02. NLP with Transformers/02. Text Classification/Text Classification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Text Classification.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "a1a64d006b764dcc866249a7a7eafdbb": {
          "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_891a1507b33540c7a8202493122457f0",
              "IPY_MODEL_e63e5c22cdb4476ab6a45dc878ca18f0",
              "IPY_MODEL_4e2db58a98674de686337955ee5e319f"
            ],
            "layout": "IPY_MODEL_fd139f0ec816452289a6aafb43138d35"
          }
        },
        "891a1507b33540c7a8202493122457f0": {
          "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_370347f1f3864d1fb7f88c05e08cc195",
            "placeholder": "​",
            "style": "IPY_MODEL_e4147b31a0404959aef48349243360e5",
            "value": "100%"
          }
        },
        "e63e5c22cdb4476ab6a45dc878ca18f0": {
          "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_6a78f5367e4f40b78805498586d40970",
            "max": 3,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_c705825c91af4c17947682e22620c18a",
            "value": 3
          }
        },
        "4e2db58a98674de686337955ee5e319f": {
          "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_70c5283af6b347f58686215e83fb2857",
            "placeholder": "​",
            "style": "IPY_MODEL_3e7c6e3fd9e14944930beeb9585f1cbd",
            "value": " 3/3 [00:00&lt;00:00, 51.41it/s]"
          }
        },
        "fd139f0ec816452289a6aafb43138d35": {
          "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
          }
        },
        "370347f1f3864d1fb7f88c05e08cc195": {
          "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
          }
        },
        "e4147b31a0404959aef48349243360e5": {
          "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": ""
          }
        },
        "6a78f5367e4f40b78805498586d40970": {
          "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
          }
        },
        "c705825c91af4c17947682e22620c18a": {
          "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": ""
          }
        },
        "70c5283af6b347f58686215e83fb2857": {
          "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
          }
        },
        "3e7c6e3fd9e14944930beeb9585f1cbd": {
          "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": ""
          }
        },
        "83b56c2fbc07484180f401eae13ac0c5": {
          "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_d8434f53d2c648488d512b2d4e0436be",
              "IPY_MODEL_ed47be1f977d4e0b98671fb01c066c19",
              "IPY_MODEL_a61484b2a74b4275812e6511ca5e20b5"
            ],
            "layout": "IPY_MODEL_e6872a4539b34f529c182fe53039ba34"
          }
        },
        "d8434f53d2c648488d512b2d4e0436be": {
          "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_e089684da2b64e8faf6e52132edfd806",
            "placeholder": "​",
            "style": "IPY_MODEL_40edd0a0a8f24c338fc580d516c112ba",
            "value": "100%"
          }
        },
        "ed47be1f977d4e0b98671fb01c066c19": {
          "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_1cc170f2d6b94faa804f40a4f7a81267",
            "max": 16,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_dc2819bfe328440d9070d83ca4bef2a4",
            "value": 16
          }
        },
        "a61484b2a74b4275812e6511ca5e20b5": {
          "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_d79f1444a5604a30af28ede9eafd3112",
            "placeholder": "​",
            "style": "IPY_MODEL_0dba833f154f434fa2b4f0a776812fae",
            "value": " 16/16 [01:20&lt;00:00,  5.05s/ba]"
          }
        },
        "e6872a4539b34f529c182fe53039ba34": {
          "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
          }
        },
        "e089684da2b64e8faf6e52132edfd806": {
          "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
          }
        },
        "40edd0a0a8f24c338fc580d516c112ba": {
          "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": ""
          }
        },
        "1cc170f2d6b94faa804f40a4f7a81267": {
          "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
          }
        },
        "dc2819bfe328440d9070d83ca4bef2a4": {
          "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": ""
          }
        },
        "d79f1444a5604a30af28ede9eafd3112": {
          "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
          }
        },
        "0dba833f154f434fa2b4f0a776812fae": {
          "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": ""
          }
        },
        "998bb51324834c31ac572e2a5edeaf02": {
          "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_eb8bc1f6971544488cd1feec51d12cc1",
              "IPY_MODEL_8bca2b6b8cfa4b728c6e586f3729539a",
              "IPY_MODEL_c2160deb54bb487e8c9c4df97d2813d4"
            ],
            "layout": "IPY_MODEL_2f1a2cf9a4d04a3a81876c61c8ff171d"
          }
        },
        "eb8bc1f6971544488cd1feec51d12cc1": {
          "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_e80d9f06c6c142bbba65797a362f0f83",
            "placeholder": "​",
            "style": "IPY_MODEL_83140c381a354c0cbd5d4a4c64eeeba8",
            "value": "100%"
          }
        },
        "8bca2b6b8cfa4b728c6e586f3729539a": {
          "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_63d265ab09fc43ed93572b7db59c6de3",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_6d77cb46deaa48af9750a9e256c3a421",
            "value": 2
          }
        },
        "c2160deb54bb487e8c9c4df97d2813d4": {
          "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_f1df8bca445b4dc3ab496a5b78e00433",
            "placeholder": "​",
            "style": "IPY_MODEL_595e46691226435eb42a35e4ac12f0f7",
            "value": " 2/2 [00:08&lt;00:00,  4.04s/ba]"
          }
        },
        "2f1a2cf9a4d04a3a81876c61c8ff171d": {
          "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
          }
        },
        "e80d9f06c6c142bbba65797a362f0f83": {
          "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
          }
        },
        "83140c381a354c0cbd5d4a4c64eeeba8": {
          "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": ""
          }
        },
        "63d265ab09fc43ed93572b7db59c6de3": {
          "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
          }
        },
        "6d77cb46deaa48af9750a9e256c3a421": {
          "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": ""
          }
        },
        "f1df8bca445b4dc3ab496a5b78e00433": {
          "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
          }
        },
        "595e46691226435eb42a35e4ac12f0f7": {
          "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": ""
          }
        },
        "bc94e058c0be47288c6ac4ffe0416d9a": {
          "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_793608d96e35419093d4348849aa8f7c",
              "IPY_MODEL_97e1af7370af4167a59209990d00332c",
              "IPY_MODEL_5bb869ae9e1247d9bd79ac91204672ee"
            ],
            "layout": "IPY_MODEL_3d64166130f94b5281446258b126f3d8"
          }
        },
        "793608d96e35419093d4348849aa8f7c": {
          "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_501b353fc1c646f895f3bc79beb9f67b",
            "placeholder": "​",
            "style": "IPY_MODEL_507bde3809014c0b991185bdee05a628",
            "value": "100%"
          }
        },
        "97e1af7370af4167a59209990d00332c": {
          "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_1184300e84894e7b99dae5349ba8dac1",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_09dfe5e8a98d43a28666ec76dd13b003",
            "value": 2
          }
        },
        "5bb869ae9e1247d9bd79ac91204672ee": {
          "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_ea8d996dc51f4b5993e89a8e7b432f21",
            "placeholder": "​",
            "style": "IPY_MODEL_a759fbf5703e4cd3969f850f9d5dd5a9",
            "value": " 2/2 [00:07&lt;00:00,  3.88s/ba]"
          }
        },
        "3d64166130f94b5281446258b126f3d8": {
          "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
          }
        },
        "501b353fc1c646f895f3bc79beb9f67b": {
