Part2_Optimization.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$\n",
    "\\newcommand{\\mat}[1]{\\boldsymbol {#1}}\n",
    "\\newcommand{\\mattr}[1]{\\boldsymbol {#1}^\\top}\n",
    "\\newcommand{\\matinv}[1]{\\boldsymbol {#1}^{-1}}\n",
    "\\newcommand{\\vec}[1]{\\boldsymbol {#1}}\n",
    "\\newcommand{\\vectr}[1]{\\boldsymbol {#1}^\\top}\n",
    "\\newcommand{\\rvar}[1]{\\mathrm {#1}}\n",
    "\\newcommand{\\rvec}[1]{\\boldsymbol{\\mathrm{#1}}}\n",
    "\\newcommand{\\diag}{\\mathop{\\mathrm {diag}}}\n",
    "\\newcommand{\\set}[1]{\\mathbb {#1}}\n",
    "\\newcommand{\\norm}[1]{\\left\\lVert#1\\right\\rVert}\n",
    "\\newcommand{\\pderiv}[2]{\\frac{\\partial #1}{\\partial #2}}\n",
    "\\newcommand{\\bb}[1]{\\boldsymbol{#1}}\n",
    "$$\n",
    "# Part 2: Optimization and Training\n",
    "<a id=part2></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this part we will learn how to implement optimization algorithms for deep networks. Additionally, we'll learn how to write training loops and implement a modular model trainer.\n",
    "We'll use our optimizers and training code to test a few configurations for classifying images with an MLP model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import unittest\n",
    "import torch\n",
    "import torchvision\n",
    "import torchvision.transforms as tvtf\n",
    "\n",
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "seed = 42\n",
    "\n",
    "plt.rcParams.update({'font.size': 12})\n",
    "test = unittest.TestCase()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Implementing Optimization Algorithms\n",
    "<a id=part2_1></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the context of deep learning, an optimization algorithm is some method of iteratively updating model parameters so that the loss converges toward some local minimum (which we hope will be good enough).\n",
    "\n",
    "Gradient descent-based methods are by far the most popular algorithms for optimization of neural network parameters.\n",
    "However the high-dimensional loss-surfaces we encounter in deep learning applications are highly non-convex.\n",
    "They may be riddled with local minima, saddle points, large plateaus and a host of very challenging \"terrain\" for gradient-based optimization.\n",
    "This gave rise to many different methods of performing the parameter updates based on the loss gradients,\n",
    "aiming to tackle these optimization challenges.\n",
    "\n",
    "The most basic gradient-based update rule can be written as,\n",
    "\n",
    "$$\n",
    "\\vec{\\theta} \\leftarrow \\vec{\\theta} - \\eta \\nabla_{\\vec{\\theta}} L(\\vec{\\theta}; \\mathcal{D})\n",
    "$$\n",
    "\n",
    "where $\\mathcal{D} = \\left\\{ (\\vec{x}^i, \\vec{y}^i) \\right\\}_{i=1}^{M}$ is our training dataset or part of it. Specifically, if we have in total $N$ training samples, then\n",
    "- If $M=N$ this is known as regular gradient descent. If the dataset does not fit in memory it becomes infeasible to compute.\n",
    "- If $M=1$, the loss is computed w.r.t. a single different sample each time. This is known as stochastic gradient descent.\n",
    "- If $1<M<N$ this is known as stochastic mini-batch gradient descent. This is the most commonly-used option.\n",
    "\n",
    "The intuition behind gradient descent is simple: since the gradient of a multivariate function points to the direction of steepest ascent (\"uphill\"), we move in the opposite direction. A small step size $\\eta$ known as the **learning rate** is required since, as with any derivative, the gradient's value is only accurate close to the point it was evaluated at (the current value of $\\vec{\\theta}$).\n",
    "\n",
    "<img src=\"imgs/sgd2d.png\" width=\"600\" />\n",
    "\n",
    "The idea behind the stochastic versions is that by constantly changing the samples we compute the loss with,\n",
    "we get a dynamic error surface, i.e. it's different for each set of training samples.\n",
    "This is thought to generally improve the optimization since it may help the optimizer get out of flat regions or sharp local minima since these features may disappear in the loss surface of subsequent batches. The image below illustrates this. The different lines are different 1-dimensional losses for different training set-samples.\n",
    "\n",
    "<img src=\"imgs/sgd1d.png\" width=\"600\" />\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Deep learning frameworks generally provide implementations of various gradient-based optimization algorithms.\n",
    "Here we'll implement our own optimization module from scratch, this time keeping a similar API to the PyTorch `optim` package.\n",
    "\n",
    "We define a base `Optimizer` class. An optimizer holds a set of parameter tensors (these are the trainable parameters of some model) and maintains internal state. It may be used as follows:\n",
    "- After the forward pass has been performed the optimizer's `zero_grad()` function is invoked to clear the parameter gradients computed by previous iterations.\n",
    "- After the backward pass has been performed, and gradients have been calculated for these parameters, the optimizer's `step()` function is invoked in order to update the value of each parameter based on it's gradient.\n",
    "\n",
    "The exact method of update is implementation-specific for each optimizer and may depend on it's internal state. In addition, adding the regularization penalty to the gradient is handled by the optimizer since it only depends on the parameter values (and not the data)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here's the API of our `Optimizer`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class Optimizer in module hw2.optimizers:\n",
      "\n",
      "class Optimizer(abc.ABC)\n",
      " |  Optimizer(params)\n",
      " |  \n",
      " |  Base class for optimizers.\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      Optimizer\n",
      " |      abc.ABC\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __init__(self, params)\n",
      " |      :param params: A sequence of model parameters to optimize. Can be a\n",
      " |      list of (param,grad) tuples as returned by the Blocks, or a list of\n",
      " |      pytorch tensors in which case the grad will be taken from them.\n",
      " |  \n",
      " |  step(self)\n",
      " |      Updates all the registered parameter values based on their gradients.\n",
      " |  \n",
      " |  zero_grad(self)\n",
      " |      Sets the gradient of the optimized parameters to zero (in place).\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors defined here:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      " |  \n",
      " |  params\n",
      " |      :return: A sequence of parameter tuples, each tuple containing\n",
      " |      (param_data, param_grad). The data should be updated in-place\n",
      " |      according to the grad.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data and other attributes defined here:\n",
      " |  \n",
      " |  __abstractmethods__ = frozenset({'step'})\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import hw2.optimizers as optimizers\n",
    "help(optimizers.Optimizer)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Vanilla SGD with Regularization\n",
    "<a id=part2_2></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start by implementing the simplest gradient based optimizer. The update rule will be exacly as stated above, but we'll also add a L2-regularization term to the gradient. Remember that in the **loss function**, the L2 regularization term is expressed by\n",
    "\n",
    "$$R(\\vec{\\theta}) = \\frac{1}{2}\\lambda||\\vec{\\theta}||^2_2.$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**: Complete the implementation of the `VanillaSGD` class in the `hw2/optimizers.py` module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "diff=0.0\n"
