Try&learn1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Try&learn1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNa6/ZhUJkpkA/g1Q4wa7Sh",
      "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/kr-viku/Glaucoma/blob/master/Try%26learn1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M015FAxhbA0j",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "outputId": "2b1d292d-7d6c-4a47-e40e-a4e7a9dbe135"
      },
      "source": [
        "from google.colab import auth\n",
        "auth.authenticate_user()\n",
        "\n",
        "!pip install PyDrive\n",
        "\n",
        "from pydrive.auth import GoogleAuth\n",
        "from pydrive.drive import GoogleDrive\n",
        "from google.colab import auth\n",
        "from oauth2client.client import GoogleCredentials\n",
        "\n",
        "# Authenticate and create the PyDrive client.\n",
        "# This only needs to be done once in a notebook.\n",
        "auth.authenticate_user()\n",
        "gauth = GoogleAuth()\n",
        "gauth.credentials = GoogleCredentials.get_application_default()\n",
        "drive = GoogleDrive(gauth)"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: PyDrive in /usr/local/lib/python3.6/dist-packages (1.3.1)\n",
            "Requirement already satisfied: oauth2client>=4.0.0 in /usr/local/lib/python3.6/dist-packages (from PyDrive) (4.1.3)\n",
            "Requirement already satisfied: google-api-python-client>=1.2 in /usr/local/lib/python3.6/dist-packages (from PyDrive) (1.7.11)\n",
            "Requirement already satisfied: PyYAML>=3.0 in /usr/local/lib/python3.6/dist-packages (from PyDrive) (3.13)\n",
            "Requirement already satisfied: pyasn1-modules>=0.0.5 in /usr/local/lib/python3.6/dist-packages (from oauth2client>=4.0.0->PyDrive) (0.2.8)\n",
            "Requirement already satisfied: httplib2>=0.9.1 in /usr/local/lib/python3.6/dist-packages (from oauth2client>=4.0.0->PyDrive) (0.11.3)\n",
            "Requirement already satisfied: rsa>=3.1.4 in /usr/local/lib/python3.6/dist-packages (from oauth2client>=4.0.0->PyDrive) (4.0)\n",
            "Requirement already satisfied: pyasn1>=0.1.7 in /usr/local/lib/python3.6/dist-packages (from oauth2client>=4.0.0->PyDrive) (0.4.8)\n",
            "Requirement already satisfied: six>=1.6.1 in /usr/local/lib/python3.6/dist-packages (from oauth2client>=4.0.0->PyDrive) (1.12.0)\n",
            "Requirement already satisfied: google-auth>=1.4.1 in /usr/local/lib/python3.6/dist-packages (from google-api-python-client>=1.2->PyDrive) (1.7.2)\n",
            "Requirement already satisfied: uritemplate<4dev,>=3.0.0 in /usr/local/lib/python3.6/dist-packages (from google-api-python-client>=1.2->PyDrive) (3.0.1)\n",
            "Requirement already satisfied: google-auth-httplib2>=0.0.3 in /usr/local/lib/python3.6/dist-packages (from google-api-python-client>=1.2->PyDrive) (0.0.3)\n",
            "Requirement already satisfied: cachetools<3.2,>=2.0.0 in /usr/local/lib/python3.6/dist-packages (from google-auth>=1.4.1->google-api-python-client>=1.2->PyDrive) (3.1.1)\n",
            "Requirement already satisfied: setuptools>=40.3.0 in /usr/local/lib/python3.6/dist-packages (from google-auth>=1.4.1->google-api-python-client>=1.2->PyDrive) (45.1.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5GgfCaT4c84U",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "c7f5acc6-036a-4ea0-a79b-ae581836ee88"
      },
      "source": [
        "fileId = drive.CreateFile({'id': '1bjsH1wijxXmXH1quyEpOW1vLgPCZIFZ3'}) #DRIVE_FILE_ID is file id example: 1iytA1n2z4go3uVCwE_vIKouTKyIDjEq\n",
        "print(fileId['title'])  # folder_data.zip\n",
        "fileId.GetContentFile('dataset.zip')  # Save Drive file as a local file\n",
        "\n",
        "!unzip dataset.zip -d ./"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Dataset.zip\n",
            "Archive:  dataset.zip\n",
            "   creating: ./Dataset/\n",
            "   creating: ./Dataset/Train/\n",
            "   creating: ./Dataset/Train/gl/\n",
            "  inflating: ./Dataset/Train/gl/Im256.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im257.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im258.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im259.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im260.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im261.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im262.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im263.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im264.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im265.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im266.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im267.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im268.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im269.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im270.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im271.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im272.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im273.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im274.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im275.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im276.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im277.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im278.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im279.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im280.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im281.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im282.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im283.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im284.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im285.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im286.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im287.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im288.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im289.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im290.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im291.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im292.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im293.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im294.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im295.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im296.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im297.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im298.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im299.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im300.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im301.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im302.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im303.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im304.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im305.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im306.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im307.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im308.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im309.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im310.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im311.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im312.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im313.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im314.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im315.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im316.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im317.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im318.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im319.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im320.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im321.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im322.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im323.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im324.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im325.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im326.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im327.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im328.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im329.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im330.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im331.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im332.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im333.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im334.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im335.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im336.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im337.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im338.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im339.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im340.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im341.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im342.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im343.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im344.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im345.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im346.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im347.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im348.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im349.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im350.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im351.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im352.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im353.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im354.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im355.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im356.