covid chest x-ray detection by transfer learning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the Keras libraries and packages\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.applications.vgg16 import VGG16\n",
    "from tensorflow.keras.applications.resnet50 import ResNet50\n",
    "from tensorflow.keras.applications.vgg19 import VGG19\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "from keras.applications import DenseNet121\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.preprocessing import image\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Input, Lambda ,Dense ,Flatten\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"vgg16\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_10 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "=================================================================\n",
      "Total params: 14,714,688\n",
      "Trainable params: 14,714,688\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Initialising the resnet50\n",
    "classifier = VGG16(input_shape= (224,224,3),include_top=False,weights='imagenet')\n",
    "#classifier = DenseNet121(input_shape= (224,224,3),include_top=False,weights='imagenet')\n",
    "#classifier = ResNet50( input_shape=(224,224,3), include_top= False) # include_top will consider the new weights\n",
    "#classifier = VGG19( input_shape=(224,224,3), include_top= False) # include_top will consider the new weights\n",
    "#classifier = InceptionV3( input_shape=(224,224,3), include_top= False) # include_top will consider the new weights\n",
    "\n",
    "\n",
    "classifier.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#don't train existing weights\n",
    "for layer in classifier.layers:\n",
    "    layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier1 = classifier.output#head model\n",
    "classifier1 = Flatten()(classifier1)#adding layer of flatten\n",
    "classifier1 = Dense(units=64, activation='relu')(classifier1)#adding layer of dense\n",
    "classifier1 = Dense(units=1, activation='sigmoid')(classifier1)#again adding another layer of dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_3 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv4 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv4 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv4 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_2 (Flatten)          (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 64)                1605696   \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 21,630,145\n",
      "Trainable params: 1,605,761\n",
      "Non-trainable params: 20,024,384\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Model(inputs = classifier.input , outputs = classifier1)\n",
    "model.summary()\n",
    "\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 100 images belonging to 2 classes.\n",
      "Found 100 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "# Part 2 - Fitting the CNN to the images\n",
    "\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "#use the image data generator to import the images from the dataset\n",
    "#data augmentation\n",
    "train_datagen = ImageDataGenerator(featurewise_center=False,  # set input mean to 0 over the dataset\n",
    "        samplewise_center=False,  # set each sample mean to 0\n",
    "        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n",
    "        samplewise_std_normalization=False,  # divide each input by its std\n",
    "        zca_whitening=False,  # apply ZCA whitening\n",
    "        rotation_range=10,  # randomly rotate images in the range (degrees, 0 to 180)\n",
    "        zoom_range = 0.1, # Randomly zoom image \n",
    "        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)\n",
    "        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)\n",
    "        horizontal_flip=True,  # randomly flip images\n",
    "        vertical_flip=False,  # randomly flip images\n",
    "        rescale = 1./255)\n",
    "\n",
    "test_datagen = ImageDataGenerator(rescale=1. / 255)\n",
    "\n",
    "\n",
    "#makes sure you provide the same target as initialised for the image size\n",
    "training_set = train_datagen.flow_from_directory('G:/projects/corona x-ray images predication/Data/train',\n",
    "                                                 target_size=(224, 224),\n",
    "                                                 batch_size=32,\n",
    "                                                 class_mode='binary')\n",
    "\n",
    "test_set = test_datagen.flow_from_directory('G:/projects/corona x-ray images predication/Data/test',\n",
    "                                            target_size=(224, 224),\n",
    "                                            batch_size=32,\n",
    "                                            class_mode='binary')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "\n",
    "es = EarlyStopping(monitor= \"val_accuracy\" , min_delta= 0.01, patience= 3, verbose=1)\n",
    "mc = ModelCheckpoint(filepath=\"my_model_vgg16.h5\", monitor=\"val_accuracy\", verbose=1, save_best_only= True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "4/4 [==============================] - 113s 31s/step - loss: 2.7462 - accuracy: 0.4900 - val_loss: 0.2161 - val_accuracy: 0.9500\n",
      "\n",
