LSTM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "LSTM.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyMamcEia7iAHPl06ndptsHN",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "666e1120eae54988a88c1814e1f22e93": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_27f04b9c43de464aaa042af00627dead",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_77802c885023427b8b7035408db82756",
              "IPY_MODEL_a6721f2a413f450b9917514ae89bf9e5"
            ]
          }
        },
        "27f04b9c43de464aaa042af00627dead": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "77802c885023427b8b7035408db82756": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_e0f0689f327544fbb3345f90c40b46e1",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 433,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 433,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_6006fb99fe744b45bcac6abc5b496b32"
          }
        },
        "a6721f2a413f450b9917514ae89bf9e5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_0afce616673e4a25af309d8ee61ceb5a",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 433/433 [00:00&lt;00:00, 14.3kB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_fc86ecbf47294e32b0fdba7555e36478"
          }
        },
        "e0f0689f327544fbb3345f90c40b46e1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "6006fb99fe744b45bcac6abc5b496b32": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "0afce616673e4a25af309d8ee61ceb5a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "fc86ecbf47294e32b0fdba7555e36478": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "4762d2208ec24134ac1d7b2d8193b531": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_b3cf5198771f4c298077ea268b01f1dd",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_b98a8ded08ae4fc29e483432ee4e323a",
              "IPY_MODEL_770d5310f55a485490b576a57e96c35e"
            ]
          }
        },
        "b3cf5198771f4c298077ea268b01f1dd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "b98a8ded08ae4fc29e483432ee4e323a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_a7f17204c6064ee0b1953f9d9b709b40",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 536063208,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 536063208,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_25c8a02ee5f44f4abd8e4c58b80b21b1"
          }
        },
        "770d5310f55a485490b576a57e96c35e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_b95989337cdd4a14a490899ca4db244f",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 536M/536M [00:06&lt;00:00, 81.6MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0f6d58e2b384474280539978a5504ae2"
          }
        },
        "a7f17204c6064ee0b1953f9d9b709b40": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "25c8a02ee5f44f4abd8e4c58b80b21b1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "b95989337cdd4a14a490899ca4db244f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "0f6d58e2b384474280539978a5504ae2": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "f9538fa35ec74011a9ab18be4653f3e8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_6a0fb8a9a26848f6867422ee6145826b",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_076f2b249be747f394f88e9ab8420d89",
              "IPY_MODEL_31bb9c8f47dd49a99a221f8a6017e99d"
            ]
          }
        },
        "6a0fb8a9a26848f6867422ee6145826b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "076f2b249be747f394f88e9ab8420d89": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_5136ecf2eb614761a8afbaa79211d76d",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 231508,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 231508,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_958cc98efe0f492da59bd2cd1886377f"
          }
        },
        "31bb9c8f47dd49a99a221f8a6017e99d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_c1777493a4094c889e68e5d2a2073ea7",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 232k/232k [00:00&lt;00:00, 3.30MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_5d7a9114cb3b4ed89f0b30c529f7fb8a"
          }
        },
        "5136ecf2eb614761a8afbaa79211d76d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "958cc98efe0f492da59bd2cd1886377f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "c1777493a4094c889e68e5d2a2073ea7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "5d7a9114cb3b4ed89f0b30c529f7fb8a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "cf5c9afee54f4b1a923a731f5677d446": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_1a63b4343824434e95a38a79e3c289f0",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_f3d7a20ea7c94eb4a1c8d3ace431149f",
              "IPY_MODEL_d4b85706b3ef48578ce2a50cf3e6aac6"
            ]
          }
        },
        "1a63b4343824434e95a38a79e3c289f0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "f3d7a20ea7c94eb4a1c8d3ace431149f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_43214e2d0a4544c69e9a7970e2303c3d",
            "_dom_classes": [],
            "description": "100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 2720,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 2720,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_8c5e8ea08b8540e09a9bc5bce721b196"
          }
        },
        "d4b85706b3ef48578ce2a50cf3e6aac6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_43db2f1055e048658e7a6a0772c7aaf1",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 2720/2720 [29:35&lt;00:00,  1.53it/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_4b82ab19a4944daaa68e0c642464d9cf"
          }
        },
        "43214e2d0a4544c69e9a7970e2303c3d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "8c5e8ea08b8540e09a9bc5bce721b196": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "43db2f1055e048658e7a6a0772c7aaf1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "4b82ab19a4944daaa68e0c642464d9cf": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/sydney-machine-learning/sentimentanalysis-USelections/blob/main/LSTM.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lOePlIwuLGNf"
      },
      "source": [
        "# Sentiment Analysis using LSTM and Keras"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5Dyl3HLpLB70"
      },
      "source": [
        "import pandas as pd    # to load dataset\n",
        "import numpy as np     # for mathematic equation\n",
        "from nltk.corpus import stopwords   # to get collection of stopwords\n",
        "from sklearn.model_selection import train_test_split       # for splitting dataset\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer  # to encode text to int\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences   # to do padding or truncating\n",
        "from tensorflow.keras.models import Sequential     # the model\n",
        "from tensorflow.keras.layers import Embedding, LSTM, Dense # layers of the architecture\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint   # save model\n",
        "from tensorflow.keras.models import load_model   # load saved model\n",
        "import re\n",
        "import tensorflow as tf\n",
        "import pandas as pd"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aHBrlQUoM9Gd",
        "outputId": "5ad75268-7841-4b1d-8698-7bd3d82d6ccc"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JBYdzfe0NLNG",
        "outputId": "672415ee-c078-449a-b398-2769a5184d94"
      },
      "source": [
        "data = pd.read_csv('/content/drive/MyDrive/US_elections/LSTM/IMDB Dataset.csv')\n",
        "\n",
        "print(data)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                                                  review sentiment\n",
            "0      One of the other reviewers has mentioned that ...  positive\n",
            "1      A wonderful little production. <br /><br />The...  positive\n",
            "2      I thought this was a wonderful way to spend ti...  positive\n",
            "3      Basically there's a family where a little boy ...  negative\n",
            "4      Petter Mattei's \"Love in the Time of Money\" is...  positive\n",
            "...                                                  ...       ...\n",
            "49995  I thought this movie did a down right good job...  positive\n",
            "49996  Bad plot, bad dialogue, bad acting, idiotic di...  negative\n",
            "49997  I am a Catholic taught in parochial elementary...  negative\n",
            "49998  I'm going to have to disagree with the previou...  negative\n",
            "49999  No one expects the Star Trek movies to be high...  negative\n",
            "\n",
            "[50000 rows x 2 columns]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IQvmimbaNT6O",
        "outputId": "2874840c-12c3-474a-b614-62fc263129c1"
      },
      "source": [
        "\n",
        "import nltk\n",
        "nltk.download('stopwords')\n",
        "english_stops = set(stopwords.words('english'))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "apZjZEDaNcPg"
      },
      "source": [
        "## Load and Clean Dataset\n",
        "In the original dataset, the reviews are still dirty. There are still html tags, numbers, uppercase, and punctuations. This will not be good for training, so in load_dataset() function, beside loading the dataset using pandas, I also pre-process the reviews by removing html tags, non alphabet (punctuations and numbers), stop words, and lower case all of the reviews.\n",
        "\n",
        "## Encode Sentiments\n",
        "In the same function, I also encode the sentiments into integers (0 and 1). Where 0 is for negative sentiments and 1 is for positive sentiments."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d8xYPJ3DNWZP",
        "outputId": "6bd2d9c1-bac8-4ad2-c72b-d73e0ad4c5f7"
      },
      "source": [
