Model Building.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "import cv2\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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"
    }
   ],
   "source": [
    "DATADIR = os.getcwd() + \"/Data\"\n",
    "CATEGORIES = [\"no\", \"yes\"] # 0 for No and 1 for yes\n",
    "IMG_SIZE = 150\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)[1:]:\n",
    "        img_arr = cv2.imread(os.path.join(path,img), cv2.IMREAD_GRAYSCALE)\n",
    "        new_arr = cv2.resize(img_arr, (IMG_SIZE, IMG_SIZE))\n",
    "        plt.imshow(new_arr, cmap = \"gray\")\n",
    "        plt.show()\n",
    "        break\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_data = []\n",
    "def make_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)[1:]:\n",
    "            img_arr = cv2.imread(os.path.join(path,img), cv2.IMREAD_GRAYSCALE)\n",
    "            new_arr = cv2.resize(img_arr, (IMG_SIZE, IMG_SIZE))\n",
    "            training_data.append([new_arr, class_num])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "make_training_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "random.shuffle(training_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = []\n",
    "y = []\n",
    "for feature, label in training_data:\n",
    "    X.append(feature)\n",
    "    y.append(label)\n",
    "#For not reshaped values of features\n",
    "pickle_out = open(\"X_not_reshaped.pickle\", \"wb\")\n",
    "pickle.dump(X, pickle_out)\n",
    "pickle_out.close()\n",
    "#For not reshaped values of labels\n",
    "pickle_out = open(\"y_not_reshaped.pickle\", \"wb\")\n",
    "pickle.dump(y, pickle_out)\n",
    "pickle_out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.array(X).reshape(-1, IMG_SIZE, IMG_SIZE, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "# Saving all the work upon the images using pickle\n",
    "# FOR X\n",
    "pickle_out = open(\"X.pickle\", \"wb\")\n",
    "pickle.dump(X, pickle_out)\n",
    "pickle_out.close()\n",
    "# FOR y\n",
    "pickle_out = open(\"y.pickle\", \"wb\")\n",
    "pickle.dump(y, pickle_out)\n",
    "pickle_out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle_in = open(\"X.pickle\", \"rb\")\n",
    "X = pickle.load(pickle_in)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "29/29 [==============================] - 6s 209ms/step - loss: 0.7181 - accuracy: 0.7353 - val_loss: 0.4727 - val_accuracy: 0.8421\n",
      "Epoch 2/15\n",
      "29/29 [==============================] - 6s 194ms/step - loss: 0.5013 - accuracy: 0.7647 - val_loss: 0.4278 - val_accuracy: 0.7895\n",
      "Epoch 3/15\n",
      "29/29 [==============================] - 6s 194ms/step - loss: 0.3680 - accuracy: 0.8412 - val_loss: 0.3468 - val_accuracy: 0.8421\n",
      "Epoch 4/15\n",
      "29/29 [==============================] - 6s 194ms/step - loss: 0.2408 - accuracy: 0.9059 - val_loss: 0.1421 - val_accuracy: 0.9474\n",
      "Epoch 5/15\n",
      "29/29 [==============================] - 6s 196ms/step - loss: 0.1221 - accuracy: 0.9588 - val_loss: 0.1035 - val_accuracy: 0.9474\n",
      "Epoch 6/15\n",
      "29/29 [==============================] - 6s 197ms/step - loss: 0.0577 - accuracy: 0.9824 - val_loss: 0.1725 - val_accuracy: 0.9474\n",
      "Epoch 7/15\n",
      "29/29 [==============================] - 6s 196ms/step - loss: 0.0326 - accuracy: 0.9941 - val_loss: 0.1173 - val_accuracy: 0.9474\n",
      "Epoch 8/15\n",
      "29/29 [==============================] - 6s 195ms/step - loss: 0.0075 - accuracy: 1.0000 - val_loss: 0.1359 - val_accuracy: 0.9474\n",
      "Epoch 9/15\n",
      "29/29 [==============================] - 6s 196ms/step - loss: 0.0031 - accuracy: 1.0000 - val_loss: 0.0860 - val_accuracy: 0.9474\n",
      "Epoch 10/15\n",
      "29/29 [==============================] - 6s 196ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.1002 - val_accuracy: 0.9474\n",
      "Epoch 11/15\n",
      "29/29 [==============================] - 6s 199ms/step - loss: 9.5516e-04 - accuracy: 1.0000 - val_loss: 0.1089 - val_accuracy: 0.9474\n",
      "Epoch 12/15\n",
      "29/29 [==============================] - 6s 198ms/step - loss: 7.5089e-04 - accuracy: 1.0000 - val_loss: 0.1073 - val_accuracy: 0.9474\n",
      "Epoch 13/15\n",
      "29/29 [==============================] - 6s 197ms/step - loss: 6.1812e-04 - accuracy: 1.0000 - val_loss: 0.1107 - val_accuracy: 0.9474\n",
      "Epoch 14/15\n",
      "29/29 [==============================] - 6s 202ms/step - loss: 5.2784e-04 - accuracy: 1.0000 - val_loss: 0.1067 - val_accuracy: 0.9474\n",
      "Epoch 15/15\n",
      "29/29 [==============================] - 6s 197ms/step - loss: 4.5890e-04 - accuracy: 1.0000 - val_loss: 0.1106 - val_accuracy: 0.9474\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x22c86ef1508>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout, Activation, Flatten, Conv2D, MaxPooling2D\n",
    "\n",
    "X = X/255.0\n",
    "y = np.array(y)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Conv2D(64, (3,3), input_shape = X_train.shape[1:]))\n",
    "model.add(Activation(\"relu\"))\n",
    "model.add(MaxPooling2D(pool_size = (2,2))) # Pooling\n",
    "\n",
    "model.add(Conv2D(64, (3,3), input_shape = X_train.shape[1:]))\n",
    "model.add(Activation(\"relu\"))\n",
    "model.add(MaxPooling2D(pool_size = (2,2))) # Pooling\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'])\n",
    "\n",
    "model.fit(X_train, y_train, batch_size = 6, epochs = 15, validation_split = 0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2/2 [==============================] - 0s 102ms/step - loss: 1.2204 - accuracy: 0.8730\n"
     ]
    }
   ],
   "source": [
    "val_loss, val_accuracy = model.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: 64x3-CNN.model\\assets\n"
     ]
    }
   ],
   "source": [
    "model.save(\"64x3-CNN.model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def reshaper(arr):\n",
    "    return arr.reshape(-1, IMG_SIZE, IMG_SIZE, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}