LLM_034_fine_tuning_yahoo_answers_topics.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOrEDvXIVhpYWe5gf5xKa0Q",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "c2580945158b422caa423e76cdd573d9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_40b36812d8d2493282f211a510a5f7b5",
              "IPY_MODEL_40dc21e1b36243faa6f0abe427284226",
              "IPY_MODEL_0f50c00ee2ea4e45b3c444ae9355c734"
            ],
            "layout": "IPY_MODEL_1e3ec3866c494c098b62b2e3ebb33205"
          }
        },
        "40b36812d8d2493282f211a510a5f7b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_798095c00c03485d9f378a8783326cde",
            "placeholder": "​",
            "style": "IPY_MODEL_36a85cba9df1435899efd2786faab78a",
            "value": "tokenizer_config.json: 100%"
          }
        },
        "40dc21e1b36243faa6f0abe427284226": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a5d6fadc6e0f448ab7b43fc999f31c24",
            "max": 48,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_868dac39cc0d42c8a1df15167bd2adee",
            "value": 48
          }
        },
        "0f50c00ee2ea4e45b3c444ae9355c734": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a4ed14d1c1484ad4aad0a612d6b0625c",
            "placeholder": "​",
            "style": "IPY_MODEL_45bfcb2333284078b2c1cc85a750be8c",
            "value": " 48.0/48.0 [00:00&lt;00:00, 2.28kB/s]"
          }
        },
        "1e3ec3866c494c098b62b2e3ebb33205": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "798095c00c03485d9f378a8783326cde": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "36a85cba9df1435899efd2786faab78a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a5d6fadc6e0f448ab7b43fc999f31c24": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "868dac39cc0d42c8a1df15167bd2adee": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "a4ed14d1c1484ad4aad0a612d6b0625c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "45bfcb2333284078b2c1cc85a750be8c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "bab2461d3f5c4addb19dbf47aceb77c9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_4752735ed522454a8a339c99e71aca7d",
              "IPY_MODEL_95d26c29f32447dc8901481d4d3734ae",
              "IPY_MODEL_229396e7fda14b8eb556f5c1bb35ed40"
            ],
            "layout": "IPY_MODEL_ebd72d0f763447a49ff69826ffa8348d"
          }
        },
        "4752735ed522454a8a339c99e71aca7d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_58093cad7a3041bdb456dabf994c5208",
            "placeholder": "​",
            "style": "IPY_MODEL_4b8d3d0753cc4d0793f015f2646b069f",
            "value": "config.json: 100%"
          }
        },
        "95d26c29f32447dc8901481d4d3734ae": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4c607d24a3aa440480597dfa27ed1ea6",
            "max": 483,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_9c99c29e66e84eaaa2e0702485893700",
            "value": 483
          }
        },
        "229396e7fda14b8eb556f5c1bb35ed40": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5f9286e8ab09438fadbcb33bcead6904",
            "placeholder": "​",
            "style": "IPY_MODEL_a465c2910ad64f7f9081e8dd23feceb7",
            "value": " 483/483 [00:00&lt;00:00, 35.0kB/s]"
          }
        },
        "ebd72d0f763447a49ff69826ffa8348d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "58093cad7a3041bdb456dabf994c5208": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4b8d3d0753cc4d0793f015f2646b069f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "4c607d24a3aa440480597dfa27ed1ea6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9c99c29e66e84eaaa2e0702485893700": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "5f9286e8ab09438fadbcb33bcead6904": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a465c2910ad64f7f9081e8dd23feceb7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a154b78f5daf43eb924bcb6742791dc0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_a7ec3a5d652b48c097a73aad7d82fc80",
              "IPY_MODEL_aa84a43e9b2148cea62f458a414a5950",
              "IPY_MODEL_be4647671cc549b3a71d369d019ec5ba"
            ],
            "layout": "IPY_MODEL_27adb0ed45ce48fea17c4796ea388fff"
          }
        },
        "a7ec3a5d652b48c097a73aad7d82fc80": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2bc18ae2235f42f08f4b078ae6711a14",
            "placeholder": "​",
            "style": "IPY_MODEL_9db077d5337d4257b705817f38b4bec4",
            "value": "vocab.txt: 100%"
          }
        },
        "aa84a43e9b2148cea62f458a414a5950": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_41bbe91682e84d4da0a37fd75c4ff3d6",
            "max": 231508,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_517eefd33cfc42d4836727647128912e",
            "value": 231508
          }
        },
        "be4647671cc549b3a71d369d019ec5ba": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ef05935f58eb40128a41c21775a78a92",
            "placeholder": "​",
            "style": "IPY_MODEL_ac16c8ee638a49afa538a1857ad38695",
            "value": " 232k/232k [00:00&lt;00:00, 11.5MB/s]"
          }
        },
        "27adb0ed45ce48fea17c4796ea388fff": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2bc18ae2235f42f08f4b078ae6711a14": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9db077d5337d4257b705817f38b4bec4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "41bbe91682e84d4da0a37fd75c4ff3d6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "517eefd33cfc42d4836727647128912e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "ef05935f58eb40128a41c21775a78a92": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ac16c8ee638a49afa538a1857ad38695": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8141bd6ae9f74f878cdabadbaaa97536": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_3a43077a3d724cf085c04b1f52e12035",
              "IPY_MODEL_3521832224f7493688958c0cf19d6382",
              "IPY_MODEL_bb77e0497871410abe924d7ee181459d"
            ],
            "layout": "IPY_MODEL_28925eab0b2143e488b147c035763d77"
          }
        },
        "3a43077a3d724cf085c04b1f52e12035": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fc3ef2e169c44fb1a21d47c74e32cf10",
            "placeholder": "​",
            "style": "IPY_MODEL_99b43b1c89964c70892b44b63f1b10f0",
            "value": "tokenizer.json: 100%"
          }
        },
        "3521832224f7493688958c0cf19d6382": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_59920481bdda4295a3498c6a3c89a66b",
            "max": 466062,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_a631d378dd88424eb37b0cae5ca5d89f",
            "value": 466062
          }
        },
        "bb77e0497871410abe924d7ee181459d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ffe39811e4b2473a8cbb2651b51869b8",
            "placeholder": "​",
            "style": "IPY_MODEL_f5d47406b1a6497181f64a2da8fd6607",
            "value": " 466k/466k [00:00&lt;00:00, 18.6MB/s]"
          }
        },
        "28925eab0b2143e488b147c035763d77": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fc3ef2e169c44fb1a21d47c74e32cf10": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "99b43b1c89964c70892b44b63f1b10f0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "59920481bdda4295a3498c6a3c89a66b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a631d378dd88424eb37b0cae5ca5d89f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "ffe39811e4b2473a8cbb2651b51869b8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f5d47406b1a6497181f64a2da8fd6607": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "68ed145e8f5a4f1fb8c83f59a66615f8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_5bcfc6a9b77c4112a906fbaf0f96f399",
              "IPY_MODEL_37cde7ffa3574dd5a2dfa9e4f048ca58",
              "IPY_MODEL_eaf977aba6e548ae8cd9bae0e56bc27b"
            ],
            "layout": "IPY_MODEL_5c626441ad1b43dab02ed43dcfca58fd"
          }
        },
        "5bcfc6a9b77c4112a906fbaf0f96f399": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_370491d99b9d4dbbadb508f7905943b4",
            "placeholder": "​",
            "style": "IPY_MODEL_ce50b44c4f7e441cbfb65d8b842a98e8",
            "value": "Generating train split: "
          }
        },
        "37cde7ffa3574dd5a2dfa9e4f048ca58": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_de65421ac13a412f93dd61a6f5ed13eb",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_4022ae749ab74756a432cc9010001690",
            "value": 1
          }
        },
        "eaf977aba6e548ae8cd9bae0e56bc27b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5325ea426fe8491d8586532c5160d2b7",
            "placeholder": "​",
            "style": "IPY_MODEL_1a7ed75b7c014b5eadbcc9eea1b46e07",
            "value": " 140000/0 [00:09&lt;00:00, 18699.78 examples/s]"
          }
        },
        "5c626441ad1b43dab02ed43dcfca58fd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "370491d99b9d4dbbadb508f7905943b4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ce50b44c4f7e441cbfb65d8b842a98e8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "de65421ac13a412f93dd61a6f5ed13eb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "4022ae749ab74756a432cc9010001690": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "5325ea426fe8491d8586532c5160d2b7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1a7ed75b7c014b5eadbcc9eea1b46e07": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "68ae0a4b753a4159936d3d22e617f91d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_596b687080d84890b2621af3d0491199",
              "IPY_MODEL_41ea2cab4d414840ba8ff92001b0e975",
              "IPY_MODEL_bb39f1375537473983135ac4d2ea569c"
            ],
            "layout": "IPY_MODEL_16c399425fa045218a9415e16971ac15"
          }
        },
        "596b687080d84890b2621af3d0491199": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6b4f443e49524901abaf8405fb64870b",
            "placeholder": "​",
            "style": "IPY_MODEL_3ebebadff6484dc284d3f5eaa70c4d7d",
            "value": "Generating train split: "
          }
        },
        "41ea2cab4d414840ba8ff92001b0e975": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d929bbf433e747afa3bc10dc06149bd3",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_db200f517cc3497b9f76db672a1c5380",
            "value": 1
          }
        },
        "bb39f1375537473983135ac4d2ea569c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fe1c305004b343e993dd64dbd6a45b8e",
            "placeholder": "​",
            "style": "IPY_MODEL_068e873911b3435788ff73d078e78269",
            "value": " 6000/0 [00:00&lt;00:00, 18510.72 examples/s]"
          }
        },
        "16c399425fa045218a9415e16971ac15": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6b4f443e49524901abaf8405fb64870b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3ebebadff6484dc284d3f5eaa70c4d7d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d929bbf433e747afa3bc10dc06149bd3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "db200f517cc3497b9f76db672a1c5380": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "fe1c305004b343e993dd64dbd6a45b8e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "068e873911b3435788ff73d078e78269": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "7f69f2e4a52741c1ab161f0a1a375636": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_4667d2f258574f2c961595f759cf13fe",
