feedforward_clothing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4073012f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing libraries\n",
    "import torch\n",
    "from torch import nn, optim\n",
    "import matplotlib.pyplot as plt\n",
    "from torchvision import datasets, transforms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b70b91e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpu\n"
     ]
    }
   ],
   "source": [
    "# device config\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "793c5fe2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define a transform to normalize the data\n",
    "transform = transforms.Compose([transforms.ToTensor(), \n",
    "                                transforms.Normalize((0.5), (0.5))]) # normalize data; mean = 0.5, std = 0.5\n",
    "\n",
    "# download and load training dataset\n",
    "train_dataset = datasets.FashionMNIST(root='./data', train=True, transform=transforms.ToTensor(), download=True)\n",
    "train_loader = torch.utils.data.DataLoader(dataset=train_dataset, batch_size = 32, shuffle=True)\n",
    "\n",
    "# download and load testing dataset\n",
    "test_dataset = datasets.FashionMNIST(root='./data', train=False, transform=transforms.ToTensor())\n",
    "test_loader = torch.utils.data.DataLoader(dataset= test_dataset, batch_size = 32, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "392d0bc9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# assign labels\n",
    "labels_map = {\n",
    "    0: \"T-Shirt\",\n",
    "    1: \"Trouser\",\n",
    "    2: \"Pullover\",\n",
    "    3: \"Dress\",\n",
    "    4: \"Coat\",\n",
    "    5: \"Sandal\",\n",
    "    6: \"Shirt\",\n",
    "    7: \"Sneaker\",\n",
    "    8: \"Bag\",\n",
    "    9: \"Ankle Boot\",\n",
    "}\n",
    "\n",
    "figure = plt.figure(figsize=(8, 8))\n",
    "cols, rows = (3,3)\n",
    "\n",
    "for i in range(1, cols * rows + 1):\n",
    "    sample_idx = torch.randint(len(train_dataset), size=(1,)).item()  \n",
    "    img, label = train_dataset[sample_idx]\n",
    "    figure.add_subplot(rows, cols, i)\n",
    "    plt.axis(\"off\")\n",
    "    plt.title(f\"{labels_map[label]}\\n(Shape: {img.shape})\", fontdict={'fontsize':8})\n",
    "    plt.imshow(img.squeeze(), cmap=\"gray\")\n",
    "    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bfe32200",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sequential(\n",
      "  (0): Linear(in_features=784, out_features=256, bias=True)\n",
      "  (1): ReLU()\n",
      "  (2): Dropout(p=0.2, inplace=False)\n",
      "  (3): Linear(in_features=256, out_features=128, bias=True)\n",
      "  (4): ReLU()\n",
      "  (5): Dropout(p=0.2, inplace=False)\n",
      "  (6): Linear(in_features=128, out_features=10, bias=True)\n",
      "  (7): LogSoftmax(dim=1)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "# building the model\n",
    "input_layer = 784  \n",
    "hidden_layers = [256, 128]  \n",
    "output_layer = 10  \n",
    "\n",
    "model = nn.Sequential(\n",
    "    nn.Linear(input_layer, hidden_layers[0]), \n",
    "    nn.ReLU(), \n",
    "    nn.Dropout(p=0.2), \n",
    "    nn.Linear(hidden_layers[0], hidden_layers[1]),\n",
    "    nn.ReLU(),\n",
    "    nn.Dropout(p=0.2),\n",
    "    nn.Linear(hidden_layers[1], hidden_layers[2]), \n",
    "    nn.ReLU(),\n",
    "    nn.Dropout(p=0.2),\n",
    "    nn.Linear(hidden_layers[2], output_layer),\n",
    "    nn.LogSoftmax(dim=1)) #\n",
    "\n",
    "print(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e1eb16e4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1/30..  Train Loss: 0.536570..  Test Loss: 0.432142 Test Accuracy: 0.843351\n",
      "Epoch: 2/30..  Train Loss: 0.399989..  Test Loss: 0.384431 Test Accuracy: 0.861522\n",
      "Epoch: 3/30..  Train Loss: 0.368359..  Test Loss: 0.361004 Test Accuracy: 0.868910\n",
      "Epoch: 4/30..  Train Loss: 0.343894..  Test Loss: 0.357442 Test Accuracy: 0.869209\n",
      "Epoch: 5/30..  Train Loss: 0.328237..  Test Loss: 0.353614 Test Accuracy: 0.871106\n",
      "Epoch: 6/30..  Train Loss: 0.316758..  Test Loss: 0.344125 Test Accuracy: 0.874800\n",
      "Epoch: 7/30..  Train Loss: 0.310239..  Test Loss: 0.338809 Test Accuracy: 0.877196\n",
      "Epoch: 8/30..  Train Loss: 0.298189..  Test Loss: 0.337018 Test Accuracy: 0.883986\n",
      "Epoch: 9/30..  Train Loss: 0.293539..  Test Loss: 0.326067 Test Accuracy: 0.880591\n",
      "Epoch: 10/30..  Train Loss: 0.285993..  Test Loss: 0.330978 Test Accuracy: 0.880990\n",
      "Epoch: 11/30..  Train Loss: 0.278011..  Test Loss: 0.331473 Test Accuracy: 0.878894\n",
      "Epoch: 12/30..  Train Loss: 0.272576..  Test Loss: 0.333446 Test Accuracy: 0.879593\n",
      "Epoch: 13/30..  Train Loss: 0.270397..  Test Loss: 0.336684 Test Accuracy: 0.882288\n",
      "Epoch: 14/30..  Train Loss: 0.264452..  Test Loss: 0.332520 Test Accuracy: 0.886981\n",
