cnn_attack_defense_all_digit.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ericyoc/adversarial-defense-cnn/blob/main/cnn_attack_defense_all_digit.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "7b23530d-c160-4139-8b3e-52b9bf469164",
      "metadata": {
        "id": "7b23530d-c160-4139-8b3e-52b9bf469164"
      },
      "source": [
        "Convolutional Neural Network (CNN) Model Using either MNIST or EMNIST Datasets with Protection Using Adverarial Training Against Compounded Attacks"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "QCsgCc7B6OTd",
      "metadata": {
        "id": "QCsgCc7B6OTd"
      },
      "source": [
        "Make sure to set runtime to GPU (e.g., T4 GPU - Google Colabs) instead of CPU"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a1iaGT-rFUa6",
        "outputId": "0f73b9a2-33d4-42d1-b145-d26cc3093776"
      },
      "id": "a1iaGT-rFUa6",
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "id": "OB7Knnyr6R4e",
      "metadata": {
        "id": "OB7Knnyr6R4e"
      },
      "source": [
        "Assumes that datasets are on Google Drive and that Google Drive will be mounted with default directory structure"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "9134f1dd-7776-4452-8fc4-47d6afe18a37",
      "metadata": {
        "id": "9134f1dd-7776-4452-8fc4-47d6afe18a37"
      },
      "source": [
        "Compounded attack: fgsm_cw attack, fgsm_pgd attack, cw_pgd_attack, Dataset: MNIST, EMNIST 'Digits'"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "8352c686-f25b-4c25-9a0a-44bc7555c846",
      "metadata": {
        "id": "8352c686-f25b-4c25-9a0a-44bc7555c846"
      },
      "source": [
        "ATTENTION: MUST INSTALL AT A MINIMUM FOLLOWING FOR PYTHON CODE TO WORK CORRECTLY"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "955c572f-5f61-4d61-b126-8f2c3571d366",
      "metadata": {
        "id": "955c572f-5f61-4d61-b126-8f2c3571d366"
      },
      "source": [
        "Python version: 3.8.18\n",
        "torch: 2.2.1\n",
        "torchvision: 0.17.1\n",
        "torchattacks: 3.5.1\n",
        "numpy: 1.23.5\n",
        "tabulate: 0.9.0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e53e8627-4a37-4ed0-8c7c-4bc8cf4e54a6",
      "metadata": {
        "id": "e53e8627-4a37-4ed0-8c7c-4bc8cf4e54a6"
      },
      "outputs": [],
      "source": [
        "import warnings\n",
        "# Ignore all warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "07ZDiyWILm2P",
      "metadata": {
        "id": "07ZDiyWILm2P"
      },
      "outputs": [],
      "source": [
        "#!pip install torchattacks\n",
        "import torchattacks"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ArbA2OMn0i6m",
      "metadata": {
        "id": "ArbA2OMn0i6m"
      },
      "outputs": [],
      "source": [
        "#Pick a compounded adversarial attack from the following: fgsm_cw_attack, fgsm_pgd_attack, cw_pgd_attack\n",
        "global compounded_attack_name\n",
        "\n",
        "compounded_attack_name = \"fgsm_pgd_attack\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "EwsUoqDp1dyj",
      "metadata": {
        "id": "EwsUoqDp1dyj"
      },
      "outputs": [],
      "source": [
        "# Set the dataset_name\n",
        "global dataset_name\n",
        "dataset_name = 'EMNIST'  # 'MNIST' or 'EMNIST' or 'SVHN' or 'USPS' or 'Semeion'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fCVtnLGaMsZJ",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fCVtnLGaMsZJ",
        "outputId": "c4c63407-8251-4ac7-8db7-311f4dbfde2f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "\n",
        "# Mount Google Drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "# Access the mounted drive\n",
        "drive_path = '/content/drive/My Drive/'\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ekL_geF31MQ0",
      "metadata": {
        "id": "ekL_geF31MQ0"
      },
      "outputs": [],
      "source": [
        "global num_epochs\n",
        "num_epochs = 10"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "1b7274e6-5dd7-48cb-b4bf-59d121320669",
      "metadata": {
        "id": "1b7274e6-5dd7-48cb-b4bf-59d121320669"
      },
      "outputs": [],
      "source": [
        "import sys\n",
        "import pkg_resources\n",
        "\n",
        "def get_package_versions(package_list):\n",
        "    installed_packages = pkg_resources.working_set\n",
        "    package_versions = {}\n",
        "\n",
        "    for package_name in package_list:\n",
        "        package = installed_packages.by_key.get(package_name)\n",
        "        if package:\n",
        "            package_versions[package_name] = package.version\n",
        "        else:\n",
        "            package_versions[package_name] = \"Not installed\"\n",
        "\n",
        "    return package_versions\n",
        "\n",
        "# Specify the list of packages you want to check\n",
        "packages_to_check = [\"torch\", \"torchvision\", \"torchattacks\", \"torchvision\", \"numpy\", \"tabulate\"]\n",
        "\n",
        "# Call the function to get package versions\n",
        "versions = get_package_versions(packages_to_check)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "88bb6b46-5683-474e-8162-03b61b15b364",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "88bb6b46-5683-474e-8162-03b61b15b364",
        "outputId": "794d14d5-4bd0-4915-bd94-9d6d10206300"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Python version: 3.10.12\n",
            "torch: 2.2.1+cu121\n",
            "torchvision: 0.17.1+cu121\n",
            "torchattacks: 3.5.1\n",
            "numpy: 1.25.2\n",
            "tabulate: 0.9.0\n"
          ]
        }
      ],
      "source": [
        "# Get the Python version\n",
        "python_version = sys.version.split()[0]\n",
        "\n",
        "# Call the function to get package versions\n",
        "versions = get_package_versions(packages_to_check)\n",
        "\n",
        "# Print the Python version\n",
        "print(f\"Python version: {python_version}\")\n",
        "\n",
        "# Print the package versions\n",
        "for package_name, version in versions.items():\n",
        "    print(f\"{package_name}: {version}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e4bc0e60-8668-4269-ad82-68be0fd52ef2",
      "metadata": {
        "id": "e4bc0e60-8668-4269-ad82-68be0fd52ef2"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "import torch.optim as optim\n",
        "from torchvision import datasets, transforms\n",
        "from torch.utils.data import DataLoader"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "b1e15569-ff43-4b6a-a18e-455147688571",
      "metadata": {
        "id": "b1e15569-ff43-4b6a-a18e-455147688571"
      },
      "outputs": [],
      "source": [
        "# Define model\n",
        "class CNN(nn.Module):\n",
        "    def __init__(self):\n",
        "        super().__init__()\n",
        "        self.conv1 = nn.Conv2d(1, 4, 5)\n",
        "        self.conv2 = nn.Conv2d(4, 10, 5)\n",
        "        self.fc1 = nn.Linear(160, 80)\n",
        "        self.fc2 = nn.Linear(80, 10)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = F.relu(self.conv1(x))\n",
        "        x = F.max_pool2d(x, 2)\n",
        "        x = F.relu(self.conv2(x))\n",
        "        x = F.max_pool2d(x, 2)\n",
        "        x = x.view(-1, 160)\n",
        "        x = F.relu(self.fc1(x))\n",
        "        x = self.fc2(x)\n",
        "        return F.log_softmax(x, dim=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "8787f3f3-e37c-4012-8cb7-9737713b6ce1",
      "metadata": {
        "id": "8787f3f3-e37c-4012-8cb7-9737713b6ce1"
      },
      "outputs": [],
      "source": [
        "device = 'cuda' if torch.cuda.is_available() else 'cpu'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "6c2d9762-e9f9-4b70-9888-983828929995",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6c2d9762-e9f9-4b70-9888-983828929995",
        "outputId": "85c20288-2eab-4d8d-d57b-4782db7cdf80"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model Architecture:\n",
            "CNN(\n",
            "  (conv1): Conv2d(1, 4, kernel_size=(5, 5), stride=(1, 1))\n",
            "  (conv2): Conv2d(4, 10, kernel_size=(5, 5), stride=(1, 1))\n",
            "  (fc1): Linear(in_features=160, out_features=80, bias=True)\n",
            "  (fc2): Linear(in_features=80, out_features=10, bias=True)\n",
            ")\n"
          ]
        }
      ],
      "source": [
        "model = CNN().to(device)\n",
        "\n",
        "print(\"Model Architecture:\")\n",
        "print(model)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d6ea1172-9b8b-48bf-ae82-de28b755a161",
      "metadata": {
        "id": "d6ea1172-9b8b-48bf-ae82-de28b755a161"
      },
      "outputs": [],
      "source": [
        "optimizer = optim.Adam(model.parameters())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d87f07a6-d229-446f-a951-3536f0e8604a",
      "metadata": {
        "id": "d87f07a6-d229-446f-a951-3536f0e8604a"
      },
      "outputs": [],
      "source": [
        "loss_fn = nn.NLLLoss()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "H09bz5fwDFj7",
      "metadata": {
        "id": "H09bz5fwDFj7"
      },
      "outputs": [],
      "source": [
        "import gzip\n",
        "import numpy as np\n",
        "from torchvision import transforms\n",
        "\n",
        "def load_emnist_digits(data_dir, train=True, transform=None, filtered=range(10)):\n",
        "    if train:\n",
        "        image_file = 'emnist-digits-train-images-idx3-ubyte.gz'\n",
        "        label_file = 'emnist-digits-train-labels-idx1-ubyte.gz'\n",
        "    else:\n",
        "        image_file = 'emnist-digits-test-images-idx3-ubyte.gz'\n",
        "        label_file = 'emnist-digits-test-labels-idx1-ubyte.gz'\n",
        "\n",
        "    image_path = os.path.join(data_dir, image_file)\n",
        "    label_path = os.path.join(data_dir, label_file)\n",
        "\n",
        "    with gzip.open(image_path, 'rb') as f:\n",
        "        images = np.frombuffer(f.read(), np.uint8, offset=16).reshape(-1, 28, 28)\n",
        "    with gzip.open(label_path, 'rb') as f:\n",
        "        labels = np.frombuffer(f.read(), np.uint8, offset=8)\n",
        "\n",
        "    dataset = EMNISTDataset(images, labels, transform)\n",
        "\n",
        "    if filtered is not None:\n",
        "        digits_to_keep = list(filtered)  # Convert the range object to a list\n",
        "        mask = np.isin(dataset.labels, digits_to_keep)\n",
        "        dataset.images = dataset.images[mask]\n",
        "        dataset.labels = dataset.labels[mask]\n",
        "\n",
        "    return dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "Wvn2OTedITEr",
      "metadata": {
        "id": "Wvn2OTedITEr"
      },
      "outputs": [],
      "source": [
        "def filtered_dataset_EMNIST(train_set, test_set, filtered_digits=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]):\n",
        "    if filtered_digits is not None:\n",
        "        train_mask = np.isin(train_set.labels, filtered_digits)\n",
        "        test_mask = np.isin(test_set.labels, filtered_digits)\n",
        "\n",
        "        train_images = train_set.images[train_mask]\n",
        "        train_labels = train_set.labels[train_mask]\n",
        "        test_images = test_set.images[test_mask]\n",
        "        test_labels = test_set.labels[test_mask]\n",
        "\n",
        "        filtered_train_set = EMNISTDataset(train_images, train_labels, transform=train_set.transform)\n",
        "        filtered_test_set = EMNISTDataset(test_images, test_labels, transform=test_set.transform)\n",
        "    else:\n",
        "        filtered_train_set = train_set\n",
        "        filtered_test_set = test_set\n",
        "\n",
        "    return filtered_train_set, filtered_test_set"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "r2Vbxm040dSn",
      "metadata": {
        "id": "r2Vbxm040dSn"
      },
      "outputs": [],
      "source": [
        "import gzip\n",
        "import numpy as np\n",
        "import torch\n",
        "from torch.utils.data import Dataset\n",
        "from torchvision.transforms import ToTensor\n",
        "\n",
        "class EMNISTDigits(Dataset):\n",
        "    def __init__(self, root, train=True, transform=None, target_transform=None):\n",
        "        self.root = root\n",
        "        self.train = train\n",
        "        self.transform = transform\n",
        "        self.target_transform = target_transform\n",
        "\n",
        "        prefix = 'train' if self.train else 'test'\n",
        "        image_file = f'{self.root}/emnist-digits-{prefix}-images-idx3-ubyte.gz'\n",
        "        label_file = f'{self.root}/emnist-digits-{prefix}-labels-idx1-ubyte.gz'\n",
        "\n",
        "        with gzip.open(image_file, 'rb') as f:\n",
        "            self.images = np.frombuffer(f.read(), np.uint8, offset=16).reshape(-1, 28, 28)\n",
        "\n",
        "        with gzip.open(label_file, 'rb') as f:\n",
        "            self.labels = np.frombuffer(f.read(), np.uint8, offset=8)\n",
        "\n",
        "    def __getitem__(self, index):\n",
        "        image = self.images[index]\n",
        "        label = self.labels[index]\n",
        "\n",
        "        if self.transform is not None:\n",
        "            image = self.transform(image)\n",
        "\n",
        "        if self.target_transform is not None:\n",
        "            label = self.target_transform(label)\n",
        "\n",
        "        return image, label\n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.images)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "BUkp5z9P1Egk",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BUkp5z9P1Egk",
        "outputId": "9aa3d353-0a4c-4d0c-e7f3-b8352d4b7fca"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "emnist-digits-test-images-idx3-ubyte.gz  emnist-digits-train-images-idx3-ubyte.gz\n",
            "emnist-digits-test-labels-idx1-ubyte.gz  emnist-digits-train-labels-idx1-ubyte.gz\n"
          ]
        }
      ],
      "source": [
        "!ls \"/content/drive/My Drive/data/EMNIST/raw/\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "31916be4-b247-445e-bdb4-f7f651135af8",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "31916be4-b247-445e-bdb4-f7f651135af8",
        "outputId": "be8d1ee9-91c4-4627-933d-471267341730"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/drive/My Drive/data/EMNIST/raw/\n"
          ]
        }
      ],
      "source": [
        "import torchvision.datasets as datasets\n",
        "import torchvision.transforms as transforms\n",
        "import os\n",
        "from PIL import Image\n",
        "from torch.utils.data import Dataset\n",
        "from torchvision.transforms import ToTensor\n",
        "\n",
        "if dataset_name == 'MNIST':\n",
        "    # Load a dataset from Google Drive\n",
        "    data_dir = drive_path + 'data/MNIST/'\n",
        "    print(data_dir)\n",
        "    train_set = datasets.MNIST(data_dir, train=True, download=False, transform=transforms.ToTensor())\n",
        "    test_set = datasets.MNIST(data_dir, train=False, download=False, transform=transforms.ToTensor())\n",
        "elif dataset_name == 'EMNIST':\n",
        "    # Load a dataset from Google Drive\n",
        "    data_dir = drive_path + 'data/EMNIST/raw/'\n",
        "    train_set = EMNISTDigits(data_dir, train=True, transform=ToTensor())\n",
