Covid_19_detection.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/muntakim1/data-science/blob/main/Covid_19_detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HnVkNexLyl6a",
        "outputId": "f89f32eb-b802-42ba-fc3c-e4e27906287c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "nvcc: NVIDIA (R) Cuda compiler driver\n",
            "Copyright (c) 2005-2020 NVIDIA Corporation\n",
            "Built on Mon_Oct_12_20:09:46_PDT_2020\n",
            "Cuda compilation tools, release 11.1, V11.1.105\n",
            "Build cuda_11.1.TC455_06.29190527_0\n"
          ]
        }
      ],
      "source": [
        "!nvcc --version"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hhEBawRUclQW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "6893decf-0904-4314-bf83-3d0421b2d32b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[33mWARNING: Skipping tensorflow-gpu as it is not installed.\u001b[0m\n",
            "Found existing installation: tensorflow 2.8.0\n",
            "Uninstalling tensorflow-2.8.0:\n",
            "  Successfully uninstalled tensorflow-2.8.0\n",
            "Collecting livelossplot\n",
            "  Downloading livelossplot-0.5.5-py3-none-any.whl (22 kB)\n",
            "Collecting ipython==7.*\n",
            "  Downloading ipython-7.33.0-py3-none-any.whl (793 kB)\n",
            "\u001b[K     |████████████████████████████████| 793 kB 13.3 MB/s \n",
            "\u001b[?25hRequirement already satisfied: matplotlib in /usr/local/lib/python3.7/dist-packages (from livelossplot) (3.2.2)\n",
            "Requirement already satisfied: bokeh in /usr/local/lib/python3.7/dist-packages (from livelossplot) (2.3.3)\n",
            "Requirement already satisfied: numpy<1.22 in /usr/local/lib/python3.7/dist-packages (from livelossplot) (1.21.6)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (4.4.2)\n",
            "Requirement already satisfied: setuptools>=18.5 in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (57.4.0)\n",
            "Requirement already satisfied: jedi>=0.16 in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (0.18.1)\n",
            "Requirement already satisfied: pexpect>4.3 in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (4.8.0)\n",
            "Collecting prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0\n",
            "  Downloading prompt_toolkit-3.0.29-py3-none-any.whl (381 kB)\n",
            "\u001b[K     |████████████████████████████████| 381 kB 63.1 MB/s \n",
            "\u001b[?25hRequirement already satisfied: pygments in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (2.6.1)\n",
            "Requirement already satisfied: backcall in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (0.2.0)\n",
            "Requirement already satisfied: traitlets>=4.2 in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (5.1.1)\n",
            "Requirement already satisfied: matplotlib-inline in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (0.1.3)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.7/dist-packages (from ipython==7.*->livelossplot) (0.7.5)\n",
            "Requirement already satisfied: parso<0.9.0,>=0.8.0 in /usr/local/lib/python3.7/dist-packages (from jedi>=0.16->ipython==7.*->livelossplot) (0.8.3)\n",
            "Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.7/dist-packages (from pexpect>4.3->ipython==7.*->livelossplot) (0.7.0)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0->ipython==7.*->livelossplot) (0.2.5)\n",
            "Requirement already satisfied: packaging>=16.8 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (21.3)\n",
            "Requirement already satisfied: pillow>=7.1.0 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (7.1.2)\n",
            "Requirement already satisfied: PyYAML>=3.10 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (3.13)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (2.8.2)\n",
            "Requirement already satisfied: tornado>=5.1 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (5.1.1)\n",
            "Requirement already satisfied: Jinja2>=2.9 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (2.11.3)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4 in /usr/local/lib/python3.7/dist-packages (from bokeh->livelossplot) (4.2.0)\n",