              "IPY_MODEL_a61f487ca142444c846bafc80c03dc8c",
              "IPY_MODEL_c0683729a1d547f1a0f345ac6c557df3"
            ],
            "layout": "IPY_MODEL_346ae85308d440a3bdcfc8047a4e8209"
          }
        },
        "4667d2f258574f2c961595f759cf13fe": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_38fa21b9b0fb47d88780a1ed403cc744",
            "placeholder": "​",
            "style": "IPY_MODEL_8b073cad71064423885608ab33dce9aa",
            "value": "config.json: 100%"
          }
        },
        "a61f487ca142444c846bafc80c03dc8c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a349c380093744e3849f5f2182c24df8",
            "max": 1206,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_89fc10a59f5f4b14855fdbec82f3c3b1",
            "value": 1206
          }
        },
        "c0683729a1d547f1a0f345ac6c557df3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b729ad76a9bc4782b311b86df6ca9f80",
            "placeholder": "​",
            "style": "IPY_MODEL_d25d2d7b6e1e4c04bb87755eb7e29361",
            "value": " 1.21k/1.21k [00:00&lt;00:00, 53.7kB/s]"
          }
        },
        "346ae85308d440a3bdcfc8047a4e8209": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "38fa21b9b0fb47d88780a1ed403cc744": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8b073cad71064423885608ab33dce9aa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "a349c380093744e3849f5f2182c24df8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "89fc10a59f5f4b14855fdbec82f3c3b1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "b729ad76a9bc4782b311b86df6ca9f80": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d25d2d7b6e1e4c04bb87755eb7e29361": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "effc347e877d41b88b78150aa3f8c873": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_ac3add7b79bb4186a50b1799a4de14e1",
              "IPY_MODEL_c910f8838578409bb4727f73b3c12c03",
              "IPY_MODEL_3a1b4f1a43de43daad938bcb1380ef34"
            ],
            "layout": "IPY_MODEL_eea701dcb08943d6bb02e3a2f61f6623"
          }
        },
        "ac3add7b79bb4186a50b1799a4de14e1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_4902199dfe6c4d14a08075531dce2cea",
            "placeholder": "​",
            "style": "IPY_MODEL_235b0904cbab4ec38d51aa9ade8877be",
            "value": "model.safetensors: 100%"
          }
        },
        "c910f8838578409bb4727f73b3c12c03": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ce396907707a4f5eb1613524bd919edb",
            "max": 242043056,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_5368f876e4664a258e4e5be34c4cf413",
            "value": 242043056
          }
        },
        "3a1b4f1a43de43daad938bcb1380ef34": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d14c0ccb3d56447d91d13b63ecf82318",
            "placeholder": "​",
            "style": "IPY_MODEL_cb96f2e0a9cd41a38a5e9347b85aa122",
            "value": " 242M/242M [00:01&lt;00:00, 159MB/s]"
          }
        },
        "eea701dcb08943d6bb02e3a2f61f6623": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4902199dfe6c4d14a08075531dce2cea": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "235b0904cbab4ec38d51aa9ade8877be": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ce396907707a4f5eb1613524bd919edb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "5368f876e4664a258e4e5be34c4cf413": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "d14c0ccb3d56447d91d13b63ecf82318": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cb96f2e0a9cd41a38a5e9347b85aa122": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cf3f9c0f29e24244a9510912e29ddcee": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_d13753328dcd4492bef5834a6c1d31c6",
              "IPY_MODEL_649e106653364f69988179d15990ec20",
              "IPY_MODEL_3ba49d3ff83047e1aaaedf45681e32ec"
            ],
            "layout": "IPY_MODEL_5a6691ddb4be4414b01e6d65b2ca419c"
          }
        },
        "d13753328dcd4492bef5834a6c1d31c6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3d508ce0ae4b48d4bffa1f5e82385dc3",
            "placeholder": "​",
            "style": "IPY_MODEL_f9f44323585f4fb08cd3d6ff0676447c",
            "value": "generation_config.json: 100%"
          }
        },
        "649e106653364f69988179d15990ec20": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_75a1d02fe21f4a37bd7ab03a65d3a71b",
            "max": 147,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_2f9e90298fc3411abb090d8655a20c39",
            "value": 147
          }
        },
        "3ba49d3ff83047e1aaaedf45681e32ec": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_2791f03f203f400abbb37af7ef86862a",
            "placeholder": "​",
            "style": "IPY_MODEL_b6734b27d72444358e16d967bda844c0",
            "value": " 147/147 [00:00&lt;00:00, 8.51kB/s]"
          }
        },
        "5a6691ddb4be4414b01e6d65b2ca419c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3d508ce0ae4b48d4bffa1f5e82385dc3": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f9f44323585f4fb08cd3d6ff0676447c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "75a1d02fe21f4a37bd7ab03a65d3a71b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "2f9e90298fc3411abb090d8655a20c39": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "2791f03f203f400abbb37af7ef86862a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b6734b27d72444358e16d967bda844c0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/micah-shull/LLMs/blob/main/LLM_034_fine_tuning_yahoo_answers_topics.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Yahoo Answers Topics Dataset Summary\n",
        "\n",
        "The **Yahoo Answers Topics** dataset is a large-scale text classification dataset sourced from the Yahoo Answers community question-answering platform. It includes a diverse range of topics, making it a valuable resource for training and evaluating models on multiclass text classification tasks. The dataset is structured to classify user-submitted questions and answers into one of ten distinct topic categories.\n",
        "\n",
        "#### Dataset Structure\n",
        "\n",
        "- **Number of Classes**: 10\n",
        "- **Classes**:\n",
        "  1. Society & Culture\n",
        "  2. Science & Mathematics\n",
        "  3. Health\n",
        "  4. Education & Reference\n",
        "  5. Computers & Internet\n",
        "  6. Sports\n",
        "  7. Business & Finance\n",
        "  8. Entertainment & Music\n",
        "  9. Family & Relationships\n",
        "  10. Politics & Government\n",
        "\n",
        "- **Columns**:\n",
        "  - **id**: Unique identifier for each entry.\n",
        "  - **topic**: The target label, represented as an integer from 0 to 9, corresponding to one of the 10 topics above.\n",
        "  - **question_title**: The title of the user’s question.\n",
        "  - **question_content**: The full content of the question.\n",
        "  - **best_answer**: The best answer selected for the question, if available.\n",
        "\n",
        "- **Data Splits**:\n",
        "  - **Training Set**: 1,400,000 samples\n",
        "  - **Test Set**: 60,000 samples\n",
        "\n",
        "#### Use Case\n",
        "\n",
        "This dataset is ideal for exploring multiclass classification, particularly for tasks requiring a model to distinguish between broad thematic categories based on limited context. It is commonly used in research and applications involving topic classification, question answering, and content categorization.\n",
        "\n",
        "#### Example Usage\n",
        "\n",
        "The Yahoo Answers Topics dataset can be used to train models that automatically classify questions into general topics, which is helpful for content filtering, recommendation systems, and automated helpdesk systems.\n",
        "\n"
      ],
      "metadata": {
        "id": "8ueDJbTnSZex"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Question Title & Content\n",
        "\n",
        "In the Yahoo Answers Topics dataset, the **question title** and **question content** serve different roles in framing the user's question. Here’s how they generally differ:\n",
        "\n",
        "1. **Question Title**:\n",
        "   - The question title is typically a **short summary or headline** that gives a quick overview of what the question is about.\n",
        "   - It’s often brief and concise, capturing the main topic or intent in just a few words.\n",
        "   - For example: \"How do I make pasta?\"\n",
        "\n",
        "2. **Question Content**:\n",
        "   - The question content is the **full description** or **additional details** provided by the user to clarify or elaborate on the question.\n",
        "   - This section usually includes more context, specifics, or sub-questions that expand on the title.\n",
        "   - For example, in addition to the title \"How do I make pasta?\", the content might be: \"I want to try making pasta from scratch. What ingredients do I need, and what is the basic process?\"\n",
        "\n",
        "#### Example of Question Title vs. Content\n",
        "\n",
        "- **Question Title**: \"How do I make pasta?\"\n",
        "- **Question Content**: \"I’m planning to make pasta for dinner but have never done it before. Do I need a special machine, or can I do it by hand? What type of flour should I use?\"\n",
        "\n",
        "\n",
        "#### Importance in Modeling\n",
        "\n",
        "Using both the title and content together can give a more comprehensive representation of the question, which can be particularly useful for classification tasks, as it provides both a summary and detailed context that help capture the nuances of the question topic.\n",
        "\n",
        "### Combine Title & Content\n",
        "\n",
        "In the code we’re **combining the title and content into a single string** for each entry. This approach provides the model with both the concise summary (title) and the detailed context (content), creating a more informative input for training.\n",
        "\n",
        "```python\n",
        "train_documents_sample = [\n",
        "    f\"{dataset['train']['question_title'][i]} {dataset['train']['question_content'][i]}\"\n",
        "    for i in train_sample_indices\n",
        "]\n",
        "```\n",
        "\n",
        "In this case, each entry in `train_documents_sample` will contain both the question title and the content in a single string, separated by a space. This approach helps ensure that the model has access to the full context of the question during training, improving its ability to accurately classify the topics.\n",
        "\n",
        "Combining the title and content is a common technique for text classification when both parts provide valuable information, as it creates a richer representation of the input."