        "    test_set = EMNISTDigits(data_dir, train=False, transform=ToTensor())\n",
        "    print(data_dir)\n",
        "elif dataset_name == 'SVHN':\n",
        "    data_dir = '/content/drive/data/'\n",
        "    train_set = datasets.SVHN(data_dir, split='train', download=True, transform=data_transform)\n",
        "    test_set = datasets.SVHN(data_dir, split='test', download=True, transform=data_transform)\n",
        "elif dataset_name == 'USPS':\n",
        "    data_dir = '/content/drive/data/'\n",
        "    train_set = datasets.USPS(data_dir, train=True, download=True, transform=data_transform)\n",
        "    test_set = datasets.USPS(data_dir, train=False, download=True, transform=data_transform)\n",
        "elif dataset_name == 'Semeion':\n",
        "    data_dir = '/content/drive/data/'\n",
        "    from torch.utils.data import DataLoader, Dataset, random_split\n",
        "    import numpy as np\n",
        "    from torchvision import transforms\n",
        "\n",
        "    class SemeionDataset(Dataset):\n",
        "        def __init__(self, file_path, transform=None):\n",
        "            data = np.loadtxt(file_path)\n",
        "            self.images = data[:, :256].reshape(-1, 16, 16).astype(np.float32)\n",
        "            self.labels = data[:, 256:].astype(np.int64)\n",
        "            self.transform = transform\n",
        "\n",
        "        def __len__(self):\n",
        "            return len(self.images)\n",
        "\n",
        "        def __getitem__(self, idx):\n",
        "            image = self.images[idx]\n",
        "            label = self.labels[idx].argmax()  # Convert one-hot encoded labels to integers\n",
        "            if self.transform:\n",
        "                image = self.transform(image)\n",
        "            return image, label\n",
        "\n",
        "    # Define the data transformation\n",
        "    data_transform = transforms.Compose([\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Resize((28, 28)),  # Resize the images to 28x28 pixels\n",
        "        transforms.Normalize((0.1307,), (0.3081,))  # Normalize the images\n",
        "    ])\n",
        "\n",
        "    # Load the Semeion dataset\n",
        "    dataset = SemeionDataset(data_dir, transform=data_transform)\n",
        "\n",
        "    # Split the dataset into training and test sets\n",
        "    train_size = int(0.8 * len(dataset))\n",
        "    test_size = len(dataset) - train_size\n",
        "    train_set, test_set = random_split(dataset, [train_size, test_size])\n",
        "\n",
        "else:\n",
        "    raise ValueError(\"Invalid dataset name.\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "db1e3f95-e99c-4dae-84ec-a2257034b356",
      "metadata": {
        "id": "db1e3f95-e99c-4dae-84ec-a2257034b356"
      },
      "outputs": [],
      "source": [
        "#from torchvision import datasets, transforms\n",
        "#from torch.utils.data import DataLoader, Subset, random_split\n",
        "from torch.utils.data import Subset\n",
        "import numpy as np\n",
        "\n",
        "def filtered_dataset(train_set, test_set, filtered=0, dataset_name=None):\n",
        "    if filtered == 0:  # digits 0 and 1\n",
        "        if dataset_name == 'SVHN' or dataset_name == 'EMNIST':\n",
        "            train_indexes_0 = torch.where(torch.tensor(train_set.labels) == 0)[0]\n",
        "            train_indexes_1 = torch.where(torch.tensor(train_set.labels) == 1)[0]\n",
        "            test_indexes_0 = torch.where(torch.tensor(test_set.labels) == 0)[0]\n",
        "            test_indexes_1 = torch.where(torch.tensor(test_set.labels) == 1)[0]\n",
        "        elif dataset_name == 'USPS':\n",
        "            train_indexes_0 = torch.where(torch.tensor(train_set.targets) == 0)[0]\n",
        "            train_indexes_1 = torch.where(torch.tensor(train_set.targets) == 1)[0]\n",
        "            test_indexes_0 = torch.where(torch.tensor(test_set.targets) == 0)[0]\n",
        "            test_indexes_1 = torch.where(torch.tensor(test_set.targets) == 1)[0]\n",
        "        else:\n",
        "            try:\n",
        "                train_indexes_0 = torch.where(train_set.targets == 0)[0]\n",
        "                train_indexes_1 = torch.where(train_set.targets == 1)[0]\n",
        "                test_indexes_0 = torch.where(test_set.targets == 0)[0]\n",
        "                test_indexes_1 = torch.where(test_set.targets == 1)[0]\n",
        "            except AttributeError:\n",
        "                train_indexes_0 = torch.where(torch.tensor(train_set.labels) == 0)[0]\n",
        "                train_indexes_1 = torch.where(torch.tensor(train_set.labels) == 1)[0]\n",
        "                test_indexes_0 = torch.where(torch.tensor(test_set.labels) == 0)[0]\n",
        "                test_indexes_1 = torch.where(torch.tensor(test_set.labels) == 1)[0]\n",
        "\n",
        "        num_to_sample_train = min(len(train_indexes_0), len(train_indexes_1))\n",
        "        num_to_sample_test = min(len(test_indexes_0), len(test_indexes_1))\n",
        "\n",
        "        sampled_train_indexes_0 = train_indexes_0[torch.randperm(len(train_indexes_0))[:num_to_sample_train]]\n",
        "        sampled_train_indexes_1 = train_indexes_1[torch.randperm(len(train_indexes_1))[:num_to_sample_train]]\n",
        "        sampled_test_indexes_0 = test_indexes_0[torch.randperm(len(test_indexes_0))[:num_to_sample_test]]\n",
        "        sampled_test_indexes_1 = test_indexes_1[torch.randperm(len(test_indexes_1))[:num_to_sample_test]]\n",
        "\n",
        "        selected_train_indexes = torch.cat([sampled_train_indexes_0, sampled_train_indexes_1])\n",
        "        selected_test_indexes = torch.cat([sampled_test_indexes_0, sampled_test_indexes_1])\n",
        "\n",
        "    elif filtered == 1:  # digits 0 to 9\n",
        "        train_indexes = []\n",
        "        test_indexes = []\n",
        "        for i in range(10):\n",
        "            if dataset_name == 'SVHN' or dataset_name == 'EMNIST':\n",
        "                train_idx = torch.where(torch.tensor(train_set.labels) == i)[0]\n",
        "                test_idx = torch.where(torch.tensor(test_set.labels) == i)[0]\n",
        "            elif dataset_name == 'USPS':\n",
        "                train_idx = torch.where(torch.tensor(train_set.targets) == i)[0]\n",
        "                test_idx = torch.where(torch.tensor(test_set.targets) == i)[0]\n",
        "            elif dataset_name == 'Semeion':\n",
        "                train_idx = torch.tensor([dataset.labels[idx].argmax() for idx in train_set.indices])\n",
        "                test_idx = torch.tensor([dataset.labels[idx].argmax() for idx in test_set.indices])\n",
        "            else:\n",
        "                try:\n",
        "                    train_idx = torch.where(train_set.targets == i)[0]\n",
        "                    test_idx = torch.where(test_set.targets == i)[0]\n",
        "                except AttributeError:\n",
        "                    train_idx = torch.where(torch.tensor(train_set.labels) == i)[0]\n",
        "                    test_idx = torch.where(torch.tensor(test_set.labels) == i)[0]\n",
        "\n",
        "            train_indexes.append(train_idx)\n",
        "            test_indexes.append(test_idx)\n",
        "\n",
        "        min_num_samples_train = min(len(idx) for idx in train_indexes)\n",
        "        min_num_samples_test = min(len(idx) for idx in test_indexes)\n",
        "\n",
        "        sampled_train_indexes = []\n",
        "        sampled_test_indexes = []\n",
        "        for train_idx, test_idx in zip(train_indexes, test_indexes):\n",
        "            sampled_train_indexes.append(train_idx[torch.randperm(len(train_idx))[:min_num_samples_train]])\n",
        "            sampled_test_indexes.append(test_idx[torch.randperm(len(test_idx))[:min_num_samples_test]])\n",
        "\n",
        "        selected_train_indexes = torch.cat(sampled_train_indexes)\n",
        "        selected_test_indexes = torch.cat(sampled_test_indexes)\n",
        "\n",
        "    else:\n",
        "        raise ValueError(\"Invalid value for 'filtered'. Choose either 0 or 1.\")\n",
        "\n",
        "    filtered_train_set = torch.utils.data.Subset(train_set, selected_train_indexes)\n",
        "    filtered_test_set = torch.utils.data.Subset(test_set, selected_test_indexes)\n",
        "\n",
        "    return filtered_train_set, filtered_test_set"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "1fcd9364-6c99-476d-9e15-ee310e2d1dce",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1fcd9364-6c99-476d-9e15-ee310e2d1dce",
        "outputId": "a01e94dd-d686-4905-e187-7262b82432f9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "EMNIST\n"
          ]
        }
      ],
      "source": [
        "global filtered_train_set, filtered_test_set\n",
        "print(dataset_name)\n",
        "#if dataset_name == \"EMNIST\":\n",
        "  #filtered_train_set, filtered_test_set = filtered_dataset_EMNIST(train_set, test_set)\n",
        "#else:\n",
        "filtered_train_set, filtered_test_set = filtered_dataset(train_set, test_set, filtered=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d49fa0ce-279f-486b-9946-5aa2da6cc28b",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d49fa0ce-279f-486b-9946-5aa2da6cc28b",
        "outputId": "6979c8e2-ead2-4277-e3f1-bd55299ea0dd"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3750\n",
            "625\n"
          ]
        }
      ],
      "source": [
        "global filtered_train_loader, filtered_test_loader\n",
        "\n",
        "filtered_train_loader = DataLoader(filtered_train_set, batch_size=64, shuffle=True)\n",
        "filtered_test_loader = DataLoader(filtered_test_set, batch_size=64)\n",
        "\n",
        "print(len(filtered_train_loader))\n",
        "print(len(filtered_test_loader))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "7d774a42-1370-42b3-abbe-3dbe61c54d75",
      "metadata": {
        "id": "7d774a42-1370-42b3-abbe-3dbe61c54d75"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "def print_images(images, labels, preds, n):\n",
        "\n",
        "    fig = plt.figure(figsize=(6, 6))\n",
        "    for i in range(n):\n",
        "        ax = fig.add_subplot(2, 3, i+1)\n",
        "        img = images[i].squeeze()\n",
        "        ax.imshow(img, cmap='gray')\n",
        "        pred = preds[i].argmax()\n",
        "        ax.set_title(f\"{labels[i].item()} -> {pred.item()}\")\n",
        "        ax.axis('off')\n",
        "\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "97f6690b-ad8c-4e0d-b6ca-6a0cb80df1b0",
      "metadata": {
        "id": "97f6690b-ad8c-4e0d-b6ca-6a0cb80df1b0"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "\n",
        "def save_clean_model(model, save_dir):\n",
        "\n",
        "    if not os.path.exists(save_dir):\n",
        "        os.makedirs(save_dir)\n",
        "\n",
        "    model_path = f\"{save_dir}/clean_model.pth\"\n",
        "\n",
        "    torch.save(model.state_dict(), model_path)\n",
        "    #print(f\"Saved clean_model state dict to {model_path}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0vbZvTM3rqeZ",
      "metadata": {
        "id": "0vbZvTM3rqeZ"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plot_accuracy_loss(train_accuracies, train_losses):\n",
        "    epochs = range(1, len(train_accuracies) + 1)\n",
        "\n",
        "    plt.figure(figsize=(12, 6))\n",
        "    plt.subplot(1, 2, 1)\n",
        "    plt.plot(epochs, train_accuracies, 'b', label='Training Accuracy')\n",
        "    plt.title('Training Accuracy')\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.legend()\n",
        "\n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.plot(epochs, train_losses, 'b', label='Training Loss')\n",
        "    plt.title('Training Loss')\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "006d4aa3-14bf-4ee8-9d9a-eaeed40a3933",
      "metadata": {
        "id": "006d4aa3-14bf-4ee8-9d9a-eaeed40a3933"
      },
      "outputs": [],
      "source": [
        "# Train model\n",
        "import numpy as np\n",
        "import torch\n",
        "\n",
        "def train_model(model, train_loader, loss_fn, optimizer, num_epochs, device, batch_size=128, learning_rate=0.001, weight_decay=0.0001):\n",
        "    # Model folder\n",
        "    save_dir = 'clean_models'\n",
        "\n",
        "    train_accuracies = []\n",
        "    train_losses = []\n",
        "\n",
        "    model.to(device)  # Move the model to the specified device\n",
        "\n",
        "    # Adjust learning rate and weight decay\n",
        "    for param_group in optimizer.param_groups:\n",
        "        param_group['lr'] = learning_rate\n",
        "        param_group['weight_decay'] = weight_decay\n",
        "\n",
        "    for epoch in range(num_epochs):\n",
        "        print(f\"Epoch {epoch+1}/{num_epochs}\")\n",
        "\n",
        "        # Training\n",
        "        model.train()\n",
        "        train_loss = 0.0\n",
        "        train_correct = 0\n",
        "        train_total = 0\n",
        "\n",
        "        for i, (imgs, labels) in enumerate(train_loader):\n",
        "            imgs, labels = imgs.to(device), labels.to(device)  # Move input data to the specified device\n",
        "\n",
        "            preds = model(imgs)\n",
        "            loss = loss_fn(preds, labels)\n",
        "\n",
        "            optimizer.zero_grad()\n",
        "            loss.backward()\n",
        "            optimizer.step()\n",
        "\n",
        "            train_loss += loss.item()\n",
        "            _, predicted = torch.max(preds.data, 1)\n",
        "            train_total += labels.size(0)\n",
        "            train_correct += (predicted == labels).sum().item()\n",
        "\n",
        "            if i == 0:\n",
        "                print_images(imgs.cpu(), labels.cpu(), preds.cpu(), n=3)  # Move data back to CPU for printing\n",
        "\n",
        "        train_loss /= len(train_loader)\n",
        "        train_acc = train_correct / train_total\n",
        "\n",
        "        print(f\"Train Loss: {train_loss:.4f}, Train Accuracy: {train_acc:.4f}\")\n",
        "\n",
        "        train_accuracies.append(train_acc)\n",
        "        train_losses.append(train_loss)\n",
        "\n",
        "    # Plot accuracy and loss curves\n",
        "    plot_accuracy_loss(train_accuracies, train_losses)\n",
        "\n",
        "    save_clean_model(model, save_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "84139728-03c9-4c31-bed5-37d8bf23160b",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "84139728-03c9-4c31-bed5-37d8bf23160b",
        "outputId": "3f6f4652-3f45-42ac-c77c-2bca65283544"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.1415, Train Accuracy: 0.9579\n",
            "Epoch 2/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0497, Train Accuracy: 0.9850\n",
            "Epoch 3/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0387, Train Accuracy: 0.9883\n",
            "Epoch 4/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0332, Train Accuracy: 0.9901\n",
