BERT_For_Question_Answering_Model_1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "BERT For Question Answering - Model 1.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "9400445bf5d24430a805d7bd32975a85": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_c96bcb048825405aac28c174767fc6d8",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_7fbd3a7baf3f46718268920875082dcd",
              "IPY_MODEL_8be630cbc0764841b92bf307e5a45c12"
            ]
          }
        },
        "c96bcb048825405aac28c174767fc6d8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "7fbd3a7baf3f46718268920875082dcd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_397af1c3bb664903894b1b48e146c305",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 231508,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 231508,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_cf4da207aaee419e8b34675655f0d9c9"
          }
        },
        "8be630cbc0764841b92bf307e5a45c12": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_f706be524ef845549a3158e2f7bb970c",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 232k/232k [00:20&lt;00:00, 11.0kB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ec7015b072ef4bca92c40fe6c4854658"
          }
        },
        "397af1c3bb664903894b1b48e146c305": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "cf4da207aaee419e8b34675655f0d9c9": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "f706be524ef845549a3158e2f7bb970c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "ec7015b072ef4bca92c40fe6c4854658": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "223397afa77a46c49ca6bde8c42e7970": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_812c722dadc44bf182fde60192dcfaac",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_b19b064e5f144096b9a6a1924366fb04",
              "IPY_MODEL_a7ea4d64bf604a9f84cb68626f1efdd9"
            ]
          }
        },
        "812c722dadc44bf182fde60192dcfaac": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "b19b064e5f144096b9a6a1924366fb04": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_37e65b2506f24ee8896b963a425b80c1",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 433,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 433,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_73904fc515dc4f3b9486895272715bcc"
          }
        },
        "a7ea4d64bf604a9f84cb68626f1efdd9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_2d12fb9626614a03994ec7090d55e790",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 433/433 [00:20&lt;00:00, 21.0B/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0b907db52a9141d9991a5aabb91ec603"
          }
        },
        "37e65b2506f24ee8896b963a425b80c1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "73904fc515dc4f3b9486895272715bcc": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "2d12fb9626614a03994ec7090d55e790": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "0b907db52a9141d9991a5aabb91ec603": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "d3b11a0e26734180838d45ce1c405b88": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_133a1c5776984360b4122cdfd946df46",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_1e10299a57664a93ba5b16eddc39baca",
              "IPY_MODEL_db6cdfa4e7f24f82a58831db432dab21"
            ]
          }
        },
        "133a1c5776984360b4122cdfd946df46": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "1e10299a57664a93ba5b16eddc39baca": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_e954e273b11a449b88654de698902661",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 440473133,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 440473133,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ad54ea8720be42e482f1679904b37756"
          }
        },
        "db6cdfa4e7f24f82a58831db432dab21": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_627c6619c7504f35a5ad5611b9d326d9",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 440M/440M [00:18&lt;00:00, 24.4MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_d9d0bdd4fecd4fe1aafa5a5c9a1b1c2c"
          }
        },
        "e954e273b11a449b88654de698902661": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "ad54ea8720be42e482f1679904b37756": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "627c6619c7504f35a5ad5611b9d326d9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "d9d0bdd4fecd4fe1aafa5a5c9a1b1c2c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "a3s-IwAXoN1m",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 615
        },
        "outputId": "b751bdba-f7cf-4303-9bac-f17d7e7a8089"
      },
      "source": [
        "!pip install transformers==2.11.0"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting transformers==2.11.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/48/35/ad2c5b1b8f99feaaf9d7cdadaeef261f098c6e1a6a2935d4d07662a6b780/transformers-2.11.0-py3-none-any.whl (674kB)\n",
            "\u001b[K     |████████████████████████████████| 675kB 2.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: filelock in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (3.0.12)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (2.23.0)\n",
            "Collecting sacremoses\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/7d/34/09d19aff26edcc8eb2a01bed8e98f13a1537005d31e95233fd48216eed10/sacremoses-0.0.43.tar.gz (883kB)\n",
            "\u001b[K     |████████████████████████████████| 890kB 13.2MB/s \n",
            "\u001b[?25hRequirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (2019.12.20)\n",
            "Collecting sentencepiece\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/e5/2d/6d4ca4bef9a67070fa1cac508606328329152b1df10bdf31fb6e4e727894/sentencepiece-0.1.94-cp36-cp36m-manylinux2014_x86_64.whl (1.1MB)\n",
            "\u001b[K     |████████████████████████████████| 1.1MB 18.2MB/s \n",
            "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (4.41.1)\n",
            "Requirement already satisfied: dataclasses; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (0.7)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (20.4)\n",
            "Collecting tokenizers==0.7.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/14/e5/a26eb4716523808bb0a799fcfdceb6ebf77a18169d9591b2f46a9adb87d9/tokenizers-0.7.0-cp36-cp36m-manylinux1_x86_64.whl (3.8MB)\n",
