Issue#14_IIT2019092.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "import os\n",
    "import math\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "\n",
    "# uncomment the following line if 'imutils' is not installed in your python kernel\n",
    "# !{sys.executable} -m pip install imutils\n",
    "import imutils\n",
    "from imutils import paths\n",
    "\n",
    "\n",
    "import keras\n",
    "from keras.layers import Dense, Conv2D, BatchNormalization, Dropout, Activation, MaxPooling2D, Flatten\n",
    "from keras.models import Sequential, load_model\n",
    "from keras.losses import categorical_crossentropy, binary_crossentropy\n",
    "from keras.optimizers import Adam\n",
    "from keras.utils import np_utils\n",
    "from keras.regularizers import l2\n",
    "from keras import regularizers\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.preprocessing.image import img_to_array\n",
    "from keras.applications.mobilenet_v2 import preprocess_input\n",
    "from keras.preprocessing.image import load_img\n",
    "from keras.utils import to_categorical\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "\n",
    "import cv2\n",
    "import time\n",
    "import random\n",
    "import shutil"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1314 images belonging to 2 classes.\n",
      "Found 194 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "train_datagen = ImageDataGenerator(\n",
    "        rescale=1./255,\n",
    "        shear_range=0.2,\n",
    "        zoom_range=0.2,\n",
    "        horizontal_flip=True)\n",
    "test_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "training_set = train_datagen.flow_from_directory(\n",
    "        'train',\n",
    "        target_size=(224,224),\n",
    "        batch_size=16 ,\n",
    "        class_mode='binary')\n",
    "\n",
    "test_set = test_datagen.flow_from_directory(\n",
    "        'test',\n",
    "        target_size=(224,224),\n",
    "        batch_size=16,\n",
    "        class_mode='binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_8\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_21 (Conv2D)           (None, 222, 222, 32)      896       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_21 (MaxPooling (None, 111, 111, 32)      0         \n",
      "_________________________________________________________________\n",
      "conv2d_22 (Conv2D)           (None, 109, 109, 64)      18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_22 (MaxPooling (None, 54, 54, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_23 (Conv2D)           (None, 52, 52, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_23 (MaxPooling (None, 26, 26, 128)       0         \n",
      "_________________________________________________________________\n",
      "dropout_7 (Dropout)          (None, 26, 26, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_24 (Conv2D)           (None, 26, 26, 128)       147584    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_24 (MaxPooling (None, 13, 13, 128)       0         \n",
      "_________________________________________________________________\n",
      "flatten_7 (Flatten)          (None, 21632)             0         \n",
      "_________________________________________________________________\n",
      "dropout_8 (Dropout)          (None, 21632)             0         \n",
      "_________________________________________________________________\n",
      "dense_16 (Dense)             (None, 256)               5538048   \n",
      "_________________________________________________________________\n",
      "dense_17 (Dense)             (None, 128)               32896     \n",
      "_________________________________________________________________\n",
      "dense_18 (Dense)             (None, 1)                 129       \n",
      "=================================================================\n",
      "Total params: 5,811,905\n",
      "Trainable params: 5,811,905\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model=Sequential()\n",
    "\n",
    "model.add(Conv2D(32, 3, activation='relu', input_shape=(224, 224, 3), kernel_regularizer=regularizers.l1_l2(0.005)))\n",
    "model.add(MaxPooling2D(2,2))\n",
    "\n",
    "model.add(Conv2D(64, 3, activation='relu'))\n",
    "model.add(MaxPooling2D(2,2))\n",
    "          \n",
    "model.add(Conv2D(128, 3, activation='relu'))\n",
    "model.add(MaxPooling2D(2,2)) \n",
    "model.add(Dropout(0.3))\n",
    "\n",
    "model.add(Conv2D(128, 3, activation='relu', padding='same'))\n",
    "model.add(MaxPooling2D(2,2)) \n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dropout(0.3))\n",
    "          \n",
    "model.add(Dense(256, activation ='relu'))\n",
    "model.add(Dense(128, activation ='relu'))\n",
    "model.add(Dense(1, activation ='sigmoid'))         \n",
    "\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "83/83 [==============================] - 262s 3s/step - loss: 0.7810 - accuracy: 0.8988 - val_loss: 0.7875 - val_accuracy: 0.9175\n",
      "Epoch 2/30\n",
      "83/83 [==============================] - 255s 3s/step - loss: 0.3935 - accuracy: 0.9597 - val_loss: 0.6168 - val_accuracy: 0.9536\n",
      "Epoch 3/30\n",
      "83/83 [==============================] - 240s 3s/step - loss: 0.3366 - accuracy: 0.9688 - val_loss: 0.5308 - val_accuracy: 0.9330\n",