            "Epoch 5/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0293, Train Accuracy: 0.9914\n",
            "Epoch 6/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0264, Train Accuracy: 0.9921\n",
            "Epoch 7/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0248, Train Accuracy: 0.9926\n",
            "Epoch 8/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAACzCAYAAABYUm8CAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAUlklEQVR4nO3deWwV5ffH8VO1QEEBSbU02I0iiYigISJFRTSi7ItLhFaI0SJ1YYskEpVgIAhGEzAhRTEKAalgMAKCS+oCKi6AgSCKRpCrLBW9aKGsFpjvH7/ojzvnFIZ2eu9z2/cr8Y/5MJ0+ttN7mN7DeVI8z/MEAAAk1AWJXgAAAKAgAwDgBAoyAAAOoCADAOAACjIAAA6gIAMA4AAKMgAADqAgAwDgAAoyAAAOoCADAOCABluQDx8+LFOnTpW+fftKmzZtJCUlRRYuXJjoZcVIhjWifsyYMUNSUlKkc+fOiV5KjUaPHi0pKSkycODARC8FIXvggQckJSWlxv/27t2b6CVK7969a1xfampqopdXLy5K9ALqSzQalWnTpkl2drZ07dpV1q5dm+glKcmwRoRvz5498txzz0mLFi0SvZQabdq0SRYuXCjNmjVL9FJQD8aMGSO33357TOZ5npSUlEhubq60a9cuQSv7f08//bQUFxfHZEeOHJGSkhK54447ErSq+tVgC3JmZqZUVFRI27ZtZdOmTXL99deHct2dO3dKZmamNG/evM7Xqq81wm2TJk2SHj16yKlTpyQajdbpWj/++KPk5+eH+sTgeZ6MGzdORo0aJR9//HFo14U7CgoKpKCgICb74osv5OjRo1JUVFTr64Z5P/bp00dlb7zxhohIndbosgb7K+umTZtK27ZtQ7/u4sWLJTMzU0pKSmTjxo11ulZ9rRHu+uyzz2T58uUyZ86cUK43a9YsadeunUyaNEm2b98eyjUXL14s27ZtkxkzZoRyPSSHsrIySUlJkcLCwlpfoz7uxzOVlZVJixYtZMiQIaFf2wUNtiDXl8LCQikqKpJly5ZJ9+7dpUuXLvLSSy/JgQMHEr00OO7UqVMyduxYKS4ulmuuuSaUa44dO1b69u0r8+bNk06dOsmNN94or7/+uhw+fLhW16uqqpInn3xSnnrqKf6y2IhUV1fLW2+9JT179pTc3NxaXyfs+/FMf/75p5SXl8vQoUOdfrunLijI56ljx45SWloqFRUVsmTJErn88stl4sSJ0q5dOxk+fLiUl5fL6dOnE71MOOjll1+WX3/9VaZPnx7aNbt16yaLFi2SiooKeeWVV+TUqVPy0EMPSWZmphQXF8tXX311XtebNm2apKWlycSJE0NbI9z34YcfyoEDB+r8q+Cw78czLVu2TE6ePNlgf10tIiJeI7Bx40ZPRLwFCxbUy/UjkYj37LPPerm5uZ6IeDk5Od5zzz3n1BqRWNFo1GvTpo334osv/pfdcsst3tVXXx365/rhhx+8SZMmeRkZGZ6IeJ06dfJeffXVc37cTz/95KWmpnrLly//L8vJyfEGDBgQ+hrhlhEjRnipqaleNBoN/dq1vR/9CgoKvMsuu8yrrq4OfY2uoCAbjh496lVUVMT8F8Tvv//uDRo0yBMRr1WrVvW6RiSXkpISr0OHDt6JEyf+y4IW5Kqqqph78Y8//gj0OXfs2OEVFBR4IuJ17dr1nOf37dvXu+WWW2IyCnLDV1VV5TVv3twbOHBg4PPjcT+eaefOnZ6IeI8//vh5fVyy4VfWhmXLlklmZmbMfzU5efKkrFq1SoYNGyZZWVny/vvvy9ChQ6WsrCyOK4bLfv75Z5k/f76MGzdO9u3bJ5FIRCKRiBw/flyqq6slEonIX3/9VePHv/jiizH34tm68Y8fPy5lZWXSp08fufLKK2XLli1y//33S2lp6VnX+Mknn8gHH3wg48eP/299kUhETp48KceOHZNIJCKHDh2q9dcA7lqxYsV5dVfH4370+/f1tEH/uloa8D97qos777xTysvLz3rODz/8IAsWLJDFixfL/v37pWPHjjJ9+nR54IEHJCMjI04rRTLYu3evnD59WsaNGyfjxo1Tf56Xlyfjx4+vsfN61KhRctNNN/13nJaWps7ZsGGDLFiwQN588005ePCgXHfddTJ37lwpLCyU1q1bn3ONv/32m4iI3HXXXeb68/LyZPbs2TJhwoRzXgvJZcmSJXLxxRfL4MGDA50fj/vRr6ysTPLz86VHjx7n/bHJhIJsONtT8dq1a2Xy5MnyzTffSFpamtxzzz1SXFwsvXr1ivMqkSw6d+4s77zzjsqfeeYZqaqqkpdeekny8/Nr/Pj27dtL+/btzT97++23ZerUqfL9999L69atpaioSIqLi+W66647rzXedttt5hoffvhhycnJkaeffjq0znC4488//5SPPvpIRowYEXi2QjzuxzNt3rxZtm/fLlOmTKn1NZJFgy7Ic+fOlcrKStm3b5+IiLz77ruyZ88eEfm/9vxWrVqd9zXXrVsn1dXVUlpaKoWFhbW6Rn2vEW5JT0+XoUOHqvzfJ2Lrz4Jas2aNpKeny6JFi+See+4xn1aCyM7OluzsbJVPmDBBMjIy6rRGuCvszuWw7sczLVmyREQa/q+rRURSPM/zEr2I+pKbmyu//vqr+We7du2q1b+3O3LkSKj/Bq4+1ojk0Lt3b4lGo7Jt27ZaXyPs+9EvNzdXOnfuLKtXr663z4HEKSgokF9++UX27dsnF154YZ2vF/b9ePr0acnOzpaMjAz59ttvQ7uuqxp0QQYAIFnQZQ0AgAMoyAAAOICCDACAAyjIAAA4gIIMAIADKMgAADiAggwAgAMCT+pKSUmpz3WgAYjnP2nnfsS5cD/CJUHuR56QAQBwAAUZAAAHUJABAHAABRkAAAc06O0XgYagadOmKsvKyoo5rqysVOdEo9H6WhKAesATMgAADqAgAwDgAAoyAAAOoCADAOAAmroAh6Snp6vs1VdfVVn//v1jjnfv3q3OeeKJJ1S2atUqlcVzohWAmvGEDACAAyjIAAA4gIIMAIADKMgAADggxQvY0cH2YjgXtrs7P7m5uSqbM2eOygYNGqSy06dPxxxfeOGF6pz169erbMCAASo7dOjQWVaZvLgf4RK2XwQAIElQkAEAcAAFGQAAB1CQAQBwAJO6gDgIOoGrZ8+eKps+fbrKvvzyy5jj1atXB/qcR48ePes6ASQOT8gAADiAggwAgAMoyAAAOID3kIGQWUM6XnjhBZX17t1bZcXFxSpbtGiRyi655JKY4/3796tzLrjA7b9vX3bZZTHHLVq0UOdEIpE4rabhuugi/TLvHyxTU4b4cvsnFgCARoKCDACAAyjIAAA4gIIMAIADaOoCQjZ8+HCV3XfffSpbunRpoMzaJaZNmzYxx61bt1bnxGMIiLXLUU5Ojsoee+wxlY0cOTLm+MCBA+qcrl27quzkyZPns8QGK+jXvqioSGX+wTIiIuvWrVMZjV7xxRMyAAAOoCADAOAACjIAAA6gIAMA4ACauoCArAlc1rSt2bNnq2zPnj0qe+KJJ1R24sSJQGvZvXt3zHF5ebk65+qrrw50raCsiU9PPvmkyh588EGV5eXlqczflHTw4ME6rK7xsRq4rJ3B7r77bpVZzYObN29WWWVlZe0Wh1rhCRkAAAdQkAEAcAAFGQAAB1CQAQBwAE1dgMGagmRN4Jo2bZrKrOlG/fv3V9kff/xRy9Xp9WVnZ9f6Wpb09HSVlZaWqmzIkCEqS01NVdmpU6dU5m+SO3To0PkssVFp1qyZykaNGqWyoUOHqqxp06Yqy8zMVFlaWprKaOqKL56QAQBwAAUZAAAHUJABAHAABRkAAAfEpanLaiq4+OKLY46tpoW///470PUvvfRSlVkTj6xJQNbWdlZDj397uxYtWqhzrElG1lZxQZt5mjdvHui8I0eOqKy6ulpl/mYjtlar2eDBg1U2f/58lV1wgf47bXFxscp27NgRzsJqcMkll9T6Y3Nzc1U2Z84clVlfk+PHj6vsvffeU9nLL7+ssiVLlsQcP//88+qcxrrVov++6tevnzrn0UcfVZn1umT9nFuvGY31a+0SnpABAHAABRkAAAdQkAEAcAAFGQAAB8SlqcuaCuPfGs46Z8uWLYGuf+2116osGo2qbPv27SoLMkFIRKRTp04xx1YjjL9RTUTk8OHDKvv8889VZsnKylKZ1YT222+/qeyvv/5SmX8SknVOY+VvyJsyZYo6Z//+/Sp77bXXVGZtbRc2fyOjv+lQxP75ad++vcrWrFmjsvz8fJVFIhGVTZw4UWUffvihyq644gqVHTt2LOZ43bp16pzGyt8I27NnT3VOmzZtVGY1pFqvcd99953KmMqVeDwhAwDgAAoyAAAOoCADAOAACjIAAA4IvanLaoiytmjzb2XXoUMHdU5VVZXKrKYmq6HFmtRlXS8o/+ewGriCTuqymoMs1tQdizV1x2rAKS8vjzmePXu2OseavNQYZGRkxBxbE7gmTJigMqshymqiCVurVq1ijq1pdd26dVOZ1XCVk5Ojsl27dqns4YcfVtnHH3981nX+q2XLlipbv359zPGBAwcCXasx8N+Pt956qzrHer2xXh+tiYdbt25VWTzuW5wdT8gAADiAggwAgAMoyAAAOICCDACAA+IyqSuI1NRUlfkbG0TsJilrwkyTJk1UZk0LCsq/Pqt5zZqSY/1/WRO4LNb1LFZDj7XlpX96mbVl3z///KOyxrBN4969e2OOu3fvrs5Jtu3prElOVmY1Zo0ePVplVqNgUNZWovPmzYs5bgz3WVD+75P1fQvKmgzob6gT4evvAp6QAQBwAAUZAAAHUJABAHAABRkAAAeE3tRlTYrZuXOnyvyTYqzJSNZ0H2s7w82bN6ssPT1dZVdddZXKrM9rCTKpy2rgslifM+g6rEYv62Ot6WV5eXkxx9YkMGuqT2Ns9nC9gcv6/gYRjwYuyxdffBHq9Ro6/2Quq8HVYv2sLlu2TGWNfSqa1ZTrb5zzT8MTsbesDXMbW56QAQBwAAUZAAAHUJABAHBA6O8hW+9h+HcZEhH5+uuvY47rsnuS9Tt8azCG9Z6AxXp/oVOnTjHHubm56hzrPW/r/d3OnTurzNrtKjMzU2XWe+PW+/YHDx5UWZDdrqxrIbGse3n8+PExx9bPirXj2dy5c1UW9vvFOD/W927EiBExx2lpaYGuZX0v165dq7KGsLNTs2bNVNavXz+VdenSJVB24403xhxbg5NmzZqlspkzZ6qstj0oPCEDAOAACjIAAA6gIAMA4AAKMgAADojLbk9Wc0l1dXXMsbVjU1BWg4I1QMQaemGxhm/4G8esYSRWc4bVIHbNNdeorGvXrirr0aOHynr16qUyq4Hgm2++UdmGDRtijg8dOqTOoakrsazhMiNHjlTZ8OHDz3mt9957T2WrV6+u3cJQb6ymvSC7O1k/9ytXrlSZK0NArNdHawc+6zyrIXfgwIEqe+SRR1Rm7YZnvS77G3CPHTumzrHqSph4QgYAwAEUZAAAHEBBBgDAARRkAAAcEJemLot/olfYOwpZzUl1+Rz+xoi6NIhZU7QqKipUZjVtWBO9jh49qrIVK1aobMuWLTHH1uSuxrizU6JY90ZhYaHKSktLVWY1pfhZOyw1hAlNycya3HfbbbepzJrS52c1y+7Zs0dldWnUDLoznb8Z8corr1Tn9OnTR2X+iWQi9oQsq9mxbdu2KmvSpInKrNc0K/P/bHz77bfqHGvqZJg7w/GEDACAAyjIAAA4gIIMAIADKMgAADggYU1dySbMJrTff/9dZdY0nZ9++kll69evV5nVVODf3lJEN4EwlSt+rKaUoA1c1vf33XffjTkeMGCAOsea1IXEsqZyWdP3/OdZrzfWtorW99yafGXdj1YjmTVB0Nq60N+Idfvtt6tzrEavoNtKWqzXr08//VRl1mtmkEavd955R51jvSaHiSdkAAAcQEEGAMABFGQAABxAQQYAwAE0dTnCvx2liN7yUURk06ZNKrOaG6wpPjRxxYfVuGNtoWg1cFn3wejRo1Xmn9h20003qXOs+weJFXSCVRDWRKuioiKVWdPZrI/t3bu3yrKyslRmbWfoZzWSWaLRqMqCTkG0to+dOXOmyj7//HOVBWnKDXMCV1A8IQMA4AAKMgAADqAgAwDgAAoyAAAOoKnLYVbjQWVlZfwXghpZDVzPPPOMyiZPnqwyq3ll/PjxKlu6dKnKpkyZEnO8Zs0adY41/Q3xY22RaW21mJ+ff85rWVseWhO+rCxoc5K1Haj1GmRt8ei/l3ft2qXO2blzp8peeeUVle3evVtlQRtSE9GIFSaekAEAcAAFGQAAB1CQAQBwAAUZAAAH0NQFBGRtRTdp0iSVDR8+XGVWg9WYMWNUtnLlykBr+eSTT2KOf/zxR3VOXbYIRd1ZE7I++ugjlQ0ePFhlN9xwQ8yxtV2ixZrQN3/+fJVZ90Z6errKIpGIyvxbf4roqXBHjhxR51RVVans+PHjKmvMeEIGAMABFGQAABxAQQYAwAEUZAAAHJDiBRyBYk1xAc4Uz+0d6/t+bN++vcrWr1+vsssvv1xlW7duVdm9996rsh07dtRydQjCxfvRas5q166dyoYNGxZz3LJly0DXt7YktJq6rIazZs2aqcxqzrK2CMW5BbkfeUIGAMABFGQAABxAQQYAwAEUZAAAHEBTF0LjYhNNUGlpaTHHZWVl6hxrotL+/ftVdvPNN6vM2noO9SuZ78eLLgpviGKyb0nYUNDUBQBAkqAgAwDgAAoyAAAOoCADAOAAtl9Eo9O0aVOV+acZ3XnnneqcRYsWqWzGjBkqo4ELdUUjVuPEEzIAAA6gIAMA4AAKMgAADmAwCEKTLIMYOnbsqDL/Dk2zZs1S50yfPl1l1q45cEOy3I9oHBgMAgBAkqAgAwDgAAoyAAAOoCADAOAABoOg0enVq5fKDh48GHNs7fZEAxeA+sQTMgAADqAgAwDgAAoyAAAOoCADAOAAJnUhNMkyGcna7SkrKyvmOBqNqnMqKytr/TkRf8lyP6JxYFIXAABJgoIMAIADKMgAADiAggwAgANo6kJoaKKBS7gf4RKaugAASBIUZAAAHEBBBgDAARRkAAAcELipCwAA1B+ekAEAcAAFGQAAB1CQAQBwAAUZAAAHUJABAHAABRkAAAdQkAEAcAAFGQAAB1CQAQBwwP8AQCPaaAZMSKwAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0235, Train Accuracy: 0.9929\n",
            "Epoch 9/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0222, Train Accuracy: 0.9933\n",
            "Epoch 10/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.0211, Train Accuracy: 0.9937\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "import time\n",
        "# Start the timer\n",
        "start_time = time.time()\n",
        "train_model(model, filtered_train_loader, loss_fn, optimizer, num_epochs,device)\n",
        "# End the timer\n",
        "end_time = time.time()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "bbd069ac-c142-49bd-8967-57ccce7f81c4",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bbd069ac-c142-49bd-8967-57ccce7f81c4",
        "outputId": "c26d3e14-26d9-4d53-de85-55716bbeabda"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training duration: 199.81 seconds\n",
            "Training duration: 3.33 minutes\n",
            "Training duration: 0.06 hours\n"
          ]
        }
      ],
      "source": [
        "\n",
        "# Calculate the training duration in seconds\n",
        "training_duration_seconds = abs(end_time - start_time)\n",
        "\n",
        "# Calculate the training duration in minutes\n",
        "training_duration_minutes = training_duration_seconds / 60\n",
        "\n",
        "# Calculate the training duration in hours\n",
        "training_duration_hours = training_duration_minutes / 60\n",
        "\n",
        "# Print the training duration in different units\n",
        "print(f\"Training duration: {training_duration_seconds:.2f} seconds\")\n",
        "print(f\"Training duration: {training_duration_minutes:.2f} minutes\")\n",
        "print(f\"Training duration: {training_duration_hours:.2f} hours\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "f989b4a1-7c0f-4e9c-9e08-93cbae908922",
      "metadata": {
        "id": "f989b4a1-7c0f-4e9c-9e08-93cbae908922"
      },
      "outputs": [],
      "source": [
        "def load_clean_model(model, save_dir):\n",
        "    # Model folder\n",
        "    model_path = f\"{save_dir}/clean_model.pth\"\n",
        "\n",
        "    if not os.path.exists(model_path):\n",
        "        raise FileNotFoundError(f\"Model file not found in {model_path}\")\n",
        "\n",
        "    model.load_state_dict(torch.load(model_path))\n",
        "    model.eval()\n",
        "\n",
        "    print(f\"Loaded clean_model state dict from {model_path}\")\n",
        "    return model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "0792c3d2-c70b-45c0-90ee-756375d09dab",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0792c3d2-c70b-45c0-90ee-756375d09dab",
        "outputId": "0736039c-7107-441e-b160-233f494f59e8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean images: 64\n"
          ]
        }
      ],
      "source": [
        "images, labels = next(iter(filtered_train_loader))\n",
        "num_clean_elements = len(images)\n",
        "print(f\"Number of clean images: {num_clean_elements}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "11oH9oQGw7Vp",
      "metadata": {
        "id": "11oH9oQGw7Vp"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plot_test_accuracy_loss(test_accuracy, test_loss):\n",
        "    plt.figure(figsize=(12, 6))\n",
        "    plt.subplot(1, 2, 1)\n",
        "    plt.bar(1, test_accuracy, color='r', label='Testing Accuracy')\n",
        "    plt.title('Testing Accuracy')\n",
        "    plt.xlabel('Evaluation')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.legend()\n",
        "\n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.bar(1, test_loss, color='r', label='Testing Loss')\n",
        "    plt.title('Testing Loss')\n",
        "    plt.xlabel('Evaluation')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ea9b8cfc-5d52-420d-a066-075752637960",
      "metadata": {
        "id": "ea9b8cfc-5d52-420d-a066-075752637960"
      },
      "outputs": [],
      "source": [
        "def evaluate(model, data_loader, loss_fn, device):\n",
        "    test_loss = 0.0\n",
        "    test_accuracy = 0.0\n",
        "    num_misclassified = 0\n",
        "    model.eval()\n",
        "    misclassified_images = []\n",
        "    misclassified_labels = []\n",
        "    misclassified_preds = []\n",
        "\n",
        "    images, labels = next(iter(data_loader))\n",
        "    images, labels = images.to(device), labels.to(device)  # Move data to the specified device\n",
        "    num_clean_test_elements = len(images)\n",
        "    print(f\"Number of clean test images: {num_clean_test_elements}\")\n",
        "\n",
        "    # Create dataset\n",
        "    test_ds = TensorDataset(images, labels)\n",
        "    # Create loader\n",
        "    test_loader = DataLoader(test_ds, batch_size=64)\n",
        "    dataset_size = len(test_loader.dataset)\n",
        "\n",
        "    with torch.no_grad():\n",
        "        for i, (x, y) in enumerate(test_loader):\n",
        "            x, y = x.to(device), y.to(device)  # Move data to the specified device\n",
        "            logits = model(x)\n",
        "            batch_loss = loss_fn(logits, y)\n",
        "            test_loss += batch_loss.item()\n",
        "            preds = torch.argmax(logits, dim=1)\n",
        "            test_accuracy += torch.sum(preds == y).item() / len(y)\n",
        "            incorrect_inds = torch.where(preds != y)[0]\n",
        "            num_misclassified += len(incorrect_inds)\n",
        "            if len(incorrect_inds) > 0:\n",
        "                misclassified_images.append(x[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_labels.append(y[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_preds.append(preds[incorrect_inds].cpu())  # Move data back to CPU\n",
        "\n",
        "    num_misclassified = min(num_misclassified, dataset_size)\n",
        "    print(f\"Number of misclassified images: {num_misclassified} out of {dataset_size}\")\n",
        "    test_loss /= len(test_loader)\n",
        "    test_accuracy /= len(test_loader)\n",
        "    print(f\"Clean Test loss: {test_loss:.4f}, Accuracy: {test_accuracy:.2%}\")\n",
        "\n",
        "    # Plot the test accuracy and loss\n",
        "    #plot_test_accuracy_loss(test_accuracy, test_loss)\n",
        "\n",
        "    # Handle empty lists\n",
        "    if len(misclassified_images) == 0:\n",
        "        misclassified_images.append(torch.tensor([]))\n",
        "    if len(misclassified_labels) == 0:\n",
        "        misclassified_labels.append(torch.tensor([]))\n",
        "    if len(misclassified_preds) == 0:\n",
        "        misclassified_preds.append(torch.tensor([]))\n",
        "\n",
        "    return test_loss, test_accuracy, [torch.cat(misclassified_images)], [torch.cat(misclassified_labels)], [torch.cat(misclassified_preds)]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3bc286ff-8d98-49a9-bb75-d2530073f446",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3bc286ff-8d98-49a9-bb75-d2530073f446",
        "outputId": "e830996c-8b57-44fd-90be-f8b230a44dcf"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded clean_model state dict from clean_models/clean_model.pth\n"
          ]
        }
      ],
      "source": [
        "# Model folder\n",
        "save_dir = 'clean_models'\n",
        "\n",
        "# Loading\n",
        "clean_model = load_clean_model(CNN().to(device), save_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e8853f9a-36e6-4a5a-bea7-b2bee29d3f2c",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e8853f9a-36e6-4a5a-bea7-b2bee29d3f2c",
        "outputId": "57d43ba4-a929-4d30-d48f-ccdaf85447bd"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean test images: 64\n",
            "Number of misclassified images: 0 out of 64\n",
            "Clean Test loss: 0.0014, Accuracy: 100.00%\n"
          ]
        }
      ],
      "source": [
        "from torch.utils.data import TensorDataset\n",
        "# Evaluation\n",
