Tasks/task1_5/task_01.5_empty.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# `Практикум по программированию на языке Python`\n",
    "\n",
    "## `Задание 1.5. Классификация изображений`.\n",
    "\n",
    "#### Фамилия, имя: `ВАШЕ ИМЯ`\n",
    "\n",
    "Дата выдачи: <span style=\"color:red\">__20 февраля__</span>.\n",
    "\n",
    "Мягкий дедлайн: <span style=\"color:red\">__2 марта 5:00__</span>.\n",
    "\n",
    "Стоимость: __10 баллов__ (основная часть заданий) + __3 балла__ (дополнительные задания).\n",
    "\n",
    "<span style=\"color:red\">__В ноутбуке все клетки должны выполняться без ошибок при последовательном их выполнении.__</span>\n",
    "\n",
    "#### `Москва, 2025`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "import random\n",
    "import numpy as np\n",
    "from typing import Optional, Tuple, Union, List\n",
    "\n",
    "import torch\n",
    "from torch import nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "import torchvision\n",
    "from torchvision import transforms as T\n",
    "\n",
    "import matplotlib_inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib inline\n",
    "matplotlib_inline.backend_inline.set_matplotlib_formats('pdf', 'svg')\n",
    "\n",
    "torch.backends.cudnn.benchmark = True\n",
    "torch.use_deterministic_algorithms(False)\n",
    "\n",
    "torch.backends.cudnn.allow_tf32 = True\n",
    "torch.backends.cuda.matmul.allow_tf32 = True\n",
    "\n",
    "torch.autograd.profiler.profile(False)\n",
    "torch.autograd.profiler.emit_nvtx(False)\n",
    "torch.autograd.set_detect_anomaly(False)\n",
    "\n",
    "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def set_global_seed(seed: int) -> None:\n",
    "    \"\"\"Set global seed for reproducibility.\n",
    "    :param int seed: Seed to be set\n",
    "    \"\"\"\n",
    "    random.seed(seed)\n",
    "    np.random.seed(seed)\n",
    "    \n",
    "    torch.manual_seed(seed)\n",
    "    torch.cuda.manual_seed(seed)\n",
    "    torch.cuda.manual_seed_all(seed)\n",
    "\n",
    "def check_numel(module: torch.nn.Module, params_numel: int, buffers_numel: Optional[int] = None) -> None:\n",
    "    \"\"\"Check whether module has correct number of parameters and buffers\n",
    "    :param torch.nn.Module module: Target model\n",
    "    :param int params_numel: Target number of parameters\n",
    "    :param Optional[int] buffers_numel: Target number of buffers\n",
    "    :rtype:\n",
    "    \"\"\"\n",
    "    numel = sum(param.numel() for param in module.parameters())\n",
    "    assert numel == params_numel, f'For params numel != correct numel: {numel} vs {params_numel}'\n",
    "    \n",
    "    if buffers_numel is not None:\n",
    "        numel = sum(param.numel() for param in module.buffers())\n",
    "        assert numel == buffers_numel, f'For buffers numel != correct numel: {numel} vs {buffers_numel}'\n",
    "\n",
    "set_global_seed(42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `Задание 1. Загрузка данных (0.5 балла)`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "В этой домашке работа с данными будет максимально простой. Мы скачаем `MNIST` с помощью [стандартных средств](https://pytorch.org/vision/0.8/datasets.html#mnist) торча. Посмотреть список других доступных датасетов можно [здесь](https://pytorch.org/vision/0.8/datasets.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_mnist(train, transform=None):    \n",
    "    if transform is None:\n",
    "        transform = T.ToTensor()\n",
    "    return torchvision.datasets.MNIST(\n",
    "        root='./data',\n",
    "        train=train,\n",
    "        transform=transform,\n",
    "        download=True\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Пиксели изображений принимают значения [0,1]. Скачайте датасет и посчитайте выборочное среднее и выборочное стандартное отклонение для значения пикселя. Т.е. должно получиться число `mean` для среднего, и число `std` для стандартного отклонения.\n",
    "\n",
    "*Подсказка:* Подумайте, на какой части датасета нужно считать эти статистики (обучение или валидация), чтобы предотвратить утечку данных.\n",
    "\n",
    "*Подсказка:* У торч датасета реализован метод `__getitem__`, т.е. его можно индексировать `dataset[i]`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOUR CODE HERE\n",
    "# mean =  ...\n",
    "# std = ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Полученные значения можно применить для нормализации изображений, когда будем подавать их в нейросеть.\n",
