dog_code.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import wget\n",
    "import tarfile\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torchvision.models as models\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader, random_split\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Download image.tar file\n",
    "if \"images.tar\" not in os.listdir() :\n",
    "    wget.download('http://vision.stanford.edu/aditya86/ImageNetDogs/images.tar')\n",
    "    \n",
    "# Extract images from tar file\n",
    "current_dir = os.getcwd()\n",
    "tar_file = os.path.join(current_dir, \"images.tar\")\n",
    "\n",
    "with tarfile.open(tar_file, 'r') as tar:\n",
    "    tar.extractall(current_dir)\n",
    "    \n",
    "# Define the transform to convert images to tensors\n",
    "transform = transforms.Compose([transforms.Resize((224, 224)), transforms.ToTensor()])\n",
    "\n",
    "# Convert images into tensors and assign class labels\n",
    "full_dataset = datasets.ImageFolder(root = \"Images\", transform=transform)\n",
    "full_dataset.classes = [image.split(\"-\")[1] for image in full_dataset.classes]\n",
    "\n",
    "# Define the split ratio\n",
    "split_ratio = 0.75\n",
    "\n",
    "# Split the dataset\n",
    "train_size = int(split_ratio * len(full_dataset))\n",
    "test_size = len(full_dataset) - train_size\n",
    "train_dataset, test_dataset = random_split(full_dataset, [train_size, test_size])\n",
    "\n",
    "# Create dataloaders for train and test sets\n",
    "train_loader = DataLoader(train_dataset, batch_size=120, shuffle=True)\n",
    "test_loader = DataLoader(test_dataset, batch_size=120, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/dlongert/miniconda3/lib/python3.11/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",
      "/home/dlongert/miniconda3/lib/python3.11/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=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "resnet50_model = models.resnet50(weights=True)\n",
    "for param in resnet50_model.parameters():\n",
    "    param.requires_grad = False                   \n",
    "num_features = resnet50_model.fc.in_features\n",
    "resnet50_model.fc = nn.Linear(num_features, 120)\n",
    "\n",
    "vgg16_model = models.vgg16(weights=True)\n",
    "for param in vgg16_model.parameters():\n",
    "    param.requires_grad = False                     \n",
    "num_features = vgg16_model.classifier[6].in_features\n",
    "vgg16_model.classifier[6] = nn.Linear(num_features, 120)\n",
    "\n",
    "vit_model = models.vit_b_16(weights=models.ViT_B_16_Weights.DEFAULT)\n",
    "for param in vit_model.parameters():\n",
    "    param.requires_grad = False  \n",
    "vit_model.heads[0] = nn.Linear(768, 120, bias = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_function = nn.CrossEntropyLoss()\n",
    "\n",
    "cuda = torch.cuda.is_available()\n",
    "device = torch.device(\"cuda\" if cuda else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_model(model, num_epochs, optimizer):\n",
    "    train_loss_list = []\n",
    "    valid_loss_list = []\n",
    "    train_acc_list = []\n",
    "    valid_acc_list = []\n",
    "\n",
    "    model = model.to(device)\n",
    "    model = model.cuda()\n",
    "\n",
    "    for epoch in range(num_epochs):\n",
    "        model.train()\n",
    "        train_loss = 0.0\n",
    "        correct_train = 0\n",
    "        total_train = 0\n",
    "\n",
    "        # Training loop\n",
    "        for data, target in train_loader:\n",
    "            data = data.to(device)\n",
    "            target = target.to(device)\n",
    "            \n",
    "            output = model(data)\n",
    "            optimizer.zero_grad()\n",
    "            loss = loss_function(output, target)\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "            train_loss += loss.item() * data.size(0)\n",
    "            _, predicted = torch.max(output, 1)\n",
    "            correct_train += (predicted == target).sum().item()\n",
    "            total_train += target.size(0)\n",
    "\n",
    "        train_loss = train_loss / total_train\n",
    "        train_accuracy = correct_train / total_train\n",
    "        train_loss_list.append(train_loss)\n",
    "        train_acc_list.append(train_accuracy)\n",
    "\n",
    "        # Validation loop\n",
    "        model.eval()\n",
    "        valid_loss = 0.0\n",
    "        correct_valid = 0\n",
    "        total_valid = 0\n",
    "\n",
    "        with torch.no_grad():\n",
    "            for data, target in test_loader:\n",
    "                data = data.to(device)\n",
    "                target = target.to(device)\n",
    "                \n",
    "                output = model(data)\n",