        "#test_loss, test_acc = evaluate(clean_model, filtered_test_loader, loss_fn)\n",
        "test_loss, test_acc, misclassified_images, misclassified_labels, misclassified_preds = evaluate(clean_model, filtered_test_loader, loss_fn, device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3d5e1d14-2ef0-48ae-831d-6fa01bfab343",
      "metadata": {
        "id": "3d5e1d14-2ef0-48ae-831d-6fa01bfab343"
      },
      "outputs": [],
      "source": [
        "import math\n",
        "def analyze_for_misclassifications(misclassified_images, misclassified_labels, misclassified_preds):\n",
        "\n",
        "    MAX_IMAGES = 60\n",
        "\n",
        "    if len(misclassified_images) > 0:\n",
        "        misclassified_images = misclassified_images[0]\n",
        "        misclassified_labels = misclassified_labels[0]\n",
        "        misclassified_preds = misclassified_preds[0]\n",
        "\n",
        "        print(f\"Number misclassified: {len(misclassified_images)}\")\n",
        "\n",
        "        # Truncate more than MAX_IMAGES\n",
        "        images = misclassified_images[:MAX_IMAGES]\n",
        "        labels = misclassified_labels[:MAX_IMAGES]\n",
        "        preds = misclassified_preds[:MAX_IMAGES]\n",
        "\n",
        "\n",
        "        num_images = len(images)\n",
        "\n",
        "        fig = plt.figure(figsize=(6, num_images))\n",
        "\n",
        "        rows = math.ceil(num_images / 2)\n",
        "        cols = 2 if num_images > 16 else min(num_images, 2)\n",
        "\n",
        "        for i, (img, label, pred) in enumerate(zip(images, labels, preds)):\n",
        "\n",
        "            # Add subplot\n",
        "            ax = fig.add_subplot(rows, cols, i+1)\n",
        "\n",
        "            # Add empty dim if needed\n",
        "            if img.ndim == 2:\n",
        "                img = img[:,None,:,]\n",
        "\n",
        "            # Permute to HWC format for imshow\n",
        "            img = img.permute(1, 2, 0)\n",
        "\n",
        "            ax.imshow(img)\n",
        "            ax.set_title(f\"{label} -> {pred}\")\n",
        "            ax.axis(\"off\")\n",
        "\n",
        "        plt.tight_layout()\n",
        "        plt.show()\n",
        "\n",
        "    else:\n",
        "        print(\"No misclassified examples\")\n",
        "\n",
        "    return misclassified_images, misclassified_labels, misclassified_preds"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "b9fd9757-996f-45c1-a3d4-73516de83f01",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "id": "b9fd9757-996f-45c1-a3d4-73516de83f01",
        "outputId": "fe0cdaeb-ae88-44fd-ae56-7b601a33dc26"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number misclassified: 0\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x0 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "clean_misclassified_images, clean_misclassified_labels, clean_misclassified_preds = analyze_for_misclassifications(misclassified_images, misclassified_labels, misclassified_preds)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3cbab15a-728a-4fc2-ab4b-f6e54188cc25",
      "metadata": {
        "id": "3cbab15a-728a-4fc2-ab4b-f6e54188cc25"
      },
      "outputs": [],
      "source": [
        "# Define compounded attack\n",
        "def compounded_attack(choose_attack_option):\n",
        "    if choose_attack_option == \"fgsm_cw_attack\":\n",
        "        attack1 = torchattacks.FGSM(model, eps=0.3)\n",
        "        attack2 = torchattacks.CW(model, c=0.1, kappa=0.0, steps=1000)\n",
        "        attack = torchattacks.MultiAttack([attack1, attack2])\n",
        "    elif choose_attack_option == \"fgsm_pgd_attack\":\n",
        "        attack1 = torchattacks.FGSM(model, eps=0.3)\n",
        "        attack2 = torchattacks.PGD(model, eps=0.3, alpha=0.01, steps=5)\n",
        "        attack = torchattacks.MultiAttack([attack1, attack2])\n",
        "    elif choose_attack_option == \"cw_pgd_attack\":\n",
        "        attack1 = torchattacks.CW(model, c=0.1, kappa=0.0, steps=1000)\n",
        "        attack2 = torchattacks.PGD(model, eps=0.3, alpha=0.01, steps=5)\n",
        "        attack = torchattacks.MultiAttack([attack1, attack2])\n",
        "    else:\n",
        "        print(\"You did not chose any of the possible options or an you made an invalid option\")\n",
        "    return attack"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "11559bea-21a9-4806-b46c-b35e5706ac10",
      "metadata": {
        "id": "11559bea-21a9-4806-b46c-b35e5706ac10"
      },
      "outputs": [],
      "source": [
        "attack = compounded_attack(compounded_attack_name)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "25bb76b8-c3ca-45b5-a5a5-804961d548d7",
      "metadata": {
        "id": "25bb76b8-c3ca-45b5-a5a5-804961d548d7"
      },
      "outputs": [],
      "source": [
        "import tqdm\n",
        "\n",
        "def perform_adv_attack(model, attack, data_loader):\n",
        "    adv_data = []\n",
        "    adv_labels = []\n",
        "    device = next(model.parameters()).device\n",
        "\n",
        "    total_batches = len(data_loader)\n",
        "    progress_bar = tqdm.tqdm(total=total_batches, unit='batch', desc='Adversarial Attack')\n",
        "\n",
        "    for x, y in data_loader:\n",
        "        x, y = x.to(device), y.to(device)\n",
        "\n",
        "        # Convert labels to long data type\n",
        "        y = y.long()\n",
        "\n",
        "        x_adv = attack(x, y)\n",
        "        adv_data.append(x_adv.detach())\n",
        "        adv_labels.append(y.detach())\n",
        "\n",
        "        progress_bar.update(1)\n",
        "\n",
        "    progress_bar.close()\n",
        "\n",
        "    adv_data = torch.cat(adv_data)\n",
        "    adv_labels = torch.cat(adv_labels)\n",
        "\n",
        "    # Move data back to CPU if necessary\n",
        "    if device != torch.device('cpu'):\n",
        "        adv_data = adv_data.to('cpu')\n",
        "        adv_labels = adv_labels.to('cpu')\n",
        "\n",
        "    # Create dataset and loader\n",
        "    adv_dataset = TensorDataset(adv_data, adv_labels)\n",
        "    adv_loader = DataLoader(adv_dataset, batch_size=64, shuffle=False)\n",
        "\n",
        "    return adv_loader"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "9bdac58d-942f-4fe3-91c8-f45cb75184d7",
      "metadata": {
        "id": "9bdac58d-942f-4fe3-91c8-f45cb75184d7"
      },
      "outputs": [],
      "source": [
        "def evaluate_adv_attack(model, data_loader, loss_fn, device):\n",
        "    attack_test_loss = 0\n",
        "    attack_test_accuracy = 0\n",
        "    num_misclassified = 0\n",
        "    model.eval()\n",
        "    misclassified_images = []\n",
        "    misclassified_labels = []\n",
        "    misclassified_preds = []\n",
        "\n",
        "    images, labels = next(iter(data_loader))\n",
        "    images, labels = images.to(device), labels.to(device)  # Move data to the specified device\n",
        "    num_clean_test_elements = len(images)\n",
        "    print(f\"Number of clean test images: {num_clean_test_elements}\")\n",
        "\n",
        "    # Create dataset\n",
        "    test_ds = TensorDataset(images, labels)\n",
        "    # Create loader\n",
        "    test_loader = DataLoader(test_ds, batch_size=64)\n",
        "    dataset_size = len(test_loader.dataset)\n",
        "\n",
        "    with torch.no_grad():\n",
        "        for i, (x, y) in enumerate(test_loader):\n",
        "            x, y = x.to(device), y.to(device)  # Move data to the specified device\n",
        "            logits = model(x)\n",
        "            batch_loss = loss_fn(logits, y)\n",
        "            attack_test_loss += batch_loss.item()\n",
        "            preds = torch.argmax(logits, dim=1)\n",
        "            attack_test_accuracy += torch.sum(preds == y).item() / len(y)\n",
        "            incorrect_inds = torch.where(preds != y)[0]\n",
        "            num_misclassified += len(incorrect_inds)\n",
        "            if len(incorrect_inds) > 0:\n",
        "                misclassified_images.append(x[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_labels.append(y[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_preds.append(preds[incorrect_inds].cpu())  # Move data back to CPU\n",
        "\n",
        "    num_misclassified = min(num_misclassified, dataset_size)\n",
        "    print(f\"Number of misclassified images: {num_misclassified} out of {dataset_size}\")\n",
        "    attack_test_loss /= len(test_loader)\n",
        "    attack_test_accuracy /= len(test_loader)\n",
        "    print(f\"Attack Test loss: {attack_test_loss:.4f}, Accuracy: {attack_test_accuracy:.2%}\")\n",
        "\n",
        "    # Handle empty lists\n",
        "    if len(misclassified_images) == 0:\n",
        "        misclassified_images.append(torch.tensor([]))\n",
        "    if len(misclassified_labels) == 0:\n",
        "        misclassified_labels.append(torch.tensor([]))\n",
        "    if len(misclassified_preds) == 0:\n",
        "        misclassified_preds.append(torch.tensor([]))\n",
        "\n",
        "    return attack_test_loss, attack_test_accuracy, [torch.cat(misclassified_images)], [torch.cat(misclassified_labels)], [torch.cat(misclassified_preds)]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "66c00ee0-efbe-4ea7-ab6b-ae46821c9c1f",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "66c00ee0-efbe-4ea7-ab6b-ae46821c9c1f",
        "outputId": "8b5baed4-d148-4bdc-a724-6684bc363e0e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Adversarial Attack: 100%|██████████| 625/625 [00:08<00:00, 73.23batch/s]\n"
          ]
        }
      ],
      "source": [
        "from torch.utils.data import TensorDataset\n",
        "# Adversarial attack\n",
        "adv_attack_loader = perform_adv_attack(clean_model, attack, filtered_test_loader)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "50c206b9-3580-40c9-afc1-865bc3f72676",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "50c206b9-3580-40c9-afc1-865bc3f72676",
        "outputId": "a43514fa-3c8a-4c32-d8a5-605551fa8db3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean test images: 64\n",
            "Number of misclassified images: 53 out of 64\n",
            "Attack Test loss: 3.4262, Accuracy: 17.19%\n",
            "\n",
            "A few key ways to evaluate whether an adversarial attack on a machine learning model was successful:\n",
            "\n",
            "Drop in Model Accuracy: 20-40% drop in accuracy indicates the attack was effective.\n",
            "\n",
            "High Loss Value: A sharp spike in loss is another sign of a working attack.\n",
            "\n",
            "Misclassified Examplese: Images wrongly labeled indicates a successful targeted attack\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# Evaluate attack\n",
        "clean_attack_test_loss, clean_attack_adv_test_accuracy, misclassified_images, misclassified_labels, misclassified_preds = evaluate_adv_attack(clean_model, adv_attack_loader, loss_fn, device)\n",
        "\n",
        "print(\"\\nA few key ways to evaluate whether an adversarial attack on a machine learning model was successful:\\n\")\n",
        "print(\"Drop in Model Accuracy: 20-40% drop in accuracy indicates the attack was effective.\\n\")\n",
        "print(\"High Loss Value: A sharp spike in loss is another sign of a working attack.\\n\")\n",
        "print(\"Misclassified Examplese: Images wrongly labeled indicates a successful targeted attack\\n\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e0c41cbf-0201-4188-b62e-2d771d104738",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "e0c41cbf-0201-4188-b62e-2d771d104738",
        "outputId": "641c158b-553e-46e6-ecb6-5514947e5bdf"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number misclassified: 53\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x5300 with 53 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "adv_misclassified_images, adv_misclassified_labels, adv_misclassified_preds = analyze_for_misclassifications(misclassified_images, misclassified_labels, misclassified_preds)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "065a9a4b-f936-4dde-b9c1-1b483974fd60",
      "metadata": {
        "id": "065a9a4b-f936-4dde-b9c1-1b483974fd60"
      },
      "outputs": [],
      "source": [
        "# Generate adversarial dataset\n",
        "import torch\n",
        "import tqdm\n",
        "from torch.utils.data import DataLoader, TensorDataset\n",
        "\n",
        "def generate_adversarial_dataset(model, attack, data_loader, device, num_samples):\n",
        "    adv_data = []\n",
        "    adv_labels = []\n",
        "    device = next(model.parameters()).device\n",
        "\n",
        "    num_generated = 0\n",
        "    progress_bar = tqdm.tqdm(total=num_samples, unit='sample', desc='Adversarial Attack')\n",
        "\n",
        "    for x, y in data_loader:\n",
        "        x, y = x.to(device), y.to(device)\n",
        "        y = y.long()\n",
        "\n",
        "        x_adv = attack(x, y)\n",
        "\n",
        "        with torch.no_grad():\n",
        "            orig_preds = model(x).argmax(1)\n",
        "            adv_preds = model(x_adv).argmax(1)\n",
        "\n",
        "        # Filter samples where original prediction is different from adversarial prediction\n",
        "        mask = orig_preds != adv_preds\n",
        "        x_adv_filtered = x_adv[mask]\n",
        "        y_filtered = y[mask]\n",
        "\n",
        "        # Append only the required number of samples\n",
        "        num_remaining = num_samples - num_generated\n",
        "        if num_remaining > 0:\n",
        "            if num_remaining >= len(x_adv_filtered):\n",
        "                adv_data.append(x_adv_filtered.detach())\n",
        "                adv_labels.append(y_filtered.detach())\n",
        "                num_generated += len(x_adv_filtered)\n",
        "                progress_bar.update(len(x_adv_filtered))\n",
        "            else:\n",
        "                adv_data.append(x_adv_filtered[:num_remaining].detach())\n",
        "                adv_labels.append(y_filtered[:num_remaining].detach())\n",
        "                num_generated = num_samples\n",
        "                progress_bar.update(num_remaining)\n",
        "                break\n",
        "\n",
        "        if num_generated >= num_samples:\n",
        "            break\n",
        "\n",
        "    progress_bar.close()\n",
        "\n",
        "    if len(adv_data) > 0:\n",
        "        adv_data = torch.cat(adv_data)\n",
        "        adv_labels = torch.cat(adv_labels)\n",
        "\n",
        "        if device != torch.device('cpu'):\n",
        "            adv_data = adv_data.to('cpu')\n",
        "            adv_labels = adv_labels.to('cpu')\n",
        "\n",
        "        adv_dataset = TensorDataset(adv_data, adv_labels)\n",
        "\n",
        "        if num_generated < num_samples:\n",
        "            print(f\"Warning: Could not generate {num_samples} unique adversarial examples.\")\n",
        "            print(f\"Returning {num_generated} unique adversarial examples.\")\n",
        "    else:\n",
        "        adv_dataset = None\n",
