DCGAN_SVHN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deep Convolutional GANs\n",
    "\n",
    "In this notebook, you'll build a GAN using convolutional layers in the generator and discriminator. This is called a Deep Convolutional GAN, or DCGAN for short. The DCGAN architecture was first explored in 2016 and has seen impressive results in generating new images; you can read the [original paper, here](https://arxiv.org/pdf/1511.06434.pdf).\n",
    "\n",
    "You'll be training DCGAN on the [Street View House Numbers](http://ufldl.stanford.edu/housenumbers/) (SVHN) dataset. These are color images of house numbers collected from Google street view. SVHN images are in color and much more variable than MNIST. \n",
    "\n",
    "<img src='assets/svhn_dcgan.png' width=80% />\n",
    "\n",
    "So, our goal is to create a DCGAN that can generate new, realistic-looking images of house numbers. We'll go through the following steps to do this:\n",
    "* Load in and pre-process the house numbers dataset\n",
    "* Define discriminator and generator networks\n",
    "* Train these adversarial networks\n",
    "* Visualize the loss over time and some sample, generated images\n",
    "\n",
    "#### Deeper Convolutional Networks\n",
    "\n",
    "Since this dataset is more complex than our MNIST data, we'll need a deeper network to accurately identify patterns in these images and be able to generate new ones. Specifically, we'll use a series of convolutional or transpose convolutional layers in the discriminator and generator. It's also necessary to use batch normalization to get these convolutional networks to train. \n",
    "\n",
    "Besides these changes in network structure, training the discriminator and generator networks should be the same as before. That is, the discriminator will alternate training on real and fake (generated) images, and the generator will aim to trick the discriminator into thinking that its generated images are real!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import libraries\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pickle as pkl\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Getting the data\n",
    "\n",
    "Here you can download the SVHN dataset. It's a dataset built-in to the PyTorch datasets library. We can load in training data, transform it into Tensor datatypes, then create dataloaders to batch our data into a desired size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "0it [00:00, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading http://ufldl.stanford.edu/housenumbers/train_32x32.mat to data/train_32x32.mat\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████▉| 181690368/182040794 [01:33<00:00, 2432496.25it/s]"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "from torchvision import datasets\n",
    "from torchvision import transforms\n",
    "\n",
    "# Tensor transform\n",
    "transform = transforms.ToTensor()\n",
    "\n",
    "# SVHN training datasets\n",
    "svhn_train = datasets.SVHN(root='data/', split='train', download=True, transform=transform)\n",
    "\n",
    "batch_size = 128\n",
    "num_workers = 0\n",
    "\n",
    "# build DataLoaders for SVHN dataset\n",
    "train_loader = torch.utils.data.DataLoader(dataset=svhn_train,\n",
    "                                          batch_size=batch_size,\n",
    "                                          shuffle=True,\n",
    "                                          num_workers=num_workers)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize the Data\n",
    "\n",
    "Here I'm showing a small sample of the images. Each of these is 32x32 with 3 color channels (RGB). These are the real, training images that we'll pass to the discriminator. Notice that each image has _one_ associated, numerical label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x288 with 20 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "182042624it [01:50, 2432496.25it/s]                                                                                    "