      ],
      "metadata": {
        "id": "-3jwJWOAUVXj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Load & Import Libraries"
      ],
      "metadata": {
        "id": "AUa1qlHGfFio"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "8F4W1m3iMoC8"
      },
      "outputs": [],
      "source": [
        "# !pip install transformers datasets\n",
        "# !pip install python-dotenv"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Hugging Face Libraries\n",
        "from transformers import AutoTokenizer, AutoModelForSequenceClassification, TrainingArguments, Trainer\n",
        "from datasets import load_dataset, Dataset\n",
        "import warnings\n",
        "\n",
        "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
        "\n",
        "# Standard Libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import os\n",
        "import time\n",
        "import json\n",
        "from dotenv import load_dotenv\n",
        "\n",
        "# Metrics and Evaluation\n",
        "from sklearn.metrics import classification_report, ConfusionMatrixDisplay\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Visualization\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns"
      ],
      "metadata": {
        "id": "BFhfCPNnM0V7"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Load  Data\n",
        "Absolutely, it's entirely possible to work with a smaller sample of the Yahoo Answers dataset to avoid issues like running out of memory in your Colab notebook. You can sample a portion of the dataset after loading it using the `load_dataset` function. Below, I'll show you how you can load a subset of the dataset to continue with the fine-tuning process.\n",
        "\n",
        "### Step-by-Step Approach to Sampling the Dataset:\n",
        "\n",
        "1. **Load the Dataset**: Load the entire dataset using `load_dataset()`.\n",
        "2. **Sampling**: Use the `.train_test_split()` method or pandas to create a sample of the dataset.\n",
        "3. **Proceed with the Sample**: Use the smaller sample for training to avoid memory issues.\n",
        "\n",
        "### Explanation:\n",
        "\n",
        "1. **Loading the Dataset**:\n",
        "   ```python\n",
        "   dataset = load_dataset(\"yahoo_answers_topics\")\n",
        "   ```\n",
        "   This line loads the Yahoo Answers dataset, which is large by default.\n",
        "\n",
        "2. **Sampling the Dataset**:\n",
        "   ```python\n",
        "   sampled_dataset = dataset[\"train\"].train_test_split(test_size=0.9)[\"test\"]\n",
        "   ```\n",
        "   Here, we use the `.train_test_split()` method to create a smaller subset of the training data. By setting `test_size=0.9`, you are effectively keeping only 10% of the original data (`test`) and discarding 90% (`train`). You can adjust this ratio depending on the available resources.\n",
        "\n",
        "3. **Using the Sampled Dataset**:\n",
        "   You can proceed with `sampled_dataset` for training and other processing steps. The sampled dataset will be much smaller and more manageable in Colab.\n",
        "\n",
        "### Alternative Approach: Using `select` for Specific Number of Samples\n",
        "\n",
        "If you want a specific number of samples instead of a percentage, you can use the `select` method:\n",
        "\n",
        "```python\n",
        "# Select a specific number of samples, e.g., 1000\n",
        "sampled_dataset = dataset[\"train\"].select(range(1000))\n",
        "```\n",
        "This line selects the first 1000 samples from the training set.\n",
        "\n",
        "### Notes:\n",
        "\n",
        "- The `.train_test_split()` and `.select()` methods allow you to create a subset of the dataset, making it easier to handle memory constraints.\n",
        "- It is generally advisable to work with a smaller portion of the data for prototyping or when you have limited computational resources.\n",
        "\n",
        "With this smaller dataset, you can proceed with preprocessing and fine-tuning the language model using the Hugging Face Transformers library without overwhelming Colab's memory.\n",
        "\n",
        "The `train_test_split()` method used in the code is from **Hugging Face's `datasets` library**, not from `sklearn`.\n",
        "\n",
        "### Difference Between the Two:\n",
        "\n",
        "- **Hugging Face's `datasets` Library (`train_test_split()`)**:\n",
        "  - This method is specifically designed to work with datasets in the Hugging Face format.\n",
        "  - It operates directly on `Dataset` objects, allowing you to split the dataset without converting it to another format.\n",
        "  - You can specify the split ratio (e.g., `test_size=0.9`), and it returns new `Dataset` objects that can be used for training and testing.\n",
        "\n",
        "- **`sklearn`'s `train_test_split()` Function**:\n",
        "  - This function is from the `scikit-learn` library and is typically used with pandas dataframes or numpy arrays.\n",
        "  - It splits data into training and testing sets but requires the data to be in array or dataframe form before splitting.\n",
        "\n",
        "For Hugging Face datasets, it is more efficient to use their built-in `train_test_split()` method, as it works directly with their dataset objects and avoids the need for conversion to a different data structure.\n",
        "\n"
      ],
      "metadata": {
        "id": "5hzZXq67fBQO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Load environment variables\n",
        "load_dotenv('/content/API_KEYS.env')\n",
        "hugginface_api_key = os.getenv(\"HUGGINGFACE_API_KEY\")\n",
        "# Set the environment variable globally for libraries like LangChain\n",
        "# os.environ[\"HUGGINGFACEHUB_API_TOKEN\"] = hugginface_api_key\n",
        "os.environ[\"HF_TOKEN\"] = hugginface_api_key\n",
        "print(\"HUGGINFACE API Key loaded from .env:\",os.environ[\"HF_TOKEN\"][0:10])\n",
        "\n",
        "# Load the Yahoo Answers dataset\n",
        "dataset = load_dataset(\"yahoo_answers_topics\")\n",
        "\n",
        "# Define the label names\n",
        "label_names = [\n",
        "    \"Society & Culture\", \"Science & Mathematics\", \"Health\", \"Education & Reference\",\n",
        "    \"Computers & Internet\", \"Sports\", \"Business & Finance\", \"Entertainment & Music\",\n",
        "    \"Family & Relationships\", \"Politics & Government\"\n",
        "]\n",
        "\n",
        "# Print the original dataset sizes\n",
        "print(\"Original dataset sizes:\")\n",
        "print(\"Train:\", len(dataset[\"train\"]))\n",
        "print(\"Test:\", len(dataset[\"test\"]))\n",
        "\n",
        "# Sample 10% of the original train and test datasets\n",
        "train_sample = dataset[\"train\"].train_test_split(test_size=0.1)[\"test\"]  # Keep 10% of the training set\n",
        "test_sample = dataset[\"test\"].train_test_split(test_size=0.1)[\"test\"]    # Keep 10% of the test set\n",
        "\n",
        "# Print the sampled dataset sizes\n",
        "print(\"\\nSampled dataset sizes (10% of original sets):\")\n",
        "print(\"Sampled Train:\", len(train_sample))\n",
        "print(\"Sampled Test:\", len(test_sample))\n",
        "\n",
        "\n",
        "# Display the first few entries in the sampled dataset\n",
        "for i in range(3):\n",
        "    print(f\"\\nSample {i+1}:\")\n",
        "    print(\"Question Title:\", train_sample[\"question_title\"][i])\n",
        "    print(\"Question Content:\", train_sample[\"question_content\"][i])\n",
        "    print(\"Best Answer:\", train_sample[\"best_answer\"][i])\n",
        "    print(\"Label:\", label_names[train_sample[\"topic\"][i]])\n",
        "    print(\"=\" * 50)\n",
        "\n",
        "# Print the first entries of train and test datasets to inspect their structure\n",
        "print(\"\\nTrain Sample:\")\n",
        "print(json.dumps(train_sample[0], indent=2))\n",
        "\n",
        "print(\"\\nTest Sample:\")\n",
        "print(json.dumps(test_sample[0], indent=2))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MnAGnnyMNRk_",
        "outputId": "5fbc4a47-f2fe-48c8-a7c8-203edbc27f65"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "HUGGINFACE API Key loaded from .env: hf_qMYzmzn\n",
            "Original dataset sizes:\n",
            "Train: 1400000\n",
            "Test: 60000\n",
            "\n",
            "Sampled dataset sizes (10% of original sets):\n",
            "Sampled Train: 140000\n",
            "Sampled Test: 6000\n",
            "\n",
            "Sample 1:\n",
            "Question Title: What are the critical scarce resources?\n",
            "Question Content: \n",
            "Best Answer: intelligent people without their head up their ^$$\n",
            "Label: Education & Reference\n",
            "==================================================\n",
            "\n",
            "Sample 2:\n",
            "Question Title: What is your favorite color hair and eyes on the opposite sex?\n",
            "Question Content: \n",
            "Best Answer: I have always loved brown eyes....\\nsorry!  I will take some Auburn hair please!\n",
            "Label: Entertainment & Music\n",
            "==================================================\n",
            "\n",
            "Sample 3:\n",
            "Question Title: I live in Panama City Florida so when does school end????\n",
            "Question Content: \n",
            "Best Answer: hi i live in valprasio fl and school over here school ends may 18\n",
            "Label: Education & Reference\n",
            "==================================================\n",
            "\n",
            "Train Sample:\n",
            "{\n",
            "  \"id\": 781902,\n",
            "  \"topic\": 3,\n",
            "  \"question_title\": \"What are the critical scarce resources?\",\n",
            "  \"question_content\": \"\",\n",
            "  \"best_answer\": \"intelligent people without their head up their ^$$\"\n",
            "}\n",
            "\n",
            "Test Sample:\n",
            "{\n",
            "  \"id\": 13203,\n",
            "  \"topic\": 3,\n",
            "  \"question_title\": \"does the word kiddish exist?\",\n",
            "  \"question_content\": \"\",\n",
            "  \"best_answer\": \"People have used kiddish to mean childish on occasion but I couldn't find it in a dictionary.\"\n",
            "}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Class Distribution Before Sampling"
      ],
      "metadata": {
        "id": "o7zyDMqyuy2U"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate value counts for labels for train and test datasets\n",
        "train_label_counts = pd.Series(dataset['train']['topic']).value_counts().sort_index()\n",
        "test_label_counts = pd.Series(dataset['test']['topic']).value_counts().sort_index()\n",
        "train_label_counts.index = label_names  # Map numeric labels to topic names\n",
        "test_label_counts.index = label_names   # Map numeric labels to topic names\n",
        "\n",
        "# Combine 'question_title' and 'question_content' for document lengths for train and test datasets\n",
        "train_documents = [f\"{title} {content}\" for title, content in zip(dataset['train']['question_title'], dataset['train']['question_content'])]\n",
        "test_documents = [f\"{title} {content}\" for title, content in zip(dataset['test']['question_title'], dataset['test']['question_content'])]\n",