            "\u001b[K     |████████████████████████████████| 3.8MB 24.8MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from transformers==2.11.0) (1.18.5)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->transformers==2.11.0) (2.10)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->transformers==2.11.0) (3.0.4)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->transformers==2.11.0) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->transformers==2.11.0) (2020.6.20)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers==2.11.0) (1.15.0)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers==2.11.0) (7.1.2)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers==2.11.0) (0.16.0)\n",
            "Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->transformers==2.11.0) (2.4.7)\n",
            "Building wheels for collected packages: sacremoses\n",
            "  Building wheel for sacremoses (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for sacremoses: filename=sacremoses-0.0.43-cp36-none-any.whl size=893257 sha256=f18fa4f316742934d53f04e74b4965fc9eca2125c4f5c250a9409e0198233378\n",
            "  Stored in directory: /root/.cache/pip/wheels/29/3c/fd/7ce5c3f0666dab31a50123635e6fb5e19ceb42ce38d4e58f45\n",
            "Successfully built sacremoses\n",
            "Installing collected packages: sacremoses, sentencepiece, tokenizers, transformers\n",
            "Successfully installed sacremoses-0.0.43 sentencepiece-0.1.94 tokenizers-0.7.0 transformers-2.11.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L_XUGqrDoWUa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "7d4ecf75-cecb-4e3f-f493-80bc5e530ed5"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "%cd '/content/drive/My Drive/Practical Topics in Machine Learning/Project'"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n",
            "/content/drive/My Drive/Practical Topics in Machine Learning/Project\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KGAO4SMuULoY"
      },
      "source": [
        "import json\n",
        "import time\n",
        "import torch\n",
        "import datetime\n",
        "import numpy as np\n",
        "from torch import nn\n",
        "import torch.nn.functional as F\n",
        "from torch.utils.data import DataLoader\n",
        "from transformers import BertTokenizerFast, BertModel, BertPreTrainedModel\n",
        "from transformers import AdamW\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7ojxBv1yULs_"
      },
      "source": [
        "MODEL_FILE = \"./model1_File\"\n",
        "\n",
        "# Set the seed value all over the place to make this reproducible.\n",
        "SEED = 42\n",
        "\n",
        "np.random.seed(SEED)\n",
        "torch.manual_seed(SEED)\n",
        "torch.cuda.manual_seed(SEED)\n",
        "\n",
        "device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qz1dIWIsULwU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 66,
          "referenced_widgets": [
            "9400445bf5d24430a805d7bd32975a85",
            "c96bcb048825405aac28c174767fc6d8",
            "7fbd3a7baf3f46718268920875082dcd",
            "8be630cbc0764841b92bf307e5a45c12",
            "397af1c3bb664903894b1b48e146c305",
            "cf4da207aaee419e8b34675655f0d9c9",
            "f706be524ef845549a3158e2f7bb970c",
            "ec7015b072ef4bca92c40fe6c4854658"
          ]
        },
        "outputId": "d4bd8717-afa9-4745-bf59-efb8c758e702"
      },
      "source": [
        "# Load the BERT tokenizer.\n",
        "tokenizer = BertTokenizerFast.from_pretrained('bert-base-uncased')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "9400445bf5d24430a805d7bd32975a85",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=231508.0, style=ProgressStyle(descripti…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eBFTENolBYqS"
      },
      "source": [
        "# Construct BertBaseForQA model - Version 1"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FT5jpoQbULzE"
      },
      "source": [
        "class BertBaseForQA_Version1(nn.Module):\n",
        "\n",
        "    def __init__(self):\n",
        "        super(BertBaseForQuestionAnswering_Version1, self).__init__()\n",
        "        \n",
        "        # Load the pretrained BERT model of 12-layers with an uncased vocab.\n",
        "        self.bert = BertModel.from_pretrained(\"bert-base-uncased\")\n",
        "        self.hidden_size = self.bert.config.hidden_size\n",
        "\n",
        "        # Task specific output layers - one for the start position and the second for the end position of the answer.\n",
        "        self.linear_start = nn.Linear(self.hidden_size, 1)\n",
        "        self.linear_end = nn.Linear(self.hidden_size, 1)\n",
        "\n",
        "    def forward(self, input_ids=None, attention_mask=None, token_type_ids=None):\n",
        "        \n",
        "        bert_output = self.bert(input_ids=input_ids, attention_mask=attention_mask, token_type_ids=token_type_ids)\n",
        "\n",
        "        hidden_states = bert_output[0]\n",
        "\n",
        "        # Pass the hidden states through the linear layers to get the start and end logits.\n",
        "        start_logits = self.linear_start(hidden_states).squeeze(-1)\n",
        "        end_logits = self.linear_end(hidden_states).squeeze(-1) \n",
        "\n",
        "        # Producing predictive distribution for the start and end positions.\n",
        "        start_probs = F.log_softmax(start_logits, dim=1)\n",
        "        end_probs = F.log_softmax(end_logits, dim=1)\n",
        "\n",
        "        return start_probs, end_probs"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vxdqXVCpUL1u",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 115,
          "referenced_widgets": [
            "223397afa77a46c49ca6bde8c42e7970",
            "812c722dadc44bf182fde60192dcfaac",
            "b19b064e5f144096b9a6a1924366fb04",
            "a7ea4d64bf604a9f84cb68626f1efdd9",
            "37e65b2506f24ee8896b963a425b80c1",
            "73904fc515dc4f3b9486895272715bcc",
            "2d12fb9626614a03994ec7090d55e790",
            "0b907db52a9141d9991a5aabb91ec603",
            "d3b11a0e26734180838d45ce1c405b88",
            "133a1c5776984360b4122cdfd946df46",
            "1e10299a57664a93ba5b16eddc39baca",