     ]
    }
   ],
   "source": [
    "# obtain one batch of training images\n",
    "dataiter = iter(train_loader)\n",
    "images, labels = dataiter.next()\n",
    "\n",
    "# plot the images in the batch, along with the corresponding labels\n",
    "fig = plt.figure(figsize=(25, 4))\n",
    "plot_size=20\n",
    "for idx in np.arange(plot_size):\n",
    "    ax = fig.add_subplot(2, plot_size/2, idx+1, xticks=[], yticks=[])\n",
    "    ax.imshow(np.transpose(images[idx], (1, 2, 0)))\n",
    "    # print out the correct label for each image\n",
    "    # .item() gets the value contained in a Tensor\n",
    "    ax.set_title(str(labels[idx].item()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pre-processing: scaling from -1 to 1\n",
    "\n",
    "We need to do a bit of pre-processing; we know that the output of our `tanh` activated generator will contain pixel values in a range from -1 to 1, and so, we need to rescale our training images to a range of -1 to 1. (Right now, they are in a range from 0-1.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min:  tensor(0.1098)\n",
      "Max:  tensor(0.6824)\n"
     ]
    }
   ],
   "source": [
    "# current range\n",
    "img = images[0]\n",
    "\n",
    "print('Min: ', img.min())\n",
    "print('Max: ', img.max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# helper scale function\n",
    "def scale(x, feature_range=(-1, 1)):\n",
    "    ''' Scale takes in an image x and returns that image, scaled\n",
    "       with a feature_range of pixel values from -1 to 1. \n",
    "       This function assumes that the input x is already scaled from 0-1.'''\n",
    "    # assume x is scaled to (0, 1)\n",
    "    # scale to feature_range and return scaled x\n",
    "    min, max = feature_range\n",
    "    x = x*(max-min) + min\n",
    "    return x\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scaled min:  tensor(-0.7804)\n",
      "Scaled max:  tensor(0.3647)\n"
     ]
    }
   ],
   "source": [
    "# scaled range\n",
    "scaled_img = scale(img)\n",
    "\n",
    "print('Scaled min: ', scaled_img.min())\n",
    "print('Scaled max: ', scaled_img.max())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "# Define the Model\n",
    "\n",
    "A GAN is comprised of two adversarial networks, a discriminator and a generator."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Discriminator\n",
    "\n",
    "Here you'll build the discriminator. This is a convolutional classifier like you've built before, only without any maxpooling layers. \n",
    "* The inputs to the discriminator are 32x32x3 tensor images\n",
    "* You'll want a few convolutional, hidden layers\n",
    "* Then a fully connected layer for the output; as before, we want a sigmoid output, but we'll add that in the loss function, [BCEWithLogitsLoss](https://pytorch.org/docs/stable/nn.html#bcewithlogitsloss), later\n",
    "\n",
    "<img src='assets/conv_discriminator.png' width=80%/>\n",
    "\n",
    "For the depths of the convolutional layers I suggest starting with 32 filters in the first layer, then double that depth as you add layers (to 64, 128, etc.). Note that in the DCGAN paper, they did all the downsampling using only strided convolutional layers with no maxpooling layers.\n",
    "\n",
    "You'll also want to use batch normalization with [nn.BatchNorm2d](https://pytorch.org/docs/stable/nn.html#batchnorm2d) on each layer **except** the first convolutional layer and final, linear output layer. \n",
    "\n",
    "#### Helper `conv` function \n",
    "\n",
    "In general, each layer should look something like convolution > batch norm > leaky ReLU, and so we'll define a function to put these layers together. This function will create a sequential series of a convolutional + an optional batch norm layer. We'll create these using PyTorch's [Sequential container](https://pytorch.org/docs/stable/nn.html#sequential), which takes in a list of layers and creates layers according to the order that they are passed in to the Sequential constructor.\n",