        "def load_dataset():\n",
        "    df = pd.read_csv('/content/drive/MyDrive/US_elections/LSTM/IMDB Dataset.csv')\n",
        "    x_data = df['review']       # Reviews/Input\n",
        "    y_data = df['sentiment']    # Sentiment/Output\n",
        "\n",
        "    # PRE-PROCESS REVIEW\n",
        "    x_data = x_data.replace({'<.*?>': ''}, regex = True)          # remove html tag\n",
        "    x_data = x_data.replace({'[^A-Za-z]': ' '}, regex = True)     # remove non alphabet\n",
        "    x_data = x_data.apply(lambda review: [w for w in review.split() if w not in english_stops])  # remove stop words\n",
        "    x_data = x_data.apply(lambda review: [w.lower() for w in review])   # lower case\n",
        "    \n",
        "    # ENCODE SENTIMENT -> 0 & 1\n",
        "    y_data = y_data.replace('positive', 1)\n",
        "    y_data = y_data.replace('negative', 0)\n",
        "\n",
        "    return x_data, y_data\n",
        "\n",
        "x_data, y_data = load_dataset()\n",
        "\n",
        "print('Reviews')\n",
        "print(x_data, '\\n')\n",
        "print('Sentiment')\n",
        "print(y_data)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Reviews\n",
            "0        [one, reviewers, mentioned, watching, oz, epis...\n",
            "1        [a, wonderful, little, production, the, filmin...\n",
            "2        [i, thought, wonderful, way, spend, time, hot,...\n",
            "3        [basically, family, little, boy, jake, thinks,...\n",
            "4        [petter, mattei, love, time, money, visually, ...\n",
            "                               ...                        \n",
            "49995    [i, thought, movie, right, good, job, it, crea...\n",
            "49996    [bad, plot, bad, dialogue, bad, acting, idioti...\n",
            "49997    [i, catholic, taught, parochial, elementary, s...\n",
            "49998    [i, going, disagree, previous, comment, side, ...\n",
            "49999    [no, one, expects, star, trek, movies, high, a...\n",
            "Name: review, Length: 50000, dtype: object \n",
            "\n",
            "Sentiment\n",
            "0        1\n",
            "1        1\n",
            "2        1\n",
            "3        0\n",
            "4        1\n",
            "        ..\n",
            "49995    1\n",
            "49996    0\n",
            "49997    0\n",
            "49998    0\n",
            "49999    0\n",
            "Name: sentiment, Length: 50000, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GLZ24hlbNuik"
      },
      "source": [
        "## Split Dataset\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Rb4AMEZONnHg",
        "outputId": "71e85b40-1952-429a-97d4-c0decadf6df2"
      },
      "source": [
        "x_train, x_test, y_train, y_test = train_test_split(x_data, y_data, test_size = 0.2)\n",
        "\n",
        "print('Train Set')\n",
        "print(x_train, '\\n')\n",
        "print(x_test, '\\n')\n",
        "print('Test Set')\n",
        "print(y_train, '\\n')\n",
        "print(y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train Set\n",
            "33282    [i, know, slashers, always, supposed, bad, com...\n",
            "38418    [this, movie, bad, i, know, begin, apparently,...\n",
            "14178    [thanks, killing, franchise, turkey, john, car...\n",
            "33857    [movie, made, watch, paul, w, s, anderson, avp...\n",
            "43719    [i, think, probably, one, worst, movies, i, wa...\n",
            "                               ...                        \n",
            "26229    [this, picture, hit, movie, screens, june, th,...\n",
            "38580    [the, best, thing, camp, films, general, know,...\n",
            "4497     [as, long, go, movie, understanding, going, co...\n",
            "7269     [this, movie, stupid, hate, turned, reached, h...\n",
            "9957     [i, actually, went, see, movie, low, expectati...\n",
            "Name: review, Length: 40000, dtype: object \n",
            "\n",
            "18524    [no, lifeforce, great, movie, it, good, movie,...\n",
            "46492    [unfortunately, one, best, efforts, yet, made,...\n",
            "12398    [i, love, called, blaxploitation, films, seen,...\n",
            "34083    [i, watched, plenty, times, i, planning, buyin...\n",
            "5210     [ok, last, night, i, saw, world, premiere, pau...\n",
            "                               ...                        \n",
            "33338    [a, romanticised, thoroughly, false, vision, u...\n",
            "42453    [a, film, high, intelligence, activity, as, we...\n",
            "48945    [omg, painful, experience, life, watching, i, ...\n",
            "7826     [black, water, one, tense, films, i, viewed, l...\n",
            "3400     [this, high, grade, cheese, fare, b, movie, ku...\n",
            "Name: review, Length: 10000, dtype: object \n",
            "\n",
            "Test Set\n",
            "33282    0\n",
            "38418    0\n",
            "14178    0\n",
            "33857    0\n",
            "43719    0\n",
            "        ..\n",
            "26229    0\n",
            "38580    0\n",
            "4497     0\n",
            "7269     0\n",
            "9957     1\n",
            "Name: sentiment, Length: 40000, dtype: int64 \n",
            "\n",
            "18524    1\n",
            "46492    0\n",
            "12398    0\n",
            "34083    1\n",
            "5210     0\n",
            "        ..\n",
            "33338    0\n",
            "42453    1\n",
            "48945    0\n",
            "7826     1\n",
            "3400     0\n",
            "Name: sentiment, Length: 10000, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rUngYm-sNwoH"
      },
      "source": [
        "def get_max_length():\n",
        "    review_length = []\n",
        "    for review in x_train:\n",
        "        review_length.append(len(review))\n",
        "\n",
        "    return int(np.ceil(np.mean(review_length)))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zW6qbnhYN4dN"
      },
      "source": [
        "## Tokenize and Pad/Truncate Reviews\n",
        "\n",
        "A Neural Network only accepts numeric data, so we need to encode the reviews. I use tensorflow.keras.preprocessing.text.Tokenizer to encode the reviews into integers, where each unique word is automatically indexed (using fit_on_texts method) based on x_train.\n",
        "x_train and x_test is converted into integers using texts_to_sequences method.\n",
        "\n",
        "Each reviews has a different length, so we need to add padding (by adding 0) or truncating the words to the same length (in this case, it is the mean of all reviews length) using tensorflow.keras.preprocessing.sequence.pad_sequences.\n",
        "\n",
        "post, pad or truncate the words in the back of a sentence\n",
        "pre, pad or truncate the words in front of a sentence"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "99kzmaqQN11r",
        "outputId": "98cddecb-627f-4ba4-f5a3-7854af0d263b"
      },
      "source": [
        "print(x_train)\n",
        "# ENCODE REVIEW\n",
        "token = Tokenizer(lower=False)    # no need lower, because already lowered the data in load_data()\n",
        "token.fit_on_texts(x_train)\n",
        "x_train = token.texts_to_sequences(x_train)\n",
        "x_test = token.texts_to_sequences(x_test)\n",
        "\n",
        "max_length = get_max_length()\n",
        "\n",
        "x_train = pad_sequences(x_train, maxlen=max_length, padding='post', truncating='post')\n",
        "x_test = pad_sequences(x_test, maxlen=max_length, padding='post', truncating='post')\n",
        "\n",
        "total_words = len(token.word_index) + 1   # add 1 because of 0 padding\n",
        "\n",
        "print('Encoded X Train\\n', x_train, '\\n')\n",
        "print('Encoded X Test\\n', x_test, '\\n')\n",
        "print('Maximum review length: ', max_length)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "33282    [i, know, slashers, always, supposed, bad, com...\n",
            "38418    [this, movie, bad, i, know, begin, apparently,...\n",
            "14178    [thanks, killing, franchise, turkey, john, car...\n",
            "33857    [movie, made, watch, paul, w, s, anderson, avp...\n",
            "43719    [i, think, probably, one, worst, movies, i, wa...\n",
            "                               ...                        \n",
            "26229    [this, picture, hit, movie, screens, june, th,...\n",
            "38580    [the, best, thing, camp, films, general, know,...\n",
            "4497     [as, long, go, movie, understanding, going, co...\n",
            "7269     [this, movie, stupid, hate, turned, reached, h...\n",
            "9957     [i, actually, went, see, movie, low, expectati...\n",
            "Name: review, Length: 40000, dtype: object\n",
            "Encoded X Train\n",
            " [[   1   46 6644 ...   27   17   10]\n",
            " [   8    3   19 ...   79   61 1106]\n",
            " [1159  754 2850 ...  271 4680  354]\n",
            " ...\n",
            " [ 107  101   62 ...    0    0    0]\n",
            " [   8    3  276 ...    0    0    0]\n",
            " [   1   74  320 ...    0    0    0]] \n",
            "\n",
            "Encoded X Test\n",
            " [[  257  8780    21 ...     0     0     0]\n",
            " [  385     5    45 ...  3737     5  2343]\n",
            " [    1    40   337 ...  1277    15  2329]\n",
            " ...\n",
            " [11148  1252   466 ...     0     0     0]\n",
            " [  224   881     5 ...     0     0     0]\n",
            " [    8   205  1124 ...   292     4     0]] \n",
            "\n",
            "Maximum review length:  130\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VwdL42nCOCWu"
