TA_NewScenario.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TA_NewScenario.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/rfadhila07/Test-result/blob/main/TA_NewScenario.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qe-rJZTxEbAq",
        "outputId": "4c8baa5e-f2f5-4e88-a345-32ef0d981347"
      },
      "source": [
        "#Connect your Google Drive \n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kUer0bZKEe_T"
      },
      "source": [
        "import numpy as np\n",
        "import os\n",
        "from imutils import paths\n",
        "from PIL import Image\n",
        "import matplotlib.pyplot as plt\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers.convolutional import Conv2D\n",
        "from keras.layers.convolutional import MaxPooling2D\n",
        "from keras.layers.core import Activation\n",
        "from keras.layers.core import Flatten\n",
        "from keras.layers.core import Dense\n",
        "from keras.layers import GlobalAveragePooling2D, BatchNormalization, Dropout\n",
        "from keras.applications.vgg16 import VGG16\n",
        "from keras.models import Model\n",
        "import tensorflow as tf\n",
        "import cv2\n",
        "import io\n",
        "from skimage.filters import gaussian\n",
        "from skimage import exposure\n",
        "import matplotlib.image as mpimg\n",
        "import tensorflow\n",
        "from tensorflow.keras import Model, Input\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, ReLU, BatchNormalization\n",
        "from tensorflow.keras.layers import Add, GlobalAveragePooling2D\n",
        "from skimage import io\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.model_selection import train_test_split\n",
        "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "from keras.optimizers import Adam, SGD, RMSprop, Adamax, Nadam\n",
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, classification_report\n",
        "from keras import backend as k\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "import keras\n",
        "import numpy as np\n",
        "import cv2\n",
        "import os\n",
        "import random\n",
        "import shutil\n",
        "import pandas as pd\n",
        "import csv\n",
        "import zipfile\n",
        "from keras import optimizers\n",
        "from keras.models import Sequential,Model\n",
        "from keras.layers import Dropout, Flatten, Dense,Input, Conv2D, MaxPooling2D\n",
        "from keras.applications.resnet_v2 import ResNet50V2\n",
        "from keras.applications.xception import Xception\n",
        "from keras.applications.resnet50 import ResNet50\n",
        "from keras.applications.vgg16 import VGG16\n",
        "from keras.callbacks import ModelCheckpoint\n",
        "from keras.applications.imagenet_utils import preprocess_input\n",
        "from keras import backend as K\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.initializers import RandomNormal\n",
        "import keras.backend as k\n",
        "from sklearn.utils import shuffle\n",
        "import io\n",
        "from PIL import Image as pil_image\n",
        "# from keras_retinanet import layers\n",
        "import keras.backend as k\n",
        "# import keras_retinanet"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TOqR32OLElDQ"
      },
      "source": [
        "fold_num=1 #Select Fold Number"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WUrp6Ew1EqES"
      },
      "source": [
        "#Here we set the data generators for applying data augmentation methods\n",
        "train_datagen = ImageDataGenerator()#horizontal_flip=True,vertical_flip=True,zoom_range=0.05,rotation_range=360,width_shift_range=0.05,height_shift_range=0.05,shear_range=0.05)\n",
        "test_datagen = ImageDataGenerator()\n",
        "train_df =pd.read_csv('/content/drive/MyDrive/COVID-CTset/CSV/train{}.csv'.format(fold_num)) #raed train csv file\n",
        "validation_df = pd.read_csv('/content/drive/MyDrive/COVID-CTset/CSV/validation{}.csv'.format(fold_num)) #raed validation csv file (Validation in the training process)\n",
        "train_df = shuffle(train_df) #Shuffle the train data\n",
        "test_df = pd.read_csv('/content/drive/MyDrive/COVID-CTset/CSV/test{}.csv'.format(fold_num))#raed test csv file (For evaluating the final version of th"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "qYQYl3s5EsN9",
        "outputId": "c2d0a609-0e48-447a-de51-50f1d3383435"
      },
      "source": [
        "# generating one row \n",
        "train_df = train_df.sample(n =1900, random_state=1)\n",
        "    \n",
        "# display\n",
        "train_df"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>filename</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>603</th>\n",
              "      <td>137covid_patient9_SR_2_IM00037.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2814</th>\n",
              "      <td>normal1_patient180_SR_2_IM00025.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3603</th>\n",
              "      <td>normal5_patient168_SR_2_IM00040.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3277</th>\n",
              "      <td>normal4_patient176_SR_2_IM00011.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2928</th>\n",
              "      <td>normal1_patient184_SR_2_IM00034.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2121</th>\n",
              "      <td>cdunnormal_patient156_SR_2_IM00015.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2801</th>\n",
              "      <td>normal1_patient180_SR_2_IM00012.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>611</th>\n",
              "      <td>157covid_patient124_SR_2_IM00024.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1788</th>\n",
              "      <td>157covid_patient140_SR_4_IM00026.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3340</th>\n",
              "      <td>normal4_patient180_SR_2_IM00012.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1900 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                    filename   class\n",
              "603       137covid_patient9_SR_2_IM00037.tif   covid\n",
              "2814     normal1_patient180_SR_2_IM00025.tif  normal\n",
              "3603     normal5_patient168_SR_2_IM00040.tif  normal\n",
              "3277     normal4_patient176_SR_2_IM00011.tif  normal\n",
              "2928     normal1_patient184_SR_2_IM00034.tif  normal\n",
              "...                                      ...     ...\n",
              "2121  cdunnormal_patient156_SR_2_IM00015.tif  normal\n",
              "2801     normal1_patient180_SR_2_IM00012.tif  normal\n",
              "611     157covid_patient124_SR_2_IM00024.tif   covid\n",
              "1788    157covid_patient140_SR_4_IM00026.tif   covid\n",
              "3340     normal4_patient180_SR_2_IM00012.tif  normal\n",
              "\n",
              "[1900 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "se_JIsa-EvkC",
        "outputId": "42d50549-686a-4b4e-ec92-b8d91dce0e29"
      },
      "source": [
        "train_df.groupby('class').count()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>filename</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>class</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>covid</th>\n",
              "      <td>903</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>normal</th>\n",
              "      <td>997</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        filename\n",
              "class           \n",
              "covid        903\n",
              "normal       997"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "SOkI0Ih0Exaa",
        "outputId": "2a91b917-138a-4aa6-db57-c5be1d978879"
      },
      "source": [
        "# generating one row\n",
        "validation_df = validation_df.sample(n = 475, random_state = 1)\n",
        "    \n",
        "# display\n",
        "validation_df"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>filename</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>191</th>\n",
              "      <td>137covid_patient105_SR_3_IM00020.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>832</th>\n",
              "      <td>normal2_patient314_SR_2_IM00033.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>111</th>\n",
              "      <td>137covid_patient120_SR_2_IM00022.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>262</th>\n",
              "      <td>137covid_patient120_SR_3_IM00025.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>670</th>\n",
              "      <td>normal5_patient226_SR_2_IM00024.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>865</th>\n",
              "      <td>normalcd1_patient412_SR_2_IM00032.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>304</th>\n",
              "      <td>137covid_patient18_SR_3_IM00013.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>200</th>\n",
              "      <td>137covid_patient10_SR_3_IM00016.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394</th>\n",
