TransferLearning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "848102af",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator, img_to_array, array_to_img, load_img\n",
    "from PIL import Image\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "\n",
    "# import cv2\n",
    "# from sklearn.model_selection import train_test_split\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPool2D, Dense, Flatten, Dropout\n",
    "from keras.utils.np_utils import to_categorical\n",
    "\n",
    "# Import pre-trained models\n",
    "from keras.applications.vgg16 import VGG16\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8dd903e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Num of GPUs available:  13\n"
     ]
    }
   ],
   "source": [
    "print(\"Num of GPUs available: \", len(tf.test.gpu_device_name()))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d56d6f23",
   "metadata": {},
   "source": [
    "# VGG-16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "723402d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 32, 32, 3)]       0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 32, 32, 64)        1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 32, 32, 64)        36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 16, 16, 64)        0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 16, 16, 128)       73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 16, 16, 128)       147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 8, 8, 128)         0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 8, 8, 256)         295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 4, 4, 256)         0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 4, 4, 512)         1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 2, 2, 512)         0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 1, 1, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               65664     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 129       \n",
      "=================================================================\n",
      "Total params: 14,780,481\n",
      "Trainable params: 65,793\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.applications import VGG16\n",
    "\n",
    "# Load the VGG16 model without the top layer (include_top=False)\n",
    "vgg = VGG16(weights=\"imagenet\", include_top=False, input_shape=(32, 32, 3))\n",
    "\n",
    "# Freeze all layers in the base VGG16 model\n",
    "for layer in vgg.layers:\n",
    "    layer.trainable = False\n",
    "\n",
    "# Add a new classification layer on top of the pre-trained model\n",
    "x = Flatten()(vgg.output)\n",
    "x = Dense(128, activation='relu')(x)\n",
    "x = Dropout(0.25)(x)\n",
    "x = Dense(1, activation='sigmoid')(x)\n",
    "\n",
    "# Create the new model and compile it\n",
    "model = keras.Model(inputs=vgg.input, outputs=x)\n",
    "model.compile(loss='binary_crossentropy', optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001), metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8a7087d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def pickle_to_set(setName):\n",
    "\n",
    "    with open(setName,'rb') as f:\n",
    "\n",
    "        set = pickle.load(f)\n",
    "\n",
    "    print(\"dataset has been imported :)\")\n",
    "\n",
    "    return set\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "36f91c65",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dataset has been imported :)\n",
      "dataset has been imported :)\n",
      "dataset has been imported :)\n",
      "dataset has been imported :)\n"
     ]
    }
   ],
   "source": [
    "train_x = pickle_to_set('./train_resamp_x.pkl')\n",
    "train_y = pickle_to_set('./train_resamp_y.pkl')\n",
    "\n",
    "val_x = pickle_to_set('./val_resamp_x.pkl')\n",
    "val_y = pickle_to_set('./val_resamp_y.pkl')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c215b184",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape of train_x: (934, 32, 32, 3)\n",
      "shape of train_y: (934,)\n",
      "shape of val_x: (234, 32, 32, 3)\n",
      "shape of val_y: (234,)\n"
     ]
    }
   ],
   "source": [
    "print(f\"shape of train_x: {train_x.shape}\")\n",
    "print(f\"shape of train_y: {train_y.shape}\")\n",
    "print(f\"shape of val_x: {val_x.shape}\")\n",
    "print(f\"shape of val_y: {val_y.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "66afeda0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "829/829 [==============================] - 14s 16ms/step - loss: 0.6671 - accuracy: 0.9583 - val_loss: 0.4725 - val_accuracy: 0.9810\n",
      "Epoch 2/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.4091 - accuracy: 0.9674 - val_loss: 0.3755 - val_accuracy: 0.9819\n",
      "Epoch 3/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.2922 - accuracy: 0.9677 - val_loss: 0.2388 - val_accuracy: 0.9817\n",