          "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
          }
        },
        "507bde3809014c0b991185bdee05a628": {
          "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": ""
          }
        },
        "1184300e84894e7b99dae5349ba8dac1": {
          "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
          }
        },
        "09dfe5e8a98d43a28666ec76dd13b003": {
          "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": ""
          }
        },
        "ea8d996dc51f4b5993e89a8e7b432f21": {
          "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
          }
        },
        "a759fbf5703e4cd3969f850f9d5dd5a9": {
          "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": ""
          }
        },
        "1dbcb3cfe78a411192b1a242b9883dd4": {
          "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_b7ae5329e38546d08e12e9b9c81b955e",
              "IPY_MODEL_10abfc876f264171b146a9621222ca54",
              "IPY_MODEL_918f3d3b2a044dbcb40c0018f16c9ddc"
            ],
            "layout": "IPY_MODEL_b888be705c5a4ad2b6338e13fe4bc7d9"
          }
        },
        "b7ae5329e38546d08e12e9b9c81b955e": {
          "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_f80989b6864c4edba92a5d0821ce9d40",
            "placeholder": "​",
            "style": "IPY_MODEL_619575b535ef41eb88755d8c05820d0a",
            "value": "100%"
          }
        },
        "10abfc876f264171b146a9621222ca54": {
          "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_2328bc41f9b14feebafa2d6013739e86",
            "max": 125,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_78f9e1cd44444132b5637ed26499fa32",
            "value": 125
          }
        },
        "918f3d3b2a044dbcb40c0018f16c9ddc": {
          "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_be49668999404c1abddeaf76ca666571",
            "placeholder": "​",
            "style": "IPY_MODEL_97df30f0c95e4434939a5d1d60f22dd5",
            "value": " 125/125 [00:09&lt;00:00, 13.48ba/s]"
          }
        },
        "b888be705c5a4ad2b6338e13fe4bc7d9": {
          "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
          }
        },
        "f80989b6864c4edba92a5d0821ce9d40": {
          "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
          }
        },
        "619575b535ef41eb88755d8c05820d0a": {
          "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": ""
          }
        },
        "2328bc41f9b14feebafa2d6013739e86": {
          "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
          }
        },
        "78f9e1cd44444132b5637ed26499fa32": {
          "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": ""
          }
        },
        "be49668999404c1abddeaf76ca666571": {
          "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
          }
        },
        "97df30f0c95e4434939a5d1d60f22dd5": {
          "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",
      "source": [
        "**Initialization**\n",
        "- I use these three lines of code on top of my each notebooks because it will help to prevent any problems while reloading the same project. And the third line of code helps to make visualization within the notebook."
      ],
      "metadata": {
        "id": "XEWelcWHRyd6"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 74,
      "metadata": {
        "id": "vNMpzVoPRKR6"
      },
      "outputs": [],
      "source": [
        "#@ INITIALIZATION: \n",
        "%reload_ext autoreload\n",
        "%autoreload 2\n",
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Downloading Libraries and Dependencies**\n",
        "- I have downloaded all the libraries and dependencies required for the project in one particular cell."
      ],
      "metadata": {
        "id": "ykii4j7pR6x-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ IMPORTING MODULES: UNCOMMENT BELOW:\n",
        "# !pip install transformers[sentencepiece]\n",
        "# !pip install datasets\n",
        "# !pip install umap-learn\n",
        "import torch\n",
        "import torch.nn.functional as F\n",
        "from transformers import pipeline\n",
        "from transformers import AutoTokenizer\n",
        "from transformers import AutoModel\n",
        "from transformers import DistilBertTokenizer\n",
        "from transformers import AutoModelForSequenceClassification\n",
        "from transformers import Trainer, TrainingArguments\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from umap import UMAP\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.dummy import DummyClassifier\n",
        "from sklearn.metrics import ConfusionMatrixDisplay, confusion_matrix\n",
        "from sklearn.metrics import accuracy_score, f1_score\n",
        "from datasets import list_datasets\n",
        "from datasets import load_dataset\n",
        "import matplotlib.pyplot as plt\n",
        "from huggingface_hub import notebook_login\n",
        "\n",
        "#@ IGNORING WARNINGS: \n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "metadata": {
        "id": "VNcLSEa0R4R-"
      },
      "execution_count": 122,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**The Dataset**"
      ],
      "metadata": {
        "id": "l83kYK3cSyrc"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING THE DATASET:\n",
        "all_datasets = list_datasets()                                  # Initializing datasets. \n",
        "print(f\"There are {len(all_datasets)} available datasets.\")     # Initializing number of datasets.\n",
        "print(f\"The first 10 datasets are: {all_datasets[:10]}\")        # Inspecting datasets. "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CYjz33WgU07s",
        "outputId": "652adc1b-0010-442a-8ca1-499c2144c453"
      },
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "There are 4216 available datasets.\n",
            "The first 10 datasets are: ['acronym_identification', 'ade_corpus_v2', 'adversarial_qa', 'aeslc', 'afrikaans_ner_corpus', 'ag_news', 'ai2_arc', 'air_dialogue', 'ajgt_twitter_ar', 'allegro_reviews']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ LOADING THE DATASET:\n",
        "emotions = load_dataset(\"emotion\")              # Loading emotion dataset.\n",
        "emotions                                        # Inspecting dataset."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 326,
          "referenced_widgets": [
            "a1a64d006b764dcc866249a7a7eafdbb",
            "891a1507b33540c7a8202493122457f0",
            "e63e5c22cdb4476ab6a45dc878ca18f0",
            "4e2db58a98674de686337955ee5e319f",
            "fd139f0ec816452289a6aafb43138d35",
            "370347f1f3864d1fb7f88c05e08cc195",
            "e4147b31a0404959aef48349243360e5",
            "6a78f5367e4f40b78805498586d40970",
            "c705825c91af4c17947682e22620c18a",
            "70c5283af6b347f58686215e83fb2857",
            "3e7c6e3fd9e14944930beeb9585f1cbd"
          ]
        },
        "id": "8Rs8xwvVXbSU",
        "outputId": "0a557823-b1d7-4937-f1aa-d1cfbf34d9af"
      },
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Using custom data configuration default\n",
            "Reusing dataset emotion (/root/.cache/huggingface/datasets/emotion/default/0.0.0/348f63ca8e27b3713b6c04d723efe6d824a56fb3d1449794716c0f0296072705)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/3 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "a1a64d006b764dcc866249a7a7eafdbb"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DatasetDict({\n",
              "    train: Dataset({\n",
              "        features: ['text', 'label'],\n",
              "        num_rows: 16000\n",
              "    })\n",
              "    validation: Dataset({\n",
              "        features: ['text', 'label'],\n",
              "        num_rows: 2000\n",
              "    })\n",
              "    test: Dataset({\n",
              "        features: ['text', 'label'],\n",
              "        num_rows: 2000\n",
              "    })\n",
              "})"
            ]
          },
          "metadata": {},
          "execution_count": 77
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING DATASET:\n",
        "train_ds = emotions[\"train\"]                    # Initializing training dataset.\n",
