CODE_model1_train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "IoGS1RK1v_jI"
   },
   "outputs": [],
   "source": [
    "# 1st Step – Import the libraries that will be used\n",
    "\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.layers import Dropout\n",
    "from tensorflow.keras.layers import MaxPooling2D\n",
    "from tensorflow.keras.layers import Conv2D\n",
    "from tensorflow.keras.layers import Flatten\n",
    "from tensorflow.keras.layers import Input\n",
    "from tensorflow.keras.layers import AveragePooling2D\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.applications import VGG16\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.model_selection import train_test_split\n",
    "from imutils import paths\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import argparse\n",
    "import pandas as pd\n",
    "import cv2\n",
    "import numpy as np\n",
    "import os\n",
    "import seaborn as sn\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "CXwTGiPpwg-s"
   },
   "outputs": [],
   "source": [
    "# 2nd Step – Load the arrays built in the step referring to data pre-processing and normalize the input data\n",
    "\n",
    "X_train = np.load(\"/content/drive/My Drive/Python/COVID/Arrays/X_Train.npy\")\n",
    "X_train = X_train/255\n",
    "\n",
    "Y_train = np.load(\"/content/drive/My Drive/Python/COVID/Arrays/Y_Train.npy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1049,
     "status": "ok",
     "timestamp": 1586864942619,
     "user": {
      "displayName": "César Soares",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjnCuTGeBsi0P9nLHQXzmAvmSU4ekinz51pT1mltQ=s64",
      "userId": "03845423099530420401"
     },
     "user_tz": 180
    },
    "id": "1RKojXsDxAkB",
    "outputId": "2515a25c-73b8-4182-c844-8cc3f326d148"
   },
   "outputs": [],
   "source": [
    "# 3rd Step – Split the data into training data and test data\n",
    "\n",
    "X_train,X_test,Y_train,Y_test = train_test_split(X_train,Y_train, test_size = 0.2, random_state = 40)\n",
    "\n",
    "print(f\"X_train shape: {X_train.shape} Y_train shape {Y_train.shape}\")\n",
    "print(f\"X_test shape: {X_test.shape} Y_test shape {Y_test.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 71
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 8582,
     "status": "ok",
     "timestamp": 1586864957585,
     "user": {
      "displayName": "César Soares",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjnCuTGeBsi0P9nLHQXzmAvmSU4ekinz51pT1mltQ=s64",
      "userId": "03845423099530420401"
     },
     "user_tz": 180
    },
    "id": "T9b0A-R_yPvu",
    "outputId": "5c23f91f-69bc-412c-bf24-f481aa3e6f09"
   },
   "outputs": [],
   "source": [
    "# 4th Step – Determining the architecture of the model that will be trained\n",
    "\n",
    "bModel = VGG16(weights=\"imagenet\", include_top=False,\n",
    "\tinput_tensor=Input(shape=(237, 237, 3)))\n",
    "\n",
    "tModel = bModel.output\n",
    "tModel = AveragePooling2D(pool_size=(4, 4))(tModel)\n",
    "tModel = Flatten(name=\"flatten\")(tModel)\n",
    "tModel = Dense(64, activation=\"relu\")(tModel)\n",
    "tModel = Dropout(0.2)(tModel)\n",
    "tModel = Dense(1, activation=\"sigmoid\")(tModel)\n",
    "\n",
    "model = Model(inputs=bModel.input, outputs=tModel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "bMDJX0paylb4"
   },
   "outputs": [],
   "source": [
    "# 5th Step – Determine the hyperparameters and compile the model\n",
    "\n",
    "INIT_LR = 1e-3\n",
    "EPOCHS = 50\n",
    "BS = 8\n",
    "\n",
    "for layer in bModel.layers:\n",
    "\tlayer.trainable = False\n",
    "\n",
    "opt = Adam(lr=INIT_LR)\n",
    "model.compile(loss=\"binary_crossentropy\", optimizer=opt,\n",
    "\tmetrics=[\"accuracy\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 64169,
     "status": "ok",
     "timestamp": 1586865031874,
     "user": {
      "displayName": "César Soares",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjnCuTGeBsi0P9nLHQXzmAvmSU4ekinz51pT1mltQ=s64",
      "userId": "03845423099530420401"
     },
     "user_tz": 180
