Fine_tune_bert_with_hugging face.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Fine tune bert.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyNK4xOWtb8aMFmulk197YdQ",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "7a38090ea79c4b2485416a636fff494a": {
          "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_475a21b11b464c68a9e44a834c10ccd0",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_9c3532e6eeba45cca6370cd58ea32707",
              "IPY_MODEL_397ecf03fb664c809fe6069c497a0c43"
            ]
          }
        },
        "475a21b11b464c68a9e44a834c10ccd0": {
          "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
          }
        },
        "9c3532e6eeba45cca6370cd58ea32707": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_8bf8f7aac81a471fb929aa45a03d1a0b",
            "_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_8753db13c9434f21b903dd18e80c3be0"
          }
        },
        "397ecf03fb664c809fe6069c497a0c43": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_534fb12c064640888437f003c4f0785c",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 232k/232k [00:01&lt;00:00, 180kB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_c3934f64d2a040218f42f3a721c330ef"
          }
        },
        "8bf8f7aac81a471fb929aa45a03d1a0b": {
          "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"
          }
        },
        "8753db13c9434f21b903dd18e80c3be0": {
          "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
          }
        },
        "534fb12c064640888437f003c4f0785c": {
          "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"
          }
        },
        "c3934f64d2a040218f42f3a721c330ef": {
          "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
          }
        },
        "4b6b7e5077104adf873dd48f96efd9c1": {
          "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_8642901308d141b18824b9b1be7fbd67",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_324a4cd5caaa4f8e93f5e0418cf2af3f",
              "IPY_MODEL_76f52d1b9b134f45985e1081379fc284"
            ]
          }
        },
        "8642901308d141b18824b9b1be7fbd67": {
          "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
          }
        },
        "324a4cd5caaa4f8e93f5e0418cf2af3f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_160e2251fd804c52b87772a1bdc3e65d",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 28,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 28,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_249187a4b71c49c187fb5bdec72f38ad"
          }
        },
        "76f52d1b9b134f45985e1081379fc284": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_24576d115d414f9281b36155fffdcfb1",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 28.0/28.0 [00:00&lt;00:00, 29.3B/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_e5cefc045f1541df82742ab049ed6550"
          }
        },
        "160e2251fd804c52b87772a1bdc3e65d": {
          "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"
          }
        },
        "249187a4b71c49c187fb5bdec72f38ad": {
          "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
          }
        },
        "24576d115d414f9281b36155fffdcfb1": {
          "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"
          }
        },
        "e5cefc045f1541df82742ab049ed6550": {
          "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
          }
        },
        "aa1103f1083146e993223cb81b8d6a16": {
          "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_4cefd8d709e549f1a9e0a9aaa6f9746f",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_8c87c7d45cc44ec48820e90adbadb9f3",
              "IPY_MODEL_a53fd45b1285488cba578897f5aaef33"
            ]
          }
        },
        "4cefd8d709e549f1a9e0a9aaa6f9746f": {
          "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
          }
        },
        "8c87c7d45cc44ec48820e90adbadb9f3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_95e98095f01c4d6786449db558283554",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 466062,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 466062,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ad8c77c74cb647798945428188ae3076"
          }
        },
        "a53fd45b1285488cba578897f5aaef33": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_d17adcf5585441f799f50f31e9f22f0b",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 466k/466k [00:00&lt;00:00, 2.36MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_24d32f325d814e24b59a77ce6a1689c7"
          }
        },
        "95e98095f01c4d6786449db558283554": {
          "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"
          }
        },
        "ad8c77c74cb647798945428188ae3076": {
          "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
          }
        },
        "d17adcf5585441f799f50f31e9f22f0b": {
          "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"
          }
        },
        "24d32f325d814e24b59a77ce6a1689c7": {
          "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
          }
        },
        "b188b3e1c3fb41d4b3369a4991600730": {
          "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_d41af32265924bb0b0c52c73619f5da0",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_a1b349cc8d024169b382512b1f7ca498",
              "IPY_MODEL_08f2b5c8d6e04f07a9f03c3ca67190ac"
            ]
          }
        },
        "d41af32265924bb0b0c52c73619f5da0": {
          "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
          }
        },
        "a1b349cc8d024169b382512b1f7ca498": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_16c09edb42ba4baeb6a7f6edba9e4bd4",
            "_dom_classes": [],
            "description": "Downloading: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 570,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 570,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_59cd169e83bd4aa58b831d5486f4e71a"
          }
        },
        "08f2b5c8d6e04f07a9f03c3ca67190ac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_157722dbca944a2e92d39a230c2531c7",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 570/570 [00:11&lt;00:00, 51.1B/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_3c939e4f38f149f881e8cd36601c94e8"
          }
        },
        "16c09edb42ba4baeb6a7f6edba9e4bd4": {
          "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"
          }
        },
        "59cd169e83bd4aa58b831d5486f4e71a": {
          "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
          }
        },
        "157722dbca944a2e92d39a230c2531c7": {
          "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"
          }
        },
        "3c939e4f38f149f881e8cd36601c94e8": {
          "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
          }
        },
        "26c12a498c8c46ab808fa4a359c1bec2": {
          "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_7a52c21d201f4cbb99584164c1696c52",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_3f81634143ad415c9777eab1b3774140",
              "IPY_MODEL_adaad589256449538f8272f863b11d2d"
            ]
          }
        },
        "7a52c21d201f4cbb99584164c1696c52": {
          "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
          }
        },
        "3f81634143ad415c9777eab1b3774140": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_0ff31244405f45d9815c7e3b0c139a36",
            "_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_aeab0ee4dfea4594b1b5ba8a6f0cf4aa"
          }
        },
        "adaad589256449538f8272f863b11d2d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_995cb3816f344adbb5fbc4f1108d07d2",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 536M/536M [00:10&lt;00:00, 50.6MB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0be58b4d9e984f6e9bfc6f8184e9267d"
          }
        },
        "0ff31244405f45d9815c7e3b0c139a36": {
          "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"
          }
        },
        "aeab0ee4dfea4594b1b5ba8a6f0cf4aa": {
          "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
          }
        },
        "995cb3816f344adbb5fbc4f1108d07d2": {
          "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"
          }
        },
        "0be58b4d9e984f6e9bfc6f8184e9267d": {
          "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/abyanjan/Fine-Tune-BERT-for-Text-Classification/blob/master/Fine_tune_bert.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NPXtEgPQ74qY"
      },
      "source": [
        "## FINE TUNIG BERT FOR TWITTER SENTIMENT CLASSIFICATION WITH HUGGING FACE TRANSFORMERS AND TENSORFLOW"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9wAuQ09A7z4R",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "199389f1-4c64-495a-ef26-da994a16493b"
      },
      "source": [
        "!pip install -q transformers"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[K     |████████████████████████████████| 2.5MB 8.1MB/s \n",
            "\u001b[K     |████████████████████████████████| 901kB 56.7MB/s \n",
            "\u001b[K     |████████████████████████████████| 3.3MB 55.3MB/s \n",
            "\u001b[?25h"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UZT4DM318SDw",
        "outputId": "43523988-aaf3-4bdc-ac20-45380dbfc97d"
      },
      "source": [
        "# check for the GPU\n",
        "!nvidia-smi"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mon Jun 21 08:53:57 2021       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 465.27       Driver Version: 460.32.03    CUDA Version: 11.2     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   45C    P8    10W /  70W |      0MiB / 15109MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tOvp7xA_8nQw"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KWv3ToOC8o6U"
      },
      "source": [
        "### Download the Data From Kaggle"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jgIFAhJ_nrwt"
      },
      "source": [
        "The data is related to disaster tweets that can be accessed at kaggle https://www.kaggle.com/c/nlp-getting-started"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZJW_PjT98r73"
      },
      "source": [
        "!pip install -q kaggle"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0RlglPqb8wfK",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 91
        },
        "outputId": "955647b0-8270-4d8b-db6c-0802f028f809"
      },
      "source": [
        "# upload the kaggle.json file that has the api key\n",
        "from google.colab import files\n",
        "files.upload()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-3f6fd1a3-1adf-47a8-b949-f8366622a463\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-3f6fd1a3-1adf-47a8-b949-f8366622a463\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving kaggle.json to kaggle.json\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'kaggle.json': b'{\"username\":\"ajaybyanjankar\",\"key\":\"19ec23eb4e8034d01ffb9a308f200617\"}'}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5ltOtjTn9r7u"
      },
      "source": [
        "!mkdir ~/.kaggle"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qLJw0XoK9v05"
      },
      "source": [
        "! cp kaggle.json ~/.kaggle/"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fI1WR8IV9x92"
      },
      "source": [
        "! chmod 600 ~/.kaggle/kaggle.json"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "opJeOd3F90ZU",
        "outputId": "212fb0da-5c87-4934-b7dc-c1d1c1ae011b"
      },
      "source": [
        "# download 'nlp-getting-started' from kaggle\n",
        "! kaggle competitions download -c 'nlp-getting-started'"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Warning: Looks like you're using an outdated API Version, please consider updating (server 1.5.12 / client 1.5.4)\n",
            "Downloading sample_submission.csv to /content\n",
            "  0% 0.00/22.2k [00:00<?, ?B/s]\n",
            "100% 22.2k/22.2k [00:00<00:00, 19.7MB/s]\n",
            "Downloading train.csv to /content\n",
            "  0% 0.00/965k [00:00<?, ?B/s]\n",
            "100% 965k/965k [00:00<00:00, 64.0MB/s]\n",
            "Downloading test.csv to /content\n",
            "  0% 0.00/411k [00:00<?, ?B/s]\n",
            "100% 411k/411k [00:00<00:00, 137MB/s]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KFqpOu8595pZ"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BgGUrWiN-GNo"
      },
      "source": [
        "### Read and Explore Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZVYYGoLW-QoL"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P1EufwG0-WEz"
      },
      "source": [
        "train_df = pd.read_csv('train.csv')\n",
        "#test_df = pd.read_csv('test.csv')"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BpMann9q-gEW",
        "outputId": "eee995b8-c617-40c2-a0b9-1173f84abfd2"
      },
      "source": [
        "# check the size of the data\n",
        "print(\"Train data size \\n\", train_df.shape )\n",
        "print()\n",
        "#print(\"Test data size \\n\", test_df.shape )"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train data size \n",
            " (7613, 5)\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "GgLuLYMk-0L1",
        "outputId": "e9c16fa7-fd0a-4b9b-c201-312938e6967a"
      },
      "source": [
        "# check data sample\n",
        "train_df.head()"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>keyword</th>\n",
              "      <th>location</th>\n",
