{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e4fc6dc9-8a30-4e4d-b419-2f2c10f5e5c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.utils.data import DataLoader\n",
    "from torchvision import datasets, models, transforms\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13d31a44-042f-4e4a-ae2c-a56ad331872b",
   "metadata": {},
   "source": [
    "# Load Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2decc2c6-b226-4ae6-ba6a-3d5eb36d3b53",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the data augmentation and transformation\n",
    "data_transforms = {\n",
    "    'train': transforms.Compose([\n",
    "        transforms.Resize((224, 224)),\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
    "    ]),\n",
    "    'test': transforms.Compose([\n",
    "        transforms.Resize((224, 224)),\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
    "    ]),\n",
    "}\n",
    "\n",
    "# Load the dataset\n",
    "data_dir = 'data/task3/ChestXray'\n",
    "image_datasets = {x: datasets.ImageFolder(f'{data_dir}/{x}', data_transforms[x])\n",
    "                  for x in ['train', 'test']}\n",
    "dataloaders = {x: DataLoader(image_datasets[x], batch_size=32, shuffle=True, num_workers=4)\n",
    "               for x in ['train', 'test']}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f71be5e4-c73c-4788-966f-792a2d47fe05",
   "metadata": {},
   "source": [
    "# Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "fdd6c6bc-f854-4945-a8ac-2b00d873dfe4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/prac/lib/python3.9/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "/opt/anaconda3/envs/prac/lib/python3.9/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet50_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet50_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "device = 'cuda:0'\n",
    "\n",
    "# Initialize the model\n",
    "model = models.resnet50(pretrained=True)\n",
    "# Alter the final layer for binary classification\n",
    "model.fc = nn.Linear(model.fc.in_features, 1)\n",
    "model = model.to(device)\n",
    "\n",
    "# Define loss function and optimizer\n",
    "criterion = nn.BCEWithLogitsLoss()\n",
    "optimizer = optim.Adam(model.parameters(), lr=0.001)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ccbd6e4-1d24-402a-aa03-a53959206234",
   "metadata": {},
   "source": [
    "# Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a9504cd5-8208-4b03-952d-2eaa1618cdad",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_accuracy(model, dataloader):\n",
    "    model.eval()\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    with torch.no_grad():\n",
    "        for data, target in dataloader:\n",
    "            data = data.to(device)\n",
    "            target = target.float().to(device).view(-1, 1)\n",
    "            output = model(data)\n",
    "            preds = (output > 0.0).float()\n",
    "            correct += (preds == target).sum().item()\n",
    "            total += target.size(0)\n",
    "    return 100 * correct / total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "729d3870-2227-4df0-a087-b3dabf3dd7f1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0/30, Loss: 0.1435, Training Accuracy: 97.30%, Test Accuracy: 74.68%\n",
      "Epoch 1/30, Loss: 0.0826, Training Accuracy: 85.62%, Test Accuracy: 84.46%\n",
      "Epoch 2/30, Loss: 0.0730, Training Accuracy: 70.17%, Test Accuracy: 77.88%\n",
      "Epoch 3/30, Loss: 0.0567, Training Accuracy: 90.09%, Test Accuracy: 89.26%\n",
      "Epoch 4/30, Loss: 0.0591, Training Accuracy: 99.04%, Test Accuracy: 74.84%\n",
      "Epoch 5/30, Loss: 0.0471, Training Accuracy: 97.34%, Test Accuracy: 87.82%\n",
      "Epoch 6/30, Loss: 0.0384, Training Accuracy: 96.40%, Test Accuracy: 66.99%\n",
      "Epoch 7/30, Loss: 0.0570, Training Accuracy: 91.45%, Test Accuracy: 90.22%\n",
