BERT.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.9-final"
    },
    "orig_nbformat": 2,
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3.6.9 64-bit",
      "metadata": {
        "interpreter": {
          "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
        }
      }
    },
    "colab": {
      "name": "BERT.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "b545861dc6b245e991c0a05f00d04e9e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_eeaf38a6fb9d4ace9c1f39dc7e57352a",
              "IPY_MODEL_eeec4665a4a4417dbb951a09e20c861f",
              "IPY_MODEL_7370896e3de2432fb4164774cd785960"
            ],
            "layout": "IPY_MODEL_ea07dd14f11e481794515c3ae2953723"
          }
        },
        "eeaf38a6fb9d4ace9c1f39dc7e57352a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ab2aac51ab8b4d74a8c434dee44eabf8",
            "placeholder": "​",
            "style": "IPY_MODEL_a255da9bcabd4711b6072a163e0f6e79",
            "value": "Downloading vocab.txt: 100%"
          }
        },
        "eeec4665a4a4417dbb951a09e20c861f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_aaeae2e29e504670a0ec8585d9813586",
            "max": 231508,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ee145eab04a54d7e8a523fd3e0d6b1ca",
            "value": 231508
          }
        },
        "7370896e3de2432fb4164774cd785960": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c16e1af626104cd5b325294025a8d7fe",
            "placeholder": "​",
            "style": "IPY_MODEL_b35b5e26127e43dd93fbc1c0d0c131b2",
            "value": " 226k/226k [00:00&lt;00:00, 1.48MB/s]"
          }
        },
        "ea07dd14f11e481794515c3ae2953723": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ab2aac51ab8b4d74a8c434dee44eabf8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a255da9bcabd4711b6072a163e0f6e79": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "aaeae2e29e504670a0ec8585d9813586": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ee145eab04a54d7e8a523fd3e0d6b1ca": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "c16e1af626104cd5b325294025a8d7fe": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b35b5e26127e43dd93fbc1c0d0c131b2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "36f457298abe4c84983a6052ee30339b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_d05a8f73787e49e1b6cd1157271f9e4d",
              "IPY_MODEL_cc0b19097489491f8945f63464ed8113",
              "IPY_MODEL_327f3d42f4eb4dd7bdf8b0f0e328d902"
            ],
            "layout": "IPY_MODEL_a9f02981d9824942a73bfd0d5bdcfc4a"
          }
        },
        "d05a8f73787e49e1b6cd1157271f9e4d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3d4cb26d7c4f4a94a0a93051ee9302c4",
            "placeholder": "​",
            "style": "IPY_MODEL_a050f3a286d84767a60905e2af84b107",
            "value": "Downloading tokenizer_config.json: 100%"
          }
        },
        "cc0b19097489491f8945f63464ed8113": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c8205ff2db474720b4f6252680520304",
            "max": 28,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_b02eade19f4b495097f561802dc2467d",
            "value": 28
          }
        },
        "327f3d42f4eb4dd7bdf8b0f0e328d902": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_49bcfd1ac9ae432cba32b37c0cfca8c4",
            "placeholder": "​",
            "style": "IPY_MODEL_a2fc8722a197488b976dbb3936523294",
            "value": " 28.0/28.0 [00:00&lt;00:00, 945B/s]"
          }
        },
        "a9f02981d9824942a73bfd0d5bdcfc4a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3d4cb26d7c4f4a94a0a93051ee9302c4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a050f3a286d84767a60905e2af84b107": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c8205ff2db474720b4f6252680520304": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b02eade19f4b495097f561802dc2467d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "49bcfd1ac9ae432cba32b37c0cfca8c4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a2fc8722a197488b976dbb3936523294": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "51b723e776b84c2580afe0d913e997a5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_6dc451ada44d4cf58f15c93df8a24041",
              "IPY_MODEL_b46b4dcada974413a56370a036c847bd",
              "IPY_MODEL_6948b12686de42ecb19d8d679aae6fc8"
            ],
            "layout": "IPY_MODEL_0722b2b5984e418baeff6db049675e04"
          }
        },
        "6dc451ada44d4cf58f15c93df8a24041": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a416af20f5364545b7f88a62529d5925",
            "placeholder": "​",
            "style": "IPY_MODEL_76a81285307b4b8cb7c88d5f623bd207",
            "value": "Downloading config.json: 100%"
          }
        },
        "b46b4dcada974413a56370a036c847bd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ae681f89bf1d426ea28537ea8cd15445",
            "max": 570,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_1802b3fec79d4571816260c49f0506fc",
            "value": 570
          }
        },
        "6948b12686de42ecb19d8d679aae6fc8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_37821a7c8b404e6e9257629f66000256",
            "placeholder": "​",
            "style": "IPY_MODEL_49a5122de0ad414aa5a3340e11bf71c3",
            "value": " 570/570 [00:00&lt;00:00, 18.9kB/s]"
          }
        },
        "0722b2b5984e418baeff6db049675e04": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a416af20f5364545b7f88a62529d5925": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "76a81285307b4b8cb7c88d5f623bd207": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ae681f89bf1d426ea28537ea8cd15445": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1802b3fec79d4571816260c49f0506fc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "37821a7c8b404e6e9257629f66000256": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "49a5122de0ad414aa5a3340e11bf71c3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "7672cb2f16c14e69b2975d713732e9ba": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_ddc8ba8db8874dfd8d403e140d7d8e7a",
              "IPY_MODEL_a9c906bed63b4b218af63d9a9a81cff5",
              "IPY_MODEL_84f21df2d3464c1688d4f19d5569b9e4"
            ],
            "layout": "IPY_MODEL_b118fd54f78342548f440463ab25aa67"
          }
        },
        "ddc8ba8db8874dfd8d403e140d7d8e7a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_87c6aee571d14cf6bf3ed0550ecc3109",
            "placeholder": "​",
            "style": "IPY_MODEL_f29abffd2398473e8f10f696f06edb5c",
            "value": "Downloading pytorch_model.bin: 100%"
          }
        },
        "a9c906bed63b4b218af63d9a9a81cff5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f8455c5dddfd433088c032c358a9bf8e",
            "max": 440473133,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_8f50a041901d475aa732e043f3c5c1aa",
            "value": 440473133
          }
        },
        "84f21df2d3464c1688d4f19d5569b9e4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7f5dcd1edd9948388566068a0d72ab81",
            "placeholder": "​",
            "style": "IPY_MODEL_afcce7f86eb8470a8cd402108059eabf",
            "value": " 420M/420M [00:07&lt;00:00, 56.6MB/s]"
          }
        },
        "b118fd54f78342548f440463ab25aa67": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "87c6aee571d14cf6bf3ed0550ecc3109": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f29abffd2398473e8f10f696f06edb5c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "f8455c5dddfd433088c032c358a9bf8e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8f50a041901d475aa732e043f3c5c1aa": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "7f5dcd1edd9948388566068a0d72ab81": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "afcce7f86eb8470a8cd402108059eabf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    },
    "gpuClass": "standard",
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T_du22iY_i3w"
      },
      "source": [
        "# **Sentiment Analysis** of the **Restaurant Reviews** from **YELP** Dataset\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **BERT**"
      ],
      "metadata": {
        "id": "9rwMymppyR4R"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### The two datasets (**review** and **business**) that we need from **YELP** dataset can be found <a href=\"https://www.yelp.com/dataset\" target=\"_blank\">here</a>"
      ],
      "metadata": {
        "id": "-Yx1tAVP-HLJ"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "01vabqH3CFOG"
      },
      "source": [
        "#### Install **Transformers** "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6JpaP9fbCMI5",
        "outputId": "f22b012b-6027-44aa-d2a9-11fe619ad617"
      },
      "source": [
        "!pip install transformers"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting transformers\n",
            "  Downloading transformers-4.21.0-py3-none-any.whl (4.7 MB)\n",
            "\u001b[K     |████████████████████████████████| 4.7 MB 4.7 MB/s \n",
            "\u001b[?25hCollecting pyyaml>=5.1\n",
            "  Downloading PyYAML-6.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (596 kB)\n",
            "\u001b[K     |████████████████████████████████| 596 kB 64.5 MB/s \n",
            "\u001b[?25hRequirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (2022.6.2)\n",
            "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from transformers) (4.64.0)\n",
            "Requirement already satisfied: importlib-metadata in /usr/local/lib/python3.7/dist-packages (from transformers) (4.12.0)\n",
            "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.7/dist-packages (from transformers) (1.21.6)\n",
            "Collecting tokenizers!=0.11.3,<0.13,>=0.11.1\n",
            "  Downloading tokenizers-0.12.1-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (6.6 MB)\n",
            "\u001b[K     |████████████████████████████████| 6.6 MB 49.4 MB/s \n",
            "\u001b[?25hRequirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.7/dist-packages (from transformers) (21.3)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.7/dist-packages (from transformers) (3.7.1)\n",
            "Collecting huggingface-hub<1.0,>=0.1.0\n",
            "  Downloading huggingface_hub-0.8.1-py3-none-any.whl (101 kB)\n",
            "\u001b[K     |████████████████████████████████| 101 kB 11.2 MB/s \n",
            "\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from transformers) (2.23.0)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.7/dist-packages (from huggingface-hub<1.0,>=0.1.0->transformers) (4.1.1)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=20.0->transformers) (3.0.9)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata->transformers) (3.8.1)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2022.6.15)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (2.10)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->transformers) (3.0.4)\n",
            "Installing collected packages: pyyaml, tokenizers, huggingface-hub, transformers\n",
            "  Attempting uninstall: pyyaml\n",
            "    Found existing installation: PyYAML 3.13\n",
            "    Uninstalling PyYAML-3.13:\n",
            "      Successfully uninstalled PyYAML-3.13\n",
            "Successfully installed huggingface-hub-0.8.1 pyyaml-6.0 tokenizers-0.12.1 transformers-4.21.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Install **langdetect** "
      ],
      "metadata": {
        "id": "wFeNaNKT53C5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install langdetect"
      ],
      "metadata": {
        "id": "fL_r-wB3sNva",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "308f2127-a33d-4942-97bb-5fbfe4aecc23"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting langdetect\n",
            "  Downloading langdetect-1.0.9.tar.gz (981 kB)\n",
            "\u001b[K     |████████████████████████████████| 981 kB 4.7 MB/s \n",
            "\u001b[?25hRequirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from langdetect) (1.15.0)\n",
            "Building wheels for collected packages: langdetect\n",