              "      <td>137covid_patient118_SR_4_IM00010.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>869</th>\n",
              "      <td>normal3_patient352_SR_2_IM00017.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>475 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                  filename   class\n",
              "191   137covid_patient105_SR_3_IM00020.tif   covid\n",
              "832    normal2_patient314_SR_2_IM00033.tif  normal\n",
              "111   137covid_patient120_SR_2_IM00022.tif   covid\n",
              "262   137covid_patient120_SR_3_IM00025.tif   covid\n",
              "670    normal5_patient226_SR_2_IM00024.tif  normal\n",
              "..                                     ...     ...\n",
              "865  normalcd1_patient412_SR_2_IM00032.tif  normal\n",
              "304    137covid_patient18_SR_3_IM00013.tif   covid\n",
              "200    137covid_patient10_SR_3_IM00016.tif   covid\n",
              "394   137covid_patient118_SR_4_IM00010.tif   covid\n",
              "869    normal3_patient352_SR_2_IM00017.tif  normal\n",
              "\n",
              "[475 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "1SsY9hbaE16B",
        "outputId": "f4ad0ceb-f13a-4878-96ba-e9178fbcf8ac"
      },
      "source": [
        "validation_df.groupby('class').count()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>filename</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>class</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>covid</th>\n",
              "      <td>244</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>normal</th>\n",
              "      <td>231</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        filename\n",
              "class           \n",
              "covid        244\n",
              "normal       231"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7nSJKiX3E4uE"
      },
      "source": [
        "test_df = test_df.drop(test_df[test_df['class']=='normal'].sample(frac =.94).index)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "lPIBN-g6E78O",
        "outputId": "b38e3f8a-f470-4be2-e593-67e67a8518f5"
      },
      "source": [
        "# generating one row\n",
        "test_df = test_df.sample(n =200, random_state=1)\n",
        "    \n",
        "# display\n",
        "test_df"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>filename</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>363</th>\n",
              "      <td>137covid_patient115_SR_4_IM00004.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>137covid_patient10_SR_2_IM00024.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6944</th>\n",
              "      <td>normal5_patient387_SR_2_IM00012.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1612</th>\n",
              "      <td>normal1_patient234_SR_2_IM00022.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>137covid_patient10_SR_2_IM00015.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5315</th>\n",
              "      <td>normal4_patient214_SR_2_IM00011.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>354</th>\n",
              "      <td>137covid_patient105_SR_4_IM00121.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6436</th>\n",
              "      <td>normal4_patient374_SR_2_IM00024.tif</td>\n",
              "      <td>normal</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>241</th>\n",
              "      <td>137covid_patient118_SR_3_IM00014.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>135</th>\n",
              "      <td>137covid_patient15_SR_2_IM00010.tif</td>\n",
              "      <td>covid</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>200 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                  filename   class\n",
              "363   137covid_patient115_SR_4_IM00004.tif   covid\n",
              "41     137covid_patient10_SR_2_IM00024.tif   covid\n",
              "6944   normal5_patient387_SR_2_IM00012.tif  normal\n",
              "1612   normal1_patient234_SR_2_IM00022.tif  normal\n",
              "35     137covid_patient10_SR_2_IM00015.tif   covid\n",
              "...                                    ...     ...\n",
              "5315   normal4_patient214_SR_2_IM00011.tif  normal\n",
              "354   137covid_patient105_SR_4_IM00121.tif   covid\n",
              "6436   normal4_patient374_SR_2_IM00024.tif  normal\n",
              "241   137covid_patient118_SR_3_IM00014.tif   covid\n",
              "135    137covid_patient15_SR_2_IM00010.tif   covid\n",
              "\n",
              "[200 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "HhUumZznE-I8",
        "outputId": "9947309d-6370-45fd-f4f9-cdcb8e43b473"
      },
      "source": [
        "test_df.groupby('class').count()"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>filename</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>class</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>covid</th>\n",
              "      <td>93</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>normal</th>\n",
              "      <td>107</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        filename\n",
              "class           \n",
              "covid         93\n",
              "normal       107"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_76ZLLbE1u7v",
        "outputId": "e0aa180b-84c3-4d17-abbd-8ebe23a36006"
      },
      "source": [
        "cd /content/drive/MyDrive/CLAHEonly/Test/"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/drive/MyDrive/CLAHEonly/Test\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b5NLWRAxFAvX"
      },
      "source": [
        "shape=(224,224,3) #shape of the dataset images (in TIFF format)"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ymfzPHtUFCWF",
        "outputId": "47eb097b-c172-43d0-a5fe-84036d87cb4b"
      },
      "source": [
        "#Create the generators\n",
        "train_generator = train_datagen.flow_from_dataframe(\n",
        "      dataframe=train_df,\n",
        "      directory='/content/drive/MyDrive/CLAHEonly/Test/',\n",
        "      x_col=\"filename\",\n",
        "      y_col=\"class\",\n",
        "      target_size=shape[:2],\n",
        "      batch_size=32,\n",
        "      class_mode='categorical',color_mode=\"rgb\",shuffle=True)\n",
        "validation_generator = test_datagen.flow_from_dataframe(\n",
        "        dataframe=validation_df,\n",
        "        directory='/content/drive/MyDrive/CLAHEonly/Test/',\n",
        "        x_col=\"filename\",\n",
        "        y_col=\"class\",\n",
        "        target_size=shape[:2],\n",
        "        batch_size=32,\n",
        "        class_mode='categorical',color_mode=\"rgb\",shuffle=True)\n",
        "test_generator = test_datagen.flow_from_dataframe(\n",
        "        dataframe=test_df,\n",
        "        directory='/content/drive/MyDrive/CLAHEonly/Test/',\n",
        "        x_col=\"filename\",\n",
        "        y_col=\"class\",\n",
        "        target_size=shape[:2],\n",
        "        batch_size=32,\n",
        "        class_mode='categorical',color_mode=\"rgb\",shuffle=True)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 1900 validated image filenames belonging to 2 classes.\n",
            "Found 475 validated image filenames belonging to 2 classes.\n",
            "Found 200 validated image filenames belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2GvuftHbFEsi"
      },
      "source": [
        "data = []\n",
        "for num,img_name in enumerate(train_generator.filenames): #load image\n",
        "    # gt_ind=test_generator.classes[num] #get the loaded image class index\n",
        "    img=cv2.imread(os.path.join('/content/drive/MyDrive/CLAHEonly/Test/',img_name),-1)\n",
        "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
        "    img = np.array(img)\n",
        "    data.append(img)"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "id": "qdOIF5mqFP40",
        "outputId": "c85a1f43-837c-48fa-b9a5-6b29d5533426"
      },
      "source": [
        "plt.imshow(data[5])"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f60598e7250>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AIBBGMruFQYm",
        "outputId": "8ec57763-cefe-4b7c-efd0-dfb3345ed291"
      },
      "source": [
        "data = np.array(data)\n",
        "data.shape"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1900, 224, 224, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qNDlALzdOzPw"
      },
      "source": [
        "data2 = []\n",
        "for num,img_name in enumerate(test_generator.filenames): #load image\n",
        "    # gt_ind=test_generator.classes[num] #get the loaded image class index\n",
        "    img=cv2.imread(os.path.join('/content/drive/MyDrive/CLAHEonly/Test/',img_name),-1)\n",
        "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
        "    img = np.array(img)\n",
