Custom.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#### IMPORTING THE LIBERARIES REQUIRED ####\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import cv2\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#### CONVERTING THE IMAGES TO ARRAYS ####\n",
    "\n",
    "training_data = []\n",
    "DATADIR = r\"C:\\Users\\bvraju\\Desktop\\AlexNet\\PetImages\"\n",
    "CATEGORIES = [\"Dog\",\"Cat\"]\n",
    "def create_training_data():\n",
    "    for category in CATEGORIES:\n",
    "        path = os.path.join(DATADIR,category)\n",
    "        class_num = CATEGORIES.index(category)\n",
    "        for img in os.listdir(path):\n",
    "            try:\n",
    "                img_array = cv2.imread(os.path.join(path,img),cv2.IMREAD_GRAYSCALE)\n",
    "                new_array = cv2.resize(img_array,(100,100))\n",
    "                training_data.append([new_array,class_num])\n",
    "            except Exception as e:\n",
    "                pass\n",
    "create_training_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#### creating the labels for the training data ####\n",
    "\n",
    "X_data=[]\n",
    "y_data=[]\n",
    "for x,y in training_data:\n",
    "    X_label = x\n",
    "    y_label = y\n",
    "    X_data.append(X_label)\n",
    "    y_data.append(y_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#### SAVING THE LABELS TO FILE ####\n",
    "\n",
    "import pickle\n",
    "pickle_out = open('X_LABELS_FOR_GRAYSCALE_IMAGES.pickle','wb')\n",
    "pickle.dump(X_data, pickle_out)\n",
    "pickle_out.close()\n",
    "\n",
    "pickle_out = open('y_LABELS_FOR_GRAYSCALE_IMAGES.pickle', 'wb')\n",
    "pickle.dump(y_data, pickle_out)\n",
    "pickle_out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "X_Temp = open('X_LABELS_FOR_GRAYSCALE_IMAGES.pickle','rb')\n",
    "y_Temp = open('y_LABELS_FOR_GRAYSCALE_IMAGES.pickle','rb')\n",
    "import pickle   \n",
    "X = pickle.load(X_Temp)\n",
    "y = pickle.load(y_Temp) \n",
    "X_Temp.close()\n",
    "y_Temp.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape of the image (100, 100)\n",
      "number of total examples (training + valdation + testing) =  24946 24946\n",
      "(24946, 100, 100, 1) (24946, 1)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(\"shape of the image\",X[0].shape)\n",
    "\n",
    "\n",
    "from sklearn.utils import shuffle            # Shuffling the data \n",
    "X,y = shuffle(X,y, random_state=2)\n",
    "\n",
    "import numpy as np    \n",
    "X = np.array(X).reshape(-1, 100,100, 1)       # reshaping and converting them to array\n",
    "y = np.array(y).reshape(-1,1)\n",
    "X = np.array(X/255.0)\n",
    "\n",
    "print(\"number of total examples (training + valdation + testing) = \",len(X),len(y))\n",
    "print(X.shape,y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(19956, 100, 100, 1) (4990, 100, 100, 1) (19956, 1) (4990, 1)\n"
     ]
    }
   ],
   "source": [
    "#### SPLLITING THE DATA INTO TRAIN (training set includes validation set) AND TEST SET ####\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=2)\n",
    "print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#### CREATING THE MODEL ####\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.datasets import cifar10\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout, Activation, Flatten\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D\n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "model.add(Conv2D(64,(3,3),input_shape = X.shape[1:]))\n",
    "model.add(Activation(\"relu\"))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "model.add(Conv2D(64,(3,3)))\n",
    "model.add(Activation(\"relu\"))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dense(64))\n",
    "\n",
    "model.add(Dense(1))\n",
    "model.add(Activation('sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy',\n",
    "            optimizer='adam',\n",
    "            metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 98, 98, 64)        640       \n",
      "_________________________________________________________________\n",