     ]
    }
   ],
   "source": [
    "# Test VanillaSGD\n",
    "torch.manual_seed(42)\n",
    "p = torch.randn(500, 10)\n",
    "dp = torch.randn(*p.shape)*2\n",
    "params = [(p, dp)]\n",
    "\n",
    "vsgd = optimizers.VanillaSGD(params, learn_rate=0.5, reg=0.1)\n",
    "vsgd.step()\n",
    "\n",
    "expected_p = torch.load('tests/assets/expected_vsgd.pt')\n",
    "diff = torch.norm(p-expected_p).item()\n",
    "print(f'diff={diff}')\n",
    "test.assertLess(diff, 1e-3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training\n",
    "<a id=part2_3></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we can build a model and loss function, compute their gradients and we have an optimizer, we can finally do some training!\n",
    "\n",
    "In the spirit of more modular software design, we'll implement a class that will aid us in automating the repetitive training loop code that we usually write over and over again. This will be useful for both training our `Block`-based models and also later for training PyTorch `nn.Module`s.\n",
    "\n",
    "Here's our `Trainer` API:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class Trainer in module hw2.training:\n",
      "\n",
      "class Trainer(abc.ABC)\n",
      " |  Trainer(model, loss_fn, optimizer, device=None)\n",
      " |  \n",
      " |  A class abstracting the various tasks of training models.\n",
      " |  \n",
      " |  Provides methods at multiple levels of granularity:\n",
      " |  - Multiple epochs (fit)\n",
      " |  - Single epoch (train_epoch/test_epoch)\n",
      " |  - Single batch (train_batch/test_batch)\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      Trainer\n",
      " |      abc.ABC\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __init__(self, model, loss_fn, optimizer, device=None)\n",
      " |      Initialize the trainer.\n",
      " |      :param model: Instance of the model to train.\n",
      " |      :param loss_fn: The loss function to evaluate with.\n",
      " |      :param optimizer: The optimizer to train with.\n",
      " |      :param device: torch.device to run training on (CPU or GPU).\n",
      " |  \n",
      " |  fit(self, dl_train: torch.utils.data.dataloader.DataLoader, dl_test: torch.utils.data.dataloader.DataLoader, num_epochs, checkpoints: str = None, early_stopping: int = None, print_every=1, **kw) -> cs236605.train_results.FitResult\n",
      " |      Trains the model for multiple epochs with a given training set,\n",
      " |      and calculates validation loss over a given validation set.\n",
      " |      :param dl_train: Dataloader for the training set.\n",
      " |      :param dl_test: Dataloader for the test set.\n",
      " |      :param num_epochs: Number of epochs to train for.\n",
      " |      :param checkpoints: Whether to save model to file every time the\n",
      " |          test set accuracy improves. Should be a string containing a\n",
      " |          filename without extension.\n",
      " |      :param early_stopping: Whether to stop training early if there is no\n",
      " |          test loss improvement for this number of epochs.\n",
      " |      :param print_every: Print progress every this number of epochs.\n",
      " |      :return: A FitResult object containing train and test losses per epoch.\n",
      " |  \n",
      " |  test_batch(self, batch) -> cs236605.train_results.BatchResult\n",
      " |      Runs a single batch forward through the model and calculates loss.\n",
      " |      :param batch: A single batch of data  from a data loader (might\n",
      " |          be a tuple of data and labels or anything else depending on\n",
      " |          the underlying dataset.\n",
      " |      :return: A BatchResult containing the value of the loss function and\n",
      " |          the number of correctly classified samples in the batch.\n",
      " |  \n",
      " |  test_epoch(self, dl_test: torch.utils.data.dataloader.DataLoader, **kw) -> cs236605.train_results.EpochResult\n",
      " |      Evaluate model once over a test set (single epoch).\n",
      " |      :param dl_test: DataLoader for the test set.\n",
      " |      :param kw: Keyword args supported by _foreach_batch.\n",
      " |      :return: An EpochResult for the epoch.\n",
      " |  \n",
      " |  train_batch(self, batch) -> cs236605.train_results.BatchResult\n",
      " |      Runs a single batch forward through the model, calculates loss,\n",
      " |      preforms back-propagation and uses the optimizer to update weights.\n",
      " |      :param batch: A single batch of data  from a data loader (might\n",
      " |          be a tuple of data and labels or anything else depending on\n",
      " |          the underlying dataset.\n",
      " |      :return: A BatchResult containing the value of the loss function and\n",
      " |          the number of correctly classified samples in the batch.\n",
      " |  \n",
      " |  train_epoch(self, dl_train: torch.utils.data.dataloader.DataLoader, **kw) -> cs236605.train_results.EpochResult\n",
      " |      Train once over a training set (single epoch).\n",
      " |      :param dl_train: DataLoader for the training set.\n",
      " |      :param kw: Keyword args supported by _foreach_batch.\n",
      " |      :return: An EpochResult for the epoch.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors defined here:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data and other attributes defined here:\n",
      " |  \n",
      " |  __abstractmethods__ = frozenset({'test_batch', 'train_batch'})\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import hw2.training as training\n",
    "help(training.Trainer)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `Trainer` class splits the task of training (and evaluating) models into three conceptual levels,\n",
    "- Multiple epochs - the `fit` method, which returns a `FitResult` containing losses and accuracies for all epochs.\n",
    "- Single epoch - the `train_epoch` and `test_epoch` methods, which return an `EpochResult` containing losses per batch and the single accuracy result of the epoch.\n",
    "- Single batch - the `train_batch` and `test_batch` methods, which return a `BatchResult` containing a single loss and the number of correctly classified samples in the batch.\n",
    "\n",
    "It implements the first two levels. Inheriting classes are expected to implement the single-batch level methods since these are model and/or task specific."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first thing we should do in order to verify our model, gradient calculations and optimizer implementation is to try to overfit a large model (many parameters) to a small dataset (few images). This will show us that things are working properly.\n",
    "\n",
    "Let's begin by loading the CIFAR-10 dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n",
      "Train: 50000 samples\n",
      "Test: 10000 samples\n"
     ]
    }
   ],
   "source": [
    "data_dir = os.path.expanduser('~/.pytorch-datasets')\n",
    "ds_train = torchvision.datasets.CIFAR10(root=data_dir, download=True, train=True, transform=tvtf.ToTensor())\n",
    "ds_test = torchvision.datasets.CIFAR10(root=data_dir, download=True, train=False, transform=tvtf.ToTensor())\n",
    "\n",
    "print(f'Train: {len(ds_train)} samples')\n",
    "print(f'Test: {len(ds_test)} samples')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's implement just a small part of our training logic since that's what we need right now."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**:\n",
    "1. Complete the implementation of the `train_batch()` method in the `BlocksTrainer` class within the `hw2/training.py` module.\n",
