HopeSpeechEDI_XLMR_Malayalam _v2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "HopeSpeechEDI_XLMR_Malayalam.ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "66d1de2202374f3689ea014824b63491": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_0c97c9e45c44415dadc3a0a237eb5b5b",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_0cfe1d1864f3461bbbdea8fe5b0ad3cb",
              "IPY_MODEL_a978e1143be149e3bf19e82fc2f42d7c"
            ]
          }
        },
        "0c97c9e45c44415dadc3a0a237eb5b5b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "0cfe1d1864f3461bbbdea8fe5b0ad3cb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_1613ae72b0e84490a8168cd9f021c32a",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 512,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 512,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_bc446d761a3741308039fc60bdd23262"
          }
        },
        "a978e1143be149e3bf19e82fc2f42d7c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_78197b526eb8410e810979f99f234f6c",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 512/512 [00:06&lt;00:00, 73.8B/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_86a650edcd104eda9f1b2b899cde525c"
          }
        },
        "1613ae72b0e84490a8168cd9f021c32a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "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"
          }
        },
        "bc446d761a3741308039fc60bdd23262": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "78197b526eb8410e810979f99f234f6c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "86a650edcd104eda9f1b2b899cde525c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "5cf4184305ed4116bdbb591bf2eeaaec": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_d5d26df73ba64708b45b4e3f89229f04",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_a29329f82f554ae98c6c65008c7aaf34",
              "IPY_MODEL_0e218f795ab34177bc9162b10e5935bc"
            ]
          }
        },
        "d5d26df73ba64708b45b4e3f89229f04": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "a29329f82f554ae98c6c65008c7aaf34": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_8242fe40097b4a0f8b2f6208f5777bb7",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 5069051,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 5069051,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_a524792107a24b73a5bcb97d6d1b0517"
          }
        },
        "0e218f795ab34177bc9162b10e5935bc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_ea80c747b0d149ba8cf3debe3d540220",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 5.07M/5.07M [00:06&lt;00:00, 750kB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_769af2f421724b83abb002068c6edb9b"
          }
        },
        "8242fe40097b4a0f8b2f6208f5777bb7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "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"
          }
        },
        "a524792107a24b73a5bcb97d6d1b0517": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "ea80c747b0d149ba8cf3debe3d540220": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "769af2f421724b83abb002068c6edb9b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "23ff93eb1da1493087c8adc58c213288": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_63727149341d47ef81265b5a8f5b15cf",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_d3dffa30b9f04355a4b0fde121b2eb5b",
              "IPY_MODEL_5be0324b9ce34efca72cfe138499f3da"
            ]
          }
        },
        "63727149341d47ef81265b5a8f5b15cf": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "d3dffa30b9f04355a4b0fde121b2eb5b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_3489bff83f5c46efaa6c62998aba8faf",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 9096718,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 9096718,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c1a1df0491714d5b9efed0e8f3b764a7"
          }
        },
        "5be0324b9ce34efca72cfe138499f3da": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_0158b8f81fa24d1bba8b41de2af41498",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 9.10M/9.10M [00:06&lt;00:00, 1.41MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_b07dc78c0a51472b918d1fc68c578baf"
          }
        },
        "3489bff83f5c46efaa6c62998aba8faf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "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"
          }
        },
        "c1a1df0491714d5b9efed0e8f3b764a7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "0158b8f81fa24d1bba8b41de2af41498": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "b07dc78c0a51472b918d1fc68c578baf": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "8a5542f3d67544288e449e8172cd053f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "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_5280b85c17a84b2d82d9e2a496c3de29",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_5d207134064e43369c37518317e348b7",
              "IPY_MODEL_5792f249efdd4bb4972e90c705a202f9"
            ]
          }
        },
        "5280b85c17a84b2d82d9e2a496c3de29": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "5d207134064e43369c37518317e348b7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_4270e52728fb4fe19fdf50664c860b9c",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1115590446,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1115590446,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_b19b3207f9b04536a0e5fef26773492f"
          }
        },
        "5792f249efdd4bb4972e90c705a202f9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_30afddfaed42443a95b0224a48f314cb",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1.12G/1.12G [00:25&lt;00:00, 43.4MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_668597d7a9df4ae8b8c98b7fd241ce72"
          }
        },
        "4270e52728fb4fe19fdf50664c860b9c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "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"
          }
        },
        "b19b3207f9b04536a0e5fef26773492f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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
          }
        },
        "30afddfaed42443a95b0224a48f314cb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "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"
          }
        },
        "668597d7a9df4ae8b8c98b7fd241ce72": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "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": "markdown",
      "metadata": {
        "id": "elqe9PQrVH6P"
      },
      "source": [
        "### Performing checks for the resources available"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fSELLChxa5su",
        "outputId": "e1a99d83-f07d-470a-82a3-85a93670d314"
      },
      "source": [
        "import torch\r\n",
        "\r\n",
        "# If there's a GPU available...\r\n",
        "if torch.cuda.is_available():    \r\n",
        "\r\n",
        "    # Tell PyTorch to use the GPU.    \r\n",
        "    device = torch.device(\"cuda\")\r\n",
        "\r\n",
        "    print('There are %d GPU(s) available.' % torch.cuda.device_count())\r\n",
        "\r\n",
        "    print('We will use the GPU:', torch.cuda.get_device_name(0))\r\n",
        "\r\n",
        "# If not...\r\n",
        "else:\r\n",
        "    print('No GPU available, using the CPU instead.')\r\n",
        "    device = torch.device(\"cpu\")"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "There are 1 GPU(s) available.\n",
            "We will use the GPU: Tesla T4\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8SokVowWo3cc",
        "outputId": "b96425d0-80ae-4944-d777-7645cd6459bb"
      },
      "source": [
        "!nvidia-smi"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Sat Jan  9 16:59:41 2021       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 460.27.04    Driver Version: 418.67       CUDA Version: 10.1     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   64C    P8    11W /  70W |     10MiB / 15079MiB |      0%      Default |\n",
            "|                               |                      |                 ERR! |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K-gbOgflVRxg"
      },
      "source": [
        "### Installing huggingface transformers library"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DbDuxKy0a9Zl",
        "outputId": "622011c6-0776-466a-8da5-beca8d0d20b4"
      },
      "source": [
        "!pip install transformers"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting transformers\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/50/0c/7d5950fcd80b029be0a8891727ba21e0cd27692c407c51261c3c921f6da3/transformers-4.1.1-py3-none-any.whl (1.5MB)\n",
            "\u001b[K     |████████████████████████████████| 1.5MB 13.1MB/s \n",
            "\u001b[?25hCollecting tokenizers==0.9.4\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/0f/1c/e789a8b12e28be5bc1ce2156cf87cb522b379be9cadc7ad8091a4cc107c4/tokenizers-0.9.4-cp36-cp36m-manylinux2010_x86_64.whl (2.9MB)\n",
            "\u001b[K     |████████████████████████████████| 2.9MB 56.6MB/s \n",
            "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.6/dist-packages (from transformers) (4.41.1)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from transformers) (20.8)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.6/dist-packages (from transformers) (2019.12.20)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from transformers) (1.19.4)\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 51.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from transformers) (2.23.0)\n",
            "Requirement already satisfied: dataclasses; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from transformers) (0.8)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.6/dist-packages (from transformers) (3.0.12)\n",
            "Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->transformers) (2.4.7)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.15.0)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (7.1.2)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.0.0)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2020.12.5)\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) (1.24.3)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2.10)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (3.0.4)\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=893261 sha256=7adcd5ef927b7124a653fa259aada37a681e7338db9eea9079a3e9aafa0a0d1f\n",
            "  Stored in directory: /root/.cache/pip/wheels/29/3c/fd/7ce5c3f0666dab31a50123635e6fb5e19ceb42ce38d4e58f45\n",
            "Successfully built sacremoses\n",
            "Installing collected packages: tokenizers, sacremoses, transformers\n",
            "Successfully installed sacremoses-0.0.43 tokenizers-0.9.4 transformers-4.1.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bPwaYLiRSNPC"
      },
      "source": [
        "## Importing necessary packages"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0XdFrhr8SRG0"
      },
      "source": [
        "import pandas as pd\r\n",
        "import numpy as np \r\n",
        "import random\r\n",
        "from sklearn.preprocessing import LabelEncoder\r\n",
        "from sklearn.utils.class_weight import compute_class_weight\r\n",
        "import time\r\n",
        "import datetime\r\n",
        "from sklearn.metrics import classification_report,confusion_matrix\r\n",
        "import random\r\n",
        "import time\r\n",
        "import torch.nn as nn\r\n",
        "from transformers import AutoModel , AutoModelForSequenceClassification, AutoConfig , AutoTokenizer , AdamW ,get_linear_schedule_with_warmup \r\n",
        "import torch\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "import seaborn as sns\r\n",
        "from matplotlib import rc\r\n",
        "from matplotlib.ticker import MaxNLocator"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dROX3juhVCxC"
      },
      "source": [
        "## Setting some configuration values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Kzafv4BzVCC9"
      },
      "source": [
        "# Use plot styling from seaborn.\r\n",
        "sns.set(style='darkgrid')\r\n",
        "\r\n",
        "# Increase the plot size and font size.\r\n",
        "sns.set(font_scale=1.5)\r\n",
        "plt.rcParams[\"figure.figsize\"] = (16,12)\r\n",
        "\r\n",
        "# Set the seed value all over the place to make this reproducible.\r\n",
        "seed_val = 42\r\n",
        "\r\n",
        "random.seed(seed_val)\r\n",
        "np.random.seed(seed_val)\r\n",
        "torch.manual_seed(seed_val)\r\n",
        "torch.cuda.manual_seed_all(seed_val)"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2v1AIN_OVX4t"
      },
      "source": [
        "## Importing drive into the colaboratory"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Lofimyh2a_tp",
        "outputId": "08f9477c-4a34-4876-e40f-f6851a0187f5"
      },
      "source": [
        "from google.colab import drive \r\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yOK-M00dVeW2"
      },
      "source": [
        "## Importing dataset directories"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-y-JRPKBbCTv"
      },
      "source": [
        "from os import getcwd , listdir \r\n",
        "from os.path import join\r\n",
        "curr_dir = getcwd()\r\n",
        "drive_dir = join(curr_dir , 'drive','MyDrive')\r\n",
        "dataset_dir = join(drive_dir,'Codalab','HopeSpeech Detection For Equality,DiversityAnd Inclusion','Dataset')"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7YU1h8WhmqvK",
        "outputId": "2ce79930-6e1d-4a9b-f553-a94ecd4c19aa"
      },
      "source": [
        "listdir(dataset_dir)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['english_hope_dev.csv',\n",
              " 'english_hope_train.csv',\n",
              " 'malayalam_hope_dev.csv',\n",
              " 'malayalam_hope_train.csv',\n",
              " 'tamil_hope_first_dev.csv',\n",
              " 'tamil_hope_first_train.csv',\n",
              " 'tamil_hope_test.csv',\n",
              " 'malayalam_hope_test.csv',\n",
              " 'english_hope_test.csv']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FUPkZxjLVjLK"
      },
      "source": [
        "## Loading training data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "id": "joone3H4bINR",
        "outputId": "f6296f79-c444-4d37-ebb6-3eda4fd5b99d"
      },
      "source": [
        "train_df_path = join(dataset_dir,'malayalam_hope_train.csv')\r\n",
        "# Load the dataset into a pandas dataframe.\r\n",
        "train_df = pd.read_csv(train_df_path)\r\n",
        "\r\n",
        "# Report the number of sentences.\r\n",
        "print('Number of training sentences: {:,}\\n'.format(train_df.shape[0]))\r\n",
        "\r\n",
        "# Display 10 random rows from the data.\r\n",
        "train_df.sample(10)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of training sentences: 8,564\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>7913</th>\n",
              "      <td>Iniyippo ningal pubg kalichalum kuzhappamilla ...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6065</th>\n",
              "      <td>ആ ചെങ്ങായിക്കെല്ലും എത്ര നല്ല മൊഞ്ചത്തി പെണ്ണി...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1649</th>\n",
              "      <td>Suryammaa ishanikkaaaa happy wedding anniversa...</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>710</th>\n",
              "      <td>Ente bayi ittekkana shoe muthal jaddy vare ava...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1918</th>\n",
              "      <td>Ayaal entha parayunath ennu ayaalkk tahnne ari...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3617</th>\n",
              "      <td>ചക്ക കുരു പൊട്ടിയോ</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>233</th>\n",
              "      <td>പിന്നെ അമ്മയെയും എന്റെ അന്വേഷണം അറിയിക്കണം</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6927</th>\n",
              "      <td>Helo  iku പകരം എന്തുണ്ട് ?</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2094</th>\n",
              "      <td>Kevin mandane kaanumbhaz santhosh pandittine o...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3114</th>\n",
              "      <td>പൊളിച്ചു bro</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                   text            label\n",
              "7913  Iniyippo ningal pubg kalichalum kuzhappamilla ...  Non_hope_speech\n",
              "6065  ആ ചെങ്ങായിക്കെല്ലും എത്ര നല്ല മൊഞ്ചത്തി പെണ്ണി...  Non_hope_speech\n",
              "1649  Suryammaa ishanikkaaaa happy wedding anniversa...      Hope_speech\n",
              "710   Ente bayi ittekkana shoe muthal jaddy vare ava...  Non_hope_speech\n",
