models/FirstExample/FirstExample.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oxabhjcraV6c",
        "colab_type": "text"
      },
      "source": [
        "## First Example\n",
        "\n",
        "Using images of viral, bacterial pneumonia and patients with COVID-19 as well as x-rays of healthy people. This example was based on the post from [Medium](https://medium.com/analytics-vidhya/cnns-architectures-lenet-alexnet-vgg-googlenet-resnet-and-more-666091488df5).\n",
        "\n",
        "It was also to understand the technique of transferring knowledge from trained networks. Shortly after the test, a performance was found to be very desirable, and soon we went on to research new sources of data on this aspect of the theme of this work.\n",
        "\n",
        "It was found content about COVID-19, COVID-NET, and explored this content later we will work in more detail approaches always related to the data set of this work found."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bbgYP3Gh5Ndk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from glob import glob\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SXP1HW485zpz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!gdown https://drive.google.com/uc?id=1coM7x3378f-Ou2l6Pg2wldaOI7Dntu1a"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yt22qu6G6CTG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!unzip Covid_Data_GradientCrescent.zip"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pZpBEJxR6EsG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_dir = 'all/train'\n",
        "test_dir = 'all/test'\n",
        "img_size = (150, 150)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ydaTD1DNVHDZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_normal = train_dir + '/normal/'\n",
        "train_pneuB   = train_dir + '/pneumonia_bac/'\n",
        "train_pneuV   = train_dir + '/pneumonia_vir/'\n",
        "train_covid  = train_dir + '/covid/'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1Dbkig8KVxXz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "bac   = glob(train_pneuB + \"*.jpeg\")\n",
        "vi   = glob(train_pneuV + \"*.jpeg\")\n",
        "covid   = glob(train_covid + \"*.jpeg\")\n",
        "normal = glob(train_normal + \"*.jpeg\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gZZ3X0EPOZTY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def show_imgs(num_of_imgs):\n",
        "    \n",
        "    for img in range(num_of_imgs):\n",
        "        pneu_pic   = np.asarray(plt.imread(vi[img]))\n",
        "        normal_pic = np.asarray(plt.imread(normal[img]))\n",
        "        bac_pic    = np.asarray(plt.imread(bac[img]))\n",
        "        covid_pic = np.asarray(plt.imread(covid[img]))\n",
        "        \n",
        "\n",
        "\n",
        "        fig = plt.figure(figsize= (15,10))\n",
        "\n",
        "        normal_plot = fig.add_subplot(221)\n",
        "        plt.imshow(normal_pic, cmap='gray')\n",
        "        normal_plot.set_title('Normal')\n",
        "        plt.axis('off')\n",
        "\n",
        "        pneu_plot = fig.add_subplot(122)\n",
        "        plt.imshow(pneu_pic, cmap='gray')\n",
        "        pneu_plot.set_title('Pneumonia Viral')\n",
        "        plt.axis('off')\n",
        "\n",
        "        pneu_plot = fig.add_subplot(223)\n",
        "        plt.imshow(bac_pic, cmap='gray')\n",
        "        pneu_plot.set_title('Pneumonia Bacteriana')\n",
        "        plt.axis('off')\n",
        "\n",
        "        pneu_plot = fig.add_subplot(342)\n",
        "        plt.imshow(covid_pic, cmap='gray')\n",
        "        pneu_plot.set_title('Pneumonia Bacteriana')\n",
        "        plt.axis('off')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "THIJ3mjgW5v5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "show_imgs(1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0ZelzW-Z7c6-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "train_datagen = ImageDataGenerator(rescale=1./255,\n",
        " rotation_range=50,\n",
        " width_shift_range=0.2,\n",
        " height_shift_range=0.2,\n",
        " shear_range=0.25,\n",
        " zoom_range=0.1,\n",
        " validation_split = 0.2,\n",
        " fill_mode='nearest')\n",
        "\n",
        "train_batches = train_datagen.flow_from_directory(train_dir,\n",
        " target_size=img_size,\n",
        " shuffle=True, \n",
        " batch_size=10,\n",
        " subset = \"training\",\n",
        " seed=42,\n",