        "    data2.append(img)"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rTsIGjBrO_jw",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "outputId": "64a63687-9ea1-4fce-c4aa-302b19035b2b"
      },
      "source": [
        "plt.imshow(data2[1])"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f60593b8990>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "whw9EpLV_LJL",
        "outputId": "7a4e6eb1-14cb-4613-bc8c-5eb88cfa6bbc"
      },
      "source": [
        "data2 = np.array(data2)\n",
        "data2.shape"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(200, 224, 224, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kfs8f24cOp23"
      },
      "source": [
        "valX, valY = np.array(data2), np.array(validation_generator.classes)\n",
        "trainX, trainY = np.array(data), np.array(train_generator.classes)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZfrJXOeQFTzl",
        "outputId": "a96afb0c-80a7-4efb-caf2-f028d8a253f3"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "# perform a training and testing split, using 75% of the data for\n",
        "# training and 25% for evaluation\n",
        "(trainX, testX, trainY, testY) = train_test_split(np.array(data),\tnp.array(train_generator.classes), test_size=0.25)\n",
        "#trainY = np.asarray(trainY).astype('float32').reshape((-1,1))\n",
        "#testY = np.asarray(testY).astype('float32').reshape((-1,1))\n",
        "print(trainX.shape)\n",
        "print(testX.shape)\n",
        "print(trainY.shape)\n",
        "print(testY.shape)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(1425, 224, 224, 3)\n",
            "(475, 224, 224, 3)\n",
            "(1425,)\n",
            "(475,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZoKN0ZRlX3oS"
      },
      "source": [
        "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "from keras.models import Model\n",
        "from keras.layers import GlobalAveragePooling2D, BatchNormalization, Dropout\n",
        "\n",
        "from keras.applications.efficientnet import EfficientNetB0\n",
        "from keras.layers import Dense, Conv2D, MaxPool2D , Flatten\n",
        "from keras.applications.resnet import ResNet50\n",
        "import tensorflow as tf"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zqxHVv9MX5fh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "060abcf1-6dcc-47af-cbb4-4489ceabac21"
      },
      "source": [
        "base_model = EfficientNetB0(weights='imagenet', include_top=False, input_shape=(224,224,3))\n",
        "x = base_model.output\n",
        "x = GlobalAveragePooling2D()(x)\n",
        "x = Dropout(0.5)(x)\n",
        "x = BatchNormalization()(x)\n",
        "x = Dense(512, activation='relu')(x)\n",
        "out = Dense(2, activation='softmax')(x)\n",
        "\n",
        "model = Model(inputs=base_model.input, outputs=out)\n",
        "model.summary()"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/keras-applications/efficientnetb0_notop.h5\n",
            "16711680/16705208 [==============================] - 0s 0us/step\n",
            "16719872/16705208 [==============================] - 0s 0us/step\n",
            "Model: \"model\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_1 (InputLayer)            [(None, 224, 224, 3) 0                                            \n",
            "__________________________________________________________________________________________________\n",
            "rescaling (Rescaling)           (None, 224, 224, 3)  0           input_1[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "normalization (Normalization)   (None, 224, 224, 3)  7           rescaling[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "stem_conv_pad (ZeroPadding2D)   (None, 225, 225, 3)  0           normalization[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "stem_conv (Conv2D)              (None, 112, 112, 32) 864         stem_conv_pad[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "stem_bn (BatchNormalization)    (None, 112, 112, 32) 128         stem_conv[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "stem_activation (Activation)    (None, 112, 112, 32) 0           stem_bn[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "block1a_dwconv (DepthwiseConv2D (None, 112, 112, 32) 288         stem_activation[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "block1a_bn (BatchNormalization) (None, 112, 112, 32) 128         block1a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block1a_activation (Activation) (None, 112, 112, 32) 0           block1a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block1a_se_squeeze (GlobalAvera (None, 32)           0           block1a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block1a_se_reshape (Reshape)    (None, 1, 1, 32)     0           block1a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block1a_se_reduce (Conv2D)      (None, 1, 1, 8)      264         block1a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block1a_se_expand (Conv2D)      (None, 1, 1, 32)     288         block1a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block1a_se_excite (Multiply)    (None, 112, 112, 32) 0           block1a_activation[0][0]         \n",
            "                                                                 block1a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block1a_project_conv (Conv2D)   (None, 112, 112, 16) 512         block1a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block1a_project_bn (BatchNormal (None, 112, 112, 16) 64          block1a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block2a_expand_conv (Conv2D)    (None, 112, 112, 96) 1536        block1a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2a_expand_bn (BatchNormali (None, 112, 112, 96) 384         block2a_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block2a_expand_activation (Acti (None, 112, 112, 96) 0           block2a_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2a_dwconv_pad (ZeroPadding (None, 113, 113, 96) 0           block2a_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block2a_dwconv (DepthwiseConv2D (None, 56, 56, 96)   864         block2a_dwconv_pad[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2a_bn (BatchNormalization) (None, 56, 56, 96)   384         block2a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block2a_activation (Activation) (None, 56, 56, 96)   0           block2a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block2a_se_squeeze (GlobalAvera (None, 96)           0           block2a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2a_se_reshape (Reshape)    (None, 1, 1, 96)     0           block2a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2a_se_reduce (Conv2D)      (None, 1, 1, 4)      388         block2a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2a_se_expand (Conv2D)      (None, 1, 1, 96)     480         block2a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2a_se_excite (Multiply)    (None, 56, 56, 96)   0           block2a_activation[0][0]         \n",
            "                                                                 block2a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2a_project_conv (Conv2D)   (None, 56, 56, 24)   2304        block2a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2a_project_bn (BatchNormal (None, 56, 56, 24)   96          block2a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block2b_expand_conv (Conv2D)    (None, 56, 56, 144)  3456        block2a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2b_expand_bn (BatchNormali (None, 56, 56, 144)  576         block2b_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block2b_expand_activation (Acti (None, 56, 56, 144)  0           block2b_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2b_dwconv (DepthwiseConv2D (None, 56, 56, 144)  1296        block2b_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block2b_bn (BatchNormalization) (None, 56, 56, 144)  576         block2b_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block2b_activation (Activation) (None, 56, 56, 144)  0           block2b_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block2b_se_squeeze (GlobalAvera (None, 144)          0           block2b_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2b_se_reshape (Reshape)    (None, 1, 1, 144)    0           block2b_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2b_se_reduce (Conv2D)      (None, 1, 1, 6)      870         block2b_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2b_se_expand (Conv2D)      (None, 1, 1, 144)    1008        block2b_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2b_se_excite (Multiply)    (None, 56, 56, 144)  0           block2b_activation[0][0]         \n",