        "train_document_lengths = [len(doc.split()) for doc in train_documents]  # Calculate document lengths for train\n",
        "test_document_lengths = [len(doc.split()) for doc in test_documents]    # Calculate document lengths for test\n",
        "\n",
        "# Create a 2x2 grid for the plots\n",
        "fig, axes = plt.subplots(2, 2, figsize=(18, 12))\n",
        "\n",
        "# Set up a color palette\n",
        "palette = sns.color_palette(\"Pastel1\", 10)\n",
        "\n",
        "# Plot 1: Train Label Distribution\n",
        "axes[0, 0].bar(train_label_counts.index, train_label_counts.values, color=palette)\n",
        "axes[0, 0].set_title(\"Yahoo Answers Topics Distribution (Train)\")\n",
        "axes[0, 0].set_xlabel(\"Category\")\n",
        "axes[0, 0].set_ylabel(\"Count\")\n",
        "axes[0, 0].set_xticks(range(len(train_label_counts.index)))\n",
        "axes[0, 0].set_xticklabels(train_label_counts.index, rotation=45, ha='right')\n",
        "\n",
        "# Plot 2: Test Label Distribution\n",
        "axes[0, 1].bar(test_label_counts.index, test_label_counts.values, color=palette)\n",
        "axes[0, 1].set_title(\"Yahoo Answers Topics Distribution (Test)\")\n",
        "axes[0, 1].set_xlabel(\"Category\")\n",
        "axes[0, 1].set_ylabel(\"Count\")\n",
        "axes[0, 1].set_xticks(range(len(test_label_counts.index)))\n",
        "axes[0, 1].set_xticklabels(test_label_counts.index, rotation=45, ha='right')\n",
        "\n",
        "# Plot 3: Train Document Length Distribution\n",
        "axes[1, 0].hist(train_document_lengths, bins=30, color='skyblue', alpha=0.7)\n",
        "axes[1, 0].set_title(\"Distribution of Document Lengths (Train)\")\n",
        "axes[1, 0].set_xlabel(\"Number of Words\")\n",
        "axes[1, 0].set_ylabel(\"Frequency\")\n",
        "\n",
        "# Plot 4: Test Document Length Distribution\n",
        "axes[1, 1].hist(test_document_lengths, bins=30, color='skyblue', alpha=0.7)\n",
        "axes[1, 1].set_title(\"Distribution of Document Lengths (Test)\")\n",
        "axes[1, 1].set_xlabel(\"Number of Words\")\n",
        "axes[1, 1].set_ylabel(\"Frequency\")\n",
        "\n",
        "# Adjust layout and show the plots\n",
        "plt.tight_layout()\n",
        "plt.show()\n",
        "\n",
        "# Display basic statistics for train and test document lengths\n",
        "print(\"Train Document Length Statistics:\")\n",
        "print(\"Mean Length:\", sum(train_document_lengths) / len(train_document_lengths))\n",
        "print(\"Max Length:\", max(train_document_lengths))\n",
        "print(\"Min Length:\", min(train_document_lengths))\n",
        "print(\"\\nTest Document Length Statistics:\")\n",
        "print(\"Mean Length:\", sum(test_document_lengths) / len(test_document_lengths))\n",
        "print(\"Max Length:\", max(test_document_lengths))\n",
        "print(\"Min Length:\", min(test_document_lengths))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 599
        },
        "id": "XeOHCzzgNxNW",
        "outputId": "18f9152a-6baa-4d63-f3f3-3cbb6e49ed92"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x1200 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Sample Data"
      ],
      "metadata": {
        "id": "HEXIYpqLrWh8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Sample 10% of the original train and test datasets\n",
        "train_sample = dataset[\"train\"].train_test_split(test_size=0.1)[\"test\"]  # Keep 10% of the training set\n",
        "test_sample = dataset[\"test\"].train_test_split(test_size=0.1)[\"test\"]    # Keep 10% of the test set\n",
        "\n",
        "# Define a function to format the data into roles for fine-tuning\n",
        "def format_for_fine_tuning(dataset_sample):\n",
        "    formatted_data = []\n",
        "\n",
        "    for i in range(len(dataset_sample)):\n",
        "        question = f\"{dataset_sample[i]['question_title']} {dataset_sample[i]['question_content']}\".strip()\n",
        "        answer = dataset_sample[i]['best_answer']\n",
        "\n",
        "        formatted_data.append([\n",
        "            {\"role\": \"system\", \"content\": \"You are an assistant who helps answer questions from Yahoo Answers users.\"},\n",
        "            {\"role\": \"user\", \"content\": question},\n",
        "            {\"role\": \"assistant\", \"content\": answer}\n",
        "        ])\n",
        "\n",
        "    return formatted_data\n",
        "\n",
        "# Format train and test datasets separately\n",
        "formatted_train_data = format_for_fine_tuning(train_sample)\n",
        "formatted_test_data = format_for_fine_tuning(test_sample)\n",
        "\n",
        "# Print a sample formatted entry from train and test data to inspect the structure\n",
        "print(\"Train Sample Entry:\")\n",
        "print(json.dumps(formatted_train_data[0], indent=2))\n",
        "\n",
        "print(\"\\nTest Sample Entry:\")\n",
        "print(json.dumps(formatted_test_data[0], indent=2))\n",
        "\n",
        "# Save the formatted data to JSONL files\n",
        "def save_to_jsonl(data, filename):\n",
        "    with open(filename, 'w') as f:\n",
        "        for entry in data:\n",
        "            f.write(json.dumps(entry) + '\\n')\n",
        "\n",
        "# Save train and test data separately\n",
        "save_to_jsonl(formatted_train_data, \"formatted_yahoo_answers_train.jsonl\")\n",
        "save_to_jsonl(formatted_test_data, \"formatted_yahoo_answers_test.jsonl\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KVbrDEpMqCAB",
        "outputId": "f8a49148-2130-4a58-97fb-768a4ac7a843"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Document Length Statistics:\n",
            "Mean Length: 32.395272142857145\n",
            "Max Length: 915\n",
            "Min Length: 1\n",
            "\n",
            "Test Document Length Statistics:\n",
            "Mean Length: 32.36026666666667\n",
            "Max Length: 760\n",
            "Min Length: 1\n",
            "Train Sample Entry:\n",
            "[\n",
            "  {\n",
            "    \"role\": \"system\",\n",
            "    \"content\": \"You are an assistant who helps answer questions from Yahoo Answers users.\"\n",
            "  },\n",
            "  {\n",
            "    \"role\": \"user\",\n",
            "    \"content\": \"Who are the known manufacturers of flame retardants for textile applications? These are chemicals used by textile manufacturers to make fabrics and home textiles flame resistant.\"\n",
            "  },\n",
            "  {\n",
            "    \"role\": \"assistant\",\n",
            "    \"content\": \"http://www.aatcc.org/bg/\"\n",
            "  }\n",
            "]\n",
            "\n",
            "Test Sample Entry:\n",
            "[\n",
            "  {\n",
            "    \"role\": \"system\",\n",
            "    \"content\": \"You are an assistant who helps answer questions from Yahoo Answers users.\"\n",
            "  },\n",
            "  {\n",
            "    \"role\": \"user\",\n",
            "    \"content\": \"Do you know some OCR programs for linux (mandriva or Suse)? i read somewhere, that abby fine reader 6 have linux version, but it is just the engine. i have to recognize hungarian caracters.\"\n",
            "  },\n",
            "  {\n",
            "    \"role\": \"assistant\",\n",
            "    \"content\": \"Looks like gocr is your best bet... there may be a KDE app called Kooka which does OCR as post-processing on a scan, but for simple file to text, gocr seems best.\"\n",
            "  }\n",
            "]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The **output format** we're dealing with is essentially a **list of dictionaries**, where each dictionary contains key-value pairs that define the **role** (e.g., system, user, assistant) and the **content** for that role. Let's break down what this means and how it relates to fine-tuning a **Large Language Model (LLM)**.\n",
        "\n",
        "#### JSONL: List of Dictionaries\n",
        "\n",
        "1. **List of Dictionaries**:\n",
        "   - In the Python code, each conversation is represented as a **list of dictionaries**.\n",
        "   - Each dictionary contains the following keys:\n",
        "     - **\"role\"**: This indicates the role in the conversation, such as `\"system\"`, `\"user\"`, or `\"assistant\"`.\n",
        "     - **\"content\"**: This contains the actual text for that role (e.g., the instructions from the system, the question from the user, or the response from the assistant).\n",
        "   \n",
        "   In this example, the list contains **three dictionaries**:\n",
        "   - The first dictionary is the **system message** that gives the assistant context about its role.\n",
        "   - The second dictionary is the **user's question**.\n",
        "   - The third dictionary is the **assistant's response**.\n",
        "\n",
        "### JSON vs. JSONL\n",
        "\n",
        "- **JSON (JavaScript Object Notation)**:\n",
        "  - A **JSON** file is typically used for structured data storage where you might have a single, cohesive object or an array of objects. JSON files are often used when you want to load all the data at once.\n",
        "  - The format of the list of dictionaries can easily be saved to a JSON file, but the entire structure would be loaded into memory at once, which may be problematic for larger datasets.\n",
        "\n",
        "- **JSONL (JSON Lines)**:\n",
        "  - In a **JSONL** file, each **line** is a complete JSON object (in this case, a conversation in the list of dictionaries).\n",
        "  - By saving each conversation as a separate line, **JSONL** helps keep memory usage low since you can read and process each line individually rather than loading the entire file at once.\n",
        "  - For example, if you save each conversation as a line in the JSONL file, you can stream it efficiently:\n",
        "    ```json\n",
        "    [{\"role\": \"system\", \"content\": \"You are an assistant who helps answer questions from Yahoo Answers users.\"}, {\"role\": \"user\", \"content\": \"How can I get better at math?\"}, {\"role\": \"assistant\", \"content\": \"To improve in algebra, practice is key...\"}]\n",
        "    [{\"role\": \"system\", \"content\": \"You are an assistant who helps answer questions from Yahoo Answers users.\"}, {\"role\": \"user\", \"content\": \"How to cook spaghetti squash?\"}, {\"role\": \"assistant\", \"content\": \"To cook spaghetti squash, cut it in half...\"}]\n",
        "    ```\n",
        "\n",
        "### Relation to Fine-Tuning an LLM\n",
        "\n",
        "To **fine-tune a Large Language Model (LLM)**, you need to provide a dataset that contains examples of conversational pairs or dialogues that the model can learn from. The format we've discussed—**lists of dictionaries** with roles (`\"system\"`, `\"user\"`, `\"assistant\"`)—is ideal for fine-tuning models like those in the **Hugging Face Transformers library** or **OpenAI's GPT models**, which rely on understanding conversation structure.\n",
        "\n",
        "Here's how it relates to fine-tuning:\n",
        "\n",
        "1. **Roles in Conversations**:\n",
        "   - The roles (`\"system\"`, `\"user\"`, `\"assistant\"`) are crucial for instructing the LLM on how to behave in each part of the conversation.\n",
        "   - The **\"system\"** message sets the context, telling the model its purpose or role.\n",
        "   - The **\"user\"** message simulates what a user might say or ask.\n",