      "Epoch 8/30, Loss: 0.0319, Training Accuracy: 98.64%, Test Accuracy: 75.96%\n",
      "Epoch 9/30, Loss: 0.0421, Training Accuracy: 99.56%, Test Accuracy: 79.81%\n",
      "Epoch 10/30, Loss: 0.0244, Training Accuracy: 99.64%, Test Accuracy: 76.12%\n",
      "Epoch 11/30, Loss: 0.0198, Training Accuracy: 99.10%, Test Accuracy: 83.97%\n",
      "Epoch 12/30, Loss: 0.0162, Training Accuracy: 95.53%, Test Accuracy: 85.26%\n",
      "Epoch 13/30, Loss: 0.0243, Training Accuracy: 99.65%, Test Accuracy: 78.69%\n",
      "Epoch 14/30, Loss: 0.0206, Training Accuracy: 99.37%, Test Accuracy: 79.97%\n",
      "Epoch 15/30, Loss: 0.0193, Training Accuracy: 99.67%, Test Accuracy: 81.25%\n",
      "Epoch 16/30, Loss: 0.0133, Training Accuracy: 99.92%, Test Accuracy: 80.45%\n",
      "Epoch 17/30, Loss: 0.0084, Training Accuracy: 99.98%, Test Accuracy: 74.84%\n",
      "Epoch 18/30, Loss: 0.0140, Training Accuracy: 96.97%, Test Accuracy: 68.75%\n",
      "Epoch 19/30, Loss: 0.0212, Training Accuracy: 97.32%, Test Accuracy: 66.67%\n",
      "Epoch 20/30, Loss: 0.0074, Training Accuracy: 99.98%, Test Accuracy: 75.64%\n",
      "Epoch 21/30, Loss: 0.0090, Training Accuracy: 97.83%, Test Accuracy: 69.87%\n",
      "Epoch 22/30, Loss: 0.0396, Training Accuracy: 99.85%, Test Accuracy: 75.80%\n",
      "Epoch 23/30, Loss: 0.0057, Training Accuracy: 100.00%, Test Accuracy: 79.81%\n",
      "Epoch 24/30, Loss: 0.0013, Training Accuracy: 99.98%, Test Accuracy: 77.88%\n",
      "Epoch 25/30, Loss: 0.0018, Training Accuracy: 100.00%, Test Accuracy: 78.37%\n",
      "Epoch 26/30, Loss: 0.0124, Training Accuracy: 97.30%, Test Accuracy: 81.57%\n",
      "Epoch 27/30, Loss: 0.0374, Training Accuracy: 99.79%, Test Accuracy: 74.68%\n",
      "Epoch 28/30, Loss: 0.0113, Training Accuracy: 99.96%, Test Accuracy: 79.97%\n",
      "Epoch 29/30, Loss: 0.0046, Training Accuracy: 99.87%, Test Accuracy: 81.73%\n"
     ]
    }
   ],
   "source": [
    "epochs = 30\n",
    "train_losses = []\n",
    "test_accuracies = []\n",
    "\n",
    "for epoch in range(epochs):\n",
    "    model.train()\n",
    "    running_loss = 0.0\n",
    "    for inputs, labels in dataloaders['train']:\n",
    "        inputs = inputs.to(device)\n",
    "        labels = labels.float().to(device).view(-1, 1)\n",
    "        optimizer.zero_grad()\n",
    "        output = model(inputs)\n",
    "        loss = criterion(output, labels)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        running_loss += loss.item()\n",
    "    train_loss = running_loss / len(dataloaders['train'])\n",
    "    train_losses.append(train_loss)\n",
    "    \n",
    "    train_accuracy = compute_accuracy(model, dataloaders['train'])\n",
    "    test_accuracy = compute_accuracy(model, dataloaders['test'])\n",
    "    test_accuracies.append(test_accuracy)\n",
    "    \n",
    "    print(f'Epoch {epoch}/{epochs}, Loss: {train_loss:.4f}, Training Accuracy: {train_accuracy:.2f}%, Test Accuracy: {test_accuracy:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0c4409b-b502-494e-b585-cfca70eff3fe",
   "metadata": {},
   "source": [
    "# Result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d0d2cbb8-2bde-4384-9922-125180da2875",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting\n",
    "fig, ax1 = plt.subplots()\n",
    "\n",
    "color = 'tab:red'\n",
    "ax1.set_xlabel('Epoch')\n",
    "ax1.set_ylabel('Training Loss', color=color)\n",
    "ax1.plot(range(1, epochs+1), train_losses, color=color)\n",
    "ax1.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "color = 'tab:blue'\n",
    "ax2.set_ylabel('Test Accuracy', color=color)\n",
    "ax2.plot(range(1, epochs+1), test_accuracies, color=color)\n",
    "ax2.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "15b56504-5c18-4b7d-b746-6ccd6cb4a12e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "prac",
   "language": "python",
   "name": "prac"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
