From 2c1c02ddd7f94fb8430a407528e7447af0330af0 Mon Sep 17 00:00:00 2001 From: Shashank Verma Date: Tue, 20 May 2025 14:06:52 -0700 Subject: [PATCH 1/2] Update links in docs Signed-off-by: Shashank Verma --- .../tool-calling/2_finetuning_and_inference.ipynb | 12 ++++++------ .../tool-calling/3_model_evaluation.ipynb | 4 ++-- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb b/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb index 81bb95ef0..498de8d13 100644 --- a/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb +++ b/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb @@ -137,7 +137,7 @@ "source": [ "### Resource Organization Using Namespace\n", "\n", - "You can use a [namespace](https://developer.nvidia.com/docs/nemo-microservices/manage-entities/namespaces/index.html) to isolate and organize the artifacts in this tutorial." + "You can use a [namespace](https://docs.nvidia.com/nemo/microservices/latest/manage-entities/namespaces/index.html) to isolate and organize the artifacts in this tutorial." ] }, { @@ -191,7 +191,7 @@ "source": [ "#### Verify Namespaces\n", "\n", - "The following [Data Store API](https://developer.nvidia.com/docs/nemo-microservices/api/datastore.html) and [Entity Store API](https://developer.nvidia.com/docs/nemo-microservices/api/entity-store.html) list the namespace created in the previous cell." + "The following [Data Store API](https://docs.nvidia.com/nemo/microservices/latest/api/datastore.html) and [Entity Store API](https://docs.nvidia.com/nemo/microservices/latest/api/entity-store.html) list the namespace created in the previous cell." ] }, { @@ -252,7 +252,7 @@ "\n", "**Note that this step does not interact with Hugging Face at all, it just uses the client library to interact with NeMo Data Store.** This is in comparison to the previous notebook, where we used the `load_dataset` API to download the xLAM dataset from Hugging Face's repository.\n", "\n", - "More information can be found in [documentation](https://developer.nvidia.com/docs/nemo-microservices/manage-entities/tutorials/manage-dataset-files.html#set-up-hugging-face-client)" + "More information can be found in [documentation](https://docs.nvidia.com/nemo/microservices/latest/manage-entities/tutorials/manage-dataset-files.html#set-up-hugging-face-client-with-nemo-data-store)" ] }, { @@ -313,7 +313,7 @@ "id": "97ac352a-31b9-4144-ad0f-699fcceebfc2", "metadata": {}, "source": [ - "Next, creating a dataset programmatically requires two steps: uploading and registration. More information can be found in [documentation](https://developer.nvidia.com/docs/nemo-microservices/manage-entities/datasets/create-dataset.html#how-to-create-a-dataset)." + "Next, creating a dataset programmatically requires two steps: uploading and registration. More information can be found in [documentation](https://docs.nvidia.com/nemo/microservices/latest/manage-entities/datasets/create-dataset.html)." ] }, { @@ -1005,7 +1005,7 @@ "### 2.3 Validate Availability of Custom Model\n", "The following NeMo Entity Store API should display the model when the training job is complete.\n", "The list below shows all models filtered by your namespace and sorted by the latest first.\n", - "For more information about this API, see the [NeMo Entity Store API reference](https://developer.nvidia.com/docs/nemo-microservices/api/entity-store.html).