      },
      "source": [
        "## Build Architecture/Model\n",
        "\n",
        "**Embedding Layer:** in simple terms, it creates word vectors of each word in the word_index and group words that are related or have similar meaning by analyzing other words around them.\n",
        "\n",
        "**LSTM Layer:** to make a decision to keep or throw away data by considering the current input, previous output, and previous memory. There are some important components in LSTM.\n",
        "\n",
        "- **Forget Gate**, decides information is to be kept or thrown away\n",
        "- Input Gate, updates cell state by passing previous output and current input into sigmoid activation function\n",
        "- **Cell State**, calculate new cell state, it is multiplied by forget vector (drop value if multiplied by a near 0), add it with the output from input gate to update the cell state value.\n",
        "- **Ouput Gate**, decides the next hidden state and used for predictions\n",
        "\n",
        "**Dense Layer:** compute the input with the weight matrix and bias (optional), and using an activation function. I use Sigmoid activation function for this work because the output is only 0 or 1.\n",
        "\n",
        "The optimizer is Adam and the loss function is Binary Crossentropy because again the output is only 0 and 1, which is a binary number."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X9Jk16NAN-4u",
        "outputId": "7d12420d-3c32-42b7-ff66-8a598b92f8a4"
      },
      "source": [
        "from tensorflow.keras.utils import plot_model\n",
        "# ARCHITECTURE\n",
        "EMBED_DIM = 32\n",
        "LSTM_OUT = 64\n",
        "\n",
        "model = Sequential()\n",
        "model.add(Embedding(total_words, EMBED_DIM, input_length = max_length))\n",
        "model.add(LSTM(LSTM_OUT))\n",
        "model.add(Dense(64, activation='relu'))\n",
        "model.add(Dense(1, activation='sigmoid'))\n",
        "model.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])\n",
        "plot_model(model, to_file='/content/drive/MyDrive/US_elections/LSTM/modelLSTM.png')\n",
        "print(model.summary())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding (Embedding)        (None, 130, 32)           2948384   \n",
            "_________________________________________________________________\n",
            "lstm (LSTM)                  (None, 64)                24832     \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 64)                4160      \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 1)                 65        \n",
            "=================================================================\n",
            "Total params: 2,977,441\n",
            "Trainable params: 2,977,441\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FwBwe39zOYJI"
      },
      "source": [
        "checkpoint = ModelCheckpoint(\n",
        "    '/content/drive/MyDrive/US_elections/LSTM/LSTM_2.h5',\n",
        "    monitor='accuracy',\n",
        "    save_best_only=True,\n",
        "    verbose=1\n",
        ")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zvhj3IKdOjYj",
        "outputId": "f37eb015-6b85-4b4b-b3b3-69879e6c8e80"
      },
      "source": [
        "model.fit(x_train, y_train, batch_size = 128, epochs = 5, callbacks=[checkpoint])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "313/313 [==============================] - 20s 36ms/step - loss: 0.5609 - accuracy: 0.6473\n",
            "\n",
            "Epoch 00001: accuracy improved from -inf to 0.77410, saving model to /content/drive/MyDrive/US_elections/LSTM/LSTM_2.h5\n",
            "Epoch 2/5\n",
            "313/313 [==============================] - 11s 36ms/step - loss: 0.1871 - accuracy: 0.9326\n",
            "\n",
            "Epoch 00002: accuracy improved from 0.77410 to 0.92948, saving model to /content/drive/MyDrive/US_elections/LSTM/LSTM_2.h5\n",
            "Epoch 3/5\n",
            "313/313 [==============================] - 11s 36ms/step - loss: 0.1013 - accuracy: 0.9686\n",
            "\n",
            "Epoch 00003: accuracy improved from 0.92948 to 0.96562, saving model to /content/drive/MyDrive/US_elections/LSTM/LSTM_2.h5\n",
            "Epoch 4/5\n",
            "313/313 [==============================] - 11s 36ms/step - loss: 0.0563 - accuracy: 0.9835\n",
            "\n",
            "Epoch 00004: accuracy improved from 0.96562 to 0.97975, saving model to /content/drive/MyDrive/US_elections/LSTM/LSTM_2.h5\n",
            "Epoch 5/5\n",
            "313/313 [==============================] - 11s 36ms/step - loss: 0.0514 - accuracy: 0.9844\n",
            "\n",
            "Epoch 00005: accuracy improved from 0.97975 to 0.98350, saving model to /content/drive/MyDrive/US_elections/LSTM/LSTM_2.h5\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7fe4a2fff710>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I2cXl9vXOoHY"
      },
      "source": [
        "\n",
        "## Testing\n",
        "\n",
        "To evaluate the model, we need to predict the sentiment using our **x_test** data and comparing the predictions with y_test (expected output) data. Then, we calculate the accuracy of the model by dividing numbers of correct prediction with the total data. Resulted an accuracy of **86.63%**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OC-t2EwBOk8O",
        "outputId": "7fd94878-6082-41ef-fed6-85b3cc3014c0"
      },
      "source": [
        "\n",
        "y_pred = model.predict_classes(x_test, batch_size = 128)\n",
        "\n",
        "true = 0\n",
        "for i, y in enumerate(y_test):\n",
        "    if y == y_pred[i]:\n",
        "        true += 1\n",
        "\n",
        "print('Correct Prediction: {}'.format(true))\n",
        "print('Wrong Prediction: {}'.format(len(y_pred) - true))\n",
        "print('Accuracy: {}'.format(true/len(y_pred)*100))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/sequential.py:450: UserWarning: `model.predict_classes()` is deprecated and will be removed after 2021-01-01. Please use instead:* `np.argmax(model.predict(x), axis=-1)`,   if your model does multi-class classification   (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`,   if your model does binary classification   (e.g. if it uses a `sigmoid` last-layer activation).\n",
            "  warnings.warn('`model.predict_classes()` is deprecated and '\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Correct Prediction: 8658\n",
            "Wrong Prediction: 1342\n",
            "Accuracy: 86.58\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e9JT5lidP7XW"
      },
      "source": [
        "## Load Saved Model\n",
        "\n",
        "Load saved model and use it to predict a movie review statement's sentiment (positive or negative)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bkahGprwOxWI"
      },
      "source": [
        "loaded_model = load_model('/content/drive/MyDrive/US_elections/LSTM/LSTM.h5')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fp0ShOc_QF0V"
      },
      "source": [
        "Receives a review as an input to be predicted"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "679dASPGQFLN",
        "outputId": "0cbff5ef-9c62-4795-f893-40fb45700055"
      },
      "source": [
        "review = str(input('Movie Review: '))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Movie Review: Nothing was typical about this. Everything was beautifully done in this movie, the story, the flow, the scenario, everything. I highly recommend it for mystery lovers, for anyone who wants to watch a good movie!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qV2VO1BpQOC0"
      },
      "source": [
        "The input must be pre processed before it is passed to the model to be predicted"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sblaYgLmQC6u",
        "outputId": "932b4fb5-97ef-44d4-b9ef-a13b9de7feb7"
      },
      "source": [
        "# Pre-process input\n",
        "regex = re.compile(r'[^a-zA-Z\\s]')\n",
        "review = regex.sub('', review)\n",
        "print('Cleaned: ', review)\n",
        "\n",
        "words = review.split(' ')\n",
        "filtered = [w for w in words if w not in english_stops]\n",
        "filtered = ' '.join(filtered)\n",
        "filtered = [filtered.lower()]\n",
        "\n",
        "print('Filtered: ', filtered)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cleaned:  Nothing was typical about this Everything was beautifully done in this movie the story the flow the scenario everything I highly recommend it for mystery lovers for anyone who wants to watch a good movie\n",
            "Filtered:  ['nothing typical everything beautifully done movie story flow scenario everything i highly recommend mystery lovers anyone wants watch good movie']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9HgxPTw6QPxX",
        "outputId": "a6389370-2a20-4fa8-df77-fe17e7abc7f7"
      },
      "source": [
        "tokenize_words = token.texts_to_sequences(filtered)\n",
        "tokenize_words = pad_sequences(tokenize_words, maxlen=max_length, padding='post', truncating='post')\n",
        "print(tokenize_words)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[  76  680  169 1188  125    3   15 2651 2642  169    1  444  285  706\n",
            "  1721  151  400   34    9    3    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0    0    0    0    0    0    0    0    0    0    0\n",
            "     0    0    0    0]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AJvw15txQSqE",
        "outputId": "196217d9-be33-4464-c787-33875ace15d5"
      },
      "source": [
        "result = loaded_model.predict(tokenize_words)\n",
        "print(result)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0.99802494]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3NumKa_UQVn1",