    "\n",
    "Note: It is also suggested that you use a **kernel_size of 4** and a **stride of 2** for strided convolutions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "# helper conv function\n",
    "def conv(in_channels, out_channels, kernel_size, stride=2, padding=1, batch_norm=True):\n",
    "    \"\"\"Creates a convolutional layer, with optional batch normalization.\n",
    "    \"\"\"\n",
    "    layers = []\n",
    "    conv_layer = nn.Conv2d(in_channels, out_channels, \n",
    "                           kernel_size, stride, padding, bias=False)\n",
    "    \n",
    "    # append conv layer\n",
    "    layers.append(conv_layer)\n",
    "\n",
    "    if batch_norm:\n",
    "        # append batchnorm layer\n",
    "        layers.append(nn.BatchNorm2d(out_channels))\n",
    "     \n",
    "    # using Sequential container\n",
    "    return nn.Sequential(*layers)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Discriminator(nn.Module):\n",
    "\n",
    "    def __init__(self, conv_dim=32):\n",
    "        super(Discriminator, self).__init__()\n",
    "\n",
    "        # complete init function\n",
    "        self.conv_dim = conv_dim\n",
    "        \n",
    "        self.conv1 = conv(3, conv_dim, 4, batch_norm=False) # first layer, no batch norm\n",
    "        self.conv2 = conv(conv_dim, conv_dim*2, 4)\n",
    "        self.conv3 = conv(conv_dim*2, conv_dim*4, 4)\n",
    "        \n",
    "        # fc layer\n",
    "        self.fc = nn.Linear(conv_dim*4*4*4, 1) # (4 by 4)\n",
    "\n",
    "    def forward(self, x):\n",
    "        # complete forward function\n",
    "        x = F.leaky_relu(self.conv1(x), 0.2)\n",
    "        x = F.leaky_relu(self.conv2(x), 0.2)\n",
    "        x = F.leaky_relu(self.conv3(x), 0.2)\n",
    "        #flatten before fc layer\n",
    "        x = x.view(-1, self.conv_dim*4*4*4)\n",
    "        x = self.fc(x)\n",
    "        \n",
    "        return x\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generator\n",
    "\n",
    "Next, you'll build the generator network. The input will be our noise vector `z`, as before. And, the output will be a $tanh$ output, but this time with size 32x32 which is the size of our SVHN images.\n",
    "\n",
    "<img src='assets/conv_generator.png' width=80% />\n",
    "\n",
    "What's new here is we'll use transpose convolutional layers to create our new images. \n",
    "* The first layer is a fully connected layer which is reshaped into a deep and narrow layer, something like 4x4x512. \n",
    "* Then, we use batch normalization and a leaky ReLU activation. \n",
    "* Next is a series of [transpose convolutional layers](https://pytorch.org/docs/stable/nn.html#convtranspose2d), where you typically halve the depth and double the width and height of the previous layer. \n",
    "* And, we'll apply batch normalization and ReLU to all but the last of these hidden layers. Where we will just apply a `tanh` activation.\n",
    "\n",
    "#### Helper `deconv` function\n",
    "\n",
    "For each of these layers, the general scheme is transpose convolution > batch norm > ReLU, and so we'll define a function to put these layers together. This function will create a sequential series of a transpose convolutional + an optional batch norm layer. We'll create these using PyTorch's Sequential container, which takes in a list of layers and creates layers according to the order that they are passed in to the Sequential constructor.\n",
    "\n",
    "Note: It is also suggested that you use a **kernel_size of 4** and a **stride of 2** for transpose convolutions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# helper deconv function\n",
    "def deconv(in_channels, out_channels, kernel_size, stride=2, padding=1, batch_norm=True):\n",
    "    \"\"\"Creates a transposed-convolutional layer, with optional batch normalization.\n",
    "    \"\"\"\n",
    "    ## TODO: Complete this function\n",
    "    ## create a sequence of transpose + optional batch norm layers\n",