        "print(train_ds)                                 # Inspecting training dataset.\n",
        "print(len(train_ds))                            # Inspecting length.\n",
        "print(train_ds[0])                              # Inspecting a dataset.\n",
        "print(train_ds.column_names)                    # Inspecting keys of dataset."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yk8ZivA7cBHB",
        "outputId": "a8cdf56b-9aab-4336-9867-7491074d5a61"
      },
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dataset({\n",
            "    features: ['text', 'label'],\n",
            "    num_rows: 16000\n",
            "})\n",
            "16000\n",
            "{'text': 'i didnt feel humiliated', 'label': 0}\n",
            "['text', 'label']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING THE DATASET:\n",
        "print(train_ds.features)                       # Inspecting data types.\n",
        "print(train_ds[:5])                            # Inspecting dataset.\n",
        "print(train_ds[\"text\"][:5])                    # Inspecting text column. "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "naAGo_9EdjCb",
        "outputId": "cf0e5cbc-9df5-4a66-b353-d33666f65d05"
      },
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'text': Value(dtype='string', id=None), 'label': ClassLabel(num_classes=6, names=['sadness', 'joy', 'love', 'anger', 'fear', 'surprise'], id=None)}\n",
            "{'text': ['i didnt feel humiliated', 'i can go from feeling so hopeless to so damned hopeful just from being around someone who cares and is awake', 'im grabbing a minute to post i feel greedy wrong', 'i am ever feeling nostalgic about the fireplace i will know that it is still on the property', 'i am feeling grouchy'], 'label': [0, 0, 3, 2, 3]}\n",
            "['i didnt feel humiliated', 'i can go from feeling so hopeless to so damned hopeful just from being around someone who cares and is awake', 'im grabbing a minute to post i feel greedy wrong', 'i am ever feeling nostalgic about the fireplace i will know that it is still on the property', 'i am feeling grouchy']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Datasets to DataFrames**"
      ],
      "metadata": {
        "id": "_j8PDwLxhj1f"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING PANDAS DATAFRAME:\n",
        "emotions.set_format(type=\"pandas\")             # Changing data format. \n",
        "df = emotions[\"train\"][:]                      # Initializing dataframe.\n",
        "df.head()                                      # Inspecting dataframe."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "0TA4dj7Ohotn",
        "outputId": "9eb8d340-4122-4db3-a836-08dd3bd4e22c"
      },
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                text  label\n",
              "0                            i didnt feel humiliated      0\n",
              "1  i can go from feeling so hopeless to so damned...      0\n",
              "2   im grabbing a minute to post i feel greedy wrong      3\n",
              "3  i am ever feeling nostalgic about the fireplac...      2\n",
              "4                               i am feeling grouchy      3"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e91a4828-49f5-4675-a400-1fd96604c933\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>i didnt feel humiliated</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>i can go from feeling so hopeless to so damned...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>im grabbing a minute to post i feel greedy wrong</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>i am ever feeling nostalgic about the fireplac...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i am feeling grouchy</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e91a4828-49f5-4675-a400-1fd96604c933')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e91a4828-49f5-4675-a400-1fd96604c933 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e91a4828-49f5-4675-a400-1fd96604c933');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING LABEL NAME:\n",
        "def label_int2str(row):                                         # Defining function. \n",
        "    return emotions[\"train\"].features[\"label\"].int2str(row)     # Converting integers into string.\n",
        "df[\"label_name\"] = df[\"label\"].apply(label_int2str)             # Updating dataframe.\n",
        "df.head()                                                       # Inspecting dataframe."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "t8YyDQ3SkUZe",
        "outputId": "84c02bc2-4ced-4668-a3fc-ab74a95bc6ac"
      },
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                text  label label_name\n",
              "0                            i didnt feel humiliated      0    sadness\n",
              "1  i can go from feeling so hopeless to so damned...      0    sadness\n",
              "2   im grabbing a minute to post i feel greedy wrong      3      anger\n",
              "3  i am ever feeling nostalgic about the fireplac...      2       love\n",
              "4                               i am feeling grouchy      3      anger"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f745597c-e9cc-44d0-9838-d6cebd8c510f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "      <th>label_name</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>i didnt feel humiliated</td>\n",
              "      <td>0</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>i can go from feeling so hopeless to so damned...</td>\n",
              "      <td>0</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>im grabbing a minute to post i feel greedy wrong</td>\n",
              "      <td>3</td>\n",
              "      <td>anger</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>i am ever feeling nostalgic about the fireplac...</td>\n",
              "      <td>2</td>\n",
              "      <td>love</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i am feeling grouchy</td>\n",
              "      <td>3</td>\n",
              "      <td>anger</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f745597c-e9cc-44d0-9838-d6cebd8c510f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f745597c-e9cc-44d0-9838-d6cebd8c510f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-f745597c-e9cc-44d0-9838-d6cebd8c510f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Looking Class Distributions**"
      ],
      "metadata": {
        "id": "kcoBx2MbndNq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING CLASS DISTRIBUTIONS:\n",
        "df[\"label_name\"].value_counts(ascending=True).plot.barh()       # Creating a bar graph. \n",
        "plt.title(\"Frequency of Classes\")                               # Initializing title.\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "-ON6EupxniKp",
        "outputId": "19092cfe-aff4-457b-db0e-130a7ac62186"
      },
      "execution_count": 82,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Note:**\n",
        "- We can see that the dataset is heavily imbalanced: the joy and sadness classes appear frequently, whereas love and surprise are rare. Ways to deal with imbalanced data:\n",
        "    - Randomly oversample the minority class.\n",
        "    - Randomly undersample the majority class. \n",
        "    - Gather more labeled data from the underrepresented classes. "
      ],
      "metadata": {
        "id": "XpTB1tFdo97E"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Tweets Length**"
      ],
      "metadata": {
        "id": "eArx0gC1vHVk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING TWEETS LENGTH:\n",
        "df[\"Words Per Tweet\"] = df[\"text\"].str.split().apply(len)       # Inspecting input sequence length.\n",
        "df.boxplot(\"Words Per Tweet\", by=\"label_name\", grid=False,\n",
        "           showfliers=False, color=\"black\")                     # Creating a box plot.\n",
        "plt.suptitle(\"\")\n",
        "plt.xlabel(\"\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        },
        "id": "N_E3VdQrrslk",
        "outputId": "16a45b75-cfcc-4d7d-d07e-6f011caa4914"
      },
      "execution_count": 83,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ RESETING DATA FORMAT:\n",
        "emotions.reset_format()"
      ],
      "metadata": {