    "1. Update the hyperparameter values in the `part2_overfit_hp()` function in the `hw2/answers.py` module. Tweak the hyperparameter values until your model overfits a small number of samples in the block below. You should get 100% accuracy within a few epochs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train_batch (Avg. Loss 4.565, Accuracy 5.0): 100%|██████████| 2/2 [00:00<00:00, 142.36it/s]\n",
      "train_batch (Avg. Loss 3.131, Accuracy 20.0): 100%|██████████| 2/2 [00:00<00:00, 119.58it/s]\n",
      "train_batch (Avg. Loss 2.730, Accuracy 30.0): 100%|██████████| 2/2 [00:00<00:00, 111.11it/s]\n",
      "train_batch (Avg. Loss 2.644, Accuracy 20.0): 100%|██████████| 2/2 [00:00<00:00, 137.55it/s]\n",
      "train_batch (Avg. Loss 1.718, Accuracy 50.0): 100%|██████████| 2/2 [00:00<00:00, 98.03it/s] \n",
      "train_batch (Avg. Loss 1.631, Accuracy 40.0): 100%|██████████| 2/2 [00:00<00:00, 114.25it/s]\n",
      "train_batch (Avg. Loss 1.473, Accuracy 45.0): 100%|██████████| 2/2 [00:00<00:00, 125.75it/s]\n",
      "train_batch (Avg. Loss 1.179, Accuracy 60.0): 100%|██████████| 2/2 [00:00<00:00, 127.76it/s]\n",
      "train_batch (Avg. Loss 0.987, Accuracy 75.0): 100%|██████████| 2/2 [00:00<00:00, 133.75it/s]\n",
      "train_batch (Avg. Loss 0.881, Accuracy 95.0): 100%|██████████| 2/2 [00:00<00:00, 109.56it/s]\n",
      "train_batch (Avg. Loss 0.860, Accuracy 85.0): 100%|██████████| 2/2 [00:00<00:00, 109.86it/s]\n",
      "train_batch (Avg. Loss 0.804, Accuracy 85.0): 100%|██████████| 2/2 [00:00<00:00, 131.20it/s]\n",
      "train_batch (Avg. Loss 0.683, Accuracy 95.0): 100%|██████████| 2/2 [00:00<00:00, 113.92it/s]\n",
      "train_batch (Avg. Loss 0.575, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 110.23it/s]\n",
      "train_batch (Avg. Loss 0.522, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 113.47it/s]\n",
      "train_batch (Avg. Loss 0.485, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 115.62it/s]\n",
      "train_batch (Avg. Loss 0.442, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 142.47it/s]\n",
      "train_batch (Avg. Loss 0.407, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 118.83it/s]\n",
      "train_batch (Avg. Loss 0.379, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 136.89it/s]\n",
      "train_batch (Avg. Loss 0.352, Accuracy 100.0): 100%|██████████| 2/2 [00:00<00:00, 153.59it/s]\n"
     ]
    }
   ],
   "source": [
    "import hw2.blocks as blocks\n",
    "import hw2.models as models\n",
    "import hw2.answers as answers\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "# Overfit to a very small dataset of 20 samples\n",
    "batch_size = 10\n",
    "max_batches = 2\n",
    "dl_train = torch.utils.data.DataLoader(ds_train, batch_size, shuffle=False)\n",
    "\n",
    "# Get hyperparameters\n",
    "hp = answers.part2_overfit_hp()\n",
    "\n",
    "torch.manual_seed(seed)\n",
    "\n",
    "# Build a model and loss using our custom MLP and CE implementations\n",
    "model = models.MLP(3*32*32, num_classes=10, hidden_features=[128]*3, wstd=hp['wstd'])\n",
    "loss_fn = blocks.CrossEntropyLoss()\n",
    "\n",
    "# Use our custom optimizer\n",
    "optimizer = optimizers.VanillaSGD(model.params(), learn_rate=hp['lr'], reg=hp['reg'])\n",
    "\n",
    "# Run training over small dataset multiple times\n",
    "trainer = training.BlocksTrainer(model, loss_fn, optimizer)\n",
    "best_acc = 0\n",
    "for i in range(20):\n",
    "    res = trainer.train_epoch(dl_train, max_batches=max_batches)\n",
    "    best_acc = res.accuracy if res.accuracy > best_acc else best_acc\n",
    "    \n",
    "test.assertGreaterEqual(best_acc, 98)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we know training works, let's try to fit a model to a bit more data for a few epochs, to see how well we're doing. First, we need a function to plot the FitResults object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m\n",
       "\u001b[0mplot_fit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mfit_res\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mcs236605\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_results\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mFitResult\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mfig\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mlog_loss\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mlegend\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m\n",
       "Plots a FitResult object.\n",
       "Creates four plots: train loss, test loss, train acc, test acc.\n",
       ":param fit_res: The fit result to plot.\n",
       ":param fig: A figure previously returned from this function. If not None,\n",
       "    plots will the added to this figure.\n",
       ":param log_loss: Whether to plot the losses in log scale.\n",
       ":param legend: What to call this FitResult in the legend.\n",
       ":return: The figure.\n",
       "\u001b[0;31mFile:\u001b[0m      ~/236605/cs236605-hw2/cs236605/plot.py\n",
       "\u001b[0;31mType:\u001b[0m      function\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from cs236605.plot import plot_fit\n",
    "plot_fit?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**:\n",
    "1. Complete the implementation of the `test_batch()` method in the `BlocksTrainer` class within the `hw2/training.py` module.\n",
    "1. Implement the `fit()` method of the `Trainer` class within the `hw2/training.py` module.\n",
    "1. Tweak the hyperparameters for this section in the `part2_optim_hp()` function in the `hw2/answers.py` module.\n",
    "1. Run the following blocks to train. Try to get above 35-40% test-set accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a larger part of the CIFAR-10 dataset (still not the whole thing)\n",
    "batch_size = 50\n",
    "max_batches = 100\n",
    "in_features = 3*32*32\n",
    "num_classes = 10\n",
    "dl_train = torch.utils.data.DataLoader(ds_train, batch_size, shuffle=False)\n",
    "dl_test = torch.utils.data.DataLoader(ds_test, batch_size//2, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a function to train a model with our Trainer and various optimizers\n",
    "def train_with_optimizer(opt_name, opt_class, fig):\n",
    "    torch.manual_seed(seed)\n",
    "    \n",
    "    # Get hyperparameters\n",
    "    hp = answers.part2_optim_hp()\n",
    "    hidden_features = [128] * 5\n",
    "    num_epochs = 10\n",
    "    \n",
    "    # Create model, loss and optimizer instances\n",
    "    model = models.MLP(in_features, num_classes, hidden_features, wstd=hp['wstd'])\n",
    "    loss_fn = blocks.CrossEntropyLoss()\n",
    "    optimizer = opt_class(model.params(), learn_rate=hp[f'lr_{opt_name}'], reg=hp['reg'])\n",
    "\n",
    "    # Train with the Trainer\n",
    "    trainer = training.BlocksTrainer(model, loss_fn, optimizer)\n",
    "    fit_res = trainer.fit(dl_train, dl_test, num_epochs, max_batches=max_batches)\n",
    "    \n",
    "    fig, axes = plot_fit(fit_res, fig=fig, legend=opt_name)\n",
    "    return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- EPOCH 1/10 ---\n",
      "train_batch (Avg. Loss 2.458, Accuracy 18.0): 100%|██████████| 100/100 [00:01<00:00, 65.17it/s]\n",
      "test_batch (Avg. Loss 2.035, Accuracy 25.0): 100%|██████████| 100/100 [00:00<00:00, 155.32it/s]\n",
      "--- EPOCH 2/10 ---\n",
      "train_batch (Avg. Loss 2.004, Accuracy 27.0): 100%|██████████| 100/100 [00:01<00:00, 67.48it/s]\n",
      "test_batch (Avg. Loss 1.953, Accuracy 29.0): 100%|██████████| 100/100 [00:00<00:00, 161.59it/s]\n",
      "--- EPOCH 3/10 ---\n",
      "train_batch (Avg. Loss 1.923, Accuracy 30.0): 100%|██████████| 100/100 [00:01<00:00, 69.22it/s]\n",
      "test_batch (Avg. Loss 1.910, Accuracy 31.0): 100%|██████████| 100/100 [00:00<00:00, 159.71it/s]\n",
      "--- EPOCH 4/10 ---\n",
      "train_batch (Avg. Loss 1.864, Accuracy 32.0): 100%|██████████| 100/100 [00:01<00:00, 68.58it/s]\n",
      "test_batch (Avg. Loss 1.872, Accuracy 33.0): 100%|██████████| 100/100 [00:00<00:00, 155.45it/s]\n",