      "activation (Activation)      (None, 98, 98, 64)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 49, 49, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 47, 47, 64)        36928     \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 47, 47, 64)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 23, 23, 64)        0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 33856)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 64)                2166848   \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 65        \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 1)                 0         \n",
      "=================================================================\n",
      "Total params: 2,204,481\n",
      "Trainable params: 2,204,481\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 17960 samples, validate on 1996 samples\n",
      "Epoch 1/5\n",
      "17960/17960 [==============================] - 175s 10ms/sample - loss: 0.6734 - accuracy: 0.6086 - val_loss: 0.5947 - val_accuracy: 0.6859\n",
      "Epoch 2/5\n",
      "17960/17960 [==============================] - 173s 10ms/sample - loss: 0.5614 - accuracy: 0.7165 - val_loss: 0.5672 - val_accuracy: 0.7194\n",
      "Epoch 3/5\n",
      "17960/17960 [==============================] - 174s 10ms/sample - loss: 0.4943 - accuracy: 0.7637 - val_loss: 0.5366 - val_accuracy: 0.7355\n",
      "Epoch 4/5\n",
      "17960/17960 [==============================] - 174s 10ms/sample - loss: 0.4570 - accuracy: 0.7896 - val_loss: 0.5363 - val_accuracy: 0.7415\n",
      "Epoch 5/5\n",
      "17960/17960 [==============================] - 172s 10ms/sample - loss: 0.4196 - accuracy: 0.8097 - val_loss: 0.5267 - val_accuracy: 0.7625\n"
     ]
    }
   ],
   "source": [
    "#### TRAINING THE MODEL AND TESTING ON VALIDATION SET ####\n",
    "num_epochs=5\n",
    "hist = model.fit(X_train,y_train,batch_size=32,epochs=num_epochs,validation_split=0.1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#### VISUALIZING THE LOSS AND ACCURACY ####\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "train_loss=hist.history['loss']\n",
    "val_loss=hist.history['val_loss']\n",
    "train_acc=hist.history['accuracy']\n",
    "val_acc=hist.history['val_accuracy']\n",
    "xc=range(num_epochs)\n",
    "\n",
    "plt.figure(1,figsize=(7,5))\n",
    "plt.plot(xc,train_loss)\n",
    "plt.plot(xc,val_loss)\n",
    "plt.xlabel('num of Epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.title('train_loss vs val_loss')\n",
    "plt.grid(True)\n",
    "plt.legend(['train','val'])\n",
    "#print plt.style.available # use bmh, classic,ggplot for big pictures\n",
    "#plt.style.use(['classic'])\n",
    "\n",
    "plt.figure(2,figsize=(7,5))\n",
    "plt.plot(xc,train_acc)\n",
    "plt.plot(xc,val_acc)\n",
    "plt.xlabel('num of Epochs')\n",
    "plt.ylabel('accuracy')\n",
    "plt.title('train_acc vs val_acc')\n",
    "plt.grid(True)\n",
    "plt.legend(['train','val'],loc=4)\n",
    "#print plt.style.available # use bmh, classic,ggplot for big pictures\n",
    "#plt.style.use(['classic'])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "press 1 to continue...1\n",
      "saved the new weights\n"
     ]
    }
   ],
   "source": [
    "#### SAVING THE TRAINED WEIGHTS (helps to avoid multiple times of training the model) ####\n",
    "\n",
    "sure = int(input(\"press 1 to continue...\"))\n",
    "if sure == 1:\n",
    "    fname = \"Custom_model_TRAINED_WEIGHTS\"\n",
    "    model.save_weights(fname , overwrite=True)\n",
    "    print(\"saved the new weights\")\n",
    "else:\n",
    "    print(\"no new weights are saved\")    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.training.tracking.util.CheckpointLoadStatus at 0x25cda4b9c88>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#### LOADING THE WEIGHTS(this step can be skipped) ####\n",
    "\n",
    "fname = \"Custom_model_TRAINED_WEIGHTS\"\n",
    "model.load_weights(fname)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "prediction on  4990  images.\n"
     ]
    }
   ],
   "source": [
    "#### PREDICTING THE CLASSES FOR THE TEST SET ####\n",
    "\n",
    "predicted=model.predict_classes([X_test])\n",
    "\n",
    "print(\"prediction on \", len(predicted),\" images.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy on test :  0.7735470941883767\n"
     ]
    }
   ],
   "source": [
    "#### PRINTTING THE ACCURACY ON THE TESTING SET ####\n",
    "\n",
    "count=0\n",
    "for i in range(4990):\n",
    "    if predicted[i]==y_test[i]:\n",
    "        count+=1\n",
    "print(\"accuracy on test : \",count/len(X_test))  "
   ]
  },
  {
   "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
}