\n", + "For more information about this API, see the [NeMo Entity Store API reference](https://docs.nvidia.com/nemo/microservices/latest/api/entity-store.html).\n", "With the following code, you can find all customized models, including the one trained in the previous cells.\n", "Look for the `name` fields in the output, which should match your `CUSTOMIZED_MODEL`." ] @@ -1384,7 +1384,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.12.3" + "version": "3.10.12" } }, "nbformat": 4, diff --git a/nemo/data-flywheel/tool-calling/3_model_evaluation.ipynb b/nemo/data-flywheel/tool-calling/3_model_evaluation.ipynb index f526fa681..5c65c9bde 100644 --- a/nemo/data-flywheel/tool-calling/3_model_evaluation.ipynb +++ b/nemo/data-flywheel/tool-calling/3_model_evaluation.ipynb @@ -194,7 +194,7 @@ "metadata": {}, "source": [ "### 1.1: Create an Evaluation Config Object\n", - "Create an evaluation configuration object for NeMo Evaluator. For more information on various parameters, refer to the [NeMo Evaluator configuration](https://developer.nvidia.com/docs/nemo-microservices/evaluate/evaluation-configs.html) in the NeMo microservices documentation.\n", + "Create an evaluation configuration object for NeMo Evaluator. For more information on various parameters, refer to the [NeMo Evaluator configuration](https://docs.nvidia.com/nemo/microservices/latest/evaluate/evaluation-configs.html) in the NeMo microservices documentation.\n", "\n", "\n", "* The `tasks.custom-tool-calling.dataset.files_url` is used to indicate which test file to use. Note that it's required to upload this to the NeMo Data Store and register with Entity store before using.\n", @@ -875,7 +875,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.12.3" + "version": "3.10.12" } }, "nbformat": 4, From e1052795e4096328360dd331553b90af672ba9a3 Mon Sep 17 00:00:00 2001 From: Shashank Verma Date: Fri, 30 May 2025 00:38:28 -0700 Subject: [PATCH 2/2] Add model version in customization job config Signed-off-by: Shashank Verma --- .../2_finetuning_and_inference.ipynb | 319 +++++++++--------- nemo/data-flywheel/tool-calling/config.py | 1 + 2 files changed, 166 insertions(+), 154 deletions(-) diff --git a/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb b/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb index 498de8d13..3b0a741a7 100644 --- a/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb +++ b/nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb @@ -76,7 +76,7 @@ "Entity Store, Customizer, Evaluator endpoint: http://nemo.test\n", "NIM endpoint: http://nim.test\n", "Namespace: xlam-tutorial-ns\n", - "Base Model for Customization: meta/llama-3.2-1b-instruct\n" + "Base Model for Customization: meta/llama-3.2-1b-instruct@v1.0.0+A100\n" ] } ], @@ -87,7 +87,7 @@ "print(f\"Entity Store, Customizer, Evaluator endpoint: {NEMO_URL}\")\n", "print(f\"NIM endpoint: {NIM_URL}\")\n", "print(f\"Namespace: {NMS_NAMESPACE}\")\n", - "print(f\"Base Model for Customization: {BASE_MODEL}\")" + "print(f\"Base Model for Customization: {BASE_MODEL}@{BASE_MODEL_VERSION}\")" ] }, { @@ -538,45 +538,54 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 4, "id": "a2e9678c-2785-4e95-b11b-1f41067bc920", "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "{'id': 'cust-BTkGbfifLfEAjV2THu3tas',\n", - " 