            "Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.7/dist-packages (from Jinja2>=2.9->bokeh->livelossplot) (2.0.1)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=16.8->bokeh->livelossplot) (3.0.8)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.1->bokeh->livelossplot) (1.15.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->livelossplot) (0.11.0)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->livelossplot) (1.4.2)\n",
            "Installing collected packages: prompt-toolkit, ipython, livelossplot\n",
            "  Attempting uninstall: prompt-toolkit\n",
            "    Found existing installation: prompt-toolkit 1.0.18\n",
            "    Uninstalling prompt-toolkit-1.0.18:\n",
            "      Successfully uninstalled prompt-toolkit-1.0.18\n",
            "  Attempting uninstall: ipython\n",
            "    Found existing installation: ipython 5.5.0\n",
            "    Uninstalling ipython-5.5.0:\n",
            "      Successfully uninstalled ipython-5.5.0\n",
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "jupyter-console 5.2.0 requires prompt-toolkit<2.0.0,>=1.0.0, but you have prompt-toolkit 3.0.29 which is incompatible.\n",
            "google-colab 1.0.0 requires ipython~=5.5.0, but you have ipython 7.33.0 which is incompatible.\u001b[0m\n",
            "Successfully installed ipython-7.33.0 livelossplot-0.5.5 prompt-toolkit-3.0.29\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "IPython",
                  "prompt_toolkit"
                ]
              }
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting tensorflow-gpu==2.4.0\n",
            "  Downloading tensorflow_gpu-2.4.0-cp37-cp37m-manylinux2010_x86_64.whl (394.7 MB)\n",
            "\u001b[K     |████████████████████████████████| 394.7 MB 19 kB/s \n",
            "\u001b[?25hRequirement already satisfied: astunparse~=1.6.3 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (1.6.3)\n",
            "Collecting grpcio~=1.32.0\n",
            "  Downloading grpcio-1.32.0-cp37-cp37m-manylinux2014_x86_64.whl (3.8 MB)\n",
            "\u001b[K     |████████████████████████████████| 3.8 MB 52.3 MB/s \n",
            "\u001b[?25hCollecting absl-py~=0.10\n",
            "  Downloading absl_py-0.15.0-py3-none-any.whl (132 kB)\n",
            "\u001b[K     |████████████████████████████████| 132 kB 73.3 MB/s \n",
            "\u001b[?25hCollecting typing-extensions~=3.7.4\n",
            "  Downloading typing_extensions-3.7.4.3-py3-none-any.whl (22 kB)\n",
            "Collecting wrapt~=1.12.1\n",
            "  Downloading wrapt-1.12.1.tar.gz (27 kB)\n",
            "Requirement already satisfied: wheel~=0.35 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (0.37.1)\n",
            "Collecting gast==0.3.3\n",
            "  Downloading gast-0.3.3-py2.py3-none-any.whl (9.7 kB)\n",
            "Requirement already satisfied: keras-preprocessing~=1.1.2 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (1.1.2)\n",
            "Requirement already satisfied: protobuf>=3.9.2 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (3.17.3)\n",
            "Collecting flatbuffers~=1.12.0\n",
            "  Downloading flatbuffers-1.12-py2.py3-none-any.whl (15 kB)\n",
            "Requirement already satisfied: opt-einsum~=3.3.0 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (3.3.0)\n",
            "Requirement already satisfied: tensorboard~=2.4 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (2.8.0)\n",
            "Requirement already satisfied: termcolor~=1.1.0 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (1.1.0)\n",
            "Collecting tensorflow-estimator<2.5.0,>=2.4.0rc0\n",
            "  Downloading tensorflow_estimator-2.4.0-py2.py3-none-any.whl (462 kB)\n",
            "\u001b[K     |████████████████████████████████| 462 kB 50.5 MB/s \n",
            "\u001b[?25hCollecting numpy~=1.19.2\n",
            "  Downloading numpy-1.19.5-cp37-cp37m-manylinux2010_x86_64.whl (14.8 MB)\n",
            "\u001b[K     |████████████████████████████████| 14.8 MB 53.9 MB/s \n",
            "\u001b[?25hCollecting h5py~=2.10.0\n",
            "  Downloading h5py-2.10.0-cp37-cp37m-manylinux1_x86_64.whl (2.9 MB)\n",
            "\u001b[K     |████████████████████████████████| 2.9 MB 57.7 MB/s \n",
            "\u001b[?25hRequirement already satisfied: google-pasta~=0.2 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (0.2.0)\n",
            "Requirement already satisfied: six~=1.15.0 in /usr/local/lib/python3.7/dist-packages (from tensorflow-gpu==2.4.0) (1.15.0)\n",
            "Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (1.0.1)\n",
            "Requirement already satisfied: setuptools>=41.0.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (57.4.0)\n",
