Assignment_2_3_CIFAR10_offline_dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feedforward Neural Network Implementation with Keras and TensorFlow: CIFAR-10 Dataset\n",
    "This Jupyter notebook demonstrates the implementation of a feedforward neural network using Keras and TensorFlow. The goal of this assignment is to walk through the process of building, training, and evaluating a neural network for image classification. We will use the MNIST or CIFAR-10 dataset as per your choice.<br>\n",
    "Dataset: https://www.cs.toronto.edu/%7Ekriz/cifar.html <br>\n",
    "Info: The CIFAR-10 dataset consists of 60000 32x32 colour images in 10 classes, with 6000 images per class. There are 50000 training images and 10000 test images.\n",
    "\n",
    "#### **Train and Test Dataset: [Drive link](https://drive.google.com/drive/folders/1DdsOhkvzbAnenlP8iEUZlqMZbLp17b0y?usp=sharing)**\n",
    "I hope the dataset above is compatible with this notebook\n",
    "\n",
    "\n",
    "## Assignment Steps:\n",
    "\n",
    "**a. Import the necessary packages**: We will start by importing the required Python libraries, including Keras, TensorFlow, and other essential packages.\n",
    "\n",
    "**b. Load the training and testing data**: In this step, we will load the dataset (either MNIST or CIFAR-10) for training and testing the neural network.\n",
    "\n",
    "**c. Define the network architecture using Keras**: Here, we will define the architecture of our feedforward neural network, specifying the number of layers, neurons, and activation functions.\n",
    "\n",
    "**d. Train the model using SGD (Stochastic Gradient Descent)**: We will use the Stochastic Gradient Descent optimizer to train the neural network on the training data.\n",
    "\n",
    "**e. Evaluate the network**: After training, we will evaluate the performance of the model on the testing data to assess its accuracy and other relevant metrics.\n",
    "\n",
    "**f. Plot the training loss and accuracy**: We will create plots to visualize the training loss and accuracy over epochs, providing insights into the model's learning progress.\n",
    "\n",
    "Feel free to follow along with the code and explanations provided in this notebook to gain a better understanding of building and training neural networks for image classification.\n",
    "\n",
    "# Build the Image classification model by dividing the model into following 4 stages:\n",
    "a. Loading and preprocessing the image data<br>\n",
    "b. Defining the model’s architecture<br>\n",
    "c. Training the model<br>\n",
    "d. Estimating the model’s performance<br>\n",
    "\n",
    "Note: Both assignment can use this code accordingly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install tensorflow matplotlib\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import matplotlib.pyplot as plt\n",
    "import random\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4c3cab30",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the paths to the training and testing data directories\n",
    "# D:\\Eng\\B.E\\DL\\deepLearning\\dataset\\cifar-10-img\\cifar-10-img\\train\n",
    "train_data_dir = 'dataset/cifar-10-img/cifar-10-img/train'\n",
    "test_data_dir = 'dataset/cifar-10-img/cifar-10-img/test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f1bd496a",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 40079 images belonging to 10 classes.\n",
      "Found 9921 images belonging to 10 classes.\n"
     ]
    }
   ],
   "source": [
    "# Set up an ImageDataGenerator to rescale pixel values to [0, 1]\n",
    "image_data_generator = ImageDataGenerator(rescale=1.0/255)\n",
    "\n",
    "# Define batch sizes\n",
    "train_batch_size = 20000\n",
    "test_batch_size = 2000\n",
    "\n",
    "# Create data generators for training and testing\n",
    "train_generator = image_data_generator.flow_from_directory(\n",
    "    train_data_dir,\n",
    "    target_size=(32, 32),  # Resize images to 28x28 pixels\n",
    "    batch_size=train_batch_size,  # Number of images per training batch\n",
    "    class_mode='categorical',  # One-hot encoded labels\n",
    "    shuffle=True,  # Shuffle the order of images during training\n",
    ")\n",
    "\n",
    "test_generator = image_data_generator.flow_from_directory(\n",
    "    test_data_dir,\n",
    "    target_size=(32, 32),  # Resize images to 28x28 pixels\n",
    "    batch_size=test_batch_size,  # Number of images per testing batch\n",
    "    class_mode='categorical',  # One-hot encoded labels\n",
    "    shuffle=True,  # Shuffle the order of images during testing\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "42c38034",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train, y_train = train_generator[0]\n",
    "x_test, y_test = test_generator[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b7b670e9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X_train (10000, 32, 32, 3)\n",
      "Shape of y_train (10000, 10)\n",
      "Shape of x_test  (5000, 32, 32, 3)\n",
      "Shape of y_test  (5000, 10)\n"
     ]
    }