            "                                                                 block2b_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2b_project_conv (Conv2D)   (None, 56, 56, 24)   3456        block2b_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block2b_project_bn (BatchNormal (None, 56, 56, 24)   96          block2b_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block2b_drop (Dropout)          (None, 56, 56, 24)   0           block2b_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block2b_add (Add)               (None, 56, 56, 24)   0           block2b_drop[0][0]               \n",
            "                                                                 block2a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3a_expand_conv (Conv2D)    (None, 56, 56, 144)  3456        block2b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block3a_expand_bn (BatchNormali (None, 56, 56, 144)  576         block3a_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block3a_expand_activation (Acti (None, 56, 56, 144)  0           block3a_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3a_dwconv_pad (ZeroPadding (None, 59, 59, 144)  0           block3a_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block3a_dwconv (DepthwiseConv2D (None, 28, 28, 144)  3600        block3a_dwconv_pad[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3a_bn (BatchNormalization) (None, 28, 28, 144)  576         block3a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block3a_activation (Activation) (None, 28, 28, 144)  0           block3a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block3a_se_squeeze (GlobalAvera (None, 144)          0           block3a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3a_se_reshape (Reshape)    (None, 1, 1, 144)    0           block3a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3a_se_reduce (Conv2D)      (None, 1, 1, 6)      870         block3a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3a_se_expand (Conv2D)      (None, 1, 1, 144)    1008        block3a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3a_se_excite (Multiply)    (None, 28, 28, 144)  0           block3a_activation[0][0]         \n",
            "                                                                 block3a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3a_project_conv (Conv2D)   (None, 28, 28, 40)   5760        block3a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3a_project_bn (BatchNormal (None, 28, 28, 40)   160         block3a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block3b_expand_conv (Conv2D)    (None, 28, 28, 240)  9600        block3a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3b_expand_bn (BatchNormali (None, 28, 28, 240)  960         block3b_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block3b_expand_activation (Acti (None, 28, 28, 240)  0           block3b_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3b_dwconv (DepthwiseConv2D (None, 28, 28, 240)  6000        block3b_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block3b_bn (BatchNormalization) (None, 28, 28, 240)  960         block3b_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block3b_activation (Activation) (None, 28, 28, 240)  0           block3b_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block3b_se_squeeze (GlobalAvera (None, 240)          0           block3b_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3b_se_reshape (Reshape)    (None, 1, 1, 240)    0           block3b_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3b_se_reduce (Conv2D)      (None, 1, 1, 10)     2410        block3b_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3b_se_expand (Conv2D)      (None, 1, 1, 240)    2640        block3b_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3b_se_excite (Multiply)    (None, 28, 28, 240)  0           block3b_activation[0][0]         \n",
            "                                                                 block3b_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3b_project_conv (Conv2D)   (None, 28, 28, 40)   9600        block3b_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block3b_project_bn (BatchNormal (None, 28, 28, 40)   160         block3b_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block3b_drop (Dropout)          (None, 28, 28, 40)   0           block3b_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block3b_add (Add)               (None, 28, 28, 40)   0           block3b_drop[0][0]               \n",
            "                                                                 block3a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4a_expand_conv (Conv2D)    (None, 28, 28, 240)  9600        block3b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block4a_expand_bn (BatchNormali (None, 28, 28, 240)  960         block4a_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block4a_expand_activation (Acti (None, 28, 28, 240)  0           block4a_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4a_dwconv_pad (ZeroPadding (None, 29, 29, 240)  0           block4a_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block4a_dwconv (DepthwiseConv2D (None, 14, 14, 240)  2160        block4a_dwconv_pad[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4a_bn (BatchNormalization) (None, 14, 14, 240)  960         block4a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block4a_activation (Activation) (None, 14, 14, 240)  0           block4a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block4a_se_squeeze (GlobalAvera (None, 240)          0           block4a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4a_se_reshape (Reshape)    (None, 1, 1, 240)    0           block4a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4a_se_reduce (Conv2D)      (None, 1, 1, 10)     2410        block4a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4a_se_expand (Conv2D)      (None, 1, 1, 240)    2640        block4a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4a_se_excite (Multiply)    (None, 14, 14, 240)  0           block4a_activation[0][0]         \n",
            "                                                                 block4a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4a_project_conv (Conv2D)   (None, 14, 14, 80)   19200       block4a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4a_project_bn (BatchNormal (None, 14, 14, 80)   320         block4a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block4b_expand_conv (Conv2D)    (None, 14, 14, 480)  38400       block4a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4b_expand_bn (BatchNormali (None, 14, 14, 480)  1920        block4b_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block4b_expand_activation (Acti (None, 14, 14, 480)  0           block4b_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4b_dwconv (DepthwiseConv2D (None, 14, 14, 480)  4320        block4b_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block4b_bn (BatchNormalization) (None, 14, 14, 480)  1920        block4b_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block4b_activation (Activation) (None, 14, 14, 480)  0           block4b_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block4b_se_squeeze (GlobalAvera (None, 480)          0           block4b_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4b_se_reshape (Reshape)    (None, 1, 1, 480)    0           block4b_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4b_se_reduce (Conv2D)      (None, 1, 1, 20)     9620        block4b_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4b_se_expand (Conv2D)      (None, 1, 1, 480)    10080       block4b_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4b_se_excite (Multiply)    (None, 14, 14, 480)  0           block4b_activation[0][0]         \n",
            "                                                                 block4b_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4b_project_conv (Conv2D)   (None, 14, 14, 80)   38400       block4b_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4b_project_bn (BatchNormal (None, 14, 14, 80)   320         block4b_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block4b_drop (Dropout)          (None, 14, 14, 80)   0           block4b_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4b_add (Add)               (None, 14, 14, 80)   0           block4b_drop[0][0]               \n",
            "                                                                 block4a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4c_expand_conv (Conv2D)    (None, 14, 14, 480)  38400       block4b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block4c_expand_bn (BatchNormali (None, 14, 14, 480)  1920        block4c_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block4c_expand_activation (Acti (None, 14, 14, 480)  0           block4c_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4c_dwconv (DepthwiseConv2D (None, 14, 14, 480)  4320        block4c_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block4c_bn (BatchNormalization) (None, 14, 14, 480)  1920        block4c_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block4c_activation (Activation) (None, 14, 14, 480)  0           block4c_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block4c_se_squeeze (GlobalAvera (None, 480)          0           block4c_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4c_se_reshape (Reshape)    (None, 1, 1, 480)    0           block4c_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4c_se_reduce (Conv2D)      (None, 1, 1, 20)     9620        block4c_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4c_se_expand (Conv2D)      (None, 1, 1, 480)    10080       block4c_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4c_se_excite (Multiply)    (None, 14, 14, 480)  0           block4c_activation[0][0]         \n",