      "--- EPOCH 5/10 ---\n",
      "train_batch (Avg. Loss 1.819, Accuracy 34.0): 100%|██████████| 100/100 [00:01<00:00, 67.24it/s]\n",
      "test_batch (Avg. Loss 1.843, Accuracy 34.0): 100%|██████████| 100/100 [00:00<00:00, 154.79it/s]\n",
      "--- EPOCH 6/10 ---\n",
      "train_batch (Avg. Loss 1.778, Accuracy 35.0): 100%|██████████| 100/100 [00:01<00:00, 65.09it/s]\n",
      "test_batch (Avg. Loss 1.819, Accuracy 36.0): 100%|██████████| 100/100 [00:00<00:00, 142.06it/s]\n",
      "--- EPOCH 7/10 ---\n",
      "train_batch (Avg. Loss 1.743, Accuracy 37.0): 100%|██████████| 100/100 [00:01<00:00, 66.09it/s]\n",
      "test_batch (Avg. Loss 1.803, Accuracy 36.0): 100%|██████████| 100/100 [00:00<00:00, 162.14it/s]\n",
      "--- EPOCH 8/10 ---\n",
      "train_batch (Avg. Loss 1.710, Accuracy 38.0): 100%|██████████| 100/100 [00:01<00:00, 63.75it/s]\n",
      "test_batch (Avg. Loss 1.780, Accuracy 36.0): 100%|██████████| 100/100 [00:00<00:00, 153.67it/s]\n",
      "--- EPOCH 9/10 ---\n",
      "train_batch (Avg. Loss 1.679, Accuracy 39.0): 100%|██████████| 100/100 [00:01<00:00, 66.19it/s]\n",
      "test_batch (Avg. Loss 1.773, Accuracy 37.0): 100%|██████████| 100/100 [00:00<00:00, 134.70it/s]\n",
      "--- EPOCH 10/10 ---\n",
      "train_batch (Avg. Loss 1.648, Accuracy 40.0): 100%|██████████| 100/100 [00:01<00:00, 63.99it/s]\n",
      "test_batch (Avg. Loss 1.760, Accuracy 37.0): 100%|██████████| 100/100 [00:00<00:00, 140.03it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_optim = None\n",
    "fig_optim = train_with_optimizer('vanilla', optimizers.VanillaSGD, fig_optim)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Momentum\n",
    "<a id=part2_4></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The simple vanilla SGD update is rarely used in practice since it's very slow to converge relative to other optimization algorithms.\n",
    "\n",
    "One reason is that naïvely updating in the direction of the current gradient causes it to fluctuate wildly in areas of the loss surface where some dimensions are much steeper than others.\n",
    "Another reason is that using the same learning rate for all parameters is not a great idea since not all parameters are created equal. For example, parameters associated with rare features should be updated with a larger step than ones associated with commonly-occurring features because they'll get less updates through the gradients.\n",
    "\n",
    "Therefore more advanced optimizers take into account the previous gradients of a parameter and/or try to use a per-parameter specific learning rate instead of a common one."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now implement a simple and common optimizer: SGD with Momentum. This optimizer takes **previous gradients** of a parameter into account when updating it's value instead of just the current one. In practice it usually provides faster convergence than the vanilla SGD.\n",
    "\n",
    "The SGD with Momentum update rule can be stated as follows:\n",
    "$$\\begin{align}\n",
    "\\vec{v}_{t+1} &= \\mu \\vec{v}_t - \\eta \\delta \\vec{\\theta}_t \\\\\n",
    "\\vec{\\theta}_{t+1} &= \\vec{\\theta}_t + \\vec{v}_{t+1}\n",
    "\\end{align}$$\n",
    "\n",
    "Where $\\eta$ is the learning rate,\n",
    "$\\vec{\\theta}$ is a model parameter,\n",
    "$\\delta \\vec{\\theta}_t=\\pderiv{L}{\\vec{\\theta}}(\\vec{\\theta}_t)$ is the gradient of the loss w.r.t. to the parameter and $\\mu$ is a hyperparameter known as momentum. \n",
    "\n",
    "Expanding the update rule recursively shows us now the parameter update infact depends on all previous gradient values for that parameter, where the old gradients are exponentially decayed by a factor of $\\mu$ at each timestep. \n",
    "\n",
    "Since we're incorporating previous gradient (update directions), a noisy value of the current gradient will have less effect so that the general direction of previous updates is maintained somewhat. The following figure illustrates this.\n",
    "\n",
    "<img src=\"imgs/sgd-momentum.png\" width=\"600\" />\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**:\n",
    "1. Complete the implementation of the `MomentumSGD` class in the `hw2/optimizers.py` module.\n",
    "1. Tweak the learning rate for momentum in `part2_optim_hp()` the function in the `hw2/answers.py` module.\n",
    "1. Run the following block to compare to the vanilla SGD."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- EPOCH 1/10 ---\n",
      "train_batch (Avg. Loss 2.332, Accuracy 19.0): 100%|██████████| 100/100 [00:01<00:00, 61.21it/s]\n",
      "test_batch (Avg. Loss 2.066, Accuracy 23.0): 100%|██████████| 100/100 [00:00<00:00, 159.39it/s]\n",
      "--- EPOCH 2/10 ---\n",
      "train_batch (Avg. Loss 1.955, Accuracy 27.0): 100%|██████████| 100/100 [00:01<00:00, 60.14it/s]\n",
      "test_batch (Avg. Loss 2.019, Accuracy 29.0): 100%|██████████| 100/100 [00:00<00:00, 157.06it/s]\n",
      "--- EPOCH 3/10 ---\n",
      "train_batch (Avg. Loss 1.874, Accuracy 31.0): 100%|██████████| 100/100 [00:01<00:00, 60.95it/s]\n",
      "test_batch (Avg. Loss 1.962, Accuracy 29.0): 100%|██████████| 100/100 [00:00<00:00, 158.81it/s]\n",
      "--- EPOCH 4/10 ---\n",
      "train_batch (Avg. Loss 1.807, Accuracy 34.0): 100%|██████████| 100/100 [00:01<00:00, 61.66it/s]\n",
      "test_batch (Avg. Loss 1.875, Accuracy 32.0): 100%|██████████| 100/100 [00:00<00:00, 162.04it/s]\n",
      "--- EPOCH 5/10 ---\n",
      "train_batch (Avg. Loss 1.744, Accuracy 37.0): 100%|██████████| 100/100 [00:01<00:00, 60.07it/s]\n",
      "test_batch (Avg. Loss 1.836, Accuracy 33.0): 100%|██████████| 100/100 [00:00<00:00, 150.71it/s]\n",
      "--- EPOCH 6/10 ---\n",
      "train_batch (Avg. Loss 1.688, Accuracy 40.0): 100%|██████████| 100/100 [00:01<00:00, 58.10it/s]\n",
      "test_batch (Avg. Loss 1.799, Accuracy 34.0): 100%|██████████| 100/100 [00:00<00:00, 158.92it/s]\n",
      "--- EPOCH 7/10 ---\n",
      "train_batch (Avg. Loss 1.644, Accuracy 41.0): 100%|██████████| 100/100 [00:01<00:00, 60.35it/s]\n",
      "test_batch (Avg. Loss 1.783, Accuracy 35.0): 100%|██████████| 100/100 [00:00<00:00, 153.12it/s]\n",
      "--- EPOCH 8/10 ---\n",
      "train_batch (Avg. Loss 1.608, Accuracy 42.0): 100%|██████████| 100/100 [00:01<00:00, 59.75it/s]\n",
      "test_batch (Avg. Loss 1.748, Accuracy 37.0): 100%|██████████| 100/100 [00:00<00:00, 146.29it/s]\n",
      "--- EPOCH 9/10 ---\n",
      "train_batch (Avg. Loss 1.586, Accuracy 43.0): 100%|██████████| 100/100 [00:01<00:00, 59.32it/s]\n",
      "test_batch (Avg. Loss 1.747, Accuracy 37.0): 100%|██████████| 100/100 [00:00<00:00, 157.33it/s]\n",
      "--- EPOCH 10/10 ---\n",
      "train_batch (Avg. Loss 1.547, Accuracy 44.0): 100%|██████████| 100/100 [00:01<00:00, 52.35it/s]\n",
      "test_batch (Avg. Loss 1.734, Accuracy 38.0): 100%|██████████| 100/100 [00:00<00:00, 151.77it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 4 Axes>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig_optim = train_with_optimizer('momentum', optimizers.MomentumSGD, fig_optim)\n",
    "fig_optim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RMSProp\n",
    "<a id=part2_5></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is another optmizer that accounts for previous gradients, but this time it uses them to adapt the learning rate per parameter.\n",
    "\n",
    "RMSProp maintains a decaying moving average of previous squared gradients,\n",
    "$$\n",
    "\\vec{r}_{t+1} = \\gamma\\vec{r}_{t} + (1-\\gamma)\\delta\\vec{\\theta}_t^2\n",
    "$$\n",
    "\n",
    "where $0<\\gamma<1$ is a decay constant usually set close to $1$. The update rule for each parameter is then,\n",
    "$$\n",
    "\\vec{\\theta}_{t+1} = \\vec{\\theta}_t - \\left( \\frac{\\eta}{\\sqrt{r_{t+1}+\\varepsilon}} \\right) \\delta\\vec{\\theta}_t\n",
    "$$\n",
    "\n",