'created_at': '2025-04-08T07:05:52.236823',\n", - " 'updated_at': '2025-04-08T07:05:52.236829',\n", + "{'id': 'cust-6wUQyBcweaEvBVdUXdTNAM',\n", + " 'created_at': '2025-05-30T07:12:36.662938',\n", + " 'updated_at': '2025-05-30T07:12:36.662940',\n", " 'namespace': 'default',\n", " 'dataset': 'xlam-tutorial-ns/xlam-ft-dataset',\n", - " 'output_model': 'xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-BTkGbfifLfEAjV2THu3tas',\n", - " 'config': {'base_model': 'meta/llama-3.2-1b-instruct',\n", - " 'precision': 'bf16-mixed',\n", - " 'num_gpus': 1,\n", - " 'num_nodes': 1,\n", - " 'micro_batch_size': 1,\n", - " 'tensor_parallel_size': 1,\n", - " 'max_seq_length': 4096,\n", - " 'prompt_template': '{prompt} {completion}'},\n", + " 'output_model': 'xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-6wUQyBcweaEvBVdUXdTNAM',\n", + " 'config': 'meta/llama-3.2-1b-instruct@v1.0.0+A100',\n", " 'hyperparameters': {'finetuning_type': 'lora',\n", " 'training_type': 'sft',\n", " 'batch_size': 16,\n", " 'epochs': 2,\n", " 'learning_rate': 0.0001,\n", - " 'lora': {'adapter_dim': 32, 'alpha': 16, 'adapter_dropout': 0.1},\n", + " 'lora': {'adapter_dim': 32,\n", + " 'alpha': 16,\n", + " 'adapter_dropout': 0.1,\n", + " 'target_modules': None},\n", " 'sequence_packing_enabled': False},\n", " 'status': 'created',\n", - " 'status_details': {'created_at': '2025-04-08T07:05:53.328702',\n", - " 'updated_at': '2025-04-08T07:05:53.328702',\n", + " 'status_details': {'created_at': '2025-05-30T07:12:40.122576',\n", + " 'updated_at': '2025-05-30T07:12:40.122576',\n", + " 'elapsed_time': 0.0,\n", " 'steps_completed': 0,\n", " 'epochs_completed': 0,\n", " 'percentage_done': 0.0,\n", - " 'status_logs': [{'updated_at': '2025-04-08T07:05:53.328702',\n", - " 'message': 'created'}]}}" + " 'status_logs': [{'updated_at': '2025-05-30T07:12:40.122576',\n", + " 'message': 'created'}]},\n", + " 'config_snapshot': {'base_model': 'meta/llama-3.2-1b-instruct',\n", + " 'precision': 'bf16-mixed',\n", + " 'training_option': {'training_type': 'sft',\n", + " 'finetuning_type': 'lora',\n", + " 'num_gpus': 1,\n", + " 'num_nodes': 1,\n", + " 'tensor_parallel_size': 1,\n", + " 'pipeline_parallel_size': 1,\n", + " 'use_sequence_parallel': False,\n", + " 'micro_batch_size': 1},\n", + " 'max_seq_length': 4096,\n", + " 'prompt_template': '{prompt} {completion}'}}" ] }, - "execution_count": 14, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -587,7 +596,7 @@ "training_params = {\n", " \"name\": \"llama-3.2-1b-xlam-ft\",\n", " \"output_model\": f\"{NMS_NAMESPACE}/llama-3.2-1b-xlam-run1\",\n", - " \"config\": BASE_MODEL,\n", + " \"config\": f\"{BASE_MODEL}@{BASE_MODEL_VERSION}\",\n", " \"dataset\": {\"name\": DATASET_NAME, \"namespace\" : NMS_NAMESPACE},\n", " \"hyperparameters\": {\n", " \"training_type\": \"sft\",\n", @@ -617,7 +626,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 5, "id": "57eb5ae6-9b3e-4915-8242-34b65b0c0680", "metadata": {}, "outputs": [], @@ -657,7 +666,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 14, "id": "900f28fb-fb8e-4d57-88d7-6c09699523e2", "metadata": { "scrolled": true @@ -668,15 +677,17 @@ "output_type": "stream", "text": [ "Response JSON: {\n", - " \"created_at\": \"2025-04-08T07:05:53.328702\",\n", - " \"updated_at\": \"2025-04-08T07:26:36.749083\",\n", + " \"created_at\": \"2025-05-30T07:12:40.122576\",\n", + " \"updated_at\": \"2025-05-30T07:28:23.111928\",\n", + " \"elapsed_time\": 