Brain_tumor_classification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "# -*- coding: utf-8 -*-\n",
        "\"\"\"\n",
        "Created on Thu Oct 15 16:43:19 2020\n",
        "\n",
        "@author: AHMED\n",
        "\"\"\"\n",
        "\n",
        "\n",
        "# -*- coding: utf-8 -*-\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "import tensorflow as tf\n",
        "import os\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from keras.models import Sequential, Model\n",
        "from keras.optimizers import RMSprop\n",
        "from keras.layers import Activation, Dropout, Flatten, Dense, GlobalMaxPooling2D, Conv2D, MaxPooling2D\n",
        "from keras.callbacks import CSVLogger\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "\n",
        "\n",
        "path = '/content/dataset'\n",
        "train_dir = os.path.join(path, 'train')\n",
        "path2='/content/dataset2'\n",
        "test_dir = os.path.join(path2, 'test')\n",
        "print(train_dir)\n",
        "print(test_dir)\n",
        "print(os.listdir(train_dir))\n",
        "\n",
        "# Hyperparams\n",
        "IMAGE_SIZE = 128\n",
        "IMAGE_WIDTH, IMAGE_HEIGHT = IMAGE_SIZE, IMAGE_SIZE\n",
        "EPOCHS = 10\n",
        "BATCH_SIZE = 16\n",
        "\n",
        "input_shape = (IMAGE_WIDTH, IMAGE_HEIGHT, 3)\n",
        "\n",
        "# data generators\n",
        "training_data_generator = ImageDataGenerator(\n",
        "        rescale=1./255,\n",
        "        shear_range=0.2,\n",
        "        zoom_range=0.2,\n",
        "        horizontal_flip=True)\n",
        "\n",
        "validation_data_generator = ImageDataGenerator(rescale=1./255)\n",
        "# Data preparation\n",
        "\n",
        "training_generator = training_data_generator.flow_from_directory(\n",
        "    train_dir,\n",
        "    target_size=(IMAGE_WIDTH, IMAGE_HEIGHT),\n",
        "    batch_size=BATCH_SIZE,\n",
        "    class_mode=\"binary\")\n",
        "validation_generator = validation_data_generator.flow_from_directory(\n",
        "    test_dir,\n",
        "    target_size=(IMAGE_WIDTH, IMAGE_HEIGHT),\n",
        "    batch_size=BATCH_SIZE,\n",
        "    class_mode=\"binary\")\n",
        "\n",
        "\n",
        "sample, label = next(validation_generator)\n",
        "print(sample[0])\n",
        "print(label[0])\n",
        "\n",
        "# model\n",
        "model = Sequential()\n",
        "\n",
        "model.add(Conv2D(32, 3, input_shape=input_shape, activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(64, 3, activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(128, 3, activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(256, 3, activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dropout(0.5))\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dense(1, activation = 'sigmoid'))\n",
        "\n",
        "# model.add(Activation('sigmoid'))\n",
        "model.summary()\n",
        "# compile model\n",
        "\n",
        "model.compile(loss='binary_crossentropy',\n",
        "            optimizer='adam',\n",
        "            metrics=['accuracy'])\n",
        "\n",
        "print(len(training_generator.filenames))\n",
        "# train model\n",
        "\n",
        "history=model.fit(\n",
        "    training_generator,\n",
        "    epochs=EPOCHS,\n",
        "    validation_data=validation_generator,\n",
        ")\n",
        "\n",
        "\n",
        "model.save('keras_model2.h5')\n",
        " #   validation_steps=len(validation_generator.filenames) // BATCH_SIZE\n",
        "\n",
        "model = tf.keras.models.load_model('keras_model2.h5')\n",
        "\n",
        "\n",
        "sample1, label1 = next(validation_generator)\n",
        "predictions = model.predict(sample1)\n",
        "print(predictions)\n",
        "\n",
        "\n",
        "def check_results():\n",
        "   class_names = [ 'No', 'Yes']\n",
        "   sample1, label1 = next(validation_generator)\n",
        "   predictions = model.predict(sample1)\n",
        "   for num in range(len(predictions)):\n",
        "       if predictions[num] > 0.5:    \n",
        "           print('prediction: '+'Yes'+' ' + str(int(predictions[num]*100))+ '%')\n",
        "       else:    \n",
        "           print('prediction: '+'No'+' ' + str(100- int(predictions[num]*100))+ '%')\n",
        "          \n",
        "       print('actual: '+ class_names[int(label1[num])])\n",
        "       plt.imshow(sample1[num])\n",
        "       plt.show()\n",
        "\n",
        "check_results()\n",
        "\n",
        "model.evaluate(training_generator)\n",
        "plt.plot(history.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",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.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"
      ],
      "metadata": {
        "id": "krlYk46lZFOh"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}