            "Requirement already satisfied: requests<3,>=2.21.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (2.23.0)\n",
            "Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (1.8.1)\n",
            "Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (0.4.6)\n",
            "Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (0.6.1)\n",
            "Requirement already satisfied: google-auth<3,>=1.6.3 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (1.35.0)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.7/dist-packages (from tensorboard~=2.4->tensorflow-gpu==2.4.0) (3.3.6)\n",
            "Requirement already satisfied: cachetools<5.0,>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from google-auth<3,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4.0) (4.2.4)\n",
            "Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.7/dist-packages (from google-auth<3,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4.0) (0.2.8)\n",
            "Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.7/dist-packages (from google-auth<3,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4.0) (4.8)\n",
            "Requirement already satisfied: requests-oauthlib>=0.7.0 in /usr/local/lib/python3.7/dist-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard~=2.4->tensorflow-gpu==2.4.0) (1.3.1)\n",
            "Requirement already satisfied: importlib-metadata>=4.4 in /usr/local/lib/python3.7/dist-packages (from markdown>=2.6.8->tensorboard~=2.4->tensorflow-gpu==2.4.0) (4.11.3)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.7/dist-packages (from importlib-metadata>=4.4->markdown>=2.6.8->tensorboard~=2.4->tensorflow-gpu==2.4.0) (3.8.0)\n",
            "Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /usr/local/lib/python3.7/dist-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4.0) (0.4.8)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4.0) (3.0.4)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4.0) (1.24.3)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4.0) (2021.10.8)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4.0) (2.10)\n",
            "Requirement already satisfied: oauthlib>=3.0.0 in /usr/local/lib/python3.7/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard~=2.4->tensorflow-gpu==2.4.0) (3.2.0)\n",
            "Building wheels for collected packages: wrapt\n",
            "  Building wheel for wrapt (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for wrapt: filename=wrapt-1.12.1-cp37-cp37m-linux_x86_64.whl size=68718 sha256=892f8c8becad6eed51c4254793512131080a7efb079055122457743727e7eae7\n",
            "  Stored in directory: /root/.cache/pip/wheels/62/76/4c/aa25851149f3f6d9785f6c869387ad82b3fd37582fa8147ac6\n",
            "Successfully built wrapt\n",
            "Installing collected packages: typing-extensions, numpy, grpcio, absl-py, wrapt, tensorflow-estimator, h5py, gast, flatbuffers, tensorflow-gpu\n",
            "  Attempting uninstall: typing-extensions\n",
            "    Found existing installation: typing-extensions 4.2.0\n",
            "    Uninstalling typing-extensions-4.2.0:\n",
            "      Successfully uninstalled typing-extensions-4.2.0\n",
            "  Attempting uninstall: numpy\n",
            "    Found existing installation: numpy 1.21.6\n",
            "    Uninstalling numpy-1.21.6:\n",
            "      Successfully uninstalled numpy-1.21.6\n",
            "  Attempting uninstall: grpcio\n",
            "    Found existing installation: grpcio 1.44.0\n",
            "    Uninstalling grpcio-1.44.0:\n",
            "      Successfully uninstalled grpcio-1.44.0\n",
            "  Attempting uninstall: absl-py\n",
            "    Found existing installation: absl-py 1.0.0\n",
            "    Uninstalling absl-py-1.0.0:\n",
            "      Successfully uninstalled absl-py-1.0.0\n",
            "  Attempting uninstall: wrapt\n",
            "    Found existing installation: wrapt 1.14.0\n",
            "    Uninstalling wrapt-1.14.0:\n",
            "      Successfully uninstalled wrapt-1.14.0\n",
            "  Attempting uninstall: tensorflow-estimator\n",
            "    Found existing installation: tensorflow-estimator 2.8.0\n",
            "    Uninstalling tensorflow-estimator-2.8.0:\n",
            "      Successfully uninstalled tensorflow-estimator-2.8.0\n",
            "  Attempting uninstall: h5py\n",
            "    Found existing installation: h5py 3.1.0\n",
