ML-Adversarial/adversarial_training_cv.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ug77AoFEhHmo"
   },
   "source": [
    "# Adversarial training for computer vision models\n",
    "\n",
    "<a href=\"https://colab.research.google.com/drive/1RL1A5R7gxcsovVFbejU-Ih81yODaBfL6\" target=\"_blank\">\n",
    "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\">\n",
    "</a>\n",
    "\n",
    "Return to the [castle](https://github.com/Nkluge-correa/teeny-tiny_castle).\n",
    "\n",
    "Adversarial machine learning is a specialized area within machine learning dedicated to creating algorithms and techniques capable of resisting and effectively responding to adversarial attacks. This field aims to enhance the robustness of models by understanding potential vulnerabilities and developing strategies to mitigate risks posed by malicious inputs and adversarial intent.\n",
    "\n",
    "In this notebook, we will create an adversarial dataset to train and test the robustness of two different models. To create our adversarial dataset, we will use the **Fast Gradient Sign Method** ([FGSM](https://arxiv.org/abs/1412.6572)) and the `CIFAR-10` dataset. If you do not want to [train this model](https://github.com/Nkluge-correa/TeenyTinyCastle/blob/master/ML-Adversarial/adversarial_training_cv.ipynb), remember that all our trained models are in the Hub, and you can download them directly. 🤗"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000,
     "referenced_widgets": [
      "292aaa49b0424c0ea87bb6e9a2cb1467",
      "18bda33398e24e1ead413469773cb9de",
      "2f3bd0c1bfa24de29ac9909f1dda8196",
      "05d8a46256fb49829e4de623a6ad795e",
      "bdf42549aed648fab9fdee634948a89a",
      "1b8a60fdb3964f45bca761609c12a480",
      "d4f3b28ef8024717b2bd98b875d444bf",
      "b725476b9402461498625da9485b864a",
      "6e78874e3a484b58bf9c4b5df53df9df",
      "6f0d5534e274492ea37007aa6c6a4e01",
      "e585e3e6927a4235af07616a61c6cfb9"
     ]
    },
    "id": "zJzHHTGa2eQ7",
    "outputId": "3f322c17-1482-4816-dbae-dcd1ae809a3c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "292aaa49b0424c0ea87bb6e9a2cb1467",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Fetching 12 files:   0%|          | 0/12 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in save file, so the model was *not* compiled. Compile it manually.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 32, 32, 32)        896       \n",
      "                                                                 \n",
      " batch_normalization (Batch  (None, 32, 32, 32)        128       \n",
      " Normalization)                                                  \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 32, 32, 32)        9248      \n",
      "                                                                 \n",
      " batch_normalization_1 (Bat  (None, 32, 32, 32)        128       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2  (None, 16, 16, 32)        0         \n",
      " D)                                                              \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 16, 16, 32)        0         \n",
      "                                                                 \n",
      " conv2d_2 (Conv2D)           (None, 16, 16, 64)        18496     \n",
      "                                                                 \n",
      " batch_normalization_2 (Bat  (None, 16, 16, 64)        256       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " conv2d_3 (Conv2D)           (None, 16, 16, 64)        36928     \n",
      "                                                                 \n",
      " batch_normalization_3 (Bat  (None, 16, 16, 64)        256       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPoolin  (None, 8, 8, 64)          0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " conv2d_4 (Conv2D)           (None, 8, 8, 128)         73856     \n",
      "                                                                 \n",
