFLUTHurZOy3JM#0@|C*MujouH}GTs64a|3po$_jrb%9HzFaSZ?~wY^^9v|kZoc# z0;XC6rU^LT`N?ZfdBDEB_DZagc8I`bPA3i8F; z`+S*r2Oj*zLqRIkyJHcdo33x4yl(9aHCQmc*-7T1K z)|Y~DspYH6YYs{kL``Y{!UhlNv%V0d*Ko2}l%XH{mKp(RUVfDM*|Ohyj6#e<6Q+8X zIx7A~oT**rMfdUoCYjAS<6`(y&V_4`M-}tS+ zp&=FRXZUkya%Lm(J91R%4Ml9@9V><(PAMPVsU^t}G9*fS6sUPUFC^5D6#{xl1-hD0 z9>%h2@a);V)X?fHoBD3A>&&4Zj>-t-H7l0sFn(i?+zv~(_J4-kHa4xpUf83xx29qo@)39?>7|R`20ZaOxV-n z>zZVOPE`am#E!{=k-~7hTUep?FE7=BpH#6O4WPbls#<=B(=x5xhK;)hm(M;OipLX! z%3sjN(sTd}UOIm89G{lU4g#jZXkbB~mf+tZ4AbvLH?8}M>xgOZ?NhtJu*ju9h^X?M;$^9XR)D!90~);qJaqEW5HDOF)S_N1u8#>-3_Lx+t8xv`F_A$|&4mlVL;kwS(m;oQH<5Qj8%rXh}d`LH+!immgor*Ro zVQd7O=JKQ}SyP~1kGGidH7}Ey*_2EqHBIg#<8|q$+ zM&OuuDO8BvI+2H8%7c^KQCJooO~xS(6pK*ISk3(Wu392F#AO>7U*s|5r?(p!ja8_S zIXht1VsI^}NOofZ&_V=Yz8G2(wW&bf_h%CAZYbdgg3a!8YAwDhzCNtgp6=Xe?!`$_Mc~R*-(N&L`-Cc1 z`+(n0G0ZOVi%!LSAl#Um{?Mxsp#XmXtwpg-CfYYAK`n93uft-8FRj+Dg(;p26N(+} z%W52QwMb0x{FqsX8OL1o3a#NJ0NL7`uX^#(NapSXN1i@lEIPKyWci&c@#^_B%{X_SOz53i3}ZWdlwArn&N0=h zT@phEA?4@95{dKRpKUlrj~bs5Pkvn!=XqE9N8yJZId-jZm*wmnW-81KzP@iOGuA=r zad5z)Om!PWU$1iBJ5%EIII9BTpLndp$sw>jlM1#)&>~GWE!J6ux}z5-fBxmU(AR6_OFfxuMUVx8H1T^30J@rl*YKN=A^NS;Nsu3)1U$ z1HmkVhnWR}l^0g|!NOR!4}pZRG;(IykJs23?{OYZO{pW!J2-xOGj=7rYu@R(H#GwI zZ*jaA*Pls@Sn!12($MKYxapneU-XNL6Ss)J1(7cQauXoCL;uYa?xDe{EReqQVL2pG zxWCDDIi3(5%@8aFHGhebKi*1Qmj6!Z3K$D?7tY!RukG9EVhd!IhAv^~+hI?-&KdxE*jB17*DHix{d+^>J~d0uK=ypR>5L+$b?Ny1bMGp-((+s{^C8f5 zJcbbIp*K!=?}zjba)MD{Q=Bvr*|4tnF|hjf@s3*Ya~D+u-lR{~$>p2605oc9J)pRw zv*v1hxtQFqW^y1msG&1W8y8<}<}i*G&1hfHyM5}N)me3TdEgHDKohy`Ts*gZYh9o4 z`AD9>@a_ir3E;*H6Eey{(tdSLWfCO$oF-r`c ziZ0wqVB}5}?Joz5^k;RxH{ClUB<=3J1XgCv0!rGWDn`?3GeaaODvxCn*XT_D{fwPl z>UoJ*<-Lkp2Q+)SjY3R9%`89KW&h+SkoCIVKI!Ek32!!L`9S@vT<t9dJ7i$n%rY>ygJR=*E4J08>Tauy9uH_8!Jfn7i3`> zv{t}|7l5+Y{L0MUKTd)A&2|cfjZ+Qq>3r~-WwGUpq5BQ1o(c*`Fh~peRSQMg28#1P z#BX;;#N3Ob21s>8>e|a3+xNNPXlTyz#NXqwg#P~|IRL*;3=v+}KV*^f{OMECad#2! zqDqjaK+h(8Zq#cc+~NDF{I%{N25?{YqT}tQKXqRPt84m@=EjDu13`N$X`)1ZjM1rl znZj~BQeLO+oHK!MGo_M49=~dA(4TJ#J>CJi15%7P_w&K5#mf2UZ@5yPH97AlD7_cA zFoY*AH4KgFwoFfUrN$xm2L^cYnas0N)!xW(e5VHYJ(WKA=ZQa?6&^vNKz$KmLGgtb zF^o*4*V&saVQc-QoqzSh{o}qMkoOvB707M<==QVh-leVG7uE5c+j`W;*72^%zbzk# z#NpCBE`Qb2vH-clNEQHN;O=%;{{ld-VD?8rbOHIZ{Sjm(pF41A$UcaC&|#mSPgjMK zbR}!>f5YFfYupHlnWeFJp6It6hwO^#D3vr@p{3!^(Z8O5#mz{CtrtXN(To`w zQZ%vvRX!^yq84^RyEr+OTKE1@T9xu6NV9n(*DanQ#!BwPb-;oL4TTp~P<|d1Hghb~ zbcpXvL9TCK7nK@0S$GY+aK9G}QJ8w15BI`3xRtj~t+mgLJB7f8IU=ml3(GG$AR6SU z7ls0!5(i7hDY`}8Ux%kkq_}Z1Qq4d7L`b&&l%`ll=_OY#c2YQYSN1ZTX9qw0A_a!D zak9fx@3^^LN3KS?S4vIH;%nTbMK$4Wv>nO4nnh!^=DTF;HCgEgM3(^^zPm%mKkC@{ z4Y%<}Zd;R&r{{X{{3)%7)pIftI9OlX%WHL9h08CDygtqK8#59YPWhO@~1;k&w2ZeN-YYl@h`EbMt9f16b zTl`1H=&@v8sc>p;n>!IM`vq$9%6nOZeC(Vkd0jZ|#7*XfN3Iv{uF8G&U_eK=L{!Pc z7RIee{`~`rb=~!{;F>QB=OKu)bWM-Z(+Q;^hE%HWhXb$6%{TeCvl#Z+lsYc-fh!kn zQ=T+JsLbmTSuEffnDfOnUEv22n7{2-!B|l{44bJZu&V0IfYD$I*v`LhQ+wKkyAdT^ zeOyFYr$gJEFtvNqH9G)`EPJt6z4d&Ist@fOK0_P}6977?fB^K7uhzPlvAbWJpGBoP z$RM7$vBp$p;3iDP6u4Kxh!9*NI;*;#F=kBh;Y5NkP-V?m>Bc<6InMggU-VOEsPLGl@_y`@P4rtf$uK)2I z{o2_)44jE^)K)gijSiGg>}~~O(w|7UZP2dXmA#vAfn84xcEc{=5V-Qhxrn{F9Hnp4 zCCdjvaLulAY_-6+jMPR4zsRQgirW_R;GLSKWGa_i2gZWBze6+iw`y#%@242umd5u} z--0-OxqIPgX@VJoBN}mY0&pV4cA&V0Qw>wlEvmPK5ppPN1z+*TipjV7iBGb)?d58< zC-q0O#st`DVc>HQ>*8kIXles<`QUOdK1nkaj}B>x|9p}1fG<)4|72_T#vto|dB#-J z0DN~+1m(|znZjrvz%U#vF(b2Wdo%krplQl{^p8IxN*Tp1uqtXX+HF_J^tv~LdXkMm zRW3y;zpOw&{Qv^sz7vUh6|%`+$*T?Smz)j|Qxhx1&Nroz>SV+2ePaNaT?TeFs%FD& zw^|lV_J`Q9)$PW^6ClYft{+G@XMG=)PUq=Qwh@`BB-i-FyAQ--?e(Ds84#dt^{ z4z>v1R>d!-(g&bTgZLC_Db|YaRx*7BcJ6LC?qM{ag3V{(lXnAsi(Z10bZXn5{OeTxcCR^(Ojk79FKl&}TU7ZkM^nQ| zPH7mM`v(^?k2S70s_9$&n)TMqAFuAC0iTNX9qup~uIob_!RmHiw{*xc#yYuzpF=8< z)Z+%3d8G(~*LmwEAQFo0_)iL!t)>q)cM%hf{|}cFh0sWU)c(NnH=OB})V!+)T7Q{`Lp3Xy$=5x>!tdzAuk^ zFNa(o{2tz&DH39Tdbyb-go?FJ@rlcK{;eOF{f}QYqjii;p|*^*K8c*X7bq(Qpq=d{0OSO2gwuuXaNkTn9QFPUCx6Z!VB8(| z6f7&#U(yADz4C4_97_WZA$52J2y(Z|DhN5f>M}avQ9G$RLC_q7_3}jH@sj(##qG%m zv0?;T_1?m-T^F7R+Z8^ILvLXCcV!1^oEdk=_@nbFcsy-aMH&{GqVx-7rS9#%vdujO z5CF8k&Y|5VIee?cy0S@pP~eO5Zt!K*6{hpTC%mvlmxu+&4?qQzLvNTOY*n%ZjAcD)G-rOze2USF9Z zMSVGD9_$_*Hl$HFSq4ldiWodIIqT*8R+ZQPF_u4LYhYFee9?tv8*)`P3X$iim%z5~Kp30UZ51dZbvE*1hHETtSzt zV1ep@O3f0PzDH~3f_`{vjug8)BgWTK)Ey*Bn(@&zg!js3sbEG$3l(X?v$S3pA2p3t zGuL=-)d>dpe>-n%MKrA1O~P{cU;&?pTr^J^ViYK{8u+Ja8tXDE<`UAYv90$fc&4~W z&<%M%rqp&v8k2LAxg@#Z+3|ID3F@?$QK25sxW^b1f4#w{mWzEN7q~16Ykp?N=>YLw zSN&&(a=Uh5B-@0J!1YFnO!GSTgC+Y!3HsaFa3F-b=Ujye0t$t!Xq%Av-MSqF4+=92i zohji%cD%Bl&WeJBX5p1e@-&hA@i9dUXsM&WqNf7F0e|s7EVt7tl1KK)rf#?iOceS0 zkT+Ofs&!tkw^JS;MEA}qWd@WYHjFti9?AoRJi=&t%MF%2i%`apr`3f@nU2xb2CcV~ zMQK|JO5hcL>2!d zU~3n)o$D!oDbK@+#0b4E7|}LKX(r%7Jv|qA;iQ*HSo&Vj$1vc{B*<%cA*uHTr)QI7 zSm^Z{aCt_2%_0R$1<|B9TGiVmBja;zAk>ZVyPYH`2*Cx`>%5<6jL5IZSNt3QOO$*e za%bl6z79wdx41i$|K;&)^i*2FM$3UMx6VQ3bu!anKi&blhv{$CHZ0R^Xr^o!RKTBpx57}$2mel^TXk>l1!G1;mK>>6g1jkH9c>1qCP$EFr z67|I~BMI?F(*zZ?8V;%95c2V3vY^l>HIMq4AVX$R^xrB2Z@6uPW*4jT8}Ta)Yj@|r z%;TG8sM)PS00Eu`i2nytq! zllHm&q1&Fy7LC*A6t^Pwpv?;)Pc6BGCCJzby&Qwo2dz`vqOw@2MHABcR`19#W%aj( z20MI{%p?|2rZD?R&@R{N&@lFLFCXs2daztxIh@2fhgbC#F#nLdlD%O4s0uk>qciM+ReW#D%fWS#*M(&fhtEZ|)3%^7^ByUZHn_7_ZGo(l_m~P#fhd!Ur_6_{w`g46q%m?yq`nUFs~pLAhXWN=e6WZN zQG*A3q*P92x2;dnVUsfVhYV@NM{+4tZW)RU|1lifH<@^mH4O7neKf8NQ?xb1exOR?vC@;TFlMdbs$isWuC z7)=EDoX*bKtFh1ft!E8;+WKRXeAJ-{n|Td@LpW|1dS`X!A@wzy;gIGPF|detDEtN6 zO%Bz0SBF58^T5_1x^?3}6pUNB6`w~cWW^QS3l8l5PU{h(U3>6>Mm}xw-Ul$8u6Zx@ zi`r?wkVm=HBM;jshZFd(pY* zxu&P!J|d%%J`^bZLM3H zzR{!!(el?pK*c;wtbVd(TZ6&!ci(>5k3IAxu2$;NB^r$=heB%~6h7L&$JsCz2!ci; zhbNS{RppxQ!xkO{nHY|Kn7#Ub%}_FCGgq8A$`26#7Z;5`p8xt?*O$|}rT4bN)0iq! zqLe#khhcTL)GCW&kDRsMZmK;sgU3rG%o<}gQ^?>=>?Rub*7U1}Rr+A`l|mNl8w}&R z*S_zCC3H2<|F%pP4RF9nImiEXjtTEvLf!ut`(R8G2^EcC7*Z``Ci*q;Us`%TF#B8M zbGHm}G?4H*{b{bVe$VU!1gVkyI!VfmB!0RbOcw`+M4M^`>h{NbU>WUGjtQ~EQUi-) zPUS6-fFRy`0nn=gxm2}+5@*p6?8K2lBTxbr$d!IIzy@};@4E6QfhqM*65I3h;KBC> zVoPD>8NOP|1+v9Kcrvv!uBfPyL#Z^;K(sA6j$U)p!?U^fXK$O=RZJSM^TD0rH5tKL zBx0yU+~-FE1;L#rgOTk{rl;deM;alTyM%@&zSsHNyEV-^@yAbTuGgyS+M5V**t0)E zu}PD%ieARYe?dh*|3Ml+bocKTa|FtiL}tHsZPgwK(GH+{@9N8}yEn8&Ysa~* z-oomF{!$T5M$!JRYM84GL*yfZzuwA=k=pz}hZQZO{mP_B)}%`f>=Rv!@9w;)o(1l? zJOog|unGJrr*Gh=9x}U6QJag}#yo&MPaf^^o3c82PvD=Iq5OV3Y_`I3&M1g!6V04?Sd33Zet z9n$yVJ@K?^ViX9W@vBeykjE2Aa0|f;!<^?N72tS0e^H*`_B95Sjq&{{^IAzvg5v}_ zr6|Wgg%S-|((Ah7dZ`VP-rmU#12oIFy`e)TVYs5@e46dcqG&l?T~uOqNn+1Y!b>Ka z-K4)^R@3+V1&U;52VOB~X(TYG;M@Mh?H66?=6LB(P0dYYoffX2X{z}(nnx)V6*=zu za9^e0(vgeWIt%PiHT;TB1vP_iYn*vFiBGTLM;}YE%>;-{&`YPU?e4p8!kNIQ>XX>kA>>^Jfa)vD;-Pp?5TGj%QS)Qlm^v-K<*-kEEh3Qr1H>xK#sG%bwAN4{?bZjiW-e( zI%tWI_j*Edsq`GFBH!1&vDo|HT>xzl^Gl6rhyOQNV+GxOoxm7M@ATLyPn4&NSsg1VYrtaAIS>Fz+GHM#WP>P_RI*|jBg*Q@ zSj8U5-XhqpM5(j@&yum)ld^tPrX{xt?0Pdk^TCkT`cMrrw86%h>R zSAvRG%f;4k`ca%WSjF$=%*L>iFe`pl)KuxUn8{r~zYI#_@l@9LW~LHHA?N!C!Z<-{ zcC#uiPZq6u_Jg1-^(~6yr(Xq;`L3mZ(aO@_V~(;Uaf@PTQSOBV&U@_w+e0;GXK8dF z)PdD}$%4i?tsm`bek=2q;dEWl=m11(KEQK*@g%0gamy(WEjnBEzWn1-XCqi<{5~TL zw+?W zW!Vq%6Pc7adLoEIITaB)DFa~xfK?<)wdFy!(-VY*%Kw+%pBpI&C#tJ6#rrfPedd5Z zvhHk|?RjDb(2q1tu$4$Z!OQr*8^(&1lpkNU4hW6fYMHubTYP+C9XRpElP`K$abUbn z=sPmn&_5bd?&ySr*a4@-v3*5oQ7nT%AC~JGfIw*CQ;7CLHw7}_bkdlDCi51W)fc2p z(_+n4>)$Ut_W^N=(s@_*ha-N$K|CP^I%kgax2uC`Wd`&*`(s*ys~Qr!$u2Q8)WC7^ zu1RJw_CCW4rMXetKzVUm<;IvY9X@?=T1pB)o5AoHeJT^7(<`fzeJdnAY=RS8F*Pff zxgDq1=nGI^IN$F2Is{Y!kEGyph`0;Z#F}XS_kpTT#H&kuyA+6n-EB&3mC+180|C8t zqu_NA4Nz(Br&6x<+7zuReBCp8dcXcE-chTn&EcG#DS_RU$g-2(wbVQXBWTxO2e=Ob zQ`uE3i2?w5%JfBL<{VP=R2QXYr=^2k0 zm#c9sS@q-J7Myz8ttP<)X~Yw}C-`b7`;*m?WC}bxUnNgw|AZF6>zb1S}|eF)3Qf`AI9fuQ4YUxsN^pWa^{y0{_3$)<2jk_Y;XI9JoZW! zu9M?|d%xNQhF4@I9{vN41Q1M~=A}Lzzn1EhOAPV+SXSc+(N$3|Rh1F#c{^s)E;F^EeMF*X{0rHhL83p8W zw_J9M?|`7zoKhmQuj=>Lo~$V*%!fQUl@CB*Uyv6`O^hWtu-6QAy|M}Rb|7z&vld?M zR3+{MU`b5NK=n$w$-vJ6$4&XN3-?s!9+jjZ7qRmIR6LOu$>8C}H4m(<2SOTT^kw3i z>z3o#WryE_0%fo;{TZ=@L#1vdYq-~OUnV(wFN+z0JQt`t9a`wqh29px(h?jchk$}l_p`zA_trBkr;J9?qQ zW7&ZTmY{g}wr`TPb4#Qtc~lKmVm;qauV<9@on59^<((xsIObg&8Qmhx)KW%_gqHkA z1!V92w!_g}f#D*K|3s4MpI+xqd3r2kj4Rc7vp+FJq!^UEWr0!Py!h+P3F6fX4Ad*}K;54>fk7Si0$lj4g*+YrH3xu)C^ z;%kgp-~In0JTBw2Hdi`aufED?F&};PuNVGyM)={jmqAF;g#-jta+Ewy#DaZ+&uTWR zFopZkd|1{jt$7EZ{tK+P*|Cd_c98@aA<~JvPuYs!)(H4C3Udsi6f2ePHfJ_n@pC&S zULPFtaZx`AbWHN-92~mhW#}KUv-$mp$L|0CIQ9=mia24=*=T|mm@p><|5FnEf96tf z*!OYg$TD`q@@ z9~7vU`t;O}%clM3@9h6B>&*G!U>kQ%6#0L@-biE}$eXAJT#|2hJU$d6|9|YgcUV*H zvOX-LC?Ymgl#U80AQ5TOQB(veDotucdRIbk5m69OK#&fhC?F;D7OH?sCln#H5Rnpk z3nY*b@;(0c`<~-__WAAYyT9+x&&!KTueB1^dY+kiX70IXW)^8A84V~E??B6wgRwe4 z6J6>_+qucto~jk)-XIk$P0!~^XG`)+1%`m`r~59f_6@0)0CNm_M1DA;UFtpX%*^Zg z|NJHYx2^l%n;SlUi5vtTfk-LWB2W!Jj*?_MUjob1?K3`Blcn zGg6X|_dI@t_d72bns!3X!vp-4QpS`Pd|6ydNPr!|tBbA~G+@l!^XrHsD|P5G1scr* z8n|cgL7r#7{-`v0_I=y2VZZtwx_!r#{`}FzBwm-D-kAV;%X1Us)RVwf*RFuy+ZKi>lSNaUPxQ~P_mS!Fw|1Wj=ozMMobHMlg z2N4IHzaIb3b}i!sk0;`zlN4*%U*G8$DgIX%_CbKedWwbrCr?hd??en(xSMV@sE7ZV z(0BrgZA^N)??2kL$dCIDA_Q_ryT1IYvj6I3|0+I>OvmJoj=w(tH}>kk<>UWX_5bfw z-G6_9Y2CaB4e5S+bIa(>iZfgX%{eeh?Rw+)`#SN2vM&l(uMM;MQxbWW4|W zmsoeqFFXlZT&o*j5(qmDAxRhD!@)qB_8#Z~Pp6WOpDj`wm3PGn<2?2l$4Ee#Q+wOA zzpJ)GwLCXws#k?8YN{>j*0t{8W5bh1)q zh~mzRWH~RH&Vr}!{o~4P`-dFCcg51t!vKzTR+kKF1=(r zwc)izU6N@7yoSYHk`L*ftR1GPL*8_WFAtMX@=7k`L@Scsjp3?a7GfQK9@W1k;@>L{ z9g}N7?j&>k@qr(qpH|TgYN)lrBl@vyD6luZKQT7)g5L1*bUZFKELFeS&ma;^#F zh898YknyV751q5CRMx1=_OVy|T=@iBoQJD5x>FmDS#@6Ow^GfE?P_u$7#q}Dpk@Vp zYy2mxYvw+;Y11XmUe}k)?GGmTP9hc7@sZrv;%11$aJ5Nyvb4QaDl_9#kCpLK$VcgG z2j<6Ye_N71>jI0B|(R*mA9Nf>{I{ll6L!aN8Hl-N6|?t13H3Bp1;dT z-iCgr+d|CZ4n$W-|DA?WMH}c}xi~|A;Hc{80d9#?b@1pz(=_H(l}%54GFePJ>osu_ z)diE*DlkUmWtasw0!|G$e&7f8?K~1>qOKjC64#^aEj%&)X*oq1XHL&^!!upyVDijjBrpS`R?BwW56Od;l_`_{I_5DDl9_&+#FKJWYI9bHFOowZ6XxK?B zP6R2OEGB(}Tk@}m#lH}rV{!YVzD~D^O8vQ-Jd*pqzD}7%WHqL^Wtw6~+)e0*%v-jP z9E6zEH_0#0WphQ#czknJ##g_$X#3Ge*yM6bZPRwr)Af+Nx#?-0{0MGZ?W#JMZ?eKg z>^3SyrNgiv8@Ku2ti%S7;8utU;$GD?90QZsfE%py12xSeIiy)pmZp`81?_BfR4Fg7 zJIX1=(3El$|26cKlU|l0h-e5z5OgId_X6;2*ewCnWrkluN&H{m{#qn(}d<%;6yIPQUDfr*1+KR=Dr3aS%v zRP5m~p*RTiExVLcmC9BkC73$yhTpx8tHkc!mmlZ6=$h_t(*xC4*^%JiajMzHd@Npe z$A8)5p?p}l%zo6+Wv0_ig-i1BJ7L3GVTSIe$bYdJe|^8t+(4Jt@TQ{Lf9f(_KuA^; zDMmXax@}EM0vrYI(j#QKkKbsy5aQ{qpc!e;8qH z45*-d%ONXi<)Doyh%*-{(x2OppIy?)9h0&j>AT!1>o_9bZ51-Vwyc!Sq${ZZ4ipy; zNu;!V)i7#um3H5G(V0;C$?0y2Ow6}jgUg+}@efhno4FgFYX%;(QseWN!3w!!41U`Y z<0<~>aXWYl|A#zE-qO6AM%&m(~^~ov)48L}wtdHbP zgjYeZxXrf*3X8PNPH53q?Dgr7Tq~MA!VcJK6aIu_b%p4VXaV_}j=PW@_T2Hu;f{kP zb!CpU_#IO4^%8e%&=tQrH}t=g z|2WWi?jHiFOr6srhV!OA%{4S;MJZd~%6wc=+cLX^cWv#4uYP4}>Aez*r{I9WudnCT zO>KoFH99e0gp+XGO9H?a_vKoO%?+iefSUb5=}zj|pL*5V6#q1Ou?3cdUw@^@(8gJA zUzNn={}IU{?X6S4HQ$giP*(_T5V;#ySj4>q90oeCv@XBCwRHC=(CvbB&6+p2MRsx) zy?`lH?KrJrS;iP|&n{v0e8aac_`+Rx0%u#pZ7Tas=cmprsMBmWDpROCrGT6>x4w8+ z!3W5L^xRq>%#;MJ^;9kU;j`j_d#A$XGeX=R9K#GEmt!8^V)Bs>dh^>#aKVC35Fs#C z(^3A1_CJFL02F=bQ@MM4j8HacY~8z5NN1gG8}HjSve|7HpZbPfGSZ;GrG25W7R*y( zWiDOWm=jwc?V0FX-#xNUfG9UF!t^q{YsEAz{OgCPXA@-BAagtLh&x0FgXY7eQ-D0XE~PLTj_A6 zkbc3CnD|GiisAU^HxX#PuYu{xHo|A-f@)M%hwyJ3<;b4>A3YI0D_1bVe?$Zd`;!li zj!cYOkzQFxR)HRmVDgmcFo2Hacb&C2YahRKF>Ps2BUzGS2m_cO%} zMQ)VaW2RHax!bo1_)H4S-g2%}2JIj@?}7-xt5 zgkU~s#c66Ho05oEGf_N;dC=O4viSCj-8dwZCSXs#BJj! z^m$=D*^6m2xKrY~XLh;$XuHATS!?2=!F!7z$rr4sEyd?%E9z3UO8%Y#~9@ zzwgyuI88@u%lVw~$?cC|Q7MCEHQ&8}6%IU#-u5X(D!bG!GlE;*yHoF~M_Sp5ttd;{ z5PHOoJ-N%w|Fm$VRz0<(;lrJ`O-?wJedXd1szXF&bL)!0A(vZuC9*Kgp886?ay7m= ziYcf)eQf}9+;qnt1IFZf^W|CTu+x_uDs%hC{V0}gebespD&dO{8caEBd}Vj%z7^N$ z6zUBl$g30QuX@g2QB76AF2x_h-~7|j`ghLz_lkq})^;k2U=RK{f%4ZVjF z!ji3CWq2XC#iv_zXc{n_Yr<0gknit@eir3aI1JAekh}?&bC~Qn4B9+}%Y~wWxT-8-TxEmlcqhU) z*N@HjM2?PoB~A4ZWTqk7Bh zONJK>N~+{A<)!k}Uh|e)sptP+z1?drigP*Pr$=iYIkI%J`U$jSX5EYiSu$)z!74^P z7AT3M^P^}vDtvvI6*uh1HqHeOmv8x|x`Mib5c@%=NWQyq9bdxY#&~7rhHisF(-xR$ zzPp0=rN=qNUCqj*Nr@5*&Cc!^FeFL<3SWC`Qh2XE|68F?Fo>618M>i#n(_6--)ex|8t!5E?Tc3u0$!Tax9!Whuyb-)EMlJ*;W zamn38c?O1ulyKKHPNs@zDe%{g`+hOM`z?aR;V?=BgZZGB!5G+O5v1`h&RH>L@TlCm zQIExvQ7sVBbh}k?cHIE{sHy0qJ~fzp3OM}mTjQ~#iajF?=oG)DPH&4u)BbOBr8`_? zavo$0%k6KStCKVJ-hbszu%^~a+!vYO_i_w*Y@gT5h#aBY=aly>NcN|ZQ!-|GSnTp< z5nG8%(d+O@s_v*iot%bRPL=lvut_s80w}opGm6;R9Rdk-{mDE?NBrs8)>W5Jj~0?U zYU)f&b^DeA(rG|j(Z$bA2HO|<`Kgzw-gv_lK8v(=l!E4NtK5iM)jN` z+SiUVYz+0(+M27zNg}ZsF%zR@)fn@YCFnPwkpjoOQN`m{fAzNis<(eRjg*8BT@Y>T zIini1tl-&WdOKatE5ld1S6c6#K=YpKjK&TACV;Fr&lP=MoMwW+|Q#$qx{ z!wDk@`f4m2qsBIeOyjRQCK@-mYk&btykA03g$?$jjf#|R0Jg`@HsS8e_38I=L{J)O zG12tWJsD7w)Fy+N-^*p;_kgXu660Svlx7J=X~=oBr8Wn(6)FeanGQXb3B}bPZ;CgQ zDSf_Y@9}*-8biThZ|^V917k#hfcLZP%hJl=h@zS z``!~^!ffwsGnJtCt)qjI#c1sM>Q^o9a2EvA27s2pU5URBPAa)FT|*Tr5ysoz`|J%d ztL=z`bR1{7*b&bgXwexjstI+ubLUmExUPX4BH^lK=qAb(<#LPcxAE3zye!LmZm5A4 z`z=SiX4I*C6-(Rw30k?=Y%XCb6%1&_c^q`3x>|1AqU-?rz|Lym;z*s5p|~^GM+0&%{EjY~CI|bFD{bRf8YV2=kkG)bqX31q=}Fy+WOru>YVo-QF#} z6$0F<3l9T;qklL3#{NZttE|9vJ`d9b{c>&l@3cjINf}6k13WKHtCj9AqVO*P+GBYD zOQ)Y2%)jP0~RJ6}cu-Y=g?1sF(a7>-?J;r0sL9WiR_xyFSR)9Tl+D9t0gmIq-NV|cHuTM7|9f-S8cFuq9LFQDcmO==Q zwR!Px<^_0e13+PRmeFGMZk&iJK07rQaAevinu8r@B=_pbU~=(y;~;TI)d#+V>dp??I-d@RBGrO^ zZsN#hb#B0Xu#EdVz~mpHCWu*A@UDTWx!z0Mv$v7o$7XgZx9~|q)|%`xihkrW;j6km z!8>cgHec|S8kyG2`?58XBzw%>j}^Kj4J&eMg2vsBykue}%4U%@(8kE=#qcxxzO*J=-G5H19z1r*{+8~{&;K6%84`|F9j7EFu<%eqGDT{a30 z3Y73fd`62_j%?l_Eg!G8Xn&`gDB);VII$*>`pye;Srd!~1}-Es2QZw$d#ACfKHV!h-(0{vs|qb@wHy5rg#D>)346?lIRsiSjDk%^txvayFi_|nYur~V zV({o>dHcZ8OtsjP*^hnF%x6)cneJs!Tz=9|FZAT4b6-MFzlkYosQ-4QOGgT$SELMzOmSG+Omt|r#Z-~ z>4s7`2_^|t-|D6?=(7UafkX@&+pxtobcqRYdW_V}61tS@M&Bn%?BxF8QK-%lN< zZl>C2d6S!Nh21Qa{W<5b`0=g#Fe>$FUKYdvfC$%#iQ6e+x~)v7FFs1GS;c&p?$#u% z2kq`>s1ulOk1enH=I|cHpW=fULoIEW`Y_TUqB!L`r#ekX-me-owday`{L}((dv8Ul zq~gtlL7b5{6p-*{kwT+?Q!RJ`7{_FeCwvmQb~P?q*r4QKIt=|Z`B5&zBZ<|Vn6LD< zY`6W(RgR)6y$Dz629?6!IHW$#BD(>9W%|dLFi9Tvq(;Jr8SjOGJb~zJ^|z`s1a&|q zD)a#+gm_irh0&EyPZmD*zPIctY-k72PxCR`_zHt){=Uh~xE+KxeK}Q;q8^)|Mnq^t zz`u;;ot`n>-xn0YaVaqP@LTO{&6jfs5fd4&`4EArfC@*)cEY3)xE*uJGK$bnOx6Zp)U+W8oYvHpFcATe?0U?eX;J{gh3mB+0&| zVQcKOo!5XI^yyOo0&9`B&9q;Wf)0a>n7f)l&-64rN^=bLI${k5_dxxr<;`7mYQGMW@ILl;S&!M>ozeZ{l*_NYR-w^U9_ZfS?Vv_j)G>IY_jQXG0Zx=ljM- zM6|Dy&uYKKin$gHRKXGihrD+{`MZNMVLW_uOmgIfdxPZlM*`qt|j>RkaYsjCUS%o>StNl%$Q>T$*@JuHrK1r zq*4<>L&KZ)6+26qx~|Y>YF0k19N-@6K?RkSx1;pt2aC$b3l)Qp#x-<>9n|a(C6Rx; z3~EftO83NOaYJz?%qDOrgb=fyeva`lwSFFY?F=wN-dZ@%h07HEeH&iA_v*V#l!z-3#@UsV zbt;BS#&P9)tLtB_^`~7&1drv{I~N#0U-V=_-d{WNQe)Z!E!n@1SmVAh6VyGR1%3R) z)VN1*g?x_gCr{Z05%KTd z!8tk{Ng9D=61lq3`39vUJV=Jdl_Si`!Kp19XT`?u!TNoc-)c_?It8yZc($F`o+6-^ zW=6{GR6!MMW(JCtZO^!Q?tQIfan+!m-5O?Nc+&Y`Ip^nP2n{@OH<-IReN#X~ciksR zqO{R*sgm}IZq9aZllKv30@?0v>}3{a2m5|6#Q{_)atpb@wl(;yNY9dH{pApQsFSr| zs)|Q{CeQ&%(QY8yI1 zF|h{Pn6;C@MB(X=WUVpiQ{ z6SQHg!+yi*Ioo-Ae?M3E58=8KzgMbboxjsQEGDnduG#SZ!=xZhCRV|dLU`t>3x0Hri{U- z+}6`LuwXYg(7tEOk)>!r$<;}F!l=18Mo8BsPm@*no>FzTNm8dC{Pn<@YLnCh=#{%y@M)0WMhpHCTYWwX z5?%LewVE}i@w)syIdoW4ew+MLZzt7cA3gtK=w^_o?JTOBAi-f=u6ByObFdv#nOA-o zS%yrf2#=o7j~iG76{%{^?PM9{`KPXk};wV>y0l2PVZ zk|T%Ipr?RV&WvKQEOHp@#yQrkK!e`USrb%)L?qXsk`zVjT z=jTWr*f&roA>mGxrq*rx4DFk!E3bB&e^tMl=Gc%^%jEz|-n@pFJ?np1VlZ@zTyo%V z+vI|0BnyY8x^bJjtIW}M19pyB_0_BJ-JM<;f1LwDHo2GWr>&8-JI@%4>?DY%g!I*n zSRS}NAG|NGViAt<(#d_;%Ozvh_u=3`XNy++Ba&@T&4h^ucKu|BetwZ(*+?nFf}aF( zb$MTL`AD^O|Do?Ap-E{^62)!hJYE_M&SUCFz&f&}-z6R)E@|`s^uP$hDM$T8n}#RkN~=vr z@7iGJveB;FJ(TGMVENj-(wn5aw{u|q4SUm;K6*($cf1vLvKV4M`;>g5BSMkJk@NJq zJcRmpW8r1S3lTorq1<94!11rO6BPWJ5*n$r7jQJJ42Q8`KG`0KP-`{h^ajrB2)d@u z=k6}I{}Z9BEyO9Ka)(QR5Caf|@YX)djt?Nfb3deCV0=8A4W)-4R9mmKQ;FazH#rpl zsZy_+hzX77Spvbp)W%$Y243y$xtGlKsi5@(SJrhSco*ZE_rm*hXUAZ5iG_-#QJ;k$ zPO^YVVd5?rRwJ8H1pxNLQmwe5EiZcwHmO(;Tem4_(;2~KuDfQhtg#3r-|CI5ZHSfi zaDvyEJ3vt-meaiEat@yl0#yDPFe3Q%O%Gg%6Q;-)gkqV1vES?B4~Tk>xV(eo7T)RvF#=jQhEeQjgr*ey7l#rz8Yt!Q5P+)G0<`~lJ`OFL(zqN z-01v3ZV1EF?`KK=f#5|J?cd3EWg(WUkmtkW(+F^21cKtqb2pf<)wjI!hve?jm;Ed- z7dEKYDKo6myW(G|UpSQf+^3*MgO31d*uD+AcM5^-@2-J(&5Opey$yywJ62KW2l9@SQ$#Tt8qnK4))yk(g+KWRyw3FF z_t6oBkLXO029xav{6P25kYjR+9?Qw{!j|ptqIi`VoRzBp>V;c6L(haxT1jE(qH!pU zDWKy-OiE5s%X1s*AKlNlx$i~T#DL*$4@933KQiZ_>m->250LY(K_1>*>TKlp^9W>! zLhJa^p7!uiqFyObj4GVUQ;)-`mU+m9Iva1aeB;sXzf00Q<_tKB2fXq?|2ICG=&5H% zlX3?Qmn!wqhz#AZN$tFc>Cc(xopANk3JoLWuC7G!z9#mR zShlwc$$M|Hf5pb_o*u}>hSLc<;EI(iyvF(g8EAa{V?wD_8{E{h(X^po+I-gS8XY-V z+T~i!xg9z89nTkzy%oIA4q=&B>WwyFt3A91D7%<^gBrcl!p0X38%^aAOXDW~S7WHo zx@`(0GaTQQ!i)9JbFWfKbkrlNR;C;%6OFtU5^YtPlJRP?7WkaKp;U6p>}X-C%_@`@ z_4l3NEWkA7S+;+bvkyBtpl?Y+$Un$=dflIJ*5Yv4~^l42Cq#o^g~kV-3MplU%j z>}QLorv)9@+2SYPPha@BH%{uD}W{5Yl4&F=J%6VVngpBkO6YbVFH6n2sJRo7d z73W{gckFw2$t>$Zot>_SAE-ZAcG-Ffpu+a`Q$kJ7M{MLlf1e5*qRz*Uh%2!SL;F=v zZX}4Aa&B&WeK1$~*}zBMv_ar0av=Y{a#WDs6%W&jRf(2ae?8xn7K}Ud`odk#iKN|u zXEGjRyj_Wsf%FM#{=v0IoUQ zG0b!G@qf-3u#hA1BERJtH}kR4r5SGPq+QpF(G>sP1ncKbR7I+eq{G;mbc`9xM}Lft zt)k~@8Wyvvcm#Q>2}{PkX?BPj=9lkrQP0RULcq1fI&CxJP?1x zRi6U``SlY>!@>aE>8_;UxxOkh<(dNCQIg`;r@YFb56!`4@&|goEV`n;IQ99Yk-Ci7 z% ze~~P(v0mh1YsW`rus9JwOe$sd$M15)K}kZ_;0{xCE=(E0+`pM6(= zO&lTbQ6Qhga{H(&!zJykcUtNnjg`x+)xS*ZYgm%kSom8318?yz9TV&2C=YLvJAF6A zb!C|$F9ik&1{(7(^sYn8y~~+r?a!DtPs&to1EW`ptycYd`0Qu>Odn<#oV?Z&CFy9$ zei1&OU~RFPDB-lnFf|27+v#W4NtJ+I)+_rc_VOY%S~;ZSiA^_*|!E7NzDbl2dzwcZ{{EyNrY}D^Xtml)y^8ELU4uDWotBwe19glKukwTktw(=Q=|8tXw(Ppus<$ z*CpXu*|?A+IEGq!CO1G!rq3Hj1A zWN2F*mSf6cb8VO~6#sB=&#FA=#?$`7RAS2blg_(Boh%KJwWfrxK*x~htI@*_M*WjR9hBt55p|=gb%56+2*&)=Qg_- znU{V-XUJXuyr+4(NI^p{BjpyuwmQEW_hnuX#vNy#g>QCAxzjK>e<5=3LQn462;glM zI+@HlCr4jU{lUfFdZYK;t*MD%t|;CHno|n}z)sB^Srb$R{>br6d(4&5l$e&zyC)5& zO@zKia7uOif71yZ)y#w~mb^pf3;U6pW-x1bx7K}%Xsr| zT~id|H0zNK0s$9j*y*d?`2c3@RzQXyAkuKrOw2j)YRq_FNyWtQ;4v<-jlH3K05`~Buft`sRPE_QL^=@UmR z3U&+53sMZr9c%;fGPm!tym;e~r~5w0*C7`s8}L}-$ql?TY%0?IsNAFG?)tC#qbuq? z-rpk>0Yf#qNvCn}lYb#X{+5suF4uO`73sSRyo8K_{_>0}78%SvXBL(+_}B|q!c3P> zUi#jV!c$EtcGx#;l80x{EPV)=@nMsQGCO~Ji1lQdT@^!9N){c}*D9PIPHyuHxUP(6 zyHk(R5r((Dn3pKx8Xnx7%g*#%`!PuQ)EG$o5(s>mu%LKZx5YD7A`|Gh3Vz<(ef!4Y zz~NyJ9gGuBR>K;q))nPUQ9p`0EjCYrK))Qa?=W@pXk~j9L!Nof9BjLs5lg#f6WNIn zYtb80w5>1I?8hCI!L-)Co_Ia6k{vWQuvQAGFKTpI%4?*8siZ>DPy! z`V>IF?4d@P5pB>;jBgl93-E&u%$0L92W*`y@<;n8)c8l|W9#&ZoQk}F0Td3=wnqHE z3O`BO-}gc7(srq3%%PWuT2rRg?xLP=I}u)I&^A=nW9Ha<74uu2SS}3cweDkiKav5< z|JeOjO0LBaVoYY1rPiB{(H{ z+nc4=I_uGedffQe&=5!2Bb;wjtvh4*k}AJ^UVZ(O2JkO;doAN;1 zqx+iArslFxPb~@A|NbfQLBg56_4@(Q`w}_C;ck*7^4%p9nUdg|7TaeSqgR-vV_K)h zuN0;}N(_oG&lHQTw3|n!wu&<+u!6x_IwxH2L?J%?Y!HUc=ZIrKEU2ai7zbDsVpv1mo zuNVSKvmWfU>*g)q*(UJSDxNydeDg4<_I#@JNzB#1kTKcA))L%K zCKEkj<4^uFk{{30XLm=1@IPBV`BnYk1Prj>*V-()4}QKM0K<2;;h}zfFnm|^Pm-;H zq}{iR;oY&eyw8*#pYD-jN};7u6bX0-E>T*#%HK&tWhg5t&(drM)LpFFFE(_>+n1_+ zalyaKm|;mZ>f4g@nh6JQc=vnLOQx%q^E$n$BZ*S>{2o#^$13^T9DXkJ`PI_x2x&*d zIe@ihe&_!E^vEkKx-Nd(+p0;?2B9JN##srCrB#zL&hK*4U$dmj0(f}h_|iRZ`v@tyJJlMkSv zifYd33hN)c`RZu^Q=G^sY&Nwh{r8`8&tB2X-z{R27aypnLbARJcyei%{P)}crs4UA zfx2)EL|BYdw_VQuuEF}teM;(pgBWqi;Ol=@-+bZ~z%TZ>oJSr0ed+!sNyw80aHN)Z z!jr$xCi~U>sFB(R)#EjTwI1s{H!$$Fu<}%Umt?>E8rz|Dj6(`_2LHC}8Z42>oC6 z=r>gD=Pf`5p5sONI;wtsr2iR_{ra|lxld{DaZn(wYrd6;&mX~-jrOy3gmz=?y(o)&et;A9}~Ja(eIqcs(1nYCj`;>WBB6Q z_Q)h*Y;xGV$@Mo@-l-Hk>8LOLHTgd_`GQlxc>%4quksX&=Hwk7iV^_XwTe8N%Rgrg`jkkQ|G z(POXdfjmx*aGELnvDJ|m_Z{I-G(6IAb0n<-{+sLp}+9L$783{z-zG?ax*W*9+vCd@{PdN?{RVJ$1SfFjN+oLHx1g7 z7Kk?#^+{t)NtK6RpD8ea&>V(EB<{w&nzL<5Sfj;@>Th^taIl_jut|^`&GqVmjrwfF z`yBQ6-rTlfMo?-RdPPZTkXZWPu${-MG{9Rqv&d!tSIG3fkcU@0{#~X;;K zp_)-m{QRVtfN^@FjHg*6)())X3K;Zd$Jt{FzqEB=?rqmfuzzq!-@D$$q@_aPknqIS zT1lK$@To@BbitqJLna<*$A;Uq2FG5Z$zi?Rr7QU!(dO{F4wp3Ui0rRO%5&YXbC8wv9y?C z*55Eq0k8I|o_Bi8^j|gxi(qgxE9UErC7r~fi&o1Lv zL2Xk!c;O_Sz$cUHNB&Z-h1u<{?Yb`Pc&@ASlRKcW{DMgla2|Ngj!T)y*{IOSn z5kh}I;)0B1P!N|`=zdo*?E}MU@b%x zyuu+46gZBo)G#b99PGh^0o9eOD}(C#QZ^*rZLs@PqtO9(V&r5t>9TVIbFX+(u0 zv$T~^S~BdbQmzZpz40Jxm0F_-k6Tpm{apV^d$3ch#YJ#Nb`m?EZHkv6u<`+8z?;yQ z9rGA;>bXyFI3OF2Xu*x8_#`&Q>mHQQK0vKYBSEw(@-xsc^iW4X+`^MYBMH5~lMM)1 z-8)XB%?7XhQof*6cF-o$XjGJ?tkx+unw;ab^Q2k!mg4Tn36`lHhpmglPgkGBJIOeW z%C1D>4J#vG^Lw%5ZIS*rR&uI6DXwmr;6$`oRt;0uCe&=zRhe#&o-9p_Q#W<+NiF+mf`!38b&zXjODEo2eyhDj z==Cw`ZbhzkzbQ06C0&U8;5Doe zH@}fqmZLwqlm(Hhm3~rW!cA)&jgs9cl&|9Ke7>tdb{)Ft!-ZaA8`)hP?nWf}iY$b2 zrS5E3)*Ue+I*?Da*&i)0$JwKoDh-KJ9(aV^tZ`~1hyo$B`Q&kwqNdylW0Izb}O&;55Jt7B!)^v+?juFuY8C3@+M z+}7rO*j%PLD+eRHKFE1^-01DYzz=LnPPB2*%gMP29}-?nIm~5+|Nf$zX{LQb&kPgV zjC+INai(r1JcI1(+SjZVAGc?(qoG^9L6=e7= zHTM*_MMs!Z>@Vu)-L=?TYawoHi} z(sp@4c}b?(lFTEDgl*DwxPZjRolZP1Bq&1SSpjJxyd)FbuqO>&q@Zm$3~;B8x8~|; zeh79M^eID>aYHb9{i|$X&~2+B8(FLQb={xmb5%Nw**;DJa?)mV^L&3z81=M5cCIu)Qvi$WC3R+)}{96g&2|u_~YH9Sr>csHmIk%FVp2tq1Stc~_ zQGd9J--(QJ4!L3xMrGghYW)_nc}rXtju>r@8oh+VOXVD%FVMD^RZyPWB{rv~E6YP& z0(>P~KekXE1zGK{ZB;RT@ZV{HUe`So5nkNDEq0$@#J~xX{yxWHJj>saef8Ta9=N|# zc-4N(<;2aaT`yKTlI%W3h&w$hoQ$Z17PpA|qT-UJJxt3+Tu2n7A;-oYv8Qc{%iH#` zBD~nnGnVRm9yVuLYm~dluc&;rT)MdgO)tk9D{8&iKbh(5Q&s`3SC`sbY&;V?DTL6^y0;aQZlZp?W)_h z8b$U&Uh@L!Vc^2Z#;r8dtQ&QD5Sr}|aoc-2we0gGS(|6Fqy(Ip5pg*VKU(%qlQhQH zc7yid*|->TqX2WwmF!@&Uja&VFuVOGXT>}|x(L*Q{Uwh4=l^~drDKk;H>`f3_sIsV!3nL&6UGC$l zu6{S#6Am>4j$|yaBins$W?lP!9^We($|yD>M)K)g*AuU;Nv%{_>%f8PI+S#To{hVq zdUuz!iUmB#xay+w&Xu7c-_7wl>!x0>Qfs&1ekH=fQN`~;wGHg@y~xmI_t2pIaCGFj z3UU#*(x#cAyCLza0P@skzpL|7wb%7D7~$*sb4K<9EHWfXJ)Y4GQ}}RQ7_hMBWj9dO zb0nRlpv*wqQoVw5>}Jh-bOFd+yfI9vb;hfFcv`O3*DXgNUsX%fphr`KvA%6ec{L=j zjQ7aB%N&INEGv?rNbQA&S9>2@pJ)F zC{!NNjt}^_9~b=Brhhk8(?BSm5mUFBtye#gw#}y9Xdtw%Kv=dxjA)uZnY5(=`cy(K##{i6k9I9xxPhDkyZ|;!Y=TZGFCsRNBKyzwSf&KIk zNR2$7GR}M0Urw1|di0=+`IpBp9eea~j|RLv?Z0UB=#5%whdgXLvms>cJGd>}j zRnBwBIBwgayq6NO(uQ!Q3;$dR>eB%;%j) zY9Kqy7Yh%pAI{XYz0n&X?|{xc^<{HEle0!DQbi+XZpg07PAA5}v_6_G-q~jF*hqL{ z+-=wHfQm>aDLCt*oa+$tGas%b5b$n)?3g6VuAKcyT%2_DiWP++s2?`MRTr96y!+`8Er^Rp7Q%-==_j2eKYz7r$w*x|G@cl=NGmirUt>SHTz4JI z?L7GsH<+_#G9fj*NuDX>y>@E8{A#szm=I0{EnWQjpEdng@CID{9Q3LGa#B?`;|1U4 zvoaX^O;15Y&5Pb@5b7FApFR9%BzaJsCpD)POG}_dt8&)~K@>e3Y(DnoJoeq1x@07} zvunE2@y=4{+~F$MemxJP6qiiw^ePjZU7=1{5w@1_ePXqyu0W+_sQV0v>Ig<=)i&c} z8GaeEuFIQEa*}pwP$5k%T8VtUS84VCZ?mCAfV+J>6 z=pB1SE0xgGNL4=q7zz( zW}RZ<=KA$5V8>0bd1$6KPcb363L!}x@@`+P8&+kLY35t`UhYuxvPwdwdmwUys~%dk zkXN}<7HqR>Ahb=f5mBk?R-|uKIwXKw1(c!?g7hN2*U$n2A~rxpid0dGBoKNH5GjgC3B8BVJ4vX4 zge2d?xyOCa`M!JZ-oF2Sum72lA7nji%{AwkV~p8Yi@_%by2PK7!)At&aZw+M)dNJW zgyV~KI=Z2Z+19{~LRJUnK<_FM9Rj&j-r+?6IL(Xw_ue_NRjmC*J9p*Ql5E6g+*67uZB+X~t_2@{KzA z2}yYg`PzqG8ehO!upKK%M#aB9cVM1&Wnq7A zcp4=5==<1c!1e6A_5{s<_r=!b4!#5wnMMsQS^@i+4!GcLIh?tT&*4VTO3TQ(&$IfZ z7)}lNk(!THlRH2m9Nv5|<{g(^{&pMt-?6VlSZ*A~xXrt9&9L*1_&&XtF7Agf*Uq>S z?Y(pq94Cz4EeR)7gq|4#PdejwRI`K(uTV{!%)h@LB`v?V!p_cDcU1>gRFUADsT_apF9K*8LRa+R#epb#L zK3{yhQ||6$T;?hP0O7-{ZDELAg^= zE}Q9fB7OA3Zj9Lj>9U&IBi-eT0m%cO&s*&bC7C>lJ5n{-UcFHF<50^qzp+{9gc#12)c5dOAl+KQ1F-93Rj_YbIIqF5o;geP z`e71V#shzUxy$du=H*xUAcSX7Oq^^vgH&dZUod6|OeOe5@8rB9lvFa)Ktayw*f_&i zgStWzH5tpQo#*OxO?N;~Gq{%jBG~}CNG_~9Y#CgNu^mcl#OkBz4CjVZKT{M=`@>EN zA|kO#qAmDW=cQV-AyX<*v04qL3M+eMBNeg*Z)13uo{1D*+hM z+iL+{)xlec0?80$z}G(dIEF65l_*}>#M7CkZ`_Bv!2RO%+dYL6%}(+61;balCVT-Q zr-IUuA=*Wx^}RxzgDRwH&W$u?kUuy86zyw|5W zyM8|d3o|d5QaNn>=JJiK5YkSoi67W!Y5=}>;uI>W5`!Ytw`XFOeVWoQwPe-qvQwSz z=+aF1aiu_YaefTrwUu1vt_LL)LPxzTv9o-0XgJygSnqMSL!GgIMuq=$ar?7nK#3!l zj<*BB{#mb7JGqDBQmGW^2v5!6>(9?R{RQUB7#UOA#zNs(p0Z@5YHFr^{ZhTVn6x=~!i}h}_9} zueJ4AQcAe7E?}K=tPGl{7H9wcRlM^|Kf9|d2}Yj0wNpHRL*?D#o-P{?Vu=@ zK2qE}ezT^N26ho$OcZ^M#dAv6NOjpS)M#EO%KHy>UWO)!ODpD9-KZMpr74e*bg0%p z?3?bi_RS(60aeL0x5;acgCBHMdpi;(kgK^6DmF=L;#_aeT3{+jmn9~RRHG~BRAe%` zpJ3^_Gj;;!YPWnVY_WUp@f-XsGzMtc9)*MUa#nX*6t4M!2J7eMor6qwL6l^{zsB>! zhE92?&w^iNoS%I6r1R42=7pzO(u#@^|D8}(G1+6&66KE2oxot7XFGz8ZY^Drx0dCq zQ%dY9a*y-kE7C;gfe_|F^*8BqT5Kp=I141mCEM?eK=K* zg6%H=3ob|P5L#<3z2Nmt*?l~@{=lTOMVgoG%7yc%2B`DMJZKR%p|H5tm1Faqah2Pr zpu*(@Ac(k6Uw-0BFstinW2wfYhR$ISYAG1;%Z`n}$tr zog!u*Kha5L!v^sxU8(#UPqevbe#z#)SMdLp!2J2&Fjj{h)0=bY(OuQKw^ziulwt3#_}+(9<^^;{D3-4#id9H437CU+uv;Cg&*In90S zk;0?!W+2xRC&E8fj#2Fd&!I_kM&pzKx0f-+CgXs5%Ti{Kw*4W~3$R?m>JQCwT{(-R z4uWT?(p;{b;9!zcLEa1s(C4HIxQHcPNB_olVlrKZJEpGu1bX}XRCG$+Dyx%8L#jd+ zkVtGN?K;bSXX1+K&p870%Zl658O@txsjpG#-rqNe({ue2F=wT}nXzN^3f0fGI;Bgw z(w-hCpWE!`Ig(tzjZz*X>kk(Nx$jtgtQGiv-A8mLb(7|rqynAWWMIe59&18qp?r*R zA31THYgpJtS*_>Ru*cEhcuuu9n6PI;qR%c~2~T_Wk%{#gtF`q5Em%v>QO1UXSIaQ& zJ#N?O(sU7@LrNjTu5sCBVCnVsgP?T ztrQY>gSTAhE^CTHkonFPO#bg=4RWQZM8THyo!zEQBnK9*e8HM0qkT_$7?V5Cx08}} zjGL0em6sP7QuyZ6$|eE6IwzdodgkEf$~yej%vZ&~hGWDIH#W^b=`5XMPdcfLJrC>r znOs%tsjAHoc+!>&P1V}lsNlJ#Bvv(de0AIDw{%B3KE!NZ?I!$|BY*GKn4h6gX#HVp zvPDpJKgzUK5Y~T5cPJN6TSzRAL%8eiP1^G;2)Ht;y2vBiDP~n=f#K6})Ys$nypTdC zvS~;rI1=%g97y2S_w{D(mUADq6U6R}*Qn6KEe3SVM$)(u1AMQLex|oD^1YEw_InB~ zg6PXy&y?k__)8goel1e~X$lQqiim}CuM0g!X-#m|Gg{pO)*lYek59bzsbc|*tQc!N zDXZ#3HTDfP&HVP;Lov=q_;AGhClIwUJ=i@}!l?KMR6b*CP#x_$q#;-}Sjv^hEvonB zDb6o~Tg0G?;;5JKGE)iL^y>)XPa>vG#E9jM?PPqEc2Hx5IFKlz=u zYLc3Pbc>mV8bw)?o~esFf>N&r|vuP&A9z`{1FbBd}P3!OTa`itiSeY9(xo3V0f zz25>s)VCa0kS~h)PMgQaCx>TO`nW7Nl~efg6sreXdW_}3KLmo751G^x@JlN#G6;81 z$N(IUKc|b94z4<6nFxA4!hj}u>f~||ZEC6%ga>>nEMOfDBQI4LJkmY_LWWn(e%$s!f9(WS zyq{HC`CU-;1V@@6k*Nzi9`I7+OUg*XNjQF`5s#Bn#73$6hCq~|zE(1kN`mNJMM3(u z*eRO#y@;Ca&ul4%k|tNmW;vkeLJHYXEj>WYz?u_$(f!B=!@SG2Y73@}`RQ4-J8W8% z`z}<3@QJkaUNTsqEJm1*NBnj$cBMFNCLP&bM=ZOv)OQ?`Y3L7KwiRgG_&}-^%@A8D z%=xj4Ff7dZ;@iR%TL~RgFwj)IZX%BDYQ1x#wf(|QeT2vM8(%t-QWr!tPpNz4n3G{5 z{Mn+sq|Sekv662S!;aIB{LHO5xB>IQjS$|~H;q)vU(ufK{^NNu$G8pI!uvg@Qdri} zfhVn}hk#x>MHcrhb((F<9wNHVmy_cTAQGf_Dd?P5fn=9YMC%AA$b*$jBkW$Gty`Hv z{P#}X|DsaJ7QO5qnSKlR>D|bQRcwiQV!B{WIzQSkN_}3@JTxW4w{*&8Q7qfrc4iy6 zaN55+fBb2&{U6JE)dOLvBClH_xwq}Vb@3FME}{!g@+1-n*Gme0IG(cqS3xO%lt}?# zs)n!K{&-)bC@@>=jjU~J%>eBPWkV;h2$qlvUekmThn>X++)4crU{=0ddz#_B z_JGGoM&EYsn)G1<{h*jKLdLMoXen7xu&U6sqO6dTp5Q4uQEQDm0)>a59o56bzusa! zbMG$H5p-qm+yjyswi01#e@ovep#NL6;TYsGw0|uEXRu|`zXe$}W$zbDARk<57iiIr z1>1l-sckW?dRSe5O+qk2sJ<2M-Xt9JTV<$Ig{#M4oqvb%D8?NllYd`E8~c$iMe-uw zD;z26*0p=?jgS3d+h$A;j>@(J%n9B7c0hKmk$*Zw{^#|c;N<-7R(<)+k?KODjqzRf zFzT1ff2NBHh96Yv!G@&|m>F8wTUTG>iURvLeNJu-ZEmN@xtk;J7l()5QqjD|t93@g zDKNX-aSw4Vo!`VRR$I-O`D<_QX;kNn$G$sBz0~88Evb7L(z>1{WD z>$2qc0>qq}%vy7Vz8ft;BBp_F_s!N<%iH5IsSo(<#D)r`W&oosJ532CqJpc*C8(|{!K>j(S0%SsXTwEhj%oa zUBXrKF|Bu3-@LfwjYkD-uM|hIu-{v21coKJOX$is1XdL}FZN{XkAZUuq+!6RsUb?x z*wi!?$h)i1X9We_rRw)eq8Xd|fQL=c+ZgT~r;)7ZTi%&V@;tY(9omz76H3FCx*3NX zHn3(wH0*Gx3YOQsUX!FO2L?3R`v`_A>P%Ot(n$quf9wICzs5s=5BvczF;6~xi%JD# zobWkOSmU_3Y4SHXN2Zc&0xa_2y*YjA(rDL$Tu4m(4}oLqoNOmoBW%Xnw@`I*yKu1^rC5~4-TI}hPSq^&G&D;*z z3I=%n^?(Q$kR6%1)pipwM353;$mS-sgp~2+n1u1rg0zkJcw{f&qPoeDEqd~oV9=j2 z@E?b=dVf&U@z^hb>NUTVE&pldHIyiMn_tPLluak}Mo2AO+V;hX&~;6u1tw(2O+-@6 zqd=AF>|SftmLS2limA5#+U0dyymoiMWP4f{99c6mji?Y@@iT9U*~XyExKa(nb|6UT zD2q+Yh!P-QdwE8pWRy_Vl+G*N_<~z3!O{Zso74v_e4qgL`A!M#%ykoYbO#{)T#$MQ znJKJa*K+3ER_QC`0>*f?n)@$a26txynqiX+^SXT36xr%Xk-6o3z+^Mp7k?9SOA3N8 zXm2UL^b^ziPjJ}3ufS103R3_F?QVOOm@>U;I{5OUt|?^QQ!v;v4n*qTrFS3OEiE2izvM z*V&|u>S}zsR?>uZ1(fJRHU-dF0Y9<|d1&8+BheyMhPlti74gk5Yk{2DY zDw84?T0i@|JmT#3zCv~tkHNduPadIydD=TrtKl11^1RBU3qS0P4mU87gJdqf@Pio^{W zMAtjUT?wh&S{|fs6aeBm#WYxYy!idbdd4J+z6=#m-WY!ZELVN!2ba$$)==~O{CjiG zjIm%Yz7O3rO9A2me!Bg7Cuws1ksq17Q-dt5Xov+9qQAlSiVI@|R+Bb&7S8)fI3a$3 zXru0RCeyd0-|oS?Xvy1KgB6Vm20qyJlTm_hfUYi9UYw?`{=3L0tB4~YA=EMPyJH*5 zL7jY~vmvdj*aCu0mi=YxTr06AYN38^3}f_kg+ke9+FT7WmXL_qt-mLxaC;ODD-^7(hkC|~?mm}G=HtBEdv;&C4?ZHXyz4r1PYYe#+~cQ&e2Cez z2DaJ_nnd9&{po_c+MA2FtQAt>e=9}!d7pBA_hr)5&##UVnq@PD5SlYLNS|W7CpUlx zP0{+ns`u@@GIVGSMZf_b(-o~x;Z4~1ky-4OdqsjFfXR$SkZVb1SKoHZIci&xUG#a7 z4{9PHE9dos5Kk7gNWoM8OAR#*cKgJt6C;{e&X{pWp!hu=o;cwV=|ilg`6(6J+`*Vl zE;*ZN>{2?@Zc;_E5QuFjV%=5($FV5crxL-W$*Apgy9t3xpSiP5*~AJvzoV$R#@ zWA=J<&m0A-hmD9yqT40wcBV-E1Ebo3dL&r~)_!s=)s1og0%;AU9_p_@1Y6M89rlM7 zTTG6Uau8IgML@P$@MPp9@@5+mz1*`UqDp(U{7DyIFWVG}i11qy$=#XQH6hkcH!ZM! z4xNe7lx(M0A&pVg%;Fe0vF-!%G+r1^ug2ej<2^@RMF@rbn`Uu^>Xh@9C>hVTqk=TA zhPzm>d+hP^k1_y-EMt^2b-qkrw>*k~!H1?!5jGx&9dZTSaT@Fx;|SZkLElxJpEpgj z=CP2siveZgc>>i7mI12xqW$2gdvd#1T#7@_+)8N^G`EUxJk(;)eQJ6u6~5OhEIZc^ z)sC}kOrfVcdi_B(uM2*r9GsF6VSne~)=AXVo<53+c1LngkS=zzW*mvCYQ!hbq`&9* z7E_*xE0jOnPl)hdOsXIxyL0BS&oC(qY>@b)^`W{EGWb@%J9%?PA83Nk?~6@<%#j`9m+ zv47)#Vja;H&=AfHhb#gH|2t*|$0#F`W*49Klt)kqz`7q(pGUkyX|=KPg6Bl7#0=6P z+T5AwF`u}?sRGumK+yTVLZVoW^F7uM4ww~A4j-b_zzNT`gQh8dK5Y*8)#;Zz)j191 zzqho2>zy?oJZA(&8cG;kHdY(6l02P15v-*>xh3R-`cfnMkforG%^$g#uAL!$$tZ2c ztYLB@B;9x5p>!yZ!EDe{nPWZ1CRvUn8r*v9R6!R^JMIVwWuJWs|3hbpFEZ`f_IC3b1C-~J4ibyVy^eF5K z034%f3O>Ys1GqVt_$C|cx#1)10xmbx{V$$Y9Ke_GYb+Msab!llogg&1DMBYeO1#rv1( z8m{>%&ob?mkETIAAtMkK^HALi*HOEhxC%Fwpesq@oOyx+zoL%Eyf$|aF?8XAe)jquGkL#PNT-6|-MUWu!G|CT19cDdnf)OcltJ-FM*cLR zng|lqfN<_u%0+%*U?ElG<<5y5YX~o@NP+P~Kw(bPv%|i(!Gg%C9NQoj>nFYrgPjK2 zrmL4L&9y-mHNX3%jg-4)v2EXW!eVwlY6CvI?lm(XU#_IG^9RQVn3cMihfbf+UwD|g z4%kJQAmemi^hI~PR~p1fVm2RGM5>&%FWS&VS;LO;XpK*$V~)jL85R-F;B(O;(RX<) zp42O;8L2d|-v{NxxZ5HoJn}KWbNXbH0?Jwi&drtvcAV$kpe%2{W781yoq0p=6N474 zhdI$$!x}qw&5Y2f6uu8X<`8d>3NL^3w}|)k4exu1g#r z_!hYh$uz;nhVSXJ^V%WDt`{tKOC7wrlRln9J*%v=q|e>YJv0y?)1YcaX)(PDA6}o` zPDXDtd{@t}y=%kCi1>uQL~4aBUQSF3xv#E`pA12rs#V`WB;mVNQsT|7D%Zl5ymLxK(nkX3FkKy=jwZLK$TSDoI|an(=Sj+eICvtoTGcGk4?_6f#Aw`+S~o_Z*= z4~~O*H^Kup{&{BNj|5-A>^7jOi-+bi8d;-BO&J&KYgLCqRt0!Dh?_@`l)VEO%v`KnGojuS^g)T<*}kLGGMv= zvYwgiayWL9I9lmmIS#|wu8x$gVe2XushM4&n0h!(5tlm;80GC1-H?L%9bDC!Pn*x^ zK1dkWOESYoCtp$XQJ&{h8fBj7r>ClnV(w7x#E(e%#R;H1GS`5(NZZ_79ER$DpwC-` zthF_OlastM*m9Psc|^h?w~vv*5z+<)557iCMqv2Cz$l@NXQ%iCi80V3vzFyu%%B2G zruXV6O>vd6C$W|#U%$_1#8UQH;Ol@W_wVJx|M)poxBWSzooL!Nnm8iP^IF{C^J|V* zbFFd0X(~Ak4~U8(#4<7z$St-vVM%H#S6z^*PCL$&SLe0~RH6wFl(zF}ZJPETxCI<@ z(*m8sOOM1Sl+&MGh0JoBgLYLT-EeUv*?=Vj#(gV-5`&B(X=mEvn(mqj$%-W-sxt6- zafPqpF>K%NII&8yK*k^kRmxpIB8>_R`_k@)&1z^LAy={PIt9k$)UYnSNlxErGc8}5 zNf!Nq6!}AW_HpPA9igS-KCX;GG=;&NDupX+8z9OY#(=R?_*b7g`>fc;E}-VV(CWXL zmT!hku&}@}@A>L-$!Ujl-Ce~#B9yB47H~YFa*;EM-vj2;m!S;6iJdVU(M-+SF6DTN zwyZ!@EuVYQx5QT^3%LPZP%%cYOufA1#+sp5G6wmb!`%W-D_OGI;Fx&`vcI>pI(HD#R zd!1%cL-C#pfB(U~=4TGifAHGuM~uxY+k1eD^ZtvHSf)P23*^97f=RwpKjQwjauw3z zyvz^7d?Twwi@>}*Ru4bN4D^d>ImKKOR0`_5qX~61Rf7PcFp$G*jt-V-VD3P+1Z?nA z4lALx{E`{-qw=pWd4z{nEAb&48OwvndEsNf4V~RqYZzP^FvtvO?a8h;4&bcC6?xhq zqIM&Y{wTvjH{3JGj|g^1HAdsjMf!9sn0_{2hlCQvb%$dbVoKN%NxJf!fgBZ!Dn_}# z`;{H=-JwsmA$0i5aA+9I%f&#fUdH+?mvYnOp|(V+494dl^WDa8fIIvOL?(7X#|!IJ zMkb5tP8Qd@Y%A^Lh_mH_NmKSMrEx`$z7a#r@}-Vqbeup zpS1!@KgF`+-&s~?En@(?8qGZ!kc{9CER*I?kdU(TQ5tcXu5@bk?7+No7{Zp3Hdbue zpPVY^fBUalgMZe~3OYjHMVW`Q`CtcGj_kQJUCWIDv9%Nkk<99XQq&81%=|Nb74^*I z*`j823l5M|+%c!CfJtJJc*x3YGQwq8-eW}A4O}; z0p<~v7G|W-{-yp*z2a}z*EX;;tK2dtTVOB54%}+qswq?(O?t{1;^9I{?*_FpH%55p zkn60!zG(G38ImFQ1RnKP&#E!=fw6WQ{x{WgH?YcY1 z_mOo4;SoIW-kvhywezs?<~vpy;xKv;m@aDr6pfL@1jjahtU?P4L#R1S)6I}>0&>We zL7=9;VEw@dz2kjN_2FUkiO1Ab~75IfEAUt zm)eHqwPq;{-3e)Sz^DMQ>d6D$sxF!c*zrzViecw_4FNgw7sjw5yzVSt8q?sC?it;L z%@(%^R<31)oTq@8;b+8+x}ZISNtp7hka>8u2PWPly?im4 z=hA9RO{%bmSKQNW`YWRA9!KfQy1q`CbtLOt?Mut%|81@R{n~kb>F^Ycu>SD!3B!M` z{yM>Rq^z}OYhm(csYDEToHFhmo2TyM=cuOV-|k%a-n@IH=@1J=o5yGA5spxGX~gYZ zO}$5^I%q*)JWsua2q%T*J)l;_f=XUWRedq&mamOuaiNaED>khJopq~$z~0>O_2|hX2lHmU{hD`im|Wgmpp2{8q{22+)}ALc{oMUHCa39hRA7U zhRf6+_%J3y2hKOAB2PUKfRbq58`;Yy{p&Waz4eqIH(J%;4pYyJ4R@Or=QOCy+_jfn zduwdl5~F0}eZoka6$fKMPF(C`mFN4BCcNZVvW{azh^qP0;N5k4H9vW+g~5!yzC*Om zJA}0s$GT@ZpEwQ-D|w`XhQ*fMed`MhX3#E);f{K$8n*;l**W6tnv-&CXRldgpui^O z=5%8f>)*lu{|nss?#J}wsw?cAp>4uDvmdR|69J-xbi@%DW?9mt_R(|xfaa6iqzI1N z`Ca@zyFH)p16O+A#HX|#L_Dr5`|g6b@tM=F2+3Ekg4!V$3}?N{pPo2xbb*Hy(^pZw zh6NIQcc;FhOpehb7l`d=owKBWGtcd!cgW!$siY!BMUmD)q4Vd*MXU6jyXc+}G;q@Z zCETuxq-5tA3FAkAvB}--v&>&U$uL(&Q-jf^fn7A6>rv1q=;qV2e|RbImsKS zV{B|@8DMO(E#Opa@B#SIw~J<+X5qJuddD=x)YB$~a9CHbkSJ|v#I?1cGp52uEXvH`PQ1U;NwC;5v121H75DSY9L zxk`;Mn_nr~^sB|H_vG$8)%U&hKv(MnBl&$0bh!HtweGYY9SB*c_%abgz+u^3&IzTu zhPyroUrmzw)Vw+qr?ZaS&uOFHak+DVFlz%Q4c2dAk%JNYaJ{8;16qBQ^=58`qp1^O zieJKqPR56t=yii6ApN=y-HjPHr%ixV!y;xATZxCyN7Rj+;p1||7!8)o6XeAJh#612 z%L5Q`+N3IM+s-iV9$X!*h%Qh0jhfb9!(>r|_&pkn2TtfBqIbM7U(@1xEo{?@YP`77Ac`N!nJF5t_0HcF=ov6L* z@1Dl!lz4#i-pfpIm2AjpzE|>vw+bkd|aaO<2R9%tepI}*X}CRom7sUWJin; z>&Q5J41GHV0_#s^*(4Gp68D=~&4naR|KnyU>M@E}{6!4%a=zOY=@1(NF>9QA^K-oJ z$Ibe$V?n&e>}M3K5W`Nay$D7BtO~2?+vowvJ9m8XCl!#7L%3O|BYqs{PE8J;I4j?E z3MxAYKTBIe%3gYJ85I8`{wDl3+viXKB+Plyp66Q5et5ak6;H@`-rTZGj}4%0H6yu# zfk}KtjC{m+O?q|G;hJdcZ9Q3O@?tHyS**-{21&WfJ z8P@~V;)KGF9ghX?W&uFFWUI*L7E$~Bl6h?)m2@h6{X$=yK4G}okbIXtGmpFsb-6sWrTL$y005BB9c3F zlS7%)H)!evcyH&E@AB~VH7Aq0!1A=fx%Zh{X`L#AjbaH=vV0fWqIqRX^mReQ25Y7V zLm@ZVIaDmB+9BEgQ(2mRK&~+gftpT=8H#BjoHj94Hs*o5wkAPBuZ&9rw+>%pl}BpP z(o_PI!iVK>fk^(KcIo=YDug?OLM6%q1(p7e|Bt@BlSq59A^6;4?&y4+@Tj{RBOQi! zO)IP$KtKwdfY??Hq!$3X%&3=f?fFf!fBSI6bCQ}(KnG;n&;btkRr`hH>ina%A;f5d zT!k(KFC|(~Cj84ZkCr?2kyMbz8&sebHf*0}%c$}wV486PW(-%N$F{(V{zZa}IvVA@ zXw351wLQCU$EjcVbEnz3MIS|&ss5w17}})wc2?KVq157W5nnv0hvJU)^pcr77jrX0 zD;!a8DW))nWZk+}sCG-W=rU_Up(pv)vM@Rb@nX#hksfL?h|<(ai3e#O;c9vTprb=# zHs5x%W_Zjz!WI(`Ce&bb5u)I^^1)yZgkQ+U3AyGjlXdiI3FE#^{(f)pRQ3FxcX#c> z>i~k3w4c{(!d!}|%*@)&t@!|%;vnOmQ+QZ?DXiM!Ll+I)svXniO=;2v4+Hk4@LTX` zHEih4qmwGVvDJDMSJ?RCq1i#aPuwEUMISRD{g9)%4mKz7;P5BQ3kZA^_8xZlK^lu0 zzg!&YvRtXh+!$sh&E1YNm>D{2df{*H!=sOKzO0>CG6pUo&~Kt?EIPnIwqxU^rW3b$ zUZ>3=-isNj^hBCK;*I-D`qyV#ZOBz0D%DX4#NN7C;SN+2`qdQD*o5Q}>k3Y1TM~9- zeFC0QUnW;AZLu07y#oEWe3)@TReDQ_l6$Be?<;O*W|lOnU`SDi6`_RpB{ne&%(5w&dLP(EKs<03~!Nyx+FKKX>XXS&-FXW#|gd0 z{7ADa0SKL8Q4Mv4(^(e>2hw0GfCZ{Ns_O(jUS-M*7CE)qdz` z3t=2lQ)&J+)p4)FHM7ZwxRsX2XP5^rlRm*V2qU<_Dcf6ITsIumu$n+35L{(}TSWUu zAWwk+qj$!tN+VhLM6R}n1}$(apAo@fEXAXkLA>(lxKG^G^_u7B1M9ca*psj5Z>|hT z?3@iOBwn#N?!RyW>^LJOcT(+wAH*v$t-e>S_fd2dF63I)Ec-#)6K#f=WEnU?ywWkW zy%>K+s@!&2ZYaj=z1;hYu)+c5eRE!W;5z9v0WjZ?_Zg@a-u+UOEhZiE5Hq( zOaO*uwAA}?$LWs}md0_LS||~r_sR8b??mDdv?eU>vsKgQmT%VvNcGojs)3nf0}YY6 z+%@K9{>2>qe#-NyU3u%XZT7%r`JqnU>0LBqhHJFO*Ezh%K^xji0n+vDgDYKX&(G)p zSN}Nd0!~->9;u<09J@l*-mSxC*?61MOs5(!Yji z7cfX+2TNwpPK?~HCxu7TNH<64n3`E%BlCJF(SsI5q!O|OR+jUD5-0gYY38|bu&vW- zu4=Vth~W4pCc{DRan%#HW0OWq9gWx z`@QDX?t_O`E|X!Z^Sw`=3;CF51_z*}cyRX&YE@Cgzgj zoVZ-T%z3B+_!ATL{i+({oY)6Om(pp5jg(FN!Y|~JUP#q+8p`(DS1I%A9+=4`Nl6^e z{*lF^z;XUZYzLWNLUv6Wpv*~P5X2*HF#`>0gVdf0ZgrLEl~OpfFY{{%MA5hxdZ!uhcXze|SEw@T0H z+e*3m!Wg7zWu+$;kzCKQpXnZ40j~c?OYpwk(_R@$Fw~iXij=DRH9n4*;37shIW&@r zPX+)b6K)vJlR#7@i5)9`WlN)M(RT{mm|mpEi4v$e&jp6FXUiC?Y~OfWb>jjNXzI;h zaYy9*4q{Rxhz-=jmU{As=M~RsO7iI0P8p*btIhKqS$)L5^A`X+bi6KzX2(1n=gj&* zbRoH0n>O&-9Ea%{sbm_IaqiRD_F(q0eUo~pgg?t`IyVD87prGW0jNyb>kC7$L0=Ko zjB!p{!}fC;YaAhucr)%>be!AwEE%BolYyvWmkb-=ceb>gd$V3Sa9!^i5pMal7L&(k z))^H8saMPDLZ+4x$kT9lOo-pC%;3BbP8|riHbz#tH^@>{x z{EB?123v0ZiO2oj*jTLzkbc~1Myq#>0n;W|UG{UN22d-?60-J+>i}&`gTmm8cqnygvbL8ju?H;QF zXxvKVn@g@eJQh1$yKl4ax&l-u0J@b%u8L+1qRd7+M-Z$9n{*skJu#q{gT8N>!G=Qug4`_~S^&mH?v z5)gW~xgR|ApE!&E`t1JYEBx~hEvW=-`c~kfXZinrO7<^%_4O>^5?8u+=5O%g|G&Q* zU~c~3FaLj^{Lh{Iw-5XN{`>#=wtp?vc89a+fK&6=uSGoQe>sr9vY3ya9^l4azZnnu zFGrZ~w)-UQR$FNM?BDNI07Q1EC7w|?kQhL62J-j~qk`5Tj<3c|y z-1@a7MSuxSURgV&DM^j@&pM3ddP(|ey6`|k@3U&yqu5=0eEp-!(tl^R|3wU(m^ijS z#34gGdj-1l>}YzFbJ5REnd7Oe4?gRaw5$%J<%U;X{`jAw)qi?TULLqzh?P71N_(vj zeuPBEKAQ-rI`Y$?;PsWTaMV-)AhY}=z`G}Yw~v45S00j(hu&dXw_$7G{@QpiWs9+P z1=^1R3#-&Y&p9f>SSEsx6)q! z^QfzOG2H^DTc-@Vj=Zz1q!GZzvNa4#5tEWf;akCf@i71WX?*897}){HYHXDJ2xY*N zQ2%<9Taly0JWt_z!$zM&@y`*4Syw+WgjzGCGBVCp+6*TL8 zMY5eKd+_yTZcR_X@AAMzqr=3X3unbdIhMd``(p@&L%X6=Br1FBh&c4%jfh`cz}LI~ z5KRQ5o;hSPG3RS!svCCI<>sDL){PZt`R55$En%W?#Bb7Dv`U06e~C2xMeXGGAid5; z&Hp_=_>b>;<$P^_+kGzvvL5f10-K4S8~K;2i7yR+33P}@s=|MJxBklKU)wMP8k-@L zmM=e@yng9RghgCZCGAbfd!P9~S~{T?U_v~$xH2Z}uLb@8)9SF?c|5V(;5jG$KfU>v zuc^lDD{ySw%?Rgr=rn36jd}iwwA~eJHw?YKEZhB8zmX7t&AJi<=k`T)^Y)f~f`jq~U?x5oa!!Hg>^*LJM~Q zU@I?o9yBJq;qoDE3By)K($Q9g+5iAZ34cZv@Efu+1d4p5US@dzXLFS)P{WV*syhJB zujFB=-L8JB-%?(%Fq6ypuRhK&anA4e8#Yex%8kVEE9F1<`XWcq<@P=$W%NAHS;_mg zRqb)|8EbQE}8VvwOp=V*isT%(;GsCh1) z02{SlRJ=p@WVr^+VJZNa{IZWo6Tc1W0V)S1v=X&X+*R6or~mdzr@kUcGs@BEO^<>> zX2@oe^cjq6N?HJA(NX^9wtSXB-h%{T-SjTY8zMF-=qu&6z?L`wE3BwiCEP--2m9Y$ z6o@3YU1NGLQq)chY`#W@l_98G&k&xTj*5`mn{RTsyg7d@s1BP7nXq6jJgvq_PVGQp zUkP+*_WRANL=2;%o<`JO)w*pYY)dI!hh!J()d#;dd!PtvJ@}9HLO_vrHuxZH zRGF`Zw^Pt^`Cz_7_s3&T3!igh7;8}~w3X5dhwh8KiosR0-uNb+`E=e9sqCQBBX5`w zUwf#EF|`6no1|)Er>1#8#N0xuFH`-c%o5mpQm^IEO7HwKNxX%aadU606)Dr%jbdP6 z^E7y`3z+3P9|iL4`Hh#{eJ8K})pV>-H~Ojp2-wkgT)$QJ}`?Aep7^jKd!MneLP|D<7Khs(UI_t8k&*|abn zb2Spn{ab8#WK?H=6y@VDVg+A9nTBho0#Zt`6`h0T&dh9Q_S~D@^8R>!$`6K3uV2~) zMt`qP>w=17=2PP=GaLi~WPD!CVGwB|*C4r%kxa*`3*3z+t<_nx z%Ou5eXW^kU!CeZWz_evXamA>xvAjPvU8TN}0t`ZQ0YIc2QIhs~IUT=LHZ$gxek3%d zv2_HhJ)fdtAJdTn#!ms-Ns8o`hbU78jKp5Xd(=lN9Zq!tq?94PK=XBmoH20?g*|D@9u;c$87PH^uiw?0c}n^*rQI& zY{3^`poP>0&%85D*~P&sLKz|Mf_Wo!0+7)SNMVvF{nsA=ZGcex%hMunO)JlpXhU|T z>K88Z60Y1^s7I>9i;I>P-Vp9*aTPu+{We!)e{)bsSS9Xx#JK3-S$ficIzzx*vqeve6d|{2`=}>*A z(#>xd0^R+6ohMG1${Df)iikR144{f8)C$)EEDjds&MM*V-7uEUQnBS_MT_BAd{W`4 z29U_eEwvOx!9%ocgKT69_YVRNnaJ@*cxVAsOBDSYI)ShdGf=VK@d)}6`4&vsnLZpp z#a)J3dLg#5BvyEX`A5XSXe8B_@ImAQ|Etoh6rmdGOv)pCn5h?1IXBISqBO^q1$eE=Xb4U)1faw;hWi_t%*DdNAkq8x$D#H*O)`w?B#a5ZA|7 z$ci&FB?SRZbwJbQ*;)?d81?9A!iYgDwWAWxUPC74DclJ^#1ZTO;JVCl!dG;XPdg@d z)mWm#DfX3a{Kkcm#gb3J?aWj_)58swzyP-@J@G(A^^fU&jfvX>HPWlg^74nkPjpWY zO(P0hMmAT*$I6|r`k0qHL6q`r+w{I}E?&1w3SZ81M9@(dSM(N4xy}oOkkpj6lWZRJ zGE)VnnACPm6Z3w|vnC0b$P)eDebqzZnSvJ+te#>?srOyO*>36riV@f#&{jt;KIIMm z@G!m<^3^14@~6CFWVvvZVd^jcz@*+faz8MlXTR~T_S}_}hQvdAY4_uof%)9S`#LjZ z0n{(ku{<$7s1t(+R55S`*W^Mba8D%0#6lT;_ECHL^JFSdtsLx05U1_OOblajA(~g8Ook< z5Qd4cR~;f3RO$DecrM25C)BRoEqYsPp}LhjCm$- zW%nP#3#!jQFW>ozx%_3e|BtFjb)89}VrrKIOn2C)7h!7HS4V|ICIwoQ_~b1bl>>Tq;6q!(P#8?jyI}B1$jE@zXnoFWRm^AEc~ zLk6*)njeo$*SRT_tjY-*-2Skxkh$StoPXzk%C!IEYeF@u^$RH_#52rZEXtJ695!Ma z!F!%9w%;xBA!M_Z*~acjR_>B|$(a-&&po~fqlHZb0K6o9oW=MO18n5-fmprAh}2C{?9* zlwL#cAtGC`AV}{;kPe~OKq8{_-a~-UTL_`IK)4IfxA!^c`|dec_wV};9`b~@tTor1 zV~#N;VSTCrJNm}3(6Ya1=MF>}9wou|$xslY- zvR-tZRKL2bBNm!8}7 z>$#}sD-;qv$xQZiHxI~oPzM~4S6imXg)H2$c_0nqjd;~cMV~N2vmM&a_Zwa8pK&MT zQiDT{(`yxo!{=4NQq%$jHbSW`REd1Tcqr@rNY>@tKgAGtn4(wgvJ`W{8nwcvDaGFN zKi`ljRUC&o36w(Xz37A5C(#N>+Dl<1S0FN9u22ZO^*r?(<;ocL!bsjEhj#YL`k)K` zo3QD2A;&vjAa74ojGw=cW$1YGk&)k9FY5)uXbVhtCIIzz&Oxr^bW01tT%v<5YFB=8 zhYXl7OZaT$i3taUD6Ci`+f1ojVqiQgjn}pJM>m5wI>l?#B7F&e!5fWoGFQE@sikWx z?C|*uV=PA{nmipVS>=PsmNrD`RcIlXNONAe{a`tpCcvfY@PcTjqSm9$)3Rt?+lDE} z#+)9X5p5rCKdkvH#t(miC7(QfuUry`2zi$uL`qswT6%#7K!y(4$yRp$<6fuR`Xa$^ z>>mn-&7lWgUx>yPK!$M4l3I{gulSa;qk4N=K-&9-&~(LRnbo5wFYVkey*FJ@nvBSD zcjPui!68LJeZW?xn`WnKzg?f`QwUVc=5SsuqYsf>1ui_t^VXw>)qd#pwg{KMU#FAW z7nfP`&hm=zqoyVdmMYyo|4~Iry}lhin|uU&4t{*G|N15I_5h96*Eq-=4rkSu5uCNS zj;E0DT3ZTlx=;ua^eq19v#y@4n;v*TT}FLFaUc@90V=O;m;lwinMTKg zJMUk9o8boKc9Vw2Z{jcja~0umM|8}C3JwZty3(4%$IKq7JF0(P>?=yQ9jU^!iSIch ztF6J*+pgl=A>aA7Z2gy7;k6g)K9;Z_M?~P%#ey`{#rsg_i0SWfK@#M0;QP^ilVai3 z(UUrWP7Y%sfD`+sq<|dkfQ@?~ARxV;cz4X8Q{5Ok{VF5t5fcQG=| zfKPo&R3r4x8>!WRVGj~vEdBYPorQl(3jQmk#!PqAQE^GnclDBiXXCN_W0Jf1f>r~! zhq+>fYzyrW3`O@!vgC(16TG;%R6Bc+4adl|Ro2@mv``@CbNTQ5eo^ zIFbk*s)y*_a!;TZ%B+kKSl~j74NGW@8v2}giTCLNB7F`dzUL-Y$*Nwe7og*l7x%?1 z)cG}+^wiPlXvtw^v2`_FJV(rdZVdMs2s8ExNslz*Tuajj4)^^9Ng@gkvhBx+4;k3`{yeB3lgIbc%G=clZJ8D5u$#Z=!k z+65tB=HBaqR;@?meMg{l0ShZcHG{Hv5E!k5*qqeJ(M3yYDK4SR+nEeux||nW``5Fx zJjY9$kCr+)s1N7cXil>9RcjgJ7~b+?kE% zE*l7dqDx2+&xTSc&JGBfU-XP_f2BB_1Bya235R(-rcUA0VH~7=RoItVV&d`YYkTJ3 zDKQU%?x*G>T~F?jMo%b_<~dbO?bNTM>t7dL_(KVCK=%T;P4bbvYOhV!yANb04G?(% zS=6niqVZ;#29bevfNj}-ol)rKEWb{Y*s{Y;jB&a4^U?x zImV&7W7d-7Nmh))cEw-_^i7Nk=c{=l0F4XD$5CtYRw}@?zuzCcqj*It zD0Sb0G&Gx)a!h8qKGC~tWPjA}2r}#3W~=bXY(KmE2r0o!(O4bXsP2A1hc`UEpKcY- zK~8zqT_`^UpqbeN$AIx(a#5gqhdwu1dUg=>T^`=gg0ECe(N4p{%a9OW_a>?E?UTy0HcK#AOpC4 z9>oOGi1@`UA2}~>ovOR6sR!m=OF6C@sY|Q8mA9-CY0IUpWd&bHl7_wV4mec)CJ=~d zPi1-1c6pFgDb8nqnCv(P{IZ)#qq+BrWgcxNpXAARwH=@^}z#h9eDr*Dfws zr+QGcK_hhwuV++Ue3ECTHFizN`pSu=tdH)CB8D(ANy!oTNU>$;&m`A8h)HsisA<+E z^$+g(i%q~i*8x1=QjgW?sP>4i<$5Nk0AktYMWsKz02C;uYVN;rH$U28nq##TGqChy zvhnQ4S2ACm1C7vOBkH|+y{)|oUSmg9UC12aoY-KwD`%IVyzYTG^^#rUD5v_7O%v$d z9qU=K&oi6r!njjPJOlGtulqCg#a3bA;MZB7swbWb}fA&_YOC(0fa=bqK@fq66 z$4oU_Q+ZZvH)I2gRU}C*5KqT!jo`|+<%sTznKvA%?YoP}Pcq-*wSQww_y5$0pX_|S z=N5fWzqt5BebqeCXVCK?x0`a(U(ycGNhvvC!tYnQek{dU^=Fwz>N~!x1{noszQopY zJ>5d~*A+jWPd>G+!faX)?{bi<;=9Z8*w+qtgOXZ$075)y9^g_<037Ol zk}%qF$_q-&f}xi;uxa@uI-Y<$P2k{U`)z9(u;qQvw}*BuXY~!we+ga4&O>ipgCE^f z2b1smV%&%Foc5UNk$z+mCx+zC}tx%Ymjurj@^QlO!csE~~?i=DazTb`iFQuud4_Y&nWtq07O<@Ij>B=8HqbQ3TZG&8s?jbx=%X+i4 zyXyiNOcN_P+7GmkuQjbS*H>cfy!16oc0RDHdLP;a_6Q4=)`)kWgR%O(Yn3ke!Bxbo zjrZZ4Fcz!V^%o|qS5}%~cO?dET4M!QWwws#vCIR!dW&OJW_EWrC9n|bwXvq2nQ}Gp zEdBx|yi5H$WaT@ZWK@pi<^sm}#1v;n;xmOg_oHzTSM^po(ZmA8E3(j~4$TIACG}Zo z71n4{}9>wVfrDKf#4UzB7v5c8IiTlUIGkVHpAQHJ~G)B`E38tlm}K2 zCt?zND@*hz?qS3A(a{)zr_T#Xc5~#67s#x#^z+V|8K_9Fz_kDrU~8u}t%|uVo}%Fg zGQc)RiVdEA{YzOHOoLJ1?g2}iy=uZZM-6nivUi8}vQazIY5@!bR~UOF^qV<6IE=q& z{A4_OJ>eqpx?Zy=Vc^k{|G<+E3PIhu99QIR=+#SiX|uf1b6GEG;nVbxI@mY$hnkPX zQ`1oA9|dNsv8U>X+Gze*t8kjxS> z9PEF1(p`u}%BB-(>1nCXfjw*C5wQL|Ln7N{n~ykb+ck1^ricZU zbL)DStD^w(0gO%N`nn$|>-wDgch;PA-vQN>Ve@=@)+c8G!YU!AMH43rq2#3AddXkoc-Uij=Sd%=m#gO7MM- z#nK~JBs{5pt7iQt_vh0bU=i@gN74P2+x{9L_CMvXh%O)ix1rzxOuup8|GcRNXTYF{ zR&JsG^|I$*9-(g>H=v!G3vxb;f@ z(c9O~Za1#8uot{^cJqIQ^Pp|9pBX_m3+3##uMWjW^P9KDZMWv+IN?10^N3U5a&GBS zk0Hsw`!3%)P*AoT$3neY~%`*>}eLjh1y^tB*TAnDh*>*&xgXVSPJ32O#xkA%|d~?@j!}GI! zDgHf~hK4%Tq(%7h(p}cyUQoWVT+*Z4N}?1+GcCcu$wi?wM?2LqRo?Y{cbW?Namw>D z2mR5Z3Ma>v9(EB>TIxxiulxM~Xn*p=8F!yh<)*c0enJALS_X4?i(K}q>`8D_YrzH*g27&&sUz~F4a4EK2lkwWe6ZQ}UMN5nmfefI2~u82B6 z5`to`lQxNwr`$%^K`9ZgC*m|>PXhF(2EbZm`5rI(bO%n)_ z;(^-@Fq3Gnd=ki1RY(x`xT&%@6mt9d7Ek5?!-41Q%M{2v%Ci4_>lIv(8)8^4uUGl0V%p`_By^R!Mqg^E5T!xEqG;Q9|@R~OE8UkYSPzO5Jfv*I> z8fmPr_G~Pk>wb8c82^kwCZ&ET=;4FbP|l5{_{T}XM|i}fI5cx^>5ICZ=Fj(;obBY> z^@lt-jD>xmszz@#ajtxy2*Msa<=sD5#V<=Ex}Rn!HAht ze%V$n^sq@S#FTPCuw0LSjE#?C#xpmg%E2 z!XLg2m?`jP6?ESM+txKt88{dN^R|bQ`zs2JM&iRJH{n2EeE%xu)AL%JPX$yT-4+L) z7RDSh=VrJoX}+j8*Hw*dF)V)~P~-uGU_(9vw_!a|cU&ucsy$cCChF{7BB_LF8N{_@ zgRXKKZW4Jn!+E$xoF?tddZVHKr-CtQ@YXjY3$4PbrPNlgl#_x+#gCkp&SSN<iJcz&NQ3z z$6B8$zu3w=0)T35z_#MjsT!bK6B{Y@-7e_K=}vOuX3#>+@O+@Q=})E@i>dJ*+FC9I zPVFW+(UO*YiN`o&s(qUKa=Rwi`=>DG3BgCG&kdO2`|`>#1X_s5b#wp&%R0W#rG5Jo zFm8=KJuRwb)aHSz@-Tggf&63=Gu!xY$hyVAPXb9gh^~_ng`43a#~>N! ztVnkzSerVCKE?QS#aK(Gw)f^EFP6142se=9W0fk6LDr>e-C706fgzjhKbpFR$rHNVVyxFs|IzuCF|v!a`_cQwNOpY%7sem~QE+>#-DCNns^bP;0$yj#c6q*C*rwbp1I zWx7gYa7ulkKjuk4^fcxL)_WM9@9`)n*_kk8urk?pA%A#-pt*1BssZ#S*yxXab0AZZ znCOGUUlau}@z)f=&2h$TiVIQCAQ8v9s9C5I@NsydA~`8h+B(1{QsOw|6h~f z2Me2%rJw%EXL1EyGM%$WGo zZ|q7U(Mvs-@9^X}TdqwR6OSI_R2pg@a0pFtxh1tL&?J7ll2wNk5ZESUJ2K8x!~`#6 zs!mDK-TBY;$iEc=IiR#D08X2g?gflr^!93w9p3OpQ_H%&`jFllbvECCQ{a)P(tMBW zu*wbDhO+x_22M@LFp|_GZqxV1ca?a#MT732Ub%fu<_V`t?;j_SC_PN!^HR>IyH>If zY3%yBW|S+mCtNoYYfMZwY1iEPEWcJqIL{%dPKmCgx+fb$Fh87FZ&X1Y^lUmjBe+o6 z;;SBxGZ8)gZP+$zdb|Wp%PWLRtWYkzL_hEnZ-NObqQ-!~N%dz9ldPF!HFaY8BU4^Q zBG~uLwEgZ+_{`g8F~O2*y(`0;t+6}^_j#x323F36LqBQL=qr7aR1f;IYPRKEL_U(x$yPGSlECNL&whwnWRk^HsD^z}kHg9Dz2k1mw5 z8&KJY(}gM8QWvX*jWpSxeL2%=BLO9Q$I|b?p=jvoyRjg`Zi}s`D3-F zTeMP-$llwNOmu6xzGJBh;UOTlZ`qm65P6Jg+W3=(*#-O4dzHBuUFbm zy}ytO82l7iOQY7#6gk^H(4#N~e!>*CaVKo?=V|hJK1reVtLP_O>ppjP@CB1f-?i)p zTg!|BGG18yM_a8Q)H=Ac@}m_A`}yzbn2FX`G`#-dru>ItQW`ZMeEGFGrBFv%_}ZS; zCvWnU@M5K)b$@e1=fq@%d4fIOS$1DcKlRxtR#5!o_|mtncdEmk_*y+~oYi$=u@$lJ z<5j80?=NwVgKhs6u5aYjaB)t`J_7{WEwoR@709n8Zbk>?bLz>l@9*deoiTjKGCDRD z`HzDoev6pK8260@ZS6ID-Z?IQ&yp;4me@wb9--s&K>-$PdB>#rVrP5C9b=omg^R5* zk6ZnjP7L^ijCcEO@C{QyI!un%-PbB<@_FO489&(^t2tO=dE@oT3yx(HaB@{_>vBsp ze{$l|`Tdrm0*S??fIC^X%JGu6$d0eyft#a$HuA=p%_T&e#m{jVXHqD4SVJ_pV;imL zY$T!N>u!de`&tBQaVdxKFAF0@{rZVK`Z|pX+}X%0Hq*u1o6f7#xrMNL<<-8A#Ub3* z7gve)cAEsYno?@$@YoHxsjt3%3DeuFQ{jNj={f%mnQy&mpuJtsP_|voLJS7pcG1U% z$C26{Y>F|gkYad$UjA7B&_EMDOhUH0zFiTdLRVNBWn z%iZ<<)18)CVeBkk8#&X&v~Op9UUG}prO?Iz=AN98-H8HdMq`Lz(PUfneLm9_voW-! z=h=PylRTN~(qd4VX+?19KWAbUXRT0_Qs;=I-4Tic3J)wSK@ui=;$j0EFi{8w3WJ?R z))tRFtfNzd5C-X%7oTnQ<^G0eZk)JF6M6HyTfF4f z?3J>5ICI;%wiSC7>=Y&HHC8uy_}UrCI}hJ zc6@ajR^|`g+}Sb~v87A|Xf?NWe{Lu)lIKEeTCU8Zex*y1j3~&{Xa-sj_h!BG*?`;~ zwL^k4W%2+!AmpLHJYJpVb03v%jAaWN76SBkq1%uBGMK#9o@Q#MD@!NVaPSRzm`l6Z z1d18W+<6kr3fNcH!N|pszI~Em`NpRes_53*Uh0oO7-B`y}=zQ@Ti@h{Zzw6pCovsms@K`**vj!gtU=?)fPrYqZjB}R2R@mugzR+gu(H~EEVJE#aRx4++9FjWLw8Nx- z+G&P6Y*9igGM9CBH_;kyE>fQc{phi>ywjD_XdS{YCzLKa>s<5g$4ZF^?+3UpL%zY` zbU`F+H7##=>GR9k%k73(w@1iLWE8ryuL<-44%ev?$W2|z+Q#gvqDur3a{u%8MaW=* z8@^Nb%_-8;HK<6sMbcUl6qWD0u9J>IVvC7foEiw%n1M{Hcz6e@gurG{<(}YzItgfj zAyCB)VxkWdcYO;k4D}H5B;#2{@-C6U`mFGQ15SOhYJUI;TSHa#ENO2q&pchRB>E)N zs%}LTn#woNOuMy6jgsD_8=iwj)u1h;b~>MB7*Ry?8^^&z?YXU=Y>9D?gr#xoU^g!D z=n2f=Z)`hkeYFR}NN&*Z5v43LcEw*PeqEk-{qbR#JFj~12@_mkiAYk`_k7Q=6Ip^6 zdX(#_g6zeUw|ZfhYcd1oE1Y)THI?AkYR>->m~MA~8XC{-wnT|^FS0wODXzGFxcGe0 zWqxf=2)n&vPrp$K{pdR-48)9_2 z6M+j`-;nxsr|50E{rpcy=OjIKqbB-T`XAtCpb^91y`U=)>-3kQI6-xQ$kuiBc!xw&EJ*|TFg!>GNW#912gJiUS{ zeHgq~-fpx)BjBp=+#BDS&aot!X$ddb^HkvCH&2l#xaNFq;Acb}ip}Sn#5Y zn&Pbvc@ue25$)*Ry4=-743w{ab8)pqQF!kZ(^l$a1JmX2O!^Zb^?FZ3mb4}Zp3JqF z-UALP&5?2-yvqWM=^7ntaFloOHwW8y+O6(LWdpLIQ zs7b?DlyFNA!S0NSEdO)pp!XiGT)^@fQad|a>Pc_Wk2QLy798~@2^aE7u$<`g1(BuD zx-GG;^w7!@Yxzts;_?v;M8s3kzvjJm{)CG z>&hS_6zjhpM~Js{4v_PUj74|Ldi%A7vM6wGMdpVBXZ_#tv{zDeLFY6<&Zh zfLt4^9^*2Ews=xam1_OQrL{DdMN2HVf0PvOgFdDMAOHT?V+9i_Av3c)Q6D)};W&u@ zF6@M+nsp;>@kihFsI+4+>%^D$ryIV($mU9JUOq!M_*d{Y)45*Kb_TzbBmCh9UfmfxpbxP)DGXYgFT0$s( z=b}Vnn*(00{i7;aQ$N0!-~)nx4TBrfq3pjwYnSV=moCfi)W@~yF0iyKo9bThB00LV zL=hRmK8%qqijuM^j6(Nh1TRS0f6d%bql}1zLu;{ucI}6Z zjXrx%wwnXviTx^ZK4wE8YJ2|Zw5v^K3aOi1z zXP9NtE2vp&ae&Qa-}cZ)nK!eR(=uyyEcnu?MypH8Xq1POsWbpl|o;-LG(*{*(y6 zd=o3QcdeWCV_2l2cN^W4{piLmlxM4Gtf2SGR-!3(TAj~j{Lw+qkl?x_T?6urUAJpQ zw|O3aT8IN)A(G=7x-FIi-ImA^TDNx-9%F2=hn`M&X%5R6mz5cm3c75JC_2)$FJEL#^#-autvZd_~=280oSM^RaO)Ma3M~*QGBAt>9a6mcjb19o`v`t832*B6d+tm^p14#h z(-;#Cl4eQP>~U9Q4t+}`kD;Ak-R8ouwc7<7Y~MaI2IJ5(+sTTY^e^<PU_Q*6&>nwlZFMv6iBupTJ#ROAxX5ucaczN%!G8B~P>{_hdq9j7x4oyk_4UZURjEU| zg%N49cE{`0w|JMc@ybyGtwoI>0d|jo;pU!>Y?`JmG2cL;cj&6epYfr$6 zLaifaQHE<_Y)si~@nWUncW$6VMN6#j#tzISU8@~Fwpg4z>_u$I<`rIAQ_VN>p%=}_ zgYgFxEx!$(u<0l(x}raU84t}gAOW(>3=2Sz+UyoCi@Mp|XqViJs={Wy_Lf>>oa;b@ zBP9m{?4-U2pdEj#fR52U#sFN2)3NUQ2|K6OuslqUzYkVLen?gET#OMBnDA){@8iAk zrG!OS(G-e3I#y*De4$C3NggszqzLH}<3NhP2yJp|7a!}J+g?Q@hQj;J1jKgIB}8Kn zt<;6w@DdV}Z8eh`_lL5mrr~MDELK~mue}=JQr6g4h7>QyrJTJ{AA`=+l(o3!wMn3D zIJECW>f6v!Q>jL*SFP=ohqn3nrLZ&x&4V_$8u_bPn~+fD(z} zG_kI)ONWoq4yMqmlsfrmf)0m}LMe|5zwab~DT=b2!nrxu=X-)5tEM$nuH#cNy3lu2 zQr5RS0c{|zb!sGpMQL>KJbBp^~um@H6|+? zyz5G7F0JkenV!{)T=|gUGP8UAl5x)lOcbD|d67Hr{PZ zLt%JOkWjDc;rAM4NE1RtLB>kuV+7t&M=Y`(Jvp@XT-Yt;iJ3;O0Vg2OIrLjSSvlNW zMO)hIQ(n2dwR)?p)!WIYJ`m!DJ3fxM&EGjT8e5=ElJs0CZ;3@Wgu-<20R6uK^MvOV zQ9Oi%#EQa7b@P1Z^H2+W$VTl8lZ`%^Hthub3`4f{a1zS*$)X zPx4Skb4tB(*oJKtT!rOv$anFjzK5Fr*+Z(ki%I=8d+MB?g6zg1(wiWQsmQ# zq4{BCp5aLy5r-A$_E?RWs^u&;Yc%q6bmyr_RR24w1pNJrpTj}!&}2OAOSvB+?!o6V z|4!};uHSD+9d3&oYpbsdAaBYjpsFgS*82@Y@=)7-Jov-uwph1PDDQD1&WrrURPoKq z8=8dDz3u(2mN?Q0bx4FlBo9VCCevr8Z8(He^Ynh2u-*m^s@8U}K83{(G5ivBvMCAex+fefurGENaf`zKD%${RUlK%;nc@(bTbFC1SJr9}-(H`HIyP5YzK7njQjL8`Uap^y@h2_@gLKA&FA_T^9!lR0(qR?#1oTHY377bvm2FZMX^H0wjbo zOe1KVv&5X0s;X&GhKM+sxS2_G1Ng@yf}do1#!Vv)eWRM+Iz6(L(bc>8eR#I_B^HqU zOcOuh>qoUMFA6-uIp8X+BloS^9j}GjN*0H(xhc128BdS-4RC3as~3|nLu^V3(~<&7 z{e7C#LzdMM03>-WQXCG?*~oEYvRG0x&~1nlNuhDwgdMJJQn$W-!JR37hKCg~Y*erE zYo>ui-Urq4)|FylP1~tr()3BDpI>ChE>C7iMSLSle52upVfqa)=$`>EA4L0Bch-n`C7Z_}2tWs`v8J*Fh&MsvFvoCPL#67wmO?dr+} zS1aKgF7U*5&-InO6Ix8m6G0R%Ym2uSORJRt_Ps5|$mLS+Q3rI}I0(lPp!QivU2 ztH1inOfws94Q~#i>fj<}wp5@dPD5FT^GnTt-kkI-S|{|^I2|sezQ*+ zM1gRo#5D)u-kipmUL7eNfsA6SsrS#CP0pavRxF(hE>(@r*;8$@vSQ+^igUQ7O= zupn0-2KsvJnrX!}^Lgq;l{9brJku@u9C3x-KR7bgdEFx{ajtEP&?cjTq{S|?iQXo!by&vgkdeFH+55QDnzJBE)G6rs9Jj?cC)n(p^Xvz z;H?E`#8Vxk_00PLdNY-KM`GEH`bbm>kTqn4pDkSu&p%GhnmWZ_T_(Px!A!z%y#VGb zlnfau^uqPs;t;;w1&7A0N(g!I>^1+o>vt zZ{^ovHp~LC^!F8&1qXDC0yJ-FDhpf|*Dic}?3S#lH^|gVFnDUwp!e6Y*I#}mus-QK z!j9hh#=n@Z%ng0H4zPh#pJD$S7n~QhgrI|3l5?Hf0{30>!Q#BODirFl;+SFv$|G|L z`f9%tT8%OQF3_P4QyelcYwe7M;Bwf5W7 z{(Py#@H=R4b8_{n8On9TWoM0kwJvJAYrdNHOVeQ&lbryOHV(0mR&N#nYsPcu19+K5 z#g^R-J@3$u!sVwVU}7f$Pv)EpVPo0+S%uEk?PMr?^-VZFx-8zcK;+~538`tqZNN7C z*g94>fXdRyycN`!N7#UmeS0|9MNmb0r=Wv^AW~y?>HHv~ZUEY_#cd>1gsIVn*y*u2 zK3;;iZV>;ap1~i6%-)t~YH<+mKUr(OciEO6RqF`N4$+sJbs)579%J$yigLRS;lSLY zW;Z{yYs>?yogsdeh8ycv%|qKKP3v+z3Rh_9lB%s$qz&==L$}%mZMdAw10tR0x~OtC za~ncLLP3Svk;@>yb}Uq+`jr|IU8s90I+{8(jmqX?u(WxrC)srsubl)F`*7NyBp(@e z)QPpZaEjA(ehqVg?O?7HvM^sJHFO%Ucy>qdS$%)P?q$d>b_(A8>gcM%$N64w|I5?f zbI({P`{Yc;mCZeQ;5$B_0IVP0*lpG{{zJi(upeUts!P1@Bn z-RUYbcZQJWiElmfuNN2U_LIV#%}5PlvyyHOgu#yi^w?iIyI&snozXk`C(Gky;!6}J z1tjQ?YBgiZt*2Y`Z=rehuT3>YxZujhX0@shL< z$?Nyu?A^-M5pv3+OpD2cZFS_OjwwK!07Npd&LDs*bU79+2#Yf*Uk!X|`%QHj3`1!K zMfbOEfgYiWI)5g21mi>h`>&2Hr+QbA7UW1hhX_&{xxt{2zY&_=cAloA&vB~h!rZ#0 zVJHhMO1~|PJtbaL>RGDdHCr4Es~xZ=P&l!S`%Lw1UquQkT;jVHE9Bp>MXfr?=A@@= zRAjfh9f@mT{AdUow;@|@XE~dr`AnAZ6Rgqvsa`Q;1wld{5;nS87pbk2}>l*nXAO*kP}d7MnDyL#w&zya(`r z1k?T3?Vhlk>>+HCD43`Gd{5?3Z_E@4MA;h>eHE*h8Z%YXhsophLA-mYS@y$dvUrb5 z{b^z)!Nm;JJ**R}R4VaBXY@~l`Xut?B_7dXd7uYLeet@mD^G=YdkIE#7uwRAZjY>= zD6#CmDB`skhgC5#zb?jv+l#;J&?Ol#+GBZ}H2m;{g1>97$=ciJ$|dphhbhUv4GC0t zs_X%fChi4HuS(3fU3J|z=IS3ns2rLCw4F@b4WMrv&Exs2^VA-j_bFfE(Rl$0J4gfu zO1$!HNY+TCJmeF*E#JOe+V}4j9~AoSOj-u0{_B(cvn&4KxcCgmU(NG-XDx7PsphRQ zC3w@WF#3ebB`)vIGIS4YO@v)-O)}FNZVSkE${np7VZ0NWDj_!u%rWq5&%b5!<~2qL zegIY>jYXcI!NIqLC5pR@=||mm$V#`w@TE?6%FRC_=+M62u^gIx*cii7WR3*9Iug#4Vm6=1@?rBB5JI4|-C@Di`a2~;T^TsPlr~>8WWD&1I ziT^y-i2ND1cZhWlDM7-&zkJ+oqJera(H@$D!J!1O@9Hk1Q0 zYa+n0&(Zj}HocUoHQD#|YfCoIf}XnN3T5$Fd|kKVkjHBi0ax*}FnDCsRsa>TU;0FR z5zvzT5cFK$8yM=WT?@Y;e7sKcJ}keM9c0C!P=D_O%0%m69+b}>)Fbb^Qso~7E?|a) zL?;%ond0tC`GBJ#+h6e^yB~4P@8~b#(8tTZuHW*vhu*vY=f_7k&fL;YF5fc0M%=c9 zs4SVg-xN5cepo&ppu&=~*PAtS9Q{5BHn`8Ng!37V-d1r--7CE2rLn`^ldY;MFi{Ad zTYe1DE|YZcbHKTxRvG?~ux|6laU8@UbCH2pRKtDz(O6jXACJ2q6vc|Mh+gSRSF!es zh#w_BuZ2exOt5+JMOc@v2UYrvFZ3zzaSJ{f5q{*>kPk_{UypEHnPhiE+QJDS2ZJ$# zQ1_hwLlBeLQ`EeaRtTHQbm_amWs~*@bsybAMFHc=ixTF;l4Qlx5XteynU2r3UDa|> z5!c;v-E_>#>SWs-PyI%K1zj}5lTYef#ozT}c+maItQ*={>`?9n22Z4juYbRhup!ZU|IE&( zH)cis`$Am1F7Jxn(?4?QmwxN!8Sq(|TsEM^x-EnO00M)r+q6B|>9ROJUxN2F`86;0 zZ`}5RN@nG;-3Q?5&#hEmePpoMN7t+^22b^bI8>qgW`F5J)sCZ|{eAJxH99``Sz5a^ z8=6}_3z=^YLL|G|Ews(h<#Fn`zCfqx79N@eFODHn#~T1|8boj#vVijRc^L`RSOAb9 zHm)_Mh}U$vn%#1c=1b8M)Dcaorg?z;e(ssY{-MsF9tMacO1fABeR zv4uBMsn3Px)g~{o)arQ<7A+|hdC*kcwWCa?y;2_Y)VqJNx@+`DKMMdVLjW5MgMy`# zUcTl>czZZ0_?ag?TF(wC?vX`r)R#2dxD-+m;5#qxEf~n5ObF4IgxP`f#w2 zW-P+6Ln07Csd8Yvb+8KOdrRhdz36_|I1ye^kJ4xm(4P{~dhU6fOdr zRxAe;!M0oI-Md;{sS0?EEJ>@t(M7sr`#$XmuPWQ_i^FLWj6+5;e$u4%xhZPdKIS!A zQ5SSgM89T`Fhvn+3e~kK|K^n2j>c=zyc!h4+PrREw9TaQJ05??oApb>kBj(n( zE1?ng`gt=@KrsY7>~E-6f_G@YVep3SJ{$Epz5P(F!yUkOm$4&=Y7{4yjFT^|B=nCSvm-y^U*^sO5MG8oMB7cdS1)2Z?fd&A!ouZ?ZPiY zp__7$($LSi4X2`Wau#f94$Chdej#{)BbsE|bN8TUl+H8o4Rn7@YQEvbv*-IV#}4pa zGl@;^?zHxeBipk6epD@icWRpT#oRst+<>&$)@Ge%-FhCOJ(sS0g&pu_j$2Hq3BV@Q z>4oRLG7UGMKMuPUtV&Q84BeRs>I-M5lYLZ6RdcvH#aNu-))0O|8BiP|pMCbGciTtt z$Gocs%mO~+rpsdM`fmI)f@1sM3#1-{!~hWZ^cAbYY)DG9150ee~Zi{ z(l2H3|NEQ_t&Z}^E-Hfwc68{|K8Q#I^bs? z6&HH=zwhg>#{>RA-8}K*PcMLfi*En=A-iMXB|`m+TJHS4J9&BrAjl`&ZGL-F%g<*q zboyZBo;Hs*{wvP@$1DAR`$IBVLmuas27ZmL^S{5L{3>{f=&Nmf^>6=L)W5QH|NQU+ z8D`}p3B)0^;oALQI#&Mm!T;l8-okwEo^@v;aV`AT0E2(N_OAfJ9||`5dyk*_j2hSl zNnC1w@7*823*MgCImN%VhAED`0_r!Wu-CuujPj4iDEIQKliaIcqeT3*@&4TG(oev$ z_bYtz`wkR87YMM8cVB#T`@e3#|GnD&nQ0v70IT-NMUKDu+^&F2ze@f0qV6F|8Nl0z%`v2_@|FX$D zxtD8^AIHqZtNYl-H2 z4A99}rO2xsy_Ue}1@O&JW6I!YfD@~L@$1(CRW2NWG>^rLyt5;Lr)P6?cRFB&FFeb! zcDqvYB=}b`9%fKzdVRD!Us9YcMss;5pso3rZ`hIHKZp~ zRW8|oH1-N8TG|q&pnmyuQArE?_YOLzsk%W-!s20H9*4>U>0+21zCRK)Fo*N^;Hkc5 z6c7osfL%VTuq$`nZ_hrrWOnnLNeK0HnLnZr&sIBq`R)_I4rD*@^E{H!o*<+^_DVV^xnB;vHLUAx9NY_e#OZxMW;P8KlIQedESH%N6fwFG^@ zFDi+Wan<^^u8Tt=(P;b6iivk>siqBpG6wM|>h|Z!rf>xVk2#hKw0!vS^*NlDUo@Pr z1_gwm-3Y6-mF8y!_m1wH6%@wZb$IdEjIt%(vGd51Sba`6@Lp>O%T%3G=dS~-t2=dt z-cr~k-#DS@ibT(q$4H2|c!5N(F*!q5Sla@@o^El-MoGC9021)9 zMk)y;-34-F$F|TRC+M`haaGLbNlBn4{});&(H<)OGAOpMTd4A)a&*$F>U5Qe46(Bp`e3K? z*rh@Ek}sDoYRE1^%21Ng9$0ieJkXrycCQKnSq?p)P1M3*wQibnB2SDhI<$gk{8?l4 z4gUaP?cw&AudATPMRjUV<1U24FZuV!=Y73lFZiuj^~XI6O4H`B;C997xlB5?rtc0D zCVj=0uG@iPi)UI5Mhcu}jvwPs=ciZzMolrhcfHG7?nJlO16u-BR>e<%?}UABF}n5r zap*n9XB|7#X5KVPeM^?CPqh6an}^&CJ6y}|nF6l$mmGSWbsmZXclt8(w&J1Ei!_A^ z)~~krvrS)#qyrHmRqW>eJZG#3j#{*TLirQ3uNr(whJFBz4&!?I?O98LSif0gkTuWF zlZW|9tL3g&0xtdT!BDP$w85PzkUtSHsOWbHWuW5qQNg<}k7JVNHg!rugb3K?glaF- z*1?iwiuo~i+Ws8NB0V<1Y3T-njh?PF7wgz{@9ybVKu1x_G|Wp#SJFD^A9ST4J6jFG z#VcrIQLjJcd(0+2a9e{L#IF+ViA2vu5fP{90BrZ`$o}Pf?kJ{E%Accl@ZmXjYgtEOA23%4ZS+nMuBVM@<9f7qOr-Vx1#sy7r zY788^wPt&$rY)CO|lpcUO5R^)$geP$_~ zQ}d=tbAAP}9e#uZr7nl% zK$0&_i=@ZOm(~*o1@~UR9OKn5RYYCodjP25=ks}hc5X#ubvi+j2sj>F0m+GYW_WM@?k{*^tOk1RvKZEgTGIp(uxs>WAa!z2p!AMs5L;ajiGBW_%?+zR=h2#Sfa z|Il(%lA|1{^4~0}`LqYy14#-1O~}>W017C;-dTNv+yjakTqNgROer#{N2a=yJGlm_ z$D{Vt_xr5?R|R-swZ_-H1i9^<=-bSU8w0vMVLuWMnyW=T1$vFfZ~)0nNoVQ040F|k zx49_^(ef$2qdiuF|MvYAv^||9&ES0*^GIrZ;0XM?It5^qd;>%_7des>W&+se8h zdlRPUx$GeF^7V!=#`(y5%7Yvu$(nSIB1Ct9?XA*kgr|ojU`lg9R$Zk`HTU0_BRU|Z z-b|U0I)Y}of93%5R>mEC7J(^oV)>3her4Rs%x{x~8fw$!%c z`O>Xhr1}zfqEyYCfu8E+-Ig7o!qO?VqUNF=a7y9fF1bH6fYs0O42m|sCR0;9a`UJE zY{0|B{fg^JopGT1=!C`f#<(*_lHgL-0o|7o#<1s0hFPi z&U%rCa2Z}C$!hoMwSvgS=N$S$chg1ZQosT+%hZ zQ>G%b9O7lMKJ=v70L3)Z7df)? zffN`E6#Ms&5djGmM-;zoX0WpP07i;%<^d640QK1j>K!0ld}*EZ@7e-oG*i zGrz@`+GP0=QZgDF0IFMgWB-HY>k-IzRMTga)~JhPR|b3!2)x^#ZAr@NbP$s)X?+29a@}hs*7USoTKDUHd2>k_f;#dH3!euNj1HYDS`V z{9T(YkbJ~HP+rY-68Sn-x+y8vKy?XC=6&@C(EFuCy~%i-6lm`yDChKX4`8BWX%BQI zS^*|8Z7p6vXTj#+zGM*qDwf2bS(wpDyj@MG8f-K<=8wHq3PUuwd$$4F)>btK)wiGi zyi>UeFgeCHXXDy`r&0c z++LEi$F9gskx56T(}{~_jn_x6M^SQqt-E}2HihHLpTRDm33VDwm!=EJpJ8@t z9eZH~a(>t+dVa1qCpI_7;mxnfmLZ|pL^ZI*{f_r7wb3Ae0KQyYz@q5It2n28x)*)V z0@!eO18|L$vh0H~Rb}SuRyhIDLg(0iJ2`hr>nciD+~`;!ZUyKpZhHx!S3r?Ix#e$IX-gzFeUhAQO2GLt`m|a;IJJTFei{3Kd-Cr zSwwrE7y1SzsvW2QFi|iep_OGU`uqo|$Radqb5!akCDKu>%FGdDwiS`AH2rMz!yVah z(&EDwE14C`0JZh|Y;}#v3I<|IqTFGr3Vs#ISkvkf7TdISWHjrKQ^`dMaPlm_->j1$ zdQmT#(|6FW_tSfFs_#7+>u~=QB>wIHq5~T1%xGh@=Z$btRJmv-Q}RK4;Fs_ z68`{YEYh|$O}Hwy>OrX&NmZ#^>4lZZDK*y&e`+V#ZF|dy&}eJD4(Ku6n`{bxno#X? zGWuu~v)l4};+T?-zSqsS<3EY}h0Cm72T9vr=dJ&|Uu3s=eDABQ#x*bg%jdKrg`(ae z;@#Nk)u^0WvPz4NYtgSjI2#)31x@*V?Uj;!;^U+6m} z>%g?e;-xNc9i1Gt_4CrITn!6Io%$G8g@KTPc`2~xQjeonPQyi97J`A3baO_3wEaJB;)y7 zjtjg#=8d8217Lj7j)p$?Q_T_3!jhx+vO?=k9*#v_&=KJEST5s6_~mpKDkB|^#xWPS z=|7yKU0b3Ng$x#bM0=7krrn3<3q`R$&-n3^1&;!Kz@$`_-!gd(fk0N5av-T))`rlw zGme6KCJ3FNcsm&wqBZVX zraxksz2<>`qkqtm=%c;N-0g`MGH&a(eqmG1k+DF7`;Svj58b2iEf-T?TgwwWR;1zW ztCY@$Ag7wQeBxCU#VO{k@IWzd(kp>(;R1)N@s8IY!P^DH1#51MZ97VvS53hA^4rUG zZo6|U?}|eG{Pte+gp{dOi$36_0luSVwEt_@AD16MiM#&VedsM8?c4oCD?y}4fV}z* z*))%ZLawYD>gP(KD;_xC@;p7KzJV2uN0TcabQ1}BC+$c zd)Q;g71@_&o&yTEJF~Hk4_bcm+s)9ULCWQ}$hqfEP0 z=H_nf{Zu{*&$`#0<_@a_q}GznCDW}oJdO4zHm|F`9Yw|h{wMGXhcR)rt@S9C7=KlG z8PM3QZ4tNx>iNqbbw*tlw*tayuy#e@Mmfb6yq3+Yl{NJMioz^8%*SyFkhW7<@aNai z(D3IU+YBd~gK8L=)P)*qy)5Hu!*Dsm5n>zPiyCS9;$JQE^ygs;+L${H{wTofEuhcL z`hHKgh?obF5c9m7`sc(X@6AV3Go2ZyIze%j3Ds~xqt;k3_GMh9y1nZM+tXlYew8Q7 z6gx2wmhI-q=)$;L;Vyw&6aWAffDL>_&(Ev4I%|4Y3MgNvBV*5Z-Be%vRs2wnvx3LZ z*S1gib++E0Hb`Qwz+(P8aPQgABUD1l+(;!}RRT$jr8!L%-jcNt0Xv`Tr=*dUqVkBS3$*qotm7^= zveD5vSJ&khFM8=QA6!C($4-Z~S@hPuaD0|Qys1=!y%65&&>Y{Elv>oL>mClnd9;ct zwk0Z6I-*GuEaZ}3mctDF7U4|*)}wphlLb&3SuLt~03lPh&{N5?Qyjc&5Ae4TXgXGhAISV|8m5SDW*kRBCGVSJ}`t=xV4 zvvusyP8TSj_{O5tq}1vl9h5KT>ksG+_AJ~a(b9-EMKAB-0wudPdn$a*O+84y0euCv zn}@&PXF>Un>v$2x{vE0Lm81my-`96g;#mOK%lA=lJB9G5zAIlCxU1{#UhGH2&}OQ} zYbnwKRSTnwR5%%5J8Jb}v)f9Cn0VJ2l8Q^mx6?VspYLsVg=W)D3JVO>`fU^KOlj9E2W;x(< zpwAKGyoZc3EJYd_)WY77Ljw5CY}SW~!WhXcE&8%t!>Ao)tCTRj;P}fKe;1G~Sd86+ z)5X3xp_8cCsccZ!Q`PD}B-?4#9hmo6^De^4oN)qW`=q(Rq#m;D?=*$Jg^eyHcfF|> z@5Sfl2kbdEHhr7?*(;F2q!zV0Y-b)EBEM3+H|dx!5zm}y6QW{Q z!37`s_niUD(66A{oyB^Z*VMLG3qM={yj3HOR4HGM?Fk}tb+R5hPVEe6ant%}cxkxQ z{Z5n@cb8*-MsUu`5M}1>>$@Co(8lKvNwp@wY$!%;Fe008`D2Bs0~cg0*A_*;J?EfB z#9wXX_N6*z>}k_eW#n=P&g?*YSfS1Zb;>=y#yj_Gp8d-iO8|& zX*)LTS<|_%DnbC<<;5>I+ka47$_!U#Wxi-2GwEz{gZy<$eNFrAAUOzWah1Yz%A>KeP6hX*Rms_A#u*5Ny;dkgB*RL+%Xh9 zen9TD9?TmHqHekVyi>EZcSh-xc!0Aeb;)q=bO>wp1viGc|8i1;d*|zw^`0-Hbv>1b zwG-qdC+K%xU_`*%YEJ2!3o6$qP644x80cV>+%mvso0|N0XUStN_56 z#*KECuZ|#M`oO);n(D_%%4l3BC~sGx4m6$$QPdJ$9i?~vMea(i3EVwMA zacWmM)3H1!AgOw&;Kq_^{UVVNpv>)mq8=QN1)Wi>&DtDsZr?@L3G}U>$9f_S#Nq`z z=DugSpRn}65k}AKIJO@35=q#jA!{&8)^n#UzHaSHyY(Wa#$*2YyX0HMF&ige#KfkD zd!{75RluU$?P+58&Ar6(T?Npa1gPiU`vSv?HL{j)=3s(I}q`KK5AKao%!}e-Kmjc&Hb%M?pa6s&NNqX zq#*PIj#0Aig<5z1l!{7*J9kTg;LV4&XlB*z{s)W3{f8#t_u^1?`ZX4jBJukr1i22l9=Sp_W&3Vr;Zi+oJ4UOWqAir?LQccbEj{Q{g-f&T{UbaL` z6w{DRqOW@>Xr_Q;{r;XxV?~obtP_fwGaflW71&xh-Z@i1zRBF_9o;+dR7W)P0vsMw+h*l zumbDEs7eP|LVZ>$d&3`;Hiz$xXjCBK(sz7ZFvkjsAy?!ahOMwR+*VM_CHS!mk}@Qo z3*ez<){Fe5M;(#uq5Oa{Y|>LymBx#S$oNXGg}tX4-OmD2Gl5<3JYy`z{g9ou9OJRt z_9cPWDZ^WIbv3OHRgVIJ-loh zGUK)-fpj!H;R~%))cz<){L1_#u-)m@y5NV&i4K ztU@1khts>(ls}ULv{>t2cfjGZbvc&YeB0%!qwmrM9sAa|`~3_^9BIxR-R4yKq>9(x zMxHK(7WyRr@40_{LqLu{^j-pM&}Lavll2K91ZY3jELWX>N}VqyDEHnD7hb~O1HGg! zRP)>>R`H;yHM!d|` zz0Nd&Ey!sWe zE=2-MIWkse(wnV%9Q#+Ds>48_#(L*ezJI3Ws=%M?O*FlH?%>s|MdcnVAOv+4#M|1{D}Bv ztrx|d@)Cy%<$53k7)X8$%)gO+;5+0V^ckp3n|z#+NE z*P7ui!|)B-UX;bkHl+c2Xk}6y({847b_q)I{!CI&Pt@|oQIhJKu{69#;cUVK#Qe6% zkK@o!T6_(lSKAz6$aJb+5@k4PL3gL}Xfs`K`qzE1>Fn)&AoeM~D?542oewS%nKztK zej0u1;CIS9-|rVEUNLm6t!D1LUG>sRU&mKF)5c~rCVh9bUXUC0-QT%#(E}8QSgDb0 z27Ep#jCe$cs5+53ql~%capIw)%0Go+Ze^q*j zq(!!FPF#JlTr7pJnG@^PL(_JaYozw&$;ijuvt>+}38U+;HM7$~ zYIy1ZUnGj&!20A2T*R-#!V{jAXU}+CFw>bh(lb^w(l2EU^!A3^qdoJOzS^f$Cow|o zn~-aoYK%ve({yTrnit&ca_Ms?pX5j@W^egX@k)tzSgb?VpF4Badk%Uiuie=xd?`Cx zd3D+I){W!CkT}yl^K#2tsg6dT{oxe*vtjY#hFj%aUtV=@9<1}-{L5=AZn)*C8sojn zk9yccihgKB7J3$4Rgg^>!~(uuaD_qXJ8KeHg$qNXkKaGf;fz6F77dJHHaFXL(@xUk z3!jti(>Fo_Cj%}UqtShw$xvkavc#xOCF6=peYtCWoi8m6^4cic5kie{E}nXA7zs@+ z(oHQy|8Ork!RQNjSW$78)1QDeB0FX4sh;iz8%B-3kb;Pc%i?Jtjw!08dpR%gUzW$SIA!s-uGh;G3jrQIhtHr_6qVu*Wn6V#l?y;=HvAjc+&~4Pp%1Ld)y`enI&(`8oNQ z5%)-dL?cPUrig z+GOmbGQamb0(L=!b?FKQTiUn(qt2B0)6umMwZTExsYNHR(k3;ycVpm0vV%G5Wd5`2 zfOTrL{VzB;4?R8Sz^LXiqGs}_ldIk;f<}q`ZAK97J||bcAJnJy_JXV#zC6W|lZl$Xh~Bag%D_jv@4}f{6<_EUTqkO;OiB z)q}uy32I!O&Ge_9hH8H@{LqEA6*kSqowm?Jo}J>o4c$-_P~>3^*F2Xx54+%8K5NdG z(hqhNGs}7EBV>tNg3cH|B+BO``MY7dt7G!#+|AhunvkW%EGgp^HtJaqEae1ua!3GY z`GP>g{PS*m>bX4|-O~M*^0x~`W@kW%=&eWR&S4hr(zeX z86gdoUFA+<-<-MJ!&2}Yu0OCm5WYx`8l+?}Zy-q(^#tl$fZ-bFp5PPBue%I(S~BL>?K$sgkxx_*cR;ywJ1!Z~ZAi)41pPEK$d@Ti@?$?0 z;W-SQR~bv*hF?&V?0}CP^7!iT1GsXU)%ox%3dT%t3oy=k%23KVfi4a~PChv60+Xn;i1A>{oX66giVKXYnY1O;O~r)@Rj?1~tHYb~6CJ_p!F{pF3pF1hR4J z(v%stJK>=uK)fFvph|>@8bPet@A5Tmb+JzOW`-ooJyORsPJQldeg26P|4O}3TqXyB z)=dUBsLp zXk=uIH*;F#*#C-d%SPj+5~+|RzpXBF1@HT0ujJ0w%K5GZ z42d?1wk}u2-1Ej6NiI{ER{a&dj)H7!;_KgA`JUeGKfVuRUR{Fs(Fk_>Vdf7w$U49% z>2ryOc(3U0enz!3IGkfq6K@z9Z-L-8|I6IcruywRx>i_Z*+}2#%WO{ozoH)Oe19<{ zSlPJJjdb(fych>J?@r(aw*!rojcX5KT*LL62|L)Kc=uyB*G=_ZVEuLd5z7Oe`}0_# zqWGg2y8rk~UJN_L)p-t$&aub&+?HfW=A%6^NY15vU%#@}4Ox0UH&IU}bhW;RlxEfF z6`q%PAf`HztIrpQPwc7aWTE1(BhlCn43Z%?X?sT%$Bq4cu!5cg{!A9#8DSynXAjJN zrb4$j2^uG>H@Ne}>62;}U0(28cmI@Y24r0JN~BF#GCh~J0MG$w8NCiZ&le^H)W&#f z#f=3>8oUPW89{&wu(D%T!d*k5KS&zXI<`3^z4{Be_dQu5zm6+HH~i=5B<6&VF{Z^1Be3;0$$bp(Ofn}{b_GS~Nn#Vh z$L$XnzV#CcKB0EoBW3cWM{*YW64n(Md0+g?O=SDj%m)`q+uAK)m}f3)joH3K2fw9P zV@&8e<@Pw$xA!BFEcvsJV}fRUXLMY@{+f9!{h z-{=G%RN?D#oNXH3wGo8m^Wg>pewZDow>K%Ve6u1bZh|mDs~6RuDs>R`s1uV4J3S^9 zCK7Reh<5g)Fn3LeBQ-azK%A6+i0;RJMe2sH}UKj(59KsT2sXM(!buq}_0TFVCpJ3sFyxT7B<7O!}{`1~!Y zZL>2`uDs$dnt*LPn3W%BahOv-#7JqGb#f*T*+b!Vn!n?j{IYocBqIxV;y4HPL7nfG zSanLs0t|(?<&^!cU!mm8X-&l!9z=&hi?cht>pV5xolen8 zFN_fO%1ubQ;vk~YFf5OQoxIvlL&wyQLo9SF3kQPJoJs=D79|sZ<+0IAlG2c9ANwsU z$$%XYqVLYYg^DZV*Hm2!52I7CKH?t>t8+_yV)BjE#4oOnWkV!H_C;{VL|&XUy^CbE z@bIy%;#3XC;L+vyc;z9&{Yt4rvqb57zeOJx{VEmgBl=P31DHc4d535&t(j{k;Zc~4 zKsRH>0J%mw@>tMwa^ufLz&7+2GWZ#@#-7X^UUM(BJ8X}OK_1>UA5RSzB**3rduG1k zW>IFbyjD_EPUDx7XR`O#*C3dko=`;Kts@nRqpQ$AnNve0*P&E|b(e;5;@-V+6Ppa^=Mub28 zz3|r%#CvDQZX_t~2y>U~%)`|0|D}^C7w`1&&Omd` zYMprAg33S+6#G)@aoFtU_-r%Rjk#Ie#vY9gSbGLbqdRb_Iegpu!`OjzAC$72mBY8K zgpkdPBY_V3bkTm0S?|z-*xq0=UxpsPGSTm%DW3)NfV=)vdWoIW%UoD~%-o8#MWEU` z_ffRA9{c)+?6=<+*d_6xb&{TmVd*2a--r62Y!jDmu1oTH*$t~@r(_>?K^wn$y*f;d z{>FDjHQSH=vL3xC)YS9zeB;mVV!#V&JC7|rkCt_2>+SW)HQtSaS+rjc=h!ROQ<8okA>q^c%>`X&U+0tIItiab%ysz2LjpD)JNR7h z5R5Pp%2+6Kc)@ijKZ~;=dqnl+>)i-y*D%1d7!Z8z1ABAg%*%$S*FCp3ita8xAX7wk z7_qIF%e0$@(y_vyx2anTtx?o>`I7MBCJT}4-=v|&Rc7R_YKHBNT)Fzj;LIFClHCH> zVHJER$)oAuU8YTm3mgMtpXeW`Im4|7`9O@vGyMQVWZdtU+=>3m*AlR&`b#9CL`dVQHYQT0h4T z9#}_9kaXiCXohy3zoJJyNyDajMz=*9AVXoYJqRQ4>r1?t6ZKEe44pYvzdbArBv%)I z<)p~eX-%2MXl{EREbEQdtc5f8Tv80C$^T$|9*qc{|vt zqI(1P=A2180#)K@{OX^h`46Vbh5OImk}oXyyO;g_&@Yy%(=jmA0rUPqz5t_*GWV7U zqa0l!xTFad(#q&{opMOb?;xGX z*sBo*>lQqgU$BRo(YWMg#ZYXK2y1{Xm{D^Y`B6=*j+p8*$sRb4tXmlFtCX;+Uifi| zPJe9Q6)xoi^ZVuAN4ZL z{&#@(k4d_D%yxz?I@a7f9otuk?nJNUr&{Uol@nYVn=yCA?@AaC*V-9~a2j3Ra3-iO zglz0)$ESVMCtS7on~ad-;h1}s&zC=&KQre0?uA21gyp3z%0R14&n{X8_3^GAj{e5d zeYn+TrF&T}T+#^nQ&vx@)4~;!04ZjwIoCeU8j}hB!Kcil!{SKsFG0GUov;;j`F^+l zNr^@er(&d=8lU+iMZ?!n?~O16-CxY^ik_*%(YcU=W|Mukh!J$OB)}5_tS5muhd{yk z3zkN5B3BZenA&V*ie37J3X0$*^&2Z9$$|CU0y1A>YTi|M;R7@bjJKY<6F`tsC>JE{ zGow}*S$HWf-*foc9=WD)$}|+dDZ)Pdd9=)eyp_QvRdhZ9)Z6|E$eh2D^2Z2@H*p`X z^P_1GgL{LSZiT^Z3zM&2O!-mQz>hx-+f>oYEmEX+$ZE9ZqQ~1P!>+oJ=kEt3k7%J0h8^z z%!z86fkY0&dp3$_8D~?1xaN>Z1m#aKJt}uh+-Ei((LdVt*4BwPA(qoY_JiG09z@St zt*t>Hme6-U+RtNpWjrA~l`CU@>vQB(F#8a-(gV>It~?y8v7(rhe$;PpXK|&TOOM5U zqL^PnkLHE)P`3E3K~}x>_&3%Y%T|wxT#pBWxBj68@cTLapXvHDd-t$6VR2h-27GSp zaePIGbsoF0YeFvI02I2}QY6Lj#^ld1ySmfs-#Q!fi@fSDp>(N^*jJ%RQ!p3rST1ig zqXe)?w?&&eFT_YL4e`U*6SZDs)xd)1k(bfWLs+*&WaAAy5@!*RmHOozHR>KxHKv3n z4CPaJ*~Rz7X$74jLKYXy|L$FqJ;nOk4;>j5MS}~wtHtkEy0!Sx=5j{PK^0WW zW?jjAWe{7_ba#l_z-XVxu^YB|70_F2NC0I!^`KrkQcqS8DTb`$PwD{^OvkqMxov@m zai(NLqK+2-8e~vz--JPU`md<+qT~U)+ICnPgByyn=+6MMYMhA@TUdP`3!VOEK4qv< zPw+b5>v2*H389ld(|w$GWhhh2JAVCI(td~!b_FQVEz0<@3M;iqCwRwgIPX+?fxVLE zn@G_g45N80a|PN*)0FQ9uXamMs?0VmwI(=m=p&bbK1QbdCzO4BSF*V80_*A^ zb$F+P5QW*;KcKm-DCl6278%E3h?}z5%uH!w&yJpjCgs zx#|pmU|%Zz+tFlDWKS>}TGiG+4LezR1Asl()TS;QR=ZB!7IEl5*_!aYg!8eJS0r^z z)$mv4RWmX7Hme=ZcEmWRFBI80W-4W#JbyAOJZCDp5S=X%mQe8Ar({XSBxYKJEGXx2 zl|3pdI70SRW)Z`|h4|T0PjxOtjWCwhy_uUy^Mi5{7j7Q}K_HPFdA&;$I{fC)X%u@H z7C$2psQR5sAmDcQnYdjhTW{d}<+AY$JW<4P|6b^<(+{VYRg4z32%Wp&x8Kq;xT5mE zS9k3+j;DAt>;ii=q5L>ia+vJl4G$!09#wenC+=Vxj;qB+IE3KO;H*TbF_bb`xwzVT z$+IwnOWQ@F;3F>`*Hyb~4Z@>{jH`zue?t=bZtS_<7H$L;t%%>^|3Ch@A%F|;;rzN@ ziq+{a22|>=H8jd_1~ScF1CK!dfT6>N|*pI<5@)t0p%Q{V3la zg)3qLQD&>GPm8abDN{U{K zab(xrl4{ejZ+`he*Jie}q`Yy0llfgb-SXFm3&r=CXz_rn*G1&KZmc#r;kK#VoSSxY z8f1ZsSbLUFA%S2%8~WSRgk%FI@0yfaH^YBC*Khvw=IjqNB3vi`N!8ps>{5;@3bw=s|z|NXjue|Zvx^wYcg(GG7lm38Syy_TK z8VXr506N_tW_>3T$Lstm3t^RAge;3OT(C?IOK9DX-q|YRzxJlkq-FpE)Y9QSN{ZRy zp?HwhYxwA=2bt&UHboWre-Ea>En9K%DwFU1ftB=M5+av8f{=o8gN&q9ev5#ih`FB6 z^EI5~fNB!;#$29MoT}xF_Q=%paWu`}gT5TBnlpPC*ae?h0)p0_Ro$BzUqew>P5I~q zg{d%Gq6SLEuI;$leAAHGs6XIRyF|Z$)karfJs&#-Pxk}bQBxv36>zja-h1w$8o>tc z-=jM_HX_i6bp~1lsl;!in-s?%VA3F91nnQasaoeNgip#%>$N7M3>Qa!aXv>_(R1NB&H>q*nT zUOijR1~?;A!~_sn{hXl@IEr&jgAX?ArWcCz;#q(d%`aj+b0w{!= zZIj7BlLXcrgn-Og0Ak|N;;>=p42Fz$uay6BKnXC{+1p%-Trg4^4Nl^dhsrZ%oyHcg z(rYh`p!@TeRynw@APBoUt5?|y3gW?jF?+YJ*e98G!75MaNwyaU;($heq#U6r9I+LMRZq2SAly;^? zeB!f}{C|X3&Rfp{nViAVH~zGFAC?JPO*PCqT;g*rdGGnOor){dPWowo{H-Lmj|b%C z30$Qb|B7mN*k{kV=8&z+UBo>{6$$+O$-n<<&^i7~u%CWbA^&Pg|7n^3=ik2G0}yVp zB24?lKf%Ypx}*Pd(q8ff+6itscl}L;=)d`@peGlA32&S%CH8x;`oDRe<8?ql%K+Q_ z9@PH<&i~c7s23l-6w^qlt0{TQClmGO^xS{A}yUBSxal$9B6;aiT`aF_CHJ;+(+OM zJ<9w;U+BNMfqzV*dnLeVfAFU%c>a%`1HjZk=gXO%KaueM@e_Z47nfRrZnXPrL*&8V z*WiCQNq}z+Is|0%ZJyoa{~z6Yn&@tHZM8|Y{PzjXyF)C)d&WCu&r#UlKi`6#J+8U`?RwOgMaSZ8_eAfC86zP8%pf5n48QR=cq1^Hq^(&9jefQN!j=ioo{%p;tKYn<@P#{ zL9Sw1Z))}QGY&$?&N4Q8N^#jMmd|_DfAod)$neSuGbR_4tmLh_{fp@E_0v$$4b4{IJ7{1u5%ECwgG5Z)&kBQM&vd5^)=R zpXp1Uag_5Kluz6oIAFHj=`d2~7_x%x_9-0DS&W0j0qndI?HOhz!DWeQrIkvzdvv&u z3r@Z0k9WZPYzmmkGZwUQ-!E#Z6ZL}UAxD*eViI|QLJoq{mj}0TnAsS(lZx*;swn)7 zzmy|Ps=JH` z=8sl(wr(K&EZ|DhZxWxN)_fltyLTe+@deV*>J03xlr&_$>l0uW^%5HZtG{<)q}m4| zmiTc)wR}XbG&F>TZL3v3E>Eg;DDOO_I8g7sPO5P1TN;B~qm9?h|vZP>o0Su)@g2W;d- z_v=IdZm>CUM!y@tE3&+ju!=D38Ca!e=$G-VdnTpFYE1C^v6Ws}9BDw>*qRim@cB5S z>r*-NhME!$9BVxEcKR4jqAy@g6Tb!_Ky^2}La?&jDf|QG$Hb~4r3L8AR-E%Iqh-aa|&XoWvUo}a|EI!-R|1zVAu9d zMpdRmE7`zt=!=P63-sywQ+OaEF{WS@Y}WVvTGiWmT-OKCko2i2SYxe4j?uV3Vfl*L zHuVZKph-J_sN{XQGF2W}X9!tnFwD`}C?;^U$R%F#^k#G*2EL8@IhvufQVQuMZ*nd( zUOh{Y6fk>%am*MyYO&xuSpKvIf*5Wf+>ju4SXhOsCH`d2z539-O{}b67#WndP;<`f z6j5!2MmP`Wg!bD7-keRT8}GDxuaqsBKfqj=Ud2EB zNto;+w@#y>JrlJw+39KeQhN)W(qB8pjJXW&8fHz1F8GlM+5`ho@E5I7#wW?PgQ6h; z3FNyx<~B>xiy|koP)X}eyosxHuy*6ph7k}T7JMjK1DkJUw?N16m`kZdC3F@~3f6@f z`@E%xlBT{stF%L~u2*PUSFVRR4>T~wnX-r(bhp-sF0^hdSz`HDD8Xk=YG(OjsOL_7SqDS!Gm6K zOU~eni1SXUHV@ewba;|;GDg0a1MX3Se{d7^x$^MKB@GfrO{FEGZtjVo$K;Y!w+gsG zu8v2qYC&MZe>y%OzCL*(kvvXY=`p@YzZw0r9B-x5qWT#|!rk^TKiC<(gtS!O(Q+!X zU(SAGo4EE}KatUPRYY-T(u`-whF&+}{By7z=~s}yeRzFinrCSi4vSkDZC0$lI5#e> zWzkLW9doO`JerCYS|s|QeM(RNX{i6j-`1q{6?YPgZW*fgztX(V|NRh324(mTR;W?l zrn;Cg=Q+pV8!+!(k}c@{Ri;4Sx8pW494@z#`Rg$R(~_K6IqbC(No^lVanHzqo9y^L z-*S|Vyqmbe$|zU!mVSxVTn3cu}0` z=ebm(f@wo)jMNZ;0<~)*EtRJ(@%aE%v7*ehFkz#p7VR!!uDV?eJth}gZ%jtN{gZE&j$<9#86){W^?(>+d(JUU zhZn@lBT=gtMyjl`zzoJvV?D!r%0lacp@ImrK~1M|(Mo43qLwuWJZB)-@B%EzRfzP# z)nIYy{`Se5n5gjd0M2AR-(N5r+xmOeJZL4oIs$TU^M=+=l!t+&_I5yPeWT_IfAS7k zx~jjTz&zF??1xM8#bkM|fAcK*D0Sb%*u?Ri$$K-mCB!--*fRxb@+A$t^{Mi`k^Mzl zOg6z%V9~ekqgSnZ_fFLqXW>AO3uvV3fpJFI_+4-^t1G9%kEx5tEN!A-ul+2)2Rxo^ zdJiA^rlvd%M<4ZyT##D5VG$fk?1b6Nl1@4uOZDxYIV6ODEc^8^MtzWjeQM&`TRqOg zq3kHXagR4bo1deAQ4%^{5wMb=HuP;IrKF%rkozcytMrM8?O~vdu;%GBt`g{%2SGd} zd9@E0_p%4W*CE+_SB$3T^+6+Emzv<{?Lwm(n7ahD`bkrYn%Cp$Ax%|E(CU;k{yW4+ zZAg6J8^*VAepF<9HCIwdZA@%y2|w^;qWChNd!m#HFH4Xr7^7E@EUj9)Y4j`j;bvJa zPniBCRCcjMT@2^eHT?Di2OsZ@`H2!ZWl6)}(e}$43=PuDXGjE<%=S7vjA}|JHjy5@0GvuDfPLT~(w_J@VhuORIYR`1PLiW6WF3569xP$2WqI1U#AuI;;u{on^}TK_BYDODNL2HvdY&K#`wP{uu5}U_BmxA~N2Aq!%|CGUw9sj3p%H$J9tbWJitE*hFIE z3LCJ)KdwbN^nQ=N3xOZ6wd8H~}KqU)r0pl?+m>&@8xDi%y~A zsBXJ>Q|$|wfVZTC=-oM~+uy&7Q=}1rj{}uvnnRzYyGt6r+YEUjyq5kn*`3s6Ns};u zEI-D34-zt`FRM5kdz4wl4vir71Ul0`O zh=|e=0R;h(-a!OJ1XKteA|fEYh90UEk)qP13(`yIy@ZZ*5a}iKP(uk2ASB#}bI$MH zZ`^af^S!?J{&~j89}bDZ>}T(_=9+8mIUV%N#^tEsu6&UCb5|Rm*;j5|E;XluRjEnM zbMgwuZW0rl&h)ODV~52Aq*d-0om4k`AZRs&9InkFXBhuWAcvFu%LF!C86Cn;?pIeO z*_42=2cXp~))izMvE-cE&DBr)ht>9dYM^4m_M8lmDf*OFY@5>Awt(#qL75nrXRL4V zD{I_7dB*I&EualMyPtirZQr(PAPZ=`{%;(sxWr#DFeFfvB%fvhH!giGXUh; z^*d0UuncX-6?TBR>-_gjs|vBf2ug;y(gP{o>&C@E6$bMaKjW%b z#$Y8WQ}e-30OU7t*L`JTw;C%^)_W*XXhx-iU$_UJldXQkn(iazQHglXe#)<#mgpKg zZI^tSOkDXoU=QMwaBoMq#7xb`*$0`ml}R?HOF5#0{K@fF!67p4p*9$R-IEq0d$m!D zrblS-g;uW(C~kdBM?IY$D@qJ!J5Uq5de4UMjFp+J-N8|ep#x}L3*cC4=LfxyWEFK=oHc=i7O%2NK&sFS@emtd&Rr{n+O$x zg{~&a>}_zfHK)x2k4_;UiY3i{+Gr|w7CEfF%%!|x<&LgUw_DN?--`BC`sHXEaBLdS=d z+dz(Y%b7HxTWycLS$Fjzs25~CL^_ND0tM*1IO2CFVPmR>m;NzNoW(D`IEI^m%m30P}GjRcspi zU=DWoz!WQEZ#n?+;~L-p8gPtD7=^a@j;xmyPpM8RAd8F}c4k_CoDbU#$&6^L%9h*; ztRls87s@ijp?vgbX>vY+c7_cjoNvlay|r->%6_`P*WI{a{1-R3HtlR+K5s;XUb4 zx9$mLpISm6rDf;@_TctlaWx2pc2yS>{zhhGW4}Ms>7Z3_yXJTB8dMa`>fo1lNfxX% zigoEm0vYt~3{ZX*eoOa!W-RtP%cjf>hsZye3sUFWO?bD=`f$RtysnJ=zd^A7RdyoP zWJBMySRBZhjBNCrvi8Q7OF=#A2Bx6f9~yiO*t+*tX-j5S{2GzG_UX7GDYC3Uugis{ zsy({XWOryhl=$S=DEQ*uEotzcs@r{|W%pVSdF6(Y+twgPb6pROnpZyQ%#1a9*^(l* z6XQG8K+%K%5`e*mC`OF%{%o_QyAI4(YTR;2QH}bxAgRnz&s5X3`&cj{BbIH(ph0I3=k0Sj9RxIT%w`97+=$~pA zAAbE3yOTtAex(gB6kw@MO^&J#ijlG4| zknn+D{5sz~TFk_9c4BPJ8x?c=G~HV-0($CK4Ro>7VMM`-nw)6lZ(RSKTHiCFa zEkE3wp!!Wr50-Pvo+^GzdTRE+m_+^0C;ETonu59I_oP{?rNrzYaf7?%Kwg+!>~<{` zN^2U>#z}%0FD45w;a}MppXQc%_neQ_HpJsGx|l}v423+`%auKqvia5Eq8GR21e9CH z<5Ui6dcJrYz~{GKgf<>1$Y|Zl(*-0=9uHp|0G zo!Fnb|E&=K=UQXy+I9mqpUw+S1qN|k8S10R3W<|n%nS6aIo>xy_C$U9FNX2|*Jb)oi}=~(yX)Cl5d~pG%}y!1 zT-^ka8ZD0`fiZ@Q)9y+FOS>BLRY@QpIBi?@g^QThPA<<<$%=vGbzUvwqtB;9OU!s< zq^FYO=okeeQmc4*Gc1AqY06E!m+3ZRaS21^elidg~Hbj zHo#(H?SPnAZ*AN=cFkD{onoeUcB&1qrry2z+tcN9hgy^B?8xbzq|>>6R#|rQDo@CX z{bcR-My;4uMe`K$u6JuI+Iygg7{5j`@QiL1WF@C>c-C1txzvR6dhj!T1FIya@8gE} zRJ~rI8eiq2YfBp124Ru9o%W~crt6aZ`~Es^DPuzR4FJGRMGQsDh0?J>R1rM)m~m4T zQi4Qj_8!ENa<%xO%YX)Od1K>e5hitAWXpyji=%BxHucMs=@3#x0Ur~tYJladF>~|S zMpF^M6x^YjX#~3$8VNKkVe_QY1d01yM39($_2^`t)&zjtoCZ(IWvC@rXzrDOtj>Ng zvI+%H{;->y%@#!uERJ-RO{mm9@&1hhIi?usCMpkFWaFC z;a0)#z2gPtRKl&6(4k;eNzN&9+|-rh?N{_O-#^(49RB4M`l-C+i%7+-Y5M4Msg0EQ zsC0XTG^QFnvHby7#&e{#J|l~p&jq$e!h4X6fA+1thTq^xj{ZP0_7 z_Yf}!DO{rAYgjZ`@F>nS=X>pRE=GFSZ)u`i4t^K|H{20bU#91Tn&`Qx*H;kgO|BKm z;%n(hsOSn6hs8%1fz07x33p9s@@y@xhT^!uiOr*QmU^EEy7f+ZLZ>nXW9;GqnfUMU z>Fe&sW)$8`iI03@mZg#lz6;zI9J%=xzi<{>j8nQ&1kk7`3xDHkkJ+2LiTLY4@1_6| zO1L?qb6iU5cHHMa^_}cu;mT@Bs%Okm6ULwrQB@3or7$sg9^E}d?{^xFaK)Ytlv&S@ zlu`cP0&>BOs^3{~WbKh^^5IfLwdbk9))BVvMSvCYplZo9yCG7_;`2=eHP?f9Uk3CC47XupG&M5j4Rh*z!D9zr@P{l?ofb_Zzu;8D zp>C*|YXxF9)X9Sxj;Y<0A>4oVG}%6v0Ye56`RjVO2}#R%CM5w{Qynx%$7KJz!mQaH zRHH~ne*zzyZ#s}Q6oIBuT;LG49|~<@pLbApQg4?b*wfSTDdNPKD!d;)j-vTcL(62O!h($y*`DY3jma7U$>35FcB=)_JqC0m z=PRUm0RZCBwByM=Jp!*EE)c_k={B;iFt-GE&w6}$@R!U+FnhIc>*JX!kr@Ro$s#w% zmh0)p_|7z%v}xLt?LN?8N!_2dL(QJgSTyf{q)G`L_I%(T)JlVkE?f3>pHi@6h*AO_ z`iY;IV)sH+?5Y}i6vRY640|pwN0s|hne3JzhT)|wXfL>H(;Aas8f1XW1_=6T+qHiT zLwbmx??CuY3vh<@j#H22DUyNmd)V;yew{#_ZmY3+N&Dt$4Th-~I1$p$zYj|l} zy~;LGK4Wulx7})h)77uaIVyfpcg1}u+@4EM?!p@kX?`17pffjj)3H=nLKD*LZa;0! zo73;qiB{9!sUGmjlLH~zj9BIc^rhTSZM1jqfZh&O(qXrxN3ZZL`7R=w15ey5!*T~v z{}tN*-~Jt?0zgHCE~u`4{!8?$yYTz@l>{TFkNe`Bx9 z|6A_O$>jJ){~Q?mCIQ4f(Sxu4*^m`5b|yXaEj;QE!`{EAIN+iJpcC+TNJ{0O4V}FM zCWWNUcqIS4AAK1+IJJNP(IdUfB=V0|?q8AnpLqzzz5w?|9R;D^U(hf z3Gn{~BmcjjbYLv=blzD14(mT3|{BnMhhOL6^53<<5 zo-d&HBz`$MI%pYP_s{Z)5*sl5N%DD$_#b`wmnoOx2*`@=*?+V_1MUKYt0qK0xBuLS z>~aO}4_+!rqNmJI|7i2b(~tm5NBGX|Ki>=eJD+#c638f>r`v+=|Lnn$0ONM_1dkj3 zaL(gjyQbn2{P-Myllo^5j_7jf{z^3dvzYd;5Z9>>>{%yS@BV)_GJKy9SUM&De*XV0 z)cfy#5U_MPuq>V~v-ancVwYdwW&p5scjYbU|ISnXPwON9Kga$5kK<0N@@99w>6FC- zT-}4wnEZbN+TZ+>cn480%iHeuMW%OzfWv1QmE75V(ZYJjmO(UX&a4E`Pkqx$m@HwV zQ)Dv%GcH!Zos~@};a1wC<*H@$^~V8QKAWj>^8ggSqRh+=xUV-}>trTXVAc&~F+_d3 zB~a~PIaKWcXsYpO=e@C;sjjNFo7MyDT#h|w11BAkgQ=oE9VBwvxf-cOBa5L(?@1Ax z@ku-PS<^zD;__r)G~?E2gS%~n>$2a`(ZumWJC$J^4`{pD8&%d7$>3DP;3nXe*3g+#AlxvK;<5Ap`7u_ ztACOBxJPQ$Wdjh_v`b~wMHiVgdCAty16rWK%#EDP$s616&*ONeJ}U$jz3mJyGxwD7 zJV>ius0vON1FY)*t{3; z4lt^Q@woLGo{5DdNYOOC1aL`%N|*-1zkFCY&2UtfxcX-2#CQ#> zi$q*97I;f)+s@J)pKvL19^Mi5D2;Sc*x3-WnB>xxk*eg0<25WXY1&RIgq4R-Fotw0 zo#mOa$U0|H=~$L8gJ78lCj#*PEM91ee_M;D9JVNb$rBJVht?V)y}F~AtKW~@^=tnS zxHEBnj?5Xxd{l)A%-w%vKPzs78b8~0hy zTneT~EvQD^v0nO%v_>?lLDYR-8UbQ=8-P?XYOw?2n)pfOCpdJQkVk#=Ois3r0E;NDJWGY88+|kt9%{7ab z8NVqVtur^|Jdqe>Xy>k@(`EMSk8R-ZUnk5~Yg4%OTad-OU$B|*5UUt6Nv#VyUp8$& z-vX>mu%)O}Kqx-L+A{}RwDDwd!hc%yU8V=*wOR)^XX2w;E~xWCh@>NNhnLQ%-H zFxXhK1j>6~<`9R0)|A?r%*3u7mEbUN{2rs2y)Bj8p8S(_BVOQKn>R&C4YGKz&|DY= z;KhtQ8fr!7^1h@Gdx~p7)f9wXS;qJOyq*Tet;3N$#y8>%tZR7`YwX8(OzQ$0yIG)> zID8Gf$`HSI&9K}xUp?({=gpTxS*c#V58TEyec#gYXg^yw86Y-us5`X)ANh>eAf~3a%_VDk^B~!ad1JH7Lb-7%LY1B{wi9oH5=3GIDS|S;Tu5 zfF^wz-E$LAG!XCaK$SQNJ88nygd&mNzL9RM}$3g3v?pPWq5t>FuwyVbG21b7W zjNb2NbzrUSMYYck4Z-TVZobzTn`f5%U1GiYq!|#dQXk_pJzc0lRQoj^eg$rv^PAQk zyKJ@uxQ#}N<;>+iSpslCp87X&!HMI7yMA3(xtxH!(GncyjOzm=nW)ISG5LTXDDJSA zcjp>1eYYg>1cF5yjV(erK1^?l*c>oz9veCgj3IU0(I46dGx%H;_L3dyI^o)Jdl|4_ zem;f^PX-0f(dnYrhwn%6%gH>aV6Xj7w~=7_E&+`*qkvsQBN1QY=zVolXJZc19+l)L z`=<5xfx2F2T~{3UxTii(QIZ z&Wqs}O;lgFi(Cb+);WmY-fk9D$Ux^_;N7t^TAEW0yG;pu>g4t83*8AO)(K!Y{a#EP zaASS@dJV3WO7>YTS0x-hC^KPyK30<1u=b}M9n%saHvOG!N5Er=#-gGkJuzZevVvfz zLhbz5Z`fBO8G(_!JU}S$eqO$Xutq%D8r3=ifdO%@3z~K3%-Vk{4T3q;`zmN^vKobZJ4D% z^punMCJ_u#8w_8rc&E1yZqGJ|xg{7Mc>2NnQqo-TLhe%y7>ObIVX{}w zj^iu4MmjPdj`9gRPI+};u1>qv(chN0Mm}$67 zV!7cuZHn=>uxNIr^^DHi`EfFzWT?DcvE;ECnoQLFjmy;_qG$g9=(W&!SH2tl2yn>4 zY?mo@NNBS$=#&AzwnEy2+f(iDo{^ZtnMC^|=%n0iALy07`4+)AtH4(2%i&=1kzOF} z(kPvzRW1JhVj1?U9G<$CpQAUPpO5-JHXKl%>08K9RHvB|8 zv$0LvyNQTm6_|W8qRu?I_$+!W*ol1`|9p20aBjLj+l^l9F?+g^*5&cLyqAub)WoA) z9@=C=Qagj3;J#CPjcL~j_Xx?UmhiloL1Z@h+htPer%7fdy>)+~EHbw(_!E6!n}4&O z5EG)zioj?gtyz0+Dm3ptZY5vqvC^(320O_>i`fbYx|Z!3!}S`O>Q1C-C%<6%Jldow zBJU#ij!@R~?B`###gxUbH0}#g1Yo#lM}#bXzBvG(ar{sy`4`X4pHY$5$#1X`{G(R@ zedu4zL#+D-?fSP-tzp|!<*I-wI*U_B=$x3{{6i@>lyKs6o^T5F6lu@b%20Hq)1>`( z!rweR&LNIR5IVnTZ70`=zbvs zp@yzePio_%8DxzzRX(ka3F2}FuVk9^|9Jo9?t$tAta_cv>}mK-)#<`#jF2$F<*q5% zdd>pPMF69_G2LwY*ckvhx*$!jtE_+38H$~nk3W(lwRWdk9 zEnt}G5&>9X@TW>9YAxbSosL}XYi-;&CP%`~ogCQiszg0hBBhUK;W<>f1H(-^z>j|8 z$`TZaC>3PiehKf8=@fw$v>42h1hlMO2}1`lszPGgX1oSk-2CnD`d?}6BwoS+^@*B3 z-8u*0A-e_9lYaHxt=M4AxAZCUeTt%3*+LTg!V(;Ub&ILS*ST(d?LRrSj8QM<$E0sI znSC65bz6P+!UT}a>^Q7-BQY|&dj0cX|LE0%*7R>ft2B0vS`NnA@yvg~I=Is@C>~<- z2nf*rLvmt)8ev{&WFLJ2yJorshgPn7%pJ*Y92S$jf6#cGL}Q_-eU{aB%r`(xatk=_`Fv?l6b_wdz00+ zCQwd>EzDm{rUdjut-~+vi>Y1ZC|)Uuaj})cXg70qSte&q>y~tjUbZ$(Th5cTAITCM z0@fw@i%)|4xk$9rUL##Ci;SvzS!fvvOZD4i_Jk<;*Z=s~ z?GlOZ%H20Z@_)`zR7s-$##Z#5Pf{GmT^-(LNAcYr~t0lY%Y zi#tNojCYR!(V?^F(Ls zv1*-2l86THA(j~5Ul-!W{c~cE-q@4T^F7w;XbQ1m)^S?yqG*+=W_#l!9))Oz&MfcM z!i&^zp(xmxx?>2a0WnM0{CSqW*kD}~&raWxT49o(Z?ygrwR-fYdC};)yd(alt?cvQ zwNLNRc*V6~VUrRu8x?|j-p6bI{$XkCgf!_Hb+W)^z1nKC1b$~=kS!lnk<-Y$(0`s6 z!zB9lLy=zHuP|l!ZQ*A?_`kKu&dy)F>?jkG2W{*4xHeYd?w~5`R`l*=#!(E3c%1{H ztes?A46w2%K0u0%`YYCOuJIWc#5#S&h{4I|`M+^J(B5Mf%-l7UjI38G(QnX!88>i= z(-T!MICIJ~rHZzwPtQp3wZd}GJ`voV2hx=4kwKNTNLk;**1Zu&>NG>{fb~15O%H|h zCfifzMEIM4^D2jQr%n69sPP|1zH0Hp`D2_`~CVI=fb5$*&XoXhh$*^&$JVC%HK9f@|>s8RzV;_xzpcBAERK5AFzfisLoxl!G zW8Rcer{X;#f2~|iZ@NTphVf*>`mu4ZbZL*h7M{o5eF*}|9HPZ1)71XTAAgVGt(-E% zV%FsD$rJv;YwyE#X;%@y_Xc74D+AvK8;h(IJfcdY)Oq463Ow%pd-4mg4D@%1LkX3A z=$7k2^SIj2WCYO&z>n6Fe-)VbJxc7r#BlZWgM9^~W`eF!lG9#;AUoLZ*_|wgD&Hrs zInE!=Q5{0frK+n55Y2er10lDL8%ixn)!t$VAj8UjV-K`ONXoYD&F9dv=3&js3wdHGyEq*zW`E z;umh`9?LB+nv#!qib*dSY|xTVCwP6Jih1u8S?)N}o?4v$8`79kJ`gWb=#;JbnJr*Z#u7Cvzvy}UeA~&n z@^-qk=P&(jLFoPir|%eXiCbUgACXkrPk`b}PDX9iY>%I%h<~{^q8oB(m3=l7ch4^A zx5*5ggf9Jq$ z;nkTsUKMqLNxwPp1=MUi^%I;xJ*N1g;T_toL9<%;0cWHyH>*zBoAz6+?x|^a4Pt+O zeWYQldRr_?!pj=n7?)b_p<2x%ZYxlzH)wl<*GpJx=;=#B#v?4&C86YM@!qSividiA zm_isuDAOglS$*+Uv#;$~P*ZoBB-IY@zw+sCafw6YKJz#|>G{GHhYWeh8Ss+&`<^t> z@WRzun3;-%+j;FuZz7=JwqZ6~(nKMM$L^YM9IU!Q_A`~To!aF`B0*(u7N49N%)vRH zhOt4tj$<2nmX2-#jA2ZCv0;NYl~(AA$No}X>}jw3$_`6wYN0n=rKf#_sm(?FLy19? zK{(xmyEfy+LW(*tJ;6yuZgNM#Q0(DcgF7*B$;s)d(kRJZpi3QHi8a2+o=e<$B@DE( zzpgf$_AC8@~uTnP`SObIEIiwIw7c);3Zk$}Pp(Tu9*TIne+z%LS^_ zg?E%A5&#}H4J?x;h6Tg-I%%P)6CA$hC-@L=+ar&X__0V@Is9QvkzU!8>1u1vYY>i5 z&e#&YW}h%*U?f!^y_ zxl6sy^)!5lQ!mp{_{{o5?Cb84G~go2=yEu^IVejLN3N1_fIRQoV_mIiVsaNy69;lN;1sQpgFI zPSqp2G^%fsiuM;f;1VK1-?%5I+0Gea_m0g(DUJfUjKXVJB{$o=H@SBnNMjTMB~6`r z+xflGS?vh{ti?CdA>`&&>aal?WQj>kwxRp;)j)@DKF{zQa_053b>2g+`6iWY!ASOi z?rPub!=VU)%uoup%AMcYpSI;3Q2t#LfX1{gbA5tE2^?SFqbN;IP+#*6Ko2!M%l^^5 zYMVrn*bYPaMu`4yWN8GmlzVS(lUL%zK^!lRvdKiBgUvRIZs9koMoa!m?J9@kg$HeF zyZS{}BxBCp6?6&GPiNg|ceZZ1?zP$lQ1{hxH>UR4ARS?)9jlHqh$`=zH+*6DRKGq; zn=m2UpQ75zxu(|fQTc!|^i0fay)HnlAZc;W#GuI{av~#PBZR>Q9WbF> z#1?@3+!CU0nB|sAWj^a3lh!`wqb+byD>MSgLw@nJ+#Wwuw9F9G%dWd;3e%r^p8cDjOxfX&jI<@_4&z(BxQI&8;SAN%L6&jSm%W zleeeX{xII9LWHDa9(RrMW^L6HgMOxIA--4)5nG2|yR+y3C%E!%G69_ArXv&8t(o%q zbW>xO?@_de(=zFYpFttyUmTR94m|{3oGo6Xa!b9dSY>j3+lZZ$^{vd%dNi8zCHz9#VS^GlN79syT?^Tv>lVmG{fLT|I`iIs{>Okdmx2-p80g*ADCVg~)X|~V*7OEAv z8lXa#iUg3Xn~Wbrmtjp+uVky_5 ziCQ;0BWVkqCO;Cj#rzeFZ2j(ie&S@PO}g&{)D4ZtfgX(xyYnVnsdN>-I1rn)3OS=- zTVu-6;Ni+HoSwuN*J%fdP7`v6c`7ufYR@GscYWlb|Gk6l=!rQ#wQj^!%lBime#C|& zLbs;bNorE+*_EkOFIkzZg?WwUg*wi-Gss5X;JzLhu>a;XsC-|p$~@10J9LaqC> z9)9tBd$LSj;wehY_k1L*;+o%nu(s~F-Zfvrb_q5~aP!+>no7kFDBiE!n*6lfnxfJ+ zV|nZ!Z*YF4?X?P)T+pPX7Hi*W)L-su4!Is9nKE<3xP402aJi5IJx5-(th z=V^#$5rberGr!#rG?g9GU#61}un~Nczd445uDVg_Ll&X^MTpVmPPTyDIAG&yZpB?+ zZTwaES1myQl|yDk1CkH%SFRZ;2Nph_%}&2s{-G!^c?_4`ZGSfQbQf*C!8)d8p$nIwJmfDs6L z6KW>Kca#Jo9m|dlyV=OP@8)TD&AK8Ue|G?H4A)gKNxS|GTfD?~g4)_X>cML|@*Q*F z8PopK@MF6f2KGlp$JCKQ1cXtmgL7W@FDq{H*JZv1u@AmLK8Dlr{EQyWCF>Xj7bwb_ zd_z8=s|wxTqHzUw1BZYo>lpFj96OsI(pLC+W}KINc-U zwtZ4ch}_9}7_;fqF16Puv?gOB<|+STo=tumpAmk-Sa5qrE0zcJJS4vYmndZULgd2| zb!n(5gUEX};|6!JiCusskNC`t!Xby!#T2`u==vJ{x+54ddpBD>NS{1IzBUYA!@8mg zI}Rs1hHL!%TyU-3(~cugN*GbQ74;um=V;`zkY?UDJf^R=Klpu<)c%f%K zEJCYT%teeFv)q?eHD5GVw#C~L*k;a~zpcg@l%i!%jjnKbHtHCH@ROf_*_`Q$2?9USOF#Ld zvjk@tLcW|lJW4>#R}nJyq(mKB4yMKINV#pSmHHiBl;8Yn%z7GGy@r4Yep+eMrTto9XS_6Wz0O5a}Z>mE%x9k2)wes>l%j>QRj@1iIQAdmy2qr&~Y7 zt6zR^P4!{(;PO1T(!skM+zZhUP>{rIrAC6gMckBA85V$bY=cM;5I(vN@Juq$e2_kMcQ0fN@4-HOb^W+PZ!0!S$lDZvrP8x zqk93^+Ze8S7d@AQcdC)%h#Bm_#3W!t6W1OZO&TQ#QGDaP8qa6*bJrYo{LG`Tz-lUnWjIwDIu+e)?W z?yhsQH?8yD?uL8y)-ft(lwLq>_S*<&kgs374%{0W>pzN?lZ4Stjb?q-8&SNN`yrws zECTOOm5qF{{`j-L$xE6=%H4!AMVbQ%(xOuoG0&&v zjAc4w&S`L48=Uhl2f?{(?kdbqhJ1>1mqT^CXZB8nPaHK&Ju9j0mE6Z|@YD6Bi~A|S zW@|Z1y27nK^yDj^Kyx9Kn%y#zctPfigQH|GKC@@xANDyLBo`V=hBiOhzrk2L>Uo2hNWm8VMzcM>aBr^&d$Jt~|>`z?Y%7ralc0 zLRk{EODSY0wM*6EE?v^jIqG(>CK1e@c*q#U-LGU-BV8v`>EpIjFH)Pk95?)`E~~zy zQuT(~Eo1NZ^Ssf$v!^U`nHc;RA!L7*T1kEIiw9r(cxgC1QsCYh5Gs?A-KFJ19WKv6 z-2o_z2__1Ann)!}Fp>VUbrO}&7>UsW?mCwrEZaYK*4g>+t&p1pPJ!FxAb-p=E z?@6*X^Vcx|?(V7+x1X);!ZG1waFz=WF!of-XVbc`LS0CPXyUXFnWleUbTJ;B`V&+; z-X?!I;@L)YJu64j`eHZryM6WeX^X|m-r{4_jDz|7`DsoGXI97?IKaQ~*!Ee7V7xwf zOQsCL(}jC#a}g64@_Zu_M260NoV)6?t)lZQntAl*05Hq5u`-^w<{Dw5; z;vQ}`c_(F1T3_N`8c$br|u-WXwNG2UlVM{qa$E;@X87cL@oN$#^mL(d(+Pje*K{jl=W>)hx0 z7nlosqzvb#*!z6B+$+J;LD!DQFVNPt!W26`Mpv&}T(wn>!t03LU%VJ?;BP8Du9<(x z4%i`no>Ptu!j#dBA3eD&S1Wm9oZtK`x}Gi%P^Faqg2tgfLFc-WK1 z_X!Vs1uoDVfF@Kxi=vk`|U4WiR}Wp7ul?}%757MWDc^OQ%|y6?^7 z@{~9#7KdPO0{=!kd?BCxuNs#RTCP#&lXYa~6-i%3^0k%0t1nkX*j`iJEXl=8qnvbJ z;LNhm>(3V5^#p&H+MFg_*&cN`-||2F_Na5v;WjR~XsSy0EYDEJsQ-HV!nl}U@cggx%+!F^1V-w5`-AOxE!suq)dnj$&>&X8 zf48bQoR*!{Td@7Cr@ok^&BF^KU9I@U&af+Dl{CN^=ZiUa+2wa@^3H7kc3rtP`rw`Q z8?golu>>wPP{O0P$I=qMs6aJ|$NHe#^ZThaoT1@pMMgdBp1#ZK`c&w{l;x}!-{cLn zi{HS4DwsA3t;1)mpZa^|tYJm-cuVr1)h~rp^(z-yY;JEwzJlU=C`P9Z(D&Il791;O!6FflJlaqR~wo%~i{4BNaM1#fIT=_$JY(8=OS110kIi_i%F zt%gVMpi^fLxco2J7FnVuVj_OlYxSqUms31D^qrC4+i_qjs*uN@uMwA_tRNX799C~3 z-$NXk0uB5KTJ>kfiz6Rlfo6v@4rnl)FXcNT!u5qL4OB~1!iM(ivnW%)byMJ?Z!Nit zM=GZ5@cDidA29@#_OXBHB70IgnycAY=YqT!kS3K{zle3_fuylD*}u6h^Y#M`B7F?+ zxab2)RW5!6_s_r2ba!Q$TO*Hg<2px{WYW=a6bH|@{}6OZ`k}? zC;w-oORngz#-32OhUavB0O50WGH?elxntzEBC(et?x@ok@w_>=5iXdre;@4>MGJ^a zcB*AVW?=l9rH;q#(<2r=zgEv&vSNUwn^rMx!=N!%g~w#b8ZYm%*|m_oT~0k|S03Uw z35HX10zMfPv!s(oE3(x1-E*t1hjTA0jg-k5?{$XXbal9KFIwHVc-pm038}y!of8Pu zGW8@wNP%-j*)4?DUh6NdesF1rk}JMztK_4cm=nfnaD&x!6FS^)OX64(}6HcHBeHBas)YyRev)1VSaElJaisjae>`!rtze@EiU|1@4g<~ zG$xiGY-}-*@(>`v4<@hfmR*gMi*Hwn@+wkcPYW5&v$F9tB;E)63AX&WA}4x<5@#Lj zBVk)P;k58g?r`f>`I`~egSkmlLVB;iQ$@9mD*lz}qV~vMBGs@Wh?=?0<+t3;?+Hhn zzL%-;?vdpPJMqZ^TAGrLk?Rg+M^g|j8B*ICJ zsrjxWZA$K)TcAEQWEoYJ0kx#SOtm8M%BgWu`m2eZ=w=oU$Bg5fj;p92chSZwQK~id zRWE2GjAg+riRJHtdl+*MwQ^KTw~!D zF#0O`qd7*FN1i{|cJ|8oOf*Xg<)%U>?me05ewI?-yC8f2+cVACElu-=Mq1af0y#s~ zO33-?V2G!li96Wk3Q}1y*-i>O&0P){+Y19&7jt|LsN)c zw~yuV=3UjXrg@q3xgd{=H=|^zPK&r8`}OO6klLY0&s`#uW0OvX8y;@g@4fi;(iz#y z>8PL?_t+d5&lnt29yQj~e{3m=OU-n#OkT)hA0PZ8({PjLr99i(F{dQepa;9v6&h$4 z`JF7FF8wN@mGkD)U@9^!v!N3}1FLr>-NzNblQ#ncB%(|2)m%}ww6nrb_Z`hbw`v4N zJRQ5eW8L)VQS3zn33{w=PvzqIMRIun<9Odr2lhl-7^~ib`Nu)UUvdwpm%N@FMo};W zM@}zdKoB6-IbSaC^zv(gPmv4>(vNkI=r@uevW_&VX6=+lb&K{Rm1Z1{yGI8W$= zu<)nt@dMpsKB;J;*}A4q2Q(ozlj7`c`)9gnNmiSb1tu}u>AR81RoCxIMgv{7zLZC| zh}LDWBps zB4s*O)RSu<(bt+A25dEZ9ZB*suB`haeiytFdB*5R z%v$&t6aOJ9z?*6CI4rf{_$N}a+09`|kZKK;iT+R@#4>nt-PIBB zXJczM&3uP7;kR>F1NDW}29|<}?mAhvaQdu;j}6{U-;f(y&%dA&+7{i2;rx;@z}GhO)jMt&s1Fk?51E@C%Wf`5)r zrbS7C9wPLOTLJ9uO5yOX_f1FBX~nxrF^U$~qOF=3qGSr6&O9)ZaYO>>=Vym+6@DY* zh(J&MPINYb2SKiqH?lnhVD!RE{Zr`KywtB%KauRPBqD?T2>>);K7vrvZOkl$?`w~d z8tKXh=7VheW}VKW*K1t$jNJWV+Rsv8ORuz}1GujUk9%iorx_6k5dIB%fyI zRZN?r*l6Dc=n}e0W5Pq|F2@HJl3TVVTq4@}d>8X|vcq?~iZ}^wc<_^CJ)>Dak0wq% z+RMvFZR+?gwkdfxQ4dfuO1^x(DKJSw0t=01ut|yDZzUbcSJQl8Qu(z!+b8??q%CB5C4m$&pgqlM|^LMz? ziq~@Q{el^xa{C0nTHj$B3bDgsdsm+o4E#WPaM4o7GO_dHRwPLN+|jHRQlrx2xwVylo@Q@tkfCvczx#aU*;hI6LKNhj=DYF{ z&*Qhh#T$50$xEH2GoHKzCXQ#-g#4#*IO)wHYa3?vQj-d|pg0i~@ea)9h*J8km~q=1 zdS%X4mAmKpeMfE^3qRjn33QqJNXs2t5<)&aU`M>I<4E9u*wAt*ekrbh;Sp$KT)>t{p-MQ4@ry5d=Y^cM^#jBoe(v5JB|bg9y<}^cppK zHyDgA+K67qC}A*+ZZM4RvG;z@`Oe;_t5G=-PdKKQmB^~ z5ZYx-71HR1zH?8%&K8tf4zHgcsD6VpDRbrVQwRBiu=HXl5aJpoEQn%a_D_~;< z-a1(E>&vo)Or20|cw^Oivt8%)adDBnq}tn^yGwl)QK`2VD?rjFFB$o?p8LN?Z+oE! z>}g;IgBZ08UWKcIIF;kx){c{Pv4Xx?rcUW1VagqkJAfXCZxr^Gji>foS1AY9&Bv!%=C<2(8WSi=Q(Q;mvdqu+a_bat^#{GG8hwm8NUq*Pron9}?Xxn@&P4!HDV2SI?Y@J6|IT!1>M73V zBz#q0muiHXoY1aRw2bMs+Zx{@%3)!_L_arFJgq!rdDz%QhzY&#**9cd|Ma0c_X4}L zXD3bLqk;ej4~3So)Go8Rweo{P%IhWXKEr=8`Bv@EQ%*Y#VG6t}wt zS6%+ZwC#iC{4Y5l4P_%APlb05G<=wiZ6ktVPZ&D&*r)dWcyDuYJvd>=_&E5i*zch| z*-qOvLhd=ky;4Wxo^=5~m%{EDj!33Fb$um^c~<*8K4TP5vb0OnNk^!PDOvJMcsNL9 zm+{1Q;33Vh7*~HhC)+vTgD~Ri;=IhI!Rm^%!7xPs!R~$=Au#&OA!FrXX#Z6dlD6f`#pCQtUL;u!{ zyL%(9q7-W{AFOz;8-Z3%yv&dofTOm+6I*N5o*~1ESfi+(A>)U4bydztXtoziQxw}p zE1Qq5_lE)teX2t&2tod`Tw}3`3OLNW0fcqQSh_dkTomu|K4Srd6x0aRX6cY~Ow3O} zbek{R&Mkv`={iTWN;KoRLc&Yew8|60*63HhEOp0#_V7aOwQb=%BDV0##$(Ac081!a zb3RYd#`Bw5J99Fub$??9rMBiW z`M$6iO|AlnFw1E`ixA!VR9U*=5Wv`I-B-O5@S;C{PH&y6$25ysBWP_bbXE0^;=t`k zB_;j4z@}!fr?%MOQ0@y3RXI0sDR+%(fDm9~a9!-1n^6)?9A`V zhcFM)`Moh8bci&S`>66vHeU`LWgvXr-DqG=jgX25&6#gCBSR@ls ziB7$Ha_jKCey&=)#(bF;r=rlOT@+H<``33B2{fAQ>8d-wf5F;pb@A&5#`4lHY4(#x zTM}Ju^ri-xu@|sPTHI7BU^NVtxv}qYL{7c;LYw>n}PV%1_7g_}e0GFAqwJmF8F34oOR-zrbI;OcKS)g@Q*&(H4YWQ5h0; z=>2$|ZSo&Eh9_qpQo`j~q^nSSXuocxGpEp%u+@jtRU0`9UzVzb4ct_IMOoW5wC9{# z(~fu=SbTs*FE~%a{g!HkyrWIOZ6+H6Od`5ZPHj0r0{Uv!@j94BkF|A4+V)nM=#R^5#MsS%**)fJP`i<%Y0 zO#O0yK+}n|&p2{5%tjJuSwnsSwF{Q}l~Ig}}QToJJ4qIUt(04BVRci|^DKh+Nfi5fv{d+f$0WcjW2w zB9q;POUz?qk%VLjVkMD(T=U+X3x@aT zp;m6+wj}(qXeOO+xG?%P&}jrHY+Wmu78qvF@KspjOe0;w4)o+>E8ERO+NZ1CZ`FVW zi8^k4nxUazIzB!B<&)rYRaw?sON_B@&ve%|cVlqX**zhSRtT;&0-DbX56MKw{O zvbUT_uQu(c@J^#{78efXI7zk8iQ~7HKMiN@U50E_ij@f%td+wXZ8m%&GUv9u)DRyz z8PuOCh|SHl-LZD+2GT}CBkWSU&?IvC!$nc(LC4(+innISchbR*w&E3Z(tR$V?j6@A zZ`PmqObE4wdW&=)%APQ5VJT6_Bxs`}&|@y?QR$*#r2UFpw;vfUtS=qSJhz^{En~rC zx4xqj405WJypKaP6v%IK{c;T5?_Hr;#QL1Nmzc5uQ!?!F!mmOBUg+F#&a=4N!#T5I zGJe?Xbop#mj4iK+adH?9Z!!m`PUOnEORzi2K+J6RC^#xvyetYf(_)&qg-7xT*>X;> z$+EwJGO72IV10mCrk|ABQi{%G@zF!Sb7c&`;GMP|0$`sjZCZNs^&ytdh+Ob;^E3Y7 ztHLv~1~x&MxrIo>Z+#M4weV#z8Hx7vJ7j01)Zy104g)9cs`F26Xxl0U=BIDpJZ%sMu}#e*GB_*w1UEzo0!S4QMCvui!}*Keh5^ zfo?yUXFEi;3v?bgGr%(6TE*Bi-YjrK8>CM88l;IH2>N@?(%uoViQkL2mI(9CujCAT z{2R=ionylKFlef1h7K{uKG35njohXN$2i-N9s`^@;>+B6@-vz`|g(YKq{O zu${R{gXIkZU1JJ$H>CL0w@!rRELv`vsnv$|BialSbAhASm$M5U#Ud0g_T8U`5WrHwAAhFuF6M$JkfuhA4&$}rKurMyes4>`eg|6ERjNj(_a2Rv;Z#4 zKmjj4-cMi5l;ssNoAvmW^vF^U*%Id$`=L-n@rexl(J9lkUzo9Wxtrx{W@~l{CK358 z90nqpLT&x*r=DG9=mf>1qGBiam`z=_x|C@{uuH$Qe+6T^6Md=Tsg-1J97{TI!mfX( zkAvPw7;lnEpSB7%jT9_&*tz3Cj||*D0BS8XqM2TyuM%-dR?_wz?U4N>FxWRqwU2Ga zbGQAGo$lKSTcYx9LcqAnesXzzAdrq|GPfD-xPT8{Nuq+Fjb4WxaC4PZO6x3?|RC~QgxQJP3G zSsS2D=jzL>KZP*b4u81`CfZLwSIdxG83^Uu-NT2SY;n&qbs@CG7>0PP9!Gh6-3^IK?aO_p zvXTmCbPTROD!7^7w^lbOMPwBiN!b`%oml>(Ca-uCO5O-rbuR*qp3_J0Z;ZxN=c!qO z$ZT=nqN?ng*hdpDEE`K^D)m#eIr;ohbJk=pD_fC|Gahb9$~;A`%>THV8zCiopOU;< ze;4Q{>Amg~>NJ!4rSvYwT@7N|QmRe`y89(3Q}Z1I+1kxt&0W4e?O=hW;%{{AK7lxW z{fK_#htucza$c}=9%0{2Sys6<{>U^j?)9;QAt};y>E)%>?Git|B(FvSsZxBIn`K2` zOac0Y;rvw~TBLn2&J_$o@QsYG>M8xvY5YX>4LSqzD{P1njG>LRE&I#tr>^V!lIEc_+;8a-JZ)&8q?1Ax->-FMDS`i{nf!Y& zyOGOmo}aQz4Yn8@#p*fj<2|sYgo8pn-#q$tB?Jm~{;)N_S&4wbWq@97qH+4U%y9Bo zaEN-x>66;qu6O@OgxDXEbj6^Lr%kkTWgfQ-VLFpHSx5XFGl(&SC|ChJ&Gn5z64-1# zpPt7T1F?G}+!T3_Pe5kp^XFBE3nqbACKoaIeo1>fZ>m%nCcMowQ$XC6to-(HREC!d8e z%h4N%r_Elspc&76$F@Bv+)(pzni6Qm%CzaZ)ZX>5MDr+L;|w7eB7Y}O`<(aC68^&| zD%X9^6zoB>+h$XIQk7XlykJi@%fXF@%>|B1%8i4!8A5M5z<+Gou$`<42ayv(O7alOcU`wRhO^=eeg}6~yxC2|k3b z{bMpZ|IBpe8`)^k`u84Z!fXD-G<-^vC?7-DPx)g##w%V@Q7U$#iV~*PzOCM4-&fd4 zjT*hn}w#l{b6^S{i16B>RCjPK|)Z27#P0YVEzmexewZ zmv^03=3*muD1j$TWxf{&TnjXhE^1sfncsctu3MyCY{o~7-i8`_nXOj^`z6T+i6nZ1 zN{YL7+iGx>c!f*oETx{QPrgDlI1JToHI2Z#VS}O^SSMHO}{cH3X}w* z)~AP;QzJXTd#o|eTf(<$bB5tK_~uS%p5K+qrE&=AkJ4FC>4T`>X^~UEjzgG1K^`t= zfZ|(l1af0y8>agj!E1Ar*XEv$8$seU#jxGq{ne9<&YJ z8}d8TU3j|{KI^uvoca>3oak^fuaDzA*9wj0ed}qVj~1Lr>oXxg0<3MTfc_r+o4X5b zS9M!#3PRymYp;{n(H!jl+Q6I|dSix{SN0C?9IZ#&%2+j>6)fxSYB6dXKC8WpU&1x* zbWk6jB&7W?L149HnmN7On~&sh#=rWN_tpoRgmr?wR)klNr?H9Y#{$AZ7qnnxb;+B$ z0m{K$!}AhNYx4x(sI40zP?oVHD)p;QXfJzc$|1jB$2$yTr&RS@3~w?BwfG8V_nmh! zFVUDk+sWESy+gdDq3OMXcn0LXb{Z0>KK#8siclEGJ1ONHDS1+Zw&1m$KVFb9d&88- zFm(H}@($RV8_z%pqiE61XXR=nO0l9Zqn+A=8dZaBW2LOAdketBILVc#CJl~Vj+Ahl#G})QR6}Uk>dLs~JneYVYkLYLn&F|2 zZ>o?g^G^U@_B?-ASnX40FwLeNxv5e1fwOnPJvYS*Qy^qqEU(#iZh?utze7v`7eOcaO?uoUpVkb!|o) zvs0vUOC-`;qrA?UQDQg^w|yD4>=(@hb|VjIm7|+_`l5PL76M}?{k(Q=9z5GHeqC*( z-~h8_LlBj;eS^gt3gm+hG!S28B;bkCDjKQ3RsG$=u=|%%X-R*jhQEl}6dxp+Kg6>1 zyTSeMvtk=rO1I~Wz6jFsw-YuDt-tC*9x-t}b25t``*O3d(uro^8bL^A>HSQZ^051p zTxjP`&quyMwc*#3*3phO3^Eg;$VioE_X_QiHX2%gigyIS>U{toQ3qf@C@pox`)sTl z=|0=h)nLOJbiUOOWD(j8Ln*bO1C zQ;Gwr-N(ryTauK(ax-8*z%{=w>F3WYocM3u77NL(+r-p&+P-Ugq2@fq+}A?rr5#_R z=1yk0>Y`Ndj3++9-=Fs2j(K27$toSZa|_4cGGb=C-{Rbh&P0qcE~rY#T12t9&hS@v zMCv7DZa-id#IfknNl5X_#-*GQv(-U11t@vNtUb!sFA&VuSI4yyB#b(l;&4^@N=E*k ziO~IQ`W$niv>!F2vY$^9;Vz;4^#^XH)La&j&%8xPmNL1|m+SXHy;RafJ(P$FPUOfdGimM#Q>DI^5PWrdQTL3w&Xnw9DSNR~vJoB~5oDU}Zq&_#MFXpZ9Wga3~ zfaV5aRX&6I`;hMo-eV&0VvGH-hDY&NB-9mAzZlcvgNO`glwi@3GnKpg^f2RY zwe(wVg`<-0b^1ecspXcs_EeP?&FyWSZI~W!`@YYqLOehmh*oc`OO$MI>N7uo2sT?M zVx2n@q2ACJv|&yMAJ(PT+0JS(RC zJ5t_<+4y;?gIq>kGcWHxi!Ij^PGH}73o*tY58)voO-{sk^RK^Zk+D1V_o_@U&&uAR zs754j*O`0_ZaubJ+2zerD}?*VZAN$z+YTK=os%1*y+T9rjS{=5(Ae&rg63kB%a zw9SBBw^=g})(TKd6d#65t7TRRhMUVjE_=n^`(hqU6onM|uvlDpus;3l?HT}WK8>Jz z_IZ7-Xwp}mIYdGB%(&?_n}?>GVNW>isQI%0I8DIngJc1{=^h}FVOql!_q*EP^=p~D z90DeEH;}Onya4#E_;}+$cXL(MH}MwTs-(~smHzF0n%Q&oCd>lm0%m6RkGN?cgQgzL96sG>K*pRES)03 zjjY}PwRM$BK>kKGPW$;Uo>Y<8uN&_c!U2^G#RAgOP3tfjc0fu;mHm7t?w)$WYwUzW zkQin)%uA->z7&J@JD1l#)qeiYg?%HeU1V zLQLJnM+-D~WTVjV7=!hZW4iV+DV_lzSYVRrBOhh=IAwGER0+wh(JE4PC$EftBaPvX z4jw_1ran^(XcEOMHl@8EtidZfQIfQzf5=Wm`Cqx%CN};-Lg%-+G(n$M{aAD+$4i0J zXbkcOeY_htQenwGjhXa?$XQI1)y6?C2ul5LHs>lT|uttIGuV|t-qw{c&aAnHK9!a)s)Ge zJ%Wi1XlYy-d3QQ6JyK@s2EGAB$q}Hd_n~+h&v#xT)m$m&mKD6Een9oVyeb${>|ig& zo9r_l^DB>hBM=0-4N$ecrH{G2u6_c!9-HEHdjO!4ovM!7Z=kT`-IqE!-%%quc(@0T zxVlNy0D=^~@{mHR*XdJJ!t5cW+hTbW<;&Rby|kY;`GTM9Fworv9rO+q9GuyIip+_G ze>ptcecN?o@C75L)!}TcOT}^hl;lf6_E~O`Nl4ziH}kmkT+Umj+Mr@h)n_|h&#^L8 z6-*-qIv2&=_MDWRG=#^2ccQB^hB-dSy*M9|xkzzg_Z2GBwm4mg#7Nd%wWw&9iqRvv zSm0txalZsxAAi_T%ttc|GvJA2B)H(YwG25*h-b+3YezZ#U33FTI08~SPj{{m3gFKJl=D4{3$}f!*D9aF!2t}ZzZH`T(JPIaeaTRR+aL`t6 zi(*e}+r9vYSkOK-SCTq4G}V%Rh6T5!j7d42Ld?;uP@^zS(B{nF4EnWAcSRr_fa(Q! z@ph1tc8(4ae#?iG;WvE%)d!zQf7Po=v-R7TQYfqLx3Xx)3`Zj9nmEkoxoC?R7Nt3M zlDL<H;BoK&ppFus_V&5?y7k+-v57YYcI4ik2r&>tJYpeC~!1-HKw31^nW51=#yL zneuyoamw^onsNlBQKNk@!aR7XINtbAFYxzP;_c$6&w~8$JcwjFLnsC@ISw6YHie^w zi-BDiwTYPj8tMHvX)L@bS)+5?+c-V@K!*Ii-!`iuwt(WkIQr?z2~SuqeC6?6*koMD z+RND0$o1a{3Dwigae+69I-BHpQ%=Hr=1C$i2EViXx z_FDJ~T_hipSf1rX9g62e4+>lo54UDPY=zQ!96`RaQNx#|K^8Fq6Jb63W&haMlCtj&`P*{qxg=oLYrx|R>!t3vn=%Q zp6w2Nv|s_~cp7*>x~}od!{{GpXW|e+TCPu2B!aG6(H} z8k%J?;W~3Ke2z?{n(tiKl8U;w(?ppC6gZlz3}~(&bL{-u#62I1;Q6zHP=CmARZ_eM zmXlZ_n=~jAdL2+4QiJ{MSDp`FPw#pj_xH)y>U1flhQLxf{qD{G(=)cQUD*{gvX@cN z;YE~#!!|{fS3k~e#9P-YzM=zHVo)HH0*P}mL;NHscT5TVBcz5R#Yi@rQCK-7I z=W?(wA1`Icj1agBvwBFhMFhQAm4hgV`U1e8 z8hp&Ay!^_VjIi@PBENL*~n0WmP|9W2J5>iIvyQmfNb z=FigjD;=jUnK@Jrod5LR|L2c8n*g8YY#Q=;Xn?3lv+ynx=)*cc9E0BOpLRl+O>9o{ zFC#Kn)`oNTULq&>T@T)W0~Zo=#Wj>=JB1ux0FpFW@ehD?_b2AZKV9r^e|N`GG@{=` ziCG8lF3{)P8_OOarB@Rjl_DbI6PYQ>k_!58UCU4Se@4jv>6`!KM?yb<%1h?gEh+c# z`q+0P!*2Do`yylGUO@A^2B4HwE~TC!&PG5=AH7e*tGKCCdC$D>Ypjq8b@^7aW}1wj z(WROvEr-KWXDqh@>6g4^k1(M9_eZ*f+r7sDc(J@TdQpSebF}evjYoB%O0f;4mA?9A zHRbx#BVaHS?}*qb14>?lbW2@fh>^2=z+$d>_Urlnkj$$4#(||S$hTQ3=aSj~{B&=n zzsk6Qc@NRg!Pk5ctJI;y=oB30-RQi2&!y;f{v2F}j8QTg=5_SrdO89?ZrRL3Z^as6 zz(d2cqHTsb1z0o+78e*ynv6Uf>E}~fdiv#_HZ+YA*tXRNb>)8}CHkl7`fne}BZ1gT zrch5WZiZ*1$SCPp5kTh{P?I_b6v|XC z{I8l%@2&720!5SYOfv+X{AL}g78UaO(vy=I|27hc+V-Q!2SB?*; z(FYpvy?4j89rb6$m*cgK9wF=sQQLLz4TdZtENZ)&N*fYi`1?YneV&ikZO!H>Y|pbd zf7pGiyVt|S?>TV00nh>*AE#HL=HuO(sfhuWZsS)80iaYLTdojJ9x{>ck9=M0Xx?%E z-9QqQE74hTVr|p?0_#I7=2>$lX`R5l749w$ryfjEYN#^*_i+3h8Nt8qW}gK76FWS- zN!4*#?p{f>{*qfx+~Y-9mqwj`VY zBy=r5&5|dY-XKh&AkS4l*;QCS?d8A)p>_ai00m_@&eiZWdh3s`yyhi6?!D~{c?TQK zS5Gv4H&mw{aoCne&sU?+`Q>I0Z{RC%0YJCTqnI?-Ch(}QE_i7o4fWTldmCi3!aEP@ zo-W|sd}H|qYT6UYb1C9vyR*QLoHWj}JwL?_c2TDo5z;gLc+tZZk}2(!Pm$gMlf1ZN zhbP`x{LwedR`R2YID zY2+%7D5npGQSn&eExT>c$CGS8rQXenvVKL4)fR`-3Mt33sc!6stBmu|<64NOc1Yb> zdKXgYwmr=XS50D06}C$dLQeD^08+P5iEVd!)|w!4t06!Db+kFO@w15a+*C-z%a4aq z7?ecvNk1R|Myg%7lULs@&U?*7P!21x>?iSDml`95fHa<<^+@Me4~#`pV|=17lE-qk zDBg!uv>UDAc5o?J1hhGr567};)Hz!wvOqIs_}&o8i;ASdE(LWZDuzzBM3z5A*51&Z zbnUl+(cKfYt~o8!1y2l+;9^K^W|dDISMYdu2nUJ_pbWOhKCkO#0L5V^J{=#Af9MbX z$Jq9p(U(x>G6c%h(Tm~XF`qGhd+QLAPQeyR<+Lv4uxB=1t1GRGitpZw=@+#M2V+ z?ubUQi7S}2f4ZI*Y%6^q5K4Sqt~cJIy?OJmY|8&0n$h;XScL>mNHPG{AT@5p z9Ip#lm;v&PfZzgws^P{MK=+)QCWp=uhICj@d|xc78olh3BtuRDjjF!fn4@hG)Ow*t zehm^h?ZUUmORdcfBP&h2VkemV9%L3>x$G^k@?Dl^wQ{pOAWQ~~1XAY7XH1)Qq697h z&rIj_r3XYVr|EjN;aT%OR|t`SqBqMOXB)tA#<5V^4VCs`wgvXJxnRjnWo~;!QDA(W z$9L~5r{hT<)o2UQlIj|#pt~G#MOq<>Ar3%J6#*@wEr%GzSkcOZ!^*CO!Gg-U<9$>#VjFn4i}1hH>DS>C_WpBOkMqx=|96~W%w!{bplY6z%^@4z|OwP_#pGvA=tR z@1g&tfOHO-j5ABcyDdFhQTGZ?aPD0@H-M23Gz^%am3}D*k|OL04s=a?T_oj(BN9 za`ZEB<0jpz+y#7-!wH*NvmR)AmFFN`gX2o*rQ}bF-$9JGh-=BQlRMfiBYHwV+xx?n z+X7Z!9iq@8@_?M%(bT;f>&^xw1fAir1<3N%6(diM7$nSUZ8eS&nSN&D8{>|@)?Oxy z`Zhg{q@P}H&{h0@$?59wB)R1~j_Bzl@R}&y*Ygr{bw~5_vDkAA|GT+*W}BPr&)+99 zNIQ0_c`3_Ae|UCRL8kbIg34>u11F$`N{rR-;vB>ou-_o?K;dZC4J8_8I6}i`=-7+t zyQh?AoD$I0R)v0 zPP|c7@@Wu2VQQpYo$2JHATxI?^}=GCcOmyXTr^UKh-d@`i=e#wb}y0iaaTKO=Tp5~ z5AZ11p5o!)>py&9SxNXZ_ujo+xkiawXEH-%@9bRd(6-z4tTK>EYj9T zx`lP4&v_{sb9m}Q7U+2u<)3zsWL^g>{5aCtSlKkP^u~)c)~ovGCgSp+$H>s|-k|Gk z{|R9K4*kC4MU&K_kz?$t8Jk+VS?8fvpqx?e4ipVOor6y>TwhCM9%=gqtn8 z(G}L1wSr9stW!=cOjWn%s})=~rzor&G`DSi5$PKBFIYgfwR}&RikJ7@ff;hO6 zF2sF2)4Z;`k9}L})s;cw zGs#aT3D*bq25c$`;X(s!dO>qD*Gl^`=m**d}4|+u$(DLp35~h z?TH}Xh?{bntFuYHQ$Ee$_5)CK3Ftm}7r7=Rz-qnd-Q?T!=dEk+l>#1txrb&oPl>*A z8tRpSSLr~tnyi`PHhMo5L37^KV}+XiJQ(L=`*oQ&A0Hg=tJ6HedYyZoeCQlJe>ynO%t1Noj=M`iH`-2-RcRv`s~ec z?^jx7dMm4dnBYCgP#Wiu)J4*f5+If`P6NGgw0h}#JO4LKYK-0w= zijbGbTmmMKr|iBL%~LhCmkBGG%=tY-Bkz}yj8a_sLJy3ThkD_ZUXg%|-A!VQ{+*LS zk-hOr<9AI&C?{~Dg&`}bab ze2{wP7|0`&G8UyC^8s%eQ5_kRg-az03mkU~s;5`|xOMF0giaMOJj(ud> z9U{@Jbrq$xb+*q3`Rc{wE?W{>k6+)cJX2>wLUG8zZ4qd63UjaYcFRnMR(?*v1;~GY4R>!kXqvQ4pSS=NT6;JaVs7{k)>;w zYCJnVTxX4-6AY;Z$BFg4eHPaD5qrM!IYsCy!fyPR%>}RtCtM2KF_*d;gVfS>Fy=Ua z(y)s8qwL;wgDijOMXc$u&>fLR6XI2<&qxQyiVla*!yESUg!JOhidZ9vQQZgF2A46S zKUDKKSDmq>%N+7DGp!C#3aLWwp05}goEx@Qr zM#3MNe8vG~s)9WVOn6W;9PE}vTsdja=l{5R{%~UbPiIM*AK>mf1hIzF{>@P`4gd0E zKv*Zsp9zGB7}OZ~3fALlQm&Y>!HrL?CCKP}A z)&A|9WW#{~ew#S|zcZ5kZyu{1um&azY`^>G>?i;DXa2VVf7?gY&sqN4xAb2Q!6yo4 zcgAZ$&;H`TJp$gg>4T&vzv-m_7d(|1R|ZUFiSm#Q3il(*N%0 ze{o~_-yQug4&49e9X*==T`VVFtkJ?$rRg55^NYVfU}4Ow$)!y?pde7`d1$5EWTupb zGzU&r4UVff1~NCkY`0t(MKVZnv4zqk0`6e!>!<>?RCde}D&G5KPYEa~Wiw(5GW`+& z;6VWfhd&a7cF0JN{#I)C9~Pzs1W1*h8g#HKME(Hq05(7mGnPf}^V4v0-?&RC0T8@5 z{4NfF8H*C><|-E_2&VOk9EFANqq$0~Ir5QTff6x6LYK(yk;L{7yU_xIMi-y;yiq#I z?cwTzDG#Z$g+KsUX98^30r^#Vy0PNKtOXCa?|COGv$N)xXjCUX!R()m9za@^)&OKV zIZeVUiM;XG^%}hP#tN&<`>9X1As2HAv_kH_gX7v@a>s-Ry4mzw?x3p#q(5>wUzxSL z!F#j2!d3tqDBEX%8D8F-wgAVAW^|v+q(NNPBUTsbBC5vO8aqC~CEd}`(|MaWh?pvL zOlmq|R_x;Ycmt0PgX(OfP3oDnsl1@!5K?;mkcAN3{cS5+>oJ<^n|>xSUp;e87;yUM zn&7AY^N+>6K*Q{gnhbep(*`5Zt0B}&WpjEbmNnjcS@Z-->nT6T_k_XW`_QwjP-q^uJ<37MMGyd8k63lW5ZiUlHiO(#yAL&C(0&kozT?CH`3t}WUFphTzhCYxng2$zaUc59_|y%rCF4PqUAS> zj@+7QC_<=UPgT0#A2&MW$UpX>yl}dA~A%OEvNC zwv+)2D}mRL%L|AN0K%fIRBC2NHG6a91KbHQ7zg+@hr+mnh-nhCpM@%>2wGC|4{##E z%1qa$nmnp=RT7~d(~+g8*aNsCt&%gWEpsrtqjf@n zt3U&u+>*R2Brz0t`sKQ731E2W1{?4LOxJBsC-iN<)D|`e)9~uuLr#>&ou6if*;t_8 z0c-v3;UTsAq+oO7SDshhJXipgf;FV*q6`{xpYATj^5_fO#km^~_BL&fGCKbH!P0e* z0ZrsHF4@{N0C3U`=`NNsOq*~Gx*s)sa-JV_1O-5SLyIWM6)2fkl{R`mUg z5^L9H)KFQX50`9x-K+B5epq8yqt?V^0CK3tj*0sU#11ZPIa!%=2Nf}C6c^x`#vtM1 z_|4@)y2_e-7NzbbdQamLi}V)vuJ_Kf4JKtE+Hl?NqERWmV;k*Ssvq18-!vijs{eKZ zR1F4IQ~O@c3^5v1zC+xhlLTRQ0Wz1NC>+9y$*|9n{`xjE<;~*W&sD=$#OY*XTX4FE zvhq>1H=UcPDyA!+0OJ=7rj9&RuVLp-Y=qe^V$ZM?rtGyhLS1_r1n zN}BA?o`^`WVs1*R2lT(y^l7FpyITX|{Z>9-Pd)ei`Py7SrCsxRU9r^%9PMI4AMx3j zB&+(jCj$_!A0W?D*(Z(N45#E~$~A{MfI`*8v@rEXRC&a3mM0%ch36n{)B3n=sg_2a z@+7qx$O_2^*?E%HS2sCnRJv4$|MC2rOYsZaSskHmZK3aJcSCnBL~`VnWJpdtdCIS0 zd2W3eFW72&AbZwd4pGh&nO7WhIqX)2PKpS*qk^EN8cl9|o*4RKDx~q&^k^y-u^u3C z9Df@9$wT-1OaRg~lBzF}?0`5u3Fy6dMh3PhR#cP`QVY_Hnu99>s6g{*#HI--fkr$peXsUpIp3t1ieo8Oa zWCjJ_6)ud|o$a7IVzivj4ssKJT&^_?lOA9EsHF#b*Q4`->6&gC)O8S86NvVpkynFU zTHhB=r{w6gc;M|=2=*w|z-v&*)?>|Y^L$lm?WH~w=@;%S#zvm~-`lQa1#zjxWh5op zm#GGY4<7#=r0vDCot$H6+rPIgi6qdgfB%yCwxM)nw9ONB=eA70MpdAGoA^`J@WPi) z9A4TsW>M4b^gc=G<<*Q|^3*R%^%|;(dr(uA9r$gm_(P>cM~fZ7Fh=B95gSe5*EpY{M_L-6Pcg3L;D z@6-jU3{<-eA(xi1)pXceADtH>8W^9|=j;2cskXHYXgM-TWV&1Pp!E7ci>nNl0o5n)zqfb373 z#cM5^1v?m+&y!_GMqT5(!P%QNURJ7`boPwIddRqb3kW7JP|$i%bvs61o-`x9gBD;S86T{(Y|>UfW75 z%JBwN$E}~mhqVJwjMkc_c|i3=E5Lr{V)y~VDt%$lLNBI~kKGF+HZ<1E`RupfS`*Y= zO4OohArkx(T}bQtGjx+$oG)K(eHKO z3JC6+I#nR5Q@#C&`X?x>S@Zs-as4M>{s8=wQst%~=>AGSqPv0!0TfbP;l}yva<`gK z4;;ueorihh=4IeTUN@9w^_b6QGM zMfm#M1yqT|R1e%^SsCGT?O^Jr+KGzCfy=}_L7UrZsvsLd6L-k9GLSQQOO^`c%8TbQ!fpbHWoULo(a# zP4lO?C;MP!DD~@%?>IWdQ+i3qeH-J=yk(7z_z9)D$A^Hw2T4iw^4frX&`8lbHbFw0*F!l`^;IU2 z!SyY%0cJW_>G_(7an0>yu2Wh0`Xfh>ZIutkwep3RT`6PcGwy6mYC7aZkLp}WPimgA zwF2$^zIp>ScxZA1vj^HceW9XXx_2?)G}J(jW!hiuBFgaUd1G=&19o<3nZHea1|kq7 zWt3cFtgj$=y4mvGDK0rc7a$R`MdY0cjz5mHH6 z%bGn&iBQVEC+lGB+ZYp)B1?pjExR$sKE^IdcG<={$TApa?86K*_v!b&pYMGi&+j>| z-}Qa|<2ab*bDrnt^M1eHuSGHEMbLvx*g;;;NLjuCZM{Uz7P-0w1gsz2`nMXm{R zAf|z!Kr1dAs$A;v=tf-g5=6N37Mb|0O6Fr3GK#dFKn}mBZhrwin{H?cvPV!W~x)9UF zTj%O--639f8`0XO+Go2FspWV_Vsak$te#`N7Cq(@A8T2;u=D;@YB&z)Ko;`nDY4;y zP|g~Z%wXy`!zh9V^wQVdI59#u{7KcWK^j-sGRL4+y7V)n*QWJ+n>C6{z-ocS7mRCT zN3MGxNJ(PwVgu}T8J;%2g^h?l1d;&{5TaLQDY~kzSE~ZL*91~D*-pH5Yg77XRpq~- zov(Z+gfoN&q3d-%gVYoU^Tb(u)YLVdY6VaIs^B9^_rmQL=9*PT{TcB$by82VfJX8V zss6w@y+~NmehN#Z%f}!=7yczW_}z+=_wo-g9q3SVB*o z>VNUWb~!ycya!0B6Xu+-LVMnu}8172@z&BA8jvwYNYI;cYz#7mPrW}X5u~3dqgny?pVb>mWR52IK)yFe~?Co z_7d13V1J^(B<2Tyu)77ak-TaQv?79cEKYFs6Pf}4Q> zBr4H(;6Q+(UPwY44JC<`QL_p^qUB{pH*)|n3je--mD}l_a_UmJ8$d2T4ZIPF?3kJ+!YCC-mO-2C zgKJA^i`k5PKnHf!e7m?W!px!jko3Ljh!R$}J|Il!_gT5dOg)rxME8!>+I4^av~Ic* z1n^ACM<3U>V0lJGvv}8!7L?zfUsw5q$MJ>*JSzUeH*={PG!lOz>=gPwmBcMg;(K4t zs3!|LH*Fe5-ks-pxBJibs((*m-BTF=j^75|+C;UsDasX3mtdGzzg^iMX>dbj+&*O6 zPev0YFwQq!_r!^Xa5FzZ?RTuo!NWBXeSChaFGx<#RTyDb9XTvEo@5qR?Y2I#)ox_f zDUG=A#?c}+?ArPg1SDZ5#0zRYtfm37{kmYhjQ15Qvu2MR{40zo0IRx9?12b|uE)%G z4^|tDs`>JiY*%53EyzwQImy>`qIPAU%5Kz7+^~iEg!^Wl{w5Pax zhBp7JJKfFI7%53GYj9 za}}pLW9JkO*Xy|&K%k>Fm4K&k#H73_TpwLw+)juW)Yi(FLermEPqqwbXXt0A9JMn4 zoD5?PJ9YN`D&;8?3suOj+y76a9ovg9M!>mtL8&47c}}2DN^a(Yb*B9+L%-0|uggCx zy0#OmwUk-@7A^VzKuBDGQTdr#6~oz)&tS&JOj49M~BhDGZ)dpMlL7*zS#* z-P@gG zx|T$!&wJJyTFxp|3UW$6B895DH#%fa*egG*!c{y$=UYm|M zl5klndoqwHHtCkvBzMSq<}$p=P#QIRusQ;tPKM5;Nct4>e6vhfc$K*I56|)c^K95N z|F{9}Uz=B2f9xPI$l9-T{TS(+BV-_(xE)%?S^X(7yyCA{c!Pj^+lI~-c(}xISq8U1 zEQx_tzeCPD0|DbK+(Trth8JX)$Z!7A9o!fXL3;}z%Qw$_ zykZOJiX{zrOPK1y2gL4K)*B)UBT7mWaUCCdMLJ@5y-w`$IFFvKM;N$R0-`Z9qoYWX zQ1Z(n{q|^^Y&+&G+@NsJoM~31-T+F88{uSZ??5`qItJmYoF@d zzL=~%T4FK{Y5Q4DI?7OnOpzJ^aPFX9GWy?_=HEy9-yS>;g@4)E=hWWeEDkEw=1KwL z*DnriIFO7i&%-NR1XczlV5gOuvK{(T1tlCuyqSGeHpgr|kE$_0x%+uk%jZV+W9Gb( zsAAk`aHjl=VDgHPoNLIE_sSr8#gS?&;K^KVil0Uyja!teQmjyX%0G%n&EjOG*H*(v z5UH(_H{y%~SLxPNjJcDPb7 zC(A{-NC*E8MgTn3Q_#r!+3Ywl=@$^?``rzUYc>lT5gHQJBq(3q1X#-zCj9SDV?-Bb zs!XKNPnv(m@NkuCe@nI-b>8!(jyCbiXa#59y}o8lgzaY^bDl{@Jzm>>RU~r@gp1p{ z9Lw*?}XvsfSQ;5Ba}Lt5%}jr7Sm`=2@px?x%feN8NmI$PM9*cO@^KHbum05 z|L)*VK5N=JE^HkEI%ZtTRRuzUNq~9QVfJIbPXVU86ZJoRiO1Ng@%ym}OXX2uRPdH( z^`oVs0o>Xw?T*K{AnIar$o(p)aLZR!Y8=qHUV!x9FsfXgkJ!y-1A%u4>!s1k&#c#h z0bJ52E0`FRvg@@CoQF}DT>3s;j|kfM{df@8@}@<;WC--yW=BP+1_helnv0k=#=^t_ z97oyz)wTp){TAopUg;*8h3ChEvAw_Eh9mT9dc7Q^{Wg{{epn$I0>h9f%Fla_;;Vj>%T45XX zg}xVX3?t1j(XNwEQ>2w%^@cy3$EoA6n?)ooM~*O3{h z1!yPdzVA|*#x6_Qe|DSb|EV--YKzbZx6a9Mi=!BCu5~|{nnZ#9PAoHv*WRC~b`wK_ zZcWLg>vbgwsGnLX#1z%PJ6xX&b;yA9g*Qmqb=^?(T|2!d3`mP#VWjozfgN#*o~cch zX1>S-996;JXFMWuFo&5{x(77{ z9lSn#d*&722iKKdwFt{V3Dj@|A5b}5lkmeY{jBtw4tQlHTN4g^KG!r+Hp#I9ww=}S z7kt0w9an7`rjv@DdpBNQS;HUxvFvXjm`#*gkdH?Hy z)YXXMOtg}I|8-!rNr$H#=(Xv(&bXRXynZrVJ{GJ6l3CEiou4VcxpR+wa3-{eQwcS? zaYOhjU|VZoiN{*JYoay5f4JCqNO)%eTxawmQ1!-4x<&ar+Eez-z@2e^Gl}K%9rdn= z3A7A+Vd`Mq!NOhk&tF1@#=G9|Ah$#$BCyUN2S7$S?Mo4A6bP|S#4t;~$)o@CuDp!3 zk|zAEp8@`|DF?O~YIbYFlrV*!R<^2jw>dB2pK+Ga5~a)AS%SgS%VYriJLFCaz|1b3 zDOKZPdFwD77nti?ZW$E}C2O*=80e)+3f}Uz%nk}ThNs6swWEi9iqq((qHGBB)Fjm2 zic_+={P}gKqLI!S!<<}4U*g`2Zog{%CI%{ksae(AjboYENFh8_E<3-zIUdFU-Mw@Q z53ya9Xub!Q@OD!Rg0Rix-Q44ggpVMN`T@AOpXNOJaew86M|7*gKSPE7T@Bt-rsFy+ z5R><9w4Z+@=c2~)EyG2-t6X=_c#OJsuI%d0sYXUNe)(&LAI3xHKMCKBP`u7OVyZy| zv}js@SI-UYrJlLE_WGU&oxWEfq;+=YbGVV!aYKzGb0pz;IID8_m;U9>1;!UlIINGndxyRFGMQ(ZmE8xakMSiXR|Jr^| zp7{!=?LFI6T9%aZDm`r+Zcov2(N4G-4L_GTw%)j2Vc#pf=Tj;RpAD{SYkOkjf_q-` zH07a1{@QrO2WWa^5@2p{8`x4l-{Vk>Dr19R<2MpKIGu(DBIBRBBg3V$V0FZ_+L#2N z|Fhfuqo2;1&3+$?G7kN(y5;^Lld;{QRq)XWOW;tn+|BjbkN*kj{kYQU6bBR#uZ8Rx zfp>2?xLpRghg?<+Zn&c0|Lmqt$~D>%DU)&|^breBM&{M8GXL3j;7|^5UMlzhcq|{k zc#7x-K*^B|`tiO+AYUiVFtA}O(%k#EMu^?hzb(1``?9o*3*6s0T&vZ+B1m`oM#vk6 z3}m5UZ}ewn$li@hp9KC~sNMe)l?O~y0h)mDZ9wBZ<%>teE1T8S%G&V$XJ6s;*KA-% z>mOilvk`c6e~=j%r zTxU`D*neM{e^AY~$X^8(?@I0auXO+6QvB;G#eD@9B;iGJ`9FBX{yW`0#OppFl9P8o zFVFBFuG+s==c@cMxKJQ&`yXON|9fF@Z-6_}YxbHa)BmrvyYI<#PDxMK;^(7(@2daw zq8-P;1m^bjzm~LRT;ljqY<~CkKN1wRn21{2Y`5Ar4GkP@wxIol zv4^vF+4D5oCde9pO;^tSmwsk;nE9NN>h99o`G2rD|JR=g-SO`^$&#*j*3mf83pt1u z+Zhc&zU`-f`LBKE>nq?Y6kPf`E#R)gbO`jZo~;aAUxx}ZzIZ+Dk;O@06N>nY?)e!& z&SYG-yDVn=6DpK!RCJ@?+ffkcr#?EHDWuoaO>yA})r&oUe;xZ=Pr#O^zAuS!kNzGA zD9=i$$wFiLzvF6Ds`Lg<75Ljv5|k?pnzguo`9+d~2y3iuUN}?Fo2S3xkN2Eik)8t$ zu{&O)qT_&2^L2NUPO)Lpg)tN> z`h3@MytC&<_#EMgVz+vz5zD0cnqCrzF8dXk+~Sc}d%)r$zc`&(JVjo67!uPBF_XyYd*M8R;W) z1{=4Mu_4mrI?L*a_Mk204;=Y}Zfx|Y-j}Pubj^BLw{zN~+_KV5Ik0y6W!+lELxz6X z{z^ehHO*yz10ag#ZXyQJ(^9~Vkl@rz6+Y&(btMW5l2#;D=uQ{jU+PK8eF`K;1gs)Y z#3CQX=sw#BT9C`Jqr{|#tn-W_6JBR~@pxxaC71bW8*^S9Admb14_0qRg)B)lbFZWb{^IF5u2`70$!2mp#CMwHG$wA zjlbVwXU)%RYA#PHIxuJUTh{&gf!VrI6xd`#VGY;}>u<35H6$koe9Xh{&LIH*J@uu| z`qP+yK*ddq+40a#>O5$uF zmxgi}nOfuiVdr3L(m!@NlvzAZJBl-D4<(a$;h_KPtFleLd%S&tAM1xbwT-as_B!%YtRZ|i-{`FL?pNW7GZZd+7f`;rgm#07v+Xa zE|0{l^B+xAJENNnPc*rtNUBXnb4xwqmUdffiWkYcRHbfz?q2`NR9UA>-an=6dq3%< z)tnTIlCJcf4opEd-=df$@Zc#92~cuIJu$AtM6&+qSQ*O)3RqGaV4vyq!~ZrTG|_&y zV1W7d{N&r+=S*LFNgAIEH0Az>L%8p9$d9*u>UzO@gNxZwy4{C~cG?VkX19!3JE4Dw z<00Wse~7%XX?R#r(v|}~)kT}|^{{e1)!^5)m-dC%WOGP03HiC+a{YSS&BHHW1p3BZ zv6qzv+>;Os$4*g(b#jK^ywAwj-^-mPL!E+`5Dk*R& z;LWlXZRj(Dve$WE`7xFGxp_CKtdZfr~EyFQWRTik_# z!GV2O{gG^Pg<^Kh9C6X)&N-iA!^p(2GX$u8zul8e-L7Ty4~!(JIoAajZ=+xLHKu%z zPw7^@45Pj4OtyN#{JB@_mGMouq0C#}khQV9<8_eOlIaSWXGJqfEQh{dgXtbTFM5{y ztYp2vg8}H?Sj{`OHxCmiU(aL+q3+CIfAa=@I#^uFScCNTZJ%<~0uz7W{C4DDbaM=A z?3Y%Di*L`w@bNEO4?Sl%{{MbFrcq6Nlo6!sbF6(f_cGXh(P*~ErIwX}Wh71cn7yqo zaJx%tFRHZH`ycA38;@pRtK+Cp(_cGJ9l0$1v&{{fdQ#kP2|>?tW)Zjd)rFLGA28la zKFc0R*on2u+7*-2TA@1#Rp1+V6-ECjI$=&#$EQ2i>36%4>Vdureo(xSU+VOY2%vmh zcxKmAMt5vQepu~Y1h{~y`L0h00@f&_t66q{>!FejW%9cL+^Bo0H%$;q-YG?jQGiq4 z4#3JQyvHRw(?ya$-?B8X&@fT<1~Bl7*?v{K?j*5zIk4qzgW}JMtUM&|nqyMvcEw*_ z00-$iU!-!1^b5*S0n$lE+N8-GeZM(i8`rG{o#+xp(4k}$AA3yzHc$)t1(A#0+V9Q^ zS#kKaM{-oS9R>PX0QMY0K(>6+;OeEK6tsm1C{es#^j22NFvc$dD*`t}c#_66xVw7N z6;YD~44dCveKW<)n#`$Wyu_14h3bpqFK_j2S5hl+JsxdbBax)v?>Za3^bnNVs#hAQ zMhhh164UM6NNn7re7yXJ7uL?iv2Xzz=sN#1`gqKPX~(MXQQyOl0PC@O@S3vGw}&5U z>nd17wjX6VCA?E>-3C|@$v!Y-^5MbOBk6s}(Ww+gu-X9mW;D;qUW&sWVN$4j_jbVqjm^E;Xe$4OQ!9 z+-#E*9QsJuudh9dE3wLHaKFAim36l(Q7Ynfe{jR*rqqtr{W=U=me0o6UZfD`0#izg>f$*Js|SdTQPC=NC4Il zanerWA4cR#8R~b7HuJg-e+3>{W`VCBp@Ex9!t4>vRhJiIU!wl^XV=xMK!v*b%7K#` zs75?cK{eFI9@9m37{yD<;Wx~1dB$CiIJOvb&o1(!XTZuAmq-J9%NjT6tV^k=(S#2F z@8;jSR)@(hV>RlM1$rXwyI`!w4iPLgWKXT`Ia+ztw_hd)TAX@HfgrLHwj&+O?oWNc+c-Zcs%!@;wV+@SAtB` z#nGwzAz#NE+dh(HvT^5~T4p%Keg}=fDrnYE6StMgeydgh*eTojRRCJDfsz|7g;MN7 z>?PNN_x0#!(G{;b>uDvS!JgI9scDXrB_Biwn~VCb$|u(CQ>bE)Og>oMX;>S!+yQxr z4n!#ST_pP*POZAmMCm(O3Yb@tRz#sz0f<_A;R=VIt4X%z)lQY7Kq3oOIq)UZ>Tcar z;v(}dart-E5C4Oi>rSFzqmr`8#OdT_QR4+H8_N8JI&o$9)%+xh6SiGy%XGwN&`VP8 znu%ChX1GA*6(X0fe0alNW=;6wF~5)#tLiJ@sM^gy?)~}w7l1I&K3Jgo$SQB)k&6IN z!zX(?1fYMI(VzOYeAjj%j%X_?`WfQ4y=3aNBhF>W#M`#ShJP&7P&9cg41u-m3l(Yu zl7NL3q0FCQPehsbRlLY(QSb<#qUt&%521!n_s|YxAzW~ugBJIGkY_o7+3k1SGmq$#h`!6I{+Jl z{W}mUlX-!vi|HLdT;F{2Vd{_N@|jNCSfz94?OZggw*eE_`ZCYuQ_Q=`pVZW`Kw41~ z28LvfC`}Shlt=`qhA-S)UP}Cta4DDF68o~>)<2jJ#kItnft{{3TEPVG5Hr6>A8@UV z!!vs@{1r!H_3s)ed#mD@V)ZJqa$I##yH*kA)*)#&rk{^SYK@r-mRTVWR+h6m0In;Y zS*>k9EZ4cDVU?@9ax$*B0P)sdx@j@%J;vAjig*Bx9aKi!cfakJuL z?RtJokf+|-AY=5Q_@2~n;!u+x+pUKaen=H?cNv>1OGypmE$r%0TW>@LuBcAJ1(#6p zG~qGmkY7E2oIR!$W&14ru)<>eO?oid&hwZ%8UZUe}?yF>k$d`ZoP=APB%I2?c)MwBroE2W# z?=XasX3hDaWC6+Ths1~mM<(!WlM#)A};z( z=b`V>j;41*mON89qd>I2J0=Z#V6Z8(=B14*)Qy3R_#Sua^e_> zfNq2|6l71i_dZzFb8_?tTz^6a(BYW~eU!oMAvp~!{TU-Nm0#eJ5)K^Riqtl1ViVMs zArtX^H%p)crim_o6#ahA-fQ*nJF~dA+nguQP_en0RD_=N?9`L6p1tAc4qmwKh6v4Ro693|UjfX*8POv(>IXyKTu3YyYu~x006-k(S#r|e zZi;}Uv3?3l%aL6DK^uEd{7bhZV4UR5Y%14W5dJiS44UO)pp_FQo2^t1{of9Q{NabP z%KNRnu;zgkDt*Ms@IG}+W@G@o)}p%0F2rK|;GN3UIpZSsJut}_`lWToZ6tO`%Tc?F za?L%O0KMU|JQd@bqp~Q?^H|!TH zP0Q}7Ghx4k%ty{+%RSyTkh0DF)NNDbXnNowVG}k4Pa604;=R=e|o`2=YckF%{2J>KbR5UkO)=)Du=LE%< z+mwYReNx`!&Cc&8;`^iCI|8nsIe%?M^obPVx=C`eb!5*6dmIy}LQc)f;(1F-03q}} zx`ZFsvt5VAyc@CQ+CRHGn`TnX(MKeXOBgOs)wy8)qS-0KEB^SsvfN@t!1*jzxO9f;&s>_v#X&-;M}+QL`!5dLE&=o z2{(9y>qw4k32mgLNDb-nfmJ7hAaQq_^Ud1fT4m`g3hU8w9156ST0-VoxUQfEno=~J zXFhTZ4#8;C2C|ujIB7t(xtNA*x&Y_2%JZ~CjU$H&x^d1Nu4eUa={qZ!exGCd*74-| zC+Pi5X|_c-rn)W=AVr!R95e}$H56+w66F{SowbCcj1Sxq{El~`+KV&-31jR|4vbz` zDW~+Mro1zhB<+Eb_AVh_ezDFz0X(+Dzdj8DlV!OY1`;rK6sO2oCz$p$yygZ#*I|wyOHF z3=3m9Q*~oxk4PqDCwApmM-M=3xLnoE=enG)3Bw0=CP=;dpCh_^4QpzJI{?WC`7BJD z{1S;0uMR%@xp^03)kNOmxr9%CdzM^P164|E9&ZEn${ZeI2JNWrHU$p7WPs1ed(L;d z$Z~9LX?C{OU^;f#zAyEpDnn!zy2g&0=k-alYcOI^eDS_qIcu z*GCxp*03gg`W!9fo@ZW)K)N>TNm;a}>Z_6xmAjU8FF$1IO_W16K0P*je8_j<(wj4X z3x|Yk>$eD}npL*=J5~zgu%38ac(*!*mn{!6|HI<4viGEvT@&v8CjxQIFz+AMIo*@EL0MY983J+ zdPJ(U4aEiQzUV@DdQ10)?n?x3q^yneZ~JwxU6%S4EmPb{zv~uPEErWk>xJxZ{@_(R zof;5WF?018Qo$!bqCJ9v(^JKBX~R0vKOXE@^2qPXl_P6SUm+LYXBI>n6p~U6Ou)Xb zEJjX^52;^VJ~JdvvISa|=r=SA?gseVsv0Ac6yLcxuw_UjB{9R^&c?;e?X zC)@CMpG7PaW$JIDEd3Ivo^Nw^rpPCtjF-f%1pCiYueP$iJbCK9!>hT~YBgm~yAR{2 zquwbMkCu9&b?|=lPDq5f-^|VaDUT>)4nL$A(Em|pw6UHv{2OpbJ>G^`sYvrf1Y59s z&ls~Vj_5Ui7E|;)kkj@=MhEJh9qw8hWm*;bI!Q{X2xqE|&{;Ktua(*!d}j!YnVUPc z_u+!i0<})=q?p4^54zUQs2|a-hNmB1{FTG;HBHv(8r%q(M6jsWnqcRsa1Hb3e(F3` z_({~b_8B}zH%=pp>+ux_K(#Vx-d4M}2wv<9;#_ zh|!#6W+Ex?Nx~z`wLuM5$E;;Jme*#?EVt3&{7AP6O@>H|YY zD)!P9>B=U+67&-zvDpgyJ8Sb2OEf?wvg%3{!-uw2+V}gmRR@!uIstlZZuWG*7@$zC zX4)yYZkBJ%O&lOXXHOiDPR@=nZ30C)_v`axS6_es4)wzZ*B_E6!Gq;c%6&Uuy!{+} z)yQ0cZKZZv=8cEh4hjpOnd8}hs~ZVH24X7 zBvLImG>^j90rrPtT`r(MN7|9}tTBaNv3g+AKLU5YlvZTISY) zLN%x!a-0nDip^gspX0asB(Sr*9|z(LSF@pC9?u4G7V019-_B~TDzQMRv{-Yrm?>{+ zT4U47w1%lLe$kf@r}izZGid_xI5X>DLe5-5!kaKlS6lNvvi~&&jKQHP#%uw$h>`fS}3^aqxWAz5KGPh zi-y_#=JH-bt!km<0V9!R0QUG*@EKB!O{Z12d7=v#YYDQ1MJpZ7PEXK0$D6^;!9>av za)bPxXBv}yj{=K*gf1&OKP<7X1A>y8uh169e%3ddyf`|@dPAiIJo8-YArOWbGkU0+uC-|b8kd7NdIbIQkJ*Ntey0I6-|JNXMw3CH4tXyXN8`q<4~#BKpT(Wly7Cje$eA5a8*^)%JK=n zK((Lg`_d+CZ4; zNKtCEnPd$>9VO^zYi76D1gorYt;h$xwhpc^RB2u(!|As_-4u1F^u)2zhgdS-Awep_ zbL(&TNq)L0>eliZfuC(IdcAd2*1n+n!y)j0VAr1){Q#G z{(+w)u1?KH3WqeqPx+jcl6nF!l-OP5)>`Pkeq>htSyU*pZ~SKVJDYc(O8KGUa5+~G zxTNztVva6CP_O38m5Fs3WD;Zb>h2$QjqhtG-rwYQi{0^C`C+cjLt(MH-;`@Fiyigq z{49=86lXpaOLfg}$W?=Y3~wKTSXxxEH*T^g4ddo~b7=vZVV$ETThtnM`N*kN+m4G9 z?`YB*!5hk&?nilxdARE&hEk`29Xi(k#|t3MUapIo2+KG4>44TLSf#%})Z2f~S7IgN zU6_!Xc@I>2I1A6sG}yix?zMRKqrAB~)mJqTfyrU}CBxAbSV@>;gAm7$Y}e~kyr%AQ zcJfL+%Bs{gUY;RLQ!=7hCR2+m~5%DC#&% zRY*Jz&Cj?Fxw=T$4U+_1#${y)CQmlI82av_hjmt=B;JS09H)>(#$#`U=q1(Xf znuiP}LS+zbT)ktcr?Mi8$~q!SIsBA)X`Rk+z+X#L2{4My@h_hR4_gW`_nK9ntFq|$ zQT4-o#+$4j$FRIkWnxK(bi@IQ81zEC_S70MK5A$$w?u3;s(KTM1BWq+xyv7{NjF%~ znP+;7YfP7CCfHP|mumZuI_yKd0VUpdz%t~(=q^B55VoG7rM-PAhH?kh%K8TN4As;$ zJ;@=fO^be}MxHVgn{u5S=5e+Z0XH6)UOnzHpp=~#Vqw9z(~!;PV_`3zjN9WS00Jk1 zm3ymE4IF8JsTn|5KVV>9Zj(i7L!qzaN`344@QpV73~1v4S~|6HVU>HtqEIPE`F4;t z`$~bD4NU^g>5`$ai>cM^L_q$OM=Z8w(@z~F-?q@_*Q$F7WXJFjeLpFjrOdKB1?xrm zHwULm3p$>Lt5~6Ydz#mmly5#A@nVjSO?N+p)XCjZ6%T zP^Qvkc{K}P__sXG2Jkd_Hn%?zHp}Q8O2anVi`j2G@$fgn&TBrRN)oN zOiiR}UkZO!@jR~_BhU-w8BD~$L+cpc6*y;2fMR7!xh^W87P&xoN9-{p8+ElCyF2aL z^xN6&AJ9qj`Tb17-!OA~X>{3M;2`)f*Tjt|T<>+A(z*v)OR7|qrCH_`+}ygH0Mb_a z3Q%uR(^KV`$!2H^TSYnA(eljCN!bY_=9x*-NlKu@&Pl{mDtc<$nLl7}nk`IeJ%oiO z^ov3=(e^q|9Y6z)wB~Ld`0DZ@+rD12rN#+$USC>3wwIaPP`_-m#^65F$ONQi=CI?_ zOG{pGvpmoqHtXdl7|niP;LZ4b{=I0K5nS^aF}?XziEl|US1-l*s)BiUR<;vdg|~EE zuE|_JR@Kh76%0#5&F;(Hh4)zGdppsJvsRcEyQ3H?^=ih0MGJXx$%Y}Z!=g$U^a~-b z0n<9~BEw!62_P9FsfXLJs*A8${hj+yf3lpXd+DoirVxRpel}|Mjk0Se*a2IFo}=-N zxe!FjD&&FLD@51o?4_D&=yItd*k^7SK&KUNWv#k`5=*vMhB_=OS9_oE>Zjk?d<@`P z&WSn5`n}>TJ6YR~DRR+cTQQQvJzkuh+F`knuqLGARR~quDMt0-f>Vmr8H72wqrYNk zGku%~OAGkvGy*qB`qLh%kn=l`4F}EEJ9AksxT4-_{pIEWz3a3G@y)3wGYGXKIiur{ zfh*7llDy68C(TR0;~-4)ooRFpQs7=D`GDK0_er`Ptg-J^Tz@t@T|Md);=8Mt<|J~Y z8fcWSd&M<)v|&dz`{$uTpF^T}x`a1&JH_$QPO>jcJ7iKqO`*&`+2+#!MRWS^eb?+3 z$3T{UbcbTf>LUHCpm<#$J0ZtX9`|<%kY)4!JkJlz3|c;8PTUH}7fm0fgdx4`QR>yx zjBHOjg4Gwx0`zNYs+Pb}{J#EU1%qCNyYse|+AkO;oVa5w_ely4!%YylzENYHf9?RP z|9kLt)A4Di@ZN+5`pc__qxlK}WqY{PT@{;sEy2jHyW6kbX_)H8E=_V(vcd^}5d;OS zIN zS5Ie2$>dIFZG|_a#DFJJR#jR-E))G3vj+uBPAfU~{RswKnH%dadg_a8(yo6#PV{$h z>*WPC5)YB67zdej*#^C<#`WuiXf#j#J@@oZVSoCizQd0zgHIBilRbe>#i#gZM}!em zf?tuDfIsK3yf}7V@2V%$b?RPw64E;4*wSp+X=s5c1A59as#9T$>&p_@zW?;HjvfA4 z@0p8EdC$O#{u|%9@ZZCkDJK&fy6?3G=-91Ye;lijQywT`SXFqIvXEemz?Zx%d?pak z0EL*zKv&w`{gZ(q&M9L0?#}d?Sx7-Zgv_P<6A)#+j-QIE!)5P!9KJAZbwB+AJ09e^ zQheou_T38&z9~xp1~TIzhL0_Pamkqqu{f5cF~DzIn4%+eGM#UlGco-gIXzzu70LxK zR2;f6Zy}Naia6WI$__LytAKk~#9b)a`}ci#&63>og02Gq+yyacSy(&}{wyIPVp!_y z03?sr&(hCZeuyhbw&UPgZh+`&rKSQip}#JRfUqPq%I}- z(XQ6Z?ckfkfJD*pzbawj-n7NTj`{6|QSjIeK(6NHSJa9H>1f_Fh8y8b4_@tucK~U5 zQAzv07dEVr`r%tD%0Wjsp&fta7wyVmtkRMSdPB3aIY}G^)cU=QuHakaw!3k^UwwCH7QR^yfcbXMf2+cA!LJNLy9D*}j-e!sP;#Xc45 z;L)M3`H4d!Z#uTDHIMIrpvA>lr3ES5E5Lc!0^-=}k?CZlF^Bu+$0aTg@T5y$?B8>! z-R+nhmxGqam|-(s#mVrute`jMBAje17x#%atv}vavPED@u*HAic< zThvSz&%~mbHc|RS+4Q-v{95=B6Rvap?4!WgU+GeV(hlD<)SGOm5NzN~3da_7=F_Z6 z-I%#HxFGb~5^>?KEATwsRelEz=Di-7(rm*BqHgaTGB0$UciA-7!3>b*f$-;a7Cd?3 zI*Dws1OqShe?t~dG+l2ZMfx3_Zf-t;;0|We74kU~MGdxD7tWwll}p z$u?h+V~vg@D_xiKje!wC?MTD=3`z(8=Br$TOke}Je8>Z@oCkux6~jJI zOT~4&Ki^K!%Nk4JTec9(aRvfjDq6idZWb^4P?xQOKLILajp7|dVoTJQXRs7Wy9Ltb zd`@LinM5Uk-~}8-%RPh#Q&0{07lNyc4eWrRMbE>158AlOd`eE)HddYXuv{#YrDOiR zZ>^*6e6-{ig=k0qczF;IhzRz!KW5YUBHc+q1IOeZcl*3enbXPm#ukJeYU50XUaFEs z#`=CO#4@3POdSl2GEsRsYdG5{dKMbCN@zu7n&0@=1h=d;j+s$)kr`v`bmB1i{whP?d(pLCb_>|!6~`;7lw zruIU%LZo}7@Z$Pn~sFHdwXAD~%c5=d!VZX5- zX&w(D8k`UR7N6G~rK4tV?_~(l1Li9o% zL^TzUhQB=9aQ|$+eQ%sv@n=!aFd1>tVuq2_k&>I&3;HV9cl<4zNV-p)Dhv|1EQB-i zZN~9RiP#$D{2FD$x1a78QYjqd4U=9I=Idr<}FVBCR`q?QEO@-03p~&6utZ;HUlUzW-ou z+?URpDsg%N@MphX6n%Cd*hyR=-9A^MI0 zmUYo*D1T0v^$I@iozOE}}4N0ZB2ji2( z+RsSgqzwMNNb49WMDi%HZOiPa@)2td#er>sVvju7M9d#^k7C?`>0 z8ElymYTFdvMP?LP1beBprjq!&;$=W(?H7XT45cu$2deoA?O)Med4S_#zB7UE0JIX$ zZ(8J=F`N9B_ojM?nWnT#UW8<$nfQDZ78gcGooE$uo-bUav`gF?$;V%~|NXsdd;*;3MH61+mi8;o0R7bn?dIu)o8;0DKT{BMK7nv=3L4@h5f7D{$huuz zw=dpa=Gjg9j(XS)LVASZxW`7BNHX2#VnaHor=0+Hb^KX@NLTr8d17ir*H?i?9!V_! zH||~PxslGwH0L$!p4Oynyts}tG51XyL(cOxnc2t_H}mfVM3r2Y>iQHF{u)Kx7Axkm zPSi|@{~qwVe+^W`u^v{#jth zd~x{uzAx5BCkg1%kk2xip9)RuqBQjA#c)=QQ}rKVc{?skrXcOJ5=4)?S5yId_&kc|q$M)1Kw{7B1DiC0fN;eSroWAdgc)=WV{PhtC7)#P$|bh@ zrMmmatc~Z@7cETxep$|^Y7;sPH#6dvaXhzmF8x_RSvbI^vs0f?m?=L~Qz8ZZv(V{# z2O!mLt;7lJ)qV}=qujPlyQCwM0$Mho{ENL0b3Z66AEQ#7ztlR4@sV@Gy2xd#4y;r^ z5k>&kPWnJca{%e(PEml?v1I50n&+wnHHGb4iFj@~UEzo>kE?O78Mab8Fv!n^oIwG3oVdjk0}*q06p*sd0fW zruT@riP4Y6Vx#)SLmW!@Hm(L+FU2O6j!T*wD&Uhuo_fEH?Ful@#OJfKp z%RMo@_Z?9T<{=;co=(wec^{q*8gfSiiUE0fM4*cI+DRTw-v_JIIWGqVj(*^6{GQ!>j(6S$dQS-6aIEcb*SiP(m=3O=2yVp#$=PS}f?-Ojtl^oxl- z&d*dNLFc_ww;dCbuin@W($+WaiCti?im!v2q2|8_9~I#JEt^BV1I-tsem2#bWwa=Y5;Hn!g1Qc2#Dd-=noOz0f$}5?S(70G;zSqO?_OFH-1?SHD6gIjs^qCt zb$!w*vXyd2u=Mzr7XC$6V~};5Sa`QJ*)wbLl@#L9?rO~i z)G1sQw|Bx-f&azcdq*|3uG_;3HdI7KEFeu3qzFof&=e8rNR=+V6QzZY2!e_NBE2KM zgx*_FRHTL?1OlN*2|Yl75CY_TIp^$q?m73{dmr~dcl^e<|2X1UVXdsS-ts*2na`Y~ zw(KQ~N8-%OQ=h^G{6A4#!FTYvyM#-vztYMx)YGrDVb!N=B9L%OH)sn%d$fEH#@53N zc_gpn{FEUd5l|yWPT++aDki_nc9$9%m09%k zYzI>_OaoDA_nA}UD{N5toVN_p#tCB{=FRaOof+~6qNvzB$Aq(z^%PDlDI8U|wUI&| z`=I;*Anz;jrZ2GQiXG43v*Eq3u`Oo!6W6%|44_Vv$B>)NIf%bwRTcCXrDPC}b&oVofZsl%cuscD=;7&=Bj7lE-gs6Ix6W&h@BD}M+& zVwIAgZZh)1ci9NxmjQd$W-qRIqcuBV9S{_2LZm#~Pus%AjR1ZBG)%wA$1~Hf5xwG} zeC)HQ$+k&GXSVKEs8P!6w#hFYW&5p_UAtxs`5-t7>%9HsI7Rdv-BLb4)L-Mm*k!VAt*yDehgc*f*ys#y+4j9Ld;C=;iKKx4?I3(&mB4Y|ACcdnHRrcn?;-Rlc0JfPt(cMCGSu|(Swx>Fsi`;MKh?0DBK~#By`oN))MX39p9o4J)hVft^uqti1X{4j`ba&OOgSzT~4c>T;B@C{J#_NIMvAP<@f_mbx*orf!*XS!uaBS#3c@QJOZ|XeLn0mdt9=RBV~c7nv6T_S%4z(yN@0b^ILoG zRClj3KN_>?3Q^kEpA8fE6ZM1Vx-yhVwBYTc$@Sa>9WmyPLn>h2s$-WsA>J!K z&0Th{&{Wk z!;MABXtY|YB)1o}kZX}6F1;m^&4u0inZhgl_mug*eZlL+t<&R9Q1#a2O#Hyo>YD$k zt$qNxi%-5?n!RI25GUSUQPVJPSRJD_v`07^icv?DV+^-}Bv!DRzCSU#)`y zQ$O-%FhfOd!$7}!t7zd2_Kvo(ki?!3s%U*fQ|IPs7G*>EYbL9g5YoWDqp=iF~9yZ^dMPQ5gPW26mzj+m`D+{ ze6zbn!nqFs{B!u^MtT)(n5Sx*arGv=)`>=X^|{Suqk>pSvRDJD*WB?*m?y9ZSZ~Wo zMKkH6*VyWkPc@8~T@K*ZW+^Ai$p!vI>`C>WzbYhsnJi@*NtAk)m*5Vp!Cy!bC%F*v zZu46dmGSJ~n-+OWjGSF8wzo<7!9UEjzDo>MZaqHU+^8cL?s)0|NZ);fm=0s?7KpW$TsrNZTqcjr+_%G~QLUI+euQ5m%B%LIoWRmlw`!6xpzr?<){pmc@r_$!X;^Q9^WeN^%#T zwzk|+zX{{@toDU47!TUJEqzg7a7MXs+NB)YZ*51+1hw&t-ir0qa9b3wOnO9O8pl2= zd*at?w~1Cwkzx~R+|IVCT<<$~bS8epV0gaEPUizP1lESl{@heGlUVir9iQumg>M3K zZwjW7LW{l>V=xh0mTIqi6Yg1}`ycvT8$8x$XIC1F2DqUL!`IHSXK<9E zFPLn&-{*`F2pM%TqQu!nUN-h(Y=pPS2Pz-kHrs;k8?)i~Ps^Kt#?B4+WB!X2;lPKX zVPm17s&AXS7~w?O3=px%MsL0&U1bUCo4`#>_s??ClV$`KW7{fyks5r4+O=#eYR?ayWjoM577J`_m8x(HHrK4Cc# zQ+Kq``0CQO8Ke>PL1y>aX#>la*Si2v@tv>Q@!d{OX0oa8QgL&kPJz(%LG1>4R0+;$ zzPG6ZI*eo}J5*Nn0xX5uDK@{9b$*ZGvWgSS~z7uqMtmd3(l8}^r7iA5R z+mrmPCsl>hAFg}jSuq5RcGelT+Nec@8jVjTFc-zfn_M5m^_$G{8BX)k+OUjb zwF@7gW7u}v5;B_(*J$6p_klyr3BnmmJbHg5gJzshdSBw3t8?9|uNSrmvp$ib)+W2W z68-JnLWgO8yfP)4@|gduW|vWc*5Vxw*F@VQM^|pYYT~?siWy?;{i@ziI%f#Hn4HRtA@S7#q19TkLp*Nd`UCbw4%&5E3%Gt#ZWF8oLg?$K9|h1=F5amH|tFDbp8?av(h%#|I1?L}=p;1`(%KD|1vFlK3|aBZVx`Erj~lp>bCX;5)uY{E4Fs6iQxRP}xIbiw_6 zbjIEd&&s|umZ8gckYZH26q)Yt60?+YR5d!aC}KmgXJ?MJ)I8S_Ubf}>L~}>GNk06F zIa0S^-oC%bmjtxWMpWorsZ-RJ|3SpTcOJ zaZs*JfW!|U+jT?iY#gG_JZ>kXoI|;MHVyRvcZxct|Evs)6$MJj=?P`Id?!sDS{Pcp zAs7Pm^A!6T5|3b>M03&I_&_|Y-|`t9Q~dbGfSr)ATJ=pf(16LGoG9Ez-w$~#Jq~P6 zXsIv~ScTyR2P0jgw$JL-5oV`LxOx z?7z4&3o?nBZ?n}SYavWED4*r}vz@3XxDF#o+ak1SnYUol(l<*baZu{kSK`qhdMtsS zKKkQQjTC-R*O=GEgf-fnIF3NWIhl(CQe3(`z^th#-HUo0Co=G<(s8t6xM@+EUNPMM z6v{Wf^-)VbV*1SYO167LHlYk;=`*KN2jCak7R#xAZLHt5IpSHwqe3s zH5IVJ2)8NXYlLl-PwZZuTVgAoCyvh#pXtu^1zOa}8;kg`(ZN|Mf{P#_{Ak~5T})*E z#?Qf=cOHlbqTJ)p4D2}#vZDmrA%7?d(@e*rj?u~-T=i6@oD1r1fw@SG&+0G)xpDca zJ+2wjVM3ktxF-TvO&Q+Z(=YIZ+$w5)-*TiwdU-txeqY*a%cq*F#wy4zikA}_$3oE4 zyvdUv)3(d#+T0)B#(_|Y8o$J|F807kB;3>4E%CwT5_e#m!WZ5C+wNYPM&8>&c~$x~ zFlpN^+8Stqi2+*3@x0*GtEt%Zfay=kDN!^{%#N}|Rbf*?fUNUE=ap3TN*xovD2~`& zQS+juhHaQ1*06$}yiC0?gkSYXHM%3r|WiNPkm1SN3`t@gN9Zq*- z%$DmXHYQd1kt9^x+eo1}P-hB(`}XbQK)b9^*nLf(pwbt-ww)_ZxYj-x0Z&@5v7ld@}fnnokpx$ zYC-0ItdO3mLxp3(&T*ttU^3@c-)F5xzdUHh5{%M<|XdF?(`f$R34$VmSHvr5rb&L#N-{ z9WZ;h(6&9&Q7{O8<2iM~>i-?|T27woSiCC5|IOgB6m|LV?}>8LjnO z4UMO4)VPo?wfTlyosltUO69e1> zyhY?GNxa+7hGAMq4~$W>Ku?;2QmV{;XP$W%GFnbj$?)>qmBtL#b>;_trjRH1(XZmu z)VMs>=XssCSKrH&`)1qD?d}o09&+{DzWd{PbVS9GFkgQi)$oS163X7XifyCpZbN6( zf$`X&b(+l~-&a(H4|2|lhlic-IC5UpWh==+M6Ovi;jF6y{)-E>U2Asoo%rk2tT;*6 z#~L}RH;7&1s`$8)3fjkEUBc$&VLGW}miez!Z>!x$rV7+kOSeu#i`G1u=nDsBh}yFV zidyG~2`qX*pBd4?lcU`$^ez$k_~LEsM}H1AMHK-CyT(>SeEF+nwcf&OyBg$bp<%bQ%n7S?cR$uwW zcJEZ3LUqb*QwC316v2JvH%RfMvnXf-#+%@p+{u;Nfmn>~+F=}To{T{K5 zeR{7_&_P|K-8Tf^u(23-Y$Vm^=egdKGhh2&Ik4J$RWq|KaRc%JE=IR9idvs`s-*TxFX z54*&%@zi+Lt7wHUIy~BELGG5K9M7`KrEYgIVb{jSj8(LA&8Qdx`cwDIvi- zq4D=T7B284!LeyhM^$+5BQDw;-FY=C))G0;?-){UYb9Hdu(&yxKlf6FZbKbQXE&Z8 zS&NpSZgiM@b@^qtTO!qRR%<>g=YGI06}>pkOzbH3(V9f9$=xvK!)#}Be~R7KjcEo6 z7q5U-chqRDa~omYkKy(%%<}bha(UKI&-JEY;NcLXU4Z&-+~4uI;92Y~O|D1MvHEaN zd<@(fy)njPQuNlwx@4&0X-e!?CZ^mv)t6uXKq%*_DL(Vf+31hqj1b-fUJ5XGRH$>E zcaU4UHhU>fCSYSIX(YNvwPppl0L6KleXnwaSmewP6e4auh%@15yLR%yg2e<+syc4W zTOPM}P41*Zz%J2jH?&ELXxTBiytS*EBx+jq-sGg*h1-l=q@r)3`BUJ`XsyeNh6=Riu9GbW&3Hh5`eXU7e;j2oJ(9$QrI$j@ zHh2?^8?~rt#i3mES~-886cJYsuGL!b?GD=nhkBmI4b3~ksTcT+N1l*_GS!|}5%wQy19(+vL* z(K~!(do1?Q_a%xye*_b6&WdjB>4{0MH%Ou?U|CU}jv6NCuw4q*4&dzTDa|uGjT~}6 z-@c*v!0a~wXa5lKQhx@y|#xY0#(!4)>J$e5EO~bk#r^FA(2bHUzeIsq_lR1X8qDA zi+nGGcIK4(FJa(UiO=WT<^Dtz>i=hy{A0JS0b^xCxbyGslfV15-#c&{Fl%b5|H2w~ zaP@mY@0O;mMfopHl0#r;9;N)3XZ_E2^=F@0L0A+*Jf9FF{Mo(z&Lfd|2p$-#XWZ+*_V+TYN#4IV2>w2je)lL#CKzH75}kj{ z6#m_luy}M!{t3C^_ipK|1}KF+WU{*UTtz|HRAx7x!h=DD5;C;=i4 ztOHZ6;T^2y7#_W{hro&^WAF_Ayfrc}X_G z^A;my)ElZ!-ly_1vCYCG5xW{2coFLrL%{ac$TNoPm$^=-{nE+=mVtojYfp!dA=Q#Z zS^KTKqk!$$cffv|`H`o&m2jXM`sM=XHJU1Wj(7Beuh!glJb&Dbx95^9Hmpu%9}wGq_Y^&9{i4B`6IpAfh@_A z$cV%~Qkwr+-}QK5-?|Obsx0uuy5}l{n+sJZx1WPVw))wWgL;EvmSSI>c7YA|0dK{r zapq#4c6l}sA!~KQe}B)m9+LNBzmlrXzJ^pOG8O?1rXl3XFfB6%J^K{Jsi_FuKBn`| zFfNsDEsvX2IgPM~UgWD{Hyx!*yvOqCs57=bdV*vT^2_run(JU}B?>um(MYa7KNuHS zv{hTAT%8oT0q(U=ND$&5Jkn!+Uc!GTFUO}UhfQC>hyZhIi|kvO46!8d@09`n0YIzu z%Cw}l|GZp{VQ*>L)c^yJ9ayrQxFUgix>E<*7eO{3$p_>*jSJ)M$4&UTCh8Ozn{Jib z6Py-JSp5C+)SG-V?aw)Qk8SqHDQ*(GcC@*4FT(Yi3my3b$dcK0-rwF{!PwysmVw%A z+2^W!fCse9Ghsvqyf%K9Er;P+mz-%KtX`g(9H&N23G5tfdug~_xTKg?F_y~`BD71w zAP`5WAI^ zZi}dnGv+=!8@6p9Ak{Y5cQSMYUGiSF4aXTt{qxB#7^B~TpIhcl&o>o z)PYFF>R97#A3x85gOXAaX}T-jJE zzZ*DQ_Zb*hXX09!JD>Ge%2eLEAF8~aSI?7v@gyMRjMQkDzK}?UkeOA{XQ_{jHd|SI~5t@ z;w^Jd2KAr(7r;sDJCFE9ehnTI5D#0+%^%9wQNqiJ17&^sv{_ddhO0QZSaTu4sTB4LuwrZjk_fvuo$&o=3 z#Wk(pi?Id1w(n7OZY~cfn+gUN*_(?95y!q){%owGF!90Uya~d+Q7C- zhs-$d{sIxLTqAx}C2)W_VWAx@_{tE>(x(o>4kO#xRDTQ5Bkt}^RKAQJsrWmG32WUR z_sCd4JG=5)1?`45H}zZGaJl@Lkm+!7)Aqds-Y+J|rgDLbmIG}|uE(p|4wPlVONu6E z2mHymS`hpCR*CA>#}GoN{d5zBd-Y8M+komrBvkp8p6sOlONiRas0PAVGIcZgJ#Z#! z*B$p^WA@*84;cT$s$-U)TT1UJs?-@~&nMTbuwb$2~0i}3T0 zO45uJi@(Cg{Q;q9x4bh!(qMLfE#!xQ_H5;RQ^mXlT&rlgJYCi||KrH(N4pt>Je@>T z*db7@TE9=)+M+rReh_kgsn*4jBN#qX)fUR0G*+#~OfV*lZ{)NDE3lTo)9rHZ*1WKx z?XZy`oMl?JDhl)~s!LQ{Xz3IaPZh!m zU;Y!bg@1-JzkP~~Y^Kv&+eONSI^5La3!*(8HZM2aIN-cxL`jgM-x=E7<{| zeBIc?zN9y_KiK`oc#J#qzEx+yYC~l;}nk4MW|2S4fF~A zT|rsNJ+KzWZP~DKtLSD5Z}vOmtT~0b3j^&uQ!ixk;2rr<;?_3JjaH$<#B|XWo-N ziA&G!zhC11O@cA4ny4%aZbr4(l0&nr0g!&&D`mn2th>ta8)HyRel&wP>b2+k^m~-b zS5Kk3<=I4Eo@K?|z~4==e&@jTy8+xqKWlYvu57N<3-gSiwPuR6%dHA|JGzbPr@>mI2?gNNyM)ZJLTxEbR&RI3fnfkL%{^MRq z2P3cLy(((uT*Cr)G7x!(7#(L3h7m_c~5N)3))tC%4T?E{QXvOqIh*>|l5UI&UL)e~bzy{D(Yzi(+e z1H|?Vlv(t-gOA&}b#4j$SQ<}aDO06!H;LxQ7O6mZuR`3F`K`MhS~pW`jE}=)0p+v1 z;H7_uC0E0u&|v!t@bg=Zho|pfnccA~@az)m10AKQ#O_zYc?G%!RuFE{{v_D4O%HH0 zB{^pk5sS}kBfMj4MOqjoP+{kQ^(#={#1_?O0KI9O4xB)Thq3# zlq!qeB2mIEe~w!#^wp~fDO@9#Hez5E7nTfJwgby5ot;~rAz*^0K<}99LrMKf^}^Iq zAty231rJ)S_73z-jtv1l&TE>LQtGP`)p-lc)Q7wY8PRQ(wgZb#ZUNOp=bld2e81!R zss*o)+l}qnX>Z7MHjw&B0*_tt-`PJwj|jf<>p6n2)GpF$gK+mf z1-{jk$B=Bd=Hk-^dI*VU&UeJ}#exvc5^M8aI!Q%3M74imDLp~Q3Uz4TPxG)p*usQQ z;}8uxpJ@iK%J0#qVP}>*)!~YC2mY>CMGkN9@-^M0e}O{krgb0sRE2c?<;|3n%J9)A z{6V||q_mW8Zd1}W51w^#FlcJa!SF0~m14?LpZhcHD} z5k}cmx6xuu)Scq+aS#GG-wjOaeTd6JCYAPA*fr=pM#c-|z^@j!px?y5`FY2Y{z2Hv zj=oC0&HHOrWh5!ui)VhEzc!S}FLPr}ruF;>+h>hu_DSs)`N-q;dAvIIWF$0zGUcpC zi7G`mmls)nK-gi3MZl`7Y+$Q$9T0%(9W7UL2SYo2sNX|WxJN_m#z@+-+X3BTnNwfG zPP;SDN^MmMXaC&o94@z<$=BYZI=mBgXgMlc92gX$Ap3Pc-6i?i#@r(qT@Jbu!usVoh%G#dwCj<{5bM~^OQ9@|$|$5g1#$rx zY?PjrkKMKUx+xws^_F~x@nljAw=S>s?Fnzt;YXp0dhNyB2!Pd`^LzEx~%wF$9;)3ivB?$%P#(ljg=XvbJ`U{FUlwCE*m0e|4lXKH@6g*&yDMlH1sC1xC`KqIjzDAYy*x?enW8?6 z_)%$JQBZNIsfH0wUtDy6OabD;jsrxT?T(Q@_N%YW=3GbK!Gme}?0tHq3L5Kd*>Mb0 zSY1{FV{N1ncU`(U(zS!O0}OKNMN}+sy-)(Q1yBI4^?I95 zu>pZCJHr^YA7YGdAEN}#<``eXFbW70S>yo|A7IiEZ!R8`+kepV(m&g^o<&4Q2|J=M zJH*TVGTIe%sb(n<`?r9Q*`McN+gt(8^P6@9FSzXGesZLGezW#NNcmnnTIYLll|k&; z&!F)O?@dj7ST%FhU^(_@3B4iA?yXbyGO~@~(pzz^D}8bYfRHS(tRGt#!&t*Sow4q% z$i()UtfD3~WBAH*Cs|o6XdZjB!X?!q2ix-}@97Fdhs{u3Q+Ybe<)s$l=ff7yqBIG? z6(VxaZUwsW_o!Y{o!c#yX9fhaMv)2AWN|mHiH#c7jj>B!(q70m|B*4$z#MtC2mEV2AT15ddQ;ryZb#7Ngs1pbJGL#9YLI@>e12S0Vuy)~aT%mrTZ&TPYO7 z#_6b^Aa;L-f+)K&LOyG9sVkeJl;GAKRGKj&s9gWF_gein(D|q%dPTRL5-#>bK=p? zNq8-i!fHA~4}pTovj3$vKZzm~>(xUj zAzL$%a?m3}-0k3_baMXu<+am{C;4cS>gkkf57Rxf8~5I^KiP_llqt*s0DP zBy5@MxXR2nprtE7X`dh?SOd#R9kY@LZ{qAhAIx-nbcd5-%P@d~D(M#7UX0PQAxPa) zaxB-bU7J|R*DB%x?%`GVI>crS4{?!?;F8O_&KFD*E5+6}Z*Q0IHI)?DgfV|hpmR(Am5wj!NvVzs zN*7&W>p1tAFkj(Lz;JY_0R5?p$(ur%q;K}cuaC|1wO$14t!^xE_EMzjC#Ho`K^G+f zGVZv@{UN?6`Ao%3SE8=5(5Lyu zNxs6RlgmPHh&n?x=9cUjbxZR>zBZgIG?wQ*$Ysj;l2N!8nq^n(+9x<5T@*>9?6%w5 z`Qc1O`4<)!#G(>}J84zRwAe5&vcXDRYtEN$WpKdtq33KZ*1hiA@=|NmDdGTQSG=2e&90NXlo!H#O*JL4`6l2T8`6PeggC^==`Y_fshC#TEEw@g{;3P{kWcx z;(Z-Kl-LkbqZfTT@f@B|9F5;Be^y?LW%6ITbLIBaCnW>R&X{_)5Bq_~+74`L6tVg$ zCY1BNwg>ra_|C@r^i)4P$lW>XL8{~&5!UTQWo(Fi*Ioj-Lb(`c&b{Y%Wp>nu^c^~2ZLpDQXEQyj!l z%j}O4Fr2YLrG!b7PbBJyC*{ip%H;T9AGbZDdYFQR4ak^L zZ`wyMbHo1@Ybg~(K}V{(f{{Wz@bu+4LAYpBx5O0|Apfq>>1zCHen;QYzq&SDaHGg{ zYi=xX%>2ih3+PAFCG*!m5=>b&9gSMXfjE!zZg(XQ~5{O zej>$%=HF(^fBN4bU=Pg{84~~RH}fC*e1Xt6E zH`7RFlSHffzc^B;ZUUO!kuO%df9daKoDXcwuNKp{zWqyc^LiQ}EB(KF^@q_{X%ftT zv4j#3N5!LGZ_L{uJVaLz%Vjz~QtfcDoVrx-U@>o|<9s^cpKbKsx?ABN4G=*`tCrA~ z;YENWtFV1jOz=M%o)6DZ&18A|Y=0KeD71N>tE+rX$!CA&1aS#}E)oJ$^{kqj z>MOa6x$bOn_1K?pZZe`r8q!^;6RV-aKZ3fI9-u2=wXDhFM-18f6sZI!$I1I2W*9`2 z+FlTEnAf>dRiHA<6E&^7zvtf?$#i~g7RfL7Ce!7KpM0cYSU;}hirk@q%YY;q&U!iY zAJ6{mfs~mpF zeoLd$V$G(QIr(S~ko@FOo)mKgDz>roS!~Dg8Rf|`9Xy~Le(S2lRU4fi+kwtXE+BQ0 zjCh7(@?6C)09z-n`*oYiQ;ZOGjRbjK2#ArZ}{}8e#1Z&bS8#Mr<^Ay(l0Zo`rq{K&%Px! zGUVWOnIqaAR6SB%Buu|GI);krfPVOXnyaEfY?$%G)!}mQOXh#9ll`|r%5vqThwOAO zRq`Td6?9nnA{_H{ty0QC8|m$f4`f&>GAT8D`vQAPr?@iRMN$f$D{0leTXGyp+S@H~ z7>CKgSgE;nPuQ>Sj&KTC&Xz^HkM!>4px6lYpYWyrJQbky2E$FV@w>G-+iH^%=A5_+ zJKn`f$XfB4SXS$aLY-14JT=6-OHy2~IV^=o`*(y$hrM=2>KeI4Yu(Xat>4AzaBAm1 ziRvf^3ElaJbx5NVs4HcRzs+Nolp`w$oKu^v7sOsEQ5f#(u{-I3*4!2usd*%sr$M}j zGvBk!#Q4I_iPK0682mY}^SmL8aJmaQ(6GY2k(_z|O0i*CoXtz?oB>XgZ&w9Ow&{G>g9v-YaB7Q32{=)uzkTf#LKdk<_$TOsSjZO+$T zJPMB3Kz~M@v3z4n#p+2gwHs0#ek)N$rpxTrRc86r_B~?qd=tg#8^h_~^_Dk5T4lD| z;AS*;Ol+ZE`F5vTI~!*wU#7XdKR4j0MT2HN#mUd-6o^m7gnxS)|Gwz_69EmdIq?Ok zQ4RDa=VN<^j2ieN?-+>_OAU16D{S(#Zt6VdN<4{dtDGO7Z#=yG1pA)b-U_DkuJ6Xo zVl1TXv(RGZvuMRg5pQ9bu?GD^-qtJ8!*}q4L^krS`6Yx@4I%h0`X$LUiHNX zNbeS-7?%NI#U=GxT0f9~vrbW0zsE(#m%!2*^#kU-zF|ucXWuke32T)HO;5%`VORd~ zTR(F>)hu6}*{>pp4hU;zDc+!77zi7vQ!ve17~tY~*)nnYw-xoDH}^k2{k(Y=)l&;2 z(3`1n-3q5N829#V$#fPil?rps)9jE}lR7`O*%5coc}H>K7I=diunp-Z>`IMZ`;HNr zHb=Fi+2mW>_q_6@pX7uEEHr%kQIqeoCwbtCJj)VT(^^#9kNqR4olra{>a+dIVWg== z60*}?Vl+DY+@G6!G|++bOd$shfrBd*ny?v}>miVzpq zks^$@#k=ACfWVMU*Sts#uojW6ajFvzTx8KO^3up-tJ|l|t$BuD-z&JsDW`aW$Z;d< zW00bSE$zLV(YZX2qN1|CH(Ba_Z=wk3Ewt*?p(;ioK>dp0XqiTD?31Yu}+}nEA3sd*B$& z`KTe+;``H?sjelzrm^s^U%n&AMX^TpgojNf4H`v;SK8<=Shz_8!N5WTLaK|+1N@Qt zFkJEcrh_0eYnF>DwCINa{%yb;jIii%pHX<~-&#)dKVH?pzs-m96fNv()x39Qr%YTu z+VqXB{6g~KZ(4DI@OPXO;Q&x=(}*3fzjUU9jp!UKDqzxNTAoz5>%C0j>8unmT3hIA zaE;K07;%54BgJbL#;MI>SZ39vC4?3eWje*+z7j*3-B*DzL|Q`YFXHquUKwK57_Nlz zWb1B^85mx{caPr2*sjTH>*}kRo{1%mZ1ux!4CftE$FTJXPp`3-GPzFNdbvf_H~pq( z5PNT*Omv7wiE+t)tm<|i==dA>SIE-e2>B|J=tT&IRznxJwXd$8l9kZft?49N_>)Z|Gx7_f`pUY^S23~LX zuz6#qv$*&B2Jb&j-)#^=l5{@G__ZZVCfdB*gtHf~fSpHO--Q11E{tmtIdMH9-b(OYzN0w$V8hM-$ zeM2d>!vMG8dMe94XP5XHa2h4Ebh<{8%o}bwjFqWKZRaB9h?@*v^|q^|O%xg^4#SsJ zq{bz&*#X-xU2fB1?_9m>vii|Jt3Xl_~21Py)Kh3~Zm$NKG zts>izG_UO*&lY!gy-}HY{YrF*()K!z1ukPlphNckNvb(B@Hk8Y`y88 z@-;7Q3+moHPLIlFddDCX5o_$D3UrX3+x%&q-~5g%ZJ@idHn`i(cWBzStTU)Q)jv1P za{zE+5`%vf2ca;hVu=!S^f*z6HwlgF!?E3Z;&CO8Nsj5$q zrNv;OE-Ns^m6^8!JIkmWpJxc;3T}J!ar;&ar-|-6%G?c0Nj*G*xsQy*MZ7!RGz%^L z@f!^xvLlF_rmU>%vzNxd2OD{he&e!ZIzKm77Ek8^8<{NQcSb8;x#Jjo74odlvfwdc zYdMEG_D=T6Z6yAp9_h34;3SGrU$=>`>NEY0>fWm)K+%lzKFb6AUpFKQ2JE-@GVVyX zF)Qh|$Gc#cS_)y*NR)7@%=mm;>-o**z>zd2?u7E1MrBS8kD_{^G^6yPEcGhP-gcg5 z>};)rWubki*}c(ljAwm?Eg(}9Zo1BSj(bUwqO|yFq+EA*8T6{nV7E;Y8z+Sm-a{i2s{TH%o);%I{M}Ip520AyhsvK>5%Tk=iXXB4%`07lq>A14fE`i z5E`i@9bZQ;LWvSz_xB#H578pO>$V_VP9-lM-jQ93%JXk;p-yan7AsR%{7KMaj1<_k zu3QkyqlE0KWy)a!RiW5pR?6@o)jw~ia=L#Q_{j!5tM9AzO{Tw($|jl2i3|nx_>xz& zWEaqanJMSFGgCzn@+*Q47U1U@_dlZ5SP}P$QZ}i(r(0YHS}sJSm{*RtyTrdi1{!-B zJ0G#EUAR-E+xo@OABp?c7GdHSPA874Rnsqd?BxRyZsV2Z){q`2ls^U`V}MS+|H_BE zW!NPo4il zI6cU@<+&kIteU3@3U=X&8iahRicII2s+-=v|FNbim*PTzp$kZ&OUB-8~juN2z~ z?P5u14o+{rF&U0h_An_PHUKtdIP$iVRq+nCZqy!%aw&&q{3Lem(=~7YJz06)>_*%3 z)V;MEGaW^=8rAJT_yZiFUD zxM_5!B3OWlh|nh@!3gGUw4XU{&F?xS;zX+8A}v>IfB4?$ZTYobNcA@PIWb^kKE}~v ze~B})yH);vE%ld^zb&@^MSl2qWGmvuk&2JC!qjiQh3ngEftFfoa0g=UY<&VTOTA7q zjNCq#f_=!lSgx`;b<2B8u_=jFWBAC*CheC?@DY~YSpV1}{O(4M$xP8n%Ns4)yDN`d zkouf@l-1}8@r~wK1A{38hR9)Ns>c0zZXJV3XH-X?j`+Ryi@Z0Fhh6 z=xy3{tXq=yxvzNXbZp0_jt-~*I|j-ww?akpXVwX+$6llr&zCg}$XXI`Y&CbL$WprB2c;6c#p z(b<3?Z?5QA!92~-ZkN=Ji|6EnRMz7W_6w9`RiwX9Mfa9Xd#Y_TWDhpPSlRPDPDR`G(E-dcei?8?) zg-VTcRs=b{2B!?||F6CGj%xzVzK6$(sDO%yA_9&SDZ)sVZbOl(^d6B;B-GGC8AL&m zCcOwq5u}C|LV$pXfb=Fr2vs^EfCLC3l=sPZXJ+@eyR)wIdH-DhiXnvLsrR0H?m1Sb z)P*rL)1AG`v8tzIyT=7d%JxWQ64~Wl?&BhLMc?atwVd+KH?@)ph+=ORen@$-(6$G* zV6bXrA=AHHGA-y0E=^LM&F1_udR71eImw3qk(zwv`NeVcW>4!d?dD;HY^I$S*17MV z7-7w}O2Mr*Y`*9;U=!lJ>JsYaSy))6&MLln!FMhaU%{HTmnKK8%i$4nx&8Hm}8eWadZ) z^(3GWs6NvHnOf_4CJP-A9vLGFHFh-1r6~B`nYm)XeL<%75-=QfCCp+n6bnMnN)I`Z z-tTg%G~8VqR2UA9J_6;*sZ)_5olP)2iA+^<@njZrv%P^GZozQXeTJRlIw)X6m1k)) zyy(Rs2?g4emwFvOp4GPU$wevPOxm93ie>c&l8p)X)|=Pf3Ql)T>(9LEJ1r#isPvNi zc`D9?Vlk+K(A`Q}p{7hbIuD#5)4ip@A$y~#qbjB+VT0@f4OeO?3Y~D1JhtVLQA=M` z2i^6vdvAvKiXQVMC--DD+xHz^7~7L>J8RNn&5gp{ix#2pWaH+P!`;k_6N07Sb|>rG zl-S@LMfD_B4vj>nzAw1g;ZcztLej8xqtEX%6hOHAW@4VB0@^mL7~?+DlrEN#b?LX^ygyB z_~X4oNN-uwqSX>bTdILA?PJZ5Lkx(!F_(Fno5I}kf^1q|8xEDj?Bq`lVQ24l+DV%D zC5;X~JkwNlkN`)l+v{3Qtz7Ih09EA}Ds=3M;?6Hx3v%6`j?PtcELX8(JkmZUaM~_r z$9A5V--^WAW1*(Y1wbcqWmYXh>v_5r0$4n0v{t0mabsgMPAzMYT`s z1PS{S!!V;8!ysG7yC&L?A{D>AtNG&pZsf{?uvUrpY^>}$WM$cc(Y!fZ!7XC)4xAEXQu^Y40JuQ{? z*isU`Ma|c7 zJC=fP7SH#M4^5MYl*?3AjGfuJLS(C&6@Q`^p_2alveSgkdl{DytE7(V zbAM}!AJDiz3>T7zM2z|tM5y-?jT}t!D{Shv10#e*Ov*p5pYWQVp2dd0o!AI&@7plC zIHF(Hj3RS}DscOEBngs-iMtJGJuS{|5xWDM8`56?*?5I^ZtcJL=EgVk=p17}PLZiO zjpVmox2|5m_`bPdq&W}TR#LxGKHP37NiB4rUWL_@E^D9d)b9hnXCig6{l*pGoRWe) z)YbBk(Z~@}Oh<>jw7)SeM$D_PO zLpA+F73Mk1A=@ZH{mAxnqSu;-jv0zt#~?beWd~*u9k}OuVK}-oMbF~(>XCLKOuhU0 zK=BW-xPFIqV4?w!XDq@#m?#Wy6T6n5TfGykI0q_HzO=D%D{gLc5mnv#dG^$Vv zb?MauizWm2JgAS1O7 zErMnRNPaxN!=FykWAA?SZ1_^c*qw!1qRNet;5*T4etMZ+^>eAsf{ ze8rV?K_mL=>$bvm!8RSQTbBL&Wtm-e=z#s(`O=4}!!SaT-~`5%=iZ9kHK-oWJY`z0 zUi!6_X{FOQl}@e!IrYk6{bOAzq@+V#@|;GOzg{|!;Yu2wWy>Nr>7o ziKHPNPBeb^RxnYCurpc0d>i77dUvUov$dPLoNxDK45gHe6l6$+91|&AOyj@hjq5A% zNdZy<5DTZBBl~d?4=S5-UXoW@eEq)Faxwnxa^~8%KYix>IQz_bGyf~v*-!VJY$B57 zlb2mBLno8J0tdjM27~2^8H?;~)XMfQnJPPgQ|MBeEZPzGS;|-P>XZ-1G{POcQ^^Q; zY?Hc?jPAfXA_+t4XjV;(uPB(=yI)yoq?@acu8@29WRY#_H{qy14PwC$_;uflx$V|a zxb3PjdZM_H9$lZ)EM?T{paG&o4TAN#Xe=;hOAw-+kz~eiUy}HJy%Sf=;yu}&iPap6 z@F60%(LTI>+l);b(jv;9+S0c~I=&h`v7)!GkUQi2In&+~#(C(#x=%XCjjc@KG#Ab^ z3P_DRjymI;!4Wr>YnI9j7s!QR^Rm>Vn?@b5ocG2t;T(z+lHr7qF=SZ#5=821j)r0I z;SX)_P9ey7pj)6|B4Sd>)t9YFF2PY}jPvzkO)y1r{ol-xW~G2K>XqSa;ZFjdjEdpn$NO4eOKBZTUXG;-@r!Y)cr^l^wO17 zSd<)}hnL=6`BYKHtDNT3$U}4pWYE%b0u&|Es zfp;P!ZlZA7V&=7fdx8~~2C(>isZM6O8PWfW;qO`QpFFyL^u$?7idLNZt;tjnquIjr z<^9vgPI&F6%gn%e5@3+jpk6!YW44WpVzZquO(nb+A+FOEqC9V{2Ya14Uw`oS40@5t z$Qiqhaq-569N>M*`EJIr!*?wGRtCAn7*BXW#2cqQg&pfmY9w~c(`AkfnCiRV#ju`j zKRM-B0EPIsMU0uMf;_^b%2M>%KDxc_WE3!sn$X^h!Er#x7je%XM$^L`I_$^zf$*`@ z$P`>U)9cy?eNuth(ok2KmNz5m-p0`WiD?;;IAqv$(2&vm@#Ck9q=;LR+ULX(8wAeR zGKh_WwcV8l&s;LGuNip$e5rh}yGha{Y=T}}vsC*boo(IyVYnAxd%S?5R44rS&Y~E4 zoU)nb7X`Al!XDsKO_Lc8Y5wH)w;JA00R;XNi{D}vL3;0}S*N)0>(~y?)keAV1iDY@ z&XTJ!*<`TVqi&WLNkHora+let#qs!yM2~eaX#px-?4IP~7}6}#XE=_-^ZoW4vJZYk z)NfHrYiB^ftoaepJp1RQ@1HhOq{*T3A&jJBx+|j7Xx)c~Mm=t@Ci(GaEuB+9qqV;c z>%H7b?ej_6h}4I~e44ZMD%rb_d8puPtQb&D{d85d0(!S+M@p-0`rG`R$hlv$kyHzTTW-w+8o_hzi_*yRC6aVz|1@m;6M#@w`FlU@M;M=JuOT}+R z$o;&&fQ#qAB2+E2mPLlSlb+zrdy{jnl|p|~2#+T;vz!(*(os1r0t@p1IrkxMx25$* zCX%BF@<^1hsSLfSX8h+o!#*pfm%j~cVu38o#% zWFExwX*~+(BunQ2*>URdEasAW(q|W{16rkP$<%UdS+QkHLUk(m0i-RHfg`5e4k>4m zn`TY=#uq5ZWb^&fF)$+p-a(AOI|yJAd9z0JrZiKmpQ@mU^s$G~etiiyT&)zzW$@=` z+>oLZf}e*=%^ZTiMp8Z#Y1rSv@RSSuJ~!%MTFG311N>Tkj3A)_u$=HsihP~#Hp;sW zWl4W8ZAsDEh^X*;Q|92Pak)`4zOuUl8)99a_31>^ zc#rJ~wrpC}?VNb4*0_2<;z?jC&3t_#lpSm$e5zFYp35r>$^)~VAk&$w6YV0=*P zC!UqnTba*<^q1O`^rYc4p=|?SUb^b<33{8Rs^`~s?b)Q1Nz;(&IG!?8DtnFF&!bCk z#mvuqz2Y?O_^5_Sed~Z9WnCYXPF_kBAAWZ8Nuc_()lc6|#vM?uuUgO3!aQ*! z?gqHd4WS4rR%d=stp2nV5)i}nrry;7G_OWFnW*Q#Mr`zTV=mk#7d+v>dEUeBEmBC% zhZBJAyI`Wo9^H`>3*c^wtsdAdA`hsCy;-MFvL+^7F6)X zjej!EAU5WA@{JZ;mt)TOv6-ZcmG zUFZmJKKgN)9xmx?P@4K{ArW~aNUupPvdW_O3D3aJfE9MkQPPe(Y-A{2|O9ZebycT z#h9cidInT&WCXkP&&Pnn&2n?-D+hc<;>vxGb2~(Ilr-rr-_KD&6GLH9CbwocSfDH( zzf%sl{R0;I^Q)eryAnB@6n8D(tnFo?*RN_6>Hp%?&6+Vmn(AE(_&~QX5O-Bk%_&=a zyi6p^2V-1dKfQ6s_e1$CHBH%sr7K5mNBfSC)SX?48!ilGiuhIj`JTd;@}VTN$qoC? zQen$ak)sYNEZ;5cj^8=$#E@DY+(|gN?ZR5PqiJ6^+P^HfB;@~c4T^7dOe2(NAE?3_ z%U@AP`2NhNky4={y*WGfHZQlf(0um%TkJq(fE;NeM?-fGqF#xp$c6c;raw7lwVG_a zvB<)#7P&GkZKeOY{#R3Xgjm-zQ2z_$T`X{D-o0$6xG9LHoc<;46Gxob7w?Df9h!BE zPIqzg9ur|j$$D*t%5_uZd+euq8c1#V5%Q12w2($tgO5X|e8u2FsM;pEg!{7{Fs;Ue zn?Uo2M|BU+&DL9C+=p0>3F1AN8lzq(3Gtq~F)#i&OQPTm<@_%b`sY^d&(Df8*y}{A zS6@=fe-G`uzsv3lSsx2Np6y+u5Lyb@7digc(Yf!mQ5j`Rm=kAC39&W#ZxYS4s_s|j zhKiKJnl+?>?sK06r1ONG9#zA}ly(2aTXGKywd3|K_b15h-mH5>>#3pO>3mjYy?TVpuPl(LOhp#ZfEsogQnFfu5pG*>}_8H7fbQ(Hxb>^ ztkV*+eUb5XXD(cDcw&0KP#C!CM6G47Vn!;GBKLp=1vl_da#2kb9hh;QxpZ!z5aM`& zIZg_BJLd95ntw3UH>%%U_eE&9jJ1J5!W=u%LdtPe=)ufTS>g$w#NMTT6H_{Q_r(PG z(iLO0ili745NhrX8pB?&RJh6qS^aIT!A=L1&mf*;Yl=suH`F={&jZk~VEMe!&axf9 zX9AK8$fJ2opf9-YpDr69l&5X>x<$XA3|TbKl>W1g-6 zrm*aHT;R{c9Bz+&)mmORXXekzFu9>fPfc#UiD*d^%yEMa8I0~&17n>lr^p-nV=ZJy z50JD9`-*zJnN9dYfEa;Mc=lWi~B8Lp*#M z^QsE#9lOrhE4O%2HovN|(zMtq*35P>^qEc;CaB#j zof5Qep{8NcJ003sfu$-zQm+`twTq5V4`eLY8bu?PA_wz#OY*SD;c^K*DXii`S`SaK zy+_rknM7-xE-n%Xd<6d3lF|6* z5l0RN@f!|;P;Ux7&!?TE9u3-~+r|S9m{uFKu^BD*RHNrm-jX|2bHgQ>^!V7-;YOLi z1;J9UuT%B$tywMFR_HdS`*qtgjoTjXA$Zh6GpD>s9jt%|ODQ6$y!!C0AaBaieIAzT zEfIlD%4R~Tr${z<5BBxYP1Le)<$8@57h?GbXgK3rfRmR#nzo0g`PF?>p>_`3@?d*3 zhC)Q#R=VIME1XIBQn-PvovR4@{t?ZEIi~aCvb8R6xZ_VL-Cn&Th73$1qPoSj`W2gw zpac1w?2RhKl47)7WbBUC^QAZCk-pQXvywbyHm=8xQS z5cmvLwHh7s+H&TP;X`kW-1Ehet`W8<7k~A#%Yo`j>k2b{OY6%Y z?6e;mCd2B`TG9kU)X&#}*P^rEzwuz%UH?7o(}Q%) zRj+{vjZwfNDSl7O_JoOG!}4fZb6)CbCp&&OR{zlj)-(pT<9*$fO-}z905==%TW$zu zYBSTum^3^Z3uU)E)QXqetHGHoiZ$$kGTvq$i>%F3+^Pt{VIeD;&Rb^3!YS zXdkJsu6CoPVaA)oq%HA>7g7}=rAJcPNj+askZQvFUlUr~(HNW19wv%keddAH}5 zHFKJwsV!IB6lVLFzDoGxI@4!7+Eo>oXRBHqD6^@K)rAOS`qr4MJr--5W7-+W%?dvU z<3{oc^e&q>E*=?o8s`|{i%8VU5Gg!TxNmy6evn6KN(ENZyq9z?(AwPt?cq-fC2$NK z7|~5H7^7N}RV|)$l68b^u!{~w%gt)MJ+K?Rq_bmZPNr(s&UFtQ)q}o~i={wnO)f@_ zPKlW&A1*RZ(6sj7+C^~puJz^383l_VueQaDeK~4by-_!0RajMo%(p`)u|KV=N z)IP9sMXgeCJQZf67PUPO>z(dYfeluOQJ)QfP1CW8jw0)3%vZ-4;Z?7e92nN8<&8~yrvx{f%E z(uHeWoW83sfSJGn+yz;;Vv{GriY1x-_0{gnOO+Uo^J!KLC63}3ms!<|w_AACxz&@z z2PUAql;)#81UbL6ie_xU2hG-L6lK@A=FD4ML`*XSlT9&T-`}A-!v6Nfpwo+%*3TTiJBh=XWW&I zNb;%=n6DvqGpi^H3lgNAQ;OFr_mzD2PUpWluC-!PkT=`V&S$B(cVE7aqybBJt&6s@ zDLvh7-`#@xdF_REj#X1gzgF+lhD%4MBBni>P8S&0S;Uvgw;#-B9umz`(fya7F!qrUrCU~57(*|;|tk5r&)VD58nlYrZ$!!6>eLiU}_f?W|hT+vCG-5+fJknjl)mVIaAuTCQ%(WD$-8htb z)#5M|Ry*UdXb_6OI@qrLZq$gE^zm`q8TIMJ8O>Hd(gW*i^6ez@fm20_50ifN+gh8) zUTFG0f0av#-gbSOb&KJZd9=t5BhGl!XD~?JdQV*NO%W@3`xgA`@!Xut2YC^F$JWm8 zz1$YJ3DA8Z#|;?R1tTQb1PN4rKY@ZPcdO20G#(q z=h2g_|4JV9)i3`_S%(3YllZy;EL0zZba(JKty$`OUkRNXUdqxm9CpZi3SE(F)!9~n z?q8U1XBHn4e<+XWWQSsg3ehUqDtMt)mtt#-@D-r87-!WMk5~%9dH3<9t_;;#hq39s zA%s$nJQ=C>s9x`(%*&zRS0|hRT@~7VrXyLl-W5lzDySz-W`6}zF%t-=$L9k6t&G1l zzdDH>wOy%-p?u-)(LZXJ)85w%yo6$svQJ@<8_P!Iy`|vhZ!rZ5vB`Xz@77l(c=6xu zy$XmCcmpiDm5b_Ksc#AH#uese%-3)ZPdgpXjXH7%tSTLAn>!aDGJJE6mobX$L&6exST*UchLYGC*4uTH&@>k7lk})5i z9c@`<5zKl;e;wan61`gxaIPd0qX)SnB-B3icHhzgIwqMB@hKXQ0cDn6v&ZMMR5Qsq zTlrMT66hDYOnsVIzt7+`-QhZ)$zA>d12rafDGuc6v0XWLa-you0RwDYPqDErRCJj2 ztYCe8ft*k@&GA_>VS)U6$A_}lenGvWIK~TOuR?T}V|X=DQ2}WTB=3`3@)~KiODyw_ zfW-raNPfS4Q|jffdeR;Y$@qP4zPNErg}}+i+xeM@{O5jU*PK(Nir)L`3Vf;tJOwwq zzH6+MH2T+F#$}Rvb2UmS2+KLB1Ghtxpl?c`0Z^d54HjU?ESt4o@7vaj!^paG(!Jhr z1OWayCDp`>eXvP-JdQ(IFUIX0d*ytO&RW7k(T%%Q!5htXwb68}$_vux3! zo-qi4_mIGsZYHL2ICeeAkFKr=j1OFW*0C*qeF-_wPF@~cMyXQu&0ewMQPH<}pKjn| z_y+_*Kdv4ZHZOO^tW8HY0vc{cXZ<_2Y@!cR-At?D2u0c1?aJlrhNW(p&;D=1pH4Hx zW;Q>%Etul3=>3}AnervSyL)(b(<0(pbICuU(EsMY!v*`}Kg_PH^^!Myga(%Zw)G zl67aWw=j`RLIjnnLmbwDogt21`FDXu|ed8nod_p#+EjZcNppJ-$2S(YBZB)hefatu=AH zV32ehQfp*LY8WrRxWfZfm}{$wzYqUb&L4bd>(@8pBK$>(%Tc*FcL9R)p0~d8m|fr5 zB)-A0MkRp#hN0znTOqPkY6Ql+Z}089=)~I;m+69Y4x>KvnKz;a*NX6NCJj?SVP+LE z3W|4|BmRZ%hm(tID}1RNjb0&@TXP1$n>=V+#yR8o9IngzNcy8^M+Yi_Vh64K`iaWv zdwSMu?)M1{V3`}lzV!J*HEoF%hB*j)taKM5+4%T z%X29!hGI|v4ldqIL$U8Cd4F_XC||92Ww^WtOk3g205e`1#sv2Clse!gZFb^BtzJ)` zhhG$OGy|1Ah&yrg3Sg7Q&8GV5^=60p3W}K3Tn1_Hu_FyBDEpR=Y5sAvMgm%W$;P;1 zO(udqJ7UxrqON$@qPa80?7(#6OP1)kD@MvV*hoPG+F@)7m4+-C4F8Q!5pc45*_h*< zN80+)_T$$Cznd`pF@YcTx!^ubm3lj~oymNh{;My5Xoc_q{{|7yH{!h(`RS7~bPNj& z#q+y5D$+NwWTW^3-8$OQMVO{(jf#g}c%B{Gq8ck?m*V|7;i@PJS@%fYVj@FHbhUc< zkpo(a7LsuIY?^xw=}4tX*o0#Ti96-0wphy-r(o3of|>KL0D3^y$>*lLvUW{j>`9o3aD>a@y87_B$2Ku9El@~8w zyaM&8DpLo~-EAO>Etoz;;fWLC^%*Hcc+7vnlufVi%=Ki5FjzISE5hKd(R>j#eQ#fu zd(HDe<#qFPVA(ito?tJhH+8v>#x|zot^9RjfG_+9@4?wCI>m2l#`5Ejv&nrSY=h1g z-jq?#Tcc9@Hb7pjI*T4V!%$AB?CI&*_$3NwVlyi><~CsXebV2*e(!)R7G=qL(xb+Z z7CPRiJ`*1f4Pp^!T@)3Fd7oluf2^y>L{ERxv3@@Mg3R+;SW`Ajy`oe)u`8G%nsV&S zr^c78h25B-GC?3wGE(i%QpnBuDLZ|oB>QJ-Xrh?a^g_Shvke;Z*>Z}fAovLFu=nt$ z!MhJ#m+$&f3w2pc>l7Bdy@1)|Tl-jn-U$?ryP7^1IIjM&$N%d`<3mTXM9rsP;;e1& zU{m+g))Ww+C|vPsd48lNhQDuJ&Nrj8zw=--s@FHRHP-5Sou5LpjOWUm+e;&Yw3daR zpB+;L8~#6^yhnzA+TW7>mBR4wT(|0x ztFaU7vBN_}rq$g3yW3_!0oY_66MV(E!l18{nQg&iOWJkjwC@iqfY4Uv_}0;^+S;7S z$5Jof1G$=AqL7e1G|kI3}=v_*q@K#NM-t#yKG&(BKSt~2eWh%OEakvwJv#I@pX&{SPu zDKb(8bgaMM*xwnof2^@zS2?2$?GvH!2^rc29ZRz#_foT^VS%1jL`J zzKbuuk&&6Hl%EcLMtY?5Qe|2tHVfD`N`2~gIS0PT^b|NhSVhj)l{0mQ_d*N8^7ADXe>kB5L?j{v8b zBeHzIaX|iYR)2rZe|SzS985-xD>C7)efSe600U{C`E1q?EbG6I@%7e|fQ#<3#3J_B zKK%Zl0qKux@%8Ed@CE;T43wooZ>4r8_EGj<`|#JnywN>PiT-{D_wU}H|1MZGtr-3G zzxLseJO}gkYq6`=k8GDeh6${AuyLiyA&GzO!|wz0#&N=3%?&Hmr6l|Nqhzs>%?oBhx4;6Doz z|Npt!RZ7ILp5GpYq{uRVzpLWke8^YjQUV_a9PpMqTk8YU(=E|=1@!aWGR^GYBcVV6 zzSO+o@Vyin)aOF&W(NhuP)?CcwL5yiPK$4uyvPq2fzJ4@@3&6=u~Ytj`U1+=c}Y)& zDN>gefH4LxfJ8KSdwVh>N|MNc!0g*mF)nsW_xUy*ulH^kC~C)c0yIR61{dx=IYe_u z%GJWST%T=ip``_t2g8&6%SsO&YU~f&`FDZHpFcOOvp*Nv+@GrGZ)Qg9zL6vjwZ@jS zQBaSA881Zf`0qKS{A^m~=CZkBZdha4f|BF0K%7*L6^Q+e+4F;H0ziPq9WwJSp$Bn? zGFyP@t5>DAjYoUPPbKBuiA`ZI*e<2d&RqMk7$B=Xq{LVa|XW`68>z9{PTLP{Mo5PFqJ%y*B1u3o`VrO4kqtyB2AJo8mjQE-bR}T zN`T_6>S+;S8mvS^`i^T;wtEgU_@o+Azr+rzCaTN>{#4{KR^d;;_seo*Yu#l?$DRow zhcxw7yS}Y#W0W`QwVapMYGMfC(6Q6SooIhB)1f;L9z;nwn&sXvdJW1a9kV7T9`DBL zptx3=_~qsc=79tnau^wT0tbV&0l=ozeDV&|7$||i9w!5-+}pFPmN#P?+s{5b#^NH| ziEEwj%K^Q}v8By&%8~s#EW)N$<+=|PiUekiJ18_%5O8VBhke43_)gtqO-6t-F57(TE{ES28MDCgR0m)Pu)2g;HcZ~znq;E5HXT2=1%O3R?=r?A zma0@`YBrJ%($Q5rc$l6ePe#y?Qeew*sVJF7ZeI24N`>L&^m)}@UuO2FH0F6K`n8|! z8X2e>P1dAijOj40_&9`2r?5giZ;0D2=kt74Zk7d5=5Y=g9Y0ahXDi&9M{0nmH%U^c z-mn54Sy2(!eLl=S?Tqg7wc!2p^XBgd;2)1#jJeiPNvT0GbPOCq<}{)2V&vVvJY3)2 zcAtzcklRkln*t?NG>Vw!zZYv-?QYtyu#P z_)P^%pl_)B%#U#kGrL2J(yj$7)EofHtu1TeD`98p{_oYpsHM zvF?tu%Av1D1KoMukoDdSDFjUZn0G$hPd+|qO>wdBuB1uPn`Pqm<6FQ@lK)|F7u+3K zyd9WH4eQ3_93lMB&>q=5Q)$i8MVHh~0pjCt%W}Dj0!9fv<*j7E{4CJ6<8iDWfzPJ& zj_G_JDBjS~KDmRjNke&1d}yQ)myJb3uY6aVT>YGImZt;1qV#YB3&V&GuiEV`D7bz~ zSyaf9-CfbVEiah)3R-83q>V4QG42(a3PSuzVY&={^#GoSZF_o^-eX2z+*lqHaZIu& zu!RrUkC}#M9xeq!Ojhg3YXqboj~hb@S=*)?v0>kONe}6B%3oB~i-1R+rvH>WrAHC#8`UFzZ6}Y)u1)Ceux2A{61<;2Rh-Xfn zdUPloZaO=iUxgllj7c0H!C2h&>VEo4BB4M%a1zzNUIWdGBuFy6VxlexLetQSZlPOD z3YUox-+uZrf-{Z#K~{^D(JwL?FsI#IVm6(X3_Y11j?w@WemdJFvIFcERPNjcjz+{U z;}gMe)d>RwSIg9tNJ=kRt1m29MTtNZ=I!%B@ae82tVy!ywO?F|1Bbo41)cm6`j9z>sBGvr)$JRHS;cqb5YOwwh*9Htg;q2%*tq2 z(~j^d|FMLNKZ3zsUu9Na9mnV$TJA%pv3&LI(D_np23lc(+(3m!&OU40i4s_WD<;Wf z3IqxE>kHR%SgiBD+Q&LM7W!aZR^QQIf*O~&ywy#?8!Iw^Ww*fZSbaONA6WslA@=>c z^|fx#eanu=sc^KYkkY5As_ax;)&)<0&=AzAP+dpxLtt7aBV;GM5XrUB6;q@V9(cYK zdQc-kV?av<5ntGElOz>?^U$%O>fup1Dz zl1U+Q%#Wh9gq`Kc(bPJ7;pK~xrapok@KhA5z@7kXkQzUfs7Ao+4|^ z(vJ#}2by>=xqY}5jM;^B()nc$SEV1?p*TG*2Q9`r=RV$Zvar{Ht<`Mvb~#HfjbH#= zOwf8L8BZ3G8*?=4VRdHSJ?MZ?zNmMwJ|y<|InheSiHZW;IMVLjFsz4;Y3LmX+Tb^kr)D| zaB9HUrYnUO_|dCgw;gnNSoT}4@B&3yd38x^ww_MD+JG?IE{(FYTEp-%6|Dkto^WxH4SyjtG4NMCoO#%k44fGq|&CNuyYs2>3g~ zS@(zX4nyUujgc2Kbg`56c3L^%`KmP=c-L#LGodVgOdTos*>7|{IzD8tUkB4;8&UI{ z3t9WNH^mWtFC=;=)9glvy&XVXsa|5L~ar7GL)zsDv=>mvaM-#z&OdfjGwWBvoRy>C_S&9o=Jy&kZCU<>dpEN2P~JDoj7 zoP*0@{jG)(Q1hX+dA;}=uhmca#32fCm-;zZPl8YvFB}`l`%6mSTNex=z@2Lto=jLB zC^Sydg{03{Ha!mwb-TG*Wbx&qV_|oyv^H;Zi}VECz$qo-Fp?Gg|R1 zC`cw5eO47ml9t^T`a-&=lKK1;&uvJA6if(gD$((>F3TXVcQ-giyOGI35prK`?kxUL z{pxQC*?1F+a_=~%Zef}*fQ%n-7&3?g{M*Rq&yU<)e?w1!iA`8nk7#9wV$aS}fl<)y zX3(xj9HQ0D{UUl)27xh8nLy?q@Yhc(uP>A8B|AxL%Q}{70W0>fE4Rs&V-v-uYRQEbS<7>Jz1xSuNAQ*7ns~-8j&D z@!~~c<0$JA`{~xs>!eV9xEPQCzjbTg%Zn4m={~N884O}CP zeM_x7?BS{cml*cw=;)fgAKp}l_)@GWZE`2l$@S}mKD^xEU#MdOs#T=STgdR5;;xhM zjX)p)!4Lqoa&yoY)>k3BtEs7l`Y3vrqfJHOL*j`DX-_jimoP}T!Ufju&YbUADYwCw zxaQE{jsfe!Dl^(m@iHjYv5qJSHH{cUJEnn4)g0uDy({0E-@fwN*`O{LRiDBZu(S}b z?|=~Wh2C@WfSccJR~osZZCR25k)KilY2O4b&)69jH!?s&j;D2j1jEDI?5re zf@fxZ)qaCX^2cU6lC7e>yr66~9F!4{)xpS%8Y_q#`j^nN*uD1w0~d{mH6Z31WkgrZ zpKVigH!)F5)6DcP|c)U5kZsiI5GoEKh#v&a}P zzQV!5k@|*&9l?~r!@Ql4AjWvmq!+gC1H~1V38dgB_ExqjZ#KiV3l1RAA1s@|%`YmABptOD+Z zG+vGs2B_7g36tW*JMCp%D%om@1;rRsC1a}MXl03KGGd^>@U3H-LT^($%~q)wgup%B zKETCgDv~}GuYRlN0>s}mgyo98qe>|m?3Fe_kvgFRhZ&;g=4{Vx4%!T~^=Aa-wwvW! zt5Yf+AyB_>(q1%Gg`G_culj?_5^{=eUw$FPO1oO~St2Lxo_V8`0fh&@@Zp>5B*WQ`>#vw0m=8qt90c=90Xi| zF?^f1((T)o=b**7(VS(|QJ+4CIwVW=5C*6jk0N~GFHzXyTy9!IirR@dCsv>`m73C< z{E!0roeQ1&vegM^WJDlEuK{mQ(;m_&Em66%?b>CyZ!MZnQ)agJeU;g4zI7e)49EbE zi_i8vtm?ec&|Q>H0_PNJX47eBt2Rh^ZKElSt))4e4MTi;!2TyiV4A*a8+drt4`~+F z`gDvLBsJSPGi=LMB#oSfYSTe^i>A=kr}Q&S=Vf$OsYf>Xnv?~nn68?8?+lgPBGNH& zl)qjdW@(L&7g1}Zdcz64Ypd57|H3exzsVI#F9s1H`rIDfGCh{)zf40i z|1(bY|NQiz5tm(bLqKg3)NR)-dy8_#s+HS*?Q^z3rJfN^|E3YZtTJN1jzAmvEVG_+ zX5Mgu#5OZ4EW}=t=q^G&t@wEM(4j+ywe<2*%|GmpbcD<4(y^@CS0dri?<+|DGr0MG zckb`Rk^MiC5_19Vu`RfVI?Iewc_r+YL$59(0^ah+)zLi zXxZBMMf%4E;J4MowL&8b_%kK!`}ae!U1utry$?tioShky5dHFiF01C!^( z!G+;kNT}5mM59(3Nmb{UQ++1Kt})Xk=s(pbKzqx4WZqrbjsokksN2a4;srcEOr!(H z`2o9Yf-sZIcN%t>-%-RWX;lgEjweV`zmC#)!oB8zc{Le?WOwP{8A;Y$oK^fMl7F71#kFO3pOPJj5DsBl@APdQ37vn27~tIEcVf_Z=u+_ zOs)~AyiBecs3r$>QPCR`j|ece0=euXFk%Iy7Zv{Eh%@b1%FX?eFMsDt};obFmzvnQa zMX3*)=H$}E!L6#5bwqGu8znZC5R%Ek8#&su z2zw0T_=G{NlDaJw23dYx93A?VhXZJv&KFOJR726HP2*$g&^3m#xSba4<6Jz_oE7-Y zZ!)(Rs_~LZBbZb0d$68Mx;-#pw1$d{4afX#AMD*`81g4>s<;H3!HDbS!q=8=Zx9_U zKNDH85MYl;eK7?li|Ayqp?yAR0;Mxkej*o4WZIR!+A8-+ccuPrXr_wrDDq;)qSqwX z6Gb_RGc*{E)kpLC6~cG(!66!X|BBCp*M&aMoZoAjv)?m(2T?!>cMzC;xIS%d-_^id zTwbnkY8`Az|86hkOV0K7ArO79u!Ge}7zWRV48!s*h#Nx*@q^gPk~ZiSSW8$u2+NWR z{!@^oC>IA!O_eNMRn0e;N0db5Ur2*@kn7K4BHLYQ zwO_Dn#{ACc{$vrT)me#_wMf{*<4{W1h%6YC-K%0wSWRC+QC+>I2R|1= zuD#if4=Hr+#~3A;-c#k7A#Szq6#ZaDzoWvflC5H*72uO^=QBL^E1M;o4bR4Mw}$U5 zn%exF7H$xS#fHH~7v?L7a`xy%*#iW4W!C7wxBG$_1UX+@TYCd)ncRXoc8{zI>jm+X z@;ug;!n!Afj9n%g%F~Kjj}0*n}*ym5q4yV zSR9gWO2G(6XUeY!Kio}Ujjt8FYARZaq`XechsGAnypGt6rmKqN4cpo+ct$LajVmEZ z^hv}#M33g?6IYLne}Wh-S(Nyf1T9+lL@HEE`gfE};yN@tzX%FR79+`IRA>kjc(ySw zun(k+-{h*HYs-47y_rEJkE&J2B#9Wv!<$6+K-CXo6HCpvvZZT9`yr{CpE^luD|iog z(hXA_!O5vP3Tt6u$bzdl@SBsEhm?6hp+=#C;A7at9m!mrrx~Ry&G^p`&G3fZ)ZRjD zKTFo@&gs1Qjl<_RO3pvtO16`}$b7>=N4Y^zg^TBiOAt>}V5Iqq+lYsWGybzvHV)09B%Gc*|#)5-ki{oGZ@i% zCWs`={-U?Eal__MXQ3wiqBtZtH26z*XyR8cQ^sq>&kS^5XbP1|SPC2p-fOUF3NYrD z>1%!b&dVK_A)PJDS}>6p7N4RtQovfIJ*_fDF)3op@t%y^n?xd(p9G)8lSG%>_&1Y> zqClibqOllz5#=|NVt}2-`ycOHGW^FDGw?Gaxt;B9Y&q*??4<2?>{+Ll$~?u7T6pA_ z6$I4`N;|~-%5RuX2GL{8#@F8=#N5R=#&E@8lIO=YV5%}GmPpw&sg$cURW?>eJB6IS zUqxLlXgyJivFqRFbWa;edr0R^&y66)*h0s{Fc7NnT9`EhUq>9>9YkGyzP{OCoqQ;q z_>DC)F=vuzQUwIT@K$qu;z_fwo?4F)pLfaLFz$WT8yyLze9ok{x3wX%;OliMH;Xol zPpna|PMopLo?$zGQ#DIJ(5LRx#-izZl^Doqt(q)xW`~&Zj9ns$VZUoq(I5>w^Z%VoG|-N7jiC zolUik3ibN7+RoQ2`V$#8My5s<0TU6{TOYali+}I7vQDKqW;(WY-+U&w-?LA((=`Ju zQZxYS2$!9!ok_)sHaI2T<-7HHmd6HTQ#T2T09`doH2c{e>&`)dU7Jk&pA++PGR2TXtQ+K9E7s>!p^jqM;bBc_fjjYmjq|D~M>DX1tY zmLiX$mywm$l@g65#^>VNHDr0?rExsI0C0VF9oak`Sngxq%*5MAVM3Wj0pr2r>e4sH zlkz*wH7m*^%5N4%SZcUQT-QM;kNt8y@y*+RPaDl&r zwIuA#565TNALRn+x)@?p8%2H?{KT)oZ^ynz(M1!KUdvC;w>ellV4lvk@wTZL-XFah zV;MWfb?@KV)crvHA+vVPM%Ehq>vT+hbTLVBTzO2GWh%u|QAL#}N667FbH-So)-iXJ z-)iiu(}y&NHnK$31m5j~pOjAo;jxP!V9=fM9Rxm!21NoSzxiQ?+bcwge z%f_M40qEY!4FP^<`px(Se^r0K!DX7pKhazLXmcuhp7mLWg{4z(Q%ght!Yayis6SWs z$?(P<{QIit$H+>GqoIl++lN5+aPa5v$@vo&-lJRSz0D9|8+{y38khd#|JCyFgScE}jLTfhA>`Oa-0*==$>aQ#?y) zRVW|+{iE;4ok6aX;)2&H;SKxjGDc&n!%M4VZ9&J|;zN%HKhJBG-s}8TQfiLh*bAkY4BnR}M|;DleUbL+a##+md@|GlD9>Y=`&0p`=;{ZXDy(_HJwX)0-@ zx1pT_>9p7Jckpb|JX~dV$9P-dflOa)W9B)pGf1IjHqgNP;VwH%uqtrm<>Yd;1%eFu z4F0&r9$BT6utXhv(9v4cfgfoGdq!;%j=2g zX?$n#NJ-9K$SwlFe=57)yHDj|$9mp@=qf^*PvdlJbvvq3+oynaTR%pYcXLW}kic%B zfIyQ+@fXQi*_%Q4TwGD%2cc`U1I0z%<*HavPv`2<*vU!qh1FxH2`oBeJkJ44`x-0^ zD+&yv5U<8y`FLc&EA?!<2kcu_JNQNhEli&e zjK|e}w}4B^ks#nL8cq%dk8FVQh%)?3ap`!W=&f-5D+O&rj`}CUP@YgcMovyb=^0hR>j6*LA5y+vS2{}Y#iWrTtMM>-q~Or#wQ!oSL>K%alE1nB*T z=ATdaFA*?E&^KJ@?fnhz-=&c{zrp`Ij-Un2gAv!1l$VD-H7(q%teo6!ojn*gEDfLu zC@yjz++kn{=>FWW@*1>f(E1nbwDdgml$8W5oE_QBES=4**t{KG{?r2_K{csJ_^(5DXURQI=fj>@vw2QanOjMQc+O}xmj8Z zXh_NYiyZnUOk?Ze;Ud7!4gdhy09%zV0HI#@-Xveb#j06 z&r1GXkCc_Wg`1s=hn=$%)t`FJ%$+?wglTC0H1wbApL|+*+x<^VPVWEe7PN!xe{$G4 z**Mt$Q#X`U=ufJEnw__m!v`rlN9dSA+YsU8wlX3FH*h#BIW0M_ur)d zmG%E7)poaXlXP~3HtHerKNI#Z;{VS47oiaQpT7UsSp0LK|B(tEX%SQ*_W#V92x=P^ zs~mJ9$?T+5|3Cu@vp*NyA6PQ}^9hZ^{d~6SEcgZkBL*WcC9dTSdz_7w_PU=S9GqWD zB-SUUqdxQvSL=(V`q!8cT6Kii*y6Qv<6m%SP|=5kP`%y9GtSnx zFFnbg6r65z;DbDPl}LJeUYl^Q3VR-|eZJoV0ohSw5#)mYHhf_WW<&+a*YD~%;8MZ< zWl#?4;U$GbNr3sgAohh`Y(r39ml91Cj#lh1!$zj#Uw1-|N*?dqS$1K0d^p43_5rs! zzXvkGpi+(a_k*QD6+GIsy1S^Yd3%FS$3^qkZMi5?Euw&ojS~z$Daroj0Km~EoFOP# zy0Ck0N^P8+yqgLQ%W*Qv7`I_ua##gGB{J)#~dp{v{5e#hMTp{oCmJ zJ^r4rs4p<#V5twREU*6#Jy4JZ8Nj{nT>WA$@Yj3=OUVcI0P-sc2tNPqN|5`>3hlm= zJSX{I^Cgx~3k9~i4jP*DUr*|K3QV|jQ(;=%U-Lz!jtd1hV|jS^{|(@RdYTM0{%^p> z2*r!a*V!J4#J^nDFeuG*VjIh>OvP{hdLhVth2n)_*xoGf?_s$D?fyy1>*>Gei&g=O z7vYMOH}1S6-NfZP$K57Eqq9ajyt}_>`AKvVjvEwO1%t}|R~B&Nes-169)dv9&ydJp zKwz}tt!(y#c~5q(4Rk8x#NgCv$|`YM-p2l#91;?93LG{|$->2?);uEb2f?`CnWY zhR}s8ANI5V{i1Yb7jx8Co`cs(Jx#Kn4TYxe@ot}mGD~ja;)U>b|5ve%4S;(QWL5_T z7D!~|#y->tg&@dL?O-?G%d~h)s^FtO4SaPUJNGYwP7IvV#`2ql(TM+L1$=oM3=B*v z746h0+FMzbZ-Cm$aOL#i)ZNaD+iB)4HfRwhStXRI)bG$Bv%hcuJse-{mz{Gr!dH6e=Jz^Ea zL=HaR<7S7&+FroYaTD)y|A}?cb9us zL$b`CwKUQc34(QPAFB)+Pt_c^n+p~2>1My{=;W43WDhri$f9egXTOUu=^E`k>zUBc z^?aW_+8Iq3y#oA@+d%U7YquXkBM78@c>;)E?2OU1xR?8=;ZC+U=hB+GO6H)RID(#? zcP`mGzdSu)$rURB_8(jWJylhJ?yW(CNNszFLcf-9N@Mlpn@f?uWfydij-;J?F{wYA z)g2yQF2f^Pf!9Xfj8pnFf3h6g4WYK^7@&pzQV8S0xP4PJR?d$cSd(Hi9eDdHr$IZRf$t<7S+%3kz9^orRR&TqY zviYGO7ZS_#;WonbLhT$?f8Dp#)-3($ly)A5K`DzK9uZ^yphSs`)k-1NBcP%yjFQ!e z>4))2DZp{D+VGvt_`c%YRS{3E@0f4L+IMM9yZOc<=iKF29y7DP#fd5+drxKN(LPR> z$h@nCOk73IFVy#EP#AsX2P>oSmVb(i4(s-%^A&nAX9nFx3u|btWuM163(xe3zC6Cc zBCKNU$`v_LY_-|*mCt>)2Kh>`N%hcZ@dF=D3SavaW~uO8jfrT~$t0^q?munODDc>J)x*!^kTS#R91 zljReQI=oux$ZR*eDQ)tMQnTLwL`yw1{=sd+!%!J zL=7_EV_qKm9gIxs-(HjlTxd1yQ|{=R1T^?R4-vGS3Os+R#|}625D}H6_=bD@;%1YD z6!CODZ}PjruZswtm5MQroNH!`$)c!xJuTIsQPX^WzUIaZ{*+k~@j7wvZOV09E@>#< zPtz3phQkM3Phdf)(96n>Cn%1CFr;Jez1{ui3L{G4GyKpGOZ8s7->l~R+_y1i2YPDC zD_a#XkE1EpKX0{N@3V7(p2TXc#%M_AjVQ=otE@%crS+XvT!maOZ_y8}=Q)>kvvD4A za@N`Wk%s>YR4V`a$V5$3=iv%UJ10GN++B{DxBEXsK)~NS4fBt?0VkgGL#zC5zl0&H z43M2+y7W5B^Gr4-$CZ|VHgwAOa)d4VV;LGvX1#@4k3m))E*l|NhfCVeI$hUhp9m(K z?WThclD6VNs@Y$3Gq|Upav(dSOdagA?&e){(RQ`bhCp3&cI$jr%5dlS@APzNAhBED zCXtJ^2X{}R^Rc7Ue(bDJY8hERfj;LB?obC%y9OW(L6OqlgaZeWf|FbjGMRnCRT7sP z@@m!}#kJm)JjiwQ?f4KDg#b%Ku8+$2qpMD9*OWR5cPZYUP&*$LF5-BcUs$$DlW*cj z=iPqA6vd;)Yzsrga4->{hg1-5Q7(4-!i-OA1o|=PjdcjjN21qooP=}bel~V99QuW* z2`jU^V&n}_2;RQ#9MLMJv6-O?+Dir;GP7p(w;HpbGSs|NUdI5w1gQ8OZTP)XOxhLL z2}qZGXVosL@?DMss-5?~U0MN$D8&5!K^>VRY{&}} zs_6dB^w+c0@k=#OS1{s<_KvcLvaI z0)u-1k6gRN|H0oEzF@RHy2F_r##CcNjP`bF|A)HUoBIfNsos*@8^9sa@|jhn!wpfV~Hcr&&&aG$wss&HagTie-jHH@;y{R&~ zfy2_7jEFVC^(X{+Z((G>|JJyTQ?BTCAA3bDw;>;iY0jIsZsNEsO~Cqr{%l~StPhvH zeLplR!?^wrQYl$9^koTtcA!Jv0|^eE`b=_AT7F2DoL+3SOW%LRI1(+8W@uZZZtQ$` zj{FT=S7S}H9{#F6S04P<%MaCa0>e|ddr7X;Li2Z^!7UGDyeIrL70X7|J=|<`A?lsw z!afm1rQ|F&rv26!kr}(k$VLmE+18c8R8$m1%qxK%)vcH*;@)^g%AwT+>%9dzo|Bh} z%lTGJ!2exRZg6(%;o*(wK0xTPTIwqMIiOYkKvuDj@nZg=GqG~@9#w*|mm?PKRQf#t z;U@u3Ds~(hSmQRQSI7Sw&s58f~d zypz5OyomB@?~e1G5N6OOqbUOi(N@DC<)0Wd9jfgWK}Tpw{5ZH-e*K*}zUy(R`dfiH zftHh-<(+vv1_bKW%m!O(D`uUTO!^J*sp0tHkeTUxzrgLtvB{{+eBi0wu6C&Z2psp|DkK3_f@hF7 z6>oX+o0xVxIit6aIMK$wgC>}%{R^n9hIMf~Jj};;cNBMTSu;Caw+*%|`$}DX@!B=l zi*U=N-}ZzF0824{stqzkKK%n`%%bY*DQpIGfR(^Zhs9=^ZL(a^r>kfo{ZsiTiR+o9 zTQqHyrpkm21M9EFz+T>NX>!MCcyYy}Zm=z67P)1YMIfVG*2NUQ^f*>WR_e$oSB z-p*yYpV12Df({$YD}t^l%&B8Z4_{LQIeOzWYVp4x?$rf=uNeH1Mj{_7SzsUTh$(v@ zPvMFO$XK?j1+txG;nQENaZzu=ms5VAGPs8 zeq$N9>caeQ_`skL!Ks%tMj77DSI9gPk1j8d3w285EcXENXrv26QBzOqDLY?`eZQXo z%RruIv^noT>SX1|@n3GWHX*F9oPD`4d@;C&(D*+X?>q(Uw}EL@JN*1(CBFx zv^NCT`{{4U%QQy%uljNQnDDIbpp&k}dQR;Rx>sXk;^y}4wAUQ%WAR^>xT)=@lsjGF zSFU*q>{%u?1|es+nI9N-*-dWTk2QgYaOg@)I_Vp7da>LIdYNZ)U*ValWZh~-JDH%u^{)_G=+d-mI|IF9QlqNc)-e|uC$50U-g&&+Gx?f&q70`hMAYuW@j~5b zv1VD)G@Z@`{53p#G;{sL_~EPU>)q2%hR7FnA!Ir@c4MLi>-H_Lju_yLNYVFdFhwMo2z#pcAT-iNZUxiNWG)1lM2O+OWip1^Q zh^@DoD7n~*7yYO@3*Gqql9Wd4KhwwXGOp4SvklWF-4*P(+s*WBlU>G|N$JhrBAw3< zKb9T-7=N!2pPb}s@fi77g@M~)y+#m|udxJ@mec|&ib z0rLG8r{C5guv`mXS42yDL+0L_qjUYK{?*0adH9wqvp-nVI6SgbG`}-a*3-v*6;K5# z!#oRExg4^&A7wQF%sl?=2}|lcX*8fSSzGb#T7E#mjhsdtd!jXJt-n}@e|61&Ua&Pi z(W!P_ty*h6#&S0UA%>Uj;anj+{9>MLDyz032X&tl8Rc+7Y*I(4_)`_t{AfLyBW zI#RTfr_-el+2_Yg_Msf%^6lmawg8eDw0xDH*kyfb@@N~Ur)}Vw%x=e$Vfcv}KX8jd zB&B$Cgz3W07ROBLXxBI z_mg>Gt5{sh^Hs(VwR#m6R1!`L_1ZtN?ly>lZCe6aR><{EqM}=hx0PBoSyzWs{e<`o zxEpOM9rTBS4>f_2`tQN3VyQ(b}^AO_~|}pm$+7!b$4eVtLN(?*_Ls+ z_WEEM_(>__ou(*Zqm%xIfP;TCn14{aZQv4_;J9PSWD(x+QsP!-Z`{VOE1|0EnSQ(O zLSdGDX|VGNbO@p=%h!-kVM#BvO_YEY-g^>eJ|C}r*izkQxOT#uw^8B)(>aAZ3sp6O z7wBXIZ{=6JribAV0Vp{Iw}I|)U@_L*W%+;W@?V64fvWN|=@Y^oF*YA(T#`-veFEBW zI@N!^q6}oLOJ2t~TI}lD*4U`Zj(~tvMW6MQC%m%m&YIOyyv0t$pXj45C@d3)CYFl*hfTiAsClDJx;5-t!Fx4?INM# zf2sdI0qZ4m?HAZ;6K_1wzLWmWCa!_}(xTVkwuBP+Rk#f>!aqxSepNSd=&?%~XcBOD zw-+J~+kEOboAoMjj zr$f4t?pOV1ZqI9kaz!;M9!r2PMo*pos(z87KG}+EtAI7nRDa-Yb!YL=p`Kpw16X(Z zP#Utg%rfRV(6FYwYBiE$6iF)!ngoCXnW0=1%^9e%FW4AVzXKI5>3L~8r40$HwTY2T zBdRb?+|DQ2L<7K2WG(TZdANv3Cwx;@Xxt`72%ZmBfW2PF~g_r7I!xYUh{h^}@$ z3pf{|vx^^&(q(+-vmhl3i9J))#5LU>$a#34Lp_ip&D0!$7!If31629qdSR>voeE2U z!@Pd(I?7yp=CA4@eF=IYwj9(mYrKvZ%lfxYB|?E?kLmP4k9I7InPrBx^phJKTJ&T8 zN7iL>$VVNv1euAh3h6x5oe^Y#3wUB<5{2X+71ML&1_o~1&5D3CngFT|Qg+)jjef%~ z-jhoaYUjvk_Gr2{fpy2P&^&(L>)5w7zY&%vD=m0E$NTlg78)Ft)q+Xf{Vd{~KP@1;J7J;n7xe|7gO!qr-=ec2crGVY z>L)W4s#eWEq2fOa*t@o8e(Zg$s_L2=WO$cxX^vpmCE{$$De(2I!)%LynCHSVcyi;t z2QC0uQ1Dg5JekwaG^ho7HtjO$H=CUP|0*F?(^QN@0RZSY%8aDA+V&pXfL zv7>cd253713}W~jTtNr39d$pC1S+l3K*+=$1{k~D5Yo^rFNY8K_0Kp@u0nZdg5N@sMJG)~V7g68hcJ3@oQ)Q?F>BAnqv+W?t|0CiqaiNlAwQfw2-#mIiW zh&|5T#=!u`zG_&`oAe7_pU&WVyQf;A<#lOc#z9(zB2 zJFA`P$=6~hi#P3WF1bX)!1&Db1Gp}4a5{5+ZeBfg$l5-zlBTo}n8CYt$1`~+f_2UF zra6cTG$VB_Fode7UkOq`3ZDbo4!VbFcy&>!1ZWIdJ#Mbk2r zhxxKHzWW=>x8SiH(yb1x35AtR6T8Lta~<#UjtE2P@5}GTO2)iRWDRRtR(wAq9^{{5 zJ+DDVpO7G`Gk{F5s-y1e8)tMO3l3@6{APrA>}!f4tFlu!vRzMchouj~GB{^`D4`rS z6Y1=sk)1_?d|1a1d@>}J|6~0Y)1Y7XOuNJa0Zn$tbI9RQq;!ddy0Zx=jX_YEA0|jS z+1A9S(IFFC_*t?Vl-CofN+zhI{K1DuriL0pw@O>_dzmO#^O{Pv`-2y;pj%;sX&W*Fju?iNR*5=d}Db@Mr zB#FW(9|&r{y%@1StQ@>DySbSc@cf;B^SMazTI9-IS=-YJJ^vn??)l~7a>6?4ISh|v z_|d=4+O2~U2@`q88H2P~+U7ybMcnpu_lJJ}Zb?)gFNiczAkYM1G@d*n%gZ(iqlxDSM5xmDR{0SsDbyer-X?ca8h z*p1(Lfja)k>V`|N4zh`j@hOuK5jJEHcq}GIVoZ6_1wF2xbK}#l(>RVi2`VUYvM%mg z{}z*>uZ3wm8Mo8*xgMJ!1>!NuEeV?MZMxJ>ApfmpZ-HU4rr=n#NVw_@=>O&Tnq^r4A|6dIb?H zV#LGj3lUg>yRXn@_yvA#sAGh3gxmzDHcCczL88nZ^}A@RIxjS0%ojWlE5Jl3@D&TA zh&M-Uo;m024mlzLGJu@kBNV%Yxdjj~SM%~xsJ!E7wDEYc_OPXw)4PAc0)OOtH;JT0 zRJ@n0BURoAsKM+fVtb;aItE2P@bNtIW@*63)o`z};~GuAt?@_w?Z@HJ2`mxA^@mDx za`u8xYcNv?A+YG~U&hnksrZeUkzbBwV4s5m{lw<61im@ASBww-O~L*vqXximk8YM)=~m_S75DhcTVS*y+G```> zf*5$jtoF$({GEn2&1j5220E*eK7V!e85t#h>keMowf(h;ioH~z)aD*lnq+W8RAL!_mM2F1}b|9mG+d~@&2R%0oqAig+1 z4#e<9n$sD13v=n0GppC8-2|PSlL(;RJ!9KEO{CAigR;*A*!3&2tY>d#Q;Qr9G&Hn@ zP0Th>K?5JjCVu(Nz~FidQflQ)+^VC8?s4x@07PEkXB#e!dF@9L?u;5#(Px&ba-wC# zf@SY;xmi&WJ;hx{GU)1VXdbnEA~MDKZF{HBvl|hM@|1QTUMp3L3-I>vA}Wy!dl_O7 zEPVs-mCVL-cn9{n1cv2;>g_Vy*M7tocJXK5guSR>6o&#TYMd0Pu(;2L3%PUuA>N(c zDIU0RH%Q?-iRJ@MoKX`*(+qC)77*3wWb$R5cCHMGy3v<{Id*C z{cl^Y6TCgth!Os`-S3TTZlG$nREL=weGWBIm}fMJw}#RxkmFTYjoj%6dMw^tY;3_U z3gIAxjSbEO%=r)|mb2pu1mn~rXUBt?o|G9mh>MpIzRR&fjZsrtr-4W_LU>l z*PAQeUJsWDybnFXC0$!9s2~Gg=ai-h%-Ij`y`O!2YVEa6yiAx?S7AGR4>|=uRq>y; z$iAcqd~5ff_%0%{uw|dKzejc{MJePWMrGTRj*M8t*OuXY|7)YSwTtvpv&x`B0jwyu z@qj~T+#5}jq9T167MuA_8rY-Rnu(jdksSwFZFAdBxN2I-AMt-&;LP65@dGF4D8vl- zb#bDT5iOL&pKkSfc~|!?`+-YxhrtSH!Di#<1XZ2(n6B4R!HCH2nAz|Id)4`tRBg+hj}*jfOvxSf@cPt4*mwyUr@P#?qs0lHf=_ zhxYQ`AAmQssa0C?H)W0RPU&^lvyMAWChU;1_HT&TTK=0T0c)mKY$b^7Y2rBa&cGhX zmse<(bMHI`r$s^H|=(b$G0O&&!Ez;?#=_i><5ld+U}o9sv^DtT^}d z1#~f4x2DY|oQ^Rj+=5%>jpZ4Yuv}wQkjfOPbY^Z-)Df<+{01pS6n23rcp~HVo5CM0 zFrT>hm82KU!esmb1A^A6r$lA?b)5vku65w)OV5*FO9?=)yZdB9qQ0DX$$d#1@4%TQ z27M30a-_;JYO?*line!3cfspC;FKMFsI1b+9Qjo!i8(4u!dXBaYT}pO`}eJHX2Pzd zy^l8D4BF=rZnE_>_D1{1TpaUIkKnZ8Q4DADI8Wb6Y6Ul&jh<$?e-6J9L0&?p6O=*; zMJ~b!wGF{e+fN9e{pTTsBzq>%n1jRT)qTC-sV>y!$!H}NJM+CL31AEPd}DkKV3Ej| z6p~^5rF@iPk7XHlAAX0|8Hq{xLr8`G@ab&4R}8`Xc5?&0Y7GUiWEgtB+n~ndb)^>U z@-b7FB?6POK-gqbhZD3r-B6CQgmk<=WufpIXcQUJ5AU(j{()2I{6xTWuE}UlY@~Va zi`Cl^_AWAv`^j3t3C>cO?COEZvbd+}z1;Gu4r6dTSEjLSW=?54afvcewdw)STuIiE z150(UA#Y`NR;VZ!NFm&VZEkduG3%dW2`ab4%4qUd_F1D)heQLf6f-q%2VSFUe-|$A z=kO+#7zN(u(X=KJ!@caX3Ko8ZbY{dxO@3qVxrdl0{`UQ;NTtrd@Px1^wS*X}7U@UX z;}{Ar6g8z|ihTL_8Pz?y5f#?@`tu#9Xs;M@GhjJYm(`+R#LvWw^irVw}`LLAtF2U7BS?{LZ`Z>M_jj1h+%QocaXQ?DT5 z1Sin~i%E@siMT;9$9o5t(U$L{5T*pv0xpf06=V?O8$?J4w_5@M$dv`L$x}CU^(-#L zk7S18HCq-qm< z{aEZr(a)Tp7#7aG$7z=2!GV%j@%l=dp=w8^ToyNVM34|N0H~<9+hX{&yC{#{n+}_F z{}q4pdpK`z!CbRdzngi8G)VHgXL4k z9^L6vfS+fwK6m+T{?2-55Wq-inHrm2C)EGg%l0&PzWhMqjP%U{7vg-Ab-%={h=qtn zyD$4WfU+NgJmF7-$mDtQm=Op;a9>H~q%vF2d^5vsR8U;PSJEZ;x|5sj`9Gdia`XuK zDko}}ezJlt(!TNmt)7=hvyOd^z#iW5MdhfH0cN4!@7BGWdLWr^!)_Sf#_(4tLQXZ# zx{Cz)M|d3T zU7|rI-c4EWow6znI`kTILdYD;ES_PrPe6~`uL9I2rFSIntcyJaekH_lCs z*8QmIrFz}^jgHK8$$Nb0YtZd~<;=(vB324n>Tnsa;XUCH8E9ipP{lFKa>f&u?p>4G z$$GJ37H_3Ni4PU|($sFQFADua;2px+}^p#1osk)T!=9= znc<-?)8YcH!V=0+r}TxKxGT||qOaZV#7 z{{nZYViymIa{2f|@RIjHf|KI_E?|iqqBtvbcD5fX%h~>B<3|`rvUgbxhB;E2=~t6- z0_7s3<&;i-xWsX2PpGq~ZF@&%?DjqUYW2xaijd+-tAlK+IN9|U+!-IDRf62RA4>fD z?e*K+yG#Q=Z~%&#Bx1%F#I83K$xLB4XgPCf(rDn|`v^+9lPK0UY@?B39d5URyF3=( ze)ukNVX@}BjJL8K%%ffGk1q^<*W)M|)HvobgNWCfHE#`!%pOc{-{{wGG}q3RjygN>B<-8Q#P-Xw2D}#p z7me3H&rLmNl*QT4ba8r@rf6fGuB}d6Ze6e3%5U`5BnBT0-XNk# zm_F2rIU-DiY{u*SrWgr^`muNIvMvn|Yai9iwaPb`TngXRK5YuZ#=AXy*lH(mCtslW zRj@qT75)CUk{E~bC=Y72rbe}(LJb3=iv)jytHFW~&C>ay6Tmu zp9+q2SZh^J$_9B6`e#Q7`*2-0v>z>J1@7C=WQ=RBdiCamVGXgHOqaDk`F`4W8Vd|G zAI9~|Km_`%6ib=fkl!g1IvCLQ@Lm{2z@m`19jRbcY_(lM&jha45kJj;>M_gVUjR=i z{qk5T-h!yre~qFZW$Mb4Zi%>VncddsuZ z`-GD(d>X7m%4_yHO}zZJl47B0Fg@((e_V+ZEK#JM8s(FP1%BJm65{+?CiRKH?HqfS z8+WGshz7#0GcYjgebIT6ADr=DMZW7$45y70XCcdz(!L~huFOy**(kvI?VpD0$Z1mV zct*lw)y&Fk$|6AD94LJx0Gz6FK5du*}cnak-9p%AkZ z0YC~3mdV`PYK)AZiiZLmHvGG36lDl@NwVd|%CFW&(=`1T(m3T#xm-jl`D`@TPFl%zWXf`)DBoAoIMP_nf@ng&)Og7 zw5F+K1d|qgA~cPYy4offThj8rwiTKcyDEJR zR#ksc*c%}k6yEg5r_X{=HHH@zL3T8eB6M+7` zCC3E+wuO`g3leVURQl!ho?ty8{qMyHe$S~p1l41F#R6g0lPA51UudvoJ%|XHe`J^@ zOFe|U;w9!vC+FfBorbYbO^BuI*}tvIsoi`XoOQ#~wzY-~^3qfDetCl$Av=@*FMKK!xC~5f$`4rofm$_!GJ&k`No}YF@9Sh3HAECnuQ)nOS>q{u` zZ3@Q`y5azZj@UcZ(oyk%ic6GMFXHps+>)PfmL7=R=bC`5yFcLt9lu#^!&V(cB^4jR zqI(G-D`WdnBP?FJE!$;5W(7~1jYQ#7%3$$qr+c-KP5ofv16Wev8xmjH!@S-dig<+= z#VL)@?qN1f-}R&4w`21nH}eJ!%lK1G_oyL~&^7kqQto0tSJ;)9(i8cS>TP2e=!XIK zygKZ$g(Hs)Ox*}}!q>S7#YJ=>n&FgukkS?1x z0tEBB3l&p&d?W9jJ5`?%NKI%O)(5th(P^sVd*9`d^IsCozL_NDE-mN?$+p9e&409* zy|zXlc9S7>EpD1W*(>P~p1g?rQ4X&{iMvmyd4=V2akxEsy+PXb-|@$uW3cb zup}jfsI60pM9Dey-B11Y8L6}Jy$KgegQmbFdjMA7ufdb^U(@wQ=rTWNn2R;L0|^AB zUE5etJSTei2V=4sB_m?z$mu?XSYY|WMr&9xAC(1XsV6J>bofrlJM=VV+7xM$LHFa1 zD}_aIAt4~yWi3*)t3Zpxq38Dy)usue628+Nzo<)*&j)rCe)+|0m6w3@0E~UAUKmp) zL8*;p`d+JlJMRW>i%}gm#pW2wP&qqWI)$BxQGmQi*ikLeKzSFT%^30STRSmVXJ1$A z{F;RHnyWGKPT1oy-}*!1W#epXaLhRza&SlY7`w?3sh{FvYO=kLCug9_)V$-Q*aT-K zn$6RL?S@1;zvuTHMZYf>r#$(M$x_9o9GlFktkan?U6*l~HgU2 z<+vXcg4T#5fn7O#)IYTMT^~hQwCYPeyq{`Sfp#`LQ2PLgO9g zrsH_~)2Gc)%TqgmQpa4hU6neadwP|}d$KzT6OK&i(X2ZRj9WxGZ+3-KOPbdPPfX$T zih`xAce$4o-yC(6nV^{L`fzS8?-D-Q0@Ufeje?I!a5I&^C<3-@Q^~$2XJiL1#||<` zsIqAq-!xJy`9Qtdu8O&t3HPxpZ=S_QOp(Cj@!E3ZaYnv};*4i&ve`@40`~tDHU?sH z25p@uLSwb8hvD~Y85ULMe0~qzTz*6#qJn?0MGwYzaYhxcG=Y zMO92~d6*!cYSQ^gk6DZ3`B|KwS*KDxuc)aHx0{m@Iiw{0x!ckTq)V-ar5TDtz@q=% zbV)zh@I*7(DjZqW z)suGQ4Ru3)=QXMZl)nn1$D8U!o(jafU))!{m3g-{h6?Li*IGhagZZ#djiG9X4!6|e z8H@RH9DaroKX2G_+RJ2~e%=~^-Q*K+g(Z;miv!h56!~?_&p4jVtS{Z$@m}+dMpyCt zVK_~14Aa0#n$?_x8LGTJO`^_T?*6BbSVzH(X&Eq~I9U;Azxx$Lih19V3HH>saX;K0 z?56e6_-EFduO0MLvg*QJTa)K6kIqIsVR!gJj%l=WTgi41m-j}E7_pB{pAiQR$i<)g zcWwbQ>$%VHkq0Xr98*g87NXId(lIDCU?E3;=%M@GI}jZZUMYbfpRdo=qjBBOwhPUo zOBc)^C4cMnP~ovQfma*o7;mE#9>zb&*$^54l6ku2(%f^(IZs=7FRX^Qz z#sV}sUN&r_sTFR58|xT=hH&7w@iZe{yUxDMyVQeDH)!xI-iKH;ntFM`QrJ_(G>hZr zDj~bPT^7bTOFVLx6f#}1W8WNRIzQB0<;NZss{hitzAx~+t^QBfavmLL40){t{$e89 ztmq9MKUn1h`_>8mw)y?2NytXsEFFNFJvLkTqWqIRC6D~-XgPM|CTZjTNg$fQ;#B-o zrmK_k2r}<1RsT~Hvy7uvN_lX6rB7r)EJ!C=YROZZxW!rgt_muDwkdyBAG!-*A)^@l zXD?N?miQ_ggwb;J4bQ@)t|PXlUa+#Z`!Vh`@OlNcm7xiqXANuDt18v9ceBdVONX>w zRi&~YHV22Vrf%_cd(1)zpq_gdueQeC<<+`l9iKE+MKv**+0)UwOiec}7g~ukgn)|T z-kPC^Wc=Plt!;S%xRIPbSk1gMCejF%$GaW3zQ|UdJd3vEe34r9Y4c;JDgrc5OerSYa(etCMzCD77-1A->W zhR1G53qe)6IK6;W(4C&+B8V-xRWQQOagSBeCgi)hCCy0JxaJVWli)aHst*YoL%_wS z2CIHiJ8T#-M8H){_F20W9^?p}e>4q2BCFqy;l!zk+e`+oRZMsRxx=3#NEXzC&hJa;Dn^wCeAEbq*WKwK-$thbpoZ zsb5#*8$2p~*pCgL4XK-?6uDuTd$uQhgZBQR<;t3A3Q&i!G5SWx^a*+0y}f&U)pXL7 z)z>t>h$`n0(Fs`8iKjn~luSWfXkHa)&@a467OZMFC~n@<8`dm0tDLWPqySD zwji`0gC=`KM6A4^cG<~mx=L>1b=gfJz|BU!i7nmaZ{kC8$Kv{%}bg{hXNv zKFiiUGZ|@p_|*-sO7*VbPuVIYaqd=G>E#6_)f-X|5emrDKBqwO@cl3*^ZBwUtJ}XD zXJgi6YvRL6dko?dUuK#O0?SXZ>H06N=c?6JFMm+Ib<68Hw+e(<`n377u(gZ8K0>Hbm*Vwjoh4 zsb`wXF4|GvfF+iHoknXY#)|>Iik;jG3c8>dQDQl&%1rU!Lh7_GH1PyHPP_zrQe$y#Z z)KHsA0?m?ZH^yi7y(F%WKHKerEO=B`)*+*7Qkzs640om}#c!Y1QwJWt6gXWu8DPKm z_yr_k@R(RS@i19S3XxGbXN%pjH|^6QV1tm=t_M7qrE6@JigKM{Z%zu|9(AYQR{N=R zcJeWB;1|s05%3)W@8gZf7{u7YLVS<}!V{{)-1Mq*pc`p~<38xKSt-WyRdDUGam!S=nU z5}DHOX!kT%>gFTjJ z_T{4XyU9b*iKa^9G;O}ktOnnV<+AGwb#qSC?|AG=5DeJ*P|oCo)6G$vBe5+OmoAsp zrdyZ3jp1sH(HN&)9xRx_(P!>v)16hRC5n;8lI0L0D1euK942LN7M#+z&+I#EJy+u% zUZ<6PaRy!Rm@R)Q(q;#9WK4jYxWLFoeP`#nI#<1nIO@@7n*7LooQGUoOvoKK3K)x+>fp^ z_44~W1Xgx(V8}6&O@%+3i0{;@QJWjK=J^6K*;g=74?6GDT!SW5HaVT4y{Stbi3_ZViH(l+J zu@(-A^H@Z}6NFWgH4XaWJHo%{%+@oErlD@UqZiT(wbe-Mw*UAx1_31f!)7_%;XFMfU=JetOL|k@-Y{9lM2S7!Yb7(9wn`l&mHJNu^OwI*kfdCr)LfbSkyliMp zsIRN;p$YTxQ>{n)q9qXvivSHbe}X{ytlKv8(9?>#6CuV~irERWf`>g9>S0BLQ|Fct z=25+Yyy8tL?(JB0Q#33xclZT01A9*TevilViB^wKA1_nfx5UR3OR2FL+B(eilLOZ< zj2!?xIqddXs@fM6Mf!%H{Z5RBFl-`uA_8$-@nwsHm|5Yo>}Y|FX}f@h?I<*Yyxye$ z&ccYib^3s1c+X?P6~JztulzcVJCP47!WJ_PqFnO|JN+bXQ>M?c2*yLud0kI{S!T9b zUKvzz=cm&P%Z3@b@ik=a<{ot3yy0#b4t}u*y;L`7=JUP?`H*Pyc8D4y30cz$P#Jh@ zf)s@0K6m_5p$hxGoJ7Z7iCQOi$FVcJ062cbSupV7$_O3QgC!6MyG=6SVNNv|b(_|HmV0iahi&y{kwPe<|z*gW5L}+)<4cHKb)*&xpHY;y#l8+OR(4 zPS9MB2cmI?YM{-5JtGsZ@ardiK;}3hrjv%y8<;--8;I;s?Ys{7+Hn@D!iwpxfjcHo zCFrjt2%6?PCGV$qE4`a?j@Rnm`gR9Jk*s~T#S``rn9;SiG`*hKYcu#bU`-O`~%T6;t zG5Pi=Tt6-)gdjY3FXSeQPi!7<&Vg+N4WKKsr1A(C^qh>#*z2fjap>^NrNfrae&N_d z^F0>`HfqhD_+?*|0X@$c_>t7ypIqa=RqEUo2g=DgAB4LSIzn3J!$f+{TaXucEWXP$ z9*xcz2lGkK-_AY}KtHS>$T$}3PmP@h8dq*0(m|?)HT7Cqcc9c&7)|waTJq*M-7-im zm7s4H{5^ftdpdzxM5QtWG0y(7mq|cD7_HVa2Jvjt48h@&IY~(BT1aG|X0R=2|76@Q zY|qT!^TN%*mu*gbMmolaTj8b-yIKcxn|h*RJi}BQ4pm8EU!SF|=n>+>XjpKzXOeNL z;zRGFKFrDy0rM>yA5pCAf5ck7ovmzDhKJK7oTQ*Bp!N}8nzw~qY>;r>7X4%)uWX&Q zwnQWdcikn#!4Cj(*-Q{MoVw_M)qnOygzk~N=?CdEd11PME!)62hxV+FGW6U>7Lnu- z*7L*F;l7jBhHlk33&w>BAgvZxi~I~g8Ly<`6&}(#s!NZNSpYqKWaAUP!kWR0)irL# zo|9qPoHgT3;BRJg+A>(lk)%_>MZ73m=3vdnDJ7iMj@u!Qj2?aBrbpU-4X z`*>8u5@I!}f&~=5r(-YRneA0_cm?))2JlK_SoE{ob&6HK5;Hte64$87YG`#qoms1| z75}||;xT(U8X~H&@<5K~iCIORDvjibdPR!o?n3@Lez$VCN2MYFT*62)LNJ@Ky zI7}RkO6+CzR}$~yui5WrVm=4f6zIZdy?$E0$Cr77&d0wP+V1&`!BrAZu(C=rITdfY zw#972Bou4D65ah=#io85Hp!chrn1WVbj#$9uIwINZBHFU` zPtJjs;xzc7VfWPr_cMnto~>g$;5JvjxGaDYFaU<>X%MbtdUX8V_B^k%pYN9)?Fqd~ z_N|lAx6LuFR0;MTIx7{b0LoQJ+8hLPD$y6&qK&{ppbE$toBVW-y}=dUL~hvIt!BjL zE86Ci^(R7PllbiPlU&xh*w^i|tw`=8njs5Xn#D(6;hx*0tUpXWGh{Jgy$|U|kP%^R zj5OKs?qn`QCVnn(kH_AE$&jAI37pu~R6{d+gp%K8;?iY3I!3p_3@m!;d+na}Eajkk zZ@4u8XaAx!U+Y(z z%r;0S99-}@S3B{Re^k#DDeOp(>=0A31ZrF?-pfmXp1=sI9AJpgO?dW2=fEwyeTp3I zr=p34WttrwbtxyT%@ZmB7}b5suE@zUg?2#@`Q6#OWZHI-6bE6RI_f(2H#6=nQhmsr z?0*|fW< zQZ?rW!9V6b>Syn^VdS`M*VB!+*t=%$zR2oz1d}hda`H{}WFSDLx+P<754ZgeD`st- zdI??@#;h9;U7W+2hGFKQsiiVgf+W z4ad5rYM*N8C6HCR{f6bz8ok_Z!&O94DQOpyk(Ip)^Q-X*HZMghlKvX{8#^CKtHx>! zpN+W)E+OQ3A;Uw>XJ7Z@9m$(kBe-xxFL%T}Vh(z-Wi(v6lA_YKxw$pdDm$k_ej4k_e+kLqwG~ti?~n#nccU<+;?2Z4KPi|wYY+q$ob7Ncm3|z+%lVB z049(S0;=3P4W|!CFyeo$X@Oc9spn46dWXVj@a&(D@wBjV_UnEgia7J-K3`4A*S;F= zao1v)kgYqlp}^7C#(gHg@*qs)R3T#aAIw#ez+#^`wG zZ5@Zz6YVcYA-9&TjB~3sAmG;#Eu`GBo|6);H|2^^L-0}3>(nH$3cIWbQdc%soSU)_ zeg0XW`~2W=BM_CD&QJHLZfSvd4}NJ=D-3+;K>L7Z(_9^X>9^|K;Tulw<*SuL(&LKZR3B^(> za-2UpCp_LR?M8a-&V zI)BdNa@R#?rtU(>rtvt&mys|*Hc41>q4CDFG%L@fpmmk31OWe%Xl2j83fC7~tLb?} z5R#_g)3jfhp{eT-c<}5}eE&>6$&jm%`%Mkxan7vgPO<_WRIGUQBgx~aMC~&?l=N2F zrN_R@UI9xDVEqCni&qld1(uR{A=hc%iILl4Ops30I#c0K(=NTK)4vvAAl+lE(X z^rRq$1Jc}}?seM2P7+ew?d)wE#I@=rr>U-tVVy$sg3~4pavu4`q-O)2Sa3S4_LV22 zu7Kl+wxIo7-a1i(5aG5@;y3F-ulHv#mxw=|>_< zauO8i#}&}m29%}H&78s-(PBYvg$#27?WVJUFOlCkp!!C=5hEMY;2$vYdcN%qeJr?= zu$ww7YI79EkwJnTf-boBFtWv`^;7!Lv$0jDdL37~I-5GokWene73p2D1FBPlyWIiS zv2BP~zeQ}!&y%e0o9c%qvlC~u*EW-+AzA?sj`}?%fA7+*pD3TzGs=nq5J-s7AuUX$ zwy8Tf9B;dXgh%ng8VR&)((ZE$?z_paPlM4-=&IkwC|pmLoCcYz>!(V>o8>lp9v*&{ zG&tv_6z4O1B-FFaAGa9kXIdvMbV{7r*RNz;D^E&!QFu8Fq0k-4pS<^)VQ8Y9+0WY} z`dGY80y1~JcUg$1_VKvFWaeiiw=rE4RaEf2gtDZe3`^O0!vOxH6##LZ7JF_>&v{xv z_KlbR5`tfNNgMa8EJNb#^;iBp4jqO7IM^;Xxg#}U>#>J08 zb3U=RRmP2GmcJRZdyu`b@6O;Tn}iQfgNWcJM7y)Ljp&P}_d3MsPyMYNV}wo#X_kEz z{+HXa{zbiUxHB#JF>!O{Je+%Cg3|mvPF%c{T z0aSIjL zy*Dw4#;TwOp-zNie{_$~0sU98n$w++*?ScReaYJOG7n+43_Km|SDpLI-qZ`^yhYDw z*rVDy>RS$w-2iS}1Ev|Y1wf5>GkkLhLgC=ZxCMZ|hpT}+dFmzvDak@}u8&FY0p}6L zTbKe72)4`>0dtgp=~|GW$DQuOx$Am=>5t&=WkkpPZ+@-JhBdXZMp5R|idGRGf@NO| z((O*6X@I4$}0p}WK{lF*yM+=5Otf-Yl{n>6In ziB0`RsBk*a9xYoxg*w6Wu+$)C1Rq_d55_OhP=|lBJwYS32-=kmw+vMtKZd|JqHIqr zt!V9t;Ztz6=Vqa^Ax~n#-2<6DKh@3vgKic1)-Y{3Hqu4ki4_p_dv-&NAD<#(ShS(y zU|`qr0E@+!abO33O=-PiIEYmwBBw%6&Apv;lp8ly!MXeE^aBejIw6Yp_jsoSD_=n* zeX`xajNfTA#7wnBo>Z?^G}o|eY)2~7Tx-8+5mg{~OcZtvY9-x`k8lT)ZN=^G#h)+p zJ!A*!tEdR7d`?-t_x1y?^Ez}y4y9NTcRc8Di8U$EWZ{==?|aIsxST4%$TMoN6fadFUuze{QE}F5_Q1@rqd^7O;d{fL2bLs?z zLau93r8h&iu@e2#(Gd@j2f+ZWbf5)0I~ZValgxD>Egi)$gXcBkYziEBprMG?G5o-y z?Pp|OZ`!d|hS-62_vix`76D1tEia%oXMNRKa?u?03{bmJ2+wIi{}8fdj#j&P_L4Jw z>w9}mwRzULneX|xvGUuzAImhmW%{v$7tJ39BiiuGChdQlW_yvDs}JMaRed_oCCS;H z{4{@q<3Y<};)?|XS|r_;bRMp^z5x?V;IrKJwWACEFIcd%y_dd_!fgKs2nMxm(4%kT zuJDstP!bA^?vs{XI%(CamoFxuJ=6$*7Qj}drc}Z zpVQr2`Fi{L64*9!nP}HT3g&px0ZZnLQzO za5c{p_D^sVG!%m0H(8cVe@{0@b$NGg@BUTeuFAt0N82-0WltM}x zgrS}(CbN@D7a80Ng%R8Q-2imtYX96c2=TZur|s}ekaQCx_iUh<{IOj$jkxI=_p7|t zhBMk1aB2aWe*hRg{w8InBiat-2ocK>_3@bMS(fVmh8wHL-CvHp4nEk}D}V$skhEKQ zH4>Sr!Ek=vk6riFE8d7rdhQ6RtEOFOkJYRAD&BJa!ZdV49DL<<4ybX&8B>czGPo_b z)K+5U-~kYgsguhR9ljS}o`5`5a zGW<7xYrSU(O}eM3fp=PkT?)8`A!9Z}>C6GezYlmsKi(TW&i+N;&NYLHvpd)?8++`u zC9cgN{3=a$HQGba=mqwfI(UY<10>|NF{vLEK&D5oj-6CQ-hSG3gdbV(eFAJ((fkM~ z;XPy$QfeoJ&aY6br*|^>cJN=0*PHmJXzj%38NNMoaN<~eGhO43J|A!E(T0Mtl#Gcv z4`>R!?kfEz@~nyflQRz?%alYWcHX?2idnuS@$ zY?^vXe=@8pf$UG36YD;|o5;!mG?|87XEL-LDW2RLCJvT14;`N?CpcIiG}`g>4g#ca zaw~AN18{Q!3_R~;@U^`|LegtVj>E~j0VNnEKw4&Xqk36H2Ia8;MR1V^qq|tAgB69n zt1;LZ&P>hC<>%v>1Za9}&EKq@g^^3Ncc#BnkcohqqhY&CN~Z%f%#2ISK|-MP6Z1kY zbVgb80r%{W9;o9EyUzE5@3*+x7SUV^@;^>AX`B%96pk1>y`Jq;a9*e@0Ng$Zlg9bs zHwsZ)?vgD3&Mhk^7*>pXc0}qzuJs|>hNe( zk7}19aq%crM|ZyH`wiX8z@1@{o9wdlH>nf$_Ei}DAP0DeRBWsvsrBp5H?h`YxyAj-+Y`^^kLFE51>srS6!e=?V8XO><2h z`WuTSAm^A$aD3PT$sw2O8jxFaO&jiM7qTg@FrrQh?;_J-^czyLZ5KL|*KRc{D@orf z^RCM=OT}9c@h;+n5kksNwy(n;eT;Mfl9n@KXuO#(G(|PzCi_U6(6ayAqVEZ!*u`3T zbNzOE{$itz4gJI~is{)O&>27v(G>MY+amu~A^}qIGr#QIC^SZ*LacQ^4L)IXoWuY0 zbE}Sp#GX}8OiAExO5fZ@LNsxpfOQZkg@>@hI5?Zn(G0t>@oLJsx%N~lLiP8;nhn1H zAPHsCTjs;Ceu@7w$8=37Nd?O4@~Q{2!`u!>qzXN4nP$W4dUw+7c+q*RM6&ztpq~V| ze7Of*)f5EpVZR$nejl(ve2SNstZo5~L%r~yCj0U3AA9dK4#xo*LP^C4I*b8~A)=`a z6dzlJ(QQe7!&2@uuMMLho~E_uJJL30;U496dsj;S3^PS&fjv=5=C}6d8 zKE~uK^xX6DmjhpXn?