    "    layers = []\n",
    "    transpose_conv_layer = nn.ConvTranspose2d(in_channels, out_channels, kernel_size, stride, padding, bias=False)\n",
    "    layers.append(transpose_conv_layer)\n",
    "    if batch_norm:\n",
    "        layers.append(nn.BatchNorm2d(out_channels))\n",
    "    \n",
    "    return nn.Sequential(*layers)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Generator(nn.Module):\n",
    "    \n",
    "    def __init__(self, z_size, conv_dim=32):\n",
    "        super(Generator, self).__init__()\n",
    "\n",
    "        # complete init function\n",
    "        self.conv_dim = conv_dim\n",
    "        \n",
    "        #1st fc layer\n",
    "        self.fc = nn.Linear(z_size, conv_dim*4*4*4)\n",
    "        \n",
    "        #transpose conv layers\n",
    "        self.t_conv1 = deconv(conv_dim*4, conv_dim*2, 4)\n",
    "        self.t_conv2 = deconv(conv_dim*2, conv_dim, 4)\n",
    "        self.t_conv3 = deconv(conv_dim, 3, 4, batch_norm=False)\n",
    "        \n",
    "\n",
    "    def forward(self, x):\n",
    "        # complete forward function\n",
    "        x = self.fc(x)\n",
    "        x = x.view(-1, self.conv_dim*4, 4, 4) # vector to the last trans conv layer dim\n",
    "        x = F.relu(self.t_conv1(x))\n",
    "        x = F.relu(self.t_conv2(x))\n",
    "        x = self.t_conv3(x)\n",
    "        #tanh activation function\n",
    "        x = F.tanh(x)\n",
    "        return x\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Build complete network\n",
    "\n",
    "Define your models' hyperparameters and instantiate the discriminator and generator from the classes defined above. Make sure you've passed in the correct input arguments."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Discriminator(\n",
      "  (conv1): Sequential(\n",
      "    (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      "  (conv2): Sequential(\n",
      "    (0): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv3): Sequential(\n",
      "    (0): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (fc): Linear(in_features=2048, out_features=1, bias=True)\n",
      ")\n",
      "\n",
      "Generator(\n",
      "  (fc): Linear(in_features=100, out_features=2048, bias=True)\n",
      "  (t_conv1): Sequential(\n",
      "    (0): ConvTranspose2d(128, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (t_conv2): Sequential(\n",
      "    (0): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (t_conv3): Sequential(\n",
      "    (0): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "# define hyperparams\n",
    "conv_dim = 32\n",
    "z_size = 100\n",
    "\n",
    "# define discriminator and generator\n",
    "D = Discriminator(conv_dim)\n",
    "G = Generator(z_size=z_size, conv_dim=conv_dim)\n",
    "\n",
    "print(D)\n",
    "print()\n",
    "print(G)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training on GPU\n",
    "\n",
    "Check if you can train on GPU. If you can, set this as a variable and move your models to GPU. \n",
    "> Later, we'll also move any inputs our models and loss functions see (real_images, z, and ground truth labels) to GPU as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GPU available for training. Models moved to GPU\n"
     ]
    }
   ],
   "source": [
    "train_on_gpu = torch.cuda.is_available()\n",
    "\n",
    "if train_on_gpu:\n",
    "    # move models to GPU\n",
    "    G.cuda()\n",
    "    D.cuda()\n",
    "    print('GPU available for training. Models moved to GPU')\n",
    "else:\n",
    "    print('Training on CPU.')\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Discriminator and Generator Losses\n",
    "\n",
    "Now we need to calculate the losses. And this will be exactly the same as before.\n",
    "\n",
    "### Discriminator Losses\n",
    "\n",
    "> * For the discriminator, the total loss is the sum of the losses for real and fake images, `d_loss = d_real_loss + d_fake_loss`. \n",