        "        print(\"Warning: No unique adversarial examples found.\")\n",
        "\n",
        "    return adv_dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "QN2lLX_vnc9d",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QN2lLX_vnc9d",
        "outputId": "55c2be49-b13c-4281-a50d-fc2ab5fdf84b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3750\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Adversarial Attack: 100%|██████████| 60/60 [00:00<00:00, 1211.74sample/s]\n"
          ]
        }
      ],
      "source": [
        "print(len(filtered_train_loader))\n",
        "adv_dataset = generate_adversarial_dataset(model, attack, filtered_train_loader, device, 60)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4c97dc22-32d7-451b-8a41-9bc190b40854",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 506
        },
        "id": "4c97dc22-32d7-451b-8a41-9bc190b40854",
        "outputId": "d8364437-8050-47c0-c47f-1dfa4c7826b7"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 6 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Print adversarial images\n",
        "adv_loader = DataLoader(adv_dataset, batch_size=len(adv_dataset))\n",
        "images, labels = next(iter(adv_loader))\n",
        "\n",
        "# Move the input tensors to the same device as the model\n",
        "device = next(model.parameters()).device\n",
        "images = images.to(device)\n",
        "\n",
        "preds = model(images)\n",
        "\n",
        "fig = plt.figure(figsize=(6, 6))\n",
        "for i in range(6):\n",
        "    ax = fig.add_subplot(2, 3, i+1)\n",
        "    img = images[i].cpu().squeeze()  # Move the image back to CPU for visualization\n",
        "    ax.imshow(img, cmap='gray')\n",
        "    orig = labels[i].item()\n",
        "    pred = preds[i].argmax().item()\n",
        "    ax.set_title(f\"{orig}->{pred}\")\n",
        "    ax.axis('off')\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "897f815c-f091-46c1-a30d-0413c365eb70",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "897f815c-f091-46c1-a30d-0413c365eb70",
        "outputId": "39e7d4cd-7566-4365-bb3e-5319950c83ff"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "60\n"
          ]
        }
      ],
      "source": [
        "from torch.utils.data import DataLoader, TensorDataset\n",
        "\n",
        "# Adversarial training\n",
        "# Create adversarial dataset\n",
        "adv_loader = DataLoader(adv_dataset, batch_size=len(adv_dataset))\n",
        "\n",
        "images, labels = next(iter(adv_loader))\n",
        "\n",
        "adv_ds = TensorDataset(images, labels)\n",
        "\n",
        "print(len(adv_ds))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ce797f73-d153-4f46-a98b-5ff1a4eb2fb2",
      "metadata": {
        "id": "ce797f73-d153-4f46-a98b-5ff1a4eb2fb2"
      },
      "outputs": [],
      "source": [
        "# Adversarial dataset\n",
        "adv_loader = DataLoader(adv_dataset, batch_size=len(adv_dataset))\n",
        "adv_images, adv_labels = next(iter(adv_loader))\n",
        "adv_ds = TensorDataset(adv_images, adv_labels)\n",
        "\n",
        "# Filtered training set\n",
        "images, labels = next(iter(filtered_train_loader))\n",
        "clean_ds = TensorDataset(images, labels)\n",
        "clean_loader = DataLoader(clean_ds, batch_size=len(clean_ds))\n",
        "\n",
        "# Combine into one dataset\n",
        "combined_tensors = (torch.cat((adv_images, images)),\n",
        "                    torch.cat((adv_labels, labels)))\n",
        "\n",
        "combined_tensor_dataset = TensorDataset(*combined_tensors)\n",
        "\n",
        "# Create combined data loader\n",
        "combineloader = DataLoader(combined_tensor_dataset, batch_size=128,\n",
        "                           shuffle=True, num_workers=4)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "4b31cf9f-cc7e-4141-bf4e-68a66e966182",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4b31cf9f-cc7e-4141-bf4e-68a66e966182",
        "outputId": "6d0f1e2e-2d53-4f39-ddc1-258ac9bbece2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean images: 64\n",
            "Number of adversarial images: 60\n",
            "Number of combined images: 124\n"
          ]
        }
      ],
      "source": [
        "for clean_images, clean_targets in clean_loader:\n",
        "    num_clean_elements = len(clean_images)\n",
        "    print(f\"Number of clean images: {num_clean_elements}\")\n",
        "\n",
        "for adv_images, adv_targets in adv_loader:\n",
        "    num_adv_elements = len(adv_images)\n",
        "    print(f\"Number of adversarial images: {num_adv_elements}\")\n",
        "\n",
        "for combined_images, combined_targets in combineloader:\n",
        "    num_combined_elements = len(combined_images)\n",
        "    print(f\"Number of combined images: {num_combined_elements}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "06331625-3cb7-47f1-bd95-b25240b2eb4f",
      "metadata": {
        "id": "06331625-3cb7-47f1-bd95-b25240b2eb4f"
      },
      "outputs": [],
      "source": [
        "def save_retrained_model(model, save_dir):\n",
        "\n",
        "    if not os.path.exists(save_dir):\n",
        "        os.makedirs(save_dir)\n",
        "\n",
        "    model_path = f\"{save_dir}/retrained_model.pth\"\n",
        "\n",
        "    torch.save(model.state_dict(), model_path)\n",
        "    #print(f\"Saved retrained_model state dict to {model_path}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "cb914981-f164-42f4-a7cb-17e77311d06b",
      "metadata": {
        "id": "cb914981-f164-42f4-a7cb-17e77311d06b"
      },
      "outputs": [],
      "source": [
        "def adversarial_retraining(model, combined_loader, optimizer, loss_fn, num_epochs, device):\n",
        "    adv_trained_dataset = []\n",
        "    adv_images = []\n",
        "    adv_labels = []\n",
        "\n",
        "    # Model folder\n",
        "    save_dir = 'retrained_models'\n",
        "    max_images = 124\n",
        "\n",
        "    model.to(device)  # Move the model to the specified device\n",
        "\n",
        "    for epoch in range(num_epochs):\n",
        "        print(f\"Epoch {epoch+1}/{num_epochs}\")\n",
        "\n",
        "        for i, (imgs, labels) in enumerate(combined_loader):\n",
        "            if len(adv_images) == max_images:\n",
        "                break\n",
        "\n",
        "            imgs, labels = imgs.to(device), labels.to(device)  # Move input data to the specified device\n",
        "\n",
        "            preds = model(imgs)\n",
        "            loss = loss_fn(preds, labels)\n",
        "\n",
        "            # Adversarial training step\n",
        "            optimizer.zero_grad()\n",
        "            loss.backward()\n",
        "            optimizer.step()\n",
        "\n",
        "            # Print images\n",
        "            if i == 0:\n",
        "                print_images(imgs.cpu(), labels.cpu(), preds.cpu(), n=3)  # Move data back to CPU for printing\n",
        "\n",
        "            adv_images.append(imgs.detach().cpu()[:max_images-len(adv_images)])  # Move data back to CPU\n",
        "            adv_labels.append(labels.detach().cpu()[:max_images-len(adv_labels)])  # Move data back to CPU\n",
        "\n",
        "    adv_images = torch.cat(adv_images)[:max_images]\n",
        "    adv_labels = torch.cat(adv_labels)[:max_images]\n",
        "\n",
        "    # Create TensorDataset\n",
        "    adv_dataset = TensorDataset(adv_images, adv_labels)\n",
        "    print(f\"Number of images: {len(adv_dataset)}\")\n",
        "\n",
        "    save_retrained_model(model, save_dir)\n",
        "\n",
        "    return adv_dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "502cbe2b-7413-40d1-8b68-9006ceb8159d",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "502cbe2b-7413-40d1-8b68-9006ceb8159d",
        "outputId": "3842a4f6-d7cb-4c4d-8354-d07bc10a2042"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 2/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAACzCAYAAABYUm8CAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAY6UlEQVR4nO3de3BU9RXA8ZPKI6GaICIQaMKbjoA8tBCJRQRrRSxRXkIBKVRH26JWK61Ta6tFp6OltgNFsFQJD6EyYBEsLSMqD+uIYgEfBAVElEIChBQIjwDB2z8cme6eQ/l59272l+X7meGPe/jt3d9ufruHyz05v4wgCAIBAAAp9ZVUTwAAAJCQAQDwAgkZAAAPkJABAPAACRkAAA+QkAEA8AAJGQAAD5CQAQDwAAkZAAAPkJABAPBAWifk48ePy/333y/NmzeXrKwsKSgokBUrVqR6WqfNmjVLMjIyzD9lZWWpnh4idPjwYXnooYekf//+0qhRI8nIyJBZs2aleloxVq1adcb1uHbt2lRPDxFat26d3HnnndKpUyf56le/Kvn5+XLzzTfLli1bUj2102rDHKNWJ9UTSKaxY8fKokWL5J577pH27dvLrFmzZMCAAbJy5Ur55je/merpnTZx4kRp3bp1TKxhw4apmQySory8XCZOnCj5+fnStWtXWbVqVaqndEZ333239OjRIybWrl27FM0GyfD444/L66+/LsOGDZMuXbpIWVmZTJ06VS677DJZu3atdO7cOdVTrBVzjFyQpt58881ARIJJkyadjh07dixo27Zt0KtXr9Dn3bx5c3DixIkophgUFxcHIhKsW7cukvPBX1VVVUFpaWkQBEGwbt26QESC4uLihM+7bdu24MiRIwmfJwiCYOXKlYGIBAsXLozkfPDX66+/Hhw/fjwmtmXLlqB+/frBqFGjQp83yvWYrDn6LG3/y3rRokVy3nnnye233346lpmZKbfeequ88cYbsnPnzlDnfeyxx6RFixYyYcIE2bx5c1TTlcrKSjl16lRk54Nf6tevL82aNYv8vHPnzpXc3Fz5wQ9+IOvWrYvsvJWVlVJdXR3Z+eCXwsJCqVevXkysffv20qlTp4S+16Jcj8mao8/SNiFv2LBBOnToINnZ2THxnj17iojIxo0bQ533rrvukv79+8v06dOlY8eOcuWVV8rMmTPl8OHDoefat29fyc7OlgYNGkhRUZFs3bo19Llwbhk5cqSMGjVKFixYID179pQuXbrI5MmTZf/+/aHPOW7cOMnOzpbMzEzp27evvP322xHOGL4KgkD27NkjjRs3Dn2OZKzHqOfos7RNyKWlpZKbm6viX8R2794d6ryXX365zJkzR0pLS+VPf/qTnDp1Sm699VbJzc2V2267Td544w3nczVo0EDGjh0rTz75pCxevFh+9rOfySuvvCKFhYWhr+BxbunQoYNMmzZNSktLZd68edKkSRO59957pUWLFjJixAhZsWKFfPbZZ07nqlevngwZMkQmT54sS5YskUcffVTee+896d27t2zYsCHJrwSpNm/ePNm1a5cMHz489DmiXI/JmqPXUv1/5snSpk2b4Prrr1fxjz76KBCR4A9/+ENkz1VSUhJMmDAhaNq0aSAiQceOHYM///nPoc712muvBRkZGcEdd9wR2fzglyjvIVt27NgRPPzww0GrVq0CEQlatmwZ/OY3vwl1rq1btwZZWVnBddddF/Es4ZPNmzcH2dnZQa9evYLq6upIzx3VekzmHH2Rtgm5U6dOQb9+/VR806ZNgYgETz311BkfW1lZGZSWlp7+s3fvXqfn3LZtW9CrV69ARIKuXbuGnXpwxRVXBG3btg39ePjtyybko0ePxqzHL4rDzqasrCwYOHBgICJBTk5O6PmOGDEiqFevXtp+CZ7rSktLgzZt2gR5eXnBrl27zjo+Fevxy86xtkrb/7LOzc2V0tJSFf8i1rx58zM+9ne/+53k5uae/hP/KyD/q6qqSubPny/XXnuttG/fXjZu3CijR4+WadOmhZ57Xl6eVFRUhH480suCBQti1qN1K+YL1dXVsnTpUhk0aJDk5eXJP/7xD7nppptk/vz5oZ8/Ly9PTpw4IUeOHAl9Dvjp4MGDcv3118uBAwdk+fLl//d78Qs1vR7DzLG2StvfQ+7WrZusXLlSDh06FFPY9eabb57++zMZM2ZMzO8pZ2VlqTFvvfWWFBcXy1/+8hc5ePCgdO/eXaZOnSojR45M+HeIt2/fLhdffHFC50D6uO66687a0KakpESKi4tl7ty5smfPHunQoYM88sgjMnbsWGnatGlCz799+3bJzMyU888/P6HzwC9VVVUycOBA2bJli7z88svSsWNHp8fV5HoMO8daK9WX6Mmydu1a9XvIVVVVQbt27YKCgoLQ5120aFHQqVOnQESChg0bBj/60Y+C9evXhzqX9V/hy5YtC0QkuPvuu0PPEX6L8h7yypUrg4KCgkBEgqysrOCWW24JVq9eHepc1nrcuHFjULdu3aCoqCjRqcIj1dXVQVFRUVCnTp1g2bJlkZ03yvWYrDn6LG2vkAsKCmTYsGHy85//XPbu3Svt2rWT2bNny44dO+SZZ54Jfd5ly5ZJ48aNZc6cOTJ06FDz6tlVYWGhdO/eXb7xjW9ITk6OrF+/XmbOnCl5eXnywAMPhD4v/DR16lQ5cODA6Qr/F198Uf7973+LyOe/TpeTk/Olz7l69Wo5efKkTJs2TUaOHBnqHF8YPny4ZGVlSWFhoTRp0kRKSkpkxowZ0qBBA3nsscdCnxf+ue+++2Tp0qUycOBAqaiokGeffTbm70ePHh3qvFGux2TN0Wup/hdBMh07diyYMGFC0KxZs6B+/fpBjx49guXLlyd0zsOHD0c0uyD4xS9+EXTr1i3IyckJ6tatG+Tn5wc//OEPg7KyssieA/5o2bJlICLmn48//jjUOaNcj5MnTw569uwZNGrUKKhTp06Qm5sbjB49Oti6dWtkzwE/9OnT54xrMZG0EOV6TNYcfZYRBEFQs/8EAAAA8dK2yhoAgNqEhAwAgAdIyAAAeICEDACAB0jIAAB4gIQMAIAHSMgAAHjAuVNXRkZGpE/8ne98J9Lzxfvb3/6W1PNDq8lfaY96PbpI9ppNlXT9rKT7ekTt4rIeuUIGAMADJGQAADxAQgYAwAMkZAAAPFDrtl90LUDxuQDHeg2JzDddi3Jqks/rxUUia8CX1846xrmOK2QAADxAQgYAwAMkZAAAPEBCBgDAA5EXdYUtEPG9oCN+fqkqhPGleMf3n9eX5fJ6KLwLL+zrd33Pz/X3F+mBK2QAADxAQgYAwAMkZAAAPEBCBgDAAxmB4x5lrtuLJbsIoyaKqXyeWyKiLFyyzlWbt7tL12JEF76s26jfy9q8HpF+2H4RAIBagoQMAIAHSMgAAHiAhAwAgAecO3X53DEnFc/p+n7UqZOaHS6rq6tVzJfinWSxXl/UW11GeS7XuUVZZOhTEVq6vi4gLK6QAQDwAAkZAAAPkJABAPBAam5wJiDZ9wTPpH79+jHHF1xwgRqTn5+vYn379lUx67GubrvtNhU7fPiwim3btk3FBg0aFHN88uRJNSZV729tVRPvV7LveUf9Gr7yFf3v/LCNM1h7OJdwhQwAgAdIyAAAeICEDACAB0jIAAB4oEaKulwKM5LdJOHLOO+881SsRYsWMcdTpkxRY6666ioVswpcrKKun/zkJyr2+9//XsVGjRqlYpWVlSr217/+VcVycnJijsvLy9UYi49NF1yLfVJRFFQT71f8c9RE4574wkYRkezsbBXr3bu3irVs2VLFbrjhhpjjl156SY05fvy4itXkLk7wl/XdasXCsporJRtXyAAAeICEDACAB0jIAAB4gIQMAIAHMgLHComBAweqmC/FPq4FLVa3oMaNG6tYnz59VGz16tUxx127dlVjrE5diez2ZL3nrp5++mkVu//++2OOrS5izz//fOjnrMlim0TeGxdRr22fO065Fse88MILKmatM6vwMDMzU8UGDx4cc7x48WI1ZsiQISpmFXpZP6+aXI9hO5FBs97Liy++WMWuvPJKFevcubOKuRZ6ffbZZzHHc+fOVWN27NjhdC6Ly3rkChkAAA+QkAEA8AAJGQAAD5CQAQDwgHPFkWuRS5TdtRIphLG6bbVu3VrFHnzwQRWzileKiorOev5ERN0VxtqmcefOnTHHhw4dUmNqouMTPpfIZ8qFtUYbNWqkYlaHOeuzYhVAWlt/WgVclptvvjnm+N1331VjcnNzVWzq1KlO50ftZK33J554QsWsNRrl9/Kll16qYuPHj1exffv2RfacXCEDAOABEjIAAB4gIQMA4AESMgAAHnAu6nIt1oqy2Md1S0ZrW7j+/fur2Lhx41TM2qawSZMmZ52ba6coq0Dsk08+UbEBAwaoWI8ePVRs3bp1Ts9riS+CsIoWrG0lKyoqQj9nqrmuoTBjEhXlc1iFVNY2n1YXuldeeUXF2rZtq2KFhYVOz+EqbHerVG3LipphFRm2atVKxVwLuKwOWS5rr3v37ioWvw2vCEVdAACkHRIyAAAeICEDAOABEjIAAB6IvFNXlKzijbp166qYVfxkdaqyCqxcuwqF3e7PKhqbN2+eit1zzz1O51uzZo2KWUUQL774oopddNFFMceXX365GjNnzhwV83nrwP/lukaTvZajfr+s4pULL7ww5viOO+5QY6ZNm6ZiVkc4a4vDsFvWiYgsWbJExW688UYV27p1a8zxRx99pMaUlZU5zaO2SOR9TUfW9rRXX3210zjLkSNHVCx+nYmIdOnSRcXiP2dWbrC2gbQKxMJu/ckVMgAAHiAhAwDgARIyAAAeICEDAOAB56KuVLCKWXJyclTsW9/6looVFxermLV9XJQ+/PBDFevXr5+KWQUK5eXlKmYVZm3fvl3FrrjiChWzikJcC0rSXdhOT1EXa1nFIPGFdyL2NnPdunWLOX7//ffVmEGDBqmYtfaOHz+uYkOHDlUxy/e//30VmzFjhtNjr7nmmpjjd955R42pqqpSsdrSlcv6vPXu3VvFsrOzVWzFihUxx9b7kA6sz4D1flis78zZs2erWNeuXZ3OF1+IZRWIWV0WwxZwWfiGBgDAAyRkAAA8QEIGAMADJGQAADzgdVGXtT2g1ZVrw4YNKta0aVMVc+2oYnXl2r9/f8zxyy+/rMYsXrxYxY4dO6ZiiXThsYp3Nm3a5PQcYYu6fCyiSWROqXg9VoGitW2otcWh1YntiSeeiDn+9re/rcZYBVzWet+9e7fTuNdee03F5s+fr2JWpyXr81hSUhJzfOLECTWmNrM+b9Z7M2HCBBX71a9+FXP8xz/+UY2xuq7VNtZ2hl/72tdUzFqPn376qYpZBWF9+/ZVMZfvQut7YufOnWd9XCK4QgYAwAMkZAAAPEBCBgDAAyRkAAA84E1R13PPPadiw4cPV7FnnnlGxZo0aeL0HKdOnVKxo0ePqlhBQYGKvfXWWzHH1pZ1VsejRFidoS644AIVu/fee1Xsxz/+cajnvPPOO1Vs6tSpoc5V01w7aUVZ1GUVZlmFJX369FExq2jRKpyKL34SEbnhhhtijq2CRWs9fvDBByo2adIkFQu7fkREvve97zmNKyoqijlOpNixtmwReujQIRWz1lB8YZM1prYVdWVlZalYq1atVMzaYtda3/Hd6kTsrlxWQaXl5MmTMcdWV674MVHjChkAAA+QkAEA8AAJGQAADyR0D9l115z8/HwVa9OmTcyxdb946dKlKuZ6v9jaHeWll15SsdGjR6tY/P1ii+v94kTuV950000qZt2LfOihh1TM2ikq/p7Tli1b1Jg9e/Z8iRn6JZH32uUepHUfq3nz5ipm3cd69dVXVWzXrl0q1qFDBxWzGnzEs9Z7aWmpilnNa6z7mq7C7pwlktg949rAuse7evVqp8fGf/bXr1+vxrzwwgsqZn0vWe+za+Mg12ZC8Z+NSy65RI2xdh8bPHiwilnr3frsud4btpqKWPVE8bUaCxYsUGOSfd+eK2QAADxAQgYAwAMkZAAAPEBCBgDAAzXSGGTAgAEqNn369LM+zrqRb7F2ZxoxYoSKWTfpy8vLnZ4jSq5NDKxmANbuKNZOMNYv3McXVaxatUqNWbhwodPcaksjBlfxBSJWc4Kvf/3rKrZt2zYVq6ysVDGreMW1YMYq1Dl48GDMsVUAGF84KZJYAZersGujtu3gFUZFRYWKWc0m4j+/EydOVGMuu+wyFTtw4ICKWTvEWYWrPXv2VLHOnTurmEvx14033qjGWAWLmZmZKhY1q4BtzZo1KjZlypSY47179yZtTmfCFTIAAB4gIQMA4AESMgAAHiAhAwDgAeeirkSKeFwKuCyHDx9WsRMnTjg91irgsoTtvBJ1EYnVnebaa69Vsfbt26vYO++84/Qcjz76aMzxL3/5SzUm3Yq1LFYRYKNGjWKOmzVrpsYsX75cxcaMGaNiubm5KuZawGUVoFgdveJ35Zo1a5YaE3UHLiTu008/VbHf/va3KjZ+/PiY49atW6sx1i5vFqsrlVV4aO0k59oNK0r79u1TsSNHjqiY9Rm1isSsTl2bNm06a8wqKrbOFSWukAEA8AAJGQAAD5CQAQDwAAkZAAAPJNSpyypsii82cWVttbh7924VKysrUzFru7tEJLugxSqUsLZSs17/+eefr2JZWVkqZhVy/PSnP4053r59uxpTXFysYrW5wGfIkCEqZm2ZGL+F5UUXXaTGdO/eXcVcOw1Z3basoitre75nn31WxeJ/dp988oka4/pzS1WXq9rSXStKVtHeU089pWLxHfn69eunxljFg9Z3gfV907Bhw/83zdPCFjZZ3ces4sT9+/er2OOPP65iO3bsUDFrfd9yyy0qZhXEjRw5UsV69eoVczxs2DCneUSJK2QAADxAQgYAwAMkZAAAPEBCBgDAA85FXa4FGKWlpSr2/PPPq1h8sU1RUZEaM2PGDBWzCgOsLfAs1hZ1VpGFxbXTUjxrC0WrA1dOTo6KXX311SrWoEEDp+e1iiWee+65mOPaXMBlzdNao9bae/XVV1UsvojLKjK0CriqqqpUzCrWWrZsmYqNGjVKxawCHGuNWgV5YdWWn3m6srb5e+CBB2KO8/Pz1Zi+ffuq2He/+10Vy8vLUzGrqMtae1ZRl1UwGl/EZW3v+Otf/1rFrCKpnTt3Oj1nSUmJim3dulXFnn76aRWL78wnInLhhRfGHFsFYpMmTVIx6zsgLK6QAQDwAAkZAAAPkJABAPAACRkAAA8k1KnLYnUkGjp0qIqtWLEi5viqq65SY6wtCeO7qYiItGzZUsWszlfxhRIiIps3b1YxyyWXXBJzbBV53X777SoWX0glIvL3v/9dxQYNGqRiVhGaK6uoK75Tl2thVG1mbVFndTtzYRVvfPjhhypmFZYsWbJExaxim4qKilBzc+X684260Cvd1lWUrM5X5eXl//dYROTdd99VscWLF6uY1cmwS5cuKnbppZeqWOPGjVXs7bffVrH4YltrHlZBrtXRy9XRo0dVzPrej+/CJyLStm3bs57feu1WkS5FXQAApBkSMgAAHiAhAwDgARIyAAAecC7qiroA6Mknn4w5XrBggRpjFQ9Y3Wk6d+6sYnfddZeK9ejRwylmdbCK7x5jdbB58MEHVez9999XsTZt2qhYIqzOUI888oiKWYUhLnzs5OS69qyikf/85z9nfdy//vUvp+dcv3690zwsrt22fHz/kXrV1dUqZq0pK2YVXVmFqq6dunxx5MgRFRs8eLCKWd0S44s9p0+frsYcPHgwgdmdHVfIAAB4gIQMAIAHSMgAAHiAhAwAgAci79RlsYpS4gsSysrK1Birw4xVyBDfgUpEZPz48Spm3dyfOXOmilmFDC6sbltW8YT1Glw71qxdu1bFrO5l+JxV8Pbee++pWPz2l1bxoPUzsjr+LFy48MtMMZRUdL6i21Z6c92KtraxCms/+OCDsz7O+p5ONq6QAQDwAAkZAAAPkJABAPAACRkAAA8kVNSV7A5Cs2fPVjFrS8LRo0er2DXXXKNiY8aMUTGrO40V69ix41nHWD7++GOncZWVlSpmFR5QwPU5185x1naGa9asUbF//vOfZ33OVBR5RM31faOAC+nM188yV8gAAHiAhAwAgAdIyAAAeICEDACAB2qkU5clbNGIdTN+ypQpTjHLuHHjnMZZ3cDi3XfffU7ncvXwww87jWN7vjNzXWe+vIfWfKOcG8VagL+4QgYAwAMkZAAAPEBCBgDAAwndQ/Z5xxnX+2779u0Ldb6aeO1R39d0mbMv91JTzfVebk2sA+77AucGrpABAPAACRkAAA+QkAEA8AAJGQAADzgXdflcWFIThUhhC6JS9b6Ffd5kN6aozRL5WabDe+j6Gnz+rgB8xhUyAAAeICEDAOABEjIAAB4gIQMA4IGU7fYUz6eCKEvYopxUdNuKmk8/hyikosAq3d5DANHjChkAAA+QkAEA8AAJGQAAD5CQAQDwQEYQBIHTwIyMSJ/Yl85FFNtEx3EpRSKR9Ri2gND3wkPEqi3rEecGl/XIFTIAAB4gIQMA4AESMgAAHiAhAwDgAeeiLgAAkDxcIQMA4AESMgAAHiAhAwDgARIyAAAeICEDAOABEjIAAB4gIQMA4AESMgAAHiAhAwDggf8CLFanZubGzDQAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 