Surface_Defect_Detection.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ardbramantyo/DeepLearning-Defect-Detection/blob/main/Surface_Defect_Detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CKqhIIWnHyR_"
      },
      "source": [
        "Created by: Ardian Kris Bramantyo\n",
        "\n",
        "Date: May 20, 2022"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dJY1JrDhKDAk"
      },
      "source": [
        "## Import Libraries"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "OCIpqt4I8M6v"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "import zipfile\n",
        "import cv2\n",
        "from skimage import io\n",
        "from tensorflow.python.keras import Sequential\n",
        "from tensorflow.keras import layers, optimizers\n",
        "from tensorflow.keras.applications import DenseNet121\n",
        "from tensorflow.keras.applications.resnet50 import ResNet50\n",
        "from tensorflow.keras.layers import *\n",
        "from tensorflow.keras.models import Model, load_model\n",
        "from tensorflow.keras.initializers import glorot_uniform\n",
        "from tensorflow.keras.utils import plot_model\n",
        "from tensorflow.keras.callbacks import ReduceLROnPlateau, EarlyStopping, ModelCheckpoint, LearningRateScheduler\n",
        "from IPython.display import display\n",
        "from tensorflow.keras import backend as K\n",
        "from sklearn.preprocessing import StandardScaler, normalize\n",
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0X3wRRfYKGK-"
      },
      "source": [
        "## Import files from GDrive"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uq-JkFgsLjIl",
        "outputId": "643b9382-3bdd-432e-e8bf-19172bddb509"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "from google.colab import files\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "RlFO_lK4L0Nw"
      },
      "outputs": [],
      "source": [
        "df_class = pd.read_csv('/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/train.csv')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P-glR6rCOjzV",
        "outputId": "3efd4a4b-cefd-435c-fa6d-39754dfe0a8e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 5748 entries, 0 to 5747\n",
            "Data columns (total 3 columns):\n",
            " #   Column         Non-Null Count  Dtype \n",
            "---  ------         --------------  ----- \n",
            " 0   ImageId        5748 non-null   object\n",
            " 1   ClassId        5748 non-null   int64 \n",
            " 2   EncodedPixels  5748 non-null   object\n",
            "dtypes: int64(1), object(2)\n",
            "memory usage: 134.8+ KB\n"
          ]
        }
      ],
      "source": [
        "df_class.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "QTKRAw33N3ND",
        "outputId": "f2f1fb97-c3bd-4d36-f5ae-76235d21ef27"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         ImageId  ClassId                                      EncodedPixels\n",
              "0  d2291de5c.jpg        1  147963 3 148213 9 148461 18 148711 24 148965 2...\n",
              "1  78416c3d0.jpg        3  54365 3 54621 7 54877 10 55133 12 55388 14 556...\n",
              "2  2283f2183.jpg        3  201217 43 201473 128 201729 213 201985 5086 20...\n",
              "3  f0dc068a8.jpg        3  159207 26 159412 77 159617 128 159822 179 1600...\n",
              "4  00d639396.jpg        3  229356 17 229595 34 229850 36 230105 37 230360..."
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-58d7ec09-ed9e-464b-8def-0c54d516f212\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ImageId</th>\n",
              "      <th>ClassId</th>\n",
              "      <th>EncodedPixels</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>d2291de5c.jpg</td>\n",
              "      <td>1</td>\n",
              "      <td>147963 3 148213 9 148461 18 148711 24 148965 2...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>78416c3d0.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>54365 3 54621 7 54877 10 55133 12 55388 14 556...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2283f2183.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>201217 43 201473 128 201729 213 201985 5086 20...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>f0dc068a8.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>159207 26 159412 77 159617 128 159822 179 1600...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>00d639396.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>229356 17 229595 34 229850 36 230105 37 230360...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-58d7ec09-ed9e-464b-8def-0c54d516f212')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-58d7ec09-ed9e-464b-8def-0c54d516f212 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-58d7ec09-ed9e-464b-8def-0c54d516f212');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "df_class.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "6kM-RhJRN468"
      },
      "outputs": [],
      "source": [
        "df_all = pd.read_csv('/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/defect_and_no_defect.csv')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DqTdw9hVOTE2",
        "outputId": "387b435a-d163-4a9a-9bcc-e4c46b5947a8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 12997 entries, 0 to 12996\n",
            "Data columns (total 2 columns):\n",
            " #   Column   Non-Null Count  Dtype \n",
            "---  ------   --------------  ----- \n",
            " 0   ImageID  12997 non-null  object\n",
            " 1   label    12997 non-null  int64 \n",
            "dtypes: int64(1), object(1)\n",
            "memory usage: 203.2+ KB\n"
          ]
        }
      ],
      "source": [
        "df_all.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V1H4RmrVPLZf"
      },
      "source": [
        "## Data Visualization"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "LQkikSbTOYO3"
      },
      "outputs": [],
      "source": [
        "df_class['mask'] = df_class['ClassId'].map(lambda x:1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "YAZ-CqPPQIEw",
        "outputId": "e46abe4a-9d80-4078-878a-d3f5cb9dd2b5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         ImageId  ClassId                                      EncodedPixels  \\\n",
              "0  d2291de5c.jpg        1  147963 3 148213 9 148461 18 148711 24 148965 2...   \n",
              "1  78416c3d0.jpg        3  54365 3 54621 7 54877 10 55133 12 55388 14 556...   \n",
              "2  2283f2183.jpg        3  201217 43 201473 128 201729 213 201985 5086 20...   \n",
              "3  f0dc068a8.jpg        3  159207 26 159412 77 159617 128 159822 179 1600...   \n",
              "4  00d639396.jpg        3  229356 17 229595 34 229850 36 230105 37 230360...   \n",
              "5  17d02873a.jpg        3  254980 43 255236 127 255492 211 255748 253 256...   \n",
              "6  47b5ab1bd.jpg        3  128976 8 129230 12 129484 16 129739 23 129995 ...   \n",
              "7  a6ecee828.jpg        3  179011 27 179126 73 179259 39 179375 80 179497...   \n",
              "8  11aaf18e2.jpg        3  303235 2 303489 7 303743 9 303997 11 304181 2 ...   \n",
              "9  cdf669a1f.jpg        4  310246 11 310499 25 310753 28 311007 31 311262...   \n",
              "\n",
              "   mask  \n",
              "0     1  \n",
              "1     1  \n",
              "2     1  \n",
              "3     1  \n",
              "4     1  \n",
              "5     1  \n",
              "6     1  \n",
              "7     1  \n",
              "8     1  \n",
              "9     1  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-76de8f12-e851-4e4e-b934-674880c9524e\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ImageId</th>\n",
              "      <th>ClassId</th>\n",
              "      <th>EncodedPixels</th>\n",
              "      <th>mask</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>d2291de5c.jpg</td>\n",
              "      <td>1</td>\n",
              "      <td>147963 3 148213 9 148461 18 148711 24 148965 2...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>78416c3d0.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>54365 3 54621 7 54877 10 55133 12 55388 14 556...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2283f2183.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>201217 43 201473 128 201729 213 201985 5086 20...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>f0dc068a8.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>159207 26 159412 77 159617 128 159822 179 1600...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>00d639396.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>229356 17 229595 34 229850 36 230105 37 230360...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>17d02873a.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>254980 43 255236 127 255492 211 255748 253 256...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>47b5ab1bd.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>128976 8 129230 12 129484 16 129739 23 129995 ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>a6ecee828.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>179011 27 179126 73 179259 39 179375 80 179497...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>11aaf18e2.jpg</td>\n",
              "      <td>3</td>\n",
              "      <td>303235 2 303489 7 303743 9 303997 11 304181 2 ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>cdf669a1f.jpg</td>\n",
              "      <td>4</td>\n",
              "      <td>310246 11 310499 25 310753 28 311007 31 311262...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-76de8f12-e851-4e4e-b934-674880c9524e')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-76de8f12-e851-4e4e-b934-674880c9524e button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-76de8f12-e851-4e4e-b934-674880c9524e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "df_class.head(10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 695
        },
        "id": "pl1AuAYXQP-q",
        "outputId": "13856488-f7bc-42fe-e647-c466dba2820f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Number of Images per Class')"
            ]
          },
          "metadata": {},
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "sns.countplot(df_class['ClassId'])\n",
        "plt.ylabel('Number of images per defect')\n",
        "plt.xlabel('ClassID')\n",
        "plt.title('Number of Images per Class')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "FUJJQ8qLREMf"
      },
      "outputs": [],
      "source": [
        "# Some of the images are classified more than 1 defect \n",
        "defect_type = df_class.groupby(['ImageId'])['mask'].sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c4N0RKQISR4C",
        "outputId": "f2595b26-c577-4ec9-d99c-846324ea5f90"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ImageId\n",
              "0002cc93b.jpg    1\n",
              "0007a71bf.jpg    1\n",
              "000a4bcdd.jpg    1\n",
              "000f6bf48.jpg    1\n",
              "0014fce06.jpg    1\n",
              "                ..\n",
              "ffcf72ecf.jpg    1\n",
              "fff02e9c5.jpg    1\n",
              "fffe98443.jpg    1\n",
              "ffff4eaa8.jpg    1\n",
              "ffffd67df.jpg    1\n",
              "Name: mask, Length: 5474, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ],
      "source": [
        "defect_type"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1afvqOo4STeA",
        "outputId": "e3447496-5f89-4aaa-a485-9469376b1c95"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    5201\n",
              "2     272\n",
              "3       1\n",
              "Name: mask, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ],
      "source": [
        "defect_type.value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 639
        },
        "id": "VHVJkYUJSZYq",
        "outputId": "186a6110-0cfb-4806-d68b-ad9406a1c043"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Number od defects in Image')"
            ]
          },
          "metadata": {},
          "execution_count": 14
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "sns.barplot(x= defect_type.value_counts().index, y = defect_type.value_counts())\n",
        "plt.xlabel('ClassID')\n",
        "plt.title('Number od defects in Image')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hmzv38P-TNGq",