            "                                                                 block4c_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4c_project_conv (Conv2D)   (None, 14, 14, 80)   38400       block4c_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block4c_project_bn (BatchNormal (None, 14, 14, 80)   320         block4c_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block4c_drop (Dropout)          (None, 14, 14, 80)   0           block4c_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block4c_add (Add)               (None, 14, 14, 80)   0           block4c_drop[0][0]               \n",
            "                                                                 block4b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block5a_expand_conv (Conv2D)    (None, 14, 14, 480)  38400       block4c_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block5a_expand_bn (BatchNormali (None, 14, 14, 480)  1920        block5a_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block5a_expand_activation (Acti (None, 14, 14, 480)  0           block5a_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5a_dwconv (DepthwiseConv2D (None, 14, 14, 480)  12000       block5a_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block5a_bn (BatchNormalization) (None, 14, 14, 480)  1920        block5a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block5a_activation (Activation) (None, 14, 14, 480)  0           block5a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block5a_se_squeeze (GlobalAvera (None, 480)          0           block5a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5a_se_reshape (Reshape)    (None, 1, 1, 480)    0           block5a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5a_se_reduce (Conv2D)      (None, 1, 1, 20)     9620        block5a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5a_se_expand (Conv2D)      (None, 1, 1, 480)    10080       block5a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5a_se_excite (Multiply)    (None, 14, 14, 480)  0           block5a_activation[0][0]         \n",
            "                                                                 block5a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5a_project_conv (Conv2D)   (None, 14, 14, 112)  53760       block5a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5a_project_bn (BatchNormal (None, 14, 14, 112)  448         block5a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block5b_expand_conv (Conv2D)    (None, 14, 14, 672)  75264       block5a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5b_expand_bn (BatchNormali (None, 14, 14, 672)  2688        block5b_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block5b_expand_activation (Acti (None, 14, 14, 672)  0           block5b_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5b_dwconv (DepthwiseConv2D (None, 14, 14, 672)  16800       block5b_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block5b_bn (BatchNormalization) (None, 14, 14, 672)  2688        block5b_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block5b_activation (Activation) (None, 14, 14, 672)  0           block5b_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block5b_se_squeeze (GlobalAvera (None, 672)          0           block5b_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5b_se_reshape (Reshape)    (None, 1, 1, 672)    0           block5b_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5b_se_reduce (Conv2D)      (None, 1, 1, 28)     18844       block5b_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5b_se_expand (Conv2D)      (None, 1, 1, 672)    19488       block5b_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5b_se_excite (Multiply)    (None, 14, 14, 672)  0           block5b_activation[0][0]         \n",
            "                                                                 block5b_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5b_project_conv (Conv2D)   (None, 14, 14, 112)  75264       block5b_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5b_project_bn (BatchNormal (None, 14, 14, 112)  448         block5b_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block5b_drop (Dropout)          (None, 14, 14, 112)  0           block5b_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5b_add (Add)               (None, 14, 14, 112)  0           block5b_drop[0][0]               \n",
            "                                                                 block5a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5c_expand_conv (Conv2D)    (None, 14, 14, 672)  75264       block5b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block5c_expand_bn (BatchNormali (None, 14, 14, 672)  2688        block5c_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block5c_expand_activation (Acti (None, 14, 14, 672)  0           block5c_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5c_dwconv (DepthwiseConv2D (None, 14, 14, 672)  16800       block5c_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block5c_bn (BatchNormalization) (None, 14, 14, 672)  2688        block5c_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block5c_activation (Activation) (None, 14, 14, 672)  0           block5c_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block5c_se_squeeze (GlobalAvera (None, 672)          0           block5c_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5c_se_reshape (Reshape)    (None, 1, 1, 672)    0           block5c_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5c_se_reduce (Conv2D)      (None, 1, 1, 28)     18844       block5c_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5c_se_expand (Conv2D)      (None, 1, 1, 672)    19488       block5c_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5c_se_excite (Multiply)    (None, 14, 14, 672)  0           block5c_activation[0][0]         \n",
            "                                                                 block5c_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5c_project_conv (Conv2D)   (None, 14, 14, 112)  75264       block5c_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block5c_project_bn (BatchNormal (None, 14, 14, 112)  448         block5c_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block5c_drop (Dropout)          (None, 14, 14, 112)  0           block5c_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block5c_add (Add)               (None, 14, 14, 112)  0           block5c_drop[0][0]               \n",
            "                                                                 block5b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block6a_expand_conv (Conv2D)    (None, 14, 14, 672)  75264       block5c_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block6a_expand_bn (BatchNormali (None, 14, 14, 672)  2688        block6a_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block6a_expand_activation (Acti (None, 14, 14, 672)  0           block6a_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6a_dwconv_pad (ZeroPadding (None, 17, 17, 672)  0           block6a_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block6a_dwconv (DepthwiseConv2D (None, 7, 7, 672)    16800       block6a_dwconv_pad[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6a_bn (BatchNormalization) (None, 7, 7, 672)    2688        block6a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block6a_activation (Activation) (None, 7, 7, 672)    0           block6a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block6a_se_squeeze (GlobalAvera (None, 672)          0           block6a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6a_se_reshape (Reshape)    (None, 1, 1, 672)    0           block6a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6a_se_reduce (Conv2D)      (None, 1, 1, 28)     18844       block6a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6a_se_expand (Conv2D)      (None, 1, 1, 672)    19488       block6a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6a_se_excite (Multiply)    (None, 7, 7, 672)    0           block6a_activation[0][0]         \n",