            "db6cdfa4e7f24f82a58831db432dab21",
            "e954e273b11a449b88654de698902661",
            "ad54ea8720be42e482f1679904b37756",
            "627c6619c7504f35a5ad5611b9d326d9",
            "d9d0bdd4fecd4fe1aafa5a5c9a1b1c2c"
          ]
        },
        "outputId": "f54af4c1-2bf8-4bf3-8b64-cb4c7742cc64"
      },
      "source": [
        "model = BertBaseForQA_Version1()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "223397afa77a46c49ca6bde8c42e7970",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=433.0, style=ProgressStyle(description_…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "d3b11a0e26734180838d45ce1c405b88",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=440473133.0, style=ProgressStyle(descri…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4Lq0dStso5yn"
      },
      "source": [
        "# Preprocessing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X2hogQVFUL4K"
      },
      "source": [
        "# Takes a time in seconds and returns a string hh:mm:ss.\n",
        "def format_time(elapsed):\n",
        "\n",
        "    # Round to the nearest second.\n",
        "    elapsed_rounded = int(round((elapsed)))\n",
        "    \n",
        "    # Format as hh:mm:ss\n",
        "    return str(datetime.timedelta(seconds=elapsed_rounded))\n",
        "\n",
        "# Measure how long the preprocessing takes.\n",
        "t0 = time.time()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XS7pqYBYUL6t"
      },
      "source": [
        "def read_squad2_file(file_name):\n",
        "\n",
        "    # Load the training data.\n",
        "    with open(file_name, 'r') as file:\n",
        "        data = json.load(file)\n",
        "\n",
        "    # Create question, context and answer lists.\n",
        "    questions, contexts, answers = [], [], []\n",
        "\n",
        "    for topic in data['data']:\n",
        "\n",
        "        # For every context paragraph.\n",
        "        for paragraph in topic['paragraphs']:\n",
        "            \n",
        "            context = paragraph['context']\n",
        "\n",
        "            # For every question.\n",
        "            for qa in paragraph['qas']:\n",
        "                \n",
        "                question = qa['question']\n",
        "                answer = qa['answers'][0] if qa['answers'] else {}\n",
        "\n",
        "                # Store the question, the context paragraph and the answer.\n",
        "                questions.append(question)\n",
        "                contexts.append(context)\n",
        "                answers.append(answer)\n",
        "\n",
        "    return questions, contexts, answers\n",
        "\n",
        "train_questions, train_contexts, train_answers = read_squad2_file('train-v2.0.json')\n",
        "dev_questions, dev_contexts, dev_answers = read_squad2_file('dev-v2.0.json')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lA46xhLRUMDF"
      },
      "source": [
        "def compute_answer_end_index(contexts, answers):\n",
        "    \n",
        "    for context, answer in zip(contexts, answers):\n",
        "\n",
        "        if not answer:   # If the question is unanswerable.\n",
        "            answer['answer_start'] = 0\n",
        "            answer['answer_end'] = 0\n",
        "            continue\n",
        "\n",
        "        gold_text = answer['text'] \n",
        "        start_index = answer['answer_start']\n",
        "\n",
        "        # Compute the end index of the answer.\n",
        "        end_index = start_index + len(gold_text)\n",
        "\n",
        "        if context[start_index:end_index] == gold_text:\n",
        "            answer['answer_end'] = end_index\n",
        "        else:\n",
        "            raise ValueError\n",
        "\n",
        "compute_answer_end_index(train_contexts, train_answers)\n",
        "compute_answer_end_index(dev_contexts, dev_answers)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f7P7aN9wZnOq"
      },
      "source": [
        "# Pack every question-context in the training set.\n",
        "packed_train = [(context, question) for context, question in zip(train_contexts, train_questions)]\n",
        "\n",
        "# Tokenize all train examples.\n",
        "train_encodings = tokenizer.batch_encode_plus(packed_train, max_length=tokenizer.model_max_length,\n",
        "                                              pad_to_max_length=True, return_token_type_ids=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CuL5BLRfUL-S"
      },
      "source": [
        "# Pack every question-context in the dev set.\n",
        "packed_dev = [(context, question) for context, question in zip(dev_contexts, dev_questions)]\n",
        "\n",
        "# Tokenize all dev examples.\n",
        "dev_encodings = tokenizer.batch_encode_plus(packed_dev, max_length=tokenizer.model_max_length,\n",
        "                                            pad_to_max_length=True, return_token_type_ids=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "amKsjrBWUsQD"
      },
      "source": [
        "def add_token_positions(encodings, answers):\n",
        "    \n",
        "    start_positions = []\n",
        "    end_positions = []\n",
        "    \n",
        "    for i in range(len(answers)):\n",
        "\n",
        "        # If the question is unanswerable, the ground truth is the [CLS] token which is in index 0. \n",
        "        if answers[i]['answer_start'] ==0 and answers[i]['answer_end']==0:\n",
        "            start_positions.append(0)\n",
        "            end_positions.append(0)\n",
        "        \n",
        "        else: # According to the start/end index, find the position of the suitable token.\n",
        "            start_positions.append(encodings.char_to_token(i, answers[i]['answer_start']))\n",
        "            end_positions.append(\n",
        "                encodings.char_to_token(i, answers[i]['answer_end'] - 1))        \n",
        "        \n",
        "        # If None, the answer passage has been truncated.\n",
        "        if start_positions[-1] is None:\n",
        "            start_positions[-1] = tokenizer.model_max_length\n",
        "            \n",
        "        if end_positions[-1] is None:\n",