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im357.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im358.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im359.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im360.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im361.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im362.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im363.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im364.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im365.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im366.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im367.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im368.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im369.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im370.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im371.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im372.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im373.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im374.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im375.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im376.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im377.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im378.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im379.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im380.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im381.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im382.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im383.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im384.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im385.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im386.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im387.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im388.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im389.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im390.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im391.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im392.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im393.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im394.jpg  \n",
            "  inflating: ./Dataset/Train/gl/Im395.jpg  \n",
            "   creating: ./Dataset/Train/normal/\n",
            "  inflating: ./Dataset/Train/normal/Im001.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im002.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im003.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im004.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im005.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im006.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im007.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im008.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im009.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im010.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im011.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im012.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im013.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im014.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im015.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im016.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im017.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im018.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im019.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im020.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im021.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im022.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im023.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im024.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im025.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im026.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im027.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im028.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im029.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im030.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im031.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im032.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im033.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im034.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im035.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im036.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im037.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im038.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im039.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im040.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im041.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im042.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im043.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im044.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im045.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im046.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im047.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im048.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im049.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im050.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im051.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im052.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im053.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im054.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im055.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im056.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im057.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im058.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im059.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im060.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im061.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im062.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im063.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im064.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im065.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im066.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im067.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im068.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im069.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im070.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im071.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im072.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im073.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im074.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im075.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im076.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im077.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im078.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im079.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im080.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im081.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im082.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im083.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im084.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im085.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im086.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im087.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im088.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im089.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im090.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im091.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im092.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im093.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im094.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im095.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im096.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im097.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im098.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im099.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im100.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im101.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im102.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im103.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im104.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im105.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im106.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im107.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im108.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im109.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im110.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im111.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im112.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im113.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im114.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im115.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im116.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im117.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im118.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im119.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im120.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im121.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im122.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im123.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im124.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im125.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im126.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im127.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im128.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im129.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im130.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im131.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im132.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im133.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im134.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im135.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im136.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im137.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im138.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im139.jpg  \n",