      " batch_normalization_4 (Bat  (None, 8, 8, 128)         512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " conv2d_5 (Conv2D)           (None, 8, 8, 128)         147584    \n",
      "                                                                 \n",
      " batch_normalization_5 (Bat  (None, 8, 8, 128)         512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPoolin  (None, 4, 4, 128)         0         \n",
      " g2D)                                                            \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 4, 4, 128)         0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 2048)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               262272    \n",
      "                                                                 \n",
      " batch_normalization_6 (Bat  (None, 128)               512       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 10)                1290      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 552874 (2.11 MB)\n",
      "Trainable params: 551722 (2.10 MB)\n",
      "Non-trainable params: 1152 (4.50 KB)\n",
      "_________________________________________________________________\n",
      "Loss: 0.45.\n",
      "Accuracy: 86.03 %.\n"
     ]
    }
   ],
   "source": [
    "!pip install huggingface_hub[\"tensorflow\"] -q\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "from huggingface_hub import from_pretrained_keras\n",
    "\n",
    "# Download the CIFAR-10 dataset\n",
    "(x_train, y_train), (x_test, y_test)  = tf.keras.datasets.cifar10.load_data()\n",
    "\n",
    "class_names = ['Airplane', 'Automobile', 'Bird', 'Cat', 'Deer',\n",
    "               'Dog', 'Frog', 'Horse', 'Ship', 'Truck']\n",
    "\n",
    "plt.figure(figsize=[10, 10])\n",
    "for i in range(25):\n",
    "    plt.subplot(5, 5, i+1)\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "    plt.grid(False)\n",
    "    plt.imshow(x_test[i], cmap=plt.cm.binary)\n",
    "    plt.xlabel(class_names[y_test[i][0]])\n",
    "\n",
    "plt.show()\n",
    "\n",
    "# Load the model from the Hub\n",
    "model = from_pretrained_keras(\"AiresPucrs/Cifar-CNN\")\n",
    "model.compile(\n",
    "    loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "    metrics=['categorical_accuracy']\n",
    "    )\n",
    "x_train = x_train.astype('float32')\n",
    "x_train = x_train / 255.\n",
    "y_train = tf.keras.utils.to_categorical(y_train, 10)\n",
    "x_test = x_test.astype('float32')\n",
    "x_test = x_test / 255.\n",
    "y_test = tf.keras.utils.to_categorical(y_test, 10)\n",
    "test_loss_score, test_acc_score = model.evaluate(x_test, y_test, verbose=0)\n",
    "model.summary()\n",
    "print(f'Loss: {round(test_loss_score, 2)}.')\n",
    "print(f'Accuracy: {round(test_acc_score * 100, 2)} %.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oOmYGzZChHmr"
   },
   "source": [
    "Now, let us write a simple function to return labels, predictions, and images."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 473
    },
    "id": "vKKJGDiAhHmr",
    "outputId": "99f1e2e4-9ff7-47e6-f13c-f90c013308f6"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def predict_image(x_test, y_test, sample):\n",
    "    img_tensor = np.expand_dims(x_test[sample], axis=0)\n",
    "    pred = [round(num, 3) for num in list(model.predict(img_tensor, verbose=0)[0])]\n",
    "    plt.figure()\n",
    "    plt.imshow(x_test[sample])\n",
    "    plt.title(\n",
    "        f'Prediction --> {class_names[np.array(pred).argmax()]} ({round(max(pred) * 100, 2)} % confidence)\\nActually is a {class_names[y_test[sample].argmax()]}')\n",
    "    return plt.show()\n",
    "\n",
    "sample = np.random.randint(0, 9999)\n",
    "predict_image(x_test, y_test, sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aetuNfikhHms"
   },
   "source": [
    "Now that we have chosen a sample, we will use the [`tf.keras.losses.CategoricalCrossentropy`](https://www.tensorflow.org/api_docs/python/tf/keras/losses/CategoricalCrossentropy) class to create a loss function. We will also record the class index of this specific sample.\n",