        "            end_positions[-1] = tokenizer.model_max_length\n",
        "    \n",
        "    encodings.update({'start_positions': start_positions, 'end_positions': end_positions})\n",
        "\n",
        "add_token_positions(train_encodings, train_answers)\n",
        "add_token_positions(dev_encodings, dev_answers)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CaxwVLb9UsVE"
      },
      "source": [
        "class Squad2Dataset(torch.utils.data.Dataset):\n",
        "    \n",
        "    def __init__(self, encodings):\n",
        "        self.encodings = encodings\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        return {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}\n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.encodings.input_ids)\n",
        "\n",
        "train_dataset = Squad2Dataset(train_encodings)\n",
        "dev_dataset = Squad2Dataset(dev_encodings)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0IHQEQo1Usdk"
      },
      "source": [
        "# Create the DataLoaders for the training and dev sets.\n",
        "train_loader = DataLoader(train_dataset, batch_size=10, shuffle=True)\n",
        "dev_loader = DataLoader(dev_dataset, batch_size=10, shuffle=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CiOMMnFnUsaA",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "9a45d136-e6e0-4f06-aae1-b6723098b887"
      },
      "source": [
        "# Report how long the preprocessing took.\n",
        "print(\"The preprocessing took {} (hh:mm:ss)\".format(format_time(time.time()-t0)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The preprocessing took 0:00:59 (hh:mm:ss)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P7jAWRxitC-p"
      },
      "source": [
        "# Evaluation Methods"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5VTH6b6C-sjJ"
      },
      "source": [
        "def get_questions_ids(file_name):\n",
        "    \n",
        "    # Load the dev data.\n",
        "    with open(file_name, 'r') as file:\n",
        "        data = json.load(file)\n",
        "\n",
        "    ids = []  # Create questions ids list.\n",
        "    \n",
        "    for topic in data['data']:\n",
        "\n",
        "        # For every context paragraph.\n",
        "        for paragraph in topic['paragraphs']:\n",
        "\n",
        "            # Store every question's id in the list.         \n",
        "            for qa in paragraph['qas']:\n",
        "                ids.append(qa['id'])\n",
        "\n",
        "    return ids\n",
        "\n",
        "ids = get_questions_ids('dev-v2.0.json')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Cb9zWFamU3l4"
      },
      "source": [
        "def evaluate(loader, ignored_index):\n",
        "    \n",
        "    # Will store the model predictions.\n",
        "    answers = []\n",
        "\n",
        "    # Reset the total dev loss.\n",
        "    total_dev_loss = 0.0\n",
        "\n",
        "    with torch.no_grad():\n",
        "\n",
        "        for batch in loader:\n",
        "\n",
        "            # Unpack the batch from the dataloader and copy each tensor to the GPU.   \n",
        "            input_ids = batch['input_ids'].to(device)\n",
        "            attention_mask = batch['attention_mask'].to(device)\n",
        "            token_type_ids = batch['token_type_ids'].to(device)\n",
        "            start_positions = batch['start_positions'].to(device)\n",
        "            end_positions = batch['end_positions'].to(device)\n",
        "\n",
        "            # Forward pass.\n",
        "            start_out, end_out = model(input_ids=input_ids, attention_mask=attention_mask,\n",
        "                                       token_type_ids=token_type_ids)\n",
        "\n",
        "            # Compute the Negative Log Liklihood loss.\n",
        "            start_loss = F.nll_loss(start_out, start_positions, ignore_index=ignored_index)\n",
        "            end_loss = F.nll_loss(end_out, end_positions, ignore_index=ignored_index)\n",
        "\n",
        "            # The total loss of the current batch is the average of the losses.\n",
        "            total_loss = (start_loss + end_loss) / 2\n",
        "\n",
        "            # Accumulate the dev loss.   \n",
        "            total_dev_loss += total_loss.item()\n",
        "\n",
        "            # Move to CPU.\n",
        "            input_ids = input_ids.detach().cpu()\n",
        "            start_out = start_out.detach().cpu()\n",
        "            end_out = end_out.detach().cpu()\n",
        "\n",
        "            # Predict the start and end positions of the answer.\n",
        "            start_predictions = torch.argmax(start_out, dim=1)\n",
        "            end_predictions = torch.argmax(end_out, dim=1)\n",
        "\n",
        "            cls_token_id = [tokenizer.cls_token_id]\n",
        "\n",
        "            # For each input id and the start and end predicted positions.\n",
        "            for current_input_ids, start_pred, end_pred in zip(input_ids.tolist(), start_predictions.tolist(),\n",
        "                                                              end_predictions.tolist()):\n",
        "\n",
        "                answer_tokens = current_input_ids[start_pred: end_pred + 1]\n",
        "\n",
        "                # If the model predicted the question is answerable with an answer span and it contains the [CLS] token \n",
        "                # within it, remove it from the answer span.\n",
        "                if cls_token_id != answer_tokens and cls_token_id[0] in answer_tokens:\n",
        "                    answer_tokens = answer_tokens[1:]\n",
        "\n",
        "                # Store predictions.\n",
        "                answers.append('' if cls_token_id == answer_tokens else tokenizer.convert_tokens_to_string(answer_tokens))\n",
        "\n",
        "    # Return the average dev loss over all of the batches and the model's predictions.\n",
        "    return total_dev_loss / len(loader), answers"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bw-gYp0q9qNF"
      },
      "source": [
        "def produce_predictions_file(questions_ids, predictions, epoch):\n",
        "    \n",