            "  inflating: ./Dataset/Train/normal/Im140.jpg  \n",
            "   creating: ./Dataset/Validation/\n",
            "   creating: ./Dataset/Validation/gl/\n",
            "  inflating: ./Dataset/Validation/gl/Im396.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im397.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im398.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im399.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im400.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im401.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im402.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im403.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im404.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im405.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im406.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im407.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im408.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im409.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im410.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im411.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im412.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im413.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im414.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im415.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im416.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im417.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im418.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im419.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im420.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im421.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im422.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im423.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im424.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im425.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im426.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im427.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im428.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im429.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im430.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im431.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im432.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im433.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im434.jpg  \n",
            "  inflating: ./Dataset/Validation/gl/Im435.jpg  \n",
            "   creating: ./Dataset/Validation/normal/\n",
            "  inflating: ./Dataset/Validation/normal/Im141.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im142.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im143.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im144.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im145.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im146.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im147.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im148.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im149.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im150.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im151.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im152.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im153.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im154.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im155.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im156.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im157.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im158.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im159.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im160.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im161.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im162.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im163.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im164.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im165.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im166.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im167.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im168.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im169.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im170.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im171.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im172.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im173.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im174.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im175.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im176.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im177.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im178.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im179.jpg  \n",
            "  inflating: ./Dataset/Validation/normal/Im180.jpg  \n",
            "   creating: ./Dataset/Test/\n",
            "   creating: ./Dataset/Test/gl/\n",
            "  inflating: ./Dataset/Test/gl/Im436.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im437.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im438.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im439.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im440.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im441.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im442.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im443.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im444.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im445.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im446.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im447.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im448.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im449.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im450.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im451.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im452.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im453.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im454.jpg  \n",
            "  inflating: ./Dataset/Test/gl/Im455.jpg  \n",
            "   creating: ./Dataset/Test/normal/\n",
            "  inflating: ./Dataset/Test/normal/Im181.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im182.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im183.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im184.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im185.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im186.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im187.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im188.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im189.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im190.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im191.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im192.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im193.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im194.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im195.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im196.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im197.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im198.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im199.jpg  \n",
            "  inflating: ./Dataset/Test/normal/Im200.jpg  \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s8tgguBndH9n",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3e1205a9-9ef0-4613-8443-273d79313d28"
      },
      "source": [
        "import tensorflow as tf\n",
        "tf.disable_v2_behavior"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<function tensorflow.python.compat.v2_compat.disable_v2_behavior>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ngW2LByofihM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.models import Model\n",
        "from keras.layers import Input, Conv2D, MaxPooling2D, UpSampling2D, Concatenate, GlobalAveragePooling2D\n",
        "from keras.layers import BatchNormalization, Activation, Dropout, Flatten, Dense\n",
        "from keras import backend as K\n",
        "from keras import optimizers\n",
        "from keras.callbacks import ModelCheckpoint, ReduceLROnPlateau\n",
        "from imgaug import augmenters as iaa\n",
        "from keras.utils import plot_model\n",
        "import numpy as np\n",
        "\n",
        "# channels last, no. of channels = 3\n",
        "img_width, img_height = 256, 256\n",
        "input_shape = (img_width, img_height, 3)\n",
        "\n",
        "train_data_dir = \"Dataset/Train\"\n",
        "validation_data_dir = \"Dataset/Validation\"\n",
        "test_data_dir = \"Dataset/Test\"\n",
        "nb_train_samples = 280\n",
        "nb_validation_samples = 80\n",
        "n_test_samples = 40\n",
        "epochs = 100"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h8yzVr2nftCs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def get_steps(n_samples, batch_size):\n",
        "    if n_samples <= batch_size:\n",
        "        return n_samples\n",
        "    elif np.remainder(n_samples, batch_size) == 0:\n",