~9`Qfv8C@yoy3{HbHJCHM~1^EoE$c)J(_CE61Ft>;_q{GiZp zzV`DDYN})bl{eCjJ^|d!C;J3~gj-ZP9mf*@RIvseO@?0QEfa9u=|i z&T)oiI*Nyh-Np!fME#%XRvH=@mP+HND$M0{DO?_V(g{rx!hrf+rOSGZLjicSyZU2# zLMZ6R`yQ6Z622GH_EveJNC&CD)epV#JDyZtQL zPalvn3ABwX;o^lT&H03vI7QouhqP$7=lys!0(hq_U{OHxD;@+`)K{ZcgI1d*9}o_q zH_x@y&|cP<)Xx_pkBLS4pZHSC{8O0j2v7jZz-70df=wnb^S;wHw0q_~eTFUjz?s46 zINF8RB)5^Ktqu>+I;E9rH~E63?C4t5!G25U~A%eAC5eIx-%FndGl3aGMBT>N0JL5Xx79#TSn(_N0!pAVdxFUd-#}v|*k7!2VZC9-k$UC+O$Y!cX^2j1X*!O;4{w$= zB>UnSB{!wz&JWgJ!q*%>@)OL(w%*=EWQw>Oh8F@ox{Sdg2l=(L#3KL6Q#U!zzzqs| zvY+DY62FK!hMX2$mt@cIUUqlZb0=JMEs8;ifwF8yx;&dokR%Amqfke`EP?^v)nvBl zq#x$VC1>?F!sPHj3$LL>nt7gMn;K(XX4#C-LlYtz`U;V>ptotxd5=Cnb}sbMl=}dy zLjtn8csV9fjBLI=KiWwD;V$pgwgX=)|A>JePWNQAQ}xEF4kK&+tOEfrv_<|dfU3Gl zva|~z?|P^W-(+H77(yaXW1syL=FZ9JpXbLjgL|5&aRV01jJig)->R=@ACS&oe4Cv7lc;7Z0u@TYo488 zFaa5n6d(1!{O%E;v21?UTYbUb)BvbO$LdEdp4L{d|SlCAJe zn0_6RmkW{N15}=(Q(R%ng}IJWKrENA|9ITz*K}~J*C(U1>A6(K#7_pl`LA%inEoM) z^uGx1XAGJEk3iV>EMg(JX~ssKo?EC$Wu28-=-^XjDe*`KKY2XuPM>9qjJu1>)S1e7b7jNRnHi`8x6Rj)&<>em3 zRe^tdQvVPR__u#3h4mOvw|=N6$er}Zp8rD<=HIs2pKlWw0SWD&)i1$n{{Zp-pOu5- zf-nHtY3ZZ^73M$66aPhqK5#`5koI=@s%J|4@0;U4-AchDz(?w6S|4ivDQffQCI9bV zJQBXs+@^V#G%5GrJnpd^m@;HUhVI#Hd0 z;$PjRfS*@e{OIk=YRAVx^H(h{PAUO#vZ6}5#MnxI)qt1??kq^^b>pAC&Hv_ZfVmF@ zfSEFtehB@uxcOha$9*CUIlz59;`;}7;SU4;k1i{Y>CQ(Hjd#lY7y0&oa-gV>onC|@ zax^kAbh#Y=gaQB8i&N}I<%4--*yt<3XZb$?{``OMp8n^9`N0|Zn$;KP-^b}#nf3o- z=0gHc;4-Vm)`-rTdV5lDUgAI9!hiCncb`is-;MNOo|LHk%Zrt8|L$z~Oa#XKj}Q9a zRw6FpZY06aZ$t0?>H`dXEd?wtU$LC&U!G1<@cQm-aM?Ya`O9Ym<=v*kc^aScZ`b>u z-?F6vJNuLICB);eTAd==-PsVVyu$o{T84jKt^d0u|6eSL&#MoTgR_srZOY?o(PO<+ z%bLGB_Qz}Om71B>OmM*W?iK!9O!z-dUXka4)@^oz?yV@N)@>GPlaqQL*H!+YA8*^> z5HaA&nV13rHs2Oiyzb=Bt}@3@*!o3@ktL(7ukUNGG~LDey96z*8~lCwbL~Sc0Sy-< ztsOl*jQ?f#{@{P0MLwufdf2L18W6)CeQ7pyU@Z?=qEeicu3eMUz^)5?Mpc=lRS?ru zYKi-<1+w+lwR}Opb9C0g*ULX8b9TNd?H|bH|763Mii@SL3QXfo^hC%gF+dboN}It6 z{@wln=02UM?3}1741gt;K0$JVLz)n1toa<`#W=7lYU?05S;+Ag~U6E%=|V@9&?-$zc{;oE8`#uS@88 z56)-FmBn?j1fsGKFyGF{8F3XrnFEJ1#54deobUYw_^XO3d7`DIweOpJHXtEvEPF*{xu+B=r48y+ zYPnMSl2w(7sq#o-*4=k4tLl@N;NRRuV1fGkp@IU_U=GQ@pTA7%4=_Of$MimWaLDn| z4FLP`CC?NJ@Q3k{0|!O|b>Ne8nA$_i6R~zXpe(>3-V9 zXJVf@CeRnnIE4Bz1wh3B`RL@uqPrOr%_kd78oE!pZAS|BnPK&QDQ=rHMk9qSN()rN zihKRp0}VsF?LQNXUSD@}ihwt@yO?FeLa@c1mpUsCT*^FLjy9Y(r&|(rpO|$!REw*T zV+I2?zrJJ#QNPunDMAe^zrdBC`GE&p`ziT7zDhZzSEUl#mENf7s$xKr~tYS@d*@6k|vD)1xUE{T;%uT-}pM)t8QM_8C{xwZk2ja^sY zsNc;U5I*cNxHe%o-tadgs+>7!f1$G4icQW``F(xmw%C0a7PNBF#_d-lH3cKKMq+j!8}$cze_Q2%DVyxVqQt(m@+? z)&X@F!bZoV`P5?OyQ*@5lnrH8SyP3&S3*;uIJDT6-w;o!#|~?q^Suw0%ihGa7~{V9 zaejr9M0Z)HUrtkG^FC|6TkX~cp}Oc;yzu-c?q-B#7gOmcnZqZ6PhS%J2DCkDUdYuT z-qk-g>P?OyVD>SnC3To#;(ThT7)f$_C_33Rn030)bhsdj=tk`C56-3W+YI`y_j~xS zg9qOT1LCaSgj<+fDq<(W-j2&RK!%>v%X@761km5T$kA|wvgywarF;EKdRaeTeds5Q z9BBC7_+k$jplH)7VK;nxb*Nzgw3XCTUd;FOC9{#bQ&(axwVe9Q4>hIlO;t<2p1;6# zS^Uwkm34~VbaV05#i;Fpw&r!~N#Pj!&W@jMA+*1~R{ELByFBgYwSHB-bGd|l?uIbE z8&_eR%&bzn9?08z+p?l+ejU5Ll# zT2v?qeX{lS@SM^eC228V+=ytN7st{T0hL9ijf~QJ5|9%W;23hAR?H^XSihT-5G&C= z<$;O&c%h2ny1=)I4^!VoDehiHZn`_&_^GWMk4c?*;i92SEnN#*@72RdccS&1h2*si72xMS}f ztK#KmEc1u&j?+dRyu5s0aC=E@=7}I_QCs6~JxNbfOV|V+6YG4AFQ5GaC3mBgV}Fv0 z7A$$3fxA6LMBkQ9oi|=xJ@5bW1qp7~+jdYoQO8jL}Y!*v*ka95* z&R{6DOvY(B#DnX2eup8xOy)d3b=xy?jK3)waOKM4l)w-^6Y)U>4H6LvHb5xLb0wOiB|y9Y3a>V$mSu zAkLGW0f_y%(MZW5KSR$8xF~A)M+{{CCn&y1hXlJ=|bT)6KyUtkB(tu~86HNp)&dpvOe1CS)i9j7jZRIMTl zX1@c|s=}t!vwWW1cMCmv6#Qx_oV+EYcvHbdT^Zht6t{MNeSfacNdwzve?g;tH7&F6 zyE}qC($vQUQ6qglb0Bv)!LUwixed`L`}G^1|IOTc-hFZ-=BY=Af1QtG-Li8T<8Zxr z%EWAYLzCwMxX0n(SKp|Z1y4|#AHod_*4rDqJyI^Uk?OkpZL*OW{v88#XMMdBGaogc z;;L@ib*(SXBHx|%y^wz;oyqDsX5#8%8$B}_KpRWnj|sxmd84-Ww@Lox>T8pf_`n04bC9ccd9Ym6szk|`X1I^9t|(mqek zpJXtUqUho($UVxYmKC^oR^Hn4LQr@)J4!j2vKq#?JMb6`M1n#bs4bpIN752{!OEYO*gN&80hM zSU45sE0bAQ(SN%9 zr8ms2n>5zA&5)H>xZQ-IGd~B`EGMg!lr}sIP#f?ky4t~(mRBQ6vl!RSN2Zmk8cc3fk z2{=|&dm);XOcYdwM=!?60L&C|&(T`V!x=y7du0uVwv`i-<>-J!(SpPgKYT^sI)ixlyGvJhH6Ew~l9g{MVvBIL8jr zC7f^41IKg6Y8qL}IkT_l#*|ts0P(29>!kNT`f###yICabviT*Yx0JLp)GcDQNhwm` zgEh+^)$!-l7?}aDGkQyip|l?}8X8&>3$VbB*f38rE)IZ^pa!S8#IcT*+LNPh(X3LZ`^R%EbPdj4h)ya%!f>%>^TN&&ztwG z1P~~p$Z53oVOH5O=vj#W&A9{+Eejn;l73_~s^30N?TYUfaC^5S>~>~D`cj;y`}MHT zN2a$=o;`f8cn|-i5<0_uTe4vK5HtWlWR4pPxU75wZn{zqu1_W3R#JH_wJo+)-|R+v z3wNt|c#B$aw2r@onufe+kz+%deOJ*WlG_Hv1e?zlZ*=_WS4;9FZc^v@`726PY!ZuZ$HwsAZ`zY(--a%3y(#oJ$!%ZiQ1M+{ zlzNr$7Er_x_GoD^g6b_twd)8pm&(4tg28`cQ?2etJ`9ofvh_QimzLi6HK%OcTOJas z=q^0L3)Ol9`Awp??7nPWAah=0XaDk)fXf5p+}SD&Gen@s!uU8=FV)g(pH|Y}IJ~3V zJw$#|kxXf9;_77TWn z{w$P1%daI6iRrX#81w#Zf*ll*KL}J$0{;+2iYb$d0Guh89H^ii1j?u%-w5W79pz* zMF8^ReHa2PqG2s>X9{xN5M&rOG;A@uq_q*0 zQmvmPsBjIhqx4Yz#9{RET~4#>U*M*I)5T^_G@3b1*O)lr>VEZVHFjSmeM2nX-#jQy)?!*d>)8W>PlFN=!tv!6No;Fu})%_^8&lS>RdLgYHe(UoNftt|`q~G;g zn^S2+n@{t#RueTOh%ss#Cw_l;JXbrh0?!OO_HZA%TFXGJoJsb$M}VSivQnQYREvRj zT^1pJr3;kEOtloQ7Agg5#G2=ZMZ%6-u|l!zy9p>G_mGcXWxKoW!X`3%lT76IAogjp z0u>sB34-@9@GVC~@;%uxCI~q-wx^?0Hm~nbEfDs+di$69u#D<`+xud8$3Il@Ax$WC2w9^S9?fpjc?Zv|$(bXZdx zzsaWD_?<|Cc-Thk{O6q1M|uQ`o=f^akv%=o7IQq;5bw1nR6QetGzX*pVJwwR$9F;%aIcYTy-?B7M|Ew*nzUE4?B z6Y~>DzmFZ`CIu!sF2mslnPQe^iM0ZrtVxK!naUJr_{j`XJ;Uuli+o!wNJ zc+)K>o23Z63tFY1r(S4>`CXRp%v8(Xyxgt)Mh})N;txRXT@RzvV~9LlYy`cuQnoU- zjf&iRqv#{$;3@?+ye;i6fXaN6UYc5>l=K65*?cE%uRs{{^n25ck!wSf4kF7eYRZe} z5F;>>{(M)+0C~_iBTCc1oqe(gWw22{_rCS_Qym?v2QcjOEX*qCdBf8j1I{cbkPO!p zaRofu1R4ifGye+cki>;zgD4zgp5@?ZJy^2!3V_i8lQ^_nq^NYz@z{pW4*6gIKHfTM z-qWVimj9}5&N9^bNl2Q7j_k@u%U{^tpA4VPO5*@XKzZHMW8IslRJQz7uwV-YO@*ph zaOylEfM{(h2P#kl(3H<~=k+Cc7T!j##VSp5(W(v;s}6g;-*2&ii&6hh#RDNpLFL2v z?4l^oJ~rr{=eNOZ;Z#1VR1NMdxb1H{Y6&ef90*cODQ>?y?tIm${9q~o+I!ZtJ^21d z^uxCya#l=KTJXS8ljy8Sp7us1>fi`83&Q%P^6`;!A4$%NwByv|=GG`OCe`IMgLRgW z8Q`J%0YppWnZUA5y4deHNqZ;QXD^%N*-JG2I@V1Xp>_4vv+0GV+1v%xMA@) z+_=_=%D)ZZ&=Gh(|0p;dZA5_xx*rtg+T=8>eXM!I%JlA8US9lfvI-fZ%o{j5_A}H| z`sc=`Mi669sjAc9?y3dxX5~#Em8r;Mh@U!uXJj3x&E+7rUwrjG4D{^1tkhJE^OrKI z=;{{UZXqfHAxayoU; z2d+``g!Tjg8L@L^c81fap)Wt$XxoZ%s5=e;XQf3w4-(kwp>&htN5)5gV@celr=0F_u1tPOJTX!QO>iP(Fy?z;LYkM zf#+HQ`(hobryEb+cY`RhWhvx;qzdDObb}@e;(;2f%J_WwR+{tLP3$FJg{PMZ>x4a= zny`lB*&kW}4EdgH!{n1fBElBN1`Ye!`y5gv7!jvd*|bEyCnCVq@a++aA&m8A)faOV zyp~__zSz_Ep^)||nK5Su#g5`Ix-qq#t#t2PF9|zkU6?zJKpc9{!#;}xWrBH3RF3op z`%2WNEEa6>E&>)EmRRh|JjaS6d*wRQ25;M5j}v?^I$4_1DJgIx$KiRUP8ln;fx%UyI0bsU_qK2U3X-@ZOVp3CUzBf1UnYs0on6Qlx%F zE1ZSJp~KN>X%|eJIrgfXouTN(Hk*JTYWzJ~5i7CrCcmHP`bL-|9?5Sd$ zLjc<63_3-%8g#AI6G!Y1M&X&)!~LUUxSm6u$hI3&{RWth#c7Z2;8$sU)>|reUys*i z_Zk{mQ<_`rju4lsuL{-GDad}<5`6bLqAIQw@Qs`bi=p+3t%KR_9IvIN=pr7TK%Z;; zS=l;3y~1<7*@A#;kV(JamJ*m)AW=OHwY~g&NTGy#zKXXu-xZJP0Scff>ZPaN8M1^M z@;U6Q7@4*`M%%dE#HZ4&)=gs&>r}ma8HC5<>sH)lL(8ioZlLn|gProkD;&=SG+op7 z4ENpm2;vB1VBy9seNkDU)~*W+82Zu5AiI(b3B;(&*_~(i`|dfZlp%T3_|cyFZF6;s zw#oK5H@9ESu>KP^xAF$T{oJc2t*tDR z4}<8=unnt#b}e_R3@KqX9#=EUk?sS6gD8 z7U5rCws;45Q~tcbp<)6buH#q4tOAEYxQK3jrk$Dg`trswKa6)#Yd<%P2BdoMs%a`O(Cyk!Wj~6EPuUQ-*uNL=>SGe zqdFL$&iSSEXA4iit3DE(BAYdz9eV=|A{Alu$A(phJQoU2&~6txxm65)Gh>Bi`W#eg zLU+;@>eitfe6Gt}n5$)Zoo*d;)cyOOe*R0wUE3h*#*(&L!Ut#&yr{Om0XF|;0n5g6 z*=&rg`SX(OHZ<-CdR9LV$vg^ws~v)>}qJ-FI#OiYN_3_fRTQ0@5&m zG)PGaf|N+N)C`Tl08*j|2n;RV-N+1~baxFcIdtRy=Xqb(dH=koV8sU8dcpq{9;eC)a^TgcV zHU)m=Pu0XVME8+UiEh|1hbajWo=B!|`C{s}Istb%KAjdJ>x(N-9=VU^={g%1#40ps zoUD70iB;DY&!IpQ?L5z#a*o)1>hQNc5)34sIWmS8xZe#+6z}cgStYdlJn+Cp=tq6! z4I7|iBMXns9UfHQ#P@QraR~hhof6l#=?DwkKEnwZYU=c~iIV@jAJ6VI)d8!S8rMIB zmMn|N>qEC?1ZAZ7qAVnCI`L)*$lK2LZn$Y!!DlR~OyOT`uEQwxm%|TAqdlc%E}>f) zq+3f`*a!6BI1ISyHQuRokuuq0kxynrUWS&?i0JAQCDCC0H*r~KzxC?dS|Mkh8Hv{Q z$8iY_uFu2Bp&#YmDFxq#Y@~mE_FYZ_CSS;A6@bYJequ(5P6FCMVgteMTm|MwHdc9#k1Xx=x&F8!>yE5e4)fs?51r1siPT9`1>LYEX@$@l zox{UJwn0d0ZteI7#le%XiqC68Tv2Ef!-MW4N~$7^V{_L*R5ORZ#v{R0GN{x_HqC>5 zi73gvYZsxYta*|y&7{H>*wK1_iGmH}${ZuxS^lKH)Woz_7WURkZ{d8Bt8AYncu)kA zmsiwi=Jf(qyYBs=oVE-TO{^n*d;aD0xG!;^YMkxd%kO%M|aB7pZYUF>-9oZWNTh?5R$8On<;kd{7s%9Y-u=sM!lf+i0IcI!U5+y*IvguRBL94ZU@YCw z)RzS$crZW^Y>+A!E#hbP$A2Ckb5Jl@f-lPUFb`v_8!?H=f?KxkJd;F-`ZcD>R_}e6 zw}j@&Z4P|nN_^LF{oZ&z-TzuEui^K4%J(v|>;r!%LPi`oF-^woc{zj00_Jq#==SO` zQOCqDINa*VgRPT}N&{7Uz`z4+9etDaUYIdH}` z&^<1}8{!95M-c+?tG@m-7Ppn$uXP{R{O}R(Fkdld+=hwB9^O!L^_b{+{ot{$b!aTT z_B~fid7(2NlyMzQvh}{<>xI-ftlj-1A1sW#HIc`J+s(vVWg|=)ATany?H}-nr1yQgvh}`X2C|U}hC3Z}g1NFkIK4H^znzRcta{;E;yXJSJO(hnLsh59kO@yOsiKC||nF~?C2g>aOO})w;qp1RRPe3(XOb&i z+%tDtxKz8M$cM#2k);vv0Hd^tC5i>ULZ{l`tAUrc?6J2_J^iCLgkq>+nF(JZ#_&9nNSU(S&=lQqHXcr#eo0%>Oo+kXS;HZwb#=n>-l@WRwszfiB_AshU@cgi;QR%}DsHKs{>o2P(@3)7Mydda@ee5?!=}1@J`h%|2#yz(9 zng*s6zfPaQXBSY!3@RQnlb_a?O1bH`s19QnaXz)_a9~cD1Yqt2z0i$NCGG_ni!h20 zBHKi4d?G-_bj2rvrf=ren$>uTDonx*)=w6{sd;5gjXLp&+I}#|vEf3JKUo6a&nYv0 zb5)|t*q};;7kDNB>7{P1)%E%$Pm7T{>l@X4E8MZ{w~)L+b2_a_^hiriVTwQP5E`~M z*3jg?jzoDKAjMct*cSL(4_aSWx@WXMo^#_~eJ**+*B!hHyh%n(22XG3x{u!_3Y=}5 zzew%kr$-$Lr8)CoC97=th?kB7Cv#1Ydd_@_pl`(0n^B?43hT*FjB?}E(nPx$veIT! zfp5Tq`A`EP3Tei)d8}=Mw^QG}WK!iiN(J*ghBCe;V0<_XyMxu_`2T#)JY$$=`K|Fl ze52ba8K7B(o@@y(dy1mzuXet7=&VqJooUNk^wB=`_?b6p_!h+53r%$tjch}-7h%CT z->oN0#ks{g=3?w_D@Wh*w+eI-GCgJ5StzNiY6&InYK}ZyhHpkX|Egt_utQTVnu1(S zQ$@8rY)nh)Q0c1_trgO^1w`=~Q7j=*=o`|7Wg2Pvw_eslBysd`&t842!B@lP_ja8< zwk8v|YV1?g*Di*$^&Yl)9g7Ts#)E!@cTeqwx~{xKU7lRFfLVnOPwng%yj62KUzOJ) z#8p(*!~p!jP4ZW+lj{spwuh6y-*4T|`_m!Nn)Z_@vH~bIHyOwOxtNI)EK7D z-#3xR<=f~U5J)(D>r>{*>8E&cHi9(y0{D1EzrU8uV=bqReC#iwuz%`k^a5wx2{97| z+(fK9c+nCQt}VMLWuV2hte}nIisX4``5bI?U1(Zjxq2PJ5A>*0)r8heRG@yr^-F`l zAQVg8Df=nao+<)>gEB|QKR5AB(RGzv<=tFKb5E36pgZp%jDM5k5058>u6g-DxWGGzW@P~<)X7!+~AH>EG5BL@y%4r5HJPP4st z6re&GaoxgA*8|t;jW&d8jmkPr=-ZP{-tF=n%oDiRL%OI}Z8ZjMV^g!)LAB6d9~%=* zQ@T@B^Vq?O5nu9pt2lF5fLOABfhd@yv&} zdO8YzOhlW@@&-(^G69Q0es?pXip50x4zpwyZB?6f=)6%4H@h1|enumY<{L~^UgKjK zZGYd+C2%Y}P>taZ3lrND$yOwuKzFOt@y^a7gb^!mfosKCqNS>-3dTS5pCZ1dzEcgT zNIW`Ns~E-0`d=}=m>+1tD~WbQdT6&Fw)CX5zNk?;Zo1>&CU_GHigmnNh#gP|)gX{e zwN;CgsfaNQH?Qmdy+;3~+xxW|<_s`qy3vuz#g1YsUY56XM=LOUasA~5;Q*5#MySdkXRgN_1bi+ci{FbA(6IyM_NQEX9G1lq%^w$S6^73K+rRlOzj^44!>@Q3a8TjdLQ451qMFVjak6LY#4zK*B>)Z~L5--vh|M#s??% z_L{INgrpGBC#Zox<~Db!;UF@orOf5X6V*c6x?7%sxmBXLGpiOXY=S#cYM9l@>c*hF zLD8D-pPLSzq>s+SxCjno?Lhcr|A)L%2Jr-29vGn9{gw|Ba=QFRoltt+iirSMa(v4 zemmb^6#U_-KRv3~zkN1xJ4TVF>h_n0V?XDh&R^yV_1D0UuH%luLm@-e7{ORU`1hk$ z?`ZF!ZUM7SqoNa4F?q_Sd$9=Vb+&O*_dulRG@nQu4bdP?BCW9F*3pY%-rS9r+XL+Y zl#mU8+UmX=`$Hu0e)AB*)?t`7bDE9c!zpYLP*+@CI!)Gw#>AL8+gAm>G?are&sv7- zoqH0LICZszZg3+JvsN?>^CW5;k9{QiEQ)P#llZI~PGnEIPn{Mk91+A3&J*9Zf03YP z-C6lyDI95xz-3}wa)@5|FKeY=Tp$$39{97F9zof#vyt~}(6JE^xRHjDxXuokPy?%% z<{?r%3-)X_Lhp@P_<+H9q3&m^e!9?9-Y*YOvl)bZ@*(XKrlKf6hnrdYH4UK&ez+{( zx`$7WhzB)-l<)YXt!{aGr-s(t4$kOmqxJMv2LuA&=P}}-c{TU@+pb#dhTnnqf~J@AJIbSu6vB7blFolB?<&od3M2} z<1XsvAn_9R%Oc9z;CrFoT4v8za(A*JF7ks7Os3q3Oena!?hv8y@u$upgnW)}(Z(@Z zXmlmQv=FX@z${1NiSIT={2UJ1HyD(_w+UIr0@7^;M_0OV2uiKiIVxtM10=6ZW<=I| zDBtY9vU>OoHVY(t2=L6(WRDF5kI+#_V8Y=9Li*C zaY{^bAMV(%2FNb;9KP)9J54z!oW5*P7n!?0YnWP1^3`=h*H*7u@hID( zeQst`j_XUGY(Obb%K@06dqKWPuKeLxEQsupw6YCv>huNf!U9UE*~wBWmlWy(x()Oj`Nc zXp$-wlV^y|6wl$^ny%vqcUq>H|Ci*lkbuF2HQ+6N*Sp6jn}XcN%WPr;NG@I!t@>ec za1qW9iX>BRHLx!RO4lJ6!l6(aW(T#iC}-SdT=-KveWr&ywi$J25YoI#5;|oUs##Mfa0E)p`_p+Jz! z0JKW>`I|PY*Ds4v;!tIj4jm#;xJbhWqkAfM4%fFzGji&BN|ppLTh7+K$6Y&xQ9pP6 zK^*Uc7#%yr9tKb2Y%xOQ%yHSAl#x zWBf$vh-ok)0ABb6pU+)Q9!DYRhPw;I|G1;o#P&v|$0wmC1-t6rgBV<5u@7G{`C-;v z;^uGx!lK|zKeJ8lPz{~9o$*lBV*&8?e2DW#Q>%dVEXwz+lt{*Wnoo|sh{OvA-s$3G z@1LUaVzV!tIJWB=x}cHvj7?Ngu(>g%D268dg_*c?l3!@b%)S=W?ljrXOS-3y^_JoF zhX$k1t5a7(gaz#_KO+NCXfji2VL?dA+(z+ze58w17I6E1Mq3YHwC z?cVp%?%SQA%tFQ;`^`E|It!;X%Y6I~TxnSO2{V^j zwq{O8#tJIm!kEb%MAi!Yo3>KTQc}Vh)X!s1nl1h)DnB);%&Fe@%^fY|m@TB_dN-4! za=8&`j#^zea^f#~(=rzKB+08ld*s&r_``nd&8x<068Wj9v4Xp^F?uZMI;VwV6hbpa zLd1YQ7}oXDWOaM~^TNt({G-q3hzWUT29i)`w^u8v-lVOL=01S*m}R-HWq{%l|B82t zBcHl0nm(HKk=~}P;Ct3N5hO4uoR4D)gVetXYD9C;wASH!17w&WoUXhVG;?R;{%hm! zpKK1lJv@>-@1p5Fp}$X;pDnat&aD!qfE)SHw@Ij&Jq)u;K}N)-M(Rny5^i2g{oRz% zz3XYf4%~;$mrblgN3i@YLyLNkYCqJ$@A@Fmu6RB0Q>e=?%yEzcSu2Z8`wB5Hfm7PP zIQa=q<~B?I6q@J8%slZPSl($tPMHAU!CX$uw8I0M&_lbaTSUgUQwSlx(o3F(d(jtD zIT|gG4RN`2%LBRJ@Cm$q3ff=^-E{{oXQm zULQxeVzTJ3&(R_d%#4XWK)L(v8YZn{vA6fkx*5 z?#SFZ3*mk}Kqv=#T883$f4B6)+r#c>;H@Ot!-J;V$5TOupx>f9)ikj*tdiWHY0UEK zHrVhMBNMq3^BV7qE2|?Q8j(U%#OBo?+4D)$fYD)Y&^Xq57{-wRTU=+G1gXA~MqYZ) z;Gv?)o<{-DHzxG9F^?KToTJ=)2j??~wRYZ{Nol5~kRQ)@=0=>(lEm)2Ts_Ch@z@sr zx!wE9M$=_Zgk@}*ap(4G!1vQ^oFuD4D{f-j>x87;gVqQy!K7Wko$vl{>81By{~U}& zBj{E@!A#=cmSH$}d8;nPmgc9y{1O%QV`uV+>a6#Y$mr))(>)WaDI9E1$S8Bj zic!~|a_~vlGG(jupJi3pCOXi-uI*m(wZ*rYK}+5y(a}_GS<2ul?NeB#RtvQQx`62|Fw)Wui>5Z zch#k^d%cYrg2LIYQe12u(GqoJG~21f&e7E%`)Ga7VX{Z75l@X}E6`z4)08kQU%@To_mWN z{t}aU6Z|cRHNfimkZfh&2IxiRv*C;WQNSBZ;yduh>Ws3)AI3_fRXi8K+W4yf<|(}% z8-Y`+h8)TlP?z?SNMmT)Dz!s=`JrjP%FCz4J9jR_M~s{B8RMX_gPjj0p!HB z#pHX}BPDk%3o8bnW>x=spUBrcu;irO;pvU`F#GN1-55`k^iKb|DhN~3i|{Lz@0=bU z$CS9;y}I($?%WD_T61C41p^L`KSUwQ7!4(6B=*xJnN<44W4HQ&<}UH!Qwy(v^J1(+ z+#Fw1ovy3ogGv=9r;jXRe7a-~y}5cH_HoVA?Bmi*E&C5t+X;cs!2V}GQt+;>2mHfH;56-TFTVz1B=z;>H=GU{f5Ns}VhWmqA#xvwrn z=V5FI2mj*TsY%Js-rpKuX`cOZe}+2D@#p;^W%IvHJnT8#g6TzLKy1gG3>zRcn#Mfg zi1K9&*HaU8?j8TX^qMn7C_%4|dBVzcG0#?;ueqZ#{3;%S>R{RaiGqdgZ`L>6kPL8I`94d}l zp`Fl#MO(}{(R5r(&EM0w2f%A`*1WSv10EOf_f3|)&X$Wm(+6)46S8!nV6IFOo@%H5 z*A>2}zt|7=5_poC4rOODvz1gQw##A@Y8a;4OdCD$MRyg{i^(!d{Fb9=KXd5hnjIR1 zI==&L!@6tN=cds=E9R0Gk!QTpf4ChrW3^0LJRJ*kqaliq`fLTQ6E1sXd1V}-h2OG2 z77%gj34J9`lApbWG`8E(cX#=ji@aJ1C0$rt<9aaAOR$pu_5E4mFzlQ3L5O&49dQ<6 zfx@}s>zC9Y6Q1zWSW{+(yZ%F|L8}WnZn4XI?!>#&W-b@iQVEdRolqvzm}CSFt_uoM@g&! zJ2+MX?ao&VOpXqASJ!WGvBSUe36y#d(7*p#ST~2_XC;(G4$Fnyo(Gp)&fkZfA2cgS z)w0YCoia9!{W-%vLCV)Ruqb)TY+?wWijHEafJMqpGK(LGsGcj{C?@&l;Y6AT6-3mAq7g$XD1(O@r*32@UNj8!YiEGkb3;4l z?65C;u4)C+&5}#xU+GK$7S)3M{Gd$3`DZB-Zn@3= z;&YN)GNmaf61_Tgylp_1R#atIg7iyl zVsrRcb_)Q-IYOh}PtJR(CXOQ5<_5V0LvZ zF|588`sfblCftLM3qQGpfDr9&R1f+6X!gf*V$I%tqMo(gI3slQ(&mLxwA3_{lO-%f z3#6FOF8En-bo`$Jay&3QPuv15jYD4X(5oi0-^+Ze{%z-X&M$#?tN9zpSyce|Ei$N| zbFMiB0O6P~KSJ-;^p4keUIm|Yg8zbWqNl^31b&L|+ADjh7kl41j8|lPhr@)~si`XG z$CnYRqwBXp2Y2~3G7xnwe{7CS=!mn|%qM0SUID)zD>d_nyU0}5xq?y8{0F8{OoZFl z?<8b!yIm%7jjdnTc(j2c(M9La!~}K`*#2qjwgM#FoP;#*xevX&ke@V)nt2QtgBURO_{z#HRkho zXXV7wAmW#v=dM-8)wXYI>lXJN4(4|o%g>nhC3^?BW*}K#-F+HkH72T6D|?L<2G}sp}jD z&?Z9{uIprfa#TRiQxBOv@|i-iIv#Ai)wk}o_;r&>LVq?mAJOfj{yJE%CzNa|jOPSRKU zp;u+Ew)@V^AZtX^D)CHzOyW&75KLR;RF>q6ujmgXN+K5S08x#3c1wEaU3+sK`v%=7 zwIpmcxQm#+z_16%ROiJ;MR-9u%7*zpKD+0%QSDxS3s% zX}hPOE4z%uH@oxOqzQuZ_+h(e7Y6g3rjNJn`US6e3XpyMiHVS~NLdG{%$eJ0G*M8V zS6S(do9$`p1)0Vj5As%ZWjS0Y=ch)@)0Ouj&k>e}ZT{KSHjFkPPw(%S?N}^gq{-yS z`Q)#WaEpZ*-s;uB zzp^6HFU)8WbR}~1-r$*UxNB>+0mNDl(=UrSbT@4uza+X0JHdI&6JkOGRFEWKOqAFn zfh+z1;W-K09Yg<`>Jc6Obx1eR{{8SfD+wff0yOt7Yh~+R{Q9gAd`4F`%l4=7Bn5ZR zNbLPPoxz<)?aSVNRDd@Oe~+(+5PL9eTmn;J5r7K`adlRB0=0lXl^)4x0B6bj1aq;} zxS%Ea>?BLBlyzFbTF-!^vQD1)&fC-gdHo=ukPK{mKv1VK)l18&bwH=xtPr+>6%uRF z{+=%3la&Y@RtO7;Jfo{(@WBc!(l0Z6nUZtNKbPQXxmU>si+>{LO<-vlkbyA1pZ_?H z2`MtA3(bIodFdARRjCJ~&j)Ny?SYO(Xj)}w`?E*yv15&1dkFiZ7$MvMFRUlTtgWd zs!h*A-Sn?|Jzs@E2CKN6XoZ zhL4mr8|P9&r)TkaEg6uqe(E8@r*aC;DT z`{23F>Gg|{uAy=o0S1#*)BVM=Uh!A`rhffG-cln>K08M)7G)B3a?0)@ieb312V=Z# z%O8KVD#pZ@qzM1M0Q5RNB`*lOBn~>aAV@N4U$G4x6qfTkCWcn-1xpR6sbK9$j`QRo>h zKI>eOn7??>dNlvIm*QJ%lVe^U@eGHyXd(g_b%VYn=|zwN8P1{yvn0SZ9AOx;FBe8! z1U@3!f@e|ZLGnU51*h7x>C(p0geIJrS3xg7z?J5pj&i{!8+qQpD%ZG<-vSfcu*Kt_ zArZV=klAXd#|ZJ-3a_FKK$|XXtm!jl7Bnv!xtL3eFT341t_oO+zPb3&dtIM#qRn-TP88emseZP|YMR*9j?7d)L%Ok!?%= zs%-*HhwBv$*O&#iRwUAw@^%IM9K)ZYF%xc9>f8V$joNGCAG*|_k!40Vjl?7l zBj@dJ2XUA<_zNqZhE=5{-(*N-YcO*QBIA)qGq_78g9X31X2x(2Xi%*{f32ybp=+vh zctbg2<8)Xt_?0nvthn2KueWT=I~LXkJlLPf$FX|#ICvnE60LS{PRI34w_Qp%AKrBn z_I|6T4W*yRH+c;w`du_S72tBaEn>&=cN57=%vb(`;)v>s#`_}={0tUbE`$|nuH-vyEO5nmbJMIQsvp@)=8(hn` zak^X(&ukq&xX~7v(`5DfagyQ4?|!ITPrNwF4ReI1VUW7ClBf576~R$7#e0d$1gbzK zjG0eU6epHY*9L%$F=5>tdrW50piY|FdkAj_3H?y#rtHy`lH@Jo=fc!|DHu!=0KdV{ z3e|MSRW9xpk-8jiz|Z2?df1iH`a75Y-Z$~vW&i0)K!*v7Vi8ay3@so}pz2FaGYbl^Vvt`WD*NMoO6+a++ z$&euo2YVrp|F#ZR_xv_|WTe)QdiNySTn#`4(d}>to}4d*B?`;h(z^i%pst1m2WtrW zVD>kU-}Sk)cI_L>_A}Y_zQi8-vO!JI0N4YMWtji?{=)tBfwaQKUA5Z-DQA|#$>O6V zs&#-cG#xh2(e=8muZ7xHpF2S4t)vJ{A$bPtvgH2Zuhj?(_ihzyZ}-C=b!6g4fkbdy znK8QAxwKx-M&$>@odNB*tRTn5 zR#iHwi5Fh1JbIN$bX#k%FEwe8Pkdy44x9)7`>N6dg|p8!B=WbCyM4l(!p|9FVF%}m zw`}oBr=@_8wbZcIv1s97P(C5}2VogTPWmQia;1e3HO?Rnwi!gpLZ>{nzze5*>L5)Y zPMH%xE_mX3$`p>oG%F}O1D@M9xJx~vlNILBDhOHgl$v(!7Fo~4g5Q5I*Yo?#>t>9 z2E2H+`b)>|dSme2$m;JxnSkilyP0qKi);oq zc*W^$W);VqlptTwX`NLxr*;-p>t-bIfR1HZCJAm$BF_Aru(qFmGj-8AR$#3DfC%^a znfrQ>;&s!<_+b*i@I%IQn{2s><@kHKbJ4Vpue3fLnQ=;SX8w935Qx5?Utsl96JGKm2k$&syKF zB#Suf2$fvpY|pC+A$Vye|B8?41@QjUa>{F93{y#oM~cMM$rpT_VH=r~M!&`*QJo#o zL(^m74v;pBAAmY&z6?>shvap~ zfU~y4r=|D@mjRcjRe+O@#KT&h3@;}U=Udj_;}SzJgU)QrYx-lnvD!g{Q*RpHZ~6Rr zb&?u71(3!ze_*@e&z`n0?F-?2rIuS<&@9TN>e+fi-n1~AEm4s5g_6)BMlRLedjQEs zJuJgq%QmBv{lH`O_Uc8^MxR0@thcFA;9>J`^A>NZM}Hly8@tj#$+Mq$g;*PtVA*0m z)iR0r100NReV6r7F)%X)qfy~8YS*w++1Eaw;>E@NJWLNtVG7lH++gsPiRirv zQ8eeSI0V=<2A(Y#Zp~AEumkp~I*c*_pJH{{Ny~p!(ZX>JzdVkb(8AxjnF3@dLlhdZ z`_Q(|<~IVubq#h=K*+Lqe`}a1>RXD}Ll<#VA^hc*cblJDwD94h!i2? zHR$7x9a-K#a*c{ROaVVXjwDa(diV(!VzULPIAQQ8;}iYY~m<0YnE%b zfK~%^rrCe!Obwj6#~P1f=y@Hv!|Gf@!@whH3j_Pu#bKM0jw25*%+8#y#~uv6#htgo zq02|+0gqFWgqzdtvS-FHzP3E%T8;_%DV4c0NITO%-G;dB07hv4yc1i@MAmk zi!yDKC%Ec+gqA!1_-rm@-5m?CMFc<71uJpOav_;hqqT^#a5)H4M7#{cCnIb_ekJfY zdi$;N`J z`2Y**s93z9DT7?KZ`B5b{DU;jx_KJlyzX?NUf>RuTy2tv`yEfy+aivRzW?%<4qxug zQM-xLU$43X6pbAEkpzUJeK=4YHrwr{S^P?zpH-M`RVn%bmHwqhvxd@{aIu06WZd*F!~jGlXIHM znLTtE{54;&LQ^`kI>ckB_Xhha9H=<6V90s)o(+~sZBVS=oPqx30@*XGc>LdP2MV4e z`2dKi^hlYUr<~2S9G-8^ZHVHdTS}?f0r|#tMN04Wg5shxD`+P*Ge6w#W0h1dyri%E zGAfy_9?bAGFVq@?HA z$BmN6ph2jsY-Vv$yk0lYQlW(mxi&=)Fq5?`kf>bop6?iiaPcAudn_|{BIV(gQxmwGbdl6(P-Ur zo&h3p5M6M&xW&vl$l0ihSSS1oPyA$?EjG&4Qg;yMzR6GJ)cn4Rcodpe=X2)&IC|`8 z!N1fjDH*k8+%aa{EO)wp6%|wZcAt9tpfu3y!R$~0*&-x@UvMG%<-!~;g0Y`7e}D-U zb~;&%u-$bdkHj$U*Ci#|cLA@w#W(X!9%Cqy0zBjYCN8Fo>E-KcC%5S~-D>t6(T{SKHf#ThJyj3R zvViOXA1E!4kXBZjkM!n4z>whf6RH$)#*Yp2Eqa)kc|&Wmr{sU7{){V#DGljq9jLta zqXMk`_87L6vU#~PosB;kcX;3?cHS<>gdk2u_#6o0okoYvc70CfVn2d`=aS&$wb+qQ(OIYypAHey#ON!55q$}c0(;mXif*OqU z;%c{#rwafrOUTBd1{Bfs-IOHv9QLhDLiTBS`@hP_VVToY52SO1;)i~G{c|xA6Yc(! z08Fc4P9iJNY_!ki!~0o5u+8Ad3ZA75qJ{_|PB&J@)1M;zVrhKuJ9=oKw-@pc5X$rB zzpcyjTxdU_i zzI!(`=iP4{6oAhIGa4+F1V8X-cA=AFx_)id&moTRZe~z`gCPY^t0i4 zZMWXu%=h4Gn_GhqS!umU5wjm`)TB&HzavQm-B$yjXxDjqFg`oyA4fkpiAH*5Y;|z? zvQgvKf*wt_weX*tQ2sPN_Ap`&kn$iBA)>FG%|THxe#1D`TkT$c8X&0G%>44EmnwRV zQ%KS6VPv)B0*}3JfPsP#3_>%E8emJXc7>dA`z%^55Kj}Li86Z+xFfAL3b8EGGF{9$4c=cSN zivKM?odVw*W$KgCz=3Qh_?eu0d?0n3Bc|EESWD72ci~G`OIgQdbLZ_#=`?!j*FL)R zcfq|{ZTJ*Fi`ZhA50yoOtFi@(rNLYi_HAO4GJNcGotjRq&@$MzRds+lDb&g*%y4OITP%Q8)0F4%y}};pc|c^1YfV5hzQfr8Ev0u>2C)vm z*xTmD&bYB#WY=nDd6!O4BkzD0aN?J)OUQUx;$@g4C+=)RCjHBw`@!$vl7dYh zVl;XacAiey{}Y&k|mx*s^~0j@Jfoco*-P-*?Gh%%!NN$NX@` z6~HGx$)c*YSNzc@r)ydHLRK(v0|+;IQxZC>{F{~MESp%dXLfK6qyoaf#Gp-eMVC`Q z3s}eGgG2rIqzz3reS<*RO0n0a$RU#wNr{$)6rI9bMTy6ynzUG(gl;ckq}lgg))O;w z=oD7@l!2UwBQ^uQiL=PH{5_=Gjz?6<{D$h80&Y%34vUhl`n^lQA0Ax5YL-shZgcm3 z>;JxD=;+TW2Zml1t+bBy2FEK6p*eZgmg!zQ2S<0Z2&+wZxNDuO#{2msfuJFZ==jR`?V?a> zcc{gM7A(T|&kf1@%))HwUMA|@d&3w0kqwO2IRin`*O+YI(~HC{;{*H#{z)QlWwtus z=F(uCi^(gJ{EDT3&%KrJsN9k6aN)CL>d}7@BdIso%-+`LezD?1I@W*MbB9Q^&ox*u zbB@nQmGk8HWw8!Ys3ZW=Lq!$A^k<|OF7fTZIgBF?&NZ$!J}H5| z{Rp65`|@Rjec*uWz0<}X-QNoC1N`{E6W~>3?|`T0(|_l&eKX_UN*Nzw(vUmfE9C4G zgV>N}T9x3l?^<3KRK_zDE=+K4>zL4j-g9U zp1TRQ0_&RxpGix7fJt9Fdj_SVO=a~0tK$D-?=8ck`rEy46(poZN*V+Nq@-IKr9@g_ z01=Sx96}nTrIC^@>1G%jq#GoM9%>l6<5~Rw`?~IXU;Ezs-p}*ud9mMc%nZk|)~w&G zb=L1ZzaIcHL5am{^H1IR=%=Xor`HhimbD}p^TUj*7rs2w4~z!yn)Vrk_|FbfRnJc` zH@Z`9*8f1MJ^R>Bt=Na0f5qwj5kAGPXZXiT{tb+1)lo1!NjVui*%AYkp?m14jPFZ9 zzn4*RfdGYI-0m@E4l3{yGE1ag{gbN_65Q(ULmhqW<3p%IJfDCnzx0a1?%%xU&{(!Y zLkGa|BD}ICvHp&SK5)QX_ou3%Mhm4^(;r8@f^aS2pKE#l!p?&Z zUPKdPki-rswt&&I?JV#*&gQ!Hvu9l!W+ zetK{$b-gHVru5CtyJK9~%>Rnn|ae|B>VJFQ3pT5dy4{JQZrJe=7a_=ai9ue8c4*&i`_l|KjccoZ-J%oPYl7 ze^%+AOYvX4`2XiB*-brxiPZe#;rNGO)_?vO!HIoKwDi~K9_C-cJ44l=>j}TZq2f+2 zV{|zM|Ggdk+nW9#|F+J&St=ob4hO}y8UKp|{4d_h>pWuKK1iz!9%}yYl9(P~rs4po zpbTKVx92~#QDD+^FM;c{^u;=HOf`zyTAy#^+AHA&!&$_~6N zOoA2($F&dD{+XZrXHUcnYG1;Zca4NCq1RVUzkPf{Q?KO^SL%m|3@||Kl9zq7B|Ubh zvIbG4BV&(IgBvh9cqe?GziVhMI>bO#>^?p)Md9`A^O3-^M;V74>$WW9wEZIu*PIy+1sLuUfK0h+YzRL>wRD z*?mS1x6t{iDYpL1hYt^#{LF`O;Qtzb(j1`PiZX89o+}3Oh$oy|H$29SZf%gF>F9FU z&eBzXc^o#JW~{tI51_FK4gZhn*1vx3Xid!PqpF;%8)FTR$b~db721BKN4*L(X{n7b zT7KV^$}cXY=Yq5M5#yov2mKF;)PHbo<}qK1o5kOSF_vf$4iLC;x>7*QS08%w>8yVrsuK6@Orr+#-Rge#(xJgbbZ)gR%vr1 zn+5?gG>0ytMDPO7o|8DG&?`U(=*4|54`_yl$8l&e?ui)p(oms^U!TWX%^us>)zuOI zW_3SY!@WIy?SJI9zZBJ$6Hi_G^6m{Qzl<_CDQ%Qp1Y& zdUe?QDeS>@V4Kc}5+XomH$I^1+k9&;i8wN!tN2<6aI@{XD7HIo zy882UGJDY!GgUHdmq52Jvu&w%_P9y!cbz>}uBkDfJCcst)t~@NK34b^HXVp z9z%ayu~b&nG0K_stE5E~kCUgm=cPJzr4vP3NIi*^S3RN1F~W7WiN-B=>|O5?;=;+e z8T-8(V%KQPkzpKb3LH`S;o^a~3*tM7WiC&8C z_IcRUg316|7gpZezhEFM8$n81Td{yhC+FktX_t5IjM+`SyEudOFc`FPOP~z!z}UU#@d3!a-eM_RL@EgTunqiU%@6CG zz?}A*(2l3K?7|y{^W$vm&^=PLifTEe1T7`o?eJo9>zsNGupd%Zx6FpaBacb+_L(YX za;PLNH#4pAizCGiozLma-??a|K0g^WA9;nt5_oBB|b9n5~9D3w^HplV1@$=3tO^*OkU7KN> zJz?JOF_Ypo;E_}ce4twVjfjnIwFh_nlUota#3K+nKTBbv$%c!bti?_~;|_G~^UFDJ zq^bALe5f7sUztJcTZN@Ml<(_j{y)>x)sFc>#cMtVUZgN)+#9JJIDJ%oo>zCW*}Fy0 zi?t%l(ilsC?$PhC?y$hl%{Jczt_vsUk~C}3M9hw5nvV~}c@0DEf5q43uC^EY%{v|V zY?w4$?x^7P+wevfc~>G|HACvKzgqw0PaMT@fXqWV&iSU{$!qpYypq?h~LK^^B&y@7ka%a>`E$%`P>jGT8oX(ZRTSJY$8(+}Fb z9t-c3aL1pSi1oT}F&z4TTRUh&r`~#~q@Bp5Vv~H@DbVXPsBuol)gr`S88b{$_IjnN zpShur*%RdvTCCywxA^{bc3YLMUxw|gn$<9{cAz|4a~W<%rS2;9?q|(sxf}YEpWEw= zPA$Cg-n4pqe_HBjsk#hHmvwqU>67;dxmr?83Q zg3F}*IYd)6@mhC%>xcvTOq%xVPqvX#C9JJ$ySV_-3ALf6wOeubK1DK7m+xY4MG~pD zCZ>)8_*?d~ME+bFHQ%N;A;ful*kXhA(;obGdyct&!xi?!wr}Ji3f<0# z9H;Z1p5WDP6@7oILAq|uyJny3Oo~DF=%ziWumvwgG%<%>L888Hd1hSw$yC7!({`{y zy3hG7c3z?8mSH-^cI&blt=O_l>*pRDnVtI*TR%!=MK;JZDo_$&>Rj)c#n1H3K**X2MG0j=CQAr6oTU|L{pGQgN1xUz$% z1HMHFAyKOp%3$O+dKH>t6t`ae3>rqvQB04O$u5<)VcI*zWMiHOuPj}->gDd&+L7Y3 z)>8>nK6MWMbGf!+_IY!k>m%o2^3jKxBDA7EYFM2`j>c(G24tZ%$iiws@I`6CU9EzA zEt%oy(LJ?G-xW~RfAOr_Yb%k~Y097oO><<0$xvL}Op%f!q!T}H(4byA2%9{+{{iFx z{TRVc$f%^H7UszoiUG?g#2@^!y4>}2)HMiqkJZGMgD8Ie?9_**`AUqjtmPm96lt#a zo@8_$)36Vvx~2^aMJ7>P6y}W2ggRE_I;HV*yJdQJy3DTzJbQ zve_}*_%L`GtC zS{fI0-wq5Q>chvTqGT>Uv&?y~)WRCI!|VT^247mEjb5))k@3D0nIJp_0AA>C7489E z9D~VfHmqvh6q;Z>Y9S~2MUNZjdw^HOUXzN2fb`gJ+~g%JiVzp%Y_3#VBA_5`j&5e+ zr`I#I5*JEnGMB*SFxo}Z?;4DUCD~YKF{jz5b<6Z9lRKWOZ~U%GMLhm~i(fX9n!4t> z*9F^i{;B?=qBh(?YEMp!o6yVUM6i#SsE z&(+Dq?TbLiyM6h`o>8Zl($IWoGnLm_wsi1Vg0r_lz6=8MOntQ+e^=Ub_Q0>(ipkM(^1y+B%Y6e5NVtTy=AX05PLD$ z6w;iRX#Xr3NKF8mw5-36Oji^0s#Nc%;m2WD5(;hFFl*?9ec%PCs*I83d&yM|M;URx z{&L$VtFm}8O_n#@-}q;Q_z~7ow64mZoZ!4#^2RZkzKVQ#9nzYgtvTrmsrf`cFIWSDTN*I62sR8vCSYP z%I}8($BrTHX?T4`ROx(FW%I*+#{SWMcNwTE9YA?aYOTlf78B>EfJ>c$sa2y@VuWbR zm)qkUxS6YehSdTx!PHcbIAMRG)L3hQ5zSv4CFHF6rnFjp@fzj{)~+-W)~dDAqvQod zOc}Woa?X!ZRc9|N9K2u0TZ`Mz+3LTN)%$z9!tN0Nx5FAkOv~L9M8Gwq8R4I!@!L`N zoLv)A@1(I@nRa?%meKp5=9>Qmgs!rp6Gq%c#3Vye)%dC@!V9a$ze*!V||-~n_VTowt@otCOT zqD1(ZBQbc;a&vmd@qFJ^s6+@iz&&dCeLBnrEoMn$z3sN5Zw}VilxMfRvo`x^`iqLT z_$bN6^>(pTU-5>XFB27Z_PoKLMtzPRNUyRj(|~KO(KxZQw}CwSIzeDVQ&}He>>K(| z>7_0#lsN^C25p}naWG}1F@$fF`E+bAu?Ai1;bf?n*H1s+_Fkqwn(wqx94C*s;5e_c%2Y-4Z^V_hzY0XRi!}>r06y-Cungt)tX^0zq1<0UFVKlA z%*!BgQvCK>3T(aTvNPB?KH2c=0>NN0yFRItFbLfsbPO=I`I9l>+*un~#zBhNL#Tef zSO$3_ezTj|h}SZ8!Dcr&SqgHfr#1lJD0ywVP&S$WjrMX<#btI{sPXU`xkM#r`hce7 z<*?{Jld~BibIf%xcF_I9ws@!9fW77VHFM6M88*e+&i5QV7PB9aLYCdDAAZh@)4Qf^ z(e`!TSa@$5TO6}WBN-!Z4}KpPJ1Pm`*Mhv)oo@)~O>G-47^~hv`s6?RuNvF}NaLBS zLrWnM2jX73=>l;6Qq44mq!@aIbRV+643f8@eWT<6hw2@!4`$=}@AwrEa!Rvsnbro{ zCGsQVJ-F%-8srKvGY}D8s;PQ zfs1Ig-2@8{fg#YFJ%uL7s$kH2Q_g^*?glfpVbS>v0TmAJaQt^IUp?}kS4CmT7OZlO zU7qlt_Le4|{&r`{v8^d0dVH2)9m3-;M8siNFUN`erRC0f7F%^av$cH1T9kU5)wjd2 zeerQ*G>Ge>hr3fGU;O5ndj?|i1IqudIJ#rV3xDyX3(K3mmX6f(2D^(A+LFVwp-kYQ{!HrJl4W`R;8)#Z zOqBSA`gF0NBZ%%X)*4BG3Z8|jk0Si5%d>5#mK(&OPL3Td8)thMSgLqcpb zLjB3)JawX-Xc?TqibXd2)d%PHO9v9>{CDSV2&2vnc&3$?w8d}YpguC z?q}>dvydVinS7${v;{C|>c^s;l##?AYhxQ{aUl)Pv^HoRY~17}AQ{&nW~uaiNTq!0 zHYqw z(|Yzg>y?K^At>-_ta@u?=opLgOMYNuE~o?jI4e_4g626^7DFpOFXm`$c!NH#Pz+pF7*Djdx*Uqukw%+@*j*%d+B}{m`;kZICjy@RBwJWRFD|5Uu z6Q90b#H!%fc2In5FYmsvD)t-Dc%3iU!xZNp4zIQ#$&t~e9$r+LaKu5_D4zL$p>!*A z67#tkngb<{_Pao=rCMCMTb%g_J)_RjMxgoa9>Xd*i+D#OD_-eng^6hwtG4&`DS6jN zK9n)9Q~L)a6I;#w$65; zs5xbM8QXW-+`#K1*Ymgy-Qysl>)TttN%#ZdqDxVj8-n+S@+big*{xLX=Mv|G1h!nS zV7VjWwbt+^sqJk$f;@ze@T`%prwBsqC#~)Vzc%8!A!A7-3&Z6IiJD=Fb#pK2X3Ni7 z)*n-KUmxWo7aPz;f{6o-p!D z^W*8qJ61>ij#L`m4#l1winE~gM{`N9?X=yy&2FY!oEjPr@_vZ$zKAsyx#+TAsWG0J zXp)=4T3KK#MYtl}8GcF}h3=7PZ9JV%YIpTpN^Qyw{A(_#&~ziB!tJ=!ap!r^)PK~l}S0qd53>1IKG)fsEgXdxn*y^E^)O< z*DPwL+cs?m&9D2-dM;|kAYX{Wcz$@|UD41G04Lwl07ekwSNIo5XRyp<`@IjD4K0MMXJ=xitW8$>Yy((j*-IZkh zz;a`l$2@b@Ya&E~;9%%gop`K~VgBmr!d^WLVwB6%)Dp4xzF55?HM3)+f8@(cHE82v zt*U=xW5-s-Pv#?gH%!=<%r*t*txAo{pH}%1?+-R#R0pzxn=h!)u$%xm7>`~;nRYfh zO;Yti=ur8?n?{9(%3MU(Ws{duPdLRZAU2P&Aa8lRc`l1CfEti!FuTk8-q^-=-~eZs zHbAfrJ&;0mrL~&O(Zu(Of0#1gT)F^UktN%%7evTaaC7H5hvj zpo3|*pM}l2^!>q9!4N7)L-Q>)$goj8GM+`zO`Tii7lAKJ;9-5pc%cS(pGOjS&N!0z z9PS?O6TENG!E?bM<*#b-x<#{GrWL@s_EnmfdO@O7DnJt^K*qh|AV!Z%-oUFKUwPSw zh1L+q?rgcn2{L(6>3jJayhkTdV1xal5{K^fM!oe^&QG(>5Y9f1bhEQUK_q#R@1xD@ zpk=`TbJXzUqc>kXLzL2}5x@HO6ojN9GiWD6t=^)jpy`7YHc$G`Je^qcFw(VZxk1*j ztM-2Un}_K}_((3M68C!+Sj*KIbO-}}wVMJrIWIU6imUXRw{TTYXhZR^5Ic47rpuZjwTKakvBoiQ}9#hAs~P#%56uRVWfD9rG?`P282jHLF@r{Ql)ahx5+jrQkj z*@k5hW_MrVT%EKNmOu@sN=xPtwsOljR%lFa{%1{s4L2zQ-=#hyrE(o=%FNL$B^Ms6 zQ$-@(4oyak@JS|ro8E<+eZ+tb+q%)lfBg}@k3)->g)TXV^t9I>b97zx=m@*^Vt6Nn zLoDQiUJmHxlv`p@hT~!(lbZ;BV8_1xw%ce5y8}3g))34Ema;w2-+xQ_xHjo z#pe}bNO0yat2wBGqyFi!2wS^XF`D+0nE!*2HiqIv4hJv3AJtrTIdWyb9$yP}f`%ys zD|a3`k{F~)*>L#eg90C7*2O~vE0gt^c0KKzseV5hR;HNa!^rqTKcCsjEv{R>X`U=# z{;^2tu{G&{s#t*!9?bFWhI^Q&5oZS=!ep}O2i&!3TKk)Yd%8`OK|&T+k%_y5S%8q& zJa@V(ZWf=q_~kKqR+r<%Du=btV$H47o`0>$mc!&p=5t6`W}4xEyR{2qkyR=va&){d z$$CHZ^0}V9w@t!kQF6ifp%Cs+YBkXfS_f7>(jT9Ns3xzYOA%oqiB+~N_rurFuuJ7BudvSu|c5zPhc3XVpbzGX}Pk4DQ|It4O`>_@iGKWzNJ(iCKgyp zVL+RdDGPIN=V!^luE+#tLdJrJ4L;I{V1z;f3&Z}qpqi~es?BwAQC8zQHu?*Ba$iny zd+I_<)D8Gb4LD@$DFv*O)>`Ic~iaXbW`svWk6LMK4e$P6RxTzr0Y|FZkHp$m6x-D^Kljkk0k;r%UrQIAv_D zR)d>t=uNR!$;eP(3`!BlogkdT{mmerx9--2*BRrW+o5GX}&pO5!NMVLu`=EzDw0$8DWPl<=C6eU1fa_ z86gM_i@vM=L*ChKtOcC_<&%f+s=sF996g1Wv1-2wm3Pj$Yc&!=WfKuw^xQ#&e2v1- z+Q%r-y%TFLS*m%wro}6cCNvlEGaF5^$@<=l2!+r|slzEHqcUUfAV*cbJ=9*$2?jWk zM_A;V4K7X${jEb~bFNLR-m8~HAM2A}Th~%8*C{xky1TcxjZ-?#v?YiLDlDpl$^eEK z2uSn6$FkehN!e1wCE*eQ7I~;_{8;{vg?dLMZg*2oF7Y=5q%PiMBgRoO3M&&{zsyqW z30_;q#gj4VfaVPi#TVd+djA;+!rNjx*=fM~AhWiy6jF{b zDwnT48j8o8bbR#aCHAKQzXmn>2vC)^_Bx4kC$NE5D5E^zAsDymmFFP=J_zL9cu%1E zY%!6e@F=kES*QI(8|gFCuzE)`C1s);`tQ*_B!3~rt^lOi0P8!HskR-6bnlZC4S4XJ z{6(cnPM!y}{V5SNVp%sh+YelT( zIf|ZJMcMQ2uVokSqyuQ%vGUaACzfQeX%{eF7c?!KHv^r}bi=CH0m;9zv|5Fy%|Tb$ zL;U`PT%2oU3^a3~LHQo`jCvc=xKJ|;={~H<*4Xbww{6ex|8_kRVg#f47AHb}W&&t> zrJ#cxr@b*<%`%<*BeYR+li_|&-ceEKAF_KK;~O`6GWSfePb7OMe25Qp@6It{q%Zug zp3*(uZtkBDelUv>#I_ekrp6@Ce1h#<#h{?|<>6w9)bss!vg$95Nd>;0R#kzxc4FNZ zvu@vbeG9?kj4S)0FrQP_Y`*h#_?sEAbe@S{-HD~eQY!&34c7Fdpsk1d;a`=~cw|c< zhfi)6kkq6eBeq6286{Ey{OV>PGqOy0hs!)i0}9114J%MgY3RyTq6&Rh7Qt$JjJg!voA@l z0mNW-AAlh1QT*n8mn7=dcq$;CDR5VKSvs2dq|7|YDAHvyxJPtjxMk@P5f}1lGu2A@ zL8pDWY3JoTnM26 zIfyxIV{m+HnHnh?@lr0U~PJreA;a4XkEKDPF~9%L1MQG znZ9-eNp#uLd;F04EPe2!y{uEYl>7S?)ShM0aEC!}yz}ht1XK-!T_(h6f3RI>HO zPX6VYgeq7yn&n)J+N-dsaxreOeDTok$zyLmbEi>)+RrXnr2BNtGIE6_mE;Wq>ZtgQ zz7c!S?)}9bV`3e)A*`e@VqE@dB_dB#^oREH8dQLA-k*Q-a%Qxs3iD^R5x&w$p;0^6 zRLdC^nOJoT$m}+-tu<%;#N2hhgIxBcdMi@b3Ra_lj%ePlx`aJWc+U#MPpt|VlKLq{tX>lo)q@l(3`0@FW#F^&Jtpxou4Hp4&z#7EcP52q5LMQ z@p25jGxanMj}#(L<(-p=`#21JlV$R)b^90NW-q0MP9v*TGXX6(6iT7v8e|AtQwN#< zlaU!q>~uqIyrw&7_w3H3(^{6|u=-+n671;)V`q|b3!tX7cD2{2^ce|5xb6~u0Eh&m`3_;1y_m%ts_}d1 znAg*(maMM+?S~wotV#nfzt7*<{ACNI7J}{OmL8~liX$Rz7jbo~|3SP&(`x16Eu-%B zlmFWRUi-XSSmM^N_tS=)s$Ea&ods1NqQR zm>ESM1)UCQM^#1CwTC?X^T4^ckhpm}^H zNe8qUkok|rLbfr^@c!nFRoz}Fd635uxdlg3#y@?eZJT+Ue8r5tOzz&f1npr8A6XHO z^lh`gnK0z9R;3}Od11C7#GBi4h}IcKw_T$ZpD4ENInn#aXS();5$r35famLak-v-) z;D0B3F6$Jq&AYwjL5iqhjM&o{^`HuwSeK~st6dS@PX6i(;Kp!=_keMG^ES*!hz$~u z`m}2ZJNnBXb09Z%3;Zb9t=kaYY;PuQ_K5;p9)KGQvosy>CeCL6oyZ@BiyrlUDo>1W zj#1Yv*~!m4OaF5ov-x64%s|>*&LWYs2B5pw(+^bYp>@f`38Ncb$CzJ^=(e%2KWY6q z)j#GY=D6Y%2bew7dBRDToqy!d%_ewv_^mYk-vsYmk)93a!L{;I_8zWZO7I%+?&;58 zS$~9;p_M+HJfTltAAY6FMoNPz1uj)7lRI&}GN>HZJ))3GyZ+O<2ryhDLR&y&@6l!H zR#fFa@0Y6-{B6xB$FNEQA5l%Ig}H8;&+e%^%6sbkZZzNpCbht@(K{6-9$>sMtt||IVvQf zxJw+MfQ5owDP}Qx>!$+af6r^i`#VNCF4mfaZTJF(*E|YpyRR_L77^p-w}7mrF-?bb z7bOHs5B#fIV;!-w`sp@5!;>yMj_E1}kdQ(wb3Z0{G;mTeuK1$mk9c{SOHc}Y&Wpx< ziN}+V0@&g+{H9}E(P^8%iDM=vaALxfy-zttgsE_@>;$sC;!)1oj~G4J*L#y6Iu4## zBcNp4#}1~QXBKLUFuEfuP9HaFMV9I;Oezrh>5?&Qwh<5gAYPaF1o~b+G!wd9H{|rq zSR5YqDxq!B&Cx)bvlW@!C-wBm`FYj&8-dROGPO8(ZQo6bEF=htR;C7rWrYR@T{b=I zFKXj%mL&Y`io5@EL!0#h1uLmhN?6~V<1)`rn;KWiJgusZGG3wxi`F4sdTu`Q;)#e{ zE~#Q8<()|HTS4NdF+me;sM*)Rm}VgaJl9L8W>$!s-*Z9>aB(Qx$^7^rifKV@Sgun= zWzEMUF(|~F^{0&vXMC9+_XVTSYkFGf_l>LsZcEywYX%ky7OKb3}k{j>^EmqRwSCxO!vg_ZpvD?AE3ck%Qx!B z8#q!2>v-F%v}Vuu>yC=yd7d3^)`}L{m5~upUv<6)GMSg*zAy1_xz173{P3?hIe(ki zZ1YTgdX=u>W;(NcO-KZ#6*FE+RI3l3HV%Bp>)edS@p68y)?MGS2^fat716RcxF?nQ z2j?gfuJGAm&9$QtN{V-mvXN`eH3hSHO0CURdz-M_W6E5oO3I?JY}eq^VzJvIRzDr; zo~YufzATs-R#DXPdx+=l60=R=3Zid8EKD%~@PKA#(&&+28G940w(|-ywy#PaD~C*n zv>T-hY9ou@L3aP41z^hfS*Wk`TQ1hI*brh#2N%zAf}gM(cY3*JkWb>Y4&7P^-c{Sr zw4HcjnUH5;u&muXrNh{A;O(+Gp)dF9HNbS9Rl_nn2Z0H_Nw4a00A|@uA33yvwNE282vnby)Le(q#u2|?8GwtEkihc#1PB#aiGQNKqD*>pDQBp z0K)6fs#?be_JF<>It)F-l`d+3(3rxz7lM7>hSvK8C|a6j*=06wQ-qN9N>do@!X2V7 zpCffwyRjoEXlGoxuF~J`2vH_?w`yWkhI}w2@3T1~MjX=!y>i99$CaIi&G_t|Gp*SZ zd7+x7H<+efNavJ3KjRrgGLh~Fm?Z+Z$Y2xmQFhLrcX_{}cm^H_PGnqxU_`uj7(1hR zJEP#tHB~Cj$zqKsuRP5Es;KULFHEcukGP-4Q%3cD z1+&!(Rx%f6a`fb|wap<@0BC2ig3#tNHczR6y)S_{clswxUfCNvvcHju8uAXLo$Uwj z)OnF^{S$g)7}OEb8^-ZwTpRx>?kvyM947WuSwu9@HbACFWMhLu3A(f_BJYmc(RjQ` z+hrR{%={BmBxeZ4F~K-;3-etxpLw6B)pq^onPoMOfFBx2@#Y zRWIw5L(?F~huHo&_Xy;cCigkxz>JV^2`o1CBqwc+^I`MYz=3}(Z-xtmC@ew@}3_uX9_Q*-w#+Fj{wU{DiEuMEi zGkYpLD1B%-hIg7Q6KTRhXf=*n_k_jUjChos^ra`0Hfd^8AOj#?Q&yLjK?u;?sPhPM zIZTsu|I84K+P1z&#E#hm=3_Iew^U%>k z!2o?*z|9s$McMdxZYD_3dGsy1$}W65&+X&v&H*JI2dW(g`KA@u#+jsbndNMKI`)V3 zLmN{enuBsrTiD+38q6@?IWB~WOV7W9oCWheN8Q&B(EdDxG6ztkY(y`2+?xrq@IFO{ z5*y}+BFl>Yf}Hu+sjUW`Z6l2;IH?|Eq-|1s&ct)N&JYyWi*;S}WTtDKot~I!_=H5B zQ94d$ui|X*mA&dCgUAxFj4RcjD$JEwjuvG@%1wUkG#d-uq&}(z7><>XP+c*~h2yks zpPO#DZ4jm(@27ti479C^fH9$colvJt;B<1zcKlQOIRN>&pXSgA59;PToDh;}2`7lS z@e8z_M2X-uyUC7{#uF*8JGqSHS>hFSobnX?+O~)*N!SU(kN`Pq3++rQus3Lnqf8Y( zBz!q;tbt(&eh`%=1_j5*#9KUgw$U4O5!$)w)(sip8~|gqiBVyF@LmwAPV3=>Ka6@4 z(CEUg>l75d*EL~nz`r~PZ4s`u{gwUb&(*9IP2AKkeeP$=a~#!;$!EmrRT!=|KaAG2 zsFT{1PQA|~8?tB(+f)6TK?S$TL*Q#!o}{Cpep5}Ey?jm)=pc=s{7O&D(oD=<<5Le* zp$(~f?I=e=gpqDV--3rbgf0;%-Ncvo`-wL3$v~*}HPJuDmF!3O8`6xGFRy;ieqP`| zdQoe{scF6bGfJo%3Nj^EwDlKhofVsqlndh8*;Sb^2GqI|e|%dYjaMGIiAi8?A`Smq zsHK?icLQF0fNAD#ztbZxJ|ch>KCN&N4Ydz`EC1RYav|GKR70j%ELB z^zHtlpI~kCEagVhsaa(^|E!7|ZrgjatUbl4zs=tA5J9&n=*WYl3YlD8IvxWj=s=Ti zZe%_wI->Y6+o0jR;J?Ly51L6D1Sp7!zw8oAo9sksl71+^ttz`n6*0*Rr~2`f>Z!jt z=jJ^-C|9GYm7>Q6>&4hqhLD%?3w4p9NkNSeL1jora)TAP1_x1z@Z>+zCD{-*qC;L7 zOO_AM1KX5^|T;ysS_eujs+Iu!{T4k>Vt2x3705;S%(pOh#gq5uxt@lndUyPMl#U> zEiZ)4djg{L$|Nj$+Lsg={EK3Uu6Mh9=|@t4#wpMzOjmq!Xq9h`1zQ@I)E|e%CmA$a z=}_v^e9t`Tic^zTe3b&mNj?uT==CH}qEBtOQ1yIr)me=i&rG<;4CpR;{ z;15JG9xq_P$L=lJi%!Yv^U9Ckd;n za&WF(uuHUi>?+$m4Q+UtJl=Oze87EPH8)E!S)|*VP&Bhw8bG2|?*y@90s2B#rVt%k zn!WeVjRZU8xKv~NaoseT%}sok9PGNDu1|h@J4bq@EgyV*Gw8T&+P(99@rQ729%LjL zflc3X*~$u#(d@ef8T{W;uj8iwMN#GCeTYUcFo=8e z&2AST+dt*z8oN?OmH?^*;;P@DflsmedIaO@-D=vj|OD5!5qKtz=IH=Bb>gCJ)NioC=ZZuvw{q;VJqZuK0JM;6p^I&oX5i=0qm9y#a;qex=*io< zGwDG`Gel|2bJ{2VWL!(5+N8s}y5}9wWH1P$%6DA6EF$68)pM%ImMZ?&nG+3Z)3)Hq z4Q+p7$e0~_R2{TC4#1)LYe)Yhu5)+6b#h2Y?-TA5rBf^c-9~BE4)zlN#X2H_9xA7W z&t<+A_5$NGpa~m|#B%*rH})086^k|BX(`<(>X|_2l(zmdc_=2@jNQJ*WSsoV3f-K1 z)@1JDcBO7;Ld^(5$Oxp~n)j*qKZ3Ye=DPXB`!Mc1Eq+kX88;v>UdeEW{mgj-y`6u+ zmb}ErqU(w03v1ws9sql`q&QxTr3z+8&NR9^3SN$@czEv}MXFqp*7#ofy8ne!vspZF zq~`Xlhg33k{lxB{mtNC4+lwD%+2wg98i)k8BxAl=EPlyXA{G5Oe1n6uaC2%RgVYVC zE<_b_BDM5m$I1@dZf}5?e+j4KqP^oP<2j#F_CA)xK@? zvcD%Uj^t_&`nmbO<>=>yI|)!VFZFXkqveF1M$FxCEL!_8SzcW4>4y+|zJ3Q#|RZ4@zHkV&7{1|krHaUHg;Y#Vt+dzg8}R?-Uv;NOP#wwLv{wJ?t;s+tu-({B_U+Ui2YZC z53b9F_agQqB1S8|;c5y!}|x=8cgZD9FY!Kh9^kWG7xr$1GNsn<%e?{n;Aun6QPME#34m z?}P0t^VbfKl6oZGClWRO%z>h(sWL;--CZ`mo0I(6`g5jDXYB=MAaLqiz>rMttW5uK;5@I_lG#!s(8$)0Yo+S8IQc@xjqPOjbMY?WGuun-N5e=Y~ByyOAkfnl%)CiV`fyOPYI#14ST;W zrR8fo&0p~o10CqWldcK$j*6Z`8KSy^v7*H!N*az~8LpnCm!E-_1?rzYdd?{rn1j*&MnPriSx~2HmQ^#gXKS3>d>psKU*Kx9lHh8if z`}Q0+%$J0}YF^>IR=CwF3UjqLpnQ5Wj@#&{IlxvIyDRl~68X%qqPBGaqjZR*HtDNf z&vQyxL#fu;(AWp}=WP(7_o-K%mpIE|&k-^fs6C=U1jj<{X@n+L%-Wb<-M6`uz71iD{iEWETVze?eqFj2VMKZj3p z>8N=reMH)ZZGnQ>wPY_3ex2$LG$G& zeU2?kP3)J#8{nfxsuLZ+@ZUG-=!)^04!j zx*dAm7yJW#i=cPIzE-uaSDkSbNi3K_=5hN}1XD-E<|_@n&X;;^p2(!}IXEvI|}SAuQUNa<=BXoKQ_otjcE7jh?dlx$=hotKXu=byu;WE6c?CotT3V z1pyfc4&?&~`jR&P!UqsuQD@v+(`d2;ba&GLpztt0B?mHE^8HOxk4y`C$&np|GhuB8 z7b=61)mSl&^BuF!fp&I1`oiMp4y6|{n*)%X+c(D|cpdbk9y7bcDt~p++CI{WERu-x zxk=hLGIj^k8wKdFYZts%jIy;N51QuRUdcPlQ^e9KakJl=6*hSzIwog8*Ez`{c1I!J zb!s?;HPI4G?I91n*&uH?sh#^UIBs2AAx`Vmk*AniusB!urNU)M%_7q7L~OHKf3B{7 z)mV+rMewl#y=;OysGd?E612R99cA|C%{p(Sdy^6V(bd^*Var_l0;vBasOL^e%`wS# z0iqN9k~&9Z%9u;B$b_LhF+?>%BVUp^d5`upG^WO3;YFr71ouW;UNlzQZ9OxB6? zv$n+Sv^5U>f7m;#u(-Bu&xai}I3!5$Bm{y>fB=OAcMZYa-Q9vagan5I0t9#W!rk2q zclSbjvCqAI&p!7%w|BqwLw{dAR25Qdt{QX9IfwjzfuA^Tjo|^5N*;E`Ya(xIvP(krP^*v~n)4kx-K+sqlbXMi+XftEa z^EQRPZO)BwlQd8cfjiOvsFc9PmbD+4`h}vk*5S<@_U&)z9_WA5n8MAtf`PZQ6w^lr z=ZNY6@i3PCGB1^N(}Xob-4_R(>g{@P!}j4{gQl(T-9CVKQVBVmR~6X!>@N55ceS-h zsLRG|)S>y(R6Hv0te0pv0?7E>E=Z(SU3VBqAHK$~)8zb!+)I~yTqcD*J=0CQRK z#*P1ZCz(B-z;QFsbmZWNy`|J_Ck?(k|FBji8s=GOAONOPM~sdN2kN&OsZl$vi^vn6 z7;f@8L=}5VE{eLT2XK9RSHBk-_oVeE^Rx`JOwWQ0x`!6+1f%ynb6zmPTBzDeGcDT2 z%7K*Urw3PXP?IVP$pUc37xM;qPb|(4KW_|=CwUl=xvnCDlI>+QUqG4ttv6gz7})^H zH%bm3K-{wz!!*?RWvUX>t!h-O^|4e$_v$PZ578oe`pDIAixQ~rm!a*sswS!LtG^&} z{B%eeX0+o3Bx{#cD||3&qP7`^^FK2hwWN!W<5f^e6MS&~TB7&$_l?laCyW6jU#Qa} znYoqO?Ty}xKM8InZmFO;!~U*Gx5Q~a|6f7WNZ2=i&e~I!_!dN6XH%|HLPP@RY4O@# zn(PHV@bB9OW0kci_NGCZO}c};K)hM3TWZHo>;rNI*~IdS_jSY3dZWcG{J1D#MTzvA zHs|sqz0Ze-?{QL*w2Z=hE|(Q7(xwjn zFiyoW-Qyl_u|~W57Q_UCa^#5_*QPQ#R2R#SlI-`$y9EqG!oODkv?RKTwS+isF1{~< zK6#Z=z^ia!*1UNo+q7TJ7ypzEnBq zs+}LjpUn1C+v`EqJ)Do% z(=ESx)q^ePqMps%;fF2?G&$8<%P5btZ5&v^pNaRNsQJr^nidYxB9sIspd&j=o*&2?X1MEd)->ssYCZLNC(uwTqB9W=N z%3!HBXBD)CgxUT(e{0-q%m{xlW3l2Ho_Drjke;`Dv-DjxcLnowCdrFB4PiFX`XtmP z4DCqIlt4uIHe3O@^3KLm6&)}B0bG-ug8@INH3Z`&_r zob$ohXRNl-K|n>Cj)>g~Mf#du09AmFL7NMD@ZBwuhw7K&wbepl?4?^fF{B`NI(b|W z6{Rr;QS}QSZ;jgwV2FlTFiCUZc(G~?xo1FbP^q6N`jk~dm_78p(y6K5Xj1Q~FKX^= zM}|!pF2jjjWApSksspu0&2tw=G<_}YfnoQu(`TI;ohNE)aoRPFz?PxO5?JjoJ+25$ zrAHU-_2=a|`Hv1ueAf%swrFV<;rnrWke*-vgmn^Ky0d3ANC=|(V z2j%y=aK?2BN?P?gt+3E`Pg!rhssvF>G%ivcygqgov|#WZ#>X`=q~0G~g6W%)Jn^X1 zjb~wz7bK$+qz9=iMSyV0$Jiwv#SCs82q0tGcO1Ruh?7uIpA-5{pJ*sYs5f*@F!LXM z4^-}s6IV}40UhGw=PZ0H+*2MG36XH)EIRyq+}6!6PiD49&C}^r>jlEDR3f(m;wiFj zs}C9!cyQZXmfT2i`!UHLWDwB_Da}*ewXE4)bQn#?aTI>v7}~C}dQF{3eO_GCy=2cO-jhgwAIAi1{l^Vd()f9FDGo6;jl=_qH~wSz{xY4fQ{K*bP%=u>cD>s#W~m9D8`Eqc^Q_wLK>Y1v^V%c&Jydo5Yz z)UkLir$jC11IOu|$s=0pEx0DPog;)84yVUof|(tAfK;sn)rNHBy{^nkgZ9YTRVc(o|M2j(Lms|GSsIRFmM^lVdj36e73 zrVcUheun*2 zOGPC}0wVxC6!F$!VtF{WT+G&gQhDX!s92q&oHy}K1PA4KLxuen_|qN&Ts4|E?prTo z_EDmDKyT+4QeU9TCnfC~yxq8TX=?Ctg(#Kqc%} z%(O&Msof8M6++K4W14t-RXHxBIR45aVxYmw+V4KB04oaXybakgvQk>~Q!zCAg7$5y z?xpQXlPrY7*JhI`=U8()aNZ%Ey(FyT>nwj&4A1ON=p`C1I+q<^;_b3l0!)J6{X+o4h~X{8m|at^>ExPlzsW zXc=E*o7ueXP(yBF0=5$lK%wwKE?>fHIK_9@v)g4q4NhZX1Twz0Qt+; z%>8;i)&Pd!#UqT83Jt>_oH#q5+ga7XUPP2{Pk#H@m^Cucd(KCo$^tSM~ zQ!+L0pdN*CsaKj9$!wPC^SMTufXhKHdX+yhb|&0@OEPNTI2XQIn;r4Z_&O{(EzFG8 zl_wr(kgb4(4*7o0ky+uMbuJ^nk^-Vb73d~Im}$p2O_Jyxew^H9cb^hl=aEm)8I5bM z1hPV$=LuQ2c4S!4<~qMg>YFYC8E)pZkjjg}W|Oo#<;lzhtF}QVOrDHiE%QXrP5jhE zXy<#Y(eL|!R{wK1s&kxX+orbMtMNUM6ilL0i~aWO=YM9lMYV8XwNAy5oLU^nKojk|+`Nho0v56Lh4>Lx>*lv`OgEBUEG#p_N z&Dy6m3^oarR2@XoOHC&gp)y12xO`({wXXoBjY%qy-~8?$JM@nQ>AiiY=$V zY{vbzSuF_yr1+h2A??SPB6Ebo*AHJEUoJW|W||xp_NwIy1d)L9S5yy-jum`rUijlU zfHGFA&`I3D4d-!_=2JVJz&=mEOZtbnY?ecPwe@W}&e+_1&?Gdz#N}(fgQP2yuFoPT zfR4ij&~ZQ)m##bw>&hjQEDR{d%~Fd(5OF9{zOo`@w%|7)g5aA$9HZeU@FDoCOawFC zcDILA9{TRb2*wT>ZELq9ze*fy%^e)(HUaj$0}Lo?-^P&euhD8WslH?>zv|X-L0*d( z0&sEG%G0PTJ`n+}t4}?`a~Daw9-sR2I5GNI`{3II;;FU1xN8x?fwtM$OgnErf?r`F zE*HUDGdnox7fUsaFJ$R!({g{z1AI$>EN@XiPje8Z$9yz}6@%?hpdCgeYczbVX)#kf^8r{V?Hp7WvZxOoPh)F(;$D0UJtbv^u0OL$Dwn>l)vp(KAOgqSw zz>rDI)3-61GsGOKDA*N_#+!uiCvk&G9Qdv>XbdPwR$Qb~3F=w|Hb%RjiLcRz1zZST zpRMV{V-BovsSiRdwZ08JXpyvIXvxB0THxL^+4vc?<(D)Mnq7OMe_&5QV zi>i}*CVxeQS`Nt*Esd&(czv##VgWWh-Mb257xe4Ge9?$E`l}S^9h-J-#JcS)za7TTy^^1a<5asoKZgz|tX4HmhZ?CaE($dAQpn9BEWL&e-MJb6 z%P(?d+~|Buv8RELVH|iHRZY4?HBH-!=O-Kvvn@xGbQOLs_~|ZCT5b1pR!w}JwCsi6 zSeLO{4L~$XJUXeqDAR6#kJ*>nRI*zS?-;7M-rbTOs${+cx9Vs%{eASUZw*_tmk7;Y zJf-`e!P=5Dv1%=my3BtRZNB*~ReMHm@6m0q^~96|ha&(?xtmOYDQC`@Bjr4uugMKa zawx2qVkNA+^1234m@Jk3Oq%F(W?jp=x4Eas)j6iVE_0$N`PyV$2JB%T0xtu%=mp%K zAV{EAGNoS)E7AuagDWm3K|`#u!DvrzPz`HUlvei~oejUTjCBa4_;9y|d8WAhc~L#& zCVq2(P&KfwHAgnJaQ;dq*qNuYH(ZoEu63M<6k3Mn_lMR@t-SS=s%taUW3Q;W0qCsP zNL4H4w;q>6$I5jD#oU;eG``6U6uP2XR4s`6Xu70f-C%R5;c7LRpId1e^(a*0&LUt} z2)g|p0r#B1!sm*fpzp(yaKvXU(eq~-eDil>9ZtWHp;XiZb)Kn-7eG?{d$C4i8wZAY z7*sc8NsJ>C8QK$u?M9*V%kLTZEBC~m*=4Ych4BDh6O*WY#e4MgyCv^OPpN?SUfNhJ zmWwyi^?wK;r7Oyv?{yrjBMm1?C>>2{Q68t7TNmX?@MLN!~_o1Fabi zt^Jg~#_;p0i_}VU3TmG-PLp}{K=7*u+jkzMLoRZ(SjC**j%Rv{0#$*m?`(`^I)^TY zltQki@_x5*#!o~b4ouhTjN-KRbA#NOp26vrTI!amG#b>M@BjP}_0%g{EMeipf!TKk zs0u_2%QD0o$K?!|hg=}@SPYar@3g=j&e<`5pHH6n_H1AgXl)9@+}TP(<2tdf%?k&V zCos+#z;ipW>CtC`IX7l(B^+((sZMHm^oM{Sd&~MKRIc4K*ax#FRw}W<6X+zaYC+On z7k!n4yjr7p&HfzUhVg25R2VB8|5v=N%-lm>AQu|C{acLd zNh&ec*{2QNmG!Rh^3u&9LTVzDL#49?R06e!Gxmz3_~hZ9tgkZB{-Vj(azr!SOhKwd=iQ*;s6?uHb?E*0 ze74TK#AyFfEl-t8 z;`wg!8@XRRJo{za#!Vy_@V>g@7vW7b+FG2syAPTTS1&rW(0;T(d!L{d}aX7I;2{!Y# ze?mvhWxO2`(1^NXGtY(0KmPDsn#vzn;oFi!v_-2VI+1E^L0{KBolte#eP8)OYpF?? z;nQPUFm|?Oz@1KuSxzM84*MSG)*V$o@f^(YHc}>*{h&bzkw~Ydr^(SA1%9%hu^;e1 zKIzoj!yoX1JrD%piwMo~UkiOG9c*R~@#m7g{4;27uYS5tx7bxa3L46isEls; zU_Rx&bIM|NO};48**sqj$?9ilTWiMxmX{vOvo-%cpH{L22Do7~A*^~q;V&QD^mSo( zXN*csXTAeT6><=c%|Z9ybm}aOW^eaUFdyh;IB!jIoYl?!3piGR-l;#Gm6t#DPV~>D zi9AW%QiZoby(Qw_y5sPI1X3B*e~fv_e8e88LXVncqRTzLuRGm7EN=8YSqSPr{zTcF z)J(YkoOD<=Uv9L7#&*HQRG+gkpx{eMF;5ft2t@^R)cvy3wESq0<3y1N6h`5YNw&@W zikAe?V=3G3ek*Dad8*^3iBjtV9mrmiFJ@E_+NEX}n+~qm1&)>spjwI^YiPMJWKD+< z4b&Gyo3o`A6ukQpjaa4GArHgVYU=Yc1ZcgwxCp1z(I-oxYV2k1o2KPA1 zeiggyw=b-eABxL0fg_`pb=NmhZWn}<;oRBhnXgo;mIMG_zvp7Uo1)MRE~Zy=903baBTKo`2Hl>K&)_l2ri8aI3~iLe2<5+rw(BMTd-ojtm&|l_n44D+g*A!e+NW zzXjR~ZAGrk$6(j4B_i7CORq)-sYCXr)uYD>X5<5XSiSxMgMu^Z6Ad0~5u2~GyGWI@edCQJ+|97kTJs!^9H4N_Q!m5J1l8t^7F!1I+@#_Z1 zE_t_i2N$MATpU2}<4D!Ij*N5}P~5KsPRxi%XxXM0Bkj?~-v$v4aWnQR%t*CkINhFh zhd|HJiv-jc!~DjcdKKNyApGtK!h8hLEv`EeO{Lozz=2k6t5-+;J==q8Z$3B0155hivC{$=`ojZn%fh z)Vojk6KE$MlTV&cJ-vLffC(nE%G|q80;;VGTJWBagwHzi36M=&PY7U&v^QfFF@GB& z@V|iB=ZLt@CpljGe}a8`eQdfA0ogytN#^SSv(bze{r|B=JIFdzI0*ZLNLc;uml$8tdR zsAb+`cKpmCi=`q0Cbj3}!x?$Tx>U`Jh{R^Pq;`4W&hg{roD$^WCi$?I4V8JOu|_Y# zNu@DPzK{JBEcKMuzb#^I9J>KC0VeP9PVs*`tM_UDA&?-l%<+o-=E{$1|CkWg0x9si zw~pQ7;H}u|`SqyRG`{S?L#OA_NXV!qkSJ8-A%9YyG77hgRT&u}j_K z`C61+D$e!F2MbKZ<36OJi&C6SS>+?4y5ib5*)|8UG70^ym_zUc4$Bs(t0afFN6q6$ z*|phL;XwATHwDlf(ElCg$xQz22oFDR-{VkZJV!hv@gfrj3C$ zz}opc7LP_4$FBK&eK^Ux*Yv7au zdYoFn{T!t!kxWx1ob*A7YA(wR6a&z`0Mm}97&v8%ijp=uj{J&5%wcx;q1UYHQlaCW zkPHy*w)tOkSrqDoA`);lUo|x=a(MYV(E$aDKv_~3=dZ{>QL^j=rsuCqi4mE1>-S!m zL{(H{^5nj*%@d9zC&uu42;c|7c5*V^m`3vr!z!7=^x=01L~_o6khbFl*1kR(Bf9f+ zA;ZB$I+FjHV?`x3#Ht&`j?|RiWB!chYI{mtt;Qmirr=3kmf$!+s3VAD$wNKN0!zT^ zYlfKLjUzmTU-H6qQd^CZ*$-g<^b%e@_=oyL%k#Hyk_4BV>9gMK=`7=tPgy{mq+e{Dc7k(?^6bxQ9)Qq&u0YMlqm)I37Wgr+ zwLe(x*d}Ny53y1wnLOo~^u;eMih~0mb{CF9l@%w~FPg?etTq*uR`|6*76C}?We$Q& z`=l0AyPfrGgG#!0-kpXA}o@3Js7_F_ z@AKrZF+!MuiO8?ko4$s&VzKfobZFEAcAE7f+QgD{R~eZfZF%KB@kPT~C%*R1s3W!H@lsTGn~u*Z8XVSA)$to2=hs}t`T^h&sx;F&dM;e3 z(jSN&ZvKs3{4wWD4j;YlPT&s25q060*iHzzB39Urn|2dKn(gzx|0?V7GQPn=mqTP| zTGyjr;F|8fArT=AcMNBm5_|{KSfu(iT~OTj*7Zz-a6FSKd<{|hxgR5z-g!=@D8kB2 z$rnXgc={l!{5wP*4y*WPcqRhGUw(|kvHgzxqBonH6Q5ppsH45%aygdd2}TCMt`=Ke zo~OA{e7F1Q#GI!;F{IPf16;e(v(k+&A6I&!>VeWMoZw_rU0|g$jZSkJ?`VXezpfc+ z;JKf$@Y%AU^D_x_t-)5qJaFQsb1C#JBVD>jQsRj^^#$r%Xzg=OX^thSX1C~1qpIgt z+MxwI5q6yy#6t5JmV@IHvtCr@UhEcapBh1`AFD5BC_x3B!?J+G&&nt0Z44clW zT7J)&m2Yp!JhJ9PVW*p;(V=%^DD@bm>quXk?8SaZu|(LRlx+VTw1KWX2qrch|1r5V z>>az8?qe{_{Wiujr9H%q$Y|ZU&E+Zg8)dM&VbD)=^@O4E@kgN^O8nGcL)UrqRC7Ihmvu4&Y362;bb>m~e@-MUq zQ-Nimz(I0+v*xk*Q@AAlU*F)4mcR~*xyvmVr9Brou0D?=2c_QacV}9DY5_h$ARd|Q>25w;IKJ5<8 zn)HnVa0RUkZdcD>8$Sgpe#g16Y_0cy*!YSe27)^j|0{mOW(;QrX#02AnG*GSfZ-3@ zg8ErWA={*4DMVUsve%daDtZC z-ya{W92xYE_ZtEs6}RT2G%qe1K$?*T;Z(cm5m;4+U7KVN-+VV6wt8rz1U7k2z0vmq zglu|p**us=Dez;-fSH`XA6k0~dwui!^Vr4Li~~C}2UJ?^3fBSW=A#(eVyY@URQn|o z?3&nPYMi2wld44p$Tka&VL|v*OLzSL1m|gwOT26nY)q93{fp(uAp8O7wnOc1c5=Wk za{BE5fe@4fzK{xnC;99aly}F|*uAIIQ8e&3!}$UankXxeIV|mUEwan6R(^?-qoWwe zuUT(q`8f+4@O32KoHC#Ni-r9UFM&@$ z{Gb2pKmYXqf5JcCjeqz2|KFv5cW3_j{r@-b5>?Uhprqtq9UVG8VMUvq8RGj&@9zJON%l>X2 z#^272@&5H6{&j8p=by50KDmgyKUi-6?U4Rus5bdeT)X`EXQ_OD(klR$_J4X9*a41< z6^3AN`4?yY@2^U1`3n`Ws(xXF19|_w(F3c@I=}%mK1J=>|JzW%Jeg(x7f@{yUXT2< zE!f}wvk&6SGRep=Vilw}Y^4 zje7uB;<3LC5S{t^p9z$yQh%`ezku~opa9#5@9_bN02%CWnI0StvIM{kmQ(n@8I>SF z^#fBdqOuDA``PrT>3#Bc4*E|mfd4X5te9s3N9kF$B!BlteCH1YHlW;sT6ji(U9kUo zI{fQ%a%y0Dd=|Im{<}AV+y_k@xD|nobOSH{{*}oA^TY4Qs?&cB?_Va}e|t&z0~kJ< zt7^iBzkg-vzyv{}>^%6_LH_%B$np*tKH+IOlj7gMvd@6^U`T|{|NUuUAok=Q)(v&D z|NfQv0@gzk^Rnjc-!{>n@4)bbTvWXt&lq6`SCDnc<1Kf`7eC62l5yFe6~lcr#^7kS z8H|O$Z5=W=z;z4Of5&QydN?%}%cD${scedAeIW8J>@8 z)>Iwi^NF4=UdDWn7F&7`@Bik=e166OrtqENfCKu7qae{sWiNhd|3lXQ!{2mD-}&ED zJXNxzZoF(}gsk6vSBL};;6v{Ss;~o&sEO}B{~xC3e~iTD2e2%Vj3Hz{9c$kIZ*2F! z8~A^|Z~OE;QZbbbe(-RQ?p5iJBH@0oT5iy*k{HU$LDkOL;o7%w_m1-WY`>k+@FURP z<)W7J%m?cAsJY~Is|njHd6ZZJDfb!1Yrb}zN<3H^uz*GQf?K`6TOvaEDU740xed~# zBH>x!;Fhium)|gl{l$>K4?Wog{M{G4<|tyAyw*&)=1)2mCWwY|DZftsWi#wdn=+sT z!f@PdMdWp|p$zcX3fz>c%wpH~D~3WXs`q|3jWumO>y10ma9uqawcChQ7Meo1KW)bp zHRzuraXYg!loSU_>=75)b`F|(?!c8+PBe{)G_sRfd~BY$9J((=!!sPcHhxNIM)oPw7mC^ZtDKRs^T4*8pTUC&(pzOcN}Zqoh~@_0`lP)tict(DC8!^_PBC6iY?$^(%Hv2Xt2iZ8}fLdxUu zKI)zEkfejBM-u=$`}7DJHn`i>7zP)<(lYFr^Vjx7ZoHV{KJ7}rT?Z&eG1^}DN({q} ziFm&K@w9J^!H(I(sfqTl`b>v;{R}TE%rBavDteOG3K}0yzru#SAKvOt(rY=$LDQFP zStsH-?F^=`WQzM4ezFdO+AS+`4Nbw&TtBccABzIdJx~PKw@E+0zT7&zQ6Q+j=YG*e z^g=SaEnt1Ye)yz`t9PoRIV4Sp)_72rI!ZQ`q0^d%W;~UkV2&xrIH^LtPHock@jfkI zkRhhLCO(uFY6L}PY~=5YrzM$gKCudCZ$l#UD0D7Rsz|!q(|!!)2$3q|S-KeMa%P=4 z$lW_z@1>8|@Q9v&uMsB$e~l(hH3CBWi>o&f_|q?#0D;bk$ZdeK&|d zAl+w2RnyQPPqOElaSW*iT+I)!t(?n@XAEo;7m0>fUgQ#XX2!vf9R59o8vcm?VfY8=uN=ilUTHV3ckOX)>gIG0*c}yW3o|7(VxMV2Z&O$9_xnEvWvya1Bzb+3wJIxIgJmQz$*AXE4ve z*B#M}XgK`dg?<#<7SM!r{@c+j)829Sz7P50BcgXiGLvu|&~pySZ~Y)6_%cIJIIQQo zti6)pP$LQon39!gszv|uiL3s}{6z!ukmX9Fslsmc%$1qQ%1EOyx2x^EjHWiE{1)K} zU-jG&-7n2-(zLL~a<*PTyOaE$|u*4Adhi3S4(f(gD#_% zTGsFp1Nh&4L*JhkOk%TKE;Bst)XFeYG32BVII@d1>yc+eUVhk&sB+l`?J@1*8Yk`C z-(Eo!1^y+?Txsi#=Z~}8sGsOV`NAIy`QGax5mZ(>(1?9_fg_O<^CH-v zoIJw+W3a!2>#{ZVx+#Je-}3 zoKi!~*O97zbc>Ra*a*Q7SR}F7hOhS}^3WxHHnd6*Oe6Dh?daxYr2S)9D(O9t4SI{^ zro*(xV86MyuE(}U$STO}(n(O5!Q$9gJFfS2j)G0GTHv=o9DgiLTs%JYGu>864;axvg9fmJ|ru7pRSYnQqEZVaTNnP%?^aV_BA0dwQJlTWM2Cy zYRZstod;A}oCPm)68XLQ$WjnX$h0?~Dxb77CTVy#qid&LH`bMBA6}K>_y^aZ>8{U(=`nnl3=oX+K%YdFJm_hotwWL%UaT&1z3Bti{z`Q8F3beo>GTKR9ha%)nb98<|Hg^G0gcjUq={j?am0AI9v`-A-Yt|l z#3?NcZjN3hjo>i?Wws;q6QyDDV4N~vd*L+AuAdM-`rYHR|s9wEC z_7uP#e@E%rpB=F$z9l*zz~db3ftspLBN!Ml)-R#$yjWE( zj{c?WOfidvL&44WRO}tZID<#%C_0mBJ1diPJJ-lzx(Q-5v){r2U3j}TX$P^kSjt}1 zRgXG<|A#A&`g$kT4JfZDRhw(tIZ#t>7H81?z|q=%5U*LCXGF;D%56Yp=WX>H*!G$% zyycsLP$3PS!xHGitc~+biw|1$ z8CGjhFLlC19(}!zZM|>$v-3wdS%mXaMQ*0@LVRUhpT_--}_5K$xVPr7)|)4tzDRPK@uPjUuWMI6wwW9DAC z{4JJvtb%#|6aUDK!Q-I;iuwm2Vd-&o^r*tcbLqca*D6K8p>bH(kbmI6mBZ|FNz$;x zj`4%Ql%M)i+UjgM^YnrPD$x}l-_v!$7=UgA%vhcX=CrTk?v2LYs*UFei=<1a^87I)wpi*G>Aaso8j%7#C4Y7m&JgpoC#}x zoS;d-@@oUBhx*mf{zqkVzNJ5zGD8}gHCU9F1PDN%vafvRa9;CfM+A#ND1O%s;^gi}24ufy1|X8s4z z(EgN3fHkr#5~gSIQhSMmhyB*T3MVbn(b^%;(YAWc(TwKVbJTr#4EsqX>8+ECx}6fNU9#y%8~U1$^&W# zIS2}ikNJ}^a1<)~OYh8nid+!ad^Rk$_4l-N&QtAS!p~Ker<8oj_H{L0lo;tTRY!Fc z+~CxY9|Vc1n1H10F810>%CrJX+zIRJP)+N@ukNR$@kG`GZ)VGv3a$ONe3htlN)$X_tym)@061lH>AvLUWDac04nLePBhZv^&UG<&*9Y38G#HdaY@@ODmQFUm9t(2pRm4^W&5@8f7t&SUO6D=I+c)YT`ynyk}q6&_D9Usod1q< z_{X*R1I+#*{T95{Z*H#LaTfp-ZHsfLadl5cjVfXEa_RQgL4DhOvbe>b+QS^~gy>qw zbu!0=*978my7O5x@?}rFa%>K#qsI5_18yf%=YR>`DwWWAf`$~q+s=WE`|FmRUGNe; zp^`h&2Fs~)C89gdThbXBxC4TiyiJ%eZkK$O&?gmg%y#d?14VXN|!{H5k-dg~3a&yGe{bLRz!6+Wljr4vVy;yCMi*k0v)|zZB9d2DSKV>%3M_baIy(faUe#v^hDa`%)LSu8U1lJ>;^7p=^jM{;7 zO}d_6tRB-cwxwoe|M6hR5qyof4A?3NTQV`!4fkkMl#Fb|a6TemGoZYEg74CE|GW+o zVd?xyHq6r2nHZM*m-wj6E1v`Ir5FXuFiO~E??wDN5oP=HfFlF`vr~Z;b>PH_Aqfke zWJ?94iN2FA;$EU&t_EPL?H{49vEd?u^Yy!Be@vRQf4WUbvKd*ql$%Ea8MC`MV!ZbL z?9_!09|7l=iW6*E?NfWcpH+4mze`tJqQz=iwO*&Hds}z0!y&XuuTscuTSiA}WTiu> z>%c|${0T5Mu*Xn@ti1~x7sy&|9k@zOX;b2L0(LMn0+S$^bQZBt~X-TCVn{}{sppJ)nU6fKr zr+2B&3@UZ?P-UZBA;R_)oY><3gHN6{sq*v0_N4pww=78p0v}(4a%=E10pT31J&~wn zYAm${L;-2&(qj9iVii(#?cy4T(U3kOkJQHvaXQzL7%D7OxhXVKE(tt(Ez(Og@^97d zxv>r=(_B(gl97U zk9SQKS3IA)dC_*wvPOp+t8Pi^AYS&C2W$mc8|y7pK+%VFsRe7Fd9aICTYc{umTnwF z5~U^C;X5J@OQ(!7ufel&tNQ}rtnB@!M&A@n{^)$L@LBtWM*YXBmU^pO>U*VY5prri z3cS`@Bv;o?W$wExOs)%jdRMKD#~ok8GRx=K&-4xz1#czss*Eese2!-_r=Rfzk2pO) zA@<8CZoZMIyN`6QoV?>4mPj0hX2Y54lw~88tHcH!Q6S(bx43L#hSpUO4Z4 zR+^zD*l1=x1d4u}6%jEBf7h%rS=n^h69BaCBdL}3>Y@XW(CXLKP5ifm3_mGMhQVy& zs0%K#q$Z16=(#TAk*P~YCS4OKsbu3$uE%ZVt~X7DymlTcM6V5B%=P#oU0xcPwY8AX zPIDB$XsChtT;+D-KHs?DsD6Lb+>`h=*$RKgG3F#lUn2eti|B-b^x-rZFQ>J2INeh| z!-0~16Gk5qIq7|KPL_{oLur9-(AMTeIhZsNufk4LvJuh7IfHl( zdX`J)k#p{l~3#Gb8M0@3sq049QRx z(akBmtFI-pWV#O68&2>CO=NvwJ*PV_%D2BZOPlpc7>rB|V+zBQig%tUoTi|`Yz4qF zD!W8rudZ*LMwmXa%virl;Y;H~kWfL|4opRWDj4(+7zu2xBLvs(xI&Pok8jlX-6d3x*>c?8_U$*L2J34(rsZ-!9fGkbD}l!)|+BS zyZfbxdlI;faMI7N!Z6WTP>V5I#qZOo<^a2x)(3iLwv4~mu~O~D;@PYSvT^s&VLz57 znfh)$FFW1!HdcyPB#EJR)_D_4WrHh8`jO;1;Wzw=`{q8Zo~}3gZNc4f6AkL*E>5-- z%3Kif$uUa;Od-p~)iU7J@uV$LE#}yc$bmliIbzCFj3EiTRuPq7@qJ`8H;q5$>2d$t z%~Y%?@;8Kjt#1e!n}m43Uijy|C7&1sv#Cx~b&*7#^g*cKI?he; z+%F)fomrXEdz3NvWM9>dp>tD+ykky5s!Bm#4CVHZn-r2-E@$)`>!F2LJATv(0rlr7fK} z9yBAoVg(Mpw62ueGDuG8rZ1Pme0|ulJop@mck~DgUbFI6 z0jNq|PRtwM+Rkk(@V9n1#P@KX$eSPNg}9V3Utjl(mLL=;x9IRwePuo&vcZ;o+H(%$ zE}o~`H6qn+ZR9sI6wS#xfnQ33Nr5e=6CdJu0-Tl!(o1>;F>mt(;I&JUYSlZ7Nn~zM zEhz5tUa{!UQy!DP-!W!kZ@oVLB~N6Nk76(ggI;mM@0)Zyfq`msgip9B{jFhZ!+e4{ zyN|}Ak^aYgRY#AV%6x(Qj}qU-*z}?JTX|roTQB=9#utXooM{h_U2zAgww$nTvR|u) zG^QN+&uIn*OS&Ti4s+EM=NDr##K)Q5B$BC z$NeHn7IHTTNE){A7_X{mzwBR*Xin}0F5T*+t!>6@M+MYh?H^q$gP%o$>%2h(38ias zAML<(OS2ne2ia}uJ@~U|)+-p6$58T2cdEK1Zt0c*;q?(9g^rQC`Fo(QC-E0!`rsIq}nwe zkr`PGk_}U`FH_{+WMqp6kP08c`; zH>!Lq$|VVP%JIVAcZY|uIe8Jc)X7&=b9uyU6nv!`B13LxjB2rJFDMB(wj%dBAUhyc zn3w&uBdR&N7UIu>f0;i%TgGY|ru@*v?CvACB=B3%MOliM<%&u4Cq@heA#4fH4uY9W z)Cy-y@7%pev2MS{k)%J}eQN@NS<`3sx0u`!aJ*++^^_+WAM&JHD1qo?T0cntu29X94 zBqXIfM7pG72q{5nkWK+fm1d}62m$Ht7`lfcXK4Nx_l}?U{_ST!_kQy~o)^zCFBpac zxUOrhbFFio@%g?ifvHcHd^Ksiptp)SV}D-HTo7ytAb zKlsUcHFj)P^TDJeRLCzDY&$=p6RQ8k8E-&v3RONZEf4_RN+1y^X|*$p-B;i}GmiIA=zs=2HFIf=>F(=kRs(QE$8$Z}MM zlUa9=U0t}9$1m;BXiB$P)&Q|1X;=&XOg@3=&SYI1i{=j-dXZ82leOoRPSGhY4}+s9 zFJCoCU8u?~>b5mKk9DT2;#aI)vpO(+m2g-a{V01O&*|qz0ZDty&%18)OP@SC8&x`# z@ghbBZXpZT;WY)*ZTc)ahzj7*a#oXN-+mo^w& zd(pTLD~-tD0W~e>A4_GtXn@Uc|6DFG_QQ7DiQ9vicPHqf1l)GuwdfgXyfdYdR_FZY zwDY&mh3D}O`@9tb;=YX83Ir!GT6Lx%OnqR<9^Lp1<5klL&Mlmk{fYU>M+Y|B?Ro`i z&QFBQi@mOg=9|G-O~<&H)msrG_^(Sg1t&}$3kELjaX}+jJg2Nhups3|wK5Ke;jZN@ z(}6_MUqN2)e|ym$A2jd1RuO^eSczb6VtI0w3Bc?HTmAKXj99;rsGn#px!9yuJ@%1e zmkA~2^-bFK73G7ZUamJ%9%ku3TEn<1=u0m!rgi<`4S7m9+uKvUU)^Ww&?_sjb15tY z+35}C*l83`CnBiUE^#rgl@i`LnKEsxb#_apn*B0@7l?I{GlIEUcDYQMp;^}HABL#$ z9qfeIV+K$!PWn0hxcibZjKR*esum7EVNMDVQA9cFpZKjCe_8DZ?M^s7aXB|MrK0NK z5=PKn(9ILZIZ@HqEv`Il8gg>nnSBw7O_C8zn6-dUL_?wfU~yC}};vNcHsKp+ERp`F84w>BZ1yFsi7pEyD+9zQXFt z-6kQNG`pClyp)FpLyGJmEn8(R*@QFkjH}bVCLwfpa)(3SUm@pFtviZjI1f2SuJ$>- zKc>W@^NM8dRxr(by&H^)&9r_$X+l=pqZZ$v&2W(Ab;U&KDx5Roxp~MH;+A+GvfO?< z1Jb8oeVybOFc{8egp81AJ?dd7#Bp9%A&Uq`ouEkOU-0Y@dbqxuK{$yXa?R6t7%s1w z4hPhlWQWX9Lx(;(3eM0sCF!I1!$kM=<*|Dm8swmKd&%e1tGBwz7&@d}k@qmEjyt&e z?FHPn7;$G=)vq{ZpI5-0#LcJKJ0-gPXXZIALnFyZEi4-Vij3}wUu~ONDjGU}JT+xd z_o=?@EjIvnFZ_%hnfExBQXsr6#BLJ(5N& zo>L8*x0Xj+<$k%6N*jHMZ!ci9E|>eUbfQ_9OCt8i)MnV4P!(d5gI%fo=1{EnRgX$# z$nIeDdfu(HB|**%NY*~3zXD^ao;40*}8)_L*>+KXsQKj2!Vp!_oFg_bHEfK5?X2HKM%zf={GH7CR z5603NL5BBbqSQd@ihsbSuU@+~6AdFV`L+ZZ3w+E^>JeR78aJ&XMa>X(?hUg;U>9LD z;uVa*MG547#FPeMwTH1`JV@Ra>|pqg1oX5zx6cP>`NoCC>dc?a+wutlls+bL;L6lh ze=N*Q{s_H^=Oqj)=+7!M52%;ABz@~j{mqEqd(ny#BqK-iW%A-M|y=4{oEROe}^V;9S?HWhj9aLyl>O zo2p;gOL`CH51Um|p}5ijw50dfwMg;owl#qS_(~)QpU)1?9VotQFS1vCIkhgsEyEms zX}8qg+>|s?m$YXe&Py0Rca`6lTyjVvydJ6Zb~vDQ5;%mix4VT|IEaYYh26{NmZgj^ zY}84f$~Owa?S;I1y4l88F0!{{xwYF{eu&rrk-tjr##Rim;fL*Q6as;(FKl!JX>OcQTpo z1Pe}~%Sp03Q$N-q9_Q_jSPz)`lj(L>nKtyEo=zy-atgH`4w%w#5ILkc?fuCgR^;SJ zLxg$q9afv*cv1s8_h~P#?)NRX`DAa2fp21t-oOVaJkI>6M235JWzu~(8_Q#~8oYav zd^QI04(wWHnYlzKBq5+2soaZ1ku=6=fe2+)b*&2K5R_~mX$`)~n1zPBgi)5+4llKo zS?=QZtlX}}BW^PG?DD@AuKOK>uLizH;D@Y>SGhpPTb>UNAa8d!s=}nI*G6&sT}JI5 zbh%q<vEBPNff{tG?8DX1c$?;#Q7GzkUt; z-7s8WvJ0!iBuFySoh)DU{aj;+LhFv52R1x0IN|x(5rHFf8h0ZAPqhgwi}Hzh_qB%)RzpC%`e!yq>*K3uoI>jTy!|B7BS=D zovjIXjly^BReauAfDPlopUDUsx(Cw{9||Uwzfs_r1<)aqFA6sAzvJ3>D4?obC-hTG zbg7iM2ud6`KwjKq{*H_|^yDcUS#$#^>)YhC*PIHGt0Hq(+znWwJN+@Vi!k{&kOn7_urT<1&}K<*_AR_%%)se4k*`T6*cg6vQa{9bsjxj1ohpo78#4< z-NxMECb^49+hplf>)aJg{$qc#7!SQ{oLgK1b9l;lJKA3-Uk3N9@$ucYYqVHxr}s&- z%$gMqJ4wa&`Pmi)J)dPakC@VTXvn+h;CpGwiP7)*ae?sbAUPa=gwusiH7!Mt@MN|J zUVpg+Hm2s$m!kDGy6WZsIU~x}EBJIGC`!H$s-yOdMRYR7waR8@Xo3Ts;_!!Adw`tj zG}2;fb%lNE)71G|lg)L1+4(eI{alYhuH7>Q#lZwOT1Ii|fC4p~LwQGkxIC({mYAst@No(Oc#uxU*!nrM=` z^x?O0_~bh9qI_*uPPip%0AD$1iJPUjMrG4^ezwPG6RW2oZF?=u8)LKOlcbTCd*_d9 z3|&c|_V;Mt_TF4{I!C-bd3G#}F=jgMr7GH+{nW10q=7G}RlBz({Q zKxvR@H4em`ma)DEH(y$D^Kn13yA&;_thc9$2As$rkjwYJH0d+uZqD0BI^7zFzPSA- z7VjLdSE5Q!?E$y(`)BHzB_>1BtBF$Shi_X%n}2)7)O$6nTqrx{2plNA!h%`SnN*g0 z+0dxJ9XVhj-akflwi=k%F{w`?K3_PUY-hCwLf1ZpxA_$yqoLPV`WglCKBzgA8oa-o z*Qr&G`5sUZ!qi9YsdJsl;}$sk8W75~%(?&~V>!X*Uhrd%3diftgRX+Ct_fp*hFXU- z%v6EICP;%&k`|;}{L z5`JD|W21_9`S;nhCY^>0V1y}Plk(XZ#>&Pfrv#TdaqFRCh_d>oTDy76Y*9^u%MT0O z?k`fta*%V|E+BUxG67sUCacyHI)}fdfNZ7B*9Fy>U#-NrQ&Ep35z792z+{B$YI58E z8Kg!Rh`~!XczU(J0aD1EzE_?DS2-4*uRh?Czr9JpZhj>{K1NAQG6T7@!F82r-&e&j zmjdxDL9hAh+Y zs?}z*>x;^Ygk$1}P?Pf+(E~i`FUI$!HPXi-jO-;GdR6Pd-u%AQM~#N3%Se1d-Ms zq?}qtqwLSUPFfO&FUzA$-A@*^$0rBG>%0g7IfKmRjOi!O#m_e01CAW$&z=RL8zSY*PFr4p`>d2I!J=QY zJBZ>%fGuxr(QiF!JjA1zWtPG?mS16?KNT3l8@&s4opkWO*tBF-Yiw)^M%_ldtjeDQ z5Prw0$8r4IN(Wo@q`AlWbA3Se3GGO>o#>%R zGkTW9sOGb{M{#Ti7={**>kjanAn^6mQN60`6LEyl6*`*KAUxF{J+uBkP@qK4obZHc zMZ~w|7!yH0FOck2Ss*vXi@n+A2rQA{A`?g4EwWL5E0>lG9ug*H&C?a~%fs(f6srIn zo=@vzlocX?q&Gb!ex=6wE0w^nb1ER3ixA?>bm^aRR@gOAYt$hsQf>P7nN9zdp-88b zqjKQV^?H{7Rt+o@jC!ZxgJVu~nb zCE%>zp!Y(+IUGt&p+iwGLC9&_(pVt1<^~+ZcHITH7^|1!+)=$JM9v=7Poij0T?lAa zj8Dj`l6+k`pk@D6y_@e(r?}zOMi&ORH^ZJl>G>~Y-?m$3=w-PX^doVj8^t&K?X}@U zAnh!$e$hq8V1Cx?)ei*n2WUeaISOGZPZHSVu6(>G8}ia|Da5t%6*Im<8XGLb4Se=r zz6J>#d-qdLy`9@y4Y*Pr&+QGJy1r6Agf23B2mSLXdT+@Yr^%QrXWe{-w%Ov_iE3a;9ACTv?m`aNgBAJH{Z1fuN~JIG zYfRg)=(3&R`%q&;t*+=6u)>cHXq+41-J5XBQEohQS;o`N^W1M$t8tYtQai;TXk1Hb z8u`lMptC3Ur@n+Dd6S)!E?%H8tqJUN6dRfclwA72M?SrLwej*u#=P*<%Q zAdTWD-HD1Akw1wdMQiIqY%u%GQ01eY9e=h^+S;Zat?=Om^}m?=60B@6vcC%a@Xf~Z z9CbqIG?T0ASv%(G&g$59bvBwQCKEOLj_1kT%(gezSiqEw+uPO?w>8gGoca&_RmA2` zY+G(EaZA5RW{_=)bAC{~**|yXOP#dV_lW5pM&ObZydiA5@@cGLz5*9e$&AC&vIBZu zS$a(SK_|p9P5^C z37ABzP<$T|4nY;x#hF>mKy&ErwyIlEwn3Q3i9q#%z_@8E8E1|QDg9I2gZl?sG5aLM zHlsF$hng|QU_(Iuyz8h48|{rrlG^Enuki&^Ar$@$aP6A27kO>_?}!napnfPhFWx1c zl*G(LZSl!H6bCEziy%ZfXrF(pi6r8sHEFE7)U-5_TDGu1zV#Gj60zP9FpeJm5m1X3i^(+K3hrXI!r28mhBk%5XmD$a^u9;c z)L_<=qY5oMBGMk0Ld(Zk%_6AKkHu{WUwltRB`chf$#cu-@}ExWR#@#j3tgKdgw8UbpTAQB(`ffkzeay->8g&<` z#BF4nIQi&~TK%dY9eup5TT_0R?khHYs?b9!IPjR-I&y99WxAp+6{}wjq(M2_qxoEKuFo&$-liQw_u6q%b|VuFbNsWb7p4bdJ6bR>KOPx*e+@@IX) zWf961Oq$&!(5@08S6>9F>(WCu-PZ#-t~lczC_ny&2IV=6z?4+N!mZI_I`daas(c{l z?nnU(C-fsV^OCS>sIFd(6}xZdhDBrONABGzc|WR9!sFf2EZS#LjRSLyX!ju=zYeJEO6gaKesqxo3HEuZ!y^({VsD1HkY#5uHvf>G8W9d}8CAfL;i@(1$FV!;p*0 z-eqB8`uB4l2-nkUyh((Bn5ZpWr8Gix(1LW|duV-seD`2|sFFM4T-gwiU}4j;CBoqI zv0t8$r~Le^?YrOHw9P{$36a_|7uQ$0E{{4me9Vq-;Z_8+Y~9AY?~kQS*jV|t&d4K` z&X6$IBs1fwRv%MQ_CQ~j9G}!>H_1sfKLI^1V9S$JOkYDBXK}o4mndDFZSz3rZ+E+z z+2JSq3V=GKH%l%29cG#oL_TWdpRLCJz3f`yOWxZ8Yi+UH)Vk<;Ga z%$XtxvPa#f=;Ix%(a_`Hk>jh~s<-c)c9T7+Y*mUkoj&D!i;BNVhUUo=R_}6ycuFNUq2gg>KN6RNfJ>71^??=E19 z+b3?~X(kGKJd5Vq>-+9VU$UiCF<>M5frCj$6!jB+TaA%*BY(S=UTStM@J;pJe!%-M z$#Akpt5SvqD@9!06=imXJLMbSp4e}6Um2`jv|e-rH|lWUsX@yQ zCi(wg4%C8`AY-A}eB9^>k`iPkBX9o*3}DVsUr zWNwn^i}*)Lc+#)AW@g!6y;FzBZlw=GU<<}tp?)sPbe?Pj$6$e&azXx z)kjqwGXrtHri?N*l5Wp$S0Bi)Iu_>IMeGz+&y>M6weoz?e@d9f=%LRpd3X~;T3CQA zX^Yx=!IuB4Z*$a2dbOP%OO{6a{d(R5ykqjYhd~g;;w~8Fy%zi0EXxVGI8(E%q$1!+ z?S+#tz6%uX5U+L2V-7*Gac@D>-PDS6DqiWW4O)K zR1C=YmBjUjqUld`JJioGE17eNm82%k9u5#8$aKy_wv%LIEEi}%P3Nb0^_g*}6WdLG zn`Lu;mv9eZCAlP+cpVViy8tptwjO6@IE3Cb($r4RVQGW<`>IWNbb(e zVx${CRu{JZ$1YR%8j^FY~QTFmyS3Mp@!mj+hJ5mT+x(xDVJ_7 z79BgMri=A+4gNC#WBIVA@FE8u0EvcV>Q-%X4}6jPc+V6SQEd-TQh~d6x$y`Fmbn^3 z(cwpZ4dI}UTxIa`u6XoU#g`Nn!-myVnKV1+rPO$@jJ>VG8-(^fj$k}*UuyND{r)P7 z$L!673FZr|{gLsy+LEoV*4QEkZ)9rmF7m|}Y)6j*f+wMa){esOW?Mf}8N2S~#5XHQ z^X($_0yPnS8m_1<4mJp=6 zBy1UDf0jYoh&~8J{8sp|9BOJ2Q6s6xrNS938|h&9ax9Hv=mTN+;BCuL$E;iil54oF zAi-9zQ$yqtMb1Ye=gx?%`)VbO!Grr@KDhfMJy0+I#r{$kq9Duzn4U~%kVhv~l$UWF z5a5e?EhC3WT}s~o53UR4_AOxgdTWe6AI+R38im))FK_RvZkDwafyp23eA)CCGZI)9s0dW znnI_grUeL2SBX{>M5QhS0_&-mU80=HsfyRYLium6_)%riH`WCBOAM5w9v`{@Wnbd2 zQ%$7(`3#=WJR+PM@9{8F4rGwO*187x21fSO)ns|OHdzNx=sTDS?Ebq z^*YtMxDlF5#C4)QjnnKGeEtciV3x4jH_gGY%Ini4V=2Nh#ndH0ZP1s+jL;9r9kni5 zb{?Zhn59socG+{1-^$?c@OOa%&g6}HC(CJ^T7r2#m$_0OUZfSn`+!q>npj7rSvr;( z_t^f|`HzOa+lAN!*xw>a^rCsE;xR|cO2veGqf=8;Z7j}@YUjyp1ve;z5kY6ofEa8j z*mQ&6q!+iNeW)}e`uxb>xfUOW+ggRlc(mH+E73sNY+}L}M&z|pCJV7VZwD=CBU>}{ z_GJo4?6GEy^A2?Y$s&JIsQEzY9t1&ZOzni>{YA;D6FcbmA`=RCXw9>ZYWPgpp%QZNDmiGaSq$b<`u%*s0 z=?x*coZt@|N^{*y7S5;>2^q6udfLgP zEa<7|u75o?(4k~D zX5iD^yOO`ea(SrPNB#DcMQ2Y(e`222Nyd+VNprq`&-B49{ab)3@m$3_KYbzkGbALO z*uwS&VZg7|2T_;bW;6CW#4K7;Jb(&Tu-FLHea^x=F}^NKX!m@>%rXXj7qR>Jrr1K* zDB68vILZWCmpIS%s0_1p6)5cf%7yLTq7eh}mzc zx2`=jOM#nxWAX^YI>j!6InhjTprzW7_VbW*qDAvUd){B38eC5l%n-rDEh%d1<;pd}2~l zk@PFty_|eus$5}ME1bs*d;htRIZR+!D3b;wsFe$Y&P&Y9=X}g@_Ekd*ZNc{$kB~Vu zC)u=HrY7(6wC8HvFFRCsFR)9VHyggfhOtPMccbqHAgjbSFxUr;0;#S*)qIN>CB#Mc zpelddg>rHriwneD)-uaSH_Ja4l}tFi!w{a#srg76g%qe2D->e!-TcDgP!PMmzZuUU z>-?75!0CLxQ_(IY?)b~*uzjZbWb+rzJv}zrpimAL(TBTFjsH--=lwC+`5K!ns!oCTok0QXQq`Oo z@O?bXVg}=zj@7KvN22tkhOV(Yv-R^+Gjr(X2#R{pI#bL6fB0^j2`@9T%2r-8Ar;t*kqv(8@T^csXSsiA63|*tqE@(J z@15nm6Fph24uV_@BZD=1V!9PYT5XBM~$q zu-UP_0NSiKK#D1XGBGN4X5E}SMB;wE+>+$(Km-nADL8vKfw>GgKc2xgKb6_X108gl zmbIhbVJ{F;;d}R12H2=Xs>VRv#s$c&4!(IX!tC{%06*;ojPtQK5|)ig9nwpK)gT(( ztBZT^S%~oy&a+jb$s+U9uae8-fnwURP>iGPFEGuO78L?NCJG}jrOZb>%7~1J%XVXs zd5DR>PBL)XJj3`QA-Tz|4ERbj!2!Nf`c*asJ+=mIF1A1=d{ZOny#tJvQ%gj;SM9jsZF5?1bWxP;2r4x4J2f9(|LCChRn03y zqO3p=>4lKkU_D(7rEw!G(PmGat#c9fh<+ALL9jo$EHVtT-(B^P_KLp=SQ`R8Jl5IW zs%Czg@Gzhf_4PF%tpP@DA<=hKKbRnnD$I#5$WekA2_c0zubGY9#u9kT{8d=hi-X_x zdVX9hLKpVR8Hle}=3AWv0@+{S@I8P*L5qfgH<46^aIw~mdIQ}X7(0i!ohzRn^yCib zy%xJk@3=3|>x&GZ+wYnq6WtN}c<07`r zNon>;Nq|$mom9pobYQt2|G7}$#iyD=J^mME0?)Q$4QB+p%~Lq}TV&j%eW-T_H46x+ zIUDz06oLGe>VOk}-0Io3G`3NFggW1urtWzxKA&Zj9ncytc)2?zfc+{I-Y<%YqJ{yL zEkOrGbx>=w1ys)65w{HMo!-~qSLiYtki!ySAojhZ-~1VK!4+ulD|TLa0ci4MJ9sp= zZ9CeHoE`y99X8iStEt$_K-=o-ab!uA$RXx{R3G)YIYZOS813Eq%8L6<)$e`w+mrri zw((sPy_~n`^NwfNe=72NhPS%gi^P&rsV)?yg{L}v>hkx(WP3}~b*ZdJ44%v;%|$Oq zKGa7&Kds*Go8$FgBuyDGQjM;<-HI2WI=om7jW zk@$gvIE+^X6bn4rSda86wg~L$*~)m|0-9>nm0BJXTB6bndbzsLB6X$*B<8hhvg~*=d2osS3CRm^f#HUKoow5;njvHYLX<-C%U#8Xo`AjZ=mv{@!*~$1+G(D zrdJ`C31B-#O`z>(A6n6F_j z7331`UJ4f|>1@2C5_fKVc>f(7h=X`Yb6ju&al_p~gBqWgIsGh?=WfKZuV#R>A+wFO z7Kbn`C7J7oEK#o);bVjX>wTH*R3gWO)dM-iO(Jrk?%zo$i=nbnK8$^z{Q)k&Nf{8f z@hNG4fJLQ(@ACpjwhXm(5@Iz0G{P~%-(l`k^!}scBO+h4MJ$fZrw#F%u`&CQtkDil ze50&dDdb>P;}rE7$kO=d2<$L8J)er?Jq#j1+3p_#v$%C$2rBw!H!|wo=d?l!6Iyq3 zja6Rhw`SaXcM&Qrfv3HUzKTT`6z=i<{?e~7)7zy^WCRHbc?8kvJz`p1d=&9&BrQQt zEjs(bqtAn`XX{~XrQA(`W_Xc$X8<}FYl3lF3$~RDz=Sm%RZq;NKF0+SJ!^OJf>|7 zFZ)0^Q?JQK&a-7rH9J3-=9bnTY##-nIC3%WFgkcF>bm(InYJ|e0s-vlQ=su}OH!u8 zF%LCU%NJOZw1+3aV!JhJpWkDEXv)vvzHy>HYnaK|Hg8y7{EEu@Io zpK?};MZR*{U;My)bWgHA%B=whSmnk=F~NIOwll&kS*hopfQnH=mG%JV@KvmXw<-y` z6xJeH?_1o1?ua54=Hh7j1lEQ3XXuDTo?RcT*UnQ*Q_es_D}rP_Id!$Mj&zkDrjbcJCk0ijImOQ`fGm8;~o-ZK+q^iBV zR>vmLPtEsR$sH^Mptf@p((_a2nJYb@@NYi*LdM`<>Iuc;#MO`h&UhnT)C(Oi0#M$A zH;p_j4Lqv%+&dQ19K+7$C4~^r%JcYBJtmrr+1d;o@AF`PTy2Ol30D6U%oH7EC` zl*Y@G$ZSgDujxJARM@t6ihoG`C&0*i^d$v=k(ci2+yNGlb0sF_G8{_Dc;+*COYD4n z#;sM;^?nWmDPIit;W(Kq(sg@Ua^Ozrds0V0I`$@|FBU{d#VB(k8LHSt{L(yqt5MwV z!Xtni&qWx}Le>Vz;S#^38GHu*$apI!pI!hRg=&Q+SJIL)3_fint*PQJ4)mr3FAj65 zs6W4Y(8GLhriqa2AploVb4n>V{6-}%ypi^II{(uWvDAJ_k#AQ0JUEXkznE6MZ891N z3p^ouJGdym*R^;MId+OJk!*hE36?eA17ff-ztWh&la<_r`NqsnPp#YDlT`cXu`r(V=BY;M5)iAnuT%{qQNn4ULiRtk-b&mA~!-tSHyga9a4Yy2YFD`aif>ZHu)zw zwGDiVjsg$U6-!E|&5j#u^F1< z`NRn=9E)ZdUY)ylOC%Z_Eqrxk#V-)dTs4?!T_vCKu?D$sh*yR;8-5hd9w z5Uhh0HDLs**5hC*&72E&=J_T41V;j>k`OJn`1NM3+&Z4N5oqz=lr(i@GwtZjJ|+Rq z&_Rr4_iklAXzfZuJQPEN?0vs`jNA37n%F`!|MOm?pb72puh%4_pU^3eZ!sAJNzUwE z3G?TL>uI7_E<PN%$_nVDmm{D=tOG`6U49QX}K7D@2A%|6c&oH2cRv{5W_H6HtxzX28;2tv@TEia( zLcJVqQALLRX=!GR>k8@5G9P>gi@f9uaEho*Bjl%D<%{ zSfmJ8I(<{a_tShPyWcC@i@`b<^P9uUlu6;XqNyVW%NNTj}Rm0yESkptZhS z$Ytg9Vc_H{eh?$v(L#3zXTAy!8FqY@g>=ZZnWW^_Jq>!TZ~aENFp>&LcoD za=XdVdD}IZ1n8DRo{iYVJhW(?!3cU*_NKe}3PSw7k;wk+#+y*)8u(XpmIBP#NAO+< zLNK5zP*RpTv9evwosj>FUYXK4Z$Hfm;@$PB=WfPD zNM-qRh7!;W_0TeKnQG~&DT?aQO60atSK9LQpz)Coni7;h79(>N_s>o&4Avpw!F%)4 zhO8IsVkLgysA7#UmYaf&6@wnl;)_pdVuy4iUkRfHgIj)mTd2uL(RXluZBT!qYlFvi zlq|JtXbLfur*aMFIFt7KR3(gE*RhG!u50mgt)*J)&jlQI`SU|AC|05wRNh{U$X;Aj z^HhL!Q{jhxBg4W$+N0zU=ew0!bW3?PEhZ1xW24ty>$VDIbZrNkl;9`Q^~T>2Zh3pl z)&w;mn6Gh?VO;`CC}Z2Q`A*7J@`I04v8>k3mxEshUcw2!^=PmMQ9cEYMdAF^|K|Fh zJcgz7HM5D(vJPIbexUT3M%ZWj?yor zZF`^3m14Zl-W^ChUJFyO#hh!aX~R7j&=A)Eq1*Dn*U}smtE%eY;*d-!?n{F)f?qC4 z#`)8Zbrm4|M}nH0Z8n5qG)o`Pa0cjD`38fQ=E~%4r<#_PAfN7H9H-j^AvY{-s(JIs z(MEZV{UKM>@-}IsY*ehV<2pB08X8q}dwuRZlRR?{(H6HdNl8%EN5c<2I@6&4GB@I6 zfw^%|%Xsi|QG|%t=y#s1;7fGwU-v-r=ikHLTlHmDsv_E>4*SxyZ1=I(pFWOBMF+Gu zMkCM=34T~6TD)4f4oD(-nsco_-t%S)I^UiJ^nf2F`w_2y+D!SZr9FK5hurJ$er}u+ z@UJU|LO=j&zdz_tJRN1c01w#tb0ElEs{Kxf+oRY--Vl5$`5ib@^R_3O{8%2XR{6#5fXDW=)_h7X%gU7~*yrC!!`Z!MLLQ#nz>`*+r^+%~Y^}E<)_OeBZ@cMQbKtg#fl%mL_=jn zi)e4C$*L;;sGshb09Y&j2Cnss6zF)|nkZqYb6JZC=WJ572S5Rd>|T=NKm0Yhw?$-; z-zv7gzVt51JxZMBWBtAfpW<%%Q#SAU9czFIh7c6XMfeMI`WHdYm8*qN>6NRW6pJcR zyG6=u2$kQeaTb(JEpbk-jI4op+=8z#4T@Rd>6) z62Lz7e9(S&H-B^2M)U2-;Fa^xOVYdb3O$LIYI-9FE+(_Tk0|hdnC6Co+18m>;L&c|#EZI`id^hF!g`47SiK zSe29S?zXR<=;l#In9$+Ft_~%~UkTa(AJm5o~>{`Y$8v{=N+V?*aP#lNb!F^kH(C zHSJ}@oxeU?g>Q^CL*lti)?b}_{jR{I7&xMAKKiRS^!)$3|D-_UL>wrbwiNcy|LYGx z?#_g z6Vf*)^Qyd!|3aQX30NmHm}qbq{`wL|$N|{!P+ZP`fz^L{<2I%Gut)qCi+&Zb>d7j6 z+WqS*#qbUA-uj@NTkU_AwcoAg-@PBY*(W$y>;F26UM~J-1@G!?mHX?%`u`8i|E+=P z+%LcIJ5Uk2I?pl!^Ve?Q*5%Pk; zI2y%{a6Z)wWpVMdzqsegh2Cu5n_qmdNA}fuM`oyedoP;kM?!XcE|h-ry8qqO{O2St zxG^{IxLFx@TV-@cV02U_rFw6X1E74 z6+8+0Yky|p>CN_`Oppv5fS&V`{Z&^?jvlypdGxjQ^j0uUN|l?~{d--ymKg~(^a(#F z)A6n4-o$NQJY|fTKXmqvzP3*Np3F`x2S(S)&zq6$47=HOo~JT3pD`*)WIlaRr=Fs% zyb2i^l`?tjgUua%lJn*6KbGY8;Q;xVG-l;WG(H4D<9{Jvvn!3+W0cGu zTIXz@bm#O?20qcVxX2a#D~ig|y2NJeNlyr=!GhpSy?ue`d?oZI83+{4=FI|?q)~=d z%n}!Sxd3@zSELl^G%;{$Umfy}oSG8=5(ot1v${9ob@EAQl17mM-5Pg1$#Ztxrn57O zsS5K@mzj%GhuE8hB{Wy4#YNQn$hOlua?7Lp8o=_ed)Q?DY^7HdRml`(Wz+sX(*Nl} zgQi10PyIzbkZDhlZHfgB5ehc*Ah2*I)uvVyWIw!|jvoP$?H6M9hAPvJcb(y-tpIV` zF@#=u4p{ives>R7W9_S}t8<#JBmwZOvjQjc#i-y%S+BT9`w>X$UykQUe|AI@DqsiZ z{~S0+Ju;+T*ET-@(uT))OohG2$8U8{R*8DBRUX`4aar-;3PK$(g>xm^{-}H9y!MOf zrt2$-&q$#SeMyrkv-@$5eXuZ(Uc870epeQkN>%?RXB7I*0lK*ikf>N96Aq#e$Rm30 z=W`kP{a`;pWo2)X%gtA-(8P~bg9T~;>Kh9PXrIQw#=Ur2p*Uc#b!Z>=u*o*CBKR^= zKi}19EUM|*FR#?B^&@>s z^VBtV2HU@YKH$h8@hd&k{67zA3&uB&?dJ#mZx<;7alIor_qTpLrhm?izx4tZ=eR!u)X$658)P1H zP5T{G@~svTEguO@20A@m8<4pEy!4f=3;0UE@y^F|s?neVk_>(41nZU+Bn&B zq8>8`aGt;zGKL`TCD{JrvLS3nv*~DKfX^_!aun^3mk!832!Z)a&_$!RxI9ZuIBOSt zy1rOk8+yR`4}$c%mUE-=H*PDHHJq+dn0No=@vcGhqae{(dn6z!3l$|^Sy!y86j>d_ z{y4?lVbTT78Z)8o)#1qpqV?vx-{ehqBdDSm(di-Ei$CH~Y{O!J{WEd7@9&!f-T3$8 z_Tv=jF}Ad8WR8+w;M0rc-mizVVvlpG0N&d*R)3x-u~@=No#@p$A~5Bv&T%^#?ri;H z2fOnHc=;eC_0%4>`?G_gNOzFEUa~?z&-k-2l0j>~xXhVvI-Bg@eSl#nIj0bS+=KaF zUxH`#8)AL-7X5DeP00l;tW@LT-;x$xh$g^k8k)&{(^2JWa3qYFo+Joo18 zd7~_f_RnLe9PiCuB(nxzR3vSJ`E1)4&18ps2qG6v?jq@0Kr2ui)0-ww9N9XQ2Fn@I zPs?uIx)Ix0y}%`}Y9`^~Ccp#`>C`QZr^FVbG$P0}1E5$%9!bGPP|jT@a<_~UvA^8; z^|I~zYB}e1-yMQamLm@dx&Tz%F|XP0gnYa(@_I)HfBxdH3pDUqbtAnxJ?>Wj^>I9> zl$SJ*Z8{jZ3i9?0YjwOJ70(C1{nY7+|Ea?1R_Y3BEM+&mifghS={U|i%!RjRiUkA_6Cs;ei){)CiCjGD)q+|lwt)~WVJE)$v2<9Q_@iLt)se|hNwsk zwxHe&A)zsg?YxO~qP_-d+v%J-S<&vFI?LqY{tQ0rf6FV@=>!Th`sI4b6g+;z`ffyS zRsP=WK^4}4VmAZpX*0JD4&19s-Bi`WL>{2vI(pQsw>%TFLZB!F6`Rg?LGXTCZcD(3 zsrJYZn8m=lBi$ZKs#@hh$xF$e{cvat#ze7X+5Xd^&shb)6B7fhL`Q=qiOV-}pp1Uimn%W}Po;ctS^~q~8CI zaKBtJVfM=7{ZINc8T2@1%Ac*xyZ37w^d5M59edZUZI!(3*|ufaWxiZZl3hi9LT#k< zjv?yQ;roUu-@!G>f~I*I4VBJsWXB)$HnJL3 zD8c24b*ve(>0$YY@p2i??guiFUNTIIijyzssrADqO|zobhg>^4Qg=sdWxqK^j7fwb zbg=U?uk$`L7m=1!`4{jliYZ4HaBF{*^*tvT0vGG8;*$$_CnIigRB2N0?j`SdyMWy# z9Fd>P%m>WugYsm&y`m?v^AZrL*`Ft2wVQ^3)|aI-eAxLk<+HF8K&QL`xIb#kvy|yF z)MR5M#lHOgh?B>C^T_WgF6#e$SfO32!w#N}V~gRuq1Sr=JFq#tr$=<|#m)twp|JAr z0QS$DE|DUTZ^9R*uw`7EHE>dkVla3V7pbB#d-YnO^C3%&wnJBj`am!qu`kg8gKSi> zZcUQh5PgX2;EfN~SigzqOLfka1clg5a`5Du&V7h{;v33XIUP;OcU19rj(vnUpU54% zQplQ(*;ka%Gxyr)sl&5v$EE}TWRESQy5654MZg(mA#}RJ`;LCa<&DR^=l_mm-!E$k zVC-`PJ^mjt_QD8cBSbrzY~V9%d_U^fxu2I%9mt1)L+=--ru;x56rcj6T=Z^aKi>NK zqp#A_;sUEB8y(!zv_<1nX1qQI(DOD@ZkfMIz@psT{exhS@-3G=0U`WZBudu9e2i)R zsatGra|6(ibENO%8;U(DrZ;(?xGRG%bX^P6lg0z#B^x@|WaiOFz+NUR)T;?J<61H; zicTM(JH*FhDtX-y0Pgl^0kq6rFw9FuEYvrSvFA9B%a~g1owPpFK0$YgjM9vi@koI< z#{;(6HHtf0+Md};nAkpY-T%YWS%yW`{a;^2L6Eo+Q5qEi5s(s)1_MO8LArB5YKCq_TDnWR zdx!ys?ob+LV34kvaj2nrj`#omJ=b~3E3P^F?7i1o-_LSMTGPZTXX(4hwk1GZ_IX4X z{52FRmx_~&dPm4<#2<~k3GV#J+DFqUyyy{Z0c5_~Ngc(sXo~w?EXGUs9gW&PeZj>i zn5&Df?*(!JbNsQDnTYcx5_jIET_?@Ul+&|aH;F(?^krlw)(Kt_nUxuuC9?N$J9_=pwajZ_c)5*loYY;!ijQ^O5CXW2>+t-#0FF5-33C5if|zrA3#6HV;~N!#j(2+5elP_d2=>tUW% zKaKc76Tw6jgPionvq#d+E(Kbr^h;}(NPs=ZJ~?{bvk9Dlf!ddN+Lnv7fG|e z_%@28QpIK$3w(J|TUWPW>)V#-lNALS-@vxqhzBnpG`)$_5l61mwu}7$J!)Eg2+}jX zn8$VEZ$SxyWU{&hIv?J$AL+*@U z&DEI4aU(|0yzhc?*XX8HSU>7$yYD)ArRNEjC7p#YZ7|soX;)~={gVxJp_PEd?-MF@ zFO`wqg3Oh!D@`B6HQVgvO&==gKjDV+fR{*Kz&jpb`Jqq5O06z`h%J~{mKaU;w8Q0ZpfBM`vf+_43@q`F%1FIDV~NVj#mN zDZ#HM=l_?2Kbpm-b`NDZ9clczo=lWQPF(OOC7XkoE7Ls!s;{VDm0z(^~ zYc{Ljkzo(a?(9c_AjpbEA-m6TOo*q;zy~xEpiRp?wJLe3cl-l$yP>;J9j9xy_^{H! z+m(sp;nqm6dem^XaS74mlxpW5Ij!sV81@~vfXqupIWp#F?sl^s>%~g57~d%g9UC>< zS*XsfJ9c~MD2etbB<88E-l_vJrQMc8Q+oW6rKjA*2+9ChnqcW*&ds*29?sQ1m8nQJ z&>X5p2P+e8V%gP0biFLo)*xzb9H5g&St+W9p}^+ZYyzFC-?sKfeLouH*hT zQPr`L`RYV|le>EFJ7a+MM5&TJkPhV~-Ff1`ViGA&l+@=mS(@tRw?~`^Y5anQMY&i@ zqnLuIIpgf&COAT?Yvfj8Sl`FM!1d4be_F7cc)w1=zYKF}I8&G?uc@^ZmIAeyJM=Ym zXI^pk2i2b1!<0ZuC<20aQ1hGp=&uET%H8)@?5JHi{oD7aT8M1xMje@`*Q2rY zheC71bMzRR41OTuoUxpG>4rD&6?N z*toute%|~0uEyuz&iuvz7JW`WT7L$iZ8qfojZ@*#XdILR5*L*&B<46%+-njJf%_Y|*f#KBCu@P6Rja18@Z&@rS*0#Omm{tjO4hhUMhe<_h?T&~hDp!5I z-h`u>69v?zdjF=A6#&(Xkw;R!g1GI#?*n_s@1b6SKHzsiN@-XfqCb97Q^^dzB$2`cu_B)^P1PkHjkz%j-$Nr6c~bTx5K(`ghDudp?LrPt+9oA+sM3L041=RqC%Y z@C6XbUY$3T{e~*cWG7uxXJOJ?~B{7tePFq2T53#wjCy3${fA1$jecX?V-iOxU*_4maL1$tGJUQxdhgw3@=6E z$&LOq#vd0^O4C2M^n2dUclB_lgW{q~<>|+RYOMSeBQ+{4&h<2JMU3D>a38m2;-~r0 zI+J8>!)1gmLpk9flfPLSCZ)A*jT*i5xVId)aFq5scr}G1oZ1-fh^nN{?qjr?Gb^kB z1<`fOfbS9n^Vu|+_tKIke6$=+igOZj`bF?ujyY4Gg9B#wMQHEx0m^bXLH%R?*#r!0 z(kXpYyU6blS2XsxE9KYQqi|`rG~Q~}LN|{2Jna!v_aHu+yIld_K|!J+(8Zi93Crd0 zl%H~=6JM}rq{s<5&aFM?cE%i}M;QHKxR6f@WHhew_ExZXY$X@cr#!kgjY{GdDgjgu0K zhv9au1dPrJF>T)Xp~5O*v5z${G3c*N`UPC|!7Iw%#!-iX4x@?U+(t7hfqFpO%%#@( z7t4?9H)_sRHkUvk#W~BZrt?Tm_rv+-N?rEyx&=6xl{SA#c--d%rBnw_4{R97wGx$y zK+p>XI-phI6q5jQ&f7qnW>q5G!*OKKIvcX{B1>jFe372_Zma#QBbtk=c44Pa zuC`PP_X=E7V1}=Zd1z>Is^?KkMBK`&K6WTMRBe-NSQ3(gU;eDk{Iq_F9pI68BaDrr8=BFppJh&AYlnYFO6ycYbjy zTnv5l+|*DRk7WW%-Lk!EKV(rwC-Y20x}K}@EeBJZU;-;O6O9MH3EBnRD+~G)tueLy z34?65+)lq9d=wu&Q;{|dd(rkeLgw{=Nk$up6J4b2#WjgxR=(xyxWbrV|Jn6kf#R7e zXJl5k7|RRIfle>7G?62^qZ?&XP`~TEJU3ZCC2g^rO+?M;i?2WCNi)pu-%4@AOlfnz z|MLcc9@(ZU=tZMn#Xq&XBSH9v_UHOx44q*9+lazmHXcFwL*eJ5^KT?v|Z= zReu!EYWgQr25~U2H_6ZosTFtYki0e@gJMS3E7m;25Vy2N5AAXT^i);O)swD2{9tBb zwcbYK#rNsG13h)&SsdeISq~eAW}_+CzSV1P^Q(%3D&8Sv<_w9;0-~C)TFU+=pp9U3h%dkBsR?vGF(mbJyOG!Pli$6QZFuuTJY0gYn`U&>Lqqj-!iYN+Lsu%e>Rsv zrKZPl@OKG$WVL@T7dkz z>L6R+jO}NFbkz)pI!k$Medur3L3PHC&sNTwViy~Ef_&_Mm1aVvR?GKela2Kfw%ba~ z_yA7RB~!w5f1gVD#o;i|7%YLG3xInGfQ*~wL}Do<&w6U!iU#c?@&&fS>who=IbF(* z=5;|Ym9j&75|y7yyjQ` zp~wo-!F6#yt!k@z79<+rQ6kS4t7ELcSKWNpa;D~|gj9>6JPzpWmOuwvkac~4lG1sq zV_NNEf3S{ET3!J=Qf5;QgHE|ibM$H|>sVCN`WJb*pGQlTZT0^7bvvxHTP-GVmVv68 zgVh^jN@_>mgc|$B z_(G=e&S_vI7sOHa^nF{yVkz~>m~g$@>5JGqRqaeS6@Y51A)o9gPKj~dz`G#^?_8fr z05l2ZtpGWH!l$1bNMkIkgMEcEiTbt zBLK`tFY@=zt)8up!(W@`D48ODOKKaAlX%{LMn&0AAd~f&LQV40qBI?LDK_ttO6_Tp zYz;Y|P*!Nnx)cmO;3>{3DW`3YZV86Bm-=`&n8vM}_V@2fCW-SMk=Qo1ACFS{V3l5D ztF9iaAM||{jk2*{IS+OUi)Hx^8eTH|?PLCnoX>A`+A2&)cquRX%V#8qrMJb-VG(00 zv3J8_mAO-wO!|Zm%akK!3z@kE=g{qTl#I({v>cb`S4g40_8cxyytS;mEp;KP!O3nX zU!G%LfKO)rofN@#rFNUuTJPyS7GNitz&yTZJ(gC6P^s=0i_A1FO?Vl7%~@4JuXoug z7mfUyZabOne9_oar%{Mc|9nyT$Pio+t^d@ot*bsYcX89|>*3r;Ru=W5|C^NmeVJW` zdFp3>GRvjDE}J@y{4KpUBH9r&FGCyemA#H`>Xvs<#nGN!`Vern&g}XL#Nze85`Y-#ulo(R& zLrb)uh?!Gwah9xdxWS1WujXJp`#lM?X9)AIn?x)EO0Q*sQSY)y#`c?FI01@FkXXK- z%yeZb#JaL(&H4xWalI$*(#FqQ7R7{>JD+G39ub@o{zXo(-V;f*2$l;i#y{hf^1rgw zUyB>Anq|O&G9>U{ifdj*a_2_RpEj;lSgJW!?|PMEKk=_VFj<-Mp5E-BxwYR_{YLK< z&`WYJ=>R#${wZniu;6V{sz*VYl=Z~;?v`idq%fZ8560r`QvE%-HfPqCZvcT3_B8~K)IDt;y0jYiA>&1>Z0DU-ms$5(yh z%iRmJoH3>kuYTvebzHi+4)ct*R}K9M4G#L3+aD9nO~^qh`GDThTkI-~|8wZ}6^Mew z=Wx(klzi5}KR$Z$eXGF^gl9QE zL1LtiCf}!9DP}dRRjmz0aesCLuA#0)d95K69nry=MSiCMyT3(;6JYn}JeIXa4l8xD zHQEYml^Rm_GH2>crOtq`#}&iRH69uHG#pd(MvD1Ndb&|$-7XldEOPW`^C1<@d26?? z2eM{5awP3r&bxQ#ER@-w&eFQ8W0~h7@1w&^?~hYr^X{B74j!JZ}4?WelXCKFw(a0W-%=ElP$w~-@l)j zKV}X0;KD;6`72n@5bWAAjb%%V4S9Yy%ovaXTvq4DXOtKKO>})F6vr3-aMIf=y z(o1h-KndAU2@Jb9-?V8sxD6q>)ffizR@eTX{TSwvnkzo{yP&q>d;D(1=%8WIlL$p$ z*`)=a5eYWV^mdxFxQ*uKh08*>20dJ*#hKur7-GQ>oC}2*dFE=lb83*uKzg~hPiP73 z_!fbCf=r}?GC9^Qq~mSipscdBl!Ta;*dd*jX?5QyN%36>r5 z{)txp`1E4bB-uOhv#R;rI60Yf-n;Jb-JFxTk53yXI^52FrGAY=a5B_=y4){wJgg7R zcD*(%`OWlzWapm+@bd1Fqqsouw-mJFrcx=*N!zSAUfsvsGLW)#D}=Uj!%lQ-YmL@2 zR-WjhnHsfKyeg6akp8LIDzTNi>#dKRdBG)SrrO^~1d9pL<9`)?B`=rrNf4d@{~F_h$j#e+)10 z{D7hfuj1$4j%LwW(I7^iiwl^D$&CuF^hoh{sbH05fcdW@%e!p8Vz$$U;PEr(ooVVO zapn=08ShFku97i=q_8pSQm;b=;A&%v2Ei$G_+!Dn90m__chRNbn;ea#J-1orJUJBi z@TP)H$m!C_*DoZ%ez3$3P=6%Cago5a5B*0iMnn?JjT8@ z+)0cJer~ZY|6NgBbuCQIC!L?kt@oo?x@GP91o<#la$dZ4xX!zJ+RvJ%S9n3dFZ@$T zq9aqyT3CsWE{A>5C^9cI1WT~rq#xs#<+v9J!jFjx4Ie|KmbaO)UEX%ZUNLwBtvaIw zG4G(5?E8eKCUJMPOP-P*swmQZZdj#M$m;Yh2M01jQNF4o^{WRL5m!`wBN$cPBU`xj zToCTn*rWW-7`DJ?~7CAW^r^xN2?$^r^FI{I}^ItF)ZC`YE1>3!3k*eTe2F z`1=ed2kH)C3FNn<_ND(_dEqj>cz=ePsOs$pt*@@9pSOeam>45G00Q700^2Qz)IWR8@xY4PwlG*qExl?N+~$ zbJh#O4GBLTZ`F?SlbJH`)nXEz{)3jwH08p@5)yTvbdXm~RFfFjxv*%jmACivok+t{ z-4qYxY(paz)HBq#cLb!7g`OS+XR(!@^m&gY<LE0r|NF-pJENG z`ya(WmgLl@&aZ;~ynDN~J^D%;q5kE6I%0`sQ@tk9K|u_@h%Xspfr-P2^wE&7YCh(> z{)iagY)2-KWNww#Xl34`7dAnWscv6ZKkS{Pd)HG&wmLg`+t<0_sUrPd8+1D6WR)K7%Ae{VzRtUx9X02%nMv=kf0t3X6AaT^3VSU& zJmq?>E8zW9gbhMeKtL6(NMlNk`fh5qYp3{9j`3>-g;xyd(Z6fR0?w1=eV%U9blcE5 zGQ^u*P>g~qHakVpY>C$1?4Rf>KF8n(C>fOCBvjEYEH1lEzD`Km4qZ{w)qjSa}EJdeWy_H$KveZI}YZ0q}#$SN8MT*IDAu zkp@PWX8mLBrvPRA9Y>Pow*IgMx`oGbRmXT3Nb^1Oa7<4cb&55Tz_>Dv`dh}JpQ^1@kV{SlWs$7incL)K1K#dC~ zzR)_;OV-1u%Dmq`7weC0QQK@DGDKsiWBzRp(E8gPaGk{EEB6U%S7q3`KjFZk(URZp zpQF#L#wosDoEIHOJ!}+Dr96Q z;5x$Rr}mGUvo@z!!{y(>!oE*}N4~0l=R`^K8-v^9AFzIJ>wC)%vJ>N2BgCS2gdN!I zB~Nq#DvpDvDMn+=6+gPruxt7Ihwyqt!|uDx18x*xmtO$ChuYZpzS=n+ zFyi#Tl5g+oo)(ZD!7S!K?gK!j4-r;M`H=C*mOYPp$K$foDkqm__vEfk&z9Dj{ z3yY}Cf4@@9vOsl!=nLtYDpbz;&9yhBr2D4@$&ZHW4vZ48@M{?&Z<+P*{k4RemaH`V z9K&M@9gZ2^zYLgDp0T*=ORRri$>wxb{R%unCLIs(D51+nqhp(e|h)l=F!vgc^D)+fy0zRFqfh8D$^(q`zq3ZMW>O zX(%I`amDy^G0f){$vG9P(Jb~a_9or|B z6Y-kfeL|*dUN1c7uuv?>)5UwZa3arJ7WT7=SflD=a@o`GRGDH8iP326E2B)#1wkz> z;~Q2z)i>1v@2!8#>U82jMWMNF_%#?q4LBX5+65O7^36i%PdN?H z4A7;xXICV!#E7e-5V&uwjm(7sAz1UX%^m89^|g2ve(l^! z^FGPkiguZ`O$Z(Z&on-%=7FCiSC>c6!%Z+($&gcf`%D1~$Tq3jW$lyi?%&uN8& zEH@~1Ev)^-`AxbKTDFi<{CI%3xpG5`P_A&7*?%vwj#E=>G3gs}dIQ{Cr9jlLL^Ds= zUt&|Y!2cxHZ!|!LVi+h+&S=^k^G(QrU_{iLCr(|a6Q>E*V^;@P`!K7@3{oE}%heXU z6Y+T-J1_fWd9!XDtR*TS>n6>@>}3mIvP?C6Oe_GtQ)aY-gnqn`gR)F6Jpk(L<2*qg2P z+3e?K8|0xUI6CK>Mj7FHD6`*)rXKGAK>^ObiieW4CI1Z$#95c^4Q|QM&-E$9rx6_o ztaav}C{zFK0)TAJL1)=>_ z^0P6!t*cz0Ni_M$D)Z1h=RKo%Gi&O^Sz}*i*oQknH40PJk?+)H;MhcqmMaJ=d}Dcc zzW>%x*}a!7>J;kIBBZQ*BQq6J1HIGsJ2%TEC78bh)d1EcH%WSh&=^X0WeYYKmC9?G zF|n%Fi;DvxwX&+CXJ0L-Q9_Npr$uAcliU01K2k@A479H-@|jW^SYl-8)*z?*mX?^- z2_VH*JeSG_eTmA6Frf{eANm%r|$4a@XHBPn37da4^slCQF(D%x%wFS zCkif8X?T!VwA0$}&}-puwhX{GQ~aK*uk(_kq4Dy}7*;p@2d^W=gI-9>dRTU=)c3zP zi)=$^ed|S+fN6LVI6mAZEMOv2&U8p%#WW$#hy=>(BB3Sz&m$@G;rKmmA9&juHHT(( zfU)LuK1?_t1bex#q#c*;)MJ^f9bXd9pU~zPsOWD8Jp}UF`tWNEeURi`#X0fm=xMla zIDL~sA;Z_G7!GsCUVKMq_j|Nr0efV{rQ4S<@Kza)c_G-vW8|0Iur)KBenDZBo{r?f zh${b>pO=e`zaROk<%Skd-H(dzFM1(Aj(TI_xt+|80%M7&%M2>nzb{QR1(I&zbK<#8 zE$A7F-rzD7;eGFD#HT8CylRs?ttwMqy>|F)T(oGO)!(wznBj zsQV2=y64}t9`OzS1A3YA?g0vnP$#;-pDm$m+WHtZHOAv5e#hJuD&y&xdls}+#O&Nd z2~BKQ$^UzpDu78P6W?O~Zw_6WoNc-2LQ*whNm`*75>7&+<0BV*aS`F}av4QCq)-P8 zICOGqHNUC);l^1Wg>|=chqMun-Yx^@(ZKv^L1LqN!8AS4&qn%Q@q>krjS!_)yMcH}KNR)f}w*_VODx{oWSAO^%?JO(` zZd@hqG{x)nCD%AZS%=eMe?vk~|D5c`$tjtapBw^qWI^d|NlqyDOFs0+{j$A9&fw}F z^aPqIqh!JJYqJK1Z-ci~1&SLSRxTIr((R!}` z#W@dfyD{#kUyA1rzRSW%M2TyE99$c)$Oc9)R6w2uC<^y&d_LPgwOClInUJgS@WHp? z8{&2Ofx`iE;YU?B#riy5oPTsZaZc-}s}#$3)HB5%`eiAQrowrLPo!ZY4E?mHyZ zDITeS;-fdMJ%Bk|=_igKtvRO}DW(iAI3^Q3YMYhP*RxW||AHuYtYe#~!hM`f!a_;3 z7t9*AcI|Y8eIiVA^1b!YvZmO$^>vh=72bJO%coxh0=iRDaaf-?U$*2uxvjF`{0hGz z_63D|(XK=Ta7;%`8qWx17HYba`NHZO)Thb$oxg3;?WTXDo|p?HU)9(Bb(OYYdE3_x z0jTdF%VQv|0nirt!HP8~Q%udLvr2N>RW&eeK%224fK;D8Y|I@{vB>BOF|VXK=peT* z)Aa19jN;U;6w>}wDcydO>M;~&e5TGi+*R)%w>^H!x9s}g#L7U)ZREc?09oO-s?R;B z=y)aEZEul@jaz(G%r(o{S&7m2>iYtnK{8Q>jKvtsr-R@9S(X+j#r+JzGMgHS6{=dW z-Fa(O&DVaLSC1(!_dd@wgz1#Kp3BChE;k*Cti=4J_$ALVt<`5{w?5%9hqr!9s*N>Hl8+XnQ6}mFdDco_V zkBq~LwMbL{rZxi8y;M|p z;nmy8!wA(sFSpIOV}s~zX&h%h6pa^7@}FBZP|?^ZQu1f3#Has3UqHu+1&@yfBM}zk zC++(SEuu;lh$VD7Q%0iwM8$)+RAg`IZsA^$M%L}Yp1-ynYLphGW2_@ZWeziF!fL3! zta}<`QTWYl$_6pHzb5?F=kXE` z`9eIPd6(D_HLp@1L&}|=cBxUZFj!eR43kUgorWEnliv26b6T!eyx48_!_VhaIdoI)|4m_=LfV4zxe?s+tvO$JE z;YcdV^xR2=gJ&|Ay>R6`6m4f_Geg@rAyM?88gR`gqW14wa>{|R^HkeQ_5oLD?Tk5G zTZ5%Ox=$g@3hC>>QSNo9Y=n#Y%+td4Rxez}tziLKXBz#r$(A(lrav!7yGC!vS}bas z>A`2tZom2Cvb=9!@3#UPY73VKhcqB|6lTc$w&KJuI)6G@NEh{4VkWw9{2lto^$g?BSIIBzqs=5A$=X`V`17tioK6#7d1^qMT^=Ia!WxWT_P+ocZzJhf$zQ$rsAi3f8BBSZLMX zE9^4?)%#YeV}8jfiFSA~{iu^ua#$*wh<^pJ@&%f4!%m2*W{TArQhxL5@e^0$zmB3t zG!ksJLPj`?1aIIc>#j65p8Yr(`|ERKDQQG8E^ol@FgUjwj5Seg&uI1J18Wsh{gIvh4Czy$*k+3 z2@58}n&_lVD}jKJQ7>e!v=m=WCZx~c+`lnfPfcSt{`5DA2PJV2d* z9Zu{tro7SBN+`zPL3RI0arMj7vn(vKF%c$#cg~28&x0|4Jtbo<170MMHV}Prp*b*4 zzbp9b_iB0rdKRtrJN@cJS#m>24~#D$)vM?$i3`RxK`@xSh*KZXh6lS z0EB)_7o-LcHysFD$D2u*HpmXX{-2rbO6#$k3CU|;Cn8~a(XQ3aNu9c`Eoq&(Aa>sw zE=IjA|DHyF>P9$i05B-Q8vA0paw5uYMDzF_vw*>`+>VZ1ozOdE4&^z5nbDdY?L-ue z+aY1zRVVWTp17-ECEx%u_r+(r1JJiMbFCzjN}}(_*FGyU_Nz?cS=PMmQ<)?F2{zC@(D{8j`IgZgP?a zrPUHJnX@ZKo6+@jzGR+5cX9Gu64LNiYvfK8h(^yY{p^77KJ<+7kinHe?}As7#`bq# z#+;y&$q^ua0#;L^^*y}jBFlBhJ%6KhCckQhAW>Oq&IsObwxLuBD5w}R)H%I;1wq=w zB4Gr3Bwd(h_-wgP0moQYY5hftL6z!7;4nFvQ*OBFSoqsal-iE&}f&T z{SiIaJfz8^$0xl*N>cCn2_;A#iOH#aOu-#yJip#xl+k$(%u-@y=*4Y9d3vherG8xC z){cGi+cuPuM*z|`QVM+oo=twNOqG?glRyL?%))Z{kEcEeWPGqE-r=h|pY)MYV~-?iVn#p+{`*y6DhjH|Z{Anuy(RN5Z=z<5LznRM~& z{h8HK()`>AoAq!0Wm`oD!T8l?U&x_o;tGq?atW#Ta;(?ryl#-lE=lP%+MuP_q0tS7~*(0u5)aa*)O)x+$cij63;245UMHH&Vl zU6y!5FKBb41N`Wi2`k|6ORrx-Tm9@bbmdq1 z+!t?Bx3ru6e-mY$HMa-Y^N=cuEA%+EO))V8)y91+%bW{6|nKdPqpXhKEV2Y7_40TQ(k(Qs65jl7X=Uh{PMdP=9`a|d)l5L_1q*JB! z`d`u)8khuyFlv`yM9+c%YnEZr|6@|1g7`iuW|Et;tB?HyE^0A;BwXux_!Dy_1QlBP&?#~xBxt;? zj%cV37o+v1ZnJ&a2al`h*zu}c_y^R%%L#28dc0s&SP|nD3>95q7|`(9&F!!O{89Gn z$Edakr#ODs+2L|KP78=H@k;)>Qqj4f{n{7Nw?-wz! zym=CNk}8UBSk(sB@Mp0P6@l4QGPPfiBtrA7qbaSpv9;{+A8R9wqZj^rO zsFwu5R=M}tBRkF6!{wRWzb8|`(#qlg{@2O!Fae)H;E|ZG$Q7-UV)D_+CDcf>wU;pJlXl0Y?30!2cjzxBO)v!B`frA}N22S!4$w z6EHupT-nC{j=j^+#wcnC#O)ZHMX|KmlzBb$)E2evcj-Mp@kKqSV>@?m`NODBA_BYh zSPX)3xnS_WaIUr2kzIOPE+q@2{=pcwSJII&n1MCMe=LP3UE^*}+=MCm{KjWycFUiOE(9sS~UMBCr-K2{i)lv(wS2ox$asZT!u& zVt#R?^jF1x?N;meMYi)tjzB%dHTzrUF=BqMo01=ou_zvZCJ^J7#5wz$|Eh-4#WHFq zn*N+kiwmf?mH~9Q&D*|zbwbln=9k{bd(K&jJjU9ElDL$da(ZCf>{Du;<*cmaP8R}! zTSwDGpN z`~?ER1&;=|e3Zl~Kt*@@Ti_3Cc&_?tk| zm&lv^=W$F7J*$qP-lB6vIt1ly>}U%PjR%VpaPkHd#|bV3o+T^NlCr4$$+EA6%m2^1 z65lw-`v21~84!>R>gZo<_B6VzjYz!k{tzvO(LMNJZ0;Pe_jTm28klyzx+*I>Edu8^ zGpcnsB?(0KG~Q&>sWga_8(c@j(Y(rTmvWaPn;CVP)=ly{r%_&ma4E9_g0X)#4>*kr zeRCvpj6wvLbqjr{0aHCmsXz`-D#A1aN0W0o*)PicZ6M#o0aE;97#QpPy89L?N9NCBx@OFmagkA!MP^SmHO>u!_8B@$WP-G7+qiAt1Vbxf0R{28QP*! z_&1*Rgz{v`06@(rT+~qM6dfO_@0I@myds)G$U@2fQthmRJu&bianLfU@f@81!IqF% z>w;k|HPo8#M6-te_T1H&qSMW?(nrP>*)cRi(mOLH>rFM51ZTHxl1sxtpU4@!_xNh{ z2cMdE;08-p3oOFSHq@q$DjJjRJ-<4Mjr=b4b)Qm`8v@)TcVsT=CpvI!%*oZfzkAV| z*KyCj#q#~!$@AWPagj^f#eSLAMMBguFe$RVZ@2gO9vW;7(0JBM*qM``ei`Rg&e#Q- z(4QaC2|WrS2o}1zQ1o4+ZyCm0ZA@W1#rr9X?wofvc@zEuFZoj;Iqxx^H9VWurYN^< zTL{O0P@1dG8ea*ZOHp?r>el8&TB3uiw{A8m=+z{oTp zz$Auop$cQs8d43av&pnAO+P-SU>@cirN%-ytYhrUO+Z938zWq)pSND&3b9S0l8UZgG&FqI5^TBb=`rtoV$B+i?%Pw#e8OY3+3LVhCaM)l{ zeruJuH-b9x%fY5>L%hZlsn8f7#lFZ)8@M*=?Zegd2+4UsRNk`wzWQ9DucvgkFSD6V zRQcvLFhVFlRYSKA5O<;~GJN(2gZx_gPG%zM1q=Ye*q&)y(&~pLhxS=q;@^*XSZm8) zoguW_`+vBsz(cAJp;;;6PKi#MKGqM6+@8JAY3rixArY91=@UpuQ+6YAchdRzQ{ufBLpIZD7ngXc+u^n;!D_fDH`eYOSHY1I z&zFl%vc!)#wxvF&kZ+te3*>i+tn=TYmBo>%+_j%b*iL=mt@;Y0PS7A&IiUK+(yJ3r zq2Bvuv{>6YHFvoD@+IlBB`*b*<~%I{nNz4z%FP=H-;{Gc^X#7l`=9lTO>Roa5M*vW z4kovP-+a5yUPg6KO3ATZo~fxSo|J+A4MkACIxgzOq$^BROmjp0+-K?rIQFUaz&BHZ z^^1eCU%@*@;Z7(0%Soc+8obY=i7EO_YWq8`B=SS(-U$cq1AlDle-|qgEH?lMw|3YR zaoix$@S6Yix$*M?CUMqJgmrfRQ1a8&g1dT3#zF`)b&n|GUkxR3vnCt4#3H8)wWd0v zovG|^+`>%Ju6PAl^+#t0$uvINkYs%y@$i>S2(3xS@*;~lQ*t7KLaV#yaH#P_BdfF{VJh76Hn!uyOYms5pUUuSzQV z(My>k3*X?P@NI|tZ&(pbla^$`b*hdge;Ux4jq-zWy#*fI;qAb`OHsO8tk&T5V4VMqFv=R_f!Ka3VJXfO%erP}0#dbDR4-HQ17`<#nfq@XXln86MQ!6Je z(;uw$|G!v0cf?;e7z6fgX`-6d1T~w;bPL9=`FcrNEnc^6A_U61M$2Y2w{bXmAV0jJ zl;03%Z3N2XYu^=2I<{$Bjlu!UCo^d;uq znczuQ8Dp>;>JGk7dgFl(IvWgfOs2Opv}X!!2P`z+tkwspKb!gxEh$`~m+*iDr15Vw zh3bT4K&?L0*xR?pVZK&NS#!u&@niQhF#Ev zo1q(gZ+7?D-S2*WyZh`v?>~G7$9tHo&g-n>I5M<87i%erjDW6Q%Y$(Ho~owKlt3&< z780)b4K63+p-^F>;xjIvD=xMK2E@y==_TcClQdummaFNd?x>OO6y9gQKje<$~0?oe5|RvEgJH* zuAUF!d`jDmsPlamwUBzXGYq0XFD|`Yde*a_18hU0Jt!t9Si1T5zPg#ZkN0PFZt(mn z+KspuyArS})Rs%vb=s)#n>;G1KlLk5SAq5!5zDrG)o@(;Xo&qYUTP@^8+mw{QRnX} zOYn^s)siBF=g6GXqs&6THP}A})#A}P<>T&dLcEmDNxubmc9?ei8SvvlAE(?E{(=La;7{bI#~kW5alVYWNqB-DyWqm_q=PZgpb3R z##DAYUGJBO@#Uy(0m;wVbpG&Pdx?!I-yj3JQ0mX%*I7EOV}>Hx#bGZZ?-nz8y{^>g zoGl(IEGww%M84=VH)i1fjWA%}htwBV7QIfDmK%TD?V&1fEcRV$S-IDG_YuK~_tSKP z(Rt4Lg#kJBUEBQ^<)sUhFdRA6wkIYe{D&G|*GTjTn#B8X-00;BKVE-LAyS5dWQYDy z0F^g53}i#%KzFP*jdpeV3an9`@NkcpriHr6lQl4O73`hu73J6U_8B~ z5~5Ql`ij%CEXunGsBshC9Jl*N|MX3Mci{4bfOFd}OP-`cjwG>5(V0dwnw@P`w~>N# z;&9y`zTq}=W_0$rN%;q7Wp&)sY03ry{Q~N_0KfjEs~-XP2u6eE=b_Jl{-80Fwk^vfcJU*U72;8GSIW(mzo!uz5pb{tKEjhSjE#u zBi|C*yc-87-Ev!buF``3@qItFrvP7gMRU!o>97H^$u#|G*y^lJGgxv3JF5T4v)=iBuY9E@UZ?tq;}qSFJdKi7w2p zksyi44|`vK!e+|;;jsKcxb+Fv7q~PNZL#xrqDj39Gd70Z?Ov^Pl8OKNMb>_6XS1QN zlD`Jn6vM5_(E-pPBre12Uj6c+QFer0L0BS@o^f+9A?B$!a;Y(DV1W8W%NFB$cU#9` z=D@XDcXj_33?&9RNj%;cNYW&DUpHSQit^5a(9#_w{`k%|J6M@(pg>pj%`2$)!DwOv zL~ZveOq1J9>%9MW%=k8BSnV_OjZ^%t-LG-Ff8E9Z_L+~D z*ud=Eo%yi*og(eOzWC$^Y((+)9Myka>wo*C?*SDs{LAg<`u~xz_W$hQXt?LJt#LkL ze;|JU#FO)~!0-=7VSE1b%>MJcWJrKjqIw?n2dl$BzxQumS7!jjpU*%DfBxrZ84K7k zGB6}9e_p!5fFBrs4L>>o?4O_IcTZ;%j@RvP9dUmg!GF6f&$0TvfZ-c^lQ6^opSk~! zb1jYm>=-1M^S__4|M|KY?KQ24?Wq<^oTCBEE#e29q;#54XV z8b?~d$wT!grthya%hitbnk^Ix%S%e&`R5nw^+z%dRgZ0-^Y_Sqe(n*tfL(!pjAC&9 zXXQD5$pJ_2DoYy%P|f)d8{wbd0>&eE;^|{d_UDxaf_a|4*9F#}C;V@J{^W?j^1>4S z={e(?LNX{>bKf83C*8YG{`13mhlIm*F%_Se5(CFBKI#+M_{%7nQQd_?dN<%dvm)dD(`GM+LWqlnDE8~P%dQ_Yn3SgRw{!Sw_n%0y+l6tPKbHyu z&ZJb4o_r)dhjEgu^OZ(-o%1e>>xAteIfeHl@JAjL|Lq8UyoYSf8Re7`7F3!|tog!?u2)@p^`OAfAQ>An;;U z`=V{xMj?q8ce**ybB?{CJLs%o*d02(AU^zEb$~^T^pH#E@kqiIKwb66ZMzWO?Q|*R z_7O&E&86sgFEfTEfpgMS+e*JbMObPT;>~wJn6xocpZ0`phzD{@gUYHZvOGh=FrR`t z&;iWZU}Bk#w#L!HghH_r(zf$AqK)+@?Rc)1tQmH@jTkYWwwhB)LcFTAMr+6iSJ%fk zhouU6_IaOHZ8ja}PjQRG@81ER{B5fBgvXJOcc$dE`#Qe8kQv0KWn%z$r2_k_%}wXou;RF>+tN25?J z%))*2L%$_8Z$mxXj)sECn2g)+*}})`%SP0J)es4ljZyC=9QtUThPovyuouKrmQx*k zPTW}5%`j0UM8rhpkx1sN*{EkZZL@{38=!VkMLc!vCQYfD5w7hV05m(AFd47dm>Kn zpzc#@Q>J!@b+ZHGiD;Db&SPzeBlnEXpOv%!7P!^4kPL)cqkPh`8Bh4!;Rt}sSk1TI z5lR1gSeKQG>p+i~vdZUW5iQnnvdz3xOFYw2e|HF4u1I6hXp%$5;aWbKXc}HVy(N-7 zhXN&%r>Y{ilb-dA$8pt^^_&za<7`N&D#2B+>(kUS07^1w&kU}+l1LbdRaaWXvPH)#o+{gq#F zM=ck<5+v}ov(ivW7j*WfTgVgfE%_f$Ximo}$8Sk2I=I5;+q~P9%0CMM3YA3&IT--` zS$M-oPE{=pRSt`#cc^=_iCIFoJXVUziIRzB^k1EmMh^2Hv;c;snl(z?(|2@hgj>Ye zRD3i-q+Y_D+AT{u>N!o}K;^i~uWVoPHF7m*LN@UGfJL=!&O8zDL406lSEt zp3NkQ->u7&tSof$jX3Yu9BkS#3KiRmJOCO#-;?ful0n9nPp{YusZ@0w8=*3w49`{- zv*}j#;ET5353(O_bdxtmJe73kYEfj#lpzbp)NRqPa7i1eHC{HL&gak5=V5x)7$rS)+JW3JlX-(tRwoPx=})4kp2(N}M}LPuwB zwvA>?x|p8-S~sE;;8^CcTzCfvl$Py5wi1E1eOjO@lJJV%tk1OIh3u^~fWsN7HC5aM zw7sVV04RbqfVxZt_%v#aqrwSTG+VFNUQ(6jMDj>b%6JHl6%>dw_zL#oCkeWS7I2IS zQL_Q5$oTo->NO(7=BooHS^oE`=vJ*Y1_JS&fSdw6@3<*wGah`p0t$;wN5ahwaDD7n z-FPf-xyalLFT##*=Lb=nr915^Ik>fsxm{1^S8V174I6HkFHe;8K1`IFz8&P)w~Eu# zMJFKdnRD_w*=GvU9=@L#e3_@eT`nI2Tm;C$E0~!JhU_o=ha+B1C6N2OiGhx90>#M* z*aCMKOG?<9*@XyK*LkP(`SXm52kdxoA=4FL0jFZ|XAIM{KhX}4`r=oV`r@Q~jZXae zALr=M9iVRuRhgei8bk5Em3buR%IC45x*AlPTt%_{J zT@AizDqZ*~WpW#IGKX3pAjWQAuctZuE^FnUTkW!}1PufAGd|}x?I_no`y4T&ZE94U zGsQGHVrWk8+t7E2xqBKZCOBTd*U>x7gbxn%P@KN-Qu7j9HH2R>FZ$Wq2q|ZBR(LKw8{oaqhvglba!NH{LX#CAti}Rk) zedWG#FdA{5$IYo;Hqj(1&c0+4*{(G4FN8rUK>*%>ZNR4eS_k5ON-;ZU!lbJ#!bPH*;UHlo)X7>c_A z`54SCA+3`Zvy0#-oa^_khbyOe0G^8%M-RVYrl}>?+=Z9KM+#=j&rn1ALZeszyI#nS zjvVoY>>EnLJnJ`5`@4=%V-yZuj5p0vtSAT1bDsJxEJIG4wP|die80O7lLInd2z(~7s$NRHoG1a*j_xg>c`NJR!JN2_2zOM^AF#IL{1{o zxR%5UKhEUvb6sa)?p^y;u(2MnP1cHck$KcgVn6p1nuJU2_-k4%H?W-(G&4A}On4Y< zpgoo=z8(NBki6kDoBE^<@TW&&_cRYC@r%*@Xt0JbW_X;2EZ}+gzf!%^xC(4?hL+>> zoG|s+QxQl&R=a}td(%M9i315aR2TX~4`mIBw8VnihN|3VtT~|4%C=_5ecBXG>zOc? zo=9RUvDdc3Y89rooSwIF)w&%p#Fy()vQu#{Cr-01P-@$4jm1~7t1n$fEu^d@bff`L zoDIodsI_jX_B#Ch{cywW&{iPRXo$EIPzN%cdb?iNoRseUc_1y#7G6v$>N#4hawi;I zSy`<7Z4dI@$12$%Up`$2t>Kc5gclX!Phe<6>y}q%yt~lWJ8j06HMf@7MgVtTcvkVg z-aKm!WZ&<4bd$81e&=QqR;5pp_uB4h)wNcQf%(mMQD6fQ%Dy{;GPiFvn7dcgEj zd-$6kS|B3MTee_$-(52-nTiko4qvL4A9gT>JQ}KkHBA>EwOzfZUmbe&)FR0*rjw%B zZIAU`Y$zlvE|d!tRo<}xm&PJ6q3bvB*{3PbFVy(e+k@v?T}<@ky@xV zjkCFr=v51~US~|0H5#^>HZ4@EignnXphys)!kNj+y#BCuoRjAb3`b7 zLai}1Vp@LX%_}rH99!U&UiK7Qm%_PS;6l;y;?wr0_p+?x1s}e?jf#Gcj5}qFcb7(l znHz9(v#oPpzt?A%>VDAh08rkPs?g1G7XNt2sgGpPX)TRcNpl6i|Fj$CVzqqR^4ZWG zti;>q$EXg_3}~%rp4NM511{>`(4d(-1zL#~UQ4T5o^=?G(ss=H!4UvAjt_!-W1Vk>d zY5YagQ*8uExS>n-7~xi$Uf1IVw4^fWY=5yO)P4UW52VP_Thrv0T>^4GQ>(Otx7cI#??Y}2yI{AyZe{1tG2d@ zT1U8ldre89v<$Fi2u~$KtoSdX!UIh4%3D*$wY|n|gAtdlI*COO#X>GobM*4a>CXPG`_?kNXqj z>XQ;_H*_Ns8S|MxQ=jvU@ z`m@#^4Yu?MC@S^jM{{dLr-(PNdVYaUJAyL`?OQoYv^1Os0g=9OWpu}7INkfV*6Hm>8wc=`WTV;%Ow%RYFXcp4 zt8@LH|7M5_xh*lD!JbM8E778!EHIJ^7dh6pLl}M1Qq!t`R61ODS@_ZKfj$<*cIYs* z&575R*A{`Vqp9o}mwm_Ax1VRtGF&F7ure^_9DfuN%NaBYO2D!xzD7KMuogkkSpSyOZak>c;FgH-POk$ zc4O9PI!gHN^7XVagPRd11H}9Kgo=SPrgbk#>$NOv_lNaG^X>-XIqbNCK>)b!nEYTf zt>M-{v$(RxW=yu!}GsuzK;B#x_=Vmi(Z2g4%)q@Kr@r7JIzA1vx*K}*Mr>Dj{8O?6tS7)i!PCH^jz9l3!n z-EPjF^`$Ltg;=Az zYHkOGZK<;xo?8z+WA6QMYcnMVKl%Xwxa2g~gc}NFH0!RI_d>JMWj^LO$nuIjWn4Es z|7}1Z92|d>#HV}Slf9j%K3M;m#sTP&_Iv)0-ya=9kp$2ja5Uaf^2zb)j&&_GW~>oO zV3#TYyO&PY=P$QD05GcEXNxW)?nj1vUrhlXZ;9#;EhNXUqmc)! zK`3o^K&OHd)I*$8bH)2A;^PbUk;@)J2mxCc$M$7z^x@d?m2w~Vt_Us;TEzJx@dMg|Ibb)-_MOF6~=vJ@U7TJ7`6Xsmks0DBV(N0<@ zfL64x1|eEB4Q$Oq8mJ^=l2%qqql4;0##LOpWX;`{hqZPuH($hQW==8pRD=__Yf{O2 zN{}cnQtH2nv8=q<;&X;j41ediTxNNLyNM5}+1eYr>oLoI24AFiOx(!=)AP+i*Zh8_ zffLY=`cOD<)a(}KvJP)oU|mt}M@+ZIenQ=fc_|(7eQ=MRbXOevY>PK}*U{wLr96bE z)XRxzV>I(Q--Yn|hP37*?MzCRB`+~CPZIDU-az18lS8zdqje=I=qd=GGU2JY+|>&5 zk;UF?bMX9a8*3=8AKx9y3^K&O@eJ zc2$T2W`)Dm49T^*t$J~$98tMpo>VUpNO*@ zx|7dvFM6BuozuLO6xGcnS4sg{pc(zy%kv2E2up+E2!2Lsv~mf5v=cE@5tgxei0&L0 zyRgz8Fm3}Pe7)$Pa6?J_J@JoLV%&{den0Yy4?`=T@}s*?vKT~!UW;3ygG>p*%HDla zIQ55qM~7Zgtf5eY&F5=e<|FpEDM7b-MDY!N1r_TIBA3ufz0>?4@4^%`w=Pl(Q* zRW(?Gigh8ZLxg!-Ny@S~97iMm~8FHvB6Zv26yp0N130s>1(jV*qm(#_-gZNY0$#}^EY7a+&RB|!S0{`H$v-8A*dum3c{ZtDIrO3CM83Yu9LfjkeBfs=K z_h$QW%lU+$ZK?$0m-Z2FmHl)KXtP#mtegwbsH=> z3%z9t3*iT<9$u^TW!QWnng5{|m&4rEIBL!4Cg4eby3nzIXFzCxJa*Ae!t7?bYX6x9 zWYuYqS35v42=SVBra)0?T0h|brv9F)U(9Tp4jxwzp!}HIBk~7Yym2YvF%F-<(HPfT zN;aoeBl89=PfAx5ivA!V5V>g|sFv2Tc8dQ3oVJ|NRyJtQ^oU~HbMwWPyA>l216?N5 zsWFYJkZJvdl9@5O=fB{sB0A{niT2oL=mdvrI*oBAO1PO_y9q4xc`nCSljEbj{8_3d zf{=*RA}mJkf4W!F;-D6mX|QO>!ASwC6v#n9n%jJ{qW<~{@>dxVk}N&HWunZT#X9o_ zrv%WIJget`OAGQ#^j0~#u3@|ro#N>jHSU*&6h@Stnx*jIp}l#vQ@Eoz;PL?YWPRkk z^zY32V%NIE1}YZg>6Mx#2{y&YxHrBACsiecMpx8iq-U$fyEdY_;r)D$y1V$<8H0MB zMNhADF2>IFA}q)6jP7mE_`m3g%x)tBoJS(c0L2Zge49mF@##y$omY9nJtMF#y)+WL z5$d^;kbBwIZw|Fq)0!*FyMY?iz2yEZG(##2vT5gP`OEwVF7C^{CD*N0R@i(yP)D@) zDi|CV9w|BWqI~VhqVe82xgE$?#>TRfbP9*pbB_3q&qIkzX@%y0UZItQb}SQ6x3O?{ zS)x>A`hIQ>ccW=#xiOuHC*(2N2ve7jn(mf6YM)~z!4}J>eZjf`1@d?~hcW!U#Yi~J zmQV{CAU#k)}z!*iJ%1?D(j>JQD!KB^Ic~8#xMBA9o_gH>Avq)J;|L#riVI{f$ zQ`4!l2Fs{q@v^m-Du`QmlOHAit4XkrWb_&+xgNKbn{#=o({wuv`~@Gj7FNnrfCxKH zgelM{`^I!uwZcgl?AXT~?r6DK9!9Sj&$4QD5!#(hX?J@gc6T?*e`RhS?_4wKmxgvQ z!Tfkq)M}|I^;M8)Rx3l{g0EeMG};wtELCdQb}CGGrgnqs&#MXW(}5o@nj~;i*NK)cDVlW8BnY0fK zqH2)D=vAuKLVcNw>#xbIR-^%*Q#T9tQI;VZOCKsE4!kzL(#KNvprS*^ke0$QU2F}T zL=$f8m3gC6{DEzNhjOOgWJOK(SjN2a=53P2=V4AtDGjuT zpHKQT4%#ZGh0D-r|Jcoe)DXFY*zMUXp4A;jyrN3e3R2}Yd>-WWGC+` zDm1+>A9|?5hu9A+8{?%k`T~0Y=xOvg8N933ht%hO{0W4_14lru%*84!eLxvP#Ay|u z&gZ`4AYi(bB_zfkq)zL=lchw=W|BEs50uR^7K+Ngex|r@)V>J!LB^wh_i%TmxDF$> zue>ywFRiH0(|kc|Fo#bq_p$69zjlN=F=dOJA+-g-uHkumF8?s;cwC&&|LY`ffgNvj z4BDQ{RVvX-YeZd@#Hdxg>~*v0Ml5^al)~@*`~gM6G&L=4bg#JFAC- z!(#4NE);;m#qUj*m~(DH(AYmrG&?il24|eVEtO%?{870lY3b0T4FD?G1i-R9yYHxE z)TAS{mUTn%naU?JB;+eoZ}kjP#ol8bgE(x(#RO>ocz#`=}ZRoPASeUy}OV@oVIY=cpJZWEWa-DsiT)ru((5V`rCu=mc{vrMzCmUtXnv1fFbyNLb5vt zC(D~0t7|NUI2|8jUg~XL(y?bCBOFuf#+Se<8|hCvY?YD3{K^5^0o=lKPA)OBGwI8e z2Hc%CokD~WUCdu;F$g2(QN2gtrQC_(ECWD9d4U1}6F~xO9lcz_@6^h~ii`AwVzmRR zKmLB{=i5JMZpKMqIg`b7fnQIn_^2m-L=mk^(P>km)gazk-BRo6??Q%7n#_h)mvYSK zcJPkfbb!7ee}#MJ>oas6@bRx{kp2vBo#P@JW>vn-t1_~s1w5Rvaeb{IRjm{4$$6i7GwH+0TZGqBFO}-aYmxdj_RaFx%91OYH$XrUR`{ z`|rOI-b7Nb6!lKURMh~psHVIBImv%C6aghKwT*STKEYyAP0)Zf$(XdmYT zME5Vu6N3b&-$G1!uIr)Znv;G__9Fa;n1+|Wr_B*wOj$XkC{XvMoO*IFqwtr?AgVne zI1$d|Kpk|sghwGEA6kMVut(A50-Uf5n+pl@DEf_vt2YDB9k|q*&1~l!krPOH;xB7y z^l>_~Qy#J{Ea!3qkIU3^ieD#rF2!uEW42mC1zb*FMb0J7wnUq7M{Z+;kAy@y;X)97 z(2!lU$v)5HK`&}ERE_!G#2EK|;Apt{LP00)fYNZcz~Gr{cd3QymMvj|d zh?T)M(bU~DM`wqlG%hcIM*@?8r{QuuP3t7W4~>f;F|(&DL0N}b8GiQl#rN6sXi(wV z>U{SlQ`?XJ4`Ak1A_(~|V~cRQ7B)O-TESRyruA>VQ<4r_i$$s($xYIM1v z+2WL`EaWsn_q{&K!tET!uV^XPGepcLKU%LCk=eKMa*r)r9c_+AvPOn@z}{aT&;-2% zQcD`~-%M}XSZ9`s8cTcpUT;COB^qT$fi5}`H}yBeLSq=*zHz)i$C{y1$p3E4nSRy6 z&EIf~b2OVD)ev_@)VZAFuTzcyO===ssTxeIh>6tU^73H1+8alzA%EW+_~K+HP%zo{ zkIpAH-`O?~fwcQOmnHfysP;fC@&!{50rk~4iwpr|NT0{)hPc^mc`mhn;oU}Gj8HJ| zXbG{6-JVQiK@hDY3F2NG0qN_kW^%KugL%aO_1vzeo3o6G^d1d^_bxbHHFIjYEUhg8 z)&Zo@v8)yY|H=czQ7aVkQTK~Inp*_kn*He#W0n4?@RKxtZ!gU2!vp~IjyY=P5r)t3 zS&H3iiEAq8Idn%RjYpyt@#P1XNk03mRJiM69kZBGf0zTNiLICQ!SdjwxR>6$8YFzv#N9|m%zetEDs*a)GQedPBVjt z{`wn{P->U6f>~lj%S9k3_ZCF~ulWDcVKUzKu=5&P-t5j%v9v*6cu<)$qXXJ%` zar})ck|RxG-vq!?`Z0Qpg!qY{o2F(vH*?5-N(q~PluA~(M%$5)1A6vDv=5F+0;yNe zN!Sa=4>zy6PLmBG=K&ad(||lj0CV@XbM6vAM&&fNuj&1oy+< zJbSP`RfdUhBQ)n$sI(a^#&v|WMAY+3FKp~|6zU>R>a24zz`8%=FdxcZY(M+q&1vgd z&tU)0A>jaMCR70B@R{#~DFzKWV^^wS(fhJR>m~2ui_0I6JW!b)sx+2e%dS<6HFBpuC}A7fbU%39taleP6~)T?ULTP5}MCY?W+ zr>~Pib$I!Y+l9VLGQHpSz0f9EuBXCCXMelqzG=!7rE=Z#em>`l3@I$?us4z<`aNG*q4#?*BN@7+#G8o8hmht* zt4d4creX`-rc;xEb9PzGQB0m3JK1mC){(oZ%hVAMUU5!OdEfE?B3UJR!_TwO{tNSE z)ylWO7=N^m#ehYp{e$3#97efn3uR2!R@gv`#wwHuh8vR_3pDOyU*zO z&-w4PLPx&r3~e1e^R0Dyn4RAp9q9r`Nk10lY5T29=N^R~lJ6u?YvIas?*%eWksVa))XjLmWx!Ue6{4#p@`$+{t42;@;@Bhz4gz`jfAleWG* zG?C*%)4$d(%LeV_ln0LOURGL##21~kLTxp-?o%&lJm@y39iI#P_{+7KE>~AZvTmg= z?3*~3%veNc)EG<25x}>9)5`y;fzRq{`v;hY?0B!LY%k;Y12Jc~T7b0+8UMWv>WlG3 z7l1kiMUR)dt6>R983^b0Tm@SmhmRyNRfLB`oD`%5oj)}+&bwkRUsai8O}0i$kpJk3w**HjCBX$vbKF!8&ZQF3FfkRV$L$qFpdO+d`x61$t6iY#%zX-D{3$j*Ku|$$daX zD2GA(#<2M<0<&mLtxCl2^mPIU`nr>eQ^=73X^WMI~Kg+fzFm(niclh(BVWy@HoVLglHac;i9k zhQ7viFoKVWYYwM%MKcF|=$n5%PwLl%GAUoC(!RnG%`)JwJamhlOtqfjdFRqTkRg}D0a8PoT> zSsLL+L*o%xKav1G&YIrn;OL|7DN7BF>vE;4QkSi{-D4@w#~b;;u`vu?T`oKqP>M4q zZzZPXi!F+;0}>CT0r^3y&Pv$Uk z;U+R0c9VJtfSSJFqQ$=GBJ}JY>9YKw5(VwY7BxFrdVuY+@pIIG%Ya~w?E7vIeXr*W;VGcNr#sONiV(Z5@AYwP-Og#m z2S-j;1s;4IDW+c9k7Mq z>_;e@W&t4E5oT=Y?^but)xMl2-U+AfFi@L=IuBI{Zw!+Wu?l)nDm*Q8(?OT^&9ay4 zA*wmE@q*3EcN5QcRBetcud(;h!-Y;gMuv!8%hd7$Na^EDtMXt~+xKkXH+$H7H>u!t(zIW*Z0WW~ZF%a~|s$`SL1 zK=p(?4!z)SkOE!AgAcEcd<+xCKo?k5t1XtnY@FH7@3MINY+Hx9U-P*M_Y4QM-yJrH zJF?WEt^8y+pHVU#bZdx7uKm{OdlbRB5-qoUtyL&T&QM+7S*ix^yaEcz1@I^YWUgew zZifqwQ?Tf%p;%Yem*sJvIve&y-Bzz4-WB=MDaAl{Ah9^B(*ajt^ik!-{_MVOf86Vi zgUxEzPJaH?yGdDoTx&Ojaj>hsa_O8w2U8cB1V<+3=V2VP|^mP#a4*^vLZaFDz z+KOx#>;4X1J(gJI=)0+Fm49_6KL=OcgeJ+*w!c9#5zYPir)En zzK1JXKHj!yubpiSyk~b^T%lyzt(unp){{x_?ZHL*5dQ6356LJ8hRZIEXE21k%a;BK zr~$8=pF?D?gXW%%KPPX-Y+4*7yWRx1MF%tlbMD25h-*6wR5AC;Z;!vi;(_dQjk&0o zD@dzHz)P_ibRL-RfmdS52E3I?%IQ`)Sw`LoVk`v7wIhn7xLXa5w1VV@7o(p>%jV$~U9R;|eyZDNphdJC zS@?)H0UOr-aJ>bg%KNQH*ZM1<_=ln`eQYI2up!^o#V6BJZ>te$9xB6~;z&JB%yb|@ z>6fDq`>JK=7@ZQmSu^k>pQy8_&o*K}a7-6`pKWrjm12cET!f=ZZSz^=;^vCOBHp0Q z68Z)srQYsxkF?_=Jo7Q5?b#7>RX<~$97^~ZL!IU1 zw#`;pyd1gj7BVeiEOE3stRLSHen+`jZMcM@P)O!VNpE;skBq8kuwIxreDVC9#@7!{ z5kx3AmMS21z!3_z(|1b7lGW!ixZgX7j|dGqJ12y*qSsVZXkCWY#@-e{N+< zjTv=feQ{)V$or(5g^zc^+>Li{`@$wk=X-at>1007#FvmU>!FVuD-G5eE|U89yA`ds z5ftjihqN zV(GgW4{WTIpZ&y+3M;tvyjm`3a@eKawOn01ber$}Qc%~Gyr?&prnh%^n3K#;Oaj^B z0{?_gQM#>rPj}T|?!5?mjh)y$?njAMJ)Qz0BF!tSF3|5%H1*BIB zj^e+O!oSrC+YU>h$5teowL!SQeWHY_r@ax5)U@R@T^qSC0aO!8qw8JzZV7 zsDjrrr{4xBNOod`>j3P7v^zUq{IWX|WJjt>g!D?A*y-jM0%@@4$iN452Ww|zOv+v7&UTIpP!D*$Qhcgn_p9nniWHdC7oS(v$;Y>43q!ZlTW~a}?+@VCzr~ z@I59?5vFhyRoGOx5e`Js&!2tz%)hz_0(RIEwuD$2rEy784?%2nP~M(mNgtNRwz5Z1kkUsV7GKaf{RRm(kOg z+xoNn7QPTK(#1^`v(luwZ35Y4TzTV#dMKkKEVDnCyK~~usTZ|MZKeaDctcP!4`Hc7 zh6uzXIK7S6L9|OvBGe%|qoMFvZwW-xAcRVU%K3VB9Jn*CKr=H(*e5{ny-?;&C)fQJ*cs9Mn?HZx?j20(=np)`xLXUi+<5 zIkrk5ERzK{c|TmtR`R`_;tDVc7m2u4j_%~F1xiLUCbFB`4$U3TdUxIE2m4jM-*=o6 zZMI?fSBYF@l37Ir-4_Y7S5QaoT)^HyPh0f>UaR{qOE>1;!-9E*@_MgqC1(T|l*)tY zbFEwyYAbLunDw&Dx3$*4>j0QEgK)>!>xJ^-k@WT7h{dl(^8rehDa(Rh1CL8zo?IS_ z>tnsWVNOVV<;gKzR8NQihr>tu$ZU6XAig?$tAC~J)&?ZyLB zM@S1dAhJzg$sax$drXW2C^bFRW#)lEEphD9!H<&_9ED&e z>nfsj3DS+Ebi<+>k(L&a?(UXGDGBKYX=&*N(n$BB7a=Tq(H&>{z3+F<-q-%l-hRK& zxh{Wr5sNvW`ONW*agTf4_v^0{BJSM&Yu>P-R7DoUYE8|jjJ|!|nD!Txm^exSwf|1VE4wpI1`92vD5u-U(cC46v@);a$CEf#UD_AbRKTgX$%wvtQb zasDEz?rAEQ=icpCo+0o9@UvoC1+D(@Ea4&4^K}olZoQa=)%tv|WYZSb%k2?>Fr~B>Mi@T{B+69D z(#e?WT7G3GHvaR+KeYh32lEB6ewi2|6&?BsE+YY}<1R(VCz9@JiXU9I zvc8%-O+CTu-3bFK>wHOkAy{+=tl0j}9F=POh_%DpH+#Vx>T-T8cXIwa1KGKV;pj6l zB&lUi+fI8C89&Xj#-hIN2ldI!E|Dc*QBxUC&*bJ#ga{b*)1XTub+Nk@Dy?c1@}J}$ z?!WB(n{#V`xP0SYdA)7w95?+6yx=So(;RJBW9SOo8?WH_dk@=35a z^XU`88kh}LQCFUH`R;^A%MA6s+eA1 z!>YOJGn_VU{WG4nhNb%^((MMbGb5i%?Ca|WsM!x13f$L0||tOiTl8O{+l} zK)$3Pn$WyouP2dJdv%(!i2QxAKvh5Vu)gVJ z9i6_iWa+&r8dT1K`1P=M$>HpPUR43N5U8>?`7qjdrRk}IsV9;I^AT6nQPg4Hc}lc+ z^#_=M_n~A+sRoz4*f9vdyhYsSnStM#Mm+O;*s`3LvEK>4@@os6g?wxT!fqnuFNZZA zF;o~~hjd{@MCiGFgInw)0xqT+hos!w*?VL9-wawf=Cl)Ld&@X_;fv^ep5dJeQ-kd9 z5Tp=N)GuXDAGy{9QhEQ#PYVFLM7SL{NqYA2ukl89bJf1kVzg?kP+99y=Z@e4X$Yys zr^9HI4)Et!hE7CC*f=OEg(S6u(| zUJb*neB*J*>mELHCkuv&W69d>D<#{0yJZPe48@p}TbC|&pXC9Q61%_po2|qvkjAIan z%n%`0;{o5@5%BYKjP2qyZPc*SZFm!c-C%b_E-@A9R|#f~b3Jdp)@4XtIHb<~Le2%8 z-TX?4-?wbPusN`P_v|RI$^ajPN(~;B3!Lo4LJ~?1GSjTzKad;U)CGvW=ep1t>*N!3 zrmjIn?`r(ekUS?5cf!ACXAEH5Cp+|uG!NQYR6wh36A6>NWOGQ5;X-(aKt{?*!G=vT z^>Ov{uqv?SWuUYVxk2Z}wl@VLt**VQyb>ce9lf)dbXu<(b3`~Ne=fn`;X7~n_LrQ{Ei&o-_7k@~i1KZ`)H@i@ zh(|XD8A=c-w8aL(I?uR8au?wnx@AHW$l{e| zpg`3cGU$WXY2RUs_0(tZV}eC7$9iEub$eL$jq?jt%YH}HJbwyHgWa)NtwwFHIr`2n z_wB@~mOc?yE>X8@QXb9WzO!vGsBEDYMg zDRDWUQ{7>AE?*fLGHq>Dc$-oXHy)IEd?VB zh;Wwt-EC%yOT)1CEqLX5zlSPBRog->JkNg-F>Afr%sA)yDv}&Sqyo<#vsc&JrcDPQ-11TzR;RQroIWJEHuWCsR?pU zhDtZk_VmMs^AnWF_QA+5s3Yfh2lS*1L1p_#(VXl7tDNx-T)s`rb@y{{ZuGei{g~q^ z(6gxX%gmdPr|1HcpnwI~^Zjs+9aGvtP4(=Z=IoO+AZ(nwRSCB#LvAP;Ru=^B8xkss zT8PL(uv)KP7YBd16LC+{12_z8LZhBL6(d3!?c?to^=0M630+-z#Vyp4`^VGn$Z*RJ z$)vpAd;G`(*~Bwx)_sQMniQkECIvlZ@ug9AecM+X6Qf`JAh7EN^s}D2RT#htGlDr> zhdNV!RcQ|x*7_A#QJYy8T!0~0RH$PU7Fq*m;sSx!jmfzO5rPG^Bl}u~vtC})p?lmJ z|I|mrzC7w~*tvtB)1r5c?t;>Wg3r&hcok)dypN>k+;(T|-UcE(jVJ2`fYb8p)9r6< z04021&h-nXkBH6$pu{UZoX)4Ud2_Z~1lqSiS(w2O&Fs9`t%O5N++q_?m4%K%nA8d^ zXY1?3&$h-*>Mp(W%fsnU@9!!#~v!gzWNJnA6>!>PAuzx2X;I+;e-g)}~g!Ybj;-mby>Pp>+|)#~~L;gHJq&Fm?6qO!wizZgK_ zfU4l`?gGk|`Runf^c59UYX6|jrQ{WBGu^oD+Waz*t@$tXeNVufkLdIed(Y@gkMvQ+ zc1t{}`jr8+MG|8iBO6(7A5UqIg~Dc8vtx2dQHpDOOdAdUK4sEn5&mS@LBBd)ms$9< z$+xajr-@-|a&8Y-ojWOqYT6UEfOeGxMu7O_*Sdv!a2${#Flc&u<~!vCWZQecyHM^? z?@)XMppA;2uD&Y{;|8&WG;_n)#Fy3PcBWq@U-2O0BNK&>Iax~0-Y@Egve(;(eJ*E* z@vi`dPxHq1yD(IltXo>zd4yJOoKj-)R!hrsVvi}IcFQpd7y{Qpb0%N?=vQzB)z(v- z9a69KFyeRZcIld-(NAsQEn3+fs78)PN~@q7@Y9#wO|>RbVBwJJ`zHzARg{vMCKTd} z!Hg`vB+5+O874#eo;E&2#NiZE_4`oz|N{ks&e$`kfo4n?a~lA8}7EL!)IMFHB?lg}K9j0`3? zicicQb6%1&A>uuG;yL0k&q!c{n@j)u!JDfA`yy(t$2F$w)Pc%dlchCo;_f$HyNpap!Zh%LG)Ggbz*# zI7e!6(wuz1DkN)06Z0fVhK6$^8fgQdsm94>?Cwaamck85h`Ogl`YK{(X9L)--gZs{ z(Abgg<~7j)kU&6QVuq?Oq0^aw^&MdPW}v7aaEC!10$;p}*R8XRBgLBeZaFToISjO# zNV0YTS>~jI{RI!~$Ar-K!`3Vwo1_pmzN^_zzN8M&bq|5z=jl*10?nStN6j&npokUCI06)f^F@!YU|7L1Gmo&|m46ZyOjuV#ZOA7sBK@*_ltAHEkbO z-H$zwj`70MmhWkdd}MF?Ihju>&o1vkUQP&n+?#YrUOk>zPOrhXG_-EUJUO>yzauo% zO7Rji2Jmh1XV%BJ}GAQMA8h-NRZW8|@;GNVnYyz?p1rO|!QQ_Y6Lq0e0dz z8`x!k_ysWwb}vciraCH~<#%vPdnb#qui%kav zqc$`jWEaM4jz4kpV5kswk!J;BCt?#5n=}ddTu`yy=!z49uQm&^0?+;NnZ0F%5~`Vj zV>bppj{YK8w<;x-`OP7HwtpdeA_6fl5W^Obxp+#oGzr6Jts3vY?{MQdh?TqGwyj=0 z#=KE$-xle!~QkYA*5ZZiw!+16d)|WInQ`c#x(onS@dTm(#w6)vboq`zkv5q z9Rl9Bb{72>mCsrLXswvK=T$^;^wF%AB@RT9J&oI}kXYEgCM1HKqrlLzya$MH9UWE% zFry+68G!gU_ugUEhC8zN86jHS_WBS(daS}2lR!|2dsE1nQZ7I@JHMIOC*q{0m=FYa zRHKsxy|!5;9;bM(wA$hhE<~kj8JEXgZXNM?vJvo+Q=>qVM53H#s+ohEwF?(xbAxf; z!!X^K{mf?1?%ER+>Lht&KUZVqRl}iDEi6Kgje?Ztc;n83*Sck78PV$&5}u3=}>?XuQl(-r!Wp{HuOFj^<6B1D9JDjy|e{_8EItCW7RaMRDiZL-XGa7JY z9GXCT;q5-;zsAx>yY_loaAI{gDcvFdqKu_2ZxVgyId*@v0Nmon-vc*h*u8#b=a;}m zNoX^=;!Y_1g$6Q$+P?R2L4Jt3*y~^3*JJ_Iqn`yF+8HI1)`^W%Xr^k-l}IYt6Qk2v z^r~gf_W9l(yzpGflf*!DG4Y2PO}w`@L|RzCaRp-qJ#TYjJ;%nqv>TF=>~OFEf{Xpv zFz=57vOdH%OD`7kFh%_~)t#~ImaE>ZUU{UgMl`eRG>)hFvGx4LD&ikjK}c~}uFKt- zZ_?LZLM4ky6CP)sQYG02f}Tv9dmj}{IDN|l`xtBO9eA0eAIi&xh?mslNAO-iP}g{G z59&(xRrd>OX#n&hI2DC!y`rjNAGq4mzcX}%!)DGZy>6N~z#U>Lr{o$q#@ z!0~>DtIX@GFN|~;h5l|-AhN$}IJrlNLC794zIjA_B-^&ZTl(PZE^7aU!QhnNB4bl~ zbG1EIBW&%)J>+VNY1}9_T^}s(mwn)&yupXj)$KZ%r>{XAVLo|I%!&elcXW&qjyO0%LHa_HMm2Is&sj|7iQrxln znTPJoaLFGs>+etzy{b{^xZSxtTIb~5nM6z8`PB9JC{OS7^vzR_+4>x9W}RjUH~1G< z(v5*6wao7gJiWSIcem#tq^VBnQ6}!o&MNL>d`V1T4Wz9VjWN;#mm@m%TNg3HtFW~oson5OZ^K+ahyaMR!NYa zDEoT;>N+INTbB&Hl{aZIrGD8fYp<+OC7bjTr`l4vDQ`Rp{B-TA$ zHv0?b>+M+)j-%&qNFbMh{mk1*i{S)|r&gR6%?h2P3z<^XPVAXFRbchEot%}=K;^-X z$CgWlYpY>5WdTt82DfOtcl`Be=+9oF|l%5;0uuyTQjPvU!g{BZr;}d-0`iYF9jVEYk69q?I zUBpxS&M0O?%i;Y`;Ko8G$m-~9G>PL3va9>ixl9=b9u$C*2va4fa%KG)nPan?m)Z-K zcPg0y9aqjFY2tDzrrA0Lhc!&+E|I`y#NjODwKma@(k^b3KB{0eRzuCNQifhKTEL%0yW zD(fbT%In@oeg5HC+hxK}Tu?0DRT6l#<@GFoq>j3ghH${PGKj^AA$1OL*4%0OGwtQZ zdt|oz3i4Re3qtfW5o9iGOQ3e2L&+oUOKAoypOS5&i(iGJZX}F6_ry1xJi|!j*dh!> z#`P5CA7r@k2MC6}NIVkiCdH4d#9qeN%y-@2-#C|!ii(!PCt=fjO=t!8AE}-uvH3CC zUM5LZ6hV*ZkI|d-WgE;0VLoG@RVTCqUD!(C4@_SM1{+T;Z660jiQ5PrsN%mN6z#?Sf}1remf-0PVM!j z3?47K9SvOLZa8X7#M6RE;TH(m3(@I!QSB0C&IFP1AlCbC9xS9qn1!&OVCXVDJ;8_s zRYX3CI6PpFSjO0y=1d>_Onx+@u;vFff%(J2t2=T0P&QbWFSg{gN~JjTv6`fIa0cBh zTv{LA7h7=Ks_1PFx67G*l80##a9Z*IQ|DcBCAM9;_|EWvT@*L1^MqvG+=TdN5T?UH zy(sce!Y-2KrD8q8maV)OjAUF`#-7R; zVH%r*fsn@yh(2a&;-oj z3wMxkOj`BrmGdj)XoF|)PE9*dIB|I#CPsthP_HZrdUR2`n(SyL#LYS* zGO^JL70L5Sj3ZND8cpVCze(ZNM>I|lAcl^Pv-yXVaX*Wl@v5a81luP``zR`2TNd^x zXyF)j7U(xy)+0A>Hn<4d^}ZvTTz;Fy+IE@u*)q89a5?BQ^mDg8!Y3`0Hw7-i^!|SP zfsa)C2!&REpT`iawauWS!jgJrcXdNRC`igutl*i+;ZzvXlWsc`OSyYpA5{rU$$Zl8 ziYzeFPwOJ~uu$x;?7igz9?Zd)80J-S8z>Vmov~!pv))rgynBRG?fv~VYUdl~un1J1 zI%HIX`oLJfW$kdMmsD~E3LZ?H2uZSQA8JUJ)geB&9tJh2IMxd(;*@6?Xnn{57q5fz(4QzGij3Ybj zwBBrDt0V;K9ZT(CO>o|oIW2-ID&k{w`n!)CA)0Zt4jG-_Ja0dW3#_(&rO6PS{nNI) z7-b@rWeyems4v*!+v~ID;~~LNht{H99`ASHyYmvPcZ*2vz+5eI8=W%pv2crr+_{r! zF9YW`yB)xUI%@9qEw~}JdLKvZvs$eWHkbTp5OnHUuTu0_*2I(Gx73Fqb*fp4@!n>$ zpexv~(qt>jq0Z7iOSu(N#gY}Uurj`+yFh7KnH?nC-=+LyuDsd*MAlTBhEpd`iR#!?F70v;}h8r->%? zx8gv5zT8u+&OZzn+N=8q>=?J{+WI>cKFSQie6w}>f%6IV3&H0^!H~^cugL}e1=8FGAJU_C|Zu2&U|`hL0v`l zbDYVTWSn^ZCso!%LA$VH z20*Mgt4qHwl=R1L<}&T4T!yuMBzxub#xUUS%N7Un0%q8F)epWq049A2dxYgulJ87BtQ|c~Y*NEi&K4U?VXHb=Hw_Q@D;o*>K`@cmq`GWy-4yWG|&+ z7q242iEU-jciEO34oVV|@+gS~yz?Yd-n8T*b!x$4zV)DyMpKHTL1#g|aB>mn))0Cg zjKNE>Zhq_av%Xt+J4JGJRU9u1D)b}gk8^)>K0+^GIt5$85=1C53vf(gOm$>vd`)P6 z@%S7}{tOXD9^S9(R|TdmOuZJ+o;OkK`L5)0Su7(oS?q6Mi|s(ZeFeF2BYP32odK9) z+UfWkoDm8GSN0HpAf=Q0g2e4jImIKfLDa=dNJD7g8yp@T*V{w+yF);xB(rVe=RQ|j z)tib=WIt^@|DLFYznxqXqIlVRZuuc7nT*(1#fmeG1Ikg)(JEOjJOq5hnL7`QNKhus z@`ppj8_P!_;Tm<(;z!HDMUt1j0=ZE#I+0X#E1}jTf>x%3Nz%RcwpUVPxdpT)DE)M~+G&vT7DItz%N^K|bKsJnhB24=pjzk-DMg+D1a zCpHEsE}wI!o$%{-^2MOV(XCiLCE=`U@RS~m4S2{^?=K!tn3{prM~d0ctP1N%`uuPo z!y$j;`lJ?9z2#jR0Mj1?-sbE8Zj_#5bJ3j<&&R);=%F;&JHRsxTkI1;O8bO^JXjIQ zB5;sVFxcv&JVdvcd}%0z))P(%M-GkadeB{C`irIOH0SjOZQdyOswicOXmzvGTrRBd z$ha7;6Xsd%g(malnb?l4TQ5Xa`Q-*(bq^A{q4RaQ5sNxCdzz-o*c5ctrt14Zs8A8| zs2Yg|+CF+SJn}4=3&TX*d&y%3F5wU1YBDlrvS1xX;{oIFa6ZapOrmTy9ql1%DVSb)~y;PjlVpWWF z;poqOANF}p+O+1+bM*g*Ip2D6_G$+rsw}{rvLXIv^)cpnx(q^5!lRvXKxu6v=z)WX z%Wo)!qr$D+GL~`aD!GglvJuNGFxiBibGVN{nV^X`ZQZQ&G&fNz^Bv!sLR!+QgxhiK_-d7yiunh-izW zmcGFRB;8n^3%N!EnBeDw9GM4o*eO)RV&$^^HeC2#uPKnEN@7Lt-w0(>R&>2$#h0v< z^;GjM>Iz$HbZFb|O?gF*&FqA==y>P0CXYLKAIA%zWix)g3FON;*}J{!8N2!(OBDx= z!je+sOj&o`Ef|7^l-z*g<3l-ijBqfyH}(9@VS97Hx-ew&JK4cLKwT+=edhD2IQHG8a!=H-JT9KE-A}G6-wn>xL(C3(;h6TO6SIK|2z#N6`dnq-+?SgmJyX@`UXZ zFsaYu^82LfT9huT+K+m-S}BIk#ovr}Uj!!q2(h48ZNeePfr>X;1XWF`vB{sl8&4&^ zSQyVYlv~0&>p$IQQ{Z0sQLkJlbO`!gW^<+C@=>|MPk2ruzgDY< z$mIA2HZ2m)6Z^Jpx{`%Plx`*RzfQr-zSY}k7jZw7zHqooYk*_G*xy1;EA5nr3)!M& zPoR6M54UWLOg{sh&==`Z5XB!EjR)N9YmwRCe^W_0VbyQm89Be`$-HZVWfavIrfcbY z`=2h~IKi1Q9OQtLClUaPDL)VWb13<*k4K>VfSLHa&b3e?RLci?>85wZt53{^97em} zi6byd{ZP16%3b1Ec&j8vO}}ON!k>&vw{EAH*Un2xFcDm5B3UZDUc3EHeP4Pl@NWl9 zu?$3i{-f`LmrM8j4V&gK523hX;hmeq>7mlZESh^%VCHJfGo1>5F@3Lbw}(5F!eg48 zL$a}_0BZsK~$ntKED_X4>J0b{F2bSh@}uxa+MJP-$}3AASjAM5VOtR$5B=JkzhVDa z5nVlM_Z}DS!_yJhX)qpmJrGud5$Zs|aK!A43O)LTGsO!r%w#PtPr>^8WB>BgC}2Xs zZhNKa_15p~jL&&-x>WNmEjbhQx(`kH6%vJS0s9O`r(BoyUyXG7lUrGYqVz|h@JU6X zvjc~iE27e11Yp5 zIkRYYm@LcotFF-UFHwpokX^zip?%z(!_4zaCl!l`>Y6^~j0#HBuG;Rw8q=Qzg^I+P zrTN+!im)9wvZjo`FLs0uLDv6Xj_hh9zjiy(tKuakqJ0x{elx9~%mi`@uqGcqkc(aw zU0*>G9&g|W@NQ4)7qhk<-d=L8B>g?h^p|PP=V!>IqW%)!bXvr}X@58HyPl&uTk?RG zohIuSa$1ZOAB@DSYLu(UY>b?V5%a5ONm$bz?kq#5rnX%+@MJJ3aw^)cOQ(DVh0Z6% zR=^jcDv|hX>L9~qpf$cj;&vRHuyc-1RA4)pN>gSHXm?;%d#+mX$0tI2=3FZnl@oma z_Ua1cO8-F-kgZ~P@2%~3a-(_%6MX=WIGazdjej*qfMxsSHlN3*Ph)J`s}p2+s9KAB z;X0pX^hNTtqV#|8za4$(J|W3kPJZVrA`O33NniVyC&gbME(b!qpBYbG*-)ebY|sMh zq*iR17DftP6csKQMnnHzP-EJ|^^XqIqx9`s?RQrRf=3XzF7iXLNnjsp-1aCk0%6O| ztqYq9!5x%VF!5@G=`K|3cT;>biod?j7aOl_@_2fq>3??b`O6{ZhmP2X&I=vKIImCt z>0$ggK0=?9Pyiy&pjI+W=1*OREFJ^zx-hEtNPZh$@Aa19VOt!%apqO5AMW&=V<@Z)(ZF{r3Mag`jxU0U?u$> zjQ@vU0Dk9({EFy{wphQD=o8O><4>(HQ#ww5hdgsn!fSgX`7`Q&{zLz~=seobn;~#Y zWd_}dH-GN?=6H>g(!V0U(`^yb=kb}|^_mUs;*aCxOKcV(fRD!;3;xfK_J1Fv@xTM) zI<7g)`FCy&2>QBe|ITg606WP~#sHrj?@!N(&ti4JhLu(LwEgE7HX;XXrP2Puf9Ji3 zNCFnXQl8__%^-ge$peQmk(Z9A%AdRvfBnojumEH%jD^25ia-0u4_E+RiqNp3Vf`DP zle)kHWM71&tG5qQM zQwCZB3lNipH~HrmmJtW+86;1gnCx5iJO5kdGr%?xF~!DJ=ML%xD^*|>G1T{6y&AXs z@%B#_=(7y63ZRB0A#r;DA5vqvBapxEV*H=oKIMzw&wfn$=AaMyfUE2PTkSfz*?Gn4 z#QG=si08>GOXIhAf zF28JJe9ug@sZpUu{FlLh_A!*vK-mG*l!gDIru=Ufl;ijP)38?plqYb1hR)!_Mf*dE z3*eJABv?!sOUIHS8C5j&%PLXKx;bIQV5>_BATMV=oTg(om_}<|dS0K(1^w#0*^hr#Z;uj7DqMKJ=cTCm<&~D2 zzVlH}t|?plap}%HIQw<>JtvRZ8p9W=u5Y<+<0HNgcUaapF`4nL+T%exJ7txh+$TdO z{qBYvc8>7(r;`o+zdk$N&3*laImt2W$n-OEFkAbzE}*$gU$;Oy2iMf8F-5(LF@FX* ziezg;rTzA{I;wdk2u-408TN*nnVJmtGl71(`AqMnPQ|;667&XXkM;v^uh=%(HceQT zPfV#=Q)^ZhiIrCFnYrF6Mu}QekRZHus{zBzc-VHSUh?ktgzdO|!~3vpEG_3?<2QdC zaB7e>8%>g4^7GhKxi2!Ii#Yz}N@`Je^W*OPz5{D|0p41MNq#H0X*V5OCuPsw7sqmVHWx(JB8jlv3IQQFC2}plx-I*;kVN zr+Q8S@;fhQ(c&c5G&0OS7$izZxg@u%@AW4@d#c1(8ttkerq%i`jHT4Ag*Dz0f|fp| z$d_sDYj@tAV7=f1GS%Nr8h+b%{*BsVORoCitYNB)hopbPB6&!?J_RO6d^tMjLQpI& zDI?$3@AZlp;eQP^47aC`Nf4PqJr9q7GY+YG2 zH2xsKExh@7;MM=xO@I?vrSrxs)}J!5gvF&=Wq4M&I-)>Nl*jj%XDX5&dra^B6vlGX zNwSLQ-Dz!vbXgKxggFvn8J6_0 zV^K0L@kIKfb^iRhzM;93=?GQaDWGSwk=6<@R;`En#dm53CusHHTAOqVdo@3H$(`iA zHQ5B6)(QsFxE!ROVJ9jGy_Le)gVL17=rAPoDQB?eHRx?78)g|yd)0+mPnLR3AzT6z z;(5JMndS8&4f=>a7HxpnMO*2+l5*nNN{cBk!IP>0fUYEJeV`sWmEi?UJ(w8Imr0!P zL4EUyMGCS3acs`6F&T^PH-o&H)4=j=>=nHsEgAU2vZ1h6JBnF${Kcjepc-x|$$7kL z1>M6R?ngb+wKZ#^B0@D!t+Js<@o72Pe#lV!FidW2D_{^CUay;UH1C;*@{RxS;HK30 zi|htpR%C-A^eb!gR_B=a$bU`bfv+Gc1E-?c8z35yAp4w>7vr<(ur5=9JZ#Z%Gw!hS zIgq0zAn#nh&x2he&)J1X*}S^l{HcdZc}U(S+xva(u?_-?Y?)zp&diH)SgI%sg0Z8U z-JZ=aToaR#Z2;Dv>g^Pwjfotn+y;+xYNq*DIFFqAA-@+!G);iJH z-U7y(uQg>F?Fa@rlW^qgsKE)wMuJuYZ+OfX6HP}&DpvcNYv%z|QnG;x<4{iP%Q1f= zoW!Bm1|*W5KB!4w%BEG}Z5}Eb71~PC6rd>(2jpMx6bWFbP!fI-TmTreO}@XzWh819 zit0;y))lGaLoYRe7K?oNVF1|2J*H8y05MLAkbBFZfaH7)V<|0Si)4E?Uk57|2E2!g z`fk%-#$PJOaerj53WWo%F;aJc(@d#GWMQd#xVpiLKazg?IIeXOW+s~QOUDnYz-U*w zL08`kK*vMU19||(@=Qb;I4<443t?5vInDBJKnB&>`0{nFn;cL?#u9Jl6OfS1Us;tCk zU3g15pfBV9vWij50*a3hgxUL@eC+tIROQJ4u42?&tJP91&d-=^e#Pi?cEyQr`roxT z`m=b}c;lYGWa^X}5XBkiQ(G=E^P@($U2XAgQnT*r{kA6FxZ5*~)0ZIPeTt@0;Q92( zGz@8?`7He@Q7HVE$H;)j?%VJYuxU@CeE!=ePTB(TYk1iAfIQ(p~gmp^K_qV&nL8~boh>#=0YP<#!3{}aDJ|nHLqK- z9A9knNge+tlk)fvv5#VRHbV944hqf>7TC54gm^NIMkTen{0e<4A6x&Ncv0g#br`s_ zx&wHDis5}bm-Azk#Fn_<_L`)QLD_GkfB4Ew6Mu> z@H`CSvH8ZMWLPk#=`ePgah)G}Dm_dDV2@*x4Rt-Qkq`(s@s=8hF zaW2l2Mh)A4zHF)IXzu&-pIQL*b`Y3{)KYmLP57e^y>fhA+P%HC7`D*_WE1ndRTNyC zXIg*E@z67htQ=WtBE8d)E#m+yt^$YP8=y>p|!o+;wQBHAvNZ=M=hl(Rcwq*zoMT z>}`tpoLL#-IhD+)|4uPos27e$JN3N@KY>2#?N;IncU0N^K?~A`fw0qYh$rid{^Y*t z^eNljX$E^}<%rLmtxX{z&28orcuDrDo78-UKuuVyTKP!^qO5wyB1`?{*G{?%BB)5)B2i&!pg%h~vPigw_73ZoJy^x+M=YQ1 zOrd{0d*z}?9Rgl11BV+C*i?1`lSABPi4fuEuMgovzcc(YpNq&U?-)T@vx!lqnHH|l z&|7zVlZP1CeW=`ZI%cw8I%YZs+={RQO(?>s-lI%DeYIXs`c1F5qmKqG@It${ zp7*0>pj)*rAa^5?C7RNZ62cS?aT`4#>vFfCmdM%DDFS7hjl&z{%V2~RgtM-1m~W}T8P;Z~?{u|cv~YFzNNWnIeaxV-k4;cQP9;|F- zI*U0gOD-O2HrTRgRF1@}EOsp48awW7BbWl*xhdwm_vcqqF2Lu_g5GMMcQ z_a4m$QBt4gD48#LfeC7MWDiTX5_<)7w-Dny8{&J%^O^(E!#KEG+~kSbW-&O3Njcj1hFG`- ziEVooQ%ZU5=kbYy$9_FoJ^8p9O&e9JV_xV_9R(Gv6-JD{e-f@>QlS&Z*i9R9eX9^2 z>(^-hMCWWT-m97Fx=p2K*)P2KX=GMh@2Fd|c(u)i|8p6_XwzQGJaO$1yTv*Ux1D*d zZk}3l^^g~toK9~!#ix{!ze9w1EIuJa$-eio@y~~GXvktxU_XcdX0zZtqr<8!Veu7h zF#BRX))*QgL92kL&Q;VDt|o1OosA~?)?i`=C=|HWuqLwP0PO@4jQT@0KldwZE3w4x zhkMC&T8Z+7&M)BQA*2q3=qBXa7UK=U?GNvdd+~x4+eP#>s`EnkdN>^nC&y`?mF`S; zg7@R?RDDbej5UYfy5?0CQ%N^gvc)T9;jF1v1dYln?h%&L0^s5PmDd} zw7#kd@Kgn0L1GZ?qt|9x(_C(D`&Fm&AwNF%N9@11bfplgN{$$h3xN|DEPm~c%BIGc zlGn?=j+I&YowRp`9*5)4lj%vIeIxQ5=?W&tSM-YS7q%S0jaX)|0Z4$~l0<5$_lR?H z!5g@-jGk+A#vyL~S+&@rltDkjFn99FziBLSeS?0IBc)W~wVrrQyyDcJlMuO&vgZ~% zdy~QCkQ=foBTK!&{!2WZUau4=5`1K$kk{^eZpw4dT`RtUP{r7TSA|225TgKzR?A!S zSRR#q(Am_*!K<}#+lidTzy@w=&2H;zWN1IhI%|9VgW@)K_1fH3?vgm^uwijs1Rr1x zuIlIGf%5Gny*ivuod1Y^K?EnM)il+b8bRV#5|MzWErizj)>An$bK@1QTb|F~|63wZg7q3`gx;=AKv)wLY$MwxTct zvzNS~NbZ0QeLPCZ$SuvCOyiBN8%4s+(@c{w`DeU9CCmd<#Hkc~=Ebc@zrqJarvd>J zffArZ%i1H0s56tVmszX&GK%{ZUj0$K=%Fe^8G?&Ne#<^*Z3{A0alkoG z_;%~j@f(AguiIt9iO;b5B>Gk=dcuu+(u*U>Bc6*5=5R9?YvxG>S_|h=3-Ys5fd$l< zU~T7EL3~sX&UTkHsA?jB1VaCM@3nsKXi^NiJ8>e-R`Dl}-X)>QVXkmz`mei7E*6Sv zL}$Xs+9&lp!B`?SSn7fyE6XRPMROSdA(Ep0+f^VkJ}#Xi7RJx_N%_9Tn&0z?bFwWO zYC8RqD1;765BeK*Fh~Sgbwqy8;5@l;OZG_qnXdjX@LDmD?7p|6^rb<0I8D?{6-Ipz z?@`3&_yPU0D*&X>!e(PSJ)VD-wdT{bWa!bZYB8AIn!6Nsz0~G0nNjD&gaLYvFTvud zSTRh~yA0f?U1Nu#>|4JPBjjv4vewy^Ve|B*M|HcQiaT`s(sn}*WD_}5WYO+~Brl$+ zGV0Yc16B|;oHJ~_PvJ$GV4F}l> zA;^BFrM4JaXYpm}^CoX$pCwVr=W~T|Uiq+){E&>T*8YvJAYdTn0La46Eg8-^1 z@>%p}XfO5zL6+QmGjwV^iHl|(Y-UHl1-C6$^6A;WCRkRPV?YMKlSJ#~N-9cl1R}>z z-kY>u8sCMXFpPtI>uFa4-hk_Ei!Y!6l#j$8xaDm2E{||a)u!jEEmO^adN=9cZ^SRQo z+D{wU`Z(an#y!AKFv+tuZ%Le;E?PaedW=Tdor--#;eQzykz`rduH@kx_+r91{wOQ@ zZ5E{;^+k@=Soj4^w?jy!j7~hqI@!(*+*PMIUvA|-=qg&073JN12ijfWqP!};N}EWn zr+vyr=^$N2=WUtrKb8 zh;@eGRSuiTb7Yo-sCy%dh6P_hM8=G4fD|DeC*)1mZ?Bg;GXF~5HZkhB6t}oSI%QX4 z@hdy@(YJG-V*5uA^cTuD)5Pn-PCET=10FmKQi@5eZt%YemMZGS3!b*tTlmz(7#l3C z-`)371PoXUJ;>z5zNJgO&3nxE&~is2p-%Wf;$BFA;F#C4#Azl2N(w4 zx6gU!+9-~R_9&QvoDME zI*xbUqJ5IlD4iS*+en}u zV?$p4D(w1G=8-;t8**Oe?UO@!-vJbQq4AgwA$;Efzz9F?TaudL>qkC|$1$m7xg_Sdw=^Ezw$%8}<_LnIE~`-N2-iCVM%*cEw|I@Q65#V}D3c?eKLH zQnsyLWz#j@18l1FM&p)_YRdyZUFdMkd5dq!RX010S}_o-zSyMcAxX=38*W(NxC)Gt z+cD}Hr{!l!#_!D}=Lj4p;U5bie)izy?XXRr4)k=s1%ErVyG8dj{LUXojuZk9C?dkL z)>xrb3dtr3^RQpXMnT0c2hH+su8HhcxOsqE2q%D_7XMK{jTt2)htfaW`TY68JuLD7 zVl22ZgMJ_M?07zSDR3&8*<00Tb+C1Q^Z-yl+AtK#(Wqs}sz)#e?;SO=+`lZMJriG$`V$P3xM{n@ zRbF~UuLw=Cm@&`h^h>)6i4N-a=MiJ=)Z!K$Kh4o`EP`cUx&qRAOcSVyq9#ddY00lU z3jbNybgmfxJo733bjc5(n>!Yt_$uxGn*2OpFe=xxHACBW_;zLNMDBAr;W$bwJ!L^! zF?+EN9-%_h!IXkJ+j$z=sR_9psfPH$TO{$nA(+C$aQ!KZR2ZH{5N4Pi2%h~kWHX2j zT%Tki3UphYpXLGZ6t;8VuE_$=Ao5H>=SIh z`4l>;rC0REW@q2`s2TSZFM)*ik^f10^F?8X*JE_Df|uqIqGPVb-X>2@WX~vho?J5{ zaBa!Rf5}<5km7MQC^oZ?q2c&i&2PpVZUhM6v{^(hJ?z3Huy{KqHo>9NB5O|NJ)h$I zp8Pv$G|DgM$SQIb)`b?wR-ifur$Pa33wH+hBgN7pZj2%Bh1_j|z?XxaC^5)AXQk%I!AS)EbPICA>73G9?mif4}u#Djt`N)7F%g5N!Og~r_;MXq)rL%k@Zg> zEUap4lo~S}L+hp5T~9dYFXQEaqUTYxNMAAY0X*b`Us}^37v${g%Kr zsaC9&UI!ShD*Dn!VKa57P@vHfSMVM4z=tD#n=46zo!;Q(;K#n4bBPsNEImQP?m!Z{ zq{T$rPrN*j#ETR1Rj=bFBH5nZUP@EuB+5lWS?v%nk`94Sq=ByY3YLUHML-Q|Xw(-o z%=7Z-946T~awm1MIFz&kn;xg_L(A($luC-v&i_K4V(mft6+Zt1bxJ9S}$sZq3e=8D?g*oR=-}#;Cr8F*uG)GBdXW@SJoLl*R(CKT|sUmG}p?Z zJ!Avl9`|B8Cz}jR%c82JHIO|-P=vviUoYa7W?o?PW!XPNaPx&f5C4mR ztJsGHe+w@E%B+GnY^-i<4`P0wD7wTN!|hqZhLLC5)lcz9Ve;Hs|z`q zk)&hrG1+{@#1q4#;Nm}{@d$LV|5GC6A7i-zO{(l#uIg;hN+an%j#%Qzh4tAC4nJIp z8l|=)d-nd6n4ONY2e-mSsGe7QTRLnqQHC$LBW*sxJevvj8$2|kzhevFdh@grRBL9Hjk=_L0cjyT|{))z8G`F{2tV+H(w8_g?NG&5P ziWM&hYRy3Y+(OYSS(`jO8JdNv@?OQlhSY(dA3bXhfApiBQeu`(@cuQtbAFBfe#S}-%(gQ^#3pFdI4Qh#Qpt?G-QeqgE4xk%#B zS1>BKihn+E4B}R6YnrfHgf1eXuvV|x8__NM=H8F$sJL89x6Cxj(K zjtKS`QSLAlxbx5a-B;#^j&a>`giMMT`HQ-j*gcB)VB6qUOpdBiAKqnl>C~?_u5|50 z7FW(>8P(ImKc!|7Ufzu#M)AiBV2cs!UI+EM{ZYwRvVli}IbMoWHc=DUQLHL{=DvE} z*=lU!B1nrP-{vn@=z56Tr!SD>9S$bTs^QlrA!% zWv#_3qf@zr+<5uhq*+I1bpTr~)&2$$2jmEMTCi#Z^m!nFFgAs^^-{ zRH)r$d+!T0GqroM5vHMl?20RMbXT1Ak*3QEEuv#honi{uz@ULX_c>MMbW*kC5x_RynzmV$m3NdQFd2L$6TS_BUcH1CUDjc+dDiRHw=DF%4?<|m{8(Mh zXB)~CK`aPxKwE6|pO3i>KExqC`Dr9MJZ_sE*kJ-LiThsvczrLXU^DK8crIUK%9Xc= z~o{M416NASx0%*&R&-XUTnUGLVqZp3Fntki^;3=D# zb#ms*+y3@;Gc~{jG92O-*b*3477Q#%WCdfRs@ddl z?5YQ5&j;G3A82DjJWHY1PdJ>t7#;UL)%yKE;o?3J7ehn4PeYCJuU}kIZuDaB(j~oN zEV+Den$>yHAq;jY&%1lnZhX-pO8o|M{x~Y`3ZG)=;+emEEPGR{W1G$VjDcL~o>X69 zu9pH9_c2OHw;}kYF0p$yvcR1Zf}=K?G_+yh&{Ys zgGHL1T*Kb?;yqm9X5%;p08E$o6dFFRFZcAy+7Mf#uW>a14 zm|4UcB);3$8GEVb(xTkQJJ*W4RBD>C2Vqmo@)==frdcGt#Ac)Fi<-!MDQlsF+UgGl0IrgcLd)wxS zkv^oZG9Ndxb|Q8A70X%_?jQZhf}Yd(osqn&VWgr@Z}hp`)|dJBvWRfHAKA`V2(eq2 z$uCl)i-(Y;3SwGe^zjBiiB&Q;3)(yJwe;)l9MNa=x6wx34GAXloAJ*=Tl56BHlKar zbTo>kBg4##37VaFre=ensk&R_GkB5T?-j8}QSqA>4V&TugRcRB1`#jPor z)puhL<1{rkW4*YJp`XdyX9f~G-IOzAaPY*8E{@{_Pivia%Uv!Wche1i{nj-dHF+*@XLHr8pTjeijWMBxTwfYuL|9^WO)HGdfraz5$0h= zr4-XcGVUJ-;!+Pu!C~g_sEU|g+NSMQIj0H0oDL5dj#Y_|$!%&sxa!XR zX1(ApwS;WAp3$&wOO%XYnjCStCb}k}y@j+?bqT0lop`4$P%JNcdX(39dXZEfR`@aO zgqsd$Y?0HwueqyVA?r>Bkexm0V9R?K_sg|nxKTh3C=wf=<4A!=qRX%`^w1tFgWtR6 zHNNn$XY)FR*r6WdzwgPPE)BILHPuxsNA%J{VleDzN8^KjCmiQj+48s9{_zCdm$6I{ z4yqg890K9dA52d2qZXoW`4f51293JdUwyWri_TH=Wsll?6-BH1$L3N9M6L zL-cD+BZ37PSE~aYWu6XHVEuxs=(JIzP@A^ic3PZN(_4`g9?K8C^WjOF=*1o;cxk+1EMH_|v5+!x%SgJDtxQWp(hGOzq^lr84f?_taPy{>c6W1as zGbinyK2j1oCue$|$x+qx6Pyp+8w%jw&o_(k(#L$J|T@)7Hicj)uiMH9OzVS&Y@ZW3PYZEd22_C>EcwuZ4hq zY?a@c0WXl;phSrv(VaH@GJ>q59lX{Oyo&46I>hZ$ps4sMAnAO2X)-U|9`y0nA6q4> zaGAM`8~*(7FF!*M&#!49%nxP2e2P$ip!*89dfv=9Qkl#pA`rA|nvD{r7{A#E1?K8_G~9 z8D#ZZQm4F3tN3ZG+}Z z{y=UW4wZHXOVC2Wk+mOLXrVseaXw$SoRT%7?V8^zo-pG(e^Sg8{Gq1>yP?9ynhI_T z@TU>o*R^Z2T0(Rvoieec_Fep}qK5WWF68T*Y1+gU;*4}%&pDInvx`i{2HRzeQ-4~u zhTtw+&|C0ZP)Vzp2E1Mn4w?t9%LTl4{dx2kg(y1@Z)e+>yK}xJt|4}-y<#!;WLaX? zkjucIpK*Oj$5#Zxx?zf$#|@;8b|HRm3u|7-(TzsT@XC)-Qp#1U3<_Ydd1!>Y zZ&vsqX^)eMbs{-5&3(usvi3ANIQdCjS}LZua}XC8FUnfm?LZs-pQN zfgOrBwLvB~f;)EGps`@P^|k^<(k+}2{S)uwjDX=6hVJP#!CEi;)5=xL@;x3K8bbkNK~13DW7_v^LeIV@SrV$x4O5zO*3ptkMYQ zVNa;s>n-N<_!e9*egsh^I#wN_(&H^#tQl)8FUmc)x~UBqc2)T- ziq0fDocBE7@>(W>(u`45FXr*DgX#LMr>X6n{JTecktPE6M`#`J4=J%?TJQJ%(+mUD z%h8%7^&4*)l^I>IyG(^zC**0g<1Nv#{xkl}fv01ueDm@r6{_t7X$mv8i!CpBxv&JZ z)+2x@)>+P0)`G&=6*DNopOJ`#08WKBhDKO_KMekQrL&dJ*4wvZv0!x#@E)yB9& zF0mQJI+#m;P1+{SS<_%q={9JVh}shpxJ zXkjPA_az1iz{G+(PQk<+#1wD7irVH3T1}Ql*Oz}G5@UE66xd+Cb26cC?vu)JtSHIi z#jPZ1I7f6M8>{>*6u)5QQhm^iU7LrAm&uvNSs{gMF|5h1h2;x)3XeD47eXYdLCPgC z+*0GRC|Ik%glORMp0Cw!(#;DQD(ZyL6cZ_?)NrH6-?Cb}|bH z8*gouRgfp#z|A2i78?$PVzjN>0E5{fu66IwW9V~0XnKdPD@Cq^`2DSK9oXD^Q=(fg z2Xd>R%r~gTbV78jZ(^8M)_WSom+(|$qF!Kl90$}69|xSD1YvEs3P$Ba_ z#JIRS_>LCJGKb+p!Li!R6J-csNOtw3OwVpjL1JoD+UKZ-{7phNDAM;|p}Y$nNNtuo zi&Yltck&wA3lL^K@RupD>SF{q0lKmygSdm+llS(?-B;Zl>7Lah)N|!ZP z!%nvH6rzudA&Qu7Ytj$E=X5xitY5TWD;w6P4?8i&&+PLIev`eQBalmDMEwpj$83SM zbKI@b-MDK}kYwcT=2U)DR)5Z|ELz>J_?BRWOlQp4Zm!jBAxeV5@{mElCIpE|kS>QHBUtBC8M+h5lA)MDpkgOu(pctzVL)j`Ob3HAH8 zf?BjY3iF*Ws*z0ulDG5#v$L|SrrsW)S!) z-5OtWr0rmErj=4tavBFY-1|OuO|SW3I`d_ulN24aXl%eY2ru(}3`z3!5HCZ>2%h4V zy{GuA?*-i-3{s-GK9TozJu7BOG;ph>Gna$h!;3sW1W|ACOS}7?vRr?;|OGB81oNVt-;{2IP~uO?I6e-943%6@`!{^aQ6e+X5tkAbUW> z#}z(QCD(PB5|^xwaZ%-Lq@sh7nVvQ|Rh_LzUp}R4SM=?=Gn|Macs9dM;iQL=Z2)n& z=HJv^Okr5#YZQ59a`^OiR+6Cw{AmBp3``2PS0{oTQVUINTIJ8JZOBd@kOSHw@UPlW zUu+HQ8egAtxFyJ51nux*kTO+&xjgjJg@DJy_1kq>YAFTnV_fJJPNV%}N1DSOQIK*e zG*=N}BG@vpmgA2~2Q=h@{rqpl@4Rn^NpgDkKNh;`2crU+6t-tC{ zb#Kgkf4H!Pz@l68bW1RtFeC{lNib}+s?RW;+&xdmTb9hC%k$nOEl8C*HIr6OL?HX^ z44jp?tb_5F4?s)5R2js`*|g7GS)WS}|CD#x&=^U%EH@`x&%j9jS)k`e2X!L`x&?v! zo{IXbikv0|qEPhsBX`FNwo+acWb9`%r0U=kwfh&g0$(S6dViHQ1h)Yh;Tzb})N4vN zu2=fb6cOH>kc&VY2t;`+l{>r^Y&R@{vh!-%8C-C-X6x!bCBYO3wo>O)9?VDxU5H3{ z;uCabwN{>f>gYWoFs1W=N$ys#6ns2BpfIw;fL9x-l=@+JGF;gUpK@)w1K#!%$L{(j zP_g_aY`*eQ(3dVuGsJG%XTYyT+_3D2y0}B0;0W>~d@B2v_6J~^cARm9funq0S-!G| zz16#Qj1#m3+)=}krJ|#E@g`w_DzdYpHl`udoZahae1rh1M)cLWv4hWO^gt21S$OGh z;w*WT-LwLoRaMh$(vzr5$Aj(Cba=~IXykI7j}RS^BD~IfeBfqkW||kA*I6mVJT7}# zo{jUkR|It>xXa(lq;rnw-X$G@qMaT%U^9;zQ6Eg}PChpNNwe(m$-`o=Nww?68_(0j zHP(yzHZP-L5ZoGmm?-$V9Vd={bI7Ove1k(Pk8$FC3t$Tcbn->`7Gyd@5syoJ3@NzX zR*}Qcho%n_BSkJ6NQG)7-+fTVUoO*qZ!3*Wwa!s8$mP_K`pU?$$l(<_QzY9kTgBbs zqacUmFA$p}FKZ_Stey{e%W~Tm&42Meh(o=v>`g2P*T?{?PjS*hxy3a)A==a-3e@7h zH{BIla+8L1J%Y2l%ntaJa&cIrT#}qpIWk@X>xv=R-PzG=76f7zh}~B;#yhA~0*zv1UYKz6r zQQS$rKG`I#4Cf`1Z8Xyg^M2p54Z^J?87X{`7$JEvwqwx~fUwv5;k)tuAr8;T%!PBzyFj^WfZY(D z5q_7Z7@L>E@q0a&ssrDh|EXFPq4RidzzI+t|p>VTe}4&pEOqf_~=K$*Yf$b zE2n1=b;07d*TUS+Cl6Q3RriGWWOs7YM)NRwZjO*7&8tTkU~K~QusMlKEEQ|9#{2RJ zoUx8^O74yWI1M1(szt9CyLjykvE9i9sUfXyG759Os(Y@rMJ3fSw9zs2|?!x2ga42MftD#nq z2r?6s~KpTN{;dtQVPhM@#dh1LRu~YUpu3Qf<;cJC7KAYiLI62q@S|4e%qPl zTDH6uL?-pu57^ZuCd4SWKX-aVF6!cMu~gbw)HS`6$e;AKXWHPdY!Hlwc!jr2$;lS* zKb^AFEGTbFFt@a-itDTP@Q%~*+5!FW;A(d3&cssxXRqSZVOQg9-$L4ygkVeBk~N@J zAL?LS5%)qaH9_DC{fmuc`oq~h_G;rD42Y{nCmd2zGXXO?Ba0Fhq;nc9e^T zX1I5sHF=5IsyRKV!_9ud($|HMc~Th@H3XhkWtMxk$>e2=tWwRjCn8yAXEtOAzud$< zdSur7iWZveBz2gAKK&DTfHA67n7h?lL3(vZygZxIFGZ<^7R1nSpCcM=a&QGwS&4$k z9c9ZEh_qU+clYw8_OK5HvkxXRgD(_P#fYxcsz-Q$!i)jMVxnsWFa4{8j9{i9JFh`QaE3dHcm{?C~AmI~V48 z1ur@Od!w1oyL`^ZELStc+W34Cr71^0)xmc&A)CfzDk7FMU-c;0`mNjaAQLf%bgcED zV7oct1FBFuyxE9q#n}9^G}=dAYl}#PV2I;|R5kt1A*U71#~-k}GRV-@Wf zN}zU1+n?Cjr-q+x>i5WMevW-p8E7m3|6(kNyqxG;!+-a&S4GtRr#-R00uLJsk-wXH zVx75}%f4$zbXU~e=owD0x4~3)D~PJc?IJeIYthNfNy+nY5}Elh6Zl68j?EFY&-&bk{QIxiR(0$Y@=)cZVt2 z#=^7|4S1=6cW;^PB33j5*0)`a*@KccRA;U8u2>W^Cone7tFG;`s&RLBZxLduKAZ{m za3+{)e0*I~cy$|mFlwg{Wyx3Y8q`Nr4-hBiL<(}bHypnq4yzAq&e|Z0ZLgBs#H(%)oGQkJCr#2#+ana+%y|z9p#}j|Ha1 zi+o{pSC@5<&S_kmw0UyXdWUGc>ydV8J)o!8EY({>bCB&|&PoD|wz^34hQM$Rh+9W8 zR!f#V6W?qoIf$TFSzAs+*6kz=k=BDCJ5GP}cyx;tNfDz0bf~NfyE&HUxSOK>5#I4I z)+=gszP@xg`Ue!svzp3vYP~cAI()YOg%8NkN*p-9nOncT87{5;@}QbB(=6#N!`4ok zwi?CCuRYUbLLx)kH;y|Om7`PZj=91x*8}I%%e1S;?bjEv;A!2dPQKU~&6@pR-^y$$ z9sqTngc66xc<`kIWo=9JV_6jX==V{0t+uUw(Qk0E0Zfl*?T@GKkraG=(DA!py~xsF zJND1Ap0p>Nvc%+6)Iw2~izn?aGT35MKF1r05jPR|%S#TvkgH*kdom+0pFnKu0s()N z-!rNE!KADaLML9tD*cilM}s)!fTE%CYl7WUbFxxvAk49%VI$a>_EXPpa$M<^aMSWH zv~>e50i%p#UAtB*AY#0-MA)C^+U}s_E#idV7vu*qUAg2RqYlCSD%5E@f1#gt#&NKh z;R|Vi#Pgz!)|wtjb32~#%c#@*Xc-lhrzr^oY)5CN;2XJS90Mo7ZJ1V~3txK7#KihM zCPr^GT@W&W7a`SR%dQo4_`!Y+yAiU9`oF1X=rTdQ3M%qA#Ei_l= z!-5~y20Uq0Vu&y?vDc+J_K^<>nFG02i!lJw`x;7?)(0N;P+X39N?5f>0>mw+F<)Iz zGA7nIElWh%M>%A*IBVf-(Vb9sk#~=8$Y(eirEzqCZ7q77);B6sw$j(GpQUNtRA&w0 zDGoRn;4L#hw5RO;++0?T(&J?~DXmeVZ3m_rar6@0BH?pskmqb~;tNV1Vp84+vnfivIm?g6xKp_n-L~@ zG%rXCP=fWkPtn(;o4{9s=6=)q2CM2JYLuP)!`PDRYph-MY@;y1Kc^!Hf2%Lvq%UTa$E7S0{?(IjI@;-!h4$N|#*0bJ(*q1;4&Xx6~1 zO>K}|ABwq?<4c6i`=W_Dlh)mjd{s94rzjR-S75WaFHOw; zg(n1w7W}Ts2j(I_cxZWSGMi`gea2{^&s9~h+QD6ATXax??tzA?h|`Azc`5@ko@z@n zqgO`BQU0Ch1cg!Af5fgFE1?PvgBYR51o`}*XQsW5UZW=ad?VegKol1Gb$ z>6H(S5KlNK(@&tIW)`&@ri=KgDClz15t*U_4~GyO$NA2DI_o4wDe=~L#kbRDKgwOX zgC4`^ak@wreMw>MKR$2Qwl~~CmHX@N7Jj#tLAAcQ-@}+nS-)f`f*dzf%DSKT*iO=t zfaX!-LAc!BNU0ukE(!_({z%697dq6CcBYC~^!|Lkb}u_l&Y9whWXV`_MOVl-04qg{ z@t7v>I|~oY72lTbj>~@~E9mxX*!Npc{-2O>GwAW=JT^RtDV*(mn4uOAAm!~2A1T>O z+3d_YZ&LhOG897C$WDRgaq2=<)tp^X?EZB$OoX|v8F+s^*p}JZix54WHr=QSU~l4f zU&M9gPyX5&V_Ji%N-y?pvl>X1?RRQ!tOvi9W5GNR~zta-KxO>+shzzA6`o z^02h*@l$n?4m*{xIYg?7rTH=#^B>?1e7VeL&(?{KXm{!^8tz?VUGkYIU+CE|vq+@nw=!)^MW-D4@r{CL|s8}TQu20UA*>s$p)*_aP%KSW%UKy#9G7xshcq4=O3z&Uajb20u zMeQzKib!hH6!S;@;c_9aGPXWgs0|9eQDPb4duf5WJKaoOmjr0*kjs4W45?|``i+eH z2WLkeI1nQOgvGeNAcQ9e<`b1^s=Rr*sq0kkTrBzSA?BUWce@NaY!)CgyJ_}(AZQ$% zTGs$ih7p9_;xVC+H?$o4WO{pfTxT|#05^{K>F?Y){IRvPlyk6(NsjcwTDIMqe4@3< zNwOTfv17>Bqi%Y$%=e4>Px(>a1A1_a#)l6`3jXfi9*SQ0hu2_je|@kOS<4d94872$Sr&)94OE;ec|wEoB}aX=55k_?0XZWU~SjcxH8bTUYEIGh3&wcm;XVZWwm$2Dd%*4_8yR_`Wbe z8y^Sz&s+Ob5%lv5xzcTcuCzml)?Ha6E=@rZv>kzweRzKS%KNSxZp0?q+s0tG`Co_B z&_(e5%#>^(ir@!5W9Z7A8*(-WmS4V7Z8ga(pt_Y zPDsyeJ#jR?@efM33|{WWM-3j}Xzg4uPN&SqbAjra)EF2*8d%pwT!@c{oX__}XTI|L z=53rXd?fCPPp;0eU&{o2n5L3#9AJe;g-9ZZEyGrqCro=H+DuiI;gRsSq(Xy^h@33q$xiNXF3;Meq=xhmOQ<~k+ zR<8u2W(7nX&Pc*@Mgh-C4xY68A^VJ#Bd|Bjx3<+g|v}N zSo&XKW)4S}>2WW+J=vY+Y_&;AB~TX1Pz(u_nDl}xzbu(9hZwhU^zvB_5DnQlohb^K z;I@r$T$QJp%s(I9I1A0<Drv4T>Oa8>CX?BGt>_#9rq@hf!tLlE3d1PUar|g z&CS~)-DhC*w7+6Yfj8*0W5rfj%c%!gz51NrZQN zfH(Q~dTSW1_vf&Ay!J57PTkgFROnmjljKtoM|{T9Mh3>O`@t@Y!5DH7!b(VAg2k+++=uq0949rbcVA)OB%$SqBS$)CaIS8yiPVeouS zl>?&6+|S|Mv^8875xMWt-bx8ss2Hs9akhS5v{X8xY3;>tKrMf>p=m(6J`-76#DrlJ zxGUx}fZtUW!x91J>r z&p1p7LmC!PD2rCwoXbvY5)w>Ca8C~7i11`;1)*nu@p0bgG|3%0)kj%tN3Dx!(~`mL zejJBlPxPX*!F!Gw_IMW!J!>N+T#1gIne;H!K4ZcR5@* z=lJgQG?PN7e9bB9&(nRv1!7mdZU$g|zUr&1vhw0xvv=G8IPy!Tbo&;8^`Cw=7%`DU zH&>xU)g{vc+ShB5h+6GB;|zH@)lu-CD^1iRfKJt7D$akJ_`g!>W{`gvcxb1hKF4KO zcf5&58FhE3k*G7Dn~&Ac@ontM`2?Aq5woUGD;`p*3p zJ`ws;i3azzSSJJp7y0zTU&IlEcWA!W zc$a`KstO8P|SPZA6A@xmrMD-uO0W3SOC+fBmI@?D4=QfZ;UUTL&1fXxZK9N{}ycsGw!uR z0!GB~|BitF-qlx20e1g?(%Z>D80j2e?E3fUYySs*g&v0d*7ALe|5!_L;=M7P zzqu^`b=IH%IO|`evS+`y7Uu^n>X+{?7a{M`MR$I0w+!gk(Bp15dA@eY(p!Ao-#dd` zhPONZFi%EiHpzw8zjGr#iOmumf&8}AnAnL~pr^@St44@2(0+#m_a0aP=TX2A0Q#-F zgv*==+`R#lI^5^KH(pl$9?)X=`R{I{5vWPjXRi*B{mpIb0Ni0Aw}Sl}QUl=Teg;4) zIVQ77V*K9tF?_L0C3y_UzZvFl&*CjRK*_`bYZ}$x8y}Sb$R!=s2EX4E9gcDUYwMX} zMt}aj@io9fjmNZ8{2t52$pe(!;L;6#`g`MZ89p7QCpNqy)zvST z9LIOvFa#hkF`-w_JMh{ekFlY@qh90F0Aglvk*t7}snH+*){5}}#wW(piUIq>=(#rb zykEce)wals@7K)zLXz4a6-diSx)mQMCdDy8D0~}VqZQ5iPvIffNGs-(rDuW~?fV}* z?fV80gxjjD)>NZSF5v|kc(dh0hn~5&UQMsH zK|!6xLPJ)O+4AIAVRhGbvRNiKmG@bt_J?tmGM!iBu0er^$*DF?a;iLyRJ2DWLx`P7 z;f=#mH5;4pA|phU&B+GPdB}EiW~<4P>lz#U z@UGE1c{a=NA$)jZ!FE24K&mg*>D!24b7b<>nKk&PQYHn+Yh*|-aQyY+hRjCl4$6WZ z5mWQJqE%$X^Q6yHEzbQLzk|X_Y?)nc_FBSIvqOphe)pM`0XcaZNA^zpGU3Bm%R5vi z`&V{ToeRP{`4j4PBt9jiJjS_TQNtUO@w%b%B4oEf+V%EY{xndCQX-(N%$I6+3;VSf zQjlhn;p&)By*GI(308uQuYT@zhfPjzCi^uxLu?tBokP(>-L!=_f1Ko_*Qefp7}OjE za7dp56f90D&gPEGVEA$X`-WmVNAu`sLNgC{EaMW1=ZNz$&=NF-sO*AA6nO*wbv8+C zmi&abkkXT5AT;J%FX0H#yJD1QF)tnLGWqj1f6rIn(aqsv%{J*nAQst1d6!1mc#?_B z_h#*Zi_NV}Vo>F_@ALYCnK-R>n zV*UcyCWMspM2CB=FInkZWMK20Wl1u!$%zYz&`%Q_G$&bs-659m2K{xJl2h#(7g_Z# zoo8M&EPV{UOQCs)NS-QlT>pSym353Th8-hcFPooY)S@F zX*-m}=avc&w45j3RbM~o7;N~ zNMKf1ub;ivU=g&HQp7y`57V*~4N$ah+3({6>}XmF?R>yn`f3ZiQM1+gCTPR6xK=M^ z{lxFCMx-|bW+Ueb^Gl4u8RP8pVm}_@2wM)f9dYsfaze08r&73>M{&0l`i(sm%o+LO zeYZY7GWcULf-}@d`y+eny5~5uP$EaEf89flL^8n61|)hK5~NyT3bZ=e*$i7EHq9ki z0X+IxPM1XI2oo`D2_G*74Wve2N&+Ni5*2JmQo?84jvKMf;3hm$9#|(0)e? zjAa{;6KCiF{in5k47Lz&mDtE@`e&xvK8bIge@G$2mRLy3_RHQ`flil)cxzdle7OEV zQcnFz-)vLAIHxD-lB=sy3R^NR0k@`FX9BV*HM7{9mHzy7^1Ch$IWV9C9 zoE}Ha+uE0WN49Qn)lP@=X=?lG+?vq$%Tkw{c1&4vD}FwfZmwK#MKV|KFp?p3D>IAN zJks5KNBa#)Cy!I-Ca3Nyn{xmOV76Jm?1OV-u>o(Wq#K;lwIN+#*$rqPw|rA3bS@2Q zwXQu+*sM430heeYy$(J&ED4w;#VQL;Ew6G1e$USI=Jh!6inmA!pBfW#JRZ#g|gn6`O=QdQ_s89?koyc-*EeI1@F zOFxa8hO~c5E?$oRw1z%N0Y=p@0kW9g(<*g90vj|nDs;ENJy^s&+I5JF_iv8=0hTm;%zEQ5@iW#tb zSA~PQ%tQ`}Yh2F;bH{KvNPRp%fGn(Oi0&a=Ue&#VFM~!YV9_fx-qG8_o}DBR-Ge?- z7snnNRG6JQhsWv3Kt-}udFu}$5W)>2-{U8AeI#7@Ac?%?CENrIY^bp+Th3bNx%)n+ z3`yHi1LKHg)1d6zQgw!MqvF2Q24&#iMPlg?Z^Hl>8c2lEG%X(a{^8s@EBHkdGy9_zz?1TQ*>xZ9PXVZaQKMvzdo*o{HYJ zr_Tx@qx%Hv;|FZym0n9`ZaPoA?*2G~VP_asWKSc=P2~K0Q`@JA5e}&I9BU zuV~;7yKPpcZOQpuZD^XpT@XOu6dm8Q-iaP^c7IrV-sUYTC8t>?B&3=IWZJOPq8Zd4 zAr7grtkMp3v3+HVaIqP5ZLZ>`%CAA~(yTd3$B|}vfn}4ooPC~QYhsY-1xmJ(##RdJ zXRBA_N`zDG?#p1ejgf!56Y-sonQF|_m;M~AaTKc84&iqrux;(oCF&IN3^l2#+Jz^UY#~P3CTvKbgU_ns0p~~)a~97A`rZl zn}w;$ux`cTtmu|Rz+vvW9&jsL&5i}f(^}I3Hr~jgdy_cGgJZ#Z#g>8DJ$^h27JS}y z#U$yCR(6~=nB(XWgsdjJYs1>Bgpi_M&&~Hg97w(=Vt-i7jh~Ke_64%;FlNX1i(i)E z_L6SqECdw@+t9329Nkh_se5cB*1&rO2GAMvVs%S>Ulungw^Hpr##ZuRqG$4zO zVlC|PwLvAzjA)$$B6Cr&P9^8s6wy?>$Xy)*{j${_U@-fxZZCxq#OLz*?sAPbCtj^- z88GCmFf&#EO=H%9i7bWoh}p)?oO68J62R>swaecbk~oq z`tR${H4~nDI9r=_>w{=@@fS6wML5b^O_N4*{R>AMcYzQjl%>Rn71>dyXIU$MIy0HO zWC!THh8F38D)^qLv-IU3MVsDrSIqUms{>ObFP{*~aONd=2Zr3j-1o|?cBje-JtlmY zdNd6@I)txlxQ8J(o#JM31J*!%f1y}azo1Ncy^i1H=QNWJUh{Yah?#^*12DREd8wAT zb!{6PEj}<|Zp^vQxrJR7xUU4~C_C};E%MLRST*a0m2A}4G`Bl2U2r{e=_;3mZ~jAy>t8S8&d|gc_`?-t|Gm1 z12Um}>EX~{AvtDn;4-T_Z&O|Ct|NVwP1wn`GFkS2xcU}3lL93I)-qh=+4tG(i?kiVOLRPS=gTQja z1pe{^guQRmc2A$ijR;V44U4-boCKwzZZH45*UL6M3YmZKc{WDa>x?bWuG&)I!M^;Y zOsKu8ccq!vuBpzW;Huo3_JzY>@1`V4>5}*FhC)d`3pDxjb;NsONU-B?>J#J(B#T@5 z#aH$1TirUN9+oxkHd+nqrDb-GXw!T)T~>kmo@aM)!+h1tIekIXhr$-zHXW>ENGcAEHh4RO#nxgZf!n<`pc`IXC=3ZtC{u*R5JsiZ zS-&^;x=pf&Ep2bC^Jb{Nr2(UpD-LLb-!E&WUE6pPamY@9jGTR_K^c?wf0}f8hE!j< z2%$djE(Le~?&B9HycGU$xXkH|?HxfI87_BU%f-Qd-?}`$Y&(1pjbP3i=Bt=IFj;Lh zoBM{j*@)c ze>(?`|0AHPr6k!={28|!`XKcQF?h&-1B@T6j!x)$W}G*Q7ipT^ zliiGbJ9A_Pd*SzN<3*^js@g;c_q$({Z;-car5mk)U4 zufB2RyW0lFOlgjAkjP{F&9fI&)Xjp&-pL-_kn(C26sU)GMf&{HfBS%pWRtIH z4|`U3iogVT^51tdYsZ_Y2M;EX9~ZCHmKfW@ZXNwQs?HGB)N_igxG262Vjt@#(C=b@ zCI7c(FE?OphkBH)np=YXhnkIoQ zH_9iWOA_R4A{gklPn&pg_{7Ln2RO(sx(p0SNnDG>X}@1YH7s=LT4iFX_3~GElzvaq zKJX=2&PKPtdGdizUIPB`_WY0%*t@_!ndvv~{6S=@PZQ-EwOQMlitt24| z8A`hWsYBH(znAt3`qMHF+EyYat84mqzJY#ltO0e7rS-)U;*Rq6I1h0zT{z+kRj*b) zh?5FZZNfK=r4z}-x2{zE?IcS>&(61>vp%0Bow6m%HE%8p`waYmrTTJ;N%>dLqeT%R zfnM}0JuWaj&uXQ8RI+40cou~4e-uOG4i{a#f2Ss5qD3MxWCP8wL0_wc^6E``3?#7M z$GLST58@lX6D0~t=c}uFUna}z9Ly-o&hAnWgF^$fTA2?zow+z80#n$=oMJ4g+VB>F?AuMn`&Q z=$a)-AFNQ{ar>2=oE#w&OQFg;c*^tG&fmPo#Vg2KI+L7`qN-3MJAAG2_LXyoKJG^c<}-jm_ClNw=41Jl z`MLfFq8@lZ9pcD|(kPSyx`tStyWLZT=U=uRLDwKW`C7NlV#pD=#Rv8}1e3q(ggC9c zso#&l5ao_PH<7}B7uVco^>9vd8FZd>4|CD28u0s& zM<>aNd;1_n;QQ=6!8Q8Vsg+Us(R_s7Y{HKztQ5(2g^@jZKc-xG_)1psSH+Gw%R(ju zeSMF}*O7dy#!K+_%ooKTU=i-Qq{8eopDQzlOgICD)ES$nn-Ai_<#nz-+b!=;AK;c_ zkMWF++BuQPbq$v%*_sb`;=z~aUb?QAmvTn`1rQ1&nzhlCO{XH((4Pz_3uk6u8ZayM zE_&|bEBMc}jQN~1=XU1GfZDfPSu>n&`!RFpO~#R9Wx#d;Er~gL)ZTLsnu!kOM?%AR zrvXxr5*yY$14Yobf~x8fA`~MvAAC>FvM&hwmdbxD=KP*$9kAa1Nz^GHES?#Wjkd;WOSxm00ka*3(`UQuE4L z{bMWzA;y5pqK=%+pC}Wmap_of)(x*pu5->+W*3AWjy_N5(@*RFSD&>P+)Jhj)K$TQ z32CdN?d2h&uBwL{`yt=n3ZXlh&xUVFC`lJBJ!#vRdS{gujM7T0<5J-7Xa5u(qinl)qkg7`dcw=VV>Q-zSRX~ zmY3twRCzyD2m9vyV&|NvW`gMcvR}7PbE%a8;-gM+AAY2zYuCg3DId?N%xK<7S{HRy zVRRYxej?6B8=-vMMk<>K%6EMJjAZ{skb-Vc`$@U&hTHnq6xg!J5R>09Z$(Y;mPPcB z&Mz%fcJ6~wTqyk=(B`tBjRX*x+gxWs*prjwy{Y7erdUbjME7dZwa+w`Hpur|U{xP} zW6Wnr^ett1qMam-FOb1DzqvU4;qeR}iprKRnD8yJL!!E4ga9J|`8|I*4;U2-5q4L?H zd%`^5rik>YE=z21(DrZt zP7!y#@p-s&S?#x#OT@?qYj;xz`Tb>KjJlQ2Rv*zTdfDzSdeT+V>Afwr+)~-I_4UAq zm5;}x38c4ES&g%ad$GL#38u}ch3%{@E`zq&D>+E)5F#GxOYq&ka@A+FrZlZxdeZn| zPP5Zw@m=rtxu~$i>9!=UJ)5wMnmto;Un%;Xsg8-x*wyqOg5La7l=V5H@v%w)2xTC@ zwk^JiFg!Lg8IDK!NQ5e#9?tL(P`YHSf8BFqp9MTG3g=_Am@1=55xQ{4gP2Yqq{~5y8|1@#0*2Q_1aEI%%Zj$bBwwF@Rbi^Zlyigz$!)g z(#VxACuEnYRneTBTUI0Z!`T(R6Eiv*Pt08<_wDR6f+yPW*&&_86SiK2?<_41a`af; zbruc(+7&ecjKF6U+Ir;;)6cE;`$vEwV1EtSvA+(rW)?)W*p;@{~f5+3?Uun zO_%3~9iNHV`c?D_F*$bQ_AxAWtzNz#?LSE5{C3iGeMEzL!sq;9<(KJwZPw94aNezQ z8XnJnOx$vyh&ZD4CXzqKP6QTOAGTF@*MD;+XJAEyY%*yMN7wJfT+H#?U0-A85)~YC z4BO3n4Vhwok7jia6rAY7ybRA-Sj9}BOa5 zy0n+SKE|`Owu(X5-)r&oZ4?~4COW=JC{iR84SWrd#<7crJ=Py{ir|wX&Q1O#u96i< z)@gi5oeVpb^6%W-CuWKtI|eQN`~9GQN%5$LI6QlYsUEKJ>1vyt?Zi2%u{}3OtEhIs z;|}cpOegYOHDTr7*I5Lr-9_|U(i_uJm#D)jcJcE==+U8Zv-;&(+pIuD=-(&=wun#y71$B|9#8)e{`n>`&@k$#(S3_@SGVya-R>bPV1da*ytg zZG9+nr80cBq=<9NM;cy)2!+4Wf?s~vJn@!O-ob&~1#odboU72uO$#^%Jzks#pG!Xx zW(B>yiC;%*i++uOxK-NfZo!@qq`BX&pu<<~^8(gbg}Xz6^HvdG2P=1X{rgRHZG-m> zia>Q*wrgzfF{OxBl1!jwkOR}+HN z(TJ}mB?sB9Zp7NamV0JJ0{e{&0kZM@18SW~C)_^y=TAQI7!yM>tHEVxyvs}$&gTBR zhCZhgSYt3{KR^Zy%gi;Y`fJNDq*x86ep#!j3ZPuAsn$+6A69Ly7g#xPF6(boeGwrU z;Tb8*T8DjUdd;@YTtg$q4dIW@JV+BA-VVc`d$4^iEk^nQ{@R25$ht|= zE|$;Ne$#6F;*3w^@WGZQYw|9vm&KzkbMjtK{)Qo~=}R+p>on5w>){9EZ8EEh!Y!fO z{LRc_o+qmTL+vd4Z_+Gi;0KMly6S6Gmk4JKvm2MWzPOd7=dew6B*`-w{ z6T!UGPuQf7eK^wfk4v7P;DxZ^#c9=ixUBQw*&il52avDY^Uaz8H@FZNW(aTbF792a z>o+Q%MIM*g5^|20-2!hrt_Mu|k2VSV{E66)A`m&h=%%Y3@=oq|E7O+71%3Y0ThesG z_9^y$6{(|@065s$y0KXVf2IB_%d(?y%CA0`qt&Xkm9-Q;# z_|+i)-j1{_z>foIDN2=Bk$D9k743VL+Zhzy!+)nIKKu%jH7$9E)Z6aM`v##oyC4`z!Vd ze2Rf_r)NU+*(GT9Y)l>zv1~T3x^TAshM~XZu?vZh5Km(JBB4(VKfEUxNj8}I;s5SU zU=K&PREFE{4ItDjF|oPzvCI_@&Se!lZ9xO7zxmCLjrWLIxt44DV&tm*Wma6R^1<%G zV9`DQ{`;H}I{VjJCIY$R8BVGOqwdkZO{osk165Z~Mzv|~)tT9gTy>727S*ReTbvU| zo1mL~pnf{)f;Wun+WqIGYTfP{`|m$sIhUrk`V5{Vy-*4nEB?0ev~k|}^`n%KRf0Y& zW^Ov-T`}eD&b>wrjvn5bt<%`vkeCQ+p)`{AZ5>^=w?C@!i3322e;-$9I^4|XNc7+S zwr_ZbKJzFqK1L9ZNjYBOi(}%RlNY@0)p(`b_>P>*-E7mD+uOjL3yYa_Q^=)yx@|sJ z662lckC`|f7i1a^YT;D;+sLfyJdVeE{olY^YrF7(9|!EB)<3l;tz;>1%%+^r-Pg0<=)W_9 zYfAcqLX7*h69-nM3?FKmyzrY4B%g5UTnl=C7GB3S5@jQw6+5@yp1~RU>))&>;abCR z`vDTfBbD{z?6%QdZYhU!R(T;w$A9S!Ek7K%yV2d2S=!IGR6PxtpJt$mJGWqyvr$wu zG2Nti_&%Rby;PS`a9&yb7WTRqwJjZ%74(3ayZe<+%S`g3_`9cf-?Tz=@}OIfOaAAA z{;P23KlioEN^t$9(E;Xe3d7#?zHeIDz!{v?)9>)iaO;TlF5aDUanDO>5to)GjlB5~ zK|qWpi%kgOQgq{Tu>P|8$~RgW`GXbspWoi|qD0G^gb_aj@I>ywTeuJQF@hdWh+5%& z`H50n6J6coj8*mPJ#)EncsHMi z^b!+V_OjJh&?GA)Uju(@l~o%DsaLG5QO>VDKh@=Ylw|K|H{g#JP^Y{OQ)a}ziZPba8Jpb?%AzZ=LUP`blUZ#9$n zs^R)p#^RhwV(r+6{@rZ#QZ+C0KAVH8G#^v^*!Ij~V|S6O_FP|A5>CT_5bkQ2Ejk@c z#&}={$9L+Ed*2WQbpWido${s)C>T(HtXh(#I&K>_7Tq{k6I|8vA>2oC;;51r*khu@ znGFPXtGetnkfoG&hRA&RQek#oRYnb>s%z`bjFqd_p=wEj=K8cb#JF}gU-A0(RHvZm zS*K`SQ1D9c%ic9?&D+5A5P75BVUz-5ZcIXF5m8`81?oTrXetSXM4HId zvK_$?={IRA*tca2+i1_ww=MP1a20HKS-uAWRqMgA0rN$(PtN66RRrtTJ`O4U^J>S9 z3p+dIUB;Zj$_q&Rd{IAqOKkbr&*Gla)r_IP#lfj)eA&)>xX}tET%&6s3B}bXN!{Kr zk5tHsS@A)55{bEv6eGaiKDql_-Q=0~y_A0IKX1#D9-Tzea{p+Lvf`QCH&fRdMsLRx zQZ$!!C?gXxWz1EKCT!WP#Ft;?fC&7QFKwzR!8shhm&CQ`O-5K4@%^z0OY6>|tKQpQ zCzKQcCRkvfFPSKw^ZsBy!!gsI00z@*(rWf{4fJhq>i}eMRf*Q+dk^ZQrxePXHwtgB zPtSMKeu(1VpUZ7VmKtrYrDz*&u1Q5N(ImZ_ta4t8nPZBR`c<}kaXWS$DmysD^>{dD zA}`ZKGiPJ}rhurgGo&QIwx!uQf>6vemq0y0`&zx%Yer8iP5`OBjDQlz9Rfn2^TQutcp3q}KoH=ztWyX9q`aAs_d9vp zw=c7_$F50{XiNd!wvSDdP*2dK5y)!_>JGba$4ghq>S)sWN$ zubrEKGDYb zi~&KHVrLwdHx%F@F@F_eK^lyI6rGbRowOelAj1&0yvO!#o!Re^E|1&61pUO!@ zfhgn4>V_tgj;HOOTOYJ@kPOrJJ#3y;kkGurN5_Y+u(?F#9r5Q%h-L2rhIXN?{yYPk ztI=fWwBV~~+{1Ox>Ou6{x4u5aPBqA!C$M(+#mt=l4=46)S*9azSir)3_>R}_zMV#M z!8mjE(x;L+MeY1J{v>+|m%j{;m#>uIQ6)dQ@kUB2Mj8F5BPhAPI62}6k8Rr;b!t%@4PRj`nvwmNp^F=Ql5C)|l}Q_f znV3rb{g^L%_njDx@;-v*7vSp^0M(Q|>HxRFyf9)WvDXFO;aQ$&KP{ee{dId7H_<%T z<7wZ^oykAeCirVtUwhthh+JMlXOH8eZa>TubCSBEx&Axf_?eIj<5Yf#i=u(c{TKh< zKaysU)_XNSggW^RmHp*6eU{d<(J7YsN2$(nXfyo#0rTT~%kOwssJeQvmv4|yU{3#i zJ|nLbL6twJ@)hF;v&z>I9m##%xj_Ujee$)~-b*_PFTV7t@tby!qV}o~ zhBSRo;Oz>#kHliDrswWE$CdFq0R(U`Z8p&4Dbelkd! zlU$5Gws9A==KeZgQoXB~;@-q`aP2qI!5W-hZWi-rw74p2SW74V8kU?tm4B5Ae>$dz zq5~?c9+9x0o4-N~oc5SHhmB*14Ovq&K0W8Dka=s>k`e3%#o2lbn@oX6)qkVz9`r|R zE6-TDI7sgKgczI75HNvX6-nWq=xj}M>FEd`hJlTGnKhqw69=YcCTyI*G|};QiNA5+ z>59)<@D*690)~fu--}x2*SBneQHc_NE>^Ac@61q{R?CxKh@76wEfg2(8?A1I)a7P* z5)zG8Z%-NMr;aIQ8j&+RKfhDeGlEo8Gn2z~kV_8-#WF1Gw;u0C{g+<%j^5>WSK(Gl z{kA=4aWwPIXB#Z}pX-k|`@(~7zU(~BbG_!2ekr%J56(YfJfDJoWzV>)ST51}@G6|j z;_2-~LS1pY3%2=8CqNp(1X$}^v10zOIaOqaVvz+aQpm80F~$B#3IOVs-o5%yyWrtm z!@OMY?0x77xF(aLI!cueiksjAOJVNoxg+CEY1ic5x z6kmA#`;-7G=sf^aOSl=Fd4Ds-dmhXA<6v-E7S%f7;fUHx`^6sbr&&$Y9r)@(zatF7 z>Q!Mylmp(4`;z8My9Q3(>U)_UG4o3(*{ErBsKiXx5B3DJreFF5GWO|}ZHwTu`*K(N zOfbissW>{W!Zu9^1!n^5+ZFBR?=eQ0_C;_rjbHN-lA{-CWq`)kil+wb|Bx4_oH^?+ zOwDkB_ck83#4ZftwIQiK3YUGF|Clkj$uR%!yje24c-62}K+$kt*70|q+eMS-+oh4p zrVCR{v4Gbf_8t54-pXmAAbw zLGFh)>>Gw|4bW5m#4pd^OV(^@BJf^NoqT>wYRuP5kDmiOlZHAN-_z{b%Vf^V7CJ?S zEdZ*e)3*E#wV-tDdKCwWLa4blS0--p)0Yu5CEo;cI{PR%r8VPM?Pbdy-Y-H*K1-Vy z8!I&$+a3-qo9t$oHe0~7G8bhWn!-U_Yjex8xU!^q_5XDQQQG zW=hR)QU8D*kjXH0h5oml>5kr|GyKi@8*)_HCK&WAPAiTVei;fnDst1WuHLZys1=Ze=1o>E4cry9uqxZm~$o}u=|qL7!ZMzrq!3qjP| z(Ge7{bqbi{U0$)prHS*}9@VE@ix5KN-n$Wgmmg|JQEy_Dc0u^v;H9^GxSHsL?=4q| z-yPJ}c!fTY5`~6U&DO%ltmn-bavnmnU3r%+23k7(vJ3-I-UadiF2MJlKM5@VTaIli ziX9Ye@}jgWE5S0b#0z*$U;pdKmEG$*WL#=eGtiekAk(b9zYcn;-s-a{0X3Ya!t@uu zT7*wk_HUe07?-qjJ5g?V)bnmjseG@(U)O@#TkA6UWddn3bo3l1xFJfQ**++Qb4z1Y zhGuyq8^k3go<1RQk56dU(ojHZAGl)L(Y`1*3*t9}`!g?X9l~}k1_3(3Ug2}AdmT*X zk-B(;Z%Qfc?QAEuX>BjN7f^54m1X4=@OQy0^J*>9KZ=0%M<%?%zTVaC!771v$qwM6 zxbZc{$1kY9d|A%9)+$h7Hl5=9tNfvuYkFyN(ar4Yu{NW30~=E6YAIUf~bj> z3$BV;N<$U20Do0YCp@zP6qo%zAFYVjWZFHk3eymE@27tbMEFsuyLg~t&p-}qGIGJT zvCQ&#ru9&mybRRvVH0COwztVl%#UxXb>p}H!-*qBVcW}VYW(Ah>zfhp z@+|a(oO@mO|1kCQPB?1pKT&6%B*&~?gBR0*iNUzq-3)I%A7vcS}i3s6DCOv*|nG{6H$iw|;)X=Z?kF z#7q3Vtih zm3DakFXf3BHr|bWpGW=_DCw_J8_3$KyM<_C+}+DyoBTAg%_}r5)gr=O!Colez>-(Y zrFs}0bw1N8I3;i$z#483NZ4=b?YCgV2!HLm5yB~-wNMLuPR9<)VOOy(`v?vKi1;X_zi z6OV%U*cItk?+Of!YKab9+U9nvskQW8xp0IdO(E-X3DL+oWp?>GrFs2F!TS9QKjgjC z^&mJTCDdafaPu$@nvr;Q;T3Kpu#SY^8`Dn!&*(0@)^aVeQV`fF| zMG+|v^N72q9jocaoYPIzmQlI;)*%aS@A%` z(@1S^8jj`~Te0vW0_a0wDVDp`|18vU_#a z6IIQ`rnOG`=>f4F4f1aNEAb6(LUdWOMny&-21G_jPgc7(msM%3{M?+@P}#T9}x1aEJI{a`wU^E`YzI0li7?Luxoh%hVamHtzxJ7?J^S7Q9r za`Z#>X_+MVWxjes{@j6FZV??<{zM3BGpD^RH&Io`x%z0v{6Nzz{M@}5=FQmAsgS|( zaTJp?YjvQn-H`fh%burGswAPfN7@(cF(Nq^>eM`rHmut3&{Dl7>VzRd=}m^I1Hqm4 z^z+y30f!7PqiJhXw*rn*cA&3a=;EWz=-5~-NYRkXL8$b@&hDPs9aFZ*?HP>W2Xl#qItPa zd}pPsZC<8F)y^^vGzFUk)gc+c=UjqUSSwmN2aMDc!9q$9f8wWKKGFO09;wi02FNh_ zii(}sfo`mrne_hE2+JI>|5N(|3Xo-1i_dlw7U!0=qmJJwi6kHF5qx#UR~P4uZ~Ug0 z>qNj<4BOZH%$d#}lA8NeJ#-tJ4+{|2ISgeYQX@n33By;7b19YtWtrd1g|@fMtzfTe zqtlp{ev;OV-`*C5Iy?H!g82>4!8B-)>!V`7l<&i~`6eLhH=i^5!o|;B%gr^xP2?%L z82Zy;<+(+NOpUvA7_y}9f8KYN%2u%bL_;Tyjx=xz4?kQQ6s3eO%JC!@e97f4UweYuSVNV)K0fx!=E);!5J7y}>84uIIh7EU=#i)mTa75O<%wy{e#>=eX?QYUX5g>lR+E*`)d6;pbM&vPk^w2nwtc=opiwwXgGQ~a|G4@Z=RT~6Bg`v|bM%v!*~8790- z;2#@;-%{8>j?mrV*WRAg#oS9RG9{-a($1zC&VM>+%o7t(9qx9N+n2dw9AK3-uPP@7 zJl$9>&O83I{$t)->WS+&bh6Cn>w#k}8(sG^al(k|+UlM9f*6G?+xra277)l7jT*T( zZ%qQ2>qDKdZ{{Dpqjo(4)zG$9H@#nsGm`xtez<3j$PX&cY1$ppxIK7QuC!m>amL`g zMF@Ch{zrMACf4L#AM_+?_M+2${P2%v5V>(0t$gJ1T_3fk0~?rhz0N|^;0sEV#yo)< zvHyNJ5h6@qTDheqn=}eY1WsJ%*7*A1ljaA8RP$c<&N26%;JIUv@hIPDJ_;myPHU|r z;=};RRR{UrM}BBYu_U>ue2?qG^6aeC^zQnwYY7@aJ+bSR3FS?~Fpt?*Z2*s$bU$i7 zW;_~}oVsbTgCpgW?@PV=&^gmL}h%xN8M zI~aUpzV&(QQzh}%Z5wOgn89vzQ8fzgZQHBpGFxH~z3IxPsy!D+gK>oa+@;N=FsnycWY@z1zN>{UP^DdHz1EA-@%(z09h796MlNYxW|juvSA zAGY{^lTja%70!shERo)Qz3v^c{^v^7=UK&8dgbdAJM4Ah(nJ505G9AX9Nd*3^U`F| zA`6obrQ?K8jwYL-^0tpbc6o7sl_pUoD|L)$P~OUJ|B*4J$P>Fl|8~>g`WC2)pPsFY zrn>Oq=8ca@zmWCsjXwg&M$Dvj{$jk3x95#Zvy{(>CPktpPOiG0aMv=tn-tYlo((N8 zC%3R~yh%4Zyq*T)U?U89tR632&+}06A7jdQ6$#8LOGl82-B!_V{3Ln9Su4S4nvB7} zIN)Txrb{}uLP81-U&RJNjy>%U!qHn|Tds1^JdBAM5BaxCsa+BBV!!Nj)=?Iw zl^P`=>e$Unu15>`rwdC=KBRwX>8uNir6wZ%!rwv24k7NV;g{*WanSLz$rYodoR%xX zv3|Qxh(()Xz!y(FeuRVF!HY69|L%UpqtLPt_kxc^HuaOuO5MInT;l_0!M5jH>4s&E z=l^)-yeVcB=Wb{`Ek>JLpX^kG-gxf=KrI}t=I~fb+bB4k{2ejX^o2iXS-3dK3b$nP z+C?p=dUWEp>q?eJ-A(gk5JAFqfM|BgW7Knb(nHzXZqmH(oH&QwM9oDZ9;5H;_*5)# zg8<@TR-Y(M=ZEmGZ?K^IuYl>YnJc+YummwOX6Tz5ZDlA`T21D z_?N{;p*2mpB|pB`Rno>1GFxPvwPevfd+K+0Z1e;(ATuhuslA^m@Q`7@0q4STeAAgX z`ZiiZE`K1ZrkX=Vi4!Z>vgALK6QdT*cQ02SzeTyuMlTZqVNa1)S34a`#pg-3 z?M`~9NMQ$gpR77N6!)1I(fD^i3|R7MR?(<9F5WuW1gi+nDH`^Nqc8D(FYtF=Ycr}< z8H@ddhoTvOpCk8Vt+Q|F2A{znU)IKz&ICo5(Z<>MYk1D{;|wQwdeh6sXAA9f%yX;5 z?+@HR$z@Y=W3W$iqU2)FP1`Xihye@_9cH?uB~?miIQiI-L6clbeqa8t z=pMttfXBf!d^DqLigJy&O^~$lZht(iH(U9ZJZ|f)hVNsL3&4>ISA*67U9fZ~EI#bnt*LDJD&9yb3_AzTI}!MDCDGH2mkTT<|@ zr&RLC;1TQ2M#f~qv=>C&lFf_%el%+-Y3Npp3s&4~MR&ZIJU;a)vsLz=7lsE0e>4ed@i7 z``dtbrgzNZ`NE}eH2tUP=r|-yBD|J*_*g4ke;VMJQ_m)T3A_SOv02keehz^!q&4_mt@#~UrbD@N>BRR0zvStW|R>_`yx#j87v+Xc8{%p|?9s>d*h?z4187S@dFoW*r26$_8 zkxD^Xx9oe|S|bRK8xX)>rV83p|Ap9QIEGtdHW`4YQOWXV0QK$U-_Qnxa0z`N+fJnK zc5_)h^&l2OThFYYig+sczNznp_UccuwMqN5vTk|o@1|0jNF;PGGj#S>QSUpad^_7b zT00WH=~T71T#C&D`s2O2Se>eZHY_JHxRjO|EazjBAK*~dzghKqu8+YTJ6JVZ-@(1m zVCzpF?Y;}r6A!;@VntoSZXN>m6qOopF)Le;chz)vv}?>f^IvbK`I-y2&gw-)!1@Id zrJGP|_$7ObkbRJmuGIYG)rHMf9bs=b?fpf*j3V%M`R+!iUSF?+C?WHG8~gc{efJZ2 zfjIWlB5Qpdqt;_UT5JW`i0b`Ut4aeBmg~t{07t7BIO0XuMDS?tEnoWUHYC#5n++gD z&mdfFLf?bY+slc<%6_M*8kG1<4mTF*GpTSy~!04KHefz>1^KW*+ zEA~~{mAm@hbK_(OuT}KPwd;EpqhATZ+=HHF^2tSgcBA7f?upC8#^K_t zd-v44!}jv&S<5SB;w$!e0@hGfG2%c#jz=7}C^p&=+Svuda_T;jv{@VINmEAXmeOrQ z^A#{tVJi(GAdNZO(68j;8+BW=w47!nd~dub%u*aqLS=h?FfR|F2>*%H|mF-*qVYfh3 z+*pMBgaO{j#e~(Q_EM^~zH*o{eXQr5F6*e&4DKx)1Kl1Tl_PBv?!itC2ZQ6o^0iC$ z5`ol(n9$Y!YpyG(+S9GS^;C>~n`yaAV8#X-_tmisy#(KKa>eg>dmlo)^GWx?5_O?L z6Q>0n{)Z^eJqt#wjA`I{cEWwKLo(tqeMjX>HtYruaAr*dpNDf7@H&S|!ajAc1Qpg5 z0NbG%#1#v-vc5HvYoax=zuC=uh1*QL0sGUm2PFKV(CgakgKsN~f+Pt_tZVnEmp^lt z++M3G3Nkl|t6F}q<&7MF=eZFx5`$%jXGYmp=--tHzGa>fpzzciUQ~UgHnCq3 z=aP{e*8FY2^!)tO$2b3*i2s}BCH)^g-NIaYw(BYXg3Ow~AwE|aMyg-aH(+}3(8c%e z3q4b?*V4|yAu0JlC`We=xFZyNCct(2fzG7zVwtX2CQ!*=id98^5c(&Po7+`b`loA= zpE3j7YEUH1K7IS(>%(d@;W`)4yjuExr?-u2Pgox?|6-(yp^G7TB&%}EHX`m6h+S^Igf28w@4 zK)YFQN;|XqN2kt&4FCHFz$9NEF^~H^hZc2;PN{vBy{?CQW~aCiCUu5cKI-6kqbf!8 z;Q>JCvTmrZL%SKC#EWUonJOktOeG{meNFu2nScoTm+YLp>_3o6F227}Bz3Ps>a9Id z%8@OKu#7Wumm<1gJe}O;l3Y6rj|dPF%y+edeF5;|553H>4yV+h{-MWq>%?VaNa}q# z{v**H=pKOblpibw@lc0=NRgT)AC}?dj(+3_5PX0T{yF-b1n7pi@1xcHpF~DJ0P9?P zh;8lvr|jdeKVMD$6kc_?X$1QWau$YufT`r`LgSv!@~g~V`6TT%Q9C}&5i|UiB^%bQ zEx@?gT9QBmT zZ}Grx_PV`(q;lmJik?}de#>YkMY?bo-|BU{=czBJ?33RPiqN7DX_cGW}l zA}*@&RV|lAc*Y1IQ`MAhl@ploX$t)|4#QYlH+fvXgK0|+p7$mrz&GiWp@tna-2SGl z6&^@@QH^dbski3}9ON4PcmUpPO*PT-&Qd0h?=|0ULquSVwKtxP#&}jnKCUxb7yF!g zi3Nkbjo&m5eyhqG2|%HeJdf+iBX?U1AZ6-RfOo?lJ4e-PdmhCo>PV{4V@gs-ttm?Y#zBuoxo9WaNB+2 z7XV@KL$H^}8*fo>&>8F0W5g%g9<5zC=IVUD0CZt?3vk8N(|3VjCNSJw$AkB^q+e)0 zNgF1&NNHw6)=xaA-Xyg`Oe09w#IlZql27@i4sY4-HqFEc3vKT7G-L+s1_WI*?D0${nn1q@DCvDNz^gG{DP zK6$>QH3o*&TTJt=Zvy;^;NQp~TixO#&*b;r{`m?x&RsuQW|uTn)?UKt z;^FvXp&8rA309fN!t2oIU6jHNZZ6Ifnfi=$sp#OrsJQ4TEtH>n)#$H67*-{^m7DZa zfD_wPl-Nx{_r$b~-WlqDAm$+z7$jf)RVzbHgjimC=Oy7K#uru8iXh`zi)=yVp4E4Z57R_9%aUue=TE zK5+K@zXR3LGwW$4?pDxlBynoOt&>9#H8A2qQOoDYX4M4{erVVe#lcrheImN8UVOG> z7AbqQA)(8HR^(pTMq1vDvAe!gg|OU7_kZoSzeSbk#Vbf1^BCyHg+r7; zx$w&Dy3`wX!~9>Rwv8DUOT0a{n2C0dZ2&y7_^Pmb4o8icK5ON8ZSo6;Vi9hllHM)ef2s{56D;FAYeIzPtW>*Dk_I{8pzQe5xe57_ ztIURI_veryQ`|$fHUG`bk|-)CUZ>kGoEVaCaKpxFC2=ZZ_*ngbtd3P;OzCWhO$pV9a73>)Xe$BbIj>@yeGb^WP}7BP+95V?*J=>Cg`9MiK)+g9(~-$NnPj^^k`23X%o=szb4fPC4f9O33Dwl|_8d;Rq}r^Ad!ZGu z;gr(a_#RiAwVWX5IBf}n2K=5LKBHC%jKI_;ZqL=dOu8vGLBEY|PA}xLe71WY7f)jv zZ#+SIZH);Nxq0^15+%!Pe1==m`^`U%Le}Z=w9MrH@R7yqsE~{=e|h`CA4a8%HL`>G zU|zGjZ!!+p+NpNq^IH9^+F^RR0blJTnHM)-cy$}M-~9oso~~2i)m0F2{{M|8yfc30 zZ#EDJ2r_(i`b7Sbn<%?M%hy6cgc{ZxF?}xJsYk{H_u3$S0ehoFSoxdP!w=VM*fa9~ zmE}B!3d=d%CA+q~j`k6DPLn`Q$-f1hlW8EXAW!}`Y@e#X>K!B6NSL2Zu(WJ`iTU&4 znf%$uk^3%A)k4&{WF5tcq$jftmVfx$-E4ayB-XLwy(0=D9RyUmC?KIo6Oi7UH0fX{ zA_+A}6Ok?*r1uWeTOuGJz1IXn?;-RS%8loF&e{7sW1Ktoc*p(YeaCx8@<#@w{K}eh ztu^O&e&%`F>Eebq+$hnOz3YTWzrIO5_TF!l!&FyisMYFoMo4Efr#!+HPWB?_vBmbX zx@0wKQ@^g%X?LqQyVPOIs(5J4&k6dSnzN?Z0}ia?tEu%YS1dG1a_*?K)orem1`yeO z*LaX~RuaC2ofV3JV<5D>suDbR!|fTXd0PPgoD~ibP%f!uc&hqr1~{MU3}A9ctppzI z&2y-ZjsXgIdvzmqKwoAH!Vf6?rpj?L;9dsVSz8vt6-2V}1BDX$FrS{U9OcfO{;+5L zU+e<2H#Hbu&r(EuN9B71(Z|5%UE5D&joDhcsd73#I9R1- zr#ju^k1sdxWX|)OK!eEt{5bYERyeT$%T&!Lw1mhafknbU6X%&Tn1bqgi-SNLB@i~qHp z?+@&M#pC@8v8YV*uzUUa^t#*$QsRHb#%AW?)});ZcAq}0#)f(W9KO3*y4ls2=R2W{ zY|xWEo--v-7naVF@Mq1Jj%fRz_&fKZt@n3 z2-KS-+@14O7y6(9(~`Q8aC}%9^fCxL591P=u|@}Ae~MU9^xvoH*J;0}Y04Yuhh)PR zEBkaAZiv9Pu5m|cZI?qQ#UKA>_`h!7N?VlJ-h8pP+jE*Mg7oV4Ih~_$>N}jRbx3J) zcS!!t7Br#Q`5MASg^PQjY`(u_*8jUE+W$oK+HR~#H|$dK<1mYc5|IX4e>9xK@UQ!l z!k52MiAjccV15|i8DhipviNzdaOFWZnCUP!^Dj5&K;dP;=W8tRa`h7LwcV6$f8%@0YYdg!|8i*m z`K1mS_J-@SXDD&}wZQ&Az2E;e7Hr%+YQszPxA46h&k3>A>j)pt_`jb1e*|nl`OUw5 zom%^l^iSjYcWmWfKl*Zl1xrnSevqsq^Pk`Jr(m=2|6cUJI>LXApZ`x^^g`@rHg7=i z4*tJR=Rb|yf4rdzR~Xx1U>eie*f^TemvHZHx25Ez2Zt211r`3^8UD8jbr^8sU!9eU z{tA8k_db9{`1}UTdj`w2{BIBKUj~x7gk8gwi+akd{11fYfA9}1pRnG6^sdui=Yaog z4=+1eu}0&1@FnSQ8O_I+q*#-@{|NCn>DymqvF>4M<(^>hpN98WoBbbu^}qj5w7j6^ z(%iOf|-|>R}uVwy$HGe$fz5g}Olep}xtyCXK_7lCd-0^Q* zpZ_pyG;aL^G#v>s98l`b>L*dO2w>2(R(OY;xp|CitX-}qb(Y@e63 z0@7F5t*K(bzprxt(;3TQS7q|J2`dbXfKZk5CNm8Ub3G&byn}m1Y+dTqzGtYWw^Ggha*|SB`QuX=!9P9mz1ZK2 z>w~fYyD(Q@XEH6e5bkDw$o>{z18O~*s+!^NQiZ&YK1mFvwKk1_o7oE7gI=xW|#8RD3 z;8R3@qQ7f~G(5e3*Xg7>w_48o-LM}bZ=ANamOs_0&1YU*gtlQ>&eRgYJYRbie-R}L z;(kdr^D;Y{|22-~fz4Qi;0tEaLVqK|lF?91TWPb?^D!(=l^F23ae15C@VOIWc)Zw< zy-Rnk%;oYaFg&fsMg^N^>d#Mg+F62PDZ~nCaT^$H^56~Tx~Cu`nIC#KTH3Vr#3oi% zSa={6^)6+n#d5IesCH^SSzR*Od9tXkZgafKH+1RF@f*Ym%L6qf811Pq*}9`=O!Dh*Gaw39^z=2Scq=!fIi%Zv}2yd0`kMM4knAh`L_Zz;3hd&N{h=D#RnIUzn) z7#Vbv0vc+pT)p3Qqr@0diJUAmPI|NY*!_nfc?Yb>sP>%=2AJB}@Z_f6DaECk&KE_N zZy%f1e{88Rjbti`Q_9u1q9dyh9r;;YQ9UTO-SA$&0HkUxRYvXJe){yG#tZ}<&nm|B zr*G5~{k^rdme!?#J|`Uie@3Kd`SCZ1^j9(06-kK`mP|a)1q93&TxhkhmE6faV6|Ve zh4K{H4@qq#B;X8F|HqN^_d_KIjOs*w*S^{ParS?VNPjal@A>0O=gRdGPI*j=l9#2U zdZ4Ejqj@)!O-u0s+gB?q8&098^m=h@F|ENGH6$GNb_LOe9@pz`>4T`tT{raA9MVgi zBID3YVyG47eS+CJl#s=jnlL7jn8Od)dQ)#$frNPe?BjvuQ7b-dJwd_w<(1>c$miTn z(?;_SbFwOhTdEsNO->`<7%bozes{6a0Th;u@W1rM!Lv9|irEDEMq2}Xq}M77$->}G z;LU|({=l84V{!&j7R#=P#C__5CyUoZPQWXA9K1Mx?!New@$T;OIZW@$81XgO74*1Odc7y;1Jnq+?D z%z+h>No%qA$ZHh%s3cAv-0*!CZFPJ|F5fbbrAQRH!A)kjC(Cvd>wrc<ny*bxOX)1J+)nJw(qrB$vP$FNe^|`SwgE-H%enovssn& z3+;hs{C0V~sRY1w>n$czfYIJ7WXuDts0`>z#8v$G-_J*9YFbR9U-6JHryCvllP^X; z_=cYquO71D4oZJ^oOEJg_WQ1+20-Z;Hal3fm$U>131BOXiDQ=hk7i!Se&+n27Vy zCSdnxQLQ~jf7uK|jjXq1ZMbOE@s{EZbdH_u?6N(Zl5{uA3$*cY2Jjr_r~|T*S`ih(C*FH zragug$BI|F5dluJ_!<_(c^iK2QXAtETD;xr{BTUUoe>ONY6H z*4WLcyzUlqS)t5&4;43KOXU9zBp~U{C*nMdQdxi4!o<@Y>zCj zMff>uANF`^?V&OWeU$X1_=jWfI_N>6b?@MHCQ<4M{K<^H1!=G=KI;h_ts1)LPK8g{ z6>|CY$gN+H%bp}XX(onJp>`!OJwJQ+x5u*92GeO=jf@!#7`UTA8&8xI%j52F@Wl`f zMSXy#Cdwu&po9fg%m}#hT0~vnsT5Eoc^o9)_wTT1NxXk+mh;vfmj8PC?kF2FUUeucEhAY%k54?QFO3KOcezm^{ zPkaP%Fsd%MvAt?qimVKC8cLU1iPKh#oNyTJ=W73vp4%o^>wUT#pTL{amlJhaPa^&( z6a3uwrCyn7NtEwxf=9$qp*zjhJy!vipImwN0gj2E#E7A)aJ8q~5uRHfW*nJX%{pJJ z!Xy<7lfzrSy>2^Q&EwvR#nVX4QajWNo#ZlgvRSJ?VQlIt5wfx8M?2`4p&Yyf*yaQU zMOO3l^PF#_jix`FRP;^A+`~Bd+qpZ{oZ+s<*ny&+U0LNOO3XY6`{Suz+vycWU3_p@ z!sZxLLdBahuvUnn1~spb=;C5Y1uVvONcC%C!N9QLA6%I#UKTZ`<21_Y!(oBDcAOiq zm+_A@RhXRSWXiRvLM*pyNIZ z9`b=|p|?Pb!owrvyf zSe!l)nwVhAM|IZA-J?NTP%({?+i&EN>oP@5a%Q9bOM*CIqN*8M{zy^R^;&(dO>TJ0 zQoC~zd-PnlcM-d6qS8NIjUtO9b>;#s$Ys-)sLZQu~;!SQ6qqC=r? zxIn?tt=d*Cekt*8gacQKzg)1u2m(7a=m$_c*%^ely}zz?J(#fCZISoNwmj~|(?rK! zpTN;8v%MmlY5mWgo22LbT3@o~EC4JAJ%MZV2Z|UzquT(r%EO0>7F;^X@8j`?p2Tza zR$Gony{7n1{fd7=W zf>x^@(AoozefEjsFkuHW&A_leTJ1tzt3Ag_;=-ZZ1=U{JhAr9Y%`69wY-wUe`w_D( zmL6f;){FBCNNLwM^ z#K(2gFfdZ((G#FJ*Tp|k>Fju5*NB}*A z8j-XzoeBB~)U^NIF!BkN}dNwE&xvYXCN9|)7 zf6^6!?h!%m3iHX$7ZzckE zxJxz|k44^?8NQMd5`FaLqR=sxW=Q-h&RjFVSUTyrhbz_Hbgten<6B;22Xw3VyFY;9 zGAIQEabLHthc&QY=0r)wBXz~v--Ry8;}zh&48G^%JX<4Q2fTTW+)U;g3f0QhkRB=_ z*I*LCQkOu@*4y3Bq$+$x^ChG5sB~bMZz_1Reu`^_x&tsU2K1@G6 zKg?iHEbNq-W~GjjX?4LC+c=aAMbPubbDJ!0ld9=A7|$6e=SL2O3}XfXTL7EDtv8k% zQBv0$GC|P*_MWh-~jvqkoi30>YD7%)OHnK65w(&?mktYUYPuf80w}V~NNAB=f*Y(AOxq zin>q8aPwZDRSBb{P!V?kHh%!oU+uQ3zvriGTc3oFEFgdTp*w@*78iodw{tZSKmFJ8J@Rip)m{T$ zGc}%JI(p#7rO2~NR5ak71zgN>2%C{Eu&H0Q&G5^@gL&%s(V+OYDk^(dfHmH>8NT^< zd9_3ZoZ-9Ku;X2C<}LR8)14W1T2l}IUB4Vugnm}~4;h9y&+(zd^_DH}BVkA;($zPj zsj1Y=Wbw6oTDx5ihp9DH3exwqwbU3UeXVuQmT(OcUljnvB^q3K8dgOGJ$sH~>SEm1 z<0o#3Bv~#iDQ}eKAM7L&Q$N}u<$C&PB{oWOJ{R+}5EFLy@y56(9jU^(c~sF?*0Wb> zUtAA}LpXwbgFVYlbbrMl1urI$XY42sm_a4Pcxr%*cFFq^NS2TUA-E`B8M~mvi&*O_tUSd86yV46oE6zci2sOv$ zi+E!F_%U?Vie@uZ0Ul#ktD9KNJ{}{oq&YGx2UmTA)+&C(!FaPau&bqCp4;6iY&kDj z5m5DjW_uN|{?3FJ@0d4S;!qBTd+xSAnAYo*^%=rF3PM7Upr26rbCF3W6LaNRQOkOv zC)*us1of~pM&%bZ%e2q@aKX!uWjSvO99lwLG-LkdA79?b*+82ZmVdaPLuIp6OS`D} z1){N+_OSAeYS?XOI*Kd^zs?oecC& z0%3{Y1_snF(=q^qdln1QecGw!XS~)VsPMbUU3}7o=tN98i!i4G{_nVxN=!;V9c|6bxUu?YL_btPR91g?OEis)9ytqB zDv@k$va|T`tnHN8{pZ1g1$(XeuW1o=l@;*zLDHKX+$XPQNR0bMF!soq`}vYe4=GYE zaAlmwpZi{rjW*k$o_+kd-A1q*qeme;1%3>-h&X|+l^u^3?)%Ofz~<+2+feQ~hWq1D zUAGp2v!!`7*(@;g7!q8fSLdr#IZ^9}%Vu%5uOzq)6s<}SjW=9T;+L6XUQJo(P^z9^ zFuS3YLPj&P37K0e4$@hS?ay&uC3`H`U3#PKd4SBmZ&v5(*>-bYwFa74idM1d#g`s^ zvHm_moNj4H1zKJmB$#SE@idX@!{C*ji#`iM}hcA{wZ%I>3 zbbq9rE7Rb+IN2@6%W#qary1(;GBvJ4C$Yhub*fjainoxNCv0@qIAQ+yM`*f5@Nksf zrb919B(mG}@Rzf}p4DB@3|M<=(Y)=J4w5V6m)nr~XgirDzWNbB({!%4yg>5GB~5Jp z+QhdWKm(W|4$~e^H!=F&aP^`C0HTzVcpL8auukw}CzHEV(PJ#OU}J`}sYB2PK~n7G ztwu7kh=0kacl5R(!yZwa0RmV-yJsiqyg?qqel4cF0mdv>84preO+~4)^rO`LWzsWS z$g;e!UGSdrm?}Wg)LkDgB2GNyCd>Bo7UOwM0LF!OXY1bK)6#a~ z227DsU)7M>@NJ}S-@3!$D++Ei$Ej9X^kuuq&dau+BNYzEg18y4EIO-=&lsPH5{(9? zCNXsUcyMYbZBeOTYCv`FfDySUM=nPV<>C7=*1n{_fCQl5%7;B-9W5JewUufMx&PjUI|!NGw^TQmJ&Pa`GsYtu2DU<5n_g#J2iCr1q~7yX z?`1cZLaq_=x}t?5?Iln(+Q9W=?`+biq8k_H>Z3d9pmhl*USIS(LjacIwP zbFbC=HqKFfkw9GB@G4I^iez8|JVS7P~9U zwhE2-0MLj_qiIVu^&emcMzQy2nWXIs?ENkzO4*%cB*K;SGl*&;sV{}vv>^0+8JEc( zy^4Tlh)caA{56T+{#>Db#o&kDCs65)FJts9F0-xPQNjt_`f9!+0w7`;-(pZs68n;r zHsEz-A5MF((Dj_IZ*7YBQvh%F1eXhcjJO@AOeGvYmQ{3%(p{-70B>c=Ovj{9Ek=Mk zPWm*OoxzkCP+I;-TN7IZktTeM*NSbA@ou=Dg?xHoWM@7ESH_sf;78TelD z#w+XbTvb2>_Z#XxsXdiDX>C{;6d1z+MZm19v-R{Imx~Zab~LH;KUn}y6{l8t`lYdD zoILqL^s0Li!MEaz^Z7?Tznf7M`N4T-Lles2+a?XYTV1}rHds~nMKeX`q*uu z0(Id*Io}k(uc~R=jwWk5nw0YwAF*u)&6~^@k}TN{Qy*7D0-SgF%8`}H^Pa?~nm6Wb z4~`=ZSqPvNYN^o{cjn54Gzp|P!WcgkQV$RPAQxQ$=dfIMvE8nJt(WXFwdv%CfaT-W+vWG|Fb>(V|LHYoCvb`UX}6`Gt< zs>xW`u9ont|DxpDI9~=&7JcjmIb|dSI-gjh&#HpGdb#4g;rPv-QHGVLVmTpqXq#s#hAjRQfzY!2!tu zwAisn=+V6y<;80k>hg0rL?Zu-{$9lO9!vYZyyoD^ps}c1iSd=nL`;6oMb{A0$!?ZvsVij7qFtG<~5%+1c$kN4KOVRiB zyw@&h%Tzzm(N_N&;wi5`7OX2XuBf8s9M>19Uv6eT8nj80XJ$CAn8Lw7nxLQetdNZF zZ!k7UyHl^9<`LxGl~wtOWZ7MS;*Jld*3qEFN{8+0!s8%tUu6EhT|L}104>)Sy$Q&7 zXt7-y;`-Ak<5E4+*nmZ(Sp*|((uW&T?*$`^D1DZb>T6_=Pgf%u?a%t&-N3FPwkcNy zoC-p!GuKNJ)yqy8k%vA&}(nEW22$Uoty27ny31@BR1pafcE}f3GdPJW@z0c z0Je~4Pt67UmLD1A*56XS*!qm)&D6y?$z65Kd7>GkM9?xe?_(zswAhC-3z@`hr8rse&u7Ll5P3su z=A`aj2vjKZ^GyC?Z_TZvTB2KE@!<0LrQ}=-M_F6i0OP6D&D+!*0->#q7YgV+dm*1= zK3-p-(Q(T0sN{uJtfkG=u7rJ*`_|&LKD9t(`o0t$X}B>&-MzG#8lU6-sKv8$>y5n; zx^^psXQnOyNP&pgC(tX*ShSOBYW?`dx~Tyh zDx#4@=q#0o<_e#5;+myGQs|jydSRccc0I?o;$ z8f9hY)tVM@a~JR$eHSOQ{$o)qOl6dY$7rpx5!>62eJ7vQD2c;e5g_wEkKX`U6VEgR zcjmblgyF||WY?+vsb=C(a-yK$4>FiIPHU&Kes0z~79U=?ctU=8hRae^UH0(S{-n9fe`;L7lA8*E|?{5!c_B+H?#Wijcc-&P}n#D7LxCBARTiZ@5vXUa8 z4$0UUQ$nfsk5c0i7FVwraR5_qz}lHU1{Ya-Yd=l2U%V_@#S!V`H3xyE&UdAOB0L}>!(d)yn+9_&)&dwep?PLv!rB4^j z$554CtqRKe+K}=`#*b0?vt*;g(EDJ@s0SibQ~NSVd{+fSfpCokJy`8v(b&i?UGtj# zl=jAQ8?!r^zoi-xe5^}7Wh+i|dT9``~$_S$%RoPtKgQ_pR%*!zX4S zQH=_OJ2>pww9syI#>i}z99#?_Uf9g!voS(?6*pTyxA{|vXhtaL`6K5X@|TkG-ejD6 z9V_0{Qz)}5ec@ugirBivxl?zkp$+w^W?Sj9Y@tRMS4k=Hr$`nzYaZ zEYeSsdgs;&Et#*Lg?CV$kc7R7Icfj-*d74In1Gu%=G(yLL{9r8|KqG?xaa z@B6yBO9b!Qb|5k^Cj}Cib}L=m1eaNnGwUstWjzhJ{n7BGWq1GxIUR`1r^r;BafQA+*~=k@qQhwA{6tK;lZk!c{7#Hnh&OM&9N+YsQyC{wz@3 zYE7PxPm>QgAJC}=ClnFT6L90(U@i=nqdI9jyiSM`-S=?V+p7VZK~|HbuOh|Bmnp|i zj{O{xaww#BKo5vzVK+Xa^)`=6J-c#TXVFD1a;mQXJ3{&Fwh@TsQ}%)xLrzX z=JmXtbjLC}949E$yh0|BNK?Bqq0Hc(1*e_Mom^XdMwcna0Q|)t4~zC`)qi%7XP~;a zo%_Rv17DCvvQYY)kTRh{O_n2PJ2%g;E%?qOZHNDun+*KOv^x#jj|{~pHC>d#Z*%em zT<$-0Y2LgBs*ko{X*gn}B8U&EIiym znJ$s;2l=YB$eiK2B}I4U*6&jcSygI)qr?{PzgRmbJdH3*k`rq0!?qHWXb(@zhiqCP z_PtY-=mGheTge zP1NR2!ZAs>?YK=!w(j3NgOEuEaycZoF1)cl!?J7GPj3@s&)J@Nlfv;m58}fxu6FNgic??MA}zCg}G(^SXz&yB~bbKyphY- zm58&=G(%!a=+b43KPr>=k(Rix;GJGUPL0;~KO}+}Wn`NfxP)%9FB{+wzy>Pflr%gw zF5Al@xE7zPd4HZh-9F}}+q71%dppv-Ja-XvA#|+stgW8vk>yArBf*&?O(Lt`jMqss zqa=S@7LevRdok33nGkd42GeF4p)=XklBv!<7D zDH{04FLILkoLbq?I(Lvq<^3_t7H5OmXvzgnaZS*X1CejPuX<0B1bbMn}d;7Wl=kqrGgOYJgRJqZujV$y69gLKc3yDRS}c@g<6d? zjQ2IEO_eydSpV(^4sPVy4B(7;C$V>YK#%Ot6{Qn zYL%e-*s$DO?58f38ZJ;1eO-q_q#Hu`r1E7up;8O(RM(qfYQW?|i2jUVJ86CN!$(e( zH60R!=5qk*9haogS4r+9P@o$GJ*{gqU1|bjy`73RPuK%x#panIK(V}JIk4A%>sPdR>{Lgj}NfD{^!`<+*Ws3%$SyIQ7%Mt-O8 zRSn*H;uAqpViP5UPYcpBuTJ1EY(w}J_ac0PP0k1boHBu1ULp$P+?<25>5<~R?ZPDf z%%D4%t47+S)?psfHm5rryAY^jT&S;gG@BzA z)oUNPv-i)C@Lu8?xZIA#OfwRR3^UhtID~ZY-mM=T#>pI;LN`7p&^O>nr=fMDzb|?E; zmy*7MM zeGQC9Dsp>_bx;NRu;3WTG6c3a17uTs-qFK(!n~#Y%_<1=t5QMI4P!||@5TH2`;{<` z*Xb^}gRnrkQGK}*1rLh~z(r!MF8k_2?m(7pb=Zk(?)UzN@_y=N)CaXyk0CX6D3|SZ zB12D%=f{6M7F_yoj|3rSlr|!cvKEY`6<@u|Tfj&~ioTjA9km%Uq->)NG!bI`>~G4m zBe)+Ns;a_p|8%Z>F|#c+V}6Y{1F$Pkok_>aad}GZLqH@pFaq0Up9hk!8N52J0!qg! z)}>Fyz`A_B72;Z##SKZrr6B6wU%LpQ$HfL9F{ik1m1_!c)a4Suk%&eS?I-e^O0Oa(|rBm z^UOcc%onGb6SX`L=+dDQ#KHhNn?qB|v5`wR+s%PmAndAwo*|(={H&t5%#<4qs96S< zOyOom!5OLJ)nD9n2+%&9Kk7cKeRSr%6Dp@zy_{}Ow847}z2kMK;I@+p0;Vc15V|ob z1Vd3vxGa4dwv|)C;B9t;kDQ&Pc(Ejhv!Xr?W}dNcwwQ|IM4V77AmMAOmvaHmgh$~L6`dw^^bL19 zl4>istF~2kM(a@xNEG9nLt0+4XO25vMw|%uCf9DXeL=}o0KPUPfD=8y#$1sTw8%gE z5bNJzJ6^qW{O%z$P?|1NAM34$gbOhth?5=UD?1{KfU7z35M;RlOYr27p)g=k{*(WS@lC;?ljF^(~Jhz6@ zWtAIGJ01ld*zp{AaDnmyR+bH{Q+!8cqsHP z_?wk^JMF_W^5SjMu}sLrOL+;qSxGQVa{e+}@kI?>$#@TV7*$Wo3e86+87@nJXAe&O z+>)%MY5H&f^RemqL;WmmE@)nHNQV#_70|Kv_OKt4zcPA28FRLnC_$Bfkm0HI6RmT7 zwibe9`zCcbj-Y}GVY>Zq6UeDv_F|zOZ2u)TBK;7sw^m%AyFbT!gkJBCh~9o;g4&qj z$ExA(au_7sjG;Xqr_w955*&`iyEKNrjYrJ2#0v|8Po6#Ya<#ac&(fyQ6i_x z%!|6d-seiqwL#%u_?``O<*liTY@EjAWW*NWlTD%Nr}z_N)z$c&w{r0wzbOsB#FFUW z>*rh?$VCm#h9hBO^>ZBP)|;$1uAWNClE7@?1X3E1=)uYhO*PR)To}veLV=RdXJu0F z6r~y)ZOE>e;^wa-t;dr^H#s)1<^0O3%m+)f8eF8ihs%!X`?|nV(O)_EW2OpV8$Mvz zJFKK&jsMM)45wM1EmOexZ@~vOMvZBTU0_#(&EGxy zY_mZHJr%Isesib(U##zaYq!~;W~NQqkFr8)8eqxdCwAL}5|?`=jyp9`A{6ac0$BAz zQ3;M!U&4^P??*3IL*mWf+Q5UYaKnQ+Fw)VsFX1rxys_wg3Xl*}%ZPIjJN(A$KUIO1FO48rKdHJC4# zEK=vr;pq4B$zLw0dvEv3f7h8lnyq(+-R0F(^Y$K^1HVoj+(b#&xIFtGx>5XPzniQ2 z-vXaRI&iZyY0jHO*qYb*WzLgTrJ^&Tuo2+0CtZbbDMzSXg50 z73NeKF3g+jUV_2qEdaA42gPWj=VwLpTI&fZ7mPPMUJfVyMOv3%vm?_nolZiD{K>rQ z8Q^l^w_CX~c>Hm}hc*pm@|TcR=t{bX5lXE6I}HF2WaEg*hwzLPM3cPg~HlwZLh;e9?U&aZ~2ZpYaS&y(J&&L=MeuoYp0F!!eQ&~Q{Z)SLB>~lN1el3%z6f9<*RaE z$>Qr2@v|QiqtH3o*&VC`t*>%1=NJ=$E>FN_bl11>XJ3xfYGgRi7Si4Cz%G7{U^-Wz zT7x&y8R25sl#Et>|EgA&&x<0xV=*&7DR2YZ2DJWIRkCGss=>Q(+?*i_SP4BQT&Ng-3FeSQ=^raorL8{Q~zB9npR zOCl9435rqfb*R4YY=N-WnFjq7=roX;hxoFcv(CTYpoLk*u?mp(-h2{F$L>>RAbV+l zihD_KHCH%8XNH6+Bv>fwAwRC;IFSDpkl;9u;W>MyhMogwil4Sy@zzW?K$6+dgmBbI zhp{>1jTgfqENZ9+aadvst87+VE`3l}7!5zIN&tuK9+0Hk%-j9z8<2e{>2(Ul3Cp3& z9UrbR|HYNl8ZP^H*eJ&krHt(H&b-oZ{VybB_*Q#Jta>lEWBf~m3KND13be4p=4w~` zF{51WHI_^_(L-y4#t&&t3|`Xkd#I)=;*$UvnC34aTZpVO1iU45k|HH~-~ET|ZE8mA zisE7}#7@r_tX?d!gnMF`+7PSuiRO%;x2YuaU%Ji9Oy$l6#Barh4*XXEEj zew7{Erd7GJ7e7Su>3%kPl9o4)A_!%{C>rRgz|E*lt?yGRlGt&mUWWUOslDx1pY<&;?wt5FNv&ikCd8_tOfrp%iKow+-uAnb(3vhws z!o`#qS&=y4pvj0!F_tfh-o(W{CLqsUx3)BoHwt$hYTb;zW)G3#cwT(N#k0pZg$#_ zmMNHBt)l7^?6Pnn)5q4|agm`l?iL-(TxXFS&c+_!^evI+luN5Y%Tb9@QXbqvJs;Sg zJ510LCI$9-*>%TBH}iCHpACk;_bBWOV=ma4RcL2>!_(D?BNcuVUaTX}t9QNSh!rW` zqr-|74_|apU#+lqyz1!{nG*GVXh|5B)l@8RS3eHz@gY9O>g214p_X*pECO5QUac-Q znjK!Fn9y$bvx^gEo~`lQJ@MQHJ{J})aA6yS$EK8dVI3`2%SjWRIIhI()$*wdR>mR+ zEWSv&Pq6YUrekvS*rO!?ZLozQRF}Q3gv;^#aD+F>2dTQtM1I&oX!{)+%Q+TU1g$=j zvKdIq?tZz!Uu@Df`yokCzv`<#F^%-JrIG(#rvXOXz)Jq4+&`Ke>U zi8i#{pnya^eM9x$CcMcmVb1LBNk^XPR-9Y9bH)xmx~qM~@%zCmEr$mHoFi^-9cuT2 zf@hAiicON~(~-`E$4<=>Cexx;<{3Xi2?&)~3j@Ty;Uat9nvag4?bS!81DumYVq^2l z8+1&4Jp)WbP0M9;C4RUURAbZKyXNv@Aa;?4*7Bq6IkF6&i2?UrN_Q2DpK^NN4b0bH zfTZ8HFP&U$29P2@UY6Wq9sVu4^gCC5@XD=2`36cuvE9jbqV|~Zt-vNJ>pjp+Gnh4i z6@SLSgd&(RMWisa^-gCPG06D-A%)ba5;k_H{^cHTHu!Az8bc%$uw>I-$oi?>V)reHM3XB$$#`8gN!9xLm^jQ#{9Sc2>zT0YsWQL!_X*`c>_es?QJiPOfRcBB zk@;3mqh&I)XE4Ntgg03h;An`J5f;!`tn-2P@E@~NwyR;cz%8hqpVixaR)kP$;CB0` zZuLDz<@uAPxBiX6bg{lC&ORw!;N;{@52i4JR3K!CrgI34=&GWE%~+|DNufW zx%h#ca5{76`~9c|Ky~k879YR#x9^tBM4p76MAldeD#=kc?+OLsVP*@Bh6C9Y4C#l= zQ&;!2?Pv}Vq*9ZAN^D6JZqok5TjgHK0)1G&>lXF+aNAWbsO$lBem3mhyQ~7C%PPDzZI=f>*xY*wyIktNcbACD6E0T@5ObX}`1$Yw?{fXY$^)L)H1j&%*>PkH zD|Ge3k4q!i1H1!#wSzH1W}Zx5n!_XV7%n>b>8uGY4z>Zai=6;Gq!ZhvD$c`~|9D2z z;Ue16I}eZWGq77GJcTsoVwldiIVMiNmSuotiavCa#+Nxx?7^FL0tB<;ZpiG}JVCmr z)$p(_gOK!Ta5Cmq^T`JNVkA;RMNw8t0^ts9zlUP#v|#TNwYl0o10m7&>&$Kn+y8GZ+pv5Pq1+Y{4NPtfiH7f<1#$y`DwxOt}epTW}OyOuAM~Dw2UcsEZG*F zEFK#U&DigN?80GEzq5tIHpVpdn33gJgi4YGGvN6>1L%O_3s zBBG0K$o;ZN=W(yRK7A$l{@f{DJDy1NjY?yNkh3Pf&>G|gOsuU!(jOkWCf#7M>qAnk zZob>J?YQlMJ9kXvwks9anOwb=zc_DQpFZW>p<~iXX)(WMzH9oEF)6RSw6r7m=^Ed5 z9R}iw>xsV^i1`LQv_NVn`?a<&&{Dv6D5S1-R$d^0j@W>^x}Mj}>1(gK+8tVQx5TKa zxo~W|8s%BMV7@U|ZR%8e=YEs>XJFmVBeB`E#07B6;vq>r^7OeQ5;V#-I~QI@pFc0> zL*;BdRMG!rF?V_Dt2H3!qD9wu$I|K5Ec( zcAjI>L2M}iI(O;^U;NqJ70pw7(O2d4qeafm3!A3g0HNn z5>HIB>P#>a*wyvcst3HOa&dFKHZhW7*QTz6R(ZY-l?*=8R|ULV2j^sj7tHDZFtN5k z$ldH=v!vq|&Dfq3*kk)0rSR^vHD(lfTEf7zb#UkBW57`i;v9FsjS=~);(ouw#Q*z3*b;Tp^l};RtHT2ps7-SfFU7_>b4x^ zMbXsSdskO#wHhLaamE|zx=*Ktax%sOb7eYvbAQL*YX-4BevqcIbDaI)xjC8C{X5xo zLd*s@C46(#v_2gsLR4nDWSJ9AhoAkc_i1ml3%v-FURr#3rbctSqxbg&*br<$ zsK3(c9?f4Iy8d*UsU%xJV+RvtwdEWR5tn$PnwAj#smJ>#BelNiiRyXqi~5h~ zoRG>^KZHjQLs-l!T7Z_h80{&akkeh*><-`aq-t&hGg7z7(p`dO_T=zPr zd9n&@CZ*?BU{N}u{cKBKq*M?fsl~M$dEI(gj~wlP=ye51kxx2Kr8K%3e5G*T=sJbo z=P=rgsftS{;^iE9tfxZjB`Oll){fh%vu3tje{~cV0_3D>1Aiao?#BxCZ4Q|po-s+; z7i%X7XgcnkNucnY=;wW<+I_J0bJ&Ll97H6`1yRE1m^^1vC2alg_~cHdo=Qv$vmpe~ z`Zi;DoSP@J;GOig^nH@{kEsqK!#u3_pCb2b;DK00$k98YaL-tMjgV`oWBsz4O(GFB zlX%W=bF$oGabGE;=7L96V0rI7Ax?{RChCpb0&uc%cbgHlxr+$)%PQJvRX-poC>_-` zg>B;6+#vUYk*3DBE}b!w<(7XHbs6L9nlci4CaE^?bW4Qf|KjZ{qvF`MZi565njpbF z1cyN5gy8NT+(U5pV8PvkTN2#eJ-EBOyEHWRid;G8zIX5a&iCgH#^}*>QFPU=wb$Bv ztvTmHz@8gOyb%%7{8B%EAuJN+LFyByr<>X-c zw^5)f{#K<8w7|aFGj4T&Y{%=0HnQ16C&j)BaoG0q5One|8(7IJMzq^A?qtSp19jtB zdRHG_?N>ZvH5=H}6eavFZavOUF^)|H;kshbY;k>l^aUz+LEQ}4eZF`PQ^m`XZ64&c zmbS9}D6YUpllkb@Zn+u77{Zs->vep#Oe3lw9D4OMmEsR4T50VDH;*+gIcRRt@3Z&_oBF7Zm}B4i)ogvNI0JFWy#c8KQLJ-ILa6-Lbpi886&Z zH1K|Dc3YdRH32IwIH*2NC91zor5HK^qU+4mvJR7_!SHtg-odDOJ*XNJc{T&y`yK2; zHQ&;xmA@Q{G08|*FLZTM24sxgW3XA zXYjnPZU%ABGB5d}FZj{Uv1%ln9<(&AY6Vyw&L0h|@If}d7EG(b9j$E8I&!{G}^ zcoP7O?wd79-!#*^{N`Y55_4Ncbi`@yZUW=c0&ep)JKOy}$nt66oW<%v-=k-)%@@AQ zL$wxILRvG4Spm%1oV^z{J8m5|HP;@4$$epL`QoRN!N0TfpGL^<;ZY{Rp~>74Xg|1{ z+ojOR5X9 zY7&c1_u=)_XigE}<8==-$P>9+Qc^4Xn1)dcVrvGrmh@o!6i&b?L@3NyG10Un_%%$y z2=c~**iGJHn`&%A#S>gD5;D&V=A+mP)INOKM|yPIobL`40^ho}c`K38| z$fKvGE&#`<&aTH%L2FUkbBNOu!`@ifOi`dhgF;$$iqMu5UGira0Xf? z#o7^JA2&65hOpr7Oo0Dl)NNi-Wdm;da=nuvnh|izivR8tL#+LfI0x#Z}E)pj5IBgM9SCH&AJf>*`72cK{LP;KduTS6YBt-AXfXb*_* zz#Dc24GvmE0}m2&Sik39yga=cD>TbSS%Qt5)oO6uua-uHy5Rjen7l{{QYt9C7f7wt zT~&~$PBrBp!4Ac}GVJu|0YG!?u=nV--hAkL)fV&BAzUn~BkWoCeL$LpE|Gh&?wrV5 z+9)**>g&GX-?pp{Cd`oZmWO)9w4G7$5&mDv)Dc4=70Zd+3j&%zH6!U9A}Q1SZuI6l zMp~goj}=G{lr+m$ooc>Ecq|1N*_mDr zH%3tgcs1O(w5!eL*(*P|;yxJXGc)wT-j#R?DjyC3QZ@yBM+z=S{=KfBLvM3`wD6F2 zI3wSes1y-nQN~UTWgXFJLXbGuL&purw}uTjlg&3)ynPj#T$O(?;$DSYU7Yr7DN8>< z_8GN6JhL?ov9y{Ra_pmU`OleItWT_G0pdV$#P)zhi{L`1)x`{sJuya$8z9s0#N&Cq zJCDE$0HArL&8h}b)5guDRs-r8O$}9+^mx~XjOz{)CY9`?jw@{{5nS#);jiCL=BwCN zI=tqC)czF4t8a2W@pjuN7P;K49G7`@fzLeE(G>c`ysaLuaHD3cv28FQW~?oM0Og)t zzIMbOVn(3%cj{pL}BfQ9IaDgc`P4JlxHL(%m%KdS;d4xhL_=w=YDMOcmNcX1PKGeZ3nSs5TCAcOu?M#SqtyXVP zikUOMwr`xBWWsC!1is5-RqfY|iJeSu2U@$N^x^ zV#4yeMN-Qq4|_o>oq67FtXDP>gWi(={-WY1Ii*p>w|A_2epoV*Kt*cz!q?jF%x`oI zvKpRs1i=3-H1mQIx4M5G%$zp8Q@J8tU`J{XC-%A%mu)sUYXoCO^sj5VnOlH`F!}V$ zo8;GTQ)<`oskquu-fDP2vj%On0R_W5F+;MAhT~=&hp{hP&uc=ViKVt?@)f!Ys?bTC zjvoD%MgX#&188qRmT$2Tr!tg~jUsCRVC$`e{KS8PM%mx%m2*VoCjk4fmsEE0o)Ynm zo$h19N}J0FvJdwd^h8YYfh+h|0P?s@s=8e^`wn*5@t)mc?_5Dd+BrZzMmvRS(e+|d zq$iAMZzIUL1E*V-pEZ%`2;g9*eFU@m9&_4l#rAMljmPw9Z)L@)=jbfDhB>4f9T;E) zEn zH3AlG1|fM=^U2zeB3}*EsHLMT@Xlo>?@1n@RcicP@%QnB^803K1b8*rWot&(--$mx z4x>xHXzs()2D@o(mdo7Ta222sIg!JQ6E9&{H~p|&-W=vgx9?4rNXt@~(xAao;&WLn~}=|^g} znl-NmJ(t_v=d#{^hawN{}!jOcq$Qg^!Pu&G6s1u-%&Gdo`f zHu}rPpI-*jS+y?xBOF6XUbGP-JB4!HvDAmTR+U zJ{g2&YSr$_i#jdlNyklY6L{rDR#ESX|}) zgS$Z>{##(-lU+{@NDSqmFSg`B>MiQdvf^>`C41__E8lqlMUc>|_?`1A1Ex*&*0Jb< z!5ZNtn*RDLak|ywchvw!_4;;VqvmuWs$o$y$}Grd0Iv~eWGqVBWBLF<&%?vN>W`gK}wQlsmE{4Hn9_f{nl&-#$rDWC4YNOrxyvC5*+q07O z+Dl9-q9CN^RjsVwx&^OzPqEx}R1b}t&XKr}rVCR_o~u;zwm6l((QFP*i!I7@ObjDn zmctJNxqCkN-J`YJm?K8eKNGlFfIt6o8s%>vS)Z5q5|T9O`S*ka)WB zn3-Z4-8D0t&=brJVVMy&!6wG_ zi1lqqULXZ8jn7oj`QQ?7pteboS!m6i0D>6g1kEm*Rpe@&*R38-{Q%YhE;|j0 zo5LnJl&hTcgBnZ+I-OnuDP%3^EyAq))FehPULP0sFH*EhmmSZhU&BDhtEWQV*M>ZX zr;@(kX6)px*b$Ge%y*V$RXUH3nn%5X_dD$3o16z}B3{2#c4DSf4*NTbh%XKdW{nq* z0ZKv9_T{|f$qPHRYgPayan^S+O^&Uvcc+igt((qTy>)77uA z&)%t8U;@ZRBp&Nez0`Y#SfOo9GWAlaWM9V3n2u;Y)%|l2nlZi#{RR^|_vZA=yE>@B*W7aMGg(;;!oGo4?LGXemIL(XhrhoRVb3a~bJhH1z5~{Z=yRN4e%YDvTrMjXV@F6BqOOQQrYA zgwgtL>$t%YG28%%lY6+2=sP?&Qe<%Y4I1hhBu5-gy9BQam?bG4V=ZySow#6B_4Z zl{e(!CFRVA^cVNxN^cNcOG{teSFkE^^4}h?8XOT*ye=+OLseAPzEd+w;EUF|?(pns zx7+|gLH9D-tQThI8=TCAm$Au$+-JF;5+)X>H6lYH10Hcojo!HI?v>aBS)O+X<4$XD zO#qfB`|)grD0sVUb90dwmcMfW+j{3HGi?gZ`=6Cbl9* z{=UsAS|J4}27BkQ{=FRa_o5U9w5Lxa1*fcu*h(o`>zS<@rAZ%owyd~3w~sG2Y89Yt z24KOmWRqR(ZWdw#f`~drp5eb>ptJTaq5R0J++@sK6wxqXvkm%_l>%Boh7QmGubajn z&;h~};}ZNw3GDZGij%z;(2jmTmPGFJtiK;4YCt?_qY$(u9zmU^!M3CzCSiWdve-l zW9Xj$35Un`{qBS`vdI9t&1rD%#YO6Q`^b?#>D}I8TLAJ#lbo-aN^7^~E-m&38vYQU6m~4hoZl1rg9w>9h|xaXU`; zB92a7p4s(iKjnOLA*1g)M)NN>_zyigYgp8uiYKL#iErI7bs_69%yiKYOCm<;%j zguT8kf6S?Ws+9eQH|2Q1$W^HOFxURebo<++`wy>Cz=IFpXw&)YXUv8IHa$z%hWqi~ z-PNDQffT?{P@p}K{kwyIJjy>E z{bdZ;&X6%;_sSpnihm|9{=*v~OkgOY;l2LF1OA7<{o5lpWCd);qypB!`p=W&zm50* zzo~<&+2&;U?&hzX!iqJu_0aAQYWiT&`%Cb$nYbOjNI)nBzlixeng7MpMI>McCPx}7 z3g2Hhb!-xAF(-Eu?`u|s|HWKSd7A4yP?!60dC79XZu-q`d^(O<(yA$%LSU{ z`mbJ?bq=7?cq=vx70Ult2f)zo$&Ain4Nv~68Lfy>TZ=tk=Jtf_ifs>H$NX0-{lgu# zVu-J#jLXVuG}2tlyq4e+`peH1J-$_s`dY@=ul+*~=U*Lf67K2T|J6~6e6I!!O8}*?-CrI0_t5H34p5y%9rw*&zs6F)Q=m!)Z~gaa`{$$OPtLr; zu$)QmuV0zoQ^bsaIrzVDF5*vC(QBxe@#pXU@9*26To@hAH}XHZ!hh=@|M}ss=qE=R z-P`{DuV0xTCE$+L6yAycW%G%od|HA~#`t^xbr+8CB#B#^V;cTf=c*WqBn94_J(z_$a|&H_P~^XcRAn4Hx{^uZ#|`n2zYI|Hagydy1?SRQ>6J=>A%c3TK0-rxVbK|CELLB_bYSRO7ne=IrcS#IfX_PXj@hJTFq zHG?mgL%@59i6FYNlN}yKAWBSK7tSm2kU(Ct(d!bN61!h%9dGv560pObForQQzomw& zgNQq41kcUh)R^{T!PZpm|1rFF%B+;Y+-}_R>h#LGwJ7@>2{M`l@xoOKK(8WfaB%%d z@%+HYNegzdU0|+jSI^2z1eOPcSPWhrEqB%6$b$>V00iz9S>9b(u*t(8g315MzP;&$ zSCX~T`#}x-!wP!(x?&`(2pjxm85_fjOstJuS$BYei1*%iFuFx=D69>oxO_V5B+Nwq zpXYHTG_==UlVtVbo!Zq;^1(b{6G<7VN$9P;vLQJkzvB|F1S=i^0yb11_ORWz`OkAG z>IfE-tURyj+tHt7hp&DACu1QAL7%BR7QH!`2D>;pVT+@|t&d=~YOVOo)G5gfIdo^f z_@%c7Kq$o}#DF~rZ1m6{e*HhVe}Z^el8dFg{_YMI%g~pI6dy8v<*CYDry{g^9V!4s zhd^terya0V8LOF@neBwUXzonH1eX^_Cux=6j;kYS;%{G^-o{s0e z!w)*mW(oWfr81us*F74&BC5|vJ?WnDx0;o402M<3D3hw^D;LFL5{eck2!|x}7i%>* zsMBigv4~QuySQxa+ikRZ``U_^k=dSh`4fU!*C_!Jc4I%*LgHUm$?RTK#GV&HpO^0H zQarAb>iHOo_cj0u$o{^RTQPyq43@QS)*EImYOk)gaeC zuit3kS6e<2C)H7D1HCcWwgx~(oLx?HFN`8x-VU(J0?dz(czvH$*!{SS`{Rq!1b9}x zditNF-nUcis#9zrW@*OP%mm)b1=D?K{C5-qZ(PT>G-0g*<14U_Y|GFeiBX?i6# zC1rFQ^{Mzt1|Q~;AaY!WTllTECMnf%=%S(wr+RULy@Qma0o&*4wkoS)(jFqW&xJCE z$mh5v1}R&)wwrxCk;P@iVn!cqbuU|j{mvpi~HDR3|FLWN&t4HRUlP3;;w)+9LSqJI+wi;iu zvIKJVlNa9NvjzE|pj2mBayK?CUd^n#?_Z1 zxBZ|aanG$3zHGandiDTM{MMG?^r0IFGX}{EltLQvN>FJ)p5)D4a4=V(IxL5{> zBpvztRR!2ojhJu65PT;uA{-8ish6oDh_m0`N&5I~wRg&?8)ZQ5G5i@;gMN1?0hvFt)=P}O3@%W6#xNSt*23T} z@a}FiK}USi)ybus@DNG`W8E;Iaqk@#<7K0U(pRmZq+Ed8MvX?gZ&JCDWG{dwFu6`kA|^TRl)WzIDkB71E7Sx9$k=`B~@pB-5+m|8VNbh_#n6{N^ zK5|rHX*v?RQy?_k>901+Ki)1s8s6Cr8}9iyu7jf&*MGs-jzIBMo_s#%dZKKCG_I&yHv$O0Tn`yoo-<7Gk znT=P)6D>W@_MU!@ApFAt#QFB8V-{W@mESBtZ;qB!_CY`C%*Mokbg!q+qdM{!l!j-Y zpvFn~c>|g-8xKoOZn;JwTW34ukB87`!P8pO#OMUeL5qkYV!*CIMBjHe^P%*H_?bTC8J+Ap4ewTc6nD3TOy%;-x=7vTi6`!6@z=h(Q)?Uaz z_|_V{?#*;s*RJsbKIZh+?r*R`l$u6h+wCrcMWEoHG0Llf{NV!vaz)ms(el&*omwG= zmcpC?XCwWO0w|dE8AEM|@aZ1a%8Y5Prz3Kv=Qe%P#yVygs<0J-xq#7iY{7gSw2SAe z=WBS2P-RuS%YJjStz-YvTIj}?{wSz69?`~KTC0J@c}L9S6B0J*i=Amj%OXu$vN<5V zx`J>Q)ptccvJT58?-fU*qaGn!+_aT17vjq#ps_l)=fklc8^T$kEyMJ3v0<=D!&!w5eQ1Pr3co&8WGJ)VK-w7e z?>$^ADf!9dv8G{98YAY_V>tKg?oE+=zh%Jy)Tqn->3t@><_4jlXpPn;yKR}9lMb}W zvXyl1WNe@dS|MzAnN_!)DDH=40bjV?VE}^{$70kcg4>pI3xM!qQC5|5RjFvsSyQ5U>^&$mdNM{0KTih#BEN?{S|jz^_jFQNNq6L(mFgO z_g}hr7;FT3E3ALb$9ti!bMi(yFzu3o;JnT z@;yNFpf51WVAA7s-JVzZVy*oQ5{iR%)5>KaR&&9FTn-)R4;J{<{v6#H^* zX~XT_pH^jzZ@QWbK5Km>9v zVE(o;pD4LiVYA*~PaqNZmi}E#m&&>OS;7#Qg7voeTc2KV|sLr~d^{%!dS&FKcKc?3(jdQzo zQ?v3Vqp#`aw|0;p`(ZV~2U=Lwnw)*ruP7if+SBj)D=SfN-QKO@S6nZ<9mCk{O(gUw zTvpFpzPdT*CD5zdt6t7fdvEG@IhO}0o-lxm6)%Y#mh%{T@YnITv-}Qxa^r+O>dJGo z>U-FAweIgDmg8QjuKL_eJ}Y4>c{rr_^zCv|BjK_dg2mj< z{N^&dF&u!+It^Jb2$i&hGX+IEPqC;sJ+7;`mI!7mW;N_rJl%nYfs@7aJoCz@Fvy*6 zMEep?MWFgIC`t|y&$<=i9o(Q*TT6?DKA;?-@!hFu2MCf$#t)DsG^Xn*DVu?+0CYy^QD7N@p(h*S9r4b9=EArL!Rm7&chlL#0K?l z2k&V_3g@?@;@g9+w-$Yn*ZpmW^y+YF2G-Ll)L;+qp0QBj>#3fdP4^3z)!Eb22f;l& zcNM;2^!SZT`18PdtNyGCit7o7Y1g7Cd1ArFl!YVirsa$yyvnu{h2%xj?*BBvPHOseaqIg44_gUoW1aZY@t0tCjAVic%Qr}nv z8*i5a^3%r}Om--7SQ`z3)`<#sd=R6ZAIOEVzMeFCKKe!&@>F4+J=_W}I7}1fKYp#A zH}~n{+~YCc+ODN`>-0cAMrwI{Xs9LMKZIT-wrAO=q~q1Q44J)b05?*i67rEr@M`ef z9gNJcu%Eve$SyMOzd4epC~0@1yFXCd^KPDI6%VnJ3UrXzLHk*K zf6f@Yzp809NU^^HgOOF{pcw;#Ox8Dh+FmG5cIPDATVLRe9CasK-?62mx|E#l6%!IE zxL)EFIwH!7oS5dsq@ILby|)inpUp{pJ%d#OADWta5kL7LAifIW^=~U zN7r4-wXkVWo~N<%rkBph5HZ(X(5C`Mal0VEVxS-aCG`5>V;_Eyo?OSXPEhFb%EK-*a}>)F&Os%fW&s%QnL= zjf!+e@w+;BL7m4Lm679D{ z8=yNAH>rG8=u6{m2u3SDUpWsT_CyCx!%FvX{P=mMRG{dsW_61DW0Pb&jZm3>*Cb2r zhk!Q`upw5VqiW+xtR}I~L1$}P&xjg`oLAZc`S|G?Pl?UCLwEBTZNZC9AwZzdHv;wA z^tv5~y)zc71CTaSU)S$6w#3O7Gfk9=r!dYJ=vPyTVqj&`c=DEZSq*<>4ABCq1B~Qd zgSiTm8h!7$u-9xV%#+MpmN>(AA&fFX9RX;_p_d=&={r_m^QjhpL@b8sg8-?7PJrg| zr6j+3;mxv9^Yt3CH|`MvzIwLEi_<}nPXAf>PoBV4O9!?Rm0$dTLQ?#;{hLpGsmn*_ zODHZ~4rWae<}+~^7I@vWPc0KNGKAFLuX2<8eA(>JadSf>pQR)$TxoPJ zlI1on*Qm?giEAqCZbiXOi-e*bmdvCKbcz%1_K5W+4uf5-P!8#?u2@?X!mV+rcSd*( z1NQ<4eSW~N?PyoAC+M*^U__A(|EIE~&NqAvWGS>yW(l7XFVR5l)&@VY;NnQ=$cmOr z)p@3i(q9uy?w_h}l|}3Kb;@I);n6u^eHe5IY(QLXsHnVtF3Q@b{|QLgRl}5Wr4z%| zqx_Inyk4p|i+i2ewX$fCD;bP8-py(8eeZplXu2Gb&~ez^_KrHdx0=v*E^>aorwTw4-{B%n zwpSLKAW5d=369V6C2mX3k`opqABx;Ot##xZRM_U-N3liLzi^HJ6vw9&$!OM%mK-J9 z)`3ph3#?n7wZjSSRWuRlROoW8F*$L+#p$; znoGr9BWwX*p*AR2_v^pXA|b&^@Q@^66#`MqLl5tJ`-)QRfZzLxv_O5Jpc3YwjE&f) zO2aq)z#Xz}zxcyEkc_gg08#01u2RkdK*>$@`w}(6{S z@f>7iWN=rPO1$_m6ITl;GAHCKzYX@jub;o7ms$z{w@_-UH^`&lb212UMkFwj!(AR{ z&Y5dfPY*A71O)nrB;=dy8izI1qDVbH)nNSMBR`VmCn+lyk8I^9$+f01y^{Zo?5XwWs;*=SS$6%~Ap=w4qb zXQGlGRiL{yyoF5jjIdHlE`y6W>i{&8U!aZv_U-6vGn)JrUn&O3jJg5D^{;!)-k zPWu^L=Hsq}iIuE6UaH!Q$m7|vX8ep$wRv5;mMyUJv5wZ)bavH*)hoP^6JCjWZc%CX!qmXKSGR!j*{cw2dZnxyHe#Wza6<0hvIG2ye1K39LMr$H%d*IYS-Zk| zm8{D7a!*;mH=1(qc4yQ;=(|)>`z&*>#Zb!Z*el4;3CsQSTW?|4qXn0VETBJia2&1b z&Lkoq(7B6juF|X^9$@PX>+g;wbOBr~%nQBo@6y!ys$LR2r$YX)M!Hab8&TwaYJH?zn+H9z!C zU}!ra;?=mxkfI86Lv>63nR!kr?$3`!GR{nJVeD|+Eio0TB++|@CN~DC2M0@K0*3J_ z6I^vgf@xLz?D+|7jVGjf&-R*cZ3APdWoQZrI-egi3L_)TiOIBs^eZQeFJpH=D=tyI z^ICmeudoRj0&#D&kwcjeWl^~v4Hy2CjC!E4axuig8D_hSB zEM0yS(@FpYDx_>ZJryBDpD@jWa<0bwl@0C|o5O2_z?s&w4ny|OVmk5#-IU}Er=w;o+;eq$x;Rt6xQeN+ zphB4KGjyxMV5!j4trt?ctCiPEG)|X^Yrjo^-(qJPKp(%reRB__RFsM<%}UT2hmyYY z(CfAdn04i8b$XG}r;uoTDC`^%%E_{|Mm|JUsJw!8xtkE+y+$@y01}K_^hy@u)t{Y$ zkB9oIUGVfPkIsfaM>sofO#8~c;E%crD(aJm%;_#j9(R(qzWctx zg)QGyJ+jTY8(?-6qA4LTa%URT0>M&>Ob=2qTMIbq&>aH!*aFQaX zp^T^)#;FX8P5(+7D+_!)4P;RKK8Q&!Dywum-@N%s89Wj3_|Q6NU&JLn!5bl*c;{48 zdBykkmajBl#Ag!Bg||I91i`nM9KZZVTSE7?;?6GOBKkb};`$jQy%>7Qs16i4XceE# zYWgS>$53iWKy$S0o>jtKl+$ybQi>{nI>V^5EUP30Z(nq?s;e{X+hjpt$%fx6>Q(Ng zzsDG0o={EMAG>Iyi@G6}z?A-+K#UO{HJQR*bgi33)y)#S4d+!#zqjNB+M|f`#bp&) z{?K)4{M_3JYqV2-z}JR;PgXhaY&@98+{=)B$byoeas0(VgjkSzkuOtAh?v*{MgMV@ zcU<$=wpfHE;UyrY;aq}`#FRwuYV*}}qB6`?#3+tPnrRoUC(IA=H6S92gZ^Svc95Z| z;3N4yGY|G^W8}zHiL*=d9&z6D$NSN3gti*@x-tFhUve`98!ZQ8{bZ16=lGUNIi#I< z+>?5ghHZF9yOk9^qaZX{7BO9vHjH}^@91?g_c40Eb{h4`0!z#+_~N|SFy0L8aWT`& zBGrLXg~V{2O6OB!u}pa@mI>8fmK_lMR+bNg0Tt}hK~*%#&4#_kbH42+)YP%_09sO1 zWpI;kA`bvvoKU8*&uA2;-{KhESlAWy7=XF4SMjk>j0e7z1h;jfS1xj`E|zJ8!L33< z%)P~rbUZJmo%InQwH3Kw3OoHDOH0QC^YB?7{vTj1^YnM|POgjJuMhD*HQ!PEE;XKruB_Bpg%Ce->=#6;`B|wsZVx1l zPtm(Zsx0?}6JmKlE>yOkkJf56m~7%2qYiH)$~)|gkghT}CtEeEyBhSrE%HNqQ(*Ga ze!ob)GeZEzsD&xDd@T)Rm3AR_9a8?G8#0i@+Rrz@W;^gUY-b*H!M8`{4nAJi*+UN~ z%fpOmTd%x$9b&YjYQVx$IHuQyg5lN^K4J)#<>Sl>$6B&vHA2Gs@IF@Lz&=#|LnOcT z%9pn|_peU)sc)-ABZyw4>i}_57>Iz%MEBdGH%m(;Iyx>WKPJi4ek=nD%Pj1WvXcg_ z?obdS0r+xShnSB_O|gTEh(@#e^VuDva=WcB+=~g8l>V7AgB+7(r%9_Fv_vuMy@|Y1 zQnKRNFHkU>xAMQ-UbB2_cLI7|bpg5a$*aWx%l{L+ykDq zlVRw;9Fs&g5w5L#v5B6aNvm;Y!{&@oz>Lzm;|E~BjZ)|XTb(nBCt7~kF6U}Fb=cZn z|2<Ys(meKRyU8HDrT#)W>LEEx#zURo)a_S7jSP4pFFm@NW%(l_cwg`2PATAW1>J zddoG7@V$X0)fI7)!GqbF%kfT$O%Bxt$8ugzm``29(Yye$mfzQ(1@3JWNq*wdvqf!` zK?iiw6%E_imvE~R_=wOLbUCLip|Ay2dV+#zAFAFzAo_SS^zV+pYz7r;QZ=RKU~Hy2 zOBfxG)uoneSIuKHjFJ*{G%fGXYb&czxV|5hxhCRuj3%)EmKXeKt~Tvk;4@?$o)e~4 zkjK`?GM9cwN`2|ncN&N_Tb{o9neZ)gs?34v*{(+35O3JgOphOFV_n+~dr(0$xpQIc zQ_0g4W+cyUQZR>J#fz3s*|?b|`Kl%=eeq^XWqK2KDHeu5rL0nd+bZ-SHI4IDm>CdD zxWvc4EnNV`jJMXz*R9qTD3wt2MHhpjNN|-(R5*EpHvLfCM#>$i&0O3RP*d*n=CBG* z!o1HV;?h-Oc$t18F}J*bX#BK-4n8Sfvk~Z!@?u0o&Wt1j zzyT%%u&A|Z{;K!{tW@Vac0QYYh1sKMEKQC((AF(k5#GI)Vu^uYO;92DV(9gnRl5|Z z{hl;)WOq(j%{E?fZ?u)Ff26jPVk!wd-k;7hWHN%xhoF^T`{_B~k9fde=#9en)alwn zLkemGq(cCH&*y^~N!}4~J&^j^N1SfgMT`IqSAM4BeZ!kGhs)EV*BdSlC+27oI*Xby zNKfXU{xG&ScA!%gGyM|FCW-`U6>p6#6HV=>se0Vv(d=3k0c(2mF_RhGJps>fO4_co z@fu#{t>&AaV+YoxOdY>SffM~Nx3Gs?RLE;d!Hz_Syj0VA<=-6D!#(GbZQIXKZio@V zcykD>N!x)z@Myi_@F%vp)K%@l?#&lO@0=dUs_;HM8~Dc8Mn|>~KF`Z<&8QN0FSxLe z9z!~k@J2;w(t&`O2j>L=i^L0EJ(l72n{jd;+$4`-=jAvu)Tz}Tll0V=5%DexMODc& z`5<2tTplZCC}<+Xf!+X!>#8dX(TF6Ea*@K!Ez}%OrTPxMz7JcYZicpt^l4?A@vP_Z zGRfS>bLW)i9y0g+pxYB9pR(tvauLi>P*@zyR9~uL*Q&R|%tx9Poi0;aSq)-P7QgQ2 zDWY$Gn$qOaIjTQLHPc%X^gkRshIRGwyakw>sm+hfs}%9g{w|)5#|^!!9unuf4SWA3 zeRoaF$qVp$DwjXiTsuAYLdE_?oxRIagA-$}S1U_{c9W}X9AtvI8Tr-Q99U1gO(5ds z(G`rszm?C*FUqU__~m@m!U8LL!S%xyD4Hx}AK!5{+GzZ5S%8V~*?Po}@XL znrzDL-li*ew^OjjFEpXi5=)YgwUeOQZXBajJ3@Gk8#DCgfZX=*uA>Y=CRd9$&&j^?Cq5ga$y)NQHVSBT8)zEwXRZ6T7ZE>kbenB@MQvX%kHAs@&my**-Cun_#h{< z1TJ0t!r2$lMGE(-N|<90cJHf9X{o>Y&2;r~Y|o;s`s}^~1WOMA=y6#fv=s4w*n97A zxcatjG!dQXC89^~LG(@p(FsN{S`dWMdx<^>g6J)Th~9f=5G@kDk6y!!UPjp~_x(Qa z^X+4Q@A2H4Iw5nQ@kdr@(zthm7FZSkj6z=AP@8Q>6GTN&yFzVgRs$Be~pLomLDBg8&_UF zuXT)uYq+D226y)k71{5vYu}r1c(?Oi;((4{Ppj&)Q=-{4Ib1$dU3(@M}z5rQcZpxp9N^%X{_B$Gtv_ z0dqK%0b;Y0BK~SVfCcub)Pt&$Z*Eo(OJ0Cb<>23dgb0f`0% zTXGAO7}1}!yNS_t)$A%+)o>L`bwTlZ367VHt=sF0$O-Q(=`~^#hSVfZBSUgAEe&X> z1SHFD+M#%LSDv}{SoXaiGL_%SY1@jl++qazgeY>}efw)vO$Hg1d|OabzdedX1Zi}x z2+VGkrlcaFI9aR>TID}snlht$*a3iMy`KAo3Q~T4x>ix*zP%1}339(7Cq|%c;(>pK zQO=rI0Odz_B2py=G6zu0aDB`UyUAh2>tjxjLYOu*Up%dqaoFNXaniZS*z4R0OU>c{ ziUqkZ^czCvXx|gZj>$-_K6+tZs#n!