      "Epoch 4/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.2091 - accuracy: 0.9716 - val_loss: 0.1759 - val_accuracy: 0.9814\n",
      "Epoch 5/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.1592 - accuracy: 0.9725 - val_loss: 0.1556 - val_accuracy: 0.9817\n",
      "Epoch 6/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.1171 - accuracy: 0.9766 - val_loss: 0.1328 - val_accuracy: 0.9810\n",
      "Epoch 7/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.1039 - accuracy: 0.9789 - val_loss: 0.1234 - val_accuracy: 0.9813\n",
      "Epoch 8/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.0889 - accuracy: 0.9803 - val_loss: 0.1187 - val_accuracy: 0.9811\n",
      "Epoch 9/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.0793 - accuracy: 0.9813 - val_loss: 0.1195 - val_accuracy: 0.9811\n",
      "Epoch 10/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.0765 - accuracy: 0.9827 - val_loss: 0.1143 - val_accuracy: 0.9808\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x209a3748df0>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epoch = 10\n",
    "batch_size =32\n",
    "\n",
    "# This callback will stop the training when there is no improvement in\n",
    "# the loss for three consecutive epochs.\n",
    "callback = tf.keras.callbacks.EarlyStopping(monitor='loss', patience=3)\n",
    "\n",
    "# Train Model\n",
    "history = model.fit(train_x, train_y, batch_size= batch_size, callbacks=[callback], epochs=epoch,\n",
    "               validation_data= (val_x, val_y))\n",
    "\n",
    "# Display Accuracy and Loss values of Training and Validation data\n",
    "\n",
    "plt.figure(0)\n",
    "plt.plot(history.history['accuracy'], label='training accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='val accuracy')\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.figure(1)\n",
    "plt.plot(history.history['loss'], label='training loss')\n",
    "plt.plot(history.history['val_loss'], label='val loss')\n",
    "plt.title('Loss')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "e6f48ba6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save model\n",
    "model.save(\"VGG-16.h5\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c97fcd5a",
   "metadata": {},
   "source": [
    "# VGG-16 with Augmented Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "47ef633c",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_augmentation = tf.keras.Sequential([                                    \n",
    "  tf.keras.layers.RandomFlip('horizontal',\n",
    "         input_shape=(32, 32, 3)),\n",
    "  tf.keras.layers.RandomRotation(0.2),\n",
    "  tf.keras.layers.RandomZoom(0.2),\n",
    "  tf.keras.layers.RandomWidth(0.2),\n",
    "  tf.keras.layers.RandomHeight(0.2),\n",
    "  tf.keras.layers.RandomContrast(0.2)                                  \n",
    "], name='data_augmentation')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "6bc71461",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_17\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_28 (InputLayer)        [(None, 32, 32, 3)]       0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 32, 32, 64)        1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 32, 32, 64)        36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 16, 16, 64)        0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 16, 16, 128)       73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 16, 16, 128)       147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 8, 8, 128)         0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 8, 8, 256)         295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 4, 4, 256)         0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 4, 4, 512)         1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 2, 2, 512)         0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 1, 1, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_15 (Flatten)         (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_34 (Dense)             (None, 128)               65664     \n",
      "_________________________________________________________________\n",
      "dropout_19 (Dropout)         (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_35 (Dense)             (None, 1)                 129       \n",
      "=================================================================\n",
      "Total params: 14,780,481\n",
      "Trainable params: 65,793\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.applications import VGG16\n",
    "\n",
    "# Load the VGG16 model without the top layer (include_top=False)\n",
    "vgg = VGG16(weights=\"imagenet\", include_top=False, input_shape=(32, 32, 3))\n",
    "\n",
    "# Freeze all layers in the base VGG16 model\n",
    "for layer in vgg.layers:\n",
    "    layer.trainable = False\n",
    "\n",
    "# Add a new classification layer on top of the pre-trained model\n",
    "inputs = keras.Input(shape=(32, 32, 3))\n",
    "x = data_augmentation(inputs)  # Apply random data augmentation\n",