0.0,\n", " \"status\": \"completed\",\n", + " \"steps_per_epoch\": 219,\n", " \"steps_completed\": 438,\n", " \"epochs_completed\": 2,\n", " \"percentage_done\": 100.0,\n", - " \"best_epoch\": 1,\n", - " \"train_loss\": 0.052824027836322784,\n", - " \"val_loss\": 0.047389596700668335,\n", + " \"best_epoch\": 2,\n", + " \"train_loss\": 0.005034453235566616,\n", + " \"val_loss\": 0.04448993504047394,\n", " \"metrics\": {\n", " \"keys\": [\n", " \"train_loss\",\n", @@ -685,293 +696,283 @@ " \"metrics\": {\n", " \"train_loss\": [\n", " {\n", - " \"value\": 1.7395085096359253,\n", + " \"value\": 1.8326354026794434,\n", " \"step\": 9,\n", - " \"timestamp\": \"2025-04-08T07:07:50.318152\"\n", + " \"timestamp\": \"2025-05-30T07:14:29.103653\"\n", " },\n", " {\n", - " \"value\": 0.5097759962081909,\n", + " \"value\": 0.537447452545166,\n", " \"step\": 19,\n", - " \"timestamp\": \"2025-04-08T07:08:21.445975\"\n", + " \"timestamp\": \"2025-05-30T07:14:45.813166\"\n", " },\n", " {\n", - " \"value\": 0.11785753816366196,\n", + " \"value\": 0.11371524631977081,\n", " \"step\": 29,\n", - " \"timestamp\": \"2025-04-08T07:08:45.141126\"\n", + " \"timestamp\": \"2025-05-30T07:15:02.989690\"\n", " },\n", " {\n", - " \"value\": 0.044126059859991074,\n", + " \"value\": 0.042690254747867584,\n", " \"step\": 39,\n", - " \"timestamp\": \"2025-04-08T07:09:08.064725\"\n", + " \"timestamp\": \"2025-05-30T07:15:19.740801\"\n", " },\n", " {\n", - " \"value\": 0.16011953353881836,\n", + " \"value\": 0.17299434542655945,\n", " \"step\": 49,\n", - " \"timestamp\": \"2025-04-08T07:09:27.849022\"\n", + " \"timestamp\": \"2025-05-30T07:15:37.660643\"\n", " },\n", " {\n", - " \"value\": 0.060752492398023605,\n", + " \"value\": 0.06746871024370193,\n", " \"step\": 59,\n", - " \"timestamp\": \"2025-04-08T07:09:49.829851\"\n", + " \"timestamp\": \"2025-05-30T07:15:55.100014\"\n", " },\n", " {\n", - " \"value\": 0.061382267624139786,\n", + " \"value\": 0.06935714185237885,\n", " \"step\": 69,\n", - " \"timestamp\": \"2025-04-08T07:10:07.325047\"\n", + " \"timestamp\": \"2025-05-30T07:16:12.147909\"\n", " },\n", " {\n", - " \"value\": 0.01653873734176159,\n", + " \"value\": 0.01449008472263813,\n", " \"step\": 79,\n", - " \"timestamp\": \"2025-04-08T07:10:26.559717\"\n", + " \"timestamp\": \"2025-05-30T07:16:29.190033\"\n", " },\n", " {\n", - " \"value\": 0.06011583283543587,\n", + " \"value\": 0.06649354100227356,\n", " \"step\": 89,\n", - " \"timestamp\": \"2025-04-08T07:10:42.204168\"\n", + " \"timestamp\": \"2025-05-30T07:16:46.253256\"\n", " },\n", " {\n", - " \"value\": 0.06969955563545227,\n", + " \"value\": 0.0677015632390976,\n", " \"step\": 99,\n", - " \"timestamp\": \"2025-04-08T07:10:58.948349\"\n", + " \"timestamp\": \"2025-05-30T07:17:04.440599\"\n", " },\n", " {\n", - " \"value\": 0.08384404331445694,\n", + " \"value\": 0.07733975350856781,\n", " \"step\": 109,\n", - " \"timestamp\": \"2025-04-08T07:11:26.313792\"\n", + " \"timestamp\": \"2025-05-30T07:17:21.098431\"\n", " },\n", " {\n", - " \"value\": 0.06991764903068542,\n", + " \"value\": 0.06474778056144714,\n", " \"step\": 119,\n", - " \"timestamp\": \"2025-04-08T07:11:43.883420\"\n", + " \"timestamp\": \"2025-05-30T07:17:38.176146\"\n", " },\n", " {\n", - " \"value\": 0.03866000473499298,\n", + " \"value\": 