SmartAssistantForDumb&Deaf.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "SmartAssistantForDumb&Deaf.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/jawadkhanubit/Smart-Assistant-for-Dumb-And-Deaf/blob/master/SmartAssistantForDumb%26Deaf.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vpw2Aehdz2zm",
        "colab_type": "text"
      },
      "source": [
        "#Uploading the Data Sete on Colab"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q5VT0XnrzGWu",
        "colab_type": "code",
        "outputId": "8246f814-f9e0-4198-c93b-7dc9f82b84d5",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "Ly8gQ29weXJpZ2h0IDIwMTcgR29vZ2xlIExMQwovLwovLyBMaWNlbnNlZCB1bmRlciB0aGUgQXBhY2hlIExpY2Vuc2UsIFZlcnNpb24gMi4wICh0aGUgIkxpY2Vuc2UiKTsKLy8geW91IG1heSBub3QgdXNlIHRoaXMgZmlsZSBleGNlcHQgaW4gY29tcGxpYW5jZSB3aXRoIHRoZSBMaWNlbnNlLgovLyBZb3UgbWF5IG9idGFpbiBhIGNvcHkgb2YgdGhlIExpY2Vuc2UgYXQKLy8KLy8gICAgICBodHRwOi8vd3d3LmFwYWNoZS5vcmcvbGljZW5zZXMvTElDRU5TRS0yLjAKLy8KLy8gVW5sZXNzIHJlcXVpcmVkIGJ5IGFwcGxpY2FibGUgbGF3IG9yIGFncmVlZCB0byBpbiB3cml0aW5nLCBzb2Z0d2FyZQovLyBkaXN0cmlidXRlZCB1bmRlciB0aGUgTGljZW5zZSBpcyBkaXN0cmlidXRlZCBvbiBhbiAiQVMgSVMiIEJBU0lTLAovLyBXSVRIT1VUIFdBUlJBTlRJRVMgT1IgQ09ORElUSU9OUyBPRiBBTlkgS0lORCwgZWl0aGVyIGV4cHJlc3Mgb3IgaW1wbGllZC4KLy8gU2VlIHRoZSBMaWNlbnNlIGZvciB0aGUgc3BlY2lmaWMgbGFuZ3VhZ2UgZ292ZXJuaW5nIHBlcm1pc3Npb25zIGFuZAovLyBsaW1pdGF0aW9ucyB1bmRlciB0aGUgTGljZW5zZS4KCi8qKgogKiBAZmlsZW92ZXJ2aWV3IEhlbHBlcnMgZm9yIGdvb2dsZS5jb2xhYiBQeXRob24gbW9kdWxlLgogKi8KKGZ1bmN0aW9uKHNjb3BlKSB7CmZ1bmN0aW9uIHNwYW4odGV4dCwgc3R5bGVBdHRyaWJ1dGVzID0ge30pIHsKICBjb25zdCBlbGVtZW50ID0gZG9jdW1lbnQuY3JlYXRlRWxlbWVudCgnc3BhbicpOwogIGVsZW1lbnQudGV4dENvbnRlbnQgPSB0ZXh0OwogIGZvciAoY29uc3Qga2V5IG9mIE9iamVjdC5rZXlzKHN0eWxlQXR0cmlidXRlcykpIHsKICAgIGVsZW1lbnQuc3R5bGVba2V5XSA9IHN0eWxlQXR0cmlidXRlc1trZXldOwogIH0KICByZXR1cm4gZWxlbWVudDsKfQoKLy8gTWF4IG51bWJlciBvZiBieXRlcyB3aGljaCB3aWxsIGJlIHVwbG9hZGVkIGF0IGEgdGltZS4KY29uc3QgTUFYX1BBWUxPQURfU0laRSA9IDEwMCAqIDEwMjQ7Ci8vIE1heCBhbW91bnQgb2YgdGltZSB0byBibG9jayB3YWl0aW5nIGZvciB0aGUgdXNlci4KY29uc3QgRklMRV9DSEFOR0VfVElNRU9VVF9NUyA9IDMwICogMTAwMDsKCmZ1bmN0aW9uIF91cGxvYWRGaWxlcyhpbnB1dElkLCBvdXRwdXRJZCkgewogIGNvbnN0IHN0ZXBzID0gdXBsb2FkRmlsZXNTdGVwKGlucHV0SWQsIG91dHB1dElkKTsKICBjb25zdCBvdXRwdXRFbGVtZW50ID0gZG9jdW1lbnQuZ2V0RWxlbWVudEJ5SWQob3V0cHV0SWQpOwogIC8vIENhY2hlIHN0ZXBzIG9uIHRoZSBvdXRwdXRFbGVtZW50IHRvIG1ha2UgaXQgYXZhaWxhYmxlIGZvciB0aGUgbmV4dCBjYWxsCiAgLy8gdG8gdXBsb2FkRmlsZXNDb250aW51ZSBmcm9tIFB5dGhvbi4KICBvdXRwdXRFbGVtZW50LnN0ZXBzID0gc3RlcHM7CgogIHJldHVybiBfdXBsb2FkRmlsZXNDb250aW51ZShvdXRwdXRJZCk7Cn0KCi8vIFRoaXMgaXMgcm91Z2hseSBhbiBhc3luYyBnZW5lcmF0b3IgKG5vdCBzdXBwb3J0ZWQgaW4gdGhlIGJyb3dzZXIgeWV0KSwKLy8gd2hlcmUgdGhlcmUgYXJlIG11bHRpcGxlIGFzeW5jaHJvbm91cyBzdGVwcyBhbmQgdGhlIFB5dGhvbiBzaWRlIGlzIGdvaW5nCi8vIHRvIHBvbGwgZm9yIGNvbXBsZXRpb24gb2YgZWFjaCBzdGVwLgovLyBUaGlzIHVzZXMgYSBQcm9taXNlIHRvIGJsb2NrIHRoZSBweXRob24gc2lkZSBvbiBjb21wbGV0aW9uIG9mIGVhY2ggc3RlcCwKLy8gdGhlbiBwYXNzZXMgdGhlIHJlc3VsdCBvZiB0aGUgcHJldmlvdXMgc3RlcCBhcyB0aGUgaW5wdXQgdG8gdGhlIG5leHQgc3RlcC4KZnVuY3Rpb24gX3VwbG9hZEZpbGVzQ29udGludWUob3V0cHV0SWQpIHsKICBjb25zdCBvdXRwdXRFbGVtZW50ID0gZG9jdW1lbnQuZ2V0RWxlbWVudEJ5SWQob3V0cHV0SWQpOwogIGNvbnN0IHN0ZXBzID0gb3V0cHV0RWxlbWVudC5zdGVwczsKCiAgY29uc3QgbmV4dCA9IHN0ZXBzLm5leHQob3V0cHV0RWxlbWVudC5sYXN0UHJvbWlzZVZhbHVlKTsKICByZXR1cm4gUHJvbWlzZS5yZXNvbHZlKG5leHQudmFsdWUucHJvbWlzZSkudGhlbigodmFsdWUpID0+IHsKICAgIC8vIENhY2hlIHRoZSBsYXN0IHByb21pc2UgdmFsdWUgdG8gbWFrZSBpdCBhdmFpbGFibGUgdG8gdGhlIG5leHQKICAgIC8vIHN0ZXAgb2YgdGhlIGdlbmVyYXRvci4KICAgIG91dHB1dEVsZW1lbnQubGFzdFByb21pc2VWYWx1ZSA9IHZhbHVlOwogICAgcmV0dXJuIG5leHQudmFsdWUucmVzcG9uc2U7CiAgfSk7Cn0KCi8qKgogKiBHZW5lcmF0b3IgZnVuY3Rpb24gd2hpY2ggaXMgY2FsbGVkIGJldHdlZW4gZWFjaCBhc3luYyBzdGVwIG9mIHRoZSB1cGxvYWQKICogcHJvY2Vzcy4KICogQHBhcmFtIHtzdHJpbmd9IGlucHV0SWQgRWxlbWVudCBJRCBvZiB0aGUgaW5wdXQgZmlsZSBwaWNrZXIgZWxlbWVudC4KICogQHBhcmFtIHtzdHJpbmd9IG91dHB1dElkIEVsZW1lbnQgSUQgb2YgdGhlIG91dHB1dCBkaXNwbGF5LgogKiBAcmV0dXJuIHshSXRlcmFibGU8IU9iamVjdD59IEl0ZXJhYmxlIG9mIG5leHQgc3RlcHMuCiAqLwpmdW5jdGlvbiogdXBsb2FkRmlsZXNTdGVwKGlucHV0SWQsIG91dHB1dElkKSB7CiAgY29uc3QgaW5wdXRFbGVtZW50ID0gZG9jdW1lbnQuZ2V0RWxlbWVudEJ5SWQoaW5wdXRJZCk7CiAgaW5wdXRFbGVtZW50LmRpc2FibGVkID0gZmFsc2U7CgogIGNvbnN0IG91dHB1dEVsZW1lbnQgPSBkb2N1bWVudC5nZXRFbGVtZW50QnlJZChvdXRwdXRJZCk7CiAgb3V0cHV0RWxlbWVudC5pbm5lckhUTUwgPSAnJzsKCiAgY29uc3QgcGlja2VkUHJvbWlzZSA9IG5ldyBQcm9taXNlKChyZXNvbHZlKSA9PiB7CiAgICBpbnB1dEVsZW1lbnQuYWRkRXZlbnRMaXN0ZW5lcignY2hhbmdlJywgKGUpID0+IHsKICAgICAgcmVzb2x2ZShlLnRhcmdldC5maWxlcyk7CiAgICB9KTsKICB9KTsKCiAgY29uc3QgY2FuY2VsID0gZG9jdW1lbnQuY3JlYXRlRWxlbWVudCgnYnV0dG9uJyk7CiAgaW5wdXRFbGVtZW50LnBhcmVudEVsZW1lbnQuYXBwZW5kQ2hpbGQoY2FuY2VsKTsKICBjYW5jZWwudGV4dENvbnRlbnQgPSAnQ2FuY2VsIHVwbG9hZCc7CiAgY29uc3QgY2FuY2VsUHJvbWlzZSA9IG5ldyBQcm9taXNlKChyZXNvbHZlKSA9PiB7CiAgICBjYW5jZWwub25jbGljayA9ICgpID0+IHsKICAgICAgcmVzb2x2ZShudWxsKTsKICAgIH07CiAgfSk7CgogIC8vIENhbmNlbCB1cGxvYWQgaWYgdXNlciBoYXNuJ3QgcGlja2VkIGFueXRoaW5nIGluIHRpbWVvdXQuCiAgY29uc3QgdGltZW91dFByb21pc2UgPSBuZXcgUHJvbWlzZSgocmVzb2x2ZSkgPT4gewogICAgc2V0VGltZW91dCgoKSA9PiB7CiAgICAgIHJlc29sdmUobnVsbCk7CiAgICB9LCBGSUxFX0NIQU5HRV9USU1FT1VUX01TKTsKICB9KTsKCiAgLy8gV2FpdCBmb3IgdGhlIHVzZXIgdG8gcGljayB0aGUgZmlsZXMuCiAgY29uc3QgZmlsZXMgPSB5aWVsZCB7CiAgICBwcm9taXNlOiBQcm9taXNlLnJhY2UoW3BpY2tlZFByb21pc2UsIHRpbWVvdXRQcm9taXNlLCBjYW5jZWxQcm9taXNlXSksCiAgICByZXNwb25zZTogewogICAgICBhY3Rpb246ICdzdGFydGluZycsCiAgICB9CiAgfTsKCiAgaWYgKCFmaWxlcykgewogICAgcmV0dXJuIHsKICAgICAgcmVzcG9uc2U6IHsKICAgICAgICBhY3Rpb246ICdjb21wbGV0ZScsCiAgICAgIH0KICAgIH07CiAgfQoKICBjYW5jZWwucmVtb3ZlKCk7CgogIC8vIERpc2FibGUgdGhlIGlucHV0IGVsZW1lbnQgc2luY2UgZnVydGhlciBwaWNrcyBhcmUgbm90IGFsbG93ZWQuCiAgaW5wdXRFbGVtZW50LmRpc2FibGVkID0gdHJ1ZTsKCiAgZm9yIChjb25zdCBmaWxlIG9mIGZpbGVzKSB7CiAgICBjb25zdCBsaSA9IGRvY3VtZW50LmNyZWF0ZUVsZW1lbnQoJ2xpJyk7CiAgICBsaS5hcHBlbmQoc3BhbihmaWxlLm5hbWUsIHtmb250V2VpZ2h0OiAnYm9sZCd9KSk7CiAgICBsaS5hcHBlbmQoc3BhbigKICAgICAgICBgKCR7ZmlsZS50eXBlIHx8ICduL2EnfSkgLSAke2ZpbGUuc2l6ZX0gYnl0ZXMsIGAgKwogICAgICAgIGBsYXN0IG1vZGlmaWVkOiAkewogICAgICAgICAgICBmaWxlLmxhc3RNb2RpZmllZERhdGUgPyBmaWxlLmxhc3RNb2RpZmllZERhdGUudG9Mb2NhbGVEYXRlU3RyaW5nKCkgOgogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAnbi9hJ30gLSBgKSk7CiAgICBjb25zdCBwZXJjZW50ID0gc3BhbignMCUgZG9uZScpOwogICAgbGkuYXBwZW5kQ2hpbGQocGVyY2VudCk7CgogICAgb3V0cHV0RWxlbWVudC5hcHBlbmRDaGlsZChsaSk7CgogICAgY29uc3QgZmlsZURhdGFQcm9taXNlID0gbmV3IFByb21pc2UoKHJlc29sdmUpID0+IHsKICAgICAgY29uc3QgcmVhZGVyID0gbmV3IEZpbGVSZWFkZXIoKTsKICAgICAgcmVhZGVyLm9ubG9hZCA9IChlKSA9PiB7CiAgICAgICAgcmVzb2x2ZShlLnRhcmdldC5yZXN1bHQpOwogICAgICB9OwogICAgICByZWFkZXIucmVhZEFzQXJyYXlCdWZmZXIoZmlsZSk7CiAgICB9KTsKICAgIC8vIFdhaXQgZm9yIHRoZSBkYXRhIHRvIGJlIHJlYWR5LgogICAgbGV0IGZpbGVEYXRhID0geWllbGQgewogICAgICBwcm9taXNlOiBmaWxlRGF0YVByb21pc2UsCiAgICAgIHJlc3BvbnNlOiB7CiAgICAgICAgYWN0aW9uOiAnY29udGludWUnLAogICAgICB9CiAgICB9OwoKICAgIC8vIFVzZSBhIGNodW5rZWQgc2VuZGluZyB0byBhdm9pZCBtZXNzYWdlIHNpemUgbGltaXRzLiBTZWUgYi82MjExNTY2MC4KICAgIGxldCBwb3NpdGlvbiA9IDA7CiAgICB3aGlsZSAocG9zaXRpb24gPCBmaWxlRGF0YS5ieXRlTGVuZ3RoKSB7CiAgICAgIGNvbnN0IGxlbmd0aCA9IE1hdGgubWluKGZpbGVEYXRhLmJ5dGVMZW5ndGggLSBwb3NpdGlvbiwgTUFYX1BBWUxPQURfU0laRSk7CiAgICAgIGNvbnN0IGNodW5rID0gbmV3IFVpbnQ4QXJyYXkoZmlsZURhdGEsIHBvc2l0aW9uLCBsZW5ndGgpOwogICAgICBwb3NpdGlvbiArPSBsZW5ndGg7CgogICAgICBjb25zdCBiYXNlNjQgPSBidG9hKFN0cmluZy5mcm9tQ2hhckNvZGUuYXBwbHkobnVsbCwgY2h1bmspKTsKICAgICAgeWllbGQgewogICAgICAgIHJlc3BvbnNlOiB7CiAgICAgICAgICBhY3Rpb246ICdhcHBlbmQnLAogICAgICAgICAgZmlsZTogZmlsZS5uYW1lLAogICAgICAgICAgZGF0YTogYmFzZTY0LAogICAgICAgIH0sCiAgICAgIH07CiAgICAgIHBlcmNlbnQudGV4dENvbnRlbnQgPQogICAgICAgICAgYCR7TWF0aC5yb3VuZCgocG9zaXRpb24gLyBmaWxlRGF0YS5ieXRlTGVuZ3RoKSAqIDEwMCl9JSBkb25lYDsKICAgIH0KICB9CgogIC8vIEFsbCBkb25lLgogIHlpZWxkIHsKICAgIHJlc3BvbnNlOiB7CiAgICAgIGFjdGlvbjogJ2NvbXBsZXRlJywKICAgIH0KICB9Owp9CgpzY29wZS5nb29nbGUgPSBzY29wZS5nb29nbGUgfHwge307CnNjb3BlLmdvb2dsZS5jb2xhYiA9IHNjb3BlLmdvb2dsZS5jb2xhYiB8fCB7fTsKc2NvcGUuZ29vZ2xlLmNvbGFiLl9maWxlcyA9IHsKICBfdXBsb2FkRmlsZXMsCiAgX3VwbG9hZEZpbGVzQ29udGludWUsCn07Cn0pKHNlbGYpOwo=",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from google.colab import files\n",
        "uploaded = files.upload()\n",
        "\n",
        "!unzip /content/DataSet.zip"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-b63017b9-e91a-4f31-b468-a17e69fe406f\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-b63017b9-e91a-4f31-b468-a17e69fe406f\">\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 DataSet.zip to DataSet.zip\n",
            "Archive:  /content/DataSet.zip\n",
            "   creating: TrainingSet/I feel pain/\n",
            "  inflating: TrainingSet/I feel pain/1.png  \n",
            "  inflating: TrainingSet/I feel pain/10.png  \n",
            "  inflating: TrainingSet/I feel pain/100.png  \n",
            "  inflating: TrainingSet/I feel pain/11.png  \n",
            "  inflating: TrainingSet/I feel pain/12.png  \n",
            "  inflating: TrainingSet/I feel pain/13.png  \n",
            "  inflating: TrainingSet/I feel pain/14.png  \n",
            "  inflating: TrainingSet/I feel pain/15.png  \n",
            "  inflating: TrainingSet/I feel pain/16.png  \n",
            "  inflating: TrainingSet/I feel pain/17.png  \n",
            "  inflating: TrainingSet/I feel pain/18.png  \n",
            "  inflating: TrainingSet/I feel pain/19.png  \n",
            "  inflating: TrainingSet/I feel pain/2.png  \n",
            "  inflating: TrainingSet/I feel pain/20.png  \n",
            "  inflating: TrainingSet/I feel pain/21.png  \n",
            "  inflating: TrainingSet/I feel pain/22.png  \n",
            "  inflating: TrainingSet/I feel pain/23.png  \n",
            "  inflating: TrainingSet/I feel pain/24.png  \n",
            "  inflating: TrainingSet/I feel pain/25.png  \n",
            "  inflating: TrainingSet/I feel pain/26.png  \n",
            "  inflating: TrainingSet/I feel pain/27.png  \n",
            "  inflating: TrainingSet/I feel pain/28.png  \n",
            "  inflating: TrainingSet/I feel pain/29.png  \n",
            "  inflating: TrainingSet/I feel pain/3.png  \n",
            "  inflating: TrainingSet/I feel pain/30.png  \n",
            "  inflating: TrainingSet/I feel pain/31.png  \n",
            "  inflating: TrainingSet/I feel pain/32.png  \n",
            "  inflating: TrainingSet/I feel pain/33.png  \n",
            "  inflating: TrainingSet/I feel pain/34.png  \n",
            "  inflating: TrainingSet/I feel pain/35.png  \n",
            "  inflating: TrainingSet/I feel pain/36.png  \n",
            "  inflating: TrainingSet/I feel pain/37.png  \n",
            "  inflating: TrainingSet/I feel pain/38.png  \n",
            "  inflating: TrainingSet/I feel pain/39.png  \n",
            "  inflating: TrainingSet/I feel pain/4.png  \n",
            "  inflating: TrainingSet/I feel pain/40.png  \n",
            "  inflating: TrainingSet/I feel pain/41.png  \n",
            "  inflating: TrainingSet/I feel pain/42.png  \n",
            "  inflating: TrainingSet/I feel pain/43.png  \n",
            "  inflating: TrainingSet/I feel pain/44.png  \n",
            "  inflating: TrainingSet/I feel pain/45.png  \n",
            "  inflating: TrainingSet/I feel pain/46.png  \n",
            "  inflating: TrainingSet/I feel pain/47.png  \n",
            "  inflating: TrainingSet/I feel pain/48.png  \n",
            "  inflating: TrainingSet/I feel pain/49.png  \n",
            "  inflating: TrainingSet/I feel pain/5.png  \n",
            "  inflating: TrainingSet/I feel pain/50.png  \n",
            "  inflating: TrainingSet/I feel pain/51.png  \n",
            "  inflating: TrainingSet/I feel pain/52.png  \n",
            "  inflating: TrainingSet/I feel pain/53.png  \n",
            "  inflating: TrainingSet/I feel pain/54.png  \n",
            "  inflating: TrainingSet/I feel pain/55.png  \n",
            "  inflating: TrainingSet/I feel pain/56.png  \n",
            "  inflating: TrainingSet/I feel pain/57.png  \n",
            "  inflating: TrainingSet/I feel pain/58.png  \n",
            "  inflating: TrainingSet/I feel pain/59.png  \n",
            "  inflating: TrainingSet/I feel pain/6.png  \n",
            "  inflating: TrainingSet/I feel pain/60.png  \n",
            "  inflating: TrainingSet/I feel pain/61.png  \n",
            "  inflating: TrainingSet/I feel pain/62.png  \n",
            "  inflating: TrainingSet/I feel pain/63.png  \n",
            "  inflating: TrainingSet/I feel pain/64.png  \n",
            "  inflating: TrainingSet/I feel pain/65.png  \n",
            "  inflating: TrainingSet/I feel pain/66.png  \n",
            "  inflating: TrainingSet/I feel pain/67.png  \n",
            "  inflating: TrainingSet/I feel pain/68.png  \n",
            "  inflating: TrainingSet/I feel pain/69.png  \n",
            "  inflating: TrainingSet/I feel pain/7.png  \n",
            "  inflating: TrainingSet/I feel pain/70.png  \n",
            "  inflating: TrainingSet/I feel pain/71.png  \n",
            "  inflating: TrainingSet/I feel pain/72.png  \n",
            "  inflating: TrainingSet/I feel pain/73.png  \n",
            "  inflating: TrainingSet/I feel pain/74.png  \n",
            "  inflating: TrainingSet/I feel pain/75.png  \n",
            "  inflating: TrainingSet/I feel pain/76.png  \n",
            "  inflating: TrainingSet/I feel pain/77.png  \n",
            "  inflating: TrainingSet/I feel pain/78.png  \n",
            "  inflating: TrainingSet/I feel pain/79.png  \n",
            "  inflating: TrainingSet/I feel pain/8.png  \n",
            "  inflating: TrainingSet/I feel pain/80.png  \n",
            "  inflating: TrainingSet/I feel pain/81.png  \n",
            "  inflating: TrainingSet/I feel pain/82.png  \n",
            "  inflating: TrainingSet/I feel pain/83.png  \n",
            "  inflating: TrainingSet/I feel pain/84.png  \n",
            "  inflating: TrainingSet/I feel pain/85.png  \n",
            "  inflating: TrainingSet/I feel pain/86.png  \n",
            "  inflating: TrainingSet/I feel pain/87.png  \n",
            "  inflating: TrainingSet/I feel pain/88.png  \n",
            "  inflating: TrainingSet/I feel pain/89.png  \n",
            "  inflating: TrainingSet/I feel pain/9.png  \n",
            "  inflating: TrainingSet/I feel pain/90.png  \n",
            "  inflating: TrainingSet/I feel pain/91.png  \n",
            "  inflating: TrainingSet/I feel pain/92.png  \n",
            "  inflating: TrainingSet/I feel pain/93.png  \n",
            "  inflating: TrainingSet/I feel pain/94.png  \n",
            "  inflating: TrainingSet/I feel pain/95.png  \n",
            "  inflating: TrainingSet/I feel pain/96.png  \n",
            "  inflating: TrainingSet/I feel pain/97.png  \n",
            "  inflating: TrainingSet/I feel pain/98.png  \n",
            "  inflating: TrainingSet/I feel pain/99.png  \n",
            "   creating: TrainingSet/I need help/\n",
            "  inflating: TrainingSet/I need help/1.png  \n",
            "  inflating: TrainingSet/I need help/10.png  \n",
            "  inflating: TrainingSet/I need help/100.png  \n",
            "  inflating: TrainingSet/I need help/11.png  \n",
            "  inflating: TrainingSet/I need help/12.png  \n",
            "  inflating: TrainingSet/I need help/13.png  \n",
            "  inflating: TrainingSet/I need help/14.png  \n",