            "                                                                 block6a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6a_project_conv (Conv2D)   (None, 7, 7, 192)    129024      block6a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6a_project_bn (BatchNormal (None, 7, 7, 192)    768         block6a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block6b_expand_conv (Conv2D)    (None, 7, 7, 1152)   221184      block6a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6b_expand_bn (BatchNormali (None, 7, 7, 1152)   4608        block6b_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block6b_expand_activation (Acti (None, 7, 7, 1152)   0           block6b_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6b_dwconv (DepthwiseConv2D (None, 7, 7, 1152)   28800       block6b_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block6b_bn (BatchNormalization) (None, 7, 7, 1152)   4608        block6b_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block6b_activation (Activation) (None, 7, 7, 1152)   0           block6b_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block6b_se_squeeze (GlobalAvera (None, 1152)         0           block6b_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6b_se_reshape (Reshape)    (None, 1, 1, 1152)   0           block6b_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6b_se_reduce (Conv2D)      (None, 1, 1, 48)     55344       block6b_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6b_se_expand (Conv2D)      (None, 1, 1, 1152)   56448       block6b_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6b_se_excite (Multiply)    (None, 7, 7, 1152)   0           block6b_activation[0][0]         \n",
            "                                                                 block6b_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6b_project_conv (Conv2D)   (None, 7, 7, 192)    221184      block6b_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6b_project_bn (BatchNormal (None, 7, 7, 192)    768         block6b_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block6b_drop (Dropout)          (None, 7, 7, 192)    0           block6b_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6b_add (Add)               (None, 7, 7, 192)    0           block6b_drop[0][0]               \n",
            "                                                                 block6a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6c_expand_conv (Conv2D)    (None, 7, 7, 1152)   221184      block6b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block6c_expand_bn (BatchNormali (None, 7, 7, 1152)   4608        block6c_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block6c_expand_activation (Acti (None, 7, 7, 1152)   0           block6c_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6c_dwconv (DepthwiseConv2D (None, 7, 7, 1152)   28800       block6c_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block6c_bn (BatchNormalization) (None, 7, 7, 1152)   4608        block6c_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block6c_activation (Activation) (None, 7, 7, 1152)   0           block6c_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block6c_se_squeeze (GlobalAvera (None, 1152)         0           block6c_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6c_se_reshape (Reshape)    (None, 1, 1, 1152)   0           block6c_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6c_se_reduce (Conv2D)      (None, 1, 1, 48)     55344       block6c_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6c_se_expand (Conv2D)      (None, 1, 1, 1152)   56448       block6c_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6c_se_excite (Multiply)    (None, 7, 7, 1152)   0           block6c_activation[0][0]         \n",
            "                                                                 block6c_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6c_project_conv (Conv2D)   (None, 7, 7, 192)    221184      block6c_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6c_project_bn (BatchNormal (None, 7, 7, 192)    768         block6c_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block6c_drop (Dropout)          (None, 7, 7, 192)    0           block6c_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6c_add (Add)               (None, 7, 7, 192)    0           block6c_drop[0][0]               \n",
            "                                                                 block6b_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block6d_expand_conv (Conv2D)    (None, 7, 7, 1152)   221184      block6c_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block6d_expand_bn (BatchNormali (None, 7, 7, 1152)   4608        block6d_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block6d_expand_activation (Acti (None, 7, 7, 1152)   0           block6d_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6d_dwconv (DepthwiseConv2D (None, 7, 7, 1152)   28800       block6d_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block6d_bn (BatchNormalization) (None, 7, 7, 1152)   4608        block6d_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block6d_activation (Activation) (None, 7, 7, 1152)   0           block6d_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block6d_se_squeeze (GlobalAvera (None, 1152)         0           block6d_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6d_se_reshape (Reshape)    (None, 1, 1, 1152)   0           block6d_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6d_se_reduce (Conv2D)      (None, 1, 1, 48)     55344       block6d_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6d_se_expand (Conv2D)      (None, 1, 1, 1152)   56448       block6d_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6d_se_excite (Multiply)    (None, 7, 7, 1152)   0           block6d_activation[0][0]         \n",
            "                                                                 block6d_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6d_project_conv (Conv2D)   (None, 7, 7, 192)    221184      block6d_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block6d_project_bn (BatchNormal (None, 7, 7, 192)    768         block6d_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "block6d_drop (Dropout)          (None, 7, 7, 192)    0           block6d_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block6d_add (Add)               (None, 7, 7, 192)    0           block6d_drop[0][0]               \n",
            "                                                                 block6c_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block7a_expand_conv (Conv2D)    (None, 7, 7, 1152)   221184      block6d_add[0][0]                \n",
            "__________________________________________________________________________________________________\n",
            "block7a_expand_bn (BatchNormali (None, 7, 7, 1152)   4608        block7a_expand_conv[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "block7a_expand_activation (Acti (None, 7, 7, 1152)   0           block7a_expand_bn[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block7a_dwconv (DepthwiseConv2D (None, 7, 7, 1152)   10368       block7a_expand_activation[0][0]  \n",
            "__________________________________________________________________________________________________\n",
            "block7a_bn (BatchNormalization) (None, 7, 7, 1152)   4608        block7a_dwconv[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "block7a_activation (Activation) (None, 7, 7, 1152)   0           block7a_bn[0][0]                 \n",
            "__________________________________________________________________________________________________\n",
            "block7a_se_squeeze (GlobalAvera (None, 1152)         0           block7a_activation[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block7a_se_reshape (Reshape)    (None, 1, 1, 1152)   0           block7a_se_squeeze[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block7a_se_reduce (Conv2D)      (None, 1, 1, 48)     55344       block7a_se_reshape[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "block7a_se_expand (Conv2D)      (None, 1, 1, 1152)   56448       block7a_se_reduce[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block7a_se_excite (Multiply)    (None, 7, 7, 1152)   0           block7a_activation[0][0]         \n",
            "                                                                 block7a_se_expand[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block7a_project_conv (Conv2D)   (None, 7, 7, 320)    368640      block7a_se_excite[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "block7a_project_bn (BatchNormal (None, 7, 7, 320)    1280        block7a_project_conv[0][0]       \n",
            "__________________________________________________________________________________________________\n",