        "outputId": "2f47d133-1b08-417c-b734-fca00d353bc2"
      },
      "source": [
        "if result >= 0.7:\n",
        "    print('positive')\n",
        "else:\n",
        "    print('negative')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "positive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "exkkcbkzRg2W"
      },
      "source": [
        "# Testing this on tweet data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 960,
          "referenced_widgets": [
            "666e1120eae54988a88c1814e1f22e93",
            "27f04b9c43de464aaa042af00627dead",
            "77802c885023427b8b7035408db82756",
            "a6721f2a413f450b9917514ae89bf9e5",
            "e0f0689f327544fbb3345f90c40b46e1",
            "6006fb99fe744b45bcac6abc5b496b32",
            "0afce616673e4a25af309d8ee61ceb5a",
            "fc86ecbf47294e32b0fdba7555e36478",
            "4762d2208ec24134ac1d7b2d8193b531",
            "b3cf5198771f4c298077ea268b01f1dd",
            "b98a8ded08ae4fc29e483432ee4e323a",
            "770d5310f55a485490b576a57e96c35e",
            "a7f17204c6064ee0b1953f9d9b709b40",
            "25c8a02ee5f44f4abd8e4c58b80b21b1",
            "b95989337cdd4a14a490899ca4db244f",
            "0f6d58e2b384474280539978a5504ae2",
            "f9538fa35ec74011a9ab18be4653f3e8",
            "6a0fb8a9a26848f6867422ee6145826b",
            "076f2b249be747f394f88e9ab8420d89",
            "31bb9c8f47dd49a99a221f8a6017e99d",
            "5136ecf2eb614761a8afbaa79211d76d",
            "958cc98efe0f492da59bd2cd1886377f",
            "c1777493a4094c889e68e5d2a2073ea7",
            "5d7a9114cb3b4ed89f0b30c529f7fb8a"
          ]
        },
        "id": "lbw3zst_QZd4",
        "outputId": "a97f7dd7-bd79-447f-d061-ce3d75f1a6cc"
      },
      "source": [
        "# from keras.preprocessing.text import Tokenizer\n",
        "# from keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "\n",
        "!pip install transformers command\n",
        "from transformers import BertTokenizer, TFBertForSequenceClassification\n",
        "\n",
        "model = TFBertForSequenceClassification.from_pretrained(\"bert-base-uncased\")\n",
        "tokenizer = BertTokenizer.from_pretrained(\"bert-base-uncased\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting transformers\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/cd/40/866cbfac4601e0f74c7303d533a9c5d4a53858bd402e08e3e294dd271f25/transformers-4.2.1-py3-none-any.whl (1.8MB)\n",
            "\r\u001b[K     |▏                               | 10kB 29.7MB/s eta 0:00:01\r\u001b[K     |▍                               | 20kB 34.7MB/s eta 0:00:01\r\u001b[K     |▋                               | 30kB 21.1MB/s eta 0:00:01\r\u001b[K     |▊                               | 40kB 19.7MB/s eta 0:00:01\r\u001b[K     |█                               | 51kB 20.6MB/s eta 0:00:01\r\u001b[K     |█▏                              | 61kB 15.8MB/s eta 0:00:01\r\u001b[K     |█▎                              | 71kB 15.4MB/s eta 0:00:01\r\u001b[K     |█▌                              | 81kB 14.8MB/s eta 0:00:01\r\u001b[K     |█▊                              | 92kB 14.9MB/s eta 0:00:01\r\u001b[K     |█▉                              | 102kB 15.5MB/s eta 0:00:01\r\u001b[K     |██                              | 112kB 15.5MB/s eta 0:00:01\r\u001b[K     |██▎                             | 122kB 15.5MB/s eta 0:00:01\r\u001b[K     |██▍                             | 133kB 15.5MB/s eta 0:00:01\r\u001b[K     |██▋                             | 143kB 15.5MB/s eta 0:00:01\r\u001b[K     |██▉                             | 153kB 15.5MB/s eta 0:00:01\r\u001b[K     |███                             | 163kB 15.5MB/s eta 0:00:01\r\u001b[K     |███▏                            | 174kB 15.5MB/s eta 0:00:01\r\u001b[K     |███▍                            | 184kB 15.5MB/s eta 0:00:01\r\u001b[K     |███▌                            | 194kB 15.5MB/s eta 0:00:01\r\u001b[K     |███▊                            | 204kB 15.5MB/s eta 0:00:01\r\u001b[K     |████                            | 215kB 15.5MB/s eta 0:00:01\r\u001b[K     |████                            | 225kB 15.5MB/s eta 0:00:01\r\u001b[K     |████▎                           | 235kB 15.5MB/s eta 0:00:01\r\u001b[K     |████▌                           | 245kB 15.5MB/s eta 0:00:01\r\u001b[K     |████▋                           | 256kB 15.5MB/s eta 0:00:01\r\u001b[K     |████▉                           | 266kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████                           | 276kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████▏                          | 286kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████▍                          | 296kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████▋                          | 307kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████▊                          | 317kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████                          | 327kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████▏                         | 337kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████▎                         | 348kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████▌                         | 358kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████▊                         | 368kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████▉                         | 378kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████                         | 389kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████▎                        | 399kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████▍                        | 409kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████▋                        | 419kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████▉                        | 430kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████                        | 440kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████▏                       | 450kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████▍                       | 460kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████▌                       | 471kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████▊                       | 481kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████                       | 491kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████                       | 501kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████▎                      | 512kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████▌                      | 522kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████▋                      | 532kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████▉                      | 542kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████                      | 552kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████▏                     | 563kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████▍                     | 573kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████▋                     | 583kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████▊                     | 593kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████                     | 604kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████▏                    | 614kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████▎                    | 624kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████▌                    | 634kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████▊                    | 645kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████▉                    | 655kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████                    | 665kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████▎                   | 675kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████▍                   | 686kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████▋                   | 696kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████▉                   | 706kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████                   | 716kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████▏                  | 727kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████▍                  | 737kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████▌                  | 747kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████▊                  | 757kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████                  | 768kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████                  | 778kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████▎                 | 788kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████▌                 | 798kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████▋                 | 808kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████▉                 | 819kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████                 | 829kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████▏                | 839kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████▍                | 849kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████▋                | 860kB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████▊                | 870kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████                | 880kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████▏               | 890kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████▎               | 901kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████▌               | 911kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████▊               | 921kB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████▉               | 931kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████               | 942kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████▎              | 952kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████▍              | 962kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████▋              | 972kB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████▉              | 983kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████              | 993kB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████▏             | 1.0MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████▍             | 1.0MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████▌             | 1.0MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████▊             | 1.0MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████             | 1.0MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████             | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████▎            | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████▌            | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████▋            | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████▉            | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████            | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████▏           | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████▍           | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████▋           | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████▊           | 1.1MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████           | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████▏          | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████▎          | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████▌          | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████▊          | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████▉          | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████          | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████▎         | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████▍         | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████▋         | 1.2MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████▉         | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████         | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████▏        | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████▍        | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████▋        | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████▊        | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████        | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████▏       | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████▎       | 1.3MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████▌       | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████▊       | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████▉       | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████       | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▎      | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▍      | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▋      | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▉      | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████      | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▏     | 1.4MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▍     | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▌     | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▊     | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████     | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████     | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▎    | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▌    | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▋    | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▉    | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████    | 1.5MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▏   | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▍   | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▋   | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▊   | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████   | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▏  | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▎  | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▌  | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▊  | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▉  | 1.6MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████  | 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▎ | 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▍ | 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▋ | 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▉ | 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████ | 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▏| 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▍| 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▌| 1.7MB 15.5MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▊| 1.8MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 1.8MB 15.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 1.8MB 15.5MB/s \n",
            "\u001b[?25hCollecting command\n",
            "  Downloading https://files.pythonhosted.org/packages/f1/aa/fd642369a793e4a79c99f8da8cf3f7541276c43fbe1c47f7f49c1d64928e/Command-0.1.0.tar.gz\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from transformers) (20.8)\n",
            "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.6/dist-packages (from transformers) (4.41.1)\n",
            "Requirement already satisfied: importlib-metadata; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from transformers) (3.3.0)\n",
            "Collecting sacremoses\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/7d/34/09d19aff26edcc8eb2a01bed8e98f13a1537005d31e95233fd48216eed10/sacremoses-0.0.43.tar.gz (883kB)\n",
            "\u001b[K     |████████████████████████████████| 890kB 44.1MB/s \n",
            "\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.6/dist-packages (from transformers) (2.23.0)\n",
            "Requirement already satisfied: dataclasses; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from transformers) (0.8)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.6/dist-packages (from transformers) (3.0.12)\n",
            "Collecting tokenizers==0.9.4\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/0f/1c/e789a8b12e28be5bc1ce2156cf87cb522b379be9cadc7ad8091a4cc107c4/tokenizers-0.9.4-cp36-cp36m-manylinux2010_x86_64.whl (2.9MB)\n",
            "\u001b[K     |████████████████████████████████| 2.9MB 54.8MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from transformers) (1.19.5)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.6/dist-packages (from transformers) (2019.12.20)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from command) (51.1.1)\n",
            "Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from packaging->transformers) (2.4.7)\n",
            "Requirement already satisfied: typing-extensions>=3.6.4; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < \"3.8\"->transformers) (3.7.4.3)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < \"3.8\"->transformers) (3.4.0)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.15.0)\n",
            "Requirement already satisfied: click in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (7.1.2)\n",
            "Requirement already satisfied: joblib in /usr/local/lib/python3.6/dist-packages (from sacremoses->transformers) (1.0.0)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (1.24.3)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2.10)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (2020.12.5)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests->transformers) (3.0.4)\n",
            "Building wheels for collected packages: command, sacremoses\n",
            "  Building wheel for command (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for command: filename=Command-0.1.0-cp36-none-any.whl size=6406 sha256=aceed2257756ff61a270c6d4d4405b6607b54400d548c28f760b8b7093af8bd5\n",
            "  Stored in directory: /root/.cache/pip/wheels/6b/61/e7/7e78f4d431de4a7d1f83af33f599d0bfe03e65fe3ff1ad73c5\n",
            "  Building wheel for sacremoses (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for sacremoses: filename=sacremoses-0.0.43-cp36-none-any.whl size=893261 sha256=6372e273b728bfbf7042d8ce941fb11ba9ea8aba8f4997f84ea169a57aaa9ac8\n",
            "  Stored in directory: /root/.cache/pip/wheels/29/3c/fd/7ce5c3f0666dab31a50123635e6fb5e19ceb42ce38d4e58f45\n",
            "Successfully built command sacremoses\n",
            "Installing collected packages: sacremoses, tokenizers, transformers, command\n",
            "Successfully installed command-0.1.0 sacremoses-0.0.43 tokenizers-0.9.4 transformers-4.2.1\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "666e1120eae54988a88c1814e1f22e93",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=433.0, style=ProgressStyle(description_…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "4762d2208ec24134ac1d7b2d8193b531",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=536063208.0, style=ProgressStyle(descri…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "All model checkpoint layers were used when initializing TFBertForSequenceClassification.\n",
            "\n",