    "\n",
    "We will use this loss function (rather than serving as a supervision signal for updating the parameters of a neural network), in combination with the class [`tf.GradientTape()`](https://www.tensorflow.org/api_docs/python/tf/GradientTape) to create an image (perturbation) that generates the largest possible loss (here we are moving the gradient in the direction in which the loss increases, i.e., gradient ascent) to a specific class.\n",
    "\n",
    "And that is what the **[Fast Sign Gradient Method](https://arxiv.org/abs/1412.6572)** is all about. In essence, FGSM is to add the noise (not random noise) whose direction is the same as the gradient of the cost function concerning the data.\n",
    "\n",
    "All this will be encapsulated by a function called `create_adversarial_pattern`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 470
    },
    "id": "i2cKJSxDhHms",
    "outputId": "c0922f2d-77cd-49e3-e3f7-6cf48069a5e5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 23ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img_tensor = np.expand_dims(x_test[sample], axis=0)\n",
    "image_probs = model.predict(img_tensor)\n",
    "class_index = np.argmax(image_probs[0])\n",
    "\n",
    "loss_object = tf.keras.losses.CategoricalCrossentropy()\n",
    "\n",
    "def create_adversarial_pattern(input_image, input_label):\n",
    "    \"\"\"\n",
    "    Computes the adversarial perturbation of an input image using the\n",
    "    given model and loss function.\n",
    "\n",
    "    Parameters:\n",
    "        input_image (tf.Tensor): The input image for which to\n",
    "            compute the adversarial perturbation.\n",
    "        input_label (tf.Tensor): The true label of the input image.\n",
    "\n",
    "    Returns:\n",
    "        tf.Tensor: The signed gradient of the loss with respect to\n",
    "        the input image, which represents the direction in which\n",
    "        the input image should be perturbed in order to increase the loss.\n",
    "    \"\"\"\n",
    "    with tf.GradientTape() as tape:\n",
    "        tape.watch(input_image)\n",
    "        prediction = model(input_image)\n",
    "        loss = loss_object(input_label, prediction)\n",
    "    gradient = tape.gradient(loss, input_image)\n",
    "    signed_grad = tf.sign(gradient)\n",
    "    return signed_grad\n",
    "\n",
    "label = tf.one_hot(class_index, image_probs.shape[-1])\n",
    "label = tf.reshape(label, (1, image_probs.shape[-1]))\n",
    "\n",
    "perturbations = create_adversarial_pattern(\n",
    "    tf.convert_to_tensor(img_tensor, dtype=tf.float32), label)\n",
    "adversarial_perturbation = (perturbations[0] * 0.5 + 0.5).numpy()\n",
    "plt.figure()\n",
    "plt.imshow(adversarial_perturbation)\n",
    "plt.title(f'Adversarial Perturbation for class: {class_names[class_index]}')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7Z1BHXKVhHms"
   },
   "source": [
    "Now, we combine our perturbation with the original image to generate a misclassification. For this explanation, we will only use an  $\\epsilon$ of 0.01 since we are interested in developing \" imperceptible \" attacks on the human observer. Since CIFAR 10 images are low resolution, even a tiny $\\epsilon$ can be perceived by the careful eye."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 523
    },
    "id": "Tm8qJlhAhHms",
    "outputId": "1847a3cf-e33b-4006-d0b3-a5a0dc559074"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "\n",
    "def display_images(img_tensors, descriptions):\n",
    "\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(10, 5))\n",
    "\n",
    "    for i, (img_tensor, description) in enumerate(zip(img_tensors, descriptions)):\n",
    "        # Model prediction\n",
    "        image_probs = model.predict(tf.convert_to_tensor(img_tensor, dtype=tf.float32), verbose=0)\n",
    "        class_confidence = image_probs.max()\n",
    "        class_index = np.argmax(image_probs[0])\n",
