{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7d413e5c",
   "metadata": {},
   "source": [
    "# Lesson 29: PyTorch neural network demonstration part 2\n",
    "\n",
    "In this notebook, we build a deep neural network (DNN) classifier for the CIFAR-10 dataset using PyTorch's `nn.Sequential` module. The CIFAR-10 dataset consists of 60,000 32x32 color images in 10 classes. We convert the images to grayscale (single channel) and use only fully connected layers (no convolution or pooling) to demonstrate the fundamentals of deep learning classification.\n",
    "\n",
    "- **Dataset**: [CIFAR-10](https://www.cs.toronto.edu/~kriz/cifar.html)\n",
    "- **PyTorch documentation**: [`torchvision.datasets.CFAR10()`](https://docs.pytorch.org/vision/main/generated/torchvision.datasets.CIFAR10.html)\n",
    "\n",
    "## Notebook set-up\n",
    "\n",
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9e9e35c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cpu\n"
     ]
    }
   ],
   "source": [
    "# Standard library imports\n",
    "from pathlib import Path\n",
    "\n",
    "# Third party imports\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "# Set random seeds for reproducibility\n",
    "torch.manual_seed(315)\n",
    "np.random.seed(315)\n",
    "\n",
    "# Check for GPU availability\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "print(f'Using device: {device}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b397c04d",
   "metadata": {},
   "source": [
    "### Hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "07125b1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size = 10000 # Images come in batches of 10,000 already\n",
    "learning_rate = 1e-2\n",
    "epochs = 50\n",
    "print_every = 5 # Print training progress every n epochs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d7fbeb7",
   "metadata": {},
   "source": [
    "## 1. Load and preprocess CIFAR-10 data\n",
    "\n",
    "CIFAR-10 contains 32x32 color images (3 channels) across 10 classes: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck. We convert the images to grayscale for this demonstration.\n",
    "\n",
    "### 1.1. Define transformations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "86aba4a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "transformations = transforms.Compose([\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize((0.5,), (0.5)),\n",
    "    transforms.Grayscale(num_output_channels=1)\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "560f0523",
   "metadata": {},
   "source": [
    "### 1.2. Load datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7936ea8c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create directory for image data\n",
    "data_dir = Path('./data')\n",
    "data_dir.mkdir(parents=True, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c07176ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load images with PyTorch datasets\n",
    "train_dataset = datasets.CIFAR10(\n",
    "    root=data_dir,\n",
    "    train=True,\n",
    "    download=True,\n",
    "    transform=transformations\n",
    ")\n",
    "\n",
    "test_dataset = datasets.CIFAR10(\n",
    "    root=data_dir,\n",
    "    train=False,\n",
    "    download=True,\n",
    "    transform=transformations\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64e239dc",
   "metadata": {},
   "source": [
    "### 1.3. Pre-load data and create data loaders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3acd03ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create tensors and move to GPU\n",
    "x_train_full = torch.stack([img for img, _ in train_dataset]).to(device)\n",
    "y_train_full = torch.tensor([label for _, label in train_dataset]).to(device)\n",
    "\n",
    "x_test = torch.stack([img for img, _ in test_dataset]).to(device)\n",
    "y_test = torch.tensor([label for _, label in test_dataset]).to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "b5c6604a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: torch.Size([40000, 1, 32, 32])\n",
      "y_train shape: torch.Size([40000])\n"
     ]
    }
   ],
   "source": [
    "# Training validation split\n",
    "n_train = int(0.8 * len(x_train_full))\n",
    "indices = torch.randperm(len(x_train_full))\n",
    "\n",
    "x_train = x_train_full[indices[:n_train]]\n",
    "y_train = y_train_full[indices[:n_train]]\n",
    "\n",
    "x_val = x_train_full[indices[n_train:]]\n",
    "y_val = y_train_full[indices[n_train:]]\n",
    "\n",
    "print(f'x_train shape: {x_train.shape}')\n",
    "print(f'y_train shape: {y_train.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "591be909",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_tensor_dataset = torch.utils.data.TensorDataset(x_train, y_train)\n",
    "val_tensor_dataset = torch.utils.data.TensorDataset(x_val, y_val)\n",
    "\n",
    "train_loader = DataLoader(\n",
    "    train_tensor_dataset,\n",
    "    batch_size=batch_size,\n",
    "    shuffle=True\n",
    ")\n",
    "\n",
    "val_loader = DataLoader(\n",
    "    val_tensor_dataset,\n",
    "    batch_size=batch_size,\n",
    "    shuffle=False\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "554f5690",
   "metadata": {},
   "source": [
    "### 1.4. Visualize sample images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "02684f62",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define class names\n",
    "class_names = [\n",
    "    'airplane', 'automobile', 'bird', 'cat', 'deer',\n",
    "    'dog', 'frog', 'horse', 'ship', 'truck'\n",
    "]\n",
    "\n",
    "images, labels = next(iter(train_loader))\n",
    "\n",
    "fig, axes = plt.subplots(2, 5, figsize=(6, 3))\n",
    "axes = axes.flatten()\n",
    "\n",
    "for i, ax in enumerate(axes):\n",
    "    img = images[i].cpu() * 0.5 + 0.5\n",
    "    img = img.numpy().squeeze()\n",
    "    ax.set_title(class_names[labels[i]])\n",
    "    ax.imshow(img, cmap='gray')\n",
    "    ax.axis('off')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80c6da6d",
   "metadata": {},
   "source": [
    "## 2. Build DNN classifier with nn.Sequential\n",
    "\n",
    "We build a fully connected deep neural network using `nn.Sequential`. Since we are not using convolutional layers, we flatten the 32x32x1 grayscale images into a 1024-dimensional vector.\n",
    "\n",
    "### 2.1. Define model architecture"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f686db9",
   "metadata": {},
   "source": [
    "### 2.2. Define loss function and optimizer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90de2ff9",
   "metadata": {},
   "source": [
    "### 2.3. Define training function"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "417c653d",
   "metadata": {},
   "source": [
    "### 2.4. Train model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47e76789",
   "metadata": {},
   "source": [
    "### 2.5. Learning curves"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8158a746",
   "metadata": {},
   "source": [
    "## 3. Evaluate model on test set\n",
    "\n",
    "### 3.1. Calculate test accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff588a34",
   "metadata": {},
   "source": [
    "### 3.4. Confusion matrix"
   ]
  }
 ],
 "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