        "        return n_samples//batch_size\n",
        "    else:\n",
        "        return n_samples//batch_size + 1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9_QdKnXJf2Jz",
        "colab_type": "text"
      },
      "source": [
        "#MODEL"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5FL_HoPzfyhO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def block_encoder(layer, filters, kernel_size, activation):\n",
        "  block = Conv2D(filters, kernel_size, padding='same', data_format=\"channels_last\", activation = activation)(layer)\n",
        "  block = Conv2D(filters * 2, kernel_size, padding='same', data_format=\"channels_last\", activation = activation)(block)\n",
        "  block = MaxPooling2D(pool_size = (2, 2))(block)\n",
        "  block = BatchNormalization()(block)\n",
        "\n",
        "  print(block.shape)\n",
        "  return block"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XGXbqaZQgFhK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def block_decoder(layer, filters, kernel_size, activation, upSample = True):\n",
        "  block = Conv2D(filters * 2, kernel_size, padding='same', data_format=\"channels_last\", activation = activation)(layer)\n",
        "  block = Conv2D(filters, kernel_size, padding='same', data_format=\"channels_last\", activation = activation)(block)\n",
        "  if upSample: \n",
        "    block = UpSampling2D(size=(2,2))(layer)\n",
        "  block = BatchNormalization()(block)\n",
        "\n",
        "  print(block.shape)\n",
        "  return block"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Js27tBgVgKxY",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 101
        },
        "outputId": "3bc209d2-d837-4024-d85f-f2c2eb0923aa"
      },
      "source": [
        "inputLayer = Input(shape=input_shape)\n",
        "\n",
        "en_block1 = block_encoder(inputLayer, 2, (2, 2), \"relu\") # output_shape = 128, 128, 4\n",
        "en_block2 = block_encoder(en_block1, 4, (2, 2), \"relu\") # output_shape = 64, 64, 8\n",
        "de_block3 = block_decoder(en_block2, 8, (2, 2), \"relu\", upSample = False) # output_shape = 64, 64, 8\n",
        "temp = Concatenate()([de_block3, en_block2])\n",
        "de_block2 = block_decoder(temp, 4, (2, 2), \"relu\")\n",
        "de_block1 = block_decoder(Concatenate()([de_block2, en_block1]), 2, (2, 2), \"relu\")\n",
        "\n",
        "output = Conv2D(2, (1, 1), activation = \"softmax\")(de_block1)\n",
        "model = Model(inputs=[inputLayer], outputs=[output])\n",
        "model.compile(loss=\"categorical_crossentropy\", optimizer=optimizers.Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.0, amsgrad=False), metrics=[\"accuracy\"])"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(?, 128, 128, 4)\n",
            "(?, 64, 64, 8)\n",
            "(?, 64, 64, 8)\n",
            "(?, 128, 128, 16)\n",
            "(?, 256, 256, 20)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "usZAey16i2hZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 823
        },
        "outputId": "1e839b0e-2cb4-4e2b-8a8b-fed3aad4f713"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model_2\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_4 (InputLayer)            (None, 256, 256, 3)  0                                            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_26 (Conv2D)              (None, 256, 256, 2)  26          input_4[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_27 (Conv2D)              (None, 256, 256, 4)  36          conv2d_26[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling2d_7 (MaxPooling2D)  (None, 128, 128, 4)  0           conv2d_27[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_12 (BatchNo (None, 128, 128, 4)  16          max_pooling2d_7[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_28 (Conv2D)              (None, 128, 128, 4)  68          batch_normalization_12[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_29 (Conv2D)              (None, 128, 128, 8)  136         conv2d_28[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling2d_8 (MaxPooling2D)  (None, 64, 64, 8)    0           conv2d_29[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_13 (BatchNo (None, 64, 64, 8)    32          max_pooling2d_8[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_30 (Conv2D)              (None, 64, 64, 16)   528         batch_normalization_13[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_31 (Conv2D)              (None, 64, 64, 8)    520         conv2d_30[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_14 (BatchNo (None, 64, 64, 8)    32          conv2d_31[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "concatenate_5 (Concatenate)     (None, 64, 64, 16)   0           batch_normalization_14[0][0]     \n",
            "                                                                 batch_normalization_13[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "up_sampling2d_3 (UpSampling2D)  (None, 128, 128, 16) 0           concatenate_5[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_15 (BatchNo (None, 128, 128, 16) 64          up_sampling2d_3[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "concatenate_6 (Concatenate)     (None, 128, 128, 20) 0           batch_normalization_15[0][0]     \n",
            "                                                                 batch_normalization_12[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "up_sampling2d_4 (UpSampling2D)  (None, 256, 256, 20) 0           concatenate_6[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_16 (BatchNo (None, 256, 256, 20) 80          up_sampling2d_4[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_36 (Conv2D)              (None, 256, 256, 2)  42          batch_normalization_16[0][0]     \n",
            "==================================================================================================\n",
            "Total params: 1,580\n",
            "Trainable params: 1,468\n",
            "Non-trainable params: 112\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0FnttHCIi7Bs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "988720b6-0a77-46a1-fb60-b8b4208249f8"
      },
      "source": [
        "input = Input(shape=input_shape)\n",
        "\n",
        "block1 = BatchNormalization(name='norm_0')(input)\n",
        "\n",
        "# Block 1\n",
        "block1 = Conv2D(8, (3,3), name='conv_11', activation='relu')(block1)\n",
        "block1 = Conv2D(16, (3,3), name='conv_12', activation='relu')(block1)\n",
        "block1 = Conv2D(32, (3,3), name='conv_13', activation='relu')(block1)\n",
        "block1 = Conv2D(64, (3,3), name='conv_14', activation='relu')(block1)\n",
        "block1 = MaxPooling2D(pool_size=(2, 2))(block1)\n",
        "block1 = BatchNormalization(name='norm_1')(block1)\n",
        "\n",
        "block1 = Conv2D(16, 1)(block1)\n",
        "\n",
        "# Block 2\n",
        "block2 = Conv2D(32, (3,3), name='conv_21', activation='relu')(block1)\n",
        "block2 = Conv2D(64, (3,3), name='conv_22', activation='relu')(block2)\n",
        "block2 = Conv2D(64, (3,3), name='conv_23', activation='relu')(block2)\n",
        "block2 = Conv2D(128, (3,3), name='conv_24', activation='relu')(block2)\n",
        "block2 = MaxPooling2D(pool_size=(2, 2))(block2)\n",
        "block2 = BatchNormalization(name='norm_2')(block2)\n",
        "\n",
        "block2 = Conv2D(64, 1)(block2)\n",
        "\n",
        "# Block 3\n",
        "block3 = Conv2D(64, (3,3), name='conv_31', activation='relu')(block2)\n",
        "block3 = Conv2D(128, (3,3), name='conv_32', activation='relu')(block3)\n",
        "block3 = Conv2D(128, (3,3), name='conv_33', activation='relu')(block3)\n",
        "block3 = Conv2D(64, (3,3), name='conv_34', activation='relu')(block3)\n",
        "block3 = MaxPooling2D(pool_size=(2, 2))(block3)\n",
        "block3 = BatchNormalization(name='norm_3')(block3)\n",
        "\n",
        "# Block 4\n",
        "block4 = Conv2D(64, (3,3), name='conv_41', activation='relu')(block3)\n",
        "block4 = Conv2D(32, (3,3), name='conv_42', activation='relu')(block4)\n",
        "block4 = Conv2D(16, (3,3), name='conv_43', activation='relu')(block4)\n",
        "block4 = Conv2D(8, (2,2), name='conv_44', activation='relu')(block4)\n",
        "block4 = MaxPooling2D(pool_size=(2, 2))(block4)\n",
        "block4 = BatchNormalization(name='norm_4')(block4)\n",
        "\n",
        "block4 = Conv2D(2, 1)(block4)\n",
        "\n",
        "block5 = GlobalAveragePooling2D()(block4)\n",
        "output = Activation('softmax')(block5)\n",
        "\n",
        "model = Model(inputs=[input], outputs=[output])\n",
        "model.compile(loss=\"categorical_crossentropy\", optimizer=optimizers.Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.0, amsgrad=False), metrics=[\"accuracy\"])\n",
        "\n",
        "\n",
        "plot_model(model, to_file = \"model.png\", show_shapes = True)\n",
        "model.summary()"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"model_4\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_7 (InputLayer)         (None, 256, 256, 3)       0         \n",
            "_________________________________________________________________\n",
            "norm_0 (BatchNormalization)  (None, 256, 256, 3)       12        \n",
            "_________________________________________________________________\n",
            "conv_11 (Conv2D)             (None, 254, 254, 8)       224       \n",
            "_________________________________________________________________\n",
            "conv_12 (Conv2D)             (None, 252, 252, 16)      1168      \n",
            "_________________________________________________________________\n",
            "conv_13 (Conv2D)             (None, 250, 250, 32)      4640      \n",
            "_________________________________________________________________\n",