        "    preds = {key: val for key, val in zip(questions_ids, predictions)}\n",
        "\n",
        "    if epoch == 1:\n",
        "        with open('./preds_model1_epoch_1.json', 'w') as outfile:\n",
        "            json.dump(preds, outfile)\n",
        "\n",
        "        !python3 evaluate-v2.0.py ./dev-v2.0.json ./preds_model1_epoch_1.json\n",
        "    else:\n",
        "        if epoch == 2:\n",
        "            with open('./preds_model1_epoch_2.json', 'w') as outfile:\n",
        "                json.dump(preds, outfile)\n",
        "\n",
        "            !python3 evaluate-v2.0.py ./dev-v2.0.json ./preds_model1_epoch_2.json\n",
        "        else:\n",
        "            if epoch == 3:\n",
        "                with open('./preds_model1_epoch_3.json', 'w') as outfile:\n",
        "                    json.dump(preds, outfile)\n",
        "\n",
        "                !python3 evaluate-v2.0.py ./dev-v2.0.json ./preds_model1_epoch_3.json\n",
        "            else:\n",
        "                with open('./preds_model1_epoch_0.json', 'w') as outfile:\n",
        "                    json.dump(preds, outfile)\n",
        "\n",
        "                !python3 evaluate-v2.0.py ./dev-v2.0.json ./preds_model1_epoch_0.json"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "frcrYI8FrHul"
      },
      "source": [
        "# Training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YpOf9FJxVpYb"
      },
      "source": [
        "if torch.cuda.is_available():\n",
        "    model.cuda()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1WvPg_XdVCvv"
      },
      "source": [
        "# Number of training epochs.\n",
        "EPOCHS = 3\n",
        "\n",
        "# Initialize the optimizer.\n",
        "optimizer = AdamW(model.parameters(), lr=2e-5)\n",
        "\n",
        "# Ignore this index when calculating the loss.\n",
        "ignored_index =tokenizer.model_max_length"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Jqiea6woVCzX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "aa5c5fe0-8c3d-42f4-963a-d3bca055ae06"
      },
      "source": [
        "def train(train_loader, dev_loader, optimizer, epochs, ignored_index):\n",
        "\n",
        "    train_epochs_loss, dev_epochs_loss = [], []\n",
        "    train_steps_loss = []\n",
        "    \n",
        "    lr = optimizer.param_groups[0]['lr']\n",
        "\n",
        "    # Measure the total training time for the whole run.\n",
        "    total_t0 = time.time()\n",
        "          \n",
        "    print(\"\")\n",
        "    print('============================ Initial Evaluation ===========================')\n",
        "    print(\"\")\n",
        "    print(\" Evaluating...\")\n",
        "    print(\"\")\n",
        "\n",
        "    # Compute the loss and the model's predictions on the dev set, before training.\n",
        "    initial_dev_loss, preds = evaluate(dev_loader, ignored_index)\n",
        "    dev_epochs_loss.append(initial_dev_loss)\n",
        "      \n",
        "    print(\"   Initial Dev Loss: {:.3f}\".format(initial_dev_loss))\n",
        "    print(\"\")\n",
        "\n",
        "    # Report the EM and F1 score on the dev set.\n",
        "    print(\"   Initial Dev Score:\")\n",
        "    produce_predictions_file(ids, preds, 0)\n",
        "    print(\"\")\n",
        "    \n",
        "    # Measure how long the evaluation took.\n",
        "    evaluation_time = format_time(time.time() - total_t0)\n",
        "    print(\"   Initial Evaluation took: {}\".format(evaluation_time))\n",
        "    print(\"\")\n",
        "\n",
        "    # For each epoch...\n",
        "    for epoch in range(0, epochs):\n",
        "        \n",
        "        # ========================================\n",
        "        #               Training\n",
        "        # ========================================\n",
        "        \n",
        "        # Perform one full pass over the training set.\n",
        "\n",
        "        print(\"\")\n",
        "        print('=============================== Epoch {:} / {:} ==============================='.format(epoch + 1, epochs))\n",
        "        print(\"\")\n",
        "        print(' Training...')\n",
        "        print(\"\")\n",
        "\n",
        "        # Measure how long the training epoch takes.\n",
        "        t0 = time.time()\n",
        "\n",
        "        # Declare train mode.        \n",
        "        model.train()\n",
        "\n",
        "        # Reset the total loss for this epoch.\n",
        "        total_train_loss = 0.0\n",
        "\n",
        "        # Reset the loss for the next 1000 bathes.\n",
        "        sum_loss = 0.0\n",
        "\n",
        "        for batch_idx, batch in enumerate(train_loader):\n",
        "\n",
        "            # Zero the gradients from the previous iteration.\n",
        "            optimizer.zero_grad()\n",
        "\n",
        "            # Unpack the training batch from the dataloader and copy each tensor to the GPU.   \n",
        "            input_ids = batch['input_ids'].to(device)\n",
        "            attention_mask = batch['attention_mask'].to(device)\n",
        "            token_type_ids = batch['token_type_ids'].to(device)\n",
        "            start_positions = batch['start_positions'].to(device)\n",
        "            end_positions = batch['end_positions'].to(device)\n",
        "\n",
        "            # Forward pass.\n",
        "            start_out, end_out = model(input_ids=input_ids, attention_mask=attention_mask, token_type_ids=token_type_ids)\n",
        "\n",
        "            # Compute the Negative Log Liklihood loss.\n",
        "            start_loss = F.nll_loss(start_out, start_positions, ignore_index=ignored_index)\n",
        "            end_loss = F.nll_loss(end_out, end_positions ,ignore_index=ignored_index)\n",
        "\n",
        "            # The total loss of the current batch is the average of the losses.\n",
        "            total_loss = (start_loss + end_loss) / 2\n",
        "\n",
        "            # Accumulate the train loss.   \n",
        "            total_train_loss += total_loss.item()\n",
        "            sum_loss += total_loss.item()\n",
        "\n",
        "            # Backward pass to calculate the gradients.\n",