            "    Uninstalling h5py-3.1.0:\n",
            "      Successfully uninstalled h5py-3.1.0\n",
            "  Attempting uninstall: gast\n",
            "    Found existing installation: gast 0.5.3\n",
            "    Uninstalling gast-0.5.3:\n",
            "      Successfully uninstalled gast-0.5.3\n",
            "  Attempting uninstall: flatbuffers\n",
            "    Found existing installation: flatbuffers 2.0\n",
            "    Uninstalling flatbuffers-2.0:\n",
            "      Successfully uninstalled flatbuffers-2.0\n",
            "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
            "kapre 0.3.7 requires tensorflow>=2.0.0, which is not installed.\n",
            "google-colab 1.0.0 requires ipython~=5.5.0, but you have ipython 7.33.0 which is incompatible.\n",
            "datascience 0.10.6 requires folium==0.2.1, but you have folium 0.8.3 which is incompatible.\n",
            "albumentations 0.1.12 requires imgaug<0.2.7,>=0.2.5, but you have imgaug 0.2.9 which is incompatible.\u001b[0m\n",
            "Successfully installed absl-py-0.15.0 flatbuffers-1.12 gast-0.3.3 grpcio-1.32.0 h5py-2.10.0 numpy-1.19.5 tensorflow-estimator-2.4.0 tensorflow-gpu-2.4.0 typing-extensions-3.7.4.3 wrapt-1.12.1\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "numpy"
                ]
              }
            }
          },
          "metadata": {}
        }
      ],
      "source": [
        "!pip uninstall tensorflow-gpu tensorflow -y\n",
        "!pip install livelossplot\n",
        "!pip install tensorflow-gpu==2.4.0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_WukYRYGWURS"
      },
      "outputs": [],
      "source": [
        "%tensorflow_version 2.x\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "from livelossplot import PlotLossesKerasTF\n",
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IKYdyXr6fBYz",
        "outputId": "f24a71c5-4a67-4a73-9436-026349f397c7"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU'),\n",
              " PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "tf.config.list_physical_devices()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Qz_YXcfQvjYN"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "gpus = tf.config.experimental.list_physical_devices('GPU')\n",
        "if gpus:\n",
        "    try:\n",
        "        for gpu in gpus:\n",
        "            tf.config.experimental.set_memory_growth(gpu, True)\n",
        "\n",
        "    except RuntimeError as e:\n",
        "        print(e)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mzBHmRu2fZgi"
      },
      "source": [
        "Data Preprocessing\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a0LgUn7efel9"
      },
      "source": [
        "# Data Cleaning"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K6N5uw_GHxQs",
        "outputId": "80b06e15-a445-42cd-a496-b4b407385ad7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zi5oI39TfvrB"
      },
      "source": [
        "# Cleaning folders"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x28Yqb4aZbm6",
        "outputId": "46e984c8-fd6c-4484-b148-5f6c48d57819"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Train’: No such file or directory\n",
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Test’: No such file or directory\n",
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Validation’: No such file or directory\n",
            "mv: cannot stat '/content/COVID-19_Radiography_Dataset/COVID/': No such file or directory\n",
            "mv: cannot stat '/content/COVID-19_Radiography_Dataset/Normal/': No such file or directory\n"
          ]
        }
      ],
      "source": [
        "!mkdir /content/COVID-19_Radiography_Dataset/Train\n",
        "!mkdir /content/COVID-19_Radiography_Dataset/Test\n",
        "!mkdir /content/COVID-19_Radiography_Dataset/Validation\n",
        "!mv /content/COVID-19_Radiography_Dataset/COVID/ /content/COVID-19_Radiography_Dataset/Train\n",
        "!mv /content/COVID-19_Radiography_Dataset/Normal/ /content/COVID-19_Radiography_Dataset/Train"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "10us08CEdwav",
        "outputId": "b00387b4-db62-4290-b791-71f23b6cef05"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Test/Normal’: No such file or directory\n",
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Test/COVID’: No such file or directory\n",
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Validation/Normal’: No such file or directory\n",