      "Epoch 00001: val_accuracy improved from -inf to 0.95000, saving model to my_model_vgg16.h5\n",
      "Epoch 2/10\n",
      "4/4 [==============================] - 121s 34s/step - loss: 0.4146 - accuracy: 0.8200 - val_loss: 0.5221 - val_accuracy: 0.7200\n",
      "\n",
      "Epoch 00002: val_accuracy did not improve from 0.95000\n",
      "Epoch 3/10\n",
      "4/4 [==============================] - 117s 33s/step - loss: 0.1595 - accuracy: 0.9600 - val_loss: 0.0321 - val_accuracy: 1.0000\n",
      "\n",
      "Epoch 00003: val_accuracy improved from 0.95000 to 1.00000, saving model to my_model_vgg16.h5\n",
      "Epoch 4/10\n",
      "4/4 [==============================] - 136s 37s/step - loss: 0.0760 - accuracy: 0.9700 - val_loss: 0.0196 - val_accuracy: 1.0000\n",
      "\n",
      "Epoch 00004: val_accuracy did not improve from 1.00000\n",
      "Epoch 5/10\n",
      "4/4 [==============================] - 125s 34s/step - loss: 0.0201 - accuracy: 1.0000 - val_loss: 0.1070 - val_accuracy: 0.9400\n",
      "\n",
      "Epoch 00005: val_accuracy did not improve from 1.00000\n",
      "Epoch 6/10\n",
      "4/4 [==============================] - 114s 32s/step - loss: 0.0495 - accuracy: 0.9800 - val_loss: 0.0902 - val_accuracy: 0.9600\n",
      "\n",
      "Epoch 00006: val_accuracy did not improve from 1.00000\n",
      "Epoch 00006: early stopping\n"
     ]
    }
   ],
   "source": [
    "#fit the model\n",
    "#it will take some time to train\n",
    "history = model.fit(training_set,\n",
    "                              validation_data=test_set,\n",
    "                              epochs=10,\n",
    "                              steps_per_epoch=len(training_set),\n",
    "                              validation_steps=len(test_set),\n",
    "                              callbacks=[es,mc])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "#save model\n",
    "model.save('G:/projects/corona x-ray images predication/my_model_vgg16.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4/4 [==============================] - 54s 12s/step - loss: 0.0902 - accuracy: 0.9600\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.09021569043397903, 0.9599999785423279]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#loading model\n",
    "loaded_model = tf.keras.models.load_model('G:/projects/corona x-ray images predication/my_model_vgg16.h5')\n",
    "loaded_model.evaluate(test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#prediction of test set\n",
    "y_pred = model.predict(test_set)\n",
    "import numpy as np\n",
    "y_pred = np.argmax(y_pred , axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting acc and loss\n",
    "import matplotlib.pyplot as plt\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.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",
    "\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', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normal\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# for only one prediction\n",
    "import numpy as np\n",
    "from keras.preprocessing import image\n",
    "\n",
    "test_image = image.load_img(\n",
    "    r'G:\\projects\\corona x-ray images predication\\Data\\test\\Normal\\IM-0236-0001.jpeg',\n",
    "    target_size=(224,224))\n",
    "plt.imshow(test_image)\n",
    "test_image = image.img_to_array(test_image)\n",
    "test_image = np.expand_dims(test_image, axis=0)\n",
    "result = loaded_model.predict(test_image)\n",
    "training_set.class_indices\n",
    "if result[0][0] == 1:\n",
    "    prediction = 'Normal'\n",
    "else:\n",
    "    prediction = 'Covid'\n",
    "print(prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot confusion metrix\n",
    "y_pred = []\n",
    "y_test = []\n",
    "import os\n",
    "\n",
    "for i in os.listdir(\"G:/projects/corona x-ray images predication/Data/test/Normal\"):\n",
    "    img = image.load_img(\"G:/projects/corona x-ray images predication/Data/test/Normal/\" + i, target_size=(224,224))\n",
    "    img = image.img_to_array(img)\n",
    "    img = np.expand_dims(img, axis=0)\n",
    "    p = model.predict(img)\n",
    "    y_test.append(p[0, 0])\n",
    "    y_pred.append(1)\n",
    "\n",
    "for i in os.listdir(\"G:/projects/corona x-ray images predication/Data/test/Covid\"):\n",
    "    img = image.load_img(\"G:/projects/corona x-ray images predication/Data/test/Covid/\" + i, target_size=(224,224))\n",
    "    img = image.img_to_array(img)\n",
    "    img = np.expand_dims(img, axis=0)\n",
    "    p = model.predict(img)\n",
    "    y_test.append(p[0, 0])\n",
    "    y_pred.append(0)\n",
    "\n",
    "y_pred = np.array(y_pred)\n",
    "y_test = np.array(y_test).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.06      0.11        50\n",
      "           1       0.52      1.00      0.68        50\n",
      "\n",
      "    accuracy                           0.53       100\n",
      "   macro avg       0.76      0.53      0.40       100\n",
      "weighted avg       0.76      0.53      0.40       100\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "cm = confusion_matrix(y_pred, y_test)\n",
    "sns.heatmap(cm, cmap=\"plasma\", annot=True)\n",
    "plt.show()\n",
    "\n",
    "print(classification_report(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}