            "top_conv (Conv2D)               (None, 7, 7, 1280)   409600      block7a_project_bn[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "top_bn (BatchNormalization)     (None, 7, 7, 1280)   5120        top_conv[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "top_activation (Activation)     (None, 7, 7, 1280)   0           top_bn[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "global_average_pooling2d (Globa (None, 1280)         0           top_activation[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "dropout (Dropout)               (None, 1280)         0           global_average_pooling2d[0][0]   \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization (BatchNorma (None, 1280)         5120        dropout[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "dense (Dense)                   (None, 512)          655872      batch_normalization[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "dense_1 (Dense)                 (None, 2)            1026        dense[0][0]                      \n",
            "==================================================================================================\n",
            "Total params: 4,711,589\n",
            "Trainable params: 4,667,006\n",
            "Non-trainable params: 44,583\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "30cRQcbIFhED"
      },
      "source": [
        "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "es = EarlyStopping(monitor='val_loss',\n",
        "                   patience=5,\n",
        "                   mode='auto',\n",
        "                   restore_best_weights=True)\n",
        "checkpoint = ModelCheckpoint('best_weight_EfficientNetB0.h5',\n",
        "                             monitor='val_accuracy',\n",
        "                             verbose=1,\n",
        "                             save_best_only=True,\n",
        "                             mode='auto')"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rjlFlCpfZoYW"
      },
      "source": [
        "\n",
        "from keras.models import Sequential\n",
        "from keras.layers.core import Flatten\n",
        "from keras.layers.core import Dense\n",
        "\n",
        "model = Sequential(\n",
        "    [\n",
        "        Flatten(input_shape=(224*224*3,)),\n",
        "        Dense(100, activation=\"relu\", name=\"layer1\"),\n",
        "        Dense(16, activation=\"relu\", name=\"layer2\"),\n",
        "        Dense(16, activation=\"relu\", name=\"layer3\"),\n",
        "        Dense(3, activation = \"softmax\", name=\"layer4\"),\n",
        "    ]\n",
        ")"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lbVk0DuzFlVT",
        "outputId": "4e43e237-d90b-441a-ba3e-dd2a82a3762e"
      },
      "source": [
        "from keras.optimizers import Adam, SGD, RMSprop, Adamax\n",
        "\n",
        "# train the model using the Adamax optimizer\n",
        "print(\"[INFO] training network...\")\n",
        "opt = Adamax(learning_rate=0.01,decay=1e-6)\n",
        "model.compile(loss=\"sparse_categorical_crossentropy\", optimizer=opt, metrics=[\"accuracy\"])\n",
        "H = model.fit(trainX, trainY, validation_data=(testX, testY), epochs=50, batch_size=32, callbacks=[es,checkpoint])"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[INFO] training network...\n",
            "Epoch 1/50\n",
            "WARNING:tensorflow:Model was constructed with shape (None, 150528) for input KerasTensor(type_spec=TensorSpec(shape=(None, 150528), dtype=tf.float32, name='flatten_input'), name='flatten_input', description=\"created by layer 'flatten_input'\"), but it was called on an input with incompatible shape (None, 224, 224, 3).\n",
            "WARNING:tensorflow:Model was constructed with shape (None, 150528) for input KerasTensor(type_spec=TensorSpec(shape=(None, 150528), dtype=tf.float32, name='flatten_input'), name='flatten_input', description=\"created by layer 'flatten_input'\"), but it was called on an input with incompatible shape (None, 224, 224, 3).\n",
            "43/45 [===========================>..] - ETA: 0s - loss: 32.4586 - accuracy: 0.4766WARNING:tensorflow:Model was constructed with shape (None, 150528) for input KerasTensor(type_spec=TensorSpec(shape=(None, 150528), dtype=tf.float32, name='flatten_input'), name='flatten_input', description=\"created by layer 'flatten_input'\"), but it was called on an input with incompatible shape (None, 224, 224, 3).\n",
            "45/45 [==============================] - 18s 38ms/step - loss: 31.4910 - accuracy: 0.4780 - val_loss: 2.3680 - val_accuracy: 0.5032\n",
            "\n",
            "Epoch 00001: val_accuracy improved from -inf to 0.50316, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 2/50\n",
            "45/45 [==============================] - 1s 25ms/step - loss: 1.4577 - accuracy: 0.6365 - val_loss: 0.4782 - val_accuracy: 0.7811\n",
            "\n",
            "Epoch 00002: val_accuracy improved from 0.50316 to 0.78105, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 3/50\n",
            "45/45 [==============================] - 1s 25ms/step - loss: 0.4397 - accuracy: 0.8025 - val_loss: 0.5373 - val_accuracy: 0.7305\n",
            "\n",
            "Epoch 00003: val_accuracy did not improve from 0.78105\n",
            "Epoch 4/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.6801 - accuracy: 0.6984 - val_loss: 0.3737 - val_accuracy: 0.8421\n",
            "\n",
            "Epoch 00004: val_accuracy improved from 0.78105 to 0.84211, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 5/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.5027 - accuracy: 0.7684 - val_loss: 0.3644 - val_accuracy: 0.8337\n",
            "\n",
            "Epoch 00005: val_accuracy did not improve from 0.84211\n",
            "Epoch 6/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.2764 - accuracy: 0.8778 - val_loss: 0.2728 - val_accuracy: 0.8926\n",
            "\n",
            "Epoch 00006: val_accuracy improved from 0.84211 to 0.89263, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 7/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.5805 - accuracy: 0.7783 - val_loss: 0.3536 - val_accuracy: 0.8358\n",
            "\n",
            "Epoch 00007: val_accuracy did not improve from 0.89263\n",
            "Epoch 8/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.2610 - accuracy: 0.8888 - val_loss: 0.2198 - val_accuracy: 0.9221\n",
            "\n",
            "Epoch 00008: val_accuracy improved from 0.89263 to 0.92211, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 9/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.1802 - accuracy: 0.9219 - val_loss: 0.2553 - val_accuracy: 0.8926\n",
            "\n",
            "Epoch 00009: val_accuracy did not improve from 0.92211\n",
            "Epoch 10/50\n",
            "45/45 [==============================] - 1s 26ms/step - loss: 0.1385 - accuracy: 0.9499 - val_loss: 0.1454 - val_accuracy: 0.9516\n",
            "\n",
            "Epoch 00010: val_accuracy improved from 0.92211 to 0.95158, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 11/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.1132 - accuracy: 0.9663 - val_loss: 0.1244 - val_accuracy: 0.9600\n",
            "\n",
            "Epoch 00011: val_accuracy improved from 0.95158 to 0.96000, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 12/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0744 - accuracy: 0.9870 - val_loss: 0.1113 - val_accuracy: 0.9642\n",
            "\n",
            "Epoch 00012: val_accuracy improved from 0.96000 to 0.96421, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 13/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.0644 - accuracy: 0.9885 - val_loss: 0.1011 - val_accuracy: 0.9726\n",
            "\n",
            "Epoch 00013: val_accuracy improved from 0.96421 to 0.97263, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 14/50\n",
            "45/45 [==============================] - 1s 25ms/step - loss: 0.0550 - accuracy: 0.9922 - val_loss: 0.1131 - val_accuracy: 0.9600\n",
            "\n",
            "Epoch 00014: val_accuracy did not improve from 0.97263\n",
            "Epoch 15/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0494 - accuracy: 0.9901 - val_loss: 0.0838 - val_accuracy: 0.9747\n",
            "\n",
            "Epoch 00015: val_accuracy improved from 0.97263 to 0.97474, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 16/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0374 - accuracy: 0.9964 - val_loss: 0.0777 - val_accuracy: 0.9768\n",
            "\n",
            "Epoch 00016: val_accuracy improved from 0.97474 to 0.97684, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 17/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0377 - accuracy: 0.9944 - val_loss: 0.1357 - val_accuracy: 0.9305\n",
            "\n",
            "Epoch 00017: val_accuracy did not improve from 0.97684\n",
            "Epoch 18/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0442 - accuracy: 0.9973 - val_loss: 0.0610 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00018: val_accuracy improved from 0.97684 to 0.98105, saving model to best_weight_EfficientNetB0.h5\n",
            "Epoch 19/50\n",
            "45/45 [==============================] - 1s 27ms/step - loss: 0.0239 - accuracy: 0.9988 - val_loss: 0.0768 - val_accuracy: 0.9747\n",
            "\n",
            "Epoch 00019: val_accuracy did not improve from 0.98105\n",