    "\n",
    "        axes[i].imshow(img_tensor[0])\n",
    "        axes[i].set_title(f'{description} \\n {class_names[class_index]}: {round(class_confidence * 100, 2)}% confidence')\n",
    "        axes[i].axis('off')\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "epsilons = [0, 0.01]\n",
    "descriptions = ['Input' if eps == 0 else f'Epsilon = {eps:.3f}' for eps in epsilons]\n",
    "\n",
    "adv_images = [tf.clip_by_value(img_tensor + eps * perturbations, -1, 1) for eps in epsilons]\n",
    "display_images(adv_images, descriptions)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hLrpkCB7hHms"
   },
   "source": [
    "As you can see, an $\\epsilon$ of $0.01$ is enough to fool this model by a significant margin.\n",
    "\n",
    "In this notebook, we will use a straightforward strategy for creating our adversarial dataset. We will convert ~40% of our original dataset (randomly picked) into adversarial examples and then add those examples to the original dataset, thus creating an adversarial training dataset. This can take considerable time, so skip to the next cell if you want to skip this part and directly download the dataset created for this example."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Xpm_re886XM4"
   },
   "outputs": [],
   "source": [
    "from tqdm import tqdm\n",
    "\n",
    "# Initialize loss and epsilon\n",
    "loss_object = tf.keras.losses.CategoricalCrossentropy()\n",
    "eps = 0.01\n",
    "adversarial_x = []\n",
    "adversarial_y = []\n",
    "\n",
    "# Concatenate the training and test datasets\n",
    "X = np.concatenate((x_train, x_test), axis=0)\n",
    "Y = np.concatenate((y_train, y_test), axis=0)\n",
    "\n",
    "# Loop through the dataset with a progress bar\n",
    "for i in tqdm(range(len(X)), desc=\"Generating Adversarial Samples\"):\n",
    "\n",
    "    # Use ~40% of the dataset to create adversarial samples\n",
    "    if np.random.uniform(0, 1) >= 0.6:\n",
    "\n",
    "        # Get info on the sample\n",
    "        img_tensor = np.expand_dims(X[i], axis=0)\n",
    "        image_probs = model.predict(img_tensor, verbose=0)\n",
    "        class_index = np.argmax(image_probs[0])\n",
    "\n",
    "        # Target the specific label\n",
    "        label = tf.one_hot(class_index, image_probs.shape[-1])\n",
    "        label = tf.reshape(label, (1, image_probs.shape[-1]))\n",
    "\n",
    "        # Create perturbation\n",
    "        perturbations = create_adversarial_pattern(\n",
    "            tf.convert_to_tensor(img_tensor, dtype=tf.float32), label)\n",
    "        adversarial_perturbation = (perturbations[0] * 0.5 + 0.5).numpy()\n",
    "\n",
    "        # Create adversarial sample\n",
    "        adversarial_sample = tf.clip_by_value(img_tensor + eps * perturbations, -1, 1)\n",
    "\n",
    "        # Append to adversarial dataset\n",
    "        adversarial_x.append(adversarial_sample[0].numpy())\n",
    "        adversarial_y.append(Y[i])\n",
    "\n",
    "# Convert lists to numpy arrays\n",
    "adversarial_x = np.array(adversarial_x).astype('float32')\n",
    "adversarial_y = np.array(adversarial_y)\n",
    "\n",
    "# Save adversarial data to a file\n",
    "with open('adversarial_data_CIFAR_10.npy', 'wb') as f:\n",
    "    np.save(f, adversarial_x)\n",
    "    np.save(f, adversarial_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4LNSmvLlhHms"
   },
   "source": [
    "Now, we will divide our adversarial dataset into two parts: one to train a new model and one to test the robustness of both models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 49,
     "referenced_widgets": [
      "d1e86840ef0b478e9cfd6f062b005aa1",
      "acd1941bb68a4db18edf77b7f5bc683a",
      "2b57e77bc54545b0af358403a5af95e2",
      "ba769a0975ee49408b4ad0afc0c968a8",
      "2301d88a7d1c48b6a9694347f05c7597",
      "daa48b3d2b774e05b2a382c756477027",
      "ed653dd8858640f9a0978b48daaba55c",
      "2d642f75647f48958527ada0f9675fbc",
      "0c3d76b10ca64034882d1b903b2fb9c7",