        "outputId": "c19aa1ed-5144-4c55-8363-78a6f256ccca"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5748, 4)"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "df_class.shape #Only for defect images"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5BFhj3VWTbA-",
        "outputId": "fc823dfa-e5bf-4d54-b216-cb757c6ba60d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(12997, 2)"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ],
      "source": [
        "df_all.shape #All images: defect-free and deffected"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 639
        },
        "id": "b5WdlBwXTfIK",
        "outputId": "44109b38-fd6e-4b11-b0c3-e077c59be7d2"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Defect and Non-Defect Images')"
            ]
          },
          "metadata": {},
          "execution_count": 17
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "sns.barplot(x = df_all.label.value_counts().index, y = df_all.label.value_counts())\n",
        "plt.ylabel('Number of Images')\n",
        "plt.xlabel('0 - Non-defect      1 - Defect')\n",
        "plt.title('Defect and Non-Defect Images')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PgPPxgctVRc8"
      },
      "source": [
        "Visualize Images with Defects"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "bCusDtsMUibH",
        "outputId": "76d186c3-9b8f-4f26-afcf-a698df56c6b7"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAABkCAYAAAB99XB/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO29W4xs2Xnf919167p0Vd+7z2VmOBxiFIl5sMQICQUHASHGkqwYFgIYBhUjph0bAhwHsBMDCRUBAfIWJYGRCEgsK7EDJ1F8iazEgiRLcBj5wQ+hKCXWhSPOcMTLkDPnzDnTfbqru+6XnYeq36r/Xqf6nD70nOnmYH9Ao7urdu291re+6//71qqQZZkKKqigggr6cFHpugdQUEEFFVTQ+0+FcS+ooIIK+hBSYdwLKqiggj6EVBj3ggoqqKAPIRXGvaCCCiroQ0iFcS+ooIIK+hBSYdwLKqiggj6EVBj3ggq6hEII/2sI4V4IoRtCeCOE8Beve0wFFXRVCsUmpoIKWk8hhH9Z0ptZlo1CCN8t6Z9K+reyLPvt6x1ZQQU9nYrIvaCCLqEsy76UZdmIf5c/H7vGIRVU0JWpMO4FFfQECiH89yGEvqQvS7on6VeveUgFFXQlKmCZggp6CoUQypJ+QNKnJP10lmWT6x1RQQU9nYrIvaCCnkJZls2yLPtnkl6Q9JeuezwFFXQVKox7QQVdnSoqMPeCvkOoMO4FFbSGQgiHIYTPhBA2QwjlEMIPS/pxSZ+/7rEVVNBVqMDcCypoDYUQDiT9gqQ/okUQ9A1JP5Nl2f9wrQMrqKArUmHcCyqooII+hFTAMgUVVFBBH0J6LsY9hPAjIYTXQwhvhhA+9zyeUVBBBRVU0OX0vsMyy57gNyT9MUnfkvRFST+eZdlr7+uDCiqooIIKupSeR+T+r2pxHsdXsywbS/p7kn7sOTynoIIKKqigS6jyHO55V9I37f9vSfrX0otCCD8h6SckqVwu/yvtdtvfi7/JLEII8fXkPvLsg2suy0j8Pv7Z+Xx+6bOhdc+57N7rPnPV8T3pGVmWxefw92XEdZfNOcuyx37Wjd/HvI4vV6H03j6eq1wTQlC5XM5dO5/PNZ/Pc/MJIWhjY0Pj8Viz2eyJfHnSOC/jA/N3ubxsHZxfl/HN7+P3W/feOvlL73uZLqR8XscTp/l8rnK5rFKppBCCZrOZ5vN5fN5sNov8KJVK8f8sy+I6uW4xhvl8rlKppEqlEtePZ5fLZc3nc81mM2VZplKppFKppOl0Gsf4JPnj3pfx40mvPel9H2N6XcqP2WymUmkRM/PbCT5Op9Mon/8i6EmWZe9lWXaw7r3nYdyvRFmW/Zykn5Ok7e3t7NOf/nSOWZVKRZVKJSpxtVpVtVp9zNgjXNPpNBoAmA3DnTY2NlStVjWZTFSr1SKDh8NhfH6tVovCN51ONZ1Oc2NhfAjfbDaLilCpVOIiTyYTlUqlnBDXarU4PuaBsWDOTq5c3BsFgi9+Hxdu5tpoNCQthNSVajqdajweazweazgcSlooWLlcjoroCsPcXVFdYfk7HT9zRemZC/xjTIyfNZxMJur3+7pz545KpZI2NzfjPebzuUajkXq9nobDoSaTiabTqQ4PD/WDP/iD+rVf+zU9evQoxysfE7IUQojPDyFoOBzGdarX6xqPx6pWq+r3+6rVaqrVahoOh5rNZtrY2FAIIcoHc+AzLqvIAYYr5RnXLHVDIYQoP6wt68EckCVfj+l0GvmXZZmq1epjBpMx83s+n6ter+fkeTabqdPpqFwuq1KpaDKZ6Pz8PPLu5ORE8/lch4eHGo1GOjk5UaVS0Ww2097eXvx7PB7rvffei7xER9vttubzucbjsWq1miaTidrttobDofr9fuR3u93W8fGxJKlarT4mb/AoyzJNp1PVarU4N0maTCaPOcnRaBRfQ9/gDcYXmk6nGo1Gqlar0Z5wHZ9BDiuVik5PT6P9qNfrUWdZu/Pzc3W7XZ2enkb5TJ35k4LKlMbj8Tcue+95wDJvS3rR/n9h+dqllEY/GJc4yKXCp9EeQuz/s+hch8ByDxyBR3ySorHBwOKBy+WyqtVqVDzu6UY5jYo8C3DvzUL7tc6DVPn9PTewOBr/zHQ6fexzHv1gtPlhDoyJ34yP930OfBZesSbpesEvfrg/PKjVarl7+3WsKc/D6DBmjHvqEFi7fr+v11577bFon/tXq9W4xu5I3AExXnhSKpXUaDTiHNI1SuWhXC5HA1kulzUej3Ny7nNmndJszHnD34zJeeHr6evKXBkP73lkDR+ZFxEyxt6Dio2NDXU6nehMCZDg1cbGhnZ2diJ/a7WaGo2GJpPJYw7fDR4OBkcOfzCq62SV11zWXIc90vagBHIeuf5ybw+YXP/gm68NeodT4PmMjXvAYw9K/0XpaRH/84jcvyjp1RDCR7Uw6p+R9O887UNEVy74MI/FLJVKmkwma9M+Vy43rh41cS0RDYo8m81iFO8L5ePifgiDGyE3gG6EfYxcx2+fgxtP7pGm3R5twY90XC7YHoGnaSpj90iFz7pB8fk4P9OMyOfgc3HB9tSd6zzbQUncUKJoPBdHxmcxdLyWZZna7ba63W7OSPq83Hh7VO+G2iNDN348J023U17xw/+u0BgOlx9fc973zzkvkQHGNh6PH9Ob1IB4poB8VKvVnDzBAyJ9MgP+ducQQtDu7q7K5bJqtZrOzs4ec57omK+fZ4Tu/NDLarWq8Xice7Z/FqgI3k4mk8i78Xiser2e4/06qBVnBSzCGniGDB/9+S4rjjCQLaV8d5nyOfh1Tk8y1KkOP+166H037lmWTUMI/4GkX5dUlvS3syz70tM+BzM8unVGzGaznCckDcNw+Ov8TlNa98gorEcxnpYiYJJi6kd058/0KNPHgAC5MPBshMSjF+7thiN1Yg4l+Ji5xhU3hDy8xHxTw42COL88UknTRuaJQLsypOlvGuG4ccA48zeKnjpNj9iYG2vpvGK8h4eHqlQq6na7uWzLIzIMC47cjWKtVov3whAwttQxeUDCdaPRKM6beXBP5zEy7zLnBpCghudxf3ekyIhHlMgUQRDPQiaQyXVOxOWae04mk5g9MUbghv39/RwMORwO1el0VK1WI8yJDAK/8Ez4jkHf2NjQaDSKY2Pc/X5fs9lM9Xpdk8lEk8kk6gH3AFKEkI11sFS5XM4FEp5JeZDoTgd+cw+3ScjxeDzO6YfrlGdfl0Ev6ww4tO71J10PPRfMPcuyX9W3ce61CxAM9Uh0ee+c50u9pb+HcnLduueFEDQej3OpLekmuHyqTH4v/9uFQ1JMXR1C8UjTP48Cu3AxvrRomH7W/yZywpjBN54LXzy6SaGcNOpzo8P/HqmngptGO6T7jCM1lP4DvusRIPi2QyUorNcRsizT/fv3defOnRgB+ng8m3K5Yp29FgNuDt6+sbEReZgGDPARGGk4HOYyA18fPufQAg4N4+p8dxlw3rlz8WDBs1v0yfntGSyGz9cYQ+vQWq1WU7fb1WAwiPfd3NyMhtV1tV6vS1pBLzzDITQw9o2NjaijZM5kIszn/Pw8Vxh32RgMBjmHzTw2NjaiAXdyW+FZexogpbqVIgKpk2Z9/DVHBrjGI37XkW+X1tk0pxuzQzXFxZwQfOnxFIfPOl6Y4qf8+DXpvTEiRJgecUp53D+N2iEWNF3UdRBJCuk4nurYYRr1+zhSw8trKQwjrSAFUvEsy2J05ff2yN1xaee1/58aaXdKXANfPIphriiG49vwIIU2nCcUNX19mNdsNotFZM8G3AAQVabZRbVaVb1ejxGn9HimVq1W42fBppkj64jBcqPr2Qqf4Tkpdu4/GGIMrsNRqcFnfVxm+Rt5cGPka1gul2NtoVaraWNjQ+12OzqJXq8Xn9XtdiUtspJms5nLNFg7+BlC0ObmpjY2Nh7LoqlHMK5ms6kQgtrtthqNxmOBFfwbj8dqNpvRoc7nc9VqtcgPDwRcngncPNtMMzIPXPxvlyfX6xR68+zTkYjUQcD71Eg/zWhflW6McU9TehjGgvGeL1jq/dKUHkPn90ExUian+J4bqNSoQiy4GypXMu+ISRXWswsfh5OPw5XUIyp/JvdLHQgGEUfmhsWdCQrt6XMaoaCsvJeO0ZXReZRGsB45+2+uwUh5AZbCnUfejguj0ETZrpgeFTcajccgOq7JslW3y8bGRi5d595cjwPx9feaQDovX1fGlSo3ssb6kEWk187n81iAdIPi0Jw7Atbfnb/DE9Kqs8TfR0/ghaTY1YKhbLVacc34DcEP7k2EjsMYDoexoEom5o6HeRCIDAaDXKcbHTgup1zrMsS9ccYpz93BsY7oQAq/udFP199tAvLk65rCquma8Npl5Lp9LbDMt0Mp01xg04jZvS6vITDrvKc/IzUqfh/vRPEo0VvOpLwj4j0EgHQzjbgd9069tgumGxN3XP4s/uda/vf5Oa/8ft4JhMA7ZINhZ5xe9PJI2tPhdcaLcTqskEb0zgPG7kbKnYYrKMbODT+8lxR/eyDAPNyYuREqlUpx3bIs08bGxlo4bJ188rpDHg4V+bMw/GSbrGkaIGDg4cVoNFKr1XoMZqTuxDM8knS55H83ZK4/FBg9a2KNwdzv3buXC0qOj4+1vb0d+TaZTHLj84yESLtUKqler6vVaqnb7eaaF0qlkgaDQa7uMZlMokF3focQYksjMgAc5rrk68ka+euuHy7b3sfv93LeuT6nsgHvU6eaZlIppTZrHbR0VboRxt0jTxiIgJFueeQFI0npEITUESA04H7SKhr2e9E54FGOtHIGjCtN0VIn4795Vor9MT6iIi90OmTBa268Ud40QlgnEJ6So6RuLB1eQRiJgn1dfEysDdCV34N7uvL4s9KMLBV8Hx/Xca0bOSJm38fAWOiWeO2116KRcmiLz3smw9iJBDc2NmLhDmo0GhqPx7n7OYyS8p0olUIb/IC/LqustZTPHsmiJOnw8DBCGRTuvBPFjTkG1uXQ7+264I651WrlcH3gDt4/OztTCEGNRkOnp6caj8eaTqdqtVpqt9uq1+sajUaRb17v8mKpG2lpFUx4sdQd93w+12AwyNU0GGu9Xtf5+XmsE3BPn6O3bDp/kYVSadWRR5DjY6CVE5nzQHJdwOJZD/UD1snh1CcZ7TSITHX7qnRjYBk3Mp72e5FRyhtnNyyScmmdtDIQHtVL+Z5TrgWiQHndqPpuMo8yPH1joYEE2OCSRmRplsCc3NB6+5hHhTginu8OwV/D+DJ+UkQMgjtL7gnOnGZJ7lTgtRskH4OvF3NiHbjWX+c6oio3/K4EboTT7GpdJLS9va1WqxV565vV0iwihJCDWjDgyAHG0gukXrCeTCYajUa5dfJ1A95xeXBZBr5bl8EBy2xvb0d82fvIiVh9nV0WIeep64Wvo7TKAmgN5nMPHjzQ/fv3o150u91YOO33+xGrD2FRAJUUeQpRRMXow0e6a1zecYjIaqPRiHJDsTSEEDMZ5JbrXOcgd/JpNgaPPBvi8+tqGMhqWkT1LI0xcf9UF5zSz/vraaT+HRe5uxKnRk3KV5r9bwydtFJub+1bF81nWZYrKvI573V3xU+vk1YQkpQ39ozLBdi3TuNo/PMeUbhR82zEnY3vhk3TPQw7UYO0ap9Mlbvf70chJEIkSvHuBMeOUXzG7fOCB+7w3Ji5Q+ba1EF733KatnoXBhG6R6mMfXNzU41GQycnJzE7cogDAzmbzdRqtWJEnWVZNFDlclmj0UjNZjNnCDHmrqhs1PE1oVUQOMKzCIzMujnCA4qazlfWSFoZYXc47hRcR7inR5buhLiG4jTrz2tEy17LAQJBlyTpxRdfjPxxOXUZSAvBvM4ceD6y3u/3Y1EV4w4feY3xl0qlWPBlzGQDBGjIAOvv7ZEur67TLvNObqfWNXAwj9SGrDPgni2nOrXuNV/fJ9GNMO5SvgWPCbvBl1aGNMVOuYb/vfVOWhluj/pcKFzIHV7xRfbf3pPui8dYfOs51zmkIq2yB4dOvD/XowwcwrpoXlJu7B4583+KozrkwBhIs70f3cefRpa+mSblBXxgvqTJni1wrY/XlcXXlLF7zzQRN5HeaDSStCj2tVotTadTXVxcxLE5vESR0rNAxusdKPCJlj1p0d3hWYHDTqTgDv047oxBgQaDQeSP89zXAKMHD7yw6c4xlTPGhawhS77+OCXwf+DJ6XQajxO4deuW2u22ms1mPJphY2NDg8FAnU5HpVJJn/rUp9TtdvWbv/mburi40Gg0Uq1Wi/wHkvEMimzx/Pw8dhq1222dn5+rUqnEegTGlkI/RVSvL00mEzWbTUmLIKBarWowGEQ9dmfoGD/PcZmFB2533HmnOh/CAkKjTdTXJc3S1jn1q8IsTzL2l9GNgGXcYHk65MbevZQbr3RzgrfGQesY6vBACufwOkLv43Mn4ouVOqY0svZ5esSRKqB7ec8GeJ5fxzidL3R0eJcRz59MJhqPx7kzS4jK0n781FD78xzC8PmksIvDTfAMJYO/aUcRrzMX7ukG0I2+ZxCVSkWPHj2Kypm2D/oarVN0V0Q+xxkzGGmHQnhuCCGe0RNCiI7Go0SU3Q2yrwPj8fV2eAFe4gxwPr5D1eXC+Zlmn6nc4PAwhKVSSf1+X5LiEQCtVusxuQJOOjk50WAwUK/XU6/Xi44cQ3l0dKRqtarNzU3t7+/HzITov9FoaHNzMzo/d6R01GRZpl6vF/kzGo3i5iaCqWazmcsUPYhJoT+vA3kA4bLq/PLghvWCtw7tuky6Iyf7Tpsq1tHTjPZVHcKNMO5SvnjpmBUK6kbCI2CE2A22G383cJ4RSI/vgOXZKSbnqa5H1esKV6kj8vG68vqYibjcALrg+b18bIzPFTmFOjwa9iI1EZhDP57JMBfHhD1C9et85yzCm0JXKcwEVAK/UifqRpcfeqoxcNIqIvINOfDPjbqPl9eIVnkmY/b+bMaOUd/a2srhvD5+eOjwQ6VSUavVyq0TsuA4vDtHN+rMzY0Kht3TeS9WIh8esXtmDLTVbDajbnk9Bh64Lt6+fTsWm7nvbDbTYDDQr/zKr+jLX/6yDg8PJS2Ksz7/er2u7e1tdTodbW9vazqdajgcxuBpOp1G49tqtdRqtaJjOD09zTl41pz2SxwrmQA6gnNxWUHmXf9Zf+TIZTcN6HifQJB1QK8Hg0H8Gz3zTJDPul6ngcI6G5KSZw1PohsDy8AEDAhwhHvgdYxOOwa8Cg6lBtkXift6pMx7T4N7UGZ/NoYQAXHH4UIlPb5rzRXQF88jaP5HWHzzCPzz7dsIrwsMykAE5LCBZ07SCj7yQqHDV1zjR0EwP2+jdMOarp8bfH67QvjalcvlCBH46X6MrdPp5HbguhykCka/O86ETIa0X1LcIMPnXQYYN8YDA5tlWYyAPT3HUHqGwG+HclweXU5SufTo0bOBVHY8CHBoEkOd1nwk6eLiInapbGxsqF6vRxmZTqdx49JsNtPZ2Zlee+01HR0daTAYaDwea29vL973+PhYd+/ejRkijrzRaERDzNpvbm5qOBzG/vfxeKxOpxMdAFAWWVStVtPp6WmUNyA0HAtG2Du8kDkMPe9RrPZuGE4BhZdeF1iXeZKppcGSG2LP5lP992ugVP/9tSfRjYncYYb0+CmQHm1CaTSNMUCZ3Al45MYiuseG+W78uc6NnT9jHezj0SFjSg03UTTPcyzeu3ykyz25R7aeeWB4KKZ6FpLiukQ0KdyQziV1VIzfDXKpVMqdIujFJC96+n09OoVHzlfPwqBmsxmjXa5hq7xnd0AKHvGm68ac4YXXYlI819tWe72eut1uLgLH6DEPcH+ygVQm+aGLBlnzjiXWxteRNSfCdj3AwXjUCKVQHuPxTWyuG/AoyxY7gS8uLnIFXIcYvIvnnXfeUb/fjwYYmMSP3WVn69nZWU4HiYDPz89j5O4RN5AXsCPrV6lUtLm5+Vgk6zrizt4dKo55XS0o7RDjnsgCG69oTU2PbOC+bpuQR7cdl9FVjPfT6MZE7giyR9Ae8fjrLKT3nbqhkPSY8UkdgZTfxCI9fiYKz6H45lv4WazUA0urCMKNqysXvbQoDwruKZzvHvQ00OfpWQ5KinFFiIhuUOButxuNpI9tMBjkIj9gDo+mHXJhXRgPBtHv6UWsdP1QemkVwXpLIUQGx3OZ1+bmZnx/Pp/HAhpRtxto/vbdo2DVRKTw1DfzEMW7k2m1Wjo5OYlrDITjUaIXPYkM/Xx4h148e/SjLzybcofqspZCCY7puiOhY8QxYObDujqMg0POskzb29vqdrsxS/K+/UajoZdffllvv/22Tk9PNZlM1Gq14jqSHZRKJV1cXMTuo0ajoX6/r4uLi1gkRad4Dj39pVIpnpXO/OAL8/DAwjukXHY9+/GgiTW4zN64PUm7eSDvRnKnwn3SoM3XcN240ije1+yy6H4d3YjIHQV0Zq870wWl9YjcuzDW9YevwzA9EvJ0NIUdUvxSerwnnde8UJKm0T4GaVUoA2Lgbz89ME3VXNh8bh4dugPJsizeD8Fjp587QYzKZWdu8BpRmq+T88CF241hCj04pBNCyEWuqaNCOdw4cy8v7Pkzt7a29PGPfzy20BHhOmZLMc5hHc4T9zWUVtCN/yay9LXH0KfZmqTc8QE+lhQyI3tyJ+/RMp/lesaD8+aeyAN8TesDnpGlveFkQOjIxcWFer1e/KIOrjk6OlKz2VS9Xtfh4WFck3q9roODgyh/ZCcY7vPz86hPw+Ew7nAdjUbqdDqx66VarcYdrb5JiftgUF32PYhIDafPCR10XmCHPJDCMZAteoDj2RFOj99eqPfMLoVjnmYTU0o/87To/kYYd2kVDcIExznXGTaHCyQ9duSmwzjOWMfFfHE5FdAdjSuRK6eUx+s9gpVWi+0Kg6A45sq8PVXzXnNeTwXCo1B/Js/CqK9zhHQqpLxkLMzDo2d/LYVPXGB92zgG2aMgj6TcufKZzc3N6Ox8vdO+eXBVT9t900i3233M0HkkzwYbdy6np6cRSmDjkRcPMZ6VSkWHh4dqNBoajUbxddaFjALKssWeAu/gSTuk/N6socNu3Mf3TKS/3YixRhiZNCt2Z+j88Gvoq8dIw1v0ZGtrKwZh5+fncR6j0Ug7OzvRsaCbZGoUUNvttiqVik5OTjQcDrW5uRl7/Jlrq9XS9vZ2HFez2YxOG14iG+n+EYdQPUhBn1MnyWc8CIO3HhhQL/AjSXgOukv2B5/drqXBW0pPM/rPQjfCuIcQYioHg9Oeb49GUqH24kkaNTmE4QYmvZd77NTruzPh/zRShVg4jBIFqRQrRrEZI0UglJxU0591WdawLpJNMxiHNtLzxblP2t/Lvb2rxg0/gg9feI3n+3gdauDevi7uYNb1+xN5NRqNXG+y87LT6WhnZyd3H2TDDejGxkZ0JMACwDDwkhMRfRMM8A/zIZvJsixuw+cURbpRWA+KkrRVYhxSZae1DyOH3LE+GCNfi9SIOc/TTGvdOqJf7iyr1ao6nY4kxc1dzIWuFDdy9LbzVYg4yq2tLQ0GA7XbbW1sbGh7ezvuO6nX6xoMBjEr4hhhd7pkt+xwrVarsZ0yhBANagpXepupy4o7S//t13rw4zLrjgSCn2QF1E6QV3QPOfWgJbWBKT0tMn8a3QjjnkYlKc6YRn2pMXW8EMONQXIGwXCPil0hEBY3iuuKgYzTo3qPdvx+KWQk5XeNutNKFZV5Eo17RkEK7xV+N+A8F0MCNuzbynFoftCWz9WF0fF2L16uS/e9o4M5+Dqkjjf9m0iJeaQng5J9pNHtfD7X17/+dZ2cnMTr2XjjDhv4BGc2HA7VbrcVQohYvBtFrhmNRjEKlxYbplxx2UXLa+DrwAvwDmfvWRbwkWPyzmfkErnxTT7ISBqZY6xTQ4IuITe+HgRZBCVAJeglvAOHd3gryzJtbW3FzilaM4FqWINWq6VOpxNlZV0E3W63Y4R/9+5dlUqluAaeiQ8Gg9wXpMB75uCBoeuN84G/3WE5fAhsibx4vYhxONrAunqQwhowpjRCT3H3dfSsxv7GFFTXedQ0/ZRWRYpKpZI7QwVhRKmzLMtV6T2yZxFZEFLwdOcpY/DiIv23/jrOiIVPswAEzqvl/nx3TI7VQSg2Sp0aTI/G3CA6hksLGLxDsdMjU5lXemwCbWEOl2VZFlN6j5RwJu4cXOHSTATD5nsN0mI1n5UWkeTFxUWcrxe0eC5QQLPZzLWnsib+5dabm5txh+5gMIg7UDE+GLU0wACGIGLD8GMMhsNhrsPH5wMfuA/wB18X54GEtDoAy9fWI1TnE2N1nXJD7wGTj8f7z5k/64ujodWRDGY8HsfNRePxWI8ePYrfp4pst1otvfHGG+p0OqrX67G2QUsuuD/3AV93ePHi4iJnEBkj9Q8PLJBv2nx5nTkh484fD/LQVwx7r9eLus5nvN3R96/AR9en1CH4M92oPw2SeVbI5kZE7tLju7tSgyPlv1LMI0OuQamAalJ4wiNkzwYcr0z7nt24sXiuOKTbpJGuyH5fFNLhHISQa9P2N/jAfF0IGT+RgEe7bjDn87mGw2F8DYGldzfdeMHzfX68lvb08nzntWdVaRTp/3tWwxzSDMkjV3fSzWYzfvmD46yl0qKLxo8HaDQajxXEWAcMA7tLHUJjvafTaTwcC2PE+DFu4PuMCeeGvPJ1dDyfAivBgh8Rzbp5iyfG1YvzpP+dTifqjRsZz/DgT4o1u/OVFHcvExh5H7sbsxCCDg4OtLGxofF4rKOjo/gsggXkbTqdqtPpqNvt6uTkRPV6PUJO3M/XzrNQd8SuF+xYHY1GkUdpRxlyyjxTQ07wkmLyqR76OfMe8HEveOT66RkOhHw5ubFOYaL073X0tPdvhHFPDTgM8vfSDhcpv6EDgXJMOi0cejQo5bfCp7h1ei93On6tG/vU2bhBcuXj2+M9egLLZK48H2PNWFBQx01diDlFz1NxFIS/SV+96JuOG5jBozted9zQBZj5w1f44wVp7+5AmTD+ZCZ83ue0ubkZYRDGCFzgEBUp8d7eXg4S82yP7AolJL0vlUpxwwz92BzP0Gg0YvYCH7xPnHP8JUXIoVKpaGdnJ/5Ka8wAACAASURBVOLsHGZF1Eg07VFkyhM3yGDTISzwbLIS5y/z9EwRR+CHcHnE6PUoxgPGjXGrVqvx3BicHZvJOGMG