            "  Building wheel for langdetect (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for langdetect: filename=langdetect-1.0.9-py3-none-any.whl size=993242 sha256=1d47711e915ade7ec10e0baaf2b120d5c86a75700236de3d0f93a1564574adcb\n",
            "  Stored in directory: /root/.cache/pip/wheels/c5/96/8a/f90c59ed25d75e50a8c10a1b1c2d4c402e4dacfa87f3aff36a\n",
            "Successfully built langdetect\n",
            "Installing collected packages: langdetect\n",
            "Successfully installed langdetect-1.0.9\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive \n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yefa9iJT8RxY",
        "outputId": "56578407-7452-4ef6-e1de-133eefa5006f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bwoJKCWp__xf"
      },
      "source": [
        "#### Importing most of the libraries "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ogxuogt_AEyV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ee3b74fb-a302-4d21-fc0d-36387b26a4e3"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "from matplotlib import rc\n",
        "from pylab import rcParams\n",
        "import matplotlib.pyplot as plt\n",
        "from textwrap import wrap\n",
        "from collections import defaultdict\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import confusion_matrix, classification_report, accuracy_score,f1_score,precision_score, recall_score\n",
        "\n",
        "import re\n",
        "import spacy\n",
        "sp = spacy.load('en_core_web_sm')\n",
        "\n",
        "import nltk\n",
        "nltk.download('punkt')\n",
        "from nltk.tokenize import word_tokenize\n",
        "\n",
        "import transformers\n",
        "from transformers import BertModel, BertTokenizer, BertForSequenceClassification\n",
        "from transformers import AdamW, get_linear_schedule_with_warmup\n",
        "\n",
        "import torch\n",
        "from torch import nn,optim\n",
        "from torch.utils.data import Dataset,DataLoader,TensorDataset, RandomSampler, SequentialSampler\n",
        "import torch.nn as nn  \n",
        "import torch.nn.functional as F  \n",
        "device=torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
        "\n",
        "import time \n",
        "import datetime\n",
        "\n",
        "import tensorflow as tf\n",
        "\n",
        "%matplotlib inline\n",
        "%config InlineBackend.figure_format='retina'\n",
        "\n",
        "sns.set(style='whitegrid',palette='muted',font_scale=1.2)\n",
        "color_palette=['#01BEFE','#FFDD00','#FF7D00','#FF006D','#ADFF02','#8F00FF']\n",
        "sns.set_palette(sns.color_palette(color_palette))\n",
        "\n",
        "rcParams['figure.figsize']= 10,5\n",
        "\n",
        "seed=42\n",
        "np.random.seed(seed)\n",
        "torch.manual_seed(seed)\n",
        "\n",
        "import sys\n",
        "import os\n",
        "import warnings\n",
        "\n",
        "if not sys.warnoptions:\n",
        "    warnings.simplefilter(\"ignore\")\n",
        "    os.environ[\"PYTHONWARNINGS\"] = \"ignore\" "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Read** the **reviews** dataset. As it is a **very heavy** one, let's use only the first 10,000 rows and **reduce** the **memory usage**"
      ],
      "metadata": {
        "id": "WO_dhSZ7zlDq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_json('/content/drive/MyDrive/yelp_academic_dataset_review.json', nrows=10000, lines=True)"
      ],
      "metadata": {
        "id": "Vf0FMog8Orgk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "id": "2JVm5rW8OrlT",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "d7c18222-f688-4f7c-968b-0258010f7b0b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                review_id                 user_id             business_id  \\\n",
              "0  lWC-xP3rd6obsecCYsGZRg  ak0TdVmGKo4pwqdJSTLwWw  buF9druCkbuXLX526sGELQ   \n",
              "1  8bFej1QE5LXp4O05qjGqXA  YoVfDbnISlW0f7abNQACIg  RA4V8pr014UyUbDvI-LW2A   \n",
              "2  NDhkzczKjLshODbqDoNLSg  eC5evKn1TWDyHCyQAwguUw  _sS2LBIGNT5NQb6PD1Vtjw   \n",
              "3  T5fAqjjFooT4V0OeZyuk1w  SFQ1jcnGguO0LYWnbbftAA  0AzLzHfOJgL7ROwhdww2ew   \n",
              "4  sjm_uUcQVxab_EeLCqsYLg  0kA0PAJ8QFMeveQWHFqz2A  8zehGz9jnxPqXtOc7KaJxA   \n",
              "\n",
              "   stars  useful  funny  cool  \\\n",
              "0      4       3      1     1   \n",
              "1      4       1      0     0   \n",
              "2      5       0      0     0   \n",
              "3      2       1      1     1   \n",
              "4      4       0      0     0   \n",
              "\n",
              "                                                text                date  \n",
              "0  Apparently Prides Osteria had a rough summer a... 2014-10-11 03:34:02  \n",
              "1  This store is pretty good. Not as great as Wal... 2015-07-03 20:38:25  \n",
              "2  I called WVM on the recommendation of a couple... 2013-05-28 20:38:06  \n",
              "3  I've stayed at many Marriott and Renaissance M... 2010-01-08 02:29:15  \n",
              "4  The food is always great here. The service fro... 2011-07-28 18:05:01  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3de550bb-b688-461a-9258-98b83e3055f0\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>review_id</th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>stars</th>\n",
              "      <th>useful</th>\n",
              "      <th>funny</th>\n",
              "      <th>cool</th>\n",
              "      <th>text</th>\n",
              "      <th>date</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>lWC-xP3rd6obsecCYsGZRg</td>\n",
              "      <td>ak0TdVmGKo4pwqdJSTLwWw</td>\n",
              "      <td>buF9druCkbuXLX526sGELQ</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Apparently Prides Osteria had a rough summer a...</td>\n",
              "      <td>2014-10-11 03:34:02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>8bFej1QE5LXp4O05qjGqXA</td>\n",
              "      <td>YoVfDbnISlW0f7abNQACIg</td>\n",
              "      <td>RA4V8pr014UyUbDvI-LW2A</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>This store is pretty good. Not as great as Wal...</td>\n",
              "      <td>2015-07-03 20:38:25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>NDhkzczKjLshODbqDoNLSg</td>\n",
              "      <td>eC5evKn1TWDyHCyQAwguUw</td>\n",
              "      <td>_sS2LBIGNT5NQb6PD1Vtjw</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>I called WVM on the recommendation of a couple...</td>\n",
              "      <td>2013-05-28 20:38:06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>T5fAqjjFooT4V0OeZyuk1w</td>\n",
              "      <td>SFQ1jcnGguO0LYWnbbftAA</td>\n",
              "      <td>0AzLzHfOJgL7ROwhdww2ew</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>I've stayed at many Marriott and Renaissance M...</td>\n",
              "      <td>2010-01-08 02:29:15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>sjm_uUcQVxab_EeLCqsYLg</td>\n",
              "      <td>0kA0PAJ8QFMeveQWHFqz2A</td>\n",
              "      <td>8zehGz9jnxPqXtOc7KaJxA</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>The food is always great here. The service fro...</td>\n",
              "      <td>2011-07-28 18:05:01</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3de550bb-b688-461a-9258-98b83e3055f0')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-3de550bb-b688-461a-9258-98b83e3055f0 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-3de550bb-b688-461a-9258-98b83e3055f0');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Drop the columns** that we **won't be needing**"
      ],
      "metadata": {
        "id": "KOv92EsW1m8h"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "cols_to_drop = ['review_id', 'user_id', 'useful', 'funny', 'cool', 'date'] \n",
        "df.drop(cols_to_drop, axis=1, inplace=True)"
      ],
      "metadata": {
        "id": "7fkvyL1xOrpw"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "id": "lF-Fe0afOruS",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "bd967509-0763-4de5-d761-e87ea39ea474"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              business_id  stars  \\\n",
              "0  buF9druCkbuXLX526sGELQ      4   \n",
              "1  RA4V8pr014UyUbDvI-LW2A      4   \n",
              "2  _sS2LBIGNT5NQb6PD1Vtjw      5   \n",
              "3  0AzLzHfOJgL7ROwhdww2ew      2   \n",
              "4  8zehGz9jnxPqXtOc7KaJxA      4   \n",
              "\n",
              "                                                text  \n",
              "0  Apparently Prides Osteria had a rough summer a...  \n",
              "1  This store is pretty good. Not as great as Wal...  \n",
              "2  I called WVM on the recommendation of a couple...  \n",
              "3  I've stayed at many Marriott and Renaissance M...  \n",
              "4  The food is always great here. The service fro...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4b715674-1a52-44fe-a71b-8ecc46225008\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>business_id</th>\n",
              "      <th>stars</th>\n",
              "      <th>text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>buF9druCkbuXLX526sGELQ</td>\n",
              "      <td>4</td>\n",
              "      <td>Apparently Prides Osteria had a rough summer a...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>RA4V8pr014UyUbDvI-LW2A</td>\n",
              "      <td>4</td>\n",
              "      <td>This store is pretty good. Not as great as Wal...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>_sS2LBIGNT5NQb6PD1Vtjw</td>\n",
              "      <td>5</td>\n",
              "      <td>I called WVM on the recommendation of a couple...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0AzLzHfOJgL7ROwhdww2ew</td>\n",
              "      <td>2</td>\n",
              "      <td>I've stayed at many Marriott and Renaissance M...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>8zehGz9jnxPqXtOc7KaJxA</td>\n",
              "      <td>4</td>\n",
              "      <td>The food is always great here. The service fro...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4b715674-1a52-44fe-a71b-8ecc46225008')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-4b715674-1a52-44fe-a71b-8ecc46225008 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-4b715674-1a52-44fe-a71b-8ecc46225008');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Importing** the **business** dataset"
      ],
      "metadata": {
        "id": "yPSlv4-Q2AQ9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df1 = pd.read_json('/content/drive/MyDrive/yelp_academic_dataset_business.json', lines=True)"
      ],
      "metadata": {
        "id": "KWjs5VloOryD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df1.head()"
      ],
      "metadata": {
        "id": "ryBewrQS2YT0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 528
        },
        "outputId": "bd964965-36e3-4cd1-a632-813b1b24a737"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              business_id                     name              address  \\\n",
              "0  6iYb2HFDywm3zjuRg0shjw      Oskar Blues Taproom         921 Pearl St   \n",
              "1  tCbdrRPZA0oiIYSmHG3J0w  Flying Elephants at PDX  7000 NE Airport Way   \n",
              "2  bvN78flM8NLprQ1a1y5dRg           The Reclaimory   4720 Hawthorne Ave   \n",
              "3  oaepsyvc0J17qwi8cfrOWg              Great Clips   2566 Enterprise Rd   \n",
              "4  PE9uqAjdw0E4-8mjGl3wVA        Crossfit Terminus  1046 Memorial Dr SE   \n",
              "\n",
              "          city state postal_code   latitude   longitude  stars  review_count  \\\n",
              "0      Boulder    CO       80302  40.017544 -105.283348    4.0            86   \n",
              "1     Portland    OR       97218  45.588906 -122.593331    4.0           126   \n",
              "2     Portland    OR       97214  45.511907 -122.613693    4.5            13   \n",