    "where $\\varepsilon$ is a small constant to prevent numerical instability. The idea here is to decrease the learning rate for parameters with high gradient values and vice-versa. The decaying moving average prevents accumulating all the past gradients which would cause the effective learning rate to become zero."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**:\n",
    "1. Complete the implementation of the `RMSProp` class in the `hw2/optimizers.py` module.\n",
    "1. Tweak the learning rate for RMSProp in `part2_optim_hp()` the function in the `hw2/answers.py` module.\n",
    "1. Run the following block to compare to the other optimizers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- EPOCH 1/10 ---\n",
      "train_batch (Avg. Loss 2.268, Accuracy 21.0): 100%|██████████| 100/100 [00:01<00:00, 52.99it/s]\n",
      "test_batch (Avg. Loss 2.055, Accuracy 25.0): 100%|██████████| 100/100 [00:00<00:00, 159.41it/s]\n",
      "--- EPOCH 2/10 ---\n",
      "train_batch (Avg. Loss 1.947, Accuracy 29.0): 100%|██████████| 100/100 [00:01<00:00, 54.00it/s]\n",
      "test_batch (Avg. Loss 1.958, Accuracy 28.0): 100%|██████████| 100/100 [00:00<00:00, 158.41it/s]\n",
      "--- EPOCH 3/10 ---\n",
      "train_batch (Avg. Loss 1.863, Accuracy 32.0): 100%|██████████| 100/100 [00:01<00:00, 51.37it/s]\n",
      "test_batch (Avg. Loss 1.918, Accuracy 30.0): 100%|██████████| 100/100 [00:00<00:00, 157.56it/s]\n",
      "--- EPOCH 4/10 ---\n",
      "train_batch (Avg. Loss 1.805, Accuracy 34.0): 100%|██████████| 100/100 [00:01<00:00, 52.27it/s]\n",
      "test_batch (Avg. Loss 1.878, Accuracy 32.0): 100%|██████████| 100/100 [00:00<00:00, 154.92it/s]\n",
      "--- EPOCH 5/10 ---\n",
      "train_batch (Avg. Loss 1.756, Accuracy 36.0): 100%|██████████| 100/100 [00:02<00:00, 49.36it/s]\n",
      "test_batch (Avg. Loss 1.851, Accuracy 33.0): 100%|██████████| 100/100 [00:00<00:00, 152.09it/s]\n",
      "--- EPOCH 6/10 ---\n",
      "train_batch (Avg. Loss 1.714, Accuracy 38.0): 100%|██████████| 100/100 [00:01<00:00, 51.54it/s]\n",
      "test_batch (Avg. Loss 1.828, Accuracy 34.0): 100%|██████████| 100/100 [00:00<00:00, 149.95it/s]\n",
      "--- EPOCH 7/10 ---\n",
      "train_batch (Avg. Loss 1.675, Accuracy 40.0): 100%|██████████| 100/100 [00:01<00:00, 51.44it/s]\n",
      "test_batch (Avg. Loss 1.810, Accuracy 35.0): 100%|██████████| 100/100 [00:00<00:00, 150.21it/s]\n",
      "--- EPOCH 8/10 ---\n",
      "train_batch (Avg. Loss 1.637, Accuracy 41.0): 100%|██████████| 100/100 [00:01<00:00, 51.41it/s]\n",
      "test_batch (Avg. Loss 1.788, Accuracy 36.0): 100%|██████████| 100/100 [00:00<00:00, 147.66it/s]\n",
      "--- EPOCH 9/10 ---\n",
      "train_batch (Avg. Loss 1.604, Accuracy 43.0): 100%|██████████| 100/100 [00:02<00:00, 43.77it/s]\n",
      "test_batch (Avg. Loss 1.779, Accuracy 37.0): 100%|██████████| 100/100 [00:00<00:00, 118.25it/s]\n",
      "--- EPOCH 10/10 ---\n",
      "train_batch (Avg. Loss 1.571, Accuracy 44.0): 100%|██████████| 100/100 [00:02<00:00, 41.83it/s]\n",
      "test_batch (Avg. Loss 1.757, Accuracy 38.0): 100%|██████████| 100/100 [00:00<00:00, 141.86it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 4 Axes>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig_optim = train_with_optimizer('rmsprop', optimizers.RMSProp, fig_optim)\n",
    "fig_optim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that you should get better train/test accuracy with Momentum and RMSProp than Vanilla."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dropout Regularization\n",
    "<a id=part2_6></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Dropout](http://jmlr.org/papers/volume15/srivastava14a.old/srivastava14a.pdf) is a useful technique to improve generalization of deep models.\n",
    "\n",
    "The idea is simple: during the forward pass drop, i.e. set to to zero, the activation of each neuron, with a probability of $p$. For example, if $p=0.5$ this means we drop half the activations on average.\n",
    "\n",
    "There are a few important things to note about dropout:\n",
    "1. It is only performed during training. When testing our model the dropout layers should be a no-op.\n",
    "1. In the backward pass, gradients are only propagated back into neurons that weren't dropped during the forward pass.\n",
    "1. During testing, the activations must be scaled since the expected value of each neuron during the training phase is now $1-p$ times it's original expectation. Thus, we need to scale the test-time activations by $1-p$ to match. Equivalently, we can scale the train time activations by $1/(1-p)$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**: Complete the implementation of the `Dropout` class in the `hw2/blocks.py` module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MLP, Sequential\n",
      "\t[0] Linear(3072, 50)\n",
      "\t[1] ReLU\n",
      "\t[2] Dropout(p=0.6)\n",
      "\t[3] Linear(50, 50)\n",
      "\t[4] ReLU\n",
      "\t[5] Dropout(p=0.6)\n",
      "\t[6] Linear(50, 50)\n",
      "\t[7] ReLU\n",
      "\t[8] Dropout(p=0.6)\n",
      "\t[9] Linear(50, 10)\n",
      "\n",
      "Comparing gradients... \n",
      "input    diff=0.000\n",
      "param#01 diff=0.000\n",
      "param#02 diff=0.000\n",
      "param#03 diff=0.000\n",
      "param#04 diff=0.000\n",
      "param#05 diff=0.000\n",
      "param#06 diff=0.000\n",
      "param#07 diff=0.000\n",
      "param#08 diff=0.000\n",
      "Comparing gradients... \n",
      "input    diff=0.000\n",
      "param#01 diff=0.000\n",
      "param#02 diff=0.000\n",
      "param#03 diff=0.000\n",
      "param#04 diff=0.000\n",
      "param#05 diff=0.000\n",
      "param#06 diff=0.000\n",
      "param#07 diff=0.000\n",
      "param#08 diff=0.000\n"
     ]
    }
   ],
   "source": [
    "from hw2.grad_compare import compare_block_to_torch\n",
    "\n",
    "# Check architecture of MLP with dropout layers\n",
    "mlp_dropout = models.MLP(in_features, num_classes, [50]*3, dropout=0.6)\n",
    "print(mlp_dropout)\n",
    "test.assertEqual(len(mlp_dropout.sequence), 10)\n",
    "for b1, b2 in zip(mlp_dropout.sequence, mlp_dropout.sequence[1:]):\n",
    "    if str(b1) == 'relu':\n",
    "        test.assertTrue(str(b2).startswith('Dropout'))\n",
    "test.assertTrue(str(mlp_dropout.sequence[-1]).startswith('Linear'))\n",
    "\n",
    "# Test end-to-end gradient in train and test modes.\n",
    "mlp_dropout.train(True)\n",
    "for diff in compare_block_to_torch(mlp_dropout, torch.randn(500, in_features)):\n",
    "    test.assertLess(diff, 1e-2)\n",
    "    \n",
    "mlp_dropout.train(False)\n",
    "for diff in compare_block_to_torch(mlp_dropout, torch.randn(500, in_features)):\n",
    "    test.assertLess(diff, 1e-2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To see whether dropout really improves generalization, let's take a small training set\n",
    "(small enough to overfit) and a large test set and check whether we get less overfitting and\n",
    "perhaps improved test-set accuracy when using dropout."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a small set from CIFAR-10, but take a larger test set since we want to test generalization\n",
    "batch_size = 10\n",
    "max_batches = 40\n",
    "in_features = 3*32*32\n",
    "num_classes = 10\n",
    "dl_train = torch.utils.data.DataLoader(ds_train, batch_size, shuffle=False)\n",
    "dl_test = torch.utils.data.DataLoader(ds_test, batch_size*2, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get hyperparameters\n",
    "hp = answers.part2_dropout_hp()\n",
    "hidden_features = [400] * 1\n",
    "num_epochs = 30"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO**:\n",
    "Tweak the hyperparameters for this section in the `part2_dropout_hp()` function in the `hw2/answers.py` module. Try to set them so that the first model (with `dropout`=0) overfits. You can disable the other dropout options until you tune the hyperparameters. We can then see the effect of dropout for generalization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** Training with dropout= 0\n",