      "Epoch: 15/30..  Train Loss: 0.257649..  Test Loss: 0.334492 Test Accuracy: 0.883387\n",
      "Epoch: 16/30..  Train Loss: 0.253804..  Test Loss: 0.332563 Test Accuracy: 0.886482\n",
      "Epoch: 17/30..  Train Loss: 0.250778..  Test Loss: 0.337971 Test Accuracy: 0.882388\n",
      "Epoch: 18/30..  Train Loss: 0.248307..  Test Loss: 0.320497 Test Accuracy: 0.887979\n",
      "Epoch: 19/30..  Train Loss: 0.242564..  Test Loss: 0.322830 Test Accuracy: 0.889577\n",
      "Epoch: 20/30..  Train Loss: 0.240468..  Test Loss: 0.336695 Test Accuracy: 0.888578\n",
      "Epoch: 21/30..  Train Loss: 0.237787..  Test Loss: 0.331396 Test Accuracy: 0.889077\n",
      "Epoch: 22/30..  Train Loss: 0.233593..  Test Loss: 0.328964 Test Accuracy: 0.888678\n",
      "Epoch: 23/30..  Train Loss: 0.230035..  Test Loss: 0.323337 Test Accuracy: 0.892073\n",
      "Epoch: 24/30..  Train Loss: 0.230142..  Test Loss: 0.336312 Test Accuracy: 0.889876\n",
      "Epoch: 25/30..  Train Loss: 0.225793..  Test Loss: 0.325671 Test Accuracy: 0.890875\n",
      "Epoch: 26/30..  Train Loss: 0.222579..  Test Loss: 0.331020 Test Accuracy: 0.893271\n",
      "Epoch: 27/30..  Train Loss: 0.221596..  Test Loss: 0.330692 Test Accuracy: 0.892572\n",
      "Epoch: 28/30..  Train Loss: 0.220325..  Test Loss: 0.350386 Test Accuracy: 0.889077\n",
      "Epoch: 29/30..  Train Loss: 0.216619..  Test Loss: 0.337977 Test Accuracy: 0.891174\n",
      "Epoch: 30/30..  Train Loss: 0.218341..  Test Loss: 0.328460 Test Accuracy: 0.893670\n"
     ]
    }
   ],
   "source": [
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.SGD(model.parameters(), lr=0.001)\n",
    "num_epochs = 30 \n",
    "\n",
    "training_losses = []\n",
    "testing_losses = []\n",
    "\n",
    "# training loop for each epoch\n",
    "for epoch in range(1, num_epochs + 1):\n",
    "    total_training_loss = 0\n",
    "    n_correct = 0  \n",
    "    n_samples = 0  \n",
    "\n",
    "    for i, (images, labels) in enumerate(train_loader):\n",
    "        images = images.view(images.shape[0], -1).reshape(-1, 28 * 28).to(device)\n",
    "        labels = labels.to(device)\n",
    "\n",
    "        # forward pass\n",
    "        outputs = model(images)\n",
    "        loss = criterion(outputs, labels)\n",
    "\n",
    "        # backward pass + optimization\n",
    "        optimizer.zero_grad() # zero the gradient buffers\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        total_training_loss += loss.item()\n",
    "\n",
    "    # test\n",
    "    total_testing_loss = 0\n",
    "    test_correct = 0\n",
    "    with torch.no_grad(): # gradient computation disabled\n",
    "        model.eval()\n",
    "        for images, labels in test_loader:\n",
    "            images = images.view(images.shape[0], -1).to(device) \n",
    "            logps = model(images)\n",
    "            loss = criterion(logps, labels) # compares model's predictions to group truth labels\n",
    "            \n",
    "            total_testing_loss += loss.item()\n",
    "            \n",
    "            ps = torch.exp(model(images)) # calculate class probabilities \n",
    "            \n",
    "            top_p, top_class = ps.topk(1, dim=1) # identifies top predicted class for each image\n",
    "            \n",
    "            equals = top_class == labels.view(*top_class.shape) # check if predicted class matches actual labels\n",
    "            \n",
    "            test_correct += torch.mean(equals.type(torch.FloatTensor)) # update accuracy based on correct predictions\n",
    "\n",
    "        model.train() # dropout back on\n",
    "\n",
    "    # calculate mean training and testing losses for this epoch\n",
    "    training_losses.append(total_training_loss / len(train_loader))\n",
    "    testing_losses.append(total_testing_loss / len(test_loader))\n",
    "\n",
    "    # print the progress for the current epoch\n",
    "    print(f\"Epoch: {epoch}/{num_epochs}.. \",\n",
    "          \"Train Loss: {:3f}.. \".format(training_losses[-1]), \n",
    "          \"Test Loss: {:3f}\".format(testing_losses[-1]),\n",
    "          \"Test Accuracy: {:3f}\".format(test_correct / len(test_loader)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "118c5508",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "check = 7\n",
    "with torch.no_grad():\n",
    "    model.eval() \n",
    "    image, actual_label = test_dataset[check]\n",
    "    image = image.view(1, -1).to(device)\n",
    "    predicted_label = model(image).argmax().item()\n",
    "    actual_clothing = test_dataset.classes[actual_label]\n",
    "    predicted_clothing = test_dataset.classes[predicted_label]\n",
    "\n",
    "# display image, clothing name, predicted clothing name\n",
    "image = image.view(28, 28)\n",
    "plt.imshow(image.cpu(), cmap='gray')\n",
    "plt.title(f\"Actual: {actual_clothing}, Predicted: {predicted_clothing}\")\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}