              "3  Orange City    FL       32763  28.914482  -81.295979    3.0             8   \n",
              "4      Atlanta    GA       30316  33.747027  -84.353424    4.0            14   \n",
              "\n",
              "   is_open                                         attributes  \\\n",
              "0        1  {'RestaurantsTableService': 'True', 'WiFi': 'u...   \n",
              "1        1  {'RestaurantsTakeOut': 'True', 'RestaurantsAtt...   \n",
              "2        1  {'BusinessAcceptsCreditCards': 'True', 'Restau...   \n",
              "3        1  {'RestaurantsPriceRange2': '1', 'BusinessAccep...   \n",
              "4        1  {'GoodForKids': 'False', 'BusinessParking': '{...   \n",
              "\n",
              "                                          categories  \\\n",
              "0  Gastropubs, Food, Beer Gardens, Restaurants, B...   \n",
              "1  Salad, Soup, Sandwiches, Delis, Restaurants, C...   \n",
              "2  Antiques, Fashion, Used, Vintage & Consignment...   \n",
              "3                         Beauty & Spas, Hair Salons   \n",
              "4  Gyms, Active Life, Interval Training Gyms, Fit...   \n",
              "\n",
              "                                               hours  \n",
              "0  {'Monday': '11:0-23:0', 'Tuesday': '11:0-23:0'...  \n",
              "1  {'Monday': '5:0-18:0', 'Tuesday': '5:0-17:0', ...  \n",
              "2  {'Thursday': '11:0-18:0', 'Friday': '11:0-18:0...  \n",
              "3                                               None  \n",
              "4  {'Monday': '16:0-19:0', 'Tuesday': '16:0-19:0'...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2a4c4bbc-6ee0-4caa-93a7-3a05c971a850\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>business_id</th>\n",
              "      <th>name</th>\n",
              "      <th>address</th>\n",
              "      <th>city</th>\n",
              "      <th>state</th>\n",
              "      <th>postal_code</th>\n",
              "      <th>latitude</th>\n",
              "      <th>longitude</th>\n",
              "      <th>stars</th>\n",
              "      <th>review_count</th>\n",
              "      <th>is_open</th>\n",
              "      <th>attributes</th>\n",
              "      <th>categories</th>\n",
              "      <th>hours</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>6iYb2HFDywm3zjuRg0shjw</td>\n",
              "      <td>Oskar Blues Taproom</td>\n",
              "      <td>921 Pearl St</td>\n",
              "      <td>Boulder</td>\n",
              "      <td>CO</td>\n",
              "      <td>80302</td>\n",
              "      <td>40.017544</td>\n",
              "      <td>-105.283348</td>\n",
              "      <td>4.0</td>\n",
              "      <td>86</td>\n",
              "      <td>1</td>\n",
              "      <td>{'RestaurantsTableService': 'True', 'WiFi': 'u...</td>\n",
              "      <td>Gastropubs, Food, Beer Gardens, Restaurants, B...</td>\n",
              "      <td>{'Monday': '11:0-23:0', 'Tuesday': '11:0-23:0'...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>tCbdrRPZA0oiIYSmHG3J0w</td>\n",
              "      <td>Flying Elephants at PDX</td>\n",
              "      <td>7000 NE Airport Way</td>\n",
              "      <td>Portland</td>\n",
              "      <td>OR</td>\n",
              "      <td>97218</td>\n",
              "      <td>45.588906</td>\n",
              "      <td>-122.593331</td>\n",
              "      <td>4.0</td>\n",
              "      <td>126</td>\n",
              "      <td>1</td>\n",
              "      <td>{'RestaurantsTakeOut': 'True', 'RestaurantsAtt...</td>\n",
              "      <td>Salad, Soup, Sandwiches, Delis, Restaurants, C...</td>\n",
              "      <td>{'Monday': '5:0-18:0', 'Tuesday': '5:0-17:0', ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>bvN78flM8NLprQ1a1y5dRg</td>\n",
              "      <td>The Reclaimory</td>\n",
              "      <td>4720 Hawthorne Ave</td>\n",
              "      <td>Portland</td>\n",
              "      <td>OR</td>\n",
              "      <td>97214</td>\n",
              "      <td>45.511907</td>\n",
              "      <td>-122.613693</td>\n",
              "      <td>4.5</td>\n",
              "      <td>13</td>\n",
              "      <td>1</td>\n",
              "      <td>{'BusinessAcceptsCreditCards': 'True', 'Restau...</td>\n",
              "      <td>Antiques, Fashion, Used, Vintage &amp; Consignment...</td>\n",
              "      <td>{'Thursday': '11:0-18:0', 'Friday': '11:0-18:0...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>oaepsyvc0J17qwi8cfrOWg</td>\n",
              "      <td>Great Clips</td>\n",
              "      <td>2566 Enterprise Rd</td>\n",
              "      <td>Orange City</td>\n",
              "      <td>FL</td>\n",
              "      <td>32763</td>\n",
              "      <td>28.914482</td>\n",
              "      <td>-81.295979</td>\n",
              "      <td>3.0</td>\n",
              "      <td>8</td>\n",
              "      <td>1</td>\n",
              "      <td>{'RestaurantsPriceRange2': '1', 'BusinessAccep...</td>\n",
              "      <td>Beauty &amp; Spas, Hair Salons</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>PE9uqAjdw0E4-8mjGl3wVA</td>\n",
              "      <td>Crossfit Terminus</td>\n",
              "      <td>1046 Memorial Dr SE</td>\n",
              "      <td>Atlanta</td>\n",
              "      <td>GA</td>\n",
              "      <td>30316</td>\n",
              "      <td>33.747027</td>\n",
              "      <td>-84.353424</td>\n",
              "      <td>4.0</td>\n",
              "      <td>14</td>\n",
              "      <td>1</td>\n",
              "      <td>{'GoodForKids': 'False', 'BusinessParking': '{...</td>\n",
              "      <td>Gyms, Active Life, Interval Training Gyms, Fit...</td>\n",
              "      <td>{'Monday': '16:0-19:0', 'Tuesday': '16:0-19:0'...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2a4c4bbc-6ee0-4caa-93a7-3a05c971a850')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-2a4c4bbc-6ee0-4caa-93a7-3a05c971a850 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-2a4c4bbc-6ee0-4caa-93a7-3a05c971a850');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### We **only** want to analyze the **restaurant** category from the **business** dataset"
      ],
      "metadata": {
        "id": "9Jkb7fTY2dxg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df1 = df1[df1['categories'].notnull()]\n",
        "df_rest = df1[df1['categories'].str.contains('Restaurant')]"
      ],
      "metadata": {
        "id": "LkzRAnT2Or1w"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_rest = df_rest[['business_id', 'categories']]"
      ],
      "metadata": {
        "id": "NIEee5u-Or6J"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_rest.head()"
      ],
      "metadata": {
        "id": "cL2klGyh3LM3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "89a8e04b-d695-4092-b146-0b63c018a65b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               business_id                                         categories\n",
              "0   6iYb2HFDywm3zjuRg0shjw  Gastropubs, Food, Beer Gardens, Restaurants, B...\n",
              "1   tCbdrRPZA0oiIYSmHG3J0w  Salad, Soup, Sandwiches, Delis, Restaurants, C...\n",
              "5   D4JtQNTI4X3KcbzacDJsMw                                  Restaurants, Thai\n",
              "7   jFYIsSb7r1QeESVUnXPHBw                    Breakfast & Brunch, Restaurants\n",
              "12  HPA_qyMEddpAEtFof02ixg                           Food, Pizza, Restaurants"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-dd373b1f-36ba-4385-8732-bc44f16b47bd\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>business_id</th>\n",
              "      <th>categories</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>6iYb2HFDywm3zjuRg0shjw</td>\n",
              "      <td>Gastropubs, Food, Beer Gardens, Restaurants, B...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>tCbdrRPZA0oiIYSmHG3J0w</td>\n",
              "      <td>Salad, Soup, Sandwiches, Delis, Restaurants, C...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>D4JtQNTI4X3KcbzacDJsMw</td>\n",
              "      <td>Restaurants, Thai</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>jFYIsSb7r1QeESVUnXPHBw</td>\n",
              "      <td>Breakfast &amp; Brunch, Restaurants</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>HPA_qyMEddpAEtFof02ixg</td>\n",
              "      <td>Food, Pizza, Restaurants</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-dd373b1f-36ba-4385-8732-bc44f16b47bd')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-dd373b1f-36ba-4385-8732-bc44f16b47bd button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-dd373b1f-36ba-4385-8732-bc44f16b47bd');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### We want to **merge** the two datasets to analyse the **restaurant reviews**. \"**business_id**\" is the **common** columm of the two datasets"
      ],
      "metadata": {
        "id": "Qodph1Yj3Pl-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# business_id is tne common columm of the two datasets\n",
        "df_merged = df.merge(df_rest, how='inner', on='business_id')"
      ],
      "metadata": {
        "id": "1gYxDOuFOsL-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "del df, df_rest"
      ],
      "metadata": {
        "id": "usnIvi3LOsQe"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_merged.head()"
      ],
      "metadata": {
        "id": "LenH4d69Ot6x",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "8b71d29f-6e7f-45e3-c06a-a656a7c21c5f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "              business_id  stars  \\\n",
              "0  buF9druCkbuXLX526sGELQ      4   \n",
              "1  buF9druCkbuXLX526sGELQ      2   \n",
              "2  buF9druCkbuXLX526sGELQ      5   \n",
              "3  buF9druCkbuXLX526sGELQ      5   \n",
              "4  0AzLzHfOJgL7ROwhdww2ew      2   \n",
              "\n",
              "                                                text  \\\n",
              "0  Apparently Prides Osteria had a rough summer a...   \n",
              "1  I was really disappointed to say the least. I ...   \n",
              "2  This is as close to dining in Italy as you'll ...   \n",
              "3  Great food and service! Again. 4 out of the la...   \n",
              "4  I've stayed at many Marriott and Renaissance M...   \n",
              "\n",
              "                                          categories  \n",
              "0  Restaurants, Wine Bars, Nightlife, Farmers Mar...  \n",
              "1  Restaurants, Wine Bars, Nightlife, Farmers Mar...  \n",
              "2  Restaurants, Wine Bars, Nightlife, Farmers Mar...  \n",
              "3  Restaurants, Wine Bars, Nightlife, Farmers Mar...  \n",
              "4  Hotels, Hotels & Travel, Restaurants, Event Pl...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5d697c0c-1eab-4c66-9e01-3f2e52cebe87\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>business_id</th>\n",
              "      <th>stars</th>\n",
              "      <th>text</th>\n",
              "      <th>categories</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>buF9druCkbuXLX526sGELQ</td>\n",
              "      <td>4</td>\n",
              "      <td>Apparently Prides Osteria had a rough summer a...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>buF9druCkbuXLX526sGELQ</td>\n",
              "      <td>2</td>\n",
              "      <td>I was really disappointed to say the least. I ...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>buF9druCkbuXLX526sGELQ</td>\n",
              "      <td>5</td>\n",
              "      <td>This is as close to dining in Italy as you'll ...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>buF9druCkbuXLX526sGELQ</td>\n",
              "      <td>5</td>\n",
              "      <td>Great food and service! Again. 4 out of the la...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0AzLzHfOJgL7ROwhdww2ew</td>\n",