            "Some layers of TFBertForSequenceClassification were not initialized from the model checkpoint at bert-base-uncased and are newly initialized: ['classifier']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "f9538fa35ec74011a9ab18be4653f3e8",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=231508.0, style=ProgressStyle(descripti…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kgDdQQR2RmzL"
      },
      "source": [
        "loaded_model = load_model('/content/drive/MyDrive/US_elections/LSTM/LSTM.h5')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "313k8fiBRrdW",
        "outputId": "3a499b09-054d-4b5c-95f6-f3d0cb601848"
      },
      "source": [
        "biden_df = pd.read_csv('/content/drive/MyDrive/US_elections/cleaned_sample_100_biden.csv')\n",
        "trump_df = pd.read_csv('/content/drive/MyDrive/US_elections/cleaned_sample_100_trump.csv')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/IPython/core/interactiveshell.py:2718: DtypeWarning: Columns (0,2,3) have mixed types.Specify dtype option on import or set low_memory=False.\n",
            "  interactivity=interactivity, compiler=compiler, result=result)\n",
            "/usr/local/lib/python3.6/dist-packages/IPython/core/interactiveshell.py:2718: DtypeWarning: Columns (0,2) have mixed types.Specify dtype option on import or set low_memory=False.\n",
            "  interactivity=interactivity, compiler=compiler, result=result)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 47
        },
        "id": "TZuBwtOASF9i",
        "outputId": "bf68479c-99c6-4010-c09d-6ee8ecc90f87"
      },
      "source": [
        "biden_df.loc[:,'who'] = 'biden'\n",
        "trump_df.loc[:,'who'] = 'trump'\n",
        "\n",
        "balance_data_df = pd.DataFrame(columns=trump_df.columns)\n",
        "balance_data_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>tweet</th>\n",
              "      <th>long</th>\n",
              "      <th>lat</th>\n",
              "      <th>state_code</th>\n",
              "      <th>sentiment</th>\n",
              "      <th>who</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "Empty DataFrame\n",
              "Columns: [user_id, tweet, long, lat, state_code, sentiment, who]\n",
              "Index: []"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J5MvrHJZSIkr"
      },
      "source": [
        "states_list = set(biden_df['state_code']).union(set(trump_df['state_code'])) - set(['MP', 'GU', 'PR', 'VI', 'AS', 'UM', ])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h5gH4KZ4SPOj"
      },
      "source": [
        "for state in states_list:\n",
        "    trump_df_state = trump_df[trump_df['state_code']==state]\n",
        "    biden_df_state = biden_df[biden_df['state_code']==state]\n",
        "    \n",
        "    n_sample = len(trump_df_state) if len(trump_df_state) < len(biden_df_state) else len(biden_df_state)\n",
        "    \n",
        "    if len(trump_df_state) < len(biden_df_state):\n",
        "        balance_data_df = pd.concat([balance_data_df, trump_df_state])\n",
        "        balance_data_df = pd.concat([balance_data_df, biden_df_state.sample(len(trump_df_state))])\n",
        "    else:\n",
        "        balance_data_df = pd.concat([balance_data_df, biden_df_state])\n",
        "        balance_data_df = pd.concat([balance_data_df, trump_df_state.sample(len(biden_df_state))])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "DHBiBy5NSQoo",
        "outputId": "dc2e6726-b9ed-4ae4-8393-6df208eaff95"
      },
      "source": [
        "balance_data_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>tweet</th>\n",
              "      <th>long</th>\n",
              "      <th>lat</th>\n",
              "      <th>state_code</th>\n",
              "      <th>sentiment</th>\n",
              "      <th>who</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1081.0</td>\n",
              "      <td>finally FoxNews just called it for Biden</td>\n",
              "      <td>-122.0322895</td>\n",
              "      <td>37.3229</td>\n",
              "      <td>CA</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>10727.0</td>\n",
              "      <td>biden beats trump. Talk about a relaxing way t...</td>\n",
              "      <td>-122.4199061</td>\n",
              "      <td>37.779</td>\n",
              "      <td>CA</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>29283.0</td>\n",
              "      <td>Trump mentioned Motown in MI today,  but Motow...</td>\n",
              "      <td>-122.4199061</td>\n",
              "      <td>37.779</td>\n",
              "      <td>CA</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>647253.0</td>\n",
              "      <td>Fully agree! You can dislike Bidens politics h...</td>\n",
              "      <td>-122.2713563</td>\n",
              "      <td>37.8045</td>\n",
              "      <td>CA</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>682223.0</td>\n",
              "      <td>This week...    JoeBiden JOEBIDEN2020 JoeBiden...</td>\n",
              "      <td>-122.4199061</td>\n",
              "      <td>37.779</td>\n",
              "      <td>CA</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44862</th>\n",
              "      <td>198313485.0</td>\n",
              "      <td>TheRealFixNow The DOJ has said the laptop is c...</td>\n",
              "      <td>-80.4363743</td>\n",
              "      <td>33.6874</td>\n",
              "      <td>SC</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44868</th>\n",
              "      <td>198313485.0</td>\n",
              "      <td>Their crazed, mind controlled masses are turni...</td>\n",
              "      <td>-80.4363743</td>\n",
              "      <td>33.6874</td>\n",
              "      <td>SC</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>128625</th>\n",
              "      <td>1.2641824959873477e+18</td>\n",
              "      <td>BarackObama ReallyAmerican1 JoeBiden KamalaHar...</td>\n",
              "      <td>-78.89004090000002</td>\n",
              "      <td>33.6956</td>\n",
              "      <td>SC</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>120406</th>\n",
              "      <td>1.156761495272903e+18</td>\n",
              "      <td>Att CNN MSNBC CBSNews NBCNews ABC NewsHour nyt...</td>\n",
              "      <td>-80.4363743</td>\n",
              "      <td>33.6874</td>\n",
              "      <td>SC</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>82083</th>\n",
              "      <td>2498324239.0</td>\n",
              "      <td>GOPLeader realDonaldTrump The best is yet to c...</td>\n",
              "      <td>-80.4363743</td>\n",
              "      <td>33.6874</td>\n",
              "      <td>SC</td>\n",
              "      <td>0.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>271972 rows × 7 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                       user_id  ...    who\n",
              "0                       1081.0  ...  trump\n",
              "11                     10727.0  ...  trump\n",
              "15                     29283.0  ...  trump\n",
              "25                    647253.0  ...  trump\n",
              "27                    682223.0  ...  trump\n",
              "...                        ...  ...    ...\n",
              "44862              198313485.0  ...  trump\n",
              "44868              198313485.0  ...  trump\n",
              "128625  1.2641824959873477e+18  ...  trump\n",
              "120406   1.156761495272903e+18  ...  trump\n",
              "82083             2498324239.0  ...  trump\n",
              "\n",
              "[271972 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2xsP7PveSVyE"
      },
      "source": [
        "batch_size = 100"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mDINq9adSX0I"
      },
      "source": [
        "from tqdm.auto import tqdm\n",
        "import tensorflow as tf\n",
        "\n",
        "def emotion_classify(tweet_list):\n",
        "    batch_idx = np.append(np.arange(0, len(tweet_list), batch_size), len(tweet_list))\n",
        "    labels = np.array([])\n",
        "    for i in tqdm(range(len(batch_idx)-1)):\n",
        "        tf_batch = tokenizer(tweet_list[batch_idx[i]:batch_idx[i+1]], max_length=280, padding=True, truncation=True, return_tensors='tf')\n",
        "        tf_outputs = model(tf_batch)\n",
        "        tf_predictions = tf.nn.softmax(tf_outputs[0], axis=-1)\n",
        "        labels = np.append(labels, tf.argmax(tf_predictions, axis=1))\n",
        "#         labels = np.append(labels, tf_predictions[:,1])\n",
        "        del tf_batch\n",
        "    return labels"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9c_PK8bz5pDF"
      },
      "source": [
        "# vocabulary_size = 20000\n",
        "# tokenizer = Tokenizer(num_words= vocabulary_size)\n",
        "# tokenizer.fit_on_texts(balance_data_df['tweet'])\n",
        "# sequences = tokenizer.texts_to_sequences(balance_data_df['tweet'])\n",
        "# data = pad_sequences(sequences, maxlen=50)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 66,
          "referenced_widgets": [
            "cf5c9afee54f4b1a923a731f5677d446",
            "1a63b4343824434e95a38a79e3c289f0",
            "f3d7a20ea7c94eb4a1c8d3ace431149f",
            "d4b85706b3ef48578ce2a50cf3e6aac6",
            "43214e2d0a4544c69e9a7970e2303c3d",
            "8c5e8ea08b8540e09a9bc5bce721b196",
            "43db2f1055e048658e7a6a0772c7aaf1",
            "4b82ab19a4944daaa68e0c642464d9cf"
          ]
        },
        "id": "JjmXnQ7eTb-F",
        "outputId": "f7da35f0-5fbb-4858-fb81-cf57adf1ebb5"
      },
      "source": [
        "balance_data_df['sentiment'] = emotion_classify(list(balance_data_df['tweet']))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "cf5c9afee54f4b1a923a731f5677d446",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, max=2720.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "UwtKvKA13uh0",