        "   - The **\"assistant\"** message provides the response that the model should learn to give.\n",
        "\n",
        "2. **Training Data Format**:\n",
        "   - When training the LLM, the dataset is typically provided as sequences of interactions. The roles and content help the model understand what to expect and how to respond.\n",
        "   - By providing a structured dataset with **context-response pairs**, you help the model learn not just how to answer questions but also the context in which it should give those answers.\n",
        "  \n",
        "3. **Efficient Training Using JSONL**:\n",
        "   - During fine-tuning, instead of loading the entire dataset into memory, the **JSONL** format allows for efficient streaming. The training script can read each conversation from the JSONL file line by line, which is memory efficient and suitable for training large models with limited hardware resources.\n",
        "   - Many frameworks, such as **Hugging Face's `datasets` library**, allow direct reading from JSONL files, making it simple to use these formatted data for training.\n",
        "\n"
      ],
      "metadata": {
        "id": "cvdQAoQasiRB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Print the first five entries of the formatted train data\n",
        "print(\"First 5 entries of formatted train data:\")\n",
        "for i in range(5):\n",
        "    print(formatted_train_data[i])\n",
        "\n",
        "# Print the first five entries of the formatted test data\n",
        "print(\"\\nFirst 5 entries of formatted test data:\")\n",
        "for i in range(5):\n",
        "    print(formatted_test_data[i])\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hwj1LlTuqB6k",
        "outputId": "28868009-240f-40a9-fbeb-4f32def7edc2"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "First 5 entries of formatted train data:\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': 'Who are the known manufacturers of flame retardants for textile applications? These are chemicals used by textile manufacturers to make fabrics and home textiles flame resistant.'}, {'role': 'assistant', 'content': 'http://www.aatcc.org/bg/'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': \"i wish wolf wood didn't have to die on trigun , do you agree?\"}, {'role': 'assistant', 'content': \"i know , he shoundn't of died\"}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': 'who is in my left hand?'}, {'role': 'assistant', 'content': 'Thumbalina? lol'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': 'How do you cite the Bible in a bibliography? I wrote a paper for my world religions class that included quotations from the Bible.  Do I have to include this in my bibliography, and if so, how do I cite in MLA style?'}, {'role': 'assistant', 'content': 'editor, date published, page, all that good stuff.  all the info should be on the back of the title page. follow \"book no author\" guidelines.\\\\n\\\\nThe Bible (specific editions)\\\\nGive the name of the specific edition, any editor(s) associated with it, followed by the publication information\\\\n\\\\nThe New Jerusalem Bible. Susan Jones, gen. ed. New York: Doubleday, 1985.\\\\n\\\\nYour parenthetical citation will include the name of the specific edition of the Bible, followed by an abbreviation of the book and chapter:verse(s), e.g., (The New Jerusalem Bible Gen. 1:2-6).'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': \"I need to find a good dentist in Temecula or Murrieta? I have used one but wasn't happy with her.  Does anyone have a great dentist in Temecula or Murrieta that's good with kids also?  Thanks!\"}, {'role': 'assistant', 'content': 'Call the ADA - American Dental Association for a recommendation'}]\n",
            "\n",
            "First 5 entries of formatted test data:\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': 'Do you know some OCR programs for linux (mandriva or Suse)? i read somewhere, that abby fine reader 6 have linux version, but it is just the engine. i have to recognize hungarian caracters.'}, {'role': 'assistant', 'content': 'Looks like gocr is your best bet... there may be a KDE app called Kooka which does OCR as post-processing on a scan, but for simple file to text, gocr seems best.'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': \"what characteristics of the market for operating system do you think created the monopoly market? What characteristics of the market for operating systems do you think create d\\\\nthe monopoly market that Microsoft's operating system enjoyed? Evaluat e\\\\nthis market in terms of utilitarianism, rights, and justice, giving explici t\\\\nexamples from the operating systems industry to illustrate your points .\"}, {'role': 'assistant', 'content': 'Sorry... Not going to write your paper for you.\\\\n\\\\nGood luck, though!'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': \"Is this wrong? ok So my sister is getting her baby babtized..and we are methodist!..but she is getting him babtized in a Lutheran church just b/c thats where she has been going lately..w/ the baby's other g-ma.which I do not like..well first off what exactly is Lutheran anyway..and is it bad that I dont like it that shes going there??? should it matter?\"}, {'role': 'assistant', 'content': 'You should really talk it over with your sister, and read up on Lutherans on www.wikipedia.org.'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': 'what is the difference from an outlet and its original store?'}, {'role': 'assistant', 'content': 'An outlet usually sells out of date or overstocked items.  While the original store sells items currently on demand.  You can get a lower price at the outlet store.'}]\n",
            "[{'role': 'system', 'content': 'You are an assistant who helps answer questions from Yahoo Answers users.'}, {'role': 'user', 'content': \"Is anything wrong with this? I am 5'3 and 175 pounds, alot of people say I am fat, but my stomach is small, I have a round face(like Queen Latifah) and I have big legs, I am also very top heavy.\"}, {'role': 'assistant', 'content': 'Your body sounds just fine....not at all a problem.  But you may want to check where your insecurity comes from.  You should feel as comfortable with yourself as you seem to look.'}]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Class Distribution after Sampling"
      ],
      "metadata": {
        "id": "QnPNAwAfvDCP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate value counts for labels in sampled train and test datasets\n",
        "train_label_counts = pd.Series(train_sample['topic']).value_counts().sort_index()\n",
        "test_label_counts = pd.Series(test_sample['topic']).value_counts().sort_index()\n",
        "train_label_counts.index = label_names  # Map numeric labels to topic names\n",
        "test_label_counts.index = label_names   # Map numeric labels to topic names\n",
        "\n",
        "# Combine 'question_title' and 'question_content' for document lengths in sampled train and test datasets\n",
        "train_documents = [f\"{title} {content}\" for title, content in zip(train_sample['question_title'], train_sample['question_content'])]\n",
        "test_documents = [f\"{title} {content}\" for title, content in zip(test_sample['question_title'], test_sample['question_content'])]\n",
        "train_document_lengths = [len(doc.split()) for doc in train_documents]  # Calculate document lengths for train sample\n",
        "test_document_lengths = [len(doc.split()) for doc in test_documents]    # Calculate document lengths for test sample\n",
        "\n",
        "# Create a 2x2 grid for the plots\n",
        "fig, axes = plt.subplots(2, 2, figsize=(18, 12))\n",
        "\n",
        "# Set up a color palette\n",
        "palette = sns.color_palette(\"Pastel1\", 10)\n",
        "\n",
        "# Plot 1: Train Label Distribution\n",
        "axes[0, 0].bar(train_label_counts.index, train_label_counts.values, color=palette)\n",
        "axes[0, 0].set_title(\"Yahoo Answers Topics Distribution (Train Sample)\")\n",
        "axes[0, 0].set_xlabel(\"Category\")\n",
        "axes[0, 0].set_ylabel(\"Count\")\n",
        "axes[0, 0].set_xticks(range(len(train_label_counts.index)))\n",
        "axes[0, 0].set_xticklabels(train_label_counts.index, rotation=45, ha='right')\n",
        "\n",
        "# Plot 2: Test Label Distribution\n",
        "axes[0, 1].bar(test_label_counts.index, test_label_counts.values, color=palette)\n",
        "axes[0, 1].set_title(\"Yahoo Answers Topics Distribution (Test Sample)\")\n",
        "axes[0, 1].set_xlabel(\"Category\")\n",
        "axes[0, 1].set_ylabel(\"Count\")\n",
        "axes[0, 1].set_xticks(range(len(test_label_counts.index)))\n",
        "axes[0, 1].set_xticklabels(test_label_counts.index, rotation=45, ha='right')\n",
        "\n",
        "# Plot 3: Train Document Length Distribution\n",
        "axes[1, 0].hist(train_document_lengths, bins=30, color='skyblue', alpha=0.7)\n",
        "axes[1, 0].set_title(\"Distribution of Document Lengths in Train Sample\")\n",
        "axes[1, 0].set_xlabel(\"Number of Words\")\n",
        "axes[1, 0].set_ylabel(\"Frequency\")\n",
        "\n",
        "# Plot 4: Test Document Length Distribution\n",
        "axes[1, 1].hist(test_document_lengths, bins=30, color='skyblue', alpha=0.7)\n",
        "axes[1, 1].set_title(\"Distribution of Document Lengths in Test Sample\")\n",
        "axes[1, 1].set_xlabel(\"Number of Words\")\n",
        "axes[1, 1].set_ylabel(\"Frequency\")\n",
        "\n",
        "# Adjust layout and show the plots\n",
        "plt.tight_layout()\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 599
        },
        "id": "iMdcCGAmqB3P",
        "outputId": "71eec3c1-43b0-4be7-ec25-cd412320c2bd"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x1200 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Tokenize the Data"
      ],
      "metadata": {
        "id": "K4bvnTqqe3fe"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from transformers import AutoTokenizer\n",
        "import json\n",
        "\n",
        "# Load the tokenizer\n",
        "tokenizer = AutoTokenizer.from_pretrained(\"distilbert-base-uncased\")\n",
        "\n",
        "# Define a function to tokenize the data entries\n",
        "def tokenize_data(entry):\n",
        "    # Extract user text and assistant response from the entry\n",
        "    user_text = entry[1][\"content\"]  # User message (index 1)\n",
        "    assistant_response = entry[2][\"content\"]  # Assistant response (index 2)\n",
        "\n",
        "    # Tokenize the user content\n",
        "    tokenized_output = tokenizer(user_text, padding=\"max_length\", truncation=True, max_length=128)\n",
        "\n",
        "    # Add the label as an integer (in this context, the label will be the tokenized assistant response)\n",
        "    # Convert the assistant response into tokens and add as labels for the model\n",
        "    tokenized_output[\"labels\"] = [\n",
        "        label if label != tokenizer.pad_token_id else -100\n",
        "        for label in tokenizer(assistant_response, padding=\"max_length\", truncation=True, max_length=128)[\"input_ids\"]\n",
        "    ]\n",
        "\n",
        "    # Convert BatchEncoding to a dictionary\n",
        "    return {key: value for key, value in tokenized_output.items()}\n",
        "\n",
        "# Save tokenized data to JSONL\n",
        "def save_tokenized_data(tokenized_data, filepath):\n",
        "    with open(filepath, \"w\") as f:\n",
        "        for entry in tokenized_data:\n",
