Image Classifier Using CNN Model(Cars And Bikes).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6a958b1d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from keras_preprocessing.image import ImageDataGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4b5eea9f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 160 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "train_datagen = ImageDataGenerator(rescale = 1./255,\n",
    "                                   shear_range = 0.2,\n",
    "                                   zoom_range = 0.2,\n",
    "                                   horizontal_flip = True)\n",
    "training_set = train_datagen.flow_from_directory('C:/Users/ADMIN/OneDrive/Desktop/C&B/Training',\n",
    "                                                 target_size = (64, 64),\n",
    "                                                 batch_size = 16,\n",
    "                                                 class_mode = 'binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1a356cab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 40 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "test_datagen = ImageDataGenerator(rescale = 1./255)\n",
    "test_set = test_datagen.flow_from_directory('C:/Users/ADMIN/OneDrive/Desktop/C&B/Testing',\n",
    "                                            target_size = (64, 64),\n",
    "                                            batch_size = 16,\n",
    "                                            class_mode = 'binary')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "fbbc1943",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn = tf.keras.models.Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d196b5b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.add(tf.keras.layers.Conv2D(filters=16, kernel_size=3, activation='relu', input_shape=[64, 64, 3]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "15478eca",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5ca29086",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.add(tf.keras.layers.Conv2D(filters=32, kernel_size=3, activation='relu'))\n",
    "cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9a6001d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding a third convolutional layer\n",
    "cnn.add(tf.keras.layers.Conv2D(filters=32, kernel_size=3, activation='relu'))\n",
    "cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bcf0f6e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.add(tf.keras.layers.Flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e007ac2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.add(tf.keras.layers.Dense(units=50, activation='relu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "63fe0e7b",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.add(tf.keras.layers.Dense(units=1, activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0c1095ce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 62, 62, 16)        448       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 31, 31, 16)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 29, 29, 32)        4640      \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 14, 14, 32)       0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 12, 12, 32)        9248      \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPooling  (None, 6, 6, 32)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 1152)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 50)                57650     \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 51        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 72,037\n",
      "Trainable params: 72,037\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "cnn.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "643c3466",
   "metadata": {},
   "outputs": [],
   "source": [
    "cnn.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fe5085de",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "10/10 [==============================] - 1s 65ms/step - loss: 0.6972 - accuracy: 0.5500 - val_loss: 0.6888 - val_accuracy: 0.5000\n",
      "Epoch 2/20\n",
      "10/10 [==============================] - 1s 54ms/step - loss: 0.6735 - accuracy: 0.5625 - val_loss: 0.6600 - val_accuracy: 0.7250\n",
      "Epoch 3/20\n",
      "10/10 [==============================] - 1s 54ms/step - loss: 0.6152 - accuracy: 0.7563 - val_loss: 0.5941 - val_accuracy: 0.7500\n",