    "* Remember that we want the discriminator to output 1 for real images and 0 for fake images, so we need to set up the losses to reflect that.\n",
    "\n",
    "The losses will by binary cross entropy loss with logits, which we can get with [BCEWithLogitsLoss](https://pytorch.org/docs/stable/nn.html#bcewithlogitsloss). This combines a `sigmoid` activation function **and** and binary cross entropy loss in one function.\n",
    "\n",
    "For the real images, we want `D(real_images) = 1`. That is, we want the discriminator to classify the the real images with a label = 1, indicating that these are real. The discriminator loss for the fake data is similar. We want `D(fake_images) = 0`, where the fake images are the _generator output_, `fake_images = G(z)`. \n",
    "\n",
    "### Generator Loss\n",
    "\n",
    "The generator loss will look similar only with flipped labels. The generator's goal is to get `D(fake_images) = 1`. In this case, the labels are **flipped** to represent that the generator is trying to fool the discriminator into thinking that the images it generates (fakes) are real!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def real_loss(D_out, smooth=False):\n",
    "    batch_size = D_out.size(0)\n",
    "    # label smoothing\n",
    "    if smooth:\n",
    "        # smooth, real labels = 0.9\n",
    "        labels = torch.ones(batch_size)*0.9\n",
    "    else:\n",
    "        labels = torch.ones(batch_size) # real labels = 1\n",
    "    # move labels to GPU if available     \n",
    "    if train_on_gpu:\n",
    "        labels = labels.cuda()\n",
    "    # binary cross entropy with logits loss\n",
    "    criterion = nn.BCEWithLogitsLoss()\n",
    "    # calculate loss\n",
    "    loss = criterion(D_out.squeeze(), labels)\n",
    "    return loss\n",
    "\n",
    "def fake_loss(D_out):\n",
    "    batch_size = D_out.size(0)\n",
    "    labels = torch.zeros(batch_size) # fake labels = 0\n",
    "    if train_on_gpu:\n",
    "        labels = labels.cuda()\n",
    "    criterion = nn.BCEWithLogitsLoss()\n",
    "    # calculate loss\n",
    "    loss = criterion(D_out.squeeze(), labels)\n",
    "    return loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Optimizers\n",
    "\n",
    "Not much new here, but notice how I am using a small learning rate and custom parameters for the Adam optimizers, This is based on some research into DCGAN model convergence.\n",
    "\n",
    "### Hyperparameters\n",
    "\n",
    "GANs are very sensitive to hyperparameters. A lot of experimentation goes into finding the best hyperparameters such that the generator and discriminator don't overpower each other. Try out your own hyperparameters or read [the DCGAN paper](https://arxiv.org/pdf/1511.06434.pdf) to see what worked for them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.optim as optim\n",
    "\n",
    "# params\n",
    "lr = 0.0002\n",
    "beta1= 0.5\n",
    "beta2= 0.999 #default value\n",
    "\n",
    "# Create optimizers for the discriminator and generator\n",
    "d_optimizer = optim.Adam(D.parameters(), lr, [beta1, beta2])\n",
    "g_optimizer = optim.Adam(G.parameters(), lr, [beta1, beta2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "## Training\n",
    "\n",
    "Training will involve alternating between training the discriminator and the generator. We'll use our functions `real_loss` and `fake_loss` to help us calculate the discriminator losses in all of the following cases.\n",
    "\n",
    "### Discriminator training\n",
    "1. Compute the discriminator loss on real, training images        \n",
    "2. Generate fake images\n",
    "3. Compute the discriminator loss on fake, generated images     \n",
    "4. Add up real and fake loss\n",
    "5. Perform backpropagation + an optimization step to update the discriminator's weights\n",
    "\n",
    "### Generator training\n",
    "1. Generate fake images\n",
    "2. Compute the discriminator loss on fake images, using **flipped** labels!\n",
    "3. Perform backpropagation + an optimization step to update the generator's weights\n",