        "            total_loss.backward()\n",
        "\n",
        "            # Update the parameters.\n",
        "            optimizer.step()\n",
        "\n",
        "            # Progress update every 1000 batches.\n",
        "            if (batch_idx + 1) % 1000 == 0:\n",
        "                \n",
        "                # Compute the average loss of the last 1000 batches.\n",
        "                current = sum_loss / 1000\n",
        "                train_steps_loss.append(current)\n",
        "\n",
        "                # Calculate elapsed time in minutes.\n",
        "                elapsed = format_time(time.time() - t0)\n",
        "                \n",
        "                # Report progress.\n",
        "                print(\"   Batch: {:>6,} / {:>5,}...\".format(batch_idx + 1, len(train_loader)),\n",
        "                      \" Elapsed: {}...\".format(elapsed),\n",
        "                      \" Train Loss: {:.3f}...\".format(current))\n",
        "                \n",
        "                # Reset the loss for the next 1000 bathes.\n",
        "                sum_loss = 0.0\n",
        "        \n",
        "        # Save the current model's state.\n",
        "        torch.save({\n",
        "            'epoch': epoch+1,\n",
        "            'model_state_dict': model.state_dict(),\n",
        "            'optimizer_state_dict': optimizer.state_dict(),\n",
        "        }, MODEL_FILE + \"-epoch\" + str(epoch + 1))\n",
        "        \n",
        "        print(\"\")\n",
        "        print(\"   The current model's state has been saved\")\n",
        "\n",
        "        # Calculate the average loss over all of the batches.\n",
        "        avg_train_loss = total_train_loss / len(train_loader)            \n",
        "        train_epochs_loss.append(avg_train_loss)\n",
        "\n",
        "        # Measure how long this training epoch took.\n",
        "        training_time = format_time(time.time() - t0)\n",
        "\n",
        "        print(\"\")\n",
        "        print(\"   Average training loss: {:.3f}\".format(avg_train_loss))\n",
        "        print(\"   Training epcoh took: {}\".format(training_time))\n",
        "\n",
        "\n",
        "        # ========================================\n",
        "        #               Evaluation\n",
        "        # ========================================\n",
        "        \n",
        "        # After the completion of each training epoch, measure the model's performance on\n",
        "        # the dev set.\n",
        "\n",
        "        print(\"\")\n",
        "        print(\"\")\n",
        "        print(\" Evaluating...\")\n",
        "        print(\"\")\n",
        "\n",
        "        # Measure how long the evaluation takes.\n",
        "        t1 = time.time()\n",
        "\n",
        "        # Declare evaluation mode.\n",
        "        model.eval()\n",
        "\n",
        "        # Compute the loss and the model's predictions on the dev set.\n",
        "        avg_dev_loss, preds = evaluate(dev_loader, ignored_index)\n",
        "        dev_epochs_loss.append(avg_dev_loss)\n",
        "        \n",
        "        print(\"   Dev Loss: {:.3f}\".format(avg_dev_loss))\n",
        "        print(\"\")\n",
        "\n",
        "        # Report the EM and F1 score on the dev set.\n",
        "        print(\"   Dev Score:\")\n",
        "        produce_predictions_file(ids, preds, epoch + 1)\n",
        "        print(\"\")\n",
        "\n",
        "        # Report how long the evaluation took.\n",
        "        evaluation_time = format_time(time.time() - t1)\n",
        "        print(\"   Evaluation took: {}\".format(evaluation_time))\n",
        "        \n",
        "        print(\"\")\n",
        "        print(\"\")\n",
        "\n",
        "        # Measure how long the all epoch took - training and evaluating.\n",
        "        epoch_time = format_time(time.time() - t0)\n",
        "        print(\" Overall, Epoch took: {}\".format(epoch_time))\n",
        "\n",
        "        # Decay the learning rate every two epochs.\n",
        "        if (epoch + 1) % 2 == 0:\n",
        "            lr /= 2\n",
        "            for group in optimizer.param_groups:\n",
        "                group['lr'] = lr \n",
        "            print(\"\\n The learning rate updated to {}\".format(group['lr']))\n",
        "\n",
        "    print(\"\")\n",
        "    print(\"Training completed!\")  \n",
        "    print(\"Total training took {} (hh:mm:ss)\".format(format_time(time.time()-total_t0)))\n",
        "\n",
        "    return train_epochs_loss, dev_epochs_loss, train_steps_loss\n",
        "\n",
        "train_epochs_loss, dev_epochs_loss, train_steps_loss = train(train_loader, dev_loader, optimizer, epochs=EPOCHS, ignored_index=ignored_index)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "============================ Initial Evaluation ===========================\n",
            "\n",
            " Evaluating...\n",
            "\n",
            "   Initial Dev Loss: 6.300\n",
            "\n",
            "   Initial Dev Score:\n",
            "   {\n",
            "      \"exact\": 20.567674555714646,\n",
            "      \"f1\": 22.451588793441395,\n",
            "      \"total\": 11873,\n",
            "      \"HasAns_exact\": 0.033738191632928474,\n",
            "      \"HasAns_f1\": 3.8069692551504253,\n",
            "      \"HasAns_total\": 5928,\n",
            "      \"NoAns_exact\": 41.042893187552565,\n",
            "      \"NoAns_f1\": 41.042893187552565,\n",
            "      \"NoAns_total\": 5945\n",
            "   }\n",
            "\n",
            "   Initial Evaluation took: 0:03:51\n",
            "\n",
            "\n",
            "=============================== Epoch 1 / 3 ===============================\n",
            "\n",
            " Training...\n",
            "\n",
            "   Batch:  1,000 / 13,032...  Elapsed: 0:09:22...  Train Loss: 2.275...\n",
            "   Batch:  2,000 / 13,032...  Elapsed: 0:18:44...  Train Loss: 1.565...\n",
            "   Batch:  3,000 / 13,032...  Elapsed: 0:28:06...  Train Loss: 1.427...\n",
            "   Batch:  4,000 / 13,032...  Elapsed: 0:37:28...  Train Loss: 1.323...\n",