    "\n",
    "Нормализация данных в машинном обучении -- это процесс приведения числовых данных к единому масштабу, чтобы они находились в определённом диапазоне (например, от 0 до 1 или со средним 0 и стандартным отклонением 1). Это помогает улучшить скорость и стабильность обучения моделей, особенно когда признаки имеют разный масштаб."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "norm_transform = T.Compose([\n",
    "    T.ToTensor(),\n",
    "    T.Normalize(mean, std)\n",
    "])\n",
    "\n",
    "train_mnist = load_mnist(True, norm_transform)\n",
    "val_mnist = load_mnist(False, norm_transform)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Во время обучения данные подаются в нейросеть батчами.\n",
    "\n",
    "Обычно `batch_size` выбирается так, чтобы во время обучения получалась меньше дисперсия градиента (вы уже знакомы с sgd), также часто `batch_size` выбирается максимально возможным по ресурсам.\n",
    "\n",
    "Для этой задачи нам хватит `batch_size = 64` во время обучения.\n",
    "\n",
    "**Ответьте на вопрос:** Почему в большинстве задач для валидации и тестирования можно выбирать `batch_size` больше? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.utils.data import DataLoader\n",
    "\n",
    "batch_size_train = 64\n",
    "batch_size_test = 1000\n",
    "\n",
    "train_mnist_loader = DataLoader(\n",
    "    dataset=train_mnist,\n",
    "    batch_size=batch_size_train,\n",
    "    shuffle=True,\n",
    "    drop_last=True\n",
    ")\n",
    "\n",
    "val_mnist_loader = torch.utils.data.DataLoader(\n",
    "    dataset=val_mnist,\n",
    "    batch_size=batch_size_test,\n",
    "    shuffle=False,\n",
    "    drop_last=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([64, 1, 28, 28]), torch.Size([64]))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "images, labels = next(iter(train_mnist_loader))\n",
    "images.shape, labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "i = 3\n",
    "plt.title(f'digit {labels[i]}')\n",
    "plt.imshow(images[i, 0], cmap='gray')\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `Задание 2. Цикл обучения (1.5 балла)`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ниже предоставлен базовый вариант цикла обучения, представленный тремя функциями: `train_epoch`, `val_epoch`, `train_val`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_epoch(\n",
    "    network,\n",
    "    train_loader,\n",
    "    criterion,\n",
    "    optimizer,\n",
    "):\n",
    "    \"\"\"train `network` with `optimizer` for one epoch with data from `train_loader` to minimize `criterion`\"\"\"\n",
    "    \n",
    "    network.train()  # switch network submodules to train mode, e.g. it influences on batch-norm, dropout\n",
    "    for images, labels in train_loader:\n",
    "        # 0. распакавываем данные на нужное устройство\n",
    "        images, labels = ???.to(device), ???.to(device) # YOUR CODE HERE\n",
    "\n",
    "        # 1. сбрасываем градиенты предыдущего батча\n",
    "        ???.zero_grad() # YOUR CODE HERE\n",
    "\n",
    "        # 2. прогоняем данные через нашу нейросеть\n",
    "        logits = network(???) # YOUR CODE HERE\n",
    "\n",
    "        # 3. считаем loss\n",
    "        loss = criterion(logits, ???, reduction='mean') # YOUR CODE HERE\n",
    "\n",
    "        # 4. считаем градиенты\n",
    "        ???.backward() # YOUR CODE HERE\n",
    "\n",
    "        # 5. обновляем параметры\n",
    "        ???.step() # YOUR CODE HERE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "@torch.no_grad()\n",
    "def val_epoch(\n",
    "    network,\n",
    "    val_loader,\n",
    "    criterion,\n",
    "):\n",
    "    \"\"\"calculate loss and accuracy on validation data\"\"\"\n",
    "    val_loss = 0\n",
    "    correct = 0\n",
    "\n",
    "    network.eval()  # switch network submodules to test mode\n",
    "    for images, labels in val_loader:\n",
    "        # 0. распакавываем данные на нужное устройство\n",
    "        images, labels = ???.to(device), ???.to(device) # YOUR CODE HERE\n",
    "\n",
    "        # 1. прогоняем данные через нашу нейросеть\n",
    "        logits = network(???) # YOUR CODE HERE\n",
    "\n",
    "         # 2. получаем предсказание\n",
    "        pred = torch.argmax(???, dim=1) # YOUR CODE HERE\n",
    "\n",
    "        # 3. логгируем лосс и accuracy\n",