      "--- EPOCH 1/30 ---\n",
      "train_batch (Avg. Loss 4.585, Accuracy 11.0): 100%|██████████| 40/40 [00:00<00:00, 50.75it/s]\n",
      "test_batch (Avg. Loss 3.250, Accuracy 15.0): 100%|██████████| 40/40 [00:00<00:00, 175.14it/s]\n",
      "train_batch (Avg. Loss 2.688, Accuracy 25.0): 100%|██████████| 40/40 [00:00<00:00, 50.19it/s]\n",
      "test_batch (Avg. Loss 2.607, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 149.66it/s]\n",
      "train_batch (Avg. Loss 2.132, Accuracy 33.0): 100%|██████████| 40/40 [00:00<00:00, 47.72it/s]\n",
      "test_batch (Avg. Loss 2.654, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 163.79it/s]\n",
      "train_batch (Avg. Loss 1.820, Accuracy 39.0): 100%|██████████| 40/40 [00:00<00:00, 48.97it/s]\n",
      "test_batch (Avg. Loss 2.562, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 172.11it/s]\n",
      "train_batch (Avg. Loss 1.582, Accuracy 47.0): 100%|██████████| 40/40 [00:00<00:00, 48.25it/s]\n",
      "test_batch (Avg. Loss 2.655, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 176.90it/s]\n",
      "train_batch (Avg. Loss 1.414, Accuracy 52.0): 100%|██████████| 40/40 [00:00<00:00, 50.11it/s]\n",
      "test_batch (Avg. Loss 2.705, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 173.99it/s]\n",
      "--- EPOCH 7/30 ---\n",
      "train_batch (Avg. Loss 1.313, Accuracy 55.0): 100%|██████████| 40/40 [00:00<00:00, 48.50it/s]\n",
      "test_batch (Avg. Loss 2.746, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 164.28it/s]\n",
      "train_batch (Avg. Loss 1.240, Accuracy 56.0): 100%|██████████| 40/40 [00:00<00:00, 45.39it/s]\n",
      "test_batch (Avg. Loss 2.809, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 151.13it/s]\n",
      "train_batch (Avg. Loss 1.183, Accuracy 59.0): 100%|██████████| 40/40 [00:00<00:00, 47.22it/s]\n",
      "test_batch (Avg. Loss 2.895, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 174.56it/s]\n",
      "train_batch (Avg. Loss 1.142, Accuracy 62.0): 100%|██████████| 40/40 [00:00<00:00, 45.96it/s]\n",
      "test_batch (Avg. Loss 2.885, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 172.28it/s]\n",
      "train_batch (Avg. Loss 1.082, Accuracy 65.0): 100%|██████████| 40/40 [00:00<00:00, 46.97it/s]\n",
      "test_batch (Avg. Loss 2.745, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 168.63it/s]\n",
      "train_batch (Avg. Loss 0.975, Accuracy 69.0): 100%|██████████| 40/40 [00:00<00:00, 44.56it/s]\n",
      "test_batch (Avg. Loss 2.660, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 144.15it/s]\n",
      "--- EPOCH 13/30 ---\n",
      "train_batch (Avg. Loss 0.864, Accuracy 72.0): 100%|██████████| 40/40 [00:00<00:00, 47.18it/s]\n",
      "test_batch (Avg. Loss 2.677, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 154.01it/s]\n",
      "train_batch (Avg. Loss 0.784, Accuracy 77.0): 100%|██████████| 40/40 [00:00<00:00, 47.51it/s]\n",
      "test_batch (Avg. Loss 2.726, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 172.22it/s]\n",
      "train_batch (Avg. Loss 0.728, Accuracy 79.0): 100%|██████████| 40/40 [00:00<00:00, 44.74it/s]\n",
      "test_batch (Avg. Loss 2.768, Accuracy 19.0): 100%|██████████| 40/40 [00:00<00:00, 122.28it/s]\n",
      "train_batch (Avg. Loss 0.677, Accuracy 81.0): 100%|██████████| 40/40 [00:00<00:00, 46.01it/s]\n",
      "test_batch (Avg. Loss 2.741, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 129.20it/s]\n",
      "train_batch (Avg. Loss 0.623, Accuracy 82.0): 100%|██████████| 40/40 [00:00<00:00, 45.49it/s]\n",
      "test_batch (Avg. Loss 2.701, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 123.14it/s]\n",
      "train_batch (Avg. Loss 0.566, Accuracy 85.0): 100%|██████████| 40/40 [00:00<00:00, 44.32it/s]\n",
      "test_batch (Avg. Loss 2.682, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 156.18it/s]\n",
      "--- EPOCH 19/30 ---\n",
      "train_batch (Avg. Loss 0.518, Accuracy 86.0): 100%|██████████| 40/40 [00:00<00:00, 46.87it/s]\n",
      "test_batch (Avg. Loss 2.699, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 138.76it/s]\n",
      "train_batch (Avg. Loss 0.482, Accuracy 90.0): 100%|██████████| 40/40 [00:00<00:00, 47.49it/s]\n",
      "test_batch (Avg. Loss 2.724, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 155.54it/s]\n",
      "train_batch (Avg. Loss 0.449, Accuracy 90.0): 100%|██████████| 40/40 [00:00<00:00, 48.45it/s]\n",
      "test_batch (Avg. Loss 2.728, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 162.03it/s]\n",
      "train_batch (Avg. Loss 0.422, Accuracy 92.0): 100%|██████████| 40/40 [00:00<00:00, 47.02it/s]\n",
      "test_batch (Avg. Loss 2.744, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 141.46it/s]\n",
      "train_batch (Avg. Loss 0.396, Accuracy 94.0): 100%|██████████| 40/40 [00:00<00:00, 46.02it/s]\n",
      "test_batch (Avg. Loss 2.762, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 129.97it/s]\n",
      "train_batch (Avg. Loss 0.377, Accuracy 95.0): 100%|██████████| 40/40 [00:00<00:00, 46.79it/s]\n",
      "test_batch (Avg. Loss 2.788, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 150.16it/s]\n",
      "--- EPOCH 25/30 ---\n",
      "train_batch (Avg. Loss 0.358, Accuracy 95.0): 100%|██████████| 40/40 [00:00<00:00, 43.76it/s]\n",
      "test_batch (Avg. Loss 2.816, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 110.31it/s]\n",
      "train_batch (Avg. Loss 0.343, Accuracy 96.0): 100%|██████████| 40/40 [00:00<00:00, 43.66it/s]\n",
      "test_batch (Avg. Loss 2.854, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 110.71it/s]\n",
      "train_batch (Avg. Loss 0.327, Accuracy 96.0): 100%|██████████| 40/40 [00:00<00:00, 45.56it/s]\n",
      "test_batch (Avg. Loss 2.890, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 113.09it/s]\n",
      "train_batch (Avg. Loss 0.317, Accuracy 97.0): 100%|██████████| 40/40 [00:00<00:00, 43.47it/s]\n",
      "test_batch (Avg. Loss 2.933, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 120.21it/s]\n",
      "train_batch (Avg. Loss 0.307, Accuracy 96.0): 100%|██████████| 40/40 [00:00<00:00, 46.04it/s]\n",
      "test_batch (Avg. Loss 2.969, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 131.16it/s]\n",
      "--- EPOCH 30/30 ---\n",
      "train_batch (Avg. Loss 0.297, Accuracy 97.0): 100%|██████████| 40/40 [00:00<00:00, 44.26it/s]\n",
      "test_batch (Avg. Loss 3.010, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 140.44it/s]\n",
      "*** Training with dropout= 0.4\n",
      "--- EPOCH 1/30 ---\n",
      "train_batch (Avg. Loss 4.787, Accuracy 13.0): 100%|██████████| 40/40 [00:00<00:00, 45.95it/s]\n",
      "test_batch (Avg. Loss 3.017, Accuracy 17.0): 100%|██████████| 40/40 [00:00<00:00, 152.50it/s]\n",
      "train_batch (Avg. Loss 3.191, Accuracy 16.0): 100%|██████████| 40/40 [00:00<00:00, 42.79it/s]\n",
      "test_batch (Avg. Loss 2.514, Accuracy 19.0): 100%|██████████| 40/40 [00:00<00:00, 123.52it/s]\n",
      "train_batch (Avg. Loss 2.290, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 43.69it/s]\n",
      "test_batch (Avg. Loss 2.286, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 122.39it/s]\n",
      "train_batch (Avg. Loss 2.144, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 43.87it/s]\n",
      "test_batch (Avg. Loss 2.269, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 131.10it/s]\n",
      "train_batch (Avg. Loss 1.977, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 45.59it/s]\n",
      "test_batch (Avg. Loss 2.191, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 136.15it/s]\n",
      "train_batch (Avg. Loss 1.922, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 44.68it/s]\n",
      "test_batch (Avg. Loss 2.160, Accuracy 25.0): 100%|██████████| 40/40 [00:00<00:00, 138.10it/s]\n",