/DvLMnW7XQ2Hw/jdqWxq6vf78YiISmVxdo3rE4Yf2cHYe7uwtAhKms1mzAY8U2ZtPcDyOo1DmB7EuJHHiXhHDLAbMu1ZMXz2Yw78FEzWZ11b5NNoHYxzFboRxl16/NuC0rQ0jbbTLgyUzL2rY10ezbgHdZzPjWUaAWEAPdJwxZZWUYFvCMKwck/ftZlmKtzDIRbniTsqd0KumN71g/BR4MF5EJ04DJT+5n7+7TwOTaVfwJwqBvBE2ludFm19bowtnZNHQ2Rns9lM/X4/GhSXGd/E5IVCz9QwgqxPo9GIEApG3GWC+3t0OhqN4jcScU2WLXDnTqej27dvq1KpRAME/Edh1U/u5LOpbLjTcmdP6k93D+8zV+7hRtNbJzHGaabF7lPvlmE9gKBwvltbW5IUHamkeKCY7yolm2FOjOnw8DBmHhRNkU0PFDD+OCcMfgghth6mRwynWaJnb95t5c/iGci9w3mpk6AgjHxyDbJBVA/5+UQueym5XqSvpdev+7zTjTHuHrlLimmXY4Ru+Pjfo3g+70z09N5hEqfUsHl6TLRACp4uNp7ZP1OprHbRYQi9PcozCASLTgyHg7y3m+s9c+EajwZ4lkdvHsUBD9ER4tFxalQxRLzuPeOePcAHz6w8I0jX2Au3GCo+461uacqMg+K+QBcY13K5rO/+7u/WJz/5yXgfFA3logOE+3KWCZ0ubHnHEYP9umMl+j49PY28B7eHx0BDGFA/m4ZAZmtrS6XSolDo0B5yx72AkOjEIWpm/Dh0Isc00HEoJ83gcCz8j/P1DMd/AyE9ePBAs9nibJl79+5Jkg4ODiLf79+/r4cPH0Y+c0bM/fv31Ww2tbe399h31WKEgcuQsVKppO/5nu/JnR3jkbHXvNJgweWOOfJc5zV8ZX3TDWXSyrDzHLdFnuEShEyn03giJvPgXryWktsFN95PM+Tr6MYYdya7Do6R8jtOSYU88uTaNN3ksxgdNxgewXtxlkUjCnKj7b89ovVFRtFc2fDiYMCM0SEDjySkldAhrIyXaNqxRYTbscpOpxOf5z3o3uftrYC8750qvpkMJQdPZO58xvngEIkXeX1u0uOn9UGso+OfGHa+Q7NcLsfTBDEG5XI5dsik0Zu0OrLAoS5wcApwlUpFu7u7MfKEh1w7Go1iIZFiYAghFt6ocSArRIDlcjk6E2SA56Xtc54t4JiIvlmvRqMR36Nw6e2W7gwJSAgGkAHPxubzufb29iL/PPPyWsVstviy7K2tLc1mM73zzjtxTEBStHR+4xvf0NnZmS4uLjSZTLS7uxt56rLG9UTwzWYzHuNQrVZ179692BWzvb0dO6Y8Y2VdvabBEQEenPiRGMzNM8fUaKc2CD6m2f26bNPtkuPqLu+pI/p2DPk6uhHdMilDPZWXVhE1UTKfQThTuMONju/IXOdlScGJ1jy9lvJ9rBh798L0P3Otd41I+cg/VRCHKSTlTiNMi0JuRDHs8MznjUFmxyT9wJ4JkIXgJDGuXmTi2czBow93wusOLvMsw6MvngWPGA/ryFwwNh6R4iQ2NjbU6/Ui/2ezxZduwM+vfe1rcSORZ1nu9MGB+Z91kxZG4tatWzE6d/zZ5Yn+7Ha7rWq1Gvu8cRI7Ozva2dmJxUiUPc2M5vN5bMP0+SPn6w61IqoGZms2m/E7VpEnbyNmbPC62WzGAh8OAMMJX46Pj+PaegbscBDr3ev1JCnCNsfHxzEIkBQP9tra2opr1e/3Ix82NzejI0lllaIsh4nt7OwoyzK9+uqrOj09zckWephCVA7lIL8EPB408WzvmmPufB5ZdXvhtoTnu7yxdsiOEw7F6TJ8/Vlx9xsRuSPUzpA0ivfuBSIRFsqvgckpnp1GjDzDU0L/DXlk4+8jGERIKB33JVLwlBYhdCiEOTJ/rzl4VOJRncMtfu06OIM5ozyM0/um0yjXoxrnMeTRcKpAFGkdY/VreBbzcgjAMVPvEw5htVmF80vIdDBK8BRj7krj8+I17u/RMc5uNlucIY6hYy6dTifXj16tVnOwyAsvvKBOp6O9vb2YIezt7UVlJ1pnToyZ+fEcZN/lCyfDNeC93jMPVs558nye15E3juSVVm2SdL7gOHkupz+yGxcZoqVxOp3GYi2YNR1YvEckv729HU999Gzv0aNH8fqTk5MYADAGeML6Urx2yJLXmIsbUyAsN9JexMVpedEWx4QDlBT3FrCRLoVNkW3PCt1hOER5FUPt17iduyrdiMgdYpFgintz+nr9PU+jMCDeK41wOFwAeXTN5z26dK/s+Jtje9zTI0TuiVDUaosjQ1EYh3ZwakQdwAseJXjW4W12Hk37gnOfdake/ED5q9VqPMGPVN0PXvIt+GmG4o7JHRHRozsWH6PznczHo2jnMf/jBObzeTzHW1LsqaazgfE0Go3Y6YAhTKEsMiScLxuNMEwXFxe6detWXDvw3lqtptFopMPDQz148ED1el3D4TAW4jqdTozsGAcbfra3t3OGjTm7zCEfBA/wAX5jsLyu1Gw2YwbqBWs3Nu5kMfiuQ6yBY/kXFxfxfjgNjO1gMNBbb72l8Xisg4OD2LGC7GHMS6XFoWHUlMhsmBM7WtvtdsTucTSVSiXy6+zsLKcX2ArW26E25s6aeOYGcfyBQzhc40V6dJMvIUlx83Xde17DI3ig9dgDDod5HLK5jNZF+U+iGxG5OwTiqZDj7G4UHULgWjc03NONuhsOzwT8/RQWkJRTEDf6Dv2k2K4bJCI4Frxer8eNEVmWxSjAF9d7a32uKLdHAI7p4hg8Akw/DxxDlOZdLT4GT//5gWcOyXBfCpGXbYxibdKo2jdascYOWUmKMAqK6MXU/f39XK8z/MFR+jEOpOUOGXj3DVvkOSANHmKMuef+/n7EhNNuK2So1WpFY0fki4LjPDxjxZmCradKDE8cS8YoEzmDUXMfDyZwHCEsunwoKgJTUdR1efedtfDUN/zRLspn5vO5Tk9Pc1mGJL3wwgu6deuW6vW6Li4uon6wcQxDvrW1pUajEeEYCvrdblfdbjfHH97jiIfBYKBut5uDpFJIkOgZWYavzIHxeCYnrYw2uuaBo++Ed/gstWPpa/5zGfl7lyELT6IbYdyllReEAUQVKL7DHuDfKWzj0XXKXDfALLCnThCv+f282OgFRb+3R1kom0fc3g6JkeQgJMaTdqP4697eyHMQGB8HhSnSR8f5iZYwevDODbHzAWH2MTvfpFXU7hFQyjOyMebA32n04umy9wyTUZC9kUIzD9/JWC6XY8GNDWalUikH31D0wyE2Go3Yxw68gOHDmPEsHDoyB39YX89qiFjJivr9fhwr3SPS6kvd33vvvRw05A6GrfeM14MPh22yLIsHd7Gmvl7ewQGffQzck/nfvXs3GlPWwmUf/J5Or2q1GmsenO9eq9XUbre1t7cXnRH8ms/n8QvP4S31gMFgELNLTob0bJx7IWO8juyiHzglDzLSWgy/h8NhLqtmnp5tIetes2IunvFj1/jfodf0uZcZ7/T/78jI3TcduLJ7VAtDpfwhQLznEURazXaPmUbfRHS++L4Qnoa5xyWa8VSMRUahuRe4K9GbQwWpJ0c4GQv3dT4hWI7FQ44Z8j+GOX0GysJraSGKCDo16Fzv8BNz73Q6OYjBU1mu4XWUDIMKn1gzDFmtVtPu7m40FKxPu92OBr5er2tnZyd3kiBG3bM4Ii/vD+/1ehF2qFarsciJQ8GAY+hxaA8fPsxh3PCIIqMfgeH92ESopVJJ5+fnkhRhDOAqnI/Ln6TcmS9+736/r263GztHPAhwvqNz3rkFvENmeX5+Hg3tcDjU2dlZ5J2kWG/wIERawB3wAwNMSzPnwsxmi/NoyJDAvMkUwO2RK5x0v9+Pu1Mx6vBud3c3ntRJEIGzSXvMHZr0Jgb2HZDtevboOusoAXLuNQxv8vBAx4MZp6tG4ykG/zRDfyOMO0KcRiMwCuFMDZ2kHAPdy6Y/rtjO6BS64TWP0MDkvXedcbmn5l581r8ekCjHjYqPV8rvovMo1+fJj7TqaiHi9/TRnZ/fH4X31NLTVueLz8fH4tBKGsF7pMT/GCjWFWOM4mHMSIu9oIsz4zrmieFwmIixucICO3E0AEYvXX9SbPjna+XP7vf70ViUSiXt7OzEzTzM4ezsTFm2OOr34cOH0Qky742NDe3v70cYBWgJSOX8/DyeTePrjpPD6HG/i4uL6Ozn83k8h4X34bU7DP9iFGQVgxFCiLLb7XY1GAy0ubkZjweAl9vb2zGKnU6nsUOoVqtpZ2dHd+/e1dHRkd59992YEVWr1bi5i2zp5ORE29vb0VmXy2XdunUrGtyHDx/qxRdfjGuOjLH+QCk7Ozu5TihkBd32PRJpMAUkRWaETHN/nCSUQqn8pAGgrx865MGgfz6lJxnwqziEG2Pc3WhBCDSpdfq+OwHH572TJGVQarwwrpctljMRAwj27UYPI7muyOgZg0fRRICksTgP7uMZC/NkfMyD9B3DheFh/DgiIA8+RwTihVIv7LoD9WfBJy8keWrpUaIbfB+zOzDG7nUExpJmHxQ7iajZKCStoCz6pNkSjxFEJqbTacS74THFzkqlotPTU927dy/yEAOZZVmsUSBbRJpE98BgkuIz6RgplUra39/PHVGLQa7X6zo7O4u8BtunI4ROlZRP8M8DEc+mPNXHyHjXFidQ0lfucspX14H/A/NxFLLznIPcTk9P9fbbb2t3d1fNZlO3b9+O92XcdB1x38FgoJOTk9gpw5ocHR3FDV+9Xk+np6dR987Pz+Ma7u7uxrPge71ebt9DWndwWfTInvVy+NZhPoe2vB3aa2K+J8AzbH/PG0T4bEqXGfP09avAMzfCuMMMyNM8L/Z5lOvpoBshZ5wzEMPlRdD0fo6BptgYxUKHQ1Ksms97xwOpLQ4BRWFukmIU4tARBox5IlgoJ46B+3gm447E4SIXNhfAdbie9zR7YYr7XAbzoBSOmXsGgTFyBeM9L1Yxd+CQLMt/AYo7GzBcDA089wyD9XEZoO6QZl/wB+Ptc8WAsxEHR9JqtTQej3VxcRHHMBqNdHBwoEplsSmKeRIZ379/Pz671+tFHbhz507umWR6qXFinRxec8IpIY+ORzuc51v3vc7DbtN6vR7PWq9UFt8jy7kzL730kubzxZeUcO6My129Xtft27fjWtE9xjc7HR4eKoSg+/fvxxoPxdZOp6PxeBw7pKrVagxK0Iler6dmsxlhHDIuovpSqZRzqMAxXLO5uRl5vS57Z85eX3Fdc1vj8oOtQG5TW5HCyu6I15HbBf//SXQjjLu0ik4ROgQYA4DRkpTzgo5powwIPcab+7ih8u4MVxpvxcQY+qJ6msVY0hPf1i0UhoddhKTG3AMBdlgJA+dRBIqOcHrx0YuInsK70fMoW1IUPOczEa4X9DCEzj94lwq5n7UzGAxitMt1rAXPRvEYF4Yah+d9+UTg8MS/07NSWX0ZM4U8lJ3ot9FoRIMLbyi2zWaL/m0Kf0Rp3u1EJMouVaLrbrer+/fvq9frxV2i3AcZpD3T+7BxJN5zzf/IKNlBCjfAR59L2tKb1kowOkAxXlyFBx4gYZyJjkMI2tnZicHK4eFh3BNwcnISW0ErldXuZuaLTJXLi6K36+2DBw8kKXbMXFxcqFqt6uzsTNvb22q329rZ2Ykbl4BYPHtCNvwoB19DnAt8cEjPazOuzymEm55l4/ALOuVBJc7bW4yvAqlcRleJ2KEb1ee+jmCIR+AYfi88YZjxwhhhN8YsGIsi5XeWOcbuWL+0irI9RXNYx9NhDJwrHvdAMWazWWzjwvD581wgIObl84aYP4aIzzNmL/I4ZOPpqGdHzM9hlhR7h1ecQc64fR4YqZSn3pFAZATWWa1WY3GLazj7xZ0AMtHtduO9MSLuvOjogE9+fk29Xo+dK5ubm7GAe35+HsfFhqXDw8M4vyzL4imVbEAis+h0Our3+3HtcTjNZjNGvKVSKe4JkBbRusuAwy0YfmkRCBBF+/p6tuadSL429Iwjb2QQvmMYJ8k5Nq1WK0bIkuKxAxjbi4sLnZ+f56AQxra9va3d3d3ceT4PHz6Mf08mk9hSOhwO49EM8/lcx8fHun37to6Pj9VqtbS3t6fpdKrT09P4LVonJyex/5y6BwVh35/hwYukGOnDLw5x86zIZdgdJfJPL74Haegyso4zgzes0TrY1422Z8f++rMYdukKkXsI4cUQwm+EEF4LIXwphPBXlq/vhhD+SQjhK8vfO8vXQwjhZ0IIb4YQfjeE8IkrDcQ8WroYjlX73x49u6fECMHAtKDh2L3DFP5lCAgIi4Wi0g7HvVF0olXfuebtWPSee/rH/xgIN3bp54joeJ5nL2CHtAemUJFnLt4eBt4Lr12omR/R5brCNte4c3TYxKGcNKXk3p7ucl+i4kqlkiuCMVbu4REYCkgGBHTCZzY3N3PH3hJFo3g4Zd+Vym93YshHp9PR1tZWjDDd+CJHOAva6xwKwZmwZmSAGFGMOn338ItsxKNy7lEul+MuWs/ycGw+l9RouPzBR4cy2Y1arVZ1cHAgSdEIk8lwDPFgMFC9Xo9frO2Gv9frRcdHLYJIeH9/P9YYPvKRj6hUKml3dzdCc5ubm7p7927u3Pf5fK5XXnklOvDJZKJutxvhsTTg8oO80CvfrQwP4Asy6XLoMJfDvB6EOm9xnOiRG/Z0LfhMSs9q2KWrwTJTSX8ty7KPS/qkpL8cQvi4pM9J+nyWZa9K+vzyf0n645JeXf78hKS/8bQHwFCftKfgLI4bNBjl0QsMwJM6NubFDqKVlLEevThOjhGSFIWECNpb91hobzFDsTDYk8lEFxcX8TWEFiOeFh09k/A58RpjwGikODqfo7/ZC2vuFNP6hDtSrxGAw7uzcB5QXIQfni47LIbQe8+zQ2Oshz/L70vmwRxarZY2Nzd1fn6us7OzaAR5HtGqyxvFV44QwGhubGxoa2srjtudj2c3RIZg2xS1Ke4BSRG1Sco9A2PoBXG6WDDY67pjaFnkM9yHWoO3bEqrHZesl+8XSetUk8kkfnUddQt26WIAqTGcnJzo0aNHsT5zeHgY9aZcLkcjXq1WY8//1taWXn311bgBCrn5ru/6LtXrdR0fH0eIhE4ijDa1C+Y1ny964U9PT9Xr9WLLJhmVt/DigIDskAOOembMrj/IG8GetDq3hqBnnb7gONKAyiHndZF6St+OQXd6KiyTZdk9SfeWf5+HEP5A0l1JPybpU8vL/o6kfyrpP1m+/j9ni5H9PyGE7RDC7eV9nkie0qeppkeAXIuS4UFZlHSDjHtWm1cuOp3P53GRpdUiMh5fNDcyUv7Lnn1bPsTuxEajoV6vF4UVo0ILIAbeBQuYhQgVnjjM5EaRyN4Fw4vA4/FYx8fHsc8anjlO6xE39/Uoxp3KeDzW1tZWLqvAePI+jojPcx3zcn7hqDhqAeeeRlQY7dFoFDeMgeM79MTrpOu+OSnFl3kuYx2PxxEC6Pf7uY1k3OPVV1+NXTy3b9+OmC6OBwPt60eW5UcVEy0zt3K5HA/U8lMM2RWL4QLOwGAROVMQxWi5UUE/uHca3ZbLZT169CgaNoISIJC333478smPMajVatEpPnjwQIeHh/FohBR+Gg6H8Zhk35jlTotsgqOY2bVKwMQeg0ePHkXe8B7HGoC1oye+rlL+a/pcLlM9xJhz3IQ7T7c1rnMQ9gMdc6gVeU6j+NSwX+YAnkTPVFANIbws6fskfUHSkRns+5KOln/flfRN+9i3lq+l9/qJEMJvhRB+i40JrjiOVTtD3KumkUcIq00HQC2pYXAnsBxH7rhVFsF3AXpBy2EJhzt4NoqN4hL5cvIemzU8qvID/lMoCuXzRedZzMudjP9OBSWN4ulCYS7uLMhOfFxeuPPirDs1oi5e5zpfI3eorD3GS1LcVs7zNzc3c9dyT74ej8hwNBrp4cOHsTiHLHgB1tfO6zJAGDgK8GDgEgyNF9X4DMb37t27caMLc8DI0yXidQGcAsYTqGZnZycWgFNYqtFo6OzsTKenp6pUKnr48GEOCnRDgePAMPqGLIdhWD8K8RzD4JkyRhxjPRwOtb29rYODgxyMQz2A1kWKomRX1FXYRdzv9xVCiMcMA/Xg5Hu9no6OjqIxRedp+f3IRz4SZZa1dfgUHkiKkJXvYfCaCXLLPZAfD1j8+GwcB5km13gA6rq4LiOH1hnzlNa99iS6snEPIWxK+oeS/mqWZV1/bxmlP5NbybLs57Is+/4sy77fcWxp1U6URqruXVNvx+JSJU+eFe8Pw1FYFoN7+ZkSknIRl0e1KW7uka6UNyxEUefn51FAUVjuRYTHON14OpTh3j1N/RA8j07TLqK0LY8ojjk7XIVyOFbrBcX0CNMsy39TvEflDumQEWG4vBWP5wL1oMj0rXuGUSqtdj1eXFzorbfeUqlU0iuvvBLnzDNYR29LY61ns1nc2Upkx3Z6omN4S2SJIZRWZ9LXarUIx5CRIZOVSiXu6HRDAT/8G5KyLIu7bOGBQ32VSkVf//rXVS6X9fLLL6OfOj4+zkX+7IZ2vJc1p6WTGgLBktcscGQYeuAZzq0na5IWXS4nJycR0/7e7/1eVaurkyrhG/ch8t/Z2dH+/n6sjbTb7dzObZzicDjUW2+9FYvn8OXWrVtRnrmewATZIFthHdBBjDt1DpdPAhygHOScXczoRNqc4bg6r7MO3jnFeFPy19L3n0vkHkKoamHYfz7Lsl9cvvxuCOH28v3bkh4sX39b0ov28ReWr11KKLSnxzAuNbQOs/BZfmOEPYrhvrznBUgMn++KdLzeF8mNurebeYuT4/JeYEPZdnd3cwVUFCTtknFn5viyGzwvrPJdlR6Zp5mNGwi+5AFhXJfhSPmdrcw9xRmZt28Xl1YFSe6NIeE5KBhrgRH0Q69QeDdSvv0bnjWbzXgwFcZBWn2FYqlUykVrXuzsdDo6PDyMkSep92w206NHjyIsQ+0ghNX3wMKvTqej/f396FAxiODR7XY7Ong3pESMjqmzMctbIjHEzKdcLuvhw4cxWMB5uVzN54ujdFmLSqUS22G9QO7ZDLrEWgNzeL2KsSFrg8FAH/3oRyPU0Wq1NJlMYrGYIOri4iJCPegmnSrlcll37tyJsNjt27dVKi1Ok6RofXx8rPv37+vdd9/VZDLRvXv3dH5+rq985Ss5eNa/Z7ZcLscahmfB6wrL6CXRuHfheXbruuCBCtd6pgoUyPp462sK46wz9M9qzFO6SrdMkPS3JP1BlmV/3d76JUmfXf79WUn/yF7/s2FBn5R0lj0Fb0fZPbXiNwvlKSLEgnrq44bc06R0d5gXlYBLHBdLMbPUWCMwpPAuCPxNJIYhRnAo5pVKJW1tbcX3UTaU2I2hOz8v8BKteNSAoYRHKeSDA8MppN0qaZaEw/GMwyGbdC0ZD+sqrdrJUAiHs+CvZ1Fp5ubQCgVYlIouFs4W2dzcjIU3n4t3OjB+4JNer6fXXntNJycnuah6Ol194Qkywo5KPxIXWdzd3Y2th+fn5/rGN74R5YLNP0T5OBq+uAJ4A+OTyjFzISPwPQqsA7ygfdCdu/eBI+OenQEPUQNqt9tqNBo6OjqK8gsEhT5Wq4uDwth1+s4776jX68Vz2QeDgYbDYeST10U4yZKxkDHRnjqfz3Xr1q3oCDlawGGs/f39nL148OBB7NCBXN8lRTjSocbpdJo7/M2dn9eSPIDiWmQKG+R6hL56ZppCNE7rDHrqCK5KV+lz/6OS/l1JvxdC+OfL1/5TSf+FpH8QQvgLkr4h6U8v3/tVST8q6U1JfUl//ioD8cgVYZTyOy7dUFPBBntlAdwwerrkqaZjwRgMXifiZGccKSIRphcKpVVU69V2iM+xGw/j4xEqQu0Zh5Q/E8OxapxPKhiu+KT5CG0KZyFwvO8CiZPxTUMe+QAfgZF65sNYWSfvxoG4nnUlPSeS4xrGybXeOYJczOfzmO6PRiN1Op1cEc2zHncQrAtRNOv40ksvxfHyhRvwpN/vx64UZKVSqcSOGAzbrVu3YuS5tbUV32d+GJpKpaKdnZ1chkVR1DM3l1s+j8EFQnEjz+dwHOfn5zGC9MyCuSFXyLlHrpVKJUbe7BhFl6gVfOxjH9Prr7+ug4MDffWrX83tGKXVcXd3N0bzyJ0HHZA7cD83ZnNzU/1+P8799PRUd+7c0d7eXq5O5LKLXsI35ru9vR0zBAyzB0rIi29m5JusyLSYOzUg1gzifvCCDNQzgKdRGtU/F+OeZdk/k3TZaD695vpM0l9+lkE4/JDCKPy/vPdjxs8jZRTHFcOv8y4KzxI8fefepJ4spN/Lo28XAvpvvdLPGdacl+0teigj6anP26Nmj1Q8GvZImAIT2CIGEaPGPOl4GAwG8SvPcFoeIVAUhB8OhTmPMTgQztFhIxSDKAg+eosY60hPN/fnWORarZZrz+Ma/85PSfHgKuAj4CfWhZoA54Tv7OxEZ8jX4j148EDNZlOdTidiri4vFEp7vZ7eeOONeEIjB3ghq7dv39Yf/uEfxvXNsiwH+52dncXDwohK4S1RMU6IZ5MVfeITn8hBEY4he8Sfdk9huFhX5CjLFi2PHIdcLi9aGTudjgaDQbwGOQMq6vV6Oj4+1gsvvBD5/dJLL0VHhWE9Pz9Xs9nU7u6uer1edIwUorvdboRF3YB2u92YFTx69Ch2q1xcXMSjA/h+Voef4JdH5jgL6iVE8BzWRgADD51Xrsvw2QMDD0rdAcJneHpZFL4uA14H2/hrTzP4N+b4AY/cpdURA7zu7VwpeUQKzEJk7bgY5Eacogv38fc8PXc8DWFxJ4Kx4qQ/vDWG19vcvMgDjIFAYUTB2HMVEAAADcBJREFUYcHaHZ5K4Qqia64BlkF54As9+zyzWq3GHYjMH14xH9+gg7NzPNJrGygP42HdGL+3pOEwUUh6yzGivMYmIJyEGyzkhGf42d5uoBzOY40Gg0EsklL4I2Lc3d3VaDTKnQWEASUSB9Om1jAej7W3t6dvfvOb0RFTqHvvvffiNzM5LAWf6dHf2NiIrXnM348+8MjPi+lehGedPdOsVFaHnOHIfF1LpdU3TUmrL5Bmvu12O24eIgOt1WrxeN+9vT2dnp6qVqtpf39f29vb2tvb09HRkQ4ODqJzoP3QdZBI3s+TKZfLGgwGsaYwHo/V7/e1ubkZ/6cdEXnE6ZPNkG3gKMme5vNFB47DqNVqNZepEYTwt8OG644a8awauuz7ErjmKtF7Ss9i2KUbZNxZJE9hMRwIK4yC3AvCPEkxLfOqO9enipIWlLxYR7GSsbCwGDewRB/b+fl5jCQZuxvC0WgUe93dszuODh+m02nExRFgxpY6QyKUUmn1je/u/VMcsFRa7dJcZ6Q98iaKcijMswk31IzP+YLh8QKXrwcRum+yklbHMmDw4BNzp6UT+SmVSjo5OdE777zzWFcI48SI0x1DJwfZBZHd/v5+zHLY6NRut7W1tRVhNs4O5zwVvtSCdaWzwh0HcJm/zrXMxR22k0eDyA2/cULwmjPR/T4u20A0Hhh5kT2ExZeadDodXVxcxEDBHQkOmywE2K7b7Wo8HscjDIiIORytXq9ra2sr7vU4PT1Vs9nU9vZ2DHTIgjkagK9UDCHopZdeitednJzE837IWNMuLSAdgjXWNMsWXUl+bWongKfQey9c+x4OPoez9WsJQDyyv8w4u31LbR2/rwrR3BjjzkSAYdLCqJRnenr8ql9LKgW5QfY2Phf2FN5xRXJDnQq4P4Oo0Iu8pHe0jPkzwIu3trZi5OQpr5Q/ohij4AabyMMdo98fo+fZBWMi0veozw09ionAorxkLg4dsFGLSIl14nmuQChX2rFEYY85sF5+bCz8964crq9UFrsnOZLWsyywaXrLMbJkCswP+AxjznOr1WpcJ9ok79y5o93d3Xi0baPR0MsvvxyzvV6vF3eQEo0CCWEgWBuyDmoL8MWxaZxmGrB4+yoBAYaKDMTrNxRwWUvHurkvTQYOaYaw6BTa3t6On/WvBOTrBweDQfxNPWIwGOhLX/pSzCp9zZA9hz687XQ+n8cv1ma99vb2YgTPOLxA6lko+DrPkVanvBJUeNGTg9A8y3Z42GU8DWKQc8fjXRbdOHsmdplRX2cn0+svoxth3NMJI7w+Cfd+Un7LL5/DgCGsKfbmkZAXEx3jRZld2NICEIYWr4wRcsMPLoeiUYCZzWYxygE+8kX39JHnSqvMxI8p4Hfa2sX9iByZP8KPcvnOUy9+ulJ7kRcMn89QWF3XrooSeuukRz3eoTSfz2N2IuW/zco3ceG4metwOIxb3EnrmSu/Xbk9qkWZyfDgbafTiQ7N5YWsC2wWjJyNMUAUrVYrOqNKZfG9oNQRHjx4kNv0Qg+4HzWA0WVNpPy3OkHIKvUL3wSEkXEnivz6IVeeWSAn1B4YJx0tu7u7uZoW9Yj79+/HFkK+oAOM3Hd+HxwcxHG12+149gvZD5E4p3mSzQGRsfaPHj3S6empyuXV1ynShYTxZ+2RL9cpx8pDCHr33