      "Epoch 4/20\n",
      "10/10 [==============================] - 1s 57ms/step - loss: 0.5118 - accuracy: 0.7937 - val_loss: 0.4965 - val_accuracy: 0.8000\n",
      "Epoch 5/20\n",
      "10/10 [==============================] - 1s 54ms/step - loss: 0.4576 - accuracy: 0.7437 - val_loss: 0.4915 - val_accuracy: 0.7250\n",
      "Epoch 6/20\n",
      "10/10 [==============================] - 1s 52ms/step - loss: 0.4406 - accuracy: 0.8000 - val_loss: 0.4404 - val_accuracy: 0.8000\n",
      "Epoch 7/20\n",
      "10/10 [==============================] - 1s 56ms/step - loss: 0.3755 - accuracy: 0.8375 - val_loss: 0.4503 - val_accuracy: 0.9250\n",
      "Epoch 8/20\n",
      "10/10 [==============================] - 1s 51ms/step - loss: 0.3863 - accuracy: 0.8313 - val_loss: 0.4616 - val_accuracy: 0.7750\n",
      "Epoch 9/20\n",
      "10/10 [==============================] - 1s 55ms/step - loss: 0.3680 - accuracy: 0.8375 - val_loss: 0.4240 - val_accuracy: 0.8000\n",
      "Epoch 10/20\n",
      "10/10 [==============================] - 1s 63ms/step - loss: 0.3353 - accuracy: 0.8875 - val_loss: 0.4280 - val_accuracy: 0.8500\n",
      "Epoch 11/20\n",
      "10/10 [==============================] - 1s 53ms/step - loss: 0.3042 - accuracy: 0.8562 - val_loss: 0.4548 - val_accuracy: 0.8250\n",
      "Epoch 12/20\n",
      "10/10 [==============================] - 1s 54ms/step - loss: 0.3411 - accuracy: 0.8750 - val_loss: 0.4861 - val_accuracy: 0.8000\n",
      "Epoch 13/20\n",
      "10/10 [==============================] - 1s 55ms/step - loss: 0.3257 - accuracy: 0.8813 - val_loss: 0.5125 - val_accuracy: 0.8000\n",
      "Epoch 14/20\n",
      "10/10 [==============================] - 1s 52ms/step - loss: 0.2764 - accuracy: 0.8750 - val_loss: 0.4870 - val_accuracy: 0.8250\n",
      "Epoch 15/20\n",
      "10/10 [==============================] - 1s 53ms/step - loss: 0.2836 - accuracy: 0.8750 - val_loss: 0.4905 - val_accuracy: 0.8500\n",
      "Epoch 16/20\n",
      "10/10 [==============================] - 1s 63ms/step - loss: 0.2465 - accuracy: 0.8750 - val_loss: 0.5744 - val_accuracy: 0.8250\n",
      "Epoch 17/20\n",
      "10/10 [==============================] - 1s 61ms/step - loss: 0.2041 - accuracy: 0.9250 - val_loss: 0.5757 - val_accuracy: 0.8750\n",
      "Epoch 18/20\n",
      "10/10 [==============================] - 1s 64ms/step - loss: 0.2041 - accuracy: 0.9250 - val_loss: 0.6185 - val_accuracy: 0.8500\n",
      "Epoch 19/20\n",
      "10/10 [==============================] - 1s 59ms/step - loss: 0.1892 - accuracy: 0.9250 - val_loss: 0.6020 - val_accuracy: 0.8750\n",
      "Epoch 20/20\n",
      "10/10 [==============================] - 1s 55ms/step - loss: 0.2004 - accuracy: 0.9250 - val_loss: 0.6273 - val_accuracy: 0.8750\n"
     ]
    }
   ],
   "source": [
    "history_model=cnn.fit(x = training_set, validation_data = test_set, epochs = 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d6cae592",
   "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": [
    "# Accuracy comparison\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.plot(history_model.history['accuracy'])\n",
    "plt.plot(history_model.history['val_accuracy'])\n",
    "plt.legend(['accuracy','val_accuracy'])\n",
    "plt.show()\n",
    "\n",
    "\n",
    "# Loss Comparison\n",
    "plt.plot(history_model.history['loss'])\n",
    "plt.plot(history_model.history['val_loss'])\n",
    "plt.legend(['loss','val_loss'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "12f3266e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from keras_preprocessing import image\n",
    "test_image = image.load_img('C:/Users/ADMIN/OneDrive/Desktop/C&B/Training/Cars/images (102).jpg', target_size = (64, 64))\n",
    "test_image = image.img_to_array(test_image)\n",
    "test_image = np.expand_dims(test_image, axis = 0)\n",
    "result = cnn.predict(test_image)\n",
    "training_set.class_indices\n",
    "if result[0][0] == 1:\n",
    "  prediction = 'car'\n",
    "else:\n",
    "  prediction = 'bike'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "afc0ffe6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "car\n"
     ]
    }
   ],
   "source": [
    "print(prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3d08be1a",
   "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": 5
}