              "1918  Ayaal entha parayunath ennu ayaalkk tahnne ari...  Non_hope_speech\n",
              "3617                                 ചക്ക കുരു പൊട്ടിയോ  Non_hope_speech\n",
              "233          പിന്നെ അമ്മയെയും എന്റെ അന്വേഷണം അറിയിക്കണം      Hope_speech\n",
              "6927                         Helo  iku പകരം എന്തുണ്ട് ?  Non_hope_speech\n",
              "2094  Kevin mandane kaanumbhaz santhosh pandittine o...  Non_hope_speech\n",
              "3114                                       പൊളിച്ചു bro      Hope_speech"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w4VW351225YK",
        "outputId": "4612c3b5-8f7c-43ce-cb7f-cbb3d7cebe4a"
      },
      "source": [
        "train_df['label'].value_counts()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Non_hope_speech    6205\n",
              "Hope_speech        1668\n",
              "not-malayalam       691\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ooOv9QZtnLIF"
      },
      "source": [
        "\r\n",
        "le = LabelEncoder()\r\n",
        "train_df['label'] = le.fit_transform(train_df['label'])"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a89Bf4LpnigU",
        "outputId": "9016aea3-e355-48f5-9960-f2edeb786fa9"
      },
      "source": [
        "train_df['label'].value_counts()"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    6205\n",
              "0    1668\n",
              "2     691\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bvNeR4lTVnK0"
      },
      "source": [
        "### Loading `train_sentences` and `train_labels`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EufQZx2dbKMT"
      },
      "source": [
        "# Get the lists of sentences and their labels.\r\n",
        "train_sentences = train_df['text'].values\r\n",
        "train_labels = train_df['label'].values"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wnZSg5f7ZWs_"
      },
      "source": [
        "## Saving Class Names"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "83tTT-wbZa1o"
      },
      "source": [
        "class_names = ['Hope_speech' , 'Non-hope-speech'  , 'not-malayalam']"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ToUbAgZ9ap63"
      },
      "source": [
        "## Loading the validation data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "id": "Dle8VzoAjOLS",
        "outputId": "175ced64-9af4-4af1-c905-d8d4e31b3cf9"
      },
      "source": [
        "val_df_path = join(dataset_dir,'malayalam_hope_dev.csv')\r\n",
        "# Load the dataset into a pandas dataframe.\r\n",
        "val_df = pd.read_csv(val_df_path)\r\n",
        "\r\n",
        "# Report the number of sentences.\r\n",
        "print('Number of training sentences: {:,}\\n'.format(val_df.shape[0]))\r\n",
        "\r\n",
        "# Display 10 random rows from the data.\r\n",
        "val_df.sample(10)"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of training sentences: 1,070\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>424</th>\n",
              "      <td>ഇന്ത്യൻ മതേതര ജനാധിപത്യ രാജ്യമാണ്</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>722</th>\n",
              "      <td>@Mallu crafts thanik or gay aya kutti ondavath...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>306</th>\n",
              "      <td>Helo. Lesbiann.. chodichal. Vanno</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>116</th>\n",
              "      <td>Nikesh  nte  kada  kettapol  sangadam  തോന്നി...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>551</th>\n",
              "      <td>Ithil parenjekana korey app onnum njan kettit ...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>263</th>\n",
              "      <td>Gay enn kandappo nice aayitt kaanan vanirikkka</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>333</th>\n",
              "      <td>Public reactions kalakki but kurach koodi leng...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>575</th>\n",
              "      <td>പോയി ചതൂ ഡേ</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>597</th>\n",
              "      <td>Dr ഈ പ്രതിരോധ ശേഷി ധാരാളം ഉള്ള ആൾക്ക് എത്ര തവണ...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>634</th>\n",
              "      <td>Happy mother's day surya amma.. nകരയിപ്പിച്ചു ...</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                  text            label\n",
              "424                  ഇന്ത്യൻ മതേതര ജനാധിപത്യ രാജ്യമാണ്  Non_hope_speech\n",
              "722  @Mallu crafts thanik or gay aya kutti ondavath...  Non_hope_speech\n",
              "306                  Helo. Lesbiann.. chodichal. Vanno  Non_hope_speech\n",
              "116   Nikesh  nte  kada  kettapol  sangadam  തോന്നി...  Non_hope_speech\n",
              "551  Ithil parenjekana korey app onnum njan kettit ...  Non_hope_speech\n",
              "263     Gay enn kandappo nice aayitt kaanan vanirikkka  Non_hope_speech\n",
              "333  Public reactions kalakki but kurach koodi leng...  Non_hope_speech\n",
              "575                                        പോയി ചതൂ ഡേ  Non_hope_speech\n",
              "597  Dr ഈ പ്രതിരോധ ശേഷി ധാരാളം ഉള്ള ആൾക്ക് എത്ര തവണ...  Non_hope_speech\n",
              "634  Happy mother's day surya amma.. nകരയിപ്പിച്ചു ...      Hope_speech"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AypCmtBcDV6v",
        "outputId": "41f38a3f-0812-4b48-c995-aaa625aae50a"
      },
      "source": [
        "val_df['label'].value_counts()"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Non_hope_speech    784\n",
              "Hope_speech        190\n",
              "not-malayalam       96\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-CgK81lGDPKC"
      },
      "source": [
        "val_df['label'] = le.transform(val_df['label'])"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yfgn-E6JDZQZ",
        "outputId": "526d8431-bfd9-470b-b6d5-1130d6681a8c"
      },
      "source": [
        "val_df['label'].value_counts()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    784\n",
              "0    190\n",
              "2     96\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DMdNNVqrDCuN"
      },
      "source": [
        "### Loading `val_sentences` and `val_labels`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LvGhjUkaDCuO"
      },
      "source": [
        "# Get the lists of sentences and their labels.\r\n",
        "val_sentences = val_df['text'].values\r\n",
        "val_labels = val_df['label'].values"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fS2UcOiEVuhm"
      },
      "source": [
        "## Handling class imbalance using `sklearn's compute_class_weight`\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WL4tP_YkI1mY"
      },
      "source": [
        "\r\n",
        "# #compute the class weights\r\n",
        "# class_wts = compute_class_weight('balanced', np.unique(train_labels), train_labels)\r\n",
        "\r\n",
        "# print(class_wts)"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z_WjMTuibsrf"
      },
      "source": [
        "## Helper Functions"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Bef28eHNV-_d"
      },
      "source": [
        "#### 1) Update Interval Function"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lBK0ujfEbMir"
      },
      "source": [
        "def good_update_interval(total_iters, num_desired_updates):\r\n",
        "    '''\r\n",
        "    This function will try to pick an intelligent progress update interval \r\n",
        "    based on the magnitude of the total iterations.\r\n",
        "\r\n",
        "    Parameters:\r\n",
        "      `total_iters` - The number of iterations in the for-loop.\r\n",
        "      `num_desired_updates` - How many times we want to see an update over the \r\n",
        "                              course of the for-loop.\r\n",
        "    '''\r\n",
        "    # Divide the total iterations by the desired number of updates. Most likely\r\n",
        "    # this will be some ugly number.\r\n",
        "    exact_interval = total_iters / num_desired_updates\r\n",
        "\r\n",
        "    # The `round` function has the ability to round down a number to, e.g., the\r\n",
        "    # nearest thousandth: round(exact_interval, -3)\r\n",
        "    #\r\n",
        "    # To determine the magnitude to round to, find the magnitude of the total,\r\n",
        "    # and then go one magnitude below that.\r\n",
        "\r\n",
        "    # Get the order of magnitude of the total.\r\n",
        "    order_of_mag = len(str(total_iters)) - 1\r\n",
        "\r\n",
        "    # Our update interval should be rounded to an order of magnitude smaller. \r\n",
        "    round_mag = order_of_mag - 1\r\n",
        "\r\n",
        "    # Round down and cast to an int.\r\n",
        "    update_interval = int(round(exact_interval, -round_mag))\r\n",
        "\r\n",
        "    # Don't allow the interval to be zero!\r\n",
        "    if update_interval == 0:\r\n",
        "        update_interval = 1\r\n",
        "\r\n",
        "    return update_interval"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eMGycwGAWFyO"
      },
      "source": [
        "#### 2) Format time function"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c9zVyNQLm-3q"
      },
      "source": [
        "def format_time(elapsed):\r\n",
        "    '''\r\n",
        "    Takes a time in seconds and returns a string hh:mm:ss\r\n",
        "    '''\r\n",
        "    # Round to the nearest second.\r\n",
        "    elapsed_rounded = int(round((elapsed)))\r\n",
        "    \r\n",
        "    # Format as hh:mm:ss\r\n",
        "    return str(datetime.timedelta(seconds=elapsed_rounded))\r\n"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EBAo04ENWw3n"
      },
      "source": [
        "#### 3) Make Smart Batches function"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nof-AUeoiZc2"
      },
      "source": [
        "def make_smart_batches(text_samples, labels, batch_size,tokenizer,max_input_length):\r\n",
        "    '''\r\n",
        "    This function combines all of the required steps to prepare batches.\r\n",
        "    '''\r\n",
        "\r\n",
        "    print('Creating Smart Batches from {:,} examples with batch size {:,}...\\n'.format(len(text_samples), batch_size))\r\n",
        "\r\n",
        "    # =========================\r\n",
        "    #   Tokenize & Truncate\r\n",
        "    # =========================\r\n",
        "\r\n",
        "    full_input_ids = []\r\n",
        "\r\n",
        "    # Tokenize all training examples\r\n",
        "    print('Tokenizing {:,} samples...'.format(len(labels)))\r\n",
        "\r\n",
        "    # Choose an interval on which to print progress updates.\r\n",
        "    update_interval = good_update_interval(total_iters=len(labels), num_desired_updates=10)\r\n",
        "\r\n",
        "    # For each training example...\r\n",
        "    for text in text_samples:\r\n",
        "        \r\n",
        "        # Report progress.\r\n",
        "        if ((len(full_input_ids) % update_interval) == 0):\r\n",
        "            print('  Tokenized {:,} samples.'.format(len(full_input_ids)))\r\n",
        "\r\n",
        "        # Tokenize the sample.\r\n",
        "        input_ids = tokenizer.encode(text=text,              # Text to encode.\r\n",
        "                                    add_special_tokens=True, # Do add specials.\r\n",
        "                                    max_length=max_input_length,      \r\n",
        "                                    truncation=True,         # Do Truncate!\r\n",
        "                                    padding=False)           # DO NOT pad.\r\n",
        "                                    \r\n",
        "        # Add the tokenized result to our list.\r\n",
        "        full_input_ids.append(input_ids)\r\n",
        "        \r\n",
        "    print('DONE.')\r\n",
        "    print('{:>10,} samples\\n'.format(len(full_input_ids)))\r\n",
        "\r\n",
        "    # =========================\r\n",
        "    #      Select Batches\r\n",
        "    # =========================    \r\n",
        "\r\n",
        "    # Sort the two lists together by the length of the input sequence.\r\n",
        "    samples = sorted(zip(full_input_ids, labels), key=lambda x: len(x[0]))\r\n",
        "\r\n",
        "    print('{:>10,} samples after sorting\\n'.format(len(samples)))\r\n",
        "\r\n",
        "    import random\r\n",
        "\r\n",
        "    # List of batches that we'll construct.\r\n",
        "    batch_ordered_sentences = []\r\n",
        "    batch_ordered_labels = []\r\n",
        "\r\n",
        "    print('Creating batches of size {:}...'.format(batch_size))\r\n",
        "\r\n",
        "    # Choose an interval on which to print progress updates.\r\n",
        "    update_interval = good_update_interval(total_iters=len(samples), num_desired_updates=10)\r\n",
        "    \r\n",
        "    # Loop over all of the input samples...    \r\n",
        "    while len(samples) > 0:\r\n",
        "        \r\n",
        "        # Report progress.\r\n",
        "        if ((len(batch_ordered_sentences) % update_interval) == 0 \\\r\n",
        "            and not len(batch_ordered_sentences) == 0):\r\n",
        "            print('  Selected {:,} batches.'.format(len(batch_ordered_sentences)))\r\n",
        "\r\n",
        "        # `to_take` is our actual batch size. It will be `batch_size` until \r\n",
        "        # we get to the last batch, which may be smaller. \r\n",
        "        to_take = min(batch_size, len(samples))\r\n",
        "\r\n",
        "        # Pick a random index in the list of remaining samples to start\r\n",
        "        # our batch at.\r\n",
        "        select = random.randint(0, len(samples) - to_take)\r\n",
        "\r\n",
        "        # Select a contiguous batch of samples starting at `select`.\r\n",
        "        #print(\"Selecting batch from {:} to {:}\".format(select, select+to_take))\r\n",
        "        batch = samples[select:(select + to_take)]\r\n",
        "\r\n",
        "        #print(\"Batch length:\", len(batch))\r\n",
        "\r\n",
        "        # Each sample is a tuple--split them apart to create a separate list of \r\n",
        "        # sequences and a list of labels for this batch.\r\n",
        "        batch_ordered_sentences.append([s[0] for s in batch])\r\n",
        "        batch_ordered_labels.append([s[1] for s in batch])\r\n",
        "\r\n",
        "        # Remove these samples from the list.\r\n",
        "        del samples[select:select + to_take]\r\n",
        "\r\n",
        "    print('\\n  DONE - Selected {:,} batches.\\n'.format(len(batch_ordered_sentences)))\r\n",
        "\r\n",
        "    # =========================\r\n",
        "    #        Add Padding\r\n",
        "    # =========================    \r\n",
        "\r\n",
        "    print('Padding out sequences within each batch...')\r\n",
        "\r\n",
        "    py_inputs = []\r\n",
        "    py_attn_masks = []\r\n",
        "    py_labels = []\r\n",
        "\r\n",
        "    # For each batch...\r\n",
        "    for (batch_inputs, batch_labels) in zip(batch_ordered_sentences, batch_ordered_labels):\r\n",
        "\r\n",
        "        # New version of the batch, this time with padded sequences and now with\r\n",
        "        # attention masks defined.\r\n",
        "        batch_padded_inputs = []\r\n",
        "        batch_attn_masks = []\r\n",
        "        \r\n",
        "        # First, find the longest sample in the batch. \r\n",
        "        # Note that the sequences do currently include the special tokens!\r\n",
        "        max_size = max([len(sen) for sen in batch_inputs])\r\n",
        "\r\n",
        "        # For each input in this batch...\r\n",
        "        for sen in batch_inputs:\r\n",
        "            \r\n",
        "            # How many pad tokens do we need to add?\r\n",
        "            num_pads = max_size - len(sen)\r\n",
        "\r\n",
        "            # Add `num_pads` padding tokens to the end of the sequence.\r\n",
        "            padded_input = sen + [tokenizer.pad_token_id]*num_pads\r\n",
        "\r\n",
        "            # Define the attention mask--it's just a `1` for every real token\r\n",
        "            # and a `0` for every padding token.\r\n",
        "            attn_mask = [1] * len(sen) + [0] * num_pads\r\n",
        "\r\n",
        "            # Add the padded results to the batch.\r\n",
        "            batch_padded_inputs.append(padded_input)\r\n",
        "            batch_attn_masks.append(attn_mask)\r\n",