            "   Batch:  5,000 / 13,032...  Elapsed: 0:46:49...  Train Loss: 1.257...\n",
            "   Batch:  6,000 / 13,032...  Elapsed: 0:56:11...  Train Loss: 1.215...\n",
            "   Batch:  7,000 / 13,032...  Elapsed: 1:05:33...  Train Loss: 1.174...\n",
            "   Batch:  8,000 / 13,032...  Elapsed: 1:14:55...  Train Loss: 1.120...\n",
            "   Batch:  9,000 / 13,032...  Elapsed: 1:24:16...  Train Loss: 1.100...\n",
            "   Batch: 10,000 / 13,032...  Elapsed: 1:33:38...  Train Loss: 1.080...\n",
            "   Batch: 11,000 / 13,032...  Elapsed: 1:43:00...  Train Loss: 1.060...\n",
            "   Batch: 12,000 / 13,032...  Elapsed: 1:52:22...  Train Loss: 1.055...\n",
            "   Batch: 13,000 / 13,032...  Elapsed: 2:01:44...  Train Loss: 1.010...\n",
            "\n",
            "   The current model's state has been saved\n",
            "\n",
            "   Average training loss: 1.280\n",
            "   Training epcoh took: 2:02:09\n",
            "\n",
            "\n",
            " Evaluating...\n",
            "\n",
            "   Dev Loss: 1.077\n",
            "\n",
            "   Dev Score:\n",
            "   {\n",
            "      \"exact\": 65.7542322917544,\n",
            "      \"f1\": 69.90757910101019,\n",
            "      \"total\": 11873,\n",
            "      \"HasAns_exact\": 55.81983805668016,\n",
            "      \"HasAns_f1\": 64.1384424200897,\n",
            "      \"HasAns_total\": 5928,\n",
            "      \"NoAns_exact\": 75.66021867115222,\n",
            "      \"NoAns_f1\": 75.66021867115222,\n",
            "      \"NoAns_total\": 5945\n",
            "   }\n",
            "\n",
            "   Evaluation took: 0:03:48\n",
            "\n",
            "\n",
            " Overall, Epoch took: 2:05:57\n",
            "\n",
            "=============================== Epoch 2 / 3 ===============================\n",
            "\n",
            " Training...\n",
            "\n",
            "   Batch:  1,000 / 13,032...  Elapsed: 0:09:22...  Train Loss: 0.756...\n",
            "   Batch:  2,000 / 13,032...  Elapsed: 0:18:44...  Train Loss: 0.770...\n",
            "   Batch:  3,000 / 13,032...  Elapsed: 0:28:06...  Train Loss: 0.773...\n",
            "   Batch:  4,000 / 13,032...  Elapsed: 0:37:27...  Train Loss: 0.793...\n",
            "   Batch:  5,000 / 13,032...  Elapsed: 0:46:49...  Train Loss: 0.784...\n",
            "   Batch:  6,000 / 13,032...  Elapsed: 0:56:11...  Train Loss: 0.768...\n",
            "   Batch:  7,000 / 13,032...  Elapsed: 1:05:33...  Train Loss: 0.769...\n",
            "   Batch:  8,000 / 13,032...  Elapsed: 1:14:54...  Train Loss: 0.764...\n",
            "   Batch:  9,000 / 13,032...  Elapsed: 1:24:16...  Train Loss: 0.772...\n",
            "   Batch: 10,000 / 13,032...  Elapsed: 1:33:37...  Train Loss: 0.771...\n",
            "   Batch: 11,000 / 13,032...  Elapsed: 1:42:59...  Train Loss: 0.789...\n",
            "   Batch: 12,000 / 13,032...  Elapsed: 1:52:20...  Train Loss: 0.769...\n",
            "   Batch: 13,000 / 13,032...  Elapsed: 2:01:41...  Train Loss: 0.779...\n",
            "\n",
            "   The current model's state has been saved\n",
            "\n",
            "   Average training loss: 0.774\n",
            "   Training epcoh took: 2:02:07\n",
            "\n",
            "\n",
            " Evaluating...\n",
            "\n",
            "   Dev Loss: 1.230\n",
            "\n",
            "   Dev Score:\n",
            "   {\n",
            "      \"exact\": 66.06586372441674,\n",
            "      \"f1\": 70.48842097396297,\n",
            "      \"total\": 11873,\n",
            "      \"HasAns_exact\": 64.35560053981106,\n",
            "      \"HasAns_f1\": 73.21339781104267,\n",
            "      \"HasAns_total\": 5928,\n",
            "      \"NoAns_exact\": 67.77123633305298,\n",
            "      \"NoAns_f1\": 67.77123633305298,\n",
            "      \"NoAns_total\": 5945\n",
            "   }\n",
            "\n",
            "   Evaluation took: 0:03:49\n",
            "\n",
            "\n",
            " Overall, Epoch took: 2:05:55\n",
            "\n",
            " The learning rate updated to 1e-05\n",
            "\n",
            "=============================== Epoch 3 / 3 ===============================\n",
            "\n",
            " Training...\n",
            "\n",
            "   Batch:  1,000 / 13,032...  Elapsed: 0:09:23...  Train Loss: 0.469...\n",
            "   Batch:  2,000 / 13,032...  Elapsed: 0:18:45...  Train Loss: 0.449...\n",
            "   Batch:  3,000 / 13,032...  Elapsed: 0:28:08...  Train Loss: 0.445...\n",
            "   Batch:  4,000 / 13,032...  Elapsed: 0:37:30...  Train Loss: 0.462...\n",
            "   Batch:  5,000 / 13,032...  Elapsed: 0:46:51...  Train Loss: 0.481...\n",
            "   Batch:  6,000 / 13,032...  Elapsed: 0:56:13...  Train Loss: 0.453...\n",
            "   Batch:  7,000 / 13,032...  Elapsed: 1:05:34...  Train Loss: 0.472...\n",
            "   Batch:  8,000 / 13,032...  Elapsed: 1:14:55...  Train Loss: 0.464...\n",
            "   Batch:  9,000 / 13,032...  Elapsed: 1:24:16...  Train Loss: 0.456...\n",
            "   Batch: 10,000 / 13,032...  Elapsed: 1:33:37...  Train Loss: 0.477...\n",
            "   Batch: 11,000 / 13,032...  Elapsed: 1:42:59...  Train Loss: 0.456...\n",
            "   Batch: 12,000 / 13,032...  Elapsed: 1:52:20...  Train Loss: 0.473...\n",
            "   Batch: 13,000 / 13,032...  Elapsed: 2:01:41...  Train Loss: 0.443...\n",
            "\n",
            "   The current model's state has been saved\n",
            "\n",
            "   Average training loss: 0.462\n",
            "   Training epcoh took: 2:02:06\n",
            "\n",
            "\n",
            " Evaluating...\n",
            "\n",
            "   Dev Loss: 1.161\n",
            "\n",
            "   Dev Score:\n",
            "   {\n",
            "      \"exact\": 69.43485218563126,\n",
            "      \"f1\": 74.01355567480103,\n",
            "      \"total\": 11873,\n",
            "      \"HasAns_exact\": 63.225371120107965,\n",
            "      \"HasAns_f1\": 72.39590865838599,\n",
            "      \"HasAns_total\": 5928,\n",
            "      \"NoAns_exact\": 75.62657695542472,\n",