            "  inflating: TrainingSet/I need help/15.png  \n",
            "  inflating: TrainingSet/I need help/16.png  \n",
            "  inflating: TrainingSet/I need help/17.png  \n",
            "  inflating: TrainingSet/I need help/18.png  \n",
            "  inflating: TrainingSet/I need help/19.png  \n",
            "  inflating: TrainingSet/I need help/2.png  \n",
            "  inflating: TrainingSet/I need help/20.png  \n",
            "  inflating: TrainingSet/I need help/21.png  \n",
            "  inflating: TrainingSet/I need help/22.png  \n",
            "  inflating: TrainingSet/I need help/23.png  \n",
            "  inflating: TrainingSet/I need help/24.png  \n",
            "  inflating: TrainingSet/I need help/25.png  \n",
            "  inflating: TrainingSet/I need help/26.png  \n",
            "  inflating: TrainingSet/I need help/27.png  \n",
            "  inflating: TrainingSet/I need help/28.png  \n",
            "  inflating: TrainingSet/I need help/29.png  \n",
            "  inflating: TrainingSet/I need help/3.png  \n",
            "  inflating: TrainingSet/I need help/30.png  \n",
            "  inflating: TrainingSet/I need help/31.png  \n",
            "  inflating: TrainingSet/I need help/32.png  \n",
            "  inflating: TrainingSet/I need help/33.png  \n",
            "  inflating: TrainingSet/I need help/34.png  \n",
            "  inflating: TrainingSet/I need help/35.png  \n",
            "  inflating: TrainingSet/I need help/36.png  \n",
            "  inflating: TrainingSet/I need help/37.png  \n",
            "  inflating: TrainingSet/I need help/38.png  \n",
            "  inflating: TrainingSet/I need help/39.png  \n",
            "  inflating: TrainingSet/I need help/4.png  \n",
            "  inflating: TrainingSet/I need help/40.png  \n",
            "  inflating: TrainingSet/I need help/41.png  \n",
            "  inflating: TrainingSet/I need help/42.png  \n",
            "  inflating: TrainingSet/I need help/43.png  \n",
            "  inflating: TrainingSet/I need help/44.png  \n",
            "  inflating: TrainingSet/I need help/45.png  \n",
            "  inflating: TrainingSet/I need help/46.png  \n",
            "  inflating: TrainingSet/I need help/47.png  \n",
            "  inflating: TrainingSet/I need help/48.png  \n",
            "  inflating: TrainingSet/I need help/49.png  \n",
            "  inflating: TrainingSet/I need help/5.png  \n",
            "  inflating: TrainingSet/I need help/50.png  \n",
            "  inflating: TrainingSet/I need help/51.png  \n",
            "  inflating: TrainingSet/I need help/52.png  \n",
            "  inflating: TrainingSet/I need help/53.png  \n",
            "  inflating: TrainingSet/I need help/54.png  \n",
            "  inflating: TrainingSet/I need help/55.png  \n",
            "  inflating: TrainingSet/I need help/56.png  \n",
            "  inflating: TrainingSet/I need help/57.png  \n",
            "  inflating: TrainingSet/I need help/58.png  \n",
            "  inflating: TrainingSet/I need help/59.png  \n",
            "  inflating: TrainingSet/I need help/6.png  \n",
            "  inflating: TrainingSet/I need help/60.png  \n",
            "  inflating: TrainingSet/I need help/61.png  \n",
            "  inflating: TrainingSet/I need help/62.png  \n",
            "  inflating: TrainingSet/I need help/63.png  \n",
            "  inflating: TrainingSet/I need help/64.png  \n",
            "  inflating: TrainingSet/I need help/65.png  \n",
            "  inflating: TrainingSet/I need help/66.png  \n",
            "  inflating: TrainingSet/I need help/67.png  \n",
            "  inflating: TrainingSet/I need help/68.png  \n",
            "  inflating: TrainingSet/I need help/69.png  \n",
            "  inflating: TrainingSet/I need help/7.png  \n",
            "  inflating: TrainingSet/I need help/70.png  \n",
            "  inflating: TrainingSet/I need help/71.png  \n",
            "  inflating: TrainingSet/I need help/72.png  \n",
            "  inflating: TrainingSet/I need help/73.png  \n",
            "  inflating: TrainingSet/I need help/74.png  \n",
            "  inflating: TrainingSet/I need help/75.png  \n",
            "  inflating: TrainingSet/I need help/76.png  \n",
            "  inflating: TrainingSet/I need help/77.png  \n",
            "  inflating: TrainingSet/I need help/78.png  \n",
            "  inflating: TrainingSet/I need help/79.png  \n",
            "  inflating: TrainingSet/I need help/8.png  \n",
            "  inflating: TrainingSet/I need help/80.png  \n",
            "  inflating: TrainingSet/I need help/81.png  \n",
            "  inflating: TrainingSet/I need help/82.png  \n",
            "  inflating: TrainingSet/I need help/83.png  \n",
            "  inflating: TrainingSet/I need help/84.png  \n",
            "  inflating: TrainingSet/I need help/85.png  \n",
            "  inflating: TrainingSet/I need help/86.png  \n",
            "  inflating: TrainingSet/I need help/87.png  \n",
            "  inflating: TrainingSet/I need help/88.png  \n",
            "  inflating: TrainingSet/I need help/89.png  \n",
            "  inflating: TrainingSet/I need help/9.png  \n",
            "  inflating: TrainingSet/I need help/90.png  \n",
            "  inflating: TrainingSet/I need help/91.png  \n",
            "  inflating: TrainingSet/I need help/92.png  \n",
            "  inflating: TrainingSet/I need help/93.png  \n",
            "  inflating: TrainingSet/I need help/94.png  \n",
            "  inflating: TrainingSet/I need help/95.png  \n",
            "  inflating: TrainingSet/I need help/96.png  \n",
            "  inflating: TrainingSet/I need help/97.png  \n",
            "  inflating: TrainingSet/I need help/98.png  \n",
            "  inflating: TrainingSet/I need help/99.png  \n",
            "   creating: TrainingSet/I need water/\n",
            "  inflating: TrainingSet/I need water/1.png  \n",
            "  inflating: TrainingSet/I need water/10.png  \n",
            "  inflating: TrainingSet/I need water/100.png  \n",
            "  inflating: TrainingSet/I need water/11.png  \n",
            "  inflating: TrainingSet/I need water/12.png  \n",
            "  inflating: TrainingSet/I need water/13.png  \n",
            "  inflating: TrainingSet/I need water/14.png  \n",
            "  inflating: TrainingSet/I need water/15.png  \n",
            "  inflating: TrainingSet/I need water/16.png  \n",
            "  inflating: TrainingSet/I need water/17.png  \n",
            "  inflating: TrainingSet/I need water/18.png  \n",
            "  inflating: TrainingSet/I need water/19.png  \n",
            "  inflating: TrainingSet/I need water/2.png  \n",
            "  inflating: TrainingSet/I need water/20.png  \n",
            "  inflating: TrainingSet/I need water/21.png  \n",
            "  inflating: TrainingSet/I need water/22.png  \n",
            "  inflating: TrainingSet/I need water/23.png  \n",
            "  inflating: TrainingSet/I need water/24.png  \n",
            "  inflating: TrainingSet/I need water/25.png  \n",
            "  inflating: TrainingSet/I need water/26.png  \n",
            "  inflating: TrainingSet/I need water/27.png  \n",
            "  inflating: TrainingSet/I need water/28.png  \n",
            "  inflating: TrainingSet/I need water/29.png  \n",
            "  inflating: TrainingSet/I need water/3.png  \n",
            "  inflating: TrainingSet/I need water/30.png  \n",
            "  inflating: TrainingSet/I need water/31.png  \n",
            "  inflating: TrainingSet/I need water/32.png  \n",
            "  inflating: TrainingSet/I need water/33.png  \n",
            "  inflating: TrainingSet/I need water/34.png  \n",
            "  inflating: TrainingSet/I need water/35.png  \n",
            "  inflating: TrainingSet/I need water/36.png  \n",
            "  inflating: TrainingSet/I need water/37.png  \n",
            "  inflating: TrainingSet/I need water/38.png  \n",
            "  inflating: TrainingSet/I need water/39.png  \n",
            "  inflating: TrainingSet/I need water/4.png  \n",
            "  inflating: TrainingSet/I need water/40.png  \n",
            "  inflating: TrainingSet/I need water/41.png  \n",
            "  inflating: TrainingSet/I need water/42.png  \n",
            "  inflating: TrainingSet/I need water/43.png  \n",
            "  inflating: TrainingSet/I need water/44.png  \n",
            "  inflating: TrainingSet/I need water/45.png  \n",
            "  inflating: TrainingSet/I need water/46.png  \n",
            "  inflating: TrainingSet/I need water/47.png  \n",
            "  inflating: TrainingSet/I need water/48.png  \n",
            "  inflating: TrainingSet/I need water/49.png  \n",
            "  inflating: TrainingSet/I need water/5.png  \n",
            "  inflating: TrainingSet/I need water/50.png  \n",
            "  inflating: TrainingSet/I need water/51.png  \n",
            "  inflating: TrainingSet/I need water/52.png  \n",
            "  inflating: TrainingSet/I need water/53.png  \n",
            "  inflating: TrainingSet/I need water/54.png  \n",
            "  inflating: TrainingSet/I need water/55.png  \n",
            "  inflating: TrainingSet/I need water/56.png  \n",
            "  inflating: TrainingSet/I need water/57.png  \n",
            "  inflating: TrainingSet/I need water/58.png  \n",
            "  inflating: TrainingSet/I need water/59.png  \n",
            "  inflating: TrainingSet/I need water/6.png  \n",
            "  inflating: TrainingSet/I need water/60.png  \n",
            "  inflating: TrainingSet/I need water/61.png  \n",
            "  inflating: TrainingSet/I need water/62.png  \n",
            "  inflating: TrainingSet/I need water/63.png  \n",
            "  inflating: TrainingSet/I need water/64.png  \n",
            "  inflating: TrainingSet/I need water/65.png  \n",
            "  inflating: TrainingSet/I need water/66.png  \n",
            "  inflating: TrainingSet/I need water/67.png  \n",
            "  inflating: TrainingSet/I need water/68.png  \n",
            "  inflating: TrainingSet/I need water/69.png  \n",
            "  inflating: TrainingSet/I need water/7.png  \n",
            "  inflating: TrainingSet/I need water/70.png  \n",
            "  inflating: TrainingSet/I need water/71.png  \n",
            "  inflating: TrainingSet/I need water/72.png  \n",
            "  inflating: TrainingSet/I need water/73.png  \n",
            "  inflating: TrainingSet/I need water/74.png  \n",
            "  inflating: TrainingSet/I need water/75.png  \n",
            "  inflating: TrainingSet/I need water/76.png  \n",
            "  inflating: TrainingSet/I need water/77.png  \n",
            "  inflating: TrainingSet/I need water/78.png  \n",
            "  inflating: TrainingSet/I need water/79.png  \n",
            "  inflating: TrainingSet/I need water/8.png  \n",
            "  inflating: TrainingSet/I need water/80.png  \n",
            "  inflating: TrainingSet/I need water/81.png  \n",
            "  inflating: TrainingSet/I need water/82.png  \n",
            "  inflating: TrainingSet/I need water/83.png  \n",
            "  inflating: TrainingSet/I need water/84.png  \n",
            "  inflating: TrainingSet/I need water/85.png  \n",
            "  inflating: TrainingSet/I need water/86.png  \n",
            "  inflating: TrainingSet/I need water/87.png  \n",
            "  inflating: TrainingSet/I need water/88.png  \n",
            "  inflating: TrainingSet/I need water/89.png  \n",
            "  inflating: TrainingSet/I need water/9.png  \n",
            "  inflating: TrainingSet/I need water/90.png  \n",
            "  inflating: TrainingSet/I need water/91.png  \n",
            "  inflating: TrainingSet/I need water/92.png  \n",
            "  inflating: TrainingSet/I need water/93.png  \n",
            "  inflating: TrainingSet/I need water/94.png  \n",
            "  inflating: TrainingSet/I need water/95.png  \n",
            "  inflating: TrainingSet/I need water/96.png  \n",
            "  inflating: TrainingSet/I need water/97.png  \n",
            "  inflating: TrainingSet/I need water/98.png  \n",
            "  inflating: TrainingSet/I need water/99.png  \n",
            "   creating: TrainingSet/No/\n",
            "  inflating: TrainingSet/No/1.png    \n",
            "  inflating: TrainingSet/No/10.png   \n",
            "  inflating: TrainingSet/No/100.png  \n",
            "  inflating: TrainingSet/No/101.png  \n",
            "  inflating: TrainingSet/No/102.png  \n",
            "  inflating: TrainingSet/No/103.png  \n",
            "  inflating: TrainingSet/No/104.png  \n",
            "  inflating: TrainingSet/No/105.png  \n",
            "  inflating: TrainingSet/No/106.png  \n",
            "  inflating: TrainingSet/No/107.png  \n",
            "  inflating: TrainingSet/No/108.png  \n",
            "  inflating: TrainingSet/No/109.png  \n",
            "  inflating: TrainingSet/No/11.png   \n",
            "  inflating: TrainingSet/No/110.png  \n",
            "  inflating: TrainingSet/No/111.png  \n",
            "  inflating: TrainingSet/No/112.png  \n",
            "  inflating: TrainingSet/No/113.png  \n",
            "  inflating: TrainingSet/No/114.png  \n",
            "  inflating: TrainingSet/No/115.png  \n",
            "  inflating: TrainingSet/No/116.png  \n",
            "  inflating: TrainingSet/No/117.png  \n",
            "  inflating: TrainingSet/No/118.png  \n",
            "  inflating: TrainingSet/No/119.png  \n",
            "  inflating: TrainingSet/No/12.png   \n",
            "  inflating: TrainingSet/No/120.png  \n",
            "  inflating: TrainingSet/No/121.png  \n",
            "  inflating: TrainingSet/No/122.png  \n",
            "  inflating: TrainingSet/No/123.png  \n",
            "  inflating: TrainingSet/No/124.png  \n",
            "  inflating: TrainingSet/No/125.png  \n",
            "  inflating: TrainingSet/No/126.png  \n",
            "  inflating: TrainingSet/No/127.png  \n",
            "  inflating: TrainingSet/No/128.png  \n",
            "  inflating: TrainingSet/No/129.png  \n",
            "  inflating: TrainingSet/No/13.png   \n",
            "  inflating: TrainingSet/No/130.png  \n",
            "  inflating: TrainingSet/No/131.png  \n",
            "  inflating: TrainingSet/No/132.png  \n",
            "  inflating: TrainingSet/No/133.png  \n",
            "  inflating: TrainingSet/No/134.png  \n",
            "  inflating: TrainingSet/No/135.png  \n",
            "  inflating: TrainingSet/No/136.png  \n",
            "  inflating: TrainingSet/No/137.png  \n",
            "  inflating: TrainingSet/No/138.png  \n",
            "  inflating: TrainingSet/No/139.png  \n",
            "  inflating: TrainingSet/No/14.png   \n",
            "  inflating: TrainingSet/No/140.png  \n",
            "  inflating: TrainingSet/No/141.png  \n",
            "  inflating: TrainingSet/No/142.png  \n",
            "  inflating: TrainingSet/No/143.png  \n",
            "  inflating: TrainingSet/No/144.png  \n",
            "  inflating: TrainingSet/No/145.png  \n",
            "  inflating: TrainingSet/No/146.png  \n",
            "  inflating: TrainingSet/No/147.png  \n",
            "  inflating: TrainingSet/No/148.png  \n",