    "x = Flatten()(vgg.output)\n",
    "x = Dense(128, activation='relu')(x)\n",
    "x = Dropout(0.25)(x)\n",
    "x = Dense(1, activation='sigmoid')(x)\n",
    "\n",
    "# Create the new model and compile it\n",
    "model_vgg_aug = keras.Model(inputs=vgg.input, outputs=x)\n",
    "model_vgg_aug.compile(loss='binary_crossentropy', optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001), metrics=['accuracy'])\n",
    "model_vgg_aug.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "cdd66f04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.6123 - accuracy: 0.9582 - val_loss: 0.3778 - val_accuracy: 0.9813\n",
      "Epoch 2/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.3615 - accuracy: 0.9674 - val_loss: 0.2895 - val_accuracy: 0.9811\n",
      "Epoch 3/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.2581 - accuracy: 0.9690 - val_loss: 0.2376 - val_accuracy: 0.9813\n",
      "Epoch 4/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.1819 - accuracy: 0.9723 - val_loss: 0.1655 - val_accuracy: 0.9805\n",
      "Epoch 5/10\n",
      "829/829 [==============================] - 13s 16ms/step - loss: 0.1405 - accuracy: 0.9749 - val_loss: 0.1502 - val_accuracy: 0.9808\n",
      "Epoch 6/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.1195 - accuracy: 0.9777 - val_loss: 0.1362 - val_accuracy: 0.9811\n",
      "Epoch 7/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.1002 - accuracy: 0.9795 - val_loss: 0.1279 - val_accuracy: 0.9814\n",
      "Epoch 8/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.0876 - accuracy: 0.9805 - val_loss: 0.1234 - val_accuracy: 0.9814\n",
      "Epoch 9/10\n",
      "829/829 [==============================] - 13s 15ms/step - loss: 0.0820 - accuracy: 0.9811 - val_loss: 0.1196 - val_accuracy: 0.9816\n",
      "Epoch 10/10\n",
      "829/829 [==============================] - 12s 15ms/step - loss: 0.0760 - accuracy: 0.9822 - val_loss: 0.1150 - val_accuracy: 0.9814\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2093c22e6d0>"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epoch = 10\n",
    "batch_size =32\n",
    "\n",
    "# This callback will stop the training when there is no improvement in\n",
    "# the loss for three consecutive epochs.\n",
    "callback = tf.keras.callbacks.EarlyStopping(monitor='loss', patience=3)\n",
    "\n",
    "# Train Model\n",
    "history = model_vgg_aug.fit(train_x, train_y, batch_size= batch_size, callbacks=[callback], epochs=epoch,\n",
    "               validation_data= (val_x, val_y))\n",
    "\n",
    "# Display Accuracy and Loss values of Training and Validation data\n",
    "\n",
    "plt.figure(0)\n",
    "plt.plot(history.history['accuracy'], label='training accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='val accuracy')\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.figure(1)\n",
    "plt.plot(history.history['loss'], label='training loss')\n",
    "plt.plot(history.history['val_loss'], label='val loss')\n",
    "plt.title('Loss')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "be575514",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save model\n",
    "model_vgg_aug.save(\"VGG-16_aug.h5\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d80497e",
   "metadata": {},
   "source": [
    "# VGG-16 with Resampled Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a40e7085",
   "metadata": {},
   "source": [
    "## Import resampled datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9355b828",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 32, 32, 3)]       0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 32, 32, 64)        1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 32, 32, 64)        36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 16, 16, 64)        0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 16, 16, 128)       73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 16, 16, 128)       147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 8, 8, 128)         0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 8, 8, 256)         295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 4, 4, 256)         0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 4, 4, 512)         1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 2, 2, 512)         0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 1, 1, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 128)               65664     \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 1)                 129       \n",
      "=================================================================\n",
      "Total params: 14,780,481\n",
      "Trainable params: 65,793\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.applications import VGG16\n",
    "\n",
    "# Load the VGG16 model without the top layer (include_top=False)\n",
    "vgg = VGG16(weights=\"imagenet\", include_top=False, input_shape=(32, 32, 3))\n",
    "\n",