        "id": "Y96PUqxpxlMk"
      },
      "execution_count": 84,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Text to Tokens**\n",
        "- **Tokenization** is the step of breaking down a string into the atomic units used in the model. **Numericalization** is the process of converting tokens into integer. "
      ],
      "metadata": {
        "id": "yurjtVC-Sgfk"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Character Tokenization**"
      ],
      "metadata": {
        "id": "UMAyhQESZPf_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ IMPLEMENTATION OF CHARACTER TOKENIZATION: \n",
        "text = \"Tokenization is the step of breaking down.\"         # Initializing text example. \n",
        "tokenized_text = list(text)                                 # Character tokenization. \n",
        "print(tokenized_text)                                       # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tsEWPRywSeu-",
        "outputId": "c8f4972a-12a2-469d-e10e-07cc44262474"
      },
      "execution_count": 85,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['T', 'o', 'k', 'e', 'n', 'i', 'z', 'a', 't', 'i', 'o', 'n', ' ', 'i', 's', ' ', 't', 'h', 'e', ' ', 's', 't', 'e', 'p', ' ', 'o', 'f', ' ', 'b', 'r', 'e', 'a', 'k', 'i', 'n', 'g', ' ', 'd', 'o', 'w', 'n', '.']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING NUMERICALIZATION:\n",
        "token2idx = {ch: idx for idx, ch in enumerate(sorted(set(tokenized_text)))}     # Initializing numericalization.\n",
        "print(token2idx)                                                                # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vvh8FQK3UuqC",
        "outputId": "5a692b5d-77bd-4ad4-965a-4e5532f7d46f"
      },
      "execution_count": 86,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{' ': 0, '.': 1, 'T': 2, 'a': 3, 'b': 4, 'd': 5, 'e': 6, 'f': 7, 'g': 8, 'h': 9, 'i': 10, 'k': 11, 'n': 12, 'o': 13, 'p': 14, 'r': 15, 's': 16, 't': 17, 'w': 18, 'z': 19}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING NUMERICALIZATION:\n",
        "input_ids = [token2idx[token] for token in tokenized_text]                      # Numericalization.\n",
        "print(input_ids)                                                                # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ji3onX6FVQYs",
        "outputId": "bf1d8f75-41fa-4a0e-834c-0da351f64a63"
      },
      "execution_count": 87,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[2, 13, 11, 6, 12, 10, 19, 3, 17, 10, 13, 12, 0, 10, 16, 0, 17, 9, 6, 0, 16, 17, 6, 14, 0, 13, 7, 0, 4, 15, 6, 3, 11, 10, 12, 8, 0, 5, 13, 18, 12, 1]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING ONE HOT ENCODING:\n",
        "categorical_df = pd.DataFrame({\"Name\": [\"Bumblebee\", \"Optimus Prime\", \"Megatron\"], \n",
        "                               \"Label ID\": [0, 1, 2]})                              # Example dataframe.\n",
        "pd.get_dummies(categorical_df[\"Name\"])                                              # One-hot encoding."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "_OMFzMCvV0zw",
        "outputId": "a4180e73-66ee-42a3-b562-27db28c32b24"
      },
      "execution_count": 88,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Bumblebee  Megatron  Optimus Prime\n",
              "0          1         0              0\n",
              "1          0         0              1\n",
              "2          0         1              0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-00442813-f264-4f38-84c5-48fd42c8bdf9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Bumblebee</th>\n",
              "      <th>Megatron</th>\n",
              "      <th>Optimus Prime</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-00442813-f264-4f38-84c5-48fd42c8bdf9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-00442813-f264-4f38-84c5-48fd42c8bdf9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-00442813-f264-4f38-84c5-48fd42c8bdf9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 88
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ ONE HOT ENCODING WITH PYTORCH:\n",
        "input_ids = torch.tensor(input_ids)                                     # Converting into tensors.\n",
        "one_hot_encodings = F.one_hot(input_ids, num_classes=len(token2idx))    # One-hot encoding.\n",
        "one_hot_encodings.shape                                                 # Inspecting shape."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Dzk-M8VyXiEv",
        "outputId": "8ed30a9a-c102-4204-d868-92d0007f3825"
      },
      "execution_count": 89,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([42, 20])"
            ]
          },
          "metadata": {},
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING CHARACTER LEVEL TOKENIZATION:\n",
        "print(f\"Token: {tokenized_text[0]}\")                                    # Inspection.\n",
        "print(f\"Tensor index: {input_ids[0]}\")                                  # Inspection.\n",
        "print(f\"One-hot: {one_hot_encodings[0]}\")                               # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LwtejWySYaxX",
        "outputId": "8fd8d1fc-dd8a-4164-aba7-42419ed40ca1"
      },
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Token: T\n",
            "Tensor index: 2\n",
            "One-hot: tensor([0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Word Tokenization**"
      ],
      "metadata": {
        "id": "gVUCoP5m-fyO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING WORD TOKENIZATION:\n",
        "tokenized_text = text.split()                               # Initializing word tokens.\n",
        "print(tokenized_text)                                       # Inspecting tokens."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "noa8Q-Pd-kux",
        "outputId": "0d81d8fc-94a0-4dfe-833e-84b8837777ad"
      },
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['Tokenization', 'is', 'the', 'step', 'of', 'breaking', 'down.']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Subword Tokenization**\n",
        "- Subword tokenization combines the best aspects of character and word tokenization. It splits rare words into smaller units to allow the model to deal with complex words and misspellings and it keeps frequent words as unique entities so that the length of inputs are kept to a manageable size. "
      ],
      "metadata": {
        "id": "cAZFRx-oAp85"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING SUBWORD TOKENIZATION:\n",
        "model_ckpt = \"distilbert-base-uncased\"                                      # Initializing model checkpoint.\n",
        "tokenizer = AutoTokenizer.from_pretrained(model_ckpt)                       # Initializing pretrained tokenizer. \n",
        "distilbert_tokenizer = DistilBertTokenizer.from_pretrained(model_ckpt)      # Initializing pretrained tokenizer. "
      ],
      "metadata": {
        "id": "LmusjUA9_23Y"
      },
      "execution_count": 92,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@ IMPLEMENTATION OF BERT TOKENIZER:\n",
        "encoded_text = tokenizer(text)                                            # Initializing tokenization. \n",
        "print(encoded_text)                                                       # Inspection.\n",
        "tokens = tokenizer.convert_ids_to_tokens(encoded_text.input_ids)          # Generating tokens.\n",
        "print(tokens)                                                             # Inspecting tokens.\n",
        "print(tokenizer.convert_tokens_to_string(tokens))                         # Converting into strings."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dJE_cTp5bxu2",
        "outputId": "ccad8117-3475-4ca0-d056-f045089cc053"
      },
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'input_ids': [101, 19204, 3989, 2003, 1996, 3357, 1997, 4911, 2091, 1012, 102], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}\n",
            "['[CLS]', 'token', '##ization', 'is', 'the', 'step', 'of', 'breaking', 'down', '.', '[SEP]']\n",
            "[CLS] tokenization is the step of breaking down. [SEP]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING FEATURES OF TOKENIZER:\n",
        "print(tokenizer.vocab_size)                                               # Inspecting vocabulary size. \n",