XcjZIr8o5PebMBnGJvfC5l3e+R6wLp6RuEO+km28En0NAN/I4y7tEpLnfFOjrNKynlVN2ie1juM4ZVzriOKY+E94nSDj6BzD++iSFNl8EOiYxSXMyyyLItfF+f98kQc3qZFZIMQMRaEi7/7/X7kjWc+zkfPcKTVrts068HQwh8MAoLrEb/XB1DUtOXLee+ZFp9NN0h50ZhuFvrc/Xp6p/f29nIbco6PjzUYDOKWdJcN4B3IMVB4x3OoN/hhY8gh37nK53FMfAMS9OKLLyqEEGGNg4MDtVqtGJlCjNP55pH2vXv34vp7lC6tvu/AYUz/PPenZuEyxWYr9MLlfT6fx6ORgdKm08UO1eFwmGtxZHcwuuHfakTAAiT3+uuvazZbbBrb29uLMoXM7+zsRGc8Go3U7XbVarWi42S3MkHV0dFRXHsgSc9sXU8JcMg2yND4PAEZc3U41mtR/vw0u/dGCKAiz/zduEPPArU8C90Y445xYaKeCrmX9MVzhniPdwq/8D6EEAIneOuUQz8eqQB9YOQxEh7ps70cj89ncEQY/kqlEotDZAZpIY2NRrVaLf6PwSNSpXOBDTkeqbBhyTF2ojeiNTdsznPO5fADlrgG3nknDuvjcA3ZijsVjKy3KbKuODgMEQ6EtSPidKXpdDrxGa1WKxZkwc+ZL+1oKDD8Q6mJrDx7I2vz7fbS6jx377Y4OTmJ9Rhvu6vVanr06JGOj49jSx9f6AHEAaxFJuDZKeN49dVXH+Olr0kIIe5+ZV14PjxHlr2Qygat+Xyu09NTdTqdxzBq7ukFfv/iGozce++9F4u4OEcCti9+8YuaTqc6ODjQ2dlZlC0cAR0wZIF8C1SptGijpEhaq9V0dHQU9fb+/fsRy2dcrCnrSr0I2UkjdyjdXOaNFjhGXxvPOh1Gdn3iGdPpNNZpkPu0npLasyeR28gnUXgeHuNZKYRwLun16x7HFWhf0nvXPYgrUDHO95e+E8b5nTBGqRjn+00fybLsYN0bN+XLOl7Psuz7r3sQT6MQwm8V43z/qBjn+0ffCWOUinF+kHRjYJmCCiqooILePyqMe0EFFVTQh5BuinH/uesewBWpGOf7S8U43z/6ThijVIzzA6MbUVAtqKCCCiro/aWbErkXVFBBBRX0PlJh3AsqqKCCPoR07cY9hPAjIYTXQwhvhhA+d43jeDGE8BshhNdCCF8KIfyV5eu7IYR/EkL4yvL3zvL1EEL4meW4fzeE8IkPeLzlEML/F0L45eX/Hw0hfGE5nr8fQqgtX99Y/v/m8v2XP8AxbocQfiGE8OUQwh+EEH7gJvIzhPAfLtf890MIfzeEUL8J/Awh/O0QwoMQwu/ba8/MvxDCZ5fXfyWE8Nl1z3oO4/yvluv+uyGE/yOEsG3v/eRynK+HEH7YXn+utmDdOO29vxZCyEII+8v/r42f7xv5bqcP+kdSWdIfSnpFUk3S70j6+DWN5bakTyz/bkt6Q9LHJf2Xkj63fP1zkn56+fePSvrHWpx1/0lJX/iAx/sfSfrfJP3y8v9/IOkzy79/VtJfWv7970v62eXfn5H09z/AMf4dSX9x+XdN0vZN46cWX97+NUkN4+Ofuwn8lPRvSPqEpN+3156Jf5J2JX11+Xtn+ffOBzDOH5JUWf790zbOjy/1fEPSR5f6X/4gbMG6cS5ff1HSr2vxpUP7183P922+1/pw6Qck/br9/5OSfvK6mbIcyz+S9Me02Dl7e/nabS02XEnS35T043Z9vO4DGNsLkj4v6Qcl/fJSAN8zZYp8XQrtDyz/riyvCx/AGLeWRjMkr98ofmph3L+5VNbKkp8/fFP4KenlxGg+E/8k/bikv2mv5657XuNM3vu3tfj+5cd0HH5+ULZg3Tgl/YKkPyLp61oZ92vl5/vxc92wDIoFfWv52rXSMtX+PklfkHSUrb4D9r6ko+Xf1zn2/0bSfyyJbw3Yk3SaZRkH6PhY4jiX758tr3/e9FFJDyX9T0v46H8MIbR0w/iZZdnbkv5rSW9JuqcFf35bN4+f0LPy7ybo2L+nRRSsJ4znWsYZQvgxSW9nWfY7yVs3apzfDl23cb9xFELYlPQPJf3VLMu6/l62cNXX2jsaQvgTkh5kWfbb1zmOK1BFixT4b2RZ9n2SelrACJFuCD93JP2YFs7ojqSWpB+5zjFdlW4C/55GIYSfkjSV9PPXPZaUQghNLb4P+j+77rE8D7pu4/62FngX9MLytWuhEEJVC8P+81mW/eLy5XdDCLeX79+W9GD5+nWN/Y9K+pMhhK9L+ntaQDP/raTtEAJnBflY4jiX729JOv4AxvktSd/KsuwLy/9/QQtjf9P4+W9K+lqWZQ+zLJtI+kUteHzT+Ak9K/+uTcdCCH9O0p+Q9GeWjkhPGM91jPNjWjj131nq0wuS/t8Qwq0bNs5vi67buH9R0qvLzoSaFgWqX7qOgYQQgqS/JekPsiz76/bWL0miIv5ZLbB4Xv+zy6r6JyWdWbr83CjLsp/MsuyFLMte1oJf/3eWZX9G0m9I+lOXjJPx/6nl9c892suy7L6kb4YQ/qXlS5+W9JpuGD+1gGM+GUJoLmWAcd4ofho9K/9+XdIPhRB2llnKDy1fe64UQvgRLaDDP5llWT8Z/2eWXUcflfSqpN/UNdiCLMt+L8uywyzLXl7q07e0aKq4rxvGz2+Lrhv016Iq/YYWlfKfusZx/OtapLi/K+mfL39+VAs89fOSviLp/5K0u7w+SPrvluP+PUnffw1j/pRW3TKvaKEkb0r63yVtLF+vL/9/c/n+Kx/g+L5X0m8tefp/atFdcOP4Kek/l/RlSb8v6X/RopPj2vkp6e9qUQeYaGF4/sK3wz8tMO83lz9//gMa55taYNPo0s/a9T+1HOfrkv64vf5cbcG6cSbvf12rguq18fP9+imOHyiooIIK+hDSdcMyBRVUUEEFPQcqjHtBBRVU0IeQCuNeUEEFFfQhpMK4F1RQQQV9CKkw7gUVVFBBH0IqjHtBBRVU0IeQCuNeUEEFFfQhpP8fP1fT1epRM1wAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "#Image size(256x1600)\n",
        "from google.colab.patches import cv2_imshow\n",
        "train_dir = '/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/train_images/'\n",
        "\n",
        "for i in range(10):\n",
        "  img = io.imread(os.path.join(train_dir, df_class.ImageId[i])) #Specify the path to the images given their image 10\n",
        "  plt.figure()\n",
        "  plt.title(df_class.ClassId[i])\n",
        "  plt.imshow(img)\n",
        "  #plt.axis('off')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0Si_cWAAXzbH"
      },
      "source": [
        "Convert RLE to Mask"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QpyGqhpTWC4z",
        "outputId": "fae8597e-f5cf-428f-b263-5f7ed4fd7ae9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data\n"
          ]
        }
      ],
      "source": [
        "%cd /content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/\n",
        "from utilities import rle2mask, mask2rle\n",
        "\n",
        "train_dir = '/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/train_images/'"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KCiPNO2ZZAwY"
      },
      "source": [
        "Using rle2mask on a sample defected image"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RRGSJs2KYE9K",
        "outputId": "10dace92-aa44-4319-89f1-e8acd75a15f5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(256, 1600)"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ],
      "source": [
        "image_index = 3\n",
        "mask = rle2mask(df_class.EncodedPixels[image_index], img.shape[0], img.shape[1])\n",
        "mask.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "id": "WKbaRa5AZX7A",
        "outputId": "4f36fc01-f17e-4834-ba94-15655956fb7e"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(-0.5, 1599.5, 255.5, -0.5)"
            ]
          },
          "metadata": {},
          "execution_count": 21
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAABDCAYAAAA/I9boAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAEAElEQVR4nO3dv28bZRzH8c9zZ59txXHTNgQVgpAQVaGioIqFEYFgQUIsLIg/oBMTAwsD/wMbA38AC5WQIlEhNiQQC79EkVAt0R9SUqVp4x9JbN9zTMBQx9fHnJ97fHm/Nst+7r7D6R0nuR8myzIBAPyIyh4AAE4SogsAHhFdAPCI6AKAR0QXADyqzXrzjehdTm2AJq+9rI8/+1yvtqwkqW8P9daVD9T86oepn48f39Db3/6mK2u3nfbz5aCtq7uXtf3+utI/u4+8Lu509Mw3R/r0ye8lSd1xX+999KHWrv4sOxw6zQAU4Zr9whz3Ht90kavxU1e7afu/16auSevYY0omSXSpedN5P++s9PXJE1vKmg2ndWmvp1vDtX9fD7NYp3/cIbgIEtFFUFJ+t0LFEV0E5ZfRhqLd+2WPASwM0UVQbo7Pyu73yh4DWBiii1ymvaLHavtljwFUAtFFrsm507pYH5Q9xiOrGysl9bLHAKYiusiVmePPVAjR07VE/fOnyh4DmIroonIiRbLxcv2gwMlBdBGUrZ0XZA8OndbEq6t6ff36giYCikV0EZSdQVuyqduipK6nkt3FDAQUjOgCgEdEFwA8IrpwNsxGSnq27DGApUR04ezOJFOru7eQbWcZZx2g2ogu5rOgB5ru/XFmIdsFQkF0kWuw2dRqlHjZVzzkmy6qjegi18F6pIbhslqgCEQXADwiugDgEdFF5UQy6m3GZY8BTEV0EYw0s+rc+P/biU2kww2e+4MwEV0EwypT8z4XXaDaiC4AeER0AcAjogsAHhFd5HJ+XM+clwhbzfn3XBMpmnct4BnRRa4HF9yCZu/taWv/Jef93JocqX3D/anDdnNDl5Id53VAGYgucmUrE7fPj0a6N15x3s9hFssM3B7VI0m2VdNaxKGM5cCRCgAeEV0A8IjoAoBHRBfBGGeRjOXyXVQb0UUwrg0uyt4t5lHqltv/IlBEF8Hop01lo1Eh2+o8X0y8gaIRXVRSUkvLHgGYiugCgEdEF7lMnUtsgaIQXcwWxXrl2a6XXaUZhyOqj6McM5nIqBG7XQY8r69vP1fYP9KAUBFdONtO2zJHxcdxf9Cc+w5lwLIgunD23eC87PbdsscAlhLRBQCPiC4AeER0EYwsc3xCxT/rovnWAWUgugiGud6ea93ehZY6UbPgaYDFILoIRu1gvnVp0yg2HMpYDhypqKQXz96RaTTKHgN4CNHFTCZJdK7xoOwxnF1u/yWTJGWPATyE6GKm6FRHb3Z+LXsMoDKILoJhuBsjTgCiiyAM7UhnfvdzjwegTEQXwYjG3EIS1Ud0AcAjk3FXJwDwhm+6AOAR0QUAj4guAHhEdAHAI6ILAB4RXQDw6G/289ZzkX0qDwAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plt.imshow(mask)\n",
        "plt.axis('off')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "id": "G1PDj2TsZb7C",
        "outputId": "00c3639d-1494-465a-891f-5b86c0c9ec93"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(-0.5, 1599.5, 255.5, -0.5)"
            ]
          },
          "metadata": {},
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "img = io.imread(os.path.join(train_dir, df_class.ImageId[image_index]))\n",
        "plt.imshow(img)\n",
        "img.shape\n",
        "plt.axis('off')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "21zzq1TjdbUA"
      },
      "source": [
        "Show the mask on top of the image"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 847
        },
        "id": "Uy8uS97NbFU0",
        "outputId": "84d86a02-dbdb-402f-bf59-131906658268"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "for i in range(10):\n",
        "  # Read the images using openCV and convert to to RGB format \n",
        "  img = io.imread(os.path.join(train_dir, df_class.ImageId[i]))\n",
        "  img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) #convert from BGR to RGB\n",
        "\n",
        "  # Get the mask for the image from rle\n",
        "  mask = rle2mask(df_class.EncodedPixels[i], img.shape[0], img.shape[1])\n",
        "\n",
        "  # Color the pixels with values = 1\n",
        "  img[mask == 1, 1] = 255\n",
        "  plt.figure()\n",
        "  plt.imshow(img)\n",
        "  plt.title(df_class.ClassId[i])\n",
        "  plt.axis('off')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bCk0FCQmfuR5"
      },
      "source": [
        "Build and Train a Deep Learning Model to Detect Wether a Defect is Present in an Image or Not"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 24,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "cQOjSAXkebkV",
        "outputId": "3668696f-1054-4d88-9964-026aa0121041"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "             ImageID  label\n",
              "0      0002cc93b.jpg      1\n",
              "1      0007a71bf.jpg      1\n",
              "2      000a4bcdd.jpg      1\n",
              "3      000f6bf48.jpg      1\n",
              "4      0014fce06.jpg      1\n",
              "...              ...    ...\n",
              "12992  0482ee1d6.jpg      0\n",
              "12993  04802a6c2.jpg      0\n",
              "12994  03ae2bc91.jpg      0\n",
              "12995  04238d7e3.jpg      0\n",
              "12996  023353d24.jpg      0\n",
              "\n",
              "[12997 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b2e5aa94-2575-458b-806c-5ce979c1e7c1\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ImageID</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0002cc93b.jpg</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0007a71bf.jpg</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>000a4bcdd.jpg</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>000f6bf48.jpg</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0014fce06.jpg</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12992</th>\n",
              "      <td>0482ee1d6.jpg</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12993</th>\n",
              "      <td>04802a6c2.jpg</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12994</th>\n",
              "      <td>03ae2bc91.jpg</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12995</th>\n",
              "      <td>04238d7e3.jpg</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12996</th>\n",
              "      <td>023353d24.jpg</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>12997 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b2e5aa94-2575-458b-806c-5ce979c1e7c1')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-b2e5aa94-2575-458b-806c-5ce979c1e7c1 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-b2e5aa94-2575-458b-806c-5ce979c1e7c1');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ],
      "source": [
        "df_all"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "xa9anGMgf413"
      },
      "outputs": [],
      "source": [
        "# Use sklearn to divide the data into training and testing\n",
        "from sklearn.model_selection import train_test_split\n",
        "train, test = train_test_split(df_all, test_size = 0.15)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L99ktEKogZxu",
        "outputId": "8a5dbe62-2e8c-4f5f-e2d4-97a950c3432c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(11047, 2)"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aT6Laxipgbmm",
        "outputId": "f0bb9d63-f3fd-4d82-83fd-c674b68d38fb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1950, 2)"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "test.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "R5mIx-2Sge_B"
      },
      "outputs": [],
      "source": [
        "train_dir = '/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/train_images/'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XE9srv58gpfj",
        "outputId": "56873221-e210-48d3-ba31-b16d5cd795fe"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 9390 validated image filenames.\n",
            "Found 1657 validated image filenames.\n"
          ]
        }
      ],
      "source": [
        "# Create an image generator fot the training and validation dataset\n",
        "# Dividing data into training, validation, and testing\n",
        "\n",
        "from keras_preprocessing.image import ImageDataGenerator\n",
        "\n",
        "#Create a Data Generator which scales the Data from 0 to 1 and makes validation split of 0.15\n",
        "datagen = ImageDataGenerator(rescale = 1./255., validation_split = 0.15)\n",
        "\n",
        "train_generator = datagen.flow_from_dataframe(\n",
        "    dataframe = train,\n",
        "    directory = train_dir,\n",
        "    x_col = 'ImageID',\n",
        "    y_col = 'label',\n",