        "outputId": "83b10322-e70a-4988-e208-6b76e02fe699"
      },
      "source": [
        "balance_data_df.to_csv('/content/drive/MyDrive/US_elections/LSTM/balance_data_with_sentiments.csv', index=False, header=True)\n",
        "\n",
        "balance_data_df.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>tweet</th>\n",
              "      <th>long</th>\n",
              "      <th>lat</th>\n",
              "      <th>state_code</th>\n",
              "      <th>sentiment</th>\n",
              "      <th>who</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1081.0</td>\n",
              "      <td>finally FoxNews just called it for Biden</td>\n",
              "      <td>-122.0322895</td>\n",
              "      <td>37.3229</td>\n",
              "      <td>CA</td>\n",
              "      <td>1.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>10727.0</td>\n",
              "      <td>biden beats trump. Talk about a relaxing way t...</td>\n",
              "      <td>-122.4199061</td>\n",
              "      <td>37.779</td>\n",
              "      <td>CA</td>\n",
              "      <td>1.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>29283.0</td>\n",
              "      <td>Trump mentioned Motown in MI today,  but Motow...</td>\n",
              "      <td>-122.4199061</td>\n",
              "      <td>37.779</td>\n",
              "      <td>CA</td>\n",
              "      <td>1.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>647253.0</td>\n",
              "      <td>Fully agree! You can dislike Bidens politics h...</td>\n",
              "      <td>-122.2713563</td>\n",
              "      <td>37.8045</td>\n",
              "      <td>CA</td>\n",
              "      <td>1.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>682223.0</td>\n",
              "      <td>This week...    JoeBiden JOEBIDEN2020 JoeBiden...</td>\n",
              "      <td>-122.4199061</td>\n",
              "      <td>37.779</td>\n",
              "      <td>CA</td>\n",
              "      <td>1.0</td>\n",
              "      <td>trump</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     user_id  ...    who\n",
              "0     1081.0  ...  trump\n",
              "11   10727.0  ...  trump\n",
              "15   29283.0  ...  trump\n",
              "25  647253.0  ...  trump\n",
              "27  682223.0  ...  trump\n",
              "\n",
              "[5 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ephc-XU2TdtJ"
      },
      "source": [
        "dem_states = set(['DC', 'VT', 'HI', 'MA', 'MD', 'CA', 'NY', 'RI', 'DE', \n",
        "                 'CT', 'WA', 'ME', 'NJ', 'OR', 'IL'])\n",
        "rep_states = set(['OH', 'IA', 'TX', 'MT', 'SC', 'AK', 'MO', 'IN', 'NE',\n",
        "                 'KS', 'UT', 'MS', 'TN', 'SD', 'KY', 'LA', 'AL', 'ND',\n",
        "                 'ID', 'AR', 'OK', 'WV', 'WY'])\n",
        "swing_states = states_list - dem_states - rep_states"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ckQHmc8MVibM"
      },
      "source": [
        "electoral_vote_by_state = pd.read_csv('/content/drive/MyDrive/US_elections/electoral_vote_by_state.csv')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "agB4IH0cVlMG"
      },
      "source": [
        "states_results = pd.DataFrame(columns=['state_code', 'biden', 'trump', 'type', 'winner'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DFf63qDnVofN"
      },
      "source": [
        "states_results.loc[:,'state_code'] = electoral_vote_by_state['state_code']\n",
        "states_results.loc[:,'e_vote'] = electoral_vote_by_state['e_vote']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_FIA6MUZVo8q"
      },
      "source": [
        "type_vs_candidate = {'blue': 'biden', 'red': 'trump', 'purple': 'contentious'}\n",
        "threshold = 1.15\n",
        "for state in states_list:\n",
        "    trump_slice = balance_data_df.query(f'state_code == \"{state}\" and who == \"trump\"')\n",
        "    biden_slice = balance_data_df.query(f'state_code == \"{state}\" and who == \"biden\"')\n",
        "    trump_score = np.sum(trump_slice['sentiment']) + 1\n",
        "    biden_score = np.sum(biden_slice['sentiment']) + 1\n",
        "    \n",
        "    state_type = ''\n",
        "    if state in dem_states:\n",
        "        state_type = 'blue'\n",
        "    elif state in rep_states:\n",
        "        state_type = 'red'\n",
        "    else:\n",
        "        state_type = 'purple'\n",
        "        \n",
        "    winner = ''\n",
        "    if len(trump_slice) + len(biden_slice) < 1000:\n",
        "        winner = type_vs_candidate[state_type]\n",
        "    else:\n",
        "        if trump_score/biden_score > threshold:\n",
        "            winner = 'trump'\n",
        "        elif biden_score/trump_score > threshold:\n",
        "            winner = 'biden'\n",
        "        else:\n",
        "            winner = 'contentious'\n",
        "        \n",
        "    states_results.loc[states_results['state_code'] == state, ['biden', 'trump', 'type', 'winner']] = \\\n",
        "        [biden_score, trump_score, state_type, winner]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "eY8cSzmtVqif",
        "outputId": "c07abbab-7bb4-4d22-8851-aedc5ce315fa"
      },
      "source": [
        "states_results.query('winner==\"contentious\"')\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>state_code</th>\n",
              "      <th>biden</th>\n",
              "      <th>trump</th>\n",
              "      <th>type</th>\n",
              "      <th>winner</th>\n",
              "      <th>e_vote</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>AL</td>\n",
              "      <td>793</td>\n",
              "      <td>791</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>AZ</td>\n",
              "      <td>2642</td>\n",
              "      <td>2648</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>NV</td>\n",
              "      <td>2087</td>\n",
              "      <td>2087</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>NH</td>\n",
              "      <td>258</td>\n",
              "      <td>257</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>CA</td>\n",
              "      <td>23581</td>\n",
              "      <td>23685</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>55</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>NJ</td>\n",
              "      <td>3566</td>\n",
              "      <td>3569</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>CO</td>\n",
              "      <td>2407</td>\n",
              "      <td>2459</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>NM</td>\n",
              "      <td>426</td>\n",
              "      <td>430</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>CT</td>\n",
              "      <td>832</td>\n",
              "      <td>831</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>NY</td>\n",
              "      <td>17185</td>\n",
              "      <td>17248</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>29</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>NC</td>\n",
              "      <td>2908</td>\n",
              "      <td>2935</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>DC</td>\n",
              "      <td>5679</td>\n",
              "      <td>5704</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>FL</td>\n",
              "      <td>11055</td>\n",
              "      <td>11069</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>29</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>OH</td>\n",
              "      <td>3069</td>\n",
              "      <td>3072</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>18</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>GA</td>\n",
              "      <td>2723</td>\n",
              "      <td>2737</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>OK</td>\n",
              "      <td>714</td>\n",
              "      <td>724</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>HI</td>\n",
              "      <td>576</td>\n",
              "      <td>577</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>OR</td>\n",
              "      <td>1924</td>\n",
              "      <td>1965</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>PA</td>\n",
              "      <td>5048</td>\n",
              "      <td>5062</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>IL</td>\n",
              "      <td>5016</td>\n",
              "      <td>5049</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>IN</td>\n",
              "      <td>1278</td>\n",
              "      <td>1291</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>SC</td>\n",
              "      <td>993</td>\n",
              "      <td>1048</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>IA</td>\n",
              "      <td>541</td>\n",
              "      <td>543</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>KS</td>\n",
              "      <td>843</td>\n",
              "      <td>838</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>TN</td>\n",
              "      <td>1762</td>\n",
              "      <td>1790</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>KY</td>\n",
              "      <td>1126</td>\n",
              "      <td>1144</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>TX</td>\n",
              "      <td>12861</td>\n",
              "      <td>12890</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>38</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>LA</td>\n",
              "      <td>968</td>\n",
              "      <td>966</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>UT</td>\n",
              "      <td>535</td>\n",