            "  inflating: TrainingSet/No/149.png  \n",
            "  inflating: TrainingSet/No/15.png   \n",
            "  inflating: TrainingSet/No/150.png  \n",
            "  inflating: TrainingSet/No/151.png  \n",
            "  inflating: TrainingSet/No/152.png  \n",
            "  inflating: TrainingSet/No/153.png  \n",
            "  inflating: TrainingSet/No/154.png  \n",
            "  inflating: TrainingSet/No/155.png  \n",
            "  inflating: TrainingSet/No/156.png  \n",
            "  inflating: TrainingSet/No/157.png  \n",
            "  inflating: TrainingSet/No/158.png  \n",
            "  inflating: TrainingSet/No/159.png  \n",
            "  inflating: TrainingSet/No/16.png   \n",
            "  inflating: TrainingSet/No/160.png  \n",
            "  inflating: TrainingSet/No/161.png  \n",
            "  inflating: TrainingSet/No/162.png  \n",
            "  inflating: TrainingSet/No/163.png  \n",
            "  inflating: TrainingSet/No/164.png  \n",
            "  inflating: TrainingSet/No/165.png  \n",
            "  inflating: TrainingSet/No/17.png   \n",
            "  inflating: TrainingSet/No/18.png   \n",
            "  inflating: TrainingSet/No/19.png   \n",
            "  inflating: TrainingSet/No/2.png    \n",
            "  inflating: TrainingSet/No/20.png   \n",
            "  inflating: TrainingSet/No/21.png   \n",
            "  inflating: TrainingSet/No/22.png   \n",
            "  inflating: TrainingSet/No/23.png   \n",
            "  inflating: TrainingSet/No/24.png   \n",
            "  inflating: TrainingSet/No/25.png   \n",
            "  inflating: TrainingSet/No/26.png   \n",
            "  inflating: TrainingSet/No/27.png   \n",
            "  inflating: TrainingSet/No/28.png   \n",
            "  inflating: TrainingSet/No/29.png   \n",
            "  inflating: TrainingSet/No/3.png    \n",
            "  inflating: TrainingSet/No/30.png   \n",
            "  inflating: TrainingSet/No/31.png   \n",
            "  inflating: TrainingSet/No/32.png   \n",
            "  inflating: TrainingSet/No/33.png   \n",
            "  inflating: TrainingSet/No/34.png   \n",
            "  inflating: TrainingSet/No/35.png   \n",
            "  inflating: TrainingSet/No/36.png   \n",
            "  inflating: TrainingSet/No/37.png   \n",
            "  inflating: TrainingSet/No/38.png   \n",
            "  inflating: TrainingSet/No/39.png   \n",
            "  inflating: TrainingSet/No/4.png    \n",
            "  inflating: TrainingSet/No/40.png   \n",
            "  inflating: TrainingSet/No/41.png   \n",
            "  inflating: TrainingSet/No/42.png   \n",
            "  inflating: TrainingSet/No/43.png   \n",
            "  inflating: TrainingSet/No/44.png   \n",
            "  inflating: TrainingSet/No/45.png   \n",
            "  inflating: TrainingSet/No/46.png   \n",
            "  inflating: TrainingSet/No/47.png   \n",
            "  inflating: TrainingSet/No/48.png   \n",
            "  inflating: TrainingSet/No/49.png   \n",
            "  inflating: TrainingSet/No/5.png    \n",
            "  inflating: TrainingSet/No/50.png   \n",
            "  inflating: TrainingSet/No/51.png   \n",
            "  inflating: TrainingSet/No/52.png   \n",
            "  inflating: TrainingSet/No/53.png   \n",
            "  inflating: TrainingSet/No/54.png   \n",
            "  inflating: TrainingSet/No/55.png   \n",
            "  inflating: TrainingSet/No/56.png   \n",
            "  inflating: TrainingSet/No/57.png   \n",
            "  inflating: TrainingSet/No/58.png   \n",
            "  inflating: TrainingSet/No/59.png   \n",
            "  inflating: TrainingSet/No/6.png    \n",
            "  inflating: TrainingSet/No/60.png   \n",
            "  inflating: TrainingSet/No/61.png   \n",
            "  inflating: TrainingSet/No/62.png   \n",
            "  inflating: TrainingSet/No/63.png   \n",
            "  inflating: TrainingSet/No/64.png   \n",
            "  inflating: TrainingSet/No/65.png   \n",
            "  inflating: TrainingSet/No/66.png   \n",
            "  inflating: TrainingSet/No/67.png   \n",
            "  inflating: TrainingSet/No/68.png   \n",
            "  inflating: TrainingSet/No/69.png   \n",
            "  inflating: TrainingSet/No/7.png    \n",
            "  inflating: TrainingSet/No/70.png   \n",
            "  inflating: TrainingSet/No/71.png   \n",
            "  inflating: TrainingSet/No/72.png   \n",
            "  inflating: TrainingSet/No/73.png   \n",
            "  inflating: TrainingSet/No/74.png   \n",
            "  inflating: TrainingSet/No/75.png   \n",
            "  inflating: TrainingSet/No/76.png   \n",
            "  inflating: TrainingSet/No/77.png   \n",
            "  inflating: TrainingSet/No/78.png   \n",
            "  inflating: TrainingSet/No/79.png   \n",
            "  inflating: TrainingSet/No/8.png    \n",
            "  inflating: TrainingSet/No/80.png   \n",
            "  inflating: TrainingSet/No/81.png   \n",
            "  inflating: TrainingSet/No/82.png   \n",
            "  inflating: TrainingSet/No/83.png   \n",
            "  inflating: TrainingSet/No/84.png   \n",
            "  inflating: TrainingSet/No/85.png   \n",
            "  inflating: TrainingSet/No/86.png   \n",
            "  inflating: TrainingSet/No/87.png   \n",
            "  inflating: TrainingSet/No/88.png   \n",
            "  inflating: TrainingSet/No/89.png   \n",
            "  inflating: TrainingSet/No/9.png    \n",
            "  inflating: TrainingSet/No/90.png   \n",
            "  inflating: TrainingSet/No/91.png   \n",
            "  inflating: TrainingSet/No/92.png   \n",
            "  inflating: TrainingSet/No/93.png   \n",
            "  inflating: TrainingSet/No/94.png   \n",
            "  inflating: TrainingSet/No/95.png   \n",
            "  inflating: TrainingSet/No/96.png   \n",
            "  inflating: TrainingSet/No/97.png   \n",
            "  inflating: TrainingSet/No/98.png   \n",
            "  inflating: TrainingSet/No/99.png   \n",
            "   creating: TrainingSet/Yes/\n",
            "  inflating: TrainingSet/Yes/1.png   \n",
            "  inflating: TrainingSet/Yes/10.png  \n",
            "  inflating: TrainingSet/Yes/100.png  \n",
            "  inflating: TrainingSet/Yes/101.png  \n",
            "  inflating: TrainingSet/Yes/102.png  \n",
            "  inflating: TrainingSet/Yes/103.png  \n",
            "  inflating: TrainingSet/Yes/104.png  \n",
            "  inflating: TrainingSet/Yes/105.png  \n",
            "  inflating: TrainingSet/Yes/106.png  \n",
            "  inflating: TrainingSet/Yes/107.png  \n",
            "  inflating: TrainingSet/Yes/108.png  \n",
            "  inflating: TrainingSet/Yes/109.png  \n",
            "  inflating: TrainingSet/Yes/11.png  \n",
            "  inflating: TrainingSet/Yes/110.png  \n",
            "  inflating: TrainingSet/Yes/112.png  \n",
            "  inflating: TrainingSet/Yes/113.png  \n",
            "  inflating: TrainingSet/Yes/115.png  \n",
            "  inflating: TrainingSet/Yes/116.png  \n",
            "  inflating: TrainingSet/Yes/117.png  \n",
            "  inflating: TrainingSet/Yes/118.png  \n",
            "  inflating: TrainingSet/Yes/119.png  \n",
            "  inflating: TrainingSet/Yes/12.png  \n",
            "  inflating: TrainingSet/Yes/120.png  \n",
            "  inflating: TrainingSet/Yes/121.png  \n",
            "  inflating: TrainingSet/Yes/122.png  \n",
            "  inflating: TrainingSet/Yes/123.png  \n",
            "  inflating: TrainingSet/Yes/124.png  \n",
            "  inflating: TrainingSet/Yes/125.png  \n",
            "  inflating: TrainingSet/Yes/126.png  \n",
            "  inflating: TrainingSet/Yes/127.png  \n",
            "  inflating: TrainingSet/Yes/128.png  \n",
            "  inflating: TrainingSet/Yes/129.png  \n",
            "  inflating: TrainingSet/Yes/13.png  \n",
            "  inflating: TrainingSet/Yes/130.png  \n",
            "  inflating: TrainingSet/Yes/131.png  \n",
            "  inflating: TrainingSet/Yes/132.png  \n",
            "  inflating: TrainingSet/Yes/133.png  \n",
            "  inflating: TrainingSet/Yes/135.png  \n",
            "  inflating: TrainingSet/Yes/136.png  \n",
            "  inflating: TrainingSet/Yes/137.png  \n",
            "  inflating: TrainingSet/Yes/138.png  \n",
            "  inflating: TrainingSet/Yes/139.png  \n",
            "  inflating: TrainingSet/Yes/14.png  \n",
            "  inflating: TrainingSet/Yes/140.png  \n",
            "  inflating: TrainingSet/Yes/141.png  \n",
            "  inflating: TrainingSet/Yes/142.png  \n",
            "  inflating: TrainingSet/Yes/143.png  \n",
            "  inflating: TrainingSet/Yes/144.png  \n",
            "  inflating: TrainingSet/Yes/145.png  \n",
            "  inflating: TrainingSet/Yes/146.png  \n",
            "  inflating: TrainingSet/Yes/147.png  \n",
            "  inflating: TrainingSet/Yes/148.png  \n",
            "  inflating: TrainingSet/Yes/149.png  \n",
            "  inflating: TrainingSet/Yes/15.png  \n",
            "  inflating: TrainingSet/Yes/150.png  \n",
            "  inflating: TrainingSet/Yes/16.png  \n",
            "  inflating: TrainingSet/Yes/17.png  \n",
            "  inflating: TrainingSet/Yes/18.png  \n",
            "  inflating: TrainingSet/Yes/19.png  \n",
            "  inflating: TrainingSet/Yes/2.png   \n",
            "  inflating: TrainingSet/Yes/20.png  \n",
            "  inflating: TrainingSet/Yes/21.png  \n",
            "  inflating: TrainingSet/Yes/22.png  \n",
            "  inflating: TrainingSet/Yes/23.png  \n",
            "  inflating: TrainingSet/Yes/24.png  \n",
            "  inflating: TrainingSet/Yes/25.png  \n",
            "  inflating: TrainingSet/Yes/26.png  \n",
            "  inflating: TrainingSet/Yes/27.png  \n",
            "  inflating: TrainingSet/Yes/28.png  \n",
            "  inflating: TrainingSet/Yes/29.png  \n",
            "  inflating: TrainingSet/Yes/3.png   \n",
            "  inflating: TrainingSet/Yes/30.png  \n",
            "  inflating: TrainingSet/Yes/31.png  \n",
            "  inflating: TrainingSet/Yes/32.png  \n",
            "  inflating: TrainingSet/Yes/33.png  \n",
            "  inflating: TrainingSet/Yes/34.png  \n",
            "  inflating: TrainingSet/Yes/35.png  \n",
            "  inflating: TrainingSet/Yes/36.png  \n",
            "  inflating: TrainingSet/Yes/37.png  \n",
            "  inflating: TrainingSet/Yes/38.png  \n",
            "  inflating: TrainingSet/Yes/39.png  \n",
            "  inflating: TrainingSet/Yes/4.png   \n",
            "  inflating: TrainingSet/Yes/40.png  \n",
            "  inflating: TrainingSet/Yes/42.png  \n",
            "  inflating: TrainingSet/Yes/43.png  \n",
            "  inflating: TrainingSet/Yes/44.png  \n",
            "  inflating: TrainingSet/Yes/45.png  \n",
            "  inflating: TrainingSet/Yes/46.png  \n",
            "  inflating: TrainingSet/Yes/47.png  \n",
            "  inflating: TrainingSet/Yes/48.png  \n",
            "  inflating: TrainingSet/Yes/49.png  \n",
            "  inflating: TrainingSet/Yes/5.png   \n",
            "  inflating: TrainingSet/Yes/50.png  \n",
            "  inflating: TrainingSet/Yes/51.png  \n",
            "  inflating: TrainingSet/Yes/52.png  \n",
            "  inflating: TrainingSet/Yes/53.png  \n",
            "  inflating: TrainingSet/Yes/54.png  \n",
            "  inflating: TrainingSet/Yes/55.png  \n",
            "  inflating: TrainingSet/Yes/56.png  \n",
            "  inflating: TrainingSet/Yes/57.png  \n",
            "  inflating: TrainingSet/Yes/58.png  \n",
            "  inflating: TrainingSet/Yes/59.png  \n",
            "  inflating: TrainingSet/Yes/6.png   \n",
            "  inflating: TrainingSet/Yes/60.png  \n",
            "  inflating: TrainingSet/Yes/61.png  \n",
            "  inflating: TrainingSet/Yes/62.png  \n",
            "  inflating: TrainingSet/Yes/63.png  \n",
            "  inflating: TrainingSet/Yes/64.png  \n",
            "  inflating: TrainingSet/Yes/65.png  \n",
            "  inflating: TrainingSet/Yes/66.png  \n",
            "  inflating: TrainingSet/Yes/67.png  \n",
            "  inflating: TrainingSet/Yes/68.png  \n",
            "  inflating: TrainingSet/Yes/69.png  \n",
            "  inflating: TrainingSet/Yes/7.png   \n",
            "  inflating: TrainingSet/Yes/70.png  \n",
            "  inflating: TrainingSet/Yes/71.png  \n",
            "  inflating: TrainingSet/Yes/72.png  \n",
            "  inflating: TrainingSet/Yes/73.png  \n",
            "  inflating: TrainingSet/Yes/74.png  \n",
            "  inflating: TrainingSet/Yes/75.png  \n",
            "  inflating: TrainingSet/Yes/76.png  \n",
            "  inflating: TrainingSet/Yes/77.png  \n",
            "  inflating: TrainingSet/Yes/78.png  \n",
            "  inflating: TrainingSet/Yes/79.png  \n",
            "  inflating: TrainingSet/Yes/8.png   \n",
            "  inflating: TrainingSet/Yes/80.png  \n",
            "  inflating: TrainingSet/Yes/81.png  \n",
            "  inflating: TrainingSet/Yes/82.png  \n",
            "  inflating: TrainingSet/Yes/83.png  \n",
            "  inflating: TrainingSet/Yes/84.png  \n",
            "  inflating: TrainingSet/Yes/85.png  \n",
            "  inflating: TrainingSet/Yes/86.png  \n",
            "  inflating: TrainingSet/Yes/87.png  \n",
            "  inflating: TrainingSet/Yes/88.png  \n",
            "  inflating: TrainingSet/Yes/89.png  \n",
            "  inflating: TrainingSet/Yes/9.png   \n",
            "  inflating: TrainingSet/Yes/90.png  \n",
            "  inflating: TrainingSet/Yes/91.png  \n",
            "  inflating: TrainingSet/Yes/92.png  \n",
            "  inflating: TrainingSet/Yes/93.png  \n",
            "  inflating: TrainingSet/Yes/94.png  \n",
            "  inflating: TrainingSet/Yes/95.png  \n",
            "  inflating: TrainingSet/Yes/96.png  \n",
            "  inflating: TrainingSet/Yes/97.png  \n",
            "  inflating: TrainingSet/Yes/98.png  \n",
            "  inflating: TrainingSet/Yes/99.png  \n",
            "   creating: ValidationSet/I feel pain/\n",
            "  inflating: ValidationSet/I feel pain/1.png  \n",
            "  inflating: ValidationSet/I feel pain/10.png  \n",
            "  inflating: ValidationSet/I feel pain/11.png  \n",
            "  inflating: ValidationSet/I feel pain/12.png  \n",
            "  inflating: ValidationSet/I feel pain/13.png  \n",
            "  inflating: ValidationSet/I feel pain/14.png  \n",
            "  inflating: ValidationSet/I feel pain/15.png  \n",
            "  inflating: ValidationSet/I feel pain/16.png  \n",
            "  inflating: ValidationSet/I feel pain/17.png  \n",
            "  inflating: ValidationSet/I feel pain/18.png  \n",
            "  inflating: ValidationSet/I feel pain/19.png  \n",
            "  inflating: ValidationSet/I feel pain/2.png  \n",
            "  inflating: ValidationSet/I feel pain/20.png  \n",
            "  inflating: ValidationSet/I feel pain/21.png  \n",
            "  inflating: ValidationSet/I feel pain/22.png  \n",
            "  inflating: ValidationSet/I feel pain/23.png  \n",