              "      <th>text</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Our Deeds are the Reason of this #earthquake M...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Forest fire near La Ronge Sask. Canada</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>All residents asked to 'shelter in place' are ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>13,000 people receive #wildfires evacuation or...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>7</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Just got sent this photo from Ruby #Alaska as ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   id keyword  ...                                               text target\n",
              "0   1     NaN  ...  Our Deeds are the Reason of this #earthquake M...      1\n",
              "1   4     NaN  ...             Forest fire near La Ronge Sask. Canada      1\n",
              "2   5     NaN  ...  All residents asked to 'shelter in place' are ...      1\n",
              "3   6     NaN  ...  13,000 people receive #wildfires evacuation or...      1\n",
              "4   7     NaN  ...  Just got sent this photo from Ruby #Alaska as ...      1\n",
              "\n",
              "[5 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fuSDpLMp_2qw"
      },
      "source": [
        "#test_df.head()"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ix0rTZUR_7Mx"
      },
      "source": [
        "We see that for test data we do not have the 'target' columns. We will need to predict the target for the test data and submit to kaggle to check for our score.\n",
        "\n",
        "So, for the rest of the part , we will be only working with the train data to train our model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8UAabEFX-6vU",
        "outputId": "94e43001-5d81-43f5-e3d4-978918dbd7c3"
      },
      "source": [
        "# checking for any duplicates in the text columns\n",
        "train_df['text'].duplicated().sum()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "110"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3EkGO13L_f3i",
        "outputId": "79d1d641-4fd4-4d6c-c1e0-53f6a19c8a94"
      },
      "source": [
        "# dropping the duplicates\n",
        "train_df.drop_duplicates('text', keep='first')\n",
        "train_df.shape"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(7613, 5)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "3dOV8yZ__lns",
        "outputId": "a8c3641d-799a-41db-bebb-d0b35679c91f"
      },
      "source": [
        "# checking the target distribution\n",
        "train_df.target.value_counts().plot(kind = 'bar')"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f436d49b8d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XBDd5Mk7BP6W"
      },
      "source": [
        "We see that there is a balanaced distribution of the target levels."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q6Qtbh3VoVlq",
        "outputId": "dd7e9e80-b2c9-4404-d29d-f039220b4e70"
      },
      "source": [
        "# checking some of the positive disaster tweets\n",
        "for i in range(5):\n",
        "  print(train_df[train_df.target==1]['text'].iloc[i])"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Our Deeds are the Reason of this #earthquake May ALLAH Forgive us all\n",
            "Forest fire near La Ronge Sask. Canada\n",
            "All residents asked to 'shelter in place' are being notified by officers. No other evacuation or shelter in place orders are expected\n",
            "13,000 people receive #wildfires evacuation orders in California \n",
            "Just got sent this photo from Ruby #Alaska as smoke from #wildfires pours into a school \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pjRAJH0WoVM_",
        "outputId": "39e01d82-8e22-4efd-8d78-49375b398e9f"
      },
      "source": [
        "# checking some of the neagtive disaster tweets\n",
        "for i in range(5):\n",
        "  print(train_df[train_df.target==0]['text'].iloc[i])"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "What's up man?\n",
            "I love fruits\n",
            "Summer is lovely\n",
            "My car is so fast\n",
            "What a goooooooaaaaaal!!!!!!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r3L-hRrnoVJU"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zMKS5NBOBIje"
      },
      "source": [
        "# Exploring the text lengths\n",
        "textlen = train_df['text'].apply(lambda x: len(x.split()))"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 320
        },
        "id": "ZcFESdHhBcL7",
        "outputId": "644d32e8-bf54-40c4-b0a9-a7eb82675cfb"
      },
      "source": [
        "# vsualize the text length\n",
        "import seaborn as sns\n",
        "plt.figure(figsize=(8,5))\n",
        "plt.hist(textlen, edgecolor = \"black\");"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KTeC06D0EbYy"
      },
      "source": [
        "We can set the maximum sequence length as 30, for training our bert model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zqWC8wulB8uP"
      },
      "source": [
        "SEQ_LEN = 30"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LN9ZY1P1DS0V"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_CuranppEw6K"
      },
      "source": [
        "### Fine Tuning BERT"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zIwtKn3lE0HW"
      },
      "source": [
        "from transformers import BertTokenizer\n",
        "from transformers import AutoTokenizer\n",
        "import tensorflow as tf"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jmgGMtq3E8Uk"
      },
      "source": [
        "# download the tokenizer\n",
        "#task = \"sentiment\"\n",
        "#MODEL = f\"cardiffnlp/twitter-roberta-base-{task}\"\n",
        "#tokenizer = AutoTokenizer.from_pretrained(MODEL, do_lower_case = True)\n",
        "#tokenizer = BertTokenizer.from_pretrained('bert-base-uncased', do_lower_case=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 165,
          "referenced_widgets": [
            "7a38090ea79c4b2485416a636fff494a",
            "475a21b11b464c68a9e44a834c10ccd0",
            "9c3532e6eeba45cca6370cd58ea32707",
            "397ecf03fb664c809fe6069c497a0c43",
            "8bf8f7aac81a471fb929aa45a03d1a0b",
            "8753db13c9434f21b903dd18e80c3be0",
            "534fb12c064640888437f003c4f0785c",
            "c3934f64d2a040218f42f3a721c330ef",
            "4b6b7e5077104adf873dd48f96efd9c1",
            "8642901308d141b18824b9b1be7fbd67",
            "324a4cd5caaa4f8e93f5e0418cf2af3f",
            "76f52d1b9b134f45985e1081379fc284",
            "160e2251fd804c52b87772a1bdc3e65d",
            "249187a4b71c49c187fb5bdec72f38ad",
            "24576d115d414f9281b36155fffdcfb1",
            "e5cefc045f1541df82742ab049ed6550",
            "aa1103f1083146e993223cb81b8d6a16",
            "4cefd8d709e549f1a9e0a9aaa6f9746f",
            "8c87c7d45cc44ec48820e90adbadb9f3",
            "a53fd45b1285488cba578897f5aaef33",
            "95e98095f01c4d6786449db558283554",
            "ad8c77c74cb647798945428188ae3076",
            "d17adcf5585441f799f50f31e9f22f0b",
            "24d32f325d814e24b59a77ce6a1689c7"
          ]
        },
        "id": "jq2Na38KpMN9",
        "outputId": "4cf6e1c4-bd9e-47b6-bef1-fe3c588e6da4"
      },
      "source": [
        "tokenizer = BertTokenizer.from_pretrained('bert-base-uncased', do_lower_case=True)"
      ],
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "7a38090ea79c4b2485416a636fff494a",
              "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"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "4b6b7e5077104adf873dd48f96efd9c1",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=28.0, style=ProgressStyle(description_w…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "aa1103f1083146e993223cb81b8d6a16",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=466062.0, style=ProgressStyle(descripti…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cHlY89FsHIwr"
      },
      "source": [
        "Before tokenizing the text, we can do some text cleaning to remove the user name and any http links in the tweet text."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k68cUTznGwKz"
      },
      "source": [
        "# Preprocess text (username and link placeholders)\n",
        "def preprocess(text):\n",
        "    new_text = []\n",
        "\n",
        "    for t in text.split(\" \"):\n",
        "        t = '' if t.startswith('@') and len(t) > 1 else t\n",
        "        t = '' if t.startswith('http') else t\n",
        "        t = t.replace(\"#\",\"\")\n",
        "        new_text.append(t.lower())\n",
        "    return \" \".join(new_text).strip().replace(\"  \", \" \")"
      ],
      "execution_count": 94,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "xegKSyaaG79v",
        "outputId": "bc75ea9b-9953-4a73-d40b-30f86f55c697"
      },
      "source": [
        "train_df['text'] = train_df['text'].apply(preprocess)\n",
        "train_df.head()"
      ],
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>keyword</th>\n",
              "      <th>location</th>\n",
              "      <th>text</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>our deeds are the reason of this earthquake ma...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>forest fire near la ronge sask. canada</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>5</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>all residents asked to 'shelter in place' are ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>13,000 people receive wildfires evacuation ord...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>7</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>just got sent this photo from ruby alaska as s...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   id keyword  ...                                               text target\n",
              "0   1     NaN  ...  our deeds are the reason of this earthquake ma...      1\n",
              "1   4     NaN  ...             forest fire near la ronge sask. canada      1\n",
              "2   5     NaN  ...  all residents asked to 'shelter in place' are ...      1\n",
              "3   6     NaN  ...  13,000 people receive wildfires evacuation ord...      1\n",
              "4   7     NaN  ...  just got sent this photo from ruby alaska as s...      1\n",
              "\n",
              "[5 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lq2eKdq-myGD"
      },
      "source": [
        "**Encoding the text for Bert model**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XdZRhu9YAQFz",
        "outputId": "5c00743a-9f57-424b-f251-9bd98fb098bb"
      },
      "source": [
        "# example of tokenization\n",
        "tokenizer.tokenize(preprocess(\"Our Deeds are the Reason of this #earthquake May ALLAH Forgive us all\"))"
      ],
      "execution_count": 97,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['our',\n",
              " 'deeds',\n",
              " 'are',\n",
              " 'the',\n",
              " 'reason',\n",
              " 'of',\n",
              " 'this',\n",
              " 'earthquake',\n",
              " 'may',\n",
              " 'allah',\n",
              " 'forgive',\n",
              " 'us',\n",
              " 'all']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 97
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N7nd6_cJlUtB",
        "outputId": "a6881b86-ca19-4cc5-cdd5-8c3a2d7a62b9"
      },
      "source": [
        "# example of an output from encoded text by tokeneizer\n",
        "encoded_input = tokenizer(preprocess(\"Our Deeds are the Reason of this #earthquake May ALLAH Forgive us all\"))\n",
        "print(encoded_input)"
      ],
      "execution_count": 98,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'input_ids': [101, 2256, 15616, 2024, 1996, 3114, 1997, 2023, 8372, 2089, 16455, 9641, 2149, 2035, 102], 'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "kuqxiulAnjSx",
        "outputId": "aee3b154-69c8-463d-ca3b-574062567cd9"
      },
      "source": [
        "tokenizer.decode(encoded_input['input_ids'])"
      ],
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'[CLS] our deeds are the reason of this earthquake may allah forgive us all [SEP]'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oVWMCHr1ni18"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A-C3BevIJP2R"
      },
      "source": [
        "# encoding the training data\n",
        "train_encoded_inputs = tokenizer(train_df['text'].tolist(),\n",
        "                                 add_special_tokens = True,\n",
        "                                 padding='max_length', \n",
        "                                 truncation=True, \n",
        "                                 max_length=SEQ_LEN, \n",
        "                                 return_token_type_ids=False,\n",
        "                                 return_tensors = 'tf')\n"
      ],
      "execution_count": 100,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L1zwEX2vKXD6",
        "outputId": "67532693-1b9a-4b91-94bd-4bdea96510e2"
      },
      "source": [
        "train_encoded_inputs"
      ],
      "execution_count": 101,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'input_ids': <tf.Tensor: shape=(7613, 30), dtype=int32, numpy=\n",
              "array([[  101,  2256, 15616, ...,     0,     0,     0],\n",
              "       [  101,  3224,  2543, ...,     0,     0,     0],\n",
              "       [  101,  2035,  3901, ...,     0,     0,     0],\n",