      "a6c10ef70d2d40e199c5bf8abd2c9f16",
      "1b8e02ebfb084b97a7f5a345eec3f3dc"
     ]
    },
    "id": "kiLQO1qohHmt",
    "outputId": "ae4f764a-6a7e-4ec6-e371-90c3917bca38"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1e86840ef0b478e9cfd6f062b005aa1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "adversarial-data-CIFAR-10.npy:   0%|          | 0.00/297M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from huggingface_hub import hf_hub_download\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Download the adversarial dataset from the Hub\n",
    "file_path = hf_hub_download(\n",
    "    repo_id=\"AiresPucrs/Cifar-CNN-with-adversarial-training\",\n",
    "    filename=\"adversarial-data-CIFAR-10.npy\",\n",
    "    local_dir=\"./\",\n",
    "    repo_type=\"model\"\n",
    ")\n",
    "\n",
    "# Load the adversarial data from the file\n",
    "with open(file_path, 'rb') as fp:\n",
    "    adversarial_x = np.load(fp)\n",
    "    adversarial_y = np.load(fp)\n",
    "\n",
    "# Split adversarial data into training and test sets\n",
    "adversarial_x_training = adversarial_x[:20000]\n",
    "adversarial_x_test = adversarial_x[20000:]\n",
    "\n",
    "adversarial_y_training = adversarial_y[:20000]\n",
    "adversarial_y_test = adversarial_y[20000:]\n",
    "\n",
    "# Concatenate adversarial training data with the original dataset\n",
    "X = np.concatenate((X, adversarial_x_training), axis=0)\n",
    "Y = np.concatenate((Y, adversarial_y_training), axis=0)\n",
    "\n",
    "# Split the combined dataset into training and test sets\n",
    "x_train, x_test, y_train, y_test = train_test_split(\n",
    "    X, Y, test_size=0.15, random_state=42\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MtbZzb83hHmt"
   },
   "source": [
    "We will now train the \"same model\" ([same architecture and hyperparemeters]()) with this adversarial dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "HKpD3HzrhHmt"
   },
   "outputs": [],
   "source": [
    "model = tf.keras.models.Sequential()\n",
    "\n",
    "model.add(tf.keras.layers.Conv2D(32, (3, 3), padding='same',\n",
    "          activation='relu', input_shape=(32, 32, 3)))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.Conv2D(32, (3, 3), padding='same', activation='relu'))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(tf.keras.layers.Dropout(0.3))\n",
    "\n",
    "model.add(tf.keras.layers.Conv2D(64, (3, 3), padding='same', activation='relu'))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.Conv2D(64, (3, 3), padding='same', activation='relu'))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(tf.keras.layers.Dropout(0.5))\n",
    "\n",
    "model.add(tf.keras.layers.Conv2D(128, (3, 3), padding='same', activation='relu'))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.Conv2D(128, (3, 3), padding='same', activation='relu'))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(tf.keras.layers.Dropout(0.5))\n",
    "\n",
    "model.add(tf.keras.layers.Flatten())\n",
    "model.add(tf.keras.layers.Dense(128, activation='relu'))\n",
    "model.add(tf.keras.layers.BatchNormalization())\n",
    "model.add(tf.keras.layers.Dropout(0.5))\n",
    "model.add(tf.keras.layers.Dense(10, activation='softmax'))\n",
    "\n",
    "model.compile(optimizer='adam',\n",
    "              loss='categorical_crossentropy',\n",
    "              metrics=['categorical_accuracy'])\n",
    "\n",
    "print(\"Version: \", tf.__version__)\n",
    "print(\"Eager mode: \", tf.executing_eagerly())\n",
    "print(\"GPU is\", \"available\" if tf.config.list_physical_devices('GPU') else \"NOT AVAILABLE\")\n",
    "\n",
    "model.summary()\n",
    "\n",
    "callbacks = [tf.keras.callbacks.ModelCheckpoint(\"Cifar-CNN-with-adversarial-training.keras\",\n",
    "                                                save_best_only=True),\n",
    "            tf.keras.callbacks.EarlyStopping(monitor=\"val_loss\",\n",
    "                                            patience=3,\n",
    "                                            verbose=1,\n",