              "      <td>2</td>\n",
              "      <td>I've stayed at many Marriott and Renaissance M...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5d697c0c-1eab-4c66-9e01-3f2e52cebe87')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-5d697c0c-1eab-4c66-9e01-3f2e52cebe87 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-5d697c0c-1eab-4c66-9e01-3f2e52cebe87');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Rename** the **text column** and **remove** the **\"business_id\"** column because it **isn't necessary anymore**"
      ],
      "metadata": {
        "id": "exdkMOVa5Hyo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Rename the text column and remove the \"business_id\" column because it isn't necessary anymore\n",
        "df_merged.rename(columns={'text':'restaurant_reviews'}, inplace=True)\n",
        "df_merged.drop('business_id', axis=1, inplace=True)"
      ],
      "metadata": {
        "id": "eUDpeGrUOt_S"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### We will **only** analyse the reviews **written** in **English**"
      ],
      "metadata": {
        "id": "x80FJmC5uXI4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from langdetect import detect"
      ],
      "metadata": {
        "id": "DSiq-u_xP5mF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_merged['detect'] = df_merged['restaurant_reviews'].apply(detect)\n",
        "df_merged = df_merged[df_merged['detect'] == 'en'].reset_index(drop=True)"
      ],
      "metadata": {
        "id": "vl_AiME2P5r0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Check if there is NaN values\n",
        "df_merged.isnull().values.any()"
      ],
      "metadata": {
        "id": "DZ7M_4epP5xv",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "197b0dbc-709e-4d86-92d2-fa66577a5365"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "False"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Return the dataframe with **duplicate** rows **removed**"
      ],
      "metadata": {
        "id": "i5TAVn6vmHoX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_merged.drop_duplicates(inplace=True)"
      ],
      "metadata": {
        "id": "vu4cJkS5mIIL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### As **many** of the **3 stars reviews**, most likely, have both **good and bad aspects/feelings**, we will **remove** those reviews so they **won't harm** our sentiment analysis"
      ],
      "metadata": {
        "id": "WT9zEKyi6hk0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_merged = df_merged[(df_merged[\"stars\"]!=3)]\n",
        "df_merged.head(20)"
      ],
      "metadata": {
        "id": "VxG96PWOP6D8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 676
        },
        "outputId": "97e51bcc-1eb7-49ea-f13a-ca55bec2e65c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    stars                                 restaurant_reviews  \\\n",
              "0       4  Apparently Prides Osteria had a rough summer a...   \n",
              "1       2  I was really disappointed to say the least. I ...   \n",
              "2       5  This is as close to dining in Italy as you'll ...   \n",
              "3       5  Great food and service! Again. 4 out of the la...   \n",
              "4       2  I've stayed at many Marriott and Renaissance M...   \n",
              "5       4  I was a there for a professional conference. s...   \n",
              "6       4  I stayed for five nights this past April with ...   \n",
              "7       4  I stayed here recently for a conference I was ...   \n",
              "8       4  The food is always great here. The service fro...   \n",
              "9       2  Brasserie zero zero... we were hoping for a te...   \n",
              "10      5  I absolutely love this restaurant. I've been c...   \n",
              "11      5  The food is great!  One of my fav brunch spots...   \n",
              "12      5  Went here for dinner in day two of our trip. I...   \n",
              "13      5  Food: 4.0 of 5.0\\nService: 5.0 of 5.0\\nEnviron...   \n",
              "14      4  one of the staples in boulder.  fabulous space...   \n",
              "15      5  Wow. Best service I have ever had. Slight wait...   \n",
              "16      4  I'm giving four stars because the service was ...   \n",
              "17      5  I had the duck, the fromage special, and the m...   \n",
              "18      4  Better than I expected it to be.  To be fair, ...   \n",
              "19      5  I like this place! Every time I visit Boulder,...   \n",
              "\n",
              "                                           categories detect  \n",
              "0   Restaurants, Wine Bars, Nightlife, Farmers Mar...     en  \n",
              "1   Restaurants, Wine Bars, Nightlife, Farmers Mar...     en  \n",
              "2   Restaurants, Wine Bars, Nightlife, Farmers Mar...     en  \n",
              "3   Restaurants, Wine Bars, Nightlife, Farmers Mar...     en  \n",
              "4   Hotels, Hotels & Travel, Restaurants, Event Pl...     en  \n",
              "5   Hotels, Hotels & Travel, Restaurants, Event Pl...     en  \n",
              "6   Hotels, Hotels & Travel, Restaurants, Event Pl...     en  \n",
              "7   Hotels, Hotels & Travel, Restaurants, Event Pl...     en  \n",
              "8                                 Restaurants, French     en  \n",
              "9                                 Restaurants, French     en  \n",
              "10                                Restaurants, French     en  \n",
              "11                                Restaurants, French     en  \n",
              "12                                Restaurants, French     en  \n",
              "13                                Restaurants, French     en  \n",
              "14                                Restaurants, French     en  \n",
              "15                                Restaurants, French     en  \n",
              "16                                Restaurants, French     en  \n",
              "17                                Restaurants, French     en  \n",
              "18                                Restaurants, French     en  \n",
              "19                                Restaurants, French     en  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e2080fb1-6551-4457-9c7f-9299c379eb53\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>stars</th>\n",
              "      <th>restaurant_reviews</th>\n",
              "      <th>categories</th>\n",
              "      <th>detect</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4</td>\n",
              "      <td>Apparently Prides Osteria had a rough summer a...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>I was really disappointed to say the least. I ...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5</td>\n",
              "      <td>This is as close to dining in Italy as you'll ...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>5</td>\n",
              "      <td>Great food and service! Again. 4 out of the la...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "      <td>I've stayed at many Marriott and Renaissance M...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>4</td>\n",
              "      <td>I was a there for a professional conference. s...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>4</td>\n",
              "      <td>I stayed for five nights this past April with ...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>4</td>\n",
              "      <td>I stayed here recently for a conference I was ...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>4</td>\n",
              "      <td>The food is always great here. The service fro...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>2</td>\n",
              "      <td>Brasserie zero zero... we were hoping for a te...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>5</td>\n",
              "      <td>I absolutely love this restaurant. I've been c...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>5</td>\n",
              "      <td>The food is great!  One of my fav brunch spots...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>5</td>\n",
              "      <td>Went here for dinner in day two of our trip. I...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>5</td>\n",
              "      <td>Food: 4.0 of 5.0\\nService: 5.0 of 5.0\\nEnviron...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>4</td>\n",
              "      <td>one of the staples in boulder.  fabulous space...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>5</td>\n",
              "      <td>Wow. Best service I have ever had. Slight wait...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>4</td>\n",
              "      <td>I'm giving four stars because the service was ...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>5</td>\n",
              "      <td>I had the duck, the fromage special, and the m...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>4</td>\n",
              "      <td>Better than I expected it to be.  To be fair, ...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>5</td>\n",
              "      <td>I like this place! Every time I visit Boulder,...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e2080fb1-6551-4457-9c7f-9299c379eb53')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e2080fb1-6551-4457-9c7f-9299c379eb53 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e2080fb1-6551-4457-9c7f-9299c379eb53');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Label** reviews as **positive** - **1** (4 and 5 stars) - or **negative** - **0** (1 and 2 stars) - and after that **remove** the **\"stars\" column**, it is not needed anymore"
      ],
      "metadata": {
        "id": "lXOtX5Ih6q6G"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# label reviews as positive or negative and after that remove the \"stars\" column, it is not needed anymore\n",
        " \n",
        "df_merged.loc[df_merged['stars'] < 3, 'sentiment'] = 0 \n",
        "df_merged.loc[df_merged['stars'] > 3, 'sentiment'] = 1\n",
        "df_merged.drop('stars', axis=1, inplace=True)\n",
        "df_merged.head(10)"
      ],
      "metadata": {
        "id": "vOH0Cqr7P6JB",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "outputId": "fd754375-9d36-4b3d-f661-a960141f01ba"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                  restaurant_reviews  \\\n",
              "0  Apparently Prides Osteria had a rough summer a...   \n",
              "1  I was really disappointed to say the least. I ...   \n",
              "2  This is as close to dining in Italy as you'll ...   \n",
              "3  Great food and service! Again. 4 out of the la...   \n",
              "4  I've stayed at many Marriott and Renaissance M...   \n",
              "5  I was a there for a professional conference. s...   \n",
              "6  I stayed for five nights this past April with ...   \n",
              "7  I stayed here recently for a conference I was ...   \n",
              "8  The food is always great here. The service fro...   \n",
              "9  Brasserie zero zero... we were hoping for a te...   \n",
              "\n",
              "                                          categories detect  sentiment  \n",
              "0  Restaurants, Wine Bars, Nightlife, Farmers Mar...     en        1.0  \n",
              "1  Restaurants, Wine Bars, Nightlife, Farmers Mar...     en        0.0  \n",
              "2  Restaurants, Wine Bars, Nightlife, Farmers Mar...     en        1.0  \n",
              "3  Restaurants, Wine Bars, Nightlife, Farmers Mar...     en        1.0  \n",
              "4  Hotels, Hotels & Travel, Restaurants, Event Pl...     en        0.0  \n",
              "5  Hotels, Hotels & Travel, Restaurants, Event Pl...     en        1.0  \n",
              "6  Hotels, Hotels & Travel, Restaurants, Event Pl...     en        1.0  \n",
              "7  Hotels, Hotels & Travel, Restaurants, Event Pl...     en        1.0  \n",