            "Epoch 20/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.0283 - accuracy: 0.9990 - val_loss: 0.1144 - val_accuracy: 0.9600\n",
            "\n",
            "Epoch 00020: val_accuracy did not improve from 0.98105\n",
            "Epoch 21/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0320 - accuracy: 0.9946 - val_loss: 0.0608 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00021: val_accuracy did not improve from 0.98105\n",
            "Epoch 22/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0203 - accuracy: 0.9980 - val_loss: 0.0537 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00022: val_accuracy did not improve from 0.98105\n",
            "Epoch 23/50\n",
            "45/45 [==============================] - 1s 22ms/step - loss: 0.0119 - accuracy: 1.0000 - val_loss: 0.0692 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00023: val_accuracy did not improve from 0.98105\n",
            "Epoch 24/50\n",
            "45/45 [==============================] - 1s 22ms/step - loss: 0.0176 - accuracy: 0.9991 - val_loss: 0.0510 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00024: val_accuracy did not improve from 0.98105\n",
            "Epoch 25/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0098 - accuracy: 1.0000 - val_loss: 0.0484 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00025: val_accuracy did not improve from 0.98105\n",
            "Epoch 26/50\n",
            "45/45 [==============================] - 1s 26ms/step - loss: 0.0096 - accuracy: 1.0000 - val_loss: 0.0503 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00026: val_accuracy did not improve from 0.98105\n",
            "Epoch 27/50\n",
            "45/45 [==============================] - 1s 26ms/step - loss: 0.0076 - accuracy: 1.0000 - val_loss: 0.0690 - val_accuracy: 0.9747\n",
            "\n",
            "Epoch 00027: val_accuracy did not improve from 0.98105\n",
            "Epoch 28/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.0100 - accuracy: 1.0000 - val_loss: 0.0581 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00028: val_accuracy did not improve from 0.98105\n",
            "Epoch 29/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0068 - accuracy: 1.0000 - val_loss: 0.0468 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00029: val_accuracy did not improve from 0.98105\n",
            "Epoch 30/50\n",
            "45/45 [==============================] - 1s 24ms/step - loss: 0.0058 - accuracy: 1.0000 - val_loss: 0.0508 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00030: val_accuracy did not improve from 0.98105\n",
            "Epoch 31/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0072 - accuracy: 1.0000 - val_loss: 0.0860 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00031: val_accuracy did not improve from 0.98105\n",
            "Epoch 32/50\n",
            "45/45 [==============================] - 1s 26ms/step - loss: 0.0057 - accuracy: 1.0000 - val_loss: 0.0571 - val_accuracy: 0.9789\n",
            "\n",
            "Epoch 00032: val_accuracy did not improve from 0.98105\n",
            "Epoch 33/50\n",
            "45/45 [==============================] - 1s 23ms/step - loss: 0.0046 - accuracy: 1.0000 - val_loss: 0.0508 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00033: val_accuracy did not improve from 0.98105\n",
            "Epoch 34/50\n",
            "45/45 [==============================] - 1s 22ms/step - loss: 0.0035 - accuracy: 1.0000 - val_loss: 0.0612 - val_accuracy: 0.9811\n",
            "\n",
            "Epoch 00034: val_accuracy did not improve from 0.98105\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 591
        },
        "id": "8Dke0gmfFuGW",
        "outputId": "c5852cc2-4106-4968-a8a1-97842e768589"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "print(H.history.keys())\n",
        "# summarize history for accuracy\n",
        "plt.plot(H.history['accuracy'])\n",
        "plt.plot(H.history['val_accuracy'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='upper left')\n",
        "plt.show()\n",
        "# summarize history for loss\n",
        "plt.plot(H.history['loss'])\n",
        "plt.plot(H.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'test'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vUInJd_TGDVs",
        "outputId": "db3d475e-d2da-426b-8391-1b490ca708e1"
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, classification_report\n",
        "predictions = model.predict(testX, batch_size=32)\n",
        "print(classification_report(testY,predictions.argmax(axis=1),target_names = train_generator.class_indices))"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:Model was constructed with shape (None, 150528) for input KerasTensor(type_spec=TensorSpec(shape=(None, 150528), dtype=tf.float32, name='flatten_input'), name='flatten_input', description=\"created by layer 'flatten_input'\"), but it was called on an input with incompatible shape (None, 224, 224, 3).\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       covid       0.97      0.99      0.98       219\n",
            "      normal       0.99      0.98      0.98       256\n",
            "\n",
            "    accuracy                           0.98       475\n",
            "   macro avg       0.98      0.98      0.98       475\n",
            "weighted avg       0.98      0.98      0.98       475\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AcyrX9DAGSQ5",
        "outputId": "28c9611a-0369-47d0-e3b2-703a10687409"
      },
      "source": [
        "cf_matrix = confusion_matrix(testY,predictions.argmax(axis=1))\n",
        "print(cf_matrix)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[216   3]\n",
            " [  6 250]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "wsMaGpzXGSv6",
        "outputId": "a238905b-1d88-4808-b6c2-97a21425f7f0"
      },
      "source": [
        "cm = cf_matrix\n",
        "\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt     \n",
        "ax= plt.subplot();\n",
        "sns.heatmap(cm, annot=True,fmt=\"d\",cmap='Blues')\n",
        "# labels, title and ticks\n",
        "ax.set_xlabel('Predicted label');ax.set_ylabel('True label'); \n",
        "ax.set_ylim(2.0, 0)\n",
        "ax.set_title('Confusion Matrix');\n",
        "ax.xaxis.set_ticklabels(['Covid','Normal']); \n",
        "ax.yaxis.set_ticklabels(['Covid','Normal']);"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bv9idscEaymp"
      },
      "source": [
        "\n",
        "valX, valY = np.array(data2), np.array(test_generator.classes)"
      ],
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1Ic_u9TQGN03",
        "outputId": "9d893b6d-f022-4bf6-9798-fcc6d376919b"
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, classification_report\n",
        "predictions = model.predict(valX, batch_size=32)\n",
        "print(classification_report(valY,predictions.argmax(axis=1),target_names = test_generator.class_indices))"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "       covid       0.48      0.60      0.54        93\n",
            "      normal       0.56      0.44      0.49       107\n",
            "\n",
            "    accuracy                           0.52       200\n",
            "   macro avg       0.52      0.52      0.51       200\n",
            "weighted avg       0.52      0.52      0.51       200\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jikXkGvmGVhe",
        "outputId": "a9810d44-03d9-4ba5-9db5-5a949c13a58c"
      },
      "source": [
        "cf_matrix = confusion_matrix(valY,predictions.argmax(axis=1))\n",
        "print(cf_matrix)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[56 37]\n",
            " [60 47]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "52k-9rjmGPxl",
        "outputId": "306cd71d-a959-4d1d-b957-22ddc3b0d938"
      },
      "source": [
        "cm = cf_matrix\n",
        "\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt     \n",
        "ax= plt.subplot();\n",
        "sns.heatmap(cm, annot=True,fmt=\"d\",cmap='Blues')\n",
        "# labels, title and ticks\n",
        "ax.set_xlabel('Predicted label');ax.set_ylabel('True label'); \n",
        "ax.set_ylim(2.0, 0)\n",
        "ax.set_title('Confusion Matrix');\n",
        "ax.xaxis.set_ticklabels(['Covid','Normal']); \n",
        "ax.yaxis.set_ticklabels(['Covid','Normal']);"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c92oDRvfGw1L",
        "outputId": "7349596f-3911-4b24-fc99-6b2651f88c63"
      },
      "source": [
        "print(\"Evaluate on test data\")\n",
        "results = model.evaluate(valX, valY, batch_size=32)\n",
        "print(\"test loss, test acc:\", results)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Evaluate on test data\n",
            "7/7 [==============================] - 0s 11ms/step - loss: 1.9482 - accuracy: 0.5150\n",
            "test loss, test acc: [1.9481674432754517, 0.5149999856948853]\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}