            "conv_14 (Conv2D)             (None, 248, 248, 64)      18496     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_13 (MaxPooling (None, 124, 124, 64)      0         \n",
            "_________________________________________________________________\n",
            "norm_1 (BatchNormalization)  (None, 124, 124, 64)      256       \n",
            "_________________________________________________________________\n",
            "conv2d_40 (Conv2D)           (None, 124, 124, 16)      1040      \n",
            "_________________________________________________________________\n",
            "conv_21 (Conv2D)             (None, 122, 122, 32)      4640      \n",
            "_________________________________________________________________\n",
            "conv_22 (Conv2D)             (None, 120, 120, 64)      18496     \n",
            "_________________________________________________________________\n",
            "conv_23 (Conv2D)             (None, 118, 118, 64)      36928     \n",
            "_________________________________________________________________\n",
            "conv_24 (Conv2D)             (None, 116, 116, 128)     73856     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_14 (MaxPooling (None, 58, 58, 128)       0         \n",
            "_________________________________________________________________\n",
            "norm_2 (BatchNormalization)  (None, 58, 58, 128)       512       \n",
            "_________________________________________________________________\n",
            "conv2d_41 (Conv2D)           (None, 58, 58, 64)        8256      \n",
            "_________________________________________________________________\n",
            "conv_31 (Conv2D)             (None, 56, 56, 64)        36928     \n",
            "_________________________________________________________________\n",
            "conv_32 (Conv2D)             (None, 54, 54, 128)       73856     \n",
            "_________________________________________________________________\n",
            "conv_33 (Conv2D)             (None, 52, 52, 128)       147584    \n",
            "_________________________________________________________________\n",
            "conv_34 (Conv2D)             (None, 50, 50, 64)        73792     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_15 (MaxPooling (None, 25, 25, 64)        0         \n",
            "_________________________________________________________________\n",
            "norm_3 (BatchNormalization)  (None, 25, 25, 64)        256       \n",
            "_________________________________________________________________\n",
            "conv_41 (Conv2D)             (None, 23, 23, 64)        36928     \n",
            "_________________________________________________________________\n",
            "conv_42 (Conv2D)             (None, 21, 21, 32)        18464     \n",
            "_________________________________________________________________\n",
            "conv_43 (Conv2D)             (None, 19, 19, 16)        4624      \n",
            "_________________________________________________________________\n",
            "conv_44 (Conv2D)             (None, 18, 18, 8)         520       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_16 (MaxPooling (None, 9, 9, 8)           0         \n",
            "_________________________________________________________________\n",
            "norm_4 (BatchNormalization)  (None, 9, 9, 8)           32        \n",
            "_________________________________________________________________\n",
            "conv2d_42 (Conv2D)           (None, 9, 9, 2)           18        \n",
            "_________________________________________________________________\n",
            "global_average_pooling2d_2 ( (None, 2)                 0         \n",
            "_________________________________________________________________\n",
            "activation_2 (Activation)    (None, 2)                 0         \n",
            "=================================================================\n",
            "Total params: 561,526\n",
            "Trainable params: 560,992\n",
            "Non-trainable params: 534\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xQ2-JokHj-G9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Augmentationb              \n",
        "sometimes = lambda aug: iaa.Sometimes(0.6, aug)\n",
        "seq = iaa.Sequential([\n",
        "                      iaa.GaussianBlur(sigma=(0 , 1.0)),\n",
        "                      iaa.Sharpen(alpha=1, lightness=0),\n",
        "                      iaa.CoarseDropout(p=0.1, size_percent=0.15),\n",
        "                              sometimes(iaa.Affine(\n",
        "                                                    scale={\"x\": (0.8, 1.2), \"y\": (0.8, 1.2)},\n",
        "                                                    translate_percent={\"x\": (-0.2, 0.2), \"y\": (-0.2, 0.2)},\n",
        "                                                    rotate=(-30, 30),\n",
        "                                                    shear=(-16, 16)))\n",
        "                    ])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "46HG6L-MkKGe",
        "colab_type": "text"
      },
      "source": [
        "# Data Generators"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vKs1083SkGws",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 67
        },
        "outputId": "be58be22-f00a-4df5-cb7f-904f859dcaf2"
      },
      "source": [
        "# Data generators\n",
        "\n",
        "train_datagen = ImageDataGenerator(\n",
        "    rescale=1./255,\n",
        "    # preprocessing_function=seq.augment_image,\n",
        "    horizontal_flip=True,\n",
        "    vertical_flip=True)\n",
        "\n",
        "val_datagen = ImageDataGenerator(\n",
        "    rescale=1./255,\n",
        "    horizontal_flip=True,\n",
        "    vertical_flip=True)\n",
        "\n",
        "test_datagen = ImageDataGenerator(\n",
        "    rescale=1./255,\n",
        "    horizontal_flip=True,\n",
        "    vertical_flip=True) \n",
        "\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "    train_data_dir,\n",
        "    target_size=(img_height, img_width),\n",
        "    batch_size=32,\n",
        "    class_mode=\"categorical\")\n",
        "\n",
        "validation_generator = val_datagen.flow_from_directory(\n",
        "    validation_data_dir,\n",
        "    target_size=(img_height, img_width),\n",
        "    class_mode=\"categorical\")\n",
        "\n",
        "test_generator = test_datagen.flow_from_directory(\n",
        "    test_data_dir,\n",
        "    target_size=(img_height, img_width),\n",
        "    class_mode=\"categorical\")"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 280 images belonging to 2 classes.\n",
            "Found 80 images belonging to 2 classes.\n",
            "Found 40 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yv65Hrm7l9_c",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Callbacks\n",
        "checkpoint = ModelCheckpoint(\"f1.h5\", monitor='acc', verbose=1, save_best_only=True, save_weights_only=False, mode='auto', period=1)\n",
        "reduce_lr = ReduceLROnPlateau(monitor='loss', factor=0.1, patience=2, verbose=0, mode='auto', cooldown=0, min_lr=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8x_9K8YVmID1",
        "colab_type": "text"
      },
      "source": [
        "# Training Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e6wjSprimNMU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "56bab943-65ed-4f24-ed2e-4d06fe63c85d"
      },
      "source": [
        "batch_size = 32\n",
        "model.fit_generator(\n",
        "    train_generator,\n",
        "    steps_per_epoch=nb_train_samples // batch_size,\n",
        "    epochs=epochs,\n",
        "    validation_data=validation_generator,\n",
        "    validation_steps=nb_validation_samples // batch_size,\n",
        "    callbacks=[checkpoint, reduce_lr]\n",
        ")"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/math_grad.py:1424: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n",
            "\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1020: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.\n",
            "\n",
            "Epoch 1/100\n",
            "8/8 [==============================] - 24s 3s/step - loss: 0.6506 - acc: 0.6008 - val_loss: 0.6248 - val_acc: 0.5781\n",
            "\n",
            "Epoch 00001: acc improved from -inf to 0.60484, saving model to f1.h5\n",
            "Epoch 2/100\n",
            "8/8 [==============================] - 4s 532ms/step - loss: 0.5316 - acc: 0.7512 - val_loss: 0.5486 - val_acc: 0.7083\n",
            "\n",
            "Epoch 00002: acc improved from 0.60484 to 0.75000, saving model to f1.h5\n",
            "Epoch 3/100\n",
            "8/8 [==============================] - 3s 383ms/step - loss: 0.5091 - acc: 0.7398 - val_loss: 0.8327 - val_acc: 0.5833\n",
            "\n",
            "Epoch 00003: acc did not improve from 0.75000\n",
            "Epoch 4/100\n",
            "8/8 [==============================] - 3s 386ms/step - loss: 0.4929 - acc: 0.7735 - val_loss: 0.6180 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00004: acc improved from 0.75000 to 0.77419, saving model to f1.h5\n",
            "Epoch 5/100\n",
            "8/8 [==============================] - 3s 425ms/step - loss: 0.4590 - acc: 0.8084 - val_loss: 0.6663 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00005: acc improved from 0.77419 to 0.80645, saving model to f1.h5\n",
            "Epoch 6/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.4576 - acc: 0.7590 - val_loss: 0.6122 - val_acc: 0.5833\n",
            "\n",
            "Epoch 00006: acc did not improve from 0.80645\n",
            "Epoch 7/100\n",