              "       ...,\n",
              "       [  101, 23290,  1012, ...,     0,     0,     0],\n",
              "       [  101,  2610, 11538, ...,  1012,   102,     0],\n",
              "       [  101,  1996,  6745, ...,     0,     0,     0]], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(7613, 30), dtype=int32, numpy=\n",
              "array([[1, 1, 1, ..., 0, 0, 0],\n",
              "       [1, 1, 1, ..., 0, 0, 0],\n",
              "       [1, 1, 1, ..., 0, 0, 0],\n",
              "       ...,\n",
              "       [1, 1, 1, ..., 0, 0, 0],\n",
              "       [1, 1, 1, ..., 1, 1, 0],\n",
              "       [1, 1, 1, ..., 0, 0, 0]], dtype=int32)>}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 101
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iKEX0E6HKZ2w"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xubseOGBmgp9"
      },
      "source": [
        "**Creating Tensorflow data**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9xdrlomFk5-W"
      },
      "source": [
        "train_dataset = tf.data.Dataset.from_tensor_slices((train_encoded_inputs, train_df['target'].values))"
      ],
      "execution_count": 102,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LzeoWqTuyYdQ"
      },
      "source": [
        "# formatting the data as required by bert model\n",
        "def map_bert(inputs, labels):\n",
        "  inputs = {'input_ids': inputs['input_ids'],\n",
        "            'attention_mask': inputs['attention_mask']}\n",
        "  \n",
        "  return inputs, labels"
      ],
      "execution_count": 103,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "stXmyneoyYas"
      },
      "source": [
        "train_dataset = train_dataset.map(map_bert)"
      ],
      "execution_count": 104,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zd5WA6B2yYXY"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "If8RQM_HKrBk",
        "outputId": "ea9d6820-f9aa-4e44-a8df-7cb9fdacf003"
      },
      "source": [
        "for t in train_dataset.take(2):\n",
        "  print (t)"
      ],
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "({'input_ids': <tf.Tensor: shape=(30,), dtype=int32, numpy=\n",
            "array([  101,  2256, 15616,  2024,  1996,  3114,  1997,  2023,  8372,\n",
            "        2089, 16455,  9641,  2149,  2035,   102,     0,     0,     0,\n",
            "           0,     0,     0,     0,     0,     0,     0,     0,     0,\n",
            "           0,     0,     0], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(30,), dtype=int32, numpy=\n",
            "array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,\n",
            "       0, 0, 0, 0, 0, 0, 0, 0], dtype=int32)>}, <tf.Tensor: shape=(), dtype=int64, numpy=1>)\n",
            "({'input_ids': <tf.Tensor: shape=(30,), dtype=int32, numpy=\n",
            "array([  101,  3224,  2543,  2379,  2474,  6902,  3351, 21871,  2243,\n",
            "        1012,  2710,   102,     0,     0,     0,     0,     0,     0,\n",
            "           0,     0,     0,     0,     0,     0,     0,     0,     0,\n",
            "           0,     0,     0], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(30,), dtype=int32, numpy=\n",
            "array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
            "       0, 0, 0, 0, 0, 0, 0, 0], dtype=int32)>}, <tf.Tensor: shape=(), dtype=int64, numpy=1>)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NfWrWzPDK4HZ"
      },
      "source": [
        "**Creating train and test split**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FkCeVxOHK7nQ"
      },
      "source": [
        "dataset = train_dataset.shuffle(100000).batch(64)"
      ],
      "execution_count": 106,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qFPHCdZjLemW",
        "outputId": "e3dbb353-f7cc-42fc-bd84-3bf2ef4e1bd5"
      },
      "source": [
        "# length of the dataset(total batches)\n",
        "DS_LEN = len(dataset)\n",
        "DS_LEN"
      ],
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "119"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 107
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CbMbmJLeLq1K"
      },
      "source": [
        "# take 80% for train and 20% for validation\n",
        "SPLIT = 0.8\n",
        "train_ds = dataset.take(round(DS_LEN*SPLIT))\n",
        "val_ds = dataset.skip(round(DS_LEN*SPLIT))"
      ],
      "execution_count": 108,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vB47QL16MTZE"
      },
      "source": [
        "**Modeling**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p5iYGGNZMVA1"
      },
      "source": [
        "from transformers import TFAutoModel"
      ],
      "execution_count": 109,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 223,
          "referenced_widgets": [
            "b188b3e1c3fb41d4b3369a4991600730",
            "d41af32265924bb0b0c52c73619f5da0",
            "a1b349cc8d024169b382512b1f7ca498",
            "08f2b5c8d6e04f07a9f03c3ca67190ac",
            "16c09edb42ba4baeb6a7f6edba9e4bd4",
            "59cd169e83bd4aa58b831d5486f4e71a",
            "157722dbca944a2e92d39a230c2531c7",
            "3c939e4f38f149f881e8cd36601c94e8",
            "26c12a498c8c46ab808fa4a359c1bec2",
            "7a52c21d201f4cbb99584164c1696c52",
            "3f81634143ad415c9777eab1b3774140",
            "adaad589256449538f8272f863b11d2d",
            "0ff31244405f45d9815c7e3b0c139a36",
            "aeab0ee4dfea4594b1b5ba8a6f0cf4aa",
            "995cb3816f344adbb5fbc4f1108d07d2",
            "0be58b4d9e984f6e9bfc6f8184e9267d"
          ]
        },
        "id": "W26Wkju6MpqI",
        "outputId": "6fe5bc82-6ca6-4d43-c54c-6d8e89715aaf"
      },
      "source": [
        "bert = TFAutoModel.from_pretrained('bert-base-uncased')"
      ],
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "b188b3e1c3fb41d4b3369a4991600730",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Downloading', max=570.0, 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": "26c12a498c8c46ab808fa4a359c1bec2",
              "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": [
            "Some layers from the model checkpoint at bert-base-uncased were not used when initializing TFBertModel: ['mlm___cls', 'nsp___cls']\n",
            "- This IS expected if you are initializing TFBertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
            "- This IS NOT expected if you are initializing TFBertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
            "All the layers of TFBertModel were initialized from the model checkpoint at bert-base-uncased.\n",
            "If your task is similar to the task the model of the checkpoint was trained on, you can already use TFBertModel for predictions without further training.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "387VrYOQMuyf",
        "outputId": "5913ae9f-956d-4d65-d9cd-e389a626345a"
      },
      "source": [
        "# create model architecture\n",
        "#n_classes = len((train_df.target.unique()))\n",
        "\n",
        "# Input layers\n",
        "input_ids = tf.keras.layers.Input(shape=(SEQ_LEN,), dtype=np.int32, name='input_ids' )\n",
        "mask = tf.keras.layers.Input(shape=(SEQ_LEN,), dtype=np.int32, name = 'attention_mask')\n",
        "\n",
        "# bert embeddings\n",
        "embeddings = bert([input_ids, mask])[0]\n",
        "cls_token = embeddings[:,0,:]\n",
        "\n",
        "# keras layers\n",
        "#x = tf.keras.layers.GlobalMaxPool1D()(embeddings)\n",
        "x = tf.keras.layers.BatchNormalization()(cls_token)\n",
        "x = tf.keras.layers.Dense(128, activation='relu')(x)\n",
        "x = tf.keras.layers.Dropout(0.2)(x)\n",
        "x = tf.keras.layers.Dense(32, activation='relu')(x)\n",
        "\n",
        "# output layer\n",
        "y = tf.keras.layers.Dense(1, activation='sigmoid')(x)\n",
        "\n",
        "# create the model\n",
        "model = tf.keras.Model(inputs=[input_ids, mask], outputs=y)"
      ],
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:The parameters `output_attentions`, `output_hidden_states` and `use_cache` cannot be updated when calling a model.They have to be set to True/False in the config object (i.e.: `config=XConfig.from_pretrained('name', output_attentions=True)`).\n",
            "WARNING:tensorflow:The parameter `return_dict` cannot be set in graph mode and will always be set to `True`.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QQn5mn5rOQLE",
        "outputId": "0ed59957-8707-4541-d949-ba91834fb1f1"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 114,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model_5\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_ids (InputLayer)          [(None, 30)]         0                                            \n",
            "__________________________________________________________________________________________________\n",
            "attention_mask (InputLayer)     [(None, 30)]         0                                            \n",
            "__________________________________________________________________________________________________\n",
            "tf_bert_model (TFBertModel)     TFBaseModelOutputWit 109482240   input_ids[0][0]                  \n",
            "                                                                 attention_mask[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "tf.__operators__.getitem_5 (Sli (None, 768)          0           tf_bert_model[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_5 (BatchNor (None, 768)          3072        tf.__operators__.getitem_5[0][0] \n",
            "__________________________________________________________________________________________________\n",
            "dense_15 (Dense)                (None, 128)          98432       batch_normalization_5[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "dropout_227 (Dropout)           (None, 128)          0           dense_15[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "dense_16 (Dense)                (None, 32)           4128        dropout_227[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "dense_17 (Dense)                (None, 1)            33          dense_16[0][0]                   \n",
            "==================================================================================================\n",
            "Total params: 109,587,905\n",
            "Trainable params: 109,586,369\n",
            "Non-trainable params: 1,536\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d2wjfg61OiU0",
        "outputId": "2607f83b-1477-4202-dbbf-4153891685c0"
      },
      "source": [
        "# freezing the pretrained bert layer\n",
        "model.layers[2].trainable = False\n",
        "model.summary()"
      ],
      "execution_count": 121,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model_6\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_ids (InputLayer)          [(None, 30)]         0                                            \n",
            "__________________________________________________________________________________________________\n",
            "attention_mask (InputLayer)     [(None, 30)]         0                                            \n",
            "__________________________________________________________________________________________________\n",
            "tf_bert_model (TFBertModel)     TFBaseModelOutputWit 109482240   input_ids[0][0]                  \n",
            "                                                                 attention_mask[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "tf.__operators__.getitem_6 (Sli (None, 768)          0           tf_bert_model[1][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_6 (BatchNor (None, 768)          3072        tf.__operators__.getitem_6[0][0] \n",
            "__________________________________________________________________________________________________\n",
            "dense_18 (Dense)                (None, 128)          98432       batch_normalization_6[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "dropout_228 (Dropout)           (None, 128)          0           dense_18[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "dense_19 (Dense)                (None, 32)           4128        dropout_228[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "dense_20 (Dense)                (None, 1)            33          dense_19[0][0]                   \n",
            "==================================================================================================\n",
            "Total params: 109,587,905\n",
            "Trainable params: 104,129\n",
            "Non-trainable params: 109,483,776\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6pRv1z9FQy3h"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nZ9tUK58kUbU"
      },
      "source": [
        "**Compile model**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZUFvIPoOSPhz"
      },
      "source": [
        "import transformers"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QlKXGm_tkQiX"
      },
      "source": [
        "#learning_rate = 1e-3\n",
        "\n",
        "optimizer = tf.keras.optimizers.Adam(learning_rate=5e-4)\n",
        "#optimizer = transformers.AdamWeightDecay(learning_rate=5e-4)\n",
        "loss = tf.keras.losses.BinaryCrossentropy()\n",
        "metric = tf.keras.metrics.BinaryAccuracy('accuracy')\n",
        "\n",
        "model.compile(optimizer=optimizer, loss=loss, metrics=[metric])"
      ],