      "--- EPOCH 7/30 ---\n",
      "train_batch (Avg. Loss 1.887, Accuracy 33.0): 100%|██████████| 40/40 [00:00<00:00, 46.49it/s]\n",
      "test_batch (Avg. Loss 2.104, Accuracy 27.0): 100%|██████████| 40/40 [00:00<00:00, 141.44it/s]\n",
      "train_batch (Avg. Loss 1.711, Accuracy 40.0): 100%|██████████| 40/40 [00:00<00:00, 46.49it/s]\n",
      "test_batch (Avg. Loss 2.087, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 134.51it/s]\n",
      "train_batch (Avg. Loss 1.715, Accuracy 40.0): 100%|██████████| 40/40 [00:00<00:00, 45.90it/s]\n",
      "test_batch (Avg. Loss 2.105, Accuracy 27.0): 100%|██████████| 40/40 [00:00<00:00, 119.53it/s]\n",
      "train_batch (Avg. Loss 1.647, Accuracy 45.0): 100%|██████████| 40/40 [00:00<00:00, 43.32it/s]\n",
      "test_batch (Avg. Loss 2.091, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 114.16it/s]\n",
      "train_batch (Avg. Loss 1.602, Accuracy 45.0): 100%|██████████| 40/40 [00:00<00:00, 46.26it/s]\n",
      "test_batch (Avg. Loss 2.069, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 125.32it/s]\n",
      "train_batch (Avg. Loss 1.590, Accuracy 44.0): 100%|██████████| 40/40 [00:00<00:00, 44.76it/s]\n",
      "test_batch (Avg. Loss 2.066, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 148.25it/s]\n",
      "--- EPOCH 13/30 ---\n",
      "train_batch (Avg. Loss 1.568, Accuracy 44.0): 100%|██████████| 40/40 [00:00<00:00, 46.77it/s]\n",
      "test_batch (Avg. Loss 2.049, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 143.47it/s]\n",
      "train_batch (Avg. Loss 1.504, Accuracy 46.0): 100%|██████████| 40/40 [00:00<00:00, 46.81it/s]\n",
      "test_batch (Avg. Loss 2.070, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 139.71it/s]\n",
      "train_batch (Avg. Loss 1.467, Accuracy 50.0): 100%|██████████| 40/40 [00:00<00:00, 45.78it/s]\n",
      "test_batch (Avg. Loss 2.078, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 118.68it/s]\n",
      "train_batch (Avg. Loss 1.437, Accuracy 50.0): 100%|██████████| 40/40 [00:00<00:00, 44.67it/s]\n",
      "test_batch (Avg. Loss 2.082, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 120.36it/s]\n",
      "train_batch (Avg. Loss 1.429, Accuracy 51.0): 100%|██████████| 40/40 [00:00<00:00, 45.29it/s]\n",
      "test_batch (Avg. Loss 2.062, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 130.78it/s]\n",
      "train_batch (Avg. Loss 1.387, Accuracy 53.0): 100%|██████████| 40/40 [00:00<00:00, 44.47it/s]\n",
      "test_batch (Avg. Loss 2.065, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 123.53it/s]\n",
      "--- EPOCH 19/30 ---\n",
      "train_batch (Avg. Loss 1.358, Accuracy 53.0): 100%|██████████| 40/40 [00:00<00:00, 46.78it/s]\n",
      "test_batch (Avg. Loss 2.072, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 131.40it/s]\n",
      "train_batch (Avg. Loss 1.359, Accuracy 56.0): 100%|██████████| 40/40 [00:00<00:00, 47.15it/s]\n",
      "test_batch (Avg. Loss 2.068, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 147.01it/s]\n",
      "train_batch (Avg. Loss 1.329, Accuracy 55.0): 100%|██████████| 40/40 [00:00<00:00, 47.39it/s]\n",
      "test_batch (Avg. Loss 2.065, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 151.00it/s]\n",
      "train_batch (Avg. Loss 1.284, Accuracy 55.0): 100%|██████████| 40/40 [00:00<00:00, 47.80it/s]\n",
      "test_batch (Avg. Loss 2.071, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 125.97it/s]\n",
      "train_batch (Avg. Loss 1.298, Accuracy 55.0): 100%|██████████| 40/40 [00:00<00:00, 45.01it/s]\n",
      "test_batch (Avg. Loss 2.058, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 131.54it/s]\n",
      "train_batch (Avg. Loss 1.248, Accuracy 59.0): 100%|██████████| 40/40 [00:00<00:00, 45.52it/s]\n",
      "test_batch (Avg. Loss 2.068, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 129.46it/s]\n",
      "--- EPOCH 25/30 ---\n",
      "train_batch (Avg. Loss 1.160, Accuracy 60.0): 100%|██████████| 40/40 [00:00<00:00, 46.71it/s]\n",
      "test_batch (Avg. Loss 2.061, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 140.87it/s]\n",
      "train_batch (Avg. Loss 1.224, Accuracy 58.0): 100%|██████████| 40/40 [00:00<00:00, 47.78it/s]\n",
      "test_batch (Avg. Loss 2.053, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 147.00it/s]\n",
      "train_batch (Avg. Loss 1.141, Accuracy 61.0): 100%|██████████| 40/40 [00:00<00:00, 49.53it/s]\n",
      "test_batch (Avg. Loss 2.055, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 147.91it/s]\n",
      "train_batch (Avg. Loss 1.105, Accuracy 66.0): 100%|██████████| 40/40 [00:00<00:00, 46.73it/s]\n",
      "test_batch (Avg. Loss 2.062, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 143.42it/s]\n",
      "train_batch (Avg. Loss 1.105, Accuracy 62.0): 100%|██████████| 40/40 [00:00<00:00, 47.37it/s]\n",
      "test_batch (Avg. Loss 2.066, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 133.39it/s]\n",
      "--- EPOCH 30/30 ---\n",
      "train_batch (Avg. Loss 1.086, Accuracy 66.0): 100%|██████████| 40/40 [00:00<00:00, 46.06it/s]\n",
      "test_batch (Avg. Loss 2.066, Accuracy 30.0): 100%|██████████| 40/40 [00:00<00:00, 131.97it/s]\n",
      "*** Training with dropout= 0.8\n",
      "--- EPOCH 1/30 ---\n",
      "train_batch (Avg. Loss 3.670, Accuracy 11.0): 100%|██████████| 40/40 [00:00<00:00, 44.78it/s]\n",
      "test_batch (Avg. Loss 2.337, Accuracy 15.0): 100%|██████████| 40/40 [00:00<00:00, 124.79it/s]\n",
      "train_batch (Avg. Loss 3.146, Accuracy 8.0): 100%|██████████| 40/40 [00:00<00:00, 46.40it/s]\n",
      "test_batch (Avg. Loss 2.298, Accuracy 12.0): 100%|██████████| 40/40 [00:00<00:00, 140.52it/s]\n",
      "train_batch (Avg. Loss 2.779, Accuracy 12.0): 100%|██████████| 40/40 [00:00<00:00, 47.70it/s]\n",
      "test_batch (Avg. Loss 2.274, Accuracy 14.0): 100%|██████████| 40/40 [00:00<00:00, 141.92it/s]\n",
      "train_batch (Avg. Loss 2.560, Accuracy 14.0): 100%|██████████| 40/40 [00:00<00:00, 48.12it/s]\n",
      "test_batch (Avg. Loss 2.229, Accuracy 14.0): 100%|██████████| 40/40 [00:00<00:00, 137.93it/s]\n",
      "train_batch (Avg. Loss 2.394, Accuracy 16.0): 100%|██████████| 40/40 [00:00<00:00, 48.43it/s]\n",
      "test_batch (Avg. Loss 2.223, Accuracy 14.0): 100%|██████████| 40/40 [00:00<00:00, 143.26it/s]\n",
      "train_batch (Avg. Loss 2.384, Accuracy 16.0): 100%|██████████| 40/40 [00:00<00:00, 47.67it/s]\n",
      "test_batch (Avg. Loss 2.209, Accuracy 16.0): 100%|██████████| 40/40 [00:00<00:00, 123.14it/s]\n",
      "--- EPOCH 7/30 ---\n",
      "train_batch (Avg. Loss 2.224, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 46.00it/s]\n",
      "test_batch (Avg. Loss 2.201, Accuracy 17.0): 100%|██████████| 40/40 [00:00<00:00, 124.79it/s]\n",
      "train_batch (Avg. Loss 2.218, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 47.45it/s]\n",
      "test_batch (Avg. Loss 2.193, Accuracy 16.0): 100%|██████████| 40/40 [00:00<00:00, 119.89it/s]\n",
      "train_batch (Avg. Loss 2.151, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 45.89it/s]\n",
      "test_batch (Avg. Loss 2.188, Accuracy 17.0): 100%|██████████| 40/40 [00:00<00:00, 137.16it/s]\n",
      "train_batch (Avg. Loss 2.200, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 48.48it/s]\n",
      "test_batch (Avg. Loss 2.176, Accuracy 19.0): 100%|██████████| 40/40 [00:00<00:00, 156.30it/s]\n",
      "train_batch (Avg. Loss 2.175, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 47.57it/s]\n",
      "test_batch (Avg. Loss 2.170, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 143.57it/s]\n",
      "train_batch (Avg. Loss 2.191, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 47.85it/s]\n",
      "test_batch (Avg. Loss 2.172, Accuracy 19.0): 100%|██████████| 40/40 [00:00<00:00, 140.44it/s]\n",
      "--- EPOCH 13/30 ---\n",
      "train_batch (Avg. Loss 2.130, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 47.98it/s]\n",