            "mkdir: cannot create directory ‘/content/COVID-19_Radiography_Dataset/Validation/COVID’: No such file or directory\n"
          ]
        }
      ],
      "source": [
        "!mkdir /content/COVID-19_Radiography_Dataset/Test/Normal\n",
        "!mkdir /content/COVID-19_Radiography_Dataset/Test/COVID\n",
        "!mkdir /content/COVID-19_Radiography_Dataset/Validation/Normal\n",
        "!mkdir /content/COVID-19_Radiography_Dataset/Validation/COVID"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9gtSR-uRP5-7"
      },
      "outputs": [],
      "source": [
        "import shutil\n",
        "import os\n",
        "'''\n",
        "Moving 100 and 50 data Train to Test and validation folder\n",
        "'''\n",
        "\n",
        "source_dir = '/content/COVID-19_Radiography_Dataset/Train/COVID'\n",
        "target_dir = '/content/COVID-19_Radiography_Dataset/Validation/COVID'\n",
        "\n",
        "def MoveFiles(source_dir,target_dir,portion):\n",
        "  file_names = os.listdir(source_dir)\n",
        "  percent = lambda part:float(len(file_names)) / 100 * float(part)\n",
        "  for file_name in file_names[:int(percent(portion))]:\n",
        "      shutil.move(os.path.join(source_dir, file_name), target_dir)\n",
        "\n",
        "MoveFiles('/content/COVID-19_Radiography_Dataset/Train/COVID','/content/COVID-19_Radiography_Dataset/Validation/COVID',20)\n",
        "MoveFiles('/content/COVID-19_Radiography_Dataset/Train/Normal','/content/COVID-19_Radiography_Dataset/Validation/Normal',20)\n",
        "MoveFiles('/content/COVID-19_Radiography_Dataset/Train/COVID','/content/COVID-19_Radiography_Dataset/Test/COVID',10)\n",
        "MoveFiles('/content/COVID-19_Radiography_Dataset/Train/Normal','/content/COVID-19_Radiography_Dataset/Test/Normal',10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4Grt2kBNSlOQ",
        "outputId": "a6f451ad-97c0-4a11-9939-653b913c5fbf"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "rm: cannot remove '/content/COVID-19_Radiography_Dataset/Train/.ipynb_checkpoints/': No such file or directory\n",
            "rm: cannot remove '/content/COVID-19_Radiography_Dataset/Validation/.ipynb_checkpoints/': No such file or directory\n"
          ]
        }
      ],
      "source": [
        "!rm -r /content/COVID-19_Radiography_Dataset/Train/.ipynb_checkpoints/\n",
        "!rm -r /content/COVID-19_Radiography_Dataset/Validation/.ipynb_checkpoints/\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9taWe6_QftCd"
      },
      "source": [
        "# Image Data generator"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "BATCH_SIZE=32"
      ],
      "metadata": {
        "id": "m440uY4rWiF5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ySX_Zv4FW7_r",
        "outputId": "474f60ed-c8f2-4ca5-eb6b-0db3ad5512d9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 5184 images belonging to 2 classes.\n"
          ]
        }
      ],
      "source": [
        "train_datagen=tf.keras.preprocessing.image.ImageDataGenerator(rescale= 1./255,horizontal_flip=True,zoom_range=0.2,shear_range=0.2)\n",
        "train_data=train_datagen.flow_from_directory(directory=\"/content/drive/MyDrive/COVID-19_Radiography_Dataset/Train\",shuffle=True,target_size=(229,229),color_mode='grayscale',batch_size=BATCH_SIZE,class_mode='binary')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_G6lrlvTXPwc",
        "outputId": "da28fb6f-545d-4499-820f-dc8b5a3384f4"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 576 images belonging to 2 classes.\n"
          ]
        }
      ],
      "source": [
        "test_datagen=tf.keras.preprocessing.image.ImageDataGenerator(rescale=1./255)\n",
        "test_data=test_datagen.flow_from_directory(directory=\"/content/drive/MyDrive/COVID-19_Radiography_Dataset/Test\",shuffle=False,target_size=(229,229),color_mode='grayscale',batch_size=BATCH_SIZE,class_mode='binary')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QguaSbUtRe3J",
        "outputId": "adf0e528-81a2-4e39-a701-07631b3d3133"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 1440 images belonging to 2 classes.\n"
          ]
        }
      ],
      "source": [
        "val_datagen=tf.keras.preprocessing.image.ImageDataGenerator(rescale=1./255)\n",
        "val_data=val_datagen.flow_from_directory(directory=\"/content/drive/MyDrive/COVID-19_Radiography_Dataset/Validation\",shuffle=False,target_size=(229,229),color_mode='grayscale',batch_size=BATCH_SIZE,class_mode='binary')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d3VkECZvf94z"