            "  inflating: ValidationSet/I feel pain/24.png  \n",
            "  inflating: ValidationSet/I feel pain/25.png  \n",
            "  inflating: ValidationSet/I feel pain/26.png  \n",
            "  inflating: ValidationSet/I feel pain/27.png  \n",
            "  inflating: ValidationSet/I feel pain/28.png  \n",
            "  inflating: ValidationSet/I feel pain/29.png  \n",
            "  inflating: ValidationSet/I feel pain/3.png  \n",
            "  inflating: ValidationSet/I feel pain/30.png  \n",
            "  inflating: ValidationSet/I feel pain/31.png  \n",
            "  inflating: ValidationSet/I feel pain/32.png  \n",
            "  inflating: ValidationSet/I feel pain/33.png  \n",
            "  inflating: ValidationSet/I feel pain/34.png  \n",
            "  inflating: ValidationSet/I feel pain/35.png  \n",
            "  inflating: ValidationSet/I feel pain/36.png  \n",
            "  inflating: ValidationSet/I feel pain/37.png  \n",
            "  inflating: ValidationSet/I feel pain/38.png  \n",
            "  inflating: ValidationSet/I feel pain/39.png  \n",
            "  inflating: ValidationSet/I feel pain/4.png  \n",
            "  inflating: ValidationSet/I feel pain/40.png  \n",
            "  inflating: ValidationSet/I feel pain/41.png  \n",
            "  inflating: ValidationSet/I feel pain/42.png  \n",
            "  inflating: ValidationSet/I feel pain/43.png  \n",
            "  inflating: ValidationSet/I feel pain/44.png  \n",
            "  inflating: ValidationSet/I feel pain/5.png  \n",
            "  inflating: ValidationSet/I feel pain/6.png  \n",
            "  inflating: ValidationSet/I feel pain/7.png  \n",
            "  inflating: ValidationSet/I feel pain/8.png  \n",
            "  inflating: ValidationSet/I feel pain/9.png  \n",
            "   creating: ValidationSet/I need help/\n",
            "  inflating: ValidationSet/I need help/1.png  \n",
            "  inflating: ValidationSet/I need help/10.png  \n",
            "  inflating: ValidationSet/I need help/11.png  \n",
            "  inflating: ValidationSet/I need help/12.png  \n",
            "  inflating: ValidationSet/I need help/13.png  \n",
            "  inflating: ValidationSet/I need help/14.png  \n",
            "  inflating: ValidationSet/I need help/15.png  \n",
            "  inflating: ValidationSet/I need help/16.png  \n",
            "  inflating: ValidationSet/I need help/17.png  \n",
            "  inflating: ValidationSet/I need help/18.png  \n",
            "  inflating: ValidationSet/I need help/19.png  \n",
            "  inflating: ValidationSet/I need help/2.png  \n",
            "  inflating: ValidationSet/I need help/20.png  \n",
            "  inflating: ValidationSet/I need help/21.png  \n",
            "  inflating: ValidationSet/I need help/22.png  \n",
            "  inflating: ValidationSet/I need help/23.png  \n",
            "  inflating: ValidationSet/I need help/24.png  \n",
            "  inflating: ValidationSet/I need help/25.png  \n",
            "  inflating: ValidationSet/I need help/26.png  \n",
            "  inflating: ValidationSet/I need help/27.png  \n",
            "  inflating: ValidationSet/I need help/28.png  \n",
            "  inflating: ValidationSet/I need help/29.png  \n",
            "  inflating: ValidationSet/I need help/3.png  \n",
            "  inflating: ValidationSet/I need help/30.png  \n",
            "  inflating: ValidationSet/I need help/31.png  \n",
            "  inflating: ValidationSet/I need help/32.png  \n",
            "  inflating: ValidationSet/I need help/33.png  \n",
            "  inflating: ValidationSet/I need help/34.png  \n",
            "  inflating: ValidationSet/I need help/35.png  \n",
            "  inflating: ValidationSet/I need help/36.png  \n",
            "  inflating: ValidationSet/I need help/37.png  \n",
            "  inflating: ValidationSet/I need help/38.png  \n",
            "  inflating: ValidationSet/I need help/39.png  \n",
            "  inflating: ValidationSet/I need help/4.png  \n",
            "  inflating: ValidationSet/I need help/40.png  \n",
            "  inflating: ValidationSet/I need help/41.png  \n",
            "  inflating: ValidationSet/I need help/42.png  \n",
            "  inflating: ValidationSet/I need help/43.png  \n",
            "  inflating: ValidationSet/I need help/44.png  \n",
            "  inflating: ValidationSet/I need help/45.png  \n",
            "  inflating: ValidationSet/I need help/46.png  \n",
            "  inflating: ValidationSet/I need help/47.png  \n",
            "  inflating: ValidationSet/I need help/48.png  \n",
            "  inflating: ValidationSet/I need help/49.png  \n",
            "  inflating: ValidationSet/I need help/5.png  \n",
            "  inflating: ValidationSet/I need help/50.png  \n",
            "  inflating: ValidationSet/I need help/51.png  \n",
            "  inflating: ValidationSet/I need help/6.png  \n",
            "  inflating: ValidationSet/I need help/7.png  \n",
            "  inflating: ValidationSet/I need help/8.png  \n",
            "  inflating: ValidationSet/I need help/9.png  \n",
            "   creating: ValidationSet/I need water/\n",
            "  inflating: ValidationSet/I need water/1.png  \n",
            "  inflating: ValidationSet/I need water/10.png  \n",
            "  inflating: ValidationSet/I need water/11.png  \n",
            "  inflating: ValidationSet/I need water/12.png  \n",
            "  inflating: ValidationSet/I need water/13.png  \n",
            "  inflating: ValidationSet/I need water/14.png  \n",
            "  inflating: ValidationSet/I need water/15.png  \n",
            "  inflating: ValidationSet/I need water/16.png  \n",
            "  inflating: ValidationSet/I need water/17.png  \n",
            "  inflating: ValidationSet/I need water/18.png  \n",
            "  inflating: ValidationSet/I need water/19.png  \n",
            "  inflating: ValidationSet/I need water/2.png  \n",
            "  inflating: ValidationSet/I need water/20.png  \n",
            "  inflating: ValidationSet/I need water/21.png  \n",
            "  inflating: ValidationSet/I need water/22.png  \n",
            "  inflating: ValidationSet/I need water/23.png  \n",
            "  inflating: ValidationSet/I need water/24.png  \n",
            "  inflating: ValidationSet/I need water/25.png  \n",
            "  inflating: ValidationSet/I need water/26.png  \n",
            "  inflating: ValidationSet/I need water/27.png  \n",
            "  inflating: ValidationSet/I need water/28.png  \n",
            "  inflating: ValidationSet/I need water/29.png  \n",
            "  inflating: ValidationSet/I need water/3.png  \n",
            "  inflating: ValidationSet/I need water/30.png  \n",
            "  inflating: ValidationSet/I need water/31.png  \n",
            "  inflating: ValidationSet/I need water/32.png  \n",
            "  inflating: ValidationSet/I need water/33.png  \n",
            "  inflating: ValidationSet/I need water/34.png  \n",
            "  inflating: ValidationSet/I need water/35.png  \n",
            "  inflating: ValidationSet/I need water/36.png  \n",
            "  inflating: ValidationSet/I need water/37.png  \n",
            "  inflating: ValidationSet/I need water/38.png  \n",
            "  inflating: ValidationSet/I need water/39.png  \n",
            "  inflating: ValidationSet/I need water/4.png  \n",
            "  inflating: ValidationSet/I need water/40.png  \n",
            "  inflating: ValidationSet/I need water/41.png  \n",
            "  inflating: ValidationSet/I need water/42.png  \n",
            "  inflating: ValidationSet/I need water/43.png  \n",
            "  inflating: ValidationSet/I need water/44.png  \n",
            "  inflating: ValidationSet/I need water/45.png  \n",
            "  inflating: ValidationSet/I need water/46.png  \n",
            "  inflating: ValidationSet/I need water/47.png  \n",
            "  inflating: ValidationSet/I need water/48.png  \n",
            "  inflating: ValidationSet/I need water/5.png  \n",
            "  inflating: ValidationSet/I need water/6.png  \n",
            "  inflating: ValidationSet/I need water/7.png  \n",
            "  inflating: ValidationSet/I need water/8.png  \n",
            "  inflating: ValidationSet/I need water/9.png  \n",
            "   creating: ValidationSet/No/\n",
            "  inflating: ValidationSet/No/1.png  \n",
            "  inflating: ValidationSet/No/10.png  \n",
            "  inflating: ValidationSet/No/11.png  \n",
            "  inflating: ValidationSet/No/12.png  \n",
            "  inflating: ValidationSet/No/13.png  \n",
            "  inflating: ValidationSet/No/14.png  \n",
            "  inflating: ValidationSet/No/15.png  \n",
            "  inflating: ValidationSet/No/16.png  \n",
            "  inflating: ValidationSet/No/17.png  \n",
            "  inflating: ValidationSet/No/18.png  \n",
            "  inflating: ValidationSet/No/19.png  \n",
            "  inflating: ValidationSet/No/2.png  \n",
            "  inflating: ValidationSet/No/20.png  \n",
            "  inflating: ValidationSet/No/21.png  \n",
            "  inflating: ValidationSet/No/22.png  \n",
            "  inflating: ValidationSet/No/23.png  \n",
            "  inflating: ValidationSet/No/24.png  \n",
            "  inflating: ValidationSet/No/25.png  \n",
            "  inflating: ValidationSet/No/26.png  \n",
            "  inflating: ValidationSet/No/27.png  \n",
            "  inflating: ValidationSet/No/28.png  \n",
            "  inflating: ValidationSet/No/29.png  \n",
            "  inflating: ValidationSet/No/3.png  \n",
            "  inflating: ValidationSet/No/30.png  \n",
            "  inflating: ValidationSet/No/31.png  \n",
            "  inflating: ValidationSet/No/32.png  \n",
            "  inflating: ValidationSet/No/33.png  \n",
            "  inflating: ValidationSet/No/34.png  \n",
            "  inflating: ValidationSet/No/35.png  \n",
            "  inflating: ValidationSet/No/36.png  \n",
            "  inflating: ValidationSet/No/37.png  \n",
            "  inflating: ValidationSet/No/38.png  \n",
            "  inflating: ValidationSet/No/39.png  \n",
            "  inflating: ValidationSet/No/4.png  \n",
            "  inflating: ValidationSet/No/40.png  \n",
            "  inflating: ValidationSet/No/41.png  \n",
            "  inflating: ValidationSet/No/42.png  \n",
            "  inflating: ValidationSet/No/43.png  \n",
            "  inflating: ValidationSet/No/44.png  \n",
            "  inflating: ValidationSet/No/45.png  \n",
            "  inflating: ValidationSet/No/46.png  \n",
            "  inflating: ValidationSet/No/47.png  \n",
            "  inflating: ValidationSet/No/48.png  \n",
            "  inflating: ValidationSet/No/49.png  \n",
            "  inflating: ValidationSet/No/5.png  \n",
            "  inflating: ValidationSet/No/50.png  \n",
            "  inflating: ValidationSet/No/6.png  \n",
            "  inflating: ValidationSet/No/7.png  \n",
            "  inflating: ValidationSet/No/8.png  \n",
            "  inflating: ValidationSet/No/9.png  \n",
            "   creating: ValidationSet/Yes/\n",
            "  inflating: ValidationSet/Yes/1.png  \n",
            "  inflating: ValidationSet/Yes/10.png  \n",
            "  inflating: ValidationSet/Yes/11.png  \n",
            "  inflating: ValidationSet/Yes/12.png  \n",
            "  inflating: ValidationSet/Yes/13.png  \n",
            "  inflating: ValidationSet/Yes/14.png  \n",
            "  inflating: ValidationSet/Yes/15.png  \n",
            "  inflating: ValidationSet/Yes/16.png  \n",
            "  inflating: ValidationSet/Yes/17.png  \n",
            "  inflating: ValidationSet/Yes/18.png  \n",
            "  inflating: ValidationSet/Yes/19.png  \n",
            "  inflating: ValidationSet/Yes/2.png  \n",
            "  inflating: ValidationSet/Yes/20.png  \n",
            "  inflating: ValidationSet/Yes/21.png  \n",
            "  inflating: ValidationSet/Yes/22.png  \n",
            "  inflating: ValidationSet/Yes/23.png  \n",
            "  inflating: ValidationSet/Yes/24.png  \n",
            "  inflating: ValidationSet/Yes/25.png  \n",
            "  inflating: ValidationSet/Yes/26.png  \n",
            "  inflating: ValidationSet/Yes/27.png  \n",
            "  inflating: ValidationSet/Yes/28.png  \n",
            "  inflating: ValidationSet/Yes/29.png  \n",
            "  inflating: ValidationSet/Yes/3.png  \n",
            "  inflating: ValidationSet/Yes/30.png  \n",
            "  inflating: ValidationSet/Yes/31.png  \n",
            "  inflating: ValidationSet/Yes/32.png  \n",
            "  inflating: ValidationSet/Yes/33.png  \n",
            "  inflating: ValidationSet/Yes/34.png  \n",
            "  inflating: ValidationSet/Yes/35.png  \n",
            "  inflating: ValidationSet/Yes/36.png  \n",
            "  inflating: ValidationSet/Yes/37.png  \n",
            "  inflating: ValidationSet/Yes/38.png  \n",
            "  inflating: ValidationSet/Yes/39.png  \n",
            "  inflating: ValidationSet/Yes/4.png  \n",
            "  inflating: ValidationSet/Yes/40.png  \n",
            "  inflating: ValidationSet/Yes/42.png  \n",
            "  inflating: ValidationSet/Yes/43.png  \n",
            "  inflating: ValidationSet/Yes/44.png  \n",
            "  inflating: ValidationSet/Yes/45.png  \n",
            "  inflating: ValidationSet/Yes/46.png  \n",
            "  inflating: ValidationSet/Yes/47.png  \n",
            "  inflating: ValidationSet/Yes/48.png  \n",
            "  inflating: ValidationSet/Yes/49.png  \n",
            "  inflating: ValidationSet/Yes/5.png  \n",
            "  inflating: ValidationSet/Yes/50.png  \n",
            "  inflating: ValidationSet/Yes/6.png  \n",
            "  inflating: ValidationSet/Yes/7.png  \n",
            "  inflating: ValidationSet/Yes/8.png  \n",
            "  inflating: ValidationSet/Yes/9.png  \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LIq1pvW0Sk4E",
        "colab_type": "text"
      },
      "source": [
        "#Image Processing "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GnloOXrW0y1q",
        "colab_type": "code",
        "outputId": "0bc20a32-cc3a-4116-af14-b10bb59e782e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "from __future__ import print_function\n",
        "import keras\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "import os\n",
        "\n",
        "num_classes=5\n",
        "img_rows, img_cols=300,300\n",
        "batch_size=10\n",
        "\n",
        "training_dir='/content/TrainingSet'\n",
        "validation_dir='/content/ValidationSet'\n",
        "\n",
        "train_datagen=ImageDataGenerator(\n",
        "\n",
        "    rescale=1./255,\n",
        "    rotation_range=30,\n",
        "    shear_range=0.3,\n",
        "    zoom_range=0.3,\n",
        "    width_shift_range=0.4,\n",
        "    height_shift_range=0.4,\n",
        "    horizontal_flip=True,\n",
        "    fill_mode='nearest')\n",
        "\n",
        "valid_datagen=ImageDataGenerator(rescale=1./255)\n",
        "\n",
        "train_gen=train_datagen.flow_from_directory(\n",
        "\n",
        "    training_dir,\n",