    },
    "id": "fXEHkwQ4y-bw",
    "outputId": "313925dd-c8ad-4263-e721-cddd977227bc"
   },
   "outputs": [],
   "source": [
    "# 6º Step – Train the model\n",
    "\n",
    "X = model.fit(X_train, Y_train, batch_size=BS,validation_split=0.1, epochs=EPOCHS)\n",
    "\n",
    "model.save(\"/content/drive/My Drive/Python/COVID/model/modeloc_1.hdf5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 607
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2991,
     "status": "ok",
     "timestamp": 1586865051760,
     "user": {
      "displayName": "César Soares",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjnCuTGeBsi0P9nLHQXzmAvmSU4ekinz51pT1mltQ=s64",
      "userId": "03845423099530420401"
     },
     "user_tz": 180
    },
    "id": "s996fVhN0Bba",
    "outputId": "9a2bdbb2-aad6-4dc8-ba8c-9bc466433903"
   },
   "outputs": [],
   "source": [
    "# 7th Step – Observe the accuracy of the model and the loss function\n",
    "\n",
    "plt.plot(X.history['accuracy'])\n",
    "plt.plot(X.history['val_accuracy'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "plt.plot(X.history['loss'])\n",
    "plt.plot(X.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "model.evaluate(X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 893
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 9796,
     "status": "ok",
     "timestamp": 1586865114582,
     "user": {
      "displayName": "César Soares",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjnCuTGeBsi0P9nLHQXzmAvmSU4ekinz51pT1mltQ=s64",
      "userId": "03845423099530420401"
     },
     "user_tz": 180
    },
    "id": "ONo53UaLXq2U",
    "outputId": "837e6e20-c7fb-4c26-83f7-3e045b165955"
   },
   "outputs": [],
   "source": [
    "# 8º Step – Observe which images the model got right\n",
    "\n",
    "plt.figure(figsize=(20,20))\n",
    "i = 0\n",
    "for i,image in enumerate(X_test):\n",
    "    plt.subplot(7,9,i+1)\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "    plt.grid(False)\n",
    "    plt.imshow(image, cmap=plt.cm.binary)\n",
    "    img = np.expand_dims(X_test[i],axis = 0)\n",
    "    x_pred = model.predict(img)[0]\n",
    "    x0 = x_pred[0]\n",
    "    \n",
    "    if x0 > 0.5:\n",
    "      label = \"COVID\"\n",
    "    else:\n",
    "      label = \"NORMAL\"\n",
    "    \n",
    "    if Y_test[i] == 1:\n",
    "      label_test = \"COVID\"\n",
    "    else:\n",
    "      label_test = \"NORMAL\"\n",
    "    plt.xlabel(f\"Label Predict = {label} \\n Label Correct = {label_test}\")\n",
    "    i += 1\n",
    "plt.savefig('/content/drive/My Drive/Python/COVID/model/modelo_1.pdf')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 428
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1659,
     "status": "ok",
     "timestamp": 1586866201907,
     "user": {
      "displayName": "César Soares",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjnCuTGeBsi0P9nLHQXzmAvmSU4ekinz51pT1mltQ=s64",
      "userId": "03845423099530420401"
     },
     "user_tz": 180
    },
    "id": "YhZcU8hUc8n2",
    "outputId": "ee12298e-c6b1-44e2-963c-678a5aa951e3"
   },
   "outputs": [],
   "source": [
    "# 9º Step – Build a confusion matrix\n",
    "\n",
    "ypredict = model.predict(X_test)\n",
    "\n",
    "ypredictc = []\n",
    "\n",
    "for value in ypredict:\n",
    "  x0 = value [0]\n",
    "  # x1 = value [1]\n",
    "  if x0 > 0.5:\n",
    "    ypredictc.append(1)\n",
    "\n",
    "  else:\n",
    "    ypredictc.append(0)\n",
    "\n",
    "resultado = np.array(ypredictc)\n",
    "\n",
    "\n",
    "\n",
    "x = confusion_matrix(y_true=Y_test,y_pred=resultado)\n",
    "x = x/X_test.shape[0]\n",
    "\n",
    "y = pd.DataFrame(x,index = [\"NORMAL\",\"COVID\"],columns=[\"NORMAL\",\"COVID\"])\n",
    "plt.figure(figsize = (10,7))\n",
    "\n",
    "fig = sn.heatmap(y, annot=True,cmap=\"Greens\").get_figure()\n",
    "fig.savefig(\"plot.jpg\") "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "1J4hVYFAiliN"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Untitled0.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}