      },
      "source": [
        "# Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WDs9NJ1DXlvY"
      },
      "outputs": [],
      "source": [
        "model = tf.keras.models.Sequential()\n",
        "model.add(tf.keras.layers.Conv2D(64, kernel_size=(3, 3),activation='relu',padding = 'Same',input_shape=(229, 229, 1)))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "##############################\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.AveragePooling2D(pool_size = (2, 2)))\n",
        "model.add(tf.keras.layers.Dropout(0.25))\n",
        "\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.AveragePooling2D(pool_size=(2, 2)))\n",
        "model.add(tf.keras.layers.Dropout(0.25))\n",
        "##############################\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.AveragePooling2D(pool_size = (2, 2)))\n",
        "model.add(tf.keras.layers.Dropout(0.25))\n",
        "\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.AveragePooling2D(pool_size=(2, 2)))\n",
        "model.add(tf.keras.layers.Dropout(0.25))\n",
        "\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.AveragePooling2D(pool_size=(2, 2)))\n",
        "model.add(tf.keras.layers.Dropout(0.25))\n",
        "\n",
        "model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='relu',padding = 'Same'))\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.AveragePooling2D(pool_size=(2, 2)))\n",
        "model.add(tf.keras.layers.Dropout(0.25))\n",
        "##############################\n",
        "model.add(tf.keras.layers.Flatten())\n",
        "\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.Dense(128, activation='relu'))\n",
        "model.add(tf.keras.layers.Activation('relu'))\n",
        "\n",
        "#Output\n",
        "model.add(tf.keras.layers.BatchNormalization())\n",
        "model.add(tf.keras.layers.Dense(1, activation='sigmoid'))\n",
        "#Compile the model\n",
        "opt = tf.keras.optimizers.Adam(lr=0.005, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.01, amsgrad=False)\n",
        "model.compile(optimizer=opt, loss='binary_crossentropy', metrics=['acc'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hImteg_NXyvr",
        "outputId": "9c7dd98e-e136-4df1-f77f-353bbd623c2c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 229, 229, 64)      640       \n",
            "_________________________________________________________________\n",
            "batch_normalization (BatchNo (None, 229, 229, 64)      256       \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 229, 229, 64)      36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_1 (Batch (None, 229, 229, 64)      256       \n",
            "_________________________________________________________________\n",
            "average_pooling2d (AveragePo (None, 114, 114, 64)      0         \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 114, 114, 64)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 114, 114, 64)      36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_2 (Batch (None, 114, 114, 64)      256       \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 114, 114, 64)      36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_3 (Batch (None, 114, 114, 64)      256       \n",
            "_________________________________________________________________\n",
            "average_pooling2d_1 (Average (None, 57, 57, 64)        0         \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 57, 57, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_4 (Conv2D)            (None, 57, 57, 64)        36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_4 (Batch (None, 57, 57, 64)        256       \n",
            "_________________________________________________________________\n",
            "average_pooling2d_2 (Average (None, 28, 28, 64)        0         \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 28, 28, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_5 (Conv2D)            (None, 28, 28, 64)        36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_5 (Batch (None, 28, 28, 64)        256       \n",
            "_________________________________________________________________\n",
            "conv2d_6 (Conv2D)            (None, 28, 28, 64)        36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_6 (Batch (None, 28, 28, 64)        256       \n",
            "_________________________________________________________________\n",