        "    subset = 'training',\n",
        "    batch_size = 16,\n",
        "    shuffle = True,\n",
        "    class_mode = 'other',\n",
        "    target_size = (256, 256)\n",
        ")\n",
        "\n",
        "valid_generator = datagen.flow_from_dataframe(\n",
        "    dataframe = train,\n",
        "    directory = train_dir,\n",
        "    x_col = 'ImageID',\n",
        "    y_col = 'label',\n",
        "    subset = 'validation',\n",
        "    batch_size = 16,\n",
        "    shuffle = True,\n",
        "    class_mode = 'other',\n",
        "    target_size = (256, 256)\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KogYYzVFiTc7",
        "outputId": "84b22f2a-b55a-46eb-f8f2-d38aa5f480f0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 1950 validated image filenames.\n"
          ]
        }
      ],
      "source": [
        "# Create data generator for test images\n",
        "test_datagen = ImageDataGenerator(rescale = 1./255.)\n",
        "\n",
        "test_generator = test_datagen.flow_from_dataframe(\n",
        "    dataframe = test,\n",
        "    directory = train_dir,\n",
        "    x_col = 'ImageID',\n",
        "    y_col = None,\n",
        "    batch_size = 16,\n",
        "    shuffle = False,\n",
        "    class_mode = None,\n",
        "    target_size = (256, 256)\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6-oY63CUj_Pg"
      },
      "source": [
        "Download Pre-Trained Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UaiDWdl2jn6Z",
        "outputId": "29da9a78-c941-488f-a0db-7fcbb97bec3e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "94773248/94765736 [==============================] - 1s 0us/step\n",
            "94781440/94765736 [==============================] - 1s 0us/step\n"
          ]
        }
      ],
      "source": [
        "basemodel = ResNet50(weights = 'imagenet', include_top = False, input_tensor = Input(shape=(256,256,3)))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u5zbZi3Kj539",
        "outputId": "5583abac-dc24-4b76-d6c1-13c314cab25f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"resnet50\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_1 (InputLayer)           [(None, 256, 256, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " conv1_pad (ZeroPadding2D)      (None, 262, 262, 3)  0           ['input_1[0][0]']                \n",
            "                                                                                                  \n",
            " conv1_conv (Conv2D)            (None, 128, 128, 64  9472        ['conv1_pad[0][0]']              \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv1_bn (BatchNormalization)  (None, 128, 128, 64  256         ['conv1_conv[0][0]']             \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv1_relu (Activation)        (None, 128, 128, 64  0           ['conv1_bn[0][0]']               \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " pool1_pad (ZeroPadding2D)      (None, 130, 130, 64  0           ['conv1_relu[0][0]']             \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " pool1_pool (MaxPooling2D)      (None, 64, 64, 64)   0           ['pool1_pad[0][0]']              \n",
            "                                                                                                  \n",
            " conv2_block1_1_conv (Conv2D)   (None, 64, 64, 64)   4160        ['pool1_pool[0][0]']             \n",
            "                                                                                                  \n",
            " conv2_block1_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block1_0_conv (Conv2D)   (None, 64, 64, 256)  16640       ['pool1_pool[0][0]']             \n",
            "                                                                                                  \n",
            " conv2_block1_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block1_0_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block1_add (Add)         (None, 64, 64, 256)  0           ['conv2_block1_0_bn[0][0]',      \n",
            "                                                                  'conv2_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block1_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block2_1_conv (Conv2D)   (None, 64, 64, 64)   16448       ['conv2_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block2_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block2_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block2_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block2_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block2_add (Add)         (None, 64, 64, 256)  0           ['conv2_block1_out[0][0]',       \n",
            "                                                                  'conv2_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block2_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block3_1_conv (Conv2D)   (None, 64, 64, 64)   16448       ['conv2_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv2_block3_1_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_1_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_2_conv (Conv2D)   (None, 64, 64, 64)   36928       ['conv2_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block3_2_bn (BatchNormal  (None, 64, 64, 64)  256         ['conv2_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_2_relu (Activatio  (None, 64, 64, 64)  0           ['conv2_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv2_block3_3_conv (Conv2D)   (None, 64, 64, 256)  16640       ['conv2_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv2_block3_3_bn (BatchNormal  (None, 64, 64, 256)  1024       ['conv2_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv2_block3_add (Add)         (None, 64, 64, 256)  0           ['conv2_block2_out[0][0]',       \n",
            "                                                                  'conv2_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv2_block3_out (Activation)  (None, 64, 64, 256)  0           ['conv2_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_1_conv (Conv2D)   (None, 32, 32, 128)  32896       ['conv2_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block1_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block1_0_conv (Conv2D)   (None, 32, 32, 512)  131584      ['conv2_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block1_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block1_0_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block1_add (Add)         (None, 32, 32, 512)  0           ['conv3_block1_0_bn[0][0]',      \n",
            "                                                                  'conv3_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block1_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block2_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block2_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block2_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block2_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block2_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block2_add (Add)         (None, 32, 32, 512)  0           ['conv3_block1_out[0][0]',       \n",
            "                                                                  'conv3_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block2_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block3_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block3_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block3_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block3_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block3_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block3_add (Add)         (None, 32, 32, 512)  0           ['conv3_block2_out[0][0]',       \n",
            "                                                                  'conv3_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block3_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block4_1_conv (Conv2D)   (None, 32, 32, 128)  65664       ['conv3_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv3_block4_1_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block4_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_1_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block4_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_2_conv (Conv2D)   (None, 32, 32, 128)  147584      ['conv3_block4_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block4_2_bn (BatchNormal  (None, 32, 32, 128)  512        ['conv3_block4_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_2_relu (Activatio  (None, 32, 32, 128)  0          ['conv3_block4_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv3_block4_3_conv (Conv2D)   (None, 32, 32, 512)  66048       ['conv3_block4_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv3_block4_3_bn (BatchNormal  (None, 32, 32, 512)  2048       ['conv3_block4_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv3_block4_add (Add)         (None, 32, 32, 512)  0           ['conv3_block3_out[0][0]',       \n",
            "                                                                  'conv3_block4_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv3_block4_out (Activation)  (None, 32, 32, 512)  0           ['conv3_block4_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block1_1_conv (Conv2D)   (None, 16, 16, 256)  131328      ['conv3_block4_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block1_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block1_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block1_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block1_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block1_0_conv (Conv2D)   (None, 16, 16, 1024  525312      ['conv3_block4_out[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block1_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_0_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block1_0_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block1_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block1_add (Add)         (None, 16, 16, 1024  0           ['conv4_block1_0_bn[0][0]',      \n",
            "                                )                                 'conv4_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block1_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block1_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block2_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block2_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block2_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block2_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block2_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block2_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block2_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block2_add (Add)         (None, 16, 16, 1024  0           ['conv4_block1_out[0][0]',       \n",
            "                                )                                 'conv4_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block2_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block2_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block3_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block3_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block3_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block3_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block3_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block3_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block3_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block3_add (Add)         (None, 16, 16, 1024  0           ['conv4_block2_out[0][0]',       \n",
            "                                )                                 'conv4_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block3_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block3_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block3_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block4_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block4_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block4_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block4_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block4_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block4_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block4_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block4_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block4_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block4_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block4_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block4_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block4_add (Add)         (None, 16, 16, 1024  0           ['conv4_block3_out[0][0]',       \n",