      "test_batch (Avg. Loss 2.160, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 128.58it/s]\n",
      "train_batch (Avg. Loss 2.121, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 47.98it/s]\n",
      "test_batch (Avg. Loss 2.159, Accuracy 20.0): 100%|██████████| 40/40 [00:00<00:00, 123.18it/s]\n",
      "train_batch (Avg. Loss 2.082, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 47.09it/s]\n",
      "test_batch (Avg. Loss 2.152, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 120.20it/s]\n",
      "train_batch (Avg. Loss 2.114, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 47.76it/s]\n",
      "test_batch (Avg. Loss 2.145, Accuracy 21.0): 100%|██████████| 40/40 [00:00<00:00, 131.89it/s]\n",
      "train_batch (Avg. Loss 2.060, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 47.23it/s]\n",
      "test_batch (Avg. Loss 2.137, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 145.73it/s]\n",
      "train_batch (Avg. Loss 2.058, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 50.01it/s]\n",
      "test_batch (Avg. Loss 2.134, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 138.67it/s]\n",
      "--- EPOCH 19/30 ---\n",
      "train_batch (Avg. Loss 2.094, Accuracy 25.0): 100%|██████████| 40/40 [00:00<00:00, 48.64it/s]\n",
      "test_batch (Avg. Loss 2.129, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 130.25it/s]\n",
      "train_batch (Avg. Loss 2.042, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 47.56it/s]\n",
      "test_batch (Avg. Loss 2.124, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 121.07it/s]\n",
      "train_batch (Avg. Loss 2.051, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 47.97it/s]\n",
      "test_batch (Avg. Loss 2.123, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 129.07it/s]\n",
      "train_batch (Avg. Loss 1.995, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 47.53it/s]\n",
      "test_batch (Avg. Loss 2.114, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 121.13it/s]\n",
      "train_batch (Avg. Loss 2.003, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 47.37it/s]\n",
      "test_batch (Avg. Loss 2.113, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 138.83it/s]\n",
      "train_batch (Avg. Loss 1.974, Accuracy 31.0): 100%|██████████| 40/40 [00:00<00:00, 49.53it/s]\n",
      "test_batch (Avg. Loss 2.104, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 141.19it/s]\n",
      "--- EPOCH 25/30 ---\n",
      "train_batch (Avg. Loss 2.001, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 44.68it/s]\n",
      "test_batch (Avg. Loss 2.105, Accuracy 23.0): 100%|██████████| 40/40 [00:00<00:00, 147.97it/s]\n",
      "train_batch (Avg. Loss 1.959, Accuracy 29.0): 100%|██████████| 40/40 [00:00<00:00, 45.83it/s]\n",
      "test_batch (Avg. Loss 2.104, Accuracy 22.0): 100%|██████████| 40/40 [00:00<00:00, 146.78it/s]\n",
      "train_batch (Avg. Loss 1.978, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 48.51it/s]\n",
      "test_batch (Avg. Loss 2.088, Accuracy 25.0): 100%|██████████| 40/40 [00:00<00:00, 120.02it/s]\n",
      "train_batch (Avg. Loss 1.939, Accuracy 34.0): 100%|██████████| 40/40 [00:00<00:00, 46.94it/s]\n",
      "test_batch (Avg. Loss 2.084, Accuracy 26.0): 100%|██████████| 40/40 [00:00<00:00, 128.75it/s]\n",
      "train_batch (Avg. Loss 1.962, Accuracy 28.0): 100%|██████████| 40/40 [00:00<00:00, 47.80it/s]\n",
      "test_batch (Avg. Loss 2.083, Accuracy 25.0): 100%|██████████| 40/40 [00:00<00:00, 121.61it/s]\n",
      "--- EPOCH 30/30 ---\n",
      "train_batch (Avg. Loss 1.922, Accuracy 33.0): 100%|██████████| 40/40 [00:00<00:00, 47.31it/s]\n",
      "test_batch (Avg. Loss 2.088, Accuracy 24.0): 100%|██████████| 40/40 [00:00<00:00, 151.49it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "torch.manual_seed(seed)\n",
    "fig=None\n",
    "for dropout in [0, 0.4, 0.8]:\n",
    "#for dropout in [0]:\n",
    "    model = models.MLP(in_features, num_classes, hidden_features, wstd=hp['wstd'], dropout=dropout)\n",
    "    loss_fn = blocks.CrossEntropyLoss()\n",
    "    optimizer = optimizers.MomentumSGD(model.params(), learn_rate=hp['lr'], reg=0)\n",
    "\n",
    "    print('*** Training with dropout=', dropout)\n",
    "    trainer = training.BlocksTrainer(model, loss_fn, optimizer)\n",
    "    fit_res_dropout = trainer.fit(dl_train, dl_test, num_epochs, max_batches=max_batches, print_every=6)\n",
    "    fig, axes = plot_fit(fit_res_dropout, fig=fig, legend=f'dropout={dropout}', log_loss=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Questions\n",
    "<a id=part2_7></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**TODO** Answer the following questions. Write your answers in the appropriate variables in the module `hw2/answers.py`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "from cs236605.answers import display_answer\n",
    "import hw2.answers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 1 \n",
    "\n",
    "Regarding the graphs you got for the three dropout configurations:\n",
    "\n",
    "1. Explain the graphs of no-dropout vs dropout. Do they match what you expected to see?\n",
    "    - If yes, explain why and provide examples based on the graphs.\n",
    "    - If no, explain what you think the problem is and what should be modified to fix it.\n",
    "\n",
    "2. Compare the low-dropout setting to the high-dropout setting and explain based on your graphs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "\n",
       "1. The graphs we received matched our expectations. We were expecting to see better generalization in the dropout \n",
       "setting. We can see that without dropout our model overfits the training data and fails to generalize, we can see \n",
       "from the graphs  \n",
       "that the training loss decreasing gradually in contrast to the test loss which starts to increase after a few \n",
       "iterations and is not stable. In the dropout setting we can see that both the training loss and the test loss are \n",
       "decreasing gradually together resulting in higher test accuracy which implies better generalizations.\n",
       "2. We see that higher dropout rates result in slower learning and requires more iterations to converge.\n",
       "In our case we see that lower dropout rate leads to better results in terms of speed of convergence and \n",
       "generalization meaning higher test accuracy. \n",
       "Since we perform the experiment with a small number of epochs, we can't tell if higher dropout rate will result in \n",
       "better convergence based on these results. \n"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display_answer(hw2.answers.part2_q1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Question 2 \n",
    "\n",
    "When training a model with the cross-entropy loss function, is it possible for the test loss to **increase** for a few epochs while the test accuracy also **increases**?\n",
    "\n",
    "If it's possible explain how, if it's not explain why not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "\n",
       "It is possible for the test loss to increase for a few epochs while the test accuracy also increases. Since Cross \n",
       "Entropy loss takes into account the scores of all the categories, in situations that the model is suddenly less sure \n",
       "of its choice and gives the correct labels lower scores compared to other labels, the loss will increase but \n",
       "the accuracy will remain the same. Furthermore, the model could get more labels correct as a consequence of the \n",
       "effect we \n",
       "described of less sure choices and that will cause the test loss to increase while the test accuracy also increases.\n"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display_answer(hw2.answers.part2_q2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}