3/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 4/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 5/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAACzCAYAAABYUm8CAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAaM0lEQVR4nO3de3BU9fnH8SdcCgQbUghQQMAgAculEFAQxIKVSpVyrUpBLNcB6XCpI9JaoINFake8oKICbQGlxFLAADYdW2BQHEvRGhEULNdULgGSIncCJDm/Pzrw6+7zBA67Z7PfLO/XDH/sx7Nnv7v73X08yZPvN8nzPE8AAEBcVYr3AAAAAAUZAAAnUJABAHAABRkAAAdQkAEAcAAFGQAAB1CQAQBwAAUZAAAHUJABAHAABRkAAAckdEE+f/68/OxnP5OGDRtKjRo1pHPnzrJ27dp4DyvE8ePHZcyYMVK3bl2pWbOm3HXXXZKbmxvvYSFGcnNzpW/fvlK7dm1JTk6WNm3ayEsvvRTvYYmIyOLFiyUpKcn8d/jw4XgPDzE0a9YsSUpKkjZt2sR7KJf16NGjzPlYtWrVeA8vJqrEewCxNHz4cFmxYoX89Kc/lYyMDFm8eLHcd999smHDBunWrVu8hyelpaXSu3dv+fTTT+Xxxx+XtLQ0efXVV6VHjx7y8ccfS0ZGRryHiAD97W9/kz59+khmZqZMnz5dbrjhBtmzZ48cOHAg3kML8atf/UrS09NDstTU1PgMBjF34MAB+fWvfy01a9aM91BCTJ06VUaPHh2SnTlzRh555BG555574jSqGPMS1ObNmz0R8WbPnn05O3funHfzzTd7Xbp0ifi8O3bs8C5cuBDEEL1ly5Z5IuItX778cnb06FEvNTXVGzx4cCCPATecOHHCq1+/vjdgwACvpKQksPMGOR8XLVrkiYj30UcfBXI+VAyDBg3yvvvd73rdu3f3WrduHdW5gpyPliVLlngi4i1dujRmjxFPCfsj6xUrVkjlypVlzJgxl7Pq1avLqFGjZNOmTbJ///6Izvub3/xGGjVqJJMnT5YdO3ZEPcb69evLwIEDL2d169aVBx98UFavXi3nz5+P6vxwR1ZWlhw5ckRmzZollSpVkjNnzkhpaWnU5w1yPv6vU6dOSUlJSWDng5s2btwoK1askDlz5gRyvljNx0uysrKkZs2a0q9fv8DP7YKELciffPKJtGjRQlJSUkLyTp06iYjIli1bIjrvhAkT5Pvf/7689tpr0qpVK7njjjtk4cKFcvr06YjG2KFDB6lUKfRt6NSpk5w9e1Z27twZ0RjhnnXr1klKSoocPHhQWrZsKTfccIOkpKTIuHHjpKioKOLzBjkfL7nrrrskJSVFkpOTpW/fvrJr166IzwV3lZSUyIQJE2T06NHStm3bQM4Zi/l4SUFBgaxdu1b69+/v3I/Xg5KwBTk/P18aNGig8kvZoUOHIjpvx44d5Y033pD8/HyZP3++lJSUyKhRo6RBgwYyevRo2bRpU9zHCPfs2rVLiouLpV+/ftKrVy9ZuXKljBw5UubNmycjRoyI+LxBzsfk5GQZPny4vPLKK5KdnS1TpkyR9evXS9euXSP+iRLcNW/ePPn3v/8tM2fODOycQc7HcMuWLZPi4mJ56KGHAhuvc+L9M/NYadasmXfvvfeqfM+ePZ6IeC+88EJgj7V9+3Zv8uTJXv369T0R8Vq1auX99re/ver9KlWq5I0bN07l69ev90TEy87ODmyMiK9mzZp5IuI98sgjIfnYsWM9EfF27twZ2GNFOh8t77//vpeUlOSNHTs2sPEh/goLC73atWt7zz777OUsiN8hW4Kaj126dPHq1q3rXbx4MfAxuiJhr5Br1Khh/g720o8Ha9SoUeZ9T58+LYcPH778r6Cg4IqP9a1vfUtmz54tH3zwgXTp0kW2b98uc+fOjekYUbFcei8HDx4ckg8ZMkRE5IpXDuU1Hy3dunWTzp07y7p16yK6P9w0bdo0qV27tkyYMOGa7xuP+bh3717ZtGmTDBo0SKpUSdw/DkrYgtygQQPJz89X+aWsYcOGZd732WeflQYNGlz+d9ttt5V5bFFRkWRlZcn3vvc9ycjIkC1btsjQoUPl1VdfjekYUbFcei/r168fkterV09ERL766qsy71te87EsjRs3lmPHjkV8f7hl165dsmDBApk4caIcOnRI8vLyJC8vT4qKiuTixYuSl5d3xfc7HvMxKytLRCSxf1wtCfx3yO3bt5cNGzbIyZMnQxq7Nm/efPm/l+XHP/5xyN8pW1eqH374oSxatEjefPNNOXHihGRmZsrcuXNlyJAhvv9ms3379vL+++9LaWlpSGPX5s2bJTk5WVq0aOHrPHBfx44dZe3atZebui651CdQt27dMu9bXvOxLHv37r3i+FCxHDx4UEpLS2XixIkyceJE9d/T09Nl0qRJZXZex2M+ZmVlyc033yy33377Nd+3Qon3z8xj5R//+If6O+SioiKvefPmXufOnSM+74oVK7zWrVt7IuKlpqZ6P/nJT7zc3NyIzvXHP/5R/R1yQUGBl5qa6g0aNCjiMcI9ubm5noh4Q4YMCckHDx7sValSxTt48GBE5w1yPh49elRlOTk5noh4EydOjOiccE9BQYGXnZ2t/rVu3dpr0qSJl52d7W3dujWicwc5Hy+59NmZPn16VOepCBL2Crlz587ywAMPyBNPPCFHjx6V5s2by+uvvy55eXny+9//PuLz5uTkSFpamrzxxhty//33R/V73vvvv19uv/12GTFihGzfvv3ySl0lJSXy5JNPRnxeuCczM1NGjhwpCxculOLiYunevbu8++67snz5cnniiSci/vVEkPOxa9eukpmZKbfeeqvUqlVLcnNzZeHChdK4cWP5xS9+EfF54Za0tDTp37+/yi9dEVv/za8g5+MlS5cuFZHE/3G1iCTuFbLn/XdlrsmTJ3vf/OY3vWrVqnm33Xab984770R1ztOnTwc0uv86duyYN2rUKK9OnTpecnKy1717d1ZKSlAXLlzwZsyY4TVt2tSrWrWq17x586i7/YOcj1OnTvXat2/v1apVy6tatarXpEkTb9y4cd7hw4cDewy4K4gu66C/H0tKSrxGjRp5HTp0CPS8rkryPM+L9/8UAABwvUvYLmsAACoSCjIAAA6gIAMA4AAKMgAADqAgAwDgAAoyAAAOoCADAOAA3yt1JSUlxXIcSADl+Sftffr0CfR8f/7znwM9nx8/+MEPyv0xoxGP1ygaFXk+hovmtS+PeeZnfNGMo6LNPYuf+cgVMgAADqAgAwDgAAoyAAAOoCADAOCAhN1+MR4qVdL/f2NlfhUXF0czHFznEqER5noQ6fvkSrOWxe/YXJqjfsYc6/FyhQwAgAMoyAAAOICCDACAAyjIAAA4gKYun6pXrx5yu2HDhuqYdu3aqaxNmzYqsxq9SktLVZaTk6OyY8eOqSw/P19l58+fVxnKFmSDTDSNH9Z9Ix1bRWys8cN6Xi4+hyDHFHQDV9CvVzxWnQt6Hrgwh7hCBgDAARRkAAAcQEEGAMABFGQAABxw3Td1Va5cWWVNmzZV2dChQ0Nu9+3bVx3TuHFjlaWmpqrM2srS2pprxIgRKissLFTZCy+8oLJVq1ap7Ny5cyrDf8V6+7igBbmCksvNXy695okkEV7XeDSmsVIXAADXAQoyAAAOoCADAOAACjIAAA5I8qxuIutAoxGpoglfbUtEpFevXiobPny4ynr27Blyu2bNmoGN61pYb9fevXtV9stf/lJlb731VsjtoFfz8jmVAtGnT59ye6xY8dsgUlFWprrEani0WKvTRcp6PcpzPsbj+zFejVnx2C6yPOZ7rJu6/MxHrpABAHAABRkAAAdQkAEAcEDCLgySlpamst69e6ts6tSpKktPT1dZ+AIi1u+//vOf/6js+PHjVxrmFTVp0kRl1apVU1mzZs1UNmXKFJV9/vnnIbe3bt0a8dgqsiB3VIpGNI8Zj92prMdMTk5W2d13362yKlX0V81f/vKXkNuJtkNZIiy+EQ/RzMeKjitkAAAcQEEGAMABFGQAABxAQQYAwAEJ0dRl/VH+2LFjVTZ+/HiV1atXz9f5iouLQ26/99576pj58+erbNu2bSorKSlRmbXr1LRp01T2wx/+UGXWgictW7ZU2cCBA0Nu7969Wx1z9uxZlSWaeDRE+eV3EZBIn4Pf8VoLr1iNknPmzFFZSkqKr8cIP9/ChQvVMdZnxeVFUeKtUiV9jWVl8ZCRkaEya6c6vxJxHrjxTgEAcJ2jIAMA4AAKMgAADqAgAwDggArX1GWtVHXrrbeq7Oc//7nKrB2arKaRffv2qWzWrFkht3NyctQxhYWFKouG1Zi2Z88elVmrclmNXo899ljI7VOnTqljXnzxRZWFN7S5KprmJz8NIta5gjz/tRwXa3Xq1FFZ+GdAxH8Dl6Vfv34R3zeR+Jm31jEjRoxQ2Xe+8x2VtW7dWmVWo9eYMWNUtmDBApV9+eWXVz3fBx98oI752te+prLnn39eZW3btlWZ9R2/ePFilcW6YTPoz3s4rpABAHAABRkAAAdQkAEAcAAFGQAAByR5nuf5OtBYvSrWrAau8NWmROympnbt2qnM2t7tr3/9q8oWLVp01eOKiorUMeWhQ4cOKvvTn/6kMmtLxvAtI1etWqWOGTdunMoKCgp8jc3nVAqEtZKU3yYMVxqnLK6sImY1dc2ePVtl1upd1takq1evVpn1OQtSvOejJci5Zz3mmjVrfN03Ozvb13HWa2g1g4Z/V9eoUUMdE00TX9WqVVXmt9k0Hts0Wu+zn/nIFTIAAA6gIAMA4AAKMgAADqAgAwDgAGdW6rKaxm655RaVWQ1c3/72t1Vm/QL9iy++UNmMGTNU5nfLxHjYv3+/yrZs2aKym266SWXhq+lYK/jUqlVLZX6buuLNb/OGK00efsdRpYr+mN54440ht+fOnauOsRphrPfcOv/DDz+sMqvRy2+zpzUWP+LxXgUl1s2Dbdq08XXc22+/rTLrPbea8aztWLOyslSWmpoacvuee+7xNTbLW2+9pTLruVrfe7GeL7F+T7lCBgDAARRkAAAcQEEGAMABFGQAABwQt6au8GYQa2WpyZMnq8xq9LIaS6xtCp955hmV7dixQ2WuNHBZvvrqK5Vt2rRJZffee6/KwlfPsbZgqyhivSpX0M0h1vmsxpomTZqo7J133lFZ+EpL6enp6pjwRpuyMuvzY43NYjVPWlnPnj1VVr9+/ZDbrVq1UsdYDUTluQKXy55++mmVDRs2TGW5ubkqs7ai/eyzz1RmbfHYsmVLlb388sshtx966CF1jMVqOLPmY9DfVZE2WbL9IgAA1wEKMgAADqAgAwDgAAoyAAAOiFtTV926dUNuT5s2TR1jbbVobcl47tw5lVkrcFnbDVpbMrrM2nLs3XffVdmRI0dUZq3elejisf2idf5Ro0ap7He/+13EjxG+naH13lqfH4vfxhqrmcraVtFaOW/8+PEqq169eshtazvBrVu3qsxaSS/e/M6pWM/H6dOnqyy8eU5E5MSJEyqzGqfmzJmjMj/bHm7cuFFlu3fvVllGRobKLl68qLKTJ09e9TFF/DdruboCHFfIAAA4gIIMAIADKMgAADiAggwAgAPKpakrvHlDRK8k5WdlKRG7CcvarsvKrOavRGA1aFiNEYkumuYYP00e1upVfrcVPHbs2DWP6UrCPy9Wo9NTTz2lsqFDh6rMWpmucuXKvo678847VWY1elkrhIVv9Wk1l1nb+Fmr6/lpNHJBNNtw+mF9x+Xl5QV2fpHYPwfrs2I1hFlcbdbyiytkAAAcQEEGAMABFGQAABxAQQYAwAGBN3VZzSC9evVS2dSpU0Nuh6/cVZbCwkKVLVy4UGWJ2sCFa+N3ZaTwlanatGmjjrEajKwV4azVjRo1aqSyRYsWqcza7u7jjz9W2SuvvBJy+/jx4+oYS+/evVVmrcBlrZplGTlypMrCt4YUsbdXveOOO0JuWytFPfbYYyqztqiM9eprkYjHKnHlwXpe4Z+f5557Th0zb948X+evXbu2yt577z2VhW/5WBarqfgPf/iDr/uWN66QAQBwAAUZAAAHUJABAHBA4L9Dbty4scqsnZyaN28ectvv7jL16tVTWbdu3VS2YcOGK44TFZvf389Z2aOPPqqy8Dlk7ZS0Zs0alU2aNEllX375pa/7LlmyxNdxkS560bZtW5Xl5OSorF27dr7O17JlS5Xt3LlTZRV9cYZI+H3OQb42sV6g41pYv/8PV1pa6utcDz74oK/jvv71r6ssOTlZZWlpaSr74osvVBb+OduyZYuvcQSJK2QAABxAQQYAwAEUZAAAHEBBBgDAAeWy25P1y/zw3YisBUWsRgGr+ctPQ0Eis147S3iT3MmTJ9UxFWXXnGg8//zzEd1v3759KrN2Fdu4cWNE5xcJ9vVfuXKlymbOnOnrvtbOTlYDl8X6PFoLrQwYMCDktrXbkzVHV61apbJ4N5K51GAVzno//H5nWruZLV++XGXVqlULuT1q1Ch1jN/Fn6x5YC1Uk5mZ6et8lo8++uiqx1jNyLNmzYr4Mf24visZAACOoCADAOAACjIAAA6gIAMA4IDAm7ry8/NV9tRTT6msY8eOIbe7d++ujrnzzjt9Paa1YkuispoxWrVqpbLU1FSVhTfqWDuoHDlyJPLBlSO/u+b4XYUq3LBhw1Rm7RDzwAMPqCweu/xY86JLly4qC2++EbGbLpcuXaqyfv36qcxq/rIew9rZqk6dOiG3i4qK1DG7d+9WmSvNUi6ydtWymp9at26tMmsOWd+t1n3DV2i0dmzy23zqd6exWDtw4EC5PyZXyAAAOICCDACAAyjIAAA4gIIMAIADAm/qOn/+vMqsLd/WrVt31XN17dpVZdbKMT169FBZlSr6qSXCKlQ1atRQmfU6feMb31BZ+EpdJ06cUMeEr6BW0X366acq+/DDD696nPXa3HfffSo7c+aMyvw2cAXZnBTeICVib0tqNe4UFBSozFppyS/rM/raa69d9X7bt2+P+DErCmulQeu9W7x48VXP1aFDB5W9+OKLEY2rLNnZ2SqzmsTCv1v8rM4oInL8+HGVWSvpWZ9H63OWm5urMss///lPlYU3Gg8dOtTXOIJcOY4rZAAAHEBBBgDAARRkAAAcQEEGAMABcdt+Mbz5y2q+sX7hb23hFb5KjIi9UszRo0evNEznWCvb9OzZU2Xh29iJ2E1t4c07n332mTrGeq9cFE1D1NNPP60yqzEjyMeMNWslrF69evm6r/U5s1grdaWnp6ts27ZtKrM+j2fPng257ec9KEusV0K7mmhWZ7Oa6sLfz5dfflkdM2HCBF/nX716tcr8Nl1ZK/dZjbvh3yWff/65OsZqqp00aZLKDh06pLLnnntOZX5Z781LL72ksvA5euzYMXVMNHPUD66QAQBwAAUZAAAHUJABAHAABRkAAAeUS1OXJbypwGrq2r9/v8qspi6rYcTautH6hby1fZwrmjZtqrLhw4f7Os5q0AhfCemTTz5Rx1SUpi6/4t3sU16sbSAPHz6sMmt7Pr+sbfGsFb1ef/11lVmNh+Gfvb///e/qmBtvvFFlFfk9tV7/tLS0q97PbwOXxVpty2q6spr7rO0v/bDe7/DVvMoaW3l8B1kNW+FZPFbc4woZAAAHUJABAHAABRkAAAdQkAEAcEDcmrrCWauzrFmzRmUtW7ZUWc2aNVX2ox/9SGVW04i1Ek08GpusbdkefvhhlVkrdVlb6lmrcE2bNi3k9sGDB69liNeVaBo1gl7RK9Impl27dqlsypQpKrOaAq0GHL+GDRvm67hHH3005Pa+ffsifsx48/se7d27N6LzWw2YViPsnj17VGZ9P1jbDwbJatZyaaW7SD9TsX4OXCEDAOAACjIAAA6gIAMA4AAKMgAADkjyfHZvWE1HsZacnKyy8EYQEZHx48erzFq968CBAyrbuHGjyubMmaOynTt3hty2VsLyq1q1airr37+/yhYsWKCyqlWrqiwnJ0dljz/+uMrCGz6iadyxBH2+K7FWjbJYzRtBNmb4Pb8rq0tt2bJFZe3atQv0Mf71r3+p7JZbbrnq/YJ+3cpzPkbz/Wh9f4WzVl27cOGCyqxGWL/zPZrXOvwxyuNz4VKTWDjrefmZj1whAwDgAAoyAAAOoCADAOAACjIAAA5wZqUuy9mzZ1U2f/58lVnNWnPnzlXZTTfdpLKGDRuqLDMzU2Xr1q0LuX3q1Cl1jF8pKSkqs5q6LFYDl/VcrVWPyrPJxRXxaPxwpYHLeu4zZ85U2cqVK1XWoUMHlc2bN09lS5YsUVleXp7PEQYn3g0+fhuWrOOs1yv8vuXRKBjNc4jkmGs57nrBFTIAAA6gIAMA4AAKMgAADqAgAwDgAKebuiyFhYUqW7ZsmcqsLeUGDhyosoyMDJW1bdtWZa1atfI7xIhYW6S9+eabKnvyySdVZjVwlZSUBDOwCiRRV/2xRDre8+fP+zouNzdXZZ06dYroMf2K9apqsRTrRr54NQpG+vqXR8NZIuIKGQAAB1CQAQBwAAUZAAAHVLjfIVuKiopU9swzz6hs1apVKhswYICvLPx3zdauS35ZO0WF7yYlIjJjxgyV7d27V2XX44IfQXN5MY8gufI88f/8vCdBLtpxLSLdtSnohUxc6Q+J9WvOFTIAAA6gIAMA4AAKMgAADqAgAwDggCTPZ0dQUlJSrMcSF9WrV1dZixYtVHb33XeH3LZ2bPLr5MmTKlu/fr3Ktm3bpjKXG7jKc2x9+vRRmcsNS/FqyvHD7+vmd2xBny9Sb7/9dkzP/78S9fsxEZTHTll++Pl+5AoZAAAHUJABAHAABRkAAAdQkAEAcMB139TlV5UqsV3UrLi4OKbnLw/l2dRV0eajy81almjGG2RTVzTPgfmIaATdDEZTFwAAFQQFGQAAB1CQAQBwAAUZAAAH0NSFwNBEExuRNlhFsxJYRVtFzMJ8hEto6gIAoIKgIAMA4AAKMgAADqAgAwDgAN9NXQAAIHa4QgYAwAEUZAAAHEBBBgDAARRkAAAcQEEGAMABFGQAABxAQQYAwAEUZAAAHEBBBgDAAf8H36F+ZOTFfrYAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 6/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 7/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 8/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 9/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 10/10\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x600 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of images: 124\n"
          ]
        }
      ],
      "source": [
        "# Start the timer\n",
        "start_time_retrain = time.time()\n",
        "adv_trained_dataset = adversarial_retraining(clean_model, combineloader, optimizer, loss_fn, num_epochs, device)\n",
        "# End the timer\n",
        "end_time_retrain = time.time()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5e16b45d-9255-4638-ae33-55d4df03b1d2",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5e16b45d-9255-4638-ae33-55d4df03b1d2",
        "outputId": "4f9f3771-a085-4839-854b-bab80a71c3c6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training duration: 5.93 seconds\n",
            "Training duration: 0.10 minutes\n",
            "Training duration: 0.00 hours\n"
          ]
        }
      ],
      "source": [
        "# Print the training duration in different units\n",
        "\n",
        "# Calculate the training duration\n",
        "retraining_duration_seconds = abs(end_time_retrain - start_time_retrain)\n",
        "\n",
        "# Calculate the training duration in minutes\n",
        "retraining_duration_minutes = retraining_duration_seconds / 60\n",
        "\n",
        "# Calculate the training duration in hours\n",
        "retraining_duration_hours = retraining_duration_minutes / 60\n",
        "\n",
        "print(f\"Training duration: {retraining_duration_seconds:.2f} seconds\")\n",
        "print(f\"Training duration: {retraining_duration_minutes:.2f} minutes\")\n",
        "print(f\"Training duration: {retraining_duration_hours:.2f} hours\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "701b7cc3-b3b6-42b6-9912-8725fda41e4f",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "701b7cc3-b3b6-42b6-9912-8725fda41e4f",
        "outputId": "3e007366-a489-47b5-81ec-4968e5bc8fee"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean images: 64\n",
            "Number of adversarial images: 60\n",
            "Number of combined images: 124\n",
            "Number of adversarially trained images: 124\n"
          ]
        }
      ],
      "source": [
        "for clean_images, clean_targets in clean_loader:\n",
        "    num_clean_elements = len(clean_images)\n",
        "    print(f\"Number of clean images: {num_clean_elements}\")\n",
        "\n",
        "for adv_images, adv_targets in adv_loader:\n",
        "    num_adv_elements = len(adv_images)\n",
        "    print(f\"Number of adversarial images: {num_adv_elements}\")\n",
        "\n",
        "for combined_images, combined_targets in combineloader:\n",
        "    num_combined_elements = len(combined_images)\n",
        "    print(f\"Number of combined images: {num_combined_elements}\")\n",
        "\n",
        "adv_trained_images_tensor, adv_trained_labels_tensor = adv_trained_dataset.tensors\n",
        "num_adv_trained_elements = len(adv_trained_images_tensor)\n",
        "print(f\"Number of adversarially trained images: {num_adv_trained_elements}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "b56c7aa2-e278-4187-9e9b-e753496e8987",
      "metadata": {
        "id": "b56c7aa2-e278-4187-9e9b-e753496e8987"
      },
      "outputs": [],
      "source": [
        "def load_retrained_model(model, save_dir):\n",
        "    # Model folder\n",
        "    model_path = f\"{save_dir}/retrained_model.pth\"\n",
        "\n",
        "    if not os.path.exists(model_path):\n",
        "        raise FileNotFoundError(f\"Model file not found in {model_path}\")\n",
        "\n",
        "    model.load_state_dict(torch.load(model_path))\n",
        "    model.eval()\n",
        "\n",
        "    print(f\"Loaded retrained_model state dict from {model_path}\")\n",
        "    return model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a8ed695e-4ec9-4d96-b130-8ae51030b87a",
      "metadata": {
        "id": "a8ed695e-4ec9-4d96-b130-8ae51030b87a"
      },
      "outputs": [],
      "source": [
        "def evaluate_after_adv_retraining(model, data_loader, loss_fn, device):\n",
        "    after_adv_test_loss = 0\n",
        "    after_adv_accuracy = 0\n",
        "    num_misclassified = 0\n",
        "    model.eval()\n",
        "    misclassified_images = []\n",
        "    misclassified_labels = []\n",
        "    misclassified_preds = []\n",
        "\n",
        "    # Create dataset\n",
        "    dataset_size = len(data_loader.dataset)\n",
        "    print(f\"Number of adv. retrained test images: {dataset_size}\")\n",
        "\n",
        "    with torch.no_grad():\n",
        "        for i, (x, y) in enumerate(data_loader):\n",
        "            x, y = x.to(device), y.to(device)  # Move input data to the specified device\n",
        "            logits = model(x)\n",
        "            batch_loss = loss_fn(logits, y)\n",