        "    color_mode='grayscale',\n",
        "    target_size=(img_rows,img_cols),\n",
        "    batch_size=batch_size,\n",
        "    class_mode='categorical',\n",
        "    shuffle=True)\n",
        "\n",
        "valid_generator=valid_datagen.flow_from_directory(\n",
        "\n",
        "\n",
        "    validation_dir,\n",
        "    color_mode='grayscale',\n",
        "    target_size=(img_rows,img_cols),\n",
        "    batch_size=batch_size,\n",
        "    class_mode='categorical',\n",
        "    shuffle=True\n",
        ")\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 611 images belonging to 5 classes.\n",
            "Found 242 images belonging to 5 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_S6MckHPS5W7",
        "colab_type": "text"
      },
      "source": [
        "#Develop Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gq-G2By73AQy",
        "colab_type": "code",
        "outputId": "532a9769-a134-4a6e-be18-39e8c412941c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers.normalization import BatchNormalization\n",
        "from keras.layers.convolutional import Conv2D, MaxPooling2D\n",
        "from keras.layers.advanced_activations import ELU\n",
        "from keras.layers.core import Activation, Flatten, Dropout, Dense\n",
        "\n",
        "model=Sequential()\n",
        "\n",
        "model.add(Conv2D(32,(3,3),padding='same', kernel_initializer=\"he_normal\", input_shape=(300,300,1)))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Conv2D(32,(3,3),padding='same', kernel_initializer=\"he_normal\", input_shape=(300,300,1)))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Conv2D(32,(3,3),padding='same', kernel_initializer=\"he_normal\", input_shape=(300,300,1)))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Conv2D(64,(3,3),padding='same', kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Conv2D(64,(3,3),padding='same', kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Conv2D(128,(3,3),padding='same', kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Conv2D(128,(3,3),padding='same', kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Conv2D(256,(3,3),padding='same', kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Conv2D(256,(3,3),padding='same', kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dense(64, kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Dropout(0.5))\n",
        "\n",
        "model.add(Dense(64, kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('elu'))\n",
        "model.add(BatchNormalization())\n",
        "model.add(Dropout(0.5))\n",
        "\n",
        "model.add(Dense(num_classes, kernel_initializer=\"he_normal\"))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "print(model.summary())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_10 (Conv2D)           (None, 300, 300, 32)      320       \n",
            "_________________________________________________________________\n",
            "activation_12 (Activation)   (None, 300, 300, 32)      0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_12 (Batc (None, 300, 300, 32)      128       \n",
            "_________________________________________________________________\n",
            "conv2d_11 (Conv2D)           (None, 300, 300, 32)      9248      \n",
            "_________________________________________________________________\n",
            "activation_13 (Activation)   (None, 300, 300, 32)      0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_13 (Batc (None, 300, 300, 32)      128       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_6 (MaxPooling2 (None, 150, 150, 32)      0         \n",
            "_________________________________________________________________\n",
            "dropout_8 (Dropout)          (None, 150, 150, 32)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_12 (Conv2D)           (None, 150, 150, 32)      9248      \n",
            "_________________________________________________________________\n",
            "activation_14 (Activation)   (None, 150, 150, 32)      0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_14 (Batc (None, 150, 150, 32)      128       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_7 (MaxPooling2 (None, 75, 75, 32)        0         \n",
            "_________________________________________________________________\n",
            "dropout_9 (Dropout)          (None, 75, 75, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_13 (Conv2D)           (None, 75, 75, 64)        18496     \n",
            "_________________________________________________________________\n",
            "activation_15 (Activation)   (None, 75, 75, 64)        0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_15 (Batc (None, 75, 75, 64)        256       \n",
            "_________________________________________________________________\n",
            "conv2d_14 (Conv2D)           (None, 75, 75, 64)        36928     \n",
            "_________________________________________________________________\n",
            "activation_16 (Activation)   (None, 75, 75, 64)        0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_16 (Batc (None, 75, 75, 64)        256       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_8 (MaxPooling2 (None, 37, 37, 64)        0         \n",
            "_________________________________________________________________\n",
            "dropout_10 (Dropout)         (None, 37, 37, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_15 (Conv2D)           (None, 37, 37, 128)       73856     \n",
            "_________________________________________________________________\n",
            "activation_17 (Activation)   (None, 37, 37, 128)       0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_17 (Batc (None, 37, 37, 128)       512       \n",
            "_________________________________________________________________\n",
            "conv2d_16 (Conv2D)           (None, 37, 37, 128)       147584    \n",
            "_________________________________________________________________\n",
            "activation_18 (Activation)   (None, 37, 37, 128)       0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_18 (Batc (None, 37, 37, 128)       512       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_9 (MaxPooling2 (None, 18, 18, 128)       0         \n",
            "_________________________________________________________________\n",
            "dropout_11 (Dropout)         (None, 18, 18, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv2d_17 (Conv2D)           (None, 18, 18, 256)       295168    \n",
            "_________________________________________________________________\n",
            "activation_19 (Activation)   (None, 18, 18, 256)       0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_19 (Batc (None, 18, 18, 256)       1024      \n",
            "_________________________________________________________________\n",
            "conv2d_18 (Conv2D)           (None, 18, 18, 256)       590080    \n",
            "_________________________________________________________________\n",
            "activation_20 (Activation)   (None, 18, 18, 256)       0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_20 (Batc (None, 18, 18, 256)       1024      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_10 (MaxPooling (None, 9, 9, 256)         0         \n",
            "_________________________________________________________________\n",
            "dropout_12 (Dropout)         (None, 9, 9, 256)         0         \n",
            "_________________________________________________________________\n",
            "flatten_2 (Flatten)          (None, 20736)             0         \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 64)                1327168   \n",
            "_________________________________________________________________\n",
            "activation_21 (Activation)   (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_21 (Batc (None, 64)                256       \n",
            "_________________________________________________________________\n",
            "dropout_13 (Dropout)         (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 64)                4160      \n",
            "_________________________________________________________________\n",
            "activation_22 (Activation)   (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_22 (Batc (None, 64)                256       \n",
            "_________________________________________________________________\n",
            "dropout_14 (Dropout)         (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense_5 (Dense)              (None, 5)                 325       \n",
            "_________________________________________________________________\n",
            "activation_23 (Activation)   (None, 5)                 0         \n",
            "=================================================================\n",
            "Total params: 2,517,061\n",
            "Trainable params: 2,514,821\n",
            "Non-trainable params: 2,240\n",
            "_________________________________________________________________\n",
            "None\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "03J6pyIDTK1d",
        "colab_type": "text"
      },
      "source": [
        "#Train Our Model On the Data Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8YOzmuK73MzI",
        "colab_type": "code",
        "outputId": "36bf071d-254b-45aa-f407-d3d24477bb1b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from tensorflow.keras.optimizers import RMSprop\n",
        "import tensorflow as tf\n",
        "\n",
        "DESIRED_ACCURACY = 0.98\n",
        "\n",
        "class myCallback(tf.keras.callbacks.Callback):\n",
        "  def on_epoch_end(self, epoch, logs={}):\n",
        "    if(logs.get('acc')>DESIRED_ACCURACY):\n",
        "      print(\"\\nReached 98% accuracy so cancelling training!\")\n",
        "      self.model.stop_training = True\n",
        "\n",
        "callbacks = myCallback()\n",
        "\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam',\n",
        "              metrics=['acc'])\n",
        "\n",
        "no_trainSamples=611\n",
        "no_validSamples=242\n",
        "no_epoch=80\n",
        "\n",
        "history=model.fit_generator(\n",
        "\n",
        "    train_gen,\n",
        "    steps_per_epoch=no_trainSamples//batch_size,\n",
        "    epochs=no_epoch,\n",
        "    validation_data=valid_generator,\n",
        "    validation_steps=no_validSamples//batch_size,\n",
        "    verbose=1,\n",
        "    callbacks=[callbacks]\n",
        "\n",
        "    \n",
        ")\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/80\n",
            "61/61 [==============================] - 16s 261ms/step - loss: 0.4574 - acc: 0.8445 - val_loss: 3.9306 - val_acc: 0.3190\n",
            "Epoch 2/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3989 - acc: 0.8836 - val_loss: 2.9980 - val_acc: 0.5776\n",
            "Epoch 3/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3606 - acc: 0.8871 - val_loss: 0.6814 - val_acc: 0.7974\n",
            "Epoch 4/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3565 - acc: 0.8805 - val_loss: 1.6770 - val_acc: 0.6853\n",
            "Epoch 5/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3859 - acc: 0.8527 - val_loss: 10.3494 - val_acc: 0.3319\n",
            "Epoch 6/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.4277 - acc: 0.8674 - val_loss: 1.1952 - val_acc: 0.5733\n",
            "Epoch 7/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3773 - acc: 0.8773 - val_loss: 1.0899 - val_acc: 0.5333\n",
            "Epoch 8/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.3608 - acc: 0.9033 - val_loss: 0.4856 - val_acc: 0.7845\n",
            "Epoch 9/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3407 - acc: 0.8969 - val_loss: 7.4309 - val_acc: 0.3362\n",
            "Epoch 10/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.3388 - acc: 0.8822 - val_loss: 0.1486 - val_acc: 0.9784\n",
            "Epoch 11/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3479 - acc: 0.8773 - val_loss: 2.4305 - val_acc: 0.4483\n",
            "Epoch 12/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3547 - acc: 0.8871 - val_loss: 1.5378 - val_acc: 0.7112\n",
            "Epoch 13/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3353 - acc: 0.8937 - val_loss: 1.2504 - val_acc: 0.6767\n",
            "Epoch 14/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3210 - acc: 0.9049 - val_loss: 8.3223 - val_acc: 0.3276\n",
            "Epoch 15/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.3378 - acc: 0.8838 - val_loss: 1.6295 - val_acc: 0.4224\n",
            "Epoch 16/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.2539 - acc: 0.9166 - val_loss: 1.2488 - val_acc: 0.5474\n",
            "Epoch 17/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3324 - acc: 0.8887 - val_loss: 2.3612 - val_acc: 0.5043\n",
            "Epoch 18/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.2841 - acc: 0.9213 - val_loss: 1.0781 - val_acc: 0.6379\n",
            "Epoch 19/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.2528 - acc: 0.9229 - val_loss: 4.3914 - val_acc: 0.3750\n",
            "Epoch 20/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.3175 - acc: 0.9035 - val_loss: 3.8627 - val_acc: 0.3966\n",
            "Epoch 21/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3156 - acc: 0.9049 - val_loss: 1.1839 - val_acc: 0.5991\n",
            "Epoch 22/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.3349 - acc: 0.8871 - val_loss: 0.1812 - val_acc: 0.9397\n",
            "Epoch 23/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2229 - acc: 0.9199 - val_loss: 2.3513 - val_acc: 0.4397\n",
            "Epoch 24/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.3153 - acc: 0.9035 - val_loss: 5.2664 - val_acc: 0.3793\n",
            "Epoch 25/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.3089 - acc: 0.8937 - val_loss: 5.1120 - val_acc: 0.3922\n",
            "Epoch 26/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2454 - acc: 0.9344 - val_loss: 0.1737 - val_acc: 0.9483\n",
            "Epoch 27/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2932 - acc: 0.9002 - val_loss: 4.5493 - val_acc: 0.3966\n",
            "Epoch 28/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.3497 - acc: 0.8822 - val_loss: 1.2277 - val_acc: 0.5000\n",
            "Epoch 29/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2298 - acc: 0.9199 - val_loss: 0.1358 - val_acc: 0.9612\n",
            "Epoch 30/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2585 - acc: 0.9183 - val_loss: 0.2841 - val_acc: 0.8879\n",
            "Epoch 31/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2860 - acc: 0.9035 - val_loss: 5.9743 - val_acc: 0.4009\n",
            "Epoch 32/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2446 - acc: 0.9262 - val_loss: 0.1606 - val_acc: 0.9458\n",
            "Epoch 33/80\n",
            "61/61 [==============================] - 12s 205ms/step - loss: 0.2566 - acc: 0.9347 - val_loss: 0.0654 - val_acc: 0.9784\n",