(Zo8%-WCJQP2Jpc$%ZK07UrUb5spIy#lwb%X z=Q;xX0CmImCz80``M~Y0q$X_AV6EcyS#?W+s!}*YHe&r*RHub14fmf(w!U(WKKg( zClDKB%}eP>UVMaY>cr)nf+KB|-R2n(uzgSr2Y&*#zocUhfLVHBjW1=FT2lJlBSJ)d zjbL}}@QNtoX4_xO!p?m&tOb$FEV=v-NL znqO*_A9z8P@1r7PF2<)quX6&fieBT}zY*-(7eTkoFV-t5ni=!)Z|yz@XpFrlJAB_P z=IR~T6gRt@l}~XFsrw_V%7Agt#}oO(Sai=mKr)A=fvEcHw;M7d|m`fJ||s9K!v zP8PR1E;g4O9rlp5Bbv78IEK#&<5D6#c{^n9>=v)6tR@5C{!Q!oK_53deh50RKKK6N zJ5i;XEy)t(RDP_8YGocNcoMiN@T>arbp@)unTYMsC{AJ5Pet#f?@;?<%3xrNEevsd zkN-WCQVO4wq5kQDWhxGZZsk~yDzkN?sr#hfrky6-RZ0u=Fx{{$bR>wDr+XVMYidQ2 zJj99U7oeBGyg8ayETV!eQj{-sB57wlVd!HjeOiXY?|w*vrTRO@MZeJ`_&eTx8j5;< zUr!txXJ7tj)8?{|9`tyq;PpLXki7V{gsj7f3R;vNvGV9{*hzR8(u+hh`+6x*Piyf; zp)`nG60_&Tn;}Qa1x1o=4Hlbbs3D1tNN*L;*p47 zNuPCcxsv|-4jEJV(*mr!C;NU$c#h3t(h^C9^ML=+M}E>>!`wT}U;Cl)Y3K* z+PMx0g><+R^_vkebf7ltD?gXS5$0^uYz^ybxz`fa(DA>sRA+y%f2g*zO3aTx5~{)S zoD`pKz6tUiY~SK@JR-tDjFXt6RoJh*_Xrl``BK9jRx9p78o5H7ZydH;>;?NB3Zhp% z(g`%bskh#EGHr3{E&E}vHXzIaEEp~s>c;l9o?3!?g7ybc74vPaRkk248qve-wLste zPo%{Uw|ccrJoiKKIxMrE* z-~k9cNw|v)Xl9CJ#wXjOqYXRTV?J$YmDuE|Gtpr|e73{9j(2CLC+Pi?1L1`Eb^2B$ z#V3l^?JdJLFCZB%ymvyaUZSdp3=C8|Tu{FV$0*?#OSkg6Ihf$hEqg)#c;KL^U`T{5 z^>r8|i*2q@SB84t)*8EK<;`UU!W}4Ni0(rRVnc6*uNFOUo@fm8b8>jDk#C^mt9>3D zc!y=CjX&hsf}V1V%tPNZZIn1q|5?~2B-`yVvjmB%^LI&aCEXXzEYBDi7PkeZh&dm* z0aoo~r{i-CQ{J^t!y`*lJMCDHY4*9h-IuiBHJQ~Hkg;#&?@`uSZd6lSSn?)aMvoH4 z^Jz`O+YgYXVNYNB=!DLmf7minNkMFk`!&e5x~e$}kzO@zBc9Bf5ULTp@26>1W$J6y zHJ#xe-#ho3Gc)lZI-CtL&n_3@8HtR%XTVl$2yFmV14QpMHZc2BZFbY zRn7cOl-XZ@v$Cb49C_G$%i)bIK_Z_ReoU4&ObyU%@?Y%Je=haf%gXtae@>k*uAfL- zf$;+Vo;h}6tvrube!#tzd1jya{9TEvE02#7>7eiV0h2=Jl1)dAXiC6gF9*2@H5#G; z9zd;)TN(*;Q8@-vJn5JOMLKd% z1Vj6_5eMEMM{h87!VWOODe~S7gp#HQ-d@`Ls$}+ncK-bvUs*f(xQHdPA#vPKv+fJZgju zZI$dbu#0rhbaE18Jdgmkobd$q^!HyCU;ds~x$qzrKgBDU?3EZC$9dq-rr}p^jGgZI z;xB-~(UEF}mnKACxngK$xAI64H!6s8cq1+gSyqabKSQ{ui~Xk3O@tI36Z_%ja#;p4 zM$QR8yvIt!*i(W9JHVTcv$&LJdCX%F_E89=#eL#DQDkD$>E5m@&b&q<4d!^$h6$|x zdrAToMeYopMcgeLNjb$_L{FB%V#-ZR0?M|yQbrANO7s}=Tfw{|2^ ziwxiX!U@Pfy5HA#*gUP5z|=*4@PMb!_qZSuHDe>wT*3JL;D(y*BPPE+dwK{h>~Ocs z6lC0S_k-Y=>`8kHuaRQ!^DydR?VdKHT1B!afy?O@IzL|Q^m(CT7jd3DK}m@oxm{qj zwasqIf2Lz%W9ocKNUHaOfco2l>u#$@4v^w?V`G94K=BmRyz|LvfpCkfXkp9}NfWaD zOQ-adlGmjy>qd)u)vro%@jX~CJqH;LzIk}J@s4NxSL~_{^jhxBCx$LT zi=s|eS}IDqyxU^RTl81Z_HJAR_A*YXD$c19QeW8iyTUkF8PhD|(i^uew{?2$zG=k)5OWePQ=pa@nkafgx=poaIm>xny(CTu$0y;`Vw+4{meb^Ys_ zZAX)h>P$iPQpzQc=LV4|k&F4&g!76XQU_g*hE)40V=qTnqH?4T`J^3yKQH+$cd<^b z7nsO%?+MvYm#9Qzo$b4qeObnE{FPm>>Fl#(!_<-D^Z8F8E8Ba@s!w}$F@i35FtYZG z^d7@UIq2k{5O$Nwq~y`Yxmsh8(`E5Xb?B4TW%>Rd(^S*VdrC4jH^tzV-`!)VAUK&l zicD2vf1^2VIFmI|JwsZlccbqZHO_>X&lWJ==_!7^qWzR=QtgVS3jZK)MR<;bv6W0- za$|srExAlorb;cv+&>`oym^t?&{f9Af+E>Gp^5j;D~@}QsWhoe&q-l7)Xy_Y;`;Y>@!*c$ zC*o?!FSfZ{<)5&IPz2Tn4+bCNelQYcX+$VB3Y^5Bw^1I|2ekq@n<0ewD}J`h?}T>Y zfUA*!U6cRL)(-s#59XYx{w9-pqeWUBmMtdtRQvWuZH(>e`+PHM9~Q1Zzx+b{L^&E# zV|;T{QqlftC1ok;@SADk3U9|$v|^3*NPc1O*b4jtEgqEx2<193??W2LGlLjS^W}yRFGJydcFX4Hf)Q3BY7k%D6`Eh`as=qL8+RetXS+1eENTlKX=tdsox3}g*(s6q6 z4kHX>Xqs#QL#P-;57Fp8ie&MaDcDs@r|0*ZVrm$JO+L;5-Ma3SDRRGs3cewRv~D zoze{>YcJ!A*}UbarmY!^9}cSc{3sPwjdEo3V=@)t$bQn!fF2>@2_Ya%-j{JwepPC0 zM2=xk54vFhSRs!@!}T5+^RyVcev@CPZ!C|%^BqqzBjO%jdtVn`QV=d_-8iH8Smw<8 z@J;W}VVQ=Ydr7lv^sd#ItHnKG6&iEoOx)DwE8D={2KgA=)uj-Tgt7!Jb;iV{1rxP9 z!{M%`;EldX$Im|(=q=JV11h1ffkU?x=w&QodE^I|h4wNm;RL&h4E8@yMY;&q(F%{B9;dHn zWw;kNsn;T!*8oV;fjlqsEC`2v54^~Jrqa;y&5}(cTZc5P^m2{3eKEExCeaWMG(j6| zp2jZjWdGu=A*O%Kue+EKi{+CWhG00rgXNTgTjJA=3r2~b=(qb34uA&6ZssiG=9;r# zfO)2j->%H#3x(iFr*4NlvmI0uA|{ciZB=h%f$@j?Ivj(w9rrf`u|~74)pv&)7Z@pB zJT^YZ48(JF4DPY|8SNFA7=kD?TV4S~$e^d39;CCSs9)x9GD8twsM(pg6rbe@)?M-6 zo3si!L&*3wjL=8sb z-cJYAjkhH5!uJZLm0+as7JFB(JY3dmX`N;5$++;(i(icT(G@M0NfBQWft7r%1pc{x z)j1gFn{g>z4%F@JnX?Ikzr)0 z{q|H&82RWq{O<{PbAa< z8o}A$u ztP$lAM?W6zV+e!175OvImON;*G1EBVUg~9 z8&Go3>P@1iw04Qw3N?T4%C3}U-v^893!Kfb+8r%8aG*Q%sj5pM(Va~#l^x|KGM7C4 zMah%glD1#e?$!5?y<{~21&Y1ab?b-A?-Nz>9RHu6yX<5D)*C6{OqlCJO{nT5lL#V} z=xZSVpm$wgu1Sz55}VfDa5~s~nbGaZEJ`~>#%GFXBaiBcqItcH{}An6Ldd<87mn`ZP`2KMfk1dc0Z zIQmJ1QlFH5EJ&1C)M7FEWh}HBcl;JRE3>XGqfY3EKubv5d2ZymJ^->%T6XfGaxb|5f?&11vuXKK z!0tCQxjvddhhQ#zN_nZ}tw9;1Q7d+~wV?+bfSdF?s=;pC6AI)u%@}k5e~sVuEZU>R z+1SIX+0EjvU1Y+i7UJ|UW_=^g6@G(}l;L%3809eH@iZN|UuZHm4OgX|d;-k^e#nVA zKGh-2oP&XWbh+)v{#5bulsy`h0IyI_H4|>~l%GwOtK;N@)&faF2 zp%y?UX;7-EoWAibQ9gjF!1vzfEOT`7n6H9o`%8+rH}efEZgxNa+58%!H_ODotv#LR z5&T-4(T{dg@Krshg!?z~_`F(bCK9wmVppBh&NqDO2w$@earXwG`OkkNJ+A@BXE+80 zgwAFi^>9jIyVu7HR!#6c9TDP)swxk`deCZ#?p_YwD{B{?T60D@?29l}0P9%uuXQ|! z`ZVvpEdEDyLRlT1AOKZ=aTpi+#GV^}0o{Qoz_;pb%G+|b0p6;b($1}4jvpfhz%lX* z9`g>X0FCt{Z>0pg=gjtAGP|~Xtl&DiVa*b8Y?VnzfD}0^J5WmG;Pr)SW{NWMj1Ufi zwowH(qnqa7{GWbE&W#VRR%P zh8yXzFV9$ELEb)HYRCfyNX70@BEf_H1U>o4&jm{@{Q{U_C^+-nv526)gRq$xPPJ*?# z&AemplY?%iim4vIca{8jr{B3@4oo3@Gu3!-|JPup5qvd`z^)oem&NV@S-D5|{%I%a|fN>V8EJT!` zu*>xL(V;YfdZ?hyQS(&>mwJGrDE>DfS1tIi3|0-wn%-rTH1}1cKQ41bpO@ONW&6&rKr-FKWmG?^%CWjz~P5rr4OD<{-)luUC!HY*jl}b zB~Q;ZasbMW!@Q@S1>5sI&!q*5*d8hGT{V8YVdhgBg;LuRCv#8y6tGKVb?Vff%O(ed zsmmSF>Fxv>Fp@kGP3mVsGTU>hTD)-x+)DE-{x{Bi-384?!APw8KRNwS`!<$2;N76{ zw(>PTxO4x$#r)>aP|m_*S!ImIvheSo>z(tv;!G7r5@yNd704I?w^Uj$zMz(O6*vbc zqu)UR(_b~|S00N^oelWhUA8%bQ!!ZlEE70kw#Fib`BG+#>m>4gJL|B4-{+Q3xL1La zS@1;*hiJ>G)JSzO04Fk7ta8i|j59ryjF5X(m+{n&F5;4wu5vkk7Z@C;G}_1)tYuLK zRx+z_sR?VDP5INBmVmgzhf^4s&H%u-xeEl5iEa^j?@|-m$Dz-T327vChSMmIJgh{h ze)-YEd}yNvxqA*!CZ0>g0WUC||Ad|ZX@+b+$s6ky*T{yY-mwNJiH=Wa(pLScLx1Uf z)DepU3E$xcU(S`dL!l_VisJDbKLh~NzWnl8a8TB4IJGJ^qS~U@rRv?RKwZF0<%-T$ zqe;_6cnO3Vc2}-bP<&N>*NIER|1G5cVrs)kW;9cb2Y^s`;qjv!Yl0V5lts2VPw zi}vMFeIc|;dte4dy=|7~HGJKtoKm%{e!k&zx5>7kxasrx<71?7Z|i3e124^LPnuN* zD(-C2fq`MJe(IF^&4M-lD<)gdEhjh@A!dPU5|N-OwtMyP{!+k}l=krstDSbrkNxL+ z`i$Cr!y&--=1Go2qzaeE>dzWz&npiW&RUv_JrT&KF?H3bBwQIq6{yC{i&@O zqz_5&di4okUz{0XIYa;C9b$jy9X2;HzK0F)A4N06)@lm&(lnD4&6Rv2Obrp?AB~@U< zgcN-o#_cxxj4O_>iZCpo7$wPkP7C;n-ph709AMx!XgM(!p(@ev;SB1G69ps=DOm}Y z?zD)7A}9We8Ee--JF5^dD;&LVF))^-J$fHyJCk2Gb9t4GK6?MS$ujF+$i5o@OJ&^L z^x|1Z1!ElTTF}b;3WOcgllmJ8LztUA9~z=8LVyHCAE*M{kGr|qSE#pIl==W5d6akhqy+7>HW#1tVe4sy{XP)1X42)-{a-`Ls z@Urp6TZ~W;R46OnTp5#qiesuA@8M&MZUWW=FQ8@CW!)sV&>>gbw8<~#AIIwVrQ&t# zD@F2d3{nw(;{JEKD^z&=U*m~hU?oyor~~N6v`@y&9|GHtGRTTY06-AnoD%Q4{!5x= z4(A$)YFPa_>*;6dRC9HvK+Z6R z+{WBF^Q#`Lu^ngS_jF`$)de`p1tRG*#PjKSm6NNen{#+#RkOmHK-ud$z{4X2Ix|&U zNhjs=;dhRW&Yn5W&5|#I)7}-jQ29(mH%Gdy7eFvRIo)*QjWe~LHUsk=+kc=~Tf;1+ zli#!B+&Wa@a*z5inbNw?2NKy7&kquPhY$Q2K3`F`C+*1~g0CU+R3k=MIBhTV_t4I0-&kUfJ(FA4&o zREUVvb1W(Aj%>0Trx@itFY=iBB2JI_#+>%Un>rd#AdDCp9Px7`XI~gc3fk-Q;u&GLlpxW3PiCC!L|Jg)qPYx@*q`zSE&#M-qp&a$?h4G4ecrWZddg!;?kaLwpyej@ zJ+v!Yea`?J=3A#);Sgx@%wb8Rton5UW|w?4JJCVC$`CiGc$THHTcNt|z9%(yC0D-# z6|D3+j81onB71K_zo|GJgAaB$K>8wzjgBwT%+>$S;yrC5-tXrO?+EW>2WMw(MEwQP z?#Kxlckoc_<1`M}sZ8m)7792*qu<2p9+({UJ&@4Zm&RuNo$+tuih=r528!+Hd}rlUfTMh!zJD4|j0k zDCjw5tyJaH&}bXUXfRlk%|f0Ta}|$*r{qVG=jhh+@xIec4=$VKz>LpXveuvvJuWl@ z>dWuOTFYkY{Zj@hD}c_oi9m0M%9;`s7*Ejax4d~ES85XRvook}L6+0 z`Obv$SKzO`pKOawJkOqfX%Lh-6)HZAS@aywMkW&Yjf-gBP40lWe&GjI~YM3()UdKs3h z60DsWDtOazNxU4^YScF6Q5OUywjO-$Yu|E61zBsYL5MWWamspdfc~88JnCSG z<*|rk>GOBALvtz@0M)3@O^ogcs#^+j)-HRk=X^ZUz|QD5T|V6lbb2NZ%@zemCBL-q z$v6Vt_UZG3#gb}|-!m#Wok#)Pp*^6-!)dVFOrv|zCje4HEHo*U@#>eRvg;MsHm=1= zz;16)y;r9d#hRslK<8=Dy>XEJ?%iy);c&K;(G9oht+RHnJjNv3IGV`dMDjF%y{61d z7l{JVOl1e`-}1qk?(sAysPeOmYdOZbS13XRzQASqDJ~_EV_d~#M~9_R1hOPAdA0rJ z%4zMY^*VtI#*}uQL_8fE6LZ#n8Cy88+{dc2$_22FcprozLK!rNQjWUs+CGhD2^q?M z3l<7pj4oJu-SI(zJ&0Oqae4dHt7vzoDFsyH(e|3`6H@ARUvuU&mS@P>dUrvN{JO~& z{J`{NMNsNkD1zZpsJWP*GiiGj4tGN*)q}c6sdu|26-h1j_&(_a>KAaurlK1=s3GI0+a zZTKr2g6;1wI55aU=!Xg^R*8G)k{K&$LVPVipFuX~xyHw@{L$z+m-T(_?d!GbDD91Ul%@FB%mj_zt+ynNU1c!aTO;C?^}Bx$~;oo&3Otc8=kK$&hgJf+q=*CCisy&l~;@i`!1GKLql=t`JU#OM>>e}N`u&YF)Nub6V&QHj5 zyI-Zb+rN0$gNKI;7ctkU+EtqKe`Jc~CUJnrh>jokF9M?pdDf?#D^EVzBy+k?z@9dm zgkyc85onFoNHq?T#zM!>S|xA4saH*904d|%G@k$T48o$^%cobP_y%rqp0R^-`q|$e zp4D`9s!BQ8k@5%O(m5kzofhk+qaIf9sxH?*9_C-UpQw!-235bU-yOBRbsVM5!pZ@t z2=8P72C-25L-_G}yXJ7FbeH}&7&@O*mLMcPojLj^N2=(Xum4LvrMyDzT*z-bvYS9N zOl-%VLv znRE_JiXB-daF@1giy$>>*XrC?ulC9xQWo^cn+Gknpe0)#vFx5yRF@VD7|*>Akg5a~ z{1ppN=Vg$(i0c!cg&%xkt(rUbBbtB9k0Q{6ojhaPyd&2fq&T5Vlf_ajNGdm<{kk%X zsZzsT|IV`C2`u#F>gmrSAHjp)1|XM#IeVgZVifh7fMOIGF)-e&JM}|p#45-zs>b7V}Vc6x5kHA8ULjE)d9dJxx4s*2xuCOM}EOEW87G9C; z(ddm$$!tq?yLqh(g6$}+i+Zdg%SO3Q`F8^NX8Q?M3z-Y>v^+|SXCmJWj2PO_W$v!I z20U@zud$5+1ET%Rlj+J=J>3tt7%qHh@b zXtfm`xlnd7cN7H&Gmj@w{CM&48b6UBKqZm)=26;UL%XbTqr#}L&JT~Qo>SU7JRD-2 z*ou>0;-lbw&H40Rc<8AvYq;^ma>Kmry{Z0mvm2c{AOGGnVll>LTB>RCs{V+FJN8Co zm|ApR`Ki@2VU6QW3e;A%SPlJ7EFCZUPQ#O6EGhZgs{sI>Nr0JrQTbgdSV){znv9n$ zYO#GQrU8B7GWWq@Pn--V;eGuTBTeJZwuKsB8*OkOYs8~qS2Qfx=WwO_*e1#zA%@NS z!AZc#$Io!{PE8}8mO3`}aR{v&^|j)qcrTZqf@qO;NxY~NvXT=wD{9Bq@+2H*$3*uy;_XIjWybTdJa2DJMCpop3?Ks#4 zE7HQ&BY@fyy28}YWv>Y;MSM&wSpNF>UwG5J_jc)@BCu-rDK^mr<{5$!mvXp*J8hZ5 zZ92Ox59lWs^Gu#=0!Xy;foT8k5tW;?)X^qtjs4%=ZA z?apce2Y9t<7`Q3gjvgyMo)4>U3bSzqShL3jeP^4zWy%gAQ91KWM*LzvzwIkHjTxV^{^l_GqSl#gRX>t@Lq+!;QBfa zh&qE~%BEQl*h>~$pN>zHpAQr&{U8)g+X`wIx)Mnhty-(A+bph>@N{avJ{p=hr4f5) zHly0}V;WABN`!TOs%*VrZ}O3vP3^k#c;`D?e?Ix}!$EbgCV_%i9gqDbycmPDBIN{Y zsS^7L<5J&k)tS=PBFmbV{HL?Vj)w?~c^*sneEW8^PI^^K>Zc!2T)ykO@_L!BvjzW3 z>}|gmCF{_DC^lO-V5sifpa`wQ!L8wTvJ2sv?Uwn&)93~9A${-iXgVCqr70RT`MO2w zr*rF(i@O}S4zg@whVLfR4JMbLbxTiH80m&-$;&*xxz`x*{DOrUy7X{&QU`uY-?6n@ zw`36*aD_D^z2vi12qE6TPPjsmHS+(hZVY>|JlYXRiSTces*j;@={fq-)IXBiey-}io|Cj)J)v>>|K@3sG|^OG|70OW<}`g zAB1Y=7g2Xvx5$w{mlW&bUFRRbwJDW`FsXIuQ#l!Ej7lDcZa zZOu5a9j{ObnWX1|ah2`(i@Lbg-soukvL`&7JrZ#%Lp=?j$nHFC-!3qRkEH66d`8l0 z34KO75RxR%?n|6uSHKtqLSGc3?{y20dT8^9xYOak>8>Vy@J6}vYPRuGXd`{$DtK48 za;=_VL*C6{E-x!H@K9(-IkQ+(OdlDIoL@k~P{_lCq)nvHrq9$$jRC3?BUqG@ z+ViYsaq27_o+-1+w{z?KXa{@0b+owEM0{olF(=2r<1kbXj^Tt%yV__ZJA=!oqm3&( zMRh-zF4ybgiO;wVCTB6;tsW~kkBMzXYi3%la+t>0{apwX2+HQP%M>7Dk@XP%hEf-fAzbA>cXE-%LQOYrgvn+H+~70O>{Q zR*XA~Dw$rDd)ZXlFVWBYFEVg~Xg4%+#D!P0U5lBnyqk62)TEhOgra(q%#-~C+O#dgvXLpIt`|ro9_hp)(aU8cbG34w>aw7EnZZO zMze8+b6;lS9=V!!=Cwp;N}}gd;Y?HV_x(KBMJJ@!WsA|z>i~yqLV2y;e5dr=4L$RM z`&N{syl5NVA15$mMjN^asU8UfSpsS6SxuEL-zmfE4D7?2$jtTdW4UhF3|IMuj8K&<&&R;b z((AUPGcF)Ybjra`BrScq!tk2Yp#IZQtAz%^_&n@NtpVsURAf+Ezy92Nv!-+ro&wF5 z5|0!Xc!PFfzDgA8q;~xi$APW26NDoB1r+2uK<{eueB3IiVYcwaEZ>i;^v~*1Q|A*K zioOP8m#qPL{+v3ihfW#M9{c-kMkXd(Osp=8S0hYU;@DYkFU~sL@9+^qZ^xGp*^v#nleA)qB-6cz~)&!s=9 z$$S>)F22&{|IN=v(dZrEy9-q1lcsD-x9E0$n0*D#+B(r_;HSiO=S;vjHNbezp337=i!St_)Ubwg z&$UHTW5`LZP&4q$$)iRGECoJpqy}NJUeOQyw$|Cv}E|)W?v8R~!g`ONh&+lBv z%5V2ayygsQ5TM9@fS9=%HNFP@WXODI4|@}Ey0 ze90-d(uXE<+Qt5&morDoGpM*`(*=Kcy2416;6PZ}Z1TX00njEG@@zbg5PC*p+}s`G zQ2XNnrs~S>V@DPDZeeKEOqr@L|C^adRyU;i$e0_m?g4i1Qq`2Qs7Kr*>8HZ{*6T!v zNQHB*lnM`E|LaFj2HJOALr>h&-MZg%TPSqfk@8x`{|<$4_b+e#-0BZC*|+#e?8wLu z-LoR1vVs|5eR+7B_l{Wx_^=XnO#N9Hm8)JM&5&C+a|{5Rwh%W%>$FOunds9Y;f;#aN;#;X}5PS-;UU5E92EKtsD0qk3c2kOv; z4hS}Vm$dN|U-l)=UnOdu)DR8fQG739W%92?SWe=Y0{o0{yb0PRn#Vs9PlV*Qo#fUlw`zr1$KF1kFisLdSw`e8pcAS(wq}3O*H6 zW3vxb=4(K?k(w-DT_vlDL=U^<>0G)LC}Qj!KJb^}!=czi${!Vf&~}U>o~eHuA3XSy z+Fz_8bc?q%hEO^BjbsQi(+k-piKbuRb+7?&w@7BLvEefvfD+97S~jc(!-9LaP-7YE z{6}Se?~N@$H>=1As9!Ru^;s@Zf1X1$l|Vmu22-ni=de&2n~pS_RysUdkn@(qhF>)j zp_!kxh7FP|ii@F@e`55ROi9l6=#Lh#@c(KQ2KltI)$DCf0bJi(mU0N8En{pzy2kbc zh6N3j2af|*TB}ocexLRiw#X5qhA_68Jil7<$pV{jjWLDzvmam1qc#7q8GueObuPtE zYbNi9>LJH8GKUy?TPNtm@8a62vXzj*4`_kMn1!|{-Aqd5Ui(FykC*DTvgkbzsx$bc(`9N+Wk>@ zBj$gamqHXs0IH?C`yER2=E~r;?QiszyH&7YZeWUW?+IWdg zbQ|UbCC$5N8>WTKT>sGJZK19au5248 z6oD!%KS9fexcL5-M+n)`;jl$%u`KQJAbs_0M(jB-c0X`TVcERW^6SYhZ->J6?btrDe&^!XM65*r&XB+4=9 z0QYd>d+-)bmaLtftL}Kyi;kmVx&LxA*~0O=L)kiC{Px|VL{j503!Nli%fIh#z))@lGzt1uW9R?vq|({Ta&K?;khE~% z(Xgu?b`)4je7^ZS=c@Z1zr_T)du31f87Ux_)z*mhqYd@s9w8;sZ()a7R@M)+tenc8 zXH)2uH*CSIgB^LGN^~nQp=8t38S{E9&2G5*-unE;Q73~<47S>n3`U`Xl znvlJ>dd>#-R~|x4I92H`i_wWzn&Q)nzX~gO$=0R`$SkW(l;6}UX9znO+u*4pev3%T z{*vJJ!4b(rW_?sKRxR!JG!8?r{WU^WPVkMH;(swKBoYDzf?wMfTT3vQbhltz`~2no47)5f?;lMS|cp2f!83TPG( zfhQ}V=P6(HwG=>KU1X}?R-&iZA|mQ{zc{|p#s?yFnbE9-@-E z8%%@m90JcG;IO+l3k}g(ye%gr@hi<-d}roPcmKYfK9*uLNV^sCw)G~L&6~ubBKwD+ z#5&ilyTv8O$Iezq3{F7z?ocg511(2mM`yMJ&|8b~`f5e0ByyoK?;?hikBLUsuG^zW zut}q8GRHmU_SIbMb8fNEd;*6oGr z^~n(Z{UnM&{+)_;`TUmEqYJ!cz-#7TCOaQr;xL4tpV7ATCKPaeNBgL3e0PJ1TxUoCmduQ@E$Npqu@W-CF z?#0>5K?P|jydw*>mj?>y8AA=arRbiVIc24W{o56#mFPWD>>YG}_>&LzHBXyp*I-8E zU%zEI{MV7JPSYF&{rwsX>BY)9q*pW0k1gP4Pw5bK7mz4_k@Eo~JAuy&S6T#i(0-k8 zM{3L0qSGwUp`YKeis@`JN3DybVvEuJg3NkW{va;+-(UXsmHHna88m>)yN2FB@;|=g z|6zNWfO#JmuCgYMWdG-%S!D(!95tL4KK%0ovnBfc|Nk@p|CRaw%54@z#^uNMuvHTG z?*__0cnrc*WN_{K5VX%Zk`s7y3Nb6_1%1>%vRwEu`%m5U-lHdo1(7lNu^sPq_WZBT zssHVg_P>3ijE2rz-0Mjq8c6rI^8Ei`!}NdiX_6IS1J-z&sVv<;_D9V3__Me8K3u;1 zr#^rGFa8ory^Gr;KuY*W&X(rCcDDcC;{JyxMG3N-AE+KkKycwfxo_nDy(Rt+@Bi;N zfN}CgS0uiZ9^uAc{_`lzD|vsPN>VvXR0ZURlo!POC!XZ-riLX*MAO`4(=Q*Z|jNJY|L$qLk|# zqLisPQF=pzI(Tlh^bXNTjCk#8b zQzj7ltRs*EMsKQ`i>g>2v{C{7yMee)mNBo>C?i;ht)Qy*b=11&^}sUp4=fq~;}$uu zSaniQC(Yd|yORg~$6ZGDH~K9we1E)0yfY@bvyEAmgWm9GB*hIa#f0a! zZI^BbH>(+}RSF#U^D$08n|S7(U1I8R4~UUc9Opq?ds|D_9@e|dldfXy8JwY~&Iu zaV`Eh71*wM1%d)A1{4n^-rEC<;vSSe7&s3Mq)n-TcQ*$4Dv7M|=LwCyg`uv>#`4@! zO+?BE?qW^!fi6XCk{W;hnEHDMt5p-3YIVtam8v&jM821vmm2fOMszPY8?R~dUp~f> z^7jr#0{u)WXiE#ugW_Ze=nwpxTq8P;@o+xMzGPzH>8S9BI*$5-^>N|S_jgBhYmhfL z=Q7ag`nGbd^|C zxonXol0}~rO2Bi3>8lzIOSN?yPwS7p6o6&{emCE)YQ<-m2mStW1+yaK>WAmma6{gP zAZ;&zq%#3cbdEtu{V~t9L$c+Fu4#BCMT=(#)sn-3@p%-DSan38?`81b#iRi|v_`)e zq;U=brJZzSWza%n)*}Z|aHg)}4wUUrkUQgAQs+Gs^Hq z)|ViC+ITj@UjU+qFklf@3-~t#4_%0}|H}gd_ zo`b2dHqgE@%q8M^HEMf?U3qLH*#J%xA+#qVE(+kE>+d@|&a#>YF4o1)x70PoQd4mO z!m{x!lY##J)6@W$7@{~=Yw(9t>AqoJuQF|t0{+CS*DZ9nRoG4hX*I)Tw&WEqGaCP6{uii`5jzd0Y-#$<>`&r(FkR%j)-)O|kf+uJ}n~r}18v`+R{-t)iB^J%NuU3w9H(7g5v7YLy z3KI$9Sm{%W$N_5vz_Ys8*ve7N%~R|pH!zO(G0*Q5ygS*}cE0$@sQpcu*m?1En(apL z`7pwsUgN4iEi?|8Ae5zK6T*WNsU1-UsNom`{wPn9xzjh&c2CldVaGlF#n-OZ8YC>L z7t}vN=ew-G>irG$mJVpCpKlH~gvJ$kmk?MCP3`Y`&QFafZpAuu@3jVaBX0axi_#yy zIxFBCA^)1XRrkT`cqF|>7N1W1r9-v|LcwBZRqT*r!(KuLzaB9I@GkACmjZKxFJS;> zCNxk9bp%)?MxlibLFogo`7=-38tS9>Dmr+tQSt$YJx}JGv!zg3-vQmFL17;lgLau+ z!P|^)r{iBoiF=e{4d7iAAKLOeZm(3=xn8}#zF5;-4MMp_>kUzQ%?=A`rXA_^{<7>> zj49-dChoxuJ;u~aK{WB@bUJ$;%;?uTEOHy)hFy?6_jO^69UqbgWMBDSd`M@wJ@2@i zwGJ$GQvup3W=4nIWg~Sw!__~FVnNjfyEzKk^gPx-0<`(PP%&k8vU51!0d2-;>6`H4 zNiP)?axtr%&mPbSCe!^ImOVS)@P9G&mQhi@-y82&DGBM65|joJ6p$7H>5%S}Zlqxb z2}uFzPC=xbA%`504uOFgVCWu(8amGSJLkXFc{{J3HS2lqd+&W+`}%;_{6pjZ-iH~C zi;-bNE{88Mhk@L<1Wu9bccpKs+-Wr4`PD2Z8y!mRz zBl@uz4x>;{>euv=Nai;4@ARh~JfX~HssuNr(^?iZ^( z4cGNA&%1O-_{R#ghe0}T{`8SVVaAOWO1wV)ExaE1M|4>&L55=gYK)i9LJ(3E4&%8* zjeZq7RfOgpo+bBLoS00iJJ8NlS+dK5ez6+QVaNU}58v|x*1&iM*3&8wY)eI*5CH=} z74`VPEP!_X?~GC=jGmj0vvSC5bn@R-ai)EYtLp{s*&f^Y-FfyfmfNGaED_()d~*HU zqq~bD(^zW9#Wy#S7f~S(rrX89>i@g-_98PsCnR8A@pN{~EB=1IVd8n+Wy$7n+~vO% z3T!)(ekvX)^$iJw7S>guuS*j3uk<#Xgtl>)BO+=n%s5h8PnmEF{QR)Xd_W~ifbG{L z9a2+D@;>voSwrFF?zXucV%bzaETE~#C5dzY?sY8zf>vjt*(T1eZZoxszQ2L7T_K0& z8-_eM{NjBgc>ZMdua&UzY{;cdZj^Mk zim?=Q7)eGYT(%cPEpP)E)@vme*mg>Yu)j7K(6q#?0)3}PN@ zU#2kO;34~!6cXW!+*O^HXcwatGhbwrRd0)Bk-^jlF{jJFs4p)~GzlI&>uNyE9ZqxA zG#I;xru=tCAvv8glOAo4j3dHh5{LCRo3*^adVkofWz{l_pE&Tso4Q8oX!)xWQp^;S z;AM4KIYi@C4_fwheAsyNwh8)Qo;2nXu77zn!OS{#vE6Bc>p@(55^>~`P7dFrA0R_J z&D6rZ`vp4)h3#QFDK#rq2{d&c+|+uv?Hp;q z2>%PLPd|NL$k*+EnTWA75{|~)D_=@RBA!jKYnfoIW*ELlkxi7a-}xPhbWEe?YnsOz z5<>|)^b_}$@a{D_EOPe?-?}7b++Rmeoe~fm&JDs7Zz(3c#oTZXea!xE>yw&x<*VGP zI7?l?Q)7Y_zR18&26KKMI0=uGh*W#4){O(3v907$>=rnm(kZkrft+smn}+|n(81oe zZ-6B0cuqgO+MlkfF~GQFVK#SdLh^&>RWGW=q4h;ersK=;tRbL{%ZMG+{g^?7G#g_KQGii_8mKHuIJprr5mZ%`w5jP`HwUM5> zx{p03*0ldR7Tt%;@KB8;VWNyAX83`kg+Suy-k4SPeFt>yx>7ut`P&`9JqTr1#S4^w zz`nUz@EW!yy<@SyVX-~M1Q`BIVwCpCq7|{s8on4Vhp_rq=X?T1r?J%GP25I-r{X1B zD=+%GG4+gJ$9F&PJ6kGg*NgGEPwPs-#}NG@j9)A@Df%wV zon1v!6m^;0XA=cttE?Fk(v)jBN!^3rVZV;9PHi-i<`@L5f&uxB_fo)vj6az?={w?n zwd*pi(SL??FMT8yiot71BakjA$5~F=42d`w1LC2+xf5HL_jCVv*g>7X(ipE2FeZxGZGk<>g7qWaOp>1D79eZ8P7AWN+A_ z#;T2hktErRp9M3oYz39JMnxlIFd4azsSqDBVh(0eyyPSj?p++F27!Hh^wnRW)kbR> z)WhlqGyoo|J^ed`?iwbJLq{OiGIj5zS(shv=7D#MZshYrfki(td|N4G`kboyaQz*+ zt4rs07_V56dHYQ3x+sT3Jtsx4S>$(y{U3U20nJwy;%n^b0hMy|Tx9efOA%xBz(Gi^ zAd{$-uieO=L@ZvloCSpN5gh(z$#&D`u+QIX)QL#&5Z0-0XtSAe_(-6&J3;))p3VTr z(@=m4q1osF+SF|Pu)4?A2Cy0RYiP6z(mMRPF?myM)n`0G!#LhBIP~_&8WGwpK-@eIh7Ml8*}P^M+y=u^YIH{NQJcG zCKytoDXIRy1Qu-QBF5^H z7s=GPczmfFBr9-=7C{W(9O3CM=ZMye8&$0UC60Pzy!V8W;ygf$5ZcWj#%`VR-D0?F zAZZzxJ*?nRL-4aA|A}N0%>rj^D+vkkWlbNM2h=oE-ql~)u8ptFFlL7)&y|7ZuiZeS zN0IYxm)&7`GF`9*-D0JUFdHBo+OM4mYg?DL&OqB|KOB&hT4Gb60H)I^} zcf6Pj#uyTZ4U0Q5W2hKIOcSHqLg8J2OaMo0p{2N%gqS66>j+pp?6yXWGz)_szskmt zW@XlIIZ8HpOy1j#WkH1WxUClMdDqol=38_-?msSW^f)A&H{_A|xr#y{ck&URyG5lQ z)_(u_A=T+z<6Dt2tX@6%B2_#I0=o0dp{JOh^j+&*2kd4shA|r5o}({mA9mJX?3tpz z2f4LvrhTtRjQlh?+VAK{^HX2HI~b_}>)n0MTWsApWsT$}+-tnK>)bnvWFG%n9X394 zFnA37l?m`T<5{p{B;DP-VaOA23BDt!elZ;4PUff4_Xeww^_Bp}U{FxgcP#qF))&D4 zb_9hs%bpBm(K77$ zq;WsoBr+T*o`HNjA=`?*{KgWAsXHSf^b?I_L!ip4_!E@Vl;UIz%HRe| zB5Z10UD1g<2y1&N86L2_x z$GQJG9gDv~upBLA{s8l5wPjtnnhoJHO|5{*Q73+=evziX#ZqVdEnQt{F5A7|2=^Ag zy8*h8>}gU$EECQRz(jC*(mE)@*m%|F-$YTd3Hwzp)$TYmabv(!$GNjqb1Or@&8MKB zUq_Y$RCrSe+!QfHVLA2Pd)O;KAh0F4>$dq`<~IMbACWp8hra{ zF-PEM3%|9q9E#dy2O=0mwe1rDC{>Avq`C97x`%4cG3xw2h+5}y zIP5qD%L_KD4*jVgO}zMJU~ZcN2xXQ122bbbKW{(2U$TnM2L}!k3vAUJyK%Nji+L&X znl>r|ZsKzV+443|WgL1QMq0mE9n-sU?z_LP@m*-US>`qD7a5E%6B|nz{BR<1Tgz$4 z?J9xzT(+M%xnEtgv~;of($=6lexR*=V{7)0o%8I6?OuL$&6}(@Rj>=x6=eMubGQ}|C3@XJK!a~T1+OlGJoIxt2 zysrxor4OVIfUD|FYQ22CuvkpCF&QRWVkWz?#9vK;nEts%PbTRuW<|9AGTeBtvt2lV z`!^Y!{?${PaIo<~f|ApcXEfml?Gmr%JbHP*GaZ;s;1ukI6V5u$5flxVk> z;9i?-nV|c7jgs&2#h1(1%fy&e7rBrSh4%M}k2VYF+-7p-tuzu4wKsz|C{2lsxA~0(5_C5n+D2Zct1MaE#nb{Nc)%Mc&fej zu#Pw_*1}E<`*4U{zDm2biHs-2BvbUpTflPqtk$n?o`emPjMzl2_Do0*qCsK@c)j4oXz zZ}3CK*>ei2eGD61f7a{Pb%(bRcdoTAdN@4=m_|e1S6E;jA^RU)dj0uXev{kg zjxM|3LpZh`1>q5bzB?=6wO4d+U}^<^2e8iktd|bIdYT4`Yv8W!@)CN#%3(GM{%+^y z?mQ3U?M{#%t2Ks)?cTtaQcN_#ufxF3F>wDNlgUgU`SPolxiB+i38nry;98`z2;Mm_ z@2@@$KdX1n9PEE`b@+=zXb)+7h1qfcnF0)8PMcaLv}ID?mkcD)Nq%!AvE1ZR#X$p* zxK2qbsrZf0gF`4$l;;$X-^nEIiz36qi_@X7^LobX!FYYKyXVNofCPmts85-*R^lBh zy!j|f2|Kmb&?0bMjzL%Fct>>iq$H?ueF020$=xWLv(k|X4N@QVBKK(@*nKzyb`Tf5VDoepc_ zRmNn~PF!YUId!u%OALx^dP#J-=k=r2Nj&is_?I_tKo{P{T%S6kb~sRmAf+Wm3Ki3_ zK;OBu<3H*hm!2X*W!U@wd~rt~tq8!#>n<_gWf8$c=^$$>D`NIsF(qDr8iII zo65;zkxJxW2nf8&GaOL<``gC_&4&#=K)x!i^BGP>JIy_*YC($l)wVo@yu(VLNMIlX zw0UbQ8Fw6o5=|I=s>+6&A5P12HDOBrl5@)(r{)elqXHdfmedChjhxBs8FE}a&Mysm zRsG(?xD1*p3G_nBT)AUdJvh~^`A%@*yu5DJ?PjSDb8cNZ@7$}1M8q)q$8l=o`61u_ ze<2@kl!k^=fYSlYhuyiwK6lrTk<)6|L=2ove^N(ajAPzVD6z|L)vJ@DnhvJziG0ur zc7n2QdDOsIHepAXOvmtdCH$tn0Xo!9#}W);j4}1CAwjxCU-rX2HxW>C{;ZM96^P0J zb34?Zy|}}684RUZSP})BKYQy~2(}?x`%|+*%Mfc_^eO8`QloMA_ho!q4)Ps0c>U{) z!=&m#@|#Kl`$p@1$3Z+AEzFjyk!r4#g0~hNRH?Nqq-7vK^Bdh5qY>k>Ghn)6%c>IYlM2kzmI>)nl)TYg*kVznWPU zFq}p1lNsQBZV~9bJlTP!^M zIR^QFm3HWkCNylZh3uM&s+Wi)nt!i_@sP$v)+2ny(a}f z7)-(+KX&+`Hv)+b{&OkuGzFctjO+BWkww*PN~`Vz9R{rUSz_}ilUx%|E$i&lFlg<^ zxjI=7rf|(m44y1?Ru~wDk9((i7|#gf-sEp~3K3TW)L7A*eLUS5DoG*lx(|?HlV#ZO zZ(!%qAe=ucYs6%su613m{ce-e)LZshC~qrf=$7rmik``TuEy)4Ns`nTgEx#a{9>jL zP&z-jh|Zt*1J(>koFovcdAGqCncOY-GPtnCFmuP8rc*#=fWK{kt881CpEzS-!K7_E zB8ib8iHkxZR`eh!JHN+XCKnp3G!r`NfW zXZFCP+Sq)G2)v{&I|F|N7Pu^(`92iwPt`--Sfy9;cTm0;jhDh87FBVE59lb?{2wN5 ze2mwO)}dq+J^ zM(G>a71AN%0@RRzUfbBr#y&#RX%9xI&Gg8{+GT=1Q=edF#0Ypv`9;%HW>UlQG1lIb32IW# zSZJlmfn!KH`NF;2&&vTod~f^NHHip`1 z2sT;iBRbXRAvleC-GxgsQ(<;7dACn$$lUm{C~Q5d_rrjcyzb1h7HpZ4#(KMZ{`qS$ zE?fOPw-J8muDV4+k5=&qd4PN8RB54fWD&Hc0O*H$B>Q^WYVN1+g!=GLlIoJb-ra9J zcGp^ut7Ul+#OFR@KD{~~>dlx;g*#Pn_Wt=dd^VJFtiCF*>HWB)cS4UWHjKwl*4=Fa z68KH$FuN#!w`H3-eQy0|{E^JfJv8-!AaJ7h%i$QblxITBmWiFCG@5woT=JTsjYPF_ zn|Az-=NYyDaP6(Pa!ZGA!;|F>oA#h!`-@r}Y zo6@Is{B+#t$W*sNoucAy4Lq>1N5P?i?^pA6<<9Nb8Sm%(gObyzlIpEac!dIokOR}j zqjCR+9BeaHFmo14CBs4owYxRfK${$0ucX1Nu<0u(G4tL6-KnK+0WuCAoG$%%Gb+5l zea>ee9M8THsDGgnjjk%J5lH{%H2IhTd37g8Vzo zAs6Eq7)ndxi`l}sETQD&^9tjIAZ|Wyyo>$Pfy08Ju`Isq06peZy$+fUTILN8I*#j1 z>fpfe-sYNkCS3xxyEtr{V@bi~GZ3baId<(YUG4x%Cb5^Twg1*J>x+4oBolBNu9_#L z!e_~~q{5wQnHygh&TH5ZlkEyNOuZV^_COpi)Z5VjN(7%F#Eo8iSp50=$T0{PZn^o? z2_?%6Byx-pk5hcKHF1}%#%uZp?*(O z_)37!;ez%C;N`8j#&+h9lX4=8>+N!Sc}~D`IMb&19YFNHnuc;7n1V3ct1` zH0g)ae9gkOO`T_=?-WHd2<|`3*13H^ih1eYoKkU9ahntc{WcoIWy#S$y*X=` zkj2O>wI|Zov(P!I;>2pLDuCN7qrEoNYG}eYrr;v=ou3-<1`gv70W34E#^(EeSBcKZ zzuP3CA>$)1TuW`nHa}umbTN1aGkOtGvrI3k2@vvkJeEi!vnZ#-rgbw(F7T*!YVLmd z4mVOt^rUt81~c~~T5S_@nfi$A5_ig*Y_FJuh`VX>;*tca<>I+`_ioS?G6!D|iKXjA zd~=+0C^|rQ+&2QzV9#-V zjiDuSzQl8_44VAaegT|?bvcd&IOUIb>9lgp07VO5I_K*zX&w$%%+{SS3Iqhrciv*` z6W@mQ0t8c#A-62<@>iJ110BQ!_fC&&85<8_KZrBoj)A(|RuF zO15F^)sx(JS7FML&0#nHZ4cZ_T&YA%<}aMSHy!}qij27ZYPgP6h^YY5Psn!$KZJx8 zMsD~p@Jo^yz|p`Ow&C9sadxHpMqkWRF-@$;PWBkm4A7aA^EsE$yPdqv5B+`pYT)-5KRu&h3mdgO@E*| z^!L&qF(w<~{YdW)r?pOE{}fJB(MGGyXB93*_bnIRbmh%-^c#an7xqB8@rr4`F9uEJ)A3KYW0F5wnjEf{9IyG|Af0dM>mD2NV}4l zURzo}e+UHeGL@V10y-YtI_9)CQH>;UVpm##O}qoQ(&RLBZ%J zJWIaJqa>2rEDbQ|PyfWp-4`mAfUY34`ftQ(g3%!3^4oEk@8QmgW-RwbIlw!{?um=x z49D5v#r_PCm*cG5hA}e0E$C;dH}9}n&4!xe!|!q3zUKXas;s3UvHNvCipsYJP-o~U za;=5&X3(dGOziXuJ_p>mx$$-Sko#;?dk|Ea#pd<-hSKRFSl#u>82$3haNtzSb`{{A zk5@X!47iQWl?c!p7}11*Q}61TBArZxwRZ+JsEu#DOgqw?w!E6hTMY$g95~QrE-nwF zoTbYNHiMV2?v~;*nL`UFT9)|1tmB&1&314->)c`B=pMs!orp#|9?1wA-uZ@Q9*#z?G z`##1ufMt%P4oVc>i`Fvlo&_`HXKcm_qxvT)N@Xn@o2ARrql< zALue;fx$y=uL|wM7x7}hKAp~jNog~_sb7H+H!Q6(k zd00O8yVR<^N9+-=g+M9W6KY!`8x9oqJDN(oVcp+_R%bl~PY(H;GN3cafSpEHmDi7X z4D@@juy{WnJXQ%Xi0l(P>KxD;qjE~aH$g1}3>PrX%_6=0K9ja4G6}*MX+5R3-fOiM zf6Dh#6HSMh>`_Ez?8|JFn&?S&JAef;VdRb^3rBaW$fN4X@)p zyW*@#xgS#(8#VH1`HlIRNAP^jfa0{B7!(5%b zVnjJok3jbTobQ?~KY?p9*%quppJ_y&P+w%MPBs^D}QfUQbAMa0#NCbIRy#jskk71}}V0|P*$EB$`$ zPFPc`DJhmdw;wNx)(rLC{!MI&O=~VyC2AzCraBMmo|v#*$5ngw(fM-+w6oJM?}XRt za8RsQr&|^uyO4Ax)R1s(Ev@3R_*&F=uDJGyUhFCFXG=rFYGaqS` zSzNS$Z-qBw@-w~BT3caAP>tO0)Kng7v1PbgzRS-KFINm275;iIZhY=qmN}EOWzH~| z5SSrk`AKRt5KVDj0R-C{Zp!rR=E*L))ehW1R!MvYPW8fkNks+JKPOVl=yIo%1nRK{ z^GDCFiO=2tC(y2v7GPrN-M2n?!d5dFC@p`5p@J5JQ|sd0J_RMS%iXq#MQ-zolF8Y) zJX8m}9DK%*<38(3{S8bqBo0E%{+`wJ2fwE!dqZSI>Meva?Rcp2_w|@>k2g%aewwe# z^r)jFs=Hn5L4W=5ON+XQ>?e76>A=bV?h|TNF2u7fp_Ttx0x>&AyW`Huy5CJ1*KCnb zfv|G;&>DO(0sXwc@p3$yKi}T1@!V(>my}~K zc5aDL=xIXI>C(y4+9rauQg_hrwr4!H+@XOImoXZ&A#KsL0II4g+Y5s9xPVtrB zcrLh@!Me(E7Li>2vQmzyuZU|x&o`*H2<*^%o$3>)(N;4`sw4kh9j&pOWOud!h zIePtn22^lN2xSe6_d~pVAimqx038!IqM|O4gH1<9hL{^sTi=qz#_YYMTpzCYOn@dY z=@-g6m0WhZ-xQ|zKldY@PL6v_=l)CgS3`y|ZXKKLN9iVns6N5Xe!cUV0A^UHz*GK6 zm;=(3Ip=b>podJou`CT#ASucs;N4i)6>*p?r##f`c1<4+xU;gAmZp5DWxt&90qskfltBx2kY+G zYj=R$wN9<4TRNpvbEc*BcjHAxrr;9$5s?8OhiZXZLJ|QN26ssT60TQO|N0@vIVwGg zR=HnUiwr+RkaGAsJSzWuzc{BcUlxkB$~RYK=7{-!vjsK^6@9Ju{Aw4U;4o0;{djcO z7W+|S5W>~#))F|d!KU1^{d7E7UsGNVF2gM^-|%x^rW(d_@)MMenW< z@H5xWV-FnP9E*n`&28|~qmV(&?lunUzM zQ@C%!JzYpZ>zYOj(h=Tb{VI^pIDaCV+#(^;amAR-QDvhzF8!(&>B7;c%0;>u`Z&#R z6ixmih+WGMjRp=AU9`D35ME4ay06|5Joa7q7JKsrRK+p4@{S+!&wN^v#Cv%E!ARTd z*4sqCMI0J>B+jNXh<0;LCiODPRbZzR!zwhL9Dlx5tXx{lc zO}6FHVi#%cf5`@r%1F$?3;b`DfSlX)U#UM!!He{__QXob`VqYzA+EkPO<-7j<&Frn z^avfZ6I+|J_KZvJ?@ZU#I((8X;=;KCHjEt_s=g#HLQMfItfVk$a;Kfl6jEb9Nanj7 z(E%a8GEsMVkmez6JHzU=sfny)tXgU6X!&DCxnPU)2hD)(v21vqn0u+2uKHd`f6--x zvGZT5cPD+}ed{Lx_tQC~6|zljzBsRhXBjAYVE8S%W;#afpi`Eu`ObllTi-w=t55+jDc1s-#!K8%J?3xXwZ$K8x*K?XH zSCRjAYNr1S@v>VcEKb7;Kzn{)jT?Qz1s;1VLph)mq%ce}j5M1z6*2L5qI+d6#XJwv z*x=E8-x@s!AG6`u1>anP5ZpE|QhI)rXjy^{xJOwui|yJZ!DI=$Sz=sN2sFKrmcW4S z>UU|Lbbgvsy9vFzM!V5`)H~|eDs5!upnxVZ$J<7S;apK*2do_wLm!VW7jlf^o-K!D zzx)coPs~@(r$(0Sdwjpzs6H+7fY?*=6PG(oUMDC&lO}ZCv5Zim|9J z{|t`>b`4rKZe3m{n8zS7)a05cV$oiI*}+#^;zFDeJpz1Y%w1bjj0jz3 zyH{jR&)&?jz&|^C8(e(|wJH!iyJ8~aMt|_kPEW7o?5&99`h#~!M`n3UF1l->*R!W!Fce7S`Hoe24=KX5n zf-O)}2B_`hVg|WPH`ToJ+OZCqPpK$L?c~}jNL~t^QMEJ26X#q#M?GjN(p)RE;Oqko zjxmON|38{-B*{&WfB ztYLZ?2|?rCG-v5PF^(*G&Wnp#KvDZnEB4&k`~&cRBu$t%y^Zra<8jJL#{7TZu-DqN zt@MiPCMXSwN-d_pNIE<=zRv}k$Ik54zgPG+XnwW5Z z{(N;78kkCzXV(2h*{y#lf?=_1tnJ(!h;HIE=W^AeVZXIlOUPEFnJJ15QkOYUe%tU6 zT4JG27rUg?$cDm@9$A{*qUnFXZ-c0$5n{w0b&89b_>tg7a%O3PV={{8EPm{%KbHAk zf5M>cJWo&|YK3KQXbc_>{uw-Bp)D5a$h>bht8Ta2 zx?wJ!f;GOHfCP;2eIsfc;OPk4aF{C9x2uRZ@Gn$TsQUXyd~x*W9%B7%lt<8OeJ|B zfhfDuu30o_WIcOX&33}{`SnxUU(3uSjPtH+ghd!3fRLZ)Xkg{#R#S*RgWsS>@;1Dq zlJY)dCSu6N$ZA*nC%`rv-4IN;chQt`R#j-|sq}9QdR zldz&q_<8YjJKo$NM66|};}2DWxb%AZ6^1g6-`G3QB&P3Pr;c^}wQD^=B)H+&q; zAs4v{?w%}U#Sk=%9WbJ_oZfPR6gSND-0bM?czrF^2dTs4n*6Nx%jnUn)rGnrVduwD zliF^_WWBskDf+oH7y135%Q<-uZD8r!5|KSw2nbv*@T%`La5EeVx*Z}rmW#XQ=4nv@erR+4*1`oYG-#F9XF=DmVED3viNTGj#k|;kT&I6fLRs7GbBcxjx>! zedd5Z-c7{Bd>I)ub)!4}J-tUTBEOFMAr6=S{q)#P9nKuGR7~J%r!UUiAr6i zlDfGiJVw{KK_;a0g&IEX%W#9ooS0qcC0q^O6LfFs;FBmu*;U-Ub?O%zNtzG#Kh4M0X5sWN32GH+I` z9M7AUwB{4mZZZx=KJ$7A$tcqt=Cz~9zY%Owu*Fa%>h0K=RIPajl-e}LxZSyi4SCsX z8{J9_z9zHYEN@!-ky8+&#E1wAG*MIh>%UV{b{VlB80016UgbW9Ii_aiefo-L&ppXuF!)p^(X z38r+MpRm496UdS;;r2+jxm`@&s&m~pL%&X#f$0BZix#B;gy4rlgaM{4q3LYR=I{|nL8uGq4t%E!9En*f!b*r8gS;!^SsYpXRa z&8w=YM-V$XsL?hI?IFv)KWOoeh-t z&4fbDfn~f!mUPIMznqRY3N8;Z%Y2u6JD*C6#9a5;mz=D7zf!kQnLnqRWo8yD?vUO^ z8Z7&*$E}I(!j`IYg#XS%#Z^?UQN?VM*odU_AuPZAulKlX#h=f$SOYQUr`AKx*b+tI zl)v-UZ1$LOpdMm5{!FwkNBqmLx9oumE?E*zX@8#A;J-}6NQbxgt!s{CZDmo28G+IU zo-QOEyWPc>tzp@vutR^XI>h&#gyi!_C6EwFk7NDB4?OSIamD5!UrIvWy5kYj_)6cp z6G-=AhsA3?vo2s}{)PFt)og@KIC{SoWoIy7mnj`%sGlOP0qxmqa0bn{Rfub1+VV=o z>y=O$b)+z+A$H-GC8MCLOYD?upB+y_OKrMRJ78#^X?^>*=YU#t6bpWrnT{1IYL^5NLU^YqJ|r-Xs# z>@kYlOO1k0US!#5-V>kqaO3Kf^HvLM2&(Wm9<6~zQ86GfN zj`B#DZwbR34}P8b+iKIuu|hOagF}{g5wmMmb~)ZgGJ8Y1SEPTTzXlNyxiN2FqaXYj z$iH(c#wS3wnHcKvY7Vn$$z}#DYtpXAc8S#?$<0h(6Hjx)tMMtHZ~>PNSRK2lf0*`% zr(Z8Yt6%!AQ|cgo-gFmh2D#0*$TtQ_x%Q^{n`a4>px=y-D?H$m_-S=Zy3$CHM3Nkq zE~5`_I)H;km_Aeb{Szj?c22Ybe#pChv*}8Cq({2b=yG1&)HYC^1rfiE;#41uqG52L6NiuB>LB6)zg{TUe_ z*5IDdBJG5(E@OeTC!2f6D6sa88li2I%)S5d)ixxdR6V)~5<3|Cx9V@k$am0U5m5b^ zxZX1-h}u~`n`z#9;K(a+kPF57$2hHQB!NoRcV6}?ceuF0)4pjs>WLq~X-5W+b{5U8 zi(P8Fmlj4o?JoLwruT2;2VNuUJ$`zAS~fSTR3LAm^}W{Xa$gO7cM_dLdB0n_c4yj7 zDZTNnQTAg3btbCgjn>>6+MFySLo*JcvnFp#UF7K19FwO4ml5rK$I z_brL=VtW&*wLti_iaMSxg3RqOKc5U`w%68EmYLV2(cU_`IeSt`?pPVAU_&0KZPUK( zRlN=HUdL@WCG_;iTNJa!Yz7*)I8|r(B;`%xW3=qjd$2JX=OaHF{3Xv zXfvH)F%6n6nDcYN5qQ6PbnhD(IK}@La$|aj#WVjpz@RJD&0ij!udndZ-2392@s&o;E7UD+ z*7Zo<7?P%W)>1}HTxp!PlU@;_jT!vf>CZoC0r`}TQSXc-b3KOIBbg#XWBk9u*)e%T$_BwUTNINRqU+5dgjHKELPVuYblnByf& zt~mrWPBB5}*EA1$z}n>6zLUH`zZyJZnNyEUwWmZ3qruK$watLCyVJ+aGDaQO*XiPD z%70!n5oq~Uc>R2Te+1mEOO4CuwG0-8&~x|13R%94;e%uAi0E?Nwe~D+ydh~@RUhr1 zh&$Vd&Ej`+{`TK0me$}wG(8&<2CVJTWtE`xJ_+0y8-vF6fVJG+opf{K_G9%hUwS^^ zNWvL@D~;E^s|sT%t{L4*^HR-ov!mU7AUWN1Dan_(0}#`-*gPND8dJ+SmbSTK?8O4V z!5Pd#Q-CTC5J1zTg7d1W2qge^*xs9mv`DhSg zMSsWn;d4j(?U-NLO_!L-Xcomb{_v1&+*MWhotUb2$9HG?`sH*;0Pz2oS;9+?Ncj4YEztyo|l_D-pskLk0Sqfz$HB? zzA*)Z%)e_tR?v2po?8Z*=C0q$7yvM3g$ME9rFWDkoh6==DRqLlr`1TfE`x2q^*PTk ztF|sLyQfZH+B2%#kTIw%N?KU{Edh8EPr8`nyBt^-_!5Q~@P|ey*D8&FAl) z9osJkE}EdGcKn{9!Rz#aV4eEC`Z>2n1B~JJ;-bJff^ch`3e~}Q@haNq$81!)D^r64 z_Z-L-RxB*ZzXV9WXbgZ}@+-dhL7^=)axArK&V@DMDxO9&9$ zgG+EHK!D)xG!h&Fgy8P(?hQeMg#f{&ad&7MXyiTo?wy(YzTcghJO6xtO;uMHMc1jb z&pvyvwf0)idYwiJWwjPLQw*^gGr7O+xRlnqkxZtixiX+dxvdR~`xIt79}!?fZz-qx$6 z1xQg@bt;Gt#H97il1s?J-$ECJF_Hy!TO;^K(Z{%nGK?&bUE~^UC*o5~X2WW%gt!^O zfhH?xM`u{#TKJGTf8aGYfE`pH6XS@pI!IX$~&2Ih`b+C>o%fp*iMM2^#^2siq} zg}K(-?}k5s44t1d?*R-V#fomhb#s1Fc~^#wJ_S1`a_LinOQ04VBIL@XxVG z(1muhG^fDmLEL2GCVThAE~IFvo-f>!Y6<=F;u2zRu$(MnfX}M>JCkPFzFlv1E=;!i z(t4|j+zx1=>m*y8FVP7HcF@mB@`nXM>}l5nbD5AX)w7kpvYn3S?{siO_-IvaWn$h3He}8U}FS4z?GJjAp9?EaTZMzJsfhbLXYy3jgaLd?4omj=l>_;a-#@yq6{LtCIK$AJ95S0_thDJmPbg_{x7hX$Hqium zK_)N|+xm+2dL-ErEv0qXEthe>cTk^#Z{)?Of88#| za^pjtUhP1}aG2Y-cc$s%bqWZ6%aZ%L8a33cIZ=aIGt4uo!nE!3sKP`8NU&0ELDel8 zLEc**Z;0oN~xv(?b{Ur-ejT4RCdQ)*FlHygtF` zzy{x{a8|Uip0a=eIa5)UfX%3{%pEYxPqn>EQ{ehDW39k6D1 z-tUg2zs2`lUsmG4v8w}FApPugK7^>OyKjiu9$jUkOM^S!Naen*FLw&YUQX3DYk7h> zV7h{dCPcqa+9al4)#`eHI^+30kg$DmN&Dsmjg3>8NcQ|R6qAL z@O%YrleIj@Sn6sO3}caJI&bGPAJsZlS-ok9Ykx5Q5Nfd)3U5Sfp5UgP_GPYVUgtOc zK6-_9^L*WK%Ay_RdSzYueHKGI@v0)E{6&}*!igS7f- z=!C9bplKE%Loi1Qnj9(6p=-q8wecvrm+u}tojyQ_x)nAa1T3h-0R^+47$UbU3Al~jSOTZJL{6S0|<8UoYshYg{xQk7y%d{EI_v4iU*FM z;1f-PAYh@Ri+@A7s-+tP- z)W;ViaqNW*V7Ox0=`a8>7^JvvPQFEaA3W-yKt9>g#i|+OsTdX7Fjh;sza9M2JuZn@ zbv}S#7xBi_nYM&P_uVWphSIuCZUnYjB!_=DViH`N-~TEN;sT{X_rZ6Nq}%RYTbNP^2P2YDVh#L1QDw?W%g{a zAZtZ^*;IMSG^dW6L#OZBdXFRCb2-7wWCo*>@3o{4a{c$2JdU|e2+=M$ZEJ8vOUDkM zLEMI8tP=H5T#{kDfta%94dT$?V0IdYKHh#7I`C!c?ENx;U!2PS(kchtd&)Mde0LH! z+iaI%7JMgjf14{`)hz3IA_o&c-&%9$J3`i4cNoo-XUaJ}3QAtBrh3?#<L) zNcq#cK*4p{qITd4cXSBHn%Pn5rp~+>bz*t;#=IBxrISmjuKoFh{MzdhjQ*>B=o!Cw zruDqKBDttXZ4$pbVbii(TC&Nb{^u+P7~;kkVL7ELx@r>w^(!4PGtxxggSbaPYe(RQvSE zs)3K84<_@H-ubq-`HI|DujyMv{X2BF^TjP2N1G<>ulr&`47!ko_6fm1x-&=UFN%DA zv|CQQIZH7cJi@au$QEbjH=(Ed4P!&1mF0$QV)b_Li20y5sdB57#L3%Ex-eBAY9aIW zK(2Ue;y)suXDFLTBz9Gbr>Z`tlojTO@2x0Fjw&ENph6N!u9yKz41gEUVZAWT4u63L zT@3*rDN-U-n6P1wM{~2fNzH(;-lG0_b1)mAxR)$!_O;*%%v;gbFg8&)Cz$xM`qrmF zp1pvUi#_zJ$cO@wje5i zuNjk2Q5ruVhVmRXtt8hAs40ZU=Fu~_$M+yZ8%(6>%;8bYo>xGbJGlgKz-bXLFqXw@ z4xtJG%<2pXUEmrQd#pYyfb5|!m)AVCPi-FnU4#$bbq7N>LLCqTIG$ah>T}qaic;h} zTUB;AmSD8dvf6)pLWdy5fW$U5F8vt@2|HhyJUeE+`pngL;LTC;eCQ>EWj(-QVn_aI z#j8UsFVX7BvaAkQRZ7}bdEyXs^l`Vv5zMNqKG%%JEh zR$W}^-LA=Nm!v%nLBqZn>Of^yl(Mqb;5Xqbpt(>S@-fTc6jPa6<*TVb!ZAj2ubEQi z4{0VANW&VbCsM`L`^@?M=hI8=Y&*VamIudO9+2cWw$WDB)><93N3C33--E+1{f%R= z52qjcz1V{?7B@rr#Ha!4hTe-3Q!;{t`DOm4QM%H>f_xjNGKX>vMS7k z*|QqFPQQ^FeIkE>j=JkFJ29L%5(BgI52xE?no_|_9dF8xIc08TeSHU@;(i8noD3Sr zo`pAvh|XD0Thd@r+)FAxk(xWqAzd-f1BJ3ZyD(9z{8KG2<)>f~8zdn8cq?QJhh@lno434>wNgWCk+TC>j(y-rFI``kl` zjBu_9LwhG4v2Vc9RPJiJ`h3&=FX{$#Boly5kiOqtoVa`b)zGnaZw!#M8nyR(>!q&}Vp7U!-1q?{JpC{E3y5 z{K>Fh^F_!k0i+tqPpb7OguB5YJH3K_{he}>+$R~82>Xo!uzi8ag~UsA-QlSjed_!8 z%&7XVFRh{Kdl26NGOxG@OkvkU0Um9yo;88*A62^wq5B%OSQBR* z4&%9*q2=1G>g}BNnR1h&N;t*7a|t&AAb9Ud0Af#cg>GKR=Rfr&^E`flZlTllS@m6E zAAp4xB9wB_I1NO{Ts9R5$YTIn;?2>X>GMFvg@dE2 zV6wFYUkCat*W)N=)$0{oGZYay5YOAKvMK%tzcM?Jm1EJbS9duaH=?c5&iUF!X*h@* zHt6-C&ocvdBOV|IwXW)TuPEyOW;Q-sJNZZx)?$v|45y3pcxakYLL;SxKTtiocdwWqBayjsvwkw6?7?I5G_;yoKBS_lT^#0Os z?XZ?U_2_f5@<(_5yM8N1t5`bnfeYlB)X?m>MEn;__AT9-wW%!1G3|wqCd}Yfx;mwf z!nHVM5eY}LgG>Ef(1NC1fKRB4b-1<9#!n)Xp^LdaLopJJuAaMgK3RYzK`7eE9he8USD%9}KPDc_keyU011XMWf( z_i`^^u=G@2V#;uQ40<>vmHi;S@PxVDNFiBJf|8u$CIsLe48>gAw?^H!R%jK^nrxG% zkCkYIJ5RJa+=Yyq%zyPPJ$^#p+Zx*DspeptZ4*4K*M8i!-@86B&kv#Y>;)*5*n7ln zRPhKn!mzYv5Nix;Z`<1*#e7`J=Ca`WWp!o#{`UqV$e{B7lJM-6d&>(#0rwT z;Dg=^?t|Ke>+vRw!`{)JN*#7g;GFDo_N*7Wz6pn0JLBw|4^_|f8}hJlfE+?bj!$Fc zjI%=Mg8vr;Hh3fvYkB!o$BM?|MOLp1FAbqOR}ot+pLPAhZ7hZtk%QJD9;=X{d+Y5Z zAtoDSAa=&ZqT$4coM;BNe*E%WKek86xd6@VSthhOQKel%EJzGTaC;!;W$Anf)3;kh zTEfLX2@V}>*qa+ZeQO4ulV9ljO7*O5M0i#Ayougo1a#dL@!P?TWEJ7hKIgew&8Wx? zx4nV9R{H6U3c4Da+GlNopIp6FR5}TqYGVW{qLVFuS!^JEcyqf2YH|4SwOW>!f3%jM zY-3~bXs>7~qs6x{a8g6-PF`5L3a=;hqDh>;qscbOs_58up(L7j-EjZ5M|KYXEBBlBDbGSXk7S~kPf)V5^AGsm8(kWcnQ4Z7u^+d9>RCLS`h<<)v|=- zzFHTpl7sC6vGuEkUl~Ho=CbHp*!fS1)>C&2y5TwPo_%WHG0&#tmM94z@7ViUbG|_u zJ?mebtfs1C=dZqN z`k|LWk4*Dok{r+j$9N*X8;(GQx~6R6;5Rto{xtG5%Evey(maPQCl(CB!61CN-lo{Q zSq*l)Pan*^lQ-zYF7j0;Cs`>GLuYA7`?;R8IQ`UYI!5(FV3JLHbNhT}sA~Dg&dDwY z?uE`ZFuoktN=nSpONjTGv21?_~m;hU%rC#JY}% z)S5r|6^IW1;HUdu{{Y;Vs8=u}+!wK9P=tq#ViVr+u%Fvf4eJ_Z6*Dv=4a? zXp;OSOHsZ*E4X@Z$lkH2$dVMNT;G)-h}^$~1Y};atO61idpwsc{G=D`UqAG2*@3k- z#*(zWpvlIdk=L5$J^idb^C&EJ%EQ9)O_qY_0Ofc|JI!Zu?bI2ZG2iC5p~;|_dCG@b z`XvA*kXw;EntNy^)@gZptuoK9^LAjofqN(%i~{<`c!C?~&Vb6&Dj-|t;Bz2xA_zd= z1@n&o02n*jb!o3g^bAdQ#lEKHtp+5&UOnE9f#lteOw*Nf{rx0Xw&EV!2_cdj+MkNN z<~XdGwp1Z8q(uOt8ax#;?6-1?Kse5Ecsq94qI2t{mc#xs--BG!p`#5-noTFaYkFGV_ayjQIKmoRigY==!6QKg zFtpngmf@GYlKrY73->arB_=wX3BBp>3Sv%6Lm%NYSjijJ-UD9NT0m2x;h}ogNym5n-(|C;qP^T{`pCZ6bC#%b`t0U{v(UKEMfh;haj@^v*QPqJi4 z(zG6|IDVp534x}EomK<)H!J2}b7iWMkr z6C}P(yce}VI;QauL8NH=_`M5b*Vq`~zkE8~Ye%ksFp$#OV3EJrLawrmG6LG-pI2`I zctq?w1uAdyqxKVnp|DwqzLI9}8$Net-u4`6J zCYOg&wz8D4uFCs%>W@y0(0sMg(&7*nhXPtN6Nh$~H=9p0+#$A6uFVC?5r-=%5|qtcO`UqCU!h+y zbb|K7toM4~&F_=TFSQSmG8&OaK1aCHY}05>oH<&eCQBlQOUx`&qZHKAyq~tDQf2}p zwvMR#2mlhXRyihStnN0%G6A#!Nh>))0ZQ6CU2hc8ktO(DZOY2Ka=OrUPovH$ePJqm zVW9y?LNd|6+GJLrntUZ)1gQQanFx6%3f}`|3k-+$q^l1;X{TyS@mv5Pf&t|$mDAc# zu#X&nl)Vw86wjHnRYp1QOwDlQxtaNGYV`xfJmg`@7 zut-Pht50Wk3wlS;5i1;Aqkccrb--9laofGfC4ToQo}}8nmc<3A_^R>S z%owp7w(Q+R%b$i#0w z;?c(aAu7o~pa4dK3R9F?h-RKA89%~49vk%bMz=R=8jqx!?29Z(i_@k$p-)AY{p!1q zyP=543OOEM8ZhxAhLhSW%MBc{ns~l|bh9a63HX+tk>oZRhsPjm2RFpVk>il}$l~VR z@n;Mu3w=>177fQV<3S1>fiFWy&84i^^DM(X^1gC5_{F$;ky0pj7L29MEk(qCoUzoI7qSC?`< zL!;C-f*mgijpQyVtTzH{GK7i{ZQ!xk>(w(@Q0_f`>@AxQCMUdQmqXXlUfp+R+X$Wg zkbH5uniIo>@2_ra`cYY}>cD?T=Ds2jh z)h2LdsiD!O>9+PFHCHI4_wX9dJ|w6Nrl+()xRF;o87ckHQCoj&PgaE;{pSI|vLtWA z>adzdxHj*?ys-$e3Po!|;LnT+7ngadrX;qv3$Oa`wZUXm4Qd+{`D?F^4LvB1=!EcA z$=;)9#+;~}0PJCFir+*@MAA)my$?XyY)LJE%_%$&wzd@d@Y8;luzfu$3G|l5aP&D9 zao8}M5PE&=^uSc6hkjN!>&4JFku94ECuH?U#pe=0LL@&H6Zz@tFG9jnVw+#7+>xs9UdrOy z5};*Q#ek?$Cd4On_kMWD4N3|)z8QMgz~LVZQT?i0eu$fhr|~So>4z;^KlrWz_(|~PRHo9=q$8&RK!7{ z3H>iiWW-yJ!|?}P!BZj3=+S$GlfC#Ll|4Q zOuXGxZ$wirH?;;nb%Inc=TbDtl?bivX3#X#2f){w0Mo5`a`v~)W$(2X9(Mu4h12q` zx_vaufROEAHApc0sU@3?yZO%g%Dhm)b2@%9{@a>TBPG7qPuO|_aeJRuLQZrN$kyg1 zWaE>+Yy;_y)CZr-bT9KRW;8m3s~7c6>yRuJwt2^j(~&@db#2-ln0WnmPu)?xk5<>6 z)1m83sqcKfN(>Jqh!m#mE8vkZY5V3Ci;74bp6H2P92Cs*!ob=)qi2!G8e+PYkUo>c zai<9;${t6v=ZS*=a}5R8!pL4R;vnDZH9Ktg@tWdL-t*UP*p-aabG(7W(aqb&emUX-b0 z^6ty?A;)DrH5;Mtjz3DSl=_K6rO^HTFM3m6XW^(`D~4x2A%Kk6p$mMEIOZU|{Wpr~KXv=nT*;mWW)1&?{qeWY_hewqvKWH> zlVtZhc;f)g$Y##d)3H_688(NPMh)*l1oT9Pn#&9IZ*)9ss*{pB)z}f3z-q@0n6>7~ zBV0g(=u30cMzr8Rd3pmLudy7!vTEiL#?<|#YS=ITNCTX8Tt7`{WDPA#DrKL>)BjDO zM~{tS+%!=vkq0tzU;!b)X0mnTQcFMwgT6klNNm;i-A9GSK?foz++Av(49tzuzl zq}cy1@+A|8RD7kRlLVSM>d1AR+*TJnEUNl;y~Y8-*E3dG!3a7TxZdyT&`cE66_>r*`O%zs%SYAvPt`w-W6){lf}%3OFmOy zl&=YMv7)8a{>jS^;H;xj#P({yv>=_2v_Sg&%xv2Hbnj``Ma|neElrN^CtWwM|3fs8 z5*0Z5pono=o@;j+NAAqF6p5B=^0shFAah#vihd|`k!_+`RC*e&O`&EQ9h9HfLmnSm+A!8+D ziqr}tp;XEgq?Jo#V9K+-qeH*Op1gDMO^-4ym^9=B6I{ zG50wu-G@5bOA0rfx7qOlpaNdgr*=N60aGPzBAOH}r0GSy#v>ebB6b4QeTSQimz!sT z(w8=+M$DS^--0mfD}#( zZnOB_aV6LwsB(~{mq3etMzwIq>BA}6%4Xmh92S{$|6MYr-jB!Ao1 z|H?N_km#14y`qfMSpDw3^RDVoE-9*`KdHBpg3s#c{&ENZsr=kO{3;WIq#Cs*@Uqq{ z_TL@xfBT0jae&{SHG$w76UYDX8~^6U|Ff&VzXrT@+^UrE|KLWqc-dQ@(tRZC)DU6aG!B`@ehk z$BVl}0{qlZL2DP5542`-|KZC1o8bkhOHcwbRLGh{z)@df@az9Ij{ow$x41H{5yzCw z8$?u-|GM}ojnH2*utJWt)mK=x;yxLB_WwZC^nYK>|6eZ#(1Gm(bS5F}hPV%R5F>=GJ@zs`1JB^* zqxrJl1}&q_<0pmx>LLtK7YAOoKO0T_@;3YJOBK6ui@5unvXa|qa*CLL(J0CU081Zh zCpUu+1{X2ewPk-4Y+d%q0CD~If~|k|Ljd0N5z(D=`6mS(VH4k*C*_SHg*V+uz`wdP zR8fINj@YW=)(104vBO~4{y9e~;>yPIf&gOt?g|K)|4T#m_g^U!0B<0WjjjL;g)#;6 z9|Gq8QUsdj8xVv(N7_nEkFe)`GidG2fRDR7Q6O&^Ecek;Re}sx+l)@R07=zLPic2k zxnMM6>hbLaNYZ!Ri;TZGNBZZ%F%ALnM0wqW583tQSHj+x>9h5oXQ=leJ@ z77X8a(7f52b&+ zxNOj(M9Zvw^1|_0=pF$5XB-dnoaHsIMY!a?hcY>NO69BRz@atCuT=BAG#ob*?qeF> zaB5E5&rMv2 zYkx{6sniPfq=2j%o;pW+)2b(EQt_qa+?$wd{>w?~tVm_yYlVUe-Gbcat<TG6R9e$`>hI5Ib7y>sxsfpmwwesED#k4Pe=pgM78Hd>mv1HJZZZ zz*nPl-RfJ{qtx>*0GO4i3X3EMi(Q)|OZ&FJ{BvEzZnHMA&0MR&TkG*@e!rUB;NCBu zljV?D&HMALhcK!><}l~ZTN7HLYKutH5%Bm;Mm%#(b(s1uE(>46XOW2-gYNhllV8$76G@<%=3+Q#+zCA%wiYxvLBR0wv%~GD~Wkb%CCT9`+jn} zPs@bKpA)S0oMO$kbG+c4z9^&r_XuFu+(zO3Hc%?_t&*(=gj-)irnLe4SxpKo~8=YLs= zZcz4ey+vH|tH`wodFP*mx0V2F7_E2lx|cLz->)+#C_ep9UM@ifUO z8UM=<9~l70Efv=fHBh~Q92C-^m$K3qeyIYOPuJShk$#t3N{m-3mU$!q&6MFumbmuK z9;evj1>Qyha5SgGhMBMfpaVp}Zj8n^F;h_Z;kG#ScMwRYcJWaxne}Ka7E`2$2Dhxb z_rtwRgtMeh{U>*AlVu`1n2U)7YH0d3SlxV${p6WmJJL7<=Y4PZ2pW!`K;Q7ua_mjNTcZW)Eh}2}Mp6CPXe= zgSYNi;sKJ$ZJ*D@(NBkl{qzF73F?8Y*t5k9L%+LD0+f6p=O8nhu?KDp(e z>zRx0Q9e4u8O{o3a6pYqi`811eWw~{+Dv=fJ1U*Ys11SPWgNf`bF6X3;!3LeNb?I% z+-t{!W6zWr;ikqPnZRZ0m6VzlhU{;a+{$K8m*>My^MPs0_b9i~FdMv;oRWKU57iC+ z3baqLZXcQ*M*!`6>H>5CJe^64i9#f{W)l`kB$Y#D_AGI8C=qwEs8=WbrEIRQq7AK% z3DcEoTP{K6MPz(2CBggy-edjaE16I)lw>TkP4wawhr%I+kV&>j?fa?fdf6XLjK5xJ zmf+t~a~?ww033141bDP2)s8@0Gz3)p}(<@Qw+1Xg%`Xne@tn+l|1int4#hM`8zKoZIV00ny5yR;OolBa zrd);y6G?o0DV48blwZ(tspybmziv?S$$qCnHK9TcLy+uEM>YS@8)VXZAjP@#P*pW; z<>h5zZsBgO?FJJPwkg`%+iPlT9k}01r^Z{jyIdNuck+zB;~^XH-uFE}^g2Id9*%0a zW+pfMfypLAo9-j_5(!g|1CJ0)TYp`>iK|Q7OvmNtci6H`J}rN%I|;kYsq`@HZfdVJ z^ghI_vs+5~<-OO1IGkEapYg4hUZYT3rrq-{s%fN^9sIWKuraOe*!7scH3>a+T~8(b z8%WM}`-7iShOT*c^NYdXttvD3x?f)j{rRjS{wH1wG*vovWVm!S@V+h!Gt}TNzO>L~ z*C1Caa8r^>r`DTq_8RhX6V-2f2{tlz$-f7#^0|n8KmPPn37XGw=R?5tr$dWI3g5$d zz0+zmM>Voy5%A?rTIEri{qDw9c(7CaaQwOkpNUzCu(En1B`n)7^Ptt*()reLpqL~n z5FW(~shL>?Id>tLjeKyc{)QZl-+Q6oyf(DISgA%A!`WapEu$TktDx^}zwVUje|4yH1;H z>#g&#P7c4@(@;h|H$Ug@(}6AK@6vQP!O@5FHAm=m33taFa6$gFupWgH)q-3_tVBHM zZgy8Emk|isu=xPKUYxFQ*XWBNzy&Yl!rv!Nc4*~6uQwe!7mY&?M*~pES1JRy)M`%d zBffkaYU_ho7sw|y4h-$iK;|KIUcA2NqfLdDasB47CAU#mk2}qYb1k#s#5qXESb{I5 z26|Hi<#s9rVzRouzRO*jR?)OPFQjbbW0~UV2Sq8IPANhen>gY7+?jCsvK(HtD29@1 z%0#iLx>`JvNaGVa#y)o}3$1}^RXKJFH7#3N&&(*7I(?CE!4p0P{OU|$($zSN=bnx; z1d$I>nqIk?`U;c{VY%-hT2?-vhvmhHIPNn3nYoA_8tce`kVMR1Y-$;`KxPwIogWOx z>u#*QjgHKZ2iK!3NS${!nGLyK@!XnY@$WowmbngG;#q=33{k7m6cz6_6=$N~C_Z%D zfy&=X4^U{$k$!?3Qk{d$orSXS8Ump>{!fr7pCF)$K6(1LKLgg}RuZ(dlGT(dzaLLB z@AzH|t01_EUa>u?_h_T^d38Rwy+87@fz8FkuKCDnvOrAS9N`2 z>O8P>YQCCF`@Y+Bmd0n<YTZ!Y{}=%s=yqNQLOMRw8of@mMKV)0)alNv**^f zgPPs)&xmd6-O5|F&EEAdCPH4A9I$1PVUUZ@#VU&8_JUU%M>noam;lX-c?fPt-5M}Zc9hPx*k_@~l#96oza)_K#bt=C)mCJ=N$&~DtT-@K(B^Yr{Ztm>JW?~Hx39WH4vWPO88 zYG?i{^_4T`+t0b1Q8THUta`+bgE1W!XKyC+b(D_I0-9iNleq1^Wi6fe4QZv1?ytt= zd`1D7GwQ#LAa@Qa(kk#+&(FFXPW-ZIn9}525VU^&e$K-!6q{!qtYdyL8@|=D9mMdYw$j>uDkFp?xCHnrH%nPFPe0&@Zr7(JVT?< zHSHJMO{CFw5w1Y!94v++d2?o)TM=d$HJw#%vY$ho*fGCk2Kmluzxyh}&J-Jd=BG~r8M*|*wCX5SmQ6P!nT?Q3AmzEfa{ zqej>QKJSji;e>z$;^%h83USE2#7KH5A(GiGPl55yR9;<6p%nA^EjoMKIoSP1vXif1 zw=d}}w>=oDbOL8d$hiF|kTB)VDd<2~wcSZ&iSbCQkae0JS^7davv+yn;CTH+U@7uu z)oJybo;=MIt0|DEr|_XgUHf86Q#)qfvMAF8d2zYt`P{gzuTG~T{F1_o+~Xuizf+OK zNEc#|$F_9S3VDIMZz04nb-Y)2uagadYZ-W^v@Bxf>$Y7wHz~prC(1o0KW3%WZMS%@ z>{oT3mMy7|nPQR?3syJrW!t1A+v=VGJ6$l#nJk^)|GU#oltdv!Z1(fr&mq{0-daNz zWuW@1OUAIznP&lL+BmCi{wuj{uYI<0S}xA4Moz4ZfAR&+0uB^nGPZ&%t zg>DjMdM+lw7B-eXRQ(`$Pa$r0!%)bva~2YNvRkoWd*~bv85Dj`w@}DWzz?6kGzg4| z$4y|;DE%y~*Y<1kX324YtXyWU@A*7amdsV9%O@#jDgzTQ=*!PoWe#EI6XPga>?<*3 z+`aMaM!(+L_$27LZ@eHFMNZql3B(pwZ}>&yEB_f=?@bLyyRB#^d12?>O6l#~**A!5 z)5KODw#aiNu1%0(>(y#kB$E%&Vje`rd}a1;TmbtKycC3BXn;}o!Bx0gMOST<=k7KA zi12hCPv&5nfoGW+*KwA~a=l>{;(-v2JdF?YpC(BTdlR-8!tm>kD$j(dsC&U zFDL0XaQ*P0mKyoOIo^hWpigJ~6%ISE6CWb7$AVE}n#CD!WdWl>PG#7su8!z=>eaS# z%fA)9is*zXI(g{96y!RfOtuLG={I?v)I@TDD3=j;edG3T@NU7Abiu)Puekj@miA z_;Aa|AlLIt;MhjE{`l$h`X~}Z8*uFx%MUnUS79w~`(0o3eIJ*l(oC0J@8fG&p|8=e zG%jkH)5(f~!!7PtFSeuwaHL|rKH;|BtwcUI4NY^}#B$k(T1HkD~Lmf8j@+&NlA=*4adtfAzI++~KzKRq}|$zXb*V z9v{e|1)PQ2IvvFOXGB~bO4*NnmdEdW0lu4(b&fd6KvW?Q$IbhMz0BK}t`F@^$IR)( zwN@(g%*3nS*mQjOI`~5~IJmu743GZle$PkXNFF85?TyVD>_RGN6CaC`92CH-DM++F zn5ytqu2C_Ebd0|U3=b>js5)cN*fDzU^TzWYI5JAp*UDnV)8c~N7hJXK)|CWEHoKwQ zRVUNb(oyadIJ8GhHf`-c2Xw~lC~Ea{m=dJ5DBR3$J`-a-BrBj|+D}UN+}RTE8EX=& zGuWo|VuXte2Nkw^B4hr%#yLcr6-oKHvM=`*6CHB)v6g5&-T9uj(qrJVA_HdC4k8GF z$4V2maO8}M+%=zzk_*12$95H5^}hO9V}6jqM;%|;QCRXSjFO!ngp;UsoZnWMLZ2kK zF3c3uL=jKB7{aQ?0^o)%7C7gY%9>p5tl4a}$9C?&4epmHwpH@{Igz8k6%mAiiIs0p zxTEN?LKj}y8*Y2)j-F?Tem^K6AY6Z2EK0G{A)7zxHpi{S?PzFlzutDV+gA9CV@Q@M zZU8I@>q(TMk$1DRg7smhr8YqfH%q^ZPE?Bt5mo9kPtW6&@YdT{2n`F@W#0PnpYk&K z$jg-L9Xrt%Gtjj4-o}+_Iz9~wIwLalnR40pY%^W&_}P6p?PJJS5C>=n`nAHcA7-m8 zvbqN&dR%9CoM91rTA$~qZTVY#nNO_;@y<;$iJ0P&To1>`JmDqjncTQg>luEbckd11 z@I%jeREnfz3*u!Z(FA;$(b2Iv!rgx}OR!fgjgo#S6R53H>C#?RDbLU21jQHLzs zZ7gL{4FBm!EjHS74@WK@WY!sH|8$BxLpv`r^%!I{&e08|-OgKAx-4(r^F|9)-%*?6 zDns?8J0&i>#n&2R`g86fg$6`00lAzrq6HVuah2Kd`P88n^RuSpU(m0U%5{PZeHQ|L zm8%HMBv&~MgkEgJ_$^|cjNCYih|U*q`OG7|+JPoNUKu2Chcd6TY+$PxK1#5|%<)SD z5zMbXpY9QVa8F$qeqYyP6xdsE&+lHgc97kUwjFuNt>o%+VK;TR(DcB;d-cK6{iw}C zb%fJ}6tnqv2ZO+|KBql-*a_~D_a*s9@>F6pBRQ2#Xm%?qp%@0Te|5w-3Bvv-VS0k2hzDmI3HF8NC~yRT=oh$3-tyDz`_L6x4wlj)ypx{K(1Shvr2s!HyEPBC<8@T?0Z+Bzr2r&+vc)wip8e&YpYs>`Mp0n3}z$P z=GvL*vp1&{_FblBXPoaBp3qQjMbx`Dem~ajpz*SK2IQZN=zKZ}>Wy~pbnh;lzs|=o z$OT5bOAI%fpT;Q*jPHBG6;MT90Ukaa*08UCkm^0Qv~H@3dgc6^lJdDhgLTFW47Brw zvsCjnL&X9kxz9qdFY-~o7CU$OVz|q2;e@;Mb304WIn-`Ntlb&*Fqs2Wd&eZzT)vzJ zJrETW3eDR!Q?<0NbspWY=fr&S=ZW-j2dOYhc+-^iAIDq^!PnYmP`1{~n`tqbD;FS4 z7$!3CFl>JQd>fjk{P5}ex$63O|J7V2|D%>CC7|Gi$KSC@m()W{x%WhXgkA|W5YIBJd0)ERq&yDiK# zKf+=^27bw_CXePtxqZz0lP?4My%B?W@B2WPt~-l8&kyA&^2G{_hcD0hd|0+#|Kq{^ zFCR@C>;HhYIVKc)WuPu;pMlcd)R}6j^a9L_U<0zY*v_Cv&}<1#(Brl_qluMH5#C`! z$rda>x0WW;TM9@0z`k1j4u>EQ*LG6cG#nSLtlX#b=5yB1cA4VC@F4r=>_e6>*|#g5 z_Vxnn1#7aZ$#qW^sH%`0JZCQtmN^-|KYCX+;Sun(nsBD)4{eHUh&Mc8W&TrySi5P+ zOdlS&$Jia}sQ=+fw_D6;YzgVyO*0d_hbR5bxk^j%NyWF)p?%}J^T%{7KUujl)fzlIQYxtyS zqk|M-It_kOp&s3?%{IB$)>WN-ETEf~X3&5iaSWR4;4j5k0-9!df6$Kz=m+-q5UlXe zXT?cU!S+UUbLxnqY^8p|%v6rzR8q5NxXyS&%Iea9-`hj=J21N5o8(-}YaFIQCDXaa z=lw0iw=Ym1N#^lQjp^EGDrfuDK<&s8k+{2nwp?c9o5Od8_}ZcEJ|=yO+-sMrfgoA; zR}{~vNTR66Iq$s2`1Qu@g3;_CC7}ZPR3Efo(eg|SmQ@b7fmWYwzpJu@^ebFMR&N&| zw|FSq`(cP^%ttx1ecmA~Qtp_<^gmg^h3=koRewf8?zC^6bUB_%&nIc4d_JnoQ{J_*}5=b|$ve5LTf5e|{Gm--xLfmY=ELn>;i@Dq?Q zZ6X?J6EP%UtV*+ihzy0ZRrzA-Pt?t(9dpfj3B%Vpg8!e46as4er>CCb5GmPtI1eB_D zQHoUQolpc+y7UgxJ4g+LAOa#PO_1IM=~Y_j2#C~x^iHINo&*RzlyA9Dd7gdtc3=DZ z`M&S@mkU<1)|z9?G464XGRK>?j0OCbm|E@+6zrI-%Wo|bCDcA>o4Uh6!bT3LckH-p ziTK{kHkRyrKxloM9LV7#+u<=;wwc3=h0Bwzkju~;$Gka0ivIF=*{WQYLVW7ecBg zgjWiqX9;K1y6$CZOP>i+KhgOpjiD15p@dTqAtE_>99QL5*c>{6S!*j`FWNfz4o}M)@tTlZoFx{ zi|k;m`NMDt?SJR19(YI|M} zL=pLs7LuHQHe(nsEA!PYT1akmLOK_@)wf?|qv~2f(-{^x7RrjLb3uDvA)$N;*Ggq& z7Zpi~e;!4qs(jc#vgbRy+^HjNm6WAq95}0^__5Jv%DvM~Yq8&dY|_tLc{l#BBnS`F zf_0c_9Bj6LGKI^TIch|~RCa+Ckndr18bOdJp+m{l3hklCN4CS+MoPmc>#VYa4+9Lx zx3xD^OCd8fo}YZMlnEl&pMvoyt;};X4;%_l`8CP4X3bD<4_q{x&P=rkuZCxBZ(_a7 z5&_lt;ymPf?^O4R)iAmak@HAX)RSOs?x}MTZ$^eaH1O^H`iNsINyp)R;mRL_p)NP( z3Sp+Y{7ceFBktF0avf$hXc_w4oK_GQEpYMRaA4O1&}yYWmp?L%Be0KE5|1 z8)wU0-9KphUFMZs5!bL&I819ip0#fM@z4x7C_pe@mdZ6VlQ*w`I##GGC|-Imf-$wQ z{D{Vhnkq?0*lAQ9G@gU^m=&@iR-|!qg->C3Fq18Ou1B56-vf!|yZ;W7(LIK(V?x<5 z{tr3(&p1rR{a*B=6CQ^wSSfLv&nvV>2u77@PgA>27b#ofh9IQhP(3O_hBBv!gNwx~ zm+hfiecGllGK(c4?Af}|GQchiAttuo88X(WTtoDHo;iHuXi}~DQ5wPp2UYJHB;@H} z0ecS~>`viHcP5#aNP5IE@BsX}yNw|*>DdiTYWoJD_fvC|Z}KXa`+UIcW&zueD>q7W zW?e9)eh6cI48KTO>gS!8KZ7(DQ99GLdkKs}Nh=LSU(YmG{lFZrY^8?N-11kWsqgy~P_`=T4b)CtRiDXic_vOnJWTs$y zYqA|o9j@Q_=oyyg-jAph_{K-e5RWU42AT?yAoQ(-jD3K9zAENa&FxfBuC_9Xx*N(K zhg}HuXb4b`ias3TvnZOewGFNlm+SKI$!#DuT zAV~rNMm<}n(}Y3D=H=k|R`o|&@k%@wo&7=u>v-}{E@wa5TV3u;$v0x_Zb7#f_*QTbP zrpmFYpeue+O58D9X`3dA{a7(SNc^8*mLwVNlgNsRryhPv)8Sk6EKk2!hhX&GOkaG* z-ye@hE)nVw@Vc#AqB(iFycyg+`%irzEo-r+7~2CV&v2LLbp=T7)1~*e#--l@b++L1 zpG&>HTx&`0e>oKNh?;F)LhJf-lk%=N4}mvPamg@+CE2KYv^nmhMJ58S=}v6u6*HmR zm)Xdg+n?n>XXX@n&X1iBCYAcsXt48DQcsYas}1$Y`((AIG;KP0qu3{rOREj`@Mh>h z!Y!o7ry?l0gVU-IFYTe`rF+m>Uk?dtDpOpDYI;{EquJyuOPj~24YnSCxAmBjB$zbz}WHXkN&4r_Q_@-{7xzD$? zykvW==WSS}>E(p>j~O309%6b<=Cs0(LtVo_x*WrXLXL^|V>|H_VgVKW)Hn=qCn5t< z=TDhl;L<*r)hUgW^phjXdY1$zo08354SFMe&OML44(TbJl`4=|JTN{7-q7jf{C=lX zK+Ogau?4%tAU8{8>SiyfA3`lr5Y9ec^V}~?)dD?G81RZ#BkPR5rZEI_b~!8sU@P2$ zqbQ`9lk8<4rD17mndgIopKRaZ-Ox@n8_xO=abZW!unl&#-Fssa(Fe9`rX8Vjsl3W| zzX$HpfLeZ|%#RT#iNI$V<_yLn7vd6^$ z+vCT1s-OYUp~yx>z|mpJUUn1%e=Y&oJCmnd~G z)6Q-EHt`lmmYQ8PV#P-RtRtfhx(+Lj-F$1D^mq~TVCNc%k?Av+ ztd;QfM+Xer^aZa!b0j#=0}M@*1~Sf&2Y*k2I(>zi+^z}MsP7yJXy!s5S&AvxD-^41 z{OqpRYI^M26v&~8u04kIDR2n+7Rm#~0NIjVUSspPdgIVrF^}+wbKZ1e0>3C8f37)c zm0Hg7B~JMxVFqDo9hJmf%;{*U3M%ydKZp5)Uoci?lL;POW>tKIvX|{k-+@ zM5;hzl`uTL@7}_bK5oJ`pF>HCI>i{Pc;3KS$-P|<*{V2zcolrhkhaLIb+`j2K+Geg zX|Cs=%!WG^NX0*8>q4hV54sOGQde@)8O}-8vD<#h80)O`=x_PTP6{~`dC!}$UYTjq ziME+LQJQt3@f^!M3*{jTD?NhZP>XV3K%yR{btV30dWv^&02qyblJR)uXfvvv;-8t= zf0M3E7lKsrvKD+a$(p>Vt|3xNtqnb{A?mi%{O5BN8Px-7vD9XMqg|Ftxm$0P7`?1Y z)4sh+u~M9I&okoP)?&7klql_4Q)C>=erdZ!&K$6y>wYR1#OP|z6%sf(zSD||{= zsUCiXk;II_B{)Ox+IV(kiH)$!NqZ>u_yuBvTzz(>qUY7$$O!{z->b%kw|DQqg*nH- zry;{|)GR9_+W4aEO_WdUzV5@Q% zPhBrp!1-`*X7b=6DK%G^hqzKGh814wQ$V(l2n1D{qXYUKg(tRw!U%GybHvPg5>@kX z6aMH|ZR!0YW7$yGy$%EvH;a5Kg&NV))_ICmSq$CiS3caw*}9qW z&hit#s?H|icuqbUkcMm|3g&UOB1Ri*_YAqke8MKx>uI49!f9gH&%)=p3q?Nos}73B zWgF!_z!HjNm?;C%*E;;CQu0Z7qUV`i+0VTynzsi)=K`H{frYdb1|uPut)o)deDiNv z!T>I|7XKD$6V+i(nkx!BWX&s}lSKRJk13o#g`s^Tnc&=X^o0m9BzA98>I-{C2=gj? zNP{{z`P3-f;erFDMD6AQLWD~HaO_8mrJ5-Gw7<}6c?~E;nRnUJ+|~C{ZyDHa3dlE# z*rLyA=l^|q0HBnaZr6Ob8}h^FvO>4gKKfssmk@abF6s7HJ&G_hJ9~Hgf$t|GBYJhw zZU0^g1yc^ffLnVEd)7;oTo`kM5f90j+LDUyno-H#H#v?V5yJ6C^SRTYgtUEZZq&h! z=uWz0$K%5>F`Jh`wjb=0RSwUP31$(JjVw?1)+Rfaz0SYRcwqW!FWK?WPlqWb)P920 z_YmD4h&Q~mJhfripsJ~Uu2R0mn83m>-@N07JO4t{+Pl4$>D}|%|&j+c* zs~nMXI=#Vqm*BO7^6;aTw6r4U=JO}f-+#RHW$dhOk$!i6I6P2_kv!`vPV^%Sx9aYP zzYVE}DpSEltb5UnT6t!xF7kOCkQy!y2&0#iSErEpgKwCkf{^dhr_a9*sJs%_n@y_4 z*Sgu|Sq#K-(u6OV9UHg$=q# z;WD@e6J3oc*HCS}-OC1z7j0aEPqX7V)Vca^UzgAKP!GNTN7dXSV9#2-&!%_(c?VP8Ok6x@Xa?ooU=e_8Q6Nmg99tYP2D<#>CIW+jqkj3#|hS{rM8cM%O~36vjETXF+)K0WacVhgfxx z&65utloZ`~Z-3SV3EFw)I4bQ*@8J)fa>=5Qdhsc-g>)FnvZf2^Ds=m}72EYtUu1bg zSdt`)LQNYS0z*i{@Er~fAK!{#lR@~ObFH-Ids{Jg!qy3tUgi4@g)ayBvq`8MIAiy> z*)o!ecpNrEgiGEMB21U;CVnsrc)F<_YPXc-)}KX`*`eE(9mLYNxTBQ8rix9$I;QbX zOscHtR_!Ylfd0~GGv9g28Z`uMYCx{DP30O_2e>jWNa9t`plbGiR`&?e>i>ftZnP)D zu2+0;`95+IhK39IK9MBVsqf^Je@H(SX%hB7}4ZWaa;O%s#qHd z9xZ>SkBZvv!!}rY5+FyPxm5KDEYSIV8?kG?F4e!sH7LNM^E0#aq-DfIZzW%rg6w)m zMS2Ad9ka6adE*>+!HlDLyt=~2Bi{kGlJOsS$ga690B|>}4i4pwI~$_7KGd*Ej60oC zVVy^`FV7XG>2`vjZ-(RxU( z;1<1h`spA}M8Qe}6m$Q2U0aM2T*+Y2@3cm@6rF!0UA#AlnDL_;YpE&#<4`(0$06yV zgBS4u2l6uF`}fI{gBDuj(JPeA)h*0|_eWzQvu#~Rc8wknxPlXVZ!~9z=)d{{QzpaI zf_I}CtIQ<}4#6CnH|jH8R2C!IlRXc$1)C&EOm5ctmYRxNkd&&F2`e%3{;4x-AoMROE!(e4o z0G^XDB9ND62!FMCGK-u@bw{XVN0s!1`2p!+#b<64{2k1xu$@Ma0b%+QSoO}yIuwda z^MMB(3p?n`tog&b=e@pD6Zu|?2!PWowb)C*uJps7z?C;nbk2EGaN=FOicE-Y3Zg9gf+swABncj>yX&iX{ZWoNo5L#3|2Gr$9` z?^V_tctDUihn1SdKJ?1rvVljht6R{k*6 zO}y{hMRFLUR8#B|)xTV5lhCfyrEqOJNEXA?%L@mPgtTb%^z z0^l;%%>-JJ%qUAeCyNba`^0(Tc1oMaX$``7|TMS#{pH^2n$UHlWTnjgz|>WJMD_cv8s9bEQ*F z0%fOLVu*SUDm(ijUx{hhugaEsIre!p zwwIAd%)ye@^R&G*2}fR0oBDNlZn!t#oR<-u*AQ8+WdBf3!4yq$F$Cke8oIA=NYZ)@ z-XqYUI<0E@sQck%6f33;sMo!9jlvMt$6^6M|?S*CwS}g0(^x9;>gIjpo673d9bxV(*{sKuzj6UU*!G0-0zcw zM<`I~dNwx1{q8%t?r)VUpKnQdXOX`?<)heva7tWw*cK}iu$kdKF6vW;^C~daGL!yb z53uYnNoP4g*t7d;-&LzHKnA2Cb)U?_iiVwOP%Bv0iY?u(7&meDeMV^^m+jHL_ed|7 zov$rJel4Y@P2cF10+$HRI~i$b_hN7OV0Hz{0*2FCo=KqMY!Q_UM%|7$nH9ADs{GNhU6m=r$hM3(MksN11;Am zs>VC$^(W$gaDAj_Ggw{&_uA?P{A#t!tu=@E=d|hLjzs(V(>6U;O|6ye zgA$bmPkUTxY(k~T=V5`hM9avXq8UK|s)2wI5X5#JN59(`omuL*sM>OoV#pA)cLD%4 z9eRN>XaRqTq3H_qV>@XFFe+nn_I*#&a#Y?%Hco;0lSpXT`6$T+F$H@0)$P!BJ(Eg?L>euw!AhpwXp`^Q2Ihfk*;Aupg-_A7q5Miy)Z99>^pCAv$ zd_3xKWu7eT`ubv{tH_5gDK>S1fD2l;tif}Bv0I~`12|W%@LW_Es}r~n*%0kPNcpa^ zAAB4L%#0Ite?CeM&%b#YsN}}*P}lTODa}imT%7_X@j%=esNc8Qx5GewM9;vN+f0* z@RaR0ew}xdD(L4-ZE1A1CR8YHB_$J!O~Nw`t=CPoah*~+QjqX*vOFfGng7ZI{w!T% zp3A$C8!r7W@3{+&8-HJo*M8~|Uz>T|bNu9+^0*Z8NYJ;?t_U{B#t*V-;4oH~d%_N1 zf<=h;HB7a^mpQ+da;0yT?9s6EoBz0$kR{to1{?`~s<1+t@CHzDLcgDmDCaJuD8Nr~ z!=Q+Q94Zq$)+_iZ`)SUbpT&N9 zyg56Nmtb-HnZh4){p{;J+LZ0xgEN|MYq~G=A`74Wjr{b!$y*>nP~-T0G@p;$gaV%c zH@B-I)uTxE^i!zaS6Va0^zVLU(oPD{ma-vC2)<82aU`G9p+0s=OTFQu1vKzf*Tqi$ z4Xjer5E4Sx!Y!^t5v#;#w3&W+0s1}2+62c_RX4`+30C`2b*c>(c@(ra;c8iyakcUY zF=kjsXF)S__$q*5{kZ2wGyT=Ge;3&Q53Lh@6)^sSKSF%~<3HTh*C-u%f45MMZ~gMg zoiI8&5k`BPNGhYZc$w<;Gf6xV&-YucX@ zFbj)=l}T-ijah0wPKJ`R&#Wp)7@uQ{UouVJsqTP(J z{-r5cSmIm+W8g`_(NKEcamdqP3hUtubWI22W#@BqX4cp}6-uI0m&qx=M;kD?EXNXC z^i*|55+n4ad$#2nYmgLq?#y&qV|`l=O&um(^`(P z|Bn>^rv`)Urp%$*L)MIqo6T5)cMHF>XJ2gCw$}18ZYyX!zL)R#r9E87u|66_Ol!RK ze9m=W@wv<&l6(NR*SK)$8sf?eDp%?djIXx-?{DfvfVcEm+My3mz+y-+!664IwO-6+ zyajSEyv^@a<+uZ!SFW^yU{VKu9L)TrL`M5i&ADtdZpvpMtx}7x zTJCo+MdAI6r#Fp98*_dVvc@=d$$9enxb4=?rfAoh_Jc4^0#@d(T7}vLpO6#R{!9op_kh>BT~E_;zMXJ{Yzef=uX(Ab z<+p-@M^3xyZ4pk8|BhZ|ufdCJ$0gaC2@2Q_2)VXY_7t0)HxYrM`l0fw9qpWn4tc3o zG$Un3P6HX_!GDamG7rgQ4*P#pe>gmy#spHnS%FD50qJjvhnVZ8>V+@=*B4~m*_iGa zpV!R!Z7e_wQyKq2UR=1;k$R!chp**JD#M@r^A9o%T2Q+%-%x92(NFrjSNhM6o`0^A z3?JKzEt5Xp2YJ8OqdzfBKUF%jmKLS zFQeN36Qbo$lqo1nF>~S?LV;Kk^~V~f|G3+4>K?=$CG*h_;tCeEyZLu^{Kab&Lcm5) zK)o3>{hKd!=fxI%aPXDCcQw1i&s{l1nceiawi(k6c>tbY#V7iES8I#_mY=$sQ_=sA zH2{D13rw#7Yj~Z6>Tg*Hnaq5^`|>#Gz1#o2|N66E{wX_|>wt~8MaT7La`~f8k~)B7nz z01K(v@T`jZ@A!!ec+UVib^b8F`1h`ULjx>imUo98@8475uQy~4BU@W9{k^OItMI?t zy8l)9fA63F+%*4H_+M?^|62H8`h)*k_%L^9|*I za5_E4e%4COJ0M;e<0mfCQ@u&r!4#eNB{d;FDg13bO~kFFsddA)bT-8fs$V&Q3ztHl zTu95**!FPyc}Yr|jIdXu9A$H!^j>Nbc7gw{I%b;2v}&^T`IiM{^B<{K`mqi31v;wO zsXS@~IkNB9*owm9(kh<9;liNj&mX>WFPfXaOd}MN!m1+R*mBx`-F5COoyS(tm51#v zH@Z}Rl&FJ5PZ`fu*m@^_ts?AMc;5-$tcGS!{?y+g zrhh&$7DwXoD1CF9g;;QtFz67kA1&79?AvG?c6dequi)127B*CoiB_DU&F73J}8XZ9rixSoX{SrL|y?+Q`$$KWw{glA^e__2e z`F;CFv&gn}Y2wW;#jyJ}uSMzqQk&)*{-sU-_rLTP(VXB57alM?ekiN`56J;i|9@!# z{;zmxRGZj11{hjZ3_FjMTz%1Ie?bX;Ehr-J%wD(TbcW8~4qOQzGqU}p#&O^HAKT&X z_-+DmOA_^*fpzmC8XpUf0wvL(Ri;fAdD?rlS%f9`gvALMIaljgky2&j0hjn=kCUT< zNAnmx;_jO`HZtD7d&!;N%_MzTU^WOT$hU& zp@N)xfw!CrpMHKh=^8Z~jD(9}v9E}ik6^l2!r~$|ZOVRD7(Kvl>md?#GN4qcPa-Ti zl1!985XNJZ!AnTw^8RfBz@_3c3N-89-v)cUZS?LQ-{)2!?JgQzo8j!$E8|0)Y-yQ> zy`D7vxOukx`Z|f=_W(Ee4^7}itq7ZpS#P0Ae;?r-?}|-c>1@zOJH@xf zR_1?b+ju%B@D8AguFhJ_5fiKs9s7t_@Au98cmXU6!3CDY$1A4xJ}YsF7nuo2ZdM)_ zdN?GF?eV@dNa$}{t&2+br{X0pT=g17=DJkJzC>!ov+oIOICUDr?Pt zcy7S-n=zj@%^x3wSkCEmX{mtt?nr5bZ&1=%uv9Z&y3glyp^CquV``W>~Mn@J5OzP8~Nhew8Cgnayp!csO;I9*(IZ%U};+VS8uQXXGHbC z5Mk0!Cd4g%?R8QCosPv17C*m~iP9Xg?JYR%O%aoMT%hpRBV|mpD@B&++*Xdy)L@31 zVy@8}hWF#$pFGM!LNO&JtH!7KcXz~73j*3cU)kwW)BVgW$i8`;)JYGZLjM6P1#%bp zL$%kr8p2lcot%Sh2j4*Ff_l4*F48!g05{nWKISrSZTP^Fak%-OC0}(Wdu>NtuG$_pMsRIRV$-Sj^p%Y^ z0h-4{4kG}hH1__qcTG<^{Na9UJuiI9wOI%#Dw^DETqLn+ne`Fr)Md{oKieBCg1}eI z@J4k&$B%%Hsl~EJ*c9Ywmel6JVSTU{#f*V%fm-H6I$23G0KpxeF%*)io;Rgtdt6abJS0ow} z_a&g13yKPkdY6i}!l6H>h*kBSxtKC^zL9A-%QI6+YTAFG!=d|t?55Ud(qfnLsdeH) zy%+~h292>3+XiP<=+{1Cp!7!sbUXsV0^LXq3QbZY>(`N+2WR;K+p`BYsBuKHkCQ)} z)bYb|1F;m+%Kl(}d-|WIXJXm*40C=qfI2vWZNQ^iTpswx&vjHASbQUTw&n*31je2% zr!YdFhXDtxJFDmvT?Y4tu*SY&*qLSwOcWV|Y?3-I36*#|HI3@Fii2X!q`qNR@{(%x zASTe{$gL8{RN1s=``rFqfL@;a5pq=PIp9Jbg%lRt9c+=S_ zZ;L+L_-SgqUj2v~TI1+ubKoIT&xbeA8CTj?gmK-gCMN|6e*DJ9?HW3Ivx(lMId!{J zrw6IteEe@>5XV4zU1oo6Y3) zZX&U_tjJMEUb>S@v48qT@jPk}f2m@-9j=UNv_0-mwD+`prJBAxnU2wr9Vv%F z?7u8{c1J~XDu6ck6g@8dg^znJ<4)epiylN28=W^byrjtgGCwxOb|V3HlZrqmK9@Qly7QGBcgyTCE1J?w&y!}XV`pQVTWJVeze^yYsdcS})Gl*C<-l$Sh3iXC7%#b>fwM<6utk$H#!|x9}X>REMWGn}r@(ZCxgv zl*Ya~+Q0-tgj@<+84^-W;WNH5qlY!K!Zy<|T{qBqO>=AgwzDfg4)Zx|O>o=guF_;V zeZ_fm?!l8C-=pLG4I}(<6pzSY}bzoDEbH{ZO@O#5;;xI-GBXeq?efG)$mxf~n0cfinn;$+W zSzoIA=Ag|gv7SWhEOof|41u#zvTiopat{dI>-cHnbauMoCNnfGL(SJJrf%bN2b2%3 zo&ER~Vq(U3V8`uFgtViyZ`>ezvej6oqR?=4cMhLI*2>L*@0}F=3vz({p-NF6{a+sB zKR!O83tHfToF7kPmJ7z`RD4Zn(=}&3>l!UeQDi+Xbl|)r>vq?=8vD?PFdjX@s!Xn9wCDbLwOl`2#!>K9A59wzvjf>rmHMW(#eOL0d z!(n3RdiZ?^^~2b1Wzsbv)LEh|sa}TBbXEP~%R~=Vr6^4Wz;HaHp)h!)bf~LycSX|4 zlfFsnd$zoB8O9DTn6?_t7|-hk{d$ntq>Vfmb(?<_IdyR!e&U7G=s9l3;#Vc0!S z7=yr|pUaB(H7|)%oD#!$>=c4-(Tkq48Q*lg?_KI?c|`7shwr$#>8XnvhKJ{=NBIuo zHaJOykzahixw$bFnBQEdz;jbN4ZG##8fh==*+|9-;We%q+D-H`bDO>Ar+lK;DY(8q z_0jSrhv52?OE5V;h=Y<1*XL~c#;`gv53+G_p4*S%8Y2C=*01ftjcerW{cQ)OIZti}xFY<=hx#w7 zCbvpt6btpvA%K}JTyZaTG)yYO)Ype=s%mGd?<@tCD-uayFTQW=6wRJwAg9qZjpgK(1oE@1DACTy+D)xbv z%15N*Uf0=f54$wol7jfden6GK8At%8wX|UXlC$$tx@0GYXRpk&ZUuE(8blp*832Etj5p?r5R`%?t=u9nbrpBs0SofIAYJ;KkqFPR- zLJdJfib{v`Y7Iu0w#8jebdNQY0Nt2Pq(nWMq2b=SnLqUx^ zw^Zs^<~L^Pv_de~pX#n?vhC!?+$LO$iM((8>a3)uZD)UfzT^Z!*zrSEWCLvChyU^B zy0%#Vb-TenSK-0PqDb&ciO9E*yW>Q1^b;wH9y9hz&%pkeN-xEas-Bj;98)(IX$q>R z`*R>Sv0c5rsu^pfc)jBjHx)&~6mNMYy1~H95wKbBrbOF`XqFza1=X}2Ly4h>o$pcB zW-d{ro40+-lm;%(Psdsk{sr2!xYS{AI0oh*UY#&m@$nhXXDe+~Z(lQKDSy1X zEBx7{H!*tEkXLr=WobaL#@G%GG!y8!zX|^V9Upmtszf$)kjVOB;IhQatO$ zLNP%1T=;JE>Ry2hKHa;j^g1%L_V`kU*l>TAnRH*Q#k>)?J4b@0^76qXry}m@BEyd2REI(V|ND-R81sHV%km&T;B-x(Ikh zq?)Vou(XSq_uXx}Sd3Jk5$-{PIPdvPQ=W5J&nJI|Gtm@Fpo3Vp z=dP3QMP$qV{^|<`#q`0X2S}2`$ybcw%V)Z?u*e3^TQ{>}G=UI`l;k$5mT*2(gT!L` za3y28m_#brn9iYc3ay&teC2urKvE=@TKltC_;tf`fnNUnOz2TKR zU)xs#PPg2btA!Y^6fMoJ)2}Ug(d<{L?WUdATUe&~)Rf)hg5bQ8d+w`0GuQTvcoiQn zQXyI01eA9w^|H4)9i2W8kui(PmBKV59Pl}3IKl&Vc=S?sD*oXV!t=t4D?cI|CW44e zXS8{C33R{eKsU>+<)XTGo5)-;GNK4SLj$txLo0A8`2L>nCg~_Gv+D;rXoWH-GbLUI zWF0({?5BCaoPs;)wfPcI861FAZxEj%+aWz#DBDxl<2VAL0u!T%JRHZo@hoHI3MDA zccstuYa};OwKP|}pX9u0^#%>wF9=ow5Nc*&%-i#Og@g!_!*#^g?S|TEW&kQbNxDu6 ze(1NLnw=et#>};DGTO>~!RWqd`8588$us=R$gDF@=flyqWw4S;F5@C&!Q=> zXXu|9YE;eU?DbJr4_u4S&hnOw!j$80N(0=3jUt}9*(B}QVoizH^<8hD6=5FFN;5UV z+0-^40OsX03@yK}w>fzk8sviu^2b=`de*aiGeTrIvhF52&2cuTCEaKatg>hx*cj?O zIUN9%PTAdCa7*-1Ao~90^_EN!`${=_Y151~>IXEElRk3=xeRV;aFDCSc({AfxB*n_Q z&Gz*wa}=mMC&pViwiKHM$@ilfYtBET0jEgJYJ^bXsMqpHc2r-NkXG}H9AS#^5SP%K z2Cmp=os~fIyjv$R2J~znvQd)0I9ft?xwn&)RCwHJTIkwnW-3z;@q0$Crb1B5*kK=y z;?XzjZU&j75|OkWfUI7t3Jds)q6?;Y8F#8?8O?;Yq}i+7o+Rv+dLSpVruijZEL|Rh4p-) zs^EwZ2vOeYe@TN(ijg2c!Xcf?1QjpvvD0eDG<^C|13wqUwPn_ur^zL&!jN}f+*EYe zO^W8mgEF)6%;cZKppWkKnaE0QALpR6H|T^yM>4lcTN@>}ieT#$uQ3+$xXUSRhlI2* zC0nTWU7`pP^V0OCdZ!Af-;6{lJ$BP;$8)u%7cwtH1|{>7M5820-?#oG0G|{CeQx$( zHGk#(t!?X_yH_F4?lzNBGp{tIv= zd8x0Qx`D)>e)jDV==hoTnvI z74qe0z?jCw=5gbjTw%7-HO$`G&GG=2n#Y;YN+0jvuIx+ZZNA^* zqu%JYsWA#IPR7UTT2>1Y<2`N!$1~%A} z8akmn`t(%IAe?uT{fOF83Bt5Z9s*c*WuI| zm?CBI65})TY>h3jfhI+B4nfm?t5R#hV~M(=0M6`@YG4Y^KR`&XFE)v+`TQ8inAdwB zR<}8Y9S{})YU~N}SmV2jAO~-}>1wR%weUG}L6!<$Npu=!G!kE(=qJslJx|t4ByJkH-5ClPILVPf za_WQryb<+M{O8!8Tm-doN~4@3`+XrUD`==N-8Sa>O$EEv(gqd=4BXH$z;N@c6Jw~QArH!^>!(_5*Gyh|B@edCk`lNM zH_SLh0%3kj;*j9vW%e9GerKy1PLwdJ$+(`d$E+8QdF1JC%l*7b7fJL+!WkltNl zGlrb$hYz)!-@a)B^nQ0n$=P!%hhtggrct)oTP)c+5buW#B)c-`AE!#ZYEuFEbbc|; zLbj-~f4}D!DNu}lZY3=TvNehgU%&I_*7y3eE6dA+~k&lQ*~XRkOekV*B4y@9hi{(V(NJbTK2bs!^P@4bRBhY4>i^Z4UJ= zGv+&DZ#+~LyF_7W7Qd#kQYb(}BY#JOf{?|Du)Q?KlxDj+iWsMn*Wz3x38%Zg80do{r^syp zsU?Mz7EiL1~aMKjSsOCzJRH$^8I8g{mE2+HY+jM{Y-4IZ|~Gq zBau>w4H-E*!ROlDbuBPx43Xg%?FZpRE<@Ayx)8wj@H;a z*>EGq!1Wrv3X+d+$mj<=j0@s~m=wT$$ZZriIe;&n2GHM!PqxR+N)6Xs4-6GG*`}48 zcGskZ39H8#U8V<1)^e81E39Xve`3lEI3N5sp4-)=ia`q_K$K8&y);8n)x=;$MXThg zg8wjsx$xOh9<31HX?PBR$W!FBHmlnAZ2(3?aT{k%_R_ItgXY?|l?+H0ydfqQJEg}! zY!pL0Z^YZvdrU$s-oXQOaU0_4^lMRm^Cb19 zNd;afe6nD8NN>KSI?D4c5wg@7u-a42bzHRcHXhY!i@&S^yTML?M`&~i1&`u#_f{YC zPiN7@h#Ui=F7LF)&|+$ops-v zdpsa&hiICuIAfpL6$)08z!z)b=YX`Fb{iVm#342%=`^Ci?vs!p~WBTj9Ho7%IrGfd(kTPHM$S*6s5Azx}Yb}2i!|J4(o3=`X z71DS+mV+4RRqczWeTf{ouMAr9yO$l-+)I8H>I&ZD?2$Fy&v1qbaL^9KCCnj+gq#!I z7;{e!3BwYBc&cNXr2zd2w=tvEWMXB5D#o4GuqGrQ3ws5L>>(_K;bl=%#d{V}uH4EL z38Ltdq)|fGnqc#`L8>-u&n9sp7jqhbq(i+Ym^Y%_Yw?LsvK{8}chtrTt~4HJpq?|U zY=A#OmS~Lc3=7x{X3l+cC535DWBnPXY&{Hq8W~>5`N2Vi;>X8hkZtRpam{BiQ*9cB zo5ILN4O5aN7vCnjy|!ZnkQi7+9%T_1l}2CoyWd@L2U0XgJ!fh?6_r7=#5*qDWEE$7 zVlDApmGq+P&hxe3hzT;ZPHiK2hBumM;0sXs4R>3cni4R9CHY2MlyvovJJ>48q{_G|nmUPT1r(HLkSj;YX4~0-9;sZCxv|+!$FYFhuUDSfHvZU0tND(R`~)W_I9KZMXC3RL!QPuy zr~Atk8|4lC;w@0~=^Hiid;KB+wZ_@VQG$J4z}`!A4PkKU1jiVdEPWt(Bjogv_pK(# zJ->}m%4%fMcp}LnriR|u(lgem)VMiSY;rDfr#9aDZYQj0;%rb}=5FW-QEwTkx)9+K z&jxX+8<*zriwhC#Sc}Q*>C*~1v3GUItCeAJodxbP{Ak_;lsyo^)=aCM{zU{!7Q`)k z7(c#~2#%UeogBc+(yef0axcXqiHmR5n3O&P!jX5DQ8jW{sLgsCOI?qBDBWtkF*F=2 z89^2O8iLIaCrI0;Z~(6ok0&n6E8#c1k^+W@m2M5pr3ka{V%~-9o^b;s##f|^pEoG; z3jfR`=dJ=bqfz)8VR)!N#ma}iT6F(}hHNhzXLt1S+!ZtTthjz%^g@}HA)|vUMM0*W z!bXYa$kj)d5v(_+FF6Oy%d!~C2S0I25&V{@8~nyzMKI#C>9$@{*TR=fuH!A*9F(spc)OB}_a~VXIX} z+Zj`?4l?R_P~OGt;L%!PVKy#lTQ6SHAuGl+4pTNmjBPx3=Q^!A*SenP`RDoT+<$$q z-+f>A=llJ9KHu;6{_9TR`dLlMNTiY<3;rR1L~U`O2u55P-FZH1zvRgAEcQOq9x>te zlN!OmF2P!hh*Kk>B{Ki?Uvpo_jXy1l&_G&QQ;JkeV$q2)zLVde>5O_uF~zT|zVpmR ztb64=zjp1$2MiM=rUVr|8UueA`BVBE@|#KvF+n1$0Z(ix&seyL=5Xr2>oln_Dtb~d z8R>wc9e3to-npa?FG(cEF%ajYolmx9G3K5Ce{-Z@7dab{1s8`q8b#0FYIj)wuud-z zw$;^~HfKvdD_bQQS%z;&V$U3U+u(3TXdW&s<5sF>m>i;PV!wjm^qAofryh^}7=`t2 zC|Vl7J(sz;B8xa|wtO%BxR63SAhj16_AJtsJOhXPg(qYcep@fS7E9b2EsEh8O_PR(UT=#Nlog8aJ5grw>F5yXCWLt5K{%&@Os(3mLc zVm`l2XpL>JwcI;oxpg3Jh;Ac0y}~PhpLP$+t|5Ds(56E9u{&+%j7%8FIt%Ug36?p1 zhG$eeFu2pjvEkvqC>JsWBrwbiE4a&;EiWf~#Q_ut&I}GpUZCTr^z*de&?l#Am$G!j zu6`?bNzB){CV>Q1YYVN(WxII?3 zcex*F>^BtNP4^ul4UYS(a*sX^g-b8*@%g7u2x2PCB*ZwQ3xOhmF(#a5=eWTZ`<#*Z z5JSr?k78%+X%R<*LzM+%6!rlgxQD9y)3PeVaR=f(?_rP?7dLwec7p6d&>xGWR%y_A z^d7Sa#|}%~tzHi^4_Cd-=wI{5q~F7?cqxwg%5H3Vk2}FkjF^3S$7Yjdu8BkZOkTs2eLIte%T(w~fSvMX3iu)Nez%?jlWPr8Hjf)p&gDY{hFxZotpP z=aJoE$E4@r=Q;}et}lo)?wuS$Y1}%AAUCc1Ld`$)Ek%IZb+CEyqP-$8vlxO(Nyy>T zo^JP?&ERHG3r0f2{BRBG^rw*`%sh|eO~tu$&-ex>;uiJsXROXkh5xc81~C%$2fa5Z zL!8UsMSB&;rVfb=`fv>_1Fs^7zyB73J%m&?Yc-XV+n$4E!`jZW_>Kge> z4>!0&66BiXI!nA+J7akxEC)U>U13a8^YY!ZjTh2{D$yA`^<)XS8x$vD%1TR2a~?($ zg>dXGH}k5qehN!pXy0CFPbJxD!uE3XM_D3e?o`E8%vC5ssTZu5si@xM)@-mwXk_*G{{G6n13N2j(HKJ(xnsj^Y)N4LD$+=hn2ma!S8-V4n=+vL%D6d%= zL-N~C(DUv#&8YBeOt9R#oEpb3Wa;&uis<}N(9uQuWB|CViujLEMPz_@22=w*CCF3A=d z)+aTqN*z#XFxq{~PWYxb_&4lRBPuB|FEqc!L*>gY{;^%5LJ%rt=y}L~V3h4qAAKf@ z+9f7jyFbfr=i}Yu>x;dgNyA!3J^A(O@X;dfB>~08XCYx@t1!#42)FbI8En&1u7)1O zx-{`Q9X39OZtE3JTPPJkK@bap4+{KcHIE>IJ+BSzSS&7^A1iU<;)%liT6NS|t1Gp7YXP(Da0T_|@H9QAom}Uu z4=A2ga>Z@eYhn8*PF7ib1(}!FBD$3jdZH!_s$}vh0f>MMAPF4XwC|n?N419I+ zaP!%SwekbS!OKI?>Cp0P#AB`?fpi&7gp(>S+Uv18;I&;XzaTgo;0f#Tro_om{zJ?*{$irZ4I`Y&o=dd=4P zkGXzD5NIZ_s%6XH3Y1zd;Syj<@d2v7MpKCs1# diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index 2884450dc..cd8c4c2ca 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -13,7 +13,7 @@ " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants.\n", "\n", - " This interactive notebook provides a step-by-step guide for:\n", + "This interactive notebook provides a step-by-step guide for:\n", "\n", "- Verifying the data quality steps performed by the model development team\n", "- Independently replicating the champion model's results and conducting additional tests to assess performance, stability, and robustness\n", @@ -25,75 +25,6 @@ "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", - " - [Register a sample model](#toc2_1_) \n", - " - [Assign validator credentials](#toc2_1_1_) \n", - " - [Install the ValidMind Library](#toc2_2_) \n", - " - [Initialize the ValidMind Library](#toc2_3_) \n", - " - [Get your code snippet](#toc2_3_1_) \n", - " - [Importing the champion model](#toc2_4_) \n", - " - [Load the sample dataset](#toc2_5_) \n", - " - [Preprocess the dataset](#toc2_5_1_) \n", - " - [Apply feature engineering to the dataset](#toc2_5_2_) \n", - " - [Split the feature engineered dataset](#toc2_6_) \n", - "- [Developing potential challenger models](#toc3_) \n", - " - [Train potential challenger models](#toc3_1_) \n", - " - [Random forest classification model](#toc3_1_1_) \n", - " - [Logistic regression model](#toc3_1_2_) \n", - " - [Extract predicted probabilities](#toc3_2_) \n", - " - [Compute binary predictions](#toc3_2_1_) \n", - "- [Initializing the ValidMind objects](#toc4_) \n", - " - [Initialize the ValidMind datasets](#toc4_1_) \n", - " - [Initialize the model objects](#toc4_2_) \n", - " - [Assign predictions](#toc4_3_) \n", - " - [Compute credit risk scores](#toc4_4_) \n", - "- [Run data quality tests](#toc5_) \n", - " - [Run and log an individual data quality test](#toc5_1_) \n", - " - [Log data quality tests](#toc5_2_) \n", - " - [Run data quality comparison tests](#toc5_3_) \n", - "- [Run performance tests](#toc6_) \n", - " - [Identify performance tests](#toc6_1_) \n", - " - [Run and log an individual performance test](#toc6_2_) \n", - " - [Log performance tests](#toc6_3_) \n", - " - [Evaluate performance of the champion model](#toc6_4_) \n", - " - [Evaluate performance of challenger models](#toc6_5_) \n", - " - [Enable custom context for test descriptions](#toc6_5_1_) \n", - " - [Run performance comparison tests](#toc6_5_2_) \n", - "- [Adjust a ValidMind test](#toc7_) \n", - "- [Run diagnostic tests](#toc8_) \n", - "- [Run feature importance tests](#toc9_) \n", - "- [Implement a custom test](#toc10_) \n", - "- [Verify test runs](#toc11_) \n", - "- [Next steps](#toc12_) \n", - " - [Working with your validation report](#toc12_1_) \n", - " - [Add test results to reporting](#toc12_1_1_) \n", - " - [Finalize your validation report](#toc12_1_2_) \n", - " - [Discover more learning resources](#toc12_2_) \n", - "- [Upgrade ValidMind](#toc13_) \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", @@ -105,8 +36,6 @@ "cell_type": "markdown", "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", @@ -118,8 +47,6 @@ "cell_type": "markdown", "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", @@ -133,8 +60,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Key concepts\n", "\n", "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", @@ -163,8 +88,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Setting up" ] }, @@ -172,8 +95,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Register a sample model\n", "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", @@ -218,8 +139,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Assign validator credentials\n", "\n", "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", @@ -241,8 +160,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Install the ValidMind Library\n", "\n", "

Recommended Python versions\n", @@ -265,8 +182,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind Library\n", "\n", "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." @@ -276,8 +191,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Get your code snippet\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", @@ -316,8 +229,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Importing the champion model\n", "\n", "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" @@ -351,8 +262,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Load the sample dataset\n", "\n", "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", @@ -379,8 +288,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Preprocess the dataset\n", "\n", "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", @@ -404,8 +311,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Apply feature engineering to the dataset\n", "\n", "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", @@ -430,8 +335,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Split the feature engineered dataset\n", "\n", "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", @@ -491,8 +394,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Developing potential challenger models" ] }, @@ -500,8 +401,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Train potential challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", @@ -513,8 +412,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", @@ -545,8 +442,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Logistic regression model\n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", @@ -585,8 +480,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models:" @@ -615,8 +508,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Compute binary predictions\n", "\n", "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", @@ -650,8 +541,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Initializing the ValidMind objects" ] }, @@ -659,8 +548,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the ValidMind datasets\n", "\n", "Before you can run tests, you'll need 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", @@ -725,8 +612,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Initialize the model objects\n", "\n", "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", @@ -763,8 +648,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", @@ -824,8 +707,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Compute credit risk scores\n", "\n", "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" @@ -858,8 +739,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run data quality tests\n", "\n", "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately.\n", @@ -910,8 +789,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Run and log an individual data quality test\n", "\n", "Next, we'll use our previously initialized preprocessed dataset (`vm_preprocess_dataset`) as input to run an individual test, then log the result to the ValidMind Platform.\n", @@ -949,8 +826,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Log data quality tests\n", "\n", "Now that we understand how to run a test with ValidMind, we want to run all the tests that were returned for our `classification` tasks focusing on `data_quality`.\n", @@ -995,8 +870,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Run data quality comparison tests\n", "\n", "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" @@ -1021,8 +894,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run performance tests\n", "\n", "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models." @@ -1032,8 +903,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Identify performance tests\n", "\n", "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" @@ -1053,8 +922,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Run and log an individual performance test\n", "\n", "Before we run our batch of performance tests, we'll use our previously initialized testing dataset (`vm_test_ds`) as input to run an individual test, then log the result to the ValidMind Platform.\n", @@ -1082,8 +949,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Log performance tests\n", "\n", "We only want to run a few other tests that were returned for our `classification` tasks focusing on `model_performance`, so we'll isolate the specific tests we want to batch run in `mpt`:\n", @@ -1116,8 +981,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Evaluate performance of the champion model\n", "\n", "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" @@ -1142,8 +1005,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "### Evaluate performance of challenger models\n", "\n", "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", @@ -1170,8 +1031,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Enable custom context for test descriptions" ] }, @@ -1290,8 +1149,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "#### Run performance comparison tests\n", "\n", "With the use case context set, we'll run each test in `mpt_chall` once for each model with the same `vm_test_ds` dataset to compare them:" @@ -1325,8 +1182,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Adjust a ValidMind test\n", "\n", "Let's dig deeper into the `MinimumF1Score` test we ran previously in Run performance tests to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", @@ -1376,8 +1231,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run diagnostic tests\n", "\n", "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", @@ -1482,8 +1335,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Run feature importance tests\n", "\n", "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", @@ -1522,8 +1373,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Implement a custom test\n", "\n", "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", @@ -1632,8 +1481,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Verify test runs\n", "\n", "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", @@ -1794,8 +1641,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", "## Next steps" ] }, @@ -1803,74 +1648,21 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "\n", - "\n", - "### Working with your validation report" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", + "### Work with your validation report\n", "\n", - "#### Add test results to reporting\n", - "\n", - "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)).\n", - "\n", - "We'll insert a specific set of test results in this example:\n", + "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report:\n", "\n", "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", "\n", "2. In the left sidebar that appears for your model, click **Validation Report**.\n", "\n", - "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", - "\n", - "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**:\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "5. Select the Minimum F1 Score test results we logged:\n", - "\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Adj Threshold**\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion**\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion Vs Challengers**\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "6. Finally, click **Update Linked Evidence** to add the test results to the validation report.\n", - "\n", - " Confirm that the results for the minimum F1 score tests have been correctly inserted into section **2.2.2. Model Performance** of the report:\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "7. Continue to work on your validation report by:\n", - "\n", - " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", - " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", - " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "\n", - "#### Finalize your validation report\n", - "\n", - "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include more test results, create risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" + "Include your logged test results as evidence, create risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "\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", @@ -1885,8 +1677,6 @@ "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", From b023a40136a727e6e28bec1d3887ba3882838363 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 17:22:50 -0700 Subject: [PATCH 86/90] Removing the tests causing issues --- .../validate_application_scorecard.ipynb | 72 +++++-------------- 1 file changed, 18 insertions(+), 54 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index cd8c4c2ca..a71e0832e 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -1251,25 +1251,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues. Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data.\n", + "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues with the [`OverfitDiagnosis` test](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html). \n", "\n", - "We'll select the following two tests:\n", - "\n", - "- **[`TrainingTestDegradation`](https://docs.validmind.ai/tests/model_validation/sklearn/TrainingTestDegradation.html):** Evaulates how much the models' performance degrade when moving from the training to the test datasets.\n", - "- **[`OverfitDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html):** Dives into detecting overfitting beyond performance degradation by examining learning curves, model complexity v.s. generalization, and cross-validation results." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store our desired 2 tests\n", - "overfit_testing = [\n", - " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", - "]" + "Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data." ] }, { @@ -1278,40 +1262,22 @@ "metadata": {}, "outputs": [], "source": [ - "# Run and log our 2 tests for both models for the training and testing datasets\n", - "for test in overfit_testing:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", - " \"model\" : [vm_xgb_model,vm_log_model]\n", - " }\n", - " ).log()" + "vm.tests.run_test(\n", + " test_id=\"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\",\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " }\n", + ").log()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "Let's also conduct *robustness* and *stability* testing of the two models. Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n", + "Let's also conduct *robustness* and *stability* testing of the two models with the [`RobustnessDiagnosis` test](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html).\n", "\n", - "We'll select the following two tests:\n", - "\n", - "- **[`RobustnessDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html):** Evaluates whether the models are resilient to small *perturbations* (small, controlled modifications) or *variations* (broader subgroup-level or distributional changes) in input data.\n", - "- **[`WeakspotsDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/WeakspotsDiagnosis.html):** Identifies specific conditions where the models consistently underperform." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store our desired 2 tests\n", - "stab_robust = [\n", - " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", - " ]" + "Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n" ] }, { @@ -1320,15 +1286,13 @@ "metadata": {}, "outputs": [], "source": [ - "# Run and log our 2 tests for both models for the training and testing datasets\n", - "for test in stab_robust:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", - " \"model\" : [vm_xgb_model,vm_log_model]\n", - " },\n", - " ).log()" + "vm.tests.run_test(\n", + " test_id=\"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", + " input_grid={\n", + " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", + " \"model\" : [vm_xgb_model,vm_log_model]\n", + " },\n", + ").log()" ] }, { From 72596a02c8704fe221e8fad061509fd310eeeb38 Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Thu, 27 Mar 2025 17:50:07 -0700 Subject: [PATCH 87/90] Final edits? --- .../validate_application_scorecard.ipynb | 189 +++++++++++++++--- 1 file changed, 164 insertions(+), 25 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index a71e0832e..d8475a160 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -25,6 +25,73 @@ "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", + " - [Register a sample model](#toc2_1_) \n", + " - [Assign validator credentials](#toc2_1_1_) \n", + " - [Install the ValidMind Library](#toc2_2_) \n", + " - [Initialize the ValidMind Library](#toc2_3_) \n", + " - [Get your code snippet](#toc2_3_1_) \n", + " - [Importing the champion model](#toc2_4_) \n", + " - [Load the sample dataset](#toc2_5_) \n", + " - [Preprocess the dataset](#toc2_5_1_) \n", + " - [Apply feature engineering to the dataset](#toc2_5_2_) \n", + " - [Split the feature engineered dataset](#toc2_6_) \n", + "- [Developing potential challenger models](#toc3_) \n", + " - [Train potential challenger models](#toc3_1_) \n", + " - [Random forest classification model](#toc3_1_1_) \n", + " - [Logistic regression model](#toc3_1_2_) \n", + " - [Extract predicted probabilities](#toc3_2_) \n", + " - [Compute binary predictions](#toc3_2_1_) \n", + "- [Initializing the ValidMind objects](#toc4_) \n", + " - [Initialize the ValidMind datasets](#toc4_1_) \n", + " - [Initialize the model objects](#toc4_2_) \n", + " - [Assign predictions](#toc4_3_) \n", + " - [Compute credit risk scores](#toc4_4_) \n", + "- [Run data quality tests](#toc5_) \n", + " - [Run and log an individual data quality test](#toc5_1_) \n", + " - [Log multiple data quality tests](#toc5_2_) \n", + " - [Run data quality comparison tests](#toc5_3_) \n", + "- [Run performance tests](#toc6_) \n", + " - [Identify performance tests](#toc6_1_) \n", + " - [Run and log an individual performance test](#toc6_2_) \n", + " - [Log multiple performance tests](#toc6_3_) \n", + " - [Evaluate performance of the champion model](#toc6_4_) \n", + " - [Evaluate performance of challenger models](#toc6_5_) \n", + " - [Enable custom context for test descriptions](#toc6_5_1_) \n", + " - [Run performance comparison tests](#toc6_5_2_) \n", + "- [Adjust a ValidMind test](#toc7_) \n", + "- [Run diagnostic tests](#toc8_) \n", + "- [Run feature importance tests](#toc9_) \n", + "- [Implement a custom test](#toc10_) \n", + "- [Verify test runs](#toc11_) \n", + "- [Next steps](#toc12_) \n", + " - [Work with your validation report](#toc12_1_) \n", + " - [Discover more learning resources](#toc12_2_) \n", + "- [Upgrade ValidMind](#toc13_) \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", @@ -36,6 +103,8 @@ "cell_type": "markdown", "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", @@ -47,6 +116,8 @@ "cell_type": "markdown", "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", @@ -60,6 +131,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Key concepts\n", "\n", "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", @@ -88,6 +161,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Setting up" ] }, @@ -95,6 +170,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Register a sample model\n", "\n", "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", @@ -119,26 +196,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "For this notebook, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\n", - "\n", - "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", + "\n", "\n", - "2. In the left sidebar, navigate to **Inventory** and click **+ Register Model**.\n", - "\n", - "3. Enter the model details and click **Continue**. ([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:\n", - "\n", - " - Documentation template: `Credit Risk Scorecard`\n", - " - Use case: `Credit Risk — CECL`\n", - "\n", - " You can fill in other options according to your preference." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ "#### Assign validator credentials\n", "\n", "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", @@ -160,6 +219,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Install the ValidMind Library\n", "\n", "
Recommended Python versions\n", @@ -182,6 +243,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind Library\n", "\n", "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." @@ -191,6 +254,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Get your code snippet\n", "\n", "1. In a browser, [log in to ValidMind](https://docs.validmind.ai/guide/configuration/log-in-to-validmind.html).\n", @@ -229,6 +294,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Importing the champion model\n", "\n", "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" @@ -262,6 +329,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Load the sample dataset\n", "\n", "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", @@ -288,6 +357,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Preprocess the dataset\n", "\n", "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", @@ -311,6 +382,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Apply feature engineering to the dataset\n", "\n", "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", @@ -335,6 +408,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Split the feature engineered dataset\n", "\n", "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", @@ -394,6 +469,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Developing potential challenger models" ] }, @@ -401,6 +478,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Train potential challenger models\n", "\n", "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", @@ -412,6 +491,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Random forest classification model\n", "\n", "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", @@ -442,6 +523,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Logistic regression model\n", "\n", "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", @@ -480,6 +563,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Extract predicted probabilities\n", "\n", "With our challenger models trained, let's extract the predicted probabilities from our three models:" @@ -508,6 +593,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Compute binary predictions\n", "\n", "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", @@ -541,6 +628,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Initializing the ValidMind objects" ] }, @@ -548,6 +637,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the ValidMind datasets\n", "\n", "Before you can run tests, you'll need 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", @@ -612,6 +703,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Initialize the model objects\n", "\n", "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", @@ -648,6 +741,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Assign predictions\n", "\n", "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", @@ -707,6 +802,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Compute credit risk scores\n", "\n", "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" @@ -739,6 +836,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run data quality tests\n", "\n", "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately.\n", @@ -789,6 +888,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run and log an individual data quality test\n", "\n", "Next, we'll use our previously initialized preprocessed dataset (`vm_preprocess_dataset`) as input to run an individual test, then log the result to the ValidMind Platform.\n", @@ -826,7 +927,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Log data quality tests\n", + "\n", + "\n", + "### Log multiple data quality tests\n", "\n", "Now that we understand how to run a test with ValidMind, we want to run all the tests that were returned for our `classification` tasks focusing on `data_quality`.\n", "\n", @@ -870,6 +973,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run data quality comparison tests\n", "\n", "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" @@ -894,6 +999,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run performance tests\n", "\n", "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models." @@ -903,6 +1010,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Identify performance tests\n", "\n", "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" @@ -922,6 +1031,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Run and log an individual performance test\n", "\n", "Before we run our batch of performance tests, we'll use our previously initialized testing dataset (`vm_test_ds`) as input to run an individual test, then log the result to the ValidMind Platform.\n", @@ -949,7 +1060,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "### Log performance tests\n", + "\n", + "\n", + "### Log multiple performance tests\n", "\n", "We only want to run a few other tests that were returned for our `classification` tasks focusing on `model_performance`, so we'll isolate the specific tests we want to batch run in `mpt`:\n", "\n", @@ -981,6 +1094,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Evaluate performance of the champion model\n", "\n", "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" @@ -1005,6 +1120,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Evaluate performance of challenger models\n", "\n", "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", @@ -1031,6 +1148,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Enable custom context for test descriptions" ] }, @@ -1149,6 +1268,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "#### Run performance comparison tests\n", "\n", "With the use case context set, we'll run each test in `mpt_chall` once for each model with the same `vm_test_ds` dataset to compare them:" @@ -1182,6 +1303,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Adjust a ValidMind test\n", "\n", "Let's dig deeper into the `MinimumF1Score` test we ran previously in Run performance tests to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", @@ -1231,6 +1354,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run diagnostic tests\n", "\n", "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", @@ -1299,6 +1424,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Run feature importance tests\n", "\n", "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", @@ -1337,6 +1464,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Implement a custom test\n", "\n", "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", @@ -1445,6 +1574,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Verify test runs\n", "\n", "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", @@ -1553,16 +1684,16 @@ " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", " },\n", " 'validmind.model_validation.ModelMetadata': {\n", - " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " 'input_grid': {'model': ['vm_xgb_model', 'rf_model']}\n", " },\n", " 'validmind.model_validation.sklearn.ModelParameters': {\n", - " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", + " 'input_grid': {'model': ['vm_xgb_model', 'rf_model']}\n", " },\n", " 'validmind.model_validation.sklearn.ROCCurve': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']}\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['vm_xgb_model']}\n", " },\n", " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']},\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['vm_xgb_model']},\n", " 'params': {'min_threshold': 0.5}\n", " }\n", "}" @@ -1605,6 +1736,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "## Next steps" ] }, @@ -1612,6 +1745,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "\n", "### Work with your validation report\n", "\n", "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report:\n", @@ -1627,6 +1762,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "\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", @@ -1641,6 +1778,8 @@ "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", From 250da627e05bd2013eebe8c663166a7fba6e26ce Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 28 Mar 2025 09:10:31 -0700 Subject: [PATCH 88/90] xgb_model > xgb_model_developer_champion --- .../model_validation/validate_application_scorecard.ipynb | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb index d8475a160..a2a6d900e 100644 --- a/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb +++ b/notebooks/code_samples/model_validation/validate_application_scorecard.ipynb @@ -1684,16 +1684,16 @@ " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", " },\n", " 'validmind.model_validation.ModelMetadata': {\n", - " 'input_grid': {'model': ['vm_xgb_model', 'rf_model']}\n", + " 'input_grid': {'model': ['xgb_model_developer_champion', 'rf_model']}\n", " },\n", " 'validmind.model_validation.sklearn.ModelParameters': {\n", - " 'input_grid': {'model': ['vm_xgb_model', 'rf_model']}\n", + " 'input_grid': {'model': ['xgb_model_developer_champion', 'rf_model']}\n", " },\n", " 'validmind.model_validation.sklearn.ROCCurve': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['vm_xgb_model']}\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model_developer_champion']}\n", " },\n", " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['vm_xgb_model']},\n", + " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model_developer_champion']},\n", " 'params': {'min_threshold': 0.5}\n", " }\n", "}" From 7162288a9fe212d49106b6e0e2c05b19d011eb5d Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 28 Mar 2025 09:30:10 -0700 Subject: [PATCH 89/90] Moving the introductory notebooks to own PR --- .../111-import_champion_model.ipynb | 468 ------- .../112-develop_challenger_models.ipynb | 726 ----------- .../113-perform_validation_tests.ipynb | 942 -------------- .../114-finalize_testing_reporting.ipynb | 1081 ----------------- .../model_validation/compliance-summary.png | Bin 61626 -> 0 bytes .../inserted-minimum-f1-scores.png | Bin 110789 -> 0 bytes .../link-validator-evidence.png | Bin 398858 -> 0 bytes .../selecting-minimum-f1-scores.png | Bin 327835 -> 0 bytes .../model_validation/xgb_model_champion.pkl | Bin 117919 -> 0 bytes 9 files changed, 3217 deletions(-) delete mode 100644 notebooks/tutorials/model_validation/111-import_champion_model.ipynb delete mode 100644 notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb delete mode 100644 notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb delete mode 100644 notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb delete mode 100644 notebooks/tutorials/model_validation/compliance-summary.png delete mode 100644 notebooks/tutorials/model_validation/inserted-minimum-f1-scores.png delete mode 100644 notebooks/tutorials/model_validation/link-validator-evidence.png delete mode 100644 notebooks/tutorials/model_validation/selecting-minimum-f1-scores.png delete mode 100644 notebooks/tutorials/model_validation/xgb_model_champion.pkl diff --git a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb b/notebooks/tutorials/model_validation/111-import_champion_model.ipynb deleted file mode 100644 index 730a25129..000000000 --- a/notebooks/tutorials/model_validation/111-import_champion_model.ipynb +++ /dev/null @@ -1,468 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "248883b1", - "metadata": {}, - "source": [ - "# ValidMind for model validation — 111 Import the champion model\n", - "\n", - "Learn how to use ValidMind for your end-to-end model validation process based on common scenarios with our series of four introductory notebooks. In this first notebook, set up the ValidMind Library in preparation for validating a champion model.\n", - "\n", - "These notebooks use an application scorecard model as an example, but the same principles shown here apply to other model types." - ] - }, - { - "cell_type": "markdown", - "id": "e1657d54", - "metadata": {}, - "source": [ - "## Introduction\n", - "\n", - "Model validation aims to independently assess the compliance of *champion models* created by model developers with regulatory guidance by conducting thorough testing and analysis, potentially including the use of challenger models to benchmark performance. Assessments, presented in the form of a validation report, typically include *model findings* and recommendations to address those issues.\n", - "\n", - "An *application scorecard model* is a type of statistical model used in credit scoring to evaluate the creditworthiness of potential borrowers by generating a score based on various characteristics of an applicant such as credit history, income, employment status, and other relevant financial data.\n", - "\n", - " - This score assists lenders in making informed decisions about whether to approve or reject loan applications, as well as in determining the terms of the loan, including interest rates and credit limits.\n", - " - Effective validation of application scorecard models ensures that lenders can manage risk efficiently while maintaining a fast and transparent loan application process for applicants." - ] - }, - { - "cell_type": "markdown", - "id": "434bea70", - "metadata": {}, - "source": [ - "## 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 comparison and other validation tests, and then use the ValidMind Platform to submit compliance assessments of champion models via comprehensive validation reports. Together, these products simplify model risk management, facilitate compliance with regulations and institutional standards, and enhance collaboration between yourself and model developers." - ] - }, - { - "cell_type": "markdown", - "id": "662d6027", - "metadata": {}, - "source": [ - "### 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": "5a1a2fa8", - "metadata": {}, - "source": [ - "### 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, create a free ValidMind account.\n", - "

\n", - "Signing up is FREE — Register with ValidMind
" - ] - }, - { - "cell_type": "markdown", - "id": "aa9b1d8e", - "metadata": {}, - "source": [ - "### Key concepts\n", - "\n", - "**Validation report**: A comprehensive and structured assessment of a model’s development and performance, focusing on verifying its integrity, appropriateness, and alignment with its intended use. It includes analyses of model assumptions, data quality, performance metrics, outcomes of testing procedures, and risk considerations. The validation report supports transparency, regulatory compliance, and informed decision-making by documenting the validator’s independent review and conclusions.\n", - "\n", - "**Validation report template**: Serves as a standardized framework for conducting and documenting model validation activities. It outlines the required sections, recommended analyses, and expected validation tests, ensuring consistency and completeness across validation reports. The template helps guide validators through a systematic review process while promoting comparability and traceability of validation outcomes.\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.\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/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." - ] - }, - { - "cell_type": "markdown", - "id": "057c1417", - "metadata": {}, - "source": [ - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "id": "ef1f5ce4", - "metadata": {}, - "source": [ - "### Register a sample model\n", - "\n", - "In a usual model lifecycle, a champion model will have been independently registered in your model inventory and submitted to you for validation by your model development team as part of the effective challenge process. (**Learn more:** [Submit for approval](https://docs.validmind.ai/guide/model-documentation/submit-for-approval.html))\n", - "\n", - "For this series of notebooks, we'll have you register a dummy model in the ValidMind Platform inventory and assign yourself as the validator to familiarize you with the ValidMind interface and circumvent the need for an existing model:\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 **Continue**. ([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:\n", - "\n", - " - Documentation template: `Credit Risk Scorecard`\n", - " - Use case: `Credit Risk — CECL`\n", - "\n", - " You can fill in other options according to your preference." - ] - }, - { - "cell_type": "markdown", - "id": "9743927c", - "metadata": {}, - "source": [ - "#### Assign validator credentials\n", - "\n", - "In order to log tests as a validator instead of as a developer, on the model details page that appears after you've successfully registered your sample model:\n", - "\n", - "1. Remove yourself as a developer: \n", - "\n", - " - Click on the **DEVELOPERS** tile.\n", - " - Click the **x** next to your name to remove yourself from that model's role.\n", - " - Click **Save** to apply your changes to that role.\n", - "\n", - "2. Add yourself as a validator: \n", - "\n", - " - Click on the **VALIDATORS** tile.\n", - " - Select your name from the drop-down menu.\n", - " - Click **Save** to apply your changes to that role." - ] - }, - { - "cell_type": "markdown", - "id": "37a94e48", - "metadata": {}, - "source": [ - "### 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, - "id": "cef99db1", - "metadata": {}, - "outputs": [], - "source": [ - "%pip install -q validmind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind Library\n", - "\n", - "ValidMind generates a unique _code snippet_ for each registered model to connect with your validation environment. You initialize the ValidMind Library with this code snippet, which ensures that your test results are uploaded to the correct model when you run the notebook." - ] - }, - { - "cell_type": "markdown", - "id": "30803184", - "metadata": {}, - "source": [ - "#### Get your code snippet\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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": "d5de450a", - "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": [ - "## Getting to know ValidMind" - ] - }, - { - "cell_type": "markdown", - "id": "d22ef356", - "metadata": {}, - "source": [ - "### Preview the validation report 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 model validation. A template predefines sections for your validation report and provides a general outline to follow, making the validation process much easier.\n", - "\n", - "You will attach evidence to this template in the form of risk assessment notes, findings, and test results 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:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "38592086", - "metadata": {}, - "outputs": [], - "source": [ - "vm.preview_template()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### View validation report in the ValidMind Platform\n", - "\n", - "Next, let's head to the ValidMind Platform to see the template in action:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Click on the **Validation Report** for your model and note:\n", - "\n", - " - [ ] The risk assessment compliance summary at the top of the report (screenshot below)\n", - " - [ ] How the structure of the validation report reflects the previewed template\n", - "\n", - " \"Screenshot\n", - "

" - ] - }, - { - "cell_type": "markdown", - "id": "943a45d6", - "metadata": {}, - "source": [ - "## Importing the champion model\n", - "\n", - "With the ValidMind Library set up and ready to go, let's go ahead and import the champion model submitted by the model development team in the format of a `.pkl` file: **[xgb_model_champion.pkl](xgb_model_champion.pkl)**\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "3f9e197e", - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "b6887d38", - "metadata": {}, - "outputs": [], - "source": [ - "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Load the sample dataset\n", - "\n", - "Let's next import the public [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset from Kaggle, which was used to develop the dummy champion model.\n", - "\n", - "- We'll use this dataset to review steps that should have been conducted during the initial development and documentation of the model to ensure that the model was built correctly.\n", - "- By independently performing steps such as preprocessing and feature engineering, we can confirm whether the model was built using appropriate and properly processed data.\n", - "\n", - "To be able to use the dataset, you'll need to import the dataset and load it into a pandas [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html), a two-dimensional tabular data structure that makes use of rows and columns:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "40e73b7c", - "metadata": {}, - "outputs": [], - "source": [ - "from validmind.datasets.credit_risk import lending_club\n", - "\n", - "df = lending_club.load_data(source=\"offline\")\n", - "df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Preprocess the dataset\n", - "\n", - "We'll first quickly preprocess the dataset for data quality testing purposes using `lending_club.preprocess`. This function performs the following operations:\n", - "\n", - "- Filters the dataset to include only loans for debt consolidation or credit card purposes\n", - "- Removes loans classified under the riskier grades \"F\" and \"G\"\n", - "- Excludes uncommon home ownership types and standardizes employment length and loan terms into numerical formats\n", - "- Discards unnecessary fields and any entries with missing information to maintain a clean and robust dataset for modeling" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "e436e862", - "metadata": {}, - "outputs": [], - "source": [ - "preprocess_df = lending_club.preprocess(df)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Apply feature engineering to the dataset\n", - "\n", - "Feature engineering improves the dataset's structure to better match what our model expects, and ensures that the model performs optimally by leveraging additional insights from raw data.\n", - "\n", - "We'll apply the following transformations using the `ending_club.feature_engineering()` function to optimize the dataset for predictive modeling in our application scorecard:\n", - "\n", - "- **WoE encoding**: Converts both numerical and categorical features into Weight of Evidence (WoE) values. WoE is a statistical measure used in scorecard modeling that quantifies the relationship between a predictor variable and the binary target variable. It calculates the ratio of the distribution of good outcomes to the distribution of bad outcomes for each category or bin of a feature. This transformation helps to ensure that the features are predictive and consistent in their contribution to the model.\n", - "- **Integration of WoE bins**: Ensures that the WoE transformed values are integrated throughout the dataset, replacing the original feature values while excluding the target variable from this transformation. This transformation is used to maintain a consistent scale and impact of each variable within the model, which helps make the predictions more stable and accurate." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "9c6b641c", - "metadata": {}, - "outputs": [], - "source": [ - "fe_df = lending_club.feature_engineering(preprocess_df)\n", - "fe_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 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": "10272aa9", - "metadata": {}, - "outputs": [], - "source": [ - "%pip show validmind" - ] - }, - { - "cell_type": "markdown", - "id": "upgrade-version-d64591ca-3073-4b3e-9586-d3577adda203", - "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": "6e12584a", - "metadata": {}, - "source": [ - "You may need to restart your kernel after running the upgrade package for changes to be applied." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## In summary\n", - "\n", - "In this first notebook, you learned how to:\n", - "\n", - "- [ ] Register a model within the ValidMind Platform and assign yourself as the validator\n", - "- [ ] Install and initialize the ValidMind Library\n", - "- [ ] Preview the validation report template for your model\n", - "- [ ] Locate the validation report for your model within the ValidMind Platform\n" - ] - }, - { - "cell_type": "markdown", - "id": "78f8161b", - "metadata": {}, - "source": [ - "## Next steps\n", - "\n", - "### Develop potential challenger models\n", - "\n", - "Now that the ValidMind Library is connected to your model in the ValidMind Library with the correct template applied, we can go ahead and start the model validation process: **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "ValidMind Library", - "language": "python", - "name": "validmind" - }, - "language_info": { - "name": "python", - "version": "3.10.13" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb b/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb deleted file mode 100644 index 68ac41403..000000000 --- a/notebooks/tutorials/model_validation/112-develop_challenger_models.ipynb +++ /dev/null @@ -1,726 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# ValidMind for model validation — 112 Develop potential challenger models\n", - "\n", - "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this second notebook, develop potential challenger models and then pass your models and their predictions to ValidMind.\n", - "\n", - "A *challenger model* is an alternate model that attempt to outperform the champion model, ensuring that the best performing fit-for-purpose model is always considered for deployment. Challenger models also help avoid over-reliance on a single model, and allow testing of new features, algorithms, or data sources without disrupting the production lifecycle." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Prerequisites\n", - "\n", - "In order to continue on the model validation journey with notebook, you'll need to first have:\n", - "\n", - "- [ ] Registered a model within the ValidMind Platform and granted yourself access to the model as a validator\n", - "- [ ] Installed the ValidMind Library in your local environment, allowing you to access all its features\n", - "\n", - "
Need help with the above steps?\n", - "

\n", - "Refer to the first notebook in this series: 111 Import the champion model
\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Setting up" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind Library\n", - "\n", - "First, let's connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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": [ - "# Make sure the ValidMind Library is installed\n", - "\n", - "%pip install -q validmind\n", - "\n", - "# 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": [ - "### Import the champion model\n", - "\n", - "Next, we'll import the champion model submitted by the model development team as we used in the last notebook (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model\n", - "\n", - "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Lending Club dataset from Kaggle\n", - "from validmind.datasets.credit_risk import lending_club\n", - "\n", - "df = lending_club.load_data(source=\"offline\")\n", - "df.head()\n", - "\n", - "# Preprocess the dataset for data quality testing purposes\n", - "preprocess_df = lending_club.preprocess(df)\n", - "\n", - "# Apply feature engineering to the dataset\n", - "fe_df = lending_club.feature_engineering(preprocess_df)\n", - "fe_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Split the feature engineered dataset\n", - "\n", - "With our dummy model imported and our independently preprocessed and feature engineered dataset ready to go, let's now **spilt our dataset into train and test** to start the validation testing process.\n", - "\n", - "Splitting our dataset into training and testing is essential for proper validation testing, as this helps assess how well the model generalizes to unseen data:\n", - "\n", - "- We begin by dividing our data, which is based on Weight of Evidence (WoE) features, into training and testing sets (`train_df`, `test_df`).\n", - "- With `lending_club.split`, we employ a simple random split, randomly allocating data points to each set to ensure a mix of examples in both." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Split the data\n", - "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", - "\n", - "x_train = train_df.drop(lending_club.target_column, axis=1)\n", - "y_train = train_df[lending_club.target_column]\n", - "\n", - "x_test = test_df.drop(lending_club.target_column, axis=1)\n", - "y_test = test_df[lending_club.target_column]\n", - "\n", - "# Now let's apply the order of features from the champion model construction\n", - "x_train = x_train[cols_when_model_builds]\n", - "x_test = x_test[cols_when_model_builds]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cols_use = ['annual_inc_woe',\n", - " 'verification_status_woe',\n", - " 'emp_length_woe',\n", - " 'installment_woe',\n", - " 'term_woe',\n", - " 'home_ownership_woe',\n", - " 'purpose_woe',\n", - " 'open_acc_woe',\n", - " 'total_acc_woe',\n", - " 'int_rate_woe',\n", - " 'sub_grade_woe',\n", - " 'grade_woe','loan_status']\n", - "\n", - "\n", - "train_df = train_df[cols_use]\n", - "test_df = test_df[cols_use]\n", - "test_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Investigate potential challenger models\n", - "\n", - "We're curious how alternate models compare to our champion model, so let's train two challenger models as basis for our testing.\n", - "\n", - "Our selected options below offer decreased complexity in terms of implementation — such as lessened manual preprocessing — which can reduce the amount of risk for implementation. However, model risk is not calculated in isolation from a single factor, but rather in consideration with trade-offs in predictive performance, ease of interpretability, and overall alignment with business objectives." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Random forest classification model\n", - "\n", - "A *random forest classification model* is an ensemble machine learning algorithm that uses multiple decision trees to classify data. In ensemble learning, multiple models are combined to improve prediction accuracy and robustness.\n", - "\n", - "Random forest classification models generally have higher accuracy because they capture complex, non-linear relationships, but as a result they lack transparency in their predictions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Random Forest Classification model\n", - "from sklearn.ensemble import RandomForestClassifier\n", - "\n", - "# Create the model instance with 50 decision trees\n", - "rf_model = RandomForestClassifier(\n", - " n_estimators=50,\n", - " random_state=42,\n", - ")\n", - "\n", - "# Train the model\n", - "rf_model.fit(x_train, y_train)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Logistic regression model\n", - "\n", - "A *logistic regression model* is a statistical machine learning algorithm that uses a linear equation (straight-line relationship between variables) and the logistic function (or sigmoid function, which maps any real-valued number to a range between `0` and `1`) to classify data. In statistical modeling, a single equation is used to estimate the probability of an outcome based on input features.\n", - "\n", - "Logistic regression models are simple and interpretable because they provide clear probability estimates and feature coefficients (numerical value that represents the influence of a particular input feature on the model's prediction), but they may struggle with capturing complex, non-linear relationships in the data." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Logistic Regression model\n", - "from sklearn.linear_model import LogisticRegression\n", - "\n", - "# Logistic Regression grid params\n", - "log_reg_params = {\n", - " \"penalty\": [\"l1\", \"l2\"],\n", - " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", - " \"solver\": [\"liblinear\"],\n", - "}\n", - "\n", - "# Grid search for Logistic Regression\n", - "from sklearn.model_selection import GridSearchCV\n", - "\n", - "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", - "grid_log_reg.fit(x_train, y_train)\n", - "\n", - "# Logistic Regression best estimator\n", - "log_reg = grid_log_reg.best_estimator_\n", - "log_reg" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Extract predicted probabilities\n", - "\n", - "With our challenger models trained, let's extract the predicted probabilities from our three models:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", - "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Random forest classification model\n", - "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", - "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", - "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Compute binary predictions\n", - "\n", - "Next, we'll convert the probability predictions from our three models into a binary, based on a threshold of `0.3`:\n", - "\n", - "- If the probability is greater than `0.3`, the prediction becomes `1` (positive).\n", - "- Otherwise, it becomes `0` (negative)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cut_off_threshold = 0.3\n", - "\n", - "# Champion — Application scorecard model\n", - "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", - "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Random forest classification model\n", - "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", - "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", - "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Initializing the ValidMind objects" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind datasets\n", - "\n", - "Before you can run tests, you'll need 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", - "- **`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." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the raw dataset\n", - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the preprocessed dataset\n", - "vm_preprocess_dataset = vm.init_dataset(\n", - " dataset=preprocess_df,\n", - " input_id=\"preprocess_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the feature engineered dataset\n", - "vm_fe_dataset = vm.init_dataset(\n", - " dataset=fe_df,\n", - " input_id=\"fe_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the training dataset\n", - "vm_train_ds = vm.init_dataset(\n", - " dataset=train_df,\n", - " input_id=\"train_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the test dataset\n", - "vm_test_ds = vm.init_dataset(\n", - " dataset=test_df,\n", - " input_id=\"test_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "After initialization, you can pass the ValidMind `Dataset` objects `vm_raw_dataset`, `vm_preprocess_dataset`, `vm_fe_dataset`, `vm_train_ds`, and `vm_test_ds` into any ValidMind tests." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the model objects\n", - "\n", - "You'll also need to initialize a ValidMind model object (`vm_model`) that can be passed to other functions for analysis and tests on the data for each of our three models.\n", - "\n", - "You simply initialize this model object with [`vm.init_model()`](https://docs.validmind.ai/validmind/validmind.html#init_model):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the champion application scorecard model\n", - "vm_xgb_model = vm.init_model(\n", - " xgb_model,\n", - " input_id=\"xgb_model_developer_champion\",\n", - ")\n", - "\n", - "# Initialize the challenger random forest classification model\n", - "vm_rf_model = vm.init_model(\n", - " rf_model,\n", - " input_id=\"rf_model\",\n", - ")\n", - "\n", - "# Initialize the challenger logistic regression model\n", - "vm_log_model = vm.init_model(\n", - " log_reg,\n", - " input_id=\"log_model\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Assign predictions\n", - "\n", - "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold described in the Compute binary predictions step above.\n", - "\n", - "- The [`assign_predictions()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#VMDataset.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." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=train_xgb_binary_predictions,\n", - " prediction_probabilities=train_xgb_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=test_xgb_binary_predictions,\n", - " prediction_probabilities=test_xgb_prob,\n", - ")\n", - "\n", - "# Challenger — Random forest classification model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=train_rf_binary_predictions,\n", - " prediction_probabilities=train_rf_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=test_rf_binary_predictions,\n", - " prediction_probabilities=test_rf_prob,\n", - ")\n", - "\n", - "\n", - "# Challenger — Logistic regression model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=train_log_binary_predictions,\n", - " prediction_probabilities=train_log_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=test_log_binary_predictions,\n", - " prediction_probabilities=test_log_prob,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Compute credit risk scores\n", - "\n", - "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Compute the scores\n", - "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", - "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", - "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", - "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", - "train_log_scores = lending_club.compute_scores(train_log_prob)\n", - "test_log_scores = lending_club.compute_scores(test_log_prob)\n", - "\n", - "# Assign scores to the datasets\n", - "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", - "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", - "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", - "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", - "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", - "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Enabling custom context for test descriptions" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "When you run ValidMind tests, test descriptions are automatically generated with LLM using the test results, the test name, and the static test definitions provided in the test’s docstring. While this metadata offers valuable high-level overviews of tests, insights produced by the LLM-based descriptions may not always align with your specific use cases or incorporate organizational policy requirements.\n", - "\n", - "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned.\n", - "\n", - "
Want to learn more about setting custom context for LLM-generated test descriptions?\n", - "

\n", - "Refer to our extended walkthrough notebook: Add context to LLM-generated test descriptions\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Review default LLM-generated descriptions\n", - "\n", - "By default, custom context for LLM-generated descriptions is disabled, meaning that the output will not include any additional context.\n", - "\n", - "Let's generate an initial test description for the `HighPearsonCorrelation` test with our `vm_test_ds` dataset for comparison with a later iteration to demonstrate. We'll go over running tests in more detail in a subsequent notebook:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### Enable use case context\n", - "\n", - "To enable custom use case context, set the `VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED` environment variable to `1`.\n", - "\n", - "This is a global setting that will affect all tests for your linked model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"" - ] - }, - { - "cell_type": "markdown", - "id": "0d1e90ba", - "metadata": {}, - "source": [ - "Enabling use case context allows you to pass in additional context to the LLM-generated text descriptions within `context`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "87f7fb2e", - "metadata": {}, - "outputs": [], - "source": [ - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "id": "d6e38bd2", - "metadata": {}, - "source": [ - "With the use case context set, generate an updated test description for the `HighPearsonCorrelation` test for comparision with default output:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.data_validation.HighPearsonCorrelation\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " }\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## In summary\n", - "\n", - "In this second notebook, you learned how to:\n", - "\n", - "- [ ] Initialize ValidMind datasets\n", - "- [ ] Initialize ValidMind model objects\n", - "- [ ] Assign predictions and probabilities to your ValidMind datasets\n", - "- [ ] Enable custom context for test descriptions generated by ValidMind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Next steps\n", - "\n", - "### Perform model validation tests\n", - "\n", - "Now that you're familiar with the basics of using the ValidMind Library, let's learn how to identify and run validation tests: **[113 Perform model validation tests](113-perform_validation_tests.ipynb)**" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "ValidMind Library", - "language": "python", - "name": "validmind" - }, - "language_info": { - "name": "python", - "version": "3.10.13" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb b/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb deleted file mode 100644 index eb96a5f74..000000000 --- a/notebooks/tutorials/model_validation/113-perform_validation_tests.ipynb +++ /dev/null @@ -1,942 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# ValidMind for model validation — 113 Perform model validation tests\n", - "\n", - "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this third notebook, you'll identify relevant validation tests, run them, then add the log results to your validation report.\n", - "\n", - "You'll become familiar with the tests available in ValidMind, as well as how to run them. Running tests during model validation is crucial to the effective challenge process, as we want to independently evaluate the evidence and assessments provided by the model development team.\n", - "\n", - "While running our tests, we'll focus on:\n", - "\n", - "- Ensuring that data used for training and testing the model is of appropriate data quality\n", - "- Ensuring that the raw data has been preprocessed appropriately and that the resulting feature engineered dataset reflects this\n", - "- Comprehensive testing around model performance of both the champion ahd challenger models\n", - "\n", - "**For a full list of out-of-the-box tests,** refer to our [Test descriptions](https://docs.validmind.ai/developer/model-testing/test-descriptions.html) or try the interactive [Test sandbox](https://docs.validmind.ai/developer/model-testing/test-sandbox.html)." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Prerequisites\n", - "\n", - "In order to log tests results or evidence to your validation report with this notebook, you'll need to first have:\n", - "\n", - "- [ ] Registered a model within the ValidMind Platform and granted yourself access to the model as a validator\n", - "- [ ] Installed the ValidMind Library in your local environment, allowing you to access all its features\n", - "- [ ] Learned how to import and initialize datasets for use with ValidMind\n", - "- [ ] Understood the basics of how to enable custom context for test descriptions generated by ValidMind\n", - "\n", - "
Need help with the above steps?\n", - "

\n", - "Refer to the first two notebooks in this series:\n", - "\n", - "
    \n", - "
  1. 111 Import the champion model
    2. \n", - "
  2. 112 Develop potential challenger models
    3. \n", - "
\n", - "\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Setting up\n", - "\n", - "This section should be quite familiar to you — as we performed the same actions in the previous notebook, **[112 Develop potential challenger models](112-develop_challenger_models.ipynb)**." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind Library\n", - "\n", - "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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": [ - "# Make sure the ValidMind Library is installed\n", - "\n", - "%pip install -q validmind\n", - "\n", - "# 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": [ - "### Import the champion model\n", - "\n", - "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model\n", - "\n", - "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Lending Club dataset from Kaggle\n", - "from validmind.datasets.credit_risk import lending_club\n", - "\n", - "df = lending_club.load_data(source=\"offline\")\n", - "df.head()\n", - "\n", - "# Preprocess the dataset for data quality testing purposes\n", - "preprocess_df = lending_club.preprocess(df)\n", - "\n", - "# Apply feature engineering to the dataset\n", - "fe_df = lending_club.feature_engineering(preprocess_df)\n", - "fe_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We'll also spilt our dataset into train and test to start the validation testing process:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Split the data\n", - "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", - "\n", - "x_train = train_df.drop(lending_club.target_column, axis=1)\n", - "y_train = train_df[lending_club.target_column]\n", - "\n", - "x_test = test_df.drop(lending_club.target_column, axis=1)\n", - "y_test = test_df[lending_club.target_column]\n", - "\n", - "# Now let's apply the order of features from the champion model construction\n", - "x_train = x_train[cols_when_model_builds]\n", - "x_test = x_test[cols_when_model_builds]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cols_use = ['annual_inc_woe',\n", - " 'verification_status_woe',\n", - " 'emp_length_woe',\n", - " 'installment_woe',\n", - " 'term_woe',\n", - " 'home_ownership_woe',\n", - " 'purpose_woe',\n", - " 'open_acc_woe',\n", - " 'total_acc_woe',\n", - " 'int_rate_woe',\n", - " 'sub_grade_woe',\n", - " 'grade_woe','loan_status']\n", - "\n", - "\n", - "train_df = train_df[cols_use]\n", - "test_df = test_df[cols_use]\n", - "test_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Train challenger models\n", - "\n", - "Just like in the last notebook, we'll train two challenger models as the basis for our testing:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Random Forest Classification model\n", - "from sklearn.ensemble import RandomForestClassifier\n", - "\n", - "# Create the model instance with 50 decision trees\n", - "rf_model = RandomForestClassifier(\n", - " n_estimators=50,\n", - " random_state=42,\n", - ")\n", - "\n", - "# Train the model\n", - "rf_model.fit(x_train, y_train)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Logistic Regression model\n", - "from sklearn.linear_model import LogisticRegression\n", - "\n", - "# Logistic Regression grid params\n", - "log_reg_params = {\n", - " \"penalty\": [\"l1\", \"l2\"],\n", - " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", - " \"solver\": [\"liblinear\"],\n", - "}\n", - "\n", - "# Grid search for Logistic Regression\n", - "from sklearn.model_selection import GridSearchCV\n", - "\n", - "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", - "grid_log_reg.fit(x_train, y_train)\n", - "\n", - "# Logistic Regression best estimator\n", - "log_reg = grid_log_reg.best_estimator_\n", - "log_reg" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Extract predicted probabilities\n", - "\n", - "With our challenger models trained, let's extract the predicted probabilities from our three models and convert the probability predictions into a binary:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", - "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Random forest classification model\n", - "train_rf_prob = rf_model.predict_proba(x_train)[:, 1]\n", - "test_rf_prob = rf_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", - "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# If probability > 0.3 = 1 (positive)\n", - "cut_off_threshold = 0.3\n", - "\n", - "# Champion — Application scorecard model\n", - "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", - "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Random forest classification model\n", - "train_rf_binary_predictions = (train_rf_prob > cut_off_threshold).astype(int)\n", - "test_rf_binary_predictions = (test_rf_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", - "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind objects\n", - "\n", - "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the raw dataset\n", - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the preprocessed dataset\n", - "vm_preprocess_dataset = vm.init_dataset(\n", - " dataset=preprocess_df,\n", - " input_id=\"preprocess_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the feature engineered dataset\n", - "vm_fe_dataset = vm.init_dataset(\n", - " dataset=fe_df,\n", - " input_id=\"fe_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the training dataset\n", - "vm_train_ds = vm.init_dataset(\n", - " dataset=train_df,\n", - " input_id=\"train_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the test dataset\n", - "vm_test_ds = vm.init_dataset(\n", - " dataset=test_df,\n", - " input_id=\"test_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the champion application scorecard model\n", - "vm_xgb_model = vm.init_model(\n", - " xgb_model,\n", - " input_id=\"xgb_model_developer_champion\",\n", - ")\n", - "\n", - "# Initialize the challenger random forest classification model\n", - "vm_rf_model = vm.init_model(\n", - " rf_model,\n", - " input_id=\"rf_model\",\n", - ")\n", - "\n", - "# Initialize the challenger logistic regression model\n", - "vm_log_model = vm.init_model(\n", - " log_reg,\n", - " input_id=\"log_model\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Assign predictions\n", - "\n", - "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=train_xgb_binary_predictions,\n", - " prediction_probabilities=train_xgb_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=test_xgb_binary_predictions,\n", - " prediction_probabilities=test_xgb_prob,\n", - ")\n", - "\n", - "# Challenger — Random forest classification model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=train_rf_binary_predictions,\n", - " prediction_probabilities=train_rf_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_rf_model,\n", - " prediction_values=test_rf_binary_predictions,\n", - " prediction_probabilities=test_rf_prob,\n", - ")\n", - "\n", - "# Challenger — Logistic regression model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=train_log_binary_predictions,\n", - " prediction_probabilities=train_log_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=test_log_binary_predictions,\n", - " prediction_probabilities=test_log_prob,\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Finally, we'll translate model predictions into actionable scores using probability estimates generated by our trained model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Compute the scores\n", - "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", - "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", - "train_rf_scores = lending_club.compute_scores(train_rf_prob)\n", - "test_rf_scores = lending_club.compute_scores(test_rf_prob)\n", - "train_log_scores = lending_club.compute_scores(train_log_prob)\n", - "test_log_scores = lending_club.compute_scores(test_log_prob)\n", - "\n", - "# Assign scores to the datasets\n", - "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", - "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", - "vm_train_ds.add_extra_column(\"rf_scores\", train_rf_scores)\n", - "vm_test_ds.add_extra_column(\"rf_scores\", test_rf_scores)\n", - "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", - "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Enable use case context\n", - "\n", - "Here, we'll include some initial custom use case context, improving the relevancy, insight, and format of the test descriptions returned in preparation for running our tests:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", - "\n", - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - " Present insights in order from general to specific, with each insight as a single bullet point with bold title.\n", - " You are a model validator and the goal is to identify risk and/or suggest room for improvements or recommendations on what Model Developer should do in order to improve outcomes and reduce risk\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run data quality tests\n", - "\n", - "With everything ready to go, let's explore some of ValidMind's available tests. Using ValidMind’s repository of tests streamlines your validation testing, and helps you ensure that your models are being validated appropriately." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Identify qualitative tests\n", - "\n", - "Let's first explore the list of all available tests in the ValidMind Library with [the `vm.tests.list_tests()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) — we'll then narrow down the tests we want to run from this list." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "Next, we'll use the [`vm.tests.list_tasks_and_tags()` function](https://docs.validmind.ai/validmind/validmind/tests.html#list_tasks_and_tags) to list which `tags` are associated with each `task` type:\n", - "\n", - "- **`tasks`** represent the kind of modeling task associated with a test. Here we'll focus on `classification` 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 `data_quality` tag." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tasks_and_tags()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Then we'll call [`vm.tests.list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to list all the data quality tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests(\n", - " tags=[\"data_quality\"], task=\"classification\"\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "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
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Log data quality tests\n", - "\n", - "We want to run all the tests that are returned, so we'll store the identified tests in `dq` in preparation for batch running these tests and logging their results to the ValidMind Platform:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "dq = vm.tests.list_tests(tags=[\"data_quality\"], task=\"classification\",pretty=False)\n", - "dq" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "\n", - "With our data quality tests identified, let's run our first batch of tests using our previously initialized preprocessed dataset (`vm_preprocess_dataset`) and log their results.\n", - "\n", - "- You run validation tests by calling [the `run_test` function](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) provided by the `validmind.tests` module.\n", - "- Every test result returned by the `run_test()` function has a [`.log()` method](https://docs.validmind.ai/validmind/validmind/vm_models.html#TestResult.log) that can be used to send the test results to the ValidMind Platform:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in dq:\n", - " vm.tests.run_test(\n", - " test,\n", - " inputs={\n", - " \"dataset\": vm_preprocess_dataset\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", - "

\n", - "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Run comparison tests\n", - "\n", - "Next, let's reuse the tests in `dq` to perform comparison tests between the raw (`vm_raw_dataset`) and preprocessed (`vm_preprocess_dataset`) dataset, again logging the results to the ValidMind Platform:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in dq:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"dataset\": [vm_raw_dataset,vm_preprocess_dataset]\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run performance tests\n", - "\n", - "We'll also run some performance tests, beginning with independent testing of our champion application scorecard model, then moving on to our potential challenger models.\n", - "\n", - "Use `vm.tests.list_tests()` to this time identify all the model performance tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "\n", - "vm.tests.list_tests(tags=[\"model_performance\"], task=\"classification\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We'll isolate the specific tests we want to run in `mpt`:\n", - "\n", - "- [`ClassifierPerformance`](https://docs.validmind.ai/tests/model_validation/sklearn/ClassifierPerformance.html)\n", - "- [`ConfusionMatrix`](https://docs.validmind.ai/tests/model_validation/sklearn/ConfusionMatrix.html)\n", - "- [`MinimumAccuracy`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumAccuracy.html)\n", - "- [`MinimumF1Score`](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html)\n", - "- [`ROCCurve`](https://docs.validmind.ai/tests/model_validation/sklearn/ROCCurve.html)\n", - "\n", - "When running individual tests, you can use a custom `result_id` to tag the individual result with a unique identifier by appending this `result_id` to the `test_id` with a `:` separator. We'll append an identifier for our champion model here:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "mpt = [\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion\",\n", - " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion\"\n", - "]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Evaluate performance of the champion model\n", - "\n", - "Now, let's run and log our batch of model performance tests using our testing dataset (`vm_test_ds`) for our champion model:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in mpt:\n", - " vm.tests.run_test(\n", - " test,\n", - " inputs={\n", - " \"dataset\": vm_test_ds, \"model\" : vm_xgb_model, \n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Evaluate performance of challenger models\n", - "\n", - "We've now conducted similar tests as the model development team for our champion model, with the aim of verifying their test results.\n", - "\n", - "Next, let's see how our challenger models compare. We'll use the same batch of tests here as we did in `mpt`, but append a different `result_id` to indicate that these results should be associated with our challenger models:\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "mpt_chall = [\n", - " \"validmind.model_validation.sklearn.ClassifierPerformance:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.ConfusionMatrix:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.MinimumAccuracy:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", - " \"validmind.model_validation.sklearn.ROCCurve:xgboost_champion_vs_challengers\"\n", - "]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We'll also adjust the use case context to focus on this comparison for tests going forward:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", - "\n", - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - "\n", - " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", - " - The metrics for the champion model compared against the challenger models\n", - " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", - "\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "We'll run each test once for each model with the same `vm_test_ds` dataset to compare them:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for test in mpt_chall:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model,vm_rf_model],\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Based on the performance metrics, we can conclude that the random forest classification model is not a viable candidate for our use case and can be disregarded in our tests going forward.\n", - "

\n", - "In the final notebook, we'll dive a bit deeper into some tests comparing our champion application scorecard model and our remaining challenger logistic regression model, including tests that will allow us to customize parameters and thresholds for performance standards.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## In summary\n", - "\n", - "In this third notebook, you learned how to:\n", - "\n", - "- [ ] Identify and run different types of validation tests\n", - "- [ ] Log test results as evidence to the ValidMind Platform\n", - "- [ ] Use tests from ValidMind to evaluate the potential of challenger models" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Next steps\n", - "\n", - "### Finalize testing and reporting\n", - "\n", - "Now that you're familiar with the basics of using the ValidMind Library to run and log validation tests, let's learn how to implement some custom tests and wrap up our validation: **[104 Finalize testing and reporting](114-finalize_testing_reporting.ipynb)**" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "ValidMind Library", - "language": "python", - "name": "validmind" - }, - "language_info": { - "name": "python", - "version": "3.10.13" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb b/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb deleted file mode 100644 index 5446026dc..000000000 --- a/notebooks/tutorials/model_validation/114-finalize_testing_reporting.ipynb +++ /dev/null @@ -1,1081 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# ValidMind for model validation — 114 Finalize testing and reporting\n", - "\n", - "Learn how to use ValidMind for your end-to-end model validation process with our series of four introductory notebooks. In this last notebook, you'll configure and run some custom tests, then add logged test results to your validation report.\n", - "\n", - "As we concluded in [113 Perform validation tests](113-perform_validation_tests.ipynb), our challenger random forest classification model was not a viable candidate for our use case and was eliminated as a contender. We'll finish up by comparing our champion application scorecard model against our remaining challenger logistic regression model, then use the ValidMind Platform to draft our validation report supplemented by our logged test results as evidence." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Prerequisites\n", - "\n", - "In order to finalize the validation testing and reporting for your sample model, you'll need to first have:\n", - "\n", - "- [ ] Registered a model within the ValidMind Platform and granted yourself access to the model as a validator\n", - "- [ ] Installed the ValidMind Library in your local environment, allowing you to access all its features\n", - "- [ ] Learned how to import and initialize datasets for use with ValidMind\n", - "- [ ] Learned how to enable custom context for test descriptions generated by ValidMind\n", - "- [ ] Understood the basics of how to identify and run validation tests\n", - "- [ ] Run data quality and model performance tests for your champion and challenger models, and logged the results of those tests to the ValidMind Platform\n", - "\n", - "
Need help with the above steps?\n", - "

\n", - "Refer to the first three notebooks in this series:\n", - "\n", - "
    \n", - "
  1. 111 Import the champion model
    2. \n", - "
  2. 112 Develop potential challenger models
    3. \n", - "
  3. 113 Perform validation tests
    4. \n", - "
\n", - "\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Setting up\n", - "\n", - "This section should be very familiar to you now — as we performed the same actions in the previous two notebooks in this series." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind Library\n", - "\n", - "As usual, let's first connect up the ValidMind Library to our model we previously registered in the ValidMind Platform:\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 select the model you registered for this \"ValidMind for model validation\" series of notebooks.\n", - "\n", - "3. Go to **Getting Started** and click **Copy snippet to clipboard**.\n", - "\n", - "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": [ - "# Make sure the ValidMind Library is installed\n", - "\n", - "%pip install -q validmind\n", - "\n", - "# 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": [ - "### Import the champion model\n", - "\n", - "Next, we'll import the champion model submitted by the model development team as we used in the last notebooks (**[xgb_model_champion.pkl](xgb_model_champion.pkl)**) and load in the sample [Lending Club](https://www.kaggle.com/datasets/devanshi23/loan-data-2007-2014/data) dataset used to develop the model, and independently preprocess and feature engineer the dataset to confirm that the model was built using appropriate and properly processed data:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import xgboost as xgb\n", - "\n", - "#Load the saved model\n", - "xgb_model = xgb.XGBClassifier()\n", - "xgb_model.load_model(\"xgb_model_champion.pkl\")\n", - "xgb_model\n", - "\n", - "# Ensure that we have to appropriate order in feature names from Champion model and dataset\n", - "cols_when_model_builds = xgb_model.get_booster().feature_names" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Lending Club dataset from Kaggle\n", - "from validmind.datasets.credit_risk import lending_club\n", - "\n", - "df = lending_club.load_data(source=\"offline\")\n", - "df.head()\n", - "\n", - "# Preprocess the dataset for data quality testing purposes\n", - "preprocess_df = lending_club.preprocess(df)\n", - "\n", - "# Apply feature engineering to the dataset\n", - "fe_df = lending_club.feature_engineering(preprocess_df)\n", - "fe_df.head()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Split our dataset into train and test to start the validation testing process\n", - "train_df, test_df = lending_club.split(fe_df, test_size=0.2)\n", - "\n", - "x_train = train_df.drop(lending_club.target_column, axis=1)\n", - "y_train = train_df[lending_club.target_column]\n", - "\n", - "x_test = test_df.drop(lending_club.target_column, axis=1)\n", - "y_test = test_df[lending_club.target_column]\n", - "\n", - "# Now let's apply the order of features from the champion model construction\n", - "x_train = x_train[cols_when_model_builds]\n", - "x_test = x_test[cols_when_model_builds]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "cols_use = ['annual_inc_woe',\n", - " 'verification_status_woe',\n", - " 'emp_length_woe',\n", - " 'installment_woe',\n", - " 'term_woe',\n", - " 'home_ownership_woe',\n", - " 'purpose_woe',\n", - " 'open_acc_woe',\n", - " 'total_acc_woe',\n", - " 'int_rate_woe',\n", - " 'sub_grade_woe',\n", - " 'grade_woe','loan_status']\n", - "\n", - "\n", - "train_df = train_df[cols_use]\n", - "test_df = test_df[cols_use]\n", - "test_df.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Train the challenger model\n", - "\n", - "As we eliminated the random forest classification model as a challenger, we'll only train our logistic regression model here:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Import the Logistic Regression model\n", - "from sklearn.linear_model import LogisticRegression\n", - "\n", - "# Logistic Regression grid params\n", - "log_reg_params = {\n", - " \"penalty\": [\"l1\", \"l2\"],\n", - " \"C\": [0.001, 0.01, 0.1, 1, 10, 100, 1000],\n", - " \"solver\": [\"liblinear\"],\n", - "}\n", - "\n", - "# Grid search for Logistic Regression\n", - "from sklearn.model_selection import GridSearchCV\n", - "\n", - "grid_log_reg = GridSearchCV(LogisticRegression(), log_reg_params)\n", - "grid_log_reg.fit(x_train, y_train)\n", - "\n", - "# Logistic Regression best estimator\n", - "log_reg = grid_log_reg.best_estimator_\n", - "log_reg" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Extract predicted probabilities\n", - "\n", - "With our challenger model trained, let's extract the predicted probabilities from our two models and convert the probability predictions into a binary:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "train_xgb_prob = xgb_model.predict_proba(x_train)[:, 1]\n", - "test_xgb_prob = xgb_model.predict_proba(x_test)[:, 1]\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_prob = log_reg.predict_proba(x_train)[:, 1]\n", - "test_log_prob = log_reg.predict_proba(x_test)[:, 1]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# If probability > 0.3 = 1 (positive)\n", - "cut_off_threshold = 0.3\n", - "\n", - "# Champion — Application scorecard model\n", - "train_xgb_binary_predictions = (train_xgb_prob > cut_off_threshold).astype(int)\n", - "test_xgb_binary_predictions = (test_xgb_prob > cut_off_threshold).astype(int)\n", - "\n", - "# Challenger — Logistic regression model\n", - "train_log_binary_predictions = (train_log_prob > cut_off_threshold).astype(int)\n", - "test_log_binary_predictions = (test_log_prob > cut_off_threshold).astype(int)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Initialize the ValidMind objects\n", - "\n", - "Let's initialize the ValidMind `Dataset` and `Model` objects in preparation for assigning model predictions to each dataset:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the raw dataset\n", - "vm_raw_dataset = vm.init_dataset(\n", - " dataset=df,\n", - " input_id=\"raw_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the preprocessed dataset\n", - "vm_preprocess_dataset = vm.init_dataset(\n", - " dataset=preprocess_df,\n", - " input_id=\"preprocess_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the feature engineered dataset\n", - "vm_fe_dataset = vm.init_dataset(\n", - " dataset=fe_df,\n", - " input_id=\"fe_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the training dataset\n", - "vm_train_ds = vm.init_dataset(\n", - " dataset=train_df,\n", - " input_id=\"train_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")\n", - "\n", - "# Initialize the test dataset\n", - "vm_test_ds = vm.init_dataset(\n", - " dataset=test_df,\n", - " input_id=\"test_dataset\",\n", - " target_column=lending_club.target_column,\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Initialize the champion application scorecard model\n", - "vm_xgb_model = vm.init_model(\n", - " xgb_model,\n", - " input_id=\"xgb_model_developer_champion\",\n", - ")\n", - "\n", - "# Initialize the challenger logistic regression model\n", - "vm_log_model = vm.init_model(\n", - " log_reg,\n", - " input_id=\"log_model\",\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Assign predictions\n", - "\n", - "With our models registered, we'll move on to assigning both the predictive probabilities coming directly from each model's predictions, and the binary prediction after applying the cutoff threshold:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Champion — Application scorecard model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=train_xgb_binary_predictions,\n", - " prediction_probabilities=train_xgb_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_xgb_model,\n", - " prediction_values=test_xgb_binary_predictions,\n", - " prediction_probabilities=test_xgb_prob,\n", - ")\n", - "\n", - "# Challenger — Logistic regression model\n", - "vm_train_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=train_log_binary_predictions,\n", - " prediction_probabilities=train_log_prob,\n", - ")\n", - "\n", - "vm_test_ds.assign_predictions(\n", - " model=vm_log_model,\n", - " prediction_values=test_log_binary_predictions,\n", - " prediction_probabilities=test_log_prob,\n", - ")" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Compute the scores\n", - "train_xgb_scores = lending_club.compute_scores(train_xgb_prob)\n", - "test_xgb_scores = lending_club.compute_scores(test_xgb_prob)\n", - "train_log_scores = lending_club.compute_scores(train_log_prob)\n", - "test_log_scores = lending_club.compute_scores(test_log_prob)\n", - "\n", - "# Assign scores to the datasets\n", - "vm_train_ds.add_extra_column(\"xgb_scores\", train_xgb_scores)\n", - "vm_test_ds.add_extra_column(\"xgb_scores\", test_xgb_scores)\n", - "vm_train_ds.add_extra_column(\"log_scores\", train_log_scores)\n", - "vm_test_ds.add_extra_column(\"log_scores\", test_log_scores)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Enable use case context\n", - "\n", - "We'll also adjust the use case context to focus on comparison between our models for tests going forward:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import os\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT_ENABLED\"] = \"1\"\n", - "\n", - "context = \"\"\"\n", - "FORMAT FOR THE LLM DESCRIPTIONS: \n", - " **** is designed to .\n", - "\n", - " The test operates by \n", - "\n", - " The primary advantages of this test include \n", - "\n", - " Users should be aware that \n", - "\n", - " **Key Insights:**\n", - "\n", - " The test results reveal:\n", - "\n", - " - ****: \n", - " - ****: \n", - " ...\n", - "\n", - " Based on these results, \n", - "\n", - "ADDITIONAL INSTRUCTIONS:\n", - "\n", - " The champion model as the basis for comparison is called \"xgb_model_developer_champion\" and emphasis should be on the following:\n", - " - The metrics for the champion model compared against the challenger models\n", - " - Which model potentially outperforms the champion model based on the metrics, this should be highlighted and emphasized\n", - "\n", - "\n", - " For each metric in the test results, include in the test overview:\n", - " - The metric's purpose and what it measures\n", - " - Its mathematical formula\n", - " - The range of possible values\n", - " - What constitutes good/bad performance\n", - " - How to interpret different values\n", - "\n", - " Each insight should progressively cover:\n", - " 1. Overall scope and distribution\n", - " 2. Complete breakdown of all elements with specific values\n", - " 3. Natural groupings and patterns\n", - " 4. Comparative analysis between datasets/categories\n", - " 5. Stability and variations\n", - " 6. Notable relationships or dependencies\n", - "\n", - " Remember:\n", - " - Champion model (xgb_model_developer_champion) is the selection and challenger models are used to challenge the selection\n", - " - Keep all insights at the same level (no sub-bullets or nested structures)\n", - " - Make each insight complete and self-contained\n", - " - Include specific numerical values and ranges\n", - " - Cover all elements in the results comprehensively\n", - " - Maintain clear, concise language\n", - " - Use only \"- **Title**: Description\" format for insights\n", - " - Progress naturally from general to specific observations\n", - "\n", - "\"\"\".strip()\n", - "\n", - "os.environ[\"VALIDMIND_LLM_DESCRIPTIONS_CONTEXT\"] = context" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Adjust a ValidMind test\n", - "\n", - "Let's dig deeper into the [`MinimumF1Score` test](https://docs.validmind.ai/tests/model_validation/sklearn/MinimumF1Score.html) we ran previously in [113 Perform validation tests](113-perform_validation_tests.ipynb) to ensure that the models maintain a minimum acceptable balance between *precision* and *recall*. Precision refers to how many out of the positive predictions made by the model were actually correct, and recall refers to how many out of the actual positive cases did the model correctly identify.\n", - "\n", - "Use [`run_test()`](https://docs.validmind.ai/validmind/validmind/tests.html#run_test) with our testing dataset (`vm_test_ds`) to run the test in isolation again for our two remaining models without logging the result to have the output to compare with a subsequent iteration:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.run_test(\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:xgboost_champion_vs_challengers\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_xgb_model, vm_log_model]\n", - " },\n", - ")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "As `MinimumF1Score` allows us to customize parameters and thresholds for performance standards, let's adjust the threshold to see if it improves metrics:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " \"validmind.model_validation.sklearn.MinimumF1Score:AdjThreshold\",\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds],\n", - " \"model\": [vm_xgb_model, vm_log_model],\n", - " \"params\": {\"min_threshold\": 0.35}\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Note the output returned indicating that a test-driven block doesn't currently exist in your model's documentation for some test IDs. \n", - "

\n", - "That's expected, as when we run validations tests the results logged need to be manually added to your report as part of your compliance assessment process within the ValidMind Platform.