        " class_mode=\"categorical\",\n",
        "  )\n",
        "\n",
        "valid_batches = train_datagen.flow_from_directory(train_dir,\n",
        " target_size=img_size,\n",
        " shuffle=True,\n",
        " batch_size=10,\n",
        " subset = \"validation\",\n",
        " seed=42,\n",
        " class_mode=\"categorical\",\n",
        "  )"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jyxE9_iu7k3O",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.applications.vgg16 import VGG16\n",
        "from tensorflow.keras import optimizers\n",
        "\n",
        "conv_base = VGG16(weights='imagenet', include_top=False,input_shape=(150, 150, 3))\n",
        "conv_base.trainable = False\n",
        "model = tf.keras.models.Sequential()\n",
        "model.add(conv_base)\n",
        "model.add(tf.keras.layers.Flatten())\n",
        "model.add(tf.keras.layers.Dense(256, activation='relu'))\n",
        "model.add(tf.keras.layers.Dense(4, activation='softmax'))\n",
        "model.compile(loss='categorical_crossentropy',optimizer=optimizers.Adam(lr=0.0005), metrics=['acc'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IT4yLcK071_l",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "c79fef4a-25bd-4cd8-d719-a3285d73d3f1"
      },
      "source": [
        "history = model.fit_generator(train_batches,\n",
        " steps_per_epoch = train_batches.n//train_batches.batch_size,\n",
        " validation_data = valid_batches,\n",
        " validation_steps = valid_batches.n//valid_batches.batch_size,\n",
        " epochs = 100,\n",
        " )"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 69/100\n",
            "21/21 [==============================] - 5s 248ms/step - loss: 0.2538 - acc: 0.9272 - val_loss: 0.6964 - val_acc: 0.7000\n",
            "Epoch 70/100\n",
            "21/21 [==============================] - 5s 235ms/step - loss: 0.1938 - acc: 0.9272 - val_loss: 0.8032 - val_acc: 0.7000\n",
            "Epoch 71/100\n",
            "21/21 [==============================] - 5s 239ms/step - loss: 0.2408 - acc: 0.9029 - val_loss: 0.3563 - val_acc: 0.8400\n",
            "Epoch 72/100\n",
            "21/21 [==============================] - 5s 232ms/step - loss: 0.2387 - acc: 0.9126 - val_loss: 0.7650 - val_acc: 0.7600\n",
            "Epoch 73/100\n",
            "21/21 [==============================] - 5s 236ms/step - loss: 0.2776 - acc: 0.8981 - val_loss: 0.8396 - val_acc: 0.6800\n",
            "Epoch 74/100\n",
            "21/21 [==============================] - 5s 249ms/step - loss: 0.2445 - acc: 0.8835 - val_loss: 0.7569 - val_acc: 0.7800\n",
            "Epoch 75/100\n",
            "21/21 [==============================] - 5s 250ms/step - loss: 0.3428 - acc: 0.8689 - val_loss: 0.9971 - val_acc: 0.7000\n",
            "Epoch 76/100\n",
            "21/21 [==============================] - 5s 238ms/step - loss: 0.2665 - acc: 0.8641 - val_loss: 0.6086 - val_acc: 0.7400\n",
            "Epoch 77/100\n",
            "21/21 [==============================] - 5s 229ms/step - loss: 0.2391 - acc: 0.9126 - val_loss: 0.5385 - val_acc: 0.7800\n",
            "Epoch 78/100\n",
            "21/21 [==============================] - 5s 239ms/step - loss: 0.2518 - acc: 0.8883 - val_loss: 0.7131 - val_acc: 0.7400\n",
            "Epoch 79/100\n",
            "21/21 [==============================] - 5s 250ms/step - loss: 0.1924 - acc: 0.9369 - val_loss: 0.5008 - val_acc: 0.8200\n",
            "Epoch 80/100\n",
            "21/21 [==============================] - 5s 229ms/step - loss: 0.2476 - acc: 0.9078 - val_loss: 1.0120 - val_acc: 0.6800\n",
            "Epoch 81/100\n",
            "21/21 [==============================] - 5s 237ms/step - loss: 0.2874 - acc: 0.8835 - val_loss: 0.5096 - val_acc: 0.7600\n",
            "Epoch 82/100\n",
            "21/21 [==============================] - 5s 243ms/step - loss: 0.3174 - acc: 0.8495 - val_loss: 0.6009 - val_acc: 0.8000\n",
            "Epoch 83/100\n",
            "21/21 [==============================] - 5s 224ms/step - loss: 0.3564 - acc: 0.8738 - val_loss: 0.4840 - val_acc: 0.8000\n",
            "Epoch 84/100\n",
            "21/21 [==============================] - 5s 239ms/step - loss: 0.2269 - acc: 0.9078 - val_loss: 0.8454 - val_acc: 0.6800\n",
            "Epoch 85/100\n",
            "21/21 [==============================] - 5s 242ms/step - loss: 0.2331 - acc: 0.9223 - val_loss: 0.9146 - val_acc: 0.7200\n",
            "Epoch 86/100\n",