        "print(tokenizer.model_max_length)                                         # Inspecting context size. \n",
        "print(tokenizer.model_input_names)                                        # Inspecting keys."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kOUpA9PsdQJn",
        "outputId": "ee06017e-43f4-4cd4-bd4c-5da7752fbfd1"
      },
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "30522\n",
            "512\n",
            "['input_ids', 'attention_mask']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Tokenizing Whole Dataset**"
      ],
      "metadata": {
        "id": "cq7nChtLeCxQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ TOKENIZING WHOLE DATASET:\n",
        "def tokenize(batch):                                                    # Defining processing function.\n",
        "    return tokenizer(batch[\"text\"], padding=True, truncation=True)      # Initializing tokenization.\n",
        "print(tokenize(emotions[\"train\"][:2]))                                  # Implementation of function."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7VYBFvm5eCLW",
        "outputId": "627d644a-7641-496a-cd54-327f2f1e5f57"
      },
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'input_ids': [[101, 1045, 2134, 2102, 2514, 26608, 102, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [101, 1045, 2064, 2175, 2013, 3110, 2061, 20625, 2000, 2061, 9636, 17772, 2074, 2013, 2108, 2105, 2619, 2040, 14977, 1998, 2003, 8300, 102]], 'attention_mask': [[1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ TOKENIZING WHOLE DATASET:\n",
        "emotions_encoded = emotions.map(tokenize, batched=True, batch_size=None)    # Encoding whole dataset.\n",
        "print(emotions_encoded[\"train\"].column_names)                               # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aHGm9jqugdgN",
        "outputId": "f42b86e7-3781-4ca9-a282-bbbef13e587f"
      },
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Loading cached processed dataset at /root/.cache/huggingface/datasets/emotion/default/0.0.0/348f63ca8e27b3713b6c04d723efe6d824a56fb3d1449794716c0f0296072705/cache-3a9c55ebbd683fa4.arrow\n",
            "Loading cached processed dataset at /root/.cache/huggingface/datasets/emotion/default/0.0.0/348f63ca8e27b3713b6c04d723efe6d824a56fb3d1449794716c0f0296072705/cache-55738f0937c6f330.arrow\n",
            "Loading cached processed dataset at /root/.cache/huggingface/datasets/emotion/default/0.0.0/348f63ca8e27b3713b6c04d723efe6d824a56fb3d1449794716c0f0296072705/cache-a3e8ac34c2a58a50.arrow\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['text', 'label', 'input_ids', 'attention_mask']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Transformers as Feature Extractors**"
      ],
      "metadata": {
        "id": "dFTvlINZih8U"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ TRANSFORMERS AS FEATURE EXTRACTORS:\n",
        "model_ckpt = \"distilbert-base-uncased\"                                      # Initializing model checkpoint.\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")       # Initializing gpu. \n",
        "model = AutoModel.from_pretrained(model_ckpt).to(device)                    # Initializing pretrained model. "
      ],
      "metadata": {
        "id": "EB6az1cbin70"
      },
      "execution_count": 97,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING TOKENIZATION: EXAMPLE:\n",
        "text = \"I am from Nepal.\"                                       # Initialization.\n",
        "inputs = tokenizer(text, return_tensors=\"pt\")                   # Getting encoded tensors.\n",
        "print(f\"Input tensor shape: {inputs['input_ids'].size()}\")      # Inspecting shape."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SIPrPeLokrmb",
        "outputId": "0db13abb-89ac-4332-8d6c-88abe8c2c30e"
      },
      "execution_count": 98,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Input tensor shape: torch.Size([1, 7])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ IMPLEMENTATION OF MODEL: EXAMPLE:\n",
        "inputs = {k:v.to(device) for k,v in inputs.items()}             # Initializing gpu tensors.\n",
        "with torch.no_grad():\n",
        "    outputs = model(**inputs)                                   # Implementation of model.\n",
        "print(outputs)                                                  # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HWSEDgoXkrlE",
        "outputId": "ad118575-df8f-4977-cd1c-182332a6dc3d"
      },
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "BaseModelOutput(last_hidden_state=tensor([[[ 0.1270, -0.1420, -0.0982,  ..., -0.0035,  0.3750,  0.2726],\n",
            "         [ 0.5891, -0.4465, -0.4257,  ..., -0.1788,  0.4324,  0.1609],\n",
            "         [ 0.4902, -0.0613,  0.2096,  ..., -0.3614,  0.0871,  0.1906],\n",
            "         ...,\n",
            "         [ 1.0310, -0.1348, -0.4088,  ..., -0.3798, -0.2369, -0.2591],\n",
            "         [ 0.0633, -0.7405, -0.5038,  ...,  0.2587,  0.1197, -0.6394],\n",
            "         [ 0.7514,  0.0696, -0.3815,  ...,  0.1729, -0.3607, -0.5465]]],\n",
            "       device='cuda:0'), hidden_states=None, attentions=None)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING MODEL OUTPUTS: \n",
        "print(outputs.last_hidden_state.size())                         # Inspection.\n",
        "print(outputs.last_hidden_state[:, 0].size())                   # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dr2po8LEngcT",
        "outputId": "7a7802dd-d84f-472e-c144-faef87c87b6c"
      },
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([1, 7, 768])\n",
            "torch.Size([1, 768])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ EXTRACTING HIDDEN STATES FOR WHOLE DATASET:\n",
        "def extract_hidden_states(batch):                                   # Defining function. \n",
        "    inputs = {k:v.to(device) for k,v in batch.items() \n",
        "              if k in tokenizer.model_input_names}                  # Placing into gpu.\n",
        "    with torch.no_grad():\n",
        "        last_hidden_state = model(**inputs).last_hidden_state       # Getting last hidden states.\n",
        "    return {\"hidden_state\": last_hidden_state[:, 0].cpu().numpy()}  # Return vector for [CLS] token. "
      ],
      "metadata": {
        "id": "-3y5hnm7pmMz"
      },
      "execution_count": 101,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@ EXTRACTING HIDDEN STATES FOR WHOLE DATASET:\n",
        "emotions_encoded.set_format(\"torch\", columns=[\"input_ids\", \"attention_mask\", \"label\"])  # Initializing tensors format.\n",
        "emotions_hidden = emotions_encoded.map(extract_hidden_states, batched=True)             # Extracting hidden states.\n",
        "emotions_hidden[\"train\"].column_names                                                   # Inspection. "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 130,
          "referenced_widgets": [
            "83b56c2fbc07484180f401eae13ac0c5",
            "d8434f53d2c648488d512b2d4e0436be",
            "ed47be1f977d4e0b98671fb01c066c19",
            "a61484b2a74b4275812e6511ca5e20b5",
            "e6872a4539b34f529c182fe53039ba34",
            "e089684da2b64e8faf6e52132edfd806",
            "40edd0a0a8f24c338fc580d516c112ba",
            "1cc170f2d6b94faa804f40a4f7a81267",
            "dc2819bfe328440d9070d83ca4bef2a4",
            "d79f1444a5604a30af28ede9eafd3112",
            "0dba833f154f434fa2b4f0a776812fae",
            "998bb51324834c31ac572e2a5edeaf02",
            "eb8bc1f6971544488cd1feec51d12cc1",
            "8bca2b6b8cfa4b728c6e586f3729539a",
            "c2160deb54bb487e8c9c4df97d2813d4",
            "2f1a2cf9a4d04a3a81876c61c8ff171d",
            "e80d9f06c6c142bbba65797a362f0f83",
            "83140c381a354c0cbd5d4a4c64eeeba8",
            "63d265ab09fc43ed93572b7db59c6de3",
            "6d77cb46deaa48af9750a9e256c3a421",
            "f1df8bca445b4dc3ab496a5b78e00433",
            "595e46691226435eb42a35e4ac12f0f7",
            "bc94e058c0be47288c6ac4ffe0416d9a",
            "793608d96e35419093d4348849aa8f7c",
            "97e1af7370af4167a59209990d00332c",
            "5bb869ae9e1247d9bd79ac91204672ee",
            "3d64166130f94b5281446258b126f3d8",