            "Epoch 34/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2376 - acc: 0.9213 - val_loss: 3.4391 - val_acc: 0.3966\n",
            "Epoch 35/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2793 - acc: 0.9166 - val_loss: 1.8778 - val_acc: 0.5431\n",
            "Epoch 36/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2660 - acc: 0.9117 - val_loss: 0.1479 - val_acc: 0.9698\n",
            "Epoch 37/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2368 - acc: 0.9166 - val_loss: 2.8087 - val_acc: 0.4267\n",
            "Epoch 38/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2476 - acc: 0.9265 - val_loss: 0.9691 - val_acc: 0.7586\n",
            "Epoch 39/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2773 - acc: 0.9084 - val_loss: 8.9296 - val_acc: 0.4095\n",
            "Epoch 40/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2723 - acc: 0.9262 - val_loss: 0.3761 - val_acc: 0.8276\n",
            "Epoch 41/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.3670 - acc: 0.8822 - val_loss: 1.7997 - val_acc: 0.4353\n",
            "Epoch 42/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2987 - acc: 0.9115 - val_loss: 0.8754 - val_acc: 0.6983\n",
            "Epoch 43/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2545 - acc: 0.9068 - val_loss: 0.0486 - val_acc: 0.9871\n",
            "Epoch 44/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2825 - acc: 0.9164 - val_loss: 1.6762 - val_acc: 0.5043\n",
            "Epoch 45/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2435 - acc: 0.9314 - val_loss: 2.4120 - val_acc: 0.4095\n",
            "Epoch 46/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2397 - acc: 0.9344 - val_loss: 1.0496 - val_acc: 0.6034\n",
            "Epoch 47/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2731 - acc: 0.9117 - val_loss: 7.0570 - val_acc: 0.3448\n",
            "Epoch 48/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.1967 - acc: 0.9393 - val_loss: 0.7224 - val_acc: 0.5905\n",
            "Epoch 49/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2598 - acc: 0.9117 - val_loss: 0.1156 - val_acc: 0.9612\n",
            "Epoch 50/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2278 - acc: 0.9281 - val_loss: 0.1943 - val_acc: 0.9655\n",
            "Epoch 51/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2232 - acc: 0.9330 - val_loss: 6.4578 - val_acc: 0.3319\n",
            "Epoch 52/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2459 - acc: 0.9199 - val_loss: 1.1019 - val_acc: 0.6207\n",
            "Epoch 53/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2596 - acc: 0.9199 - val_loss: 9.1562 - val_acc: 0.4009\n",
            "Epoch 54/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2264 - acc: 0.9459 - val_loss: 0.6991 - val_acc: 0.7414\n",
            "Epoch 55/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2816 - acc: 0.9002 - val_loss: 1.2463 - val_acc: 0.6767\n",
            "Epoch 56/80\n",
            "61/61 [==============================] - 12s 205ms/step - loss: 0.1897 - acc: 0.9393 - val_loss: 8.0927 - val_acc: 0.4052\n",
            "Epoch 57/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2261 - acc: 0.9297 - val_loss: 4.8351 - val_acc: 0.4000\n",
            "Epoch 58/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2585 - acc: 0.9068 - val_loss: 0.0612 - val_acc: 0.9871\n",
            "Epoch 59/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2400 - acc: 0.9248 - val_loss: 6.8499 - val_acc: 0.3233\n",
            "Epoch 60/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.1989 - acc: 0.9265 - val_loss: 0.3128 - val_acc: 0.8922\n",
            "Epoch 61/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2645 - acc: 0.9183 - val_loss: 4.7972 - val_acc: 0.3405\n",
            "Epoch 62/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2362 - acc: 0.9265 - val_loss: 0.1759 - val_acc: 0.9483\n",
            "Epoch 63/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2231 - acc: 0.9265 - val_loss: 5.8542 - val_acc: 0.3190\n",
            "Epoch 64/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.1835 - acc: 0.9445 - val_loss: 1.6759 - val_acc: 0.4353\n",
            "Epoch 65/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.2286 - acc: 0.9330 - val_loss: 3.6155 - val_acc: 0.4052\n",
            "Epoch 66/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.1919 - acc: 0.9459 - val_loss: 4.5684 - val_acc: 0.3966\n",
            "Epoch 67/80\n",
            "61/61 [==============================] - 12s 204ms/step - loss: 0.2630 - acc: 0.9262 - val_loss: 0.4788 - val_acc: 0.8233\n",
            "Epoch 68/80\n",
            "61/61 [==============================] - 12s 203ms/step - loss: 0.1916 - acc: 0.9314 - val_loss: 0.1900 - val_acc: 0.9655\n",
            "Epoch 69/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.1660 - acc: 0.9492 - val_loss: 0.2555 - val_acc: 0.8405\n",
            "Epoch 70/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.1789 - acc: 0.9590 - val_loss: 1.2270 - val_acc: 0.5172\n",
            "Epoch 71/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.2202 - acc: 0.9281 - val_loss: 2.0839 - val_acc: 0.4871\n",
            "Epoch 72/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.1685 - acc: 0.9656 - val_loss: 1.9619 - val_acc: 0.4871\n",
            "Epoch 73/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.2917 - acc: 0.9133 - val_loss: 0.4474 - val_acc: 0.8060\n",
            "Epoch 74/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2344 - acc: 0.9347 - val_loss: 3.5812 - val_acc: 0.3664\n",
            "Epoch 75/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2562 - acc: 0.9215 - val_loss: 2.5120 - val_acc: 0.4483\n",
            "Epoch 76/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.1936 - acc: 0.9379 - val_loss: 0.8014 - val_acc: 0.6983\n",
            "Epoch 77/80\n",
            "61/61 [==============================] - 12s 202ms/step - loss: 0.2442 - acc: 0.9314 - val_loss: 3.3763 - val_acc: 0.3922\n",
            "Epoch 78/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.1852 - acc: 0.9314 - val_loss: 4.3845 - val_acc: 0.4095\n",
            "Epoch 79/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.3031 - acc: 0.9133 - val_loss: 1.3957 - val_acc: 0.6336\n",
            "Epoch 80/80\n",
            "61/61 [==============================] - 12s 201ms/step - loss: 0.2194 - acc: 0.9314 - val_loss: 0.1101 - val_acc: 0.9784\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CU--__pHTn4y",
        "colab_type": "text"
      },
      "source": [
        "#Get Our Class Labels"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-wJNDJ-X4O7b",
        "colab_type": "code",
        "outputId": "ecdb0a84-bbe1-4023-a804-22f73c89cd4d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "\n",
        "valid_generator=valid_datagen.flow_from_directory(\n",
        "\n",
        "\n",
        "    validation_dir,\n",
        "    color_mode='grayscale',\n",
        "    target_size=(img_rows,img_cols),\n",
        "    batch_size=batch_size,\n",
        "    class_mode='categorical',\n",
        "    shuffle=False)\n",
        "\n",
        "class_labels=valid_generator.class_indices\n",
        "class_labels={v: k for k, v in class_labels.items()}\n",
        "classes=list(class_labels.values())\n",
        "print(class_labels)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 242 images belonging to 5 classes.\n",
            "{0: 'I feel pain', 1: 'I need help', 2: 'I need water', 3: 'No', 4: 'Yes'}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cPOyGVjxT4zR",
        "colab_type": "text"
      },
      "source": [
        "#Test on Validation Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4gJh33o04dgk",
        "colab_type": "code",
        "outputId": "d910181c-f348-44f3-9491-5d53c14e704f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 978
        }
      },
      "source": [
        "from keras.models import load_model\n",
        "from keras.optimizers import RMSprop,SGD,Adam\n",
        "from keras.preprocessing import image\n",
        "import numpy as np\n",
        "import os\n",
        "import cv2\n",
        "from os import listdir\n",
        "from os.path import isfile,join\n",
        "import re\n",
        "from google.colab.patches import cv2_imshow\n",
        "\n",
        "def make_test(name,pred,im,true_label):\n",
        "  BLACK=[0,0,0]\n",
        "  expanded_image=cv2.copyMakeBorder(im,160,0,0,300,cv2.BORDER_CONSTANT,value=BLACK)\n",
        "  cv2.putText(expanded_image,\"predicted - \"+pred,(20,60),cv2.FONT_HERSHEY_SIMPLEX,1,(0,0,255),2)\n",
        "  cv2.putText(expanded_image,\"true - \"+true_label,(20,120),cv2.FONT_HERSHEY_SIMPLEX,1,(0,255,0),2)\n",
        "  cv2_imshow(expanded_image)\n",
        "  \n",
        "def getRandomImage(path,img_width,img_height):\n",
        "  folders=list(filter(lambda x:os.path.isdir(os.path.join(path,x)),os.listdir(path)))\n",
        "  random_dir=np.random.randint(0,len(folders))\n",
        "  path_class=folders[random_dir]\n",
        "  file_path=path+path_class\n",
        "  file_names=[f for f in listdir(file_path) if isfile(join(file_path,f))]\n",
        "  random_file_index=np.random.randint(0,len(file_names))\n",
        "  image_name=file_names[random_file_index]\n",
        "  final_path=file_path+\"/\"+image_name\n",
        "  return image.load_img(final_path,target_size=(img_width,img_height),grayscale=True),final_path,path_class\n",
        "\n",
        "img_width,img_height=300,300\n",
        "\n",
        "\n",
        "model.compile(loss='categorical_crossentropy',optimizer=RMSprop(lr=0.001),metrics=['accuracy'])\n",
        "\n",
        "files=[]\n",
        "predictions=[]\n",
        "true_labels=[]\n",
        "\n",
        "\n",
        "path='/content/ValidationSet/'\n",
        "img,final_path,true_label=getRandomImage(path,img_width,img_height)\n",
        "files.append(final_path)\n",
        "true_labels.append(true_label)\n",
        "x=image.img_to_array(img)\n",
        "x=x*1./255\n",
        "x=np.expand_dims(x,axis=0)\n",
        "images=np.vstack([x])\n",
        "classes=model.predict_classes(images,batch_size=batch_size)\n",
        "predictions.append(classes)\n",
        "  \n",
        "  \n",
        "\n",
        "image=cv2.imread((files[0]))\n",
        "image=cv2.resize(image,None,fx=3,fy=3, interpolation=cv2.INTER_CUBIC)\n",
        "make_test(\"Prediction \",class_labels[predictions[0][0]],image,true_labels[0])\n",
        "cv2.waitKey(0)\n",
        "  \n",
        "cv2.destroyAllWindows()\n",
        "predictedResult=class_labels[predictions[0][0]]\n",
        "print(predictedResult)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/keras_preprocessing/image/utils.py:104: UserWarning: grayscale is deprecated. Please use color_mode = \"grayscale\"\n",
            "  warnings.warn('grayscale is deprecated. Please use '\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABBoAAAOOCAIAAAAh7ubTAABRWUlEQVR4nO3d53JbZ5YobGQigzlK\nlG31dKxz7v8qTn1TNeNuT8uyEjNBgiByIL4fq7WHrWSJIsWg56kuFUUhbGwA7Xftd4VUCgAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD4DqRv+wDu\notnbH5wdeJ8vCACQyNz2Adyy2Sd/8/6/XuMTfRu39bzX6yZexezt/27RfXx3bugLAgDcU991ODG7\n9OfDeKI78rzX69pfxexurIkfxrsDAHznvutwgu/Q+8t3GTsAAFcmnHhX+iM/88Ck3/6PL+ILAgBc\nlrvtA7iLLJIeKjXE18LZAwASdicAAIArEk4AAABXdG3JTrNLKRCXq11/Ny9i9u+3+Zz7ftHjX+1e\ns8+7ze8+4Owjf/30g3+DF/i9edhvxJW/fe88yPt3/NgX4TN/7zMJAA/e9fwn/nOaXX5s8fH59/r0\njT9zQf859/30WvPzD+NLD/ibvcD75esLHu7CG/GJZ/nKd+crPwDX+AV58B9FAOB93y7Z6Qr99T9/\naf6xG1x7U/8vOoxPrJy+dAn7sRuYWvA57sIbcTfn5d3WLBQA4MG4kc5OHwsDPieD6HM2BD4zzLja\nvT7hMx/w8stM//u/fuaF7dt6gQ/Y9/NGfP637xscts8kADx41787kf7kX7/ovtf4+F9zVN/mAb/y\nkW/ueL5b9/GNuPKz38Rhf+wxxRUA8JBc8+7EB1ch6etLqv70439s9+NrjuoKh/E1S7E78gLfd73n\n6u67iTfic3bnvsaVPwY38fm5p+87APClvtEYu6utw1Jv7/LBnz/t2geWfc4DXu257sgL/Jznutp9\n78vK8ivfiC+617f0wW/fjX5+7ss7DgB8PXMn+H3Xvt8CAMDD8I12J76NB79yvcUX+ODP7Re52tlw\nDgGAh+c+hRMPfjX24F/gfSFaAAD4TJKdAACAK/pG4cTX1H1+Tn/J2ceb6F97B/0vOoybe2RtN6/d\nzb0RV/tgXJcPfvt8fgCAa3HN4cRNT27+9ON/0ejfr+xWdIXDuLlHNjD72t3EG3HT78iVPwY+PwDA\nlV3/7sTsk3+9rce/9qP6ogf8oivBd+QFch/fiCs/u88PAHA1N1KK/bG1yJVrVd+ZqPWZj/+Z97rp\nw3jHxzJP7sIL5EbfiG9Tq/35z+7zAwB8vW9Xiv2Va6nfvfuXTsf7lofxwd9/LAX/ix5ZN6Frd0Nv\nxO2+U5//sQQA+CLXs5x454r75YXy7z7BFaYmf9Hjf/peH3v2zzmqrzyMzxwEfkMv8H65xlfxRQ91\njW/E1x/M7z7dFb59H3yc1Fd8QX73FT2MDyQAkLiRcAL4Znz7AIBbZO4EAABwRcIJAADgioQTAADA\nFQknAACAKxJOAAAAt80MLLgtvn0AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAB8a+nbPgC4\nz2Zvf/BNAgC+S5nrfLDZ79+Er3UTJ3n29n+36D5+eGYf+RkA4LtxfeHE7NKf3JBrP8mzu7Em9uEB\nALifrnV3gvvl/eW7jB0AAL5E7rYPgDtAFHE1abUTAMD3TjjxvbIOvhbOHgDwfZPsBAAAXJFwAgAA\nuKLrSHaafeSv7+eBzP79l5fvmP74zT74jL+bZPKxB+fz36+P3evzz+e3fxcufzau/OwfvOPHPnWf\n+XsfSADgIfrqdc2nm3t+bDn1iRv/7ur20zf4/OO5j76+4OFLz8/VzueVn+Ur36DPaTX7iae4wt0/\nduRfeSQAAPfEVyc7fWJJ9EWrpWtZWv3uGs5kgy96v652Pu/4u/CxZ//2R+XTCADcf9eR7PSZuwof\nu+N1eWdx9kUbI9+VL90Fev9mnzifd+pd+Nizv5+e9A0O2wcSAHiIbq8U+0YzPdKf/Ctf6mrn83bf\nhSs/+00c9sceU1wBANxztzR34lvGEskvY+n2OWXcn3blJeA9jWqudj5v+l34tE8/+03c8UsfEwDg\nQbiNcOImVlezj/x8o090hfvel5Xl1c7nN3sXruyD8cyNTvS7L+84AMCVmDvxhb5mdWhlCQDAw3JL\nyU636OvX9KKCy652NpxDAIAH4cGFE9ap10u0AADAx0l2AgAArui+hRMfK/D9nM6bs7taH3wHXe18\n3v134YNV19q2AgBc1b0KJz5ztffpUc2WjF/qaufzdt+Fq43r/po7AgB8l64vnLiuS7wfe5wvetiv\nuS/vu9r5vN134crP7sMDAPDZbqYU+7oa+X/RSu6dWWO/mxbFp13tfN6pd+Hzn/0zDxsAgH93rclO\n76/SrrYsu/Ja83fvKJb4Ilc7n3f8XfjYs/tsAAB8uZtZQ31sd+JLx0JfjkbS//77332cj933vrvG\n0dpf9FBXO5+ff69reV3vfPCu/Bn44B0/doRf+vvPvwEAwJ1nOcMDcl1ZdgAAfJ571dkJAAC4S4QT\nAADAFQknAACAKxJOAAAAVyScAAAAgJQJdAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADvSt/2AQBwn6TT//oPx2w2u90jAeAuEE4A8LnS6XQm\nk4mI4uLi4uLi4raPCIBbJpwA4HdkMpl8Pl8oFAqFQi6Xy2Qys9lsOp1OJpPxeDwejyeTyXQ6ve3D\nBOAWCCcAeFcmk8lms/l8Pp/P53K5ubm5Uqk0Nzc3NzdXKBSScGI8Hg8Gg8FgMBwOh8PheDyeTqfT\n6fTi4mI2m81ms9i+kBYF8IDlbvsAALhzstns3NxctVqtVCqlUqlSqVSr1WKxWCwWC4VCNpu9uLiY\nTqej0ajf73e73fiz1+tFXDEajSIVajqdRlAhogB4qOxOAPC/0ul0sVis1+vz8/PLy8uLi4u1Wq1e\nr9dqtQ/uTvR6vV6v1+/32+322dlZt9vtdDq9Xm8ymYxGo9FoFDlRk8nkY3FFso+R7GncygsH4Grs\nTgDwL5lMJpfLNRqNra2tR48ePXnyZHNzM8KJarWaeytKsSOvafjW+fl5q9U6Ozs7OTk5Ozvr9/ud\nTqfb7SY3iCSoyyUWSUl3PFQUYFxOl7q1EwHAZxNOAPAvmUymWCwuLi5ub2//6U9/+vOf/7y9vZ2k\nPCW3SafTsda/uLiYTCax9O/1eq1W6/T09PDw8Pj4uN1un56etlqtXq8X2VCxRxGhQuptk6jYl0i2\nMt7Z0xBRANx9wgkAUqlUam5urtFoLC0tPX369G9/+9tf//rXv/3tb48fPy6Xy3Nzc7979+l02m63\nW63W4eHh0dHR2dlZs9k8OTm5nP6UZD3NZrMIJ+KOk8kkKi6S2u7RaBQV3knnqCQV6p0YI/4q8AC4\nLWonAEil0+mVlZVHjx5tb2//7W9/+7//9/8+ffr06dOn8/PzX/Q4o9EoNiU6nc7Z2dnZ2dlgMOj1\neoPBIElkej+ciDKMCBvi9lGM0W63O51O/JmEGUmVxY2cCAC+kN0JgO9FkqT0vrm5uYWFhe3t7T//\n+c9//etfnz59urKy8jmbEu/IZDLlcjmdTlcqlUajMRgMkjghqbdOvd1MiNqJiC6SSCPCiW6322w2\nj46Oms3mwcHBwcFBlHpH6cWXjs/7xAsH4CsJJwC+Fx9bUkcTp6dPn0aC09/+9rdHjx4Vi8V8Pv+l\njx+NoQqFwjvNmi4/9QcPI/llNJ/t9XpHR0f7+/uHh4evX79eXFxsNpvNZrPdbr8Tn0QCVZIQ9cFI\nQywBcHOEEwDfo7m5uXw+n81mFxYW1tbWNjY2/vSnP/3tb3/74x//+Pjx48XFxSs8Zjqdzmaz2Wz2\nK4/t4uJiOBzOz89HnLOwsLCysnJ6enpycnJ+fp7UYKRSqciSSkovouIixl9ExpRZ3QA3TTgB8N1J\np9Plcnl+fr5arW5sbGxvb29sbDx9+vSHH35YXl4ul8u3e3iZTKZQKNRqtdlsFgO5FxYWOp1OlHQn\nnWRTb8u4o+Ki3+8PBoNWq3V8fBwNpmJK9wefQvoTwHURTgB8XyqVSq1WW1tbW19fX15e3t7efvLk\nyfr6+tbW1sbGRr1ev0LJxLXLZrPlcjmXy1UqlXq9vra2NhqNotHT5XF4ycyKfr8fKVLHx8c7Ozv7\n+/vVarVQKJycnLTb7Q/OzruNlwXwAAknAB6sdDodg+citSlW541Go16vb2xsbG1tra2tPX78+NGj\nR4uLi0tLS41Go1Ao3PZR/0s2my2VSsVisVKpXJ5ZkXSGTafTSXlGMizv+Ph4ZWVld3e3Xq83Go2T\nk5MYqzcej4fDYYy/uO1XBvCgCCcAHqx0Oj03N1culyOEqFQqCwsLi4uLtVptdXV1c3NzZWVlfX19\nZWWlUqlUKpUvrb3+BqIeI+Kij1V1z2azZKJ2uVyOCCRSpM7Ozo6Ojg4PD1utVrPZjOSo23s1AA+Q\ncALgAcrlcvl8vlarzc/PNxqN1dXV1dXVqGleXFwsl8vx8/z8/MLCQq1Wi02M2z7qD8tkMjGh4nNU\nKpVyuVyv1+fn57e2tpL0p4ODg9evXxcKhWaz2Wq1bvJ4Ab4vd/Q/HgB8jhgGFwMc4ip+JpOJ8uVS\nqbS4uLi2tra8vLy1tbW5ubm0tLS8vBwZTaVSKfYryuXy+4v1SCW6jRf0tUqlUj6fL5VKtVqt3++P\nx+PT09PNzc3d3d1Go1EsFqMkIymoiIazt33UAPeYcALgHstkMrlcLgKJ/Fu1t1ZXV6NA4tGjR1tb\nWwsLC/Pz8zFmLpvNFgqFfD7/wQv/9zSWCJlMplQqZbPZRqORSqVi+2V+fr5QKMRZSqVSuVwuGstG\nMfdtHzLAPSacALhnIvknn88XCoVCoVAsFufm5nK5XKlUKpfLpVIpJjYkBRKrq6sbGxtra2u1Wq1S\nqdz24d+42J9J+lM1Go1arbawsFAsFovF4sLCQqPR2N3d7Xa7Z2dnp6enMaFCTQXA1QgnAO6TuL5e\nKBSq1WqtVkvKrOOHiCLm5+fr9XqpVGo0GktLS/Pz84uLi41G4/MrEB6Y6H4bzWejpdXu7u7p6enu\n7u6bN2+Ojo6Oj4/Pz88/OFEbgE8TTgDcJ9lsdm5urlKpLC0tRZCwurq6srLSaDSWl5fjh0ajERUR\nUUUQpRTfbSwR8vn8/Pz83NxcjAA/PDw8Ojp6/vx5bOwMh8PBYDAajdRRAHwp4QTAXZF+K9KZ3imz\njuZL1Wo1tiNWVlZWVlYWFhZiIF10alpdXa3X69Vq9bZfyp2TyWSq1Wqcma2trdPT06OjoyjOjg65\npVKp0+l0u91+v59KpZJheQIMgE8TTgDcFdFPKZfLRTnEO3URlUqlWCwmrV2Xl5cXFxer1WpMoIvy\ngIWFhc98lm/wcu6CD77YTCaztLRULBZTqVQul4tZHGtra8fHx4eHh81mMwZsDwaD2zhkgHtGOAFw\n49Lp9Odf5M7n85VKJbq41uv1pC4ioohY+1YqlVqtVq1W47J6uVy+XHz8uwfzFS/lnvnEi83n8xF9\nxSldXV3d2dn57bffYjbFZDIRTgB8DuEEwI373VgicpnK5XIECYuLiwsLC/V6fXFxMYZYLy4uxsiI\naEwUuxb5fD6uvmfeuri4iHSpb/O67q/ZbBadZKvV6ubm5traWvS/ilSx/f39SqVydHTU7/eHw+F4\nPI6sMxlQAO8TTgB8U4VCIXYSIiSI0upIbYpmTZVKJQZEVKvVSG2K30RqU+xa3PaLuPdi8kY2m42/\nxvZOpVIpFAq1Wu3w8HBzc/P4+DgJJ1KXNjouLi7G4/FkMun1ep1Op9fr9Xo9wyuA75ZwAuCafTq1\naW5ubn5+vtFoLC4uzs/Pl0qlKBGO5Wy1Wi0Wi9VqNSolkt8kP3xmRhNfJJvNRrOs6XQazWS3t7fb\n7Xa0exqNRqlUajabZTKZbDYbeVC9Xu/4+Hh3d/fw8HA8Hr8fTnxRhhvA/SWcALhmH1tERjn1ysrK\n+vr68vLyxsbG6upqrVaLTk3RzrVUKiXz6XK53Ps/JBfUuUbpdDom3M3NzS0tLUW00O/3x+NxbERE\njlPkpMW+xPn5+c7Ozq+//vry5ctsNnt0dJTsYwSxBPCdEE4A3JR0Op3EAJVKJYZVb25ubm1trays\nRMp+hBPVajVihnw+n30ryiGSH2771TxwEbCVSqV6vT6dTiOQmL4VE+4ymUyEE4PBoNvtrqysRPrZ\n3NxcuVxut9tnZ2fdblfiE/BdEU4A3JQYw1ytVkulUvQOWl5e3tra2traSv5aqVQirylKqLPZbDJ6\nIpVKXR49wTeQyWQKhcLFxUU+n7+4uIhNiYglkmSnCDYajUY+n89kMsViMZfLzc3NHR4e5nK5i4uL\nXq9nwDbw/RBOANyI6OW6srISHZnW19c3NzeXlpY2NjZiUyIqKIrFYqFQuO2D5V8iivvdjLKIMYrF\nYqlUiqL5er3+5s2bcrmczWabzWar1ZLsBHwnhBNw7yWjlOOvs7du96i+B0kO0uXZ1fFeRGrT4uJi\n7EVEatP6+np0eo0K7FKp9J3XVc9ms8gjSj6xl5ve3uU9mdhHmp+fj05QpVIpBguWSqV4CaPRqNvt\nphRkA98B4QTce0l6feptLBHZ3rd9XA9cpNHHmY9Sh3Q6nVRRz8/Px77E48ePf/jhh/X19Y2NjZjE\nHBRVp1KpyyUKkR0Utc75fD6fz9/lcCJRLBbjUHO5XKVSmc1m0Vu23W4Ph8MkS+q2DxPgBgkn4L7K\nZrOxeI2RBbGcnc1mk8lkNBoNh8PRaDQej+VwX69YOMZpT2bJJW9BDKiOSomVlZWlpaWtra3t7e21\ntbWVlZVarXYvlsjfwGQy6ff7/X6/0+l0u93hcDidTjOZTNItt1KplEqlux9xxUbK6upqsVis1Wqz\n2azX6w0Gg3a7PRqNos+s7yDwsAkn4D65nAoSC69k7VUoFDKZzHQ6HY1GnU7n/Py82+3GQi0qSm/7\n2B+CuHYe5zxmzNXr9Uh0icER8ftisdhoNObn52u12tLS0urqamQ33fbh3wlRddDr9U5OTprN5uHh\n4dHRUbvdnkwm+Xy+Vqutrq5G9BXh2W0f7+eq1+v5fL7f75+cnJyfn5+eng4Gg5OTk+l0OplMbvvo\nAG6QcALujUjXjjyQpPHo5Vz8GLDV7/dPT09PTk5arVbk8UfjfOlPXy+Xy5VKpUajsbS0tLCwEK2Z\n6vX64uJijKyOuKJQKERGU/xQLpcVWyciwandbh8cHOzs7Lx48eLVq1fNZnM0Gs3NzS0vL//www/T\n6TTCtnsUTqRSqTjmpaWltbW1jY2N09PT4XDY6XRu+7gAbpZwAm5c+pOSGyS3v5xpneTlR15TPp9P\nEmyq1WqMVY6lbbSUiQFbJycnx8fHcfU3crhD0kQ/rhBPL/kOq7cvn/PL70gUAcefEcLFdlAyvnp5\neXl1dXVxcXF9fX1tba3RaCwvLy8vL9dqtVqtFm+EMREfNBgMzs/Pz87O3rx58/z581evXv3666/P\nnz8/Pj4ej8eVSmVjY2MymUQiWbVaTaZw5HK5pFb7ne/LHRFtZIvFYr1eX15eXllZmZ+fPz099UkA\nHjzhBNy4WKlHSJCUmcYPuVzunTllSV+mWDNFXn7kNUVSfvwZ10EbjUatVqvX6/V6vVgsRrJT5G23\nWq2zs7NWq9VutweDwWAwiHBiPB5HTUWv12u320la1OVpvt+Jd8KneCOSjYVSqRRVtsk+QzQDrVar\nS0tLS0tL9Xp9aWkpNiViPl25XL6Dy9w74uLiYjKZnJycvHr16uXLl//4xz/++c9/7u/v7+zsvH79\nOpogVSqV8Xhcq9XirFar1ZgCEbF0xHhJ+ftdW6Ynm4dJ2Fkul+9LQTnA1xBOwDcSC6MkMEiWrfm3\n4nJ49GWKcCIudsbSpNFoNBqNyNqPcCIpBY4fcrlcUoo9GAz6/X4M7o3yiX6/H2XZ8U/9fr/ZbO7t\n7R0eHqZSqcFg8B2GE5dFgXU+n0+CtMXFxUajUS6XI1qLH+ItSColyuVyuVyO85/P57/Z0cbH45s9\n3bWINLxms/ns2bP//M///H//7/89f/683W5H1UTcJuLebrfbbrdPT0+Pj4+Tz3mxWLy8RxfL9Lt2\nEmLHL/re3vaxAHw7wgm4WdlsNi51RyZMUrYb+wwRV4RozRQrklgvxkzlWM4uLi4uLCwkE5Szb13O\nz4lnjESmeJzxeBwtnobD4WQymc1mSYyxv7//8uXLN2/exKOdnJy02+3bPVffUpy9WJvmcrkkMIs9\nh6iLWFpaqtVqCwsLCwsLsTVRq9WSIDBOe7wFcbE8yca56YO/a8vozxFNAs7Pzw8ODl68ePHrr7++\nfv36coeAfD6/srKyuLhYqVRSqVS32z06OhoOh/ENqtVqlUoldofiq3HXmj4lCYTR0Cl6OsWX7rYP\nDeBmCSf4LiSZRe+sv5PKhMv/lHq7MphMJlHBHGuCZMcgybV4Jz0pxLXJ2IuIDJlkJRSL0SQkiEFm\nyaSC93cnorNNPEKj0YgmQsVi8UtXk5PJJOkYGzsVvV4vrr6vrKzEpfc3b97s7u62Wq2H1wMqKVCJ\nzKXklEaR9NzcXMQVUZES8cP8/Pzy8nIsbaNHU7lcjm2la1nFjt+K+vjLK84kFIzr3Ol0OmlKG4UE\n9zGWSL39Tl3+YC8uLhYKhTiluVwuumAtLi7GEPHZbHZ8fHxwcNDv93O5XMwBfPz4cUSAd3B/Jja4\nstnsbDYbDoe9Xi+2BIUTwIMnnODhS6fTkZdSKpVi1Z5c2k8mBsSfUcmQSqUuLi7iEmMsvqPqIIKE\nmDnwfnpSkucQEUgEElGkm1zbjlzqSHNKnjr3VgQzl1ddEZNE1FEul5OL4l96BpKRFElKz9zcXKxT\no4Y7fhlH3u/3R6PRDbwPtybOWGSOLS0traysbGxsrK2tJTUnqbcdYKNxU2S9J+FfktEUgcfXH0+y\nR9Tr9Uaj0eVqmchVi0gjNpSy2WzUZsSnKNbiX38M3158nkul0tLS0uPHj6N6J85zlKYkaWaFQuHi\n4qLf75+dne3v7x8eHs5ms0ePHs1ms9gpussRbzabnU6nSWFSvL+3fVAAN0s4wUMW6+ZY1keuUST2\nxOo5m81GmBFLxtgoiLVaFDTHgq/T6fR6vbjKmJTqJpFAkp6UbGXEJee5ubkYZLa+vr60tBQXtufm\n5mJPI65iJuKXycbI5Q42lwu4r3xl+vIaNF5+XImv1+vr6+uNRqNUKlUqlXiuZrN5enra7/ev7324\nZfHyc7lcvV7f2tr68ccf//jHPz558mRhYSHWsnHO411INosifkhK568luybq4M/Pz1utVlTJ9/v9\niB7jAC4uLiKQiMvbw+Ewn88vLS1tbm5GgcG9652aiI99tVqN17K8vHxxcVGr1WLnJ7Zf4ofxeNxq\ntQ4PD8/Pz9vt9v7+/ng8zmazm5ubycX+u7Y1kXq7/dLtdlut1vHx8dHRUavVGgwGdzn4AbgWwgke\nmmTroFwux3y3uKKZlC7EyjJqFZJdi1ilxS9TqVRcpO92u51OJwknLi4u4i5Jq5/L1c+JJJyI+WVr\na2vz8/NJ+JG61OkyiSI+2Cg2uaZ+7R1skrKBWMUmfaJiDf3y5cs4A5PJ5GGshCJImJubW1hYePTo\n0R//+Mf/83/+z48//hhJaHNzc7GtdDn/7XIHoS99uvgMjEajSJ2P05hOpyeTyXA4HAwGZ2dnzWYz\nWm/1er13wokodIlott/vF4vFzc3NiCviU3dPw4nYGatUKmtra/l8fmNjIzZeIoUvvh3RUaDb7ZZK\npYuLi1arFQFG6tL2UbwpdzCcGI1G3W738PBwd3d3d3f38PCw1WpFuHjbhwZws4QTPCiR0BK1ztHN\nM5k4ljRajcVlrEticXa5jCGWj9PpNC4P93q9WNVF+UTSnSmu7sfKJsl0T3YnLi4uCoVCTIRYXFws\nl8vJUjWO8/2JEx+cO3HTa6Z49nq9vrm5GYu2WNpGBBVZXjd6AN9GVE2USqV6vZ6k4G9sbMSbmM/n\nkyqFVCqVBHhXXrOORqPIdTk/P48mWtE1azwex1/Pzs5OT0/Pzs7e352YzWYRhwyHw9gci96phUIh\nPs+1Wu2az863EhFauVyOCDbyuC5/7y5/MWOLJmZLx1bh+vp6fJXubAFJzMDe3d3d2dk5ODg4OTnp\ndDpqJ4DvgXCChyM2GWLE2NLS0sbGxvr6+vz8fNQcRwwQ+w9JitHn1E4kExveqZ1IGislHV0uV9BG\nXkc0ovnSF/KNl0qFQiFGO2ez2cFg0Ol0YjEUS7pveSQ3JJKdouo3mWA9Pz+frEqvpRohymYi16XV\nah0dHR0fH0dQMRgMUm/DiZjj1m63kx6+SV/Rd2onIqKrVCrT6TSfz8cE9KSW4+sP+BuL705sRDQa\njcuJfO/cslKpxJZOxPD1en04HMbOUnyRY1fwVl7Fx7Tb7d3d3VevXv3yyy/Pnj178+bNyclJr9cT\nSwDfA+EE91isSOJactRxxppje3t7fX398ePHa2trSWOlpOI5acqUpNRHfvw7C5RyuZwkLyVtkZIO\nlZebtF5u6BRteZL5uNc7i+CDg6uTX17+8xOLrWQ/5PLV9zgJ29vbx8fHOzs7UU0Rk8UehuS9ju2I\nJPHsM13u3JV623zpcvQYeU3j8bjdbkcgsbOzs7e3d3JycnZ21u/30+l0tBAdjUaxRxFhatI0LHmu\nZE55bFz0er0kU2thYSHmkFxXUfi3lHx33knt+6DoFZt6G+uOx+NyuRwdn+5UOJG8p/v7+8+fP//n\nP//5888///rrr3t7e+fn59G2644cKsDNEU5wjyXNWPP5fAwKWFlZ+fHHH58+fbq1tbW1tbW0tJRM\ndUhqFVKXqhcuFzC8//jJSIFoTJnc8f2Ch2Qpn4gU/Gt8sZdjldTbNUoSySQJM5cbaCa3ufzX9KUm\nuZcTe1KpVFKSnvQkvcbjv0XJWUpy2GJbIPK7fldyei9HEVEhE5tXkd0UsUGr1To4ODg4OHj9+vWb\nN2+iRmIwGMSpTpLiIi8uJAcZb9Dl50qlUpPJpNfrnZ6eHhwc7O3t1ev1ZPzI/XqD3vlYflrMmlhc\nXCyVSlG0HelqSXO2O7JGH4/Hsdf0+vXr58+fP3v27MWLF3t7e+12O5kLaYMCePCEE9wzsfyN67XF\nYjG6eVYqlaT755MnT3766af19fW1tbVarfY1y4703ZiWNZlMRm8lC9BYUcWi8/LyNFal79RjvNMq\nKhK9klSfTCYTNevtdjtalz6w4tHIIIpc/Gazub+/v7Ozk8/noyFsNIpNAsIkMLtcEhO5N8mkiCQF\nLhKWomp/MBhMJpN2u314eHh0dLSzs7O7u3t6evqVwwFjg+L4+Pj169fVajXqlfP5/PLy8vz8/LWX\n6d+oz/8yptPpZKrgjR7SlUV9S7PZPDg4ODw8fPbs2c8///z8+fPffvvt4OAgdqtu+xgBvhHhBPdM\nxBL1en1+fn5+fn59fX11dXVhYWF9fX19fT32KDY2NmKcwm0f7DWIBvydTuf8/LzT6cQl8FQqFfke\nsdhNVreRlPXOfkuyZREhRP6SWIxmMplIx+90Om/evDk+Po4S0uTC+X13cXExHA5ns9nx8fHu7m6t\nVotCkaigqNVqyaI82TGICGQ0Go3H4+TcRpLScDjsdDoxVSD+HAwGMawjlv7RsqnZbJ6cnFxL8Umv\n15tMJlG0M5vNotQnFtxJUtCDMbt78+neNx6Pz87OTk5OXr169dtvv+3s7Dx79uzXX389ODg4Ojrq\n9Xq3fYAA35RwgnsmsiDm5+c3NzfX19d/+OGHJ0+erKysrK2tLS0txczpSC6/7SO9HuPxuNvtnp6e\nHh0dnZ6exu5BKpWKYCCWvDEcI9Lxo5tt7KhcTp5JigdiVycZqhDPEglj5+fnr169evXqVVSRPpjL\nqxFODAaDqI8vlUq1Wi2Xy0WF/XQ6TWYIxvlMyiGS4CFymc7Pz8/PzyP1qNlsnp2dtVqtTqeT9AhO\ntkFi+mGS7vL1xz8YDE5PT/f29mJ69Pz8fKlUajQa5XL57i++v8i9eDmDwaDZbL558+aXX3755Zdf\nXr9+/fLly729vfic3PbRAXxrwgnutKTxTlxcL5fLjUaj0Wg8evToyZMnW1tbP/30048//ri6urq8\nvBwzg2/7kL/YO42hYjMhuUweyTOHh4d7e3vHx8fJeiUyyGPx2u12z8/PI98mCScu705EF9TYmoiu\nuBFRJHMz4sbRNX9/fz+usF7XavjWJcnrJycnyVj00WjUbDabzebCwkKyURPhRKQ2RUgQW0Nxbtvt\n9tnZWbfbPTk5iSFlUWb9wRL5a9dut2PicgxSTKVS0VA1yl2S+Pk+fgXujsvv4+XK+/gSJTmHh4eH\nv/3224sXL37++edffvklOsM+pNYFAF9EOMGdFv85jyuytVptdXV1fX19ZWXl8ePHjx8/jhkCW1tb\nsSK87YP9Yu/UB0eJcKTgD4fDWNe22+0o7d3f3z89PR0MBrHKT3YnIgmn1+tFTk6EE+8kOyXhRCTM\nJHMzYnciWTPFBfhYNMeV+1s8OTek1WrF1k2n04mOsbFTkcwbiSyvqJRIyqwjlynGGvb7/Xa7fXp6\n2ul0PvNJI2a7wtG+c8ezs7NoGhstfSPCWV5eXl5ejq/A+w3K+CKXz1767cD7/lunp6cxMGR3d/fl\ny5e7u7u//fbbq1evjo+PoxcwwPdJOME9EDO8NjY2nj59GnsRm5ubq6ur0QQ2hsTd9jFeRexCjEaj\nTqdzcnJycnJyfHwcjYBiCTuZTDqdzuHh4fHxcZQ0JA1/Yv8hljtJlv/lUuwP1k7EhfmYFJaMfE4W\nrDFoIvKCInXnts7MzZlOpzFYo9fr7e7uxtDlJPqKd+SdOuykAju5OB2lFJ8fJFz5TL5zx8h6ajab\nuVwuctt6vd7jx48nk0nU09/EDPUH7HPqNKbTaa/Xi0qYN2/evH79+vDwcGdnZ39/P8YRJg1hAb5b\nwgnurqR90/r6eiQ1/fnPf3769Gk0hK3X67F4eidj5x5J+oH2+/3j4+M3b968ePFif38/Cq97vV4s\nZZIJyqPR6INdm96ZivAxyR3feYT3jycJWh6eOKXD4fDk5ORyq9zLt7nc9jf172f4chfXd/rwfhuj\n0ajVao1Go1jInp+fn5ycRPgX2xTxvbh334VbcfksJR3SkiEzEV0PBoPj4+O9vb2Dg4PffvstZkrs\n7e21Wq1opxa3v/IGFMADIJzgroi51Ely/9zcXHQvLRaL0f71yZMnT58+3d7eXlhYqNfrD2PBlIRD\ncdX51atXL1++jHSjXq8X6drdbjdCi9s+2IcgIoF7XRYSO0jn5+exJRVtuGI8RQTVyfT32z7S+yHS\nmc7Pz8/OzjqdTqfTGY1Gs9ks2aw7OTmJ+qUouY4xhQ813ga4AuEEd0Imk4l+O9VqdWFhIcbfLi4u\n1uv16OMUJRNra2v1ej26ZN72IV+DdDodWfuFQiGVSg0Gg2gwenZ2FrsTqVQqIgprF96RJMJNp9NI\nYJvNZlEVE61vhROfKbq+7u7u7u7uJlmFEU6EbrcbZffHx8fRXc33EeAy4QS3LJvNlsvlarW6vLy8\ntLQUkcPW1tby8vLq6mqj0cjn8+VyuV6vxyTgUqmUtDd9GKIEYjgcttvtZrN5fHyctGmKPJxv0ziI\neydy5LrdbkQU2Ww2AvIIUB9Mr+Sb1mq1dnd3f/nll2fPnu3u7kZRRDIgMvaykkqVKMq/7UMGuFse\n1LKMeyFqgiMfI66kzs/P1+v11dXV1dXV+fn5jY2NCCei92syLeHhNa6JMQXj8bjdbh8fH+/v70eK\n9mAweCd9H94XVTfRqbZarc7Pz29tbZ2fn1cqlaikf0hfls8RSV/xZ8QDlzsThMxb8X8p5+fnR0dH\nr169+p//+Z+///3vu7u7BwcHZ2dn8YCxC5FMRo9iiVt9iQB3kXCCby0unUZS0/z8/MrKStSPxg+N\nRiP2Jebn5xuNRvTXT6VSkRT0wJZHsSkRxdbHx8fNZvP09LTf79/2cXHPRAfbqLFJhqN/b+FEMiqk\n2+3GDMFo+BstkiOQyOVyuVwuOu3G6JVWqxWV1m/evInxEc1ms9vtRrZYhBNJFb4cJ4APEk7wjUQy\nRumt5eXlzc3NtbW1R48era6uViqV+fn5hYWFSqVSr9cbjUYUYd/2Ud+sJMcpejdFDehtHxT3z2g0\nGgwGsVORhBOpz2uE+gDEjkSn02m1WvFtarfbMbllOp1GFBH//xPN4mJfolwuFwqFs7Oz33777fnz\n5y9evNjZ2Yl+Wbf9ggDuGeEENyj71txb9Xq9Xq9Xq9X19fUffvjh0aNH29vbS0tLhUKhVCpVKpW5\nubmIN76HZdBsNouhE+12O8mAT7IyptPp5W6wt3uo3HExeyQks9Vv8Ut0ua/u5Za7l29zuXNx0rH3\n08f8Tsfe2WwWOxIxGuLw8LDZbB4dHZ2cnMRGzWQyiSgiyZmMkfDFYjH+36bT6Tx//vy3337b2dmJ\nqfM3eloAHiThBDclnU5H19cokIiur4uLi9HmdWNj46effnr8+PHW1la9Xk+lUplMJvmv/vcQS6Te\n1k5EM8rZbFYsFuv1+mAwSKYyf2IpBokoK8pkMpeLbW43lojPcIwCTMoYLn+Sk8jhndGKqX+fqZJ6\nb8RKPGDMjI+hhGdnZ2dnZwcHB7u7u0dHR/v7+0dHR9HMIHYnCoVCpDkVi8VyuRwXLKrVarFY7Pf7\nkeMUc11u5XQB3HfCCW5EXAVsNBrz8/O1Wm1paSkSmRYXFxcXF6vV6tra2vb29ubm5srKym0f7K1J\nLscWCoWFhYWtra3ZbFav15MLruPxOGYwqwHlE2LKRCyyY539jWsnkoV+Mkc8ZmIkxxOS8PhybXSS\ngFQsFguFQjJb8IN9CJJNmF6vF/sPrVar2Wy2Wq39/f3d3d1kj6Lb7V7enYhwJWbahNidGI/HzWbz\n4ODg/Px8Mpl8m9MF8MAIJ7h+hUKh0WjE+vjRo0crKysbGxuLi4vFYrFarSajJFZWVhYXF2/7YG9T\nUpW+srIynU7n5+dXV1ej9efp6WmkccdIu7jOetvHy901mUwudzJNqpC/QUSRtDyKfbbRaNTv9zud\nTvwZy/rBYDAcDuOoIkjIZrOxLxEDZyINslKpxNI/womk9DmJK4bDYbfbjVngUSNxcnJycnISv2k2\nm/F9idqJy5XT8VwRuhQKhUKhUCwW8/n8xcVFt9s9OzvTAgHgyoQTXL9isbiwsLC5ufmHP/zhj3/8\nY2Q0LSwsJOnLSRLzbR/pLYuZG4uLi+l0ular9fv909PTWBJF5kaz2cxkMnF9dzgc3vbxckfFhITY\n1BoMBkn5xDd46ul0Ghtoyfj2wWAQ9dDn5+enp6etVqvb7cY/RdQRcU5sF+Ryudi9XF5eXllZiWZu\nkbh1uZNSUlne6/ViO2Jvb29/fz9CiLOzs4hYIpS6XIx++RSl3mZhDYfDpKwrmX6taxPAlQkn+FqR\nmRB93MPq6ur29vb29vaf//znv/zlL5HUVC6Xb/tI75yYBZ7L5UqlUmxQJIngOzs79Xp9Z2fn4uIi\nlmi3fbDcXRcXF4PB4Pz8vNVqtVqt8/Pz6JQaX8zfvfvl+ubLZv8u9baMIW4ZSU0x4i0+op1OJzYl\nzs7OTk5OYsT7yclJ0sf2cvPW3FsxcyZm3i8sLETKU+RuJTtySTDQ7XZj1OPOzs7Ozk70Vu71ep8T\nOyWT6b7qXAPwHuEE1yAZa724uNhoNDY3N7e3t9fX13/88ccff/xxbW3t0y1fv5N2lu9LFlXFYjGW\nfdPpNC6yxgi/crk8HA6jh+xtHyx3V9JxOFKAzs/Po7g/n89/zt3j2zeZTJJWs/FDVCkkMUDyPY1w\nIi7qJ2UMg8EgGXwRUXH82W63IwtrMBhc3p1I9geq1Wqr1YoS6nq9XigU3t+dSOKZiFXi9kdHRzG+\n+sbOKwCfRTjB74s85o/9ayaTqdVq0fh1e3t7dXV1a2trY2Oj0WgsLS01Go1c7nc+Zt9nLBGSXpnx\n12w2W6lUKpVKNpuN9JWDg4NarVYoFG71MLnTko7DrVYrGWDyRclO6XR6PB7H/sblJKVOpxMtAZIw\nIPU2nIhgI4k9LhdIxM8xDSNSj2JSddLBNo4t9h86nc50Ou12uycnJ8ViMWKMd0qxkx/G4/FgMIid\nEOl/AHeEcILf97F1ScyXXVlZ2d7efvLkyZ/+9Kc//OEP6+vrUXhdeCv6V37PMcPVJEN8Y3V124fD\n3RXjF6bT6enp6cHBwd7e3u7ubr1ej/r+T9/x4uKi3++32+12u314eHh0dJRc+z8/Pz87Ozs/P494\nIAknYqciadwUFdhxmyRgiNt8MGMqxK5CRBSxtRI9nVKfvL6QPGZ0kYpacz2UAW6XcIIvEx1RYiBd\ntVotlUrr6+vb29uPHz9++vTpjz/+uLy8vLS0pMz6S0WqSVxXnkwmzWbz9evXb968iZaX4/H4tg+Q\nuysGmEyn07Ozs729vVqtVi6XR6PR6upq9GWO21xuwBrZSpHLFGHD2dnZ8fHx8fFxu91uNpvNZjMZ\nsJjsLcTyPRb0SfvX+NfbPgcA3BrhBB/1/mW/dDpdKpWWlpYWFxfX19dXVlaijHJzc3N5eXljY2N5\neVlmztV0Op3j4+Po5tTtdo+Ojvb29o6PjyOiUIrNJyTX/geDQbPZfPHiRTqdbrfby8vLy8vL9Xo9\nuq/m8/kIBiIGiJqHqEaIFKlWq9Vut/v9/vn5edRzR6uoGGRxOdnpnbrt2z4BANwm4QQf9U4sEYOf\n1tbWHj9+vLm5+eTJk83NzSiQWFpaqlQqMfo6urLc1jHfO7EaOz093dnZef369a+//vrmzZtWq3V8\nfNxsNnu9XrvdbrVamuLzCZfHMjSbzfF43O129/f3G43G4uLi/Px8sViMTcVMJjOdTqMCIWKGbrfb\nbrfPz8+TgulkeGJS85DkLCXhRMqYdgDekpDN78tms7VabXFxcWlp6cmTJz/99NOjR4+ePHmysbFR\nrVZrtVqtVsvn84VC4Xerrkm9DSGiuWc0w9nb23v+/PnLly8jnIjeNZ1OJ64ix3Bf+SR8jtiFKJfL\nlUqlXC7HhLhyuVwqlZJwIpk0F31du91uv9+PgdPJRkT8cNuvBoB7QDjB74hRa2tra48ePdrc3Hz6\n9OnTp0+j3nppaSkuecYy5TOb3BMXfUejUdS8RnbK8+fP37x58/r16/39/chCiV6fUXUqpYTPFAXN\nyVSHYrEYsUSpVIq+CNGUKRmhPX7rg+MmbvvVAHAPCCf4HQsLC8vLyz/99NMf/vCHJ0+ePH369MmT\nJwsLCwsLC7VaLWnGwmeazWbJheHd3d0XL168efPm2bNnL168OD4+jr46cZH4to+UByJpn5DUTiTp\nTEki020fIwD3mNQUPiWfzy8sLGxvb//lL3+J+dYxn65cLn96Mh0fFFd8R6PR+fn50dHRy5cv//GP\nf/z666/Pnj3b3d2NxCdNnLhe0c41k8lEx+HozSqEAOC6CCf4lLm5uYWFhWgC+4c//GFjY2N1dbXR\naNiRuLK4NtzpdJrN5u7u7suXL1+8ePH69etmsym9hJvz/oRpALgWwgk+IFKYstns6urq9vb206dP\n//jHPz59+nRhYaFer9/20d1jsZJLwon9/f2dnZ3d3d2Dg4PbPjQePuU3ANwE4QQfkM1mK5VKvV6P\nYokff/zx0aNHq6urhtN9vZj/NRwOe71e0qnztg8KAOCKpKzwAdEZdn19/cmTJ48ePVpZWalUKprA\nfr2kYU7SqSmXy83NzUU7fwCAe0c4wQfk8/nFxcXt7e3/+I//+Omnn9bW1qrVquF01yJJW48fstls\nNN651YMCALgi15v5gFwuV6vVNjY2tre3Hz16tLy8XKlUlF9/vQjJCoVCPp/P5XKZt+xOAAD3lAUi\nH5DNZkulUqPRWFxcbDQapVLJ1sR1yWazMfgvggqxBABwr9md4F2ZTKZSqczPzy8tLS0vLwsnrlFU\nTQwGg8Fg0O/3B4NBTKzTuBMAuKeEE/yvyOOfm5tbXFxcWVlZWVlZXFys1+u5XM4V9K83nU6Hw2G/\n3z86Otrb29vb2zs8PDw7OxsMBjp4AgD3lHCC/5XL5UqlUrVaXVxcjBETpVIpn8/f9nE9EDFu4vT0\ndGdn59WrV69fv97d3W02m71ebzqd3vbRAQBchXCC/5XP5yuVSmxNLC0tRThx2wf1EMxms+l0en5+\nvr+/v7u7+8svv/zyyy+//vrr7u7u6enpdDqV7AQA3FPCCf5XNpstFou1Wm1+fr7RaFQqlbm5uds+\nqIdgMpkMBoPYl3j27Nnf//73f/7znzs7O61WazQapVIpuWQAwD2lsxP/K51Ox1S1UqlULBbz+bxl\n7rWYTCb9fr/Vau3t7b18+fLFixe7u7tnZ2cRS6QuDaMAALhf7E7wL5lMJpfLxe5EvV6vVCrCiesy\nGAxardb+/v7Lly+fP3/+4sWL/f396Ol024cGAPBVhBP8SzqdjnETtVqt0WhUq9W5uTnhxLUYjUYR\nTuzs7Ozs7BwcHPR6vds+KACAayDZiVQqlUqn05HpVCgUyuVypVIpFovZbFYSzrUYjUbn5+cnJycn\nJyfn5+eDweC2jwgA4HrYnSCVSqWiZKJarS4sLKysrMT0OrsT1yIynY6Ojg4ODprN5vn5eVIyAQBw\n3wknSKVSqUwmUywWG43G6urq5ubm5ubm8vJyuVzOZL7r/avZbBb7M7PZ7DMjq9jnufwIrVbr+Pj4\n8PDw6Oio1WpJcwIAHhLhBKnU2wF2jUZjeXl5bW1tdXV1YWEhk8l8z+HEbDa7eCt+8zkRReSMpdPp\ni4uL8Xg8HA5PTk6azebp6Wmn0zEAGwB4YIQTpIrF4tLS0ubm5g8//PDDDz9sbW2trKzExIkHXDsR\nOw8XFxfT6TT5M36YTCbxZ/xwcXERuxO/G04kJSi5XG48Hvf7/U6n8+bNm2fPnr1586bZbPb7fQOw\nAYCHRDjxXUun0/l8fmVl5Ycffnj69Onf/va3P/zhD+vr65VKJbnB7R7hzYmXNh6Pu91up9M5Pz/v\ndrvD4XAwGAyHw8lkMh6Px+PxdDqNAOB3w4n412iQlclkJpNJr9frdrsHBwevX7/e29s7ODhQOAEA\nPDDCie9aPp+v1+tra2s//fTTX/7ylz//+c+PHz+u1+sPOIp4R7RwPTw83NvbOzo66nQ67Xa71+vF\n1sRoNHonnEin0+/v2CSBRPLXdDo9mUyGw2G/3z87Ozs9PW2322dnZ8Ph0O4EAPCQCCe+a7VabW1t\n7ccff/zzn//817/+9Y9//OPGxkYu9118Kvr9frJ1sLOz8+LFi52dnVardXZ21ul0Itkpdiei2uET\nuxOXA4mkdHs6ncb+xmg0Go1G4/E4UqcecP4YAPAd+i4WjnxQsVhcWFjY2NjY3t7+4Ycftre319fX\ni8XibR/Xt9DtdqNz64sXL54/f763t/f69evd3d3z8/PYnUiKKKK+IvV7yU7JPyXhRDyC4AEAeNiE\nE9+vcrk8Pz+/vLy8urq6uLhYrVbz+fxtH9S3cHFx0W63d3Z2nj9//o9//OO33347OTk5OjqKUumo\nnUgKtT8nHngnzEgiCrEEAPDgCSe+R/l8fm5ubnl5eXNzc3t7+/Hjx6urq9Vq9X6lOSUtmGLV/v6a\nPp1OZzKZbDYbP6RSqShm6Ha7r1+//vvf//73v//9559/fv36ddRMd7vdpJvT7bwkAID75j4tH7ku\nUYG9urr66NGjy+HEbR/XB8Q1/qiKHo1Gg8FgNBrFoj9pu/TBGXPxy2iylAzQGAwGvV6v0+m8fv36\n559//sc//vE///M/BwcHSWfY23iJAAD3mHDie1QoFGq12tLS0urq6tra2p3NdEoyjobDYbvdbrfb\n0SJpMBj0+/3RaBTJSJ8IJz62O7G3t/f8+fOdnZ1ms9ntdr+fTlYAANdLOPE9qlQqy8vLjx49evz4\n8cbGxtLSUqVSuYMDsJN9iVartbe3t7+/v7Ozc3R01O12e73ecDhMbvaJcCKTySRV1DGmut/vn5yc\nxCCIs7Oz1IOe1gcAcKOEE9+dTCZTqVRWVlYePXq0tbW1urraaDTuYCyRSqUuLi5Go1G32z0+Pn75\n8uWvv/767Nmzvb29fr8/HA5jHtzvjpaLlxY3i+at8ZhnZ2ftdvvbvBAAgIdKOPHdyeVy5XJ5YWFh\naWlpYWGhUqnc2QrsJJw4OTl58+bNs2fP/v73v+/v70fVxGQy+d1J1al/nzF3cXER/VtjmyI6wAIA\ncGV3dB3JtUsmrC0sLKyurm5ubiZpTnc5nIip0jFsLmZEHBwcyE0CALgj7ug6kmuXyWTm5uZKpdLW\n1tb29vbW1tba2tr8/HypVLrtQ/uoi4uLwWDQbrebzebx8fHJycnZ2dm1xxJJoAUAwJe6ixnz3IRs\nNlutVqM5bJRM1Gq1YrF4N6smwmw2G4/H3W630+kMBoNUKlUoFG7iWa79MQEAvhN3dynJ9crn8wsL\nC48fP/6P//iPn376aWNjo9FozM3NZbPZ2z60j0p2J05PTzudzmQyyeVy0anptg8NAIBUSrLT9yOX\ny9Xr9a2trZ9++unJkyexO3FbVRMfbO36vqR24uTkpNVqdbvd8XisfhoA4O6wO/G9yGazlUplcXFx\ndXV1eXm5Vqvd1ty6z08uuri4iGSn8/PzyHcaj8c3emwAAHwRuxPfi2gOu7a2trGxsbKyUqlUbiuc\n+PxUpclk0u/3z8/PW61Wu93u9XrCCQCAO0U48cBlMplsNpvP5+v1emxNrKyszM/P5/P5W69AGI1G\ng8FgNBpdXFyk0+lcLlcsFovFYnJgMXSi3W6fnZ3F7sR0Or3dYwYA4DLhxAOX9Iet1+uNRqNarZZK\npZvoj3QFg8Hg9PS03W6Px+NsNlsqlebn57PZbHJ4EW/0er1erzccDqfTqS5MAAB3inDigcvn85VK\nJdKcYl+iXC7f9kH9S6fT2d3d3d/f7/f7+Xx+fn5+NBrl8/kIJ6JwYjgc9vv9fr8/HA4nk8ltHzIA\nAP9GOPHA5XK5arW6vLy8sbGxuro6Pz9fLBY/duPZbDadTi8uLqJ7UjqdvpwQlX4rRlV8fa5Uv9/f\n399/9uxZu92em5tbW1vLZrONRqNer0f72ouLi8lkMhqNhsOhqgkAgDtIOPHA5XK5KMJeWlpaWFgo\nlUqfGDSRLN8nk8lsNstkMjHkIfq6RhlG3P2dSONqhsPhycnJmzdvTk5OSqXSdDqdn5/f3NycTCbZ\nbDZ5itlspjksAMDdJJx4sNLpdBQkLC4ubmxsPHr0KBo6fXAMduxLROlzt9uNEdT5fD66P02n0yiV\nLhQKc3Nzc3NzMU7uaw5vPB43m82dnZ3ffvvt+Pi4XC5fXFwsLS09evRobW1tbm5uMplcXFxELCGc\nAAC4m4QTD1bUNFcqlaWlpc3NzUePHq2urlar1Q+GARcXF9PptN/vt1qtGEGdTqfL5XKxWLy4uBiN\nRqlUKkq6a7VaKpWK6OJrDu/4+Hhvb+/NmzevX78+OjqqVCpzc3Obm5snJyfn5+flcjnSrmazmd0J\nAIA7SzjxYGUymUKhUC6XG43G8vLy0tJSo9GYm5v74I1ns9lkMhkMBu12+/j4+OzsLJPJ1Ov1SqUS\nwx8uLi6iPVQmk0mqpa9mPB4PBoODg4P9/f3j4+OYKTGdTttvnZ+f12q1dDotigAAuOOEEw9TOp2e\nm5urVqsxa2JtbW11dbXRaERNwvu3jzSn09PT/f39N2/enJ6e5nK5paWlWq02Go06nc50Oq1Wq8Ph\nMJvNxnSIqx1Y1EscHR394x//ePbs2e7ubqvV6na76XQ6RkycnJwcHx8Xi8VcLpdMwtYfFgDgbhJO\nPEyZTKZYLDYajZWVlYgllpaWIgZ4/5L/dDodDocRThwcHOzs7DSbzUKhENUUg8Gg1WpNJpNGozGb\nzSqVSqPRuPI4uV6vt7e3989//vP/+//+v59//nl3d7fT6UQJeHIMR0dHhUKhUChEpBETJ772jAAA\ncAOEEw9T7CHUarWFhYWFhYV6vV4qleKf3t+dSMKJs7Ozk5OTZrPZbDaLxWK0ger3+6enp9HrqVQq\n9fv90Wh0tfX9xcVFt9vd39//5z//+fPPP7948eL09DTCm6jtnk6nEVFEwXer1ep0OsIJAIA7Szjx\nMOXz+XK5PD8/v7i4OD8/X61Wk6qJd8KJ2WzW6/Vardbh4eHu7u7Ozs7u7u7JyUmxWEyn07E7cXJy\nEoXR5XI51vcRXXx+r9gowOh2uy9fvvzll19+/vnnX3755fXr11HkXSgUisViPp+fzWbD4bDVakVf\n2mazeXh4eH5+boAdAMDdJJx4mPL5fKlUqlQqlUqlXC5/rAJ7Op2Ox+Pz8/Pos/Tq1atXr17t7Oy0\nWq1Idup2u6PR6OzsLKZAVKvVq4UT/X4/wpX//u///u///u9ff/11f38/YonU2yZUc3NzhUIhk8mM\nx+PT09PBYLC/v7+7uxuF2tdzXgAAuFbCiQcomi/l8/lcLhdJRB8rZY5uTufn5ycnJwcHB3t7ewcH\nB81ms91u5/P52J2YTCbdbrdQKFSr1X6/PxwOR6PRF/Vcig2Qvb29//mf//mv//qvZ8+eNZvNyxsO\nEU5EelWxWMxms4PB4OjoaGdn5+DgIArBv/akAABwA4QTD002m41Mp2q1WqvVYp7Dx0bOTSaTyHSK\nCuw3b97s7e2dnJz0er2oZOj3+6lUajweRyzR7/d7vV7yZ6VSSaVS0+k0hmd/8CnG4/FwOGw2my9e\nvPiv//qv//zP//ztt9+63e7lCCG2JvL5fDxpr9c7Ozvb2dl5/fp1hBM6xgIA3E3CiYcmYomYNbG+\nvr62trawsFAulz944+indH5+3mw2YxBEs9lMdgMuLi7G43H0li0Wi9PpdDKZRDVFr9frdrvZbDb2\nPWIP5INBxWQy6XQ6R0dHr1+//vXXX3/99de9vb3LN4iq6xi/HXP0Li4uTk5OYmui1WqNx2PhBADA\n3SSceGiSnk5LS0urq6urq6sLCwsfGxMRE6/7/X6n04n5cTGxLv41NhaiZ2s2m424Iu7S6/U6nU7E\nDxFUZLPZCCreiSjG43Gn0zk9PT0+Po6hde8fbeyfRBXHYDAYjUatVuvk5CS6xMZs7Js5WwAAfBXh\nxEMTyU7FYrFcLkcddqFQ+FgmUjRcOj8/Pzs7Ozs7Oz8/T8qjU6nUbDYbjUYxXTudTkfbpdjKiHqM\nfr8fNQ9R7R3tmC4/fr/fPz4+3tnZefXq1e7u7unpaWRPhchxKpVK0dMpWtBeXFwkh9Tr9fR0AgC4\ny4QTD83srchNGo/Ho9FoPB5HZfY7N47qiHa73Wq1Wq3W+fn5OzeYTCaTyeTi4mI6nQ4Gg3a7fXR0\nlMlkOp1Ot9tdWVmp1+vz8/PJNkWhUIg7RqLU0dHRq1evnj9//vz58729vU6nkzzy3Nxc8a1sNntx\ncRHBQ0zA6PV6UfN9o+cKAICvJJx4aGLCdFzgb7Vap6ensdyPLk/v3ziW791udzAYjMfjjz1sBB6x\n/zAej9vtdr/fj1gll8vFDsPlCodoLxvNZ1+8ePHmzZtWqxVDtYfDYS6Xixynubm5XC43m83G43Hk\nVkXB95Un5QEA8C0JJx6aZBvh8PCwXC6Xy+V0Ot3tdvv9fq1Wi3igWq0WCoV+v392dnZ6epoMn/5Y\nxXOMzY7KimgsW6lUer1ezLaLARf5fD6ZbnFxcXF+fn5wcPDy5ctnz57985//fPXqVVRN1Ov16XQa\nWxm5XC7a0cYeSMQSg8FgOBx+s9MFAMDXEE48NDFKotVq7e/vp1KpmB1xenq6urpar9djER8NZHu9\n3uHh4dHR0enpabfbvbw1EZUS8XOMuosdg1Qq1el0zs/Py+XyxcVFVD4sLi4OBoNyuTydTmO2XXRz\nOjw8fP369W+//fbixYvd3d1+vx/ta1NvM7JiNF7Udg+Hw8FgEDset3HaAAC4CuHEQ5PsTkRW0nQ6\nHY1G0bUpwol8Pl+v16vV6nA4TDrDxiI+l8tF6XOUbseKPzKRIrqYzWaFQmEymUyn01qt1uv1er1e\n9GKKEouLi4tsNjsej2OcxfHx8dHRUbPZPD8/n81mmUwmqZSIY4vDi0cYDocSnAAA7hfhxEMTF/47\nnU6MqOv3+91u9/j4+ODgoFarRThRq9ViMyE2MU5PT2O+RKlUigKJiCKSECLih2S/Isokkl2LTqfT\n6XRKpdLc3FyhUMjn82dnZycnJ81ms9lsJplU2Ww2lUrFeIp48KgUHwwGEU6YLwEAcO8IJx6mpO9q\nrPhPTk729vZKpVLUZEcZdCqVim2Edrt9cXER0yqSzYdY2ceifzKZxE5F9G6KYolyuRxxRUQU0Ss2\nm81mMpnT09Ozs7NOpzMYDGLDIdnlGAwGkQ01Go3i8GJfIjY3bvmsAQDwhYQTD9nFxcVgMIjQ4uTk\nJJfLRSl2LPovD55L0pCiEjpprBQj5CIDKrYv6vX60tLSwsLC0tJStVqN1KZ+v9/r9QqFQuxgRDHG\nZDKJu1Sr1YgZIu0qek9dbkF7eesDAIB7RDjxkMUcutFo9P5AiVQqlU6noza6Wq0Wi8VcLpdMjYgp\n11EnHSMsZrPZ3NxcpVKp1Wr1er1Wq5VKpXQ6PRwOT09PM5nMcDg8OzsrFArT6bTdbscM7H6/P5vN\nooNTHMlgMEh2P2xHAADcd8KJ79dsNut2uxEqXFxcFAqFqGqIpKbYwYi/xg8xeC5au8a+R2wvZLPZ\n09PTSqVSLBbT6fR4PO52u+12u91uN5vN09PTXq+XFF7HcL3bfukAAFwP4cT3LjpBzWazZGx2EkLE\nBkX8kEql8vl8Pp9PpVKj0ajb7Y5Go3Q6HfsMMSMvk8lEffZgMIiOT9H6KTKdoj2UHQkAgIckfdsH\nwG1KooVcLpf8efk3sS8RMUBsWUTUkaRCRXemJCdqPB5HmXVkNIX418vVESolAAAeBuEE/woqoulT\nElfEbkMUasfqP7lZTKVIesWGyGKKKOK2XxAAAN+IcIJU6lKokH0riRwuhxMh7pI0ZbpcWi2WAAD4\nrggn+JcIFZK+sZlMJvV2PPbl28QPkbx0cUlKChMAwPdHOMG/SV9y+ZfJz5enZaurBgD4zgkneFcE\nD+/sS1yW7ELYjgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA\nAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAC+uf8f\n/4iRBDuMXqsAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<PIL.Image.Image image mode=RGB size=1050x910 at 0x7F5D40214390>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "I feel pain\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_4VZu2mJUPyr",
        "colab_type": "text"
      },
      "source": [
        "#Text to Speach"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EfGPqQI27aiT",
        "colab_type": "code",
        "outputId": "faaef214-ed2e-4b61-ca01-7d10b7ea3452",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 700
        }
      },
      "source": [
        "import os\n",
        "from os.path import exists, join, expanduser\n",
        "\n",
        "# Clone\n",
        "name = \"deepvoice3_pytorch\"\n",
        "if not exists(name):\n",
        "  ! git clone https://github.com/r9y9/$name\n",
        "    \n",
        "  \n",
        "# Change working directory to the project dir \n",
        "os.chdir(join('/content/', name))\n",
        "\n",
        "!git checkout 7a10ac6763eda92595e257543494b6a95f64229b --quiet\n",
        "\n",
        "# Install dependencices\n",
        "!pip install -q -e '.[bin]'\n",
        "\n",
        "\n",
        "%pylab inline\n",
        "! pip install -q librosa nltk\n",
        "\n",
        "import torch\n",
        "import numpy as np\n",
        "import librosa\n",
        "import librosa.display\n",
        "import IPython\n",
        "from IPython.display import Audio\n",
        "# need this for English text processing frontend\n",
        "import nltk\n",
        "! python -m nltk.downloader cmudict\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "preset = \"20180505_deepvoice3_ljspeech.json\"\n",
        "checkpoint_path = \"20180505_deepvoice3_checkpoint_step000640000.pth\"\n",
        "\n",
        "if not exists(preset):\n",
        "  !curl -O -L \"https://www.dropbox.com/s/0ck82unm0bo0rxd/20180505_deepvoice3_ljspeech.json\"\n",
        "if not exists(checkpoint_path):\n",
        "  !curl -O -L \"https://www.dropbox.com/s/5ucl9remrwy5oeg/20180505_deepvoice3_checkpoint_step000640000.pth\"\n",
        "  \n",
        "\n",
        "import hparams\n",
        "import json\n",
        "\n",
        "    \n",
        "# Load parameters from preset\n",
        "with open(preset) as f:\n",
        "  hparams.hparams.parse_json(f.read())\n",
        "  \n",
        "# Inject frontend text processor\n",
        "import synthesis\n",
        "import train\n",
        "from deepvoice3_pytorch import frontend\n",
        "synthesis._frontend = getattr(frontend, \"en\")\n",
        "train._frontend =  getattr(frontend, \"en\")\n",
        "\n",
        "# alises\n",
        "fs = hparams.hparams.sample_rate\n",
        "hop_length = hparams.hparams.hop_size\n",
        "\n",
        "def tts(model, text, p=0, speaker_id=None, fast=True, figures=True):\n",
        "  from synthesis import tts as _tts\n",
        "  waveform, alignment, spectrogram, mel = _tts(model, text, p, speaker_id, fast)\n",
        "  if figures:\n",
        "      visualize(alignment, spectrogram)\n",
        "  IPython.display.display(Audio(waveform, rate=fs))\n",
        "  \n",
        "def visualize(alignment, spectrogram):\n",
        "  label_fontsize = 16\n",
        "  figure(figsize=(16,16))\n",
        "\n",
        "  subplot(2,1,1)\n",
        "  imshow(alignment.T, aspect=\"auto\", origin=\"lower\", interpolation=None)\n",
        "  xlabel(\"Decoder timestamp\", fontsize=label_fontsize)\n",
        "  ylabel(\"Encoder timestamp\", fontsize=label_fontsize)\n",
        "  colorbar()\n",
        "\n",
        "  subplot(2,1,2)\n",
        "  librosa.display.specshow(spectrogram.T, sr=fs, \n",
        "                           hop_length=hop_length, x_axis=\"time\", y_axis=\"linear\")\n",
        "  xlabel(\"Time\", fontsize=label_fontsize)\n",
        "  ylabel(\"Hz\", fontsize=label_fontsize)\n",
        "  tight_layout()\n",
        "  colorbar()\n",
        "  \n",
        "  \n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cloning into 'deepvoice3_pytorch'...\n",
            "remote: Enumerating objects: 1111, done.\u001b[K\n",
            "remote: Total 1111 (delta 0), reused 0 (delta 0), pack-reused 1111\u001b[K\n",
            "Receiving objects: 100% (1111/1111), 6.92 MiB | 12.67 MiB/s, done.\n",
            "Resolving deltas: 100% (617/617), done.\n",
            "\u001b[K     |████████████████████████████████| 245kB 5.0MB/s \n",
            "\u001b[K     |████████████████████████████████| 143kB 45.1MB/s \n",
            "\u001b[K     |████████████████████████████████| 225kB 48.3MB/s \n",
            "\u001b[K     |████████████████████████████████| 1.7MB 48.6MB/s \n",
            "\u001b[K     |████████████████████████████████| 419kB 42.4MB/s \n",
            "\u001b[K     |████████████████████████████████| 368kB 44.0MB/s \n",
            "\u001b[?25h  Building wheel for lws (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Building wheel for nnmnkwii (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Building wheel for pysptk (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Building wheel for bandmat (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Populating the interactive namespace from numpy and matplotlib\n",
            "/usr/lib/python3.6/runpy.py:125: RuntimeWarning: 'nltk.downloader' found in sys.modules after import of package 'nltk', but prior to execution of 'nltk.downloader'; this may result in unpredictable behaviour\n",
            "  warn(RuntimeWarning(msg))\n",
            "[nltk_data] Downloading package cmudict to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/cmudict.zip.\n",
            "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
            "                                 Dload  Upload   Total   Spent    Left  Speed\n",
            "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
            "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
            "100  1795  100  1795    0     0   2770      0 --:--:-- --:--:-- --:--:--  2770\n",
            "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
            "                                 Dload  Upload   Total   Spent    Left  Speed\n",
            "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
            "  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0\n",
            "100   990    0   990    0     0    984      0 --:--:--  0:00:01 --:--:--   984\n",
            "100  151M  100  151M    0     0  33.1M      0  0:00:04  0:00:04 --:--:-- 48.5M\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "W0828 10:18:35.508102 140041326647168 lazy_loader.py:50] \n",
            "The TensorFlow contrib module will not be included in TensorFlow 2.0.\n",
            "For more information, please see:\n",
            "  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md\n",
            "  * https://github.com/tensorflow/addons\n",
            "  * https://github.com/tensorflow/io (for I/O related ops)\n",
            "If you depend on functionality not listed there, please file an issue.\n",
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PZ9Zlzeywv-I",
        "colab_type": "text"
      },
      "source": [
        "#Generate Speach\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nR1bzP3a8o1D",
        "colab_type": "code",
        "outputId": "c7bbba35-7e3a-452b-96be-a3ad7c819262",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        }
      },
      "source": [
        "from train import build_model\n",
        "from train import restore_parts, load_checkpoint\n",
        "\n",
        "model1 = build_model()\n",
        "model1 = load_checkpoint(checkpoint_path, model1, None, True)\n",
        "\n",
        "# Try your favorite senteneces:)\n",
        "texts = [\n",
        " predictedResult   \n",
        "]\n",
        "\n",
        "for idx, text in enumerate(texts):\n",
        "  print(idx, text)\n",
        "  tts(model1, text, figures=False)\n",
        "  "
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Load checkpoint from: 20180505_deepvoice3_checkpoint_step000640000.pth\n",
            "0 I feel pain\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:1386: UserWarning: nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.\n",
            "  warnings.warn(\"nn.functional.sigmoid is deprecated. Use torch.sigmoid instead.\")\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "                <audio controls=\"controls\" >\n",
              "                    <source src=\"data:audio/wav;base64,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\" type=\"audio/wav\" />\n",
              "                    Your browser does not support the audio element.\n",
              "                </audio>\n",
              "              "
            ],
            "text/plain": [
              "<IPython.lib.display.Audio object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}