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run diagnostic tests\n", - "\n", - "Next we want to inspect the robustness and stability testing comparison between our champion and challenger model.\n", - "\n", - "Use [`list_tests()`](https://docs.validmind.ai/validmind/validmind/tests.html#list_tests) to identify all the model diagnosis tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "vm.tests.list_tests(tags=[\"model_diagnosis\"], task=\"classification\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "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
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's see if models suffer from any *overfit* potentials and also where there are potential sub-segments of issues. Overfitting occurs when a model learns the training data too well, capturing not only the true pattern but noise and random fluctuations resulting in excellent performance on the training dataset but poor generalization to new, unseen data.\n", - "\n", - "We'll select the following two tests:\n", - "\n", - "- **[`TrainingTestDegradation`](https://docs.validmind.ai/tests/model_validation/sklearn/TrainingTestDegradation.html):** Evaulates how much the models' performance degrade when moving from the training to the test datasets.\n", - "- **[`OverfitDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/OverfitDiagnosis.html):** Dives into detecting overfitting beyond performance degradation by examining learning curves, model complexity v.s. generalization, and cross-validation results." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store our desired 2 tests\n", - "overfit_testing = [\n", - " \"validmind.model_validation.sklearn.TrainingTestDegradation:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.OverfitDiagnosis:Champion_vs_LogRegression\"\n", - "]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Run and log our 2 tests for both models for the training and testing datasets\n", - "for test in overfit_testing:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", - " \"model\" : [vm_xgb_model,vm_log_model]\n", - " }\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's also conduct *robustness* and *stability* testing of the two models. Robustness refers to a model's ability to maintain consistent performance, and stability refers to a model's ability to produce consistent outputs over time across different data subsets.\n", - "\n", - "We'll select the following two tests:\n", - "\n", - "- **[`RobustnessDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/RobustnessDiagnosis.html):** Evaluates whether the models are resilient to small *perturbations* (small, controlled modifications) or *variations* (broader subgroup-level or distributional changes) in input data.\n", - "- **[`WeakspotsDiagnosis`](https://docs.validmind.ai/tests/model_validation/sklearn/WeakspotsDiagnosis.html):** Identifies specific conditions where the models consistently underperform." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store our desired 2 tests\n", - "stab_robust = [\n", - " \"validmind.model_validation.sklearn.RobustnessDiagnosis:Champion_vs_LogRegression\",\n", - " \"validmind.model_validation.sklearn.WeakspotsDiagnosis:Champion_vs_LogRegression\"\n", - " ]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Run and log our 2 tests for both models for the training and testing datasets\n", - "for test in stab_robust:\n", - " vm.tests.run_test(\n", - " test,\n", - " input_grid={\n", - " \"datasets\": [[vm_train_ds,vm_test_ds]],\n", - " \"model\" : [vm_xgb_model,vm_log_model]\n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Run feature importance tests\n", - "\n", - "We want to verify the relative influence of different input features on our models' predictions, as well as inspect the differences between our champion and challenger model to see if a certain model offers more understandable or logical importance scores for features.\n", - "\n", - "Use `list_tests()` to identify all the feature importance tests for classification:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Store the feature importance tests\n", - "FI = vm.tests.list_tests(tags=[\"feature_importance\"], task=\"classification\",pretty=False)\n", - "FI" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Run and log our feature importance tests for both models for the testing dataset\n", - "for test in FI:\n", - " vm.tests.run_test(\n", - " \"\".join((test,':Champion_vs_LogisticRegression')),\n", - " input_grid={\n", - " \"dataset\": [vm_test_ds], \"model\" : [vm_xgb_model,vm_log_model]\n", - " },\n", - " ).log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Implement a custom test\n", - "\n", - "Let's finish up testing by implementing a custom *inline test* that outputs a FICO score-type score. 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", - "\n", - "The [`@vm.test` wrapper](https://docs.validmind.ai/validmind/validmind.html#test) allows you to create a reusable test:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import numpy as np\n", - "import pandas as pd\n", - "import plotly.graph_objects as go\n", - "\n", - "@vm.test(\"my_custom_tests.ScoreToOdds\")\n", - "def score_to_odds_analysis(dataset, score_column='score', score_bands=[410, 440, 470]):\n", - " \"\"\"\n", - " Analyzes the relationship between score bands and odds (good:bad ratio).\n", - " Good odds = (1 - default_rate) / default_rate\n", - " \n", - " Higher scores should correspond to higher odds of being good.\n", - "\n", - " If there are multiple scores provided through score_column, this means that there are two different models and the scores reflect each model\n", - "\n", - " If there are more scores provided in the score_column then focus the assessment on the differences between the two scores and indicate through evidence which one is preferred.\n", - " \"\"\"\n", - " df = dataset.df\n", - " \n", - " # Create score bands\n", - " df['score_band'] = pd.cut(\n", - " df[score_column],\n", - " bins=[-np.inf] + score_bands + [np.inf],\n", - " labels=[f'<{score_bands[0]}'] + \n", - " [f'{score_bands[i]}-{score_bands[i+1]}' for i in range(len(score_bands)-1)] +\n", - " [f'>{score_bands[-1]}']\n", - " )\n", - " \n", - " # Calculate metrics per band\n", - " results = df.groupby('score_band').agg({\n", - " dataset.target_column: ['mean', 'count']\n", - " })\n", - " \n", - " results.columns = ['Default Rate', 'Total']\n", - " results['Good Count'] = results['Total'] - (results['Default Rate'] * results['Total'])\n", - " results['Bad Count'] = results['Default Rate'] * results['Total']\n", - " results['Odds'] = results['Good Count'] / results['Bad Count']\n", - " \n", - " # Create visualization\n", - " fig = go.Figure()\n", - " \n", - " # Add odds bars\n", - " fig.add_trace(go.Bar(\n", - " name='Odds (Good:Bad)',\n", - " x=results.index,\n", - " y=results['Odds'],\n", - " marker_color='blue'\n", - " ))\n", - " \n", - " fig.update_layout(\n", - " title='Score-to-Odds Analysis',\n", - " yaxis=dict(title='Odds Ratio (Good:Bad)'),\n", - " showlegend=False\n", - " )\n", - " \n", - " return fig" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "With the custom test available, run and log the test for our champion and challenger models with our testing dataset (`vm_test_ds`):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "result = vm.tests.run_test(\n", - " \"my_custom_tests.ScoreToOdds:Champion_vs_Challenger\",\n", - " inputs={\n", - " \"dataset\": vm_test_ds,\n", - " },\n", - " param_grid={\n", - " \"score_column\": [\"xgb_scores\",\"log_scores\"],\n", - " \"score_bands\": [[500, 540, 570]],\n", - " },\n", - ").log()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "
Want to learn more about custom tests?\n", - "

\n", - "Refer to our in-depth introduction to custom tests: Implement custom tests
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Verify test runs\n", - "\n", - "Our final task is to verify that all the tests provided by the model development team were run and reported accurately. Note the appended `result_ids` to delineate which dataset we ran the test with for the relevant tests.\n", - "\n", - "Here, we'll specify all the tests we'd like to independently rerun in a dictionary called `test_config`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "test_config = {\n", - " # Run with the raw dataset\n", - " 'validmind.data_validation.DatasetDescription:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'}\n", - " },\n", - " 'validmind.data_validation.DescriptiveStatistics:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'}\n", - " },\n", - " 'validmind.data_validation.MissingValues:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.ClassImbalance:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 10}\n", - " },\n", - " 'validmind.data_validation.Duplicates:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.HighCardinality:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {\n", - " 'num_threshold': 100,\n", - " 'percent_threshold': 0.1,\n", - " 'threshold_type': 'percent'\n", - " }\n", - " },\n", - " 'validmind.data_validation.Skewness:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.UniqueRows:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'min_percent_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.TooManyZeroValues:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'max_percent_threshold': 0.03}\n", - " },\n", - " 'validmind.data_validation.IQROutliersTable:raw_data': {\n", - " 'inputs': {'dataset': 'raw_dataset'},\n", - " 'params': {'threshold': 5}\n", - " },\n", - " # Run with the preprocessed dataset\n", - " 'validmind.data_validation.DescriptiveStatistics:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.TabularDescriptionTables:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.MissingValues:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'min_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.TabularNumericalHistograms:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.TabularCategoricalBarPlots:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'}\n", - " },\n", - " 'validmind.data_validation.TargetRateBarPlots:preprocessed_data': {\n", - " 'inputs': {'dataset': 'preprocess_dataset'},\n", - " 'params': {'default_column': 'loan_status'}\n", - " },\n", - " # Run with the training and test datasets\n", - " 'validmind.data_validation.DescriptiveStatistics:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.TabularDescriptionTables:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.ClassImbalance:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 10}\n", - " },\n", - " 'validmind.data_validation.UniqueRows:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'min_percent_threshold': 1}\n", - " },\n", - " 'validmind.data_validation.TabularNumericalHistograms:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.MutualInformation:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'min_threshold': 0.01}\n", - " },\n", - " 'validmind.data_validation.PearsonCorrelationMatrix:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']}\n", - " },\n", - " 'validmind.data_validation.HighPearsonCorrelation:development_data': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset']},\n", - " 'params': {'max_threshold': 0.3, 'top_n_correlations': 10}\n", - " },\n", - " 'validmind.model_validation.ModelMetadata': {\n", - " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", - " },\n", - " 'validmind.model_validation.sklearn.ModelParameters': {\n", - " 'input_grid': {'model': ['xgb_model', 'rf_model']}\n", - " },\n", - " 'validmind.model_validation.sklearn.ROCCurve': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']}\n", - " },\n", - " 'validmind.model_validation.sklearn.MinimumROCAUCScore': {\n", - " 'input_grid': {'dataset': ['train_dataset', 'test_dataset'], 'model': ['xgb_model']},\n", - " 'params': {'min_threshold': 0.5}\n", - " }\n", - "}" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Then batch run and log our tests in `test_config`:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "for t in test_config:\n", - " print(t)\n", - " try:\n", - " # Check if test has input_grid\n", - " if 'input_grid' in test_config[t]:\n", - " # For tests with input_grid, pass the input_grid configuration\n", - " if 'params' in test_config[t]:\n", - " vm.tests.run_test(t, input_grid=test_config[t]['input_grid'], params=test_config[t]['params']).log()\n", - " else:\n", - " vm.tests.run_test(t, input_grid=test_config[t]['input_grid']).log()\n", - " else:\n", - " # Original logic for regular inputs\n", - " if 'params' in test_config[t]:\n", - " vm.tests.run_test(t, inputs=test_config[t]['inputs'], params=test_config[t]['params']).log()\n", - " else:\n", - " vm.tests.run_test(t, inputs=test_config[t]['inputs']).log()\n", - " except Exception as e:\n", - " print(f\"Error running test {t}: {str(e)}\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Add test results to reporting\n", - "\n", - "With all the test results logged, let's head to the model we connected to at the beginning of this notebook and insert some test results into our validation report ([Need more help?](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#link-validator-evidence)):\n", - "\n", - "1. From the **Inventory** in the ValidMind Platform, go to the model you connected to earlier.\n", - "\n", - "2. In the left sidebar that appears for your model, click **Validation Report**.\n", - "\n", - "3. Locate the Data Preparation section and click on **2.2.2. Model Performance** to expand that section.\n", - "\n", - "4. Under the Model Performance Metrics section, locate Validator Evidence then click **Link Evidence to Report**:\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "5. Select the Minimum F1 Score test results we logged:\n", - "\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Adj Threshold**\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion**\n", - " - **ValidMind Model Validation Sklearn Minimum F 1 Score Xgboost Champion Vs Challengers**\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "6. Finally, click **Update Linked Evidence** to add the test results to the validation report.\n", - "\n", - " Confirm that the results for the minimum F1 score tests have been correctly inserted into section **2.2.2. Model Performance** of the report:\n", - "\n", - " \"Screenshot\n", - "

\n", - "\n", - "7. Continue to work on your validation report by:\n", - "\n", - " - **Adding risk assessment notes:** Click under **Risk Assessment Notes** in any validation report section to access the text editor and content editing toolbar, including an option to generate a draft with AI. (Learn more: [Work with content blocks](https://docs.validmind.ai/guide/model-documentation/work-with-content-blocks.html#content-editing-toolbar))\n", - " - **Adding findings:** Click on **Link Finding to Report** in any validation report section, then click **+ Create New Finding**. (Learn more: [Add and manage model findings](https://docs.validmind.ai/guide/model-validation/add-manage-model-findings.html))\n", - " - **Assessing compliance:** Under the Guideline for any validation report section, click **ASSESSMENT** and select the compliance status from the drop-down menu. (Learn more: [Assess compliance](https://docs.validmind.ai/guide/model-validation/assess-compliance.html#provide-compliance-assessments))\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## In summary\n", - "\n", - "In this final notebook, you learned how to:\n", - "\n", - "- [ ] Adjust an out-of-the box ValidMind test\n", - "- [ ] Implement a custom inline test\n", - "- [ ] Re-run tests provided by your model development team to verify that they were run and reported accurately\n", - "- [ ] Link your logged test results as evidence on your validation report\n", - "\n", - "With our ValidMind for model validation series of notebooks, you learned how to validate a model end-to-end with the ValidMind Library by running through some common scenarios in a typical model validation setting:\n", - "\n", - "- Verifying the data quality steps performed by the model development team\n", - "- Independently replicating the champion model's results and conducting additional tests to assess performance, stability, and robustness\n", - "- Setting up test inputs and challenger models for comparative analysis\n", - "- Running validation tests, analyzing results, and logging findings to ValidMind" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Next steps" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Work with your validation report\n", - "\n", - "Now that you've logged all your test results and verified the work done by the model development team, head to the ValidMind Platform to wrap up your validation report. Include risk assessment notes, add findings, and assess compliance, then submit your report for review when it's ready. **Learn more:** [Preparing validation reports](https://docs.validmind.ai/guide/model-validation/preparing-validation-reports.html)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Learn more\n", - "\n", - "Now that you're familiar with the basics, you can explore the following notebooks to get a deeper understanding on how the ValidMind Library assists you in streamlining model validation:" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 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)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 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)\n", - "\n", - "Or, visit our [documentation](https://docs.validmind.ai/) to learn more about ValidMind." - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "ValidMind Library", - "language": "python", - "name": "validmind" - }, - "language_info": { - "name": "python", - "version": "3.10.13" - } - }, - "nbformat": 4, - "nbformat_minor": 2 -} diff --git a/notebooks/tutorials/model_validation/compliance-summary.png b/notebooks/tutorials/model_validation/compliance-summary.png deleted file mode 100644 index a8abd2dd39ea4ef683e4300fb90d80f309b64e48..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 61626 zcmeFZbyQT}`ae!eDJdc#At{X@;n32JbRz;%!caq}D2Oyjr{vJxsVFrH($Yx7Fo3{- z^zRIK-}k-lXMOMc|KGZ^)?sGOIs5FrpLjj5=h=G*(@>MYg-?Z#hK6=aQ9(u%4Gjm1 zhK5dsdjmLAij8-UhIUigR$5v^QCgZ_!v*}p_N5gXnnGBD9*(|N-HFRl1jK-fd>q4l@u^ulUh8{sLBn;FE^2?%qgLNM?+j?u6O6lw&5KDb`JRSkL9g+ z=x&pG?LzZH9v7#BDFFkIJyz#L(7sO}YuXYP<4Kk?sox>n8l*~k|CIrPahn$n`+FdP z8GVvVPR=KEPE5o+tR)TYY5l~#_lB2Smld+g58?>WBy;H1$ojFHi0QDof|OmM)ApURu7pYsGno7$y_<-A2QF3J2X=8j)DGRoR23 z60a2J`^CoQ>R4yum*$-M4GN#a=)RNm^}Hea%y1hUtN3|OAA_QA>AN0ep4Y|0;;P1s zN36y3v_}*xSR7vbZ7i$Dn0O%AFBK|E|P;cJpy1`{|{aM5n*C za17p#^=(Z4>UGS-&yvs49=^USQ~TwX6Wt5}QK+7TeulH)_Cw+x51&t3PpE&A{JilY z!^t|XQA6-EtSXyk69>7*is(68(A5|rKk8}w0MhPGX+(GGG+7Y6H$yS7g>T^4G2dTW z8DewE^@=W2ql4y4*elJDr)LiYrRD8@*ftUtKX5&_@YilHxV4%l@`K;af17BM;*;P@ zD04loNMJ^n+H;CZmQvcv$|Zfw@mDlDiyz64^1xqieZ4cXrY<;2+5Dni!oZ4UMNL30 zL(NRv&nw5)tN&YX24@C0nYHC=mGEx({pY`!FnUNhty!)0(B99}jlmJBZu1`4=Jmd7 ztCxgAP>mpTE*)IDE_hNzOM$lIdQUR9uWU6J{(u zz9kDE9h;Oc!%FX+JgUjidvaXH?bQ&~b8h^a+DazehxKQ(p+4HgKn~9%=zwCcYw^CoqWs}^* zJH%4Qh~|xokxEv2!0?{5o{W%WptEgM%2S?+G)2ZfM`R{k@iXV*%|+718;i7yc+aVS zJ**0qVc3Y@Qk=_EAkBI&&(2WBh+v>$HDQLth{ufevRGQX5I;`gyiXCQ+$Yl4)2r7v z*qg{)3ATBZO8zZ6Q3aOEm9I0R zHbnbf+=lnjJpoUuCy|e-$f+PydIF}yY;MY+Q1Mt(NuGSVY_kFnTg^vhkD5|_zfGo+ zr-llEZ4Yf8*2>w++O61e4NVn6q~J}0iqlFW8b;7INuQEKHbl?O2=jpjKCFo22!{y% z2tt~is5(M*R^>t&>jt$FwTANg@^Hsj+mB}PXLFkoDiO9{mLIw%!;()^gi^9XC~qy@ zB)erKTI&8|%zOtK0zcjeJ9vjYT%Y}Znm0I1G&(qL_Q9+oU>;4VlK-t>vR&oSLWIgyqwEK28)^{z0yPQhQ!_A{(tDaWIj@o1lvE_1G6jWt7C;qZqvgpoW zxk*MxR@G-0zr+$y>1J?Gm#`!Pk~-#KXG8V8n0NO1l-uAxMSG>4Xthwo_Y*6fmO^~^mcY3n>{D5k-cU}~wmKOa8f%W?V4g#*Ef zl5Urd?2?u*uxPaSV9~Y&i@n1bFJFGw%W+wVRj=?~p@W)-ylpf-jgD}!^QgVGh;@+b zn|^ZqY$vdh&&EA<>*kW%=Qi5C+Uj@yoW4TcSB9t20)N677lAE2!V zhopyIM6W_SdhSKDL`P>jW_o3wie2ux?Dii{Z6JLkkavfY9V}>l(!BQD8ax|LD}6Qn z1W#*EQ&0B%IQ;02Q%>>D>rcg&ez;T%w+Qo`XPjAGgr3d2iZ#_VvSL&QE(9(Hj!Pms ze|PS6;z+%aqLdPln*CV*k^hzYYvL!Lp6I@wS)}}h)oN$z0rgog{XSy8X(FuYW!G#`$fF)b-2CqMpHhgS6^7Yxx&D zz1!auzfQ(052$_<;~Yw~R8~_L%oKHSNgFjaWOB$_d~EgYy`w?$%a(hw>M=sg8=Z9L zciu!UvVQ6E%{qH`Mx;enK$~13pvqTf)jWIfWh9Fni~NiaGwXixkm!ipYHJFSwet&y zl*q4YX|tj|z5}lw)3^q&>3+){fMp)pf6)5b|EXc0Y)y!p*O1ra<5JeEfMT}c2XW-H zhU;}sBMiQ=o=@S6L*Wx#7rLCB?fQ$_nudE;VbA)$WXYc!AG+=gALN(8W)dBY)r`3f z{9WJdyem#HT z&;8W#e9*fK89vc5Pe0=X$(`?>Qk6?n67&Q7%hyMFH?di%q<^^VbJ?*`mZS0y_ic(p zp0Vj$^+_);o29FB+t>>>-RgWGNVTpEnB8|OS=F`j_a=JoI9rZT2z0hOw5iaN=SRtp z)1LNGKcS;g1X#RNT*K%4-olA=g!C7P!t6ow_Q}_i!;rj?`mPV5Y1D00L^RH#)wzAO zMT^kwA*&u}&z*a4F;l13wbutS6MYf!(P^v#^I`(qqsWn=LZiXg8IJ=5*SCsX(_MQ8 zeX_vK3tm5WWe>lcC_5%?8X6lBo^PJOKj=1$H^a7*sLMT#ZC_H4xNjBjjKxo2lxMUJ zwD@nxeU7Y8`y~XPS85vbH}X6^&PW%j@P}O@_Gg=paF5>Y*w69!J^P)#=P^BhRNGQV z=XZCva^>k(Zc8+Fw3%qE#PM*|0vp11gFoJSY%n8B2yz2rdTenz~VQ$A|kj?$~vYpY0V9SW#fDuYe0W%9DE26_N9w+GpY45Qv05tFv}9^qUW&1vk)I=g`r(@X&6E3TgI~421e&Kh3Z` zCH}3h4++`ga6}vZ6mYW6{oBKpAh2y3t@72$!ELl08??_}Xl@7VA3;uua1jr70+P&I zg7SW|+3~->@T)$(+g`nX8-$;SOGaKVy zcjG{_G5_@$=m&U7X-O+80`FQDE>>2Kt~Ow|65TKl-~^tNf`Ka<+8t)pfv%{@vIHn#*nx8nA6a6&x?P1F+vd~~pKGpF}-cSFllg%E=O0!>xca7IWktR7B03cqbLHPX{;j3{zgs?hEXe!!rhmKipG|dKtz4wR4nUu7;{VpypT>W` z_@|*L4{GXv6UDWhug(I37RMLm`Accy_;hr1F91E>vz1ZT2HpWHyLv?ee;)t`>K!=b z-IZ@i(V?M9qAAKqX?vn?&EoiuBF}q%sUL>Crs=FC!SmU<{}y9WrtA=$QKtK)Ubm{Q zytuBbS>A!>3&^hI?RZ%44a#@8{ifzSXpz1qKE#eIVL@{UfP>g$tGb z^i}pBI!BEU7wUEEpPmQw_F5(|5XAJ=80(*h2@G=LzmMi-7$PqrcXKS};qL;s&m z{v*Ep&%6H{SpE-!{s%$-4;T9Xq2pY(C|FtLYimO*`tgxEY$u0`X2m7#s#5Vi@4P7%$pBg3 zqnbHE`X?rm0TU~~kBX_!=>97c1jL=rQJPX6anru%w3ZgqIZ_WA^gZ!b*{vYM!pcTEBH@=JCDj@r(u7pXU zP+*Ciiy%SKlt?w z^xf-j=sz*Wk0iph+DRMVr=*_Ilswt|VMDH6BR96j`Y!N9gq=ny>3uQzhz>Q{?=8Kt zr#veYIQejm)QQXh_XD|q&^-(G!FE3=k6X*^}yXd1v;#aY}|S+!|m^ySA5RI3pLF zlhN8d#Y;F@aEiOw|6~!H1%aju#S|@@75IL-mC>?0&85Mc4>6Tl z&4S)SZNgMk-Y16?k&e`+$dv3jCZsPix*+~WXo%H|(y>TaCkrk;gimUz1GW?(@<&o;sj`+pd(~P`2H_5VKJF|w% zagj&xnAA6wmp)}FzZzm-Hi6*op0F^0>dmK!rPJTtkQV z#~tUJ*I98VMbJSAN|=vq|^I#dSo4OR}3kg>q_UDVFJJO&rw}!2vxG zaq?*XtmTpIxO>1#fx#uOS00mFqT5?FW~S#bM6)g$A${fKopk5{(srM01Kc*kQD&lH zhJALc2(J(h=Xlc-i8`}T64-HH;Ad?`^}$sWCuw?akm-QJ4r9uOA9IaWM_}RTv_G6C z!;kG3ds2jRjPi*~q|>&RHKrYn3OsPiXh^-?RG{pr_^4qSJk2l(OH?>af{jiN9bXU@ zN?gN2Ic~c11lVtC+XxKL*k94qEAL~v4vC|Xpd05mj%#Nd=EhG;+6Xp5IMGDc2+)J! zH!T1A`pQPrk#IXL*>zS<{5DV=)@%)z8iv>2x8vB2@>hRqV8HgpPN1%WGM}&*Zc*e4 zfYc%ZGI4G*&_!JqXZS>s$WmmdOPK=p^UHq8K;og>t*Jk*_9jp~zPU?PV&A){72Ll1 z)wV*(?OGoT9T+Dno6Fm6tuCdHcd@P2d~|X!uP4L+T=@FpyiN1WVq+=&F+wMd@M=CE zh;4EQinr(8^A7s%ZXc~2)X#fAEF0hwOMpTjC~A0F(G%E;eWTuXfB@jKrb0nZq5LL^ zRU}4dBw_Q7+p_`A$HD>i?=9&FkJb`SuL6L^AeLIkIk%!6_)^AfvMM*OWe?_E@Tu!M zes`uwD((1_&hn0d6xDu_7}SBQ>N0Np;RwP8+1BA702n|l_wUS3Wwt3^G%tME8QW#N zo`e8TOzIE)8x7PP(x3<}Yxz@unPcPIZ z63?^z7_7XS<_ZdFu3eFe#KwJ9*jDTqXWX8`aqfX^i@W}`?X>{Wg0!#~jjR=(^b8aV z`s&pndkrTuYR0D2{zwrwo%4IKl#gxJI1)9&lif5S%U_8p^Cds!ucz=yo}P_cHpJr| z4@cTCMMRWN4T~uG9o5}4d}Ma$ zw};MA;D?=iMF%7!@t9uJ#&UNw4|<=clf&{wg?j!r4tC;%VM(e38YjCh4*`xN?)R;6 z7US+ze-LTGGHCfn$gs$4yXt1m{&xJVe2p6IXRbeERFSNJ(#{@>)}0?c$l5U&H&eaNIq=>ch^Mo{!tq zTkIko>3`2eSe+dPD@8o|Q-ta)!XSA%6bJGhExI*7#yR=HeX4l*sD%9OA4M1A$;ivV z79w1W$1ZkO!4i#9?_lpP^ggOzyV&chK=EAfElLW-(1OERsp5IRHtO0r@+mDmOCcNe z{WlgyWrR>ayh00$=S*sT$C~V;;m?mE8kQGcJ#Kd^B)Ya0bLQ`BO-W@(YrU<5r4IiZ zS)d8M!>GRf(kBF+{rK8t$?h3A@Qfnhjc1*1ta?!61#fkjxsAEKWyjRL+xzl?6 z`o>yqrO~89%?asELu!pvQnl0lWg3&FC+?~N@rYKj^ASmERM(RF1jHJl$dRs$D|CY9 z1BABjAZ*F;3XETup`k;5hzra-yix`ZhWGSN*lc;s*Ti$*)Gvtik;3qhr-g=;8mBI+ zwGudCJ7-657pJ)`UT?!jj%)G!=8=>{bWk#0>w1?3_@?^S{aUwYWc1mGgW+>mch)IC zb2_c!E9yxQpS5ipYna=c)+#Eo)|_Y#oJwMaO*l0_>v5m_)y<(<$p{;9RlgcO8b;ld zz`$l*QPmkm)H8z&feuVynY`QFSF`(pn%%k9XBrt4TsuAZ4n!;C?{KQ2`8IMPS9S0N ze&L}Zxki$nwu02tSox06jlyB?-tGO67os(7*01j<9@s1`J(xzqT4vAQfsO9lElBY(l(eNX8bD#EYPaQD>n{+K$F|;*F*~hTVA| z`uXyYcNNp#)~cszmON&5X(8U%O0YxexXf>}ap`Ij=A6JJy449iu2P3=U3_AViT=V! zk?UY4nLS}Ey*%75m5jr#W8(&ch9 z`;5>(USwB;l9m5q$HsJ_sNXr#v)UXExXOGb8p^^V%7ND}Wv-wV=d@0}5o|KthIWbfEn^aw( zZ0-su6y!0rdYhVz(RmZ5#nvazX>_rdcYAworHI<$YAU#+-hfpYUf$kAyE5`m8bIOt zX4dU((kl}zyF;H{KBi9RaaC{c{2&>Aqp&af#^n=DGAf_+qH8}bTi=J~GB7%PGdH6f zM~48mxE?d5oiip5_qGSC%T}b%oqe@zNmsiqT*j$zl)_lJK4>H*&>Wd20(P9D^?#`TCR>LZVR z;66V=n6imf+4QYTZunH4ETl^lbWcxtFqaS(I5myF(9UTz@zZVlqz-am1?zfpMWGh7 z0qJR}La#I3&LSX%)spJI=7LVuXz0j3+L^06p&b}BX)pTgqzk>VcHQq)&p8k`GOf}k@S0a-RGA+Cnt9&}Q;GC#gx8xQ(T1t+Me3@n(Mq7JTN3ETR|b$hfCrd*Q{j7@*H`pV z!VYJ?Aodya<&*{;2Ch^2JqYVRu$jzeU z1tUv}nk(_41n4t}{Cczdh2m}`4R9W@7ZEMuS__l);=*ZBo8hMg(awSzTs08yiywdPu! zH0U&H^P-y#v~)3rW$nm^eITu}Tg)#pHd{YEfma_`LK+(i%i4hbp^(SIV0VYzRq?Bl z(i1-dhMGOZ_wdRs#ZYe9qb^Q#MYQN204kF*x6-_}1_s{7bN!UjB_(+7kdZH3WlY;G zv6z2sO5+nlHa8nr!gHh*yE8jtBGhNPe<*>&5>7BUm0ENKCX(56xPXGw3R7|3p%hH~ z1n8Ccxhvb%lf6-{$&>klmg&mQFoWm_lNnODP7SOduyMpM#*te17^_5lF0OFm5unw8 zGv@9iH*Jv}o6JDvDsFM9L@%q+x`Z&KV68Hw0mxGe4NsuN8pLDR$8MVSCp0g!ZsqO@R~^nq2i~c zrLQyUDOskc1}7gah|4D2kJ1hGp6(wIGK&u;SpD(WL`_kaTvC{Zb7ghJ45*1pxl&$F zlo6OHIMWW;_jiRsK~jkBUN5l#%Zv}Dt;YCcPfWFjIG{+SO($*vDh;^6h!-}t9NksM zqw6A68n3D@{@lLcdWkZ~n-zH2PNoYj@0IFsQEQEXUzx9rW-|epYMXbnSLKzh5Mq^6 zdl$OdOpSqp384@T2ll?OFgq^D??kFwP_YcL3~2pfl%|pIbbq%zU1BR&n!~rzYpl`O z0UisV)f&LO79k2CaATUTFjTS)Q@qxueamwx!7jHEaHWr=-`DQmE)M7;zjH^?%b9b-h zs7n$|CMAUyv`Yd+sI0YDKCc3W(9h>PWt$$y1uR2vudFHhx>Pge7byBFy_1o7y}xGE zCdM~K0`)HY;A(rp>s`=}_s|+JYbkt}t0+<%Ga+2jT&osa|LF z% z>&rBY#L;xyWQu8Cr+Tc@)7-fEJ__?h@6GtwsS{wxPXG|0F$9-~1WI7YZU&EXOq3;v^a52?Dl|vQ5RIm(#g&mt+4=g6nw~gvwY?4R<;r(zZx7zI(zi&W% zaY6-Dx;u=F8wt>;N!24d}Hr~&ke|_q-eec+5?j9Ql*S>v!l*4rY7=;g{p> zGSPzJO}#TVUz#8Q7rNUpcZ9Ty93Hmp?evdJo%%j;A64T>0{qD2l_eFH@$DbKkJqYq zkrR2n)zo1;NlGt&XvmpD>S}kUCz7@gfQX+2z?R7kD-hN_d`rX>GfxMSXR-}!*2ynh zq{9?HaF3BpJ~~zb8N`bZYzWW=%(!;W_!c@C0~BK_3Xa_RQ9o~_7mYxOc5u1c!_SK|T1V;X29M=yiH%xtyys|B{a`|+M$#Sc+2)dBH}4jK70 zkA4X9xk`wn80oJ29f+cB(d`l*Ek#9@M>-QWCcT)O!__Wq><@YP)pbh>9kL9?THUH% zUiK&py!-Nn1R9QEKLLJU(Qb+ukyh3)l9DMQioLx|0p(v-7aPnzVPfl3blLvxoHLkM z2xPdsfQ}N&G!QXwE(>rzKW3VqzN*h^Jf#OA@t5ru@+Qh+q^lg6noankQEjBa=s}) z0#s5s(X;{Zs0%!mZPzxSsO5Ms&!hiajc+o)xS-nePg~NA_lclL+I4T9gH72ik+Q+8 z`h>`6aZWEI!D&<=4no5^U>&^N|Ea>p|FS8-w^;`HG4?CJTGAn6>g|Q}rRQVM^~!Z8 zpSAMYo|1@pA2&GrNjOlk{|Im`!uE%HlAURR7ZwX;r#tb%1_tBh4>zM56993B` z=R3!fox2dyz8}l-W3hh`z+#oC-nuzZ*6bEb;O{pP=6nW>{39x}D(9(TTnO*%5cf0_ zH)-s&m_Qfv2{XMlTMvPgh10nywC#qk@VS2UjsEGbsTp^6k>v6ALJffC1t2}gnNQ}R zUr>G4Q;)efDbFG5Za*ce<9%SuLy1Vu_&sa*^r-3=yfN7k;?|UA?r?tp#XpBa0q|tx$IHEvI%%( zP$vUq+ig55W1KQo(2+A&?q;m8Ivwp~t4_Oq)$mDQ6SVh1$|Li1rjoJI+d!}a7Ste4 zeknBQW?182GdT7O&S$3ts?4QRe(#$p)Epwp{0<%FIO)WX8*i^H8=gkAY_M!Ra*DYs z5;~~|22P17zj#7V95j1+1Upw3Wg$<@P&|k-(KGd|x~aDY?2A`(EEj!@updf2>{wx+Z1kJ6gF{Lk+w$oD7ZTT&mOl1e<2Eya-9xA_(=KA`2hbC0sSa#aJG8;j=1eqLD_^dJmw^0Ud(cqrB!OA zz-5Io;0`jWeWGa{FrEP*RAd}Kq(U#m7(81dH)vWlfSRFGZA&K2PEz-)j&F$_9-3562D<%NCnDh%|JLQ7}N~$d7_-M-V47L=U~x`;qYJuZ& zs!_XbbH_e|SK-$ETTtZB8BlMrsi?z_pW6!lw z;*xy52^IU6qMzmR7fh!_DSF-l_QNv%0X-Q7=A!@@hXb8*Kj$OI@W;RRZ6My`qUk$` zy~f&zQh4ho&`b)9Y%`BsxvN;WF<~V^#iiKytKb9BPBIWf;vTOopV1+IRK3GK%?J0N z%4!O{QAJ_>G1#Lko2A(p%6C_7=l4r(1LC(Vxl5lnf9y>%N(Lx~1{_-~%ao1K>Whah9}LXB3{l zo2;#1hn!2DGRW1bC69Udk($ZQz=`(U<;z58H50ERt_G!+5w)875tQEnAu)rR4-??v zrTWcJb<2J{rxCZ^BetDt{ZKbMTIu0Y?|WA6_fxauexY*o970vPNE!$=*xng}GTaEC zVCP!X)fnr^Rgs6{ra7Yw@0=doCx9xCra``h2yNwz8u*4eJL`t)G|V!sBtDx4p(^RS z4|b88;FamRRiPGSC&6GEz2c@SH~{u(Fsc8LrX`{)CpTxMx^L~Y6`ATx$_`NJTmV|+ zX{ktXi#ydIf~x@+2dr>`?m4I6ba6{)-R$@-Vr!DmETheqBWcE?Hqlu&)R^MqeVjp{jqeVI1(P7Mgtf>{>)7*wb*gG37{v5&Y^$Cko!lm{a zv4H|EVkU-?7*~5Cctd%mvz@W~C)+L3l{gN0s%cT3%~{ZVx}Y9V<|-I&R@I{PDEG=Q zERui#IB0@e)!P<1su+*L{_>^U!e#OKb0%HCH_~_pk~Fw-15;qDeg5b-!DL71K_F&c z>kEd-8jq}B4+R{C^#^FFf6y(7xFZs=!xQhWjU}w9iW*Vn11gl8*ZP`)atRsxzqxic0D-1x z zlF%9Mj5LUl8u5CL`I4nn@l6BC4C+poJ(H13W}Fv;H;1LI4Y4*WI^Fy(YNy_fYI`58 zv!s}`B&W8HCzK%CE{}n=h034OHzi z<+j5e@7Te(P@)pBi$2-lRpad2@X*3?Il7#${wa`dZv$nRmZe@!+}|9P&yM-LtWKu8 z&8r?(?QLNaDOy^-u9`JkIZzi*j9wjh=&lSNH>Sm3&?j87)Mf|_dtX{+l}3%0yJ`>(~1BBDVk==;*ABTI+r7m_&`^ z4RU|X960M5N#U$d85aoBcncw#XMM9tRs*mKC-rQ72gGh7h+|4Oy!yTedbuCec6+jYBt#R2pHY*QbPkgq0tP0sow-I2>gS^*9~JC~@`BZxpp zkqngo_#)@iPBQQ0$#*xpBI)dUfR#-gTF0E48DZBu%(1#!m)Fjn!^%7e{LRO9o{@z;!x8Nk;h z`ji{!64;sPht;K~4Fkf(esUpatE+6?f@bU;#%+cddnYC1e&S+1*!|Y;u=iyRUY1+Y{OcvKJXBlnWA}iaGgin{@N%}e=N8SZs@hKfKx*bq zTWNsnwRB5x^*P^)kam#g>FMcKQgN$wR(Zf7O{xJ$G19Sc@wY~Fy$IC3!W57V0 z?(@?Fn+Tww^j%XSAD?)Ud;Z zTcx^WQqBSxfo8g!VtOir4EncFFaC8TYr9s5v_p48Kiv)(a)=*#E)i&7T@+Q z0yw9hv3Kmr8ee6Z@j)I;uk?LnTK(NMb!F9ru(b$*EAabwfX7{?73aKcL{N->SX0el zFSK&6srRw}zNQX$g*vR9bcpgehlTyAT>(fVEg8Z{oZXobaMpIOe?TxQ({uH1Q}4%Z zCNF@8*5GUl4U}PZm4es2jQ;2uvp**tI5TZeC8O6e3NFtV!f&gm@OHk-#u@dt4{iOVg?@0Z`fFEi zYO_cpzPfb=jqXktK$=)8rQYQ3iG+({qX zPJch5$N+{G05rS&b~h-H8pBoGfX%?@kg~AfW?H3N?Gc#l<-KRnGn;O{$%dY@pVb2< zMzG4E*;~0#pk~5~q3F!Em-g#W_EwYKgaj&8TOU7g^pzrl!eAVE*^TS(dn)v7SPT+l zl4wQP=g&^+wYoKeP+U$BJjJP>9M1fI7d(zQrSe zEzVr*?kvCc4(Qb$kg!ir7ZTic~3(JH|wLO8e2y7nk49qN+|QXQz7?NK7IfP{Ef!$Sh<3^`p1y z{AzI(QP0YG?|n?aOTqK>Y0yrdeQR=70w5A)SbrXLdDIo>U~=T-rkz&bY^9o@*zwz< zFc4~)ztWIVA; zc*=}RxZAo!CrT!WSI2xu`AMa-j=mdKVVW)3z^|<6g#n}b=xn8n)rvP$Ht=??WZ<1( z0zVE>ES;&-2b6&XPlm4O1)K%f?Kn)F>+^|t{2wSyU>|Phns^#F36vT9brr zc@bFci0`N2Zg3g)CO_@M$Vhbt;)5)pF9PCkxLAtW#(wns$MEtxpo%T8r+DO++CF zves1Ar0u%?Q@z%TCG~Umer&~RuydLN{dk8GBnGVMLsIE!5k)B zRu^#>)H{6}*o_FJfLNpGpB~;sr@*?}nyIx>q3e08k#|?eG~6`6MAr{A@~G~!kU}gx z(|`Ob5H$kJUWh2h8v2EbP79tkdV9gt=;+Ov_TU+>uE+~Dubj*NI{W1my@HHJlk-p< zaEpndAlS4mbsCU<900${F)G1uYMVRJ8HJyrXYK z_rB+E-v7zNL3|5WnE39PS6l=fG$|Z>6|@0k?f?D6(wZ%dzA+ zof(y(LDjc=zrm7@#DD()!FKOs=Bswwj=O_zjm)2F;1du;&VoEP=`l|Z1^)vx@LTu7 zvy@HU>zzG?>b$-43}tC-Qjlk6mP>KW-9IA@>*8Q^7j6HUJ_J$BpA~W%949#Id)KV) zedhrS)gNm%{bA(eb%_vOSjH zUqMVz=!a%{sex^$KfJk$$GFDXN4C-{aqSWh)v_Co#veN%7dP&%Zu$j7MAp$EL7>2v z88K%$YU7l@3SG#_rvY1c{zR}m1HD~!UmlPdq z23XZ|?};@a-ql-(Ki*$1i~iaIXD-_P%|_0EVmnc`mrP4*Ws^1Y zrcKGCHizFZEyvinR~Sn((%da7-s;IV|0#!O-wGo4euVSrhl(A*M3~wP`0ZN?Lwd}- zpSI67e*BSdl*GLkZeM;KvHq6CX_)GAU;duvgfHgYXXMXUZs@H4Ic{wP=Py@f2C*oW zpZzjGKUaO38Vy7MVFUU$^AHuv^M$^I`PpI(bV&RC>?rN&6IN!XyFIUq`2fSn#D}Vu zu*HvIPOMQS_cwUV-)Ej&-TgY61+?E)e>R!KVKuVK)hz|^kUk2vI2#z{uWqS%9slF8 ziE96)7ZyplVKFf7rq1TN4rgK%Cr<+W*UZ>Wu2U!I8!WN!>)MvJR93)RMIL0L+~3R< z{*HX_mzKI=k81NdVc+`P?A89rzcJUC_$<~A&xHmc{6tZ5P(+DLPdVE6ej7}hruCLQ zNP`sc`H$M9i2A>}4swbBSTidWNYSGOdCciQkdI!bl&@D%kH@3Hu6MHSP77b|{b{_^ zs+v~VpVt?Bjnu9|qo2F^ATt)-VQ%6ju2Y+`Yc-(4vOdy-K+r_>RF{@y+vXvgrfG~> zJhw4*5QwW1Vw4DQPGz{oJs=aSwO^VqTsi1ipBq|JrxuUQ<$0etrNNTvnuttA)ks|d z?z0y1H4t#$rclCu3IAhd_^t}_#SzrLysCbh%5A-@3#=8U(L!Q!Ito<$;$GSn0PNTp zSX_6iY|kn*CJ|uVA#z* zW_~y%%%Ytja7V+(dw~PoCdCd@lD6uE5JNpMnDg`Z4kM1rvhIS%_)O%Y?`|f++Rt&8 zYnlj3ExeKIEVIq-*O~8MbXyH;?g9a}y42k#WOi2Ub3u%#E9XrWeV))XH0I=)cd2=w z@cwS7r5l_(e!H>jI%NE{B3#iqO&7pKH&{c*2-zu$8}7%ES^#^Cf~z6R2^>0Rm?vNC18-{p=_zm~Y?j^i%$y9PF%_$Bb*5$t=KjPDe}K zZTm}bzCpNl$7&w1cknkiF%2anQTkC1B+%i0Rbl!#C?))sIi*Gbqy_}Kf~v`a zQFUiuU_;Jf%DX7sM(#?VdGV&BU-D)9nU6i+p6$LOC_{`3(V(skQtH7GfA}T_z6AX?;>qGB(om z7<(+O*s<`D!h7ZNd$xbD>H9R~a{J77ybl*zGUTbW+R?l*W!KgO2-xvY7J4!Ra_|nn z!}ns8*Z%Z~`fzKK)%6bNiT{|QgpmD|Om`$DM3Y7cp#Oe=MvhcdNYDeY-nqE0>;P4{ zZH>6(RQ9CHq+hCuz<!#s?02cYUN(5lwo1v z`~q6Q_H=I4+W*@=ZCosRr_1>0z5y|xNJd5m2CL0TA*HqYIX!P3ZS8)#(ajC6>6!@F zjeZe-mt41v?;o$A0A zg03U52NrGn0y$g;Yjn&;^u}muc6q@URje5bs76-e>F$yNsxkvQrY!V`6L0hOJo{|y zen^rc77%;8#6iJf_Pz zeqj;aA5?ZIvFD?n`~Zj;Uq6aj0Xv3%1G_^JCAZIkB)xM_#|&mR?>P(dvg-OKB|E5qH1w}fiVQVZR$89fcyPkHU83! zx5u|wYh(9K1%*!lXBKY5*O2pQ=%tAOvf`MAeFs@sI0=9iYL2!bbB%qlK)`}<)X z^KQd~Wkz-HpLT_&Yxmsxd$j>DjJ>_(x+1MIjx40H?ZyV*@*XVTAv!Z>4qM`3+eNPb zbbp2AFKV#Dz{gZ7VkR^p(*QXK7KaonRl)=5Fal2C`_<;~JMg!j zb-BIEcx09~^ZumNWImc5r6w)Qh-mTm5Th+?i_L@|M=B|cM_%KsCU2<$@c^4~>jt&m zWaw=$I68iCawLJP2P-bL5#u!(ZacHv?*ku8R4A0C&N0B-tLToPthBsaHwEO4P6cd$ zO&fYx!ogFYYzUz7W0UouF`)#V2KCszSH9Qmyt*@IFi8PfUM#FWC zjqRdemd?$^g%svy=Tl-)z02Cyr$~D`gxcfsuTrfr`FGT&FyJQ$`HEz=GsSgG#R+D> z@_heK6A@+}k?Cn(N{9&Z+rEV!JR@|c0^EkfAgQ6MKG`hP8M>LECVJqOlEs};fl}t z=ES_MoI&`r$zayYRE1SgInoV8GZ4gU{YaKBMuDCQ?UOJvJTUzzypIfg`hkv;oupy^ZLLOUL z0dJe;xP&;jqeBf3UC(gYCfn$|;^?>gdw*~x zC|f$D{34)tU1%d~y3zxd%GL%5L)!O$;rA61e0V%6ws)Hw9sd0E_R4b@xXpHZrwdtO z38ZT1Mn;M*0QFEd5|!yLT?0y3v5p`2hZ})}{%y>CYhM;L2&@8K@ey8(6GoWy{NkzE zqV+UNw^ey$cWg(4^GP)(X8vY`U6o3>3l0)jNuu}tgnq%?o{T3XlXw=7=QfaR>`wqE z_vE7leSWf;&;HFcyoOxdfrC0gVKo`vhl^m{2Nzz3h02?UslIno2BH-qfuNZ*-p$D5 zAQt7)zrQIce4%eXno}p~wt~25j*z!^$EbdtXiXHMljGZ!qJyPY>Ww;bF<5f+BuKT; zz&c!*R1<2))Rbkf4_8-9YQ<@r2G8wgW){sA_K5i_6BK-r! zV51+dOdm!SKm-Tl>K;ztr`4@HE+wWEiA=@bLmA@^2`6+(w#v(b(?EMp+hE#?=8qmwCy6wX-R=SOKo$i1uF93tN=h!4UA z#!G#)rH`&Z{E;kZm7=gGqtU4;a?`42bFk9#!|a9%l5HUhfS@m7#!S)AMGVe-p;@dR zbZWTiMxvBnPf8juJ6xY&@Yugs?Z=b9l?U4)QrjPE{tUJHX7@W`eMhESEx^^5jKvD? z^o+^!h--pnEhD2G1*<|>NI#$U?_*uIz0_Lfr^@Ur2yg04|1c_-@Cx_t-Ro;rZ^^>H zujw2EKX_uz-3kctNy5WG1)!56oSs$$HpC^gfYtOL9~L_rDo;;OD_<%ng~M8P?1D+9 zY=xnE5pD(7Rz~ezSZk952Ccb*^_|Ts#?sGFd3nLDd%W@EHA-K!t@Vi0`J76+@Svcf zXu2jUg?KP-6$lmJ3KclUVnqQ$=rmq1HMI`kU6ffUDbMojawYgPzW#h=;tVfQh`<+D zdUpR!JR)5{N(vvx2t_3LFnkFku{TyT8jJj&-^yoj0eIEv3)AV;|7OUg?vEZ!Cs3Dn zcm5`5Fc+x?wkl7L0|%or>y7+=k}7(`gN_B~FSn6mskC;bg8nmU(zte~72R{YH&OXY zU9!EsLz#kRF$TN&d^As^mBh-DOBVi2iB@7f;Tu6GmFZ3G%<4MZw&yzZEpD%s8`A|jLy`53D5@}xkC2)9*kO_2lEp~QsX50M<#cWS&4 zeO34_66$P7h~xz22UuBX(qEyVp|L*_sh9K2l&S=G-{a9y!NWtX`u;=*tumudApx?! z!b~Z%Vo9aPdimAV$`@^tW%Dr+Vo1)Ed{B`i%SZB8y!_zH0x*z$Of8m?^_h0?O@oy> zv3~#Zf;H%~W1cUs*yDk(2Rhx=$VdEeHR6|SYPlqgaMoe{9|4d;s6pgl@J0})OR@%{ zynhC4iQ(Y^Uc~ECY7qUPLs$oN!MrCck`ww(@$W!Z68!Sw(MJpX$*DSLRJw_F6H(5rw!3UvsCHZ@DQg|#;h~oT`Bt!pQnj#^LFm&nU7P&xGS<3Hps}v1(t|VHu5|b1S^$>`p zw}e@09ro#89U2uQ4RxApxZfe1wFB^UI9oM-8~@9V5PFr65WEGF0fpATtKmX}*p&d< zdEWnoVBiOl3RRY5%>PIu!x)ikPcpOo-=m?mpd!8mcW zMrU%^kx3w1^yYBXdSQC!im(#N&Qn9`0rwD zMIfu=6xvq&cYuZp>Uonf$FRIVc~<<`yf@3ak(SVpyZ19pt;fO?A0wt84<}QUdG_Fnc{FMyAPe@*tyfB_ zz>e@9N~sIQ;f~s`gU@DK|3?=d3bF)%6;vn%!hguXL+yn?&t%0ai)LgxPyUc@Ui!+6!nMy7_uia z#E`u#yYgE9>t%XT5W0GTqEmqX5+V`GayhoeUwE*X&cK_~!eK{!n?_wH1qwnL2DnvbiYCQVv@Ea`4;#h{mvu?NL_2Rb_HENo= zdMgEU>##@Y?J589JKs?Mu4?0f{rSnQ0EXl5FAZjzNj#`qR+?k@0p&@g25+306-oZb zclLt2{+uiwblqAxekB922Y6|=-~=%w{))k=kU{#cD?W*mwL#h9f0g_hoJ)g+yUzJ4}p>5WIbXZd?F@RKYZ$u(V^_x zZ|@^YKfuAp=*QDmgxpIE%$DT)M@Jj++V;Z4<>ES@g>vez2|*r6X1M%*b;M4r+1+D&U%YV$g2pt|PIBAwiSJHwAWdRQ20wZ4kWR$LZaz#QT zZNbuL5Dz}Ku|E7^D08GeHpGzDz5e-0rSRKN))sO7790Cz+}}(!KT62%mwCH}u=3G& z)u(5JS4H>BoG5yxx+dK+Q~&%pr2?3nA(g@}oRy?MqJEV2v#hB%>=mAUG5teWYAN-P z*>>%rzvFH<8*9Nha0x~|5xs2{jlO;s56HTGW{#tSBv9W=v3J|oagA&AzvaW9!2T39 zYm`u?&77ZMoSM;m9p5q#Eh!Xf;o9qMFh+BYhfR~ZNv90VP|Xif_`{kW?eTvME9%|g z!Hx_#K~kdi<`>4o83tggmEd626&<*I3Ogcira9;V_W>rRGqX9R-qbih39|Bf>JPBhYwf%<#cHkdR%)P7 zN-24QPY6<0q#Fb}5WhE0%S-g@y8mp;#t_?Jvno<=p;0(*hQhUeeBYpXahOQKZIwk% zQSU3-x3anbmI_6UVs96sbF;y!7>?{-yVzIFCq?l21ikxb9LD_1h7y`L$8j~OI9A;&;n+@Ar zMqgGu{&o@b&%DH_mF|z|dn-ROE1?_=d7PE*D`MS0oO-cn)qTOxaMgO5~*Da=4@&9vxKDYKN;U_|;-);bWtW_}uQ8x>$*= zl>UGkb1%#KG&)sQ#p`iPAMtAVKA&VleI_XKP*y48^Jq#_)^vfqG)?>wZcK}+!A+<* zLEyk7kymEbv5XzJ<_N3|bM7dD1;lpY$0JBr%o18GnJK*<^R*K4)V{)aKR7`#8FAs3 zwgFSsK?Q0^)1`dFEJLw{WA?j1?pusY+$bP5aegza-KFF3`Uj}0KhO`#CM!G^J{MEi zb-Z7+PB^2Y^2dpmO1COxmn&G3B(d8mrv&dXB_DHC|0gI&J;~DhUGfgJ0MWx8EV*fK++TG`N`Rq`OBs0p)Dxu51ZNk z*ZNL_KKlo=CmV}SLr7zLv~wHJz#!-ZbdO|y=ojc7v(o|KS}MoHo&I&&{-Ql{OVQ); zmVP|?bND>$x6kb!uI4BkT5T_|`HP*?@f$``v0)6|pOShu%cHMNcNL1Pz&UEU(!hj_3zhnjmmLg6X+a!PqJ zK=XrmS9(^3ax)^3%?MttQr$K%4Py8%9JXhbykP~e zrvue~D5*C1z0>*#qP|b&^|<)V6XzZPd$TX5baRNHeg6-U<$sAlwYZQ&h`2;4 zXx)rMgrUygyHyMqaXeVnA5{_}R$1t5pv&D?4{7 zN|vg@@@GFL~l{gzu8OX z1nYVKmfuyFx;~ptXC!8cvru)AwcfLbuPMITo1&nqHvA>^JdUZL^!{V2q^^*^;}L#yJR6w8X1#6i>x^(X z?Rj?0Ya2~jTpQ=<`d#wtOu0&MiJFrvUlW^lfWJtv%Km)yRzOGux}GnFdM=L`$XT+- z)nfPV!jZ?5BEF`!Hr_($=5VG+&jd)lkTo|pI-1k-B~mknCRlT#DFXt~RI$)bDbb+L zY0wL-XZz5{;986wU|Yze(@uy2*``brX}VUii>rs zSuN}YO}r$>o5F;NgP49%Hx^kAONE`RQFr8mx7_k(k0z0b4=Q+K5nMu)>^ zfYN4J67Sd)Yk19#^xcn)sKSqeBezyd;>8}fit^$-fCs|W?%&nbuz3kO(a&%g4JtT- zaM1Vn8vc(TS6WciIPR^qweR*%Hrp=v-D#F5D|=d9RD=}K#Tv~^D@Klm%is^Mh%$ z`0nh`j<-0bZi%waVQ-26BkBd<#MLBkef8T?S@zF|w(SX6s`Xj!k$Pt)s|eD1jggTg z0r8HFZ}e_hgWr8U z#u@Jc8}h$oSafe-7?j@QB{+05cTQHT{kR=P&FgLVFemqs2)J@Uu{sFqdmw1A&hL3z z_IDBC@}cI8vXcI|v3vIW{8GS`LX7?6S`$K|LmL_u1?3bfDiM|*dH5y)XTf6H>ChB6 z-!$wgrsShIN>0*Kn9^GMuWV1h%=;=-2Axtck;2Koo2$=qg9mPBB2kv3HSQdN8fQ$( zrHh?mLPoq4LF46BiVxiRQPH7Ksu_dqM3hSh_7TyUGTGDVyl*B$lIv1@nrJE}M$!8L z44b94K=)ww@zk+1+f*0J!d`xlV${yu#=Sl1o05t%Tldza;KSJte+w5ls)mUyYM>f(pxS37J|F$7%`y#+ZgoJO3qkNec(~< zo#|w?Bp=A(55r&KBPa-WboC@2HqWKb-Ejp#b{afR~^T=!N-Qy#sS3oHLmhY=;J3WUG&!QYfGM{wQR=V zqcY5cBo}#^u`Dmoh?f_&IAqUk<@4|n-)D2jxTE)-e=PunYi%jD{NNv;C_BdW_*gID zK2bQvQ*f5vs-++UdvF4i)96U*j!wRBk@!kZlB*K#2Zl370({K~Vxn7Ay19pB=WDDh z{$GMtW@LPcs)4v$?y2!e02Fi~MgJmaPFcplbD5vKA(M4m`NRxAI}#zGzR_VAfF~Ky%wim_+k=Dm8U=?S>=7x(OZDyxK2|5 z3k#8GnI0@AG?3RN>-;>$y=D=Pxz*r;g^HJ`n(PMZ&lBZjt_N<(glMUCh?K|oNo6=# zJ2?WC5uCB_!`BP)Oz+~#fu%So1k3vTy||28yA&A_&K6%UmMRq#aFZRQE4HZRmBr8- zJx2S;%{Ml8K0Ltsxe0Zy2D}`Yi77I@dSFu?FPDdYcIGnrMG%A#4!K`AJ!^%)XxAU% z={gKPoSfOwbr&qX5WVTcMajJnf7Z+3n?q>=?>^?24CO#8GzGIjfaySjP;(PTrtlJRBk}#Agh#OpzXr=*GjZb#%TPO=QN!92tSh^>?0oUQZvSOM zK|o5v6$)q8yI1niX-2Jp5+w&c`8`!ATbR#72*+p|AliL)Opr91N8cU6sqK4QhAe61 zE4cDeRaR52w72;ihFP<<_wUptF7g zIb+AYi?gW%lD5i053oBK-xvO0mG4FCt@8t;RvB?%4XXcyL#{*rH>H>ve!9iT>*sc^&0KOGpL0#qxr2sJ49L11pK=RlP+GB2$;w z;{`QB=j#eeUiF2iDEFw$)7E70N*H$(-1WZH9FRwxNU)!lD)lqFx)x3Bmz^YC+P3j>wnm4Uf+>RSw*on zzd&yC6+iQ|p`$%}CTB)S5gzt_MW2&|OGb7($#f;u&5+1P%xLUV$0hb)yTKj`eA;bxhSzxM<(-v6+5U8I%?UiQ7=NhLSNw(JSWQ2 zj|8{E5Q8hY?0CNuS~Pog{dg?b5mo(jGwNVkw}Ls`M{vb?XGT_a;7~37QN79d;&y%2 zYhPJ?!?z~8l|iRA#XE`17kylYGklIxBK!?MK)mU^*fGtO#BsD)@ z(+yI?cXG=*_1dFn1Xzi&5NN>LyFX<5l!`lZ6DYvMU7_o~Qb zIgVg!>P(kD9nLe1U#FYRMMBp(zEJV}KB5{J(eIN>na%C+U%K4X(GWJjXp5+`nkIHi z@fKP9((Sq$?!ag~9l z&H|kwsXgQE>{@X;Vd4|aSPg3NGJa+m!gN-5&Gl`=Ec`yXi~@r};LzWaZDYrfQzWsL zz&L-fXE9qs=5&ldFxM0=qh$`bl%!y&Y0!0M+Wt zlwfZ>WJIt~yD9Y^oB#Ok^RfxouIGAk$hMs$kI2ctV$n&X_gCF&UM}jb`sjKU811|i zZNKWk(lpQZcZ$W25E8Zfn z7)8jD1VPP+hsd(@q_9P1XRM5u3o(Uftxo8*IN;JVo=2-h?D+i5J6*|59qEyua9Nu$ zVp=q>-BpfU8cEheWwH=~y-{Ns6X(&3+#A&wg+t?3-z7MBn5_&YK_Q;kaFZf9fE)-` zzJ_XSn$~Ez0iqVI(|KtzyvH6Q#MKfR_*#Qe1i(3!n4RYd)?Ie<$Z0^@%lCt~+5Qi$ zMFd#ES$Z|cHjDY?oxqvDfw zvT+t3B}As}ej=W!poQAB`+j-Ck^BZ?zF$-wPcIc$$2{Xcvcv69AP%F?#usjzbA*Fo-6QRe~|l{zj&`_BRQ~rm&du5*Dh;b>?YvsBqJ8%P0?`di(mXtmoN7-yYL3tqQs zR2gaF%H_W8m^X(%X%~n@)>4bZ_bK8KX>`%F1X@lj8ZU8_VC%dtPfbK1Dh1XisyglZ+&FPjB@Hd+_pg)v0mB=zV%thxsw~_1M|gef5cSK$dY+f8xuPq_4*^>S>dY z4W#IK^-xu7xTNlWcJ#`9(aVZ^g>zoSVL4j0!Dg{ey7kxo@x?`O{x)D1VRsXKsfDk} zG?*e+_bsXEZh%u_HDLVQZ!S;joOk4-lKNRE7d^6${Pw(tzOXr5TI_*KwV)%6kCa}s zPjq+7oIW>O4N%c`u>Mk6`MTyl6yN@RXUb}rS7(CtcFD;?I=v-R=Vw_XlD&&d_xqPI z<#TS4_+PNdRto#6hCJqkL#w^yOh?uxzp(*c4d*DQitn znSSdo?&E8R>`W~@Zbt+C6s2X;Bkzd8(e2AvZ+-EF<9aL8_r5P|ms8eQdXLN9Z1e(2 z0_K!N1`0MkLi0!SJx5;KN!AUL7Oo0QH~+`;q|3qsF;rk*Mn5ksyalU&9p@d1O`Sb? z(R*6Z%vNWTJX2jl36pGTut^^(T2a^FSFl9f6AV+Ip| zg5d&-ITX$D(|6e#XDJ|GS@`?J>T{jhiuj(4+SQ%r;qv5itBl(Ct5Ebkq>5e@0C}tL z{lQYc3S0tWWeG(Vzzl!eZiVL~n~8zoH>Z_a?xf0MktKCm#JR$)HVbzo53I9=R)zS$ z>*ZhB&}ylYe-SL6x_+iPh3#I*ZqH@`TVqv$aaVgG3o3;QN~-=3AU?(fpSI}S6|cwN z<(8mxGMELcB>ATjn3s2`7u9lF6*Ez{HEessGe6RsI1RPRGmbZ`G?fqzK_Ipi9*4-1 z$fNz)d~#pgfkbkO(dy`Tm^4)#?H*kDZ8zSVrk|WPl2@WDStHsX8>o*}Yqm>1`yg*> z?%d2{E`7yacFXCMm>Y-rOaMmZ%Z&wdsjuXc7&8&=Q1ownK((>`^vh0{x1V55ix&@! zO{Y=%OdDVNSF8sPm-Zfr<+ljTX)U6tM+f!q>$^f)8g#o!J#=?-g_)c2BYWy5qt?qcfd|+>H2wc9>!{QLoPsFlnONYYa%eOKWvqA3u`Qq*2>$thAU7o5Tl43CuCy zj}dbOS*36$4voGR_Z)uwcx>naCw3XY!<Ms4a=qaWBsALrNkLDYch!b4Z{j*BBlD za7}^y8P*iNm@!;)ro-3Mn177IL>f-0s6;)eXue@o0l(I z1@lRlfHtQW?8#<`$)~nIm9D$yB}xO;G{tr2lS}Lm^UnK<7Gs#5?+ML}z=!O0zuH@o zD{FV#J*KmKmaVZU)7ctC6f?JRlWa)bf7(wIbz0e0-1bN)va5fSG7VW=(D#)8a^DNv zk8W(9aaIt3zeKLqOuD7n)G{rfje{a5Dxc)r{TT#Rb-l_u zanWW;=^b30Ly_=18#9SR0KE%_;E28f(*$`kKql2R3y9$qs>+Z@m&gVXLt+AV!_KpnX=x5{1hR< zperXKd@qvMp!>tQ%y>RNmX$N;XMrBCzvVzUC$#Lbd%A?lwWHaq$6{k)5))=g;~vLl zJCb+)@>WYFRa$qQ*=ul`s(;I*FxA-IUIpCozzmtty-!YMic^r%`;HSnEF;x7bPhRvLX~vy9*smJvdTM>vnVw;jo+_)*Qh@sOPKxLc4-Y zXh?*~#CmLcO; zlut^aU;c)eq}3Yc!dombil_T*EduMPnZ>X_3?ZI<14UWQ{xto=s^&v#j)H`IS@^?- zsOJwo*w+EAPkfA^jg;58T{coTWM zdIJU3{*jHzb+y`wi2+!pRUZ)rnQ1#tdJe(j8XDse=+2g_ef*~_IDd+nDu|Q4t*Ub4 zM*uZ1a?QEV^KhPP#~z!>$VN7hpI&sxzfz}OB?4twSilP=D%tSLtRfeGF|&rU23thG{tw{QtL&4~dCE2c*LJj;(T_hEGkIj~u(H(|H5vN7hUUp^TnjM0*)TAJXdq+QagI59ROLvCYA%aFr zH4XKMb{ohQmK>92{*NG14S-O$5WU}VzmH^(Gd0!prYRkW3`Z=y^+`D~{!pCT3(3gi zC5Mt0A!aTyoOokXEp6Q7N?+w$$+Oq8UgXA(eMh=pZ@UE*9Bc1gmtWOarx2))44n|@? zlgG@c(;@nVbF-`HDk_V&O|JH*$ z^|{!DlSL;bc0;=zuj~;;u)M5G4lnH?&$3~n8eMtd^_f)O%Yka|6z^Pabl4krQMhzn zPNGgk7vu}4rVGXFbbrURjox|=Y3#LZM!Ib-*0ZG=f5)fD@y;g=-b%8yUg#7A+%Bwm zyI&$m%Q?5vlXQQ0sSk?}>nY?>z7Y)(q5!SeV^WJOBUsr>8cYdp`A zi1kFhu~>^|l=K`)LmA({x>`ng>}u6Vp*`y|6Y=~N1HUdM2d4|_xray@51ciAy6x)x z#lML%eKxYHp)Yrmgf)k<*V{}6UbVaaZZT1lTNNlfUYW@(a?-O!lJVM7zQu=r_Z9qd zxD)OuGJ%&#{^?Lh-~}}6mKW^;IV07(L`W!k#E!(WuT@zLASM7g zSbJq)FXl^)si_Yv<9LVHqR`r}{p=*)+JFvg^OGSWk~;pR{g08>(GD5)-+xO(g&+@o z14`pz3}>?bMump-mmT(uOphpNDAM#j_YK7|O1htzeS2J<`Zib$08f#jTuVXEdpo!@ zIeLUqTWjxnmMOY?o*Z&wdN_t5LYn)UjcRW5iK0^9v>tV(^zPk&i1$N!BgkmAXwdyS zbLq^nU(9=VrSZFhcTvG_XxJ&u#Am3eOCJudXcdx(>n_*ODE$KB|eZhIwr#xTtywh2QOg{2q|i%{r%*Q!sH#qvIf zq0StbvncUoGoohCVxRv91c7udaYjkJ1F9||k?xON5s{r*ou2YJ;%G>hIISS@dg zou?mWFyQDg9+jR?1D@sHj3uIu7d!meAr(B`B7Ab`?{aA`2D*(Y|e3hz~_rrB6*CsRR&0@58I{cVR=&&>UEt=3qJWI)yaAgvV zFv2K&flU@aMjze$XeVndCO{?LY%)4?C)digJWM9CM#e8vGNuKBspW0tm-P+nkt{vgM>dd`87Q& z=z8uGRiBeiqhYMdB~OEt@!@KHY=#`G*~B+A(}Bb03VLYb3BB#FE}JfQ(Y}mRn=}sL z7wLH$VGg-yRh#IblAIwpBex!Z{P{oYS}TnjF7;=t6C+&e&hQGz`vtLmW1B_`rS0v* zvvzRBl`?4rHy>kBR;NkVQJ@c=ukJ!Dls~MV{+cLPAk9u&b>{L|jv(CCnkp-TVIO}V zgI}E&m<5Y(^k4j6hYnRD?!9FSXYl$Ba`(*2dI;#~@JA7mM;AVg zi_PmbtDkPYbuu5|VXX5^Ln_+nWk7HPT z_sO1#r7E_J-Y1P&IiXRI?VK4Xw}L9n9Hx7mte`u(`&IQT3xuvznVtYognA zy#@{C>+{Wl&x*q24MEXo9s~=jKggTdFlC14m#8}~NyrJ|9k0#bR9X0}*Srq3{Ms_D zgSq+oee^wMl1+lMULxAt3u`bCktvu5ja+yK&ob*>vl-fEQ2Yze_!R^Xp5!hk9hlIK zNRA4&P?C?pZc12GLn|TM8YCOGL{EBF^Elp&Z5PQ~T10;)9yEx(L4(RmA+qbQL!tu6 z#9c1B4&{q3HsA{n*U2;-SeU@&i%_rQAXstj45=Y}|CR3a5~5ecUP znp;2TM*Yszvul`HDRoEy?R8oMSOH7q4JVHM;UY`wI*#KB34Bhy5Ljl`yg8K@CDznI zIdXNh1(`^z!=%YhBG0dnYlOpNLn=ei5{_9(RH*XrGRBWIcySz(hkSCO+Dc39pRD+u zZ}cL$`N;(fxbw!~kR3*OkM?Gk#KW!WHP21ND2pZ&$Zk-~ZOMxGvvn>mtxs;?*&}QN zY*<@)=+4aJ%7?E!+Y)+yl@>xGJaOnREbDbrv!*I|}1IIF!{;u1I|RN?v65 zh-p)K6Npf%iCVgPDf5%$zGhJN3!bF?AqsfAtA_ z8K_mlD^O9K0W|R0O}ls8w+)p(o315b4|RU}N~`#n!SK1Cq0W1YDt2fpFzV9sn>nzjy+LE zxL%=Ghp3a*e;!ON^-&gWl(!}BjH)ce@-GK#O1m;*5dVPRLqouR^JLvbWl07Q;uxDzn)!jmj{)23_6G^=SL$Tpi zKa#FXhw?IcmE6H|YfYZ7r)wP%FRqgSu9V+{(RX^6R}&?2=~=UnFj{84TF#tps`!^1 zWPE?Tu3FEUaqYM>&WA_pY({Y6VLub@)f)R4f4&2+sZH)4GqBODsmN3!nV;)txJy%h z(5M77w!L;NGQI6i*(o*gGMUaE3#W6vAv@q|+!kXf0Chx|u%ZZ+7o10V{;_;sPsp*{ z*hDwxQ~2x{CctZmJj%B1NZs9TvDTdY`F`;SEeaUxlp)C*ZU2CFx(AMLQLyvMnI|H| z)?1(lX?0Y{vQ!K&&F!SLQb{w zHu*iBNX=6q+^fjVA?S;cqR-3l4@RV4jaLpwY6&?(hS`5q!0A_f;`kIDKE^j6lc@a` z!wo)&zj|+I?RDR_R(AUG)H@fIH${#a5T_a*5{@kByw;NVC_Y|95YR=rIYqy|PrmM) ztIS<#Tc;0Mi%^+TPRH00=T?BJU+`-9V#wRxXUJ{v0O>^@9rU}EH86r11Rb*NzndkY zvu89jz3dr=f^b+{5}87K8~Q!*UDmO-eb@ zL2%&0$vDA;(o^9>eBuFTNR4GH50OZ zn{JHIw9nzRiTEvTn+)>kVpS54JBQcR0WvDkPCFlh1}x|O`e(7WS%#m`av^#&bqMIn&}jXAj^bBl+B+cP|}z|>y$Kv(SHh`RA}(T11zZ=AEmB3AP!79P*P zfGK}jj!jdA^rp0GDV4r!$J~q==aV(w?>C_^@hF@4QyZtSj8I`X26CI|yltc*<#S)& z*slC9+Yh@99$NrMyH)~vUbF%ECt_AsSNR_Q#nkd;1dEq)1x+V}nt>G|;UMiNc<}{R zr!S~-8e!PjtOJzkv7+AJU}3I6&%np2cfj;KFUY>>*&Ym9F=oDk3rUYsptq zvnq$TXa(oX7<>Nocy0|Qc2u}^wcPcs(16?{k`RGBUIjt)BI2znz%{p8Ov64r%;_U! z6jUu~Mdayt4EW7`#soF80#-orSwV4EBpj;J9dQ1?+OZ! z^jhv`Kt}qke(o0~@z*m2Vibh|l*_wuu%?5XE+&EkB@H^pT+* z&l}z9y@p~d7(}c29H086s890;zB*kofEwJwSjdX9f8KtHLRpPiIblcTt+twzkN0FB zun`ZpOPZN?{ai%$`{g-cExkHbP2|bF+c$~Vd19u!4#r>t~-i;Sn@g5g$(YmHGB z;PT(6=06}zC*uCYbYND3$2<6a81@cG*J#P$lbNk4CJGDZf^$8yYmcX#uTp!AjjIw|{a)K@O3HQ%n{Jnt=gn>tP# z{?Hb3!Q3Dukn1E_UT}?%+ecfg!_@QQS=SkhDZXBw@V%kn`+|VWf%vxc z=DZCCx9*W)?Gy4!YW>5|KpV)-7|11+MqyI=Je(E5b5q1~2er=0k#85>673AE=uumSIMBW-V02scHAWbP0BT%%7KMYr1 zHVP}e1))gJ0u878wqnVjidW)wlDLWehZO~7I1R1@-xCqA zS+JvSV@1WKxpm)t6~+v5snS<18L8!T(N;{%`nYZvf`MK|J4iNj56lEetFl2Hv*J6s zG>q0+9dslvnWRH%UsAtX=BCsQ(`ierhH$1b=_kEhiI|xEy}}Gt^V97d>1{OEcF%{X zr8FlS_dTHBb%T6dj-}Mcsul_R6)L&n*D+24AoJQ~<-sC7+Q|@AC*-*s`(x2hMnLl2 zdLo?><~m#ldMnN83ZVHN{|D+3v?_|@xBkE|Xgy@>=yK@wpA(JF>Z?$np*qoD0dLRi zkA9ZiveY?{&2NvxL#fYdkKM3*X#;-UcK`&0dEpmSfYpD&I}J%quf&g)!w#LnW#jpX zSHai-tGy1kMcai-qs9pclE-b%zYXqQw^;el%#nLK{%t))-rxMe&~QVDK97< z1`rY0Lh4(46)!!S7h_k!={gBYAVNJO>;;JPnIvH&l(5|Q4EwvI36f_FCB9Vr$=I;r zSF-)By*k9#auhNUQ80)Kcz)s3pVn5l^98zm@o;w_H04cb^C97XJmoDZ;=Q*3NgxmK za$8qjc?*+k>|Hd7i=&>36FlI6PueH=ER6RBh=Jnn&)bkC{4xYdrd{!Spy6%s;?7ZhUVu zZ>j;#t~xLIP;!{@_gS-K>71Ae$q#>Z@qE#llPyxBZ@INVDgq+X z4N6LPgY?ib)G(xgqO^2MGjt3+fFKNwba!_T4eyPQ&vTyhzUQ3(m-h=^*C2E6z1G_6 zUVH7eetWdaS$0X$>~9tTSAoW+2KjWFEFEaL|KQ@ImoU+O#$9jar&jWr6cZHFlaz8> z5vn=_p!8%JBZt|Cm^u?jdLY$?G86p+SNvZZ#v6KV6Grl6mq*yXq)F``i8-!sahAb? z)>cf|A6KlzP(3qAUNGrT`$T_Eh?a~A(z(l@Ja>wlhl~imhtEr?#yyO+K6dcfA$j5Z zK$$i%sHwyQ+}shd_mGiqDUX$e!}oK)0z%RU+9oL`M>7v)F@rzFmgv-UHv^m}Mon@t^QGqpkUXGI{#C%2^ z#8Ihw6(5MUM&a7J%6~OYX~S(jtc-K_;ipPVPv0rj&Qo?&5wLpm?fU;DHC_h!2j<4? zuwUpNvdjE5Dn*8PRNCVuSG0VlvnK3*S1{odh0&z(Kz*}G3~0;_;#%enU!fwDtn|EK zLyEqrvb(8mKAZ31OQK&GWlYX!tKxtef4-UFF=P6RB-XY_Fd+JZTT2Nr#|-s z3flu(i9;rcaw(ha1*1Fw80(r&3;n`tl&2qjbs*;Mv~z;zXOP#BzL(Oglp+>nJpM{m z@~#SRmC?OmuPdG?f_;*o^|dSC-Ttkd@>T)()MIl-_n;f99Yx(?RozNE*(hQ}ry z+f(1#_OrocTuYuDS=3jTcg`b?GPezP*kfrVitvL@6%)Dp@Al~o*3eyFQokENeDiX7 zNfNFk3W!DttpYcGsC`gI)6?xR^+%T3WfJjA+2n;-Wpff@d4lCYk7rmw@2JN{XD=38 zv0-IMZIHY3dIl;uG?KsiFeOIcu0yh6zb*aI{cTr0?eJ7_U7)#S2JRwoRK+e>9OtsJ z54T)pS2JYm!WR)R*(N6y>i(`aoUmX$(`8>dg{Prbq!mnhfL3+}ri&ZR&+v?#Z2qz= z9Mr6Ly|eIW8`YB*nag4Mv5zd2r|kf+oSiC^*82FcYQ)3o`A|C#JF(XBPkX_Z#iV3R zzV0NfS?3qo*J7%z4Drb)*?H>cAwS=`pBe?Ci%Z=ftRlc}7O>yZ(`o?YaLfsEgakRT zEV?e>=)L*TfNl^lS;PBmS;Br^lI*Rc;>I+j8U6_Qvmy5Q#;N{3^2@KfbClA+qBuRW zYMFp9nVWya3)zSRP}&16fS`biZ*S5=RYdj3eU^6pTfvLBis znhNT$;VKK-E-@(hL6R=H%qHym>NKIchr8obnH6EOiHHJIy8}Z{dvJ*68y7n3nK}tD z++;$05D_XSk_SWoN@d)W1Ki68o*|dGOhBM{2}>$>h+8<;(E0LgoxM01 zq<^h}H1YLw(@R0-Gq933UfeQnW4x|^PgE8pQc!7V*1&C_ftOaD0|wecY1=%$oIXYq zh{w(9vo5OKd{gh&m>W;Bt!)RZH_#JPSnz+zrLWXU!R?j&3GahENWS7IYULRpPoMJr z_JjL(uz^$r$$(jJ`nN|&G3f@NRYqLL^ru+>#(|fltD>08O!1H<*RSP*i9tgHgIl_6 z%OB`E9dK+dU3HsxsVMeR>0G0V4bQog;d6b#LoG(EB=5}0BJY0sY)y!u7>8%7G0z8s zKpdAD&)Jh$H)>p*fql>Jt=6Lx%O}iwC@#n)L#}5)KtO+NA6+68I9S8mv2*?0JlSqn z_eMU6Eiia9Cm@2Lo)8mT58 zI+t2OUPq2|iHD4lLtzgmXGIlTm4c0@y zM@&JuY_mQW)u)Z4?|FE1P1-ta-<(<3RMbJy|`S>a+y~C$XJZ7Y2!+8T(|#X z{gp$8VsL2)g~`qwGO;A~eA*5jV#e1N{5w|B;;#5;%Ugx*>*__ z5U?O;KqiVCYT^3}} z8H!F^ffdE@k(un8tl;cG9$tjxRL8mqZ;Cv3tiOl=aRsRz3+G*-e|_0mpc`lVGwd!; zL1Y(oF2U|KE1%wIrN>3C2^D>F?G0z67Kv2e$?8s#XpoD`5Bj)!yo&ywpDP5ks`S)i zM$kR-GdJD00R_)ty*SpQB%mSVdU=PgdD?BA1lGFqu7+rNEU470E4GJnJY=W;mEOyv ziw3GRpkc_2{Zr^y?)GBXIx?ID1 z%J}T|SN_U{WERet>&$#EzsOQ}>U(#)E_ae#nA&Xy0}G7O4iMMSOw=V%{-C{y5$qy0 znwR47cEauUgM4Nt=ZlRjZQvYgR^YOOk*Z%#&C3T28i?Mm?_qncu675kf^c7V_uUuk zp1Zu1iI$SccZh{-TIH1hZBvXueY#h7eD3&qFI#KTipE0C<*PY@_%f%@(k}piAG}so zX)&F+-EgY5VqqW5>ai>MbQ?^Fe|C9&k99KkTZ3J?Os%ui3dab~hT%PIjO+6)^(F+y z7|JuhQ7LvT>Qo$eZ4L^l4Fja_vddRR!Mb@o`cS{?kLa`W-PLt`-KV?2<(B83L*1|; zzqNjQ#B=IjC1HGAjG+PvF$!u~epSzzfppo{r7Z=h#?etf3q{#wC= zqt1$6w<>CV39H;WbAf{YBf_7x78Y7|{VWwi>^E4C8W!m>?etA;Z3kG_V)otXi7T=| z3r2&d_WD=-sld-iRURJ@!0`vBYuI@#LF>htDi*QElEWD03Pc{IH?qm&`10+B{u&4gy8muOd?#x9wUiF@A<)-HDFy+7gglqUfb_lz|j z>#)4r6}{tHo8Z<$W)gg3j$xhldvF~|)3CwZ+_8$fVu`Oo>~hbwsYFsmG;pT|t|$3E zX;%Ko^i}3yx67SXf&QoqNWH5uMy^cCs&4wYt2FwPvZcvbyHxp+D$09~6`O5Fyb)gn z$}f<_AfJ#)X?!-c4r~nTQ?k0_B}h)~gK75gdKVQ!z;%HpHOHG{D_^10dG|UY2X9Rc z1Mc$+;YH2EbtkdTamI_9b1FK(fV}D*dowXQT|$gnQRf9 z9-tUB%J0daB!x8YSgaZ=wqxKAeD!0s_u=YIthMJ|&=W}{ONz{KL{CS*hYu5+O}AVL zc|xdCs1!&W~R#GDypuiB6G>JrD zv>)X>W?s8f^mFQ`p2G$SRkL(NYHQjjwBx5np%7ZbH6`A1%WPg|Pj+44qzW6I{#ME! ztKbfK%xXp5iB%&t_T0>?doH_gdk@!oG{%|FkFsx9CejIg-LRV(Kl+GTynQ*O>r%zF z;b7&eTT9>J(|bei>BUv`387W#E89UVJlNs#je0x#R5#A+xH<3iRi4t5fR>uD*${(w z_nWuf&ZI@PwP+uEa1oTscza-D2MM_Bxq9vq`97k1j(w02V>mjCX{*Q}_T2|jtj1%8%I!Lo=G zadguv7n=z+=D`4BM9HVyy^||c9-nAWuL5o##|T|{D=7^O+HaV44;~qX-yKgZF{h#6 z^VC^7GITi0Dv8O1%2NSF@B(haGshP)$#NnABXFc>gy!~5Kl!%UED^;s7Lfv^SF69u zUfvCs%uTQidu!TWAv{SUPQgElK?Hg-%NEaR??3RRMo)!#TS7qZ7F4`NO^#>L?&lAk z1(T6OMW{%Ma9T6n@d+C;k-qJHSiEW;hH`(Q*t&c-&;BEmh94`72*pYuAHz>z=P9Wv z@pd>BxXHEL#Hds`x=y@XfmEO(J5q%es6X2H)qWYv5I&e?WCXr^9vD)Vh-Yo)21}%uC)s--wVimK|?6#@v$QmUMGtw3h~r=ifG(QQKND66g#sGS+m^5cU>= zZ%lq$AI4F(FQs{hk89Aly>EH`;IZ5z)4v2FpR}lPVhd&gyub_negwg75B0@9}u z%VLymJ;h$#M#Gcn^jKMjvt_MYnI6)Hs29OEQc~EIVHr3Ig|0nQel~r04ki zUKtzc6Wbbp)2>a{x8t&9%S-DQCL|#1$}1b)7{76lG}Q?AE%$|T7VTN|L?<9=g-1(i zWcc25)p~Ajs$2^9>E>u=giJ)LxS#)A2)Cx#y(6PglM(Lz-A@ve_DAUoJH-_$-y9K$#ZHLt6c-S?_=`C|o>0$M^Ji@kO(;9->X) zk@X~JsXspjA?*6kj@~RKRVF1jxl^L<&zwa_P=2qG?G+%n*?rRI*vq+c$S-MpyL&(Q zjf>@UEmtS_x_Q{;<-Dq#`vD%>9LVH88ykMa*sgdFYYt>7eyE_GQZ+{4iXKDN(KR#u z{cl&rA4t5?=~HuDuoXr#$T(Bm)pj%6Nl&+RDQP1DQFco{r<}+P7AarSl^2`!cNEzd zQ4?-HqO386nU3TNu2*Iyo3N-MQ~5%ftAkgeaT^`ZdHc?FIddX-YaTAmv;?k-+Hqk6 zYHcTSH2{K%UCmTr3SmZoK#ei-&&8TWZ@ov?ypUH5_vXs65t2QNG-c7rVJva6B|~i^ zL^6rQ$w`T5h%Wn)RJ>RMbhANq0Ptp8D9qOM%YbRl|@GD_1(QB~Gge=`g)Y2RTfG2bV(E zReterCCF3(-V-Gk=*?x5p;(vI4L75^C8o+)^ClC8H<`SE@$lTuzBUeTc}C!_lx_s& zAh)_<)k55tmOBFb#JHrFGEy?>>UJ}wk={{r*SLdNlS3H!9TR(L0r$Z9&IC`aw+!BS zfAomrqLZD$<|#;@n>znpAN);qCrC!YDA}g}#Kf4e(cSBQgX?aaT+hj`Oeuwhy0Bew zE2GY1Erz)B*X`t;Z*y$oNO<0SmRh$$)xb-)q1s9^A#VBFnn2a=j@&qSbj`moMd(#q zO8Xk6N|F(TOpV;v9a#Y_H{k3QOE@9*%>mFVy7!N`C4k+pInMM1J3~PRIz-vt90aEY z{c4k`jqvhLk-6%~oG`xxAt&K7;#aqq10X0{Rr9YPL?Hz$$W9_RX#DP(En$yMhE{Gp&C%(Ud<2`uCygu_eg znG)sj>bswDr7EhjWK|j_u||PIpely1Elp25ljxcFBSo+)EeE87yD#3!3}(i>5eXLW zwnK8X6t8$p3^Ow?g!*OZHQ2LfW>w4isxM&eM0d|WlRJ2!p2WJ?Tu2P2>5eu1QeOG| zC9z5x;UE#DDw3OWXXYV8|J)QG5B_|0C3jdBoQyLq`tnRs=Xg7sI`zr{^;6tRe`=k} zQf`X1A#uG0xzU~}-nh;vrBfMQWZ|paG1lgXv^$@p%Tc|)>PYC4WVK2_3iZd7K3b`0 z`J(aoauU>k%d^6_>s^A6O|NEpH(WWpl_Gj+^#hNGG9A=kX)#{3sNz? zr#&r!V>a4!M9_9u3TqHDFluhstb=A?lsZU(XO@xQmUN;wLgyn@=--ESMW%zE^Bdubm-3UQU&wr z=WNvIa}*0-`I0X6yMAY?YvJ0N{4tUus~CRtjqxRa3zh3!HE>#G&K`r{v}2Z0FLH7> zrxW0Q7JK(-y=A}a(p3mk3d2eua_}LE@=w8G=43=a>q3jOJzL?zbb^dXrk6SuTyyrLqr4jWJFTYc(H>gE4QA@jJ zO$KvdefAsWrS6nkb%%r3Yk$!qx@j85A0@w>^uboo3-qh_F??(Eo0!1eU-IP5Bx>Wo zYr))sLs>H&tBfDC8$%iRQ}elc=p!yt5-d!r#rHTC>u8t@dNutntUF8K3)9K_STkd&P%2q5ZFjYdJ^C#?!53lI9_@7@sdrs9ZX3YY z+d@3u=SNZ$NjvF2W1;b)X`I|5*;2lk#lpi=Q<`!2v`4X#-1@va`OuAA?CS(BUdhr@ zrdmz03zwd|2#BsHmS#;i(eqKZoSYm>a0++3Cea!|CvQI!R@wk zbu%Rlrp3+bA|Y6harlrZ z+OeRIj{FK+r%@Azk!}={6z4j~C{siU5`n=gN0ljedeiZHI;*E;)^e_m4Pp1Q%2&@W z+&Zxes|bkiBk`2SF1l}eF3_~eTY9PsfPvmbcXXxOsma6Idj%GByYf9OGUwvA-S)Av z7f~D6P@SXfXDfc{yZ7}bqktl8j=Ygnt&7^9No{Zt!B2Mn&imWxoXH~>o@&vi?>rWc z8x5}t{EwlMO)=uQLgk2^Un1B`Gi1BdtTOC&5+2>W7j>0dlC#r{bq}fL_BsT^6NH8h zTRfMe0Sv9wOzv1JPspFXN1m@nHyE)v zxM3i)^YeLOOhKEHD?wnuexlKye&AF>CD`hnhY>#UgPBJ4(1+G}iV6PYRZB{)Tc8)H zvTo3dF7EUo$=X<*>n6PsWe$Do)&-^VK_^KpW6x8e=$ODn7i!k9XN6I}$5U5z`OQ7^ zi|~1XfQ1fY@Au~2TCzLvI77-i*n{<^dQu9IeuaCD&Xa==ry+i?E3{APzC^s3KDvSwX`$_QVQxBUk>LjpG!q(wOO!IuF^Wo zVy@GoF|;7=-hTg~z)XJXNW0?KXgHl}mj%L6+hr?bMCkJEQ^UDKiC%7-VhWx z2vaXYj-2WKenvg#+1uYmK2o7foM=FnsihvYQ(Y55pX;B9O16hjM2>z^bCO{V@zWQ-L%K8^BC|E}sK{bGs0yN-S(Qyq%{Uq@P+7O6l9Co*7+DM_XJDZb z)86!>B69#bj0$WobnoB6;90jxhENx z3@OU-&pR~l=aJfH(@hy4eCZGPcEkcC15kx~Vbh}$v*`eZK$VQUo=*F+2CNxU&BtLm z{;jmMjL07GajBa9b(xCi^|rv;8Jp4ImH$2sP&NPc@HeRpZG?Nr84{o|)O?Q00``%c zYT*ugH}p@v*M$9eBGgpCQbsJON&Y5`rQq;!MP@~B;G#PuYd6g;?;*XTZrbv|%qs4OUtyxXlMv?5`4XaOG4$+(6!T9fi{Q7X zl}y0u4D~rob;ZmmK12b zo!mMWTBaW`xlBJxdoAYQUuds@3gHbOBI}nGpvqVmNp40g0xWJu9{#U!s)iP}&*aWw zZX7_h`9w`p>WrI3Zu!B2i+=`ch)?BA7j&~V1^s$lYuM&V#pj+_>`&85f&1~z%deG& z9G0Ju<)UJOT{k5nxI|J9?|9(c$kMis`W$L-&dD1amdYCleS&GBVXTulP8@21A>n@k ztXon14Gc#`5^~4$-Z*yiv2l)X-O`Q$B5^FQX?@SY8*)I48)pP-4uFcB=9&zR?RKDl z{2=o9-S)&o$q2^3sJOESH_Xxlv1i18DATk4Hy`->(dxc*^T34T`j3v|r@3~u6%bkHAguL}1a>J2DBA;vyRWWll-ocM9z?n`>hJxC(h zgYS0Z{Ydwb6)L6H;qDyTe^AH3-vF60JFyqUe@M%-7&imJ@T1oK{S0Bk9EBTYeGgD( zxuL(_@UM&_>8SmknDjvx8)(cS%UvmEDtgctsq#tA(zt@FiVsq8_xa{@X6c%^=|SBd`7do)ZbsE*B&QenrO+kN>n- z7AkXeWbOnA>v5uJ=($Rn($OX!&zsFKLEhL7r1TtZAok|4JNGaYi6}FgUV_IyNS|#p z&9?$YD5w7rhvjebV)D3=PD+0Ug}gCvm5y5~HX1gLZy)`;IVu=$0K5 z`0KW^xg~YBV{Br@hnIbi+bqHciCR2@Qtz(%S68Ng)veyyT%^D0n(mO3d&|N0&%Cp5 z=A9iV&-(|$Jezs*h)i_4sRw`Nt$8!=vG0)~IDZkRN8jG$opr~rOq(iRF-c9A6+5~q z*Z1BZyj7HO&$5?QNTH){Z2ApLKk?rWNxgw(>l(LZ`2MY8g=c^QiLt{U{jngzJ3$`- z-|@~@2K&!bPM-X8>Nm6zvCLdhNfT+n9YLzw7`R-oNGd-}t{bN+r3OT-^7FkUvU&blaymOv_1e!!k$p zvZn$R^`_TiBSTN6%?t|w>zM8S1qI(6B?+-b?1bvyU~~CeGQa*kuUXX_>8Uo;B;Ef@ zdO5(N710&BtHdJqN2iuIW3mho<)!`0&=f>5oox9(1yt5Vpk1I@!(IPE{L8C7V$_k} zaiw+2w!STAAsK5@7mQymssOa#2((e!N67qefwjcgm*fKd^&tY2Cue4U!L1p z&FSz$&dUEA#ame9&WpWTV~v#C80`DX(VKDN*}aLST`~VG0%1bCcd61QLhEC8RhDVa zhnpFEt3PvfnBO`7S!Tdp0q<`5W$iTwb#P0G$KBa}0}P)xD|H()691Bta5W9#gQ-vPLTBYmrvvKD*#jgb(WenK ze~IJ9$zj~jQ7Rx+-5D$}5clkeN5^^2zEkzzrsBC9WgHTGKn_1_5p+ajsd zXg-R4j~OQgKv z@hhf`-XG-JPNM~PgpfJ`f&91hEq`G-^g3+dtouXpI(4AbdXpN}$@yt+)!-AW+Pj*A z&5ou@9T#Ksiq^z$p8t7gD41ftfvmD&TR*ULt5-&bDz?T){eMI}{GI)D2$Lx`JSD<& z{#gNE)#!hlQzMa&b2uW$-9UEa?bfsB%szs;yS~x1e}mY4g-_{jg|1Ns-&5|*64T5b z>vayCRr%sNK%FB%qLS*Ti9tGn)A9ny*orh8G)IzB4kd4TsL^JyXur?>vLf>pB!5dL z`Hw-w(I}*Y`9SW+4cG$u%K>nw*GH1ymoomjSAiOn@N@(wt2$B6`Y1>?P+dW*XHK+J z>Muk*poLjc32Oc-d^l_XuJEK@7sY8UP*w5ELFtcg2ME!?bu&gaGVg`fpIB2QH^B&; zqf+MB@Qzl}xQDco24eb|$VIHZy?mwC>qOP)Q?2ala)TX>x2B~mI%AZ^r9(F2l1%73 z)&ADiGvzy4{HT{05*#lN6SLXbpF@{Q|2-aILTcD_HJ@s(7F^+3>3+Y(kq?G{bmZT_ ze&0s_5MiV{^!b}q_wTRBxs%`T^g>Jj+j#Nr_+EZ-VaW{Mf#1!^;?X(de*j=WjF zC%hNezJ}BH;-yINkMi`j5C@GY!r7f7n>*yh$dCffcgW#_v;}}$wKh39|Aq$qw~hMH z(h)i?SU;LB$lh{yyGNifJST)+G>`(b-=lujGTbWxDKLZ|8e6t*IC?r+Shbl>B6|nsd`C;^dgad)0dMRhK={} zMH>S;s$dF5joc#J{NEi_DOZ((Vlx6Kv}T>pF$b|39D^rb;IB_u1! zd2&o`v1%T2M6}+c{llqFGXs_tC1tu4y@n3*a$MNj*PdWXRZ~c!BtV9AH`J8CytEJe zt#vC~F^2gntWeo4k8a-~6u#y25#!DB;pZdL)}aCyt$XS&GYwv+Bt}Zv7(X3;|NbRo zDqA*zz*2ct&5<0A#1ypX^wg*|Lk{QD5-`MDXqwW(MsmcwFb$n4_iT*x?fMnZH^FSV z185(3e$PP%%?+Pta-T}EQH~cDmXx?Nn7Xk&uvILE;@|w4j9)0M^`jhZ-Bn9%yPt2I zvGdMgjhz7Xgzr5QGxDLqaX|q;5Jhk0R>Osr*B#w_oAWuy`2jK*wbGmLzVngJH5{C` zLF-|raCCXB&-ZfmUgJ*d?|D&w!8Ycjk6Gfqa^>4BPm8W}BI#7--Yb#sPn)=ZalFzy z+aST`IcCh~cYXJgV&ytkdc*al#H|~V%nIM%K9QQn3U_6H+ZGUJV{0xKRvmyftgrDE2DlugYQz7s~biuUlR$u<_p*CI-De&g##Slqu z$iH^`9)P39Ejt0+mIrQ=BuMcEA?}B-`mGQw_^zii+4&EB+ky~f{V$XL`#BdbcixZ} zi(1G&Eck{}|1PQq3_3RJ(@@@~o3t=Rv!uuoik=D995~AC`U%#zp zUXYt+DLZsA2%Tm950WL6q?v`h9aixv%fe2dQV}^6`r#CmgqrumR#Ii_elh^@d9n$N zJv#*rB=n@lYjj~Us{20}l<@4WM+%C$0Vk@@x_9f>9MQ3IBJCy%Z}#5k zfIH*Ss{!9=NHTIe~J(k4fG2Bl2@rxD&lJC|xymo`&> zMR764@odMFH}3rv{LbK-tKU@PI(EL_b|aD2PWSBknyVH$_J10SC>nMuk$B#OjSyae zDJMDxW)3@{OSbR2*OmD+K}WKWe*5?P|LE#}o)$HI>iVkHS^BH96ruR+EvnmS>(I?Y zn?@oPref4Ve2vhLYIE3f_BOiLe_UVKw=sG_@t_{rYpitLXA~cQVP{=B+ZbJq?*Lb$ zo?kw;`wx-?f>Z+ZYG7Qy>1EiI2pHeOTA8dwD*q{N2P+{j-yAkVuALZZQ zt0@NDt!X)FRtQ|({XZ-1AK8rl|0LReS%dBxaE0lX)__X4JPy&bSwTy!90QSqlqu0G z(SIIyBX%_44n#`ye+{hkQ{x$){=~#iezD7pwngAlyAiO>FOGQnWuotNzm>VzdV*8C zdR+{1K1EbvGo|i49M_V+$kS3_-Z#K3f!egEVw}dSw4Jfxaa{iDa*$CODHB3Iww3t6 zj1l_v>-3&5H}Wgz>pGlP zC#`3bBU2qoE`@wO(}_m?j7=feSUx_XxMjp<-SDj7jfvob!b|zYhm3nOk2Gp7tT)Cl z1^L!8E;wxli=BkpjeXa4r3jS*Ib#2c4t^_&+7d7|v#s;qA-caU3e(uZr$r`OyOV5u zVwqXn8Ip-1C)nN^>^UjNcl1qV;_`GQWw?GpS45JNaj0e$V{2mCz{P3Z;2wpO!XZq+ zQ}1$Z#o3y$CyOX!<5u=#7Gt^;@&^a}$0ryTRdT@V|bmX6AuZK%C2kBj!~;ncmM3b%vD zeCs)MTW1@ErKH5_jy^jsGuH+((e)-`0eF==_2<)I`KzDBKF%UhL3`x9h1St5kj_$* zP8%45e8yIux7NzHO&$gZ0Zl_~ReA#QH(l!^c_w1_1Y&-_yW_5n0zBtC<8j0L_t*GSI(^%SnY=JtA19(xCEkjpe|K$x!y&DUXYR3jGdTW`w zqnQ(A;2;@{b~Oc zT-3>b4Lc!F=;~8mVhH&&km>NIn{#tF#Z-T|Jic7XmbPt!nSP%Q>2N>(lGaGFGdh zdsNjd?MEL(;~ZsLJp?~Mv!fBOuffM3n%A<4@ZW#QhHDzvY8-Rw=^s7Yos&oz-Hg#M ziCO8XRC+4&RGT+io3@LB;LN$Pe%2<&Ac{h#4J2 zuc!D>lO3v5HI|cv7krLC?+0qP8zi2T{%|a-&6O)B5256CQB@~ z8Jbn8JXTR9v&Sl_jn|uLAXU1CxpLs@2i(r+2BR8Oq@N+~;Z9R>egsd0F{+P;5h8#1 za=r2Zyz3D5^6(;ADJRwBtuVr1PE@LPtoHx{vU%_FaBD5T(o}<(}42AzFsLkKbM6J#S(oQ`8mGl~1 zbMB!)NF7SS@zT`$ysL~o-zd8<{8A5Ml10YKTWY|?t`0Z1#=b>VjPF@BV^c8+U!Ojw zEGJ@`u6!-xGO4>$$59G(3M1Cs>+ND~xB!_&<pE2y3?PwZ^g9pu%X4pF_7 zA8M)g-BIHL+eJ;-q$v88-;uoBS_)pjdaPDI7Hp?37 zHJn3!;+_+QO-&X?Zfx_`g)Sc@;(4sruePuW3H|a4!*g@0!hs*{i{)pXtHG>PM^rCQ z`a;M2RbAXHC*FFT6Oe;KaozZb)vGq@);k#3Ml9x*WDDV$oiicTf@hlQp=cvIBdvm( zmli==M`+^d0Y~}Q=m=-Bi|oMJiUU9oyN4@-Yt!{E~UCTXW3m-#C{w zvy(IXln7x?IQ1s~NB0(QE%E)dkzC7T$|;*rl243|p6#3Ij?$>)uasPU?lUt&ubuot zz`sQ9jU^@V#d>axlv^$6eUuDk%UQsS2jlk{I@YO<2o4Ujng2UB4Flgqq&HqS< zgpyBpTVgEprQy8cOS;>u=GL}0M5p9C?`uO`#vOYbRyl-~ zWu5Ugr`uCyUvyop_lKP3sEZ&YJ-KoMao%TnM`ZauChD;;wNM`36kQKRRlcLari#~? z<9dl4NR|S7`;0SQ-qfq3k#4o3xhT}wihfPwYEvr2Np%h8?xaPyx)IM2QX;x-C&Q4c z*61Q-XY<8ZdaILc5PqHHl!hajEqG9OPU+vxo|5qGMWY93{!wk7HY;&N9&MFU3qVeO zC3u=2vl0iB2Ll(2#V4ojtE(Op!3N8V8cv5*t%vI};)!rz>#A@8NDmXkiYwAwtp&1# zV_Dmo!Ub{fS|7%W%oZg2}rm$F7~)?QaH zZ&_?Y=+VgGifIio$zo;yt2+eD z@M@zd37~(ek6t3KwxXzqkNsm#S8B%nReg!kq9?GBR6EJdbw4z=4)k@+Z4n2_8@>%k zc+4+{9aFjl@V&eaFS2BXCR*U6JFqM!b$+3odBG7?uDsW}$NIge`EklG<+mrp zr&JMo*`Dmg`LGTxPEuRFs7BaD_v1sf-Ttrb&y#)j9%kK{b7CDq_x6Mk-?@$WYStTP zyHi)&giZ9)z)>$gdOq`k>R{zagT#+c4z~l!rR}ZiOzBbejfM)dtZqBiiFHi_XFE?I zc%IBVkXepD^lzVzbUV;5i{>hKIX)k3^3QL5Iy7agBJd{XLxWra>GljAo=+I%ZEzh; zS>bbXg&o=A0;9fvZ#1Y6ZX(DlNEgw+qlT=;)bH}6V(0;HhkIR}eWO^wa*K}q?AUQK z(YYVAJKcPDax{jWL37?(G-E{Sa0gvT-n`4j(k9a1y zW;k<*l(W7y8e`5nAvKL(=l)LR@x}za`p0(7<>N|MTMu>+E+|QP!E~fLRUs;$^?i!D zyM1G&U96k@G2dpuM$ulLGv84!oKxuu|9rd1b6@?Lg0f)(io?{?U%Dxg2;?|d0qSIz zZFt8g(9fn*zL?lIhVpPez!sJ-ROG>Ntn@P@JD~R9XyZ)q8}XmwKfyT}v_(EX>hn?&6c0)*#)b z7OKsXrTe0;TxOuQa9?7%H;AohPG@Z-rze-p@xYlqx;R4tijCcs1ey{|8`enK11#%S% z-PNje#)DX`dbh0`qpvdt;FXWZ=sn;GN!gouND=fWX5WI@m^_|(l=0kH9Rl9Q9pLT* zt;f4IHiy>O+Y)_};0h+r?-+HRSWLw^m2;W2X1wdO($BS=g~_5?ptI*2|D(9stM--7 z!X|7h*cOPc>$%XtV}cw0?LRbDv$JVkX)#?O$7?$;O-SVrdf}dQMZGC&uz>XAjo}4B zGQO*|)SxR-GNL-%IqlNp(VdiGm1Eb9orYS-V?=OGteSaY4A!BNkF9K>)f#NGnU(Ye zOxo)qqflLBpAQ`5d zokoIn`I&~-PQeKLi{r@u`-+!iAC>hXHv3-tyomjPh+SqoFfm`L%g;#XZwgIW&S7-L_5 zfXfScFqMwpb6X=)kdNnQesX!+;afbvU;XXNapjjQG-{Sa`5Luwk%*ZX6p83 zh`gG1-f5$8pPU#G500Cync9ZTOo?!<5ofN3Q_MK81|^9jSHaSF50@9qTuG?vZHRMjP@w}t zrG4IWKxKfnL1(3P6OpcbZ{wGZy80Y8Y7nvSYoKGo5v;6RKk25nUDt18wWtx*oQqEm zpRJgvo%m*!?5N8jG+3l=T+ncgPF2$z{{}{-xn1;}M4xItm&pqJIZ0Bva&tXdVED79diGxej!)ivXT1&0kd*^Bi3-M}>>5A?iPItb}(Rb^adV!W= zwDv^(wDZkwf$H6Pchs7heuk?iy_1xUGmyO6jGm~;kRhJ=d}pvv*NKr->7jzejFjMi z<%p=mfTGqM_G^J9h*l{)h}odE_eZNK`nN^hWQv`EsYSDf-5$|zlm*i41<&W3=W!F@KoK0rE@jD zC#>8+7>2Sbe(AC;Mu!|FoB4{JbIG%{?be5(+)~=IRcM5( zqGs6Lsa9aw^bGolMQhx@u->T$U{3>bg zw&6DwN(LV`$aqeaXN3y!HUbH+`x@|HSV8J1tV|IV=B;`j zACE53Y_AV@CdYHP1n-@9bY43LZ~4^$homW-)};?JU4k9Ywo5U>gSyLMVPGJV635%9 zCKMX}?9bHLd>cG-kE%q1ip&t$*_cpFREEGUgG{O&C?1eN-1P4ZHe6o?5;5rw<6~dR z&sD*yTt*xg+bGM_TefP0PdU;=#P90O^SN;8^RV}N{4%!U~v$aw;RIwKpt7*Ejg=CknFv^k}_hiiH`d7g~Z)EUXs;EEbcfGAb|h$`H0 zQte~92pwvbkNp7bPsdqr!+Ve}M+JpHuyTy1XrljpCgV*^or-$!{$Fdi?I#*p{Z(u5 zQU0F#kcxdy;WwhIr~P*cp#7BOySCm=KqB6%9NRd>2YdO=*x9R&7=+`RKeo{e_6oF(RbZ5sjb#! zRK6v&;s3lZ&@m;&%P<tT5&H9V!^gQKbCy#5Z+>I@oQ2yB(xsguvf$WgVgN|>~CFj#8rYgdh4hnOS z%U}u}uA}urLACR%bE6shBH4Ci-4;+C{!2rL;<_BXVXyzI+v>LVZU;^@9R$_J+kFyb-i>^ykk_xH ziU6hmKkc1qIGgDj$2&TqQcJf~scqURCfZtBr7D)TXw}vd6hW&!h`mSUr?6QZkTc>=f+=1pa=XdZ2ts!%g%9lPB&-(bRXI&2PX1irLu(uP+ zcl7%f76?rB0cBKBJ7S}#_r&pSH=n2CVQ@%3bP!P2XDgf{{JW$b7VCFe-E|~+B~Q@P zlBWgr`cgy0v951hOg7UiTun%QT@f?M`#Le*S&@T}{Jp^wHy9;+RmGssJ* zB7>%6oktEVSLDQ(c(v=i$<__)s@-WJ0NbZnywBE{3iD+GX-6OLrkiymR3)7JY< zJpu8OhmQ?E6k&fx&|P|*SA$AelUyDs!~Gzgz(uUindrY?r8s5o%t3}RLtf^bI_Uc} zuXbetYOTrj5zehsf6I6IMpZ{%1SmE&*j7uC?^XTbygsA@mx^%1KUrc@Fyzbeo+&Gd z{$_Ggs;>n_!w(b@@q2RwX|2ZN7bK4xo;c@c&}f; z#@_(oN+vn@;$GlMO|*AJXjUEbd7`ZYt{3Mu-uf67w4i|UO(Z)?P^LoBwV8QckwcbN z)!A)x-*;-G00LEU^cNmS!tLkUS;}A@a!kS-=rB&t&0AL94bG9TfI{7V`^+S*%_t$F z8D1g=Xx+TteodS~e{skKK;LU^tvy( zij0xerzfgJa7WK2P2i!nK_SuiEx^#093%Yr=#%%Boa^E<_v&%xi=(k)-Tm%7@sGFG z9&~ap?x`3EA(VK8SLK{nGwZUYRpIApFUi;19c^D_3|Es^-cKfwy#e!X<~k_Ikm#n? z9|oZc*j05}R#*!!tH+)`!|dj!a~{uqR!e-7uPY|k(#DChS8@E*5uk;%%N$0ymNT+; z!Wpum?niCCWl3=sOR^kRf(9E>Ev2#^t-E{{{~M&Iv?4Et?Axm%ZOK3*N{0=7`aN(f zI)Sq>_ax5oIQ|UlRCPT}E~s>svo5u515<$e0zlg56)+aZ zm(X;E(L_zSaTx2T^!Cjj>{=RXCaGx}Q85w|Cw3Z65Zq-@*uY^*Y#4!9rS6_%L^V zc5x`}5oFU%ke(8edm$mO&fh@Y`i z@T<^IHc4aXAjV{EZtzwls3b|#`wEL_H%c*iK+gful?ohJDBDqTV1Ys(*bEvWYb?|W zuBCC8?_gW}8ZtBQ&-Zayk)K-63+WuQI}uXbxW`}yNH;cD8OwOTT?8zV2pe-h zr93g^(4D^pK`R8ia0bxm+enUuO(aq@)zOo>Hcudv0$NY@&C3Q^tESKP-NST0rQ=Y> zIby4~Z>Hq;7PF?8+N+$_$COe1{tfjyMnx3xPc*a2Rz0A$q}z>6naQWm%s|qW&)Rl` z$jGfp9)DtydD>|_)6LG+pCA*t=5KR2H{BU66A-ri!m+H!*QsMb2{Y!;AtN z&;it8*MqEOvuCFD?oN}yQ2kZz_ z`Y>mVfMZ`C00gB+MkJ+vQ3OAE=$7zEda)H45}3htrf7n~yDFa5z%0dg$we&Q;zZyw zU}`-oI%)dq<4JFIwR5hqZmpXYSW%M+54-#W*F)6|>Lip@Spbfvw`aLkSX(L{ODGmk z*-bA4+}La%ipQTJ@S;c6AR43Gq~@#&hUun^c%R9wbv>+jLps)@42{faJnG{R+w{D_ zT%^Jvq6_rP`W7$+Z%uO7EUD;%!Vj|~a#Em~Ay+@UW7e9~mO8S{NcJUwE=18Dm^muh z%@*$x26e7t<*-NNAYpCE%g7&)}VbTuAwu>)^LQDvU0unp4A3kIlV#zT`KlZty)KK{Q3-Se3>v0}h zTR`hvxe2B%Ij>$!Znp7x_IN{CjJR$Ou8)YEer9OC98b4~K}0ZNnKSNf02A zC#D>iz^Sep^c-FV&!F^b6i1h~AG06%6*)61jO&)6?}F(uGv4aV}A)Wc)xFE1p)!a)qa+7;+o6b2x4 z(f;O9zV?nG&~Y0tE&chH<>R74ogYnJ!^ek-eUgl21a^|0*(M^5DNe(dZ*~G;Bkz5B~cqN`Qx_cqRMNx=GX-8lc=fgjGvy1AC zGxAHHC+K_tXIxVKpn(ZawOarVxLw0sOCSs09&X|%Mj?Apeb5~RF@x+l(ZTYhW1R;L z7K7*7Ie5(6efJul);(*Bphn3`PfuD9;u^)e2+QiBsfV-g7pmStJ3TsZQqFI|o(u;p{e1zyAq?5CTG>`tQQu zh6>;*-m-TLNpnvJomP?kHJ)Afhi~uf(G-}ozDU4o+epw!w?j1{JvTczW3ux{_)h+s z1i%?K;?c1In>&k&9z{Vr3o2IyIYV`c#15i20^M`@PE?EGke=cWUr3O;qIvf~Q|=3K zK9@h4*DDp!Z%h*S0<3_MmORKPG1VQ0VF$RzFa&@tZgfhUE0&3{Yku)cyfJwB&c+@7jB=?QV*n_LT?}|JB*SOXLcNW)Qzx+ae&8Pz`j+c&BRmzBrhm zwN`!7qjgX8VQAs+F7zea0)ITgh9n(OIh85aXwJ8R@_yCe9hP#1WeL0a`hNaU6iY?V-8byAz^_OC=0R5Sg zMlI3&uiyV%^}gQ=~^NH diff --git a/notebooks/tutorials/model_validation/inserted-minimum-f1-scores.png b/notebooks/tutorials/model_validation/inserted-minimum-f1-scores.png deleted file mode 100644 index 52ae43c719a2621e56360d2f2f357256253a4fe5..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 110789 zcmeFYcT^MI`ZkQBARt8$kX}@f-a%TVsPx`@@4bc)0!mf7^o~gHp#}&=1Vp;j(2?E( zks3N*JkR-+^}TC7=e+;Df1Io}naRxFvom|&``-6;-Pe3nSCu2gr^d&^!Xi|Vf31my zh3AWfbtm`%4(5z%4jvB{)Cq^h2&QIH`=s8f>$nSJg)=Eb`I z^BsX1W2)zuFW>HPV(%>tz%#Kl8WFTfhPTMu>NiR+5(u!QzA&ni_1|wMVYr9oMk;Ru z!Xhb*7s<_F)mfP&+5Pc3ghY?cBOG6Z%_9TfY`gAb)UwRgJ*-->T%G|eseGEliU4Vx ziM%fR_ld7xq(~ugYt&r#t(m&Lm>S;;yQ{Wn=4PLUsD57(9LEtlzlxTwh%Bs*$L{uY zAPTy;O^>}b@$`C(?WNyUmcvHyQRDl=qtj<#&ot6<*(>P;ibr@g!dNf3DNbiN_yn{b zkyh8rx!m*Q9;7mtk$9U!Hn6*C6?*lK_X8o5YUkU7dD|l?}`4q?5I`WAPCk4sC>jd+lWPol4H$mh!Y$=_C;RBucC$^0@ z$E)R_bPuL7V+*x=SAbg!PJNjCw~q`zANBQy5qC2`zJIUGvbT>}A)q3%_oM)H!&g?* zlr6|wwnUGnT))Qw68NyXc3y*Rb#(M=MSFYo%~)&6%j8@?5XCD&#dYF#%1~_X2Uwh+ z-w6>a;}}TI&+07QM`=q+zazG0#QyM3_LTrHb4|6}-RhdpSXX$T9{j-VJx8BJN&(h7 zh%Az`{NlAt=%oCEay6gRNQAwl41G`QSG}c`@X$XVD@W?>6>;1d6yN-zD3ty3R`S)? zK)aXmcqqHa*aEd%*eTsomRNkDPhQve5V|wW6A(w}z0%M25Z>n_=>>wnY01zmKU&7g z%XYU*Xi^vM9;o?Dw}X4K$-38j^-EWMoE+WTmM5nDJ-z9U`&W}+614M_!$^Ds*N&zB z3TU|fEmtr@j(P`O&&OcRaE+@M!m@IX(+*8UWiPz0tpc^%iwHL|#is?l1NVt%D8C6i zmppHLApS19OVyGR%2GiOg|6yjPX*KItaOs23*34L2Pwuk)r2RgT5Q{28Q8F_s|u-R ztD0#;Kwlg{{X;*pIkUOR?5sCxUjB|^vOIWpxAzgJ9jl!lR?-r~P;%|}3+As%a8-Ui zdr83kp5qrz6M>!@uHT)OPVwJV(j=rZvQIt$EQ9r#Pd^BJdm9ih^^EqTbV9uLGwe@{ z_|`1^3~bW6%Q5Sb@r%2AomfTTe0*AicdV?8IZ2hikMU6n({OxO ztWm6>a0(qcr=E)Sv0(6E9{y_3gl*i(h*VfCkSEPel4EDCe7eU>$7=H2Cms+#`GdvU&XYtmos)?&L8(u?ulI*u z-^h<#wv5M0k*v=Xm7?B)02E-lGm;ciMYT!5#MpM7tA0Zic3)*$mc55 z8CM;p{|T_?6{HpNqn3#gr6#BLq1F>J9b@xW5{m$QGL_;fWcX}W1a#08tQ2g{2pF2l zAkT;pa&tJf=c|9~@Wyf7k!y6e)JJ-^Sy*9CQC!`qq)iH3e#*Aj`!L#K_!s}Z=<{fo zXn|-Vx-YQ}L~5)`#jovPs^zM%s>Z4)*Wi7@1^k8lmObTYhn_V)ue5=*%XE?S+;A$w z)rVw+MiLdi(~}mclkna1?T^QiC#PErKQ9YL#)v0Irp)rps)LrWM4$p6gwq_Mqral1 z5$@T`rd{{Dq9RZXHcTSQsJeX7xbewwh)9%aktwkoR0^+-8zT($b1 z&GL|pfvl#_F>#e82t4mZK?UOO-;%HB8S;o6yIqP^aGdRn+Sn<<2ZU*92CORjv z5YPas1Hr;R_UK2YQwULX7-(|Nq@<@fag7+X!)hB9>kaL7+)m~VM>6b8%uTEyBjL8I zP69neV;e18qbV+#F7VFNNIJ(&$5aPB3*ZcW1F(*A&JF5DBmHEVPlo@CXEztF4{oe% zyKGFCtXz>*lU1Ho+v>n4il>QkRZl>!Ya*?oL2WNtg+F*XLOK5;5MEJ7) zGUMV9!U178Prt;wZoHISo%XDK34h6Roqc6<6LGcVCD~lp#CjL{?$^7OcT-Y(U(dcC ze#Moxm8Oywl3wVn>J$i83nh{HCZiiVze06zuhr2MSOVUv_&IK|V^L$VZV~rw@0-nX zjMZpOo|Rrq8>Iym)5nsyd!Jf==zK!wSL7GHru)e9R_=}7>sK*UWY#X)$=RZQ><=jn z&Lk=gs|-nUj;2^Esj3O*NVs@rPM8`#bIDy1wHZosHAr)Y(|%Hm7g^i>%5Y5)7PG?I z(-n|=6?sLhMOH+gRwSgtUun~_aNIMVOMZ|1iXS_dDQ#3@+|D(a4k}|CdyzoCV7S%bKF%EQ z$xmZtW3HglnuS|CJMdB4R;KCW>0H>*M|F0?50#-YUUUO=6~z39P#fu zi9&QNG0waD`3t8%w<<&a^&UI_~1zP_7 z=CtQR$u)Jy(AbFRdgo#{PZu`TGO(XYQ{`vu;7l{_i!4J;CL-=uWw#B(1GnFH$24Xh zh`22&Hcti``CXo8XNgw_4&3e?Ei|Japd(REi#!nXv(JaXxg~TxynzAo|7Ba93Rtjt1AXNo&p{Kko>yzWb-0bn1}dg9j&K? zZrYF4tJQ0(PHi0xs#~obnA^xH$-zZ=EdxkNnT2v5Ns>W=*?Bf zEPZQV*d5y2OFp!@YB#&{@I{>PHdgE69V{+9EF1|D&EE3i2*`bnY==vdGc|pma3qH- z*2K46H_mo|6AUnaM9716^h{}Kp_H9wg zaKWoH$@=?>I+VQi*OVc`A!IN3_$V>xBGN`*!B#~Diyd?P0PEfzYOMR1qdS-laEIo< zkKf*Tfrb5#^LMeZA{?;p{j-cJX8-3CkJ4z5RQe5`9+G^{663PuhMwFpRxbz z_)ak98kV${tbzh&uVv+FW8>;&@8(@Roo04%+(O z`YOs|R&Flb7S?W-Hr#$L?tkusCE+KAIdrk{wqW#gad!0*^OI!$M+q^^@t@N?%#8mi z;_W2KtgoWZDC_2F!zj$n%gxISz-MG+l<>5+71MnE_Mg=;za*LMy}jMVczA$7AU9Be z+s)IChfh>gl!up}ho7GdQ-aG2Cljj1Z}=d75zgP)DF!D|N> zOrK%e00_JkmiR~c|KrO4wD`Yj>iWaVm`5$L79Sy*j;Q8-C1K=k%KRCyXBdx=0HEqluV`cyNbYi|= zV75Pd%+?s{x1z{}g(Zch@LF2i?+$VSuT|NHq4yvc5`~;et+gKDR#GS|vF2b?NQrNI zFe96UOmOZad!rpn+ONrOQ~E6Vc|iij42*N+7U(@J>9s$f-i#))ts1!n*$fBXY(j?T zo8!(et`oCGRO}xxzQelv?|wac8Xs-0*F8@l7X0s5dZ2&z9^8+FL<;N9y?^iP{R1+_ zX>tpZ(Eb0LKZj`q=ij^kKmNym8u?Fm{a={=f2`|2lkwkiVE=I||8Xn-kGPfBDPO<4 zh_8JM60eNa;q=WunXNKi__6b%P5O@^%WGnm<%Eg7HOqfv(fs!Ovqe>@QcAiyB1|f^ zW3{mySWNQ!khP2d&yxPW8=nNJSiyMt^j|;uGtu{@N&w!xs`X(Xhh%QJW-7OL_0#!v zUcUE#LJE?%cY;kUzDj@&lOBQtqd7rA>Dr~*(mDH+D)^M_Pr{#ieXyta)%;-tF1PgP zZzGB->pQ>_r7q*Wok_mDD&`vb`8k#C#Zi=Sjc?0qf*Ub!{TspCmT#8&(Bo7L5B|K zXRB)23oOwzOt5+5g>*aSMk&}n#QG$op++whx;7+Y`{Sf1DjaHaCA{4HA(ztVUvw%7 z-rap@N_b0=H|jCeXn#*cnV!sjPF;^4;A;$6yvnWv^~u~{o1I6zJnCej2x>RY$v!E{ zk_4){a%*XR_x0zT<~}eCIy`nbcV8751{C5`#(!&;LFLw2(ULI)G%E}@@!#`s-l%I* z{AigVa5M=RAQd1Z>XPk^?PJY^i7k|XuA^7Rp>x5jOn+~>B!qYO<8@!;abIT4#(@-sUFRHR%$G}NPr61r=)g&5xox23 zeX^BnM?SmRRl~VQW!mM2GwE?DALSqVzB%%G#D&fdWd0Xj#{E7tyyf`H7g}!8CO!m> z@gZ9%?8bo?ao$K3wZZ0h%mEbO_q6|O_S%aR&`CM+RX8E7~pb% zO_-?0YgGJ*XyfujsgbqB=@KB%ZS`Aj2OW`s8S=~N(F?JeE8~m~G27#oxVOYHQ%cPR z*RdbW5gJ;7P~*Z3&#~kth^>?xsN&?QG@ohCMcG3zcafN0kuuf#?upYgpPdP5G@(z3 zc_+8zaJqoqngCfmyHQb+F~p}xop=u@9&8e5aF_cS2ku&>p`w_~N$<`8Ubm_=3VqC6 z4~rXdss~UWG{0?p^Y^g<vYjy}7&=ZZxmwEhBcX4$fGls|=>l=>$Ya(~e2w(6 zVgm0LziN0=XH^ANH7Q*rzrItDn+cl+qZxEa#*~3bHrRy83wK_<9I*|SF}g1xu`|Kg zm5(E#e^<{x#AG%udYiko`q<$7=%>Z`dE6@q`_OX{i~em%_zyw^h2k%R#3@(917*5B zimE)Pz18QFhz}2Z*<6dOB-a;ul<4f}NZk*DMY7JDSayU0B=hx0bf6 z2b-LpcO57~Ic_2s%@M!zunXQC6&<`%zlZW*%Z++q``PJVK*&&?uS7Q6(qz4c*EY>l`KNun(i8X_$E^bWADu@%@D1Cfe`5)C(p}oG2@Vwt*eZB~8kkSceMxC5 z=tiF|@QcmvMlpwRIvEOQ;Ond+)~k9%Sk*MQnai1Yb-qIHb)0GzlvpORu)X+ZV>pLf zWU^$ZZZJpG^SA^#eUNz7hyscRUWKmpWh>#Qb-RA{+G_6%Ho+yPPiX77a?E{OYu(3r zKA0(>KG*0HaaHdWmvw_SYwipnA?J)=VuMgEVW|aFD{r?{q9$~F z0rA^vY)%01sdIujLTFeF)qQgvq>uQ$>vs9fWma%Xh?2`c+gd(%5nejdir_#9i9VN$ zMNSF7a-HKgQue%B3L23=anvc0`d~F2{&aeMPp?e7bnRm04dh}!`(QuuyLKdpZpDX& zbl^f^Aa_04MaAv$xa+Ras*avQdjLrJdEM@t0RWEUML>=iWOlZD7up+*eI3%Q)*DAh z*|zj1%GilentID`(P!UIE;lv!{*h8%@cm~HiQB@jOo{%BjgaCi$3TiSk@&T~M5X>D zcDC)Y!jqf?(7yoeSNPigt?-6n{=vWd_#QJXX*Au$F+}Ct@R)ywS4nb!0kqL7&X%HcGY(m zZg%;kkbp8tnd)v(ZpGYGz?V3!^B3)%RZA+#wYhTSitM-tkn*qT2>M!Z$e(ex; z>vkm`FBfT6rC*Y6FZT=4X~jeKHfE^l=N$|$8GDr*8PysqG^f{a@Qo9@;0{?w2W_!U zM91e-3~N`-NBoCF_iNE39Th{7YeQTnG?R5jb!m?lR?5bFe{b=fPg(OHgI)&Rw667M zyg5D}$9KnbSd28e2{!RJmekAC;OXDUhG&mg(2LK{tKBYHFuV+S-H5!I*|Ebvk_>cR zt$YWXGB*6H%SZZ~V6YttLflH|BQbj|LTV^c{@p{VpB z^-?kc;9|(8R{-^4hj1eT;gkNa=FRHDK}A92k~hH(yVjEkzOYOdZ&xRm+;3OWEu$v_ z>`6DK3UjZOp-K@I9A&4Iu|~NzRJvmg%b+Hrlfg?zat~h`Y@@t3YmCN&AwQa}kNS`))6+-U4<9 z;jFsV1{^%mIp*KAjTWMwN8oA1&`27)@)7%#bXJ9$B2rN8sVyfNXIV>2${bew-tJ5K z~*Y&dRUH5)!KMhx(7 zq&dGL0hKsd*#UmOei>{6M_+M@Ot#F4#>Dbn_Y`wfV12IpDZpp`(;alukk;=RLp_y9 z*$Z%US#KLwIx*hrLj+uvDrfM^0aIyOI;ffbIE%nwJrrlV1yPSQ~5v7~Gef zuU-eQIknY)Ntdtc@Z1_};n3B&R{17w>0%$$nc@n-c z^D`Ax`fcemJ@Hi--efqAC#1 z-<@8=Q*@ikd%Uc&eY$1^Vd3AsDXv8FAb}S9caH{59z1SAI5C!*w26Iz95uSnE1~)9 zIIq9f)(a6)m>n2aJ%7V&LJwj?$o&gk*WMB+at%(wO1`W z;gQ#33OP+X1z~BkQCSk(+9|$GAvy%D@ZFDs4o_&ha5K|3=bEj`?v`AD)@o*#J2VG* zLn{tClzAl*NViKZkQP4y5i0LnJwYW*uxZW0^{W@ri$AA`l|Wk2g_~=7>zB9T<@LZH zsZTSH`O^u+TOD|S9W37muIWcfw&w3hRt5&No{O_8)p`!{Zi`EI@pP$m z)>WQ0&H!CrE4)853_K3}st$5F%^a>|GbsxVfLga>{8!1+nZ?ZUq{#4!aAn58ulH)= zM`ATUlbUx&o!({fyDgem%EEQWs@3B3Gs8HhPj+riQ>A6!>MQi=v+@ztM~-zL@ca-b zpor4tt(zOvGfkLui4R^Y>(C8gO$hh@yv6+AF3o!vW(pkhV1 zdfBQaW;$ut*|s65^q?GBBVmxZ@2Y+?`J_Q}HmVt>L-sq5?Utu>a1idh zalZls9?au*RFAgt2M&7ezZtGys8-LRtK<|$pKNQMhK~*3?0;Nms7l~o@E@pI?Olkn zQrQ)F>G6#~^Qedb{d5Y4^KT&^Nz<46NcB`^=QN4$z2DkYWzDE7K9>9I1<05KSn+th z*g{MabCEByS(R(buFiJHe2PBz(p-Z-=#tfDiWrZzDabnbf>po`oc#A+2>!N);eU6x zM4Z;hu%G&)oGdX;n-l#LVxSyWWb-uVqYmR+SZ0G|a&!>5M%B>8F-}#rhZ|_$LlZk5 z153v@iHpmA7BM6*cawqRnd9o);gPmmaY~7twPROopyxBK@7MUe!Si{8(D(V8>FnkN zNly5|T_M7^F@dcOOP|#(tDaVQeHWQ7WP-Yq75AsU>ejA51Ugr8o@l~)S1Tgd-Ezg@PMi5 zwy}m0kbCpg7WsT=rg55ICgU}%b&pNrF#0}o8pR{>1EOLic61s_beS^#@VYn;T4y&h z6IF>=mc^Bi+E9U{ggu>Z*pDI{Tg%f0Tu&z2hvk=7^)t%F(l!i(8rVJ+%VL~pmvAxh zqPd=BwNl(dQ(qW#?0~En^3eN5>*GEx{Y}@!ijQhP_b|}A?}Nr_cHj(Iy7H`k#QL~r z;C{;NL!~C~gG|jJ5ixaN1X!MZ47Rv5J8)5a|hs=6KQc$TBk$Is)aIwDnDW7+k z2rY%vsvl=lQYSZ|HF6QQ^e{}<_;BL~`>XVAl_JA*_=)qZe8!@GUEL2FZVz0z_k%`y zo{{*vTS#%S?50XknNpVQs3@+F$}{0cYIzvJTF%@2vnA9^Ke3){Z*9IaZHQ{ku? znF{nzeanMnYqZbjk-I^as`nmgeN>7tug@fkC{F%VZ-{`s+IvK|qQSxAV8a3jg_uOd z(j^K%HA`BsJ$~l0Jzn&Q1vWuKIDTytY>KL!GbhAzKvgIA3jVSYaf5O&p*7=9HpRQ%sJSU5Q}_&j5`MQbULT0P+#e<79I#hJn-~~SiRoYT0UEW~SXxQ~+_HwG!So*lR zCKKHTY@ixdygq^Gu3uRCzefpPYeTtm@YLXQQX zdT*2DM5xqI5dMU;f}*YtH0kP6PL%iucWF2md^)1ATHP}mnlc7q#v<=28aNAcey zuMX!wc<^ry!74%5vJo~DQXk>Zg{KSOCmJ^F4yeI>#UmyehmN{%6tl%$$q3(W^S@aRT^`_aRnTj$Qbj(ncw$npcWAw=O{u*)CSv985Q}m(E6Sg+)R(^WQi$I&%5h+b%rMpZA#QJ zS%BPIs9wx_VkCJ1ooHWx^{0cu;1<78=08jXcnW_yxjZ0BYgr}egQNby5;S4a;tAZ_ zk)thlXM=VCoQ?4TO}SAnMVxkY0^LP)&ST~#L0a#tCi91~%mCh*L?-_IadVcW{U+it zfi4PPpNmSeee+ZnzmvAoyuKSqi_-Sk^K~d&=f(a?;grBGJ$}tjO!8V~vj8M^9W~S3 zFg3*Q?EZ?)ob-FdAkQ#_&e_Y|8X2lu4H!X# zNx!+C;ZfU40V_S4%F=Cdr@_|#D==%{@r7cShVN0h64D$2WO-KDrI{mvL37&JG6McS z$yr5E*?>F{e3hl{;iDl9XE}0?ruA{Y#3r1CR3UjG6#}-<e@2Q-pNtl{6`&_>XEGIQ0zlSZ4*DE})qRhscRbyoyKDjhi#ydF4ZfrAhARMS ziaAJwJz0HOyhe~kut2w|FPXOgZS#B8nn!=l-3Zk9GV|VDrkxo!g5 z9^Nd`f^`ii+f2lST6@#gkUlSQBdxxfyjLmjP&py1onU&_I~d2N>*>_Ye<*XAG7Mbz zhlJP;e?qHh(8GOrW$7B5FNwgyznm~YVBPk?kHG3dPYIzdbWXFKudU^9f1J&6YuIq} zcW>vXVr!-FZLnG{1AK^P!l-Hp5+k$=L2Z8nEAuKxkTjIR!Au)c83HE zHyhG#m~^dL)y5j*9gH(-C#k$^4a0C=S31m@l3ukFE$b|okEhB{n<2&ffV{^h%S66m zM=i(Gc{p%?faK46zbpYwLbR}zYv5~+faXloU5E;UkA%-cg`OD&;oQ5 zG5CAv7o>|UK=DZ0^O1*id5OB?_{NzckUz2IV&H;)VM41zxS{ek6~VO89hIdh8boCh=WSv5 zl0nn!766sNL>Z2do$?V~yOvrB+b?lcyP9CPElh`lLO~kqE=eK04^|0gu9CQsQWl_sqQEh?Hvn zuJ0yK!Zip2TgB^h)k^y53=c5vtc-acEi^40iw*d{eLCjjBBzp6?R#bIwt)dg>`K5> z+BI!q@I~I``i)VU_}nbnq6HVLw*%{>l8Ia8DGr>}lSjz9z~U*q$A@e0d+Rf~*z9=C zFamiUt%W}*Yl?fhiRg0Pqb-~i>3E8K;pC?*86G|+_;8(QcKa}&Rka^GMD7OzF7b8+ z`4eOFvih*G)=uuqtRiV~7eg+jn^x$WoL0Aed3oB$s@>^duDOod#Kbf%F2%e1ykQYJ zPGF0Ap=?;DF3eHTy@`wnTtDnGR;VHA)PQa)9@)4YAwHz4(h!+Y^Z=gFb7HVm?df=Y z;iV)k(VxIj?J!_YE_?HvSm`ElunB!UiQQA~=?0fe)#jm4)p?Z=6YAf!bEu++7VA4O zapD?ov*s%kjsQzHvRuTD;q$>z6J+DnHLS4UT>2mL07D@>*D4ySCuUtASYV8zZ2f~P zywkeH_Z^X@Yh=hngGbgZ(>~k#7${nQS9?ZLsFGdflx=e`heZVOf$hMa2Z6W=3GW+3 zSdO*C&`c%bS>*zSF)?4g&I;TQ*x{dLLR;GeT z^8S9fTgU$5v#Iz*%sTTjmw>Kc9ra1KVyx&+dG%EF0iIK{_$H)ivyD%-Os{;!d4u_vc}w)`sk7`@VjFbUUj3M%i0l!{sK2E-3HqWz$F-@38iU7~6UbNHp;?w?1P> z;^)A66}j^tCj>o_g(!Kv@}z<`;9xnL3S8@iyM16?oJN=?^S6X?)0cZl#uV`^a!2TZ zNy9)DuT?}F_uCRBx52`PzADIB)3TK4^T7E)UVoKzHnD(gVY>Cr0fl^|&I)0wpg=xb zC7T$`3`5o-)|5zR-Ki~3h8WN;^ zDp{nO?*+t+l*7i7;g+ee9R$sFZf!W#vFN*WQ41MgG@-w$YF3N-{mQ!1q{%Vu!7&Ve z0acIA{lz~k#;rvhFA6FSbnz@~TDK|dKuV|P+gZV-Y)cvq-6qA9pB zCj6`!Pp*mzowXGo?I9Fw;?Z}{l-{G0d(YKGAZs2&Hv#=`w9&(duT#@%V{bEs^ezYk zrVOf?I0>!7#_CrlF+{Fm*Zuw3HaWn}nt(ym79`h1$h&Zv0}hhqwUQ%DWI^|Yj74 ziORRD7xY(&M*D%jUV7X|r%KpnM(r09cuq5yp@)WL#y+e~DO1s0*Tb!d6^Y>yaL}hy zI-V;60?9_8ey{p)ukhca6$RO^L_XV-O(FWeF9+yO4}m=xlJlmKV6#?<3hAG3M$Bd# zzI4>ccSPAw!(eijW5i)t83d}Y`zT_(QgPrY>Xgjh|7ZZD&-^Dje-aUWk31&G9p@+!tSL)8#1^q$X$d=~0an+wY7@u!rT-+OQ z+F*xyWZk`5%quX3jgGWSQ^!-8JyoiIf$b^(+G?FX#mhYDXLQuO04`d@BV+)^Oy`sf zD{PBMQx@(L)|_GRiS=400aC^0Ld&GU_|v!H0AsQPJYO66OkMhFGrf#?&936FF|b^Z z>;ZVI{Ng-6P)g0I2IEp?Eq8vV-ONT8GV~V!tqyrlkCyl3WZLew$z7d9n5Vu<{x!hN zKV3LkS)cJ~p?^``a1jiwlH`c1-(f7*?R|)zLT~Sn)j*u=+JGx_Tt=0Q10tEQv^=Vg z#HTEkR3%~J(>}m!+(8gbtT+n*YMrQDy^SfP5ds%f2DF}f*YnsXo3O8(48QG3ias?@osAYuj&v2uKYmuw9AYAp`D0uY!ER6`|#5VyWxeY+a?pSXvj?4i-PT}^y4| zkwoEu~WnwcwGF;q9dy^sD9?&v~X~RTFSOJStUe4*VBz{9{=62~y=#S#y0a zytHD<_<^V1c8lV9=o0%rML-lVphKg+KcUh5J!>wVY!)Vd3MZyxz4Uu3NS2vJkvgjLehQir4OS4vHY!Fw4Fm@jHG)bH5O-#BP*00mR%JxI*f!C+CCr*_8%vHa{rX_m`s z#*a*{OnY(x7fi-?o6i-+Cq+ajUX-Z2&R%Iwo=XwC`@pvylPbx!4J|E_eJd6LLgt9C zAAonoj`ly!aOe&+!6-$eY>#)?MU0Jpq{lg7CThkc44(8Neq`oqk{FPXO3}VKDAJLi zBH>{Mu`^}$y=qaTr==br!H5i!NbetAI@*%8$_?9i6HgAhd{!!!L!0{g)hB zvcojfN>xLNO5fJWw^t z&Gp*CiTe;s0lT!f7%9y9B6#QJ(ITGPbUz&uf753h)V7)NYE!3FcM{fJ0HfB%Bqy!+ zcG5f{i#K3djnsTgJ796v_3g=Y6-dy(psZ9QU)18 z{}I)>*;H8eM1COhpmHqNuF0Bc-=nZQ`eR5Vs?tY!YxF7Oq89OZZaq~PWUo}7LaDhW z?c1QcIrz!Mu-EERJ+|*(EOwu`NE8{B$~{y}YL)*eK}@1M+*9d@j$r$icz)Ol)Fh*y zD2TA{swo{op!-J5FgkaX~{yTZ@% zTLL#I>$GYViRE;hu3-Wu$supNANueC+C$4V1*ij!%eS)YvQ@JSUB}uapT=6R6E@fz(Y6m z#1450r&cU|dg6h#4uw`B@WK<5hU|0oKY?V%KeH4* zvoaf(3%(~iic8y+k-hd?wxiISDS3;t-NTubeMl--N5f&!ynnn{h~)=d7U-i zu_Fd}f^POWfw=J4Yza1C+Pj(Upb`CDhAK#u1SkmG&Il}nsGdG+cs!D;_lJrZ0gf*} z@C8L^PFOtm@lybbd!1HD@SffOqe-d6clNLJ6)JN#j<__wDZ;$8=lKRds-w&j1F5}p z#0ftS@lCG$NwjaecQBmw;$JxAHki1vyABB_!(<#VF`9og5O8iB-T_z=onH{q73}i9 zpy2e-i{fwy4iEqXk7xpn&nIm;UKgWHIDG#^B8nF6;z8({z=Q#E#&H*+4+!WjFY8zR zLE6ojK)x>5@Gg#@;O_nE%SjsKh3D9d+|8Lbu5rG*k7$r$o@*sZ7^+fW?Tha0h3|MV zi}{3l<1Lygp8pO0OnqR3z5T%4_qLY?Z)4K8IHxQ&yqiRWDm@$KY?gu?3{4m?E5WJk zr4@|1=-zr6dDK-g-O~SnE*x-%fNHK0- zmdKK?uB3aAES12afgM2qSKkaTJsIWGX`kP}n08xkuhC}Ztp05w7wCix*rb>LRB4yh zSTTZZxE0Ji?Gt{~{1cE;sjWrC0I>9t<}lm_%bJmmeIrLkqUy%;yok@ct}M0MogLWh z5_%e3n-!|~OWj@mibKtw+!}6P>!gma3;qiU2=cRTz+D4sobe)^qWYOU?(9WgP;|R- zEhg69R(B$vD0}q#YFo4T(?qGJ;oLoZ;1tmYF%Vkr*thcKeTg|2B%?lo-6m2$^V+3dwrzL2bC~3l;j(jwAYhKd z)99SWJ9&J5&i<=h1QBNnPh10l7VU}34eARiDw`?2ZKoVb17s`jjv8QO9WL3jscJ!P z&0FdrzDK$lo@hQMgaw006@r_rsDINa2&!AGlxPT`6`Clc-&ni};So`|rK-z*V}nnb zRFh4gIH+fG)_(rGW_1j_b8Fqk=4>TGcTEN#1W#DUcJd;}FIBg7`(nhxE{2^6~S2DpgC&(OkpdfEwBE(I!Vw`U5y{!uN2OKkp!Fq#GAcLtR%&$ zr@TcVL?@tnquF`~>v78R%4IE|nK=RqC5$o`Iq4~zp1=|+s#hPJ1tVnO)1}}vX0}n0 zs&Gyp)5PR#G?R$$@sJ(;9X^&yAeWbi4#X@I`Sh^TZUEj|7Gc&>_1MD6ocf)_MydBv z?xF8nmq~$8rIA|mVDQooY`i$W~zO!knV{9!0w;A7iSzWB1!oKPsfe2i!utOW0Wuv-) z{>}RY5;KsGBjqB~Iv9m-p(QQD?BeQ_NTKji2J!PCmmazD!bvMWhG z_Ibjox)rxqCKUZX;k(U0kFMLR z70X()a{~OPG8D+_)-_Q>{07L<37fMvBH}g9hL4 z{q1{+Qeza*5Cc3so0-cpb(vG*2%Dt4jwDNTZ)Y3P zdUoPLWV$*gjSBbEw+w8+A?*dXlcHSSlJ2x&Juen7hC>$|{f>=5CsPzFV zH`;t@b~6MGyE;mEPe~XY&rqH+cU134wtE?k?pXa{@x0}g37@n?ylTJ%5S1(T5_IoQRC?CZ|GK!H4lx(?2M+|08%++U% zh|s+H=v5-h<)Yu>>?7)rOsgwzXH!aUHN~g+!6WXwUmMKZqRjE&8l#rAgGW6=UJNJQ zZ^el7Jx`sLzfFqfImv|B^(b2&YawW+|AKE04P_dstX6tHV?LEe08Uw)o-2|pz9S#1 zmuI3T>*Heu3Qvnw5~eS)0c~ufO~^J_fqPyW1AmLv3k*j<7RtLy=oH-Hq}gXXx!#$^E_>glD!x7+!h0Jccuv4M`-JNl2ES)hmoG9EnCu4A zMp@4sP@9AwW(Giiuiln#x$E0g71>jj$hIIJ7LfOeWM^!cCCnW0QLDJ{{5KzfygUbV z%v7pZ>G|{W(Y`stL7+?{+caeDM^KlnBQB1_m^Sr96W?%qD5;^S?_q(m!PR?Cqw9^vYZ)+FaDhKrahV*CVW $v+SmSk%&vPpl+B=`J?|ZXPgwB#x{#xSB$=prtJ*=z@%secAtv#!d)Bx3oenu>^GE5B5?+%bF`+n>E)oON zsZiHlgk1XM_Ke#N@=wOU{f<9%=4@>%$cgTspsZXzY~7W?lK`RdgoA1@bA8b+JeH)* z6{zI$o%?2IU{dutCt3_FHu0As0x{euzvWufXS94!JTcAKlG{tJc}!-(sAMEIynbUO z+u=U#o|9*XV`1AL60I{NVm|6xb!n-oJ=))*q(J!*nA4?097^){^p(N`=lMq2*|>D; za1jE@X>zu_s%IY9RnVxP7N|hJH^QoE!7ttnelQW^G?mG~`6;-TznraYs}G>Fq5hYhCgX_@cM(Y*68@5zy|{_W!Z>mQisv>(*!>kU-Eta0n#02M_M< z!KKmQ?ry5`;6$$5-rA{eWX9ljD<_CQam2Z+#;$o z>Sg#h8W;e?6p&u3W{?!txU5@_Bnd0e950$J67jfV4&^VNlOVT`gd zYEM@Ugf8T)U+-T8Z4NUs@=%xi@+fYm^IM zvv*t~KU$5;P0l;{c-A$rad=pTq3ws-? zE=wxXb$xmb@&3cHhFNg#>T-}JgeMoJpf=t(HpqMNPc48a?boo9iX;lJ&wf>vJ>uQ|<%V4TNc%)i+``mqWTaJn#${1? zdmx6p7O#aV(v&JH=x^l$Y>kj+RT`DwmD(m)vpDrJ&(&Gw2V%`QtAWYXBA>xVy%#8_ z{dClod(ex832AW5Nk#nIE%T4cDhUQ$9>Yx=Tx~ozXlylpiFMHEWrHJLCWBUx8XS6v zGRzlHl)DLoPAh^lh({~p+uwHZA4DhHB>1K^1{)U|?8f1MhJA@{1JS0UL~-ET-b?xF zsfjE_NRQ)v0VM<^#r%P6fwmkBimO>wk)zoUfTRf5z_0^aTLk!z#D4zbk$r7RHXj>V zTYeT!(x~+S9Kk(AfMDr<6-)c~W#hl1WBsUb01<^Pj_*f*pIZ8BUj@bi;f1*KBWksP z|2!Z6WPd3#0I#RthGgmYRQv}()c@_EUk)YEW^YNU6yZ;sz`$i`1G5~vBG3RViN6cB z{`#q;JkVymF(24Bf7%2=r2biZDOIbVBk`a2Ejk@&^Ir_oKbryIUx(NOYf#bq`5xb& z_l=AfXd?+tw-^26;gFzj0Yu-d`xoP0fB1HQzDcI`ZhigZ;RulQ!y`D0-k{{% z{dwQMpTi+IlQ(nVZ6)!&mP{@Qbr|J^|UyMcZitpD9WB)$Uw z|K30$vwAZ>C-S$h`(o_Jo6%OGtSQecK%aDPyRG@&V~NDt+K^~AJ3H}O=2;Q1(MmI9v!m#$=6sA zF6Z@td#B?cPHf3P8fm=TB~8X1e#%BO0Cge6R;f#ozwPJ09N1*^wnvMcoK|8lc=W{( zdED>oScLR02b0K!8ypo~4qszJ?1LLvN8`G4wh$adm(E0j|JVT5L16JuZ+pMt7r-!D zjo}`B4be+LZLNcrwbF^u`#4{ysA){x|3iS`pQrkNm=`kT)zPtR`%$>2>aF2)-ohZG zZvg3<_$BHuASf=*#Hj*#ISE0Z-~G8+e+J73>qE@-S>;G7fg*f!FnP)ZL^h&*`Cp6S zs`V*;voBh5Ih7O7n<+X2`}Xf$%D)UlAQo}A{9EheNpLE+kB+_30EOA7{D`l#b&&Q_ zA5`M_&fW;gJY+mF={y-lKQYQbo>Ku5cepI3Y{xl&(B0#$wA|BOnYQ7$tywl13Jz~1 z8SmdG75?ST#~3TOKL9_#x{^GC%w@xF0y+tokB^Ni+jLKCGarx*m3f#QG5*va@!Q21 zUBkUPT4C&qA&;?A7fT6>mo0yPiex?J|iWRi5Z}A)cGF^XlKmXk(!cqG@K(A~Ry7YwJ?SJR&zkbS43b>}Ow_*h9JdBY;f3UhsKlq*Hqy4&>c=~nY zgEnweMJY0zhRx}7^h#;f{&k%YTr@e3WV&W_t^{S@C!^@kKdTa-Ds|rB5PK!yS#grn zIE${Zh(-Rv7rmEVg-~<4tmB-mv>*wVN*oYB2tMJlj{aEy;iUYgfPk>*rz&3``c~U8Z5fNY8K)ydUziTR>Eg$7M*Gb z+tpFPl?~k9w8XN~lj}|TgB4h306&~@D?R*n{=NOO)c#jrv5qKx|C30`?`W;c6o|=RV z6M-m!y37dU(&WwA4!K6<=Y*pGoHZHAHGBqGq~LEtex^S0+09Ri*S|)$q)BX{3bd8p92#Qs%7M=2Q zp&j-&Ht&(|II>vVX-(6n#Hs4@NIcsl2(x*K*yo!xjH#j6A=KoK~KrWPOXPB9Kx>Y>sR_yjz;9 zZf|yx+@JiJb|cLhg&U^#z~i(ZOQ(72SECUpg_Wb%e&6m~FyBb3 zUKQ0`7DPEeUYW+zC2jD?{K%-cJKEHI zJ;Sm-A0SC;dkNumGg&j}eXmxU`nyIt>`8^W0EWieu%}>5Kh>D4&`P&7xj{4#XYR&1 zf0$&QVK$eHjPz>qzh7!}IxjY2PxQSyW0}2dcx(~7>N#L(^BM#|qWu{@4_5T!J0ckg zUDaN#P!4|GmYAWG)|gK52q{2iXAtU{n`0VvfbpOaljuzz=!KIw(!sJ_rLX|7(&CEB zhTnh(UY&bHJPwamU6J&q-k2wGnH9gG`p%eh_E>FEYq@wXE3x3tF}`ATUJqW#D{F~L z<3Ei5&E#*~p9bSl*3U>>7b6E>pz&KL*wk9qcr(;P4Wk?NwbB2_QtQ%Mh4`V^KdyA1 zEHlocYSnh#r@8ix=lWX>-@3lq<-cF7^KK^OwHGN|a*WB5jG;&u*E^HIzBcPcy*UXc z&e06VkLHw)mpCe%s{Qpg5_&h(~Qx$U!m=a#R6P3; z-G8XYSXb9P{DX!gO`b;&@b?{e7dyGSa5752H@Vq!2U{$G!|0xDUiRJ$h67@|#hr zM5$-L%VXQkOr1$+>~^>3W2YsIsa{-|Fb@54&sx%6;QC4%~lTr<}d9W-MFvI6ZxULe9N)si5D9 zVS3Cryt1z~e4`_a>I8Qh=)ej!nV@wa>x;?HkqG@7;TMFj35R2zg3MMN{ltoVNfiV zCsqYbirSxN!~=YK?z7M|gmaJKY%4CVr*h%WL3%Bw7r|)o1^hRIbRC3ZH`t07A?*YspP zn$kw6bwPBZgK?%Lrh-&V9b`RPSHW}ri*#)A;p3e<5ES4omSv=V@X;-m751ZF0Eau= z<&uOI7JR5zflP!ZBCIa^+}-3k5%|Z3pPmja&g41 zu9LyCg1>FMG{42?S19n~R<=^ zgAPqv!{zxN2f!sqRT{HuUHnZw(B*n-KCdpuM?9a9kMBNN9u3a)zmqD$9`Zy>^{uaK z2|=Ez@?-+g+*1u&s6>30UV@Jz3B86OuaDm8U0l25I=1Uwl1kmSzPoo)S`Ch7CHtJ3 zB?rDg<*qxyhpKvNGgu7D3OY&5I$yD9v#R8f&RJ&WsMSt`?%-A%uh-D)i1m*~888R$B}5$Zp(N<^hweY~I4 z(=D9*2@BPmw!Srp%aOwJj=165HFaoNM-?cn21hX1tP>CS+K zcl?|3+ZGpSS(eIX^&S)!#8nfQpAJb~ePuZ9SK7w^g6Ax?>9%#I`$}GKd^ZS~2XWmN<48zneVWi4!eH%#VCkH5mH-4xlIJe86gKSw60) z;vv*0-O+o^^BOTDZmpbR7ONc%4iK^@t zZI{^0evO#(C70i0nrRrjRn8Ib72G`C`ixhLbBPSDSEla(FABcISv);y%Sfi<@q$fD z)Y=HohGn782%p=w%vt22;Zs^}^Oy&;#iB)vTTiUL@BIZndo_+$*{w^hgUJXuuZ&Sq zh(@hszv|I`2yXw(4+!W=4;QrbyE|q;l%3+?4cbfJuH2N2%icwQS%*CdlI}uo(^k5Mf!_m$=k#`m2k zbh?9jYI~<_#r?LoRWd$#Hx{7}ZtKv*1bdCE^arD5Eu<19zT02^;~RZ)d{>}O5dil@ zDI6SG4J(L$t=V|%m9axTJF=)ZgSm@`2!_U0FB3s3eVl(Q>0q}wtvw;YRk@zznlT$p zx@Kem%;RLYjqVx&{t{)MKw#`O|fohyo* z0R~6C{Z9No#|p$zxQPF7p;n&V=ORueRZxx9<6NJ#)oT)ktwBu1%@R zz-?m()vog>Lko}o7tbd1xXoWlN{_c^@?-BP7n~qewaa$msds!>QvAJf8X@SlC zD>?2%4bzO=7Y9>uqSAcjV9Hsnl*2>#uw1PxzUi+&V|x0RT)H2^f~*~XbM27>wpeN`Bs`s!a zzdqz_ylyQ}pp-!ZG6IyOzje7ODuMy}2e9_`$&0lgOW)AqUneL%@Dp+@xz}FY0E$$5 zjI&rqbZW{jOBope09B>ZlC4N6gHN7Op3n&1zi` zz?MvwPW?<`lmLnuz9WqP@Ks zkfqbxgt;5Zw_W&lNKO(-X19oGfpvxGeACK7_dk*L)_ELrkD`g-&xR zph4RN?pc-0Pvm=ML1$;+W{=BXEXFr>XD->Ai2HFfjLMr=*1h*zuSb2JJ=Oi>abE92 z99tvjNZAK9Iv>SiR5kpQLyM9cgX=tuCO|Ost4nRAg=0}ej}X19#@x9#y7QA3-3INwnj4K&6SG` zRIpC1i_ZLnUiduTB(OoUt<{zYNoDb$$gj_(k1q&Xqtm!uQqadVSgK{H4>YTz6@M_+ zRcEw5g5{Zw2dOkAMN?FP(9H8BZ(tyS9zR<4(z8Z%qIXK}NYpj#p( ze2c77@@Xu8U*X4D(14IXt7hD0g06QA9YADCzgyp)o7w5toj4p&DKsm==WwrE1n>{7 zL-!05KU8n}Yt^yF_NVJXdGc?eF5DiO9w7sLjN#ZQ5v~gWT%uTi^1Qj=9cD>`Qub|l zG_{f#G!Nk6>54RDs6neBko%RcOCtWhczUfZ01>zCrLGM`Pt=+f5?X+U0MuueqS5)V zVmWwT7bWAC;H%0?B*9Qs*5Qd9FCBl4>3mLR4IQAu_d77N;F`>!Y_Jnb7rfi)n^H50 z2E<}CcP5{NkEdo<`r}Jh!cfgrd(K7uqY%w*9oGDeOG-dbSN&x54`8H`qfNsc;watHM(}qh{ANZ?HiZ<` zk{3#)^xD5Pi>-CN>-%06zoEA;xqz70vDtr8>LflA? ze`U?AxmWk9`vu^S#!nj%!73J=nE{|O*k;D1kw#U}IP=)NBUSTLl?#O}mx>Lc4_&5@ zk$jkf#3yI!EIQF#A?LY)WOC(Kb650Q(~vt1t5>Bco;}UbIRHxGa0FbncfY(-tda^yqwQS{~YY*f(7(d5=E z<`p|50T|A_Iw!C9CaS^?rdatpO3iEV`;u!8F1ROyp+iV|dk-JX`eiO2!OB-fZdDY4 zsTr*c_x_OCdg;ibU{;(Vd&;{BZ_vTE#%cQb`u2!6i_z^v{2I^XN2zjpay%`0T zgqqb}7B%XslLdgTI|pi4ycmxWa3-|fa_4a#oMd?-Lt#jNpjdC7s7U|5mcZ zdF_>vM5FlDTsDsp5wE*4v(a!d=Tb78MSDKYJNXLO-bVs@K&(X*X{R-14_ ze!k?LLTKzu@CG1h{Dsq@+J*5$nNuIPTEv1)g02g6`ifO!D{zZZT|tWXzMQXCWB=KO zuI0i~H3$rugV&-k@G{!DJUpF<IMs)?slga{0bg0R^x9h*Pi3BG`dO)Ort#?Kr1F49zvVKVu6e!SJ(v5#(`j{7H@eEz`+IRnO{F9owr{-a?38kUEWt zo`VMuUC4E^j$`dWkI+SVyO*}zS0a<)jn1~%HJ56fyjJxSKc0mjN!Yx~t~teene!tb zG83pkiuomKwRzYYdN7#Vl#q@A%qHX^t>cJvxWmed4@h>2^n>LaLvCd1dVUoaSKgJj zV;=5dv`Xv|=ziyl3!pgqvocy$Oq*IW%Rms-Mt4No1zQEmMxK*yz#Ba_vw5+2pVK6{ zv%~q3=M)R1`=X|c*VN2TJ%op21CwatA*4mB-a@w*-n_d)p5=iOn@Bs$myidL&f#Jw z)))yf@`?`bXl)8MY&RL?_CBoa^Ktpumq5=l{nypvwQ5^yFueiZnu-l~Y?NA=4~r3HyHXW#=#2^{TQkw`T@KgDB`?1pdA@m){UI+u*S; z2x=8N{0ejhWbrL4c)U($e=v4#yY;8j7ZvN&4}K1HrYqwSw^{;&3YAOAvV+l(LpUyM zZndB99UIUt5x1yiDH)=hoI|E1@j1S^i|PyiY>!D9s&E z6~UzHmY!9q%Je>Z2dh)HzfmqK=B+)tTz0ifsRg!_z z=6k2-eMZa32jzq|HHv7q&Pw8hv-iZOoYY(+G^%=EW}bh7232j-A-h65wdXX$hYsgi zBiU7TjxGbK&<#5$&8lhyMz@^|w`JxhG(Ws+#q;^#oz=WRbEI3y=rOy2w4JE4pN66H z@qrElN}k!Hc>WNTO^L9rvv7PjE^&O|6?LRDFz^#$`4(U5Um&X{Sz` z%;i$b*Xx5Gj-t2@_VT%33tEGNWuL6Kt1>=qCDA<%Mz@?S*Vj}wGU>uWu5i*jn=bK6 zvzXm$-VvUBuu|x#vxtic4V}mzVC>gY<8VUzfB-I}QUzZqFVB>N^gk5oG=k}>kET-c zJXTtwIaXS^le=@5WEV2g;=DFr?`6(!bcJFb<+f$7W6DcBP{=N}P6QXSh`5YQ7XLg; zsjSE~E%&Za2%THjYEnOj*IX!7I&hIq=17Rn&GHyxHtNZ1ah_XCr<^{-tx#3WmNRUQ zuhAQhwY3r75!BfX4WpuiAaygn3j|fbCE)t? zrMhe?Jon)xiiaLGohF>V^6&m^i0&f$Nr-s?(oI=5+8fcC|eD4^l2h(S9Vzf#>ayO@w_z2m!s{XJxRAI9RpI+177?433!>Hy(@3hZ0 z)!oc%Q+V2S?5~3>?3+2`a(b=AR_*czGoJU(Dq7Vo*Z%|SQp!+P|jgEb)JWRc~$%= zg;AK;>e@@Bxnu8_&tnpF4@-5ib@pae!%}3tY=N?W#a;0_i_RM)8XEzAo29$~LG@(T zpKlu(~GuaR{^I3CO_tSz015bDiRZx@UH<*M)7%yELD7<)tbRl6*8B&508b$y-8< zB!0Km&-X^5Ntc)@)OawNvCJv_cI$;cx^SM-a^@%>`gPkXr*(w6IFBw^wD8H`o9_&N zh^uY*pgBZjMdP9S)0O=;pC+pnSy3zpYfj)b(sJ4*6{8DS!fboeZmfx;0sba+wu1qI z+M!_QcrIwj$(*0jBP3wvgDX_{G2&47`{~ywO!=Adn{^rpvrg*=@eT*9CY6WN0c&zk zzOsS9onja4;cx6u1Y(&1AlK61NJ2{Ssah=(k>v)bU|MuvoYBZwQ~WFy1gX`~mm6cX zDeR4B1lxOa6O=DPN)@TkcmgrVNI)Nf9(=ii@)vZMTJO^eRkN)1eyh#?hUV4rY4+)2gHPWt)1!o%b3@-}ePVSkgOW`@)T&-;dHn?N5qAab|Xlw|>O^Xib@ z@jAsFwLLLjaR}}BeHguYcsBs^Hwth_WwD7$e%`p3Guwh;NP3i8FmHh4V4_lEbVOC8 z)879L>LF)<>8zpv9Y)8#8H#HM7c`|FB4s-8vRQWyU?V?VKdyIgmb){uL6vdUR&;r| z=P%EV?~7G0s4uo+(kgE}o;ZzfH<`w^H76dUTp3RY#1Bp0&U8WvT(5#~F`>_Aw=Y6_ z-^Q(F_bAR9;RlFgkih)+r>-AS`{QlN#~s*K5^Q2J^wB0>M%$}ru3=x=V-jXCI)0n@ z(kWdp^6VfnFa2trjh;_?d-+TI55+w67tlGSG9N+hSQa&g=AKH!J7-QO=6$66NR>h| zl#EEt=d5@>a}~lQvwoe3xLpc!4x8V3`wicKVd?I6vw=bf@j=}HHN;&Yv}zR|DW1V= zaz(dSHWZ_}ZDQ&`iWMGefNsB(&XWK6SEAy~&8pvXF8R{M`rKd)tu-5y9+Gbaqm8vX zl)@i*zPD@5Y|Ex>&gL=_#WN_0Y1UYnR0cQ&d`DgQAha3eUhA<(?iH>o^~OnN3M;)W zDB>-V!t^F0W_54*((z#Kv`@ub#4@!HorFG(OD#ZDAYIl{?lU3FvF8wy@y(^U4&3-! zDy~dj5nRyn+`4wzIpv5p7>#Irrrk$2R^Vu?H@oExe&O&|?8lLiN13_y01$E0S;Sy6 z`w%_xa^L94X#9uj9a>PMl%G)GBI`${--aFrKH%n^Mj>+a`_5KW_m08%)xN=m6uWl_ z)g7;01QA{+CRB7$9Z4Z`k#|xSXKQ|L%>v&=@)O=`OQ!I;WX4=vdYnh@>j@TRyP&}2Flet3;ke34a(WUJt?)3Is)Izm&yX~Q-P`==w} z{*M4O{oca`m$v$Y^mQ)Ah_IELHpOM!#d}@P`uY0Cw&!w3jbRO2WuV%z@ZL}i;@vur zS8s=e3sS+dAFSEL7y zUZiF384fwCl^rUkh$^Q8XO3`;j~~mN_x85cmb(@|I$$8a$fIlB$j>NoJSaChUmnT#3d!$g*q?mRxo}tXAk7; zG%rK@w%?L=8;pLBy5-s}+MtlJi%sll?jgRrZ-aSpH+~1_!^bK5mtLA~?;SQ0yzYMzn8@|> zoW@2#kL}ekb0Ra3P%gahe-);|gDHvJoE>W<$K6uY3@M=~Xa9!RV<)OFkj< zS^E@mDwB72&%ZO$=p?|;a4Vnz>Ro`+ZMya%ap;G{sOcPJo^9|+^C&CwBVDoyeRj&e z|L9wV@pkW&DZYr_WQpf8RF_x~f6H#Dy~^*l#;qbI8|-9XWl*YMKveq}lRxaZF({dH zYjdf6ljaF8y=&5aPU?W-z zsO*r9a;K>%#QW&j+q5>E zSIqz!yke;ExMiu!5@CRx!s*eH-6eNKPczn#-Ey3%krkO#(Mk4U11T;p1^}8!maFEp3oATgG%cL6k-&+iNgoyyouv6 zEhp^+iYsm(Yqy1P?#F>Uxe@H#CC?QfuKsXt@CIn-hSGkw4xfC2Gh+mLlzQV;AJKa< zG?QU>=GXm|6SUM+;8OAl&+)!ojRrn|fSp**Yi{UiW5=e!%+`7qkZPbd{kSgfTnZsv z2G53z*ZV+0`%T(q<-;r0O3t7TqBxvAnp!$8QRS88i6A<_o-~`E z8HR4NoS1TVni*tpjjeCYA5pDT6GeC_w{jWSiE*bK@IAR+n$Do8m^512v|gVc>Y(1@ z^&gGb)t%kBR4rB3mZb0g;C-!DW*P5$SuyZU+6!f*Ji^>l@*DekILA_J?z&B3=Jc_m zPghE7^LogA=op5-95uW4kJ+ZIcwApWIvfm zxv$+|y`XO82|qRO8ZMGh4OYRX-;Q~z2CfianQ6avn-01-*J7}hGoCb`wzc@40|HSw zu78vIXgN-xtije;XEO&~;+!!fpWA_Hub^*`eG@i$;?K;1;PPP$XFyRbrJ9N9!`^+w zWk>btkIa~#R_%^ylkDl5)muYFO}S?JNOgi#@cszM87CW9&eZZM){729^X_3285PG! zcJQadBso0na5-UcNObqfToD-`*?xGQ3A&j(baTeh%&@oLn%4f*^?>_~aGLq_W32&I zUAZ?)|AW%P$sTJzi}W)egx%voR-38zD-E%X4p`zNOr#(k7Nn@m0K#RDsthF?KBdRd z7|ILSmbIcL{zNLT0AF}a0yu+yBr$>1n`cbVPB%1=DW*ds%W22G%AvmcdGj#h9`B~O z<6n@S-}pYN-OcT=z3e<)?m*+}CylPUULuF-Aj)|E<*Qjm;==XKbb~g+yC&H8jdZPi zU+PJq%~uD^$ArSa-VCA#HwY){^Y=zUT3x-H9KvfEw)qrr5D@+GMDl{F=szLFPZJm; zHO7kL6P19or?&JfJfSwN9#y*|VwP}YSe6e#veraRhZyFA6&cbYMPCpnE}{IZuiQvJ zUnc8sdX*j?sJdop`cieqGTDtg)-JaP^hxCNeBNV4e{voAw$~1a=sa&5079p5U13ys z(StwHf|lUJyzA9;wj6AYfBZFk?u@dBaTDzVvpC$qfj1{SvK;B=gyFM@lAgZ_;=@jO zlpztq#Yt+<=YGs$hwpK|K-4F}m9p>e zS`1a&L=CTfA794TnII%tR;^Tw$=?t1U{ZB*14T!s<}-2%eVilG)cWze#z5XjGRM1Q zM(R~O1mo-GXHe~H-d~UJpN%JEJfq&=dwBr9RM&b!sn$evsc$!ehAlEVM2H7Y-`l8%d=cCE!u+EAd3)T=+|`4& z$wY5wn%^O9os*TymfTANGKZ3S(v*WblQ@#?w#c(3Tg<14{t9cIkX)JTd2;{7W9o;4 z`ZbETGp0)PWhvYCuNt{%*CM?l^W;+0zBwM8T{J#qP2qKvy@*`Qny4;L<^J()TYhVD zt&UPDoily*!mn!MpfVvX2}jj3W23PXHzn9uzVN$-gQ1++Y?cmvNi<8&R-L+_043}Ar^y-xdh>8KP$5C;479(EW z5gO^eo+4elp*Bp;DpbbA2kKb6^uNsw{ckE>JS+_w#=*%|B7%>c!X-|P!9qRgsV$s_Tn*G zVC|ci{Q{M>}Ndim}Iki9j~47qEge) zb`rUd2Y;l48ZJ)64n~eCOpiztYk&PnVt}7N%cmxtJZbau&-{u+%h1xnpbaOa<$BL- z2}lk);Po)rcqH`;mYa?X7eGvKd--%u+^X$6h9d1{R|3V^t8_0RQwjeFC91)GM27=h}(qsNdT?R<;Nt1}_iy^#ANShha zQUfAv()-GRu#tLt2y+K^D`}S&IDtkRN2lfc+qcd4`zHn^9!~fn9Wu#o32^$q?iTVi ztm{I1?M~|0OJ{#?LF|?}8dI-jK!0o|-}@%zxrml?8s&N6*0kFb+S#Ig1Bb^bN(S=} zZ-v*XAi0k_=*xJF$`AALE@gpD=%Kv%JUV_hF};=b%cBr`;s0mfc>((lvypFm@zV2! zjV1+QuY!|U6ZHbNvtUjz`}451GCnMYbP#wiA00HmmN!^8Im1Aide+6$&_i}4C6&cs zD`(zO`nkYS-GiM%A*-v>C0Elk_JyqGUZbbQB7$47f!kFCqjnvY=vRI^AT<%!FVD+h z6F~^+>2|c&IW#H4Mi6!L{VpghzP~xbR!Yg+*t~D-<1{G zobP)y`PYH_xfp-)*IQO1&QA^E<}<}K!Y?kJCko(c6=7K)+^$ExKRqP*Txo$f+^<|^ z@i{!FhfFe>?jFTYg%%Q=MrU*xVk~QorgowX^mJ)w>orZIZ^O0r3$}<`=x>f%q+{#k zzb@;l9yWSt{xe7Zs}PKbK@J-;ju!u&lZA4>Zp@IP-?yFg{AN2`S4ymgxPqzX2X#Mp z+@fw%;=Hs+A!3Ac*M_^x<4X~GR>|C6+F#d9osP*86(24NfKm(c?gIV%g>2i17WgM` zc*YoU0=)*+?*!w>;cv^)M=-4D28+41C2};J0*D-&Dhk8{?oi(W@9jZ3qpBxKz5ViA zbm$cf1eNewY1-;qRmXkH=|=i6=^I!iWj+s^MAXgtyCz#<|ES>Fg;7lbgOn{r?psfR z?P1RW28LD71MuOe76TO`>gLbH?o)B-1>6Uqs~zh?`r!){aVn$^A0Q2A~lAX_D-=-O-RD$zUV%n zSmoFZ%k0W`eersWO$LB=$zMsH%km31pW#sFE)@12q;1|%Ts%ZKtW_Mgjwri*5^G^< z8zP2uE%t6i^1jFHMMaAjdAvXSUO>O)NH{xLtV3?vpyy{YoQk!DC@QW$+`0I&NVO`H zDyrhQD$ePU+Q4t2KrX^3YBc?mB)chWI9zC9_I+Lfyl0=}Bxud#!0Lr^q*0>(8FCf0PsvRQ7^MbW%;a`r5&s=qPCs z^((0y?CX6ya0NF7dC%l|P2RQ5Pi2@=P=vdsSDFma?B;t1F~J*3m)HNABz9oz+0=`t zwNX^eGTWPdgg0AZH!rQ4I9%~j_!>u*?=OaYz&KUNXCxlXQ@bNbDzE8 zM@g>kkyDI9-8ZY5J~1uvkY$7D8EToFsF#HF`**{DCUotZA`5pi-kCDD@L z+B>G}B(njKv6gv7Gb`%auVnz^6OmE~2u}2~E z>1Md*z#`Zv%+cqNjp)&HApTQ}&}6c|ZXGpR?`z?pUmQiMk^pvQtA|blLUnb2zBgeH z*;Ra~ANe5|eK1^5gBTm>{L%*Zhy$oy*zbT`Bs1TO=IE_;BnL(ypS8N+&_~y^81GIp zHeF}Mx?ivPrcNxoOUE;4WOZ^m9nu7T6)Txb6e**zZatdMJzz3*UFCW`51>@Cr;TY; z3(3OMeU?8@^wxp7nH#Ot3S+DHaC4)|k!lx8uiE^gx#iz-a4Dxf?+4#@Lt7sx7;DV3 z2MR!C>6#9O+wmqD-*9=8JN2?5252av3~g{c+(KG@$qyD?cDxAFVa-_aE7uqDCc#$KQrdf^~E~=9_KQA?@8h-0JSn%%j zJSK9_fnOg7j=*J;x|s4b=)&S4&>*dShaUl+1eZ!P=SS1sC0!Z~1|7sbN7ln%8{kmh zObRqt75b<{g2Uf3{Q}K`aR`KKe%iDHccw9|e>Do*D&={4#Y}SSJDRQM=V8>>=-?Ih z0*Imx`^nJ(xKxLa_=4m^oo3nQ8Aj z5?hanZVXqv(})a`=O%g&b|zy@ki1D8L5<$o5nreb>M>~619@VTy3NaG^W~cJPPYon zE^3&KC5Q_VcxP6(mgEu8_Z|s~nZ2*{^HsDg%wFSn+{4K@yWcI!&*nD4h0b3pP^K{% z700T)he@1ARYEi#=zD#EB>!MyA@#(! z?2buF>yu=;p8r8ejP()JcyCvO>0?*z;57563IA9a(G=J7^xy@n;gRBAoH$@J_PkpL;`AeB*yQAdE*>gegB*B=?Oh9=<5mChqud18 zQWrq&)J!{H4p{KS#&qgGX7ss=DxD7(sgsyZ;>R#4UkqKIMe<8Al1qFcf1$6Ikf0l{ z>kL8AoYC_ubAmFUcj5Vmn1eA49_rkxuyd1a2`BR98HcoLBPd!bCtOE#&1U2-kLI%d zl?qx*Ji6_!xe2;#%A*BEBr`=m0oAP&GndJ3EN6{%+5m#E>hgmH6oStR8GH?{HvG&p zU3RydeWzo0(1Ra#z?HRJQxxUblWI8VM?Tc!68(dUK09a}191{NST1e{-`|LU3RnU` z%kg1r4hFE(RU;%xAI%zcVuQuWNiCUR(+OBiqMcSe@g2urBaP7O?2_DF5{EeVs?!Da z1{o3i&;W9HlRDDt2NeU@7R{}$;5Q7)B=b(PzVj{eh0`IY2xqz&&Z1xa);3+vF`eha zF1_5Hqr_0MbhKAxU+LsBQ|b4oh0Y()_(er#jU15jAFNxezPFU=b(G>&ZA%kx zdBHDtS#}y2lxlP4>OFdhSUvxl{hk`~^^h0tz`tNJ4%->ec=crkDxA&-*9?CgE<01Ieu_)tOfD@8ta#Z!7}fp7wW1>}KCm$TM%tJYWvGpnSz-28<2jQnT<- zg=2e{`|HqWBmU?H>{d5fE%yLiF9G)EY|r6U<2tm0mI-v5peX_t~aAGsZ!S4|W zUQ)>$UGw&VXNQ@<<+PbA(gtU#KdnS+W{h4nLZ-L=9fU)>8IFPQK{eGuaXE2a>Rl{l z>1-?>br}d;UL;XraB82xV9i$k_T|Z_jmOpghlS^Hl9Q#PNSv;#zB~$*Ig=KPvv7BB z-)(*gC*u4B;8K#XNo5t+neGoqEs9$2KZ&6b0O!mXLQ*hwCY-dDzpK}md>Yr{FQWa1 zGuLC;TW46bWQKV%{_5;S7zJJIp4`o?Y|aQ7?PT;Z!(Jy`{+^sSTj9NJ@1R3%019#{w~6K4en%jZ^TE}}L-iuip!<~t%>NLIQn?j9sKDsucG#XX!2OP;fqwRus0%gnFE3M8>Ik-mz}1vGqT z#!xo82Ech$;#{G-JqLD|SkoX=S$t+4kBjJOy|QUZLp-I&JBg(`S>sJLJ%AqkR>a!a ze9y;d9^d+d*ZDibm~W4>ExBCRjmYQ`M#3jSdalO7r!=()A*GRxeX_nE!0)2Ox6P zxr~H*x#!$}j=H8#9*7iM)_Ny(1Yl3@3eIhZv2-sL7Y^IC5@rudpqe{Ma}n|S3Zkj2 zd&?BisYK6J*}WzxZg>3hFS{jR8~HVAWT?XM^UB!Gwq)wxbsdQ>dz{{fruvxBp+**1 z9Nt~#oUx7GP`Y%Q%BM3O>z&QNj?^1jtY6~1|5$<+?sdAG$DJQAAAYiG|6I!0BR&VT zOcNJdBaeYo+5TW$>wd(=<-QSwp}L^ZV&4y+^2L5vAP8q6zvMkKF*k)+2zGotlhrJ4 zp-i+uOdxACRf^HszP-fRzC%FtTZQb?;Krz^eE1j9NnrRio6`Klr*&j#fv?>IYQ<|G zn;(h9-4MrBG;gQPA8`!;mxc8c`?NEWHQpv6{Uv$cd)wc14b7n^x$G zlaPhVYYqj3IllR}6+bHvJ{(NW; zB%U?nZkb>w(RNF>D=Xne3?MY6);?RcuMO3^e%N{sSV*n!qy4`!)@5Z~Acz7g`Q)5+ z>&A6makpsugQp6o{j1MaS<5Lu;~Z}KB0twfAq(WL#|W~JEZz&%4?5a!GGgU^Wg9QBv@{N^cyn^VOLqP;uXE?d$c3dRQ!$NLz6LwOLKQVw@_Ew$H39`DQf6k+LUe;ZB{yuh!6w+xJ(5w)=HJF|V~N zC-v1sK~vn9Xv$@?A=d}b5a)6_X*aTob@(2fMF$xZfwDd$gvTFMyW8|?i9f*+<^49T zD~Aup@g8BechqRTE<^UrIBOrf^YrR??~rh|JgW}SSJJzWnd~Yyhsq}DGi(J5BHc$d z9!8+8y@bQyA+RG>{mhPmOg*Z|K>p6$NbIsHPkGgGyB~;~QOm{AjqX&b?JeEW;&4eV9Q=N8X3xvO*s}I-0XQi`SFQE4_oD_-YdP%uFu7F!7$IeK6&vi{gxwQ z5U9j}mE$Onxd*C?aN)Fca--bR`XM5{b+>1I5v;@AtQ5NXZso9uy1~s9!Fpq>dQxUr zZ!~eND6Z1pK>j)n>49@4vACo0ZT*VIyY=Z$w=-UxbBr!Pw0L~&@XQWLbvUg_WZSJ~ zC~TxaPvBy>(qc)u_TXrk2tnf7zPW?Dw0p81vcoPcx-q`)_Pui{0V6+P{wpkE2fhyy zT}kD9+9Z4GEMRV&eRA8J`}{UzkD%}X$RO>!v51k`w$y&d@Lgu(rOPBUR;V>D3UT^f z=7BAUrG+((Tzs>1uB4YRa!i)v*cmRla@FUPXGf4lF+MGzRgzrYJFUYwR|$F)mQG#0ICcH4@-9 zVY)k5e_GJvWvz~98tq%E?H3wO5vyU=Q&@y{C z$9N#wrJPyj9HWnh&kmiV;Ty!CRpktd+zS!>H2qa}Q{n@k+C zuFKZ6Ms^S^N*sD<=mb=_%O=%~Uq^IL3tSyuFMGFbN9W1a3ukDK@?)cHbCe1PWz`H! z&C;BH5L@Bjm=zym9{+N~>RbL9UhnR+9WMNuu5{nielr&T{Ze4+jILuuHW+^P^Eb?5 zlI$<~@Y}9mSW4%etKk|+zOX$?y2({W)@|e1$l#UTy!_f>W(3!mX-g+B$}}jK>d_GT zDEAtH=qR^#rGy39FjTnFpAOnVJ7W1|s>{kVjeapK$Mfwub_#XT)cCB=i=G&4EKsq1UPI-StY|VXo~cYkcAp zQe?5!YN)^!kkbU1(LG@Ot4t2M8bi`ob;!j_A-Nd?bi8A5R6+VUZB;;Hdk|XXWr^uGPm3PDT zxrpXFy6D$doX^$UL);=oGrF}BxBNS>;tWgAm&n+Uh|0v+zGa0{(R>+PK@7wy-lC1nP<>k47? zlG9V+5|R~r7yvKn4>Ww1YL(1J!Hk-p4n#DVo9!^Vs1LWPc3E2 zXxhDV6`2TZW6Q!r*23}6SUC6^T|aQam>2T?**P}TreK|+7KOUy-H}Hv(XzLBX~0u) z&9$BV2GdSkNHXIIoD1uZ7|h9Wm|?IhA83{5Zuxwxb%o1eOJ=OtS?dCf47*v>B_7lr0!kzC-U8VaGx>gtV2WTloof-&AQQ94dT}!|>ejwr$oSG{( z_V8NX>Xy6Oyu6*JJE#D%Y^bl;g4TxEgbdU$GV0t9I_~kCvJ(00t?ozk`U6a(p1SB$~FhWjji5YL1^Cp|$1XbjRb&q_wp74tvW}jhZ@2{k@r4b@f-BW=GL! zcu}*M*8)43Cii%w{xvoE0w`6jCl|ax&C>5Yn==x98_BK91XL@O>n)s>!hE_NnuWw6 zcrf#stgyG{Ac7=wjjA}VG!2>19Qt_f%b9=4=JG0=&5dNt2W80f3a;@d?2v288?&Ku zyY7XQoJn1~^Y>D~nMEXYI?qX?)>8ZYM}-C?f9FdqD3PmXZ{DKHmrS|gPDBG3KO+%V zm7w3j^cii}j`lhhZNE*R`OcbmM7EA=o@ilh{>Xcc zUa9N@QKMm@Lw*Hm&`TEwW@ir)1m7~Rl~JeTl__fFL<#i5mfX;Ua2&Ny@$rI&leFEA z*ZtpqN?CU#;#`*&&*lf@TXAgwHNhVNe=v&I&Hi?)Xw!b5kpo)n5{OFX0MSYAQtQLL ziF_f#dLA1h1q=VP$x*G1;qvH@{o;d{-i_v~AJQze<=veBG(-;v(6#9@+(abMfA zb=~~3Ow5=Nh={)`T^Y2s3tn3k-4nIyrv|bgvqJggd*%#i2r?Sp;?*?t0MKV@d{Dqs z>ukwOO;EA=F;}@nL9dpXC~v>4QSOlAF;RQ_hnfas)2~FOkpIAWm|2 zRFovw~j&IfOt~~nnwNUdM|>*K5Aj}l4v>St@EUb!t&_bh5KNvH~<&= z3rO(sIS)7~Fx z%r2!0uU=-wqW3De+%!!sp$~}Z7L@e-It3x_XJ3O0(?C2?| zrV2#e(4yH)N8t8mN}U|8C;YyIV{9hM2n?3Fcxw=W4p8~9ex|i$SVVYNIBYXFEcac^ z+@Sh;8z#G{PUc9{8@)%GF-kb&rMW?XXP+VAZ~2GHq17udv;-hi5zsi;9qiQ-CF$&& zT=J&0-vrW|=4x_?Y-|+P+2(Jy8MTZLZfNf`gfamE9P|h=+1;Mdrlvw-c0OP*ALL3U zu=1~|7%qz-2|JsjWH;x)qI8{lEOHqgQ55AX+VSH7{a_pvxRtp>_7vfh$KH$s0vIZK zo)iU&ITPK$0it@}W<+EuP%?{GO!A2==R%wp*_Ek-{o;Jl1$(-mb=%;m67gDV)v z_8O>Kaa&Jr58ae5&p)jVDSVa5mc7rv_NfM-v96+kV*RMG3REu8mIRKopMqz^TkjUW0tY&}I@R7QW(ki=UN@`fCqKnv zQ7%GWLv>IAP^)i)wlaG4z~#0^Q*a`ou~HQo{K3~#LF-_b)p2cZwVZCo8kx5eV6b40 zhykhFZ#IZ96>_~?2XM(VeLDak#*GT-mi8tegv`hM16whzAUoKdS+McrNQ^=f-^vG~ zB6Z!-94SFY>39&`I<-tBV8;3mxhe*?>6hbUXcgB^o1<6*Sid(GX z=h9ZUV zWi~m`{0dH@dcOMACeqP@+Vo*!%R zLoU-Mj0YsDHLEE^dP0AVBY%!T6D;U#y<+VDg@3_scdzBTt9bnx;^XsIJ!w?$5X z3f)1*I4gZPnM=v}Xl^y`YbywgPEjP02#@T8}^4TT;~56r>Ll5*koi)TnS0#DodtqVgidn%^010BMfy#LN*Q^EeTN2F6>p7XA^x4g6re?>7e}rch4B?}N>KiaNxkj2@ z{6)UQnSc|3Z0J{cVgys(c6I=$C9GS`7^PgIt9RaEYV>n`nl_$MNiz)OxA|qijHCa8 zl?n-kLP!D%0(an*h1R#VS@9t;lB<%H{G9tJ`1T$|GaD2JeSAN#L(f#XB3SSzvo`EB zXxQn{Naot7>MGMC8;I8HS?Lvt6>*Nb^&VCLJeIb2`f5dBcoY&WtBK-j`kkuw?>2^| zZpV)~zWm#mAQCH1D;a}$S(T4-C#=3)V?1e4xoGW8oM&MW<4`qJggtcCOZh~{vN&60 zLHuD$ahIx>@aGi$k>ts+_7STat+)c^B)>U$g!=aw=C+V>%7mxOG0R@6Cm%lG8*|_uw-~*6O!7fDS zp+9w;1XqqA^p@M4{?K7m31Ik5AoSZGV^*21H41Um{L}sFSYdOJFMAtQT~dgpceH$Z zOpF~F_x?8rNDA9av^rPwcT68L>w>B3W7fE8wCxi~xxrh}IGmvP^mGMk`+KD9RCVq1Eu1Dy|6PRu$ z5^$E8poP%d!nZ*{X}q0&d3~3Mrj_$SyTW7;D~n~YEJfmrA8`v5k^*6uNsUmx%gq?+ z<13>oGnO7ZE0=rjaz130?SuhgU6^H(cTK0c$s6vYnuapBbN2K7XW^3~dAF>)o^jzDk48Uu7^c>iEo&i#u>1+aK7ZKD2jX&vV!f4;c#_M{E&Y zv3;QvBGf#Om}K;Q{b!>YVf21%NHAF%^;>dJ0d1fM=&g@gBjz+txdF^xPeaTA+3 z9Oe*lNgq%*-RECHC6`1bTL($1a_`KzIeuoG8EXYC8r(g#>)@y}W@_d4t;pb!d!k7r z>yvK3VPk@2{Jskkv+9Y-t3hXc-a$shyAJ*BK-)ZzG6+o!SR0)!WJX9ZcB)f=X70wCQEqIj_8^1e&HU z)k1PzmJ$~(Q_yAR0xnK$XB$^g6c)&L57cd!&K~g|4bh^%_iHnkdY#w)LL^l^YbLzP-q^WXpRYvA!#l9f_`iX1n!fU>cvzQz+2w|lM*am58`<;#5B`^=DSSxlny7|yuDzdu8M1{&V(ci&PE45EIe zK=82*(1CSuij#tbl_&YZZbm8AV+ac()g{IA>B?6MR+XgxOw`DL6Uo_KJ6pbl#fz}{ zAp``}CZbN2e55GdWPC~V)mBH`3L;U@O5saDYtir%J;Kinc;CX67*1$m6b6a~EypjD zrIGqRPY6EnO_GmUyp!#XFrx2S>n{-UyyM9cp_Bz4E-J6+zA)Z0&9Xs6Fz`$Z)70n; z~@P$!4k&8?3aTkXyBIhW1Xi;QNm*UfWyJw`*joF*z-rduCMOGz6L zXJwwq#l!oDIZVfOsYweZVM0a|Vo@$(RZO&(9ngg5Fbl)KGBWWW!k^xj?P;_%CF@w7 zd)IF-0Ehi{=aJTBb*klo z_XQT~n%GgRwN!&c#V9of=E`_b^d4Y8%wbA@kwjEyu!-pZQ=SbC0%E~qPkzOVkUB$` ztPWAt=XWmya&sh+D7GU+QSiO*sH5f|t_y_>WIKW4RDM z`=m9ZeTStzkb05Na5jMcIj_|5T5-xLK+CtIRQ;Wx;!CW-1ta@Ey^CH#e+S=YVDUEO z_#QUcQN%{)zR>XdsJ<@a1Id0yzPUJaf1nOjUqVjp@C4Xn$ zfC9#EjxgH1)zVB))%3f{3G9?W>cSE#O3x*bqV*|!5NM1<^A<24!slPHlBYh4HA1}p zQ@bDvO~_R4>llktN!sh>X!$}l4^72fiO5g;1HqBMixL4pr?7*>9vSGKY%NIlBEue~ zJWV#Z1jJ%ZjKqyj()*BcZ~I*&M!~uY2Ag=CpyDL$xRKh&h4h?%_qaa|2)NYV{MEmx zi_1mXW_(ARlaTVlo=JC8+4Y{{gqZRhLwyjMJ*Bi=iT+?W(C#-sJ@k1zS@y10%s`D{ zfu#TWuoZc01*E4gMYcC zfP|Ek%@d8#`TBhl4EHyt04x0&*i>h-=52;<9x}fR>i>s_euIIbU{6KkbEfo_>!kai zX8SLL|HnU5FaYvy*w69*zaTWF~Q4Y@!`ukqq}^>MxD;PxJkcmp-EMFBVvyyidMsI}+4u9MUCle?6dc z?GDiS^utd)yl=PCbo1`{Q68Z4jo-mf#{5m)<9~XG1RfL;u>e!v)00-IY^r9($DxF` zbS37AH2yl+MOJ9n$_J;Cq?a{z%#&}n_DXdq-wVWw@fzry<(1xWTYC|0dFzdUsgU8 zSdb_VsQ?>`Tsl|QD;;b?aWzKT;D!&Uq5uRj0L)0fn4Ze-fkkZgANK!mmkh-MayD0v zad*1ZN7C$G?4Ale6+EbXbWrbbR;xIgs?d8bC25M`2sDQOi?Z~GAh1AzMf9br+4Pf? z3pPOJ@oiIk2{DZ0aE(!KlxI@^&XTW*09x>7%t-g|3`gPxB!)BP3uUZ-XYzl$ME`yB zlvL0_&2xv9`7eJ#Kz&lY0CXqX3%ZcMv1dqpZcqU0!+((w|96%Q?eh!Zu<#;Q?6`nYiUVkD}fFU)>d7hBz=c({Y`mR+u@Z3 z>a{dAox!YLp8Q~jQSZf)ppeVs^Aq(f-K92NPPR0#^%sw$xc<&OCjT%U+fyT-^QM`6 z0Cl#oM9SiDcVWL&Jxf*ZA^*06No(qrklO2g~>hQ6pVet7&TRKLS%?~p4K#SOIWv4d9Ao^!hB zg5^4$)r-1n*ryPguabns`?tRR|L_XH?WYh6n|OwJ@#6UzA5D$vzxiYjQeQ1%bmR9D z3s1EM$G}Z2g*mzeIZC;>NVsrJNVxb-NJ!K|N|6`5r6CE1yiZ?g8sV-zK3rb%xbC>` zVDr>3okl?9NFA_E;Xt$$uaJEeP1MGSiEOuL6)ZZHaXIOs$ z$W__Au8z`^+1|RIzxj|GiO&H#Wt=R;$-U!=e`Wpt?>u2%LaEPfuFJyMw0T+*iJSh- zr9wb~vEim%4rWDg#4CmT3t#g8jzk|aLmHS4{~6-{7elF;!+gz+sOIxGgbNsw&lqJ} z2NUt_7Y=>?cM(}^e_25Pm=?-w7=MmbG_3=;m#JTh{-TKg_reJ7py?aLM|gJP|Mk!O zlbu}u!7C!P^*H);^yQnhN6Cu^8?yCVi}yasgHXEYkf>y6VxhvQWX-au%Ry48nWESr zR^8~Hz*dtl{}?#PE7W4>FYQf(u(nDdAvS8UcI^J`8-4L~#C|7KLR$zqiN7xDN;vt` z8mgH4d3P*;d8Gi>I3h@!+@C|gw0fKVFDU`nC0JgNV`SR1-KkS=+aVygZHPWW#}v6i z$5hVf9lFeZg07Ra<=v(67||e! z$YptP)aB4k5wdW+Pk*QgtVYPCmz)BBX$r16uo4QhHobbg@9ftB%?d*Z8JTIUZ>P5N z{gO@%%h@epw^d>laSp@|vNQc+@WCGjRae_5WeqGc!$-qg*)9MQ?YoB3+miyA#o=VKVHJ-BO^DEu}kT{H;w zqu|KcT4PVio6cmp-RkN&r{gED`39@JgI}bJ*;QtWZcD}&RkTVJc?i3cuBkIRT}89r z8IRCNgWuoe1W!IcUDx1RjV`~ODzLB##=GQ8;x>CrDV5NMaDKIM)O=rM% z+_2~5UBSw!^-_Z@VscH+K8N+#_b0vr(ee)OZQLk?2Ls#n9>)1<({v5XJ4Z7bl>wjx zmV>%3qE)xm;60;S4TjiPwC*PFNa`o>iSj7G>NPfN!7=dZN2PIwKkT}P6j^6Gb_)6X zPzlU|*huj*v|tH3#;yp=0WYaYGO>$y1q-gpHI`Y>$V3CChtZaa!<;^QgY;@;Ii{m& z3Jp~Sv`&Q>CQr<36e#QcA^CZ=#@}00xxM0mu04`@qbWHN>s`4(9$WGCY3Tb1qTzS= zd91VVBC+&m#j9vM>0o$0u2NJ>pL#x`@>~VRFv~NY&nHJ$ZZnOQuXp%gwlb-W$kXU9 zMod$_+N*!dR+Um^KC99q01%PM63*9v{m*i!A!n}h@@`? zF<+oxV1pJf@j1G0&d>=3p3wPRZ-OTBq#3t7&1l^!vb|d^9tzZ4B|T0~6dF8l3yZD8 zKA0v7r%pSimz9H&4%?yBdm07of0`+>^47%y8^Lbtb)mT4J@WU&w;zM)#i z4t#}xI(n^yvh5;uAuQ)xS2aB>78uAkrpz?22!NWYEKF$=51sS=Nt5yKq*)&yfF|$g zU5Zb2HZxY4TP55B3`hq*gzW^vjZ9|0+gZ#%96Gx#)u<@#EtXf~ZXNbVX7%Q4=ZqR( zs0?mJGO3=vi}==)%x}bWiO;D`6*y+PSy4|qJW%HN#4^`nggsrlvdAFdQhHG3xGHl# zKiR(mdY09&$H-jqzF!J;f7oFPK8@0)dIGcH>pR$QAMp*uFt#EP@|d%zl-?xD%BUSn zDw&M^G%(7hp7Sl|I7$aSWyqvsd|v^91QXxn=2aJAyG0MN%4Ha)kJq0IC{@#8ewsRg zFj*A5-0Y%#gTA+vV}x%$S@<>aIfN6o%k_+GZ&wyqDQ`5Fw6k{tKu<2cXV$Gk3B%{> zIy9`9OxSr!R=+HYQwKp%*M*1e}FCTYl`8z=T4pXM9Y+Zs& z?ibj?9n~hNIZ{|TyV1Q#vn?rtA1AWiJh)HSC2RXh6mi+?QDPZ%yR0}pUZb|LU6|eC z*{Ad%)6A4;$hgL(Xm2y$jb&2Of5g~7KZ+SiX89$71FnSYf(QckLr|Ixmh60Y>k;}e zBu%49l;mDq&IuSEuR=(8Zq+oOZr@ByQWZTw+nmMfpcj$;7DM4;V8-y_aJgmK!a~`i<%b~Sf}^t;HhiPDElh|t z2d#LTp}(Yfm^r3d*Vt1js%0sOdlwn@;%D!r0`JtMcdX6VHRy*>h0KAuV0xSOiHmzb z=qa!_qxo!u2KYU+Aunv)F8tNa`Dm*@W~Z+IHr&x+?Ml>)gMP_=8d%Qt?tqJSPE37G z6xOU~$}WaX?2UHPh}jT3IakWuC2Al=nejK3xmPeK(!y_d!~ClA2;4%X2KDG#RCZAF zw-q3Jk02m%SuUB8dvNI?%8%Yv?%#J6z` z&wJD1S?pFM=Xg3{yh#~BiU|Oao8?<7k5`^$w_X;Bd_|}H=G5wbiS3KY#X0m!v({23 zb-os^Jb*QEoc?Jg9zAao9J&6H{#18wDkIh^q}Fmt=o0@`K&L%^SG|wpQ~0A#@!;T) z#p>FoB2E}L0cyI@Lph1Z?ipqwJyab|NUUEDyZX@m0YdSoiUsHCKX4!j=?w-c(~A<+!F!!=BHw=HdmoanQlT^B?l(2-2o2b(ZXzK@bQnl zp)^Ks%V&YfQTGb#VZ8DhXZZtPbe#u@+9qk1P-TWR0lm8%snjI?{=`Yj2V))1>D%R{ z#;II@-OfRAEnCv;q%V@WD2YEFpG%b}Q^Hca&D7SL;+IQ%qDuu932e=SQ}I=`Fbusg%NdDY^Gnnx0)qD zV-gx~dNfA-`mW=GXKfRDkVJjxaK50Vy1=^3V#ZoVoyPlk$>mj`_!}hnM&GPLg`)CB zRn)i-;e^Y()51nYH9DVwwa$6PAYee&940lDOFOdEiEDruP~f1$<2JTQL{AY?CqAjd4v-2 zY0Q-q+-!i3`N2>eJ9>#O(S=YOIy@!Jc&{8#OydT%R_|B{@1JL92-?61=(L2{IUY@% zY^6}2fX@cn$D&>mzZJfMP*=hqtK|v9<mnVEzhm6VaZ8m3s zT0S*{+20AAwL%>IG-rZmgvxUb1*W9mn*SLo*v*g7GX;rpv2{6mQ{R;v``1xaS=ywhP6fqcP#>vaps88jl!K0S>oJX5 zd|!3n-7lMohtokn89sGmpNm$*y(esq7TR!32 zmk7drwBz!+YeWI$dDLQml=Eqe4Gra|7f^(p<{MaGUM zVt7E3Tt99PCZjGc=fmFc1Hx&fMT+$_@4Q3xGFKK)BuDy({9BaW({+LFccHhNjFq}D2WakG|S6GSxXDJak2V}oa{~H#)09K=Fjfe@S1f*0wy#n z48#PS@tC~Gp_WR<- z&~H;ih;JXy5ud;{XCu6%JTRZB$Vx%a!cFrgad_yGEL^F>TxxO)$$ak>;S4)idxy+_ zv)q(D($@W3<);rxN_KK_Wu3VV3ZyB~A>*;oq4WL}+z1%zILjaTm2)Gbd!@Q9GF}1C zMyE?O3_`=|;I_O}?Bd%~Cw-u%Z-HAd&i=aUv1neC6!7SU<9w<524x!uAZ%t z6;d@Sh6R6jy*hAf@Z2o)Ea;9PjyT&GU@C#rSYMkrV_m;G_$7a$-Bg-^hn>{l#)*=%Rmyye)nmWfWR~Z02W^r&fWT4(po29GNi_sKTFBS_- zH^-W zbXgH{Kf4wXwv7*+fIUs&SkU7VO#X(2eDKhb+PpQtmm?oYu@jxP9QOGoDp8d;iNMStnA9c=$YQao3V?wR+v`=GY zi(Fl!zg$0Bzf^YR@xm_yb@J78H{-GQSFL% z{6hY!hJPQFs>|8N2q`vZiQRN{>L>BzvYD3RiNbZjrv-R*(y~{)pLKWxX=%Og-a>_l z*)GF1%BAv4V>9cJQLF%&8L>Z=}y=li z@q@)^YWYvk01D!AkGMTu76EMlu8 z`upWbZYpcTC`D1oQ8c$7{%p@u1Vci+zGGrA3FrdCk)$s@CVv`a?XGr;Ph~6U>@9UV zd1zWZ-6Ip^4ovS$+F1I)?0ed(8pKaJ6kNNg>-Z*q;YIV9=fjV$ zP;N3>O|Xw|kOdHalxbdEn|^|Al!~R1dU=1zQj&B>eo&}fLPW^vsy>o218VLH$hkD^ zDZVv)s!_P2hlxz7spN5f5a$N1W)i3wh@ohgs}$GD<@7pNmUnwF=*%mJ?_NG73R$iq zLLfqwyQ=AHOUJ#1UFEZIdJ5N=dAQiGKac-#tt+%{T-kG(Z+b0fa&i`{r_>m}wCSXT zIwK9wMs7x(u&!?npDUmKWHjP-8i3Z5>UEDiFqhaxfVBWgw{3mUi%m3VjoCr2-qTIc zyVR(H&ChJ%NzjQJGH>yRO&z2Z5agN6>b9BVVu4ma!AN z&XY)lcyYIm1Rk5WF6vRtSmUzfwOhTF_@NWf0cb+IPEBsQyQ%+7QyW8&{BHyB{zi~J zE7v23E3bJ?hYsTxo|76v_12mwzRB!g=Vnp?2s%!CsR+1{`OP^kCEF=ym$iU~Y z{(y1nMYu*GlcJ6@5cDx@&mOUTZq*NZ_o_q-gWN<$D(wwrcL=eb=k;R{aCWCyCo79E z?0q+RIA1PO8qq0?2_>U?z0ce_c3WwhU~kVTnb6h^4QTw+OBK0wPV?G`1oj%MS8Mf zOat2er^kB?HdTiH-Ss-V)F`emhbxtVJ&%otB>vh0L;`a3@l3HRdJ*Z|>3k!M&mB%@Ao6Cw}W`(A3-Oy0gw z{*375E%vL6T->*l@`{z8?|oU8xTd)xcZ^^~n^%>WogX3Xl!@yY8Io7;6)uWoG%Dx! zSd`ORrm#0}_Q=FWMNu=mPS}6Ie{BU&l_-xSJlVUtqFK5%a{aUKTCT7fN#q7}zGrXD zWSYI)dlG}z`f_96w2`Osv93`s*Qgw4rfJ4%jzN#Y{k{5y)p&7+_Snp$-H|n`NVgrF zk{wuoMO;1CXH3uq;QivVYEr*Io3WyMl_;fMZJCB*X#6hAAQCae3dOt4S2(YNc{>Jsz$uarxiPDS<#B1I`ob-h%Dx>&OUgiKXk;}Yjb0BHw^ntyQ1Roq%}-n0 zCAqnPe)M7lMvWTH7O2anKX|{GhQ>Ol-AL5+Gb&Bf+x@#$Q}&TU_+EB9s|CP4nFgx_ z>|al*8DO6QA=!a0$Xn#V76*Xw>(inplbrCy#bY}m6Z3+HA=z8wJ%%Jco5-r2oV`Hu zu1Fw!mkTh=)awM*q93|W*3hegGMF2#*~Q)#YBRh;CY;iO^WlYUboQ*(vexi_rC!wO zYqGMe_rb@rewak3&dlAM8tnNR{*&!;DD~5Y5c@y4zJLbXRiFCd;ynPaFQ<^6sazHG zbkK0Q=`jRrxh2iQ8&3(nLXGAM-s?*Nc>cA|@-4Bh*xA=4Q8Ie0r7nyym4^W-VK&ndPIsC`NcfbSGr#P?%y4U2|kPU&mB! zwi4!ixJ+LQR<>g~LO{vHRUbZ^-Uaxd%mTaJcvo1vbPvUmEFGGj!OPI}-s(YLzNWw+__Rs# zQ;X9k3B%S$?C@^b|0l)H%lzP%r2UJD(Mxh)rq5$YytU^W9ipYs4%9`}!?_)<2x6Jl zzEFh{ZlRU#t<)J1?v0fO93WHs?gz?m^EFxRuI8Y2oR! zR#Gd6i{$mC8mxN3GQZ-+uL9JuocvtTe7UJ33H&2dTXx7nf*JAGLd6MhbznM3g@;lF zBz6$n3A+{tc8ws9n7#AIi;bb9LxSUlU^H39KeG(=4=2_p^n|AqSjq@J(yP!&0~Bwz zmLUSiT2t_be7u#F+dxYt;LBB!KSC!3iVoiQ*GZ-lcO4V$jlRB3N?zyAF_iqgUA{& zwf)w;?hkVy$Q&~I+kB$G8e~Q}qEQ$eW>W{MrfOporn3Q059yGsAB9M%Ta^jG|LAc&P|)bob*JN(D@*IL{8P2^m)chw zU065=HxQD~&=pSQGtXAUF(z$VIef&+7E2nN%jw(?o)i6O5RZt%a}0h_Y#7TuvGxk! z6&uScc30o${nSU3#A_8tAb54W%z$mLLZCiFqf?qxh4DgQBa&ao{B-bUB^xZ>?H7xt zPCuSH8Gj*X(Dzd7zS?YdT6@gr6PZ<#@`NT1mI(M0?{^zy+BsoQ9s}$NL5gvpf1MOl z%SA8q6^PThLmQ|0M;utzDpM>rH` zFI2I|f4p8H4zj|5?~7+zBPq3?+bdB1K{pWR%J{$FY-Pu?y3j zvPZM8v%tqfFM_pDtUX>-=^hX>SEun+;pNQkG{mQZ*Ws>-T%%&E;M8bxjJ+-HAV%*F z?Z6u=U^C6hA-`$9oJkgDc%!2^-hSDyWIDviFq+20e2m=1InW`;Uw%_oOzws!|1Dj> z>jSKUJ}Q~?76MerI~?gKSRuS(|2E$*njgC|NuerQIMvxD(>ql#?kt<+=~v@dt-@LBUYcH0l?sNQQma0Rc#h#4w&eN#3bb9FeCn zFTMJMQEVDu6#MX=p&TZ>rSiHl%*3vc5S&&it}Y^?9n`A7v>Obz&kKp!8cjnRa?eVP z9RA95;449v;fh3h@VT^AWy-Y3RvA?y0mvAjR_g4egAG{Ij1anAkpj=k`L^UpZ+7OJ z)2j}A-8=EnL^%d752vTI!0O#9;(wSBCY5mS3-P&WX_zF_VI7)EzwanoEnlYE;V>BR z+V6-7(gf58xSj4oZ-0@9T@8JmXo#vNXuQBVKi#L~gdrZcaVkS2ft$f)LfVngq9n*e7CJklvas;#` zKR19yT6?P`frCMhS1VWslbpaw?qJCf z+b#1xo5}g9H!!p()Mb&_&{@$6bD2ua_48)ZDE0QDJ)1W*xsXMYmKMIfh~3u(VN*z^ zD0^?aJT7s%pEK6(m^~r#&RTucR_a%S64#k6s#2WnA(9Z(>j(6| zx%j+T91G431c^RS?!A`Q%cYdD9kAT0HJ>z?qv*H^g*^4Rlb7jq&}>~3{$84|nPg14 z?sya@@**E2@m8Fv5^`USmoJeBMRU|RC$PQa(BYX^J6dTX8|v_R6Ch9lDbD)y=+T0N zoW@sTH zxekKJf9SF$?dD)XaRPNBRCaaG;E4^^&KqmstCj0mB8DTFxSy1Ik~syXiB~B##w7O%RN?U1w}276oCa&vgCO7!L)RC#Z*mQ$n^}1DCsBexeo6Wq zcFn5{sMZYQs^`MLF&Clp-7FD-;cERHAq`Jo_JNYKoTDhHw~1DZQ=!Q4%K+>PGaUtgr15`DyyT|iazvuaZXeC zkmNAM6b?{!Rd0f=5CsE#O`plin7yguEg2{fNgiY{pN7vc{6&w29G^HeVO}OE;&YdU zCpT$2YKb^xR6X`|c3!NGaJ~WP1Or@@QW*o#NNf|4;8=#`+re`)(~5hFUo}~YYu|pL zRBt>R6>=p%>w_@Vr+t{(p>pyH^KMOur)nh->Xf%n*9 z(NnJ9D8xx#UVTDzHF76eiUW5LS@(IMZ+z^f-~9e+v$&14o1V6xr(~N(t!=#^6Ndib zN#L$bcACiPzSi^2OT#RCAuSWcj+#n}LpD=H0(_|2N@qN9aTj32uKE*b$OXRs=e(qq%gN>sb zhrUkuaw7`w3iSAM&F9Pl^WB36FRX6wL+lyo1K+xKqzltvAJLY>QCj=4<}_9Wy+OFJ za7fIEK|s&_Va(U|%${>o%G$nja+;d{NB>Ji-}L!NKy7w(qul#wrXN&{o*wID)rtQqQ=tQ7 zj2J%LTfXRGq}bxXR4C{p;2H#Wdo%p0T(*@}$Wlj-KC1Gc=MwQ3PfII?_YQ`Z0*{ei z+@nQlbQaKdQN#B`_vxK4R1kihZ3FB@f>#X;R*Ut5<*$=ZBCT-DdttW^p1Gf^qKqL? zuQ($X3)PbJ+n?^gPE@M%u`Y*N>!#?pr|TNB(Wc7>sSG@{RT-KSZB*)h9J0lyx-2m( zco|+E?w(?rzAKj;!B=($))MR}L?t=Y7mNJVBv);f5)?uc+ z_p^trxT>(<;=SduL&HqdScMzCoZ7=I4ah;XN5A;uKRvN(@Tzb`5?GqV*)6Ea*%T

>5?nALj(>}qhDS0f>Axo_tKn*WoVi!gl0#AE`LkuB2b=Yq8E7)8 zQ&oK}#IRPMGI);GiDm0A|N1Q{lh(!@k4A z(BGwd%ucsfydGy-5DV5GPUArp_6NJ1RV85c81sFk=i8d0F#)tFGLXkb)~b(c191q< za8Hue#X40him+&k>-Aa)sTLi%RJ9PRE-e6mrha#<_(S?2QFs@GTFJA|n^P2ifD8`o zS|vzKS!UfBP>dY%{2Sp|4tG@~e`kfjRR_>lQ)&UjCpnTwDKjjV7D+`uZ{N#gXE_tl z0Yqu$fK5!p=@ci1PJ&eMQkXeOZ}DnT{4llsKf zKEtz4nGS#0Ka@?1s167O<07mRl`0v!|9M}pf`~q&erNc#xg9>tbFZGD8HCW{X?J^v zzJWc-{m=vxZ;K0Vd_;tOBXl6q#t#5`(@cl22VMo-y2|o$>LaPz_dw|Dr=8vt7>Gmr zT>aZhxNRo)`K3UbGrI%YD?K3F##CJr-*&?U>1;kNpvG?Ez(wy_M*bWm0GX z?vY`I(Z?0bcUl(tQAi-0CiHq@Uq0QMFz5+`+QUn)k@VllZnbEgIu};f^UW`H3Xr?+ zR3BWQc}=Bvc@j8~c=Aoa--hf1wOd{qbtB8~-YiY`#b$cjwBS~DsJT3PV|hdya?PzR zugfY2pX8gQQ&Q^gzq5CgoniLsE@B*U+)FQI?d0^%%Hhit7`JrVOAqD>>TUV1L8$wW z-eSZkxW}fZWek5;iXR|SESPKS`|XeLV5xq;+*l~`8Yp0X6Bo@#dv3a2^!ZCax(kZ@wTtLjCpRHq@hfheMfUO`m)J37oN;0AyvvPxv65QP7i}$nUGJ?L=-au zv`Jczsnw8R-(P!unpP*!RFN^Cd1#|pDjs!TQPo#lEpRi=0Da!gVNjv$4%_2Qql0rF z)E#3bznDHdab@UDgzmLqx&&e?U~>e+r^hf}r3Pw0O)7@VfN+}ba}Vc$k&Z--~#DTQQGnv(4F=5QArtGWYyo(DYvA-^mYr&0D5Q4 zpF2t^j@Y}Gy4;0WbHs4t$Dgur=Hg55F9OXe0HO2ebPiUxJa9EzRS}nUgGvY_Ofj7W z*`E1?2$ku(uLQsKu}KK$qQ<7i8r1m4jhar2EI!S*gs!cZx2m-NFf+xS-ldpqz->1e zp{GzK@bxi_<0wGA1FiAo3npBi%hf@VC_NQPaGf=TN7NgHLsQE8IqU=RmZEjdYuHjx z!z}^#V{9^^0e0z`_}nJI0M?EJ!CU`;mv-l^GAj$VJM#fd-4J1LB{exR?Yf|jgzXhc?yrata}?wHYt zv_Js6&T(ohZ+8FCc6aGyc(@_tbQgjealJ?M{dvw6pgeam!lkwrp~KYseCZv8ubQi; z-<(H&UE5Lv0#m%2*Ra7(%@~{?}0wB=xj#E6W5M^#p zOuL<9x%l8Ig$A9CdKMkmTe<@wRM0}_dZIz@EN;ROle7A9VhJ8xI4xhmp!3@e960_N z&z4V7o+_(>fh0@Po98Z7D(Cs@Kd*Co!L2fBUF+z*8a)FR>O*)$hHP*aA-XVc2TSNs z9oi^qMJ#kEJ)S2X6@o>tE|`F@R&ezDQSD--YU*)T34~Y7wZR&6^x~vVWrpIUsmFXJ zJF#^10n<5qmG)Iq;;YWxV`H0MO9(MaXw4nCfS)y+x4=JrZg8spHfg)0O593v7fo*4 z2w<*<(VX_eP&UHT-}A4v6Qz7Vtv>qk*O2S3?m7C8#+Vkg-_*mlAp2eO!!jo~vIE6V zMqYlOQ77QI)z1--tW1_GXKO)jgghbV)urMvi(Mo9kWS?^il^6|-!2#4A^8lZZNVr37RD}7$|S^k2(2dY3AN0OG+cxS@CVfC9X0+TQ6r;Jbgc5 zOz)HQ&Bl(hX+UXd{|(4-Q+$exmg$y&l+=c84)p&1RpT zk%JUS@wO|1Toj|QT+7jtq`3zwz-(I# zO-1(7*Fb{lWeZx*{`!1J2Zr-w2P4yLlOHH9dUjd_#&!vPUGPR1!q|M#BQ*U|)hG?W z%%JG8P&0AzxcI`Gd(-rd5|#%PYT&Ig@&CCxl2gfIs~hg<`v?}{(a3v{6(gz=+mp!l#>&v=L^*dr{^ z15tQ6w1wFH`+D>A?glL!WU%^mK+?=%dYQSM^{n`hnL4w~SmD;`RzDRgo3zRY>_q{0 z`w{|K%^7`Pb>4j`vrQUZ%p*V#V>14V_ySiu!jB1TkjoRIHGS#m3vvcL<5=@Br$GxW zI6>7y9L}!@&I#L&meDixhANzMva%<28q~^vEzFL}qKmnxvnlGoIc=X?&n8sfFKAR zw~VJ??--wO{^NS-w-m8F7VMzRNzgSXK$D~4PRXtXHOw_^qa0f5b9Vo#ezj%`VFI*= zWGMy}+dQfqZO=(bfj3Cr4X&r7oSN<@no=dNd)g9e1e zy)Hha==gPBL1W@k1y4R%?D1MqH5d;!V!*2u=T3xM|4^E7DD`YJRwL!cT6dBg;tttU z+$2E5nk;kU3e5wAMqsm}fItz4x7_3iDL9Y2rRPAzBftoCdFDYzgl@o z4u+2JPMJ!E$nNxxXu|m{e<;%pzrA?uJij$rrbKJsN}NTww90NIdg|+BIQZVtrChAR zo4i#@vrz=tdQ7}$&`*Zk(KI6@kK>276fdpRhpB}?f}N-66d^>3mq$xRxBFCo22bn3 zbUJQ&Z4RY(EMq+%6&?TBVRhLjr;oJ8LNZreF}0|BQAxf?RgSW=;sO z{Uz!EAkx>^)UQ}rgP*!p$;rVY{Wz(!m04`XN`nsi%VAiI+M+-lfRcb0n!e)0Ujq2c zsn9e(44z5qM&^z2(bJ4k$)v~v57F4|zNn)doQd2a2Od9zj-(7eFDal4sZj92L&^vC z)cu0m>#56b5HO9Gt5*q#A;TfJ5-Q}ji67q|`#7r*3*W=+HWE-WnLcyGKb6(;xHPZU zCpRG_eQ+VU*kgqRSuNLw0Wg%$y=V!DJN*-eam>P88Vx2@yEeS2$o}M5pCv=aB_K$o zRnNpaY15_ImZX=ImXYY-$+4n~HR4~hSu}39MQn51g~^;B%sW`fA3N-=7N%yT4>dvX zDM=vF-s#-+1@+}A%>f%iA?juuTjk{St0W|Fx#8H1Q^3;Azvs@P#Pui~#&^|efD_sy zWHH}@>a;(}0KtQuh_qjk9Qsoar=Gx_gvAvJn2u(|ZanV|6yRLiu56zd5K0pTf-xjD zBhkF?>C?DX_zFw#(_i*>`3FZ0W?|68Ol&y~(b)g$r!85`)g~%dC}loesF6#+By(qL zFpsBN`QW<6U#Son^yVnHWyCP+t}?mG*>COBPJpB)JUaXgK2$g`}h9cDXwKN-3acGvU#7q&O*h%J8QtqnhK5TkeAWg5VBgx-|aea3*GFqvLmbceDV%C-}VVLq*jk$>*{2mok6OCr&q>cBAqaL?_1WXjqZv^_q@E~>KKrVeBUK~u~%xAOFCg>6`y8B z=cMJ*4vBKHkUH<&iI$JgFm^rmJp=kwOZeTUp_b zPB3ydsTb^!Tcg#4!1~dd8@!^JCxL3-oaB$msL?-cx)e}MUM~_m+ZTC|lf>+|a$y4Z zj;bxM&lBB&tr6a#Pyw(TY&Nk!prgLzG#T(~y?*Dgr{R<;<(&lkyi}`Fus2!trp_Fb zZMNQ|#^#9xji2G$N=;jeKu;J(tr&A<;CM*MtOa=`_SjPxpeUFJ?HYCl*(4|e$){na z{ZPJl^EHMqDP(YT#RONV$wW`{YOT-kELjrwJGy$u^JrM}R>w>u72~KCj9?GxXC%TA ziP!kVaCs)lNopofGO*{ZV~rl_uaLO(Xk9(xj^Z-#EEmx ztU^cGwcFMcla|Mw+;_YEkJcU4D7&W2HW|qRUw4n&&fca!o(Mpj_X)3;{KxnKPqawz zB|I+~V#7v&TLkFS5V$}FUo^0F;j(ar1-IeL)s}9Qi z=jcWsT*k`0^!Wi}y9U(r#NZ_T1Z|OrR;84(D{Y;)Ge@P=>g+egGfT|6Qn+$UN+57i z;!Sw&#dFHxamz-pSu0_Z$SOx@0c`E1X8_PY_H?B}SI^i~PmA%;FD@CE=-D1*Y-*D- z2%G+`W5a;#!#VoJF>dYWg_ck62X%Maj}0tUD!?;a845?9H$_gtZ#yP!<{P6;WYcLM z?%v=E(J1D}Y?~=%eipRE_?p6ANZR8+gMV;;5#ix@B*UM{U+KY_;j%v!{MWv|3Gg@5 zr?E$Qb9rmcETe#^CUQh`3(7SV3d#icY2A~S-4g4qclYAySU6&?Kl2I&ATlVN@P5SU z*c?qw)NnzWBrB58Mf_Z(bRU3ZehryDo+9G+mU;7Zh)gjUm+k)5ugyhN!{>;5^3MMF z70P?D0g-m)(c)2OHy@v0h;N8E`JBZTq zz&NRi!Kpfab-zOChI=9wl`hf=`C>f*M9;F`9%{@51;>AZ<|v&eARxoo6dzu_rPAjW zWf&B8%?+TJ`hEtuD;y=jq>Y-6-JK3K2?RU0UsE+A1q*?0G-`{f1Z(oLzWyF-pFr(q z$bPB{GL1t4O2?oIp+PD}iF|&2i)Va*wyyzT93TLOH|@E6#@{Fhv;KS2NGXkaRVd*m z6M@i|;j)l^{~C1XU4zUc?M0<cQ(-wvD|s_>GY5e z=3CJ}05I-b_ZlV>$)iiO&Peb)vg8_aw;a%nh7|rWK-5D<ccU+=d1lFO~U<@bOa+_DM?7gdE*!exC^X^hhlTT=jY0RNL)j>*w9?ctko|s2a zEXA@Q&o?%6w811kdL&DI7V2-E47!UyE8gWOw!Ky$K1a!eN;ALY4{;*??f0Imh<&<5 zT?HNnj_5=eV(_%}70wh|wYPUh&g`rXI#81}FV`1+yYcS)Ig%Jh6&*nBQA z64_?LInS_S=hl_oeSCJfJGVmwT5Y0fRy1P^hVw+FRlm)3PUX9L`o8Q~lPAF!Yid3N z3O1@VA?b1I^_I&KTuz9$Cm}jPmD|pK6qbBaA^Wm(n`3|OPq=lbTl$k z(4hme&Op?)?%^k#?D;yU=U<(Vy{|5Fx(zv*o#*x?U7DC~;8By#S~j2=_`5J3xEzMoAYC z&HE4)a+LCRl3PP>CpYFoVV&VXt(5xQrXl-l>F?biJ(P1~jOSlmAUi@%{d#C19aS0& zil2DBw(z}Fs|spsUp^Px4BMoZJHn6CfMiH~BhH`!4bg)A^qET){1<)zO~dg^qDFcT z@y*NF*P!yaN-gOvX76G763C+xEB`){hycjod#e<%cV&j2R_iPx!9a3Wo8r*2Sb9rJ zjaJXL+J(3e;)3R)z3!QZ0L@g~Dv+XFa=J_>Iq2OkGNY#2ZB2{?Y*1@&yn^{-(16j@UI%AASGPoUG@YBpOvt`r^!B5d_?q!ydpP z9uUt~N`V3S<=4NqoiAe+fFB--~b^8w2&g&s&S~dYhsqsxG0rhh^CN> zdviHQU%1)ln9vvtYohty5?*#24%Kj<^3%>yI(OU$59ht}N)`4(9u@axt~8a9f^}Di zh-+D)umiW$ldwbk=P+{ixX)CNKENSh9pTGe#A zj;}*Am#Xnty{mD>fl2*6LWViZR?NxmInf^4_I};_k#hqWM`dS)R;x(?Dsni3c^S4r z^mqgvmZz2URgf~>w{qE&nJ41=0t!|`Bpsz(=+$vlLW5Y+N^ zimXDjeSY{ajaxnoCkb~U&1$i23akq@XzEVkH1$OZPvl4^DU-x?_W}B7(xjIFlP~L8 zlgDv}H00<7g>e!HvH4n)F`!9?%)A85@_FcE`p8d#Dgk~0-PyB`$S3RFyWSqrOB7*q ze!aga>dE8ug@&mg0yf_!Htrcbb#$5F{jPTqVx!!^WYngC#QiY|K;Ndf9uEP&h*78 zs@_$;5GhyI3}T(Nb-|6YxyW~kI@~qN`y%BbMM zVv8s7+OQ; zJ>9z5L_k((uR>`M&Q`MmgkBU^82aLK4#QrqkTj#Ov*RX^Nb4+f&o#LpY1SX2fro+V zPNYXWX=J)}7ZJd(RFW|BoK_ zw`XB&gW3G09?gc^gqw#K6N%@y_eY&$*dr`0yO|*e$4SOmJ3XP$EHislEPHK5nKYIz z6m(TO?M_?YOT7&E3jd&p*H{%E5%|0v!o55Bu6k2XG$3A%G*kb&13gCWT0{nTTM-;<5iVT@)9ao+Cc4e2wTE`_uUSf3-}7(M$N$bIB2Ghdef(8%GWNBcej6*#c!vSw}fCY^2iT;lL#ArDbqaw>oQiIj*y7mqa4d zucx2u8iq!FF}?Q+Zz9ka+t|8%miRhkW`%z}U0v9WFy+BKoZ}S$z1t)W(AS#-TuM_% zP`GcF$rGWgqm|5faS=8_e+lf74n58rOBI9KHr|#dN0qA>RpV%j=WouM#VVKE#%ZegECn(XG`w zB|O+9)$PhWrjJ5rTO-*RQZp9Qe4QH+oH&!b+v~!J4d%(aB5+e1dau33+bo@|cSGv= za4>2zU+66B9J--wTG$d?$H;J27IWvdu~RU?;6_T&Sn6G6CFGE3>$4Z$()!SKjkdB> z#Eib_63Y?9jy>Zjm2s^vU6Z)ii3g>bG8q_6Bpso8g~jcDi3_eQ?|ep}vHXhO2;YSi z74!I4F)PbkU^!q=ess{5CyrmpL{WgInv0@uEIBsGgH5X|3@mG_bt&|*@o*7eQG)+u z`4(s|L5mYfLUbW5-3i>Md@%HD4kVZ|zkikpcBOhLc^OmE17di4GqBatx0Bq867A=I zbS*PIwttKI>alDzHIUg74`!*~+Ar@Hc!FMl3+mD*!F+(|ip)-xZeyPW#Pjm@^8|Cq zLh55~e$H2MU9?=29w;KWHu``N=9(DY&smQ%RsPi30iqWFMT^PstqrqtSAh*6@>jS9 zip*h#@D*Xvrw8N|nMnj{zrkxrh=7hqipo%3{hlCGqEU;-uyPh-+@{k+ z=%Fy+D|$e?0{KWWZ7Ke4$Hqyl`u*7{d)}wqcx|DAz+KA+%QhW&BEA$-h5CKN^P^AO z!0l`-fz1NA0Oqtkxla3($ZnoK*W7g|lnzlE_`jX&(F4_%SKgF{u-I&}-jtJzCd?Io zoLpPow9^jmisLW(y(u;y8zVlf6!_epCzI6U$88Ew=W*Mt_5jf`_hi{^(0QIzqVP_RYsHh>Ox^WcGkR_h*C&~;0>14f+oCzEuOhn#x%yCN z+vHm~T@VXnr&m6@w@N2@+W`{$OD9onCh>7?;+2N}cW0ut^XXmOVi&?8YEIaQ`q(QS z1B#~cheMjs+Kn`Y`0_2pqC}?(qbnU%^JxVnJ0R2H!yoQ;HVmgYXPo}B+8v-G-!1ArW}~+I+xO^VJ6X@>>pZNYv<2vI&Zn z`nc8A;6VAXqt8S2{FiYFW}K-DC7OG8ayZexhDde|H7VoQUqF4d2Py2;ZOK@6S7n4o z_nx*1RUVtqa;@Xp!~nz0@NWSm*myxzPDYX3G|vP|2<#|Boo3PTFc`i!ZOv(fW#V(s zO%&bx2rki$ex{J=x8?6Ye}DH^cOIF4@=>B8@`xjC>jffT*kZqIK2PuQZ<|NZ2SKBP zQ(=wxu}g~9uYd^%=c?yHc})s`wn+KKORs0s_4;$Rt%NIZYta`k$=b34EtRHuo&cqibs0*DnZ#o`L5y39!oao=bX35>e~V0z8Rf z9=oH_4ygoMY;ms>)RSq_@K_L8nFi3G_p;;J4F@L`;MDKf1I-yk%Rh8_udQQtR6C+M z|B;U8kM;i4=cyt99jX#r`t;TmTzvIN>bY6=^$!b6z~`n_Ua`@DMf0WVlA~`T zJF+1ykxitOSx}Fq@!?D4{N@W3UZMvbb?i4PJdg_cF_A3beCOAPsq6#(ca;pTu)_iw zRN}z~hP`+iHSL*FuR=oF&>_?-q8CakE`XK{!54A`b~##;CCG*F9RlZqHofRSDDr?m z*ef0$PnK6MZo}Oe3MyR?bYpAkcPu)};-5eIR=s;z-aEkJtNPARe-!ToR;e#J?4&*7 z-BYPGg5Q`sAsRUlu6arcG@}CECvtGIX%770nQXrlRY{&*p0+r_D%?*_@11~ki;zx8 z$P#DB`VZEQ&I)270O_IsSRf*}6KguAZ5yn0tKXu0Y?xQo78|D`vGjhJV!3IJtS|NI z^FXxn_Ayh3+M;N3jL@Qpa1?O+V z73@ZwXm0ianQNv`d8kKOQXQOL)g-Zx$q)x>C($+TbMNR{R3PY>f@gfvId%>7xPAG{ z#7l@IzvrY)2VVX+#*<0_Ai{G@di$GtZXA3`jru7ok`d_J`}h{ni!q*CXRM9{b^tE4 zh^@_-nRI$^1`wpNzIj^UDH8>6V9ldHS;+g?OwjNU4d((HN8!=*X zm6>vT+5^J~ejj;>2X zr?Ej?|E^sp+D`P_|L6W~XphOAidWjH2!7#h2@ekea|P6+3a&F1O84ToYn`bfwdm_` z54D++*$5!&muhb`CWi2qupk7LZ1QnAb$5fKYFTlv&Xj1nXc`jqXZYWnE*30S$|tU> z5C-5I1);H~e-kv^8EOov(g1Oc`3!pHnvh zg^3?&&71FbCcA%gWAHAQ1DuxGdn$C}{rWFue5zV;Lwk_jc%6Ufm1v!?R(1|ctcyUz zvnc5oq#1vpI;k*#P?U9?#jB{Lfl|E0j=%M?W)aLq8^u7{y3L24jdOtN6!K8eFvXqV z72nfI_lAI3wt&~$5xXaZaG4?#*r&+@s3nQ^n4WO?Y1&O1pOvNa>#;s^tZPXyLk!4w ze9@+z;#hb~23KnmA*=^*h)af;7_AsfVHQ@TFUkWrA7dXr?{)D%>rbTUcw9cGNwXS4 z*{j8fUtI%N{PHvc1_oK+tjlNSRQP=A!DSk(}TtJh@JmWX}W- zr(+p_OBHYUd;#t)@7>!1WNhz*3$dKaf=IG+^yOmH3jiN5nRi0=3x^W?4y|G>trsug zZV{msS596w8Vz%2c;3OD=X>MNa&%I-n=+&!Hx|jDbMqqY)Ghm#uVlTZOB(=sYy4Mn z&953i2^vN6Tp)SzZ7iR*4?P%&)W*&gNQ?FHcD7EYYZD3L>zn(RC|Ak}84=&j0;naL za97m{1jc*M(2qW_^dOY?WEq9q7*vAui4lX@e@e)7OFf06=YMTOqJ`1871l{tuC@d6@P0 zLC!r8uZ=$Y*865(zHV@Hx9c5$R?Rl!c;kr@c`5+O0;0WRN;me#W}S2Tw3cqQrR{_? z<|SnT)c|Qn^04R3$cg>)QjS&AYw2UV++g|q*T&@vR+Sf5+rzuMiU4!YYU35H0OF~n z)h1=%I6x1(jVt%&|2u2o8@guhNro z7al2GQCVjg*;dz}CqnQO-gJHpMOeQ}F-U%@cVy9QuU7s1e4bS*j_LW~-0hn`K&|0U z8s6LK-cOq53u+EV?mX{esUqUMM2vradQ!G~_>(J@?F;pbpTm`H{(}j)0IOLo46yj_@GaR{lT?W%iI-#o!hA$OCAq{MCXLjrF%=X9A3h zNwbKhN^C_wawiP*;MG_m5h=?ybvc0h7h?`6%fKFpJ{RDtrMmqT37INVoNhkv+vtn? z_=%SJxKgqw924z0`Dvnp_s;w#tWpv}5@-G|$>P;lU3bdIHk&{kt+wrPgu8GSCZg}$ zn*dEQC+djm#P9`|*H&7Z(!H%2l_1nUr`GT9uDSLCPFw4EJsGdzsT=sEatR)P`- zA48~L#_djIwdnY^)404;w_-@$6jMHUy#TbvRSLDg7G=Ar z)2?6de2ktxviARLR^@8=76DAFaI(>>A>eI_Tyd1RGi5drt7Y1|BbRl=E4Qz-Y*x18 z+b?!W9v#*L?nu0?gy?!a;4`)V5DxRDv2wx5tRYwZA2^R>GMUx9Kg+~$pqA?}GtQZ! z+N|I{4kIgbU;i_%XbIi*G9Y6G=@#YKQZoSvleRJHtXr}yUbntN$zt1`oDcFq|9!rz zRHB_97Lh`s`EhS4H+a!Vtx_wy;&7qE45cRoH?~(M#i+uDxMSo!yV+VAr{k}2lB$G@ z;bAh3S~KMqNAs!`HIg|O>k!9{Tc;FGC)#ul=V6(%j$k6ivyaxlGzPz~wC3|yoMGEP zMBX&qUG-djn%ceV3f^?7SI8C1cf@6p&)=IiZX4f3ONbz0nasW%ISh3CmgI5e?)0!~q^+&oy`dsAg z(PZ!FZkH~izHsq)(e#?_`{O$qR=-U+%dC<2Y+3vPvOmikR&va92(S4T2bP<>QWw~) zX^*w<;gHO(8?!Df^5&b}Sn1WFUkLeJiE9LWgxdXt)x$*|H9vkPKS$^IfqtmOKz`M# zx;a=K^=mXSO2cI*tC}T>`kmK{&)=c?abVHctSD5nPM2})xAQSKXS@M0e}xc+$1gHVWR{j%V~ zzYMlst-$QF@n=DypKk5*nyA|>=6P<(HBB(%e~=2h0_H%HM;0dixwIYRsiOdG3Drhd zPBhhbMsJU>$WVVu&INLiiOUh|2fi$v{1f*)%|}8-!-w+(Qg=1l`0p^nnK!7UpT~if z@2}dQ0Ol^4NSp!YW+I7~Mhd&f>+DLp?UBx;u>C`3pEbGN?eN6m9p`X!+D<@(|3%6| znP3a*6tqI9BEt{RfsA9$9Q^V-C~z9eZ@(P_WC5kc&#P_1K$U9(axpC5s+_b_IIJ^5@e*kwtO$NNc=k*AA1`gIyFlx~u z@a<*LwF|+Yg`U{JwO$>;2Bisk?Y=dJwj7J46@|uPD6$j_G8l+f;zr+(lPL;Z;1XF2P^- z@S{kO08)6WJ-9pSbwaocDjbz;g{(goc^UQ~wS%c_PG_K>H-6S+1kMG;*HqBCug=X{ z$oBK~?oRkKrQNZTL1Jj&cMVRjgFl)6>(lHE3{5inqc?D;gMu8jW=Cu^Z&kf-QPxxV*P*<`vpo}k|eem^M3e|y%3so64g)%vL8yMyk4I_<(7jL%gRH55fF z&)3PU_B~C|<@)3m-`Vv#u_7VZTZF~<)+%a%_;Wzg`?7M`c6Jg}=bJnA;Z##fpk&>1 z%kN9O`Ow?ba7jdjRge&4>+#R|3|@zvu`_JjFa1R8e4Aw(qX<<7lPQ?U)PjCLb>)F! zL0~#G>VIDX{>`S=fqyYZqz%+3-Zx}yNTW;Nv@;IRD?f z^=A8js4K_F&o#^5@!6un=BHm;x6*g3Mt5I@*=_V)Q4d!SQGI*-8BF9Wtc*~Vd_?>M z=CK`lPl$2#gJ1)&u(smsj-~G%&@ko&{(Q4K=Jk6*{Jg@LenRP1AV}Pkidoc?-Hy2C zS4#^qfDK_J-c{%%Wde(tKIVOC`o8fs9<7B*@2hzR)H|OMMR0DS&Oy`sPbs1RGVD9r zOrYL#lc5=cErYi@-=OX+c1yPA3eT|We(H$oV=rvJdzt5Azk&$szNGr$BJeFg4nVOg z6h!R$8}9<(oX~IB-U{!sqmGF4ELQ6|G3gZLuiiQnQe#0OzRg)f6ucqHXY z9g||ffKH8&5ZryKQR_M{l?cxTG4jn%1(;1zL+pQUTkm`+%jmybyrlFW)JL@qfuwy( z=1Df^>`R9tpXjfB{{AzXyk0+MpbF}7+x?uA`NVmREzj$>C?af}I${E7tHw%bGD-2{ zorjzdbz?PF!Uyzxo1ssUV9tB3ByPi@)nxxw3d!#KaXUY8i1BQ>Iyg|gzbiFR4w0fa zGLyEPuf!gxL##BoVfj&Dtke)*5`vERfPJHp&VGn!4dCS*CCH!_wN&ypnbOh~l zgOGFcDG7asUOXyT~&hOVv`F_CK10EvHCx zGbANedLARx-RJl4b)SXMo9FKkbjFOlS8ax9shpIFBgVc?q-GS&2hmys5XGJajP)+G zh1dG=h%;t(K9l*MKHUTuR{-xHQl+8P*rSqjPLDrO&6|n>L7GJTC$VmsB;60^Q!#M+ z2F>u=$=RFy!|%kANQldzyKfU}@DPy$NAPCPICPPA@31zreIjt#LublYtzHlRjCu^@ zdmSFFN>_sz>~CE8+<>9heT0j(T*a0UV+!`v0kT%7+w;?<+k1Oem;yynl^Z5l_FD6A zTxm%}9UC>k?!3PNZ;h~`OXy>RLAg(kM>Bn*_n|jpz-wl)f6}oge5zS4d)jYKj2Hd+ zL#i-D$&-0dCP}lS#V05*LH*emXeS@^{iBtg2i?{AmQRuXg>3yea65$+=yOYUr`$^xhE zA@d?!gnAaFDNul|8K=K??Br1S`Xyw?FKaIav@L+>7ic7h95LfTfQYp-25J;Ax{QLp zc(zXp!gXWWH7_3;Y7Y9c2SD5MRG6&>u5vPvAVy8ZSXxzw26$D!-|Zu|ae%`*@%olw zSAlik@?v1i21VL!pkG<<^n7uV1KXN`)s*fKilk@AGoYZ+`b7G<)j&MColq2(zL27< z#MJk7ygdY8NE!!`~kMM%3EOs*-@n!dn$w|O>%$fJ|fqc(HVk>m+R9X-! zrDA75MIx9pSDpaN;6%2MxnhosHJi!l%|toCENgc;W)CB{6ZV^y3~>?ax?wd56!Y>N zad`6WT2BnSFsjCs-Ufk*-0VRr!bxy7o4|2;O-X(hM>sRhq$cDOO1C8$ibnC3sIq`}Z%p-_MNMk@A#<=CuHu|B=YCPfnx>|e^`{VIpLyrezeYmlj z<^6c0cXDP=+JYgKcf0onWjgAjNiUj%sORT^{CNQm$TR=zliQ2b9JzMJyL_kj*8*`} zZvW`7b4^FWkt3R{@emIPJ` zqC2+^bJ@-R<3lOw%M)A#mCMe&P2a;H(jd z5`={a=m+2>_==bRxH%suA@0#`f9g+w2|Z^-27b_*-kc(RSDd}uh5WdPIfesL5-rcy z-XukhZ0!8k&vPxPciiT$oa;>1sp|8`#{9urE3w*SR75g{JWf-Kkni;6r&Q)(Xdrz( zIzF!fu}ZP>XcQVJ4LTzDIhHeg0@e0^{ak-_gbgB+}n&Jp#dTKSDeQEbg{Pf=PsNtIULB8Ao&--gq#B;iYI}$8mVym z=ObyX|7g(yjsTXgsoc&@=4|(>(;tLrd9$50z{GQ{f+H2n4zcM&^*IS4ZM`uV+M2{i z`q$yHZ%O|@(6<5*@-OAMWTIg>Sf#yA2LEIS`@0Fm=-x~2emPN-4R7ob1AIxq-Rm`B{}4e)>Q8ME>6$FuqIz#4WpV za|UWSJ8W9;|MlYg7XQfu>7z$L9>iB%bc(CyRp-_Und7ZH| z8qjNG^vS%AoMK)y{|$X@jEJZZ$CKNZwXDOK$5_*q*RfX<``C5MD9cdtO$H~c!x7Im z`#*LZ{pT-n|8zdGoqqbLXAUOx|94s^pZOXAHK7J-FO>02{%-}c|3^#L|JUS2YzS1; z#*4KLyaIpiF|_@km#+E35r4DdqR=>y`Xbc@{(qE>{ZsFx(Vx=E$$s|vFZorjDF06F zwzjHVQUUa}k5^z(lKHb*<$V#9RLcAX@)CS!5itY7}mLwI* z&>0RE?)r#M`!ao`B3Z}c(m3tO;n_@L{>Qaw62xwZf!r|5F~XbuaT3n`2-aUq zHDj{Zfb$~epT4L6{c$2k@YVHs(!M0(ATBf=qTJpbjK@z_ZIVIMpIvKzN+5u_Sek7c zjWL*JI;l=RGZOnBe}*q`XtCdqfm&3XyUX2S?Ii97Bi@T2-NR-{S`-0LaZ?yf;fI;4^A z?(PO*q`RespV9_F}a(O`7&)BFs+q6Ufh*iqU)5kigsZHxbdUks29g1S_uhrbq@j?kMG zsUp-C+pxC=;<{3CeBSw6*h`~F_9Wv4!QTQvvOfYK-#kf>`Wpu~<|E@HdpU??^KZTQ z#Ur@#Sr+@>fZIOB3JC%}clEfj&VO#SpJM%dWPO@G`Aq${SfAAxk{X6{*+8ASt2@9^ z(I%RRhhmV3Cee|d`9HrgUB1a5cVttx;#K6oPt|-$;E?B&tNH6hV7n6E-`ZWe)wfs`0Y=LGy8+we)n7F#@`ybrJ z|8}Y|dI+7zYb>2~e)!8l%}z8kH4!M}2LSXc`p)G~jGpJ|qj^8}G2s6@rwy6>1gOqg zfp!0B-ElXPej))L#O2S4;leefGL}b$y7{W@yy?z}Vh(R7@;`QZu@O%aP467$n0a=% z9lM)%r5nmL<8lyzx2T6j3vERUJ;W24Wzg|BZArc~|Ml)aKcpYeFeJqzCsFWS59iyp z(H`K9X_eFVg;b$$nLSG0IsFX|v%O+D{Z*}&Dl4S8aQN{*{yYYV0d7FM@-{n+W7oS* zZwB&;*#E|7eTV;)aqtR0CL{ygs&`}0@!x&gQ2-d&OT>Kdi@fpml5x6{|AxWw+oF67 zRRK70Wfa@s-#B6bfZ1@)mq_Dn_~7rq@p6=pkHP+Gpvkkufekk9zwrnWFeE>uVRuF4pj>^{Gq~+>tJOB zg6WUnFuSlKQ=D)TWeZKls;pSo{~#;Q;}X+weXe=x#4$?zZ<}PlxWunc=8Z&fxC-X( zycGEBbDF1ElF$5PQ>efT{qIy7mMud1vQDZ0k&}kGy*BXt_iyTh1%lQzbgAIM`Hdzw z7MWiJdCYcNDeciPL7h^*w4wUvWO4B6YZl?Z9{dR@MVH{7kTZuS&j0m={qktOZsSKE zEqs*GmGY(0e1gR-jA+ozpl-nocK1J_0RWPD^?)YpSWok2)KOP1OixLpf>G<;KE3fE zlTJ~Vj%u6vv_tRnyZc~L(~(8=I`wod&aDUzdAERaoqDw+$c#2HvCh0X_$o=}eA@+{ zvvTb+rsuJ{5?`ZMuB_>pnaZ$O#1_0%55aGS9NE?w@5;ZKe^4s+Fs@IYl2t&cW}R<# zN^0wLtoK^j=ns?r-t&BA)7;gb)OWFZ>a`XuQ7e|yN)2}9e$>}Pp)80_%=@mgy;>59 zLAxM+3!FrwjB|+I@f%1=z7(rBMy-nDDm`a+AP1GH!@dl~QaRNr^?uWC+akW+jP|+F zY%TaTW1H+gvz?^xAsR&@K;K`{CQ?DG{8R41{h%*bfQRHd;__{goxb%deX8KK4$$eK z$JcfD!(jpC>*Qxlr16ZM%25}MKqE~&u4XV9klk?oawVdNz?E%BHBG<$&+{Dn&yF14 z^J#b9X$h|8$!a`o4(oeG>NuW$+ELtq>a?JEy7I0q$-pvgmfk0ktnhkxdk}h+Qbbe4Qr$$%=c{%IHmk9iGCLdWfGfAn<68c5%g1SJ z`}oU8j`3PLu?ZqLPYb{f+edAit)yjuZiU#dn@l9MpZ-BI&fImIT%TJ7jnd4Y^(|gX zc%{d2E_c9hP4^~YCbw>DUW|EMk*P+xL9@Eq=SXAmwi26*dc>Z*&TMHAuTaj5@mhnAZ6a!WTRy1cz~*=Jbd43&7(9guS^N z4dI>(opzb=m7k?*Ou?=-l-CLg52=+Mu}rC4GS^|a+i7V$*d5j>@*~-Ox~5oynGgN1 zl;;x$T0L{Vt_S}r1*=UUL^+y*f66!&h^L=dV;B7F>Q>l00rUX4A^AoQm7UERysk&Z zaT=|Tp*r3ZLAH)Ii!R?AS3&BeZV>Or<+a*y2>#*dw@6Z%l^1UNzt8p4m>V>x7T-@2 z+HT}oOxkrR5)+GLCh4aouI3qq;iCq)CAG7RXuCOA>Sudvrtmq{coA9bo~-vOm*}Wo z18BXKtke%3EY0$YW>~1GjZXymT76e2slYmgJ`((#l5b}_XJ}0R?6GZ94`tD0%+hq@ zMdUj80?|=?+m7co943@!61s6Vz)+hLibqheAK5sZr0+u+XKKgNBe1{5TND2l^q+@Anp#6ohU?Y+J5f zefA%djgicV3AiYnUsXcXmX+I0zMtcM(+kig5B4V5Vg2eR51GCNB2U(R_LQl%QZN?0n_ zanl!%QJr%$e;>Fh9{MJ9u3xkyeEdZiv>tlG_^*dD=BwO)d7{hWrURk-d1E8g*0=91 z-@PW}ijr9bg?15iiwbxKLPuQ}En**Pze$#uQ1Mhi7wyT$-}4UwRpo!=EEh(S{^DM{X6(W z^H`|f>_1(f!iDwy?wMPTv-I9`k4dcqgepCBSlpXhy=gE}h32VRNESUZEwv|`CbV;T zcld_$_$Qra4dV0YKltxc*!1O;FWrB$38e7f2JPncsamLiG0TggQC{!3|I)fVTa|L4 z5+~^6B}7}U^*$7bO7zay@OOIh0OCxT8@`xLqF$_N}+)e>cG(8DHrbEykQ zbE8b~=b$JFuUPttP4A7$PBeOkZ@>T+L(`Okkev3IUwwtZFz@2HY&+D6Qv%V5{@?v3Z=Utq=Y<=?7NGgOn*P-h2bH(KA#j_-oV_eU^$0ND4 zM7eL;Fj^kDRXfH7wrQMA1SVL-hFuAckg%(66_9V`M?5X<5w}|%4qF}fi@*1#1hSGq z=C1`88sOSa(sok}2U;CQ^+&a^Be+~40GzyL87V!#2QJ>rvY*|CjN{uGu8#V!(Y*%N z0_>aU2iUJ#?CnrPN*B5bmX*2OTD#CBMqTRirGl%!h?A$z)eZ2|U*J$#tDOT6ErLjX zYsF)>YK@qc`$M-$Wfc1fJVGQHtEyz!lmZ{3no%U*Zd&d3f|9~8G=U*mM8|de-r%Ajn5b0XAaxB5I z36qn!XrV-5pW7DUQ@-1#VwIV}OS7ikHa@$Ko01>eOSx*@L&buVfGmVioc>GaiNB!w zRv5$9539=8r$S0d!XB&VkjhEuQ7d#<>KYX%M!5G|dXaU(bxYw^1#N85!1Kvh%-@i4 zT=OwkLlGP91JLtV59#QagGrjykM`z27#*=WIQT$QUAP%XreW23n44lRF)Y}h#f2Br zeR_}hYR{oMN78+8`+ubZgym9C$E*c5$)-!LAlgM zY*`#LAqZo?VJDiXp}OT~xgYgYVEP^`VBIcbv}I1UQzOz=!W^DCDjDHiJ0U)UxiD~& zryox;Zfi?U!>iziS2kMQ_*Pz6%p;o~6@&2T;bo7?KQay0s~VoJoitk=qOun~znagk zvS{kbko_xh!FYkwg_1V5d1r_--z6OssC~8PO&3Sm^JRt^&Xx5LDXLlhFrldQeBvUy2BkBDg;Acx-SW$atPQs}zpJ-apHnc6l;8O>%r&#V_Dkrgy_F z(8J6~7{W#(=2Mk%)0ZWvynQ20u6chK&UBVlgVu1`8x8ef!F?T= zUeeFNyrg`N{MEL|BUKLoQm*N}7b@%TKT!`ovIb15v%1usqDn?4c{EfGb0mLIB!e#7 zXUraLOeabHP&Bbf;`aLNZf<*MJr}C(sPPRQVQq#$^G}4)wf!G|C z)?b|~K6K``mhh17mT_>ujo%#`cgW`wTf(*ahwxh1wD@qzX6cvDvE4OJ( zNGeg>m|d8NGv*Y8q$tEbsb;OBDX^>1Raw=j(ijzb;MZ(Cvw4g>$^1w`cT)5^ZgC_B z%bw?QXN!d4&h9T;m5sfm(U$f`2=en`GUJexS2R| zYgi&}CDTQ=U?j1C>`Q!l+Bvln@;_cvln4Iv&vWpRWAIp#3#U-sI;)nKUvonCAVAbD zp~fszO-8`XI2PJapqA_U%&))SKS`fDeAQj4#uOEPR&=_X8NvD@i>|!!4kH$Zv)U*Q?Wn4 zi8?pMHZ3D9o`1-1A%3^|@+`bBc#A) zTcG;1%sHQj!^TPVbfGK|VGvc36vvl5i|yxjp8QcPPraY432TzvLMl5E!7{eWv#kw2 z@OmAG?Sx6Mt|x2Z&qMQKRp+<0=fz+@(^TcJ_5TF#gl24;jFATETC@ewyiI1B)~Y!% zZO)O*?!xMdeOBgbXi@2gSEvzrD&7^iF$p8aY6esQ+dtUzx6Tl*U389pf%Q2?+i!ht z!YG2a{ZU7q0_H)${DM>Yk-?{(#WucFAiKX*KTo8WYbbSb0 zn{m?62*YQguhKxoFRy}xk7%1NHP}I&LZlQ!6ndHZ!*(44N*L^uL;OGSRGB;9+T1KZ z+$U*=>3RgMMmGou|GtKGBe+cm#DXna=p5##cpri3@(V8zSFRt%zNiS+_f?nz|tLrF9R=v}G&ny6SJG^tqfTYFKL?kJnn_gB9vGy$$b-A4|1g|6Oi!Irnxl?`cKYF3g4F0{&#u8HsWGXe zFypmEMd@cjRU$^A`h)fa1W-|>sVb@Os>j}KN&k@$hxEK^>CS8RtToCC)hUfYU`)9_ zwjLrlZd~-das>={Fs*b;`5p7jF z`_l_+UOm?X$sHYDR|$4#UkiM>ZuTo9j&NXYcs@lbaql|Vb|s8*IJ(aXasCr)`6}G> zX(a001uM0zBk6@(`)jnV|smL zUfiCeTY@+H-tZ6&P0h{_KHFpudjazZ-|rl^HPZ2OtIN7v%nP#7TI)YBqt0D&z7R0K zU1$AvE4h-yqS_gFpOfdQcfQjub{lj%b~NX|ii(ME!?WKPMJxo_ePZliDx-(XtO-lo z<6yVciNQ4XxoCIin%lIkC*oneUx&{uJ+uTxRK@*q;VU=peN6O)SPJSououTZx2>K} zO&tx^5?LH7q1ma>M748{T(;xxIw3oPaW#v45F^5;$~OJ3#wuxx z7p6m}(Nv`qOA~!_=+$Yh(Ww{wNo!Z%-jDDYZVJA*<7WIU+_mK*1i#Alu6HRF)UEkU zHiKe7JH_X@sW zt00IOvZ{ClMlD{eoYwzP|580hbpH5r+Hrh-S-Tf(xVhQMN*-U=+d;AJV<#tr`tVQR zMB>00$dp{_)^@WK^ms0JAX>q+dEeO74=_)lIDE5_$w$EDgwSmBnXYHCcdJh;^OOlL zeuaA|mh#>W1llp8oT>Pb5A8_xN}{pT>y8b*;mTm?j2{!am9@gN*Z*zSn#HeQvag|+ zV8yeli>B%rEb7STIy(9}qy)q_2NxU2yXm#aEg ziARzLfvkH=#l5;JrQ zhhNA*K09N}DqCY%^STFJL$RrH&e$@gR^8>9CBi=#m&3E09TN5KeOJcX=L43WJu3`X zC^y~nm#L?B?(LxCb^+4KxFgraq3~%l3|V6SFNKG%LHG!>bJhP8By+*G4IQKOHacr!$Wm-u(7vD%Y9iC zzs?ls;tTMgOiTqs5`&x@yrw^zFJIBIyWB9^On^JM`@Tc>XA^~^0a!J~szM#M+(O6g zy@bcfdPhabfDV4#2(V6bRwPTZr^58Od|U?45MdU=n!V5n3vl$=ep!*kwCt>)#rF@^ z|%oJ~E#aS2&_T{}U{z3{HD%Htd(rcWgVS+~VphxvF0+&0=WLleOZZ6Z*w~ zlc9VR&(%_3I~J{CoA1pELPY}BUlZ_SyNaow0;siF=YE7k}h3Zdpj0^;mV}d_CDbw9W}5H zyySOtFtYpMFqPK3xLNk4b%#z;2%qE!yX#58av1ps?eyk;CVdqE?dYRd6zJOc?sW-hWf|8&0`6n(imu%MnMqez*UT8;a@gsE7ZU5MP5y= zXOJ-CV;|8|nfDVs?WrPz-CZoO8%@uDwt0EBmbJ>nUiCofXBwK{`Kg0!wwHqsdY$6A zHp<}6f2KGV<3O-j1Y6ZNMM7QBZ*lLaOlXPuy8k;{;?~N6V1Nsv&hmN*<&&oI+uxy zLJONDx%(!n|CwI`T^MGBH`T9<<2<1`{xM2wtvSHn0iJjG<;-;W%tE}doyl5JxTslU{+$3oWp)STD}JX~^n&b+Ozvg}Ux1dPi1R%o-xiO; z{>;dPH{h{oWWib0-@9B|w?{vay(KK@v*82X3<(4iBhSa|&K@p--xvy>1!EkftoPOz z4M&Bnu~s?%UPeLGjx>jeugKC@BmX_`v-9vV7TFvK;irEaJCf*ZNPR;A2(PC-+@ z*D+{ar*|OIG(%O*E?Mvyk~AO?UYXi!KHgc)MpJEHf3a73G)TIiE3ny>`{i(P64;Uo zbGymn7j;&Dh~jg0j{A^^^IDx{hmLK@>UTEOUnDHSbe@j za~QS@kN7Xu<<-3U46vueY`l4R?j{-YhDjeiv65LI4H+OZH@K6hL8!WEY#hg9VPy*s zCFvynii8G|9*|GYA&lWr4?;zX+o^E3A1LFl4excJw1J(4+zN(V_I_J#0P;w&>O% z;gmwLqS5z(M=~G}WxyFV??+P&X&jAA4R*PfI|SI^N3nBttm_NaV-^fX5ViJ0tD!A; zLTSEKEshl;%SZ}6khm9TeC_4D3)B_=WpL)8c`xH4@j`7tVD;G0qFju8U58)801J86 zfz1DF?>p>$q!*v7E*vlZu=$Ws_yJ9wJ)iSTABFZfBx>1lyxEnvTvnw4{5*M??=7au ztWVZ_A4HjY-P@0xF<%%)_9`pX+cxWJpRw*5z{w;o4=SL1X`RL5E>z;)ya`PMvX;+L zczP0E$Aw15a~A8nK3FoB(i9tJBdeYM#z2p=Az7o!TF0I4@#sak+o3Zk`rI8uUlcmIE--?i>;vtn%_sitJDyn*Q1a>NHR`Z;UmB&Ex zRvpigqv_78`9Q-D%KKdEsVjK~G$0WDsc>Xpy+vZ%S^)O@AE6@(+G7y@QifrUB^_?4 z`c8;NcJsL%O&V#`?V5{DWc04zXplHHV|ibA zHf1&(_y$HjrAZ%3T8SFL%@@W1djMqQ(kQfxJ z%uNuH-0uzO9#hz55RS-Laa`uiY(m@IX?XMpQhkPVJke8b zWfvUbVaS9Ef7VAdMvv-y$Xjs1B;3OsLUyWSghg_=`Uzh*YXCV_x51qmSP#vVK(9yG zU|ze^R(5l~T0uJb7cVoSHyBfcV*0q9BJmy%Ve3nYKv2IR_jKiBTdvZwEd*WZch@fT znq$Ru6L>4NrAu11CVYF1j`vPGvYx&hO_+`r9c`XMe}txm>L&hBjtj}B@>A2ghmeX< z5J>o=oLQ%>=)qKaC^ed(R+Vf6?uXlNikk+`n7;^d5R{slq@o$BAEw$WsxhjF!3;pd zbebXqF$t;6fR6bua?`u58=Q;CTC#`>*vfKY7Bf^Z2=0X*jq^N zPd#ZOwgrS{U4=5XL#8te3Ux0(3zPhdsPP?&U#^!B&h7~S+Z^~|gKVz%1G~lZN!h3F z?%1R`)4-Vu^B~<)wXztq4C!lwI!(x|^6so}B%gOSBar#cDUA6F>o{xrA>43vR(9Me zUmnX`|Gp^Tb;&bbis$*LQUO%yrOD>m#Hq=?4X3hXdN=yC-n@o?y}7y(_i!7+BEAC( zixsLHj2hkke79-ws!s>xq%iBcCTEPL-#>>KVL0k~E||u00F>KZU7?4vO`N7X$Gy@% zwW~-lH}d8;{`)4-?tq5|SwMY-9Dv8{gmV-sHcR8~O_!&JV$I-*@G|xye@Kf*Y8;?@W#koMGVZCF)G@ZU#~^x2PL~hr9E! zxmygOHxP}z_G58c)%GuP3uXIgO}frQR4o;EEnbZsX!lmRh>l#d(%0#Nv|}^z++!j4 zvWx41x^``6VZYq(J-B={Xha2i+XH*y-oGa#wo78ER#@lREDESP$x*~SV(6^ohhoq5 zp9v_5gHH^SwiGPl?8%Yqa{h_jbO;3mfr#m`>)11Da054ymCH<#GZ1OjoZbg$?p)cTT`-nN^x>yJ)+&uCwGS z>W88WyehYkdlB73u_w3)K0WzXew_il-!wn!yZu^1liz-`wDK&Sp`vBB`7UhI_69O* z=M1mT?L%5T9*)ILvl>=+J2)#1WtWiSwRiw}+z!L4Z4=G)AQhxJPI*lfryn2eb_*&& z-kf@#rDDt#YrqHHhp5;5a*4v)&TB<1zZ6 zz4GI5G?YJ1Uu$wm{42G-jnKqz{z`+#q6OGn&3+TCEuz`Er-hh}HXN?C>}AO_$e2_- zY)~CHfxZ+Dziu>tAb|;4%J11IF2o#bH7>6glqUXu<&oHE_b^(lcmAR9<9B7WCJJB5 z%r~U59au`-&%+ZZK&BPtJG8=q0RVaPwK;J;e+jtYOZqo*hmofl#(Cw`s3XNOHj3H- zU#n+<SPOqFoDzCgu$w zn>ptoG8LFq@|X3A0QHW^WwH%M2J_}HayAXxF=`NDG5u+-JLVWJV#Le09NXu#J}M|@ z^bO35eAY`_BtWZLQ4(-lfXL6=fn{r%_&d5LcTKD{(iO~I)eV)^RocMY+_oJ#|J>pD z8m7}k%yIwmx@~le)T8aXff{Y`<7Rx*FC~ATbcu_9us1UR_GYCf%x~!^%SIkZofPcg z(pFj%-y||ok8Jpy!#{)(qQQeKf+qm>CNv7X-1U?%B)|dtOsQDf#;m)CD(c6{3|suq znUwPl?-kN6QkSW#ly?Cexucs4jGr&wrgzl*++XFuUIQ_YEg?ok zp6X=zu*ruK5!DG5tOPue-rf_dr`i|I?JU^LmIROlCtqUv@Z1>akoppg1vyn;HWo0N zYfobZM`2O5CMg*?O8A6iS4Ml;WmsT8@rfSFBp$nxSTt(0*a(IUQuL$m+6@OdpT6>3 zvWg$Ft=t{;qI#F2@z0vu1^8vsBot{|34yPXNZ+@XX)DEi0Z<8;xmwGwHlP?2wB`Df zS7bA?^Rrsr>vZ3Kx_1ZdHR3`KX!(53wA3}K zx?)d=hvwAl^RSL84<=-p1Vz`;?cUW~Q_UU2?DX7jT#60`Znx6SbUN2MJZCJl`7$qC z+j|y9Ynpx10)#s9;nhBu)i<{MRT{hyB`)T(l_F*NL!moAUh>`7sn1o8-8Ow24m7!5 zUA|uWH02p2fc~I_KIZ~j;a!7Dv>(_{J#~eEJIEJ92ZQs#XO5i`#`PLSo*WneJbTZ?v6X)nnbKdG~?+php zE&gPz8EsEp886jZpb!^$ZEo1|Z$^8Wamx{nzf17-Jm#C^f}t%vEzvuy=H3)cg*22W zaLu>`oLLthmg66HPOk|Ih%QFgF+=XIPGz|c%S)+wn=((V)SzuKx5kAj8Fz0plx$RQ zV?&OL3o~7Ou?OrNg`KW`3f*Cj-jRbSTExcb@`F%uQfa2C5LiR9Po8np&-t`_25)C4 z?l>^49W=892MXWOjg3tw;KbRi6H%+kZR>UzsY+s;Qi_*i zpo^Z_u2B1{HVe<_!%z!TVNtJA({AUCY4!}Qy6wmS(j0PY;f~W>ZXz**Z*jn^|B%F{ zoBD0V>edAtHHf_8Be(2l1B6M}Qv<(H{kZ5;?83#N*4ts_REE)FmBwl>5beT~h?^;W@2=B;vzg z_&u!#v9vj-Tn)Vc?KN68<$P|tux`EX`_sc{tm#-6*Nd1A0)#00O$M=$`jMB-=5Nse zJ>1?bAL~~G{>k>np!;>Pp1wP%3-KqUQxDJm>mxam%e z*K94-e9`KpUbpPnXs}ieiSL4GUChN?|3`!Q@fj2pW9)R+u%{phjUXJmX#PNXu=x!4 z9OMsUsO0b=IZMXeL~v@?7`Tz_Medpac?MOV3?`#(itu+@V7gKntwgz4o6p%C-^#@3 z^RJ;y-DZ$RKZfGS`wB|OMixn!nNj~+LJSP~-NDaWD+u1vXMe-01~#d!RQNL(u!-PV z2L$S#1e5(@8M#Pn4AO=3=^lTyH@$cDW?#BCA41*@J>pz$b8RWldd3#pJbR$zX4T1; zK(@0xw$rj3`LGKfY;hn@WZ#`@=b07|GPUP`wVe)fw!Ej^_{NjbQxTkSGRT}eG*ytF zyL?v$a#1Yd7rmO<_=f6|9Oc4-F12~2%{;ZML>-s0a9@4+dbf0Dj3*eR6+1i@c)BF$ zIaklSGLYcJgJGOvPl%81c7CHnV2IZPG@vO^S|`2M9CRrU5E!O8;tL=#oVNO{Xg_CJZoGMa)_b>NO!PT|Ai1 z(iY8YIIeI}uq^v}@Swd=9U%h)#&nTs&ScO0Yb$F1Hnj|YP#)~O7cm(#6j4*KyFU{7 zh0x?JO6y54x0ud)TJ;CDdhr$e)tj{H9t2dWe-vMy=?j@r1JKW^W4ShhE8V%YMw$X( zLyLZQsua_m!GX({9z5(G@mGe!e(rie8vl$NT`6R4Ox_EQMA|avMLpfm$K=+V?OR$ew4?nz$VU{Grjcr08drq}GT8&)C7j$Qu2nYEu(g=KsD(P<89GLH>=B?`m7 z)cRo{kB*9Z2I*FIzM)U6W{Gfp0(+Ji-8Rbl(h@zweU1=+w&Iu1<<3$dXWSGrl5o}G zug6+L66Vmto-wHpQpOvdhwrP;+E#A};x-D1EN~s>;e}F^=2qI3D5)v5^1xl}p(3jb z>SyU480BCb$vQk+Wr^cN#*=do!M`I~WJ3nE?THbR?)_*%MHOvI8d&hDR=+UqG4N_z zWW4<`AFN7LSN|<;t|pO-f6-jFh|>FfmbiVDd{CS_yh^yI!d~r6R>Z)gR}OjsC0ni| zzVVGaHU+~wGyKPYZoN4sXb6|^wJMX@Xs0q?Hq@Ux0M-tRv~#{AD;{|d6UElhYe?Ta z2fCC76{gY8Cpl&t%gMa@?RMVoEl@kHToMv4x-8w^V{PG&x%FJM?GBc%M4R#W->6JC zYAkMFOsfx&|L+Piw*w-@OAdrU9akM8E8Pj?)`%0ZJWHYvAmIvR0%?adS~I9NF_(Fh z^D}y015|eLe!yz?$R_hVAopOvf}_7Q?M4bUoDaG<^nVysnddxSkT*O(AHJ&K6&fkk z$DmlOIp@#zrQ{mp6H4DKm@4%71iOnTAO7J*fSHyq|Na(Z^$o-4fEG2k%Ah8U(3Sj0 zPH1ZCS<~$JfpdVE^7lYpATe(0z;yKsO((6j(-sWiWO5=|%Hr=`%dsPge|(^R;Gdv8l_RG477WY^)ie(KmIC&zkDkjTRkhl7gW zP8=(78~n*TGzBl%)Q*)djOCl$jozusc=QDbr0;uvEN1o2F9dV__YDZx=}R7c{e7|( zF@Gd`16uEV?J0OtDnz4pjkn;e5Rg^ zjA*BU$IdJ{Z<=@=I1@$|9QwBQjyjNbpg%ll2&wERB9BHtD|DfO8TY*~MtsKHn8rbO zU&-78cC5wqOVR_@lnAq7*@G30(hjccqC~y(S<~7rJh6D;QM__7PuoLhgUBa|M!%AN z8<5HyV1`$Lmv1j-g}3IAm13gF%o3;kIfFEaj_9b5*bH&xx1LWtqn-n_@ap7U?1`#m zeHYBfTan?$9H)D@x5@cdN*!du-l<;COXVk;>SpyBpS>AMaYSU&3`Y5KUPhR$n&%wp z%Wa5%7r{w^Y+}v!@{Tn$5+b~1S!UlSIuGsNDVG#1<4~(5Lq6l}{pnSC*f2VFmS)2? z)oN1KjFuS3;oXk+@~?WKI=7VxHM+QYD3wVskC#P>2TLa&bI37I>%;r)SBunPRbU)s zaB}zA>qRW%?SU;)8JiyIthsE#WY+b-wbhf=&APrp$DtjqOXXQtcwN#Hs2&656610G zfpM+ukD*&8=#LvMZZZB41zC*{Wq$w{N^Koe1?HQD6IoM%GFD}U*jH^>@}isY2 z%%%*P9$xF+BVg1hPYEDjWCF$U)B%z72}kv z96KB!bv%yb)r*6SLUE<7#iCl%_z-9Z2X8ACRmHuUY&O#>F4hdJg+wO%tE@@)RFX$`-s>SO2|Gu(6 z12CmucXqm#@OQ82iJzmIo{o&VDQ1XHHYEO8zVAL~MiQDJ*+7GM#GxUT>NWH?Xq;K? z3T%K7v=7*s$}^i77Q}RE6>|Q>zgm)9l0_z?8TV;54({;mKxN_gYsTPlyeQIfvf)$` zvX?&8;u;QQQ)sa;07EXP^ey653G zJvW$>NHA5X4|4bs&!jT^ItLLmhfzpsl^-8mvK_v1ekot8M(mtIAJ4)7EbVJ(U-qew zh_`;Ypgc+$yQSmz-A~j(MwVb*DZ<2^@v1gOo+o5!=8l#AJZGKT z=*cTP-isoIBcbw0@+l$Xy}_f zpG4*imLODIXHhyC#2UlRLFJ7SqcdQ9VdZ?4n3Ijxazia1SuWo6i4wL!PT6D-VsCq& zKz(|RGMOd2@Y@wVg^agvo0);exqeX|k$|V{VqJ^M4b0=GJhO?u99lS5*&inu?0^DD zg^4$!pHY4J{>hFBczz$_F=QKTZq-05wOp!Ay91@W9pZnSDjC0p{$Pyv;%!I(^weCc zM4gv?9k9LffF_OihbTR&m6{3-)^O!UBy;?YZTn1pFQ05$hs3)&<3jlcodtGwr$u*o z+7HM)(Kt!PtwHW3{pUZ99*_}_M|oj#mSH2vB~Pl>_3VCl=0>a8F$4QU1y5+li?O?O zIol+f7f`ms6`M9lWolLZraBr!ND)71;%m`N>u3~K2)(WliMXj8SP&A*XeFQe?M)9dhj2+AFU5WzhQQ!j?vZW$4+V$-r=Q1$z^Kw6Djt*i!neR31Zm?3)xSfN_ zVjqwP^=K)#C4M8Yfx;!|dUs}rM6Z}vN9z!LaR7>TKK?N|w{UL`0moMsC`1woYP_9p z?Y~4@ZoC~6j20x@u571f_5Mx~v=z^&iO?%{_F3j)|FUCNj_YS%H6_LK=tX@L-0efR z)i`xMAC*Hx(^SRy0rfBi;$|g#47?j zqlgz=#*2u}9gff2N#)>TUG5Evg&w$Fo}S!U{ci#(=fhjL$_ zKcG2+%;SZH{9IbLWqSRDeFADClorcn<$@XT9w+7Ejrxkr6>3Irp9&lnwQ1=2(fnpi z0vJ!a!`s6dBQMelhaMoCc|cU{>S8v975JHq2jnJqI~Hn`YQBt=fQcc==qNeueO_O& zVp`_^S2fqO)pv_Rx+A^3mU&`_S4u038F%+-{6}EFW2sjee5Z9AFvtZxeMQ4 zwuct93j&pVlinY0%2X+R6k~=BD^l?nwwqa_?lyU>P&+w+QjGjN7EOF0$^@V@Z3EPqc`|X_{w~j9Elm# z9a@W;@!YF&XLVCGQpKe*l8^Ijk%Xl$1lMfejmcBY*_dz&0P?h!!(FpZe3J{l>@_bC z{~~--6Ky$DW6`v)kn(@Wev%{z!0ZGTk0)!vtFcr8ZTJ`dL9WObk#8+JvsjT?t?s?o z-cN4$HHYmq7XGrrWBe?6)Woq3bI+@GIsf9}Fw2;}T#w5j-FSP1Ds}U_J7BU!DlW$9 zJ6b{KRZd2*hem$=A*6D9dAZs2qA%1i*XMM>Z1Q0u-eW3Cw{_S>^#RV*pxOJ)()Ot6 zbCHf_%f+AX^!Kr!!cESAm2lqwR2g`Zn(m(~4ikO)hsGeD2dVSiwI_Pnx6}K4M*Xt5 z3m$xP@x4?!(Ye>V)h6{LEVSxpdT*# zlBErFovjqMj4Oz4)$gXu>1_ zkC(G8Y?$%d~^m56I@%*$jC{A#)46|l1%Li zotaM!40kD6pOK77kaj5W)>-9`-N<3M=ZNtKR^@(PDABq?IK>uIv>(oiG?*lL_I>bH z@Ljp~JzvxNP!ZOt3Q;a4HPA0QH?L71Gl*2CK4^g{)BC0wZ?B*Jpr-kg(*rY;hH?@^ zT9wsJL^C^}UAk6zooJ18`q2*|Nh)+hJ?v&t&#PZ$K4F$Uur%Q3XRRPw49@|TdDCP+ z(mq}c_wLtk=Cyxtw)eUR%SLRx3)FFU@DoiJS$XLQ!pYdZ>09CS9-t%faqQK-s-qlhVe%lB23i=F+maHnU zzWZwpaoUtWJrzUPnJAba_XyLUKA6T=fk){P({zu06n$TvbUth+qR%lowa+Nav9CAp z)_0vlMhKVBCyw%{1pCTnET?CrmPF3XZ#UfQG$#^K*T*xK6-zs4!PZ&?(Al0OXfXD0 zO^R(W`D{mx`EWFYYYWD7JpeBDp`T1v*KEY8m(4^Ye0QJ)zc(6k8tT6^6}ddT{E6rH z=8I}YWvC$jG77A3K5kAU2k8`zd}Y&5~oaoHq< zSA9fu0q`(n(m?9e4h=z!Q9WrIwm|^DaE4O>gmNRQuPPMsFag@4Jle z`ilK`S0Q0I%(T6kL|Fq^>YnZ1(b)C2gX|w(l`Mfku6*053=tBZt0?e5-BWWIkhs2z zx@Z7X3;yfEpdsPke^$}^dD3o-@9MTXq}n|#P>u-k5URdK65ifOKnzh^r4#5VVM1pi zgt+Gp80hd5IP7NmYqn_q$8O(9w%;a=Xv8LSV?yA1TzbjcSvpel7L>pRzZg{8gCeDx zyA`*30*f%LXhYB{pn-Yl;OF2sMeAnx$!3&#CS5f>L#ZLc8@}qTlHq&AlWx)$sJkf< zooFhPBwgV&mT)in&Am2tpL=k@3|w58@niq)9+)UT-^~QivUJ4Mm#SNVkOQkkGeBC* zE%OhJGI8^609&>4Xl2B(k7=iLP%fRmFcZ{5_(Ad=?z;mK1R(znu{zRypVQ`!+@= zva}5=K?YHxAc{x^$&z!BARt*JNJb<{lpGw8AUP_MlOP#~48oA3l5-9M!T`g}kcTwz zF7`gp**@nyXJ3B5zr(M&SiO4nT2)n>K-JFd3^aqa6Y1Kg}HVJVMrPBqInQjB6rDI*i-ZJSBnxHe)S9L^a*GJt@*i7ncn!Uosr`BWpD3)uMI;9e1x^A%`FA zCiTq^t4aI8#8!b7@7Jn>up$hJMeJtJzYv<#z3{T{C&Y?t@Y?&jG(O@pvw594_BK0# zYRz8i*mbtlrHRj)U@vDyzst2)Z=vnwT_8~a7w$+L*VcM6n_k2f&+1Z5v2C_{Cr~85 z*m+!3wC^*1&gZqL@%^S_?nVkIu@n$P<0l2^P8+hg;x+8iP2kgNd6BA(u4x27ciQLa z+aNyU5nH0)sD>9dmUl8vQf-*ZVN$ODB;f(1^W~cNquG1Ym2h7yGv12x;NCzJk*%^j z0Nv@yE2S67;%>pdnx~%a&aFSI%Qcq4T&!EyDcoC~2}>e@VuDXyVnWIYbOV@}^f}k=@3JrxBQBu3TSMO{vkd-CEE8nIL*T5gIEA!M5USEuTL$?+Y$Z&8$Rm%Lx;& zw6q3rmqf2N^Jg|(AX9yZ501M(YEWg9g_hLVV>DunevkkGavw9j2qUONU;)$QDk2wE zNAZNaKT&tVnBT;-|Jzr+cSeeM=o={2GS}mQbJY1op4Ov?K-!eUv$c3c!vH zC>aOZ>J^r)@B<@`6f0)qHmP0%CX zwRz^8#Xro?D|C#mNZWW}RaNiO^4`6F!SpYa3fElNiX2+?+uLS? z^j$#Oc5f-vLwTdZVztSyq#ifdW_fmEsX>OHj5)=Ajc-qBt@64Y89!Z{XQ$D3#!aji zxe-=*H(6~aiBDX^6Oa-Bt^iwBE_IC`C+@d64%a@Rgb9LopX=$U$J~VnYfhB61-E|B zX6k`Eflr9=@%Mmqz`3JWbzck400cRnP;`pkjj$C7jSoQ%dIC#i%b1}YOtlGsc9$8L z#q+3XFbL7?{V|m%2F8^7I!tp1q$sQ|O}){)DzS9N6;z)~b4mX^}Z&GxYm#sD;%U zySIRC!8OXN@j0H*M}6|t2<2HWEg@}GJL=9&+6f>%qUpUAi4SC(dzCqr3NZ0WnweS& zVlr(hfdX(zrM=7=C1KH}a)z$Te-ss#WL$4mN@L!70e zaYY-eapyO}kuyur-(dRXqf?&6utl2{F|*0G%>opHf{1+P+BDutozb)HNUuycv8&CZ z0iuCEE^gCo?