            "501b353fc1c646f895f3bc79beb9f67b",
            "507bde3809014c0b991185bdee05a628",
            "1184300e84894e7b99dae5349ba8dac1",
            "09dfe5e8a98d43a28666ec76dd13b003",
            "ea8d996dc51f4b5993e89a8e7b432f21",
            "a759fbf5703e4cd3969f850f9d5dd5a9"
          ]
        },
        "id": "_mbs-JRureZJ",
        "outputId": "d7a36299-1c76-43ca-fe3e-7b6ddbcf4e05"
      },
      "execution_count": 102,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/16 [00:00<?, ?ba/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "83b56c2fbc07484180f401eae13ac0c5"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/2 [00:00<?, ?ba/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "998bb51324834c31ac572e2a5edeaf02"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/2 [00:00<?, ?ba/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "bc94e058c0be47288c6ac4ffe0416d9a"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['text', 'label', 'input_ids', 'attention_mask', 'hidden_state']"
            ]
          },
          "metadata": {},
          "execution_count": 102
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Feature Matrix**\n",
        "- We will use hidden states as input features and the labels as targets. "
      ],
      "metadata": {
        "id": "-xSy5akjtMAx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ CREATING FEATURE MATRIX: \n",
        "X_train = np.array(emotions_hidden[\"train\"][\"hidden_state\"])            # Initializing training arrays.\n",
        "X_valid = np.array(emotions_hidden[\"validation\"][\"hidden_state\"])       # Initializing validation arrays.\n",
        "y_train = np.array(emotions_hidden[\"train\"][\"label\"])                   # Initializing training arrays.\n",
        "y_valid = np.array(emotions_hidden[\"validation\"][\"label\"])              # Initializing validation arrays.\n",
        "X_train.shape, X_valid.shape                                            # Inspecting shape."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-cHo546ctLg5",
        "outputId": "5cccdefd-ed05-47aa-c855-a7791afc13d8"
      },
      "execution_count": 103,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((16000, 768), (2000, 768))"
            ]
          },
          "metadata": {},
          "execution_count": 103
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Visualizing Training Set**"
      ],
      "metadata": {
        "id": "E3uJjNHYvso_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ VISUALIZING TRAINING SET: PREPROCESSING:\n",
        "X_scaled = MinMaxScaler().fit_transform(X_train)                        # Scaling training arrays. \n",
        "mapper = UMAP(n_components=2, metric=\"cosine\").fit(X_scaled)            # Fitting UMAP. \n",
        "df_emb = pd.DataFrame(mapper.embedding_, columns=[\"X\", \"Y\"])            # Creating a dataframe.\n",
        "df_emb[\"label\"] = y_train                                               # Initialization.\n",
        "df_emb.head()                                                           # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "7eeZJeQvvr2P",
        "outputId": "ac1fa4f6-80cc-4f40-d617-b2eba6576cdc"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          X         Y  label\n",
              "0  4.223744  6.418426      0\n",
              "1 -3.088841  5.653121      0\n",
              "2  5.288105  2.886452      3\n",
              "3 -2.685671  2.942268      2\n",
              "4 -3.189245  3.688916      3"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-aba9fca4-3139-462e-bb93-f9d9c40a4ee0\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>X</th>\n",
              "      <th>Y</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4.223744</td>\n",
              "      <td>6.418426</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-3.088841</td>\n",
              "      <td>5.653121</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5.288105</td>\n",
              "      <td>2.886452</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-2.685671</td>\n",
              "      <td>2.942268</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-3.189245</td>\n",
              "      <td>3.688916</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-aba9fca4-3139-462e-bb93-f9d9c40a4ee0')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-aba9fca4-3139-462e-bb93-f9d9c40a4ee0 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-aba9fca4-3139-462e-bb93-f9d9c40a4ee0');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ VISUALIZING TRAINING SET: \n",
        "fig, axes = plt.subplots(2, 3, figsize=(7, 5))\n",
        "axes = axes.flatten()\n",
        "cmaps = [\"Greys\", \"Blues\", \"Oranges\", \"Reds\", \"Purples\", \"Greens\"]\n",
        "labels = emotions[\"train\"].features[\"label\"].names\n",
        "for i, (label, cmap) in enumerate(zip(labels, cmaps)):\n",
        "    df_emb_sub = df_emb.query(f\"label == {i}\")\n",
        "    axes[i].hexbin(df_emb_sub[\"X\"], df_emb_sub[\"Y\"], cmap=cmap, gridsize=20, linewidths=(0,))\n",
        "    axes[i].set_title(label)\n",
        "    axes[i].set_xticks([]) \n",
        "    axes[i].set_yticks([])\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        },
        "id": "6sL72mw8yc91",
        "outputId": "bdcd95f5-8275-459e-b40b-80dad3fb557a"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 504x360 with 6 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Training Classifier**"
      ],
      "metadata": {
        "id": "nQULbNUbCFn0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ TRAINING LOGISTIC REGRESSION: \n",
        "lr_clf = LogisticRegression(max_iter=3000)                      # Initializing logistic regression model. \n",
        "lr_clf.fit(X_train, y_train)                                    # Training the model. \n",
        "lr_clf.score(X_valid, y_valid)                                  # Inspecting accuracy of model."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BErb2fiRCFRF",
        "outputId": "db51392f-60b3-45a6-ecab-308526f5dfe2"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.633"
            ]
          },
          "metadata": {},
          "execution_count": 106
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ TRAINING DUMMY CLASSIFIER:\n",
        "dummy_clf = DummyClassifier(strategy=\"most_frequent\")           # Initializing dummy classifier model.\n",
        "dummy_clf.fit(X_train, y_train)                                 # Training the model. \n",
        "dummy_clf.score(X_valid, y_valid)                               # Inspecting accuracy of model."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oq1QqTDQDgZ-",
        "outputId": "b3ec70f8-d190-4d47-bac4-ab32c308a90a"
      },
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.352"
            ]
          },
          "metadata": {},
          "execution_count": 107
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTING CONFUSION MATRIX:\n",
        "def plot_confusion_matrix(y_preds, y_true, labels):                             # Defining function.\n",
        "    cm = confusion_matrix(y_true, y_preds, normalize=\"true\")                    # Initializing confusion matrix.\n",
        "    fig, ax = plt.subplots(figsize=(6, 6)) \n",
        "    disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=labels)\n",
        "    disp.plot(cmap=\"Blues\", values_format=\".2f\", ax=ax, colorbar=False)\n",
        "    plt.title(\"Normalized confusion matrix\")\n",
        "    plt.show()\n",
        "\n",
        "#@ INSPECTING CONFUSION MATRIX:\n",
        "y_preds = lr_clf.predict(X_valid)                                               # Getting predictions. \n",
        "plot_confusion_matrix(y_preds, y_valid, labels)                                 # Plotting confusion matrix."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "Lj5Cy-xWEv2o",
        "outputId": "467ae717-dfa1-4b9b-f9eb-4ecfe6555c8c"
      },
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Fine-Tuning Transfomers**"
      ],
      "metadata": {
        "id": "XZMAvmfXHauY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "**DistilBERT**\n",
        "- The main advantage of this model is that it achieves comparable performance to BERT, while being significantly smaller and more efficient."