              "      <td>536</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>MD</td>\n",
              "      <td>1880</td>\n",
              "      <td>1885</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>VA</td>\n",
              "      <td>2094</td>\n",
              "      <td>2110</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>MA</td>\n",
              "      <td>3377</td>\n",
              "      <td>3414</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>WA</td>\n",
              "      <td>1991</td>\n",
              "      <td>2001</td>\n",
              "      <td>blue</td>\n",
              "      <td>contentious</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>MI</td>\n",
              "      <td>2668</td>\n",
              "      <td>2690</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>MN</td>\n",
              "      <td>1524</td>\n",
              "      <td>1538</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>WI</td>\n",
              "      <td>1424</td>\n",
              "      <td>1426</td>\n",
              "      <td>purple</td>\n",
              "      <td>contentious</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>MO</td>\n",
              "      <td>1592</td>\n",
              "      <td>1608</td>\n",
              "      <td>red</td>\n",
              "      <td>contentious</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   state_code  biden  trump    type       winner  e_vote\n",
              "0          AL    793    791     red  contentious       9\n",
              "4          AZ   2642   2648  purple  contentious      11\n",
              "5          NV   2087   2087  purple  contentious       6\n",
              "7          NH    258    257  purple  contentious       4\n",
              "8          CA  23581  23685    blue  contentious      55\n",
              "9          NJ   3566   3569    blue  contentious      14\n",
              "10         CO   2407   2459  purple  contentious       9\n",
              "11         NM    426    430  purple  contentious       5\n",
              "12         CT    832    831    blue  contentious       7\n",
              "13         NY  17185  17248    blue  contentious      29\n",
              "15         NC   2908   2935  purple  contentious      15\n",
              "16         DC   5679   5704    blue  contentious       3\n",
              "18         FL  11055  11069  purple  contentious      29\n",
              "19         OH   3069   3072     red  contentious      18\n",
              "20         GA   2723   2737  purple  contentious      16\n",
              "21         OK    714    724     red  contentious       7\n",
              "22         HI    576    577    blue  contentious       4\n",
              "23         OR   1924   1965    blue  contentious       7\n",
              "25         PA   5048   5062  purple  contentious      20\n",
              "26         IL   5016   5049    blue  contentious      20\n",
              "28         IN   1278   1291     red  contentious      11\n",
              "29         SC    993   1048     red  contentious       9\n",
              "30         IA    541    543     red  contentious       6\n",
              "32         KS    843    838     red  contentious       6\n",
              "33         TN   1762   1790     red  contentious      11\n",
              "34         KY   1126   1144     red  contentious       8\n",
              "35         TX  12861  12890     red  contentious      38\n",
              "36         LA    968    966     red  contentious       8\n",
              "37         UT    535    536     red  contentious       6\n",
              "40         MD   1880   1885    blue  contentious      10\n",
              "41         VA   2094   2110  purple  contentious      13\n",
              "42         MA   3377   3414    blue  contentious      11\n",
              "43         WA   1991   2001    blue  contentious      12\n",
              "44         MI   2668   2690  purple  contentious      16\n",
              "46         MN   1524   1538  purple  contentious      10\n",
              "47         WI   1424   1426  purple  contentious      10\n",
              "50         MO   1592   1608     red  contentious      10"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8rPO66rQVuc2",
        "outputId": "85387e53-58e7-47d6-fa29-fa373d2f8ba3"
      },
      "source": [
        "!pip install geopandas\n",
        "import geopandas as gpd\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "us_shape = gpd.read_file('/content/drive/MyDrive/US_elections/us_shape/cb_2018_us_state_20m.shp')\n",
        "us_shape = us_shape[['STUSPS', 'geometry']]\n",
        "states_results_tmp = states_results.copy()\n",
        "states_results_tmp = states_results_tmp.merge(us_shape, left_on='state_code', right_on='STUSPS')\n",
        "states_results_tmp.drop(['STUSPS'], axis=1, inplace=True)\n",
        "states_results_tmp = gpd.GeoDataFrame(states_results_tmp)\n",
        "\n",
        "states_results_tmp.columns"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting geopandas\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/f7/a4/e66aafbefcbb717813bf3a355c8c4fc3ed04ea1dd7feb2920f2f4f868921/geopandas-0.8.1-py2.py3-none-any.whl (962kB)\n",
            "\u001b[K     |████████████████████████████████| 972kB 22.4MB/s \n",
            "\u001b[?25hCollecting fiona\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/37/94/4910fd55246c1d963727b03885ead6ef1cd3748a465f7b0239ab25dfc9a3/Fiona-1.8.18-cp36-cp36m-manylinux1_x86_64.whl (14.8MB)\n",
            "\u001b[K     |████████████████████████████████| 14.8MB 221kB/s \n",
            "\u001b[?25hRequirement already satisfied: pandas>=0.23.0 in /usr/local/lib/python3.6/dist-packages (from geopandas) (1.1.5)\n",
            "Collecting pyproj>=2.2.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/e4/ab/280e80a67cfc109d15428c0ec56391fc03a65857b7727cf4e6e6f99a4204/pyproj-3.0.0.post1-cp36-cp36m-manylinux2010_x86_64.whl (6.4MB)\n",
            "\u001b[K     |████████████████████████████████| 6.5MB 44.3MB/s \n",
            "\u001b[?25hRequirement already satisfied: shapely in /usr/local/lib/python3.6/dist-packages (from geopandas) (1.7.1)\n",
            "Collecting click-plugins>=1.0\n",
            "  Downloading https://files.pythonhosted.org/packages/e9/da/824b92d9942f4e472702488857914bdd50f73021efea15b4cad9aca8ecef/click_plugins-1.1.1-py2.py3-none-any.whl\n",
            "Collecting cligj>=0.5\n",
            "  Downloading https://files.pythonhosted.org/packages/42/1e/947eadf10d6804bf276eb8a038bd5307996dceaaa41cfd21b7a15ec62f5d/cligj-0.7.1-py3-none-any.whl\n",
            "Collecting munch\n",
            "  Downloading https://files.pythonhosted.org/packages/cc/ab/85d8da5c9a45e072301beb37ad7f833cd344e04c817d97e0cc75681d248f/munch-2.5.0-py2.py3-none-any.whl\n",
            "Requirement already satisfied: attrs>=17 in /usr/local/lib/python3.6/dist-packages (from fiona->geopandas) (20.3.0)\n",
            "Requirement already satisfied: six>=1.7 in /usr/local/lib/python3.6/dist-packages (from fiona->geopandas) (1.15.0)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.6/dist-packages (from fiona->geopandas) (2020.12.5)\n",
            "Requirement already satisfied: click<8,>=4.0 in /usr/local/lib/python3.6/dist-packages (from fiona->geopandas) (7.1.2)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.23.0->geopandas) (2018.9)\n",
            "Requirement already satisfied: numpy>=1.15.4 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.23.0->geopandas) (1.19.5)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.23.0->geopandas) (2.8.1)\n",
            "Installing collected packages: click-plugins, cligj, munch, fiona, pyproj, geopandas\n",
            "Successfully installed click-plugins-1.1.1 cligj-0.7.1 fiona-1.8.18 geopandas-0.8.1 munch-2.5.0 pyproj-3.0.0.post1\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['state_code', 'biden', 'trump', 'type', 'winner', 'e_vote', 'geometry'], dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 541
        },
        "id": "Ddf61jToV4xz",
        "outputId": "513d7e34-6b68-462d-f8d5-c84f8f921599"
      },
      "source": [
        "fig, ax = plt.subplots(1, 1, figsize=(15,15))\n",
        "states_results_tmp.apply(lambda x: ax.annotate(s=x.state_code, xy=x.geometry.centroid.coords[0], ha='center', fontsize=9, color='white'),axis=1);\n",
        "states_results_tmp[states_results_tmp['winner'] == 'trump'].plot(ax=ax, color='red')\n",
        "states_results_tmp[states_results_tmp['winner'] == 'biden'].plot(ax=ax, color='blue')\n",
        "states_results_tmp[states_results_tmp['winner'] == 'contentious'].plot(ax=ax, color='purple')\n",
        "plt.axis('square')\n",
        "ax.set_xlim([-180, -60])\n",
        "ax.set_ylim([10, 80])\n",
        "ax.set_xticks([])\n",
        "ax.set_yticks([])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x1080 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h0C6yBAEV56v"
      },
      "source": [
        "states_results.to_csv('/content/drive/MyDrive/US_elections/state_results_LSTM.csv', index=False, header=True)\n",
        "fig.savefig('/content/drive/MyDrive/US_elections/state_results_LSTM.png')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iP51W2VcV_uO",
        "outputId": "8aa6e461-815f-48c8-c6f3-148b8ae4c7d6"
      },
      "source": [
        "trump_vote = np.sum(states_results[states_results['winner'] == 'trump']['e_vote'])\n",
        "biden_vote = np.sum(states_results[states_results['winner'] == 'biden']['e_vote'])\n",
        "print(trump_vote, biden_vote)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "41 14\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RBaJkW5bWA-q"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}