0.03583735227584839,\n", " \"step\": 129,\n", - " \"timestamp\": \"2025-04-08T07:12:11.505720\"\n", + " \"timestamp\": \"2025-05-30T07:17:55.308118\"\n", " },\n", " {\n", - " \"value\": 0.05358102172613144,\n", + " \"value\": 0.05422971397638321,\n", " \"step\": 139,\n", - " \"timestamp\": \"2025-04-08T07:12:41.459384\"\n", + " \"timestamp\": \"2025-05-30T07:18:12.879283\"\n", " },\n", " {\n", - " \"value\": 0.07019069045782089,\n", + " \"value\": 0.06653454154729843,\n", " \"step\": 149,\n", - " \"timestamp\": \"2025-04-08T07:13:01.654562\"\n", + " \"timestamp\": \"2025-05-30T07:18:31.162906\"\n", " },\n", " {\n", - " \"value\": 0.07882250845432281,\n", + " \"value\": 0.07617977261543274,\n", " \"step\": 159,\n", - " \"timestamp\": \"2025-04-08T07:13:28.198781\"\n", + " \"timestamp\": \"2025-05-30T07:18:48.257926\"\n", " },\n", " {\n", - " \"value\": 0.057352885603904724,\n", + " \"value\": 0.05259016901254654,\n", " \"step\": 169,\n", - " \"timestamp\": \"2025-04-08T07:13:55.679940\"\n", + " \"timestamp\": \"2025-05-30T07:19:05.653311\"\n", " },\n", " {\n", - " \"value\": 0.048704877495765686,\n", + " \"value\": 0.05665893107652664,\n", " \"step\": 179,\n", - " \"timestamp\": \"2025-04-08T07:14:16.962284\"\n", + " \"timestamp\": \"2025-05-30T07:19:22.797584\"\n", " },\n", " {\n", - " \"value\": 0.11735919862985611,\n", + " \"value\": 0.11689528077840805,\n", " \"step\": 189,\n", - " \"timestamp\": \"2025-04-08T07:14:47.678515\"\n", + " \"timestamp\": \"2025-05-30T07:19:39.733535\"\n", " },\n", " {\n", - " \"value\": 0.030179649591445923,\n", + " \"value\": 0.028054695576429367,\n", " \"step\": 199,\n", - " \"timestamp\": \"2025-04-08T07:15:09.066542\"\n", + " \"timestamp\": \"2025-05-30T07:19:58.093349\"\n", " },\n", " {\n", - " \"value\": 0.027945276349782944,\n", + " \"value\": 0.027705904096364975,\n", " \"step\": 209,\n", - " \"timestamp\": \"2025-04-08T07:15:29.159765\"\n", + " \"timestamp\": \"2025-05-30T07:20:14.738329\"\n", " },\n", " {\n", - " \"value\": 0.04044452682137489,\n", + " \"value\": 0.0403934009373188,\n", " \"step\": 219,\n", - " \"timestamp\": \"2025-04-08T07:17:11.140956\"\n", + " \"timestamp\": \"2025-05-30T07:21:12.385598\"\n", " },\n", " {\n", - " \"value\": 0.047244369983673096,\n", + " \"value\": 0.046751510351896286,\n", " \"step\": 229,\n", - " \"timestamp\": \"2025-04-08T07:17:31.705232\"\n", + " \"timestamp\": \"2025-05-30T07:21:29.056354\"\n", " },\n", " {\n", - " \"value\": 0.04799420014023781,\n", + " \"value\": 0.045215629041194916,\n", " \"step\": 239,\n", - " \"timestamp\": \"2025-04-08T07:18:00.639729\"\n", + " \"timestamp\": \"2025-05-30T07:21:46.235006\"\n", " },\n", " {\n", - " \"value\": 0.017997125163674355,\n", + " \"value\": 0.01668056845664978,\n", " \"step\": 249,\n", - " \"timestamp\": \"2025-04-08T07:18:18.967952\"\n", + " \"timestamp\": \"2025-05-30T07:22:04.285016\"\n", " },\n", " {\n", - " \"value\": 0.05586077272891998,\n", + " \"value\": 0.0531625971198082,\n", " \"step\": 259,\n", - " \"timestamp\": \"2025-04-08T07:18:37.262997\"\n", + " \"timestamp\": \"2025-05-30T07:22:22.314943\"\n", " },\n", " {\n", - " \"value\": 0.024363242089748383,\n", + " \"value\": 0.029073115438222885,\n", " \"step\": 269,\n", - " \"timestamp\": \"2025-04-08T07:18:55.515872\"\n", + " \"timestamp\": \"2025-05-30T07:22:39.786721\"\n", " },\n", " {\n", - " \"value\": 