      ],
      "metadata": {
        "id": "B5Ahs7c_R_tO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING PRETRAINED DISTILBERT MODEL:\n",
        "num_labels = 6                                                  # Initialization.\n",
        "model = (AutoModelForSequenceClassification.from_pretrained(\n",
        "    model_ckpt, num_labels=num_labels).to(device))              # Initializing pretrained model."
      ],
      "metadata": {
        "id": "NJGECWscSZwh"
      },
      "execution_count": 109,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Performance Metrics**"
      ],
      "metadata": {
        "id": "3Fb4mbz1JQmj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ DEFINING FUNCTION FOR PERFORMANCE METRICS:\n",
        "def compute_metrics(pred):                              # Defining function. \n",
        "    labels = pred.label_ids                             # Initializing attributes. \n",
        "    preds = pred.predictions.argmax(-1)                 # Initializing attributes.\n",
        "    f1 = f1_score(labels, preds, average=\"weighted\")    # Computing f1 score.\n",
        "    acc = accuracy_score(labels, preds)                 # Computing accuracy score.\n",
        "    return {\"accuracy\": acc, \"f1\": f1}"
      ],
      "metadata": {
        "id": "A6t-lJNIJDjk"
      },
      "execution_count": 110,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Training Model**"
      ],
      "metadata": {
        "id": "l95SS7K7LwaG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING TRAINING ARGUMENTS:\n",
        "batch_size = 64                                                             # Initializing batch size.\n",
        "logging_steps = len(emotions_encoded[\"train\"]) // batch_size                # Initialization.\n",
        "model_name = f\"{model_ckpt}-finetuned-emotion\"                              # Initializing model name.\n",
        "training_args = TrainingArguments(output_dir=model_name, \n",
        "                                  num_train_epochs=2,\n",
        "                                  learning_rate=2e-5,\n",
        "                                  per_device_train_batch_size=batch_size,\n",
        "                                  per_device_eval_batch_size=batch_size,\n",
        "                                  weight_decay=0.01,\n",
        "                                  evaluation_strategy=\"epoch\",\n",
        "                                  disable_tqdm=False,\n",
        "                                  logging_steps=logging_steps,\n",
        "                                  push_to_hub=False,\n",
        "                                  log_level=\"error\")                        # Initializing training arguments."
      ],
      "metadata": {
        "id": "1pb7o1MPMBNv"
      },
      "execution_count": 111,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@ TRAINING MODEL:\n",
        "trainer = Trainer(model=model, args=training_args,\n",
        "                  compute_metrics=compute_metrics,\n",
        "                  train_dataset=emotions_encoded[\"train\"],\n",
        "                  eval_dataset=emotions_encoded[\"validation\"],\n",
        "                  tokenizer=tokenizer)                                      # Initializing trainer.\n",
        "trainer.train();                                                            # Training the model."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 138
        },
        "id": "n2vQKVJSOhAv",
        "outputId": "e49a4e28-7e41-4874-c0f0-61eefa645426"
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='500' max='500' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [500/500 07:31, Epoch 2/2]\n",
              "    </div>\n",
              "    <table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              " <tr style=\"text-align: left;\">\n",
              "      <th>Epoch</th>\n",
              "      <th>Training Loss</th>\n",
              "      <th>Validation Loss</th>\n",
              "      <th>Accuracy</th>\n",
              "      <th>F1</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>1</td>\n",
              "      <td>0.829600</td>\n",
              "      <td>0.336119</td>\n",
              "      <td>0.893000</td>\n",
              "      <td>0.888759</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2</td>\n",
              "      <td>0.253100</td>\n",
              "      <td>0.229689</td>\n",
              "      <td>0.920000</td>\n",
              "      <td>0.920102</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><p>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INITIALIZING MODEL EVALUATION: \n",
        "preds_output = trainer.predict(emotions_encoded[\"validation\"])                  # Generating predictions.\n",
        "print(preds_output.metrics)                                                     # Inspecting metrics."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "9sECrZZBPNi4",
        "outputId": "02028442-97b3-42cd-af18-11564b3210cc"
      },
      "execution_count": 113,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='32' max='32' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [32/32 00:08]\n",
              "    </div>\n",
              "    "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'test_loss': 0.22968870401382446, 'test_accuracy': 0.92, 'test_f1': 0.9201015148683785, 'test_runtime': 8.7718, 'test_samples_per_second': 228.004, 'test_steps_per_second': 3.648}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ PLOTTING CONFUSION MATRIX:\n",
        "y_preds = np.argmax(preds_output.predictions, axis=1)                           # Generating predictions.\n",
        "plot_confusion_matrix(y_preds, y_valid, labels)                                 # Plotting confusion matrix."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "k71H-WDyS3Tk",
        "outputId": "9bde1154-4256-4ffd-8e13-ad3e07bb2f58"
      },
      "execution_count": 114,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Error Analysis**"
      ],
      "metadata": {
        "id": "N0qm7qLrVkMO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@ DEFINING FUNCTION FOR ERROR ANALYSIS:\n",
        "def forward_pass_with_label(batch):                             # Defining function. \n",
        "    inputs = {k:v.to(device) for k,v in batch.items()\n",
        "              if k in tokenizer.model_input_names}              # Placing into gpu.\n",
        "    \n",
        "    with torch.no_grad():\n",
        "        output = model(**inputs)                                # Getting output predictions.\n",
        "        pred_label = torch.argmax(output.logits, axis=-1)       # Getting predictions label.\n",
        "        loss = F.cross_entropy(output.logits, \n",
        "                               batch[\"label\"].to(device),\n",
        "                               reduction=\"none\")                # Calculating loss. \n",
        "        return {\"loss\": loss.cpu().numpy(),\n",
        "                \"predicted_label\": pred_label.cpu().numpy()}    # Placing outputs on cpu."