    "                loss = loss_function(output, target)\n",
    "                valid_loss += loss.item() * data.size(0)\n",
    "                _, predicted = torch.max(output, 1)\n",
    "                correct_valid += (predicted == target).sum().item()\n",
    "                total_valid += target.size(0)\n",
    "\n",
    "        valid_loss = valid_loss / total_valid\n",
    "        valid_accuracy = correct_valid / total_valid\n",
    "        valid_loss_list.append(valid_loss)\n",
    "        valid_acc_list.append(valid_accuracy)\n",
    "\n",
    "        # Print epoch summary\n",
    "        print(f'Epoch {epoch + 1} \\t Training Accuracy: {train_accuracy*100:.3f}% \\t Validation Accuracy: {valid_accuracy*100:.3f}%')\n",
    "        print(f'\\t\\t Training Loss: {train_loss:.3f} \\t Validation Loss: {valid_loss:.3f}')\n",
    "        print('-' * 100)\n",
    "\n",
    "    return train_loss_list, valid_loss_list, train_acc_list, valid_acc_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 \t Training Accuracy: 32.368% \t Validation Accuracy: 60.972%\n",
      "\t\t Training Loss: 4.047 \t Validation Loss: 3.296\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 2 \t Training Accuracy: 68.818% \t Validation Accuracy: 74.363%\n",
      "\t\t Training Loss: 2.743 \t Validation Loss: 2.264\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 3 \t Training Accuracy: 77.570% \t Validation Accuracy: 77.940%\n",
      "\t\t Training Loss: 1.926 \t Validation Loss: 1.666\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 4 \t Training Accuracy: 80.862% \t Validation Accuracy: 80.428%\n",
      "\t\t Training Loss: 1.460 \t Validation Loss: 1.331\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 5 \t Training Accuracy: 82.410% \t Validation Accuracy: 81.380%\n",
      "\t\t Training Loss: 1.188 \t Validation Loss: 1.126\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 6 \t Training Accuracy: 83.635% \t Validation Accuracy: 81.652%\n",
      "\t\t Training Loss: 1.014 \t Validation Loss: 1.005\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 7 \t Training Accuracy: 84.282% \t Validation Accuracy: 82.391%\n",
      "\t\t Training Loss: 0.900 \t Validation Loss: 0.918\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 8 \t Training Accuracy: 85.196% \t Validation Accuracy: 82.721%\n",
      "\t\t Training Loss: 0.815 \t Validation Loss: 0.843\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 9 \t Training Accuracy: 85.617% \t Validation Accuracy: 83.479%\n",
      "\t\t Training Loss: 0.748 \t Validation Loss: 0.795\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 10 \t Training Accuracy: 86.032% \t Validation Accuracy: 83.343%\n",
      "\t\t Training Loss: 0.699 \t Validation Loss: 0.756\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 11 \t Training Accuracy: 86.725% \t Validation Accuracy: 83.401%\n",
      "\t\t Training Loss: 0.657 \t Validation Loss: 0.725\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 12 \t Training Accuracy: 86.731% \t Validation Accuracy: 83.537%\n",
      "\t\t Training Loss: 0.621 \t Validation Loss: 0.700\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 13 \t Training Accuracy: 87.302% \t Validation Accuracy: 83.440%\n",
      "\t\t Training Loss: 0.591 \t Validation Loss: 0.681\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 14 \t Training Accuracy: 87.457% \t Validation Accuracy: 83.460%\n",
      "\t\t Training Loss: 0.569 \t Validation Loss: 0.663\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 15 \t Training Accuracy: 87.885% \t Validation Accuracy: 83.790%\n",
      "\t\t Training Loss: 0.543 \t Validation Loss: 0.645\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 16 \t Training Accuracy: 88.293% \t Validation Accuracy: 83.440%\n",
      "\t\t Training Loss: 0.523 \t Validation Loss: 0.639\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 17 \t Training Accuracy: 88.636% \t Validation Accuracy: 84.043%\n",
      "\t\t Training Loss: 0.504 \t Validation Loss: 0.619\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 18 \t Training Accuracy: 88.772% \t Validation Accuracy: 83.829%\n",
      "\t\t Training Loss: 0.488 \t Validation Loss: 0.618\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 19 \t Training Accuracy: 89.083% \t Validation Accuracy: 84.218%\n",
      "\t\t Training Loss: 0.473 \t Validation Loss: 0.602\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 20 \t Training Accuracy: 89.258% \t Validation Accuracy: 83.984%\n",
      "\t\t Training Loss: 0.460 \t Validation Loss: 0.598\n",
      "----------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "optimizer_resnet50 = torch.optim.Adam(params = resnet50_model.parameters(), lr = 0.0001)\n",