0.007903068326413631,\n", + " \"value\": 0.008088404312729836,\n", " \"step\": 279,\n", - " \"timestamp\": \"2025-04-08T07:19:15.973629\"\n", + " \"timestamp\": \"2025-05-30T07:22:56.558962\"\n", " },\n", " {\n", - " \"value\": 0.04577765613794327,\n", + " \"value\": 0.04699515923857689,\n", " \"step\": 289,\n", - " \"timestamp\": \"2025-04-08T07:19:38.154098\"\n", + " \"timestamp\": \"2025-05-30T07:23:13.916215\"\n", " },\n", " {\n", - " \"value\": 0.05233767256140709,\n", + " \"value\": 0.04497051611542702,\n", " \"step\": 299,\n", - " \"timestamp\": \"2025-04-08T07:19:58.773278\"\n", + " \"timestamp\": \"2025-05-30T07:23:32.565833\"\n", " },\n", " {\n", - " \"value\": 0.02214895375072956,\n", + " \"value\": 0.018568137660622597,\n", " \"step\": 309,\n", - " \"timestamp\": \"2025-04-08T07:20:31.187366\"\n", + " \"timestamp\": \"2025-05-30T07:23:50.710766\"\n", " },\n", " {\n", - " \"value\": 0.03232814371585846,\n", + " \"value\": 0.03213285654783249,\n", " \"step\": 319,\n", - " \"timestamp\": \"2025-04-08T07:20:53.427308\"\n", + " \"timestamp\": \"2025-05-30T07:24:08.359392\"\n", " },\n", " {\n", - " \"value\": 0.07305795699357986,\n", + " \"value\": 0.07464674115180969,\n", " \"step\": 329,\n", - " \"timestamp\": \"2025-04-08T07:21:14.126436\"\n", + " \"timestamp\": \"2025-05-30T07:24:26.309504\"\n", " },\n", " {\n", - " \"value\": 0.010019486770033836,\n", + " \"value\": 0.00965566374361515,\n", " \"step\": 339,\n", - " \"timestamp\": \"2025-04-08T07:21:45.127167\"\n", + " \"timestamp\": \"2025-05-30T07:24:44.773706\"\n", " },\n", " {\n", - " \"value\": 0.005295319017022848,\n", + " \"value\": 0.0056703174486756325,\n", " \"step\": 349,\n", - " \"timestamp\": \"2025-04-08T07:22:04.148566\"\n", + " \"timestamp\": \"2025-05-30T07:25:03.499804\"\n", " },\n", " {\n", - " \"value\": 0.006320458836853504,\n", + " \"value\": 0.005867576226592064,\n", " \"step\": 359,\n", - " \"timestamp\": \"2025-04-08T07:22:26.461070\"\n", + " \"timestamp\": \"2025-05-30T07:25:21.372588\"\n", " },\n", " {\n", - " \"value\": 0.005519861355423927,\n", + " \"value\": 0.005034453235566616,\n", " \"step\": 369,\n", - " \"timestamp\": \"2025-04-08T07:22:56.395855\"\n", + " \"timestamp\": \"2025-05-30T07:25:39.495024\"\n", " },\n", " {\n", - " \"value\": 0.024514716118574142,\n", + " \"value\": 0.026450905948877335,\n", " \"step\": 379,\n", - " \"timestamp\": \"2025-04-08T07:23:14.846578\"\n", + " \"timestamp\": \"2025-05-30T07:25:57.922653\"\n", " },\n", " {\n", - " \"value\": 0.006911200471222401,\n", + " \"value\": 0.008726379834115505,\n", " \"step\": 389,\n", - " \"timestamp\": \"2025-04-08T07:23:40.472405\"\n", + " \"timestamp\": \"2025-05-30T07:26:15.829773\"\n", " },\n", " {\n", - " \"value\": 0.02401524968445301,\n", + " \"value\": 0.024301953613758087,\n", " \"step\": 399,\n", - " \"timestamp\": \"2025-04-08T07:24:12.313296\"\n", + " \"timestamp\": \"2025-05-30T07:26:34.681124\"\n", " },\n", " {\n", - " \"value\": 0.0362529456615448,\n", + " \"value\": 0.03407437354326248,\n", " \"step\": 409,\n", - " \"timestamp\": \"2025-04-08T07:24:43.646847\"\n", + " \"timestamp\": \"2025-05-30T07:26:52.882022\"\n", " },\n", " {\n", - " \"value\": 0.03379900008440018,\n", + " \"value\": 0.029367350041866302,\n", " \"step\": 