            "8/8 [==============================] - 3s 393ms/step - loss: 0.4381 - acc: 0.8047 - val_loss: 0.6028 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00007: acc did not improve from 0.80645\n",
            "Epoch 8/100\n",
            "8/8 [==============================] - 3s 418ms/step - loss: 0.4067 - acc: 0.8268 - val_loss: 0.7663 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00008: acc improved from 0.80645 to 0.82661, saving model to f1.h5\n",
            "Epoch 9/100\n",
            "8/8 [==============================] - 3s 368ms/step - loss: 0.4078 - acc: 0.8072 - val_loss: 0.6238 - val_acc: 0.7083\n",
            "\n",
            "Epoch 00009: acc did not improve from 0.82661\n",
            "Epoch 10/100\n",
            "8/8 [==============================] - 3s 387ms/step - loss: 0.4280 - acc: 0.8245 - val_loss: 0.7439 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00010: acc improved from 0.82661 to 0.82661, saving model to f1.h5\n",
            "Epoch 11/100\n",
            "8/8 [==============================] - 3s 398ms/step - loss: 0.3851 - acc: 0.8256 - val_loss: 0.6669 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00011: acc improved from 0.82661 to 0.82661, saving model to f1.h5\n",
            "Epoch 12/100\n",
            "8/8 [==============================] - 3s 368ms/step - loss: 0.4041 - acc: 0.8319 - val_loss: 0.6263 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00012: acc improved from 0.82661 to 0.83065, saving model to f1.h5\n",
            "Epoch 13/100\n",
            "8/8 [==============================] - 3s 375ms/step - loss: 0.3514 - acc: 0.8621 - val_loss: 0.5868 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00013: acc improved from 0.83065 to 0.86290, saving model to f1.h5\n",
            "Epoch 14/100\n",
            "8/8 [==============================] - 3s 428ms/step - loss: 0.3778 - acc: 0.8390 - val_loss: 0.6329 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00014: acc did not improve from 0.86290\n",
            "Epoch 15/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.3852 - acc: 0.8516 - val_loss: 0.5976 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00015: acc did not improve from 0.86290\n",
            "Epoch 16/100\n",
            "8/8 [==============================] - 3s 366ms/step - loss: 0.3588 - acc: 0.8588 - val_loss: 0.5638 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00016: acc did not improve from 0.86290\n",
            "Epoch 17/100\n",
            "8/8 [==============================] - 3s 425ms/step - loss: 0.3689 - acc: 0.8789 - val_loss: 0.6342 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00017: acc improved from 0.86290 to 0.87891, saving model to f1.h5\n",
            "Epoch 18/100\n",
            "8/8 [==============================] - 3s 370ms/step - loss: 0.3652 - acc: 0.8542 - val_loss: 0.5434 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00018: acc did not improve from 0.87891\n",
            "Epoch 19/100\n",
            "8/8 [==============================] - 3s 390ms/step - loss: 0.3673 - acc: 0.8715 - val_loss: 0.5622 - val_acc: 0.7031\n",
            "\n",
            "Epoch 00019: acc did not improve from 0.87891\n",
            "Epoch 20/100\n",
            "8/8 [==============================] - 3s 411ms/step - loss: 0.3486 - acc: 0.8817 - val_loss: 0.6550 - val_acc: 0.5625\n",
            "\n",
            "Epoch 00020: acc improved from 0.87891 to 0.88306, saving model to f1.h5\n",
            "Epoch 21/100\n",
            "8/8 [==============================] - 3s 372ms/step - loss: 0.3873 - acc: 0.8284 - val_loss: 0.5378 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00021: acc did not improve from 0.88306\n",
            "Epoch 22/100\n",
            "8/8 [==============================] - 3s 381ms/step - loss: 0.4093 - acc: 0.8206 - val_loss: 0.5639 - val_acc: 0.6719\n",
            "\n",
            "Epoch 00022: acc did not improve from 0.88306\n",
            "Epoch 23/100\n",
            "8/8 [==============================] - 3s 422ms/step - loss: 0.3695 - acc: 0.8722 - val_loss: 0.7017 - val_acc: 0.5417\n",
            "\n",
            "Epoch 00023: acc did not improve from 0.88306\n",
            "Epoch 24/100\n",
            "8/8 [==============================] - 3s 388ms/step - loss: 0.3578 - acc: 0.8750 - val_loss: 0.4886 - val_acc: 0.7708\n",
            "\n",
            "Epoch 00024: acc did not improve from 0.88306\n",
            "Epoch 25/100\n",
            "8/8 [==============================] - 3s 387ms/step - loss: 0.3743 - acc: 0.8445 - val_loss: 0.5926 - val_acc: 0.6094\n",
            "\n",
            "Epoch 00025: acc did not improve from 0.88306\n",
            "Epoch 26/100\n",
            "8/8 [==============================] - 3s 425ms/step - loss: 0.3839 - acc: 0.8605 - val_loss: 0.5774 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00026: acc did not improve from 0.88306\n",
            "Epoch 27/100\n",
            "8/8 [==============================] - 3s 378ms/step - loss: 0.3617 - acc: 0.8510 - val_loss: 0.5850 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00027: acc did not improve from 0.88306\n",
            "Epoch 28/100\n",
            "8/8 [==============================] - 3s 386ms/step - loss: 0.3617 - acc: 0.8644 - val_loss: 0.5525 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00028: acc did not improve from 0.88306\n",
            "Epoch 29/100\n",
            "8/8 [==============================] - 3s 410ms/step - loss: 0.3727 - acc: 0.8605 - val_loss: 0.6983 - val_acc: 0.5417\n",
            "\n",
            "Epoch 00029: acc did not improve from 0.88306\n",
            "Epoch 30/100\n",
            "8/8 [==============================] - 3s 374ms/step - loss: 0.3757 - acc: 0.8420 - val_loss: 0.4971 - val_acc: 0.7292\n",
            "\n",
            "Epoch 00030: acc did not improve from 0.88306\n",
            "Epoch 31/100\n",
            "8/8 [==============================] - 3s 385ms/step - loss: 0.3339 - acc: 0.8907 - val_loss: 0.5826 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00031: acc improved from 0.88306 to 0.89113, saving model to f1.h5\n",
            "Epoch 32/100\n",
            "8/8 [==============================] - 3s 433ms/step - loss: 0.3933 - acc: 0.8320 - val_loss: 0.5232 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00032: acc did not improve from 0.89113\n",
            "Epoch 33/100\n",
            "8/8 [==============================] - 3s 374ms/step - loss: 0.3743 - acc: 0.8616 - val_loss: 0.6417 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00033: acc did not improve from 0.89113\n",
            "Epoch 34/100\n",
            "8/8 [==============================] - 3s 378ms/step - loss: 0.3529 - acc: 0.8706 - val_loss: 0.5579 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00034: acc did not improve from 0.89113\n",
            "Epoch 35/100\n",
            "8/8 [==============================] - 3s 434ms/step - loss: 0.3859 - acc: 0.8242 - val_loss: 0.6439 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00035: acc did not improve from 0.89113\n",
            "Epoch 36/100\n",
            "8/8 [==============================] - 3s 370ms/step - loss: 0.3846 - acc: 0.8688 - val_loss: 0.5837 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00036: acc did not improve from 0.89113\n",
            "Epoch 37/100\n",
            "8/8 [==============================] - 3s 378ms/step - loss: 0.3671 - acc: 0.8464 - val_loss: 0.6163 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00037: acc did not improve from 0.89113\n",
            "Epoch 38/100\n",
            "8/8 [==============================] - 3s 398ms/step - loss: 0.3797 - acc: 0.8397 - val_loss: 0.4662 - val_acc: 0.7917\n",
            "\n",
            "Epoch 00038: acc did not improve from 0.89113\n",
            "Epoch 39/100\n",
            "8/8 [==============================] - 3s 383ms/step - loss: 0.3601 - acc: 0.8711 - val_loss: 0.6776 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00039: acc did not improve from 0.89113\n",
            "Epoch 40/100\n",
            "8/8 [==============================] - 3s 379ms/step - loss: 0.3491 - acc: 0.8934 - val_loss: 0.5888 - val_acc: 0.7031\n",
            "\n",
            "Epoch 00040: acc improved from 0.89113 to 0.89516, saving model to f1.h5\n",
            "Epoch 41/100\n",
            "8/8 [==============================] - 3s 421ms/step - loss: 0.4049 - acc: 0.8425 - val_loss: 0.6085 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00041: acc did not improve from 0.89516\n",
            "Epoch 42/100\n",
            "8/8 [==============================] - 3s 361ms/step - loss: 0.3525 - acc: 0.8569 - val_loss: 0.5967 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00042: acc did not improve from 0.89516\n",
            "Epoch 43/100\n",
            "8/8 [==============================] - 3s 387ms/step - loss: 0.3954 - acc: 0.8477 - val_loss: 0.6252 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00043: acc did not improve from 0.89516\n",
            "Epoch 44/100\n",
            "8/8 [==============================] - 3s 420ms/step - loss: 0.3766 - acc: 0.8386 - val_loss: 0.5656 - val_acc: 0.7292\n",
            "\n",
            "Epoch 00044: acc did not improve from 0.89516\n",
            "Epoch 45/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.3563 - acc: 0.8884 - val_loss: 0.5963 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00045: acc did not improve from 0.89516\n",
            "Epoch 46/100\n",
            "8/8 [==============================] - 3s 389ms/step - loss: 0.3617 - acc: 0.8633 - val_loss: 0.6118 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00046: acc did not improve from 0.89516\n",
            "Epoch 47/100\n",
            "8/8 [==============================] - 3s 396ms/step - loss: 0.3698 - acc: 0.8644 - val_loss: 0.5824 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00047: acc did not improve from 0.89516\n",