    "train_loss_resnet50, test_loss_resnet50, train_accs_resnet50, test_accs_resnet50 = train_model(resnet50_model, 20, optimizer_resnet50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(range(1, 20 + 1), train_loss_resnet50, label='Training Loss')\n",
    "plt.plot(range(1, 20 + 1), test_loss_resnet50, label='Test Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss vs Epochs')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(range(1, 20 + 1), train_accs_resnet50, label='Training Accuracy')\n",
    "plt.plot(range(1, 20 + 1), test_accs_resnet50, label='Test Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Accuracy vs Epochs')\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 \t Training Accuracy: 21.140% \t Validation Accuracy: 48.980%\n",
      "\t\t Training Loss: 3.676 \t Validation Loss: 2.806\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 2 \t Training Accuracy: 46.090% \t Validation Accuracy: 56.443%\n",
      "\t\t Training Loss: 2.425 \t Validation Loss: 2.158\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 3 \t Training Accuracy: 54.214% \t Validation Accuracy: 58.484%\n",
      "\t\t Training Loss: 1.959 \t Validation Loss: 1.870\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 4 \t Training Accuracy: 57.279% \t Validation Accuracy: 59.650%\n",
      "\t\t Training Loss: 1.736 \t Validation Loss: 1.710\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 5 \t Training Accuracy: 59.417% \t Validation Accuracy: 61.341%\n",
      "\t\t Training Loss: 1.600 \t Validation Loss: 1.602\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 6 \t Training Accuracy: 62.041% \t Validation Accuracy: 61.905%\n",
      "\t\t Training Loss: 1.483 \t Validation Loss: 1.528\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 7 \t Training Accuracy: 62.883% \t Validation Accuracy: 62.274%\n",
      "\t\t Training Loss: 1.405 \t Validation Loss: 1.476\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 8 \t Training Accuracy: 64.101% \t Validation Accuracy: 62.449%\n",
      "\t\t Training Loss: 1.346 \t Validation Loss: 1.434\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 9 \t Training Accuracy: 64.911% \t Validation Accuracy: 63.013%\n",
      "\t\t Training Loss: 1.302 \t Validation Loss: 1.402\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 10 \t Training Accuracy: 66.129% \t Validation Accuracy: 63.440%\n",
      "\t\t Training Loss: 1.254 \t Validation Loss: 1.372\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 11 \t Training Accuracy: 66.855% \t Validation Accuracy: 63.460%\n",
      "\t\t Training Loss: 1.214 \t Validation Loss: 1.351\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 12 \t Training Accuracy: 67.068% \t Validation Accuracy: 63.868%\n",
      "\t\t Training Loss: 1.181 \t Validation Loss: 1.333\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 13 \t Training Accuracy: 68.092% \t Validation Accuracy: 64.334%\n",
      "\t\t Training Loss: 1.155 \t Validation Loss: 1.315\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 14 \t Training Accuracy: 68.526% \t Validation Accuracy: 64.179%\n",
      "\t\t Training Loss: 1.127 \t Validation Loss: 1.305\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 15 \t Training Accuracy: 69.291% \t Validation Accuracy: 64.315%\n",
      "\t\t Training Loss: 1.099 \t Validation Loss: 1.292\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 16 \t Training Accuracy: 69.906% \t Validation Accuracy: 64.393%\n",
      "\t\t Training Loss: 1.070 \t Validation Loss: 1.282\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 17 \t Training Accuracy: 70.418% \t Validation Accuracy: 64.723%\n",
      "\t\t Training Loss: 1.054 \t Validation Loss: 1.271\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 18 \t Training Accuracy: 71.267% \t Validation Accuracy: 64.898%\n",
      "\t\t Training Loss: 1.028 \t Validation Loss: 1.265\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 19 \t Training Accuracy: 71.519% \t Validation Accuracy: 64.665%\n",
      "\t\t Training Loss: 1.004 \t Validation Loss: 1.259\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 20 \t Training Accuracy: 72.277% \t Validation Accuracy: 65.053%\n",
      "\t\t Training Loss: 0.994 \t Validation Loss: 1.248\n",
      "----------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "optimizer_vgg16 = torch.optim.Adam(params = vgg16_model.parameters(), lr = 0.0001)\n",
    "train_loss_vgg16, test_loss_vgg16, train_accs_vgg16, test_accs_vgg16 = train_model(vgg16_model, 20, optimizer_vgg16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(range(1, 20 + 1), train_loss_vgg16, label='Training Loss')\n",
    "plt.plot(range(1, 20 + 1), test_loss_vgg16, label='Test Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss vs Epochs')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(range(1, 20 + 1), train_accs_vgg16, label='Training Accuracy')\n",