    "                                            mode=\"auto\",\n",
    "                                            baseline=None,\n",
    "                                            restore_best_weights=True)]\n",
    "\n",
    "model.fit(x_train, y_train,\n",
    "    validation_split=0.2,\n",
    "    batch_size=128,\n",
    "    epochs=80,\n",
    "    verbose=1,\n",
    "    callbacks=callbacks)\n",
    "\n",
    "test_loss_score, test_acc_score = model.evaluate(x_test, y_test)\n",
    "\n",
    "print(f'Final Loss: {round(test_loss_score, 2)}.')\n",
    "print(f'Final Performance: {round(test_acc_score * 100, 2)} %.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "e7_u1PpbhHmt"
   },
   "source": [
    "For comparison reasons, we will define our models' robustness score as their accuracy against the test portion of our adversarial dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 371,
     "referenced_widgets": [
      "172cee0db0774da48180202b053afa50",
      "c44c8fc5259d41368d2673d877278173",
      "bbc8259af56a419f8795768be20c6002",
      "8821a6195bd54ee88bdcf855eab3afd6",
      "d50c4186da544f2d9cb5599b282d7be6",
      "3c50df80d1ba4fbabe1e227700fbb38e",
      "08eb590f4b5d4058979b28edbf702acf",
      "434271e589ab4080be65e79a55077fe7",
      "6787a341e5034d2c900a97fffd47985e",
      "2d58aac1dd564f9fa4e4c22b117efcb0",
      "e9effd6ddfbe4e3ca944a684011f72b5",
      "f8d6f87f0a8b410e986bf0ef7a0e0b70",
      "102c974512aa49a4804d36355fb53937",
      "fbbab687d3da4411bd6a0fb7a3f23399",
      "22577e9509b349a98195366eb2162dd1",
      "a3e3ed5ef7f849da8f91dc0f09c98281",
      "e8c66c0a2df2443297bfc583129ad133",
      "1ebbf74e10b7476abbfe1dd8e0857827",
      "1ab4519ddc304e79a79220bbd3ce75b0",
      "338b37d236b048518b8e5aa2a352d3dd",
      "0f41b766ffdb470880bfce1d7bc33d3a",
      "fe0e840547f247cf908a721893f4816d"
     ]
    },
    "id": "yegvPbeBAAGM",
    "outputId": "031ea7b1-e380-4a3e-f863-a20d030b3305"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "172cee0db0774da48180202b053afa50",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Fetching 12 files:   0%|          | 0/12 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in save file, so the model was *not* compiled. Compile it manually.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "# Model Evaluation: Base Model Against Adversaries\n",
      "\n",
      "--------------------------------------------------\n",
      "\n",
      "- Robustness Score: 34.19 %.\n",
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f8d6f87f0a8b410e986bf0ef7a0e0b70",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Fetching 10 files:   0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in save file, so the model was *not* compiled. Compile it manually.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "# Model Evaluation: Adversarially Trained Model Against Adversaries\n",
      "\n",
      "--------------------------------------------------\n",
      "\n",
      "- Robustness Score: 90.43 %.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from huggingface_hub import from_pretrained_keras\n",
    "\n",
    "def evaluate_model(repo_id, test_data, test_labels, description):\n",
    "    # Load and compile the model\n",
    "    model = from_pretrained_keras(repo_id)\n",
    "    model.compile(\n",
    "        loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "        metrics=['categorical_accuracy']\n",
    "    )\n",
    "\n",
    "    # Evaluate the model\n",
    "    print(f'\\n# Model Evaluation: {description}\\n\\n{\"-\" * 50}\\n')\n",
    "    _, test_acc_score = model.evaluate(test_data, test_labels, verbose=0)\n",
    "    print(f'- Robustness Score: {round(test_acc_score * 100, 2)} %.\\n')\n",
    "\n",
    "# Evaluate the base model\n",
    "evaluate_model(\n",
    "    repo_id=\"AiresPucrs/Cifar-CNN\",\n",
    "    test_data=adversarial_x_test,\n",
    "    test_labels=adversarial_y_test,\n",
    "    description=\"Base Model Against Adversaries\"\n",
    ")\n",
    "\n",