        "            f.write(json.dumps(entry) + \"\\n\")\n",
        "\n",
        "# Tokenize and save train data\n",
        "tokenized_train_data = [tokenize_data(entry) for entry in formatted_train_data]\n",
        "save_tokenized_data(tokenized_train_data, \"/content/tokenized_yahoo_answers_train.jsonl\")\n",
        "\n",
        "# Tokenize and save test data\n",
        "tokenized_test_data = [tokenize_data(entry) for entry in formatted_test_data]\n",
        "save_tokenized_data(tokenized_test_data, \"/content/tokenized_yahoo_answers_test.jsonl\")\n",
        "\n",
        "# Print the first tokenized train data entry to verify\n",
        "print(\"First tokenized train data entry:\")\n",
        "print(tokenized_train_data[0])\n",
        "\n",
        "print(\"First tokenized test data entry:\")\n",
        "print(tokenized_test_data[0])\n",
        "\n",
        "# Check the structure of the data\n",
        "print(\"Training Dataset Sample:\\n\", pd.DataFrame(tokenized_train_data).head(10))\n",
        "print(\"\\nTest Dataset Sample:\\n\", pd.DataFrame(tokenized_test_data).head(10))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "c2580945158b422caa423e76cdd573d9",
            "40b36812d8d2493282f211a510a5f7b5",
            "40dc21e1b36243faa6f0abe427284226",
            "0f50c00ee2ea4e45b3c444ae9355c734",
            "1e3ec3866c494c098b62b2e3ebb33205",
            "798095c00c03485d9f378a8783326cde",
            "36a85cba9df1435899efd2786faab78a",
            "a5d6fadc6e0f448ab7b43fc999f31c24",
            "868dac39cc0d42c8a1df15167bd2adee",
            "a4ed14d1c1484ad4aad0a612d6b0625c",
            "45bfcb2333284078b2c1cc85a750be8c",
            "bab2461d3f5c4addb19dbf47aceb77c9",
            "4752735ed522454a8a339c99e71aca7d",
            "95d26c29f32447dc8901481d4d3734ae",
            "229396e7fda14b8eb556f5c1bb35ed40",
            "ebd72d0f763447a49ff69826ffa8348d",
            "58093cad7a3041bdb456dabf994c5208",
            "4b8d3d0753cc4d0793f015f2646b069f",
            "4c607d24a3aa440480597dfa27ed1ea6",
            "9c99c29e66e84eaaa2e0702485893700",
            "5f9286e8ab09438fadbcb33bcead6904",
            "a465c2910ad64f7f9081e8dd23feceb7",
            "a154b78f5daf43eb924bcb6742791dc0",
            "a7ec3a5d652b48c097a73aad7d82fc80",
            "aa84a43e9b2148cea62f458a414a5950",
            "be4647671cc549b3a71d369d019ec5ba",
            "27adb0ed45ce48fea17c4796ea388fff",
            "2bc18ae2235f42f08f4b078ae6711a14",
            "9db077d5337d4257b705817f38b4bec4",
            "41bbe91682e84d4da0a37fd75c4ff3d6",
            "517eefd33cfc42d4836727647128912e",
            "ef05935f58eb40128a41c21775a78a92",
            "ac16c8ee638a49afa538a1857ad38695",
            "8141bd6ae9f74f878cdabadbaaa97536",
            "3a43077a3d724cf085c04b1f52e12035",
            "3521832224f7493688958c0cf19d6382",
            "bb77e0497871410abe924d7ee181459d",
            "28925eab0b2143e488b147c035763d77",
            "fc3ef2e169c44fb1a21d47c74e32cf10",
            "99b43b1c89964c70892b44b63f1b10f0",
            "59920481bdda4295a3498c6a3c89a66b",
            "a631d378dd88424eb37b0cae5ca5d89f",
            "ffe39811e4b2473a8cbb2651b51869b8",
            "f5d47406b1a6497181f64a2da8fd6607"
          ]
        },
        "id": "7fsCm8mc0kn5",
        "outputId": "41cd3c6e-9ca5-40f8-e681-e8d079bffc1c"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "tokenizer_config.json:   0%|          | 0.00/48.0 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "c2580945158b422caa423e76cdd573d9"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "config.json:   0%|          | 0.00/483 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "bab2461d3f5c4addb19dbf47aceb77c9"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "a154b78f5daf43eb924bcb6742791dc0"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "tokenizer.json:   0%|          | 0.00/466k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "8141bd6ae9f74f878cdabadbaaa97536"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "First tokenized train data entry:\n",
            "{'input_ids': [101, 2040, 2024, 1996, 2124, 8712, 1997, 8457, 2128, 7559, 28210, 2015, 2005, 12437, 5097, 1029, 2122, 2024, 12141, 2109, 2011, 12437, 8712, 2000, 2191, 25123, 1998, 2188, 18762, 8457, 13070, 1012, 102, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'labels': [101, 8299, 1024, 1013, 1013, 7479, 1012, 9779, 13535, 2278, 1012, 8917, 1013, 1038, 2290, 1013, 102, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100]}\n",
            "First tokenized test data entry:\n",
            "{'input_ids': [101, 2079, 2017, 2113, 2070, 1051, 26775, 3454, 2005, 11603, 1006, 2158, 13626, 11444, 2030, 10514, 3366, 1007, 1029, 1045, 3191, 4873, 1010, 2008, 9460, 2986, 8068, 1020, 2031, 11603, 2544, 1010, 2021, 2009, 2003, 2074, 1996, 3194, 1012, 1045, 2031, 2000, 6807, 5588, 14418, 21162, 2015, 1012, 102, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'labels': [101, 3504, 2066, 2175, 26775, 2003, 2115, 2190, 6655, 1012, 1012, 1012, 2045, 2089, 2022, 1037, 1047, 3207, 10439, 2170, 12849, 12352, 2029, 2515, 1051, 26775, 2004, 2695, 1011, 6364, 2006, 1037, 13594, 1010, 2021, 2005, 3722, 5371, 2000, 3793, 1010, 2175, 26775, 3849, 2190, 1012, 102, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100]}\n",
            "Training Dataset Sample:\n",
            "                                            input_ids  \\\n",
            "0  [101, 2040, 2024, 1996, 2124, 8712, 1997, 8457...   \n",
            "1  [101, 1045, 4299, 4702, 3536, 2134, 1005, 1056...   \n",
            "2  [101, 2040, 2003, 1999, 2026, 2187, 2192, 1029...   \n",
            "3  [101, 2129, 2079, 2017, 21893, 1996, 6331, 199...   \n",
            "4  [101, 1045, 2342, 2000, 2424, 1037, 2204, 2438...   \n",
            "5  [101, 28194, 2232, 2003, 1996, 2190, 2126, 200...   \n",
            "6  [101, 2515, 3087, 2113, 2073, 1045, 2064, 2135...   \n",
            "7  [101, 2339, 2515, 1037, 22764, 13212, 13366, 1...   \n",
            "8  [101, 2003, 2045, 1037, 3698, 2008, 2064, 5468...   \n",
            "9  [101, 2055, 5088, 11408, 1012, 1012, 1012, 101...   \n",
            "\n",
            "                                      attention_mask  \\\n",
            "0  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "1  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "2  [1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, ...   \n",
            "3  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "4  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "5  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, ...   \n",
            "6  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "7  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, ...   \n",
            "8  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "9  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "\n",
            "                                              labels  \n",
            "0  [101, 8299, 1024, 1013, 1013, 7479, 1012, 9779...  \n",
            "1  [101, 1045, 2113, 1010, 2002, 26822, 8630, 207...  \n",
            "2  [101, 7639, 11475, 2532, 1029, 8840, 2140, 102...  \n",
            "3  [101, 3559, 1010, 3058, 2405, 1010, 3931, 1010...  \n",
            "4  [101, 2655, 1996, 15262, 1011, 2137, 11394, 25...  \n",
            "5  [101, 2444, 1998, 2292, 2444, 102, -100, -100,...  \n",
            "6  [101, 3046, 2115, 2334, 16215, 16338, 7340, 10...  \n",
            "7  [101, 2035, 15865, 2131, 9742, 2099, 2004, 202...  \n",
            "8  [101, 2009, 1005, 1055, 2170, 1037, 4834, 8316...  \n",
            "9  [101, 1045, 2123, 1005, 1056, 2066, 1996, 2028...  \n",
            "\n",
            "Test Dataset Sample:\n",
            "                                            input_ids  \\\n",
            "0  [101, 2079, 2017, 2113, 2070, 1051, 26775, 345...   \n",
            "1  [101, 2054, 6459, 1997, 1996, 3006, 2005, 4082...   \n",
            "2  [101, 2003, 2023, 3308, 1029, 7929, 2061, 2026...   \n",
            "3  [101, 2054, 2003, 1996, 4489, 2013, 2019, 1330...   \n",
            "4  [101, 2003, 2505, 3308, 2007, 2023, 1029, 1045...   \n",
            "5  [101, 2003, 4767, 11933, 2030, 18558, 7215, 20...   \n",
            "6  [101, 2339, 2987, 1005, 1056, 20643, 6998, 203...   \n",
            "7  [101, 2024, 2017, 1037, 10118, 17167, 8013, 10...   \n",
            "8  [101, 2129, 2052, 2017, 10509, 2065, 4441, 259...   \n",
            "9  [101, 2043, 1037, 6965, 8289, 1010, 2515, 2014...   \n",
            "\n",
            "                                      attention_mask  \\\n",
            "0  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "1  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "2  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "3  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, ...   \n",
            "4  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "5  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "6  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "7  [1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, ...   \n",
            "8  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "9  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
            "\n",
            "                                              labels  \n",
            "0  [101, 3504, 2066, 2175, 26775, 2003, 2115, 219...  \n",
            "1  [101, 3374, 1012, 1012, 1012, 2025, 2183, 2000...  \n",
            "2  [101, 2017, 2323, 2428, 2831, 2009, 2058, 2007...  \n",
            "3  [101, 2019, 13307, 2788, 15187, 2041, 1997, 30...  \n",
            "4  [101, 2115, 2303, 4165, 2074, 2986, 1012, 1012...  \n",
            "5  [101, 2009, 2003, 13418, 2008, 12290, 2097, 89...  \n",
            "6  [101, 1045, 2123, 1005, 1056, 2113, 1012, 2021...  \n",
            "7  [101, 2053, 10118, 17167, 2105, 2182, 1012, 10...  \n",
            "8  [101, 1012, 1012, 1012, 1045, 1005, 1040, 2681...  \n",
            "9  [101, 2429, 2000, 1996, 2184, 3094, 8163, 1006...  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Converting your tokenized data to a Hugging Face `Dataset` is an important step for several reasons, especially if you're planning to use Hugging Face’s `Trainer` API for model fine-tuning. Let’s break down this step and why it’s useful.\n",
        "\n",
        "### Convert to a Hugging Face Dataset\n",
        "\n",
        "1. **Compatibility with Hugging Face Trainer**: Hugging Face's `Trainer` class expects the training and evaluation data to be in the form of a `Dataset` (from the `datasets` library). This format allows seamless integration with the `Trainer` API, making it straightforward to define and run training, evaluation, and even tokenized data manipulation.\n",
        "\n",
        "2. **Efficiency and Scalability**: Hugging Face's `Dataset` object is optimized for large-scale datasets, enabling efficient handling of in-memory and memory-mapped data. It’s designed to handle transformations, batching, shuffling, and slicing efficiently, especially when working with large volumes of tokenized data.\n",
        "\n",
        "3. **Preprocessing Flexibility**: The `Dataset` object provides easy methods for on-the-fly transformations, such as tokenization or data augmentation. It’s very flexible, allowing you to apply these transformations directly to each entry while maintaining an efficient workflow.\n",
        "\n",