            "                                )                                 'conv4_block4_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block4_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block4_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block4_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block5_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block5_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block5_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block5_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block5_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block5_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block5_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block5_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block5_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block5_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block5_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block5_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block5_add (Add)         (None, 16, 16, 1024  0           ['conv4_block4_out[0][0]',       \n",
            "                                )                                 'conv4_block5_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block5_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block5_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_1_conv (Conv2D)   (None, 16, 16, 256)  262400      ['conv4_block5_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv4_block6_1_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block6_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block6_1_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block6_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_2_conv (Conv2D)   (None, 16, 16, 256)  590080      ['conv4_block6_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv4_block6_2_bn (BatchNormal  (None, 16, 16, 256)  1024       ['conv4_block6_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv4_block6_2_relu (Activatio  (None, 16, 16, 256)  0          ['conv4_block6_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv4_block6_3_conv (Conv2D)   (None, 16, 16, 1024  263168      ['conv4_block6_2_relu[0][0]']    \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_3_bn (BatchNormal  (None, 16, 16, 1024  4096       ['conv4_block6_3_conv[0][0]']    \n",
            " ization)                       )                                                                 \n",
            "                                                                                                  \n",
            " conv4_block6_add (Add)         (None, 16, 16, 1024  0           ['conv4_block5_out[0][0]',       \n",
            "                                )                                 'conv4_block6_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv4_block6_out (Activation)  (None, 16, 16, 1024  0           ['conv4_block6_add[0][0]']       \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv5_block1_1_conv (Conv2D)   (None, 8, 8, 512)    524800      ['conv4_block6_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block1_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block1_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block1_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block1_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block1_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block1_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block1_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block1_0_conv (Conv2D)   (None, 8, 8, 2048)   2099200     ['conv4_block6_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block1_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block1_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block1_0_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block1_0_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block1_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block1_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block1_0_bn[0][0]',      \n",
            "                                                                  'conv5_block1_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block1_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block1_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block2_1_conv (Conv2D)   (None, 8, 8, 512)    1049088     ['conv5_block1_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block2_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block2_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block2_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block2_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block2_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block2_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block2_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block2_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block2_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block2_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block2_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block2_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block1_out[0][0]',       \n",
            "                                                                  'conv5_block2_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block2_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block2_add[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block3_1_conv (Conv2D)   (None, 8, 8, 512)    1049088     ['conv5_block2_out[0][0]']       \n",
            "                                                                                                  \n",
            " conv5_block3_1_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block3_1_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_1_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block3_1_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_2_conv (Conv2D)   (None, 8, 8, 512)    2359808     ['conv5_block3_1_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block3_2_bn (BatchNormal  (None, 8, 8, 512)   2048        ['conv5_block3_2_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_2_relu (Activatio  (None, 8, 8, 512)   0           ['conv5_block3_2_bn[0][0]']      \n",
            " n)                                                                                               \n",
            "                                                                                                  \n",
            " conv5_block3_3_conv (Conv2D)   (None, 8, 8, 2048)   1050624     ['conv5_block3_2_relu[0][0]']    \n",
            "                                                                                                  \n",
            " conv5_block3_3_bn (BatchNormal  (None, 8, 8, 2048)  8192        ['conv5_block3_3_conv[0][0]']    \n",
            " ization)                                                                                         \n",
            "                                                                                                  \n",
            " conv5_block3_add (Add)         (None, 8, 8, 2048)   0           ['conv5_block2_out[0][0]',       \n",
            "                                                                  'conv5_block3_3_bn[0][0]']      \n",
            "                                                                                                  \n",
            " conv5_block3_out (Activation)  (None, 8, 8, 2048)   0           ['conv5_block3_add[0][0]']       \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 23,587,712\n",
            "Trainable params: 23,534,592\n",
            "Non-trainable params: 53,120\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "basemodel.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "id": "RpaWZQEkkI4R"
      },
      "outputs": [],
      "source": [
        "#freeze the model weights\n",
        "\n",
        "for layer in basemodel.layers:\n",
        "  layers.trainable = False\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 34,
      "metadata": {
        "id": "KzAbYreTkYbv"
      },
      "outputs": [],
      "source": [
        "headmodel = basemodel.output\n",
        "headmodel = AveragePooling2D(pool_size = (4,4))(headmodel)\n",
        "headmodel = Flatten(name= 'flatten')(headmodel)\n",
        "headmodel = Dense(256, activation = 'relu')(headmodel)\n",
        "headmodel = Dropout(0.3)(headmodel)\n",
        "headmodel = Dense(1, activation = 'sigmoid')(headmodel)\n",
        "\n",
        "#concatenate basemodel & headmodel named model\n",
        "model = Model(inputs = basemodel.input, outputs = headmodel)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 35,
      "metadata": {
        "id": "At_3izhHk7Uq"
      },
      "outputs": [],
      "source": [
        "model.compile(loss = 'binary_crossentropy', optimizer = 'Nadam', metrics =['accuracy'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "id": "YHRKk6K1lGlA"
      },
      "outputs": [],
      "source": [
        "# Use EarlyStopping to exit traing if validation loss is not decreasing even after certain epochs (patience)\n",
        "earlystopping = EarlyStopping(monitor = 'val_loss', mode = 'min', verbose = 1, patience =20)\n",
        "\n",
        "# Save the best model with least validation loss\n",
        "checkpointer = ModelCheckpoint(filepath='resnet-weights.hdf5', verbose = 1, save_best_only = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H_5R-guimeXw",
        "outputId": "a5821f8c-fe52-4e43-806f-93a461848c02"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:1: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.5794 - accuracy: 0.7344 \n",
            "Epoch 1: val_loss improved from inf to 0.73549, saving model to resnet-weights.hdf5\n",
            "586/586 [==============================] - 7647s 13s/step - loss: 0.5794 - accuracy: 0.7344 - val_loss: 0.7355 - val_accuracy: 0.5413\n",
            "Epoch 2/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.4166 - accuracy: 0.8074\n",
            "Epoch 2: val_loss did not improve from 0.73549\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.4166 - accuracy: 0.8074 - val_loss: 0.7411 - val_accuracy: 0.5843\n",
            "Epoch 3/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.3534 - accuracy: 0.8478\n",
            "Epoch 3: val_loss did not improve from 0.73549\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.3534 - accuracy: 0.8478 - val_loss: 1.1042 - val_accuracy: 0.6590\n",
            "Epoch 4/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.3258 - accuracy: 0.8596\n",
            "Epoch 4: val_loss improved from 0.73549 to 0.53198, saving model to resnet-weights.hdf5\n",
            "586/586 [==============================] - 163s 278ms/step - loss: 0.3258 - accuracy: 0.8596 - val_loss: 0.5320 - val_accuracy: 0.7209\n",
            "Epoch 5/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.2885 - accuracy: 0.8800\n",
            "Epoch 5: val_loss did not improve from 0.53198\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.2885 - accuracy: 0.8800 - val_loss: 1.7519 - val_accuracy: 0.5704\n",
            "Epoch 6/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.2615 - accuracy: 0.8908\n",
            "Epoch 6: val_loss improved from 0.53198 to 0.47273, saving model to resnet-weights.hdf5\n",
            "586/586 [==============================] - 164s 279ms/step - loss: 0.2615 - accuracy: 0.8908 - val_loss: 0.4727 - val_accuracy: 0.8289\n",
            "Epoch 7/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.2495 - accuracy: 0.8964\n",
            "Epoch 7: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.2495 - accuracy: 0.8964 - val_loss: 1.1701 - val_accuracy: 0.5740\n",
            "Epoch 8/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.2249 - accuracy: 0.9083\n",
            "Epoch 8: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.2249 - accuracy: 0.9083 - val_loss: 0.7170 - val_accuracy: 0.6638\n",