%M@uzDt7>bZKyW4!6cFWa=yj!PAG~UrQ5woL3};yu1}X#YgkFigWK! z3k+{PC=ZkrT0KstGyCW?ezQ2JhtBtreKa_;5nP=9EvFH0bH7*TYb$^q$57%LO)_RD zD7x%5S|Yq%L|>j(utAsJN_MzEIm^R9*tEY(roUPgN26VRXg_9}oFm=_LwQqeGu_c1 zTSwFVD-DSzE`th{=IG3xS2p~bF9^+NR%)p0b6Gi=ZVVv^W+>xD-dASocJL4$OwWy4 z^y;RJYC4XRF5X`RWBLWrcZbK+aiXjHow+GHqbsu3oi!<(WnYMkIlPDGm*@l^EnQh{ z6(%qQz`AT7o53%pwuc?`QKfqJJ|ga^#D!cQi1REp7DxZ(XkKgo)4M>foNih9X%tU7rm<)alp# z!3MTRfvZre{D|s`6H4cP05_5CW%w4W0HU}|JZQA>jM*^+@)AW7Ln;aoH%UFytlegj zwj+BEWY1~?&!Q*)*a0>GJ3#H|mM6L54As4H%cDTip)0MEBnTjzV44h%e`6Ip%Dt&l>nF~lIzt-S=i5*GWgRBe6j1YG8%7fLg7 zVLnSWP?jMi!Yw}Db$p47xKLlDMs)ee<@$Mc&q8RHG4*auqYX314oSI{_iyg_6E|!m zg&@?HJwGf5g3hDTR`*WoEKZIq>jqxBoyQ5Trge_esE|8i0$Ao5iQ zYiFVZ>DuS8tqXU5$25C*e!UIa#)OA)4$}{r@TGR(BzC?(IaP~M6b2qreI? zQEaPYHbj0|YDKO~&8C|eWOcKiq#&b;Mn>R>Z?@0P^fd(efV&{q{rm5km;IrIx4EN{ zl*=^AzH#nVe%2=M`y~$cD^F^L?LPD~rKquzN^#h@%NS)}r4lj=hrD?gZoFbzu`(YIX)0?AKxGKY)EHm1fzLZBsDV>O^NI;~U#SrWAkBzGi?vuYx}xGG%LbDDW2w3C z=k97?N5^%G_x=q-z01thD&4J530nZS39(lplgB=okYyX6)IHhz+M_T|<2_5I=VnfD z5@tX9+e8#S%kR}_KE(Q+d$p)ZzNDeQapq+-gySz{_u2g5UA;LDLC#}!C)(h4IkwGm zi7qG6lRIZ>z$ZD;g&&(BSk6nIdH0I?V=%A zx;?69I}-25Vt$*;clL1PG+5Ug@aGExEk}42AB0Ed=}uiH67_?~5J??3oKFu?9`9tL zkPYmN1t!pNC4=f`c}iZ)wHRzq) zZF`e@wy72-eAYAp8T6e!8x1;NG==0Iyz^X7q%*zBx4%ft>zTg8{j?!1CqV%14PU{= zJ9nlH;vbX)P90{sr|`2c9?C3Ui_)Jlrjv=+H8vn#wxZdZJFX0Dpy#G!m;w$UI7iw& zNh}CqL)}T*BFd2V&-z66M@iB3#F|U@dPa7; zBGPKDB(voo5!2WkiL~BvQYWvS#XQldve|f$uN+7Z zxdJk4XoGCD>qB3YxUaHfr#nulzSe0t`#Op^%vOVnCf1+tkh*_TVR$APmnp`Uu_8Tp z08dDCOZGjo4&FN(N@Bc(N;D1%lMk3X4wd-_eCD}oo zY!Qk}({4WnuBH!0?WmtzGPYWd$QphafVy~#HQV`qpDsq`?+7h_t*PD>Ex-=-`Ee>g zBL#R*Hr@2`;qg|@Rm}hv%SID}Wn?^d6SU=oYP#6Qh2`>~g`?PEBfdsd#!Fk6UGl_7 zCC36B3=Z1R_a$62F@9uU-h}08g?EIsaS_o~Kg$*)$xis;GWK5bM8WJ$BOJCpF5)X3 zOnimx&^QF2T~C@Y$@6&LL%r2xbZtR=REAvrDqn&!`+r%%I8XPIs|m`qnkT zFczR%F?9pk@hDk0m1T;Y<~~yb=8NRQezL(OWh+^YF+kb+-4-%A$>!_Eghv)r1Dpj^ zZ@o(;a+DC5lmYAX3W*$ZaWA!mr*dvzPoaIdGg=`a)n&drqj%aOS7y6DsUJeo2!UoVK0QW`?Y z+*_B4&oCuqII5to9Ky$=8Ce<|y-My?N_WEY3@eEcu?Og zge({>+8m~?QdXSn<96S6-Z=!-0T8zCx&epqX)!|JP3FwFgUQCwKLhd~;s65JE7*9U zcV|7eAZX~^Y;RM;g4FYOMm4dw>9d^(vxWCmfc=xqgT^+l$i>J9AU9`tqDhD>WbjiX zKZa-7^OVt?oD_Y{{JWFGdYLA&F?DbGES7e-1E7(CEK%n?p1s{YWBKmo4D5=%HD=ae zz2cIw)kwG&eCnarnr-1K9GQ0XBz@t4$7yttZ4+=E_sJt;F?0Cxn0Y<6zBo0Skl zkXgRT<@Y`9cl0i9F29WM$Kafo-M5i^|5je4w0HfrKiXSZaHZP1`y_M}$d@(=*-;iYMqo82>$Q~IF%`By{HMr)?~hna}hn_cDC*0{#}FA-)CmI4|~PJa^mV5*Gc zkdj2Tp$5?1cf5p~`t#(#iAS3(-3@o}dG=^Iaup(lBUi>~kyvRG(xcz^axXx*zy;Gr zx1@3porUuDk3l zZfmU31QXwV_m<9|<=@?4!Fr7C$A#)upfL09T6Fim!;d`dN2!S04m52{E=F6a56hY? z2T~F7bp1wtTd8}Nwgz^}C@`ALB>4^Q^R4j9oe^CqwuctQsqc?_I3{9}$%ur-pDYhl zgYbJwl^8}S5ZC8-{wQhh#X|p(;bnj{_J{jNSH-Qr-!K?^eGv%WCpAtQYmMpAYuW+! z-up3wYay0VT>a;H3#F0pfF9zE(AczR? zlMHw3Jr6#5aiU819!R6iSr)dpVtti0KC%jP)FwBmKx*swUc`E!6t%= z`omb)GASUKNPKx=mAmGJ_di>d=yd>6+c&7%Sa)vaYu;)k_pZSI(El~Dj8x}jrU>%2 zmwk)3P5h@lP%kd}c{s>IzwXr+>|Si<#UC_kKrtXLj?-{>&iPbAZ@>&$Jl%w7(l9VF z;{q+@v@RaT-KZ(Mwg*7a2|HZMq|GzPgRq54LA<1R3=?v1Lb)5bte45`AI@CqkjDh9 z-qFH~YX%tZpdX5ol4=uMd%m5(IkBSwj;S`Q&t(Lw0u=Ttr*+<{n@j}ifE&Pk)K4n2 z-g`>)=BfH_yZ7xAXONxdWTZAJP;BOvJ%Bt4wdalsYr^S0F?&jB590X#=$5}rUgo1l zdFX~4$QG##fYc8vO!K(qn}c^!p1&tt#;!7{AqbDL7^8i`volQOgIi#;5n*mqxg~xI zY?D*GaIF`F^(Y3aP?YLE>Gsv6D<+^v3X*)r1m%5TAq?@z&h^@@8X9wR4p~v#fnLH4 z*2r0&488DrhHZavI$g>DOT9wDmvnl&-aej8!+)c^sC=QQ9t(a_16Ed}HD= zA0uLxIU;eg&&tMp$reWL1&j8`INE&?4pK_M`>o*+2#?N3kJ#vzLyBAxnb$ zot=&UwK;i<^FFdAF0e3&CQgp3OKOIRB5Ip)ghHo9WoBdkv3WUxP5a#hk~V1#+z72h z0kiT~1^UQ=P*ar!GlUJ2k9z-ojJszg(bZdWCd}eB#9@zu_J-e+AiYX~bUs7A;91=-~mK zS}8^DJTKjSPd7(HLxX#j?I(!#7Ze;hbPjw>5BMmm&e~SQJD#j9xam%eZTq6bkiD$8i$WE!v}btEMM0R8 z2jeV}EqJ|DAF*#=J$myvB6kZk9 zE_)@g8D8O?H7N9Fs5q4^GG!>9PCa@ICS?D{#Cf$Z8Tijb3IY7J3@JpFT{biM>iV$x ziD3dqvc>Y(xTh<@MGllcEXEx1SDuXRayyJzrM<2P=sKRqD}I;o(B5IUI2n5#Q14(k zgb_3wsc(eL^zZi9ARMX^*Hk_NnS5Wn7t$|lRs!&D(p>&eCEVL?PfKZ_5ZGb)VIj87gPuz()z7VXb&(OjWtDCLq#H33T0@5|+351u z2qzBtV7dx@@(U()`YE-W57w!E0GVX+6|iqyPsB|F)?t1pW966V52VP}%V(IIb>F1$ z;(Gi|zq#x@{pfl_oXC?a+wDg|P=Tmt^8!(i(U;WR@FKZe*&NI3dzU4A5}NlO7d@mB z^xZlHU?evZMXO(ZH4B(YQ$$ZQ?&@Ve;SCE6%8jtrSmBl zJyavPr-7n<_1caE8D6uTV5_<+o5g7aww=_N*Sih0_o*&K!P1V|3eW_XgxGz)KT&|OyYH0B0J);!@9Dw_2Ha%SS@V~l}{w!Btd zjNcQvWAK+9a>7iIDI>ftCpg~|I8&*J#DJibm<)$~LVC1^$#WQ~#XE)CcNV?z@4bq+ z0S1vuA-!*}BLvgs@B{YiY(8`=Ru{3FE0V2$&(DYhU=^vd9ucssZeR63!aoBCCF^^p z2IVp$LXiOe!ART$sNc!-Wa#s(*Blko50~}~V3W^lJA=atpjyyrje9^jhU|okQ4HTe zj^T8L1vyiJd!X6Y#Wz{j6(98HYBFWN;sno0OyhiMn#aP$zFx-k( zm~yPH8#V1v!mbmW*H5l6TsWe9O*x7;#olHnPdAq#xuBI7*P&$uU$1@!>Dj&r*B%Wm zyqWjW)#geVxdzC3GTuJp6I zs3_aw>f1kqG;O4&`=R>N#rBW`Fy>0W(*+4w#%Y}u5{gPkrTOg9ET){FI3g))r;brx z9^sQ`Ws=S|l7b3Nw_h38zHW+M+3U!Oxgn{2*r>sN??Cpv|q?L{66TD$!8vU-jh*>A}!%@eRyqcqIlk!3HvV*w$! zJ1es51D!nCY4T|Aq)`(|64ldgC=K7UB^)J^q?n`@IzM;obesAS$ay}*3KmN-SetPw zX_A<5xtUkANEVHjl$z08Ymm}DC7ES)F?OgL^}xab5z|V)y>b-`XOn5D`C@wPV)r5j zRyBHJmEkKIGdnk_7H0;}aPPXaflVcN!>57lzHn@JSFPh>#`5XN!osp`!ys9aB`bg67#T?kC}ZiXpB*6 z{q2(UKDP1k0BM%B_^wwqrjsC19!l^bcZ0dp+x(BMTb`|_vPy~Y$^NYn5$+jW&ZF?W5jUzr^nrypLsyJzpW^Krj!a8akE z{v-%Q>uc3P5WH6Z;_bu5!+emu!2Otr$Ou1pn`=9ZP@ednYCWt;^~2q*aD-eI@8EVk z5_2LHd}Aoxbvfmhf2hl#NHIt;^43=PE|11XtB5BTDR_}9FUrd0l6Dfrg`wwTTGy~C z7;vF4FqE`)8*5U^$AT*IJCsuxRQ`Q-E){)h;dE;nyZcr5I3L`NVj~0kc^%yZr$MUv z9jwmcVf*ZOa*))7wk(JZ`1sVX6|wJj!nTPYu@eqTpk=%r3Gsd=yF3lDXg+9ZWDQU# zuj1Xl^C(QJ8PU5@j*>Kvw>cxbz2AI(ZlXGA3|!;& zWc1cPWlM2uN7<|B8XbRHW2TTA{AsFt<>$2jNK|H>h{W+vE%{K!fN)1JT!dAog9O>I zzTAroP4QNpHr0Jb?*ADyFP?_|I4V}foQxz`crS7^!qg$nwo9MBAl1rTg-CM$Tdb3D zTK@8p5hiX?tvBHyTEEkJb)_2JMr0TZne|i!_|BKgcFDXrxm*VBPP{#nesd+(QOO`^ zfkAKgM{P3?N0QXjI*?K`Tmr={lS}f<`$+5}DO${7Csy0(8 zcD5z;8hmto52vrP{-&4rcn|r8tcb>LTrW2j+S?+#w6?r;oyO#7*(6VVrQfuL<9^IJ zIe0%$s@|+a>T#%>J8~x7p=VMCEw%lnwL;|i=KF7x^4PKrl-5FxHcPuZEIQgqA> zyDrI^7o>);$76u6^V|6W*=4ki!QC4dNbwmN!)h9qe?tx=PxvqeYLBx zu=!)}remVCWsjqHt%+P)NE5x$wk?X<{sX7f0zO#muGp|`d-=(s-k<=s_HjnpM^oJ@ zi) zpXf8+pg34>-IBLvc*E@;=QSQ#U7RB0Z%b*QfDUQrJr##9_d>pxg`VHLnwUh*x|q$< zxkXRHyBYM;ll>Y=miI|A>UL##X9<PvgH*@?{t`%`Q4$edf;KDWH6 zHpon_RuJAezAXrwY=w;iz=ZI(J#FHr#!%(7m=ytuCK|u9BEVb-)HQvow2RWV$~l7= z`Hb{@?A7WMy|a|g24q~7u+5A*q{m1edY&tH_J+SmZF08?Pmbv_pua`B6eZlnQYvPn z7i-tB%LCOnWS937ku+gRcO6Nou>`T5dQ%}Tl#n?)-d`>IH-TWFd2l``0PTwYJbruv znYWE%uzz7vv({V`3GXsr8yqb4VGrrs^8R_EedZ^+7!g%1mGf?~)^0w*H`dAr(LWi! z)5K$$k{-!q856O+lE69ZoN4@Qzc(dvj`tDy-TsO3;= zMi?{&l`vlUb8bwa81p(L`J2G>sbOu}`#Vr(C|*I#<;wc!66>!-9!mT)h)SXUouRp` zp9f5L?}kV1tqbt}A`M5ta1`tf%foc699(E#rD~08)+lZoS_}iSq4{SASFz|GI+bxZ zHo<%zvXK&R6@hfBsp*C)k6>DpbHk|3VUj5Ea4Nc~@wL`3RC=jUE4Tl7G|K{t z)d&mZB5*9xD!T!NoE@c7m+;cFf9=z{xA(=!{MzRvGk%Bpd$pOB#OpX$TOn$~*Jm_X zuEH>))SUKAe>w@5uaX7wK!9iZjL2PaTOrpqv_&vjJX(|4Lilc~dvOH@YTqM`M(51z zGJbjWVchc-speF!`k(Er4j~k^&!M}!M0qZQzApN;$c|--#2BfamnS`o3(x2BI2>3! z;hDaIMn_b)BaJ^JH4dis$c%CP|8aIe#1B}v2Z;t_8D}^T31QzBtvH|9o;w->vn<3%gCfxh}F4XF3>C*GgPb{}m>YL1z|8lWFDjKGzhtwO$Q^8!nlSA2!&VGZDg~rp7 z!7uvB6QxoD=?TqTjW1LGj*e(N)3Sc}mh{ z#i~d3{?&CpIxN2k70dstfVsN52d+Tz`^(o9)ok%9Jb2#Ee+LQBghBby&X z$YqKs>{6N|n<3jQzY_Wk2I(vameFzL=DnYro5`8W2@ysv(Y+?hoAp8pU?@C6i`RTo zbCC%_`-?LF@>%|p%Ad~)FtR`N|F`acExcR&0IiT_k@>LD9`|G#Vb&)4*Sg^L57+5h+X_@_dxZ- zcfq6_-hpN;)(adE-S|AtuJaj*L=p-IrTm?E{t{TrFyqkTVrIr-4X2xjTkbf!?!=49 zJc+65+){38(Ko_h^y>NhQ~B~KtnVuQb=TiWM;eS{EaCdVezuOUKUqTT%O>KBe0x67 zq0eGd6>2qcw9Pz56=2xWBjAr>{H-x&Nn*M6Jy7*2gPW?g+xOdiOWE|8>HugcRCTXS{E(S`pGCZ;aX-7^0Y%!-FwprsCzCp9#I}{_q`7ZT>4`> zcm~+N)hTg7NXtN@&*-_XvC) zc?p>h5^}#Knk)XpxXXtTP?s!r%gyjykDX=Lt+V5t+v_b>qRdm|g8fSnBm}ZN5xEEJ zS9Y=NC*9p_8OTxk8&mSvB~f|WHp38OA5Ln|gK_$yajAzI{9mF!L99I$aG}|cUK?~J zoo@lukUJ&>zs|;A^#5O_H$#ym<>8>8hPtblKg%tMdfAN&y&EOwxvt^Oln*oRUo`1e zs(%~+*=a!iHH^RM4@7JY>Y&E5S{1_{NabUp{(vGC7V<0I8!2yvh~~T!`^{nXfI#ja zhqV&^W(xdzi5zII{kG`I?^zm9JPGj@TTVX5Zw`iy7;sdsXn*|85AcA2c1ruP`t#o$ z%rjb``PCw}-&0&#jmLlii_EC_&A}8%1I@9XX_Nk|;QwzKW<|a=)x9HO@wfAnnrYvKRX f>hu3E3x9z}9&RGdL8Iw)1^82vSCfOtmu+T{GfKW<#rpEo=ifBLHaAJUIDgrDo*-2d#f+mvq&iFY*tQ9Tt?=m8l8eUN97{Of zq<0;{0sA>GvmEMvbjeaDP_Q!^dxLTuQvUbOd7zB{`k{=6AmGv68n@tqC%N^3rPd?9QI8Y3l^Ve9M1dBD%2sW4Og*o(U@$JZ$rLwu7@qqKLfz2QD} z6r;p)Nrsy;A}J?tdg?T~RL}FEuT00;``0yeGdbE9369Tp9$kKZ;=EDPuN%*(W1M*} zpT1MNjCh!$6324{Blyis7=GWQvGeH2ujI#Yue|o^^zc1Z`$57}6Gob_dS-TSYq`8m zeRx<0=A69DDF=yf`?jZ_{qbD!Tai46z@#a1v0c4&k zb1a%zg-?+bdM=2S2gGkFaoQ$JcG*uM#L9y(a) z(E3C8R%lUl>t+soPp%MCouMFBh!>>t%pR14%Ris_wT(I8xV}C$eUCV!(_8;hB>@=( z=hjltpJm7LJUbxs2aDAEM~WwIAGW+YK5B+PM82na?Gd}vxdYE1X=}+}7RI3U_Mo#>W`YPyGBin<7^q|+KwgOj!5l9meK8*+!o@bBwgV6J)Z;l7QX34y& zyTcy0^@;WVan(-}bcUwZ=Wth**GI{&oCoAf=MN-)yXwFq_w3A#^5zrXLgUBSqrh4g z89quRxs$E1kgulK`F@`Ic{nS>+x1QL9i?yWnD+t;M>Yv!%dI<8=6CwIsI3iIYWG@F zs`q(+ww==8ALr>>Vzq=e;@gToMLuLmKM~5j(F+O@HB{=6E%q z=-}=n7%TOkt#!d_Svrf(8YvoO7}?(ogMWa)+d98zNM*=yxH`>YR90S|cUZl+zx9-q ztC%a8d9}he9pfZ#&$6$LzU7ehqynwIio?%(?i$XtZ!mIKr?UaAq=e_so z>iKrCPxadMM_z1d<6*SVXOa)_or#n8K4ABm{os>3Cy!2FJ)QhV1Mb{_SJq-Upag2&M|p38sL4Sjv}BL9nPZ(iTKcp&~9tCQ#zA@OVG!n5DGb9hIh_;rnhd0s31Vp}^& zykU0%dFQy9Ucj9T{j35nO7F7qMz&;ebRYL;wR$9THR*%%Ly_8J#oERnlDhdGs?+u_ ze__dsl#?^<*yreIBX!!KrC08(5}#y?K1RQY+x;1Cn|Cla;DL~jaMx$c>H{`k&If78 ze9oULTNMdXwTl?~nZN43rCra*(0zaE_)*G1cF=uhl#o&Onkd^Ycl8NY2-US3E|=k%fbR; zwju$qL9YkCi#oabo>WbdI?waQ;D`E;*6-jSxbH~uR8E6vF_AaIIk)qrvOTgDOk_;e zE+Gr7?ztB#E5@ekWavp}51QcakniF@78Iluz(4AsUJT3?CKN22siP@(ma}UsYctfq znmV>w*n447w4OXr&P8TANugC1U1;8zUdy&j{=%f%bO3s$u2 z756t#ivq>lU6suhdT*ZxgP153E7YWhc8;WSrA8@wK`0O9%5@<+ZnJLEJ);Ey*A{D( zZjI@y-?9GKa5V%)5npaS{^~*3)Rlv;wqJR^l7Gb}@FBK>%~;GJ|AuRo5z45lq_X5C z^a)8}f^{OhcKP-zNb@f_zvT90dWv!iGV<(+nd2NMtTl=PhX)>zHzOCfi7z&yH!1TI z-Siw>FMB_3&_2r^jli=gf0BQ$lzq7tIzgQ@9nm8!b(wMiNsJ z-KBAs*s9V>{c@{^W?q})R=8AG+xxbTVYo<_8F%^SyxzH5>7GQ-G|##(lxP7rf?E;< z`~Ws0SOF{J8T0z&#dqz@Pr2(?KKOo`9FjFb?>34TkcYW*^vA)?;-g7OR@Ctfnx2<_n=T~XZY^7vV zK+)p$A(tUElmxv}WmRPaJ4U@n-6T~Ft1VrkoTk9ppF}mb^2dqB#brV>;hA(z#+vU+ z8)cNZ8TxATOi!|>qhLrHe7&J6sEYn6)Fe!aUQSP?t%XU332mp)kM35|HD`u>OI7Mr zWOp-mocE%3@P3*#Wz}N)KRudyH2rAs>hkBU&ugELTywd0_L}0gi7zEzl1dF^)Z$EJtum1J8rA=U;q`I~ZU8 z-7KD(U6h@7L*Rv|uAUC~hE~j3E_wMm8>tI{CQDtzFrPi2_Gwbf*f+`PG>-YB;ztLL zk~t2X28&k4@u@-ws|{`)zBQc_>15)2eU0)fbM^7e&@1vyA475d8r3hsU(1j3i1Oa< zND1;k9d;Vg`}p2i zkgh?lMQ!sP8=X>}OQ-#sXQ#oI=PlDpCtdYi$lpnww>n1N8gv~WFuo4?iWFhE=;hJJZ28EQw%hla2s5h9`G8v zZRMBMpX=XewP4jeN-13_jSS#kOhnU7GA+ZizcxDHf}1v94mIM>je7@V<6A}zb<^~f z!hro18e*`LSgVg}>n;9Pylsty8Vm=Lcqp<1V<#$;JQ%gwrP&PIsMWyjI@Ujc|Xq0CEC@+})Na}HJ{n*OQzo=>FT~BznabxJmtG97! zVv2Z8MN4a9@5FL0@lJn(jCe@R9|S6w(!_ghS5+N!b_UcY7_+&YN9osq7N4sT~H zuTRua|DZ;b-6v(k?r*(cgN@;-<#iQ8VP{r8&E8$gu8TVqXRi^jwcU%E5?_W8<=?sv z;4+ZP0fz(ZwjFnb1Y~n$K`^e}f=vP~NlBJ{ZW4h7rAgP=JVrDa0gs_$@L#bLi=E5M32V+fSo?j)FU2Vl zS?VYEu}B|fIjo^<(u(Se3OjT+142K!Wo!`;xg-f?>HmtL&C6`T{Ej_p7-RYLgto!S z^5G%NH#m#`#{3sG@5DuQn8dMDnJ1L=!VKFD4>0mRcI9YoX_g<-H{-cnzRUAe=_$u0 zIXNC6yGV4lxaDGK$RYt8|G{!--)WWuz|lV7AIm;omP3CYv#{vz%=ZWWaeSB?xW;nLRQuK~ z;M3I6*V!5B_t48fsA6myIC0ebrllVX3%3aKci%0Oiz_TF`$r)6Ec`7DZ>u?adCEL+ z@^WyN3G(!2-iJjaNDVmjboPI6F38gZ>ZcZ@Dg0*(HQ<2 z@^wC^By(BjvM`AC+_`fazD_P`CO35d>JI#+Dg4mi-&;*q76yaKz~p7Td|hSbR8>`F zFJF; zu=uw<{`u^`_AvH^02A?mIW&;mUr+ewbAPqhkYz6ZKN-VsOZ)RIU}_*%4cY%OTM+Bs z_d{=4Sgx|%x^eAZ(7vUKBgvdPlSif+A05>()_HQ2<9?T`1}@$2b9z}aNAyvSxpY7C z4of*gQNAms#C~+mFo?;KQ-0|FtFXl>gHvn)Q!N zo`l}G|8w%s!TkLhE{StNc3EPR4>{#p9W3nevCH2>(Fk8Y2x-&sX|D%Qdp31KW58@=3 zo~hgX>BsK;*x$(53(=sDzg+#1f52hB8|~mtBvi#W>O`^0>$b8jXZsP1Fgv~mN%*Q8 z`9q9dm-^i<`fLsRSv3VL<7aE~0~#^gu~mU_wmqwrG*Mgmtw4_zie2fIl*@6-Tr&0c zOATD!bGJ`rXyn3WD}GJ)6q>b%pTkw*!+7k&RjaSfD0z;4t6%qb|7738hS=M7C5|_O zKZYcYxWqv6AmESWXTy*wf_)3S*3N6y%|krNvSnCR3B>m7(ANBn@fBD`iQ=+rGN`Or zE}?OkV!cCn8pm7qBN=OIr0Tpq+5=rEXzhpJHNF4sx&3sGDJ~X;ZjkGReG1;*9ExA< ztimVkXuWdH3^|Y8p0ob4zHrs%lFPk9g$MKo=I>2xg1HEC>|Tf zd=9WW^b5h|wnO4Jp@5uiLCvaT^ z;_@}Iv8>ZfpZ{3@G}B-mD=L%#8(ijLn)+Kd&4}tDd=RR7z895R9%ibnf33T?3a=R~ z1h>fBc>^DqaL5+#&&|3nLI>O{B4 zelP5%(e_*%*eLhWD@3_b%+pWsBG7`Q`Y~+;G2Xb`i?!!^M97*KOEfu zaO$Ca5C4k;=l^oS?+*65jX)B}!jhXta?QVb@&eMXXH9CeEpph_cJLa=V1|PNKr=MwVrEIK`JG63m>k}RF)f6P52l~RV|q~Di)6!Najl(Ox3=>Wvr!x3uB9o z5~|kBOCx!8%0)DWB8lRMaGDGgQn;)dbs=iL4 zYs5#@1ni#A59gt8UX8nnHAD_$kSQp?U8c;*CB&fQwqzGoll1-JSYVr|q2KTj{#zgJ zRoVbKO<^CO;ivs&uoUS*o3PoAZN{BiXpY$RP|C>XGbXirxnwW9vl>nhiQ>njS{1=I zC*R(_a?7sjMcy6?jg`TFeiv{Rm1wTSqlSBP_b)H?AE!0}NSf`$T}OR>Z*taC)oZDI zesTUyCk9`QUBd;)v|K-6xZ15gNaSTbK0^R9`9uzdEyO5CbB1 z8f!85hp2UmlyT2iHP4w3)4=X}X{^QfTJ;`loqhPKYJmA(X6V`v$4Pi?0-|y}v~O)( zYilaILrw;}wD6GlW&La(QU<$CspDN|z`s48XLM&Xz5KvOS=t;aMDY90?GCl}j@4}* zdu{>UMpn-GNl%i>+R9TtjkWXI*l18|20xXGP^lu%7;6XsTU7rggKbXm>IL4&;{vH; zA@Mrg9iDyv7t{Qk{U>qly_pDqj^oAOEhV(4I;$pMFu}&$mq6kb_w?{JpMdLx0rmU= z91I^2HIMmmcmU;Ik0mCmyVVPUy?Wj7yHKpAUo{g<)VOBc6Z5yL^J8^EdM}--K(*F4 z!ZkWIm!}Hd*AFoCD4>r<^PwF%dsK(Y-i37@BtrML>MUnPS{u2qkkAwvmN7DDB9UYr zSkO^09!S?r+0>EnYFwY}fVNAk|193igrAd0vwd%x&*sAnu3Sl(*5UhaOg@9wESJ>h zy(rwH;u+sHGl7`}RT&=mZz#MZK#{b@v&p%6|v6&$`t%6||w*kLXvpm3O zq(XXr^8vYNZB-2EY!(+Vp(ljkWjdNU6?Jm8p+9+0v8*_7)+G&k=B0G4_wvFT9d|f1 zvnDvug`jIbHzHd;wSyVpPUGe+wjQc+$QP*mu|y{z5iMfK@Wek~meE?8VhSfkBsr|S z-Z6V8uPkAfH-N_jzFV!c*t#k@H;ERK5zKY4MfZNpfO+h4<%kCCdAV?Ku)if;a)blc8+EzYquM#%0;Q(QdsyUBk^G~YWPbrLJI8szVc z{3{o@b93sY?2%cDb#Q@*@0gqN?SV>9|BR5XAo4D|n{xB#$M*dNd^}v6ona}td^}g| zoPHjft{B39lrQF#;a3RqW*Tq6XiJ_0;6$132*)qn8TRIu=ns$fLL2{|hsp}u0}kj@ z45Tf0H0Ib8GI_b}{j-?<O5^u=OYeN76lMNAoqq30SqAP3Xu#ZcQTx6HNH8 zcYaI`9ED8p_!-5V(u#W?LPmJs#&1lgd95Gl19q6SUC`<*?`~h(M)?cR%F!aDu;u90 zUCofK9kny`DKB{h}y zMBWYmoV1)?{JV_?<1t)cgFDf$%4u6P&p+&;+pWDg5Y0(rwZCk3aV#mW%0m4u-2}BUqX?3bB8=E7Lf=Yv_ecMw&PnP zTl$Rt4ohBbKXp>WOO^2T_U{e!7{htLkd$m3oOHyO8va@CbwqPh(s*SG4JOO6nr5IYxPcTRSo|p`vbUCEK`7t3WQo z%#XzxFDshoRLO7MhC%)fRB{kBaD>t>+K^qp^E>8_)G~`Edm%~g06Y4sd zaRO`;rmbPi+Lp()>7K;-?zWa}Bh6~uuJ-8tTH>BLn~PxR=gAqVbjuKA=Pu}wCy&+r zy`Z@oo-6rppt_e#_E^Wt_lL0F8Ilpdn)ZYsz7;xkUav(UHx1;L#2LK{*!}d=I8bZ| z33XNjJyAuYD;s>FZl(4HFnL5M+-C1q>i~IfY-YnWRucct#PRjv$s7cq4Q#j5XK@)eq)I9@ zM3tx#!)RD!=;%l;nO=*Tp;$KqSukoLkzNNvWyj4c&ZYJ+zUUd)<+0x);{ugeR&ijr3z${2eemj4JWQzTX4n&au7W6-a1GJws zIX(H*@0iwo<1yzS6P`)Siyjlh-EiJ>hd9dgt&06Z-zl2gt8!c1pX|u$i*An*hx-{j zzKYw@>LbGh`db;Mm6<2BtTD^Ese!0@t>Y@KZBgqVb{!(ti1fi{SlVxL)O0vD^Zxsar) zN24ijoaNyrmbhsk+sNx)Ja$2*@mIPqRPPt=hM{LXIyVHy{yfy zB}mBd(}xGlw|Hn}Z7h!lbnQ&RAp!F=eFKc)`fxU9iIK950t`|@wjA`y@7@knU$!I^ z9JdVPE2M7@*|%R?Kfnh1uyZ3SlGeCRqGS4?A*}*cHp(u!8=SZCx&9}DN!v-hqxIvN zB?9i~D$DfHX*o=@3m@00=)^(H5Vk>h86_uD{6{AA`zV*{X9KYG2!hts76=NUJ1*8Mx~Mh%7>&>L|xwkmmjCt5(F2A0phSb$NXK#4%6%hQ8L^1jn+rqOz_x zW}1Tm^0@w*u6k5#t{JLj_pNksaGIHDhJWw;uun@-9;VQJ2)Zy{qa3KSR9};Sv}^D? z%btb6Jq{(ABRH_#c~SA=&`dLE!)|8LLPu@+G}QbS>X58O?X{l(zT$K0STN0zDSlSB z>N2LR=5H6aR35!@IL*VB?SwfRP2L1D1&%Z*M*C+se858dS>r@(h%T#Rtl_PntJ0Mt zxjk9y3-sHl3hE_+CbL|wbg`D+8o&JUCVi9|AtBKzWX7UlTZFAP`Ob+;C@T*katN*82V^Tv!XY z51$Rd%#?&MM;KC%70WnPH693FUieDZON!l`qzgrBv#HktccXoVC7Lg3ZEpCXzyox( zv-#fPj@pEQBf9`7S1C2dR|BA-5KO=t5!U8=RaKd+50@Gmbl=KH%d?Rh-f#tZ^=<=t z=>M+Szr+*;^wk-Y+nws*5V5}y`Hskbh}l4Sp`W?sgKg)sLhIYMtOr2b{becZmcc(n z9M*n_;XK^M6CksFMZJ~F?=at{UpgsPB-i5gzjUz)xOCVK>;PH9^YYVv+9DZE8c}9o zI}E>v@%i4o#B$b+y(0pGklYcB^ab6W({<#4`x?7T$4;;*dJGR?z~c9kDMdiWYGtZ8D}1aUtaX2i;L~h2G%zGmwmfG|j`nJ#m+=2GLVZ!r z&AIPSz7U6o^#R`cR4YG1KZm?9oT$avkq;Ia6cdCm{SZB|IkWO}wkF3X)r5&R06LMK zZ(L!w&bdq)#_vklUdfJ)^IyNAEuxyAf9=&vUoesU3(jMsBole%@ZUoHnOpBWgLsEN z{k24okD5>Q_ezw&ojV7>UG8qDHo>-j+9bDL_Y#c$%VfvNi*^iZ+A6>pIV+*-r?nhQ ze~7j-*8;KQ#kDqIZYwt#w7#s?r~jb6)j%>Fja|?W7%c>AY^*r0_TjX^Xx{bU&EDYx z1;wDfO=WybR+Nyb-~}?mB?X>mu_x62E#0q65D_aVfDENsYHV(-*F0$9%+r{EePIR@ zMT!O>Qj)R(%Xm?oCNC6U>2{U~-P+XM( zet`2bPLfDGN)$zLfw%NBWFU6=$bj$+T4D4?mD9Q)UkfY4My7WY3Jg@r))f>7W^9sg zqH*x`gOKvbY&Z}_|w6!u5>~&C| zZZw3mhITY&3EM`ot86_VZxnzKh@&;1ch*~`UrO76Nc#?e;)V=r$cbti`wx z>3RQ=&R|++_Zm%L5Fd%`Bu zYMPj3jM+rn!g2EobVuywJ;A^&OuRW8PC{HGcngSJv149lV&fTII4-cQP=0#2(H!g7 zr%}gaF3>k({BUZj?(=IfiLk0Qeby(Lesi}lq)}87*dGmOS4yhz5VS7~Xh2!SANk}H zAsu#W^FeF95uG&Q&KP#i_jF&P3$;tzlFDE>TZhI2!L99RtddJwA&-5SqEsw7#ERqt z-b}j_&JOUULF=6`-~2Ino>9Q4RjTJE*JggETj4MaIpJ7i@~C2Q+)pRi_d4q;n@mL5 zn4xCC&aziuVpHn?`$7WP9(N)WUDcI}SsqL?BWG^c)?i&z-fr7F2PF9mj06ytKZFp; zmaK>K-IrF7;Sv$qQ3J|1Vk)VVjQ2a?DY*>6d}OGNKk>?9GKJHE1Y2LuCASb$ls(h^ z_C(@oId8gt#5b@r)A7&_3Q?!lC?j;iGS|oLa6bx0gAorSi>9`LLD*lIw+&0$A-Row@S!SzG^`1o>3EWx=3A<7<)abtd0pSPcH?VV+hhejCt*BJ zW#O6cAJ{=wM-rhf!F$9b*=CaLMAU(tC;z!P_~@$qu`P5^XXyT=%-^jZs_utyb#Vr^ zNkAfcwb{niQ3X2vxm=~NNl*M4&`y5GHvcEGpyCr~N=7QC6ijvmIL^t=^e?ofK4xDx#_x9E$z;U|I2lX0qyCQuY#zUL9Fbi#x?*s}# zRPpCpR2x49g&Q#x&ilqFZ!mjCJ{zQ?xN=ybOJn&N?=EWc9=n8nV9y9Z%@yE}_Zjo# zC+uH#FRpPLa0v)UY^>%7ESNN-bNQV2u-`t+a*=w)J8GHT0xw3sAfi7ZWgT zC~R32%M7h-p?~n|ZCj&G49K#LHxiN;g9a*z@+l=~2oZ_cKe5-0}+SdiwOuq+a zRceiW_->S}7NP_8J}|0Wc>BbVomKesjpfRXJ&HQ{g?;#Tg;C#T?p4}nTw{?bQi6ha zjZ>NFpOM!l6ljJC8LMK9iMsiz#{4%u3qn_C-v?2)llJH>j6ZqY{}jo)KQLWmKk@Qk za*+2m2Ti76`L@*91=Lq_Wc$9{PZm_DEec$bJOt6{4Wph3xaatEy`Kz;3K z0*{@qv-oggLmLdJv9|cM@=u=58=2zQjS6z zlTl$VRA#G2h>?aY)xQA=@j*Ws*{vtj%AYI)w4cgc*fI}nA`G!yxW%k%0(HLr0(9ZL zUw;Tk0u?+8f zE|(Mp@C{S~vX9o{dCL=`xCg*1F_=EHn3?g}!=Pe$yui`M_h(9gc}16Iwa;+34Zch` z%6t9`y(NziZJ*=*G_1GtlDLWumkn<8Tto{hv-kA?c5VUHZ|Q>#Y~+Y*$DWnir2l%m zG`0PhzWGYst3hTGu{ayFl%?LjZ>#(b9Pb!+34ir&h)@Nd;W(1(hgQ_qH-X%iMcy*1IVk##5%xI);Q%3gXSJ(P$=9s6v@A>u zFiEaKBQh%M$env>Z(2st8##}KbBDgrZ?Mu`6Re?ik|-;I|Fr6VqUBDhb3v0GVxLLB zbL7W!K-ssL30PsXUAakC!W3eS`?|4$<6+3`@Y7&{vvC>}C*4H^I7XAu@pj?F5f|td zj~S(IJ62`&m%KbRaeoXXV6uE|`Pm&^@UcvE<78@$X^HH38qa5AR8vWPuU6|TRT{44 z5Uy{$^vP4m=UZYK3X~Dl7D2t(Db0MoQv&9E@Du>vMmh$3`zU?J%C;W$51JlR`;{X& zlqF8a8i%2h%7x*BZUph$;olw{F(X@%7j<|nFPYo?6iNv*c^GnC8shwOIdI|Z5&qB{ z4Nty)#Oe>Rk!wqZAfkFE^o)UWd-@&L40QSQnwjC|p3zw6mG#>xVUgjA zQe{|doBx3UQXT1zeUh9fv48_U`_{NDj!FuKN{l`jlfEEe9AC;TiI?{oaTHrsL_=)} z5>Zq1#3u~)aX&wzS_3rZJc#e|WvaMH1H0zA^&e*y4{zbq%6r2n32R@X1Y>?`IcqhG zxJw26Z9cT4xPMbNWOJzFi`_qn=}ytt<|=q-zUB;2n){yK#N=_0;WUf+zqKuOcO- zU8e)A^mEQpl)&h1$-&0mt`J9>83HrNi)!ICy$#AMq+wg`4X#hyqZ>{!Ikpi;j1Eui zhN`SpFKg{F+4`$Fa_&h?0LVnx+&1^Ia)vk%bvk*d!rr4SAWjYSaEN5Oy9=%J;8Dg0 zyKWX9w#UO5+9_eZ+uLLKyiSjHluhuDW_hp|Vx|+erUa>^QMI-%$ExnV>)2)7WA)4H zCgtaHy^9LlHFQBPBtR0DB6zIc2b*mN7?c}Is>|YU55>&wWDcmRR!EpjOen7i+ps+pTeq4NxC91A*S=x(sjp4hSLkmqYWkXL7vy0njmmR)stn$z!1(pdt zQKDj}2nh51#gX|7%2&)P!hoV&VUlAwgEHo|3giK=;P$&ZwGSXKJC#c8%*c!7(Yq5> z_&ebypZiS>&AjmS9P&{;zVVjV{M5`g46B;BH4k)m;3jq)|ctJyJ{|T{6@7? zvfMYb)|X+Vl(X**O>arg*}*oOKUx*ZQhR+TtI~hqAbB7N?#J-ibOQnoVjwb+h zX^W)_FZ6fTq1+nJ2In~4;;$3qt#2zBf@vPF&uSQl2?cMhs!2|PQp?=J78;47i~073 zGD3(oOH!py6LR3lyc#vV2C!80!k6{)XCLu(g zL9pcv6`y2L)It~bO0;n3>41vwZ0}OF6x6G3AtLji5e$GePS z4qShb{bnA3fidG(49prld&yb$1GduDZtpq!Jv~xFNIPVm<%AcFC9XQlEk)&aimhF_ zcU%P?mx5`&s1SJ=I?QyabD$DitEw*ccghM$g%qtB2J-Y>Gi4AWzfqM)6!og# z4%{XHuIH7k!9FrT=2wOmpRos0!=jdVm8KQa<6$>7hbVbA_Ag0WD0C?r#Y|+A1G)~q zWRuUcpR5RYNUSaV`(04V+y#T%m-+s&3r;ds7LH>Al;?nq0tl{QuZNW(L80$_+|oA zjb(1-vwidY`snOHuPeD#C@Rh}CXiWzYuroNg^ve44ahb_Vb*VE&DZz7n}Sybe_g0} z`UHM{6)4}n%M|Sfd0EwL%HzgWC#8z@ZE`~C6Vw`Im7L}EudzBUB@?bUe^di`s5{vB z6M4wObCwC~{F*iZsQV^4XQU%-J!Ej@nljR$$&U}A+57g7lA67|pbboZoALXb#$sD2 zgX~!Yj*h%Y%=`>6VNTerXI)vq+thy zsrB&LoaOljM1jl^(0OwTAQtPGf?FNh1xoI*ieUTPPzz>R9eM`wHfD%#Q5!9A7!8^N z$Q!#|&#WQ8PE|c<`KHuAan`^7<}oPzWExuQBU2_9&=-R1FEc+eMf{&W$U>{btPHS$Q-uw z8gO)N8|y4rc3)v#MY|uaK|wTVZ^=(~5GvDXe}5029a2@2;u%wr?y$CVnX&RyOVQ<{ zFJ35$9XF9ji8)|enQ^7}giwWnDJ8Atj2#Z^PG)jyC|kwJ$+)&{kGLv<*r$*4GXhiR z0V2ph-KWLEDgKbd%!#3wA0(&BibM9Nu7{YzloG7h7tt3S@x6vd>>G%joIMW-Oc#MQSj3skX!y>@%nxGd!awTQ3H!eOcyjF{U9WQ0Eg)@BFLG zchq1iN+|r5bk&5&Wa#2V$BQb+gqjQp(EB)6B>bMh3Uw@nOWP7T@kJYZ@ZI#oMYOUayFWA_^>u7C>=lx+rknihS>Acc;1 z%Es6)Yn({`+6xNDPQ#R7;n5S}(531ki}#K%dFt1tsF#Ca_bApM@LWQe$N3BTOSt&}+L^$~x8jipmBT0v6J&Fe_h=DFm2~o%&A-Yo_I1&IJ5l zD@W0TzfgSQh1cf0y7K~>UpUL&H?G)Nu2R6nES z)Bp1bK~dQ~tTD|ijwwtmcL*`xrk{SgZX2=~*pLPC_pCFT#+$oJWKPJTEY=e$YyS@(3rRWbYFRDgT4e;$P2XP46HsOrI z!<&;O@{4_}fXYDWy^-~#VwRB}SIvG8_gd(DS3#fp6B28G_j9|?DQ3kjC9fGF%`lHDHI+|;9Xm6oz;Sa4giop5M;J@eq} zDr9-^t~$9MIg)M;u@A-QxRrM*dr^7GsKy#fWE!f`b7ix$YWyryr}uQX5M}`+ZKy_p zb($7q`w}(pxm-s5NAM$0!%QZaPro&br7x8|PC0_y8%z&tc@c4)0C^OrRS=SWXYO;S zjq1#+WoE|b+$B5_r2No(2HH=o!F|E>mC(Pb66{YKYp#dx0BAMEF`uusRG(JWjsSf} z2+Q%4;*gMj?>5-Ni>5a z7*R9pve*u{XvB_V%0sQD4aHGFLYTmuU-<06&o2EXsith{9hh2)$?n^4TQ%&fZihUfHfpxURb0U zAR74%f1>N(d|d;$&848D8Ts%@z@Pej{-XCyFhqcp$1YT7N5kR9HHk~uybIc5ga;BpS%WKXamL1W^tlfk7=zB4tt>;B}_ zs$1#z)POICjSt!3Weia+KthKjn^cg0N%szYD^?bwW*@xNhjz+Hnl*g~s zb~ZW{RZO}t#mTyWr^jhT>9nT5relgj038ef0Ynol-%_2TkO-MQ*b*?`neSH?=pD9z z4q>&Ywb|TD3n-wK>C(kNaz z?2;sRp4vcNb*x>1y5OfVjxiZumV%&T~tbo^f3|toM0Ylbe^=!uBK5r8d!BRYmL@MZe-Z8p_pGK z7>7cf`v65&8Xy}a=^oijF|Om*dVgZYFbca_Tkom&!u&+qKA#RanEtLCnv??yi*WpHV`N=rHyd8_%9a zwlSm6u25+CxO7+&Q_onPg)jE`{vfRF{Ex^=qH*)g^pY}NR*f<|-N5w}8lTT-3Ly#A7vm-VD>IYwW=}ePPx`~( z=WA;Rqcz=YXFJDU{lldO95ne<*qW%W%ZeZ*F!in%;xsMyek}Ig*|h@XEKHJlDe;Mk zJsYo0zI22;u3@TI#k(;xVAMp1b%#{jx_2Tn=g0}((_`dxP2IWA6%wwp#h{EQRLj}) zb(t50OhkcEAHhEK?2%20h#tDFS$=WhG*H}q$+=Tf_3pYeASA4tlsTbj2N3^qv+qp% zi4Ams#ELGB_4olTcgg~qj%K>L#Bxz$u6&oR^5gKe5XLG zQ`&LPk+OYQu#M-g%EF4a!_&<{UcYfO$DDyDvNcN!t(3VALSHP<$7#!B?^ppC%?W(lo8)G#xP+s-Iy&rjOl4OHx#l539=n<2UR?~5#h zS+xiIF7!6Wm|HqF4Tk`Q#dc|>v+D<+JFNi~i3G2A2FAyJNWvMDx0HE{fqMNcUxS-h z+MYl;P=3>`rBKKH2y*nI1?n;kuhWh`8YOxakHj@4W>Xu6ZBI3jV zP{&`w*)zu;1L~)|NkEmQR(vmgf2E^oZC8eX2T;xQn#l$4C8%%gvTMBI(?IO}aOFLo zk^k5}e6wPF^`(N;Wk!9?M6W&?{-+9yjHQF*>or5Sj88;$85}^Y{A`^?)`xKxFdXku z3Rl@Rwr-NXgAn~%2x^ii?+(8Y|+*xFUy#TF){^Frn+L2I^KR=bJ=6gH$Ye$6-Op$u^cNWogF3`rn<-FJ#8b5c)QotQ35-5hA zV}6qa8sc^nRmkL<>bH7Pn-tVUG6g|7E7k(2Joq7I6Qd%6v4gbi-Bvk3on^LP_9uRhWGit zEuzM5lQAj5ljc}HE?@8jD93Wj?QxqJ_0#;e1vG_8HQ+u1=_rRoA06mtrvFB16pep& z_ztPI?)~6*QK|)!9A#9Xw+euggMBZ>M2XSC{y!KS@ZQ`uxI)Rcq1n=?SU@5k54ptd z^gy82_lRL$t0Yvu9wv-TuRuX!qa)~->krq`p1zn*W<}2&d5<+HICEza>cXX!k64XZ zbyfcJ61hj^Z-AmHbaUYr#fr`u6%F~lxx019F`N0sgG)qEeZzVgQfBy0&v(dziX$9j z087!Hthb0=W2F|VRE$3(Yr-rjxiesk1ewG`T;OjEtLzz)&D=F`<3aglncuF9An1vp zCfTEABl31bySDjR&=@E>McT){)J2K4ub*?RaTVVeg_~enF8t{N9%IS>+a85lp9gXh zJihhZzjMZT7$D@lWB7$+g!+6;_d-rU4qe`wiXnl>@H!vH#P9XBq=;M95<$T7^ z0y*C64M8{2AyXw~Kd26`HmGmKEJY2zsK^T(&4-K&qq$ z_oLz?P<;dF9Ix6KUquT^ews{NbZ}f}Z0Mj)&VRU#N0;m|c276e-@nyF|1fZ?t$nss zX*;^|i<6QS5ZAT!DzltF^xt@Om-}~i6B*f~TWRI8U0oYLxbZ@~Az}zy zfIVps|F}6>mGuO_xkG5E&=4f+Kov1^kX82c%@8rrst*2KWPP}`?YL{Wmnw3py-Nar zz3%&M8aT&0AJ6=8y4YX18X+`B7WNzkU@x5@YGu|t}_sr@a=tQ z2M|ezEr9e_Z9d!e!x9Zms)xtk(jOwJ?FdQyz5$;=K$P|G{at*T=FQVB^Wkd%3v`}! z|Bm}t_P|}YhOO=T`G>1dqXh?Y&gywQI?amL0o=H>jV^H0wSPFD>0xRGvamVNd+MjBt?7nySA6-I92(h5lly+Q$=zxLbgk7cHcS` zNS}NQk>8ahr@gc}<$%t<+*_xP-o;EM9GdFcql>EtmtVA{$ zaL7e{>dNwy2aI_&b&J+?Eh_Ce{@f6;Jh$DvaG6@VXqHvxCtK>Gg%(*K-mQzWB4m8tjl6#r;3&i=iMKe^ifSDXG;)c?G- z|K5xLQ8s_W%`X%2fB*R#ZvKXw|8J(>Dax}i@^`PmmM!!|IE4fvW;O8-+=M@N1hnZ& zfB1==7W!}D>i;zXz$e+DXAB5*>gV6R4uN~w4^VIcYeAml?7;uk@qhayk>9)x+bOpb zlm4d%`*R>KfHDrX@Y0=s>ofYpVgAcCp`8psX>r^G^Zl<*_Thb?XmK`P;}1QBf1B37 zUh~-)NRRzJ#eZ3Ye^J7}SMfi^ihnEWe>4WbA%K7H#s7qtf5Xkcg73fK^51atH{ASf z`v1sB`WsRI*g*lnAKJ}M>dGxHsS`tdRoq>NoJKKU((71*Jh94ZizE&ZZqxC#j;CM$xpjOgYxH>R*JX7D;DMiCi?H9^dgN07sSE$Pdc0TJY3&pC`Qe}6 zJc{+C<4@!9zAfE3v9P-0SRUq_F8pw2o=2@8CrzfqZzfItx6zkuM(9jc;OBmd=;Qew z+9Vnpn#V4D>FCH7i1?eh_LMQyV&r|I`$?2jM(paJo~~lH6N&TX`#Zhj(aeg-+llx= zN4}J5?pt$?W@bWOJdp`GCvacypGj+6FmmG#Qr>lcm8neh+^D;Pw? z*V+A+G3_Ucq55{^cSKT5zR=U`X9P^0Ar61JpE5m7xx`&V>3?hGPdo};<2=jG$cxWa zk2VrdS+6DU+2ZdbNtT8SL=`afh&`LkU~0jx;q2W7f6in2TJ|7p>tx2{72edmTOLpn z;oZ(}wh)_RefR?y{C5d>%AF8UTe?s<@CmtIYaMr`rd%x{s zem+PZywv;8VUH-1Z!h5@2^-R}fyb*WPZgzT+tbo^>D&e%WS3mb&Hk#U&L$*;b zjO!b3in&iz4o}_#|HpZjdAPTe1wq_0B~3VW_NymvM&8#|b(H^9qtlb7Ht!)2S5TYl z3S-Tm!GiQTDw415Y4lwiFH%$Zead?J)A~<7RJ1=*iZGs$^I4r3OyQ_Dy?aq5+G@I? z?4vOs);*W+@TE56+ymIFS&e4dceCcFkEL~1BYOvDa1)o|bANgL+)RWaA&rkecfl4U z`kwm7FZujN@`2W*Gan)9~Dl>$o8ut z0!?&3g;1$yj*~m&3Su;6-V)tf5WKD2zWo#gpIr$WM_$|)Z&Sf!&I}gxP#K!u$)#`~0 zi_xAlUpwO6=WK(`cR!M0xtFNS{CX@W)6IzB!=(6p4Jkdj{juaGu6SOJ(=1wg`aKp1 z8TSEy;3aw2ZVrz>HJ)PB8%6p*l*{>4Sg7i~I>L!3Y8xA0xpu7sUuDI1v7rZ=7xhma>ey8v2>EVrbSMk+k`=%k~J&9TJZyEXzbNG&Y-IFY{b z3&_;SHrYe;A4k548<^qz0w2>hP|3xXs-FJkci$$%X$WszN?G9NtZbb%&liaJ4a;pb+UR;3;g8|nnjGN*$FzIvvJ1-|WO zb1QkhA|qd@)Ju}`YGdQE%@>rRaX;8@2DY$Zoqwuos-|dyv0}QG9EO5r@S`Q1+~EZf z2)rPn!8|jh&n$e(2;xF3$#rp#3=5)^#k_oxkU*beOVTYxK`Y&lNF8Z>x6A7GS{l{Q zvrAZZ0F&h%ZfUZgzFtqw^mPc^djkqF5RpdJ*0hghDhGk*SfsfQH{Q5(n)r-g#`qIf zS5rVnPW31lbhF=#$mgnIb}(*}vEyrgm!3NE40}?R?5ZHmQhoZy z<4f#Lz^(hie)$mk;Yt{D*OoAKCi<2@t+yM8F9;!#YTq#vhN4jaUR7MI7Ad11-oLc8 zwAOznU_ZocH_Ri@&d$yqmcHI3yHa{`Bc%q^sKKZ1b+aicHI+|cf6d3LKS|(b@c6KJ z(+e+oENjWs*YDx;5;^~T7Ar7 z$fq6C&W8Qr_R&OD@sQ=rZsnF`&&OI_ZO9jGc(PP{|HBhyppT3&is0cJ>?>bo@khYZMe=>p#}oZ(EKrbCHkqu zOt+_&t+er$2P{MvIjB-dnbqFbW>TV6wI)$sQRU1wDr_<8+0ogZuo*>_Z-H8q2dB%M!k1eC(J#qv(`qV9xCFFDB zPbW!n=X?)BE4T&gF_>S6r1TI!R`g~D`X0n%;0q|PQpFv;^`2pdUk=GzMFX$0hDwZz zK|I07K>9$uHu%fxIVY5^LaaonJ4|sTbwN}N@R6_56#L{|FuB6t{kFZmz2hSpwPjnO z4*Wh%`Py(**YgfWf9Hak8G$hf8D8|XJo0hS{_w`^V}1Rr+-e0w7fBp7}Qsh08F9*hMWh^qf z9Btg(aE4Pv{B~lQMf-hBaj`A|CgZQor{>%A`D*3O5s+L!dfCQ;x~Rpds7>vM54W~i zCC|QBigi*vJP1VzFv8QN_}gR5WMi~F=ItO7efvQQS0VQ0#W(_@erjqRo)tfZ)H8NMO2M{%zgn;LtTXbJN$uB5S zxg*Bv)QBLw$L!?gSxfou#$i)6E=o*pylLBj1k%1B1YZ|b(8C288$ae-%Ij=utmb*m zjM~HA4LLKIs&Kz@L^e7qYHltU`EV-BpW;=Lz+-gM*ak}|oFn@KHub?21!T52>k&Tm zfkOJ-QFuv!l19@$2?Lyy%!1{oPI^*>c8-TPw)grr4pe9TOUff8)HN#~j!fFsR1`Ya z-)GNsIL2)X_MaY31sU+>{lb-+FC;o%44a#c`Ih~?I_~{neLDAhCGL5W8a~wW&|s2B zrkCsWzpk>*eXPz?A@TxY!6b{MGb{If^bhoQo^+(iWEVg2l%}t474ZDHYwyMd0+779 zZ+3X$ysB5bO-_E-8>E?#&%{Qi?u@?A>(vcptn3vVxM}f$7i?T=GZzsI8GJ|sQ#c?f zy#LMj(fz4iHwyr3_td^ikHzm>w9jd3DsTdB3@@JNlrs~DnPf=jertwL#PhZ+@kHp( z$^^6QfWg5*Y{(B;`_i3WGh&~?(8G#)Q$1Tr0%X1`qjCDs?*4w+l-#-Vk9`P(ZhPfc zo0C3GT_WpDj$4^Njarj?+hg;|v~pZ&)uxTMiYQeHY~Y*{yC^2KZEZj}XnzbfOsM%56TBv?BX7~=GnH_I_b(@2q6glfJP`NiTKgiVII<{UsHh0+-rv@^mW&^Im1_v97o<9>VfIG8G z{hT1H)9<_C(|u6U6a7yIhx_bW#XxOmziC?8nPVHVSV>}dg1BcA+Zhw)Rqoiy3dBMK zrR{|;#QN1ASqt1BGpBMNO^rZQoz0IJlu?RqV6lC!Ec0aF_mZT!qB$;8V#>4dF@-Lv zfMXr~M^C+~;@1Z_4@3v>I)E28i^kOjqAk6Qrv55yyKO?nuR?s*Y}v&~37-h;whYop zKX-MpC6_6mtb(gSW%yDj;B==Dtjv_x?KuG@RGoUf-zUf6N`poMk^QBv``YKcM9AR| zg#zw$23me_LyAs+Flund(8D8s{IEZ(iYf!`%``_G0SB)dJDzpUglyLjR!*!I72ZC) zY!bep;6HuKp!$hLQBSS&rxVseh``yg!&z;6AJm3_vGA>BmqScY$!(Ohu8`sG%;AI( z;0)E$LQIHoh-ECA%VcugA9Iw-r*sW+E$^nk z>)?TYW1fyq;=;#QktaWR_+;TP7R}OR_B$9fv114`v~o%)X;VVTR@&6})>t{Va=-Nn zj&vDe7HRrEu;%t^Q{c|YpvFT*w0VH*ut|ungN`>7Nrnl>yPn#~M6SLbzlyyZ-us%G z-)9IRii-^uO}i}YKQU3CEER2E5;W@8f=bqEWbS+QPgOYusLF)P)oR;{3Yt|Xx zCy4~Liw8LkyPNz&MBC=QWZJ`f%iYl>0v*OqH+;4oJ!1ORdTMS#oTu1Mg*m+36W01* zR3z+Rdr{ta;Zj$H zO|&KOx;l-EMyP;|`DAW>5v@r#!dP*?iv&ZpSaxyC^tE_isnA1-oi%Do_~AhZ?>t}} z(mfy=1K(mAIF*$kD@d@*bZw)l!r_(@k7IyvU{PX4a%$8#5c1!+h2a^ogX=U^vWg`x zvInm`+UsuGe_&2XqSj35#3m*tel!)yQ>XD;^(u}CVfWjwzG?M?YiUeErnHtVl6-w< zZHiq#pCK28sQ*T3=Q05?iFzP78HSzy(CU7O2oW^fM;zHWwa(B#rFnuDL1V^TfKhxA(67P`mp4?sVQ@H>aIphEWsT zw2r@dV>TI;B4&|iX=SCdmmAGjNdnbnbQ!@p|K_~qw%X6P>>R!2Q0T+nLoLv!HcvRm z8Ii{S?0l|f)+BuGsawOE+Zfvx(Q}G6gubRqrgV+gX6sqT7ZbI7&D-IhTdkopys{0iZI#o)(@rC;C}Mv@i-dY3LzKexTa z9%rY7dLb8XNhcdCvZqM%ek^s&r;0zz=iz$bku&fcvxJK?gDg_fOQmD}2JVX=R*7UV zpNd+2*S~><|7zTwllV7>`>}DV%Uv!8JXr@{7)f+WXj7M)%BV7%)_0O~;Mz)FV|@or)8yX z*b7>AK@R48(2JLA46*!RkSHUUBPA|mqvCLxbb5lJ%ek?GbU!POp-5M%sB&~aU~&&0 zCpBJ%|9trg0Pp*s*lGM%(_jvV=w=5n<3(F@Gn0w!z9>DV7%j6MM94zFhOK^k;O<&| zM&X!)grlqYNUcNl{#F|!tgN9m9@pa2?MLOr$O1nWv313t1g9f_SL(eM7m%v61_YV#_scZ^cYir> zWJKfEAPEVU29k%Hv?K4E@)n?l;_Q1p6nB-TS_Ce%Gs|WnH^JkEx-FT z9!xFV4(sxHSu;qAZQ4+nqSPJk^O8En&h!)&QSy6r_7|V)nCjZd9#2p+Bq4&dR7a@5JGv*G|p?kVCq5da&!p! z@PdWhG0gsn5^s-bX!-KfJz^4t)yA;ZW=snE8JP^Aut3nGOE7b;eQRCdsg~nckTll* zqt@ruyf<9d^ZVIL7Ppbk{ZundKWLm=PCPXMKUtX0d(!2STDr&CwEtp%vGI_GG6OA1 z^9250T%c6NdGc+ud%%Ut)KXneA0tD<)IeNj$Z678FQ_id;iBA5B^9V{Z1GdXLKA*o zr~@2=&)#1-3_gEe+NWK>wDP-&Ctrj^?v}oIiF1PkqK#@w8mxQO#lTH^gmhbeYa`^$ z^N(KRqg&{$mI^z4J;Vf%EY5z7^X%o@X(?I9rA+5P5&#<#BV-9E1sOBLB?8bM7k{|F zATi%!wtt07v}lGny^L`PiwWorBfI>;W9K*LdN054wd5|dF>dTteA`g=1J`JzyoUTj z!*<77Y1~dU$hh9wBwKSTpIrpQPQ2%nCOCzip;eh$>C)*#7B@A#krCY8m<-u>E$Z=e z4GBo@G9*P1(xSHCLZ-{!ES#Ze8SF|IE=V`Z(ZSE#$mt^%#+ zM(b$0XrwmAcpb4b=|G%x!fe#FuWo%-IO(%C$*|}@#R`~0c$L!O^eNCo5I9=*xzAX7 zHr2fUBaN!s^%f_Bpbj31!Y;tQbRDYprm0XZ7PtVvO)dU;;YDNuv)qHa7=WxN1O^ydhEiQvR*p*MS3;M4KjD`6n z7il|WNkgij5dSL5jKj7){{wJHB68AKDTW0`o1(!l)`Z<2JBb#x1nUe<WsX@*l zw5WCoIDQP&aF`R<-j33*+WlwWxi>6$%m|CNEE4kFu z7zNDYBPoa5D81!2=Sv4{;+CIL!po{OnMGCNG?NZq_TV^wUEMs9y!k`?7Mi(y`OAwK zKP87n0%)h9y@0BOk-MfOr}C?bCmTCqg6^Z)jG>)fir332bkWS@wD+WT(prChYy#xe zI4!Ha1pClO@K+YV;W-(W@0Fwlqm7&9n=T>$YJS%{t7{PHB^{JSZuWY*I{Kr>-B9PC zAVany-E#`t?F!Ns?%2(onw&ehr*uv&w>;%`uEq06p|?VS#U(c*JwN;?`UQl|CECtV z&dE>Si~v4#7W?LKcRj%K>-v*EAh+s)M+fTwWOEQBObXepxeyGGC}^ z{sQVpIvuZ2a4fn;szT6gHydGh%JaerA~ye9;Eg_IqyDVujaz_O_?Ch1s&L0AGCEBF zCAMRCGmLU$eTsEtl4kxrJolVZm zPjcb=`}^+jtF@O84h%h4OMd&7ZBuqv#lH@!*jranihBSJw!GI$*NGrw+sWb==6*!odya zYOKD24#TEwAFq!5nAL7W*Mj^u&(I(f9&jN6oEZ3Rb9x}=;D|-Z4tqDYh zqhV;K{gb7fru|9n3tJ5(?PxK*^}5bZ2vpff+IA3oR_>XytuZ zr2FA>a+yg3>Amaf%<&zq-J3O7ri07K*i;}LgobJ+e%&LA6Hk-nt73LhPy>jP$k*Ed8A1 zL3ku>7mzdYGgot66fZ4jr8Y?UFxY`e<-?jXRCzJlM8V0o6W+(U4&FxjpEbt$aom=n)cBC$c9kr}m^W)#@a9()jrUW~?-r~0t@6MTfhRsyHW*Vkekb&_omPJ3 zC=Wk>SKt&PfwkUSD#*yk(I*Ur$7a{zf3%&$Pny+Ps(gnC$UWUijaX^k%qeMiND(Zp zo1Y*#SelHzUydLkCQ9dbighJ@!cqQ=FUjeq6rEa-3_sbbUA*h zNFnl`Dwq8lw~|I1PdS>G(B>U}VAelZO6&t++YAG_}RuJ^~y>zM_}tsBP@C=LlG`BZM^xdw|bXK6E=%$gco zH0S3Ev@~|*Ea!81c{MYsir0`}GxJ)J>kVFeMstaY{29NMupy`_0?|fEQVbarvknGw z4M&0C3tHz8BQ~Px)cyqwb;&xo#B9nj-@aORd&s+Ry}YR6%3=uc77)CBosG5icOVyO zg#tzUaDE5!`Vhx49_j@cwe*jkyK=zksea;?xu%Xk1_6Vs!r>b%DWS>7&u@iX>tf7S zOO-4e+)-9;$TMwUr~(N9+gNkGZn~ao(I4t-Jw9C4qIAV@C~Bc&f%ZJ~y|NODYytW{ zEg9ulq`+@dozib3+2XOyNo~$w{ z#GbHhkvxTg?CJkvGW=)q&oBSXSRi!#W6Lsrx9Ac3)MRvGRZ{Pwt1vROrv1^Al>^!zLo zQ&;gIfR#HSxqHQ#?7WSWfo<72a5I^Xq#L5Cso%vffTd*n1s7Xe9~X1Splb}}_!2jf#u5^H#349w(bZq9t+ z0z8X-xUitWAQ%_2^T6-!U2_uygOvTQIUNq4Mz=!W6yAtSBabq0pSP_INC zoz0@i5Kb{Mu~4s;F;d-oWj&kF^MQJy_w##uHn(n_STDz-=Ers>Yzk8qI0+u^IioC? z6cHP%*#+1p408Zpd~w>je+R~^aDZ6r*NvsjI~={69CE>NxS>NkT*&jA-PYO*ER?^f z9WUnOWuC;~Bxl;?3Acxywdn;SeP3U{wxW-Ww<;ZJ6}D@znS0fT`$8x=|L$2hm4}TP zNWam3i#ha~H6s0-{?PQnuG?%nDh=Z7WvUR=H#OdC{&jsO)a^P)&n41*K~hywy%lak zpl#o*4`0UNkKh6+zmFp6XWD;(!Joj3J6x`H`!)OCvzj>#5U;l@)0Co0xdwbj*e>%C zh+P-{ImD43CeCC%sM^j18N9$!Y0N@@hbB`?31uiYN|WvaWY+EN9WwZu3{U9u=Q~tw z_@2CIEGOq1vweN)V902GOU&iIN=F;x#%D^&Lo{O(7f@^!51oxp>=bJ#xn&EsAF9JF zY2|l>ria6y*luB6f_Lik@PzA28G}Zm^u%wYT_*x<;@S@lBb1te489~o?(C1*xv8d2 zhFAeZa2=|Z{awEeqVwP+MLL{;U#|o3K`a+=moyTw)9c;7Wz-FmHO zquwBe0#3Pv0vXKU1(s*Jk%z5XnlB{B0UuF|8(*_eOk{w+KdLlTGT{a~;xni(0nDt6 z?q#>5C}o8VeWgx~Z+44j7s+Q4HuWsK7av^)LuI^yqVMLAQF6UzMmzUHP|z3cqJu7K zA!&pMc5^O1i;%9CBHB@)2vtBS-sd6$u!K9x4*`WN2Th%;Ub-u#vd#BpqS5K5y$ z&{W>SR15l^^WFAa+MtIqY1tKxs4BZ%Q$(vABUlO-^a)e(Ht$ERnUVj*d4n?hL=nq& zxJl>-YEbRH$B!SAAi^1+5>v?$tj@Ip?aa>b*hy^c0%5I~Mb>k;*7dGX*n7uGM*njf ziF=o75v2R!n4}xlINM&;O@Xl$gNi#7TZHWWz8B~0CGCEEjr-uyjW8=&!^HH$r|Ta+ zKfORnqNk#$us#fe^cyuVBJ7)nECr)F2=n33;l#Mbzd**-;Lo1H@`@{PutT>GRaBoA zZb^zg7_{6RaMl&|&9*KTjZREYpl^^~=wdb9Fm9b`(FvXq>FJq!n$?4;pWN{7rD_?a zwBRhFS&PliJzPHYZ0CAAAKTeb#Opic0rs<_##49ICs8>`AsI~pJX;` zBP)q2Ibl*=(3%Min8=6EQcN8N13~V5)GBb=^gJ5e1=J`dQAM{*ocx8q>NV|iqk0Wy zHkyp4%GG$%IKa9Bw85(rA1+y zF|cwkleljq&w7!up(o$R<&uoQ=QyWZ_Q$j)8 zW*Hn_uG!kiEJ?pk2Reio6VHJJU3sI|NV(#B1>bC4cjRwdFp2TA4RuLJu%92z*~mionX^xjE4=fxzx^E@4+RSF8gnUnf0sIn-MybS(B%c50OIY z3R8YT?Qd8eGHw?cO*zHr6@9FBJR+fl++Y}2*eDCJ_iDM-RrXds1W5ILCt&Kd`nvDZ zw5-k*#J6a+3ZiR|3g*d&2cvo!(MspL+o%Q7bRX)?vIC`2fNBc0s@~e$dXV%SXQ$&w zG?+#{-Al!cFT+5K+rszsT!pO%f*95N# zw-D)h=~sPDJVUaWSYh;K-aX!XT`;){p$(aKu65dT zM-|`!CFs-UcX@X!ocvWpB&U-<3y`ytYJd88b3qa>wZ>$Q`JBhF0VMyp_(epxe8-cB zOSQ?Vdr@z!Hgk8>L{U>oL8FjY+hSF)euE*2_eIf_owbcS=g-sK3E%a96c({+cZx;C z$FWD8Z*lF?gvwoo=>a1k!1Nf!w7!*2|Cz9J5}Oc#18o0vnFUe+v+ycemDMS0Gj0eOqJkzqJ&3sz-TSr;>@3tNHlv#BuEm$8f`F>L4ygi97d z&%2LxcS5xp<1J{oWZa5^=$g2y#Sr19BmEFo^iNLspUIYmg08Huz^A@p>~89f`iViQpPAaBkW>fjHSfD{`w@R_FWH?a zu{=PT#NsoSbhck~O4KLHEtHUolroeiyYR5bGZjelXVU`?(%%3`HCI8uK%SlkVuG z-5N=i^6|~Ph}>F`hK4L5AMR`TSMuB(Js(Jf)`;P17bROQi{HuZd>Sdb7Z0a? z0n;5c7BNblh|=NoW#d;_xb;&nHBXdyPns8aXP#vj!dp2oP|)7*O+`BXwKQ)Xv^non z5f|9DX-k&quAe&5cUtc{&?u#U_2x0i=#w#xK3Z)6m_;SnzR+Kj{NIa-qYq7_$i&L& zG!>#!XQJ*tc-h9){p$w9q{~5t8SU*O;82Q#JEv!_tFXH3o=2Yq2rZP#HnF^x#`MJ2S8Th9m4DbvX4^%I|kLL>I@xE9bC3!=m{@BC> z6}aS5-q#n;Ehb=a{``uMA7<31yO;Aww7o*k0bP+gwDaBhjwIn?t#R;-Y;xEhaqv$(d7ATfN{Zq2HNpY=F zdi~3M!Gmx_v9HSVgD9xD)lmb9>pP2oy`*Ke zJV&|z<7zR>$)5wv&ZipC)&8D3MTC>WC8G5O?|VDpKq4<9X=TjNhr0nO4VDGh_nIf( zr6;YOnK^E)1Y~|cyF(iku(5@lYa0ZsScLCjW=^)kY|8rY-rf8Hu#S5+{A)H%`{}1l~GRY{J>Ty^_Sdw)gTibb+a92KUdVHw})%uatgsOT6=Q zOZ`sLJLCI-dw8P4Tc7d3=<(owf%A;WB6V!(xnlwJu#(xUy$mES2e~gVrif>lnpnMQ zIQZdxN$SbV>y5#wT=`rOJ2vs7j-D@+#jqfiq}TO>^anVQ0%)X|DR#Ox4YkRCk;!?= zZHBz(SDv%-ZnN&HvCD@{b5LPLRRtKz_a>16;yiSe4qH9nlf75>#>f)Wk}m!N9`0l6 z-l%PAP-Up9KJWRfOYm9q-B3AjBOmx+IoXLSgMwK%OF>9TZui+-u@PfKs?{%?P^ME8GJQPzxv(gX#*e?)$X5J;h@mMgX=#%QT?G%&1;72rGX6tV zKc2P4)!1uDXyPa{971}Ti*TV40C;-7Y+fjl$(f#e^&U3SGr&<~YQ{Uy(5r`iJ#y>D zKV_~$dIn^M(Uy)KnRy>gZbpE$x?7Gx-m7=q`R7;}Bp>NtRq&g8LT6eSJ94dftWGD> zf8)h_So^N65i)Dn&ckq_uIv z#`Qhp*ZG4tJg!Ok-5h@iY^UF?T&OeuLn9;fb9j-1{g+IOr%ykgURF>nzj?DYX7B|6 z5a_L#k+`yG(yd=o@nP_t9W?oPYooM8Pe`RhH8G)E`&LHVZ6HU1si+XC;!jFN<)ruS&s*>>du-%u8VC+hB+ z8ek(Z+lYq!cg7-4bvgO@I~TPyj!yzO>ql+UC@sVoFW1&uSBe(R z>&zHem_VWyGvmoha(?)oC6E5OCOyF?N}Nzzp?7T{CMx+sN1bmsh4dJY+MX4=$-Gy_ zIp4r08QT`&=|KM`=4)zTFhAIKWJ3R; z9oa_ID`rr`s7-7M>itlCHRYQSbr3OfC-OPmE4=u`$+^0EOp;H~mP4Ll0+bKckM5@A>`#pBQ%=<);v0MV1fA z5=5o2uyC$c&lKWS!S>5xfRL!G!^-_8PE28$6`XP!h|p-uf#F-1v#Ngl@b6f33!i88 z5lj-+iXgzt026{XnGKT2p#y$?iqo}tu9QAu7^@2bJlgs%Q%wT_C{hH}ME{tG?L=St z0zOf+=-8j+IBAQ1dX!k{5j--N9SUya(#rn!b--J-z22p#JFxw@5;_5}&;D+&oTA}-3NdmZ7 zLXKGZKI8*7Bl1_1M0P!;tP9BuOx`DHFLG>3@Bs?3Bez=It+w|=A;=j2o|?pI^^Jg0 z&gY!rA=r@v_j7Hi=2isaQaA09mJBMczjJ#=^8L0|CwzIE6A-+R}Z zoW@QbM?lD?$wf*Oc0E7eVK>E9qf57DfELen>&MBj z2Yl-!@fgVHCTtN!TWhY_o?EnVc(yFM=?3)H;K;Z;QW7Mne~NfpA27+RRrN)!MhA@- znNswh^WX@gF9TGvlwTT~^Vj_)81 zSG&eWhfrtIJCpvc(th;RKa9aP`_Ex$waVHxYx_52aut%v{4b_x>}{r{_NGB$&i8iB z2iGf2z2c=kiPv&=2 zwb98MJF_OM(Kvy2~4bb?{~xa*--o#gp{i$=Ia*=9hgmhoG33!O6GHw4<{6` z`PZq!HF+5Kz^hQHuoOntjbFzzouX6+FE4jlHVDe@9&RV4ub0l}X$-P5@02Om9c2Fq z84O{%_Y;LMRlPJGE6I(zDc$FM^@WbV6R8S``JWRcbA3~2yb6+%U*A)B*;tx_Q_FQf z8BTpZ{zuFo1l2I^f+j8=Zebw*`+Wn;G`+4*0ndwOjz>n`Y3l`uwH@ zTsF7UEMwZsxxn0zD!OE2MSTVb;>DXxGp1#2w`M^9Sd%I%6|l7i1JE0d_w?68DIo*xhYlRJtxbuI<;vx$ zv{F$cZmBiTZx|Pe0Nu~05s0)lr7-kssPNi{AX9vjCYwXoEGDE^8rZk~&`+ofXy!`e zX(S_~=#8%(t61mlVo)b;^tvI`P_`?SJ1qN)i(+Brley@yy{N%-2Np3Wgkjvy- zC4+WX(BO$RUcn}`$Im~Fadd~Ctc{F*9%cgzzIM9r-S|OmOl~FiIbr6p3tiRs zRVTNOmTN+-(r0%b6ey^Za^4}Q6Xu*TT%l~KSM%~(Mx*Q;o?nZGt`dIMdg{l$KBKHz zz(B$$Xd-e=O{+JZhRQEBt52s&{=<<_yq!nq-A##oZEt@uby`)qHQQK(m&&gOr4K?H z^|_z=BflNuw_e4qr>sLmQ>t%;D3PkId$oY{$QLMqX^zxcAjP0QcCTVQg zc&0a4eF>-OT^e8xXE}oipPu{Xo+;Bt7zy~aI0N-m#h5wnSNV}&KmA+1C10M!;n|armB?f* z%{T>2g~L;@-DQLk(U<$O94vSF2fZ_%?~N^oj8m5ZvpnTEUH2nK5T%~SzLA%**4wA% zdkWm+#srgk$KH|gSfg@k#M#dLj9Vx(Zn8q3h}N&7`@1Qq0%btjvC6{Hp1=BQ0hfJ4 zhb*rcRI>~y2L{Ya+)`fwh^v6gVji$PQc|PoVPQz4{Qowu0hF9Ycj9C%=~@T>6B}|%g@e-07Qwp_ggMB5l4GU@3wBk zB_Ds6#KieM>^HOwKWt1ucIPWi4HTqqJK3%G@EoexE-OiTdF2aU39PEBqLoDd2)Vs$ z>bCBeTY8U2i)_}VnUz)9%dpVU>XX60*a_b6O!RX5h-s?=Ubuy;Z8(?vkDSvY1b2 ztriJd>3Za+j!1TXwUn$|s--vARIps>TsbNt_YA$g%trFQR!Te|1{=s7`%-@OWo!TqI;Laae zRDEA|veFsUD3MdYs};4&62Bh21+e?H>`T7Ps2cUDUxJ19Cs3I0T}s^#L4){tcW3v@ z9lI55?Ff#v!^Qm8iGVpieC~hbB1-Z#(F!>g5QrEam;`Jfk=^4Bn0MzCxzE+VSw1Jb zxZ=ssGvgg9FFr*~r9|b8h>5g^<}BiVQPiFkptzMXV_?Hezxi6gvpZaOXXnJ!kEga( zJ&rewQ*xWmk}D(bNE(Q-V1nPb&fSgnWVgC;j^R5=LK4%1KAK_=%vbB3O))?{N9{;* zwHhxnwnv1K&u`iMz6yG(afI?pM^|H{xD6RwTVWruvD&iB>dS-ILwgL|b2YT~7e#Nj za+f|n4s%HFxh&r2j`NkUqUwaQ90pYinpg5g8-pAs%Jg)*#P_CU0`8G7lp$@KL8oe} zd~NyII9qYs|Bf0>umXmP|CTe?ab$E*+_=_if}tcBA=NYg z$NgbmC~EUQ`=LXg8_ki-5Nbj6;&)*Sv-_tc;8P9krio9B{ie7OLFMtuyoXB~zYwX$ zU&G231TK&kQmSq%l}^_wz75ptIPxLgxF9>~W_G;!j+1QSKW};@mzH?1@cd;Kql=!WpCD>@9OuK3eFt+nw z6R6S0nE{pQs&S~Qo*fO{Rr=e-iR%??H$_cDx2geUHct6VP^yPrx|55GEsp8w-`ME4 za_ESCyUkM^72!+q96vpbau)eDI6VHwG@P*VCC+^Lf5m_O_JQQdybKXU=X&4pg zn-Al*DbwgKSDbU8X}y)CmXzGDmV1QFS<`c>Dxk;v zi;~vCO|!7zBK~`TB6y0rwyWgcP**4`Psllt6U|rUSv`Dtl~Otv6lt9Z2z&6B#boo? zykm#s6_%z+^0}>_Mo*JMVXdv(^*ytl{{i}cW|AoACEQ?iR64h^(w(kk&7aMiX6<>} z$Lj8?wN>6%Z*<51MJ)O)C!KoPeGh#KOssW6Olb^K6gsNXVv8P$jIYcC$Vh z4LkyEJlm8byC@1_^k@FL~%B~zVfkqaJn@Q zeA@l`*1Ge{>4fR9r{NRm9hIS&zH#(dx|tdeU4wSXd0{2PFD3@KcbC$C zg%YH+eIEDqQ*vAVqC7S{PPw<%!;`pMdU8;}lqs(3(~5Q<8wdK}D4_HT9-p06VjY~i zT%hHIKvf>wquRFQAK68Sk_!wHlk_6L#57P#w?J05`s>rq&rtVxt z4{T3_xYJ}`ANDebLIPNmADE=a3rP%GH9&cg6#;C;#;$r=0R*!Hv&^(BGDyL>E3WOG zlCCv@fX&d`Kie_{Y2h+z&2s22D{n&uwM-20?)8EA-Qz<5=kJ-3!+u7>BWYwQE3BvH z_0+7|ohGO<5R#J0FPz)b6SKI$uo1HgcpxgN%NB(8KxJzeo^IBv0AhP&kok$7q2a;l zSC}RAyxK1EG=nDj%xDgS7DaH+3AJ+7_FtswWb9Wx<5qsW#}J?K40zGDd)FiyWttqE zo~8v}9+pSB(#~<)Y;{L6n{TQAhecw|<7(Rzn^_z1Ug6rVA8`Ws%ST9G z``t&`gnpgqHeu#F+)DBdd(_FO{C0S&yOEjWA9)?Q`RVRY7ngSd5g!T8KSOL!04meP zb9A}d!b3^E#>NT(@6f+kOyJ_-kwfO+L$^j&4zmqcZoT9s;y22BLBP$qmwZ=3{EaHU zu zk9JNaB?8f>ulN7PaV)=|g{%quV>cQg?G{Q)wXG<+P3UI)x?>afWbL8um!c?N5$VqLG(zQMK)1jpac{N6YF}UZrJ^ z?Z->J^40A)4FR}2nrJ{Clm=z3ABGACr|hh_-sspcLQ|o<;>TOvtNjS_a$nzd4y&51 z;~hGc)eNA);dxXLoiCj4u=k8p38au@YUbJ)q3>TQ@zkna; z5`DvrjIt!^5SJc;rwt|JaVOr&a0@j_FtSz*y6i>~~I*jMOY<8n0gXh6qixN90q%0i5oA8d){*>sIr>GXM3b;r>_;7eRGy{k0=DQO1> zS%~r9tG**15shobHE!ds$D$CGi$ussf7O~}(kqkeaQx%M;NGEcCm{SuTDWaKLUgI` zyQO}4upS!XNj&lzTEgNve{z6tnXvDal#=({=VrYp8$-`@_sjHmv=caetm+{KFb{*{ zL1^#j=+P25#2NhIMmsr-XXDhmHF(v^gLP?qmrDF`2j$PEUE1Uh_?1$4`O~jC*oinS zD!k6q1YWPNp9l(@I?vZUEOt1`E8fDA>8q;RsQA&F(1R$Jcgu9PX_UBTxp%!kHBDl^ zzk$`z_I6IP(d73=2Izkx9KCt0h0qLLh()qa;-iVouUs0?R3hQSjp3v<$Tso`JTeG+ z3P)-wASVw($gBU2HW{jICVR+^ic1eAY$wadNJ5P(di+%AvS(LeU4+rSQBir|>eFw2 z4oea{u?tGMVvTfoe{xmmLuYj6M>lQ1LhFUyhG2fW+VgY8v|?2b-(724^rCNsraPJ< z5A{ft8d^pxQQz97Co;yH7x-Kb^X)6)l7OFG-w}UF^fYv!lQTt{nPger<;mmEjB>`G z@Bah8Blmyx{bvT$CUFQiaX%z~O_iUE?r?LjydPBbyPN$EMtsj>wz%R+wNAIcN(&w>sP8QQLgFA|syW0ncH$D~O~9f>2_n2EL*X9RXWVH7HA zG?=BM#}D9HagfpC5pwjd-t#3qQQz{Fh-|){Y_t?BRoH0;b-i9 z*JsiQtpz2d<|`;ytTW+1x|2`DM?t|40zJO3!DH*B9h$weI{`2ueB_XC?wPPlFU6Y3eE2uw?_z6DVvg~Jtw#g(zW`9dD_`g@_ zkyRCMsPYaT9z+qZ`^CQ={2=-N@eh4@V(M<~Xycp}=3hQIm5a-6KINR{xGjRX_o8f` zM~J7ZzS9TI{bz{T`+h>hzlL~g#1)_Yqc0DN^MCb+AS%KhVut=4qY|!(COykGKvB$& zmm}jN>%ABrM<8@N(3r>+$M?lVKfOYW=Cy!N5V}{+|Ep)*U!08C4oAsDd6FA%UZcI< z^o830zcigfpZ&1UDkqtBJ}r5}p6AW~D6aqQ1^NoauUSXy!8u)a_hbBjnvh^KAB@z3c@9n=tF}oZLwU#-OaqhJ z$d`YYo&vTLN!X1rKF;PHUey1e)TYV5;-|h(9FcXi$vJ2Kj?n7J!Ym1YucmuJA0~*i zN~*^yQul>gmI3ib9{BM8ZybEzD`H2q()Cb;x0eEh1A?|zC`yc^p zG!e`6l)Ts>apDVC4SQ;M&Ie0{9Z2`(VnO=C?|q7Qr1 zm<0Gw{w>8Jnqqvgf39q%eNz)*rT@=bMm?zsT|uJeCh^UwF9{7c?cEFN`K z&setO5=KE4B>arCBSL>4`{=O(LY{`*%#6I4$N0W>oMq&S#upN!PTB$9&# zqsw^u-~~`>ZV-n@s+6p)@p|e%oe&Z(Mtm5eG&*BUjcw4J%8hdp z#SpvxE{nby>{GPl`hpp1!N9X=Op1SN3MI@F#Qv0LYiR7dvzaHv{9{5b6!(VNoJY*~ zb$)t8Y89dhyM~fhpYh2sGx9^4+9SEJG)oT)6B~#LH~p`W3*;{ZZ=QtvKbYqSO4y%q z`b>E3X3l|Nop08Q1ixjWkAaFZ=q}$cr37$vl~-wp@VIP|Y1C#%yBU+gE@4w*Y!P>|dZ{_7T+&4`IJ4n^Rl<#hUI zxFk^1Y-eLQvbd+s;oSx2JxCZ??9)Lz@#a-OS#{}Ymi$ptkwp=@9(({48nNT;KzDr` zv44$x+KNqkkVo(b9z`SynwnCka66{GyV zl!=4Fk3V{a*!E|+P>Dq12W~ebr=`IADt5mXa(7~%w768lPI{4x>!PM%1Y$@t3+ zXJsTeY;0E_@2{GEmezlvM*IYwG}7HwA%fdvSiJcY>oZxI+&GzY@xaKZFqu)!t8`QO z)zuOyNogevMVh5Co#AY$c$34^vh}BgKOEG7Hf&);5ep5N-3T^YtPJtrPA2Vfxs^xNu zmVC|3P_LGWqPp9+sq-*qpR*k%3v(N_z_{W>aYd#$2z_?Ph|_YjSS%qbrYvY|{v3cR zkjcv8LUJ~JGRtIPX{FTRb=^S4MBbYhlEf|dXIdY{f-wjeIHpxH`B7?Y5VAg*zIAY` zGxrt!Zj6z!jTSbh4T@r!TADUMPFEw@s61chT$0UIt8nj!L&L8jO^_yP@#R) zjMVROLOvpn(GLBo{A{2-)($Gk1GD$+q$)h&rwgP5!_&fjMSTUWgfm3O9;S@fCwrA% zhGWXg>RTs>X2K3sO5GW)e!>TRy`T5Ih~+tdqZbZtzmptskK6pYD1^_)gPM8@2h_bi z>5la-gVu9bacCKvhx(VRBm-|1TcXjZhoxLyoir7is=KPFYGRik^aCktv=qhS$bYyH~ z#Be(_f+H&+)y}1(xCFe)K0OsW5{O^Bx-!a5k+5@j*XT4Y(?kUx;`6K*FtFk@e(w1+ zvI=rekL!E^o78S0{1YFE5p7c{Il2 zVl{thZEUQC5#0vy@2A~LDN^NYsy6t4AYlxKb4n?)9#+?o`ZZPrhe`0*6>j+*85A#! zH1a0F1@0h#L5>XX#B7QylZ(x*KK>3E+J-QA)z+ESRSmXd7Vo_9&o9mnq-{E~Qf0&)%TUgrW_Djt_WVOqa}r@S%B)bRK(a_aLh?oJCh?!uU1*RX0Kq z)7-}Kgwu^IqL}DYrGrmv5rqt6hHX2~y0do|wTo*3rbgm5B}#E^jX)_#0NxyE#$3YVG z8+p9qgOzUpPEgaIu1L9W0#9Cb$R!#%&ll3rWV+s*no=t$sA20&*CV$c4j%3!PbGX= z8mTF>t+yWEz>~wQB_-tG)>F_PTJW$DMKQm_mUh;N6%|!O^*mJ?csjNEuf45i=rGVQ zvW7&}o7~4%> zjyaNk{Ef@}@V-0>eY|@aQ|q~n3hm3>5HoC@#?0J2yI=!1 zmh7B_v{kAed?v&D{@wGLm;t=hhQsh6&gi!*Ws=}J3nLbKtJa(VU#nqK&r&Y*=xc(WRE0L|O%OG;b__!G*d$@B& zwJYPUw8g^B#pOBm?16?G+M?)=RFLO+I6-~PVXaj?o*kbn0OQ~;ooWt~`j(s$RT*FD z^Mu^U*t~v&kN347)|{_^_gSDo-u1_&F0=JEc2{}t)sq|wYA)}@#PfR37fC{JV&aOu zKNFVQ3~{?qP(5!J+Nd?~M=ASLi#m~F&(EFCw{SuvWiyQEd338olqh}G00*VLABCR< z2df%ieH6j}UNvRAwA&EHV(K=ri2H1QM4(-fo0h(xK0<=0LEB`aBUIV*oVk*NLs8$Y zX2a6TXJ%DV`-QnmvH(t#I(PvI^i(^y9R@-A-(AN#Y?zsooNtjtN4-fEKqqus zw@Bk`7sbWZNnL>ExGFKh*#PPDdENR~b#seAuk5 z&-#jdhq1)N1(|%oplet}+naumvz>^k8N9vY{q*|LSz!r<6xGwL4%HMufLSMBZ)}=gn+ld~)GS0tl0wr)l71DxWhhPyXdwMRTf@h21Z9gb+!ci?5!Y zxeVIhs0rbn9p+lk>s=i@aK!@f=1kGlPC^nbRK!+41t)_YfNx4_f(}AM-tdyeFjLBY zlazPGlJlxlGM7zKwY$vpP4pzje8VT`yT?(NQ3p6Q0M2VJCpoc3 zmus#)UR`==chxT=zcQd-dww4;;`zBt?8fbpl*Z?6kg40*u#`&ntJg+{>X{K!^lzwg z9NpbxQ1E0}C`|J`wKeOly&@pmKJ=J6^x0-jfgGV-I4yoBr*22bz1J6TS2!#^pqF)9H=$>?ITEJ9%V%ztMJ011Rh*HYXp6cxVMI5gTxp=H`p09Au2Qdm!?-UmVLISq zhet*-pBn)yJ8$)|TzUOCW`*)=pz6qL{ zT0(E@=^C3dc5H0)R_xwL21ZxjB^(-}QBZg{UiK#hu<;?rXERvM4A@r6gGIxQhCp=4`mMniZa z-qagz_AdsDt;4lA#~@nkZ(!-(O6GE!?8>Xz;!e@WUh=`y5_$rrE`4&=%J$j1Y?Kfj z0!w#!4Yes*W3O;s-6hHPodXA)pV?Q^1S5`I3)m>jGVd3JMlZBuW2-)?L2Il4L}hLK z5+SvfmH0bJMfo1)@UH0YNNM!YFw~@a3gfw#xWYW<7pmag#`x=DTCEt?OBs2U{gg0 zvkE00RB}ub(=P&Ja?t#5u=fv=Fk>Z`VtnFABw3Q9`{smRi}`dp)baeyfIhPy&CcVd zn;1Sv?(7`0iILoGd=Qruwz@K<_gG*F0GUp!;Vbks+fh%>6w+~6PM0>po~Tpt2&r|z zfr!qZia8Yc47AP7X*%t}vSfB^H+*_mZi+`xmb2}7Zp?kXfVHs#6gpa1*%oo!Z3;D+ z^9kfns8iRj+)fZEs2GTUC7rY(?ApaED*g)(u{{R#CskKF z)$-n<(o)$xgucmHzB{5T*)pSGrBBP@t%owaJ^IMLK1IiIfle3PZR(O08;)tFrOrA0 z2`?KCkS+3le6f$tk;_)nsv1Zi(<`rkx(t0$9C4)Bd-3-kIu{oMU!1FlK6>sjVQ8=H z%qi6CUC}s~{g5Ma-~(TtbUoSATVXM1@O|&9vTs+>7|6(b%p-Nw zn-eDOM2|zcuP(~KE8w*O+jzE`B>&OnW|0&X-FM*z-mIamZP&Xie5hCiUZq~|Vp&8P zG&iMivjXnL7a~7EPH6T87vZhN8G|axHtT~RzSmcdWH2t?S%+t9mT4EPjR!jvD!KTc zi&Me0+TkOnEHm^=rDIAAyx`Bzv0elCGlLF@-bVrvro)6Mr|W#vVr%2Nra(zM+VycZ zTD8(L?{=-PYR%ImoQ%limhAg{57qN)e!A$%!mcZLUD}hQ-NE$6pK1QO`g@Zmz89@e zkcVWk@+{L!#=v6=!DGI+)!0R8X>khuOIp12XogQ*Gv*zFD}l^Z&5CY5N5fPe=1Wt< z&FG>8B0CA*C$pZhk92%xeN*K6y0fmm!fk!_nr65@AGlqFB6~%_t*u{Vj9Rno7)<_f zS!b@hjMH*QYkpP#gTQnI>S!KvBg@>HSDR02_FS?5i* zj48wS81hY%ch1I7wkKMnk1x_3Q#j*gG$zo^XBVZ0Cnn^QbOoAEKWy#hx95ugDmb3t zP57x=-89Q3Q=XUJ6mY`a)?_o)*icHln8wDClGB&?Z8LIHPAyGE%@lci#y(Nnoy2d{ zx2TF~y1JrJRI|b>grI(-SA5eniJdJN?!w0_96mnymXF#N~$qEwoCC z7~l65T}_Wvs7}@TCmmk*RQrNB$ln#^l{M_gvlcmO9y4jDGXMTk^tdF!om2>>%gC(f z+K==#?F|&alP3Sb6w!|24O`=}%_5%DS0Vk`+HxWU)-$Dn_;76pZ!O1WujN~9ii%E( zUrshc9tA^B{&H(rRRsoYCa!6N)RgpO&H>NXw`!luXE^r!>ZXr{f2q;FsfknX+BD&G z)@|R`G{RPPc(UVdq4Knz*?*kNZtYFIYD#rTe@D8u=&oi^`SnSP=R|4CSp*-dZyL8_ zzmyaGO28$*zaxH-r=kyV`JY6aUlRsZiaoNwR)=SO2sQM1tzk)nhJI!v zeguKX5*v4m@EeK2+GX^yHK>Ob&Q*V8gT=Q+TN@#^=mGs2b8xo_pqoKMo9B#?q7_~} zFe-54M|8lc(8=8mvl316bsak;RA0Sa7j5xF<2nHNwts`VC5&h>{s#Jjke_dz=sReh z(+^GwTvCn1#Eg$n$g_+IX8Zmv^9#Sv zF~Tqcfs7L=oPu{JY<#BwHcEI&hcw`wz=_QO7MEVBl3G>UD^G83j9(8 zTZJ31=u1H1^4wMX1e5#IY51X$739$*tb?`M91ho#-Ny~LL=3-5U1#m7O=I4+Eb+UN zCQ+Jt-K-1b0jxZ6Q!7@?XS)BC>lqpS+SR8*s1L4hA!EA6uI8SlVZ$V+_TKI!O$|SH z)?ZHJ4#x4-CyRnKYgrne^_25c4GfCu5CAUOC=~|EOBxFCt1FqqcNbsYz^AfO&>XC1 z1|p%{vZg0wDJrip6Rw#|dit^vLtolucEWA2S_LarLU6H*2U zmip*+f5F$g;)E-9s9ZV|0&NTx>okJaKLlMKLv@ZOV@+*#O0-ri0IGpp^wVkH=j0a5 z8s<=15q^(_K8_ax0_HO9mqWnxPF9DcZU(tn4&bdZ-y4?LV}+w%L&v*| z@wc6qCwKP{KACGg`?$ci6H?DP3#|9XyAJ4b4!6Hpgu6gjXdMuRFh50asr|I)cHHP{ zjt+39JRs;?F7CBbZ~EeM*tlEp-;#U;V*(pa)SLaB>u!ua3>n+yUG#%Z1g6O$fLP^U zv&afuq5b5P;2_%fiWkL2xrN{#$<*_(_M58Cr=k6oybuM9GLSRN_QhUSG*n{RR0SF6 z5(kxA^1GF6autD~4KYYmenh}nmxhsEw>O@}(aF(}I93Nh*C5mX<|g#5?~9VJl&?u{ zu2{$`Y;#$7imdHJ=W16zL2nk0t@9=mU4p?}qiC0|Z~LgFlk_umh{|B$o6|XdVQ03J z)9Qzf8oUjZaQ@-rnqt;Z|IaT2i?^v_K#Ey(b~|$JC%#L1*6hzP;!qpj+^Z z)?jQ-h}`!!bmD!Me{&uwF>Jc#0IU7%E_1^ov2%*3Db=o;aU9Vr;gG(boNvv7 zhC8e28BVRsI`O16Rn0k9M=1a_I>nRmGi7K*u>;W`W*E*4^TOAohN2&mzcBe!u*E|o;#P?-aFqNA{ViDjNz>ruy6H~)B^u}~weGwth- zhj5N2O3a~<<=b{>X(Y5JDzUPYi|@+FMsWYTBH(t_Y>w5#1Fp-^(IPD$a2}R~nUtM*_laR%wbGpD<$UzY<-M z{F7A(CU31n!%3 zBWb)YGqhZei6l``=A)mHFjC_s7sXTBTG0_v;UW83%{W^h@;W9`cnY_lN{E$`o&c!W zNkxJO7`A3a>TeIQH{K6|Jb@ROGq-I-2j@_-exx~`1Q3tFWr7MPoip9-aPAv2ia6<7 zTZy>|Ns{5XrY&PgZ~g6c{iuiND?N{$!bxmbn*~ycx|(ih>lvML^T%&CSWebg=L~C2 zL|$pa8QW?FkssUoBsR4%EB{!}>(VQIDK%hVeR zal4|2ROxAJUlOfKX_b%Ke)|@LfR=tLY8;nn*ijE|u~FglJldwp6V z|8OguAO<}4Q$@E}qc?T@z|Y`2=BOTkL{qc(7@Cer5ERkHc6ZwFi9CE83OOy(# zVwU6SD%0!68R5O&lEeZ#NE8Ha$|$Id+gh9}jcJrw}8GIYa$$}2vsPMVl61RxG;&X9ATG27dV zkH_h!XoQ|cnl(Wrt-|YP6XnuYbgi?blnU?c^ilYZO6!%mx}X)f8Q~sYagU=Gg?Ger zr`%toMe31>GF}o2NOtd2@lp&4YXcoG)xJmX&R*Kq`>)ET_2;~4B!gL#o0>*{VR_tB zU(uiqi^?{l)|9R=*{A0@` zX7TIxqD^x`p2pLYlYz%gw;71>xBV;2P4Pz~OrqM}%STdmr&UvCs-wzi?~gjIwPG1B z_JYyLU~b7woHh?uVw#s){ePMr784ZUhVA3tc;Sn*Uj-^G|2RB-cVmSW%`jIl(6ZD4 zJj~+6@yFbMwO+EN%20gdkA&(5tV-wtT^lm4cU`Q|yh-j_B4PF28S%|?nLKYAr@zte zEB42pbC1S4oP%>fl1iz6NwjP!(Qf7)(xytN<(oWVkq)6B^Bd*~RAaYT+E2W)F&p#K z28Nn~jdO2;3(OGb(yjtu_mD_>M%|&efUUy!ytEG1{+k^ z^{gN8?jj2|MDE8IAaiU?{gF@r7O14*Cb@rh{wk_F!Voqxto0*Cs5+Y z(%H(QMHE=nQ9z&z<_gDgs5=yH8%>8c`F1{qba=!7M$_t`{vqy~ zYa(5_#T%{E#Z`~ve6KClSVrB7w7@Z3Ust-?$*`81@)Ang^10QKi`@^-z+>N=f(5NQ z4<9R9Dn^$|ZP))AwFPokinN&zr4{vdX_X0$Uwb1hCvTuuBpj=(qh$Gp7tjNL@QBv3 zY(-p6Qt}7dH{ivpI5`ZUm85v=6C`|EjZRN^Vc~rNQrKnZ>BdB3Q|t`LD=$2yHbY~N zA~MPJ(%0pr5GR>1e`qD5|4RiTnau?=s$e_B9WH=1@UZ5Y1iY7;yQ71A6lY54Bfc0S zX6n}-k>Alyx_He&O=KtkG~8Wb77!7gHphrEt+&V~TIM#fqc$c_G_}2TqhDOmsGgRd(AFjS zLOE0ktvA`z0#JlbH+(NSLS(CCQ$vjWJP;!&Eg1AvyYMb3JMsx`>^(MBt)t_3&xk`I zbaRI!-_q(FFk^pYI=7S>N?ZBbjM>C?X*TO;exVF!JK_<#C?8lcyuBz-Knw37kHWbNsImT|1?77a!Ir}CB-5C(3d+k$;rtext%$=qt zo8XvxD7Uq+nkmrnvm?}eqAfx_8X6wuNF=e^DhwI!O(N*@G-=2YOpG+*FnfOwLH&PKjHO+KC24RZ4OkFewJC5=Pr%thA zf;RDM`kRX->cIsGuKMG8ygR-zGIMS2>s^?H#y|*y`pH3MVffu^#K5vAw zLs2^;X?Fo%z)`%>NPvz!K&-*hyl0DxR&+m#4D*Kf{Cl+8$biJ7680r(3JS_!K;L6a z`O~r!A^{6$kUZq_QCj0E|0Z8b>~pNBl4_3n2w+yhY}--{yn9(R!Y2aK0dN}O?IIxN zjb=Fcp<7oV$Yyz0Ab9AeG;(r*h!U(N%3bEm;Hy`xWrDibd(&mB0VX++pRBSi*{UPJ zrq{`kh6YI3!)Kpwuhuf})-I&0t9{_~b{DWg7#lP^HOV8XF1^L;B-jXwK%!9{vgUm{ zEc|IGjE=4%5N?cFe5MDTAnX%_q*%Aw2nSFd6BZht??ioESg8^g~N^w$n?0qnlM+G zT_P|4d*(^+6M?0MjmU|rXX0S z*k5J3SkzANTir{GGs0ORtZ$(UpW{E8wH`J~oY$>*rl5h`%D(BU^Cxuu#5+5T?tw#_ zn#k7m)h6e@>ZYM~kz|_lJ_rCt{>+~vEIJ*g`}Q~ktQ+l*eYUBZ zmQpPi6NRjp;C_n7;pfFGO+nR_@r`b1#&Qa(^*%ncte!#?Y#fzJ_B)uk;LAmlVnpzv zMvPA^0g#+T-Kx=Pgmv>w|6rJmJZ82ldc zh=Khnoojz4AZ@O=S7&=YlLlZJNX82-m%wyTiocU{&O{3fZPqcM6^QFgOcgd&z|_*| z`A&^XMKyBZrC_{q`?`3|lH;f9qMOt+BxL0&qNWZ*nU`F%&UczYJ8)U3F)3JF`jN?QU$nwj~<%X8ewJgRipZ=G`%gv)r{p-Y{{jP4N^1 zTlt-0ptQ965G*DlDY?hEE zCg=5E%`Rr$4yZ{9y=i|V+Qx?e{(R1b(?XcwAGiUHa{7XTgsMSqgKvJCjGMbJ?VQzA z{V%@WI;@SZ`xGIJ%FGiT1;XYIAu3Wmd47l@^o!y9RwOrZ!f&f22>%X!18$3npMWlA1R z6`kd@Ky%oeZ9-=3>!%VwT5=RH^Fs~X((?_KV{t98RZ21)ndJSKfJ^SE0@uOd-OCHB zSokh@w_n2lyN<-RHjkS*H%C7W`}FqoSr7O<#AoGdpy|w`qM;}54Btj2<<|u0WpY&0} ze~4UCPbL%7+>M7mD(36yiXaJ5qAnG!nN55ZjnZ10*j@2YQ{sW)vbX{kQJo4=YyKtZ zn0-Rm>uDodKG&22Cmra&$5KPDGUW?9n8y+h2^YUyHCy%u*UI7iT&DUS8fkRlp67a~ zdG;!4J@4gO((QBoIeW!riDH(t?Xa5t;!MYaEI*+-QJ33^%mj_Z525Eb&=jV>BS+4anlw~N ziD;E0s!?!QsaC=q?dgru%S^`KJTcE|#Eu47Kf1uj<(sywz7j1KW>6|%EtcZ}*0q1W ztoW#a93MEQgBeAL+=VJDTRge+iy>g=c(>;%Cco5aA0-xG5x_{sqA?#Klv|+!PL34Z zQm6{Y?6_%)4VcCnHk#G5L?Q4}LZy8Fw8^!2P6-xGsi_xj&qoEHM`M(BQie%IBsBMweCXFG5e~ zTzJctg0@@cm*}oTda*F+{2tyrJR3Mu@nH6N34pa6do9~Kh_0HD*`Jn+m2JNUaC1E7*gd zuSZna-MoQychnmc&(Qu$s&7?# z8KgQD8h|SKj4;8FmSs?4wjw26M(a(DCx_bc(vj=STeA0-$UVpH-*{%Ly^i&qBg|NI zvxBieSr9#D*uJTpuu3@+3F9^*lz55S*?Fj)`5jQ(-YyzmsWonCX~TukCMs6E3kzRa zA-gY~n+jc#zs=k?k`F%=`2neX;v@9?O`&>rm|xFnZjj3Jd*qBpxSv{vXM~`hV3Vj{ zXKzt0#1~;k#`)doK-(Q$A1^x|v|we9~7ErwRtw7$JTS zjK=esc7XzQ@KDV=aIX%)v6oIg9@)3Ej}YDB}}Z?BI1Whw5{0;+&Y!A z!mP-GG*VA5Yq-<)=Fe7pGtb7PSQOtB)6rE@o%;E5d7q2j8}7T5m6oWJtOrzDKhLuLbD!L&Z|La-sB=urVg;1C5r z`v6_iM|YNnF&{R+$S$dF?#(EFM1``W7E+x#w`e*ThLWirD+Sn1es- z;jsMd9uq^h|0d+HM}-rom-`t4R0nAE51n*T&}X@XmHKHeI*;^6q-yUp6z#XJe0gSn z26Teffoh!XE~*yeKS&guiE3@*gsXRO z!iP3>?j*C^o8W-0{+{=$|d@byxpG;mTs{WWnt;yfwtj0ic&9Zy5AeTz8DDFF! zNe`Q^(n$%iG}7Z4$;BSzGdWPi*gmt`6lb;}7haL1Ew^hUkPS`z>`st~*8k4Wye>?Q zRRACT)o%7*5V+hTMcb z@l6U4_Vi@Tp`tnS1AW?C=OgKMR2WOPHlrn7B^Nd5%?i+e9BZC~T$U>0b?5nt9_JpK zoRAF^d&)hw%kmt3At7^r6lrqIGOl>{1Gk1IUKRTNY z>#N#cM2JF9jk=vH%b`z zXb9wtG*QzNu|+Y$uFG;hdEbn>RVVRWj%!hdlL`w<2A{`&Opku`l<~k~2d|!8UE?jv za0$+9dogjSb`ZWBnY|;3aGe8F#%h&#h4I^T1P_F37lwuLnLk(#?kI>rU_Zi%b;36# zKy+)Q#cjgaB2GHBS4F9M2dj?a0`q+ zn9WJWkuAkgq%kY2cD~~kgn-a>EJbgIDPmfZJQrD8bzgmLe2nq<#p@Zf@*|GgnWN_G zQ&~4j@~dF+C#A<}%7R|nHMXxlQ=Lt1Pkyn=beBQpFeK|1MR3a`My;Y2|6xPLuYnYJ z6Bsldenf|yb-kn6IP*E7NpqQ-U1}H(zwUO{Es-@apOYJv`^V$cZz5+r`+f`Zb!PRO z*!|+LJdxk*e(&6G^d9JcD%Ox@K7sU~cFTnh#OvEDh+;{-;S=_&(6*fRCBY-1V=Jq} z?4m|wqllwff+s>L`HeEuaW0-z%fy{$+B$}-wI6JBLQM{GiJ|AsRav2EM?8w14Yj?A zaI3%v!2^WhB7_eN8mqrfq}9e3%stTd4&Rfvs`0r$tO+movuKRY?mi$YYFw<~zx?_& z;5Uc2Bn|dej>vQvy7`4~)b_PT{bS3$T}{xRmW)3nC0kyEv%mLMrEO_K1 zfSr~}2K5Y`9p~GA|EuZRpdvnA-dTaZ8($+2@L|`FD27I`*SqcE!;S=RD+PI~ zI-jIO)p2u%^@S^TE#45|DN+pi?1CQ!XZBwA@0LBS+8@g|ursD|t5O0a-$=yx9AC8z zqZ8T4i4evjzcn}@>HpkrwdwkMvHNpB1$hzHE{hkV<%2?8v=`+&x}atQnXDvIHgmP| z&gUIBP(Qp>!sNnwkK3xOYMKR{r!M@FEJRtdbuhikU$X2GH%N2Se7V!SAKAzIT%3W* z3_jiNSTg+TBYQ=KB^d=VyXvP@i##$sSvy8}eFTer-^ZqEP_zUj)a72G*|}U;Th|76 zbUyVeAMLtYN`08um(eLOanL3hO!^Q_lbfwXq@U*S`fi%V=g;*uv@hlxQ0dosACRyN zLC-wSNcBC!&Wk<}#0Ce4y_c=c&4F^?Y4fE18@`hwzv^|AqY3xD zL|#F2gWuZ6#1IS)<2<7y8v)g}q>1H#7?1YEETr@~lX>(kYJ-N82vD@i`7`_epvQhR=?$q@6+7b}jJs0smYZVVx=3Os3It}VmuBpx#08Q{y8VE+ zBZYruEaHQA67gm?xNhmO3UtD|jOdT?)Zn6Ne8_OhR%9sbMG~R%#uIdrZ1RGtnCHh> z&d!(2Kj@=;;5`x|$n_>}ix@7bpk+h4JP9u^KW%uX0|cvU=o@DFEeOK$NaSR zq0Fvt>Nh4RU-OBg1W&ix>&bwC?;OPQ z*Jw8tRUQK!!Ec6aSJ$ZCCdJtf>Zs^ibKB-KGL~-zk8ak{M4U*zMO_-hV=T9lsP9F5 z#+3O9Nk}XGx&ygrhd*(4;r`1dg(GGzIG?0R#T3VIpQ zSkHRLx1U&vJ7irrS2l{2;rk=;eGT}X;Q2b`_k5s!H#pK*Y2fyXx~r=KoXizG19(NT zM?jD~K<@9$DfxOm)}qao2*|YFmgC8RDYBd!bii;(^PFl2-J{xTuZ^c@#}gqk@^Mp6 zzk-2-wqJ$R>2z@7e5+DOkH>T}7 zs#s&<`Lf_|2+s=p3aXQb0duM&uzqG843$^>x;*GVYJGp?FZE(H{&ZcMsN7@D?K5d; zjl}zs$izZPdZyCc{CJFzFf?*f{V$O1U&EAcEtT1lCb9f*&ZC}-cv)FZi7_@Y-}N>S z_D~N7D630(qJFwi)8CRUu&KE`$q{ zUvv1j@_|T^1T60DEx5;qW#1eYnfclJj_j(ts#iv4OstQ~ z?HcAdglI1=w7ybB3ENy=UUlNra8A5L{2xu?2Jy)8uZ@kLWb+)yb0j!Dwr`{*Te{%y zlEH-{jLZAoRK>G->(1!s<&nChCqp9~6;f7_G%t%#lewMae7YIgj=ip*YA6|eHNsK2 zh4G@!{Az0Ka(O1f3Ifn+lrAAfl4M;8E=U3Zahww6Ok5;X;yqqRqN#IJ-a+8WUAFBS z0B+%YXlpT&q@5TdbTJ}KA*}{w8sQG1Cn>f#^@Pnk4VwOuw`riF=wQtAa$suPgp`BHOwmU^0%JuF1G! zQIIU(gIYA7bI3eX)fhV6%Z77`M%GGSB;I zY`sg`5(KbJUE}o~AgS-MhCByd(=XTzIgIK(`-?GdG53euGJqwU=430M1CreQj-#ZfgnOwE6lw)GhFN8|6dN z>4;TXw5JP}NR;x%y3k4dR55rhqD+E4w2_$dxb`w6nka(}Z zp1Tfx%rQ#h)Q*&@BFx7PTy|HPuZC#g;fxAFt_0>T57?Ysw$ZE_XCk4WqF>1GU%ZBRBfW{4}zm2?}PKc~dMwv_=E-Hq*CV4||K(V7keznJt4(|SV zS=I=nr%{55NOZDzQ5CplKII9E!!|`jGO&4_+IJsnt&u8sxad81u0l7!Tw2@s57S)B zN(fFCVwstZA$mT`-bBVcYp`v^6j!h1QzNHf+OyqtD%AMp>lH8ehYO9}K*js-bMs8t zbJ;=#!JMO)`pfQO`e7Z1cM^+NOJ$~9=Hf;b`Ayb z%tqln_deOzA9XCt6B7XaqXhsFg%{Huc#quOf+^BgRX#nnhB&CRGJM3TIU1OTXdNYSN-GL>xC}{0waU<9W9(NJ5eL zEiffVB@k*p-<9Qbrs;d(s#-1I<9V)B3V5U*KNWM69blec*DNM^<7pGn&UC>P1040P zG$ck{MRl}^rlHJyA*7|`e_=Ctzs`ab;V{G`f2gXW%#zej5XJk{XElJeu+Y4`=+QwL zCxt&=4hfi?3)Ej6P7B@SvCH-xx*zsssX_&p+l2|tD~{hD6J;3bLNaVug3;GpnOT6M zXH1$g(r**uTJWmx&p%#;4K3Jor$ETiU+T&sAZHXfZHFl~Sbt*IIJ*j0j<=`@J8Mib zO}gCpUwEg!B3LJ%KS*=&Ra?ke>=j{ZB%LyP{{U%G37x-fJQ=5TRj)WT=q)B16Iqd@ zDYsj)cE9o~nE*~wRDR$E^(O49s(zA%_qR?g%_@+VOS>ZQ?ZEdc) z#w_~evCo-V?}7mza*FSKD;_zF_Dxk6N z>C38vBXdn(55_9bSjf?&eU56Lc4Y|(dby62&*XGA!BhC9vc-*!XnSMep3(7(>H;A_ zyoWvV#krO(XGpoA;7~22!)2I7wav8Nrn`3x+`{YNRZ={DS)-bka%Y4w-Xw>TioVsh z`ElVg1y$Mb+iQO6v@BQM%T2W&1bU6tQtm}HV(u4zHwIo<7#c?h^R_Aoy1T0li{N*l zP9^)d2?q!QQH7x>yM3LZeFD0Ce#M*py|k^h#fOL=agA`6&daT?xfhklSXzczb%vQmym* zNis8Op?HZ6U z^cH$1;@`O8&I~2927GqKnxP^AH4s|4@!R_!n~KhNEZf@i353crP{}g0jE+%Nw=8SNO-(jn^_q*uM;RIJ1poGVmdxm;*U+NS)!8^Y{GsAJFiKSh|YRx(;O zSk?Yo_j@)!tmG^~Dbe{|Nkxnhz^#%wqyodhz!j#${`_Jw#eThuUg|W_0T%>?&O8GT zjh@e9Puyr|aE`5dZ5Nb*fg=}5IANjac^%c7^d_hE)!>U}qYw`#YkUA*Nk$w0yyLj> z15vD)fCde=iWyU1HukMsSkLB?$2!Jfa+2f+`;3XHIT;CEeHpz5%i55|HCp)!uPIZn zC@0LHq`U?mm)3C_XFMU@Ci;3e4W)qbMV!2(ns6tK>-?8J8>wJ3uOSoY8CR)!X_%-J zV&){l5&i=~AUH7pPsil%-MWD)=E*AIjSc_KhY&RHCm5~htn78h-TV)|bTMZI!z!n! zJ`Vi;UKsc_h@#*8EcP~JoDxYT69UkMUTT9K-Zr;52V$ji)w0uZIdtDaE6plg7q;GS zy6U#Dv;_9!vZNm}O5-BEv;WxA%A5!X%3Y=L2h)sLOw;)3{CIi6JFocogAE`*0zSfz z*$HEm7P0j*LZYLsOV@#}N*o@Kn(&N2sLnf%wTv88qvi4yjCDJFb zZJi0b^t@a0vrflQo%j8Rxx4*G6$GdXskii*Ls+kzmT{$dlu_k4By8IDW>3RnoJIR) zVMrNCk%X_)Hsfq8u)c-m@jUNBh8-HGlb=GnPVBKwyh)P=_7UopBQ-p5fmL)r1~lC+ zZeMgux_gXzU5!IP`kD8SuCoONZ}+E0t0#+6#sp-7j7FVYMZYz6ZOqNV0)=L9(tddo z@ywavZ$$7-=hRDqJue+xfWQe^S8bYA!4X*p^4r&mqd z0FY&UV{M$vSK#nKE15m7*(SGha>a?C-1ZO2?!hN>*sBdMok<_p zvKYA^OR)Ar3ViFo@qG^(lIRf7B`=+e^xATS*S8;K!rjEbdasUAt2MveYdIUCJfClQ z2q19#oDL@cb}vQh88kQdAijO`Amb>7sY(S0znif5v{>+YjH{(4I(S{5Un^S=RwyS=FLsUo&b z$!Wf2*07?Enum+OI`n}=?O^U9{-fQ2KZ)3NT5O=zmD9gs3V9V|e1oSCX|Tn`Lvy%FMs_zW@g1a@c}Nv6!nd|tz5c2 z_iYO4n>r$X*2YG9`XSJ<(goxeYl^%@#Ul`vHpFY6hg>l6o_7(d0yND*wBBRJ+OEgV zZ=&ODL4sp^YxEG`Tc?eZy5gfL?Ia5J)Xs0Q1q>S6pyJ7hgyu-2yq~X^?{JA-H%XoB zm-Y6JcB)M;O!GcsTF9j=c@Lqx{igrlz>r3InMjr7QU)=l_ma7R%uL=ZB3MJD5(7;a z>ID7RZ{S%ZuQOa9g4y*3R^ESQsXg#5J>z@wy1?=+%kQnN>Hq`e{8Dql;G3p7du&;O zh1GOK(1|YW@T}38{QA!0Cn7#OsY{|-0pQHZxyp-N(5yXX!>19e%LPj8JD6mc_nDKYcF66qZ*F?c*G z!^tq>1L{Tbs4LNw=y#?5u^Pdc`r}&Y@$4}Me9MCyQg_-%9$eu`>Mt=7LE!BZg!4kv3(|3ab%g~b}DJM_Ql78BvT|x-Nnc!^IY7Lr`k`nX?ru5X)Js*82ky1% zsYHPd2^>m2X@n&uHhYG#i~vUT_hmKxZXO;;%@F(Onj-XNU}Vj){2?wB zVRR7Pm~a-8E?z=sP1CtVba=75dp-*qy;-%%2JU-A&xN=)pCu zZ+anE?~*rp%f&uDp(N);92$2obFFIya=h_;xEuVUskon(`WR}`)a7NBU-+r?Vyx_x za%G5OTAer);_2#;>WMbZTD|XZP=D$Pkcy5>TEAq*eZ_fdHr3h2b5s26rV4j;#UPV{ z3$%1Q$?`&FqI%%B9$AMn@l%8>U*Va)n%umKyqQa2t_2Oz6W{$GUl*1OO3Sv7BR(+l zM!r1}EOtThx@%^P?ly4-pZritoe;XdeP`oYmYbUq(ReMWQeT|dwN3^+VzcmiTMFdK zEAu)Qnx*PoL%V8ud>}0pN}XWoPnbTTf^5~QqoAT?yd=p((E(;wG$dHwRN=SZP@3GC zlHCFGQh#klUA928Xp1}3Zqm}CxEuSg%R{=C9E9G7biaMJSX8IJ^6UEqpz8(OQ~rk- z8RThJ_9Kegkge0Sw4_Y?1~c{cGz}dysYf$n-klGcdiEa#;F5jcFRDpz&)caY_TW%* z)JsOWhW2Xkve!P_U#-+u|0fJjbaeQ7gj%;L-|FmoOfF5yz}x;Tb>FcfP&u$?pW;p&_N22dO?THl*LaOH8npk1y{hB_@%J*)bqZ__zAFUd(X|dieKX({dB#X zF3^!Q_4@q`hBBNjdKt}zqytj7{go*$268&&UJ_Yxr+&|1@nKPDyPp|M>l2ySkkWs@ z_Jr$k%6>MeSLEtJgNrWA_)og|mzKu(RBI@wd3>uYuT1KNB?#i1IS_ukvV@uBCHE z4*+wsyRD6F?CFY=F^itAVx@^W9ffd4{E_p+NXPu+K#+@EX-Q9{bUNG2{wJaX%GqWYZc)TFxIiEC zDUbMe`)NTs+V9JOs~xP3F|sI%A@;UYV_a^_9QAsirn>xEzdOTtc*AFv#_N-M)Nsg& zf>z#w9;2gafRhqxCu<5> zx5<3fj}_RT@z(OeS_Ao zFJ#FCEcd=}X{|JFk7!jbu*XL4{fg>KUF*Cv+s z2On~IS-GsKT25-|WuLqJ8Y#K00mu1*1?`uFWN;Zn_-77YLFAHvhZ%$5t5=1&VH0n( zTRiEKhma)D1O8UO;jg7q@H3a~q+V=pPJQyWT}*F$=WJ{&*}Chgbgs0MMe_rnGm0ef zapvoy9W^f=Q=ECFBr6#|sZ3-3ziN@K)KB4Ea6h$N;oC&Pgp}@x9f;soziPM*PW`tejzM=H=T^=L_tV0`h=0U5vShm!2=9@p;Jc+2U#wmygQKlcc}x?tdtC@gy%B%Kch>g=op|JJ&zb8%)A3X2R6$;3g$9h;+; zE7_g-tx5hhXm~SLL-*s21a*x5r_9Ev-@3aJ=FCy4$sczMdhNJ1*iWAQnV`QBm1=B^ z(}t!!%lU%ZTKVRFp8nrIx8^G7rO!S@wgVlb{@2T(oxAZ+i2N^zA+yI)F9H4Nr-aTL zj1;+fqf&=>=g-x78tNJh&OYYSJ+HkA&oK3v8M5#pJZtN1IG5iAWt>P5ze_)9p%p_6 zTANN5ws&i=WTR$2O02tAW}jIc7b}hPJvFg!#AA8BPTxpKxHEQ~mR-9hsxfQ#>YpF^ zTQk>pw#8%UTz3y>%OUXn+288v-`69|mpTno?CYs(@_*yT@9KH`Rr>xliG!F5_RS4U znMg>3Sg2KCd;gemg!0d_1wf3;Suk`L{cp7SLk58lxFoi3M|p7Ht|u8>$JM;(duWo? za}ZJqTp90^w}@|zxPkDK9K~ewMPLp2kxM6tMW@O?FhJSibEdP;d%_531zAN}A)%z_ z_k73AQr(KF*&lRo=G2PJd)op?xKz|s#57<5zMotz)u+B@az_%rO&4a-*?N4KH7kH4 z4oyu>J5g|$JSIE)sVqQ0uqeSOAtKA@*cexmn-8nU`NNN$kqXAkK*{tJ;vgt9;1gPm zyeH*Qk`QV_J|-R^-^TK2#>GHGbVEm?_F3cL5aa1nI(5XHt0j+Pgom`_ zXV|o}R7rzJm8hPn(+nS3p&vbljl+9^-fVX1_gk_JMRe)z-SS*H)oemx1SVNsVZp8` zo^8mpXf@25^&^7&S5h^>MJ&ziSp3UdZB|CzkyraF4OeMCH`FC7Hy`nOh}oUYjlqNy zMP>nVu^~-Mgo>ua3F~6p+Moj;XG545lWQp|?hCZHKj}csSow#d7tIlTLxpN3i55P* z*(~MW?JsUkH>8ueEq?CGCb*V3nd2X)ivR6^ z8_h_qIvyYRUM-!-%u9BJ!7!_Fb$pK0u|MO`0 znAw2SWUg8|?HgK#77bLKH}6Kgr$$%eBE8SO5N`ag3z)0&V*Bim-id@-58X=g%ZtjO zZ@2P!ldI8Rc8fK;V$DA)fxv6Z(sG$dyjW92>^5UKt2y_;QPg&u=6Nb%oHrx@9jInm zia%zayIzS_{0RIYJ*8Uk;0WCQ$^+e=?`?M9+&g@^?0N;zsr77FCTa#gZd|f3916Rq z8FvOiUT#SFem-ikNLmb}RuhVxJ{KtzulTH9F6`?XO}W0O=b3{;r#~8pcE`#ZLHmCl zn*aTPupWw&mFLe3^HjGN4tf2G)Tk+Vb~kh;C6_M!6)O##P0pdXQ7a94ipl&u<0}oK zrG%Wes=(zp$;?LxCJl>2e7Dv~+$JG7j22|VPYJl1o>|>XH6ZBpapKo+7y1yx|YOC7V zftE0v6C5npQfVAhY41K<3(8cNg-IFc<8kk#`xRfOo-{BkgihDixA4Gn*H}Cj?`O^9Xyj9ts2|K0m zdK%Cx}dpB&*fZfo|bxq6*+Tn{E%+B zyEOLkNZa7jhlBLVZ(f#m$vJhVuI(h@2mnQ=!{2g$)d=R=#v)}ix zf!-ghpa1s@S@(0xxFV~(+U5@Sf#O5E06*-_+;47bySi_Sn#_3!_@@=B&8G$=!{4r7 zNr+5YF#nhL{HW~Imw#H2H2N$gpU!yL23 zJdh%^?CprtIVl^d>Zl0yL`9={tMn zTuu_>1ytg+m8uk}_icZrcV^M8?x`k58Q7m8J)p1IrwA9fxe396wB#0P*F{7n7Pc^K zlsMKIfn6T%%g35oR15eQa2waLl9v-T`g@A$Cdy2nmni$aa2eXoGvnq$sfTLd@$TGt zt)~E+BDZebca*FaE8**eF1MzCz*Sbd$Kynq4i_uaoj6eNC4B-Oi{1j7%Q0I`LY0|E zl|f>cjeY7bkBn@O<1W4IbioRy%b6~SqQg4T@ZJp`c(6%wMHp|y#}S87@seD2{1b8- zC?C+Bxa6G1biG2X^TE2uSE2vmxAXJ=zUus+-fPvv;x6_E@)cY>rz>HK9dV;~D~M5# zGtE2yG%ur`8?q^KqPs|=kedA7we(5Fw`viqx%?;`v;cJu@39y<&Nc9{bpq;I7w2X`Fp?5tx*JBpdv_#cXP_!A5YODLa5%5SwT)l4<2_TjzZy7?8AxoJcDaw|Jo+_G zZ6RRD<6_6K#IeA#xwa$&bph#@V~NVIEwY%BX-gD3A0G1`Lcq33Fo*n8Yqd>5Sbi3( zYHxTA|NTa^SL@wTMPA)=66jo z=k{*%@1c5r;mk}8CKFk`L1MVcz~L`Afxvig2&AjSZgEqZ^R*ZNm3f zU;6ZuhSH(=y$1-DB}!V6VrFPa8+4NQmY-oUzkuoPT3{Xmi$t|)^Ky;&KTVbY%X`Ke z>AJxoQ{c_C5{Ym$IwsjpNc3iQHO6LkOhI3N+`99WRfFu%W`Atx!8oHvg{E4;dW~I; zqx_e*DsR1Zf-o;v&C)twrxdl&llok(M_xwlXZ3CH%yvbPemQPP|6;wPyRXEtI}OO$ zPS148yr8GL+tc}SgV8__=&nrxyUZfu$$D3P(N}TaSgiM(saE%et^x~cbQ&r$kipi|%>An-A zrGo6#>If&wNOxJ6)xcq9T@=jL%|ERi}dXU6Jt7j=uM)8rbh) z_H8MkZb;RahX}V!J11naOlLY$<-$R2?!kK^-@iz?^#;n)66QbdTuun?+k36sc|^=? z&i^~(!_Xx+Ejd0ISAh2aULEOLh2R6GUvI8#Obh_@XkacQXw_?r#HKF09}DQT7>XJ5 zs4-~N(hPmdfbFuG!QR)>&@bp_iX-!7I&MoZ&(NZbGoQfNRn3hF#|0q79)F z@J_G@aVeZg$i<14Lf2bz0?!1{)` zH{Ude&8xcUd7l*?M@wV0_(4Yn081>_BBwYm0bOWutwx>QYw_UN^HiR1$B}A3Pp0yWq3ASZ^G?oGV*7mZ4w0 zF1A%{+;&w+^U+$gg;ACZp&~vC$Xu>kf^Sc9TJRddqNXKZ_f_^6pk6P%WLgw$){7Go z2HTwe<4Pw?wv7Ddoi^hxC-DY^?|SjR<^Ah_Q!6ror(UVDQz*Fb*bX$&)1;YYjfXoO z{|zEU`3Nw^*kT+e)q&c>AZGFSVxjSWay8#O3Z*J}n1g zsuE?c_L>dAlT`b3JdxoiaI@eTRX7F3W9)&a#RxMf<dx zUIXGp_Gg>H&POvJTz9XaCUwLjY4H4QeQ}@aC5cswc0=F!*@(UJ_(Ss=eNdUhsu>5d z#&a-E$WSk*eczj;0}_$(&X=+DYAAYhib9Xp(W2wS0??0EYi$oA`!l5}=+}`1Y+aFK zrz+(Q4(4w~^k3ufepyFY$V;{p-IysNxXjBiDnotc1nJ>07qaAA2*4Dv+7f**0~@$0 z;oZ4TYitj5Ys3QNS0BIlT>tta_g>L8-?1VaOtvzhqFCd8)D~(tZa*ewXwf|UZEEWI zk+Y$2p^ey18{!}~#7e}pUul`l6=Jw9sq%zxW_n<2(Ar%C=sA(gL&>Bg{|jns(S&z5 zKzA2?#FIL{t&fr+xUMRRN*$q)ak=2z1pXu z)Hdsv&b^KQbK3cS6_lg4ij`HXlCXkjzHR?zzGV8yF$~J15w@O{&c*Xgn}?rpwF0aSAz!XUVtaLwrSU)7E#xeYIR5K}E6`c+RTx-h5rs1rqsX@~Nic+kS+xAwcO63X1 z;f+B!*6Jsm3!VW^8UflFfSQ0%yg=H#d)*8zwQMx)iqDFK1}ftDH!JGGI`Uqsv%LAAi08Rdrdc&&kJIUbJk8%Zm7YqV;MI;R z@Q)&O`=K1K(|kNJ+%kZil;@*DEtdU88gg-f-D=%mW_qHcN>MWM2>D5&`(>PqYA-d$ zIeSfm!A}hBKh`(`JTcI4Ex~?MnQ(^Sh}(O;jN3H8iEg)+Jk&`DwzJ6c`9W=RVM)MU zoY+H+QE2qA<=e>o6_G~@%j$}oE1y$p>P(S_DuzzRfu3!H+LxE96stp?RNTVanrXPcY2Cu9%l}zPQ#}s_QSzY7M9nVTnIO6!ldcv zBC@3Q-yTgnZHokhY?iPE6))&pWTl)q1ma%9`r=-b81QjE*n@t|Y(feCSx0(U5ON4O zWg6)M`r>F^Q&oe4|4BoI9sVpZeYi!!C;V?Q{=0rq{aITkOJ@96+*rwS|ARRQLonZI zRaXJp!yJGW1^*46*eQmDSzpO&I$uySM4rFIFZGD;Hd9ZaNp zjN*!?CE^7I6pAralawlzKzM4_x- zNO74V>{OyeeVNk_LaH*xFGTUdyvhx?kyv~PyaEDtl4q;ksscc>>E8GGP%h25SM8HC z8f9A81q3AL$lqb>qxm9!D<-9{j2_?L;}%}0&7!mek(8ECB#;vZlSuRO93!^7Za}0q zGc<^2eqSH-V(omyoXM3cbM4$Kzq&WM6Vd zsmqbxq?gHK!~Ex#3X?Smo2KQ_$L2?QKK&^1Z>k~v?gtz^jN{;=Nd_VRdzZa2lmei7 zOhXmmh(}5fEw8TOdFvLTE|ofrOB=+~?Kz{m!$$Z@>Fo-_QMzD_6)` zYpyx(IqosWea{;8vg|_lv#TwbjIxlX0B2&=wL~e%t7LN@syq4wuWfkKg8$Y^3mTdm z%au1Ju3M%VSb>b@JAsa5jXf)#ptr_JTrhXJpy1Mo0wb9B9P*`J!yeTO?|vz4^bM+f z*fm0;`xWc+s(Cs+Ggi4rvBJ#_d+F>mtkszjrobuN0NVlU=z8`$Q8Mo}N?1p74@Nt1 zaYX_*&xuH)9X#x+SlN~!DJ(XfhUS9RM1ih*fB*JpPu4@N_JXMM)|E1-JF|_Zg9^_S zRjX?Y>adpZAU4KKp>4eyaHavZ5I!Yj^v^$ z@|X|oYD+6mNo0zEf{J|m2Rk&9S!Wcx)8gwIgIAa84mkPG%`aO$`kkKM^X0|`xj+0; z-yTc(4UL}dhR0}U^-B%DANQBgU;BMj?`wwdX3Z#v2w;jnMVG7z?SuO;&y+p zc(NO#gk$p!^xlGvA$Ci5T^2W&H--`{?_HF<1g+E+b?W;ROx{7cvM4g<&p-&~^Ur#^ z)H_9Ah^_5Tg9TA$kImQ(6HlR1Zy@PjE>WltgSkdJdT`0fvAcIhij55PVLt@ajALzL zmJ36X+V3wNQC@H85LGgbp6}26))@Gh7$7QrQT? zBF9JyG$SCra)L4zm@?wgoIh9pcBv6)%yKI=_X2t~%}i)T(B+|SY^;j!Esve)HTe?d zy&I6-@UAD^-cbI;eD3~EAILtqdYiVSNn!BZV*nJpPz!CGsa4l1T(^x$b34g*e;*;e zQK3E>C0#kmGpPN%oO@Jr^g?n)k&c0QR$rR#ugJEGYer3N&dX4u!1m4cfvP8F$@sV0 zAM^aWd42`TE(dY=EA1?>j2WrJ85Ra#YeO2$>4JW(%`;luk;f(M+fkLJ$NNcUQKv+E zEGl>%CDc;1?>zx`fX#0ZD$Pgh?W09V35U+!pX>anLUIHJ6-k+wU79>vIK1U6EG9xh zh4B<h+5cSIYf-|kYmZcdSo zBqo@-enhfLct|l{#*TOKNQZTT*~(QfE{@4>qIsa8QLyIjHDagG@=Sy%gm#!g1~K%_ z)ci9F(yq72KT38sJ1+evwbkkO7&mbrQ!I3UmT*O?@_zZKu$`c4=TVWAf1FJ}zDuY} zGq8@5V|V}+uS|qp2uekrVE<#aj;jml&nj;&zDH@`hP5^^FFOr{GC}`XnYLQ$!g#sd zk&#nI&!0!5%UU}!oWt7`B;8Z5v>nS=4=Vr%88pDtryZBf!KES!j<)ZzA1xo}O&tW7 zKHewbZW|JIeIc`l4LzhH>w439xhLz5N$4jx+q+46hQ+2DjOoA-qJhKrqID5@W9K46 zq56h6k`1I%-f+%zg1Iu?+UQ%nwBclJ&njw*Bgwey&--)>U5A{I*}sEU48m2o+%R?o zH3)SAS@z0SUY>;i;Q5%V;5taJh$ld>_%a`eYs#1U0m22^UdOD6lJTm=ASbCKIdsxb z$0K; zy#}}G45Y5Ufo9w}T1x3CFOYtp>iE!s`~>QYWqVF-KZKbj3Iy)~aA%En6`_tdZ9%)v zRtGc-$OcNhsgf^JY2g*u+s%^19F^qrE2B&8hqJ3oT~@cNRsY1>Oue#AuKQ=^U>95> zMKb;wl&gFE7ejC-=Msv96l158vd!m|2kuUIg?+)Nt`zG)TT|?<;lIB) zci{aM(_vIXo{L;tf4Qm9DPt;cbiQHttaazy6_e^exI5y!bh3mzPFjX_FtXy&bzT~| zT%L!OE$2{b97R|2uQ~SX^G4}GTg?mO5%s9B)7cVlQSGS;ox>HbqK`Hj(#-o8h^XGj zq9w~2{-yR6K|k&j_J*0W5NmHzkR7+O^|8aRqD80xZ7I)PCs^JpU4wtV6QpKx)>cZZ z-#9covL_RISlNVcKzPlh$UVK^BYnou>2=-#9Ly8?r0#jkvcAWJ++JJlM)AQ8DvV)XJzVU@~vH4~9_)l#zhKwCP zXT#y_6xgD>@Hj-yn|o2gZOl0;g(?8#8A%9xsPePxap9?tx(mIZBnTGVWRQEKk3S6hS(m+8m zly;+qeigv3l9CqDCmsypjLLae-FUotATU`I>RM6$sKP zvugk10upw(Vv+bG^4bNNw}17ocY%KGQq6QOZoXqBlVRGXL)Jq1Pa;kkWk367u5ZN9 zs9WVzB3rK-brwZLAu6iT*+Z?-4_!zx1t!pA?0-CfmD+Lr^vMt&p^|#;DC6$>#(ckt zx3tO19wV&rElIPQkT{vgEfVkZt>Yd-Dz7(|a86!$^r8Pt^~zUGXx*df_>J=Tg}T*O z(%(^zUA!ce&IeT$>(AQb%SpE>sFHl`W=BDBof4|_XEQRlkMmY3e*&yInUA=uX7b!ebB-Is4bqDOc-G zDL!-=JSsfNpETy3lja%Oh*3yPO}IlXz5N3e#LOcf)Z8J6Day8L1aD1C8g0J7D#8Kg zTD~3ZRrr&C9PadLvYA^^lZd{QNbZ5!{*@aiX*R%8UY(4}j{MCp4rW?_6@3H*W2>1S-$vVVu7s0K|`c2?-{h_}Z#vj7Fr(-H`%F7*Bjs;3ga#46w z=xA!}(UIyh9^UXKqn87yV@5j1ln2+&6qDXLZ3>nKeY zHOLNChK7jWx@uV7SE{ShA1N$(+$1SvW)<4#p|WPKVZz-fCh^<7pKKHz-t6FHbputmLZf~%UN?i8RNVP-i$pgAhsZSs) zjN6n`NbAnjftm+`D(wjc)#xi2i7kzH4S;SHl7Jvi5pAf4A)+eJuvgW=6-aL(tWb%7 zMBIY)esJ;N)yE5c_}ap&h4aG`HlQ^Kw^xdI*wps!n-yl2xmT+@o^p4T*G_a*OZ0%q zwSh&=W#PPTFps#F2A#MTpE0R};EMu!u?%y{PL{CwINk8DTfpMYS}nj7u$T+yXFUFM z6_7**uZ|e%K!r1BKhmc+2;Or&6D=oG7iy0q)Eq5Dkzos~25l{?GHO&|Qpgv)#3zAg z7XXJZk)RQ=p~XoVos{+_{g8MgStPctaQ-p!+kZn(48HK`(VG&@))SYV-aF2kpS)=I zVPm?hI>sBr(^-Br%*iGhrJvwi>95;2-=AlbP`xW(LPzqmSlL_JKzM<9*4iqR2o^$n zq5D0B24!$dw?SVI6#DOat+P)W1 zMq^bK9lKEhVu$v%&V070eZ+1}X!CP=1ihRw1?#_%kSJ{dSVp9_xnLiE994a#NW&)H zrf30#&{64(O1>2krQ!~N@5}&oA&8q{5Xjh=SRl^L)u;}GU(#t@bZqX~krGeLGga!s z97daYfs#g-tZ(9TZZ-O*U=Ahsr^i%+P z=Ck9R$f%|?(>+&G@cBOznE&c8_i^4!cRpWvb6vCbcv#e7af8S94~8z-CDxmKqsN8+ z>IE=vyZVFjoTs{rM&I(^Q|^MjZk2~NDu`_2MboGi{m{6g0II3M`jWOQJ8$1$2(_;z zYpIZ{q#xR>lp?4lR2uqM!EI{s#H1DU13$O`9?iFau_TNgtCyy5eQlB6)$HuQ<&V{M z{!rk>LP=^L+mK$kJ+2tDJj~=5Sgd7RFO! z4^gs(k;`+k;7tB@-A8aq#5D50CVYnTAZo>kIY8zMF7`M|d zdj}@xBeuJ_uQLNP!=;-KSnhpS8TfVhS1kM${r)op%m4oDOb++^-G&>6!CopoEh0i5 zVK3tPN7IH?k0_JVz6>R7Oa67mFBMstBz0$3FdO`0+@B_D0>5Oq3o2Y8o z@nAD(+P9%Pxwto<4R)(kz3rCz0#TDG5`Td?9I#tugy;Tm#^}Gk_z%5cD1qaOJ0?qB zC1#;!hKPpT9pH(YqP9P%oh@fWC_^_NBntgw*fyZM;a?0kM|irR>t0~>L7tZUit!3X z?C`abkX_hUOsl-B*jJ{7uN>>T2;;x=>)*uW|LdSP$A6(q>tnLhQ=jMlRSm*A>mRi3RP~fu&>5^)wM}xaQ`h zW4C_MJpabargIxQm+04ps5)cDw}Sijr-JTlCBtzmW ze>vR6Qx~EcxtYRl=l%v;Jdh4L5!h+*-+0I8c89gc$w@7~;&D`J2Ct4l4R+LptO);m zdNzl1t;SzjHUsAbtOr&3@C;m>3wQ&+iI`KUTkW zHtaj!@k@tK(xO;zl1Y!2lofnThVt6ScynN_d4jTk5B3A#5e#DM(=C2xp_nrPOXhbt?rd2@zE?Zk zoZo=vzY47{)GPKLy9DXg@7#9RYaR)2k6Z69Q(*@=>^v|5F4$`5efQU$J`wEV=?w{R z7qWGud%g|0JsaTE+kN|NeX(XWY2=r&0>koc=pjgMZ%DO6{(? z)&IHrKL_OB|LcE7{(olO|HYF1pB48%vF`Uq`#;g<-x-SkpCauBhku*RATu~AwgUe? z;{UxH6T`WAsSTu5bTZy)P+H{XGyV91Zmf?J8Zhg}&)n)#;;<*eso{WvTH2}HD_y%e z36Hm#`W7wk!t_p7bV-sjDfv1-P~fDKEL;4F|CBLcU)dO9TQ~e)+{#~HG%&-m)7euW z`A8Kmr8co@svYmYX40w`q!;>29VQPwia466YV!&j&?tm3Prn`JJf2`_+MS`9$^5|( zHUnr*r0UjZekOCFD%tdt`fl9P_9{HrqVs3csgB}W9nrkv=!1gj`5b5dcWO!YpZUv) zYpY^%tH-Wwag;Gu8u9MpRtn)exIwk%VS;(+{fhB(nvurK$52MXzU*hn{XyM@X>7`L z0{hQS)g2mpC^iJ|MuWX-%VV^X<6m-cJ@N-T#9NJe{;EJDi!JL@5 z%Ld)#hSjZBug(l9!C6e+ z9wiv<^$EZ$DZLWprgn?^-eB?70)|9Fk5UGK-h;kol?Zq}`;u|dr`nBH$%>J%GPEJ) zW057|V-YYzaMK$kmx&BbRefZplA7+~sFH*^!LeT=uv!nqDnTMy6HOp0h{H@!S^KhQ z%L4|OKmPq&{WUA!@*cad=p&=Iy&Wa9$OReNM(^J9D1B8|REah%W#!|4wx}?y!%lXE z{Ful8Ds)9m}RlGCXUsiRIu<^Yc7jlM0F%y#*&t{`b*XR_c8J=F4kLh zJjxGmJ+18q>wBX$tSM}66@5y+EE5nSVwHe$l8p-xvl8#TcCUU?|9pRe6Hw$Djbrl( zOTXTmm@Mn+Jl&Si*+plNk(CeLiIYOO`J8)a+Kq5VQclEA!Qfo|vvlk+O9%8izLz2s)RNvbEzRwp^YxT{g%W zR)nM@FlP6<5U1o?d4-1I*z)J2!;!)=H1TK=wXq84A?rA3n7%ty)Jn4RcNxOwG&dy{RlE??y?MO@V>XC6*%5a@P<;6FgLlqt0|ghS z;-yTDBaU7;M-oCDb+&1Fd!Yc_7LVuw>Mp2(kVYuw{;GZZVwY}8eJSxF}{0e`yFKz?!P$l9G7)wy;JulUFG+wu)^V!(xf72yAqdisKpx;n_dm!o9p zlzaT(0&Ocrs{u=^d^f(1h_37STeMK$7aSy{4EeHUEG!GrH`CcY3@;awGVK!!SbGfm zV)SAsjjKKsuf!1nFX7U%*rrKC2uDOd)!heyawMu?}aSz4^q`rBwb zZI(f4a?*W4tW?UNxlddQ1JE62qTw04)?F~Ds^RrPi{!I@sZauOox4!T9@2XYHC}pq zC?ct2Az|d&>%o$#{OQZ%qQ)va&F5A75DTKP#E=OMmKw}LgtN)>?M2yIHP@OQW))21gp%vWQA#{o-gaC)uqma` zt#&m=VA{zXSS|3SptZzn8*c+asM_tGhpJcUmS49hQ5<%sMZlJw+B8_<4?i)Uh{yj~ zg3OA0fu|>{+QdsfyEY(Xm*`mTH0Vhdt|%FhGON0MuF*5d5^R|1K2d)QKSJd0uq!`?O0zDT;CLL0Mt)rH)!#a6< zHG{7Xg#rX1#t6Jp=<<_4P_YuSWNQ70#dD)_JrGdpB zy%%w#VI`rFtT? zT;AcjMxIdoj^YVX*oj9Fx%H`7C25qNFFWHna2M&g#52$6$jzrqvU=gX>10Yh8Q5Ib zc6+ACIHK-z^n{lNH=9N^_w%Mm-XFz7an035{a{tX6bAQ5W9P2UQpK&<9#?3A znL1|ponueDD#C16$o68tX%btdHwFEd8z1wyE@1B|L*~h=~&f6i_ z54{HdMRS9MSO$q zkhb2*szP;}Z4t z5D@NjEy-o@H289u6j0A>m{R-v-Ngv|mi1+n^Vn9`9qL6`ew46`>&n)0PTJxT2;BVD z;HY1wPU`hWnx{~=-r^|9Ef$}DTHg9)p+3e7IE5)nWU}0j_AzK-6h^%T!{)TbodEhSfyW(J1nfqHu2p_TH?vY)s7V02y~xX zSs0)25Yi$*@yV5QiK(sb4X>ho`^zhdf4U+6oX30iJF)_ruX6mBIZWaWJ+&F$*QW*h z#g$}ytBY8v$&ld=sEI&FXG-N4U>|*ox}PejL(XyZZ%Zir)70D@*5X}p%~6AAGs>Ul zSG!|mX(_iekX+dQ_MY!ddr#!;l)(Z^dS^HQ#iNdfVV92WmI1ILeaF0;kG=8koz(sm z2Zs-;K{T#e;WSuke*siUDazX7Z@`eqvv&pDxVk+r2mf}EwKNs`P8HAEVO1p~7upq) zxe9qVRoUNf;A8Q35i4^!Y~d}`E<^;{RE#7R^y7$GxnrCFN^Kq3h4lRnD?-Tc^b}h# z_(@*1GHWLlp=ZU8+-l3dIKZ6Y!W^kx(+1eA1mi#Bi!?&hE zU>Si57po8y1@!|XW8bg)`oMmNL75voVY3_r8E)jI_rg?TrwzKaJ+j(aNwM&&gM5HI z3vjCdyip%+b9htGE%ZMd1Y$p0{FzVeJXWR0CO?a8wTb069_%v%W*o;GDUS_obSGUJ85*58uNuR2swvel_eCU`*Zm zWw`x+s3R=w7c+X0l|Xgt4C)@n2p@C{H-VL0%o7)@P;$d{&-ZO*Ry8cWA(P|=$*I%O zu@XCUO27=asyBwt#qsVqxiJ^+Hbm?`zesw@rcMi|`NGtkY4q_t6uGl@iF~E-eD^^i z$@i6o1SXMLvz$OvGafg;HLweY$_S|jF|5`WZsw#JIJE*Hhl?`kg-Edt` zlEV{FP#%}A@w%Nn6eP|6+sV4w+;oXt4y#4HAc;=^f;N?NG_@e;uX}RGXkmE|goYn1 zMnnDZ>cO)Me}t$9gz*XmWznsYz3OF*jY^)5VtuIVs_+rvA$UFqVz~vncO2UaG@IN^ zN4~#e(gXQG_yYF3f9$1h#df}~K^8rP7M@Hzjl0BU26XGhDw+~#(iP18ewfRihch2^_d7%8=C4} zxeC@?0igKTH6r`T?hWGVA0vwyh8e92WEX$sFpTQn+cItCyFnr#` zJV;gE=*&4E7XU&miNScfO3&8<&KN78hOfdx1xHGgsp`3hV1(jW6JEZpE#g)5Q!uiM zY|pKB@%4p%AF}0uju)-=Mlu&id|eem&wpM1GyVncJ$9^fF5B+OmU0f*MC|{Z*jo<) z34^yO^F80;1dpfvwgn+*$JE@*?=g`*Z2WeQh2OF!sl*3O59sC=u9BIB@gN#jV|@Ed zLkDhY=o|M6U3Z-enLd9nrp$kJhP#yb_QFHw<;lh!pYRQ;l=)?qp68wK+}G!U!%f+y z;O|1pmU!zhft^aXLFtkQ$?`5GZqPq13kH{)w74;)3&JK3c=ZgS=HM@dp8XV`(}r7! zzS(Zt0D3`(l&ifGYW`(w@kVEvQz2e)uK4~_p9i>0UqlPTWSsTEYkpF>Z!yo!Dx4!k zlnc$PJzqur5n{*PKj%}Yo?%pPQs65HNXZ{_AHG^P8x6GU7^&Xa;Q`&vF7VjRxLcr` zq8M>;Ygt2RRX~l|oTfHBxP*@tyYY?uTsxNWSjTKF_JK2Be4<{A}*hqJ$` z?YNG{sjCuKK9Anw++xJ>Kt+%6_$&pwu& zn1`POME1PR?&IJ<4vm%O-18$3zofZfB$W@xRnP((tS1`9gH(Nn55i!LmVl`^&ZsSH zS`OKJG5Y96tisc^#9~@tmp`o|ho8L-fL2nh%FHFpPxk%uavKe?na(r$w+22TR3OSz z`ff+pViiWPKFR!B_Eqg4Zglcut#9g_k}J|p8BRtp?hkk)9wlBK%6^pWn)NWQ`abxR z$v)@3JG3L>u2hP@&vK()=Dp;KTIm;VC6c|O)B zq^u}wu<9M_QNnLl-Sx<_9DXmt^Q_^JGu1OE zXAqTb`FUwG>3@foJC}h1s<0{4u6h1JJG%Tw1@pjyXy4!;VNt%wH{>c08(bIoM(m*B z5vY!3h2=zjYs*hFDBE`Qkjyj5jqinnsJpd=@?duXlr1_W@TQ-t^Vj_(JRCd zI;|a-Z_pI;W?A`nUBtiAR#r$sVu^p0sLxjr?B-(Ma%??l0$%X=Jo8Yv$>l5Eg&cJb zJg@#UW5mbFulQst$h`Db-Z>@?uI33lm_oTdjx(=&GYYFa&2BRW2(-YUOG=NIwt=*~ z6rgcPe4VOCgp`q+%i{3;8_}vV_WM&8Tr*I!m~p2nIp2xJK}-L$$`TD;a5Z@QdL>Rt z@=*I#9gPJRt@?cLlJnIm76>tkzq9id<2w<=Js5i2nI)W7W22|Gij@A!!CbxsDvK0q z#eLZc9mr&L*X8C;jHfqH2T1hw;IXB>VFBgv;SOR20Jd4I20fyaTaBPn%Vvo)>uTh;EUI|o^T=7E?Vs}Me zWR+>#ajAUW@4}F}oty=V@W*K6u`bomE@qsY^bs#=aJX1Wi$OLKN3W$wk_RzhNs)(S zH3#-0z?+68Sggc7s}H>NP?Vgllgyl7o_Ic|>IP7cbM>)vy*W5g7(QliXXrRCo`iPj z{2ceV2dc38x2O7PPXEht?;InZ2GMwx?12h{D z3(UM+7O)kku+>$KN(TAd6&BWarAzW7xBAcEeng6xZ^E@kbh=@km17=-0x4)-Tu2pF z;5jw&`IQRbRM2`=RtABuuCr)2TlK!_Sw&j?)a#&y6oJLwPUg_boD^FrbD_oQHV6@z z@Ajw{hz1>G@sT^MYBCqPUc)~g_Cy*pmeT2 z$Kr`q5x43#{H($k$P@a|)QN?lXIjbgTx`ojo5L5e+5?OcMv)5VyISUe5M6VwCu@X} z-?T;D`LT-d2&G6&W;}PKeDpE3y=JmrV2~5fUYDsPDnm|kQt%+ABTc~qnPr(O;Un&p z*RfDJ*FgYAfwpTn%S+2Zh+QnLETtEgD}!A#Wfjq#j56*n+oHc;)e-hG4`^JcF+Rol zj-_>Y_Af3lC72Cb2nxb1ai*aeRKs$7<#l<~=Ht^n8?U=m$Hrp_8(Kp10%~QUYHr8w z4K5F+@JJJw7g>v$m2CeLSG0sB*HpJoo7LFn^`p(2yvJmHz#Tpq{u8oaRX1mESRS!p z9hapJK6il5@*VW?jB5q=6U?hiJg%l^wz<$Tb^0 zzkQ8Z);$XM9k;T)yz|;ix}?q$?x={-c<8|D`JlD%kdc^MzikH@8z=W-{U)ktqxl-4 zO!=`C2s6;INPR{V7Q)uCOVsXP@oQ_*j`K#PHVN9E{y(7`D;%(G)Ow#*V{dzjthUKy zYx85lW65p=mr<;5Io4<1mKbu~&KJVhoshusUv7NlmYK`_5)B191j(0t*TmEZ0G#r< zLN8_n{?G62*^V>3scyq%Eu*xxKqai1f`KLj$Q0JmU^ylcAQOnwb`+oJQ*w(lB5vMb zCCqZv$%P7~lyQGoB?kxK*mZgWn??G`u*zfuj#b~42ys(%ycB%A#)IGRc=_1;e0pI$E00tb23XJXs$8N`+C6qyniqzx~6&Tp##$U9xJ8$dwxS=h60aHDKWhl1aL$E>%uaLiUN8NCuRA~?LeisnCJ~B{9 zwz6w|ACsh62b}7jU-`>88}n_8fR?yX=f%Cat{4|TU$fGKe}Cn1{l1{B zbOB?YHu5ZjFzYipq#A@#=b$-o>BRRv?f`%@$_$V*uwyaT=b8Q(Z%X&uthU7}iyTsC zCbB$`j+8~s`(T@H4(*F(`oE6$09OhrA~~a~+3J2>v}%_801b71 zBn!&;5W=jx-aCZ%o_WAXyz~PBYughf^sTqHO6Ir z#f3BZF4X`BKAP|}oF&uulOgF9)-Ep4o^D0F&yKrfF>*@5@%>z8FnaQInFM%e^KIb9 z$6Ix?SNQp=Wsdg);?Y}gJXpFcIa#M41MW9>nN=WfPAnZczfCoX5|XrU5`&k4gm=Vo zhzYyKK#co}Lc^=Arq>^CO!-8d8vF_#91t#d{8_n`6mR)4Q{GXAa6;Cv$TG8o1RD$L z#Z=EraK!SaB^r2K%|;tf-5QEf+At+f{uJ z<+$onX`bNzZp`znX?cQWfTDyq#sj5{pry`_y5ZTW_v~2guJeOL(+Ogc0 zddthVPjMfD9bFiSk?5J|^6z?(?gRF>thJAZ4q2Ou|&F8yO&?W9r#ToYlEtqpYJiT5y9t)&byA4_{`(t#@e@_BOgSmjotpi z;`llk&YGVWszPkcHH|=B`G}wD$agzfwrx=0;i=dWiY20(F3BgH z`+#wRr!7W7xy*n^M8W>$-0j!yv&yr9-%@c$UW+JPJ24$jcr8i~*GZCn{jwwUkmzt_ zxkskf@D~}iNG54{A=Uqwm;l#goifDIKSF1F9S4k4cI`q;T}+A+wS(mLnm5HMsbdIT zSY@oo=HAb{$|IFRedv%1Ag20ltiER4N*aryCUshIP}N&48=c~-O}B!TsC-m%voLLI z)}B>1*1#_C&N%$D0mm-}YRkYXO(0crQyE4ZX||b0jau4Lgk8DKK!~-+mJh14(Fpj= zTn;IE5K1BJHE5h39P>NhknKB9yyi*jrX#nnHLNY|2Rp^M@LDJUd~oc}PB>`t6)Dtz zERX{?klO&d=#&YRDSD|@w zzkDeSA7%^eoG)SUW}pH*P%eWQGJN-ubr~9O{#PfI;#jVKc$EP7d}o2Y+YT40=xL^m zu15%;3Q%DbFpv3MtOL#!%7ughMhstJod~K|3}33`BOJi_GDKNeVv7l)0M)f@n)wQc zNuU zUw<^f6uv@)tA6Kq30IVN>LD^ft8F(e3(WheKsqb+wIW;R)YfDA8oP_R4w~IFt&zsX z&1;(*Q}`U{bFu;rSidPhdrGb^>`?ikF{t6iplkOONeLawb7JoRZJNb?2D>i;YZ}?q z^f;OVd%Wv1zFGT44e5v$aSx141qGz`$H)2D`3$Z7oocKRY~Cz*KIRf-xvC2(IFmd4 zsmr%ZR@Gcoc}y6uLEe^KWQ_&NKb@q>oOG+6yU5Y>`YxaE-8S-APz`iyTEe=;0_B+__#yu^{&!TY;b;D~JWc|A)`g zk1b(~{i)wXUm+jq_cH+O^p%=0bRR5-oT;V+-P|P6@ukax7dK7HVwX*IHyh1duj*O` z_&yu#z@Jndfp;sB_Z>QA+r%KUFTlIRPhdQRYz*Vno+%DY#0z#zOjDye7XZTpZ#KOJD znE0J+rSo+jU<<@VwRp)fMWb~Jkh8M>+VB%Tp_1>_bnqSuQ)|~YaBxb_day$c0s514 zR^+_+qY>!AObs>`ukBN#Ip_LX&e5>#F#Zrq#cialMfcWJjbAC#gUNq~m16~KIu!oL zVnXZ7p9Rd+Y+MLxj`7Z21ebHpCv9-7grT~5Q~C0Wa_&9Qp;;r8ah)q+J>cg?zi0r7 z-d>eqn&Ix40{$H8DE(MCj(lD5Zsg3}&{|&-+f8mH9>V=RY3pyk^yP z_+5HPJ@nARF%xZz&3i_v<1IuF9P(W*y(7Vvpab|7zOcn)fxdDn<7h*1WAj(tawtgj z2>XU_#+a6(Ue2QwTD(dlyW!=g z!ji!l!|pVmq{19Ci|N+5`cXFb>D3vjB?>vZa%b7h$WWYN-J9Kxcs~ZK-FWoqC(suk zwK=-CQ$)`BFiz>spwE0?q%t{Kub!~rqqK6LF#nXsIQDnV>NJsSOWF@n%jdwOL0wAU zY7*2-6w_dlbJ9$=_V&LCRlOx=zIeMAbvo4q4MoR`p5}FCbyf1kpf4w$J|vpPp{TLC z(9GO(%mp~}E*M@t2tHi3@k+!w20eoq0zKPnrx85d=C(<)F(m$aO zzI8M}f)zkl-9OcCw@MnvfSCZpLdRRVoGzdJ=pJJ_uDY3F6Mjh5PxTA;rii@r zGw5hS0=DZmxLN2vlKH`5BvxtFE2y(<4Cpdb4Z<5Sqi&Ri*u>uYHcIni#LC^m-#-5L zy%#V*d{I5KG^LkGt$$b;vA4c0LB_$2u#C)J@;5S*Q;5BtFKGTk&LZ$fT_81`hZghb zNmcpk5nNQzdugb|b=91-E55s*l}lWk7$B~{_kroYcG2j64)!q$qkJDOnQptW@OG*< zRD0SWr@h&GsF(y}mqUyJW77yM+ii%Uwqu*w^m;v3m{S<5#SlGGt$C12Dlg9AtpC|x2vN9RDDAn)MC&aM z30qWjj&L7pi*v5n z$7ROgT9dZO*w*)#sSGvoofd7Ol|kt86`*!_S1pFcZsX}LJ*?c>iuJT68}Iu+)8hS$ zqQODJ%48*6N_qKf@3Sr{*PRluR^?)js1A&JwM0@b7ei;zlHSC_xwHspSngOuEWDyw zCGN#46jV6bpGB>+1tg?0mAQ}&VBM$S!lpa}m)>Y?C%~r7?F7Nyhmufk(4VJEwgD%$r6oSXzDVW_;Q4IZ`(cAN`064q~H zrhBtV+RHK!4rJq4?MpaxliC7mZQ*g2wvgF$2NXR2h5}e+to4-F`)JUIo+G6t^Z@#- z^{!DLb-IQe)Li3tmz94_D}Zhrn}F`HIN4Sh^|GTaPSS~9%hn!=Y|J|$>SHTxxjP^P z?I%{4msBdI!N;TRw0K~7U+;Ah+UvafkNL=V-}u=-OiF=%arO@rq}#3nW`uhP^Pj{# z-K@|yQB3JJc{iOp@YQ8Xh7eP#U2t9XCN%sI40bhDu*>#@G!)B_u{t1sTc`t9 zEzFzxyD!_nXjBKbm!n-(m;GA8bP1yJS9-PdsoQ|1BziuOkpi2EQt_uoD~w3VIxtMm^PG$k#kc}L#j%7H0c2MRHcwHh+l2j_~OafevaIk0ChlX@7u(Q#htj}Y$Q?$~0 zYAjZN%LK!$=At-i%$i$XL~}6#7oIIt5f-CF_(oVkavTawD_$F3g2=f=BiAOv1u+W7 zi392G74@|fyp&~$VV>1LPNA^XsE_E&S4RM9^)B@M)K4)`KSVs-cit9M<+$XKiU{aS z_y}BpGMg@{G)$)BIx$s?B`-C$78S*bHE{bfR5FLsC`?_FhMubw>~A~{_zLEC-f^~b zd^~zRGs8k>ur|RRK7ZeDe`=&W8D9X~M^ThCPX4GOkjI&#d@z>cL=C@%j;V z?8%RWZ&?3e;Ps~N@sEh=wP`XFB+8ZYvBTW)g5{cg!s1Eqp9ZzXnGHoF2H&l$ON{RodaS>4^IA)| zW&vMfPBaryG-tgFT#w*1=pAi83JWTn!m(QWd$T*;Cx)x#-MbExCh#1UjE4?NRWmd? zJM`pgrkfe;`+8tgy=IhPs?6{Xpp>viZ+_N6Tfv^3npf@x>|ie5Fetx73J;}#Io)^d zP@z||!g=;Xa29fD2UlPde&<}emyR*s+O8_Fo}#FytCCe&yR;}cKrdy~q>KSXJ5mHD zHYERiq*+>-bDH&|L4m!M9kYiZ=7_zop+oTi28QilYBiDZsMARZs@+$Dg`4V1{0!)( z)UCkJgv$}bpIZzBby`<8YfqjbCoi;K-+SY424nSME<$Dn6=S8w%p^PXCG zV}hGOp&!NGrH+RRRPdOCsnI}N6e5^qQ5{eo=T!;KpWv!SAyQ2Y^tZD`T#L|{s(J_? zU~UdeqmL)a!NKd_XWT}rihmBOt0irwz@i{Y0Kv1$28;@!IUeD|-va4)5)C_b2qJ2q zmj-z{`Kgx5#mtk1FK^=c{2j`Hpj)~Q&NuFM#V_B(b%V&!Zc0_! zI~x*RzA#`?i}eBQ5?KpG^o8M0n;6yM(c$&kyHUD|77P7h2eB#+H(#cDU%dFctnpv{ zm_=y^4#N6#WNKHwaUAH<<5{^osxh3Z&)zUt<%=^%dwMQJQbU7VEkED6{rG*!m)n|$ z4|DG|eDC@hXMn|#91uMAABJ#&Hq{A%G7K*Z8RCZ<4)z^24lsW)4tc9!%SWrW%(mri zEAcVeH0+csPuZSY^MZKuoQU&82ip7mEG7U>yTwI7~2?T48M=Fea~~2 z=XAdR{+@q&F>{;dzV7RDt?%o7eg2RKdFLM1yOPZ+?_3-&VW-pAJwIKI*n0-PlHJYMB)hg2 z^>Wb(IZ?ARmew|_r71pxBwgR6 z_{NH6=_^C+HpStnFj?ELoGg2A%((?;3pNa_BI*QUtgFrG;`8%Y85zZMxfSoQFtP|6 zRCmm5JZj>rIRq1~R!}I}G|1mjee7m%i)OV=PN$VKpN5wav8=y?{HBI`5Ri0jimAY8 z6AoMWuKdiFDN`x1cM>462_?Raos{2SMAq zqmM=L8lyqYWUTmIQ4!#__20MtuYC=byJG)K%+P5b_>=&o{Jh9h+QOAeDfmM?`34v zog5`(aGtAT#Zmi~>+Fn08lnQ%5;=ko4pRx=e0xIN&h3WPi`5f?102G3hMGyRWO&rd z1SSHrS;M<#I_*T@lkobUx{gQAUJs_Vf5c`xl+?E6D1=ttE74FVVzi?|s^F9pFpF`Q zi=Qq(96LTGH>@?)OWWC7`Z&-+?1Z>&gfM2q*n>PPOzg2=UM7C(b3Or>kzs1k7O-x< z|5N_rqrm=?OYD&O^`x^+DCkRdK4(WZT5gI6s{q+ ze(i==!X>?#`6`cT^70JXN5X}+StX=0x4r{~4|lh1qr_C>mk}1?2sw>J5$$TaUFuXm z9=)<6Ps*#ppm8?^BgF?fW?c4Ei>}H(nP;2WegsORKSu2F;#fi!lD;W*4cX~;d#!&y zpYMwITGGpcjNdI#IUt7+TRu^Mf@8hXUzJq_%gQgV<=Qt}wV(1+OLKr5^y7@fu4Zbn zN7W=9gdxnGtGOyF1@ZLxaQCW*er>bIhSsLMd}dfCW!m3iATMw|nYV>;a2-5U#)BMA zHcq;E<=RKgoCWJ5%{^zE#JB$tU;nBiJDA^nci(0?^K9FW%h2J)@g~Ubw1o*pKpJ#G zzbf%JqMF}38+n@}(U+}rD*INrmI9+Z2KD8~h$g;i@;;v5(a+ol-j&JzGJuk z`Ay4<4GB^i@lvj>_b=M_l~$7j6ZqA-&AgZ9=j`eC5A+rzGAu(@Q7cAi7D-HLHZfu5 zZ#cf$J>**I%Ir+(9{$7GA&r{HV}dX%?4xC*hQ`BT2u~M<=i^hDN^xUu0oiusTa)MfSLTs+opVm^)*Sf)@LOtT5~2 z^p`hK+0d2di)%~c5!%*092Q)d)=lsJew5U_H69GvW-_@(!l$SQp zB*VVG0MEC;$}T;>;(Dn>l{sffA3sNn~0cF1YjyQR)dsUHC-emnx#4_733gmSEUPWa(chhpZ-`B^rz|(k4#NwS$-(PI-x?tHmgOeNBZVelTd-1&c!Io-LSY7EfMkjuBB! za;>PhP@-0>^r#Mw-tgwvcQ#fklyaGi%=E@yN?v-+-r3Y|U&Z$xWwpAoV>HKiP4}^T z#&nsTRd;@7@L|`A9c21qpu#IB@|^BYx-Ar;LNy?b$oDf7=_>5E+QA-QyUH;GWd5 zSa|4Xw)nZdr*@C~SYrf|Pb~-5F6Q=;){*Z%+JAbq6uHV)pQOq6n$}*5FBEXH-e-un zrp~K!6Ss%jS96|9kSeUGA|+tR$pIq(0zg-^_qw7{OQ3Rh50}bH&lARy(jtQ!snpvC z<@82ih!6#DqX$Fl%K49`_5&qkVt>8vrJEb##k+Li5l>a2Jn5xZ`i+B|pQQJtV0Iyy z@;)?(61_W*w_$?aKh*4YV-nQ&lg^H)cOE!^GPtdcbPE>4Qs0*#aRFjMxrzLRqq$sv z^7biF@|;Be#f?B?t%qV;qn~5@E*Rt~N;*xCNxBi&1>{WX}dt-F33iAF5GyN6r-+Bz*Kya-5t$tP*x6Z>&D? z;sB1N2ew;Mqlz*!Xd%#ybd9_aqm0xD=mIIQ&EWf;juXv#$VEqPtx&1d1(RMemw{bV z1E}Dmg#y#qSh4dLDx%>>UZD461nc_U>_}C)*bv6Y*jYBCkUR?qHQe3``y__343%(I z$BBCGC&%%!8T2fYH7XXJ#E6{s*V%37eqMl-qbt}99{tXGVPW(tD-NB<9hI9ah=EJ#x--b4*}h6Wte%lq>Z_~I+aY2?)@44 zKF%cD>_3rt$u}S4b>l7|@f6(L=JRAwN>9>JW&2?42%Vt*V*cvub4DAC?bWjnUs@!5Ol>G0Z4H?6WWf4mS3c^%ttslB?u zl5s1bm8AOWXoK0*C-(Gm=L@OBVTWIc?Jz-kaMwSlIT4T5+s*j?O85Dj_r83c*Z5!z zf9G$kD}OSrTlV$E$sHWmglYAojx*mJu&*nez2e%+Zu;)tm7*ap7=vjmFR~n8dSTdn?vU(fk@i<(a~e=cR-rAe>gXr3Jc?`j-uM;MY=u8BWd_xr=6kk4iTaaVT5xZ$NO zjgnW}3d7V&Z#VykfBq*=_uD^ixw}?o`uX82bF7GNbBtA+VO;e@bExlr!VO={uV3I- z3-toAV;9yGFGG0s({TL5r$E1sKM9XM^+)af&lCC6Pt1Y4o;_?Ya?JeGJAZzhe|5?B zeazBK+k%Y#_}%_w{C@q?zkkE$+ZllN6&tk8`IQ{&-+jzJc>vA-Ija4u@AeO`edgQ# zjF2~F2SfeCg}BfMe^%+kT?u5s4?**Wlg(fpOTe*fAI zW?Ecl(1H{^{2|6(Bg9@_s|Xf1xV>Q3&QOP21M?ZTIiL(LdesZ#w=$eM@aS z2PgFZ3r+`MIJfQEm$>)oSE2v$H2!ppfAcQLYJlM4=EQKXe)LcN$921SfyQpT>hhPU z?%!_l@1Eqta3HvEP9U(89KU#VpP5Cx_vY?=gTg+7LaD#y?#4%dyz=kN?C-7vLGHrC zdTE3PY&=UFM4}E1`}c>%idseMB<(~v_0hh*WVs1e1heNL5S$tyHNOBrmEoUXvfPfS zOCuMr*AQoR8K)h zo>v9}o@tnQ?>T3b?|l{CtHLFE<-p@Muqw(7Y%dKvuf?Y`?@4>ZB?5?8 z3Wy)t-=~qaG*v!5XHb%Si?x;8TnxMb%~NcQJP*VB5`;X?1BFyKKUo5`?ynxd4v_)jr1&Y6#15Aw}B&J znn>Y%4b7NMG2-Kn)KGvqP(fEC=>GAwL?c{QR?h4*)>p=rk5oQXSeJUVfB#}0-+D#) z)zRcRmcu(!tS6U;08d4d7!nBk=SG$8X9HWU(y7sfnxyEkfPlm zcvR>D=;p0R$QEi;K@y;n$nWkz)98ofz|L&2l5FmSSPF!J0aMqY9S> zuO`Z5`}?r2{TdUk5x4j2)h*bu2^oe2#fin(RT!wzt#jT0SiB@@3UuP0CmZCC!4j7RCHGL?X9bxvVf9J5Aac0zSG}{+vv;g zEokB6(#m|OjJw3BySavgE_P0~JI<^+;KUfQQ9|0PZd+JjpQ zxLeiJ?@?h`>C+eE#9hh#f*3vR1%CCIhl&KK-XW;xTH5*oAt`b1E~j|eu6Y|`PfA_; zG;yn)_AyD`?Gxd7EB$=OZ=0|F=_LIm=%)s}^;H~Y)?jpcIi6Q}aV&8hZu9}pzYmo> zTL$Q{WCnmyP~O?sm!+L~is8taCqZ<}VR?2NZOsxTf@IzIFRxL=ETDjNA@+B9kiCy}!vdcgKSMNy9kf zs8*@RVQ0*ZlqLo}R^EaoJJHVc+dFCcRfe0B?uBW)vpMvtO}L_2o~T2W-a&j=6mrEVAx)42rRv6$ca^ojU1GkKK#ZPn_3%_msiR`U^<{ zMrh;}eCstltiXQ7CA*UR^+XPRrxE)GosuF1gpRVx@*rTNu-VXZCly)ltGBM|<|sHo zZSfoQ#RfuBocdH}TDVYxq)Nk)vu8{`J)h9m$gVdi)!NcHs~-J4eS`K>%4(M+Y_tv{-a=+_ zj0)1(OI-#fS@KtkaLv61>LrkNp}Uo(*0;v15+z|vj>~an%y)qW9@r!npxrYI*MOPfYf9@a*x{inLXT2`?-dgLX zMWBUWL6Ti} zxOykY$CEL^x|4P<+w$&|t&uN89>?p!(*zNGN3*iPCPqX;o>Y-LH98HWLn6l!|#-(|{nKD1-fo6ethwcja)%+2+p;y@@^Cld}g zEKnNZXfED4(W;av>8#q5ZDd_;rFYVqv=sMbso%bd4Y+?$ig%cs)@uzHjMD4L8sJE& z#uYypQN%iw>qj)QoVmq36C#PVFeo|HWAAD?PKq*iQ>dm*oM?y=M-R?=w1i8Q=waNK z(lgJ^^|=KL>YEtH#ZHN?T6thAT&Et2J)TD-YP%0X;aO(4li~!lvnLlMEVUDUVrMd8 z-FHk@&hhh%{x_;K!{Cj-1SS4`9PZP$d+zqUq2P1xE1sn>Eho?zA-w`~MuhnYXR>zj z0oDtaE3ucYA-PqA<~Ra_d;dxD1K)M0csXz5tys9zy!q1ULh34*%v*nb2u>+yJkk6B zrIy3mLoWjMP36^C1KoC|AH6gHSTfe7?cyz(!LYDQ)8V|_jx4tvbtTgId#UsATPQ~~ z>p&4ozo?N=I@Ti!I$|9JF~(tw=8zg2DX;I(v6p{392>jhM_^n^lLJXd>HqFXtR8zo zsF3>ZL7Bx1CzQKHSmSb30w-G1RNmh>?D_$~gd+OZ|UY{EmcCCJ36EtC6PN4}Q8sv5DMWcT_{QzZXTe!I6_i#KClYi6+O_cwzZfmLqF$5;`7JA1331l8K^LyMdtNX2GMEg9t2|P009a-K{u2^$&HbAW6Yr4)@+P@<)&(468kR zMQ_y3NMzrED+tb}B&CEvNJG!7 zUJeaPx9s|=X$;_d?M*eYMI{=iJ~3$slPE~N9o2gO450LrBkjSS9@f97tKee0H+y<4 zxroJ8AP^oLN5w(;E}`iS*-t!bspK@>_7uJ+iS1co+;U!?Fi*{S=^Ycc*pt8jPOO3; zlMQu{M_!pBpDX5LA5T^tr{5XnmICD(M_nMO#sM`TfhdFJiwJ@#;b;u9_WI*YWy zyyg`sqYb4HZ=<*U7*;F0IH~bha70*ysc{zZRv+1ksow)}V zXeL!yS?c@Z`#5x2YY~2}&i%2@#N~090)Y>BYe9qb*AmbD^Vv3+g7LL2MOh_CB)ZtW z=~$-`Ij_#sztwp(w#b?^$FY5>8h$DCrojD6NYGJ#TN}V6C0-U8B?O(jTugCbeDGgY zxKmK_&-Y-99NmIl@ev>%yR~A|J67vOS9?JNLF#up0q~QCK1)ens6*QkG;O+qqaC@|1du1a zm4Jj7wpmD}xyx<$B#j=ePl8J7iS>!u44eaysw0f+#LLUE8?j>Qk}gvZ70Dd=Jge&0zB-Q4=`Joi+55B2gm)H1eqQOV4PG)y8bVl5i1#qxG47}LX)we;l^59@`NUJd1`zN)N(3wMx~1F zg+Y{CCc}@W7ZrfAP>j_ahZw0XGn`!~|2hLZwb!DA!Z1$SqbY5&gCzHc{z0U!MMF@Qqn51N;7QtPM!utZ1&>0r^$Prnl^l}zJXl7&1RUZx-h7@Nb*J3bZ@KM|+sc51 z1Y-Jas(O@07@x|KzWE3JMW5DueD|t9mGRLolPO^J!ab4xsmIRi`4dt_YAm<<=3n!( z{jHz6UIaQ=AF6l8|3X5+1A2LgcWr-ceE+XUgxPF|!d(}0jo1I{RexI`Vm`gq$9mEK z`OJUUKl}hd9^B3=@qdBq{i}%M53jvAyw$z@Z2K_mht&L!cOl=k)%TTzxBhoG|9Kzi z0ABtsc3I`$l`wy?;Jnj7bTJ;K`%l}kg#Yi5p1*r~DH=d@UA-{#zZT8k{n@Wta|3`p zf)AIxRQYc#+_oKe0gaXP^o+!n|4wWEzdr6hZWQZC98*H%?GsG(MOFOc6TP=jm26Z% zGbW^eR;BCJ@^?62#+}MLQy+Tc2wRK5!j+6y>oCVX|L$4Og?~gaBqk-~DF~qlU6X`f zcK)n@%)58=?9>+5guZQ})=|Yt$zi=`l~E7xUOKD+PQnQ5HK)}V01l_olRe10Rhnab1(xn;y|uLL6>b?$CX)5iTs975?aa_ zwB--k5a!D}LiHC61rw|%4NoL5y(lx`{3)B^r5^aUInleh=f)=c&4GhD*@hQ+ysmyG zY`-0Oj?Y_TqjyU>re55_Nlf$}f;r=^o4Db%VNL&rwF7L@OJ0Q8;!tjB+|!NJ25*m6 zy==>or_Tt!a$J0b)&vSCf8lOMXW4rrChi+Ckb7vmmNx&JakJufLvLLrS3>Zd=B*81 zf_b2geoG1rgpCWQnLm7dj`xase8rqLKyRM9(osc}k}X*yuFsF2Oc{Yew?>9|qKDT$ zw2vA_mAaC=V#K{e@1s|ukx8sSq~H4tc{`pFwhmB}HKj}-p{^-JF-*IWpTRLb;@ZOZ zJ+js-KbutEScY}i(R9Opm}0B&&`I$(iJ}F8V?3-c&37bgTFjO+=EHE9FmcVCw!H?Y zyGSJ}uD)F+eJ>R#M#G?QlE0c{HlbO-$C&DkF%G-7phQ7%kI0PhK35V?cGWAKHRWKDA$HnbJ1}fc~y6aTO4!jjK(ASq6{N$X%!mrVoSVem@v<~)r ze^kH~1LbVeH1&!ArCTTR&5{T@&0g-zV=NNv2k(IwBvdsD?r6X9kp{odGt6K;KS0wt z%{@lCOHcE&$E<^$>$0XY(qZ?%e+ewy)DiEi6p!BbBW{plz}vBbTaB|2F3UNq&{~+| z4Efm`yWQEnDo&b9P*0LgFvPeRX(Y(nN-}ZDp1K{G6Qau6gy$MFo!V-)5xUrOE-D8e zjdUi8ms53Ldinqhq5@RZGjS5&^cP3!l*R&h>Pm~fX@m8aJkoNom3Zq`R-W?&eT&^b zlc(-{c;K})N@F1jrfS1LQBAeujEXd(`WOUdY{AW(S2Es^p!J4J?j!*V&8^_B$AOGR zHU#%Y`A0gv%v`%pU33#F*?<}GbhjFDX8+`Dq(AOhk8`C=3{^TPWoaYs^XW&l&Zc*J zC>=<|MvH2KDV7s&x`xd_M`Bj#IgMPZolI*>u610kzpD}<;`gXEUPZ*B@r*uQJj7+O z>+u=|U**`dUzkncEnX7xRsIl*4G*Iu&o#GOhjg_++YgqTO~hQe>7w%XoJnq8d$JOt zyF^Dvufp#A7P|XNnm1)eln6+B=a<*?OQVR|G(cVNb_fYmc|2%ftIzJ;IdL(a1D^8y zWvb_z$dRM`w^|dVNbKIk33Y6~kYNFL$Ti2Y=nPVFz!4Py;Fci{9iTq>tU#1W^In$B z*Ck~FvwwSWtf4Z1YA9#7`)LB}y0aYWTeiB=mpC!|2-7+ZUEEtk0UeLe%S5+9Y_Ev9 z8cDCpX)51YSS*@y2dwx>y`Fc<3$1{!jVrNt zVf?qvIYrDGc^5|OD9ugyXC(?*_s@e)R|Sb|TK6WfI!ef}^b9aKc@r^ug;ZC!xNZr*505X_jY?AD8`CT&vT>4i2! z-?C}6CZPLS(hwD|4YL%?$jKFauB>ptAwB;d(IhGFAwz56X=Rcmpi)Fy`$n@-nIokj z%d|?)z9Ec1;>=5yxY|kb6rtaaDw7~II2aJ^Iy-b46bQvI>B*S;?~-I}hiH$XU_Ri} za$>u;{UOe4kA_aa%%;4PeF78SyfIk|0o>LlIZLrI(1{B%iikV%gxfz-p5!e*aNT%b zx*RI8^vmTwx7bRp{N}QB;DC=C_i=15iM;&W2BH4&lf9N9SVi$Mb&o@B=hcJRHBEQl zdoPsP>kHOz+(()(4qQQf;yi9t{4nAmeNk%5RWZo)Ew!7<^ix$HZ(?v~&z)O0SqX|O zj?NkZ1@op|th1xjNdHh`cjkEjFM54c81k*`y#57i=G8Pd{fcJ<$voSB*H@Tf_v?oC zSXrLNEuVXE8i%|HWjb*oa8r*jp@I?$CMe;>>IotmJ@(%&40==(rk0G4Uv_*w$x|Vk z4ZT^t{xGC!+P%B66OfQcG+;EHcjKL(>fg_~uaF2Yj_AtKlllD7deHx(k%B8-P1=4o zqkDAkp|ww15&AplrNq!z33%_}r_%aGi55cTxz3{^ESB^|B5AzI-s{qKGx*az$O>mW z`>ML6!iwfvHXg>+`|0ZoB?GzK2OWyP1|?o%I+CiA(Adjmv$i~SWwGgbgGlTC%cKFa zeeSIO%L?U!8;k2OO_d?(vS+p>Qk9JU-sH?Kl-3sdL5PQO{r*I=(Iho~$^rU#sj?8h zr2UdjJO^SB9PLe>}Nt-#{Ekj$L>(%S4T#who zVIdhgRWkj{^isQSMV9@Wc&WihQ~CW#d%4=)8|-zbN3bF0G7?q@wFl=Ha5EAIn#ohH z<+3xQHq`m+qyFSElz88ZY8^#u7~$eulGmzAu}!~;%vDF}Z}T@V3ML?Bx5@vIKBw=0 zF2~NQg?{V)OP>7^!PYi!S|ku$u9(6mY!rscz%EL$F^ab)VZId^)e>=4iupO)jRk3_ zS0jwE(q{o_2AN$FH^Q%x7$)q`7&R?=q9I#isp<{0r}j?2i~3dPIE2ov7K1~w7DBxX-U~X-Ss(IQLCCD%%+$j8WpbcSJ|d-)0(S>S)^Ao(<(6&=Mg?SC zG_1VtfuV%*J8?%M$lC*32dVXd`}6z$1Y$MtK_PRbYLffBX8UtlRd z4@lS;ZrAz@Xe{;_WE)dvrWv^}$gJK1N<9evw~Ecd(^~_&!7|2BfO28$7*QjyQ-hL3 z6s(}hut7Z@)yaPQr<2!;-h6Orp>m_v+_Z5ckr#|io_)*FsCz|~2wG;_&YgKj$Plkd zHO_$`L)V5D4kZvXuYQp69+6vA?O*93hnc<7CL8#Af+F>) zGU@dvkK0ZC(jGrue(8@=>mJK}E$2P*tg>t3BFo&0xUnE3#eLH>Y2-RVmSbhz^IJ)4 z4s#SannQOvr;upPFWNpoTIajImlrYXDo3@GoQpzKUsG_HQN~(7GDh!_z)!8`#7oZR zG|ot@$p~5U8rJrhl<9b6O-gm5cMSN%u2W38Pt-X9G zGqXbW<9y!3=BNvmF^jdmxhdeYT4I+@>j|tYW+-ex1wFzlo zd72t?F!OOsTSnrqTmWQh$g_`Di?3Pdo%f)XB!>WJ!&pNGbmhcY59uv6NMBET-SxTC zajY2QJ8|o^%1j~BRAsP%|s*V$$;g7s5WnUe6a zC9Xv6jt-K%QShkD!YeT9!jB0AsZOW7}tevmTw`g&?S2)Bvlykd99Pz!~ajH)#< z8QCHN5YScFu+kQkat-P*gR;K6V$nDmtMv$q-cvOzFtnANf9gS~LUHrBFZ#?29;imrp|X6+yOT;S zBhy_T10I`^k+-9S1O5F)*KJBVGqCkd=)$F48GzdGyqRjmrKIzWeneVN?=ssq5JdBv zsYX3LtQE7S*Ps}x z`E!OM{@bx#oZ^~uhK^&CcbD{Jp3H~9KD_3ZGETImrydoE#&>%!PhOttLX}Lq1{Y{& z=?=wOC^y83moP+cO53G0O#;IH^pus)P%o!BbJ?a@sg+AUo&lIt+c#-A8Lo}Le$-fs_(tbA3r%0D}(`5DrIDWyO*p~-=&V8QD-1aDN!feD#NK6R^hIK49(YPQzco!B|KxQ$p7 zJ!jMxCrhxf`&^%w@6dS+`%StGrV3hFE7?&3cbf}ls-Ee}x?I3LB-Tc4&NeEkDT?#bxoUSuQ#DFJNTCNK4$!v|m*s+8>HGa4{a__mwmZ6C zib!vprnaUh#r0?w<#HeUz^mZ!>``mw?w&lp`f~CKnc5p0%LSE--q_nc7h+@(7ykJm z?c-%;+3H0+ulUUQb5;38%?^&~94(r|6mS3pfkOL^F*&0erNL5Dg54Gmg+;tjSJ955 zK2Pm#u@ju$kLR^^`nEzUbq}x6OD-P~9va*%$k1-$OF5}%vAaJIAaP06bhFQ{CBBs_ zm#evDPwDQSScI(3-?TQ&f21;N(y`%s z>3&J0yeXc~$`f#lG@Hz$c1&u#v zbt&nuPx`92oy>QyQrD!S32?E>sZZZuTnIoI){!N?wP$W8paU2qsbnj-2i~UtMqhPz z`>wWG4Jl5knI)f&4+=&V216e2gO(?|HJuV$i_M5)I`f7+LZzPLuPzrW4U#GcA$=wA z!2+;&z`{srh@+>tf8_K{WuD=-y@HBXn{&N_XJ4^JkgC_z;+fK%h>q?19bDf6dIJv^ z98Co4;Z`fjj;_lg?(wNJjpWRj4DcpB&Vy8=!UrSY;TzdQmCiIa-sPHo7W1AI4Ezz` zq)EYZwWfF1C19A7lL7(F&el-~?nG}s&?$1GYs4-S&qdt^tRJbN-snj-o!JlSZMgJwGO)jW(9=Tkn~>13zJnn>3Ba9d5RBUF&u3e@!xX>;n$HF zQgPL|uNzkl7xkFVep8n_y;Vz0hxmD0oLvHGiNkIzn~+mLYo~tlI`r%M18$eRU@~9| zx>w@D=w7M8vh+at*)E2yd~w2r_1O;s3@;6k0K)(5C+ebH$c0Th1i6-Q&8xD1OZZ7cMgC zae*OGGUIS9<_W-X$ltjWQi(6AI4a2D_))h`$f)>UJZt=@K?Ew`;0yr-de>|l!ubn6 z)mPc02i6RbWUE_pS#kd43bH4b9A{1WW-jXC7+#DmTSDX#y!vyh|4WaEZdZOs}g=DbUU3EorPb~MGV^dy+}~U2YYMAP0fTkn%&-eAc44T z$-PLinc7xHL<{#36ZAL= z`o4|wwd#4Rvk3KLcu6h4uPl)Ktcsjr@!Lt6qPS|gV6YW=Kbc3yYhcL6n2ZnD#V}W$ zF%zVJ7cnqNdPMDF$VYTKt`RBFPs@VaRyKa9?ageo)p(4Os*(PsiY33>Zjg#KjDaAX z#Zd1my?mv{GLz*kkY?Ib>Tk~_Pvr-X4$K+$ylQOH$#HRMxT0CY%-Bq|;%J?<_2WyJ z^=A#+B$6s>x>5ZuyAE(!GvBQp-snSaQYta(iC&}!=4*~-o_ly^xl}<>yUcMSSEq$I zQ+B|R7+RDMXtUbqvzM?skT1g;M)2E39E@f&r3^J1^mNv3uEjKmDvoD}C9+s+=Z*S& z%7EydLBgPrS1_t(0mKt9Tb$`NBwvNtkX}37FwIdJ8mBlT2(5mNoz+*7(CesRU30z@ z({UE@1~Hj@Jxsk9YS2&oAFyaQY~VbBF8zlUqvJCa+8*`=VsHw0x6e#6 zXS6mG4}Z4@e{2>T%`yQsV^rY8#qV~_)P)0pqkFb0kuii{f$ha~( zWvIoR`rK(*MDdQiKzZHHOv&uDI0GD8qAY1Q=}r;F#ha)nVzF4#kj0!RpM9`2BTXZr z0|YH0S05iCLG|Wwa2?eQ6>@7R#s&SYGT)s^#0jT zet|0~?%jKpEBi;gv)o+EdWs|~FvTq%*S^wxe0J_V^b&KYn!Aq1#!~ofwG2j3^3?r; zUqg_^?R)N|w?xnOIst|cy?h(L`#DZDxCOOYBGI<|op$*N>AQ8^Y#pyCG#MrTWf9Y! zfe;sl;@SC;_cheHqSI~IfThZ<(U`%_IR_UNMl>(R`h5`?RpD^o=!F1KnKPZ$2ZY{3 zAWE*n)9RmNkvhl=pe!14uSe;ugkves%#+!VKlFGn-vUkOgXMZ^KM(u+L|=H9+&>Mm z>}P)%FAA@Anrmw{>=|$mG>nQ8GSYX?c08*Cy(I`{0+@;=phE*S)OapnNH0gcW#TD2 z@mAVndCoy2{@N6$U8K@}>h#mX$M+0;@XsB|Srv=XZSZjBl$yv{PY{1gc{`UVt~pOhfM5@ucjKO1@0K|g z>-teL)pXi-AI2b}7-#$K45-aAE>v-MI8>JSnKKI;*KeW&DHf-kv+k#K5A3w{z*ZVm ztQb{QC>qIGA;kTb1sRgLKVE)!gDj8&_^p>8Yofh^`?heY$2_Rc$rGd6wa#`V?DS9- zZM?gfRQiU!?-d2H&(>47n4e+3W*;GsEbCU%d{Vk}M5YH<4!MmJSYz>`4HoHnpZ^Z3 z_tA3Y3Sje4H&1xEv(+tU!I>OP>0fiw{sHj8%eQ}kMvAEWj5KfkC1qqjyF;I`zgNt@ zm#>cWyn^ilwXRGh_Ns^C_pgk!&CKsck9CsL-`>xzX>(x)nx0^vdE#Kyg6;2El!ksu_dpr>_o7X;G z(U2HSpC(>$Lz526ti0P!x*f1$|NJ-~j`DH!rPf%Qn7s=92I^5$PX#911RJdLBdx`U z&*baPcA9=-y>#=J%U@jHA=xEQ;9A-Vx@w3A9-1TBG%M146|%kJ-MHPY1H5R!r%Us z0yuTtkA8<){e1DmxTla>I~#2x-JuJccpJKo6={F+{z|G&uDtB|d!_Fz)O3%D=$^DB zr7JXpQ3W?wc#r7%EZ3RVkdZkDPwm?7g_A9_&$VmlR5CFUwyNLuvG_{7VX~**6doOg zPQE*o;j;?FKhP=tu!HOY{m?qN#@$Z2Z)qFAx@*M_uqaaIUfm@sVW>j7lZJ9l7h}aE zL5D6tcXrLo=);hpMTOf`7^ke`ebBXBIf2L*yzIqn?r0V&dbq_>D3$_5+zGGa8_I5n4%^mw;58E_TN7&JkUqxp z>95(xzzW^V-plx>>TxBp+o(L=8d2kdgAtQhXelqr`_ipyZn0}^oq0c0G`f3=`RG}T z@zh_9h4t=2eRRP^Xr5~1Mdi?B-=XL&SLc62$%^Beb8%PZ zJwJVSyo9Wzkg#6)CsEWM&92DgB*qX>CfRc!0qlcR5;CluGYXufG=3Y73w7-8NrBp zdQjqgc7XFq>&WJN<@XtI^LO9ZHGsCXA!4NwWl)j|-Beh~a_@%8-7SgC&_wRHXqrAG zpb@*65f4TPuY84@tWw@rRy;%DN)#(vi$T1nSK{@)ps6T(WYr#X;yeIh3(;1%p4bL5 z2wq17{&A||H*>-%mMk(jK+`Z&@>%p>o z^VvN(;p5jzlqVdw>O~6AcI9X-Z?0-#V7(=q@4#%XPC15F-DUml-Xhz13t5Ym?SK?e z;LEFH#UKo)$h~$!!O_{E7{4qfuk?i)xIeUeSxUripkfm}kn*MJ5dd2xc8<4%-(Ic| zt3+v_rgH?Qz(k58UC|FcN?*e{0oGf*TK(qNmbhyFUBsnBYkN;Ht8g(+vW{wM36>8PFd-LyzqaLT%q^lY{VePVKhF3Wial>}G5`hkReyl;?J zG4__#3TN8(x7Ou%Z@f_1;o7>_ zK7m%5PAZ~~TecPf^_EhEU85Q91NAxj{<5`vSi*b&Cu2cuJ=c<&8 zyG}grJIAnYCZG>(7x?DVId4)YR5iBxGTbGKNDjwh!YH!G&!I8B3IdSK)5^ECw~S}S z*Xx7`aE>AbwD+wf9$#8=13k#QmmCqAPBVQ{y539M1XlM--8$+DX3IZ`?(`-tf!Pn$ zU}L(WDG8?vDoSaPAgi3tDe1k8?95oTE0Bt+MJZs>qa1 zSzE9Qc7KCB0N_IA+RyQ0(*-)mE*cBi7ZT4Tl&^&8lVeqCQpUGX3E$#m(3x6zdUa2alw-s5wB?)!7!_c!m~e}_8GdA+Xd^}L?rF@PBF90qR)BZj;3E=A_7 z1MJGD{ciZzH>{yN`lW4UnBY$-@}r0QGZJhVbIu_bWNw?KIV2w`gl@8+Y!ga%YWb;=>Tp*VLqL`CW3F!#i(_>Z084iJ(EHsRFDW}BZyK)bEsj#f zsZO&ttt6T(zKbuw>znZu7F%SzcGa>>(EuXng~e}$I9`3xD!pb7QF_AtsNc-u*#%0| z3jsuFBo-gH6z^%_5_nMxwdXSG)Io^DN?_r9J$vih37F=0c<l^KcrG-+GI{*EaGNr&mz}Zxa83Om=H4(G?B{0=cc<-|J7Ir@77N z*YNl8JysSSUzHmIU5XLb;YuB6NO`o^Mg~hVimhD2EW_8bV-2TG@hq9S4R@%19_D6Z|I=AzN=gxJ44-WaJQYd}8#;G^Blaba|O=-Nr2Ag)o-M|VvL5tjRiPf6m zMP+2z_Ao_Ux=-*L@^&%xCebcUDPZV01&gQS3tdb1r5(A?h_5lbu>!!o8eZw==Z8aH zFB|a)z7+K4ISI@2_~y1!)~Iu1iO}Qt-YBQJ9(z<(c#@0pUOScb9$f4)zL0!p-2-%aENAxS_Pc(t82pmSHdjZ=q1); z<*@o@2`%+bjdw~F1DtKHPxwJ4X;+8^H>#W`h|Fy$lI$lPCM7i{%RX9~_~a!FEZ(?i z*XEZJ8vn0F%fA^uM9aOv;NXX)GijKFV{Z8NOp_m_Oua7~3#Cf*u{&ZJ^aByU$5-xj zI5KiEZO;ffru8n~wt+#K_>Nh?>~-d`;V~`hOZcA1N2kpK1ha~?7$W0suLyw6ZbN1sCZ4ntZaP z1U(vZ^lmWe6k2I1jO+a!Wb)@Eb;#4>Yh`9CWDl>m$iz248lV44ZpiB|lb+@7z1(?O zQ+1{M(={W5!kD+|(Ip_sZlEkg+YgLyGJwd$B^=wiQ8QYp*b!Gd!Gh#6eBlRp1s~_` zJX5f={9@p=-nT>z67Va88B`gGF&#}gb@y12V~KjjV=4}ale^ReMos%M$F z?Q4k&hf3^-*hZ#z@LQduyI|sN7Lj=^mv)8S=3CJDFwd$nt1LUWeSsGC#SQ^dG|^ko zYx}DyZJXea*yaILZE<3_#riX>tmQyb9yw>I3~IaUDL8kWKF*21txD>#?FjQ|X2H&djcR1JfFQDK!mF4!-abMRqVB?% zX0ldlt5;6K7ItM#)9~X9tyTy975QQFr`|}oH(Q9o&)nT(#yyAVW@;5wt0S?#Qo(DX zz{{^WJAIW1m~N6)JG)FcLAE;aZJB1Snfez`q&vb_W_BYGk8w^EWQlcggb#jw!tbk| zi<7!EPaFrtOYoZ>cy>`1Ab7t@d-mSCJI|K8qNa_fMe6I+ox{ifS5EsM%j;&gBTrN; z=l77(l65DkJxwFf?>G|d7D^IB<1Y{v7SG384>3CQhm>kkNdO2QC!3Di(!!Ww-uelN z46_b@(}sq|={7`dC~|k#C%lGba}4_Oit4%3CkJXN0T*{WQ)JPDk1-V$FJ88I@?;pG zp;w}_bV{7mxN6{E$L9TSQ*i=~eED!IlNCKG2h z*w~SuZqVPDZVOi>_&wUbHL`&dp9Z*9jT#SnLXzrGtrL~>6`BeBK!Nhc(#f#dl&+bH zymPL(96vV4pPszEP-8JelC);g=)%A7Mm#lDAIf(bz^G&=Sv8 zMi!RgPG306bKW#4C@9Q!hLpD_#4K{gina8*xfGs1O*wrI(`7odhn+*Z=$%{$++V&J~C@}`j9A5M>An*)>efNo!IPQl{5 zdZ%9^!=Lw)3CMHb9sK&nu1==_oZcMl?Dwx+y{g)5o?Q*R@QTalI4^GP5Hfdfmf{CY zu%K8?usktn^(YXYeOU~S-Oft}eka7QYAL=Zxt3$4E>Jql=4rJWAa|naV*fJ7phRh3 z-81xt;o26tYY;AZ1gM}>?&Ca! zj3B&O1`?^%6KD=}0W2ZD(>K#=7x(EpKlJ-40{7JLz=EZN*Nht`Z9eAC zvyM-B>4n7LB8%ZpVz{os$1Fk*|6{*Z)286EjELKP1Su9lB{8@8DKHf6XoI37ChV&( zyN$TUGFN$`Sh}du(FMel@fO9T=2K9W+$U0WwKZAKO)nBA0_~Rf^a*a?A>7tkb5ORh zfVyvNAsBxds7~fnd$~p|BfGj40U+}(iJkwyBGUgo#O=~GchaNuOKh%h#=da>a7lDm zzhJO6Q5q@HgGAsw`>4$TW0l5MW_(wUscZxv`S$%n{uud9}M9o;Pdm zmE*maePkQcD#tbHT!+kUrAG?>^0MEOf*6o^eHaquZWz)QC zJo@=3s4^=8kb+Bh?)EW@yA4^b?+Sj6*U~R=O#9l@q>sqb&fY%NSBpd=BKLL9Zfnlf z9rLo#)Z3RZ zW1ETwI^F+$rT+K7jz2x$+*wbc2HDU&0{OkVD_r+2C8}(fgowNC`%E$-H2v~0lS=1E zGz6uJVz_wmx=+p%Q@*a1nOHiSH(Um?NUnsCDwSJCjcq3xc#AK4p>ErsX64;iTPw}g zZ+l^pu(Ie_Cb&4VXM^eC%)ulZikGPi3+!aqliwuz@SY`Ydahdmr-U(-6uHDIrt>_;M7f^4E8-KYV}p5qkfDEF=Tc<|MGwX<+{A#%c)4a=ua zrQH^aVM&y7TPZ>2rFWCyAxv{h<=(#NV`Zj)-rZe5@F%MH%ywRena@7#A}I{qh+f_) zQkUcahOp`b;fuAD6UgluCh9lZhTNLcU5La>0+*+pY=N)w8Je3|QGFMEZrg?0+Ey<2 zW|{2go44xmazW>Z0bXgm@ALhYpRN}f-)@|`B@R{oAsFW*fr0janeY~#>4?P+P8Z?U zp!6hQ*)!FLv?4qM$=*r3SqX5cfS78Yy}saf_bA5A4Zo!0vneL}LVVlUw?{2rV$Oz> zlxIR@p_Ld9=s5Y?H8y}N5C3UFF0&iGaTMbH!HBTeejGANa8IA$UJmP zm>P>r|zN5jZoP*AYpowzux zn`bpc?nh*%el=t>j{8&x5szqqf&;0+{2PS#5uKFtQd+7=vco`gO~OcTFl{P^gnIZGe52O%^)^_KrW%@)@=pwHk(LFjN~x4 zY@kvqv0XXx1_wYIMdbD_0r*Z_rIgvd-64k9L%XZ=+>5SBdjqwmfnJiZ!uZm6nF+w# z@7r&N_f&oKDtkZlmGtLoHZ6E&({f2b8`icr^O9l4^_eq?y;BgC81}>uOcsI8b0be! zV6(}o0X=~-b;shCE7WGSqyX6dmJ#Q@f0-Wt*B{IMkMWy-Bg~pMnGGx;hHsSygfbbz z%ma3t1^jbzL*%s=Kk=3u`yN^9RG^dK)5(dd^2uW=QB8A)bAWyF#ki_$hY(D2LA-t~ zwL=sQ#6MRRGC)4gT~JZ%;RaMXrhuMm=@E}cN}6*_G4qo^=P)~J&!zzl#A8ANS={Sr zro_WUeb?^2qJE>3=BDTFUY~(aZII-D$!F%3f%&Y!Q4-RA5`B;!HBJg0##DEnho43&Yr!zuW%w2 zED^e6Xf%En^o7f~MPRFDS-R)w>8BIIBt-^B)$s2@xiqBo6R3@qM2bEC>HqcX*Vj7t zkRdoQTL3^OqbJ4g1vK#sdIHtVAhm24bZ05>LGBbhdoOS+ye2+9c`*%Gzq{#|*<1&T zZ!VQ)>v&|<@cS@#wtJ{d^Lu;JF7Fu9Jv9UoOQTNf28~3M7F&%DL9bUz+3rif8qrEl^+h^hEuD)ci$_OvJ&FOFt~pyaK!NButszOK z3zp=h1u~D9_>*^^Nbg!gIDYuCFe=zv-rM@2QbLeO=Z9<9es{yPKz#OW1I>WXv$i%m zEGAOwSoe*j3BvdF`LuQCj$h&R{NtG$4`pc9swV(8ee)-)9(LV4gVh-*LUFD;?asGA z-Bg9MRUOOA4qdGiqE0g^E0uyL@^}Wc#L@Y1wpblnp9Nf*?U%>0V+lO+A1-O6IhwbC zWOt|1CfWiG8|<%Jt&4<|(l^l=<-4!^)WG=5mt29Z4KBMj+{RWS=vzu+i06S?@J|lq z)WpnQoK{{UsKDZD`^_U2D+Cs@m(trHXW19u#%!m3kEL zs?5)Mm@aubE<489 zd|ma`Y>Q^b(2c=1ot%(`yo&G8uYlHVPKe;ycpR3U7a9_BrYhIOA$Kh|*EhP6YGvr? zb#mjek95#RBOk+=_IPNOpv-9Nv0RX)Ybr&z_R9q^l%QjYosw1_zxLFpzZX+@bfjBl5-|%c`ouu8E=QwKQUhj?h9)-Na zcj`jOh=4b#q+z#*OlqSGmga?`bBp+LWssn=ha0h`Gr33Li-r=R+iotMb-snui9Tka z|JM@mo@pBQ@#=JRM#8Faoaui#E&l_T=;C8G&0{Dav~m?6rPQMLYMWkD@QLien-ZX| zrfg65Y25}NlQqWJDH$o`y5p`-9E_ff0Wyrb!7Lqiq4@WRYK_&ImaAQybrOBYW)U^p z)(DwGo%=rXgBG^0%S`+zS2tqqhdMV?5lDs_R$J9rft8AYfY_36m!y|^u$M&6T%5D< zs7VxaY1LWn_hWQr-EQUf=6=)^Lyf~$S#db!orrjhN!m}TXN00d+ zAbfi9N?Ij0|3LLRjy#F)JkmM>pV)^ z+&6<)1n%|b%W=#C?&wz65U}T@k>3qe{R@@o?mR;)v$1Bl!}I?q{($@Xd2w+0Wke6np@w+ zm)ib3RyG`aCO@wWLTbzbU}r6Lx29vpva+p1z+N)Wa@If=1rz~FCeI<}?5a?6v>90y zI$6>{Nz-!KV7MgR8oo4A0iUT|eak)4x`qVOdp&sAcoI=9P1kd){y46eF?4S^C0GM~ zQPLjt<-AcnXGoy?T>*dd{-mfr*)Y2w@Vo%hY*F;;7_5o1Ym{o>pKs3pt{eXmMF06! z^YW34V(!4%&>GgF^|Sws&3@g!pfaG_2v|oenWyBSGKNRzI@*Y-=fs!RGSGBY7QKb%*)m8JVksUC+-lPbrW zdeeJl_kpE_3Pxu%16-)i)=gq}Apdy^Xic?FDi{j^<%RPMoV!wGptc=8$sF`%Wf*}8 zd;{^0o_ZQu^&phX)7e_vDc3HKejsvpH3+a@IyOq7J>N}xePZ>V>}7p_k2_pKOG~?l z=rLUg7KEmnM)G#IR6ipx`^l>t3oZ0#PUe}uBA)UgkvD%+-)Tc2m0~w=(=3{!UMqu1 z@P;r+?g`y&4Gxnsl(#qCJytRId^1$Ss5b&~LFD0SX0fR|ACCTwn4rlfdmNR4ikb*O zVBmZ3iZLy++G68=siA<;p9M7fLV-XWQ>xFf z7JSL%<;)%Ao>4z}J#xpiVG=}G`rLE5_0}}z&;`eZ-0pO3ZVQ+ZKKI9SE{WaGnNd=v zGb&Vf?qZvp6}psGHlyo2>R(cfQja8XEr?n7e=(tjZ>=u>EjdjrW5gSNxbQYs>}WQk5$tyIY)AYv8Ej&ZE9#y(GyMf9A6w(K~fXpqY{V zIkn;6Sg`wS`jk(aI0{bDta-21A75Q7LJb)cIq5!oPv(}LtJ!?|?QKg+jVIR-u#ljr zAZ!dzJ86>j{p#&wLh>zoI@h>4L&g@810`syh9&Vx2xay1OUpw$Fil@Q*;e2l@bUHQ z6XFkKmupngHo8o|@9f3hd@|yMGEd3PWRRMRwZDIK>8i>h2&BHgp5LPXrv6#|Y!?7^ zR`E8f?4=|K#zB%~4!?Q-o<8?y9xhLODSu*n)N|uCQk%o=eQypVI~J92M}AFjXsFnZ z=SgUOb0lBlDPh`M5@u53C0?K}B5s7QC)}l6_iptkack#9#Z0HV+oUOlPXTlE5K+#p zJYP6GYMhgBOU_X>J4_n10#A98T}T%tp)66ICG*?sRX>Z*eG6(UQkTXprq9;o7U&~Q z{NM`rvTJ=rJc`M-l-I?g?;iZNpI6y`P6BCI;;f@U0n^g`EhtP9FROW1QV^HPmG~PR z2xBV^&@}}|f;=fAvnM0SBF#0|$Y^X$SLp5(1;~6mS1stcTH&KrTpojgDt&9M>GNwB zG~WaJc6e!4NkG#AI-6^A#3j;FM6#eCbPJjMS;2UChlEF#elDjIL+F;}(Xc`lL^@TE zwqI4{AWU3+Fkp=r+XH?E_x-jQJ5Uikv8M#Jf5^vrZue`|NIwJWm>THx@JZI$K}|c{ zKo=zncURhn-Hk>%gWW?RI~E#FkW4>PMVVjH?d6yom1(q>2Rkz!_I%Q8C6wdb`SV!N z49)`L$gcD>gBcb*fhQSakhJsJ^ur|{RRKF;&|JOJPcLWi-4-C--7m`B8iBo{zrO;g z|E55FVf$)#YDQ@Qv)LtY4zg>Z7rWuGL&*V|Ti%W9Xs%1&rGABZ@J|aRLwTvFDr=VUoO<4z z!M>m_IE+G>=6ewIJ;Js_X2H5 zZqH7o@`>cp{7%^er{sLjRsm&yZY@*U>x#&X@UGK3$!C0Tm_t_9c76GUiZ=`An|JSu z>9>QU@sUy@$;lpIvck?czH>=s0fl6qvG6N(oWY#>rE$wYuKT>EQ4dA3hFyn|Vaq zp;w-@rEt0p_mx7WW!pYO<)h8a(bPZGz(OuktVusO2NhLWj{E&6`Rk7SzqX8n3B|qD z8PM6wu$^}L;YH7_b?w{9&uf<81I2;=f^4B<-wkoAe3%B?jT=h9j^g4#(5+#PtGCul z(5be8r@Qf)8NAV+GY`(nu$fYQO^~?WWNJ!hsAZk-R@wQS*nvVy$CD)E-5p@@V0H%* zLsqi7>YkvHc_=s*Y3v&(q<-=fFtLddhCZ{{d%m%F2=Z!cYN+G^ih3Ge`J0 zK@7U6!Z!2u4ZChCGW6BKZo~w|OVE$3u5cWD#U4psHQ=dtQh!eG{m(38iKD zfy;pYR;t_62@7uxx2@O)5XhF7azRrOr$=MU9|xLo47$}ewsk}C{k(WP744Sp#>$};sj^fBGq}bpY*tX1ECAFyUdG- zHnITK1I*!#vDjk9u(FNo=5vPT6a`9kPeH8%u6dx@c z_$};D8#kgHoE!S5va+X$7Kz~@AtC;Tt`LN@*ObDWH*Z$v(MhYh%;HM;Kvp#Ywl5P~ zXF|*4%1Hc5|*M$;Gj98yEF)6nyjnHYjMWx#pup5|)CJw*f;KR{{#3QuJvy_28I;Jy= z=qs=qOv}j(sl^!RCO+O;z$4RSe2=uaHDjZ8*Fhc9c6#41q zk5~EKcD+3HYARjm5_r!)9ks3m02BQdLvBClt!l3dM-6@ft<(1(3<=r^Yj>?b9+~S| zeMjZE0M|k2{1`(g7hQJDJ&kl((5YAnVDeg0ruBjz8DngAYxv37fR5A&L!TkN4d*t; zRl0PqrpJ2SBX=;CJS7Wh()bioU}x_UdzICJI4q29kWMD0AEM5+il= z(8muu#z24r+ena39^bR9fim0CA?7xKO^$;c#$t{2e0G_I_Pi~*jlG%fJ+tiaAm?gl z#|K6w(jREj<@@gH_*AV%jAqJDB5Ze2K;jY$iydMfOf<1E*@%d<_*&@E??tL|s|lN! z#M##GA6V^WzcVtj3OFQv5;$yO%XkTr?~r_oMM#BVZ!OGnzeTTCPO!WMU+v9wMFN=4 z>O+K)U7ca$4^t_!R=P?oXNuIeLZ0QoL*R5dnUKwt_>P&I@MDLOZIRq5+x^`?CjLb= z=c)QlZx6nT zF|HD#@rP0n=78D7?~tH7_;}T6Cm+>pv^^2a3fuBMKJ{VPJ`za%51B0NZnx=_?W^zY zPj&a3!{qiR@0|i>LN$@)=IumS+FN?_;tG}n6}~dm$0!{6XOxU&WT%Jr`|a=Z zs=EPlK}%e_LGV_db{SbPx4LFj>sx<+wmhxyiyYFRii4itjE8|=8^|uujdr9#ttQtS z;nMw8WlW6d!?1RbZHd*_#ddv4m7`9{(P(X8Js`AztXvW(GlEhl&JnCS-pR$$Hu*=&qzoI z59yITKpSJ@t=@!$h{(XgJda9=jji4r5~>7RvmF;dJ8Nm730y^H?L6(A@k83%Q4Zyr zgTr%99ev9cI{j(`0_D3qMX0w4iTOiB;1!5-pee#UAGW1ZFCl^H-6zkb+F?8Ukz&QC zrlzCMJ4)8;hz4avFOrhjtUUU(JMu_*iFdzI(=XPJ6$NzB3Bj+)K!*CcZtYE@A#&N@ zpgp&cMoo~Oz{ADN(%uF_wu3+IUrV}LTqy-plY+~x%*8;l`#vh;K+vl{H_+Od<-}0c zVD8%)H_6lIg;TvIX@s|48$cUo^)BlV2RqGdtt}j;Q4E|y*^_E_=DCSxi;wj)hwJ9EYqL%V+hk^MT-~}u*W@vvSZ7y*6QPP%P-q;_}`lF;H!Ei{CnQpt^*d27O*rEM7 zM%cLyb3te;&ako>9mds`P~$oM16aknNpID$pq*lcZEt92>#U3NNl6(2_$zLO;^g_( zLi0AS?;^*Yb9w}f4Z+~H%JXtRE=LwamwkUpl#^%pct=28V?$$FSb=JmX!1Jl%`wzE zK|M?KFid?i^?Og?L5in8@^hIaP&^-3;*W-X{`}eOTWh$h7C0WeKdUk!ExvyHHaNNO z*q9i}OYaljg{Ycgyl$Qa_2IXR+sE0i#Q{6v0rXU+rfuJ%yys@bJ_yT+Ml$d48lsl? z0Em^c0l?8J2oeM6a+FjU@K9+5z`KtCz*K<+)> z2T)5_2D$e{EIweyD-ynhw_x1YwMDI8qW1GLFhUIjP|MdkJ^dl1@nl)sFV(VDu2SfV z-O7dg5K=?Kh?5U-U02k3HpB*?4PfQ!ti-j+V8EcF-WVCFa8h<~$d3aCf?{KiLB;mN z{T6CI$3e3E(`m~qZy5V?&Vh~H$6FlwgHbSbFHAjsmKqk8X_N3-@q+)s7q1TX8=Jd# zLOvEurx9A8-JipDy5z&E=V<{v*BYCxgNOi=k?RKK4xc9p07t1SY;OmtVfB6t_}sjF zwBUfbeVc!#NBtXQc_jK~@qH>P+dieCs7M{D#HU$lK9)zYjJH^&O7b}M&0y#Dmw z6tK=Ny1NOo=<=qX%ICHH@%`&*Yt>KXf`*bO&YnGYBJz__u{%_Vx=*Gs#12KtRPgb zO(P6nyO8wd*@8uRw_nr&?!~!yF)67rgy(0jPTzws252gS#bA+oS3#Jm(Cr$BIpS7_ z03SI8N?Nf=bH-lEeL9w9C_Xoq5(rojPNzL<~w5iVZemdiR$3Zh6I0gu6ZB-vryP(BVv&K z7Re8W4;8Mtqv+1fZ5pPj#;%ma}|(YL!P$LbU% zZs5w!!55=0NqLEkqfPD|xc}y>tjRW<1{=oW^8_!jFtFJ1MTJ^pJ6iQd+*5Kxb4k)M zY(y|f8_omRPDN6w?Q|9JdK!iJDB_z~=xC%Z+9Fgxr=-O9>_fP_aS5Iy@X<~(W^b#J zGSvd$DQCF>nDyz?rz1*Ou{>pLNB38)RkZ=JD5ZsuA3qA+ud!@7*A;1`&z-z*NiEwh zD{b=*JxX%;l|dO+tlP8->>qxcWn+CsBi3`BsQr@(SRAm3+Ps`m(_Bb?W&pyK(=}tf z?Qro&MPLT_=+Pto1_`YszM76p1j&5@R9#-<{ewT!k?Du%Q9VQ_x2*UDb3n7&{tc4r zMO@V=QKA?39(zggS^HawmEc+iX)}_G#m7J?;-*!w)i*}~Mr35Xu@9yA*chnndxXLU zNsu3_hT&Gk?U9Dg`ms9S_EGaRN>Gla$m6F!HIdc|NFA^Q{H6<@S3L#_JgQq+`k{b< zQ8Px~9OK5nzclBBJ6U6g@>B@M1<)`Kv|?QrdJB3I0h}3l&>k}In=qByd00H}MgBpSJ(j$!cmLAlQk7no4pa{$>rj5F&bbhs;*R)81089#yg$wyyfxuaa8V? zy!s`eZBgaBI`wQHBC$AHr3)$$n?Ax@`WvZP{H5Nu8%x7n6X`LY00I2z+M|2-esFX2 zRr5_%hB;B)N9w%NjZH~E0x$S1m-Y!J!v?TnzENUwPP2`N^vtCbVdY5<6AjnJ0>NnV zjQgKO9q!iN2TN`XmTRr2emp@SOXg8u3TG^i9jK~cn>-;>f%r7jTWB7aRs_7!k5Nzz zW?|K)3B7TNlAf;D?X$8LxOr4%{TaXX>so{T=VpzLiCAIWs15PO$B8;A0XSWEp>xOI z7;n@eWI`}0m;Pp7U1mH7%r#s%T$y8FHiUOuKG$bp?F8~nWk9}v5b7ZHz>ap~b zrIQD^#&TzhZXWl98Uqu)z7(X**&B5pAM^e8Z??l(>RFT+Ia*$pE%yi@P=3_T-U~_T z>avRe`7+UMXN&4dfn&+n_!;3bgJoJjieK5XzhA-`HeZ{=F-;aRBPB|0D}v%2pr9-? zYq|1#B9LP>$UGEbMc-mn``WA}q7QrN&wvo}ghId}FT6w0IZv8UJcp9en>qLTt=P?z z=l5;`LyibZ&ke;q1JShmZ+<#zrvzw}so9b+0&_6==g*%zYA&elLrHNZ(t9?=0+pZd zX7N?P2-2(AO5`YEsLYlFv^5+yQ7)$I_E59o4QAJv@p0pdG+8FMR!}dt{)k^HttX!2 zc-!(Bc83EpR&6KE^y^Q0i0JdR^|hZgpW!oEtXQND&EEWil@)aY zpFyEkz$wo*!*0MWj~a(Pa@#Wyu1%0o<%}+|YK!DU004Pw??t;3R~Pd04X=t^0uR6Z?zfgC}r}<45V($8nk>9WWqGBB}8oP6)y!1Lm4FRW5(2Swuw%0djREWI_6ZV8HfW=s_R)@;ObO7A2*GRt=#WU! z&3Knzp8?C9J2I!UGE(s@nHR88Di*45B#OhU|8SLXxJ_-D=0whA-9E&?V|sFDeUoy| z5es*m@4fDpyiM$@%n1NI#|Ix+ByTVY?_HgkI9n>cv}SctWL6uDdmdce{cgM&Cu0jF z8*~|dBwokBrS<3gY7Fvj-Cb+%k1=pb+u0fwUTrH3@2>ntqrqQ%|5+79f6%W-GVfzx zCwNN;0hiQf6TZy)D)0a(w?nF?r#>i%!bKiTsZ=LUJN(f$@Kw2S=964uz*NNIH1%*I zte)fKvu>se=Llx#uKsiZSHe1VJamAB6An}Wo>qn@Xkyh9R&!6e@6vbK+CtP5c+n8F z!BDB)#D|l;4=msOT$MpB`Nk~0Y}c&PQSCYiC6$-O*ZE|E{3VHNukPJrhb)cgC?9pd z?7z&LV1L@}58R0i=lLz?o%4mTHl0mgrSMbX1*T(RpXI25B+FI({v1avuB^4guxh#n z2(Q8$G?=zDU2-3B@weYVxfJ?-Ys;Nz`a^H96rOQE2GEoA?psH9x=$<7YdBV- zE`XtgcO&QK7QJq7Bp~dRliGB1A|{)Hhso6Rt%^X57gy@d0NpjXF`%|A9zSCM-_2nyhL^Ocjc)1+$93T&4@_K8taB-&k>lj ztn8wx)TKF>o0;)oloqLrxo~~YZg}ZV&IhVOTA+f`i_mS#;LALUj9RYI33VHad!jj2 z7}FPTIJemzRA}17TLrLdfvKB~l4uc z6?Id;wv~_%XaiIUi=@{hV`(YstOrOR+18KDF-oYD9d{B7Ofgc!k9Y-o_YQsdpLmdY zrn9b)jt$)dlqzouLba3~13+w!mV&*!#DCZ+gPup@>!5@^9-~6k*_OWZup>@%(n&Sf zV2t>|bGly*2j4!&^0;`*0yMp(--yG-On|1W6AsdysVodzcwgULp?Mr>+OeKwM5-TC zZ*7p1CaSn>EuV$;9vkpe`9_Q#F58#`MLlAq4|lhVdSQ16$lax4diYwwL^K>7STkjR zqUk@UhX0yTve6%jzT%$nmSl46uOuQQr*3;aw)(rf&eX=dLVg(7fnDfV){Ldz6KTmt z$ppr}Y^9>~_xndAfSHa^&P%fpo`focIkFhjmg_w%@U06Q7QReQ zKA$E{nlQq~)T+)Kl{iH~Hy0=F^U-p83uq@OPClTSE~fgtSM40VD1Bxw;kMa;J{Zg` z0E>l7OSftEFg;{sCu~dtma2~yPxUCdagL<`lCBazC-rz4=#DwUa50XX@rEOW!GFEH z|C~Mg_MQYKNzFxzKXU{SFHYQ{Ts!N?Z7N~4crlL6^~K4Kp4qpQ{Lgpf0p*I*KUHMkq0UYs5A#s7gQm8m{D-~cZGy{VwUdU0wxrXZ(QuQA!X?DIqg~0U2*j~@o8yQD%k2e*6CI&k(loPC{ z*L4F=IfL2IM;?74Pn+GB9Q!>V@qQ=MuFy0i`fm`XjZj@+85D7fXJ0cYB(n!ulwobY zdR6`dW%W>%&JDDu!#%7=0f?oklRUQ5wsZ_`L!ez%RqvGRDh0ICGL#cMRv#fcD_jAZ z-VjX4GS+-%rm9B7=h?l zI#=X8UXMlcdgfS%w_-tf3`aILUL7d`9NuoOTY2%o@GZPQ*X~n`rbW4_q_FokQP)kl z`2O?jMmb+-!mEy>?27HttuzUV60mNq3wyf2axBbYKoqB&tuvS>Ji%_OkC~#Epe2;! zA(4nr0GuXKx{+HxW zfB%l>W5Hby!qH88{~3qap7;XYSlIi)3TPy{wums_>3fcu8MmnP__4plax>kMck(@o z5!~9Ch{qbzWwwOc-=E#1C3XAr=hIvsMjFhI0()GIG@R1Y_=*tHJe%#J|Lvsvr`*Tv z7^n&_PG}m_{P`9KBML@gyK6vJc^@o=a{*`#*(IE7BzumrQHc@2eRh5g_NP8S`&v%Y^UfydW|@7FAoWfzd22AO zFZ3y9k*e3}Hy$I6*52zvA`C>GI6cYy=1N@R0f&G2d$Wa_`iXTlXsU}gTmp9oAkuz7BikZh zyYSL4KWDsBzDYs&i1Mp{y@L(r3ly`K1ZTfR2DBDVXIFoKEvY1k$v*Rz9+q?lY%ztE zpWjBvS6-b|1gxeJGK|2`0vszw;TDrkgM@fhQo>_%XT;rS@>qmkNuj6v(#WfG3^V%c zUO-H*&|yV=sobAwUog${8s|)TJEZ*U=;lZ#tai`j!{Sg)6JePGV>gugajN-aqhTP{ z;wOs$6PTg#n-8fBmRkkINx{Dt1=;@fUH|hD@mIyF;0M-1AnEyE^6h5cz6`=f7TIud zH%E{ljyzYVy~w}F1LRiC9l=RGh}?TYlOny<9JYFY4;!Bx$^+GzBPgi5VFtfEPOdJx z-qrM>B_AzcK0z?OeS38z$j=-|8;;!09JNHJzw%T9U=~LRLh2lBImGkkm8ZxxujK-E zzy2RF0uhDg%2im2y`N_Wi(x@k2&T$@PtT2rK#$F^?3l7ooakald?R$6VxT2b7@=#v zt=^>i$=}4=u>TH(?d0UeF*>fkXfNr|z4McI} zshm0FaTJYznH4+?m={a9saOlK%nllI!Ug-<>Fw0@H_;vyg)Q$jGB(W5+3%LQj(X_N zRcU-%EnLUoyfZpt1QlwxY>J)dyHz(>Uw^70HKdvk^Eyt_pcO!l{R1{^Mts}nIQ#IE zr*}r|up(d|jFL~*tjve_7CHfIp^9xc=+MGMZTD>Jem%}ZsH-y}x(q6Hy~U|jRri}~ z6g?2Mh;;kq+j0hM4bY)~@%{Q=;pO=WfB&h-i<-zYG=3SW(;e~3V_vo)Ye3`-Gg0PJ zD;FbVHPz<^s=q<-WYQ==w<>@fAZ$J385bx;S0_k~a%fyJNE?{yj7qj?nZ)e-iwh(M z7K}26@t&_%r>BCbSe18oh&Dc;uic(dBs0b;_+5u~^HjXsQt#busoi`GjJLPmpu(}$ z!c@M+Trj&z`7vmTyc43O@ml}kjOc{rwplmN(#n@l9wcQb-cK1~nJoV73DwD6Nitat zmVX~&Uwergob;bS2An9){p?1Bv{}Ol!6%N_ZOl7c9qGgda0No0WM^bkDv(y|Fwy7P z$rBfAn^K${$uAp@Zha2~QCr)sAVc|H4?BJKu4kAmg2=&3r@|(F?HaVO$ohm4IwhH8 zGw9_*rH0cJoE|-kXn9>009;&KP?3uiM8R~Zi~Yf89e71Jvw@hJLWkVoud0AOLSm4#y^ab zvZ&d$6nJM?dY{V2@Z#zK0B*8NJVh=J7NID=%c{PpQ25yJdGq6tE08D@^0T8hIA zz(>*eVfx>|oq*uYD+G@C^TS-N|9?MyT;t5^;9%r$c43|eZ(oaWhtNx=V7A!O2KTbR1UWgwBEI(_QYueZJaI*gj%fo^lp za^~;Q4wYpgpIWa_X`fwz_PXo)06u_9u)9nRq= zoxYy%Zr8U!>epl3AE_pr%n^3D<_XZGd=@R~mUoj#w0QISbqL6iJfFR$N`Q=3R{#ND zS!((lpBe_e0^>!U9stqmo0yKc>-)`&*U<~_(MAO3sMf0{6BDaxR44Yd=iF>!B7FTX%H-=jFG)DCzK zTg_+p`Q}gZEk9OEJEm&o>h(bl9K|!#;tPE@fP`h}Xq8)4Bkg&r2 ztPYIKacXB>w(_gs$wD)HkB!l5(ibY zWY|;Xl(Zh+J9Y2={SeB-egL@4fA@^v-Yy=$H1>RKvAsH4nBa@ z*=Gug369gPaNG<3&-hvxA$@7lrz)ENS*jqnZnsOa`)TF% zn<_7;ZXM>VI~H|_&iKgR`N56yu#SY=N}RIP$}vIffsoY`=5tdG-J&@O__3rLYDpiNPFwBDBrGYSPAKF5m4z80g(N7<%ZT`7ZDKyPxm*j^q93y?=jm9K+0cUFY6w zuf6u#-v!U|V#cf22k62Yj!)aGQY8RAfk5e=8nDSqkGcVuED2#MOs}C2Ho{V3%XC<- zmeHS6P2|#h&k|Nif%XtwYzaLf?sOvlC%*ys-ApEEWu6BstIUd!FO=)fpXUJ}j|oB%i7@d-Eqah~b@bTW)VS>ksV>2X zlD3OL;~kEx!5@-=IvSSFuKD{AbOCIZ=b|IqZag1gfUvmPz|vTWfKgl6t395Mn_QJg zZ2I}w%&O^sCYQG-3b)swO5=2Qr@^Q-HPdI3hRv?}k4Gb4Nv+b8Q7e-3sLnD3yk~gZ zuGZp#EXe9RTh*zOdgy(z8jgNbgzop4cT9>`{O%Uzdb*|Ftfu_=XDTxuu%_MsbF4l9 z`;l?LNt9T>OJMr_k=z_mu`OUFr3*X%^lL=UcNiZ?z8DYnm{F{H+&N<$phES+D8E-) zj}B$j2znO0)|`+|iqC`dIXN(v$H8cOe>vbDij02pee|i}TrD4W%%AtV3@8E9E;=${ z)l3oDLD}@}u!LJ+^M$%^0M^y3V~PLr&HT^XOp5mR4LyV$G$As;*Gq*vR#+K;9m~& z%>r)7MFWm@PBjoYa;%(Iph#COGga8QM}^0b!hsS zh5AkLdmdnATYmf69H>?|*!7H72t)_lw=}3H_K(F?6PGoYi%GoxqIO9q;`vg$Qb#3o z<~}6%qLzk#E6;6r3)yi$Evlfj&7BpH`B=b;*wKitjP$0HL=jPA>=X6zSE?-FX7+5c z)l1XsJ5u@U<}l6?n7#V`YPa&0_kz$IqYl$cT=Gqdz<(aWe;r-MFb+Nv zFX1dSyD5ETl0Dv-^RJ>~%bcO0Y%hmchJ-aZ^B9#S7r5HPStUk#exUkQ(l($&5Ozk2 zFHWFhX-qFdod&=hmlaihXGf_sChE3d!k!^{{>NNPrx#P!{by2lun?qm_(T7%lvf{b zJd0fkWgXazsW|nu3eXcAw^Hbm+K!HxVW+On$nLk)G{( zSJ9~J%UU%_Xp8t0Zg$<>_tmF;*sI_TI| zC4VP9U!R)hgou*;imkW?dP^_hD+E=`GV%`MJ1$HURO&nCHfSK_O4BXX*T%aW;_3L^ z0=TuUi`3a_7M}03d;MYx+MTJKxWUBRYSS*Ynobw~tGh>ZU!_3=o6DnTG)>pe27PwmnNKT*TMDTN#_k2x8qzBdETB ztp;mO>%LB>x?Wwl8w&8fQBVAp0vBHq^#Z)0w!iMrLuFkZ*>vm6wwXunFBs|W?|gQ+ z#((EKnRl0<6FUahU2{s_GI4j$kQ|kINXcSA&w-_k#4JJI$nZ_n8E|)F8RcR$j7|4% zfG7G&g{_l_{-qS=I`=;edC7qUL00@fU(^2?%zVJb-*d&s1VBQK!D)%!j9px)Gj>^g zlJ6Q0!P^q8;v)Ig?qh1jArl)GV3Rj=Lh|Nm24MD0mF<|$rvALxD*6PZr88ZwEoS^_ zo+Tm6bgU)3D!BqOvPuZwwKr@A0h2t%@RQNfr>3*rxR=y9LIk^uiYAx7QXT%s#FT>* zg-UXMfVZL|LMY;uj90T#q3SphWHf*@az+PDX`MeTSa?kvt#rAnV!1+v)(dwv#|l~4 z8F^)v10o8N_VVbWQA7|g$OPS7TK86&j>AN`T4Tr+S3zF{;eNk+IYqU#~3q1Et_ha93%gmg$?m6-CuI~RH z_5Ax{5vM=k%*3+%uxsAugCybdVpWI>{Y*&zBzwb#_K?sel$#%u#hQT2zX0^+1Tl(r5ws@E z^+%?RI)PL3T|p&mLQN^`A`tuHpMpc>P6%*9X=YCD$MeYcJODl^jT(miI5CeA-{e}M z+p?WM-%pi3oJimtcn*5QmInkcaHesCsSKA#`p-lGPN;n5`{uT@5z*5tT8G~hv_s2R z^0)(Ft$`*0Fiu?C&WsPI37#3PeUNMmlNRC%UUnH3=ObVbb93fzy1%=wn-H{L_bwFB#Zt7`H8lWHB0Z+Ako_59b{^ z6rhez?pczx%k?ycF+bh^9VEc-Tj8mqdK!R7#Zxd)W0BW|j-_$%zR4_1MV>XT z4;N6oY@wX2G#~s#ni*_&ne4!TG6IN;@*{ZKm=Y3+ ze9tK8M0}G0zcSgvfQz6*-~Mm3tO>B)$r6fP5kP9dbOWr8ljqnw=jq5d4ud1fSk*KT z_KQ|stzHlnOA0X^CQ4Sb`K9_p>dgts+fy>AQrh$}2Vx{+Vj-?jK4;%&7Fdp1thIKB z$O^!wgoofjI-jxoHn$&nS9wZEDHCol))e|m9!BrYcVxATc!DZ4s`mC9aLDA};nUgR z!~Qw0WWfL*Enq}W6-$OPA4uF@PY_v-fO3K{trxb&AZ*_Sx&=tZZl9Z*et$eYg4$bD z`dj&mm(`5e;m306@p&u0?5ngzjMSPVR38g}#9st9@aea_))doqlHwILJloOIyfL}F z5jW-bs|N{%ubl$Z#8#(YNMN>s+MmGZ^`!ZNq!@~en8%0)Wf+iGq4SF#%PPg8!S%** zS7UO}@k_vk@l#&!2D&n7V7|Kv-Qw#XPot}>jLNLkFKs7EDEtD}FrsKAGgZIrv!=aP zFQw#WQ`-FX#>~|6|8Pi=`R4*Z`3K)d5l_pR93*87O+^ zR)Jtz=XXmu?gLmeofWRxF~F{M(4aTKc0_qZFeWZ&tv8~)wOQi+1F*RQUkI4u>EgQ? zS!{v+kQ6I=nV2c#_?xqyHpDXE#>r;!2h?4kz0;P}GyET3%>U#<95EcuKQf;Q@?^q^ zG=W8v);E8u$8paAFr(6+Wx#w04CzZ#pM{KbF)7|m3V)jN+TL1WAfG&2)g~@Ifh{8U z`_&}Iazod=UIy{K;>wem?gK08ngXnn)RgU1u1>ejP1skesqMghW3%7~7*6>>3U%3C zDy5n#F-W>1c*53XMaV0<@)dzU$jm~y#SCmdn_Hb9O0vDa%9uiSIFt}qqR4cwug{UJ zH@`m2eWP)`gi&m3HEBS>Mh6}D{$2-iw0%e5AT!lT;^}o$S-N=icQLer)Ph0>|*-p3`-#?dqrhvWBs$8hUkMe2zM@50-z&()Fsx8DkZxpXD1)$PK% zlTjS#c6*zT?Le1%^Vyz4c*mVcVnY(E%JUnDZLM+luaLNA z-lGIIqx34X&T~3ZtFu%AjM5W;qsvhsV*!olD}>1I!rG+nZcdYuiP2KTEO41gk?&T! z!?wGy=(AlABV+bQbfbW|+jb_SkNdbmO%m%R!#~DE>_7q&?CWN|ed-AGN}Aaq$ZAC~#3AvG6j3j?89{eGZgT0vCG%RPthrvG~DKog7hd=ih?mYCT3WpW%iRAjax^ z@4WV@MTM4tDR3>2@pTtgCw^|GJ{_HFQwqO)@S6wmk8Tn-x?#6c zyx+}%@Hp|e*bn)P|6S(~gZI|6;=_sxtgja#kMKd^trq_kEBn6+e{AWe^_$kQ4?J>D z*1I)O_1p&Um%Z;L{IDfOg9NMuoeBXu@%#O3W!p!OS+Kq8_+w5`EE#N0xV6iyQFAa3 zAOU}Um(gbpPdKQ^5LJSkf@K(Hg0N~wR||MTg#^6Nbkp z^)Uy3Py6QEkgdLu?+4;Qz9ZQjYJ#sjAoDgYB=UjL<(uR~gXf9QXAY3X&l+x^2&hd1Y0T_fNg z6AQv-s{Nq`o%zOdysd0Kkt%#7{NICuFRW;a$f4qPK?AXvLae@|vv(^8Ey#k??jM5w z_Lz#BdZDCz{yEs>Y!l4Hn2K!c0O9ruG7nMQu_vF3qUz#@L3s0SXcpIY0CJjvR8Nrx%`i#-M`u{os7Wep|E2^`2T56{8ul> z-UBVd9!0xxZI0%F7Nk|IpLR2|_jREox!? zTHyJ!Zi1{&1;AS3zZ_-J+oR{-WAo=l2!Nq~s;B{n?Y~#>YIN%S%xE-IR1FwLNpO*`0RCm8>u>1bSM@z-~?2m42*EjP`}@UE_0>5ycX4%eEvl=#fe(#T>SDY%*;zRwkB*T9=BDi*^?$jo7ka9wi zB9%{qt?Si%J7hTNu;QqgDMmA#As}17%(%ZKEZ&84b2ejKd){+4&Ulr|XE{=Y`NL&G zEUeMxXwBlv?3z$FXUsUrACLcXbDr^Z-$M?n2|)N_`~AlsmoDCmkqX_fWA6Nc68X6u zRwYIPdj7FF?V;NBc~jQ(Wfk8;C;iXanfJ#V)~+X;#`=6o+?h)7EDpw)v6#Bcn(|ZW zdCJK`OR8Zt=~;!XZ8f&^5aqOhN>kc;*$TA!|bcmEb#rwLA}^~CaGVyiLT_q zk7Z6{zd0~0%*f*hhV&6Z<=+%*N=gH3*7M&+lxY`!J9_CMKKSSQs7ndD_NN-Ty|U`C z@eG0oR6e23KNkDp$xc;2`d{blc@B^^PQVTUS|Y+3K`Y|AMceGXPs9-`NklH z^S=Q3({3E!i!tw10TjVN@vjT;=Pw#!0S;68`p28jKya1`))X61p4TlLO~jqe@@)=t zKUk0fhJc)JxJcs$aL5YBkC18_>z_xf*SwyHHL9%nrM_hMejc(Mi7=+SN%cn2K1ZTu zB)sc8!HF+`4L$`D^M^hctkJ6zp7wklI}tK2AF1E!p(BE|Htcgk73`uh1}T?Igc4$1qd(`g&5gAB!z>zDfAo()c{kJg}LmJN@AJQu9#e*6d*SD-L|I zS@c6=YrUoM1DKFi%B3e+uaAG=Vq;#hnBiOg4qPh0jq3MIZlM_5I&m6S z%1W&!E3hIHrB_XeJKw+h<*pHgORH<%-vxV|K)=NIP#id9e7X5n8N|}~&*aRYIXSP8 zAGzLBupLO{d}vv~L*j*wqvhS-WfpRCb)s|GPeyX>V|ab3-98+}NL>^JT_nAWj`&}( z+mtyF3L2~afC_P}D&Zs6cHzd5c8U%FWiMn3{_>!x#;{M~ zV>f%~(#;g4NRbtHrd)QYF<@Dm(6wWDgPn_WF&J?|;9vSL0C6Pq^gY~q-?BuZRF}H5 zLLX{&v+%_Dto~`>!%r>kvhV^(!jwwSvE5L;q1E2SVCs02^Sa#!l0UUxr9l6HV*cFs zl08GLISYW2^m;h($=>HH`S@OzeH=6gN1ybpX5+kE*R~$cUG}}xd)N7_q3TOIkAwYo zq0Vp#Nueu!S*6JM1W(toWn#eIOgMeb@3)u-26D>xVd-M??>N>kjyZZ+5dp(=Lwx7cC_rX zA=TV`Z*9C8#SN{ta`Fgs%U;n;h`CP^?@Tv|LlsVnKGOZ%VjYjUZ}c}@Vtw4irTicR zNnB^~{2ZhGPrIgfpc{2Z`NX_&ieZY>C$t8c6!Pne3uPvfz_sr7j5Hh_bW^)QH75<^ z@rgH%z>5xHr}polWein8AVuS(NHxMH5WT$@z-j)ziFKdL&p6ZWfpgTEr*t2Wz*6!Tk8Y>w6) zdz{-=f(yPj8RYW1VYBkAV=fzCSEmcb0un5aWXVe3Ok{;`R<`TbyKaGxd--Kf@v7%O z2;xREm?Il*1GxnbzK(@_R+V(Hj4XZ_R4TN(W>NOPl2M`qNd(T2&}zUL`N z3Q8PFfS-tx27gU!_dZKKHLR6?LzVM~lMI^yKyuv!iRRHP4@wRFIWzd@FPJ+o-`yM| zN&%GYXh{md3EZzqVi6Z3zJ+Vo6$l0L!ap<63GDa6^3Wk5o_R!BP$(RCH z`V@A!RtiZKr%Z)WTL~sky@(|fM}}>LM^*gq3FHbTlA4Et6B3VVUTZ^bPnWOg7>{P$ zf3NQ=0~u6H>nW*LJ#Mo(!y0d^0N2?qY~?&uL^#UFZAf_M*x}_UOMtGfkGf^A4%efc zC7<1Xt>4H@qU26isM;j#+TgJO5AGQyx6n}Rx~#5Qxc4QL&HxQIU5-i>zCWkih|_E2 zJ^bk7>e^0v+7FhB@!lB)For#X@8&GbK@sZ0Hp`ACs?tZ z?Ycb7>klUkAT5b9k8q0*^o=S}} zfX*<_Pg3YJ^v(K@^>JftWE2dKKLUl0q`_#dde49n!l(#Y{zk({sAIq7miLbjMpJ3jlp7ELq2j^QA;1()e_WWqZV6v1q^-E`p-t-YQM z50N#Vg8W~i5oWS@^Bi8+PNVkYrKmJkj~Bt+KT;Y885Sm6PG@lCXo}i%fB`4uuv?#c$A=9O+p&vjF?AJ{MS8#}^VwBPnro{3*tkuy#!t{ImR5B~;?!aXKBQ}P zl4$38G|p^u=1ZpM@g@~wT;to5ycv?Ke0SH&F6@|LG&%a+_~UTg?AGupbx7WYL`>u{ zp2dS5F*f8dOmPDHld3fD1ErZv3Om}?ctB}T>-*@+;ZbRVNs23fS6n3`j1PTbbyT(0 zQT=F59JkJ>eC_3Y^CjhQ_d1=d(KpL*F^Wr#LDIf_=g%ntf{1LcU=)#^T&+0BgPjp9 zYDXj$WL8;}mm%t5#PRwg>FRe&X{X+ZMgrtOOJ2Jm>vF*z{X$~+Rn;F4?@Ie?EUk<` zB>|nu(Ix{QeHZGnPDd4{FiphC>|`B7*WUS$JrazZcMsw9m7+PwR^6!;>vQY=KJaV_ zG*?rSi8u5Wn1XuqYP@^vVDTDfxpvkqoW2V_R;PvHH+Z{eFcaIna18vUDnEm}?jin(P%*az8yZs3l|`n#uOjnuL0F_}Hyfr1;3&eZc*_nZ$mQ_|e*#_A?$21B|>{icgB0F^_$| z7#h3DZxiKYlp~jAXiXcaE2Ruy#B2q~Hc?TneSS8c7a;druAt1vTsD%&V~_II^8U+f zO?MbizehGXmGbZNr5a=)wVVj|)x3lHn=w@MwTHd~5vAmUtyX|On~y^*S@LU#*~gP5 zCy}V>pX~wP9%d6%0R?P9(i@4;c0>6#OWU+7bynKfMM?~~vitY4w`!*~#=3=M8}*i` z=pR{V5prpRFvZs{9w^VAzM*7XaD|*C3&tqnev;2Y#KC%R%Tlfvqj{{5Fd?*TjP;k& zbG~0hV0BvA(d-FypVm<7O*(MlD9`7ErPC6sHJ3T);(sskN!E zv7yNxeS!DURJ)-PuYy(E9)qkQ`PnD_qbuLxH7&%=7lV60NiNvtCq4clUTX7>%CtK> zzw6}+A}Ul)hr}G8YX9z(54G+st^J&4B-sJM`k`WWVIGrGqH_Jd8a$?91Shm0XHCri zeDMDx)`I4vA3UPbuifTC?)JU$D^eP$Pee6(EbBU_yz!)#$PWly6BglLHfoRj_$$4T z8(#ceIuDH?c#tQ?UHtIXR-d}FmlPM?6|9$+0~WYnXB%LJdU*u>K3S#^@&E-IuRrf` zdu2La8|cZ>#wR06&%YyUGw~xu%&7L%?qAg@6M0w1Li3xG5IXoH=*blF1$1nk1y_|5 zkYsKV23)+svO~Sthab!949!X(IreTPS2Lrm z&#S@0^Y#WEzg67da;i}$gOKQtKZc|3X9CsztHRK9VvNwY@pwIBYT1M04z|AYkd}pm zaUPGz$bCz)?rS&t-AK#X%{fHk-C%NEHJLwt*MCEUVNtZnE&8*zq>7A1HG%)@SyA8^H_-=iJziDvk_?mkfAs;&|o=Rn2W{b z0@sbADf`EtRn?(ThqGW+60Sn=y*~d6ior=#@S72$nO2pW&Gt8*AFn*{Q95;svZ*ia zL(9#aijXOH;`+^w^{sUk;#gMvFk4AY6^=~W4m!{Gb(=DmnqsN}0WVDs`-gn;j-@xo z$_)ha-l@#PqRHRmp>uinxxlIDbo{*CzWe8VE5yK;BVdrX(-%l8?2pb;1`GjkR`9V} z_?|(0wGEBMEh{gduNkzS(xr;{y*9kuwxnoVSQGyQsw4!+ry?)xB?AqKn5C z@z~{5^SfV^DDIstwq@9-)$zxRjL={c9p-v_K>{S5w@@sD7F5%nOI?_D#j^ssQKt-} z+cp(Hi`JXzQxu8f-Acg{w1T3759B99J!x>i+5!+-FN6C8{PeU+KMJJs^LrQX#7x`e zOWb~jF?VUtRh8>NNddB}&;99=9sS1V_-3Nkb=xHaMLW^c?BOiCskjT}s1p6VqMKAW zEysf$rJi@ug+K?fRCA6plA4EbkIq&j{y6Z>>q*+F5A1IA25g+!Q6_>kZ;{K5F+qPu z7OzPYs|0W%N`=bwrR;ia9=C!uw%4rAYw6At4oQ($x82MRg9@3~?)KB7*4IXjMe?na zrP@zE&V$-z{lq{cQgo(cYYnet3E-HtJVTX;!VvhkdS4+9iUo-_batUY7(_+yKls1v`sN8V<)(U9In7`D2ac> zsYVjecq}A2$80FtAw@UekW1f&g(vgC_z{>zwxg5Ubh}x#e8*e4eUBA_D7B~DzZG_& zg0(;MlO$f?t=?_|`lcC^icL~|@t}GY+8L^NNu~MefSX3DN(;!$@%#%h7H>6bNsV)o z$4Kq_#rZSAt)8ZGo}Hf5eYJG442k=-p$DvN{Y7`Lgj#PmeK~YD6^3IB^@5RYJ~M<5 z{Ydqts`Ex?LZ%eXR`1OW=`HtGzh*hY%g<(X#0BCsf~OQu^;4@Yzwq>!_ebRjMR|0U z1mJSb)p;4+RXH#lWmamZ@3hV|+I2XXJ}3%6+B-X^G-v)h_mmLLoL{~9SKc|eivrDQ zeoO}yj?JKaTz93=ug*bMUc2ZIPclC+Jl|ydUu_}B7Shf@K`s@U$+!?5RI`nWt0k6* z@D2_pTiYqEP`c5R*Y7k%!pD!vy5|^ga-X*va{2r@y*#fLu{?7MwUva_n!bKPZhA`j zB2?F zGVsY70%h&e&q#6a#l4As;M*bsg&ZZfw8ErT!Hec2dogK`WW5b)57@9uG$IiAwmQ^S z!BS2I0O9j6Qsqj%GsmOj+3B<>a?t0EIc?%e&f70OLF2@aRWu8R$FW@6h=%Yon$Hsy z`9M!~bquqa#6ij*rQ{Sl-t<_=(Pg4@MFA9kN9#ydz*2sqohTVsP8tr2h}d#TycSy5 z?F#*Igx;>A9xXcBzKCmKWvJ{kP8t?+q+GOD3S?YZg!o9+WQAtNZn8E?>steE%n3-k z)6G#F`+x*pzREKu8ejQy5eORnaoEV)`9f{+>knUh{MR;nZ+B)btF%N0ZW$SZ+LCnD}2PJ=X5%zvlJF8j@gA}*oHlN#^D08Gwo0T7wq@E^&Wlswe_oK){%s% z2KFPZ$qnMO{S_Z|{_~trXgkBFv$X-Akc&hAdNNo@avRZkm15>;Q2v#Y_4Uf{fE8?O zJ<4V4`ZB$ITI%S4OYQ319xve(i|FLqU>w}!r}+)e(GLaiozoNmxe+gY{-{#>msW?L zht>K`XtMkkrqsx1$0=7)1|((0DTCt{oEsiRkP+S^Ov*G9=Cf3=xJhX8z2f#0FUB46pPbLAH>B>chC1ERP>JQDgAUY4EXeTD zqEu}7YjUr0efAy>ddan_A*D&T;U&wu=7&p`%K9N@RGAqWz1rUUeyt;5CUPE9 zFeJxq2EL2i>xTGo$1Ju<@iI+2=fJNA&Bj58r9iZ3$k5Fc*uFm_Ik1m-$Rpiiq!lM zjxIeyw`C*oqCK@?k8Qqhi-GqX&vkoAt(nGtXr@O#pxvrv%Jz?jYLSj20!#pcLx1)h zW_WBbNlG~_>&iPl(Ss|hg&A%y*wVo3Zt8I>=V(9+Qdu-*v$GNLucg_}WLs>z33OoE z9ewBudDo=B845;BOGio@f5Av=E1z1a>hu2 zn;^AzdO<50)UZ`qW>8pwAAMe`9Y8N;;-$pJ%%zPTfD6L=Yx zZ9J$5^aN;5PFB~T9gmOWr@A`{F#MiybOe!L{A|C)+Me8I*zkXUj|VL~D!pC(_@0S6 z$;6FR483yyZO7GU#zI5VziJN2UEY4n#NR`X+UHX2kL`ST*5H?qS}JB5hua-$J+#Hq z-BJh*y|KRWf1c~GbLaUu;~{Dk zq&$o&BdTDvPs&c~?zu(2WX>t|iX6U>?JBLH0~Iq@mh2*+JiHWc zQz?4ODQ5$kEbN`dupX1m=H5)++rli|$8|oVcSUAU4w&tSL3JTTHX~bm8#y*{YwNAw zO&*guJ0WUG_1$PK(Ae?&bgg= zc@(TJ{ax0@lP) z9~rZVxPZgOCjF?PR&xV#K}odIy*E23@zIz}OKo-AeN8aLqID!6kklX^(MzSt*f$pYhP578fe9VOy& zp1w(N=`t)5W#Mf*X-^&Ii#2+*IlR>v>w#Qo6D;YCPk=Ifidqve)*hWzYG-?tUF&dm zIh~OUXvOjpM03~0cc2lPb*Ox|mWqCZFVg-M?PmvVkKWxnZc4s0dB&n)y}ArO{aAL` zrCMZ3Y%>do1B}=(Obfml=GQpHv_Z>ed8k6cqs^|!by?B2dyy!yc++(5uj5njrc5qp zblDm1Bb|r^5bu*WIAOg-z?`WqoT8n~d~q;FXkmld9m((cCF1$qwUtgxqT_C^_=!ED$dc^c+L&}f!5;rLknOemHy_ELnU@;m*1UWZ;Na0W~(gftM3MY(@ zHu6&ItSZI%Ynn4=#D3zix-qZ`9+D-N=K0^jskk$)=-@M4%TnTRri5WSMKote&1$*5 z*%16_k&9ywU0b*4?!K(?1+5zn`&!0W-Q8VIGv|ou_AOD9BHT64(qnyyt@lWgZiIGV zt!1EJ+g1lnjd$#^sOyCg4c0MEo+mta`~YfXryPIxXSbnKpfT@0*GyK5sS6U=)oV9U8UW+Y`W&c>sb$+D=f`c#b^J!2ei ziF*TO78kl1?OvhY8mjx=aUZlyccy}I{kknNyb%dPy4737tk56v+Y&vs_q8li>r`ur z4!T8mB+@2l;Msq8>w;!kOH($-Y&1vEDA&R#2F{qQ5R+Zt$U5YYIjJ!W7Lx>bNHs-{ zTDiF{#52T!GIk5o%v(N2)tq?e51 zx^YUrZ9~t2s2dGAv>bZA*rL+(b20njTgInxzZWYxdnV{v&Ld*Vty3O5N^4P>08U`u z5|zsOIDfuq@x+sU_w}bpiYVAjNP4QsE-suM)=w z%$ZBdcz0V4yXRv>veMdZ8n!*=N8Y8d>d4@4q036XzT&2^Y?bqm)ct(rfw3*eurii6 zP{)+3=JY#OUH?k`BMs*M`_s}D`f7^ED)nxm@J#~0IWOGA@vEv;qRU{~u_RGIToP}s zrQC-bmZ%%M{) zASlYx@66^2x!by_HKa&Sk6xl4yehzQ>Qgej8KP^cvn&i4xwwzm>-}XEDmE^nUlzuia?9{<>W~rb|fm5>{7xgTMaA+iAR`d`S(` z@y_3U#zVxYa;eAh+`-dMEhwb~kv-hE79BW_a(*1ioq1jMb99XNf7DtyF%QaKw^m!F z>=Ws%oTBZ*ZKhijv};B~nPn`>3!U;1r(J!=)iHqNUIjZ5iWp7dC}-eQk_6n1@-|gy zHQFrmxbx@rf(XA%7KX?x&~gEcQ+?R`*AipmC@T*vys7m8{P+R0!%X+Rxx6M*#9H6b zt4|w3z;x@YLKQCsec<`Ow7r-bT*n^rHtKMa$)rrMgg;5-{@7_wqYF0pjhCopKT(n* zhH3lDax`N?G=A%4o^|DL0^)PqA2qI_MSXoa^~`aoiVWCB$Zl?O{CN`3Aba zY645^vb+P5J_8Onp~!XWu`IS;_ue@kk241-PIWRThS*yyLTh^)%eik&#}bY=Yn@BZ z3Nql~kWNx2F3%BvJ3iBtPhh{Dg|0%-qi0+_W$lu!v@3i?0cyin zisXI3z70pBQl2V|_D_Ri*324urm_ncn~#XLq6|rK&6vmq;q4Ovz#b%0-^_Ua!YI?X zmW8#6Xq<(N;RN`epR3-zGP4S>0e8GO+nO7ro;>2^>aZA2OY2GA!oO`{DC=6+IZLE=F7-Sf9vTmDbNST9ZtE7^z}FznH`)e)rA^^ z&vel^6UmGpvlE#PKdq^94P_28Rev1Z+VUa7?9~Uy(-zK0 zry?I!MT-E$ILGB%r{Oq(c06Ahz!vg5d%x&2k+E8XKne_sc0gM3oQRI03SR{LjaD48 z@+jq{Q|e7-9rl_D>%y{L(cXHlaOpEz00YnYbEUVRZ>PQ@n_8O+~ZJ+yCF>0c!aUW`AR*zWhWCh6W#|I`Z@ zzS5(5yHITFnz?_c3znVktm^iM{5Hs{ zSYoEspE*>5DL_fOrhOY)`m8!Vq1j83?ow69*!!QIEixgdfG?^XkL#X-tRf_dGHRym zHW{?p^c@&`7;(QLSa$)mW_>D<_N%@~v#NUhHw^xDB?(;XmIsMe%5>d>N`{W`ojPYK zPDveN&KUNfK7%qaE=jzhat9Gyw}Zgq7nosbs6A_+NO1kRyEUQCUk2;6@F*e*nvfkoR*Fo)S+6g!{&K zK^U*d*s`VA37?nwVd0lm_)B_wNwM9#dgdhPInL@W*PY{Pb|$$rYV4abR#uav#^h zssX8vt2}pIG!y}e(;Bs>1O9sW$6@LRd)w=4H&UGs-&8e%+z}OR$G)hFeYo>cwOFyx z$KfO;=)po8&$NdvW&faX0*mj!MJDg>A}ZJCCVZi~+ic`&XNsyE68J326i0YU3(mo4 ziM3OOU8AV7AW-%Uk7>d9iu>+a+n8~s)ScLzRK#8DtbdTN*P!w&p;FcA^#y90`>#e| zyCF9*K&};wkbW9c<+Nwn{#dcS@eLbqNyKX&JZA5@8@};^q@=!$_{3paJlO#=4O6`) z(-kbbv#jkuoqMC1P_8wWNROo(shqWsYo{ojH!qoYX{s!aT?Q1b3d~AosW_=AxEGKtQbD~`rmMK-Oj~0{ztqq>xJLWaw++b`SHc_GbzeM7JL<(yd}r z&m3}}l<2iZ?sc?3WzohP%F#FGB5#^0kIDQNS+5h~?WT-;NcR7b_7*^KZC%=MLP&6T zPeO2amq2ib0F7G+7Bsj9cXto&?(QBSxYM{poFa_@a--ui25YQCAOu2dIIHhs?C zd!4n`ene=Zq0#bQUgbtn-q5bN^0&01DjgEqe;9{`Hi~tJ*VfU6_5J~1w!)`UhS)-r7UE1 z2pKZDL7L?+nb`eU1;N-&Tjr1e@D@!41~MNY=dDG!J`pKea_Y&ihm&Y~D6uTxZoT|$ zXP{5-%l!5=xw2KY_OfeO|B2fj-8OrQs+&1@ZC6>wdC7EcIOV=4&Q68Y(z)EOf>Tq z!v<+^VAcM3SaQ9j%EFNBdhc9LYtST0E;h!&jcfV3>J0wtF zl0e+9Pv(61-^Ce}!I}vdv?cFtmYA}Z{ zSUE?tw$7}^nFSO>AyZq&UFB!-15XHCa};TU6vP+zN@$hn(P1Tb=zy=Yb5a1Hd2Sg{ zfxE8uX0R{!D;J$&D*{rOimioF5&J+cEHY^HXB&JzF@?HuIJBU~$z2WL@eqg^RQ)xN zG7SIG?~wcNRI^1^gal?*cxOznFstKyN@20xwm`hDAiSW~$Q@3R`G>HJPA1qQqc?)g zcHW~w9&CclYb1;Z$W${naoppbT&Czu?elQm`KqRc;4Sy)yvdh~rBr*$n#?=LDQSj% zD1!P`trE?-C~9UQ8*qG-Sc#k&#r^g9Us?dTY8r9#DD64&enaT}A2Hoc%+W7HkNFBA zKMd?9Q$nUjaIqAl#znu5dx)aW;~!4wZZser#)k@?C?3v0xtCA)?r>$0LbuRm{mf_T zLu=m9BW`J$wZ4bNVh2Hi2GnbgW;>y4{e@1h1NQ++?-qT z(xA1iT98-qV)xd1t!eH(zOricUnKplM;<8Wv1-xskp;C5Na6}?g@|3!z{i*ccX(=W z2#Fu2cabPi;xx$Mx4+ChfBgmN?xUaF@b^EGp2?_tEb_#Uzg8TfuGnsvI+fg2VZqwa zXIL#3>MsvRpSu$4n{8CR)2_6`!}4Z1L(1;P-#k}7+22Hw$gq$tdE!p{E8JY&pbxjZ zMy#(L=Z@h?r(7AN<%7-mmQLJ8snvY^dsVey=j=ul2|j%UH!Dq3k>Bs1pQKqF*^Wk~+Jibb#y~aJBKTx&i2znLva)PrHcK$fCqQGg@U8j7l zq}z9YSF1s#2KTtgWdzj_#-X+g(TMl}9}=;0OSG&U`gCYBJbxBjG+QHAr?&zfEiy_J zG^`8XU##gttOS`22QxcS#XVhMG3w$6s=iSkzc#x5IasX5=wL3uJGQInvQa`GDi2eUfu&1se9?vXf1OyhI-1-#dU4XF?!j~)Ok+S1fM2t(Gm5$CW?g zf{>MB=+qtSESg871yVoTN52|2{HYY|n!!Cr5vO_FTl~&*U-|Ow3*M2;qNQFJcM zOz9f2Q0(_0sJF2jD1(6g3wt>61?##yfFLV0j}KXlDXb5!vF-hlr{praM}jC;Nr`4r zE&bp*e$qDQhBXy=BfdgbFTJ4THD)eE4EBD~)V-Kdyyy#M0e|Mq;w~TmerZ;-qz?Yw zbb*(_f6s>$-!|N{kuBTphg;g~GeiOOfO1s2V*obQLoqtf z;@6<$Y2$5zyP%lN@&{S8FY?8xb!9-%*ru>rHOX~~r4|BgKIdM@e9<)QlW&|V=}1^< zpL^({wYqxj7bCzgT{VvH*WaBH=v5{Ca`W_Bro>ff9ZhXjG@6ARZt6RV=5Ga|XOR)G z4ucoI{m1l|vA&Ad@M0{avm8`mMDjXr$Wgx4pnS*vQkOUYCg_^|NvZh5BB!x`9xw9e zI|ff1B{Rf4-Lh`pg#ai+c6oztDRVWzv1%HAp+@*4n`_?C4*N|@o2;w9;Rq*7mBGmF z(q0hA1;&BbrcT9bZtlVkt|&CM>WpxeUY=<;%`-JCF3@wCjxtt%gMC=n=(#hH%-SBv zXZ#fkk6&(bU};S6UWJKRW)`=;o#riKDT8j>Y}tA?$fWJ+d3z8;?=PD(fzF}Yy!Xt! zbqAj`=ru}0jX@E|q_H1}H&9G+^*rIv>vLFuyC3V+~U1OMsFJ2OjlvdG1T7QO8O ztCHJ5@^RI<@JLsc#)IL543QNm06gH;D>R!8{2uE`b^MLGL1UI(Vy>rTqHugaL;za8%8g z>oPSZXpumAae$c-9S!sV0< zzyp%%TaQB%_&}uW+R!*3xmC;4^h{(J>=G2OYUn|X#*c5QBk9Gcd~Hy73J5~P4|#8Y zdPoa8_+qocwe0!(Rz2pON3FlB!?$G-rMK~w-3Vggz_00iSDRJVj~8)BvWN&s2QbXc zX~BXI43vWGng-s#gZGHm+{#4!_zt%$(#;hJjnpWzo{T;<1HqbO=vYWNh=~r&V9Ee#0>ftoQ1=Dg<>G0>h2+` zW;T_HcH@-akudexNE@5LsP(aOIhTI5`0^vht*t>*ZPcJv8t4ikAu%_6NCN$N#$snX zhRn9`GSzaWbsSm^H}W&PXOVhczT(Waf750c?chFqob|W!xK-qgz$aHyzk6i})d7PV z-3@HZbEyXFR6dG@>BiQ+oj$fhC!67Gf`?NQ@)O6hRF&uU6&<~ORSmNzb<)$w0$>asLdwC|B60dR=eCvUV5Fr~zXvp|89slQE7ZZJz~<7z38 zqqtN+%LHF1Pe@0x0JltZpN+&8NLf7xNoi%_ z`x%~`kS2m#nG^hm7SNA&8BhJpe4g&8T+>hzCME>EW_wwi%{sXSi_Q(T)iUQ&D9Du; z$)Hdmvi3F}2*F9Hnp{r)sXPeh&?E%J+qmQp+@UwLlfyQrRhX@iy38UV3w`uHjk&RJ|o$sQrh2hx4OXn{-|iyHxV{T?*VwD(kMe4O*?G z&Ymp-O6_F-L6&|ikIur@%lNV+u%;qAxszp!KD|l0ouo^!u~*?LdxTsb};Le zG3(XnW4CQ@{SHU%-EWPriRx**kG{P=-ft+PZ!`8Yy$T`W%^i{p{Z8#6QY7f7$YRQf z6ljv`BCm7nw5e?8HTZ2XSARw7EaW`|Uht{R+0=xGvLfJ?y+I>ki1kO{9ZhBbxzQIY z*~dwb5?T!l?oj4$7@^W{dSy`hX`9?;ZkNLK*|Rq*a=a0;;~@wfz9L7ssX?k(+JOeN zv%scp<3*q(&8zU6;~kJd%k7wNTCPT0W?_Ri<}B}2Zg9rs_J`6{H0*xLc6MdiZzS*4 zS>_oIkeB@HEI4(pPB;s%Z)I8xyVex22!Gn?aGQ zzI!{PsnfDs6jpt-nb)DAOP7sjkb#Dv=C6@fBwJ|14=s+6938joiF^g2#01wY;t17p z&N4lc>FyM1zJnk+tyba6jqssQC(FT8%~L3)Y#L?U1zn3KLrl}5*&Fap>f9olAf;@8kAB@NkQgc@ z1OaJ3y4bnA0_#JZEr!q{hV6t``6^Ppx*G`{a-LP$RG+FOG^Zf0~kvE7^Up&#hj!9to1O1d~Bw&R&>Gt-(`vGSNT zxuRXjc8^dLHWQbV^X*L8mF{%TsZgjpt4hDAe-oLXs^qV=y6PB|?eH2FL^DCEu7*9} zj7<(CPH+<~q(=(g2rmhmEKn%XIgpIX`*|fJcG$De(AycAzR?sfBG+s`F?>*YaJS|w zRVNS6U~Aj@DsLg?9*pyP9r)AOX7tMFk!YO*bdiifP-LAIHI@`XmpgHTyOSd z0XpGldI~;n&17Lv)!{-Jxxw9(BCNv~>dSR50FQpTZ=zMC75Q>FW$i_Xiuw^sQ*xMl8sdo?Ik92^@%ZIR9`1$j@{s?8@h|9D=SdCH7 zFj>N=`bY9xw9X$_6GjtvS;=HA-*tQ3Je-auzuI2_ZOAtDTsSPrDxV1eqVqeD%rfrr z6q)=0(3wpw6oPFb{c$q}5e%c(srZ`rB=X-xN{qiG^&rGS?V zJbu&J)sr%L_wz}hNGtK0*6ig{Dk^pe49w|AdKme?dH|`X5=IBdlBG{eq4Cb4SkcOQ z_L>R8JSX#KSm#zEO?pVkFAG##v#wvke#++y%5_!YJ{)e=a;!a!Zy!Gl7mE_%@r)m(X-$L0l+V{OPo4Uig!Ve0#oBK>2b@VQblUB01&s%8 zCE4L_0txQ~!ifccbH0O=Y15UZdsQ9FRJs1bkxpOqG=^Ck?A|IoH~f?twMeGB^w%CG zB{(og7aBlJ2NYuor)Mi%!(6S?rtjX@Aq!C<4|z~5`bKStg(_;Ya9UCI*>B=eyTxs| zaA#CV2hKIld>3)=YPWWf6Kb$3ujr4l_3+>pnli#s|4>&@Iyt!e3SpT95d8WVDC{rz zuN4Ay$xf!nlYg!7?HN>#6_NfTY#~|v?qtxm=r-^|E_Iqql{auHLdUH#8b+pHoXAxHONhcANuEl3<44)qTsRQ*mFl=B&p z>T9By1hZ!}r6T5dI^Ck{O};|I;23cUyU4p&Fcv%q?Y7$WvNYdvJEe2uh4)?Wuha9# z0PVnp68O%usZlMt$HG`eWW>M+=%Kap{vbXV_Ky|-Q_u7LYoh!>L?*kdt$S%HLF(%t zGjhn>!Wpz$yq%#NBs7p@GKgF!O8>=2VEe|?@>9_0+bo^%b$8C~*~Xe4)|0TJsqWFL zQPVk1&_p^7Nh_-PLs494T0XmOYyb~uyJq0wykAn0&1Y;eTWDB>Tuxv1e#OxNvBe!W zfe}#m&{}m3-VobwzyC$W46q+a^a>zsrE(A<8%1%|CPRA{b6o4jDGNtA2SKuPLX zGminlko?rswFEr^6bHmfNiQ|D?$U9Ho(|QUG7IP(4yU zW$FyypW^sa6bZ=g5M}&R1IsKqYd?7#K0F^?9TEVM&$s2$Lh`?Qd z`mmP0fjBT*VU2+$K1vDIKOZSv_pNp68U3!};@;lc!~k%8XB68c`uTOYlbtywGvoEI z$^pHFST(;ZTW1u=)k>9g#ez#8?ezKY!?t2}nv$K=hHn7al$NRTfJ|}yVSX{*%<=eT z!QAztm13A^S?n7blR}p?)z${Bmfi!wnY&xh_5-4R#2*YPTxvsz^BXH&(wOKTx;SgW zXtD^v0~TPqr3VCxtzHr3+`Cm`<|-4jv|)tC0=xS2pQheT4Z~7x`?@qv&e9=n1Maz5 zJZ=YY@(MV>d3LL+qY~7^@5MI^zXGtKE>e8mynjdm^>degn1byD-7iD{0hce6S&b-p z;%p!9fFtPjDCqcX0v+8t;okESdTfJXp3cn8(N?o{`xT?Z!D6B&04_&KpZEW$H?iuW zC&d(9IB1dRxIL?SIZC%FrcyAP5L*?W8q4tANn~G0$ny*e&>DH2_bdo(S9m${>Rdsk z4bVETth~!m*8y{uXdz6Bta_o=dh(6zXKexN4xpGW(jrG$FC17#HRguaDxR!yIrI}> zHPSZ#tkL|Y{SnfRj0)T)s{xB;t9A&`H=2myO2kqdzn1y?(OXeF-g{Rj4&JegaMwWiJg0C}MnoJDGm*ML3`3LDdT8ksFfoRK)KeZ1`+C0a&6)TKztcb7f~*~yiGiJtgjkj!VMWupbT zw#BQ!%5&>u7dMZb46_%Au|aw}WRj?z?@>5NkY0^Dndv}Kr}+jc2_1GD?Cvk4(nsqc z>2QA7W1v?tpC;6C;o8%yp0ASOllO(xlEWrzBcQG_j)rWK5ru-8;=7(+uH&ovq-nGK zX;cN>CRlB1WH{XpvFQ%QK>Eke9E3s!A!dlrn6ZXRyS~^qO5hQK*a5x!O|&UaxJ8)C zF%t?vZ8!cHfqLuuh)q()cU6dz@N!=$?li*wU{)R3?}=-F!NR!$5+Ru0bq(MIbs?&G zXe0GTcDPWupOQ@w;oA4|vEsAF(TIl*MCNXt`wngOL}>}@Y7WXR_O3&lM;mSFixiGG z2n|5=pzRQC58xk{{Rv-?iEq3UIQ|;*^q;WdWqIw+3pC}DProwKpC3S`Bca)X?TR<) zaTKPRS48aEx-c@U>SdTZPRLg&YU|zcyMLGK#&i+#<8Iq!=beg&?ve$G?dAphPK#rX z!=nmbp3Iji8N78RBh>mH4#Tw?`4eT@kKWr`2vaW@vRr^lBpRnQ@pFc-2J=1r$~8ts$|aHR7;+e_kp44Ov{hv zgW8JX#?@W1^RIf(-s7$Neo$Z1S?qqFfsyI(bZ4{Q+><5MTmw-B*XG7elcc%AoriUU zh5MhHT~MQ)=|5`rP8M&5OzQ%Hy^myk#R|MlT9Ny}wc zY=OK?9+v&Bp?*2IbN#X@DfdUjD$qjAMPa>7MrE?Cvs_90Bc&tG;&wtbnZjE7^C#BE zC7Oxp*n`+~-oOhN@A?>?mvYX%(aB@a{$(_VM-Aj) zctE$AXN(6xQZ!mHvMylmqX6duS%_Dq{I7fnc?URD#HG+C>w&g#*V5pOINH%2Ni-lb z2aq2eF>=Stv|eG%u&^G2!2?QDJIT48wMlTdt9iQX==Nm3BIf|wtmUHPUKYdpGQi^n z;aa^v6S&xHLqPa~4|5k?;VOq4mzmJYBC~_m@p#Qw_if346(9&J?yL!XB$%s{@DeG& z%+08b%2mhi*(#<$G8s|jM`VTnlSAQE^-1*jQJ%a+a?Qh4?n;Bbw>xn(L994?n-aw?g{fy~wA^jSl>miIYY z@_Bb9p4M_#aWaA8N1aJa==?2JT5GNnnIixv(9~EefrPSKjKP#W1vD?oU=^5$bXNwc zLz)&fPXRnekUNhn>*88gPG9cq_!kl^Qu*{JZ(IhK)PQherpF$_Y|RKzXy?}g=ZWsg zN#294q^LKcRf7XcR6Ta}ib=X>tAqr;Rf0aQ0fd8M89#>X8}UFNe3biwCa~aRRYGtW zz9hpiY15Nm4Rswh1Z^+Oz%tb$;WE0&q!#*C<6ixmeQ)jmQ@c>P$!FQ=yh2-+6fN)R zPqVBwnk$FlhiYlFJ%njfLT5pixyZJ=5n~cz9Q8`RRo#^8qYKtCN3I-=-SyEd%g^*Z zj#e^A#DYh1`rlPmJS9k`RJSgM-;b3&1E>9pN;>HVt<&hw$)fwV=I0J2I-KysG5|0) z73yNi1xwchP%4in_bDQ4Lb;R|DV!5eBi2kfTn0}B(;EA3E*Mm+{p#QAN9y{*WA&)I zy`p35Iwzr=NVziRX%gZSvhA8Xfvy7u-Ji7VxZB9#{j`v7xKg1o*nBs_QnY(r!h(kT%;9J9+~&UHB>Vv|!JVgE!lY0!ASd|Ub#E)MgH&fC9maZ%Vz9y48& zkoxp8(M@x_$D`s_-^qBQGvBP#!{j?L(g1fnpP?+%`=WRgfQ*BbpLA+A7ifuFk@_7_ zg^gQn@M@AN_YAJ0%R`5-lTt@#w;^a;U$=a|Z=$S5@#Y=uXs{ac>_ih<6?QljQV30K zCx7h9cCE=r*=k=9sx|Qm@9H8s*3G$9Qs?Y9uh|FY`SXGcG}H*k{Fv@{WX zyc3VN*ju9JMNK{yEkN|qj3cS@?fCa`LGrdPQ#>eZY42fDm1?W}n1u9wkCbLSS;cWs zNp6jLpUamT2XGag0sTvPFk1fQ^zH99wW04Yx9I!(%ruypQ&nplUIjXe7rx z8EC7$r>K8~7|;i>`4|r;;n_@uD?laHt}g9em$VHjZkB)BsleP_5J5D%1b=hpGilb#?qmSlSHMdMhX0J z`tQR1jpf)~OLMI&$}dF*GOBH-+rKmM*dzdl6AJ`E%$3aZ9p9absKbp~8rRD4sQny{ zev3O=>9$ccb5g4+!zx_*d7>L`1tCc)6r~Vn8`+i zhn$*d1L`PWc52VgAEJ~N=`h^=o z;2xlI?S$pw_ior|#@bk1Hcy=_i4C``NIfGgozP->(r{#!ez6aV7lS&BXik#CpAQyI2A-j)hSVg9@y-tgc-@wwdB~Z97M$gg# zvU04s=Z0!eGwslVjHp1mIF3LJOGx9UBf)${L9N=xU4Wb>yk}hFNwbAVX*RBkVdBfj zw566@wei+ME{vC9w)zdn#dv)qnGn)sPUBK`x8`(xfCfC=tqJi|FS$N&&2&{3*Wh5q z(mlGvRaA2*P#`G~irEZwfzPRQl|VGk>kpbiCv|Y^pFZ=sX2#sDhe(I*(#ITD1%^9h z`bBb7hJ{glmAi*$;h;=g_kEjDfKlwly!nzC&hGZCHXf7WdQU)c4x@I)F*Nqp)?;`i z$AJiIgdHZ&@}Z|}+;YZwujs3sEc-wi-rP}=O;U!{QlL%0PGU70GM4C|zSYuC8*YF4 zPf0KJ+mMV>2pvch0f8k1%9M?9ucKEoTKK6+@ljC^mJN9^O5|_MSYY~-6dH~S!C1-# z@?`_U`S3w}uDK9=a2ir=-{e5$w;KHkWq~Z%V{TZ5IUT2T4g_3VnFQ)$!Ed$FQONxS zAUOm}!Gwr`%oT%d17Hyu1Es&CwFMXjk1iVjq|pOcBGO$qva55Tl}I=&qKZy1q-Vu> z6U8;ntbvm3AV!+{9c2QdUfYhSi^n<<6zvhwZ}Y1m(_*u#xev>#So)&#@=cCJq{Rf~ zWEa^m%*TQi*LR~&3I$J$E{+4Et7m8Wl?h#L*% zwt`YOJUTU=ow@b-Rgu?Dc$-!%96Pftx<79EPbIE36i8u-%VhbS1xP(iogQ$jS)Sw( za`oAj9=-m~)!K%C(ezlxyL=vakwL;R4)K~VxaaG&zLHG{iaGN!_(Z@9+us|ePT8L8 z?}R?|(LF-w-g=%Oj)nLE_2oKo+7x6SRVwhUp9g^;bs&5-sVyL~G45^6}CIfmp;M3fovL`WfeaEFoLH}ut8KZsab zlz?#n7n2W0RI(Et>IR`gRDqu6-Sg4V8`kFwD@cWx>$OpXS>V0<=F;LN&1S8)fZLOm z7Z{c6Ss#3Oyd6B_sqC5=GQ&YxU|%6GkJn@I?j`2izhB-mVlQS#!yQ1#4+c#fNp6Lo zZ8e&{Wh}>XH9S?Jl0Cc&@TUU`tC&!+=q+^1k3=U+&Bz%hUEH=sEFH$=Tuwy? zeO%bwOAAFlg^?MyIieqVZ@(f8>sLOf zkZ|hxrc04_kq>>EkSHKBi_c6452u&TGgNS5X!zO_8*m5r{*}i2ZVOHt#Gr8rl`YEi z4EvOPO5UVnk80bKeOFUUO{f1co7)?$3`teTY^u6RR~l2Qa}wtOsrM^-)255Z(}JMj zMvY`mv1VTO*Eq%%9nDT3vZ3`;+Z>N2lgls2?k^CwVl)LVvR7f7XM3-2vURSqe*wa& zlVlOh$*)k=tmFi`F1*Q>b_;ir4cV#Gn?EDeW%pv&WlI=%5W>)q^A1EBm|B2!tAzEp zKM?NbL!?W0=~E6x(m>Zz_)M)$M-G-B@nDp>!)R>0^ifGMy~EKbZT8`cxJFNKnzmNyuUeSe1rm8Rj;hDN?N@U*TnA8#~Kfnue`m> z`3SulEET!avS|`kh-IHZ+}8O&tvM4kM(5$kBr;r@i(2a*91b75GQ`_>Ilh!@8h4le z^1Xo#N?d1d0uY3>)dyvL=930rU{qVN)_U%ucJDQe7%yLb-KMC;OXsbPdHvmlJA#I# z{nm6;+VYNg?nPibAz2vQ+^T--u(-nZ&01x+LS`H8Csac_(IpgK(*Wp1%3C z(dKn>r0ygCBMevC@I1Ue%xa_QIA{49+_jOzWbyRHD4FBYSRstNk%Xyv{aYkaGgI=< zaH{#uUUMN_AaoIP^_iC*;XIqVxUuY9Y&0AwZlm1@@M&fwDOwW`V+LRN7<{;UjTm~9 z6ECxSXtL~{*kwJBk7XFc7^D6z9+5%Yklm-b%^fFYE~7(Ggb||y*;5?! z2}u;`mUN-IFZOA$hqr%={9#8!iC~Y_qfp>|`#Yof4-bGc7>ajV>OB#)+$MzcSbv$}~bFqGic2tG_7-mcRF!q7Q%F-FT(y{OBebqFW2;%Mz`ledq4KL-ym zYj8OYdl zW8(Kz8#aATmZ}hMv{Nnd~$c19OkN+~7vM@31Qpsh8L6)(W<%o0f^6lf9lXj2eil{w-1=_R+%k7(owBhK#UjALUB- z~dP$D)m!ckJ1mTph#;Q1LchZ3KD9|-0-`{=%`l;y z*Is7%l116)x(Iy)C)pqUwD8`<^uLWb*c;1;geI~2U->(nNA}(SP!W{(Y|OL*E0hWH z53>}hG%eTL%9{~($eii=oH@4pocUY?Z%XV3Pv`SjE$bfKhpLJ?dOe1FDoBv+ZMOuZ zFXe4mFo_zTwEPteD4&wqI`*F6aF23GWc)UEzP94i9@~sB-D5sbB36YZ!{S;I>IQ_0 zrMl4!$W;;2x%KgDuIh%}wceI3(t0I97IbagtWG79dE7GUPw)06tR>QO@sM-ImT@s+ z3z!`v@AcBK6|n-1W)c^w!xk7RMK_%~$-JOa753XReDZ}bEhKb-(n3XQk?^12FnJ2yC@hUaa;^GjtB41*Lj;Mn&vf2w>M`}-NFsFvlPY&qtlz^@ zhJQL=%4-_8I;oH%{|z*HcJ2hT+ILs8YTs$pe8B~&q7AI`d@t!>zH7PDC6m`!@sl!# zs92S6+XY+L>#ZNO(eQU@pU{s;!F&Mh#BIbdf@(!Orfp<_wexYk2@3~G@#bYv>RAk~ zq)lLA^*t$lnV4|pSMTkbeaSUDa&$>kyCxmD5gC}FbYbO_SeW2S@+Kc zxow-W3cDGz{Q3Je$I`S`(+{gNQEnSmU*+)-K5T=>nG#~${#;;GD+2zCdSmxRNh}wt zE(t;uFwozx&Uo=PJ!1s-X&J0s_q-HV(?O({i6f~G!?Fy#aGCG@$r>k_&s?Q`z)Jpr z#q@r0pZn>@FM8nppV!-94E!6rR{!sy*b40;k>8V#l};Ukdgb4Wey%&84^!wLNP0iq zDB6?ynhUqH{22c5XPCFrFm6m?p~H6C*c4q`c%zq7#B@|AZ4@`X+a7tl{owJ6OTx}W zs+X9JX$YU%DId4p>*)FZ^=IiO)r|Vc^>GWVkea^P7zv+~%w~WCg>{Cu-3UX& zV3%*8@p_DLHM9+lBUGEGMrLJkCPn~H_w8SU2ybBNXU5(BN+$njOl?FMwy&e{zJU>c zWJqfBkrScXbSgIu_ulPUiU@JW7c zT^2Y+ z`yUhV0)ZR7GZI^cDl&d-1v;F9CYS)Y)gB53{&XF>z>`2pIIQSc-DiLN=bm@syu6kw zKV@P4*Vo5|^?W?$BH-Pi5V%{wZj>)GpRAR7^&Q{F?PQs!iO++VM)$Johk^6PFr^J( zDIxWJmqal#Q&@S1yAf={&|j;{4;Gb-ht_E|S`!hcRl%z1ZU^3o&DoBkv+6vr zsbHOw`?rQ9!8JNeISHPvj_ULSJ{O4MFfpCj18NH#F%wy@}-g?dE;xVDT zP^!D3 zUgT}jjg(VflV}5ZBN^ASjp*kFz^OKSn0r_$8gsE$XN6*z;HejQlz`O?goAEx%UG zy=`cV0T)R&H}q91bJgKoH-EdCS@!QO@n@5CZqME%pYvvQ1i~spvSec>|JRW|a~_Usv_-f9ikuKAS>#-KBO5Kl1-NP&qmbc81TqxIi+ z@PGC11m@3EeoebDiN7Y{e>KG46A0WV#PNzKqS28L|J6z|;Tt zlKy}ApZ~wnz3Kd3`0uI@hVK~vS*rh+mvAjCHGgCR@n-b@Wt2br1xMsZ_XXXS_yz0V zkC6Z2+W9Yj?X4!T3M6DA&i;$f{-0Lq8saM; z_l!4yOXt=EAxEwg;CBsx@x_S2a(+_<3Nj`2mU;4z(DMc8l~qiL#&X$Gy_9c%_;!&= zMl>Mr*-uYKDf@R`;tp@qt$f~;IJ7PeBJ%}{og<6|nn=B+!H`59x3Fsr$o)(O!o@s| z5pEpcRVWxAq{=K$l+*hbUt=3chwO|~TMjuNAT#`E1S83Csq5MkGP(8!y0-8+*W4@&Hw9$fx>RP8Yk5wy{Q z^3WwgHUtQulM6jv#?pfte_PtI1b2MdSLX5D@N<9eLDO6&iFUba1cQ9w{7Y-d%z?5XvNW;5ZOY ze0l&Lw)X9fXmDtx*HVU$yQ#T=UXk)$m8;#_+Plo}>d-Ere3HIejFgFvK|?z6%4NjA zk({!_5^Uo4wZi}Kb$QCfvp$0h9<9wj>tH&eY^ki3iR<0sI=MtyHQ)_Rretvc?ePtb zQgij^A~ii!5?)MI!bH}XCC**tgh%N?zF80MrGs3((EM?CxpO?)yf9SzD6d!e#@wz) zjh{x>3C$=eQd0$sns|8fglcPa?gOD!PIjKH7ne&hRdRoHy~@3dRiNaC_)8S>)5g&O*09zIMCBd2 zq_nETq4ddb{+>%#ipu*E>o0x7PM0%sN!)Uabyn}VR-Hvz+ZJy}T#Xh|&lg+8Vt7>E z6lvyd7`v5`ON6V${<@axf!5^#@Np%b0XVYq_~@L0k-9l;at90bNZN4I_vwI;@%~~? zM+EOl-a0^brKC$vuGvL^&9sqO16{X@>eAe}DE-v< zH~MF-f{EH0A1j)AY1;sESB>$dbBgcVV|>|OLRU;ig}nZ50+6e|ztWw@W{$@ks{PGl zK1Z!(=>%$8F0i{w?{+$iMR`h^r|h%0SOeqnd-+Xeahs(b?cr)Y1ebwkG!w1#S_=$M zFJeA}kTgJjeN}z4DV(T7vEv#0hUPL-=;=e`j4AckNeoW~-;#65hue2)?CzMY-Ds`& zZdALwv`v=;x?!oUlADNL%g%TI&DKn$@h)MSvZFGuGFUm@S1bE+Mr1t*$OF~w@n0P` z&{fTb+o@>?cpauY< z;LJ&3wN=-6#QyKd9s~ap-VuraG^$>6uU- z)^=KPhvW?pS@H*exhu&tQoPSEsty>$7>n{cLtlKM_`n{>vqr!i4a%7(FuUsvxo zfbrELb!qh{Cu6L}{d^*hA4JwcU|lozVj(rL1``^U{HPGLL>aq8GIM^6;!!C(yj~c8 zxgVAe^Mx_rgo6eTr;GDP5;|y|9~^$irq5zz`J>e-RKV#8}|@pp>{#zho8LGMO68v(W3XC8!l^{+c+?jkW##8a z>Umw3N3+In5>YL+q!(+=dO>UrdJD~0M3)93TGsKv5zS!TT>tZOD7ZEZDLw#UfC0hp zJ$muDB;w|@;#}v&f3vQEk51r$oa&8`du6hKQNY$PpasM(c#Tb`{K?e;!q|3yr7~hp z77rATh@Zggyg_5{O0AX)qo)hiVXV&l~Qx!&n3^fzwd-`D6M;o_-_H7s+;jy-T;T|l@=$uu62YaT@QEJOu@FV z=FjM6uiGn)@CziPMF29T`u!gh)br5{0bVdLxBG`}Y`=zH0#QE3dZssIb0E?pW1+?b zZ4lf+&rkVRs&X3eLTq={<_qmA`VUoWUK1Gnxs_<%)A=zTTo0+WN%Y43{#5&ZbX~oo z6Fcki`5o;dj>I{SCUnWflky*80&XWjbQF3PNZ4DZv8iNF>z;~)g~8Cz#0Wxu{O=OL zwc)owv4DSbM`-+RTF>4pQtfl_cttzZ-23%+5FMJjGwQnQxYf$mM6v2s_3|})&Pt@J zMh-*dlJiIcQ{pzAx_R_v%xfg~7`ENFml+%$&0+W)(rRV8We{HkP`fq{0jp`0;KOaK zLahT-=;^j6ZFUK;;hmLPM*No+z!17|*(%77mFnFK$V5wnAo1^p3P9^d1C^NH^e)+k ztU`zG)|XB_KMa|K9|TE#=CoX;2l@t+hL)Qq^+3xh8t`I1)#nZ+=(~$mYA_&VOe1jU6 zJUlHyVrW#vhmy%t$1K24ug<#yGY6liiz$ssY2V$f#tsx90~$0jb(y7!WL^(aCrwlh z*bEOpmL1dyGnu_nJ$XIE>phx8BZzq@8^8(DcDrLVF4LJ_&%U>RI!!kwgK#HlDg2jb zi{qDqi;IEMICJ;6-XrlECb8AUPAg1Q92N70 zQCZtPYGLqay%DAmseqO#nXG#x$rSeWiU^(IJAUY3AO(%b!}@QJN5YCxBm zjB$AJdjPW2h%XwpeXXSB#yaWA(*cnO20js1_GXa7b?>6b48?q!{pkWAmcw6 z`179C$8*3>zA)JLZhEV2nb+(4=wQjuVkF0Ps$mzvQIn%zk_HVsB(pQfb|P`fY3QV) zBlz}dsRC}czhk*w#Vsq{rdYI+_5Lnklo$Cx8t9B1NrCP&Y3TBvpDrq_WX@Mo)$eo4 z%d|R_AmCW8NY98d8l{C<%cafnXA+^8=WAU(YD+(M-*yy_cM)fd0_f#JmUW?nre8F7 z7hpON$~(*kqg^YW?>i<}t2J|g=u4f(oK9uAX6>G@-N-1H*6xGTGQY2Wvr?U0pz4yJ zQxTvcbWne*#(}}1X|mYYdw`@*;ThWeSKB&PJlfi}kA5Cj`Mz%5<<{^pvDyjpaz^37 zL;x*X&^P*0>iXf;F_FyfS0ZZShw%(mEJM|n+cH=2GxzLt*HiC%0ke|jbB=9sfyaYG zf$HybbBif_(8GI3Xy(zk?)ioS`NzVZ9wv}hCA=Fy=pMRe(EcK=F)Esio3i&LVa3?tIrV4K@K^7A42XnDuyDRG3TrrkMLIs>hQoJyMsfH zB`T|X(eB3&;vL>;YNu0{C#|BOGtmIlq>IBkiNFgkUU2b@KN9yG$xIR>h5-`A3hP2*&q zE_LW9?)|zxG+lQd8~A3}Es}tfFlTsAV%PCK8an~Ue$BC;bde>!=9!)QX)C;fpodrp zRr=!Zzxbvsb3~J)a2Bh>LsUXbGU{VtZPbf3?O%O-F9~jvt}n-gFc^Cj z6a$=OHHx%b>R0k5_r<@9UULG{wG@?A8J5tx5dDqKTu5~Lha~Fz9UNN?1 z(iidO8obyJZZSK%4SC59%TL5Zh2Vb>!+v^e&XR;PeaD8NLPxO)Oc`-@x{JVIu3%kV zJMe-$B-5|HL13mV4Dy!m?sGENCFcqxm31C+I0Z~RzKbV2zoMge5HpCG0auix6Df;P zlYLFONdp54b?$=3- z&>7>1W9?=Ek8nAe=i3weZJ={7TfyA@0_y%MO60>*y{LQoO2PH1b#?2&QhPeT{f}Ya zV^=1F)#mf}@Zu_n`EW`ppi?`EU!S_E=Q&U&KMZ=bsDFXuOh&%p)|4~R&h1x$h$)UO z)>%@PXIPp+v6{%Sh-GK@CU$w58f$1!C(tS|@P&qdRa-pB(y7)Yy|x)$M!rbdM4z|? zCN z9nD$jQ_t@XrUH2C*#LDo+PA+G1g{hkt{hg12x*?C?GyaF{RGnTE$lVz3hJ2;Y4e+q zH)5oIo|7N_s|HV$i{gX4Tn@7q{rhKhZ-LqTG{(u>$hsOUP2flpqc` zd+j$}>sgz2-X4!&mYbB1O6^?7EyYs$=LA8Kk6`4J_zlq)R+NS39NMz%`&kt)nH9DD zB<1=JrtD+3u8l<&#&_YNub#g+3{QjYIrQVp+hSFiuIv)eY#+1D$2X`mI<&fhmG8OT zAB;RVdT$=gX0*fcQO8yUx~t3ht%Wpca{Fbc-*xPQ?xY`&fFMn}kap2!(H++i(G8-GCl;i|w{(V2&HH)`ab6u0=x8U;`Ya#plYQ`zMH%D&mYRM>)p z3%w)4;vv;t0iIL&F2#7yr_RPKy8Bp6*0VNhcD40Es3?>iee@6b&i7Led|Y6rK3QG< zhS&;0nD!PS{=*X+3H0d7RA!`;()7N0fiqRXOetKG zRnC`Wr}fKr7hF5L3IBOar8q|udoTr^4<&6o-yIxQmXkv)yTd;$RSN!WB4gABJLR$Ip_RfR{StR1H4YCx*Gi@}tz6#q!+yt{Lz6r` zyZt0(j_}0imaTlw1rIH%VLIz2g0P~Uxoq9mM%0H;Eb@eE{7qyE_k5{bF~PxiaBcem z6I1XV6IC3!pi>^6%u0SR%@%WzXshIf@Qs08Gkqy;I(M1A{$D!(n}8-~)1Aru#Q@&2 zw(ib%F!Mv^??m@gK|nv);DwJi6{CJRV}^p?&TGvYhwS!*kM4ci$-1>BVeh*&4>HyR zaeEHJXTVxK_8Adh-aR&`dLhEEr!>p&uT)CY9y~9Tp584Zxo0`i8Eu(U3F?OT%-;27 zd@{1kZy&lec+1KrNA;`3^pZgmemDIRb}jMFs5KP8A{R;GuJ3b%?b)Aqad3s#(J1ke ze&Se&lCYcInPl!&iDTiCA;ml3Tx}5f0Sn^K(5$g2zFTA?5#gl2CyzV-tBs&d%5{FE z?X!IPc%&%fg7vAoIw%x-uraFnWvb<20K{V+&}^W2YHc)(6TD^bK+4svd6Kw(FUe^! ze1JUzkZ6M0^Gj*g)h2CyCcgfE%BLnQ$zCwOg8I1vfh)V5{34x?X9X;tyr3S}2{0m>Q#nxyY9J06W^#KH;t<=H#C7rWk>8&VyF2G-fr zKnKhT?;Gf!3o??x_)*6qN^cy13t?kj1ERRW%g!20;^F(VXw1^nB#mDRnis*Yk{S`A zjTl%6oN0ld`y5CaUEV9S#Ji)8$nhd*Ub|mnij05fvlH7*str7BJntjkZcS0VO~bPl zd@zwYOlMrSt_}@1)&J_#41HTfBa#%xvS&R^kJreoLr$HE?N1JRK-p)R>EI#$c$c`hXew zkwLZGINe=(s+a|RMnZxkqiR=IU(SoI^c|z_L;d}ifO}Jh)k$BRe3*in<0WQ)%?YDU zt4@aE*y}R>?(F+6D=xF604XviE!Vga?fU&o{|9gPYZR|m=r668h{Kc=eH`d0oivlb zFL$?%WKDVk9K6wuHW72 z?}VAzP9I1ex`2h35GIQEW7$?MXDY0JL(v}tlD#7y=E^<>)XMtVK%voxE)nrWti_|> zMknO6eawdao#t%EWh=9`UkQ;237|BJ%ch7f1)#Yn$)bja`NtNHPL2)vJ>IieI3@it z72DlYsPYO~;jI?e`|Fzv+GtvCH%=7P3IJCQrcxO0_=Av5g*DeHMEZi1t`0r0d$$Li z;;oQKbF+PtVuGY>lfL-7Yer0M{3bdzZ>%g@dMs8|W_>dS;yr7F3jLft#*@#OMiaZy zuics!Ubx!3F9S9_i7Ctq6Gc2ufO2hnT`M1bVht_|9!j)- zz|-tv=qjfmS)%a6IB_F*e5r;Yv*@gkBlatXN?3TGBk;(m{(iOtNq3B%wI~`>FV_=d zwyp?p1h$%e=di8Q`t-qW>o1ezXK8`5ZiB<^xFpTYu_PwHWriuUQa1=8Lz0#>Uxj?2 z8O$ev1loJNcwgr_dL%Ga@C&5h2HEOk=$r*~4-(|fleP}APiMSE2wxN#R=sfB$iyI|66p4`=8+OcgoezM#0QRwZ@_z?MpcDu zSiyW&?|QEw_t%)QcLmqJu`QplEZ~8E^6rjX_)e3Vj>AzuUUpZofBd z#ljlI&U-&Gil7h-4Oov<7S*q-86m;sa!45~P6M$p*he@4+_Ovg_qr!Qn~Ug#l{ip9wJSTv_m#Oh%T*|I)Z6Jl zxYfbOI=U_}*kE@&=7(Oq$u&q*Z_!f;57%m<`t-K?r2B`VlVxdcIiQV1Lh$eJbEt;O z*_dIX%P0cgMHm*L=Va@r%aG4SDps4{M-jm}c(41@VdpgIFJbJ{+#GhofB=3sScv); zSlsN})4xXY(Q;O+FtVOH1|#9cV!6oYq5Y3!^iYfCmeSQ|6%mK@>0}NeYcnG2T$`fP zzdBj1clUDo)-d)6?-4jUOoUU3y+SG0+y(XraZBCLd@$Ob?5-F{rx+{agNUrRo4iIs z9d{UJ#yzuj`5o3{!mMoNJV$O8wdl zEK1mCb5&9p_p&Ka`P=+F+3YUfz4R9Hzu@w2)(wEFcK;qBdTE)&hdaVE zG!3=*mr@c!`gDx_r=L|ER=z3@W z?}%yNbn#$~t$hELlgU`c{S3-0(`?q<Sa*v;y%ldi~8i3EI;?Cj?1Vgxiki_hRZ41{e$-@ zp#i3)Jd3EEegQE7jzlC;-q3Sv7tOf%bwkk?7BWd^Vrwd8e*z~v?A=Ml!@_HOHNg%W z9b?=!YNd9vGZMfJJ7$(C)f&T%S{CMFX?k)*_mg==_oDIuC8zF%-K2{1+Lh}^I}Hi# z=UfjPHim3O31&)I8gdzT7Z!ACwN<7&oYo1$B`o`lz#N_z+g;cBy~fcSN?OHu3^|yP z(pKf#HHfdUJ3g5Fe#J9X7jMv`4n z%^)@kL!b(E`v*XOH`LS&rjnZGU*@N(o!HpyS=cNA?VqgyX25&7)Q>TEz2SepWh3sK zi(l7%7RP@z+g&{@nCA8T2WiV@d0)!qgdI?JN>iln(5>2TkOnB5;D|>+c4Q3K%cSdg zJP+Loy-QzC&h$$#&#-%Y_0SDl50Pt!uF)~I=;wdmDHEl-tAkVr?uv%JtwWxKU0WRh zD-8~aUyLomu2Zjg%dl`5Le0-+l4I+T^mp%qzRViR zEq&Sf5o&rA-c99PfT>n(kWMt`@9Y3L?hR!*+WuY3VXFGi!qb5L!c+OeyqUz~GL>G> zv9G8sLaC$3^VOGxwdb;(E0N|{4JW@0;oXdP-sh{0^fQJJOQMY3->DaBjq%>yWnODQ zrS?RhgeoAQA_4(njr~2~B)9CKjyybk=+!U@rYW1tCH?extvIlWt62dJsBEixyt}Mq z)V?aIe`|ebC*v&O9Cj~8D)@ngDiun-I~qPgO20z$XZ$)K`QjQ5^K-pzA9w3 zT9-kWOal$--l?(WYh3qIAoYe6<5+7c>ipFCVSk-y%e-O7Rk}fC&STZ0W4D7#F1r8d z{vB+vy{-A6$O`XkQ5c*2z_*klIuD_)#(AEr{#*%ArTVcVzQ;~aSZuT1M*0*_SppjX zTl`g=Hes+9qB}BCD?R*Zy(rn?C+laLrkDQg8VTzZaa{$dcLg7r)U5)d1)=30g5FK# zx0+dyHEhU8ZKNjCnP1#_O1#-K+Fjd@+jHsGG4nsm6@XXy?QsS&xeZ>}g!$G}Z-JCY~magZcQylb`o;;>0a9T8a8@gfJl((5crV)rT7fbK;s!|D{6YZ{B zJPFLUTHurZOy3JM#0@|C*MujouH}GTs64a|3po$_jrb%9HzFaSZ?~wY^^9v|kZoc# z0;XC6rU^LT`N?ZfdBDEB_DZagc8I`bPA3i8F; z`+S*r2Oj*zLqRIkyJHcdo33x4yl(9aHCQmc*-7T1K z)|Y~DspYH6YYs{kL``Y{!UhlNv%V0d*Ko2}l%XH{mKp(RUVfDM*|Ohyj6#e<6Q+8X zIx7A~oT**rMfdUoCYjAS<6`(y&V_4`M-}tS+ zp&=FRXZUkya%Lm(J91R%4Ml9@9V><(PAMPVsU^t}G9*fS6sUPUFC^5D6#{xl1-hD0 z9>%h2@a);V)X?fHoBD3A>&&4Zj>-t-H7l0sFn(i?+zv~(_J4-kHa4xpUf83xx29qo@)39?>7|R`20ZaOxV-n z>zZVOPE`am#E!{=k-~7hTUep?FE7=BpH#6O4WPbls#<=B(=x5xhK;)hm(M;OipLX! z%3sjN(sTd}UOIm89G{lU4g#jZXkbB~mf+tZ4AbvLH?8}M>xgOZ?NhtJu*ju9h^X?M;$^9XR)D!90~);qJaqEW5HDOF)S_N1u8#>-3_Lx+t8xv`F_A$|&4mlVL;kwS(m;oQH<5Qj8%rXh}d`LH+!immgor*Ro zVQd7O=JKQ}SyP~1kGGidH7}Ey*_2EqHBIg#<8|q$+ zM&OuuDO8BvI+2H8%7c^KQCJooO~xS(6pK*ISk3(Wu392F#AO>7U*s|5r?