        "4. **Integration with Data Collators and Tokenizers**: When you load tokenized data into a `Dataset` object, you can leverage Hugging Face’s data collators, which are utilities that help pad and batch the data dynamically. This step makes the data ready for model consumption without manually adjusting batch sizes or paddings.\n",
        "\n"
      ],
      "metadata": {
        "id": "hYjn0TV44CAk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from datasets import load_dataset, DatasetDict\n",
        "\n",
        "# Load the tokenized train and test datasets directly from JSONL\n",
        "train_dataset = load_dataset(\"json\", data_files=\"/content/tokenized_yahoo_answers_train.jsonl\")[\"train\"]\n",
        "test_dataset = load_dataset(\"json\", data_files=\"/content/tokenized_yahoo_answers_test.jsonl\")[\"train\"]\n",
        "\n",
        "# Combine train and test datasets into a DatasetDict\n",
        "dataset_dict = DatasetDict({\n",
        "    \"train\": train_dataset,\n",
        "    \"test\": test_dataset\n",
        "})\n",
        "\n",
        "# Verify the dataset structure\n",
        "print(dataset_dict)\n",
        "\n",
        "# Inspect a few samples\n",
        "print(\"Training Dataset Sample:\\n\", train_dataset.select(range(10)))\n",
        "print(\"\\nTest Dataset Sample:\\n\", test_dataset.select(range(10)))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 445,
          "referenced_widgets": [
            "68ed145e8f5a4f1fb8c83f59a66615f8",
            "5bcfc6a9b77c4112a906fbaf0f96f399",
            "37cde7ffa3574dd5a2dfa9e4f048ca58",
            "eaf977aba6e548ae8cd9bae0e56bc27b",
            "5c626441ad1b43dab02ed43dcfca58fd",
            "370491d99b9d4dbbadb508f7905943b4",
            "ce50b44c4f7e441cbfb65d8b842a98e8",
            "de65421ac13a412f93dd61a6f5ed13eb",
            "4022ae749ab74756a432cc9010001690",
            "5325ea426fe8491d8586532c5160d2b7",
            "1a7ed75b7c014b5eadbcc9eea1b46e07",
            "68ae0a4b753a4159936d3d22e617f91d",
            "596b687080d84890b2621af3d0491199",
            "41ea2cab4d414840ba8ff92001b0e975",
            "bb39f1375537473983135ac4d2ea569c",
            "16c399425fa045218a9415e16971ac15",
            "6b4f443e49524901abaf8405fb64870b",
            "3ebebadff6484dc284d3f5eaa70c4d7d",
            "d929bbf433e747afa3bc10dc06149bd3",
            "db200f517cc3497b9f76db672a1c5380",
            "fe1c305004b343e993dd64dbd6a45b8e",
            "068e873911b3435788ff73d078e78269"
          ]
        },
        "id": "9VFOK4ds2xCF",
        "outputId": "974e6763-1f8d-4560-96ee-68bff3b36aa8"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train split: 0 examples [00:00, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "68ed145e8f5a4f1fb8c83f59a66615f8"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Generating train split: 0 examples [00:00, ? examples/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "68ae0a4b753a4159936d3d22e617f91d"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "DatasetDict({\n",
            "    train: Dataset({\n",
            "        features: ['input_ids', 'attention_mask', 'labels'],\n",
            "        num_rows: 140000\n",
            "    })\n",
            "    test: Dataset({\n",
            "        features: ['input_ids', 'attention_mask', 'labels'],\n",
            "        num_rows: 6000\n",
            "    })\n",
            "})\n",
            "Training Dataset Sample:\n",
            " Dataset({\n",
            "    features: ['input_ids', 'attention_mask', 'labels'],\n",
            "    num_rows: 10\n",
            "})\n",
            "\n",
            "Test Dataset Sample:\n",
            " Dataset({\n",
            "    features: ['input_ids', 'attention_mask', 'labels'],\n",
            "    num_rows: 10\n",
            "})\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Tune the Model\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "DDnJ4MnFpeO8"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Load the Pretrained Model"
      ],
      "metadata": {
        "id": "PROYSEQzqNiS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from transformers import T5ForConditionalGeneration\n",
        "\n",
        "# Load a pretrained model suitable for conversational response generation\n",
        "model = T5ForConditionalGeneration.from_pretrained(\"t5-small\")\n"
      ],
      "metadata": {
        "id": "VQ20V3yzVQZc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 113,
          "referenced_widgets": [
            "7f69f2e4a52741c1ab161f0a1a375636",
            "4667d2f258574f2c961595f759cf13fe",
            "a61f487ca142444c846bafc80c03dc8c",
            "c0683729a1d547f1a0f345ac6c557df3",
            "346ae85308d440a3bdcfc8047a4e8209",
            "38fa21b9b0fb47d88780a1ed403cc744",
            "8b073cad71064423885608ab33dce9aa",
            "a349c380093744e3849f5f2182c24df8",
            "89fc10a59f5f4b14855fdbec82f3c3b1",
            "b729ad76a9bc4782b311b86df6ca9f80",
            "d25d2d7b6e1e4c04bb87755eb7e29361",
            "effc347e877d41b88b78150aa3f8c873",
            "ac3add7b79bb4186a50b1799a4de14e1",
            "c910f8838578409bb4727f73b3c12c03",
            "3a1b4f1a43de43daad938bcb1380ef34",
            "eea701dcb08943d6bb02e3a2f61f6623",
            "4902199dfe6c4d14a08075531dce2cea",
            "235b0904cbab4ec38d51aa9ade8877be",
            "ce396907707a4f5eb1613524bd919edb",
            "5368f876e4664a258e4e5be34c4cf413",
            "d14c0ccb3d56447d91d13b63ecf82318",
            "cb96f2e0a9cd41a38a5e9347b85aa122",
            "cf3f9c0f29e24244a9510912e29ddcee",
            "d13753328dcd4492bef5834a6c1d31c6",
            "649e106653364f69988179d15990ec20",
            "3ba49d3ff83047e1aaaedf45681e32ec",
            "5a6691ddb4be4414b01e6d65b2ca419c",
            "3d508ce0ae4b48d4bffa1f5e82385dc3",
            "f9f44323585f4fb08cd3d6ff0676447c",
            "75a1d02fe21f4a37bd7ab03a65d3a71b",
            "2f9e90298fc3411abb090d8655a20c39",
            "2791f03f203f400abbb37af7ef86862a",
            "b6734b27d72444358e16d967bda844c0"
          ]
        },
        "outputId": "e5c10815-518a-4af8-f371-e7ac5e7936af"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "config.json:   0%|          | 0.00/1.21k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "7f69f2e4a52741c1ab161f0a1a375636"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "model.safetensors:   0%|          | 0.00/242M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "effc347e877d41b88b78150aa3f8c873"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "generation_config.json:   0%|          | 0.00/147 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cf3f9c0f29e24244a9510912e29ddcee"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Define Training Arguments and Train\n"
      ],
      "metadata": {
        "id": "V5uv8QcbqHbY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# from transformers import DataCollatorForSeq2Seq, T5ForConditionalGeneration, Seq2SeqTrainer, Seq2SeqTrainingArguments\n",
        "# import time\n",
        "\n",
        "# # Disable Weights & Biases logging\n",
        "# os.environ[\"WANDB_DISABLED\"] = \"true\"\n",
        "\n",
        "# # Load a pretrained T5 model suitable for conversational response generation\n",
        "# model = T5ForConditionalGeneration.from_pretrained(\"t5-small\")\n",
        "\n",
        "# # Define a data collator to dynamically pad sequences during training\n",
        "# data_collator = DataCollatorForSeq2Seq(tokenizer=tokenizer, model=model)\n",
        "\n",
        "# # Define training arguments\n",
        "# training_args = Seq2SeqTrainingArguments(\n",
        "#     output_dir=\"./results\",\n",
        "#     eval_strategy=\"epoch\",\n",
        "#     learning_rate=2e-5,\n",
        "#     per_device_train_batch_size=16,\n",
        "#     per_device_eval_batch_size=16,\n",
        "#     weight_decay=0.01,\n",
        "#     save_total_limit=3,\n",
        "#     num_train_epochs=1,\n",
        "#     predict_with_generate=True,\n",
        "#     report_to=\"none\"  # Disable W&B logging\n",
        "# )\n",
        "\n",
        "# # Initialize the Seq2SeqTrainer\n",
        "# trainer = Seq2SeqTrainer(\n",
        "#     model=model,\n",
        "#     args=training_args,\n",
        "#     train_dataset=dataset_dict[\"train\"],\n",
        "#     eval_dataset=dataset_dict[\"test\"],\n",
        "#     processing_class=tokenizer,\n",
        "#     data_collator=data_collator,\n",
        "# )\n",
        "\n",
        "# # Start the timer\n",
        "# start_time = time.time()\n",
        "\n",
        "# # Start training\n",
        "# trainer.train()\n",
        "\n",
        "# # End the timer and calculate duration\n",
        "# end_time = time.time()\n",
        "# duration = (end_time - start_time) / 60  # Convert seconds to minutes\n",
        "# print(f\"Training completed in {duration:.2f} minutes.\")\n"
      ],
      "metadata": {
        "id": "tPoAbv2ZuyOR"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Reduce Training Time\n",
        "\n",
        "There are several strategies you can employ to **reduce training time**. Below, I'll provide a few options you can adjust in your `training_args` and the training setup to help reduce the training duration:\n",
        "\n",
        "### 1. **Reduce the Number of Epochs**\n",
        "- You’ve already set `num_train_epochs` to `1`. This is good, but you can also make adjustments that allow you to get useful results even with fewer training steps.\n",
        "\n",
        "### 2. **Use a Smaller Subset of the Dataset**\n",
        "- If you want to train quickly to see if the setup works or to do initial debugging, **subsampling the dataset** is a great approach.\n",
        "\n",
        "  ```python\n",
        "  # Reduce the size of train and test datasets for quicker training\n",
        "  small_train_dataset = dataset_dict[\"train\"].shuffle(seed=42).select(range(1000))  # Use a subset of 1000 samples\n",
        "  small_test_dataset = dataset_dict[\"test\"].shuffle(seed=42).select(range(200))     # Use a subset of 200 samples\n",
        "  ```\n",
        "\n",
        "  This drastically reduces the dataset size, allowing the training to complete faster. Once everything is confirmed to be working well, you can gradually increase the dataset size.\n",
        "\n",
        "### 3. **Reduce `max_length` in Tokenization**\n",
        "- Reducing the maximum length of the input sequences can significantly reduce training time because smaller sequences are quicker to process.\n",
        "\n",
        "  ```python\n",
        "  # Reduce max_length to shorten input/output sequences\n",
        "  def tokenize_data(entry):\n",
        "      user_text = entry[1][\"content\"]\n",
        "      assistant_response = entry[2][\"content\"]\n",
        "      tokenized_output = tokenizer(user_text, padding=\"max_length\", truncation=True, max_length=64)  # Reduced from 128\n",
        "      tokenized_output[\"labels\"] = tokenizer(assistant_response, padding=\"max_length\", truncation=True, max_length=64)[\"input_ids\"]\n",
        "      return tokenized_output\n",
        "  ```\n",
        "  - Reducing `max_length` will lead to smaller tokenized sequences, meaning fewer computations.\n",
        "\n",
        "### 4. **Reduce the Batch Size**\n",
        "- Smaller batches require fewer resources per training step. You can try reducing `per_device_train_batch_size` and `per_device_eval_batch_size` if you are running into long training times or GPU/CPU memory issues.\n",
        "\n",