      "Epoch 4/30\n",
      "83/83 [==============================] - 238s 3s/step - loss: 0.3269 - accuracy: 0.9650 - val_loss: 0.4325 - val_accuracy: 0.9330\n",
      "Epoch 5/30\n",
      "83/83 [==============================] - 270s 3s/step - loss: 0.3682 - accuracy: 0.9505 - val_loss: 0.5814 - val_accuracy: 0.8196\n",
      "Epoch 6/30\n",
      "83/83 [==============================] - 321s 4s/step - loss: 0.3498 - accuracy: 0.9513 - val_loss: 0.4315 - val_accuracy: 0.9639\n",
      "Epoch 7/30\n",
      "83/83 [==============================] - 291s 4s/step - loss: 0.3691 - accuracy: 0.9543 - val_loss: 0.2593 - val_accuracy: 0.9691\n",
      "Epoch 8/30\n",
      "83/83 [==============================] - 280s 3s/step - loss: 0.2614 - accuracy: 0.9718 - val_loss: 0.2620 - val_accuracy: 0.9742\n",
      "Epoch 9/30\n",
      "83/83 [==============================] - 239s 3s/step - loss: 0.2021 - accuracy: 0.9863 - val_loss: 0.2266 - val_accuracy: 0.9897\n",
      "Epoch 10/30\n",
      "83/83 [==============================] - 239s 3s/step - loss: 0.1969 - accuracy: 0.9886 - val_loss: 0.3050 - val_accuracy: 0.9485\n",
      "Epoch 11/30\n",
      "83/83 [==============================] - 321s 4s/step - loss: 0.2029 - accuracy: 0.9817 - val_loss: 0.2343 - val_accuracy: 0.9794\n",
      "Epoch 12/30\n",
      "83/83 [==============================] - 304s 4s/step - loss: 0.1642 - accuracy: 0.9916 - val_loss: 0.3415 - val_accuracy: 0.9381\n",
      "Epoch 13/30\n",
      "83/83 [==============================] - 294s 4s/step - loss: 0.1528 - accuracy: 0.9901 - val_loss: 0.5138 - val_accuracy: 0.8763\n",
      "Epoch 14/30\n",
      "83/83 [==============================] - 308s 4s/step - loss: 0.1746 - accuracy: 0.9833 - val_loss: 0.2086 - val_accuracy: 0.9691\n",
      "Epoch 15/30\n",
      "83/83 [==============================] - 306s 4s/step - loss: 0.1492 - accuracy: 0.9886 - val_loss: 0.1917 - val_accuracy: 0.9742\n",
      "Epoch 16/30\n",
      "83/83 [==============================] - 305s 4s/step - loss: 0.1594 - accuracy: 0.9825 - val_loss: 0.2469 - val_accuracy: 0.9433\n",
      "Epoch 17/30\n",
      "83/83 [==============================] - 268s 3s/step - loss: 0.1462 - accuracy: 0.9840 - val_loss: 0.4200 - val_accuracy: 0.9072\n",
      "Epoch 18/30\n",
      "83/83 [==============================] - 295s 4s/step - loss: 0.1816 - accuracy: 0.9741 - val_loss: 1.3399 - val_accuracy: 0.7423\n",
      "Epoch 19/30\n",
      "83/83 [==============================] - 238s 3s/step - loss: 0.1377 - accuracy: 0.9848 - val_loss: 9.0759 - val_accuracy: 0.5000\n",
      "Epoch 20/30\n",
      "83/83 [==============================] - 238s 3s/step - loss: 0.1191 - accuracy: 0.9863 - val_loss: 0.3716 - val_accuracy: 0.9536\n",
      "Epoch 21/30\n",
      "83/83 [==============================] - 241s 3s/step - loss: 0.1007 - accuracy: 0.9954 - val_loss: 0.3183 - val_accuracy: 0.9691\n",
      "Epoch 22/30\n",
      "83/83 [==============================] - 237s 3s/step - loss: 0.1128 - accuracy: 0.9901 - val_loss: 0.1914 - val_accuracy: 0.9845\n",
      "Epoch 23/30\n",
      "83/83 [==============================] - 237s 3s/step - loss: 0.1177 - accuracy: 0.9840 - val_loss: 0.1705 - val_accuracy: 0.9691\n",
      "Epoch 24/30\n",
      "83/83 [==============================] - 238s 3s/step - loss: 0.0923 - accuracy: 0.9932 - val_loss: 0.1332 - val_accuracy: 0.9794\n",
      "Epoch 25/30\n",
      "83/83 [==============================] - 241s 3s/step - loss: 0.0852 - accuracy: 0.9932 - val_loss: 0.3557 - val_accuracy: 0.9381\n",
      "Epoch 26/30\n",
      "83/83 [==============================] - 241s 3s/step - loss: 0.0961 - accuracy: 0.9916 - val_loss: 0.2872 - val_accuracy: 0.9536\n",
      "Epoch 27/30\n",
      "83/83 [==============================] - 241s 3s/step - loss: 0.1122 - accuracy: 0.9848 - val_loss: 1.2901 - val_accuracy: 0.8093\n",
      "Epoch 28/30\n",
      "83/83 [==============================] - 240s 3s/step - loss: 0.1519 - accuracy: 0.9734 - val_loss: 0.2384 - val_accuracy: 0.9639\n",
      "Epoch 29/30\n",
      "83/83 [==============================] - 239s 3s/step - loss: 0.1102 - accuracy: 0.9833 - val_loss: 0.3601 - val_accuracy: 0.9485\n",
      "Epoch 30/30\n",
      "83/83 [==============================] - 240s 3s/step - loss: 0.0865 - accuracy: 0.9909 - val_loss: 0.2386 - val_accuracy: 0.9588\n"
     ]
    }
   ],
   "source": [
    "history=model.fit(training_set, epochs=30, validation_data=test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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": [
    "plt.plot(history.history['loss'],'r',label='Training Loss')\n",
    "plt.plot(history.history['val_loss'],label='Validation Loss')\n",
    "plt.xlabel('No. of Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# Plotting the accuracy on validation set w.r.t the number of epochs\n",
    "plt.plot(history.history['accuracy'],'r',label='Training Accuracy')\n",
    "plt.plot(history.history['val_accuracy'],label='Validation Accuracy')\n",
    "plt.xlabel('No. of Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('dev_model.h5', history)"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}