(p!ja8_S zIXht1VsI^}NOofZ&_V=Yz8G2(wW&bf_h%CAZYbdgg3a!8YAwDhzCNtgp6=Xe?!`$_Mc~R*-(N&L`-Cc1 z`+(n0G0ZOVi%!LSAl#Um{?Mxsp#XmXtwpg-CfYYAK`n93uft-8FRj+Dg(;p26N(+} z%W52QwMb0x{FqsX8OL1o3a#NJ0NL7`uX^#(NapSXN1i@lEIPKyWci&c@#^_B%{X_SOz53i3}ZWdlwArn&N0=h zT@phEA?4@95{dKRpKUlrj~bs5Pkvn!=XqE9N8yJZId-jZm*wmnW-81KzP@iOGuA=r zad5z)Om!PWU$1iBJ5%EIII9BTpLndp$sw>jlM1#)&>~GWE!J6ux}z5-fBxmU(AR6_OFfxuMUVx8H1T^30J@rl*YKN=A^NS;Nsu3)1U$ z1HmkVhnWR}l^0g|!NOR!4}pZRG;(IykJs23?{OYZO{pW!J2-xOGj=7rYu@R(H#GwI zZ*jaA*Pls@Sn!12($MKYxapneU-XNL6Ss)J1(7cQauXoCL;uYa?xDe{EReqQVL2pG zxWCDDIi3(5%@8aFHGhebKi*1Qmj6!Z3K$D?7tY!RukG9EVhd!IhAv^~+hI?-&KdxE*jB17*DHix{d+^>J~d0uK=ypR>5L+$b?Ny1bMGp-((+s{^C8f5 zJcbbIp*K!=?}zjba)MD{Q=Bvr*|4tnF|hjf@s3*Ya~D+u-lR{~$>p2605oc9J)pRw zv*v1hxtQFqW^y1msG&1W8y8<}<}i*G&1hfHyM5}N)me3TdEgHDKohy`Ts*gZYh9o4 z`AD9>@a_ir3E;*H6Eey{(tdSLWfCO$oF-r`c ziZ0wqVB}5}?Joz5^k;RxH{ClUB<=3J1XgCv0!rGWDn`?3GeaaODvxCn*XT_D{fwPl z>UoJ*<-Lkp2Q+)SjY3R9%`89KW&h+SkoCIVKI!Ek32!!L`9S@vT<t9dJ7i$n%rY>ygJR=*E4J08>Tauy9uH_8!Jfn7i3`> zv{t}|7l5+Y{L0MUKTd)A&2|cfjZ+Qq>3r~-WwGUpq5BQ1o(c*`Fh~peRSQMg28#1P z#BX;;#N3Ob21s>8>e|a3+xNNPXlTyz#NXqwg#P~|IRL*;3=v+}KV*^f{OMECad#2! zqDqjaK+h(8Zq#cc+~NDF{I%{N25?{YqT}tQKXqRPt84m@=EjDu13`N$X`)1ZjM1rl znZj~BQeLO+oHK!MGo_M49=~dA(4TJ#J>CJi15%7P_w&K5#mf2UZ@5yPH97AlD7_cA zFoY*AH4KgFwoFfUrN$xm2L^cYnas0N)!xW(e5VHYJ(WKA=ZQa?6&^vNKz$KmLGgtb zF^o*4*V&saVQc-QoqzSh{o}qMkoOvB707M<==QVh-leVG7uE5c+j`W;*72^%zbzk# z#NpCBE`Qb2vH-clNEQHN;O=%;{{ld-VD?8rbOHIZ{Sjm(pF41A$UcaC&|#mSPgjMK zbR}!>f5YFfYupHlnWeFJp6It6hwO^#D3vr@p{3!^(Z8O5#mz{CtrtXN(To`w zQZ%vvRX!^yq84^RyEr+OTKE1@T9xu6NV9n(*DanQ#!BwPb-;oL4TTp~P<|d1Hghb~ zbcpXvL9TCK7nK@0S$GY+aK9G}QJ8w15BI`3xRtj~t+mgLJB7f8IU=ml3(GG$AR6SU z7ls0!5(i7hDY`}8Ux%kkq_}Z1Qq4d7L`b&&l%`ll=_OY#c2YQYSN1ZTX9qw0A_a!D zak9fx@3^^LN3KS?S4vIH;%nTbMK$4Wv>nO4nnh!^=DTF;HCgEgM3(^^zPm%mKkC@{ z4Y%<}Zd;R&r{{X{{3)%7)pIftI9OlX%WHL9h08CDygtqK8#59YPWhO@~1;k&w2ZeN-YYl@h`EbMt9f16b zTl`1H=&@v8sc>p;n>!IM`vq$9%6nOZeC(Vkd0jZ|#7*XfN3Iv{uF8G&U_eK=L{!Pc z7RIee{`~`rb=~!{;F>QB=OKu)bWM-Z(+Q;^hE%HWhXb$6%{TeCvl#Z+lsYc-fh!kn zQ=T+JsLbmTSuEffnDfOnUEv22n7{2-!B|l{44bJZu&V0IfYD$I*v`LhQ+wKkyAdT^ zeOyFYr$gJEFtvNqH9G)`EPJt6z4d&Ist@fOK0_P}6977?fB^K7uhzPlvAbWJpGBoP z$RM7$vBp$p;3iDP6u4Kxh!9*NI;*;#F=kBh;Y5NkP-V?m>Bc<6InMggU-VOEsPLGl@_y`@P4rtf$uK)2I z{o2_)44jE^)K)gijSiGg>}~~O(w|7UZP2dXmA#vAfn84xcEc{=5V-Qhxrn{F9Hnp4 zCCdjvaLulAY_-6+jMPR4zsRQgirW_R;GLSKWGa_i2gZWBze6+iw`y#%@242umd5u} z--0-OxqIPgX@VJoBN}mY0&pV4cA&V0Qw>wlEvmPK5ppPN1z+*TipjV7iBGb)?d58< zC-q0O#st`DVc>HQ>*8kIXles<`QUOdK1nkaj}B>x|9p}1fG<)4|72_T#vto|dB#-J z0DN~+1m(|znZjrvz%U#vF(b2Wdo%krplQl{^p8IxN*Tp1uqtXX+HF_J^tv~LdXkMm zRW3y;zpOw&{Qv^sz7vUh6|%`+$*T?Smz)j|Qxhx1&Nroz>SV+2ePaNaT?TeFs%FD& zw^|lV_J`Q9)$PW^6ClYft{+G@XMG=)PUq=Qwh@`BB-i-FyAQ--?e(Ds84#dt^{ z4z>v1R>d!-(g&bTgZLC_Db|YaRx*7BcJ6LC?qM{ag3V{(lXnAsi(Z10bZXn5{OeTxcCR^(Ojk79FKl&}TU7ZkM^nQ| zPH7mM`v(^?k2S70s_9$&n)TMqAFuAC0iTNX9qup~uIob_!RmHiw{*xc#yYuzpF=8< z)Z+%3d8G(~*LmwEAQFo0_)iL!t)>q)cM%hf{|}cFh0sWU)c(NnH=OB})V!+)T7Q{`Lp3Xy$=5x>!tdzAuk^ zFNa(o{2tz&DH39Tdbyb-go?FJ@rlcK{;eOF{f}QYqjii;p|*^*K8c*X7bq(Qpq=d{0OSO2gwuuXaNkTn9QFPUCx6Z!VB8(| z6f7&#U(yADz4C4_97_WZA$52J2y(Z|DhN5f>M}avQ9G$RLC_q7_3}jH@sj(##qG%m zv0?;T_1?m-T^F7R+Z8^ILvLXCcV!1^oEdk=_@nbFcsy-aMH&{GqVx-7rS9#%vdujO z5CF8k&Y|5VIee?cy0S@pP~eO5Zt!K*6{hpTC%mvlmxu+&4?qQzLvNTOY*n%ZjAcD)G-rOze2USF9Z zMSVGD9_$_*Hl$HFSq4ldiWodIIqT*8R+ZQPF_u4LYhYFee9?tv8*)`P3X$iim%z5~Kp30UZ51dZbvE*1hHETtSzt zV1ep@O3f0PzDH~3f_`{vjug8)BgWTK)Ey*Bn(@&zg!js3sbEG$3l(X?v$S3pA2p3t zGuL=-)d>dpe>-n%MKrA1O~P{cU;&?pTr^J^ViYK{8u+Ja8tXDE<`UAYv90$fc&4~W z&<%M%rqp&v8k2LAxg@#Z+3|ID3F@?$QK25sxW^b1f4#w{mWzEN7q~16Ykp?N=>YLw zSN&&(a=Uh5B-@0J!1YFnO!GSTgC+Y!3HsaFa3F-b=Ujye0t$t!Xq%Av-MSqF4+=92i zohji%cD%Bl&WeJBX5p1e@-&hA@i9dUXsM&WqNf7F0e|s7EVt7tl1KK)rf#?iOceS0 zkT+Ofs&!tkw^JS;MEA}qWd@WYHjFti9?AoRJi=&t%MF%2i%`apr`3f@nU2xb2CcV~ zMQK|JO5hcL>2!d zU~3n)o$D!oDbK@+#0b4E7|}LKX(r%7Jv|qA;iQ*HSo&Vj$1vc{B*<%cA*uHTr)QI7 zSm^Z{aCt_2%_0R$1<|B9TGiVmBja;zAk>ZVyPYH`2*Cx`>%5<6jL5IZSNt3QOO$*e za%bl6z79wdx41i$|K;&)^i*2FM$3UMx6VQ3bu!anKi&blhv{$CHZ0R^Xr^o!RKTBpx57}$2mel^TXk>l1!G1;mK>>6g1jkH9c>1qCP$EFr z67|I~BMI?F(*zZ?8V;%95c2V3vY^l>HIMq4AVX$R^xrB2Z@6uPW*4jT8}Ta)Yj@|r z%;TG8sM)PS00Eu`i2nytq! zllHm&q1&Fy7LC*A6t^Pwpv?;)Pc6BGCCJzby&Qwo2dz`vqOw@2MHABcR`19#W%aj( z20MI{%p?|2rZD?R&@R{N&@lFLFCXs2daztxIh@2fhgbC#F#nLdlD%O4s0uk>qciM+ReW#D%fWS#*M(&fhtEZ|)3%^7^ByUZHn_7_ZGo(l_m~P#fhd!Ur_6_{w`g46q%m?yq`nUFs~pLAhXWN=e6WZN zQG*A3q*P92x2;dnVUsfVhYV@NM{+4tZW)RU|1lifH<@^mH4O7neKf8NQ?xb1exOR?vC@;TFlMdbs$isWuC z7)=EDoX*bKtFh1ft!E8;+WKRXeAJ-{n|Td@LpW|1dS`X!A@wzy;gIGPF|detDEtN6 zO%Bz0SBF58^T5_1x^?3}6pUNB6`w~cWW^QS3l8l5PU{h(U3>6>Mm}xw-Ul$8u6Zx@ zi`r?wkVm=HBM;jshZFd(pY* zxu&P!J|d%%J`^bZLM3H zzR{!!(el?pK*c;wtbVd(TZ6&!ci(>5k3IAxu2$;NB^r$=heB%~6h7L&$JsCz2!ci; zhbNS{RppxQ!xkO{nHY|Kn7#Ub%}_FCGgq8A$`26#7Z;5`p8xt?*O$|}rT4bN)0iq! zqLe#khhcTL)GCW&kDRsMZmK;sgU3rG%o<}gQ^?>=>?Rub*7U1}Rr+A`l|mNl8w}&R z*S_zCC3H2<|F%pP4RF9nImiEXjtTEvLf!ut`(R8G2^EcC7*Z``Ci*q;Us`%TF#B8M zbGHm}G?4H*{b{bVe$VU!1gVkyI!VfmB!0RbOcw`+M4M^`>h{NbU>WUGjtQ~EQUi-) zPUS6-fFRy`0nn=gxm2}+5@*p6?8K2lBTxbr$d!IIzy@};@4E6QfhqM*65I3h;KBC> zVoPD>8NOP|1+v9Kcrvv!uBfPyL#Z^;K(sA6j$U)p!?U^fXK$O=RZJSM^TD0rH5tKL zBx0yU+~-FE1;L#rgOTk{rl;deM;alTyM%@&zSsHNyEV-^@yAbTuGgyS+M5V**t0)E zu}PD%ieARYe?dh*|3Ml+bocKTa|FtiL}tHsZPgwK(GH+{@9N8}yEn8&Ysa~* z-oomF{!$T5M$!JRYM84GL*yfZzuwA=k=pz}hZQZO{mP_B)}%`f>=Rv!@9w;)o(1l? zJOog|unGJrr*Gh=9x}U6QJag}#yo&MPaf^^o3c82PvD=Iq5OV3Y_`I3&M1g!6V04?Sd33Zet z9n$yVJ@K?^ViX9W@vBeykjE2Aa0|f;!<^?N72tS0e^H*`_B95Sjq&{{^IAzvg5v}_ zr6|Wgg%S-|((Ah7dZ`VP-rmU#12oIFy`e)TVYs5@e46dcqG&l?T~uOqNn+1Y!b>Ka z-K4)^R@3+V1&U;52VOB~X(TYG;M@Mh?H66?=6LB(P0dYYoffX2X{z}(nnx)V6*=zu za9^e0(vgeWIt%PiHT;TB1vP_iYn*vFiBGTLM;}YE%>;-{&`YPU?e4p8!kNIQ>XX>kA>>^Jfa)vD;-Pp?5TGj%QS)Qlm^v-K<*-kEEh3Qr1H>xK#sG%bwAN4{?bZjiW-e( zI%tWI_j*Edsq`GFBH!1&vDo|HT>xzl^Gl6rhyOQNV+GxOoxm7M@ATLyPn4&NSsg1VYrtaAIS>Fz+GHM#WP>P_RI*|jBg*Q@ zSj8U5-XhqpM5(j@&yum)ld^tPrX{xt?0Pdk^TCkT`cMrrw86%h>R zSAvRG%f;4k`ca%WSjF$=%*L>iFe`pl)KuxUn8{r~zYI#_@l@9LW~LHHA?N!C!Z<-{ zcC#uiPZq6u_Jg1-^(~6yr(Xq;`L3mZ(aO@_V~(;Uaf@PTQSOBV&U@_w+e0;GXK8dF z)PdD}$%4i?tsm`bek=2q;dEWl=m11(KEQK*@g%0gamy(WEjnBEzWn1-XCqi<{5~TL zw+?W zW!Vq%6Pc7adLoEIITaB)DFa~xfK?<)wdFy!(-VY*%Kw+%pBpI&C#tJ6#rrfPedd5Z zvhHk|?RjDb(2q1tu$4$Z!OQr*8^(&1lpkNU4hW6fYMHubTYP+C9XRpElP`K$abUbn z=sPmn&_5bd?&ySr*a4@-v3*5oQ7nT%AC~JGfIw*CQ;7CLHw7}_bkdlDCi51W)fc2p z(_+n4>)$Ut_W^N=(s@_*ha-N$K|CP^I%kgax2uC`Wd`&*`(s*ys~Qr!$u2Q8)WC7^ zu1RJw_CCW4rMXetKzVUm<;IvY9X@?=T1pB)o5AoHeJT^7(<`fzeJdnAY=RS8F*Pff zxgDq1=nGI^IN$F2Is{Y!kEGyph`0;Z#F}XS_kpTT#H&kuyA+6n-EB&3mC+180|C8t zqu_NA4Nz(Br&6x<+7zuReBCp8dcXcE-chTn&EcG#DS_RU$g-2(wbVQXBWTxO2e=Ob zQ`uE3i2?w5%JfBL<{VP=R2QXYr=^2k0 zm#c9sS@q-J7Myz8ttP<)X~Yw}C-`b7`;*m?WC}bxUnNgw|AZF6>zb1S}|eF)3Qf`AI9fuQ4YUxsN^pWa^{y0{_3$)<2jk_Y;XI9JoZW! zu9M?|d%xNQhF4@I9{vN41Q1M~=A}Lzzn1EhOAPV+SXSc+(N$3|Rh1F#c{^s)E;F^EeMF*X{0rHhL83p8W zw_J9M?|`7zoKhmQuj=>Lo~$V*%!fQUl@CB*Uyv6`O^hWtu-6QAy|M}Rb|7z&vld?M zR3+{MU`b5NK=n$w$-vJ6$4&XN3-?s!9+jjZ7qRmIR6LOu$>8C}H4m(<2SOTT^kw3i z>z3o#WryE_0%fo;{TZ=@L#1vdYq-~OUnV(wFN+z0JQt`t9a`wqh29px(h?jchk$}l_p`zA_trBkr;J9?qQ zW7&ZTmY{g}wr`TPb4#Qtc~lKmVm;qauV<9@on59^<((xsIObg&8Qmhx)KW%_gqHkA z1!V92w!_g}f#D*K|3s4MpI+xqd3r2kj4Rc7vp+FJq!^UEWr0!Py!h+P3F6fX4Ad*}K;54>fk7Si0$lj4g*+YrH3xu)C^ z;%kgp-~In0JTBw2Hdi`aufED?F&};PuNVGyM)={jmqAF;g#-jta+Ewy#DaZ+&uTWR zFopZkd|1{jt$7EZ{tK+P*|Cd_c98@aA<~JvPuYs!)(H4C3Udsi6f2ePHfJ_n@pC&S zULPFtaZx`AbWHN-92~mhW#}KUv-$mp$L|0CIQ9=mia24=*=T|mm@p><|5FnEf96tf z*!OYg$TD`q@@ z9~7vU`t;O}%clM3@9h6B>&*G!U>kQ%6#0L@-biE}$eXAJT#|2hJU$d6|9|YgcUV*H zvOX-LC?Ymgl#U80AQ5TOQB(veDotucdRIbk5m69OK#&fhC?F;D7OH?sCln#H5Rnpk z3nY*b@;(0c`<~-__WAAYyT9+x&&!KTueB1^dY+kiX70IXW)^8A84V~E??B6wgRwe4 z6J6>_+qucto~jk)-XIk$P0!~^XG`)+1%`m`r~59f_6@0)0CNm_M1DA;UFtpX%*^Zg z|NJHYx2^l%n;SlUi5vtTfk-LWB2W!Jj*?_MUjob1?K3`Blcn zGg6X|_dI@t_d72bns!3X!vp-4QpS`Pd|6ydNPr!|tBbA~G+@l!^XrHsD|P5G1scr* z8n|cgL7r#7{-`v0_I=y2VZZtwx_!r#{`}FzBwm-D-kAV;%X1Us)RVwf*RFuy+ZKi>lSNaUPxQ~P_mS!Fw|1Wj=ozMMobHMlg z2N4IHzaIb3b}i!sk0;`zlN4*%U*G8$DgIX%_CbKedWwbrCr?hd??en(xSMV@sE7ZV z(0BrgZA^N)??2kL$dCIDA_Q_ryT1IYvj6I3|0+I>OvmJoj=w(tH}>kk<>UWX_5bfw z-G6_9Y2CaB4e5S+bIa(>iZfgX%{eeh?Rw+)`#SN2vM&l(uMM;MQxbWW4|W zmsoeqFFXlZT&o*j5(qmDAxRhD!@)qB_8#Z~Pp6WOpDj`wm3PGn<2?2l$4Ee#Q+wOA zzpJ)GwLCXws#k?8YN{>j*0t{8W5bh1)q zh~mzRWH~RH&Vr}!{o~4P`-dFCcg51t!vKzTR+kKF1=(r zwc)izU6N@7yoSYHk`L*ftR1GPL*8_WFAtMX@=7k`L@Scsjp3?a7GfQK9@W1k;@>L{ z9g}N7?j&>k@qr(qpH|TgYN)lrBl@vyD6luZKQT7)g5L1*bUZFKELFeS&ma;^#F zh898YknyV751q5CRMx1=_OVy|T=@iBoQJD5x>FmDS#@6Ow^GfE?P_u$7#q}Dpk@Vp zYy2mxYvw+;Y11XmUe}k)?GGmTP9hc7@sZrv;%11$aJ5Nyvb4QaDl_9#kCpLK$VcgG z2j<6Ye_N71>jI0B|(R*mA9Nf>{I{ll6L!aN8Hl-N6|?t13H3Bp1;dT z-iCgr+d|CZ4n$W-|DA?WMH}c}xi~|A;Hc{80d9#?b@1pz(=_H(l}%54GFePJ>osu_ z)diE*DlkUmWtasw0!|G$e&7f8?K~1>qOKjC64#^aEj%&)X*oq1XHL&^!!upyVDijjBrpS`R?BwW56Od;l_`_{I_5DDl9_&+#FKJWYI9bHFOowZ6XxK?B zP6R2OEGB(}Tk@}m#lH}rV{!YVzD~D^O8vQ-Jd*pqzD}7%WHqL^Wtw6~+)e0*%v-jP z9E6zEH_0#0WphQ#czknJ##g_$X#3Ge*yM6bZPRwr)Af+Nx#?-0{0MGZ?W#JMZ?eKg z>^3SyrNgiv8@Ku2ti%S7;8utU;$GD?90QZsfE%py12xSeIiy)pmZp`81?_BfR4Fg7 zJIX1=(3El$|26cKlU|l0h-e5z5OgId_X6;2*ewCnWrkluN&H{m{#qn(}d<%;6yIPQUDfr*1+KR=Dr3aS%v zRP5m~p*RTiExVLcmC9BkC73$yhTpx8tHkc!mmlZ6=$h_t(*xC4*^%JiajMzHd@Npe z$A8)5p?p}l%zo6+Wv0_ig-i1BJ7L3GVTSIe$bYdJe|^8t+(4Jt@TQ{Lf9f(_KuA^; zDMmXax@}EM0vrYI(j#QKkKbsy5aQ{qpc!e;8qH z45*-d%ONXi<)Doyh%*-{(x2OppIy?)9h0&j>AT!1>o_9bZ51-Vwyc!Sq${ZZ4ipy; zNu;!V)i7#um3H5G(V0;C$?0y2Ow6}jgUg+}@efhno4FgFYX%;(QseWN!3w!!41U`Y z<0<~>aXWYl|A#zE-qO6AM%&m(~^~ov)48L}wtdHbP zgjYeZxXrf*3X8PNPH53q?Dgr7Tq~MA!VcJK6aIu_b%p4VXaV_}j=PW@_T2Hu;f{kP zb!CpU_#IO4^%8e%&=tQrH}t=g z|2WWi?jHiFOr6srhV!OA%{4S;MJZd~%6wc=+cLX^cWv#4uYP4}>Aez*r{I9WudnCT zO>KoFH99e0gp+XGO9H?a_vKoO%?+iefSUb5=}zj|pL*5V6#q1Ou?3cdUw@^@(8gJA zUzNn={}IU{?X6S4HQ$giP*(_T5V;#ySj4>q90oeCv@XBCwRHC=(CvbB&6+p2MRsx) zy?`lH?KrJrS;iP|&n{v0e8aac_`+Rx0%u#pZ7Tas=cmprsMBmWDpROCrGT6>x4w8+ z!3W5L^xRq>%#;MJ^;9kU;j`j_d#A$XGeX=R9K#GEmt!8^V)Bs>dh^>#aKVC35Fs#C z(^3A1_CJFL02F=bQ@MM4j8HacY~8z5NN1gG8}HjSve|7HpZbPfGSZ;GrG25W7R*y( zWiDOWm=jwc?V0FX-#xNUfG9UF!t^q{YsEAz{OgCPXA@-BAagtLh&x0FgXY7eQ-D0XE~PLTj_A6 zkbc3CnD|GiisAU^HxX#PuYu{xHo|A-f@)M%hwyJ3<;b4>A3YI0D_1bVe?$Zd`;!li zj!cYOkzQFxR)HRmVDgmcFo2Hacb&C2YahRKF>Ps2BUzGS2m_cO%} zMQ)VaW2RHax!bo1_)H4S-g2%}2JIj@?}7-xt5 zgkU~s#c66Ho05oEGf_N;dC=O4viSCj-8dwZCSXs#BJj! z^m$=D*^6m2xKrY~XLh;$XuHATS!?2=!F!7z$rr4sEyd?%E9z3UO8%Y#~9@ zzwgyuI88@u%lVw~$?cC|Q7MCEHQ&8}6%IU#-u5X(D!bG!GlE;*yHoF~M_Sp5ttd;{ z5PHOoJ-N%w|Fm$VRz0<(;lrJ`O-?wJedXd1szXF&bL)!0A(vZuC9*Kgp886?ay7m= ziYcf)eQf}9+;qnt1IFZf^W|CTu+x_uDs%hC{V0}gebespD&dO{8caEBd}Vj%z7^N$ z6zUBl$g30QuX@g2QB76AF2x_h-~7|j`ghLz_lkq})^;k2U=RK{f%4ZVjF z!ji3CWq2XC#iv_zXc{n_Yr<0gknit@eir3aI1JAekh}?&bC~Qn4B9+}%Y~wWxT-8-TxEmlcqhU) z*N@HjM2?PoB~A4ZWTqk7Bh zONJK>N~+{A<)!k}Uh|e)sptP+z1?drigP*Pr$=iYIkI%J`U$jSX5EYiSu$)z!74^P z7AT3M^P^}vDtvvI6*uh1HqHeOmv8x|x`Mib5c@%=NWQyq9bdxY#&~7rhHisF(-xR$ zzPp0=rN=qNUCqj*Nr@5*&Cc!^FeFL<3SWC`Qh2XE|68F?Fo>618M>i#n(_6--)ex|8t!5E?Tc3u0$!Tax9!Whuyb-)EMlJ*;W zamn38c?O1ulyKKHPNs@zDe%{g`+hOM`z?aR;V?=BgZZGB!5G+O5v1`h&RH>L@TlCm zQIExvQ7sVBbh}k?cHIE{sHy0qJ~fzp3OM}mTjQ~#iajF?=oG)DPH&4u)BbOBr8`_? zavo$0%k6KStCKVJ-hbszu%^~a+!vYO_i_w*Y@gT5h#aBY=aly>NcN|ZQ!-|GSnTp< z5nG8%(d+O@s_v*iot%bRPL=lvut_s80w}opGm6;R9Rdk-{mDE?NBrs8)>W5Jj~0?U zYU)f&b^DeA(rG|j(Z$bA2HO|<`Kgzw-gv_lK8v(=l!E4NtK5iM)jN` z+SiUVYz+0(+M27zNg}ZsF%zR@)fn@YCFnPwkpjoOQN`m{fAzNis<(eRjg*8BT@Y>T zIini1tl-&WdOKatE5ld1S6c6#K=YpKjK&TACV;Fr&lP=MoMwW+|Q#$qx{ z!wDk@`f4m2qsBIeOyjRQCK@-mYk&btykA03g$?$jjf#|R0Jg`@HsS8e_38I=L{J)O zG12tWJsD7w)Fy+N-^*p;_kgXu660Svlx7J=X~=oBr8Wn(6)FeanGQXb3B}bPZ;CgQ zDSf_Y@9}*-8biThZ|^V917k#hfcLZP%hJl=h@zS z``!~^!ffwsGnJtCt)qjI#c1sM>Q^o9a2EvA27s2pU5URBPAa)FT|*Tr5ysoz`|J%d ztL=z`bR1{7*b&bgXwexjstI+ubLUmExUPX4BH^lK=qAb(<#LPcxAE3zye!LmZm5A4 z`z=SiX4I*C6-(Rw30k?=Y%XCb6%1&_c^q`3x>|1AqU-?rz|Lym;z*s5p|~^GM+0&%{EjY~CI|bFD{bRf8YV2=kkG)bqX31q=}Fy+WOru>YVo-QF#} z6$0F<3l9T;qklL3#{NZttE|9vJ`d9b{c>&l@3cjINf}6k13WKHtCj9AqVO*P+GBYD zOQ)Y2%)jP0~RJ6}cu-Y=g?1sF(a7>-?J;r0sL9WiR_xyFSR)9Tl+D9t0gmIq-NV|cHuTM7|9f-S8cFuq9LFQDcmO==Q zwR!Px<^_0e13+PRmeFGMZk&iJK07rQaAevinu8r@B=_pbU~=(y;~;TI)d#+V>dp??I-d@RBGrO^ zZsN#hb#B0Xu#EdVz~mpHCWu*A@UDTWx!z0Mv$v7o$7XgZx9~|q)|%`xihkrW;j6km z!8>cgHec|S8kyG2`?58XBzw%>j}^Kj4J&eMg2vsBykue}%4U%@(8kE=#qcxxzO*J=-G5H19z1r*{+8~{&;K6%84`|F9j7EFu<%eqGDT{a30 z3Y73fd`62_j%?l_Eg!G8Xn&`gDB);VII$*>`pye;Srd!~1}-Es2QZw$d#ACfKHV!h-(0{vs|qb@wHy5rg#D>)346?lIRsiSjDk%^txvayFi_|nYur~V zV({o>dHcZ8OtsjP*^hnF%x6)cneJs!Tz=9|FZAT4b6-MFzlkYosQ-4QOGgT$SELMzOmSG+Omt|r#Z-~ z>4s7`2_^|t-|D6?=(7UafkX@&+pxtobcqRYdW_V}61tS@M&Bn%?BxF8QK-%lN< zZl>C2d6S!Nh21Qa{W<5b`0=g#Fe>$FUKYdvfC$%#iQ6e+x~)v7FFs1GS;c&p?$#u% z2kq`>s1ulOk1enH=I|cHpW=fULoIEW`Y_TUqB!L`r#ekX-me-owday`{L}((dv8Ul zq~gtlL7b5{6p-*{kwT+?Q!RJ`7{_FeCwvmQb~P?q*r4QKIt=|Z`B5&zBZ<|Vn6LD< zY`6W(RgR)6y$Dz629?6!IHW$#BD(>9W%|dLFi9Tvq(;Jr8SjOGJb~zJ^|z`s1a&|q zD)a#+gm_irh0&EyPZmD*zPIctY-k72PxCR`_zHt){=Uh~xE+KxeK}Q;q8^)|Mnq^t zz`u;;ot`n>-xn0YaVaqP@LTO{&6jfs5fd4&`4EArfC@*)cEY3)xE*uJGK$bnOx6Zp)U+W8oYvHpFcATe?0U?eX;J{gh3mB+0&| zVQcKOo!5XI^yyOo0&9`B&9q;Wf)0a>n7f)l&-64rN^=bLI${k5_dxxr<;`7mYQGMW@ILl;S&!M>ozeZ{l*_NYR-w^U9_ZfS?Vv_j)G>IY_jQXG0Zx=ljM- zM6|Dy&uYKKin$gHRKXGihrD+{`MZNMVLW_uOmgIfdxPZlM*`qt|j>RkaYsjCUS%o>StNl%$Q>T$*@JuHrK1r zq*4<>L&KZ)6+26qx~|Y>YF0k19N-@6K?RkSx1;pt2aC$b3l)Qp#x-<>9n|a(C6Rx; z3~EftO83NOaYJz?%qDOrgb=fyeva`lwSFFY?F=wN-dZ@%h07HEeH&iA_v*V#l!z-3#@UsV zbt;BS#&P9)tLtB_^`~7&1drv{I~N#0U-V=_-d{WNQe)Z!E!n@1SmVAh6VyGR1%3R) z)VN1*g?x_gCr{Z05%KTd z!8tk{Ng9D=61lq3`39vUJV=Jdl_Si`!Kp19XT`?u!TNoc-)c_?It8yZc($F`o+6-^ zW=6{GR6!MMW(JCtZO^!Q?tQIfan+!m-5O?Nc+&Y`Ip^nP2n{@OH<-IReN#X~ciksR zqO{R*sgm}IZq9aZllKv30@?0v>}3{a2m5|6#Q{_)atpb@wl(;yNY9dH{pApQsFSr| zs)|Q{CeQ&%(QY8yI1 zF|h{Pn6;C@MB(X=WUVpiQ{ z6SQHg!+yi*Ioo-Ae?M3E58=8KzgMbboxjsQEGDnduG#SZ!=xZhCRV|dLU`t>3x0Hri{U- z+}6`LuwXYg(7tEOk)>!r$<;}F!l=18Mo8BsPm@*no>FzTNm8dC{Pn<@YLnCh=#{%y@M)0WMhpHCTYWwX z5?%LewVE}i@w)syIdoW4ew+MLZzt7cA3gtK=w^_o?JTOBAi-f=u6ByObFdv#nOA-o zS%yrf2#=o7j~iG76{%{^?PM9{`KPXk};wV>y0l2PVZ zk|T%Ipr?RV&WvKQEOHp@#yQrkK!e`USrb%)L?qXsk`zVjT z=jTWr*f&roA>mGxrq*rx4DFk!E3bB&e^tMl=Gc%^%jEz|-n@pFJ?np1VlZ@zTyo%V z+vI|0BnyY8x^bJjtIW}M19pyB_0_BJ-JM<;f1LwDHo2GWr>&8-JI@%4>?DY%g!I*n zSRS}NAG|NGViAt<(#d_;%Ozvh_u=3`XNy++Ba&@T&4h^ucKu|BetwZ(*+?nFf}aF( zb$MTL`AD^O|Do?Ap-E{^62)!hJYE_M&SUCFz&f&}-z6R)E@|`s^uP$hDM$T8n}#RkN~=vr z@7iGJveB;FJ(TGMVENj-(wn5aw{u|q4SUm;K6*($cf1vLvKV4M`;>g5BSMkJk@NJq zJcRmpW8r1S3lTorq1<94!11rO6BPWJ5*n$r7jQJJ42Q8`KG`0KP-`{h^ajrB2)d@u z=k6}I{}Z9BEyO9Ka)(QR5Caf|@YX)djt?Nfb3deCV0=8A4W)-4R9mmKQ;FazH#rpl zsZy_+hzX77Spvbp)W%$Y243y$xtGlKsi5@(SJrhSco*ZE_rm*hXUAZ5iG_-#QJ;k$ zPO^YVVd5?rRwJ8H1pxNLQmwe5EiZcwHmO(;Tem4_(;2~KuDfQhtg#3r-|CI5ZHSfi zaDvyEJ3vt-meaiEat@yl0#yDPFe3Q%O%Gg%6Q;-)gkqV1vES?B4~Tk>xV(eo7T)RvF#=jQhEeQjgr*ey7l#rz8Yt!Q5P+)G0<`~lJ`OFL(zqN z-01v3ZV1EF?`KK=f#5|J?cd3EWg(WUkmtkW(+F^21cKtqb2pf<)wjI!hve?jm;Ed- z7dEKYDKo6myW(G|UpSQf+^3*MgO31d*uD+AcM5^-@2-J(&5Opey$yywJ62KW2l9@SQ$#Tt8qnK4))yk(g+KWRyw3FF z_t6oBkLXO029xav{6P25kYjR+9?Qw{!j|ptqIi`VoRzBp>V;c6L(haxT1jE(qH!pU zDWKy-OiE5s%X1s*AKlNlx$i~T#DL*$4@933KQiZ_>m->250LY(K_1>*>TKlp^9W>! zLhJa^p7!uiqFyObj4GVUQ;)-`mU+m9Iva1aeB;sXzf00Q<_tKB2fXq?|2ICG=&5H% zlX3?Qmn!wqhz#AZN$tFc>Cc(xopANk3JoLWuC7G!z9#mR zShlwc$$M|Hf5pb_o*u}>hSLc<;EI(iyvF(g8EAa{V?wD_8{E{h(X^po+I-gS8XY-V z+T~i!xg9z89nTkzy%oIA4q=&B>WwyFt3A91D7%<^gBrcl!p0X38%^aAOXDW~S7WHo zx@`(0GaTQQ!i)9JbFWfKbkrlNR;C;%6OFtU5^YtPlJRP?7WkaKp;U6p>}X-C%_@`@ z_4l3NEWkA7S+;+bvkyBtpl?Y+$Un$=dflIJ*5Yv4~^l42Cq#o^g~kV-3MplU%j z>}QLorv)9@+2SYPPha@BH%{uD}W{5Yl4&F=J%6VVngpBkO6YbVFH6n2sJRo7d z73W{gckFw2$t>$Zot>_SAE-ZAcG-Ffpu+a`Q$kJ7M{MLlf1e5*qRz*Uh%2!SL;F=v zZX}4Aa&B&WeK1$~*}zBMv_ar0av=Y{a#WDs6%W&jRf(2ae?8xn7K}Ud`odk#iKN|u zXEGjRyj_Wsf%FM#{=v0IoUQ zG0b!G@qf-3u#hA1BERJtH}kR4r5SGPq+QpF(G>sP1ncKbR7I+eq{G;mbc`9xM}Lft zt)k~@8Wyvvcm#Q>2}{PkX?BPj=9lkrQP0RULcq1fI&CxJP?1x zRi6U``SlY>!@>aE>8_;UxxOkh<(dNCQIg`;r@YFb56!`4@&|goEV`n;IQ99Yk-Ci7 z% ze~~P(v0mh1YsW`rus9JwOe$sd$M15)K}kZ_;0{xCE=(E0+`pM6(= zO&lTbQ6Qhga{H(&!zJykcUtNnjg`x+)xS*ZYgm%kSom8318?yz9TV&2C=YLvJAF6A zb!C|$F9ik&1{(7(^sYn8y~~+r?a!DtPs&to1EW`ptycYd`0Qu>Odn<#oV?Z&CFy9$ zei1&OU~RFPDB-lnFf|27+v#W4NtJ+I)+_rc_VOY%S~;ZSiA^_*|!E7NzDbl2dzwcZ{{EyNrY}D^Xtml)y^8ELU4uDWotBwe19glKukwTktw(=Q=|8tXw(Ppus<$ z*CpXu*|?A+IEGq!CO1G!rq3Hj1A zWN2F*mSf6cb8VO~6#sB=&#FA=#?$`7RAS2blg_(Boh%KJwWfrxK*x~htI@*_M*WjR9hBt55p|=gb%56+2*&)=Qg_- znU{V-XUJXuyr+4(NI^p{BjpyuwmQEW_hnuX#vNy#g>QCAxzjK>e<5=3LQn462;glM zI+@HlCr4jU{lUfFdZYK;t*MD%t|;CHno|n}z)sB^Srb$R{>br6d(4&5l$e&zyC)5& zO@zKia7uOif71yZ)y#w~mb^pf3;U6pW-x1bx7K}%Xsr| zT~id|H0zNK0s$9j*y*d?`2c3@RzQXyAkuKrOw2j)YRq_FNyWtQ;4v<-jlH3K05`~Buft`sRPE_QL^=@UmR z3U&+53sMZr9c%;fGPm!tym;e~r~5w0*C7`s8}L}-$ql?TY%0?IsNAFG?)tC#qbuq? z-rpk>0Yf#qNvCn}lYb#X{+5suF4uO`73sSRyo8K_{_>0}78%SvXBL(+_}B|q!c3P> zUi#jV!c$EtcGx#;l80x{EPV)=@nMsQGCO~Ji1lQdT@^!9N){c}*D9PIPHyuHxUP(6 zyHk(R5r((Dn3pKx8Xnx7%g*#%`!PuQ)EG$o5(s>mu%LKZx5YD7A`|Gh3Vz<(ef!4Y zz~NyJ9gGuBR>K;q))nPUQ9p`0EjCYrK))Qa?=W@pXk~j9L!Nof9BjLs5lg#f6WNIn zYtb80w5>1I?8hCI!L-)Co_Ia6k{vWQuvQAGFKTpI%4?*8siZ>DPy! z`V>IF?4d@P5pB>;jBgl93-E&u%$0L92W*`y@<;n8)c8l|W9#&ZoQk}F0Td3=wnqHE z3O`BO-}gc7(srq3%%PWuT2rRg?xLP=I}u)I&^A=nW9Ha<74uu2SS}3cweDkiKav5< z|JeOjO0LBaVoYY1rPiB{(H{ z+nc4=I_uGedffQe&=5!2Bb;wjtvh4*k}AJ^UVZ(O2JkO;doAN;1 zqx+iArslFxPb~@A|NbfQLBg56_4@(Q`w}_C;ck*7^4%p9nUdg|7TaeSqgR-vV_K)h zuN0;}N(_oG&lHQTw3|n!wu&<+u!6x_IwxH2L?J%?Y!HUc=ZIrKEU2ai7zbDsVpv1mo zuNVSKvmWfU>*g)q*(UJSDxNydeDg4<_I#@JNzB#1kTKcA))L%K zCKEkj<4^uFk{{30XLm=1@IPBV`BnYk1Prj>*V-()4}QKM0K<2;;h}zfFnm|^Pm-;H zq}{iR;oY&eyw8*#pYD-jN};7u6bX0-E>T*#%HK&tWhg5t&(drM)LpFFFE(_>+n1_+ zalyaKm|;mZ>f4g@nh6JQc=vnLOQx%q^E$n$BZ*S>{2o#^$13^T9DXkJ`PI_x2x&*d zIe@ihe&_!E^vEkKx-Nd(+p0;?2B9JN##srCrB#zL&hK*4U$dmj0(f}h_|iRZ`v@tyJJlMkSv zifYd33hN)c`RZu^Q=G^sY&Nwh{r8`8&tB2X-z{R27aypnLbARJcyei%{P)}crs4UA zfx2)EL|BYdw_VQuuEF}teM;(pgBWqi;Ol=@-+bZ~z%TZ>oJSr0ed+!sNyw80aHN)Z z!jr$xCi~U>sFB(R)#EjTwI1s{H!$$Fu<}%Umt?>E8rz|Dj6(`_2LHC}8Z42>oC6 z=r>gD=Pf`5p5sONI;wtsr2iR_{ra|lxld{DaZn(wYrd6;&mX~-jrOy3gmz=?y(o)&et;A9}~Ja(eIqcs(1nYCj`;>WBB6Q z_Q)h*Y;xGV$@Mo@-l-Hk>8LOLHTgd_`GQlxc>%4quksX&=Hwk7iV^_XwTe8N%Rgrg`jkkQ|G z(POXdfjmx*aGELnvDJ|m_Z{I-G(6IAb0n<-{+sLp}+9L$783{z-zG?ax*W*9+vCd@{PdN?{RVJ$1SfFjN+oLHx1g7 z7Kk?#^+{t)NtK6RpD8ea&>V(EB<{w&nzL<5Sfj;@>Th^taIl_jut|^`&GqVmjrwfF z`yBQ6-rTlfMo?-RdPPZTkXZWPu${-MG{9Rqv&d!tSIG3fkcU@0{#~X;;K zp_)-m{QRVtfN^@FjHg*6)())X3K;Zd$Jt{FzqEB=?rqmfuzzq!-@D$$q@_aPknqIS zT1lK$@To@BbitqJLna<*$A;Uq2FG5Z$zi?Rr7QU!(dO{F4wp3Ui0rRO%5&YXbC8wv9y?C z*55Eq0k8I|o_Bi8^j|gxi(qgxE9UErC7r~fi&o1Lv zL2Xk!c;O_Sz$cUHNB&Z-h1u<{?Yb`Pc&@ASlRKcW{DMgla2|Ngj!T)y*{IOSn z5kh}I;)0B1P!N|`=zdo*?E}MU@b%x zyuu+46gZBo)G#b99PGh^0o9eOD}(C#QZ^*rZLs@PqtO9(V&r5t>9TVIbFX+(u0 zv$T~^S~BdbQmzZpz40Jxm0F_-k6Tpm{apV^d$3ch#YJ#Nb`m?EZHkv6u<`+8z?;yQ z9rGA;>bXyFI3OF2Xu*x8_#`&Q>mHQQK0vKYBSEw(@-xsc^iW4X+`^MYBMH5~lMM)1 z-8)XB%?7XhQof*6cF-o$XjGJ?tkx+unw;ab^Q2k!mg4Tn36`lHhpmglPgkGBJIOeW z%C1D>4J#vG^Lw%5ZIS*rR&uI6DXwmr;6$`oRt;0uCe&=zRhe#&o-9p_Q#W<+NiF+mf`!38b&zXjODEo2eyhDj z==Cw`ZbhzkzbQ06C0&U8;5Doe zH@}fqmZLwqlm(Hhm3~rW!cA)&jgs9cl&|9Ke7>tdb{)Ft!-ZaA8`)hP?nWf}iY$b2 zrS5E3)*Ue+I*?Da*&i)0$JwKoDh-KJ9(aV^tZ`~1hyo$B`Q&kwqNdylW0Izb}O&;55Jt7B!)^v+?juFuY8C3@+M z+}7rO*j%PLD+eRHKFE1^-01DYzz=LnPPB2*%gMP29}-?nIm~5+|Nf$zX{LQb&kPgV zjC+INai(r1JcI1(+SjZVAGc?(qoG^9L6=e7= zHTM*_MMs!Z>@Vu)-L=?TYawoHi} z(sp@4c}b?(lFTEDgl*DwxPZjRolZP1Bq&1SSpjJxyd)FbuqO>&q@Zm$3~;B8x8~|; zeh79M^eID>aYHb9{i|$X&~2+B8(FLQb={xmb5%Nw**;DJa?)mV^L&3z81=M5cCIu)Qvi$WC3R+)}{96g&2|u_~YH9Sr>csHmIk%FVp2tq1Stc~_ zQGd9J--(QJ4!L3xMrGghYW)_nc}rXtju>r@8oh+VOXVD%FVMD^RZyPWB{rv~E6YP& z0(>P~KekXE1zGK{ZB;RT@ZV{HUe`So5nkNDEq0$@#J~xX{yxWHJj>saef8Ta9=N|# zc-4N(<;2aaT`yKTlI%W3h&w$hoQ$Z17PpA|qT-UJJxt3+Tu2n7A;-oYv8Qc{%iH#` zBD~nnGnVRm9yVuLYm~dluc&;rT)MdgO)tk9D{8&iKbh(5Q&s`3SC`sbY&;V?DTL6^y0;aQZlZp?W)_h z8b$U&Uh@L!Vc^2Z#;r8dtQ&QD5Sr}|aoc-2we0gGS(|6Fqy(Ip5pg*VKU(%qlQhQH zc7yid*|->TqX2WwmF!@&Uja&VFuVOGXT>}|x(L*Q{Uwh4=l^~drDKk;H>`f3_sIsV!3nL&6UGC$l zu6{S#6Am>4j$|yaBins$W?lP!9^We($|yD>M)K)g*AuU;Nv%{_>%f8PI+S#To{hVq zdUuz!iUmB#xay+w&Xu7c-_7wl>!x0>Qfs&1ekH=fQN`~;wGHg@y~xmI_t2pIaCGFj z3UU#*(x#cAyCLza0P@skzpL|7wb%7D7~$*sb4K<9EHWfXJ)Y4GQ}}RQ7_hMBWj9dO zb0nRlpv*wqQoVw5>}Jh-bOFd+yfI9vb;hfFcv`O3*DXgNUsX%fphr`KvA%6ec{L=j zjQ7aB%N&INEGv?rNbQA&S9>2@pJ)F zC{!NNjt}^_9~b=Brhhk8(?BSm5mUFBtye#gw#}y9Xdtw%Kv=dxjA)uZnY5(=`cy(K##{i6k9I9xxPhDkyZ|;!Y=TZGFCsRNBKyzwSf&KIk zNR2$7GR}M0Urw1|di0=+`IpBp9eea~j|RLv?Z0UB=#5%whdgXLvms>cJGd>}j zRnBwBIBwgayq6NO(uQ!Q3;$dR>eB%;%j) zY9Kqy7Yh%pAI{XYz0n&X?|{xc^<{HEle0!DQbi+XZpg07PAA5}v_6_G-q~jF*hqL{ z+-=wHfQm>aDLCt*oa+$tGas%b5b$n)?3g6VuAKcyT%2_DiWP++s2?`MRTr96y!+`8Er^Rp7Q%-==_j2eKYz7r$w*x|G@cl=NGmirUt>SHTz4JI z?L7GsH<+_#G9fj*NuDX>y>@E8{A#szm=I0{EnWQjpEdng@CID{9Q3LGa#B?`;|1U4 zvoaX^O;15Y&5Pb@5b7FApFR9%BzaJsCpD)POG}_dt8&)~K@>e3Y(DnoJoeq1x@07} zvunE2@y=4{+~F$MemxJP6qiiw^ePjZU7=1{5w@1_ePXqyu0W+_sQV0v>Ig<=)i&c} z8GaeEuFIQEa*}pwP$5k%T8VtUS84VCZ?mCAfV+J>6 z=pB1SE0xgGNL4=q7zz( zW}RZ<=KA$5V8>0bd1$6KPcb363L!}x@@`+P8&+kLY35t`UhYuxvPwdwdmwUys~%dk zkXN}<7HqR>Ahb=f5mBk?R-|uKIwXKw1(c!?g7hN2*U$n2A~rxpid0dGBoKNH5GjgC3B8BVJ4vX4 zge2d?xyOCa`M!JZ-oF2Sum72lA7nji%{AwkV~p8Yi@_%by2PK7!)At&aZw+M)dNJW zgyV~KI=Z2Z+19{~LRJUnK<_FM9Rj&j-r+?6IL(Xw_ue_NRjmC*J9p*Ql5E6g+*67uZB+X~t_2@{KzA z2}yYg`PzqG8ehO!upKK%M#aB9cVM1&Wnq7A zcp4=5==<1c!1e6A_5{s<_r=!b4!#5wnMMsQS^@i+4!GcLIh?tT&*4VTO3TQ(&$IfZ z7)}lNk(!THlRH2m9Nv5|<{g(^{&pMt-?6VlSZ*A~xXrt9&9L*1_&&XtF7Agf*Uq>S z?Y(pq94Cz4EeR)7gq|4#PdejwRI`K(uTV{!%)h@LB`v?V!p_cDcU1>gRFUADsT_apF9K*8LRa+R#epb#L zK3{yhQ||6$T;?hP0O7-{ZDELAg^= zE}Q9fB7OA3Zj9Lj>9U&IBi-eT0m%cO&s*&bC7C>lJ5n{-UcFHF<50^qzp+{9gc#12)c5dOAl+KQ1F-93Rj_YbIIqF5o;geP z`e71V#shzUxy$du=H*xUAcSX7Oq^^vgH&dZUod6|OeOe5@8rB9lvFa)Ktayw*f_&i zgStWzH5tpQo#*OxO?N;~Gq{%jBG~}CNG_~9Y#CgNu^mcl#OkBz4CjVZKT{M=`@>EN zA|kO#qAmDW=cQV-AyX<*v04qL3M+eMBNeg*Z)13uo{1D*+hM z+iL+{)xlec0?80$z}G(dIEF65l_*}>#M7CkZ`_Bv!2RO%+dYL6%}(+61;balCVT-Q zr-IUuA=*Wx^}RxzgDRwH&W$u?kUuy86zyw|5W zyM8|d3o|d5QaNn>=JJiK5YkSoi67W!Y5=}>;uI>W5`!Ytw`XFOeVWoQwPe-qvQwSz z=+aF1aiu_YaefTrwUu1vt_LL)LPxzTv9o-0XgJygSnqMSL!GgIMuq=$ar?7nK#3!l zj<*BB{#mb7JGqDBQmGW^2v5!6>(9?R{RQUB7#UOA#zNs(p0Z@5YHFr^{ZhTVn6x=~!i}h}_9} zueJ4AQcAe7E?}K=tPGl{7H9wcRlM^|Kf9|d2}Yj0wNpHRL*?D#o-P{?Vu=@ zK2qE}ezT^N26ho$OcZ^M#dAv6NOjpS)M#EO%KHy>UWO)!ODpD9-KZMpr74e*bg0%p z?3?bi_RS(60aeL0x5;acgCBHMdpi;(kgK^6DmF=L;#_aeT3{+jmn9~RRHG~BRAe%` zpJ3^_Gj;;!YPWnVY_WUp@f-XsGzMtc9)*MUa#nX*6t4M!2J7eMor6qwL6l^{zsB>! zhE92?&w^iNoS%I6r1R42=7pzO(u#@^|D8}(G1+6&66KE2oxot7XFGz8ZY^Drx0dCq zQ%dY9a*y-kE7C;gfe_|F^*8BqT5Kp=I141mCEM?eK=K* zg6%H=3ob|P5L#<3z2Nmt*?l~@{=lTOMVgoG%7yc%2B`DMJZKR%p|H5tm1Faqah2Pr zpu*(@Ac(k6Uw-0BFstinW2wfYhR$ISYAG1;%Z`n}$tr zog!u*Kha5L!v^sxU8(#UPqevbe#z#)SMdLp!2J2&Fjj{h)0=bY(OuQKw^ziulwt3#_}+(9<^^;{D3-4#id9H437CU+uv;Cg&*In90S zk;0?!W+2xRC&E8fj#2Fd&!I_kM&pzKx0f-+CgXs5%Ti{Kw*4W~3$R?m>JQCwT{(-R z4uWT?(p;{b;9!zcLEa1s(C4HIxQHcPNB_olVlrKZJEpGu1bX}XRCG$+Dyx%8L#jd+ zkVtGN?K;bSXX1+K&p870%Zl658O@txsjpG#-rqNe({ue2F=wT}nXzN^3f0fGI;Bgw z(w-hCpWE!`Ig(tzjZz*X>kk(Nx$jtgtQGiv-A8mLb(7|rqynAWWMIe59&18qp?r*R zA31THYgpJtS*_>Ru*cEhcuuu9n6PI;qR%c~2~T_Wk%{#gtF`q5Em%v>QO1UXSIaQ& zJ#N?O(sU7@LrNjTu5sCBVCnVsgP?T ztrQY>gSTAhE^CTHkonFPO#bg=4RWQZM8THyo!zEQBnK9*e8HM0qkT_$7?V5Cx08}} zjGL0em6sP7QuyZ6$|eE6IwzdodgkEf$~yej%vZ&~hGWDIH#W^b=`5XMPdcfLJrC>r znOs%tsjAHoc+!>&P1V}lsNlJ#Bvv(de0AIDw{%B3KE!NZ?I!$|BY*GKn4h6gX#HVp zvPDpJKgzUK5Y~T5cPJN6TSzRAL%8eiP1^G;2)Ht;y2vBiDP~n=f#K6})Ys$nypTdC zvS~;rI1=%g97y2S_w{D(mUADq6U6R}*Qn6KEe3SVM$)(u1AMQLex|oD^1YEw_InB~ zg6PXy&y?k__)8goel1e~X$lQqiim}CuM0g!X-#m|Gg{pO)*lYek59bzsbc|*tQc!N zDXZ#3HTDfP&HVP;Lov=q_;AGhClIwUJ=i@}!l?KMR6b*CP#x_$q#;-}Sjv^hEvonB zDb6o~Tg0G?;;5JKGE)iL^y>)XPa>vG#E9jM?PPqEc2Hx5IFKlz=u zYLc3Pbc>mV8bw)?o~esFf>N&r|vuP&A9z`{1FbBd}P3!OTa`itiSeY9(xo3V0f zz25>s)VCa0kS~h)PMgQaCx>TO`nW7Nl~efg6sreXdW_}3KLmo751G^x@JlN#G6;81 z$N(IUKc|b94z4<6nFxA4!hj}u>f~||ZEC6%ga>>nEMOfDBQI4LJkmY_LWWn(e%$s!f9(WS zyq{HC`CU-;1V@@6k*Nzi9`I7+OUg*XNjQF`5s#Bn#73$6hCq~|zE(1kN`mNJMM3(u z*eRO#y@;Ca&ul4%k|tNmW;vkeLJHYXEj>WYz?u_$(f!B=!@SG2Y73@}`RQ4-J8W8% z`z}<3@QJkaUNTsqEJm1*NBnj$cBMFNCLP&bM=ZOv)OQ?`Y3L7KwiRgG_&}-^%@A8D z%=xj4Ff7dZ;@iR%TL~RgFwj)IZX%BDYQ1x#wf(|QeT2vM8(%t-QWr!tPpNz4n3G{5 z{Mn+sq|Sekv662S!;aIB{LHO5xB>IQjS$|~H;q)vU(ufK{^NNu$G8pI!uvg@Qdri} zfhVn}hk#x>MHcrhb((F<9wNHVmy_cTAQGf_Dd?P5fn=9YMC%AA$b*$jBkW$Gty`Hv z{P#}X|DsaJ7QO5qnSKlR>D|bQRcwiQV!B{WIzQSkN_}3@JTxW4w{*&8Q7qfrc4iy6 zaN55+fBb2&{U6JE)dOLvBClH_xwq}Vb@3FME}{!g@+1-n*Gme0IG(cqS3xO%lt}?# zs)n!K{&-)bC@@>=jjU~J%>eBPWkV;h2$qlvUekmThn>X++)4crU{=0ddz#_B z_JGGoM&EYsn)G1<{h*jKLdLMoXen7xu&U6sqO6dTp5Q4uQEQDm0)>a59o56bzusa! zbMG$H5p-qm+yjyswi01#e@ovep#NL6;TYsGw0|uEXRu|`zXe$}W$zbDARk<57iiIr z1>1l-sckW?dRSe5O+qk2sJ<2M-Xt9JTV<$Ig{#M4oqvb%D8?NllYd`E8~c$iMe-uw zD;z26*0p=?jgS3d+h$A;j>@(J%n9B7c0hKmk$*Zw{^#|c;N<-7R(<)+k?KODjqzRf zFzT1ff2NBHh96Yv!G@&|m>F8wTUTG>iURvLeNJu-ZEmN@xtk;J7l()5QqjD|t93@g zDKNX-aSw4Vo!`VRR$I-O`D<_QX;kNn$G$sBz0~88Evb7L(z>1{WD z>$2qc0>qq}%vy7Vz8ft;BBp_F_s!N<%iH5IsSo(<#D)r`W&oosJ532CqJpc*C8(|{!K>j(S0%SsXTwEhj%oa zUBXrKF|Bu3-@LfwjYkD-uM|hIu-{v21coKJOX$is1XdL}FZN{XkAZUuq+!6RsUb?x z*wi!?$h)i1X9We_rRw)eq8Xd|fQL=c+ZgT~r;)7ZTi%&V@;tY(9omz76H3FCx*3NX zHn3(wH0*Gx3YOQsUX!FO2L?3R`v`_A>P%Ot(n$quf9wICzs5s=5BvczF;6~xi%JD# zobWkOSmU_3Y4SHXN2Zc&0xa_2y*YjA(rDL$Tu4m(4}oLqoNOmoBW%Xnw@`I*yKu1^rC5~4-TI}hPSq^&G&D;*z z3I=%n^?(Q$kR6%1)pipwM353;$mS-sgp~2+n1u1rg0zkJcw{f&qPoeDEqd~oV9=j2 z@E?b=dVf&U@z^hb>NUTVE&pldHIyiMn_tPLluak}Mo2AO+V;hX&~;6u1tw(2O+-@6 zqd=AF>|SftmLS2limA5#+U0dyymoiMWP4f{99c6mji?Y@@iT9U*~XyExKa(nb|6UT zD2q+Yh!P-QdwE8pWRy_Vl+G*N_<~z3!O{Zso74v_e4qgL`A!M#%ykoYbO#{)T#$MQ znJKJa*K+3ER_QC`0>*f?n)@$a26txynqiX+^SXT36xr%Xk-6o3z+^Mp7k?9SOA3N8 zXm2UL^b^ziPjJ}3ufS103R3_F?QVOOm@>U;I{5OUt|?^QQ!v;v4n*qTrFS3OEiE2izvM z*V&|u>S}zsR?>uZ1(fJRHU-dF0Y9<|d1&8+BheyMhPlti74gk5Yk{2DY zDw84?T0i@|JmT#3zCv~tkHNduPadIydD=TrtKl11^1RBU3qS0P4mU87gJdqf@Pio^{W zMAtjUT?wh&S{|fs6aeBm#WYxYy!idbdd4J+z6=#m-WY!ZELVN!2ba$$)==~O{CjiG zjIm%Yz7O3rO9A2me!Bg7Cuws1ksq17Q-dt5Xov+9qQAlSiVI@|R+Bb&7S8)fI3a$3 zXru0RCeyd0-|oS?Xvy1KgB6Vm20qyJlTm_hfUYi9UYw?`{=3L0tB4~YA=EMPyJH*5 zL7jY~vmvdj*aCu0mi=YxTr06AYN38^3}f_kg+ke9+FT7WmXL_qt-mLxaC;ODD-^7(hkC|~?mm}G=HtBEdv;&C4?ZHXyz4r1PYYe#+~cQ&e2Cez z2DaJ_nnd9&{po_c+MA2FtQAt>e=9}!d7pBA_hr)5&##UVnq@PD5SlYLNS|W7CpUlx zP0{+ns`u@@GIVGSMZf_b(-o~x;Z4~1ky-4OdqsjFfXR$SkZVb1SKoHZIci&xUG#a7 z4{9PHE9dos5Kk7gNWoM8OAR#*cKgJt6C;{e&X{pWp!hu=o;cwV=|ilg`6(6J+`*Vl zE;*ZN>{2?@Zc;_E5QuFjV%=5($FV5crxL-W$*Apgy9t3xpSiP5*~AJvzoV$R#@ zWA=J<&m0A-hmD9yqT40wcBV-E1Ebo3dL&r~)_!s=)s1og0%;AU9_p_@1Y6M89rlM7 zTTG6Uau8IgML@P$@MPp9@@5+mz1*`UqDp(U{7DyIFWVG}i11qy$=#XQH6hkcH!ZM! z4xNe7lx(M0A&pVg%;Fe0vF-!%G+r1^ug2ej<2^@RMF@rbn`Uu^>Xh@9C>hVTqk=TA zhPzm>d+hP^k1_y-EMt^2b-qkrw>*k~!H1?!5jGx&9dZTSaT@Fx;|SZkLElxJpEpgj z=CP2siveZgc>>i7mI12xqW$2gdvd#1T#7@_+)8N^G`EUxJk(;)eQJ6u6~5OhEIZc^ z)sC}kOrfVcdi_B(uM2*r9GsF6VSne~)=AXVo<53+c1LngkS=zzW*mvCYQ!hbq`&9* z7E_*xE0jOnPl)hdOsXIxyL0BS&oC(qY>@b)^`W{EGWb@%J9%?PA83Nk?~6@<%#j`9m+ zv47)#Vja;H&=AfHhb#gH|2t*|$0#F`W*49Klt)kqz`7q(pGUkyX|=KPg6Bl7#0=6P z+T5AwF`u}?sRGumK+yTVLZVoW^F7uM4ww~A4j-b_zzNT`gQh8dK5Y*8)#;Zz)j191 zzqho2>zy?oJZA(&8cG;kHdY(6l02P15v-*>xh3R-`cfnMkforG%^$g#uAL!$$tZ2c ztYLB@B;9x5p>!yZ!EDe{nPWZ1CRvUn8r*v9R6!R^JMIVwWuJWs|3hbpFEZ`f_IC3b1C-~J4ibyVy^eF5K z034%f3O>Ys1GqVt_$C|cx#1)10xmbx{V$$Y9Ke_GYb+Msab!llogg&1DMBYeO1#rv1( z8m{>%&ob?mkETIAAtMkK^HALi*HOEhxC%Fwpesq@oOyx+zoL%Eyf$|aF?8XAe)jquGkL#PNT-6|-MUWu!G|CT19cDdnf)OcltJ-FM*cLR zng|lqfN<_u%0+%*U?ElG<<5y5YX~o@NP+P~Kw(bPv%|i(!Gg%C9NQoj>nFYrgPjK2 zrmL4L&9y-mHNX3%jg-4)v2EXW!eVwlY6CvI?lm(XU#_IG^9RQVn3cMihfbf+UwD|g z4%kJQAmemi^hI~PR~p1fVm2RGM5>&%FWS&VS;LO;XpK*$V~)jL85R-F;B(O;(RX<) zp42O;8L2d|-v{NxxZ5HoJn}KWbNXbH0?Jwi&drtvcAV$kpe%2{W781yoq0p=6N474 zhdI$$!x}qw&5Y2f6uu8X<`8d>3NL^3w}|)k4exu1g#r z_!hYh$uz;nhVSXJ^V%WDt`{tKOC7wrlRln9J*%v=q|e>YJv0y?)1YcaX)(PDA6}o` zPDXDtd{@t}y=%kCi1>uQL~4aBUQSF3xv#E`pA12rs#V`WB;mVNQsT|7D%Zl5ymLxK(nkX3FkKy=jwZLK$TSDoI|an(=Sj+eICvtoTGcGk4?_6f#Aw`+S~o_Z*= z4~~O*H^Kup{&{BNj|5-A>^7jOi-+bi8d;-BO&J&KYgLCqRt0!Dh?_@`l)VEO%v`KnGojuS^g)T<*}kLGGMv= zvYwgiayWL9I9lmmIS#|wu8x$gVe2XushM4&n0h!(5tlm;80GC1-H?L%9bDC!Pn*x^ zK1dkWOESYoCtp$XQJ&{h8fBj7r>ClnV(w7x#E(e%#R;H1GS`5(NZZ_79ER$DpwC-` zthF_OlastM*m9Psc|^h?w~vv*5z+<)557iCMqv2Cz$l@NXQ%iCi80V3vzFyu%%B2G zruXV6O>vd6C$W|#U%$_1#8UQH;Ol@W_wVJx|M)poxBWSzooL!Nnm8iP^IF{C^J|V* zbFFd0X(~Ak4~U8(#4<7z$St-vVM%H#S6z^*PCL$&SLe0~RH6wFl(zF}ZJPETxCI<@ z(*m8sOOM1Sl+&MGh0JoBgLYLT-EeUv*?=Vj#(gV-5`&B(X=mEvn(mqj$%-W-sxt6- zafPqpF>K%NII&8yK*k^kRmxpIB8>_R`_k@)&1z^LAy={PIt9k$)UYnSNlxErGc8}5 zNf!Nq6!}AW_HpPA9igS-KCX;GG=;&NDupX+8z9OY#(=R?_*b7g`>fc;E}-VV(CWXL zmT!hku&}@}@A>L-$!Ujl-Ce~#B9yB47H~YFa*;EM-vj2;m!S;6iJdVU(M-+SF6DTN zwyZ!@EuVYQx5QT^3%LPZP%%cYOufA1#+sp5G6wmb!`%W-D_OGI;Fx&`vcI>pI(HD#R zd!1%cL-C#pfB(U~=4TGifAHGuM~uxY+k1eD^ZtvHSf)P23*^97f=RwpKjQwjauw3z zyvz^7d?Twwi@>}*Ru4bN4D^d>ImKKOR0`_5qX~61Rf7PcFp$G*jt-V-VD3P+1Z?nA z4lALx{E`{-qw=pWd4z{nEAb&48OwvndEsNf4V~RqYZzP^FvtvO?a8h;4&bcC6?xhq zqIM&Y{wTvjH{3JGj|g^1HAdsjMf!9sn0_{2hlCQvb%$dbVoKN%NxJf!fgBZ!Dn_}# z`;{H=-JwsmA$0i5aA+9I%f&#fUdH+?mvYnOp|(V+494dl^WDa8fIIvOL?(7X#|!IJ zMkb5tP8Qd@Y%A^Lh_mH_NmKSMrEx`$z7a#r@}-Vqbeup zpS1!@KgF`+-&s~?En@(?8qGZ!kc{9CER*I?kdU(TQ5tcXu5@bk?7+No7{Zp3Hdbue zpPVY^fBUalgMZe~3OYjHMVW`Q`CtcGj_kQJUCWIDv9%Nkk<99XQq&81%=|Nb74^*I z*`j823l5M|+%c!CfJtJJc*x3YGQwq8-eW}A4O}; z0p<~v7G|W-{-yp*z2a}z*EX;;tK2dtTVOB54%}+qswq?(O?t{1;^9I{?*_FpH%55p zkn60!zG(G38ImFQ1RnKP&#E!=fw6WQ{x{WgH?YcY1 z_mOo4;SoIW-kvhywezs?<~vpy;xKv;m@aDr6pfL@1jjahtU?P4L#R1S)6I}>0&>We zL7=9;VEw@dz2kjN_2FUkiO1Ab~75IfEAUt zm)eHqwPq;{-3e)Sz^DMQ>d6D$sxF!c*zrzViecw_4FNgw7sjw5yzVSt8q?sC?it;L z%@(%^R<31)oTq@8;b+8+x}ZISNtp7hka>8u2PWPly?im4 z=hA9RO{%bmSKQNW`YWRA9!KfQy1q`CbtLOt?Mut%|81@R{n~kb>F^Ycu>SD!3B!M` z{yM>Rq^z}OYhm(csYDEToHFhmo2TyM=cuOV-|k%a-n@IH=@1J=o5yGA5spxGX~gYZ zO}$5^I%q*)JWsua2q%T*J)l;_f=XUWRedq&mamOuaiNaED>khJopq~$z~0>O_2|hX2lHmU{hD`im|Wgmpp2{8q{22+)}ALc{oMUHCa39hRA7U zhRf6+_%J3y2hKOAB2PUKfRbq58`;Yy{p&Waz4eqIH(J%;4pYyJ4R@Or=QOCy+_jfn zduwdl5~F0}eZoka6$fKMPF(C`mFN4BCcNZVvW{azh^qP0;N5k4H9vW+g~5!yzC*Om zJA}0s$GT@ZpEwQ-D|w`XhQ*fMed`MhX3#E);f{K$8n*;l**W6tnv-&CXRldgpui^O z=5%8f>)*lu{|nss?#J}wsw?cAp>4uDvmdR|69J-xbi@%DW?9mt_R(|xfaa6iqzI1N z`Ca@zyFH)p16O+A#HX|#L_Dr5`|g6b@tM=F2+3Ekg4!V$3}?N{pPo2xbb*Hy(^pZw zh6NIQcc;FhOpehb7l`d=owKBWGtcd!cgW!$siY!BMUmD)q4Vd*MXU6jyXc+}G;q@Z zCETuxq-5tA3FAkAvB}--v&>&U$uL(&Q-jf^fn7A6>rv1q=;qV2e|RbImsKS zV{B|@8DMO(E#Opa@B#SIw~J<+X5qJuddD=x)YB$~a9CHbkSJ|v#I?1cGp52uEXvH`PQ1U;NwC;5v121H75DSY9L zxk`;Mn_nr~^sB|H_vG$8)%U&hKv(MnBl&$0bh!HtweGYY9SB*c_%abgz+u^3&IzTu zhPyroUrmzw)Vw+qr?ZaS&uOFHak+DVFlz%Q4c2dAk%JNYaJ{8;16qBQ^=58`qp1^O zieJKqPR56t=yii6ApN=y-HjPHr%ixV!y;xATZxCyN7Rj+;p1||7!8)o6XeAJh#612 z%L5Q`+N3IM+s-iV9$X!*h%Qh0jhfb9!(>r|_&pkn2TtfBqIbM7U(@1xEo{?@YP`77Ac`N!nJF5t_0HcF=ov6L* z@1Dl!lz4#i-pfpIm2AjpzE|>vw+bkd|aaO<2R9%tepI}*X}CRom7sUWJin; z>&Q5J41GHV0_#s^*(4Gp68D=~&4naR|KnyU>M@E}{6!4%a=zOY=@1(NF>9QA^K-oJ z$Ibe$V?n&e>}M3K5W`Nay$D7BtO~2?+vowvJ9m8XCl!#7L%3O|BYqs{PE8J;I4j?E z3MxAYKTBIe%3gYJ85I8`{wDl3+viXKB+Plyp66Q5et5ak6;H@`-rTZGj}4%0H6yu# zfk}KtjC{m+O?q|G;hJdcZ9Q3O@?tHyS**-{21&WfJ z8P@~V;)KGF9ghX?W&uFFWUI*L7E$~Bl6h?)m2@h6{X$=yK4G}okbIXtGmpFsb-6sWrTL$y005BB9c3F zlS7%)H)!evcyH&E@AB~VH7Aq0!1A=fx%Zh{X`L#AjbaH=vV0fWqIqRX^mReQ25Y7V zLm@ZVIaDmB+9BEgQ(2mRK&~+gftpT=8H#BjoHj94Hs*o5wkAPBuZ&9rw+>%pl}BpP z(o_PI!iVK>fk^(KcIo=YDug?OLM6%q1(p7e|Bt@BlSq59A^6;4?&y4+@Tj{RBOQi! zO)IP$KtKwdfY??Hq!$3X%&3=f?fFf!fBSI6bCQ}(KnG;n&;btkRr`hH>ina%A;f5d zT!k(KFC|(~Cj84ZkCr?2kyMbz8&sebHf*0}%c$}wV486PW(-%N$F{(V{zZa}IvVA@ zXw351wLQCU$EjcVbEnz3MIS|&ss5w17}})wc2?KVq157W5nnv0hvJU)^pcr77jrX0 zD;!a8DW))nWZk+}sCG-W=rU_Up(pv)vM@Rb@nX#hksfL?h|<(ai3e#O;c9vTprb=# zHs5x%W_Zjz!WI(`Ce&bb5u)I^^1)yZgkQ+U3AyGjlXdiI3FE#^{(f)pRQ3FxcX#c> z>i~k3w4c{(!d!}|%*@)&t@!|%;vnOmQ+QZ?DXiM!Ll+I)svXniO=;2v4+Hk4@LTX` zHEih4qmwGVvDJDMSJ?RCq1i#aPuwEUMISRD{g9)%4mKz7;P5BQ3kZA^_8xZlK^lu0 zzg!&YvRtXh+!$sh&E1YNm>D{2df{*H!=sOKzO0>CG6pUo&~Kt?EIPnIwqxU^rW3b$ zUZ>3=-isNj^hBCK;*I-D`qyV#ZOBz0D%DX4#NN7C;SN+2`qdQD*o5Q}>k3Y1TM~9- zeFC0QUnW;AZLu07y#oEWe3)@TReDQ_l6$Be?<;O*W|lOnU`SDi6`_RpB{ne&%(5w&dLP(EKs<03~!Nyx+FKKX>XXS&-FXW#|gd0 z{7ADa0SKL8Q4Mv4(^(e>2hw0GfCZ{Ns_O(jUS-M*7CE)qdz` z3t=2lQ)&J+)p4)FHM7ZwxRsX2XP5^rlRm*V2qU<_Dcf6ITsIumu$n+35L{(}TSWUu zAWwk+qj$!tN+VhLM6R}n1}$(apAo@fEXAXkLA>(lxKG^G^_u7B1M9ca*psj5Z>|hT z?3@iOBwn#N?!RyW>^LJOcT(+wAH*v$t-e>S_fd2dF63I)Ec-#)6K#f=WEnU?ywWkW zy%>K+s@!&2ZYaj=z1;hYu)+c5eRE!W;5z9v0WjZ?_Zg@a-u+UOEhZiE5Hq( zOaO*uwAA}?$LWs}md0_LS||~r_sR8b??mDdv?eU>vsKgQmT%VvNcGojs)3nf0}YY6 z+%@K9{>2>qe#-NyU3u%XZT7%r`JqnU>0LBqhHJFO*Ezh%K^xji0n+vDgDYKX&(G)p zSN}Nd0!~->9;u<09J@l*-mSxC*?61MOs5(!Yji z7cfX+2TNwpPK?~HCxu7TNH<64n3`E%BlCJF(SsI5q!O|OR+jUD5-0gYY38|bu&vW- zu4=Vth~W4pCc{DRan%#HW0OWq9gWx z`@QDX?t_O`E|X!Z^Sw`=3;CF51_z*}cyRX&YE@Cgzgj zoVZ-T%z3B+_!ATL{i+({oY)6Om(pp5jg(FN!Y|~JUP#q+8p`(DS1I%A9+=4`Nl6^e z{*lF^z;XUZYzLWNLUv6Wpv*~P5X2*HF#`>0gVdf0ZgrLEl~OpfFY{{%MA5hxdZ!uhcXze|SEw@T0H z+e*3m!Wg7zWu+$;kzCKQpXnZ40j~c?OYpwk(_R@$Fw~iXij=DRH9n4*;37shIW&@r zPX+)b6K)vJlR#7@i5)9`WlN)M(RT{mm|mpEi4v$e&jp6FXUiC?Y~OfWb>jjNXzI;h zaYy9*4q{Rxhz-=jmU{As=M~RsO7iI0P8p*btIhKqS$)L5^A`X+bi6KzX2(1n=gj&* zbRoH0n>O&-9Ea%{sbm_IaqiRD_F(q0eUo~pgg?t`IyVD87prGW0jNyb>kC7$L0=Ko zjB!p{!}fC;YaAhucr)%>be!AwEE%BolYyvWmkb-=ceb>gd$V3Sa9!^i5pMal7L&(k z))^H8saMPDLZ+4x$kT9lOo-pC%;3BbP8|riHbz#tH^@>{x z{EB?123v0ZiO2oj*jTLzkbc~1Myq#>0n;W|UG{UN22d-?60-J+>i}&`gTmm8cqnygvbL8ju?H;QF zXxvKVn@g@eJQh1$yKl4ax&l-u0J@b%u8L+1qRd7+M-Z$9n{*skJu#q{gT8N>!G=Qug4`_~S^&mH?v z5)gW~xgR|ApE!&E`t1JYEBx~hEvW=-`c~kfXZinrO7<^%_4O>^5?8u+=5O%g|G&Q* zU~c~3FaLj^{Lh{Iw-5XN{`>#=wtp?vc89a+fK&6=uSGoQe>sr9vY3ya9^l4azZnnu zFGrZ~w)-UQR$FNM?BDNI07Q1EC7w|?kQhL62J-j~qk`5Tj<3c|y z-1@a7MSuxSURgV&DM^j@&pM3ddP(|ey6`|k@3U&yqu5=0eEp-!(tl^R|3wU(m^ijS z#34gGdj-1l>}YzFbJ5REnd7Oe4?gRaw5$%J<%U;X{`jAw)qi?TULLqzh?P71N_(vj zeuPBEKAQ-rI`Y$?;PsWTaMV-)AhY}=z`G}Yw~v45S00j(hu&dXw_$7G{@QpiWs9+P z1=^1R3#-&Y&p9f>SSEsx6)q! z^QfzOG2H^DTc-@Vj=Zz1q!GZzvNa4#5tEWf;akCf@i71WX?*897}){HYHXDJ2xY*N zQ2%<9Taly0JWt_z!$zM&@y`*4Syw+WgjzGCGBVCp+6*TL8 zMY5eKd+_yTZcR_X@AAMzqr=3X3unbdIhMd``(p@&L%X6=Br1FBh&c4%jfh`cz}LI~ z5KRQ5o;hSPG3RS!svCCI<>sDL){PZt`R55$En%W?#Bb7Dv`U06e~C2xMeXGGAid5; z&Hp_=_>b>;<$P^_+kGzvvL5f10-K4S8~K;2i7yR+33P}@s=|MJxBklKU)wMP8k-@L zmM=e@yng9RghgCZCGAbfd!P9~S~{T?U_v~$xH2Z}uLb@8)9SF?c|5V(;5jG$KfU>v zuc^lDD{ySw%?Rgr=rn36jd}iwwA~eJHw?YKEZhB8zmX7t&AJi<=k`T)^Y)f~f`jq~U?x5oa!!Hg>^*LJM~Q zU@I?o9yBJq;qoDE3By)K($Q9g+5iAZ34cZv@Efu+1d4p5US@dzXLFS)P{WV*syhJB zujFB=-L8JB-%?(%Fq6ypuRhK&anA4e8#Yex%8kVEE9F1<`XWcq<@P=$W%NAHS;_mg zRqb)|8EbQE}8VvwOp=V*isT%(;GsCh1) z02{SlRJ=p@WVr^+VJZNa{IZWo6Tc1W0V)S1v=X&X+*R6or~mdzr@kUcGs@BEO^<>> zX2@oe^cjq6N?HJA(NX^9wtSXB-h%{T-SjTY8zMF-=qu&6z?L`wE3BwiCEP--2m9Y$ z6o@3YU1NGLQq)chY`#W@l_98G&k&xTj*5`mn{RTsyg7d@s1BP7nXq6jJgvq_PVGQp zUkP+*_WRANL=2;%o<`JO)w*pYY)dI!hh!J()d#;dd!PtvJ@}9HLO_vrHuxZH zRGF`Zw^Pt^`Cz_7_s3&T3!igh7;8}~w3X5dhwh8KiosR0-uNb+`E=e9sqCQBBX5`w zUwf#EF|`6no1|)Er>1#8#N0xuFH`-c%o5mpQm^IEO7HwKNxX%aadU606)Dr%jbdP6 z^E7y`3z+3P9|iL4`Hh#{eJ8K})pV>-H~Ojp2-wkgT)$QJ}`?Aep7^jKd!MneLP|D<7Khs(UI_t8k&*|abn zb2Spn{ab8#WK?H=6y@VDVg+A9nTBho0#Zt`6`h0T&dh9Q_S~D@^8R>!$`6K3uV2~) zMt`qP>w=17=2PP=GaLi~WPD!CVGwB|*C4r%kxa*`3*3z+t<_nx z%Ou5eXW^kU!CeZWz_evXamA>xvAjPvU8TN}0t`ZQ0YIc2QIhs~IUT=LHZ$gxek3%d zv2_HhJ)fdtAJdTn#!ms-Ns8o`hbU78jKp5Xd(=lN9Zq!tq?94PK=XBmoH20?g*|D@9u;c$87PH^uiw?0c}n^*rQI& zY{3^`poP>0&%85D*~P&sLKz|Mf_Wo!0+7)SNMVvF{nsA=ZGcex%hMunO)JlpXhU|T z>K88Z60Y1^s7I>9i;I>P-Vp9*aTPu+{We!)e{)bsSS9Xx#JK3-S$ficIzzx*vqeve6d|{2`=}>*A z(#>xd0^R+6ohMG1${Df)iikR144{f8)C$)EEDjds&MM*V-7uEUQnBS_MT_BAd{W`4 z29U_eEwvOx!9%ocgKT69_YVRNnaJ@*cxVAsOBDSYI)ShdGf=VK@d)}6`4&vsnLZpp z#a)J3dLg#5BvyEX`A5XSXe8B_@ImAQ|Etoh6rmdGOv)pCn5h?1IXBISqBO^q1$eE=Xb4U)1faw;hWi_t%*DdNAkq8x$D#H*O)`w?B#a5ZA|7 z$ci&FB?SRZbwJbQ*;)?d81?9A!iYgDwWAWxUPC74DclJ^#1ZTO;JVCl!dG;XPdg@d z)mWm#DfX3a{Kkcm#gb3J?aWj_)58swzyP-@J@G(A^^fU&jfvX>HPWlg^74nkPjpWY zO(P0hMmAT*$I6|r`k0qHL6q`r+w{I}E?&1w3SZ81M9@(dSM(N4xy}oOkkpj6lWZRJ zGE)VnnACPm6Z3w|vnC0b$P)eDebqzZnSvJ+te#>?srOyO*>36riV@f#&{jt;KIIMm z@G!m<^3^14@~6CFWVvvZVd^jcz@*+faz8MlXTR~T_S}_}hQvdAY4_uof%)9S`#LjZ z0n{(ku{<$7s1t(+R55S`*W^Mba8D%0#6lT;_ECHL^JFSdtsLx05U1_OOblajA(~g8Ook< z5Qd4cR~;f3RO$DecrM25C)BRoEqYsPp}LhjCm$- zW%nP#3#!jQFW>ozx%_3e|BtFjb)89}VrrKIOn2C)7h!7HS4V|ICIwoQ_~b1bl>>Tq;6q!(P#8?jyI}B1$jE@zXnoFWRm^AEc~ zLk6*)njeo$*SRT_tjY-*-2Skxkh$StoPXzk%C!IEYeF@u^$RH_#52rZEXtJ695!Ma z!F!%9w%;xBA!M_Z*~acjR_>B|$(a-&&po~fqlHZb0K6o9oW=MO18n5-fmprAh}2C{?9* zlwL#cAtGC`AV}{;kPe~OKq8{_-a~-UTL_`IK)4IfxA!^c`|dec_wV};9`b~@tTor1 zV~#N;VSTCrJNm}3(6Ya1=MF>}9wou|$xslY- zvR-tZRKL2bBNm!8}7 z>$#}sD-;qv$xQZiHxI~oPzM~4S6imXg)H2$c_0nqjd;~cMV~N2vmM&a_Zwa8pK&MT zQiDT{(`yxo!{=4NQq%$jHbSW`REd1Tcqr@rNY>@tKgAGtn4(wgvJ`W{8nwcvDaGFN zKi`ljRUC&o36w(Xz37A5C(#N>+Dl<1S0FN9u22ZO^*r?(<;ocL!bsjEhj#YL`k)K` zo3QD2A;&vjAa74ojGw=cW$1YGk&)k9FY5)uXbVhtCIIzz&Oxr^bW01tT%v<5YFB=8 zhYXl7OZaT$i3taUD6Ci`+f1ojVqiQgjn}pJM>m5wI>l?#B7F&e!5fWoGFQE@sikWx z?C|*uV=PA{nmipVS>=PsmNrD`RcIlXNONAe{a`tpCcvfY@PcTjqSm9$)3Rt?+lDE} z#+)9X5p5rCKdkvH#t(miC7(QfuUry`2zi$uL`qswT6%#7K!y(4$yRp$<6fuR`Xa$^ z>>mn-&7lWgUx>yPK!$M4l3I{gulSa;qk4N=K-&9-&~(LRnbo5wFYVkey*FJ@nvBSD zcjPui!68LJeZW?xn`WnKzg?f`QwUVc=5SsuqYsf>1ui_t^VXw>)qd#pwg{KMU#FAW z7nfP`&hm=zqoyVdmMYyo|4~Iry}lhin|uU&4t{*G|N15I_5h96*Eq-=4rkSu5uCNS zj;E0DT3ZTlx=;ua^eq19v#y@4n;v*TT}FLFaUc@90V=O;m;lwinMTKg zJMUk9o8boKc9Vw2Z{jcja~0umM|8}C3JwZty3(4%$IKq7JF0(P>?=yQ9jU^!iSIch ztF6J*+pgl=A>aA7Z2gy7;k6g)K9;Z_M?~P%#ey`{#rsg_i0SWfK@#M0;QP^ilVai3 z(UUrWP7Y%sfD`+sq<|dkfQ@?~ARxV;cz4X8Q{5Ok{VF5t5fcQG=| zfKPo&R3r4x8>!WRVGj~vEdBYPorQl(3jQmk#!PqAQE^GnclDBiXXCN_W0Jf1f>r~! zhq+>fYzyrW3`O@!vgC(16TG;%R6Bc+4adl|Ro2@mv``@CbNTQ5eo^ zIFbk*s)y*_a!;TZ%B+kKSl~j74NGW@8v2}giTCLNB7F`dzUL-Y$*Nwe7og*l7x%?1 z)cG}+^wiPlXvtw^v2`_FJV(rdZVdMs2s8ExNslz*Tuajj4)^^9Ng@gkvhBx+4;k3`{yeB3lgIbc%G=clZJ8D5u$#Z=!k z+65tB=HBaqR;@?meMg{l0ShZcHG{Hv5E!k5*qqeJ(M3yYDK4SR+nEeux||nW``5Fx zJjY9$kCr+)s1N7cXil>9RcjgJ7~b+?kE% zE*l7dqDx2+&xTSc&JGBfU-XP_f2BB_1Bya235R(-rcUA0VH~7=RoItVV&d`YYkTJ3 zDKQU%?x*G>T~F?jMo%b_<~dbO?bNTM>t7dL_(KVCK=%T;P4bbvYOhV!yANb04G?(% zS=6niqVZ;#29bevfNj}-ol)rKEWb{Y*s{Y;jB&a4^U?x zImV&7W7d-7Nmh))cEw-_^i7Nk=c{=l0F4XD$5CtYRw}@?zuzCcqj*It zD0Sb0G&Gx)a!h8qKGC~tWPjA}2r}#3W~=bXY(KmE2r0o!(O4bXsP2A1hc`UEpKcY- zK~8zqT_`^UpqbeN$AIx(a#5gqhdwu1dUg=>T^`=gg0ECe(N4p{%a9OW_a>?E?UTy0HcK#AOpC4 z9>oOGi1@`UA2}~>ovOR6sR!m=OF6C@sY|Q8mA9-CY0IUpWd&bHl7_wV4mec)CJ=~d zPi1-1c6pFgDb8nqnCv(P{IZ)#qq+BrWgcxNpXAARwH=@^}z#h9eDr*Dfws zr+QGcK_hhwuV++Ue3ECTHFizN`pSu=tdH)CB8D(ANy!oTNU>$;&m`A8h)HsisA<+E z^$+g(i%q~i*8x1=QjgW?sP>4i<$5Nk0AktYMWsKz02C;uYVN;rH$U28nq##TGqChy zvhnQ4S2ACm1C7vOBkH|+y{)|oUSmg9UC12aoY-KwD`%IVyzYTG^^#rUD5v_7O%v$d z9qU=K&oi6r!njjPJOlGtulqCg#a3bA;MZB7swbWb}fA&_YOC(0fa=bqK@fq66 z$4oU_Q+ZZvH)I2gRU}C*5KqT!jo`|+<%sTznKvA%?YoP}Pcq-*wSQww_y5$0pX_|S z=N5fWzqt5BebqeCXVCK?x0`a(U(ycGNhvvC!tYnQek{dU^=Fwz>N~!x1{noszQopY zJ>5d~*A+jWPd>G+!faX)?{bi<;=9Z8*w+qtgOXZ$075)y9^g_<037Ol zk}%qF$_q-&f}xi;uxa@uI-Y<$P2k{U`)z9(u;qQvw}*BuXY~!we+ga4&O>ipgCE^f z2b1smV%&%Foc5UNk$z+mCx+zC}tx%Ymjurj@^QlO!csE~~?i=DazTb`iFQuud4_Y&nWtq07O<@Ij>B=8HqbQ3TZG&8s?jbx=%X+i4 zyXyiNOcN_P+7GmkuQjbS*H>cfy!16oc0RDHdLP;a_6Q4=)`)kWgR%O(Yn3ke!Bxbo zjrZZ4Fcz!V^%o|qS5}%~cO?dET4M!QWwws#vCIR!dW&OJW_EWrC9n|bwXvq2nQ}Gp zEdBx|yi5H$WaT@ZWK@pi<^sm}#1v;n;xmOg_oHzTSM^po(ZmA8E3(j~4$TIACG}Zo z71n4{}9>wVfrDKf#4UzB7v5c8IiTlUIGkVHpAQHJ~G)B`E38tlm}K2 zCt?zND@*hz?qS3A(a{)zr_T#Xc5~#67s#x#^z+V|8K_9Fz_kDrU~8u}t%|uVo}%Fg zGQc)RiVdEA{YzOHOoLJ1?g2}iy=uZZM-6nivUi8}vQazIY5@!bR~UOF^qV<6IE=q& z{A4_OJ>eqpx?Zy=Vc^k{|G<+E3PIhu99QIR=+#SiX|uf1b6GEG;nVbxI@mY$hnkPX zQ`1oA9|dNsv8U>X+Gze*t8kjxS> z9PEF1(p`u}%BB-(>1nCXfjw*C5wQL|Ln7N{n~ykb+ck1^ricZU zbL)DStD^w(0gO%N`nn$|>-wDgch;PA-vQN>Ve@=@)+c8G!YU!AMH43rq2#3AddXkoc-Uij=Sd%=m#gO7MM- z#nK~JBs{5pt7iQt_vh0bU=i@gN74P2+x{9L_CMvXh%O)ix1rzxOuup8|GcRNXTYF{ zR&JsG^|I$*9-(g>H=v!G3vxb;f@ z(c9O~Za1#8uot{^cJqIQ^Pp|9pBX_m3+3##uMWjW^P9KDZMWv+IN?10^N3U5a&GBS zk0Hsw`!3%)P*AoT$3neY~%`*>}eLjh1y^tB*TAnDh*>*&xgXVSPJ32O#xkA%|d~?@j!}GI! zDgHf~hK4%Tq(%7h(p}cyUQoWVT+*Z4N}?1+GcCcu$wi?wM?2LqRo?Y{cbW?Namw>D z2mR5Z3Ma>v9(EB>TIxxiulxM~Xn*p=8F!yh<)*c0enJALS_X4?i(K}q>`8D_YrzH*g27&&sUz~F4a4EK2lkwWe6ZQ}UMN5nmfefI2~u82B6 z5`to`lQxNwr`$%^K`9ZgC*m|>PXhF(2EbZm`5rI(bO%n)_ z;(^-@Fq3Gnd=ki1RY(x`xT&%@6mt9d7Ek5?!-41Q%M{2v%Ci4_>lIv(8)8^4uUGl0V%p`_By^R!Mqg^E5T!xEqG;Q9|@R~OE8UkYSPzO5Jfv*I> z8fmPr_G~Pk>wb8c82^kwCZ&ET=;4FbP|l5{_{T}XM|i}fI5cx^>5ICZ=Fj(;obBY> z^@lt-jD>xmszz@#ajtxy2*Msa<=sD5#V<=Ex}Rn!HAht ze%V$n^sq@S#FTPCuw0LSjE#?C#xpmg%E2 z!XLg2m?`jP6?ESM+txKt88{dN^R|bQ`zs2JM&iRJH{n2EeE%xu)AL%JPX$yT-4+L) z7RDSh=VrJoX}+j8*Hw*dF)V)~P~-uGU_(9vw_!a|cU&ucsy$cCChF{7BB_LF8N{_@ zgRXKKZW4Jn!+E$xoF?tddZVHKr-CtQ@YXjY3$4PbrPNlgl#_x+#gCkp&SSN<iJcz&NQ3z z$6B8$zu3w=0)T35z_#MjsT!bK6B{Y@-7e_K=}vOuX3#>+@O+@Q=})E@i>dJ*+FC9I zPVFW+(UO*YiN`o&s(qUKa=Rwi`=>DG3BgCG&kdO2`|`>#1X_s5b#wp&%R0W#rG5Jo zFm8=KJuRwb)aHSz@-Tggf&63=Gu!xY$hyVAPXb9gh^~_ng`43a#~>N! ztVnkzSerVCKE?QS#aK(Gw)f^EFP6142se=9W0fk6LDr>e-C706fgzjhKbpFR$rHNVVyxFs|IzuCF|v!a`_cQwNOpY%7sem~QE+>#-DCNns^bP;0$yj#c6q*C*rwbp1I zWx7gYa7ulkKjuk4^fcxL)_WM9@9`)n*_kk8urk?pA%A#-pt*1BssZ#S*yxXab0AZZ znCOGUUlau}@z)f=&2h$TiVIQCAQ8v9s9C5I@NsydA~`8h+B(1{QsOw|6h~f z2Me2%rJw%EXL1EyGM%$WGo zZ|q7U(Mvs-@9^X}TdqwR6OSI_R2pg@a0pFtxh1tL&?J7ll2wNk5ZESUJ2K8x!~`#6 zs!mDK-TBY;$iEc=IiR#D08X2g?gflr^!93w9p3OpQ_H%&`jFllbvECCQ{a)P(tMBW zu*wbDhO+x_22M@LFp|_GZqxV1ca?a#MT732Ub%fu<_V`t?;j_SC_PN!^HR>IyH>If zY3%yBW|S+mCtNoYYfMZwY1iEPEWcJqIL{%dPKmCgx+fb$Fh87FZ&X1Y^lUmjBe+o6 z;;SBxGZ8)gZP+$zdb|Wp%PWLRtWYkzL_hEnZ-NObqQ-!~N%dz9ldPF!HFaY8BU4^Q zBG~uLwEgZ+_{`g8F~O2*y(`0;t+6}^_j#x323F36LqBQL=qr7aR1f;IYPRKEL_U(x$yPGSlECNL&whwnWRk^HsD^z}kHg9Dz2k1mw5 z8&KJY(}gM8QWvX*jWpSxeL2%=BLO9Q$I|b?p=jvoyRjg`Zi}s`D3-F zTeMP-$llwNOmu6xzGJBh;UOTlZ`qm65P6Jg+W3=(*#-O4dzHBuUFbm zy}ytO82l7iOQY7#6gk^H(4#N~e!>*CaVKo?=V|hJK1reVtLP_O>ppjP@CB1f-?i)p zTg!|BGG18yM_a8Q)H=Ac@}m_A`}yzbn2FX`G`#-dru>ItQW`ZMeEGFGrBFv%_}ZS; zCvWnU@M5K)b$@e1=fq@%d4fIOS$1DcKlRxtR#5!o_|mtncdEmk_*y+~oYi$=u@$lJ z<5j80?=NwVgKhs6u5aYjaB)t`J_7{WEwoR@709n8Zbk>?bLz>l@9*deoiTjKGCDRD z`HzDoev6pK8260@ZS6ID-Z?IQ&yp;4me@wb9--s&K>-$PdB>#rVrP5C9b=omg^R5* zk6ZnjP7L^ijCcEO@C{QyI!un%-PbB<@_FO489&(^t2tO=dE@oT3yx(HaB@{_>vBsp ze{$l|`Tdrm0*S??fIC^X%JGu6$d0eyft#a$HuA=p%_T&e#m{jVXHqD4SVJ_pV;imL zY$T!N>u!de`&tBQaVdxKFAF0@{rZVK`Z|pX+}X%0Hq*u1o6f7#xrMNL<<-8A#Ub3* z7gve)cAEsYno?@$@YoHxsjt3%3DeuFQ{jNj={f%mnQy&mpuJtsP_|voLJS7pcG1U% z$C26{Y>F|gkYad$UjA7B&_EMDOhUH0zFiTdLRVNBWn z%iZ<<)18)CVeBkk8#&X&v~Op9UUG}prO?Iz=AN98-H8HdMq`Lz(PUfneLm9_voW-! z=h=PylRTN~(qd4VX+?19KWAbUXRT0_Qs;=I-4Tic3J)wSK@ui=;$j0EFi{8w3WJ?R z))tRFtfNzd5C-X%7oTnQ<^G0eZk)JF6M6HyTfF4f z?3J>5ICI;%wiSC7>=Y&HHC8uy_}UrCI}hJ zc6@ajR^|`g+}Sb~v87A|Xf?NWe{Lu)lIKEeTCU8Zex*y1j3~&{Xa-sj_h!BG*?`;~ zwL^k4W%2+!AmpLHJYJpVb03v%jAaWN76SBkq1%uBGMK#9o@Q#MD@!NVaPSRzm`l6Z z1d18W+<6kr3fNcH!N|pszI~Em`NpRes_53*Uh0oO7-B`y}=zQ@Ti@h{Zzw6pCovsms@K`**vj!gtU=?)fPrYqZjB}R2R@mugzR+gu(H~EEVJE#aRx4++9FjWLw8Nx- z+G&P6Y*9igGM9CBH_;kyE>fQc{phi>ywjD_XdS{YCzLKa>s<5g$4ZF^?+3UpL%zY` zbU`F+H7##=>GR9k%k73(w@1iLWE8ryuL<-44%ev?$W2|z+Q#gvqDur3a{u%8MaW=* z8@^Nb%_-8;HK<6sMbcUl6qWD0u9J>IVvC7foEiw%n1M{Hcz6e@gurG{<(}YzItgfj zAyCB)VxkWdcYO;k4D}H5B;#2{@-C6U`mFGQ15SOhYJUI;TSHa#ENO2q&pchRB>E)N zs%}LTn#woNOuMy6jgsD_8=iwj)u1h;b~>MB7*Ry?8^^&z?YXU=Y>9D?gr#xoU^g!D z=n2f=Z)`hkeYFR}NN&*Z5v43LcEw*PeqEk-{qbR#JFj~12@_mkiAYk`_k7Q=6Ip^6 zdX(#_g6zeUw|ZfhYcd1oE1Y)THI?AkYR>->m~MA~8XC{-wnT|^FS0wODXzGFxcGe0 zWqxf=2)n&vPrp$K{pdR-48)9_2 z6M+j`-;nxsr|50E{rpcy=OjIKqbB-T`XAtCpb^91y`U=)>-3kQI6-xQ$kuiBc!xw&EJ*|TFg!>GNW#912gJiUS{ zeHgq~-fpx)BjBp=+#BDS&aot!X$ddb^HkvCH&2l#xaNFq;Acb}ip}Sn#5Y zn&Pbvc@ue25$)*Ry4=-743w{ab8)pqQF!kZ(^l$a1JmX2O!^Zb^?FZ3mb4}Zp3JqF z-UALP&5?2-yvqWM=^7ntaFloOHwW8y+O6(LWdpLIQ zs7b?DlyFNA!S0NSEdO)pp!XiGT)^@fQad|a>Pc_Wk2QLy798~@2^aE7u$<`g1(BuD zx-GG;^w7!@Yxzts;_?v;M8s3kzvjJm{)CG z>&hS_6zjhpM~Js{4v_PUj74|Ldi%A7vM6wGMdpVBXZ_#tv{zDeLFY6<&Zh zfLt4^9^*2Ews=xam1_OQrL{DdMN2HVf0PvOgFdDMAOHT?V+9i_Av3c)Q6D)};W&u@ zF6@M+nsp;>@kihFsI+4+>%^D$ryIV($mU9JUOq!M_*d{Y)45*Kb_TzbBmCh9UfmfxpbxP)DGXYgFT0$s( z=b}Vnn*(00{i7;aQ$N0!-~)nx4TBrfq3pjwYnSV=moCfi)W@~yF0iyKo9bThB00LV zL=hRmK8%qqijuM^j6(Nh1TRS0f6d%bql}1zLu;{ucI}6Z zjXrx%wwnXviTx^ZK4wE8YJ2|Zw5v^K3aOi1z zXP9NtE2vp&ae&Qa-}cZ)nK!eR(=uyyEcnu?MypH8Xq1POsWbpl|o;-LG(*{*(y6 zd=o3QcdeWCV_2l2cN^W4{piLmlxM4Gtf2SGR-!3(TAj~j{Lw+qkl?x_T?6urUAJpQ zw|O3aT8IN)A(G=7x-FIi-ImA^TDNx-9%F2=hn`M&X%5R6mz5cm3c75JC_2)$FJEL#^#-autvZd_~=280oSM^RaO)Ma3M~*QGBAt>9a6mcjb19o`v`t832*B6d+tm^p14#h z(-;#Cl4eQP>~U9Q4t+}`kD;Ak-R8ouwc7<7Y~MaI2IJ5(+sTTY^e^<PU_Q*6&>nwlZFMv6iBupTJ#ROAxX5ucaczN%!G8B~P>{_hdq9j7x4oyk_4UZURjEU| zg%N49cE{`0w|JMc@ybyGtwoI>0d|jo;pU!>Y?`JmG2cL;cj&6epYfr$6 zLaifaQHE<_Y)si~@nWUncW$6VMN6#j#tzISU8@~Fwpg4z>_u$I<`rIAQ_VN>p%=}_ zgYgFxEx!$(u<0l(x}raU84t}gAOW(>3=2Sz+UyoCi@Mp|XqViJs={Wy_Lf>>oa;b@ zBP9m{?4-U2pdEj#fR52U#sFN2)3NUQ2|K6OuslqUzYkVLen?gET#OMBnDA){@8iAk zrG!OS(G-e3I#y*De4$C3NggszqzLH}<3NhP2yJp|7a!}J+g?Q@hQj;J1jKgIB}8Kn zt<;6w@DdV}Z8eh`_lL5mrr~MDELK~mue}=JQr6g4h7>QyrJTJ{AA`=+l(o3!wMn3D zIJECW>f6v!Q>jL*SFP=ohqn3nrLZ&x&4V_$8u_bPn~+fD(z} zG_kI)ONWoq4yMqmlsfrmf)0m}LMe|5zwab~DT=b2!nrxu=X-)5tEM$nuH#cNy3lu2 zQr5RS0c{|zb!sGpMQL>KJbBp^~um@H6|+? zyz5G7F0JkenV!{)T=|gUGP8UAl5x)lOcbD|d67Hr{PZ zLt%JOkWjDc;rAM4NE1RtLB>kuV+7t&M=Y`(Jvp@XT-Yt;iJ3;O0Vg2OIrLjSSvlNW zMO)hIQ(n2dwR)?p)!WIYJ`m!DJ3fxM&EGjT8e5=ElJs0CZ;3@Wgu-<20R6uK^MvOV zQ9Oi%#EQa7b@P1Z^H2+W$VTl8lZ`%^Hthub3`4f{a1zS*$)X zPx4Skb4tB(*oJKtT!rOv$anFjzK5Fr*+Z(ki%I=8d+MB?g6zg1(wiWQsmQ# zq4{BCp5aLy5r-A$_E?RWs^u&;Yc%q6bmyr_RR24w1pNJrpTj}!&}2OAOSvB+?!o6V z|4!};uHSD+9d3&oYpbsdAaBYjpsFgS*82@Y@=)7-Jov-uwph1PDDQD1&WrrURPoKq z8=8dDz3u(2mN?Q0bx4FlBo9VCCevr8Z8(He^Ynh2u-*m^s@8U}K83{(G5ivBvMCAex+fefurGENaf`zKD%${RUlK%;nc@(bTbFC1SJr9}-(H`HIyP5YzK7njQjL8`Uap^y@h2_@gLKA&FA_T^9!lR0(qR?#1oTHY377bvm2FZMX^H0wjbo zOe1KVv&5X0s;X&GhKM+sxS2_G1Ng@yf}do1#!Vv)eWRM+Iz6(L(bc>8eR#I_B^HqU zOcOuh>qoUMFA6-uIp8X+BloS^9j}GjN*0H(xhc128BdS-4RC3as~3|nLu^V3(~<&7 z{e7C#LzdMM03>-WQXCG?*~oEYvRG0x&~1nlNuhDwgdMJJQn$W-!JR37hKCg~Y*erE zYo>ui-Urq4)|FylP1~tr()3BDpI>ChE>C7iMSLSle52upVfqa)=$`>EA4L0Bch-n`C7Z_}2tWs`v8J*Fh&MsvFvoCPL#67wmO?dr+} zS1aKgF7U*5&-InO6Ix8m6G0R%Ym2uSORJRt_Ps5|$mLS+Q3rI}I0(lPp!QivU2 ztH1inOfws94Q~#i>fj<}wp5@dPD5FT^GnTt-kkI-S|{|^I2|sezQ*+ zM1gRo#5D)u-kipmUL7eNfsA6SsrS#CP0pavRxF(hE>(@r*;8$@vSQ+^igUQ7O= zupn0-2KsvJnrX!}^Lgq;l{9brJku@u9C3x-KR7bgdEFx{ajtEP&?cjTq{S|?iQXo!by&vgkdeFH+55QDnzJBE)G6rs9Jj?cC)n(p^Xvz z;H?E`#8Vxk_00PLdNY-KM`GEH`bbm>kTqn4pDkSu&p%GhnmWZ_T_(Px!A!z%y#VGb zlnfau^uqPs;t;;w1&7A0N(g!I>^1+o>vt zZ{^ovHp~LC^!F8&1qXDC0yJ-FDhpf|*Dic}?3S#lH^|gVFnDUwp!e6Y*I#}mus-QK z!j9hh#=n@Z%ng0H4zPh#pJD$S7n~QhgrI|3l5?Hf0{30>!Q#BODirFl;+SFv$|G|L z`f9%tT8%OQF3_P4QyelcYwe7M;Bwf5W7 z{(Py#@H=R4b8_{n8On9TWoM0kwJvJAYrdNHOVeQ&lbryOHV(0mR&N#nYsPcu19+K5 z#g^R-J@3$u!sVwVU}7f$Pv)EpVPo0+S%uEk?PMr?^-VZFx-8zcK;+~538`tqZNN7C z*g94>fXdRyycN`!N7#UmeS0|9MNmb0r=Wv^AW~y?>HHv~ZUEY_#cd>1gsIVn*y*u2 zK3;;iZV>;ap1~i6%-)t~YH<+mKUr(OciEO6RqF`N4$+sJbs)579%J$yigLRS;lSLY zW;Z{yYs>?yogsdeh8ycv%|qKKP3v+z3Rh_9lB%s$qz&==L$}%mZMdAw10tR0x~OtC za~ncLLP3Svk;@>yb}Uq+`jr|IU8s90I+{8(jmqX?u(WxrC)srsubl)F`*7NyBp(@e z)QPpZaEjA(ehqVg?O?7HvM^sJHFO%Ucy>qdS$%)P?q$d>b_(A8>gcM%$N64w|I5?f zbI({P`{Yc;mCZeQ;5$B_0IVP0*lpG{{zJi(upeUts!P1@Bn z-RUYbcZQJWiElmfuNN2U_LIV#%}5PlvyyHOgu#yi^w?iIyI&snozXk`C(Gky;!6}J z1tjQ?YBgiZt*2Y`Z=rehuT3>YxZujhX0@shL< z$?Nyu?A^-M5pv3+OpD2cZFS_OjwwK!07Npd&LDs*bU79+2#Yf*Uk!X|`%QHj3`1!K zMfbOEfgYiWI)5g21mi>h`>&2Hr+QbA7UW1hhX_&{xxt{2zY&_=cAloA&vB~h!rZ#0 zVJHhMO1~|PJtbaL>RGDdHCr4Es~xZ=P&l!S`%Lw1UquQkT;jVHE9Bp>MXfr?=A@@= zRAjfh9f@mT{AdUow;@|@XE~dr`AnAZ6Rgqvsa`Q;1wld{5;nS87pbk2}>l*nXAO*kP}d7MnDyL#w&zya(`r z1k?T3?Vhlk>>+HCD43`Gd{5?3Z_E@4MA;h>eHE*h8Z%YXhsophLA-mYS@y$dvUrb5 z{b^z)!Nm;JJ**R}R4VaBXY@~l`Xut?B_7dXd7uYLeet@mD^G=YdkIE#7uwRAZjY>= zD6#CmDB`skhgC5#zb?jv+l#;J&?Ol#+GBZ}H2m;{g1>97$=ciJ$|dphhbhUv4GC0t zs_X%fChi4HuS(3fU3J|z=IS3ns2rLCw4F@b4WMrv&Exs2^VA-j_bFfE(Rl$0J4gfu zO1$!HNY+TCJmeF*E#JOe+V}4j9~AoSOj-u0{_B(cvn&4KxcCgmU(NG-XDx7PsphRQ zC3w@WF#3ebB`)vIGIS4YO@v)-O)}FNZVSkE${np7VZ0NWDj_!u%rWq5&%b5!<~2qL zegIY>jYXcI!NIqLC5pR@=||mm$V#`w@TE?6%FRC_=+M62u^gIx*cii7WR3*9Iug#4Vm6=1@?rBB5JI4|-C@Di`a2~;T^TsPlr~>8WWD&1I ziT^y-i2ND1cZhWlDM7-&zkJ+oqJera(H@$D!J!1O@9Hk1Q0 zYa+n0&(Zj}HocUoHQD#|YfCoIf}XnN3T5$Fd|kKVkjHBi0ax*}FnDCsRsa>TU;0FR z5zvzT5cFK$8yM=WT?@Y;e7sKcJ}keM9c0C!P=D_O%0%m69+b}>)Fbb^Qso~7E?|a) zL?;%ond0tC`GBJ#+h6e^yB~4P@8~b#(8tTZuHW*vhu*vY=f_7k&fL;YF5fc0M%=c9 zs4SVg-xN5cepo&ppu&=~*PAtS9Q{5BHn`8Ng!37V-d1r--7CE2rLn`^ldY;MFi{Ad zTYe1DE|YZcbHKTxRvG?~ux|6laU8@UbCH2pRKtDz(O6jXACJ2q6vc|Mh+gSRSF!es zh#w_BuZ2exOt5+JMOc@v2UYrvFZ3zzaSJ{f5q{*>kPk_{UypEHnPhiE+QJDS2ZJ$# zQ1_hwLlBeLQ`EeaRtTHQbm_amWs~*@bsybAMFHc=ixTF;l4Qlx5XteynU2r3UDa|> z5!c;v-E_>#>SWs-PyI%K1zj}5lTYef#ozT}c+maItQ*={>`?9n22Z4juYbRhup!ZU|IE&( zH)cis`$Am1F7Jxn(?4?QmwxN!8Sq(|TsEM^x-EnO00M)r+q6B|>9ROJUxN2F`86;0 zZ`}5RN@nG;-3Q?5&#hEmePpoMN7t+^22b^bI8>qgW`F5J)sCZ|{eAJxH99``Sz5a^ z8=6}_3z=^YLL|G|Ews(h<#Fn`zCfqx79N@eFODHn#~T1|8boj#vVijRc^L`RSOAb9 zHm)_Mh}U$vn%#1c=1b8M)Dcaorg?z;e(ssY{-MsF9tMacO1fABeR zv4uBMsn3Px)g~{o)arQ<7A+|hdC*kcwWCa?y;2_Y)VqJNx@+`DKMMdVLjW5MgMy`# zUcTl>czZZ0_?ag?TF(wC?vX`r)R#2dxD-+m;5#qxEf~n5ObF4IgxP`f#w2 zW-P+6Ln07Csd8Yvb+8KOdrRhdz36_|I1ye^kJ4xm(4P{~dhU6fOdr zRxAe;!M0oI-Md;{sS0?EEJ>@t(M7sr`#$XmuPWQ_i^FLWj6+5;e$u4%xhZPdKIS!A zQ5SSgM89T`Fhvn+3e~kK|K^n2j>c=zyc!h4+PrREw9TaQJ05??oApb>kBj(n( zE1?ng`gt=@KrsY7>~E-6f_G@YVep3SJ{$Epz5P(F!yUkOm$4&=Y7{4yjFT^|B=nCSvm-y^U*^sO5MG8oMB7cdS1)2Z?fd&A!ouZ?ZPiY zp__7$($LSi4X2`Wau#f94$Chdej#{)BbsE|bN8TUl+H8o4Rn7@YQEvbv*-IV#}4pa zGl@;^?zHxeBipk6epD@icWRpT#oRst+<>&$)@Ge%-FhCOJ(sS0g&pu_j$2Hq3BV@Q z>4oRLG7UGMKMuPUtV&Q84BeRs>I-M5lYLZ6RdcvH#aNu-))0O|8BiP|pMCbGciTtt z$Gocs%mO~+rpsdM`fmI)f@1sM3#1-{!~hWZ^cAbYY)DG9150ee~Zi{ z(l2H3|NEQ_t&Z}^E-Hfwc68{|K8Q#I^bs? z6&HH=zwhg>#{>RA-8}K*PcMLfi*En=A-iMXB|`m+TJHS4J9&BrAjl`&ZGL-F%g<*q zboyZBo;Hs*{wvP@$1DAR`$IBVLmuas27ZmL^S{5L{3>{f=&Nmf^>6=L)W5QH|NQU+ z8D`}p3B)0^;oALQI#&Mm!T;l8-okwEo^@v;aV`AT0E2(N_OAfJ9||`5dyk*_j2hSl zNnC1w@7*823*MgCImN%VhAED`0_r!Wu-CuujPj4iDEIQKliaIcqeT3*@&4TG(oev$ z_bYtz`wkR87YMM8cVB#T`@e3#|GnD&nQ0v70IT-NMUKDu+^&F2ze@f0qV6F|8Nl0z%`v2_@|FX$D zxtD8^AIHqZtNYl-H2 z4A99}rO2xsy_Ue}1@O&JW6I!YfD@~L@$1(CRW2NWG>^rLyt5;Lr)P6?cRFB&FFeb! zcDqvYB=}b`9%fKzdVRD!Us9YcMss;5pso3rZ`hIHKZp~ zRW8|oH1-N8TG|q&pnmyuQArE?_YOLzsk%W-!s20H9*4>U>0+21zCRK)Fo*N^;Hkc5 z6c7osfL%VTuq$`nZ_hrrWOnnLNeK0HnLnZr&sIBq`R)_I4rD*@^E{H!o*<+^_DVV^xnB;vHLUAx9NY_e#OZxMW;P8KlIQedESH%N6fwFG^@ zFDi+Wan<^^u8Tt=(P;b6iivk>siqBpG6wM|>h|Z!rf>xVk2#hKw0!vS^*NlDUo@Pr z1_gwm-3Y6-mF8y!_m1wH6%@wZb$IdEjIt%(vGd51Sba`6@Lp>O%T%3G=dS~-t2=dt z-cr~k-#DS@ibT(q$4H2|c!5N(F*!q5Sla@@o^El-MoGC9021)9 zMk)y;-34-F$F|TRC+M`haaGLbNlBn4{});&(H<)OGAOpMTd4A)a&*$F>U5Qe46(Bp`e3K? z*rh@Ek}sDoYRE1^%21Ng9$0ieJkXrycCQKnSq?p)P1M3*wQibnB2SDhI<$gk{8?l4 z4gUaP?cw&AudATPMRjUV<1U24FZuV!=Y73lFZiuj^~XI6O4H`B;C997xlB5?rtc0D zCVj=0uG@iPi)UI5Mhcu}jvwPs=ciZzMolrhcfHG7?nJlO16u-BR>e<%?}UABF}n5r zap*n9XB|7#X5KVPeM^?CPqh6an}^&CJ6y}|nF6l$mmGSWbsmZXclt8(w&J1Ei!_A^ z)~~krvrS)#qyrHmRqW>eJZG#3j#{*TLirQ3uNr(whJFBz4&!?I?O98LSif0gkTuWF zlZW|9tL3g&0xtdT!BDP$w85PzkUtSHsOWbHWuW5qQNg<}k7JVNHg!rugb3K?glaF- z*1?iwiuo~i+Ws8NB0V<1Y3T-njh?PF7wgz{@9ybVKu1x_G|Wp#SJFD^A9ST4J6jFG z#VcrIQLjJcd(0+2a9e{L#IF+ViA2vu5fP{90BrZ`$o}Pf?kJ{E%Accl@ZmXjYgtEOA23%4ZS+nMuBVM@<9f7qOr-Vx1#sy7r zY788^wPt&$rY)CO|lpcUO5R^)$geP$_~ zQ}d=tbAAP}9e#uZr7nl% zK$0&_i=@ZOm(~*o1@~UR9OKn5RYYCodjP25=ks}hc5X#ubvi+j2sj>F0m+GYW_WM@?k{*^tOk1RvKZEgTGIp(uxs>WAa!z2p!AMs5L;ajiGBW_%?+zR=h2#Sfa z|Il(%lA|1{^4~0}`LqYy14#-1O~}>W017C;-dTNv+yjakTqNgROer#{N2a=yJGlm_ z$D{Vt_xr5?R|R-swZ_-H1i9^<=-bSU8w0vMVLuWMnyW=T1$vFfZ~)0nNoVQ040F|k zx49_^(ef$2qdiuF|MvYAv^||9&ES0*^GIrZ;0XM?It5^qd;>%_7des>W&+se8h zdlRPUx$GeF^7V!=#`(y5%7Yvu$(nSIB1Ct9?XA*kgr|ojU`lg9R$Zk`HTU0_BRU|Z z-b|U0I)Y}of93%5R>mEC7J(^oV)>3her4Rs%x{x~8fw$!%c z`O>Xhr1}zfqEyYCfu8E+-Ig7o!qO?VqUNF=a7y9fF1bH6fYs0O42m|sCR0;9a`UJE zY{0|B{fg^JopGT1=!C`f#<(*_lHgL-0o|7o#<1s0hFPi z&U%rCa2Z}C$!hoMwSvgS=N$S$chg1ZQosT+%hZ zQ>G%b9O7lMKJ=v70L3)Z7df)? zffN`E6#Ms&5djGmM-;zoX0WpP07i;%<^d640QK1j>K!0ld}*EZ@7e-oG*i zGrz@`+GP0=QZgDF0IFMgWB-HY>k-IzRMTga)~JhPR|b3!2)x^#ZAr@NbP$s)X?+29a@}hs*7USoTKDUHd2>k_f;#dH3!euNj1HYDS`V z{9T(YkbJ~HP+rY-68Sn-x+y8vKy?XC=6&@C(EFuCy~%i-6lm`yDChKX4`8BWX%BQI zS^*|8Z7p6vXTj#+zGM*qDwf2bS(wpDyj@MG8f-K<=8wHq3PUuwd$$4F)>btK)wiGi zyi>UeFgeCHXXDy`r&0c z++LEi$F9gskx56T(}{~_jn_x6M^SQqt-E}2HihHLpTRDm33VDwm!=EJpJ8@t z9eZH~a(>t+dVa1qCpI_7;mxnfmLZ|pL^ZI*{f_r7wb3Ae0KQyYz@q5It2n28x)*)V z0@!eO18|L$vh0H~Rb}SuRyhIDLg(0iJ2`hr>nciD+~`;!ZUyKpZhHx!S3r?Ix#e$IX-gzFeUhAQO2GLt`m|a;IJJTFei{3Kd-Cr zSwwrE7y1SzsvW2QFi|iep_OGU`uqo|$Radqb5!akCDKu>%FGdDwiS`AH2rMz!yVah z(&EDwE14C`0JZh|Y;}#v3I<|IqTFGr3Vs#ISkvkf7TdISWHjrKQ^`dMaPlm_->j1$ zdQmT#(|6FW_tSfFs_#7+>u~=QB>wIHq5~T1%xGh@=Z$btRJmv-Q}RK4;Fs_ z68`{YEYh|$O}Hwy>OrX&NmZ#^>4lZZDK*y&e`+V#ZF|dy&}eJD4(Ku6n`{bxno#X? zGWuu~v)l4};+T?-zSqsS<3EY}h0Cm72T9vr=dJ&|Uu3s=eDABQ#x*bg%jdKrg`(ae z;@#Nk)u^0WvPz4NYtgSjI2#)31x@*V?Uj;!;^U+6m} z>%g?e;-xNc9i1Gt_4CrITn!6Io%$G8g@KTPc`2~xQjeonPQyi97J`A3baO_3wEaJB;)y7 zjtjg#=8d8217Lj7j)p$?Q_T_3!jhx+vO?=k9*#v_&=KJEST5s6_~mpKDkB|^#xWPS z=|7yKU0b3Ng$x#bM0=7krrn3<3q`R$&-n3^1&;!Kz@$`_-!gd(fk0N5av-T))`rlw zGme6KCJ3FNcsm&wqBZVX zraxksz2<>`qkqtm=%c;N-0g`MGH&a(eqmG1k+DF7`;Svj58b2iEf-T?TgwwWR;1zW ztCY@$Ag7wQeBxCU#VO{k@IWzd(kp>(;R1)N@s8IY!P^DH1#51MZ97VvS53hA^4rUG zZo6|U?}|eG{Pte+gp{dOi$36_0luSVwEt_@AD16MiM#&VedsM8?c4oCD?y}4fV}z* z*))%ZLawYD>gP(KD;_xC@;p7KzJV2uN0TcabQ1}BC+$c zd)Q;g71@_&o&yTEJF~Hk4_bcm+s)9ULCWQ}$hqfEP0 z=H_nf{Zu{*&$`#0<_@a_q}GznCDW}oJdO4zHm|F`9Yw|h{wMGXhcR)rt@S9C7=KlG z8PM3QZ4tNx>iNqbbw*tlw*tayuy#e@Mmfb6yq3+Yl{NJMioz^8%*SyFkhW7<@aNai z(D3IU+YBd~gK8L=)P)*qy)5Hu!*Dsm5n>zPiyCS9;$JQE^ygs;+L${H{wTofEuhcL z`hHKgh?obF5c9m7`sc(X@6AV3Go2ZyIze%j3Ds~xqt;k3_GMh9y1nZM+tXlYew8Q7 z6gx2wmhI-q=)$;L;Vyw&6aWAffDL>_&(Ev4I%|4Y3MgNvBV*5Z-Be%vRs2wnvx3LZ z*S1gib++E0Hb`Qwz+(P8aPQgABUD1l+(;!}RRT$jr8!L%-jcNt0Xv`Tr=*dUqVkBS3$*qotm7^= zveD5vSJ&khFM8=QA6!C($4-Z~S@hPuaD0|Qys1=!y%65&&>Y{Elv>oL>mClnd9;ct zwk0Z6I-*GuEaZ}3mctDF7U4|*)}wphlLb&3SuLt~03lPh&{N5?Qyjc&5Ae4TXgXGhAISV|8m5SDW*kRBCGVSJ}`t=xV4 zvvusyP8TSj_{O5tq}1vl9h5KT>ksG+_AJ~a(b9-EMKAB-0wudPdn$a*O+84y0euCv zn}@&PXF>Un>v$2x{vE0Lm81my-`96g;#mOK%lA=lJB9G5zAIlCxU1{#UhGH2&}OQ} zYbnwKRSTnwR5%%5J8Jb}v)f9Cn0VJ2l8Q^mx6?VspYLsVg=W)D3JVO>`fU^KOlj9E2W;x(< zpwAKGyoZc3EJYd_)WY77Ljw5CY}SW~!WhXcE&8%t!>Ao)tCTRj;P}fKe;1G~Sd86+ z)5X3xp_8cCsccZ!Q`PD}B-?4#9hmo6^De^4oN)qW`=q(Rq#m;D?=*$Jg^eyHcfF|> z@5Sfl2kbdEHhr7?*(;F2q!zV0Y-b)EBEM3+H|dx!5zm}y6QW{Q z!37`s_niUD(66A{oyB^Z*VMLG3qM={yj3HOR4HGM?Fk}tb+R5hPVEe6ant%}cxkxQ z{Z5n@cb8*-MsUu`5M}1>>$@Co(8lKvNwp@wY$!%;Fe008`D2Bs0~cg0*A_*;J?EfB z#9wXX_N6*z>}k_eW#n=P&g?*YSfS1Zb;>=y#yj_Gp8d-iO8|& zX*)LTS<|_%DnbC<<;5>I+ka47$_!U#Wxi-2GwEz{gZy<$eNFrAAUOzWah1Yz%A>KeP6hX*Rms_A#u*5Ny;dkgB*RL+%Xh9 zen9TD9?TmHqHekVyi>EZcSh-xc!0Aeb;)q=bO>wp1viGc|8i1;d*|zw^`0-Hbv>1b zwG-qdC+K%xU_`*%YEJ2!3o6$qP644x80cV>+%mvso0|N0XUStN_56 z#*KECuZ|#M`oO);n(D_%%4l3BC~sGx4m6$$QPdJ$9i?~vMea(i3EVwMA zacWmM)3H1!AgOw&;Kq_^{UVVNpv>)mq8=QN1)Wi>&DtDsZr?@L3G}U>$9f_S#Nq`z z=DugSpRn}65k}AKIJO@35=q#jA!{&8)^n#UzHaSHyY(Wa#$*2YyX0HMF&ige#KfkD zd!{75RluU$?P+58&Ar6(T?Npa1gPiU`vSv?HL{j)=3s(I}q`KK5AKao%!}e-Kmjc&Hb%M?pa6s&NNqX zq#*PIj#0Aig<5z1l!{7*J9kTg;LV4&XlB*z{s)W3{f8#t_u^1?`ZX4jBJukr1i22l9=Sp_W&3Vr;Zi+oJ4UOWqAir?LQccbEj{Q{g-f&T{UbaL` z6w{DRqOW@>Xr_Q;{r;XxV?~obtP_fwGaflW71&xh-Z@i1zRBF_9o;+dR7W)P0vsMw+h*l zumbDEs7eP|LVZ>$d&3`;Hiz$xXjCBK(sz7ZFvkjsAy?!ahOMwR+*VM_CHS!mk}@Qo z3*ez<){Fe5M;(#uq5Oa{Y|>LymBx#S$oNXGg}tX4-OmD2Gl5<3JYy`z{g9ou9OJRt z_9cPWDZ^WIbv3OHRgVIJ-loh zGUK)-fpj!H;R~%))cz<){L1_#u-)m@y5NV&i4K ztU@1khts>(ls}ULv{>t2cfjGZbvc&YeB0%!qwmrM9sAa|`~3_^9BIxR-R4yKq>9(x zMxHK(7WyRr@40_{LqLu{^j-pM&}Lavll2K91ZY3jELWX>N}VqyDEHnD7hb~O1HGg! zRP)>>R`H;yHM!d|` zz0Nd&Ey!sWe zE=2-MIWkse(wnV%9Q#+Ds>48_#(L*ezJI3Ws=%M?O*FlH?%>s|MdcnVAOv+4#M|1{D}Bv ztrx|d@)Cy%<$53k7)X8$%)gO+;5+0V^ckp3n|z#+NE z*P7ui!|)B-UX;bkHl+c2Xk}6y({847b_q)I{!CI&Pt@|oQIhJKu{69#;cUVK#Qe6% zkK@o!T6_(lSKAz6$aJb+5@k4PL3gL}Xfs`K`qzE1>Fn)&AoeM~D?542oewS%nKztK zej0u1;CIS9-|rVEUNLm6t!D1LUG>sRU&mKF)5c~rCVh9bUXUC0-QT%#(E}8QSgDb0 z27Ep#jCe$cs5+53ql~%capIw)%0Go+Ze^q*j zq(!!FPF#JlTr7pJnG@^PL(_JaYozw&$;ijuvt>+}38U+;HM7$~ zYIy1ZUnGj&!20A2T*R-#!V{jAXU}+CFw>bh(lb^w(l2EU^!A3^qdoJOzS^f$Cow|o zn~-aoYK%ve({yTrnit&ca_Ms?pX5j@W^egX@k)tzSgb?VpF4Badk%Uiuie=xd?`Cx zd3D+I){W!CkT}yl^K#2tsg6dT{oxe*vtjY#hFj%aUtV=@9<1}-{L5=AZn)*C8sojn zk9yccihgKB7J3$4Rgg^>!~(uuaD_qXJ8KeHg$qNXkKaGf;fz6F77dJHHaFXL(@xUk z3!jti(>Fo_Cj%}UqtShw$xvkavc#xOCF6=peYtCWoi8m6^4cic5kie{E}nXA7zs@+ z(oHQy|8Ork!RQNjSW$78)1QDeB0FX4sh;iz8%B-3kb;Pc%i?Jtjw!08dpR%gUzW$SIA!s-uGh;G3jrQIhtHr_6qVu*Wn6V#l?y;=HvAjc+&~4Pp%1Ld)y`enI&(`8oNQ z5%)-dL?cPUrig z+GOmbGQamb0(L=!b?FKQTiUn(qt2B0)6umMwZTExsYNHR(k3;ycVpm0vV%G5Wd5`2 zfOTrL{VzB;4?R8Sz^LXiqGs}_ldIk;f<}q`ZAK97J||bcAJnJy_JXV#zC6W|lZl$Xh~Bag%D_jv@4}f{6<_EUTqkO;OiB z)q}uy32I!O&Ge_9hH8H@{LqEA6*kSqowm?Jo}J>o4c$-_P~>3^*F2Xx54+%8K5NdG z(hqhNGs}7EBV>tNg3cH|B+BO``MY7dt7G!#+|AhunvkW%EGgp^HtJaqEae1ua!3GY z`GP>g{PS*m>bX4|-O~M*^0x~`W@kW%=&eWR&S4hr(zeX z86gdoUFA+<-<-MJ!&2}Yu0OCm5WYx`8l+?}Zy-q(^#tl$fZ-bFp5PPBue%I(S~BL>?K$sgkxx_*cR;ywJ1!Z~ZAi)41pPEK$d@Ti@?$?0 z;W-SQR~bv*hF?&V?0}CP^7!iT1GsXU)%ox%3dT%t3oy=k%23KVfi4a~PChv60+Xn;i1A>{oX66giVKXYnY1O;O~r)@Rj?1~tHYb~6CJ_p!F{pF3pF1hR4J z(v%stJK>=uK)fFvph|>@8bPet@A5Tmb+JzOW`-ooJyORsPJQldeg26P|4O}3TqXyB z)=dUBsLp zXk=uIH*;F#*#C-d%SPj+5~+|RzpXBF1@HT0ujJ0w%K5GZ z42d?1wk}u2-1Ej6NiI{ER{a&dj)H7!;_KgA`JUeGKfVuRUR{Fs(Fk_>Vdf7w$U49% z>2ryOc(3U0enz!3IGkfq6K@z9Z-L-8|I6IcruywRx>i_Z*+}2#%WO{ozoH)Oe19<{ zSlPJJjdb(fych>J?@r(aw*!rojcX5KT*LL62|L)Kc=uyB*G=_ZVEuLd5z7Oe`}0_# zqWGg2y8rk~UJN_L)p-t$&aub&+?HfW=A%6^NY15vU%#@}4Ox0UH&IU}bhW;RlxEfF z6`q%PAf`HztIrpQPwc7aWTE1(BhlCn43Z%?X?sT%$Bq4cu!5cg{!A9#8DSynXAjJN zrb4$j2^uG>H@Ne}>62;}U0(28cmI@Y24r0JN~BF#GCh~J0MG$w8NCiZ&le^H)W&#f z#f=3>8oUPW89{&wu(D%T!d*k5KS&zXI<`3^z4{Be_dQu5zm6+HH~i=5B<6&VF{Z^1Be3;0$$bp(Ofn}{b_GS~Nn#Vh z$L$XnzV#CcKB0EoBW3cWM{*YW64n(Md0+g?O=SDj%m)`q+uAK)m}f3)joH3K2fw9P zV@&8e<@Pw$xA!BFEcvsJV}fRUXLMY@{+f9!{h z-{=G%RN?D#oNXH3wGo8m^Wg>pewZDow>K%Ve6u1bZh|mDs~6RuDs>R`s1uV4J3S^9 zCK7Reh<5g)Fn3LeBQ-azK%A6+i0;RJMe2sH}UKj(59KsT2sXM(!buq}_0TFVCpJ3sFyxT7B<7O!}{`1~!Y zZL>2`uDs$dnt*LPn3W%BahOv-#7JqGb#f*T*+b!Vn!n?j{IYocBqIxV;y4HPL7nfG zSanLs0t|(?<&^!cU!mm8X-&l!9z=&hi?cht>pV5xolen8 zFN_fO%1ubQ;vk~YFf5OQoxIvlL&wyQLo9SF3kQPJoJs=D79|sZ<+0IAlG2c9ANwsU z$$%XYqVLYYg^DZV*Hm2!52I7CKH?t>t8+_yV)BjE#4oOnWkV!H_C;{VL|&XUy^CbE z@bIy%;#3XC;L+vyc;z9&{Yt4rvqb57zeOJx{VEmgBl=P31DHc4d535&t(j{k;Zc~4 zKsRH>0J%mw@>tMwa^ufLz&7+2GWZ#@#-7X^UUM(BJ8X}OK_1>UA5RSzB**3rduG1k zW>IFbyjD_EPUDx7XR`O#*C3dko=`;Kts@nRqpQ$AnNve0*P&E|b(e;5;@-V+6Ppa^=Mub28 zz3|r%#CvDQZX_t~2y>U~%)`|0|D}^C7w`1&&Omd` zYMprAg33S+6#G)@aoFtU_-r%Rjk#Ie#vY9gSbGLbqdRb_Iegpu!`OjzAC$72mBY8K zgpkdPBY_V3bkTm0S?|z-*xq0=UxpsPGSTm%DW3)NfV=)vdWoIW%UoD~%-o8#MWEU` z_ffRA9{c)+?6=<+*d_6xb&{TmVd*2a--r62Y!jDmu1oTH*$t~@r(_>?K^wn$y*f;d z{>FDjHQSH=vL3xC)YS9zeB;mVV!#V&JC7|rkCt_2>+SW)HQtSaS+rjc=h!ROQ<8okA>q^c%>`X&U+0tIItiab%ysz2LjpD)JNR7h z5R5Pp%2+6Kc)@ijKZ~;=dqnl+>)i-y*D%1d7!Z8z1ABAg%*%$S*FCp3ita8xAX7wk z7_qIF%e0$@(y_vyx2anTtx?o>`I7MBCJT}4-=v|&Rc7R_YKHBNT)Fzj;LIFClHCH> zVHJER$)oAuU8YTm3mgMtpXeW`Im4|7`9O@vGyMQVWZdtU+=>3m*AlR&`b#9CL`dVQHYQT0h4T z9#}_9kaXiCXohy3zoJJyNyDajMz=*9AVXoYJqRQ4>r1?t6ZKEe44pYvzdbArBv%)I z<)p~eX-%2MXl{EREbEQdtc5f8Tv80C$^T$|9*qc{|vt zqI(1P=A2180#)K@{OX^h`46Vbh5OImk}oXyyO;g_&@Yy%(=jmA0rUPqz5t_*GWV7U zqa0l!xTFad(#q&{opMOb?;xGX z*sBo*>lQqgU$BRo(YWMg#ZYXK2y1{Xm{D^Y`B6=*j+p8*$sRb4tXmlFtCX;+Uifi| zPJe9Q6)xoi^ZVuAN4ZL z{&#@(k4d_D%yxz?I@a7f9otuk?nJNUr&{Uol@nYVn=yCA?@AaC*V-9~a2j3Ra3-iO zglz0)$ESVMCtS7on~ad-;h1}s&zC=&KQre0?uA21gyp3z%0R14&n{X8_3^GAj{e5d zeYn+TrF&T}T+#^nQ&vx@)4~;!04ZjwIoCeU8j}hB!Kcil!{SKsFG0GUov;;j`F^+l zNr^@er(&d=8lU+iMZ?!n?~O16-CxY^ik_*%(YcU=W|Mukh!J$OB)}5_tS5muhd{yk z3zkN5B3BZenA&V*ie37J3X0$*^&2Z9$$|CU0y1A>YTi|M;R7@bjJKY<6F`tsC>JE{ zGow}*S$HWf-*foc9=WD)$}|+dDZ)Pdd9=)eyp_QvRdhZ9)Z6|E$eh2D^2Z2@H*p`X z^P_1GgL{LSZiT^Z3zM&2O!-mQz>hx-+f>oYEmEX+$ZE9ZqQ~1P!>+oJ=kEt3k7%J0h8^z z%!z86fkY0&dp3$_8D~?1xaN>Z1m#aKJt}uh+-Ei((LdVt*4BwPA(qoY_JiG09z@St zt*t>Hme6-U+RtNpWjrA~l`CU@>vQB(F#8a-(gV>It~?y8v7(rhe$;PpXK|&TOOM5U zqL^PnkLHE)P`3E3K~}x>_&3%Y%T|wxT#pBWxBj68@cTLapXvHDd-t$6VR2h-27GSp zaePIGbsoF0YeFvI02I2}QY6Lj#^ld1ySmfs-#Q!fi@fSDp>(N^*jJ%RQ!p3rST1ig zqXe)?w?&&eFT_YL4e`U*6SZDs)xd)1k(bfWLs+*&WaAAy5@!*RmHOozHR>KxHKv3n z4CPaJ*~Rz7X$74jLKYXy|L$FqJ;nOk4;>j5MS}~wtHtkEy0!Sx=5j{PK^0WW zW?jjAWe{7_ba#l_z-XVxu^YB|70_F2NC0I!^`KrkQcqS8DTb`$PwD{^OvkqMxov@m zai(NLqK+2-8e~vz--JPU`md<+qT~U)+ICnPgByyn=+6MMYMhA@TUdP`3!VOEK4qv< zPw+b5>v2*H389ld(|w$GWhhh2JAVCI(td~!b_FQVEz0<@3M;iqCwRwgIPX+?fxVLE zn@G_g45N80a|PN*)0FQ9uXamMs?0VmwI(=m=p&bbK1QbdCzO4BSF*V80_*A^ zb$F+P5QW*;KcKm-DCl6278%E3h?}z5%uH!w&yJpjCgs zx#|pmU|%Zz+tFlDWKS>}TGiG+4LezR1Asl()TS;QR=ZB!7IEl5*_!aYg!8eJS0r^z z)$mv4RWmX7Hme=ZcEmWRFBI80W-4W#JbyAOJZCDp5S=X%mQe8Ar({XSBxYKJEGXx2 zl|3pdI70SRW)Z`|h4|T0PjxOtjWCwhy_uUy^Mi5{7j7Q}K_HPFdA&;$I{fC)X%u@H z7C$2psQR5sAmDcQnYdjhTW{d}<+AY$JW<4P|6b^<(+{VYRg4z32%Wp&x8Kq;xT5mE zS9k3+j;DAt>;ii=q5L>ia+vJl4G$!09#wenC+=Vxj;qB+IE3KO;H*TbF_bb`xwzVT z$+IwnOWQ@F;3F>`*Hyb~4Z@>{jH`zue?t=bZtS_<7H$L;t%%>^|3Ch@A%F|;;rzN@ ziq+{a22|>=H8jd_1~ScF1CK!dfT6>N|*pI<5@)t0p%Q{V3la zg)3qLQD&>GPm8abDN{U{K zab(xrl4{ejZ+`he*Jie}q`Yy0llfgb-SXFm3&r=CXz_rn*G1&KZmc#r;kK#VoSSxY z8f1ZsSbLUFA%S2%8~WSRgk%FI@0yfaH^YBC*Khvw=IjqNB3vi`N!8ps>{5;@3bw=s|z|NXjue|Zvx^wYcg(GG7lm38Syy_TK z8VXr506N_tW_>3T$Lstm3t^RAge;3OT(C?IOK9DX-q|YRzxJlkq-FpE)Y9QSN{ZRy zp?HwhYxwA=2bt&UHboWre-Ea>En9K%DwFU1ftB=M5+av8f{=o8gN&q9ev5#ih`FB6 z^EI5~fNB!;#$29MoT}xF_Q=%paWu`}gT5TBnlpPC*ae?h0)p0_Ro$BzUqew>P5I~q zg{d%Gq6SLEuI;$leAAHGs6XIRyF|Z$)karfJs&#-Pxk}bQBxv36>zja-h1w$8o>tc z-=jM_HX_i6bp~1lsl;!in-s?%VA3F91nnQasaoeNgip#%>$N7M3>Qa!aXv>_(R1NB&H>q*nT zUOijR1~?;A!~_sn{hXl@IEr&jgAX?ArWcCz;#q(d%`aj+b0w{!= zZIj7BlLXcrgn-Og0Ak|N;;>=p42Fz$uay6BKnXC{+1p%-Trg4^4Nl^dhsrZ%oyHcg z(rYh`p!@TeRynw@APBoUt5?|y3gW?jF?+YJ*e98G!75MaNwyaU;($heq#U6r9I+LMRZq2SAly;^? zeB!f}{C|X3&Rfp{nViAVH~zGFAC?JPO*PCqT;g*rdGGnOor){dPWowo{H-Lmj|b%C z30$Qb|B7mN*k{kV=8&z+UBo>{6$$+O$-n<<&^i7~u%CWbA^&Pg|7n^3=ik2G0}yVp zB24?lKf%Ypx}*Pd(q8ff+6itscl}L;=)d`@peGlA32&S%CH8x;`oDRe<8?ql%K+Q_ z9@PH<&i~c7s23l-6w^qlt0{TQClmGO^xS{A}yUBSxal$9B6;aiT`aF_CHJ;+(+OM zJ<9w;U+BNMfqzV*dnLeVfAFU%c>a%`1HjZk=gXO%KaueM@e_Z47nfRrZnXPrL*&8V z*WiCQNq}z+Is|0%ZJyoa{~z6Yn&@tHZM8|Y{PzjXyF)C)d&WCu&r#UlKi`6#J+8U`?RwOgMaSZ8_eAfC86zP8%pf5n48QR=cq1^Hq^(&9jefQN!j=ioo{%p;tKYn<@P#{ zL9Sw1Z))}QGY&$?&N4Q8N^#jMmd|_DfAod)$neSuGbR_4tmLh_{fp@E_0v$$4b4{IJ7{1u5%ECwgG5Z)&kBQM&vd5^)=R zpXp1Uag_5Kluz6oIAFHj=`d2~7_x%x_9-0DS&W0j0qndI?HOhz!DWeQrIkvzdvv&u z3r@Z0k9WZPYzmmkGZwUQ-!E#Z6ZL}UAxD*eViI|QLJoq{mj}0TnAsS(lZx*;swn)7 zzmy|Ps=JH` z=8sl(wr(K&EZ|DhZxWxN)_fltyLTe+@deV*>J03xlr&_$>l0uW^%5HZtG{<)q}m4| zmiTc)wR}XbG&F>TZL3v3E>Eg;DDOO_I8g7sPO5P1TN;B~qm9?h|vZP>o0Su)@g2W;d- z_v=IdZm>CUM!y@tE3&+ju!=D38Ca!e=$G-VdnTpFYE1C^v6Ws}9BDw>*qRim@cB5S z>r*-NhME!$9BVxEcKR4jqAy@g6Tb!_Ky^2}La?&jDf|QG$Hb~4r3L8AR-E%Iqh-aa|&XoWvUo}a|EI!-R|1zVAu9d zMpdRmE7`zt=!=P63-sywQ+OaEF{WS@Y}WVvTGiWmT-OKCko2i2SYxe4j?uV3Vfl*L zHuVZKph-J_sN{XQGF2W}X9!tnFwD`}C?;^U$R%F#^k#G*2EL8@IhvufQVQuMZ*nd( zUOh{Y6fk>%am*MyYO&xuSpKvIf*5Wf+>ju4SXhOsCH`d2z539-O{}b67#WndP;<`f z6j5!2MmP`Wg!bD7-keRT8}GDxuaqsBKfqj=Ud2EB zNto;+w@#y>JrlJw+39KeQhN)W(qB8pjJXW&8fHz1F8GlM+5`ho@E5I7#wW?PgQ6h; z3FNyx<~B>xiy|koP)X}eyosxHuy*6ph7k}T7JMjK1DkJUw?N16m`kZdC3F@~3f6@f z`@E%xlBT{stF%L~u2*PUSFVRR4>T~wnX-r(bhp-sF0^hdSz`HDD8Xk=YG(OjsOL_7SqDS!Gm6K zOU~eni1SXUHV@ewba;|;GDg0a1MX3Se{d7^x$^MKB@GfrO{FEGZtjVo$K;Y!w+gsG zu8v2qYC&MZe>y%OzCL*(kvvXY=`p@YzZw0r9B-x5qWT#|!rk^TKiC<(gtS!O(Q+!X zU(SAGo4EE}KatUPRYY-T(u`-whF&+}{By7z=~s}yeRzFinrCSi4vSkDZC0$lI5#e> zWzkLW9doO`JerCYS|s|QeM(RNX{i6j-`1q{6?YPgZW*fgztX(V|NRh324(mTR;W?l zrn;Cg=Q+pV8!+!(k}c@{Ri;4Sx8pW494@z#`Rg$R(~_K6IqbC(No^lVanHzqo9y^L z-*S|Vyqmbe$|zU!mVSxVTn3cu}0` z=ebm(f@wo)jMNZ;0<~)*EtRJ(@%aE%v7*ehFkz#p7VR!!uDV?eJth}gZ%jtN{gZE&j$<9#86){W^?(>+d(JUU zhZn@lBT=gtMyjl`zzoJvV?D!r%0lacp@ImrK~1M|(Mo43qLwuWJZB)-@B%EzRfzP# z)nIYy{`Se5n5gjd0M2AR-(N5r+xmOeJZL4oIs$TU^M=+=l!t+&_I5yPeWT_IfAS7k zx~jjTz&zF??1xM8#bkM|fAcK*D0Sb%*u?Ri$$K-mCB!--*fRxb@+A$t^{Mi`k^Mzl zOg6z%V9~ekqgSnZ_fFLqXW>AO3uvV3fpJFI_+4-^t1G9%kEx5tEN!A-ul+2)2Rxo^ zdJiA^rlvd%M<4ZyT##D5VG$fk?1b6Nl1@4uOZDxYIV6ODEc^8^MtzWjeQM&`TRqOg zq3kHXagR4bo1deAQ4%^{5wMb=HuP;IrKF%rkozcytMrM8?O~vdu;%GBt`g{%2SGd} zd9@E0_p%4W*CE+_SB$3T^+6+Emzv<{?Lwm(n7ahD`bkrYn%Cp$Ax%|E(CU;k{yW4+ zZAg6J8^*VAepF<9HCIwdZA@%y2|w^;qWChNd!m#HFH4Xr7^7E@EUj9)Y4j`j;bvJa zPniBCRCcjMT@2^eHT?Di2OsZ@`H2!ZWl6)}(e}$43=PuDXGjE<%=S7vjA}|JHjy5@0GvuDfPLT~(w_J@VhuORIYR`1PLiW6WF3569xP$2WqI1U#AuI;;u{on^}TK_BYDODNL2HvdY&K#`wP{uu5}U_BmxA~N2Aq!%|CGUw9sj3p%H$J9tbWJitE*hFIE z3LCJ)KdwbN^nQ=N3xOZ6wd8H~}KqU)r0pl?+m>&@8xDi%y~A zsBXJ>Q|$|wfVZTC=-oM~+uy&7Q=}1rj{}uvnnRzYyGt6r+YEUjyq5kn*`3s6Ns};u zEI-D34-zt`FRM5kdz4wl4vir71Ul0`O zh=|e=0R;h(-a!OJ1XKteA|fEYh90UEk)qP13(`yIy@ZZ*5a}iKP(uk2ASB#}bI$MH zZ`^af^S!?J{&~j89}bDZ>}T(_=9+8mIUV%N#^tEsu6&UCb5|Rm*;j5|E;XluRjEnM zbMgwuZW0rl&h)ODV~52Aq*d-0om4k`AZRs&9InkFXBhuWAcvFu%LF!C86Cn;?pIeO z*_42=2cXp~))izMvE-cE&DBr)ht>9dYM^4m_M8lmDf*OFY@5>Awt(#qL75nrXRL4V zD{I_7dB*I&EualMyPtirZQr(PAPZ=`{%;(sxWr#DFeFfvB%fvhH!giGXUh; z^*d0UuncX-6?TBR>-_gjs|vBf2ug;y(gP{o>&C@E6$bMaKjW%b z#$Y8WQ}e-30OU7t*L`JTw;C%^)_W*XXhx-iU$_UJldXQkn(iazQHglXe#)<#mgpKg zZI^tSOkDXoU=QMwaBoMq#7xb`*$0`ml}R?HOF5#0{K@fF!67p4p*9$R-IEq0d$m!D zrblS-g;uW(C~kdBM?IY$D@qJ!J5Uq5de4UMjFp+J-N8|ep#x}L3*cC4=LfxyWEFK=oHc=i7O%2NK&sFS@emtd&Rr{n+O$x zg{~&a>}_zfHK)x2k4_;UiY3i{+Gr|w7CEfF%%!|x<&LgUw_DN?--`BC`sHXEaBLdS=d z+dz(Y%b7HxTWycLS$Fjzs25~CL^_ND0tM*1IO2CFVPmR>m;NzNoW(D`IEI^m%m30P}GjRcspi zU=DWoz!WQEZ#n?+;~L-p8gPtD7=^a@j;xmyPpM8RAd8F}c4k_CoDbU#$&6^L%9h*; ztRls87s@ijp?vgbX>vY+c7_cjoNvlay|r->%6_`P*WI{a{1-R3HtlR+K5s;XUb4 zx9$mLpISm6rDf;@_TctlaWx2pc2yS>{zhhGW4}Ms>7Z3_yXJTB8dMa`>fo1lNfxX% zigoEm0vYt~3{ZX*eoOa!W-RtP%cjf>hsZye3sUFWO?bD=`f$RtysnJ=zd^A7RdyoP zWJBMySRBZhjBNCrvi8Q7OF=#A2Bx6f9~yiO*t+*tX-j5S{2GzG_UX7GDYC3Uugis{ zsy({XWOryhl=$S=DEQ*uEotzcs@r{|W%pVSdF6(Y+twgPb6pROnpZyQ%#1a9*^(l* z6XQG8K+%K%5`e*mC`OF%{%o_QyAI4(YTR;2QH}bxAgRnz&s5X3`&cj{BbIH(ph0I3=k0Sj9RxIT%w`97+=$~pA zAAbE3yOTtAex(gB6kw@MO^&J#ijlG4| zknn+D{5sz~TFk_9c4BPJ8x?c=G~HV-0($CK4Ro>7VMM`-nw)6lZ(RSKTHiCFa zEkE3wp!!Wr50-Pvo+^GzdTRE+m_+^0C;ETonu59I_oP{?rNrzYaf7?%Kwg+!>~<{` zN^2U>#z}%0FD45w;a}MppXQc%_neQ_HpJsGx|l}v423+`%auKqvia5Eq8GR21e9CH z<5Ui6dcJrYz~{GKgf<>1$Y|Zl(*-0=9uHp|0G zo!Fnb|E&=K=UQXy+I9mqpUw+S1qN|k8S10R3W<|n%nS6aIo>xy_C$U9FNX2|*Jb)oi}=~(yX)Cl5d~pG%}y!1 zT-^ka8ZD0`fiZ@Q)9y+FOS>BLRY@QpIBi?@g^QThPA<<<$%=vGbzUvwqtB;9OU!s< zq^FYO=okeeQmc4*Gc1AqY06E!m+3ZRaS21^elidg~Hbj zHo#(H?SPnAZ*AN=cFkD{onoeUcB&1qrry2z+tcN9hgy^B?8xbzq|>>6R#|rQDo@CX z{bcR-My;4uMe`K$u6JuI+Iygg7{5j`@QiL1WF@C>c-C1txzvR6dhj!T1FIya@8gE} zRJ~rI8eiq2YfBp124Ru9o%W~crt6aZ`~Es^DPuzR4FJGRMGQsDh0?J>R1rM)m~m4T zQi4Qj_8!ENa<%xO%YX)Od1K>e5hitAWXpyji=%BxHucMs=@3#x0Ur~tYJladF>~|S zMpF^M6x^YjX#~3$8VNKkVe_QY1d01yM39($_2^`t)&zjtoCZ(IWvC@rXzrDOtj>Ng zvI+%H{;->y%@#!uERJ-RO{mm9@&1hhIi?usCMpkFWaFC z;a0)#z2gPtRKl&6(4k;eNzN&9+|-rh?N{_O-#^(49RB4M`l-C+i%7+-Y5M4Msg0EQ zsC0XTG^QFnvHby7#&e{#J|l~p&jq$e!h4X6fA+1thTq^xj{ZP0_7 z_Yf}!DO{rAYgjZ`@F>nS=X>pRE=GFSZ)u`i4t^K|H{20bU#91Tn&`Qx*H;kgO|BKm z;%n(hsOSn6hs8%1fz07x33p9s@@y@xhT^!uiOr*QmU^EEy7f+ZLZ>nXW9;GqnfUMU z>Fe&sW)$8`iI03@mZg#lz6;zI9J%=xzi<{>j8nQ&1kk7`3xDHkkJ+2LiTLY4@1_6| zO1L?qb6iU5cHHMa^_}cu;mT@Bs%Okm6ULwrQB@3or7$sg9^E}d?{^xFaK)Ytlv&S@ zlu`cP0&>BOs^3{~WbKh^^5IfLwdbk9))BVvMSvCYplZo9yCG7_;`2=eHP?f9Uk3CC47XupG&M5j4Rh*z!D9zr@P{l?ofb_Zzu;8D zp>C*|YXxF9)X9Sxj;Y<0A>4oVG}%6v0Ye56`RjVO2}#R%CM5w{Qynx%$7KJz!mQaH zRHH~ne*zzyZ#s}Q6oIBuT;LG49|~<@pLbApQg4?b*wfSTDdNPKD!d;)j-vTcL(62O!h($y*`DY3jma7U$>35FcB=)_JqC0m z=PRUm0RZCBwByM=Jp!*EE)c_k={B;iFt-GE&w6}$@R!U+FnhIc>*JX!kr@Ro$s#w% zmh0)p_|7z%v}xLt?LN?8N!_2dL(QJgSTyf{q)G`L_I%(T)JlVkE?f3>pHi@6h*AO_ z`iY;IV)sH+?5Y}i6vRY640|pwN0s|hne3JzhT)|wXfL>H(;Aas8f1XW1_=6T+qHiT zLwbmx??CuY3vh<@j#H22DUyNmd)V;yew{#_ZmY3+N&Dt$4Th-~I1$p$zYj|l} zy~;LGK4Wulx7})h)77uaIVyfpcg1}u+@4EM?!p@kX?`17pffjj)3H=nLKD*LZa;0! zo73;qiB{9!sUGmjlLH~zj9BIc^rhTSZM1jqfZh&O(qXrxN3ZZL`7R=w15ey5!*T~v z{}tN*-~Jt?0zgHCE~u`4{!8?$yYTz@l>{TFkNe`Bx9 z|6A_O$>jJ){~Q?mCIQ4f(Sxu4*^m`5b|yXaEj;QE!`{EAIN+iJpcC+TNJ{0O4V}FM zCWWNUcqIS4AAK1+IJJNP(IdUfB=V0|?q8AnpLqzzz5w?|9R;D^U(hf z3Gn{~BmcjjbYLv=blzD14(mT3|{BnMhhOL6^53<<5 zo-d&HBz`$MI%pYP_s{Z)5*sl5N%DD$_#b`wmnoOx2*`@=*?+V_1MUKYt0qK0xBuLS z>~aO}4_+!rqNmJI|7i2b(~tm5NBGX|Ki>=eJD+#c638f>r`v+=|Lnn$0ONM_1dkj3 zaL(gjyQbn2{P-Myllo^5j_7jf{z^3dvzYd;5Z9>>>{%yS@BV)_GJKy9SUM&De*XV0 z)cfy#5U_MPuq>V~v-ancVwYdwW&p5scjYbU|ISnXPwON9Kga$5kK<0N@@99w>6FC- zT-}4wnEZbN+TZ+>cn480%iHeuMW%OzfWv1QmE75V(ZYJjmO(UX&a4E`Pkqx$m@HwV zQ)Dv%GcH!Zos~@};a1wC<*H@$^~V8QKAWj>^8ggSqRh+=xUV-}>trTXVAc&~F+_d3 zB~a~PIaKWcXsYpO=e@C;sjjNFo7MyDT#h|w11BAkgQ=oE9VBwvxf-cOBa5L(?@1Ax z@ku-PS<^zD;__r)G~?E2gS%~n>$2a`(ZumWJC$J^4`{pD8&%d7$>3DP;3nXe*3g+#AlxvK;<5Ap`7u_ ztACOBxJPQ$Wdjh_v`b~wMHiVgdCAty16rWK%#EDP$s616&*ONeJ}U$jz3mJyGxwD7 zJV>ius0vON1FY)*t{3; z4lt^Q@woLGo{5DdNYOOC1aL`%N|*-1zkFCY&2UtfxcX-2#CQ#> zi$q*97I;f)+s@J)pKvL19^Mi5D2;Sc*x3-WnB>xxk*eg0<25WXY1&RIgq4R-Fotw0 zo#mOa$U0|H=~$L8gJ78lCj#*PEM91ee_M;D9JVNb$rBJVht?V)y}F~AtKW~@^=tnS zxHEBnj?5Xxd{l)A%-w%vKPzs78b8~0hy zTneT~EvQD^v0nO%v_>?lLDYR-8UbQ=8-P?XYOw?2n)pfOCpdJQkVk#=Ois3r0E;NDJWGY88+|kt9%{7ab z8NVqVtur^|Jdqe>Xy>k@(`EMSk8R-ZUnk5~Yg4%OTad-OU$B|*5UUt6Nv#VyUp8$& z-vX>mu%)O}Kqx-L+A{}RwDDwd!hc%yU8V=*wOR)^XX2w;E~xWCh@>NNhnLQ%-H zFxXhK1j>6~<`9R0)|A?r%*3u7mEbUN{2rs2y)Bj8p8S(_BVOQKn>R&C4YGKz&|DY= z;KhtQ8fr!7^1h@Gdx~p7)f9wXS;qJOyq*Tet;3N$#y8>%tZR7`YwX8(OzQ$0yIG)> zID8Gf$`HSI&9K}xUp?({=gpTxS*c#V58TEyec#gYXg^yw86Y-us5`X)ANh>eAf~3a%_VDk^B~!ad1JH7Lb-7%LY1B{wi9oH5=3GIDS|S;Tu5 zfF^wz-E$LAG!XCaK$SQNJ88nygd&mNzL9RM}$3g3v?pPWq5t>FuwyVbG21b7W zjNb2NbzrUSMYYck4Z-TVZobzTn`f5%U1GiYq!|#dQXk_pJzc0lRQoj^eg$rv^PAQk zyKJ@uxQ#}N<;>+iSpslCp87X&!HMI7yMA3(xtxH!(GncyjOzm=nW)ISG5LTXDDJSA zcjp>1eYYg>1cF5yjV(erK1^?l*c>oz9veCgj3IU0(I46dGx%H;_L3dyI^o)Jdl|4_ zem;f^PX-0f(dnYrhwn%6%gH>aV6Xj7w~=7_E&+`*qkvsQBN1QY=zVolXJZc19+l)L z`=<5xfx2F2T~{3UxTii(QIZ z&Wqs}O;lgFi(Cb+);WmY-fk9D$Ux^_;N7t^TAEW0yG;pu>g4t83*8AO)(K!Y{a#EP zaASS@dJV3WO7>YTS0x-hC^KPyK30<1u=b}M9n%saHvOG!N5Er=#-gGkJuzZevVvfz zLhbz5Z`fBO8G(_!JU}S$eqO$Xutq%D8r3=ifdO%@3z~K3%-Vk{4T3q;`zmN^vKobZJ4D% z^punMCJ_u#8w_8rc&E1yZqGJ|xg{7Mc>2NnQqo-TLhe%y7>ObIVX{}w zj^iu4MmjPdj`9gRPI+};u1>qv(chN0Mm}$67 zV!7cuZHn=>uxNIr^^DHi`EfFzWT?DcvE;ECnoQLFjmy;_qG$g9=(W&!SH2tl2yn>4 zY?mo@NNBS$=#&AzwnEy2+f(iDo{^ZtnMC^|=%n0iALy07`4+)AtH4(2%i&=1kzOF} z(kPvzRW1JhVj1?U9G<$CpQAUPpO5-JHXKl%>08K9RHvB|8 zv$0LvyNQTm6_|W8qRu?I_$+!W*ol1`|9p20aBjLj+l^l9F?+g^*5&cLyqAub)WoA) z9@=C=Qagj3;J#CPjcL~j_Xx?UmhiloL1Z@h+htPer%7fdy>)+~EHbw(_!E6!n}4&O z5EG)zioj?gtyz0+Dm3ptZY5vqvC^(320O_>i`fbYx|Z!3!}S`O>Q1C-C%<6%Jldow zBJU#ij!@R~?B`###gxUbH0}#g1Yo#lM}#bXzBvG(ar{sy`4`X4pHY$5$#1X`{G(R@ zedu4zL#+D-?fSP-tzp|!<*I-wI*U_B=$x3{{6i@>lyKs6o^T5F6lu@b%20Hq)1>`( z!rweR&LNIR5IVnTZ70`=zbvs zp@yzePio_%8DxzzRX(ka3F2}FuVk9^|9Jo9?t$tAta_cv>}mK-)#<`#jF2$F<*q5% zdd>pPMF69_G2LwY*ckvhx*$!jtE_+38H$~nk3W(lwRWdk9 zEnt}G5&>9X@TW>9YAxbSosL}XYi-;&CP%`~ogCQiszg0hBBhUK;W<>f1H(-^z>j|8 z$`TZaC>3PiehKf8=@fw$v>42h1hlMO2}1`lszPGgX1oSk-2CnD`d?}6BwoS+^@*B3 z-8u*0A-e_9lYaHxt=M4AxAZCUeTt%3*+LTg!V(;Ub&ILS*ST(d?LRrSj8QM<$E0sI znSC65bz6P+!UT}a>^Q7-BQY|&dj0cX|LE0%*7R>ft2B0vS`NnA@yvg~I=Is@C>~<- z2nf*rLvmt)8ev{&WFLJ2yJorshgPn7%pJ*Y92S$jf6#cGL}Q_-eU{aB%r`(xatk=_`Fv?l6b_wdz00+ zCQwd>EzDm{rUdjut-~+vi>Y1ZC|)Uuaj})cXg70qSte&q>y~tjUbZ$(Th5cTAITCM z0@fw@i%)|4xk$9rUL##Ci;SvzS!fvvOZD4i_Jk<;*Z=s~ z?GlOZ%H20Z@_)`zR7s-$##Z#5Pf{GmT^-(LNAcYr~t0lY%Y zi#tNojCYR!(V?^F(Ls zv1*-2l86THA(j~5Ul-!W{c~cE-q@4T^F7w;XbQ1m)^S?yqG*+=W_#l!9))Oz&MfcM z!i&^zp(xmxx?>2a0WnM0{CSqW*kD}~&raWxT49o(Z?ygrwR-fYdC};)yd(alt?cvQ zwNLNRc*V6~VUrRu8x?|j-p6bI{$XkCgf!_Hb+W)^z1nKC1b$~=kS!lnk<-Y$(0`s6 z!zB9lLy=zHuP|l!ZQ*A?_`kKu&dy)F>?jkG2W{*4xHeYd?w~5`R`l*=#!(E3c%1{H ztes?A46w2%K0u0%`YYCOuJIWc#5#S&h{4I|`M+^J(B5Mf%-l7UjI38G(QnX!88>i= z(-T!MICIJ~rHZzwPtQp3wZd}GJ`voV2hx=4kwKNTNLk;**1Zu&>NG>{fb~15O%H|h zCfifzMEIM4^D2jQr%n69sPP|1zH0Hp`D2_`~CVI=fb5$*&XoXhh$*^&$JVC%HK9f@|>s8RzV;_xzpcBAERK5AFzfisLoxl!G zW8Rcer{X;#f2~|iZ@NTphVf*>`mu4ZbZL*h7M{o5eF*}|9HPZ1)71XTAAgVGt(-E% zV%FsD$rJv;YwyE#X;%@y_Xc74D+AvK8;h(IJfcdY)Oq463Ow%pd-4mg4D@%1LkX3A z=$7k2^SIj2WCYO&z>n6Fe-)VbJxc7r#BlZWgM9^~W`eF!lG9#;AUoLZ*_|wgD&Hrs zInE!=Q5{0frK+n55Y2er10lDL8%ixn)!t$VAj8UjV-K`ONXoYD&F9dv=3&js3wdHGyEq*zW`E z;umh`9?LB+nv#!qib*dSY|xTVCwP6Jih1u8S?)N}o?4v$8`79kJ`gWb=#;JbnJr*Z#u7Cvzvy}UeA~&n z@^-qk=P&(jLFoPir|%eXiCbUgACXkrPk`b}PDX9iY>%I%h<~{^q8oB(m3=l7ch4^A zx5*5ggf9Jq$ z;nkTsUKMqLNxwPp1=MUi^%I;xJ*N1g;T_toL9<%;0cWHyH>*zBoAz6+?x|^a4Pt+O zeWYQldRr_?!pj=n7?)b_p<2x%ZYxlzH)wl<*GpJx=;=#B#v?4&C86YM@!qSividiA zm_isuDAOglS$*+Uv#;$~P*ZoBB-IY@zw+sCafw6YKJz#|>G{GHhYWeh8Ss+&`<^t> z@WRzun3;-%+j;FuZz7=JwqZ6~(nKMM$L^YM9IU!Q_A`~To!aF`B0*(u7N49N%)vRH zhOt4tj$<2nmX2-#jA2ZCv0;NYl~(AA$No}X>}jw3$_`6wYN0n=rKf#_sm(?FLy19? zK{(xmyEfy+LW(*tJ;6yuZgNM#Q0(DcgF7*B$;s)d(kRJZpi3QHi8a2+o=e<$B@DE( zzpgf$_AC8@~uTnP`SObIEIiwIw7c);3Zk$}Pp(Tu9*TIne+z%LS^_ zg?E%A5&#}H4J?x;h6Tg-I%%P)6CA$hC-@L=+ar&X__0V@Is9QvkzU!8>1u1vYY>i5 z&e#&YW}h%*U?f!^y_ zxl6sy^)!5lQ!mp{_{{o5?Cb84G~go2=yEu^IVejLN3N1_fIRQoV_mIiVsaNy69;lN;1sQpgFI zPSqp2G^%fsiuM;f;1VK1-?%5I+0Gea_m0g(DUJfUjKXVJB{$o=H@SBnNMjTMB~6`r z+xflGS?vh{ti?CdA>`&&>aal?WQj>kwxRp;)j)@DKF{zQa_053b>2g+`6iWY!ASOi z?rPub!=VU)%uoup%AMcYpSI;3Q2t#LfX1{gbA5tE2^?SFqbN;IP+#*6Ko2!M%l^^5 zYMVrn*bYPaMu`4yWN8GmlzVS(lUL%zK^!lRvdKiBgUvRIZs9koMoa!m?J9@kg$HeF zyZS{}BxBCp6?6&GPiNg|ceZZ1?zP$lQ1{hxH>UR4ARS?)9jlHqh$`=zH+*6DRKGq; zn=m2UpQ75zxu(|fQTc!|^i0fay)HnlAZc;W#GuI{av~#PBZR>Q9WbF> z#1?@3+!CU0nB|sAWj^a3lh!`wqb+byD>MSgLw@nJ+#Wwuw9F9G%dWd;3e%r^p8cDjOxfX&jI<@_4&z(BxQI&8;SAN%L6&jSm%W zleeeX{xII9LWHDa9(RrMW^L6HgMOxIA--4)5nG2|yR+y3C%E!%G69_ArXv&8t(o%q zbW>xO?@_de(=zFYpFttyUmTR94m|{3oGo6Xa!b9dSY>j3+lZZ$^{vd%dNi8zCHz9#VS^GlN79syT?^Tv>lVmG{fLT|I`iIs{>Okdmx2-p80g*ADCVg~)X|~V*7OEAv z8lXa#iUg3Xn~Wbrmtjp+uVky_5 ziCQ;0BWVkqCO;Cj#rzeFZ2j(ie&S@PO}g&{)D4ZtfgX(xyYnVnsdN>-I1rn)3OS=- zTVu-6;Ni+HoSwuN*J%fdP7`v6c`7ufYR@GscYWlb|Gk6l=!rQ#wQj^!%lBime#C|& zLbs;bNorE+*_EkOFIkzZg?WwUg*wi-Gss5X;JzLhu>a;XsC-|p$~@10J9LaqC> z9)9tBd$LSj;wehY_k1L*;+o%nu(s~F-Zfvrb_q5~aP!+>no7kFDBiE!n*6lfnxfJ+ zV|nZ!Z*YF4?X?P)T+pPX7Hi*W)L-su4!Is9nKE<3xP402aJi5IJx5-(th z=V^#$5rberGr!#rG?g9GU#61}un~Nczd445uDVg_Ll&X^MTpVmPPTyDIAG&yZpB?+ zZTwaES1myQl|yDk1CkH%SFRZ;2Nph_%}&2s{-G!^c?_4`ZGSfQbQf*C!8)d8p$nIwJmfDs6L z6KW>Kca#Jo9m|dlyV=OP@8)TD&AK8Ue|G?H4A)gKNxS|GTfD?~g4)_X>cML|@*Q*F z8PopK@MF6f2KGlp$JCKQ1cXtmgL7W@FDq{H*JZv1u@AmLK8Dlr{EQyWCF>Xj7bwb_ zd_z8=s|wxTqHzUw1BZYo>lpFj96OsI(pLC+W}KINc-U zwtZ4ch}_9}7_;fqF16Puv?gOB<|+STo=tumpAmk-Sa5qrE0zcJJS4vYmndZULgd2| zb!n(5gUEX};|6!JiCusskNC`t!Xby!#T2`u==vJ{x+54ddpBD>NS{1IzBUYA!@8mg zI}Rs1hHL!%TyU-3(~cugN*GbQ74;um=V;`zkY?UDJf^R=Klpu<)c%f%K zEJCYT%teeFv)q?eHD5GVw#C~L*k;a~zpcg@l%i!%jjnKbHtHCH@ROf_*_`Q$2?9USOF#Ld zvjk@tLcW|lJW4>#R}nJyq(mKB4yMKINV#pSmHHiBl;8Yn%z7GGy@r4Yep+eMrTto9XS_6Wz0O5a}Z>mE%x9k2)wes>l%j>QRj@1iIQAdmy2qr&~Y7 zt6zR^P4!{(;PO1T(!skM+zZhUP>{rIrAC6gMckBA85V$bY=cM;5I(vN@Juq$e2_kMcQ0fN@4-HOb^W+PZ!0!S$lDZvrP8x zqk93^+Ze8S7d@AQcdC)%h#Bm_#3W!t6W1OZO&TQ#QGDaP8qa6*bJrYo{LG`Tz-lUnWjIwDIu+e)?W z?yhsQH?8yD?uL8y)-ft(lwLq>_S*<&kgs374%{0W>pzN?lZ4Stjb?q-8&SNN`yrws zECTOOm5qF{{`j-L$xE6=%H4!AMVbQ%(xOuoG0&&v zjAc4w&S`L48=Uhl2f?{(?kdbqhJ1>1mqT^CXZB8nPaHK&Ju9j0mE6Z|@YD6Bi~A|S zW@|Z1y27nK^yDj^Kyx9Kn%y#zctPfigQH|GKC@@xANDyLBo`V=hBiOhzrk2L>Uo2hNWm8VMzcM>aBr^&d$Jt~|>`z?Y%7ralc0 zLRk{EODSY0wM*6EE?v^jIqG(>CK1e@c*q#U-LGU-BV8v`>EpIjFH)Pk95?)`E~~zy zQuT(~Eo1NZ^Ssf$v!^U`nHc;RA!L7*T1kEIiw9r(cxgC1QsCYh5Gs?A-KFJ19WKv6 z-2o_z2__1Ann)!}Fp>VUbrO}&7>UsW?mCwrEZaYK*4g>+t&p1pPJ!FxAb-p=E z?@6*X^Vcx|?(V7+x1X);!ZG1waFz=WF!of-XVbc`LS0CPXyUXFnWleUbTJ;B`V&+; z-X?!I;@L)YJu64j`eHZryM6WeX^X|m-r{4_jDz|7`DsoGXI97?IKaQ~*!Ee7V7xwf zOQsCL(}jC#a}g64@_Zu_M260NoV)6?t)lZQntAl*05Hq5u`-^w<{Dw5; z;vQ}`c_(F1T3_N`8c$br|u-WXwNG2UlVM{qa$E;@X87cL@oN$#^mL(d(+Pje*K{jl=W>)hx0 z7nlosqzvb#*!z6B+$+J;LD!DQFVNPt!W26`Mpv&}T(wn>!t03LU%VJ?;BP8Du9<(x z4%i`no>Ptu!j#dBA3eD&S1Wm9oZtK`x}Gi%P^Faqg2tgfLFc-WK1 z_X!Vs1uoDVfF@Kxi=vk`|U4WiR}Wp7ul?}%757MWDc^OQ%|y6?^7 z@{~9#7KdPO0{=!kd?BCxuNs#RTCP#&lXYa~6-i%3^0k%0t1nkX*j`iJEXl=8qnvbJ z;LNhm>(3V5^#p&H+MFg_*&cN`-||2F_Na5v;WjR~XsSy0EYDEJsQ-HV!nl}U@cggx%+!F^1V-w5`-AOxE!suq)dnj$&>&X8 zf48bQoR*!{Td@7Cr@ok^&BF^KU9I@U&af+Dl{CN^=ZiUa+2wa@^3H7kc3rtP`rw`Q z8?golu>>wPP{O0P$I=qMs6aJ|$NHe#^ZThaoT1@pMMgdBp1#ZK`c&w{l;x}!-{cLn zi{HS4DwsA3t;1)mpZa^|tYJm-cuVr1)h~rp^(z-yY;JEwzJlU=C`P9Z(D&Il791;O!6FflJlaqR~wo%~i{4BNaM1#fIT=_$JY(8=OS110kIi_i%F zt%gVMpi^fLxco2J7FnVuVj_OlYxSqUms31D^qrC4+i_qjs*uN@uMwA_tRNX799C~3 z-$NXk0uB5KTJ>kfiz6Rlfo6v@4rnl)FXcNT!u5qL4OB~1!iM(ivnW%)byMJ?Z!Nit zM=GZ5@cDidA29@#_OXBHB70IgnycAY=YqT!kS3K{zle3_fuylD*}u6h^Y#M`B7F?+ zxab2)RW5!6_s_r2ba!Q$TO*Hg<2px{WYW=a6bH|@{}6OZ`k}? zC;w-oORngz#-32OhUavB0O50WGH?elxntzEBC(et?x@ok@w_>=5iXdre;@4>MGJ^a zcB*AVW?=l9rH;q#(<2r=zgEv&vSNUwn^rMx!=N!%g~w#b8ZYm%*|m_oT~0k|S03Uw z35HX10zMfPv!s(oE3(x1-E*t1hjTA0jg-k5?{$XXbal9KFIwHVc-pm038}y!of8Pu zGW8@wNP%-j*)4?DUh6NdesF1rk}JMztK_4cm=nfnaD&x!6FS^)OX64(}6HcHBeHBas)YyRev)1VSaElJaisjae>`!rtze@EiU|1@4g<~ zG$xiGY-}-*@(>`v4<@hfmR*gMi*Hwn@+wkcPYW5&v$F9tB;E)63AX&WA}4x<5@#Lj zBVk)P;k58g?r`f>`I`~egSkmlLVB;iQ$@9mD*lz}qV~vMBGs@Wh?=?0<+t3;?+Hhn zzL%-;?vdpPJMqZ^TAGrLk?Rg+M^g|j8B*ICJ zsrjxWZA$K)TcAEQWEoYJ0kx#SOtm8M%BgWu`m2eZ=w=oU$Bg5fj;p92chSZwQK~id zRWE2GjAg+riRJHtdl+*MwQ^KTw~!D zF#0O`qd7*FN1i{|cJ|8oOf*Xg<)%U>?me05ewI?-yC8f2+cVACElu-=Mq1af0y#s~ zO33-?V2G!li96Wk3Q}1y*-i>O&0P){+Y19&7jt|LsN)c zw~yuV=3UjXrg@q3xgd{=H=|^zPK&r8`}OO6klLY0&s`#uW0OvX8y;@g@4fi;(iz#y z>8PL?_t+d5&lnt29yQj~e{3m=OU-n#OkT)hA0PZ8({PjLr99i(F{dQepa;9v6&h$4 z`JF7FF8wN@mGkD)U@9^!v!N3}1FLr>-NzNblQ#ncB%(|2)m%}ww6nrb_Z`hbw`v4N zJRQ5eW8L)VQS3zn33{w=PvzqIMRIun<9Odr2lhl-7^~ib`Nu)UUvdwpm%N@FMo};W zM@}zdKoB6-IbSaC^zv(gPmv4>(vNkI=r@uevW_&VX6=+lb&K{Rm1Z1{yGI8W$= zu<)nt@dMpsKB;J;*}A4q2Q(ozlj7`c`)9gnNmiSb1tu}u>AR81RoCxIMgv{7zLZC| zh}LDWBps zB4s*O)RSu<(bt+A25dEZ9ZB*suB`haeiytFdB*5R z%v$&t6aOJ9z?*6CI4rf{_$N}a+09`|kZKK;iT+R@#4>nt-PIBB zXJczM&3uP7;kR>F1NDW}29|<}?mAhvaQdu;j}6{U-;f(y&%dA&+7{i2;rx;@z}GhO)jMt&s1Fk?51E@C%Wf`5)r zrbS7C9wPLOTLJ9uO5yOX_f1FBX~nxrF^U$~qOF=3qGSr6&O9)ZaYO>>=Vym+6@DY* zh(J&MPINYb2SKiqH?lnhVD!RE{Zr`KywtB%KauRPBqD?T2>>);K7vrvZOkl$?`w~d z8tKXh=7VheW}VKW*K1t$jNJWV+Rsv8ORuz}1GujUk9%iorx_6k5dIB%fyI zRZN?r*l6Dc=n}e0W5Pq|F2@HJl3TVVTq4@}d>8X|vcq?~iZ}^wc<_^CJ)>Dak0wq% z+RMvFZR+?gwkdfxQ4dfuO1^x(DKJSw0t=01ut|yDZzUbcSJQl8Qu(z!+b8??q%CB5C4m$&pgqlM|^LMz? ziq~@Q{el^xa{C0nTHj$B3bDgsdsm+o4E#WPaM4o7GO_dHRwPLN+|jHRQlrx2xwVylo@Q@tkfCvczx#aU*;hI6LKNhj=DYF{ z&*Qhh#T$50$xEH2GoHKzCXQ#-g#4#*IO)wHYa3?vQj-d|pg0i~@ea)9h*J8km~q=1 zdS%X4mAmKpeMfE^3qRjn33QqJNXs2t5<)&aU`M>I<4E9u*wAt*ekrbh;Sp$KT)>t{p-MQ4@ry5d=Y^cM^#jBoe(v5JB|bg9y<}^cppK zHyDgA+K67qC}A*+ZZM4RvG;z@`Oe;_t5G=-PdKKQmB^~ z5ZYx-71HR1zH?8%&K8tf4zHgcsD6VpDRbrVQwRBiu=HXl5aJpoEQn%a_D_~;< z-a1(E>&vo)Or20|cw^Oivt8%)adDBnq}tn^yGwl)QK`2VD?rjFFB$o?p8LN?Z+oE! z>}g;IgBZ08UWKcIIF;kx){c{Pv4Xx?rcUW1VagqkJAfXCZxr^Gji>foS1AY9&Bv!%=C<2(8WSi=Q(Q;mvdqu+a_bat^#{GG8hwm8NUq*Pron9}?Xxn@&P4!HDV2SI?Y@J6|IT!1>M73V zBz#q0muiHXoY1aRw2bMs+Zx{@%3)!_L_arFJgq!rdDz%QhzY&#**9cd|Ma0c_X4}L zXD3bLqk;ej4~3So)Go8Rweo{P%IhWXKEr=8`Bv@EQ%*Y#VG6t}wt zS6%+ZwC#iC{4Y5l4P_%APlb05G<=wiZ6ktVPZ&D&*r)dWcyDuYJvd>=_&E5i*zch| z*-qOvLhd=ky;4Wxo^=5~m%{EDj!33Fb$um^c~<*8K4TP5vb0OnNk^!PDOvJMcsNL9 zm+{1Q;33Vh7*~HhC)+vTgD~Ri;=IhI!Rm^%!7xPs!R~$=Au#&OA!FrXX#Z6dlD6f`#pCQtUL;u!{ zyL%(9q7-W{AFOz;8-Z3%yv&dofTOm+6I*N5o*~1ESfi+(A>)U4bydztXtoziQxw}p zE1Qq5_lE)teX2t&2tod`Tw}3`3OLNW0fcqQSh_dkTomu|K4Srd6x0aRX6cY~Ow3O} zbek{R&Mkv`={iTWN;KoRLc&Yew8|60*63HhEOp0#_V7aOwQb=%BDV0##$(Ac081!a zb3RYd#`Bw5J99Fub$??9rMBiW z`M$6iO|AlnFw1E`ixA!VR9U*=5Wv`I-B-O5@S;C{PH&y6$25ysBWP_bbXE0^;=t`k zB_;j4z@}!fr?%MOQ0@y3RXI0sDR+%(fDm9~a9!-1n^6)?9A`V zhcFM)`Moh8bci&S`>66vHeU`LWgvXr-DqG=jgX25&6#gCBSR@ls ziB7$Ha_jKCey&=)#(bF;r=rlOT@+H<``33B2{fAQ>8d-wf5F;pb@A&5#`4lHY4(#x zTM}Ju^ri-xu@|sPTHI7BU^NVtxv}qYL{7c;LYw>n}PV%1_7g_}e0GFAqwJmF8F34oOR-zrbI;OcKS)g@Q*&(H4YWQ5h0; z=>2$|ZSo&Eh9_qpQo`j~q^nSSXuocxGpEp%u+@jtRU0`9UzVzb4ct_IMOoW5wC9{# z(~fu=SbTs*FE~%a{g!HkyrWIOZ6+H6Od`5ZPHj0r0{Uv!@j94BkF|A4+V)nM=#R^5#MsS%**)fJP`i<%Y0 zO#O0yK+}n|&p2{5%tjJuSwnsSwF{Q}l~Ig}}QToJJ4qIUt(04BVRci|^DKh+Nfi5fv{d+f$0WcjW2w zB9q;POUz?qk%VLjVkMD(T=U+X3x@aT zp;m6+wj}(qXeOO+xG?%P&}jrHY+Wmu78qvF@KspjOe0;w4)o+>E8ERO+NZ1CZ`FVW zi8^k4nxUazIzB!B<&)rYRaw?sON_B@&ve%|cVlqX**zhSRtT;&0-DbX56MKw{O zvbUT_uQu(c@J^#{78efXI7zk8iQ~7HKMiN@U50E_ij@f%td+wXZ8m%&GUv9u)DRyz z8PuOCh|SHl-LZD+2GT}CBkWSU&?IvC!$nc(LC4(+innISchbR*w&E3Z(tR$V?j6@A zZ`PmqObE4wdW&=)%APQ5VJT6_Bxs`}&|@y?QR$*#r2UFpw;vfUtS=qSJhz^{En~rC zx4xqj405WJypKaP6v%IK{c;T5?_Hr;#QL1Nmzc5uQ!?!F!mmOBUg+F#&a=4N!#T5I zGJe?Xbop#mj4iK+adH?9Z!!m`PUOnEORzi2K+J6RC^#xvyetYf(_)&qg-7xT*>X;> z$+EwJGO72IV10mCrk|ABQi{%G@zF!Sb7c&`;GMP|0$`sjZCZNs^&ytdh+Ob;^E3Y7 ztHLv~1~x&MxrIo>Z+#M4weV#z8Hx7vJ7j01)Zy104g)9cs`F26Xxl0U=BIDpJZ%sMu}#e*GB_*w1UEzo0!S4QMCvui!}*Keh5^ zfo?yUXFEi;3v?bgGr%(6TE*Bi-YjrK8>CM88l;IH2>N@?(%uoViQkL2mI(9CujCAT z{2R=ionylKFlef1h7K{uKG35njohXN$2i-N9s`^@;>+B6@-vz`|g(YKq{O zu${R{gXIkZU1JJ$H>CL0w@!rRELv`vsnv$|BialSbAhASm$M5U#Ud0g_T8U`5WrHwAAhFuF6M$JkfuhA4&$}rKurMyes4>`eg|6ERjNj(_a2Rv;Z#4 zKmjj4-cMi5l;ssNoAvmW^vF^U*%Id$`=L-n@rexl(J9lkUzo9Wxtrx{W@~l{CK358 z90nqpLT&x*r=DG9=mf>1qGBiam`z=_x|C@{uuH$Qe+6T^6Md=Tsg-1J97{TI!mfX( zkAvPw7;lnEpSB7%jT9_&*tz3Cj||*D0BS8XqM2TyuM%-dR?_wz?U4N>FxWRqwU2Ga zbGQAGo$lKSTcYx9LcqAnesXzzAdrq|GPfD-xPT8{Nuq+Fjb4WxaC4PZO6x3?|RC~QgxQJP3G zSsS2D=jzL>KZP*b4u81`CfZLwSIdxG83^Uu-NT2SY;n&qbs@CG7>0PP9!Gh6-3^IK?aO_p zvXTmCbPTROD!7^7w^lbOMPwBiN!b`%oml>(Ca-uCO5O-rbuR*qp3_J0Z;ZxN=c!qO z$ZT=nqN?ng*hdpDEE`K^D)m#eIr;ohbJk=pD_fC|Gahb9$~;A`%>THV8zCiopOU;< ze;4Q{>Amg~>NJ!4rSvYwT@7N|QmRe`y89(3Q}Z1I+1kxt&0W4e?O=hW;%{{AK7lxW z{fK_#htucza$c}=9%0{2Sys6<{>U^j?)9;QAt};y>E)%>?Git|B(FvSsZxBIn`K2` zOac0Y;rvw~TBLn2&J_$o@QsYG>M8xvY5YX>4LSqzD{P1njG>LRE&I#tr>^V!lIEc_+;8a-JZ)&8q?1Ax->-FMDS`i{nf!Y& zyOGOmo}aQz4Yn8@#p*fj<2|sYgo8pn-#q$tB?Jm~{;)N_S&4wbWq@97qH+4U%y9Bo zaEN-x>66;qu6O@OgxDXEbj6^Lr%kkTWgfQ-VLFpHSx5XFGl(&SC|ChJ&Gn5z64-1# zpPt7T1F?G}+!T3_Pe5kp^XFBE3nqbACKoaIeo1>fZ>m%nCcMowQ$XC6to-(HREC!d8e z%h4N%r_Elspc&76$F@Bv+)(pzni6Qm%CzaZ)ZX>5MDr+L;|w7eB7Y}O`<(aC68^&| zD%X9^6zoB>+h$XIQk7XlykJi@%fXF@%>|B1%8i4!8A5M5z<+Gou$`<42ayv(O7alOcU`wRhO^=eeg}6~yxC2|k3b z{bMpZ|IBpe8`)^k`u84Z!fXD-G<-^vC?7-DPx)g##w%V@Q7U$#iV~*PzOCM4-&fd4 zjT*hn}w#l{b6^S{i16B>RCjPK|)Z27#P0YVEzmexewZ zmv^03=3*muD1j$TWxf{&TnjXhE^1sfncsctu3MyCY{o~7-i8`_nXOj^`z6T+i6nZ1 zN{YL7+iGx>c!f*oETx{QPrgDlI1JToHI2Z#VS}O^SSMHO}{cH3X}w* z)~AP;QzJXTd#o|eTf(<$bB5tK_~uS%p5K+qrE&=AkJ4FC>4T`>X^~UEjzgG1K^`t= zfZ|(l1af0y8>agj!E1Ar*XEv$8$seU#jxGq{ne9<&YJ z8}d8TU3j|{KI^uvoca>3oak^fuaDzA*9wj0ed}qVj~1Lr>oXxg0<3MTfc_r+o4X5b zS9M!#3PRymYp;{n(H!jl+Q6I|dSix{SN0C?9IZ#&%2+j>6)fxSYB6dXKC8WpU&1x* zbWk6jB&7W?L149HnmN7On~&sh#=rWN_tpoRgmr?wR)klNr?H9Y#{$AZ7qnnxb;+B$ z0m{K$!}AhNYx4x(sI40zP?oVHD)p;QXfJzc$|1jB$2$yTr&RS@3~w?BwfG8V_nmh! zFVUDk+sWESy+gdDq3OMXcn0LXb{Z0>KK#8siclEGJ1ONHDS1+Zw&1m$KVFb9d&88- zFm(H}@($RV8_z%pqiE61XXR=nO0l9Zqn+A=8dZaBW2LOAdketBILVc#CJl~Vj+Ahl#G})QR6}Uk>dLs~JneYVYkLYLn&F|2 zZ>o?g^G^U@_B?-ASnX40FwLeNxv5e1fwOnPJvYS*Qy^qqEU(#iZh?utze7v`7eOcaO?uoUpVkb!|o) zvs0vUOC-`;qrA?UQDQg^w|yD4>=(@hb|VjIm7|+_`l5PL76M}?{k(Q=9z5GHeqC*( z-~h8_LlBj;eS^gt3gm+hG!S28B;bkCDjKQ3RsG$=u=|%%X-R*jhQEl}6dxp+Kg6>1 zyTSeMvtk=rO1I~Wz6jFsw-YuDt-tC*9x-t}b25t``*O3d(uro^8bL^A>HSQZ^051p zTxjP`&quyMwc*#3*3phO3^Eg;$VioE_X_QiHX2%gigyIS>U{toQ3qf@C@pox`)sTl z=|0=h)nLOJbiUOOWD(j8Ln*bO1C zQ;Gwr-N(ryTauK(ax-8*z%{=w>F3WYocM3u77NL(+r-p&+P-Ugq2@fq+}A?rr5#_R z=1yk0>Y`Ndj3++9-=Fs2j(K27$toSZa|_4cGGb=C-{Rbh&P0qcE~rY#T12t9&hS@v zMCv7DZa-id#IfknNl5X_#-*GQv(-U11t@vNtUb!sFA&VuSI4yyB#b(l;&4^@N=E*k ziO~IQ`W$niv>!F2vY$^9;Vz;4^#^XH)La&j&%8xPmNL1|m+SXHy;RafJ(P$FPUOfdGimM#Q>DI^5PWrdQTL3w&Xnw9DSNR~vJoB~5oDU}Zq&_#MFXpZ9Wga3~ zfaV5aRX&6I`;hMo-eV&0VvGH-hDY&NB-9mAzZlcvgNO`glwi@3GnKpg^f2RY zwe(wVg`<-0b^1ecspXcs_EeP?&FyWSZI~W!`@YYqLOehmh*oc`OO$MI>N7uo2sT?M zVx2n@q2ACJv|&yMAJ(PT+0JS(RC zJ5t_<+4y;?gIq>kGcWHxi!Ij^PGH}73o*tY58)voO-{sk^RK^Zk+D1V_o_@U&&uAR zs754j*O`0_ZaubJ+2zerD}?*VZAN$z+YTK=os%1*y+T9rjS{=5(Ae&rg63kB%a zw9SBBw^=g})(TKd6d#65t7TRRhMUVjE_=n^`(hqU6onM|uvlDpus;3l?HT}WK8>Jz z_IZ7-Xwp}mIYdGB%(&?_n}?>GVNW>isQI%0I8DIngJc1{=^h}FVOql!_q*EP^=p~D z90DeEH;}Onya4#E_;}+$cXL(MH}MwTs-(~smHzF0n%Q&oCd>lm0%m6RkGN?cgQgzL96sG>K*pRES)03 zjjY}PwRM$BK>kKGPW$;Uo>Y<8uN&_c!U2^G#RAgOP3tfjc0fu;mHm7t?w)$WYwUzW zkQin)%uA->z7&J@JD1l#)qeiYg?%HeU1V zLQLJnM+-D~WTVjV7=!hZW4iV+DV_lzSYVRrBOhh=IAwGER0+wh(JE4PC$EftBaPvX z4jw_1ran^(XcEOMHl@8EtidZfQIfQzf5=Wm`Cqx%CN};-Lg%-+G(n$M{aAD+$4i0J zXbkcOeY_htQenwGjhXa?$XQI1)y6?C2ul5LHs>lT|uttIGuV|t-qw{c&aAnHK9!a)s)Ge zJ%Wi1XlYy-d3QQ6JyK@s2EGAB$q}Hd_n~+h&v#xT)m$m&mKD6Een9oVyeb${>|ig& zo9r_l^DB>hBM=0-4N$ecrH{G2u6_c!9-HEHdjO!4ovM!7Z=kT`-IqE!-%%quc(@0T zxVlNy0D=^~@{mHR*XdJJ!t5cW+hTbW<;&Rby|kY;`GTM9Fworv9rO+q9GuyIip+_G ze>ptcecN?o@C75L)!}TcOT}^hl;lf6_E~O`Nl4ziH}kmkT+Umj+Mr@h)n_|h&#^L8 z6-*-qIv2&=_MDWRG=#^2ccQB^hB-dSy*M9|xkzzg_Z2GBwm4mg#7Nd%wWw&9iqRvv zSm0txalZsxAAi_T%ttc|GvJA2B)H(YwG25*h-b+3YezZ#U33FTI08~SPj{{m3gFKJl=D4{3$}f!*D9aF!2t}ZzZH`T(JPIaeaTRR+aL`t6 zi(*e}+r9vYSkOK-SCTq4G}V%Rh6T5!j7d42Ld?;uP@^zS(B{nF4EnWAcSRr_fa(Q! z@ph1tc8(4ae#?iG;WvE%)d!zQf7Po=v-R7TQYfqLx3Xx)3`Zj9nmEkoxoC?R7Nt3M zlDL<H;BoK&ppFus_V&5?y7k+-v57YYcI4ik2r&>tJYpeC~!1-HKw31^nW51=#yL zneuyoamw^onsNlBQKNk@!aR7XINtbAFYxzP;_c$6&w~8$JcwjFLnsC@ISw6YHie^w zi-BDiwTYPj8tMHvX)L@bS)+5?+c-V@K!*Ii-!`iuwt(WkIQr?z2~SuqeC6?6*koMD z+RND0$o1a{3Dwigae+69I-BHpQ%=Hr=1C$i2EViXx z_FDJ~T_hipSf1rX9g62e4+>lo54UDPY=zQ!96`RaQNx#|K^8Fq6Jb63W&haMlCtj&`P*{qxg=oLYrx|R>!t3vn=%Q zp6w2Nv|s_~cp7*>x~}od!{{GpXW|e+TCPu2B!aG6(H} z8k%J?;W~3Ke2z?{n(tiKl8U;w(?ppC6gZlz3}~(&bL{-u#62I1;Q6zHP=CmARZ_eM zmXlZ_n=~jAdL2+4QiJ{MSDp`FPw#pj_xH)y>U1flhQLxf{qD{G(=)cQUD*{gvX@cN z;YE~#!!|{fS3k~e#9P-YzM=zHVo)HH0*P}mL;NHscT5TVBcz5R#Yi@rQCK-7I z=W?(wA1`Icj1agBvwBFhMFhQAm4hgV`U1e8 z8hp&Ay!^_VjIi@PBENL*~n0WmP|9W2J5>iIvyQmfNb z=FigjD;=jUnK@Jrod5LR|L2c8n*g8YY#Q=;Xn?3lv+ynx=)*cc9E0BOpLRl+O>9o{ zFC#Kn)`oNTULq&>T@T)W0~Zo=#Wj>=JB1ux0FpFW@ehD?_b2AZKV9r^e|N`GG@{=` ziCG8lF3{)P8_OOarB@Rjl_DbI6PYQ>k_!58UCU4Se@4jv>6`!KM?yb<%1h?gEh+c# z`q+0P!*2Do`yylGUO@A^2B4HwE~TC!&PG5=AH7e*tGKCCdC$D>Ypjq8b@^7aW}1wj z(WROvEr-KWXDqh@>6g4^k1(M9_eZ*f+r7sDc(J@TdQpSebF}evjYoB%O0f;4mA?9A zHRbx#BVaHS?}*qb14>?lbW2@fh>^2=z+$d>_Urlnkj$$4#(||S$hTQ3=aSj~{B&=n zzsk6Qc@NRg!Pk5ctJI;y=oB30-RQi2&!y;f{v2F}j8QTg=5_SrdO89?ZrRL3Z^as6 zz(d2cqHTsb1z0o+78e*ynv6Uf>E}~fdiv#_HZ+YA*tXRNb>)8}CHkl7`fne}BZ1gT zrch5WZiZ*1$SCPp5kTh{P?I_b6v|XC z{I8l%@2&720!5SYOfv+X{AL}g78UaO(vy=I|27hc+V-Q!2SB?*; z(FYpvy?4j89rb6$m*cgK9wF=sQQLLz4TdZtENZ)&N*fYi`1?YneV&ikZO!H>Y|pbd zf7pGiyVt|S?>TV00nh>*AE#HL=HuO(sfhuWZsS)80iaYLTdojJ9x{>ck9=M0Xx?%E z-9QqQE74hTVr|p?0_#I7=2>$lX`R5l749w$ryfjEYN#^*_i+3h8Nt8qW}gK76FWS- zN!4*#?p{f>{*qfx+~Y-9mqwj`VY zBy=r5&5|dY-XKh&AkS4l*;QCS?d8A)p>_ai00m_@&eiZWdh3s`yyhi6?!D~{c?TQK zS5Gv4H&mw{aoCne&sU?+`Q>I0Z{RC%0YJCTqnI?-Ch(}QE_i7o4fWTldmCi3!aEP@ zo-W|sd}H|qYT6UYb1C9vyR*QLoHWj}JwL?_c2TDo5z;gLc+tZZk}2(!Pm$gMlf1ZN zhbP`x{LwedR`R2YID zY2+%7D5npGQSn&eExT>c$CGS8rQXenvVKL4)fR`-3Mt33sc!6stBmu|<64NOc1Yb> zdKXgYwmr=XS50D06}C$dLQeD^08+P5iEVd!)|w!4t06!Db+kFO@w15a+*C-z%a4aq z7?ecvNk1R|Myg%7lULs@&U?*7P!21x>?iSDml`95fHa<<^+@Me4~#`pV|=17lE-qk zDBg!uv>UDAc5o?J1hhGr567};)Hz!wvOqIs_}&o8i;ASdE(LWZDuzzBM3z5A*51&Z zbnUl+(cKfYt~o8!1y2l+;9^K^W|dDISMYdu2nUJ_pbWOhKCkO#0L5V^J{=#Af9MbX z$Jq9p(U(x>G6c%h(Tm~XF`qGhd+QLAPQeyR<+Lv4uxB=1t1GRGitpZw=@+#M2V+ z?ubUQi7S}2f4ZI*Y%6^q5K4Sqt~cJIy?OJmY|8&0n$h;XScL>mNHPG{AT@5p z9Ip#lm;v&PfZzgws^P{MK=+)QCWp=uhICj@d|xc78olh3BtuRDjjF!fn4@hG)Ow*t zehm^h?ZUUmORdcfBP&h2VkemV9%L3>x$G^k@?Dl^wQ{pOAWQ~~1XAY7XH1)Qq697h z&rIj_r3XYVr|EjN;aT%OR|t`SqBqMOXB)tA#<5V^4VCs`wgvXJxnRjnWo~;!QDA(W z$9L~5r{hT<)o2UQlIj|#pt~G#MOq<>Ar3%J6#*@wEr%GzSkcOZ!^*CO!Gg-U<9$>#VjFn4i}1hH>DS>C_WpBOkMqx=|96~W%w!{bplY6z%^@4z|OwP_#pGvA=tR z@1g&tfOHO-j5ABcyDdFhQTGZ?aPD0@H-M23Gz^%am3}D*k|OL04s=a?T_oj(BN9 za`ZEB<0jpz+y#7-!wH*NvmR)AmFFN`gX2o*rQ}bF-$9JGh-=BQlRMfiBYHwV+xx?n z+X7Z!9iq@8@_?M%(bT;f>&^xw1fAir1<3N%6(diM7$nSUZ8eS&nSN&D8{>|@)?Oxy z`Zhg{q@P}H&{h0@$?59wB)R1~j_Bzl@R}&y*Ygr{bw~5_vDkAA|GT+*W}BPr&)+99 zNIQ0_c`3_Ae|UCRL8kbIg34>u11F$`N{rR-;vB>ou-_o?K;dZC4J8_8I6}i`=-7+t zyQh?AoD$I0R)v0 zPP|c7@@Wu2VQQpYo$2JHATxI?^}=GCcOmyXTr^UKh-d@`i=e#wb}y0iaaTKO=Tp5~ z5AZ11p5o!)>py&9SxNXZ_ujo+xkiawXEH-%@9bRd(6-z4tTK>EYj9T zx`lP4&v_{sb9m}Q7U+2u<)3zsWL^g>{5aCtSlKkP^u~)c)~ovGCgSp+$H>s|-k|Gk z{|R9K4*kC4MU&K_kz?$t8Jk+VS?8fvpqx?e4ipVOor6y>TwhCM9%=gqtn8 z(G}L1wSr9stW!=cOjWn%s})=~rzor&G`DSi5$PKBFIYgfwR}&RikJ7@ff;hO6 zF2sF2)4Z;`k9}L})s;cw zGs#aT3D*bq25c$`;X(s!dO>qD*Gl^`=m**d}4|+u$(DLp35~h z?TH}Xh?{bntFuYHQ$Ee$_5)CK3Ftm}7r7=Rz-qnd-Q?T!=dEk+l>#1txrb&oPl>*A z8tRpSSLr~tnyi`PHhMo5L37^KV}+XiJQ(L=`*oQ&A0Hg=tJ6HedYyZoeCQlJe>ynO%t1Noj=M`iH`-2-RcRv`s~ec z?^jx7dMm4dnBYCgP#Wiu)J4*f5+If`P6NGgw0h}#JO4LKYK-0w= zijbGbTmmMKr|iBL%~LhCmkBGG%=tY-Bkz}yj8a_sLJy3ThkD_ZUXg%|-A!VQ{+*LS zk-hOr<9AI&C?{~Dg&`}bab ze2{wP7|0`&G8UyC^8s%eQ5_kRg-az03mkU~s;5`|xOMF0giaMOJj(ud> z9U{@Jbrq$xb+*q3`Rc{wE?W{>k6+)cJX2>wLUG8zZ4qd63UjaYcFRnMR(?*v1;~GY4R>!kXqvQ4pSS=NT6;JaVs7{k)>;w zYCJnVTxX4-6AY;Z$BFg4eHPaD5qrM!IYsCy!fyPR%>}RtCtM2KF_*d;gVfS>Fy=Ua z(y)s8qwL;wgDijOMXc$u&>fLR6XI2<&qxQyiVla*!yESUg!JOhidZ9vQQZgF2A46S zKUDKKSDmq>%N+7DGp!C#3aLWwp05}goEx@Qr zM#3MNe8vG~s)9WVOn6W;9PE}vTsdja=l{5R{%~UbPiIM*AK>mf1hIzF{>@P`4gd0E zKv*Zsp9zGB7}OZ~3fALlQm&Y>!HrL?CCKP}A z)&A|9WW#{~ew#S|zcZ5kZyu{1um&azY`^>G>?i;DXa2VVf7?gY&sqN4xAb2Q!6yo4 zcgAZ$&;H`TJp$gg>4T&vzv-m_7d(|1R|ZUFiSm#Q3il(*N%0 ze{o~_-yQug4&49e9X*==T`VVFtkJ?$rRg55^NYVfU}4Ow$)!y?pde7`d1$5EWTupb zGzU&r4UVff1~NCkY`0t(MKVZnv4zqk0`6e!>!<>?RCde}D&G5KPYEa~Wiw(5GW`+& z;6VWfhd&a7cF0JN{#I)C9~Pzs1W1*h8g#HKME(Hq05(7mGnPf}^V4v0-?&RC0T8@5 z{4NfF8H*C><|-E_2&VOk9EFANqq$0~Ir5QTff6x6LYK(yk;L{7yU_xIMi-y;yiq#I z?cwTzDG#Z$g+KsUX98^30r^#Vy0PNKtOXCa?|COGv$N)xXjCUX!R()m9za@^)&OKV zIZeVUiM;XG^%}hP#tN&<`>9X1As2HAv_kH_gX7v@a>s-Ry4mzw?x3p#q(5>wUzxSL z!F#j2!d3tqDBEX%8D8F-wgAVAW^|v+q(NNPBUTsbBC5vO8aqC~CEd}`(|MaWh?pvL zOlmq|R_x;Ycmt0PgX(OfP3oDnsl1@!5K?;mkcAN3{cS5+>oJ<^n|>xSUp;e87;yUM zn&7AY^N+>6K*Q{gnhbep(*`5Zt0B}&WpjEbmNnjcS@Z-->nT6T_k_XW`_QwjP-q^uJ<37MMGyd8k63lW5ZiUlHiO(#yAL&C(0&kozT?CH`3t}WUFphTzhCYxng2$zaUc59_|y%rCF4PqUAS> zj@+7QC_<=UPgT0#A2&MW$UpX>yl}dA~A%OEvNC zwv+)2D}mRL%L|AN0K%fIRBC2NHG6a91KbHQ7zg+@hr+mnh-nhCpM@%>2wGC|4{##E z%1qa$nmnp=RT7~d(~+g8*aNsCt&%gWEpsrtqjf@n zt3U&u+>*R2Brz0t`sKQ731E2W1{?4LOxJBsC-iN<)D|`e)9~uuLr#>&ou6if*;t_8 z0c-v3;UTsAq+oO7SDshhJXipgf;FV*q6`{xpYATj^5_fO#km^~_BL&fGCKbH!P0e* z0ZrsHF4@{N0C3U`=`NNsOq*~Gx*s)sa-JV_1O-5SLyIWM6)2fkl{R`mUg z5^L9H)KFQX50`9x-K+B5epq8yqt?V^0CK3tj*0sU#11ZPIa!%=2Nf}C6c^x`#vtM1 z_|4@)y2_e-7NzbbdQamLi}V)vuJ_Kf4JKtE+Hl?NqERWmV;k*Ssvq18-!vijs{eKZ zR1F4IQ~O@c3^5v1zC+xhlLTRQ0Wz1NC>+9y$*|9n{`xjE<;~*W&sD=$#OY*XTX4FE zvhq>1H=UcPDyA!+0OJ=7rj9&RuVLp-Y=qe^V$ZM?rtGyhLS1_r1n zN}BA?o`^`WVs1*R2lT(y^l7FpyITX|{Z>9-Pd)ei`Py7SrCsxRU9r^%9PMI4AMx3j zB&+(jCj$_!A0W?D*(Z(N45#E~$~A{MfI`*8v@rEXRC&a3mM0%ch36n{)B3n=sg_2a z@+7qx$O_2^*?E%HS2sCnRJv4$|MC2rOYsZaSskHmZK3aJcSCnBL~`VnWJpdtdCIS0 zd2W3eFW72&AbZwd4pGh&nO7WhIqX)2PKpS*qk^EN8cl9|o*4RKDx~q&^k^y-u^u3C z9Df@9$wT-1OaRg~lBzF}?0`5u3Fy6dMh3PhR#cP`QVY_Hnu99>s6g{*#HI--fkr$peXsUpIp3t1ieo8Oa zWCjJ_6)ud|o$a7IVzivj4ssKJT&^_?lOA9EsHF#b*Q4`->6&gC)O8S86NvVpkynFU zTHhB=r{w6gc;M|=2=*w|z-v&*)?>|Y^L$lm?WH~w=@;%S#zvm~-`lQa1#zjxWh5op zm#GGY4<7#=r0vDCot$H6+rPIgi6qdgfB%yCwxM)nw9ONB=eA70MpdAGoA^`J@WPi) z9A4TsW>M4b^gc=G<<*Q|^3*R%^%|;(dr(uA9r$gm_(P>cM~fZ7Fh=B95gSe5*EpY{M_L-6Pcg3L;D z@6-jU3{<-eA(xi1)pXceADtH>8W^9|=j;2cskXHYXgM-TWV&1Pp!E7ci>nNl0o5n)zqfb373 z#cM5^1v?m+&y!_GMqT5(!P%QNURJ7`boPwIddRqb3kW7JP|$i%bvs61o-`x9gBD;S86T{(Y|>UfW75 z%JBwN$E}~mhqVJwjMkc_c|i3=E5Lr{V)y~VDt%$lLNBI~kKGF+HZ<1E`RupfS`*Y= zO4OohArkx(T}bQtGjx+$oG)K(eHKO z3JC6+I#nR5Q@#C&`X?x>S@Zs-as4M>{s8=wQst%~=>AGSqPv0!0TfbP;l}yva<`gK z4;;ueorihh=4IeTUN@9w^_b6QGM zMfm#M1yqT|R1e%^SsCGT?O^Jr+KGzCfy=}_L7UrZsvsLd6L-k9GLSQQOO^`c%8TbQ!fpbHWoULo(a# zP4lO?C;MP!DD~@%?>IWdQ+i3qeH-J=yk(7z_z9)D$A^Hw2T4iw^4frX&`8lbHbFw0*F!l`^;IU2 z!SyY%0cJW_>G_(7an0>yu2Wh0`Xfh>ZIutkwep3RT`6PcGwy6mYC7aZkLp}WPimgA zwF2$^zIp>ScxZA1vj^HceW9XXx_2?)G}J(jW!hiuBFgaUd1G=&19o<3nZHea1|kq7 zWt3cFtgj$=y4mvGDK0rc7a$R`MdY0cjz5mHH6 z%bGn&iBQVEC+lGB+ZYp)B1?pjExR$sKE^IdcG<={$TApa?86K*_v!b&pYMGi&+j>| z-}Qa|<2ab*bDrnt^M1eHuSGHEMbLvx*g;;;NLjuCZM{Uz7P-0w1gsz2`nMXm{R zAf|z!Kr1dAs$A;v=tf-g5=6N37Mb|0O6Fr3GK#dFKn}mBZhrwin{H?cvPV!W~x)9UF zTj%O--639f8`0XO+Go2FspWV_Vsak$te#`N7Cq(@A8T2;u=D;@YB&z)Ko;`nDY4;y zP|g~Z%wXy`!zh9V^wQVdI59#u{7KcWK^j-sGRL4+y7V)n*QWJ+n>C6{z-ocS7mRCT zN3MGxNJ(PwVgu}T8J;%2g^h?l1d;&{5TaLQDY~kzSE~ZL*91~D*-pH5Yg77XRpq~- zov(Z+gfoN&q3d-%gVYoU^Tb(u)YLVdY6VaIs^B9^_rmQL=9*PT{TcB$by82VfJX8V zss6w@y+~NmehN#Z%f}!=7yczW_}z+=_wo-g9q3SVB*o z>VNUWb~!ycya!0B6Xu+-LVMnu}8172@z&BA8jvwYNYI;cYz#7mPrW}X5u~3dqgny?pVb>mWR52IK)yFe~?Co z_7d13V1J^(B<2Tyu)77ak-TaQv?79cEKYFs6Pf}4Q> zBr4H(;6Q+(UPwY44JC<`QL_p^qUB{pH*)|n3je--mD}l_a_UmJ8$d2T4ZIPF?3kJ+!YCC-mO-2C zgKJA^i`k5PKnHf!e7m?W!px!jko3Ljh!R$}J|Il!_gT5dOg)rxME8!>+I4^av~Ic* z1n^ACM<3U>V0lJGvv}8!7L?zfUsw5q$MJ>*JSzUeH*={PG!lOz>=gPwmBcMg;(K4t zs3!|LH*Fe5-ks-pxBJibs((*m-BTF=j^75|+C;UsDasX3mtdGzzg^iMX>dbj+&*O6 zPev0YFwQq!_r!^Xa5FzZ?RTuo!NWBXeSChaFGx<#RTyDb9XTvEo@5qR?Y2I#)ox_f zDUG=A#?c}+?ArPg1SDZ5#0zRYtfm37{kmYhjQ15Qvu2MR{40zo0IRx9?12b|uE)%G z4^|tDs`>JiY*%53EyzwQImy>`qIPAU%5Kz7+^~iEg!^Wl{w5Pax zhBp7JJKfFI7%53GYj9 za}}pLW9JkO*Xy|&K%k>Fm4K&k#H73_TpwLw+)juW)Yi(FLermEPqqwbXXt0A9JMn4 zoD5?PJ9YN`D&;8?3suOj+y76a9ovg9M!>mtL8&47c}}2DN^a(Yb*B9+L%-0|uggCx zy0#OmwUk-@7A^VzKuBDGQTdr#6~oz)&tS&JOj49M~BhDGZ)dpMlL7*zS#* z-P@gG zx|T$!&wJJyTFxp|3UW$6B895DH#%fa*egG*!c{y$=UYm|M zl5klndoqwHHtCkvBzMSq<}$p=P#QIRusQ;tPKM5;Nct4>e6vhfc$K*I56|)c^K95N z|F{9}Uz=B2f9xPI$l9-T{TS(+BV-_(xE)%?S^X(7yyCA{c!Pj^+lI~-c(}xISq8U1 zEQx_tzeCPD0|DbK+(Trth8JX)$Z!7A9o!fXL3;}z%Qw$_ zykZOJiX{zrOPK1y2gL4K)*B)UBT7mWaUCCdMLJ@5y-w`$IFFvKM;N$R0-`Z9qoYWX zQ1Z(n{q|^^Y&+&G+@NsJoM~31-T+F88{uSZ??5`qItJmYoF@d zzL=~%T4FK{Y5Q4DI?7OnOpzJ^aPFX9GWy?_=HEy9-yS>;g@4)E=hWWeEDkEw=1KwL z*DnriIFO7i&%-NR1XczlV5gOuvK{(T1tlCuyqSGeHpgr|kE$_0x%+uk%jZV+W9Gb( zsAAk`aHjl=VDgHPoNLIE_sSr8#gS?&;K^KVil0Uyja!teQmjyX%0G%n&EjOG*H*(v z5UH(_H{y%~SLxPNjJcDPb7 zC(A{-NC*E8MgTn3Q_#r!+3Ywl=@$^?``rzUYc>lT5gHQJBq(3q1X#-zCj9SDV?-Bb zs!XKNPnv(m@NkuCe@nI-b>8!(jyCbiXa#59y}o8lgzaY^bDl{@Jzm>>RU~r@gp1p{ z9Lw*?}XvsfSQ;5Ba}Lt5%}jr7Sm`=2@px?x%feN8NmI$PM9*cO@^KHbum05 z|L)*VK5N=JE^HkEI%ZtTRRuzUNq~9QVfJIbPXVU86ZJoRiO1Ng@%ym}OXX2uRPdH( z^`oVs0o>Xw?T*K{AnIar$o(p)aLZR!Y8=qHUV!x9FsfXgkJ!y-1A%u4>!s1k&#c#h z0bJ52E0`FRvg@@CoQF}DT>3s;j|kfM{df@8@}@<;WC--yW=BP+1_helnv0k=#=^t_ z97oyz)wTp){TAopUg;*8h3ChEvAw_Eh9mT9dc7Q^{Wg{{epn$I0>h9f%Fla_;;Vj>%T45XX zg}xVX3?t1j(XNwEQ>2w%^@cy3$EoA6n?)ooM~*O3{h z1!yPdzVA|*#x6_Qe|DSb|EV--YKzbZx6a9Mi=!BCu5~|{nnZ#9PAoHv*WRC~b`wK_ zZcWLg>vbgwsGnLX#1z%PJ6xX&b;yA9g*Qmqb=^?(T|2!d3`mP#VWjozfgN#*o~cch zX1>S-996;JXFMWuFo&5{x(77{ z9lSn#d*&722iKKdwFt{V3Dj@|A5b}5lkmeY{jBtw4tQlHTN4g^KG!r+Hp#I9ww=}S z7kt0w9an7`rjv@DdpBNQS;HUxvFvXjm`#*gkdH?Hy z)YXXMOtg}I|8-!rNr$H#=(Xv(&bXRXynZrVJ{GJ6l3CEiou4VcxpR+wa3-{eQwcS? zaYOhjU|VZoiN{*JYoay5f4JCqNO)%eTxawmQ1!-4x<&ar+Eez-z@2e^Gl}K%9rdn= z3A7A+Vd`Mq!NOhk&tF1@#=G9|Ah$#$BCyUN2S7$S?Mo4A6bP|S#4t;~$)o@CuDp!3 zk|zAEp8@`|DF?O~YIbYFlrV*!R<^2jw>dB2pK+Ga5~a)AS%SgS%VYriJLFCaz|1b3 zDOKZPdFwD77nti?ZW$E}C2O*=80e)+3f}Uz%nk}ThNs6swWEi9iqq((qHGBB)Fjm2 zic_+={P}gKqLI!S!<<}4U*g`2Zog{%CI%{ksae(AjboYENFh8_E<3-zIUdFU-Mw@Q z53ya9Xub!Q@OD!Rg0Rix-Q44ggpVMN`T@AOpXNOJaew86M|7*gKSPE7T@Bt-rsFy+ z5R><9w4Z+@=c2~)EyG2-t6X=_c#OJsuI%d0sYXUNe)(&LAI3xHKMCKBP`u7OVyZy| zv}js@SI-UYrJlLE_WGU&oxWEfq;+=YbGVV!aYKzGb0pz;IID8_m;U9>1;!UlIINGndxyRFGMQ(ZmE8xakMSiXR|Jr^| zp7{!=?LFI6T9%aZDm`r+Zcov2(N4G-4L_GTw%)j2Vc#pf=Tj;RpAD{SYkOkjf_q-` zH07a1{@QrO2WWa^5@2p{8`x4l-{Vk>Dr19R<2MpKIGu(DBIBRBBg3V$V0FZ_+L#2N z|Fhfuqo2;1&3+$?G7kN(y5;^Lld;{QRq)XWOW;tn+|BjbkN*kj{kYQU6bBR#uZ8Rx zfp>2?xLpRghg?<+Zn&c0|Lmqt$~D>%DU)&|^breBM&{M8GXL3j;7|^5UMlzhcq|{k zc#7x-K*^B|`tiO+AYUiVFtA}O(%k#EMu^?hzb(1``?9o*3*6s0T&vZ+B1m`oM#vk6 z3}m5UZ}ewn$li@hp9KC~sNMe)l?O~y0h)mDZ9wBZ<%>teE1T8S%G&V$XJ6s;*KA-% z>mOilvk`c6e~=j%r zTxU`D*neM{e^AY~$X^8(?@I0auXO+6QvB;G#eD@9B;iGJ`9FBX{yW`0#OppFl9P8o zFVFBFuG+s==c@cMxKJQ&`yXON|9fF@Z-6_}YxbHa)BmrvyYI<#PDxMK;^(7(@2daw zq8-P;1m^bjzm~LRT;ljqY<~CkKN1wRn21{2Y`5Ar4GkP@wxIol zv4^vF+4D5oCde9pO;^tSmwsk;nE9NN>h99o`G2rD|JR=g-SO`^$&#*j*3mf83pt1u z+Zhc&zU`-f`LBKE>nq?Y6kPf`E#R)gbO`jZo~;aAUxx}ZzIZ+Dk;O@06N>nY?)e!& z&SYG-yDVn=6DpK!RCJ@?+ffkcr#?EHDWuoaO>yA})r&oUe;xZ=Pr#O^zAuS!kNzGA zD9=i$$wFiLzvF6Ds`Lg<75Ljv5|k?pnzguo`9+d~2y3iuUN}?Fo2S3xkN2Eik)8t$ zu{&O)qT_&2^L2NUPO)Lpg)tN> z`h3@MytC&<_#EMgVz+vz5zD0cnqCrzF8dXk+~Sc}d%)r$zc`&(JVjo67!uPBF_XyYd*M8R;W) z1{=4Mu_4mrI?L*a_Mk204;=Y}Zfx|Y-j}Pubj^BLw{zN~+_KV5Ik0y6W!+lELxz6X z{z^ehHO*yz10ag#ZXyQJ(^9~Vkl@rz6+Y&(btMW5l2#;D=uQ{jU+PK8eF`K;1gs)Y z#3CQX=sw#BT9C`Jqr{|#tn-W_6JBR~@pxxaC71bW8*^S9Admb14_0qRg)B)lbFZWb{^IF5u2`70$!2mp#CMwHG$wA zjlbVwXU)%RYA#PHIxuJUTh{&gf!VrI6xd`#VGY;}>u<35H6$koe9Xh{&LIH*J@uu| z`qP+yK*ddq+40a#>O5$uF zmxgi}nOfuiVdr3L(m!@NlvzAZJBl-D4<(a$;h_KPtFleLd%S&tAM1xbwT-as_B!%YtRZ|i-{`FL?pNW7GZZd+7f`;rgm#07v+Xa zE|0{l^B+xAJENNnPc*rtNUBXnb4xwqmUdffiWkYcRHbfz?q2`NR9UA>-an=6dq3%< z)tnTIlCJcf4opEd-=df$@Zc#92~cuIJu$AtM6&+qSQ*O)3RqGaV4vyq!~ZrTG|_&y zV1W7d{N&r+=S*LFNgAIEH0Az>L%8p9$d9*u>UzO@gNxZwy4{C~cG?VkX19!3JE4Dw z<00Wse~7%XX?R#r(v|}~)kT}|^{{e1)!^5)m-dC%WOGP03HiC+a{YSS&BHHW1p3BZ zv6qzv+>;Os$4*g(b#jK^ywAwj-^-mPL!E+`5Dk*R& z;LWlXZRj(Dve$WE`7xFGxp_CKtdZfr~EyFQWRTik_# z!GV2O{gG^Pg<^Kh9C6X)&N-iA!^p(2GX$u8zul8e-L7Ty4~!(JIoAajZ=+xLHKu%z zPw7^@45Pj4OtyN#{JB@_mGMouq0C#}khQV9<8_eOlIaSWXGJqfEQh{dgXtbTFM5{y ztYp2vg8}H?Sj{`OHxCmiU(aL+q3+CIfAa=@I#^uFScCNTZJ%<~0uz7W{C4DDbaM=A z?3Y%Di*L`w@bNEO4?Sl%{{MbFrcq6Nlo6!sbF6(f_cGXh(P*~ErIwX}Wh71cn7yqo zaJx%tFRHZH`ycA38;@pRtK+Cp(_cGJ9l0$1v&{{fdQ#kP2|>?tW)Zjd)rFLGA28la zKFc0R*on2u+7*-2TA@1#Rp1+V6-ECjI$=&#$EQ2i>36%4>Vdureo(xSU+VOY2%vmh zcxKmAMt5vQepu~Y1h{~y`L0h00@f&_t66q{>!FejW%9cL+^Bo0H%$;q-YG?jQGiq4 z4#3JQyvHRw(?ya$-?B8X&@fT<1~Bl7*?v{K?j*5zIk4qzgW}JMtUM&|nqyMvcEw*_ z00-$iU!-!1^b5*S0n$lE+N8-GeZM(i8`rG{o#+xp(4k}$AA3yzHc$)t1(A#0+V9Q^ zS#kKaM{-oS9R>PX0QMY0K(>6+;OeEK6tsm1C{es#^j22NFvc$dD*`t}c#_66xVw7N z6;YD~44dCveKW<)n#`$Wyu_14h3bpqFK_j2S5hl+JsxdbBax)v?>Za3^bnNVs#hAQ zMhhh164UM6NNn7re7yXJ7uL?iv2Xzz=sN#1`gqKPX~(MXQQyOl0PC@O@S3vGw}&5U z>nd17wjX6VCA?E>-3C|@$v!Y-^5MbOBk6s}(Ww+gu-X9mW;D;qUW&sWVN$4j_jbVqjm^E;Xe$4OQ!9 z+-#E*9QsJuudh9dE3wLHaKFAim36l(Q7Ynfe{jR*rqqtr{W=U=me0o6UZfD`0#izg>f$*Js|SdTQPC=NC4Il zanerWA4cR#8R~b7HuJg-e+3>{W`VCBp@Ex9!t4>vRhJiIU!wl^XV=xMK!v*b%7K#` zs75?cK{eFI9@9m37{yD<;Wx~1dB$CiIJOvb&o1(!XTZuAmq-J9%NjT6tV^k=(S#2F z@8;jSR)@(hV>RlM1$rXwyI`!w4iPLgWKXT`Ia+ztw_hd)TAX@HfgrLHwj&+O?oWNc+c-Zcs%!@;wV+@SAtB` z#nGwzAz#NE+dh(HvT^5~T4p%Keg}=fDrnYE6StMgeydgh*eTojRRCJDfsz|7g;MN7 z>?PNN_x0#!(G{;b>uDvS!JgI9scDXrB_Biwn~VCb$|u(CQ>bE)Og>oMX;>S!+yQxr z4n!#ST_pP*POZAmMCm(O3Yb@tRz#sz0f<_A;R=VIt4X%z)lQY7Kq3oOIq)UZ>Tcar z;v(}dart-E5C4Oi>rSFzqmr`8#OdT_QR4+H8_N8JI&o$9)%+xh6SiGy%XGwN&`VP8 znu%ChX1GA*6(X0fe0alNW=;6wF~5)#tLiJ@sM^gy?)~}w7l1I&K3Jgo$SQB)k&6IN z!zX(?1fYMI(VzOYeAjj%j%X_?`WfQ4y=3aNBhF>W#M`#ShJP&7P&9cg41u-m3l(Yu zl7NL3q0FCQPehsbRlLY(QSb<#qUt&%521!n_s|YxAzW~ugBJIGkY_o7+3k1SGmq$#h`!6I{+Jl z{W}mUlX-!vi|HLdT;F{2Vd{_N@|jNCSfz94?OZggw*eE_`ZCYuQ_Q=`pVZW`Kw41~ z28LvfC`}Shlt=`qhA-S)UP}Cta4DDF68o~>)<2jJ#kItnft{{3TEPVG5Hr6>A8@UV z!!vs@{1r!H_3s)ed#mD@V)ZJqa$I##yH*kA)*)#&rk{^SYK@r-mRTVWR+h6m0In;Y zS*>k9EZ4cDVU?@9ax$*B0P)sdx@j@%J;vAjig*Bx9aKi!cfakJuL z?RtJokf+|-AY=5Q_@2~n;!u+x+pUKaen=H?cNv>1OGypmE$r%0TW>@LuBcAJ1(#6p zG~qGmkY7E2oIR!$W&14ru)<>eO?oid&hwZ%8UZUe}?yF>k$d`ZoP=APB%I2?c)MwBroE2W# z?=XasX3hDaWC6+Ths1~mM<(!WlM#)A};z( z=b`V>j;41*mON89qd>I2J0=Z#V6Z8(=B14*)Qy3R_#Sua^e_> zfNq2|6l71i_dZzFb8_?tTz^6a(BYW~eU!oMAvp~!{TU-Nm0#eJ5)K^Riqtl1ViVMs zArtX^H%p)crim_o6#ahA-fQ*nJF~dA+nguQP_en0RD_=N?9`L6p1tAc4qmwKh6v4Ro693|UjfX*8POv(>IXyKTu3YyYu~x006-k(S#r|e zZi;}Uv3?3l%aL6DK^uEd{7bhZV4UR5Y%14W5dJiS44UO)pp_FQo2^t1{of9Q{NabP z%KNRnu;zgkDt*Ms@IG}+W@G@o)}p%0F2rK|;GN3UIpZSsJut}_`lWToZ6tO`%Tc?F za?L%O0KMU|JQd@bqp~Q?^H|!TH zP0Q}7Ghx4k%ty{+%RSyTkh0DF)NNDbXnNowVG}k4Pa604;=R=e|o`2=YckF%{2J>KbR5UkO)=)Du=LE%< z+mwYReNx`!&Cc&8;`^iCI|8nsIe%?M^obPVx=C`eb!5*6dmIy}LQc)f;(1F-03q}} zx`ZFsvt5VAyc@CQ+CRHGn`TnX(MKeXOBgOs)wy8)qS-0KEB^SsvfN@t!1*jzxO9f;&s>_v#X&-;M}+QL`!5dLE&=o z2{(9y>qw4k32mgLNDb-nfmJ7hAaQq_^Ud1fT4m`g3hU8w9156ST0-VoxUQfEno=~J zXFhTZ4#8;C2C|ujIB7t(xtNA*x&Y_2%JZ~CjU$H&x^d1Nu4eUa={qZ!exGCd*74-| zC+Pi5X|_c-rn)W=AVr!R95e}$H56+w66F{SowbCcj1Sxq{El~`+KV&-31jR|4vbz` zDW~+Mro1zhB<+Eb_AVh_ezDFz0X(+Dzdj8DlV!OY1`;rK6sO2oCz$p$yygZ#*I|wyOHF z3=3m9Q*~oxk4PqDCwApmM-M=3xLnoE=enG)3Bw0=CP=;dpCh_^4QpzJI{?WC`7BJD z{1S;0uMR%@xp^03)kNOmxr9%CdzM^P164|E9&ZEn${ZeI2JNWrHU$p7WPs1ed(L;d z$Z~9LX?C{OU^;f#zAyEpDnn!zy2g&0=k-alYcOI^eDS_qIcu z*GCxp*03gg`W!9fo@ZW)K)N>TNm;a}>Z_6xmAjU8FF$1IO_W16K0P*je8_j<(wj4X z3x|Yk>$eD}npL*=J5~zgu%38ac(*!*mn{!6|HI<4viGEvT@&v8CjxQIFz+AMIo*@EL0MY983J+ zdPJ(U4aEiQzUV@DdQ10)?n?x3q^yneZ~JwxU6%S4EmPb{zv~uPEErWk>xJxZ{@_(R zof;5WF?018Qo$!bqCJ9v(^JKBX~R0vKOXE@^2qPXl_P6SUm+LYXBI>n6p~U6Ou)Xb zEJjX^52;^VJ~JdvvISa|=r=SA?gseVsv0Ac6yLcxuw_UjB{9R^&c?;e?X zC)@CMpG7PaW$JIDEd3Ivo^Nw^rpPCtjF-f%1pCiYueP$iJbCK9!>hT~YBgm~yAR{2 zquwbMkCu9&b?|=lPDq5f-^|VaDUT>)4nL$A(Em|pw6UHv{2OpbJ>G^`sYvrf1Y59s z&ls~Vj_5Ui7E|;)kkj@=MhEJh9qw8hWm*;bI!Q{X2xqE|&{;Ktua(*!d}j!YnVUPc z_u+!i0<})=q?p4^54zUQs2|a-hNmB1{FTG;HBHv(8r%q(M6jsWnqcRsa1Hb3e(F3` z_({~b_8B}zH%=pp>+ux_K(#Vx-d4M}2wv<9;#_ zh|!#6W+Ex?Nx~z`wLuM5$E;;Jme*#?EVt3&{7AP6O@>H|YY zD)!P9>B=U+67&-zvDpgyJ8Sb2OEf?wvg%3{!-uw2+V}gmRR@!uIstlZZuWG*7@$zC zX4)yYZkBJ%O&lOXXHOiDPR@=nZ30C)_v`axS6_es4)wzZ*B_E6!Gq;c%6&Uuy!{+} z)yQ0cZKZZv=8cEh4hjpOnd8}hs~ZVH24X7 zBvLImG>^j90rrPtT`r(MN7|9}tTBaNv3g+AKLU5YlvZTISY) zLN%x!a-0nDip^gspX0asB(Sr*9|z(LSF@pC9?u4G7V019-_B~TDzQMRv{-Yrm?>{+ zT4U47w1%lLe$kf@r}izZGid_xI5X>DLe5-5!kaKlS6lNvvi~&&jKQHP#%uw$h>`fS}3^aqxWAz5KGPh zi-y_#=JH-bt!km<0V9!R0QUG*@EKB!O{Z12d7=v#YYDQ1MJpZ7PEXK0$D6^;!9>av za)bPxXBv}yj{=K*gf1&OKP<7X1A>y8uh169e%3ddyf`|@dPAiIJo8-YArOWbGkU0+uC-|b8kd7NdIbIQkJ*Ntey0I6-|JNXMw3CH4tXyXN8`q<4~#BKpT(Wly7Cje$eA5a8*^)%JK=n zK((Lg`_d+CZ4; zNKtCEnPd$>9VO^zYi76D1gorYt;h$xwhpc^RB2u(!|As_-4u1F^u)2zhgdS-Awep_ zbL(&TNq)L0>eliZfuC(IdcAd2*1n+n!y)j0VAr1){Q#G z{(+w)u1?KH3WqeqPx+jcl6nF!l-OP5)>`Pkeq>htSyU*pZ~SKVJDYc(O8KGUa5+~G zxTNztVva6CP_O38m5Fs3WD;Zb>h2$QjqhtG-rwYQi{0^C`C+cjLt(MH-;`@Fiyigq z{49=86lXpaOLfg}$W?=Y3~wKTSXxxEH*T^g4ddo~b7=vZVV$ETThtnM`N*kN+m4G9 z?`YB*!5hk&?nilxdARE&hEk`29Xi(k#|t3MUapIo2+KG4>44TLSf#%})Z2f~S7IgN zU6_!Xc@I>2I1A6sG}yix?zMRKqrAB~)mJqTfyrU}CBxAbSV@>;gAm7$Y}e~kyr%AQ zcJfL+%Bs{gUY;RLQ!=7hCR2+m~5%DC#&% zRY*Jz&Cj?Fxw=T$4U+_1#${y)CQmlI82av_hjmt=B;JS09H)>(#$#`U=q1(Xf znuiP}LS+zbT)ktcr?Mi8$~q!SIsBA)X`Rk+z+X#L2{4My@h_hR4_gW`_nK9ntFq|$ zQT4-o#+$4j$FRIkWnxK(bi@IQ81zEC_S70MK5A$$w?u3;s(KTM1BWq+xyv7{NjF%~ znP+;7YfP7CCfHP|mumZuI_yKd0VUpdz%t~(=q^B55VoG7rM-PAhH?kh%K8TN4As;$ zJ;@=fO^be}MxHVgn{u5S=5e+Z0XH6)UOnzHpp=~#Vqw9z(~!;PV_`3zjN9WS00Jk1 zm3ymE4IF8JsTn|5KVV>9Zj(i7L!qzaN`344@QpV73~1v4S~|6HVU>HtqEIPE`F4;t z`$~bD4NU^g>5`$ai>cM^L_q$OM=Z8w(@z~F-?q@_*Q$F7WXJFjeLpFjrOdKB1?xrm zHwULm3p$>Lt5~6Ydz#mmly5#A@nVjSO?N+p)XCjZ6%T zP^Qvkc{K}P__sXG2Jkd_Hn%?zHp}Q8O2anVi`j2G@$fgn&TBrRN)oN zOiiR}UkZO!@jR~_BhU-w8BD~$L+cpc6*y;2fMR7!xh^W87P&xoN9-{p8+ElCyF2aL z^xN6&AJ9qj`Tb17-!OA~X>{3M;2`)f*Tjt|T<>+A(z*v)OR7|qrCH_`+}ygH0Mb_a z3Q%uR(^KV`$!2H^TSYnA(eljCN!bY_=9x*-NlKu@&Pl{mDtc<$nLl7}nk`IeJ%oiO z^ov3=(e^q|9Y6z)wB~Ld`0DZ@+rD12rN#+$USC>3wwIaPP`_-m#^65F$ONQi=CI?_ zOG{pGvpmoqHtXdl7|niP;LZ4b{=I0K5nS^aF}?XziEl|US1-l*s)BiUR<;vdg|~EE zuE|_JR@Kh76%0#5&F;(Hh4)zGdppsJvsRcEyQ3H?^=ih0MGJXx$%Y}Z!=g$U^a~-b z0n<9~BEw!62_P9FsfXLJs*A8${hj+yf3lpXd+DoirVxRpel}|Mjk0Se*a2IFo}=-N zxe!FjD&&FLD@51o?4_D&=yItd*k^7SK&KUNWv#k`5=*vMhB_=OS9_oE>Zjk?d<@`P z&WSn5`n}>TJ6YR~DRR+cTQQQvJzkuh+F`knuqLGARR~quDMt0-f>Vmr8H72wqrYNk zGku%~OAGkvGy*qB`qLh%kn=l`4F}EEJ9AksxT4-_{pIEWz3a3G@y)3wGYGXKIiur{ zfh*7llDy68C(TR0;~-4)ooRFpQs7=D`GDK0_er`Ptg-J^Tz@t@T|Md);=8Mt<|J~Y z8fcWSd&M<)v|&dz`{$uTpF^T}x`a1&JH_$QPO>jcJ7iKqO`*&`+2+#!MRWS^eb?+3 z$3T{UbcbTf>LUHCpm<#$J0ZtX9`|<%kY)4!JkJlz3|c;8PTUH}7fm0fgdx4`QR>yx zjBHOjg4Gwx0`zNYs+Pb}{J#EU1%qCNyYse|+AkO;oVa5w_ely4!%YylzENYHf9?RP z|9kLt)A4Di@ZN+5`pc__qxlK}WqY{PT@{;sEy2jHyW6kbX_)H8E=_V(vcd^}5d;OS zIN zS5Ie2$>dIFZG|_a#DFJJR#jR-E))G3vj+uBPAfU~{RswKnH%dadg_a8(yo6#PV{$h z>*WPC5)YB67zdej*#^C<#`WuiXf#j#J@@oZVSoCizQd0zgHIBilRbe>#i#gZM}!em zf?tuDfIsK3yf}7V@2V%$b?RPw64E;4*wSp+X=s5c1A59as#9T$>&p_@zW?;HjvfA4 z@0p8EdC$O#{u|%9@ZZCkDJK&fy6?3G=-91Ye;lijQywT`SXFqIvXEemz?Zx%d?pak z0EL*zKv&w`{gZ(q&M9L0?#}d?Sx7-Zgv_P<6A)#+j-QIE!)5P!9KJAZbwB+AJ09e^ zQheou_T38&z9~xp1~TIzhL0_Pamkqqu{f5cF~DzIn4%+eGM#UlGco-gIXzzu70LxK zR2;f6Zy}Naia6WI$__LytAKk~#9b)a`}ci#&63>og02Gq+yyacSy(&}{wyIPVp!_y z03?sr&(hCZeuyhbw&UPgZh+`&rKSQip}#JRfUqPq%I}- z(XQ6Z?ckfkfJD*pzbawj-n7NTj`{6|QSjIeK(6NHSJa9H>1f_Fh8y8b4_@tucK~U5 zQAzv07dEVr`r%tD%0Wjsp&fta7wyVmtkRMSdPB3aIY}G^)cU=QuHakaw!3k^UwwCH7QR^yfcbXMf2+cA!LJNLy9D*}j-e!sP;#Xc45 z;L)M3`H4d!Z#uTDHIMIrpvA>lr3ES5E5Lc!0^-=}k?CZlF^Bu+$0aTg@T5y$?B8>! z-R+nhmxGqam|-(s#mVrute`jMBAje17x#%atv}vavPED@u*HAic< zThvSz&%~mbHc|RS+4Q-v{95=B6Rvap?4!WgU+GeV(hlD<)SGOm5NzN~3da_7=F_Z6 z-I%#HxFGb~5^>?KEATwsRelEz=Di-7(rm*BqHgaTGB0$UciA-7!3>b*f$-;a7Cd?3 zI*Dws1OqShe?t~dG+l2ZMfx3_Zf-t;;0|We74kU~MGdxD7tWwll}p z$u?h+V~vg@D_xiKje!wC?MTD=3`z(8=Br$TOke}Je8>Z@oCkux6~jJI zOT~4&Ki^K!%Nk4JTec9(aRvfjDq6idZWb^4P?xQOKLILajp7|dVoTJQXRs7Wy9Ltb zd`@LinM5Uk-~}8-%RPh#Q&0{07lNyc4eWrRMbE>158AlOd`eE)HddYXuv{#YrDOiR zZ>^*6e6-{ig=k0qczF;IhzRz!KW5YUBHc+q1IOeZcl*3enbXPm#ukJeYU50XUaFEs z#`=CO#4@3POdSl2GEsRsYdG5{dKMbCN@zu7n&0@=1h=d;j+s$)kr`v`bmB1i{whP?d(pLCb_>|!6~`;7lw zruIU%LZo}7@Z$Pn~sFHdwXAD~%c5=d!VZX5- zX&w(D8k`UR7N6G~rK4tV?_~(l1Li9o% zL^TzUhQB=9aQ|$+eQ%sv@n=!aFd1>tVuq2_k&>I&3;HV9cl<4zNV-p)Dhv|1EQB-i zZN~9RiP#$D{2FD$x1a78QYjqd4U=9I=Idr<}FVBCR`q?QEO@-03p~&6utZ;HUlUzW-ou z+?URpDsg%N@MphX6n%Cd*hyR=-9A^MI0 zmUYo*D1T0v^$I@iozOE}}4N0ZB2ji2( z+RsSgqzwMNNb49WMDi%HZOiPa@)2td#er>sVvju7M9d#^k7C?`>0 z8ElymYTFdvMP?LP1beBprjq!&;$=W(?H7XT45cu$2deoA?O)Med4S_#zB7UE0JIX$ zZ(8J=F`N9B_ojM?nWnT#UW8<$nfQDZ78gcGooE$uo-bUav`gF?$;V%~|NXsdd;*;3MH61+mi8;o0R7bn?dIu)o8;0DKT{BMK7nv=3L4@h5f7D{$huuz zw=dpa=Gjg9j(XS)LVASZxW`7BNHX2#VnaHor=0+Hb^KX@NLTr8d17ir*H?i?9!V_! zH||~PxslGwH0L$!p4Oynyts}tG51XyL(cOxnc2t_H}mfVM3r2Y>iQHF{u)Kx7Axkm zPSi|@{~qwVe+^W`u^v{#jth zd~x{uzAx5BCkg1%kk2xip9)RuqBQjA#c)=QQ}rKVc{?skrXcOJ5=4)?S5yId_&kc|q$M)1Kw{7B1DiC0fN;eSroWAdgc)=WV{PhtC7)#P$|bh@ zrMmmatc~Z@7cETxep$|^Y7;sPH#6dvaXhzmF8x_RSvbI^vs0f?m?=L~Qz8ZZv(V{# z2O!mLt;7lJ)qV}=qujPlyQCwM0$Mho{ENL0b3Z66AEQ#7ztlR4@sV@Gy2xd#4y;r^ z5k>&kPWnJca{%e(PEml?v1I50n&+wnHHGb4iFj@~UEzo>kE?O78Mab8Fv!n^oIwG3oVdjk0}*q06p*sd0fW zruT@riP4Y6Vx#)SLmW!@Hm(L+FU2O6j!T*wD&Uhuo_fEH?Ful@#OJfKp z%RMo@_Z?9T<{=;co=(wec^{q*8gfSiiUE0fM4*cI+DRTw-v_JIIWGqVj(*^6{GQ!>j(6S$dQS-6aIEcb*SiP(m=3O=2yVp#$=PS}f?-Ojtl^oxl- z&d*dNLFc_ww;dCbuin@W($+WaiCti?im!v2q2|8_9~I#JEt^BV1I-tsem2#bWwa=Y5;Hn!g1Qc2#Dd-=noOz0f$}5?S(70G;zSqO?_OFH-1?SHD6gIjs^qCt zb$!w*vXyd2u=Mzr7XC$6V~};5Sa`QJ*)wbLl@#L9?rO~i z)G1sQw|Bx-f&azcdq*|3uG_;3HdI7KEFeu3qzFof&=e8rNR=+V6QzZY2!e_NBE2KM zgx*_FRHTL?1OlN*2|Yl75CY_TIp^$q?m73{dmr~dcl^e<|2X1UVXdsS-ts*2na`Y~ zw(KQ~N8-%OQ=h^G{6A4#!FTYvyM#-vztYMx)YGrDVb!N=B9L%OH)sn%d$fEH#@53N zc_gpn{FEUd5l|yWPT++aDki_nc9$9%m09%k zYzI>_OaoDA_nA}UD{N5toVN_p#tCB{=FRaOof+~6qNvzB$Aq(z^%PDlDI8U|wUI&| z`=I;*Anz;jrZ2GQiXG43v*Eq3u`Oo!6W6%|44_Vv$B>)NIf%bwRTcCXrDPC}b&oVofZsl%cuscD=;7&=Bj7lE-gs6Ix6W&h@BD}M+& zVwIAgZZh)1ci9NxmjQd$W-qRIqcuBV9S{_2LZm#~Pus%AjR1ZBG)%wA$1~Hf5xwG} zeC)HQ$+k&GXSVKEs8P!6w#hFYW&5p_UAtxs`5-t7>%9HsI7Rdv-BLb4)L-Mm*k!VAt*yDehgc*f*ys#y+4j9Ld;C=;iKKx4?I3(&mB4Y|ACcdnHRrcn?;-Rlc0JfPt(cMCGSu|(Swx>Fsi`;MKh?0DBK~#By`oN))MX39p9o4J)hVft^uqti1X{4j`ba&OOgSzT~4c>T;B@C{J#_NIMvAP<@f_mbx*orf!*XS!uaBS#3c@QJOZ|XeLn0mdt9=RBV~c7nv6T_S%4z(yN@0b^ILoG zRClj3KN_>?3Q^kEpA8fE6ZM1Vx-yhVwBYTc$@Sa>9WmyPLn>h2s$-WsA>J!K z&0Th{&{Wk z!;MABXtY|YB)1o}kZX}6F1;m^&4u0inZhgl_mug*eZlL+t<&R9Q1#a2O#Hyo>YD$k zt$qNxi%-5?n!RI25GUSUQPVJPSRJD_v`07^icv?DV+^-}Bv!DRzCSU#)`y zQ$O-%FhfOd!$7}!t7zd2_Kvo(ki?!3s%U*fQ|IPs7G*>EYbL9g5YoWDqp=iF~9yZ^dMPQ5gPW26mzj+m`D+{ ze6zbn!nqFs{B!u^MtT)(n5Sx*arGv=)`>=X^|{Suqk>pSvRDJD*WB?*m?y9ZSZ~Wo zMKkH6*VyWkPc@8~T@K*ZW+^Ai$p!vI>`C>WzbYhsnJi@*NtAk)m*5Vp!Cy!bC%F*v zZu46dmGSJ~n-+OWjGSF8wzo<7!9UEjzDo>MZaqHU+^8cL?s)0|NZ);fm=0s?7KpW$TsrNZTqcjr+_%G~QLUI+euQ5m%B%LIoWRmlw`!6xpzr?<){pmc@r_$!X;^Q9^WeN^%#T zwzk|+zX{{@toDU47!TUJEqzg7a7MXs+NB)YZ*51+1hw&t-ir0qa9b3wOnO9O8pl2= zd*at?w~1Cwkzx~R+|IVCT<<$~bS8epV0gaEPUizP1lESl{@heGlUVir9iQumg>M3K zZwjW7LW{l>V=xh0mTIqi6Yg1}`ycvT8$8x$XIC1F2DqUL!`IHSXK<9E zFPLn&-{*`F2pM%TqQu!nUN-h(Y=pPS2Pz-kHrs;k8?)i~Ps^Kt#?B4+WB!X2;lPKX zVPm17s&AXS7~w?O3=px%MsL0&U1bUCo4`#>_s??ClV$`KW7{fyks5r4+O=#eYR?ayWjoM577J`_m8x(HHrK4Cc# zQ+Kq``0CQO8Ke>PL1y>aX#>la*Si2v@tv>Q@!d{OX0oa8QgL&kPJz(%LG1>4R0+;$ zzPG6ZI*eo}J5*Nn0xX5uDK@{9b$*ZGvWgSS~z7uqMtmd3(l8}^r7iA5R z+mrmPCsl>hAFg}jSuq5RcGelT+Nec@8jVjTFc-zfn_M5m^_$G{8BX)k+OUjb zwF@7gW7u}v5;B_(*J$6p_klyr3BnmmJbHg5gJzshdSBw3t8?9|uNSrmvp$ib)+W2W z68-JnLWgO8yfP)4@|gduW|vWc*5Vxw*F@VQM^|pYYT~?siWy?;{i@ziI%f#Hn4HRtA@S7#q19TkLp*Nd`UCbw4%&5E3%Gt#ZWF8oLg?$K9|h1=F5amH|tFDbp8?av(h%#|I1?L}=p;1`(%KD|1vFlK3|aBZVx`Erj~lp>bCX;5)uY{E4Fs6iQxRP}xIbiw_6 zbjIEd&&s|umZ8gckYZH26q)Yt60?+YR5d!aC}KmgXJ?MJ)I8S_Ubf}>L~}>GNk06F zIa0S^-oC%bmjtxWMpWorsZ-RJ|3SpTcOJ zaZs*JfW!|U+jT?iY#gG_JZ>kXoI|;MHVyRvcZxct|Evs)6$MJj=?P`Id?!sDS{Pcp zAs7Pm^A!6T5|3b>M03&I_&_|Y-|`t9Q~dbGfSr)ATJ=pf(16LGoG9Ez-w$~#Jq~P6 zXsIv~ScTyR2P0jgw$JL-5oV`LxOx z?7z4&3o?nBZ?n}SYavWED4*r}vz@3XxDF#o+ak1SnYUol(l<*baZu{kSK`qhdMtsS zKKkQQjTC-R*O=GEgf-fnIF3NWIhl(CQe3(`z^th#-HUo0Co=G<(s8t6xM@+EUNPMM z6v{Wf^-)VbV*1SYO167LHlYk;=`*KN2jCak7R#xAZLHt5IpSHwqe3s zH5IVJ2)8NXYlLl-PwZZuTVgAoCyvh#pXtu^1zOa}8;kg`(ZN|Mf{P#_{Ak~5T})*E z#?Qf=cOHlbqTJ)p4D2}#vZDmrA%7?d(@e*rj?u~-T=i6@oD1r1fw@SG&+0G)xpDca zJ+2wjVM3ktxF-TvO&Q+Z(=YIZ+$w5)-*TiwdU-txeqY*a%cq*F#wy4zikA}_$3oE4 zyvdUv)3(d#+T0)B#(_|Y8o$J|F807kB;3>4E%CwT5_e#m!WZ5C+wNYPM&8>&c~$x~ zFlpN^+8Stqi2+*3@x0*GtEt%Zfay=kDN!^{%#N}|Rbf*?fUNUE=ap3TN*xovD2~`& zQS+juhHaQ1*06$}yiC0?gkSYXHM%3r|WiNPkm1SN3`t@gN9Zq*- z%$DmXHYQd1kt9^x+eo1}P-hB(`}XbQK)b9^*nLf(pwbt-ww)_ZxYj-x0Z&@5v7ld@}fnnokpx$ zYC-0ItdO3mLxp3(&T*ttU^3@c-)F5xzdUHh5{%M<|XdF?(`f$R34$VmSHvr5rb&L#N-{ z9WZ;h(6&9&Q7{O8<2iM~>i-?|T27woSiCC5|IOgB6m|LV?}>8LjnO z4UMO4)VPo?wfTlyosltUO69e1> zyhY?GNxa+7hGAMq4~$W>Ku?;2QmV{;XP$W%GFnbj$?)>qmBtL#b>;_trjRH1(XZmu z)VMs>=XssCSKrH&`)1qD?d}o09&+{DzWd{PbVS9GFkgQi)$oS163X7XifyCpZbN6( zf$`X&b(+l~-&a(H4|2|lhlic-IC5UpWh==+M6Ovi;jF6y{)-E>U2Asoo%rk2tT;*6 z#~L}RH;7&1s`$8)3fjkEUBc$&VLGW}miez!Z>!x$rV7+kOSeu#i`G1u=nDsBh}yFV zidyG~2`qX*pBd4?lcU`$^ez$k_~LEsM}H1AMHK-CyT(>SeEF+nwcf&OyBg$bp<%bQ%n7S?cR$uwW zcJEZ3LUqb*QwC316v2JvH%RfMvnXf-#+%@p+{u;Nfmn>~+F=}To{T{K5 zeR{7_&_P|K-8Tf^u(23-Y$Vm^=egdKGhh2&Ik4J$RWq|KaRc%JE=IR9idvs`s-*TxFX z54*&%@zi+Lt7wHUIy~BELGG5K9M7`KrEYgIVb{jSj8(LA&8Qdx`cwDIvi- zq4D=T7B284!LeyhM^$+5BQDw;-FY=C))G0;?-){UYb9Hdu(&yxKlf6FZbKbQXE&Z8 zS&NpSZgiM@b@^qtTO!qRR%<>g=YGI06}>pkOzbH3(V9f9$=xvK!)#}Be~R7KjcEo6 z7q5U-chqRDa~omYkKy(%%<}bha(UKI&-JEY;NcLXU4Z&-+~4uI;92Y~O|D1MvHEaN zd<@(fy)njPQuNlwx@4&0X-e!?CZ^mv)t6uXKq%*_DL(Vf+31hqj1b-fUJ5XGRH$>E zcaU4UHhU>fCSYSIX(YNvwPppl0L6KleXnwaSmewP6e4auh%@15yLR%yg2e<+syc4W zTOPM}P41*Zz%J2jH?&ELXxTBiytS*EBx+jq-sGg*h1-l=q@r)3`BUJ`XsyeNh6=Riu9GbW&3Hh5`eXU7e;j2oJ(9$QrI$j@ zHh2?^8?~rt#i3mES~-886cJYsuGL!b?GD=nhkBmI4b3~ksTcT+N1l*_GS!|}5%wQy19(+vL* z(K~!(do1?Q_a%xye*_b6&WdjB>4{0MH%Ou?U|CU}jv6NCuw4q*4&dzTDa|uGjT~}6 z-@c*v!0a~wXa5lKQhx@y|#xY0#(!4)>J$e5EO~bk#r^FA(2bHUzeIsq_lR1X8qDA zi+nGGcIK4(FJa(UiO=WT<^Dtz>i=hy{A0JS0b^xCxbyGslfV15-#c&{Fl%b5|H2w~ zaP@mY@0O;mMfopHl0#r;9;N)3XZ_E2^=F@0L0A+*Jf9FF{Mo(z&Lfd|2p$-#XWZ+*_V+TYN#4IV2>w2je)lL#CKzH75}kj{ z6#m_luy}M!{t3C^_ipK|1}KF+WU{*UTtz|HRAx7x!h=DD5;C;=i4 ztOHZ6;T^2y7#_W{hro&^WAF_Ayfrc}X_G z^A;my)ElZ!-ly_1vCYCG5xW{2coFLrL%{ac$TNoPm$^=-{nE+=mVtojYfp!dA=Q#Z zS^KTKqk!$$cffv|`H`o&m2jXM`sM=XHJU1Wj(7Beuh!glJb&Dbx95^9Hmpu%9}wGq_Y^&9{i4B`6IpAfh@_A z$cV%~Qkwr+-}QK5-?|Obsx0uuy5}l{n+sJZx1WPVw))wWgL;EvmSSI>c7YA|0dK{r zapq#4c6l}sA!~KQe}B)m9+LNBzmlrXzJ^pOG8O?1rXl3XFfB6%J^K{Jsi_FuKBn`| zFfNsDEsvX2IgPM~UgWD{Hyx!*yvOqCs57=bdV*vT^2_run(JU}B?>um(MYa7KNuHS zv{hTAT%8oT0q(U=ND$&5Jkn!+Uc!GTFUO}UhfQC>hyZhIi|kvO46!8d@09`n0YIzu z%Cw}l|GZp{VQ*>L)c^yJ9ayrQxFUgix>E<*7eO{3$p_>*jSJ)M$4&UTCh8Ozn{Jib z6Py-JSp5C+)SG-V?aw)Qk8SqHDQ*(GcC@*4FT(Yi3my3b$dcK0-rwF{!PwysmVw%A z+2^W!fCse9Ghsvqyf%K9Er;P+mz-%KtX`g(9H&N23G5tfdug~_xTKg?F_y~`BD71w zAP`5WAI^ zZi}dnGv+=!8@6p9Ak{Y5cQSMYUGiSF4aXTt{qxB#7^B~TpIhcl&o>o z)PYFF>R97#A3x85gOXAaX}T-jJE zzZ*DQ_Zb*hXX09!JD>Ge%2eLEAF8~aSI?7v@gyMRjMQkDzK}?UkeOA{XQ_{jHd|SI~5t@ z;w^Jd2KAr(7r;sDJCFE9ehnTI5D#0+%^%9wQNqiJ17&^sv{_ddhO0QZSaTu4sTB4LuwrZjk_fvuo$&o=3 z#Wk(pi?Id1w(n7OZY~cfn+gUN*_(?95y!q){%owGF!90Uya~d+Q7C- zhs-$d{sIxLTqAx}C2)W_VWAx@_{tE>(x(o>4kO#xRDTQ5Bkt}^RKAQJsrWmG32WUR z_sCd4JG=5)1?`45H}zZGaJl@Lkm+!7)Aqds-Y+J|rgDLbmIG}|uE(p|4wPlVONu6E z2mHymS`hpCR*CA>#}GoN{d5zBd-Y8M+komrBvkp8p6sOlONiRas0PAVGIcZgJ#Z#! z*B$p^WA@*84;cT$s$-U)TT1UJs?-@~&nMTbuwb$2~0i}3T0 zO45uJi@(Cg{Q;q9x4bh!(qMLfE#!xQ_H5;RQ^mXlT&rlgJYCi||KrH(N4pt>Je@>T z*db7@TE9=)+M+rReh_kgsn*4jBN#qX)fUR0G*+#~OfV*lZ{)NDE3lTo)9rHZ*1WKx z?XZy`oMl?JDhl)~s!LQ{Xz3IaPZh!m zU;Y!bg@1-JzkP~~Y^Kv&+eONSI^5La3!*(8HZM2aIN-cxL`jgM-x=E7<{| zeBIc?zN9y_KiK`oc#J#qzEx+yYC~l;}nk4MW|2S4fF~A zT|rsNJ+KzWZP~DKtLSD5Z}vOmtT~0b3j^&uQ!ixk;2rr<;?_3JjaH$<#B|XWo-N ziA&G!zhC11O@cA4ny4%aZbr4(l0&nr0g!&&D`mn2th>ta8)HyRel&wP>b2+k^m~-b zS5Kk3<=I4Eo@K?|z~4==e&@jTy8+xqKWlYvu57N<3-gSiwPuR6%dHA|JGzbPr@>mI2?gNNyM)ZJLTxEbR&RI3fnfkL%{^MRq z2P3cLy(((uT*Cr)G7x!(7#(L3h7m_c~5N)3))tC%4T?E{QXvOqIh*>|l5UI&UL)e~bzy{D(Yzi(+e z1H|?Vlv(t-gOA&}b#4j$SQ<}aDO06!H;LxQ7O6mZuR`3F`K`MhS~pW`jE}=)0p+v1 z;H7_uC0E0u&|v!t@bg=Zho|pfnccA~@az)m10AKQ#O_zYc?G%!RuFE{{v_D4O%HH0 zB{^pk5sS}kBfMj4MOqjoP+{kQ^(#={#1_?O0KI9O4xB)Thq3# zlq!qeB2mIEe~w!#^wp~fDO@9#Hez5E7nTfJwgby5ot;~rAz*^0K<}99LrMKf^}^Iq zAty231rJ)S_73z-jtv1l&TE>LQtGP`)p-lc)Q7wY8PRQ(wgZb#ZUNOp=bld2e81!R zss*o)+l}qnX>Z7MHjw&B0*_tt-`PJwj|jf<>p6n2)GpF$gK+mf z1-{jk$B=Bd=Hk-^dI*VU&UeJ}#exvc5^M8aI!Q%3M74imDLp~Q3Uz4TPxG)p*usQQ z;}8uxpJ@iK%J0#qVP}>*)!~YC2mY>CMGkN9@-^M0e}O{krgb0sRE2c?<;|3n%J9)A z{6V||q_mW8Zd1}W51w^#FlcJa!SF0~m14?LpZhcHD} z5k}cmx6xuu)Scq+aS#GG-wjOaeTd6JCYAPA*fr=pM#c-|z^@j!px?y5`FY2Y{z2Hv zj=oC0&HHOrWh5!ui)VhEzc!S}FLPr}ruF;>+h>hu_DSs)`N-q;dAvIIWF$0zGUcpC zi7G`mmls)nK-gi3MZl`7Y+$Q$9T0%(9W7UL2SYo2sNX|WxJN_m#z@+-+X3BTnNwfG zPP;SDN^MmMXaC&o94@z<$=BYZI=mBgXgMlc92gX$Ap3Pc-6i?i#@r(qT@Jbu!usVoh%G#dwCj<{5bM~^OQ9@|$|$5g1#$rx zY?PjrkKMKUx+xws^_F~x@nljAw=S>s?Fnzt;YXp0dhNyB2!Pd`^LzEx~%wF$9;)3ivB?$%P#(ljg=XvbJ`U{FUlwCE*m0e|4lXKH@6g*&yDMlH1sC1xC`KqIjzDAYy*x?enW8?6 z_)%$JQBZNIsfH0wUtDy6OabD;jsrxT?T(Q@_N%YW=3GbK!Gme}?0tHq3L5Kd*>Mb0 zSY1{FV{N1ncU`(U(zS!O0}OKNMN}+sy-)(Q1yBI4^?I95 zu>pZCJHr^YA7YGdAEN}#<``eXFbW70S>yo|A7IiEZ!R8`+kepV(m&g^o<&4Q2|J=M zJH*TVGTIe%sb(n<`?r9Q*`McN+gt(8^P6@9FSzXGesZLGezW#NNcmnnTIYLll|k&; z&!F)O?@dj7ST%FhU^(_@3B4iA?yXbyGO~@~(pzz^D}8bYfRHS(tRGt#!&t*Sow4q% z$i()UtfD3~WBAH*Cs|o6XdZjB!X?!q2ix-}@97Fdhs{u3Q+Ybe<)s$l=ff7yqBIG? z6(VxaZUwsW_o!Y{o!c#yX9fhaMv)2AWN|mHiH#c7jj>B!(q70m|B*4$z#MtC2mEV2AT15ddQ;ryZb#7Ngs1pbJGL#9YLI@>e12S0Vuy)~aT%mrTZ&TPYO7 z#_6b^Aa;L-f+)K&LOyG9sVkeJl;GAKRGKj&s9gWF_gein(D|q%dPTRL5-#>bK=p? zNq8-i!fHA~4}pTovj3$vKZzm~>(xUj zAzL$%a?m3}-0k3_baMXu<+am{C;4cS>gkkf57Rxf8~5I^KiP_llqt*s0DP zBy5@MxXR2nprtE7X`dh?SOd#R9kY@LZ{qAhAIx-nbcd5-%P@d~D(M#7UX0PQAxPa) zaxB-bU7J|R*DB%x?%`GVI>crS4{?!?;F8O_&KFD*E5+6}Z*Q0IHI)?DgfV|hpmR(Am5wj!NvVzs zN*7&W>p1tAFkj(Lz;JY_0R5?p$(ur%q;K}cuaC|1wO$14t!^xE_EMzjC#Ho`K^G+f zGVZv@{UN?6`Ao%3SE8=5(5Lyu zNxs6RlgmPHh&n?x=9cUjbxZR>zBZgIG?wQ*$Ysj;l2N!8nq^n(+9x<5T@*>9?6%w5 z`Qc1O`4<)!#G(>}J84zRwAe5&vcXDRYtEN$WpKdtq33KZ*1hiA@=|NmDdGTQSG=2e&90NXlo!H#O*JL4`6l2T8`6PeggC^==`Y_fshC#TEEw@g{;3P{kWcx z;(Z-Kl-LkbqZfTT@f@B|9F5;Be^y?LW%6ITbLIBaCnW>R&X{_)5Bq_~+74`L6tVg$ zCY1BNwg>ra_|C@r^i)4P$lW>XL8{~&5!UTQWo(Fi*Ioj-Lb(`c&b{Y%Wp>nu^c^~2ZLpDQXEQyj!l z%j}O4Fr2YLrG!b7PbBJyC*{ip%H;T9AGbZDdYFQR4ak^L zZ`wyMbHo1@Ybg~(K}V{(f{{Wz@bu+4LAYpBx5O0|Apfq>>1zCHen;QYzq&SDaHGg{ zYi=xX%>2ih3+PAFCG*!m5=>b&9gSMXfjE!zZg(XQ~5{O zej>$%=HF(^fBN4bU=Pg{84~~RH}fC*e1Xt6E zH`7RFlSHffzc^B;ZUUO!kuO%df9daKoDXcwuNKp{zWqyc^LiQ}EB(KF^@q_{X%ftT zv4j#3N5!LGZ_L{uJVaLz%Vjz~QtfcDoVrx-U@>o|<9s^cpKbKsx?ABN4G=*`tCrA~ z;YENWtFV1jOz=M%o)6DZ&18A|Y=0KeD71N>tE+rX$!CA&1aS#}E)oJ$^{kqj z>MOa6x$bOn_1K?pZZe`r8q!^;6RV-aKZ3fI9-u2=wXDhFM-18f6sZI!$I1I2W*9`2 z+FlTEnAf>dRiHA<6E&^7zvtf?$#i~g7RfL7Ce!7KpM0cYSU;}hirk@q%YY;q&U!iY zAJ6{mfs~mpF zeoLd$V$G(QIr(S~ko@FOo)mKgDz>roS!~Dg8Rf|`9Xy~Le(S2lRU4fi+kwtXE+BQ0 zjCh7(@?6C)09z-n`*oYiQ;ZOGjRbjK2#ArZ}{}8e#1Z&bS8#Mr<^Ay(l0Zo`rq{K&%Px! zGUVWOnIqaAR6SB%Buu|GI);krfPVOXnyaEfY?$%G)!}mQOXh#9ll`|r%5vqThwOAO zRq`Td6?9nnA{_H{ty0QC8|m$f4`f&>GAT8D`vQAPr?@iRMN$f$D{0leTXGyp+S@H~ z7>CKgSgE;nPuQ>Sj&KTC&Xz^HkM!>4px6lYpYWyrJQbky2E$FV@w>G-+iH^%=A5_+ zJKn`f$XfB4SXS$aLY-14JT=6-OHy2~IV^=o`*(y$hrM=2>KeI4Yu(Xat>4AzaBAm1 ziRvf^3ElaJbx5NVs4HcRzs+Nolp`w$oKu^v7sOsEQ5f#(u{-I3*4!2usd*%sr$M}j zGvBk!#Q4I_iPK0682mY}^SmL8aJmaQ(6GY2k(_z|O0i*CoXtz?oB>XgZ&w9Ow&{G>g9v-YaB7Q32{=)uzkTf#LKdk<_$TOsSjZO+$T zJPMB3Kz~M@v3z4n#p+2gwHs0#ek)N$rpxTrRc86r_B~?qd=tg#8^h_~^_Dk5T4lD| z;AS*;Ol+ZE`F5vTI~!*wU#7XdKR4j0MT2HN#mUd-6o^m7gnxS)|Gwz_69EmdIq?Ok zQ4RDa=VN<^j2ieN?-+>_OAU16D{S(#Zt6VdN<4{dtDGO7Z#=yG1pA)b-U_DkuJ6Xo zVl1TXv(RGZvuMRg5pQ9bu?GD^-qtJ8!*}q4L^krS`6Yx@4I%h0`X$LUiHNX zNbeS-7?%NI#U=GxT0f9~vrbW0zsE(#m%!2*^#kU-zF|ucXWuke32T)HO;5%`VORd~ zTR(F>)hu6}*{>pp4hU;zDc+!77zi7vQ!ve17~tY~*)nnYw-xoDH}^k2{k(Y=)l&;2 z(3`1n-3q5N829#V$#fPil?rps)9jE}lR7`O*%5coc}H>K7I=diunp-Z>`IMZ`;HNr zHb=Fi+2mW>_q_6@pX7uEEHr%kQIqeoCwbtCJj)VT(^^#9kNqR4olra{>a+dIVWg== z60*}?Vl+DY+@G6!G|++bOd$shfrBd*ny?v}>miVzpq zks^$@#k=ACfWVMU*Sts#uojW6ajFvzTx8KO^3up-tJ|l|t$BuD-z&JsDW`aW$Z;d< zW00bSE$zLV(YZX2qN1|CH(Ba_Z=wk3Ewt*?p(;ioK>dp0XqiTD?31Yu}+}nEA3sd*B$& z`KTe+;``H?sjelzrm^s^U%n&AMX^TpgojNf4H`v;SK8<=Shz_8!N5WTLaK|+1N@Qt zFkJEcrh_0eYnF>DwCINa{%yb;jIii%pHX<~-&#)dKVH?pzs-m96fNv()x39Qr%YTu z+VqXB{6g~KZ(4DI@OPXO;Q&x=(}*3fzjUU9jp!UKDqzxNTAoz5>%C0j>8unmT3hIA zaE;K07;%54BgJbL#;MI>SZ39vC4?3eWje*+z7j*3-B*DzL|Q`YFXHquUKwK57_Nlz zWb1B^85mx{caPr2*sjTH>*}kRo{1%mZ1ux!4CftE$FTJXPp`3-GPzFNdbvf_H~pq( z5PNT*Omv7wiE+t)tm<|i==dA>SIE-e2>B|J=tT&IRznxJwXd$8l9kZft?49N_>)Z|Gx7_f`pUY^S23~LX zuz6#qv$*&B2Jb&j-)#^=l5{@G__ZZVCfdB*gtHf~fSpHO--Q11E{tmtIdMH9-b(OYzN0w$V8hM-$ zeM2d>!vMG8dMe94XP5XHa2h4Ebh<{8%o}bwjFqWKZRaB9h?@*v^|q^|O%xg^4#SsJ zq{bz&*#X-xU2fB1?_9m>vii|Jt3Xl_~21Py)Kh3~Zm$NKG zts>izG_UO*&lY!gy-}HY{YrF*()K!z1ukPlphNckNvb(B@Hk8Y`y88 z@-;7Q3+moHPLIlFddDCX5o_$D3UrX3+x%&q-~5g%ZJ@idHn`i(cWBzStTU)Q)jv1P za{zE+5`%vf2ca;hVu=!S^f*z6HwlgF!?E3Z;&CO8Nsj5$q zrNv;OE-Ns^m6^8!JIkmWpJxc;3T}J!ar;&ar-|-6%G?c0Nj*G*xsQy*MZ7!RGz%^L z@f!^xvLlF_rmU>%vzNxd2OD{he&e!ZIzKm77Ek8^8<{NQcSb8;x#Jjo74odlvfwdc zYdMEG_D=T6Z6yAp9_h34;3SGrU$=>`>NEY0>fWm)K+%lzKFb6AUpFKQ2JE-@GVVyX zF)Qh|$Gc#cS_)y*NR)7@%=mm;>-o**z>zd2?u7E1MrBS8kD_{^G^6yPEcGhP-gcg5 z>};)rWubki*}c(ljAwm?Eg(}9Zo1BSj(bUwqO|yFq+EA*8T6{nV7E;Y8z+Sm-a{i2s{TH%o);%I{M}Ip520AyhsvK>5%Tk=iXXB4%`07lq>A14fE`i z5E`i@9bZQ;LWvSz_xB#H578pO>$V_VP9-lM-jQ93%JXk;p-yan7AsR%{7KMaj1<_k zu3QkyqlE0KWy)a!RiW5pR?6@o)jw~ia=L#Q_{j!5tM9AzO{Tw($|jl2i3|nx_>xz& zWEaqanJMSFGgCzn@+*Q47U1U@_dlZ5SP}P$QZ}i(r(0YHS}sJSm{*RtyTrdi1{!-B zJ0G#EUAR-E+xo@OABp?c7GdHSPA874Rnsqd?BxRyZsV2Z){q`2ls^U`V}MS+|H_BE zW!NPo4il zI6cU@<+&kIteU3@3U=X&8iahRicII2s+-=v|FNbim*PTzp$kZ&OUB-8~juN2z~ z?P5u14o+{rF&U0h_An_PHUKtdIP$iVRq+nCZqy!%aw&&q{3Lem(=~7YJz06)>_*%3 z)V;MEGaW^=8rAJT_yZiFUD zxM_5!B3OWlh|nh@!3gGUw4XU{&F?xS;zX+8A}v>IfB4?$ZTYobNcA@PIWb^kKE}~v ze~B})yH);vE%ld^zb&@^MSl2qWGmvuk&2JC!qjiQh3ngEftFfoa0g=UY<&VTOTA7q zjNCq#f_=!lSgx`;b<2B8u_=jFWBAC*CheC?@DY~YSpV1}{O(4M$xP8n%Ns4)yDN`d zkouf@l-1}8@r~wK1A{38hR9)Ns>c0zZXJV3XH-X?j`+Ryi@Z0Fhh6 z=xy3{tXq=yxvzNXbZp0_jt-~*I|j-ww?akpXVwX+$6llr&zCg}$XXI`Y&CbL$WprB2c;6c#p z(b<3?Z?5QA!92~-ZkN=Ji|6EnRMz7W_6w9`RiwX9Mfa9Xd#Y_TWDhpPSlRPDPDR`G(E-dcei?8?) zg-VTcRs=b{2B!?||F6CGj%xzVzK6$(sDO%yA_9&SDZ)sVZbOl(^d6B;B-GGC8AL&m zCcOwq5u}C|LV$pXfb=Fr2vs^EfCLC3l=sPZXJ+@eyR)wIdH-DhiXnvLsrR0H?m1Sb z)P*rL)1AG`v8tzIyT=7d%JxWQ64~Wl?&BhLMc?atwVd+KH?@)ph+=ORen@$-(6$G* zV6bXrA=AHHGA-y0E=^LM&F1_udR71eImw3qk(zwv`NeVcW>4!d?dD;HY^I$S*17MV z7-7w}O2Mr*Y`*9;U=!lJ>JsYaSy))6&MLln!FMhaU%{HTmnKK8%i$4nx&8Hm}8eWadZ) z^(3GWs6NvHnOf_4CJP-A9vLGFHFh-1r6~B`nYm)XeL<%75-=QfCCp+n6bnMnN)I`Z z-tTg%G~8VqR2UA9J_6;*sZ)_5olP)2iA+^<@njZrv%P^GZozQXeTJRlIw)X6m1k)) zyy(Rs2?g4emwFvOp4GPU$wevPOxm93ie>c&l8p)X)|=Pf3Ql)T>(9LEJ1r#isPvNi zc`D9?Vlk+K(A`Q}p{7hbIuD#5)4ip@A$y~#qbjB+VT0@f4OeO?3Y~D1JhtVLQA=M` z2i^6vdvAvKiXQVMC--DD+xHz^7~7L>J8RNn&5gp{ix#2pWaH+P!`;k_6N07Sb|>rG zl-S@LMfD_B4vj>nzAw1g;ZcztLej8xqtEX%6hOHAW@4VB0@^mL7~?+DlrEN#b?LX^ygyB z_~X4oNN-uwqSX>bTdILA?PJZ5Lkx(!F_(Fno5I}kf^1q|8xEDj?Bq`lVQ24l+DV%D zC5;X~JkwNlkN`)l+v{3Qtz7Ih09EA}Ds=3M;?6Hx3v%6`j?PtcELX8(JkmZUaM~_r z$9A5V--^WAW1*(Y1wbcqWmYXh>v_5r0$4n0v{t0mabsgMPAzMYT`s z1PS{S!!V;8!ysG7yC&L?A{D>AtNG&pZsf{?uvUrpY^>}$WM$cc(Y!fZ!7XC)4xAEXQu^Y40JuQ{? z*isU`Ma|c7 zJC=fP7SH#M4^5MYl*?3AjGfuJLS(C&6@Q`^p_2alveSgkdl{DytE7(V zbAM}!AJDiz3>T7zM2z|tM5y-?jT}t!D{Shv10#e*Ov*p5pYWQVp2dd0o!AI&@7plC zIHF(Hj3RS}DscOEBngs-iMtJGJuS{|5xWDM8`56?*?5I^ZtcJL=EgVk=p17}PLZiO zjpVmox2|5m_`bPdq&W}TR#LxGKHP37NiB4rUWL_@E^D9d)b9hnXCig6{l*pGoRWe) z)YbBk(Z~@}Oh<>jw7)SeM$D_PO zLpA+F73Mk1A=@ZH{mAxnqSu;-jv0zt#~?beWd~*u9k}OuVK}-oMbF~(>XCLKOuhU0 zK=BW-xPFIqV4?w!XDq@#m?#Wy6T6n5TfGykI0q_HzO=D%D{gLc5mnv#dG^$Vv zb?MauizWm2JgAS1O7 zErMnRNPaxN!=FykWAA?SZ1_^c*qw!1qRNet;5*T4etMZ+^>eAsf{ ze8rV?K_mL=>$bvm!8RSQTbBL&Wtm-e=z#s(`O=4}!!SaT-~`5%=iZ9kHK-oWJY`z0 zUi!6_X{FOQl}@e!IrYk6{bOAzq@+V#@|;GOzg{|!;Yu2wWy>Nr>7o ziKHPNPBeb^RxnYCurpc0d>i77dUvUov$dPLoNxDK45gHe6l6$+91|&AOyj@hjq5A% zNdZy<5DTZBBl~d?4=S5-UXoW@eEq)Faxwnxa^~8%KYix>IQz_bGyf~v*-!VJY$B57 zlb2mBLno8J0tdjM27~2^8H?;~)XMfQnJPPgQ|MBeEZPzGS;|-P>XZ-1G{POcQ^^Q; zY?Hc?jPAfXA_+t4XjV;(uPB(=yI)yoq?@acu8@29WRY#_H{qy14PwC$_;uflx$V|a zxb3PjdZM_H9$lZ)EM?T{paG&o4TAN#Xe=;hOAw-+kz~eiUy}HJy%Sf=;yu}&iPap6 z@F60%(LTI>+l);b(jv;9+S0c~I=&h`v7)!GkUQi2In&+~#(C(#x=%XCjjc@KG#Ab^ z3P_DRjymI;!4Wr>YnI9j7s!QR^Rm>Vn?@b5ocG2t;T(z+lHr7qF=SZ#5=821j)r0I z;SX)_P9ey7pj)6|B4Sd>)t9YFF2PY}jPvzkO)y1r{ol-xW~G2K>XqSa;ZFjdjEdpn$NO4eOKBZTUXG;-@r!Y)cr^l^wO17 zSd<)}hnL=6`BYKHtDNT3$U}4pWYE%b0u&|Es zfp;P!ZlZA7V&=7fdx8~~2C(>isZM6O8PWfW;qO`QpFFyL^u$?7idLNZt;tjnquIjr z<^9vgPI&F6%gn%e5@3+jpk6!YW44WpVzZquO(nb+A+FOEqC9V{2Ya14Uw`oS40@5t z$Qiqhaq-569N>M*`EJIr!*?wGRtCAn7*BXW#2cqQg&pfmY9w~c(`AkfnCiRV#ju`j zKRM-B0EPIsMU0uMf;_^b%2M>%KDxc_WE3!sn$X^h!Er#x7je%XM$^L`I_$^zf$*`@ z$P`>U)9cy?eNuth(ok2KmNz5m-p0`WiD?;;IAqv$(2&vm@#Ck9q=;LR+ULX(8wAeR zGKh_WwcV8l&s;LGuNip$e5rh}yGha{Y=T}}vsC*boo(IyVYnAxd%S?5R44rS&Y~E4 zoU)nb7X`Al!XDsKO_Lc8Y5wH)w;JA00R;XNi{D}vL3;0}S*N)0>(~y?)keAV1iDY@ z&XTJ!*<`TVqi&WLNkHora+let#qs!yM2~eaX#px-?4IP~7}6}#XE=_-^ZoW4vJZYk z)NfHrYiB^ftoaepJp1RQ@1HhOq{*T3A&jJBx+|j7Xx)c~Mm=t@Ci(GaEuB+9qqV;c z>%H7b?ej_6h}4I~e44ZMD%rb_d8puPtQb&D{d85d0(!S+M@p-0`rG`R$hlv$kyHzTTW-w+8o_hzi_*yRC6aVz|1@m;6M#@w`FlU@M;M=JuOT}+R z$o;&&fQ#qAB2+E2mPLlSlb+zrdy{jnl|p|~2#+T;vz!(*(os1r0t@p1IrkxMx25$* zCX%BF@<^1hsSLfSX8h+o!#*pfm%j~cVu38o#% zWFExwX*~+(BunQ2*>URdEasAW(q|W{16rkP$<%UdS+QkHLUk(m0i-RHfg`5e4k>4m zn`TY=#uq5ZWb^&fF)$+p-a(AOI|yJAd9z0JrZiKmpQ@mU^s$G~etiiyT&)zzW$@=` z+>oLZf}e*=%^ZTiMp8Z#Y1rSv@RSSuJ~!%MTFG311N>Tkj3A)_u$=HsihP~#Hp;sW zWl4W8ZAsDEh^X*;Q|92Pak)`4zOuUl8)99a_31>^ zc#rJ~wrpC}?VNb4*0_2<;z?jC&3t_#lpSm$e5zFYp35r>$^)~VAk&$w6YV0=*P zC!UqnTba*<^q1O`^rYc4p=|?SUb^b<33{8Rs^`~s?b)Q1Nz;(&IG!?8DtnFF&!bCk z#mvuqz2Y?O_^5_Sed~Z9WnCYXPF_kBAAWZ8Nuc_()lc6|#vM?uuUgO3!aQ*! z?gqHd4WS4rR%d=stp2nV5)i}nrry;7G_OWFnW*Q#Mr`zTV=mk#7d+v>dEUeBEmBC% zhZBJAyI`Wo9^H`>3*c^wtsdAdA`hsCy;-MFvL+^7F6)X zjej!EAU5WA@{JZ;mt)TOv6-ZcmG zUFZmJKKgN)9xmx?P@4K{ArW~aNUupPvdW_O3D3aJfE9MkQPPe(Y-A{2|O9ZebycT z#h9cidInT&WCXkP&&Pnn&2n?-D+hc<;>vxGb2~(Ilr-rr-_KD&6GLH9CbwocSfDH( zzf%sl{R0;I^Q)eryAnB@6n8D(tnFo?*RN_6>Hp%?&6+Vmn(AE(_&~QX5O-Bk%_&=a zyi6p^2V-1dKfQ6s_e1$CHBH%sr7K5mNBfSC)SX?48!ilGiuhIj`JTd;@}VTN$qoC? zQen$ak)sYNEZ;5cj^8=$#E@DY+(|gN?ZR5PqiJ6^+P^HfB;@~c4T^7dOe2(NAE?3_ z%U@AP`2NhNky4={y*WGfHZQlf(0um%TkJq(fE;NeM?-fGqF#xp$c6c;raw7lwVG_a zvB<)#7P&GkZKeOY{#R3Xgjm-zQ2z_$T`X{D-o0$6xG9LHoc<;46Gxob7w?Df9h!BE zPIqzg9ur|j$$D*t%5_uZd+euq8c1#V5%Q12w2($tgO5X|e8u2FsM;pEg!{7{Fs;Ue zn?Uo2M|BU+&DL9C+=p0>3F1AN8lzq(3Gtq~F)#i&OQPTm<@_%b`sY^d&(Df8*y}{A zS6@=fe-G`uzsv3lSsx2Np6y+u5Lyb@7digc(Yf!mQ5j`Rm=kAC39&W#ZxYS4s_s|j zhKiKJnl+?>?sK06r1ONG9#zA}ly(2aTXGKywd3|K_b15h-mH5>>#3pO>3mjYy?TVpuPl(LOhp#ZfEsogQnFfu5pG*>}_8H7fbQ(Hxb>^ ztkV*+eUb5XXD(cDcw&0KP#C!CM6G47Vn!;GBKLp=1vl_da#2kb9hh;QxpZ!z5aM`& zIZg_BJLd95ntw3UH>%%U_eE&9jJ1J5!W=u%LdtPe=)ufTS>g$w#NMTT6H_{Q_r(PG z(iLO0ili745NhrX8pB?&RJh6qS^aIT!A=L1&mf*;Yl=suH`F={&jZk~VEMe!&axf9 zX9AK8$fJ2opf9-YpDr69l&5X>x<$XA3|TbKl>W1g-6 zrm*aHT;R{c9Bz+&)mmORXXekzFu9>fPfc#UiD*d^%yEMa8I0~&17n>lr^p-nV=ZJy z50JD9`-*zJnN9dYfEa;Mc=lWi~B8Lp*#M z^QsE#9lOrhE4O%2HovN|(zMtq*35P>^qEc;CaB#j zof5Qep{8NcJ003sfu$-zQm+`twTq5V4`eLY8bu?PA_wz#OY*SD;c^K*DXii`S`SaK zy+_rknM7-xE-n%Xd<6d3lF|6* z5l0RN@f!|;P;Ux7&!?TE9u3-~+r|S9m{uFKu^BD*RHNrm-jX|2bHgQ>^!V7-;YOLi z1;J9UuT%B$tywMFR_HdS`*qtgjoTjXA$Zh6GpD>s9jt%|ODQ6$y!!C0AaBaieIAzT zEfIlD%4R~Tr${z<5BBxYP1Le)<$8@57h?GbXgK3rfRmR#nzo0g`PF?>p>_`3@?d*3 zhC)Q#R=VIME1XIBQn-PvovR4@{t?ZEIi~aCvb8R6xZ_VL-Cn&Th73$1qPoSj`W2gw zpac1w?2RhKl47)7WbBUC^QAZCk-pQXvywbyHm=8xQS z5cmvLwHh7s+H&TP;X`kW-1Ehet`W8<7k~A#%Yo`j>k2b{OY6%Y z?6e;mCd2B`TG9kU)X&#}*P^rEzwuz%UH?7o(}Q%) zRj+{vjZwfNDSl7O_JoOG!}4fZb6)CbCp&&OR{zlj)-(pT<9*$fO-}z905==%TW$zu zYBSTum^3^Z3uU)E)QXqetHGHoiZ$$kGTvq$i>%F3+^Pt{VIeD;&Rb^3!YS zXdkJsu6CoPVaA)oq%HA>7g7}=rAJcPNj+askZQvFUlUr~(HNW19wv%keddAH}5 zHFKJwsV!IB6lVLFzDoGxI@4!7+Eo>oXRBHqD6^@K)rAOS`qr4MJr--5W7-+W%?dvU z<3{oc^e&q>E*=?o8s`|{i%8VU5Gg!TxNmy6evn6KN(ENZyq9z?(AwPt?cq-fC2$NK z7|~5H7^7N}RV|)$l68b^u!{~w%gt)MJ+K?Rq_bmZPNr(s&UFtQ)q}o~i={wnO)f@_ zPKlW&A1*RZ(6sj7+C^~puJz^383l_VueQaDeK~4by-_!0RajMo%(p`)u|KV=N z)IP9sMXgeCJQZf67PUPO>z(dYfeluOQJ)QfP1CW8jw0)3%vZ-4;Z?7e92nN8<&8~yrvx{f%E z(uHeWoW83sfSJGn+yz;;Vv{GriY1x-_0{gnOO+Uo^J!KLC63}3ms!<|w_AACxz&@z z2PUAql;)#81UbL6ie_xU2hG-L6lK@A=FD4ML`*XSlT9&T-`}A-!v6Nfpwo+%*3TTiJBh=XWW&I zNb;%=n6DvqGpi^H3lgNAQ;OFr_mzD2PUpWluC-!PkT=`V&S$B(cVE7aqybBJt&6s@ zDLvh7-`#@xdF_REj#X1gzgF+lhD%4MBBni>P8S&0S;Uvgw;#-B9umz`(fya7F!qrUrCU~57(*|;|tk5r&)VD58nlYrZ$!!6>eLiU}_f?W|hT+vCG-5+fJknjl)mVIaAuTCQ%(WD$-8htb z)#5M|Ry*UdXb_6OI@qrLZq$gE^zm`q8TIMJ8O>Hd(gW*i^6ez@fm20_50ifN+gh8) zUTFG0f0av#-gbSOb&KJZd9=t5BhGl!XD~?JdQV*NO%W@3`xgA`@!Xut2YC^F$JWm8 zz1$YJ3DA8Z#|;?R1tTQb1PN4rKY@ZPcdO20G#(q z=h2g_|4JV9)i3`_S%(3YllZy;EL0zZba(JKty$`OUkRNXUdqxm9CpZi3SE(F)!9~n z?q8U1XBHn4e<+XWWQSsg3ehUqDtMt)mtt#-@D-r87-!WMk5~%9dH3<9t_;;#hq39s zA%s$nJQ=C>s9x`(%*&zRS0|hRT@~7VrXyLl-W5lzDySz-W`6}zF%t-=$L9k6t&G1l zzdDH>wOy%-p?u-)(LZXJ)85w%yo6$svQJ@<8_P!Iy`|vhZ!rZ5vB`Xz@77l(c=6xu zy$XmCcmpiDm5b_Ksc#AH#uese%-3)ZPdgpXjXH7%tSTLAn>!aDGJJE6mobX$L&6exST*UchLYGC*4uTH&@>k7lk})5i z9c@`<5zKl;e;wan61`gxaIPd0qX)SnB-B3icHhzgIwqMB@hKXQ0cDn6v&ZMMR5Qsq zTlrMT66hDYOnsVIzt7+`-QhZ)$zA>d12rafDGuc6v0XWLa-you0RwDYPqDErRCJj2 ztYCe8ft*k@&GA_>VS)U6$A_}lenGvWIK~TOuR?T}V|X=DQ2}WTB=3`3@)~KiODyw_ zfW-raNPfS4Q|jffdeR;Y$@qP4zPNErg}}+i+xeM@{O5jU*PK(Nir)L`3Vf;tJOwwq zzH6+MH2T+F#$}Rvb2UmS2+KLB1Ghtxpl?c`0Z^d54HjU?ESt4o@7vaj!^paG(!Jhr z1OWayCDp`>eXvP-JdQ(IFUIX0d*ytO&RW7k(T%%Q!5htXwb68}$_vux3! zo-qi4_mIGsZYHL2ICeeAkFKr=j1OFW*0C*qeF-_wPF@~cMyXQu&0ewMQPH<}pKjn| z_y+_*Kdv4ZHZOO^tW8HY0vc{cXZ<_2Y@!cR-At?D2u0c1?aJlrhNW(p&;D=1pH4Hx zW;Q>%Etul3=>3}AnervSyL)(b(<0(pbICuU(EsMY!v*`}Kg_PH^^!Myga(%Zw)G zl67aWw=j`RLIjnnLmbwDogt21`FDXu|ed8nod_p#+EjZcNppJ-$2S(YBZB)hefatu=AH zV32ehQfp*LY8WrRxWfZfm}{$wzYqUb&L4bd>(@8pBK$>(%Tc*FcL9R)p0~d8m|fr5 zB)-A0MkRp#hN0znTOqPkY6Ql+Z}089=)~I;m+69Y4x>KvnKz;a*NX6NCJj?SVP+LE z3W|4|BmRZ%hm(tID}1RNjb0&@TXP1$n>=V+#yR8o9IngzNcy8^M+Yi_Vh64K`iaWv zdwSMu?)M1{V3`}lzV!J*HEoF%hB*j)taKM5+4%T z%X29!hGI|v4ldqIL$U8Cd4F_XC||92Ww^WtOk3g205e`1#sv2Clse!gZFb^BtzJ)` zhhG$OGy|1Ah&yrg3Sg7Q&8GV5^=60p3W}K3Tn1_Hu_FyBDEpR=Y5sAvMgm%W$;P;1 zO(udqJ7UxrqON$@qPa80?7(#6OP1)kD@MvV*hoPG+F@)7m4+-C4F8Q!5pc45*_h*< zN80+)_T$$Cznd`pF@YcTx!^ubm3lj~oymNh{;My5Xoc_q{{|7yH{!h(`RS7~bPNj& z#q+y5D$+NwWTW^3-8$OQMVO{(jf#g}c%B{Gq8ck?m*V|7;i@PJS@%fYVj@FHbhUc< zkpo(a7LsuIY?^xw=}4tX*o0#Ti96-0wphy-r(o3of|>KL0D3^y$>*lLvUW{j>`9o3aD>a@y87_B$2Ku9El@~8w zyaM&8DpLo~-EAO>Etoz;;fWLC^%*Hcc+7vnlufVi%=Ki5FjzISE5hKd(R>j#eQ#fu zd(HDe<#qFPVA(ito?tJhH+8v>#x|zot^9RjfG_+9@4?wCI>m2l#`5Ejv&nrSY=h1g z-jq?#Tcc9@Hb7pjI*T4V!%$AB?CI&*_$3NwVlyi><~CsXebV2*e(!)R7G=qL(xb+Z z7CPRiJ`*1f4Pp^!T@)3Fd7oluf2^y>L{ERxv3@@Mg3R+;SW`Ajy`oe)u`8G%nsV&S zr^c78h25B-GC?3wGE(i%QpnBuDLZ|oB>QJ-Xrh?a^g_Shvke;Z*>Z}fAovLFu=nt$ z!MhJ#m+$&f3w2pc>l7Bdy@1)|Tl-jn-U$?ryP7^1IIjM&$N%d`<3mTXM9rsP;;e1& zU{m+g))Ww+C|vPsd48lNhQDuJ&Nrj8zw=--s@FHRHP-5Sou5LpjOWUm+e;&Yw3daR zpB+;L8~#6^yhnzA+TW7>mBR4wT(|0x ztFaU7vBN_}rq$g3yW3_!0oY_66MV(E!l18{nQg&iOWJkjwC@iqfY4Uv_}0;^+S;7S z$5Jof1G$=AqL7e1G|kI3}=v_*q@K#NM-t#yKG&(BKSt~2eWh%OEakvwJv#I@pX&{SPu zDKb(8bgaMM*xwnof2^@zS2?2$?GvH!2^rc29ZRz#_foT^VS%1jL`J zzKbuuk&&6Hl%EcLMtY?5Qe|2tHVfD`N`2~gIS0PT^b|NhSVhj)l{0mQ_d*N8^7ADXe>kB5L?j{v8b zBeHzIaX|iYR)2rZe|SzS985-xD>C7)efSe600U{C`E1q?EbG6I@%7e|fQ#<3#3J_B zKK%Zl0qKux@%8Ed@CE;T43wooZ>4r8_EGj<`|#JnywN>PiT-{D_wU}H|1MZGtr-3G zzxLseJO}gkYq6`=k8GDeh6${AuyLiyA&GzO!|wz0#&N=3%?&Hmr6l|Nqhzs>%?oBhx4;6Doz z|Npt!RZ7ILp5GpYq{uRVzpLWke8^YjQUV_a9PpMqTk8YU(=E|=1@!aWGR^GYBcVV6 zzSO+o@Vyin)aOF&W(NhuP)?CcwL5yiPK$4uyvPq2fzJ4@@3&6=u~Ytj`U1+=c}Y)& zDN>gefH4LxfJ8KSdwVh>N|MNc!0g*mF)nsW_xUy*ulH^kC~C)c0yIR61{dx=IYe_u z%GJWST%T=ip``_t2g8&6%SsO&YU~f&`FDZHpFcOOvp*Nv+@GrGZ)Qg9zL6vjwZ@jS zQBaSA881Zf`0qKS{A^m~=CZkBZdha4f|BF0K%7*L6^Q+e+4F;H0ziPq9WwJSp$Bn? zGFyP@t5>DAjYoUPPbKBuiA`ZI*e<2d&RqMk7$B=Xq{LVa|XW`68>z9{PTLP{Mo5PFqJ%y*B1u3o`VrO4kqtyB2AJo8mjQE-bR}T zN`T_6>S+;S8mvS^`i^T;wtEgU_@o+Azr+rzCaTN>{#4{KR^d;;_seo*Yu#l?$DRow zhcxw7yS}Y#W0W`QwVapMYGMfC(6Q6SooIhB)1f;L9z;nwn&sXvdJW1a9kV7T9`DBL zptx3=_~qsc=79tnau^wT0tbV&0l=ozeDV&|7$||i9w!5-+}pFPmN#P?+s{5b#^NH| ziEEwj%K^Q}v8By&%8~s#EW)N$<+=|PiUekiJ18_%5O8VBhke43_)gtqO-6t-F57(TE{ES28MDCgR0m)Pu)2g;HcZ~znq;E5HXT2=1%O3R?=r?A zma0@`YBrJ%($Q5rc$l6ePe#y?Qeew*sVJF7ZeI24N`>L&^m)}@UuO2FH0F6K`n8|! z8X2e>P1dAijOj40_&9`2r?5giZ;0D2=kt74Zk7d5=5Y=g9Y0ahXDi&9M{0nmH%U^c z-mn54Sy2(!eLl=S?Tqg7wc!2p^XBgd;2)1#jJeiPNvT0GbPOCq<}{)2V&vVvJY3)2 zcAtzcklRkln*t?NG>Vw!zZYv-?QYtyu#P z_)P^%pl_)B%#U#kGrL2J(yj$7)EofHtu1TeD`98p{_oYpsHM zvF?tu%Av1D1KoMukoDdSDFjUZn0G$hPd+|qO>wdBuB1uPn`Pqm<6FQ@lK)|F7u+3K zyd9WH4eQ3_93lMB&>q=5Q)$i8MVHh~0pjCt%W}Dj0!9fv<*j7E{4CJ6<8iDWfzPJ& zj_G_JDBjS~KDmRjNke&1d}yQ)myJb3uY6aVT>YGImZt;1qV#YB3&V&GuiEV`D7bz~ zSyaf9-CfbVEiah)3R-83q>V4QG42(a3PSuzVY&={^#GoSZF_o^-eX2z+*lqHaZIu& zu!RrUkC}#M9xeq!Ojhg3YXqboj~hb@S=*)?v0>kONe}6B%3oB~i-1R+rvH>WrAHC#8`UFzZ6}Y)u1)Ceux2A{61<;2Rh-Xfn zdUPloZaO=iUxgllj7c0H!C2h&>VEo4BB4M%a1zzNUIWdGBuFy6VxlexLetQSZlPOD z3YUox-+uZrf-{Z#K~{^D(JwL?FsI#IVm6(X3_Y11j?w@WemdJFvIFcERPNjcjz+{U z;}gMe)d>RwSIg9tNJ=kRt1m29MTtNZ=I!%B@ae82tVy!ywO?F|1Bbo41)cm6`j9z>sBGvr)$JRHS;cqb5YOwwh*9Htg;q2%*tq2 z(~j^d|FMLNKZ3zsUu9Na9mnV$TJA%pv3&LI(D_np23lc(+(3m!&OU40i4s_WD<;Wf z3IqxE>kHR%SgiBD+Q&LM7W!aZR^QQIf*O~&ywy#?8!Iw^Ww*fZSbaONA6WslA@=>c z^|fx#eanu=sc^KYkkY5As_ax;)&)<0&=AzAP+dpxLtt7aBV;GM5XrUB6;q@V9(cYK zdQc-kV?av<5ntGElOz>?^U$%O>fup1Dz zl1U+Q%#Wh9gq`Kc(bPJ7;pK~xrapok@KhA5z@7kXkQzUfs7Ao+4|^ z(vJ#}2by>=xqY}5jM;^B()nc$SEV1?p*TG*2Q9`r=RV$Zvar{Ht<`Mvb~#HfjbH#= zOwf8L8BZ3G8*?=4VRdHSJ?MZ?zNmMwJ|y<|InheSiHZW;IMVLjFsz4;Y3LmX+Tb^kr)D| zaB9HUrYnUO_|dCgw;gnNSoT}4@B&3yd38x^ww_MD+JG?IE{(FYTEp-%6|Dkto^WxH4SyjtG4NMCoO#%k44fGq|&CNuyYs2>3g~ zS@(zX4nyUujgc2Kbg`56c3L^%`KmP=c-L#LGodVgOdTos*>7|{IzD8tUkB4;8&UI{ z3t9WNH^mWtFC=;=)9glvy&XVXsa|5L~ar7GL)zsDv=>mvaM-#z&OdfjGwWBvoRy>C_S&9o=Jy&kZCU<>dpEN2P~JDoj7 zoP*0@{jG)(Q1hX+dA;}=uhmca#32fCm-;zZPl8YvFB}`l`%6mSTNex=z@2Lto=jLB zC^Sydg{03{Ha!mwb-TG*Wbx&qV_|oyv^H;Zi}VECz$qo-Fp?Gg|R1 zC`cw5eO47ml9t^T`a-&=lKK1;&uvJA6if(gD$((>F3TXVcQ-giyOGI35prK`?kxUL z{pxQC*?1F+a_=~%Zef}*fQ%n-7&3?g{M*Rq&yU<)e?w1!iA`8nk7#9wV$aS}fl<)y zX3(xj9HQ0D{UUl)27xh8nLy?q@Yhc(uP>A8B|AxL%Q}{70W0>fE4Rs&V-v-uYRQEbS<7>Jz1xSuNAQ*7ns~-8j&D z@!~~c<0$JA`{~xs>!eV9xEPQCzjbTg%Zn4m={~N884O}CP zeM_x7?BS{cml*cw=;)fgAKp}l_)@GWZE`2l$@S}mKD^xEU#MdOs#T=STgdR5;;xhM zjX)p)!4Lqoa&yoY)>k3BtEs7l`Y3vrqfJHOL*j`DX-_jimoP}T!Ufju&YbUADYwCw zxaQE{jsfe!Dl^(m@iHjYv5qJSHH{cUJEnn4)g0uDy({0E-@fwN*`O{LRiDBZu(S}b z?|=~Wh2C@WfSccJR~osZZCR25k)KilY2O4b&)69jH!?s&j;D2j1jEDI?5re zf@fxZ)qaCX^2cU6lC7e>yr66~9F!4{)xpS%8Y_q#`j^nN*uD1w0~d{mH6Z31WkgrZ zpKVigH!)F5)6DcP|c)U5kZsiI5GoEKh#v&a}P zzQV!5k@|*&9l?~r!@Ql4AjWvmq!+gC1H~1V38dgB_ExqjZ#KiV3l1RAA1s@|%`YmABptOD+Z zG+vGs2B_7g36tW*JMCp%D%om@1;rRsC1a}MXl03KGGd^>@U3H-LT^($%~q)wgup%B zKETCgDv~}GuYRlN0>s}mgyo98qe>|m?3Fe_kvgFRhZ&;g=4{Vx4%!T~^=Aa-wwvW! zt5Yf+AyB_>(q1%Gg`G_culj?_5^{=eUw$FPO1oO~St2Lxo_V8`0fh&@@Zp>5B*WQ`>#vw0m=8qt90c=90Xi| zF?^f1((T)o=b**7(VS(|QJ+4CIwVW=5C*6jk0N~GFHzXyTy9!IirR@dCsv>`m73C< z{E!0roeQ1&vegM^WJDlEuK{mQ(;m_&Em66%?b>CyZ!MZnQ)agJeU;g4zI7e)49EbE zi_i8vtm?ec&|Q>H0_PNJX47eBt2Rh^ZKElSt))4e4MTi;!2TyiV4A*a8+drt4`~+F z`gDvLBsJSPGi=LMB#oSfYSTe^i>A=kr}Q&S=Vf$OsYf>Xnv?~nn68?8?+lgPBGNH& zl)qjdW@(L&7g1}Zdcz64Ypd57|H3exzsVI#F9s1H`rIDfGCh{)zf40i z|1(bY|NQiz5tm(bLqKg3)NR)-dy8_#s+HS*?Q^z3rJfN^|E3YZtTJN1jzAmvEVG_+ zX5Mgu#5OZ4EW}=t=q^G&t@wEM(4j+ywe<2*%|GmpbcD<4(y^@CS0dri?<+|DGr0MG zckb`Rk^MiC5_19Vu`RfVI?Iewc_r+YL$59(0^ah+)zLi zXxZBMMf%4E;J4MowL&8b_%kK!`}ae!U1utry$?tioShky5dHFiF01C!^( z!G+;kNT}5mM59(3Nmb{UQ++1Kt})Xk=s(pbKzqx4WZqrbjsokksN2a4;srcEOr!(H z`2o9Yf-sZIcN%t>-%-RWX;lgEjweV`zmC#)!oB8zc{Le?WOwP{8A;Y$oK^fMl7F71#kFO3pOPJj5DsBl@APdQ37vn27~tIEcVf_Z=u+_ zOs)~AyiBecs3r$>QPCR`j|ece0=euXFk%Iy7Zv{Eh%@b1%FX?eFMsDt};obFmzvnQa zMX3*)=H$}E!L6#5bwqGu8znZC5R%Ek8#&su z2zw0T_=G{NlDaJw23dYx93A?VhXZJv&KFOJR726HP2*$g&^3m#xSba4<6Jz_oE7-Y zZ!)(Rs_~LZBbZb0d$68Mx;-#pw1$d{4afX#AMD*`81g4>s<;H3!HDbS!q=8=Zx9_U zKNDH85MYl;eK7?li|Ayqp?yAR0;Mxkej*o4WZIR!+A8-+ccuPrXr_wrDDq;)qSqwX z6Gb_RGc*{E)kpLC6~cG(!66!X|BBCp*M&aMoZoAjv)?m(2T?!>cMzC;xIS%d-_^id zTwbnkY8`Az|86hkOV0K7ArO79u!Ge}7zWRV48!s*h#Nx*@q^gPk~ZiSSW8$u2+NWR z{!@^oC>IA!O_eNMRn0e;N0db5Ur2*@kn7K4BHLYQ zwO_Dn#{ACc{$vrT)me#_wMf{*<4{W1h%6YC-K%0wSWRC+QC+>I2R|1= zuD#if4=Hr+#~3A;-c#k7A#Szq6#ZaDzoWvflC5H*72uO^=QBL^E1M;o4bR4Mw}$U5 zn%exF7H$xS#fHH~7v?L7a`xy%*#iW4W!C7wxBG$_1UX+@TYCd)ncRXoc8{zI>jm+X z@;ug;!n!Afj9n%g%F~Kjj}0*n}*ym5q4yV zSR9gWO2G(6XUeY!Kio}Ujjt8FYARZaq`XechsGAnypGt6rmKqN4cpo+ct$LajVmEZ z^hv}#M33g?6IYLne}Wh-S(Nyf1T9+lL@HEE`gfE};yN@tzX%FR79+`IRA>kjc(ySw zun(k+-{h*HYs-47y_rEJkE&J2B#9Wv!<$6+K-CXo6HCpvvZZT9`yr{CpE^luD|iog z(hXA_!O5vP3Tt6u$bzdl@SBsEhm?6hp+=#C;A7at9m!mrrx~Ry&G^p`&G3fZ)ZRjD zKTFo@&gs1Qjl<_RO3pvtO16`}$b7>=N4Y^zg^TBiOAt>}V5Iqq+lYsWGybzvHV)09B%Gc*|#)5-ki{oGZ@i% zCWs`={-U?Eal__MXQ3wiqBtZtH26z*XyR8cQ^sq>&kS^5XbP1|SPC2p-fOUF3NYrD z>1%!b&dVK_A)PJDS}>6p7N4RtQovfIJ*_fDF)3op@t%y^n?xd(p9G)8lSG%>_&1Y> zqClibqOllz5#=|NVt}2-`ycOHGW^FDGw?Gaxt;B9Y&q*??4<2?