      "execution_count": 122,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XXaY3lgRSY0Y",
        "outputId": "897412a7-e191-485b-eb71-6a7c7f6c394a"
      },
      "source": [
        "# train the model\n",
        "callbacks = [tf.keras.callbacks.ReduceLROnPlateau(patience=2, factor=0.1,min_delta=0.001,monitor='val_loss'),\n",
        "             tf.keras.callbacks.EarlyStopping(patience=5, min_delta=0.001, monitor='val_loss')]\n",
        "\n",
        "history = model.fit(\n",
        "    train_ds,\n",
        "    validation_data = val_ds,\n",
        "    epochs = 50,\n",
        "    callbacks = callbacks\n",
        ")"
      ],
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n",
            "WARNING:tensorflow:The parameters `output_attentions`, `output_hidden_states` and `use_cache` cannot be updated when calling a model.They have to be set to True/False in the config object (i.e.: `config=XConfig.from_pretrained('name', output_attentions=True)`).\n",
            "WARNING:tensorflow:The parameter `return_dict` cannot be set in graph mode and will always be set to `True`.\n",
            "WARNING:tensorflow:The parameters `output_attentions`, `output_hidden_states` and `use_cache` cannot be updated when calling a model.They have to be set to True/False in the config object (i.e.: `config=XConfig.from_pretrained('name', output_attentions=True)`).\n",
            "WARNING:tensorflow:The parameter `return_dict` cannot be set in graph mode and will always be set to `True`.\n",
            "95/95 [==============================] - ETA: 0s - loss: 0.5507 - accuracy: 0.7363WARNING:tensorflow:The parameters `output_attentions`, `output_hidden_states` and `use_cache` cannot be updated when calling a model.They have to be set to True/False in the config object (i.e.: `config=XConfig.from_pretrained('name', output_attentions=True)`).\n",
            "WARNING:tensorflow:The parameter `return_dict` cannot be set in graph mode and will always be set to `True`.\n",
            "95/95 [==============================] - 31s 211ms/step - loss: 0.5507 - accuracy: 0.7363 - val_loss: 0.5299 - val_accuracy: 0.7776\n",
            "Epoch 2/50\n",
            "95/95 [==============================] - 17s 181ms/step - loss: 0.4612 - accuracy: 0.7845 - val_loss: 0.4500 - val_accuracy: 0.8226\n",
            "Epoch 3/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.4445 - accuracy: 0.7992 - val_loss: 0.4038 - val_accuracy: 0.8258\n",
            "Epoch 4/50\n",
            "95/95 [==============================] - 17s 176ms/step - loss: 0.4348 - accuracy: 0.8023 - val_loss: 0.3919 - val_accuracy: 0.8278\n",
            "Epoch 5/50\n",
            "95/95 [==============================] - 17s 176ms/step - loss: 0.4279 - accuracy: 0.8051 - val_loss: 0.3676 - val_accuracy: 0.8454\n",
            "Epoch 6/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.4202 - accuracy: 0.8151 - val_loss: 0.3398 - val_accuracy: 0.8637\n",
            "Epoch 7/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.4106 - accuracy: 0.8153 - val_loss: 0.3508 - val_accuracy: 0.8500\n",
            "Epoch 8/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.4065 - accuracy: 0.8238 - val_loss: 0.3172 - val_accuracy: 0.8682\n",
            "Epoch 9/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.4041 - accuracy: 0.8229 - val_loss: 0.3342 - val_accuracy: 0.8584\n",
            "Epoch 10/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3990 - accuracy: 0.8263 - val_loss: 0.2934 - val_accuracy: 0.8826\n",
            "Epoch 11/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3792 - accuracy: 0.8363 - val_loss: 0.3185 - val_accuracy: 0.8584\n",
            "Epoch 12/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.3806 - accuracy: 0.8367 - val_loss: 0.3002 - val_accuracy: 0.8813\n",
            "Epoch 13/50\n",
            "95/95 [==============================] - 17s 176ms/step - loss: 0.3662 - accuracy: 0.8398 - val_loss: 0.2958 - val_accuracy: 0.8865\n",
            "Epoch 14/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3806 - accuracy: 0.8378 - val_loss: 0.3134 - val_accuracy: 0.8702\n",
            "Epoch 15/50\n",
            "95/95 [==============================] - 17s 179ms/step - loss: 0.3627 - accuracy: 0.8428 - val_loss: 0.2920 - val_accuracy: 0.8806\n",
            "Epoch 16/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3750 - accuracy: 0.8354 - val_loss: 0.2984 - val_accuracy: 0.8748\n",
            "Epoch 17/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.3656 - accuracy: 0.8391 - val_loss: 0.2980 - val_accuracy: 0.8819\n",
            "Epoch 18/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.3800 - accuracy: 0.8314 - val_loss: 0.2895 - val_accuracy: 0.8839\n",
            "Epoch 19/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3716 - accuracy: 0.8398 - val_loss: 0.3040 - val_accuracy: 0.8708\n",
            "Epoch 20/50\n",
            "95/95 [==============================] - 17s 179ms/step - loss: 0.3608 - accuracy: 0.8447 - val_loss: 0.2833 - val_accuracy: 0.8826\n",
            "Epoch 21/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3756 - accuracy: 0.8349 - val_loss: 0.2903 - val_accuracy: 0.8898\n",
            "Epoch 22/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.3625 - accuracy: 0.8385 - val_loss: 0.2808 - val_accuracy: 0.8839\n",
            "Epoch 23/50\n",
            "95/95 [==============================] - 17s 177ms/step - loss: 0.3638 - accuracy: 0.8426 - val_loss: 0.2898 - val_accuracy: 0.8845\n",
            "Epoch 24/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3667 - accuracy: 0.8395 - val_loss: 0.3027 - val_accuracy: 0.8721\n",
            "Epoch 25/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3732 - accuracy: 0.8431 - val_loss: 0.3042 - val_accuracy: 0.8748\n",
            "Epoch 26/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3721 - accuracy: 0.8360 - val_loss: 0.2959 - val_accuracy: 0.8819\n",
            "Epoch 27/50\n",
            "95/95 [==============================] - 17s 178ms/step - loss: 0.3743 - accuracy: 0.8350 - val_loss: 0.2879 - val_accuracy: 0.8780\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "id": "CZ41vAIGUHUT",
        "outputId": "a19c0cee-8b08-4162-80c9-6461cceacc13"
      },
      "source": [
        "# plot the performance curve\n",
        "epochs = history.epoch\n",
        "plt.figure(figsize=(15, 6))\n",
        "\n",
        "# Accuracy\n",
        "plt.subplot(1,2, 1)\n",
        "plt.plot(epochs, history.history['accuracy'], label=\"Train\")\n",
        "plt.plot(epochs, history.history['val_accuracy'], label = \"Val\")\n",
        "plt.legend()\n",
        "plt.title(\"Accuracy\")\n",
        "\n",
        "# loss\n",
        "plt.subplot(1,2, 2)\n",
        "plt.plot(epochs, history.history['loss'], label=\"Train\")\n",
        "plt.plot(epochs, history.history['val_loss'], label = \"Val\")\n",
        "plt.legend()\n",
        "plt.title(\"Loss\")\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 124,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3AAAAF1CAYAAACpuAhNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3SU1dbA4d9Jr5QkhJ5GQm+BCCJFEBBUil1ABCxYrr1cP3v32rter4oKqAjYUSkCgoCgECB0QkJJCDWFnoSUOd8fZ4IhpGcyJdnPWqxk3jZ7EmDePeecvZXWGiGEEEIIIYQQzs/N0QEIIYQQQgghhKgcSeCEEEIIIYQQwkVIAieEEEIIIYQQLkISOCGEEEIIIYRwEZLACSGEEEIIIYSLkAROCCGEEEIIIVyEJHBCCCGEEEII4SIkgROiEpRSS5VSR5RS3o6ORQghhHBlSqk9Sqkhjo5DCFclCZwQFVBKRQD9AQ2MsuPzetjruYQQQgghhGuQBE6Iik0A/gKmAhOLNiqlWiulvldKpSulMpVS7xfbN1kptU0pdUIptVUp1cO6XSuloosdN1Up9YL1+4FKqTSl1P8ppQ4CnyulGiulfrE+xxHr962KnR+klPpcKbXfuv9H6/bNSqmRxY7zVEplKKVia+2nJIQQQlSTUspbKfW29f1sv/V7b+u+EOv731GlVJZSarlSys267/+UUvus77eJSqnBjn0lQtQ+SeCEqNgE4Cvrn2FKqaZKKXfgFyAFiABaAjMBlFLXAM9Yz2uAGbXLrORzNQOCgHDgVsy/0c+tj8OAHOD9Ysd/AfgBnYBQ4C3r9unA+GLHXQoc0Fqvr2QcQgghhD09DpwPdAe6Ab2AJ6z7HgTSgCZAU+AxQCul2gF3AedprQOBYcAe+4YthP3JFC0hyqGU6odJnmZrrTOUUjuBcZgRuRbAv7XWBdbDV1i/3gK8qrVeY32cXIWntABPa61PWx/nAN8Vi+dFYIn1++bAJUCw1vqI9ZA/rF+/BJ5USjXQWh8HbsAke0IIIYQzuh64W2t9GEAp9SzwEfAkkA80B8K11snAcusxhYA30FEpla613uOIwIWwNxmBE6J8E4HftNYZ1sczrNtaAynFkrfiWgM7q/l86Vrr3KIHSik/pdRHSqkUpdRxYBnQyDoC2BrIKpa8naG13g/8CVyllGqESfS+qmZMQgghRG1rgZnVUiTFug3gNcyHob8ppXYppR4BsCZz92FmvRxWSs1USrVAiDpOEjghyqCU8gWuBS5USh20rku7HzO14xAQVkahkb1AmzIum42Z8likWYn9usTjB4F2QG+tdQNgQFF41ucJsiZopZmGmUZ5DbBKa72vjOOEEEIIR9uPmfFSJMy6Da31Ca31g1rrKMyyhAeK1rpprWdorYtmy2jgFfuGLYT9SQInRNkuBwqBjpg5+d2BDpipG5cDB4CXlVL+SikfpVRf63lTgIeUUj2VEa2UKnpTSgDGKaXclVLDgQsriCEQM43yqFIqCHi6aIfW+gAwD/ivtdiJp1JqQLFzfwR6APdi1sQJIYQQzsLT+t7po5TyAb4GnlBKNVFKhQBPYZYDoJQaYX0vVcAxzHuzRSnVTil1kbXYSS7m/dLimJcjhP1IAidE2SYCn2utU7XWB4v+YIqIjAVGAtFAKmZx9XUAWutvgBcx0y1PYBKpIOs177WedxQz3//HCmJ4G/AFMjDr7uaX2H8DZm3AduAwZioJ1jiK1s9FAt9X8bULIYQQtWkuJuEq+uMDxAMbgU3AOuAF67ExwCLgJLAK+K/Weglm/dvLmPfIg5hiXo/a7yUI4RhK65IztoQQdYVS6imgrdZ6fIUHCyGEEEIIpydVKIWoo6xTLm/GjNIJIYQQQog6QKZQClEHKaUmY4qczNNaL3N0PEIIIYQQwjZkCqUQQgghhBBCuAgZgRNCCCGEEEIIFyEJnBBCCCGEEEK4CKcrYhISEqIjIiIcHYYQQgg7WLt2bYbWuomj43AV8h4phBD1Q3nvj06XwEVERBAfH+/oMIQQQtiBUirF0TG4EnmPFEKI+qG890eZQimEEEIIIYQQLkISOCGEEEIIIYRwEZLACSGEEEIIIYSLcLo1cKXJz88nLS2N3NxcR4dS63x8fGjVqhWenp6ODkUIIYQQQgi7k3v/8rlEApeWlkZgYCAREREopRwdTq3RWpOZmUlaWhqRkZGODkcIIYQQQgi7k3v/8rnEFMrc3FyCg4Pr9C8QQClFcHBwvfi0QQghhBBCiNLIvX/5XCKBA+r8L7BIfXmdQgghhBBClKW+3BNX53W6TALnSJmZmXTv3p3u3bvTrFkzWrZseeZxXl5euefGx8dzzz332ClSIYQQQgghRE04+72/S6yBc7Tg4GASEhIAeOaZZwgICOChhx46s7+goAAPj9J/lHFxccTFxdklTiGEEEIIIUTNOPu9v4zAVdOkSZO4/fbb6d27Nw8//DCrV6+mT58+xMbGcsEFF5CYmAjA0qVLGTFiBGD+Atx0000MHDiQqKgo3n33XUe+BCGEEEIIIUQlONO9v8uNwD378xa27j9u02t2bNGAp0d2qvJ5aWlprFy5End3d44fP87y5cvx8PBg0aJFPPbYY3z33XfnnLN9+3aWLFnCiRMnaNeuHXfccYe0DBBCCBellBoOvAO4A1O01i+X2D8JeA3YZ930vtZ6inVfIbDJuj1Vaz3KLkELIYQLkXv/c7lcAudMrrnmGtzd3QE4duwYEydOJCkpCaUU+fn5pZ5z2WWX4e3tjbe3N6GhoRw6dIhWrVrZM2whBEBGMgQ2Be9AR0ciXJRSyh34ABgKpAFrlFJztNZbSxw6S2t9VymXyNFad6/tOIvbdzSHrfuPM7RjU3s+rRBC1AnOcu/vcglcdbLl2uLv73/m+yeffJJBgwbxww8/sGfPHgYOHFjqOd7e3me+d3d3p6CgoLbDFEKUlDgPZo2HRuFw/TcQ3MbREQnX1AtI1lrvAlBKzQRGAyUTOKcxJ2E/r8zfzuZnhxHg7XK3AEKIekju/c8la+Bs5NixY7Rs2RKAqVOnOjYYIUTZdv4OsydAkw6QexSmDIaUlY6OSrimlsDeYo/TrNtKukoptVEp9a1SqnWx7T5KqXil1F9KqcvLehKl1K3W4+LT09NrFHBYkB8AqZnZNbqOEELUd46895cEzkYefvhhHn30UWJjY2VUTQhntedP+HochLSFiXPglkXgFwLTRsGGmY6OTtRNPwMRWuuuwEJgWrF94VrrOGAc8LZSqtShYK31x1rrOK11XJMmTWoUTHiwNYHLOlWj6wghRH3nyHt/pbW26xNWJC4uTsfHx5+1bdu2bXTo0MFBEdlffXu9QthFWjxMHw0NWsCkuRBgvRHOOQKzboA9y2HAwzDoMagnzUOdgVJqrTWJcTlKqT7AM1rrYdbHjwJorV8q43h3IEtr3bCUfVOBX7TW35b3nKW9R1bFsZx8uj37G49d2p5bB8jUYSGEc6pv98Klvd7y3h9lBE4IUfcd2AhfXgn+ITDhp3+SNwDfxjD+e4gdD8tehe9uhvxcx8UqXMkaIEYpFamU8gLGAHOKH6CUal7s4Shgm3V7Y6WUt/X7EKAvdlg719DXk0Z+nqTIFEohhHBZsoJZCFG3Hd4OX1wOXoEwYY4ZgSvJwwtGvQ/B0bDoGTi6F8Z+bRI+IcqgtS5QSt0FLMC0EfhMa71FKfUcEK+1ngPco5QaBRQAWcAk6+kdgI+UUhbMh6kvl1K9slaEB/mRmiUJnBBCuCpJ4ISwFUshLH0JtvwAk36FwGaOjkhk7jTTJt08zJq3xuFlH6sU9LsfGkfCD7fBJxeZCpVN2tkvXuFytNZzgbkltj1V7PtHgUdLOW8l0KXWAyxF6yA/Nu075oinFkIIYQMyhVIIWzhxyCQKy16DzGRYO9XREYmje83vpDDPTJusbKuATpebNXL5OTBlKOxaWqthCmFv4cF+7DuSQ0GhxdGhCCGEqAZJ4ISoqd3L4H/9TJGMyz+E6KEQ/xkU5Dk6svrrxEGYPgpyj8OEHyG0iguhW/WEyYuhYUv48ipJyEWdEh7kT4FFc+CYrPUUQghXJAmcENVlsZgRt+mjwachTP4duo+DXrfCyUOw/WdHR1g/ncowv5OTh2H8d9C8W/Wu0ygMbloAUQPh53vhtyfN77wqCk7D0VTI2l29GISoBa2tveCkkIkQQrgmSeAqadCgQSxYsOCsbW+//TZ33HFHqccPHDiQmpR6Fk7uVCbMuAZ+fwE6XwW3LoWmHc2+6CHQOAJWf+LAAOupnCOmYMmRPTBuFrQ+r2bX82kAY2dB3M2w8l2YfQOcPml+/4e2ws4lpn/cn+/Agsfhu1tg6gh4vxe8HA4vhMLbXeDdWEj9yyYvUYiaKuoFlyK94IQQokzOfO8vRUwqaezYscycOZNhw4ad2TZz5kxeffVVB0YlHCL1b/j2RjiVDiPegp43nt03zM0NzpsMvz1uytc37+q4WO0tL9v0U0taaNaOdb4KBp1Tv6F2nD4BX14N6YmmgmREP9tc190DLnsDQmJg/qPwUsvSj/PwhcCmENDMFD6JHAABTSEg1BS3WfC4aRwuPeaEgzVt4IOXu5tUohRCiHI4872/JHCVdPXVV/PEE0+Ql5eHl5cXe/bsYf/+/Xz99dc88MAD5OTkcPXVV/Pss886OlRRW7SGVe+bMvMNW8HNC6FF99KPjb3ejM6t+QRGvWfXMO1Ka8hIguRFkLwQ9vwJhafB088kLsvfMNNKy6v+aAt52TDjOti/Hq77woyC2pJScP4dENrRJKZFiVlAU1NtNCAUvALKTs7cPOCnf5kKpZ2vtG1sQlSRu5uiVZAvqTKFUgghyuTM9/6ul8DNewQObrLtNZt1gUteLveQoKAgevXqxbx58xg9ejQzZ87k2muv5bHHHiMoKIjCwkIGDx7Mxo0b6dq1Ho242FthASQtgHaX2nckI+cI/HgnJP4KHUbC6A/Murey+DaGrtfCxtkw5FnwC7JfrLUt75Qp3JK00CRtR1PN9pC2cN4tEDMEwi6A7EwzdXDpy3DFh7UXT0EezLoeUlfBVVOg/WW191xRF5o/VdVtDPz1oUn+218GHt42D02IqpBecEIIlyH3/ueQNXBVUDSUCmYIdezYscyePZsePXoQGxvLli1b2LrVLn1Y66/N38HMcSZ5sJd96+CjC03iOPxluPaL8pO3Ir0mQ0EOJHxV+zHWtoxkWPmeKQ7ySgR8Pcas/WraGS57E+7dCHetgeH/gTYXgaePqeDYazJsnGmaadeWle/Czt/NSGfnq2rveWrCzR0ufh6OpsDqjx0dzdm0huwsR0ch7CwsyI/UzGy01o4ORQghnJaz3vu73ghcBdlybRo9ejT3338/69atIzs7m6CgIF5//XXWrFlD48aNmTRpErm5Upa5ViX99s/XthfX7nNpDWumwILHwD8UbpxftaIYzbqYkag1U+D8f5mbeFe0/iv46U5AQ5MOpspmzFAI61PxSFK/B2DtNPj9eRhTC4nssX1mmmaHkRA73vbXt6U2g0yLiWWvQffrnWNU1lIIP94Bm76BHhNh0OMQ0MTRUQk7CAv258TpAo5k5xPk7+XocIQQomxy73+OSo3AKaWGK6USlVLJSqlHStkfppRaopRar5TaqJS61LrdUyk1TSm1SSm1TSllp2oGtSMgIIBBgwZx0003MXbsWI4fP46/vz8NGzbk0KFDzJs3z9Eh1m2WQti52HyfvNAkWLVp6Usw9yFTRv725dWraNhrsqmImLzIxsHZyebvYM5d5mdw32a48y8Y9qJ5XJlpgP7BcMHdsP0X0yfP1hY+BdoCF79o+2vXhqHPmWIry15zdCTm388v98PGWRB5Iaz/At7rYSpqFpx2dHSiloVbWwnINEohhCibs977V5jAKaXcgQ+AS4COwFilVMcShz0BzNZaxwJjgP9at18DeGutuwA9gduUUhG2Cd0xxo4dy4YNGxg7dizdunUjNjaW9u3bM27cOPr27evo8Oq2fevMWrSwC0xSlLmz9p7LYjHNm2OGmTLy1R0t6TDSVCWsjWlzm7410xhry/Zf4ftbofX5MGYGNGpdvev0+Rf4hcBiGy/yTVkJm7+FvvfWfpEUW2naEWJvMC0mavPvb0W0hvmPwLpp0P9B0+z8jlVmVHXhU/BBL9j2c+1/SCIcJqyolUCmtBIQQojyOOO9f2WmUPYCkrXWuwCUUjOB0UDxCZ8aaGD9viGwv9h2f6WUB+AL5AHHbRC3w1x++eVnrRmYOnVqqcctXbrUPgHVJ8kLQbmZdVYfDzSPQ6Jr57kOrDfNuDtfZdoCVJe7J8TdBEv/Y27Yg9vYKL4NJrnShbBvLQx7yZS7t5XkRfDNJNMEe9ws8PKr/rW8A02SsOBR0zetzaCax2cphLkPQ4NW0Pe+ml/PngY9bpLvxc/CtdMdE8Pi5+Dv/5mpvRc9abY1aQvXz4bkxablwazxEN7P/HurbjN04bTCikbgpBKlEEKUyxnv/StzZ9oS2FvscZp1W3HPAOOVUmnAXOBu6/ZvgVPAASAVeF1rfc5qeaXUrUqpeKVUfHp6etVegag/khZCy57QIhaCY2q3kEniPJMsxgyt+bV6TgI3T9s19i7IMxUx/UOg121mdG/mWDM1zxb2rICZ4yGkHYz/zjSzrqm4m0yytfg524zqrJ0KhzbBsBdqllw6QmBT6HcfbP3JMc29l70GK940fy+H/efcaq7Rg+H2Fab33eGtpoDPT3fCiUP2j1XUGh9Pd5o28JYplEII4YJsVYVyLDBVa90KuBT4Qinlhhm9KwRaAJHAg0qpqJIna60/1lrHaa3jmjSRBfSiFKcyTI+vaGtCFTPUJBp5tXTzkTjfTCezRaGJwKbQcbSpRnn6ZM2vt+Itk7yMeAsufdV8TV4Mnw2HY2k1u/beNaafWqMwuOEH0w7BFjx9TEPv/evM1LyayM4yRVEi+kPHy20Tn731uRMCm5uRLntOU1z1gelP2HUMXPZW2a043D1MS4h71ptYN8wy6+OWvwH5UqiprggL8iNFEjghhHA5lUng9gHFF7+0sm4r7mZgNoDWehXgA4QA44D5Wut8rfVh4E8grqZBi3ooeTGgTY8xMI2aC0+bJM7Wju41CVLb4ba7Zq9b4fRxUzCiJg5tMSMona/+p99Z3E1w/TemH9snF5m1gtVxYAN8dRX4N4EJP9m+GmHXMaZX3O8vmCmQ1bXkP5B7HC55xb69AG3Jyx8uegL2xZvm3vYQ/5mpqNpxtOljWJmpwb6NTNGaO/82hWsWPwfvnwebv5f1cXVAWJC/TKEUQggXVJkEbg0Qo5SKVEp5YYqUzClxTCowGEAp1QGTwKVbt19k3e4PnA9UqyFUfelVU19eZ5UlLzSFMJrHmsfhfcHTz2y3tR3zzdd2l9jumq17QbOuZhpldX/HhQXw47/MTfUlr569L3ow3PwbuHvD55fCtl+qdu3D2+GLK8ArECbOgQbNqxdjedw9TNKSkVj94isHN0P8p2Z0qGkn28Znb93Gmj56i56p/aqPCV/DLw+YojxXTqn6esngNqYNxIQ5ZkrttzeaEd8DG2snXlF7Dmw0I7GYEbiDx3PJza/BBypCCFFL6ss9cXVeZ4UJnNa6ALgLWABsw1Sb3KKUek4pNcp62IPAZKXUBuBrYJI20XwABCiltmASwc+11lV+x/fx8SEzM7PO/yK11mRmZuLj4+PoUJyLpdCMwEUP/mfUwNPHTKGrjXVwO+ZDUBsIibHdNZUyo3Dp26o/arjyXTiQAJe+bsrzlxTaASYvNonNrPHw57uVSxYzd8L0UeDmYZK3RmHVi68yOowyaxiXvlT1pEVrmPd/4NPITMd0dWc197bR+sjSbPkBfvoXRA4wRVM8atDzK+pCuG0ZjHwXsnbCqcO2i1PYx+4/zEhsdhbh1kqUaUdkFE4I4Vzk3r98lfoYVms9F1OcpPi2p4p9vxU4p46m1vokppVAjbRq1Yq0tDTqQ4ETHx8fWrVq5egwnMv+9ZCT9c/6tyIxQyFpgW2rO54+AbuXmWTL1rpcDQufNEVHIvtX7dz0RJP0dBgFncpZ9xUQCpN+Mc2ZFz4JmcmmGIW7Z+nHH02FaaPAUgCT5tru51gWpWDwU2a0L/5zOP/2yp+75QdIWQEj3rbd2jxHa3ORmQ687FXoPs72zb0T58N3t0Dr3jD2a/PBR025uUPPiebvs5d/za8n7CvY+sFUZjJh1u9Ts7KJDg10YFBCCHE2ufcvnw3rjtceT09PIiMjHR2GcJSkhYAyN7vFRQ/5Z7+tEo+dS6Awz7br34p4+poeYKs+MMVGGlbyH6ul0FQB9AowyVhlnueqz8wo4vLXzQjPNdPM1Mvijh8wyVveCZj4M4S2r/prqo6oQWb0dNlrEDsevAMqPifvFPz2pJmG2mNC7cdoT0Ofh//1NT+P4S/Z7ro7l8DsG6BZFxg32/bJliRvrqloZkFGEmExpj1EiqyDE0I4Gbn3L5+tqlAKUXuSre0DSk4bDIqE4GjbroPbMd9M0Qs733bXLO68m0FbzOhTZf31X0hbY9a9BYRW7hw3Nxj8JFz+Iez5Ez692DQ/L3Iy3UybPJUO139n3z5fSsHgpyE7A/76sHLnrHgbjqfBpa+ZEaC6pDaae6eshJnjTNGY8d/bphWEqBsahZvp0plJBPt74e/lLgmcEEK4GEnghHM7lWGqKpbVjy3a2k4gP6fmz2UphB0LzHOVNeWwphpHmNG9tVMrtwYsI9lUbWx3qZmyVlXdx5l2ACcPwSeDYe9qU4b/iytMtc1xs6D1eVW/bk21Pg/aXWbW9WWf0xrybEf2wJ/vQJdray+xdrRBj4G7l2nuXVNpa+Gra6FBS7jhR9tPyxSuzd0DGkdCRhJKKcKC/dkrrQSEEMKlSAInnNvO3wF97vq3IjFDoCDXNu0E9q01o0K1MX2yuF6TzfNs+bH84ywWmHMXeHjDZW9Wv2R+ZH+4ZbEZhZk6wlQPzEiEMV9CRL/qXdMWLnrCrDlc8Vb5xy143IwYDLVBcuOsAptB33utzb3/rv519q2FL680o9UT59i+FYSoG0JizPpYICzIV3rBCSGEi5EETji3pIXgF2wqF5YmvB94+NqmGmXiPJMoFK2tqy1Rg8zUz9Ufl3/cmk8gdRUMe6nmZf1DouHmRWYqamYyXDO19l9nRZp2hK7XmZ/D8f2lH7Pzd9j+Cwx4CBq0sG989nbBXRDQDH6rYnPv3OOwdpqZJvvJRWZt2oQ5df/nJaovOBqydoGlkPBgf1KzsrFY6nalNyGEqEskgRPOy2KBnYtNolFW02FPHzPCZIt1cInzIKzPucU+bM3NDc6bbJo471tb+jFZu01/sOihZhqkLfgHm2Il92/5pwm4ow161Exd/ePVc/cV5sO8R8x0rz532j82eytq7p22puLm3haLqZb6/W3welv4+R7IOQpDnzNl/huH2ydm4ZpCYkyxpqOptA7yI6/AwuETtdyLUAghhM1IAiec1/71kJ1Z9vTJItFDzafJNSkAcWSP6dFmy+bd5ek+Fjz9YfWUc/dZLDDnblDuMPLt6k+dLI27R+006a6uxhHQcxKs/+Lc39/qT8xUz+Evm2mk9UH3cRDaqezm3kdSYOnL8G43mDYSEudCtzFmiuydf5tpmP4hdg9buJhirQTCg0wvuJTMUw4MSAghRFVIAiecV3IZ7QNKirFOBUxeVP3nSpxvvtorgfNpaG68N39nCrUUt/Zz2LMchr1Q+VYDrmzAv00BjyX/+WfbyXTT9y56KLQd5rjY7K205t552bBhlknY3ulqErigKLhyCjy0wyT5reJsm+iLui042nzNSDrTzFvWwQkhhOuQBE44r6Qy2geUFBRlep7VZB3cjnkQ0s5cy1563QqFp2Hd9H+2HU2FhU9B1EDoMdF+sThSYFPofTts/hYObDTbFj9rKosOf6n+JSbRg/9p7j3nHnijHfxwq/m7MehxuG8TTPgJul5jev4JUVX+IeZDpMwkWjTyxd1NSSVKIYRwIZLACed0KtOsDyurfUBJMUPNqFV12gnkHjO90trVcvXJkkLbQ+QAiP8MCgtM4Yqf7zVfR75bvxKXvveYG8rfnze/9/Vfwvl3/NN0uL4Z+jycPgmbvoH2I2DSr3D3erjwYWjU2tHRCVenlJlGmZGEp7sbLRr5SC84IYRwIR6ODkCIUlXUPqCk6KHw9/9MIhZTxeqKyYvBkg9t7TR9srhet8Ks8aaBeM4R87ovfb3+FaHwbQx97zMjb+nbTcPyAf92dFSO07Qj3LXG/By8Ax0djaiLQmJg1x8AhAeZSpRCCCFcg4zACeeUXEH7gJIi+oKHT/WqUe6YD75B0LpX1c+tqbaXQINWsOw10+8svB/E3Wz/OJxB79shoKmZKjjkWdO3rj4LbiPJm6g9wdFwYj+cPknrID9J4IQQwoVIAiecj8ViRsXaDC67fUBJnr4Q0b/q6+AKCyDpN4i52BSQsDd3DzjvJjiQYMp6j3q38q+5rvHyg1HvQ+87TH84IUTtCSlWiTLYj6xTeZzIzXdsTEIIISqlnt4pCqd2YD1kZ1R+/VuRmKGQtdO0FKistNVm6qK9qk+WpsdEU05/+Etm1KU+a3sxXPJy/U1ihbCXokqUmcmEWVsJyCicEEK4BrlLEs4naRGmfcDgqp0XPaTY+ZWUOA/cPCtuVVCb/EPgngSIu9FxMQgh6pegKEBBRtI/CZwUMhFCCJcgCZxwPskLoWWPitsHlBTcxtyUVGUd3I75ENHP8eut6lPFSSGE43n6moqmmUmESS84IYRwKZLACdtY+R7sWFDz62RnQVp85atPlhQ9FHYvh/zcio/N3AkZOxw7fVIIIRwlOAYyk2ng40ljP0+ZQimEEC5CEjhRc8fS4Lcn4dubIGt3za5V1D6gquvfisQMhYIcSFlR8bGJ88zXtnbu/yaEEM4gJMZ8kKU1YcH+MoVSCCFchCRwouY2fA1oQMEPt5nKjtWVtNCU9K9s+4CSIvqZdgKVWQe3Yz6Edqx/PdeEEAJMIZO8k3DiAOHSSkAIIVyGJHCiZrSGhBmmhP+It2Dv31aEDMoAACAASURBVLDirepdy2KB5EUQPbj6Jf09fU0SV9E6uJwjkLJSpk8KIeqvokqU1kIm+47mkF9ocWxMQgghKiQJnKiZ1L9M2f7u46DrNdD5alj6EqStrfq1DiSY9gHVXf9WJHooZCaXP50zeTHoQtNIWwgh6qMzveBMIZNCi2b/0RzHxiSEEKJCksCJmkn4CrwCoONo8/iy1yGwOXx/C5w+WbVrJVvbB0RXsX1ASUXr55LLmUaZOBf8m0DLnjV7LiGEcFWBLcDTDzKkF5wQQrgSSeBE9eWdgi0/QMfLwcvfbPNtDFd8aEa/fnu8atdLWmjWvvmH1Cyu4DbQONJcrzSF+WaNXMwwaRgthKi/3NzM/5eZyYQXtRKQQiZCCOH05O5VVN+2n80C+Njrz94eOQAuuBvWToXtcyt3rews2Bdf/eqTJcUMhd3LSm8nkLoKTh+DdlJ9UghRzwXHQGYSTQN98PJwkxE4IYRwAZLAiepb/6UZ6Qrrc+6+i56AZl1gzl1w4lDF19r5O2hLzde/FYkuaifw57n7EueDuxdEDbLNcwkhhKsKiYGjqbhZ8mjd2FdaCQghhAuQBK4uS5wPM8ZAQZ7tr30kBfYsh+7Xg1Ln7vfwhiunmGmWc+4y1SrLk7zItA9o2cM28UX0A3fvc9fBaQ075kHkheAdYJvnEkIIVxUcbT48y9pFeLA/KTICJ4QQTk8SuLps+RsmWUn4yvbX3vA1oKDbmLKPCW0PQ5+DpN8g/tOyjytqH9Dmouq3DyjJy88kcSXXwWUkmaqZMn1SCCHOaSWQmnkKXdEHbkIIIRxKEri6Kn0HpK0GN0+TyNlyFM5iMb3foi6ERq3LP/a8ydBmMCx4wsRUmoMb4FS67da/FYkZCplJcGTPP9sSrWvy2koCJ4QQZxK4TFOJ8lReIVmnamHWhhBCCJuRBK6u2jADlDuMeheO7YWEL2137ZQ/4WiKmT5ZETc3GP2BabD9/S2lJ5JJ1mmObWrYPqCkovV0xUfhdsw3a/MatrLtcwkh6iWl1HClVKJSKlkp9Ugp+ycppdKVUgnWP7cU2zdRKZVk/TPRvpFb+TSAgGZnV6KUaZRCCOHUJIGriyyFsGGmGYHqNhZa9YJlb0DBadtcP2EGeDeA9iMqd3yD5jDyHTiwAf54+dz9ydb2AQFNbBNfkeA20Djin3Vw2Vmw929p3i2EsAmllDvwAXAJ0BEYq5TqWMqhs7TW3a1/pljPDQKeBnoDvYCnlVKN7RT62UJizkyhBNgrCZwQQji1SiVwlfiEMUwptUQptV4ptVEpdWmxfV2VUquUUluUUpuUUj62fAGiFDuXwIkD0H2cKTAy8BE4nmaqRtbU6ROw9UfodIVZZ1ZZHUdB7HhY/iakrPxne3YWpK2xXfXJ4pQy1y1qJ5D0m1msL+vfhBC20QtI1lrv0lrnATOB0ZU8dxiwUGudpbU+AiwEHPOfU3AbyEyidZD0ghNCCFdQYQJXyU8YnwBma61jgTHAf63negBfArdrrTsBA4F8m0UvSpfwpanoWDTS1OYiaN3buhauhqNwW3+C/OzKTZ8safjL0Dgcvr8Nco+ZbbuWmKTK1uvfisQMNfGmroTEeRDQFJrH1s5zCSHqm5bA3mKP06zbSrrK+uHmt0qpooXDlT0XpdStSql4pVR8enq6LeI+W3AM5BzBJ+8ozRr4SAInhBBOrjIjcJX5hFEDDazfNwT2W7+/GNiotd4AoLXO1FoX1jxsUaacI7D9V+hyDXh4mW1KwcBH4fg+WDe9Ztdf/5VZ9N66V9XP9Q6EKz8xccx92GxLWgS+jaFlz5rFVZaI/qadwPa5kLzYFC9xk5nDQgi7+RmI0Fp3xYyyTavqBbTWH2ut47TWcU2a2HiqOZgplACZZhqlTKEUQgjnVpk72cp8SvgMMF4plQbMBe62bm8LaKXUAqXUOqXUw6U9Qa1/ulifbPoWCvMgtsQIWdRAaH2+mcJY3VG4zJ1mJKtoamZ1tO4FA/4NG2eaWG3dPqAkLz+I6AvrpkHeCWgn69+EEDazDyheireVddsZ1g8ui/7TnQL0rOy5dlO8EmWwHylZpxwShhBCiMqx1VDEWGCq1roVcCnwhVLKDfAA+gHXW79eoZQ6p9RgrX+6WJ8kzICmnaFZ17O3KwWDHoUT+6s/CrdhJig3UxilJgY8BC3j4Md/wanDtbP+rbjooSap9fAxDbyFEMI21gAxSqlIpZQXZgnBnOIHKKWaF3s4Cthm/X4BcLFSqrG1eMnF1m321yjctJzJSCI8yI9Dx0+Tmy+TZYQQwllVJoGrzKeENwOzAbTWqwAfIAQzWrdMa52htc7GjM71qGnQogyHt8H+dWZ9WmkjZJEXQlgfMwqXn1u1a1sspnl31CBo0KJmcbp7wpUfg5uHeRxt4/YBJRWtr4saWLXCK0IIUQ6tdQFwFybx2oZZC75FKfWcUmqU9bB7rEW8NgD3AJOs52YBz2OSwDXAc9Zt9ufuAUFRZ0bgQCpRCiGEM6tMAlfhJ4xAKjAYQCnVAZPApWPe1LoopfysBU0uBLbaKnhRQsJXJinqem3p+4vWwlVnFG7PMtNPruTUzOoKbgNXfmSmUwaE2uaaZT5XNFxwN/S9t3afRwhR72it52qt22qt22itX7Rue0prPcf6/aNa605a625a60Fa6+3Fzv1Max1t/fO5o14DYP6fLNZKQAqZCCGE86owgavkJ4wPApOtnzB+DUzSxhHgTUwSmACs01r/WhsvpN4rzIcNs0yRDv+Qso+LHADhfWFFFUfh1n8FPg2h3WU1j7VIh5Fw0RO2u15ZlIKLX4DwC2r/uYQQwhWFREPWLsIbewOQKiNwQgjhtDwqc5DWei5m+mPxbU8V+34r0LeMc7/EtBIQtSl5sVlP1n1c+ccV9YWbNtIU9uh9W8XXzj0G23421/aUNn5CCFHnBMeAJZ/GeQcJ8PaQBE4IIZyY1FOvKxK+Ar8QiLm44mMjB0B4P+tauJyKj9/yAxTkVK/3mxBCCOdnbSWgMpMJC/IjJVMqUQohhLOSBK4uOJVpmlR3vc4UCKmMgY/AyYOwdmrFxybMgCbtoaXUnxFCiDop+OxecDICJ4QQzksSuLpg87dgya94+mRxkf1Nk+sVb5U/CpeRBHv/rlnvNyGEEM7NPxh8G5tWAsF+7D2Sg8WiHR2VEEKIUkgCVxes/xKad4Nmnat23sBH4OQhiC+n+FnCDFDuZnRPCCFE3RUcfaaVQF6BhYPHq9huRgghhF1IAufqDm6Cgxuh+/iqnxvRz4zC/fl26aNwlkLTvDt6CAQ2q3msQgghnFdwzFmtBGQapRBCOCdJ4Fxdwgxw94IuV1fv/EGPWUfhPjt3364lpmdcVaZmCiGEcE0h0XDyIBEBZupkqvSCE0IIpyQJnCsryIONs6DdJeAXVL1rhF8AkRfCirchr8Sb9fqvzJqIdpfUPFYhhBDOzVrIpHlhGu5uipQsqUQphBDOSBI4V5b0G2Rn1ry8/8BHTQ+54qNwOUdg+6/Q5Rrw8K7Z9YUQQjg/aysBj6ydtGzkS2pWJdrMCCGEsDtJ4FxZwgwIaAptBtfsOuF9IGqgWQuXZ/3EdfN3UHhaer8JIUR9ERQFKMg0lShTpRecEEI4JUngXNXJdEhaYO395lHz6w18FE6lw5pPzeOEGRDayVS3FEIIUfd5eEOjMMhIorX0ghNCCKclCZyr2jQbLAW2GyELOx+iBsGf78C+teZP7PXS+00IIeqTkBjITCY8yI8j2fkcz813dERCCCFKkATOFWltCoy07Amh7W133YGPQnYGzBwPbh7Q5VrbXVsIIYTzC46BzJ2EB/kCUolSCCGckSRwrujABji8xfbl/cN6Q5uLTOuAmGEQ0MS21xdCCOHcQqIh/xSR3scB6QUnhBDOSBI4V5TwFbh7Q+erbH/tQU+AmyfE3WT7awshhHBu1lYCYXofACkyAieEEE7HBtUvhF0VnIZN30D7y0yPNltr1RMeSQEvf9tfWwghhHOzthLwPbaLYP8IUqUXnBBCOB0ZgXM1O+abHm2xtVjeX5I3IYSonwKbg6c/ZCZLJUohhHBSksC5mvVfQWALUzFSCCGEsCWlILiNqUQZ7CdTKIUQwglJAudKThyE5EXQbQy4uTs6GiGEEHVRSAxkJBEW5Mf+oznkFVgcHZEQQohiJIFzJRtngS60Xe83IYQQoqTgGDiaSkRDdywa9h/NcXREQgghipEEzlVoDQkzoHVvU+ZZCCGEqA0hMYCmrWc6ACmyDk4IIZyKJHCuYt86SN9u+95vQgghRHHBbQBoZTGtBKSQiRBCOBdJ4FxFwlfg4QudrnB0JEIIIeqyYDPLo2F2Ct4ebqRmSisBIYRwJpLAuYK8bNj0LXQYCT4NHR2NEE7FYtHsOHQCrbWjQxGibvAOhMDmuGUmExYklSiFEMLZSALnCrbNgdPHoOdER0cihFM5fDyXiZ+v5uK3lvHFXymODkeIuiM4GjJNJUqZQimEEM5FEjhXsG46BEVBeF9HRyKE01i09RDD31nOmj1ZtG8WyIu/biP58ElHhyVE3XCmlYAvqVnZMsIthBBORBI4Z5eRDCl/Qo8JpsGqEPVcbn4hT/20mVumx9OsgQ+/3N2f6Tf1ws/LnftnJZBfWH97VhVaNAX1+PULGwqOgdyjxASeJjuvkIyTeY6OSAghhJUkcM5u/XRQ7tBNqk+Ks/29K5ONaUcdHYZdbTtwnFHvr2D6qhQm94/khzsvIDo0gNAGPrx0ZRc27TvGu4uTHB3mWbTW5OQV1upz5Bda+Hp1Kn1f/p2R7//JkVNysy1qyFrIpJ3HQUAqUQohhDORBM6ZFeab3m/tLoHApo6ORjiJ3PxCnpmzhes+/otrP1rFmj1Zjg6p1mmt+fzP3Yz+4E+OZOcz/aZePH5ZR7w93M8cM7xzc67u2YoPliSzNuWIA6P9x870k0z4bDUdnprPpM9XszI5w6ZT0SwWzU8J+xj65h88+v0mmjbwZlf6ScZ/+jfHsvNt9jyiHrL2G21lOQBAapZUohRCCGchCZwz2zEfTqWb6ZNCAIkHT3D5B38ydeUeJvQJp0VDX26auoYt+485OrRak37iNDdOXcOzP2+lf3QI8+/tz4C2TUo99umRHWnRyJcHZidw6nSBnSP9x6nTBbw8bzvD315GQupRru8dxuZ9xxg35W9Gvr+CnxL21Wiqp9aa37cf4rL3VnDvzAR8PN2ZMiGOH+/sy0c39CTp0EkmfL6aE7mSxIlqahQO7l4E56agFKRm5jg6IiGEEFaVSuCUUsOVUolKqWSl1COl7A9TSi1RSq1XSm1USl1ayv6TSqmHbBV4vbBuOgS2gDaDHR2JcDCtNdNW7mHU+yvIOHmaz288j+dGd+aLW3oT4O3BxM9Wszuj7n1CviTxMJe8s4xVOzN5bnQnpkyMIzjAu8zjA308efPa7qRmZfPCr1vtGKmhteaXjfsZ8uYf/O+PnYzq1pLfHxrIi1d0YcX/XcTLV3YhO6+Qe2cmMPC1pUxZvouTVUw0/96VyTX/W8VNU+PJzivgnTHdmXtPf4Z0bIpSioHtQvnv9T3Ysu8Ykz5f49BEVrgwN3cIisIjayfNGviQIiNwQgjhNCpM4JRS7sAHwCVAR2CsUqpjicOeAGZrrWOBMcB/S+x/E5hX83DrkWNpkLwIYq8Hdw9HRyMcKPPkaW6ZFs/Tc7bQp00w8+4dwKB2oQC0bOTLFzf3xqJh/JS/OXCsbnxKnptfyLM/b+HGz9cQ7O/NnLv6MaFPBKoShXx6RQZx24A2fL16Lwu3HrJDtEby4ROM//Rv7pqxnsZ+Xnx3Rx/euLYbTQJNwunj6c6YXmEsuv9CpkyIo2VjX174dRt9XlrMy/O2c/BYbrnX37zvGBM+W811H//F3iPZvHhFZxY9cCGju7fEze3sn8uQjk15b2wsCXuPcvO0NbW+Bk/UUcVbCUgvOCGEcBqVyQx6Acla610ASqmZwGig+MfbGmhg/b4hsL9oh1LqcmA3IB/fVUXCDNAWiB3v6EiEAy3bkc6D32zgWHY+T4/syKQLzk1iokMDmHZjL8Z+8hc3fLqa2bf1Icjfy0ER19yOQye45+v1bD94gkkXRPDIJe3x8XSv+MRiHhjalj92pPPIdxuJDRtASDmjdjV18nQB7y5O4rMVu/Hzcue50Z24vnc47m6lJ5tuboohHZsypGNTEvYe5ZPlu/h42U4+XbGLUd1aMnlAJO2bNThzfPLhk7y5MJG5mw7SyM+Txy5tz4Q+ERX+TC7p0pw3Cy3cPyuBydPjmTIxrso/R1HPhcTAjgVEtPdiSZJzrCsVQghRuQSuJbC32OM0oHeJY54BflNK3Q34A0MAlFIBwP8BQwGZPllZFgus+wKiBkLjCAcHIyorJ6+Qb9el8dfOTHqEN2ZguyZEhfhXatSopNMFhbw2P5EpK3YTExrA9Jt60aF5gzKP79KqIVMmxjHhs9VM+nw1MyafT4C3643c/rrxAA/MTiDQx4PPJ53HoPah1bqOl4cb74zpzoj3VvDIdxv5ZEJctX4P5dFaM2fDfv4zdxuHjp/m2rhW/N/w9uVO8Sype+tGfDCuB6mZ2Xz2525mrdnLd+vSGNC2Cdf3DmPxtkN8uzYNX0937hkcwy39I2ng41np64/u3pL8Qs2/v93A7V+u5aMbep5V+EWIcgVHgyWfzv5HmXXiNDl5hfh6yd8fIYRwNFvd4Y0Fpmqt31BK9QG+UEp1xiR2b2mtT5Z386SUuhW4FSAsLMxGIbmw3UvhWCoMfcbRkYhKyDh5mumrUvhi1R6OZOcTEuDNr5sO8Pwv0DrIl0HtQhnYrgl9okIqdfOTfPgE93ydwNYDx7nh/HAev6xDpUZOzo8K5r/jenDbl2uZPC2ez288zyYjLsdz81mw+SDDOjerUvJQVetTj3D/7AS6tGzI/8b3PDP1sLraNg3k/4a35/lftjJrzV7G9LLd/y2JB0/w9JzN/LUri84tG/Dh+J70CGtc7euFBfvxzKhO3Dckhi//SmHqyhRu+2ItXh5u3Ng3kn8NbFOlxLC4q3u2Ir/QwqPfb+KuGev57/U98HSX+lWiEoJjAIhxPwQ0JDUrm3bNAh0bkxBCiEolcPuA1sUet7JuK+5mYDiA1nqVUsoHCMGM1F2tlHoVaARYlFK5Wuv3i5+stf4Y+BggLi7OdjW2XdW66eDbGNqPcHQkohy70k8yZcVuvlubxukCC0M6NOXWAVGcF9GYtCM5LN2Rzh+Jh/kmPo3pq1Lw8nDj/KhgBrZtwqD2oUSG+J91Pa01M1an8vwvW/H1dOeTCXEM7Vi19hFDOjbljWu6cd+sBO7+ej0fXt8Dj2rerOcVmN5i7yxOIutUHt+uTWP6zb1qZQTn4LFcbvtiLU0bePPJhDibTQG98YIIft9+iOd+2cr5UcFElPiZV9WJ3HzeXpTE1JV7CPD24MUrOjPmvLAyp0tWVSM/L+66KIZb+kexIimDji0a0KKRb42vO7ZXGPmFFp76aQv3zlzPu2Niq/33QtQjISaBa23ZBzRkY9pRSeCEEMIJqIp6EimlPIAdwGBM4rYGGKe13lLsmHnALK31VKVUB2Ax0FIXu7hS6hngpNb69fKeLy4uTsfHx1fz5dQBpzLhjXbQazIMf8nR0YgStNasTTnCR8t2sWjbITzd3biqRytu7hdJdGhAqefk5heyZk8WSxPTWZJ4mF3pZjloeLAfA9s2YWC7UNo3D+Tpn7bw29ZD9IsO4Y1ru9G0gU+145y2cg9Pz9nCVT1a8drVXc8pclHRa5y/+SCvzN/Onsxszo8Kon9ME15bkMhlXZvz3pjYKl2vIrn5hVz30SqSD5/k+3/1tfkN4oFjOQx7axnRoQHMvq1PtRKX3PxCvvwrhQ+WJHM0J58x54Xx72HtXG6t4ZTlu3jh122M7t6CN6/tbrPEsyaUUmu11nGOjsNV2P098pVICjuMZOTuazhwLIf59w2o0f9NQgghKqe898cKR+C01gVKqbuABYA78JnWeotS6jkgXms9B3gQ+EQpdT+moMkkbctutfXJxplgyYfYGxwdiSim0KL5bctBPl6+i/WpR2nk58ndg6K5oU9EhVP9fDzd6R/ThP4xTXhyREdSM7NZuuMwSxPTmRW/l2mrUgDwdFc8fmkHbu4XWeMEaeIFERzLyefNhTto4OvBUyM6VmoN2NqULF78dRvrUo8SExrAZ5PiGNQuFKUUHm6Kl+Ztp1kDH54cUbIQbfVorXnku41sSDvGxzf0rJVP95s39OX5yztz78wEPly6k7sHx1T63PxCC9+uTeOdRUkcPJ5L/5gQHh7Wni6tGto8Tnu4pX8UpwssvLYgEU93N169qmrJvaiHQmJwz9zJe+NiGfHuCu6bmcCXt/R2iuRfCCHqq0qtgdNazwXmltj2VLHvtwJ9K7jGM9WIr37R2kyfbHUeNLXNDbKomZy8Qr5du5cpK3aTkplNWJAfz43uxNU9W+HnVb0lpGHBfkzoE8GEPhHk5hfy9+4s1qUcYWjHpnRuabvE4O6Lojmanc9nf+6msZ8X95STuOzOOMUr87Yzf8tBQgO9efnKLlzds9VZo1W3DojiwLFcPl2xm+YNfbilf1SNY/xo2S5+TNjPQxe35eJOzWp8vbKM7t6SxdsO887iJC5s14SurRqVe7zFovll0wHeWriD3RmniA1rxFvXdadPm+Bai9Fe7hwUTV6BhXcWJ+Hp7sZ/ruhs8wIvog4JjoHkhbRpEsCzozrx8Hcb+d8fO7lzULSjIxNCiHrL9crU1WVpayB9O4x6z9GR1Dn5hRaOZOeRfbqQU3kFnLJ+/edxAdl5hWe+njxttv21K5Mj2fl0b92IR4a35+JOzWz6ybOPpzsXtm3ChW2b2OyaRZRSPHFZhzMjcQ19PZl4QcRZx2SePM27i5P46u9UvDzcuH9IWyYPiCw1OVVK8eSIjhw+kcsLv24jtIEPo7q1qHZ8v28/xCvztzOia3O73Aw+P7oza/Zkcd+sBH69u3+pBWW01ixJPMxrC3aw7cBx2jcLZMqEOAZ3CK1TSc59Q2LIK7Tw4dKdeLkrnhnVqU69PmFDwW0g4UvIPc41ca1YlpTOmwt30KdNcI0K9wghhKg+SeCcybpp4BUAna50dCR1yuETuVzxwUr2Ha24ybWHm8Lf2wN/L3f8vD3oHRnMzf0jiQtv7JI3uG5uileu6sLx3HyenrOFhr6eXB7bkpy8Qj77