            "average_pooling2d_3 (Average (None, 14, 14, 64)        0         \n",
            "_________________________________________________________________\n",
            "dropout_3 (Dropout)          (None, 14, 14, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_7 (Conv2D)            (None, 14, 14, 64)        36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_7 (Batch (None, 14, 14, 64)        256       \n",
            "_________________________________________________________________\n",
            "average_pooling2d_4 (Average (None, 7, 7, 64)          0         \n",
            "_________________________________________________________________\n",
            "dropout_4 (Dropout)          (None, 7, 7, 64)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_8 (Conv2D)            (None, 7, 7, 64)          36928     \n",
            "_________________________________________________________________\n",
            "batch_normalization_8 (Batch (None, 7, 7, 64)          256       \n",
            "_________________________________________________________________\n",
            "average_pooling2d_5 (Average (None, 3, 3, 64)          0         \n",
            "_________________________________________________________________\n",
            "dropout_5 (Dropout)          (None, 3, 3, 64)          0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 576)               0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_9 (Batch (None, 576)               2304      \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 128)               73856     \n",
            "_________________________________________________________________\n",
            "activation (Activation)      (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_10 (Batc (None, 128)               512       \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 1)                 129       \n",
            "=================================================================\n",
            "Total params: 375,169\n",
            "Trainable params: 372,609\n",
            "Non-trainable params: 2,560\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 707
        },
        "id": "ugnUuHsuX3hB",
        "outputId": "95011e80-281d-4f5b-8fbe-4b2d2067fa1d"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x576 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy\n",
            "\ttraining         \t (min:    0.779, max:    0.920, cur:    0.916)\n",
            "\tvalidation       \t (min:    0.500, max:    0.952, cur:    0.938)\n",
            "Loss\n",
            "\ttraining         \t (min:    0.197, max:    0.475, cur:    0.200)\n",
            "\tvalidation       \t (min:    0.133, max:    4.067, cur:    0.149)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f7b931f71d0>"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "model.fit(train_data,epochs=25,\n",
        "          callbacks=[PlotLossesKerasTF()],batch_size=BATCH_SIZE,\n",
        "          validation_data=val_data)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "EHkfBqPUb10-"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import ConfusionMatrixDisplay\n",
        "from sklearn.metrics import confusion_matrix"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iBZBTgaM6w_n"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import classification_report"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "veWx2fuYJ0BO"
      },
      "outputs": [],
      "source": [
        "test_steps_per_epoch = np.math.ceil(test_data.samples / test_data.batch_size)\n",
        "\n",
        "\n",
        "predictions = model.predict_generator(test_data, steps=test_steps_per_epoch)\n",
        "# Get most likely class\n",
        "predicted_classes = np.argmax(predictions, axis=1)\n",
        "true_classes = test_data.classes\n",
        "\n",
        "class_labels = list(test_data.class_indices.keys())\n",
        "print(class_labels)\n",
        "predictions_flat = predicted_classes.flatten()\n",
        "true_classes_flat = true_classes.flatten()\n",
        "print(predictions_flat)\n",
        "# print(confusion_matrix(true_classes_flat,predictions_flat))\n",
        "\n",
        "report = classification_report(true_classes_flat, predictions_flat, target_names=class_labels)\n",
        "print(report)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "DtT6nR3uJ14t"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [
        "zi5oI39TfvrB"
      ],
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}