>{+Ll$~?u7T6pA_ z6$I4`N;|~-%5RuX2GL{8#@F8=#N5R=#&E@8lIO=YV5%}GmPpw&sg$cURW?>eJB6IS zUqxLlXgyJivFqRFbWa;edr0R^&y66)*h0s{Fc7NnT9`EhUq>9>9YkGyzP{OCoqQ;q z_>DC)F=vuzQUwIT@K$qu;z_fwo?4F)pLfaLFz$WT8yyLze9ok{x3wX%;OliMH;Xol zPpna|PMopLo?$zGQ#DIJ(5LRx#-izZl^Doqt(q)xW`~&Zj9ns$VZUoq(I5>w^Z%VoG|-N7jiC zolUik3ibN7+RoQ2`V$#8My5s<0TU6{TOYali+}I7vQDKqW;(WY-+U&w-?LA((=`Ju zQZxYS2$!9!ok_)sHaI2T<-7HHmd6HTQ#T2T09`doH2c{e>&`)dU7Jk&pA++PGR2TXtQ+K9E7s>!p^jqM;bBc_fjjYmjq|D~M>DX1tY zmLiX$mywm$l@g65#^>VNHDr0?rExsI0C0VF9oak`Sngxq%*5MAVM3Wj0pr2r>e4sH zlkz*wH7m*^%5N4%SZcUQT-QM;kNt8y@y*+RPaDl&r zwIuA#565TNALRn+x)@?p8%2H?{KT)oZ^ynz(M1!KUdvC;w>ellV4lvk@wTZL-XFah zV;MWfb?@KV)crvHA+vVPM%Ehq>vT+hbTLVBTzO2GWh%u|QAL#}N667FbH-So)-iXJ z-)iiu(}y&NHnK$31m5j~pOjAo;jxP!V9=fM9Rxm!21NoSzxiQ?+bcwge z%f_M40qEY!4FP^<`px(Se^r0K!DX7pKhazLXmcuhp7mLWg{4z(Q%ght!Yayis6SWs z$?(P<{QIit$H+>GqoIl++lN5+aPa5v$@vo&-lJRSz0D9|8+{y38khd#|JCyFgScE}jLTfhA>`Oa-0*==$>aQ#?y) zRVW|+{iE;4ok6aX;)2&H;SKxjGDc&n!%M4VZ9&J|;zN%HKhJBG-s}8TQfiLh*bAkY4BnR}M|;DleUbL+a##+md@|GlD9>Y=`&0p`=;{ZXDy(_HJwX)0-@ zx1pT_>9p7Jckpb|JX~dV$9P-dflOa)W9B)pGf1IjHqgNP;VwH%uqtrm<>Yd;1%eFu z4F0&r9$BT6utXhv(9v4cfgfoGdq!;%j=2g zX?$n#NJ-9K$SwlFe=57)yHDj|$9mp@=qf^*PvdlJbvvq3+oynaTR%pYcXLW}kic%B zfIyQ+@fXQi*_%Q4TwGD%2cc`U1I0z%<*HavPv`2<*vU!qh1FxH2`oBeJkJ44`x-0^ zD+&yv5U<8y`FLc&EA?!<2kcu_JNQNhEli&e zjK|e}w}4B^ks#nL8cq%dk8FVQh%)?3ap`!W=&f-5D+O&rj`}CUP@YgcMovyb=^0hR>j6*LA5y+vS2{}Y#iWrTtMM>-q~Or#wQ!oSL>K%alE1nB*T z=ATdaFA*?E&^KJ@?fnhz-=&c{zrp`Ij-Un2gAv!1l$VD-H7(q%teo6!ojn*gEDfLu zC@yjz++kn{=>FWW@*1>f(E1nbwDdgml$8W5oE_QBES=4**t{KG{?r2_K{csJ_^(5DXURQI=fj>@vw2QanOjMQc+O}xmj8Z zXh_NYiyZnUOk?Ze;Ud7!4gdhy09%zV0HI#@-Xveb#j06 z&r1GXkCc_Wg`1s=hn=$%)t`FJ%$+?wglTC0H1wbApL|+*+x<^VPVWEe7PN!xe{$G4 z**Mt$Q#X`U=ufJEnw__m!v`rlN9dSA+YsU8wlX3FH*h#BIW0M_ur)d zmG%E7)poaXlXP~3HtHerKNI#Z;{VS47oiaQpT7UsSp0LK|B(tEX%SQ*_W#V92x=P^ zs~mJ9$?T+5|3Cu@vp*NyA6PQ}^9hZ^{d~6SEcgZkBL*WcC9dTSdz_7w_PU=S9GqWD zB-SUUqdxQvSL=(V`q!8cT6Kii*y6Qv<6m%SP|=5kP`%y9GtSnx zFFnbg6r65z;DbDPl}LJeUYl^Q3VR-|eZJoV0ohSw5#)mYHhf_WW<&+a*YD~%;8MZ< zWl#?4;U$GbNr3sgAohh`Y(r39ml91Cj#lh1!$zj#Uw1-|N*?dqS$1K0d^p43_5rs! zzXvkGpi+(a_k*QD6+GIsy1S^Yd3%FS$3^qkZMi5?Euw&ojS~z$Daroj0Km~EoFOP# zy0Ck0N^P8+yqgLQ%W*Qv7`I_ua##gGB{J)#~dp{v{5e#hMTp{oCmJ zJ^r4rs4p<#V5twREU*6#Jy4JZ8Nj{nT>WA$@Yj3=OUVcI0P-sc2tNPqN|5`>3hlm= zJSX{I^Cgx~3k9~i4jP*DUr*|K3QV|jQ(;=%U-Lz!jtd1hV|jS^{|(@RdYTM0{%^p> z2*r!a*V!J4#J^nDFeuG*VjIh>OvP{hdLhVth2n)_*xoGf?_s$D?fyy1>*>Gei&g=O z7vYMOH}1S6-NfZP$K57Eqq9ajyt}_>`AKvVjvEwO1%t}|R~B&Nes-169)dv9&ydJp zKwz}tt!(y#c~5q(4Rk8x#NgCv$|`YM-p2l#91;?93LG{|$->2?);uEb2f?`CnWY zhR}s8ANI5V{i1Yb7jx8Co`cs(Jx#Kn4TYxe@ot}mGD~ja;)U>b|5ve%4S;(QWL5_T z7D!~|#y->tg&@dL?O-?G%d~h)s^FtO4SaPUJNGYwP7IvV#`2ql(TM+L1$=oM3=B*v z746h0+FMzbZ-Cm$aOL#i)ZNaD+iB)4HfRwhStXRI)bG$Bv%hcuJse-{mz{Gr!dH6e=Jz^Ea zL=HaR<7S7&+FroYaTD)y|A}?cb9us zL$b`CwKUQc34(QPAFB)+Pt_c^n+p~2>1My{=;W43WDhri$f9egXTOUu=^E`k>zUBc z^?aW_+8Iq3y#oA@+d%U7YquXkBM78@c>;)E?2OU1xR?8=;ZC+U=hB+GO6H)RID(#? zcP`mGzdSu)$rURB_8(jWJylhJ?yW(CNNszFLcf-9N@Mlpn@f?uWfydij-;J?F{wYA z)g2yQF2f^Pf!9Xfj8pnFf3h6g4WYK^7@&pzQV8S0xP4PJR?d$cSd(Hi9eDdHr$IZRf$t<7S+%3kz9^orRR&TqY zviYGO7ZS_#;WonbLhT$?f8Dp#)-3($ly)A5K`DzK9uZ^yphSs`)k-1NBcP%yjFQ!e z>4))2DZp{D+VGvt_`c%YRS{3E@0f4L+IMM9yZOc<=iKF29y7DP#fd5+drxKN(LPR> z$h@nCOk73IFVy#EP#AsX2P>oSmVb(i4(s-%^A&nAX9nFx3u|btWuM163(xe3zC6Cc zBCKNU$`v_LY_-|*mCt>)2Kh>`N%hcZ@dF=D3SavaW~uO8jfrT~$t0^q?munODDc>J)x*!^kTS#R91 zljReQI=oux$ZR*eDQ)tMQnTLwL`yw1{=sd+!%!J zL=7_EV_qKm9gIxs-(HjlTxd1yQ|{=R1T^?R4-vGS3Os+R#|}625D}H6_=bD@;%1YD z6!CODZ}PjruZswtm5MQroNH!`$)c!xJuTIsQPX^WzUIaZ{*+k~@j7wvZOV09E@>#< zPtz3phQkM3Phdf)(96n>Cn%1CFr;Jez1{ui3L{G4GyKpGOZ8s7->l~R+_y1i2YPDC zD_a#XkE1EpKX0{N@3V7(p2TXc#%M_AjVQ=otE@%crS+XvT!maOZ_y8}=Q)>kvvD4A za@N`Wk%s>YR4V`a$V5$3=iv%UJ10GN++B{DxBEXsK)~NS4fBt?0VkgGL#zC5zl0&H z43M2+y7W5B^Gr4-$CZ|VHgwAOa)d4VV;LGvX1#@4k3m))E*l|NhfCVeI$hUhp9m(K z?WThclD6VNs@Y$3Gq|Upav(dSOdagA?&e){(RQ`bhCp3&cI$jr%5dlS@APzNAhBED zCXtJ^2X{}R^Rc7Ue(bDJY8hERfj;LB?obC%y9OW(L6OqlgaZeWf|FbjGMRnCRT7sP z@@m!}#kJm)JjiwQ?f4KDg#b%Ku8+$2qpMD9*OWR5cPZYUP&*$LF5-BcUs$$DlW*cj z=iPqA6vd;)Yzsrga4->{hg1-5Q7(4-!i-OA1o|=PjdcjjN21qooP=}bel~V99QuW* z2`jU^V&n}_2;RQ#9MLMJv6-O?+Dir;GP7p(w;HpbGSs|NUdI5w1gQ8OZTP)XOxhLL z2}qZGXVosL@?DMss-5?~U0MN$D8&5!K^>VRY{&}} zs_6dB^w+c0@k=#OS1{s<_KvcLvaI z0)u-1k6gRN|H0oEzF@RHy2F_r##CcNjP`bF|A)HUoBIfNsos*@8^9sa@|jhn!wpfV~Hcr&&&aG$wss&HagTie-jHH@;y{R&~ zfy2_7jEFVC^(X{+Z((G>|JJyTQ?BTCAA3bDw;>;iY0jIsZsNEsO~Cqr{%l~StPhvH zeLplR!?^wrQYl$9^koTtcA!Jv0|^eE`b=_AT7F2DoL+3SOW%LRI1(+8W@uZZZtQ$` zj{FT=S7S}H9{#F6S04P<%MaCa0>e|ddr7X;Li2Z^!7UGDyeIrL70X7|J=|<`A?lsw z!afm1rQ|F&rv26!kr}(k$VLmE+18c8R8$m1%qxK%)vcH*;@)^g%AwT+>%9dzo|Bh} z%lTGJ!2exRZg6(%;o*(wK0xTPTIwqMIiOYkKvuDj@nZg=GqG~@9#w*|mm?PKRQf#t z;U@u3Ds~(hSmQRQSI7Sw&s58f~d zypz5OyomB@?~e1G5N6OOqbUOi(N@DC<)0Wd9jfgWK}Tpw{5ZH-e*K*}zUy(R`dfiH zftHh-<(+vv1_bKW%m!O(D`uUTO!^J*sp0tHkeTUxzrgLtvB{{+eBi0wu6C&Z2psp|DkK3_f@hF7 z6>oX+o0xVxIit6aIMK$wgC>}%{R^n9hIMf~Jj};;cNBMTSu;Caw+*%|`$}DX@!B=l zi*U=N-}ZzF0824{stqzkKK%n`%%bY*DQpIGfR(^Zhs9=^ZL(a^r>kfo{ZsiTiR+o9 zTQqHyrpkm21M9EFz+T>NX>!MCcyYy}Zm=z67P)1YMIfVG*2NUQ^f*>WR_e$oSB z-p*yYpV12Df({$YD}t^l%&B8Z4_{LQIeOzWYVp4x?$rf=uNeH1Mj{_7SzsUTh$(v@ zPvMFO$XK?j1+txG;nQENaZzu=ms5VAGPs8 zeq$N9>caeQ_`skL!Ks%tMj77DSI9gPk1j8d3w285EcXENXrv26QBzOqDLY?`eZQXo z%RruIv^noT>SX1|@n3GWHX*F9oPD`4d@;C&(D*+X?>q(Uw}EL@JN*1(CBFx zv^NCT`{{4U%QQy%uljNQnDDIbpp&k}dQR;Rx>sXk;^y}4wAUQ%WAR^>xT)=@lsjGF zSFU*q>{%u?1|es+nI9N-*-dWTk2QgYaOg@)I_Vp7da>LIdYNZ)U*ValWZh~-JDH%u^{)_G=+d-mI|IF9QlqNc)-e|uC$50U-g&&+Gx?f&q70`hMAYuW@j~5b zv1VD)G@Z@`{53p#G;{sL_~EPU>)q2%hR7FnA!Ir@c4MLi>-H_Lju_yLNYVFdFhwMo2z#pcAT-iNZUxiNWG)1lM2O+OWip1^Q zh^@DoD7n~*7yYO@3*Gqql9Wd4KhwwXGOp4SvklWF-4*P(+s*WBlU>G|N$JhrBAw3< zKb9T-7=N!2pPb}s@fi77g@M~)y+#m|udxJ@mec|&ib z0rLG8r{C5guv`mXS42yDL+0L_qjUYK{?*0adH9wqvp-nVI6SgbG`}-a*3-v*6;K5# z!#oRExg4^&A7wQF%sl?=2}|lcX*8fSSzGb#T7E#mjhsdtd!jXJt-n}@e|61&Ua&Pi z(W!P_ty*h6#&S0UA%>Uj;anj+{9>MLDyz032X&tl8Rc+7Y*I(4_)`_t{AfLyBW zI#RTfr_-el+2_Yg_Msf%^6lmawg8eDw0xDH*kyfb@@N~Ur)}Vw%x=e$Vfcv}KX8jd zB&B$Cgz3W07ROBLXxBI z_mg>Gt5{sh^Hs(VwR#m6R1!`L_1ZtN?ly>lZCe6aR><{EqM}=hx0PBoSyzWs{e<`o zxEpOM9rTBS4>f_2`tQN3VyQ(b}^AO_~|}pm$+7!b$4eVtLN(?*_Ls+ z_WEEM_(>__ou(*Zqm%xIfP;TCn14{aZQv4_;J9PSWD(x+QsP!-Z`{VOE1|0EnSQ(O zLSdGDX|VGNbO@p=%h!-kVM#BvO_YEY-g^>eJ|C}r*izkQxOT#uw^8B)(>aAZ3sp6O z7wBXIZ{=6JribAV0Vp{Iw}I|)U@_L*W%+;W@?V64fvWN|=@Y^oF*YA(T#`-veFEBW zI@N!^q6}oLOJ2t~TI}lD*4U`Zj(~tvMW6MQC%m%m&YIOyyv0t$pXj45C@d3)CYFl*hfTiAsClDJx;5-t!Fx4?INM# zf2sdI0qZ4m?HAZ;6K_1wzLWmWCa!_}(xTVkwuBP+Rk#f>!aqxSepNSd=&?%~XcBOD zw-+J~+kEOboAoMjj zr$f4t?pOV1ZqI9kaz!;M9!r2PMo*pos(z87KG}+EtAI7nRDa-Yb!YL=p`Kpw16X(Z zP#Utg%rfRV(6FYwYBiE$6iF)!ngoCXnW0=1%^9e%FW4AVzXKI5>3L~8r40$HwTY2T zBdRb?+|DQ2L<7K2WG(TZdANv3Cwx;@Xxt`72%ZmBfW2PF~g_r7I!xYUh{h^}@$ z3pf{|vx^^&(q(+-vmhl3i9J))#5LU>$a#34Lp_ip&D0!$7!If31629qdSR>voeE2U z!@Pd(I?7yp=CA4@eF=IYwj9(mYrKvZ%lfxYB|?E?kLmP4k9I7InPrBx^phJKTJ&T8 zN7iL>$VVNv1euAh3h6x5oe^Y#3wUB<5{2X+71ML&1_o~1&5D3CngFT|Qg+)jjef%~ z-jhoaYUjvk_Gr2{fpy2P&^&(L>)5w7zY&%vD=m0E$NTlg78)Ft)q+Xf{Vd{~KP@1;J7J;n7xe|7gO!qr-=ec2crGVY z>L)W4s#eWEq2fOa*t@o8e(Zg$s_L2=WO$cxX^vpmCE{$$De(2I!)%LynCHSVcyi;t z2QC0uQ1Dg5JekwaG^ho7HtjO$H=CUP|0*F?(^QN@0RZSY%8aDA+V&pXfL zv7>cd253713}W~jTtNr39d$pC1S+l3K*+=$1{k~D5Yo^rFNY8K_0Kp@u0nZdg5N@sMJG)~V7g68hcJ3@oQ)Q?F>BAnqv+W?t|0CiqaiNlAwQfw2-#mIiW zh&|5T#=!u`zG_&`oAe7_pU&WVyQf;A<#lOc#z9(zB2 zJFA`P$=6~hi#P3WF1bX)!1&Db1Gp}4a5{5+ZeBfg$l5-zlBTo}n8CYt$1`~+f_2UF zra6cTG$VB_Fode7UkOq`3ZDbo4!VbFcy&>!1ZWIdJ#Mbk2r zhxxKHzWW=>x8SiH(yb1x35AtR6T8Lta~<#UjtE2P@5}GTO2)iRWDRRtR(wAq9^{{5 zJ+DDVpO7G`Gk{F5s-y1e8)tMO3l3@6{APrA>}!f4tFlu!vRzMchouj~GB{^`D4`rS z6Y1=sk)1_?d|1a1d@>}J|6~0Y)1Y7XOuNJa0Zn$tbI9RQq;!ddy0Zx=jX_YEA0|jS z+1A9S(IFFC_*t?Vl-CofN+zhI{K1DuriL0pw@O>_dzmO#^O{Pv`-2y;pj%;sX&W*Fju?iNR*5=d}Db@Mr zB#FW(9|&r{y%@1StQ@>DySbSc@cf;B^SMazTI9-IS=-YJJ^vn??)l~7a>6?4ISh|v z_|d=4+O2~U2@`q88H2P~+U7ybMcnpu_lJJ}Zb?)gFNiczAkYM1G@d*n%gZ(iqlxDSM5xmDR{0SsDbyer-X?ca8h z*p1(Lfja)k>V`|N4zh`j@hOuK5jJEHcq}GIVoZ6_1wF2xbK}#l(>RVi2`VUYvM%mg z{}z*>uZ3wm8Mo8*xgMJ!1>!NuEeV?MZMxJ>ApfmpZ-HU4rr=n#NVw_@=>O&Tnq^r4A|6dIb?H zV#LGj3lUg>yRXn@_yvA#sAGh3gxmzDHcCczL88nZ^}A@RIxjS0%ojWlE5Jl3@D&TA zh&M-Uo;m024mlzLGJu@kBNV%Yxdjj~SM%~xsJ!E7wDEYc_OPXw)4PAc0)OOtH;JT0 zRJ@n0BURoAsKM+fVtb;aItE2P@bNtIW@*63)o`z};~GuAt?@_w?Z@HJ2`mxA^@mDx za`u8xYcNv?A+YG~U&hnksrZeUkzbBwV4s5m{lw<61im@ASBww-O~L*vqXximk8YM)=~m_S75DhcTVS*y+G```> zf*5$jtoF$({GEn2&1j5220E*eK7V!e85t#h>keMowf(h;ioH~z)aD*lnq+W8RAL!_mM2F1}b|9mG+d~@&2R%0oqAig+1 z4#e<9n$sD13v=n0GppC8-2|PSlL(;RJ!9KEO{CAigR;*A*!3&2tY>d#Q;Qr9G&Hn@ zP0Th>K?5JjCVu(Nz~FidQflQ)+^VC8?s4x@07PEkXB#e!dF@9L?u;5#(Px&ba-wC# zf@SY;xmi&WJ;hx{GU)1VXdbnEA~MDKZF{HBvl|hM@|1QTUMp3L3-I>vA}Wy!dl_O7 zEPVs-mCVL-cn9{n1cv2;>g_Vy*M7tocJXK5guSR>6o&#TYMd0Pu(;2L3%PUuA>N(c zDIU0RH%Q?-iRJ@MoKX`*(+qC)77*3wWb$R5cCHMGy3v<{Id*C z{cl^Y6TCgth!Os`-S3TTZlG$nREL=weGWBIm}fMJw}#RxkmFTYjoj%6dMw^tY;3_U z3gIAxjSbEO%=r)|mb2pu1mn~rXUBt?o|G9mh>MpIzRR&fjZsrtr-4W_LU>l z*PAQeUJsWDybnFXC0$!9s2~Gg=ai-h%-Ij`y`O!2YVEa6yiAx?S7AGR4>|=uRq>y; z$iAcqd~5ff_%0%{uw|dKzejc{MJePWMrGTRj*M8t*OuXY|7)YSwTtvpv&x`B0jwyu z@qj~T+#5}jq9T167MuA_8rY-Rnu(jdksSwFZFAdBxN2I-AMt-&;LP65@dGF4D8vl- zb#bDT5iOL&pKkSfc~|!?`+-YxhrtSH!Di#<1XZ2(n6B4R!HCH2nAz|Id)4`tRBg+hj}*jfOvxSf@cPt4*mwyUr@P#?qs0lHf=_ zhxYQ`AAmQssa0C?H)W0RPU&^lvyMAWChU;1_HT&TTK=0T0c)mKY$b^7Y2rBa&cGhX zmse<(bMHI`r$s^H|=(b$G0O&&!Ez;?#=_i><5ld+U}o9sv^DtT^}d z1#~f4x2DY|oQ^Rj+=5%>jpZ4Yuv}wQkjfOPbY^Z-)Df<+{01pS6n23rcp~HVo5CM0 zFrT>hm82KU!esmb1A^A6r$lA?b)5vku65w)OV5*FO9?=)yZdB9qQ0DX$$d#1@4%TQ z27M30a-_;JYO?*line!3cfspC;FKMFsI1b+9Qjo!i8(4u!dXBaYT}pO`}eJHX2Pzd zy^l8D4BF=rZnE_>_D1{1TpaUIkKnZ8Q4DADI8Wb6Y6Ul&jh<$?e-6J9L0&?p6O=*; zMJ~b!wGF{e+fN9e{pTTsBzq>%n1jRT)qTC-sV>y!$!H}NJM+CL31AEPd}DkKV3Ej| z6p~^5rF@iPk7XHlAAX0|8Hq{xLr8`G@ab&4R}8`Xc5?&0Y7GUiWEgtB+n~ndb)^>U z@-b7FB?6POK-gqbhZD3r-B6CQgmk<=WufpIXcQUJ5AU(j{()2I{6xTWuE}UlY@~Va zi`Cl^_AWAv`^j3t3C>cO?COEZvbd+}z1;Gu4r6dTSEjLSW=?54afvcewdw)STuIiE z150(UA#Y`NR;VZ!NFm&VZEkduG3%dW2`ab4%4qUd_F1D)heQLf6f-q%2VSFUe-|$A z=kO+#7zN(u(X=KJ!@caX3Ko8ZbY{dxO@3qVxrdl0{`UQ;NTtrd@Px1^wS*X}7U@UX z;}{Ar6g8z|ihTL_8Pz?y5f#?@`tu#9Xs;M@GhjJYm(`+R#LvWw^irVw}`LLAtF2U7BS?{LZ`Z>M_jj1h+%QocaXQ?DT5 z1Sin~i%E@siMT;9$9o5t(U$L{5T*pv0xpf06=V?O8$?J4w_5@M$dv`L$x}CU^(-#L zk7S18HCq-qm< z{aEZr(a)Tp7#7aG$7z=2!GV%j@%l=dp=w8^ToyNVM34|N0H~<9+hX{&yC{#{n+}_F z{}q4pdpK`z!CbRdzngi8G)VHgXL4k z9^L6vfS+fwK6m+T{?2-55Wq-inHrm2C)EGg%l0&PzWhMqjP%U{7vg-Ab-%={h=qtn zyD$4WfU+NgJmF7-$mDtQm=Op;a9>H~q%vF2d^5vsR8U;PSJEZ;x|5sj`9Gdia`XuK zDko}}ezJlt(!TNmt)7=hvyOd^z#iW5MdhfH0cN4!@7BGWdLWr^!)_Sf#_(4tLQXZ# zx{Cz)M|d3T zU7|rI-c4EWow6znI`kTILdYD;ES_PrPe6~`uL9I2rFSIntcyJaekH_lCs z*8QmIrFz}^jgHK8$$Nb0YtZd~<;=(vB324n>Tnsa;XUCH8E9ipP{lFKa>f&u?p>4G z$$GJ37H_3Ni4PU|($sFQFADua;2px+}^p#1osk)T!=9= znc<-?)8YcH!V=0+r}TxKxGT||qOaZV#7 z{{nZYViymIa{2f|@RIjHf|KI_E?|iqqBtvbcD5fX%h~>B<3|`rvUgbxhB;E2=~t6- z0_7s3<&;i-xWsX2PpGq~ZF@&%?DjqUYW2xaijd+-tAlK+IN9|U+!-IDRf62RA4>fD z?e*K+yG#Q=Z~%&#Bx1%F#I83K$xLB4XgPCf(rDn|`v^+9lPK0UY@?B39d5URyF3=( ze)ukNVX@}BjJL8K%%ffGk1q^<*W)M|)HvobgNWCfHE#`!%pOc{-{{wGG}q3RjygN>B<-8Q#P-Xw2D}#p z7me3H&rLmNl*QT4ba8r@rf6fGuB}d6Ze6e3%5U`5BnBT0-XNk# zm_F2rIU-DiY{u*SrWgr^`muNIvMvn|Yai9iwaPb`TngXRK5YuZ#=AXy*lH(mCtslW zRj@qT75)CUk{E~bC=Y72rbe}(LJb3=iv)jytHFW~&C>ay6Tmu zp9+q2SZh^J$_9B6`e#Q7`*2-0v>z>J1@7C=WQ=RBdiCamVGXgHOqaDk`F`4W8Vd|G zAI9~|Km_`%6ib=fkl!g1IvCLQ@Lm{2z@m`19jRbcY_(lM&jha45kJj;>M_gVUjR=i z{qk5T-h!yre~qFZW$Mb4Zi%>VncddsuZ z`-GD(d>X7m%4_yHO}zZJl47B0Fg@((e_V+ZEK#JM8s(FP1%BJm65{+?CiRKH?HqfS z8+WGshz7#0GcYjgebIT6ADr=DMZW7$45y70XCcdz(!L~huFOy**(kvI?VpD0$Z1mV zct*lw)y&Fk$|6AD94LJx0Gz6FK5du*}cnak-9p%AkZ z0YC~3mdV`PYK)AZiiZLmHvGG36lDl@NwVd|%CFW&(=`1T(m3T#xm-jl`D`@TPFl%zWXf`)DBoAoIMP_nf@ng&)Og7 zw5F+K1d|qgA~cPYy4offThj8rwiTKcyDEJR zR#ksc*c%}k6yEg5r_X{=HHH@zL3T8eB6M+7` zCC3E+wuO`g3leVURQl!ho?ty8{qMyHe$S~p1l41F#R6g0lPA51UudvoJ%|XHe`J^@ zOFe|U;w9!vC+FfBorbYbO^BuI*}tvIsoi`XoOQ#~wzY-~^3qfDetCl$Av=@*FMKK!xC~5f$`4rofm$_!GJ&k`No}YF@9Sh3HAECnuQ)nOS>q{u` zZ3@Q`y5azZj@UcZ(oyk%ic6GMFXHps+>)PfmL7=R=bC`5yFcLt9lu#^!&V(cB^4jR zqI(G-D`WdnBP?FJE!$;5W(7~1jYQ#7%3$$qr+c-KP5ofv16Wev8xmjH!@S-dig<+= z#VL)@?qN1f-}R&4w`21nH}eJ!%lK1G_oyL~&^7kqQto0tSJ;)9(i8cS>TP2e=!XIK zygKZ$g(Hs)Ox*}}!q>S7#YJ=>n&FgukkS?1x z0tEBB3l&p&d?W9jJ5`?%NKI%O)(5th(P^sVd*9`d^IsCozL_NDE-mN?$+p9e&409* zy|zXlc9S7>EpD1W*(>P~p1g?rQ4X&{iMvmyd4=V2akxEsy+PXb-|@$uW3cb zup}jfsI60pM9Dey-B11Y8L6}Jy$KgegQmbFdjMA7ufdb^U(@wQ=rTWNn2R;L0|^AB zUE5etJSTei2V=4sB_m?z$mu?XSYY|WMr&9xAC(1XsV6J>bofrlJM=VV+7xM$LHFa1 zD}_aIAt4~yWi3*)t3Zpxq38Dy)usue628+Nzo<)*&j)rCe)+|0m6w3@0E~UAUKmp) zL8*;p`d+JlJMRW>i%}gm#pW2wP&qqWI)$BxQGmQi*ikLeKzSFT%^30STRSmVXJ1$A z{F;RHnyWGKPT1oy-}*!1W#epXaLhRza&SlY7`w?3sh{FvYO=kLCug9_)V$-Q*aT-K zn$6RL?S@1;zvuTHMZYf>r#$(M$x_9o9GlFktkan?U6*l~HgU2 z<+vXcg4T#5fn7O#)IYTMT^~hQwCYPeyq{`Sfp#`LQ2PLgO9g zrsH_~)2Gc)%TqgmQpa4hU6neadwP|}d$KzT6OK&i(X2ZRj9WxGZ+3-KOPbdPPfX$T zih`xAce$4o-yC(6nV^{L`fzS8?-D-Q0@Ufeje?I!a5I&^C<3-@Q^~$2XJiL1#||<` zsIqAq-!xJy`9Qtdu8O&t3HPxpZ=S_QOp(Cj@!E3ZaYnv};*4i&ve`@40`~tDHU?sH z25p@uLSwb8hvD~Y85ULMe0~qzTz*6#qJn?0MGwYzaYhxcG=Y zMO92~d6*!cYSQ^gk6DZ3`B|KwS*KDxuc)aHx0{m@Iiw{0x!ckTq)V-ar5TDtz@q=% zbV)zh@I*7(DjZqW z)suGQ4Ru3)=QXMZl)nn1$D8U!o(jafU))!{m3g-{h6?Li*IGhagZZ#djiG9X4!6|e z8H@RH9DaroKX2G_+RJ2~e%=~^-Q*K+g(Z;miv!h56!~?_&p4jVtS{Z$@m}+dMpyCt zVK_~14Aa0#n$?_x8LGTJO`^_T?*6BbSVzH(X&Eq~I9U;Azxx$Lih19V3HH>saX;K0 z?56e6_-EFduO0MLvg*QJTa)K6kIqIsVR!gJj%l=WTgi41m-j}E7_pB{pAiQR$i<)g zcWwbQ>$%VHkq0Xr98*g87NXId(lIDCU?E3;=%M@GI}jZZUMYbfpRdo=qjBBOwhPUo zOBc)^C4cMnP~ovQfma*o7;mE#9>zb&*$^54l6ku2(%f^(IZs=7FRX^Qz z#sV}sUN&r_sTFR58|xT=hH&7w@iZe{yUxDMyVQeDH)!xI-iKH;ntFM`QrJ_(G>hZr zDj~bPT^7bTOFVLx6f#}1W8WNRIzQB0<;NZss{hitzAx~+t^QBfavmLL40){t{$e89 ztmq9MKUn1h`_>8mw)y?2NytXsEFFNFJvLkTqWqIRC6D~-XgPM|CTZjTNg$fQ;#B-o zrmK_k2r}<1RsT~Hvy7uvN_lX6rB7r)EJ!C=YROZZxW!rgt_muDwkdyBAG!-*A)^@l zXD?N?miQ_ggwb;J4bQ@)t|PXlUa+#Z`!Vh`@OlNcm7xiqXANuDt18v9ceBdVONX>w zRi&~YHV22Vrf%_cd(1)zpq_gdueQeC<<+`l9iKE+MKv**+0)UwOiec}7g~ukgn)|T z-kPC^Wc=Plt!;S%xRIPbSk1gMCejF%$GaW3zQ|UdJd3vEe34r9Y4c;JDgrc5OerSYa(etCMzCD77-1A->W zhR1G53qe)6IK6;W(4C&+B8V-xRWQQOagSBeCgi)hCCy0JxaJVWli)aHst*YoL%_wS z2CIHiJ8T#-M8H){_F20W9^?p}e>4q2BCFqy;l!zk+e`+oRZMsRxx=3#NEXzC&hJa;Dn^wCeAEbq*WKwK-$thbpoZ zsb5#*8$2p~*pCgL4XK-?6uDuTd$uQhgZBQR<;t3A3Q&i!G5SWx^a*+0y}f&U)pXL7 z)z>t>h$`n0(Fs`8iKjn~luSWfXkHa)&@a467OZMFC~n@<8`dm0tDLWPqySD zwji`0gC=`KM6A4^cG<~mx=L>1b=gfJz|BU!i7nmaZ{kC8$Kv{%}bg{hXNv zKFiiUGZ|@p_|*-sO7*VbPuVIYaqd=G>E#6_)f-X|5emrDKBqwO@cl3*^ZBwUtJ}XD zXJgi6YvRL6dko?dUuK#O0?SXZ>H06N=c?6JFMm+Ib<68Hw+e(<`n377u(gZ8K0>Hbm*Vwjoh4 zsb`wXF4|GvfF+iHoknXY#)|>Iik;jG3c8>dQDQl&%1rU!Lh7_GH1PyHPP_zrQe$y#Z z)KHsA0?m?ZH^yi7y(F%WKHKerEO=B`)*+*7Qkzs640om}#c!Y1QwJWt6gXWu8DPKm z_yr_k@R(RS@i19S3XxGbXN%pjH|^6QV1tm=t_M7qrE6@JigKM{Z%zu|9(AYQR{N=R zcJeWB;1|s05%3)W@8gZf7{u7YLVS<}!V{{)-1Mq*pc`p~<38xKSt-WyRdDUGam!S=nU z5}DHOX!kT%>gFTjJ z_T{4XyU9b*iKa^9G;O}ktOnnV<+AGwb#qSC?|AG=5DeJ*P|oCo)6G$vBe5+OmoAsp zrdyZ3jp1sH(HN&)9xRx_(P!>v)16hRC5n;8lI0L0D1euK942LN7M#+z&+I#EJy+u% zUZ<6PaRy!Rm@R)Q(q;#9WK4jYxWLFoeP`#nI#<1nIO@@7n*7LooQGUoOvoKK3K)x+>fp^ z_44~W1Xgx(V8}6&O@%+3i0{;@QJWjK=J^6K*;g=74?6GDT!SW5HaVT4y{Stbi3_ZViH(l+J zu@(-A^H@Z}6NFWgH4XaWJHo%{%+@oErlD@UqZiT(wbe-Mw*UAx1_31f!)7_%;XFMfU=JetOL|k@-Y{9lM2S7!Yb7(9wn`l&mHJNu^OwI*kfdCr)LfbSkyliMp zsIRN;p$YTxQ>{n)q9qXvivSHbe}X{ytlKv8(9?>#6CuV~irERWf`>g9>S0BLQ|Fct z=25+Yyy8tL?(JB0Q#33xclZT01A9*TevilViB^wKA1_nfx5UR3OR2FL+B(eilLOZ< zj2!?xIqddXs@fM6Mf!%H{Z5RBFl-`uA_8$-@nwsHm|5Yo>}Y|FX}f@h?I<*Yyxye$ z&ccYib^3s1c+X?P6~JztulzcVJCP47!WJ_PqFnO|JN+bXQ>M?c2*yLud0kI{S!T9b zUKvzz=cm&P%Z3@b@ik=a<{ot3yy0#b4t}u*y;L`7=JUP?`H*Pyc8D4y30cz$P#Jh@ zf)s@0K6m_5p$hxGoJ7Z7iCQOi$FVcJ062cbSupV7$_O3QgC!6MyG=6SVNNv|b(_|HmV0iahi&y{kwPe<|z*gW5L}+)<4cHKb)*&xpHY;y#l8+OR(4 zPS9MB2cmI?YM{-5JtGsZ@ardiK;}3hrjv%y8<;--8;I;s?Ys{7+Hn@D!iwpxfjcHo zCFrjt2%6?PCGV$qE4`a?j@Rnm`gR9Jk*s~T#S``rn9;SiG`*hKYcu#bU`-O`~%T6;t zG5Pi=Tt6-)gdjY3FXSeQPi!7<&Vg+N4WKKsr1A(C^qh>#*z2fjap>^NrNfrae&N_d z^F0>`HfqhD_+?*|0X@$c_>t7ypIqa=RqEUo2g=DgAB4LSIzn3J!$f+{TaXucEWXP$ z9*xcz2lGkK-_AY}KtHS>$T$}3PmP@h8dq*0(m|?)HT7Cqcc9c&7)|waTJq*M-7-im zm7s4H{5^ftdpdzxM5QtWG0y(7mq|cD7_HVa2Jvjt48h@&IY~(BT1aG|X0R=2|76@Q zY|qT!^TN%*mu*gbMmolaTj8b-yIKcxn|h*RJi}BQ4pm8EU!SF|=n>+>XjpKzXOeNL z;zRGFKFrDy0rM>yA5pCAf5ck7ovmzDhKJK7oTQ*Bp!N}8nzw~qY>;r>7X4%)uWX&Q zwnQWdcikn#!4Cj(*-Q{MoVw_M)qnOygzk~N=?CdEd11PME!)62hxV+FGW6U>7Lnu- z*7L*F;l7jBhHlk33&w>BAgvZxi~I~g8Ly<`6&}(#s!NZNSpYqKWaAUP!kWR0)irL# zo|9qPoHgT3;BRJg+A>(lk)%_>MZ73m=3vdnDJ7iMj@u!Qj2?aBrbpU-4X z`*>8u5@I!}f&~=5r(-YRneA0_cm?))2JlK_SoE{ob&6HK5;Hte64$87YG`#qoms1| z75}||;xT(U8X~H&@<5K~iCIORDvjibdPR!o?n3@Lez$VCN2MYFT*62)LNJ@Ky zI7}RkO6+CzR}$~yui5WrVm=4f6zIZdy?$E0$Cr77&d0wP+V1&`!BrAZu(C=rITdfY zw#972Bou4D65ah=#io85Hp!chrn1WVbj#$9uIwINZBHFU` zPtJjs;xzc7VfWPr_cMnto~>g$;5JvjxGaDYFaU<>X%MbtdUX8V_B^k%pYN9)?Fqd~ z_N|lAx6LuFR0;MTIx7{b0LoQJ+8hLPD$y6&qK&{ppbE$toBVW-y}=dUL~hvIt!BjL zE86Ci^(R7PllbiPlU&xh*w^i|tw`=8njs5Xn#D(6;hx*0tUpXWGh{Jgy$|U|kP%^R zj5OKs?qn`QCVnn(kH_AE$&jAI37pu~R6{d+gp%K8;?iY3I!3p_3@m!;d+na}Eajkk zZ@4u8XaAx!U+Y(z z%r;0S99-}@S3B{Re^k#DDeOp(>=0A31ZrF?-pfmXp1=sI9AJpgO?dW2=fEwyeTp3I zr=p34WttrwbtxyT%@ZmB7}b5suE@zUg?2#@`Q6#OWZHI-6bE6RI_f(2H#6=nQhmsr z?0*|fW< zQZ?rW!9V6b>Syn^VdS`M*VB!+*t=%$zR2oz1d}hda`H{}WFSDLx+P<754ZgeD`st- zdI??@#;h9;U7W+2hGFKQsiiVgf+W z4ad5rYM*N8C6HCR{f6bz8ok_Z!&O94DQOpyk(Ip)^Q-X*HZMghlKvX{8#^CKtHx>! zpN+W)E+OQ3A;Uw>XJ7Z@9m$(kBe-xxFL%T}Vh(z-Wi(v6lA_YKxw$pdDm$k_ej4k_e+kLqwG~ti?~n#nccU<+;?2Z4KPi|wYY+q$ob7Ncm3|z+%lVB z049(S0;=3P4W|!CFyeo$X@Oc9spn46dWXVj@a&(D@wBjV_UnEgia7J-K3`4A*S;F= zao1v)kgYqlp}^7C#(gHg@*qs)R3T#aAIw#ez+#^`wG zZ5@Zz6YVcYA-9&TjB~3sAmG;#Eu`GBo|6);H|2^^L-0}3>(nH$3cIWbQdc%soSU)_ zeg0XW`~2W=BM_CD&QJHLZfSvd4}NJ=D-3+;K>L7Z(_9^X>9^|K;Tulw<*SuL(&LKZR3B^(> za-2UpCp_LR?M8a-&V zI)BdNa@R#?rtU(>rtvt&mys|*Hc41>q4CDFG%L@fpmmk31OWe%Xl2j83fC7~tLb?} z5R#_g)3jfhp{eT-c<}5}eE&>6$&jm%`%Mkxan7vgPO<_WRIGUQBgx~aMC~&?l=N2F zrN_R@UI9xDVEqCni&qld1(uR{A=hc%iILl4Ops30I#c0K(=NTK)4vvAAl+lE(X z^rRq$1Jc}}?seM2P7+ew?d)wE#I@=rr>U-tVVy$sg3~4pavu4`q-O)2Sa3S4_LV22 zu7Kl+wxIo7-a1i(5aG5@;y3F-ulHv#mxw=|>_< zauO8i#}&}m29%}H&78s-(PBYvg$#27?WVJUFOlCkp!!C=5hEMY;2$vYdcN%qeJr?= zu$ww7YI79EkwJnTf-boBFtWv`^;7!Lv$0jDdL37~I-5GokWene73p2D1FBPlyWIiS zv2BP~zeQ}!&y%e0o9c%qvlC~u*EW-+AzA?sj`}?%fA7+*pD3TzGs=nq5J-s7AuUX$ zwy8Tf9B;dXgh%ng8VR&)((ZE$?z_paPlM4-=&IkwC|pmLoCcYz>!(V>o8>lp9v*&{ zG&tv_6z4O1B-FFaAGa9kXIdvMbV{7r*RNz;D^E&!QFu8Fq0k-4pS<^)VQ8Y9+0WY} z`dGY80y1~JcUg$1_VKvFWaeiiw=rE4RaEf2gtDZe3`^O0!vOxH6##LZ7JF_>&v{xv z_KlbR5`tfNNgMa8EJNb#^;iBp4jqO7IM^;Xxg#}U>#>J08 zb3U=RRmP2GmcJRZdyu`b@6O;Tn}iQfgNWcJM7y)Ljp&P}_d3MsPyMYNV}wo#X_kEz z{+HXa{zbiUxHB#JF>!O{Je+%Cg3|mvPF%c{T z0aSIjL zy*Dw4#;TwOp-zNie{_$~0sU98n$w++*?ScReaYJOG7n+43_Km|SDpLI-qZ`^yhYDw z*rVDy>RS$w-2iS}1Ev|Y1wf5>GkkLhLgC=ZxCMZ|hpT}+dFmzvDak@}u8&FY0p}6L zTbKe72)4`>0dtgp=~|GW$DQuOx$Am=>5t&=WkkpPZ+@-JhBdXZMp5R|idGRGf@NO| z((O*6X@I4$}0p}WK{lF*yM+=5Otf-Yl{n>6In ziB0`RsBk*a9xYoxg*w6Wu+$)C1Rq_d55_OhP=|lBJwYS32-=kmw+vMtKZd|JqHIqr zt!V9t;Ztz6=Vqa^Ax~n#-2<6DKh@3vgKic1)-Y{3Hqu4ki4_p_dv-&NAD<#(ShS(y zU|`qr0E@+!abO33O=-PiIEYmwBBw%6&Apv;lp8ly!MXeE^aBejIw6Yp_jsoSD_=n* zeX`xajNfTA#7wnBo>Z?^G}o|eY)2~7Tx-8+5mg{~OcZtvY9-x`k8lT)ZN=^G#h)+p zJ!A*!tEdR7d`?-t_x1y?^Ez}y4y9NTcRc8Di8U$EWZ{==?|aIsxST4%$TMoN6fadFUuze{QE}F5_Q1@rqd^7O;d{fL2bLs?z zLau93r8h&iu@e2#(Gd@j2f+ZWbf5)0I~ZValgxD>Egi)$gXcBkYziEBprMG?G5o-y z?Pp|OZ`!d|hS-62_vix`76D1tEia%oXMNRKa?u?03{bmJ2+wIi{}8fdj#j&P_L4Jw z>w9}mwRzULneX|xvGUuzAImhmW%{v$7tJ39BiiuGChdQlW_yvDs}JMaRed_oCCS;H z{4{@q<3Y<};)?|XS|r_;bRMp^z5x?V;IrKJwWACEFIcd%y_dd_!fgKs2nMxm(4%kT zuJDstP!bA^?vs{XI%(CamoFxuJ=6$*7Qj}drc}Z zpVQr2`Fi{L64*9!nP}HT3g&px0ZZnLQzO za5c{p_D^sVG!%m0H(8cVe@{0@b$NGg@BUTeuFAt0N82-0WltM}x zgrS}(CbN@D7a80Ng%R8Q-2imtYX96c2=TZur|s}ekaQCx_iUh<{IOj$jkxI=_p7|t zhBMk1aB2aWe*hRg{w8InBiat-2ocK>_3@bMS(fVmh8wHL-CvHp4nEk}D}V$skhEKQ zH4>Sr!Ek=vk6riFE8d7rdhQ6RtEOFOkJYRAD&BJa!ZdV49DL<<4ybX&8B>czGPo_b z)K+5U-~kYgsguhR9ljS}o`5`5a zGW<7xYrSU(O}eM3fp=PkT?)8`A!9Z}>C6GezYlmsKi(TW&i+N;&NYLHvpd)?8++`u zC9cgN{3=a$HQGba=mqwfI(UY<10>|NF{vLEK&D5oj-6CQ-hSG3gdbV(eFAJ((fkM~ z;XPy$QfeoJ&aY6br*|^>cJN=0*PHmJXzj%38NNMoaN<~eGhO43J|A!E(T0Mtl#Gcv z4`>R!?kfEz@~nyflQRz?%alYWcHX?2idnuS@$ zY?^vXe=@8pf$UG36YD;|o5;!mG?|87XEL-LDW2RLCJvT14;`N?CpcIiG}`g>4g#ca zaw~AN18{Q!3_R~;@U^`|LegtVj>E~j0VNnEKw4&Xqk36H2Ia8;MR1V^qq|tAgB69n zt1;LZ&P>hC<>%v>1Za9}&EKq@g^^3Ncc#BnkcohqqhY&CN~Z%f%#2ISK|-MP6Z1kY zbVgb80r%{W9;o9EyUzE5@3*+x7SUV^@;^>AX`B%96pk1>y`Jq;a9*e@0Ng$Zlg9bs zHwsZ)?vgD3&Mhk^7*>pXc0}qzuJs|>hNe( zk7}19aq%crM|ZyH`wiX8z@1@{o9wdlH>nf$_Ei}DAP0DeRBWsvsrBp5H?h`YxyAj-+Y`^^kLFE51>srS6!e=?V8XO><2h z`WuTSAm^A$aD3PT$sw2O8jxFaO&jiM7qTg@FrrQh?;_J-^czyLZ5KL|*KRc{D@orf z^RCM=OT}9c@h;+n5kksNwy(n;eT;Mfl9n@KXuO#(G(|PzCi_U6(6ayAqVEZ!*u`3T zbNzOE{$itz4gJI~is{)O&>27v(G>MY+amu~A^}qIGr#QIC^SZ*LacQ^4L)IXoWuY0 zbE}Sp#GX}8OiAExO5fZ@LNsxpfOQZkg@>@hI5?Zn(G0t>@oLJsx%N~lLiP8;nhn1H zAPHsCTjs;Ceu@7w$8=37Nd?O4@~Q{2!`u!>qzXN4nP$W4dUw+7c+q*RM6&ztpq~V| ze7Of*)f5EpVZR$nejl(ve2SNstZo5~L%r~yCj0U3AA9dK4#xo*LP^C4I*b8~A)=`a z6dzlJ(QQe7!&2@uuMMLho~E_uJJL30;U496dsj;S3^PS&fjv=5=C}6d8 zKE~uK^xX6DmjhpXn?~9`Qfv8C@yoy3{HbHJCHM~1^EoE$c)J(_CE61Ft>;_q{GiZp zzV`DDYN})bl{eCjJ^|d!C;J3~gj-ZP9mf*@RIvseO@?0QEfa9u=|i z&T)oiI*Nyh-Np!fME#%XRvH=@mP+HND$M0{DO?_V(g{rx!hrf+rOSGZLjicSyZU2# zLMZ6R`yQ6Z622GH_EveJNC&CD)epV#JDyZtQL zPalvn3ABwX;o^lT&H03vI7QouhqP$7=lys!0(hq_U{OHxD;@+`)K{ZcgI1d*9}o_q zH_x@y&|cP<)Xx_pkBLS4pZHSC{8O0j2v7jZz-70df=wnb^S;wHw0q_~eTFUjz?s46 zINF8RB)5^Ktqu>+I;E9rH~E63?C4t5!G25U~A%eAC5eIx-%FndGl3aGMBT>N0JL5Xx79#TSn(_N0!pAVdxFUd-#}v|*k7!2VZC9-k$UC+O$Y!cX^2j1X*!O;4{w$= zB>UnSB{!wz&JWgJ!q*%>@)OL(w%*=EWQw>Oh8F@ox{Sdg2l=(L#3KL6Q#U!zzzqs| zvY+DY62FK!hMX2$mt@cIUUqlZb0=JMEs8;ifwF8yx;&dokR%Amqfke`EP?^v)nvBl zq#x$VC1>?F!sPHj3$LL>nt7gMn;K(XX4#C-LlYtz`U;V>ptotxd5=Cnb}sbMl=}dy zLjtn8csV9fjBLI=KiWwD;V$pgwgX=)|A>JePWNQAQ}xEF4kK&+tOEfrv_<|dfU3Gl zva|~z?|P^W-(+H77(yaXW1syL=FZ9JpXbLjgL|5&aRV01jJig)->R=@ACS&oe4Cv7lc;7Z0u@TYo488 zFaa5n6d(1!{O%E;v21?UTYbUb)BvbO$LdEdp4L{d|SlCAJe zn0_6RmkW{N15}=(Q(R%ng}IJWKrENA|9ITz*K}~J*C(U1>A6(K#7_pl`LA%inEoM) z^uGx1XAGJEk3iV>EMg(JX~ssKo?EC$Wu28-=-^XjDe*`KKY2XuPM>9qjJu1>)S1e7b7jNRnHi`8x6Rj)&<>em3 zRe^tdQvVPR__u#3h4mOvw|=N6$er}Zp8rD<=HIs2pKlWw0SWD&)i1$n{{Zp-pOu5- zf-nHtY3ZZ^73M$66aPhqK5#`5koI=@s%J|4@0;U4-AchDz(?w6S|4ivDQffQCI9bV zJQBXs+@^V#G%5GrJnpd^m@;HUhVI#Hd0 z;$PjRfS*@e{OIk=YRAVx^H(h{PAUO#vZ6}5#MnxI)qt1??kq^^b>pAC&Hv_ZfVmF@ zfSEFtehB@uxcOha$9*CUIlz59;`;}7;SU4;k1i{Y>CQ(Hjd#lY7y0&oa-gV>onC|@ zax^kAbh#Y=gaQB8i&N}I<%4--*yt<3XZb$?{``OMp8n^9`N0|Zn$;KP-^b}#nf3o- z=0gHc;4-Vm)`-rTdV5lDUgAI9!hiCncb`is-;MNOo|LHk%Zrt8|L$z~Oa#XKj}Q9a zRw6FpZY06aZ$t0?>H`dXEd?wtU$LC&U!G1<@cQm-aM?Ya`O9Ym<=v*kc^aScZ`b>u z-?F6vJNuLICB);eTAd==-PsVVyu$o{T84jKt^d0u|6eSL&#MoTgR_srZOY?o(PO<+ z%bLGB_Qz}Om71B>OmM*W?iK!9O!z-dUXka4)@^oz?yV@N)@>GPlaqQL*H!+YA8*^> z5HaA&nV13rHs2Oiyzb=Bt}@3@*!o3@ktL(7ukUNGG~LDey96z*8~lCwbL~Sc0Sy-< ztsOl*jQ?f#{@{P0MLwufdf2L18W6)CeQ7pyU@Z?=qEeicu3eMUz^)5?Mpc=lRS?ru zYKi-<1+w+lwR}Opb9C0g*ULX8b9TNd?H|bH|763Mii@SL3QXfo^hC%gF+dboN}It6 z{@wln=02UM?3}1741gt;K0$JVLz)n1toa<`#W=7lYU?05S;+Ag~U6E%=|V@9&?-$zc{;oE8`#uS@88 z56)-FmBn?j1fsGKFyGF{8F3XrnFEJ1#54deobUYw_^XO3d7`DIweOpJHXtEvEPF*{xu+B=r48y+ zYPnMSl2w(7sq#o-*4=k4tLl@N;NRRuV1fGkp@IU_U=GQ@pTA7%4=_Of$MimWaLDn| z4FLP`CC?NJ@Q3k{0|!O|b>Ne8nA$_i6R~zXpe(>3-V9 zXJVf@CeRnnIE4Bz1wh3B`RL@uqPrOr%_kd78oE!pZAS|BnPK&QDQ=rHMk9qSN()rN zihKRp0}VsF?LQNXUSD@}ihwt@yO?FeLa@c1mpUsCT*^FLjy9Y(r&|(rpO|$!REw*T zV+I2?zrJJ#QNPunDMAe^zrdBC`GE&p`ziT7zDhZzSEUl#mENf7s$xKr~tYS@d*@6k|vD)1xUE{T;%uT-}pM)t8QM_8C{xwZk2ja^sY zsNc;U5I*cNxHe%o-tadgs+>7!f1$G4icQW``F(xmw%C0a7PNBF#_d-lH3cKKMq+j!8}$cze_Q2%DVyxVqQt(m@+? z)&X@F!bZoV`P5?OyQ*@5lnrH8SyP3&S3*;uIJDT6-w;o!#|~?q^Suw0%ihGa7~{V9 zaejr9M0Z)HUrtkG^FC|6TkX~cp}Oc;yzu-c?q-B#7gOmcnZqZ6PhS%J2DCkDUdYuT z-qk-g>P?OyVD>SnC3To#;(ThT7)f$_C_33Rn030)bhsdj=tk`C56-3W+YI`y_j~xS zg9qOT1LCaSgj<+fDq<(W-j2&RK!%>v%X@761km5T$kA|wvgywarF;EKdRaeTeds5Q z9BBC7_+k$jplH)7VK;nxb*Nzgw3XCTUd;FOC9{#bQ&(axwVe9Q4>hIlO;t<2p1;6# zS^Uwkm34~VbaV05#i;Fpw&r!~N#Pj!&W@jMA+*1~R{ELByFBgYwSHB-bGd|l?uIbE z8&_eR%&bzn9?08z+p?l+ejU5Ll# zT2v?qeX{lS@SM^eC228V+=ytN7st{T0hL9ijf~QJ5|9%W;23hAR?H^XSihT-5G&C= z<$;O&c%h2ny1=)I4^!VoDehiHZn`_&_^GWMk4c?*;i92SEnN#*@72RdccS&1h2*si72xMS}f ztK#KmEc1u&j?+dRyu5s0aC=E@=7}I_QCs6~JxNbfOV|V+6YG4AFQ5GaC3mBgV}Fv0 z7A$$3fxA6LMBkQ9oi|=xJ@5bW1qp7~+jdYoQO8jL}Y!*v*ka95* z&R{6DOvY(B#DnX2eup8xOy)d3b=xy?jK3)waOKM4l)w-^6Y)U>4H6LvHb5xLb0wOiB|y9Y3a>V$mSu zAkLGW0f_y%(MZW5KSR$8xF~A)M+{{CCn&y1hXlJ=|bT)6KyUtkB(tu~86HNp)&dpvOe1CS)i9j7jZRIMTl zX1@c|s=}t!vwWW1cMCmv6#Qx_oV+EYcvHbdT^Zht6t{MNeSfacNdwzve?g;tH7&F6 zyE}qC($vQUQ6qglb0Bv)!LUwixed`L`}G^1|IOTc-hFZ-=BY=Af1QtG-Li8T<8Zxr z%EWAYLzCwMxX0n(SKp|Z1y4|#AHod_*4rDqJyI^Uk?OkpZL*OW{v88#XMMdBGaogc z;;L@ib*(SXBHx|%y^wz;oyqDsX5#8%8$B}_KpRWnj|sxmd84-Ww@Lox>T8pf_`n04bC9ccd9Ym6szk|`X1I^9t|(mqek zpJXtUqUho($UVxYmKC^oR^Hn4LQr@)J4!j2vKq#?JMb6`M1n#bs4bpIN752{!OEYO*gN&80hM zSU45sE0bAQ(SN%9 zr8ms2n>5zA&5)H>xZQ-IGd~B`EGMg!lr}sIP#f?ky4t~(mRBQ6vl!RSN2Zmk8cc3fk z2{=|&dm);XOcYdwM=!?60L&C|&(T`V!x=y7du0uVwv`i-<>-J!(SpPgKYT^sI)ixlyGvJhH6Ew~l9g{MVvBIL8jr zC7f^41IKg6Y8qL}IkT_l#*|ts0P(29>!kNT`f###yICabviT*Yx0JLp)GcDQNhwm` zgEh+^)$!-l7?}aDGkQyip|l?}8X8&>3$VbB*f38rE)IZ^pa!S8#IcT*+LNPh(X3LZ`^R%EbPdj4h)ya%!f>%>^TN&&ztwG z1P~~p$Z53oVOH5O=vj#W&A9{+Eejn;l73_~s^30N?TYUfaC^5S>~>~D`cj;y`}MHT zN2a$=o;`f8cn|-i5<0_uTe4vK5HtWlWR4pPxU75wZn{zqu1_W3R#JH_wJo+)-|R+v z3wNt|c#B$aw2r@onufe+kz+%deOJ*WlG_Hv1e?zlZ*=_WS4;9FZc^v@`726PY!ZuZ$HwsAZ`zY(--a%3y(#oJ$!%ZiQ1M+{ zlzNr$7Er_x_GoD^g6b_twd)8pm&(4tg28`cQ?2etJ`9ofvh_QimzLi6HK%OcTOJas z=q^0L3)Ol9`Awp??7nPWAah=0XaDk)fXf5p+}SD&Gen@s!uU8=FV)g(pH|Y}IJ~3V zJw$#|kxXf9;_77TWn z{w$P1%daI6iRrX#81w#Zf*ll*KL}J$0{;+2iYb$d0Guh89H^ii1j?u%-w5W79pz* zMF8^ReHa2PqG2s>X9{xN5M&rOG;A@uq_q*0 zQmvmPsBjIhqx4Yz#9{RET~4#>U*M*I)5T^_G@3b1*O)lr>VEZVHFjSmeM2nX-#jQy)?!*d>)8W>PlFN=!tv!6No;Fu})%_^8&lS>RdLgYHe(UoNftt|`q~G;g zn^S2+n@{t#RueTOh%ss#Cw_l;JXbrh0?!OO_HZA%TFXGJoJsb$M}VSivQnQYREvRj zT^1pJr3;kEOtloQ7Agg5#G2=ZMZ%6-u|l!zy9p>G_mGcXWxKoW!X`3%lT76IAogjp z0u>sB34-@9@GVC~@;%uxCI~q-wx^?0Hm~nbEfDs+di$69u#D<`+xud8$3Il@Ax$WC2w9^S9?fpjc?Zv|$(bXZdx zzsaWD_?<|Cc-Thk{O6q1M|uQ`o=f^akv%=o7IQq;5bw1nR6QetGzX*pVJwwR$9F;%aIcYTy-?B7M|Ew*nzUE4?B z6Y~>DzmFZ`CIu!sF2mslnPQe^iM0ZrtVxK!naUJr_{j`XJ;Uuli+o!wNJ zc+)K>o23Z63tFY1r(S4>`CXRp%v8(Xyxgt)Mh})N;txRXT@RzvV~9LlYy`cuQnoU- zjf&iRqv#{$;3@?+ye;i6fXaN6UYc5>l=K65*?cE%uRs{{^n25ck!wSf4kF7eYRZe} z5F;>>{(M)+0C~_iBTCc1oqe(gWw22{_rCS_Qym?v2QcjOEX*qCdBf8j1I{cbkPO!p zaRofu1R4ifGye+cki>;zgD4zgp5@?ZJy^2!3V_i8lQ^_nq^NYz@z{pW4*6gIKHfTM z-qWVimj9}5&N9^bNl2Q7j_k@u%U{^tpA4VPO5*@XKzZHMW8IslRJQz7uwV-YO@*ph zaOylEfM{(h2P#kl(3H<~=k+Cc7T!j##VSp5(W(v;s}6g;-*2&ii&6hh#RDNpLFL2v z?4l^oJ~rr{=eNOZ;Z#1VR1NMdxb1H{Y6&ef90*cODQ>?y?tIm${9q~o+I!ZtJ^21d z^uxCya#l=KTJXS8ljy8Sp7us1>fi`83&Q%P^6`;!A4$%NwByv|=GG`OCe`IMgLRgW z8Q`J%0YppWnZUA5y4deHNqZ;QXD^%N*-JG2I@V1Xp>_4vv+0GV+1v%xMA@) z+_=_=%D)ZZ&=Gh(|0p;dZA5_xx*rtg+T=8>eXM!I%JlA8US9lfvI-fZ%o{j5_A}H| z`sc=`Mi669sjAc9?y3dxX5~#Em8r;Mh@U!uXJj3x&E+7rUwrjG4D{^1tkhJE^OrKI z=;{{UZXqfHAxayoU; z2d+``g!Tjg8L@L^c81fap)Wt$XxoZ%s5=e;XQf3w4-(kwp>&htN5)5gV@celr=0F_u1tPOJTX!QO>iP(Fy?z;LYkM zf#+HQ`(hobryEb+cY`RhWhvx;qzdDObb}@e;(;2f%J_WwR+{tLP3$FJg{PMZ>x4a= zny`lB*&kW}4EdgH!{n1fBElBN1`Ye!`y5gv7!jvd*|bEyCnCVq@a++aA&m8A)faOV zyp~__zSz_Ep^)||nK5Su#g5`Ix-qq#t#t2PF9|zkU6?zJKpc9{!#;}xWrBH3RF3op z`%2WNEEa6>E&>)EmRRh|JjaS6d*wRQ25;M5j}v?^I$4_1DJgIx$KiRUP8ln;fx%UyI0bsU_qK2U3X-@ZOVp3CUzBf1UnYs0on6Qlx%F zE1ZSJp~KN>X%|eJIrgfXouTN(Hk*JTYWzJ~5i7CrCcmHP`bL-|9?5Sd$ zLjc<63_3-%8g#AI6G!Y1M&X&)!~LUUxSm6u$hI3&{RWth#c7Z2;8$sU)>|reUys*i z_Zk{mQ<_`rju4lsuL{-GDad}<5`6bLqAIQw@Qs`bi=p+3t%KR_9IvIN=pr7TK%Z;; zS=l;3y~1<7*@A#;kV(JamJ*m)AW=OHwY~g&NTGy#zKXXu-xZJP0Scff>ZPaN8M1^M z@;U6Q7@4*`M%%dE#HZ4&)=gs&>r}ma8HC5<>sH)lL(8ioZlLn|gProkD;&=SG+op7 z4ENpm2;vB1VBy9seNkDU)~*W+82Zu5AiI(b3B;(&*_~(i`|dfZlp%T3_|cyFZF6;s zw#oK5H@9ESu>KP^xAF$T{oJc2t*tDR z4}<8=unnt#b}e_R3@KqX9#=EUk?sS6gD8 z7U5rCws;45Q~tcbp<)6buH#q4tOAEYxQK3jrk$Dg`trswKa6)#Yd<%P2BdoMs%a`O(Cyk!Wj~6EPuUQ-*uNL=>SGe zqdFL$&iSSEXA4iit3DE(BAYdz9eV=|A{Alu$A(phJQoU2&~6txxm65)Gh>Bi`W#eg zLU+;@>eitfe6Gt}n5$)Zoo*d;)cyOOe*R0wUE3h*#*(&L!Ut#&yr{Om0XF|;0n5g6 z*=&rg`SX(OHZ<-CdR9LV$vg^ws~v)>}qJ-FI#OiYN_3_fRTQ0@5&m zG)PGaf|N+N)C`Tl08*j|2n;RV-N+1~baxFcIdtRy=Xqb(dH=koV8sU8dcpq{9;eC)a^TgcV zHU)m=Pu0XVME8+UiEh|1hbajWo=B!|`C{s}Istb%KAjdJ>x(N-9=VU^={g%1#40ps zoUD70iB;DY&!IpQ?L5z#a*o)1>hQNc5)34sIWmS8xZe#+6z}cgStYdlJn+Cp=tq6! z4I7|iBMXns9UfHQ#P@QraR~hhof6l#=?DwkKEnwZYU=c~iIV@jAJ6VI)d8!S8rMIB zmMn|N>qEC?1ZAZ7qAVnCI`L)*$lK2LZn$Y!!DlR~OyOT`uEQwxm%|TAqdlc%E}>f) zq+3f`*a!6BI1ISyHQuRokuuq0kxynrUWS&?i0JAQCDCC0H*r~KzxC?dS|Mkh8Hv{Q z$8iY_uFu2Bp&#YmDFxq#Y@~mE_FYZ_CSS;A6@bYJequ(5P6FCMVgteMTm|MwHdc9#k1Xx=x&F8!>yE5e4)fs?51r1siPT9`1>LYEX@$@l zox{UJwn0d0ZteI7#le%XiqC68Tv2Ef!-MW4N~$7^V{_L*R5ORZ#v{R0GN{x_HqC>5 zi73gvYZsxYta*|y&7{H>*wK1_iGmH}${ZuxS^lKH)Woz_7WURkZ{d8Bt8AYncu)kA zmsiwi=Jf(qyYBs=oVE-TO{^n*d;aD0xG!;^YMkxd%kO%M|aB7pZYUF>-9oZWNTh?5R$8On<;kd{7s%9Y-u=sM!lf+i0IcI!U5+y*IvguRBL94ZU@YCw z)RzS$crZW^Y>+A!E#hbP$A2Ckb5Jl@f-lPUFb`v_8!?H=f?KxkJd;F-`ZcD>R_}e6 zw}j@&Z4P|nN_^LF{oZ&z-TzuEui^K4%J(v|>;r!%LPi`oF-^woc{zj00_Jq#==SO` zQOCqDINa*VgRPT}N&{7Uz`z4+9etDaUYIdH}` z&^<1}8{!95M-c+?tG@m-7Ppn$uXP{R{O}R(Fkdld+=hwB9^O!L^_b{+{ot{$b!aTT z_B~fid7(2NlyMzQvh}{<>xI-ftlj-1A1sW#HIc`J+s(vVWg|=)ATany?H}-nr1yQgvh}`X2C|U}hC3Z}g1NFkIK4H^znzRcta{;E;yXJSJO(hnLsh59kO@yOsiKC||nF~?C2g>aOO})w;qp1RRPe3(XOb&i z+%tDtxKz8M$cM#2k);vv0Hd^tC5i>ULZ{l`tAUrc?6J2_J^iCLgkq>+nF(JZ#_&9nNSU(S&=lQqHXcr#eo0%>Oo+kXS;HZwb#=n>-l@WRwszfiB_AshU@cgi;QR%}DsHKs{>o2P(@3)7Mydda@ee5?!=}1@J`h%|2#yz(9 zng*s6zfPaQXBSY!3@RQnlb_a?O1bH`s19QnaXz)_a9~cD1Yqt2z0i$NCGG_ni!h20 zBHKi4d?G-_bj2rvrf=ren$>uTDonx*)=w6{sd;5gjXLp&+I}#|vEf3JKUo6a&nYv0 zb5)|t*q};;7kDNB>7{P1)%E%$Pm7T{>l@X4E8MZ{w~)L+b2_a_^hiriVTwQP5E`~M z*3jg?jzoDKAjMct*cSL(4_aSWx@WXMo^#_~eJ**+*B!hHyh%n(22XG3x{u!_3Y=}5 zzew%kr$-$Lr8)CoC97=th?kB7Cv#1Ydd_@_pl`(0n^B?43hT*FjB?}E(nPx$veIT! zfp5Tq`A`EP3Tei)d8}=Mw^QG}WK!iiN(J*ghBCe;V0<_XyMxu_`2T#)JY$$=`K|Fl ze52ba8K7B(o@@y(dy1mzuXet7=&VqJooUNk^wB=`_?b6p_!h+53r%$tjch}-7h%CT z->oN0#ks{g=3?w_D@Wh*w+eI-GCgJ5StzNiY6&InYK}ZyhHpkX|Egt_utQTVnu1(S zQ$@8rY)nh)Q0c1_trgO^1w`=~Q7j=*=o`|7Wg2Pvw_eslBysd`&t842!B@lP_ja8< zwk8v|YV1?g*Di*$^&Yl)9g7Ts#)E!@cTeqwx~{xKU7lRFfLVnOPwng%yj62KUzOJ) z#8p(*!~p!jP4ZW+lj{spwuh6y-*4T|`_m!Nn)Z_@vH~bIHyOwOxtNI)EK7D z-#3xR<=f~U5J)(D>r>{*>8E&cHi9(y0{D1EzrU8uV=bqReC#iwuz%`k^a5wx2{97| z+(fK9c+nCQt}VMLWuV2hte}nIisX4``5bI?U1(Zjxq2PJ5A>*0)r8heRG@yr^-F`l zAQVg8Df=nao+<)>gEB|QKR5AB(RGzv<=tFKb5E36pgZp%jDM5k5058>u6g-DxWGGzW@P~<)X7!+~AH>EG5BL@y%4r5HJPP4st z6re&GaoxgA*8|t;jW&d8jmkPr=-ZP{-tF=n%oDiRL%OI}Z8ZjMV^g!)LAB6d9~%=* zQ@T@B^Vq?O5nu9pt2lF5fLOABfhd@yv&} zdO8YzOhlW@@&-(^G69Q0es?pXip50x4zpwyZB?6f=)6%4H@h1|enumY<{L~^UgKjK zZGYd+C2%Y}P>taZ3lrND$yOwuKzFOt@y^a7gb^!mfosKCqNS>-3dTS5pCZ1dzEcgT zNIW`Ns~E-0`d=}=m>+1tD~WbQdT6&Fw)CX5zNk?;Zo1>&CU_GHigmnNh#gP|)gX{e zwN;CgsfaNQH?Qmdy+;3~+xxW|<_s`qy3vuz#g1YsUY56XM=LOUasA~5;Q*5#MySdkXRgN_1bi+ci{FbA(6IyM_NQEX9G1lq%^w$S6^73K+rRlOzj^44!>@Q3a8TjdLQ451qMFVjak6LY#4zK*B>)Z~L5--vh|M#s??% z_L{INgrpGBC#Zox<~Db!;UF@orOf5X6V*c6x?7%sxmBXLGpiOXY=S#cYM9l@>c*hF zLD8D-pPLSzq>s+SxCjno?Lhcr|A)L%2Jr-29vGn9{gw|Ba=QFRoltt+iirSMa(v4 zemmb^6#U_-KRv3~zkN1xJ4TVF>h_n0V?XDh&R^yV_1D0UuH%luLm@-e7{ORU`1hk$ z?`ZF!ZUM7SqoNa4F?q_Sd$9=Vb+&O*_dulRG@nQu4bdP?BCW9F*3pY%-rS9r+XL+Y zl#mU8+UmX=`$Hu0e)AB*)?t`7bDE9c!zpYLP*+@CI!)Gw#>AL8+gAm>G?are&sv7- zoqH0LICZszZg3+JvsN?>^CW5;k9{QiEQ)P#llZI~PGnEIPn{Mk91+A3&J*9Zf03YP z-C6lyDI95xz-3}wa)@5|FKeY=Tp$$39{97F9zof#vyt~}(6JE^xRHjDxXuokPy?%% z<{?r%3-)X_Lhp@P_<+H9q3&m^e!9?9-Y*YOvl)bZ@*(XKrlKf6hnrdYH4UK&ez+{( zx`$7WhzB)-l<)YXt!{aGr-s(t4$kOmqxJMv2LuA&=P}}-c{TU@+pb#dhTnnqf~J@AJIbSu6vB7blFolB?<&od3M2} z<1XsvAn_9R%Oc9z;CrFoT4v8za(A*JF7ks7Os3q3Oena!?hv8y@u$upgnW)}(Z(@Z zXmlmQv=FX@z${1NiSIT={2UJ1HyD(_w+UIr0@7^;M_0OV2uiKiIVxtM10=6ZW<=I| zDBtY9vU>OoHVY(t2=L6(WRDF5kI+#_V8Y=9Li*C zaY{^bAMV(%2FNb;9KP)9J54z!oW5*P7n!?0YnWP1^3`=h*H*7u@hID( zeQst`j_XUGY(Obb%K@06dqKWPuKeLxEQsupw6YCv>huNf!U9UE*~wBWmlWy(x()Oj`Nc zXp$-wlV^y|6wl$^ny%vqcUq>H|Ci*lkbuF2HQ+6N*Sp6jn}XcN%WPr;NG@I!t@>ec za1qW9iX>BRHLx!RO4lJ6!l6(aW(T#iC}-SdT=-KveWr&ywi$J25YoI#5;|oUs##Mfa0E)p`_p+Jz! z0JKW>`I|PY*Ds4v;!tIj4jm#;xJbhWqkAfM4%fFzGji&BN|ppLTh7+K$6Y&xQ9pP6 zK^*Uc7#%yr9tKb2Y%xOQ%yHSAl#x zWBf$vh-ok)0ABb6pU+)Q9!DYRhPw;I|G1;o#P&v|$0wmC1-t6rgBV<5u@7G{`C-;v z;^uGx!lK|zKeJ8lPz{~9o$*lBV*&8?e2DW#Q>%dVEXwz+lt{*Wnoo|sh{OvA-s$3G z@1LUaVzV!tIJWB=x}cHvj7?Ngu(>g%D268dg_*c?l3!@b%)S=W?ljrXOS-3y^_JoF zhX$k1t5a7(gaz#_KO+NCXfji2VL?dA+(z+ze58w17I6E1Mq3YHwC z?cVp%?%SQA%tFQ;`^`E|It!;X%Y6I~TxnSO2{V^j zwq{O8#tJIm!kEb%MAi!Yo3>KTQc}Vh)X!s1nl1h)DnB);%&Fe@%^fY|m@TB_dN-4! za=8&`j#^zea^f#~(=rzKB+08ld*s&r_``nd&8x<068Wj9v4Xp^F?uZMI;VwV6hbpa zLd1YQ7}oXDWOaM~^TNt({G-q3hzWUT29i)`w^u8v-lVOL=01S*m}R-HWq{%l|B82t zBcHl0nm(HKk=~}P;Ct3N5hO4uoR4D)gVetXYD9C;wASH!17w&WoUXhVG;?R;{%hm! zpKK1lJv@>-@1p5Fp}$X;pDnat&aD!qfE)SHw@Ij&Jq)u;K}N)-M(Rny5^i2g{oRz% zz3XYf4%~;$mrblgN3i@YLyLNkYCqJ$@A@Fmu6RB0Q>e=?%yEzcSu2Z8`wB5Hfm7PP zIQa=q<~B?I6q@J8%slZPSl($tPMHAU!CX$uw8I0M&_lbaTSUgUQwSlx(o3F(d(jtD zIT|gG4RN`2%LBRJ@Cm$q3ff=^-E{{oXQm zULQxeVzTJ3&(R_d%#4XWK)L(v8YZn{vA6fkx*5 z?#SFZ3*mk}Kqv=#T883$f4B6)+r#c>;H@Ot!-J;V$5TOupx>f9)ikj*tdiWHY0UEK zHrVhMBNMq3^BV7qE2|?Q8j(U%#OBo?+4D)$fYD)Y&^Xq57{-wRTU=+G1gXA~MqYZ) z;Gv?)o<{-DHzxG9F^?KToTJ=)2j??~wRYZ{Nol5~kRQ)@=0=>(lEm)2Ts_Ch@z@sr zx!wE9M$=_Zgk@}*ap(4G!1vQ^oFuD4D{f-j>x87;gVqQy!K7Wko$vl{>81By{~U}& zBj{E@!A#=cmSH$}d8;nPmgc9y{1O%QV`uV+>a6#Y$mr))(>)WaDI9E1$S8Bj zic!~|a_~vlGG(jupJi3pCOXi-uI*m(wZ*rYK}+5y(a}_GS<2ul?NeB#RtvQQx`62|Fw)Wui>5Z zch#k^d%cYrg2LIYQe12u(GqoJG~21f&e7E%`)Ga7VX{Z75l@X}E6`z4)08kQU%@To_mWN z{t}aU6Z|cRHNfimkZfh&2IxiRv*C;WQNSBZ;yduh>Ws3)AI3_fRXi8K+W4yf<|(}% z8-Y`+h8)TlP?z?SNMmT)Dz!s=`JrjP%FCz4J9jR_M~s{B8RMX_gPjj0p!HB z#pHX}BPDk%3o8bnW>x=spUBrcu;irO;pvU`F#GN1-55`k^iKb|DhN~3i|{Lz@0=bU z$CS9;y}I($?%WD_T61C41p^L`KSUwQ7!4(6B=*xJnN<44W4HQ&<}UH!Qwy(v^J1(+ z+#Fw1ovy3ogGv=9r;jXRe7a-~y}5cH_HoVA?Bmi*E&C5t+X;cs!2V}GQt+;>2mHfH;56-TFTVz1B=z;>H=GU{f5Ns}VhWmqA#xvwrn z=V5FI2mj*TsY%Js-rpKuX`cOZe}+2D@#p;^W%IvHJnT8#g6TzLKy1gG3>zRcn#Mfg zi1K9&*HaU8?j8TX^qMn7C_%4|dBVzcG0#?;ueqZ#{3;%S>R{RaiGqdgZ`L>6kPL8I`94d}l zp`Fl#MO(}{(R5r(&EM0w2f%A`*1WSv10EOf_f3|)&X$Wm(+6)46S8!nV6IFOo@%H5 z*A>2}zt|7=5_poC4rOODvz1gQw##A@Y8a;4OdCD$MRyg{i^(!d{Fb9=KXd5hnjIR1 zI==&L!@6tN=cds=E9R0Gk!QTpf4ChrW3^0LJRJ*kqaliq`fLTQ6E1sXd1V}-h2OG2 z77%gj34J9`lApbWG`8E(cX#=ji@aJ1C0$rt<9aaAOR$pu_5E4mFzlQ3L5O&49dQ<6 zfx@}s>zC9Y6Q1zWSW{+(yZ%F|L8}WnZn4XI?!>#&W-b@iQVEdRolqvzm}CSFt_uoM@g&! zJ2+MX?ao&VOpXqASJ!WGvBSUe36y#d(7*p#ST~2_XC;(G4$Fnyo(Gp)&fkZfA2cgS z)w0YCoia9!{W-%vLCV)Ruqb)TY+?wWijHEafJMqpGK(LGsGcj{C?@&l;Y6AT6-3mAq7g$XD1(O@r*32@UNj8!YiEGkb3;4l z?65C;u4)C+&5}#xU+GK$7S)3M{Gd$3`DZB-Zn@3= z;&YN)GNmaf61_Tgylp_1R#atIg7iyl zVsrRcb_)Q-IYOh}PtJR(CXOQ5<_5V0LvZ zF|588`sfblCftLM3qQGpfDr9&R1f+6X!gf*V$I%tqMo(gI3slQ(&mLxwA3_{lO-%f z3#6FOF8En-bo`$Jay&3QPuv15jYD4X(5oi0-^+Ze{%z-X&M$#?tN9zpSyce|Ei$N| zbFMiB0O6P~KSJ-;^p4keUIm|Yg8zbWqNl^31b&L|+ADjh7kl41j8|lPhr@)~si`XG z$CnYRqwBXp2Y2~3G7xnwe{7CS=!mn|%qM0SUID)zD>d_nyU0}5xq?y8{0F8{OoZFl z?<8b!yIm%7jjdnTc(j2c(M9La!~}K`*#2qjwgM#FoP;#*xevX&ke@V)nt2QtgBURO_{z#HRkho zXXV7wAmW#v=dM-8)wXYI>lXJN4(4|o%g>nhC3^?BW*}K#-F+HkH72T6D|?L<2G}sp}jD z&?Z9{uIprfa#TRiQxBOv@|i-iIv#Ai)wk}o_;r&>LVq?mAJOfj{yJE%CzNa|jOPSRKU zp;u+Ew)@V^AZtX^D)CHzOyW&75KLR;RF>q6ujmgXN+K5S08x#3c1wEaU3+sK`v%=7 zwIpmcxQm#+z_16%ROiJ;MR-9u%7*zpKD+0%QSDxS3s% zX}hPOE4z%uH@oxOqzQuZ_+h(e7Y6g3rjNJn`US6e3XpyMiHVS~NLdG{%$eJ0G*M8V zS6S(do9$`p1)0Vj5As%ZWjS0Y=ch)@)0Ouj&k>e}ZT{KSHjFkPPw(%S?N}^gq{-yS z`Q)#WaEpZ*-s;uB zzp^6HFU)8WbR}~1-r$*UxNB>+0mNDl(=UrSbT@4uza+X0JHdI&6JkOGRFEWKOqAFn zfh+z1;W-K09Yg<`>Jc6Obx1eR{{8SfD+wff0yOt7Yh~+R{Q9gAd`4F`%l4=7Bn5ZR zNbLPPoxz<)?aSVNRDd@Oe~+(+5PL9eTmn;J5r7K`adlRB0=0lXl^)4x0B6bj1aq;} zxS%Ea>?BLBlyzFbTF-!^vQD1)&fC-gdHo=ukPK{mKv1VK)l18&bwH=xtPr+>6%uRF z{+=%3la&Y@RtO7;Jfo{(@WBc!(l0Z6nUZtNKbPQXxmU>si+>{LO<-vlkbyA1pZ_?H z2`MtA3(bIodFdARRjCJ~&j)Ny?SYO(Xj)}w`?E*yv15&1dkFiZ7$MvMFRUlTtgWd zs!h*A-Sn?|Jzs@E2CKN6XoZ zhL4mr8|P9&r)TkaEg6uqe(E8@r*aC;DT z`{23F>Gg|{uAy=o0S1#*)BVM=Uh!A`rhffG-cln>K08M)7G)B3a?0)@ieb312V=Z# z%O8KVD#pZ@qzM1M0Q5RNB`*lOBn~>aAV@N4U$G4x6qfTkCWcn-1xpR6sbK9$j`QRo>h zKI>eOn7??>dNlvIm*QJ%lVe^U@eGHyXd(g_b%VYn=|zwN8P1{yvn0SZ9AOx;FBe8! z1U@3!f@e|ZLGnU51*h7x>C(p0geIJrS3xg7z?J5pj&i{!8+qQpD%ZG<-vSfcu*Kt_ zArZV=klAXd#|ZJ-3a_FKK$|XXtm!jl7Bnv!xtL3eFT341t_oO+zPb3&dtIM#qRn-TP88emseZP|YMR*9j?7d)L%Ok!?%= zs%-*HhwBv$*O&#iRwUAw@^%IM9K)ZYF%xc9>f8V$joNGCAG*|_k!40Vjl?7l zBj@dJ2XUA<_zNqZhE=5{-(*N-YcO*QBIA)qGq_78g9X31X2x(2Xi%*{f32ybp=+vh zctbg2<8)Xt_?0nvthn2KueWT=I~LXkJlLPf$FX|#ICvnE60LS{PRI34w_Qp%AKrBn z_I|6T4W*yRH+c;w`du_S72tBaEn>&=cN57=%vb(`;)v>s#`_}={0tUbE`$|nuH-vyEO5nmbJMIQsvp@)=8(hn` zak^X(&ukq&xX~7v(`5DfagyQ4?|!ITPrNwF4ReI1VUW7ClBf576~R$7#e0d$1gbzK zjG0eU6epHY*9L%$F=5>tdrW50piY|FdkAj_3H?y#rtHy`lH@Jo=fc!|DHu!=0KdV{ z3e|MSRW9xpk-8jiz|Z2?df1iH`a75Y-Z$~vW&i0)K!*v7Vi8ay3@so}pz2FaGYbl^Vvt`WD*NMoO6+a++ z$&euo2YVrp|F#ZR_xv_|WTe)QdiNySTn#`4(d}>to}4d*B?`;h(z^i%pst1m2WtrW zVD>kU-}Sk)cI_L>_A}Y_zQi8-vO!JI0N4YMWtji?{=)tBfwaQKUA5Z-DQA|#$>O6V zs&#-cG#xh2(e=8muZ7xHpF2S4t)vJ{A$bPtvgH2Zuhj?(_ihzyZ}-C=b!6g4fkbdy znK8QAxwKx-M&$>@odNB*tRTn5 zR#iHwi5Fh1JbIN$bX#k%FEwe8Pkdy44x9)7`>N6dg|p8!B=WbCyM4l(!p|9FVF%}m zw`}oBr=@_8wbZcIv1s97P(C5}2VogTPWmQia;1e3HO?Rnwi!gpLZ>{nzze5*>L5)Y zPMH%xE_mX3$`p>oG%F}O1D@M9xJx~vlNILBDhOHgl$v(!7Fo~4g5Q5I*Yo?#>t>9 z2E2H+`b)>|dSme2$m;JxnSkilyP0qKi);oq zc*W^$W);VqlptTwX`NLxr*;-p>t-bIfR1HZCJAm$BF_Aru(qFmGj-8AR$#3DfC%^a znfrQ>;&s!<_+b*i@I%IQn{2s><@kHKbJ4Vpue3fLnQ=;SX8w935Qx5?Utsl96JGKm2k$&syKF zB#Suf2$fvpY|pC+A$Vye|B8?41@QjUa>{F93{y#oM~cMM$rpT_VH=r~M!&`*QJo#o zL(^m74v;pBAAmY&z6?>shvap~ zfU~y4r=|D@mjRcjRe+O@#KT&h3@;}U=Udj_;}SzJgU)QrYx-lnvD!g{Q*RpHZ~6Rr zb&?u71(3!ze_*@e&z`n0?F-?2rIuS<&@9TN>e+fi-n1~AEm4s5g_6)BMlRLedjQEs zJuJgq%QmBv{lH`O_Uc8^MxR0@thcFA;9>J`^A>NZM}Hly8@tj#$+Mq$g;*PtVA*0m z)iR0r100NReV6r7F)%X)qfy~8YS*w++1Eaw;>E@NJWLNtVG7lH++gsPiRirv zQ8eeSI0V=<2A(Y#Zp~AEumkp~I*c*_pJH{{Ny~p!(ZX>JzdVkb(8AxjnF3@dLlhdZ z`_Q(|<~IVubq#h=K*+Lqe`}a1>RXD}Ll<#VA^hc*cblJDwD94h!i2? zHR$7x9a-K#a*c{ROaVVXjwDa(diV(!VzULPIAQQ8;}iYY~m<0YnE%b zfK~%^rrCe!Obwj6#~P1f=y@Hv!|Gf@!@whH3j_Pu#bKM0jw25*%+8#y#~uv6#htgo zq02|+0gqFWgqzdtvS-FHzP3E%T8;_%DV4c0NITO%-G;dB07hv4yc1i@MAmk zi!yDKC%Ec+gqA!1_-rm@-5m?CMFc<71uJpOav_;hqqT^#a5)H4M7#{cCnIb_ekJfY zdi$;N`J z`2Y**s93z9DT7?KZ`B5b{DU;jx_KJlyzX?NUf>RuTy2tv`yEfy+aivRzW?%<4qxug zQM-xLU$43X6pbAEkpzUJeK=4YHrwr{S^P?zpH-M`RVn%bmHwqhvxd@{aIu06WZd*F!~jGlXIHM znLTtE{54;&LQ^`kI>ckB_Xhha9H=<6V90s)o(+~sZBVS=oPqx30@*XGc>LdP2MV4e z`2dKi^hlYUr<~2S9G-8^ZHVHdTS}?f0r|#tMN04Wg5shxD`+P*Ge6w#W0h1dyri%E zGAfy_9?bAGFVq@?HA z$BmN6ph2jsY-Vv$yk0lYQlW(mxi&=)Fq5?`kf>bop6?iiaPcAudn_|{BIV(gQxmwGbdl6(P-Ur zo&h3p5M6M&xW&vl$l0ihSSS1oPyA$?EjG&4Qg;yMzR6GJ)cn4Rcodpe=X2)&IC|`8 z!N1fjDH*k8+%aa{EO)wp6%|wZcAt9tpfu3y!R$~0*&-x@UvMG%<-!~;g0Y`7e}D-U zb~;&%u-$bdkHj$U*Ci#|cLA@w#W(X!9%Cqy0zBjYCN8Fo>E-KcC%5S~-D>t6(T{SKHf#ThJyj3R zvViOXA1E!4kXBZjkM!n4z>whf6RH$)#*Yp2Eqa)kc|&Wmr{sU7{){V#DGljq9jLta zqXMk`_87L6vU#~PosB;kcX;3?cHS<>gdk2u_#6o0okoYvc70CfVn2d`=aS&$wb+qQ(OIYypAHey#ON!55q$}c0(;mXif*OqU z;%c{#rwafrOUTBd1{Bfs-IOHv9QLhDLiTBS`@hP_VVToY52SO1;)i~G{c|xA6Yc(! z08Fc4P9iJNY_!ki!~0o5u+8Ad3ZA75qJ{_|PB&J@)1M;zVrhKuJ9=oKw-@pc5X$rB zzpcyjTxdU_i zzI!(`=iP4{6oAhIGa4+F1V8X-cA=AFx_)id&moTRZe~z`gCPY^t0i4 zZMWXu%=h4Gn_GhqS!umU5wjm`)TB&HzavQm-B$yjXxDjqFg`oyA4fkpiAH*5Y;|z? zvQgvKf*wt_weX*tQ2sPN_Ap`&kn$iBA)>FG%|THxe#1D`TkT$c8X&0G%>44EmnwRV zQ%KS6VPv)B0*}3JfPsP#3_>%E8emJXc7>dA`z%^55Kj}Li86Z+xFfAL3b8EGGF{9$4c=cSN zivKM?odVw*W$KgCz=3Qh_?eu0d?0n3Bc|EESWD72ci~G`OIgQdbLZ_#=`?!j*FL)R zcfq|{ZTJ*Fi`ZhA50yoOtFi@(rNLYi_HAO4GJNcGotjRq&@$MzRds+lDb&g*%y4OITP%Q8)0F4%y}};pc|c^1YfV5hzQfr8Ev0u>2C)vm z*xTmD&bYB#WY=nDd6!O4BkzD0aN?J)OUQUx;$@g4C+=)RCjHBw`@!$vl7dYh zVl;XacAiey{}Y&k|mx*s^~0j@Jfoco*-P-*?Gh%%!NN$NX@` z6~HGx$)c*YSNzc@r)ydHLRK(v0|+;IQxZC>{F{~MESp%dXLfK6qyoaf#Gp-eMVC`Q z3s}eGgG2rIqzz3reS<*RO0n0a$RU#wNr{$)6rI9bMTy6ynzUG(gl;ckq}lgg))O;w z=oD7@l!2UwBQ^uQiL=PH{5_=Gjz?6<{D$h80&Y%34vUhl`n^lQA0Ax5YL-shZgcm3 z>;JxD=;+TW2Zml1t+bBy2FEK6p*eZgmg!zQ2S<0Z2&+wZxNDuO#{2msfuJFZ==jR`?V?a> zcc{gM7A(T|&kf1@%))HwUMA|@d&3w0kqwO2IRin`*O+YI(~HC{;{*H#{z)QlWwtus z=F(uCi^(gJ{EDT3&%KrJsN9k6aN)CL>d}7@BdIso%-+`LezD?1I@W*MbB9Q^&ox*u zbB@nQmGk8HWw8!Ys3ZW=Lq!$A^k<|OF7fTZIgBF?&NZ$!J}H5| z{Rp65`|@Rjec*uWz0<}X-QNoC1N`{E6W~>3?|`T0(|_l&eKX_UN*Nzw(vUmfE9C4G zgV>N}T9x3l?^<3KRK_zDE=+K4>zL4j-g9U zp1TRQ0_&RxpGix7fJt9Fdj_SVO=a~0tK$D-?=8ck`rEy46(poZN*V+Nq@-IKr9@g_ z01=Sx96}nTrIC^@>1G%jq#GoM9%>l6<5~Rw`?~IXU;Ezs-p}*ud9mMc%nZk|)~w&G zb=L1ZzaIcHL5am{^H1IR=%=Xor`HhimbD}p^TUj*7rs2w4~z!yn)Vrk_|FbfRnJc` zH@Z`9*8f1MJ^R>Bt=Na0f5qwj5kAGPXZXiT{tb+1)lo1!NjVui*%AYkp?m14jPFZ9 zzn4*RfdGYI-0m@E4l3{yGE1ag{gbN_65Q(ULmhqW<3p%IJfDCnzx0a1?%%xU&{(!Y zLkGa|BD}ICvHp&SK5)QX_ou3%Mhm4^(;r8@f^aS2pKE#l!p?&Z zUPKdPki-rswt&&I?JV#*&gQ!Hvu9l!W+ zetK{$b-gHVru5CtyJK9~%>Rnn|ae|B>VJFQ3pT5dy4{JQZrJe=7a_=ai9ue8c4*&i`_l|KjccoZ-J%oPYl7 ze^%+AOYvX4`2XiB*-brxiPZe#;rNGO)_?vO!HIoKwDi~K9_C-cJ44l=>j}TZq2f+2 zV{|zM|Ggdk+nW9#|F+J&St=ob4hO}y8UKp|{4d_h>pWuKK1iz!9%}yYl9(P~rs4po zpbTKVx92~#QDD+^FM;c{^u;=HOf`zyTAy#^+AHA&!&$_~6N zOoA2($F&dD{+XZrXHUcnYG1;Zca4NCq1RVUzkPf{Q?KO^SL%m|3@||Kl9zq7B|Ubh zvIbG4BV&(IgBvh9cqe?GziVhMI>bO#>^?p)Md9`A^O3-^M;V74>$WW9wEZIu*PIy+1sLuUfK0h+YzRL>wRD z*?mS1x6t{iDYpL1hYt^#{LF`O;Qtzb(j1`PiZX89o+}3Oh$oy|H$29SZf%gF>F9FU z&eBzXc^o#JW~{tI51_FK4gZhn*1vx3Xid!PqpF;%8)FTR$b~db721BKN4*L(X{n7b zT7KV^$}cXY=Yq5M5#yov2mKF;)PHbo<}qK1o5kOSF_vf$4iLC;x>7*QS08%w>8yVrsuK6@Orr+#-Rge#(xJgbbZ)gR%vr1 zn+5?gG>0ytMDPO7o|8DG&?`U(=*4|54`_yl$8l&e?ui)p(oms^U!TWX%^us>)zuOI zW_3SY!@WIy?SJI9zZBJ$6Hi_G^6m{Qzl<_CDQ%Qp1Y& zdUe?QDeS>@V4Kc}5+XomH$I^1+k9&;i8wN!tN2<6aI@{XD7HIo zy882UGJDY!GgUHdmq52Jvu&w%_P9y!cbz>}uBkDfJCcst)t~@NK34b^HXVp z9z%ayu~b&nG0K_stE5E~kCUgm=cPJzr4vP3NIi*^S3RN1F~W7WiN-B=>|O5?;=;+e z8T-8(V%KQPkzpKb3LH`S;o^a~3*tM7WiC&8C z_IcRUg316|7gpZezhEFM8$n81Td{yhC+FktX_t5IjM+`SyEudOFc`FPOP~z!z}UU#@d3!a-eM_RL@EgTunqiU%@6CG zz?}A*(2l3K?7|y{^W$vm&^=PLifTEe1T7`o?eJo9>zsNGupd%Zx6FpaBacb+_L(YX za;PLNH#4pAizCGiozLma-??a|K0g^WA9;nt5_oBB|b9n5~9D3w^HplV1@$=3tO^*OkU7KN> zJz?JOF_Ypo;E_}ce4twVjfjnIwFh_nlUota#3K+nKTBbv$%c!bti?_~;|_G~^UFDJ zq^bALe5f7sUztJcTZN@Ml<(_j{y)>x)sFc>#cMtVUZgN)+#9JJIDJ%oo>zCW*}Fy0 zi?t%l(ilsC?$PhC?y$hl%{Jczt_vsUk~C}3M9hw5nvV~}c@0DEf5q43uC^EY%{v|V zY?w4$?x^7P+wevfc~>G|HACvKzgqw0PaMT@fXqWV&iSU{$!qpYypq?h~LK^^B&y@7ka%a>`E$%`P>jGT8oX(ZRTSJY$8(+}Fb z9t-c3aL1pSi1oT}F&z4TTRUh&r`~#~q@Bp5Vv~H@DbVXPsBuol)gr`S88b{$_IjnN zpShur*%RdvTCCywxA^{bc3YLMUxw|gn$<9{cAz|4a~W<%rS2;9?q|(sxf}YEpWEw= zPA$Cg-n4pqe_HBjsk#hHmvwqU>67;dxmr?83Q zg3F}*IYd)6@mhC%>xcvTOq%xVPqvX#C9JJ$ySV_-3ALf6wOeubK1DK7m+xY4MG~pD zCZ>)8_*?d~ME+bFHQ%N;A;ful*kXhA(;obGdyct&!xi?!wr}Ji3f<0# z9H;Z1p5WDP6@7oILAq|uyJny3Oo~DF=%ziWumvwgG%<%>L888Hd1hSw$yC7!({`{y zy3hG7c3z?8mSH-^cI&blt=O_l>*pRDnVtI*TR%!=MK;JZDo_$&>Rj)c#n1H3K**X2MG0j=CQAr6oTU|L{pGQgN1xUz$% z1HMHFAyKOp%3$O+dKH>t6t`ae3>rqvQB04O$u5<)VcI*zWMiHOuPj}->gDd&+L7Y3 z)>8>nK6MWMbGf!+_IY!k>m%o2^3jKxBDA7EYFM2`j>c(G24tZ%$iiws@I`6CU9EzA zEt%oy(LJ?G-xW~RfAOr_Yb%k~Y097oO><<0$xvL}Op%f!q!T}H(4byA2%9{+{{iFx z{TRVc$f%^H7UszoiUG?g#2@^!y4>}2)HMiqkJZGMgD8Ie?9_**`AUqjtmPm96lt#a zo@8_$)36Vvx~2^aMJ7>P6y}W2ggRE_I;HV*yJdQJy3DTzJbQ zve_}*_%L`GtC zS{fI0-wq5Q>chvTqGT>Uv&?y~)WRCI!|VT^247mEjb5))k@3D0nIJp_0AA>C7489E z9D~VfHmqvh6q;Z>Y9S~2MUNZjdw^HOUXzN2fb`gJ+~g%JiVzp%Y_3#VBA_5`j&5e+ zr`I#I5*JEnGMB*SFxo}Z?;4DUCD~YKF{jz5b<6Z9lRKWOZ~U%GMLhm~i(fX9n!4t> z*9F^i{;B?=qBh(?YEMp!o6yVUM6i#SsE z&(+Dq?TbLiyM6h`o>8Zl($IWoGnLm_wsi1Vg0r_lz6=8MOntQ+e^=Ub_Q0>(ipkM(^1y+B%Y6e5NVtTy=AX05PLD$ z6w;iRX#Xr3NKF8mw5-36Oji^0s#Nc%;m2WD5(;hFFl*?9ec%PCs*I83d&yM|M;URx z{&L$VtFm}8O_n#@-}q;Q_z~7ow64mZoZ!4#^2RZkzKVQ#9nzYgtvTrmsrf`cFIWSDTN*I62sR8vCSYP z%I}8($BrTHX?T4`ROx(FW%I*+#{SWMcNwTE9YA?aYOTlf78B>EfJ>c$sa2y@VuWbR zm)qkUxS6YehSdTx!PHcbIAMRG)L3hQ5zSv4CFHF6rnFjp@fzj{)~+-W)~dDAqvQod zOc}Woa?X!ZRc9|N9K2u0TZ`Mz+3LTN)%$z9!tN0Nx5FAkOv~L9M8Gwq8R4I!@!L`N zoLv)A@1(I@nRa?%meKp5=9>Qmgs!rp6Gq%c#3Vye)%dC@!V9a$ze*!V||-~n_VTowt@otCOT zqD1(ZBQbc;a&vmd@qFJ^s6+@iz&&dCeLBnrEoMn$z3sN5Zw}VilxMfRvo`x^`iqLT z_$bN6^>(pTU-5>XFB27Z_PoKLMtzPRNUyRj(|~KO(KxZQw}CwSIzeDVQ&}He>>K(| z>7_0#lsN^C25p}naWG}1F@$fF`E+bAu?Ai1;bf?n*H1s+_Fkqwn(wqx94C*s;5e_c%2Y-4Z^V_hzY0XRi!}>r06y-Cungt)tX^0zq1<0UFVKlA z%*!BgQvCK>3T(aTvNPB?KH2c=0>NN0yFRItFbLfsbPO=I`I9l>+*un~#zBhNL#Tef zSO$3_ezTj|h}SZ8!Dcr&SqgHfr#1lJD0ywVP&S$WjrMX<#btI{sPXU`xkM#r`hce7 z<*?{Jld~BibIf%xcF_I9ws@!9fW77VHFM6M88*e+&i5QV7PB9aLYCdDAAZh@)4Qf^ z(e`!TSa@$5TO6}WBN-!Z4}KpPJ1Pm`*Mhv)oo@)~O>G-47^~hv`s6?RuNvF}NaLBS zLrWnM2jX73=>l;6Qq44mq!@aIbRV+643f8@eWT<6hw2@!4`$=}@AwrEa!Rvsnbro{ zCGsQVJ-F%-8srKvGY}D8s;PQ zfs1Ig-2@8{fg#YFJ%uL7s$kH2Q_g^*?glfpVbS>v0TmAJaQt^IUp?}kS4CmT7OZlO zU7qlt_Le4|{&r`{v8^d0dVH2)9m3-;M8siNFUN`erRC0f7F%^av$cH1T9kU5)wjd2 zeerQ*G>Ge>hr3fGU;O5ndj?|i1IqudIJ#rV3xDyX3(K3mmX6f(2D^(A+LFVwp-kYQ{!HrJl4W`R;8)#Z zOqBSA`gF0NBZ%%X)*4BG3Z8|jk0Si5%d>5#mK(&OPL3Td8)thMSgLqcpb zLjB3)JawX-Xc?TqibXd2)d%PHO9v9>{CDSV2&2vnc&3$?w8d}YpguC z?q}>dvydVinS7${v;{C|>c^s;l##?AYhxQ{aUl)Pv^HoRY~17}AQ{&nW~uaiNTq!0 zHYqw z(|Yzg>y?K^At>-_ta@u?=opLgOMYNuE~o?jI4e_4g626^7DFpOFXm`$c!NH#Pz+pF7*Djdx*Uqukw%+@*j*%d+B}{m`;kZICjy@RBwJWRFD|5Uu z6Q90b#H!%fc2In5FYmsvD)t-Dc%3iU!xZNp4zIQ#$&t~e9$r+LaKu5_D4zL$p>!*A z67#tkngb<{_Pao=rCMCMTb%g_J)_RjMxgoa9>Xd*i+D#OD_-eng^6hwtG4&`DS6jN zK9n)9Q~L)a6I;#w$65; zs5xbM8QXW-+`#K1*Ymgy-Qysl>)TttN%#ZdqDxVj8-n+S@+big*{xLX=Mv|G1h!nS zV7VjWwbt+^sqJk$f;@ze@T`%prwBsqC#~)Vzc%8!A!A7-3&Z6IiJD=Fb#pK2X3Ni7 z)*n-KUmxWo7aPz;f{6o-p!D z^W*8qJ61>ij#L`m4#l1winE~gM{`N9?X=yy&2FY!oEjPr@_vZ$zKAsyx#+TAsWG0J zXp)=4T3KK#MYtl}8GcF}h3=7PZ9JV%YIpTpN^Qyw{A(_#&~ziB!tJ=!ap!r^)PK~l}S0qd53>1IKG)fsEgXdxn*y^E^)O< z*DPwL+cs?m&9D2-dM;|kAYX{Wcz$@|UD41G04Lwl07ekwSNIo5XRyp<`@IjD4K0MMXJ=xitW8$>Yy((j*-IZkh zz;a`l$2@b@Ya&E~;9%%gop`K~VgBmr!d^WLVwB6%)Dp4xzF55?HM3)+f8@(cHE82v zt*U=xW5-s-Pv#?gH%!=<%r*t*txAo{pH}%1?+-R#R0pzxn=h!)u$%xm7>`~;nRYfh zO;Yti=ur8?n?{9(%3MU(Ws{duPdLRZAU2P&Aa8lRc`l1CfEti!FuTk8-q^-=-~eZs zHbAfrJ&;0mrL~&O(Zu(Of0#1gT)F^UktN%%7evTaaC7H5hvj zpo3|*pM}l2^!>q9!4N7)L-Q>)$goj8GM+`zO`Tii7lAKJ;9-5pc%cS(pGOjS&N!0z z9PS?O6TENG!E?bM<*#b-x<#{GrWL@s_EnmfdO@O7DnJt^K*qh|AV!Z%-oUFKUwPSw zh1L+q?rgcn2{L(6>3jJayhkTdV1xal5{K^fM!oe^&QG(>5Y9f1bhEQUK_q#R@1xD@ zpk=`TbJXzUqc>kXLzL2}5x@HO6ojN9GiWD6t=^)jpy`7YHc$G`Je^qcFw(VZxk1*j ztM-2Un}_K}_((3M68C!+Sj*KIbO-}}wVMJrIWIU6imUXRw{TTYXhZR^5Ic47rpuZjwTKakvBoiQ}9#hAs~P#%56uRVWfD9rG?`P282jHLF@r{Ql)ahx5+jrQkj z*@k5hW_MrVT%EKNmOu@sN=xPtwsOljR%lFa{%1{s4L2zQ-=#hyrE(o=%FNL$B^Ms6 zQ$-@(4oyak@JS|ro8E<+eZ+tb+q%)lfBg}@k3)->g)TXV^t9I>b97zx=m@*^Vt6Nn zLoDQiUJmHxlv`p@hT~!(lbZ;BV8_1xw%ce5y8}3g))34Ema;w2-+xQ_xHjo z#pe}bNO0yat2wBGqyFi!2wS^XF`D+0nE!*2HiqIv4hJv3AJtrTIdWyb9$yP}f`%ys zD|a3`k{F~)*>L#eg90C7*2O~vE0gt^c0KKzseV5hR;HNa!^rqTKcCsjEv{R>X`U=# z{;^2tu{G&{s#t*!9?bFWhI^Q&5oZS=!ep}O2i&!3TKk)Yd%8`OK|&T+k%_y5S%8q& zJa@V(ZWf=q_~kKqR+r<%Du=btV$H47o`0>$mc!&p=5t6`W}4xEyR{2qkyR=va&){d z$$CHZ^0}V9w@t!kQF6ifp%Cs+YBkXfS_f7>(jT9Ns3xzYOA%oqiB+~N_rurFuuJ7BudvSu|c5zPhc3XVpbzGX}Pk4DQ|It4O`>_@iGKWzNJ(iCKgyp zVL+RdDGPIN=V!^luE+#tLdJrJ4L;I{V1z;f3&Z}qpqi~es?BwAQC8zQHu?*Ba$iny zd+I_<)D8Gb4LD@$DFv*O)>`Ic~iaXbW`svWk6LMK4e$P6RxTzr0Y|FZkHp$m6x-D^Kljkk0k;r%UrQIAv_D zR)d>t=uNR!$;eP(3`!BlogkdT{mmerx9--2*BRrW+o5GX}&pO5!NMVLu`=EzDw0$8DWPl<=C6eU1fa_ z86gM_i@vM=L*ChKtOcC_<&%f+s=sF996g1Wv1-2wm3Pj$Yc&!=WfKuw^xQ#&e2v1- z+Q%r-y%TFLS*m%wro}6cCNvlEGaF5^$@<=l2!+r|slzEHqcUUfAV*cbJ=9*$2?jWk zM_A;V4K7X${jEb~bFNLR-m8~HAM2A}Th~%8*C{xky1TcxjZ-?#v?YiLDlDpl$^eEK z2uSn6$FkehN!e1wCE*eQ7I~;_{8;{vg?dLMZg*2oF7Y=5q%PiMBgRoO3M&&{zsyqW z30_;q#gj4VfaVPi#TVd+djA;+!rNjx*=fM~AhWiy6jF{b zDwnT48j8o8bbR#aCHAKQzXmn>2vC)^_Bx4kC$NE5D5E^zAsDymmFFP=J_zL9cu%1E zY%!6e@F=kES*QI(8|gFCuzE)`C1s);`tQ*_B!3~rt^lOi0P8!HskR-6bnlZC4S4XJ z{6(cnPM!y}{V5SNVp%sh+YelT( zIf|ZJMcMQ2uVokSqyuQ%vGUaACzfQeX%{eF7c?!KHv^r}bi=CH0m;9zv|5Fy%|Tb$ zL;U`PT%2oU3^a3~LHQo`jCvc=xKJ|;={~H<*4Xbww{6ex|8_kRVg#f47AHb}W&&t> zrJ#cxr@b*<%`%<*BeYR+li_|&-ceEKAF_KK;~O`6GWSfePb7OMe25Qp@6It{q%Zug zp3*(uZtkBDelUv>#I_ekrp6@Ce1h#<#h{?|<>6w9)bss!vg$95Nd>;0R#kzxc4FNZ zvu@vbeG9?kj4S)0FrQP_Y`*h#_?sEAbe@S{-HD~eQY!&34c7Fdpsk1d;a`=~cw|c< zhfi)6kkq6eBeq6286{Ey{OV>PGqOy0hs!)i0}9114J%MgY3RyTq6&Rh7Qt$JjJg!voA@l z0mNW-AAlh1QT*n8mn7=dcq$;CDR5VKSvs2dq|7|YDAHvyxJPtjxMk@P5f}1lGu2A@ zL8pDWY3JoTnM26 zIfyxIV{m+HnHnh?@lr0U~PJreA;a4XkEKDPF~9%L1MQG znZ9-eNp#uLd;F04EPe2!y{uEYl>7S?)ShM0aEC!}yz}ht1XK-!T_(h6f3RI>HO zPX6VYgeq7yn&n)J+N-dsaxreOeDTok$zyLmbEi>)+RrXnr2BNtGIE6_mE;Wq>ZtgQ zz7c!S?)}9bV`3e)A*`e@VqE@dB_dB#^oREH8dQLA-k*Q-a%Qxs3iD^R5x&w$p;0^6 zRLdC^nOJoT$m}+-tu<%;#N2hhgIxBcdMi@b3Ra_lj%ePlx`aJWc+U#MPpt|VlKLq{tX>lo)q@l(3`0@FW#F^&Jtpxou4Hp4&z#7EcP52q5LMQ z@p25jGxanMj}#(L<(-p=`#21JlV$R)b^90NW-q0MP9v*TGXX6(6iT7v8e|AtQwN#< zlaU!q>~uqIyrw&7_w3H3(^{6|u=-+n671;)V`q|b3!tX7cD2{2^ce|5xb6~u0Eh&m`3_;1y_m%ts_}d1 znAg*(maMM+?S~wotV#nfzt7*<{ACNI7J}{OmL8~liX$Rz7jbo~|3SP&(`x16Eu-%B zlmFWRUi-XSSmM^N_tS=)s$Ea&ods1NqQR zm>ESM1)UCQM^#1CwTC?X^T4^ckhpm}^H zNe8qUkok|rLbfr^@c!nFRoz}Fd635uxdlg3#y@?eZJT+Ue8r5tOzz&f1npr8A6XHO z^lh`gnK0z9R;3}Od11C7#GBi4h}IcKw_T$ZpD4ENInn#aXS();5$r35famLak-v-) z;D0B3F6$Jq&AYwjL5iqhjM&o{^`HuwSeK~st6dS@PX6i(;Kp!=_keMG^ES*!hz$~u z`m}2ZJNnBXb09Z%3;Zb9t=kaYY;PuQ_K5;p9)KGQvosy>CeCL6oyZ@BiyrlUDo>1W zj#1Yv*~!m4OaF5ov-x64%s|>*&LWYs2B5pw(+^bYp>@f`38Ncb$CzJ^=(e%2KWY6q z)j#GY=D6Y%2bew7dBRDToqy!d%_ewv_^mYk-vsYmk)93a!L{;I_8zWZO7I%+?&;58 zS$~9;p_M+HJfTltAAY6FMoNPz1uj)7lRI&}GN>HZJ))3GyZ+O<2ryhDLR&y&@6l!H zR#fFa@0Y6-{B6xB$FNEQA5l%Ig}H8;&+e%^%6sbkZZzNpCbht@(K{6-9$>sMtt||IVvQf zxJw+MfQ5owDP}Qx>!$+af6r^i`#VNCF4mfaZTJF(*E|YpyRR_L77^p-w}7mrF-?bb z7bOHs5B#fIV;!-w`sp@5!;>yMj_E1}kdQ(wb3Z0{G;mTeuK1$mk9c{SOHc}Y&Wpx< ziN}+V0@&g+{H9}E(P^8%iDM=vaALxfy-zttgsE_@>;$sC;!)1oj~G4J*L#y6Iu4## zBcNp4#}1~QXBKLUFuEfuP9HaFMV9I;Oezrh>5?&Qwh<5gAYPaF1o~b+G!wd9H{|rq zSR5YqDxq!B&Cx)bvlW@!C-wBm`FYj&8-dROGPO8(ZQo6bEF=htR;C7rWrYR@T{b=I zFKXj%mL&Y`io5@EL!0#h1uLmhN?6~V<1)`rn;KWiJgusZGG3wxi`F4sdTu`Q;)#e{ zE~#Q8<()|HTS4NdF+me;sM*)Rm}VgaJl9L8W>$!s-*Z9>aB(Qx$^7^rifKV@Sgun= zWzEMUF(|~F^{0&vXMC9+_XVTSYkFGf_l>LsZcEywYX%ky7OKb3}k{j>^EmqRwSCxO!vg_ZpvD?AE3ck%Qx!B z8#q!2>v-F%v}Vuu>yC=yd7d3^)`}L{m5~upUv<6)GMSg*zAy1_xz173{P3?hIe(ki zZ1YTgdX=u>W;(NcO-KZ#6*FE+RI3l3HV%Bp>)edS@p68y)?MGS2^fat716RcxF?nQ z2j?gfuJGAm&9$QtN{V-mvXN`eH3hSHO0CURdz-M_W6E5oO3I?JY}eq^VzJvIRzDr; zo~YufzATs-R#DXPdx+=l60=R=3Zid8EKD%~@PKA#(&&+28G940w(|-ywy#PaD~C*n zv>T-hY9ou@L3aP41z^hfS*Wk`TQ1hI*brh#2N%zAf}gM(cY3*JkWb>Y4&7P^-c{Sr zw4HcjnUH5;u&muXrNh{A;O(+Gp)dF9HNbS9Rl_nn2Z0H_Nw4a00A|@uA33yvwNE282vnby)Le(q#u2|?8GwtEkihc#1PB#aiGQNKqD*>pDQBp z0K)6fs#?be_JF<>It)F-l`d+3(3rxz7lM7>hSvK8C|a6j*=06wQ-qN9N>do@!X2V7 zpCffwyRjoEXlGoxuF~J`2vH_?w`yWkhI}w2@3T1~MjX=!y>i99$CaIi&G_t|Gp*SZ zd7+x7H<+efNavJ3KjRrgGLh~Fm?Z+Z$Y2xmQFhLrcX_{}cm^H_PGnqxU_`uj7(1hR zJEP#tHB~Cj$zqKsuRP5Es;KULFHEcukGP-4Q%3cD z1+&!(Rx%f6a`fb|wap<@0BC2ig3#tNHczR6y)S_{clswxUfCNvvcHju8uAXLo$Uwj z)OnF^{S$g)7}OEb8^-ZwTpRx>?kvyM947WuSwu9@HbACFWMhLu3A(f_BJYmc(RjQ` z+hrR{%={BmBxeZ4F~K-;3-etxpLw6B)pq^onPoMOfFBx2@#Y zRWIw5L(?F~huHo&_Xy;cCigkxz>JV^2`o1CBqwc+^I`MYz=3}(Z-xtmC@ew@}3_uX9_Q*-w#+Fj{wU{DiEuMEi zGkYpLD1B%-hIg7Q6KTRhXf=*n_k_jUjChos^ra`0Hfd^8AOj#?Q&yLjK?u;?sPhPM zIZTsu|I84K+P1z&#E#hm=3_Iew^U%>k z!2o?*z|9s$McMdxZYD_3dGsy1$}W65&+X&v&H*JI2dW(g`KA@u#+jsbndNMKI`)V3 zLmN{enuBsrTiD+38q6@?IWB~WOV7W9oCWheN8Q&B(EdDxG6ztkY(y`2+?xrq@IFO{ z5*y}+BFl>Yf}Hu+sjUW`Z6l2;IH?|Eq-|1s&ct)N&JYyWi*;S}WTtDKot~I!_=H5B zQ94d$ui|X*mA&dCgUAxFj4RcjD$JEwjuvG@%1wUkG#d-uq&}(z7><>XP+c*~h2yks zpPO#DZ4jm(@27ti479C^fH9$colvJt;B<1zcKlQOIRN>&pXSgA59;PToDh;}2`7lS z@e8z_M2X-uyUC7{#uF*8JGqSHS>hFSobnX?+O~)*N!SU(kN`Pq3++rQus3Lnqf8Y( zBz!q;tbt(&eh`%=1_j5*#9KUgw$U4O5!$)w)(sip8~|gqiBVyF@LmwAPV3=>Ka6@4 z(CEUg>l75d*EL~nz`r~PZ4s`u{gwUb&(*9IP2AKkeeP$=a~#!;$!EmrRT!=|KaAG2 zsFT{1PQA|~8?tB(+f)6TK?S$TL*Q#!o}{Cpep5}Ey?jm)=pc=s{7O&D(oD=<<5Le* zp$(~f?I=e=gpqDV--3rbgf0;%-Ncvo`-wL3$v~*}HPJuDmF!3O8`6xGFRy;ieqP`| zdQoe{scF6bGfJo%3Nj^EwDlKhofVsqlndh8*;Sb^2GqI|e|%dYjaMGIiAi8?A`Smq zsHK?icLQF0fNAD#ztbZxJ|ch>KCN&N4Ydz`EC1RYav|GKR70j%ELB z^zHtlpI~kCEagVhsaa(^|E!7|ZrgjatUbl4zs=tA5J9&n=*WYl3YlD8IvxWj=s=Ti zZe%_wI->Y6+o0jR;J?Ly51L6D1Sp7!zw8oAo9sksl71+^ttz`n6*0*Rr~2`f>Z!jt z=jJ^-C|9GYm7>Q6>&4hqhLD%?3w4p9NkNSeL1jora)TAP1_x1z@Z>+zCD{-*qC;L7 zOO_AM1KX5^|T;ysS_eujs+Iu!{T4k>Vt2x3705;S%(pOh#gq5uxt@lndUyPMl#U> zEiZ)4djg{L$|Nj$+Lsg={EK3Uu6Mh9=|@t4#wpMzOjmq!Xq9h`1zQ@I)E|e%CmA$a z=}_v^e9t`Tic^zTe3b&mNj?uT==CH}qEBtOQ1yIr)me=i&rG<;4CpR;{ z;15JG9xq_P$L=lJi%!Yv^U9Ckd;n za&WF(uuHUi>?+$m4Q+UtJl=Oze87EPH8)E!S)|*VP&Bhw8bG2|?*y@90s2B#rVt%k zn!WeVjRZU8xKv~NaoseT%}sok9PGNDu1|h@J4bq@EgyV*Gw8T&+P(99@rQ729%LjL zflc3X*~$u#(d@ef8T{W;uj8iwMN#GCeTYUcFo=8e z&2AST+dt*z8oN?OmH?^*;;P@DflsmedIaO@-D=vj|OD5!5qKtz=IH=Bb>gCJ)NioC=ZZuvw{q;VJqZuK0JM;6p^I&oX5i=0qm9y#a;qex=*io< zGwDG`Gel|2bJ{2VWL!(5+N8s}y5}9wWH1P$%6DA6EF$68)pM%ImMZ?&nG+3Z)3)Hq z4Q+p7$e0~_R2{TC4#1)LYe)Yhu5)+6b#h2Y?-TA5rBf^c-9~BE4)zlN#X2H_9xA7W z&t<+A_5$NGpa~m|#B%*rH})086^k|BX(`<(>X|_2l(zmdc_=2@jNQJ*WSsoV3f-K1 z)@1JDcBO7;Ld^(5$Oxp~n)j*qKZ3Ye=DPXB`!Mc1Eq+kX88;v>UdeEW{mgj-y`6u+ zmb}ErqU(w03v1ws9sql`q&QxTr3z+8&NR9^3SN$@czEv}MXFqp*7#ofy8ne!vspZF zq~`Xlhg33k{lxB{mtNC4+lwD%+2wg98i)k8BxAl=EPlyXA{G5Oe1n6uaC2%RgVYVC zE<_b_BDM5m$I1@dZf}5?e+j4KqP^oP<2j#F_CA)xK@? zvcD%Uj^t_&`nmbO<>=>yI|)!VFZFXkqveF1M$FxCEL!_8SzcW4>4y+|zJ3Q#|RZ4@zHkV&7{1|krHaUHg;Y#Vt+dzg8}R?-Uv;NOP#wwLv{wJ?t;s+tu-({B_U+Ui2YZC z53b9F_agQqB1S8|;c5y!}|x=8cgZD9FY!Kh9^kWG7xr$1GNsn<%e?{n;Aun6QPME#34m z?}P0t^VbfKl6oZGClWRO%z>h(sWL;--CZ`mo0I(6`g5jDXYB=MAaLqiz>rMttW5uK;5@I_lG#!s(8$)0Yo+S8IQc@xjqPOjbMY?WGuun-N5e=Y~ByyOAkfnl%)CiV`fyOPYI#14ST;W zrR8fo&0p~o10CqWldcK$j*6Z`8KSy^v7*H!N*az~8LpnCm!E-_1?rzYdd?{rn1j*&MnPriSx~2HmQ^#gXKS3>d>psKU*Kx9lHh8if z`}Q0+%$J0}YF^>IR=CwF3UjqLpnQ5Wj@#&{IlxvIyDRl~68X%qqPBGaqjZR*HtDNf z&vQyxL#fu;(AWp}=WP(7_o-K%mpIE|&k-^fs6C=U1jj<{X@n+L%-Wb<-M6`uz71iD{iEWETVze?eqFj2VMKZj3p z>8N=reMH)ZZGnQ>wPY_3ex2$LG$G& zeU2?kP3)J#8{nfxsuLZ+@ZUG-=!)^04!j zx*dAm7yJW#i=cPIzE-uaSDkSbNi3K_=5hN}1XD-E<|_@n&X;;^p2(!}IXEvI|}SAuQUNa<=BXoKQ_otjcE7jh?dlx$=hotKXu=byu;WE6c?CotT3V z1pyfc4&?&~`jR&P!UqsuQD@v+(`d2;ba&GLpztt0B?mHE^8HOxk4y`C$&np|GhuB8 z7b=61)mSl&^BuF!fp&I1`oiMp4y6|{n*)%X+c(D|cpdbk9y7bcDt~p++CI{WERu-x zxk=hLGIj^k8wKdFYZts%jIy;N51QuRUdcPlQ^e9KakJl=6*hSzIwog8*Ez`{c1I!J zb!s?;HPI4G?I91n*&uH?sh#^UIBs2AAx`Vmk*AniusB!urNU)M%_7q7L~OHKf3B{7 z)mV+rMewl#y=;OysGd?E612R99cA|C%{p(Sdy^6V(bd^*Var_l0;vBasOL^e%`wS# z0iqN9k~&9Z%9u;B$b_LhF+?>%BVUp^d5`upG^WO3;YFr71ouW;UNlzQZ9OxB6? zv$n+Sv^5U>f7m;#u(-Bu&xai}I3!5$Bm{y>fB=OAcMZYa-Q9vagan5I0t9#W!rk2q zclSbjvCqAI&p!7%w|BqwLw{dAR25Qdt{QX9IfwjzfuA^Tjo|^5N*;E`Ya(xIvP(krP^*v~n)4kx-K+sqlbXMi+XftEa z^EQRPZO)BwlQd8cfjiOvsFc9PmbD+4`h}vk*5S<@_U&)z9_WA5n8MAtf`PZQ6w^lr z=ZNY6@i3PCGB1^N(}Xob-4_R(>g{@P!}j4{gQl(T-9CVKQVBVmR~6X!>@N55ceS-h zsLRG|)S>y(R6Hv0te0pv0?7E>E=Z(SU3VBqAHK$~)8zb!+)I~yTqcD*J=0CQRK z#*P1ZCz(B-z;QFsbmZWNy`|J_Ck?(k|FBji8s=GOAONOPM~sdN2kN&OsZl$vi^vn6 z7;f@8L=}5VE{eLT2XK9RSHBk-_oVeE^Rx`JOwWQ0x`!6+1f%ynb6zmPTBzDeGcDT2 z%7K*Urw3PXP?IVP$pUc37xM;qPb|(4KW_|=CwUl=xvnCDlI>+QUqG4ttv6gz7})^H zH%bm3K-{wz!!*?RWvUX>t!h-O^|4e$_v$PZ578oe`pDIAixQ~rm!a*sswS!LtG^&} z{B%eeX0+o3Bx{#cD||3&qP7`^^FK2hwWN!W<5f^e6MS&~TB7&$_l?laCyW6jU#Qa} znYoqO?Ty}xKM8InZmFO;!~U*Gx5Q~a|6f7WNZ2=i&e~I!_!dN6XH%|HLPP@RY4O@# zn(PHV@bB9OW0kci_NGCZO}c};K)hM3TWZHo>;rNI*~IdS_jSY3dZWcG{J1D#MTzvA zHs|sqz0Ze-?{QL*w2Z=hE|(Q7(xwjn zFiyoW-Qyl_u|~W57Q_UCa^#5_*QPQ#R2R#SlI-`$y9EqG!oODkv?RKTwS+isF1{~< zK6#Z=z^ia!*1UNo+q7TJ7ypzEnBq zs+}LjpUn1C+v`EqJ)Do% z(=ESx)q^ePqMps%;fF2?G&$8<%P5btZ5&v^pNaRNsQJr^nidYxB9sIspd&j=o*&2?X1MEd)->ssYCZLNC(uwTqB9W=N z%3!HBXBD)CgxUT(e{0-q%m{xlW3l2Ho_Drjke;`Dv-DjxcLnowCdrFB4PiFX`XtmP z4DCqIlt4uIHe3O@^3KLm6&)}B0bG-ug8@INH3Z`&_r zob$ohXRNl-K|n>Cj)>g~Mf#du09AmFL7NMD@ZBwuhw7K&wbepl?4?^fF{B`NI(b|W z6{Rr;QS}QSZ;jgwV2FlTFiCUZc(G~?xo1FbP^q6N`jk~dm_78p(y6K5Xj1Q~FKX^= zM}|!pF2jjjWApSksspu0&2tw=G<_}YfnoQu(`TI;ohNE)aoRPFz?PxO5?JjoJ+25$ zrAHU-_2=a|`Hv1ueAf%swrFV<;rnrWke*-vgmn^Ky0d3ANC=|(V z2j%y=aK?2BN?P?gt+3E`Pg!rhssvF>G%ivcygqgov|#WZ#>X`=q~0G~g6W%)Jn^X1 zjb~wz7bK$+qz9=iMSyV0$Jiwv#SCs82q0tGcO1Ruh?7uIpA-5{pJ*sYs5f*@F!LXM z4^-}s6IV}40UhGw=PZ0H+*2MG36XH)EIRyq+}6!6PiD49&C}^r>jlEDR3f(m;wiFj zs}C9!cyQZXmfT2i`!UHLWDwB_Da}*ewXE4)bQn#?aTI>v7}~C}dQF{3eO_GCy=2cO-jhgwAIAi1{l^Vd()f9FDGo6;jl=_qH~wSz{xY4fQ{K*bP%=u>cD>s#W~m9D8`Eqc^Q_wLK>Y1v^V%c&Jydo5Yz z)UkLir$jC11IOu|$s=0pEx0DPog;)84yVUof|(tAfK;sn)rNHBy{^nkgZ9YTRVc(o|M2j(Lms|GSsIRFmM^lVdj36e73 zrVcUheun*2 zOGPC}0wVxC6!F$!VtF{WT+G&gQhDX!s92q&oHy}K1PA4KLxuen_|qN&Ts4|E?prTo z_EDmDKyT+4QeU9TCnfC~yxq8TX=?Ctg(#Kqc%} z%(O&Msof8M6++K4W14t-RXHxBIR45aVxYmw+V4KB04oaXybakgvQk>~Q!zCAg7$5y z?xpQXlPrY7*JhI`=U8()aNZ%Ey(FyT>nwj&4A1ON=p`C1I+q<^;_b3l0!)J6{X+o4h~X{8m|at^>ExPlzsW zXc=E*o7ueXP(yBF0=5$lK%wwKE?>fHIK_9@v)g4q4NhZX1Twz0Qt+; z%>8;i)&Pd!#UqT83Jt>_oH#q5+ga7XUPP2{Pk#H@m^Cucd(KCo$^tSM~ zQ!+L0pdN*CsaKj9$!wPC^SMTufXhKHdX+yhb|&0@OEPNTI2XQIn;r4Z_&O{(EzFG8 zl_wr(kgb4(4*7o0ky+uMbuJ^nk^-Vb73d~Im}$p2O_Jyxew^H9cb^hl=aEm)8I5bM z1hPV$=LuQ2c4S!4<~qMg>YFYC8E)pZkjjg}W|Oo#<;lzhtF}QVOrDHiE%QXrP5jhE zXy<#Y(eL|!R{wK1s&kxX+orbMtMNUM6ilL0i~aWO=YM9lMYV8XwNAy5oLU^nKojk|+`Nho0v56Lh4>Lx>*lv`OgEBUEG#p_N z&Dy6m3^oarR2@XoOHC&gp)y12xO`({wXXoBjY%qy-~8?$JM@nQ>AiiY=$V zY{vbzSuF_yr1+h2A??SPB6Ebo*AHJEUoJW|W||xp_NwIy1d)L9S5yy-jum`rUijlU zfHGFA&`I3D4d-!_=2JVJz&=mEOZtbnY?ecPwe@W}&e+_1&?Gdz#N}(fgQP2yuFoPT zfR4ij&~ZQ)m##bw>&hjQEDR{d%~Fd(5OF9{zOo`@w%|7)g5aA$9HZeU@FDoCOawFC zcDILA9{TRb2*wT>ZELq9ze*fy%^e)(HUaj$0}Lo?-^P&euhD8WslH?>zv|X-L0*d( z0&sEG%G0PTJ`n+}t4}?`a~Daw9-sR2I5GNI`{3II;;FU1xN8x?fwtM$OgnErf?r`F zE*HUDGdnox7fUsaFJ$R!({g{z1AI$>EN@XiPje8Z$9yz}6@%?hpdCgeYczbVX)#kf^8r{V?Hp7WvZxOoPh)F(;$D0UJtbv^u0OL$Dwn>l)vp(KAOgqSw zz>rDI)3-61GsGOKDA*N_#+!uiCvk&G9Qdv>XbdPwR$Qb~3F=w|Hb%RjiLcRz1zZST zpRMV{V-BovsSiRdwZ08JXpyvIXvxB0THxL^+4vc?<(D)Mnq7OMe_&5QV zi>i}*CVxeQS`Nt*Esd&(czv##VgWWh-Mb257xe4Ge9?$E`l}S^9h-J-#JcS)za7TTy^^1a<5asoKZgz|tX4HmhZ?CaE($dAQpn9BEWL&e-MJb6 z%P(?d+~|Buv8RELVH|iHRZY4?HBH-!=O-Kvvn@xGbQOLs_~|ZCT5b1pR!w}JwCsi6 zSeLO{4L~$XJUXeqDAR6#kJ*>nRI*zS?-;7M-rbTOs${+cx9Vs%{eASUZw*_tmk7;Y zJf-`e!P=5Dv1%=my3BtRZNB*~ReMHm@6m0q^~96|ha&(?xtmOYDQC`@Bjr4uugMKa zawx2qVkNA+^1234m@Jk3Oq%F(W?jp=x4Eas)j6iVE_0$N`PyV$2JB%T0xtu%=mp%K zAV{EAGNoS)E7AuagDWm3K|`#u!DvrzPz`HUlvei~oejUTjCBa4_;9y|d8WAhc~L#& zCVq2(P&KfwHAgnJaQ;dq*qNuYH(ZoEu63M<6k3Mn_lMR@t-SS=s%taUW3Q;W0qCsP zNL4H4w;q>6$I5jD#oU;eG``6U6uP2XR4s`6Xu70f-C%R5;c7LRpId1e^(a*0&LUt} z2)g|p0r#B1!sm*fpzp(yaKvXU(eq~-eDil>9ZtWHp;XiZb)Kn-7eG?{d$C4i8wZAY z7*sc8NsJ>C8QK$u?M9*V%kLTZEBC~m*=4Ych4BDh6O*WY#e4MgyCv^OPpN?SUfNhJ zmWwyi^?wK;r7Oyv?{yrjBMm1?C>>2{Q68t7TNmX?@MLN!~_o1Fabi zt^Jg~#_;p0i_}VU3TmG-PLp}{K=7*u+jkzMLoRZ(SjC**j%Rv{0#$*m?`(`^I)^TY zltQki@_x5*#!o~b4ouhTjN-KRbA#NOp26vrTI!amG#b>M@BjP}_0%g{EMeipf!TKk zs0u_2%QD0o$K?!|hg=}@SPYar@3g=j&e<`5pHH6n_H1AgXl)9@+}TP(<2tdf%?k&V zCos+#z;ipW>CtC`IX7l(B^+((sZMHm^oM{Sd&~MKRIc4K*ax#FRw}W<6X+zaYC+On z7k!n4yjr7p&HfzUhVg25R2VB8|5v=N%-lm>AQu|C{acLd zNh&ec*{2QNmG!Rh^3u&9LTVzDL#49?R06e!Gxmz3_~hZ9tgkZB{-Vj(azr!SOhKwd=iQ*;s6?uHb?E*0 ze74TK#AyFfEl-t8 z;`wg!8@XRRJo{za#!Vy_@V>g@7vW7b+FG2syAPTTS1&rW(0;T(d!L{d}aX7I;2{!Y# ze?mvhWxO2`(1^NXGtY(0KmPDsn#vzn;oFi!v_-2VI+1E^L0{KBolte#eP8)OYpF?? z;nQPUFm|?Oz@1KuSxzM84*MSG)*V$o@f^(YHc}>*{h&bzkw~Ydr^(SA1%9%hu^;e1 zKIzoj!yoX1JrD%piwMo~UkiOG9c*R~@#m7g{4;27uYS5tx7bxa3L46isEls; zU_Rx&bIM|NO};48**sqj$?9ilTWiMxmX{vOvo-%cpH{L22Do7~A*^~q;V&QD^mSo( zXN*csXTAeT6><=c%|Z9ybm}aOW^eaUFdyh;IB!jIoYl?!3piGR-l;#Gm6t#DPV~>D zi9AW%QiZoby(Qw_y5sPI1X3B*e~fv_e8e88LXVncqRTzLuRGm7EN=8YSqSPr{zTcF z)J(YkoOD<=Uv9L7#&*HQRG+gkpx{eMF;5ft2t@^R)cvy3wESq0<3y1N6h`5YNw&@W zikAe?V=3G3ek*Dad8*^3iBjtV9mrmiFJ@E_+NEX}n+~qm1&)>spjwI^YiPMJWKD+< z4b&Gyo3o`A6ukQpjaa4GArHgVYU=Yc1ZcgwxCp1z(I-oxYV2k1o2KPA1 zeiggyw=b-eABxL0fg_`pb=NmhZWn}<;oRBhnXgo;mIMG_zvp7Uo1)MRE~Zy=903baBTKo`2Hl>K&)_l2ri8aI3~iLe2<5+rw(BMTd-ojtm&|l_n44D+g*A!e+NW zzXjR~ZAGrk$6(j4B_i7CORq)-sYCXr)uYD>X5<5XSiSxMgMu^Z6Ad0~5u2~GyGWI@edCQJ+|97kTJs!^9H4N_Q!m5J1l8t^7F!1I+@#_Z1 zE_t_i2N$MATpU2}<4D!Ij*N5}P~5KsPRxi%XxXM0Bkj?~-v$v4aWnQR%t*CkINhFh zhd|HJiv-jc!~DjcdKKNyApGtK!h8hLEv`EeO{Lozz=2k6t5-+;J==q8Z$3B0155hivC{$=`ojZn%fh z)Vojk6KE$MlTV&cJ-vLffC(nE%G|q80;;VGTJWBagwHzi36M=&PY7U&v^QfFF@GB& z@V|iB=ZLt@CpljGe}a8`eQdfA0ogytN#^SSv(bze{r|B=JIFdzI0*ZLNLc;uml$8tdR zsAb+`cKpmCi=`q0Cbj3}!x?$Tx>U`Jh{R^Pq;`4W&hg{roD$^WCi$?I4V8JOu|_Y# zNu@DPzK{JBEcKMuzb#^I9J>KC0VeP9PVs*`tM_UDA&?-l%<+o-=E{$1|CkWg0x9si zw~pQ7;H}u|`SqyRG`{S?L#OA_NXV!qkSJ8-A%9YyG77hgRT&u}j_K z`C61+D$e!F2MbKZ<36OJi&C6SS>+?4y5ib5*)|8UG70^ym_zUc4$Bs(t0afFN6q6$ z*|phL;XwATHwDlf(ElCg$xQz22oFDR-{VkZJV!hv@gfrj3C$ zz}opc7LP_4$FBK&eK^Ux*Yv7au zdYoFn{T!t!kxWx1ob*A7YA(wR6a&z`0Mm}97&v8%ijp=uj{J&5%wcx;q1UYHQlaCW zkPHy*w)tOkSrqDoA`);lUo|x=a(MYV(E$aDKv_~3=dZ{>QL^j=rsuCqi4mE1>-S!m zL{(H{^5nj*%@d9zC&uu42;c|7c5*V^m`3vr!z!7=^x=01L~_o6khbFl*1kR(Bf9f+ zA;ZB$I+FjHV?`x3#Ht&`j?|RiWB!chYI{mtt;Qmirr=3kmf$!+s3VAD$wNKN0!zT^ zYlfKLjUzmTU-H6qQd^CZ*$-g<^b%e@_=oyL%k#Hyk_4BV>9gMK=`7=tPgy{mq+e{Dc7k(?^6bxQ9)Qq&u0YMlqm)I37Wgr+ zwLe(x*d}Ny53y1wnLOo~^u;eMih~0mb{CF9l@%w~FPg?etTq*uR`|6*76C}?We$Q& z`=l0AyPfrGgG#!0-kpXA}o@3Js7_F_ z@AKrZF+!MuiO8?ko4$s&VzKfobZFEAcAE7f+QgD{R~eZfZF%KB@kPT~C%*R1s3W!H@lsTGn~u*Z8XVSA)$to2=hs}t`T^h&sx;F&dM;e3 z(jSN&ZvKs3{4wWD4j;YlPT&s25q060*iHzzB39Urn|2dKn(gzx|0?V7GQPn=mqTP| zTGyjr;F|8fArT=AcMNBm5_|{KSfu(iT~OTj*7Zz-a6FSKd<{|hxgR5z-g!=@D8kB2 z$rnXgc={l!{5wP*4y*WPcqRhGUw(|kvHgzxqBonH6Q5ppsH45%aygdd2}TCMt`=Ke zo~OA{e7F1Q#GI!;F{IPf16;e(v(k+&A6I&!>VeWMoZw_rU0|g$jZSkJ?`VXezpfc+ z;JKf$@Y%AU^D_x_t-)5qJaFQsb1C#JBVD>jQsRj^^#$r%Xzg=OX^thSX1C~1qpIgt z+MxwI5q6yy#6t5JmV@IHvtCr@UhEcapBh1`AFD5BC_x3B!?J+G&&nt0Z44clW zT7J)&m2Yp!JhJ9PVW*p;(V=%^DD@bm>quXk?8SaZu|(LRlx+VTw1KWX2qrch|1r5V z>>az8?qe{_{Wiujr9H%q$Y|ZU&E+Zg8)dM&VbD)=^@O4E@kgN^O8nGcL)UrqRC7Ihmvu4&Y362;bb>m~e@-MUq zQ-Nimz(I0+v*xk*Q@AAlU*F)4mcR~*xyvmVr9Brou0D?=2c_QacV}9DY5_h$ARd|Q>25w;IKJ5<8 zn)HnVa0RUkZdcD>8$Sgpe#g16Y_0cy*!YSe27)^j|0{mOW(;QrX#02AnG*GSfZ-3@ zg8ErWA={*4DMVUsve%daDtZC z-ya{W92xYE_ZtEs6}RT2G%qe1K$?*T;Z(cm5m;4+U7KVN-+VV6wt8rz1U7k2z0vmq zglu|p**us=Dez;-fSH`XA6k0~dwui!^Vr4Li~~C}2UJ?^3fBSW=A#(eVyY@URQn|o z?3&nPYMi2wld44p$Tka&VL|v*OLzSL1m|gwOT26nY)q93{fp(uAp8O7wnOc1c5=Wk za{BE5fe@4fzK{xnC;99aly}F|*uAIIQ8e&3!}$UankXxeIV|mUEwan6R(^?-qoWwe zuUT(q`8f+4@O32KoHC#Ni-r9UFM&@$ z{Gb2pKmYXqf5JcCjeqz2|KFv5cW3_j{r@-b5>?Uhprqtq9UVG8VMUvq8RGj&@9zJON%l>X2 z#^272@&5H6{&j8p=by50KDmgyKUi-6?U4Rus5bdeT)X`EXQ_OD(klR$_J4X9*a41< z6^3AN`4?yY@2^U1`3n`Ws(xXF19|_w(F3c@I=}%mK1J=>|JzW%Jeg(x7f@{yUXT2< zE!f}wvk&6SGRep=Vilw}Y^4 zje7uB;<3LC5S{t^p9z$yQh%`ezku~opa9#5@9_bN02%CWnI0StvIM{kmQ(n@8I>SF z^#fBdqOuDA``PrT>3#Bc4*E|mfd4X5te9s3N9kF$B!BlteCH1YHlW;sT6ji(U9kUo zI{fQ%a%y0Dd=|Im{<}AV+y_k@xD|nobOSH{{*}oA^TY4Qs?&cB?_Va}e|t&z0~kJ< zt7^iBzkg-vzyv{}>^%6_LH_%B$np*tKH+IOlj7gMvd@6^U`T|{|NUuUAok=Q)(v&D z|NfQv0@gzk^Rnjc-!{>n@4)bbTvWXt&lq6`SCDnc<1Kf`7eC62l5yFe6~lcr#^7kS z8H|O$Z5=W=z;z4Of5&QydN?%}%cD${scedAeIW8J>@8 z)>Iwi^NF4=UdDWn7F&7`@Bik=e166OrtqENfCKu7qae{sWiNhd|3lXQ!{2mD-}&ED zJXNxzZoF(}gsk6vSBL};;6v{Ss;~o&sEO}B{~xC3e~iTD2e2%Vj3Hz{9c$kIZ*2F! z8~A^|Z~OE;QZbbbe(-RQ?p5iJBH@0oT5iy*k{HU$LDkOL;o7%w_m1-WY`>k+@FURP z<)W7J%m?cAsJY~Is|njHd6ZZJDfb!1Yrb}zN<3H^uz*GQf?K`6TOvaEDU740xed~# zBH>x!;Fhium)|gl{l$>K4?Wog{M{G4<|tyAyw*&)=1)2mCWwY|DZftsWi#wdn=+sT z!f@PdMdWp|p$zcX3fz>c%wpH~D~3WXs`q|3jWumO>y10ma9uqawcChQ7Meo1KW)bp zHRzuraXYg!loSU_>=75)b`F|(?!c8+PBe{)G_sRfd~BY$9J((=!!sPcHhxNIM)oPw7mC^ZtDKRs^T4*8pTUC&(pzOcN}Zqoh~@_0`lP)tict(DC8!^_PBC6iY?$^(%Hv2Xt2iZ8}fLdxUu zKI)zEkfejBM-u=$`}7DJHn`i>7zP)<(lYFr^Vjx7ZoHV{KJ7}rT?Z&eG1^}DN({q} ziFm&K@w9J^!H(I(sfqTl`b>v;{R}TE%rBavDteOG3K}0yzru#SAKvOt(rY=$LDQFP zStsH-?F^=`WQzM4ezFdO+AS+`4Nbw&TtBccABzIdJx~PKw@E+0zT7&zQ6Q+j=YG*e z^g=SaEnt1Ye)yz`t9PoRIV4Sp)_72rI!ZQ`q0^d%W;~UkV2&xrIH^LtPHock@jfkI zkRhhLCO(uFY6L}PY~=5YrzM$gKCudCZ$l#UD0D7Rsz|!q(|!!)2$3q|S-KeMa%P=4 z$lW_z@1>8|@Q9v&uMsB$e~l(hH3CBWi>o&f_|q?#0D;bk$ZdeK&|d zAl+w2RnyQPPqOElaSW*iT+I)!t(?n@XAEo;7m0>fUgQ#XX2!vf9R59o8vcm?VfY8=uN=ilUTHV3ckOX)>gIG0*c}yW3o|7(VxMV2Z&O$9_xnEvWvya1Bzb+3wJIxIgJmQz$*AXE4ve z*B#M}XgK`dg?<#<7SM!r{@c+j)829Sz7P50BcgXiGLvu|&~pySZ~Y)6_%cIJIIQQo zti6)pP$LQon39!gszv|uiL3s}{6z!ukmX9Fslsmc%$1qQ%1EOyx2x^EjHWiE{1)K} zU-jG&-7n2-(zLL~a<*PTyOaE$|u*4Adhi3S4(f(gD#_% zTGsFp1Nh&4L*JhkOk%TKE;Bst)XFeYG32BVII@d1>yc+eUVhk&sB+l`?J@1*8Yk`C z-(Eo!1^y+?Txsi#=Z~}8sGsOV`NAIy`QGax5mZ(>(1?9_fg_O<^CH-v zoIJw+W3a!2>#{ZVx+#Je-}3 zoKi!~*O97zbc>Ra*a*Q7SR}F7hOhS}^3WxHHnd6*Oe6Dh?daxYr2S)9D(O9t4SI{^ zro*(xV86MyuE(}U$STO}(n(O5!Q$9gJFfS2j)G0GTHv=o9DgiLTs%JYGu>864;axvg9fmJ|ru7pRSYnQqEZVaTNnP%?^aV_BA0dwQJlTWM2Cy zYRZstod;A}oCPm)68XLQ$WjnX$h0?~Dxb77CTVy#qid&LH`bMBA6}K>_y^aZ>8{U(=`nnl3=oX+K%YdFJm_hotwWL%UaT&1z3Bti{z`Q8F3beo>GTKR9ha%)nb98<|Hg^G0gcjUq={j?am0AI9v`-A-Yt|l z#3?NcZjN3hjo>i?Wws;q6QyDDV4N~vd*L+AuAdM-`rYHR|s9wEC z_7uP#e@E%rpB=F$z9l*zz~db3ftspLBN!Ml)-R#$yjWE( zj{c?WOfidvL&44WRO}tZID<#%C_0mBJ1diPJJ-lzx(Q-5v){r2U3j}TX$P^kSjt}1 zRgXG<|A#A&`g$kT4JfZDRhw(tIZ#t>7H81?z|q=%5U*LCXGF;D%56Yp=WX>H*!G$% zyycsLP$3PS!xHGitc~+biw|1$ z8CGjhFLlC19(}!zZM|>$v-3wdS%mXaMQ*0@LVRUhpT_--}_5K$xVPr7)|)4tzDRPK@uPjUuWMI6wwW9DAC z{4JJvtb%#|6aUDK!Q-I;iuwm2Vd-&o^r*tcbLqca*D6K8p>bH(kbmI6mBZ|FNz$;x zj`4%Ql%M)i+UjgM^YnrPD$x}l-_v!$7=UgA%vhcX=CrTk?v2LYs*UFei=<1a^87I)wpi*G>Aaso8j%7#C4Y7m&JgpoC#}x zoS;d-@@oUBhx*mf{zqkVzNJ5zGD8}gHCU9F1PDN%vafvRa9;CfM+A#ND1O%s;^gi}24ufy1|X8s4z z(EgN3fHkr#5~gSIQhSMmhyB*T3MVbn(b^%;(YAWc(TwKVbJTr#4EsqX>8+ECx}6fNU9#y%8~U1$^&W# zIS2}ikNJ}^a1<)~OYh8nid+!ad^Rk$_4l-N&QtAS!p~Ker<8oj_H{L0lo;tTRY!Fc z+~CxY9|Vc1n1H10F810>%CrJX+zIRJP)+N@ukNR$@kG`GZ)VGv3a$ONe3htlN)$X_tym)@061lH>AvLUWDac04nLePBhZv^&UG<&*9Y38G#HdaY@@ODmQFUm9t(2pRm4^W&5@8f7t&SUO6D=I+c)YT`ynyk}q6&_D9Usod1q< z_{X*R1I+#*{T95{Z*H#LaTfp-ZHsfLadl5cjVfXEa_RQgL4DhOvbe>b+QS^~gy>qw zbu!0=*978my7O5x@?}rFa%>K#qsI5_18yf%=YR>`DwWWAf`$~q+s=WE`|FmRUGNe; zp^`h&2Fs~)C89gdThbXBxC4TiyiJ%eZkK$O&?gmg%y#d?14VXN|!{H5k-dg~3a&yGe{bLRz!6+Wljr4vVy;yCMi*k0v)|zZB9d2DSKV>%3M_baIy(faUe#v^hDa`%)LSu8U1lJ>;^7p=^jM{;7 zO}d_6tRB-cwxwoe|M6hR5qyof4A?3NTQV`!4fkkMl#Fb|a6TemGoZYEg74CE|GW+o zVd?xyHq6r2nHZM*m-wj6E1v`Ir5FXuFiO~E??wDN5oP=HfFlF`vr~Z;b>PH_Aqfke zWJ?94iN2FA;$EU&t_EPL?H{49vEd?u^Yy!Be@vRQf4WUbvKd*ql$%Ea8MC`MV!ZbL z?9_!09|7l=iW6*E?NfWcpH+4mze`tJqQz=iwO*&Hds}z0!y&XuuTscuTSiA}WTiu> z>%c|${0T5Mu*Xn@ti1~x7sy&|9k@zOX;b2L0(LMn0+S$^bQZBt~X-TCVn{}{sppJ)nU6fKr zr+2B&3@UZ?P-UZBA;R_)oY><3gHN6{sq*v0_N4pww=78p0v}(4a%=E10pT31J&~wn zYAm${L;-2&(qj9iVii(#?cy4T(U3kOkJQHvaXQzL7%D7OxhXVKE(tt(Ez(Og@^97d zxv>r=(_B(gl97U zk9SQKS3IA)dC_*wvPOp+t8Pi^AYS&C2W$mc8|y7pK+%VFsRe7Fd9aICTYc{umTnwF z5~U^C;X5J@OQ(!7ufel&tNQ}rtnB@!M&A@n{^)$L@LBtWM*YXBmU^pO>U*VY5prri z3cS`@Bv;o?W$wExOs)%jdRMKD#~ok8GRx=K&-4xz1#czss*Eese2!-_r=Rfzk2pO) zA@<8CZoZMIyN`6QoV?>4mPj0hX2Y54lw~88tHcH!Q6S(bx43L#hSpUO4Z4 zR+^zD*l1=x1d4u}6%jEBf7h%rS=n^h69BaCBdL}3>Y@XW(CXLKP5ifm3_mGMhQVy& zs0%K#q$Z16=(#TAk*P~YCS4OKsbu3$uE%ZVt~X7DymlTcM6V5B%=P#oU0xcPwY8AX zPIDB$XsChtT;+D-KHs?DsD6Lb+>`h=*$RKgG3F#lUn2eti|B-b^x-rZFQ>J2INeh| z!-0~16Gk5qIq7|KPL_{oLur9-(AMTeIhZsNufk4LvJuh7IfHl( zdX`J)k#p{l~3#Gb8M0@3sq049QRx z(akBmtFI-pWV#O68&2>CO=NvwJ*PV_%D2BZOPlpc7>rB|V+zBQig%tUoTi|`Yz4qF zD!W8rudZ*LMwmXa%virl;Y;H~kWfL|4opRWDj4(+7zu2xBLvs(xI&Pok8jlX-6d3x*>c?8_U$*L2J34(rsZ-!9fGkbD}l!)|+BS zyZfbxdlI;faMI7N!Z6WTP>V5I#qZOo<^a2x)(3iLwv4~mu~O~D;@PYSvT^s&VLz57 znfh)$FFW1!HdcyPB#EJR)_D_4WrHh8`jO;1;Wzw=`{q8Zo~}3gZNc4f6AkL*E>5-- z%3Kif$uUa;Od-p~)iU7J@uV$LE#}yc$bmliIbzCFj3EiTRuPq7@qJ`8H;q5$>2d$t z%~Y%?@;8Kjt#1e!n}m43Uijy|C7&1sv#Cx~b&*7#^g*cKI?he; z+%F)fomrXEdz3NvWM9>dp>tD+ykky5s!Bm#4CVHZn-r2-E@$)`>!F2LJATv(0rlr7fK} z9yBAoVg(Mpw62ueGDuG8rZ1Pme0|ulJop@mck~DgUbFI6 z0jNq|PRtwM+Rkk(@V9n1#P@KX$eSPNg}9V3Utjl(mLL=;x9IRwePuo&vcZ;o+H(%$ zE}o~`H6qn+ZR9sI6wS#xfnQ33Nr5e=6CdJu0-Tl!(o1>;F>mt(;I&JUYSlZ7Nn~zM zEhz5tUa{!UQy!DP-!W!kZ@oVLB~N6Nk76(ggI;mM@0)Zyfq`msgip9B{jFhZ!+e4{ zyN|}Ak^aYgRY#AV%6x(Qj}qU-*z}?JTX|roTQB=9#utXooM{h_U2zAgww$nTvR|u) zG^QN+&uIn*OS&Ti4s+EM=NDr##K)Q5B$BC z$NeHn7IHTTNE){A7_X{mzwBR*Xin}0F5T*+t!>6@M+MYh?H^q$gP%o$>%2h(38ias zAML<(OS2ne2ia}uJ@~U|)+-p6$58T2cdEK1Zt0c*;q?(9g^rQC`Fo(QC-E0!`rsIq}nwe zkr`PGk_}U`FH_{+WMqp6kP08c`; zH>!Lq$|VVP%JIVAcZY|uIe8Jc)X7&=b9uyU6nv!`B13LxjB2rJFDMB(wj%dBAUhyc zn3w&uBdR&N7UIu>f0;i%TgGY|ru@*v?CvACB=B3%MOliM<%&u4Cq@heA#4fH4uY9W z)Cy-y@7%pev2MS{k)%J}eQN@NS<`3sx0u`!aJ*++^^_+WAM&JHD1qo?T0cntu29X94 zBqXIfM7pG72q{5nkWK+fm1d}62m$Ht7`lfcXK4Nx_l}?U{_ST!_kQy~o)^zCFBpac zxUOrhbFFio@%g?ifvHcHd^Ksiptp)SV}D-HTo7ytAb zKlsUcHFj)P^TDJeRLCzDY&$=p6RQ8k8E-&v3RONZEf4_RN+1y^X|*$p-B;i}GmiIA=zs=2HFIf=>F(=kRs(QE$8$Z}MM zlUa9=U0t}9$1m;BXiB$P)&Q|1X;=&XOg@3=&SYI1i{=j-dXZ82leOoRPSGhY4}+s9 zFJCoCU8u?~>b5mKk9DT2;#aI)vpO(+m2g-a{V01O&*|qz0ZDty&%18)OP@SC8&x`# z@ghbBZXpZT;WY)*ZTc)ahzj7*a#oXN-+mo^w& zd(pTLD~-tD0W~e>A4_GtXn@Uc|6DFG_QQ7DiQ9vicPHqf1l)GuwdfgXyfdYdR_FZY zwDY&mh3D}O`@9tb;=YX83Ir!GT6Lx%OnqR<9^Lp1<5klL&Mlmk{fYU>M+Y|B?Ro`i z&QFBQi@mOg=9|G-O~<&H)msrG_^(Sg1t&}$3kELjaX}+jJg2Nhups3|wK5Ke;jZN@ z(}6_MUqN2)e|ym$A2jd1RuO^eSczb6VtI0w3Bc?HTmAKXj99;rsGn#px!9yuJ@%1e zmkA~2^-bFK73G7ZUamJ%9%ku3TEn<1=u0m!rgi<`4S7m9+uKvUU)^Ww&?_sjb15tY z+35}C*l83`CnBiUE^#rgl@i`LnKEsxb#_apn*B0@7l?I{GlIEUcDYQMp;^}HABL#$ z9qfeIV+K$!PWn0hxcibZjKR*esum7EVNMDVQA9cFpZKjCe_8DZ?M^s7aXB|MrK0NK z5=PKn(9ILZIZ@HqEv`Il8gg>nnSBw7O_C8zn6-dUL_?wfU~yC}};vNcHsKp+ERp`F84w>BZ1yFsi7pEyD+9zQXFt z-6kQNG`pClyp)FpLyGJmEn8(R*@QFkjH}bVCLwfpa)(3SUm@pFtviZjI1f2SuJ$>- zKc>W@^NM8dRxr(by&H^)&9r_$X+l=pqZZ$v&2W(Ab;U&KDx5Roxp~MH;+A+GvfO?< z1Jb8oeVybOFc{8egp81AJ?dd7#Bp9%A&Uq`ouEkOU-0Y@dbqxuK{$yXa?R6t7%s1w z4hPhlWQWX9Lx(;(3eM0sCF!I1!$kM=<*|Dm8swmKd&%e1tGBwz7&@d}k@qmEjyt&e z?FHPn7;$G=)vq{ZpI5-0#LcJKJ0-gPXXZIALnFyZEi4-Vij3}wUu~ONDjGU}JT+xd z_o=?@EjIvnFZ_%hnfExBQXsr6#BLJ(5N& zo>L8*x0Xj+<$k%6N*jHMZ!ci9E|>eUbfQ_9OCt8i)MnV4P!(d5gI%fo=1{EnRgX$# z$nIeDdfu(HB|**%NY*~3zXD^ao;40*}8)_L*>+KXsQKj2!Vp!_oFg_bHEfK5?X2HKM%zf={GH7CR z5603NL5BBbqSQd@ihsbSuU@+~6AdFV`L+ZZ3w+E^>JeR78aJ&XMa>X(?hUg;U>9LD z;uVa*MG547#FPeMwTH1`JV@Ra>|pqg1oX5zx6cP>`NoCC>dc?a+wutlls+bL;L6lh ze=N*Q{s_H^=Oqj)=+7!M52%;ABz@~j{mqEqd(ny#BqK-iW%A-M|y=4{oEROe}^V;9S?HWhj9aLyl>O zo2p;gOL`CH51Um|p}5ijw50dfwMg;owl#qS_(~)QpU)1?9VotQFS1vCIkhgsEyEms zX}8qg+>|s?m$YXe&Py0Rca`6lTyjVvydJ6Zb~vDQ5;%mix4VT|IEaYYh26{NmZgj^ zY}84f$~Owa?S;I1y4l88F0!{{xwYF{eu&rrk-tjr##Rim;fL*Q6as;(FKl!JX>OcQTpo z1Pe}~%Sp03Q$N-q9_Q_jSPz)`lj(L>nKtyEo=zy-atgH`4w%w#5ILkc?fuCgR^;SJ zLxg$q9afv*cv1s8_h~P#?)NRX`DAa2fp21t-oOVaJkI>6M235JWzu~(8_Q#~8oYav zd^QI04(wWHnYlzKBq5+2soaZ1ku=6=fe2+)b*&2K5R_~mX$`)~n1zPBgi)5+4llKo zS?=QZtlX}}BW^PG?DD@AuKOK>uLizH;D@Y>SGhpPTb>UNAa8d!s=}nI*G6&sT}JI5 zbh%q<vEBPNff{tG?8DX1c$?;#Q7GzkUt; z-7s8WvJ0!iBuFySoh)DU{aj;+LhFv52R1x0IN|x(5rHFf8h0ZAPqhgwi}Hzh_qB%)RzpC%`e!yq>*K3uoI>jTy!|B7BS=D zovjIXjly^BReauAfDPlopUDUsx(Cw{9||Uwzfs_r1<)aqFA6sAzvJ3>D4?obC-hTG zbg7iM2ud6`KwjKq{*H_|^yDcUS#$#^>)YhC*PIHGt0Hq(+znWwJN+@Vi!k{&kOn7_urT<1&}K<*_AR_%%)se4k*`T6*cg6vQa{9bsjxj1ohpo78#4< z-NxMECb^49+hplf>)aJg{$qc#7!SQ{oLgK1b9l;lJKA3-Uk3N9@$ucYYqVHxr}s&- z%$gMqJ4wa&`Pmi)J)dPakC@VTXvn+h;CpGwiP7)*ae?sbAUPa=gwusiH7!Mt@MN|J zUVpg+Hm2s$m!kDGy6WZsIU~x}EBJIGC`!H$s-yOdMRYR7waR8@Xo3Ts;_!!Adw`tj zG}2;fb%lNE)71G|lg)L1+4(eI{alYhuH7>Q#lZwOT1Ii|fC4p~LwQGkxIC({mYAst@No(Oc#uxU*!nrM=` z^x?O0_~bh9qI_*uPPip%0AD$1iJPUjMrG4^ezwPG6RW2oZF?=u8)LKOlcbTCd*_d9 z3|&c|_V;Mt_TF4{I!C-bd3G#}F=jgMr7GH+{nW10q=7G}RlBz({Q zKxvR@H4em`ma)DEH(y$D^Kn13yA&;_thc9$2As$rkjwYJH0d+uZqD0BI^7zFzPSA- z7VjLdSE5Q!?E$y(`)BHzB_>1BtBF$Shi_X%n}2)7)O$6nTqrx{2plNA!h%`SnN*g0 z+0dxJ9XVhj-akflwi=k%F{w`?K3_PUY-hCwLf1ZpxA_$yqoLPV`WglCKBzgA8oa-o z*Qr&G`5sUZ!qi9YsdJsl;}$sk8W75~%(?&~V>!X*Uhrd%3diftgRX+Ct_fp*hFXU- z%v6EICP;%&k`|;}{L z5`JD|W21_9`S;nhCY^>0V1y}Plk(XZ#>&Pfrv#TdaqFRCh_d>oTDy76Y*9^u%MT0O z?k`fta*%V|E+BUxG67sUCacyHI)}fdfNZ7B*9Fy>U#-NrQ&Ep35z792z+{B$YI58E z8Kg!Rh`~!XczU(J0aD1EzE_?DS2-4*uRh?Czr9JpZhj>{K1NAQG6T7@!F82r-&e&j zmjdxDL9hAh+Y zs?}z*>x;^Ygk$1}P?Pf+(E~i`FUI$!HPXi-jO-;GdR6Pd-u%AQM~#N3%Se1d-Ms zq?}qtqwLSUPFfO&FUzA$-A@*^$0rBG>%0g7IfKmRjOi!O#m_e01CAW$&z=RL8zSY*PFr4p`>d2I!J=QY zJBZ>%fGuxr(QiF!JjA1zWtPG?mS16?KNT3l8@&s4opkWO*tBF-Yiw)^M%_ldtjeDQ z5Prw0$8r4IN(Wo@q`AlWbA3Se3GGO>o#>%R zGkTW9sOGb{M{#Ti7={**>kjanAn^6mQN60`6LEyl6*`*KAUxF{J+uBkP@qK4obZHc zMZ~w|7!yH0FOck2Ss*vXi@n+A2rQA{A`?g4EwWL5E0>lG9ug*H&C?a~%fs(f6srIn zo=@vzlocX?q&Gb!ex=6wE0w^nb1ER3ixA?>bm^aRR@gOAYt$hsQf>P7nN9zdp-88b zqjKQV^?H{7Rt+o@jC!ZxgJVu~nb zCE%>zp!Y(+IUGt&p+iwGLC9&_(pVt1<^~+ZcHITH7^|1!+)=$JM9v=7Poij0T?lAa zj8Dj`l6+k`pk@D6y_@e(r?}zOMi&ORH^ZJl>G>~Y-?m$3=w-PX^doVj8^t&K?X}@U zAnh!$e$hq8V1Cx?)ei*n2WUeaISOGZPZHSVu6(>G8}ia|Da5t%6*Im<8XGLb4Se=r zz6J>#d-qdLy`9@y4Y*Pr&+QGJy1r6Agf23B2mSLXdT+@Yr^%QrXWe{-w%Ov_iE3a;9ACTv?m`aNgBAJH{Z1fuN~JIG zYfRg)=(3&R`%q&;t*+=6u)>cHXq+41-J5XBQEohQS;o`N^W1M$t8tYtQai;TXk1Hb z8u`lMptC3Ur@n+Dd6S)!E?%H8tqJUN6dRfclwA72M?SrLwej*u#=P*<%Q zAdTWD-HD1Akw1wdMQiIqY%u%GQ01eY9e=h^+S;Zat?=Om^}m?=60B@6vcC%a@Xf~Z z9CbqIG?T0ASv%(G&g$59bvBwQCKEOLj_1kT%(gezSiqEw+uPO?w>8gGoca&_RmA2` zY+G(EaZA5RW{_=)bAC{~**|yXOP#dV_lW5pM&ObZydiA5@@cGLz5*9e$&AC&vIBZu zS$a(SK_|p9P5^C z37ABzP<$T|4nY;x#hF>mKy&ErwyIlEwn3Q3i9q#%z_@8E8E1|QDg9I2gZl?sG5aLM zHlsF$hng|QU_(Iuyz8h48|{rrlG^Enuki&^Ar$@$aP6A27kO>_?}!napnfPhFWx1c zl*G(LZSl!H6bCEziy%ZfXrF(pi6r8sHEFE7)U-5_TDGu1zV#Gj60zP9FpeJm5m1X3i^(+K3hrXI!r28mhBk%5XmD$a^u9;c z)L_<=qY5oMBGMk0Ld(Zk%_6AKkHu{WUwltRB`chf$#cu-@}ExWR#@#j3tgKdgw8UbpTAQB(`ffkzeay->8g&<` z#BF4nIQi&~TK%dY9eup5TT_0R?khHYs?b9!IPjR-I&y99WxAp+6{}wjq(M2_qxoEKuFo&$-liQw_u6q%b|VuFbNsWb7p4bdJ6bR>KOPx*e+@@IX) zWf961Oq$&!(5@08S6>9F>(WCu-PZ#-t~lczC_ny&2IV=6z?4+N!mZI_I`daas(c{l z?nnU(C-fsV^OCS>sIFd(6}xZdhDBrONABGzc|WR9!sFf2EZS#LjRSLyX!ju=zYeJEO6gaKesqxo3HEuZ!y^({VsD1HkY#5uHvf>G8W9d}8CAfL;i@(1$FV!;p*0 z-eqB8`uB4l2-nkUyh((Bn5ZpWr8Gix(1LW|duV-seD`2|sFFM4T-gwiU}4j;CBoqI zv0t8$r~Le^?YrOHw9P{$36a_|7uQ$0E{{4me9Vq-;Z_8+Y~9AY?~kQS*jV|t&d4K` z&X6$IBs1fwRv%MQ_CQ~j9G}!>H_1sfKLI^1V9S$JOkYDBXK}o4mndDFZSz3rZ+E+z z+2JSq3V=GKH%l%29cG#oL_TWdpRLCJz3f`yOWxZ8Yi+UH)Vk<;Ga z%$XtxvPa#f=;Ix%(a_`Hk>jh~s<-c)c9T7+Y*mUkoj&D!i;BNVhUUo=R_}6ycuFNUq2gg>KN6RNfJ>71^??=E19 z+b3?~X(kGKJd5Vq>-+9VU$UiCF<>M5frCj$6!jB+TaA%*BY(S=UTStM@J;pJe!%-M z$#Akpt5SvqD@9!06=imXJLMbSp4e}6Um2`jv|e-rH|lWUsX@yQ zCi(wg4%C8`AY-A}eB9^>k`iPkBX9o*3}DVsUr zWNwn^i}*)Lc+#)AW@g!6y;FzBZlw=GU<<}tp?)sPbe?Pj$6$e&azXx z)kjqwGXrtHri?N*l5Wp$S0Bi)Iu_>IMeGz+&y>M6weoz?e@d9f=%LRpd3X~;T3CQA zX^Yx=!IuB4Z*$a2dbOP%OO{6a{d(R5ykqjYhd~g;;w~8Fy%zi0EXxVGI8(E%q$1!+ z?S+#tz6%uX5U+L2V-7*Gac@D>-PDS6DqiWW4O)K zR1C=YmBjUjqUld`JJioGE17eNm82%k9u5#8$aKy_wv%LIEEi}%P3Nb0^_g*}6WdLG zn`Lu;mv9eZCAlP+cpVViy8tptwjO6@IE3Cb($r4RVQGW<`>IWNbb(e zVx${CRu{JZ$1YR%8j^FY~QTFmyS3Mp@!mj+hJ5mT+x(xDVJ_7 z79BgMri=A+4gNC#WBIVA@FE8u0EvcV>Q-%X4}6jPc+V6SQEd-TQh~d6x$y`Fmbn^3 z(cwpZ4dI}UTxIa`u6XoU#g`Nn!-myVnKV1+rPO$@jJ>VG8-(^fj$k}*UuyND{r)P7 z$L!673FZr|{gLsy+LEoV*4QEkZ)9rmF7m|}Y)6j*f+wMa){esOW?Mf}8N2S~#5XHQ z^X($_0yPnS8m_1<4mJp=6 zBy1UDf0jYoh&~8J{8sp|9BOJ2Q6s6xrNS938|h&9ax9Hv=mTN+;BCuL$E;iil54oF zAi-9zQ$yqtMb1Ye=gx?%`)VbO!Grr@KDhfMJy0+I#r{$kq9Duzn4U~%kVhv~l$UWF z5a5e?EhC3WT}s~o53UR4_AOxgdTWe6AI+R38im))FK_RvZkDwafyp23eA)CCGZI)9s0dW znnI_grUeL2SBX{>M5QhS0_&-mU80=HsfyRYLium6_)%riH`WCBOAM5w9v`{@Wnbd2 zQ%$7(`3#=WJR+PM@9{8F4rGwO*187x21fSO)ns|OHdzNx=sTDS?Ebq z^*YtMxDlF5#C4)QjnnKGeEtciV3x4jH_gGY%Ini4V=2Nh#ndH0ZP1s+jL;9r9kni5 zb{?Zhn59socG+{1-^$?c@OOa%&g6}HC(CJ^T7r2#m$_0OUZfSn`+!q>npj7rSvr;( z_t^f|`HzOa+lAN!*xw>a^rCsE;xR|cO2veGqf=8;Z7j}@YUjyp1ve;z5kY6ofEa8j z*mQ&6q!+iNeW)}e`uxb>xfUOW+ggRlc(mH+E73sNY+}L}M&z|pCJV7VZwD=CBU>}{ z_GJo4?6GEy^A2?Y$s&JIsQEzY9t1&ZOzni>{YA;D6FcbmA`=RCXw9>ZYWPgpp%QZNDmiGaSq$b<`u%*s0 z=?x*coZt@|N^{*y7S5;>2^q6udfLgP zEa<7|u75o?(4k~D zX5iD^yOO`ea(SrPNB#DcMQ2Y(e`222Nyd+VNprq`&-B49{ab)3@m$3_KYbzkGbALO z*uwS&VZg7|2T_;bW;6CW#4K7;Jb(&Tu-FLHea^x=F}^NKX!m@>%rXXj7qR>Jrr1K* zDB68vILZWCmpIS%s0_1p6)5cf%7yLTq7eh}mzc zx2`=jOM#nxWAX^YI>j!6InhjTprzW7_VbW*qDAvUd){B38eC5l%n-rDEh%d1<;pd}2~l zk@PFty_|eus$5}ME1bs*d;htRIZR+!D3b;wsFe$Y&P&Y9=X}g@_Ekd*ZNc{$kB~Vu zC)u=HrY7(6wC8HvFFRCsFR)9VHyggfhOtPMccbqHAgjbSFxUr;0;#S*)qIN>CB#Mc zpelddg>rHriwneD)-uaSH_Ja4l}tFi!w{a#srg76g%qe2D->e!-TcDgP!PMmzZuUU z>-?75!0CLxQ_(IY?)b~*uzjZbWb+rzJv}zrpimAL(TBTFjsH--=lwC+`5K!ns!oCTok0QXQq`Oo z@O?bXVg}=zj@7KvN22tkhOV(Yv-R^+Gjr(X2#R{pI#bL6fB0^j2`@9T%2r-8Ar;t*kqv(8@T^csXSsiA63|*tqE@(J z@15nm6Fph24uV_@BZD=1V!9PYT5XBM~$q zu-UP_0NSiKK#D1XGBGN4X5E}SMB;wE+>+$(Km-nADL8vKfw>GgKc2xgKb6_X108gl zmbIhbVJ{F;;d}R12H2=Xs>VRv#s$c&4!(IX!tC{%06*;ojPtQK5|)ig9nwpK)gT(( ztBZT^S%~oy&a+jb$s+U9uae8-fnwURP>iGPFEGuO78L?NCJG}jrOZb>%7~1J%XVXs zd5DR>PBL)XJj3`QA-Tz|4ERbj!2!Nf`c*asJ+=mIF1A1=d{ZOny#tJvQ%gj;SM9jsZF5?1bWxP;2r4x4J2f9(|LCChRn03y zqO3p=>4lKkU_D(7rEw!G(PmGat#c9fh<+ALL9jo$EHVtT-(B^P_KLp=SQ`R8Jl5IW zs%Czg@Gzhf_4PF%tpP@DA<=hKKbRnnD$I#5$WekA2_c0zubGY9#u9kT{8d=hi-X_x zdVX9hLKpVR8Hle}=3AWv0@+{S@I8P*L5qfgH<46^aIw~mdIQ}X7(0i!ohzRn^yCib zy%xJk@3=3|>x&GZ+wYnq6WtN}c<07`r zNon>;Nq|$mom9pobYQt2|G7}$#iyD=J^mME0?)Q$4QB+p%~Lq}TV&j%eW-T_H46x+ zIUDz06oLGe>VOk}-0Io3G`3NFggW1urtWzxKA&Zj9ncytc)2?zfc+{I-Y<%YqJ{yL zEkOrGbx>=w1ys)65w{HMo!-~qSLiYtki!ySAojhZ-~1VK!4+ulD|TLa0ci4MJ9sp= zZ9CeHoE`y99X8iStEt$_K-=o-ab!uA$RXx{R3G)YIYZOS813Eq%8L6<)$e`w+mrri zw((sPy_~n`^NwfNe=72NhPS%gi^P&rsV)?yg{L}v>hkx(WP3}~b*ZdJ44%v;%|$Oq zKGa7&Kds*Go8$FgBuyDGQjM;<-HI2WI=om7jW zk@$gvIE+^X6bn4rSda86wg~L$*~)m|0-9>nm0BJXTB6bndbzsLB6X$*B<8hhvg~*=d2osS3CRm^f#HUKoow5;njvHYLX<-C%U#8Xo`AjZ=mv{@!*~$1+G(D zrdJ`C31B-#O`z>(A6n6F_j z7331`UJ4f|>1@2C5_fKVc>f(7h=X`Yb6ju&al_p~gBqWgIsGh?=WfKZuV#R>A+wFO z7Kbn`C7J7oEK#o);bVjX>wTH*R3gWO)dM-iO(Jrk?%zo$i=nbnK8$^z{Q)k&Nf{8f z@hNG4fJLQ(@ACpjwhXm(5@Iz0G{P~%-(l`k^!}scBO+h4MJ$fZrw#F%u`&CQtkDil ze50&dDdb>P;}rE7$kO=d2<$L8J)er?Jq#j1+3p_#v$%C$2rBw!H!|wo=d?l!6Iyq3 zja6Rhw`SaXcM&Qrfv3HUzKTT`6z=i<{?e~7)7zy^WCRHbc?8kvJz`p1d=&9&BrQQt zEjs(bqtAn`XX{~XrQA(`W_Xc$X8<}FYl3lF3$~RDz=Sm%RZq;NKF0+SJ!^OJf>|7 zFZ)0^Q?JQK&a-7rH9J3-=9bnTY##-nIC3%WFgkcF>bm(InYJ|e0s-vlQ=su}OH!u8 zF%LCU%NJOZw1+3aV!JhJpWkDEXv)vvzHy>HYnaK|Hg8y7{EEu@Io zpK?};MZR*{U;My)bWgHA%B=whSmnk=F~NIOwll&kS*hopfQnH=mG%JV@KvmXw<-y` z6xJeH?_1o1?ua54=Hh7j1lEQ3XXuDTo?RcT*UnQ*Q_es_D}rP_Id!$Mj&zkDrjbcJCk0ijImOQ`fGm8;~o-ZK+q^iBV zR>vmLPtEsR$sH^Mptf@p((_a2nJYb@@NYi*LdM`<>Iuc;#MO`h&UhnT)C(Oi0#M$A zH;p_j4Lqv%+&dQ19K+7$C4~^r%JcYBJtmrr+1d;o@AF`PTy2Ol30D6U%oH7EC` zl*Y@G$ZSgDujxJARM@t6ihoG`C&0*i^d$v=k(ci2+yNGlb0sF_G8{_Dc;+*COYD4n z#;sM;^?nWmDPIit;W(Kq(sg@Ua^Ozrds0V0I`$@|FBU{d#VB(k8LHSt{L(yqt5MwV z!Xtni&qWx}Le>Vz;S#^38GHu*$apI!pI!hRg=&Q+SJIL)3_fint*PQJ4)mr3FAj65 zs6W4Y(8GLhriqa2AploVb4n>V{6-}%ypi^II{(uWvDAJ_k#AQ0JUEXkznE6MZ891N z3p^ouJGdym*R^;MId+OJk!*hE36?eA17ff-ztWh&la<_r`NqsnPp#YDlT`cXu`r(V=BY;M5)iAnuT%{qQNn4ULiRtk-b&mA~!-tSHyga9a4Yy2YFD`aif>ZHu)zw zwGDiVjsg$U6-!E|&5j#u^F1< z`NRn=9E)ZdUY)ylOC%Z_Eqrxk#V-)dTs4?!T_vCKu?D$sh*yR;8-5hd9w z5Uhh0HDLs**5hC*&72E&=J_T41V;j>k`OJn`1NM3+&Z4N5oqz=lr(i@GwtZjJ|+Rq z&_Rr4_iklAXzfZuJQPEN?0vs`jNA37n%F`!|MOm?pb72puh%4_pU^3eZ!sAJNzUwE z3G?TL>uI7_E<PN%$_nVDmm{D=tOG`6U49QX}K7D@2A%|6c&oH2cRv{5W_H6HtxzX28;2tv@TEia( zLcJVqQALLRX=!GR>k8@5G9P>gi@f9uaEho*Bjl%D<%{ zSfmJ8I(<{a_tShPyWcC@i@`b<^P9uUlu6;XqNyVW%NNTj}Rm0yESkptZhS z$Ytg9Vc_H{eh?$v(L#3zXTAy!8FqY@g>=ZZnWW^_Jq>!TZ~aENFp>&LcoD za=XdVdD}IZ1n8DRo{iYVJhW(?!3cU*_NKe}3PSw7k;wk+#+y*)8u(XpmIBP#NAO+< zLNK5zP*RpTv9evwosj>FUYXK4Z$Hfm;@$PB=WfPD zNM-qRh7!;W_0TeKnQG~&DT?aQO60atSK9LQpz)Coni7;h79(>N_s>o&4Avpw!F%)4 zhO8IsVkLgysA7#UmYaf&6@wnl;)_pdVuy4iUkRfHgIj)mTd2uL(RXluZBT!qYlFvi zlq|JtXbLfur*aMFIFt7KR3(gE*RhG!u50mgt)*J)&jlQI`SU|AC|05wRNh{U$X;Aj z^HhL!Q{jhxBg4W$+N0zU=ew0!bW3?PEhZ1xW24ty>$VDIbZrNkl;9`Q^~T>2Zh3pl z)&w;mn6Gh?VO;`CC}Z2Q`A*7J@`I04v8>k3mxEshUcw2!^=PmMQ9cEYMdAF^|K|Fh zJcgz7HM5D(vJPIbexUT3M%ZWj?yor zZF`^3m14Zl-W^ChUJFyO#hh!aX~R7j&=A)Eq1*Dn*U}smtE%eY;*d-!?n{F)f?qC4 z#`)8Zbrm4|M}nH0Z8n5qG)o`Pa0cjD`38fQ=E~%4r<#_PAfN7H9H-j^AvY{-s(JIs z(MEZV{UKM>@-}IsY*ehV<2pB08X8q}dwuRZlRR?{(H6HdNl8%EN5c<2I@6&4GB@I6 zfw^%|%Xsi|QG|%t=y#s1;7fGwU-v-r=ikHLTlHmDsv_E>4*SxyZ1=I(pFWOBMF+Gu zMkCM=34T~6TD)4f4oD(-nsco_-t%S)I^UiJ^nf2F`w_2y+D!SZr9FK5hurJ$er}u+ z@UJU|LO=j&zdz_tJRN1c01w#tb0ElEs{Kxf+oRY--Vl5$`5ib@^R_3O{8%2XR{6#5fXDW=)_h7X%gU7~*yrC!!`Z!MLLQ#nz>`*+r^+%~Y^}E<)_OeBZ@cMQbKtg#fl%mL_=jn zi)e4C$*L;;sGshb09Y&j2Cnss6zF)|nkZqYb6JZC=WJ572S5Rd>|T=NKm0Yhw?$-; z-zv7gzVt51JxZMBWBtAfpW<%%Q#SAU9czFIh7c6XMfeMI`WHdYm8*qN>6NRW6pJcR zyG6=u2$kQeaTb(JEpbk-jI4op+=8z#4T@Rd>6) z62Lz7e9(S&H-B^2M)U2-;Fa^xOVYdb3O$LIYI-9FE+(_Tk0|hdnC6Co+18m>;L&c|#EZI`id^hF!g`47SiK zSe29S?zXR<=;l#In9$+Ft_~%~UkTa(AJm5o~>{`Y$8v{=N+V?*aP#lNb!F^kH(C zHSJ}@oxeU?g>Q^CL*lti)?b}_{jR{I7&xMAKKiRS^!)$3|D-_UL>wrbwiNcy|LYGx z?#_g z6Vf*)^Qyd!|3aQX30NmHm}qbq{`wL|$N|{!P+ZP`fz^L{<2I%Gut)qCi+&Zb>d7j6 z+WqS*#qbUA-uj@NTkU_AwcoAg-@PBY*(W$y>;F26UM~J-1@G!?mHX?%`u`8i|E+=P z+%LcIJ5Uk2I?pl!^Ve?Q*5%Pk; zI2y%{a6Z)wWpVMdzqsegh2Cu5n_qmdNA}fuM`oyedoP;kM?!XcE|h-ry8qqO{O2St zxG^{IxLFx@TV-@cV02U_rFw6X1E74 z6+8+0Yky|p>CN_`Oppv5fS&V`{Z&^?jvlypdGxjQ^j0uUN|l?~{d--ymKg~(^a(#F z)A6n4-o$NQJY|fTKXmqvzP3*Np3F`x2S(S)&zq6$47=HOo~JT3pD`*)WIlaRr=Fs% zyb2i^l`?tjgUua%lJn*6KbGY8;Q;xVG-l;WG(H4D<9{Jvvn!3+W0cGu zTIXz@bm#O?20qcVxX2a#D~ig|y2NJeNlyr=!GhpSy?ue`d?oZI83+{4=FI|?q)~=d z%n}!Sxd3@zSELl^G%;{$Umfy}oSG8=5(ot1v${9ob@EAQl17mM-5Pg1$#Ztxrn57O zsS5K@mzj%GhuE8hB{Wy4#YNQn$hOlua?7Lp8o=_ed)Q?DY^7HdRml`(Wz+sX(*Nl} zgQi10PyIzbkZDhlZHfgB5ehc*Ah2*I)uvVyWIw!|jvoP$?H6M9hAPvJcb(y-tpIV` zF@#=u4p{ives>R7W9_S}t8<#JBmwZOvjQjc#i-y%S+BT9`w>X$UykQUe|AI@DqsiZ z{~S0+Ju;+T*ET-@(uT))OohG2$8U8{R*8DBRUX`4aar-;3PK$(g>xm^{-}H9y!MOf zrt2$-&q$#SeMyrkv-@$5eXuZ(Uc870epeQkN>%?RXB7I*0lK*ikf>N96Aq#e$Rm30 z=W`kP{a`;pWo2)X%gtA-(8P~bg9T~;>Kh9PXrIQw#=Ur2p*Uc#b!Z>=u*o*CBKR^= zKi}19EUM|*FR#?B^&@>s z^VBtV2HU@YKH$h8@hd&k{67zA3&uB&?dJ#mZx<;7alIor_qTpLrhm?izx4tZ=eR!u)X$658)P1H zP5T{G@~svTEguO@20A@m8<4pEy!4f=3;0UE@y^F|s?neVk_>(41nZU+Bn&B zq8>8`aGt;zGKL`TCD{JrvLS3nv*~DKfX^_!aun^3mk!832!Z)a&_$!RxI9ZuIBOSt zy1rOk8+yR`4}$c%mUE-=H*PDHHJq+dn0No=@vcGhqae{(dn6z!3l$|^Sy!y86j>d_ z{y4?lVbTT78Z)8o)#1qpqV?vx-{ehqBdDSm(di-Ei$CH~Y{O!J{WEd7@9&!f-T3$8 z_Tv=jF}Ad8WR8+w;M0rc-mizVVvlpG0N&d*R)3x-u~@=No#@p$A~5Bv&T%^#?ri;H z2fOnHc=;eC_0%4>`?G_gNOzFEUa~?z&-k-2l0j>~xXhVvI-Bg@eSl#nIj0bS+=KaF zUxH`#8)AL-7X5DeP00l;tW@LT-;x$xh$g^k8k)&{(^2JWa3qYFo+Joo18 zd7~_f_RnLe9PiCuB(nxzR3vSJ`E1)4&18ps2qG6v?jq@0Kr2ui)0-ww9N9XQ2Fn@I zPs?uIx)Ix0y}%`}Y9`^~Ccp#`>C`QZr^FVbG$P0}1E5$%9!bGPP|jT@a<_~UvA^8; z^|I~zYB}e1-yMQamLm@dx&Tz%F|XP0gnYa(@_I)HfBxdH3pDUqbtAnxJ?>Wj^>I9> zl$SJ*Z8{jZ3i9?0YjwOJ70(C1{nY7+|Ea?1R_Y3BEM+&mifghS={U|i%!RjRiUkA_6Cs;ei){)CiCjGD)q+|lwt)~WVJE)$v2<9Q_@iLt)se|hNwsk zwxHe&A)zsg?YxO~qP_-d+v%J-S<&vFI?LqY{tQ0rf6FV@=>!Th`sI4b6g+;z`ffyS zRsP=WK^4}4VmAZpX*0JD4&19s-Bi`WL>{2vI(pQsw>%TFLZB!F6`Rg?LGXTCZcD(3 zsrJYZn8m=lBi$ZKs#@hh$xF$e{cvat#ze7X+5Xd^&shb)6B7fhL`Q=qiOV-}pp1Uimn%W}Po;ctS^~q~8CI zaKBtJVfM=7{ZINc8T2@1%Ac*xyZ37w^d5M59edZUZI!(3*|ufaWxiZZl3hi9LT#k< zjv?yQ;roUu-@!G>f~I*I4VBJsWXB)$HnJL3 zD8c24b*ve(>0$YY@p2i??guiFUNTIIijyzssrADqO|zobhg>^4Qg=sdWxqK^j7fwb zbg=U?uk$`L7m=1!`4{jliYZ4HaBF{*^*tvT0vGG8;*$$_CnIigRB2N0?j`SdyMWy# z9Fd>P%m>WugYsm&y`m?v^AZrL*`Ft2wVQ^3)|aI-eAxLk<+HF8K&QL`xIb#kvy|yF z)MR5M#lHOgh?B>C^T_WgF6#e$SfO32!w#N}V~gRuq1Sr=JFq#tr$=<|#m)twp|JAr z0QS$DE|DUTZ^9R*uw`7EHE>dkVla3V7pbB#d-YnO^C3%&wnJBj`am!qu`kg8gKSi> zZcUQh5PgX2;EfN~SigzqOLfka1clg5a`5Du&V7h{;v33XIUP;OcU19rj(vnUpU54% zQplQ(*;ka%Gxyr)sl&5v$EE}TWRESQy5654MZg(mA#}RJ`;LCa<&DR^=l_mm-!E$k zVC-`PJ^mjt_QD8cBSbrzY~V9%d_U^fxu2I%9mt1)L+=--ru;x56rcj6T=Z^aKi>NK zqp#A_;sUEB8y(!zv_<1nX1qQI(DOD@ZkfMIz@psT{exhS@-3G=0U`WZBudu9e2i)R zsatGra|6(ibENO%8;U(DrZ;(?xGRG%bX^P6lg0z#B^x@|WaiOFz+NUR)T;?J<61H; zicTM(JH*FhDtX-y0Pgl^0kq6rFw9FuEYvrSvFA9B%a~g1owPpFK0$YgjM9vi@koI< z#{;(6HHtf0+Md};nAkpY-T%YWS%yW`{a;^2L6Eo+Q5qEi5s(s)1_MO8LArB5YKCq_TDnWR zdx!ys?ob+LV34kvaj2nrj`#omJ=b~3E3P^F?7i1o-_LSMTGPZTXX(4hwk1GZ_IX4X z{52FRmx_~&dPm4<#2<~k3GV#J+DFqUyyy{Z0c5_~Ngc(sXo~w?EXGUs9gW&PeZj>i zn5&Df?*(!JbNsQDnTYcx5_jIET_?@Ul+&|aH;F(?^krlw)(Kt_nUxuuC9?N$J9_=pwajZ_c)5*loYY;!ijQ^O5CXW2>+t-#0FF5-33C5if|zrA3#6HV;~N!#j(2+5elP_d2=>tUW% zKaKc76Tw6jgPionvq#d+E(Kbr^h;}(NPs=ZJ~?{bvk9Dlf!ddN+Lnv7fG|e z_%@28QpIK$3w(J|TUWPW>)V#-lNALS-@vxqhzBnpG`)$_5l61mwu}7$J!)Eg2+}jX zn8$VEZ$SxyWU{&hIv?J$AL+*@U z&DEI4aU(|0yzhc?*XX8HSU>7$yYD)ArRNEjC7p#YZ7|soX;)~={gVxJp_PEd?-MF@ zFO`wqg3Oh!D@`B6HQVgvO&==gKjDV+fR{*Kz&jpb`Jqq5O06z`h%J~{mKaU;w8Q0ZpfBM`vf+_43@q`F%1FIDV~NVj#mN zDZ#HM=l_?2Kbpm-b`NDZ9clczo=lWQPF(OOC7XkoE7Ls!s;{VDm0z(^~ zYc{Ljkzo(a?(9c_AjpbEA-m6TOo*q;zy~xEpiRp?wJLe3cl-l$yP>;J9j9xy_^{H! z+m(sp;nqm6dem^XaS74mlxpW5Ij!sV81@~vfXqupIWp#F?sl^s>%~g57~d%g9UC>< zS*XsfJ9c~MD2etbB<88E-l_vJrQMc8Q+oW6rKjA*2+9ChnqcW*&ds*29?sQ1m8nQJ z&>X5p2P+e8V%gP0biFLo)*xzb9H5g&St+W9p}^+ZYyzFC-?sKfeLouH*hT zQPr`L`RYV|le>EFJ7a+MM5&TJkPhV~-Ff1`ViGA&l+@=mS(@tRw?~`^Y5anQMY&i@ zqnLuIIpgf&COAT?Yvfj8Sl`FM!1d4be_F7cc)w1=zYKF}I8&G?uc@^ZmIAeyJM=Ym zXI^pk2i2b1!<0ZuC<20aQ1hGp=&uET%H8)@?5JHi{oD7aT8M1xMje@`*Q2rY zheC71bMzRR41OTuoUxpG>4rD&6?N z*toute%|~0uEyuz&iuvz7JW`WT7L$iZ8qfojZ@*#XdILR5*L*&B<46%+-njJf%_Y|*f#KBCu@P6Rja18@Z&@rS*0#Omm{tjO4hhUMhe<_h?T&~hDp!5I z-h`u>69v?zdjF=A6#&(Xkw;R!g1GI#?*n_s@1b6SKHzsiN@-XfqCb97Q^^dzB$2`cu_B)^P1PkHjkz%j-$Nr6c~bTx5K(`ghDudp?LrPt+9oA+sM3L041=RqC%Y z@C6XbUY$3T{e~*cWG7uxXJOJ?~B{7tePFq2T53#wjCy3${fA1$jecX?V-iOxU*_4maL1$tGJUQxdhgw3@=6E z$&LOq#vd0^O4C2M^n2dUclB_lgW{q~<>|+RYOMSeBQ+{4&h<2JMU3D>a38m2;-~r0 zI+J8>!)1gmLpk9flfPLSCZ)A*jT*i5xVId)aFq5scr}G1oZ1-fh^nN{?qjr?Gb^kB z1<`fOfbS9n^Vu|+_tKIke6$=+igOZj`bF?ujyY4Gg9B#wMQHEx0m^bXLH%R?*#r!0 z(kXpYyU6blS2XsxE9KYQqi|`rG~Q~}LN|{2Jna!v_aHu+yIld_K|!J+(8Zi93Crd0 zl%H~=6JM}rq{s<5&aFM?cE%i}M;QHKxR6f@WHhew_ExZXY$X@cr#!kgjY{GdDgjgu0K zhv9au1dPrJF>T)Xp~5O*v5z${G3c*N`UPC|!7Iw%#!-iX4x@?U+(t7hfqFpO%%#@( z7t4?9H)_sRHkUvk#W~BZrt?Tm_rv+-N?rEyx&=6xl{SA#c--d%rBnw_4{R97wGx$y zK+p>XI-phI6q5jQ&f7qnW>q5G!*OKKIvcX{B1>jFe372_Zma#QBbtk=c44Pa zuC`PP_X=E7V1}=Zd1z>Is^?KkMBK`&K6WTMRBe-NSQ3(gU;eDk{Iq_F9pI68BaDrr8=BFppJh&AYlnYFO6ycYbjy zTnv5l+|*DRk7WW%-Lk!EKV(rwC-Y20x}K}@EeBJZU;-;O6O9MH3EBnRD+~G)tueLy z34?65+)lq9d=wu&Q;{|dd(rkeLgw{=Nk$up6J4b2#WjgxR=(xyxWbrV|Jn6kf#R7e zXJl5k7|RRIfle>7G?62^qZ?&XP`~TEJU3ZCC2g^rO+?M;i?2WCNi)pu-%4@AOlfnz z|MLcc9@(ZU=tZMn#Xq&XBSH9v_UHOx44q*9+lazmHXcFwL*eJ5^KT?v|Z= zReu!EYWgQr25~U2H_6ZosTFtYki0e@gJMS3E7m;25Vy2N5AAXT^i);O)swD2{9tBb zwcbYK#rNsG13h)&SsdeISq~eAW}_+CzSV1P^Q(%3D&8Sv<_w9;0-~C)TFU+=pp9U3h%dkBsR?vGF(mbJyOG!Pli$6QZFuuTJY0gYn`U&>Lqqj-!iYN+Lsu%e>Rsv zrKZPl@OKG$WVL@T7dkz z>L6R+jO}NFbkz)pI!k$Medur3L3PHC&sNTwViy~Ef_&_Mm1aVvR?GKela2Kfw%ba~ z_yA7RB~!w5f1gVD#o;i|7%YLG3xInGfQ*~wL}Do<&w6U!iU#c?@&&fS>who=IbF(* z=5;|Ym9j&75|y7yyjQ` zp~wo-!F6#yt!k@z79<+rQ6kS4t7ELcSKWNpa;D~|gj9>6JPzpWmOuwvkac~4lG1sq zV_NNEf3S{ET3!J=Qf5;QgHE|ibM$H|>sVCN`WJb*pGQlTZT0^7bvvxHTP-GVmVv68 zgVh^jN@_>mgc|$B z_(G=e&S_vI7sOHa^nF{yVkz~>m~g$@>5JGqRqaeS6@Y51A)o9gPKj~dz`G#^?_8fr z05l2ZtpGWH!l$1bNMkIkgMEcEiTbt zBLK`tFY@=zt)8up!(W@`D48ODOKKaAlX%{LMn&0AAd~f&LQV40qBI?LDK_ttO6_Tp zYz;Y|P*!Nnx)cmO;3>{3DW`3YZV86Bm-=`&n8vM}_V@2fCW-SMk=Qo1ACFS{V3l5D ztF9iaAM||{jk2*{IS+OUi)Hx^8eTH|?PLCnoX>A`+A2&)cquRX%V#8qrMJb-VG(00 zv3J8_mAO-wO!|Zm%akK!3z@kE=g{qTl#I({v>cb`S4g40_8cxyytS;mEp;KP!O3nX zU!G%LfKO)rofN@#rFNUuTJPyS7GNitz&yTZJ(gC6P^s=0i_A1FO?Vl7%~@4JuXoug z7mfUyZabOne9_oar%{Mc|9nyT$Pio+t^d@ot*bsYcX89|>*3r;Ru=W5|C^NmeVJW` zdFp3>GRvjDE}J@y{4KpUBH9r&FGCyemA#H`>Xvs<#nGN!`Vern&g}XL#Nze85`Y-#ulo(R& zLrb)uh?!Gwah9xdxWS1WujXJp`#lM?X9)AIn?x)EO0Q*sQSY)y#`c?FI01@FkXXK- z%yeZb#JaL(&H4xWalI$*(#FqQ7R7{>JD+G39ub@o{zXo(-V;f*2$l;i#y{hf^1rgw zUyB>Anq|O&G9>U{ifdj*a_2_RpEj;lSgJW!?|PMEKk=_VFj<-Mp5E-BxwYR_{YLK< z&`WYJ=>R#${wZniu;6V{sz*VYl=Z~;?v`idq%fZ8560r`QvE%-HfPqCZvcT3_B8~K)IDt;y0jYiA>&1>Z0DU-ms$5(yh z%iRmJoH3>kuYTvebzHi+4)ct*R}K9M4G#L3+aD9nO~^qh`GDThTkI-~|8wZ}6^Mew z=Wx(klzi5}KR$Z$eXGF^gl9QE zL1LtiCf}!9DP}dRRjmz0aesCLuA#0)d95K69nry=MSiCMyT3(;6JYn}JeIXa4l8xD zHQEYml^Rm_GH2>crOtq`#}&iRH69uHG#pd(MvD1Ndb&|$-7XldEOPW`^C1<@d26?? z2eM{5awP3r&bxQ#ER@-w&eFQ8W0~h7@1w&^?~hYr^X{B74j!JZ}4?WelXCKFw(a0W-%=ElP$w~-@l)j zKV}X0;KD;6`72n@5bWAAjb%%V4S9Yy%ovaXTvq4DXOtKKO>})F6vr3-aMIf=y z(o1h-KndAU2@Jb9-?V8sxD6q>)ffizR@eTX{TSwvnkzo{yP&q>d;D(1=%8WIlL$p$ z*`)=a5eYWV^mdxFxQ*uKh08*>20dJ*#hKur7-GQ>oC}2*dFE=lb83*uKzg~hPiP73 z_!fbCf=r}?GC9^Qq~mSipscdBl!Ta;*dd*jX?5QyN%36>r5 z{)txp`1E4bB-uOhv#R;rI60Yf-n;Jb-JFxTk53yXI^52FrGAY=a5B_=y4){wJgg7R zcD*(%`OWlzWapm+@bd1Fqqsouw-mJFrcx=*N!zSAUfsvsGLW)#D}=Uj!%lQ-YmL@2 zR-WjhnHsfKyeg6akp8LIDzTNi>#dKRdBG)SrrO^~1d9pL<9`)?B`=rrNf4d@{~F_h$j#e+)10 z{D7hfuj1$4j%LwW(I7^iiwl^D$&CuF^hoh{sbH05fcdW@%e!p8Vz$$U;PEr(ooVVO zapn=08ShFku97i=q_8pSQm;b=;A&%v2Ei$G_+!Dn90m__chRNbn;ea#J-1orJUJBi z@TP)H$m!C_*DoZ%ez3$3P=6%Cago5a5B*0iMnn?JjT8@ z+)0cJer~ZY|6NgBbuCQIC!L?kt@oo?x@GP91o<#la$dZ4xX!zJ+RvJ%S9n3dFZ@$T zq9aqyT3CsWE{A>5C^9cI1WT~rq#xs#<+v9J!jFjx4Ie|KmbaO)UEX%ZUNLwBtvaIw zG4G(5?E8eKCUJMPOP-P*swmQZZdj#M$m;Yh2M01jQNF4o^{WRL5m!`wBN$cPBU`xj zToCTn*rWW-7`DJ?~7CAW^r^xN2?$^r^FI{I}^ItF)ZC`YE1>3!3k*eTe2F z`1=ed2kH)C3FNn<_ND(_dEqj>cz=ePsOs$pt*@@9pSOeam>45G00Q700^2Qz)IWR8@xY4PwlG*qExl?N+~$ zbJh#O4GBLTZ`F?SlbJH`)nXEz{)3jwH08p@5)yTvbdXm~RFfFjxv*%jmACivok+t{ z-4qYxY(paz)HBq#cLb!7g`OS+XR(!@^m&gY<LE0r|NF-pJENG z`ya(WmgLl@&aZ;~ynDN~J^D%;q5kE6I%0`sQ@tk9K|u_@h%Xspfr-P2^wE&7YCh(> z{)iagY)2-KWNww#Xl34`7dAnWscv6ZKkS{Pd)HG&wmLg`+t<0_sUrPd8+1D6WR)K7%Ae{VzRtUx9X02%nMv=kf0t3X6AaT^3VSU& zJmq?>E8zW9gbhMeKtL6(NMlNk`fh5qYp3{9j`3>-g;xyd(Z6fR0?w1=eV%U9blcE5 zGQ^u*P>g~qHakVpY>C$1?4Rf>KF8n(C>fOCBvjEYEH1lEzD`Km4qZ{w)qjSa}EJdeWy_H$KveZI}YZ0q}#$SN8MT*IDAu zkp@PWX8mLBrvPRA9Y>Pow*IgMx`oGbRmXT3Nb^1Oa7<4cb&55Tz_>Dv`dh}JpQ^1@kV{SlWs$7incL)K1K#dC~ zzR)_;OV-1u%Dmq`7weC0QQK@DGDKsiWBzRp(E8gPaGk{EEB6U%S7q3`KjFZk(URZp zpQF#L#wosDoEIHOJ!}+Dr96Q z;5x$Rr}mGUvo@z!!{y(>!oE*}N4~0l=R`^K8-v^9AFzIJ>wC)%vJ>N2BgCS2gdN!I zB~Nq#DvpDvDMn+=6+gPruxt7Ihwyqt!|uDx18x*xmtO$ChuYZpzS=n+ zFyi#Tl5g+oo)(ZD!7S!K?gK!j4-r;M`H=C*mOYPp$K$foDkqm__vEfk&z9Dj{ z3yY}Cf4@@9vOsl!=nLtYDpbz;&9yhBr2D4@$&ZHW4vZ48@M{?&Z<+P*{k4RemaH`V z9K&M@9gZ2^zYLgDp0T*=ORRri$>wxb{R%unCLIs(D51+nqhp(e|h)l=F!vgc^D)+fy0zRFqfh8D$^(q`zq3ZMW>O zX(%I`amDy^G0f){$vG9P(Jb~a_9or|B z6Y-kfeL|*dUN1c7uuv?>)5UwZa3arJ7WT7=SflD=a@o`GRGDH8iP326E2B)#1wkz> z;~Q2z)i>1v@2!8#>U82jMWMNF_%#?q4LBX5+65O7^36i%PdN?H z4A7;xXICV!#E7e-5V&uwjm(7sAz1UX%^m89^|g2ve(l^! z^FGPkiguZ`O$Z(Z&on-%=7FCiSC>c6!%Z+($&gcf`%D1~$Tq3jW$lyi?%&uN8& zEH@~1Ev)^-`AxbKTDFi<{CI%3xpG5`P_A&7*?%vwj#E=>G3gs}dIQ{Cr9jlLL^Ds= zUt&|Y!2cxHZ!|!LVi+h+&S=^k^G(QrU_{iLCr(|a6Q>E*V^;@P`!K7@3{oE}%heXU z6Y+T-J1_fWd9!XDtR*TS>n6>@>}3mIvP?C6Oe_GtQ)aY-gnqn`gR)F6Jpk(L<2*qg2P z+3e?K8|0xUI6CK>Mj7FHD6`*)rXKGAK>^ObiieW4CI1Z$#95c^4Q|QM&-E$9rx6_o ztaav}C{zFK0)TAJL1)=>_ z^0P6!t*cz0Ni_M$D)Z1h=RKo%Gi&O^Sz}*i*oQknH40PJk?+)H;MhcqmMaJ=d}Dcc zzW>%x*}a!7>J;kIBBZQ*BQq6J1HIGsJ2%TEC78bh)d1EcH%WSh&=^X0WeYYKmC9?G zF|n%Fi;DvxwX&+CXJ0L-Q9_Npr$uAcliU01K2k@A479H-@|jW^SYl-8)*z?*mX?^- z2_VH*JeSG_eTmA6Frf{eANm%r|$4a@XHBPn37da4^slCQF(D%x%wFS zCkif8X?T!VwA0$}&}-puwhX{GQ~aK*uk(_kq4Dy}7*;p@2d^W=gI-9>dRTU=)c3zP zi)=$^ed|S+fN6LVI6mAZEMOv2&U8p%#WW$#hy=>(BB3Sz&m$@G;rKmmA9&juHHT(( zfU)LuK1?_t1bex#q#c*;)MJ^f9bXd9pU~zPsOWD8Jp}UF`tWNEeURi`#X0fm=xMla zIDL~sA;Z_G7!GsCUVKMq_j|Nr0efV{rQ4S<@Kza)c_G-vW8|0Iur)KBenDZBo{r?f zh${b>pO=e`zaROk<%Skd-H(dzFM1(Aj(TI_xt+|80%M7&%M2>nzb{QR1(I&zbK<#8 zE$A7F-rzD7;eGFD#HT8CylRs?ttwMqy>|F)T(oGO)!(wznBj zsQV2=y64}t9`OzS1A3YA?g0vnP$#;-pDm$m+WHtZHOAv5e#hJuD&y&xdls}+#O&Nd z2~BKQ$^UzpDu78P6W?O~Zw_6WoNc-2LQ*whNm`*75>7&+<0BV*aS`F}av4QCq)-P8 zICOGqHNUC);l^1Wg>|=chqMun-Yx^@(ZKv^L1LqN!8AS4&qn%Q@q>krjS!_)yMcH}KNR)f}w*_VODx{oWSAO^%?JO(` zZd@hqG{x)nCD%AZS%=eMe?vk~|D5c`$tjtapBw^qWI^d|NlqyDOFs0+{j$A9&fw}F z^aPqIqh!JJYqJK1Z-ci~1&SLSRxTIr((R!}` z#W@dfyD{#kUyA1rzRSW%M2TyE99$c)$Oc9)R6w2uC<^y&d_LPgwOClInUJgS@WHp? z8{&2Ofx`iE;YU?B#riy5oPTsZaZc-}s}#$3)HB5%`eiAQrowrLPo!ZY4E?mHyZ zDITeS;-fdMJ%Bk|=_igKtvRO}DW(iAI3^Q3YMYhP*RxW||AHuYtYe#~!hM`f!a_;3 z7t9*AcI|Y8eIiVA^1b!YvZmO$^>vh=72bJO%coxh0=iRDaaf-?U$*2uxvjF`{0hGz z_63D|(XK=Ta7;%`8qWx17HYba`NHZO)Thb$oxg3;?WTXDo|p?HU)9(Bb(OYYdE3_x z0jTdF%VQv|0nirt!HP8~Q%udLvr2N>RW&eeK%224fK;D8Y|I@{vB>BOF|VXK=peT* z)Aa19jN;U;6w>}wDcydO>M;~&e5TGi+*R)%w>^H!x9s}g#L7U)ZREc?09oO-s?R;B z=y)aEZEul@jaz(G%r(o{S&7m2>iYtnK{8Q>jKvtsr-R@9S(X+j#r+JzGMgHS6{=dW z-Fa(O&DVaLSC1(!_dd@wgz1#Kp3BChE;k*Cti=4J_$ALVt<`5{w?5%9hqr!9s*N>Hl8+XnQ6}mFdDco_V zkBq~LwMbL{rZxi8y;M|p z;nmy8!wA(sFSpIOV}s~zX&h%h6pa^7@}FBZP|?^ZQu1f3#Has3UqHu+1&@yfBM}zk zC++(SEuu;lh$VD7Q%0iwM8$)+RAg`IZsA^$M%L}Yp1-ynYLphGW2_@ZWeziF!fL3! zta}<`QTWYl$_6pHzb5?F=kXE` z`9eIPd6(D_HLp@1L&}|=cBxUZFj!eR43kUgorWEnliv26b6T!eyx48_!_VhaIdoI)|4m_=LfV4zxe?s+tvO$JE z;YcdV^xR2=gJ&|Ay>R6`6m4f_Geg@rAyM?88gR`gqW14wa>{|R^HkeQ_5oLD?Tk5G zTZ5%Ox=$g@3hC>>QSNo9Y=n#Y%+td4Rxez}tziLKXBz#r$(A(lrav!7yGC!vS}bas z>A`2tZom2Cvb=9!@3#UPY73VKhcqB|6lTc$w&KJuI)6G@NEh{4VkWw9{2lto^$g?BSIIBzqs=5A$=X`V`17tioK6#7d1^qMT^=Ia!WxWT_P+ocZzJhf$zQ$rsAi3f8BBSZLMX zE9^4?)%#YeV}8jfiFSA~{iu^ua#$*wh<^pJ@&%f4!%m2*W{TArQhxL5@e^0$zmB3t zG!ksJLPj`?1aIIc>#j65p8Yr(`|ERKDQQG8E^ol@FgUjwj5Seg&uI1J18Wsh{gIvh4Czy$*k+3 z2@58}n&_lVD}jKJQ7>e!v=m=WCZx~c+`lnfPfcSt{`5DA2PJV2d* z9Zu{tro7SBN+`zPL3RI0arMj7vn(vKF%c$#cg~28&x0|4Jtbo<170MMHV}Prp*b*4 zzbp9b_iB0rdKRtrJN@cJS#m>24~#D$)vM?$i3`RxK`@xSh*KZXh6lS z0EB)_7o-LcHysFD$D2u*HpmXX{-2rbO6#$k3CU|;Cn8~a(XQ3aNu9c`Eoq&(Aa>sw zE=IjA|DHyF>P9$i05B-Q8vA0paw5uYMDzF_vw*>`+>VZ1ozOdE4&^z5nbDdY?L-ue z+aY1zRVVWTp17-ECEx%u_r+(r1JJiMbFCzjN}}(_*FGyU_Nz?cS=PMmQ<)?F2{zC@(D{8j`IgZgP?a zrPUHJnX@ZKo6+@jzGR+5cX9Gu64LNiYvfK8h(^yY{p^77KJ<+7kinHe?}As7#`bq# z#+;y&$q^ua0#;L^^*y}jBFlBhJ%6KhCckQhAW>Oq&IsObwxLuBD5w}R)H%I;1wq=w zB4Gr3Bwd(h_-wgP0moQYY5hftL6z!7;4nFvQ*OBFSoqsal-iE&}f&T z{SiIaJfz8^$0xl*N>cCn2_;A#iOH#aOu-#yJip#xl+k$(%u-@y=*4Y9d3vherG8xC z){cGi+cuPuM*z|`QVM+oo=twNOqG?glRyL?%))Z{kEcEeWPGqE-r=h|pY)MYV~-?iVn#p+{`*y6DhjH|Z{Anuy(RN5Z=z<5LznRM~& z{h8HK()`>AoAq!0Wm`oD!T8l?U&x_o;tGq?atW#Ta;(?ryl#-lE=lP%+MuP_q0tS7~*(0u5)aa*)O)x+$cij63;245UMHH&Vl zU6y!5FKBb41N`Wi2`k|6ORrx-Tm9@bbmdq1 z+!t?Bx3ru6e-mY$HMa-Y^N=cuEA%+EO))V8)y91+%bW{6|nKdPqpXhKEV2Y7_40TQ(k(Qs65jl7X=Uh{PMdP=9`a|d)l5L_1q*JB! z`d`u)8khuyFlv`yM9+c%YnEZr|6@|1g7`iuW|Et;tB?HyE^0A;BwXux_!Dy_1QlBP&?#~xBxt;? zj%cV37o+v1ZnJ&a2al`h*zu}c_y^R%%L#28dc0s&SP|nD3>95q7|`(9&F!!O{89Gn z$Edakr#ODs+2L|KP78=H@k;)>Qqj4f{n{7Nw?-wz! zym=CNk}8UBSk(sB@Mp0P6@l4QGPPfiBtrA7qbaSpv9;{+A8R9wqZj^rO zsFwu5R=M}tBRkF6!{wRWzb8|`(#qlg{@2O!Fae)H;E|ZG$Q7-UV)D_+CDcf>wU;pJlXl0Y?30!2cjzxBO)v!B`frA}N22S!4$w z6EHupT-nC{j=j^+#wcnC#O)ZHMX|KmlzBb$)E2evcj-Mp@kKqSV>@?m`NODBA_BYh zSPX)3xnS_WaIUr2kzIOPE+q@2{=pcwSJII&n1MCMe=LP3UE^*}+=MCm{KjWycFUiOE(9sS~UMBCr-K2{i)lv(wS2ox$asZT!u& zVt#R?^jF1x?N;meMYi)tjzB%dHTzrUF=BqMo01=ou_zvZCJ^J7#5wz$|Eh-4#WHFq zn*N+kiwmf?mH~9Q&D*|zbwbln=9k{bd(K&jJjU9ElDL$da(ZCf>{Du;<*cmaP8R}! zTSwDGpN z`~?ER1&;=|e3Zl~Kt*@@Ti_3Cc&_?tk| zm&lv^=W$F7J*$qP-lB6vIt1ly>}U%PjR%VpaPkHd#|bV3o+T^NlCr4$$+EA6%m2^1 z65lw-`v21~84!>R>gZo<_B6VzjYz!k{tzvO(LMNJZ0;Pe_jTm28klyzx+*I>Edu8^ zGpcnsB?(0KG~Q&>sWga_8(c@j(Y(rTmvWaPn;CVP)=ly{r%_&ma4E9_g0X)#4>*kr zeRCvpj6wvLbqjr{0aHCmsXz`-D#A1aN0W0o*)PicZ6M#o0aE;97#QpPy89L?N9NCBx@OFmagkA!MP^SmHO>u!_8B@$WP-G7+qiAt1Vbxf0R{28QP*! z_&1*Rgz{v`06@(rT+~qM6dfO_@0I@myds)G$U@2fQthmRJu&bianLfU@f@81!IqF% z>w;k|HPo8#M6-te_T1H&qSMW?(nrP>*)cRi(mOLH>rFM51ZTHxl1sxtpU4@!_xNh{ z2cMdE;08-p3oOFSHq@q$DjJjRJ-<4Mjr=b4b)Qm`8v@)TcVsT=CpvI!%*oZfzkAV| z*KyCj#q#~!$@AWPagj^f#eSLAMMBguFe$RVZ@2gO9vW;7(0JBM*qM``ei`Rg&e#Q- z(4QaC2|WrS2o}1zQ1o4+ZyCm0ZA@W1#rr9X?wofvc@zEuFZoj;Iqxx^H9VWurYN^< zTL{O0P@1dG8ea*ZOHp?r>el8&TB3uiw{A8m=+z{oTp zz$Auop$cQs8d43av&pnAO+P-SU>@cirN%-ytYhrUO+Z938zWq)pSND&3b9S0l8UZgG&FqI5^TBb=`rtoV$B+i?%Pw#e8OY3+3LVhCaM)l{ zeruJuH-b9x%fY5>L%hZlsn8f7#lFZ)8@M*=?Zegd2+4UsRNk`wzWQ9DucvgkFSD6V zRQcvLFhVFlRYSKA5O<;~GJN(2gZx_gPG%zM1q=Ye*q&)y(&~pLhxS=q;@^*XSZm8) zoguW_`+vBsz(cAJp;;;6PKi#MKGqM6+@8JAY3rixArY91=@UpuQ+6YAchdRzQ{ufBLpIZD7ngXc+u^n;!D_fDH`eYOSHY1I z&zFl%vc!)#wxvF&kZ+te3*>i+tn=TYmBo>%+_j%b*iL=mt@;Y0PS7A&IiUK+(yJ3r zq2Bvuv{>6YHFvoD@+IlBB`*b*<~%I{nNz4z%FP=H-;{Gc^X#7l`=9lTO>Roa5M*vW z4kovP-+a5yUPg6KO3ATZo~fxSo|J+A4MkACIxgzOq$^BROmjp0+-K?rIQFUaz&BHZ z^^1eCU%@*@;Z7(0%Soc+8obY=i7EO_YWq8`B=SS(-U$cq1AlDle-|qgEH?lMw|3YR zaoix$@S6Yix$*M?CUMqJgmrfRQ1a8&g1dT3#zF`)b&n|GUkxR3vnCt4#3H8)wWd0v zovG|^+`>%Ju6PAl^+#t0$uvINkYs%y@$i>S2(3xS@*;~lQ*t7KLaV#yaH#P_BdfF{VJh76Hn!uyOYms5pUUuSzQV z(My>k3*X?P@NI|tZ&(pbla^$`b*hdge;Ux4jq-zWy#*fI;qAb`OHsO8tk&T5V4VMqFv=R_f!Ka3VJXfO%erP}0#dbDR4-HQ17`<#nfq@XXln86MQ!6Je z(;uw$|G!v0cf?;e7z6fgX`-6d1T~w;bPL9=`FcrNEnc^6A_U61M$2Y2w{bXmAV0jJ zl;03%Z3N2XYu^=2I<{$Bjlu!UCo^d;uq znczuQ8Dp>;>JGk7dgFl(IvWgfOs2Opv}X!!2P`z+tkwspKb!gxEh$`~m+*iDr15Vw zh3bT4K&?L0*xR?pVZK&NS#!u&@niQhF#Ev zo1q(gZ+7?D-S2*WyZh`v?>~G7$9tHo&g-n>I5M<87i%erjDW6Q%Y$(Ho~owKlt3&< z780)b4K63+p-^F>;xjIvD=xMK2E@y==_TcClQdummaFNd?x>OO6y9gQKje<$~0?oe5|RvEgJH* zuAUF!d`jDmsPlamwUBzXGYq0XFD|`Yde*a_18hU0Jt!t9Si1T5zPg#ZkN0PFZt(mn z+KspuyArS})Rs%vb=s)#n>;G1KlLk5SAq5!5zDrG)o@(;Xo&qYUTP@^8+mw{QRnX} zOYn^s)siBF=g6GXqs&6THP}A})#A}P<>T&dLcEmDNxubmc9?ei8SvvlAE(?E{(=La;7{bI#~kW5alVYWNqB-DyWqm_q=PZgpb3R z##DAYUGJBO@#Uy(0m;wVbpG&Pdx?!I-yj3JQ0mX%*I7EOV}>Hx#bGZZ?-nz8y{^>g zoGl(IEGww%M84=VH)i1fjWA%}htwBV7QIfDmK%TD?V&1fEcRV$S-IDG_YuK~_tSKP z(Rt4Lg#kJBUEBQ^<)sUhFdRA6wkIYe{D&G|*GTjTn#B8X-00;BKVE-LAyS5dWQYDy z0F^g53}i#%KzFP*jdpeV3an9`@NkcpriHr6lQl4O73`hu73J6U_8B~ z5~5Ql`ij%CEXunGsBshC9Jl*N|MX3Mci{4bfOFd}OP-`cjwG>5(V0dwnw@P`w~>N# z;&9y`zTq}=W_0$rN%;q7Wp&)sY03ry{Q~N_0KfjEs~-XP2u6eE=b_Jl{-80Fwk^vfcJU*U72;8GSIW(mzo!uz5pb{tKEjhSjE#u zBi|C*yc-87-Ev!buF``3@qItFrvP7gMRU!o>97H^$u#|G*y^lJGgxv3JF5T4v)=iBuY9E@UZ?tq;}qSFJdKi7w2p zksyi44|`vK!e+|;;jsKcxb+Fv7q~PNZL#xrqDj39Gd70Z?Ov^Pl8OKNMb>_6XS1QN zlD`Jn6vM5_(E-pPBre12Uj6c+QFer0L0BS@o^f+9A?B$!a;Y(DV1W8W%NFB$cU#9` z=D@XDcXj_33?&9RNj%;cNYW&DUpHSQit^5a(9#_w{`k%|J6M@(pg>pj%`2$)!DwOv zL~ZveOq1J9>%9MW%=k8BSnV_OjZ^%t-LG-Ff8E9Z_L+~D z*ud=Eo%yi*og(eOzWC$^Y((+)9Myka>wo*C?*SDs{LAg<`u~xz_W$hQXt?LJt#LkL ze;|JU#FO)~!0-=7VSE1b%>MJcWJrKjqIw?n2dl$BzxQumS7!jjpU*%DfBxrZ84K7k zGB6}9e_p!5fFBrs4L>>o?4O_IcTZ;%j@RvP9dUmg!GF6f&$0TvfZ-c^lQ6^opSk~! zb1jYm>=-1M^S__4|M|KY?KQ24?Wq<^oTCBEE#e29q;#54XV z8b?~d$wT!grthya%hitbnk^Ix%S%e&`R5nw^+z%dRgZ0-^Y_Sqe(n*tfL(!pjAC&9 zXXQD5$pJ_2DoYy%P|f)d8{wbd0>&eE;^|{d_UDxaf_a|4*9F#}C;V@J{^W?j^1>4S z={e(?LNX{>bKf83C*8YG{`13mhlIm*F%_Se5(CFBKI#+M_{%7nQQd_?dN<%dvm)dD(`GM+LWqlnDE8~P%dQ_Yn3SgRw{!Sw_n%0y+l6tPKbHyu z&ZJb4o_r)dhjEgu^OZ(-o%1e>>xAteIfeHl@JAjL|Lq8UyoYSf8Re7`7F3!|tog!?u2)@p^`OAfAQ>An;;U z`=V{xMj?q8ce**ybB?{CJLs%o*d02(AU^zEb$~^T^pH#E@kqiIKwb66ZMzWO?Q|*R z_7O&E&86sgFEfTEfpgMS+e*JbMObPT;>~wJn6xocpZ0`phzD{@gUYHZvOGh=FrR`t z&;iWZU}Bk#w#L!HghH_r(zf$AqK)+@?Rc)1tQmH@jTkYWwwhB)LcFTAMr+6iSJ%fk zhouU6_IaOHZ8ja}PjQRG@81ER{B5fBgvXJOcc$dE`#Qe8kQv0KWn%z$r2_k_%}wXou;RF>+tN25?J z%))*2L%$_8Z$mxXj)sECn2g)+*}})`%SP0J)es4ljZyC=9QtUThPovyuouKrmQx*k zPTW}5%`j0UM8rhpkx1sN*{EkZZL@{38=!VkMLc!vCQYfD5w7hV05m(AFd47dm>Kn zpzc#@Q>J!@b+ZHGiD;Db&SPzeBlnEXpOv%!7P!^4kPL)cqkPh`8Bh4!;Rt}sSk1TI z5lR1gSeKQG>p+i~vdZUW5iQnnvdz3xOFYw2e|HF4u1I6hXp%$5;aWbKXc}HVy(N-7 zhXN&%r>Y{ilb-dA$8pt^^_&za<7`N&D#2B+>(kUS07^1w&kU}+l1LbdRaaWXvPH)#o+{gq#F zM=ck<5+v}ov(ivW7j*WfTgVgfE%_f$Ximo}$8Sk2I=I5;+q~P9%0CMM3YA3&IT--` zS$M-oPE{=pRSt`#cc^=_iCIFoJXVUziIRzB^k1EmMh^2Hv;c;snl(z?(|2@hgj>Ye zRD3i-q+Y_D+AT{u>N!o}K;^i~uWVoPHF7m*LN@UGfJL=!&O8zDL406lSEt zp3NkQ->u7&tSof$jX3Yu9BkS#3KiRmJOCO#-;?ful0n9nPp{YusZ@0w8=*3w49`{- zv*}j#;ET5353(O_bdxtmJe73kYEfj#lpzbp)NRqPa7i1eHC{HL&gak5=V5x)7$rS)+JW3JlX-(tRwoPx=})4kp2(N}M}LPuwB zwvA>?x|p8-S~sE;;8^CcTzCfvl$Py5wi1E1eOjO@lJJV%tk1OIh3u^~fWsN7HC5aM zw7sVV04RbqfVxZt_%v#aqrwSTG+VFNUQ(6jMDj>b%6JHl6%>dw_zL#oCkeWS7I2IS zQL_Q5$oTo->NO(7=BooHS^oE`=vJ*Y1_JS&fSdw6@3<*wGah`p0t$;wN5ahwaDD7n z-FPf-xyalLFT##*=Lb=nr915^Ik>fsxm{1^S8V174I6HkFHe;8K1`IFz8&P)w~Eu# zMJFKdnRD_w*=GvU9=@L#e3_@eT`nI2Tm;C$E0~!JhU_o=ha+B1C6N2OiGhx90>#M* z*aCMKOG?<9*@XyK*LkP(`SXm52kdxoA=4FL0jFZ|XAIM{KhX}4`r=oV`r@Q~jZXae zALr=M9iVRuRhgei8bk5Em3buR%IC45x*AlPTt%_{J zT@AizDqZ*~WpW#IGKX3pAjWQAuctZuE^FnUTkW!}1PufAGd|}x?I_no`y4T&ZE94U zGsQGHVrWk8+t7E2xqBKZCOBTd*U>x7gbxn%P@KN-Qu7j9HH2R>FZ$Wq2q|ZBR(LKw8{oaqhvglba!NH{LX#CAti}Rk) zedWG#FdA{5$IYo;Hqj(1&c0+4*{(G4FN8rUK>*%>ZNR4eS_k5ON-;ZU!lbJ#!bPH*;UHlo)X7>c_A z`54SCA+3`Zvy0#-oa^_khbyOe0G^8%M-RVYrl}>?+=Z9KM+#=j&rn1ALZeszyI#nS zjvVoY>>EnLJnJ`5`@4=%V-yZuj5p0vtSAT1bDsJxEJIG4wP|die80O7lLInd2z(~7s$NRHoG1a*j_xg>c`NJR!JN2_2zOM^AF#IL{1{o zxR%5UKhEUvb6sa)?p^y;u(2MnP1cHck$KcgVn6p1nuJU2_-k4%H?W-(G&4A}On4Y< zpgoo=z8(NBki6kDoBE^<@TW&&_cRYC@r%*@Xt0JbW_X;2EZ}+gzf!%^xC(4?hL+>> zoG|s+QxQl&R=a}td(%M9i315aR2TX~4`mIBw8VnihN|3VtT~|4%C=_5ecBXG>zOc? zo=9RUvDdc3Y89rooSwIF)w&%p#Fy()vQu#{Cr-01P-@$4jm1~7t1n$fEu^d@bff`L zoDIodsI_jX_B#Ch{cywW&{iPRXo$EIPzN%cdb?iNoRseUc_1y#7G6v$>N#4hawi;I zSy`<7Z4dI@$12$%Up`$2t>Kc5gclX!Phe<6>y}q%yt~lWJ8j06HMf@7MgVtTcvkVg z-aKm!WZ&<4bd$81e&=QqR;5pp_uB4h)wNcQf%(mMQD6fQ%Dy{;GPiFvn7dcgEj zd-$6kS|B3MTee_$-(52-nTiko4qvL4A9gT>JQ}KkHBA>EwOzfZUmbe&)FR0*rjw%B zZIAU`Y$zlvE|d!tRo<}xm&PJ6q3bvB*{3PbFVy(e+k@v?T}<@ky@xV zjkCFr=v51~US~|0H5#^>HZ4@EignnXphys)!kNj+y#BCuoRjAb3`b7 zLai}1Vp@LX%_}rH99!U&UiK7Qm%_PS;6l;y;?wr0_p+?x1s}e?jf#Gcj5}qFcb7(l znHz9(v#oPpzt?A%>VDAh08rkPs?g1G7XNt2sgGpPX)TRcNpl6i|Fj$CVzqqR^4ZWG zti;>q$EXg_3}~%rp4NM511{>`(4d(-1zL#~UQ4T5o^=?G(ss=H!4UvAjt_!-W1Vk>d zY5YagQ*8uExS>n-7~xi$Uf1IVw4^fWY=5yO)P4UW52VP_Thrv0T>^4GQ>(Otx7cI#??Y}2yI{AyZe{1tG2d@ zT1U8ldre89v<$Fi2u~$KtoSdX!UIh4%3D*$wY|n|gAtdlI*COO#X>GobM*4a>CXPG`_?kNXqj z>XQ;_H*_Ns8S|MxQ=jvU@ z`m@#^4Yu?MC@S^jM{{dLr-(PNdVYaUJAyL`?OQoYv^1Os0g=9OWpu}7INkfV*6Hm>8wc=`WTV;%Ow%RYFXcp4 zt8@LH|7M5_xh*lD!JbM8E778!EHIJ^7dh6pLl}M1Qq!t`R61ODS@_ZKfj$<*cIYs* z&575R*A{`Vqp9o}mwm_Ax1VRtGF&F7ure^_9DfuN%NaBYO2D!xzD7KMuogkkSpSyOZak>c;FgH-POk$ zc4O9PI!gHN^7XVagPRd11H}9Kgo=SPrgbk#>$NOv_lNaG^X>-XIqbNCK>)b!nEYTf zt>M-{v$(RxW=yu!}GsuzK;B#x_=Vmi(Z2g4%)q@Kr@r7JIzA1vx*K}*Mr>Dj{8O?6tS7)i!PCH^jz9l3!n z-EPjF^`$Ltg;=Az zYHkOGZK<;xo?8z+WA6QMYcnMVKl%Xwxa2g~gc}NFH0!RI_d>JMWj^LO$nuIjWn4Es z|7}1Z92|d>#HV}Slf9j%K3M;m#sTP&_Iv)0-ya=9kp$2ja5Uaf^2zb)j&&_GW~>oO zV3#TYyO&PY=P$QD05GcEXNxW)?nj1vUrhlXZ;9#;EhNXUqmc)! zK`3o^K&OHd)I*$8bH)2A;^PbUk;@)J2mxCc$M$7z^x@d?m2w~Vt_Us;TEzJx@dMg|Ibb)-_MOF6~=vJ@U7TJ7`6Xsmks0DBV(N0<@ zfL64x1|eEB4Q$Oq8mJ^=l2%qqql4;0##LOpWX;`{hqZPuH($hQW==8pRD=__Yf{O2 zN{}cnQtH2nv8=q<;&X;j41ediTxNNLyNM5}+1eYr>oLoI24AFiOx(!=)AP+i*Zh8_ zffLY=`cOD<)a(}KvJP)oU|mt}M@+ZIenQ=fc_|(7eQ=MRbXOevY>PK}*U{wLr96bE z)XRxzV>I(Q--Yn|hP37*?MzCRB`+~CPZIDU-az18lS8zdqje=I=qd=GGU2JY+|>&5 zk;UF?bMX9a8*3=8AKx9y3^K&O@eJ zc2$T2W`)Dm49T^*t$J~$98tMpo>VUpNO*@ zx|7dvFM6BuozuLO6xGcnS4sg{pc(zy%kv2E2up+E2!2Lsv~mf5v=cE@5tgxei0&L0 zyRgz8Fm3}Pe7)$Pa6?J_J@JoLV%&{den0Yy4?`=T@}s*?vKT~!UW;3ygG>p*%HDla zIQ55qM~7Zgtf5eY&F5=e<|FpEDM7b-MDY!N1r_TIBA3ufz0>?4@4^%`w=Pl(Q* zRW(?Gigh8ZLxg!-Ny@S~97iMm~8FHvB6Zv26yp0N130s>1(jV*qm(#_-gZNY0$#}^EY7a+&RB|!S0{`H$v-8A*dum3c{ZtDIrO3CM83Yu9LfjkeBfs=K z_h$QW%lU+$ZK?$0m-Z2FmHl)KXtP#mtegwbsH=> z3%z9t3*iT<9$u^TW!QWnng5{|m&4rEIBL!4Cg4eby3nzIXFzCxJa*Ae!t7?bYX6x9 zWYuYqS35v42=SVBra)0?T0h|brv9F)U(9Tp4jxwzp!}HIBk~7Yym2YvF%F-<(HPfT zN;aoeBl89=PfAx5ivA!V5V>g|sFv2Tc8dQ3oVJ|NRyJtQ^oU~HbMwWPyA>l216?N5 zsWFYJkZJvdl9@5O=fB{sB0A{niT2oL=mdvrI*oBAO1PO_y9q4xc`nCSljEbj{8_3d zf{=*RA}mJkf4W!F;-D6mX|QO>!ASwC6v#n9n%jJ{qW<~{@>dxVk}N&HWunZT#X9o_ zrv%WIJget`OAGQ#^j0~#u3@|ro#N>jHSU*&6h@Stnx*jIp}l#vQ@Eoz;PL?YWPRkk z^zY32V%NIE1}YZg>6Mx#2{y&YxHrBACsiecMpx8iq-U$fyEdY_;r)D$y1V$<8H0MB zMNhADF2>IFA}q)6jP7mE_`m3g%x)tBoJS(c0L2Zge49mF@##y$omY9nJtMF#y)+WL z5$d^;kbBwIZw|Fq)0!*FyMY?iz2yEZG(##2vT5gP`OEwVF7C^{CD*N0R@i(yP)D@) zDi|CV9w|BWqI~VhqVe82xgE$?#>TRfbP9*pbB_3q&qIkzX@%y0UZItQb}SQ6x3O?{ zS)x>A`hIQ>ccW=#xiOuHC*(2N2ve7jn(mf6YM)~z!4}J>eZjf`1@d?~hcW!U#Yi~J zmQV{CAU#k)}z!*iJ%1?D(j>JQD!KB^Ic~8#xMBA9o_gH>Avq)J;|L#riVI{f$ zQ`4!l2Fs{q@v^m-Du`QmlOHAit4XkrWb_&+xgNKbn{#=o({wuv`~@Gj7FNnrfCxKH zgelM{`^I!uwZcgl?AXT~?r6DK9!9Sj&$4QD5!#(hX?J@gc6T?*e`RhS?_4wKmxgvQ z!Tfkq)M}|I^;M8)Rx3l{g0EeMG};wtELCdQb}CGGrgnqs&#MXW(}5o@nj~;i*NK)cDVlW8BnY0fK zqH2)D=vAuKLVcNw>#xbIR-^%*Q#T9tQI;VZOCKsE4!kzL(#KNvprS*^ke0$QU2F}T zL=$f8m3gC6{DEzNhjOOgWJOK(SjN2a=53P2=V4AtDGjuT zpHKQT4%#ZGh0D-r|Jcoe)DXFY*zMUXp4A;jyrN3e3R2}Yd>-WWGC+` zDm1+>A9|?5hu9A+8{?%k`T~0Y=xOvg8N933ht%hO{0W4_14lru%*84!eLxvP#Ay|u z&gZ`4AYi(bB_zfkq)zL=lchw=W|BEs50uR^7K+Ngex|r@)V>J!LB^wh_i%TmxDF$> zue>ywFRiH0(|kc|Fo#bq_p$69zjlN=F=dOJA+-g-uHkumF8?s;cwC&&|LY`ffgNvj z4BDQ{RVvX-YeZd@#Hdxg>~*v0Ml5^al)~@*`~gM6G&L=4bg#JFAC- z!(#4NE);;m#qUj*m~(DH(AYmrG&?il24|eVEtO%?{870lY3b0T4FD?G1i-R9yYHxE z)TAS{mUTn%naU?JB;+eoZ}kjP#ol8bgE(x(#RO>ocz#`=}ZRoPASeUy}OV@oVIY=cpJZWEWa-DsiT)ru((5V`rCu=mc{vrMzCmUtXnv1fFbyNLb5vt zC(D~0t7|NUI2|8jUg~XL(y?bCBOFuf#+Se<8|hCvY?YD3{K^5^0o=lKPA)OBGwI8e z2Hc%CokD~WUCdu;F$g2(QN2gtrQC_(ECWD9d4U1}6F~xO9lcz_@6^h~ii`AwVzmRR zKmLB{=i5JMZpKMqIg`b7fnQIn_^2m-L=mk^(P>km)gazk-BRo6??Q%7n#_h)mvYSK zcJPkfbb!7ee}#MJ>oas6@bRx{kp2vBo#P@JW>vn-t1_~s1w5Rvaeb{IRjm{4$$6i7GwH+0TZGqBFO}-aYmxdj_RaFx%91OYH$XrUR`{ z`|rOI-b7Nb6!lKURMh~psHVIBImv%C6aghKwT*STKEYyAP0)Zf$(XdmYT zME5Vu6N3b&-$G1!uIr)Znv;G__9Fa;n1+|Wr_B*wOj$XkC{XvMoO*IFqwtr?AgVne zI1$d|Kpk|sghwGEA6kMVut(A50-Uf5n+pl@DEf_vt2YDB9k|q*&1~l!krPOH;xB7y z^l>_~Qy#J{Ea!3qkIU3^ieD#rF2!uEW42mC1zb*FMb0J7wnUq7M{Z+;kAy@y;X)97 z(2!lU$v)5HK`&}ERE_!G#2EK|;Apt{LP00)fYNZcz~Gr{cd3QymMvj|d zh?T)M(bU~DM`wqlG%hcIM*@?8r{QuuP3t7W4~>f;F|(&DL0N}b8GiQl#rN6sXi(wV z>U{SlQ`?XJ4`Ak1A_(~|V~cRQ7B)O-TESRyruA>VQ<4r_i$$s($xYIM1v z+2WL`EaWsn_q{&K!tET!uV^XPGepcLKU%LCk=eKMa*r)r9c_+AvPOn@z}{aT&;-2% zQcD`~-%M}XSZ9`s8cTcpUT;COB^qT$fi5}`H}yBeLSq=*zHz)i$C{y1$p3E4nSRy6 z&EIf~b2OVD)ev_@)VZAFuTzcyO===ssTxeIh>6tU^73H1+8alzA%EW+_~K+HP%zo{ zkIpAH-`O?~fwcQOmnHfysP;fC@&!{50rk~4iwpr|NT0{)hPc^mc`mhn;oU}Gj8HJ| zXbG{6-JVQiK@hDY3F2NG0qN_kW^%KugL%aO_1vzeo3o6G^d1d^_bxbHHFIjYEUhg8 z)&Zo@v8)yY|H=czQ7aVkQTK~Inp*_kn*He#W0n4?@RKxtZ!gU2!vp~IjyY=P5r)t3 zS&H3iiEAq8Idn%RjYpyt@#P1XNk03mRJiM69kZBGf0zTNiLICQ!SdjwxR>6$8YFzv#N9|m%zetEDs*a)GQedPBVjt z{`wn{P->U6f>~lj%S9k3_ZCF~ulWDcVKUzKu=5&P-t5j%v9v*6cu<)$qXXJ%` zar})ck|RxG-vq!?`Z0Qpg!qY{o2F(vH*?5-N(q~PluA~(M%$5)1A6vDv=5F+0;yNe zN!Sa=4>zy6PLmBG=K&ad(||lj0CV@XbM6vAM&&fNuj&1oy+< zJbSP`RfdUhBQ)n$sI(a^#&v|WMAY+3FKp~|6zU>R>a24zz`8%=FdxcZY(M+q&1vgd z&tU)0A>jaMCR70B@R{#~DFzKWV^^wS(fhJR>m~2ui_0I6JW!b)sx+2e%dS<6HFBpuC}A7fbU%39taleP6~)T?ULTP5}MCY?W+ zr>~Pib$I!Y+l9VLGQHpSz0f9EuBXCCXMelqzG=!7rE=Z#em>`l3@I$?us4z<`aNG*q4#?*BN@7+#G8o8hmht* zt4d4creX`-rc;xEb9PzGQB0m3JK1mC){(oZ%hVAMUU5!OdEfE?B3UJR!_TwO{tNSE z)ylWO7=N^m#ehYp{e$3#97efn3uR2!R@gv`#wwHuh8vR_3pDOyU*zO z&-w4PLPx&r3~e1e^R0Dyn4RAp9q9r`Nk10lY5T29=N^R~lJ6u?YvIas?*%eWksVa))XjLmWx!Ue6{4#p@`$+{t42;@;@Bhz4gz`jfAleWG* zG?C*%)4$d(%LeV_ln0LOURGL##21~kLTxp-?o%&lJm@y39iI#P_{+7KE>~AZvTmg= z?3*~3%veNc)EG<25x}>9)5`y;fzRq{`v;hY?0B!LY%k;Y12Jc~T7b0+8UMWv>WlG3 z7l1kiMUR)dt6>R983^b0Tm@SmhmRyNRfLB`oD`%5oj)}+&bwkRUsai8O}0i$kpJk3w**HjCBX$vbKF!8&ZQF3FfkRV$L$qFpdO+d`x61$t6iY#%zX-D{3$j*Ku|$$daX zD2GA(#<2M<0<&mLtxCl2^mPIU`nr>eQ^=73X^WMI~Kg+fzFm(niclh(BVWy@HoVLglHac;i9k zhQ7viFoKVWYYwM%MKcF|=$n5%PwLl%GAUoC(!RnG%`)JwJamhlOtqfjdFRqTkRg}D0a8PoT> zSsLL+L*o%xKav1G&YIrn;OL|7DN7BF>vE;4QkSi{-D4@w#~b;;u`vu?T`oKqP>M4q zZzZPXi!F+;0}>CT0r^3y&Pv$Uk z;U+R0c9VJtfSSJFqQ$=GBJ}JY>9YKw5(VwY7BxFrdVuY+@pIIG%Ya~w?E7vIeXr*W;VGcNr#sONiV(Z5@AYwP-Og#m z2S-j;1s;4IDW+c9k7Mq z>_;e@W&t4E5oT=Y?^but)xMl2-U+AfFi@L=IuBI{Zw!+Wu?l)nDm*Q8(?OT^&9ay4 zA*wmE@q*3EcN5QcRBetcud(;h!-Y;gMuv!8%hd7$Na^EDtMXt~+xKkXH+$H7H>u!t(zIW*Z0WW~ZF%a~|s$`SL1 zK=p(?4!z)SkOE!AgAcEcd<+xCKo?k5t1XtnY@FH7@3MINY+Hx9U-P*M_Y4QM-yJrH zJF?WEt^8y+pHVU#bZdx7uKm{OdlbRB5-qoUtyL&T&QM+7S*ix^yaEcz1@I^YWUgew zZifqwQ?Tf%p;%Yem*sJvIve&y-Bzz4-WB=MDaAl{Ah9^B(*ajt^ik!-{_MVOf86Vi zgUxEzPJaH?yGdDoTx&Ojaj>hsa_O8w2U8cB1V<+3=V2VP|^mP#a4*^vLZaFDz z+KOx#>;4X1J(gJI=)0+Fm49_6KL=OcgeJ+*w!c9#5zYPir)En zzK1JXKHj!yubpiSyk~b^T%lyzt(unp){{x_?ZHL*5dQ6356LJ8hRZIEXE21k%a;BK zr~$8=pF?D?gXW%%KPPX-Y+4*7yWRx1MF%tlbMD25h-*6wR5AC;Z;!vi;(_dQjk&0o zD@dzHz)P_ibRL-RfmdS52E3I?%IQ`)Sw`LoVk`v7wIhn7xLXa5w1VV@7o(p>%jV$~U9R;|eyZDNphdJC zS@?)H0UOr-aJ>bg%KNQH*ZM1<_=ln`eQYI2up!^o#V6BJZ>te$9xB6~;z&JB%yb|@ z>6fDq`>JK=7@ZQmSu^k>pQy8_&o*K}a7-6`pKWrjm12cET!f=ZZSz^=;^vCOBHp0Q z68Z)srQYsxkF?_=Jo7Q5?b#7>RX<~$97^~ZL!IU1 zw#`;pyd1gj7BVeiEOE3stRLSHen+`jZMcM@P)O!VNpE;skBq8kuwIxreDVC9#@7!{ z5kx3AmMS21z!3_z(|1b7lGW!ixZgX7j|dGqJ12y*qSsVZXkCWY#@-e{N+< zjTv=feQ{)V$or(5g^zc^+>Li{`@$wk=X-at>1007#FvmU>!FVuD-G5eE|U89yA`ds z5ftjihqN zV(GgW4{WTIpZ&y+3M;tvyjm`3a@eKawOn01ber$}Qc%~Gyr?&prnh%^n3K#;Oaj^B z0{?_gQM#>rPj}T|?!5?mjh)y$?njAMJ)Qz0BF!tSF3|5%H1*BIB zj^e+O!oSrC+YU>h$5teowL!SQeWHY_r@ax5)U@R@T^qSC0aO!8qw8JzZV7 zsDjrrr{4xBNOod`>j3P7v^zUq{IWX|WJjt>g!D?A*y-jM0%@@4$iN452Ww|zOv+v7&UTIpP!D*$Qhcgn_p9nniWHdC7oS(v$;Y>43q!ZlTW~a}?+@VCzr~ z@I59?5vFhyRoGOx5e`Js&!2tz%)hz_0(RIEwuD$2rEy784?%2nP~M(mNgtNRwz5Z1kkUsV7GKaf{RRm(kOg z+xoNn7QPTK(#1^`v(luwZ35Y4TzTV#dMKkKEVDnCyK~~usTZ|MZKeaDctcP!4`Hc7 zh6uzXIK7S6L9|OvBGe%|qoMFvZwW-xAcRVU%K3VB9Jn*CKr=H(*e5{ny-?;&C)fQJ*cs9Mn?HZx?j20(=np)`xLXUi+<5 zIkrk5ERzK{c|TmtR`R`_;tDVc7m2u4j_%~F1xiLUCbFB`4$U3TdUxIE2m4jM-*=o6 zZMI?fSBYF@l37Ir-4_Y7S5QaoT)^HyPh0f>UaR{qOE>1;!-9E*@_MgqC1(T|l*)tY zbFEwyYAbLunDw&Dx3$*4>j0QEgK)>!>xJ^-k@WT7h{dl(^8rehDa(Rh1CL8zo?IS_ z>tnsWVNOVV<;gKzR8NQihr>tu$ZU6XAig?$tAC~J)&?ZyLB zM@S1dAhJzg$sax$drXW2C^bFRW#)lEEphD9!H<&_9ED&e z>nfsj3DS+Ebi<+>k(L&a?(UXGDGBKYX=&*N(n$BB7a=Tq(H&>{z3+F<-q-%l-hRK& zxh{Wr5sNvW`ONW*agTf4_v^0{BJSM&Yu>P-R7DoUYE8|jjJ|!|nD!Txm^exSwf|1VE4wpI1`92vD5u-U(cC46v@);a$CEf#UD_AbRKTgX$%wvtQb zasDEz?rAEQ=icpCo+0o9@UvoC1+D(@Ea4&4^K}olZoQa=)%tv|WYZSb%k2?>Fr~B>Mi@T{B+69D z(#e?WT7G3GHvaR+KeYh32lEB6ewi2|6&?BsE+YY}<1R(VCz9@JiXU9I zvc8%-O+CTu-3bFK>wHOkAy{+=tl0j}9F=POh_%DpH+#Vx>T-T8cXIwa1KGKV;pj6l zB&lUi+fI8C89&Xj#-hIN2ldI!E|Dc*QBxUC&*bJ#ga{b*)1XTub+Nk@Dy?c1@}J}$ z?!WB(n{#V`xP0SYdA)7w95?+6yx=So(;RJBW9SOo8?WH_dk@=35a z^XU`88kh}LQCFUH`R;^A%MA6s+eA1 z!>YOJGn_VU{WG4nhNb%^((MMbGb5i%?Ca|WsM!x13f$L0||tOiTl8O{+l} zK)$3Pn$WyouP2dJdv%(!i2QxAKvh5Vu)gVJ z9i6_iWa+&r8dT1K`1P=M$>HpPUR43N5U8>?`7qjdrRk}IsV9;I^AT6nQPg4Hc}lc+ z^#_=M_n~A+sRoz4*f9vdyhYsSnStM#Mm+O;*s`3LvEK>4@@os6g?wxT!fqnuFNZZA zF;o~~hjd{@MCiGFgInw)0xqT+hos!w*?VL9-wawf=Cl)Ld&@X_;fv^ep5dJeQ-kd9 z5Tp=N)GuXDAGy{9QhEQ#PYVFLM7SL{NqYA2ukl89bJf1kVzg?kP+99y=Z@e4X$Yys zr^9HI4)Et!hE7CC*f=OEg(S6u(| zUJb*neB*J*>mELHCkuv&W69d>D<#{0yJZPe48@p}TbC|&pXC9Q61%_po2|qvkjAIan z%n%`0;{o5@5%BYKjP2qyZPc*SZFm!c-C%b_E-@A9R|#f~b3Jdp)@4XtIHb<~Le2%8 z-TX?4-?wbPusN`P_v|RI$^ajPN(~;B3!Lo4LJ~?1GSjTzKad;U)CGvW=ep1t>*N!3 zrmjIn?`r(ekUS?5cf!ACXAEH5Cp+|uG!NQYR6wh36A6>NWOGQ5;X-(aKt{?*!G=vT z^>Ov{uqv?SWuUYVxk2Z}wl@VLt**VQyb>ce9lf)dbXu<(b3`~Ne=fn`;X7~n_LrQ{Ei&o-_7k@~i1KZ`)H@i@ zh(|XD8A=c-w8aL(I?uR8au?wnx@AHW$l{e| zpg`3cGU$WXY2RUs_0(tZV}eC7$9iEub$eL$jq?jt%YH}HJbwyHgWa)NtwwFHIr`2n z_wB@~mOc?yE>X8@QXb9WzO!vGsBEDYMg zDRDWUQ{7>AE?*fLGHq>Dc$-oXHy)IEd?VB zh;Wwt-EC%yOT)1CEqLX5zlSPBRog->JkNg-F>Afr%sA)yDv}&Sqyo<#vsc&JrcDPQ-11TzR;RQroIWJEHuWCsR?pU zhDtZk_VmMs^AnWF_QA+5s3Yfh2lS*1L1p_#(VXl7tDNx-T)s`rb@y{{ZuGei{g~q^ z(6gxX%gmdPr|1HcpnwI~^Zjs+9aGvtP4(=Z=IoO+AZ(nwRSCB#LvAP;Ru=^B8xkss zT8PL(uv)KP7YBd16LC+{12_z8LZhBL6(d3!?c?to^=0M630+-z#Vyp4`^VGn$Z*RJ z$)vpAd;G`(*~Bwx)_sQMniQkECIvlZ@ug9AecM+X6Qf`JAh7EN^s}D2RT#htGlDr> zhdNV!RcQ|x*7_A#QJYy8T!0~0RH$PU7Fq*m;sSx!jmfzO5rPG^Bl}u~vtC})p?lmJ z|I|mrzC7w~*tvtB)1r5c?t;>Wg3r&hcok)dypN>k+;(T|-UcE(jVJ2`fYb8p)9r6< z04021&h-nXkBH6$pu{UZoX)4Ud2_Z~1lqSiS(w2O&Fs9`t%O5N++q_?m4%K%nA8d^ zXY1?3&$h-*>Mp(W%fsnU@9!!#~v!gzWNJnA6>!>PAuzx2X;I+;e-g)}~g!Ybj;-mby>Pp>+|)#~~L;gHJq&Fm?6qO!wizZgK_ zfU4l`?gGk|`Runf^c59UYX6|jrQ{WBGu^oD+Waz*t@$tXeNVufkLdIed(Y@gkMvQ+ zc1t{}`jr8+MG|8iBO6(7A5UqIg~Dc8vtx2dQHpDOOdAdUK4sEn5&mS@LBBd)ms$9< z$+xajr-@-|a&8Y-ojWOqYT6UEfOeGxMu7O_*Sdv!a2${#Flc&u<~!vCWZQecyHM^? z?@)XMppA;2uD&Y{;|8&WG;_n)#Fy3PcBWq@U-2O0BNK&>Iax~0-Y@Egve(;(eJ*E* z@vi`dPxHq1yD(IltXo>zd4yJOoKj-)R!hrsVvi}IcFQpd7y{Qpb0%N?=vQzB)z(v- z9a69KFyeRZcIld-(NAsQEn3+fs78)PN~@q7@Y9#wO|>RbVBwJJ`zHzARg{vMCKTd} z!Hg`vB+5+O874#eo;E&2#NiZE_4`oz|N{ks&e$`kfo4n?a~lA8}7EL!)IMFHB?lg}K9j0`3? zicicQb6%1&A>uuG;yL0k&q!c{n@j)u!JDfA`yy(t$2F$w)Pc%dlchCo;_f$HyNpap!Zh%LG)Ggbz*# zI7e!6(wuz1DkN)06Z0fVhK6$^8fgQdsm94>?Cwaamck85h`Ogl`YK{(X9L)--gZs{ z(Abgg<~7j)kU&6QVuq?Oq0^aw^&MdPW}v7aaEC!10$;p}*R8XRBgLBeZaFToISjO# zNV0YTS>~jI{RI!~$Ar-K!`3Vwo1_pmzN^_zzN8M&bq|5z=jl*10?nStN6j&npokUCI06)f^F@!YU|7L1Gmo&|m46ZyOjuV#ZOA7sBK@*_ltAHEkbO z-H$zwj`70MmhWkdd}MF?Ihju>&o1vkUQP&n+?#YrUOk>zPOrhXG_-EUJUO>yzauo% zO7Rji2Jmh1XV%BJ}GAQMA8h-NRZW8|@;GNVnYyz?p1rO|!QQ_Y6Lq0e0dz z8`x!k_ysWwb}vciraCH~<#%vPdnb#qui%kav zqc$`jWEaM4jz4kpV5kswk!J;BCt?#5n=}ddTu`yy=!z49uQm&^0?+;NnZ0F%5~`Vj zV>bppj{YK8w<;x-`OP7HwtpdeA_6fl5W^Obxp+#oGzr6Jts3vY?{MQdh?TqGwyj=0 z#=KE$-xle!~QkYA*5ZZiw!+16d)|WInQ`c#x(onS@dTm(#w6)vboq`zkv5q z9Rl9Bb{72>mCsrLXswvK=T$^;^wF%AB@RT9J&oI}kXYEgCM1HKqrlLzya$MH9UWE% zFry+68G!gU_ugUEhC8zN86jHS_WBS(daS}2lR!|2dsE1nQZ7I@JHMIOC*q{0m=FYa zRHKsxy|!5;9;bM(wA$hhE<~kj8JEXgZXNM?vJvo+Q=>qVM53H#s+ohEwF?(xbAxf; z!!X^K{mf?1?%ER+>Lht&KUZVqRl}iDEi6Kgje?Ztc;n83*Sck78PV$&5}u3=}>?XuQl(-r!Wp{HuOFj^<6B1D9JDjy|e{_8EItCW7RaMRDiZL-XGa7JY z9GXCT;q5-;zsAx>yY_loaAI{gDcvFdqKu_2ZxVgyId*@v0Nmon-vc*h*u8#b=a;}m zNoX^=;!Y_1g$6Q$+P?R2L4Jt3*y~^3*JJ_Iqn`yF+8HI1)`^W%Xr^k-l}IYt6Qk2v z^r~gf_W9l(yzpGflf*!DG4Y2PO}w`@L|RzCaRp-qJ#TYjJ;%nqv>TF=>~OFEf{Xpv zFz=57vOdH%OD`7kFh%_~)t#~ImaE>ZUU{UgMl`eRG>)hFvGx4LD&ikjK}c~}uFKt- zZ_?LZLM4ky6CP)sQYG02f}Tv9dmj}{IDN|l`xtBO9eA0eAIi&xh?mslNAO-iP}g{G z59&(xRrd>OX#n&hI2DC!y`rjNAGq4mzcX}%!)DGZy>6N~z#U>Lr{o$q#@ z!0~>DtIX@GFN|~;h5l|-AhN$}IJrlNLC794zIjA_B-^&ZTl(PZE^7aU!QhnNB4bl~ zbG1EIBW&%)J>+VNY1}9_T^}s(mwn)&yupXj)$KZ%r>{XAVLo|I%!&elcXW&qjyO0%LHa_HMm2Is&sj|7iQrxln znTPJoaLFGs>+etzy{b{^xZSxtTIb~5nM6z8`PB9JC{OS7^vzR_+4>x9W}RjUH~1G< z(v5*6wao7gJiWSIcem#tq^VBnQ6}!o&MNL>d`V1T4Wz9VjWN;#mm@m%TNg3HtFW~oson5OZ^K+ahyaMR!NYa zDEoT;>N+INTbB&Hl{aZIrGD8fYp<+OC7bjTr`l4vDQ`Rp{B-TA$ zHv0?b>+M+)j-%&qNFbMh{mk1*i{S)|r&gR6%?h2P3z<^XPVAXFRbchEot%}=K;^-X z$CgWlYpY>5WdTt82DfOtcl`Be=+9oF|l%5;0uuyTQjPvU!g{BZr;}d-0`iYF9jVEYk69q?I zUBpxS&M0O?%i;Y`;Ko8G$m-~9G>PL3va9>ixl9=b9u$C*2va4fa%KG)nPan?m)Z-K zcPg0y9aqjFY2tDzrrA0Lhc!&+E|I`y#NjODwKma@(k^b3KB{0eRzuCNQifhKTEL%0yW zD(fbT%In@oeg5HC+hxK}Tu?0DRT6l#<@GFoq>j3ghH${PGKj^AA$1OL*4%0OGwtQZ zdt|oz3i4Re3qtfW5o9iGOQ3e2L&+oUOKAoypOS5&i(iGJZX}F6_ry1xJi|!j*dh!> z#`P5CA7r@k2MC6}NIVkiCdH4d#9qeN%y-@2-#C|!ii(!PCt=fjO=t!8AE}-uvH3CC zUM5LZ6hV*ZkI|d-WgE;0VLoG@RVTCqUD!(C4@_SM1{+T;Z660jiQ5PrsN%mN6z#?Sf}1remf-0PVM!j z3?47K9SvOLZa8X7#M6RE;TH(m3(@I!QSB0C&IFP1AlCbC9xS9qn1!&OVCXVDJ;8_s zRYX3CI6PpFSjO0y=1d>_Onx+@u;vFff%(J2t2=T0P&QbWFSg{gN~JjTv6`fIa0cBh zTv{LA7h7=Ks_1PFx67G*l80##a9Z*IQ|DcBCAM9;_|EWvT@*L1^MqvG+=TdN5T?UH zy(sce!Y-2KrD8q8maV)OjAUF`#-7R; zVH%r*fsn@yh(2a&;-oj z3wMxkOj`BrmGdj)XoF|)PE9*dIB|I#CPsthP_HZrdUR2`n(SyL#LYS* zGO^JL70L5Sj3ZND8cpVCze(ZNM>I|lAcl^Pv-yXVaX*Wl@v5a81luP``zR`2TNd^x zXyF)j7U(xy)+0A>Hn<4d^}ZvTTz;Fy+IE@u*)q89a5?BQ^mDg8!Y3`0Hw7-i^!|SP zfsa)C2!&REpT`iawauWS!jgJrcXdNRC`igutl*i+;ZzvXlWsc`OSyYpA5{rU$$Zl8 ziYzeFPwOJ~uu$x;?7igz9?Zd)80J-S8z>Vmov~!pv))rgynBRG?fv~VYUdl~un1J1 zI%HIX`oLJfW$kdMmsD~E3LZ?H2uZSQA8JUJ)geB&9tJh2IMxd(;*@6?Xnn{57q5fz(4QzGij3Ybj zwBBrDt0V;K9ZT(CO>o|oIW2-ID&k{w`n!)CA)0Zt4jG-_Ja0dW3#_(&rO6PS{nNI) z7-b@rWeyems4v*!+v~ID;~~LNht{H99`ASHyYmvPcZ*2vz+5eI8=W%pv2crr+_{r! zF9YW`yB)xUI%@9qEw~}JdLKvZvs$eWHkbTp5OnHUuTu0_*2I(Gx73Fqb*fp4@!n>$ zpexv~(qt>jq0Z7iOSu(N#gY}Uurj`+yFh7KnH?nC-=+LyuDsd*MAlTBhEpd`iR#!?F70v;}h8r->%? zx8gv5zT8u+&OZzn+N=8q>=?J{+WI>cKFSQie6w}>f%6IV3&H0^!H~^cugL}e1=8FGAJU_C|Zu2&U|`hL0v`l zbDYVTWSn^ZCso!%LA$VH z20*Mgt4qHwl=R1L<}&T4T!yuMBzxub#xUUS%N7Un0%q8F)epWq049A2dxYgulJ87BtQ|c~Y*NEi&K4U?VXHb=Hw_Q@D;o*>K`@cmq`GWy-4yWG|&+ z7q242iEU-jciEO34oVV|@+gS~yz?Yd-n8T*b!x$4zV)DyMpKHTL1#g|aB>mn))0Cg zjKNE>Zhq_av%Xt+J4JGJRU9u1D)b}gk8^)>K0+^GIt5$85=1C53vf(gOm$>vd`)P6 z@%S7}{tOXD9^S9(R|TdmOuZJ+o;OkK`L5)0Su7(oS?q6Mi|s(ZeFeF2BYP32odK9) z+UfWkoDm8GSN0HpAf=Q0g2e4jImIKfLDa=dNJD7g8yp@T*V{w+yF);xB(rVe=RQ|j z)tib=WIt^@|DLFYznxqXqIlVRZuuc7nT*(1#fmeG1Ikg)(JEOjJOq5hnL7`QNKhus z@`ppj8_P!_;Tm<(;z!HDMUt1j0=ZE#I+0X#E1}jTf>x%3Nz%RcwpUVPxdpT)DE)M~+G&vT7DItz%N^K|bKsJnhB24=pjzk-DMg+D1a zCpHEsE}wI!o$%{-^2MOV(XCiLCE=`U@RS~m4S2{^?=K!tn3{prM~d0ctP1N%`uuPo z!y$j;`lJ?9z2#jR0Mj1?-sbE8Zj_#5bJ3j<&&R);=%F;&JHRsxTkI1;O8bO^JXjIQ zB5;sVFxcv&JVdvcd}%0z))P(%M-GkadeB{C`irIOH0SjOZQdyOswicOXmzvGTrRBd z$ha7;6Xsd%g(malnb?l4TQ5Xa`Q-*(bq^A{q4RaQ5sNxCdzz-o*c5ctrt14Zs8A8| zs2Yg|+CF+SJn}4=3&TX*d&y%3F5wU1YBDlrvS1xX;{oIFa6ZapOrmTy9ql1%DVSb)~y;PjlVpWWF z;poqOANF}p+O+1+bM*g*Ip2D6_G$+rsw}{rvLXIv^)cpnx(q^5!lRvXKxu6v=z)WX z%Wo)!qr$D+GL~`aD!GglvJuNGFxiBibGVN{nV^X`ZQZQ&G&fNz^Bv!sLR!+QgxhiK_-d7yiunh-izW zmcGFRB;8n^3%N!EnBeDw9GM4o*eO)RV&$^^HeC2#uPKnEN@7Lt-w0(>R&>2$#h0v< z^;GjM>Iz$HbZFb|O?gF*&FqA==y>P0CXYLKAIA%zWix)g3FON;*}J{!8N2!(OBDx= z!je+sOj&o`Ef|7^l-z*g<3l-ijBqfyH}(9@VS97Hx-ew&JK4cLKwT+=edhD2IQHG8a!=H-JT9KE-A}G6-wn>xL(C3(;h6TO6SIK|2z#N6`dnq-+?SgmJyX@`UXZ zFsaYu^82LfT9huT+K+m-S}BIk#ovr}Uj!!q2(h48ZNeePfr>X;1XWF`vB{sl8&4&^ zSQyVYlv~0&>p$IQQ{Z0sQLkJlbO`!gW^<+C@=>|MPk2ruzgDY< z$mIA2HZ2m)6Z^Jpx{`%Plx`*RzfQr-zSY}k7jZw7zHqooYk*_G*xy1;EA5nr3)!M& zPoR6M54UWLOg{sh&==`Z5XB!EjR)N9YmwRCe^W_0VbyQm89Be`$-HZVWfavIrfcbY z`=2h~IKi1Q9OQtLClUaPDL)VWb13<*k4K>VfSLHa&b3e?RLci?>85wZt53{^97em} zi6byd{ZP16%3b1Ec&j8vO}}ON!k>&vw{EAH*Un2xFcDm5B3UZDUc3EHeP4Pl@NWl9 zu?$3i{-f`LmrM8j4V&gK523hX;hmeq>7mlZESh^%VCHJfGo1>5F@3Lbw}(5F!eg48 zL$a}_0BZsK~$ntKED_X4>J0b{F2bSh@}uxa+MJP-$}3AASjAM5VOtR$5B=JkzhVDa z5nVlM_Z}DS!_yJhX)qpmJrGud5$Zs|aK!A43O)LTGsO!r%w#PtPr>^8WB>BgC}2Xs zZhNKa_15p~jL&&-x>WNmEjbhQx(`kH6%vJS0s9O`r(BoyUyXG7lUrGYqVz|h@JU6X zvjc~iE27e11Yp5 zIkRYYm@LcotFF-UFHwpokX^zip?%z(!_4zaCl!l`>Y6^~j0#HBuG;Rw8q=Qzg^I+P zrTN+!im)9wvZjo`FLs0uLDv6Xj_hh9zjiy(tKuakqJ0x{elx9~%mi`@uqGcqkc(aw zU0*>G9&g|W@NQ4)7qhk<-d=L8B>g?h^p|PP=V!>IqW%)!bXvr}X@58HyPl&uTk?RG zohIuSa$1ZOAB@DSYLu(UY>b?V5%a5ONm$bz?kq#5rnX%+@MJJ3aw^)cOQ(DVh0Z6% zR=^jcDv|hX>L9~qpf$cj;&vRHuyc-1RA4)pN>gSHXm?;%d#+mX$0tI2=3FZnl@oma z_Ua1cO8-F-kgZ~P@2%~3a-(_%6MX=WIGazdjej*qfMxsSHlN3*Ph)J`s}p2+s9KAB z;X0pX^hNTtqV#|8za4$(J|W3kPJZVrA`O33NniVyC&gbME(b!qpBYbG*-)ebY|sMh zq*iR17DftP6csKQMnnHzP-EJ|^^XqIqx9`s?RQrRf=3XzF7iXLNnjsp-1aCk0%6O| ztqYq9!5x%VF!5@G=`K|3cT;>biod?j7aOl_@_2fq>3??b`O6{ZhmP2X&I=vKIImCt z>0$ggK0=?9Pyiy&pjI+W=1*OREFJ^zx-hEtNPZh$@Aa19VOt!%apqO5AMW&=V<@Z)(ZF{r3Mag`jxU0U?u$> zjQ@vU0Dk9({EFy{wphQD=o8O><4>(HQ#ww5hdgsn!fSgX`7`Q&{zLz~=seobn;~#Y zWd_}dH-GN?=6H>g(!V0U(`^yb=kb}|^_mUs;*aCxOKcV(fRD!;3;xfK_J1Fv@xTM) zI<7g)`FCy&2>QBe|ITg606WP~#sHrj?@!N(&ti4JhLu(LwEgE7HX;XXrP2Puf9Ji3 zNCFnXQl8__%^-ge$peQmk(Z9A%AdRvfBnojumEH%jD^25ia-0u4_E+RiqNp3Vf`DP zle)kHWM71&tG5qQM zQwCZB3lNipH~HrmmJtW+86;1gnCx5iJO5kdGr%?xF~!DJ=ML%xD^*|>G1T{6y&AXs z@%B#_=(7y63ZRB0A#r;DA5vqvBapxEV*H=oKIMzw&wfn$=AaMyfUE2PTkSfz*?Gn4 z#QG=si08>GOXIhAf zF28JJe9ug@sZpUu{FlLh_A!*vK-mG*l!gDIru=Ufl;ijP)38?plqYb1hR)!_Mf*dE z3*eJABv?!sOUIHS8C5j&%PLXKx;bIQV5>_BATMV=oTg(om_}<|dS0K(1^w#0*^hr#Z;uj7DqMKJ=cTCm<&~D2 zzVlH}t|?plap}%HIQw<>JtvRZ8p9W=u5Y<+<0HNgcUaapF`4nL+T%exJ7txh+$TdO z{qBYvc8>7(r;`o+zdk$N&3*laImt2W$n-OEFkAbzE}*$gU$;Oy2iMf8F-5(LF@FX* ziezg;rTzA{I;wdk2u-408TN*nnVJmtGl71(`AqMnPQ|;667&XXkM;v^uh=%(HceQT zPfV#=Q)^ZhiIrCFnYrF6Mu}QekRZHus{zBzc-VHSUh?ktgzdO|!~3vpEG_3?<2QdC zaB7e>8%>g4^7GhKxi2!Ii#Yz}N@`Je^W*OPz5{D|0p41MNq#H0X*V5OCuPsw7sqmVHWx(JB8jlv3IQQFC2}plx-I*;kVN zr+Q8S@;fhQ(c&c5G&0OS7$izZxg@u%@AW4@d#c1(8ttkerq%i`jHT4Ag*Dz0f|fp| z$d_sDYj@tAV7=f1GS%Nr8h+b%{*BsVORoCitYNB)hopbPB6&!?J_RO6d^tMjLQpI& zDI?$3@AZlp;eQP^47aC`Nf4PqJr9q7GY+YG2 zH2xsKExh@7;MM=xO@I?vrSrxs)}J!5gvF&=Wq4M&I-)>Nl*jj%XDX5&dra^B6vlGX zNwSLQ-Dz!vbXgKxggFvn8J6_0 zV^K0L@kIKfb^iRhzM;93=?GQaDWGSwk=6<@R;`En#dm53CusHHTAOqVdo@3H$(`iA zHQ5B6)(QsFxE!ROVJ9jGy_Le)gVL17=rAPoDQB?eHRx?78)g|yd)0+mPnLR3AzT6z z;(5JMndS8&4f=>a7HxpnMO*2+l5*nNN{cBk!IP>0fUYEJeV`sWmEi?UJ(w8Imr0!P zL4EUyMGCS3acs`6F&T^PH-o&H)4=j=>=nHsEgAU2vZ1h6JBnF${Kcjepc-x|$$7kL z1>M6R?ngb+wKZ#^B0@D!t+Js<@o72Pe#lV!FidW2D_{^CUay;UH1C;*@{RxS;HK30 zi|htpR%C-A^eb!gR_B=a$bU`bfv+Gc1E-?c8z35yAp4w>7vr<(ur5=9JZ#Z%Gw!hS zIgq0zAn#nh&x2he&)J1X*}S^l{HcdZc}U(S+xva(u?_-?Y?)zp&diH)SgI%sg0Z8U z-JZ=aToaR#Z2;Dv>g^Pwjfotn+y;+xYNq*DIFFqAA-@+!G);iJH z-U7y(uQg>F?Fa@rlW^qgsKE)wMuJuYZ+OfX6HP}&DpvcNYv%z|QnG;x<4{iP%Q1f= zoW!Bm1|*W5KB!4w%BEG}Z5}Eb71~PC6rd>(2jpMx6bWFbP!fI-TmTreO}@XzWh819 zit0;y))lGaLoYRe7K?oNVF1|2J*H8y05MLAkbBFZfaH7)V<|0Si)4E?Uk57|2E2!g z`fk%-#$PJOaerj53WWo%F;aJc(@d#GWMQd#xVpiLKazg?IIeXOW+s~QOUDnYz-U*w zL08`kK*vMU19||(@=Qb;I4<443t?5vInDBJKnB&>`0{nFn;cL?#u9Jl6OfS1Us;tCk zU3g15pfBV9vWij50*a3hgxUL@eC+tIROQJ4u42?&tJP91&d-=^e#Pi?cEyQr`roxT z`m=b}c;lYGWa^X}5XBkiQ(G=E^P@($U2XAgQnT*r{kA6FxZ5*~)0ZIPeTt@0;Q92( zGz@8?`7He@Q7HVE$H;)j?%VJYuxU@CeE!=ePTB(TYk1iAfIQ(p~gmp^K_qV&nL8~boh>#=0YP<#!3{}aDJ|nHLqK- z9A9knNge+tlk)fvv5#VRHbV944hqf>7TC54gm^NIMkTen{0e<4A6x&Ncv0g#br`s_ zx&wHDis5}bm-Azk#Fn_<_L`)QLD_GkfB4Ew6Mu> z@H`CSvH8ZMWLPk#=`ePgah)G}Dm_dDV2@*x4Rt-Qkq`(s@s=8hF zaW2l2Mh)A4zHF)IXzu&-pIQL*b`Y3{)KYmLP57e^y>fhA+P%HC7`D*_WE1ndRTNyC zXIg*E@z67htQ=WtBE8d)E#m+yt^$YP8=y>p|!o+;wQBHAvNZ=M=hl(Rcwq*zoMT z>}`tpoLL#-IhD+)|4uPos27e$JN3N@KY>2#?N;IncU0N^K?~A`fw0qYh$rid{^Y*t z^eNljX$E^}<%rLmtxX{z&28orcuDrDo78-UKuuVyTKP!^qO5wyB1`?{*G{?%BB)5)B2i&!pg%h~vPigw_73ZoJy^x+M=YQ1 zOrd{0d*z}?9Rgl11BV+C*i?1`lSABPi4fuEuMgovzcc(YpNq&U?-)T@vx!lqnHH|l z&|7zVlZP1CeW=`ZI%cw8I%YZs+={RQO(?>s-lI%DeYIXs`c1F5qmKqG@It${ zp7*0>pj)*rAa^5?C7RNZ62cS?aT`4#>vFfCmdM%DDFS7hjl&z{%V2~RgtM-1m~W}T8P;Z~?{u|cv~YFzNNWnIeaxV-k4;cQP9;|F- zI*U0gOD-O2HrTRgRF1@}EOsp48awW7BbWl*xhdwm_vcqqF2Lu_g5GMMcQ z_a4m$QBt4gD48#LfeC7MWDiTX5_<)7w-Dny8{&J%^O^(E!#KEG+~kSbW-&O3Njcj1hFG`- ziEVooQ%ZU5=kbYy$9_FoJ^8p9O&e9JV_xV_9R(Gv6-JD{e-f@>QlS&Z*i9R9eX9^2 z>(^-hMCWWT-m97Fx=p2K*)P2KX=GMh@2Fd|c(u)i|8p6_XwzQGJaO$1yTv*Ux1D*d zZk}3l^^g~toK9~!#ix{!ze9w1EIuJa$-eio@y~~GXvktxU_XcdX0zZtqr<8!Veu7h zF#BRX))*QgL92kL&Q;VDt|o1OosA~?)?i`=C=|HWuqLwP0PO@4jQT@0KldwZE3w4x zhkMC&T8Z+7&M)BQA*2q3=qBXa7UK=U?GNvdd+~x4+eP#>s`EnkdN>^nC&y`?mF`S; zg7@R?RDDbej5UYfy5?0CQ%N^gvc)T9;jF1v1dYln?h%&L0^s5PmDd} zw7#kd@Kgn0L1GZ?qt|9x(_C(D`&Fm&AwNF%N9@11bfplgN{$$h3xN|DEPm~c%BIGc zlGn?=j+I&YowRp`9*5)4lj%vIeIxQ5=?W&tSM-YS7q%S0jaX)|0Z4$~l0<5$_lR?H z!5g@-jGk+A#vyL~S+&@rltDkjFn99FziBLSeS?0IBc)W~wVrrQyyDcJlMuO&vgZ~% zdy~QCkQ=foBTK!&{!2WZUau4=5`1K$kk{^eZpw4dT`RtUP{r7TSA|225TgKzR?A!S zSRR#q(Am_*!K<}#+lidTzy@w=&2H;zWN1IhI%|9VgW@)K_1fH3?vgm^uwijs1Rr1x zuIlIGf%5Gny*ivuod1Y^K?EnM)il+b8bRV#5|MzWErizj)>An$bK@1QTb|F~|63wZg7q3`gx;=AKv)wLY$MwxTct zvzNS~NbZ0QeLPCZ$SuvCOyiBN8%4s+(@c{w`DeU9CCmd<#Hkc~=Ebc@zrqJarvd>J zffArZ%i1H0s56tVmszX&GK%{ZUj0$K=%Fe^8G?&Ne#<^*Z3{A0alkoG z_;%~j@f(AguiIt9iO;b5B>Gk=dcuu+(u*U>Bc6*5=5R9?YvxG>S_|h=3-Ys5fd$l< zU~T7EL3~sX&UTkHsA?jB1VaCM@3nsKXi^NiJ8>e-R`Dl}-X)>QVXkmz`mei7E*6Sv zL}$Xs+9&lp!B`?SSn7fyE6XRPMROSdA(Ep0+f^VkJ}#Xi7RJx_N%_9Tn&0z?bFwWO zYC8RqD1;765BeK*Fh~Sgbwqy8;5@l;OZG_qnXdjX@LDmD?7p|6^rb<0I8D?{6-Ipz z?@`3&_yPU0D*&X>!e(PSJ)VD-wdT{bWa!bZYB8AIn!6Nsz0~G0nNjD&gaLYvFTvud zSTRh~yA0f?U1Nu#>|4JPBjjv4vewy^Ve|B*M|HcQiaT`s(sn}*WD_}5WYO+~Brl$+ zGV0Yc16B|;oHJ~_PvJ$GV4F}l> zA;^BFrM4JaXYpm}^CoX$pCwVr=W~T|Uiq+){E&>T*8YvJAYdTn0La46Eg8-^1 z@>%p}XfO5zL6+QmGjwV^iHl|(Y-UHl1-C6$^6A;WCRkRPV?YMKlSJ#~N-9cl1R}>z z-kY>u8sCMXFpPtI>uFa4-hk_Ei!Y!6l#j$8xaDm2E{||a)u!jEEmO^adN=9cZ^SRQo z+D{wU`Z(an#y!AKFv+tuZ%Le;E?PaedW=Tdor--#;eQzykz`rduH@kx_+r91{wOQ@ zZ5E{;^+k@=Soj4^w?jy!j7~hqI@!(*+*PMIUvA|-=qg&073JN12ijfWqP!};N}EWn zr+vyr=^$N2=WUtrKb8 zh;@eGRSuiTb7Yo-sCy%dh6P_hM8=G4fD|DeC*)1mZ?Bg;GXF~5HZkhB6t}oSI%QX4 z@hdy@(YJG-V*5uA^cTuD)5Pn-PCET=10FmKQi@5eZt%YemMZGS3!b*tTlmz(7#l3C z-`)371PoXUJ;>z5zNJgO&3nxE&~is2p-%Wf;$BFA;F#C4#Azl2N(w4 zx6gU!+9-~R_9&QvoDME zI*xbUqJ5IlD4iS*+en}u zV?$p4D(w1G=8-;t8**Oe?UO@!-vJbQq4AgwA$;Efzz9F?TaudL>qkC|$1$m7xg_Sdw=^Ezw$%8}<_LnIE~`-N2-iCVM%*cEw|I@Q65#V}D3c?eKLH zQnsyLWz#j@18l1FM&p)_YRdyZUFdMkd5dq!RX010S}_o-zSyMcAxX=38*W(NxC)Gt z+cD}Hr{!l!#_!D}=Lj4p;U5bie)izy?XXRr4)k=s1%ErVyG8dj{LUXojuZk9C?dkL z)>xrb3dtr3^RQpXMnT0c2hH+su8HhcxOsqE2q%D_7XMK{jTt2)htfaW`TY68JuLD7 zVl22ZgMJ_M?07zSDR3&8*<00Tb+C1Q^Z-yl+AtK#(Wqs}sz)#e?;SO=+`lZMJriG$`V$P3xM{n@ zRbF~UuLw=Cm@&`h^h>)6i4N-a=MiJ=)Z!K$Kh4o`EP`cUx&qRAOcSVyq9#ddY00lU z3jbNybgmfxJo733bjc5(n>!Yt_$uxGn*2OpFe=xxHACBW_;zLNMDBAr;W$bwJ!L^! zF?+EN9-%_h!IXkJ+j$z=sR_9psfPH$TO{$nA(+C$aQ!KZR2ZH{5N4Pi2%h~kWHX2j zT%Tki3UphYpXLGZ6t;8VuE_$=Ao5H>=SIh z`4l>;rC0REW@q2`s2TSZFM)*ik^f10^F?8X*JE_Df|uqIqGPVb-X>2@WX~vho?J5{ zaBa!Rf5}<5km7MQC^oZ?q2c&i&2PpVZUhM6v{^(hJ?z3Huy{KqHo>9NB5O|NJ)h$I zp8Pv$G|DgM$SQIb)`b?wR-ifur$Pa33wH+hBgN7pZj2%Bh1_j|z?XxaC^5)AXQk%I!AS)EbPICA>73G9?mif4}u#Djt`N)7F%g5N!Og~r_;MXq)rL%k@Zg> zEUap4lo~S}L+hp5T~9dYFXQEaqUTYxNMAAY0X*b`Us}^37v${g%Kr zsaC9&UI!ShD*Dn!VKa57P@vHfSMVM4z=tD#n=46zo!;Q(;K#n4bBPsNEImQP?m!Z{ zq{T$rPrN*j#ETR1Rj=bFBH5nZUP@EuB+5lWS?v%nk`94Sq=ByY3YLUHML-Q|Xw(-o z%=7Z-946T~awm1MIFz&kn;xg_L(A($luC-v&i_K4V(mft6+Zt1bxJ9S}$sZq3e=8D?g*oR=-}#;Cr8F*uG)GBdXW@SJoLl*R(CKT|sUmG}p?Z zJ!Avl9`|B8Cz}jR%c82JHIO|-P=vviUoYa7W?o?PW!XPNaPx&f5C4mR ztJsGHe+w@E%B+GnY^-i<4`P0wD7wTN!|hqZhLLC5)lcz9Ve;Hs|z`q zk)&hrG1+{@#1q4#;Nm}{@d$LV|5GC6A7i-zO{(l#uIg;hN+an%j#%Qzh4tAC4nJIp z8l|=)d-nd6n4ONY2e-mSsGe7QTRLnqQHC$LBW*sxJevvj8$2|kzhevFdh@grRBL9Hjk=_L0cjyT|{))z8G`F{2tV+H(w8_g?NG&5P ziWM&hYRy3Y+(OYSS(`jO8JdNv@?OQlhSY(dA3bXhfApiBQeu`(@cuQtbAFBfe#S}-%(gQ^#3pFdI4Qh#Qpt?G-QeqgE4xk%#B zS1>BKihn+E4B}R6YnrfHgf1eXuvV|x8__NM=H8F$sJL89x6Cxj(K zjtKS`QSLAlxbx5a-B;#^j&a>`giMMT`HQ-j*gcB)VB6qUOpdBiAKqnl>C~?_u5|50 z7FW(>8P(ImKc!|7Ufzu#M)AiBV2cs!UI+EM{ZYwRvVli}IbMoWHc=DUQLHL{=DvE} z*=lU!B1nrP-{vn@=z56Tr!SD>9S$bTs^QlrA!% zWv#_3qf@zr+<5uhq*+I1bpTr~)&2$$2jmEMTCi#Z^m!nFFgAs^^-{ zRH)r$d+!T0GqroM5vHMl?20RMbXT1Ak*3QEEuv#honi{uz@ULX_c>MMbW*kC5x_RynzmV$m3NdQFd2L$6TS_BUcH1CUDjc+dDiRHw=DF%4?<|m{8(Mh zXB)~CK`aPxKwE6|pO3i>KExqC`Dr9MJZ_sE*kJ-LiThsvczrLXU^DK8crIUK%9Xc= z~o{M416NASx0%*&R&-XUTnUGLVqZp3Fntki^;3=D# zb#ms*+y3@;Gc~{jG92O-*b*3477Q#%WCdfRs@ddl z?5YQ5&j;G3A82DjJWHY1PdJ>t7#;UL)%yKE;o?3J7ehn4PeYCJuU}kIZuDaB(j~oN zEV+Den$>yHAq;jY&%1lnZhX-pO8o|M{x~Y`3ZG)=;+emEEPGR{W1G$VjDcL~o>X69 zu9pH9_c2OHw;}kYF0p$yvcR1Zf}=K?G_+yh&{Ys zgGHL1T*Kb?;yqm9X5%;p08E$o6dFFRFZcAy+7Mf#uW>a14 zm|4UcB);3$8GEVb(xTkQJJ*W4RBD>C2Vqmo@)==frdcGt#Ac)Fi<-!MDQlsF+UgGl0IrgcLd)wxS zkv^oZG9Ndxb|Q8A70X%_?jQZhf}Yd(osqn&VWgr@Z}hp`)|dJBvWRfHAKA`V2(eq2 z$uCl)i-(Y;3SwGe^zjBiiB&Q;3)(yJwe;)l9MNa=x6wx34GAXloAJ*=Tl56BHlKar zbTo>kBg4##37VaFre=ensk&R_GkB5T?-j8}QSqA>4V&TugRcRB1`#jPor z)puhL<1{rkW4*YJp`XdyX9f~G-IOzAaPY*8E{@{_Pivia%Uv!Wche1i{nj-dHF+*@XLHr8pTjeijWMBxTwfYuL|9^WO)HGdfraz5$0h= zr4-XcGVUJ-;!+Pu!C~g_sEU|g+NSMQIj0H0oDL5dj#Y_|$!%&sxa!XR zX1(ApwS;WAp3$&wOO%XYnjCStCb}k}y@j+?bqT0lop`4$P%JNcdX(39dXZEfR`@aO zgqsd$Y?0HwueqyVA?r>Bkexm0V9R?K_sg|nxKTh3C=wf=<4A!=qRX%`^w1tFgWtR6 zHNNn$XY)FR*r6WdzwgPPE)BILHPuxsNA%J{VleDzN8^KjCmiQj+48s9{_zCdm$6I{ z4yqg890K9dA52d2qZXoW`4f51293JdUwyWri_TH=Wsll?6-BH1$L3N9M6L zL-cD+BZ37PSE~aYWu6XHVEuxs=(JIzP@A^ic3PZN(_4`g9?K8C^WjOF=*1o;cxk+1EMH_|v5+!x%SgJDtxQWp(hGOzq^lr84f?_taPy{>c6W1as zGbinyK2j1oCue$|$x+qx6Pyp+8w%jw&o_(k(#L$J|T@)7Hicj)uiMH9OzVS&Y@ZW3PYZEd22_C>EcwuZ4hq zY?a@c0WXl;phSrv(VaH@GJ>q59lX{Oyo&46I>hZ$ps4sMAnAO2X)-U|9`y0nA6q4> zaGAM`8~*(7FF!*M&#!49%nxP2e2P$ip!*89dfv=9Qkl#pA`rA|nvD{r7{A#E1?K8_G~9 z8D#ZZQm4F3tN3ZG+}Z z{y=UW4wZHXOVC2Wk+mOLXrVseaXw$SoRT%7?V8^zo-pG(e^Sg8{Gq1>yP?9ynhI_T z@TU>o*R^Z2T0(Rvoieec_Fep}qK5WWF68T*Y1+gU;*4}%&pDInvx`i{2HRzeQ-4~u zhTtw+&|C0ZP)Vzp2E1Mn4w?t9%LTl4{dx2kg(y1@Z)e+>yK}xJt|4}-y<#!;WLaX? zkjucIpK*Oj$5#Zxx?zf$#|@;8b|HRm3u|7-(TzsT@XC)-Qp#1U3<_Ydd1!>Y zZ&vsqX^)eMbs{-5&3(usvi3ANIQdCjS}LZua}XC8FUnfm?LZs-pQN zfgOrBwLvB~f;)EGps`@P^|k^<(k+}2{S)uwjDX=6hVJP#!CEi;)5=xL@;x3K8bbkNK~13DW7_v^LeIV@SrV$x4O5zO*3ptkMYQ zVNa;s>n-N<_!e9*egsh^I#wN_(&H^#tQl)8FUmc)x~UBqc2)T- ziq0fDocBE7@>(W>(u`45FXr*DgX#LMr>X6n{JTecktPE6M`#`J4=J%?TJQJ%(+mUD z%h8%7^&4*)l^I>IyG(^zC**0g<1Nv#{xkl}fv01ueDm@r6{_t7X$mv8i!CpBxv&JZ z)+2x@)>+P0)`G&=6*DNopOJ`#08WKBhDKO_KMekQrL&dJ*4wvZv0!x#@E)yB9& zF0mQJI+#m;P1+{SS<_%q={9JVh}shpxJ zXkjPA_az1iz{G+(PQk<+#1wD7irVH3T1}Ql*Oz}G5@UE66xd+Cb26cC?vu)JtSHIi z#jPZ1I7f6M8>{>*6u)5QQhm^iU7LrAm&uvNSs{gMF|5h1h2;x)3XeD47eXYdLCPgC z+*0GRC|Ik%glORMp0Cw!(#;DQD(ZyL6cZ_?)NrH6-?Cb}|bH z8*gouRgfp#z|A2i78?$PVzjN>0E5{fu66IwW9V~0XnKdPD@Cq^`2DSK9oXD^Q=(fg z2Xd>R%r~gTbV78jZ(^8M)_WSom+(|$qF!Kl90$}69|xSD1YvEs3P$Ba_ z#JIRS_>LCJGKb+p!Li!R6J-csNOtw3OwVpjL1JoD+UKZ-{7phNDAM;|p}Y$nNNtuo zi&Yltck&wA3lL^K@RupD>SF{q0lKmygSdm+llS(?-B;Zl>7Lah)N|!ZP z!%nvH6rzudA&Qu7Ytj$E=X5xitY5TWD;w6P4?8i&&+PLIev`eQBalmDMEwpj$83SM zbKI@b-MDK}kYwcT=2U)DR)5Z|ELz>J_?BRWOlQp4Zm!jBAxeV5@{mElCIpE|kS>QHBUtBC8M+h5lA)MDpkgOu(pctzVL)j`Ob3HAH8 zf?BjY3iF*Ws*z0ulDG5#v$L|SrrsW)S!) z-5OtWr0rmErj=4tavBFY-1|OuO|SW3I`d_ulN24aXl%eY2ru(}3`z3!5HCZ>2%h4V zy{GuA?*-i-3{s-GK9TozJu7BOG;ph>Gna$h!;3sW1W|ACOS}7?vRr?;|OGB81oNVt-;{2IP~uO?I6e-943%6@`!{^aQ6e+X5tkAbUW> z#}z(QCD(PB5|^xwaZ%-Lq@sh7nVvQ|Rh_LzUp}R4SM=?=Gn|Macs9dM;iQL=Z2)n& z=HJv^Okr5#YZQ59a`^OiR+6Cw{AmBp3``2PS0{oTQVUINTIJ8JZOBd@kOSHw@UPlW zUu+HQ8egAtxFyJ51nux*kTO+&xjgjJg@DJy_1kq>YAFTnV_fJJPNV%}N1DSOQIK*e zG*=N}BG@vpmgA2~2Q=h@{rqpl@4Rn^NpgDkKNh;`2crU+6t-tC{ zb#Kgkf4H!Pz@l68bW1RtFeC{lNib}+s?RW;+&xdmTb9hC%k$nOEl8C*HIr6OL?HX^ z44jp?tb_5F4?s)5R2js`*|g7GS)WS}|CD#x&=^U%EH@`x&%j9jS)k`e2X!L`x&?v! zo{IXbikv0|qEPhsBX`FNwo+acWb9`%r0U=kwfh&g0$(S6dViHQ1h)Yh;Tzb})N4vN zu2=fb6cOH>kc&VY2t;`+l{>r^Y&R@{vh!-%8C-C-X6x!bCBYO3wo>O)9?VDxU5H3{ z;uCabwN{>f>gYWoFs1W=N$ys#6ns2BpfIw;fL9x-l=@+JGF;gUpK@)w1K#!%$L{(j zP_g_aY`*eQ(3dVuGsJG%XTYyT+_3D2y0}B0;0W>~d@B2v_6J~^cARm9funq0S-!G| zz16#Qj1#m3+)=}krJ|#E@g`w_DzdYpHl`udoZahae1rh1M)cLWv4hWO^gt21S$OGh z;w*WT-LwLoRaMh$(vzr5$Aj(Cba=~IXykI7j}RS^BD~IfeBfqkW||kA*I6mVJT7}# zo{jUkR|It>xXa(lq;rnw-X$G@qMaT%U^9;zQ6Eg}PChpNNwe(m$-`o=Nww?68_(0j zHP(yzHZP-L5ZoGmm?-$V9Vd={bI7Ove1k(Pk8$FC3t$Tcbn->`7Gyd@5syoJ3@NzX zR*}Qcho%n_BSkJ6NQG)7-+fTVUoO*qZ!3*Wwa!s8$mP_K`pU?$$l(<_QzY9kTgBbs zqacUmFA$p}FKZ_Stey{e%W~Tm&42Meh(o=v>`g2P*T?{?PjS*hxy3a)A==a-3e@7h zH{BIla+8L1J%Y2l%ntaJa&cIrT#}qpIWk@X>xv=R-PzG=76f7zh}~B;#yhA~0*zv1UYKz6r zQQS$rKG`I#4Cf`1Z8Xyg^M2p54Z^J?87X{`7$JEvwqwx~fUwv5;k)tuAr8;T%!PBzyFj^WfZY(D z5q_7Z7@L>E@q0a&ssrDh|EXFPq4RidzzI+t|p>VTe}4&pEOqf_~=K$*Yf$b zE2n1=b;07d*TUS+Cl6Q3RriGWWOs7YM)NRwZjO*7&8tTkU~K~QusMlKEEQ|9#{2RJ zoUx8^O74yWI1M1(szt9CyLjykvE9i9sUfXyG759Os(Y@rMJ3fSw9zs2|?!x2ga42MftD#nq z2r?6s~KpTN{;dtQVPhM@#dh1LRu~YUpu3Qf<;cJC7KAYiLI62q@S|4e%qPl zTDH6uL?-pu57^ZuCd4SWKX-aVF6!cMu~gbw)HS`6$e;AKXWHPdY!Hlwc!jr2$;lS* zKb^AFEGTbFFt@a-itDTP@Q%~*+5!FW;A(d3&cssxXRqSZVOQg9-$L4ygkVeBk~N@J zAL?LS5%)qaH9_DC{fmuc`oq~h_G;rD42Y{nCmd2zGXXO?Ba0Fhq;nc9e^T zX1I5sHF=5IsyRKV!_9ud($|HMc~Th@H3XhkWtMxk$>e2=tWwRjCn8yAXEtOAzud$< zdSur7iWZveBz2gAKK&DTfHA67n7h?lL3(vZygZxIFGZ<^7R1nSpCcM=a&QGwS&4$k z9c9ZEh_qU+clYw8_OK5HvkxXRgD(_P#fYxcsz-Q$!i)jMVxnsWFa4{8j9{i9JFh`QaE3dHcm{?C~AmI~V48 z1ur@Od!w1oyL`^ZELStc+W34Cr71^0)xmc&A)CfzDk7FMU-c;0`mNjaAQLf%bgcED zV7oct1FBFuyxE9q#n}9^G}=dAYl}#PV2I;|R5kt1A*U71#~-k}GRV-@Wf zN}zU1+n?Cjr-q+x>i5WMevW-p8E7m3|6(kNyqxG;!+-a&S4GtRr#-R00uLJsk-wXH zVx75}%f4$zbXU~e=owD0x4~3)D~PJc?IJeIYthNfNy+nY5}Elh6Zl68j?EFY&-&bk{QIxiR(0$Y@=)cZVt2 z#=^7|4S1=6cW;^PB33j5*0)`a*@KccRA;U8u2>W^Cone7tFG;`s&RLBZxLduKAZ{m za3+{)e0*I~cy$|mFlwg{Wyx3Y8q`Nr4-hBiL<(}bHypnq4yzAq&e|Z0ZLgBs#H(%)oGQkJCr#2#+ana+%y|z9p#}j|Ha1 zi+o{pSC@5<&S_kmw0UyXdWUGc>ydV8J)o!8EY({>bCB&|&PoD|wz^34hQM$Rh+9W8 zR!f#V6W?qoIf$TFSzAs+*6kz=k=BDCJ5GP}cyx;tNfDz0bf~NfyE&HUxSOK>5#I4I z)+=gszP@xg`Ue!svzp3vYP~cAI()YOg%8NkN*p-9nOncT87{5;@}QbB(=6#N!`4ok zwi?CCuRYUbLLx)kH;y|Om7`PZj=91x*8}I%%e1S;?bjEv;A!2dPQKU~&6@pR-^y$$ z9sqTngc66xc<`kIWo=9JV_6jX==V{0t+uUw(Qk0E0Zfl*?T@GKkraG=(DA!py~xsF zJND1Ap0p>Nvc%+6)Iw2~izn?aGT35MKF1r05jPR|%S#TvkgH*kdom+0pFnKu0s()N z-!rNE!KADaLML9tD*cilM}s)!fTE%CYl7WUbFxxvAk49%VI$a>_EXPpa$M<^aMSWH zv~>e50i%p#UAtB*AY#0-MA)C^+U}s_E#idV7vu*qUAg2RqYlCSD%5E@f1#gt#&NKh z;R|Vi#Pgz!)|wtjb32~#%c#@*Xc-lhrzr^oY)5CN;2XJS90Mo7ZJ1V~3txK7#KihM zCPr^GT@W&W7a`SR%dQo4_`!Y+yAiU9`oF1X=rTdQ3M%qA#Ei_l= z!-5~y20Uq0Vu&y?vDc+J_K^<>nFG02i!lJw`x;7?)(0N;P+X39N?5f>0>mw+F<)Iz zGA7nIElWh%M>%A*IBVf-(Vb9sk#~=8$Y(eirEzqCZ7q77);B6sw$j(GpQUNtRA&w0 zDGoRn;4L#hw5RO;++0?T(&J?~DXmeVZ3m_rar6@0BH?pskmqb~;tNV1Vp84+vnfivIm?g6xKp_n-L~@ zG%rXCP=fWkPtn(;o4{9s=6=)q2CM2JYLuP)!`PDRYph-MY@;y1Kc^!Hf2%Lvq%UTa$E7S0{?(IjI@;-!h4$N|#*0bJ(*q1;4&Xx6~1 zO>K}|ABwq?<4c6i`=W_Dlh)mjd{s94rzjR-S75WaFHOw; zg(n1w7W}Ts2j(I_cxZWSGMi`gea2{^&s9~h+QD6ATXax??tzA?h|`Azc`5@ko@z@n zqgO`BQU0Ch1cg!Af5fgFE1?PvgBYR51o`}*XQsW5UZW=ad?VegKol1Gb$ z>6H(S5KlNK(@&tIW)`&@ri=KgDClz15t*U_4~GyO$NA2DI_o4wDe=~L#kbRDKgwOX zgC4`^ak@wreMw>MKR$2Qwl~~CmHX@N7Jj#tLAAcQ-@}+nS-)f`f*dzf%DSKT*iO=t zfaX!-LAc!BNU0ukE(!_({z%697dq6CcBYC~^!|Lkb}u_l&Y9whWXV`_MOVl-04qg{ z@t7v>I|~oY72lTbj>~@~E9mxX*!Npc{-2O>GwAW=JT^RtDV*(mn4uOAAm!~2A1T>O z+3d_YZ&LhOG897C$WDRgaq2=<)tp^X?EZB$OoX|v8F+s^*p}JZix54WHr=QSU~l4f zU&M9gPyX5&V_Ji%N-y?pvl>X1?RRQ!tOvi9W5GNR~zta-KxO>+shzzA6`o z^02h*@l$n?4m*{xIYg?7rTH=#^B>?1e7VeL&(?{KXm{!^8tz?VUGkYIU+CE|vq+@nw=!)^MW-D4@r{CL|s8}TQu20UA*>s$p)*_aP%KSW%UKy#9G7xshcq4=O3z&Uajb20u zMeQzKib!hH6!S;@;c_9aGPXWgs0|9eQDPb4duf5WJKaoOmjr0*kjs4W45?|``i+eH z2WLkeI1nQOgvGeNAcQ9e<`b1^s=Rr*sq0kkTrBzSA?BUWce@NaY!)CgyJ_}(AZQ$% zTGs$ih7p9_;xVC+H?$o4WO{pfTxT|#05^{K>F?Y){IRvPlyk6(NsjcwTDIMqe4@3< zNwOTfv17>Bqi%Y$%=e4>Px(>a1A1_a#)l6`3jXfi9*SQ0hu2_je|@kOS<4d94872$Sr&)94OE;ec|wEoB}aX=55k_?0XZWU~SjcxH8bTUYEIGh3&wcm;XVZWwm$2Dd%*4_8yR_`Wbe z8y^Sz&s+Ob5%lv5xzcTcuCzml)?Ha6E=@rZv>kzweRzKS%KNSxZp0?q+s0tG`Co_B z&_(e5%#>^(ir@!5W9Z7A8*(-WmS4V7Z8ga(pt_Y zPDsyeJ#jR?@efM33|{WWM-3j}Xzg4uPN&SqbAjra)EF2*8d%pwT!@c{oX__}XTI|L z=53rXd?fCPPp;0eU&{o2n5L3#9AJe;g-9ZZEyGrqCro=H+DuiI;gRsSq(Xy^h@33q$xiNXF3;Meq=xhmOQ<~k+ zR<8u2W(7nX&Pc*@Mgh-C4xY68A^VJ#Bd|Bjx3<+g|v}N zSo&XKW)4S}>2WW+J=vY+Y_&;AB~TX1Pz(u_nDl}xzbu(9hZwhU^zvB_5DnQlohb^K z;I@r$T$QJp%s(I9I1A0<Drv4T>Oa8>CX?BGt>_#9rq@hf!tLlE3d1PUar|g z&CS~)-DhC*w7+6Yfj8*0W5rfj%c%!gz51NrZQN zfH(Q~dTSW1_vf&Ay!J57PTkgFROnmjljKtoM|{T9Mh3>O`@t@Y!5DH7!b(VAg2k+++=uq0949rbcVA)OB%$SqBS$)CaIS8yiPVeouS zl>?&6+|S|Mv^8875xMWt-bx8ss2Hs9akhS5v{X8xY3;>tKrMf>p=m(6J`-76#DrlJ zxGUx}fZtUW!x91J>r z&p1p7LmC!PD2rCwoXbvY5)w>Ca8C~7i11`;1)*nu@p0bgG|3%0)kj%tN3Dx!(~`mL zejJBlPxPX*!F!Gw_IMW!J!>N+T#1gIne;H!K4ZcR5@* z=lJgQG?PN7e9bB9&(nRv1!7mdZU$g|zUr&1vhw0xvv=G8IPy!Tbo&;8^`Cw=7%`DU zH&>xU)g{vc+ShB5h+6GB;|zH@)lu-CD^1iRfKJt7D$akJ_`g!>W{`gvcxb1hKF4KO zcf5&58FhE3k*G7Dn~&Ac@ontM`2?Aq5woUGD;`p*3p zJ`ws;i3azzSSJJp7y0zTU&IlEcWA!W zc$a`KstO8P|SPZA6A@xmrMD-uO0W3SOC+fBmI@?D4=QfZ;UUTL&1fXxZK9N{}ycsGw!uR z0!GB~|BitF-qlx20e1g?(%Z>D80j2e?E3fUYySs*g&v0d*7ALe|5!_L;=M7P zzqu^`b=IH%IO|`evS+`y7Uu^n>X+{?7a{M`MR$I0w+!gk(Bp15dA@eY(p!Ao-#dd` zhPONZFi%EiHpzw8zjGr#iOmumf&8}AnAnL~pr^@St44@2(0+#m_a0aP=TX2A0Q#-F zgv*==+`R#lI^5^KH(pl$9?)X=`R{I{5vWPjXRi*B{mpIb0Ni0Aw}Sl}QUl=Teg;4) zIVQ77V*K9tF?_L0C3y_UzZvFl&*CjRK*_`bYZ}$x8y}Sb$R!=s2EX4E9gcDUYwMX} zMt}aj@io9fjmNZ8{2t52$pe(!;L;6#`g`MZ89p7QCpNqy)zvST z9LIOvFa#hkF`-w_JMh{ekFlY@qh90F0Aglvk*t7}snH+*){5}}#wW(piUIq>=(#rb zykEce)wals@7K)zLXz4a6-diSx)mQMCdDy8D0~}VqZQ5iPvIffNGs-(rDuW~?fV}* z?fV80gxjjD)>NZSF5v|kc(dh0hn~5&UQMsH zK|!6xLPJ)O+4AIAVRhGbvRNiKmG@bt_J?tmGM!iBu0er^$*DF?a;iLyRJ2DWLx`P7 z;f=#mH5;4pA|phU&B+GPdB}EiW~<4P>lz#U z@UGE1c{a=NA$)jZ!FE24K&mg*>D!24b7b<>nKk&PQYHn+Yh*|-aQyY+hRjCl4$6WZ z5mWQJqE%$X^Q6yHEzbQLzk|X_Y?)nc_FBSIvqOphe)pM`0XcaZNA^zpGU3Bm%R5vi z`&V{ToeRP{`4j4PBt9jiJjS_TQNtUO@w%b%B4oEf+V%EY{xndCQX-(N%$I6+3;VSf zQjlhn;p&)By*GI(308uQuYT@zhfPjzCi^uxLu?tBokP(>-L!=_f1Ko_*Qefp7}OjE za7dp56f90D&gPEGVEA$X`-WmVNAu`sLNgC{EaMW1=ZNz$&=NF-sO*AA6nO*wbv8+C zmi&abkkXT5AT;J%FX0H#yJD1QF)tnLGWqj1f6rIn(aqsv%{J*nAQst1d6!1mc#?_B z_h#*Zi_NV}Vo>F_@ALYCnK-R>n zV*UcyCWMspM2CB=FInkZWMK20Wl1u!