        "\r\n",
        "        # Our batch has been padded, so we need to save this updated batch.\r\n",
        "        # We also need the inputs to be PyTorch tensors, so we'll do that here.\r\n",
        "        # Todo - Michael's code specified \"dtype=torch.long\"\r\n",
        "        py_inputs.append(torch.tensor(batch_padded_inputs))\r\n",
        "        py_attn_masks.append(torch.tensor(batch_attn_masks))\r\n",
        "        py_labels.append(torch.tensor(batch_labels))\r\n",
        "    \r\n",
        "    print('  DONE.')\r\n",
        "\r\n",
        "    # Return the smart-batched dataset!\r\n",
        "    return (py_inputs, py_attn_masks, py_labels)\r\n"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5816-I6miXJ1"
      },
      "source": [
        "### 4) Make Smart Batches (On Test Set)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ssv3g8tzEdWa"
      },
      "source": [
        "def make_smart_batches_on_test(text_samples, text_ids, batch_size,tokenizer,max_input_length):\r\n",
        "    '''\r\n",
        "    This function combines all of the required steps to prepare batches.\r\n",
        "    '''\r\n",
        "\r\n",
        "    print('Creating Smart Batches from {:,} examples with batch size {:,}...\\n'.format(len(text_samples), batch_size))\r\n",
        "\r\n",
        "    # =========================\r\n",
        "    #   Tokenize & Truncate\r\n",
        "    # =========================\r\n",
        "\r\n",
        "    full_input_ids = []\r\n",
        "\r\n",
        "    # Tokenize all training examples\r\n",
        "    print('Tokenizing {:,} samples...'.format(len(text_samples)))\r\n",
        "\r\n",
        "    # Choose an interval on which to print progress updates.\r\n",
        "    update_interval = good_update_interval(total_iters=len(text_samples), num_desired_updates=10)\r\n",
        "\r\n",
        "    # For each training example...\r\n",
        "    for text in text_samples:\r\n",
        "        \r\n",
        "        # Report progress.\r\n",
        "        if ((len(full_input_ids) % update_interval) == 0):\r\n",
        "            print('  Tokenized {:,} samples.'.format(len(full_input_ids)))\r\n",
        "\r\n",
        "        # Tokenize the sample.\r\n",
        "        input_ids = tokenizer.encode(text=text,              # Text to encode.\r\n",
        "                                    add_special_tokens=True, # Do add specials.\r\n",
        "                                    max_length=max_input_length,      \r\n",
        "                                    truncation=True,         # Do Truncate!\r\n",
        "                                    padding=False)           # DO NOT pad.\r\n",
        "                                    \r\n",
        "        # Add the tokenized result to our list.\r\n",
        "        full_input_ids.append(input_ids)\r\n",
        "        \r\n",
        "    print('DONE.')\r\n",
        "    print('{:>10,} samples\\n'.format(len(full_input_ids)))\r\n",
        "\r\n",
        "    # =========================\r\n",
        "    #      Select Batches\r\n",
        "    # =========================    \r\n",
        "\r\n",
        "    # Sort the two lists together by the length of the input sequence.\r\n",
        "    samples = sorted(zip(full_input_ids, text_ids), key=lambda x: len(x[0]))\r\n",
        "\r\n",
        "    print('{:>10,} samples after sorting\\n'.format(len(samples)))\r\n",
        "\r\n",
        "    import random\r\n",
        "\r\n",
        "    # List of batches that we'll construct.\r\n",
        "    batch_ordered_sentences = []\r\n",
        "    batch_ordered_ids = []\r\n",
        "\r\n",
        "    print('Creating batches of size {:}...'.format(batch_size))\r\n",
        "\r\n",
        "    # Choose an interval on which to print progress updates.\r\n",
        "    update_interval = good_update_interval(total_iters=len(samples), num_desired_updates=10)\r\n",
        "    \r\n",
        "    # Loop over all of the input samples...    \r\n",
        "    while len(samples) > 0:\r\n",
        "        \r\n",
        "        # Report progress.\r\n",
        "        if ((len(batch_ordered_sentences) % update_interval) == 0 \\\r\n",
        "            and not len(batch_ordered_sentences) == 0):\r\n",
        "            print('  Selected {:,} batches.'.format(len(batch_ordered_sentences)))\r\n",
        "\r\n",
        "        # `to_take` is our actual batch size. It will be `batch_size` until \r\n",
        "        # we get to the last batch, which may be smaller. \r\n",
        "        to_take = min(batch_size, len(samples))\r\n",
        "\r\n",
        "        # Pick a random index in the list of remaining samples to start\r\n",
        "        # our batch at.\r\n",
        "        select = random.randint(0, len(samples) - to_take)\r\n",
        "\r\n",
        "        # Select a contiguous batch of samples starting at `select`.\r\n",
        "        #print(\"Selecting batch from {:} to {:}\".format(select, select+to_take))\r\n",
        "        batch = samples[select:(select + to_take)]\r\n",
        "\r\n",
        "        #print(\"Batch length:\", len(batch))\r\n",
        "\r\n",
        "        # Each sample is a tuple--split them apart to create a separate list of \r\n",
        "        # sequences and a list of labels for this batch.\r\n",
        "        batch_ordered_sentences.append([s[0] for s in batch])\r\n",
        "        batch_ordered_ids.append([s[1] for s in batch])\r\n",
        "\r\n",
        "        # Remove these samples from the list.\r\n",
        "        del samples[select:select + to_take]\r\n",
        "\r\n",
        "    print('\\n  DONE - Selected {:,} batches.\\n'.format(len(batch_ordered_sentences)))\r\n",
        "\r\n",
        "    # =========================\r\n",
        "    #        Add Padding\r\n",
        "    # =========================    \r\n",
        "\r\n",
        "    print('Padding out sequences within each batch...')\r\n",
        "\r\n",
        "    py_inputs = []\r\n",
        "    py_attn_masks = []\r\n",
        "    py_ids = []\r\n",
        "\r\n",
        "    # For each batch...\r\n",
        "    for (batch_inputs, batch_ids) in zip(batch_ordered_sentences, batch_ordered_ids):\r\n",
        "\r\n",
        "        # New version of the batch, this time with padded sequences and now with\r\n",
        "        # attention masks defined.\r\n",
        "        batch_padded_inputs = []\r\n",
        "        batch_attn_masks = []\r\n",
        "        \r\n",
        "        # First, find the longest sample in the batch. \r\n",
        "        # Note that the sequences do currently include the special tokens!\r\n",
        "        max_size = max([len(sen) for sen in batch_inputs])\r\n",
        "\r\n",
        "        # For each input in this batch...\r\n",
        "        for sen in batch_inputs:\r\n",
        "            \r\n",
        "            # How many pad tokens do we need to add?\r\n",
        "            num_pads = max_size - len(sen)\r\n",
        "\r\n",
        "            # Add `num_pads` padding tokens to the end of the sequence.\r\n",
        "            padded_input = sen + [tokenizer.pad_token_id]*num_pads\r\n",
        "\r\n",
        "            # Define the attention mask--it's just a `1` for every real token\r\n",
        "            # and a `0` for every padding token.\r\n",
        "            attn_mask = [1] * len(sen) + [0] * num_pads\r\n",
        "\r\n",
        "            # Add the padded results to the batch.\r\n",
        "            batch_padded_inputs.append(padded_input)\r\n",
        "            batch_attn_masks.append(attn_mask)\r\n",
        "\r\n",
        "        # Our batch has been padded, so we need to save this updated batch.\r\n",
        "        # We also need the inputs to be PyTorch tensors, so we'll do that here.\r\n",
        "        # Todo - Michael's code specified \"dtype=torch.long\"\r\n",
        "        py_inputs.append(torch.tensor(batch_padded_inputs))\r\n",
        "        py_attn_masks.append(torch.tensor(batch_attn_masks))\r\n",
        "        py_ids.append(torch.tensor(batch_ids))\r\n",
        "    \r\n",
        "    print('  DONE.')\r\n",
        "\r\n",
        "    # Return the smart-batched dataset!\r\n",
        "    return (py_inputs, py_attn_masks, py_ids)\r\n"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fT7qY9gtEdWc"
      },
      "source": [
        "### 5) Function for calculating accuracy per class"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tSHVuo18EdWc"
      },
      "source": [
        "def accuracy_per_class(preds, labels , label_dict):\r\n",
        "    \r\n",
        "    preds_flat = np.argmax(preds, axis=1).flatten()\r\n",
        "    labels_flat = labels.flatten()\r\n",
        "\r\n",
        "    for label in np.unique(labels_flat):\r\n",
        "        y_preds = preds_flat[labels_flat==label]\r\n",
        "        y_true = labels_flat[labels_flat==label]\r\n",
        "        print(f'Class: {label_dict[label]}')\r\n",
        "        print(f'Accuracy: {round(100*len(y_preds[y_preds==label])/len(y_true) , 2)}\\n')"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MLcm5L2MEdWd"
      },
      "source": [
        "### 6) Function for plotting training history"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rLLSWnMTEdWd"
      },
      "source": [
        "def plot_training_history(history):\r\n",
        "  fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(18, 6))\r\n",
        "\r\n",
        "  ax1.plot(history['train_loss'], label='train loss')\r\n",
        "  ax1.plot(history['val_loss'], label='validation loss')\r\n",
        "\r\n",
        "  ax1.xaxis.set_major_locator(MaxNLocator(integer=True))\r\n",
        "  ax1.set_ylim([-0.05, 1.05])\r\n",
        "  ax1.legend()\r\n",
        "  ax1.set_ylabel('Loss')\r\n",
        "  ax1.set_xlabel('Epoch')\r\n",
        "\r\n",
        "  ax2.plot(history['train_acc'], label='train accuracy')\r\n",
        "  ax2.plot(history['val_acc'], label='validation accuracy')\r\n",
        "\r\n",
        "  ax2.xaxis.set_major_locator(MaxNLocator(integer=True))\r\n",
        "  ax2.set_ylim([-0.05, 1.05])\r\n",
        "  ax2.legend()\r\n",
        "\r\n",
        "  ax2.set_ylabel('Accuracy')\r\n",
        "  ax2.set_xlabel('Epoch')\r\n",
        "\r\n",
        "  fig.suptitle('Training history')"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mHcxGVmREdWd"
      },
      "source": [
        "### 7) Function to check accuracy"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VF8tJIO6EdWe"
      },
      "source": [
        "def check_accuracy(predictions,true_labels):\r\n",
        "  # Combine the results across the batches.\r\n",
        "  predictions = np.concatenate(predictions, axis=0)\r\n",
        "  true_labels = np.concatenate(true_labels, axis=0)\r\n",
        "\r\n",
        "  # Choose the label with the highest score as our prediction.\r\n",
        "  preds = np.argmax(predictions, axis=1).flatten()\r\n",
        "\r\n",
        "  # Calculate simple flat accuracy -- number correct over total number.\r\n",
        "  accuracy = (preds == true_labels).mean()\r\n",
        "\r\n",
        "  return accuracy"
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uW3YFVk7EdWe"
      },
      "source": [
        "### 8) Function to evaluate model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6ukojsSXEdWe"
      },
      "source": [
        "def eval_model(model,py_inputs, py_attn_masks, py_labels):\r\n",
        "# Prediction on test set\r\n",
        "  t0 = time.time()\r\n",
        "\r\n",
        "  print('Doing validation on {:,} sentences...'.format(len(py_labels)))\r\n",
        "\r\n",
        "  # Put model in evaluation mode\r\n",
        "  model.eval()\r\n",
        "\r\n",
        "  # Tracking variables \r\n",
        "  predictions , true_labels = [], []\r\n",
        "\r\n",
        "  # Choose an interval on which to print progress updates.\r\n",
        "  update_interval = good_update_interval(total_iters=len(py_inputs), num_desired_updates=10)\r\n",
        "\r\n",
        "  # Measure elapsed time.\r\n",
        "  t0 = time.time()\r\n",
        "  total_val_loss = 0\r\n",
        "\r\n",
        "  # Put model in evaluation mode\r\n",
        "  model.eval()\r\n",
        "\r\n",
        "  # For each batch of training data...\r\n",
        "  for step in range(0, len(py_inputs)):\r\n",
        "\r\n",
        "      # Progress update every 100 batches.\r\n",
        "      if step % update_interval == 0 and not step == 0:\r\n",
        "          # Calculate elapsed time in minutes.\r\n",
        "          elapsed = format_time(time.time() - t0)\r\n",
        "          \r\n",
        "          # Calculate the time remaining based on our progress.\r\n",
        "          steps_per_sec = (time.time() - t0) / step\r\n",
        "          remaining_sec = steps_per_sec * (len(py_inputs) - step)\r\n",
        "          remaining = format_time(remaining_sec)\r\n",
        "\r\n",
        "          # Report progress.\r\n",
        "          print('  Batch {:>7,}  of  {:>7,}.    Elapsed: {:}.  Remaining: {:}'.format(step, len(py_inputs), elapsed, remaining))\r\n",
        "\r\n",
        "      # Copy the batch to the GPU.\r\n",
        "      b_input_ids = py_inputs[step].to(device)\r\n",
        "      b_input_mask = py_attn_masks[step].to(device)\r\n",
        "      b_labels = py_labels[step].to(device)\r\n",
        "    \r\n",
        "      # Telling the model not to compute or store gradients, saving memory and \r\n",
        "      # speeding up prediction\r\n",
        "      with torch.no_grad():\r\n",
        "        output = model(b_input_ids, \r\n",
        "                             token_type_ids=None, \r\n",
        "                             attention_mask=b_input_mask, \r\n",
        "                             labels=b_labels)\r\n",
        "\r\n",
        "\r\n",
        "      logits = output.logits \r\n",
        "      loss = output.loss\r\n",
        "      # Move logits and labels to CPU\r\n",
        "      logits = logits.detach().cpu().numpy()\r\n",
        "      label_ids = b_labels.to('cpu').numpy()\r\n",
        "    \r\n",
        "      # Store predictions and true labels\r\n",
        "      predictions.append(logits)\r\n",
        "      true_labels.append(label_ids)\r\n",
        "\r\n",
        "      total_val_loss += loss.item()\r\n",
        "  \r\n",
        "  val_accuracy = check_accuracy(predictions,true_labels)\r\n",
        "\r\n",
        "  validation_time = format_time(time.time() - t0)\r\n",
        "  \r\n",
        "  # Calculate the average loss over all of the batches.\r\n",
        "  avg_val_loss = total_val_loss / len(py_inputs)     \r\n",
        "    \r\n",
        "\r\n",
        "  # print('Accuracy: {:.3f}'.format(val_accuracy))\r\n",
        "  return (avg_val_loss,val_accuracy,validation_time)\r\n"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Rj5Zao7WEdWe"
      },
      "source": [
        "### 9) Function for making predictions on our test dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ChLIow-bEdWf"
      },
      "source": [
        "def get_predictions(py_inputs, py_attn_masks, py_labels):\r\n",
        "\r\n",
        "  print('Predicting labels for {:,} test batches...'.format(len(py_labels)))\r\n",
        "\r\n",
        "  # Put model in evaluation mode\r\n",
        "  model.eval()\r\n",
        "\r\n",
        "  # Tracking variables \r\n",
        "  predictions , true_labels = [], []\r\n",
        "\r\n",
        "  # Choose an interval on which to print progress updates.\r\n",
        "  update_interval = good_update_interval(total_iters=len(py_inputs), num_desired_updates=10)\r\n",
        "\r\n",
        "  # Measure elapsed time.\r\n",
        "  t0 = time.time()\r\n",
        "\r\n",
        "  # Put model in evaluation mode\r\n",
        "  model.eval()\r\n",
        "\r\n",
        "  # For each batch of training data...\r\n",
        "  for step in range(0, len(py_inputs)):\r\n",
        "\r\n",
        "      # Progress update every 100 batches.\r\n",
        "      if step % update_interval == 0 and not step == 0:\r\n",
        "          # Calculate elapsed time in minutes.\r\n",
        "          elapsed = format_time(time.time() - t0)\r\n",
        "          \r\n",
        "          # Calculate the time remaining based on our progress.\r\n",
        "          steps_per_sec = (time.time() - t0) / step\r\n",
        "          remaining_sec = steps_per_sec * (len(py_inputs) - step)\r\n",
        "          remaining = format_time(remaining_sec)\r\n",