        "            after_adv_test_loss += batch_loss.item()\n",
        "            preds = torch.argmax(logits, dim=1)\n",
        "            after_adv_accuracy += torch.sum(preds == y).item() / len(y)\n",
        "            incorrect_inds = torch.where(preds != y)[0]\n",
        "            num_misclassified += len(incorrect_inds)\n",
        "\n",
        "            # Filter indices to current batch size\n",
        "            incorrect_inds = incorrect_inds[incorrect_inds < len(x)]\n",
        "\n",
        "            if len(incorrect_inds) > 0:\n",
        "                misclassified_images.append(x[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_labels.append(y[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_preds.append(preds[incorrect_inds].cpu())  # Move data back to CPU\n",
        "\n",
        "    num_misclassified = min(num_misclassified, dataset_size)\n",
        "    print(f\"Number of misclassified images: {num_misclassified} out of {dataset_size}\")\n",
        "\n",
        "    after_adv_test_loss /= len(data_loader)\n",
        "    after_adv_accuracy /= len(data_loader)\n",
        "    print(f\"Adversarial Retrained Test loss: {after_adv_test_loss:.4f}, Accuracy: {after_adv_accuracy:.2%}\")\n",
        "\n",
        "    # Handle empty lists\n",
        "    if len(misclassified_images) == 0:\n",
        "        misclassified_images.append(torch.tensor([]))\n",
        "    if len(misclassified_labels) == 0:\n",
        "        misclassified_labels.append(torch.tensor([]))\n",
        "    if len(misclassified_preds) == 0:\n",
        "        misclassified_preds.append(torch.tensor([]))\n",
        "\n",
        "    return after_adv_test_loss, after_adv_accuracy, [torch.cat(misclassified_images)], [torch.cat(misclassified_labels)], [torch.cat(misclassified_preds)]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3739430e-455c-4eba-8ec3-7fb2a0f197dc",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3739430e-455c-4eba-8ec3-7fb2a0f197dc",
        "outputId": "4825ab7a-910f-43d4-bdf8-411b3f4cd5b7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded retrained_model state dict from retrained_models/retrained_model.pth\n"
          ]
        }
      ],
      "source": [
        "# Model folder\n",
        "save_dir = 'retrained_models'\n",
        "\n",
        "# Loading\n",
        "retrained_model = load_retrained_model(CNN().to(device), save_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e6f1eb5b-1a84-4c95-ad31-724aaa0ff558",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e6f1eb5b-1a84-4c95-ad31-724aaa0ff558",
        "outputId": "470294dd-6c9a-4648-e49e-c278d28f360c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of adv. retrained test images: 124\n",
            "Number of misclassified images: 60 out of 124\n",
            "Adversarial Retrained Test loss: 2.3948, Accuracy: 51.61%\n"
          ]
        }
      ],
      "source": [
        "# Create dataset and loader\n",
        "adv_retrain_loader = DataLoader(adv_trained_dataset, batch_size=124, shuffle=False)\n",
        "adv_retrain_test_loss, adv_retrain_test_acc, misclassified_images, misclassified_labels, misclassified_preds = evaluate_after_adv_retraining(retrained_model, adv_retrain_loader, loss_fn, device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d89bfa58-0661-425b-86ca-006829db4d64",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "d89bfa58-0661-425b-86ca-006829db4d64",
        "outputId": "6c2b53c2-eb65-4ef3-baaa-a4662c45abb8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number misclassified: 60\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x6000 with 60 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(tensor([[[[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           ...,\n",
              "           [0.3000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]]],\n",
              " \n",
              " \n",
              "         [[[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.3000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.3000, 0.3000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           ...,\n",
              "           [0.0000, 0.0000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.3000, 0.0000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]]],\n",
              " \n",
              " \n",
              "         [[[0.0000, 0.0000, 0.0000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.3000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           ...,\n",
              "           [0.0000, 0.3000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.3000, 0.3000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]]],\n",
              " \n",
              " \n",
              "         ...,\n",
              " \n",
              " \n",
              "         [[[0.0000, 0.0000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.3000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.3000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           ...,\n",
              "           [0.0000, 0.0000, 0.3000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]]],\n",
              " \n",
              " \n",
              "         [[[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.3000, 0.3000,  ..., 0.3000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.3000, 0.3000,  ..., 0.3000, 0.0000, 0.0000],\n",
              "           ...,\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.3000,  ..., 0.0000, 0.0000, 0.0000]]],\n",
              " \n",
              " \n",
              "         [[[0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000],\n",
              "           ...,\n",
              "           [0.0000, 0.3000, 0.3000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           [0.3000, 0.3000, 0.0000,  ..., 0.3000, 0.3000, 0.0000],\n",
              "           [0.0000, 0.0000, 0.0000,  ..., 0.0000, 0.0000, 0.0000]]]]),\n",
              " tensor([1, 8, 5, 4, 3, 7, 3, 3, 5, 0, 3, 3, 2, 5, 8, 4, 8, 1, 3, 3, 7, 6, 7, 4,\n",
              "         5, 2, 1, 9, 6, 8, 5, 4, 0, 4, 0, 6, 8, 0, 8, 9, 1, 7, 0, 7, 0, 2, 9, 0,\n",
              "         1, 3, 2, 9, 6, 3, 5, 7, 4, 6, 8, 1]),\n",
              " tensor([8, 2, 6, 9, 5, 3, 5, 5, 3, 5, 5, 5, 3, 8, 6, 7, 6, 5, 5, 5, 3, 5, 2, 7,\n",
              "         7, 7, 5, 7, 5, 5, 8, 9, 2, 7, 3, 5, 5, 3, 5, 8, 3, 3, 3, 8, 6, 7, 7, 5,\n",
              "         2, 8, 0, 8, 7, 9, 3, 3, 8, 5, 5, 8]))"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ],
      "source": [
        "analyze_for_misclassifications(misclassified_images, misclassified_labels, misclassified_preds)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "45ecfee8-6779-470e-b649-5e97623ba464",
      "metadata": {
        "id": "45ecfee8-6779-470e-b649-5e97623ba464"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import tqdm\n",
        "from torch.utils.data import DataLoader, TensorDataset\n",
        "\n",
        "def perform_adv_attack_after_retraining(model, attack, data_loader, device):\n",
        "    adv_data = []\n",
        "    adv_labels = []\n",
        "    device = next(model.parameters()).device\n",
        "\n",
        "    total_batches = len(data_loader)\n",
        "    progress_bar = tqdm.tqdm(total=total_batches, unit='batch', desc='Adversarial Attack')\n",
        "\n",
        "    for x, y in data_loader:\n",
        "        x, y = x.to(device), y.to(device)\n",
        "\n",
        "        # Convert labels to long data type\n",
        "        y = y.long()\n",
        "\n",
        "        x_adv = attack(x, y)\n",
        "        adv_data.append(x_adv.detach())\n",
        "        adv_labels.append(y.detach())\n",
        "\n",
        "        progress_bar.update(1)\n",
        "\n",
        "    progress_bar.close()\n",
        "\n",
        "    adv_data = torch.cat(adv_data)\n",
        "    adv_labels = torch.cat(adv_labels)\n",
        "\n",
        "    # Move data back to CPU if necessary\n",
        "    if device != torch.device('cpu'):\n",
        "        adv_data = adv_data.to('cpu')\n",
        "        adv_labels = adv_labels.to('cpu')\n",
        "\n",
        "    # Create dataset and loader\n",
        "    adv_dataset = TensorDataset(adv_data, adv_labels)\n",
        "    adv_loader = DataLoader(adv_dataset, batch_size=64, shuffle=False)\n",
        "\n",
        "    return adv_dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "39ff4209-d0f0-408f-82e7-bb8fc6f1529e",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "39ff4209-d0f0-408f-82e7-bb8fc6f1529e",
        "outputId": "3e10cdbd-a922-435a-d025-572c5767e8e9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Adversarial Attack: 100%|██████████| 625/625 [00:08<00:00, 71.81batch/s]\n"
          ]
        }
      ],
      "source": [
        "# Adversarial attack on retrained model\n",
        "#adv_retrain_loader = perform_adv_attack_after_retraining(retrained_model, attack, filtered_test_loader, device)\n",
        "adv_retrain_loader = perform_adv_attack(model, attack, filtered_test_loader)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "998e2d3b-78ef-495d-bb29-c775be0f5588",
      "metadata": {
        "id": "998e2d3b-78ef-495d-bb29-c775be0f5588"
      },
      "outputs": [],
      "source": [
        "def evaluate_adv_attack_after_retraining(model, data_loader, loss_fn, device):\n",
        "    atack_after_retrained_test_loss = 0\n",
        "    atack_after_retrained_accuracy = 0\n",
        "    num_misclassified = 0\n",
        "    model.eval()\n",
        "    misclassified_images = []\n",
        "    misclassified_labels = []\n",
        "    misclassified_preds = []\n",
        "\n",
        "    images, labels = next(iter(data_loader))\n",
        "    images, labels = images.to(device), labels.to(device)  # Move data to the specified device\n",
        "    num_clean_test_elements = len(images)\n",
        "    print(f\"Number of clean test images: {num_clean_test_elements}\")\n",
        "\n",
        "    # Create dataset\n",
        "    test_ds = TensorDataset(images, labels)\n",
        "    # Create loader\n",
        "    test_loader = DataLoader(test_ds, batch_size=64)\n",
        "    dataset_size = len(test_loader.dataset)\n",
        "\n",
        "    with torch.no_grad():\n",
        "        for i, (x, y) in enumerate(test_loader):\n",
        "            x, y = x.to(device), y.to(device)  # Move data to the specified device\n",
        "            logits = model(x)\n",
        "            batch_loss = loss_fn(logits, y)\n",
        "            atack_after_retrained_test_loss += batch_loss.item()\n",
        "            preds = torch.argmax(logits, dim=1)\n",
        "            atack_after_retrained_accuracy += torch.sum(preds == y).item() / len(y)\n",
        "            incorrect_inds = torch.where(preds != y)[0]\n",
        "            num_misclassified += len(incorrect_inds)\n",
        "\n",
        "            # Filter indices to current batch size\n",
        "            incorrect_inds = incorrect_inds[incorrect_inds < len(x)]\n",
        "\n",
        "            if len(incorrect_inds) > 0:\n",
        "                misclassified_images.append(x[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_labels.append(y[incorrect_inds].cpu())  # Move data back to CPU\n",
        "                misclassified_preds.append(preds[incorrect_inds].cpu())  # Move data back to CPU\n",
        "\n",
        "    num_misclassified = min(num_misclassified, dataset_size)\n",
        "    print(f\"Number of misclassified images: {num_misclassified} out of {dataset_size}\")\n",
        "    atack_after_retrained_test_loss /= len(test_loader)\n",
        "    atack_after_retrained_accuracy /= len(test_loader)\n",
        "    print(f\"Adv. Attack Against Retrained Test loss: {atack_after_retrained_test_loss:.4f}, Accuracy: {atack_after_retrained_accuracy:.2%}\")\n",
        "\n",
        "    # Handle empty lists\n",
        "    if len(misclassified_images) == 0:\n",
        "        misclassified_images.append(torch.tensor([]))\n",
        "    if len(misclassified_labels) == 0:\n",
        "        misclassified_labels.append(torch.tensor([]))\n",
        "    if len(misclassified_preds) == 0:\n",
        "        misclassified_preds.append(torch.tensor([]))\n",
        "\n",
        "    return atack_after_retrained_test_loss, atack_after_retrained_accuracy, [torch.cat(misclassified_images)], [torch.cat(misclassified_labels)], [torch.cat(misclassified_preds)]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "a37544a3-e216-44e9-8cda-152bda7c539e",
      "metadata": {
        "id": "a37544a3-e216-44e9-8cda-152bda7c539e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "38264aae-4dde-43f4-a371-eea089c6b133"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean test images: 64\n",
            "Number of misclassified images: 0 out of 64\n",
            "Adv. Attack Against Retrained Test loss: 0.0014, Accuracy: 100.00%\n",
            "\n",
            "A few key ways to evaluate whether an adversarial attack on a machine learning model was successful:\n",
            "\n",
            "Drop in Model Accuracy: 20-40% drop in accuracy indicates the attack was effective.\n",
            "\n",
            "High Loss Value: A sharp spike in loss is another sign of a working attack.\n",
            "\n",
            "Misclassified Examplese: Images wrongly labeled indicates a successful targeted attack\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# Evaluate attack after retraining\n",
        "retrained_test_loss, retrained_accuracy, misclassified_images, misclassified_labels, misclassified_preds = evaluate_adv_attack_after_retraining(retrained_model, filtered_test_loader, loss_fn, device)\n",
        "print(\"\\nA few key ways to evaluate whether an adversarial attack on a machine learning model was successful:\\n\")\n",
        "print(\"Drop in Model Accuracy: 20-40% drop in accuracy indicates the attack was effective.\\n\")\n",
        "print(\"High Loss Value: A sharp spike in loss is another sign of a working attack.\\n\")\n",
        "print(\"Misclassified Examplese: Images wrongly labeled indicates a successful targeted attack\\n\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5ad629c3-7174-4fc6-82ec-51f470bd8d5e",
      "metadata": {
        "id": "5ad629c3-7174-4fc6-82ec-51f470bd8d5e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "07105ab3-9c8c-48a4-acf8-ad4068bd637e"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number misclassified: 0\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x0 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "retrained_misclassified_images, retrained_misclassified_labels, retrained_misclassified_preds = analyze_for_misclassifications(misclassified_images, misclassified_labels, misclassified_preds)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "d10babee-9534-4187-b9d7-4babdd96b8df",
      "metadata": {
        "id": "d10babee-9534-4187-b9d7-4babdd96b8df"
      },
      "outputs": [],
      "source": [
        "def precision(true_positives, false_positives):\n",
        "    if true_positives + false_positives == 0:\n",
        "        return 0.0\n",
        "    return true_positives / (true_positives + false_positives)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "3e20ecbb-accf-4238-8f7b-5e96678c518f",
      "metadata": {
        "id": "3e20ecbb-accf-4238-8f7b-5e96678c518f"