            "      \"NoAns_f1\": 75.62657695542472,\n",
            "      \"NoAns_total\": 5945\n",
            "   }\n",
            "\n",
            "   Evaluation took: 0:03:48\n",
            "\n",
            "\n",
            " Overall, Epoch took: 2:05:54\n",
            "\n",
            "Training completed!\n",
            "Total training took 6:21:38 (hh:mm:ss)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8lh9s6nXJoME",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "outputId": "c5be5cdd-20ce-4b80-8b3b-5b8325ac99a2"
      },
      "source": [
        "def plot_loss_graph(title):\n",
        "    \n",
        "    fig = plt.figure()\n",
        "    ax = fig.add_subplot(111)\n",
        "    ax.set_title(title)\n",
        "    ax.set_xlabel(\"Epochs\")\n",
        "    ax.set_ylabel(\"Loss\")\n",
        "\n",
        "    ticks = [1, 2, 3]\n",
        "    train_loss = [1.280, 0.774, 0.462]\n",
        "    dev_loss = [1.077, 1.230, 1.161]\n",
        "    plt.xticks(ticks)\n",
        "\n",
        "    plt.plot(ticks, train_loss, label='Train', color='purple')\n",
        "    plt.plot(ticks, dev_loss, label='Dev', color='salmon')\n",
        "\n",
        "    for idx, (a, b) in enumerate(zip(ticks, train_loss)):\n",
        "\n",
        "        if idx == 0:\n",
        "            ax.text(a+0.1, b-0.002, '1.280', fontsize=6, bbox={'facecolor': 'orange', 'alpha': 0.1, 'pad': 1},\n",
        "                    verticalalignment='bottom', horizontalalignment='center')\n",
        "        else:\n",
        "            if idx == 1:\n",
        "                ax.text(a + 0.07, b + 0.018, str(b), fontsize=6, bbox={'facecolor': 'orange', 'alpha': 0.1, 'pad': 1},\n",
        "                        verticalalignment='bottom', horizontalalignment='center')\n",
        "            else:\n",
        "                if idx == 2:\n",
        "                    ax.text(a+0.02, b + 0.03, str(b), fontsize=6,\n",
        "                            bbox={'facecolor': 'orange', 'alpha': 0.1, 'pad': 1},\n",
        "                            verticalalignment='bottom', horizontalalignment='center')\n",
        "\n",
        "    for idx, (a, b) in enumerate(zip(ticks, dev_loss)):\n",
        "\n",
        "        if idx == 0:\n",
        "            ax.text(a, b - 0.043, str(b), fontsize=6, bbox={'facecolor': 'teal', 'alpha': 0.1, 'pad': 1},\n",
        "                    verticalalignment='bottom', horizontalalignment='center')\n",
        "        else:\n",
        "            if idx == 1:\n",
        "                ax.text(a, b + 0.018, '1.230', fontsize=6, bbox={'facecolor': 'teal', 'alpha': 0.1, 'pad': 1},\n",
        "                        verticalalignment='bottom', horizontalalignment='center')\n",
        "            else:\n",
        "                if idx == 2:\n",
        "                    ax.text(a, b - 0.043, str(b), fontsize=6,\n",
        "                            bbox={'facecolor': 'teal', 'alpha': 0.1, 'pad': 1},\n",
        "                            verticalalignment='bottom', horizontalalignment='center')\n",
        "\n",
        "    plt.legend(frameon=False)\n",
        "    plt.savefig(\"Model1 Loss.png\", dpi=192)\n",
        "    fig.set_dpi(100)\n",
        "    plt.show()\n",
        "\n",
        "plot_loss_graph(\"Model 1: Loss over Epochs\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_7y6vvVAMdir",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "outputId": "a9cd274e-f8b6-4dba-89db-e00d041fc3dc"
      },
      "source": [
        "def plot_score_graph(title):\n",
        "    \n",
        "    fig = plt.figure()\n",
        "    ax = fig.add_subplot(111)\n",
        "    ax.set_title(title)\n",
        "    ax.set_xlabel(\"Epochs\")\n",
        "    ax.set_ylabel(\"Loss\")\n",
        "\n",
        "    ticks = [1, 2, 3]\n",
        "    EM = [65.754,66.065,69.434]\n",
        "    F1 = [69.907,70.488, 74.013]\n",
        "    plt.xticks(ticks)\n",
        "\n",
        "    plt.plot(ticks, EM, label='EM', color='teal')\n",
        "    plt.plot(ticks, F1, label='F1', color='orange')\n",
        "\n",
        "    for idx, (a, b) in enumerate(zip(ticks, EM)):\n",
        "\n",
        "        if idx == 0:\n",
        "            ax.text(a, b + 0.2, str(b), fontsize=6, bbox={'facecolor': 'crimson', 'alpha': 0.1, 'pad': 1},\n",
        "                    verticalalignment='bottom', horizontalalignment='center')\n",
        "        else:\n",
        "            if idx == 1:\n",
        "                ax.text(a , b - 0.45, str(b), fontsize=6, bbox={'facecolor': 'crimson', 'alpha': 0.1, 'pad': 1},\n",
        "                        verticalalignment='bottom', horizontalalignment='center')\n",
        "            else:\n",
        "                if idx == 2:\n",
        "                    ax.text(a, b + 0.14, str(b), fontsize=6,\n",
        "                            bbox={'facecolor': 'crimson', 'alpha': 0.1, 'pad': 1},\n",
        "                            verticalalignment='bottom', horizontalalignment='center')\n",
        "\n",
        "    for idx, (a, b) in enumerate(zip(ticks, F1)):\n",
        "\n",
        "        if idx == 0:\n",
        "            ax.text(a - 0.005, b - 0.45, str(b), fontsize=6, bbox={'facecolor': 'teal', 'alpha': 0.1, 'pad': 1},\n",
        "                    verticalalignment='bottom', horizontalalignment='center')\n",
        "        else:\n",
        "            if idx == 1:\n",
        "                ax.text(a, b - 0.45, str(b), fontsize=6, bbox={'facecolor': 'teal', 'alpha': 0.1, 'pad': 1},\n",
        "                        verticalalignment='bottom', horizontalalignment='center')\n",
        "            else:\n",
        "                if idx == 2:\n",
        "                    ax.text(a-0.12, b-0.015, str(b), fontsize=6,\n",
        "                            bbox={'facecolor': 'teal', 'alpha': 0.1, 'pad': 1},\n",
        "                            verticalalignment='bottom', horizontalalignment='center')\n",
        "\n",
        "    plt.legend(frameon=False)\n",
        "    plt.savefig(\"Model1 Score.png\", dpi=192)\n",
        "    fig.set_dpi(100)\n",
        "    plt.show()\n",
        "\n",
        "plot_score_graph(\"Model 1: EM and F1 Score over Epochs\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}