czcfLt1JTn4hY85rzb1DYggNLH9ti7ub4s1ru5NxYjUPzd5ASIAXF7QJqXJcRZU2OzZvwGtXd7PLz7ahnyevX9ON66f8zcvztvHs6M5n7f97VyavLUgkPuUI4cF+vDOmOyO7tqiTUwyVUjw8rB15BRY+XbGbvELNyK7NCQ/xp3kDnzr5mkU1WQuZkJmMatmDF6/owvrUo9w7cz2/3tO/VivTCiGEKJ0kcM4i9zhs/h66XA3epRfDEFWnteax7zeRfvI0T43oSGN/T/y8PPD38sDP2x1/Lw/8rV/9vN3xcndzyUStPB7ubrw3NpYbP1/Dg99sIPHQCX5Yt4+Dx3MZ0qEpj1zSjujQyq8987FWyLz6fyu5bfpavrmjz1mNpytyNDuPW6bF4+PpxicT4uzaV6pvdAg394vk0xW7uahDUy5s24TN+47x2oJE/tiRTtMG3rx4RWeujWtd50vtF43QFlo0U1fu4evVqQB4e7gRHuxHRLA/ESH+1q/mcTNJ7uqf4H8SOFr2oKGvJ++O7c61H/3F4z9s5t0x3evc/5lCCOHsJIFzFlu+h/xs6DHR0ZHUKd+sTWPRtsM8OaIjN/WLdHQ4DuPj6c4nE+O4/pO/+HDpTrq1bsQ7Y7rTO6p6a7oa+nky7aZeXPnflUz8bDXf/6svLStR7r6g0MJdM9az72gOM2893yYl8qvq38PasTwpnX9/s4HzIoL4ddMBGvl58til7ZnQJ8ImvfNchVJm+uStA6LYk3GK3Zmn2JNxij2Z2ezOOMXSHenkFVjOHO/t4UZEsD/hwX5EhvgzqnsLOrVwzYIuopKCIkG5QUbSmU09w4O4f0gMr/+2gwExIVwT17qcCwghhLA1SeCcxbrpENoRWvZ0dCR1xt6sbJ77eSvnRwVxY4m1X/VRgLcHX9zSm837jtEnKrjGn5q3aOTL1JvO45oPVzHps9V8e/sFNPQrfzrVi3O3sSI5g1ev6krP8KAaPX91+Xi689Z13bn8gz9ZmniYewbHcEv/yHo9FaxFI19aNPLlguizp8MWWjQHj+ea5C7jn+RuV8Ypliam07VVI0ng6joPb2gUDplJZ22+Y2A0K5IzeHrOFnqGNyaqicwcEUIIe5EEzhkc3Az71sLwl0GmotiExaJ56JsNALx2dTeZ9mXVwMezWmvWytK+WQM+mtCTSZ+tYfL0eKbf3KvMEaxZa1L5/M893NQ3kmvPc+wn9p1aNGTevf0J8vd2uV5u9uTupmjZyJeWjXzpW0pyZ5FuMfVDSAxkJJ+1yd1N8dZ13bnkneXc/fV6vv/XBXh71J/RayGEcKS6vcjDVaz/Aty9oOt1jo6kzvh85R7+3p3FUyM70jrIz9Hh1GkXtDGNx1fvyeKB2QkUWs69qY/fk8UTP26mf0wIj13a3gFRnis6NFCStxpwd1N1fp1gZSilhiulEpVSyUqpR8o57iqllFZKxVkfRyilcpRSCdY//7Nf1FUUHA1ZO8FiOWtz84a+vHpVV7bsP85r8xMdFJwQQtQ/8u7raPm5sGEmdBgJfo6ZUlbXJB06wSvztzOkQyjX9Gzl6HDqhZHdWvDEZR2Yu+kgz/+yFV1sZGbf0Rxu/3ItLRv58v7YHmf1lhPClSml3IEPgEuAjsBYpdQ5TTyVUoHAvcDfJXbt1Fp3t/65vdYDrq7gaLNG+8T+c3Zd3KkZE/qEM2XF01sYzAAAIABJREFUbpYmHnZAcEIIUf/InZSjbf8Fco9CjwmOjqROyC+08MDsDQR4e/DSlV2lOpod3dI/ilv6RTJ15R4+XrYLgOy8AiZPi+d0voUpE+MqXCMnhIvpBSRrrXdprfOAmcDoUo57HngFyLVncDZT1EogI6nU3Y9d2oF2TQN56JsNHD7hmi9RCCFciSRwjrZumlkgHjHA0ZHUCR8sSWbTvmO8eHlnmgR6OzqceuexSzswomtzXpq3nR/Wp/Hvbzay7eBx3h0bW6VWBUK4iJbA3mKP06zbzlBK9QBaa61/LeX8SKXUeqXUH0qp/mU9iVLqVqVUvFIqPj093SaBV0lTa8/EtPhSd/t4uvPeuFhO5Bbw4OwNWEqZRi2EEMJ2JIFzpKxdsHsZ9LgB3ORXUVMb047y3u/JXBHbkku6NHd0OPWSm5vijWu7cX5UEPfP2sCvmw7wyPD2DGof6ujQhLA7pZQb8CbwYCm7DwBhWutY4AFghlKq1IaKWuuPtdZxWuu4Jk2a1F7AZfELgmZdYPcfZR7StmkgT43syPKkDD5dsduOwQkhRP0jWYMjrf/S9Nfpfr2jI3F5ufmFPDB7A00CvHlmVCdHh1OveXu48/GEOOLCGzP+/DBuHRDl6JDE/7N353FVV/kfx1+HHQRBBTdERXHfFffMtEWzxdJxa7Gm0tH2daaZaZtm+lXTXtO+2uaSlVlWppWWa4I7boDivgAqigoKnN8fXywylQtcuMB9Px8PH/fe7/d8v/eDG3zuOefzkfKyEyhaUrVR4bGTwoD2wDxjTBrQC5hpjIm31uZaazMBrLWJQCrQskKiLo3Y/rB9KRw/esYhV/VozOB29fnv7A2s3nGwAoMTEfEuSuA8JT8PVnwELS6Cmg09HU2V99TsjaTsy+apER0JD9Y+K0+rGeTP9Il9+M8VHbQPUaqzZUALY0ysMSYAGA3MPHnSWptlrY201ja11jYFlgCXW2sTjDFRhUVQMMY0A1oAmyv+S3BRs/Mg/zhsX3LGIcYYnhjegcjQQG6fvILs3LwKC09ExJsogfOUlDmQvUfFS9xgcWomby/YwtjeTejXwgPLi0TEK1lr84BbgdnAemCatTbJGPOoMebyYi4/F1htjFkJTAcmWGv3l2/EZdC4N/j4weYzL6MEiAgJ4PlRndm2/ygPf5FUQcGJiHgXNfL2lOXvQ2g9ZwZOSu1wzgnu/WQVTeuEcP/FlaO/mIh4D2vt18DXpxx76Axjzyvy/FPg03INzp0CQ6FR97PugzupZ7M63DqwBS9+n0z3prUY3aNxBQQoIuI9NAPnCYd2w6bZ0Pkq8NVyv7L491fr2J11jGdGdiYkQJ9HiIiUm9j+sGslHDtQ7NDbB8ZxTlwk93+2hie+2UC+KlOKiLiNEjhPWPUx2Hzocq2nI6nS5q7by7SEHUzo35xuTWp5OhwRkeqtWX/AQtqCYof6+frwzvXdubpnY16bn8qf31vGwaPHyz9GEREvoASuIhXkO310Et+Dpv2gTnNPR1RlZWbncv9nq2nToCZ3XlB5C7eJiFQb0fHgH1LsPriTAvx8eOzKDjw+rAOLUzO4/H8L2bDnUDkHKSJS/WnNWXnLTofU7yF5DqT+AMf2g/GFIc94OrIqy1rLAzPWcuhYHh/c2IkAP30OISJS7vwCoEkfl/bBFTWmR2Na1gtj4oeJDHtlEU+P6MQQ9eoUESk1JXDudnKWLWWuU2ly10rAQo0oaDkI4i6A5gOdxqhSKl+s3MU3a/fwt8GtadPgtH1vRUSkPMT2hzkPwqFdJWqB061JLb687RwmfpjIzR8t5+bzmnPPRa3w9VGbERGRklIC5w6H9/5+li3noNOgu1EPGPBPaHEB1O8EPpopKgtrLQtSMnjwi7V0a1JLDaJFRCpas/7O45afoNPoEl1ar2YQk8f34pGZSbwyL5V1uw/xwqguhIeomJeISEkogSutEzmw8kOnHcDuVc6x0HrQ+pLCWbYBEKzCGu5wMnF7fm4yiVsPEB0RzLMjO+mTWxGRilavAwTXdvbBlTCBAwj08+XxYR1pHx3OIzOTGPryAt4YG0/LemHlEKyISPWkBK6kcg5Bwjuw+GU4sg8adIbzH3KStnodNMvmRtZaFqVm8tycTSRsPUCD8CD+fUV7RsY3ItDP19PhiYh4Hx8fiD3X2QdnLZjSfZB2dc8mtKoXxsSPlnPFywt5dmQnBrfXvjgREVcogXPVkQxY8iosexNysqDZAOj3tlNNspTfwOT0rLUsTs3k+bnJ/JK2n/o1g/j30HaM7B6jxE1ExNOa9Yd1MyAzFSLjSn2b+Ka1+fLWc5jwYSITPlzOrQPiuPvClvhodYWIyFkpgStO1g5Y9BIkToK8HGhzGZxzF0R39XRk1dKiVGep5C9bnMTt0aHtGBkfQ5C/EjcRkUoh9uQ+uHllSuAA6ocHMfUvvXhwxlr+92MKSbuyeH50F8KDtS9ORORMlMCdSUYyLHgeVk9xXnccBX3vhCj1HDtVzol87pq6krTMozStE0LTyBrE1qlB08gaNK0TQlRYIKaYWUpnxm0TS7fsp17NQP51eTtGdVfiJiJS6dRuBuExsHkedL+pzLcL9PPlyeEd6RAdzr++XMelL/3MA5e05aK29Yr93iEi4o2UwJ1q10pY8Cysmwl+QRB/I/S5DSJiPB1ZpWSt5Z+fr+WbtXs4Jy6SjXsOM2fdXvIK7K9jagT40qRODZpGhtC0MLGLjaxBkzohbE4/wvNzN7Fk837qhgXyyGVtGd2jsRI3EZHKyhhnFm7DV07rHJ+y/39tjOHa3k1p06Amf/9sDX/5IJHezerw0GVt1S5GROQUSuBO2rcBZv/daQMQGA797oFeE6FGpKcjq9QmLUrj0+U7uOP8Ftx1oTM7mZdfwM6Dx0jLPEpaxhG2ZBxha+YR1u8+zHdJv0/uAOqGBfLwZW0Zo8RNRKRqaNbfqcS8ZzU07OK228Y3rc03d/Tj41+28eycTVzy4s+M6t6Yey5qSWRooNveR0SkKlMCd9K398PO5XDBI86sW5A+8SvO4tRM/j1rPRe0qccd57f49bifrw9N6tSgSZ0a9G8Z9btrTiZ3TlJ3lEA/H67oEq3ETUSkKok913ncPN+tCRw430PG9m7K0E7RvPB9Mu8vTuOrVbu4dWAc1/dtqmJWIuL1lMCdlL7B6eF2zl2ejqRK2HHgKLd8vJymdUJ4blQnl6uGFU3uRESkigqrD1GtnXYC59xZLm8RHuLPQ5e15epejXls1noe/2YDH/+yjX8MaaP9cSLi1VxqWmaMGWyM2WiMSTHG3H+a842NMT8aY1YYY1YbY4YUHr/QGJNojFlT+DjQ3V+AW+QehsO7y1xNy1scO57PXz5I5EReAW+OjScsSNXCRES8Tmx/2LoY8nLL9W2aR4XyzvXdmXRDDwJ8ffjLB4lc9eZS1u8+VK7vKyJSWRWbwBljfIGXgYuBtsAYY0zbU4Y9AEyz1nYBRgOvFB7PAC6z1nYArgM+cFfgbpWR7DxGqsJkcay13P/ZatbtPsQLYzrTLCrU0yGJiIgnNOsPecdgx7IKebv+LaP45o5+PDq0Hev3HOKSF3/m75+tISO7fBNIEZHKxpUZuB5AirV2s7X2ODAFGHrKGAuc3DQWDuwCsNausNbuKjyeBAQbYyrfLuTMFOexTouzjxPe+nkLX6zcxb0XtWJg63qeDkdERDylSV8wPs4+uApycn/c/HsHcH2fWD5J2M6Ap+bx+vxUjucVVFgcIiKe5EoCFw1sL/J6R+Gxoh4BrjHG7AC+Bm47zX2GA8uttX/4qMwYM94Yk2CMSUhPT3cpcLfKSHa+CdWOrfj3rkJ+Tk7n8W/WM6RDfW4+r7mnwxEREU8KjnAKmGypuATupJP742bfdS49Ymvz+DcbuOylBSzfdqDCYxERqWgu7YFzwRjgPWttI2AI8IEx5td7G2PaAU8CfzndxdbaN6y18dba+KioqNMNKV+ZyRDRBPwq3+RgZbEt8yi3fryCFnXDeOpPnbR5XEREnH1wOxOdveQe0DwqlLev787b18VzKOcEw19dxCMzk8jOzfNIPCIiFcGVBG4nULSLdaPCY0XdCEwDsNYuBoKASABjTCPgc2CstTa1rAGXi4wUiNTyyTM5kpvH+A8SAHhjbDdqBKp4qYiI4OyDK8iDrYs8Gsb5beox5+7+XNe7KZMWp3HRs/P5fv1ej8YkIlJeXEnglgEtjDGxxpgAnCIlM08Zsw04H8AY0wYngUs3xkQAs4D7rbUL3Re2GxUUOHvgtP/ttKy13Dd9FZv2HualMV1U/l9ERH4T0xN8Ayt0H9yZhAb68cjl7fh0Yh9Cg/y4cVICt3y8nH2HczwdmoiIWxWbwFlr84BbgdnAepxqk0nGmEeNMZcXDrsHGGeMWQVMBq631trC6+KAh4wxKwt/1S2Xr6S0Du10qmiphcBpvTIvla/X7OH+i1tzbksPLG8VEZHKyz8YGvfyyD64M+nauBZf3daPey5syZykvVzwzHymLtuG82OJiEjV59JaOGvt1zjFSYoee6jI83VA39Nc9x/gP2WMsXxlbHIe1ULgD37YsJenv9vI5Z0aMq5fM0+HIyIilVGz/vD9o5CdDqGV44O+AD8fbju/BUM6NuDvn63hb5+u4fMVO3l8WEdiI7WSRESqNncVMam61ELgtFLTs7lj8kra1K/Jk8M7qmiJiIicXux5zmPaTx4N43SaR4UyZVwvHh/WgaRdhxj0/E+8/GMKJ/LVckBEqi4lcBnJEFgTQivXyk5POpxzgvHvJ+Dv58MbY7sRHODr6ZBERKSyatgZAsMrxT640/HxMYzp0Zjv7+7PBW3q8tTsjVz20gJWbj/o6dBEREpF5QQzk6FOHFTTGaYjuXnk5hVwIr+A43kFHM93np/Is789zz953nIiv4BPl+8gLfMoH9zYg0a1Qjz9JYiISGXm4wtNz4HN8zwdyVnVrRnEK1d3Y866vTw4Yy1XvrKQm89rzt0XtsLXp3r+DCAi1ZMSuIwUaPqH7XvVwss/pvDU7I2luvaRy9rSp3mkmyMSEZFqqVl/2DgLDqRBraaejuasLmxbj17NavOfr9bz8o+prNl5iBdHdyYiJMDToYmIuMS7E7jjR+DQjmq5/235tgM8891GBrauS/+WUfj7+uDvawjw8yl8Xvja1+d3xwL8DGFB/tSrGeTpL0FERKqK2P7O4+b50K2pR0NxRViQP0/+qSNdGkfw0BdJXPa/Bbx2TTfaNQz3dGgiIsXy7gTuZAGTatZC4NjxfO6ZtooG4cG8MLozYUH+ng5JRESqs6hWEFrfaSfQ7TpPR+Oy0T0a06p+GBM/XM7wVxfxxLCOXNEl2tNhiYiclXcXMclIdh6rWQuBJ75Zz5aMIzw1oqOSNxERKX/GQOy5sOUnqGL91ro0rsWXt51Dx0YR3Dl1Jf/6MklVKkWkUvPuBC4zBTBQu/r0OFuQnMGkxVu5oW+s9rCJiEjFadYfjqTDvnWejqTEosIC+eimntzQN5Z3F6Zx9VtLST+c6+mwREROy7sTuIxkiIgB/2BPR+IWWcdOcN/0VTSPqsFfB7fydDgiIuJNiu6Dq4L8fX146LK2vDC6M6t3HOTSl35m+bYDng5LROQPvDuBy0yuVgVM/jUziX2Hc3l2ZGeC/NW7TUREKlBEjLOiZUvVTOBOGto5ms8m9iXAz4fRry/h46XbPB2SiMjveG8CZy1kpkJk9Ujgvl27m89W7OSWAXF0ionwdDgiIuKNYvtD2kLIz/N0JGXStmFNvrz1HHo1r8M/Pl/D/Z+uJudEvqfDEhEBvDmBO7wbjmc7TbyruPTDufzj87W0j67JbQOr/tcjIiJVVLP+cPww7Fru6UjKLCIkgHev786tA+KYsmw7o95Ywq6DxzwdloiIFydwGZucxyo+A2et5e+frSY7N4/nRnbG39d7/0hFRMTDmp7rPFbRfXCn8vUx3DuoFa9d043Ufdlc9tICvl+/l/yCqlVpU0SqF+/9ab+atBD4JHEHc9fv46+DWtGiXpinwxEREW9Wow7U71Dl98GdanD7+sy4pS/hIf7cOCmB3o9/zyMzk1iWtp8CJXMiUsG8t5F3ZgoEhEJYA09HUmo7Dhzl0S/X0TO2Njf0jfV0OCIiItDsPFj6Ohw/CgEhno7GbeLqhjLrtn7MWb+XWat38fEv23hvURr1awYxpEMDLunYgC4xEfj4GE+HKiLVnPcmcBnJUKe503y0CioosNz7ySqstTw9opO+YYiISOUQex4segm2L4HmAz0djVsFB/hyeaeGXN6pIdm5eXy/fi9frd7Nh0u28s7CLTQMd5K5Szs1pFOjcEwV/RlDRCo3703gMpOhUQ9PR1Fq7y5KY8nm/Tw5vAMxtavPJ5wiIlLFNekNPv7OPrhqlsAVFRrox9DO0QztHM2hnBPMXbeXWat3M2lxGm8t2EJ0RDCXdmzApR0b0j66ppI5EXEb70zgThyDg9uh89WejqRUUvYd5r/fbmBg67qMjI/xdDgiIiK/CagBjbpXu31wZ1MzyJ9hXRsxrGsjso6d4LukPcxas5u3F2zh9Z82Ex0RTLcmtegQHU6HRuG0a1iTsCB/T4ctIlWUdyZwmamArZItBE7kF3D3tFWEBPjyxPAO+kRPREQqn2b9Yd4TcHQ/hNT2dDQVKjzYnxHxMYyIj+Hg0eN8l7SX7zfsJSFtPzNX7QKc3RuxkTXoGB1O++hwOjaKoF3DmtQI9M4fy0SkZLzzf4rMkxUoq14LgVd+TGX1jixeubordcOCPB2OiIjIH8X2h3mPQ9oCaHu5p6PxmIiQAEZ2j2Fkd2e1TEZ2Lmt2ZrFmRxZrdmaxZPN+Zqz8LalrHhVaJKkLp3NMBH5qDyQip/DOBC4jxXmsYjNwq3cc5KUfkhnauSFDOlTd6pkiIlLNRXcD/xqwajK0vgR8fD0dUaUQGRrIgFZ1GdCq7q/H9h3OYe3OLNbsOMSanQdZmJrBZyt2AlC/ZhBX9WzM6B4x+tBWRH7lnQlcZjLUbOSs068ick7kc/e0VdQJDeDRy9t7OhwREZEz8wuAc++B7x+FGTfDFa8oiTuDumFBDGwdxMDW9X49tu9QDglbDzBl2XaenbOJl35IZnD7Bozt3YT4JrW0fULEy3lnApeRDJFVZ/Zta+YRnpq9kZR92Uy6oQfhIdr4LCJSWRhjBgMvAL7AW9baJ84wbjgwHehurU0oPPZ34EYgH7jdWju7YqKuAP3ugYJ8+PExsPlwxWvg650/dpRU3cLeckM6NGBLxhE+XLKVTxK28+WqXbSuH8bY3k0Z2rmh9syJeCnv+5dvrdPEu+MoT0dyVnn5Bcxdv4+Plm7l5+QMfH0Mtw2Mo3/LKE+HJiIihYwxvsDLwIXADmCZMWamtXbdKePCgDuApUWOtQVGA+2AhsBcY0xLa21+RcVf7vr/FXz84Pt/QUEeDHsTfPUhZEnERtbgwUvbcs9FLZm5chfvL97KPz5fw+Nfr+dP8Y24tlcTmkWFejpMEalA3pfAZe+D3EOVtoDJroPHmLJsO1OXbWPvoVwahAdx1wUtGdU9hvrhWv8uIlLJ9ABSrLWbAYwxU4ChwLpTxv0beBK4r8ixocAUa20usMUYk1J4v8XlHnVF6ne3k7R994CTxA1/x1liKSUSEuDH6B6NGdU9huXbDvD+4q18uGQr7y5Mo1+LSK7t1YSBreuq6ImIF/C+BC5jk/NYiQqY5BdYfkpO56Ml2/hhw14s0L9lFP+5ogkDWkXpP2MRkcorGthe5PUOoGfRAcaYrkCMtXaWMea+U65dcsq10ae+gTFmPDAeoHHjxm4Ku4L1uc2Zifv2fvjkOhjxHvgFejqqKskYQ7cmtenWpDYPXNKWqcu28dHSbYz/IJHoiGD+OrgVQzv/4a+RiFQj3pfAVaIWAumHc5mWsJ3Jv2xjx4FjRIYGMKF/c8b0aExM7RBPhyciImVkjPEBngWuL+09rLVvAG8AxMfHW/dE5gG9JjpJ3Nf3wtRrYeT74K+VJWURFRbIrQNbMKF/c+au38fLP6Zw3yer6dgogtjIqlOoTURKxvsSuIwU8At2qlB6gLWWJZv38+HSrXyXtIcT+Zbezepw/8WtuahtfQL8NNsmIlKF7ARiirxuVHjspDCgPTCvsHJgfWCmMeZyF66tfnqMc5K4r+6EKVfB6I/AP9jTUVV5fr4+DG5fn65NIjj/6fk89MVa3r+hh6pVilRT3pfAZSY7yyd9KjZROpFfwNdrdvPGT5tJ2nWI8GB/xvZuypgejYmrq83HIiJV1DKghTEmFif5Gg1cdfKktTYLiDz52hgzD7jXWptgjDkGfGyMeRaniEkL4JcKjN0z4v/sJHEzb4PJo2H0ZAjQqhN3qBsWxL2DWvHwzCRmrdnNpR0bejokESkH3pfAZSRDw84V9nbZuXlM+WUb7yzYwq6sHJpH1eDxYR24sks0Qf7qiSMiUpVZa/OMMbcCs3HaCLxjrU0yxjwKJFhrZ57l2iRjzDScgid5wC3VqgLl2XS91ukLN+Nm+HgkXDW1SvVmrcyu6dWETxK38+iX6+jfMoqwIFX9FKluXJqGMsYMNsZsNMakGGPuP835xsaYH40xK4wxq40xQ4qc+3vhdRuNMYPcGXyJ5eXCwa1Qp/z3v+3JyuHxb9bT+/Hv+c+s9TSqHcLb18Uz567+jOnRWMmbiEg1Ya392lrb0lrb3Fr7WOGxh06XvFlrzzvZA67w9WOF17Wy1n5TkXF7XOernLYCWxfCh3+C3MOejqha8PUxPHZFB9Kzc3luTrKnwxGRclDsDJyLPW4eAKZZa18t7GvzNdC00vW42b8FbEG5FjDZsOcQb/y0mZkrd1FgLRd3aMC4fs3oHBNRbu8pIiJSJXUc4Wxp+HQcfDgcrp4OQTU9HVWV1ykmgqt7Nua9RVsY3i2adg3DPR2SiLiRK0soXelxY4GT/+OGA7sKn1euHjfl1ELAWsvClEze+HkzP21KJ9jfl2t6NeHGc2JVTVJERORs2g939sRNvwE+uBKu+RSC9aFnWd13UWu+XbuHB2as5dMJffDxUUETkerClSWUp+txc2qDkUeAa4wxO3Bm324rwbUYY8YbYxKMMQnp6ekuhl4KJ1sIuCmBKyiwzFixk0teXMA1by9l3a5D3DeoFYv/PpBHLm+n5E1ERMQVbYc6bQV2r4KPRkBBgacjqvLCQ/z5x5A2rNh2kKkJ24u/QESqDHeVYhwDvGetbQQMAT4o7H3jEmvtG9baeGttfFRUlJtCOo2MFAhr4LblGW8t2MydU1dyPL+AJ4d3YMHfBnDLgDgiQgLccn8RERGv0foSuOx52PELrD9j7RcpgSu7RNMztjZPfLOBzOxcT4cjIm7iSpLlSp+aG4FpANbaxUAQTtnkytXj5mQLATfIOZHPGz9t4Zy4SL6781xGdVdhEhERkTLpNAYiW8L8/2oWzg2MMfznivYcyc3jiW82eDocEXETVxK4X3vcGGMCcIqSnPrR2DbgfABjTBucBC69cNxoY0xgYY8cz/W4sdZpIeCmAibTE3eQkZ3LzQOaa125iIiIO/j4wrl/hX1JsOFLT0dTLbSoF8a4c5vxSeIOlqXt93Q4IuIGxSZw1to84GSPm/U41SaTjDGPGmMuLxx2DzDOGLMKmAxcbx1JODNz64Bv8WSPm6OZkHPQLS0E8vILeP2nVDrFRNC7WR03BCciIiIAtB/mfK/WLJzb3DYwjuiIYB74fC0n8vV7KlLVubRPrbgeN9baddbavtbaTtbaztba74pcWzl63JysQOmGGbhZa3azff8xJvZvjjGafRMREXEbH1849z7YuxY2zvJ0NNVCSIAfj1zejo17D/Puwi2eDkdEyshdRUwqvwz3VKC01vLqvFSaR9Xgorb13BCYiIiI/E774c736/lPOlsgpMwubFuPC9rU4/m5yew6eMzT4YhIGXhPApeZDL6BENG4TLeZtzGdDXsOM6G/9r6JiIiUC18/ZxZuzxrYoFk4d3n4srYUWMu/vkzydCgiUgbek8BlpECd5s7SjDJ4dV4qDcKDGNr5D+3sRERExF3a/wlqN9MsnBvF1A7h9vNbMDtpLz9s2OvpcESklLwngXNDC4GEtP38krafcf2aEeDnPb91IiIiFe7XWbjVsNFzW+irm5vOaUZc3VAenpnEseOeqSsnImXjHVlI/gk4kFbmAiavzkulVog/o3vEFD9YREREyqbDSKgVC/Of0CycmwT4+fDvoe3Zvv8Yr8xL8XQ4IlIK3pHAHUiDgrwytRDYsOcQ32/Yx/V9YgkJ8HNfbCIiInJ6J2fhdq+CTd96Oppqo3fzOgzrEs1r81NJTc/2dDgiUkLekcC5oYXA6/M3ExLgy9jeTdwUlIiIiBSr4yio1RTmaRbOnf4+pA3B/r48OGMt1oXfV2stB44cZ+3OLGYn7WF64g4O55wo9zgzs3NJ3KoG5CJFecdUUhlbCGzff5SZq3ZxfZ+m1KoR4MbARERE5Kx8/aDfvTDzVkj+DloO8nRE1UJUWCD3DW7NgzPWMnPVLga3r8+erBx2HjzGroM57Dp4jF0HjxW+do4dO/H7PXOfr6jDe3/ugb9v+cwHHM45wZg3l5C8L5up43vTI7Z2ubxPRVu36xAvfp/MhPOa0zkmwtPhSBXkHQlcZjLUiILg0v0jefPnzfgYuKlfrJsDExERkWJ1Gg0/PeXMwrW4CIza+LjDVT0aMz1hO/dMW8WdU1f+YYIzMjSQ6IggWtYL47xWdWkYEUx0RBANI4JZvSOLB2as5ZGZSfznivYYN/+Z5OUXcNvkFWxOP0JUaCD3TV/FN3f0q/LbWPYeyuGG95ax51AOc9bv5faBLbhlQHP8yikJluqpav8rcFVGCkS2LN2l2blMXbadK7tE0yA82M2BiYiISLF8/eHce2HmbZA8B1pj+9iEAAAgAElEQVRe5OmIqgVfH8MzIzvx7sI0osICiY4IJjoimIYRwdQPDyLI/8ytlzo2imDHgWO8Nj+VuLqh/Lmvez/k/s+s9czbmM7jwzrQLLIGo99cwhPfbODRoe3d+j4V6djxfMa9n8ChnBNMGd+LKb9s47m5m/hx4z6eG9WZ2Mgang5RqgjvSOAyk6H1paW69N2FWzieX8Bf+jd3c1AiIiLisk5jnFm4+U9Aiws1C+cmcXXDeOzKDqW69q+DWpGans2/v1pHbGQNzmtV1y0xfbhkK+8tSuPGc2IZ06MxAH/uE8s7C7cwqF19+sZFuuV9KlJBgeWeT1ayZmcWb14bT69mdejVrA7nt6nHAzPWMuSFn3nw0raM6RHj9tlMqX6q/3zt0f1wNLNUBUwO55zg/cVbGdyuPs2jQsshOBEREXGJrz/0uwd2JkLK956ORgAfH8PzozrTqn5Nbvt4Bcl7D5f5nguSM3h4ZhIDW9flH0Pa/Hr8r4Nb0SyyBn+dvrpCiqe423NzN/H1mj384+I2XNC23q/HL+vUkNl3nku3JrX4x+druGlSAumHcz0YqVQF1T+B+7WASckTuI+XbuNwTh4TNPsmIiLieZ2ugvAY9YWrRGoE+vH2dfEE+vtyw6RlZGaXPvlITc/m5o8SiYsK5YXRnfH1+W0mKsjfl6dHdmJ31jEem7XeHaH/we6sYxw9nuf2+85YsZOXfkhhVHzMaesp1A8P4v0bevDwZW35OSWDwc//xJx1e90eh6u+Xbub4a8uYlvmUY/FIGdX/RO4zMIEroQzcDkn8nlrwRb6xtWhkyoEiYiIeJ5fgDMLt2MZpGoWrrJoGBHMm2O7sfdQLhM+TCQ3L7/4i05x4MhxbnxvGf6+Prx1XTxhQf5/GNO1cS3Gn9ucKcu28+PGfe4I/VeLUjI476l5XPzCzyTtynLbfRPS9vPX6avpGVubf5+l2IuPj+HPfWOZdds51KsZxLj3E/jb9NVk57o/oTybtTuzuHPqShK3HuCat5ey