   ],
   "source": [
    "print(f\"Shape of X_train {x_train.shape}\")\n",
    "print(f\"Shape of y_train {y_train.shape}\")\n",
    "print(f\"Shape of x_test  {x_test.shape}\")\n",
    "print(f\"Shape of y_test  {y_test.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ac00705a",
   "metadata": {},
   "outputs": [],
   "source": [
    "num_classes = 10\n",
    "class_names = ['airplane','automobile','bird','cat','deer',\n",
    "               'dog','frog','horse','ship','truck']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0617f2cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(32, 32, 3)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "f777dae1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a list of layers\n",
    "model = keras.Sequential([\n",
    "    layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Conv2D(32, (3, 3), activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.MaxPooling2D((2, 2)),\n",
    "    \n",
    "    layers.Conv2D(64, (3, 3), activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.MaxPooling2D((2, 2)),\n",
    "    \n",
    "    layers.Conv2D(128, (3, 3), activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.MaxPooling2D((2, 2)),\n",
    "    \n",
    "    layers.Flatten(),\n",
    "    layers.Dense(512, activation='relu'),\n",
    "    layers.Dropout(0.2),\n",
    "    \n",
    "    layers.Dense(512, activation='relu'),\n",
    "    layers.Dropout(0.2),\n",
    "    \n",
    "    layers.Dense(10, activation='softmax')\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "a1a28156",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "500/500 [==============================] - 24s 46ms/step - loss: 1.7975 - accuracy: 0.3537 - val_loss: 1.6716 - val_accuracy: 0.4024\n",
      "Epoch 2/10\n",
      "500/500 [==============================] - 23s 47ms/step - loss: 1.4317 - accuracy: 0.4865 - val_loss: 1.4276 - val_accuracy: 0.4954\n",
      "Epoch 3/10\n",
      "500/500 [==============================] - 26s 51ms/step - loss: 1.2666 - accuracy: 0.5485 - val_loss: 1.5146 - val_accuracy: 0.4802\n",
      "Epoch 4/10\n",
      "500/500 [==============================] - 27s 54ms/step - loss: 1.1415 - accuracy: 0.5966 - val_loss: 1.2333 - val_accuracy: 0.5652\n",
      "Epoch 5/10\n",
      "500/500 [==============================] - 28s 56ms/step - loss: 1.0262 - accuracy: 0.6351 - val_loss: 1.1926 - val_accuracy: 0.5882\n",
      "Epoch 6/10\n",
      "500/500 [==============================] - 29s 57ms/step - loss: 0.9334 - accuracy: 0.6718 - val_loss: 1.1255 - val_accuracy: 0.6098\n",
      "Epoch 7/10\n",
      "500/500 [==============================] - 29s 58ms/step - loss: 0.8418 - accuracy: 0.7046 - val_loss: 1.1030 - val_accuracy: 0.6198\n",
      "Epoch 8/10\n",
      "500/500 [==============================] - 30s 61ms/step - loss: 0.7630 - accuracy: 0.7303 - val_loss: 1.1298 - val_accuracy: 0.6138\n",
      "Epoch 9/10\n",
      "500/500 [==============================] - 30s 61ms/step - loss: 0.6842 - accuracy: 0.7582 - val_loss: 1.1761 - val_accuracy: 0.6062\n",
      "Epoch 10/10\n",
      "500/500 [==============================] - 30s 60ms/step - loss: 0.5942 - accuracy: 0.7934 - val_loss: 1.1651 - val_accuracy: 0.6154\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.optimizers import SGD\n",
    "# Compile the model\n",
    "sgd_optimizer = SGD(learning_rate=0.01)\n",
    "\n",
    "model.compile(optimizer=sgd_optimizer, loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# Training the model\n",
    "history = model.fit(x_train, y_train,\n",
    "              batch_size=20,\n",
    "              epochs=10,\n",
    "              validation_data=(x_test, y_test),\n",
    "              shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "3d1247f1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 65,
     "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"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.lineplot(model.history.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "56c877f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "157/157 [==============================] - 4s 23ms/step - loss: 1.1651 - accuracy: 0.6154\n",
      "Loss:  1.1651123762130737\n",
      "Accuracy:  0.6154000163078308\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc = model.evaluate(x_test, y_test)\n",
    "print(\"Loss: \", test_loss)\n",
    "print(\"Accuracy: \", test_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "fc1d2b2d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "157/157 [==============================] - 3s 19ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "airplane\n"
     ]
    }
   ],
   "source": [
    "predicted_value=model.predict(x_test)\n",
    "plt.imshow(x_test[50])\n",
    "plt.show()\n",
    "print(class_names[np.argmax(predicted_value[50], axis=0)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "80c73997",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "lab_env",
   "language": "python",
   "name": "lab_env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}