    "        val_loss += criterion(logits, labels, reduction='sum').item()\n",
    "        correct += torch.sum(pred == labels).item()\n",
    "\n",
    "    val_loss /= len(val_loader.dataset)\n",
    "    val_accur = correct / len(val_loader.dataset)\n",
    "\n",
    "    print(\n",
    "        f'Test set: Avg. loss: {val_loss:.4f}',\n",
    "        f'Accuracy: {correct}/{len(val_loader.dataset)}',\n",
    "        f'({100. * val_accur:.0f}%)',\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_val(\n",
    "    network,\n",
    "    n_epochs,\n",
    "    criterion,\n",
    "    optimizer,\n",
    "    train_loader,\n",
    "    val_loader,\n",
    "):\n",
    "    \"\"\"full cycle of neural network training\"\"\"\n",
    "\n",
    "    val_epoch(network, val_loader, criterion)\n",
    "    for epoch in range(1, n_epochs + 1):\n",
    "        train_epoch(network, train_loader, criterion, optimizer)\n",
    "        val_epoch(network, val_loader, criterion)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Для проверки, что все работает, обучите бейзлайн -- однослойную [полносвязную](https://pytorch.org/docs/stable/generated/torch.nn.Linear.html) сеть на [кросс-энтропийный лосс](https://pytorch.org/docs/stable/generated/torch.nn.functional.cross_entropy.html). \n",
    "\n",
    "\n",
    "*Заметьте*, что на вход этому лоссу нужно подавать сырой выход нейросети, а не результат применения софтмакса.\n",
    "\n",
    "*Подсказка:* Не забудьте вытянуть картинку в вектор с помощью `torch.nn.Flatten()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "class LogReg(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "\n",
    "        # YOUR CODE HERE\n",
    "\n",
    "    def forward(self, x):\n",
    "        \n",
    "        # YOUR CODE HERE\n",
    "\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "net = LogReg()\n",
    "check_numel(net, 7850)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set: Avg. loss: 2.4114 Accuracy: 843/10000 (8%)\n",
      "Test set: Avg. loss: 0.2837 Accuracy: 9179/10000 (92%)\n",
      "Test set: Avg. loss: 0.2796 Accuracy: 9201/10000 (92%)\n",
      "Test set: Avg. loss: 0.2812 Accuracy: 9199/10000 (92%)\n"
     ]
    }
   ],
   "source": [
    "# YOUR CODE HERE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `Задание 3. Логирование (3 балл)`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Скопируйте и перепишите функции `train_epoch`, `val_epoch`, `train_val` таким образом, чтобы\n",
    "- во время обучения раз в `logging_interval` шагов оптимизации выводились лосс и accuracy на одном батче (пример ниже, но не обязательно делать идентично)\n",
    "- происходило сохранение весов лучшей (по accuracy на валидации) модели и состояния оптимизатора ([в помощь](https://pytorch.org/tutorials/beginner/saving_loading_models.html#save-load-state-dict-recommended))\n",
    "- в конце обучения выводился и сохранялся график-саммари (пример ниже, но не обязательно делать идентично)\n",
    "\n",
    "**Ответьте на вопрос:** Что хранится в состоянии оптимизатора? Зачем нужно его сохранять? Приведите хотя бы один пример оптимизатора, для которого есть смысл сохранять состояние."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOUR TRAINING FUNCTIONS HERE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Validation set: Avg. loss: 2.3681, Accuracy: 1300/10000 (13%)\n",
      "\n",
      "Train Epoch: 1 [0/60000 (0%)]\tLoss: 2.399217\n",
      "Train Epoch: 1 [19200/60000 (32%)]\tLoss: 0.316934\n",
      "Train Epoch: 1 [38400/60000 (64%)]\tLoss: 0.522537\n",
      "Train Epoch: 1 [57600/60000 (96%)]\tLoss: 0.435730\n",
      "\n",
      "Validation set: Avg. loss: 0.2926, Accuracy: 9185/10000 (92%)\n",
      "\n",
      "Train Epoch: 2 [0/60000 (0%)]\tLoss: 0.236281\n",
      "Train Epoch: 2 [19200/60000 (32%)]\tLoss: 0.288449\n",
      "Train Epoch: 2 [38400/60000 (64%)]\tLoss: 0.384067\n",
      "Train Epoch: 2 [57600/60000 (96%)]\tLoss: 0.204619\n",
      "\n",
      "Validation set: Avg. loss: 0.2779, Accuracy: 9231/10000 (92%)\n",
      "\n",
      "Train Epoch: 3 [0/60000 (0%)]\tLoss: 0.410723\n",
      "Train Epoch: 3 [19200/60000 (32%)]\tLoss: 0.264276\n",
      "Train Epoch: 3 [38400/60000 (64%)]\tLoss: 0.317687\n",
      "Train Epoch: 3 [57600/60000 (96%)]\tLoss: 0.088601\n",
      "\n",