              "8                                Restaurants, French     en        1.0  \n",
              "9                                Restaurants, French     en        0.0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8a4d602f-4551-4e18-9871-ef9f6db6735e\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>restaurant_reviews</th>\n",
              "      <th>categories</th>\n",
              "      <th>detect</th>\n",
              "      <th>sentiment</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Apparently Prides Osteria had a rough summer a...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>I was really disappointed to say the least. I ...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>This is as close to dining in Italy as you'll ...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Great food and service! Again. 4 out of the la...</td>\n",
              "      <td>Restaurants, Wine Bars, Nightlife, Farmers Mar...</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>I've stayed at many Marriott and Renaissance M...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>I was a there for a professional conference. s...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>I stayed for five nights this past April with ...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>I stayed here recently for a conference I was ...</td>\n",
              "      <td>Hotels, Hotels &amp; Travel, Restaurants, Event Pl...</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>The food is always great here. The service fro...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Brasserie zero zero... we were hoping for a te...</td>\n",
              "      <td>Restaurants, French</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8a4d602f-4551-4e18-9871-ef9f6db6735e')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-8a4d602f-4551-4e18-9871-ef9f6db6735e button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-8a4d602f-4551-4e18-9871-ef9f6db6735e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Let's see if the dataset is **balanced or not**: **it isn't...**"
      ],
      "metadata": {
        "id": "vQLKhukM60KG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
        "# See if the dataset is balanced or not: its isn't...\n",
        "plt.figure(figsize=(12,8))\n",
        "grouped = df_merged.sentiment.value_counts().sort_index()\n",
        "sns.barplot(grouped.index, grouped.values, palette=sns.color_palette(\"RdBu_r\", len(grouped)))\n",
        "plt.xlabel('Stars Ratings', labelpad=10, fontsize=14)\n",
        "plt.ylabel('Number of Reviews', fontsize=14)\n",
        "plt.title('Star Rating Distribution', fontsize=15)\n",
        "plt.tick_params(labelsize=14)\n",
        "for  i, v in enumerate(grouped):\n",
        "    plt.text(i, v*1.02, str(v), horizontalalignment ='center',fontweight='bold', fontsize=14)"
      ],
      "metadata": {
        "id": "72-rNYCEP6NP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 531
        },
        "outputId": "4c562a7c-61a8-4719-d0d3-ae69d5420017"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 747,
              "height": 514
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### To deal with **imbalanced** data we chose to **upsample** the **minority class** (negative sentiment - 0)"
      ],
      "metadata": {
        "id": "wEh_85bz7fn0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.utils import resample\n",
        "\n",
        "#create two different dataframes: one regarding the majority and the other the minority class \n",
        "df_majority = df_merged[(df_merged['sentiment']==1)] \n",
        "df_minority = df_merged[(df_merged['sentiment']==0)] \n",
        "\n",
        "\n",
        "df_minority_upsampled = resample(df_minority, \n",
        "                                 replace=True,   \n",
        "                                 n_samples= 4643, # to match majority class\n",
        "                                 random_state=42)  \n",
        "# Combine majority class with upsampled minority class\n",
        "df_final = pd.concat([df_minority_upsampled, df_majority])"
      ],
      "metadata": {
        "id": "QQd2OSEsQ3Cl"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_final.sentiment.value_counts()"
      ],
      "metadata": {
        "id": "2x4Ne31yQ3HB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2b11b471-924d-4b92-f916-25f7eb9592fc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0    4643\n",
              "1.0    4643\n",
              "Name: sentiment, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Stopwords** with **spaCy**. As **'no'** and **'not'** are **important** words to describe an opinion about a restaurant (e.g. not cheap, not a good service, no napkins available, etc.), we should, at least, **remove** those two words from the **stopwords**. "
      ],
      "metadata": {
        "id": "41D8LX1G96cr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- **Note**: using stopwords provides **context**, so when using a **contextual model** like **BERT** one **should usually keep** them to **provide** enough **context information**; however, in this case, when spliting the dataset into train, validation and test sets, we can see that for two of those three sets the **number of tokens** in some reviews (in one of the sets **more than 600** and in the other **more than 700**) **exceeds the maximum length possible** (512), being so we decided to **exclude the stopwords** with the **exception of the two words already mentioned**."
      ],
      "metadata": {
        "id": "x9fEA65Uio-p"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "sp = spacy.load('en_core_web_sm')\n",
        "stopwords = sp.Defaults.stop_words\n",
        "\n",
        "exclude_stopwords = ['no','not']\n",
        "for word in exclude_stopwords:\n",
        "    stopwords.remove(word)"
      ],
      "metadata": {
        "id": "etpLEPQhNIEt"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **UDF** to **remove non-letters**, convert to **lower case**, **tokenize** with **NLTK** and **remove** the **spaCy stopwords** from the restaurant reviews:"
      ],
      "metadata": {
        "id": "r-NyXi1FBx19"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# The input is a single string (a raw restaurant review), and the output is a single string (a preprocessed restaurant review)\n",
        "\n",
        "def text_preprocessing( raw_review ):\n",
        "    # 1. Remove non-letters \n",
        "    review_text_letters_only = re.sub(\"[^a-zA-Z]\", \" \", raw_review)\n",
        "    # 2. Convert to lower case\n",
        "    review_preprocessed = review_text_letters_only.lower()\n",
        "    # 3. Word tokenization\n",
        "    review_tokens = word_tokenize(review_preprocessed)\n",
        "    # 4. Filter the stopwords\n",
        "    filtered_sentence =[] \n",
        "    for word in review_tokens:\n",
        "        lexeme = sp.vocab[word]\n",
        "        if lexeme.is_stop == False:\n",
        "            filtered_sentence.append(word)\n",
        "\n",
        "    return \" \".join(filtered_sentence)"
      ],
      "metadata": {
        "id": "LP9Sw3OCWmRA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Apply** the above **UDF** to the **entire** 'restaurant_reviews' column, **creating** a **new column** named **'cleaned_reviews'**"
      ],
      "metadata": {
        "id": "3I8RB3bvEfWN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_final['cleaned_reviews'] = df_final['restaurant_reviews'].apply(text_preprocessing) "
      ],
      "metadata": {
        "id": "qXw3rx3Nv6-Z"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = df_final.reset_index(drop=True)"
      ],
      "metadata": {
        "id": "e-rhvEFtQ3NB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "id": "ErBRfSycQ3Tj",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "dd04cd5c-c307-408e-b03a-12461862182b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                  restaurant_reviews  \\\n",
              "0  There were maybe 8 tables in the restaurant,  ...   \n",
              "1  The food was good but Worst service I ever had...   \n",
              "2  Be aware the April server charge extra tip to ...   \n",
              "3  This place has really gone down hill on servic...   \n",
              "4  Placed order online for delivery, to be delive...   \n",
              "\n",
              "                                          categories detect  sentiment  \\\n",
              "0  Bars, Restaurants, Chicken Wings, Sports Bars,...     en        0.0   \n",
              "1                        Restaurants, American (New)     en        0.0   \n",
              "2  American (Traditional), Restaurants, Bars, Tex...     en        0.0   \n",
              "3             Restaurants, Cafes, Breakfast & Brunch     en        0.0   \n",
              "4      Sandwiches, Wraps, Chicken Wings, Restaurants     en        0.0   \n",
              "\n",
              "                                     cleaned_reviews  \n",
              "0  maybe tables restaurant waitress busy talking ...  \n",
              "1  food good worst service bad mean bad menus sil...  \n",
              "2       aware april server charge extra tip paid tip  \n",
              "3  place gone hill service waitress food prep del...  \n",
              "4  placed order online delivery delivered work br...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1a442ad7-e642-41fa-b01d-6d3f76841611\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>restaurant_reviews</th>\n",
              "      <th>categories</th>\n",
              "      <th>detect</th>\n",
              "      <th>sentiment</th>\n",
              "      <th>cleaned_reviews</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>There were maybe 8 tables in the restaurant,  ...</td>\n",
              "      <td>Bars, Restaurants, Chicken Wings, Sports Bars,...</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "      <td>maybe tables restaurant waitress busy talking ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>The food was good but Worst service I ever had...</td>\n",
              "      <td>Restaurants, American (New)</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "      <td>food good worst service bad mean bad menus sil...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Be aware the April server charge extra tip to ...</td>\n",
              "      <td>American (Traditional), Restaurants, Bars, Tex...</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "      <td>aware april server charge extra tip paid tip</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>This place has really gone down hill on servic...</td>\n",
              "      <td>Restaurants, Cafes, Breakfast &amp; Brunch</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "      <td>place gone hill service waitress food prep del...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Placed order online for delivery, to be delive...</td>\n",
              "      <td>Sandwiches, Wraps, Chicken Wings, Restaurants</td>\n",
              "      <td>en</td>\n",
              "      <td>0.0</td>\n",
              "      <td>placed order online delivery delivered work br...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1a442ad7-e642-41fa-b01d-6d3f76841611')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1a442ad7-e642-41fa-b01d-6d3f76841611 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1a442ad7-e642-41fa-b01d-6d3f76841611');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Sentiment analysis** of **YELP restaurant reviews** with **BERT** in **PyTorch**"
      ],
      "metadata": {
        "id": "xV0KQtpBOTKy"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7rbiGE67U8_G"
      },
      "source": [
        "#### Define the **pre-trained model** we are going to use: "
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- “**bert-base-uncased**” is the version that has only lowercase letters (“uncased”) and is the smaller version of the two (“base” vs “large”)."