73BOhb5/VXb0eB4rtx9kyi/beOOn1HJ9r+qfwGUkg48/RJSsf9tny3eSfjiXm88rXesBERERKQedr3Zm4eY9qVm4SqRL41o8PaITy9IO8M/PXestd9LxvAImfJjIrqwc3hjbjZjaIWcce9eFLWhZL5T7P11N1lH3LKVcsjmTGyclEFM7hNwTBQx7ZRGfJGwv83237z/KXz5IpGFEEK9d040Av+J/7G5RL4wZt/Tl5vOaMy1xO0Ne+LnC+uClH85l3PsJ1A4J4O3r4snIzmXs27+QdazqLVk9VUGBJXHrAdbsyGJ31rFSfchQ9F7bMo8yO2kPL8xNZuKHiQx4eh7tHp7NFS8v5P7P1vDGT1tK9G+gpKp/EZPMFKgd6/SRcVF+geX1n1Lp2CicPs3rlGNwIiIiUiJ+AXDOXTDrbtj8IzQf6OmIpNDlnRqSui+bF75PJq5uqEtbUKy1PDhjLUu37Of5UZ3p1uTsvd4C/Xx5ZkRnrnhlIf/6MolnR3UuU8zL0vZzw3vLaFQrmMnjewFw++QV3Dd9Ncu3HeDhy9qdtRrnmRzOOcGNk5ZxIr+At6/vXqI+wgF+Pvx1cGsGtK7LXVNXMuK1xdx8Xhx3XNCi3Hru5eblM/HDRA4cPc6nE/vQrmE4r1/bjRveW8ZNk5bx/g09CQ4o+e9DZfBzcjpPfLOBpF2Hfne8ZpAfkWGBRNYIJDIsgMjQQOoUeR4ZGkCdGoFkZOeyfs9hNuw+xIY9h9m45/CvM6PGQJPaIbRpUJMrOkfTukEYberXpFGt4HItRlP9E7iM5BK3EPh6zW62Zh7ltWu6qhKQiIhIZdPlGvj5GWcWrtkAVaSsRO44vwUp6dk8+e0GYiNrMKhd/bOOf+vnLUxN2M5tA+O4oku0S+/RoVE4twyI48Xvkxncvj4XFfMeZ5K49QDXv/ML9cOD+GhcTyJDAwH44MaePDtnIy//mMqanVm8evXZZwVPdbKHXWr6Ed6/oUepC+F1b1qbb+7ox6NfruN/P6Ywf1M6L43pQlM3txuw1vLQjCQSth7gf1d1oV3DcAD6tYji+VFduHXycm75eDmvX9ut3BLI8rB2ZxZPfLOBBSkZNKoVzH+HdyQ8xJ/M7ONkZOeSkZ1LZvZx0rNz2bjnMAuzM8862xgW5EebBjUZ3jWa1g1q0rp+GC3rhVEjsOLTqeqdwOXnwf7N0Opily+x1vLqvFSaRdXgoral+w9BREREypFfIPS7G2bdA5vnQfMBno5ICvn4GJ4Z0Ykd+49y19SVfDKh968JwanmrtvL/32zniEd6nPXBSX7sP3WAXHMXbeXf3y+hvimtaldghkugJXbD3LdO79Qt2YQk8f1om5Y0K/nfH0M9w1qTZeYWtw1bSWXvPgzz4/uzMDW9c5yx9889rXTw+6xK9uXueVBWJA/T43oxPlt6nH/Z6sZ9uoi3rm+O53dWJ9h0qI0piZs59YBcVzaseHvzl3SsQEHj7Xnn5+v5b5PVvHsyM74+FTuD0y2ZR7l6e82MnPVLmqF+PPQpW25uldjAv2Kn0E8nlfA/iNOgpdemODVCvGndYOaNAwPqjQTO1UnjS6Ng1uh4ESJCpjM35TOut2HmHBu80r/F1RERMRrdbkWakbDfO2Fq2yC/H15c2w8NYP8GTcp4bSFMNbvPsQdU1bQITqcZ0aUPCkI8PPh2VGdyDp2gge/WFuia9fsyOLat5dSu0YAH4/rSb2aQacdd0Hbesy6rR+NaoVww3sJPD17I/kFZ/+79uGSrby7MI0b+sZydc+S1V84m8Ht6/P5zX0JDfRjzBtL+GGDe6pULkzJ4N+z1nNh23rcfeHpk+irezbhvkGtmLFyF49+ta5c93aVRWZ2Lo/MTOL8Z+fx3bo93Dogjvl/HcAN58S6lLyB8/eqfngQ7aPDGdCqLn/q1ojz29QjOqJ8l0SWVPVO4ErRQuDVeanUrxnk8jS+iIiIeIBfoLMXbtti2PKTp6ORU9StGcRb18Vz4OgJxr2fSM6J34pG7Ducw02TEggL8ufNsfGl3lvVun5N7rygJbNW7+ar1btcumbtziyueXsp4cH+TB7fiwbhwWcd37hOCJ/d3IeR8Y34348pjH1n6RlbJZzsYTegVRT/vKTNaceURWxkDT6d2Ie4uqGMez+Rqcu2lel+2zKPcsvHy2keVYPnRp09ib75vObceE4s7y1K46UfUsr0vu529HgeL32fTP+n5vHBkq38qVsM8+8bwL2DWlHzNNVMq4PqncCVsIVA4tYDLN2yn5v6xbpUKUhEREQ8qMu1ENYA5qkvXGXUPjqc50Z1ZtX2g9w3fTXWWnJO5DP+/UT2HznOW9fFn3H2y1V/ObcZnWIieHDG2mIbYK/bdYhr3l5KaKAfk8f1Ijri7MnbSUH+vvz3T514cngHlqUd4NKXFrB824HfjUnZl83Ewh52L47p8rsedu4UFRbIlPG96BsXyd8+XcMLc5NLNSOWnZvHTe8vA+DNsfGEFrOPyxjDP4e0YXjXRjw7ZxMfLE4rRfTudSK/gA+XbKX/U/N4Zs4m+sbVYfad5/L4sA5l/ntV2VXvLCUjGYJrQ8jZKxqd9Oq8VCJC/BnTo3E5ByYiIiJl5h9UOAu3CLYu8nQ0chqD29fnvkGt+HLVLl74Ppm/Tl/Nyu0HeW5UJ9pHn35vXEn4+frwzIiOHDmezz8+X3PGZGbjnsNc8/ZSgv19mTyuV4mKkpw0qntjPpvYBz9fw6jXF/PeQqdU/IEjx7lx0jICztLDzp1ONk8f1jWa5+Zu4h+fryUvv8Dl6wsKLHdNXUlq+hFevqorTeq4VhTFx8fw5PAOXNCmHg/NTGLmKtdmPd2toMDyzZrdDHruJx6YsZamdUL4dGIfXr82nri6pSsYU9VU7yImmSkuz75t2nuYuev3csf5LTxSTUZERERKoetYZwZuySvQtK+no5HTuPm85qSmZ/P8XGdl1H2DWjG4fQO33T+ubhj3XdSKx75ez+crdjKsa6PfnU/Zd5ir31qCv6/h43G9aFyn5MnbSe2jw/nq1n7c88lKHvlyHYnbDrLvUA67s3KYPK5nqRLD0vD39eGZEZ1oEB7Eyz+mkn44h5fGdHVpOepzczcxZ91eHrmsbYmLrPj5+vC/q7ow9p1fuHvqSmoG+XFeq7ql/TJcknMin9U7skjYup/EtAMkbjvAwaMnaFE3lLfGxnN+m7qVan9aRajemUpGMrS8yKWh7y7cQrC/L9f1aVq+MYmIiIj7+AdDt+thwXNwIA1qNfVwQHIqYwyPD+vAoWN5NKoVzM3nFd8frqRuOCeW2Ul7eHhmEn2aR1I/3FlCl5qezZg3l2KMk7zFuqEEf3iIP29cG8+r81N55ruNFFhc6mHnbsY41TLr1Qzi4ZlJXPXWEt6+rvtZK3LOWr2bl35IYVR8TKl/5g3y9+Wt6+IZ/foSJnyYyEc39XTr156ZnUvC1gMkbj1AQtp+1u48xPHCGcZmUTUY1LY+fVtEMqR9ffyqUFsDdzKVrZJMfHy8TUhIKPuNjh2EJ5vABf+Cc+4sdviQF36mbs1A3vtzj7K/t4iIuMQYk2itjfd0HFWF275HVjdZO+GFjtBzAgx6zNPRiIdsyTjCxS/8RM/YOrz35+6kZR5l1OuLKbCWKeN7EVc3zO3vuSxtP3sP5fyh/H5F+3btHm6fsoJGtYKZ9Ocep50JTNqVxZ9eXUzbhjX5eFxPlysznkn64VxGvLaI/UeOM21Cb1rXr1nie1hr2ZxxhIS0/SSkOUnb5owjAAT4+tChUTjxTWsR36Q23ZrUKnG7iKrsbN8fq+8MXGZhhRwXllBaa0nLPEKP2Ir95ERERETcIDwa2l4By9+H8+6HQPf/oC6VX2xkDe4f3JpHvlzHc3M28UniDvIKLJPHlU/yBk6z7cpgcPv6fHRTT26alMCwVxfx7vXdf7fHMCM7l/HvJxIR4s9r13Qrc/IGTkGVD27syZ9eW8TYt3/h04l9/pA4Hj2ex66DOezOOsbugznsPHiM3VnH2HUwh12Fx44VViiNCPEnvkktRsTH0L1pLdpHhxPkX/Y4q6Pqm8CVoIVAenYuR4/nu2VaXURERDyg10RYOx1WToae4z0djXjI2N5NmZ20lxd/SCEixJ/J43rRqr53JPTdm9Zm+oTeXPfOL4x+YwmvXdONc1pEcjyvgIkfJpJ5JJfpE/oQFRbotveMqR3CBzf2ZMRri7nm7aX0bxnFroO/JWgHj5743XhjICo0kIYRwbSuH8bAVnWJqxtKfNNaNIsMVQ9mF1XfBC4zGYyvS2vh0zKOAtCkDJtaRURExIMaxUOj7rD0Veh+E/h4594Yb+fjY3hqREf+7+v13HxeHG0alHxZX1XWol4Yn93cl+vf/YXr3/2Fp0d0YumW/SxLO8BLY7q4pfLnqVrWC+PdP3dn3KQEvli5iwbhQURHBNO1SQQNwoOJjgimQXgQDSOCqVczSK263KD6JnAZyU7y5lf8Wtm0TGetrWbgREREqrBeE2H6DZD8HbQa7OloxEMa1Qrhlau7eToMj6kfHsS0Cb0Z/34Cd05dCTiVQC/rVH779Lo2rkXCAxd4XTVIT3EpBTbGDDbGbDTGpBhj7j/N+eeMMSsLf20yxhwscu6/xpgkY8x6Y8yLpqL+ZEvQQiAt4wh+Psblho4iIiJSCbW5HMIaOrNwIl6sZpA/k27owZgeMYzo1oh7L2pV7u+p5K3iFDsDZ4zxBV4GLgR2AMuMMTOttetOjrHW3lVk/G1Al8LnfYC+QMfC0wuA/sA8N8V/egX5kJkKcee7NHxr5lFiaod4bSlSERGRasHXH3qMg+//BXvXQb22no5IxGMC/Xx5fFjH4gdKleNKxtIDSLHWbrbWHgemAEPPMn4MMLnwuQWCgAAgEPAH9pY+XBdlbYf8XJcKmIBTdrap9r+JiIhUfd2uB79gzcKJSLXlSgIXDWwv8npH4bE/MMY0AWKBHwCstYuBH4Hdhb9mW2vXlyVgl5ysQFmCFgJN6mj/m4iISJUXUhs6jYLV0+BIpqejERFxO3evGRwNTLfW5gMYY+KANkAjnKRvoDGm36kXGWPGG2MSjDEJ6enpZY9CLQRERES8V8+JkJcDie96OhIREbdzJYHbCcQUed2o8NjpjOa35ZMAVwJLrLXZ1tps4Bug96kXWWvfsNbGW2vjo6KiXIv8bDKTISgcakQWO1QtBERERKqZuq2h2QBY9hbknyh+vIhIFeJKArcMaGGMiTXGBOAkaTNPHWSMaQ3UAhYXObwN6G+M8TPG+OMUMKmYJZR1WjjdAouhFgIiIiLVUK+b4fBuWPeFpyMREXGrYhM4a20ecCswGyf5mmatTTLGPGqMubzI0NHAFGutLXJsOpAKrAFWAaustV+6LfozUQsBERER7xZ3AdSJgyWveDoSERG3cqmRt7X2a+DrU449dMrrR05zXT7wlzLEV3K52c4nbi4mcGohICIiUg35+EDPCfD1vbB9GcR093REIiJuUf2ylsBQ+PtO6H6TS8PVQkBERErLGDPYGLPRGJNijLn/NOcnGGPWGGNWGmMWGGPaFh5vaow5Vnh8pTHmtYqP3gt0GgOB4ZqFE5FqpfolcOAkcUHhxQ5TCwERESktY4wv8DJwMdAWGHMyQSviY2ttB2ttZ+C/wLNFzqVaazsX/ppQMVF7mcBQ6Hqtsw8ua0fZ77fuC1j+QdnvIyJSBtUzgXORWgiIiEgZ9ABSrLWbrbXHgSnA0KIDrLWHirysARTdJy4Vocd4wDoVKUvLWljwHEwbC1/dBUf3uy08EZGS8uoETi0ERESkDKKB7UVe7yg89jvGmFuMMak4M3C3FzkVa4xZYYyZf7oeqUWud2+vVG9Tqwm0vhQS3oXjR0t+fUEBzP4HzH0EmvaDghOQ9LnbwxQRcZV3J3BqISAiIuXMWvuytbY58DfggcLDu4HG1touwN3Ax8aYmme43r29Ur1Rr4mQcxBWTy3ZdXnH4fPxzh66nhNh7EyIagOrp5VPnCIiLvDuBE4tBEREpPR2AjFFXjcqPHYmU4ArAKy1udbazMLniTgtd1qWU5zSuDc06ARLXnWWQ7oiNxsmj4I1n8D5D8Pgx53Klh1HwvYlsH9L+cYsInIGXp3AqYWAiIiUwTKghTEm1hgTgNMPdWbRAcaYoj1tLgGSC49HFRZBwRjTDGgBbK6QqL2RMc4MWsZGSP2h+PFHMmDSZbB5Pgx9Gfrd7dwDoMMI53HN9PKLV0TkLLw6c1ELARERKS1rbR5wKzAbWA9Ms9YmGWMeNcZcXjjsVmNMkjFmJc5SyesKj58LrC48Ph2YYK1VZYzy1H4Y1KgLS4vp2HBgK7wzCPatg9EfQZdrfn8+IgaanOMsx3R1Nk9ExI1cauRdHZ1sIdAjtranQxERkSrKWvs18PUpxx4q8vyOM1z3KfBp+UYnv+MX6PSInfd/kJEMkS3+OGbPWvhwOOQdg7FfQONep79Xx5Hw5e2wawVEdy3fuEVETuG1M3BqISAiIuJl4v8MvgGnn4XbugjeHQLGB26YfebkDaDtUPANVDETEfEIr03g1EJARETEy4TWdfawrfwYjh347fj6r+D9K5zzN34Hdduc/T7BEdBqMKydDvl55RuziMgpvDeBUwsBERER79NzApw4Css/cF4nToJp10L9Ds7MW0TM2a8/qeMoOJIOm38sv1hFRE7Da/fAqYWAiIiIF2rQ0SlC8subkJcDPz4GcRfCyEkQUIIPdeMuhOBaTjGTFheWX7wiIqfw2hk4tRAQERHxUr0mQtY2J3nrOArGTC5Z8gbgFwDtrnSWX+YeLp84RUROw2uzF7UQEBER8VKtLnZm0PrdA1e8Br7+pbtPx1FOxcoNs9wbn4jIWXjlEkq1EBAREfFiPr5wjRsaccf0hIgmzjLKTqPLfj8RERd45QycWgiIiIhImRnjzMJtngeH93g6GhHxEl6ZwKmFgIiIiLhFx5FgC2CNG2b0RERc4J0JnFoIiIiIiDtEtoCGXZ1llCIiFcA7Ezi1EBARERF36TgK9qyGfevdcz9rYd0XkJ3unvuJSLXilQmcWgiIiIiI27QfBsYXVk9zz/1WT4VpY2HBc+65n4hUK16ZwWzJOKL9byIiIuIeoXWh+UBY8wkUFJTtXgfSYNa9zvNN35Y5NBGpfrwugbPWsjXzCE3raP+biIiIuEmn0ZC1HbYtLv098vPgs/FOdcs+t8P+VMhIcV+MIlIteF0Cl56dyxG1EBARERF3ajUEAkJh9ZTS32PBs7B9KVzyLHS/yTmWPNs98YlIteF1CZxaCIiIiIjbBYRAm8sg6Qs4kVPy67cvg3lPQIeR0HEE1GoCddtqGaWI/IH3JXBqISAiIiLloeNIyM0q+axZ7mH4bBzUjIZLnv7teIuLYOsiyMlyb5wiUqV5XwKnFgIiIiJSHmL7Q2i9klej/OZ+OLgVhr0OQeG/HW85GAryIPUH98YpIlWa1yVwaiEgIiIi5cLHFzqMgE2z4eh+165JmgErP4Rz7oYmfX5/rlF3CK7l3E9EpJDXZTFqISAiIiLlpuMoKDgB62YUPzZrJ3x5BzTsCufd/8fzvn4QdyEkfwcF+e6PVUSqJK9K4NRCQERERMpV/Q4Q1ab4ZZQFBTBjAuSfgOFvga//6ce1HARHM2FnovtjFZEqyaUEzhgz2Biz0RiTYoz5w0dExpjnjDErC39tMsYcLHKusTHmO2PMemPMOmNMU/eFXzJqISAiIiLlyhinmMm2xU5T7jNZ/D/Y8hNc/ATUaX7mcXHng/HVMkoR+VWxCZwxxhd4GbgYaAuMMca0LTrGWnuXtbaztbYz8BLwWZHT7wNPWWvbAD2Afe4KvqTUQkBERETKXYcRzuPqT05/fvcq+P5RaH0pdLn27PcKrgWNeymBE5FfuTID1wNIsdZuttYeB6YAQ88yfgwwGaAw0fOz1s4BsNZmW2uPljHmUlMLARERESl3ETHQ5BxYPRWs/f2540fh03FQIxIuf8mZsStOy0Gwdw1k7SifeEWkSnElgYsGthd5vaPw2B8YY5oAscDJerctgYPGmM+MMSuMMU8Vzuidet14Y0yCMSYhPT29ZF9BCaiFgIiIiFSIjiMhMxl2rfj98TkPQsZGuOIVCKnt2r1aDnYeNQsnIri/iMloYLq19mSpJD+gH3Av0B1oBlx/6kXW2jestfHW2vioqCg3h/QbtRAQERGRCtF2KPgG/r6YycZvYdlb0PtWaD7Q9XtFtoRaTZXAiQjgWgK3E4gp8rpR4bHTGU3h8slCO4CVhcsv84AZQNfSBOoOaiEgIiIiFSI4AloNhrXTIT8PsvfBF7dAvfZw/kMlu5cxzizclvnOEkwR8WquJHDLgBbGmFhjTABOkjbz1EHGmNZALWDxKddGGGNOTqsNBNaVLeTSUQsBERERqVAdR8GRdNj8o5O8Hc92Wgb4BZb8Xi0HQV6OU7nSnax1kksRqTKKTeAKZ85uBWYD64Fp1tokY8yjxpjLiwwdDUyx9rfduoVLKe8FvjfGrAEM8KY7vwBXqYWAiIiIVKi4C50qkjNvc5pxX/go1G1Tuns16Qv+NSDZzcsof3oKnm0DOxLce18RKTd+rgyy1n4NfH3KsYdOef3IGa6dA3QsZXxuoxYCIiIiUqH8AqDdlZDwDsRdAD3Gl+FegdB8gLMPzlrXqlcWJ+eQ04+uIM+ZIRw/H/yDyn5fESlXXlPNQy0EREREpML1nODsXxv6StmTrpaD4dBO2LvWPbEtewtysmDgg5C+AeY/6Z77iki58p4ETi0EREREpKJFtYKrpkJYvbLfq8VFzuOmb8t+r+NHYfHLzszgufdC52tg4Qt/bHsgIpWO1yRwaiEgIiIiVVpYPWjY1T3tBJZPgqMZ0O9e5/WgxyC0Lsy4GfKOl/3+IlJuvCabUQsBERERqfJaDnYKjhzJKP098nJh4YtOYZQmvZ1jwRFw6fOwb51T2KSySN8EKz/2dBQilYpXJHBqISAiIiLVQstBgIXkOaW/x6rJcHgX9Lvn98dbDXZaHyx4FnavLlOYbpGbDR+PhBkTYfM8T0cjUml4RQKnFgIiIiJSLTToBKH1S78PLj8PFjwHDbtA84F/PD/4CQiu7SylzD9RtljLas6DcCANakTB7H9CQb5n4xGpJLwigVMLAREREakWjIGWF0HqD6Xbq5b0mZMUnXvf6atihtSGS5+DvWvg52fLHG6pJc9x2i/0uQ0u/q9TeXPFB56LR6QS8Y4ETi0EREREpLpoORhyD8G2xSW7rqAAfn4G6raFlhefeVybS6H9cGcv3N6kssVaGkf3wxe3QlQbGPBPp5deTC/44T9O7zoRL+cdCZxaCIiIiEh1EdsffANLXo1y4yyn31u/e8CnmB8BL34KgsKd/WcVvZTy63udCpnDXncaixsDg/8PjqQ7+/NEvJxXJHBqISAiIiLVRmAoxPYr2T44a+Gnp6F2M2dGqzg16sAlz8DuVU5/uIqyZjqs/RTOu9/Z73dSdDenwMriV+DA1oqLR6QS8oqMRi0EREREpFppORj2p0JGimvjU76H3SvhnLvAx9e1a9pdAW2HwvwnYd/60sfqqkO7YdY9EB0Pfe/64/nzHwbjA3MfKf9YRCqxap/AqYWAiIiIVDstLnIek11cRvnz01CzEXQcXbL3GfIMBITCF7c4FSzLi7Uw81anR92Vr4Ov3x/HhEdD39udQizblpZfLCKVXLVP4E62EGiqGTgRERGpLmo1cYp8uLKMMm2hU/Ck7+3gF1Cy9wmNgiFPwc5EWPy/0sXqisR3IWUuXPRviIw787i+d0BYA5j9d6coi4gXqvYJ3NZMp4VAU1WgFBERkeqk5SDYughyss4+7uennV5qXceW7n3aD4fWl8KP/wfpm0p3j7PJTHX6vDUbAPE3nn1sQA04/yEnoQUqhmsAABRmSURBVFw73f2xVGZZO8p3FrS6yc9z9kumLYCVHzv7J7PTPR2VW5xmfrp62ZKhFgIiIiJSDbUcDAufd3rCnakwyc5E5/wF/wL/UlbjNgYueRZe7uEspbzhW9f30RWnIN+pdOnjD0NfLr46JjjLQJe+5uyFa30pBHjBKqvUH+HD4dDlGrj8RU9HUznk5TpJ7cGtcHA7ZG2Hg9t+e35oJ9hTZml/fsb5/Wt9iWdidpNqPwOnFgIiIlKejDGDjTEbjTEpxpj7T3N+gjFmjTFmpTFmgTGmbZFzfy+8bqMxZlDFRi5VXqPuEFzr7O0Efn4WgiKgezEzW8UJq+c01N7xCyx5tWz3KmrRi7B9KVzytLPHzRU+PjDocecH9PJc1llZZKTAJ9c5SfPy92HXCk9H5FmJk+CZ1vCfuvBSV/jgSvjydic527rIGdOkr9Mu4/KX4NoZcNty+MtPULMhTLkKZtxSpXsKVvsZOLUQEBGR8mKM8QVeBi4EdgDLjDEzrbXrigz72Fr7WuH4y4FngcGFidxooB3QEJhrjGlprc2v0C9Cqi5fP4i7EJK/c2ayTp0V27sONnwF/e+HwLCyv1/HkU4BkR/+Da0uhjrNy3a/PWvhh8ecSpcdRpTs2qZ9oc1lsOA56HIt1GxQtlgqq2MHYPIo8PGDcT84yco3f4MbZjszo95m1VQnWWvcG7r9GSIaQ0QMhMc4ydn/t3fn0VFV2R7HvztBQFFsUBtBkUGhnQWNKEojCoITguIAoqK0ggLqe90qDu3Qjm85gbYjk+AE4oyK4IggD5GAKKKoDCogIuKINkPgvD925SWEBCqkKpW69/dZy5XUrap7Tw7XOtk55+ydu83m33/BW55V9b17YNFk6Pqgl+TIMpGPalRCQERE0qgVMD+EsDCEsBYYA3Qp/oIQQvE/89YCQuL7LsCYEMKaEMIiYH7ifCLJa94J/ljpSyVLmpLIIHlY39RcywxOGgzVavhSyookESlYAy/09RnEEwdtXTBy7E2wocADyihavw6eOc/3cZ35JOx6gJdSWDwd5jyT6dZVvnmv+nLbJm19Vq3dQGjRAxq38aQ+WwrewJP4tL8Oer/urx/V2fdfrlud/vanUKQDOJUQEBGRNNsNWFzs8ZLEsY2YWX8zWwDcAVxanveKbNZe7cFyN11GuXKBz5bl9Ybt6qbuerXr+/LFb6b5nrgpd/s+pPKadDss/8SXuNXaaevaUrepB6ezn4JvZ2/dOaqyCVfDwknQeTA0au3HWvSEBi3hjethzaqMNq9SLZzkwWyDltB9NGxTs2Lna3goXDTFlxZPux+GHJVV91CkAziVEBARkaoghPBACGFPYCDwz/K818z6mFm+meWvWBGNDGqSQtvWgT0O3zSAe2+QJwZpPSD112xxFnR92DNbvnUTDNofRp0Ms0cnF1R88z5MvdeXPv7luIq1pe0VHqBOvNZryVVEwVrfF1WwpuLnqqgZw2DGUDjiEk9cUignB46/E35b5sFzHCyeAaPPgp2aQc9noMb2qTlv9Vpw4t1w9nOeyXVYe3j3zqzI9BnpPXAqISAiImm2FGhY7PHuiWNlGQMUZoBI6r0hhCHAEIC8vLwM/1YpVVLzTj4j88sS2HF3//rRGDjkPE8+kmpmvnStRQ/4cSF8PBY+Gg0vXgSv/gP2PRkO6gGN/7ppVsk1q+CFi7ydnW6reFtq7ghHX+PXnfeK74srr7W/e2KWqffBmmIlGarV9OWipX6tCbnV/Wu9fT1hxtZm+Sxp4SQYf6VnGe3wr02fb3ioZ+Kcdj8cfI7PREbVd5/Ak91g+z/DOS+kdja50F4doN80ePVyeOcWr614yiObr0eYYZGegVMJARERSbMZQDMza2Jm1fGkJOOKv8DMmhV7eCLwZeL7cUB3M6thZk2AZsAHldBmiZrmiVmswlm4qfcBwYtep1vdptDuKrh0Npw/AQ7o5nuVHjsZBh8Ab/4Lfviy6PVvXAc/fQVdH4KatVPThoPP86Lmr1/ns2fJKlgL04fAvS18H12j1tDxFjjmOmh7JbTq44FS8+NgjyOg3n6eLKPGDp6efvXP8NMimHwnDGkHyz6u+M+ycgGM7QU7N4dTh5ZdrqHDjR5ATizXhH52WbnAk7ZsUwvOfSk9f4wotG0dOG04nDYCVs6Hh9vAB0MzPxNbhkjPwKmEgIiIpFMIocDMBgATgVxgRAhhrpndBOSHEMYBA8ysA7AO+AnolXjvXDMbC3wKFAD9lYFStsrOzaFOYw/g9ukMs0bBQd09O19lMfMAqFFrLzcw71WfBZw62DP+7XaIp3bPH+HLOhu3Sd21c6tBp1u8TtoHQ3zZ4eZsWO+zhpNu87phjdrAmU/AHodt3fXnvwUv9oOhx3iR8dYDkqtnV9J/foanzvSg7awxmw9wa9eHtpd7Lbz5b/leyCj5ZSk81hXCejj3FU9SUhn27+bB+rhLYPzlfh93eSD5EheVxEIViyzz8vJCfn5+Ss7V/8lZfLrsV965vF1KziciIqllZjNDCHmZbke2SOUYKRHz2kCYORIO7uV7pwbkVzzNfyr89p1nTJw9Gr6fC7vsDX3erXgSitI8cRos/gAunQW1dt70+RD8F/K3b4EVn0H9gzzg2rN9xVPy/77S09vPe8WXjp7ysC8TTdb6AnjqdFg0BXqNg0ZHbPk9BWvgwcN9r+PFU5PLwpgNfv8BHj3e751eL0ODFpXfhhBg5qO+tzKnGnS61fdsVmLphs2Nj5FfQqkSAiIiIhJ5zTpCwWr44BHY75SqEbwB7LCrz4j1+1/o9z70eiU9wRv48se1qzzDZUkL34VhHeDpnrBhHZw+Ei6c5PufUvFLea2dfBbv5Pth6Sx46Aj45Lnk3z/xGljwNpw0KLngDXw/Xqfb4YfPfblfFKz+xZdN/rwYzno6M8Eb+D2R19sD4/oH+YzcE6d6u6qAyAZwKiEgIiIisdG4je8VAk+oURX9eR/Yfpc0nn9vyDsf8h+F7+f5saUz4bEuvifvt2VetqDfdA9yt2aZ4+aYeVKRi6Z4xsRne8PzfT2z5ebkj/DAu/UAf395NO/kQeik22FVlmepXfuHLyH9/jMPhpMNZNOpblM4dxyccBd8M91nPGcMr1gNxBSIbACnEgIiIiISG9VqePCS19uTbcRVu2u8ePmrf4enz/Z9ad/N8YyXl8yCg8/1PXPptNOe0HsCHHUVzBkLDx8JX08r/bWLJsP4K6BZJy9MXl5mPgu37g94eyveX1UUrIWx53iR8m5DoVmHTLeoSE4OtLrQM1Xunuf31mMnw4+LMtekjF05zVRCQERERGKl062+BC/Oau0ER10BX0+FBZM8oLvsI2jdP31LN0uTuw0cfTX0ngiWAyNPgLduhvXril6zcgE8fY7P1nUbVnbGyS3ZpTkcdhHMehy+/TA17a9MG9bD8xfC/Deh870+O1oV1WkE57zobfx2ti+Tnf5IRmbjIhvAqYSAiIiISAwddjF0G+6BW7uBnvY/Uxq2goveg4POgil3wfCO8MN8zzg5ursHdz1GV7ykwlFXeuKW1wZWPPX9hg2eSKQyhAAvXwafvggdb/UZ0qrMzOsr9n/fl3i+diWMPNGD8UqUVABnZseZ2edmNt/Mrirl+UFmNjvx3xdm9nOJ52ub2RIzuz9VDd8SlRAQERERiaHcanDAaT4bVxXU2AG6PgCnj/LC54/8FUZ19iV4Zz4BdZtU/Bo1d4T2N/gSxDnPbP15Fs