        "\r\n",
        "          # Report progress.\r\n",
        "          print('  Batch {:>7,}  of  {:>7,}.    Elapsed: {:}.  Remaining: {:}'.format(step, len(py_inputs), elapsed, remaining))\r\n",
        "\r\n",
        "      # Copy the batch to the GPU.\r\n",
        "      b_input_ids = py_inputs[step].to(device)\r\n",
        "      b_input_mask = py_attn_masks[step].to(device)\r\n",
        "      b_labels = py_labels[step].to(device)\r\n",
        "    \r\n",
        "      # Telling the model not to compute or store gradients, saving memory and \r\n",
        "      # speeding up prediction\r\n",
        "      with torch.no_grad():\r\n",
        "          # Forward pass, calculate logit predictions\r\n",
        "          outputs = model(b_input_ids, b_input_mask)\r\n",
        "\r\n",
        "      logits = outputs.logits \r\n",
        "\r\n",
        "      # Move logits and labels to CPU\r\n",
        "      logits = logits.detach().cpu().numpy()\r\n",
        "      label_ids = b_labels.to('cpu').numpy()\r\n",
        "    \r\n",
        "      # Store predictions and true labels\r\n",
        "      predictions.append(logits)\r\n",
        "      true_labels.append(label_ids)\r\n",
        "      # Combine the results across the batches.\r\n",
        "  predictions = np.concatenate(predictions, axis=0)\r\n",
        "  true_labels = np.concatenate(true_labels, axis=0)\r\n",
        "\r\n",
        "      # Choose the label with the highest score as our prediction.\r\n",
        "  preds = np.argmax(predictions, axis=1).flatten()\r\n",
        "\r\n",
        "  return (preds,true_labels)"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G8-bi_GEEdWf"
      },
      "source": [
        "### 10) Function for making predictions on test dataset(without labels)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tsgswf85EdWf"
      },
      "source": [
        "def get_predictions_test(py_inputs, py_attn_masks,py_ids):\r\n",
        "\r\n",
        "  print('Predicting labels for {:,} test batches...'.format(len(py_inputs)))\r\n",
        "\r\n",
        "  # Put model in evaluation mode\r\n",
        "  model.eval()\r\n",
        "\r\n",
        "  # Tracking variables \r\n",
        "  predictions , true_labels ,ids = [], [] , []\r\n",
        "\r\n",
        "  # Choose an interval on which to print progress updates.\r\n",
        "  update_interval = good_update_interval(total_iters=len(py_inputs), num_desired_updates=10)\r\n",
        "\r\n",
        "  # Measure elapsed time.\r\n",
        "  t0 = time.time()\r\n",
        "\r\n",
        "  # Put model in evaluation mode\r\n",
        "  model.eval()\r\n",
        "\r\n",
        "  # For each batch of training data...\r\n",
        "  for step in range(0, len(py_inputs)):\r\n",
        "\r\n",
        "      # Progress update every 100 batches.\r\n",
        "      if step % update_interval == 0 and not step == 0:\r\n",
        "          # Calculate elapsed time in minutes.\r\n",
        "          elapsed = format_time(time.time() - t0)\r\n",
        "          \r\n",
        "          # Calculate the time remaining based on our progress.\r\n",
        "          steps_per_sec = (time.time() - t0) / step\r\n",
        "          remaining_sec = steps_per_sec * (len(py_inputs) - step)\r\n",
        "          remaining = format_time(remaining_sec)\r\n",
        "\r\n",
        "          # Report progress.\r\n",
        "          print('  Batch {:>7,}  of  {:>7,}.    Elapsed: {:}.  Remaining: {:}'.format(step, len(py_inputs), elapsed, remaining))\r\n",
        "\r\n",
        "      # Copy the batch to the GPU.\r\n",
        "      b_input_ids = py_inputs[step].to(device)\r\n",
        "      b_input_mask = py_attn_masks[step].to(device)\r\n",
        "      b_ids = py_ids[step].to(device)\r\n",
        "    \r\n",
        "      # Telling the model not to compute or store gradients, saving memory and \r\n",
        "      # speeding up prediction\r\n",
        "      with torch.no_grad():\r\n",
        "          # Forward pass, calculate logit predictions\r\n",
        "          outputs = model(b_input_ids, b_input_mask)\r\n",
        "\r\n",
        "      logits = outputs.logits \r\n",
        "\r\n",
        "      # Move logits and labels to CPU\r\n",
        "      logits = logits.detach().cpu().numpy()\r\n",
        "      b_ids = b_ids.detach().cpu().numpy()\r\n",
        "      # Store predictions\r\n",
        "      predictions.append(logits)\r\n",
        "      ids.append(b_ids)\r\n",
        "      # Combine the results across the batches.\r\n",
        "  predictions = np.concatenate(predictions, axis=0)\r\n",
        "  ids = np.concatenate(ids,axis=0)\r\n",
        "  # Choose the label with the highest score as our prediction.\r\n",
        "  preds = np.argmax(predictions, axis=1).flatten()\r\n",
        "\r\n",
        "  return ids,preds"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4Fh-Tb5OEdWf"
      },
      "source": [
        "### 11) Confusion Matrix Helper Function"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XGsBuZ1kEdWg"
      },
      "source": [
        "def show_confusion_matrix(confusion_matrix, class_names):\r\n",
        "\r\n",
        "  cm = confusion_matrix.copy()\r\n",
        "\r\n",
        "  cell_counts = cm.flatten()\r\n",
        "\r\n",
        "  cm_row_norm = cm / cm.sum(axis=1)[:, np.newaxis]\r\n",
        "\r\n",
        "  row_percentages = [\"{0:.2f}\".format(value) for value in cm_row_norm.flatten()]\r\n",
        "\r\n",
        "  cell_labels = [f\"{cnt}\\n{per}\" for cnt, per in zip(cell_counts, row_percentages)]\r\n",
        "  cell_labels = np.asarray(cell_labels).reshape(cm.shape[0], cm.shape[1])\r\n",
        "\r\n",
        "  df_cm = pd.DataFrame(cm_row_norm, index=class_names, columns=class_names)\r\n",
        "\r\n",
        "  hmap = sns.heatmap(df_cm, annot=cell_labels, fmt=\"\", cmap=\"Blues\")\r\n",
        "  hmap.yaxis.set_ticklabels(hmap.yaxis.get_ticklabels(), rotation=0, ha='right')\r\n",
        "  hmap.xaxis.set_ticklabels(hmap.xaxis.get_ticklabels(), rotation=30, ha='right')\r\n",
        "  plt.ylabel('True Sign')\r\n",
        "  plt.xlabel('Predicted Sign');"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MBxHY4rMojd-"
      },
      "source": [
        "## Loading the tokenizer using AutoTokenizer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 180,
          "referenced_widgets": [
            "66d1de2202374f3689ea014824b63491",
            "0c97c9e45c44415dadc3a0a237eb5b5b",
            "0cfe1d1864f3461bbbdea8fe5b0ad3cb",
            "a978e1143be149e3bf19e82fc2f42d7c",
            "1613ae72b0e84490a8168cd9f021c32a",
            "bc446d761a3741308039fc60bdd23262",
            "78197b526eb8410e810979f99f234f6c",
            "86a650edcd104eda9f1b2b899cde525c",
            "5cf4184305ed4116bdbb591bf2eeaaec",
            "d5d26df73ba64708b45b4e3f89229f04",
            "a29329f82f554ae98c6c65008c7aaf34",
            "0e218f795ab34177bc9162b10e5935bc",
            "8242fe40097b4a0f8b2f6208f5777bb7",
            "a524792107a24b73a5bcb97d6d1b0517",
            "ea80c747b0d149ba8cf3debe3d540220",
            "769af2f421724b83abb002068c6edb9b",
            "23ff93eb1da1493087c8adc58c213288",
            "63727149341d47ef81265b5a8f5b15cf",
            "d3dffa30b9f04355a4b0fde121b2eb5b",
            "5be0324b9ce34efca72cfe138499f3da",
            "3489bff83f5c46efaa6c62998aba8faf",
            "c1a1df0491714d5b9efed0e8f3b764a7",
            "0158b8f81fa24d1bba8b41de2af41498",
            "b07dc78c0a51472b918d1fc68c578baf"
          ]
        },
        "id": "fIlwEvcMb4pb",
        "outputId": "7d04c93e-a5f7-44f6-9d1b-766924e64eab"
      },
      "source": [
        "model_name = 'xlm-roberta-base'\r\n",
        "# Load the BERT tokenizer.\r\n",
        "print(f'Loading {model_name} tokenizer...')\r\n",
        "tokenizer = AutoTokenizer.from_pretrained(model_name, do_lower_case=True)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Loading xlm-roberta-base tokenizer...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "66d1de2202374f3689ea014824b63491",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=512.0, 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": "5cf4184305ed4116bdbb591bf2eeaaec",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=5069051.0, style=ProgressStyle(descript…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "23ff93eb1da1493087c8adc58c213288",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=9096718.0, style=ProgressStyle(descript…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OVs3Jb3Vb89i"
      },
      "source": [
        "## Smart Batching of the training data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cWNt4eOLofmF"
      },
      "source": [
        "lengths = []\r\n",
        "for text in train_sentences:\r\n",
        "  lengths.append(len(text))       "
      ],
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 427
        },
        "id": "hbdGRA3Bo02O",
        "outputId": "5bebd93b-880d-4493-ee78-33050f369956"
      },
      "source": [
        "\r\n",
        "# Use plot styling from seaborn.\r\n",
        "sns.set(style='darkgrid')\r\n",
        "\r\n",
        "# Increase the plot size and font size.\r\n",
        "sns.set(font_scale=1.5)\r\n",
        "plt.rcParams[\"figure.figsize\"] = (12,6)\r\n",
        "\r\n",
        "plt.scatter(range(0, len(lengths)), lengths, marker=\"|\")\r\n",
        "\r\n",
        "plt.xlabel('Sample Number')\r\n",
        "plt.ylabel('Sequence Length')\r\n",
        "plt.title('Samples BEFORE Tokenizing')\r\n",
        "\r\n",
        "plt.show()"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9rMhLNhPcZjU"
      },
      "source": [
        "### Tokenizing the sequences"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t7JsLi3EdEje"
      },
      "source": [
        "max_input_length = 400"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J968o8RJcUEd",
        "outputId": "25729806-26cd-4a71-e305-4952c9c53e2a"
      },
      "source": [
        "full_input_ids = []\r\n",
        "labels = []\r\n",
        "\r\n",
        "# Tokenize all training examples\r\n",
        "print('Tokenizing {:,} training samples...'.format(len(train_sentences)))\r\n",
        "\r\n",
        "# Choose an interval on which to print progress updates.\r\n",
        "update_interval = good_update_interval(total_iters=len(train_sentences), num_desired_updates=10)\r\n",
        "\r\n",
        "# For each training example...\r\n",
        "for text in train_sentences:\r\n",
        "    \r\n",
        "    # Report progress.\r\n",
        "    if ((len(full_input_ids) % update_interval) == 0):\r\n",
        "        print('  Tokenized {:,} samples.'.format(len(full_input_ids)))\r\n",
        "\r\n",
        "    # Tokenize the sentence.\r\n",
        "    input_ids = tokenizer.encode(text=text,           \r\n",
        "                                 add_special_tokens=True, \r\n",
        "                                 max_length=max_input_length,  \r\n",
        "                                 truncation=True,     \r\n",
        "                                 padding=False)       \r\n",
        "                                 \r\n",
        "    # Add the tokenized result to our list.\r\n",
        "    full_input_ids.append(input_ids)\r\n",
        "    \r\n",
        "print('DONE.')\r\n",
        "print('{:>10,} samples'.format(len(full_input_ids)))"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tokenizing 8,564 training samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,800 samples.\n",
            "  Tokenized 2,700 samples.\n",
            "  Tokenized 3,600 samples.\n",
            "  Tokenized 4,500 samples.\n",
            "  Tokenized 5,400 samples.\n",
            "  Tokenized 6,300 samples.\n",
            "  Tokenized 7,200 samples.\n",
            "  Tokenized 8,100 samples.\n",
            "DONE.\n",
            "     8,564 samples\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rZzhO2tZeXur"
      },
      "source": [
        "# Get all of the lengths.\r\n",
        "unsorted_lengths = [len(x) for x in full_input_ids]"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 427
        },
        "id": "x7c_mYuVed_H",
        "outputId": "472b9132-3ef5-4650-e485-5b9e34fb23e9"
      },
      "source": [
        "\r\n",
        "# Use plot styling from seaborn.\r\n",
        "sns.set(style='darkgrid')\r\n",
        "\r\n",
        "# Increase the plot size and font size.\r\n",
        "sns.set(font_scale=1.5)\r\n",
        "plt.rcParams[\"figure.figsize\"] = (12,6)\r\n",
        "\r\n",
        "plt.scatter(range(0, len(unsorted_lengths)), unsorted_lengths, marker=\"|\")\r\n",
        "\r\n",
        "plt.xlabel('Sample Number')\r\n",
        "plt.ylabel('Sequence Length')\r\n",
        "plt.title('Samples BEFORE Sorting')\r\n",
        "\r\n",
        "plt.show()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7PaeohUVegmv"
      },
      "source": [
        "# Sort the two lists together by the length of the input sequence.\r\n",
        "train_samples = sorted(zip(full_input_ids, train_labels), key=lambda x: len(x[0]))"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_9nyx-BtepE4",
        "outputId": "19563972-94a7-4546-f2f7-90a6f1021fdb"
      },
      "source": [
        "train_samples[0:5]"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[([0, 18621, 2], 2),\n",
              " ([0, 32506, 204320, 2], 2),\n",
              " ([0, 18621, 26008, 2], 2),\n",
              " ([0, 13177, 8478, 2], 0),\n",
              " ([0, 32506, 204320, 2], 2)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rBxJhEieetFh",
        "outputId": "a5ecb055-fcdf-424d-f20d-1d7d67985c6e"
      },
      "source": [
        "print('Shortest sample:', len(train_samples[0][0]))\r\n",
        "print('Longest sample:', len(train_samples[-1][0]))"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Shortest sample: 3\n",
            "Longest sample: 282\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P44bFOI-e7hW"
      },
      "source": [
        "# Get the new list of lengths after sorting.\r\n",
        "sorted_lengths = [len(s[0]) for s in train_samples]"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 427
        },
        "id": "9P5ZX99le_G1",
        "outputId": "9bfab3db-3657-480a-a8b5-6f222fd59c56"
      },
      "source": [
        "\r\n",
        "plt.plot(range(0, len(sorted_lengths)), sorted_lengths)\r\n",
        "\r\n",
        "plt.xlabel('Sample Number')\r\n",
        "plt.ylabel('Sequence Length')\r\n",
        "plt.title('Samples after Sorting')\r\n",
        "\r\n",
        "plt.show()"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Eh1z0xugfF2t"
      },
      "source": [
        "### Random Batch Selections"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gPrYkFMcgm5H",
        "outputId": "37d84990-ea86-4453-a4b3-a1469343dd4f"
      },
      "source": [
        "train_samples[0]"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "([0, 18621, 2], 2)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "isMxwgezfAvO"
      },
      "source": [
        "batch_size = 16"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "s0tjD8pIfJ0S",
        "outputId": "a03be60b-bd16-46da-f516-b12040631fcb"
      },
      "source": [
        "# List of batches that we'll construct.\r\n",
        "batch_ordered_sentences = []\r\n",
        "batch_ordered_labels = []\r\n",
        "\r\n",
        "print('Creating training batches of size {:}'.format(batch_size))\r\n",
        "\r\n",
        "# Loop over all of the input samples...    \r\n",
        "while len(train_samples) > 0:\r\n",
        "    \r\n",
        "    # Report progress.\r\n",
        "    if ((len(batch_ordered_sentences) % 100) == 0):\r\n",
        "        print('  Selected {:,} batches.'.format(len(batch_ordered_sentences)))\r\n",
        "\r\n",
        "    # `to_take` is our actual batch size. It will be `batch_size` until \r\n",
        "    # we get to the last batch, which may be smaller. \r\n",
        "    to_take = min(batch_size, len(train_samples))\r\n",
        "\r\n",
        "    # Pick a random index in the list of remaining samples to start\r\n",
        "    # our batch at.\r\n",
        "    select = random.randint(0, len(train_samples) - to_take)\r\n",
        "\r\n",
        "    # Select a contiguous batch of samples starting at `select`.\r\n",
        "    batch = train_samples[select:(select + to_take)]\r\n",
        "\r\n",
        "    # Each sample is a tuple--split them apart to create a separate list of \r\n",
        "    # sequences and a list of labels for this batch.\r\n",