      ],
      "metadata": {
        "id": "RX8ZgpeeHESm"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CtL66HBEVvDV"
      },
      "source": [
        "pre_trained_model='bert-base-uncased'"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "APA0dw47V8-n"
      },
      "source": [
        "#### &rarr; To **feed** our **reviews** to BERT they must be **split into tokens**, and then these tokens **must be mapped to their index in the tokenizer vocabulary**.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- When we **actually convert all of our reviews**, we’ll **use** the `tokenize.encode` function to **handle both steps**, rather than calling `tokenize` and `convert_tokens_to_ids` separately"
      ],
      "metadata": {
        "id": "KvyXBroW2eyY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- **Load** the **BERT Tokenizer**: "
      ],
      "metadata": {
        "id": "cQHN2abZlbTm"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fawk_k8xWWMp",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 113,
          "referenced_widgets": [
            "b545861dc6b245e991c0a05f00d04e9e",
            "eeaf38a6fb9d4ace9c1f39dc7e57352a",
            "eeec4665a4a4417dbb951a09e20c861f",
            "7370896e3de2432fb4164774cd785960",
            "ea07dd14f11e481794515c3ae2953723",
            "ab2aac51ab8b4d74a8c434dee44eabf8",
            "a255da9bcabd4711b6072a163e0f6e79",
            "aaeae2e29e504670a0ec8585d9813586",
            "ee145eab04a54d7e8a523fd3e0d6b1ca",
            "c16e1af626104cd5b325294025a8d7fe",
            "b35b5e26127e43dd93fbc1c0d0c131b2",
            "36f457298abe4c84983a6052ee30339b",
            "d05a8f73787e49e1b6cd1157271f9e4d",
            "cc0b19097489491f8945f63464ed8113",
            "327f3d42f4eb4dd7bdf8b0f0e328d902",
            "a9f02981d9824942a73bfd0d5bdcfc4a",
            "3d4cb26d7c4f4a94a0a93051ee9302c4",
            "a050f3a286d84767a60905e2af84b107",
            "c8205ff2db474720b4f6252680520304",
            "b02eade19f4b495097f561802dc2467d",
            "49bcfd1ac9ae432cba32b37c0cfca8c4",
            "a2fc8722a197488b976dbb3936523294",
            "51b723e776b84c2580afe0d913e997a5",
            "6dc451ada44d4cf58f15c93df8a24041",
            "b46b4dcada974413a56370a036c847bd",
            "6948b12686de42ecb19d8d679aae6fc8",
            "0722b2b5984e418baeff6db049675e04",
            "a416af20f5364545b7f88a62529d5925",
            "76a81285307b4b8cb7c88d5f623bd207",
            "ae681f89bf1d426ea28537ea8cd15445",
            "1802b3fec79d4571816260c49f0506fc",
            "37821a7c8b404e6e9257629f66000256",
            "49a5122de0ad414aa5a3340e11bf71c3"
          ]
        },
        "outputId": "c54bd21a-0dbd-405b-fc8d-9de27b363e0f"
      },
      "source": [
        "tokenizer=BertTokenizer.from_pretrained(pre_trained_model)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading vocab.txt:   0%|          | 0.00/226k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "b545861dc6b245e991c0a05f00d04e9e"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading tokenizer_config.json:   0%|          | 0.00/28.0 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "36f457298abe4c84983a6052ee30339b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading config.json:   0%|          | 0.00/570 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "51b723e776b84c2580afe0d913e997a5"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Before we are ready to **encode our reviews** we need to decide on a **maximum sentence length**. BERT **expects fixed-length sequences**, let's choose **maximum sequence length**:"
      ],
      "metadata": {
        "id": "kFrnIGUhRdk4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "token_lens = []\n",
        "\n",
        "for txt in df.cleaned_reviews:\n",
        "  tokens = tokenizer.encode(txt, max_length=512, truncation=True)\n",
        "  token_lens.append(len(tokens))"
      ],
      "metadata": {
        "id": "TkCoZv1JRkOX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sns.set(font_scale=1.4)\n",
        "plt.rcParams[\"figure.figsize\"] = (14,8)\n",
        "sns.histplot(token_lens)\n",
        "plt.xlim([0, 512])\n",
        "plt.xlabel('Number of Tokens')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 510
        },
        "id": "xroqeSVJ3Ij5",
        "outputId": "29c36250-6855-466d-e495-7f8ee577ea4c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1008x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 854,
              "height": 493
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### The **great majority** of the reviews seems to have **less than 240 tokens**, that's the **maximum length** that we'll choose."
      ],
      "metadata": {
        "id": "rjkq6uMRLpAj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "MAX_SEQ_LENGTH = 240"
      ],
      "metadata": {
        "id": "NqMRPl-fSOa4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zeMckzedWxWf"
      },
      "source": [
        "#### **Split** the **dataset** into **train** (70 %), **validation** (15%) and **test** (15%) sets and **preserve** the **same proportions** of **observations** in **each class** as **observed** in the **original dataset**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pkQsj5i6XPtc"
      },
      "source": [
        "x_train, x_test, y_train, y_test = train_test_split(df['cleaned_reviews'],df.sentiment, test_size=0.3, random_state = 42, stratify=df.sentiment)\n",
        "x_test, x_val, y_test, y_val = train_test_split(x_test, y_test, test_size=0.5, random_state = 42, stratify= y_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_train = y_train.astype(int)\n",
        "y_val = y_val.astype(int)\n",
        "y_test = y_test.astype(int)"
      ],
      "metadata": {
        "id": "tDD_DMLdf0LQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Convert to list each of the sets**"
      ],
      "metadata": {
        "id": "zDkd-rsKZGIr"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "30E9fQJlX3t5"
      },
      "source": [
        "\n",
        "train_reviews = x_train.tolist()\n",
        "val_reviews = x_val.tolist()\n",
        "test_reviews = x_test.tolist()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Create** a sequence of **token ids** (input ids) for **each review**"
      ],
      "metadata": {
        "id": "dXZggQIfRex9"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- **`tokenizer.encode`** converts a **string** in a sequence of **intput ids** (integers) using the **tokenizer** and the **vocabulary**, meaning, it **splits** a string **into tokens** and **map the tokens to their ids**.\n"
      ],
      "metadata": {
        "id": "hoqW6kKgRowX"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "-  We must **add special tokens** - `[CLS]` and `[SEP]` - for **classification** tasks."
      ],
      "metadata": {
        "id": "xUkF5M8U5puq"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kqZNVG4baRJ6"
      },
      "source": [
        "# Train dataset\n",
        "train_input_ids = [tokenizer.encode(train_reviews[i],add_special_tokens = True, max_length=MAX_SEQ_LENGTH, truncation=True) for i in range(0,len(train_reviews))]\n",
        "# Val dataset\n",
        "val_input_ids = [tokenizer.encode(val_reviews[i],add_special_tokens = True, max_length=MAX_SEQ_LENGTH, truncation=True) for i in range(0,len(val_reviews))]\n",
        "# Test dataset\n",
        "test_input_ids = [tokenizer.encode(test_reviews[i],add_special_tokens = True, max_length=MAX_SEQ_LENGTH, truncation=True) for i in range(0,len(test_reviews))]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SVXQcZcqbEk2"
      },
      "source": [
        "### **Padding** and **truncating**: pad and truncate our sequences so that they all have the **same length**:"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- **Padding**: when necessary, paddding wiil **add zeros** to make all the sequences **reach** the **length** of **MAX_SEQ_LENGTH**"
      ],
      "metadata": {
        "id": "UPeKatFqbxRS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- **Truncating**: when necessary, sequences **longer than the MAX_SEQ_LENGTH** are **truncated** to **MAX_SEQ_LENGTH**"
      ],
      "metadata": {
        "id": "k2AnzMD4cSSI"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "- **Note**: '**post**' specifies that we want to pad and truncate at the **end** of the sequence. "
      ],
      "metadata": {
        "id": "XrAfaS_08QAS"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DjlWdSqrcI5n"
      },
      "source": [
        "from keras.preprocessing.sequence import pad_sequences    # Pad utility function to pad sequences to maximum length.\n",
        " \n",
        "# Padding value: is optional, the default is 0.\n",
        "\n",
        "# Train dataset\n",
        "train_input_ids = pad_sequences(train_input_ids, maxlen=MAX_SEQ_LENGTH, dtype=\"long\", \n",
        "                          value=0, truncating=\"post\", padding=\"post\")\n",
        "\n",
        "# Validation dataset\n",
        "val_input_ids = pad_sequences(val_input_ids, maxlen=MAX_SEQ_LENGTH, dtype=\"long\", \n",
        "                          value=0, truncating=\"post\", padding=\"post\")\n",
        "\n",
        "# Test dataset\n",
        "test_input_ids = pad_sequences(test_input_ids, maxlen=MAX_SEQ_LENGTH, dtype=\"long\", \n",
        "                          value=0, truncating=\"post\", padding=\"post\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ttcFfNmieXJC"
      },
      "source": [
        "### **Attention Masks**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-fz6Ek8RebPH"
      },
      "source": [
        "The attention masks **differentiate** which tokens are **`real`** tokens versus which are **`[PAD]`** tokens."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vz1yKGqneiVe"
      },
      "source": [
        "# Create attention masks\n",
        "\n",
        "# Train dataset\n",
        "train_attention_masks = [[int(token_id > 0) for token_id in review]\n",
        "                         for review in train_input_ids]\n",
        "# dev dataset\n",
        "val_attention_masks = [[int(token_id > 0) for token_id in review]\n",
        "                         for review in val_input_ids]\n",
        "# Test dataset\n",
        "test_attention_masks = [[int(token_id > 0) for token_id in review]\n",
        "                         for review in test_input_ids]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Convert all lists** of **inputs ids**, **labels** and **attention masks** into **torch tensors**"
      ],
      "metadata": {
        "id": "0eKL-j51gqPY"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GXES16EAe6i3"
      },
      "source": [
        "# input_ids\n",
        "train_inputs = torch.tensor(train_input_ids)\n",
        "val_inputs = torch.tensor(val_input_ids)\n",
        "test_inputs = torch.tensor(test_input_ids)\n",
        "# labels\n",
        "train_labels = torch.tensor(y_train.values)\n",
        "val_labels = torch.tensor(y_val.values)\n",
        "test_labels = torch.tensor(y_test.values)\n",
        "# attention masks\n",
        "train_masks = torch.tensor(train_attention_masks)\n",
        "val_masks = torch.tensor(val_attention_masks)\n",
        "test_masks = torch.tensor(test_attention_masks)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Create a **`DataLoader`** to **load** our **datasets**. Serves as an **iterator**, it **saves memory during training** when compared with a **for loop**"
      ],
      "metadata": {
        "id": "NMR3XX_oriqj"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hl0vk6SofwmZ"
      },
      "source": [
        "batch_size = 16\n",
        "\n",
        "# Create the DataLoader for our training set.\n",
        "train_data = TensorDataset(train_inputs, train_masks, train_labels)\n",
        "train_sampler = RandomSampler(train_data)\n",
        "train_dataloader = DataLoader(train_data, sampler=train_sampler, batch_size=batch_size)\n",
        "\n",
        "# Create the DataLoader for our validation set.\n",
        "val_data = TensorDataset(val_inputs, val_masks, val_labels)\n",
        "val_sampler = SequentialSampler(val_data)\n",
        "val_dataloader = DataLoader(val_data, sampler=val_sampler, batch_size=batch_size)\n",
        "\n",
        "# Create the DataLoader for our test set.\n",
        "test_data = TensorDataset(test_inputs, test_masks, test_labels)\n",
        "test_sampler = SequentialSampler(test_data)\n",
        "test_dataloader = DataLoader(test_data, sampler=test_sampler, batch_size=batch_size)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h4wDCYZglTYE",
        "outputId": "84d197c5-0cca-4ba8-a2ee-7acf96818568"
      },
      "source": [
        "# Number of classes / labels\n",
        "n_classes = y_train.nunique()\n",
        "n_classes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2"
            ]
          },
          "metadata": {},
          "execution_count": 47
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5tMzrfwsgKM_"
      },
      "source": [
        "## **Build a sentiment Classifier**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m6KUN7ifgPwb"
      },
      "source": [
        "### **BertForSequenceClassification**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "smbZEZuuk-zW"
      },
      "source": [
        "- We'll use the **BertForSequenceClassification** model for our classification task. It's a **Bert Model transformer** with an **added layer on top of it** for **classification**."