    "# Evaluate the adversarially trained model\n",
    "evaluate_model(\n",
    "    repo_id=\"AiresPucrs/Cifar-CNN-with-adversarial-training\",\n",
    "    test_data=adversarial_x_test,\n",
    "    test_labels=adversarial_y_test,\n",
    "    description=\"Adversarially Trained Model Against Adversaries\"\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ftzJO1hrhHmt"
   },
   "source": [
    "As we can see, the model trained with a combination of normal and perturbed data receives a robustness score of around 90%! Adversarial training is a technique that can significantly enhance the robustness of AI models against various adversarial attacks. AI developers can create models that better generalize against attacks by incorporating adversarial examples into training. 🙃\n",
    "\n",
    "---\n",
    "\n",
    "Return to the [castle](https://github.com/Nkluge-correa/teeny-tiny_castle)."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3.9.13 64-bit",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "aca09746cf57686f00a55ae76e987247ecfb5dd0b3b2e2474d8dbbf0c5e3377e"
   }
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "05d8a46256fb49829e4de623a6ad795e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6f0d5534e274492ea37007aa6c6a4e01",
      "placeholder": "​",
      "style": "IPY_MODEL_e585e3e6927a4235af07616a61c6cfb9",
      "value": " 12/12 [00:00&lt;00:00, 572.20it/s]"
     }
    },
    "08eb590f4b5d4058979b28edbf702acf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "0c3d76b10ca64034882d1b903b2fb9c7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "0f41b766ffdb470880bfce1d7bc33d3a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "102c974512aa49a4804d36355fb53937": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e8c66c0a2df2443297bfc583129ad133",
      "placeholder": "​",
      "style": "IPY_MODEL_1ebbf74e10b7476abbfe1dd8e0857827",
      "value": "Fetching 10 files: 100%"
     }
    },
    "172cee0db0774da48180202b053afa50": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_c44c8fc5259d41368d2673d877278173",
       "IPY_MODEL_bbc8259af56a419f8795768be20c6002",
       "IPY_MODEL_8821a6195bd54ee88bdcf855eab3afd6"
      ],
      "layout": "IPY_MODEL_d50c4186da544f2d9cb5599b282d7be6"
     }
    },
    "18bda33398e24e1ead413469773cb9de": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1b8a60fdb3964f45bca761609c12a480",
      "placeholder": "​",
      "style": "IPY_MODEL_d4f3b28ef8024717b2bd98b875d444bf",
      "value": "Fetching 12 files: 100%"
     }
    },
    "1ab4519ddc304e79a79220bbd3ce75b0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1b8a60fdb3964f45bca761609c12a480": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1b8e02ebfb084b97a7f5a345eec3f3dc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "1ebbf74e10b7476abbfe1dd8e0857827": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "22577e9509b349a98195366eb2162dd1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0f41b766ffdb470880bfce1d7bc33d3a",
      "placeholder": "​",
      "style": "IPY_MODEL_fe0e840547f247cf908a721893f4816d",
      "value": " 10/10 [00:00&lt;00:00, 299.55it/s]"
     }
    },
    "2301d88a7d1c48b6a9694347f05c7597": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "292aaa49b0424c0ea87bb6e9a2cb1467": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_18bda33398e24e1ead413469773cb9de",
       "IPY_MODEL_2f3bd0c1bfa24de29ac9909f1dda8196",
       "IPY_MODEL_05d8a46256fb49829e4de623a6ad795e"
      ],
      "layout": "IPY_MODEL_bdf42549aed648fab9fdee634948a89a"
     }
    },
    "2b57e77bc54545b0af358403a5af95e2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2d642f75647f48958527ada0f9675fbc",
      "max": 296685904,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_0c3d76b10ca64034882d1b903b2fb9c7",
      "value": 296685904
     }
    },
    "2d58aac1dd564f9fa4e4c22b117efcb0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2d642f75647f48958527ada0f9675fbc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2f3bd0c1bfa24de29ac9909f1dda8196": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b725476b9402461498625da9485b864a",