        "  ```python\n",
        "  training_args = Seq2SeqTrainingArguments(\n",
        "      output_dir=\"./results\",\n",
        "      evaluation_strategy=\"epoch\",\n",
        "      learning_rate=2e-5,\n",
        "      per_device_train_batch_size=8,  # Reduced from 16 to 8\n",
        "      per_device_eval_batch_size=8,   # Reduced from 16 to 8\n",
        "      weight_decay=0.01,\n",
        "      save_total_limit=3,\n",
        "      num_train_epochs=1,\n",
        "      predict_with_generate=True,\n",
        "      report_to=\"none\"\n",
        "  )\n",
        "  ```\n",
        "  - Reducing the batch size reduces the amount of data processed in each step, which can help speed up the training when resources are limited.\n",
        "\n",
        "### 5. **Use Gradient Accumulation**\n",
        "- If you reduce the batch size but still want the effect of a larger batch, you can use **gradient accumulation**.\n",
        "  \n",
        "  ```python\n",
        "  training_args = Seq2SeqTrainingArguments(\n",
        "      output_dir=\"./results\",\n",
        "      evaluation_strategy=\"epoch\",\n",
        "      learning_rate=2e-5,\n",
        "      per_device_train_batch_size=8,\n",
        "      per_device_eval_batch_size=8,\n",
        "      weight_decay=0.01,\n",
        "      save_total_limit=3,\n",
        "      num_train_epochs=1,\n",
        "      predict_with_generate=True,\n",
        "      report_to=\"none\",\n",
        "      gradient_accumulation_steps=2  # This effectively doubles the batch size\n",
        "  )\n",
        "  ```\n",
        "  - **Gradient accumulation** allows you to accumulate gradients for multiple mini-batches before performing a weight update. This effectively mimics a larger batch size without requiring more GPU memory at each training step.\n",
        "\n",
        "### 6. **Use FP16 Mixed Precision Training**\n",
        "- Mixed precision training allows you to use **16-bit floating point precision** instead of 32-bit. This reduces memory usage and speeds up training while still maintaining reasonable accuracy.\n",
        "\n",
        "  ```python\n",
        "  training_args = Seq2SeqTrainingArguments(\n",
        "      output_dir=\"./results\",\n",
        "      evaluation_strategy=\"epoch\",\n",
        "      learning_rate=2e-5,\n",
        "      per_device_train_batch_size=16,\n",
        "      per_device_eval_batch_size=16,\n",
        "      weight_decay=0.01,\n",
        "      save_total_limit=3,\n",
        "      num_train_epochs=1,\n",
        "      predict_with_generate=True,\n",
        "      report_to=\"none\",\n",
        "      fp16=True  # Enable mixed precision training\n",
        "  )\n",
        "  ```\n",
        "  - **FP16** training can provide significant speed improvements, especially on GPUs that support it (like NVIDIA Turing and newer architectures).\n",
        "\n",
        "### 7. **Reduce Model Complexity**\n",
        "- Consider using a **smaller model** if DistilBERT is taking too long.\n",
        "  - You can use `\"distilgpt2\"` instead of `\"t5-small\"` or other lightweight models to speed up training.\n",
        "\n",
        "  ```python\n",
        "  from transformers import GPT2LMHeadModel\n",
        "\n",
        "  # Load a smaller, faster model for conversational generation\n",
        "  model = GPT2LMHeadModel.from_pretrained(\"distilgpt2\")\n",
        "  ```\n",
        "\n",
        "  Smaller models require less computational power and will train faster.\n",
        "\n",
        "### Summary of Changes to Reduce Training Time\n",
        "\n",
        "- **Subsample the Dataset**: Use smaller subsets of the train and test data for quicker training.\n",
        "- **Reduce `max_length`**: Shorten the sequence length to reduce computation.\n",
        "- **Lower Batch Size**: Use smaller batch sizes to process less data per step.\n",
        "- **Enable Mixed Precision (FP16)**: Use half-precision training for faster processing.\n",
        "- **Use a Smaller Model**: Select a smaller pre-trained model if feasible.\n",
        "- **Gradient Accumulation**: Combine smaller batch sizes to get the effect of larger batches.\n",
        "\n",
        "By applying a combination of these changes, you should notice a significant reduction in training time. Once you confirm that everything is working properly and you are satisfied with the results, you can scale up the settings gradually.\n"
      ],
      "metadata": {
        "id": "cK2bLZmh6RV8"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Modified Code for Fast Training\n",
        "\n",
        "### Detailed Explanation:\n",
        "\n",
        "1. **Disable W&B Integration**:\n",
        "   ```python\n",
        "   os.environ[\"WANDB_DISABLED\"] = \"true\"\n",
        "   ```\n",
        "   - We set `WANDB_DISABLED` to `\"true\"` to disable any logging with Weights & Biases, so it won’t interfere with the training.\n",
        "\n",
        "2. **Use a Small Subset of the Dataset**:\n",
        "   ```python\n",
        "   small_train_data = tokenized_train_data[:100]  # Use only the first 100 samples for training\n",
        "   small_test_data = tokenized_test_data[:20]     # Use only the first 20 samples for testing\n",
        "   ```\n",
        "   - Use **100 samples** for training and **20 samples** for testing. This reduces the dataset to a very small subset, which will allow training to complete much faster.\n",
        "\n",
        "3. **Convert Data to DatasetDict**:\n",
        "   - Convert the small subsets to a **Hugging Face DatasetDict** object to easily handle train and test datasets.\n",
        "\n",
        "4. **Load a Small Model**:\n",
        "   ```python\n",
        "   model = T5ForConditionalGeneration.from_pretrained(\"t5-small\")\n",
        "   ```\n",
        "   - Load a small model like **T5-small** to minimize computational requirements. You can also consider even smaller models like `t5-tiny` if available.\n",
        "\n",
        "5. **Data Collator**:\n",
        "   - Set up a data collator with `DataCollatorForSeq2Seq` for handling padding dynamically during training.\n",
        "\n",
        "6. **Training Arguments for Quick Training**:\n",
        "   - **`evaluation_strategy=\"no\"`**: No evaluation during training, which saves time.\n",
        "   - **`learning_rate=1e-4`**: A slightly higher learning rate can help the model converge more quickly.\n",
        "   - **`per_device_train_batch_size=4`**: A small batch size allows for faster training and is easier on memory.\n",
        "   - **`num_train_epochs=1`**: Only **one epoch** for training.\n",
        "   - **`predict_with_generate=False`**: Disable predictions during training to save time.\n",
        "   - **`fp16=True`**: Enable **mixed precision training** for faster training on compatible GPUs.\n",
        "\n"
      ],
      "metadata": {
        "id": "gSS3io8Y7OG_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import pandas as pd\n",
        "from transformers import T5ForConditionalGeneration, Seq2SeqTrainer, Seq2SeqTrainingArguments, DataCollatorForSeq2Seq\n",
        "from datasets import Dataset, DatasetDict\n",
        "import warnings\n",
        "\n",
        "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
        "\n",
        "# Step 1: Disable Weights & Biases\n",
        "os.environ[\"WANDB_DISABLED\"] = \"true\"\n",
        "\n",
        "# Step 2: Load and Use a Small Subset of Tokenized Data\n",
        "small_train_data = tokenized_train_data[:100]  # Use only the first 100 samples for training\n",
        "small_test_data = tokenized_test_data[:20]     # Use only the first 20 samples for testing\n",
        "\n",
        "# Step 3: Convert Small Subset to DatasetDict\n",
        "train_dataset = Dataset.from_pandas(pd.DataFrame(small_train_data))\n",
        "test_dataset = Dataset.from_pandas(pd.DataFrame(small_test_data))\n",
        "dataset_dict = DatasetDict({\n",
        "    \"train\": train_dataset,\n",
        "    \"test\": test_dataset\n",
        "})\n",
        "\n",
        "# Step 4: Load a Small Model\n",
        "model = T5ForConditionalGeneration.from_pretrained(\"t5-small\")\n",
        "\n",
        "# Step 5: Tokenizer and Data Collator for Sequence-to-Sequence\n",
        "data_collator = DataCollatorForSeq2Seq(tokenizer=tokenizer, model=model)\n",
        "\n",
        "# Step 6: Define Training Arguments for Quick Training\n",
        "training_args = Seq2SeqTrainingArguments(\n",
        "    output_dir=\"./results\",\n",
        "    eval_strategy=\"no\",              # No evaluation during training to save time\n",
        "    learning_rate=1e-4,              # Increase learning rate to converge faster\n",
        "    per_device_train_batch_size=4,   # Reduce batch size to save memory\n",
        "    per_device_eval_batch_size=4,    # Same reduced batch size for evaluation\n",
        "    weight_decay=0.01,\n",
        "    num_train_epochs=1,              # Use just one epoch to save time\n",
        "    predict_with_generate=False,     # Turn off predictions during training\n",
        "    logging_steps=10,                # Reduce logging frequency\n",
        "    fp16=True,                       # Enable mixed precision training for faster computation\n",
        "    report_to=\"none\"                 # Disable W&B logging\n",
        ")\n",
        "\n",
        "# Step 7: Set Up the Trainer\n",
        "trainer = Seq2SeqTrainer(\n",
        "    model=model,\n",
        "    args=training_args,\n",
        "    train_dataset=dataset_dict[\"train\"],\n",
        "    eval_dataset=dataset_dict[\"test\"],\n",
        "    tokenizer=tokenizer,\n",
        "    data_collator=data_collator,\n",
        ")\n",
        "\n",
        "# Step 8: Train the Model\n",
        "start_time = time.time()\n",
        "\n",
        "# Start training\n",
        "trainer.train()\n",
        "\n",
        "# End the timer and calculate duration\n",
        "end_time = time.time()\n",
        "duration = (end_time - start_time) / 60  # Convert seconds to minutes\n",
        "print(f\"Training completed in {duration:.2f} minutes.\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 193
        },
        "id": "FI0aebCa7AAL",
        "outputId": "9bc7647e-bd80-4c82-86df-22b677077bb2"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Passing a tuple of `past_key_values` is deprecated and will be removed in Transformers v4.48.0. You should pass an instance of `EncoderDecoderCache` instead, e.g. `past_key_values=EncoderDecoderCache.from_legacy_cache(past_key_values)`.\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='25' max='25' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [25/25 02:21, Epoch 1/1]\n",
              "    </div>\n",
              "    <table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              " <tr style=\"text-align: left;\">\n",
              "      <th>Step</th>\n",
              "      <th>Training Loss</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>10</td>\n",
              "      <td>12.398500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>20</td>\n",
              "      <td>11.137100</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><p>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training completed in 2.51 minutes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "cd-dEHpEWAj8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "B8W6K7z-WAhH"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}