    "plt.plot(range(1, 20 + 1), test_accs_vgg16, label='Test Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Accuracy vs Epochs')\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 \t Training Accuracy: 35.407% \t Validation Accuracy: 74.947%\n",
      "\t\t Training Loss: 4.172 \t Validation Loss: 3.522\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 2 \t Training Accuracy: 81.503% \t Validation Accuracy: 85.928%\n",
      "\t\t Training Loss: 2.945 \t Validation Loss: 2.404\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 3 \t Training Accuracy: 86.693% \t Validation Accuracy: 88.222%\n",
      "\t\t Training Loss: 1.958 \t Validation Loss: 1.582\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 4 \t Training Accuracy: 88.390% \t Validation Accuracy: 89.640%\n",
      "\t\t Training Loss: 1.299 \t Validation Loss: 1.085\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 5 \t Training Accuracy: 89.602% \t Validation Accuracy: 89.951%\n",
      "\t\t Training Loss: 0.923 \t Validation Loss: 0.811\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 6 \t Training Accuracy: 90.146% \t Validation Accuracy: 90.379%\n",
      "\t\t Training Loss: 0.713 \t Validation Loss: 0.657\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 7 \t Training Accuracy: 90.677% \t Validation Accuracy: 90.748%\n",
      "\t\t Training Loss: 0.590 \t Validation Loss: 0.562\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 8 \t Training Accuracy: 90.820% \t Validation Accuracy: 90.748%\n",
      "\t\t Training Loss: 0.511 \t Validation Loss: 0.500\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 9 \t Training Accuracy: 91.228% \t Validation Accuracy: 90.943%\n",
      "\t\t Training Loss: 0.456 \t Validation Loss: 0.456\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 10 \t Training Accuracy: 91.474% \t Validation Accuracy: 91.137%\n",
      "\t\t Training Loss: 0.416 \t Validation Loss: 0.425\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 11 \t Training Accuracy: 91.629% \t Validation Accuracy: 91.137%\n",
      "\t\t Training Loss: 0.385 \t Validation Loss: 0.400\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 12 \t Training Accuracy: 91.953% \t Validation Accuracy: 91.176%\n",
      "\t\t Training Loss: 0.361 \t Validation Loss: 0.382\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 13 \t Training Accuracy: 92.051% \t Validation Accuracy: 90.982%\n",
      "\t\t Training Loss: 0.341 \t Validation Loss: 0.367\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 14 \t Training Accuracy: 92.141% \t Validation Accuracy: 91.118%\n",
      "\t\t Training Loss: 0.324 \t Validation Loss: 0.355\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 15 \t Training Accuracy: 92.349% \t Validation Accuracy: 91.176%\n",
      "\t\t Training Loss: 0.310 \t Validation Loss: 0.345\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 16 \t Training Accuracy: 92.634% \t Validation Accuracy: 91.137%\n",
      "\t\t Training Loss: 0.297 \t Validation Loss: 0.337\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 17 \t Training Accuracy: 92.783% \t Validation Accuracy: 91.118%\n",
      "\t\t Training Loss: 0.285 \t Validation Loss: 0.330\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 18 \t Training Accuracy: 93.016% \t Validation Accuracy: 91.118%\n",
      "\t\t Training Loss: 0.275 \t Validation Loss: 0.324\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 19 \t Training Accuracy: 93.230% \t Validation Accuracy: 91.079%\n",
      "\t\t Training Loss: 0.266 \t Validation Loss: 0.319\n",
      "----------------------------------------------------------------------------------------------------\n",
      "Epoch 20 \t Training Accuracy: 93.340% \t Validation Accuracy: 91.137%\n",
      "\t\t Training Loss: 0.257 \t Validation Loss: 0.314\n",
      "----------------------------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "optimizer_vit = torch.optim.Adam(params = vit_model.parameters(), lr = 0.0001)\n",
    "train_loss_vit, test_loss_vit, train_accs_vit, test_accs_vit = train_model(vit_model, 20, optimizer_vit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(range(1, 20 + 1), train_loss_vit, label='Training Loss')\n",
    "plt.plot(range(1, 20 + 1), test_loss_vit, label='Test Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss vs Epochs')\n",
    "plt.legend()\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(range(1, 20 + 1), train_accs_vit, label='Training Accuracy')\n",
    "plt.plot(range(1, 20 + 1), test_accs_vit, label='Test Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Accuracy vs Epochs')\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}