        "    batch_ordered_sentences.append([s[0] for s in batch])\r\n",
        "    batch_ordered_labels.append([s[1] for s in batch])\r\n",
        "\r\n",
        "    # Remove these samples from the list.\r\n",
        "    del train_samples[select:select + to_take]\r\n",
        "\r\n",
        "print('\\n  DONE - {:,} batches.'.format(len(batch_ordered_sentences)))"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Creating training batches of size 16\n",
            "  Selected 0 batches.\n",
            "  Selected 100 batches.\n",
            "  Selected 200 batches.\n",
            "  Selected 300 batches.\n",
            "  Selected 400 batches.\n",
            "  Selected 500 batches.\n",
            "\n",
            "  DONE - 536 batches.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rf91Dd3LhBaP",
        "outputId": "d9fa1f74-a8c6-4082-e253-6621c130b57f"
      },
      "source": [
        "print(batch_ordered_sentences[0])"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0, 116, 76438, 2837, 5062, 290, 16715, 18, 7030, 2], [0, 1374, 162, 192, 5568, 11159, 395, 113, 2037, 2], [0, 30607, 115992, 1821, 161740, 13, 4027, 5, 27, 2], [0, 2016, 1162, 190081, 19, 6083, 100258, 22562, 1351, 2], [0, 1374, 176769, 169, 1199, 5701, 14380, 1492, 16797, 2], [0, 140975, 39, 12905, 5997, 13177, 1146, 18621, 2692, 2], [0, 10071, 294, 408, 13857, 1468, 14195, 3731, 18, 2], [0, 6, 129483, 202826, 120907, 28331, 123752, 2117, 211124, 2], [0, 189093, 6, 12559, 130978, 2117, 1022, 23894, 30476, 2], [0, 79731, 2381, 145061, 229364, 11483, 18002, 235196, 574, 2], [0, 150848, 45875, 2569, 6190, 17530, 28526, 238617, 1351, 2], [0, 1374, 203334, 12927, 606, 39, 10, 85751, 2439, 2], [0, 105269, 177, 36, 47311, 30987, 331, 2340, 76457, 2], [0, 18356, 65073, 198743, 25372, 68071, 10020, 23833, 8126, 2], [0, 4324, 6190, 6368, 6, 106536, 21944, 224701, 32, 2], [0, 18779, 183495, 142154, 51577, 1351, 15626, 25634, 1626, 2]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qmkQK_gVgiTu",
        "outputId": "0e5609b7-21d1-437c-c805-ab0333370f47"
      },
      "source": [
        "batch_ordered_labels[0]"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1, 2, 2, 1, 1, 0, 2, 1, 1, 0, 1, 1, 1, 1, 1, 1]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xZ0TFA8Mfn6m"
      },
      "source": [
        "### Padding"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c9rdswcpfTip"
      },
      "source": [
        "py_inputs = []\r\n",
        "py_attn_masks = []\r\n",
        "py_labels = []\r\n",
        "\r\n",
        "# For each batch...\r\n",
        "for (batch_inputs, batch_labels) in zip(batch_ordered_sentences, batch_ordered_labels):\r\n",
        "\r\n",
        "    # New version of the batch, this time with padded sequences and now with\r\n",
        "    # attention masks defined.\r\n",
        "    batch_padded_inputs = []\r\n",
        "    batch_attn_masks = []\r\n",
        "    \r\n",
        "    # First, find the longest sample in the batch. \r\n",
        "    # Note that the sequences do currently include the special tokens!\r\n",
        "    max_size = max([len(sen) for sen in batch_inputs])\r\n",
        "\r\n",
        "    #print('Max size:', max_size)\r\n",
        "\r\n",
        "    # For each input in this batch...\r\n",
        "    for sen in batch_inputs:\r\n",
        "        \r\n",
        "        # How many pad tokens do we need to add?\r\n",
        "        num_pads = max_size - len(sen)\r\n",
        "\r\n",
        "        # Add `num_pads` padding tokens to the end of the sequence.\r\n",
        "        padded_input = sen + [tokenizer.pad_token_id]*num_pads\r\n",
        "\r\n",
        "        # Define the attention mask--it's just a `1` for every real token\r\n",
        "        # and a `0` for every padding token.\r\n",
        "        attn_mask = [1] * len(sen) + [0] * num_pads\r\n",
        "\r\n",
        "        # Add the padded results to the batch.\r\n",
        "        batch_padded_inputs.append(padded_input)\r\n",
        "        batch_attn_masks.append(attn_mask)\r\n",
        "\r\n",
        "    # Our batch has been padded, so we need to save this updated batch.\r\n",
        "    # We also need the inputs to be PyTorch tensors, so we'll do that here.\r\n",
        "    py_inputs.append(torch.tensor(batch_padded_inputs))\r\n",
        "    py_attn_masks.append(torch.tensor(batch_attn_masks))\r\n",
        "    py_labels.append(torch.tensor(batch_labels))\r\n"
      ],
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "roS5lLWdX-5J"
      },
      "source": [
        "### Check the number of token reductions because of smart batching"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wlh7cteVfroV",
        "outputId": "5f96426a-5625-42d3-d306-4c9df8af68c7"
      },
      "source": [
        "# Get the new list of lengths after sorting.\r\n",
        "\r\n",
        "padded_lengths = []\r\n",
        "\r\n",
        "# For each batch...\r\n",
        "for batch in py_inputs:\r\n",
        "    \r\n",
        "    # For each sample...\r\n",
        "    for s in batch:\r\n",
        "    \r\n",
        "        # Record its length.\r\n",
        "        padded_lengths.append(len(s))\r\n",
        "\r\n",
        "# Sum up the lengths to the get the total number of tokens after smart batching.\r\n",
        "smart_token_count = np.sum(padded_lengths)\r\n",
        "\r\n",
        "# To get the total number of tokens in the dataset using fixed padding, it's\r\n",
        "# as simple as the number of samples times our `max_len` parameter (that we\r\n",
        "# would pad everything to).\r\n",
        "fixed_token_count = len(train_sentences) * max_input_length\r\n",
        "\r\n",
        "# Calculate the percentage reduction.\r\n",
        "prcnt_reduced = (fixed_token_count - smart_token_count) / float(fixed_token_count) \r\n",
        "\r\n",
        "print('Total tokens:')\r\n",
        "print('   Fixed Padding: {:,}'.format(fixed_token_count))\r\n",
        "print('  Smart Batching: {:,}  ({:.1%} less)'.format(smart_token_count, prcnt_reduced))\r\n"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total tokens:\n",
            "   Fixed Padding: 3,425,600\n",
            "  Smart Batching: 232,560  (93.2% less)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8wgdhsI-YIAL"
      },
      "source": [
        "## Load the model configuration from the transformers library using AutoConfig"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9er7EHvdfudf",
        "outputId": "e3787915-493b-4f92-e77b-1c7cfb3df06f"
      },
      "source": [
        "# Load the Config object, with an output configured for classification.\r\n",
        "config = AutoConfig.from_pretrained(pretrained_model_name_or_path=model_name,\r\n",
        "                                    num_labels=3)\r\n",
        "\r\n",
        "print('Config type:', str(type(config)), '\\n')\r\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Config type: <class 'transformers.models.xlm_roberta.configuration_xlm_roberta.XLMRobertaConfig'> \n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170,
          "referenced_widgets": [
            "8a5542f3d67544288e449e8172cd053f",
            "5280b85c17a84b2d82d9e2a496c3de29",
            "5d207134064e43369c37518317e348b7",
            "5792f249efdd4bb4972e90c705a202f9",
            "4270e52728fb4fe19fdf50664c860b9c",
            "b19b3207f9b04536a0e5fef26773492f",
            "30afddfaed42443a95b0224a48f314cb",
            "668597d7a9df4ae8b8c98b7fd241ce72"
          ]
        },
        "id": "bWvc9Ikv4HtD",
        "outputId": "ef334f56-5d71-4ede-833f-e66fbd14b387"
      },
      "source": [
        "model = AutoModelForSequenceClassification.from_pretrained(pretrained_model_name_or_path=model_name,config = config)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "8a5542f3d67544288e449e8172cd053f",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=1115590446.0, style=ProgressStyle(descr…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Some weights of the model checkpoint at xlm-roberta-base were not used when initializing XLMRobertaForSequenceClassification: ['lm_head.bias', 'lm_head.dense.weight', 'lm_head.dense.bias', 'lm_head.layer_norm.weight', 'lm_head.layer_norm.bias', 'lm_head.decoder.weight', 'roberta.pooler.dense.weight', 'roberta.pooler.dense.bias']\n",
            "- This IS expected if you are initializing XLMRobertaForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
            "- This IS NOT expected if you are initializing XLMRobertaForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
            "Some weights of XLMRobertaForSequenceClassification were not initialized from the model checkpoint at xlm-roberta-base and are newly initialized: ['classifier.dense.weight', 'classifier.dense.bias', 'classifier.out_proj.weight', 'classifier.out_proj.bias']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xouf5M2B55ei",
        "outputId": "9291cc1d-701b-4696-8799-c72e77b50377"
      },
      "source": [
        "model.cuda()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XLMRobertaForSequenceClassification(\n",
              "  (roberta): RobertaModel(\n",
              "    (embeddings): RobertaEmbeddings(\n",
              "      (word_embeddings): Embedding(250002, 768, padding_idx=1)\n",
              "      (position_embeddings): Embedding(514, 768, padding_idx=1)\n",
              "      (token_type_embeddings): Embedding(1, 768)\n",
              "      (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "      (dropout): Dropout(p=0.1, inplace=False)\n",
              "    )\n",
              "    (encoder): RobertaEncoder(\n",
              "      (layer): ModuleList(\n",
              "        (0): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (1): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (2): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (3): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (4): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (5): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (6): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (7): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (8): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (9): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (10): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (11): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "    )\n",
              "  )\n",
              "  (classifier): RobertaClassificationHead(\n",
              "    (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "    (dropout): Dropout(p=0.1, inplace=False)\n",
              "    (out_proj): Linear(in_features=768, out_features=3, bias=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SXJHBVlxYR-l"
      },
      "source": [
        "## Load the model from the transformers library using AutoModel"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n8J4r6KyhveR"
      },
      "source": [
        "# # Load the pre-trained model for classification, passing in the `config` from\r\n",
        "# # above.\r\n",
        "# bert = AutoModel.from_pretrained(\r\n",
        "#     pretrained_model_name_or_path=model_name,\r\n",
        "#     config=config)\r\n",
        "\r\n",
        "# print('\\nModel type:', str(type(bert)))\r\n"
      ],
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RcFMgf5EuPAF"
      },
      "source": [
        "# # freeze all the parameters\r\n",
        "# for param in bert.parameters():\r\n",
        "#     param.requires_grad = False"
      ],
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NS_-0rBDYeY5"
      },
      "source": [
        "## Creating a custom BERT model "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bmGtZwykJ39i"
      },
      "source": [
        "\r\n",
        "# class BERT_Model(nn.Module):\r\n",
        "#     ## defining the constructor for the class\r\n",
        "#     def __init__(self, bert,num_labels):\r\n",
        "#       ## calling the super class constructor\r\n",
        "#       super(BERT_Model, self).__init__()\r\n",
        "#       ## adding the bert layer to the model\r\n",
        "#       self.bert = bert        \r\n",
        "#       # relu activation function\r\n",
        "#       self.relu =  nn.ReLU()\r\n",
        "#       # adding a dense layer to our custom model\r\n",
        "#       self.fc1 = nn.Linear(768,512)\r\n",
        "#       # adding another dense layer to our custom model ,i.e., the Output layer\r\n",
        "#       self.fc2 = nn.Linear(512,num_labels)\r\n",
        "#       # adding a softmax  activation function for our custom model's output layer\r\n",
        "#       self.softmax = nn.LogSoftmax(dim=1)\r\n",
        "\r\n",
        "#     #defining the forward pass\r\n",
        "#     def forward(self, input_id, mask):\r\n",
        "#       #pass the inputs to the model  \r\n",
        "#       outputs = self.bert(input_id,mask)\r\n",
        "  \r\n",
        "#       last_hidden_state = outputs.last_hidden_state       ## last hidden state from the model\r\n",
        "#       pooler_output = outputs.pooler_output               ## pooler output from the model\r\n",
        "#       ## adding a fully connected layer to the BERT model\r\n",
        "#       x = self.fc1(pooler_output)\r\n",
        "#       ## applying relu activation function\r\n",
        "#       x = self.relu(x)\r\n",
        "#       # the final output layer\r\n",
        "#       x = self.fc2(x)\r\n",
        "#       # apply softmax activation to our output layer\r\n",
        "#       x = self.softmax(x)\r\n",
        "\r\n",
        "#       return x"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jvOs0G5sh02S"
      },
      "source": [
        "\r\n",
        "\r\n",
        "# print('\\nLoading model ...')\r\n",
        "# # pass the pre-trained BERT to our define architecture\r\n",
        "# model = BERT_Model(bert,num_labels=3)\r\n",
        "\r\n",
        "# model.cuda()"
      ],
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LjQSUqCWJTbI"
      },
      "source": [
        "## Custom Loss function"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iOHVl1RRJSUB"
      },
      "source": [
        "# # convert class weights to tensor\r\n",
        "# weights= torch.tensor(class_wts,dtype=torch.float)\r\n",
        "# weights = weights.to(device)\r\n",
        "\r\n",
        "# # loss function\r\n",
        "# cross_entropy  = nn.NLLLoss(weight=weights) \r\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "evYV0yrQ2u_b"
      },
      "source": [
        "## Loading Optimizer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8tAijXBEh7t8"
      },
      "source": [
        "# Note: AdamW is a class from the huggingface library (as opposed to pytorch) \r\n",
        "# I believe the 'W' stands for 'Weight Decay fix\"\r\n",
        "optimizer = AdamW(model.parameters(),\r\n",
        "                  lr = 2e-5, # This is the value Michael used.\r\n",
        "                  eps = 1e-8 # args.adam_epsilon  - default is 1e-8.\r\n",
        "                )\r\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "s9n3_5jI2ywo"
      },
      "source": [
        "## Loading lr scheduler"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "agidP6D6iBFI"
      },
      "source": [
        "# Number of training epochs. I chose to train for 1 simply because the training\r\n",
        "# time is long. More epochs may improve the model's accuracy.\r\n",
        "epochs = 4\r\n",
        "\r\n",
        "# Total number of training steps is [number of batches] x [number of epochs]. \r\n",
        "# Note that it's the number of *batches*, not *samples*!\r\n",
        "total_steps = len(py_inputs) * epochs\r\n",
        "\r\n",
        "# Create the learning rate scheduler.\r\n",
        "scheduler = get_linear_schedule_with_warmup(optimizer, \r\n",
        "                                            num_warmup_steps = 0, # Default value in run_glue.py\r\n",
        "                                            num_training_steps = total_steps)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MDDp05Cm245e"
      },
      "source": [
        "## Training Loop"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H4ERR8fBiEJn",
        "outputId": "f3ae7619-0b4c-44c5-c4db-c52c80aae7af"
      },
      "source": [
        "# We'll store a number of quantities such as training and validation loss, \r\n",
        "# validation accuracy, and timings.\r\n",
        "training_stats = {\r\n",
        "    'epoch':[],\r\n",
        "    'train_loss':[],\r\n",
        "    'Training Time':[],\r\n",
        "    'val_loss':[],\r\n",
        "    'Validation Time':[],\r\n",
        "    'train_acc':[],\r\n",