+A4R3grmbw3uD01kFbv85T9X/4OBw1EI4YkL5rpdqOu0PPZ6HLg7B8Ljx0JEx7wGcTK8EWAzgzywUeAI4H9gV6mNm+xV8TQvjvEEKLEEIL4N/A8yVOczMwOTVNTs7XK/+gYd3tqJYb2UlGEREREckW+3VN7KM6FL77GE66Bxofmbrzt+gJDVrCG9fDmlXle++vyzzhyvAOXoOt6dHw5g2+L21LSVi2xs+L4dETYMYwz1La7urUXyPdzKBlT+g/HZq280yiI46DFV+k/dLJRDetgPkhhIUhhLXAGKDLZl7fAxhd+MDMDgHqAa9XpKHlpRICIiIiIlKl1G7g+6gu+zj1ywVzcuD4Oz3b5pS7k3vPutUw+S749yEw9wXPYHrJTDj7OV/SOG88DD3aM0Omyrzx8HAbP+dpI7z8QyXWV0u52vV9GeypQ2Hll/6zvTfYa/ulSTIB3G5A8aIHSxLHNmFmjYAmwNuJxznA3cDlFWtm+aiEgIiIiIhUSTk5nhAjHRoeCgd2h2n3+3LNsoQAn70MD7SCt2+GvY7xmaT210ON7T2gOmKAFxVf/QsMbV++unalKVgLE66GMT385+/7LuzfrWLnrCrM4MAzvERF844w+ynYkNkArjy6A8+GEAoXgPYDxocQlmzuTWbWx8zyzSx/xYqK17BQCQERERERiaUON0JudZh4benPL5/rafCfPhuq14JzXyp7L17jNtB3Muy6v9e1m3D1xpk0k/XjIhjREd5/EFr1hb+9UXWKzafSDvXgjMe9jEQas54mUwhjKdCw2OPdE8dK0x3oX+xxa+CvZtYP2B6obmarQggbJUIJIQwBhgDk5eVVeLekSgiIiIiISCzVrg9tL4c3b4T5b8Fe7f34Hz/CO7dB/nCoUduLUx9y/pbr4tVuAL1egdf/6QHYt7Ph9JEerCRj7osw7hKfpTrzCdinc0V+uqrPDLarm9ZLJDMDNwNoZmZNzKw6HqSNK/kiM9sbqAP8f6XCEELPEMIeIYTG+DLKx0oGb+mgEgIiIiIiEluH94O6TWHCVbDuPzB9CNzXEvJHwKEXwKUfenHqZIuaV6sOJ9wBpw6DZbPhkbZlFycvtG41vPoPeKYX7NwM+k6JfvBWSbYYwIUQCoABwETgM2BsCGGumd1kZicXe2l3YEwI6cw3mhyVEBARERGR2KpWAzrdDj98AYP2h9eugPoHek26E+7c+hmiA0+HC96E6tvBqJPg/YdKLzWwcgEMP9azTLYeAOdPSN++vxhKKuwOIYwHxpc4dn2Jxzdu4RwjgZHlat1WUgkBEREREYm15p18xuu7T6DzvbD3ianJ9lhvP+gzCV642Gf4lsyAzvd58hOAOc96ce7cbaDH0/CX4yp+TdlIkvOm2UUlBEREREQk1sw8oUY6UvTX3NH3s00dBG/fAss/hW5D4YOhMGsUNDwcThvuBa8l5SIXwBWWEGjVJL2bB0VEREREqrR01lfLyfG6cQ0Ohuf+5vXPANr8HY6+Nvn9dVJukevZ1es20HrPnWi5x58y3RQRERERkWjb82jo865nuDygG+zVIdMtirzIBXDbVs9lWK9DM90MEREREZF4+FNDOOWhTLciNpTlQ0REREREJEsogBMREREREckSCuBERERERESyhAI4ERERERGRLKEATkREREREJEsogBMREREREckSCuBERERERESyhAI4ERERERGRLKEATkREREREJEsogBMREREREckSCuBERERERESyhAI4ERERERGRLKEATkREREREJEtYCCHTbdiIma0Avk7BqXYGfkjBebKd+qGI+sKpH4qoL1wm+6FRCGGXDF0766RojNR9X0R94dQPRdQXTv1QJFN9Ueb4WOUCuFQxs/wQQl6m25Fp6oci6gunfiiivnDqh3jRv3cR9YVTPxRRXzj1Q5Gq2BdaQikiIiIiIpIlFMCJiIiIiIhkiSgHcEMy3YAqQv1QRH3h1A9F1BdO/RAv+vcuor5w6oci6gunfihS5foisnvgREREREREoibKM3AiIiIiIiKRErkAzsyOM7PPzWy+mV2V6fZkkpl9ZWZzzGy2meVnuj2VycxGmNn3ZvZJsWN1zewNM/sy8bVOJttYGcrohxvNbGnivphtZidkso2Vwcwamtk7Zvapmc01s8sSx+N4T5TVF7G7L+JIY6TT+KjxETRGFtIY6bJpfIzUEkozywW+AI4FlgAzgB4hhE8z2rAMMbOvgLwQQuzqeJhZW2AV8FgIYf/EsTuAH0MI/5P4xaVOCGFgJtuZbmX0w43AqhDCXZlsW2Uys/pA/RDCLDPbAZgJdAXOI373RFl9cQYxuy/iRmNkEY2PGh9BY2QhjZEum8bHqM3AtQLmhxAWhhDWAmOALhluk2RACGEy8GOJw12AUYnvR+H/U0ZaGf0QOyGEZSGEWYnvfwM+A3YjnvdEWX0h0acxUjQ+FqMx0mmMdNk0PkYtgNsNWFzs8RKqaMdXkgC8bmYzzaxPphtTBdQLISxLfP8dUC+TjcmwAWb2cWL5SKSXRJRkZo2BlsB0Yn5PlOgLiPF9ERMaI4tofNxYrD8LSxHbz0KNka6qj49RC+BkY21CCAcDxwP9E0sFBAi+djg664fL5yFgT6AFsAy4O7PNqTxmtj3wHPBfIYRfiz8Xt3uilL6I7X0hsaTxsQxx+ywsRWw/CzVGumwYH6MWwC0FGhZ7vHviWCyFEJYmvn4PvIAvn4mz5Yn1zYXrnL/PcHsyIoSwPISwPoSwARhKTO4LM9sG/0B+MoTwfOJwLO+J0voirvdFzGiMTND4uIlYfhaWJq6fhRojXbaMj1EL4GYAzcysiZlVB7oD4zLcpowws1qJDZiYWS2gI/DJ5t8VeeOAXonvewEvZbAtGVP4YZxwCjG4L8zMgOHAZyGEe4o9Fbt7oqy+iON9EUMaI9H4WIbYfRaWJY6fhRojXTaNj5HKQgmQSO05GMgFRoQQbs1wkzLCzJrif1UEqAY8Fae+MLPRQDtgZ2A5cAPwIjAW2AP4GjgjhBDpzctl9EM7fBlAAL4C+hZb4x5JZtYGmALMATYkDl+Dr22P2z1RVl/0IGb3RRxpjNT4qPGxiMZIpzHSZdP4GLkATkREREREJKqitoRSREREREQkshTAiYiIiIiIZAkFcCIiIiIiIllCAZyIiIiIiEiWUAAnIiIiIiKSJRTAiYiIiIiIZAkFcCIiIiIiIllCAZyIiIiIiEiW+D+MzIdMOY7L8wAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1080x432 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fl-wIq13UK-Z",
        "outputId": "b2ff7bc8-8b19-431c-b0cf-78bcf6b96ee6"
      },
      "source": [
        "# evaluate on val data\n",
        "model.evaluate(val_ds)"
      ],
      "execution_count": 125,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "24/24 [==============================] - 3s 135ms/step - loss: 0.3003 - accuracy: 0.8787\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.30029502511024475, 0.878669261932373]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 125
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2Kj08289UaUh"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pp0rrVS0Xn2n"
      },
      "source": [
        "### Make prediction for test data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FDUvprKR99hU"
      },
      "source": [
        "test_df = pd.read_csv('test.csv')"
      ],
      "execution_count": 126,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NN_X2myZdDEN"
      },
      "source": [
        "# clean text\n",
        "test_df['text'] = test_df['text'].apply(preprocess)"
      ],
      "execution_count": 127,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fUd-emiCXq1y"
      },
      "source": [
        "# encoding the test data\n",
        "test_encoded_inputs = tokenizer(test_df['text'].tolist(),\n",
        "                                 add_special_tokens = True,\n",
        "                                 padding='max_length', \n",
        "                                 truncation=True, \n",
        "                                 max_length=SEQ_LEN, \n",
        "                                 return_token_type_ids=False,\n",
        "                                 return_tensors = 'tf')"
      ],
      "execution_count": 128,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3CmVjWBdX6Zo"
      },
      "source": [
        "# create tf dataset\n",
        "test_dataset = tf.data.Dataset.from_tensor_slices(dict(test_encoded_inputs))\n"
      ],
      "execution_count": 129,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Rkm8HX5QZY2s",
        "outputId": "b4a24ca1-916a-4618-ea77-85b013cd99c1"
      },
      "source": [
        "test_dataset"
      ],
      "execution_count": 130,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<TensorSliceDataset shapes: {input_ids: (30,), attention_mask: (30,)}, types: {input_ids: tf.int32, attention_mask: tf.int32}>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 130
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5u4LsupHZvVl",
        "outputId": "ace95000-35af-4e76-c6fb-c9126b7ce8b0"
      },
      "source": [
        "for d in train_dataset.take(1):\n",
        "  print (d)"
      ],
      "execution_count": 131,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "({'input_ids': <tf.Tensor: shape=(30,), dtype=int32, numpy=\n",
            "array([  101,  2256, 15616,  2024,  1996,  3114,  1997,  2023,  8372,\n",
            "        2089, 16455,  9641,  2149,  2035,   102,     0,     0,     0,\n",
            "           0,     0,     0,     0,     0,     0,     0,     0,     0,\n",
            "           0,     0,     0], dtype=int32)>, 'attention_mask': <tf.Tensor: shape=(30,), dtype=int32, numpy=\n",
            "array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,\n",
            "       0, 0, 0, 0, 0, 0, 0, 0], dtype=int32)>}, <tf.Tensor: shape=(), dtype=int64, numpy=1>)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Whhwppc_YNo3",
        "outputId": "56ef4d0a-3ea7-4e67-c65d-7ccc280ddaa2"
      },
      "source": [
        "# make prediction\n",
        "test_ds = test_dataset.shuffle(100000).batch(64)\n",
        "test_pred = model.predict(test_ds)"
      ],
      "execution_count": 132,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:The parameters `output_attentions`, `output_hidden_states` and `use_cache` cannot be updated when calling a model.They have to be set to True/False in the config object (i.e.: `config=XConfig.from_pretrained('name', output_attentions=True)`).\n",
            "WARNING:tensorflow:The parameter `return_dict` cannot be set in graph mode and will always be set to `True`.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GFNKv8pivpH_",
        "outputId": "93e24b8f-ee63-4261-aa2f-5528a7d4f798"
      },
      "source": [
        "test_pred"
      ],
      "execution_count": 133,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.9743454 ],\n",
              "       [0.9379656 ],\n",
              "       [0.8759267 ],\n",
              "       ...,\n",
              "       [0.34330094],\n",
              "       [0.68898344],\n",
              "       [0.83045846]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 133
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cx91RxXOZDmI",
        "outputId": "2a3a519b-473d-4435-d266-b1c89ae0626b"
      },
      "source": [
        "test_pred.shape"
      ],
      "execution_count": 134,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3263, 1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 134
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rqkehHylYhot",
        "outputId": "2af62f9e-f365-422a-b2df-bb35af1b5a08"
      },
      "source": [
        "# create the target labels for test data\n",
        "test_target = np.round(test_pred).flatten()\n",
        "test_target"
      ],
      "execution_count": 135,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1., 1., 1., ..., 0., 1., 1.], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 135
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p3_mpWblZIKq",
        "outputId": "447cf4e5-aaeb-4490-800a-c3c1caf93b2b"
      },
      "source": [
        "test_target.shape"
      ],
      "execution_count": 136,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3263,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 136
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "m7AN7nHBY0zq",
        "outputId": "93de374f-3e95-4671-99ec-b5b9a33b59d1"
      },
      "source": [
        "test_df['target'] = test_target.astype('int')\n",
        "test_df.head(6)"
      ],
      "execution_count": 137,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>keyword</th>\n",
              "      <th>location</th>\n",
              "      <th>text</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>just happened a terrible car crash</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>heard about earthquake is different cities, st...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>there is a forest fire at spot pond, geese are...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>9</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>apocalypse lighting. spokane wildfires</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>11</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>typhoon soudelor kills 28 in china and taiwan</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>12</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>we're shaking...it's an earthquake</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   id keyword  ...                                               text target\n",
              "0   0     NaN  ...                 just happened a terrible car crash      1\n",
              "1   2     NaN  ...  heard about earthquake is different cities, st...      1\n",
              "2   3     NaN  ...  there is a forest fire at spot pond, geese are...      1\n",
              "3   9     NaN  ...             apocalypse lighting. spokane wildfires      0\n",
              "4  11     NaN  ...      typhoon soudelor kills 28 in china and taiwan      0\n",
              "5  12     NaN  ...                 we're shaking...it's an earthquake      0\n",
              "\n",
              "[6 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 137
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L9nDW6jmZpwd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "outputId": "a0b88aa9-df0c-45f5-84c3-19407ac5b5d0"
      },
      "source": [
        "test_df.tail(6)"
      ],
      "execution_count": 138,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>keyword</th>\n",
              "      <th>location</th>\n",
              "      <th>text</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>3257</th>\n",
              "      <td>10858</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>the death toll in a is-suicide car bombing on ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3258</th>\n",
              "      <td>10861</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>earthquake safety los angeles ûò safety faste...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3259</th>\n",
              "      <td>10865</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>storm in ri worse than last hurricane. my city...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3260</th>\n",
              "      <td>10868</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>green line derailment in chicago</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3261</th>\n",
              "      <td>10874</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>meg issues hazardous weather outlook (hwo)</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3262</th>\n",
              "      <td>10875</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>cityofcalgary has activated its municipal emer...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         id keyword  ...                                               text target\n",
              "3257  10858     NaN  ...  the death toll in a is-suicide car bombing on ...      1\n",
              "3258  10861     NaN  ...  earthquake safety los angeles ûò safety faste...      0\n",
              "3259  10865     NaN  ...  storm in ri worse than last hurricane. my city...      0\n",
              "3260  10868     NaN  ...                   green line derailment in chicago      0\n",
              "3261  10874     NaN  ...         meg issues hazardous weather outlook (hwo)      1\n",
              "3262  10875     NaN  ...  cityofcalgary has activated its municipal emer...      1\n",
              "\n",
              "[6 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 138
        }
      ]
    }
  ]
}