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- We are using **Pytorch framework** for model **training** and **evaluation** for **classification** purposes, hence we have chosen **BertForSequeunceClassification class** to be used."
      ],
      "metadata": {
        "id": "SsIb24k-h7yL"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Instantiate** our model:"
      ],
      "metadata": {
        "id": "Po_IU2jZ-08u"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = BertForSequenceClassification.from_pretrained(\n",
        "    \"bert-base-uncased\", # The 12-layer BERT model with an uncased vocab\n",
        "    num_labels = 2, # For binary classification  \n",
        "    output_attentions = False, # Not to return attentions weights\n",
        "    output_hidden_states = False, # Not to return all hidden-states\n",
        ")\n"
      ],
      "metadata": {
        "id": "tjHuh0twIQn3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 156,
          "referenced_widgets": [
            "7672cb2f16c14e69b2975d713732e9ba",
            "ddc8ba8db8874dfd8d403e140d7d8e7a",
            "a9c906bed63b4b218af63d9a9a81cff5",
            "84f21df2d3464c1688d4f19d5569b9e4",
            "b118fd54f78342548f440463ab25aa67",
            "87c6aee571d14cf6bf3ed0550ecc3109",
            "f29abffd2398473e8f10f696f06edb5c",
            "f8455c5dddfd433088c032c358a9bf8e",
            "8f50a041901d475aa732e043f3c5c1aa",
            "7f5dcd1edd9948388566068a0d72ab81",
            "afcce7f86eb8470a8cd402108059eabf"
          ]
        },
        "outputId": "e50e3e3a-578e-4ac6-a781-e134dc808c6e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading pytorch_model.bin:   0%|          | 0.00/420M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "7672cb2f16c14e69b2975d713732e9ba"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForSequenceClassification: ['cls.predictions.transform.LayerNorm.bias', 'cls.predictions.decoder.weight', 'cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.seq_relationship.weight', 'cls.predictions.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight']\n",
            "- This IS expected if you are initializing BertForSequenceClassification 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 BertForSequenceClassification 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 BertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier.bias', 'classifier.weight']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zlDD0ekkldqe"
      },
      "source": [
        "#### **Move** our model to the **GPU**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5r2GuQ59lhkV"
      },
      "source": [
        "model = model.to(device) "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### To **fine-tune** our Bert Classifier we need to create an **`optimizer`**. The authors have some **recommendations** for **fine-tuning**:\n",
        "\n",
        "- **Batch size**: 16, 32\n",
        "- **Learning rate** (Adam): 5e-5, 3e-5, 2e-5\n",
        "- **Number of epochs**: 2, 3 or 4\n",
        "\n",
        "We chose **16**, **3e-5** and **2**, respectively.\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "4hLzsVyjJa4l"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CDG0P_q5lrOI"
      },
      "source": [
        "### **Optimizer**, **learning rate scheduler**, **loss function** and **number of epochs**\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Notes**:\n",
        "\n",
        "- To **fine-tune** our Bert classifier we need to create an **`optimizer`**. Considering the **original paper** on **BERT**, we will use <a href=\"https://huggingface.co/transformers/main_classes/optimizer_schedules.html\" target=\"_blank\">AdamW</a> **optimizer** because it implements **gradient bias correction** as well as **weight decay**. We will also use **linear scheduler with no warmup steps**.\n",
        "\n",
        "- The authors have some recommendations for fine-tuning:\n",
        "\n",
        "    - Batch size: 16, 32\n",
        "    - Learning rate (Adam): 5e-5, 3e-5, 2e-5\n",
        "    - Number of epochs: 2, 3 or 4\n",
        "\n",
        "We chose **16**, **3e-5** and **2**, respectively.\n",
        "<br><br>"
      ],
      "metadata": {
        "id": "GYUHBjxNbrZ_"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3RxQoF0wl6k9"
      },
      "source": [
        "epochs=2\n",
        "\n",
        "optimizer=AdamW(model.parameters(),lr=3e-5)\n",
        "total_steps=len(train_dataloader)*epochs\n",
        "\n",
        "# Create the learning rate scheduler\n",
        "scheduler=get_linear_schedule_with_warmup(\n",
        "    optimizer,\n",
        "    num_warmup_steps=0,\n",
        "    num_training_steps=total_steps\n",
        ")\n",
        " # Define loss function and move it to GPU\n",
        "loss_fn=nn.CrossEntropyLoss().to(device)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WJSxizPcmCZJ"
      },
      "source": [
        "## **Training** our model and **validation**: "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hc-diW99mG-W"
      },
      "source": [
        "#### Function for calculating **elapsed time**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RFFefySQmLGi"
      },
      "source": [
        "def format_time(elapsed):\n",
        "    # Round to the nearest second\n",
        "    elapsed_round = int(round(elapsed))\n",
        "    # Format time in hh:mm:ss\n",
        "    return str(datetime.timedelta(seconds = elapsed_round))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ox9L0ZHymRAL"
      },
      "source": [
        "#### Function for calculating **accuracy**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1EjrDj2wmVbY"
      },
      "source": [
        "def accuracy(preds, labels):\n",
        "    preds = np.argmax(preds, axis=1).flatten()\n",
        "    labels = labels.flatten()\n",
        "    return np.sum(preds == labels) / len(labels)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4pc_F-LjmbX6"
      },
      "source": [
        "### **Training** and **validation**\n",
        "- We can **store** the **training** and **validation loss** and **accuracy values** and then **plot** or **make a table** to **measure the performance** on both the **train** and **validation** sets **after the completion of each epoch** so we can see **more clearly** the **progress** of the **training loop**.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Store for each epoch\n",
        "loss_train_values = []\n",
        "acc_train_values = []\n",
        "loss_val_values = []\n",
        "acc_val_values = []\n",
        "for epoch in range(0,epochs):\n",
        "     \n",
        "        #             --- Train---\n",
        "        \n",
        "        # Perform forward pass over the training dataset\n",
        "        print(\"\\n Epoch {:}/{:} :\".format(epoch+1,epochs))\n",
        "        print('Training....')\n",
        "        \n",
        "        # Measure how long the training epoch takes\n",
        "        t0 = time.time()\n",
        "        # Reset total loss and accuracy for this epoch\n",
        "        total_loss = 0\n",
        "        total_acc = 0\n",
        "\n",
        "        # Put the model in training mode\n",
        "        model.train()\n",
        "        \n",
        "        # For each batch of training data\n",
        "        for step, batch in enumerate(train_dataloader):\n",
        "            # Update progress for each 100 steps\n",
        "            if (step % 100==0) & (not step==0):\n",
        "                # Calculate elapsed time in minutes\n",
        "                elapsed = format_time((time.time()-t0))\n",
        "                # Report progress\n",
        "                print(' Batch {:>5,} of {:>5,}. Elapsed:{:}.'.format(step,len(train_dataloader),elapsed))\n",
        "            \n",
        "            # Unpack training batch from trainloader and move to GPU\n",
        "            b_input_ids = batch[0].to(device)  # 0 - input ids tensor\n",
        "            b_attention_mask = batch[1].to(device) # 1 - input masks tensor\n",
        "            b_labels = batch[2].to(device) # 2 - labels tensor\n",
        "            \n",
        "            # Clear any previously calculated gradients in Pytorch before performing a backward pass\n",
        "            model.zero_grad()\n",
        "            \n",
        "            # Output the results\n",
        "            outputs = model(input_ids = b_input_ids, attention_mask = b_attention_mask, labels=b_labels) # Return tuple\n",
        "            # Loss value from output\n",
        "            loss = outputs.loss   # Loss\n",
        "            \n",
        "            # Update total loss\n",
        "            total_loss += loss.item()\n",
        "            \n",
        "            preds = outputs.logits  # Output probabilities\n",
        "            # Move logits and labels to CPU\n",
        "            preds = preds.detach().cpu().numpy()\n",
        "            label_ids = b_labels.to('cpu').numpy()\n",
        "\n",
        "            # Calculate the accuracy for this batch \n",
        "            tmp_train_accuracy = accuracy(preds, label_ids)\n",
        "\n",
        "            # Accumulate the total accuracy\n",
        "            total_acc += tmp_train_accuracy\n",
        "            \n",
        "            # Perform a backward pass to calculate gradients\n",
        "            loss.backward()\n",
        "            \n",
        "            # To avoid exploding vanishing gradients problem, clip the norm of the gradients to 1.0\n",
        "            torch.nn.utils.clip_grad_norm_(model.parameters(),1.0)\n",
        "            \n",
        "            # Update the parameters (weights)\n",
        "            optimizer.step()\n",
        "            \n",
        "            # Update the learning rate\n",
        "            scheduler.step()\n",
        "            \n",
        "        # Calculate the average loss over training data\n",
        "        avg_total_loss = total_loss/len(train_dataloader)\n",
        "            \n",
        "        # Store the loss values\n",
        "        loss_train_values.append(avg_total_loss)\n",
        "\n",
        "        # Calculate the average accuracy over the training data\n",
        "        avg_train_acc = total_acc / len(train_dataloader)\n",
        "\n",
        "        # Store the accuracy values \n",
        "        acc_train_values.append(avg_train_acc)\n",
        "\n",
        "        print(\"\")\n",
        "        print(\"\\nAverage training accuracy: {0:.2f}\".format(avg_train_acc))\n",
        "        \n",
        "        print('Average training loss : {0:.2f}'.format(avg_total_loss))\n",
        "        print('Training epoch took: {:}'.format(format_time(time.time()- t0)))\n",
        " \n",
        "        #             --- VALIDATION ---\n",
        "\n",
        "        \n",
        "        # After each epoch perform validation to check model performance\n",
        "        print('\\n Running validation...')\n",
        "        \n",
        "        t0 = time.time()\n",
        "        # Put model in evaluation mode\n",
        "        model.eval()\n",
        "        \n",
        "        # Tracking variables\n",
        "        total_eval_accuracy = 0\n",