      },
      "outputs": [],
      "source": [
        "def recall(true_positives, false_negatives):\n",
        "    if true_positives + false_negatives == 0:\n",
        "        return 0.0\n",
        "    return true_positives / (true_positives + false_negatives)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "89773386-b77a-46c8-ad7d-3ae9ba0995cd",
      "metadata": {
        "id": "89773386-b77a-46c8-ad7d-3ae9ba0995cd"
      },
      "outputs": [],
      "source": [
        "def f1_score(precision, recall):\n",
        "    if precision + recall == 0:\n",
        "        return 0.0\n",
        "    return 2 * (precision * recall) / (precision + recall)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "54a1cc1a-f492-40ff-b85b-d43d101f654d",
      "metadata": {
        "id": "54a1cc1a-f492-40ff-b85b-d43d101f654d"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import roc_auc_score\n",
        "\n",
        "def auc_score(true_labels, predicted_probabilities):\n",
        "    if len(set(true_labels)) == 2:\n",
        "        return roc_auc_score(true_labels, predicted_probabilities)\n",
        "    else:\n",
        "        return \"ROC AUC score is not defined for a single class.\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "369a998a-4447-46f7-a83e-e00e289e55ec",
      "metadata": {
        "id": "369a998a-4447-46f7-a83e-e00e289e55ec"
      },
      "outputs": [],
      "source": [
        "def print_summary_images(images, labels, preds, stage, attack_name, dataset_name, num_clean_elements, num_adv_elements, num_combined_elements, num_epoch, num_test_images):\n",
        "    # Check for no images\n",
        "    if len(images) == 0:\n",
        "        print(f\"\\n---------------------------------------------------------\")\n",
        "        print(f\"No misclassified images for stage: {stage}\")\n",
        "        print(f\"Attack: {attack_name}\")\n",
        "        print(f\"Dataset: {dataset_name}\")\n",
        "        print(f\"Training Epochs: {num_epoch}\")\n",
        "        if stage == \"Clean\":\n",
        "            print(f\"Trained Clean Images: {num_clean_elements}\")\n",
        "            accuracy = test_acc\n",
        "        elif stage == \"No Defense Attack\":\n",
        "            print(f\"Adversarial Training Images: {num_adv_elements}\")\n",
        "            accuracy = clean_attack_adv_test_accuracy\n",
        "        elif stage == \"w/ Defense Attack\":\n",
        "            print(f\"Retrained Clean and Adversarial Images: {num_combined_elements}\")\n",
        "            accuracy = retrained_accuracy\n",
        "        print(f\"Test Images: {num_test_images}\")\n",
        "        print(f\"Accuracy: {accuracy:.2f}\")\n",
        "\n",
        "        true_positives = int(accuracy * num_test_images)\n",
        "        false_positives = 0\n",
        "        false_negatives = num_test_images - true_positives\n",
        "\n",
        "        precision_value = precision(true_positives, false_positives)\n",
        "        recall_value = recall(true_positives, false_negatives)\n",
        "        f1_score_value = f1_score(precision_value, recall_value)\n",
        "\n",
        "        # Create the true labels and predicted probabilities based on the variables\n",
        "        true_labels = [1] * true_positives + [0] * false_negatives\n",
        "        predicted_probabilities = [1] * true_positives + [0] * false_negatives\n",
        "        auc = auc_score(true_labels, predicted_probabilities)\n",
        "\n",
        "        print(f\"Precision: {precision_value:.2f}\")\n",
        "        print(f\"Recall: {recall_value:.2f}\")\n",
        "        print(f\"F1-score: {f1_score_value:.2f}\")\n",
        "        if isinstance(auc, float):\n",
        "            print(f\"ROC AUC Score: {auc:.2f}\")\n",
        "        else:\n",
        "            print(auc)\n",
        "        print(f\"---------------------------------------------------------\\n\")\n",
        "    else:\n",
        "        # Count the number of misclassifications for each situation\n",
        "        misclassifications = {}\n",
        "        for label, pred in zip(labels, preds):\n",
        "            if label != pred:\n",
        "                key = f\"{label} -> {pred}\"\n",
        "                misclassifications[key] = misclassifications.get(key, 0) + 1\n",
        "\n",
        "        # Plot images\n",
        "        num_images = len(images)\n",
        "        print(f\"\\n---------------------------------------------------------\")\n",
        "        print(f\"Number of misclassified images for {stage}: {num_images}\")\n",
        "        print(f\"Attack: {attack_name}\")\n",
        "        print(f\"Dataset: {dataset_name}\")\n",
        "        print(f\"Training Epochs: {num_epoch}\")\n",
        "        if stage == \"Clean\":\n",
        "            print(f\"Trained Clean Images: {num_clean_elements}\")\n",
        "            accuracy = test_acc\n",
        "        elif stage == \"No Defense Attack\":\n",
        "            print(f\"Adversarial Training Images: {num_adv_elements}\")\n",
        "            accuracy = clean_attack_adv_test_accuracy\n",
        "        elif stage == \"w/ Defense Attack\":\n",
        "            print(f\"Retrained Clean and Adversarial Images: {num_combined_elements}\")\n",
        "            accuracy = retrained_accuracy\n",
        "        print(f\"Test Images: {num_test_images}\")\n",
        "        print(f\"Accuracy: {accuracy:.2f}\")\n",
        "\n",
        "        true_positives = int(accuracy * num_test_images)\n",
        "        false_positives = num_images\n",
        "        false_negatives = num_test_images - true_positives - false_positives\n",
        "\n",
        "        precision_value = precision(true_positives, false_positives)\n",
        "        recall_value = recall(true_positives, false_negatives)\n",
        "        f1_score_value = f1_score(precision_value, recall_value)\n",
        "\n",
        "        # Create the true labels and predicted probabilities based on the variables\n",
        "        true_labels = [1] * true_positives + [0] * false_negatives\n",
        "        predicted_probabilities = [1] * true_positives + [0] * false_negatives\n",
        "        auc = auc_score(true_labels, predicted_probabilities)\n",
        "\n",
        "        print(f\"Precision: {precision_value:.2f}\")\n",
        "        print(f\"Recall: {recall_value:.2f}\")\n",
        "        print(f\"F1-score: {f1_score_value:.2f}\")\n",
        "        if isinstance(auc, float):\n",
        "            print(f\"ROC AUC Score: {auc:.2f}\")\n",
        "        else:\n",
        "            print(auc)\n",
        "        print(f\"---------------------------------------------------------\\n\")\n",
        "\n",
        "        fig = plt.figure(figsize=(6, num_images))\n",
        "        rows = math.ceil(num_images / 2)\n",
        "        cols = 2 if num_images > 16 else min(num_images, 2)\n",
        "        for i, (img, label, pred) in enumerate(zip(images, labels, preds)):\n",
        "            # Add subplot\n",
        "            ax = fig.add_subplot(rows, cols, i+1)\n",
        "            # Add empty dim if needed\n",
        "            if img.ndim == 2:\n",
        "                img = img[:,None,:,]\n",
        "            # Permute to HWC format for imshow\n",
        "            img = img.permute(1, 2, 0)\n",
        "            ax.imshow(img)\n",
        "            ax.set_title(f\"{label} -> {pred}\")\n",
        "            ax.axis(\"off\")\n",
        "\n",
        "        # Print the number of misclassifications for each situation\n",
        "        print(\"Misclassifications:\")\n",
        "        for situation, count in misclassifications.items():\n",
        "            print(f\"{situation}: {count}\")\n",
        "\n",
        "        plt.tight_layout()\n",
        "        plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "496d47bd-2cdf-443e-9958-f43f3b8fc818",
      "metadata": {
        "id": "496d47bd-2cdf-443e-9958-f43f3b8fc818",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2fda0af6-5215-4e00-d09d-de6311b274d1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of clean images: 64\n",
            "Number of adversarial images: 60\n",
            "Number of combined images: 124\n",
            "Number of adversarially trained images: 124\n"
          ]
        }
      ],
      "source": [
        "for clean_images, clean_labels in clean_loader:\n",
        "    num_clean_elements = len(clean_images)\n",
        "    print(f\"Number of clean images: {num_clean_elements}\")\n",
        "\n",
        "for adv_images, adv_labels in adv_loader:\n",
        "    num_adv_elements = len(adv_images)\n",
        "    print(f\"Number of adversarial images: {num_adv_elements}\")\n",
        "\n",
        "for combined_images, combined_labels in combineloader:\n",
        "    num_combined_elements = len(combined_images)\n",
        "    print(f\"Number of combined images: {num_combined_elements}\")\n",
        "\n",
        "adv_trained_images_tensor, adv_trained_labels_tensor = adv_trained_dataset.tensors\n",
        "num_adv_trained_elements = len(adv_trained_images_tensor)\n",
        "print(f\"Number of adversarially trained images: {num_adv_trained_elements}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "acc5b456-0852-401a-ac21-5dde331211f4",
      "metadata": {
        "id": "acc5b456-0852-401a-ac21-5dde331211f4"
      },
      "outputs": [],
      "source": [
        "import tabulate"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "fa6be136-af9e-4d9b-9ac6-746c9b181f7d",
      "metadata": {
        "id": "fa6be136-af9e-4d9b-9ac6-746c9b181f7d"
      },
      "outputs": [],
      "source": [
        "def summarize_performance(clean_loss, clean_acc, adv_loss, adv_acc, retrain_loss, retrain_acc,\n",
        "                          clean_misclassified_images, adv_misclassified_images, retrained_misclassified_images,\n",
        "                          clean_misclassified_labels, adv_misclassified_labels, retrained_misclassified_labels,\n",
        "                          clean_misclassified_preds,adv_misclassified_preds, retrained_misclassified_preds):\n",
        "\n",
        "    table_data = [[\"Metric\", \"Clean\", \"No Defense Attack\", \"w/ Defense Attack\"],\n",
        "                  [\"Loss\", f'{clean_loss:.3g}', f'{adv_loss:.3g}', f'{retrain_loss:.3g}'],\n",
        "                  [\"Accuracy\", f'{clean_acc:.0%}', f'{adv_acc:.0%}', f'{retrain_acc:.0%}']]\n",
        "\n",
        "    print(tabulate.tabulate(table_data, headers=\"firstrow\", tablefmt=\"grid\"))\n",
        "\n",
        "    print(\"\\nExample Misclassifications:\")\n",
        "    print_summary_images(clean_misclassified_images, clean_misclassified_labels, clean_misclassified_preds, \"Clean\", compounded_attack_name, dataset_name, num_clean_elements, num_adv_elements, num_combined_elements, num_epochs, len(filtered_test_loader))\n",
        "    print_summary_images(adv_misclassified_images, adv_misclassified_labels, adv_misclassified_preds, \"No Defense Attack\",  compounded_attack_name, dataset_name, num_clean_elements, num_adv_elements, num_combined_elements, num_epochs, len(filtered_test_loader))\n",
        "    print_summary_images(retrained_misclassified_images, retrained_misclassified_labels, retrained_misclassified_preds, \"w/ Defense Attack\",  compounded_attack_name, dataset_name, num_clean_elements, num_adv_elements, num_combined_elements, num_epochs, len(filtered_test_loader))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "5e465011-93ef-4447-84b8-838bf0f291fb",
      "metadata": {
        "id": "5e465011-93ef-4447-84b8-838bf0f291fb",
        "scrolled": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "172c9d10-dbea-4e99-8002-e0569c681b8d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "+----------+---------+---------------------+---------------------+\n",
            "| Metric   | Clean   | No Defense Attack   | w/ Defense Attack   |\n",
            "+==========+=========+=====================+=====================+\n",
            "| Loss     | 0.00142 | 3.43                | 0.00142             |\n",
            "+----------+---------+---------------------+---------------------+\n",
            "| Accuracy | 100%    | 17%                 | 100%                |\n",
            "+----------+---------+---------------------+---------------------+\n",
            "\n",
            "Example Misclassifications:\n",
            "\n",
            "---------------------------------------------------------\n",
            "No misclassified images for stage: Clean\n",
            "Attack: fgsm_pgd_attack\n",
            "Dataset: EMNIST\n",
            "Training Epochs: 10\n",
            "Trained Clean Images: 64\n",
            "Test Images: 625\n",
            "Accuracy: 1.00\n",
            "Precision: 1.00\n",
            "Recall: 1.00\n",
            "F1-score: 1.00\n",
            "ROC AUC score is not defined for a single class.\n",
            "---------------------------------------------------------\n",
            "\n",
            "\n",
            "---------------------------------------------------------\n",
            "Number of misclassified images for No Defense Attack: 53\n",
            "Attack: fgsm_pgd_attack\n",
            "Dataset: EMNIST\n",
            "Training Epochs: 10\n",
            "Adversarial Training Images: 60\n",
            "Test Images: 625\n",
            "Accuracy: 0.17\n",
            "Precision: 0.67\n",
            "Recall: 0.19\n",
            "F1-score: 0.29\n",
            "ROC AUC Score: 1.00\n",
            "---------------------------------------------------------\n",
            "\n",
            "Misclassifications:\n",
            "0 -> 5: 17\n",
            "0 -> 1: 1\n",
            "0 -> 9: 4\n",
            "0 -> 2: 9\n",
            "0 -> 6: 9\n",
            "0 -> 3: 6\n",
            "0 -> 8: 7\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x5300 with 53 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "---------------------------------------------------------\n",
            "No misclassified images for stage: w/ Defense Attack\n",
            "Attack: fgsm_pgd_attack\n",
            "Dataset: EMNIST\n",
            "Training Epochs: 10\n",
            "Retrained Clean and Adversarial Images: 124\n",
            "Test Images: 625\n",
            "Accuracy: 1.00\n",
            "Precision: 1.00\n",
            "Recall: 1.00\n",
            "F1-score: 1.00\n",
            "ROC AUC score is not defined for a single class.\n",
            "---------------------------------------------------------\n",
            "\n"
          ]
        }
      ],
      "source": [
        "summarize_performance(test_loss, test_acc, clean_attack_test_loss, clean_attack_adv_test_accuracy, retrained_test_loss, retrained_accuracy,\n",
        "                          clean_misclassified_images, adv_misclassified_images, retrained_misclassified_images,\n",
        "                          clean_misclassified_labels, adv_misclassified_labels, retrained_misclassified_labels,\n",
        "                          clean_misclassified_preds,adv_misclassified_preds, retrained_misclassified_preds)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "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": 5
}