      ],
      "metadata": {
        "id": "bEuH6ofBTznc"
      },
      "execution_count": 115,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#@ COMPUTING LOSS:\n",
        "emotions_encoded[\"validation\"] = emotions_encoded[\"validation\"].map(\n",
        "    forward_pass_with_label, batched=True, batch_size=16\n",
        ")                                                                               # Implementation.\n",
        "emotions_encoded.set_format(\"pandas\")                                           # Changing format.\n",
        "cols = [\"text\", \"label\", \"predicted_label\", \"loss\"]                             # Initialization. \n",
        "df_test = emotions_encoded[\"validation\"][:][cols]                               # Initializing dataframe.\n",
        "df_test[\"label\"] = df_test[\"label\"].apply(label_int2str)                        # Converting integers into strings.\n",
        "df_test[\"predicted_label\"] = df_test[\"predicted_label\"].apply(label_int2str)    # Converting integers into strings.\n",
        "df_test.sort_values(\"loss\", ascending=False).head()                             # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238,
          "referenced_widgets": [
            "1dbcb3cfe78a411192b1a242b9883dd4",
            "b7ae5329e38546d08e12e9b9c81b955e",
            "10abfc876f264171b146a9621222ca54",
            "918f3d3b2a044dbcb40c0018f16c9ddc",
            "b888be705c5a4ad2b6338e13fe4bc7d9",
            "f80989b6864c4edba92a5d0821ce9d40",
            "619575b535ef41eb88755d8c05820d0a",
            "2328bc41f9b14feebafa2d6013739e86",
            "78f9e1cd44444132b5637ed26499fa32",
            "be49668999404c1abddeaf76ca666571",
            "97df30f0c95e4434939a5d1d60f22dd5"
          ]
        },
        "id": "3sUuq78eXax7",
        "outputId": "887a6885-054a-4c76-dcb2-051e87a28ba1"
      },
      "execution_count": 116,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "  0%|          | 0/125 [00:00<?, ?ba/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "1dbcb3cfe78a411192b1a242b9883dd4"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                   text  label  \\\n",
              "882   i feel badly about reneging on my commitment t...   love   \n",
              "1801  i feel that he was being overshadowed by the s...   love   \n",
              "1963  i called myself pro life and voted for perry w...    joy   \n",
              "1274  i am going to several holiday parties and i ca...    joy   \n",
              "1195  i always think about are act the way i want to...  anger   \n",
              "\n",
              "     predicted_label      loss  \n",
              "882          sadness  5.773220  \n",
              "1801         sadness  5.688723  \n",
              "1963         sadness  5.446440  \n",
              "1274         sadness  5.381667  \n",
              "1195             joy  5.324480  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-19c703eb-f23d-4958-8240-78ec5c17536d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "      <th>predicted_label</th>\n",
              "      <th>loss</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>882</th>\n",
              "      <td>i feel badly about reneging on my commitment t...</td>\n",
              "      <td>love</td>\n",
              "      <td>sadness</td>\n",
              "      <td>5.773220</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1801</th>\n",
              "      <td>i feel that he was being overshadowed by the s...</td>\n",
              "      <td>love</td>\n",
              "      <td>sadness</td>\n",
              "      <td>5.688723</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1963</th>\n",
              "      <td>i called myself pro life and voted for perry w...</td>\n",
              "      <td>joy</td>\n",
              "      <td>sadness</td>\n",
              "      <td>5.446440</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1274</th>\n",
              "      <td>i am going to several holiday parties and i ca...</td>\n",
              "      <td>joy</td>\n",
              "      <td>sadness</td>\n",
              "      <td>5.381667</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1195</th>\n",
              "      <td>i always think about are act the way i want to...</td>\n",
              "      <td>anger</td>\n",
              "      <td>joy</td>\n",
              "      <td>5.324480</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-19c703eb-f23d-4958-8240-78ec5c17536d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-19c703eb-f23d-4958-8240-78ec5c17536d button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-19c703eb-f23d-4958-8240-78ec5c17536d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 116
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#@ INSPECTION:\n",
        "df_test.sort_values(\"loss\", ascending=True).head()                             # Inspection."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "70tzyvbtZsec",
        "outputId": "07b9756a-01f8-47aa-8863-ab141f4596a5"
      },
      "execution_count": 118,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                   text    label  \\\n",
              "1502             i feel ungrateful for stupid shit like  sadness   \n",
              "21    i feel try to tell me im ungrateful tell me im...  sadness   \n",
              "1466  i feel so ungrateful to be wishing this pregna...  sadness   \n",
              "264       i am left feeling underwhelmed and ungrateful  sadness   \n",
              "133   i and feel quite ungrateful for it but i m loo...  sadness   \n",
              "\n",
              "     predicted_label      loss  \n",
              "1502         sadness  0.015914  \n",
              "21           sadness  0.015969  \n",
              "1466         sadness  0.016134  \n",
              "264          sadness  0.016180  \n",
              "133          sadness  0.016256  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2f3df1af-40cc-4464-891f-06e5f293c655\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "      <th>predicted_label</th>\n",
              "      <th>loss</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1502</th>\n",
              "      <td>i feel ungrateful for stupid shit like</td>\n",
              "      <td>sadness</td>\n",
              "      <td>sadness</td>\n",
              "      <td>0.015914</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>i feel try to tell me im ungrateful tell me im...</td>\n",
              "      <td>sadness</td>\n",
              "      <td>sadness</td>\n",
              "      <td>0.015969</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1466</th>\n",
              "      <td>i feel so ungrateful to be wishing this pregna...</td>\n",
              "      <td>sadness</td>\n",
              "      <td>sadness</td>\n",
              "      <td>0.016134</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>264</th>\n",
              "      <td>i am left feeling underwhelmed and ungrateful</td>\n",
              "      <td>sadness</td>\n",
              "      <td>sadness</td>\n",
              "      <td>0.016180</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>133</th>\n",
              "      <td>i and feel quite ungrateful for it but i m loo...</td>\n",
              "      <td>sadness</td>\n",
              "      <td>sadness</td>\n",
              "      <td>0.016256</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2f3df1af-40cc-4464-891f-06e5f293c655')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-2f3df1af-40cc-4464-891f-06e5f293c655 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-2f3df1af-40cc-4464-891f-06e5f293c655');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 118
        }
      ]
    }
  ]
}