$%zYz&`%Q_G$&bs-659m2K{xJl2h#(7g_Z# zoo8M&EPV{UOQCs)NS-QlT>pSym353Th8-hcFPooY)S@F zX*-m}=avc&w45j3RbM~o7;N~ zNMKf1ub;ivU=g&HQp7y`57V*~4N$ah+3({6>}XmF?R>yn`f3ZiQM1+gCTPR6xK=M^ z{lxFCMx-|bW+Ueb^Gl4u8RP8pVm}_@2wM)f9dYsfaze08r&73>M{&0l`i(sm%o+LO zeYZY7GWcULf-}@d`y+eny5~5uP$EaEf89flL^8n61|)hK5~NyT3bZ=e*$i7EHq9ki z0X+IxPM1XI2oo`D2_G*74Wve2N&+Ni5*2JmQo?84jvKMf;3hm$9#|(0)e? zjAa{;6KCiF{in5k47Lz&mDtE@`e&xvK8bIge@G$2mRLy3_RHQ`flil)cxzdle7OEV zQcnFz-)vLAIHxD-lB=sy3R^NR0k@`FX9BV*HM7{9mHzy7^1Ch$IWV9C9 zoE}Ha+uE0WN49Qn)lP@=X=?lG+?vq$%Tkw{c1&4vD}FwfZmwK#MKV|KFp?p3D>IAN zJks5KNBa#)Cy!I-Ca3Nyn{xmOV76Jm?1OV-u>o(Wq#K;lwIN+#*$rqPw|rA3bS@2Q zwXQu+*sM430heeYy$(J&ED4w;#VQL;Ew6G1e$USI=Jh!6inmA!pBfW#JRZ#g|gn6`O=QdQ_s89?koyc-*EeI1@F zOFxa8hO~c5E?$oRw1z%N0Y=p@0kW9g(<*g90vj|nDs;ENJy^s&+I5JF_iv8=0hTm;%zEQ5@iW#tb zSA~PQ%tQ`}Yh2F;bH{KvNPRp%fGn(Oi0&a=Ue&#VFM~!YV9_fx-qG8_o}DBR-Ge?- z7snnNRG6JQhsWv3Kt-}udFu}$5W)>2-{U8AeI#7@Ac?%?CENrIY^bp+Th3bNx%)n+ z3`yHi1LKHg)1d6zQgw!MqvF2Q24&#iMPlg?Z^Hl>8c2lEG%X(a{^8s@EBHkdGy9_zz?1TQ*>xZ9PXVZaQKMvzdo*o{HYJ zr_Tx@qx%Hv;|FZym0n9`ZaPoA?*2G~VP_asWKSc=P2~K0Q`@JA5e}&I9BU zuV~;7yKPpcZOQpuZD^XpT@XOu6dm8Q-iaP^c7IrV-sUYTC8t>?B&3=IWZJOPq8Zd4 zAr7grtkMp3v3+HVaIqP5ZLZ>`%CAA~(yTd3$B|}vfn}4ooPC~QYhsY-1xmJ(##RdJ zXRBA_N`zDG?#p1ejgf!56Y-sonQF|_m;M~AaTKc84&iqrux;(oCF&IN3^l2#+Jz^UY#~P3CTvKbgU_ns0p~~)a~97A`rZl zn}w;$ux`cTtmu|Rz+vvW9&jsL&5i}f(^}I3Hr~jgdy_cGgJZ#Z#g>8DJ$^h27JS}y z#U$yCR(6~=nB(XWgsdjJYs1>Bgpi_M&&~Hg97w(=Vt-i7jh~Ke_64%;FlNX1i(i)E z_L6SqECdw@+t9329Nkh_se5cB*1&rO2GAMvVs%S>Ulungw^Hpr##ZuRqG$4zO zVlC|PwLvAzjA)$$B6Cr&P9^8s6wy?>$Xy)*{j${_U@-fxZZCxq#OLz*?sAPbCtj^- z88GCmFf&#EO=H%9i7bWoh}p)?oO68J62R>swaecbk~oq z`tR${H4~nDI9r=_>w{=@@fS6wML5b^O_N4*{R>AMcYzQjl%>Rn71>dyXIU$MIy0HO zWC!THh8F38D)^qLv-IU3MVsDrSIqUms{>ObFP{*~aONd=2Zr3j-1o|?cBje-JtlmY zdNd6@I)txlxQ8J(o#JM31J*!%f1y}azo1Ncy^i1H=QNWJUh{Yah?#^*12DREd8wAT zb!{6PEj}<|Zp^vQxrJR7xUU4~C_C};E%MLRST*a0m2A}4G`Bl2U2r{e=_;3mZ~jAy>t8S8&d|gc_`?-t|Gm1 z12Um}>EX~{AvtDn;4-T_Z&O|Ct|NVwP1wn`GFkS2xcU}3lL93I)-qh=+4tG(i?kiVOLRPS=gTQja z1pe{^guQRmc2A$ijR;V44U4-boCKwzZZH45*UL6M3YmZKc{WDa>x?bWuG&)I!M^;Y zOsKu8ccq!vuBpzW;Huo3_JzY>@1`V4>5}*FhC)d`3pDxjb;NsONU-B?>J#J(B#T@5 z#aH$1TirUN9+oxkHd+nqrDb-GXw!T)T~>kmo@aM)!+h1tIekIXhr$-zHXW>ENGcAEHh4RO#nxgZf!n<`pc`IXC=3ZtC{u*R5JsiZ zS-&^;x=pf&Ep2bC^Jb{Nr2(UpD-LLb-!E&WUE6pPamY@9jGTR_K^c?wf0}f8hE!j< z2%$djE(Le~?&B9HycGU$xXkH|?HxfI87_BU%f-Qd-?}`$Y&(1pjbP3i=Bt=IFj;Lh zoBM{j*@)c ze>(?`|0AHPr6k!={28|!`XKcQF?h&-1B@T6j!x)$W}G*Q7ipT^ zliiGbJ9A_Pd*SzN<3*^js@g;c_q$({Z;-car5mk)U4 zufB2RyW0lFOlgjAkjP{F&9fI&)Xjp&-pL-_kn(C26sU)GMf&{HfBS%pWRtIH z4|`U3iogVT^51tdYsZ_Y2M;EX9~ZCHmKfW@ZXNwQs?HGB)N_igxG262Vjt@#(C=b@ zCI7c(FE?OphkBH)np=YXhnkIoQ zH_9iWOA_R4A{gklPn&pg_{7Ln2RO(sx(p0SNnDG>X}@1YH7s=LT4iFX_3~GElzvaq zKJX=2&PKPtdGdizUIPB`_WY0%*t@_!ndvv~{6S=@PZQ-EwOQMlitt24| z8A`hWsYBH(znAt3`qMHF+EyYat84mqzJY#ltO0e7rS-)U;*Rq6I1h0zT{z+kRj*b) zh?5FZZNfK=r4z}-x2{zE?IcS>&(61>vp%0Bow6m%HE%8p`waYmrTTJ;N%>dLqeT%R zfnM}0JuWaj&uXQ8RI+40cou~4e-uOG4i{a#f2Ss5qD3MxWCP8wL0_wc^6E``3?#7M z$GLST58@lX6D0~t=c}uFUna}z9Ly-o&hAnWgF^$fTA2?zow+z80#n$=oMJ4g+VB>F?AuMn`&Q z=$a)-AFNQ{ar>2=oE#w&OQFg;c*^tG&fmPo#Vg2KI+L7`qN-3MJAAG2_LXyoKJG^c<}-jm_ClNw=41Jl z`MLfFq8@lZ9pcD|(kPSyx`tStyWLZT=U=uRLDwKW`C7NlV#pD=#Rv8}1e3q(ggC9c zso#&l5ao_PH<7}B7uVco^>9vd8FZd>4|CD28u0s& zM<>aNd;1_n;QQ=6!8Q8Vsg+Us(R_s7Y{HKztQ5(2g^@jZKc-xG_)1psSH+Gw%R(ju zeSMF}*O7dy#!K+_%ooKTU=i-Qq{8eopDQzlOgICD)ES$nn-Ai_<#nz-+b!=;AK;c_ zkMWF++BuQPbq$v%*_sb`;=z~aUb?QAmvTn`1rQ1&nzhlCO{XH((4Pz_3uk6u8ZayM zE_&|bEBMc}jQN~1=XU1GfZDfPSu>n&`!RFpO~#R9Wx#d;Er~gL)ZTLsnu!kOM?%AR zrvXxr5*yY$14Yobf~x8fA`~MvAAC>FvM&hwmdbxD=KP*$9kAa1Nz^GHES?#Wjkd;WOSxm00ka*3(`UQuE4L z{bMWzA;y5pqK=%+pC}Wmap_of)(x*pu5->+W*3AWjy_N5(@*RFSD&>P+)Jhj)K$TQ z32CdN?d2h&uBwL{`yt=n3ZXlh&xUVFC`lJBJ!#vRdS{gujM7T0<5J-7Xa5u(qinl)qkg7`dcw=VV>Q-zSRX~ zmY3twRCzyD2m9vyV&|NvW`gMcvR}7PbE%a8;-gM+AAY2zYuCg3DId?N%xK<7S{HRy zVRRYxej?6B8=-vMMk<>K%6EMJjAZ{skb-Vc`$@U&hTHnq6xg!J5R>09Z$(Y;mPPcB z&Mz%fcJ6~wTqyk=(B`tBjRX*x+gxWs*prjwy{Y7erdUbjME7dZwa+w`Hpur|U{xP} zW6Wnr^ett1qMam-FOb1DzqvU4;qeR}iprKRnD8yJL!!E4ga9J|`8|I*4;U2-5q4L?H zd%`^5rik>YE=z21(DrZt zP7!y#@p-s&S?#x#OT@?qYj;xz`Tb>KjJlQ2Rv*zTdfDzSdeT+V>Afwr+)~-I_4UAq zm5;}x38c4ES&g%ad$GL#38u}ch3%{@E`zq&D>+E)5F#GxOYq&ka@A+FrZlZxdeZn| zPP5Zw@m=rtxu~$i>9!=UJ)5wMnmto;Un%;Xsg8-x*wyqOg5La7l=V5H@v%w)2xTC@ zwk^JiFg!Lg8IDK!NQ5e#9?tL(P`YHSf8BFqp9MTG3g=_Am@1=55xQ{4gP2Yqq{~5y8|1@#0*2Q_1aEI%%Zj$bBwwF@Rbi^Zlyigz$!)g z(#VxACuEnYRneTBTUI0Z!`T(R6Eiv*Pt08<_wDR6f+yPW*&&_86SiK2?<_41a`af; zbruc(+7&ecjKF6U+Ir;;)6cE;`$vEwV1EtSvA+(rW)?)W*p;@{~f5+3?Uun zO_%3~9iNHV`c?D_F*$bQ_AxAWtzNz#?LSE5{C3iGeMEzL!sq;9<(KJwZPw94aNezQ z8XnJnOx$vyh&ZD4CXzqKP6QTOAGTF@*MD;+XJAEyY%*yMN7wJfT+H#?U0-A85)~YC z4BO3n4Vhwok7jia6rAY7ybRA-Sj9}BOa5 zy0n+SKE|`Owu(X5-)r&oZ4?~4COW=JC{iR84SWrd#<7crJ=Py{ir|wX&Q1O#u96i< z)@gi5oeVpb^6%W-CuWKtI|eQN`~9GQN%5$LI6QlYsUEKJ>1vyt?Zi2%u{}3OtEhIs z;|}cpOegYOHDTr7*I5Lr-9_|U(i_uJm#D)jcJcE==+U8Zv-;&(+pIuD=-(&=wun#y71$B|9#8)e{`n>`&@k$#(S3_@SGVya-R>bPV1da*ytg zZG9+nr80cBq=<9NM;cy)2!+4Wf?s~vJn@!O-ob&~1#odboU72uO$#^%Jzks#pG!Xx zW(B>yiC;%*i++uOxK-NfZo!@qq`BX&pu<<~^8(gbg}Xz6^HvdG2P=1X{rgRHZG-m> zia>Q*wrgzfF{OxBl1!jwkOR}+HN z(TJ}mB?sB9Zp7NamV0JJ0{e{&0kZM@18SW~C)_^y=TAQI7!yM>tHEVxyvs}$&gTBR zhCZhgSYt3{KR^Zy%gi;Y`fJNDq*x86ep#!j3ZPuAsn$+6A69Ly7g#xPF6(boeGwrU z;Tb8*T8DjUdd;@YTtg$q4dIW@JV+BA-VVc`d$4^iEk^nQ{@R25$ht|= zE|$;Ne$#6F;*3w^@WGZQYw|9vm&KzkbMjtK{)Qo~=}R+p>on5w>){9EZ8EEh!Y!fO z{LRc_o+qmTL+vd4Z_+Gi;0KMly6S6Gmk4JKvm2MWzPOd7=dew6B*`-w{ z6T!UGPuQf7eK^wfk4v7P;DxZ^#c9=ixUBQw*&il52avDY^Uaz8H@FZNW(aTbF792a z>o+Q%MIM*g5^|20-2!hrt_Mu|k2VSV{E66)A`m&h=%%Y3@=oq|E7O+71%3Y0ThesG z_9^y$6{(|@065s$y0KXVf2IB_%d(?y%CA0`qt&Xkm9-Q;# z_|+i)-j1{_z>foIDN2=Bk$D9k743VL+Zhzy!+)nIKKu%jH7$9E)Z6aM`v##oyC4`z!Vd ze2Rf_r)NU+*(GT9Y)l>zv1~T3x^TAshM~XZu?vZh5Km(JBB4(VKfEUxNj8}I;s5SU zU=K&PREFE{4ItDjF|oPzvCI_@&Se!lZ9xO7zxmCLjrWLIxt44DV&tm*Wma6R^1<%G zV9`DQ{`;H}I{VjJCIY$R8BVGOqwdkZO{osk165Z~Mzv|~)tT9gTy>727S*ReTbvU| zo1mL~pnf{)f;Wun+WqIGYTfP{`|m$sIhUrk`V5{Vy-*4nEB?0ev~k|}^`n%KRf0Y& zW^Ov-T`}eD&b>wrjvn5bt<%`vkeCQ+p)`{AZ5>^=w?C@!i3322e;-$9I^4|XNc7+S zwr_ZbKJzFqK1L9ZNjYBOi(}%RlNY@0)p(`b_>P>*-E7mD+uOjL3yYa_Q^=)yx@|sJ z662lckC`|f7i1a^YT;D;+sLfyJdVeE{olY^YrF7(9|!EB)<3l;tz;>1%%+^r-Pg0<=)W_9 zYfAcqLX7*h69-nM3?FKmyzrY4B%g5UTnl=C7GB3S5@jQw6+5@yp1~RU>))&>;abCR z`vDTfBbD{z?6%QdZYhU!R(T;w$A9S!Ek7K%yV2d2S=!IGR6PxtpJt$mJGWqyvr$wu zG2Nti_&%Rby;PS`a9&yb7WTRqwJjZ%74(3ayZe<+%S`g3_`9cf-?Tz=@}OIfOaAAA z{;P23KlioEN^t$9(E;Xe3d7#?zHeIDz!{v?)9>)iaO;TlF5aDUanDO>5to)GjlB5~ zK|qWpi%kgOQgq{Tu>P|8$~RgW`GXbspWoi|qD0G^gb_aj@I>ywTeuJQF@hdWh+5%& z`H50n6J6coj8*mPJ#)EncsHMi z^b!+V_OjJh&?GA)Uju(@l~o%DsaLG5QO>VDKh@=Ylw|K|H{g#JP^Y{OQ)a}ziZPba8Jpb?%AzZ=LUP`blUZ#9$n zs^R)p#^RhwV(r+6{@rZ#QZ+C0KAVH8G#^v^*!Ij~V|S6O_FP|A5>CT_5bkQ2Ejk@c z#&}={$9L+Ed*2WQbpWido${s)C>T(HtXh(#I&K>_7Tq{k6I|8vA>2oC;;51r*khu@ znGFPXtGetnkfoG&hRA&RQek#oRYnb>s%z`bjFqd_p=wEj=K8cb#JF}gU-A0(RHvZm zS*K`SQ1D9c%ic9?&D+5A5P75BVUz-5ZcIXF5m8`81?oTrXetSXM4HId zvK_$?={IRA*tca2+i1_ww=MP1a20HKS-uAWRqMgA0rN$(PtN66RRrtTJ`O4U^J>S9 z3p+dIUB;Zj$_q&Rd{IAqOKkbr&*Gla)r_IP#lfj)eA&)>xX}tET%&6s3B}bXN!{Kr zk5tHsS@A)55{bEv6eGaiKDql_-Q=0~y_A0IKX1#D9-Tzea{p+Lvf`QCH&fRdMsLRx zQZ$!!C?gXxWz1EKCT!WP#Ft;?fC&7QFKwzR!8shhm&CQ`O-5K4@%^z0OY6>|tKQpQ zCzKQcCRkvfFPSKw^ZsBy!!gsI00z@*(rWf{4fJhq>i}eMRf*Q+dk^ZQrxePXHwtgB zPtSMKeu(1VpUZ7VmKtrYrDz*&u1Q5N(ImZ_ta4t8nPZBR`c<}kaXWS$DmysD^>{dD zA}`ZKGiPJ}rhurgGo&QIwx!uQf>6vemq0y0`&zx%Yer8iP5`OBjDQlz9Rfn2^TQutcp3q}KoH=ztWyX9q`aAs_d9vp zw=c7_$F50{XiNd!wvSDdP*2dK5y)!_>JGba$4ghq>S)sWN$ zubrEKGDYb zi~&KHVrLwdHx%F@F@F_eK^lyI6rGbRowOelAj1&0yvO!#o!Re^E|1&61pUO!@ zfhgn4>V_tgj;HOOTOYJ@kPOrJJ#3y;kkGurN5_Y+u(?F#9r5Q%h-L2rhIXN?{yYPk ztI=fWwBV~~+{1Ox>Ou6{x4u5aPBqA!C$M(+#mt=l4=46)S*9azSir)3_>R}_zMV#M z!8mjE(x;L+MeY1J{v>+|m%j{;m#>uIQ6)dQ@kUB2Mj8F5BPhAPI62}6k8Rr;b!t%@4PRj`nvwmNp^F=Ql5C)|l}Q_f znV3rb{g^L%_njDx@;-v*7vSp^0M(Q|>HxRFyf9)WvDXFO;aQ$&KP{ee{dId7H_<%T z<7wZ^oykAeCirVtUwhthh+JMlXOH8eZa>TubCSBEx&Axf_?eIj<5Yf#i=u(c{TKh< zKaysU)_XNSggW^RmHp*6eU{d<(J7YsN2$(nXfyo#0rTT~%kOwssJeQvmv4|yU{3#i zJ|nLbL6twJ@)hF;v&z>I9m##%xj_Ujee$)~-b*_PFTV7t@tby!qV}o~ zhBSRo;Oz>#kHliDrswWE$CdFq0R(U`Z8p&4Dbelkd! zlU$5Gws9A==KeZgQoXB~;@-q`aP2qI!5W-hZWi-rw74p2SW74V8kU?tm4B5Ae>$dz zq5~?c9+9x0o4-N~oc5SHhmB*14Ovq&K0W8Dka=s>k`e3%#o2lbn@oX6)qkVz9`r|R zE6-TDI7sgKgczI75HNvX6-nWq=xj}M>FEd`hJlTGnKhqw69=YcCTyI*G|};QiNA5+ z>59)<@D*690)~fu--}x2*SBneQHc_NE>^Ac@61q{R?CxKh@76wEfg2(8?A1I)a7P* z5)zG8Z%-NMr;aIQ8j&+RKfhDeGlEo8Gn2z~kV_8-#WF1Gw;u0C{g+<%j^5>WSK(Gl z{kA=4aWwPIXB#Z}pX-k|`@(~7zU(~BbG_!2ekr%J56(YfJfDJoWzV>)ST51}@G6|j z;_2-~LS1pY3%2=8CqNp(1X$}^v10zOIaOqaVvz+aQpm80F~$B#3IOVs-o5%yyWrtm z!@OMY?0x77xF(aLI!cueiksjAOJVNoxg+CEY1ic5x z6kmA#`;-7G=sf^aOSl=Fd4Ds-dmhXA<6v-E7S%f7;fUHx`^6sbr&&$Y9r)@(zatF7 z>Q!Mylmp(4`;z8My9Q3(>U)_UG4o3(*{ErBsKiXx5B3DJreFF5GWO|}ZHwTu`*K(N zOfbissW>{W!Zu9^1!n^5+ZFBR?=eQ0_C;_rjbHN-lA{-CWq`)kil+wb|Bx4_oH^?+ zOwDkB_ck83#4ZftwIQiK3YUGF|Clkj$uR%!yje24c-62}K+$kt*70|q+eMS-+oh4p zrVCR{v4Gbf_8t54-pXmAAbw zLGFh)>>Gw|4bW5m#4pd^OV(^@BJf^NoqT>wYRuP5kDmiOlZHAN-_z{b%Vf^V7CJ?S zEdZ*e)3*E#wV-tDdKCwWLa4blS0--p)0Yu5CEo;cI{PR%r8VPM?Pbdy-Y-H*K1-Vy z8!I&$+a3-qo9t$oHe0~7G8bhWn!-U_Yjex8xU!^q_5XDQQQG zW=hR)QU8D*kjXH0h5oml>5kr|GyKi@8*)_HCK&WAPAiTVei;fnDst1WuHLZys1=Ze=1o>E4cry9uqxZm~$o}u=|qL7!ZMzrq!3qjP| z(Ge7{bqbi{U0$)prHS*}9@VE@ix5KN-n$Wgmmg|JQEy_Dc0u^v;H9^GxSHsL?=4q| z-yPJ}c!fTY5`~6U&DO%ltmn-bavnmnU3r%+23k7(vJ3-I-UadiF2MJlKM5@VTaIli ziX9Ye@}jgWE5S0b#0z*$U;pdKmEG$*WL#=eGtiekAk(b9zYcn;-s-a{0X3Ya!t@uu zT7*wk_HUe07?-qjJ5g?V)bnmjseG@(U)O@#TkA6UWddn3bo3l1xFJfQ**++Qb4z1Y zhGuyq8^k3go<1RQk56dU(ojHZAGl)L(Y`1*3*t9}`!g?X9l~}k1_3(3Ug2}AdmT*X zk-B(;Z%Qfc?QAEuX>BjN7f^54m1X4=@OQy0^J*>9KZ=0%M<%?%zTVaC!771v$qwM6 zxbZc{$1kY9d|A%9)+$h7Hl5=9tNfvuYkFyN(ar4Yu{NW30~=E6YAIUf~bj> z3$BV;N<$U20Do0YCp@zP6qo%zAFYVjWZFHk3eymE@27tbMEFsuyLg~t&p-}qGIGJT zvCQ&#ru9&mybRRvVH0COwztVl%#UxXb>p}H!-*qBVcW}VYW(Ah>zfhp z@+|a(oO@mO|1kCQPB?1pKT&6%B*&~?gBR0*iNUzq-3)I%A7vcS}i3s6DCOv*|nG{6H$iw|;)X=Z?kF z#7q3Vtih zm3DakFXf3BHr|bWpGW=_DCw_J8_3$KyM<_C+}+DyoBTAg%_}r5)gr=O!Colez>-(Y zrFs}0bw1N8I3;i$z#483NZ4=b?YCgV2!HLm5yB~-wNMLuPR9<)VOOy(`v?vKi1;X_zi z6OV%U*cItk?+Of!YKab9+U9nvskQW8xp0IdO(E-X3DL+oWp?>GrFs2F!TS9QKjgjC z^&mJTCDdafaPu$@nvr;Q;T3Kpu#SY^8`Dn!&*(0@)^aVeQV`fF| zMG+|v^N72q9jocaoYPIzmQlI;)*%aS@A%` z(@1S^8jj`~Te0vW0_a0wDVDp`|18vU_#a z6IIQ`rnOG`=>f4F4f1aNEAb6(LUdWOMny&-21G_jPgc7(msM%3{M?+@P}#T9}x1aEJI{a`wU^E`YzI0li7?Luxoh%hVamHtzxJ7?J^S7Q9r za`Z#>X_+MVWxjes{@j6FZV??<{zM3BGpD^RH&Io`x%z0v{6Nzz{M@}5=FQmAsgS|( zaTJp?YjvQn-H`fh%burGswAPfN7@(cF(Nq^>eM`rHmut3&{Dl7>VzRd=}m^I1Hqm4 z^z+y30f!7PqiJhXw*rn*cA&3a=;EWz=-5~-NYRkXL8$b@&hDPs9aFZ*?HP>W2Xl#qItPa zd}pPsZC<8F)y^^vGzFUk)gc+c=UjqUSSwmN2aMDc!9q$9f8wWKKGFO09;wi02FNh_ zii(}sfo`mrne_hE2+JI>|5N(|3Xo-1i_dlw7U!0=qmJJwi6kHF5qx#UR~P4uZ~Ug0 z>qNj<4BOZH%$d#}lA8NeJ#-tJ4+{|2ISgeYQX@n33By;7b19YtWtrd1g|@fMtzfTe zqtlp{ev;OV-`*C5Iy?H!g82>4!8B-)>!V`7l<&i~`6eLhH=i^5!o|;B%gr^xP2?%L z82Zy;<+(+NOpUvA7_y}9f8KYN%2u%bL_;Tyjx=xz4?kQQ6s3eO%JC!@e97f4UweYuSVNV)K0fx!=E);!5J7y}>84uIIh7EU=#i)mTa75O<%wy{e#>=eX?QYUX5g>lR+E*`)d6;pbM&vPk^w2nwtc=opiwwXgGQ~a|G4@Z=RT~6Bg`v|bM%v!*~8790- z;2#@;-%{8>j?mrV*WRAg#oS9RG9{-a($1zC&VM>+%o7t(9qx9N+n2dw9AK3-uPP@7 zJl$9>&O83I{$t)->WS+&bh6Cn>w#k}8(sG^al(k|+UlM9f*6G?+xra277)l7jT*T( zZ%qQ2>qDKdZ{{Dpqjo(4)zG$9H@#nsGm`xtez<3j$PX&cY1$ppxIK7QuC!m>amL`g zMF@Ch{zrMACf4L#AM_+?_M+2${P2%v5V>(0t$gJ1T_3fk0~?rhz0N|^;0sEV#yo)< zvHyNJ5h6@qTDheqn=}eY1WsJ%*7*A1ljaA8RP$c<&N26%;JIUv@hIPDJ_;myPHU|r z;=};RRR{UrM}BBYu_U>ue2?qG^6aeC^zQnwYY7@aJ+bSR3FS?~Fpt?*Z2*s$bU$i7 zW;_~}oVsbTgCpgW?@PV=&^gmL}h%xN8M zI~aUpzV&(QQzh}%Z5wOgn89vzQ8fzgZQHBpGFxH~z3IxPsy!D+gK>oa+@;N=FsnycWY@z1zN>{UP^DdHz1EA-@%(z09h796MlNYxW|juvSA zAGY{^lTja%70!shERo)Qz3v^c{^v^7=UK&8dgbdAJM4Ah(nJ505G9AX9Nd*3^U`F| zA`6obrQ?K8jwYL-^0tpbc6o7sl_pUoD|L)$P~OUJ|B*4J$P>Fl|8~>g`WC2)pPsFY zrn>Oq=8ca@zmWCsjXwg&M$Dvj{$jk3x95#Zvy{(>CPktpPOiG0aMv=tn-tYlo((N8 zC%3R~yh%4Zyq*T)U?U89tR632&+}06A7jdQ6$#8LOGl82-B!_V{3Ln9Su4S4nvB7} zIN)Txrb{}uLP81-U&RJNjy>%U!qHn|Tds1^JdBAM5BaxCsa+BBV!!Nj)=?Iw zl^P`=>e$Unu15>`rwdC=KBRwX>8uNir6wZ%!rwv24k7NV;g{*WanSLz$rYodoR%xX zv3|Qxh(()Xz!y(FeuRVF!HY69|L%UpqtLPt_kxc^HuaOuO5MInT;l_0!M5jH>4s&E z=l^)-yeVcB=Wb{`Ek>JLpX^kG-gxf=KrI}t=I~fb+bB4k{2ejX^o2iXS-3dK3b$nP z+C?p=dUWEp>q?eJ-A(gk5JAFqfM|BgW7Knb(nHzXZqmH(oH&QwM9oDZ9;5H;_*5)# zg8<@TR-Y(M=ZEmGZ?K^IuYl>YnJc+YummwOX6Tz5ZDlA`T21D z_?N{;p*2mpB|pB`Rno>1GFxPvwPevfd+K+0Z1e;(ATuhuslA^m@Q`7@0q4STeAAgX z`ZiiZE`K1ZrkX=Vi4!Z>vgALK6QdT*cQ02SzeTyuMlTZqVNa1)S34a`#pg-3 z?M`~9NMQ$gpR77N6!)1I(fD^i3|R7MR?(<9F5WuW1gi+nDH`^Nqc8D(FYtF=Ycr}< z8H@ddhoTvOpCk8Vt+Q|F2A{znU)IKz&ICo5(Z<>MYk1D{;|wQwdeh6sXAA9f%yX;5 z?+@HR$z@Y=W3W$iqU2)FP1`Xihye@_9cH?uB~?miIQiI-L6clbeqa8t z=pMttfXBf!d^DqLigJy&O^~$lZht(iH(U9ZJZ|f)hVNsL3&4>ISA*67U9fZ~EI#bnt*LDJD&9yb3_AzTI}!MDCDGH2mkTT<|@ zr&RLC;1TQ2M#f~qv=>C&lFf_%el%+-Y3Npp3s&4~MR&ZIJU;a)vsLz=7lsE0e>4ed@i7 z``dtbrgzNZ`NE}eH2tUP=r|-yBD|J*_*g4ke;VMJQ_m)T3A_SOv02keehz^!q&4_mt@#~UrbD@N>BRR0zvStW|R>_`yx#j87v+Xc8{%p|?9s>d*h?z4187S@dFoW*r26$_8 zkxD^Xx9oe|S|bRK8xX)>rV83p|Ap9QIEGtdHW`4YQOWXV0QK$U-_Qnxa0z`N+fJnK zc5_)h^&l2OThFYYig+sczNznp_UccuwMqN5vTk|o@1|0jNF;PGGj#S>QSUpad^_7b zT00WH=~T71T#C&D`s2O2Se>eZHY_JHxRjO|EazjBAK*~dzghKqu8+YTJ6JVZ-@(1m zVCzpF?Y;}r6A!;@VntoSZXN>m6qOopF)Le;chz)vv}?>f^IvbK`I-y2&gw-)!1@Id zrJGP|_$7ObkbRJmuGIYG)rHMf9bs=b?fpf*j3V%M`R+!iUSF?+C?WHG8~gc{efJZ2 zfjIWlB5Qpdqt;_UT5JW`i0b`Ut4aeBmg~t{07t7BIO0XuMDS?tEnoWUHYC#5n++gD z&mdfFLf?bY+slc<%6_M*8kG1<4mTF*GpTSy~!04KHefz>1^KW*+ zEA~~{mAm@hbK_(OuT}KPwd;EpqhATZ+=HHF^2tSgcBA7f?upC8#^K_t zd-v44!}jv&S<5SB;w$!e0@hGfG2%c#jz=7}C^p&=+Svuda_T;jv{@VINmEAXmeOrQ z^A#{tVJi(GAdNZO(68j;8+BW=w47!nd~dub%u*aqLS=h?FfR|F2>*%H|mF-*qVYfh3 z+*pMBgaO{j#e~(Q_EM^~zH*o{eXQr5F6*e&4DKx)1Kl1Tl_PBv?!itC2ZQ6o^0iC$ z5`ol(n9$Y!YpyG(+S9GS^;C>~n`yaAV8#X-_tmisy#(KKa>eg>dmlo)^GWx?5_O?L z6Q>0n{)Z^eJqt#wjA`I{cEWwKLo(tqeMjX>HtYruaAr*dpNDf7@H&S|!ajAc1Qpg5 z0NbG%#1#v-vc5HvYoax=zuC=uh1*QL0sGUm2PFKV(CgakgKsN~f+Pt_tZVnEmp^lt z++M3G3Nkl|t6F}q<&7MF=eZFx5`$%jXGYmp=--tHzGa>fpzzciUQ~UgHnCq3 z=aP{e*8FY2^!)tO$2b3*i2s}BCH)^g-NIaYw(BYXg3Ow~AwE|aMyg-aH(+}3(8c%e z3q4b?*V4|yAu0JlC`We=xFZyNCct(2fzG7zVwtX2CQ!*=id98^5c(&Po7+`b`loA= zpE3j7YEUH1K7IS(>%(d@;W`)4yjuExr?-u2Pgox?|6-(yp^G7TB&%}EHX`m6h+S^Igf28w@4 zK)YFQN;|XqN2kt&4FCHFz$9NEF^~H^hZc2;PN{vBy{?CQW~aCiCUu5cKI-6kqbf!8 z;Q>JCvTmrZL%SKC#EWUonJOktOeG{meNFu2nScoTm+YLp>_3o6F227}Bz3Ps>a9Id z%8@OKu#7Wumm<1gJe}O;l3Y6rj|dPF%y+edeF5;|553H>4yV+h{-MWq>%?VaNa}q# z{v**H=pKOblpibw@lc0=NRgT)AC}?dj(+3_5PX0T{yF-b1n7pi@1xcHpF~DJ0P9?P zh;8lvr|jdeKVMD$6kc_?X$1QWau$YufT`r`LgSv!@~g~V`6TT%Q9C}&5i|UiB^%bQ zEx@?gT9QBmT zZ}Grx_PV`(q;lmJik?}de#>YkMY?bo-|BU{=czBJ?33RPiqN7DX_cGW}l zA}*@&RV|lAc*Y1IQ`MAhl@ploX$t)|4#QYlH+fvXgK0|+p7$mrz&GiWp@tna-2SGl z6&^@@QH^dbski3}9ON4PcmUpPO*PT-&Qd0h?=|0ULquSVwKtxP#&}jnKCUxb7yF!g zi3Nkbjo&m5eyhqG2|%HeJdf+iBX?U1AZ6-RfOo?lJ4e-PdmhCo>PV{4V@gs-ttm?Y#zBuoxo9WaNB+2 z7XV@KL$H^}8*fo>&>8F0W5g%g9<5zC=IVUD0CZt?3vk8N(|3VjCNSJw$AkB^q+e)0 zNgF1&NNHw6)=xaA-Xyg`Oe09w#IlZql27@i4sY4-HqFEc3vKT7G-L+s1_WI*?D0${nn1q@DCvDNz^gG{DP zK6$>QH3o*&TTJt=Zvy;^;NQp~TixO#&*b;r{`m?x&RsuQW|uTn)?UKt z;^FvXp&8rA309fN!t2oIU6jHNZZ6Ifnfi=$sp#OrsJQ4TEtH>n)#$H67*-{^m7DZa zfD_wPl-Nx{_r$b~-WlqDAm$+z7$jf)RVzbHgjimC=Oy7K#uru8iXh`zi)=yVp4E4Z57R_9%aUue=TE zK5+K@zXR3LGwW$4?pDxlBynoOt&>9#H8A2qQOoDYX4M4{erVVe#lcrheImN8UVOG> z7AbqQA)(8HR^(pTMq1vDvAe!gg|OU7_kZoSzeSbk#Vbf1^BCyHg+r7; zx$w&Dy3`wX!~9>Rwv8DUOT0a{n2C0dZ2&y7_^Pmb4o8icK5ON8ZSo6;Vi9hllHM)ef2s{56D;FAYeIzPtW>*Dk_I{8pzQe5xe57_ ztIURI_veryQ`|$fHUG`bk|-)CUZ>kGoEVaCaKpxFC2=ZZ_*ngbtd3P;OzCWhO$pV9a73>)Xe$BbIj>@yeGb^WP}7BP+95V?*J=>Cg`9MiK)+g9(~-$NnPj^^k`23X%o=szb4fPC4f9O33Dwl|_8d;Rq}r^Ad!ZGu z;gr(a_#RiAwVWX5IBf}n2K=5LKBHC%jKI_;ZqL=dOu8vGLBEY|PA}xLe71WY7f)jv zZ#+SIZH);Nxq0^15+%!Pe1==m`^`U%Le}Z=w9MrH@R7yqsE~{=e|h`CA4a8%HL`>G zU|zGjZ!!+p+NpNq^IH9^+F^RR0blJTnHM)-cy$}M-~9oso~~2i)m0F2{{M|8yfc30 zZ#EDJ2r_(i`b7Sbn<%?M%hy6cgc{ZxF?}xJsYk{H_u3$S0ehoFSoxdP!w=VM*fa9~ zmE}B!3d=d%CA+q~j`k6DPLn`Q$-f1hlW8EXAW!}`Y@e#X>K!B6NSL2Zu(WJ`iTU&4 znf%$uk^3%A)k4&{WF5tcq$jftmVfx$-E4ayB-XLwy(0=D9RyUmC?KIo6Oi7UH0fX{ zA_+A}6Ok?*r1uWeTOuGJz1IXn?;-RS%8loF&e{7sW1Ktoc*p(YeaCx8@<#@w{K}eh ztu^O&e&%`F>Eebq+$hnOz3YTWzrIO5_TF!l!&FyisMYFoMo4Efr#!+HPWB?_vBmbX zx@0wKQ@^g%X?LqQyVPOIs(5J4&k6dSnzN?Z0}ia?tEu%YS1dG1a_*?K)orem1`yeO z*LaX~RuaC2ofV3JV<5D>suDbR!|fTXd0PPgoD~ibP%f!uc&hqr1~{MU3}A9ctppzI z&2y-ZjsXgIdvzmqKwoAH!Vf6?rpj?L;9dsVSz8vt6-2V}1BDX$FrS{U9OcfO{;+5L zU+e<2H#Hbu&r(EuN9B71(Z|5%UE5D&joDhcsd73#I9R1- zr#ju^k1sdxWX|)OK!eEt{5bYERyeT$%T&!Lw1mhafknbU6X%&Tn1bqgi-SNLB@i~qHp z?+@&M#pC@8v8YV*uzUUa^t#*$QsRHb#%AW?)});ZcAq}0#)f(W9KO3*y4ls2=R2W{ zY|xWEo--v-7naVF@Mq1Jj%fRz_&fKZt@n3 z2-KS-+@14O7y6(9(~`Q8aC}%9^fCxL591P=u|@}Ae~MU9^xvoH*J;0}Y04Yuhh)PR zEBkaAZiv9Pu5m|cZI?qQ#UKA>_`h!7N?VlJ-h8pP+jE*Mg7oV4Ih~_$>N}jRbx3J) zcS!!t7Br#Q`5MASg^PQjY`(u_*8jUE+W$oK+HR~#H|$dK<1mYc5|IX4e>9xK@UQ!l z!k52MiAjccV15|i8DhipviNzdaOFWZnCUP!^Dj5&K;dP;=W8tRa`h7LwcV6$f8%@0YYdg!|8i*m z`K1mS_J-@SXDD&}wZQ&Az2E;e7Hr%+YQszPxA46h&k3>A>j)pt_`jb1e*|nl`OUw5 zom%^l^iSjYcWmWfKl*Zl1xrnSevqsq^Pk`Jr(m=2|6cUJI>LXApZ`x^^g`@rHg7=i z4*tJR=Rb|yf4rdzR~Xx1U>eie*f^TemvHZHx25Ez2Zt211r`3^8UD8jbr^8sU!9eU z{tA8k_db9{`1}UTdj`w2{BIBKUj~x7gk8gwi+akd{11fYfA9}1pRnG6^sdui=Yaog z4=+1eu}0&1@FnSQ8O_I+q*#-@{|NCn>DymqvF>4M<(^>hpN98WoBbbu^}qj5w7j6^ z(%iOf|-|>R}uVwy$HGe$fz5g}Olep}xtyCXK_7lCd-0^Q* zpZ_pyG;aL^G#v>s98l`b>L*dO2w>2(R(OY;xp|CitX-}qb(Y@e63 z0@7F5t*K(bzprxt(;3TQS7q|J2`dbXfKZk5CNm8Ub3G&byn}m1Y+dTqzGtYWw^Ggha*|SB`QuX=!9P9mz1ZK2 z>w~fYyD(Q@XEH6e5bkDw$o>{z18O~*s+!^NQiZ&YK1mFvwKk1_o7oE7gI=xW|#8RD3 z;8R3@qQ7f~G(5e3*Xg7>w_48o-LM}bZ=ANamOs_0&1YU*gtlQ>&eRgYJYRbie-R}L z;(kdr^D;Y{|22-~fz4Qi;0tEaLVqK|lF?91TWPb?^D!(=l^F23ae15C@VOIWc)Zw< zy-Rnk%;oYaFg&fsMg^N^>d#Mg+F62PDZ~nCaT^$H^56~Tx~Cu`nIC#KTH3Vr#3oi% zSa={6^)6+n#d5IesCH^SSzR*Od9tXkZgafKH+1RF@f*Ym%L6qf811Pq*}9`=O!Dh*Gaw39^z=2Scq=!fIi%Zv}2yd0`kMM4knAh`L_Zz;3hd&N{h=D#RnIUzn) z7#Vbv0vc+pT)p3Qqr@0diJUAmPI|NY*!_nfc?Yb>sP>%=2AJB}@Z_f6DaECk&KE_N zZy%f1e{88Rjbti`Q_9u1q9dyh9r;;YQ9UTO-SA$&0HkUxRYvXJe){yG#tZ}<&nm|B zr*G5~{k^rdme!?#J|`Uie@3Kd`SCZ1^j9(06-kK`mP|a)1q93&TxhkhmE6faV6|Ve zh4K{H4@qq#B;X8F|HqN^_d_KIjOs*w*S^{ParS?VNPjal@A>0O=gRdGPI*j=l9#2U zdZ4Ejqj@)!O-u0s+gB?q8&098^m=h@F|ENGH6$GNb_LOe9@pz`>4T`tT{raA9MVgi zBID3YVyG47eS+CJl#s=jnlL7jn8Od)dQ)#$frNPe?BjvuQ7b-dJwd_w<(1>c$miTn z(?;_SbFwOhTdEsNO->`<7%bozes{6a0Th;u@W1rM!Lv9|irEDEMq2}Xq}M77$->}G z;LU|({=l84V{!&j7R#=P#C__5CyUoZPQWXA9K1Mx?!New@$T;OIZW@$81XgO74*1Odc7y;1Jnq+?D z%z+h>No%qA$ZHh%s3cAv-0*!CZFPJ|F5fbbrAQRH!A)kjC(Cvd>wrc<ny*bxOX)1J+)nJw(qrB$vP$FNe^|`SwgE-H%enovssn& z3+;hs{C0V~sRY1w>n$czfYIJ7WXuDts0`>z#8v$G-_J*9YFbR9U-6JHryCvllP^X; z_=cYquO71D4oZJ^oOEJg_WQ1+20-Z;Hal3fm$U>131BOXiDQ=hk7i!Se&+n27Vy zCSdnxQLQ~jf7uK|jjXq1ZMbOE@s{EZbdH_u?6N(Zl5{uA3$*cY2Jjr_r~|T*S`ih(C*FH zragug$BI|F5dluJ_!<_(c^iK2QXAtETD;xr{BTUUoe>ONY6H z*4WLcyzUlqS)t5&4;43KOXU9zBp~U{C*nMdQdxi4!o<@Y>zCj zMff>uANF`^?V&OWeU$X1_=jWfI_N>6b?@MHCQ<4M{K<^H1!=G=KI;h_ts1)LPK8g{ z6>|CY$gN+H%bp}XX(onJp>`!OJwJQ+x5u*92GeO=jf@!#7`UTA8&8xI%j52F@Wl`f zMSXy#Cdwu&po9fg%m}#hT0~vnsT5Eoc^o9)_wTT1NxXk+mh;vfmj8PC?kF2FUUeucEhAY%k54?QFO3KOcezm^{ zPkaP%Fsd%MvAt?qimVKC8cLU1iPKh#oNyTJ=W73vp4%o^>wUT#pTL{amlJhaPa^&( z6a3uwrCyn7NtEwxf=9$qp*zjhJy!vipImwN0gj2E#E7A)aJ8q~5uRHfW*nJX%{pJJ z!Xy<7lfzrSy>2^Q&EwvR#nVX4QajWNo#ZlgvRSJ?VQlIt5wfx8M?2`4p&Yyf*yaQU zMOO3l^PF#_jix`FRP;^A+`~Bd+qpZ{oZ+s<*ny&+U0LNOO3XY6`{Suz+vycWU3_p@ z!sZxLLdBahuvUnn1~spb=;C5Y1uVvONcC%C!N9QLA6%I#UKTZ`<21_Y!(oBDcAOiq zm+_A@RhXRSWXiRvLM*pyNIZ z9`b=|p|?Pb!owrvyf zSe!l)nwVhAM|IZA-J?NTP%({?+i&EN>oP@5a%Q9bOM*CIqN*8M{zy^R^;&(dO>TJ0 zQoC~zd-PnlcM-d6qS8NIjUtO9b>;#s$Ys-)sLZQu~;!SQ6qqC=r? zxIn?tt=d*Cekt*8gacQKzg)1u2m(7a=m$_c*%^ely}zz?J(#fCZISoNwmj~|(?rK! zpTN;8v%MmlY5mWgo22LbT3@o~EC4JAJ%MZV2Z|UzquT(r%EO0>7F;^X@8j`?p2Tza zR$Gony{7n1{fd7=W zf>x^@(AoozefEjsFkuHW&A_leTJ1tzt3Ag_;=-ZZ1=U{JhAr9Y%`69wY-wUe`w_D( zmL6f;){FBCNNLwM^ z#K(2gFfdZ((G#FJ*Tp|k>Fju5*NB}*A z8j-XzoeBB~)U^NIF!BkN}dNwE&xvYXCN9|)7 zf6^6!?h!%m3iHX$7ZzckE zxJxz|k44^?8NQMd5`FaLqR=sxW=Q-h&RjFVSUTyrhbz_Hbgten<6B;22Xw3VyFY;9 zGAIQEabLHthc&QY=0r)wBXz~v--Ry8;}zh&48G^%JX<4Q2fTTW+)U;g3f0QhkRB=_ z*I*LCQkOu@*4y3Bq$+$x^ChG5sB~bMZz_1Reu`^_x&tsU2K1@G6 zKg?iHEbNq-W~GjjX?4LC+c=aAMbPubbDJ!0ld9=A7|$6e=SL2O3}XfXTL7EDtv8k% zQBv0$GC|P*_MWh-~jvqkoi30>YD7%)OHnK65w(&?mktYUYPuf80w}V~NNAB=f*Y(AOxq zin>q8aPwZDRSBb{P!V?kHh%!oU+uQ3zvriGTc3oFEFgdTp*w@*78iodw{tZSKmFJ8J@Rip)m{T$ zGc}%JI(p#7rO2~NR5ak71zgN>2%C{Eu&H0Q&G5^@gL&%s(V+OYDk^(dfHmH>8NT^< zd9_3ZoZ-9Ku;X2C<}LR8)14W1T2l}IUB4Vugnm}~4;h9y&+(zd^_DH}BVkA;($zPj zsj1Y=Wbw6oTDx5ihp9DH3exwqwbU3UeXVuQmT(OcUljnvB^q3K8dgOGJ$sH~>SEm1 z<0o#3Bv~#iDQ}eKAM7L&Q$N}u<$C&PB{oWOJ{R+}5EFLy@y56(9jU^(c~sF?*0Wb> zUtAA}LpXwbgFVYlbbrMl1urI$XY42sm_a4Pcxr%*cFFq^NS2TUA-E`B8M~mvi&*O_tUSd86yV46oE6zci2sOv$ zi+E!F_%U?Vie@uZ0Ul#ktD9KNJ{}{oq&YGx2UmTA)+&C(!FaPau&bqCp4;6iY&kDj z5m5DjW_uN|{?3FJ@0d4S;!qBTd+xSAnAYo*^%=rF3PM7Upr26rbCF3W6LaNRQOkOv zC)*us1of~pM&%bZ%e2q@aKX!uWjSvO99lwLG-LkdA79?b*+82ZmVdaPLuIp6OS`D} z1){N+_OSAeYS?XOI*Kd^zs?oecC& z0%3{Y1_snF(=q^qdln1QecGw!XS~)VsPMbUU3}7o=tN98i!i4G{_nVxN=!;V9c|6bxUu?YL_btPR91g?OEis)9ytqB zDv@k$va|T`tnHN8{pZ1g1$(XeuW1o=l@;*zLDHKX+$XPQNR0bMF!soq`}vYe4=GYE zaAlmwpZi{rjW*k$o_+kd-A1q*qeme;1%3>-h&X|+l^u^3?)%Ofz~<+2+feQ~hWq1D zUAGp2v!!`7*(@;g7!q8fSLdr#IZ^9}%Vu%5uOzq)6s<}SjW=9T;+L6XUQJo(P^z9^ zFuS3YLPj&P37K0e4$@hS?ay&uC3`H`U3#PKd4SBmZ&v5(*>-bYwFa74idM1d#g`s^ zvHm_moNj4H1zKJmB$#SE@idX@!{C*ji#`iM}hcA{wZ%I>3 zbbq9rE7Rb+IN2@6%W#qary1(;GBvJ4C$Yhub*fjainoxNCv0@qIAQ+yM`*f5@Nksf zrb919B(mG}@Rzf}p4DB@3|M<=(Y)=J4w5V6m)nr~XgirDzWNbB({!%4yg>5GB~5Jp z+QhdWKm(W|4$~e^H!=F&aP^`C0HTzVcpL8auukw}CzHEV(PJ#OU}J`}sYB2PK~n7G ztwu7kh=0kacl5R(!yZwa0RmV-yJsiqyg?qqel4cF0mdv>84preO+~4)^rO`LWzsWS z$g;e!UGSdrm?}Wg)LkDgB2GNyCd>Bo7UOwM0LF!OXY1bK)6#a~ z227DsU)7M>@NJ}S-@3!$D++Ei$Ej9X^kuuq&dau+BNYzEg18y4EIO-=&lsPH5{(9? zCNXsUcyMYbZBeOTYCv`FfDySUM=nPV<>C7=*1n{_fCQl5%7;B-9W5JewUufMx&PjUI|!NGw^TQmJ&Pa`GsYtu2DU<5n_g#J2iCr1q~7yX z?`1cZLaq_=x}t?5?Iln(+Q9W=?`+biq8k_H>Z3d9pmhl*USIS(LjacIwP zbFbC=HqKFfkw9GB@G4I^iez8|JVS7P~9U zwhE2-0MLj_qiIVu^&emcMzQy2nWXIs?ENkzO4*%cB*K;SGl*&;sV{}vv>^0+8JEc( zy^4Tlh)caA{56T+{#>Db#o&kDCs65)FJts9F0-xPQNjt_`f9!+0w7`;-(pZs68n;r zHsEz-A5MF((Dj_IZ*7YBQvh%F1eXhcjJO@AOeGvYmQ{3%(p{-70B>c=Ovj{9Ek=Mk zPWm*OoxzkCP+I;-TN7IZktTeM*NSbA@ou=Dg?xHoWM@7ESH_sf;78TelD z#w+XbTvb2>_Z#XxsXdiDX>C{;6d1z+MZm19v-R{Imx~Zab~LH;KUn}y6{l8t`lYdD zoILqL^s0Li!MEaz^Z7?Tznf7M`N4T-Lles2+a?XYTV1}rHds~nMKeX`q*uu z0(Id*Io}k(uc~R=jwWk5nw0YwAF*u)&6~^@k}TN{Qy*7D0-SgF%8`}H^Pa?~nm6Wb z4~`=ZSqPvNYN^o{cjn54Gzp|P!WcgkQV$RPAQxQ$=dfIMvE8nJt(WXFwdv%CfaT-W+vWG|Fb>(V|LHYoCvb`UX}6`Gt< zs>xW`u9ont|DxpDI9~=&7JcjmIb|dSI-gjh&#HpGdb#4g;rPv-QHGVLVmTpqXq#s#hAjRQfzY!2!tu zwAisn=+V6y<;80k>hg0rL?Zu-{$9lO9!vYZyyoD^ps}c1iSd=nL`;6oMb{A0$!?ZvsVij7qFtG<~5%+1c$kN4KOVRiB zyw@&h%Tzzm(N_N&;wi5`7OX2XuBf8s9M>19Uv6eT8nj80XJ$CAn8Lw7nxLQetdNZF zZ!k7UyHl^9<`LxGl~wtOWZ7MS;*Jld*3qEFN{8+0!s8%tUu6EhT|L}104>)Sy$Q&7 zXt7-y;`-Ak<5E4+*nmZ(Sp*|((uW&T?*$`^D1DZb>T6_=Pgf%u?a%t&-N3FPwkcNy zoC-p!GuKNJ)yqy8k%vA&}(nEW22$Uoty27ny31@BR1pafcE}f3GdPJW@z0c z0Je~4Pt67UmLD1A*56XS*!qm)&D6y?$z65Kd7>GkM9?xe?_(zswAhC-3z@`hr8rse&u7Ll5P3su z=A`aj2vjKZ^GyC?Z_TZvTB2KE@!<0LrQ}=-M_F6i0OP6D&D+!*0->#q7YgV+dm*1= zK3-p-(Q(T0sN{uJtfkG=u7rJ*`_|&LKD9t(`o0t$X}B>&-MzG#8lU6-sKv8$>y5n; zx^^psXQnOyNP&pgC(tX*ShSOBYW?`dx~Tyh zDx#4@=q#0o<_e#5;+myGQs|jydSRccc0I?o;$ z8f9hY)tVM@a~JR$eHSOQ{$o)qOl6dY$7rpx5!>62eJ7vQD2c;e5g_wEkKX`U6VEgR zcjmblgyF||WY?+vsb=C(a-yK$4>FiIPHU&Kes0z~79U=?ctU=8hRae^UH0(S{-n9fe`;L7lA8*E|?{5!c_B+H?#Wijcc-&P}n#D7LxCBARTiZ@5vXUa8 z4$0UUQ$nfsk5c0i7FVwraR5_qz}lHU1{Ya-Yd=l2U%V_@#S!V`H3xyE&UdAOB0L}>!(d)yn+9_&)&dwep?PLv!rB4^j z$554CtqRKe+K}=`#*b0?vt*;g(EDJ@s0SibQ~NSVd{+fSfpCokJy`8v(b&i?UGtj# zl=jAQ8?!r^zoi-xe5^}7Wh+i|dT9``~$_S$%RoPtKgQ_pR%*!zX4S zQH=_OJ2>pww9syI#>i}z99#?_Uf9g!voS(?6*pTyxA{|vXhtaL`6K5X@|TkG-ejD6 z9V_0{Qz)}5ec@ugirBivxl?zkp$+w^W?Sj9Y@tRMS4k=Hr$`nzYaZ zEYeSsdgs;&Et#*Lg?CV$kc7R7Icfj-*d74In1Gu%=G(yLL{9r8|KqG?xaa z@B6yBO9b!Qb|5k^Cj}Cib}L=m1eaNnGwUstWjzhJ{n7BGWq1GxIUR`1r^r;BafQA+*~=k@qQhwA{6tK;lZk!c{7#Hnh&OM&9N+YsQyC{wz@3 zYE7PxPm>QgAJC}=ClnFT6L90(U@i=nqdI9jyiSM`-S=?V+p7VZK~|HbuOh|Bmnp|i zj{O{xaww#BKo5vzVK+Xa^)`=6J-c#TXVFD1a;mQXJ3{&Fwh@TsQ}%)xLrzX z=JmXtbjLC}949E$yh0|BNK?Bqq0Hc(1*e_Mom^XdMwcna0Q|)t4~zC`)qi%7XP~;a zo%_Rv17DCvvQYY)kTRh{O_n2PJ2%g;E%?qOZHNDun+*KOv^x#jj|{~pHC>d#Z*%em zT<$-0Y2LgBs*ko{X*gn}B8U&EIiym znJ$s;2l=YB$eiK2B}I4U*6&jcSygI)qr?{PzgRmbJdH3*k`rq0!?qHWXb(@zhiqCP z_PtY-=mGheTge zP1NR2!ZAs>?YK=!w(j3NgOEuEaycZoF1)cl!?J7GPj3@s&)J@Nlfv;m58}fxu6FNgic??MA}zCg}G(^SXz&yB~bbKyphY- zm58&=G(%!a=+b43KPr>=k(Rix;GJGUPL0;~KO}+}Wn`NfxP)%9FB{+wzy>Pflr%gw zF5Al@xE7zPd4HZh-9F}}+q71%dppv-Ja-XvA#|+stgW8vk>yArBf*&?O(Lt`jMqss zqa=S@7LevRdok33nGkd42GeF4p)=XklBv!<7D zDH{04FLILkoLbq?I(Lvq<^3_t7H5OmXvzgnaZS*X1CejPuX<0B1bbMn}d;7Wl=kqrGgOYJgRJqZujV$y69gLKc3yDRS}c@g<6d? zjQ2IEO_eydSpV(^4sPVy4B(7;C$V>YK#%Ot6{Qn zYL%e-*s$DO?58f38ZJ;1eO-q_q#Hu`r1E7up;8O(RM(qfYQW?|i2jUVJ86CN!$(e( zH60R!=5qk*9haogS4r+9P@o$GJ*{gqU1|bjy`73RPuK%x#panIK(V}JIk4A%>sPdR>{Lgj}NfD{^!`<+*Ws3%$SyIQ7%Mt-O8 zRSn*H;uAqpViP5UPYcpBuTJ1EY(w}J_ac0PP0k1boHBu1ULp$P+?<25>5<~R?ZPDf z%%D4%t47+S)?psfHm5rryAY^jT&S;gG@BzA z)oUNPv-i)C@Lu8?xZIA#OfwRR3^UhtID~ZY-mM=T#>pI;LN`7p&^O>nr=fMDzb|?E; zmy*7MM zeGQC9Dsp>_bx;NRu;3WTG6c3a17uTs-qFK(!n~#Y%_<1=t5QMI4P!||@5TH2`;{<` z*Xb^}gRnrkQGK}*1rLh~z(r!MF8k_2?m(7pb=Zk(?)UzN@_y=N)CaXyk0CX6D3|SZ zB12D%=f{6M7F_yoj|3rSlr|!cvKEY`6<@u|Tfj&~ioTjA9km%Uq->)NG!bI`>~G4m zBe)+Ns;a_p|8%Z>F|#c+V}6Y{1F$Pkok_>aad}GZLqH@pFaq0Up9hk!8N52J0!qg! z)}>Fyz`A_B72;Z##SKZrr6B6wU%LpQ$HfL9F{ik1m1_!c)a4Suk%&eS?I-e^O0Oa(|rBm z^UOcc%onGb6SX`L=+dDQ#KHhNn?qB|v5`wR+s%PmAndAwo*|(={H&t5%#<4qs96S< zOyOom!5OLJ)nD9n2+%&9Kk7cKeRSr%6Dp@zy_{}Ow847}z2kMK;I@+p0;Vc15V|ob z1Vd3vxGa4dwv|)C;B9t;kDQ&Pc(Ejhv!Xr?W}dNcwwQ|IM4V77AmMAOmvaHmgh$~L6`dw^^bL19 zl4>istF~2kM(a@xNEG9nLt0+4XO25vMw|%uCf9DXeL=}o0KPUPfD=8y#$1sTw8%gE z5bNJzJ6^qW{O%z$P?|1NAM34$gbOhth?5=UD?1{KfU7z35M;RlOYr27p)g=k{*(WS@lC;?ljF^(~Jhz6@ zWtAIGJ01ld*zp{AaDnmyR+bH{Q+!8cqsHP z_?wk^JMF_W^5SjMu}sLrOL+;qSxGQVa{e+}@kI?>$#@TV7*$Wo3e86+87@nJXAe&O z+>)%MY5H&f^RemqL;WmmE@)nHNQV#_70|Kv_OKt4zcPA28FRLnC_$Bfkm0HI6RmT7 zwibe9`zCcbj-Y}GVY>Zq6UeDv_F|zOZ2u)TBK;7sw^m%AyFbT!gkJBCh~9o;g4&qj z$ExA(au_7sjG;Xqr_w955*&`iyEKNrjYrJ2#0v|8Po6#Ya<#ac&(fyQ6i_x z%!|6d-seiqwL#%u_?``O<*liTY@EjAWW*NWlTD%Nr}z_N)z$c&w{r0wzbOsB#FFUW z>*rh?$VCm#h9hBO^>ZBP)|;$1uAWNClE7@?1X3E1=)uYhO*PR)To}veLV=RdXJu0F z6r~y)ZOE>e;^wa-t;dr^H#s)1<^0O3%m+)f8eF8ihs%!X`?|nV(O)_EW2OpV8$Mvz zJFKK&jsMM)45wM1EmOexZ@~vOMvZBTU0_#(&EGxy zY_mZHJr%Isesib(U##zaYq!~;W~NQqkFr8)8eqxdCwAL}5|?`=jyp9`A{6ac0$BAz zQ3;M!U&4^P??*3IL*mWf+Q5UYaKnQ+Fw)VsFX1rxys_wg3Xl*}%ZPIjJN(A$KUIO1FO48rKdHJC4# zEK=vr;pq4B$zLw0dvEv3f7h8lnyq(+-R0F(^Y$K^1HVoj+(b#&xIFtGx>5XPzniQ2 z-vXaRI&iZyY0jHO*qYb*WzLgTrJ^&Tuo2+0CtZbbDMzSXg50 z73NeKF3g+jUV_2qEdaA42gPWj=VwLpTI&fZ7mPPMUJfVyMOv3%vm?_nolZiD{K>rQ z8Q^l^w_CX~c>Hm}hc*pm@|TcR=t{bX5lXE6I}HF2WaEg*hwzLPM3cPg~HlwZLh;e9?U&aZ~2ZpYaS&y(J&&L=MeuoYp0F!!eQ&~Q{Z)SLB>~lN1el3%z6f9<*RaE z$>Qr2@v|QiqtH3o*&VC`t*>%1=NJ=$E>FN_bl11>XJ3xfYGgRi7Si4Cz%G7{U^-Wz zT7x&y8R25sl#Et>|EgA&&x<0xV=*&7DR2YZ2DJWIRkCGss=>Q(+?*i_SP4BQT&Ng-3FeSQ=^raorL8{Q~zB9npR zOCl9435rqfb*R4YY=N-WnFjq7=roX;hxoFcv(CTYpoLk*u?mp(-h2{F$L>>RAbV+l zihD_KHCH%8XNH6+Bv>fwAwRC;IFSDpkl;9u;W>MyhMogwil4Sy@zzW?K$6+dgmBbI zhp{>1jTgfqENZ9+aadvst87+VE`3l}7!5zIN&tuK9+0Hk%-j9z8<2e{>2(Ul3Cp3& z9UrbR|HYNl8ZP^H*eJ&krHt(H&b-oZ{VybB_*Q#Jta>lEWBf~m3KND13be4p=4w~` zF{51WHI_^_(L-y4#t&&t3|`Xkd#I)=;*$UvnC34aTZpVO1iU45k|HH~-~ET|ZE8mA zisE7}#7@r_tX?d!gnMF`+7PSuiRO%;x2YuaU%Ji9Oy$l6#Barh4*XXEEj zew7{Erd7GJ7e7Su>3%kPl9o4)A_!%{C>rRgz|E*lt?yGRlGt&mUWWUOslDx1pY<&;?wt5FNv&ikCd8_tOfrp%iKow+-uAnb(3vhws z!o`#qS&=y4pvj0!F_tfh-o(W{CLqsUx3)BoHwt$hYTb;zW)G3#cwT(N#k0pZg$#_ zmMNHBt)l7^?6Pnn)5q4|agm`l?iL-(TxXFS&c+_!^evI+luN5Y%Tb9@QXbqvJs;Sg zJ510LCI$9-*>%TBH}iCHpACk;_bBWOV=ma4RcL2>!_(D?BNcuVUaTX}t9QNSh!rW` zqr-|74_|apU#+lqyz1!{nG*GVXh|5B)l@8RS3eHz@gY9O>g214p_X*pECO5QUac-Q znjK!Fn9y$bvx^gEo~`lQJ@MQHJ{J})aA6yS$EK8dVI3`2%SjWRIIhI()$*wdR>mR+ zEWSv&Pq6YUrekvS*rO!?ZLozQRF}Q3gv;^#aD+F>2dTQtM1I&oX!{)+%Q+TU1g$=j zvKdIq?tZz!Uu@Df`yokCzv`<#F^%-JrIG(#rvXOXz)Jq4+&`Ke>U zi8i#{pnya^eM9x$CcMcmVb1LBNk^XPR-9Y9bH)xmx~qM~@%zCmEr$mHoFi^-9cuT2 zf@hAiicON~(~-`E$4<=>Cexx;<{3Xi2?&)~3j@Ty;Uat9nvag4?bS!81DumYVq^2l z8+1&4Jp)WbP0M9;C4RUURAbZKyXNv@Aa;?4*7Bq6IkF6&i2?UrN_Q2DpK^NN4b0bH zfTZ8HFP&U$29P2@UY6Wq9sVu4^gCC5@XD=2`36cuvE9jbqV|~Zt-vNJ>pjp+Gnh4i z6@SLSgd&(RMWisa^-gCPG06D-A%)ba5;k_H{^cHTHu!Az8bc%$uw>I-$oi?>V)reHM3XB$$#`8gN!9xLm^jQ#{9Sc2>zT0YsWQL!_X*`c>_es?QJiPOfRcBB zk@;3mqh&I)XE4Ntgg03h;An`J5f;!`tn-2P@E@~NwyR;cz%8hqpVixaR)kP$;CB0` zZuLDz<@uAPxBiX6bg{lC&ORw!;N;{@52i4JR3K!CrgI34=&GWE%~+|DNufW zx%h#ca5{76`~9c|Ky~k879YR#x9^tBM4p76MAldeD#=kc?+OLsVP*@Bh6C9Y4C#l= zQ&;!2?Pv}Vq*9ZAN^D6JZqok5TjgHK0)1G&>lXF+aNAWbsO$lBem3mhyQ~7C%PPDzZI=f>*xY*wyIktNcbACD6E0T@5ObX}`1$Yw?{fXY$^)L)H1j&%*>PkH zD|Ge3k4q!i1H1!#wSzH1W}Zx5n!_XV7%n>b>8uGY4z>Zai=6;Gq!ZhvD$c`~|9D2z z;Ue16I}eZWGq77GJcTsoVwldiIVMiNmSuotiavCa#+Nxx?7^FL0tB<;ZpiG}JVCmr z)$p(_gOK!Ta5Cmq^T`JNVkA;RMNw8t0^ts9zlUP#v|#TNwYl0o10m7&>&$Kn+y8GZ+pv5Pq1+Y{4NPtfiH7f<1#$y`DwxOt}epTW}OyOuAM~Dw2UcsEZG*F zEFK#U&DigN?80GEzq5tIHpVpdn33gJgi4YGGvN6>1L%O_3s zBBG0K$o;ZN=W(yRK7A$l{@f{DJDy1NjY?yNkh3Pf&>G|gOsuU!(jOkWCf#7M>qAnk zZob>J?YQlMJ9kXvwks9anOwb=zc_DQpFZW>p<~iXX)(WMzH9oEF)6RSw6r7m=^Ed5 z9R}iw>xsV^i1`LQv_NVn`?a<&&{Dv6D5S1-R$d^0j@W>^x}Mj}>1(gK+8tVQx5TKa zxo~W|8s%BMV7@U|ZR%8e=YEs>XJFmVBeB`E#07B6;vq>r^7OeQ5;V#-I~QI@pFc0> zL*;BdRMG!rF?V_Dt2H3!qD9wu$I|K5Ec( zcAjI>L2M}iI(O;^U;NqJ70pw7(O2d4qeafm3!A3g0HNn z5>HIB>P#>a*wyvcst3HOa&dFKHZhW7*QTz6R(ZY-l?*=8R|ULV2j^sj7tHDZFtN5k z$ldH=v!vq|&Dfq3*kk)0rSR^vHD(lfTEf7zb#UkBW57`i;v9FsjS=~);(ouw#Q*z3*b;Tp^l};RtHT2ps7-SfFU7_>b4x^ zMbXsSdskO#wHhLaamE|zx=*Ktax%sOb7eYvbAQL*YX-4BevqcIbDaI)xjC8C{X5xo zLd*s@C46(#v_2gsLR4nDWSJ9AhoAkc_i1ml3%v-FURr#3rbctSqxbg&*br<$ zsK3(c9?f4Iy8d*UsU%xJV+RvtwdEWR5tn$PnwAj#smJ>#BelNiiRyXqi~5h~ zoRG>^KZHjQLs-l!T7Z_h80{&akkeh*><-`aq-t&hGg7z7(p`dO_T=zPr zd9n&@CZ*?BU{N}u{cKBKq*M?fsl~M$dEI(gj~wlP=ye51kxx2Kr8K%3e5G*T=sJbo z=P=rgsftS{;^iE9tfxZjB`Oll){fh%vu3tje{~cV0_3D>1Aiao?#BxCZ4Q|po-s+; z7i%X7XgcnkNucnY=;wW<+I_J0bJ&Ll97H6`1yRE1m^^1vC2alg_~cHdo=Qv$vmpe~ z`Zi;DoSP@J;GOig^nH@{kEsqK!#u3_pCb2b;DK00$k98YaL-tMjgV`oWBsz4O(GFB zlX%W=bF$oGabGE;=7L96V0rI7Ax?{RChCpb0&uc%cbgHlxr+$)%PQJvRX-poC>_-` zg>B;6+#vUYk*3DBE}b!w<(7XHbs6L9nlci4CaE^?bW4Qf|KjZ{qvF`MZi565njpbF z1cyN5gy8NT+(U5pV8PvkTN2#eJ-EBOyEHWRid;G8zIX5a&iCgH#^}*>QFPU=wb$Bv ztvTmHz@8gOyb%%7{8B%EAuJN+LFyByr<>X-c zw^5)f{#K<8w7|aFGj4T&Y{%=0HnQ16C&j)BaoG0q5One|8(7IJMzq^A?qtSp19jtB zdRHG_?N>ZvH5=H}6eavFZavOUF^)|H;kshbY;k>l^aUz+LEQ}4eZF`PQ^m`XZ64&c zmbS9}D6YUpllkb@Zn+u77{Zs->vep#Oe3lw9D4OMmEsR4T50VDH;*+gIcRRt@3Z&_oBF7Zm}B4i)ogvNI0JFWy#c8KQLJ-ILa6-Lbpi886&Z zH1K|Dc3YdRH32IwIH*2NC91zor5HK^qU+4mvJR7_!SHtg-odDOJ*XNJc{T&y`yK2; zHQ&;xmA@Q{G08|*FLZTM24sxgW3XA zXYjnPZU%ABGB5d}FZj{Uv1%ln9<(&AY6Vyw&L0h|@If}d7EG(b9j$E8I&!{G}^ zcoP7O?wd79-!#*^{N`Y55_4Ncbi`@yZUW=c0&ep)JKOy}$nt66oW<%v-=k-)%@@AQ zL$wxILRvG4Spm%1oV^z{J8m5|HP;@4$$epL`QoRN!N0TfpGL^<;ZY{Rp~>74Xg|1{ z+ojOR5X9 zY7&c1_u=)_XigE}<8==-$P>9+Qc^4Xn1)dcVrvGrmh@o!6i&b?L@3NyG10Un_%%$y z2=c~**iGJHn`&%A#S>gD5;D&V=A+mP)INOKM|yPIobL`40^ho}c`K38| z$fKvGE&#`<&aTH%L2FUkbBNOu!`@ifOi`dhgF;$$iqMu5UGira0Xf? z#o7^JA2&65hOpr7Oo0Dl)NNi-Wdm;da=nuvnh|izivR8tL#+LfI0x#Z}E)pj5IBgM9SCH&AJf>*`72cK{LP;KduTS6YBt-AXfXb*_* zz#Dc24GvmE0}m2&Sik39yga=cD>TbSS%Qt5)oO6uua-uHy5Rjen7l{{QYt9C7f7wt zT~&~$PBrBp!4Ac}GVJu|0YG!?u=nV--hAkL)fV&BAzUn~BkWoCeL$LpE|Gh&?wrV5 z+9)**>g&GX-?pp{Cd`oZmWO)9w4G7$5&mDv)Dc4=70Zd+3j&%zH6!U9A}Q1SZuI6l zMp~goj}=G{lr+m$ooc>Ecq|1N*_mDr zH%3tgcs1O(w5!eL*(*P|;yxJXGc)wT-j#R?DjyC3QZ@yBM+z=S{=KfBLvM3`wD6F2 zI3wSes1y-nQN~UTWgXFJLXbGuL&purw}uTjlg&3)ynPj#T$O(?;$DSYU7Yr7DN8>< z_8GN6JhL?ov9y{Ra_pmU`OleItWT_G0pdV$#P)zhi{L`1)x`{sJuya$8z9s0#N&Cq zJCDE$0HArL&8h}b)5guDRs-r8O$}9+^mx~XjOz{)CY9`?jw@{{5nS#);jiCL=BwCN zI=tqC)czF4t8a2W@pjuN7P;K49G7`@fzLeE(G>c`ysaLuaHD3cv28FQW~?oM0Og)t zzIMbOVn(3%cj{pL}BfQ9IaDgc`P4JlxHL(%m%KdS;d4xhL_=w=YDMOcmNcX1PKGeZ3nSs5TCAcOu?M#SqtyXVP zikUOMwr`xBWWsC!1is5-RqfY|iJeSu2U@$N^x^ zV#4yeMN-Qq4|_o>oq67FtXDP>gWi(={-WY1Ii*p>w|A_2epoV*Kt*cz!q?jF%x`oI zvKpRs1i=3-H1mQIx4M5G%$zp8Q@J8tU`J{XC-%A%mu)sUYXoCO^sj5VnOlH`F!}V$ zo8;GTQ)<`oskquu-fDP2vj%On0R_W5F+;MAhT~=&hp{hP&uc=ViKVt?@)f!Ys?bTC zjvoD%MgX#&188qRmT$2Tr!tg~jUsCRVC$`e{KS8PM%mx%m2*VoCjk4fmsEE0o)Ynm zo$h19N}J0FvJdwd^h8YYfh+h|0P?s@s=8e^`wn*5@t)mc?_5Dd+BrZzMmvRS(e+|d zq$iAMZzIUL1E*V-pEZ%`2;g9*eFU@m9&_4l#rAMljmPw9Z)L@)=jbfDhB>4f9T;E) zEn zH3AlG1|fM=^U2zeB3}*EsHLMT@Xlo>?@1n@RcicP@%QnB^803K1b8*rWot&(--$mx z4x>xHXzs()2D@o(mdo7Ta222sIg!JQ6E9&{H~p|&-W=vgx9?4rNXt@~(xAao;&WLn~}=|^g} znl-NmJ(t_v=d#{^hawN{}!jOcq$Qg^!Pu&G6s1u-%&Gdo`f zHu}rPpI-*jS+y?xBOF6XUbGP-JB4!HvDAmTR+U zJ{g2&YSr$_i#jdlNyklY6L{rDR#ESX|}) zgS$Z>{##(-lU+{@NDSqmFSg`B>MiQdvf^>`C41__E8lqlMUc>|_?`1A1Ex*&*0Jb< z!5ZNtn*RDLak|ywchvw!_4;;VqvmuWs$o$y$}Grd0Iv~eWGqVBWBLF<&%?vN>W`gK}wQlsmE{4Hn9_f{nl&-#$rDWC4YNOrxyvC5*+q07O z+Dl9-q9CN^RjsVwx&^OzPqEx}R1b}t&XKr}rVCR_o~u;zwm6l((QFP*i!I7@ObjDn zmctJNxqCkN-J`YJm?K8eKNGlFfIt6o8s%>vS)Z5q5|T9O`S*ka)WB zn3-Z4-8D0t&=brJVVMy&!6wG_ zi1lqqULXZ8jn7oj`QQ?7pteboS!m6i0D>6g1kEm*Rpe@&*R38-{Q%YhE;|j0 zo5LnJl&hTcgBnZ+I-OnuDP%3^EyAq))FehPULP0sFH*EhmmSZhU&BDhtEWQV*M>ZX zr;@(kX6)px*b$Ge%y*V$RXUH3nn%5X_dD$3o16z}B3{2#c4DSf4*NTbh%XKdW{nq* z0ZKv9_T{|f$qPHRYgPayan^S+O^&Uvcc+igt((qTy>)77uA z&)%t8U;@ZRBp&Nez0`Y#SfOo9GWAlaWM9V3n2u;Y)%|l2nlZi#{RR^|_vZA=yE>@B*W7aMGg(;;!oGo4?LGXemIL(XhrhoRVb3a~bJhH1z5~{Z=yRN4e%YDvTrMjXV@F6BqOOQQrYA zgwgtL>$t%YG28%%lY6+2=sP?&Qe<%Y4I1hhBu5-gy9BQam?bG4V=ZySow#6B_4Z zl{e(!CFRVA^cVNxN^cNcOG{teSFkE^^4}h?8XOT*ye=+OLseAPzEd+w;EUF|?(pns zx7+|gLH9D-tQThI8=TCAm$Au$+-JF;5+)X>H6lYH10Hcojo!HI?v>aBS)O+X<4$XD zO#qfB`|)grD0sVUb90dwmcMfW+j{3HGi?gZ`=6Cbl9* z{=UsAS|J4}27BkQ{=FRa_o5U9w5Lxa1*fcu*h(o`>zS<@rAZ%owyd~3w~sG2Y89Yt z24KOmWRqR(ZWdw#f`~drp5eb>ptJTaq5R0J++@sK6wxqXvkm%_l>%Boh7QmGubajn z&;h~};}ZNw3GDZGij%z;(2jmTmPGFJtiK;4YCt?_qY$(u9zmU^!M3CzCSiWdve-l zW9Xj$35Un`{qBS`vdI9t&1rD%#YO6Q`^b?#>D}I8TLAJ#lbo-aN^7^~E-m&38vYQU6m~4hoZl1rg9w>9h|xaXU`; zB92a7p4s(iKjnOLA*1g)M)NN>_zyigYgp8uiYKL#iErI7bs_69%yiKYOCm<;%j zguT8kf6S?Ws+9eQH|2Q1$W^HOFxURebo<++`wy>Cz=IFpXw&)YXUv8IHa$z%hWqi~ z-PNDQffT?{P@p}K{kwyIJjy>E z{bdZ;&X6%;_sSpnihm|9{=*v~OkgOY;l2LF1OA7<{o5lpWCd);qypB!`p=W&zm50* zzo~<&+2&;U?&hzX!iqJu_0aAQYWiT&`%Cb$nYbOjNI)nBzlixeng7MpMI>McCPx}7 z3g2Hhb!-xAF(-Eu?`u|s|HWKSd7A4yP?!60dC79XZu-q`d^(O<(yA$%LSU{ z`mbJ?bq=7?cq=vx70Ult2f)zo$&Ain4Nv~68Lfy>TZ=tk=Jtf_ifs>H$NX0-{lgu# zVu-J#jLXVuG}2tlyq4e+`peH1J-$_s`dY@=ul+*~=U*Lf67K2T|J6~6e6I!!O8}*?-CrI0_t5H34p5y%9rw*&zs6F)Q=m!)Z~gaa`{$$OPtLr; zu$)QmuV0zoQ^bsaIrzVDF5*vC(QBxe@#pXU@9*26To@hAH}XHZ!hh=@|M}ss=qE=R z-P`{DuV0xTCE$+L6yAycW%G%od|HA~#`t^xbr+8CB#B#^V;cTf=c*WqBn94_J(z_$a|&H_P~^XcRAn4Hx{^uZ#|`n2zYI|Hagydy1?SRQ>6J=>A%c3TK0-rxVbK|CELLB_bYSRO7ne=IrcS#IfX_PXj@hJTFq zHG?mgL%@59i6FYNlN}yKAWBSK7tSm2kU(Ct(d!bN61!h%9dGv560pObForQQzomw& zgNQq41kcUh)R^{T!PZpm|1rFF%B+;Y+-}_R>h#LGwJ7@>2{M`l@xoOKK(8WfaB%%d z@%+HYNegzdU0|+jSI^2z1eOPcSPWhrEqB%6$b$>V00iz9S>9b(u*t(8g315MzP;&$ zSCX~T`#}x-!wP!(x?&`(2pjxm85_fjOstJuS$BYei1*%iFuFx=D69>oxO_V5B+Nwq zpXYHTG_==UlVtVbo!Zq;^1(b{6G<7VN$9P;vLQJkzvB|F1S=i^0yb11_ORWz`OkAG z>IfE-tURyj+tHt7hp&DACu1QAL7%BR7QH!`2D>;pVT+@|t&d=~YOVOo)G5gfIdo^f z_@%c7Kq$o}#DF~rZ1m6{e*HhVe}Z^el8dFg{_YMI%g~pI6dy8v<*CYDry{g^9V!4s zhd^terya0V8LOF@neBwUXzonH1eX^_Cux=6j;kYS;%{G^-o{s0e z!w)*mW(oWfr81us*F74&BC5|vJ?WnDx0;o402M<3D3hw^D;LFL5{eck2!|x}7i%>* zsMBigv4~QuySQxa+ikRZ``U_^k=dSh`4fU!*C_!Jc4I%*LgHUm$?RTK#GV&HpO^0H zQarAb>iHOo_cj0u$o{^RTQPyq43@QS)*EImYOk)gaeC zuit3kS6e<2C)H7D1HCcWwgx~(oLx?HFN`8x-VU(J0?dz(czvH$*!{SS`{Rq!1b9}x zditNF-nUcis#9zrW@*OP%mm)b1=D?K{C5-qZ(PT>G-0g*<14U_Y|GFeiBX?i6# zC1rFQ^{Mzt1|Q~;AaY!WTllTECMnf%=%S(wr+RULy@Qma0o&*4wkoS)(jFqW&xJCE z$mh5v1}R&)wwrxCk;P@iVn!cqbuU|j{mvpi~HDR3|FLWN&t4HRUlP3;;w)+9LSqJI+wi;iu zvIKJVlNa9NvjzE|pj2mBayK?CUd^n#?_Z1 zxBZ|aanG$3zHGandiDTM{MMG?^r0IFGX}{EltLQvN>FJ)p5)D4a4=V(IxL5{> zBpvztRR!2ojhJu65PT;uA{-8ish6oDh_m0`N&5I~wRg&?8)ZQ5G5i@;gMN1?0hvFt)=P}O3@%W6#xNSt*23T} z@a}FiK}USi)ybus@DNG`W8E;Iaqk@#<7K0U(pRmZq+Ed8MvX?gZ&JCDWG{dwFu6`kA|^TRl)WzIDkB71E7Sx9$k=`B~@pB-5+m|8VNbh_#n6{N^ zK5|rHX*v?RQy?_k>901+Ki)1s8s6Cr8}9iyu7jf&*MGs-jzIBMo_s#%dZKKCG_I&yHv$O0Tn`yoo-<7Gk znT=P)6D>W@_MU!@ApFAt#QFB8V-{W@mESBtZ;qB!_CY`C%*Mokbg!q+qdM{!l!j-Y zpvFn~c>|g-8xKoOZn;JwTW34ukB87`!P8pO#OMUeL5qkYV!*CIMBjHe^P%*H_?bTC8J+Ap4ewTc6nD3TOy%;-x=7vTi6`!6@z=h(Q)?Uaz z_|_V{?#*;s*RJsbKIZh+?r*R`l$u6h+wCrcMWEoHG0Llf{NV!vaz)ms(el&*omwG= zmcpC?XCwWO0w|dE8AEM|@aZ1a%8Y5Prz3Kv=Qe%P#yVygs<0J-xq#7iY{7gSw2SAe z=WBS2P-RuS%YJjStz-YvTIj}?{wSz69?`~KTC0J@c}L9S6B0J*i=Amj%OXu$vN<5V zx`J>Q)ptccvJT58?-fU*qaGn!+_aT17vjq#ps_l)=fklc8^T$kEyMJ3v0<=D!&!w5eQ1Pr3co&8WGJ)VK-w7e z?>$^ADf!9dv8G{98YAY_V>tKg?oE+=zh%Jy)Tqn->3t@><_4jlXpPn;yKR}9lMb}W zvXyl1WNe@dS|MzAnN_!)DDH=40bjV?VE}^{$70kcg4>pI3xM!qQC5|5RjFvsSyQ5U>^&$mdNM{0KTih#BEN?{S|jz^_jFQNNq6L(mFgO z_g}hr7;FT3E3ALb$9ti!bMi(yFzu3o;JnT z@;yNFpf51WVAA7s-JVzZVy*oQ5{iR%)5>KaR&&9FTn-)R4;J{<{v6#H^* zX~XT_pH^jzZ@QWbK5Km>9v zVE(o;pD4LiVYA*~PaqNZmi}E#m&&>OS;7#Qg7voeTc2KV|sLr~d^{%!dS&FKcKc?3(jdQzo zQ?v3Vqp#`aw|0;p`(ZV~2U=Lwnw)*ruP7if+SBj)D=SfN-QKO@S6nZ<9mCk{O(gUw zTvpFpzPdT*CD5zdt6t7fdvEG@IhO}0o-lxm6)%Y#mh%{T@YnITv-}Qxa^r+O>dJGo z>U-FAweIgDmg8QjuKL_eJ}Y4>c{rr_^zCv|BjK_dg2mj< z{N^&dF&u!+It^Jb2$i&hGX+IEPqC;sJ+7;`mI!7mW;N_rJl%nYfs@7aJoCz@Fvy*6 zMEep?MWFgIC`t|y&$<=i9o(Q*TT6?DKA;?-@!hFu2MCf$#t)DsG^Xn*DVu?+0CYy^QD7N@p(h*S9r4b9=EArL!Rm7&chlL#0K?l z2k&V_3g@?@;@g9+w-$Yn*ZpmW^y+YF2G-Ll)L;+qp0QBj>#3fdP4^3z)!Eb22f;l& zcNM;2^!SZT`18PdtNyGCit7o7Y1g7Cd1ArFl!YVirsa$yyvnu{h2%xj?*BBvPHOseaqIg44_gUoW1aZY@t0tCjAVic%Qr}nv z8*i5a^3%r}Om--7SQ`z3)`<#sd=R6ZAIOEVzMeFCKKe!&@>F4+J=_W}I7}1fKYp#A zH}~n{+~YCc+ODN`>-0cAMrwI{Xs9LMKZIT-wrAO=q~q1Q44J)b05?*i67rEr@M`ef z9gNJcu%Eve$SyMOzd4epC~0@1yFXCd^KPDI6%VnJ3UrXzLHk*K zf6f@Yzp809NU^^HgOOF{pcw;#Ox8Dh+FmG5cIPDATVLRe9CasK-?62mx|E#l6%!IE zxL)EFIwH!7oS5dsq@ILby|)inpUp{pJ%d#OADWta5kL7LAifIW^=~U zN7r4-wXkVWo~N<%rkBph5HZ(X(5C`Mal0VEVxS-aCG`5>V;_Eyo?OSXPEhFb%EK-*a}>)F&Os%fW&s%QnL= zjf!+e@w+;BL7m4Lm679D{ z8=yNAH>rG8=u6{m2u3SDUpWsT_CyCx!%FvX{P=mMRG{dsW_61DW0Pb&jZm3>*Cb2r zhk!Q`upw5VqiW+xtR}I~L1$}P&xjg`oLAZc`S|G?Pl?UCLwEBTZNZC9AwZzdHv;wA z^tv5~y)zc71CTaSU)S$6w#3O7Gfk9=r!dYJ=vPyTVqj&`c=DEZSq*<>4ABCq1B~Qd zgSiTm8h!7$u-9xV%#+MpmN>(AA&fFX9RX;_p_d=&={r_m^QjhpL@b8sg8-?7PJrg| zr6j+3;mxv9^Yt3CH|`MvzIwLEi_<}nPXAf>PoBV4O9!?Rm0$dTLQ?#;{hLpGsmn*_ zODHZ~4rWae<}+~^7I@vWPc0KNGKAFLuX2<8eA(>JadSf>pQR)$TxoPJ zlI1on*Qm?giEAqCZbiXOi-e*bmdvCKbcz%1_K5W+4uf5-P!8#?u2@?X!mV+rcSd*( z1NQ<4eSW~N?PyoAC+M*^U__A(|EIE~&NqAvWGS>yW(l7XFVR5l)&@VY;NnQ=$cmOr z)p@3i(q9uy?w_h}l|}3Kb;@I);n6u^eHe5IY(QLXsHnVtF3Q@b{|QLgRl}5Wr4z%| zqx_Inyk4p|i+i2ewX$fCD;bP8-py(8eeZplXu2Gb&~ez^_KrHdx0=v*E^>aorwTw4-{B%n zwpSLKAW5d=369V6C2mX3k`opqABx;Ot##xZRM_U-N3liLzi^HJ6vw9&$!OM%mK-J9 z)`3ph3#?n7wZjSSRWuRlROoW8F*$L+#p$; znoGr9BWwX*p*AR2_v^pXA|b&^@Q@^66#`MqLl5tJ`-)QRfZzLxv_O5Jpc3YwjE&f) zO2aq)z#Xz}zxcyEkc_gg08#01u2RkdK*>$@`w}(6{S z@f>7iWN=rPO1$_m6ITl;GAHCKzYX@jub;o7ms$z{w@_-UH^`&lb212UMkFwj!(AR{ z&Y5dfPY*A71O)nrB;=dy8izI1qDVbH)nNSMBR`VmCn+lyk8I^9$+f01y^{Zo?5XwWs;*=SS$6%~Ap=w4qb zXQGlGRiL{yyoF5jjIdHlE`y6W>i{&8U!aZv_U-6vGn)JrUn&O3jJg5D^{;!)-k zPWu^L=Hsq}iIuE6UaH!Q$m7|vX8ep$wRv5;mMyUJv5wZ)bavH*)hoP^6JCjWZc%CX!qmXKSGR!j*{cw2dZnxyHe#Wza6<0hvIG2ye1K39LMr$H%d*IYS-Zk| zm8{D7a!*;mH=1(qc4yQ;=(|)>`z&*>#Zb!Z*el4;3CsQSTW?|4qXn0VETBJia2&1b z&Lkoq(7B6juF|X^9$@PX>+g;wbOBr~%nQBo@6y!ys$LR2r$YX)M!Hab8&TwaYJH?zn+H9z!C zU}!ra;?=mxkfI86Lv>63nR!kr?$3`!GR{nJVeD|+Eio0TB++|@CN~DC2M0@K0*3J_ z6I^vgf@xLz?D+|7jVGjf&-R*cZ3APdWoQZrI-egi3L_)TiOIBs^eZQeFJpH=D=tyI z^ICmeudoRj0&#D&kwcjeWl^~v4Hy2CjC!E4axuig8D_hSB zEM0yS(@FpYDx_>ZJryBDpD@jWa<0bwl@0C|o5O2_z?s&w4ny|OVmk5#-IU}Er=w;o+;eq$x;Rt6xQeN+ zphB4KGjyxMV5!j4trt?ctCiPEG)|X^Yrjo^-(qJPKp(%reRB__RFsM<%}UT2hmyYY z(CfAdn04i8b$XG}r;uoTDC`^%%E_{|Mm|JUsJw!8xtkE+y+$@y01}K_^hy@u)t{Y$ zkB9oIUGVfPkIsfaM>sofO#8~c;E%crD(aJm%;_#j9(R(qzWctx zg)QGyJ+jTY8(?-6qA4LTa%URT0>M&>Ob=2qTMIbq&>aH!*aFQaX zp^T^)#;FX8P5(+7D+_!)4P;RKK8Q&!Dywum-@N%s89Wj3_|Q6NU&JLn!5bl*c;{48 zdBykkmajBl#Ag!Bg||I91i`nM9KZZVTSE7?;?6GOBKkb};`$jQy%>7Qs16i4XceE# zYWgS>$53iWKy$S0o>jtKl+$ybQi>{nI>V^5EUP30Z(nq?s;e{X+hjpt$%fx6>Q(Ng zzsDG0o={EMAG>Iyi@G6}z?A-+K#UO{HJQR*bgi33)y)#S4d+!#zqjNB+M|f`#bp&) z{?K)4{M_3JYqV2-z}JR;PgXhaY&@98+{=)B$byoeas0(VgjkSzkuOtAh?v*{MgMV@ zcU<$=wpfHE;UyrY;aq}`#FRwuYV*}}qB6`?#3+tPnrRoUC(IA=H6S92gZ^Svc95Z| z;3N4yGY|G^W8}zHiL*=d9&z6D$NSN3gti*@x-tFhUve`98!ZQ8{bZ16=lGUNIi#I< z+>?5ghHZF9yOk9^qaZX{7BO9vHjH}^@91?g_c40Eb{h4`0!z#+_~N|SFy0L8aWT`& zBGrLXg~V{2O6OB!u}pa@mI>8fmK_lMR+bNg0Tt}hK~*%#&4#_kbH42+)YP%_09sO1 zWpI;kA`bvvoKU8*&uA2;-{KhESlAWy7=XF4SMjk>j0e7z1h;jfS1xj`E|zJ8!L33< z%)P~rbUZJmo%InQwH3Kw3OoHDOH0QC^YB?7{vTj1^YnM|POgjJuMhD*HQ!PEE;XKruB_Bpg%Ce->=#6;`B|wsZVx1l zPtm(Zsx0?}6JmKlE>yOkkJf56m~7%2qYiH)$~)|gkghT}CtEeEyBhSrE%HNqQ(*Ga ze!ob)GeZEzsD&xDd@T)Rm3AR_9a8?G8#0i@+Rrz@W;^gUY-b*H!M8`{4nAJi*+UN~ z%fpOmTd%x$9b&YjYQVx$IHuQyg5lN^K4J)#<>Sl>$6B&vHA2Gs@IF@Lz&=#|LnOcT z%9pn|_peU)sc)-ABZyw4>i}_57>Iz%MEBdGH%m(;Iyx>WKPJi4ek=nD%Pj1WvXcg_ z?obdS0r+xShnSB_O|gTEh(@#e^VuDva=WcB+=~g8l>V7AgB+7(r%9_Fv_vuMy@|Y1 zQnKRNFHkU>xAMQ-UbB2_cLI7|bpg5a$*aWx%l{L+ykDq zlVRw;9Fs&g5w5L#v5B6aNvm;Y!{&@oz>Lzm;|E~BjZ)|XTb(nBCt7~kF6U}Fb=cZn z|2<Ys(meKRyU8HDrT#)W>LEEx#zURo)a_S7jSP4pFFm@NW%(l_cwg`2PATAW1>J zddoG7@V$X0)fI7)!GqbF%kfT$O%Bxt$8ugzm``29(Yye$mfzQ(1@3JWNq*wdvqf!` zK?iiw6%E_imvE~R_=wOLbUCLip|Ay2dV+#zAFAFzAo_SS^zV+pYz7r;QZ=RKU~Hy2 zOBfxG)uoneSIuKHjFJ*{G%fGXYb&czxV|5hxhCRuj3%)EmKXeKt~Tvk;4@?$o)e~4 zkjK`?GM9cwN`2|ncN&N_Tb{o9neZ)gs?34v*{(+35O3JgOphOFV_n+~dr(0$xpQIc zQ_0g4W+cyUQZR>J#fz3s*|?b|`Kl%=eeq^XWqK2KDHeu5rL0nd+bZ-SHI4IDm>CdD zxWvc4EnNV`jJMXz*R9qTD3wt2MHhpjNN|-(R5*EpHvLfCM#>$i&0O3RP*d*n=CBG* z!o1HV;?h-Oc$t18F}J*bX#BK-4n8Sfvk~Z!@?u0o&Wt1j zzyT%%u&A|Z{;K!{tW@Vac0QYYh1sKMEKQC((AF(k5#GI)Vu^uYO;92DV(9gnRl5|Z z{hl;)WOq(j%{E?fZ?u)Ff26jPVk!wd-k;7hWHN%xhoF^T`{_B~k9fde=#9en)alwn zLkemGq(cCH&*y^~N!}4~J&^j^N1SfgMT`IqSAM4BeZ!kGhs)EV*BdSlC+27oI*Xby zNKfXU{xG&ScA!%gGyM|FCW-`U6>p6#6HV=>se0Vv(d=3k0c(2mF_RhGJps>fO4_co z@fu#{t>&AaV+YoxOdY>SffM~Nx3Gs?RLE;d!Hz_Syj0VA<=-6D!#(GbZQIXKZio@V zcykD>N!x)z@Myi_@F%vp)K%@l?#&lO@0=dUs_;HM8~Dc8Mn|>~KF`Z<&8QN0FSxLe z9z!~k@J2;w(t&`O2j>L=i^L0EJ(l72n{jd;+$4`-=jAvu)Tz}Tll0V=5%DexMODc& z`5<2tTplZCC}<+Xf!+X!>#8dX(TF6Ea*@K!Ez}%OrTPxMz7JcYZicpt^l4?A@vP_Z zGRfS>bLW)i9y0g+pxYB9pR(tvauLi>P*@zyR9~uL*Q&R|%tx9Poi0;aSq)-P7QgQ2 zDWY$Gn$qOaIjTQLHPc%X^gkRshIRGwyakw>sm+hfs}%9g{w|)5#|^!!9unuf4SWA3 zeRoaF$qVp$DwjXiTsuAYLdE_?oxRIagA-$}S1U_{c9W}X9AtvI8Tr-Q99U1gO(5ds z(G`rszm?C*FUqU__~m@m!U8LL!S%xyD4Hx}AK!5{+GzZ5S%8V~*?Po}@XL znrzDL-li*ew^OjjFEpXi5=)YgwUeOQZXBajJ3@Gk8#DCgfZX=*uA>Y=CRd9$&&j^?Cq5ga$y)NQHVSBT8)zEwXRZ6T7ZE>kbenB@MQvX%kHAs@&my**-Cun_#h{< z1TJ0t!r2$lMGE(-N|<90cJHf9X{o>Y&2;r~Y|o;s`s}^~1WOMA=y6#fv=s4w*n97A zxcatjG!dQXC89^~LG(@p(FsN{S`dWMdx<^>g6J)Th~9f=5G@kDk6y!!UPjp~_x(Qa z^X+4Q@A2H4Iw5nQ@kdr@(zthm7FZSkj6z=AP@8Q>6GTN&yFzVgRs$Be~pLomLDBg8&_UF zuXT)uYq+D226y)k71{5vYu}r1c(?Oi;((4{Ppj&)Q=-{4Ib1$dU3(@M}z5rQcZpxp9N^%X{_B$Gtv_ z0dqK%0b;Y0BK~SVfCcub)Pt&$Z*Eo(OJ0Cb<>23dgb0f`0% zTXGAO7}1}!yNS_t)$A%+)o>L`bwTlZ367VHt=sF0$O-Q(=`~^#hSVfZBSUgAEe&X> z1SHFD+M#%LSDv}{SoXaiGL_%SY1@jl++qazgeY>}efw)vO$Hg1d|OabzdedX1Zi}x z2+VGkrlcaFI9aR>TID}snlht$*a3iMy`KAo3Q~T4x>ix*zP%1}339(7Cq|%c;(>pK zQO=rI0Odz_B2py=G6zu0aDB`UyUAh2>tjxjLYOu*Up%dqaoFNXaniZS*z4R0OU>c{ ziUqkZ^czCvXx|gZj>$-_K6+tZs#n!(Zo8%-WCJQP2Jpc$%ZK07UrUb5spIy#lwb%X z=Q;xX0CmImCz80``M~Y0q$X_AV6EcyS#?W+s!}*YHe&r*RHub14fmf(w!U(WKKg( zClDKB%}eP>UVMaY>cr)nf+KB|-R2n(uzgSr2Y&*#zocUhfLVHBjW1=FT2lJlBSJ)d zjbL}}@QNtoX4_xO!p?m&tOb$FEV=v-NL znqO*_A9z8P@1r7PF2<)quX6&fieBT}zY*-(7eTkoFV-t5ni=!)Z|yz@XpFrlJAB_P z=IR~T6gRt@l}~XFsrw_V%7Agt#}oO(Sai=mKr)A=fvEcHw;M7d|m`fJ||s9K!v zP8PR1E;g4O9rlp5Bbv78IEK#&<5D6#c{^n9>=v)6tR@5C{!Q!oK_53deh50RKKK6N zJ5i;XEy)t(RDP_8YGocNcoMiN@T>arbp@)unTYMsC{AJ5Pet#f?@;?<%3xrNEevsd zkN-WCQVO4wq5kQDWhxGZZsk~yDzkN?sr#hfrky6-RZ0u=Fx{{$bR>wDr+XVMYidQ2 zJj99U7oeBGyg8ayETV!eQj{-sB57wlVd!HjeOiXY?|w*vrTRO@MZeJ`_&eTx8j5;< zUr!txXJ7tj)8?{|9`tyq;PpLXki7V{gsj7f3R;vNvGV9{*hzR8(u+hh`+6x*Piyf; zp)`nG60_&Tn;}Qa1x1o=4Hlbbs3D1tNN*L;*p47 zNuPCcxsv|-4jEJV(*mr!C;NU$c#h3t(h^C9^ML=+M}E>>!`wT}U;Cl)Y3K* z+PMx0g><+R^_vkebf7ltD?gXS5$0^uYz^ybxz`fa(DA>sRA+y%f2g*zO3aTx5~{)S zoD`pKz6tUiY~SK@JR-tDjFXt6RoJh*_Xrl``BK9jRx9p78o5H7ZydH;>;?NB3Zhp% z(g`%bskh#EGHr3{E&E}vHXzIaEEp~s>c;l9o?3!?g7ybc74vPaRkk248qve-wLste zPo%{Uw|ccrJoiKKIxMrE* z-~k9cNw|v)Xl9CJ#wXjOqYXRTV?J$YmDuE|Gtpr|e73{9j(2CLC+Pi?1L1`Eb^2B$ z#V3l^?JdJLFCZB%ymvyaUZSdp3=C8|Tu{FV$0*?#OSkg6Ihf$hEqg)#c;KL^U`T{5 z^>r8|i*2q@SB84t)*8EK<;`UU!W}4Ni0(rRVnc6*uNFOUo@fm8b8>jDk#C^mt9>3D zc!y=CjX&hsf}V1V%tPNZZIn1q|5?~2B-`yVvjmB%^LI&aCEXXzEYBDi7PkeZh&dm* z0aoo~r{i-CQ{J^t!y`*lJMCDHY4*9h-IuiBHJQ~Hkg;#&?@`uSZd6lSSn?)aMvoH4 z^Jz`O+YgYXVNYNB=!DLmf7minNkMFk`!&e5x~e$}kzO@zBc9Bf5ULTp@26>1W$J6y zHJ#xe-#ho3Gc)lZI-CtL&n_3@8HtR%XTVl$2yFmV14QpMHZc2BZFbY zRn7cOl-XZ@v$Cb49C_G$%i)bIK_Z_ReoU4&ObyU%@?Y%Je=haf%gXtae@>k*uAfL- zf$;+Vo;h}6tvrube!#tzd1jya{9TEvE02#7>7eiV0h2=Jl1)dAXiC6gF9*2@H5#G; z9zd;)TN(*;Q8@-vJn5JOMLKd% z1Vj6_5eMEMM{h87!VWOODe~S7gp#HQ-d@`Ls$}+ncK-bvUs*f(xQHdPA#vPKv+fJZgju zZI$dbu#0rhbaE18Jdgmkobd$q^!HyCU;ds~x$qzrKgBDU?3EZC$9dq-rr}p^jGgZI z;xB-~(UEF}mnKACxngK$xAI64H!6s8cq1+gSyqabKSQ{ui~Xk3O@tI36Z_%ja#;p4 zM$QR8yvIt!*i(W9JHVTcv$&LJdCX%F_E89=#eL#DQDkD$>E5m@&b&q<4d!^$h6$|x zdrAToMeYopMcgeLNjb$_L{FB%V#-ZR0?M|yQbrANO7s}=Tfw{|2^ ziwxiX!U@Pfy5HA#*gUP5z|=*4@PMb!_qZSuHDe>wT*3JL;D(y*BPPE+dwK{h>~Ocs z6lC0S_k-Y=>`8kHuaRQ!^DydR?VdKHT1B!afy?O@IzL|Q^m(CT7jd3DK}m@oxm{qj zwasqIf2Lz%W9ocKNUHaOfco2l>u#$@4v^w?V`G94K=BmRyz|LvfpCkfXkp9}NfWaD zOQ-adlGmjy>qd)u)vro%@jX~CJqH;LzIk}J@s4NxSL~_{^jhxBCx$LT zi=s|eS}IDqyxU^RTl81Z_HJAR_A*YXD$c19QeW8iyTUkF8PhD|(i^uew{?2$zG=k)5OWePQ=pa@nkafgx=poaIm>xny(CTu$0y;`Vw+4{meb^Ys_ zZAX)h>P$iPQpzQc=LV4|k&F4&g!76XQU_g*hE)40V=qTnqH?4T`J^3yKQH+$cd<^b z7nsO%?+MvYm#9Qzo$b4qeObnE{FPm>>Fl#(!_<-D^Z8F8E8Ba@s!w}$F@i35FtYZG z^d7@UIq2k{5O$Nwq~y`Yxmsh8(`E5Xb?B4TW%>Rd(^S*VdrC4jH^tzV-`!)VAUK&l zicD2vf1^2VIFmI|JwsZlccbqZHO_>X&lWJ==_!7^qWzR=QtgVS3jZK)MR<;bv6W0- za$|srExAlorb;cv+&>`oym^t?&{f9Af+E>Gp^5j;D~@}QsWhoe&q-l7)Xy_Y;`;Y>@!*c$ zC*o?!FSfZ{<)5&IPz2Tn4+bCNelQYcX+$VB3Y^5Bw^1I|2ekq@n<0ewD}J`h?}T>Y zfUA*!U6cRL)(-s#59XYx{w9-pqeWUBmMtdtRQvWuZH(>e`+PHM9~Q1Zzx+b{L^&E# zV|;T{QqlftC1ok;@SADk3U9|$v|^3*NPc1O*b4jtEgqEx2<193??W2LGlLjS^W}yRFGJydcFX4Hf)Q3BY7k%D6`Eh`as=qL8+RetXS+1eENTlKX=tdsox3}g*(s6q6 z4kHX>Xqs#QL#P-;57Fp8ie&MaDcDs@r|0*ZVrm$JO+L;5-Ma3SDRRGs3cewRv~D zoze{>YcJ!A*}UbarmY!^9}cSc{3sPwjdEo3V=@)t$bQn!fF2>@2_Ya%-j{JwepPC0 zM2=xk54vFhSRs!@!}T5+^RyVcev@CPZ!C|%^BqqzBjO%jdtVn`QV=d_-8iH8Smw<8 z@J;W}VVQ=Ydr7lv^sd#ItHnKG6&iEoOx)DwE8D={2KgA=)uj-Tgt7!Jb;iV{1rxP9 z!{M%`;EldX$Im|(=q=JV11h1ffkU?x=w&QodE^I|h4wNm;RL&h4E8@yMY;&q(F%{B9;dHn zWw;kNsn;T!*8oV;fjlqsEC`2v54^~Jrqa;y&5}(cTZc5P^m2{3eKEExCeaWMG(j6| zp2jZjWdGu=A*O%Kue+EKi{+CWhG00rgXNTgTjJA=3r2~b=(qb34uA&6ZssiG=9;r# zfO)2j->%H#3x(iFr*4NlvmI0uA|{ciZB=h%f$@j?Ivj(w9rrf`u|~74)pv&)7Z@pB zJT^YZ48(JF4DPY|8SNFA7=kD?TV4S~$e^d39;CCSs9)x9GD8twsM(pg6rbe@)?M-6 zo3si!L&*3wjL=8sb z-cJYAjkhH5!uJZLm0+as7JFB(JY3dmX`N;5$++;(i(icT(G@M0NfBQWft7r%1pc{x z)j1gFn{g>z4%F@JnX?Ikzr)0 z{q|H&82RWq{O<{PbAa< z8o}A$u ztP$lAM?W6zV+e!175OvImON;*G1EBVUg~9 z8&Go3>P@1iw04Qw3N?T4%C3}U-v^893!Kfb+8r%8aG*Q%sj5pM(Va~#l^x|KGM7C4 zMah%glD1#e?$!5?y<{~21&Y1ab?b-A?-Nz>9RHu6yX<5D)*C6{OqlCJO{nT5lL#V} z=xZSVpm$wgu1Sz55}VfDa5~s~nbGaZEJ`~>#%GFXBaiBcqItcH{}An6Ldd<87mn`ZP`2KMfk1dc0Z zIQmJ1QlFH5EJ&1C)M7FEWh}HBcl;JRE3>XGqfY3EKubv5d2ZymJ^->%T6XfGaxb|5f?&11vuXKK z!0tCQxjvddhhQ#zN_nZ}tw9;1Q7d+~wV?+bfSdF?s=;pC6AI)u%@}k5e~sVuEZU>R z+1SIX+0EjvU1Y+i7UJ|UW_=^g6@G(}l;L%3809eH@iZN|UuZHm4OgX|d;-k^e#nVA zKGh-2oP&XWbh+)v{#5bulsy`h0IyI_H4|>~l%GwOtK;N@)&faF2 zp%y?UX;7-EoWAibQ9gjF!1vzfEOT`7n6H9o`%8+rH}efEZgxNa+58%!H_ODotv#LR z5&T-4(T{dg@Krshg!?z~_`F(bCK9wmVppBh&NqDO2w$@earXwG`OkkNJ+A@BXE+80 zgwAFi^>9jIyVu7HR!#6c9TDP)swxk`deCZ#?p_YwD{B{?T60D@?29l}0P9%uuXQ|! z`ZVvpEdEDyLRlT1AOKZ=aTpi+#GV^}0o{Qoz_;pb%G+|b0p6;b($1}4jvpfhz%lX* z9`g>X0FCt{Z>0pg=gjtAGP|~Xtl&DiVa*b8Y?VnzfD}0^J5WmG;Pr)SW{NWMj1Ufi zwowH(qnqa7{GWbE&W#VRR%P zh8yXzFV9$ELEb)HYRCfyNX70@BEf_H1U>o4&jm{@{Q{U_C^+-nv526)gRq$xPPJ*?# z&AemplY?%iim4vIca{8jr{B3@4oo3@Gu3!-|JPup5qvd`z^)oem&NV@S-D5|{%I%a|fN>V8EJT!` zu*>xL(V;YfdZ?hyQS(&>mwJGrDE>DfS1tIi3|0-wn%-rTH1}1cKQ41bpO@ONW&6&rKr-FKWmG?^%CWjz~P5rr4OD<{-)luUC!HY*jl}b zB~Q;ZasbMW!@Q@S1>5sI&!q*5*d8hGT{V8YVdhgBg;LuRCv#8y6tGKVb?Vff%O(ed zsmmSF>Fxv>Fp@kGP3mVsGTU>hTD)-x+)DE-{x{Bi-384?!APw8KRNwS`!<$2;N76{ zw(>PTxO4x$#r)>aP|m_*S!ImIvheSo>z(tv;!G7r5@yNd704I?w^Uj$zMz(O6*vbc zqu)UR(_b~|S00N^oelWhUA8%bQ!!ZlEE70kw#Fib`BG+#>m>4gJL|B4-{+Q3xL1La zS@1;*hiJ>G)JSzO04Fk7ta8i|j59ryjF5X(m+{n&F5;4wu5vkk7Z@C;G}_1)tYuLK zRx+z_sR?VDP5INBmVmgzhf^4s&H%u-xeEl5iEa^j?@|-m$Dz-T327vChSMmIJgh{h ze)-YEd}yNvxqA*!CZ0>g0WUC||Ad|ZX@+b+$s6ky*T{yY-mwNJiH=Wa(pLScLx1Uf z)DepU3E$xcU(S`dL!l_VisJDbKLh~NzWnl8a8TB4IJGJ^qS~U@rRv?RKwZF0<%-T$ zqe;_6cnO3Vc2}-bP<&N>*NIER|1G5cVrs)kW;9cb2Y^s`;qjv!Yl0V5lts2VPw zi}vMFeIc|;dte4dy=|7~HGJKtoKm%{e!k&zx5>7kxasrx<71?7Z|i3e124^LPnuN* zD(-C2fq`MJe(IF^&4M-lD<)gdEhjh@A!dPU5|N-OwtMyP{!+k}l=krstDSbrkNxL+ z`i$Cr!y&--=1Go2qzaeE>dzWz&npiW&RUv_JrT&KF?H3bBwQIq6{yC{i&@O zqz_5&di4okUz{0XIYa;C9b$jy9X2;HzK0F)A4N06)@lm&(lnD4&6Rv2Obrp?AB~@U< zgcN-o#_cxxj4O_>iZCpo7$wPkP7C;n-ph709AMx!XgM(!p(@ev;SB1G69ps=DOm}Y z?zD)7A}9We8Ee--JF5^dD;&LVF))^-J$fHyJCk2Gb9t4GK6?MS$ujF+$i5o@OJ&^L z^x|1Z1!ElTTF}b;3WOcgllmJ8LztUA9~z=8LVyHCAE*M{kGr|qSE#pIl==W5d6akhqy+7>HW#1tVe4sy{XP)1X42)-{a-`Ls z@Urp6TZ~W;R46OnTp5#qiesuA@8M&MZUWW=FQ8@CW!)sV&>>gbw8<~#AIIwVrQ&t# zD@F2d3{nw(;{JEKD^z&=U*m~hU?oyor~~N6v`@y&9|GHtGRTTY06-AnoD%Q4{!5x= z4(A$)YFPa_>*;6dRC9HvK+Z6R z+{WBF^Q#`Lu^ngS_jF`$)de`p1tRG*#PjKSm6NNen{#+#RkOmHK-ud$z{4X2Ix|&U zNhjs=;dhRW&Yn5W&5|#I)7}-jQ29(mH%Gdy7eFvRIo)*QjWe~LHUsk=+kc=~Tf;1+ zli#!B+&Wa@a*z5inbNw?2NKy7&kquPhY$Q2K3`F`C+*1~g0CU+R3k=MIBhTV_t4I0-&kUfJ(FA4&o zREUVvb1W(Aj%>0Trx@itFY=iBB2JI_#+>%Un>rd#AdDCp9Px7`XI~gc3fk-Q;u&GLlpxW3PiCC!L|Jg)qPYx@*q`zSE&#M-qp&a$?h4G4ecrWZddg!;?kaLwpyej@ zJ+v!Yea`?J=3A#);Sgx@%wb8Rton5UW|w?4JJCVC$`CiGc$THHTcNt|z9%(yC0D-# z6|D3+j81onB71K_zo|GJgAaB$K>8wzjgBwT%+>$S;yrC5-tXrO?+EW>2WMw(MEwQP z?#Kxlckoc_<1`M}sZ8m)7792*qu<2p9+({UJ&@4Zm&RuNo$+tuih=r528!+Hd}rlUfTMh!zJD4|j0k zDCjw5tyJaH&}bXUXfRlk%|f0Ta}|$*r{qVG=jhh+@xIec4=$VKz>LpXveuvvJuWl@ z>dWuOTFYkY{Zj@hD}c_oi9m0M%9;`s7*Ejax4d~ES85XRvook}L6+0 z`Obv$SKzO`pKOawJkOqfX%Lh-6)HZAS@aywMkW&Yjf-gBP40lWe&GjI~YM3()UdKs3h z60DsWDtOazNxU4^YScF6Q5OUywjO-$Yu|E61zBsYL5MWWamspdfc~88JnCSG z<*|rk>GOBALvtz@0M)3@O^ogcs#^+j)-HRk=X^ZUz|QD5T|V6lbb2NZ%@zemCBL-q z$v6Vt_UZG3#gb}|-!m#Wok#)Pp*^6-!)dVFOrv|zCje4HEHo*U@#>eRvg;MsHm=1= zz;16)y;r9d#hRslK<8=Dy>XEJ?%iy);c&K;(G9oht+RHnJjNv3IGV`dMDjF%y{61d z7l{JVOl1e`-}1qk?(sAysPeOmYdOZbS13XRzQASqDJ~_EV_d~#M~9_R1hOPAdA0rJ z%4zMY^*VtI#*}uQL_8fE6LZ#n8Cy88+{dc2$_22FcprozLK!rNQjWUs+CGhD2^q?M z3l<7pj4oJu-SI(zJ&0Oqae4dHt7vzoDFsyH(e|3`6H@ARUvuU&mS@P>dUrvN{JO~& z{J`{NMNsNkD1zZpsJWP*GiiGj4tGN*)q}c6sdu|26-h1j_&(_a>KAaurlK1=s3GI0+a zZTKr2g6;1wI55aU=!Xg^R*8G)k{K&$LVPVipFuX~xyHw@{L$z+m-T(_?d!GbDD91Ul%@FB%mj_zt+ynNU1c!aTO;C?^}Bx$~;oo&3Otc8=kK$&hgJf+q=*CCisy&l~;@i`!1GKLql=t`JU#OM>>e}N`u&YF)Nub6V&QHj5 zyI-Zb+rN0$gNKI;7ctkU+EtqKe`Jc~CUJnrh>jokF9M?pdDf?#D^EVzBy+k?z@9dm zgkyc85onFoNHq?T#zM!>S|xA4saH*904d|%G@k$T48o$^%cobP_y%rqp0R^-`q|$e zp4D`9s!BQ8k@5%O(m5kzofhk+qaIf9sxH?*9_C-UpQw!-235bU-yOBRbsVM5!pZ@t z2=8P72C-25L-_G}yXJ7FbeH}&7&@O*mLMcPojLj^N2=(Xum4LvrMyDzT*z-bvYS9N zOl-%VLv znRE_JiXB-daF@1giy$>>*XrC?ulC9xQWo^cn+Gknpe0)#vFx5yRF@VD7|*>Akg5a~ z{1ppN=Vg$(i0c!cg&%xkt(rUbBbtB9k0Q{6ojhaPyd&2fq&T5Vlf_ajNGdm<{kk%X zsZzsT|IV`C2`u#F>gmrSAHjp)1|XM#IeVgZVifh7fMOIGF)-e&JM}|p#45-zs>b7V}Vc6x5kHA8ULjE)d9dJxx4s*2xuCOM}EOEW87G9C; z(ddm$$!tq?yLqh(g6$}+i+Zdg%SO3Q`F8^NX8Q?M3z-Y>v^+|SXCmJWj2PO_W$v!I z20U@zud$5+1ET%Rlj+J=J>3tt7%qHh@b zXtfm`xlnd7cN7H&Gmj@w{CM&48b6UBKqZm)=26;UL%XbTqr#}L&JT~Qo>SU7JRD-2 z*ou>0;-lbw&H40Rc<8AvYq;^ma>Kmry{Z0mvm2c{AOGGnVll>LTB>RCs{V+FJN8Co zm|ApR`Ki@2VU6QW3e;A%SPlJ7EFCZUPQ#O6EGhZgs{sI>Nr0JrQTbgdSV){znv9n$ zYO#GQrU8B7GWWq@Pn--V;eGuTBTeJZwuKsB8*OkOYs8~qS2Qfx=WwO_*e1#zA%@NS z!AZc#$Io!{PE8}8mO3`}aR{v&^|j)qcrTZqf@qO;NxY~NvXT=wD{9Bq@+2H*$3*uy;_XIjWybTdJa2DJMCpop3?Ks#4 zE7HQ&BY@fyy28}YWv>Y;MSM&wSpNF>UwG5J_jc)@BCu-rDK^mr<{5$!mvXp*J8hZ5 zZ92Ox59lWs^Gu#=0!Xy;foT8k5tW;?)X^qtjs4%=ZA z?apce2Y9t<7`Q3gjvgyMo)4>U3bSzqShL3jeP^4zWy%gAQ91KWM*LzvzwIkHjTxV^{^l_GqSl#gRX>t@Lq+!;QBfa zh&qE~%BEQl*h>~$pN>zHpAQr&{U8)g+X`wIx)Mnhty-(A+bph>@N{avJ{p=hr4f5) zHly0}V;WABN`!TOs%*VrZ}O3vP3^k#c;`D?e?Ix}!$EbgCV_%i9gqDbycmPDBIN{Y zsS^7L<5J&k)tS=PBFmbV{HL?Vj)w?~c^*sneEW8^PI^^K>Zc!2T)ykO@_L!BvjzW3 z>}|gmCF{_DC^lO-V5sifpa`wQ!L8wTvJ2sv?Uwn&)93~9A${-iXgVCqr70RT`MO2w zr*rF(i@O}S4zg@whVLfR4JMbLbxTiH80m&-$;&*xxz`x*{DOrUy7X{&QU`uY-?6n@ zw`36*aD_D^z2vi12qE6TPPjsmHS+(hZVY>|JlYXRiSTces*j;@={fq-)IXBiey-}io|Cj)J)v>>|K@3sG|^OG|70OW<}`g zAB1Y=7g2Xvx5$w{mlW&bUFRRbwJDW`FsXIuQ#l!Ej7lDcZa zZOu5a9j{ObnWX1|ah2`(i@Lbg-soukvL`&7JrZ#%Lp=?j$nHFC-!3qRkEH66d`8l0 z34KO75RxR%?n|6uSHKtqLSGc3?{y20dT8^9xYOak>8>Vy@J6}vYPRuGXd`{$DtK48 za;=_VL*C6{E-x!H@K9(-IkQ+(OdlDIoL@k~P{_lCq)nvHrq9$$jRC3?BUqG@ z+ViYsaq27_o+-1+w{z?KXa{@0b+owEM0{olF(=2r<1kbXj^Tt%yV__ZJA=!oqm3&( zMRh-zF4ybgiO;wVCTB6;tsW~kkBMzXYi3%la+t>0{apwX2+HQP%M>7Dk@XP%hEf-fAzbA>cXE-%LQOYrgvn+H+~70O>{Q zR*XA~Dw$rDd)ZXlFVWBYFEVg~Xg4%+#D!P0U5lBnyqk62)TEhOgra(q%#-~C+O#dgvXLpIt`|ro9_hp)(aU8cbG34w>aw7EnZZO zMze8+b6;lS9=V!!=Cwp;N}}gd;Y?HV_x(KBMJJ@!WsA|z>i~yqLV2y;e5dr=4L$RM z`&N{syl5NVA15$mMjN^asU8UfSpsS6SxuEL-zmfE4D7?2$jtTdW4UhF3|IMuj8K&<&&R;b z((AUPGcF)Ybjra`BrScq!tk2Yp#IZQtAz%^_&n@NtpVsURAf+Ezy92Nv!-+ro&wF5 z5|0!Xc!PFfzDgA8q;~xi$APW26NDoB1r+2uK<{eueB3IiVYcwaEZ>i;^v~*1Q|A*K zioOP8m#qPL{+v3ihfW#M9{c-kMkXd(Osp=8S0hYU;@DYkFU~sL@9+^qZ^xGp*^v#nleA)qB-6cz~)&!s=9 z$$S>)F22&{|IN=v(dZrEy9-q1lcsD-x9E0$n0*D#+B(r_;HSiO=S;vjHNbezp337=i!St_)Ubwg z&$UHTW5`LZP&4q$$)iRGECoJpqy}NJUeOQyw$|Cv}E|)W?v8R~!g`ONh&+lBv z%5V2ayygsQ5TM9@fS9=%HNFP@WXODI4|@}Ey0 ze90-d(uXE<+Qt5&morDoGpM*`(*=Kcy2416;6PZ}Z1TX00njEG@@zbg5PC*p+}s`G zQ2XNnrs~S>V@DPDZeeKEOqr@L|C^adRyU;i$e0_m?g4i1Qq`2Qs7Kr*>8HZ{*6T!v zNQHB*lnM`E|LaFj2HJOALr>h&-MZg%TPSqfk@8x`{|<$4_b+e#-0BZC*|+#e?8wLu z-LoR1vVs|5eR+7B_l{Wx_^=XnO#N9Hm8)JM&5&C+a|{5Rwh%W%>$FOunds9Y;f;#aN;#;X}5PS-;UU5E92EKtsD0qk3c2kOv; z4hS}Vm$dN|U-l)=UnOdu)DR8fQG739W%92?SWe=Y0{o0{yb0PRn#Vs9PlV*Qo#fUlw`zr1$KF1kFisLdSw`e8pcAS(wq}3O*H6 zW3vxb=4(K?k(w-DT_vlDL=U^<>0G)LC}Qj!KJb^}!=czi${!Vf&~}U>o~eHuA3XSy z+Fz_8bc?q%hEO^BjbsQi(+k-piKbuRb+7?&w@7BLvEefvfD+97S~jc(!-9LaP-7YE z{6}Se?~N@$H>=1As9!Ru^;s@Zf1X1$l|Vmu22-ni=de&2n~pS_RysUdkn@(qhF>)j zp_!kxh7FP|ii@F@e`55ROi9l6=#Lh#@c(KQ2KltI)$DCf0bJi(mU0N8En{pzy2kbc zh6N3j2af|*TB}ocexLRiw#X5qhA_68Jil7<$pV{jjWLDzvmam1qc#7q8GueObuPtE zYbNi9>LJH8GKUy?TPNtm@8a62vXzj*4`_kMn1!|{-Aqd5Ui(FykC*DTvgkbzsx$bc(`9N+Wk>@ zBj$gamqHXs0IH?C`yER2=E~r;?QiszyH&7YZeWUW?+IWdg zbQ|UbCC$5N8>WTKT>sGJZK19au5248 z6oD!%KS9fexcL5-M+n)`;jl$%u`KQJAbs_0M(jB-c0X`TVcERW^6SYhZ->J6?btrDe&^!XM65*r&XB+4=9 z0QYd>d+-)bmaLtftL}Kyi;kmVx&LxA*~0O=L)kiC{Px|VL{j503!Nli%fIh#z))@lGzt1uW9R?vq|({Ta&K?;khE~% z(Xgu?b`)4je7^ZS=c@Z1zr_T)du31f87Ux_)z*mhqYd@s9w8;sZ()a7R@M)+tenc8 zXH)2uH*CSIgB^LGN^~nQp=8t38S{E9&2G5*-unE;Q73~<47S>n3`U`Xl znvlJ>dd>#-R~|x4I92H`i_wWzn&Q)nzX~gO$=0R`$SkW(l;6}UX9znO+u*4pev3%T z{*vJJ!4b(rW_?sKRxR!JG!8?r{WU^WPVkMH;(swKBoYDzf?wMfTT3vQbhltz`~2no47)5f?;lMS|cp2f!83TPG( zfhQ}V=P6(HwG=>KU1X}?R-&iZA|mQ{zc{|p#s?yFnbE9-@-E z8%%@m90JcG;IO+l3k}g(ye%gr@hi<-d}roPcmKYfK9*uLNV^sCw)G~L&6~ubBKwD+ z#5&ilyTv8O$Iezq3{F7z?ocg511(2mM`yMJ&|8b~`f5e0ByyoK?;?hikBLUsuG^zW zut}q8GRHmU_SIbMb8fNEd;*6oGr z^~n(Z{UnM&{+)_;`TUmEqYJ!cz-#7TCOaQr;xL4tpV7ATCKPaeNBgL3e0PJ1TxUoCmduQ@E$Npqu@W-CF z?#0>5K?P|jydw*>mj?>y8AA=arRbiVIc24W{o56#mFPWD>>YG}_>&LzHBXyp*I-8E zU%zEI{MV7JPSYF&{rwsX>BY)9q*pW0k1gP4Pw5bK7mz4_k@Eo~JAuy&S6T#i(0-k8 zM{3L0qSGwUp`YKeis@`JN3DybVvEuJg3NkW{va;+-(UXsmHHna88m>)yN2FB@;|=g z|6zNWfO#JmuCgYMWdG-%S!D(!95tL4KK%0ovnBfc|Nk@p|CRaw%54@z#^uNMuvHTG z?*__0cnrc*WN_{K5VX%Zk`s7y3Nb6_1%1>%vRwEu`%m5U-lHdo1(7lNu^sPq_WZBT zssHVg_P>3ijE2rz-0Mjq8c6rI^8Ei`!}NdiX_6IS1J-z&sVv<;_D9V3__Me8K3u;1 zr#^rGFa8ory^Gr;KuY*W&X(rCcDDcC;{JyxMG3N-AE+KkKycwfxo_nDy(Rt+@Bi;N zfN}CgS0uiZ9^uAc{_`lzD|vsPN>VvXR0ZURlo!POC!XZ-riLX*MAO`4(=Q*Z|jNJY|L$qLk|# zqLisPQF=pzI(Tlh^bXNTjCk#8b zQzj7ltRs*EMsKQ`i>g>2v{C{7yMee)mNBo>C?i;ht)Qy*b=11&^}sUp4=fq~;}$uu zSaniQC(Yd|yORg~$6ZGDH~K9we1E)0yfY@bvyEAmgWm9GB*hIa#f0a! zZI^BbH>(+}RSF#U^D$08n|S7(U1I8R4~UUc9Opq?ds|D_9@e|dldfXy8JwY~&Iu zaV`Eh71*wM1%d)A1{4n^-rEC<;vSSe7&s3Mq)n-TcQ*$4Dv7M|=LwCyg`uv>#`4@! zO+?BE?qW^!fi6XCk{W;hnEHDMt5p-3YIVtam8v&jM821vmm2fOMszPY8?R~dUp~f> z^7jr#0{u)WXiE#ugW_Ze=nwpxTq8P;@o+xMzGPzH>8S9BI*$5-^>N|S_jgBhYmhfL z=Q7ag`nGbd^|C zxonXol0}~rO2Bi3>8lzIOSN?yPwS7p6o6&{emCE)YQ<-m2mStW1+yaK>WAmma6{gP zAZ;&zq%#3cbdEtu{V~t9L$c+Fu4#BCMT=(#)sn-3@p%-DSan38?`81b#iRi|v_`)e zq;U=brJZzSWza%n)*}Z|aHg)}4wUUrkUQgAQs+Gs^Hq z)|ViC+ITj@UjU+qFklf@3-~t#4_%0}|H}gd_ zo`b2dHqgE@%q8M^HEMf?U3qLH*#J%xA+#qVE(+kE>+d@|&a#>YF4o1)x70PoQd4mO z!m{x!lY##J)6@W$7@{~=Yw(9t>AqoJuQF|t0{+CS*DZ9nRoG4hX*I)Tw&WEqGaCP6{uii`5jzd0Y-#$<>`&r(FkR%j)-)O|kf+uJ}n~r}18v`+R{-t)iB^J%NuU3w9H(7g5v7YLy z3KI$9Sm{%W$N_5vz_Ys8*ve7N%~R|pH!zO(G0*Q5ygS*}cE0$@sQpcu*m?1En(apL z`7pwsUgN4iEi?|8Ae5zK6T*WNsU1-UsNom`{wPn9xzjh&c2CldVaGlF#n-OZ8YC>L z7t}vN=ew-G>irG$mJVpCpKlH~gvJ$kmk?MCP3`Y`&QFafZpAuu@3jVaBX0axi_#yy zIxFBCA^)1XRrkT`cqF|>7N1W1r9-v|LcwBZRqT*r!(KuLzaB9I@GkACmjZKxFJS;> zCNxk9bp%)?MxlibLFogo`7=-38tS9>Dmr+tQSt$YJx}JGv!zg3-vQmFL17;lgLau+ z!P|^)r{iBoiF=e{4d7iAAKLOeZm(3=xn8}#zF5;-4MMp_>kUzQ%?=A`rXA_^{<7>> zj49-dChoxuJ;u~aK{WB@bUJ$;%;?uTEOHy)hFy?6_jO^69UqbgWMBDSd`M@wJ@2@i zwGJ$GQvup3W=4nIWg~Sw!__~FVnNjfyEzKk^gPx-0<`(PP%&k8vU51!0d2-;>6`H4 zNiP)?axtr%&mPbSCe!^ImOVS)@P9G&mQhi@-y82&DGBM65|joJ6p$7H>5%S}Zlqxb z2}uFzPC=xbA%`504uOFgVCWu(8amGSJLkXFc{{J3HS2lqd+&W+`}%;_{6pjZ-iH~C zi;-bNE{88Mhk@L<1Wu9bccpKs+-Wr4`PD2Z8y!mRz zBl@uz4x>;{>euv=Nai;4@ARh~JfX~HssuNr(^?iZ^( z4cGNA&%1O-_{R#ghe0}T{`8SVVaAOWO1wV)ExaE1M|4>&L55=gYK)i9LJ(3E4&%8* zjeZq7RfOgpo+bBLoS00iJJ8NlS+dK5ez6+QVaNU}58v|x*1&iM*3&8wY)eI*5CH=} z74`VPEP!_X?~GC=jGmj0vvSC5bn@R-ai)EYtLp{s*&f^Y-FfyfmfNGaED_()d~*HU zqq~bD(^zW9#Wy#S7f~S(rrX89>i@g-_98PsCnR8A@pN{~EB=1IVd8n+Wy$7n+~vO% z3T!)(ekvX)^$iJw7S>guuS*j3uk<#Xgtl>)BO+=n%s5h8PnmEF{QR)Xd_W~ifbG{L z9a2+D@;>voSwrFF?zXucV%bzaETE~#C5dzY?sY8zf>vjt*(T1eZZoxszQ2L7T_K0& z8-_eM{NjBgc>ZMdua&UzY{;cdZj^Mk zim?=Q7)eGYT(%cPEpP)E)@vme*mg>Yu)j7K(6q#?0)3}PN@ zU#2kO;34~!6cXW!+*O^HXcwatGhbwrRd0)Bk-^jlF{jJFs4p)~GzlI&>uNyE9ZqxA zG#I;xru=tCAvv8glOAo4j3dHh5{LCRo3*^adVkofWz{l_pE&Tso4Q8oX!)xWQp^;S z;AM4KIYi@C4_fwheAsyNwh8)Qo;2nXu77zn!OS{#vE6Bc>p@(55^>~`P7dFrA0R_J z&D6rZ`vp4)h3#QFDK#rq2{d&c+|+uv?Hp;q z2>%PLPd|NL$k*+EnTWA75{|~)D_=@RBA!jKYnfoIW*ELlkxi7a-}xPhbWEe?YnsOz z5<>|)^b_}$@a{D_EOPe?-?}7b++Rmeoe~fm&JDs7Zz(3c#oTZXea!xE>yw&x<*VGP zI7?l?Q)7Y_zR18&26KKMI0=uGh*W#4){O(3v907$>=rnm(kZkrft+smn}+|n(81oe zZ-6B0cuqgO+MlkfF~GQFVK#SdLh^&>RWGW=q4h;ersK=;tRbL{%ZMG+{g^?7G#g_KQGii_8mKHuIJprr5mZ%`w5jP`HwUM5> zx{p03*0ldR7Tt%;@KB8;VWNyAX83`kg+Suy-k4SPeFt>yx>7ut`P&`9JqTr1#S4^w zz`nUz@EW!yy<@SyVX-~M1Q`BIVwCpCq7|{s8on4Vhp_rq=X?T1r?J%GP25I-r{X1B zD=+%GG4+gJ$9F&PJ6kGg*NgGEPwPs-#}NG@j9)A@Df%wV zon1v!6m^;0XA=cttE?Fk(v)jBN!^3rVZV;9PHi-i<`@L5f&uxB_fo)vj6az?={w?n zwd*pi(SL??FMT8yiot71BakjA$5~F=42d`w1LC2+xf5HL_jCVv*g>7X(ipE2FeZxGZGk<>g7qWaOp>1D79eZ8P7AWN+A_ z#;T2hktErRp9M3oYz39JMnxlIFd4azsSqDBVh(0eyyPSj?p++F27!Hh^wnRW)kbR> z)WhlqGyoo|J^ed`?iwbJLq{OiGIj5zS(shv=7D#MZshYrfki(td|N4G`kboyaQz*+ zt4rs07_V56dHYQ3x+sT3Jtsx4S>$(y{U3U20nJwy;%n^b0hMy|Tx9efOA%xBz(Gi^ zAd{$-uieO=L@ZvloCSpN5gh(z$#&D`u+QIX)QL#&5Z0-0XtSAe_(-6&J3;))p3VTr z(@=m4q1osF+SF|Pu)4?A2Cy0RYiP6z(mMRPF?myM)n`0G!#LhBIP~_&8WGwpK-@eIh7Ml8*}P^M+y=u^YIH{NQJcG zCKytoDXIRy1Qu-QBF5^H z7s=GPczmfFBr9-=7C{W(9O3CM=ZMye8&$0UC60Pzy!V8W;ygf$5ZcWj#%`VR-D0?F zAZZzxJ*?nRL-4aA|A}N0%>rj^D+vkkWlbNM2h=oE-ql~)u8ptFFlL7)&y|7ZuiZeS zN0IYxm)&7`GF`9*-D0JUFdHBo+OM4mYg?DL&OqB|KOB&hT4Gb60H)I^} zcf6Pj#uyTZ4U0Q5W2hKIOcSHqLg8J2OaMo0p{2N%gqS66>j+pp?6yXWGz)_szskmt zW@XlIIZ8HpOy1j#WkH1WxUClMdDqol=38_-?msSW^f)A&H{_A|xr#y{ck&URyG5lQ z)_(u_A=T+z<6Dt2tX@6%B2_#I0=o0dp{JOh^j+&*2kd4shA|r5o}({mA9mJX?3tpz z2f4LvrhTtRjQlh?+VAK{^HX2HI~b_}>)n0MTWsApWsT$}+-tnK>)bnvWFG%n9X394 zFnA37l?m`T<5{p{B;DP-VaOA23BDt!elZ;4PUff4_Xeww^_Bp}U{FxgcP#qF))&D4 zb_9hs%bpBm(K77$ zq;WsoBr+T*o`HNjA=`?*{KgWAsXHSf^b?I_L!ip4_!E@Vl;UIz%HRe| zB5Z10UD1g<2y1&N86L2_x z$GQJG9gDv~upBLA{s8l5wPjtnnhoJHO|5{*Q73+=evziX#ZqVdEnQt{F5A7|2=^Ag zy8*h8>}gU$EECQRz(jC*(mE)@*m%|F-$YTd3Hwzp)$TYmabv(!$GNjqb1Or@&8MKB zUq_Y$RCrSe+!QfHVLA2Pd)O;KAh0F4>$dq`<~IMbACWp8hra{ zF-PEM3%|9q9E#dy2O=0mwe1rDC{>Avq`C97x`%4cG3xw2h+5}y zIP5qD%L_KD4*jVgO}zMJU~ZcN2xXQ122bbbKW{(2U$TnM2L}!k3vAUJyK%Nji+L&X znl>r|ZsKzV+443|WgL1QMq0mE9n-sU?z_LP@m*-US>`qD7a5E%6B|nz{BR<1Tgz$4 z?J9xzT(+M%xnEtgv~;of($=6lexR*=V{7)0o%8I6?OuL$&6}(@Rj>=x6=eMubGQ}|C3@XJK!a~T1+OlGJoIxt2 zysrxor4OVIfUD|FYQ22CuvkpCF&QRWVkWz?#9vK;nEts%PbTRuW<|9AGTeBtvt2lV z`!^Y!{?${PaIo<~f|ApcXEfml?Gmr%JbHP*GaZ;s;1ukI6V5u$5flxVk> z;9i?-nV|c7jgs&2#h1(1%fy&e7rBrSh4%M}k2VYF+-7p-tuzu4wKsz|C{2lsxA~0(5_C5n+D2Zct1MaE#nb{Nc)%Mc&fej zu#Pw_*1}E<`*4U{zDm2biHs-2BvbUpTflPqtk$n?o`emPjMzl2_Do0*qCsK@c)j4oXz zZ}3CK*>ei2eGD61f7a{Pb%(bRcdoTAdN@4=m_|e1S6E;jA^RU)dj0uXev{kg zjxM|3LpZh`1>q5bzB?=6wO4d+U}^<^2e8iktd|bIdYT4`Yv8W!@)CN#%3(GM{%+^y z?mQ3U?M{#%t2Ks)?cTtaQcN_#ufxF3F>wDNlgUgU`SPolxiB+i38nry;98`z2;Mm_ z@2@@$KdX1n9PEE`b@+=zXb)+7h1qfcnF0)8PMcaLv}ID?mkcD)Nq%!AvE1ZR#X$p* zxK2qbsrZf0gF`4$l;;$X-^nEIiz36qi_@X7^LobX!FYYKyXVNofCPmts85-*R^lBh zy!j|f2|Kmb&?0bMjzL%Fct>>iq$H?ueF020$=xWLv(k|X4N@QVBKK(@*nKzyb`Tf5VDoepc_ zRmNn~PF!YUId!u%OALx^dP#J-=k=r2Nj&is_?I_tKo{P{T%S6kb~sRmAf+Wm3Ki3_ zK;OBu<3H*hm!2X*W!U@wd~rt~tq8!#>n<_gWf8$c=^$$>D`NIsF(qDr8iII zo65;zkxJxW2nf8&GaOL<``gC_&4&#=K)x!i^BGP>JIy_*YC($l)wVo@yu(VLNMIlX zw0UbQ8Fw6o5=|I=s>+6&A5P12HDOBrl5@)(r{)elqXHdfmedChjhxBs8FE}a&Mysm zRsG(?xD1*p3G_nBT)AUdJvh~^`A%@*yu5DJ?PjSDb8cNZ@7$}1M8q)q$8l=o`61u_ ze<2@kl!k^=fYSlYhuyiwK6lrTk<)6|L=2ove^N(ajAPzVD6z|L)vJ@DnhvJziG0ur zc7n2QdDOsIHepAXOvmtdCH$tn0Xo!9#}W);j4}1CAwjxCU-rX2HxW>C{;ZM96^P0J zb34?Zy|}}684RUZSP})BKYQy~2(}?x`%|+*%Mfc_^eO8`QloMA_ho!q4)Ps0c>U{) z!=&m#@|#Kl`$p@1$3Z+AEzFjyk!r4#g0~hNRH?Nqq-7vK^Bdh5qY>k>Ghn)6%c>IYlM2kzmI>)nl)TYg*kVznWPU zFq}p1lNsQBZV~9bJlTP!^M zIR^QFm3HWkCNylZh3uM&s+Wi)nt!i_@sP$v)+2ny(a}f z7)-(+KX&+`Hv)+b{&OkuGzFctjO+BWkww*PN~`Vz9R{rUSz_}ilUx%|E$i&lFlg<^ zxjI=7rf|(m44y1?Ru~wDk9((i7|#gf-sEp~3K3TW)L7A*eLUS5DoG*lx(|?HlV#ZO zZ(!%qAe=ucYs6%su613m{ce-e)LZshC~qrf=$7rmik``TuEy)4Ns`nTgEx#a{9>jL zP&z-jh|Zt*1J(>koFovcdAGqCncOY-GPtnCFmuP8rc*#=fWK{kt881CpEzS-!K7_E zB8ib8iHkxZR`eh!JHN+XCKnp3G!r`NfW zXZFCP+Sq)G2)v{&I|F|N7Pu^(`92iwPt`--Sfy9;cTm0;jhDh87FBVE59lb?{2wN5 ze2mwO)}dq+J^ zM(G>a71AN%0@RRzUfbBr#y&#RX%9xI&Gg8{+GT=1Q=edF#0Ypv`9;%HW>UlQG1lIb32IW# zSZJlmfn!KH`NF;2&&vTod~f^NHHip`1 z2sT;iBRbXRAvleC-GxgsQ(<;7dACn$$lUm{C~Q5d_rrjcyzb1h7HpZ4#(KMZ{`qS$ zE?fOPw-J8muDV4+k5=&qd4PN8RB54fWD&Hc0O*H$B>Q^WYVN1+g!=GLlIoJb-ra9J zcGp^ut7Ul+#OFR@KD{~~>dlx;g*#Pn_Wt=dd^VJFtiCF*>HWB)cS4UWHjKwl*4=Fa z68KH$FuN#!w`H3-eQy0|{E^JfJv8-!AaJ7h%i$QblxITBmWiFCG@5woT=JTsjYPF_ zn|Az-=NYyDaP6(Pa!ZGA!;|F>oA#h!`-@r}Y zo6@Is{B+#t$W*sNoucAy4Lq>1N5P?i?^pA6<<9Nb8Sm%(gObyzlIpEac!dIokOR}j zqjCR+9BeaHFmo14CBs4owYxRfK${$0ucX1Nu<0u(G4tL6-KnK+0WuCAoG$%%Gb+5l zea>ee9M8THsDGgnjjk%J5lH{%H2IhTd37g8Vzo zAs6Eq7)ndxi`l}sETQD&^9tjIAZ|Wyyo>$Pfy08Ju`Isq06peZy$+fUTILN8I*#j1 z>fpfe-sYNkCS3xxyEtr{V@bi~GZ3baId<(YUG4x%Cb5^Twg1*J>x+4oBolBNu9_#L z!e_~~q{5wQnHygh&TH5ZlkEyNOuZV^_COpi)Z5VjN(7%F#Eo8iSp50=$T0{PZn^o? z2_?%6Byx-pk5hcKHF1}%#%uZp?*(O z_)37!;ez%C;N`8j#&+h9lX4=8>+N!Sc}~D`IMb&19YFNHnuc;7n1V3ct1` zH0g)ae9gkOO`T_=?-WHd2<|`3*13H^ih1eYoKkU9ahntc{WcoIWy#S$y*X=` zkj2O>wI|Zov(P!I;>2pLDuCN7qrEoNYG}eYrr;v=ou3-<1`gv70W34E#^(EeSBcKZ zzuP3CA>$)1TuW`nHa}umbTN1aGkOtGvrI3k2@vvkJeEi!vnZ#-rgbw(F7T*!YVLmd z4mVOt^rUt81~c~~T5S_@nfi$A5_ig*Y_FJuh`VX>;*tca<>I+`_ioS?G6!D|iKXjA zd~=+0C^|rQ+&2QzV9#-V zjiDuSzQl8_44VAaegT|?bvcd&IOUIb>9lgp07VO5I_K*zX&w$%%+{SS3Iqhrciv*` z6W@mQ0t8c#A-62<@>iJ110BQ!_fC&&85<8_KZrBoj)A(|RuF zO15F^)sx(JS7FML&0#nHZ4cZ_T&YA%<}aMSHy!}qij27ZYPgP6h^YY5Psn!$KZJx8 zMsD~p@Jo^yz|p`Ow&C9sadxHpMqkWRF-@$;PWBkm4A7aA^EsE$yPdqv5B+`pYT)-5KRu&h3mdgO@E*| z^!L&qF(w<~{YdW)r?pOE{}fJB(MGGyXB93*_bnIRbmh%-^c#an7xqB8@rr4`F9uEJ)A3KYW0F5wnjEf{9IyG|Af0dM>mD2NV}4l zURzo}e+UHeGL@V10y-YtI_9)CQH>;UVpm##O}qoQ(&RLBZ%J zJWIaJqa>2rEDbQ|PyfWp-4`mAfUY34`ftQ(g3%!3^4oEk@8QmgW-RwbIlw!{?um=x z49D5v#r_PCm*cG5hA}e0E$C;dH}9}n&4!xe!|!q3zUKXas;s3UvHNvCipsYJP-o~U za;=5&X3(dGOziXuJ_p>mx$$-Sko#;?dk|Ea#pd<-hSKRFSl#u>82$3haNtzSb`{{A zk5@X!47iQWl?c!p7}11*Q}61TBArZxwRZ+JsEu#DOgqw?w!E6hTMY$g95~QrE-nwF zoTbYNHiMV2?v~;*nL`UFT9)|1tmB&1&314->)c`B=pMs!orp#|9?1wA-uZ@Q9*#z?G z`##1ufMt%P4oVc>i`Fvlo&_`HXKcm_qxvT)N@Xn@o2ARrql< zALue;fx$y=uL|wM7x7}hKAp~jNog~_sb7H+H!Q6(k zd00O8yVR<^N9+-=g+M9W6KY!`8x9oqJDN(oVcp+_R%bl~PY(H;GN3cafSpEHmDi7X z4D@@juy{WnJXQ%Xi0l(P>KxD;qjE~aH$g1}3>PrX%_6=0K9ja4G6}*MX+5R3-fOiM zf6Dh#6HSMh>`_Ez?8|JFn&?S&JAef;VdRb^3rBaW$fN4X@)p zyW*@#xgS#(8#VH1`HlIRNAP^jfa0{B7!(5%b zVnjJok3jbTobQ?~KY?p9*%quppJ_y&P+w%MPBs^D}QfUQbAMa0#NCbIRy#jskk71}}V0|P*$EB$`$ zPFPc`DJhmdw;wNx)(rLC{!MI&O=~VyC2AzCraBMmo|v#*$5ngw(fM-+w6oJM?}XRt za8RsQr&|^uyO4Ax)R1s(Ev@3R_*&F=uDJGyUhFCFXG=rFYGaqS` zSzNS$Z-qBw@-w~BT3caAP>tO0)Kng7v1PbgzRS-KFINm275;iIZhY=qmN}EOWzH~| z5SSrk`AKRt5KVDj0R-C{Zp!rR=E*L))ehW1R!MvYPW8fkNks+JKPOVl=yIo%1nRK{ z^GDCFiO=2tC(y2v7GPrN-M2n?!d5dFC@p`5p@J5JQ|sd0J_RMS%iXq#MQ-zolF8Y) zJX8m}9DK%*<38(3{S8bqBo0E%{+`wJ2fwE!dqZSI>Meva?Rcp2_w|@>k2g%aewwe# z^r)jFs=Hn5L4W=5ON+XQ>?e76>A=bV?h|TNF2u7fp_Ttx0x>&AyW`Huy5CJ1*KCnb zfv|G;&>DO(0sXwc@p3$yKi}T1@!V(>my}~K zc5aDL=xIXI>C(y4+9rauQg_hrwr4!H+@XOImoXZ&A#KsL0II4g+Y5s9xPVtrB zcrLh@!Me(E7Li>2vQmzyuZU|x&o`*H2<*^%o$3>)(N;4`sw4kh9j&pOWOud!h zIePtn22^lN2xSe6_d~pVAimqx038!IqM|O4gH1<9hL{^sTi=qz#_YYMTpzCYOn@dY z=@-g6m0WhZ-xQ|zKldY@PL6v_=l)CgS3`y|ZXKKLN9iVns6N5Xe!cUV0A^UHz*GK6 zm;=(3Ip=b>podJou`CT#ASucs;N4i)6>*p?r##f`c1<4+xU;gAmZp5DWxt&90qskfltBx2kY+G zYj=R$wN9<4TRNpvbEc*BcjHAxrr;9$5s?8OhiZXZLJ|QN26ssT60TQO|N0@vIVwGg zR=HnUiwr+RkaGAsJSzWuzc{BcUlxkB$~RYK=7{-!vjsK^6@9Ju{Aw4U;4o0;{djcO z7W+|S5W>~#))F|d!KU1^{d7E7UsGNVF2gM^-|%x^rW(d_@)MMenW< z@H5xWV-FnP9E*n`&28|~qmV(&?lunUzM zQ@C%!JzYpZ>zYOj(h=Tb{VI^pIDaCV+#(^;amAR-QDvhzF8!(&>B7;c%0;>u`Z&#R z6ixmih+WGMjRp=AU9`D35ME4ay06|5Joa7q7JKsrRK+p4@{S+!&wN^v#Cv%E!ARTd z*4sqCMI0J>B+jNXh<0;LCiODPRbZzR!zwhL9Dlx5tXx{lc zO}6FHVi#%cf5`@r%1F$?3;b`DfSlX)U#UM!!He{__QXob`VqYzA+EkPO<-7j<&Frn z^avfZ6I+|J_KZvJ?@ZU#I((8X;=;KCHjEt_s=g#HLQMfItfVk$a;Kfl6jEb9Nanj7 z(E%a8GEsMVkmez6JHzU=sfny)tXgU6X!&DCxnPU)2hD)(v21vqn0u+2uKHd`f6--x zvGZT5cPD+}ed{Lx_tQC~6|zljzBsRhXBjAYVE8S%W;#afpi`Eu`ObllTi-w=t55+jDc1s-#!K8%J?3xXwZ$K8x*K?XH zSCRjAYNr1S@v>VcEKb7;Kzn{)jT?Qz1s;1VLph)mq%ce}j5M1z6*2L5qI+d6#XJwv z*x=E8-x@s!AG6`u1>anP5ZpE|QhI)rXjy^{xJOwui|yJZ!DI=$Sz=sN2sFKrmcW4S z>UU|Lbbgvsy9vFzM!V5`)H~|eDs5!upnxVZ$J<7S;apK*2do_wLm!VW7jlf^o-K!D zzx)coPs~@(r$(0Sdwjpzs6H+7fY?*=6PG(oUMDC&lO}ZCv5Zim|9J z{|t`>b`4rKZe3m{n8zS7)a05cV$oiI*}+#^;zFDeJpz1Y%w1bjj0jz3 zyH{jR&)&?jz&|^C8(e(|wJH!iyJ8~aMt|_kPEW7o?5&99`h#~!M`n3UF1l->*R!W!Fce7S`Hoe24=KX5n zf-O)}2B_`hVg|WPH`ToJ+OZCqPpK$L?c~}jNL~t^QMEJ26X#q#M?GjN(p)RE;Oqko zjxmON|38{-B*{&WfB ztYLZ?2|?rCG-v5PF^(*G&Wnp#KvDZnEB4&k`~&cRBu$t%y^Zra<8jJL#{7TZu-DqN zt@MiPCMXSwN-d_pNIE<=zRv}k$Ik54zgPG+XnwW5Z z{(N;78kkCzXV(2h*{y#lf?=_1tnJ(!h;HIE=W^AeVZXIlOUPEFnJJ15QkOYUe%tU6 zT4JG27rUg?$cDm@9$A{*qUnFXZ-c0$5n{w0b&89b_>tg7a%O3PV={{8EPm{%KbHAk zf5M>cJWo&|YK3KQXbc_>{uw-Bp)D5a$h>bht8Ta2 zx?wJ!f;GOHfCP;2eIsfc;OPk4aF{C9x2uRZ@Gn$TsQUXyd~x*W9%B7%lt<8OeJ|B zfhfDuu30o_WIcOX&33}{`SnxUU(3uSjPtH+ghd!3fRLZ)Xkg{#R#S*RgWsS>@;1Dq zlJY)dCSu6N$ZA*nC%`rv-4IN;chQt`R#j-|sq}9QdR zldz&q_<8YjJKo$NM66|};}2DWxb%AZ6^1g6-`G3QB&P3Pr;c^}wQD^=B)H+&q; zAs4v{?w%}U#Sk=%9WbJ_oZfPR6gSND-0bM?czrF^2dTs4n*6Nx%jnUn)rGnrVduwD zliF^_WWBskDf+oH7y135%Q<-uZD8r!5|KSw2nbv*@T%`La5EeVx*Z}rmW#XQ=4nv@erR+4*1`oYG-#F9XF=DmVED3viNTGj#k|;kT&I6fLRs7GbBcxjx>! zedd5Z-c7{Bd>I)ub)!4}J-tUTBEOFMAr6=S{q)#P9nKuGR7~J%r!UUiAr6i zlDfGiJVw{KK_;a0g&IEX%W#9ooS0qcC0q^O6LfFs;FBmu*;U-Ub?O%zNtzG#Kh4M0X5sWN32GH+I` z9M7AUwB{4mZZZx=KJ$7A$tcqt=Cz~9zY%Owu*Fa%>h0K=RIPajl-e}LxZSyi4SCsX z8{J9_z9zHYEN@!-ky8+&#E1wAG*MIh>%UV{b{VlB80016UgbW9Ii_aiefo-L&ppXuF!)p^(X z38r+MpRm496UdS;;r2+jxm`@&s&m~pL%&X#f$0BZix#B;gy4rlgaM{4q3LYR=I{|nL8uGq4t%E!9En*f!b*r8gS;!^SsYpXRa z&8w=YM-V$XsL?hI?IFv)KWOoeh-t z&4fbDfn~f!mUPIMznqRY3N8;Z%Y2u6JD*C6#9a5;mz=D7zf!kQnLnqRWo8yD?vUO^ z8Z7&*$E}I(!j`IYg#XS%#Z^?UQN?VM*odU_AuPZAulKlX#h=f$SOYQUr`AKx*b+tI zl)v-UZ1$LOpdMm5{!FwkNBqmLx9oumE?E*zX@8#A;J-}6NQbxgt!s{CZDmo28G+IU zo-QOEyWPc>tzp@vutR^XI>h&#gyi!_C6EwFk7NDB4?OSIamD5!UrIvWy5kYj_)6cp z6G-=AhsA3?vo2s}{)PFt)og@KIC{SoWoIy7mnj`%sGlOP0qxmqa0bn{Rfub1+VV=o z>y=O$b)+z+A$H-GC8MCLOYD?upB+y_OKrMRJ78#^X?^>*=YU#t6bpWrnT{1IYL^5NLU^YqJ|r-Xs# z>@kYlOO1k0US!#5-V>kqaO3Kf^HvLM2&(Wm9<6~zQ86GfN zj`B#DZwbR34}P8b+iKIuu|hOagF}{g5wmMmb~)ZgGJ8Y1SEPTTzXlNyxiN2FqaXYj z$iH(c#wS3wnHcKvY7Vn$$z}#DYtpXAc8S#?$<0h(6Hjx)tMMtHZ~>PNSRK2lf0*`% zr(Z8Yt6%!AQ|cgo-gFmh2D#0*$TtQ_x%Q^{n`a4>px=y-D?H$m_-S=Zy3$CHM3Nkq zE~5`_I)H;km_Aeb{Szj?c22Ybe#pChv*}8Cq({2b=yG1&)HYC^1rfiE;#41uqG52L6NiuB>LB6)zg{TUe_ z*5IDdBJG5(E@OeTC!2f6D6sa88li2I%)S5d)ixxdR6V)~5<3|Cx9V@k$am0U5m5b^ zxZX1-h}u~`n`z#9;K(a+kPF57$2hHQB!NoRcV6}?ceuF0)4pjs>WLq~X-5W+b{5U8 zi(P8Fmlj4o?JoLwruT2;2VNuUJ$`zAS~fSTR3LAm^}W{Xa$gO7cM_dLdB0n_c4yj7 zDZTNnQTAg3btbCgjn>>6+MFySLo*JcvnFp#UF7K19FwO4ml5rK$I z_brL=VtW&*wLti_iaMSxg3RqOKc5U`w%68EmYLV2(cU_`IeSt`?pPVAU_&0KZPUK( zRlN=HUdL@WCG_;iTNJa!Yz7*)I8|r(B;`%xW3=qjd$2JX=OaHF{3Xv zXfvH)F%6n6nDcYN5qQ6PbnhD(IK}@La$|aj#WVjpz@RJD&0ij!udndZ-2392@s&o;E7UD+ z*7Zo<7?P%W)>1}HTxp!PlU@;_jT!vf>CZoC0r`}TQSXc-b3KOIBbg#XWBk9u*)e%T$_BwUTNINRqU+5dgjHKELPVuYblnByf& zt~mrWPBB5}*EA1$z}n>6zLUH`zZyJZnNyEUwWmZ3qruK$watLCyVJ+aGDaQO*XiPD z%70!n5oq~Uc>R2Te+1mEOO4CuwG0-8&~x|13R%94;e%uAi0E?Nwe~D+ydh~@RUhr1 zh&$Vd&Ej`+{`TK0me$}wG(8&<2CVJTWtE`xJ_+0y8-vF6fVJG+opf{K_G9%hUwS^^ zNWvL@D~;E^s|sT%t{L4*^HR-ov!mU7AUWN1Dan_(0}#`-*gPND8dJ+SmbSTK?8O4V z!5Pd#Q-CTC5J1zTg7d1W2qge^*xs9mv`DhSg zMSsWn;d4j(?U-NLO_!L-Xcomb{_v1&+*MWhotUb2$9HG?`sH*;0Pz2oS;9+?Ncj4YEztyo|l_D-pskLk0Sqfz$HB? zzA*)Z%)e_tR?v2po?8Z*=C0q$7yvM3g$ME9rFWDkoh6==DRqLlr`1TfE`x2q^*PTk ztF|sLyQfZH+B2%#kTIw%N?KU{Edh8EPr8`nyBt^-_!5Q~@P|ey*D8&FAl) z9osJkE}EdGcKn{9!Rz#aV4eEC`Z>2n1B~JJ;-bJff^ch`3e~}Q@haNq$81!)D^r64 z_Z-L-RxB*ZzXV9WXbgZ}@+-dhL7^=)axArK&V@DMDxO9&9$ zgG+EHK!D)xG!h&Fgy8P(?hQeMg#f{&ad&7MXyiTo?wy(YzTcghJO6xtO;uMHMc1jb z&pvyvwf0)idYwiJWwjPLQw*^gGr7O+xRlnqkxZtixiX+dxvdR~`xIt79}!?fZz-qx$6 z1xQg@bt;Gt#H97il1s?J-$ECJF_Hy!TO;^K(Z{%nGK?&bUE~^UC*o5~X2WW%gt!^O zfhH?xM`u{#TKJGTf8aGYfE`pH6XS@pI!IX$~&2Ih`b+C>o%fp*iMM2^#^2siq} zg}K(-?}k5s44t1d?*R-V#fomhb#s1Fc~^#wJ_S1`a_LinOQ04VBIL@XxVG z(1muhG^fDmLEL2GCVThAE~IFvo-f>!Y6<=F;u2zRu$(MnfX}M>JCkPFzFlv1E=;!i z(t4|j+zx1=>m*y8FVP7HcF@mB@`nXM>}l5nbD5AX)w7kpvYn3S?{siO_-IvaWn$h3He}8U}FS4z?GJjAp9?EaTZMzJsfhbLXYy3jgaLd?4omj=l>_;a-#@yq6{LtCIK$AJ95S0_thDJmPbg_{x7hX$Hqium zK_)N|+xm+2dL-ErEv0qXEthe>cTk^#Z{)?Of88#| za^pjtUhP1}aG2Y-cc$s%bqWZ6%aZ%L8a33cIZ=aIGt4uo!nE!3sKP`8NU&0ELDel8 zLEc**Z;0oN~xv(?b{Ur-ejT4RCdQ)*FlHygtF` zzy{x{a8|Uip0a=eIa5)UfX%3{%pEYxPqn>EQ{ehDW39k6D1 z-tUg2zs2`lUsmG4v8w}FApPugK7^>OyKjiu9$jUkOM^S!Naen*FLw&YUQX3DYk7h> zV7h{dCPcqa+9al4)#`eHI^+30kg$DmN&Dsmjg3>8NcQ|R6qAL z@O%YrleIj@Sn6sO3}caJI&bGPAJsZlS-ok9Ykx5Q5Nfd)3U5Sfp5UgP_GPYVUgtOc zK6-_9^L*WK%Ay_RdSzYueHKGI@v0)E{6&}*!igS7f- z=!C9bplKE%Loi1Qnj9(6p=-q8wecvrm+u}tojyQ_x)nAa1T3h-0R^+47$UbU3Al~jSOTZJL{6S0|<8UoYshYg{xQk7y%d{EI_v4iU*FM z;1f-PAYh@Ri+@A7s-+tP- z)W;ViaqNW*V7Ox0=`a8>7^JvvPQFEaA3W-yKt9>g#i|+OsTdX7Fjh;sza9M2JuZn@ zbv}S#7xBi_nYM&P_uVWphSIuCZUnYjB!_=DViH`N-~TEN;sT{X_rZ6Nq}%RYTbNP^2P2YDVh#L1QDw?W%g{a zAZtZ^*;IMSG^dW6L#OZBdXFRCb2-7wWCo*>@3o{4a{c$2JdU|e2+=M$ZEJ8vOUDkM zLEMI8tP=H5T#{kDfta%94dT$?V0IdYKHh#7I`C!c?ENx;U!2PS(kchtd&)Mde0LH! z+iaI%7JMgjf14{`)hz3IA_o&c-&%9$J3`i4cNoo-XUaJ}3QAtBrh3?#<L) zNcq#cK*4p{qITd4cXSBHn%Pn5rp~+>bz*t;#=IBxrISmjuKoFh{MzdhjQ*>B=o!Cw zruDqKBDttXZ4$pbVbii(TC&Nb{^u+P7~;kkVL7ELx@r>w^(!4PGtxxggSbaPYe(RQvSE zs)3K84<_@H-ubq-`HI|DujyMv{X2BF^TjP2N1G<>ulr&`47!ko_6fm1x-&=UFN%DA zv|CQQIZH7cJi@au$QEbjH=(Ed4P!&1mF0$QV)b_Li20y5sdB57#L3%Ex-eBAY9aIW zK(2Ue;y)suXDFLTBz9Gbr>Z`tlojTO@2x0Fjw&ENph6N!u9yKz41gEUVZAWT4u63L zT@3*rDN-U-n6P1wM{~2fNzH(;-lG0_b1)mAxR)$!_O;*%%v;gbFg8&)Cz$xM`qrmF zp1pvUi#_zJ$cO@wje5i zuNjk2Q5ruVhVmRXtt8hAs40ZU=Fu~_$M+yZ8%(6>%;8bYo>xGbJGlgKz-bXLFqXw@ z4xtJG%<2pXUEmrQd#pYyfb5|!m)AVCPi-FnU4#$bbq7N>LLCqTIG$ah>T}qaic;h} zTUB;AmSD8dvf6)pLWdy5fW$U5F8vt@2|HhyJUeE+`pngL;LTC;eCQ>EWj(-QVn_aI z#j8UsFVX7BvaAkQRZ7}bdEyXs^l`Vv5zMNqKG%%JEh zR$W}^-LA=Nm!v%nLBqZn>Of^yl(Mqb;5Xqbpt(>S@-fTc6jPa6<*TVb!ZAj2ubEQi z4{0VANW&VbCsM`L`^@?M=hI8=Y&*VamIudO9+2cWw$WDB)><93N3C33--E+1{f%R= z52qjcz1V{?7B@rr#Ha!4hTe-3Q!;{t`DOm4QM%H>f_xjNGKX>vMS7k z*|QqFPQQ^FeIkE>j=JkFJ29L%5(BgI52xE?no_|_9dF8xIc08TeSHU@;(i8noD3Sr zo`pAvh|XD0Thd@r+)FAxk(xWqAzd-f1BJ3ZyD(9z{8KG2<)>f~8zdn8cq?QJhh@lno434>wNgWCk+TC>j(y-rFI``kl` zjBu_9LwhG4v2Vc9RPJiJ`h3&=FX{$#Boly5kiOqtoVa`b)zGnaZw!#M8nyR(>!q&}Vp7U!-1q?{JpC{E3y5 z{K>Fh^F_!k0i+tqPpb7OguB5YJH3K_{he}>+$R~82>Xo!uzi8ag~UsA-QlSjed_!8 z%&7XVFRh{Kdl26NGOxG@OkvkU0Um9yo;88*A62^wq5B%OSQBR* z4&%9*q2=1G>g}BNnR1h&N;t*7a|t&AAb9Ud0Af#cg>GKR=Rfr&^E`flZlTllS@m6E zAAp4xB9wB_I1NO{Ts9R5$YTIn;?2>X>GMFvg@dE2 zV6wFYUkCat*W)N=)$0{oGZYay5YOAKvMK%tzcM?Jm1EJbS9duaH=?c5&iUF!X*h@* zHt6-C&ocvdBOV|IwXW)TuPEyOW;Q-sJNZZx)?$v|45y3pcxakYLL;SxKTtiocdwWqBayjsvwkw6?7?I5G_;yoKBS_lT^#0Os z?XZ?U_2_f5@<(_5yM8N1t5`bnfeYlB)X?m>MEn;__AT9-wW%!1G3|wqCd}Yfx;mwf z!nHVM5eY}LgG>Ef(1NC1fKRB4b-1<9#!n)Xp^LdaLopJJuAaMgK3RYzK`7eE9he8USD%9}KPDc_keyU011XMWf( z_i`^^u=G@2V#;uQ40<>vmHi;S@PxVDNFiBJf|8u$CIsLe48>gAw?^H!R%jK^nrxG% zkCkYIJ5RJa+=Yyq%zyPPJ$^#p+Zx*DspeptZ4*4K*M8i!-@86B&kv#Y>;)*5*n7ln zRPhKn!mzYv5Nix;Z`<1*#e7`J=Ca`WWp!o#{`UqV$e{B7lJM-6d&>(#0rwT z;Dg=^?t|Ke>+vRw!`{)JN*#7g;GFDo_N*7Wz6pn0JLBw|4^_|f8}hJlfE+?bj!$Fc zjI%=Mg8vr;Hh3fvYkB!o$BM?|MOLp1FAbqOR}ot+pLPAhZ7hZtk%QJD9;=X{d+Y5Z zAtoDSAa=&ZqT$4coM;BNe*E%WKek86xd6@VSthhOQKel%EJzGTaC;!;W$Anf)3;kh zTEfLX2@V}>*qa+ZeQO4ulV9ljO7*O5M0i#Ayougo1a#dL@!P?TWEJ7hKIgew&8Wx? zx4nV9R{H6U3c4Da+GlNopIp6FR5}TqYGVW{qLVFuS!^JEcyqf2YH|4SwOW>!f3%jM zY-3~bXs>7~qs6x{a8g6-PF`5L3a=;hqDh>;qscbOs_58up(L7j-EjZ5M|KYXEBBlBDbGSXk7S~kPf)V5^AGsm8(kWcnQ4Z7u^+d9>RCLS`h<<)v|=- zzFHTpl7sC6vGuEkUl~Ho=CbHp*!fS1)>C&2y5TwPo_%WHG0&#tmM94z@7ViUbG|_u zJ?mebtfs1C=dZqN z`k|LWk4*Dok{r+j$9N*X8;(GQx~6R6;5Rto{xtG5%Evey(maPQCl(CB!61CN-lo{Q zSq*l)Pan*^lQ-zYF7j0;Cs`>GLuYA7`?;R8IQ`UYI!5(FV3JLHbNhT}sA~Dg&dDwY z?uE`ZFuoktN=nSpONjTGv21?_~m;hU%rC#JY}% z)S5r|6^IW1;HUdu{{Y;Vs8=u}+!wK9P=tq#ViVr+u%Fvf4eJ_Z6*Dv=4a? zXp;OSOHsZ*E4X@Z$lkH2$dVMNT;G)-h}^$~1Y};atO61idpwsc{G=D`UqAG2*@3k- z#*(zWpvlIdk=L5$J^idb^C&EJ%EQ9)O_qY_0Ofc|JI!Zu?bI2ZG2iC5p~;|_dCG@b z`XvA*kXw;EntNy^)@gZptuoK9^LAjofqN(%i~{<`c!C?~&Vb6&Dj-|t;Bz2xA_zd= z1@n&o02n*jb!o3g^bAdQ#lEKHtp+5&UOnE9f#lteOw*Nf{rx0Xw&EV!2_cdj+MkNN z<~XdGwp1Z8q(uOt8ax#;?6-1?Kse5Ecsq94qI2t{mc#xs--BG!p`#5-noTFaYkFGV_ayjQIKmoRigY==!6QKg zFtpngmf@GYlKrY73->arB_=wX3BBp>3Sv%6Lm%NYSjijJ-UD9NT0m2x;h}ogNym5n-(|C;qP^T{`pCZ6bC#%b`t0U{v(UKEMfh;haj@^v*QPqJi4 z(zG6|IDVp534x}EomK<)H!J2}b7iWMkr z6C}P(yce}VI;QauL8NH=_`M5b*Vq`~zkE8~Ye%ksFp$#OV3EJrLawrmG6LG-pI2`I zctq?w1uAdyqxKVnp|DwqzLI9}8$Net-u4`6J zCYOg&wz8D4uFCs%>W@y0(0sMg(&7*nhXPtN6Nh$~H=9p0+#$A6uFVC?5r-=%5|qtcO`UqCU!h+y zbb|K7toM4~&F_=TFSQSmG8&OaK1aCHY}05>oH<&eCQBlQOUx`&qZHKAyq~tDQf2}p zwvMR#2mlhXRyihStnN0%G6A#!Nh>))0ZQ6CU2hc8ktO(DZOY2Ka=OrUPovH$ePJqm zVW9y?LNd|6+GJLrntUZ)1gQQanFx6%3f}`|3k-+$q^l1;X{TyS@mv5Pf&t|$mDAc# zu#X&nl)Vw86wjHnRYp1QOwDlQxtaNGYV`xfJmg`@7 zut-Pht50Wk3wlS;5i1;Aqkccrb--9laofGfC4ToQo}}8nmc<3A_^R>S z%owp7w(Q+R%b$i#0w z;?c(aAu7o~pa4dK3R9F?h-RKA89%~49vk%bMz=R=8jqx!?29Z(i_@k$p-)AY{p!1q zyP=543OOEM8ZhxAhLhSW%MBc{ns~l|bh9a63HX+tk>oZRhsPjm2RFpVk>il}$l~VR z@n;Mu3w=>177fQV<3S1>fiFWy&84i^^DM(X^1gC5_{F$;ky0pj7L29MEk(qCoUzoI7qSC?`< zL!;C-f*mgijpQyVtTzH{GK7i{ZQ!xk>(w(@Q0_f`>@AxQCMUdQmqXXlUfp+R+X$Wg zkbH5uniIo>@2_ra`cYY}>cD?T=Ds2jh z)h2LdsiD!O>9+PFHCHI4_wX9dJ|w6Nrl+()xRF;o87ckHQCoj&PgaE;{pSI|vLtWA z>adzdxHj*?ys-$e3Po!|;LnT+7ngadrX;qv3$Oa`wZUXm4Qd+{`D?F^4LvB1=!EcA z$=;)9#+;~}0PJCFir+*@MAA)my$?XyY)LJE%_%$&wzd@d@Y8;luzfu$3G|l5aP&D9 zao8}M5PE&=^uSc6hkjN!>&4JFku94ECuH?U#pe=0LL@&H6Zz@tFG9jnVw+#7+>xs9UdrOy z5};*Q#ek?$Cd4On_kMWD4N3|)z8QMgz~LVZQT?i0eu$fhr|~So>4z;^KlrWz_(|~PRHo9=q$8&RK!7{ z3H>iiWW-yJ!|?}P!BZj3=+S$GlfC#Ll|4Q zOuXGxZ$wirH?;;nb%Inc=TbDtl?bivX3#X#2f){w0Mo5`a`v~)W$(2X9(Mu4h12q` zx_vaufROEAHApc0sU@3?yZO%g%Dhm)b2@%9{@a>TBPG7qPuO|_aeJRuLQZrN$kyg1 zWaE>+Yy;_y)CZr-bT9KRW;8m3s~7c6>yRuJwt2^j(~&@db#2-ln0WnmPu)?xk5<>6 z)1m83sqcKfN(>Jqh!m#mE8vkZY5V3Ci;74bp6H2P92Cs*!ob=)qi2!G8e+PYkUo>c zai<9;${t6v=ZS*=a}5R8!pL4R;vnDZH9Ktg@tWdL-t*UP*p-aabG(7W(aqb&emUX-b0 z^6ty?A;)DrH5;Mtjz3DSl=_K6rO^HTFM3m6XW^(`D~4x2A%Kk6p$mMEIOZU|{Wpr~KXv=nT*;mWW)1&?{qeWY_hewqvKWH> zlVtZhc;f)g$Y##d)3H_688(NPMh)*l1oT9Pn#&9IZ*)9ss*{pB)z}f3z-q@0n6>7~ zBV0g(=u30cMzr8Rd3pmLudy7!vTEiL#?<|#YS=ITNCTX8Tt7`{WDPA#DrKL>)BjDO zM~{tS+%!=vkq0tzU;!b)X0mnTQcFMwgT6klNNm;i-A9GSK?foz++Av(49tzuzl zq}cy1@+A|8RD7kRlLVSM>d1AR+*TJnEUNl;y~Y8-*E3dG!3a7TxZdyT&`cE66_>r*`O%zs%SYAvPt`w-W6){lf}%3OFmOy zl&=YMv7)8a{>jS^;H;xj#P({yv>=_2v_Sg&%xv2Hbnj``Ma|neElrN^CtWwM|3fs8 z5*0Z5pono=o@;j+NAAqF6p5B=^0shFAah#vihd|`k!_+`RC*e&O`&EQ9h9HfLmnSm+A!8+D ziqr}tp;XEgq?Jo#V9K+-qeH*Op1gDMO^-4ym^9=B6I{ zG50wu-G@5bOA0rfx7qOlpaNdgr*=N60aGPzBAOH}r0GSy#v>ebB6b4QeTSQimz!sT z(w8=+M$DS^--0mfD}#( zZnOB_aV6LwsB(~{mq3etMzwIq>BA}6%4Xmh92S{$|6MYr-jB!Ao1 z|H?N_km#14y`qfMSpDw3^RDVoE-9*`KdHBpg3s#c{&ENZsr=kO{3;WIq#Cs*@Uqq{ z_TL@xfBT0jae&{SHG$w76UYDX8~^6U|Ff&VzXrT@+^UrE|KLWqc-dQ@(tRZC)DU6aG!B`@ehk z$BVl}0{qlZL2DP5542`-|KZC1o8bkhOHcwbRLGh{z)@df@az9Ij{ow$x41H{5yzCw z8$?u-|GM}ojnH2*utJWt)mK=x;yxLB_WwZC^nYK>|6eZ#(1Gm(bS5F}hPV%R5F>=GJ@zs`1JB^* zqxrJl1}&q_<0pmx>LLtK7YAOoKO0T_@;3YJOBK6ui@5unvXa|qa*CLL(J0CU081Zh zCpUu+1{X2ewPk-4Y+d%q0CD~If~|k|Ljd0N5z(D=`6mS(VH4k*C*_SHg*V+uz`wdP zR8fINj@YW=)(104vBO~4{y9e~;>yPIf&gOt?g|K)|4T#m_g^U!0B<0WjjjL;g)#;6 z9|Gq8QUsdj8xVv(N7_nEkFe)`GidG2fRDR7Q6O&^Ecek;Re}sx+l)@R07=zLPic2k zxnMM6>hbLaNYZ!Ri;TZGNBZZ%F%ALnM0wqW583tQSHj+x>9h5oXQ=leJ@ z77X8a(7f52b&+ zxNOj(M9Zvw^1|_0=pF$5XB-dnoaHsIMY!a?hcY>NO69BRz@atCuT=BAG#ob*?qeF> zaB5E5&rMv2 zYkx{6sniPfq=2j%o;pW+)2b(EQt_qa+?$wd{>w?~tVm_yYlVUe-Gbcat<TG6R9e$`>hI5Ib7y>sxsfpmwwesED#k4Pe=pgM78Hd>mv1HJZZZ zz*nPl-RfJ{qtx>*0GO4i3X3EMi(Q)|OZ&FJ{BvEzZnHMA&0MR&TkG*@e!rUB;NCBu zljV?D&HMALhcK!><}l~ZTN7HLYKutH5%Bm;Mm%#(b(s1uE(>46XOW2-gYNhllV8$76G@<%=3+Q#+zCA%wiYxvLBR0wv%~GD~Wkb%CCT9`+jn} zPs@bKpA)S0oMO$kbG+c4z9^&r_XuFu+(zO3Hc%?_t&*(=gj-)irnLe4SxpKo~8=YLs= zZcz4ey+vH|tH`wodFP*mx0V2F7_E2lx|cLz->)+#C_ep9UM@ifUO z8UM=<9~l70Efv=fHBh~Q92C-^m$K3qeyIYOPuJShk$#t3N{m-3mU$!q&6MFumbmuK z9;evj1>Qyha5SgGhMBMfpaVp}Zj8n^F;h_Z;kG#ScMwRYcJWaxne}Ka7E`2$2Dhxb z_rtwRgtMeh{U>*AlVu`1n2U)7YH0d3SlxV${p6WmJJL7<=Y4PZ2pW!`K;Q7ua_mjNTcZW)Eh}2}Mp6CPXe= zgSYNi;sKJ$ZJ*D@(NBkl{qzF73F?8Y*t5k9L%+LD0+f6p=O8nhu?KDp(e z>zRx0Q9e4u8O{o3a6pYqi`811eWw~{+Dv=fJ1U*Ys11SPWgNf`bF6X3;!3LeNb?I% z+-t{!W6zWr;ikqPnZRZ0m6VzlhU{;a+{$K8m*>My^MPs0_b9i~FdMv;oRWKU57iC+ z3baqLZXcQ*M*!`6>H>5CJe^64i9#f{W)l`kB$Y#D_AGI8C=qwEs8=WbrEIRQq7AK% z3DcEoTP{K6MPz(2CBggy-edjaE16I)lw>TkP4wawhr%I+kV&>j?fa?fdf6XLjK5xJ zmf+t~a~?ww033141bDP2)s8@0Gz3)p}(<@Qw+1Xg%`Xne@tn+l|1int4#hM`8zKoZIV00ny5yR;OolBa zrd);y6G?o0DV48blwZ(tspybmziv?S$$qCnHK9TcLy+uEM>YS@8)VXZAjP@#P*pW; z<>h5zZsBgO?FJJPwkg`%+iPlT9k}01r^Z{jyIdNuck+zB;~^XH-uFE}^g2Id9*%0a zW+pfMfypLAo9-j_5(!g|1CJ0)TYp`>iK|Q7OvmNtci6H`J}rN%I|;kYsq`@HZfdVJ z^ghI_vs+5~<-OO1IGkEapYg4hUZYT3rrq-{s%fN^9sIWKuraOe*!7scH3>a+T~8(b z8%WM}`-7iShOT*c^NYdXttvD3x?f)j{rRjS{wH1wG*vovWVm!S@V+h!Gt}TNzO>L~ z*C1Caa8r^>r`DTq_8RhX6V-2f2{tlz$-f7#^0|n8KmPPn37XGw=R?5tr$dWI3g5$d zz0+zmM>Voy5%A?rTIEri{qDw9c(7CaaQwOkpNUzCu(En1B`n)7^Ptt*()reLpqL~n z5FW(~shL>?Id>tLjeKyc{)QZl-+Q6oyf(DISgA%A!`WapEu$TktDx^}zwVUje|4yH1;H z>#g&#P7c4@(@;h|H$Ug@(}6AK@6vQP!O@5FHAm=m33taFa6$gFupWgH)q-3_tVBHM zZgy8Emk|isu=xPKUYxFQ*XWBNzy&Yl!rv!Nc4*~6uQwe!7mY&?M*~pES1JRy)M`%d zBffkaYU_ho7sw|y4h-$iK;|KIUcA2NqfLdDasB47CAU#mk2}qYb1k#s#5qXESb{I5 z26|Hi<#s9rVzRouzRO*jR?)OPFQjbbW0~UV2Sq8IPANhen>gY7+?jCsvK(HtD29@1 z%0#iLx>`JvNaGVa#y)o}3$1}^RXKJFH7#3N&&(*7I(?CE!4p0P{OU|$($zSN=bnx; z1d$I>nqIk?`U;c{VY%-hT2?-vhvmhHIPNn3nYoA_8tce`kVMR1Y-$;`KxPwIogWOx z>u#*QjgHKZ2iK!3NS${!nGLyK@!XnY@$WowmbngG;#q=33{k7m6cz6_6=$N~C_Z%D zfy&=X4^U{$k$!?3Qk{d$orSXS8Ump>{!fr7pCF)$K6(1LKLgg}RuZ(dlGT(dzaLLB z@AzH|t01_EUa>u?_h_T^d38Rwy+87@fz8FkuKCDnvOrAS9N`2 z>O8P>YQCCF`@Y+Bmd0n<YTZ!Y{}=%s=yqNQLOMRw8of@mMKV)0)alNv**^f zgPPs)&xmd6-O5|F&EEAdCPH4A9I$1PVUUZ@#VU&8_JUU%M>noam;lX-c?fPt-5M}Zc9hPx*k_@~l#96oza)_K#bt=C)mCJ=N$&~DtT-@K(B^Yr{Ztm>JW?~Hx39WH4vWPO88 zYG?i{^_4T`+t0b1Q8THUta`+bgE1W!XKyC+b(D_I0-9iNleq1^Wi6fe4QZv1?ytt= zd`1D7GwQ#LAa@Qa(kk#+&(FFXPW-ZIn9}525VU^&e$K-!6q{!qtYdyL8@|=D9mMdYw$j>uDkFp?xCHnrH%nPFPe0&@Zr7(JVT?< zHSHJMO{CFw5w1Y!94v++d2?o)TM=d$HJw#%vY$ho*fGCk2Kmluzxyh}&J-Jd=BG~r8M*|*wCX5SmQ6P!nT?Q3AmzEfa{ zqej>QKJSji;e>z$;^%h83USE2#7KH5A(GiGPl55yR9;<6p%nA^EjoMKIoSP1vXif1 zw=d}}w>=oDbOL8d$hiF|kTB)VDd<2~wcSZ&iSbCQkae0JS^7davv+yn;CTH+U@7uu z)oJybo;=MIt0|DEr|_XgUHf86Q#)qfvMAF8d2zYt`P{gzuTG~T{F1_o+~Xuizf+OK zNEc#|$F_9S3VDIMZz04nb-Y)2uagadYZ-W^v@Bxf>$Y7wHz~prC(1o0KW3%WZMS%@ z>{oT3mMy7|nPQR?3syJrW!t1A+v=VGJ6$l#nJk^)|GU#oltdv!Z1(fr&mq{0-daNz zWuW@1OUAIznP&lL+BmCi{wuj{uYI<0S}xA4Moz4ZfAR&+0uB^nGPZ&%t zg>DjMdM+lw7B-eXRQ(`$Pa$r0!%)bva~2YNvRkoWd*~bv85Dj`w@}DWzz?6kGzg4| z$4y|;DE%y~*Y<1kX324YtXyWU@A*7amdsV9%O@#jDgzTQ=*!PoWe#EI6XPga>?<*3 z+`aMaM!(+L_$27LZ@eHFMNZql3B(pwZ}>&yEB_f=?@bLyyRB#^d12?>O6l#~**A!5 z)5KODw#aiNu1%0(>(y#kB$E%&Vje`rd}a1;TmbtKycC3BXn;}o!Bx0gMOST<=k7KA zi12hCPv&5nfoGW+*KwA~a=l>{;(-v2JdF?YpC(BTdlR-8!tm>kD$j(dsC&U zFDL0XaQ*P0mKyoOIo^hWpigJ~6%ISE6CWb7$AVE}n#CD!WdWl>PG#7su8!z=>eaS# z%fA)9is*zXI(g{96y!RfOtuLG={I?v)I@TDD3=j;edG3T@NU7Abiu)Puekj@miA z_;Aa|AlLIt;MhjE{`l$h`X~}Z8*uFx%MUnUS79w~`(0o3eIJ*l(oC0J@8fG&p|8=e zG%jkH)5(f~!!7PtFSeuwaHL|rKH;|BtwcUI4NY^}#B$k(T1HkD~Lmf8j@+&NlA=*4adtfAzI++~KzKRq}|$zXb*V z9v{e|1)PQ2IvvFOXGB~bO4*NnmdEdW0lu4(b&fd6KvW?Q$IbhMz0BK}t`F@^$IR)( zwN@(g%*3nS*mQjOI`~5~IJmu743GZle$PkXNFF85?TyVD>_RGN6CaC`92CH-DM++F zn5ytqu2C_Ebd0|U3=b>js5)cN*fDzU^TzWYI5JAp*UDnV)8c~N7hJXK)|CWEHoKwQ zRVUNb(oyadIJ8GhHf`-c2Xw~lC~Ea{m=dJ5DBR3$J`-a-BrBj|+D}UN+}RTE8EX=& zGuWo|VuXte2Nkw^B4hr%#yLcr6-oKHvM=`*6CHB)v6g5&-T9uj(qrJVA_HdC4k8GF z$4V2maO8}M+%=zzk_*12$95H5^}hO9V}6jqM;%|;QCRXSjFO!ngp;UsoZnWMLZ2kK zF3c3uL=jKB7{aQ?0^o)%7C7gY%9>p5tl4a}$9C?&4epmHwpH@{Igz8k6%mAiiIs0p zxTEN?LKj}y8*Y2)j-F?Tem^K6AY6Z2EK0G{A)7zxHpi{S?PzFlzutDV+gA9CV@Q@M zZU8I@>q(TMk$1DRg7smhr8YqfH%q^ZPE?Bt5mo9kPtW6&@YdT{2n`F@W#0PnpYk&K z$jg-L9Xrt%Gtjj4-o}+_Iz9~wIwLalnR40pY%^W&_}P6p?PJJS5C>=n`nAHcA7-m8 zvbqN&dR%9CoM91rTA$~qZTVY#nNO_;@y<;$iJ0P&To1>`JmDqjncTQg>luEbckd11 z@I%jeREnfz3*u!Z(FA;$(b2Iv!rgx}OR!fgjgo#S6R53H>C#?RDbLU21jQHLzs zZ7gL{4FBm!EjHS74@WK@WY!sH|8$BxLpv`r^%!I{&e08|-OgKAx-4(r^F|9)-%*?6 zDns?8J0&i>#n&2R`g86fg$6`00lAzrq6HVuah2Kd`P88n^RuSpU(m0U%5{PZeHQ|L zm8%HMBv&~MgkEgJ_$^|cjNCYih|U*q`OG7|+JPoNUKu2Chcd6TY+$PxK1#5|%<)SD z5zMbXpY9QVa8F$qeqYyP6xdsE&+lHgc97kUwjFuNt>o%+VK;TR(DcB;d-cK6{iw}C zb%fJ}6tnqv2ZO+|KBql-*a_~D_a*s9@>F6pBRQ2#Xm%?qp%@0Te|5w-3Bvv-VS0k2hzDmI3HF8NC~yRT=oh$3-tyDz`_L6x4wlj)ypx{K(1Shvr2s!HyEPBC<8@T?0Z+Bzr2r&+vc)wip8e&YpYs>`Mp0n3}z$P z=GvL*vp1&{_FblBXPoaBp3qQjMbx`Dem~ajpz*SK2IQZN=zKZ}>Wy~pbnh;lzs|=o z$OT5bOAI%fpT;Q*jPHBG6;MT90Ukaa*08UCkm^0Qv~H@3dgc6^lJdDhgLTFW47Brw zvsCjnL&X9kxz9qdFY-~o7CU$OVz|q2;e@;Mb304WIn-`Ntlb&*Fqs2Wd&eZzT)vzJ zJrETW3eDR!Q?<0NbspWY=fr&S=ZW-j2dOYhc+-^iAIDq^!PnYmP`1{~n`tqbD;FS4 z7$!3CFl>JQd>fjk{P5}ex$63O|J7V2|D%>CC7|Gi$KSC@m()W{x%WhXgkA|W5YIBJd0)ERq&yDiK# zKf+=^27bw_CXePtxqZz0lP?4My%B?W@B2WPt~-l8&kyA&^2G{_hcD0hd|0+#|Kq{^ zFCR@C>;HhYIVKc)WuPu;pMlcd)R}6j^a9L_U<0zY*v_Cv&}<1#(Brl_qluMH5#C`! z$rda>x0WW;TM9@0z`k1j4u>EQ*LG6cG#nSLtlX#b=5yB1cA4VC@F4r=>_e6>*|#g5 z_Vxnn1#7aZ$#qW^sH%`0JZCQtmN^-|KYCX+;Sun(nsBD)4{eHUh&Mc8W&TrySi5P+ zOdlS&$Jia}sQ=+fw_D6;YzgVyO*0d_hbR5bxk^j%NyWF)p?%}J^T%{7KUujl)fzlIQYxtyS zqk|M-It_kOp&s3?%{IB$)>WN-ETEf~X3&5iaSWR4;4j5k0-9!df6$Kz=m+-q5UlXe zXT?cU!S+UUbLxnqY^8p|%v6rzR8q5NxXyS&%Iea9-`hj=J21N5o8(-}YaFIQCDXaa z=lw0iw=Ym1N#^lQjp^EGDrfuDK<&s8k+{2nwp?c9o5Od8_}ZcEJ|=yO+-sMrfgoA; zR}{~vNTR66Iq$s2`1Qu@g3;_CC7}ZPR3Efo(eg|SmQ@b7fmWYwzpJu@^ebFMR&N&| zw|FSq`(cP^%ttx1ecmA~Qtp_<^gmg^h3=koRewf8?zC^6bUB_%&nIc4d_JnoQ{J_*}5=b|$ve5LTf5e|{Gm--xLfmY=ELn>;i@Dq?Q zZ6X?J6EP%UtV*+ihzy0ZRrzA-Pt?t(9dpfj3B%Vpg8!e46as4er>CCb5GmPtI1eB_D zQHoUQolpc+y7UgxJ4g+LAOa#PO_1IM=~Y_j2#C~x^iHINo&*RzlyA9Dd7gdtc3=DZ z`M&S@mkU<1)|z9?G464XGRK>?j0OCbm|E@+6zrI-%Wo|bCDcA>o4Uh6!bT3LckH-p ziTK{kHkRyrKxloM9LV7#+u<=;wwc3=h0Bwzkju~;$Gka0ivIF=*{WQYLVW7ecBg zgjWiqX9;K1y6$CZOP>i+KhgOpjiD15p@dTqAtE_>99QL5*c>{6S!*j`FWNfz4o}M)@tTlZoFx{ zi|k;m`NMDt?SJR19(YI|M} zL=pLs7LuHQHe(nsEA!PYT1akmLOK_@)wf?|qv~2f(-{^x7RrjLb3uDvA)$N;*Ggq& z7Zpi~e;!4qs(jc#vgbRy+^HjNm6WAq95}0^__5Jv%DvM~Yq8&dY|_tLc{l#BBnS`F zf_0c_9Bj6LGKI^TIch|~RCa+Ckndr18bOdJp+m{l3hklCN4CS+MoPmc>#VYa4+9Lx zx3xD^OCd8fo}YZMlnEl&pMvoyt;};X4;%_l`8CP4X3bD<4_q{x&P=rkuZCxBZ(_a7 z5&_lt;ymPf?^O4R)iAmak@HAX)RSOs?x}MTZ$^eaH1O^H`iNsINyp)R;mRL_p)NP( z3Sp+Y{7ceFBktF0avf$hXc_w4oK_GQEpYMRaA4O1&}yYWmp?L%Be0KE5|1 z8)wU0-9KphUFMZs5!bL&I819ip0#fM@z4x7C_pe@mdZ6VlQ*w`I##GGC|-Imf-$wQ z{D{Vhnkq?0*lAQ9G@gU^m=&@iR-|!qg->C3Fq18Ou1B56-vf!|yZ;W7(LIK(V?x<5 z{tr3(&p1rR{a*B=6CQ^wSSfLv&nvV>2u77@PgA>27b#ofh9IQhP(3O_hBBv!gNwx~ zm+hfiecGllGK(c4?Af}|GQchiAttuo88X(WTtoDHo;iHuXi}~DQ5wPp2UYJHB;@H} z0ecS~>`viHcP5#aNP5IE@BsX}yNw|*>DdiTYWoJD_fvC|Z}KXa`+UIcW&zueD>q7W zW?e9)eh6cI48KTO>gS!8KZ7(DQ99GLdkKs}Nh=LSU(YmG{lFZrY^8?N-11kWsqgy~P_`=T4b)CtRiDXic_vOnJWTs$y zYqA|o9j@Q_=oyyg-jAph_{K-e5RWU42AT?yAoQ(-jD3K9zAENa&FxfBuC_9Xx*N(K zhg}HuXb4b`ias3TvnZOewGFNlm+SKI$!#DuT zAV~rNMm<}n(}Y3D=H=k|R`o|&@k%@wo&7=u>v-}{E@wa5TV3u;$v0x_Zb7#f_*QTbP zrpmFYpeue+O58D9X`3dA{a7(SNc^8*mLwVNlgNsRryhPv)8Sk6EKk2!hhX&GOkaG* z-ye@hE)nVw@Vc#AqB(iFycyg+`%irzEo-r+7~2CV&v2LLbp=T7)1~*e#--l@b++L1 zpG&>HTx&`0e>oKNh?;F)LhJf-lk%=N4}mvPamg@+CE2KYv^nmhMJ58S=}v6u6*HmR zm)Xdg+n?n>XXX@n&X1iBCYAcsXt48DQcsYas}1$Y`((AIG;KP0qu3{rOREj`@Mh>h z!Y!o7ry?l0gVU-IFYTe`rF+m>Uk?dtDpOpDYI;{EquJyuOPj~24YnSCxAmBjB$zbz}WHXkN&4r_Q_@-{7xzD$? zykvW==WSS}>E(p>j~O309%6b<=Cs0(LtVo_x*WrXLXL^|V>|H_VgVKW)Hn=qCn5t< z=TDhl;L<*r)hUgW^phjXdY1$zo08354SFMe&OML44(TbJl`4=|JTN{7-q7jf{C=lX zK+Ogau?4%tAU8{8>SiyfA3`lr5Y9ec^V}~?)dD?G81RZ#BkPR5rZEI_b~!8sU@P2$ zqbQ`9lk8<4rD17mndgIopKRaZ-Ox@n8_xO=abZW!unl&#-Fssa(Fe9`rX8Vjsl3W| zzX$HpfLeZ|%#RT#iNI$V<_yLn7vd6^$ z+vCT1s-OYUp~yx>z|mpJUUn1%e=Y&oJCmnd~G z)6Q-EHt`lmmYQ8PV#P-RtRtfhx(+Lj-F$1D^mq~TVCNc%k?Av+ ztd;QfM+Xer^aZa!b0j#=0}M@*1~Sf&2Y*k2I(>zi+^z}MsP7yJXy!s5S&AvxD-^41 z{OqpRYI^M26v&~8u04kIDR2n+7Rm#~0NIjVUSspPdgIVrF^}+wbKZ1e0>3C8f37)c zm0Hg7B~JMxVFqDo9hJmf%;{*U3M%ydKZp5)Uoci?lL;POW>tKIvX|{k-+@ zM5;hzl`uTL@7}_bK5oJ`pF>HCI>i{Pc;3KS$-P|<*{V2zcolrhkhaLIb+`j2K+Geg zX|Cs=%!WG^NX0*8>q4hV54sOGQde@)8O}-8vD<#h80)O`=x_PTP6{~`dC!}$UYTjq ziME+LQJQt3@f^!M3*{jTD?NhZP>XV3K%yR{btV30dWv^&02qyblJR)uXfvvv;-8t= zf0M3E7lKsrvKD+a$(p>Vt|3xNtqnb{A?mi%{O5BN8Px-7vD9XMqg|Ftxm$0P7`?1Y z)4sh+u~M9I&okoP)?&7klql_4Q)C>=erdZ!&K$6y>wYR1#OP|z6%sf(zSD||{= zsUCiXk;II_B{)Ox+IV(kiH)$!NqZ>u_yuBvTzz(>qUY7$$O!{z->b%kw|DQqg*nH- zry;{|)GR9_+W4aEO_WdUzV5@Q% zPhBrp!1-`*X7b=6DK%G^hqzKGh814wQ$V(l2n1D{qXYUKg(tRw!U%GybHvPg5>@kX z6aMH|ZR!0YW7$yGy$%EvH;a5Kg&NV))_ICmSq$CiS3caw*}9qW z&hit#s?H|icuqbUkcMm|3g&UOB1Ri*_YAqke8MKx>uI49!f9gH&%)=p3q?Nos}73B zWgF!_z!HjNm?;C%*E;;CQu0Z7qUV`i+0VTynzsi)=K`H{frYdb1|uPut)o)deDiNv z!T>I|7XKD$6V+i(nkx!BWX&s}lSKRJk13o#g`s^Tnc&=X^o0m9BzA98>I-{C2=gj? zNP{{z`P3-f;erFDMD6AQLWD~HaO_8mrJ5-Gw7<}6c?~E;nRnUJ+|~C{ZyDHa3dlE# z*rLyA=l^|q0HBnaZr6Ob8}h^FvO>4gKKfssmk@abF6s7HJ&G_hJ9~Hgf$t|GBYJhw zZU0^g1yc^ffLnVEd)7;oTo`kM5f90j+LDUyno-H#H#v?V5yJ6C^SRTYgtUEZZq&h! z=uWz0$K%5>F`Jh`wjb=0RSwUP31$(JjVw?1)+Rfaz0SYRcwqW!FWK?WPlqWb)P920 z_YmD4h&Q~mJhfripsJ~Uu2R0mn83m>-@N07JO4t{+Pl4$>D}|%|&j+c* zs~nMXI=#Vqm*BO7^6;aTw6r4U=JO}f-+#RHW$dhOk$!i6I6P2_kv!`vPV^%Sx9aYP zzYVE}DpSEltb5UnT6t!xF7kOCkQy!y2&0#iSErEpgKwCkf{^dhr_a9*sJs%_n@y_4 z*Sgu|Sq#K-(u6OV9UHg$=q# z;WD@e6J3oc*HCS}-OC1z7j0aEPqX7V)Vca^UzgAKP!GNTN7dXSV9#2-&!%_(c?VP8Ok6x@Xa?ooU=e_8Q6Nmg99tYP2D<#>CIW+jqkj3#|hS{rM8cM%O~36vjETXF+)K0WacVhgfxx z&65utloZ`~Z-3SV3EFw)I4bQ*@8J)fa>=5Qdhsc-g>)FnvZf2^Ds=m}72EYtUu1bg zSdt`)LQNYS0z*i{@Er~fAK!{#lR@~ObFH-Ids{Jg!qy3tUgi4@g)ayBvq`8MIAiy> z*)o!ecpNrEgiGEMB21U;CVnsrc)F<_YPXc-)}KX`*`eE(9mLYNxTBQ8rix9$I;QbX zOscHtR_!Ylfd0~GGv9g28Z`uMYCx{DP30O_2e>jWNa9t`plbGiR`&?e>i>ftZnP)D zu2+0;`95+IhK39IK9MBVsqf^Je@H(SX%hB7}4ZWaa;O%s#qHd z9xZ>SkBZvv!!}rY5+FyPxm5KDEYSIV8?kG?F4e!sH7LNM^E0#aq-DfIZzW%rg6w)m zMS2Ad9ka6adE*>+!HlDLyt=~2Bi{kGlJOsS$ga690B|>}4i4pwI~$_7KGd*Ej60oC zVVy^`FV7XG>2`vjZ-(RxU( z;1<1h`spA}M8Qe}6m$Q2U0aM2T*+Y2@3cm@6rF!0UA#AlnDL_;YpE&#<4`(0$06yV zgBS4u2l6uF`}fI{gBDuj(JPeA)h*0|_eWzQvu#~Rc8wknxPlXVZ!~9z=)d{{QzpaI zf_I}CtIQ<}4#6CnH|jH8R2C!IlRXc$1)C&EOm5ctmYRxNkd&&F2`e%3{;4x-AoMROE!(e4o z0G^XDB9ND62!FMCGK-u@bw{XVN0s!1`2p!+#b<64{2k1xu$@Ma0b%+QSoO}yIuwda z^MMB(3p?n`tog&b=e@pD6Zu|?2!PWowb)C*uJps7z?C;nbk2EGaN=FOicE-Y3Zg9gf+swABncj>yX&iX{ZWoNo5L#3|2Gr$9` z?^V_tctDUihn1SdKJ?1rvVljht6R{k*6 zO}y{hMRFLUR8#B|)xTV5lhCfyrEqOJNEXA?%L@mPgtTb%^z z0^l;%%>-JJ%qUAeCyNba`^0(Tc1oMaX$``7|TMS#{pH^2n$UHlWTnjgz|>WJMD_cv8s9bEQ*F z0%fOLVu*SUDm(ijUx{hhugaEsIre!p zwwIAd%)ye@^R&G*2}fR0oBDNlZn!t#oR<-u*AQ8+WdBf3!4yq$F$Cke8oIA=NYZ)@ z-XqYUI<0E@sQck%6f33;sMo!9jlvMt$6^6M|?S*CwS}g0(^x9;>gIjpo673d9bxV(*{sKuzj6UU*!G0-0zcw zM<`I~dNwx1{q8%t?r)VUpKnQdXOX`?<)heva7tWw*cK}iu$kdKF6vW;^C~daGL!yb z53uYnNoP4g*t7d;-&LzHKnA2Cb)U?_iiVwOP%Bv0iY?u(7&meDeMV^^m+jHL_ed|7 zov$rJel4Y@P2cF10+$HRI~i$b_hN7OV0Hz{0*2FCo=KqMY!Q_UM%|7$nH9ADs{GNhU6m=r$hM3(MksN11;Am zs>VC$^(W$gaDAj_Ggw{&_uA?P{A#t!tu=@E=d|hLjzs(V(>6U;O|6ye zgA$bmPkUTxY(k~T=V5`hM9avXq8UK|s)2wI5X5#JN59(`omuL*sM>OoV#pA)cLD%4 z9eRN>XaRqTq3H_qV>@XFFe+nn_I*#&a#Y?%Hco;0lSpXT`6$T+F$H@0)$P!BJ(Eg?L>euw!AhpwXp`^Q2Ihfk*;Aupg-_A7q5Miy)Z99>^pCAv$ zd_3xKWu7eT`ubv{tH_5gDK>S1fD2l;tif}Bv0I~`12|W%@LW_Es}r~n*%0kPNcpa^ zAAB4L%#0Ite?CeM&%b#YsN}}*P}lTODa}imT%7_X@j%=esNc8Qx5GewM9;vN+f0* z@RaR0ew}xdD(L4-ZE1A1CR8YHB_$J!O~Nw`t=CPoah*~+QjqX*vOFfGng7ZI{w!T% zp3A$C8!r7W@3{+&8-HJo*M8~|Uz>T|bNu9+^0*Z8NYJ;?t_U{B#t*V-;4oH~d%_N1 zf<=h;HB7a^mpQ+da;0yT?9s6EoBz0$kR{to1{?`~s<1+t@CHzDLcgDmDCaJuD8Nr~ z!=Q+Q94Zq$)+_iZ`)SUbpT&N9 zyg56Nmtb-HnZh4){p{;J+LZ0xgEN|MYq~G=A`74Wjr{b!$y*>nP~-T0G@p;$gaV%c zH@B-I)uTxE^i!zaS6Va0^zVLU(oPD{ma-vC2)<82aU`G9p+0s=OTFQu1vKzf*Tqi$ z4Xjer5E4Sx!Y!^t5v#;#w3&W+0s1}2+62c_RX4`+30C`2b*c>(c@(ra;c8iyakcUY zF=kjsXF)S__$q*5{kZ2wGyT=Ge;3&Q53Lh@6)^sSKSF%~<3HTh*C-u%f45MMZ~gMg zoiI8&5k`BPNGhYZc$w<;Gf6xV&-YucX@ zFbj)=l}T-ijah0wPKJ`R&#Wp)7@uQ{UouVJsqTP(J z{-r5cSmIm+W8g`_(NKEcamdqP3hUtubWI22W#@BqX4cp}6-uI0m&qx=M;kD?EXNXC z^i*|55+n4ad$#2nYmgLq?#y&qV|`l=O&um(^`(P z|Bn>^rv`)Urp%$*L)MIqo6T5)cMHF>XJ2gCw$}18ZYyX!zL)R#r9E87u|66_Ol!RK ze9m=W@wv<&l6(NR*SK)$8sf?eDp%?djIXx-?{DfvfVcEm+My3mz+y-+!664IwO-6+ zyajSEyv^@a<+uZ!SFW^yU{VKu9L)TrL`M5i&ADtdZpvpMtx}7x zTJCo+MdAI6r#Fp98*_dVvc@=d$$9enxb4=?rfAoh_Jc4^0#@d(T7}vLpO6#R{!9op_kh>BT~E_;zMXJ{Yzef=uX(Ab z<+p-@M^3xyZ4pk8|BhZ|ufdCJ$0gaC2@2Q_2)VXY_7t0)HxYrM`l0fw9qpWn4tc3o zG$Un3P6HX_!GDamG7rgQ4*P#pe>gmy#spHnS%FD50qJjvhnVZ8>V+@=*B4~m*_iGa zpV!R!Z7e_wQyKq2UR=1;k$R!chp**JD#M@r^A9o%T2Q+%-%x92(NFrjSNhM6o`0^A z3?JKzEt5Xp2YJ8OqdzfBKUF%jmKLS zFQeN36Qbo$lqo1nF>~S?LV;Kk^~V~f|G3+4>K?=$CG*h_;tCeEyZLu^{Kab&Lcm5) zK)o3>{hKd!=fxI%aPXDCcQw1i&s{l1nceiawi(k6c>tbY#V7iES8I#_mY=$sQ_=sA zH2{D13rw#7Yj~Z6>Tg*Hnaq5^`|>#Gz1#o2|N66E{wX_|>wt~8MaT7La`~f8k~)B7nz z01K(v@T`jZ@A!!ec+UVib^b8F`1h`ULjx>imUo98@8475uQy~4BU@W9{k^OItMI?t zy8l)9fA63F+%*4H_+M?^|62H8`h)*k_%L^9|*I za5_E4e%4COJ0M;e<0mfCQ@u&r!4#eNB{d;FDg13bO~kFFsddA)bT-8fs$V&Q3ztHl zTu95**!FPyc}Yr|jIdXu9A$H!^j>Nbc7gw{I%b;2v}&^T`IiM{^B<{K`mqi31v;wO zsXS@~IkNB9*owm9(kh<9;liNj&mX>WFPfXaOd}MN!m1+R*mBx`-F5COoyS(tm51#v zH@Z}Rl&FJ5PZ`fu*m@^_ts?AMc;5-$tcGS!{?y+g zrhh&$7DwXoD1CF9g;;QtFz67kA1&79?AvG?c6dequi)127B*CoiB_DU&F73J}8XZ9rixSoX{SrL|y?+Q`$$KWw{glA^e__2e z`F;CFv&gn}Y2wW;#jyJ}uSMzqQk&)*{-sU-_rLTP(VXB57alM?ekiN`56J;i|9@!# z{;zmxRGZj11{hjZ3_FjMTz%1Ie?bX;Ehr-J%wD(TbcW8~4qOQzGqU}p#&O^HAKT&X z_-+DmOA_^*fpzmC8XpUf0wvL(Ri;fAdD?rlS%f9`gvALMIaljgky2&j0hjn=kCUT< zNAnmx;_jO`HZtD7d&!;N%_MzTU^WOT$hU& zp@N)xfw!CrpMHKh=^8Z~jD(9}v9E}ik6^l2!r~$|ZOVRD7(Kvl>md?#GN4qcPa-Ti zl1!985XNJZ!AnTw^8RfBz@_3c3N-89-v)cUZS?LQ-{)2!?JgQzo8j!$E8|0)Y-yQ> zy`D7vxOukx`Z|f=_W(Ee4^7}itq7ZpS#P0Ae;?r-?}|-c>1@zOJH@xf zR_1?b+ju%B@D8AguFhJ_5fiKs9s7t_@Au98cmXU6!3CDY$1A4xJ}YsF7nuo2ZdM)_ zdN?GF?eV@dNa$}{t&2+br{X0pT=g17=DJkJzC>!ov+oIOICUDr?Pt zcy7S-n=zj@%^x3wSkCEmX{mtt?nr5bZ&1=%uv9Z&y3glyp^CquV``W>~Mn@J5OzP8~Nhew8Cgnayp!csO;I9*(IZ%U};+VS8uQXXGHbC z5Mk0!Cd4g%?R8QCosPv17C*m~iP9Xg?JYR%O%aoMT%hpRBV|mpD@B&++*Xdy)L@31 zVy@8}hWF#$pFGM!LNO&JtH!7KcXz~73j*3cU)kwW)BVgW$i8`;)JYGZLjM6P1#%bp zL$%kr8p2lcot%Sh2j4*Ff_l4*F48!g05{nWKISrSZTP^Fak%-OC0}(Wdu>NtuG$_pMsRIRV$-Sj^p%Y^ z0h-4{4kG}hH1__qcTG<^{Na9UJuiI9wOI%#Dw^DETqLn+ne`Fr)Md{oKieBCg1}eI z@J4k&$B%%Hsl~EJ*c9Ywmel6JVSTU{#f*V%fm-H6I$23G0KpxeF%*)io;Rgtdt6abJS0ow} z_a&g13yKPkdY6i}!l6H>h*kBSxtKC^zL9A-%QI6+YTAFG!=d|t?55Ud(qfnLsdeH) zy%+~h292>3+XiP<=+{1Cp!7!sbUXsV0^LXq3QbZY>(`N+2WR;K+p`BYsBuKHkCQ)} z)bYb|1F;m+%Kl(}d-|WIXJXm*40C=qfI2vWZNQ^iTpswx&vjHASbQUTw&n*31je2% zr!YdFhXDtxJFDmvT?Y4tu*SY&*qLSwOcWV|Y?3-I36*#|HI3@Fii2X!q`qNR@{(%x zASTe{$gL8{RN1s=``rFqfL@;a5pq=PIp9Jbg%lRt9c+=S_ zZ;L+L_-SgqUj2v~TI1+ubKoIT&xbeA8CTj?gmK-gCMN|6e*DJ9?HW3Ivx(lMId!{J zrw6IteEe@>5XV4zU1oo6Y3) zZX&U_tjJMEUb>S@v48qT@jPk}f2m@-9j=UNv_0-mwD+`prJBAxnU2wr9Vv%F z?7u8{c1J~XDu6ck6g@8dg^znJ<4)epiylN28=W^byrjtgGCwxOb|V3HlZrqmK9@Qly7QGBcgyTCE1J?w&y!}XV`pQVTWJVeze^yYsdcS})Gl*C<-l$Sh3iXC7%#b>fwM<6utk$H#!|x9}X>REMWGn}r@(ZCxgv zl*Ya~+Q0-tgj@<+84^-W;WNH5qlY!K!Zy<|T{qBqO>=AgwzDfg4)Zx|O>o=guF_;V zeZ_fm?!l8C-=pLG4I}(<6pzSY}bzoDEbH{ZO@O#5;;xI-GBXeq?efG)$mxf~n0cfinn;$+W zSzoIA=Ag|gv7SWhEOof|41u#zvTiopat{dI>-cHnbauMoCNnfGL(SJJrf%bN2b2%3 zo&ER~Vq(U3V8`uFgtViyZ`>ezvej6oqR?=4cMhLI*2>L*@0}F=3vz({p-NF6{a+sB zKR!O83tHfToF7kPmJ7z`RD4Zn(=}&3>l!UeQDi+Xbl|)r>vq?=8vD?PFdjX@s!Xn9wCDbLwOl`2#!>K9A59wzvjf>rmHMW(#eOL0d z!(n3RdiZ?^^~2b1Wzsbv)LEh|sa}TBbXEP~%R~=Vr6^4Wz;HaHp)h!)bf~LycSX|4 zlfFsnd$zoB8O9DTn6?_t7|-hk{d$ntq>Vfmb(?<_IdyR!e&U7G=s9l3;#Vc0!S z7=yr|pUaB(H7|)%oD#!$>=c4-(Tkq48Q*lg?_KI?c|`7shwr$#>8XnvhKJ{=NBIuo zHaJOykzahixw$bFnBQEdz;jbN4ZG##8fh==*+|9-;We%q+D-H`bDO>Ar+lK;DY(8q z_0jSrhv52?OE5V;h=Y<1*XL~c#;`gv53+G_p4*S%8Y2C=*01ftjcerW{cQ)OIZti}xFY<=hx#w7 zCbvpt6btpvA%K}JTyZaTG)yYO)Ype=s%mGd?<@tCD-uayFTQW=6wRJwAg9qZjpgK(1oE@1DACTy+D)xbv z%15N*Uf0=f54$wol7jfden6GK8At%8wX|UXlC$$tx@0GYXRpk&ZUuE(8blp*832Etj5p?r5R`%?t=u9nbrpBs0SofIAYJ;KkqFPR- zLJdJfib{v`Y7Iu0w#8jebdNQY0Nt2Pq(nWMq2b=SnLqUx^ zw^Zs^<~L^Pv_de~pX#n?vhC!?+$LO$iM((8>a3)uZD)UfzT^Z!*zrSEWCLvChyU^B zy0%#Vb-TenSK-0PqDb&ciO9E*yW>Q1^b;wH9y9hz&%pkeN-xEas-Bj;98)(IX$q>R z`*R>Sv0c5rsu^pfc)jBjHx)&~6mNMYy1~H95wKbBrbOF`XqFza1=X}2Ly4h>o$pcB zW-d{ro40+-lm;%(Psdsk{sr2!xYS{AI0oh*UY#&m@$nhXXDe+~Z(lQKDSy1X zEBx7{H!*tEkXLr=WobaL#@G%GG!y8!zX|^V9Upmtszf$)kjVOB;IhQatO$ zLNP%1T=;JE>Ry2hKHa;j^g1%L_V`kU*l>TAnRH*Q#k>)?J4b@0^76qXry}m@BEyd2REI(V|ND-R81sHV%km&T;B-x(Ikh zq?)Vou(XSq_uXx}Sd3Jk5$-{PIPdvPQ=W5J&nJI|Gtm@Fpo3Vp z=dP3QMP$qV{^|<`#q`0X2S}2`$ybcw%V)Z?u*e3^TQ{>}G=UI`l;k$5mT*2(gT!L` za3y28m_#brn9iYc3ay&teC2urKvE=@TKltC_;tf`fnNUnOz2TKR zU)xs#PPg2btA!Y^6fMoJ)2}Ug(d<{L?WUdATUe&~)Rf)hg5bQ8d+w`0GuQTvcoiQn zQXyI01eA9w^|H4)9i2W8kui(PmBKV59Pl}3IKl&Vc=S?sD*oXV!t=t4D?cI|CW44e zXS8{C33R{eKsU>+<)XTGo5)-;GNK4SLj$txLo0A8`2L>nCg~_Gv+D;rXoWH-GbLUI zWF0({?5BCaoPs;)wfPcI861FAZxEj%+aWz#DBDxl<2VAL0u!T%JRHZo@hoHI3MDA zccstuYa};OwKP|}pX9u0^#%>wF9=ow5Nc*&%-i#Og@g!_!*#^g?S|TEW&kQbNxDu6 ze(1NLnw=et#>};DGTO>~!RWqd`8588$us=R$gDF@=flyqWw4S;F5@C&!Q=> zXXu|9YE;eU?DbJr4_u4S&hnOw!j$80N(0=3jUt}9*(B}QVoizH^<8hD6=5FFN;5UV z+0-^40OsX03@yK}w>fzk8sviu^2b=`de*aiGeTrIvhF52&2cuTCEaKatg>hx*cj?O zIUN9%PTAdCa7*-1Ao~90^_EN!`${=_Y151~>IXEElRk3=xeRV;aFDCSc({AfxB*n_Q z&Gz*wa}=mMC&pViwiKHM$@ilfYtBET0jEgJYJ^bXsMqpHc2r-NkXG}H9AS#^5SP%K z2Cmp=os~fIyjv$R2J~znvQd)0I9ft?xwn&)RCwHJTIkwnW-3z;@q0$Crb1B5*kK=y z;?XzjZU&j75|OkWfUI7t3Jds)q6?;Y8F#8?8O?;Yq}i+7o+Rv+dLSpVruijZEL|Rh4p-) zs^EwZ2vOeYe@TN(ijg2c!Xcf?1QjpvvD0eDG<^C|13wqUwPn_ur^zL&!jN}f+*EYe zO^W8mgEF)6%;cZKppWkKnaE0QALpR6H|T^yM>4lcTN@>}ieT#$uQ3+$xXUSRhlI2* zC0nTWU7`pP^V0OCdZ!Af-;6{lJ$BP;$8)u%7cwtH1|{>7M5820-?#oG0G|{CeQx$( zHGk#(t!?X_yH_F4?lzNBGp{tIv= zd8x0Qx`D)>e)jDV==hoTnvI z74qe0z?jCw=5gbjTw%7-HO$`G&GG=2n#Y;YN+0jvuIx+ZZNA^* zqu%JYsWA#IPR7UTT2>1Y<2`N!$1~%A} z8akmn`t(%IAe?uT{fOF83Bt5Z9s*c*WuI| zm?CBI65})TY>h3jfhI+B4nfm?t5R#hV~M(=0M6`@YG4Y^KR`&XFE)v+`TQ8inAdwB zR<}8Y9S{})YU~N}SmV2jAO~-}>1wR%weUG}L6!<$Npu=!G!kE(=qJslJx|t4ByJkH-5ClPILVPf za_WQryb<+M{O8!8Tm-doN~4@3`+XrUD`==N-8Sa>O$EEv(gqd=4BXH$z;N@c6Jw~QArH!^>!(_5*Gyh|B@edCk`lNM zH_SLh0%3kj;*j9vW%e9GerKy1PLwdJ$+(`d$E+8QdF1JC%l*7b7fJL+!WkltNl zGlrb$hYz)!-@a)B^nQ0n$=P!%hhtggrct)oTP)c+5buW#B)c-`AE!#ZYEuFEbbc|; zLbj-~f4}D!DNu}lZY3=TvNehgU%&I_*7y3eE6dA+~k&lQ*~XRkOekV*B4y@9hi{(V(NJbTK2bs!^P@4bRBhY4>i^Z4UJ= zGv+&DZ#+~LyF_7W7Qd#kQYb(}BY#JOf{?|Du)Q?KlxDj+iWsMn*Wz3x38%Zg80do{r^syp zsU?Mz7EiL1~aMKjSsOCzJRH$^8I8g{mE2+HY+jM{Y-4IZ|~Gq zBau>w4H-E*!ROlDbuBPx43Xg%?FZpRE<@Ayx)8wj@H;a z*>EGq!1Wrv3X+d+$mj<=j0@s~m=wT$$ZZriIe;&n2GHM!PqxR+N)6Xs4-6GG*`}48 zcGskZ39H8#U8V<1)^e81E39Xve`3lEI3N5sp4-)=ia`q_K$K8&y);8n)x=;$MXThg zg8wjsx$xOh9<31HX?PBR$W!FBHmlnAZ2(3?aT{k%_R_ItgXY?|l?+H0ydfqQJEg}! zY!pL0Z^YZvdrU$s-oXQOaU0_4^lMRm^Cb19 zNd;afe6nD8NN>KSI?D4c5wg@7u-a42bzHRcHXhY!i@&S^yTML?M`&~i1&`u#_f{YC zPiN7@h#Ui=F7LF)&|+$ops-v zdpsa&hiICuIAfpL6$)08z!z)b=YX`Fb{iVm#342%=`^Ci?vs!p~WBTj9Ho7%IrGfd(kTPHM$S*6s5Azx}Yb}2i!|J4(o3=`X z71DS+mV+4RRqczWeTf{ouMAr9yO$l-+)I8H>I&ZD?2$Fy&v1qbaL^9KCCnj+gq#!I z7;{e!3BwYBc&cNXr2zd2w=tvEWMXB5D#o4GuqGrQ3ws5L>>(_K;bl=%#d{V}uH4EL z38Ltdq)|fGnqc#`L8>-u&n9sp7jqhbq(i+Ym^Y%_Yw?LsvK{8}chtrTt~4HJpq?|U zY=A#OmS~Lc3=7x{X3l+cC535DWBnPXY&{Hq8W~>5`N2Vi;>X8hkZtRpam{BiQ*9cB zo5ILN4O5aN7vCnjy|!ZnkQi7+9%T_1l}2CoyWd@L2U0XgJ!fh?6_r7=#5*qDWEE$7 zVlDApmGq+P&hxe3hzT;ZPHiK2hBumM;0sXs4R>3cni4R9CHY2MlyvovJJ>48q{_G|nmUPT1r(HLkSj;YX4~0-9;sZCxv|+!$FYFhuUDSfHvZU0tND(R`~)W_I9KZMXC3RL!QPuy zr~Atk8|4lC;w@0~=^Hiid;KB+wZ_@VQG$J4z}`!A4PkKU1jiVdEPWt(Bjogv_pK(# zJ->}m%4%fMcp}LnriR|u(lgem)VMiSY;rDfr#9aDZYQj0;%rb}=5FW-QEwTkx)9+K z&jxX+8<*zriwhC#Sc}Q*>C*~1v3GUItCeAJodxbP{Ak_;lsyo^)=aCM{zU{!7Q`)k z7(c#~2#%UeogBc+(yef0axcXqiHmR5n3O&P!jX5DQ8jW{sLgsCOI?qBDBWtkF*F=2 z89^2O8iLIaCrI0;Z~(6ok0&n6E8#c1k^+W@m2M5pr3ka{V%~-9o^b;s##f|^pEoG; z3jfR`=dJ=bqfz)8VR)!N#ma}iT6F(}hHNhzXLt1S+!ZtTthjz%^g@}HA)|vUMM0*W z!bXYa$kj)d5v(_+FF6Oy%d!~C2S0I25&V{@8~nyzMKI#C>9$@{*TR=fuH!A*9F(spc)OB}_a~VXIX} z+Zj`?4l?R_P~OGt;L%!PVKy#lTQ6SHAuGl+4pTNmjBPx3=Q^!A*SenP`RDoT+<$$q z-+f>A=llJ9KHu;6{_9TR`dLlMNTiY<3;rR1L~U`O2u55P-FZH1zvRgAEcQOq9x>te zlN!OmF2P!hh*Kk>B{Ki?Uvpo_jXy1l&_G&QQ;JkeV$q2)zLVde>5O_uF~zT|zVpmR ztb64=zjp1$2MiM=rUVr|8UueA`BVBE@|#KvF+n1$0Z(ix&seyL=5Xr2>oln_Dtb~d z8R>wc9e3to-npa?FG(cEF%ajYolmx9G3K5Ce{-Z@7dab{1s8`q8b#0FYIj)wuud-z zw$;^~HfKvdD_bQQS%z;&V$U3U+u(3TXdW&s<5sF>m>i;PV!wjm^qAofryh^}7=`t2 zC|Vl7J(sz;B8xa|wtO%BxR63SAhj16_AJtsJOhXPg(qYcep@fS7E9b2EsEh8O_PR(UT=#Nlog8aJ5grw>F5yXCWLt5K{%&@Os(3mLc zVm`l2XpL>JwcI;oxpg3Jh;Ac0y}~PhpLP$+t|5Ds(56E9u{&+%j7%8FIt%Ug36?p1 zhG$eeFu2pjvEkvqC>JsWBrwbiE4a&;EiWf~#Q_ut&I}GpUZCTr^z*de&?l#Am$G!j zu6`?bNzB){CV>Q1YYVN(WxII?3 zcex*F>^BtNP4^ul4UYS(a*sX^g-b8*@%g7u2x2PCB*ZwQ3xOhmF(#a5=eWTZ`<#*Z z5JSr?k78%+X%R<*LzM+%6!rlgxQD9y)3PeVaR=f(?_rP?7dLwec7p6d&>xGWR%y_A z^d7Sa#|}%~tzHi^4_Cd-=wI{5q~F7?cqxwg%5H3Vk2}FkjF^3S$7Yjdu8BkZOkTs2eLIte%T(w~fSvMX3iu)Nez%?jlWPr8Hjf)p&gDY{hFxZotpP z=aJoE$E4@r=Q;}et}lo)?wuS$Y1}%AAUCc1Ld`$)Ek%IZb+CEyqP-$8vlxO(Nyy>T zo^JP?&ERHG3r0f2{BRBG^rw*`%sh|eO~tu$&-ex>;uiJsXROXkh5xc81~C%$2fa5Z zL!8UsMSB&;rVfb=`fv>_1Fs^7zyB73J%m&?Yc-XV+n$4E!`jZW_>Kge> z4>!0&66BiXI!nA+J7akxEC)U>U13a8^YY!ZjTh2{D$yA`^<)XS8x$vD%1TR2a~?($ zg>dXGH}k5qehN!pXy0CFPbJxD!uE3XM_D3e?o`E8%vC5ssTZu5si@xM)@-mwXk_*G{{G6n13N2j(HKJ(xnsj^Y)N4LD$+=hn2ma!S8-V4n=+vL%D6d%= zL-N~C(DUv#&8YBeOt9R#oEpb3Wa;&uis<}N(9uQuWB|CViujLEMPz_@22=w*CCF3A=d z)+aTqN*z#XFxq{~PWYxb_&4lRBPuB|FEqc!L*>gY{;^%5LJ%rt=y}L~V3h4qAAKf@ z+9f7jyFbfr=i}Yu>x;dgNyA!3J^A(O@X;dfB>~08XCYx@t1!#42)FbI8En&1u7)1O zx-{`Q9X39OZtE3JTPPJkK@bap4+{KcHIE>IJ+BSzSS&7^A1iU<;)%liT6NS|t1Gp7YXP(Da0T_|@H9QAom}Uu z4=A2ga>Z@eYhn8*PF7ib1(}!FBD$3jdZH!_s$}vh0f>MMAPF4XwC|n?N419I+ zaP!%SwekbS!OKI?>Cp0P#AB`?fpi&7gp(>S+Uv18;I&;XzaTgo;0f#Tro_om{zJ?*{$irZ4I`Y&o=dd=4P zkGXzD5NIZ_s%6XH3Y1zd;Syj<@d2v7MpKCs1# diff --git a/notebooks/tutorials/model_validation/xgb_model_champion.pkl b/notebooks/tutorials/model_validation/xgb_model_champion.pkl deleted file mode 100644 index bfe7349b673d72f09dc01d5d1640bd3d62a5aa07..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 117919 zcmeFa2VB)k`u|U{fS^jcN*qAPISv~h=@@5rHk%X|=_=xgs zbGIffyqbG?d3!haXxYr&-P_GeE^eF<7CAUFF<8@j`2qH$9D->}k1#5<&<$Q4xW9}$w66dw^B6A~Sf(5s%DZd_hjwUC&Yq>!lK$e6I; zVX=SfmDSLQcv@SS=2U_c5@|IFzbjBJB07$CEM`FBz~7~tM8?q4QBlzmF^Rv+FrWhy z{ks(Nfw9pM!Lh^Wz$OfgjQd@lQCw1dTx>$b?-Gq;<04{$L&ARF)vAfHw88%)**KDR zfL@BO|^B>P6~DR*0+^St+t|q(P)%WR*yx zNaM(=k<}tiB26RBBF!T$A}u4UN7jh6inK0oRMY3knEtVUtInEM_fP#NS5{He`|m0$ z9{}Uf5SW+|kpl)MCiJQwUjAp5A`SjhL2a|jpB=Z?UftI|MtWvy!xT>&uk`$QTpfLX zo4U!3TWZsGyQOg-KNBW*4oh0xV@sz!RYu;hRwsG&R{N{EsD1jLQde5sP+jXxMJaIF z0BKgq8Fug6bcrcx%62;v$W9dPl)c*>V7ZohO2rCGWk5v><+=?=)#;=AsH4lUG~4m4 zx{dDvHCH`FZM)~2+M~o?+CJ;By570D(m8_->WUXzk-gK$NNY<+lA|vorK(kTDh=x$ zlddc@Bu`x%OD^Z$u&a(VWqK66V2eJCRC-?Ts+@K88MC*nGrM`Bcm%StGuY>GwbQeh6zORkFBE zmK$#-GhM@r77QX%wM8G9K9Ra?yBnvO{QM76o%_$pcHK&3+$k5ujn`{Q@r*P@VV$?W zPv3814V;@0=MGPqY5mVI!|YqJ)wrF^wIjJ~7UrEe%JDPQnYur|4by0iC$!QWcdJ9g+2y{4UCKmkB^A?ehi_rx)gm% zMM`CgA;pMNm106MqgYU?Q>-X8DYYrKl)4mqNS(v;#(X+~*IX-V;> z_)-{(lEQtXfyS63Qd&{kP})%fC>zhM z;y9c&dIN48H=JVz=%6=?Qu#$MQuzpTOWbf$CVy9q<8#dV@(V8_G&)>I+4jB;G6>f} zcRFt5zRl3#7Dd^=X0kyP% zIRk98fN2BjXsJtMz)p(;je&Yv>eCo#pv95KfRh$y8UqcrxX>7I)zXN@Kw~XUXbd#f z;znb@U5f{efo57fX$&;i(t^f7OD$eB2E4WS&=>$S3@EgKNd{Ozo)FdI;c@JTw-1;D zaNXfr!}W!03fB>?9b7NCMsQu=TEO_jn8P^3*xFHG9AWHWykLxATwpBV{NbG8JmK8n zeBd0Q-_T#^BckwN!f7l}{3xv{Z7IxV zM2VurP~s@@ltju<%5cg^N;0(jGX@w~2&M^fQFI9%78?^D3Cr@|-$K}R`n{wMoS^J6 zp*@?uWuuJJY4S*EZ65~!w!cR%L;BjiC+?#LvzEc4s+Vyb%dV^oX%I8pzI4&#(|x7h z&{M2_YP2$0oT?~kGJttFs6>-*AJmNLyzUD;utOTO!BqODfEenPk7s%YCO|o8Lq^jE zlr?j_u7S1zbwWPGa{HQNVISJ0bpV96s~i%S^|L>)SN&~f(Di3z*zcdlQ|P<_10frK zm`8E|BLaBO z0C`aEd!P*{3o!t7%l&{@ZeMQe`#R*~XtAF!$OfazZ_@%~pZ|ReCZhx@s|<{wN?!4d$>QoWE5y3_eqv7P z7tvyZuQ1NJop6vDB{nfyCDg85D)jN5k5g;x5a-mXBvjp6S+p8+8tYZ|7SXkL!pZ)X zg*WNz@qT4}!EyCryr*5BkiX+TE-?2OdL|nPPcH`WR;B9&vusB2&_lS<)gk-`9Dwz+ z4~eP!Z3SN43S0Pm#3zqT7v9|-i?2jwaSsMQ=MNS(7o1-#6>Ahs!AD4tu&~7@q59EK zVfDPvIAe7>|Fv@lzPaiH?%LfOd+X8~b6W`AdJ6blQ!C!K3BpsN#^L$9I||0N%7hH( zFG8n&i}CJ5<{0H27u@;;pi?6-A9F07zu6_9e-v98T@?m!(Mff9+O{xY$$YWZk}lX~ zZDYaOQo-Npy_?tbh`^kfhC{pL3$32s5N19x$DX!NaC)B^T$?z}>=d+Ww)q`jfqQW$`*;ph-5KF{L%0W1Nh?%?cE{MjhvCI+)`D z#;F`VmVa`17=J!18LvHXi7#?<=bw*8{4&3JJRJjmzN;ty^_v$q-O^X^XdJ~K(2c@N zV^ru#QXP(_^J{xO!hUNk%4e)b_WHd@X9W5iRrDefBf`rSI9)gS5KXpL94^n+bN7(* zHS(QCr6|wUclT&6XPbiJ0J3!d;Gq#=RE91urR$!O@;?h@W26>67f7F`e`8(y=BO3@ zhDbYiKV*_>4q-1hXsY&4yQ{7ias(T4&bEYISA0-fwcL}e`H)IJoC#$6OtNBvo7`Z>H}+w#+`OuEX(FoaCNyT( zsOOQT&m38+;M=UP<9zmZ->vNDd$~$xcTFiRJXV^#BV8JPXNDviY><{seJh17F_*fW zLXwA5wuJUiA}+HkOO>L}N`AZBOMx>o$-~SW64$aC$@Y0G`DSNH)puu;GmE;aaviLt z&biY_@r}x4u-ybj^@3yMm`flT>f4%h;HLXdSbd&sPc&gZuUn}&XY_`7>y$&@pK~Hv zOX`yO7T1f1M4cdy5+H35@xyX=hoXM~z z5k~CboFpkd{3uh$r=j9YEW=i)ev4!~FJpEe$WqjE3uZo?eXJ_ndzjgjf1g=BEQM`S zxheDT`c>9*;6v8-xWJBTe@vc4e=Wgm%zyZfD32OQX1emuk0mqcu6)~vTRc2Az@q>> z62O`j)~T>Ih4m<`L1EnqYfV^R!kQA+k@l2&lm-+hN<)e(1=fGC=7V(}tnFYu2WvQ3 zx4}9M)@DiyN5K@4(u&fC(vA{9=|Jg3=|bsB=}zfE=|$;738sWl!YC1x0hCC}U`jM4 zmNJBrKuMwuql}=8qNMzc>z(h2uIQuN|qzkSLN zo^8julzjaC-sa}`&lqwLHJ+}M>@xGm`=?9^#G$ktki z$+Oo{#&p+Fu4|a8oKU3%ndOx!-}ScREZ3HxITAwAP6#b--ogqJUkY~pqNoZrn(0TOT=zs|UKt2KVwaBX;Ff9PcDgc-n09F!! zp%!_?1m*`Y)>4(m0LU-^m?Z!t830TaV6Fws6#!BV0HzB7*#-bJ27rVEfJp;D&H=!@ z0U+%FVCn#nc>pkbfSne58Uqem>d_deucZNv0Y@!PGzOftG^8=$qQ#ZQKqD=UX$&;c z(v-%4n-+H(10GtM(HQX5(wxRX3oR{a40vhrrZM27#h1naNK*iqY5-)ae?&?fL(9QD zfa?#}8?G^2TezNZ{otCxwSsE{*8{Erj5~}qjISL9#uCO4#tg;@#sGF60s&%3lLO?*KwA(X27?7RC@@}d=r$ITCgJ78N{BOw6^bI-2%hJs zQ?c$0_xaO7{8?E?sNLv2u62atFK=+=-?bPkMhvB8E^oq>97FN3J{5!qc}MVtrG?z> zItBdYhu(rsMn}h3hV_Om{ z=ta8V%l;DtwnMbgVpId1{&@wyyLqHgq_*epJ!;GuHtNIeh?s|0%xr=ib}tiZ=GI1s zpECU2CwaVW*Ry>3nkBqPWpl1Wtq1(oE{;Om{!>MdE6auYncMjEMHhIJF8%qvv_D?= z^l{kfih_5=3}JE&L+qct6New!z^!b45(>*op>?? z>7@&NuRX!I!Yd`#xxAM59oL?};`xAQvmfA^Rfppp?dD)Fm(6%|Oh3FnrZzt2CSf0! zSfTRsiSp#A^6wJcTFNc{--~Tk9%o4xw!C6ioikR~VYf>QDvu#;bR*fY^p)x+a~nvv z9iB@ikFQIO*7PQ?9~&@BEjqE48s20V$8}J*$SzeIf7>9vTYNxT`t_!CVsHj|EFO~3 zhvzK2rIo~%`YC3Y&SDfU&g}TeWb&$03%21N59U*t9lQNxnY#61Z*`NfMQSXLQX5ut zSL+t7l}b&fN)LDGN}UfikRF7ukam9PrM6A+BkLTeNUEhtBDgzmnTrfgY3ig+8TVd#8v;#!pCeA-5;(kPECckU?} zO|h4(Ud&PKUe-pD<#3BEZy6&wyqzoQPT3(E+`{WEtKdDK8`)0208iERmV43j9sC?hDND8IxuItUPcEVc;+q4jawjE=bb>1^KI zpdTM(n$P#z>dtf55oaB$gU0Hnau2QZxxm?FyvhCsT&D-YT^KjmpwdPOG1) zTxynzx)yCjS1;)x)yP$x&JKSxIm?A>rW40yRvFA$Oz4a16mCSRk1ug^r~pYx4sPoejG?#>^$~qJhXOq6{?+FGPE%hN7I7 zb-8gl!_lPK2I&5d zLx6PH2Ehy3f_>Nq(M#U`UI>!~GWmEYC$|B@nGB>sfV$*1pez7sAjm--ve*Xu5FlT! zTa)lJ{!ca_e_63@aZIwGQPeuWmT_&xj&tqB^^@9)svZx;%^7P%>zWIMP(4#|z)*eR zkyuYm*q9?$)JLM{*a*>W!VBTetmXJ&d?%rieIMLC{jk_!&|+ND%S5<#eHOnz=%iTR z|BbNl04rQSVvkoo7>PIXtXOB1t)O=biza9L3DxiK7FHLC+^xfp@RjB>1cP@W*m%VY ze$S0W{!x8V@H+gEchKPtQmzk8I8pci8&yy|Cw`>Y$yqY1jWXy^>V<~ruT z3{(gfyG(>;K?!)%k|BJx`i^+&2$8Gmm5ok}yv|o2S%@7fY{Eu{HMo$}A>55+m-yKo zGle#;eTBo}k2t;eXZXONB>qvITX?PadtvtHFSyP{Z+_d_JwnIh`Q>x>uP3&VlXmx{ z`PQeHf)RVvdQtzJlaZstv#|}|9aql5r=@=#*uxq69AIwNaE0;>Or8H^rmQ=~l zue(%pQyH`Wv%R#BM36`CPLT_pIo96zn6yOQL~=MdT#DSVUBRtCEqQnI-1$}Q&fE-X z$-b_tN7Oz&nLD?#S;LD#Y)+t~RQ;U+TgmVd%Ukqf&)?fG#ZPD;wVilPazE9K#3^n` zIj#q#`qNKJerpDjuQw)5UPyVoz1c$|;WYxo<*NE?Rm*MS6<2%Z zGDXHa$cxo?nN(duj_x0%y60&`cCBKHPk5|REh;#|1P`7^#s|Cm$ZL<`d zg8?|t{M1379lyl3Ut-%YvF(@G_6IkRKTeVj$NwO<$uEn-3!UG;EJ{7kz8BjxLKX;E z;8{k?fuN-k-Ke0Z5y!MIi)utJS_aZI-eq)K6Kg!!)_u`#^fi)|pN$zEaK--#jZd$A3~DG;z^feQpH2#^lq7>HdU*vVxe4f15s4Yr{Sq<=4h z$@?@}Qx?k0Vi}~#0-1cgoF@xtPzJW49su$I*oI>uK)a9zu`J9%{8IaNFi{!~yHC3VQc;;mnK)!k3ikLWh)(LWRVZc+uWo_`>N+ z;#8yFV$Iz7xG?R#kiOd&f4p%{m@)4ar!oz|dvkjUHQO}eDqS+1ILnZA0&(mFbK4_B$)_)gK?>LswShL$am__RqSD zH5!y)hxVrU9d5?&j>UM^{fS&7DV4vKbC}P+x{A-ZW2;);$PjjZH|y)>vSe31P8Klbz?>LQ9oaYDVD6!(Y|WSL2aZLbyKDJ1ySVP z?Rrw$p@Sr+bEeeTY7dF`I4!k(zLc~wxIhwac{AaGpUIw+H<@v7?kG-EIS32%?Uz*!zw$$C&N-`p=n4>+tq$!Ui^7gW^ zq*Jw-&)g{nQdTXN| zTql`%xnPazvBd`F$kC3hzg>XT<=!o+!iCY&@SDpdkHooB{hg_jp5r>n{n2&O>+>jP z;(@`EkU^bD8Z9G>44owdQJ>v#J;e9T;UQ9o+on1Q7DOWF(b{NZ(yQtrz--Fs+n=7^O$AJhA9@>tWeD;j$^pCZ;Cc6 zcQ9eTrzML~nM}&W*5%Xom)Q1C+Mb{)|rM1_uFw^{f==~5hKyNOvG`M;*=ru zm)8Qs5@hI>iYC3WK&J~4nmoiCrHxKSb8Q--9aXMyRd&8W6Hmr*-N(#8^WA%KRaMu~ z*qx4?`MMzGn+qk%k_9EG$@NrJ!TvegdZ<8|qfABQb8S?S?S#e<#OYTvKwO_>G%ENd zimK>_Jc4vk{ViwEfXW{c+sqEx>1{?^nMvqNe+Se;w-O4pzl*$tZJbBId?bvrQGT## z%^8e-iyS7JBAbYXh&eou>+QUp^Y{dhz&~}6XU8wG?U&g0OKkfkw*A2k%pFEa2_0}%{f@ze@# z8c#R*byF(Rfe@w<)wEu48m}|D4fVse2G|B+4LsziD5fb-kAr?dnkJq03+mP!4|Rf& zrWN#L&%EEf+kWd4)Ti-kqh&PtR9yV^X#d-t_0PpN5XeAO``)un7R=yy5UfDB0-*}F z;f^hfSn@W+Ac{d55WZj=L^C)}4&NUGc~BMtl!tBk*zfn@7`ZRdhAf~#x?CTWhdekQ z(&T!ftSqd_b<(8r<3LROse^1l{<30QUi~NH6Yi3q_sS>2=Zik#dZS6=&c>I-qWU_* z))5bcYEpsNuVkL^cI^kz!wL2-Kx)%xt1(moxGbf6z9z4YR zw@wt;y-C9Cwrauk<{;r)*;!%yv0Gw=Y!<(%xk5lI#|Z8Cy<+^TkGM`*JRZL>Pl#?b zf}gW}6dxE~0l(dzjXvmK<;HiODUh9E!l_zA@btmdgDzEczVa957UPkZ zU-JRJZ}8|(5&WSQRrt3}Z{z6;rt|YF7U4d}p5iP09`VzjUJ@FwyDW5_?9J_%ohF(N zs#HFA|AJy$L`p5GdB@?5-I9vZ>RMmP)3&F*BNMWilAvAEzA64vcfHlpu*Ux6fOP}1 z?0r0gQ(iLerK{NLM%~n=Qzl7L=4WYqVSUMg875`zOD7(6)l&UyhQu%9y5znPkq&Il zBKwvhj5^Lx@?f)>h=iA9VBJNsBXGf1Z~Idf+`Ghjr2a-h%?s zdH30>m8*gjD@!k`UOVM0F6(qud}N+6n;P7e6rv$naQ+(Eu;3iCzpgJ+dU+A4RO@b0 z>!V*4&(<)ir=d%j?w1ZRdCL+Mob4p0{^Id$@$prRSv?i=*<)AvwEZQv{S&uddDK8` zt0?dM65D=>ZGW&ZlBdU?h;6;Te&!eQiQG11!VMkQkL&-Yj4N4qlJgF|jS~Hk@@e)o zl-@p$OYfYDj@_{2j9RJCf#`Q!;9G=xE!ROY2X#>YYXgzXGB@PCBo1x2P2(zG+lb~z zo1trW^Ej8uce&cZ#whuK7g|zr9*VO#g!X)xi!QaXQ?^@^idxSZggl4qprK!eDl1hj zL3N%UV$Vx%%DIm$k;{2IG;4TQWD{P1qVHco-RA0`E(Y(>)MpbpiwEB5+MZP8W1NXH zdyGS)FO{O*Zs|u2GTXc81S=g6EpH4UH!oh;&ef zigz0CG_7YFl+heV>xXSC`lx3fq=AU2DMz>ASdDO} z73F?C+W&UP{ByBQ_H6rJ7z2?C1S|-$hy?k~LEoS(oC9nFP(RcOG39^hK=XiOGXffYjR2>KrcVAd0PQ12DI9jEl;Mv&gC8n$c3-!^I|NCy+~mXFtQAxRV5zP96cJP8r29?ufJcPNBvZY9En{_lln zUMk`9>NNhHrL|aL=XzmqyE0+^^P&9mi|@GC`#K5hClAHC*I(m~5659EBOhLWaT;FQ zrW5{PpNTtdtIJ&-(SlE#YAN*iq$jLjJXQEQ=e`iR&JHhY^AyK4br4eT`w30HgyTam zN(EJNTf8)-wQxGyjo<#QnETvm7|zUoBHWwS4}TbLkFVBUBseJz1oOqW`Irxp97!@j z=dQKj)K&Ki%?t#=#e602z++J-ai9U!@y@-d%0TNj)3$l~*7f-J}YB7&${&`{ong zGG!MwI;FPhwRHr?YIRXwVrLxd9)#Deu_>Rse>t%&g*bcOk}OZZQOsL?NSbcj zgBjBTGi&dVw6W2xP`b*jB~y-Dps>CEk#R#J*yl8=5y z4>JDzH75JK3;W)9iPT7;uSiPi&o1lcmvH zn6`C>5~fZGxs&OmT3E)AH7`1_R?Wi{RlaqXj9bhnNeerXHkX>KZZ>z4>a42HI(!^Q zS{3wDESR;CBusj#n()#|b@Y;vZ|<75q_je9X4|$ds&_htOrBVklqR=i{D)1I^lZ+O zv`MeX$d4mc*~6czTIAGWOtRWbHC{AUz3rnSx}M_{s(bH=qgj#SblsVX)qB#IG4v~~ zV<#Uc%k$0`PszW_Fp*J=TeTXhJ1@5TF8olact&5VtXSy*ely&bQkP;+sYhu*aiTP&xKiMEyy3UI;rF_mQQ&vE;kUTq_qX9U zw_)9;q;M2W5h<-GZ7A*F>7B*_ln#_mlrEI6l$18;Y{#1F6{7_m;;U5Veyz}RNT;3@|&fsuIu11alDs0PgwQKc7UQ$1$ zj=@H>^HeJ8zUd>HbEG|*=syu%oY9u6YRD=_uP#B#OGsI-B2tdZor`+!EkSKl@{!^A z3(EKFwjrg%86*zkk?X^cXaK2!UQP@|2fVwX?#b!s{WVWzQ3a$tRS%)K)BTX?%Q$4b z;tEo3?8k-;>rJns4sy)why=UmsLGWnG-#y{^6q>axwKFzAM72gJYKOrnldC@SyJ~s z8gJ1JHSC;*{JTv-UC1Sb1A8eS9&v(3?Ts4GT!r|jJy98I z4Q>34L7o+VB&q%SlIZ`x&me#~1e~>iX#`xffSCjuX#tZ7z)$}JU_JqOlNkV03cx1; z05Gcnd>#M*6AQqn0st_#0DLw80MiS=Cj15CJt<&J`cnE)LMh>t{*-}~L6j&;3?+^d zPf4T;^KKrXBEugSP0t z#wU&*tMQxD_7l=HG8?>PO3P@D2e}N6(Y7f|aM0Gz{=oiY=?n~p5e1BdA?5o3=@7t} z$mr2+U5fl{YCyLuYxf~dp8^45C9$0UP|af}a@h*hhGLWU%nfb%2nU=_wvRyIuGgy%jHC zFiPmC9ECqtb4Ops2Z}9>MJ#<=EnIsV&0oEK56`~YQ@H4HLeL3y5Zu zwSqriGS>%>-~JQ_^u2(eBzDEm5*!2@vw>Xf$4_X-kt}XRVS7H;wmEmF&MjWiZwucb zwE+99OvfYI*$C}Qh6xqhSz%+_M%;z%ru@n&Cvo3x^Ki3-0&INqHGj2hBwjS|CN}7_ zLHX@%A^&;Q0ny>O3r@-N=UYzC!fT4sg_%9_M7__c%93N75Q)y^Q}PG!xjiTH=Zp-n zIzv~8diWWis&QAI6#ur58~n?@B$}dno%c)%=$WQ6KAtHh2302SO1hCi7eB?j)IuqL zO_0>R&JZcOjXk;0V;Y%jH$>$xrVvvDU52e=C2iF&lzJ6sNb?WemzsQzmfqU-lMJp{ zOLe^3kd#KP$<62O33FjEsae^EvGGnO+f8RMsmt7UUO61WWF0C|SL$Ie<%LZpGivvj z_AXFL*B?HU#@sz6`99vp*0LWe<;=s<)CbAZtqMyO^P4x3hE>@?%uo0#KAl=X&goy0 zID>Cw#px81^T^X@L992)J*lh0dzbkN>VvG~0#|86wasMQ=QKru_feLg;-J{wP=|!< zyvA5m9?6>6?~!7yu1N-y^oet|f#l7zEN0o_aAITYD%m6l6+1e0lsw!D$!_z{l8;js znK;3Y#NQsHD(W31RjT@mxiI%5`TD9hvo-9H)aLbi-&+f7Fkz2M$?m1~$eJ(y#67>3 z!mfG{>2C8;)opzmo7%91ytr~iHR(wj*;X=F^~LTg8TRsqYQcerq@8K5YI;F7u?@CT zOb@TAI(@mA88df*YWEf=MZ=f5s;$OeiglhR*}4Yz7@hkki2kIbQr$|+iR0Bz#kF%& zRT*(9iiNKSG4BqiDb6GfBhshIBXGTe&;hfvp*fx9$<%DtKLO= zi{2x}8B^ui_QO##Cnai8@4a%NzAu`8xeNtOk5k^-Sb{Q6uR@oru0-AYvB+bzf@^9q zR5|i&8S45qRXOL89kO0>M!BM}qw@5v_sa2}^xL0vpD8zXnht*s%1ud4&lhLd|y&}TbjD0M{l0*)c&qt7U!U@IEYdJ>w@`4bxBm50VMw#t2X z{7{d@H&N(C2_0G75)}_PgwC{g{)hk6{>uu=v*MRf_W#cJ?3YmXZxYJr-=;<+)80OQ z31$BeZXbW@YnB3^D**8be6%#86ZIMUO-!Pql-9osL?_u_OZFiHK}zFSrWL_J=zo~|@@f5WE|3Pt|LhO!e=d~C0+}3xumnLCrC=Wf zCRxyu1tz(ioCiV|2ySp32vrcvb;u&x_ktMI1N(41)CI@NA{gw;_0Y}VwGGF~eS%{l z$Z3!cbwc}c4E=&_$dl`bH2GM%`Ev#N$>J|6ludiNoPMp-&oA}PRbk!O$AaT@rTFUo z5WmXD(=nOl310){7ta46n3H@l@2Hp9qFcOhe)lHv;`$(*k<^#JGPVbvZtEj7TW`;; z9?_T|@TC`?*f0~%>$Foec)w3{Uc6R#J@lavH~5V(@$NRE=Jq#Y1wB7;_PXYLD%%PB zAD+p_n^x!j-Y(=0XxrK0PX%*3uvV1PbJ(#z>y%yK4lEy!rUW?z` z<~pskQHO~7|u@8Vin zRfO%_IKe){TlbFNYq34K8cE;uYY z?7z#S>|Z97C8##NN|$P#>{NX0K%g|L#ucgJ1}oLsfrA*Uf@M;4Ag8uAT`Eb{^SN6hTTq1}Q7@!^vU# zTV;ihhN&J|lqinY^!9Pvoyw#w`K-9OXdZLY?UK5l?rtgV(_5(^!$wk$NSD%PG$-q- zJ(h-t#Y#q9H%ftStH?6vm+GdzXBeB^Na}rEN9psyTk;KmPR2KWO!fxt^yy#Pj-1;) zPPO3PD^-uZrdL`;B=62}vfcX(Y15)JY17k2RWi7eWVn4H<2!nw@7O&V zq-MzwrfB~IY4VgBs{3|LnWKx+*`yH{Rr}rJB#V2WNe6dVQe)^+pZB((ijTW1h{w{0 zs)6&v6tylGk)g&pzRia(QRRMk@2%gXgy?>&?b~6_Ek#S?`HbuQ`LXN;U6gOxaEpdpH9U^MztvTxm{80p7L@7~D@sjD zZHg_W+&@gazGy(xoG7rq1IGt&b^r$laB2WY25?>ghXrs_0LKJ|Qtl%rKbq4tk>&M#-_+J zC|vn0F%@Oqe!!XDxq|pZiOBnqi}FmG4%@Y?pEBXADU)$Dm0l4YRO#F_<+wa+B(AG~ za)xz9=Z7bwJ8uJ()9>Cu-UXM?z#Uec!=sldzxx=}ae^_sr(v9Olc6rcbNef$Byjc6 z1m)y7H)Tdg19ZI64pe=cpYp3`vT~Wv3CL6i72QPe6_A6H)zzu;Tr%4D!|fBbh2; z44vEGoi+T41MsviKmEg#IXsO6@TU*RD+e$G06eJ!U=jdWDF9#|0Anp+Du8NQz-$1f zTEK+JW&}(L_zMAGQULHB0>Hcg)>^>S0JXG$*#W@62mliVfUgk%<_NIUVozhhK}!P~ z1CCn2Q~}Of8qyeW(b9;<0GKhLi54(vfSVRDZ-9puFm-^Z7BG823oT#*0r(>Tz#IZT zTEH{{;0L9UFLbr|$N>WrgTn?!Muo>m#C-pyodjAAt}9$ixQ1}e;5xzefolTS!JGnP z4&w~t3u6jn2;*i)fibENCXvQ4J}@RQ4siZ(-f*sPesE539?);-FZ7Y6APP?rpbiLh z+mF(k(v||J(Vo&#n+JcGfs}5PAWBb4Z%SWEKT0SioYJ2%kTQr8MTw!rQQ|d?{1^TY z1`5KT%1Q|!EB(Enl}1*g@{R2ORPM6`at-y@Qc~Y6jZ8z!Xrv|ZaiWj6koKEjRaYBX z>XD@?kanP~<`qIJ?Z`1mNg7`%ka@I^x0=@xL5c!szQ*s1Zo}hkxuit*HFZ$AO7j{d z)I((^c%894o%Rd*4*MFp367z@UjOZnw?CGP;13)OFa+6XLkvNW0>&B)I0RXKk&QIu zK^X|J4FO^>5@0Ok#n;eu;ZVfRoMr|LW!lxEi;ObW~572H6D=EouV|k=XEl4 zY;zKqpI^+Tw3HUd)+U8IW68$6evGr_dexJVG-i~Oi(+p#%<8>)Ev-6WCK(ydV|N^_ zBQZCfrRG&05TisVGNsKD_UapjlxSflg#|2RtBvR;*+$$YuX|P`ibMy+jA7eJ?BZr> z`*V&`^wZv?%GJ}-v^+zmQN78M#{~z~l2RMCalPJT^Zl8sgkhIS@yWW()*6LHEsmXG z>@A9jX>DuP-q%a2z2YsIJmsOPU;H-7`|&U$G%aI&RPC9K*+yhvy;tPQ40qCM!Wcz? zK~*w!>seCKBh9x@j{>R59Se5by->-#`WIry93l-y4`;4-IZDx~q7D>6vm*B6TQ)2-;;yP9H0Rt4ggR+qfU7)oj{yujEbc@iY( zOEz^+ky^9*GcDS=OP1X?5Nq;UQNciu=+qj;JTOc5O?60TjB1`HCY_tBF5h%1n)dPn z2~uol{BM^j()*oee3$oBB{{Y9ef(yGYTS?tiYd<-)jXvId#v<2llA%;qx(p=e7gRn z#GjOZw+a7Ud-9lp3}h(p{8$Ep&VGOEh1)SaZonf1tm|McC$HaNy#{MESeLjiK5`X1KSqn5K0wB=bWrQZS;(`r8#pC*W9E!$nE1M$w{Ef<0i+7=rJWplh#MVkaLO{6Z3+1kElTcL)f8@Gs zhca-N4(j-MIm$6CL5*YgBfO_bd9w_mx|4RH9o+|`%#KmYPhCo({+}_(v*M3rrC+ZJ z{<*va<`1Z^1xz6Tath$A1xzF0q6N$(&`1lIOaNpT0L&-wZ;_j5J7AIlkahmjuL;ur zx7Xf2I@0a`Ysrb$kw_Uz8BQ5V`SqIMKlGX)NKzmNX{0IeUZRqZM()wdNE*oqJd)to zv#9(d|B{u)O9{NE!25|xUQiCCDb2CqX{42ne)~17zsXk`&nRu(AaBVsmd3k@N>7jm z5A2$9Adf-2;Q0i7)8@;bRPbwBnmpM{iyYIoh@q_925if9 zLw!&e1jvVNNP__7pe_iI2itN!v2m?2-^Kk zIJN2)yu{+UVCB43oO?D?EWW-9x7@GhJ8u~+G*~f4&=09XJ(G^$p|%&qx;1VHUk#1; ziG8o()MRg*ld}-dQ5_MJolgn<4W5b1mEDELdNZ+Jye;3g^8_w)d?UWv^ZSDSoJ?%c z@|vJ~WenOaHs@?sjN{LJzRb_RJsvx)$;DS<97X*Jw!)5gu|k!4Ex8VB4)Fa>9mZ`< zUh%yaf8t-hbHGj4t-&3S8u9yj)D#4#V6k1DkGM)LCjqBN2;Y1oME|l+{F!bG1@mEQ z-kTZDuM|h~EtYpjZ$f6H%?5;PQSl>syYv)S<~5JE=o^4v9}xL%$=~>9cShk2cB^qi zw^gW#jtlBGQORBAAMp8&?D=Yj4!o=SJ|8+sm%nD_g7dAu@h^fF2>uo;g~bLD!f4$M z_}--uT(8Dcyu0W<=T$sFxjp?GU%lioH>*Bf9!8GgXZqjg&AR(?Ik{=PqERHDQ}&cA z-Xvk`mI=7hB7Humw+Y{1i@=9}eZn7bu@+o;8+_VR5Z!hR;ZN+l$yd5^5q~~C1W$DD zDYzQf7FKNy5)3!A7hRH^_>*7Fm9y$u;?x-|UfKuAlj7eclzIHigt7#c@90Fb+AqDh z?}aqUW=l^pso^7{pK7EqeHbK}9G{}r`OsI=Ej>tjcRxg)?1)oI6E+cNrja7&Vc{hiX^Zpjn0%EL~ z5O7#BKXy&g-_?>tTO=vgju|0M+!rBr8qkzYwi!&;1nnlVE=nfi)fiQaoky6_mh^YN zS~Mf6YxK$Wv!8r>ES;@t*vwY3OYB^nX?>BIVKGRmeCG+%%C&v*xG#+eo+lFf_ID)4 zxPtGJo5|9k%k>mb4`z|FWAmf|P3|bN_6}t_E$KsE4{~KTH_0Yd#s;Tmr;i~Y(BT3nVy084ii#>V zs;Oe1Tu_W(|5$Onz>DmAe}fb_%x4y74pg03^ig3vaT)P6Jx@?a(s^KSnid*=aGMY867P@-yM6?4Lf0T4t5vv9gwRLnW&oH1b5QNgd? z<7q$l?l61r?%kQ4aq4+~R(Ex!4!l+M_w7DtMtcfdE1IC|k9kbhypQhpN<^DSUB*pj zH$|`8xZr_rcH%1Cub~O|HltS?R$$%Yd*n|3hk$iq;BIDmx~@Wuf^Q3v2A1E8)0;H3l9(E=Vj z0I2Q&JPd%XmU<)xK#d3NwAhmvaM0pNVgMcr0Hi43pOvFz8OoEC0lc&{Co#}MOG^?1 zt+cczG2pGmhr|HZeI|tJ1Tz9$cevJYec_tI zb%bjN*UOFo*QOz%F#)a#TniTh$Zv4IaGr2(a6WJj&~NB3^br$S0#ES$M>9&&{0IRA zcxKxXIuL>hp@hy_^W9Y&cPD8*2)zh>2>l5C2?GcN34;jHgrS7tguk2@6kH?-|D5*| z$TjlIW*{wTq#^hPLXdvQF_3v=35g~jeFqsw_AzlhwW*@pf>8^p^>Q=V-9Rj0QV%SG@94iXOpT>|`}!yh@!*C2q@0`oUG z25DdfAb=65qy>%vP!94!88`;pa)5dv9Rk!L&k3O(xh$0b#~`->F_eQk1JJgd2I+D+IVMSejUfMlS0Dqyz5cy}+@*@dpHaTEzEtBM7%F`ex_EJ`27B^i z^l(+$)@Ht*<+j|?wF-A;R0vmXWHBd{Oyef4 zU&|*3*HsNTlC1LX?7~O(Ojqf;UT2P(N8&rDkFjqf1a{$#olLu?5zHO~JJz~Pft!Sx zGxmc{u*o%UG0r2MSw+WJc!MKiN*WumPU~y3JEzy@%#SB=LC>c$6Qq7z)h^y#%CG{q zJW$CVKjX>;L^WWZjcd!!Wy{z(mj-jQ)f4%V^EUJId&aZOo!VT$u#SAfkt4kM!$9sJ z$DY`Od6id8m; zEV)~IBe)m8T;-odd2?>-%H^B=dx<*#KJTXl!PsGiaIIdfQjxn;Trek7SaEQK(8j4) z`Fej1QP{s!)MNML?^rRYB#)lIx@h-;^Pkm~QlqHsI2|E9Xc)EF$4okuTqHi4 z+*v$XE*maD7C# z7#dn%k<{t7RLk>-(qe6{n0T@lwe|f8rP1)d!Ut)Q`a$Ygs#?@&^_nfi#0S$3tK}eV=bQokr4IVI2suSBv44Uvo3i3QE zR41a)*vgsIyxikrk#B;!xWxvkjmcc9)q?_xX_rMmDY6otPOGAbYoVu@TYm!;ls;FO zEbXPP>opZFEpk*GZuV7ubH1fAr*b9}@>K&tsH(hEu^63!ChTX;{L@3ISKZrf?L0q-#F*H|yA9nPfizhg|#!EY-Bf%vfcS({^$;U+8 z`dTwQ#&t5f_3SVXd9VfRJlujOz1xVZzn_i=mh{2>HYA~e)0*Jv@o&++#vRb99vg6d z$LqM&=w|rvzIc@KCJp;dxj~+gpE}4-$FB)a2?^wH<%g>yuQlX#0hkj2=J)`t1prtp z0AOZ*ocn{x0btz$fcXJTw174ZfVcsGP7Q$20f1%=fan2OXaNBP_`LK#bCeK19sYh(sV5X+L)k{zxE7X~Zrf zDrp2J5WE2JVS-odi0}lW64U|50T8V~MAFm;`9N?2NM4PY1tJ%u10asUF(N=|`T>HM z_DfJ8q=EmFrafqr^i3mR5#j2)sPr?Fp$^00RRe z4+t=jKmdaRh6MIyQ3_v;*Eema;$J@G2d3@d z_EkH?&!~Nv_pnIh^SsxyS;=*{nOm-~N!$YVtL=04!3rxb+1rTe7!ioyXTIduQEB|M zs*&uGwcfmq)f_%0-kiJNFrGg;?Fu*PXbE3!@NzEDumRKTLNSx$Ofk>v1+nw2j&XHs zWN=x^Ih@;>a`xC2Z4rMvQt%XJ(b@9`@Y2wru3YE6n=Kact~>-h8EYTT~5u z`fvv(jpg+({lbO?PvWxH#c*TZn)02urt-$Nj$HjK;e4J^06H+zgN+_Cn%V0;fwN4v z;VKm`W~(n~%Dj2>1;3v^hb`S+%&vLo&Kd@tW=E7TT+oL^y!e?PqhIQX`?qd}lcrzc zo*1v^JHGG9Kk{&8>$Wi=-ePyzR%_>RlNv4Nc2!N`;;irT!7T=JCmYOU-3?dr+nUzo zeBRq|=9?a|cTQd7#C7_d&%l0cqocN*{lXRU!|{8GZJw><7R^RQ#YXgwj_5m}_t=Q= z(ZfcD$VYTNCQgi#cTD>A9Tm|#CIUXKFerQwUld;z4wALkF+~?w9p$>|n0jrX5a%49rJf+Xr-aql z1?-t4K7DpmY-8S>dg-1ZjPKfn4ry0KN=)w}KI#2Y*=Js?n2R%&=c-oZ>?i^I}}6$z5hh&GZt*!S<>2kcS;;+v=;N z$1c;vkQAQMKS$F=r>&)F9=mDk{ZOf3`XlLvVH?ry!X6>j{RY*y#V5(()O2dc+nvdKy!03Yv?>PGhMaZ5N3B^)6AX-b@qT zCy!7cF;1h;)LTh)-%w3#etQ%Z_TZ+F7MLk1(jE%WA3RkojeRMk)f%kK3#dyEVO~=! zLuyd3hqjSwEbdI3c0VY&pKK$pe_caLzLhGa9GxLf>zGEZ>$*y~z#db+*zG9o&DJDwen2vPI6Z@36ZSRU-w!fvD z)U2r#7=JpaUuWU?rbSez{R`xWpvvE3+rKch$X5--HWPX0Z?WxfvF#t1L}XunJbpuL zi@l9-y%+ZA+JebA{dz9$qlm_>$D3n*fPfE}X5mv4KcZ8gvav(=nK*4k5!UP636%r} z;sIt4@kPT4ND#;4VeCgdsDCnE*sdcU`$@!u$_r3otzOu+N@u+0;zjJ+t0OA2-G>dk zvuK<8DDK|Y2Dhnt6|ZP@7w_*g8>30{P|1`Fc<$C^cxLSqTz#s5lk5BA)nV?~*-Zyq zy^6<9FGiuv(nM6RiGX%H_Cz_gV^Fv9acJZ0V!XVh5$-*@2R5Cw5glHXgY8!2qfaiK zkw7`1ds~iT>o;ZSka+~!x27-d(d95MPFjYWj%e^N{L%d-v&v7$-(uUp?%f3s5Ae6x z_W#$de~WFuVuJsQNwVSiFT^$?Mrl4@t`W+<32q>6$zqu7-33CKMnof?Wgxi80-8n? zBl%?!?z`6(5%InWa70KW0-DBSi-=$Hx3@s#gR(!?`8_{~Z$ub_I%RQ9Q$G>#K%mp) zBXw!SKG+A(Fc8v+7zg#s>EB+Q{8pEyJ*}skMxY~t+uu|BZzJY!i)|oAfgmRHt;{#U z4FoA!!~%gz7OWtTEP{a`_G6IqfQToHT5?_xzd*19;5d|(^OMc*`Jr7%hX8q@90XaQ zgL)wi`V4vHdSo$9PKP$+x?mgn0YKeQ$4?z(1M=q;+o~S)^_#z4waz%tw@$4lD(ba6 ze{Zt4Dm0i?-A}Vp)on7IFIUv%$mo3(7fTryzb`t^1ta(hsWU4m79dJB;g9{|X;a zJc3Uy8OmSx6I2lq@myAq6TEYYk~1FW%V#dmVs8{(=XNKy;DzCNtoKoCuA%2=#(LCi z?wR#`ZrjQM%=)iw@w1c7nGA=??An>Vn7t;w*}FBS@V$ad`9@c-u(#6huw0Y6TttMQ zs?o70?B=TJY>P_q%=;akIppuk<$U?TJ)K#?JvsM)uT}Ddc{_*YFON-NHoc#t>Tf)c z8**|ED_*qZ>>I4-E3XM=Ypxm29TV^W#DOy?8l*h8(3aJDT1_*VKWIRn$hyk6%w z>}ubKOj&pd)A&?lw$A8auEW_;Ovd6IF3o1T%KUIwyruRrcI_tz-1Xip)^W5iQ#2zH zU(W5u85ZiYy_7q?KfAxz$J+mGVq2Vgk>^*DsMJrdOnbGkH?gX_tgc zug=n@TegZPsZCM~i;v>i=uyZ`DiTuiCklZ=XXVX(orOjABb9{%FG_Pp-=p)oEZk!o z<1ZH8nJvC=v0gN+Jb-Ssxz3)t57pA3-ihMa>5oL$?M}+edl(@%(}FHYcu%$Cx9%}N z(OdeQanjqoldja}Ymrcxe_gEIY&%LR&68dyeki;@aGqFWf3To?B3PLm){j1SqfAV? z6Q=xn)k5fU?=)q*pj^29E?jDVXSo<1)?e~_8%Eb2)j)h#GmvV3qODkdq@Eb`?y17r zd#(6p%OHC2nC;@B=)qE)^9JftXo%^y6or#R+CU5@YpN=~h2LadwfpXmH6zc+_iD;m&!k>RtJTKKDis5T*sZ zP!3L{B!{@Av~Pb-cwRQ1;=6^brFusx!@Hu;Xui2}L46gIt=(QA%KA$26~t@fZGVTihz3vxP^c#2)t_q?-0o*RY)dJirz^wvYDd7Dcc&A4}fcJI~tSm^(5IBO0(1zeo2qXj% z+7mhwLI|A*6?(F1>xv+GA_-B1!Gs}%VT2Kc{~)kI{+}txk3)q*S{$~`4nbZz+p+o6 zAT+d9p9aT#+(_6R%Yj8{`FWiuR|$%qWFx5+;Boe zN9xdn-pxm^To^P{YKxSsbx>KH8M;xa2%mN*!rSY8Sig2TUU~I33Y^*!cL`XAeWTvs zpbT?VZ-8bhOr8sb)CpZu2JM_#&lKQ-LikN#EDk)-KnT-(tqWd>1c3=& zTO{6G8V@nz5hi{rKyK_zG?E1vYJ;T!848Y>ANQz*#@yo<7Eb41S9!1({T85mn`yuPzL=W z`%q3JvXOeA45a_8)9{LgFfb5c5Fmizhrt122rw{!!H|K0L^dLj2Mi0OK|Ux4z`l%J z4$|ei$fl+~h@lR-E!c;;<+2dVeSxxIz@R?ZhIRqihQSGr!8t&F2rxjwHUQ~zT~G#& zK|N3g(toNT8;A;oFL)OIs33Q(A`i1T-_#u&_>6sBeJ4jHbCKP8^3yIg<5v{A_>NE% zGsnILazR%{@e{^2XP546%vatM#<$pB?prw`PIYbLe5O`RCnj(l!D^6E2HhpZT$Mbnt|cGr2A zanpE*=i~XyVJWJZj<@mAj!)3N_;|E^r!7-+4~0%on1!RK7BGwTNgL%)*&4mBurD7s z#K!bPbnoynys*kVJlw1`-sEM>CcGWYMz21{Ts<|7x4Rk1Sl&IAOmczAe zZNWZYK)&%iyaY|k{K_O1O~E-;F7amWtJ!zwUUL^M5A%WM+f=;`Z>!vztmB4G%;pVN z+v1oz1x!lAOulx#N$iHx9^4^xpPSgyhR^;qj}xa{RasAQ;|o^yl^=%R&7;gqZt;KW zQ5L7}+h(G0vul92Q_L!)(9u8NOw3bAt@)n}u^F@UV z75-EmYd7`uC;g?V_Z4)PR;6N#%bv)w^){+tRuy5%!qyUU@)J(1*($EMS5)}kNKog5 zkk=C522e#yd=$~82I7=WQA*csabga$P|EPgp+8odA}xpvqn3~9Ax%BL81+BYPz*Vd zCe3T%LD%{`fJ$BeR`jlpXy0z{r4`lF#ADAAy%#bK#ZMobimj5*iG!0%sCqVWg6*2u z6q9y|Qme;FJ+t3Y(KbPJtKrsSt>+!6)^r`o&@D)~wmgIC-R+>d*1ZAZBx zy4hT6(ReTQKG#{|iNF}Kap4^G?C5EV^gFF7#~wOriz>7yerWz&9%TO)tUQQZ)#c0ge^}%qqaj1!2r5Dw zSWS@FUmFLK?I1#XLPx@XDRhxM5rjxW6k#x72w@mu1mSOyi(G_?MJ`l4_#tlA{v$dX zuY-GB^+MTp9_X~iS=`e9EcQND4|hyoh%GN2#Lg)hxMd4H^tyU8yn4byJTZ(znL1Y3 z*!mN8ZI^&PZJ3So+*Rnq`dC+Oj|MUn;rl4hR>N+3>Y%4og2;F!h}i{#a`0pA`Y zURxlrf%v2m#E57G^};k9%EL4q#4e2&8hC1H+k;QrYw9O$!?xyA_aK0Qcr`&aMCUtlTO2bTjU~tQ{P1)5P?Aa0l^7GA`pW>9D)G*5I~rM zJRlZ<$OOkB4a&kkhwdJWxNBf#dQv>_hn< z`vc{m4736HFRM-8h za#fGMW)1otbl<=OLXs+OW zy7c3=ZkoaC7TfbWb=vao_8(OFDf(RLx@UZ#Wg@ry z-G07WLKYv;{*uZ$xQN+3b^!hy?7$55jbK-24aYGifjHf08}7V40EwMnGkvEpTk&aj;1T;NO?Ho}mW_tUor?+S? z)CxJGoSccIGN(*>#_W+&&h?$5MfpHM|CJ;zR+rEwWk%9Lh3y`@oTg&Z!!bgbT1E9= zWhw4R{o>Q@R-7=y^QH3a$d%%u+r@jZSDDmxwLenWZk8tZTrEB^z9GKI@fZ5p+!VIG z((^G6D^fO3HaN3Tg6Qdj}@Ie zr_nmzf;dt7Ah{X~%7d}xlwHY4(V%r7$!uq#V%2LmDsA9E<^20Ll5yH-di32gk-w8J zO>1^cyf*01ZM0T(r z)qAroRbNa~3@cO!do%l}XN{X9=54$t3>LA78uV4H?08%eQ1cDtx!TSr@APG56#nJ= zWBI4@+=}^w#y9zDfuLn7?^G;kK{sJ=hRFjA(DHx{19dfmDWL|zf&gw2wF%(ya3@d@z-<5#7?@O(nA658vfYN@PY5Jb zC}_!(agryD(38-c(3cQSh#*7~q6mWt|4G=A%l}nl`Sw>RV8x)BJ5$ljM*1k&vNbwf zyCq6*UKbs`(HG@s^~Kf)$D*C}+MqRe6j)Jd2yUd?9&PLM0lS=Ahdy^-k1lU?$0`>S z^oXm1x-3dS)i?LUm%6V-`IDdFXR}|Tnw2bZ6`$w0aaKHbFPn~ghp6!QIf-~wXl?XB z#}c=kR}VQh00$2pyskDlN$8*__VL&%ayDL}%0@SDx}vj++IaE&DR^e&KHR=I0qsAi zpfmbmT-EG7u~RZIeT=6^VQFaZF}-2pHM0IWj*Fbx3A>j5wmz>o8P znD4`y008p=m}sd+VxYPfFe`wW7BDdYc>VxjZU9RyV0wUBTEGke@bCa&k^t}&0>C^0 z;57t*sRF=*2&k{cj>Le2mWCt-8fj@vV!%;L6A}Yp;s7Tt&Ljq0v@|0z;Hm{oAmFJ5 z%pvec2~;tpZMe^HZ{dEzeT4f4_Y3Y3+!weHaQ)$$!!?HM3fB^@pB(|N8C)y4K5$Lo zTEO|kIlB|!T;cp$5?T{{2ucD?zyub|E{S<<>`S)&2myq)gm#1ugkVA_p);W?p*!S< zK(>1jdJ*~%`VsmQ1`q}k1`(nOLkYtPBMGAkW51RD>;4ND55jK>QZX72DfoCjyy`~; zB#o#;ystp?(ReZu!Ac|g5K#w2CXGiH$*1v>BI1sw9J~?;o>c%5hBRp)RDl2mACo5{ zmFAU4t>^>-7yu%XMra~aW_XPdrpy}eC?YI@hm~Abe*I8W2WdmDOP(@o#3#}}%}4mj zxybv_M~w&tF{vN=4kDN)jr8qjnKD-_K!E`Oga4049E?7sK@0(m0_@8H3g zmiHkI3=06p1?mCdIJ774!(ahyxTgXLD2@y7PQ3 zKfPtL7uayGqgJSveYRuY-L>GBc->Oft6NhQ>a<<8=k{!7@}2#7@qxZfJe|*##AK=F zHD1T9Y;%ph9eJ2>KHC{TTQ`_nFsY-;*s_4j_PoQ1y*sLu@kjW)^@XYh?^olShbu9) zdQU~Uu{Uv-6QOwQmM#pQ-<>^bFoxUp-k85VC0ONIbp;pgOg^lCID`o(RI`W7M&P?= zH{z%aU7Q>2iYu*++jo(pWMhr&o>-LdrPll53hK}QWeV{=pBPC zZ|z~-b~D_Xue$uv%!ZuntFe6fjBb2jojsi6vkPp>#9>TCV+%I1!94s+<~F?iR59~) z>jZXMgR?lJ%2xEb{u4CY5n+e9eOc34Yq^eTyIBi|r%d4WA=q)}5$lJj^K#xCf+h~3b?lv|@B!e2}q?txnYKj-l?cFm&s zYcK@hI2l z$+<*!6%!3!C7Uwb~bLzC>@fS}qqqM8S(ejMaCeKQE#xcrsi&qL; zd%G)A+TWoZFOQL&m+TVgSto?kdLGKuq;9l$%TLnXR3x3+(hXJf8Guwy9cj0w

J zbfqJk??|(ne-*v)9ranJdq_ikV5f z3SV~zQ%S?x)5{Kic>S&wO%Hb{F(d|BrlJrk)5=MHKN?jj#(}uN=Qq$Zv(WBBV>FRIJ(i>`T zr`H~Ntb976xftncrK~e+hR>84NNV(Qt{8amgjBPdITh<{Bv!rQCq4&S z3Zk0drD{HjSJcn36z#JvQ2ULXm9Z0hDZAG0L8o4##q|q1tJj-17rb-MDwc(pQx;yW zh5SQZq?9IE!h)3A>PL+pDA&!MEI&MdE>Eh7e}q~8-llxbKzcHhcPf^iptJIn0ET!N z!eRJ_;TqP8zl-!l+TjQ)LK}iVA&?M6Xiw-!2qAPLbRl#jR46rR*CbIS&tSq3!Z5-J z!YINR0t~(532}ej)yHq2nfzE(C`nC3{(7I#i5*5LrNJC*l(7#p-BQri>dR5%OjF$2 z@hiTe7{wUct;Hc7&!FcQx})Pb5?#Fe4rThCMThujs1w%5C+tgc@u#iW!Y~adHX4b| zR7I%3HW@vy9FL9?S?j_ybDZKb8Y?Uspbdq6(8_RcY;o}%>bVqSJpMG+38{w4;&X6^ z+5&d+%EMOtYaEufnqJ%yC;1Y|b&`IdRgOfNo zd;wBNZ$RtgFOs{hgR&Ekqmq(+a$6H|cOPuPu4$3BL< zKVy)eieHneCMJ+4>W7O0CPSXZ!%SXYQNW4;08;|M%pZ_f2C&)yz{CKs8UVoD047?% z^Z+1(0APjykVJqQTFgldSZJ{%G4Q`FRoRm^00%7%NDMU8(ul-BV=ay(2AXJTN@BoC zi!+G<7cI?547h4>BQfBv1!OG1QwzvgKnpD(_W-T5fb;`+YXKPuP-p>32%xlpJoL}J zNC?9R+<&;|aG&Ac!u^E%2=@){72F@VC$$Jxgnzq)Me5`Uz63u)00Exmc7zUuU_vON zGeIL=k+?$13JfPXK8O%a7)lsU7)cmS7)uyOh$T#bygy@riwog5rLBpw{G<62705$G z;`-*Nw5bgIus~jdeUPF+8j@utke)PB7LmEYp9%6n3^JJJYgc3+WG=`r`!<1ert!hj z%21Fl%Tb#8z(7@_+&{KfYxdU%T#b?lpH!HCY+O@0{9>&&a#TpJU&uthzks_xGyI z)u&C^3wynJhh2mC{nId8X}1S^Gc*q0-{j4@`!~hc&Xlos-b=W+HHK_*s1G}LsSSJJ z)p*Xr)S3?+)sp*sFPS|O6vZ8E7QyCsyQrG%d==}()Z`=gPvQ0bt$F>S7kTdGMW$8l z4g8t&2ULdbj&dpf=IkwFiQh3KlC$(3!nlq9j4$c!Vh?wCj#nBR^Y#$nwG%lUb)Em*6PnL{E&d-*T=AzEGKir>x|-; z4Eiox{a!+uXAAi;{GSSC356H->Iu{RU3{9}&JhdFtAv)f{MACdTJfT&j$}?_>gqEe zrNdfhu~PRYf`@O4a%28MVf~io%6`u!sm0~7w85G6^pg9%B;V+4O7D(Rv^_XPs?tAQ zSfA@Fwy(O9eAjA{IJUGh)yHw2^62iDV*4(w)ZQo6$^+zkQ>E+oNLi^@(z!X~q|@#4sl$DTDyuxS61Obzpf)xS5$89y*<%*DUfjHIkov*VOX7lz z+ti4_V5J`wE^fWwf^MPq65rRqqujaHr!b9MwMU1Nuf@5jYamS=m>eB=l*Y(uH!F?c_9W%V?w-o$9hb=Ylch08;F zia|TNp=FS$dSWF_zgI5Wj&3bZ?LC`n7O6)yJbhoi*gRcKDCLEY#&^U*kH^#ii)mtG zD?_SpN{Fz*w4oHfW&riXDpIVub&`6Uej4?_sFUFMMiARQ3|6nI>r2f%wxV$Q?5|2o z@0DV$Ic0^pwpS>dK8bXiiHEqtiWah~@C9S4+@KV@T2ZIcrzlRCf$U% zGhO6|=+7mXCH!&5U(>RD)j%?rRwNY=4qU? zdkqpjkE2khJhX4)GF&--D_XR;2YU9YGg|+81!~#M1TBo}g{t(JhjJEl!fvC|(Wy#m zl%4SgPe>kvj+tFXi%M3L`dx6JI~(wb*JilU;SiKPZ9A^s_B{5vT7Y%SkK%@$K2BR_ zhX*G+W47rET&Kli?D1hG`lZYXjnyBH?ijdY>y4kjU7MfxU)gy4nuPZEFGYSceRkyW z0KfzS4qCt*0u8l*X#{?639UJ46ZpNQG#FN3g$OIgzl4-V&f}kw(MZ0(e<_k&1PH$= zsU?8a1TvP!J4=3jkw{P=If0a=k;sTF1|DDFK?Zpt)_9Y_Ylz@=MIM9nrxbVMSP%WLE}BGG+&CGy*^PikZ|QeN}RmZXoe zlqP$Bfm{Z%pC%vF4PI*6kJ*23leD4vwih|qzo+)!o{);AH83C$DiibxFdx+=fD|Pg z7Lc&uH~{ljIfiXGCQDSZdy|}rK(|ypKH9Q&5PkrOePtOrlrw3f-SN3SY@7!Kjwbx?> z`=~`K(?9DHleO>`6Kv+gx9(WV>#X9& zl*Y0}#Vjis8#C>rM{rFGQ~02VmHA2WZMeYM`l@*kD4upX!7k|7f?MEjz`KmR$@!an z;GR%N*=H-}a0BR6-agBQO)RRVG95BOWqm!8&u*N<+9y9_oiBZ0Zp_<@=R4%$?WyZ= zQm=)$^?^_5RM}NrbwD&WPMXSadq?1uu!h*4E)jQPt4jL*TPT)pb`c|XMzW=H#X zypKsdH}AF`S1s9?9oOKMs_xVpDkFX%uOC~&^*c11O?|4yZS$PjwdQtQr_NvaDwlV2 z1_RUBC5mwN)6E&&h0PsUo48Y~RsX8&>Os@lL~jGGv4b)D_RavlvT-SEH?JPE_ogk^ z$iJFO-|8v9Y5yC3+S_Y9?drmt*KaI89KV~`_V4?6ec{HE6N<{iuwZ}pqk8O!-pUD< zUxf5C>y<-mHV~FrpZAIBx0Z5Q8(O$`Qkc+Tk++z#I#ck+F;x3kuj#CfSA=c0>5@h7 z_X?X^OQ@{dIm#vPCkv-d##72G2ZhL5t?A@9FX*#}rid9~+k_`>QFMRT!PJpc_mm0V z%cNEvk0G;DO~npDDL#oO+Ddr^q13h1w~9+`y9-^nVe0MYuF97yzEG8xFO}#@!F1ze z>B5Im$0dICC~9K$+QRi59mU=unNn&-6Z+x)`P8w(6{vCpJ83~>9np6cB}Glo7Mk{2 zMBl&ifcm)ekoxI67dmM8Kx)8};o{Um{iX0RXOdWgHt%jn>_&qcixF9fFhEFqw1 znX=B6Ws?48LuycSXUUqSsVfdMl$piFK8t6Eiq2K<7oPEZrt}I3pt^Q(6Q?z^R&F1> zg|c0$P$+!gD4sJNrO+8q=vfDbD;_`67m{jCF5GsZrEv1_S!w+2+M@9y2jwpAvM{`D zUv-t`n-$4UdBWl)tLfsj7u4#KNYOoXG`&h^h}fK|BRLp9SL}9m5%t5Ti(5{8q&m&X zqwJiE1wU*i&O5$_9(Cz}Q27>waXNxRat{6T?g!2vJ_34eh@di^&0KkI1h(8l+btXS)tjRSL23OO0;{46GmOe zp`azZaBRc6cx6XD-0Iw9?4EZP&9C(syRT`37Ce24P7hv#ZcSDqJ6#ua@ZLxCi^~_h z)OsTde_RjkkaEyIyAHT%oxSKyxQL6d=HcYr=4eV}dAjKkHX+n9Sk1@it(L4F#3 zO*o57Ab%S_TnsP~0IU%JSPuXooB&`t09aedYYlk~^5Z%L))fG(9RL$8U}^yHUjjhS z2f&vJ0Bs)tza{{5e!x--X#N2BI{~$|fbSCk{}%v&9~4kmiw%hZ_&)*wUhxC!Yq29S zV6Vl2#6SZrzn5_4N!kOvwEW&e84MF}-~ZG?8JK%=4*dxOwA%y8_8>wup+dhi$on$| z*)aU3Fc$YstkQn73tl%Q-`0|S!$9x?;MWp0!kN}j3j`<-x-?=F5u(2Pc7g8~*_Zvt zWIr|dt`-PzkPfeHLOKXw8ox4<2I3b8X|kZD$qV^Gv?HRLMr70anQ6o@Z9g3_@QHx_$9*+vljPbndci0sIspeBp*6BF48!-JeOu^$_2f0uVNEMbN> zFJ#TbTe6=%Msd2;vbo6F=h;VT)v?k1R(SrwFjN#1z^KkF;_5}swh^$m7dz3&5CHmuRrI^RZa-wuz4hFl4{NcysOMLYPFpA za<9xb=-|L4wA+E#70zJ|{I>J{@q2jh4)58fZaMgn(GE7yY6kDxCyVdbSmI;#?(oyS zQ`wI#yP$mcP=@k|XH0aKqTQJ+Yni@@Sxr7h-@Nt^&hK7lw%aEKqqAxi`(Z;P-po7` zXMF0z*zTHv3Xb(>O%`5b+xVBVOF~AnA8y>o>n`TAjSe5;Xq$CJ4C}zQ6XWEE;dc|c z{%zm&N+_IuJ4u+)VWYQ~Yq~hOcd($>)*5^ zdxgg#A%$mNURE4%%n~!_-lP&g-j?1+RTXouoRJn&Zd7io%LTdyeMQrE3k7HUN|cHF zT5&=29mVR_m6UTzOoU4Liv;_G5~by^`I7Oh!F0`5o$2F!mI^ruH$>5J@19yOo=C?h zIxE*&XYZ-=fTQE;-4xlk#dO=0XraaWOzOzBlj6?a2b57;Z3^>R9hV$yhe&me{VC7M z`y|WR)kX7L2dUm44vFngT^EmcjaTleHd3g5-k8qI%Ty*k+%C?r>ZM4U@PMk+CrsSv zpjUY7{d?N$=v=W~j0L^0&_vl?Xh!S2l7x+=;Sw{k4qaJqn3QO*Cms%dC54U{K;`N` zrflp2q{`L#U*N` z@Rm}FM+)i-YmM}%$nMV-hIj~-Y_?Y@$vmVS%Iu&Gj?55!9gZmsJ^CpRWQ~xCHzq) zU)!#H)j;^NkasE;zM#7>L4b(^4C37!PopP``xZld#f}oI@H*$4CY3`wD*WIo3*|?ABa_>O2^sFZ} zt@CQSUn5(zrf4lT?lB#i?|qG(jc231Uc<1v>KvM1Z!>mVeH}Haeh&$a>!Pn7199jz zOB^@f9qXn75>dRY<^yibO=cf+x)A4J<*xw#y ze|wbaR38}?IdF9E@PSd${YOR&4Qm+PKz?RmSozZnW#k_}in5 zTm%TeDV!z9AEnoPPmBmpninA9*Axlhk)=tOJ-uL`2xcI70U)yd=s5<06r_VtCW}-W zu}o8s_N7nq5~s#%41_T7A|qlQpm}i;>LkLO#zPFsXuZO|_lfmjFWkVp1pgL-6f zP2*9f73M&Y(+Gj^5+?{}&>rcV+{WKi`)^N3#X=ew5D>4xh`|4($U+kcRRGLa<#fm+ z&tD-1U``8jS(wwxKsuz!b-}irpKN|_4~|0_s2_j;`5?%1WjU7n0A&D32O!Ax$ZbNJ zETF+TKt7mf1JEAigCLjxw)YeMD~}=-2y5_c{E@hCfx!LRn-lq%Zbt*_tYP_BO247SPE_U!Dg=lKb?$5fS+kN9&7tyMe9+pDe_bZ7UJ&tnsQ zF=E=z%VnO{J&di?dw$q!oKu)2~-p$B#|%{bB2vX4iA^gdUd}osyb( z?)W^GdXj(&KlDL2>=xm%`DJ)l@ijcZ%N11qQpu$0yu+h6A4R)@n&QegC!*w$W+>%g zAydzKJd-o-6cgp0!lXwBvv1BQxMvIQvpuKoU~J<*vPlbDvWN1vb5%1_*rtnjGX8$o z_?g{DvL1y`na!P&nS~=G81>*MoY$L;yrXG%j%_fCRgB)xU0!sSjaa{mD|TAQK3Lp| zTQOu3SIsm4AJ+3^N3GQ5r!HN>Et{-VO|f15|yn8FCJvp%=1W~tvVwoofB=qvpVKdPe( zPKYKCKML~#4;8p2J1N$?I8cs#8wwfm_mykwnA3U9JBcn9tE6-*M!+ZXseHdrh2iHG ziEj4imHw6D3NupXQRfspwjavk(-@bjL zPCNb!UH^<7UFH4`Av$uMxJi{QEa@{^TJ}T|;@i@~&eiv*PUkw)Ij2>U|IQ`Ug#oXm zROj)OHWJ|7CM=_1s=H#e; zb?znAPw%78p6n08pvOa$#pdzilVd^R%%D(ZS@v2 z?i;rYPYVJSJ9`<>gR`QAV@)P2H>WG8tXGtBtIY%IO71wR>gV@LJ@QRA^UoiIHT$~= zGnR--N6(3L`}_IgjVm35vM(2vOF!r-u55cE)y{rFIZfT5_;`AwIOj@N8ZF)_R`O^= zUH!CFylz&NvN`5R&?_EoNA-h`A%jSk-u&bKIXZB>^+Ij+4I&Fo=btEr7 z9?ZsWrYdxMizzAyJ&OFSU*PeGk_O^qyGn57Sw+}y(pB8`f+3#WIUDl_@8eFifO7X;!q2vD z#a?wfptIrd^y(nXv5nAKbrGIYeIU9w$pg~<%0a&Czb18!Ng&Vr4<`?Q@&K$lxCnb4KcozR2Oi_nMAkI&2t!*MzcP@uNFRTeiEzb| z6a2x0AqT)98);)O_CJP7WCxBx41jTm0EQpRfpM4fKv_5r$lFjB3<~6dARhzc0%;IK zUg!q|s1Ja2xnHmk?SXNE@MGIhKkP%0%Rw4!L-?^Dkp7nr@*jK!Qqmv0(5Og28sj^2 zMqABTbj(wS+JZjzJ8crkrBuqyEclA92bk5>^#KhSRGgGEvd?N zTlSRmc{_}sadH8FDDesoYu<}{c%(XeYCs|vY}A7b-+7#CHDwxKyZ#iu=&BVPGUYYL zOo`$0jui2WJ47?ORbo)NK`~>oA(U-hcAiUIu$xOFTEXO*Nqwt~NL~M1^o?p=V z6uWrWK4!s~O>9=V5wpu>p0f`<570Un|JGJVfDm)mrlOv!LkA9@5*lLlJV6zBJ~r6os0$*IdGD%MS3kv+kZRxR2k)-hPB?i-XRj(A^6n|9qU z^||o>+B+AxnzFu+OU0-ZDHVxGZlMgN&e?mdeK|+DBoYz2sU#_ePwvL0hFmi)<(4}S#x3bx|1)PBGw+z0=Xu_b=N)E!KHt{f>%P|B{r6hG&)WM$ zP#q&5s5aOAfvR6(D10^2NopuFR*gD(nwm)sQ_ep#MkuYmC*13=q%vAA7RrsnXalG5 zLjU5g=&aE#CBpYx>76%E2+y9ZrPjSDp&W|((;g9Rlt`~k9ldwCFmAh2oy1L1_uM~- zT2wGy$l7z4%Dr4mx2Bp3rB5~qDQ8bA+s#`;RdeP-A?B3ZswYdGH*sq7uC>&{X)BfO z7VZ)zZ5${$8t_bcbx^tBWVnQCVp&BMWFqD1?@TD;%w586yMxl(P4}v)P5lL<4Jt|T zpY#o8QU;^*CRRAQ z*F97@vlVXAbrf=qZHAVQxs84-G{vSmk8z25m!^?lz#sAI|CVI*SNU~CFgXCo7XX+a z;6Hs`5KI=(S__yj0HhEAOc?;O2+&STd*TCN-hhr;z|;W_TEOf9Afo_a0s)Xz05FFD z$SVMtMgXK10L&!dtOZOa&{GSTPvGB`j^asPSjoZa3|4xu%7WFK5y6(7*<1ijD2GD2dC-e>a1O0%uzn6}*=`jgo2|Phga3}O8z`g59@FEN(3?d9B_!9n< zbVSmPBSaGvgjm97|Er3k>4(7s_q!5OJjgB@iG|2LAjg1=1CowLTGH$;^0C%PJVXN0 z>|7$dk~HaIrx3_Ou=9!RL;`6?vm*+o%tTrezk8*TcZhTZa*igyR!Y*ugPcRA%$m9& zjdoX(raq9VK=J{ZibzG8G+NmRUQ+}KOCv9lFijk^1JXhna7@BAJF7^!;A(#DQUfc_gHVW3g<= z1Lpw5frI>T4rPE3^@(|*EJFgM5sx7r9E3xd*e*DRGN4>IhBBZ$@S)9O969{_ooE0) zA}NX2{!GEO%92P4LMo(H2kk;}$DD&bt6)cBI+P&7Ah4Vyf766bDk zo-ZA@RpvEq0=Ir%DQ7cRDGL%(WednIp}KAv{7zhh{pSVYV}mQ18M7q#mem$!K;!)^ z-ZY$j9QvG{{&o&t;CTz1EU01~<P zAIR>18p~GP64;($db5__hOU@RXzO z>@4ia-CR71U7D@rmuL3n_pjZ=*_?@Cv({f>Jxm+0Ze>@Qjh7g9S5#-VizSQgp1$T5 zyJhp8laBM-_YB|{Z{N)-mOteq5@yO;^`F6~z6xW5$_Dd=$KtuL1C6+%7@Du^{F1wF zznkmP_c*ukySw7W|5Uq<{_lQWF|o+?+&tB?%z=CB8cBuDO{P-v#gkO+&$M>yM42g9 z4^0xDhWXO%9O?@TnuQ3KFA9}gJGYcHy(3i?uR(OvtmlI9L{w}RI!MS~7B0Q>I+cFt zlCHLhHAK<@-zw+%ey?u*Vi)E7xU;IslyyaiPQ^;DOJ1syqZHB=X-Cwj8#~eqTg?^* zKmD5SVBVfu#CjEXx7bA)WW7;$xObcO=C7!850xO-!7f6Co3%RW@J*`qtpa-PIS=Ka z+FR;vTOLT6z&GmB{=)>HsswdWo5s}IJReE)xSc}%$mJ5(zWY>OFJhIZISZx91sQ_X z$R$$UF{_mA94=BxgVTkEb8o8KTUn_%<5Ifb@i^&~m>*R~rr4_;uoZO>uMn=E+OMA3 zd?&55Z>Z|Q5LWHHznLoYNfrII<1p1f9`}%*f0Clc5pSuZ@BT@vSc(o_6et{Ax>q=! z|CZX3FwtPyORY*Bm1R?zK-HCELWq)C&WY*qK@=PYp5oOIyP6T-ewPF0gwsdnFf zM%A$F2IXCDT6FSpKWXWVOv!|rN>wf1p{!iPOWN1HS@d}1MnUhvGs!CF*~+|=>C)c% z^_4vupHd}6-jd3{_ExPZ&y(s64-@R1S4yUZ^p}P{DW}RK`n|tB|574T;-@y@@9PuC z3uLCI;>pKlCTJ);hQ&uYJeFZv0FUv81QSAILK6b4_bdojgpbNhB(4Ji)?Bd8g0&T_ zr(g{Q>n0ZhtdBGj6Y+l`FKO2vy-2tRp%0-Sp+8{&!JFVi7$VN6$?;Iaza=%1eDPvA zpSv&K1|OB15@~LY9QoU>rj7bVp)3O*^wxVd-64Jt?U+9RmGo6m3rFhE@gjz~1)*vHbp4w087i$h zfL9fmT9k$5t5@U5 zZHd?_{Gn3+xE`Grx?O zNYcUehHDJh6|N;*Ke%pio#6Vw^?>UD;|^mDV+>W`AI7Wfz0&N3wBzmikw3ljT}Vk(8y3ANrAj2?vNsrYmk;;rxO5@ z5~SD2QBWuOd_1HTC)Z*e)GzLi(nwuI!h$x6yRI}}*MfGDwn92hc_4FrVUz2R%TZtu zz=06}M8jTJd;Z~r0*nO!4+3~tfP?f92Ir6#9vW~CK9m7zMMDGUVtJ4c&LIwfV*ngH zNFW~oh6?hCbwE90I&%1nI7km^Aq^ahY2g_924w(nF8bu~^LL^__=qee-UFY0AjF1! zOqfcPr)Bx@2GTTnGC!Ux+1rO}I5msU9{o=Kjo$$-bx1Z>@8Bpl(_tl>T9(1l_U8Pw zbrQMz)VH!eQI_o2zEUo%yfxFFYRbNv+LrZrQ_ke84zfw^-{HLFK3vndMy$;tGj?uu z7C(NYuIz{8MzWm=RWjpzUp~fAhcTG8f^D?>9Osk2fteqC39XJk!kBe=#|DpxVrQR< zWb0>|Ff-3(F+ryG_)+OW)^6NJW{>O?o_1^pTGo3y7G76yjb<@ScdJn5#y4?z*>aYh z!IbefQG2*tMV4%3`h5P`t!kOuR12A-VmH6EaT-6PmmS|)Kb^}lxQ&PGiNzc3YZlYWn_{(CMRg;79 zUEdvy*}Kz>(RLTSNw|y(eX8(*L=U!cxe;EJe+Mrl-?Uyh;VjPRB1gqtMziMrUhMf{ z<+$)t9Ny(SgxS0z4qr>q#RFDNLnpfFGtW+2vqsCB^2U{^Y{K;$T<^y_IGrC8_)yyj z*|cM$_?wo|?Awd^eC{J>zSQJ3ze8Qb=iBe(XWZgAry$G?9h}U2^~_~c+VtZaRoZb& zy`q_T&&%RX@TmlnGW4c3o(BBdDkZ&6$w+h34Fs1Vwoe>)9bd;W+vrCm&Z-!+228Gmm&wX`^qO)r2!j;JWWGb~P ze7<@@Qkd%4@T1a-Zv&`(wVhPe=Spd%MXp2=k*`X+&{|>>RU{Pd>nUB{dAAUg(vY^I zU8PGF8LMtJeI*FTt*N{l_o#v7)jGSZXm!&1MM9U9W`g&DXVkEi7u1|od)3#~2~_4h zbJgB$ro#0_Wh#Hq7eb$7snTy;rRwJ8>4ICc8uhG+70Lx$i)q9BL&A)e_f>WGrU~kJ ziFDSReXbL;a+IY__DBvbTPWF7U{F+Aubpb%?RC;tltNXu;H;2eryDiP&03f{q5(D8 zem@;~*@K#QXp%JBSuQ*&TqwPC-$G&>QzGbQ&yjxLZ?W?6i(>I+`D^`p;eX$zgFs~= z4&MJ$feKavf*_K-< zF8M8;TvP|^IWNFE0h3S<7lh1Dd*e`(PVyX%l*wuwoNtc;g95P0-AQQsp#W6y;0pfgdI#Dr*@fK4o1lx^XX2C= z!*N{H9kjIlx8zFHK}$9)#5#90$MoB>JX90tfiGpGV(Ydo=tD{0K^ly51L$KGr_kdy z2hg*nW3b}&5gfs}ptGaO(AiK2Y;mJIHr(QcQl^r2UB$>{Xaef7uL_5d6MCrFmrz_B;{s1_UJ$jnD(a6bL&+L;_I<#3KmT>_sABAW&)cDv@*`25H1D zcvX+=W6}VxxRE#zg*1Dqw6Eo9Ue6<6nS#l+MuZ~56HKn*bvP&=ehW|(r(kjo!j#BQ zb^r8*C`}tlJ49g(_F`$i0wwPAB5_bZlm~mLG=iG8J&^88n_T~~KqU%GAl`_G0uP8e zqR0f|5rl(q1YsaFLD;VaCJ=-G5R4!W%7lE7R?GuFI4BF^zOs4`=Xmj6As)cfMlG z7KgI>4HTT4$u{o3Y6usS)tWtX!Bb`!RwHkG^#?xPZ30^nlgFDM9L!g2-irqw%Vw^G zU&TL0&Bk^Y>hQfhv$^?;>amL0Dcr^E1Z=Cj317^6!Cbqz3FqCZViyFqXDih`IG6rm zT&CGtuIb%}>M%2 z+1jLhwpFKBT!M`U>#NLX_fODerukX16LsBjvzDFN`L^enlc%2J&3c`D)us`Mp+V8Wa^36YC$Uh>G+NoDiZ24Es5Q&ihmi#>GYY2agSmiFw}~9Y?WYh7cJV z5*QozL1lHFx{2`y5FZ<;2n&h*Ik(gMel&;*8WR#68$KcAed~YOag(6%ae<0Nx3N)S z;W4q{!D6|_ijXk>u~9KGzv_FVnBc&%A^y=(G5+x(@V(LxYn}YQ-uk4sG2v0;{Mv{G N>W0@1*9q5~{6Chflf(c3 From e75347aa800aac669cf19343a6e59331f2074dee Mon Sep 17 00:00:00 2001 From: Beck <164545837+validbeck@users.noreply.github.com> Date: Fri, 28 Mar 2025 10:13:53 -0700 Subject: [PATCH 90/90] 2.8.14 --- pyproject.toml | 2 +- validmind/__version__.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/pyproject.toml b/pyproject.toml index 3ea1b9f93..f0fe307de 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -10,7 +10,7 @@ description = "ValidMind Library" license = "Commercial License" name = "validmind" readme = "README.pypi.md" -version = "2.8.13" +version = "2.8.14" [tool.poetry.dependencies] aiohttp = {extras = ["speedups"], version = "*"} diff --git a/validmind/__version__.py b/validmind/__version__.py index 76aba1042..781278681 100644 --- a/validmind/__version__.py +++ b/validmind/__version__.py @@ -1 +1 @@ -__version__ = "2.8.13" +__version__ = "2.8.14"