    "# Freeze all layers in the base VGG16 model\n",
    "for layer in vgg.layers:\n",
    "    layer.trainable = False\n",
    "\n",
    "# Add a new classification layer on top of the pre-trained model\n",
    "x = Flatten()(vgg.output)\n",
    "x = Dense(128, activation='relu')(x)\n",
    "x = Dropout(0.25)(x)\n",
    "x = Dense(1, activation='sigmoid')(x)\n",
    "\n",
    "# Create the new model and compile it\n",
    "model_resamp = keras.Model(inputs=vgg.input, outputs=x)\n",
    "model_resamp.compile(loss='binary_crossentropy', optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001), metrics=['accuracy'])\n",
    "model_resamp.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9545222d",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "30/30 [==============================] - 4s 37ms/step - loss: 0.7182 - accuracy: 0.5064 - val_loss: 0.6434 - val_accuracy: 0.6752\n",
      "Epoch 2/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.6632 - accuracy: 0.5942 - val_loss: 0.6031 - val_accuracy: 0.7137\n",
      "Epoch 3/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.6307 - accuracy: 0.6531 - val_loss: 0.5842 - val_accuracy: 0.7179\n",
      "Epoch 4/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.6176 - accuracy: 0.6670 - val_loss: 0.5718 - val_accuracy: 0.7137\n",
      "Epoch 5/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.6219 - accuracy: 0.6585 - val_loss: 0.5609 - val_accuracy: 0.7265\n",
      "Epoch 6/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5962 - accuracy: 0.6820 - val_loss: 0.5515 - val_accuracy: 0.7479\n",
      "Epoch 7/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5931 - accuracy: 0.6756 - val_loss: 0.5458 - val_accuracy: 0.7436\n",
      "Epoch 8/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5886 - accuracy: 0.6852 - val_loss: 0.5405 - val_accuracy: 0.7479\n",
      "Epoch 9/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5792 - accuracy: 0.6906 - val_loss: 0.5335 - val_accuracy: 0.7521\n",
      "Epoch 10/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5822 - accuracy: 0.6991 - val_loss: 0.5301 - val_accuracy: 0.7521\n",
      "Epoch 11/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5751 - accuracy: 0.6991 - val_loss: 0.5268 - val_accuracy: 0.7521\n",
      "Epoch 12/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5722 - accuracy: 0.7173 - val_loss: 0.5251 - val_accuracy: 0.7479\n",
      "Epoch 13/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5699 - accuracy: 0.6916 - val_loss: 0.5205 - val_accuracy: 0.7564\n",
      "Epoch 14/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5714 - accuracy: 0.6949 - val_loss: 0.5154 - val_accuracy: 0.7564\n",
      "Epoch 15/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5547 - accuracy: 0.7152 - val_loss: 0.5127 - val_accuracy: 0.7607\n",
      "Epoch 16/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5498 - accuracy: 0.7248 - val_loss: 0.5129 - val_accuracy: 0.7479\n",
      "Epoch 17/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5626 - accuracy: 0.7099 - val_loss: 0.5079 - val_accuracy: 0.7564\n",
      "Epoch 18/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5513 - accuracy: 0.7184 - val_loss: 0.5048 - val_accuracy: 0.7564\n",
      "Epoch 19/20\n",
      "30/30 [==============================] - 0s 15ms/step - loss: 0.5507 - accuracy: 0.7238 - val_loss: 0.5046 - val_accuracy: 0.7521\n",
      "Epoch 20/20\n",
      "30/30 [==============================] - 0s 16ms/step - loss: 0.5485 - accuracy: 0.7248 - val_loss: 0.5014 - val_accuracy: 0.7564\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1fb12c2ffa0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epoch = 20\n",
    "batch_size =32\n",
    "\n",
    "# This callback will stop the training when there is no improvement in\n",
    "# the loss for three consecutive epochs.\n",
    "callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=3)\n",
    "\n",
    "# Train Model\n",
    "history = model_resamp.fit(train_x, train_y, batch_size= batch_size, callbacks=[callback], epochs=epoch,\n",
    "               validation_data= (val_x, val_y))\n",
    "\n",
    "# Display Accuracy and Loss values of Training and Validation data\n",
    "\n",
    "plt.figure(0)\n",
    "plt.plot(history.history['accuracy'], label='training accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='val accuracy')\n",
    "plt.title('Accuracy')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('accuracy')\n",
    "plt.legend()\n",
    "\n",
    "plt.figure(1)\n",
    "plt.plot(history.history['loss'], label='training loss')\n",
    "plt.plot(history.history['val_loss'], label='val loss')\n",
    "plt.title('Loss')\n",
    "plt.xlabel('epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c413ca92",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save model\n",
    "model_resamp.save(\"VGG-16_new.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5de2c4ca",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}