        "    'val_acc':[]\r\n",
        "}\r\n",
        "\r\n",
        "# Update every `update_interval` batches.\r\n",
        "update_interval = good_update_interval(total_iters=len(py_inputs), num_desired_updates=10)\r\n",
        "\r\n",
        "# Measure the total training time for the whole run.\r\n",
        "total_t0 = time.time()\r\n",
        "\r\n",
        "# For each epoch...\r\n",
        "for epoch_i in range(0, epochs):\r\n",
        "\r\n",
        "    predictions = []\r\n",
        "    true_labels = []\r\n",
        "    \r\n",
        "    # ========================================\r\n",
        "    #               Training\r\n",
        "    # ========================================\r\n",
        "    \r\n",
        "    # Perform one full pass over the training set.\r\n",
        "\r\n",
        "    print(\"\")\r\n",
        "    print('======== Epoch {:} / {:} ========'.format(epoch_i + 1, epochs))\r\n",
        "    \r\n",
        "    # At the start of each epoch (except for the first) we need to re-randomize\r\n",
        "    # our training data.\r\n",
        "    if epoch_i > 0:\r\n",
        "        # Use our `make_smart_batches` function (from 6.1.) to re-shuffle the \r\n",
        "        # dataset into new batches.\r\n",
        "        (py_inputs, py_attn_masks, py_labels) = make_smart_batches(train_sentences, train_labels, batch_size,tokenizer,max_input_length)\r\n",
        "    \r\n",
        "    print('Training on {:,} batches...'.format(len(py_inputs)))\r\n",
        "\r\n",
        "    # Measure how long the training epoch takes.\r\n",
        "    t0 = time.time()\r\n",
        "\r\n",
        "    # Reset the total loss for this epoch.\r\n",
        "    total_train_loss = 0\r\n",
        "\r\n",
        "    # Put the model into training mode. Don't be mislead--the call to \r\n",
        "    # `train` just changes the *mode*, it doesn't *perform* the training.\r\n",
        "    # `dropout` and `batchnorm` layers behave differently during training\r\n",
        "    # vs. test (source: https://stackoverflow.com/questions/51433378/what-does-model-train-do-in-pytorch)\r\n",
        "    model.train()\r\n",
        "\r\n",
        "    # For each batch of training data...\r\n",
        "    for step in range(0, len(py_inputs)):\r\n",
        "\r\n",
        "        # Progress update every, e.g., 100 batches.\r\n",
        "        if step % update_interval == 0 and not step == 0:\r\n",
        "            # Calculate elapsed time in minutes.\r\n",
        "            elapsed = format_time(time.time() - t0)\r\n",
        "            \r\n",
        "            # Calculate the time remaining based on our progress.\r\n",
        "            steps_per_sec = (time.time() - t0) / step\r\n",
        "            remaining_sec = steps_per_sec * (len(py_inputs) - step)\r\n",
        "            remaining = format_time(remaining_sec)\r\n",
        "\r\n",
        "            # Report progress.\r\n",
        "            print('  Batch {:>7,}  of  {:>7,}.    Elapsed: {:}.  Remaining: {:}'.format(step, len(py_inputs), elapsed, remaining))\r\n",
        "\r\n",
        "        # Copy the current training batch to the GPU using the `to` method.\r\n",
        "        b_input_ids = py_inputs[step].to(device)\r\n",
        "        b_input_mask = py_attn_masks[step].to(device)\r\n",
        "        b_labels = py_labels[step].to(device)\r\n",
        "\r\n",
        "        # Always clear any previously calculated gradients before performing a\r\n",
        "        # backward pass.\r\n",
        "        model.zero_grad()        \r\n",
        "\r\n",
        "        # Perform a forward pass (evaluate the model on this training batch).\r\n",
        "        # The call returns the loss (because we provided labels) and the \r\n",
        "        # \"logits\"--the model outputs prior to activation.\r\n",
        "        output = model(b_input_ids, \r\n",
        "                             token_type_ids=None, \r\n",
        "                             attention_mask=b_input_mask, \r\n",
        "                             labels=b_labels)\r\n",
        "\r\n",
        "        # Accumulate the training loss over all of the batches so that we can\r\n",
        "        # calculate the average loss at the end. `loss` is a Tensor containing a\r\n",
        "        # single value; the `.item()` function just returns the Python value \r\n",
        "        # from the tensor.\r\n",
        "        loss = output.loss \r\n",
        "        logits = output.logits\r\n",
        "\r\n",
        "        # Move logits and labels to CPU\r\n",
        "        logits = logits.detach().cpu().numpy()\r\n",
        "        label_ids = b_labels.to('cpu').numpy()\r\n",
        "      \r\n",
        "        # Store predictions and true labels\r\n",
        "        predictions.append(logits)\r\n",
        "        true_labels.append(label_ids)\r\n",
        "        \r\n",
        "        # Accumulate the training loss over all of the batches so that we can\r\n",
        "        # calculate the average loss at the end. `loss` is a Tensor containing a\r\n",
        "        # single value; the `.item()` function just returns the Python value \r\n",
        "        # from the tensor.\r\n",
        "        total_train_loss += loss.item()\r\n",
        "\r\n",
        "        # Perform a backward pass to calculate the gradients.\r\n",
        "        loss.backward()\r\n",
        "\r\n",
        "        # Clip the norm of the gradients to 1.0.\r\n",
        "        # This is to help prevent the \"exploding gradients\" problem.\r\n",
        "        torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)\r\n",
        "\r\n",
        "        # Update parameters and take a step using the computed gradient.\r\n",
        "        # The optimizer dictates the \"update rule\"--how the parameters are\r\n",
        "        # modified based on their gradients, the learning rate, etc.\r\n",
        "        optimizer.step()\r\n",
        "\r\n",
        "        # Update the learning rate.\r\n",
        "        scheduler.step()\r\n",
        "\r\n",
        "    # Calculate the average loss over all of the batches.\r\n",
        "    avg_train_loss = total_train_loss / len(py_inputs)     \r\n",
        "       \r\n",
        "    training_accuracy = check_accuracy(predictions,true_labels)       \r\n",
        "    \r\n",
        "    # Measure how long this epoch took.\r\n",
        "    training_time = format_time(time.time() - t0)\r\n",
        "\r\n",
        "    print(\"\")\r\n",
        "    print(\"  Average training loss: {0:.2f}\".format(avg_train_loss))\r\n",
        "    print(\"  Training Accuracy: {0:.2f}\".format(training_accuracy))\r\n",
        "    print(\"  Training epoch took: {:}\".format(training_time))\r\n",
        "        \r\n",
        "    (py_inputs, py_attn_masks, py_labels) = make_smart_batches(val_sentences, val_labels, batch_size ,tokenizer,max_input_length)\r\n",
        "    val_loss,val_accuracy,validation_time = eval_model(model,py_inputs, py_attn_masks, py_labels)  \r\n",
        "    # Record all statistics from this epoch.\r\n",
        "\r\n",
        "    print(\"\")\r\n",
        "    print(\"  Average validation loss: {0:.2f}\".format(val_loss))\r\n",
        "    print(\"  Validation Accuracy: {0:.2f}\".format(val_accuracy))\r\n",
        "    print(\"  Validation epoch took: {:}\".format(validation_time))\r\n",
        "    \r\n",
        "    \r\n",
        "    training_stats['epoch'].append(epoch_i + 1)\r\n",
        "    training_stats['train_loss'].append(avg_train_loss)\r\n",
        "    training_stats['Training Time'].append(training_time)\r\n",
        "    training_stats['val_loss'].append(val_loss)\r\n",
        "    training_stats['Validation Time'].append(validation_time)\r\n",
        "    training_stats['train_acc'].append(training_accuracy)\r\n",
        "    training_stats['val_acc'].append(val_accuracy)\r\n",
        "\r\n",
        "print(\"\")\r\n",
        "print(\"Training complete!\")\r\n",
        "\r\n",
        "print(\"Total training took {:} (h:mm:ss)\".format(format_time(time.time()-total_t0)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "======== Epoch 1 / 4 ========\n",
            "Training on 536 batches...\n",
            "  Batch      50  of      536.    Elapsed: 0:00:06.  Remaining: 0:00:55\n",
            "  Batch     100  of      536.    Elapsed: 0:00:11.  Remaining: 0:00:49\n",
            "  Batch     150  of      536.    Elapsed: 0:00:17.  Remaining: 0:00:43\n",
            "  Batch     200  of      536.    Elapsed: 0:00:23.  Remaining: 0:00:38\n",
            "  Batch     250  of      536.    Elapsed: 0:00:28.  Remaining: 0:00:32\n",
            "  Batch     300  of      536.    Elapsed: 0:00:34.  Remaining: 0:00:27\n",
            "  Batch     350  of      536.    Elapsed: 0:00:40.  Remaining: 0:00:21\n",
            "  Batch     400  of      536.    Elapsed: 0:00:45.  Remaining: 0:00:15\n",
            "  Batch     450  of      536.    Elapsed: 0:00:51.  Remaining: 0:00:10\n",
            "  Batch     500  of      536.    Elapsed: 0:00:56.  Remaining: 0:00:04\n",
            "\n",
            "  Average training loss: 0.60\n",
            "  Training Accuracy: 0.76\n",
            "  Training epoch took: 0:01:01\n",
            "Creating Smart Batches from 1,070 examples with batch size 16...\n",
            "\n",
            "Tokenizing 1,070 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 100 samples.\n",
            "  Tokenized 200 samples.\n",
            "  Tokenized 300 samples.\n",
            "  Tokenized 400 samples.\n",
            "  Tokenized 500 samples.\n",
            "  Tokenized 600 samples.\n",
            "  Tokenized 700 samples.\n",
            "  Tokenized 800 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,000 samples.\n",
            "DONE.\n",
            "     1,070 samples\n",
            "\n",
            "     1,070 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 67 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Doing validation on 67 sentences...\n",
            "  Batch       7  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      14  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      21  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      28  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      35  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      42  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      49  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      56  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      63  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "\n",
            "  Average validation loss: 0.48\n",
            "  Validation Accuracy: 0.81\n",
            "  Validation epoch took: 0:00:01\n",
            "\n",
            "======== Epoch 2 / 4 ========\n",
            "Creating Smart Batches from 8,564 examples with batch size 16...\n",
            "\n",
            "Tokenizing 8,564 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,800 samples.\n",
            "  Tokenized 2,700 samples.\n",
            "  Tokenized 3,600 samples.\n",
            "  Tokenized 4,500 samples.\n",
            "  Tokenized 5,400 samples.\n",
            "  Tokenized 6,300 samples.\n",
            "  Tokenized 7,200 samples.\n",
            "  Tokenized 8,100 samples.\n",
            "DONE.\n",
            "     8,564 samples\n",
            "\n",
            "     8,564 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 536 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Training on 536 batches...\n",
            "  Batch      50  of      536.    Elapsed: 0:00:06.  Remaining: 0:00:55\n",
            "  Batch     100  of      536.    Elapsed: 0:00:11.  Remaining: 0:00:50\n",
            "  Batch     150  of      536.    Elapsed: 0:00:17.  Remaining: 0:00:44\n",
            "  Batch     200  of      536.    Elapsed: 0:00:23.  Remaining: 0:00:38\n",
            "  Batch     250  of      536.    Elapsed: 0:00:28.  Remaining: 0:00:32\n",
            "  Batch     300  of      536.    Elapsed: 0:00:34.  Remaining: 0:00:27\n",
            "  Batch     350  of      536.    Elapsed: 0:00:39.  Remaining: 0:00:21\n",
            "  Batch     400  of      536.    Elapsed: 0:00:44.  Remaining: 0:00:15\n",
            "  Batch     450  of      536.    Elapsed: 0:00:50.  Remaining: 0:00:10\n",
            "  Batch     500  of      536.    Elapsed: 0:00:56.  Remaining: 0:00:04\n",
            "\n",
            "  Average training loss: 0.41\n",
            "  Training Accuracy: 0.83\n",
            "  Training epoch took: 0:01:01\n",
            "Creating Smart Batches from 1,070 examples with batch size 16...\n",
            "\n",
            "Tokenizing 1,070 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 100 samples.\n",
            "  Tokenized 200 samples.\n",
            "  Tokenized 300 samples.\n",
            "  Tokenized 400 samples.\n",
            "  Tokenized 500 samples.\n",
            "  Tokenized 600 samples.\n",
            "  Tokenized 700 samples.\n",
            "  Tokenized 800 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,000 samples.\n",
            "DONE.\n",
            "     1,070 samples\n",
            "\n",
            "     1,070 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 67 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Doing validation on 67 sentences...\n",
            "  Batch       7  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      14  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      21  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      28  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      35  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      42  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      49  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      56  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      63  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "\n",
            "  Average validation loss: 0.43\n",
            "  Validation Accuracy: 0.81\n",
            "  Validation epoch took: 0:00:01\n",
            "\n",
            "======== Epoch 3 / 4 ========\n",
            "Creating Smart Batches from 8,564 examples with batch size 16...\n",
            "\n",
            "Tokenizing 8,564 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,800 samples.\n",
            "  Tokenized 2,700 samples.\n",
            "  Tokenized 3,600 samples.\n",
            "  Tokenized 4,500 samples.\n",
            "  Tokenized 5,400 samples.\n",
            "  Tokenized 6,300 samples.\n",
            "  Tokenized 7,200 samples.\n",
            "  Tokenized 8,100 samples.\n",
            "DONE.\n",
            "     8,564 samples\n",
            "\n",
            "     8,564 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 536 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Training on 536 batches...\n",
            "  Batch      50  of      536.    Elapsed: 0:00:05.  Remaining: 0:00:51\n",
            "  Batch     100  of      536.    Elapsed: 0:00:11.  Remaining: 0:00:48\n",
            "  Batch     150  of      536.    Elapsed: 0:00:17.  Remaining: 0:00:43\n",
            "  Batch     200  of      536.    Elapsed: 0:00:22.  Remaining: 0:00:38\n",
            "  Batch     250  of      536.    Elapsed: 0:00:28.  Remaining: 0:00:32\n",
            "  Batch     300  of      536.    Elapsed: 0:00:34.  Remaining: 0:00:26\n",
            "  Batch     350  of      536.    Elapsed: 0:00:39.  Remaining: 0:00:21\n",
            "  Batch     400  of      536.    Elapsed: 0:00:44.  Remaining: 0:00:15\n",
            "  Batch     450  of      536.    Elapsed: 0:00:50.  Remaining: 0:00:10\n",
            "  Batch     500  of      536.    Elapsed: 0:00:56.  Remaining: 0:00:04\n",
            "\n",
            "  Average training loss: 0.33\n",
            "  Training Accuracy: 0.87\n",
            "  Training epoch took: 0:01:00\n",
            "Creating Smart Batches from 1,070 examples with batch size 16...\n",
            "\n",
            "Tokenizing 1,070 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 100 samples.\n",
            "  Tokenized 200 samples.\n",
            "  Tokenized 300 samples.\n",
            "  Tokenized 400 samples.\n",
            "  Tokenized 500 samples.\n",
            "  Tokenized 600 samples.\n",
            "  Tokenized 700 samples.\n",
            "  Tokenized 800 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,000 samples.\n",
            "DONE.\n",
            "     1,070 samples\n",
            "\n",