      "max": 12,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_6e78874e3a484b58bf9c4b5df53df9df",
      "value": 12
     }
    },
    "338b37d236b048518b8e5aa2a352d3dd": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "3c50df80d1ba4fbabe1e227700fbb38e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "434271e589ab4080be65e79a55077fe7": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "6787a341e5034d2c900a97fffd47985e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "6e78874e3a484b58bf9c4b5df53df9df": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "6f0d5534e274492ea37007aa6c6a4e01": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "8821a6195bd54ee88bdcf855eab3afd6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2d58aac1dd564f9fa4e4c22b117efcb0",
      "placeholder": "​",
      "style": "IPY_MODEL_e9effd6ddfbe4e3ca944a684011f72b5",
      "value": " 12/12 [00:00&lt;00:00, 225.87it/s]"
     }
    },
    "a3e3ed5ef7f849da8f91dc0f09c98281": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a6c10ef70d2d40e199c5bf8abd2c9f16": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "acd1941bb68a4db18edf77b7f5bc683a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_daa48b3d2b774e05b2a382c756477027",
      "placeholder": "​",
      "style": "IPY_MODEL_ed653dd8858640f9a0978b48daaba55c",
      "value": "adversarial-data-CIFAR-10.npy: 100%"
     }
    },
    "b725476b9402461498625da9485b864a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ba769a0975ee49408b4ad0afc0c968a8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_a6c10ef70d2d40e199c5bf8abd2c9f16",
      "placeholder": "​",
      "style": "IPY_MODEL_1b8e02ebfb084b97a7f5a345eec3f3dc",
      "value": " 297M/297M [00:04&lt;00:00, 81.7MB/s]"
     }
    },
    "bbc8259af56a419f8795768be20c6002": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_434271e589ab4080be65e79a55077fe7",
      "max": 12,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_6787a341e5034d2c900a97fffd47985e",
      "value": 12
     }
    },
    "bdf42549aed648fab9fdee634948a89a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c44c8fc5259d41368d2673d877278173": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_3c50df80d1ba4fbabe1e227700fbb38e",
      "placeholder": "​",
      "style": "IPY_MODEL_08eb590f4b5d4058979b28edbf702acf",
      "value": "Fetching 12 files: 100%"
     }
    },
    "d1e86840ef0b478e9cfd6f062b005aa1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_acd1941bb68a4db18edf77b7f5bc683a",
       "IPY_MODEL_2b57e77bc54545b0af358403a5af95e2",
       "IPY_MODEL_ba769a0975ee49408b4ad0afc0c968a8"
      ],
      "layout": "IPY_MODEL_2301d88a7d1c48b6a9694347f05c7597"
     }
    },
    "d4f3b28ef8024717b2bd98b875d444bf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "d50c4186da544f2d9cb5599b282d7be6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "daa48b3d2b774e05b2a382c756477027": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e585e3e6927a4235af07616a61c6cfb9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "e8c66c0a2df2443297bfc583129ad133": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e9effd6ddfbe4e3ca944a684011f72b5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "ed653dd8858640f9a0978b48daaba55c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "f8d6f87f0a8b410e986bf0ef7a0e0b70": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_102c974512aa49a4804d36355fb53937",
       "IPY_MODEL_fbbab687d3da4411bd6a0fb7a3f23399",
       "IPY_MODEL_22577e9509b349a98195366eb2162dd1"
      ],
      "layout": "IPY_MODEL_a3e3ed5ef7f849da8f91dc0f09c98281"
     }
    },
    "fbbab687d3da4411bd6a0fb7a3f23399": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1ab4519ddc304e79a79220bbd3ce75b0",
      "max": 10,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_338b37d236b048518b8e5aa2a352d3dd",
      "value": 10
     }
    },
    "fe0e840547f247cf908a721893f4816d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}