      "Validation set: Avg. loss: 0.2833, Accuracy: 9208/10000 (92%)\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# YOUR CODE HERE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Загрузите лучшие веса модели и подсчитайте accuracy на валидации, чтобы подтвердить что веса сохранились корректно."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set: Avg. loss: 0.2779 Accuracy: 9231/10000 (92%)\n"
     ]
    }
   ],
   "source": [
    "# YOUR CODE HERE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Загрузите сохранённое состояние оптимизатора и сделайте две эпохи обучения, чтобы подтвердить, что состояния оптимизатора были сохранены корректно."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Validation set: Avg. loss: 0.2768, Accuracy: 9211/10000 (92%)\n",
      "\n",
      "Train Epoch: 1 [0/60000 (0%)]\tLoss: 0.252891\n",
      "Train Epoch: 1 [19200/60000 (32%)]\tLoss: 0.084492\n",
      "Train Epoch: 1 [38400/60000 (64%)]\tLoss: 0.138018\n",
      "Train Epoch: 1 [57600/60000 (96%)]\tLoss: 0.271968\n",
      "\n",
      "Validation set: Avg. loss: 0.2685, Accuracy: 9260/10000 (93%)\n",
      "\n",
      "Train Epoch: 2 [0/60000 (0%)]\tLoss: 0.269862\n",
      "Train Epoch: 2 [19200/60000 (32%)]\tLoss: 0.228278\n",
      "Train Epoch: 2 [38400/60000 (64%)]\tLoss: 0.261185\n",
      "Train Epoch: 2 [57600/60000 (96%)]\tLoss: 0.052912\n",
      "\n",
      "Validation set: Avg. loss: 0.2675, Accuracy: 9261/10000 (93%)\n",
      "\n",
      "Train Epoch: 3 [0/60000 (0%)]\tLoss: 0.330365\n",
      "Train Epoch: 3 [19200/60000 (32%)]\tLoss: 0.552966\n",
      "Train Epoch: 3 [38400/60000 (64%)]\tLoss: 0.363951\n",
      "Train Epoch: 3 [57600/60000 (96%)]\tLoss: 0.197360\n",
      "\n",
      "Validation set: Avg. loss: 0.2670, Accuracy: 9265/10000 (93%)\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# YOUR CODE HERE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `Задание 4. Логирование с помощью tensorboard (1 балл)`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "На самом деле это дурной тон логировать обучение нейросети так, как это сделали вы выше. Дело в том, что весь функционал уже реализован в [`tensorboard`](https://pytorch.org/docs/stable/tensorboard.html), а вы написали велосипед."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ! pip3 install tensorboard\n",
    "from torch.utils.tensorboard import SummaryWriter"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Если у вас есть доступ к терминалу и браузеру, то удобнее всего ввести команду\n",
    "```bash\n",
    "tensorboard --logdir .\n",
    "```\n",
    "и открыть UI в выделенном локалхосте. \n",
    "\n",
    "Если вы работаете в гугл колабе, то можете запустить UI как виджет в Jupyter:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext tensorboard\n",
    "%tensorboard --logdir ."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Как только вы начнете логировать в объект типа `SummaryWriter`, в UI начнут строиться кривые обучения."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Скопируйте и перепишите функции обучения таким образом, чтобы логирование происходило в `tensorboard`.\n",
    "\n",
    "*Замечание.* Вам пригодится метод `add_scalar`, чтобы сохранять лосс и accuracy, и метод `add_hparams`, чтобы сохранить важные гиперпараметры (например, кодовое имя архитектуры сети) и финальное достигнутое качество."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOUR TRAINING FUNCTIONS HERE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `Задание 5. Побейте бейзлайн (2+2 балла)`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "На датасете [`CIFAR-10`](https://pytorch.org/vision/0.8/datasets.html#torchvision.datasets.CIFAR10) обучите модель, которая выдает accuracy `>=0.7` (2 балла) и accuracy `>=0.8` (ещё 2 балла).\n",
    "- Можете использовать любые модули `pytorch`, любые оптимизаторы и шедулеры, можете использовать аугментации ([например](https://pytorch.org/vision/stable/transforms.html))\n",
    "- Для отслеживания экспериментов используйте логирование с `tensorboard`, не забывайте давать осмысленные имена эспериментам и логировать нужные гиперпараметры, сохранять веса сети и состояние оптимизатора."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOUR CODE HERE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `Бонус. Побейте бейзлайн (3 балла)`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "На датасете `CIFAR-10` обучите модель, которая выдает accuracy `>=0.9`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# YOUR CODE HERE"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}