            "     1,070 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 67 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Doing validation on 67 sentences...\n",
            "  Batch       7  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      14  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      21  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      28  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      35  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      42  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      49  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      56  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      63  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "\n",
            "  Average validation loss: 0.44\n",
            "  Validation Accuracy: 0.83\n",
            "  Validation epoch took: 0:00:01\n",
            "\n",
            "======== Epoch 4 / 4 ========\n",
            "Creating Smart Batches from 8,564 examples with batch size 16...\n",
            "\n",
            "Tokenizing 8,564 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,800 samples.\n",
            "  Tokenized 2,700 samples.\n",
            "  Tokenized 3,600 samples.\n",
            "  Tokenized 4,500 samples.\n",
            "  Tokenized 5,400 samples.\n",
            "  Tokenized 6,300 samples.\n",
            "  Tokenized 7,200 samples.\n",
            "  Tokenized 8,100 samples.\n",
            "DONE.\n",
            "     8,564 samples\n",
            "\n",
            "     8,564 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 536 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Training on 536 batches...\n",
            "  Batch      50  of      536.    Elapsed: 0:00:06.  Remaining: 0:00:55\n",
            "  Batch     100  of      536.    Elapsed: 0:00:11.  Remaining: 0:00:50\n",
            "  Batch     150  of      536.    Elapsed: 0:00:17.  Remaining: 0:00:44\n",
            "  Batch     200  of      536.    Elapsed: 0:00:22.  Remaining: 0:00:37\n",
            "  Batch     250  of      536.    Elapsed: 0:00:28.  Remaining: 0:00:32\n",
            "  Batch     300  of      536.    Elapsed: 0:00:34.  Remaining: 0:00:27\n",
            "  Batch     350  of      536.    Elapsed: 0:00:39.  Remaining: 0:00:21\n",
            "  Batch     400  of      536.    Elapsed: 0:00:45.  Remaining: 0:00:15\n",
            "  Batch     450  of      536.    Elapsed: 0:00:51.  Remaining: 0:00:10\n",
            "  Batch     500  of      536.    Elapsed: 0:00:56.  Remaining: 0:00:04\n",
            "\n",
            "  Average training loss: 0.28\n",
            "  Training Accuracy: 0.90\n",
            "  Training epoch took: 0:01:00\n",
            "Creating Smart Batches from 1,070 examples with batch size 16...\n",
            "\n",
            "Tokenizing 1,070 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 100 samples.\n",
            "  Tokenized 200 samples.\n",
            "  Tokenized 300 samples.\n",
            "  Tokenized 400 samples.\n",
            "  Tokenized 500 samples.\n",
            "  Tokenized 600 samples.\n",
            "  Tokenized 700 samples.\n",
            "  Tokenized 800 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,000 samples.\n",
            "DONE.\n",
            "     1,070 samples\n",
            "\n",
            "     1,070 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 67 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Doing validation on 67 sentences...\n",
            "  Batch       7  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      14  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      21  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      28  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      35  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      42  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      49  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      56  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      63  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "\n",
            "  Average validation loss: 0.47\n",
            "  Validation Accuracy: 0.83\n",
            "  Validation epoch took: 0:00:01\n",
            "\n",
            "Training complete!\n",
            "Total training took 0:04:12 (h:mm:ss)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "id": "GOmua_S1eXNM",
        "outputId": "2401ca40-918d-4f85-cfb5-344b0b783a3f"
      },
      "source": [
        "plot_training_history(training_stats)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1296x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3c-6bOIxVyrM"
      },
      "source": [
        "## Evaluating Performance Over Training Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2QmKPNAeWHkZ",
        "outputId": "3755e58c-477a-46d4-f209-ee9b6e3ede57"
      },
      "source": [
        "(py_inputs, py_attn_masks, py_labels) = make_smart_batches(train_sentences, train_labels, batch_size ,tokenizer,max_input_length)\r\n",
        "\r\n",
        "y_pred , y_true = get_predictions(py_inputs, py_attn_masks, py_labels)\r\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Creating Smart Batches from 8,564 examples with batch size 16...\n",
            "\n",
            "Tokenizing 8,564 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,800 samples.\n",
            "  Tokenized 2,700 samples.\n",
            "  Tokenized 3,600 samples.\n",
            "  Tokenized 4,500 samples.\n",
            "  Tokenized 5,400 samples.\n",
            "  Tokenized 6,300 samples.\n",
            "  Tokenized 7,200 samples.\n",
            "  Tokenized 8,100 samples.\n",
            "DONE.\n",
            "     8,564 samples\n",
            "\n",
            "     8,564 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 536 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Predicting labels for 536 test batches...\n",
            "  Batch      50  of      536.    Elapsed: 0:00:01.  Remaining: 0:00:09\n",
            "  Batch     100  of      536.    Elapsed: 0:00:02.  Remaining: 0:00:08\n",
            "  Batch     150  of      536.    Elapsed: 0:00:03.  Remaining: 0:00:07\n",
            "  Batch     200  of      536.    Elapsed: 0:00:04.  Remaining: 0:00:06\n",
            "  Batch     250  of      536.    Elapsed: 0:00:05.  Remaining: 0:00:05\n",
            "  Batch     300  of      536.    Elapsed: 0:00:05.  Remaining: 0:00:04\n",
            "  Batch     350  of      536.    Elapsed: 0:00:06.  Remaining: 0:00:03\n",
            "  Batch     400  of      536.    Elapsed: 0:00:07.  Remaining: 0:00:02\n",
            "  Batch     450  of      536.    Elapsed: 0:00:08.  Remaining: 0:00:02\n",
            "  Batch     500  of      536.    Elapsed: 0:00:09.  Remaining: 0:00:01\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uXsVn0wngZhv",
        "outputId": "64a382dd-1159-4277-9c37-5cf1a9b4c0e4"
      },
      "source": [
        "print(classification_report(y_true, y_pred, target_names=class_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                 precision    recall  f1-score   support\n",
            "\n",
            "    Hope_speech       0.81      0.85      0.83      1668\n",
            "Non-hope-speech       0.96      0.93      0.95      6205\n",
            "  not-malayalam       0.87      0.95      0.91       691\n",
            "\n",
            "       accuracy                           0.92      8564\n",
            "      macro avg       0.88      0.91      0.89      8564\n",
            "   weighted avg       0.92      0.92      0.92      8564\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "id": "guClUdRhYPql",
        "outputId": "8c688bbd-f8a8-4ade-d3b1-3fcf684adf5d"
      },
      "source": [
        "cm = confusion_matrix(y_true, y_pred)\r\n",
        "show_confusion_matrix(cm, class_names)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VM2N-X1tV3il"
      },
      "source": [
        "## Evaluating Performance Over Validation Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BnadNvtxWIYw",
        "outputId": "9f825dbc-c83c-4d38-ed43-1df3611fa02b"
      },
      "source": [
        "(py_inputs, py_attn_masks, py_labels) = make_smart_batches(val_sentences, val_labels, batch_size ,tokenizer,max_input_length)\r\n",
        "\r\n",
        "y_pred , y_true = get_predictions(py_inputs, py_attn_masks, py_labels)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Creating Smart Batches from 1,070 examples with batch size 16...\n",
            "\n",
            "Tokenizing 1,070 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 100 samples.\n",
            "  Tokenized 200 samples.\n",
            "  Tokenized 300 samples.\n",
            "  Tokenized 400 samples.\n",
            "  Tokenized 500 samples.\n",
            "  Tokenized 600 samples.\n",
            "  Tokenized 700 samples.\n",
            "  Tokenized 800 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,000 samples.\n",
            "DONE.\n",
            "     1,070 samples\n",
            "\n",
            "     1,070 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 67 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n",
            "Predicting labels for 67 test batches...\n",
            "  Batch       7  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      14  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      21  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:01\n",
            "  Batch      28  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      35  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      42  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      49  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      56  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n",
            "  Batch      63  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:00\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wAjlATBPgh7c",
        "outputId": "b8aa0ad5-1f56-4f41-964e-2fe7b5f1f529"
      },
      "source": [
        "print(classification_report(y_true, y_pred, target_names=class_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                 precision    recall  f1-score   support\n",
            "\n",
            "    Hope_speech       0.61      0.62      0.61       190\n",
            "Non-hope-speech       0.90      0.88      0.89       784\n",
            "  not-malayalam       0.73      0.86      0.79        96\n",
            "\n",
            "       accuracy                           0.83      1070\n",
            "      macro avg       0.75      0.79      0.76      1070\n",
            "   weighted avg       0.83      0.83      0.83      1070\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "id": "7nELY57NYshD",
        "outputId": "fa575f32-6347-45b5-d8a2-bcf4652f321d"
      },
      "source": [
        "cm = confusion_matrix(y_true, y_pred)\r\n",
        "show_confusion_matrix(cm, class_names)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sNkirROtFMqZ"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I37UTLXjFM95"
      },
      "source": [
        "## Saving the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k4mv0TG2FM95"
      },
      "source": [
        "torch.save(model,\"./drive/MyDrive/Codalab/HopeSpeech Detection For Equality,DiversityAnd Inclusion/Models/XLMRobertaForSequenceClassification_Mutlti_Malayalam_v2\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a9M_91z1yj0F"
      },
      "source": [
        "## Loading the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "42ObCUU1QcsJ"
      },
      "source": [
        "model = torch.load('./drive/MyDrive/HopeSpeech Detection For Equality,DiversityAnd Inclusion/Models/XLMRobertaForSequenceClassification_Mutlti_Malayalam_v2',map_location=device)"
      ],
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sAaHpZdIzsnT",
        "outputId": "f97f31ab-e808-428b-b25d-3fdbdd8f501f"
      },
      "source": [
        "model.cuda()"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "XLMRobertaForSequenceClassification(\n",
              "  (roberta): RobertaModel(\n",
              "    (embeddings): RobertaEmbeddings(\n",
              "      (word_embeddings): Embedding(250002, 768, padding_idx=1)\n",
              "      (position_embeddings): Embedding(514, 768, padding_idx=1)\n",
              "      (token_type_embeddings): Embedding(1, 768)\n",
              "      (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "      (dropout): Dropout(p=0.1, inplace=False)\n",
              "    )\n",
              "    (encoder): RobertaEncoder(\n",
              "      (layer): ModuleList(\n",
              "        (0): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (1): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (2): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (3): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (4): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (5): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (6): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (7): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (8): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (9): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (10): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "        (11): RobertaLayer(\n",
              "          (attention): RobertaAttention(\n",
              "            (self): RobertaSelfAttention(\n",
              "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "            (output): RobertaSelfOutput(\n",
              "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "              (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "              (dropout): Dropout(p=0.1, inplace=False)\n",
              "            )\n",
              "          )\n",
              "          (intermediate): RobertaIntermediate(\n",
              "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
              "          )\n",
              "          (output): RobertaOutput(\n",
              "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
              "            (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)\n",
              "            (dropout): Dropout(p=0.1, inplace=False)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "    )\n",
              "  )\n",
              "  (classifier): RobertaClassificationHead(\n",
              "    (dense): Linear(in_features=768, out_features=768, bias=True)\n",
              "    (dropout): Dropout(p=0.1, inplace=False)\n",
              "    (out_proj): Linear(in_features=768, out_features=3, bias=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hY2dTIjnFM98"
      },
      "source": [
        "## Making Predictions on Test Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XuUBBoiZFM98"
      },
      "source": [
        "test_df_path = join(dataset_dir,'malayalam_hope_test.csv')\r\n",
        "test_df = pd.read_csv(test_df_path,sep = '\\t')"
      ],
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "bFOWRE24FM98",
        "outputId": "f6a590a0-b30a-4db0-f1f8-cbe6c6934532"
      },
      "source": [
        "test_df"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>അതെ അണപൊട്ടി ഒഴുകുകയാണ്</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>കെവിൻ avde pettupoyi🤣🤣🤣</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>@my voice ullup illa daaa tayoli ath kond alle...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>@Albin Baiju</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>I respect you..</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1066</th>\n",
              "      <td>ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1067</th>\n",
              "      <td>@surya ishaan</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1068</th>\n",
              "      <td>ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1069</th>\n",
              "      <td>ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1070</th>\n",
              "      <td>Poorru  angodu chellu</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1071 rows × 1 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                   text\n",
              "0                               അതെ അണപൊട്ടി ഒഴുകുകയാണ്\n",
              "1                               കെവിൻ avde pettupoyi🤣🤣🤣\n",
              "2     @my voice ullup illa daaa tayoli ath kond alle...\n",
              "3                                          @Albin Baiju\n",
              "4                                       I respect you..\n",
              "...                                                 ...\n",
              "1066  ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...\n",