        "        total_eval_loss = 0\n",
        "        \n",
        "        # Unpack validation batch from trainloader and move to GPU\n",
        "        for batch in val_dataloader: \n",
        "            b_input_ids = batch[0].to(device)\n",
        "            b_input_mask = batch[1].to(device)\n",
        "            b_labels = batch[2].to(device)\n",
        "\n",
        "            # Tell model not to compute gradients to save memory and accelerate validation\n",
        "            with torch.no_grad():\n",
        "                # Forward pass, calculate logit prediction\n",
        "                 outputs = model(b_input_ids, \n",
        "                                   token_type_ids=None, \n",
        "                                   attention_mask=b_input_mask,\n",
        "                                   labels=b_labels)\n",
        "            loss = outputs.loss\n",
        "            logits = outputs.logits\n",
        "            # Update total evaluation loss\n",
        "            total_eval_loss += loss.item()\n",
        "            \n",
        "            # Move logits and labels to CPU\n",
        "            logits = logits.detach().cpu().numpy()\n",
        "            label_ids = b_labels.to('cpu').numpy()\n",
        "\n",
        "            # Calculate the accuracy for this batch and accumulate it over all batches\n",
        "            total_eval_accuracy += accuracy(logits, label_ids)\n",
        "\n",
        "        # Compute the average accuracy over all of the batches\n",
        "        avg_val_accuracy = total_eval_accuracy / len(val_dataloader)\n",
        "\n",
        "        # Store the accuracy values \n",
        "        acc_val_values.append(avg_val_accuracy)\n",
        "     \n",
        "        # Compute the average loss over all of the batches\n",
        "        avg_val_loss = total_eval_loss / len(val_dataloader)\n",
        "\n",
        "         # Store the loss values\n",
        "        loss_val_values.append(avg_val_loss)\n",
        "    \n",
        "    # Measure how long the validation run took.\n",
        "        validation_time = format_time(time.time() - t0)\n",
        "        print(\"  Accuracy: {0:.2f}\".format(avg_val_accuracy))\n",
        "        print(\"  Validation Loss: {0:.2f}\".format(avg_val_loss))\n",
        "        print(\"  Validation took: {:}\".format(validation_time))\n",
        "\n"
      ],
      "metadata": {
        "id": "RHdh10SBOK2G",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d4b2ab51-266c-4193-91b6-7d2ad4b296c3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Epoch 1/2 :\n",
            "Training....\n",
            " Batch   100 of   407. Elapsed:0:01:00.\n",
            " Batch   200 of   407. Elapsed:0:02:01.\n",
            " Batch   300 of   407. Elapsed:0:03:02.\n",
            " Batch   400 of   407. Elapsed:0:04:05.\n",
            "\n",
            "\n",
            "Average training accuracy: 0.90\n",
            "Average training loss : 0.26\n",
            "Training epoch took: 0:04:09\n",
            "\n",
            " Running validation...\n",
            "  Accuracy: 0.96\n",
            "  Validation Loss: 0.16\n",
            "  Validation took: 0:00:19\n",
            "\n",
            " Epoch 2/2 :\n",
            "Training....\n",
            " Batch   100 of   407. Elapsed:0:01:02.\n",
            " Batch   200 of   407. Elapsed:0:02:05.\n",
            " Batch   300 of   407. Elapsed:0:03:07.\n",
            " Batch   400 of   407. Elapsed:0:04:10.\n",
            "\n",
            "\n",
            "Average training accuracy: 0.98\n",
            "Average training loss : 0.08\n",
            "Training epoch took: 0:04:14\n",
            "\n",
            " Running validation...\n",
            "  Accuracy: 0.98\n",
            "  Validation Loss: 0.13\n",
            "  Validation took: 0:00:19\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Plot validation accuracy after** the **completion** of **each epoch**:"
      ],
      "metadata": {
        "id": "bX4buD-4ZIma"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_acc = pd.DataFrame(acc_val_values,columns=['Accuracy'])"
      ],
      "metadata": {
        "id": "lionMp31Br34"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_acc.index+=1"
      ],
      "metadata": {
        "id": "QdMLpEG-FQgT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Use plot styling from seaborn\n",
        "sns.set(style='darkgrid')\n",
        "\n",
        "# Increase the plot size and font size\n",
        "sns.set(font_scale=1.5)\n",
        "plt.rcParams[\"figure.figsize\"] = (12,6)\n",
        "\n",
        "# Plot the learning curve\n",
        "sns.lineplot(data=df_acc,x=df_acc.index,y=df_acc.Accuracy)\n",
        "\n",
        "# Label the plot\n",
        "plt.title(\"Validation Accuracy\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "aCx69_8WFbLh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "outputId": "a1160c7d-c99a-4c15-90da-843eb0fc2520"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 772,
              "height": 407
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Plot validation loss after** the **completion** of **each epoch**:"
      ],
      "metadata": {
        "id": "KkYErggNFInS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_loss = pd.DataFrame(loss_val_values,columns=['Loss'])"
      ],
      "metadata": {
        "id": "hlvswohkDXiA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_loss.index+=1"
      ],
      "metadata": {
        "id": "8Bdbg5JjDr5y"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Use plot styling from seaborn\n",
        "sns.set(style='darkgrid')\n",
        "\n",
        "# Increase the plot size and font size\n",
        "sns.set(font_scale=1.5)\n",
        "plt.rcParams[\"figure.figsize\"] = (12,6)\n",
        "\n",
        "# Plot the learning curve\n",
        "sns.lineplot(data=df_loss,x=df_loss.index,y=df_loss.Loss)\n",
        "\n",
        "# Label the plot\n",
        "plt.title(\"Validation Loss\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Loss\")\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "Zq9jViwREpDP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "outputId": "34e369ea-67e4-4815-8a49-bbffc52cbc1a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 761,
              "height": 407
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Predicting** on **test set**\n"
      ],
      "metadata": {
        "id": "zr-L-TMi2vQH"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_g_rCTZVDLOH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2972f418-f77d-4988-fe1c-42b8012ccd57"
      },
      "source": [
        "# Put model in evaluation mode\n",
        "model.eval()\n",
        "\n",
        "# Tracking variables\n",
        "predictions , true_labels = [], []\n",
        "\n",
        "# Predict test data\n",
        "for batch in test_dataloader:\n",
        "    # Move batch to GPU\n",
        "    batch = tuple(t.to(device) for t in batch)\n",
        "    # Unpack inputs from test dataloader\n",
        "    b_input_ids, b_attention_mask, b_labels = batch\n",
        "    # Tell model not to compute gradients to save memory and accelerate validation\n",
        "    with torch.no_grad():\n",
        "        # Forward pass, calculate logit prediction\n",
        "        outputs = model(input_ids=b_input_ids, attention_mask=b_attention_mask)\n",
        "    # logits are class probabilities and get them from outputs\n",
        "    logits = outputs[0]\n",
        "    \n",
        "    # Store predictions and true labels\n",
        "    predictions.extend(logits.tolist())\n",
        "    true_labels.extend(b_labels.tolist())\n",
        "\n",
        "print('Done')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Done\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Convert predictions** to **tensor**"
      ],
      "metadata": {
        "id": "srvvMSQIGGH6"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YEq4XpTfQAEF"
      },
      "source": [
        "preds = torch.tensor(predictions)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Apply softmax** on **predictions**"
      ],
      "metadata": {
        "id": "v_yrEDk8GWvL"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "luEXIxxrPe2U"
      },
      "source": [
        "import torch.nn.functional as F\n",
        "\n",
        "preds = F.softmax(preds,dim=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Convert to numpy arrays**"
      ],
      "metadata": {
        "id": "UNE9PrjaGutY"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HHAGs51ORM57"
      },
      "source": [
        "preds = np.array(preds)\n",
        "true_labels = np.array(true_labels)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### **Function** for **Heatmap of confusion matrix** for **test data**"
      ],
      "metadata": {
        "id": "Wi6s7Ra-HKYC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def evaluate(y_test, predictions):\n",
        "    cf_matrix = confusion_matrix(true_labels, preds.argmax(1))\n",
        "    sns.heatmap(cf_matrix, annot = True, fmt = 'd',cmap=\"Blues\")\n",
        "    plt.title('Heatmap of confusion matrix for Test data')\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "evaluate(true_labels, preds.argmax(1))"
      ],
      "metadata": {
        "id": "04a6IKMTPmxa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "outputId": "ccaca32e-3936-4818-c2b8-1786a8ce6acd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "image/png": {
              "width": 687,
              "height": 407
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6BzhJEsOSzOI"
      },
      "source": [
        "#### **Classification Report**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S4VGNcivS0uw",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "60bc9165-e989-4455-c57f-532517362ad1"
      },
      "source": [
        "class_report= classification_report(true_labels, preds.argmax(1), digits=3)\n",
        "print(class_report)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0      0.970     0.987     0.979       696\n",
            "           1      0.987     0.970     0.978       697\n",
            "\n",
            "    accuracy                          0.978      1393\n",
            "   macro avg      0.979     0.978     0.978      1393\n",
            "weighted avg      0.979     0.978     0.978      1393\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Conclusions**:\n",
        "- Our model seems to **generalize well** because the **accuracy on the test set** is **very similar** to the **accuracy on the validation set** (even higher: **97.8 > 97.5%**).\n",
        "- When **comparing** the **1st** and **2nd epochs**, **validation loss decreases** which **indicates** that **we aren't over-fitting on the training data**.\n",
        "- The model achieves a **high F1 Scores**: **≈ 97.8% / 97.9%**.\n",
        "- It is possible that **if we didn't remove the stopwords** we would **achieve a slight increase in F1 Score**. However, as we previously explained, **we chose not to include the stopwords** (with the exception of 'not' and 'no') **due to the reason mentioned above**. We **can't have the best of both worlds!**\n"
      ],
      "metadata": {
        "id": "wjIQkhFVtWt-"
      }
    }
  ]
}