            "Epoch 9/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.2263 - accuracy: 0.9092\n",
            "Epoch 9: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.2263 - accuracy: 0.9092 - val_loss: 0.5864 - val_accuracy: 0.8034\n",
            "Epoch 10/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1983 - accuracy: 0.9232\n",
            "Epoch 10: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1983 - accuracy: 0.9232 - val_loss: 0.6295 - val_accuracy: 0.7797\n",
            "Epoch 11/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1818 - accuracy: 0.9288\n",
            "Epoch 11: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1818 - accuracy: 0.9288 - val_loss: 6.6409 - val_accuracy: 0.5510\n",
            "Epoch 12/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1741 - accuracy: 0.9318\n",
            "Epoch 12: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1741 - accuracy: 0.9318 - val_loss: 0.5814 - val_accuracy: 0.8325\n",
            "Epoch 13/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1718 - accuracy: 0.9357\n",
            "Epoch 13: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1718 - accuracy: 0.9357 - val_loss: 3.6565 - val_accuracy: 0.7209\n",
            "Epoch 14/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1721 - accuracy: 0.9370\n",
            "Epoch 14: val_loss did not improve from 0.47273\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1721 - accuracy: 0.9370 - val_loss: 1.3149 - val_accuracy: 0.6754\n",
            "Epoch 15/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1374 - accuracy: 0.9466\n",
            "Epoch 15: val_loss improved from 0.47273 to 0.39300, saving model to resnet-weights.hdf5\n",
            "586/586 [==============================] - 163s 278ms/step - loss: 0.1374 - accuracy: 0.9466 - val_loss: 0.3930 - val_accuracy: 0.8441\n",
            "Epoch 16/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1295 - accuracy: 0.9508\n",
            "Epoch 16: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1295 - accuracy: 0.9508 - val_loss: 0.7498 - val_accuracy: 0.7306\n",
            "Epoch 17/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1240 - accuracy: 0.9558\n",
            "Epoch 17: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1240 - accuracy: 0.9558 - val_loss: 1.4591 - val_accuracy: 0.5728\n",
            "Epoch 18/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1177 - accuracy: 0.9546\n",
            "Epoch 18: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1177 - accuracy: 0.9546 - val_loss: 2.3876 - val_accuracy: 0.5886\n",
            "Epoch 19/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.1002 - accuracy: 0.9626\n",
            "Epoch 19: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.1002 - accuracy: 0.9626 - val_loss: 1.7605 - val_accuracy: 0.7403\n",
            "Epoch 20/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0991 - accuracy: 0.9620\n",
            "Epoch 20: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0991 - accuracy: 0.9620 - val_loss: 0.9000 - val_accuracy: 0.7822\n",
            "Epoch 21/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0877 - accuracy: 0.9666\n",
            "Epoch 21: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0877 - accuracy: 0.9666 - val_loss: 3.0242 - val_accuracy: 0.6869\n",
            "Epoch 22/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0758 - accuracy: 0.9731\n",
            "Epoch 22: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0758 - accuracy: 0.9731 - val_loss: 1.3727 - val_accuracy: 0.8016\n",
            "Epoch 23/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0744 - accuracy: 0.9713\n",
            "Epoch 23: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0744 - accuracy: 0.9713 - val_loss: 1.4862 - val_accuracy: 0.7051\n",
            "Epoch 24/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0786 - accuracy: 0.9721\n",
            "Epoch 24: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0786 - accuracy: 0.9721 - val_loss: 6.8926 - val_accuracy: 0.6826\n",
            "Epoch 25/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0733 - accuracy: 0.9749\n",
            "Epoch 25: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0733 - accuracy: 0.9749 - val_loss: 0.7690 - val_accuracy: 0.7615\n",
            "Epoch 26/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0634 - accuracy: 0.9792\n",
            "Epoch 26: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0634 - accuracy: 0.9792 - val_loss: 0.7780 - val_accuracy: 0.8908\n",
            "Epoch 27/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0573 - accuracy: 0.9815\n",
            "Epoch 27: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0573 - accuracy: 0.9815 - val_loss: 1.4141 - val_accuracy: 0.7251\n",
            "Epoch 28/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0602 - accuracy: 0.9781\n",
            "Epoch 28: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 160s 272ms/step - loss: 0.0602 - accuracy: 0.9781 - val_loss: 0.8685 - val_accuracy: 0.7567\n",
            "Epoch 29/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0535 - accuracy: 0.9811\n",
            "Epoch 29: val_loss did not improve from 0.39300\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0535 - accuracy: 0.9811 - val_loss: 0.9085 - val_accuracy: 0.8805\n",
            "Epoch 30/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0538 - accuracy: 0.9824\n",
            "Epoch 30: val_loss improved from 0.39300 to 0.30868, saving model to resnet-weights.hdf5\n",
            "586/586 [==============================] - 164s 280ms/step - loss: 0.0538 - accuracy: 0.9824 - val_loss: 0.3087 - val_accuracy: 0.9053\n",
            "Epoch 31/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0373 - accuracy: 0.9873\n",
            "Epoch 31: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0373 - accuracy: 0.9873 - val_loss: 0.8853 - val_accuracy: 0.8004\n",
            "Epoch 32/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0445 - accuracy: 0.9865\n",
            "Epoch 32: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0445 - accuracy: 0.9865 - val_loss: 0.3924 - val_accuracy: 0.9059\n",
            "Epoch 33/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0573 - accuracy: 0.9814\n",
            "Epoch 33: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0573 - accuracy: 0.9814 - val_loss: 0.9042 - val_accuracy: 0.7603\n",
            "Epoch 34/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0357 - accuracy: 0.9871\n",
            "Epoch 34: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 160s 272ms/step - loss: 0.0357 - accuracy: 0.9871 - val_loss: 1.3186 - val_accuracy: 0.7421\n",
            "Epoch 35/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0355 - accuracy: 0.9891\n",
            "Epoch 35: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 272ms/step - loss: 0.0355 - accuracy: 0.9891 - val_loss: 5.1790 - val_accuracy: 0.7112\n",
            "Epoch 36/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0455 - accuracy: 0.9841\n",
            "Epoch 36: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0455 - accuracy: 0.9841 - val_loss: 1.8231 - val_accuracy: 0.7973\n",
            "Epoch 37/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0331 - accuracy: 0.9882\n",
            "Epoch 37: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0331 - accuracy: 0.9882 - val_loss: 0.4520 - val_accuracy: 0.8768\n",
            "Epoch 38/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0275 - accuracy: 0.9917\n",
            "Epoch 38: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0275 - accuracy: 0.9917 - val_loss: 10.5196 - val_accuracy: 0.6195\n",
            "Epoch 39/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0364 - accuracy: 0.9886\n",
            "Epoch 39: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0364 - accuracy: 0.9886 - val_loss: 9.1963 - val_accuracy: 0.7106\n",
            "Epoch 40/40\n",
            "586/586 [==============================] - ETA: 0s - loss: 0.0359 - accuracy: 0.9895\n",
            "Epoch 40: val_loss did not improve from 0.30868\n",
            "586/586 [==============================] - 159s 271ms/step - loss: 0.0359 - accuracy: 0.9895 - val_loss: 0.7688 - val_accuracy: 0.7743\n"
          ]
        }
      ],
      "source": [
        "history = model.fit_generator(train_generator, steps_per_epoch= train_generator.n // 16, epochs = 40, validation_data = valid_generator, validation_steps = valid_generator.n // 16, callbacks=[checkpointer, earlystopping] )"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0wYUWKT3kSOE"
      },
      "source": [
        "### Save the Trained Model Architecture"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "metadata": {
        "id": "wc3mTizlnMO-"
      },
      "outputs": [],
      "source": [
        "model_json = model.to_json()\n",
        "with open('resnet-classifier-model.json','w') as json_file:\n",
        "  json_file.write(model_json)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vtut7PHslB8w"
      },
      "source": [
        "### Assess Train Model Performance"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "metadata": {
        "id": "1axajmabknKN"
      },
      "outputs": [],
      "source": [
        "with open('/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/resnet-classifier-model.json', 'r') as json_file:\n",
        "  json_savedModel = json_file.read()\n",
        "\n",
        "#Load the Model\n",
        "model = tf.keras.models.model_from_json(json_savedModel)\n",
        "model.load_weights('/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/resnet-weights.hdf5')\n",
        "model.compile(loss = 'binary_crossentropy', optimizer='Nadam', metrics = ['accuracy'])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QaulZ7R_nqDW"
      },
      "source": [
        "### Prediction Making"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "id": "imQZCu3Lm5fl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2c436042-d3bc-4d7d-ccd2-3100877ee69d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "121/121 [==============================] - 1281s 11s/step\n"
          ]
        }
      ],
      "source": [
        "from keras_preprocessing.image import ImageDataGenerator\n",
        "\n",
        "test_predict = model.predict(test_generator, steps = test_generator.n // 16, verbose = 1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "metadata": {
        "id": "5EDa5FrWn89I",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b687f121-2e17-464f-fe7b-56a37e4b1f97"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1.0000000e+00],\n",
              "       [2.6874841e-04],\n",
              "       [9.0780614e-14],\n",
              "       ...,\n",
              "       [6.0948956e-33],\n",
              "       [9.9728990e-01],\n",
              "       [9.9995804e-01]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ],
      "source": [
        "test_predict"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "metadata": {
        "id": "DtuCB1qroSq6"
      },
      "outputs": [],
      "source": [
        "# Sigmoid activation will give the result values from 0 to 1.\n",
        "# The network is initially used to classify wether the image has defect or not\n",
        "# These images (defective) is passed through the segmentation network to get the localization and type of defect\n",
        "\n",
        "predict = []\n",
        "\n",
        "for i in test_predict:\n",
        "  if i < 0.5:\n",
        "    predict.append(0)\n",
        "  else:\n",
        "    predict.append(1)\n",
        "\n",
        "predict = np.asarray(predict)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "metadata": {
        "id": "fdOcAh8zpIVe",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0f24f056-63f4-4b75-a836-a631c11907c3"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1936"
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ],
      "source": [
        "len(predict)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "id": "xgWAb6P5pKJ7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "25e968d5-c29e-4609-9fa0-7d7e27f6cfb0"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1936"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ],
      "source": [
        "# 1936 are get from test generator, due to batch size\n",