              "1067                                      @surya ishaan\n",
              "1068                ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ\n",
              "1069   ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...\n",
              "1070                              Poorru  angodu chellu\n",
              "\n",
              "[1071 rows x 1 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DbI53tABFM99"
      },
      "source": [
        "## Loading `test_sentences`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3uoBX0l7FM99"
      },
      "source": [
        "test_sentences = test_df['text'].values \r\n",
        "test_ids = test_df.index.values"
      ],
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7HnVIFvZFM99",
        "outputId": "c046be5b-b2ad-4a13-d9a7-a12e920cc3b3"
      },
      "source": [
        "(py_inputs, py_attn_masks,py_ids) = make_smart_batches_on_test(test_sentences,test_ids, 16,tokenizer,max_input_length = 400)"
      ],
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Creating Smart Batches from 1,071 examples with batch size 16...\n",
            "\n",
            "Tokenizing 1,071 samples...\n",
            "  Tokenized 0 samples.\n",
            "  Tokenized 100 samples.\n",
            "  Tokenized 200 samples.\n",
            "  Tokenized 300 samples.\n",
            "  Tokenized 400 samples.\n",
            "  Tokenized 500 samples.\n",
            "  Tokenized 600 samples.\n",
            "  Tokenized 700 samples.\n",
            "  Tokenized 800 samples.\n",
            "  Tokenized 900 samples.\n",
            "  Tokenized 1,000 samples.\n",
            "DONE.\n",
            "     1,071 samples\n",
            "\n",
            "     1,071 samples after sorting\n",
            "\n",
            "Creating batches of size 16...\n",
            "\n",
            "  DONE - Selected 67 batches.\n",
            "\n",
            "Padding out sequences within each batch...\n",
            "  DONE.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vjsjqQpRQcsV"
      },
      "source": [
        "## Evaluating accuracy over test set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rwkx-30AQcsV",
        "outputId": "2c35fb4a-91fb-4749-e5c0-ee72a46267b0"
      },
      "source": [
        "y_ids,y_preds  = get_predictions_test(py_inputs, py_attn_masks,py_ids)\r\n"
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Predicting labels for 67 test batches...\n",
            "  Batch       7  of       67.    Elapsed: 0:00:00.  Remaining: 0:00:04\n",
            "  Batch      14  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:03\n",
            "  Batch      21  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:02\n",
            "  Batch      28  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      35  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      42  of       67.    Elapsed: 0:00:01.  Remaining: 0:00:01\n",
            "  Batch      49  of       67.    Elapsed: 0:00:02.  Remaining: 0:00:01\n",
            "  Batch      56  of       67.    Elapsed: 0:00:02.  Remaining: 0:00:00\n",
            "  Batch      63  of       67.    Elapsed: 0:00:02.  Remaining: 0:00:00\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pDM-cnDHQcsW",
        "outputId": "c7d2517b-22ef-4faa-fd37-f7330e45b135"
      },
      "source": [
        "print(y_preds)"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1 0 1 ... 1 1 0]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kWWWjz7eUPc7",
        "outputId": "742d7e7c-4ff5-4d82-86ca-e555acde8362"
      },
      "source": [
        "print(y_ids)"
      ],
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[ 111  189  202 ... 1053 1055  877]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u2ybbsAoQcsY",
        "outputId": "13989f7e-e0a4-40ae-a3e8-070993775c65"
      },
      "source": [
        "le.inverse_transform(y_preds)"
      ],
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['Non_hope_speech', 'Hope_speech', 'Non_hope_speech', ...,\n",
              "       'Non_hope_speech', 'Non_hope_speech', 'Hope_speech'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 64
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "kSMb6sCLQcsZ",
        "outputId": "ab27f1a6-aee8-4d98-c622-c8c2dfe362bf"
      },
      "source": [
        "sns.countplot(y =le.inverse_transform(y_preds))"
      ],
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f312b8d1978>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 65
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4KXOXwH9U2MT",
        "outputId": "2a5d4c4e-993e-4485-b0bd-d370b61e857c"
      },
      "source": [
        "len(y_ids)"
      ],
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1071"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9-fkBgOxU4bX",
        "outputId": "b48c0784-954b-47fc-9449-6742c9ab91b5"
      },
      "source": [
        "len(le.inverse_transform(y_preds))"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1071"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sWPDqogJrjIZ"
      },
      "source": [
        "## Creating Submission DataFrame"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o4Hb7RD1yHol"
      },
      "source": [
        "df_new = pd.DataFrame({\r\n",
        "    'id':y_ids,\r\n",
        "    'label':le.inverse_transform(y_preds)\r\n",
        "})"
      ],
      "execution_count": 68,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "twQAkVzcn4Lg"
      },
      "source": [
        "test_df['id'] = test_df.index"
      ],
      "execution_count": 69,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "qov36eWmqawv",
        "outputId": "0adc7032-b78b-481b-b1df-833a086668fd"
      },
      "source": [
        "df_new"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>id</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>111</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>189</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>202</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>221</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>246</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1066</th>\n",
              "      <td>1033</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1067</th>\n",
              "      <td>1034</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1068</th>\n",
              "      <td>1053</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1069</th>\n",
              "      <td>1055</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1070</th>\n",
              "      <td>877</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1071 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "        id            label\n",
              "0      111  Non_hope_speech\n",
              "1      189      Hope_speech\n",
              "2      202  Non_hope_speech\n",
              "3      221  Non_hope_speech\n",
              "4      246  Non_hope_speech\n",
              "...    ...              ...\n",
              "1066  1033      Hope_speech\n",
              "1067  1034  Non_hope_speech\n",
              "1068  1053  Non_hope_speech\n",
              "1069  1055  Non_hope_speech\n",
              "1070   877      Hope_speech\n",
              "\n",
              "[1071 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "z0CEApHmqdLr",
        "outputId": "8700e24a-6d91-44f3-b307-73c76eec2060"
      },
      "source": [
        "test_df"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>id</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>അതെ അണപൊട്ടി ഒഴുകുകയാണ്</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>കെവിൻ avde pettupoyi🤣🤣🤣</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>@my voice ullup illa daaa tayoli ath kond alle...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>@Albin Baiju</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>I respect you..</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1066</th>\n",
              "      <td>ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...</td>\n",
              "      <td>1066</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1067</th>\n",
              "      <td>@surya ishaan</td>\n",
              "      <td>1067</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1068</th>\n",
              "      <td>ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ</td>\n",
              "      <td>1068</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1069</th>\n",
              "      <td>ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...</td>\n",
              "      <td>1069</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1070</th>\n",
              "      <td>Poorru  angodu chellu</td>\n",
              "      <td>1070</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1071 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                   text    id\n",
              "0                               അതെ അണപൊട്ടി ഒഴുകുകയാണ്     0\n",
              "1                               കെവിൻ avde pettupoyi🤣🤣🤣     1\n",
              "2     @my voice ullup illa daaa tayoli ath kond alle...     2\n",
              "3                                          @Albin Baiju     3\n",
              "4                                       I respect you..     4\n",
              "...                                                 ...   ...\n",
              "1066  ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...  1066\n",
              "1067                                      @surya ishaan  1067\n",
              "1068                ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ  1068\n",
              "1069   ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...  1069\n",
              "1070                              Poorru  angodu chellu  1070\n",
              "\n",
              "[1071 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "vxReWWmxn7Hj",
        "outputId": "710b7f56-638a-4cb9-d41b-ad5592108f8e"
      },
      "source": [
        "df_f = pd.merge(test_df,df_new,on = 'id')\r\n",
        "df_f"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>id</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>അതെ അണപൊട്ടി ഒഴുകുകയാണ്</td>\n",
              "      <td>0</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>കെവിൻ avde pettupoyi🤣🤣🤣</td>\n",
              "      <td>1</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>@my voice ullup illa daaa tayoli ath kond alle...</td>\n",
              "      <td>2</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>@Albin Baiju</td>\n",
              "      <td>3</td>\n",
              "      <td>not-malayalam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>I respect you..</td>\n",
              "      <td>4</td>\n",
              "      <td>not-malayalam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1066</th>\n",
              "      <td>ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...</td>\n",
              "      <td>1066</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1067</th>\n",
              "      <td>@surya ishaan</td>\n",
              "      <td>1067</td>\n",
              "      <td>not-malayalam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1068</th>\n",
              "      <td>ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ</td>\n",
              "      <td>1068</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1069</th>\n",
              "      <td>ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...</td>\n",
              "      <td>1069</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1070</th>\n",
              "      <td>Poorru  angodu chellu</td>\n",
              "      <td>1070</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1071 rows × 3 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                   text    id            label\n",
              "0                               അതെ അണപൊട്ടി ഒഴുകുകയാണ്     0  Non_hope_speech\n",
              "1                               കെവിൻ avde pettupoyi🤣🤣🤣     1  Non_hope_speech\n",
              "2     @my voice ullup illa daaa tayoli ath kond alle...     2  Non_hope_speech\n",
              "3                                          @Albin Baiju     3    not-malayalam\n",
              "4                                       I respect you..     4    not-malayalam\n",
              "...                                                 ...   ...              ...\n",
              "1066  ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...  1066  Non_hope_speech\n",
              "1067                                      @surya ishaan  1067    not-malayalam\n",
              "1068                ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ  1068  Non_hope_speech\n",
              "1069   ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...  1069  Non_hope_speech\n",
              "1070                              Poorru  angodu chellu  1070      Hope_speech\n",
              "\n",
              "[1071 rows x 3 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "1kuASymYn9aV",
        "outputId": "651834c6-fdb1-416a-fa22-ec5b2097fe2b"
      },
      "source": [
        "df_f = df_f[['id','text','label']]\r\n",
        "df_f"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>id</th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>അതെ അണപൊട്ടി ഒഴുകുകയാണ്</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>കെവിൻ avde pettupoyi🤣🤣🤣</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>@my voice ullup illa daaa tayoli ath kond alle...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>@Albin Baiju</td>\n",
              "      <td>not-malayalam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>I respect you..</td>\n",
              "      <td>not-malayalam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1066</th>\n",
              "      <td>1066</td>\n",
              "      <td>ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1067</th>\n",
              "      <td>1067</td>\n",
              "      <td>@surya ishaan</td>\n",
              "      <td>not-malayalam</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1068</th>\n",
              "      <td>1068</td>\n",
              "      <td>ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1069</th>\n",
              "      <td>1069</td>\n",
              "      <td>ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...</td>\n",
              "      <td>Non_hope_speech</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1070</th>\n",
              "      <td>1070</td>\n",
              "      <td>Poorru  angodu chellu</td>\n",
              "      <td>Hope_speech</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1071 rows × 3 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "        id                                               text            label\n",
              "0        0                            അതെ അണപൊട്ടി ഒഴുകുകയാണ്  Non_hope_speech\n",
              "1        1                            കെവിൻ avde pettupoyi🤣🤣🤣  Non_hope_speech\n",
              "2        2  @my voice ullup illa daaa tayoli ath kond alle...  Non_hope_speech\n",
              "3        3                                       @Albin Baiju    not-malayalam\n",
              "4        4                                    I respect you..    not-malayalam\n",
              "...    ...                                                ...              ...\n",
              "1066  1066  ചേട്ടാ അതിന് ഇവിടെ ആരാണ് ഇന്ത്യൻ മൊബൈൽ റിവ്യൂ ...  Non_hope_speech\n",
              "1067  1067                                      @surya ishaan    not-malayalam\n",
              "1068  1068                ഏതാണ് ആ മൂലയിൽ ഇരിക്കുന്ന അപ്പൂപ്പൻ  Non_hope_speech\n",
              "1069  1069   ഏതാ ആ ചൊമന്ന ടി ഷർട്ടിന് മേൽ കോട്ടിട്ട വവ്വാൽ...  Non_hope_speech\n",
              "1070  1070                              Poorru  angodu chellu      Hope_speech\n",
              "\n",
              "[1071 rows x 3 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p3FtBVTVroJi"
      },
      "source": [
        "## Saving Submission DataFrame"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SzoT0p9rqrO-"
      },
      "source": [
        "path = './drive/MyDrive/Codalab/HopeSpeech Detection For Equality,DiversityAnd Inclusion/SubmissionFiles'"
      ],
      "execution_count": 74,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LZmgZFM4qi7F"
      },
      "source": [
        "df_f.to_csv(join(path,'xlmr_malayalam.csv'),index=False)"
      ],
      "execution_count": 75,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jW9g_qgMv7mq"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}