            "Epoch 48/100\n",
            "8/8 [==============================] - 3s 388ms/step - loss: 0.3648 - acc: 0.8633 - val_loss: 0.5900 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00048: acc did not improve from 0.89516\n",
            "Epoch 49/100\n",
            "8/8 [==============================] - 3s 390ms/step - loss: 0.3778 - acc: 0.8531 - val_loss: 0.5691 - val_acc: 0.7031\n",
            "\n",
            "Epoch 00049: acc did not improve from 0.89516\n",
            "Epoch 50/100\n",
            "8/8 [==============================] - 3s 415ms/step - loss: 0.3906 - acc: 0.8581 - val_loss: 0.7154 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00050: acc did not improve from 0.89516\n",
            "Epoch 51/100\n",
            "8/8 [==============================] - 3s 381ms/step - loss: 0.3387 - acc: 0.8750 - val_loss: 0.4905 - val_acc: 0.7500\n",
            "\n",
            "Epoch 00051: acc did not improve from 0.89516\n",
            "Epoch 52/100\n",
            "8/8 [==============================] - 3s 382ms/step - loss: 0.3951 - acc: 0.8402 - val_loss: 0.6461 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00052: acc did not improve from 0.89516\n",
            "Epoch 53/100\n",
            "8/8 [==============================] - 3s 430ms/step - loss: 0.3561 - acc: 0.8598 - val_loss: 0.4918 - val_acc: 0.7500\n",
            "\n",
            "Epoch 00053: acc did not improve from 0.89516\n",
            "Epoch 54/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.3560 - acc: 0.8762 - val_loss: 0.5776 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00054: acc did not improve from 0.89516\n",
            "Epoch 55/100\n",
            "8/8 [==============================] - 3s 382ms/step - loss: 0.3789 - acc: 0.8621 - val_loss: 0.4974 - val_acc: 0.7344\n",
            "\n",
            "Epoch 00055: acc did not improve from 0.89516\n",
            "Epoch 56/100\n",
            "8/8 [==============================] - 3s 399ms/step - loss: 0.3959 - acc: 0.8441 - val_loss: 0.6750 - val_acc: 0.7083\n",
            "\n",
            "Epoch 00056: acc did not improve from 0.89516\n",
            "Epoch 57/100\n",
            "8/8 [==============================] - 3s 374ms/step - loss: 0.3701 - acc: 0.8609 - val_loss: 0.6305 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00057: acc did not improve from 0.89516\n",
            "Epoch 58/100\n",
            "8/8 [==============================] - 3s 399ms/step - loss: 0.3827 - acc: 0.8516 - val_loss: 0.6392 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00058: acc did not improve from 0.89516\n",
            "Epoch 59/100\n",
            "8/8 [==============================] - 3s 427ms/step - loss: 0.3722 - acc: 0.8738 - val_loss: 0.5896 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00059: acc did not improve from 0.89516\n",
            "Epoch 60/100\n",
            "8/8 [==============================] - 3s 362ms/step - loss: 0.3756 - acc: 0.8683 - val_loss: 0.5677 - val_acc: 0.7083\n",
            "\n",
            "Epoch 00060: acc did not improve from 0.89516\n",
            "Epoch 61/100\n",
            "8/8 [==============================] - 3s 394ms/step - loss: 0.3711 - acc: 0.8594 - val_loss: 0.6357 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00061: acc did not improve from 0.89516\n",
            "Epoch 62/100\n",
            "8/8 [==============================] - 3s 407ms/step - loss: 0.3855 - acc: 0.8581 - val_loss: 0.5485 - val_acc: 0.7083\n",
            "\n",
            "Epoch 00062: acc did not improve from 0.89516\n",
            "Epoch 63/100\n",
            "8/8 [==============================] - 3s 383ms/step - loss: 0.3683 - acc: 0.8516 - val_loss: 0.6070 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00063: acc did not improve from 0.89516\n",
            "Epoch 64/100\n",
            "8/8 [==============================] - 3s 379ms/step - loss: 0.3524 - acc: 0.8884 - val_loss: 0.5864 - val_acc: 0.6719\n",
            "\n",
            "Epoch 00064: acc did not improve from 0.89516\n",
            "Epoch 65/100\n",
            "8/8 [==============================] - 3s 402ms/step - loss: 0.3811 - acc: 0.8374 - val_loss: 0.5664 - val_acc: 0.7292\n",
            "\n",
            "Epoch 00065: acc did not improve from 0.89516\n",
            "Epoch 66/100\n",
            "8/8 [==============================] - 3s 371ms/step - loss: 0.3763 - acc: 0.8487 - val_loss: 0.6612 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00066: acc did not improve from 0.89516\n",
            "Epoch 67/100\n",
            "8/8 [==============================] - 3s 404ms/step - loss: 0.3803 - acc: 0.8695 - val_loss: 0.6537 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00067: acc did not improve from 0.89516\n",
            "Epoch 68/100\n",
            "8/8 [==============================] - 4s 438ms/step - loss: 0.3686 - acc: 0.8594 - val_loss: 0.4865 - val_acc: 0.7708\n",
            "\n",
            "Epoch 00068: acc did not improve from 0.89516\n",
            "Epoch 69/100\n",
            "8/8 [==============================] - 3s 372ms/step - loss: 0.3587 - acc: 0.8632 - val_loss: 0.6407 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00069: acc did not improve from 0.89516\n",
            "Epoch 70/100\n",
            "8/8 [==============================] - 3s 380ms/step - loss: 0.3968 - acc: 0.8346 - val_loss: 0.6357 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00070: acc did not improve from 0.89516\n",
            "Epoch 71/100\n",
            "8/8 [==============================] - 3s 420ms/step - loss: 0.3352 - acc: 0.8778 - val_loss: 0.5345 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00071: acc did not improve from 0.89516\n",
            "Epoch 72/100\n",
            "8/8 [==============================] - 3s 382ms/step - loss: 0.3638 - acc: 0.8438 - val_loss: 0.6112 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00072: acc did not improve from 0.89516\n",
            "Epoch 73/100\n",
            "8/8 [==============================] - 3s 385ms/step - loss: 0.3909 - acc: 0.8268 - val_loss: 0.5517 - val_acc: 0.6719\n",
            "\n",
            "Epoch 00073: acc did not improve from 0.89516\n",
            "Epoch 74/100\n",
            "8/8 [==============================] - 3s 403ms/step - loss: 0.3531 - acc: 0.8817 - val_loss: 0.7196 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00074: acc did not improve from 0.89516\n",
            "Epoch 75/100\n",
            "8/8 [==============================] - 3s 374ms/step - loss: 0.3692 - acc: 0.8538 - val_loss: 0.5883 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00075: acc did not improve from 0.89516\n",
            "Epoch 76/100\n",
            "8/8 [==============================] - 3s 398ms/step - loss: 0.3739 - acc: 0.8516 - val_loss: 0.6212 - val_acc: 0.7031\n",
            "\n",
            "Epoch 00076: acc did not improve from 0.89516\n",
            "Epoch 77/100\n",
            "8/8 [==============================] - 3s 411ms/step - loss: 0.3951 - acc: 0.8345 - val_loss: 0.5579 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00077: acc did not improve from 0.89516\n",
            "Epoch 78/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.3520 - acc: 0.8789 - val_loss: 0.6357 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00078: acc did not improve from 0.89516\n",
            "Epoch 79/100\n",
            "8/8 [==============================] - 3s 391ms/step - loss: 0.4091 - acc: 0.8242 - val_loss: 0.5718 - val_acc: 0.7031\n",
            "\n",
            "Epoch 00079: acc did not improve from 0.89516\n",
            "Epoch 80/100\n",
            "8/8 [==============================] - 3s 425ms/step - loss: 0.3666 - acc: 0.8805 - val_loss: 0.6966 - val_acc: 0.5833\n",
            "\n",
            "Epoch 00080: acc did not improve from 0.89516\n",
            "Epoch 81/100\n",
            "8/8 [==============================] - 3s 372ms/step - loss: 0.3668 - acc: 0.8503 - val_loss: 0.5386 - val_acc: 0.7292\n",
            "\n",
            "Epoch 00081: acc did not improve from 0.89516\n",
            "Epoch 82/100\n",
            "8/8 [==============================] - 3s 380ms/step - loss: 0.3763 - acc: 0.8542 - val_loss: 0.6025 - val_acc: 0.6562\n",
            "\n",
            "Epoch 00082: acc did not improve from 0.89516\n",
            "Epoch 83/100\n",
            "8/8 [==============================] - 3s 406ms/step - loss: 0.3524 - acc: 0.8801 - val_loss: 0.6694 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00083: acc did not improve from 0.89516\n",
            "Epoch 84/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.3529 - acc: 0.8766 - val_loss: 0.5500 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00084: acc did not improve from 0.89516\n",
            "Epoch 85/100\n",
            "8/8 [==============================] - 3s 392ms/step - loss: 0.3923 - acc: 0.8555 - val_loss: 0.5797 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00085: acc did not improve from 0.89516\n",
            "Epoch 86/100\n",
            "8/8 [==============================] - 3s 436ms/step - loss: 0.3682 - acc: 0.8515 - val_loss: 0.6625 - val_acc: 0.6042\n",
            "\n",
            "Epoch 00086: acc did not improve from 0.89516\n",
            "Epoch 87/100\n",
            "8/8 [==============================] - 3s 362ms/step - loss: 0.4018 - acc: 0.8491 - val_loss: 0.5568 - val_acc: 0.7292\n",
            "\n",
            "Epoch 00087: acc did not improve from 0.89516\n",
            "Epoch 88/100\n",
            "8/8 [==============================] - 3s 381ms/step - loss: 0.3702 - acc: 0.8542 - val_loss: 0.6327 - val_acc: 0.6406\n",
            "\n",