        "\n",
        "original = np.asarray(test.label)[:1936]\n",
        "len(original)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6L2gpoK1pjy8"
      },
      "source": [
        "### Find the Accuracy Score"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "metadata": {
        "id": "dr7tyeagpgm6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "73971898-9876-48db-c0c9-4500eee23a93"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9106404958677686"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ],
      "source": [
        "#From the testing data\n",
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "accuracy = accuracy_score(original, predict)\n",
        "accuracy"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "metadata": {
        "id": "rn7gqEbXpvyY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 446
        },
        "outputId": "2e7b06cb-a620-4df1-8e44-a6db51460531"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fd91dda64d0>"
            ]
          },
          "metadata": {},
          "execution_count": 46
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 504x504 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "#Confusion Matrix\n",
        "from sklearn.metrics import confusion_matrix\n",
        "\n",
        "cm = confusion_matrix(original, predict)\n",
        "plt.figure(figsize = (7,7))\n",
        "sns.heatmap(cm, annot = True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "metadata": {
        "id": "90eI_ihxrLWO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "06484191-0417-4b8f-f236-d3575c135363"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.88      0.92      0.90       859\n",
            "           1       0.93      0.90      0.92      1077\n",
            "\n",
            "    accuracy                           0.91      1936\n",
            "   macro avg       0.91      0.91      0.91      1936\n",
            "weighted avg       0.91      0.91      0.91      1936\n",
            "\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import classification_report\n",
        "\n",
        "report = classification_report(original,predict, labels = [0,1])\n",
        "print(report)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gO82YXRB07uu"
      },
      "source": [
        "Build A ResUnet Segmentation Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 63,
      "metadata": {
        "id": "gleJFSWArnWa"
      },
      "outputs": [],
      "source": [
        "#Data splitting into train and test\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train, X_val = train_test_split(df_class, test_size = 0.2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 49,
      "metadata": {
        "id": "L8eGHyKg1Ee6"
      },
      "outputs": [],
      "source": [
        "# Createing separate list imageId, classId and rle to pass into generator\n",
        "\n",
        "train_ids = list(X_train.ImageId)\n",
        "train_class = list(X_train.ClassId)\n",
        "train_rle = list(X_train.EncodedPixels)\n",
        "\n",
        "val_ids = list(X_val.ImageId)\n",
        "val_class = list(X_val.ClassId)\n",
        "val_rle = list(X_val.EncodedPixels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 50,
      "metadata": {
        "id": "EKOGDSuU30p4"
      },
      "outputs": [],
      "source": [
        "from utilities import DataGenerator\n",
        "\n",
        "#Creating Image Generators\n",
        "\n",
        "training_generator = DataGenerator(train_ids, train_class, train_rle, train_dir)\n",
        "validation_generator = DataGenerator(val_ids, val_class, val_rle, train_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 51,
      "metadata": {
        "id": "HC8qjQgG4XAv"
      },
      "outputs": [],
      "source": [
        "def resblock(X, f): #residual block\n",
        "  \n",
        "  # making a copy of input\n",
        "  X_copy = X\n",
        "\n",
        "  # Main Path\n",
        "  X = Conv2D(f, kernel_size = (1,1), strides = (1,1), kernel_initializer ='he_normal')(X)\n",
        "  X = BatchNormalization()(X)\n",
        "  X = Activation('relu')(X) \n",
        "\n",
        "  X = Conv2D(f, kernel_size = (3,3), strides =(1,1), padding = 'same', kernel_initializer ='he_normal')(X)\n",
        "  X = BatchNormalization()(X)\n",
        "\n",
        "  # Short Path\n",
        "  X_copy = Conv2D(f, kernel_size = (1,1), strides =(1,1), kernel_initializer ='he_normal')(X_copy)\n",
        "  X_copy = BatchNormalization()(X_copy)\n",
        "\n",
        "  # Adding the output from main path and short path together\n",
        "  X = Add()([X,X_copy])\n",
        "  X = Activation('relu')(X)\n",
        "\n",
        "  return X"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 52,
      "metadata": {
        "id": "d8-QmQaF6Pqk"
      },
      "outputs": [],
      "source": [
        "# function to upscale and concatenating the values passsed\n",
        "def upsample_concat(x, skip):\n",
        "  x = UpSampling2D((2,2))(x)\n",
        "  merge = Concatenate()([x, skip])\n",
        "\n",
        "  return merge"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 53,
      "metadata": {
        "id": "fu9iCLmv7iwK"
      },
      "outputs": [],
      "source": [
        "input_shape = (256,256,1)\n",
        "\n",
        "#Input tensor shape\n",
        "X_input = Input(input_shape)\n",
        "\n",
        "#Stage 1\n",
        "conv1_in = Conv2D(16,3,activation= 'relu', padding = 'same', kernel_initializer ='he_normal')(X_input)\n",
        "conv1_in = BatchNormalization()(conv1_in)\n",
        "conv1_in = Conv2D(16,3,activation= 'relu', padding = 'same', kernel_initializer ='he_normal')(conv1_in)\n",
        "conv1_in = BatchNormalization()(conv1_in)\n",
        "pool_1 = MaxPool2D(pool_size = (2,2))(conv1_in)\n",
        "\n",
        "#Stage 2\n",
        "conv2_in = resblock(pool_1, 32)\n",
        "pool_2 = MaxPool2D(pool_size = (2,2))(conv2_in)\n",
        "\n",
        "#Stage 3\n",
        "conv3_in = resblock(pool_2, 64)\n",
        "pool_3 = MaxPool2D(pool_size = (2,2))(conv3_in)\n",
        "\n",
        "#Stage 4\n",
        "conv4_in = resblock(pool_3, 128)\n",
        "pool_4 = MaxPool2D(pool_size = (2,2))(conv4_in)\n",
        "\n",
        "#Stage 5\n",
        "conv5_in = resblock(pool_4, 256)\n",
        "\n",
        "#Upscale stage 1\n",
        "up_1 = upsample_concat(conv5_in, conv4_in)\n",
        "up_1 = resblock(up_1, 128)\n",
        "\n",
        "#Upscale stage 2\n",
        "up_2 = upsample_concat(up_1, conv3_in)\n",
        "up_2 = resblock(up_2, 64)\n",
        "\n",
        "#Upscale stage 3\n",
        "up_3 = upsample_concat(up_2, conv2_in)\n",
        "up_3 = resblock(up_3, 32)\n",
        "\n",
        "#Upscale stage 4\n",
        "up_4 = upsample_concat(up_3, conv1_in)\n",
        "up_4 = resblock(up_4, 16)\n",
        "\n",
        "#Final Output\n",
        "output = Conv2D(4, (1,1), padding = \"same\", activation = \"sigmoid\")(up_4)\n",
        "\n",
        "model_seg = Model(inputs = X_input, outputs = output)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Loss Function"
      ],
      "metadata": {
        "id": "y_LTi1uJohWD"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 54,
      "metadata": {
        "id": "qtwUolxw-33G",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4998c60c-4ec8-41e1-d4c9-efac1d33a7c2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data\n"
          ]
        }
      ],
      "source": [
        "%cd /content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data\n",
        "\n",
        "from utilities import focal_tversky, tversky_loss, tversky"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 55,
      "metadata": {
        "id": "ZITQnoMyFFSl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a8f672f2-d85b-403b-b0e6-eea8d12a3f2e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(Adam, self).__init__(name, **kwargs)\n"
          ]
        }
      ],
      "source": [
        "adam = tf.keras.optimizers.Adam(lr = 0.05, epsilon = 0.1)\n",
        "model_seg.compile(optimizer = adam, loss = focal_tversky, metrics = [tversky])"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Using EarlyStopping to exit training if validation loss is not decreasing even after certain epochs (patience)\n",
        "earlystopping = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=20)\n",
        "\n",
        "# save the best model with lower validation loss\n",
        "checkpointer = ModelCheckpoint(filepath=\"resunet-segmentation-weights.hdf5\", verbose=1, save_best_only=True)"
      ],
      "metadata": {
        "id": "7FxhTmX2pmh-"
      },
      "execution_count": 61,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history = model_seg.fit_generator(training_generator, epochs = 40, validation_data= validation_generator, callbacks=[checkpointer, earlystopping])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 553
        },
        "id": "N4-XxyVZpuIa",
        "outputId": "8befd02c-af91-452d-d58b-869cd425def9"
      },
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:1: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/40\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "TypeError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-65-0fb4d5307fb1>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mhistory\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_seg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_generator\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtraining_generator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m40\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m \u001b[0mvalidation_generator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallbacks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcheckpointer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mearlystopping\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, validation_freq, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m   2221\u001b[0m         \u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2222\u001b[0m         \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mshuffle\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2223\u001b[0;31m         initial_epoch=initial_epoch)\n\u001b[0m\u001b[1;32m   2224\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2225\u001b[0m   \u001b[0;34m@\u001b[0m\u001b[0mdoc_controls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdo_not_generate_docs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/keras/utils/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     65\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 67\u001b[0;31m       \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfiltered_tb\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     68\u001b[0m     \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     69\u001b[0m       \u001b[0;32mdel\u001b[0m \u001b[0mfiltered_tb\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mautograph_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m   1145\u001b[0m           \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1146\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"ag_error_metadata\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1147\u001b[0;31m               \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1148\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1149\u001b[0m               \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mTypeError\u001b[0m: in user code:\n\n    File \"/usr/local/lib/python3.7/dist-packages/keras/engine/training.py\", line 1021, in train_function  *\n        return step_function(self, iterator)\n    File \"/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/utilities.py\", line 313, in focal_tversky  *\n        pt_1 = tversky(y_true, y_pred)\n    File \"/content/drive/My Drive/Colab Notebooks/Data Science for Business/3. Manufacturing Data/utilities.py\", line 303, in tversky  *\n        true_pos = K.sum(y_true_pos * y_pred_pos)\n\n    TypeError: Input 'y' of 'Mul' Op has type float32 that does not match type int64 of argument 'x'.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "UkGyFaygqtlx"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "name": "Surface Defect Detection.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyNYAbXMO/02Hs2uB4vuQYv7",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}