419,\n", - " \"timestamp\": \"2025-04-08T07:25:14.787556\"\n", + " \"timestamp\": \"2025-05-30T07:27:10.712817\"\n", " },\n", " {\n", - " \"value\": 0.01118356455117464,\n", + " \"value\": 0.01157580316066742,\n", " \"step\": 429,\n", - " \"timestamp\": \"2025-04-08T07:25:43.727954\"\n", + " \"timestamp\": \"2025-05-30T07:27:28.722170\"\n", " }\n", " ],\n", " \"val_loss\": [\n", " {\n", - " \"value\": 0.047389596700668335,\n", + " \"value\": 0.048070408403873444,\n", " \"step\": 218,\n", - " \"timestamp\": \"2025-04-08T07:16:54.402277\"\n", + " \"timestamp\": \"2025-05-30T07:21:09.637431\"\n", " },\n", " {\n", - " \"value\": 0.04571458697319031,\n", + " \"value\": 0.04448993504047394,\n", " \"step\": 437,\n", - " \"timestamp\": \"2025-04-08T07:26:36.013749\"\n", + " \"timestamp\": \"2025-05-30T07:28:22.517983\"\n", " }\n", " ]\n", " }\n", " },\n", " \"status_logs\": [\n", " {\n", - " \"updated_at\": \"2025-04-08T07:05:53\",\n", - " \"message\": \"PVCCreated\",\n", - " \"detail\": null\n", - " },\n", - " {\n", - " \"updated_at\": \"2025-04-08T07:05:53\",\n", + " \"updated_at\": \"2025-05-30T07:12:40\",\n", " \"message\": \"EntityHandler_0_Created\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:05:53.328702\",\n", + " \"updated_at\": \"2025-05-30T07:12:40.122576\",\n", " \"message\": \"created\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:06:04\",\n", + " \"updated_at\": \"2025-05-30T07:12:47\",\n", " \"message\": \"EntityHandler_0_Pending\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:06:04\",\n", + " \"updated_at\": \"2025-05-30T07:12:47\",\n", " \"message\": \"EntityHandler_0_Completed\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:06:04\",\n", + " \"updated_at\": \"2025-05-30T07:12:47\",\n", " \"message\": \"TrainingJobCreated\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:06:04\",\n", + " \"updated_at\": \"2025-05-30T07:12:56\",\n", " \"message\": \"TrainingJobRunning\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:27:02\",\n", + " \"updated_at\": \"2025-05-30T07:28:53\",\n", " \"message\": \"TrainingJobCompleted\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:27:02\",\n", + " \"updated_at\": \"2025-05-30T07:28:53\",\n", " \"message\": \"EntityHandler_1_Created\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:27:02\",\n", - " \"message\": \"EntityHandler_1_Running\",\n", - " \"detail\": null\n", - " },\n", - " {\n", - " \"updated_at\": \"2025-04-08T07:27:10\",\n", + " \"updated_at\": \"2025-05-30T07:29:02\",\n", " \"message\": \"EntityHandler_1_Pending\",\n", " \"detail\": null\n", " },\n", " {\n", - " \"updated_at\": \"2025-04-08T07:27:10\",\n", + " \"updated_at\": \"2025-05-30T07:29:02\",\n", " \"message\": \"EntityHandler_1_Completed\",\n", " \"detail\": null\n", " }\n", @@ -1012,7 +1013,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 15, "id": "ebad3944-70d5-4b23-9a38-a83774bf20c4", "metadata": { "scrolled": true @@ -1026,9 +1027,9 @@ " \"object\": \"list\",\n", " \"data\": [\n", " {\n", - " \"created_at\": \"2025-04-08T07:05:53.417350\",\n", - " \"updated_at\": \"2025-04-08T07:05:53.417354\",\n", - " \"name\": \"llama-3.2-1b-xlam-run1@cust-BTkGbfifLfEAjV2THu3tas\",\n", + " \"created_at\": \"2025-05-30T07:12:40.305264\",\n", + " \"updated_at\": \"2025-05-30T07:12:40.305267\",\n", + " \"name\": \"llama-3.2-1b-xlam-run1@cust-6wUQyBcweaEvBVdUXdTNAM\",\n", " \"namespace\": \"xlam-tutorial-ns\",\n", " \"description\": \"None\",\n", " \"spec\": {\n", @@ -1043,7 +1044,7 @@ " \"tensor_parallelism\": 1,\n", " \"backend_engine\": \"nemo\",\n", " \"status\": \"upload_completed\",\n", - " \"files_url\": \"hf://xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-BTkGbfifLfEAjV2THu3tas\"\n", + " \"files_url\": \"hf://xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-6wUQyBcweaEvBVdUXdTNAM\"\n", " },\n", " \"base_model\": \"meta/llama-3.2-1b-instruct\",\n", " \"peft\": {\n", @@ -1053,9 +1054,9 @@ " \"custom_fields\": {}\n", " },\n", " {\n", - " \"created_at\": \"2025-04-07T23:41:27.787234\",\n", - " \"updated_at\": \"2025-04-07T23:41:27.787238\",\n", - " \"name\": \"llama-3.2-1b-xlam-run1@cust-ULiqFRom2AnFEUsVpeoLED\",\n", + " \"created_at\": \"2025-05-30T07:07:17.482334\",\n", + " \"updated_at\": \"2025-05-30T07:07:17.482337\",\n", + " \"name\": \"llama-3.2-1b-xlam-run1@cust-AfoHSExoTv6hBudfX8Hx7d\",\n", " \"namespace\": \"xlam-tutorial-ns\",\n", " \"description\": \"None\",\n", " \"spec\": {\n", @@ -1069,8 +1070,8 @@ " \"precision\": \"bf16-mixed\",\n", " \"tensor_parallelism\": 1,\n", " \"backend_engine\": \"nemo\",\n", - " \"status\": \"upload_completed\",\n", - " \"files_url\": \"hf://xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-ULiqFRom2AnFEUsVpeoLED\"\n", + " \"status\": \"created\",\n", + " \"files_url\": \"hf://xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-AfoHSExoTv6hBudfX8Hx7d\"\n", " },\n", " \"base_model\": \"meta/llama-3.2-1b-instruct\",\n", " \"peft\": {\n", @@ -1078,14 +1079,24 @@ " },\n", " \"schema_version\": \"1.0\",\n", " \"custom_fields\": {}\n", + " },\n", + " {\n", + " \"created_at\": \"2025-05-19T21:20:19.704047\",\n", + " \"updated_at\": \"2025-05-19T21:20:19.704047\",\n", + " \"name\": \"llama-3.2-1b-xlam-run2@cust-14pJUjVaYDSN5DuftHECGy\",\n", + " \"namespace\": \"xlam-tutorial-ns\",\n", + " \"schema_version\": \"1.0\",\n", + " \"custom_fields\": {\n", + " \"created_by\": \"nmp\"\n", + " }\n", " }\n", " ],\n", " \"pagination\": {\n", " \"page\": 1,\n", " \"page_size\": 1000,\n", - " \"current_page_size\": 2,\n", + " \"current_page_size\": 3,\n", " \"total_pages\": 1,\n", - " \"total_results\": 2\n", + " \"total_results\": 3\n", " },\n", " \"sort\": \"-created_at\",\n", " \"filter\": {\n", @@ -1130,7 +1141,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 16, "id": "43026f8a-3b98-4aa6-b4c6-7441862863fd", "metadata": {}, "outputs": [], @@ -1351,7 +1362,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 11, "id": "36398168-1051-4e54-ac3e-7a3e406f393a", "metadata": {}, "outputs": [ @@ -1359,7 +1370,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "Name of your custom model is: xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-BTkGbfifLfEAjV2THu3tas\n" + "Name of your custom model is: xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-6wUQyBcweaEvBVdUXdTNAM\n" ] } ], diff --git a/nemo/data-flywheel/tool-calling/config.py b/nemo/data-flywheel/tool-calling/config.py index 0816c165e..26f12e547 100644 --- a/nemo/data-flywheel/tool-calling/config.py +++ b/nemo/data-flywheel/tool-calling/config.py @@ -18,3 +18,4 @@ # (Optional) Configure the base model. Must be one supported by the NeMo Customizer deployment! BASE_MODEL = "meta/llama-3.2-1b-instruct" +BASE_MODEL_VERSION = "v1.0.0+A100" \ No newline at end of file