            "Epoch 00088: acc did not improve from 0.89516\n",
            "Epoch 89/100\n",
            "8/8 [==============================] - 3s 431ms/step - loss: 0.3814 - acc: 0.8555 - val_loss: 0.5793 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00089: acc did not improve from 0.89516\n",
            "Epoch 90/100\n",
            "8/8 [==============================] - 3s 375ms/step - loss: 0.3890 - acc: 0.8346 - val_loss: 0.5729 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00090: acc did not improve from 0.89516\n",
            "Epoch 91/100\n",
            "8/8 [==============================] - 3s 381ms/step - loss: 0.3689 - acc: 0.8515 - val_loss: 0.6004 - val_acc: 0.6719\n",
            "\n",
            "Epoch 00091: acc did not improve from 0.89516\n",
            "Epoch 92/100\n",
            "8/8 [==============================] - 3s 404ms/step - loss: 0.3542 - acc: 0.8683 - val_loss: 0.6917 - val_acc: 0.6250\n",
            "\n",
            "Epoch 00092: acc did not improve from 0.89516\n",
            "Epoch 93/100\n",
            "8/8 [==============================] - 3s 373ms/step - loss: 0.3540 - acc: 0.8828 - val_loss: 0.5245 - val_acc: 0.7500\n",
            "\n",
            "Epoch 00093: acc did not improve from 0.89516\n",
            "Epoch 94/100\n",
            "8/8 [==============================] - 3s 391ms/step - loss: 0.3832 - acc: 0.8398 - val_loss: 0.5656 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00094: acc did not improve from 0.89516\n",
            "Epoch 95/100\n",
            "8/8 [==============================] - 3s 407ms/step - loss: 0.3620 - acc: 0.8509 - val_loss: 0.7030 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00095: acc did not improve from 0.89516\n",
            "Epoch 96/100\n",
            "8/8 [==============================] - 3s 383ms/step - loss: 0.3673 - acc: 0.8242 - val_loss: 0.5655 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00096: acc did not improve from 0.89516\n",
            "Epoch 97/100\n",
            "8/8 [==============================] - 3s 381ms/step - loss: 0.3457 - acc: 0.8801 - val_loss: 0.6404 - val_acc: 0.6406\n",
            "\n",
            "Epoch 00097: acc did not improve from 0.89516\n",
            "Epoch 98/100\n",
            "8/8 [==============================] - 3s 428ms/step - loss: 0.4042 - acc: 0.8155 - val_loss: 0.5190 - val_acc: 0.6667\n",
            "\n",
            "Epoch 00098: acc did not improve from 0.89516\n",
            "Epoch 99/100\n",
            "8/8 [==============================] - 3s 372ms/step - loss: 0.3502 - acc: 0.8738 - val_loss: 0.6529 - val_acc: 0.6458\n",
            "\n",
            "Epoch 00099: acc did not improve from 0.89516\n",
            "Epoch 100/100\n",
            "8/8 [==============================] - 3s 381ms/step - loss: 0.3471 - acc: 0.8891 - val_loss: 0.5197 - val_acc: 0.6875\n",
            "\n",
            "Epoch 00100: acc did not improve from 0.89516\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f684b7b1400>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J3pSrVt4oPw5",
        "colab_type": "text"
      },
      "source": [
        "# Prediction on Test Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D3EVlKWhoS2m",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Callbacks\n",
        "checkpoint = ModelCheckpoint(\"f1.h5\", monitor='acc', verbose=1, save_best_only=True, save_weights_only=False, mode='auto', period=1)\n",
        "reduce_lr = ReduceLROnPlateau(monitor='loss', factor=0.1, patience=2, verbose=0, mode='auto', cooldown=0, min_lr=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G1O3GM5noZG0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e69bdb98-c567-4a06-83bc-6427f2bf02ed"
      },
      "source": [
        "batch_size = 10\n",
        "pred = model.predict_generator(test_generator, \n",
        "                               steps=get_steps(n_test_samples, batch_size), \n",
        "                               callbacks=[checkpoint, reduce_lr],  \n",
        "                               verbose=1)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "4/4 [==============================] - 2s 408ms/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "knnWG6DAohQR",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "23a01c5f-5862-454b-e91f-11d7037608ea"
      },
      "source": [
        "pred"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.66776055, 0.33223945],\n",
              "       [0.39407468, 0.6059253 ],\n",
              "       [0.31803977, 0.6819602 ],\n",
              "       [0.816509  , 0.183491  ],\n",
              "       [0.66935116, 0.3306488 ],\n",
              "       [0.49172097, 0.508279  ],\n",
              "       [0.5092242 , 0.49077582],\n",
              "       [0.2921776 , 0.70782244],\n",
              "       [0.43954197, 0.56045806],\n",
              "       [0.74468166, 0.2553184 ],\n",
              "       [0.85571307, 0.14428699],\n",
              "       [0.5027314 , 0.4972686 ],\n",
              "       [0.02267935, 0.9773207 ],\n",
              "       [0.79438525, 0.20561472],\n",
              "       [0.05778668, 0.9422133 ],\n",
              "       [0.04669693, 0.95330304],\n",
              "       [0.5223356 , 0.47766447],\n",
              "       [0.5796036 , 0.42039636],\n",
              "       [0.7943506 , 0.20564936],\n",
              "       [0.37829894, 0.6217011 ],\n",
              "       [0.47306046, 0.5269396 ],\n",
              "       [0.46297646, 0.5370236 ],\n",
              "       [0.4420952 , 0.55790484],\n",
              "       [0.7734003 , 0.22659972],\n",
              "       [0.02141242, 0.97858757],\n",
              "       [0.44251254, 0.5574875 ],\n",
              "       [0.93071204, 0.06928799],\n",
              "       [0.04541377, 0.95458627],\n",
              "       [0.11802074, 0.8819793 ],\n",
              "       [0.66649526, 0.33350477],\n",
              "       [0.27289146, 0.7271086 ],\n",
              "       [0.2238025 , 0.77619743],\n",
              "       [0.4303799 , 0.56962013],\n",
              "       [0.7971196 , 0.2028804 ],\n",
              "       [0.07328009, 0.92671996],\n",
              "       [0.68595076, 0.3140492 ],\n",
              "       [0.66007024, 0.3399298 ],\n",
              "       [0.4405055 , 0.5594945 ],\n",
              "       [0.76036346, 0.23963659],\n",
              "       [0.38417894, 0.61582106],\n",
              "       [0.7335434 , 0.2664566 ],\n",
              "       [0.0567649 , 0.94323516],\n",
              "       [0.29018897, 0.70981103],\n",
              "       [0.81568766, 0.18431228],\n",
              "       [0.68595064, 0.31404942],\n",
              "       [0.79711956, 0.20288043],\n",
              "       [0.3696688 , 0.63033116],\n",
              "       [0.38417843, 0.61582154],\n",
              "       [0.5540041 , 0.44599596],\n",
              "       [0.08542068, 0.9145793 ],\n",
              "       [0.6484944 , 0.35150558],\n",
              "       [0.13985848, 0.8601416 ],\n",
              "       [0.2608718 , 0.7391282 ],\n",
              "       [0.26963928, 0.7303607 ],\n",
              "       [0.02408296, 0.97591704],\n",
              "       [0.79438525, 0.20561472],\n",
              "       [0.49172097, 0.508279  ],\n",
              "       [0.3733957 , 0.6266043 ],\n",
              "       [0.05465395, 0.9453461 ],\n",
              "       [0.67301214, 0.32698786],\n",
              "       [0.5472141 , 0.45278594],\n",
              "       [0.1433827 , 0.8566173 ],\n",
              "       [0.29871556, 0.7012844 ],\n",
              "       [0.07328001, 0.92671996],\n",
              "       [0.4322774 , 0.5677226 ],\n",
              "       [0.7603633 , 0.23963672],\n",
              "       [0.4377863 , 0.56221366],\n",
              "       [0.384207  , 0.615793  ],\n",
              "       [0.816509  , 0.183491  ],\n",
              "       [0.43954197, 0.56045806],\n",
              "       [0.93071204, 0.06928799],\n",
              "       [0.23852932, 0.7614707 ],\n",
              "       [0.67726403, 0.32273597],\n",
              "       [0.02514786, 0.9748522 ],\n",
              "       [0.72219926, 0.2778007 ],\n",
              "       [0.47306085, 0.52693915],\n",
              "       [0.5796038 , 0.42039618],\n",
              "       [0.66367626, 0.3363237 ],\n",
              "       [0.76062506, 0.23937492],\n",
              "       [0.28693843, 0.71306163]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3zHpR9jaop3L",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "5bbceb6a-0751-45fa-9ffc-3c95aaa193db"
      },
      "source": [
        "threshold = 0.6\n",
        "result = []\n",
        "for x, y in pred:\n",
        "  if x >= 0.6:\n",
        "    result.append(0)\n",
        "  else:\n",
        "    result.append(1)\n",
        "\n",
        "print(result)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 0, 0, 1]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f8ZQB0i2tvzO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b9452c5e-0651-4304-a5b0-1794a4faa1e8"
      },
      "source": [
        "print(result.count(1))"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "52\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G8I3eOydt4St",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e7378f59-1c4b-422f-94b8-3503c76286ad"
      },
      "source": [
        "print(result.count(0))"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "28\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RBY8SybZt-NT",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "94c9fb8a-f043-4f7a-adff-46f72b652f78"
      },
      "source": [
        "print(len(result))"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "80\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}