            "21/21 [==============================] - 5s 235ms/step - loss: 0.3063 - acc: 0.8883 - val_loss: 0.5887 - val_acc: 0.7800\n",
            "Epoch 87/100\n",
            "21/21 [==============================] - 5s 236ms/step - loss: 0.3746 - acc: 0.8350 - val_loss: 0.7258 - val_acc: 0.6800\n",
            "Epoch 88/100\n",
            "21/21 [==============================] - 5s 231ms/step - loss: 0.2826 - acc: 0.8883 - val_loss: 0.7513 - val_acc: 0.6600\n",
            "Epoch 89/100\n",
            "21/21 [==============================] - 5s 242ms/step - loss: 0.2427 - acc: 0.9143 - val_loss: 0.8077 - val_acc: 0.6600\n",
            "Epoch 90/100\n",
            "21/21 [==============================] - 5s 236ms/step - loss: 0.1815 - acc: 0.9175 - val_loss: 0.9015 - val_acc: 0.6400\n",
            "Epoch 91/100\n",
            "21/21 [==============================] - 5s 243ms/step - loss: 0.2911 - acc: 0.8544 - val_loss: 0.5750 - val_acc: 0.7000\n",
            "Epoch 92/100\n",
            "21/21 [==============================] - 5s 231ms/step - loss: 0.2329 - acc: 0.8981 - val_loss: 0.7672 - val_acc: 0.6600\n",
            "Epoch 93/100\n",
            "21/21 [==============================] - 5s 252ms/step - loss: 0.2483 - acc: 0.8981 - val_loss: 0.6538 - val_acc: 0.7200\n",
            "Epoch 94/100\n",
            "21/21 [==============================] - 5s 228ms/step - loss: 0.2706 - acc: 0.8883 - val_loss: 0.6331 - val_acc: 0.7400\n",
            "Epoch 95/100\n",
            "21/21 [==============================] - 5s 237ms/step - loss: 0.1995 - acc: 0.9078 - val_loss: 0.6208 - val_acc: 0.7200\n",
            "Epoch 96/100\n",
            "21/21 [==============================] - 5s 248ms/step - loss: 0.2599 - acc: 0.8883 - val_loss: 0.8255 - val_acc: 0.7400\n",
            "Epoch 97/100\n",
            "21/21 [==============================] - 5s 233ms/step - loss: 0.2762 - acc: 0.9029 - val_loss: 0.6102 - val_acc: 0.7400\n",
            "Epoch 98/100\n",
            "21/21 [==============================] - 5s 222ms/step - loss: 0.2813 - acc: 0.8738 - val_loss: 1.2934 - val_acc: 0.6200\n",
            "Epoch 99/100\n",
            "21/21 [==============================] - 5s 254ms/step - loss: 0.2452 - acc: 0.8883 - val_loss: 1.0708 - val_acc: 0.6800\n",
            "Epoch 100/100\n",
            "21/21 [==============================] - 5s 221ms/step - loss: 0.2693 - acc: 0.8738 - val_loss: 0.6739 - val_acc: 0.6800\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f0gxOnFL8CK5",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 338
        },
        "outputId": "a35ef867-19b2-4237-c27c-948991f09a54"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "def plot_acc_loss(result, epochs):\n",
        " loss = result.history['loss']\n",
        " val_loss = result.history['val_loss']\n",
        " plt.figure(figsize=(15, 5))\n",
        " plt.plot(range(1,epochs), loss[1:], label='Train_loss')\n",
        " plt.plot(range(1,epochs), val_loss[1:], label='Test_loss')\n",
        " plt.title('Loss over ' + str(epochs) + 'Epochs', size=15)\n",
        " plt.legend()\n",
        " plt.grid(True)\n",
        " plt.show()\n",
        " \n",
        "plot_acc_loss(history, 100)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lTae9Tml8DpB",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 156
        },
        "outputId": "7d17e835-be0f-4583-efe9-a43f2f743fdb"
      },
      "source": [
        "test_datagen = ImageDataGenerator(rescale=1. / 255)\n",
        "test_generator = test_datagen.flow_from_directory(\n",
        " test_dir,\n",
        " target_size=img_size,\n",
        " batch_size=1,\n",
        " shuffle=False,\n",
        " class_mode='categorical')\n",
        "\n",
        "x = model.evaluate_generator(\n",
        " test_generator,\n",
        " steps = test_generator.n,\n",
        " verbose = 1\n",
        " )\n",
        "print('Test loss:', x[0])\n",
        "print('Test accuracy:',x[1])"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 36 images belonging to 4 classes.\n",
            "WARNING:tensorflow:From <ipython-input-31-6ec24265ab2c>:12: Model.evaluate_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use Model.evaluate, which supports generators.\n",
            "36/36 [==============================] - 1s 16ms/step - loss: 0.5970 - acc: 0.7222\n",
            "Test loss: 0.5970192551612854\n",
            "Test accuracy: 0.7222222089767456\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "name": "Primeiro Teste.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}