    "\n",
    "#### Saving Samples\n",
    "\n",
    "As we train, we'll also print out some loss statistics and save some generated \"fake\" samples.\n",
    "\n",
    "**Evaluation mode**\n",
    "\n",
    "Notice that, when we call our generator to create the samples to display, we set our model to evaluation mode: `G.eval()`. That's so the batch normalization layers will use the population statistics rather than the batch statistics (as they do during training), *and* so dropout layers will operate in eval() mode; not turning off any nodes for generating samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\rune0\\.conda\\envs\\deep-learning\\lib\\site-packages\\torch\\nn\\functional.py:1340: UserWarning: nn.functional.tanh is deprecated. Use torch.tanh instead.\n",
      "  warnings.warn(\"nn.functional.tanh is deprecated. Use torch.tanh instead.\")\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [    1/   30] | d_loss: 1.3578 | g_loss: 0.9402\n",
      "Epoch [    1/   30] | d_loss: 0.2279 | g_loss: 2.9116\n",
      "Epoch [    2/   30] | d_loss: 1.7193 | g_loss: 5.5932\n",
      "Epoch [    2/   30] | d_loss: 0.4322 | g_loss: 2.6141\n",
      "Epoch [    3/   30] | d_loss: 0.4372 | g_loss: 1.9959\n",
      "Epoch [    3/   30] | d_loss: 0.4984 | g_loss: 2.4603\n",
      "Epoch [    4/   30] | d_loss: 1.1271 | g_loss: 1.6687\n",
      "Epoch [    4/   30] | d_loss: 0.4098 | g_loss: 2.2619\n",
      "Epoch [    5/   30] | d_loss: 0.1147 | g_loss: 3.6923\n",
      "Epoch [    5/   30] | d_loss: 0.0653 | g_loss: 4.2997\n",
      "Epoch [    6/   30] | d_loss: 0.2417 | g_loss: 2.4517\n",
      "Epoch [    6/   30] | d_loss: 0.1484 | g_loss: 3.5433\n",
      "Epoch [    7/   30] | d_loss: 0.8368 | g_loss: 0.5590\n",
      "Epoch [    7/   30] | d_loss: 0.1787 | g_loss: 3.3486\n",
      "Epoch [    8/   30] | d_loss: 0.1723 | g_loss: 4.5625\n",
      "Epoch [    8/   30] | d_loss: 0.3191 | g_loss: 4.3946\n",
      "Epoch [    9/   30] | d_loss: 0.1376 | g_loss: 4.1575\n",
      "Epoch [    9/   30] | d_loss: 0.2169 | g_loss: 3.9370\n",
      "Epoch [   10/   30] | d_loss: 0.1345 | g_loss: 3.7850\n",
      "Epoch [   10/   30] | d_loss: 0.1981 | g_loss: 3.1825\n",
      "Epoch [   11/   30] | d_loss: 0.4047 | g_loss: 2.2627\n",
      "Epoch [   11/   30] | d_loss: 0.1011 | g_loss: 4.3230\n",
      "Epoch [   12/   30] | d_loss: 0.1272 | g_loss: 3.0066\n",
      "Epoch [   12/   30] | d_loss: 0.2228 | g_loss: 2.6053\n",
      "Epoch [   13/   30] | d_loss: 0.1498 | g_loss: 2.9051\n",
      "Epoch [   13/   30] | d_loss: 0.2341 | g_loss: 3.7581\n",
      "Epoch [   14/   30] | d_loss: 0.2417 | g_loss: 2.8523\n",
      "Epoch [   14/   30] | d_loss: 0.1615 | g_loss: 2.8892\n",
      "Epoch [   15/   30] | d_loss: 0.1239 | g_loss: 3.4043\n",
      "Epoch [   15/   30] | d_loss: 0.1448 | g_loss: 3.8219\n",
      "Epoch [   16/   30] | d_loss: 0.1357 | g_loss: 3.2968\n",
      "Epoch [   16/   30] | d_loss: 0.0836 | g_loss: 3.3356\n",
      "Epoch [   17/   30] | d_loss: 0.2878 | g_loss: 5.2880\n",
      "Epoch [   17/   30] | d_loss: 0.1690 | g_loss: 2.4908\n",
      "Epoch [   18/   30] | d_loss: 0.5283 | g_loss: 1.8490\n",
      "Epoch [   18/   30] | d_loss: 0.2010 | g_loss: 4.7170\n",
      "Epoch [   19/   30] | d_loss: 0.1282 | g_loss: 4.0315\n",
      "Epoch [   19/   30] | d_loss: 0.0828 | g_loss: 3.7975\n",
      "Epoch [   20/   30] | d_loss: 0.2604 | g_loss: 3.0439\n",
      "Epoch [   20/   30] | d_loss: 0.6546 | g_loss: 4.5854\n",
      "Epoch [   21/   30] | d_loss: 0.1999 | g_loss: 2.4039\n",
      "Epoch [   21/   30] | d_loss: 0.1182 | g_loss: 3.9559\n",
      "Epoch [   22/   30] | d_loss: 0.1775 | g_loss: 1.8451\n",
      "Epoch [   22/   30] | d_loss: 0.1168 | g_loss: 4.1989\n",
      "Epoch [   23/   30] | d_loss: 0.0807 | g_loss: 4.3557\n",
      "Epoch [   23/   30] | d_loss: 0.2308 | g_loss: 2.0808\n",
      "Epoch [   24/   30] | d_loss: 0.5799 | g_loss: 1.5615\n",
      "Epoch [   24/   30] | d_loss: 0.2053 | g_loss: 3.9608\n",
      "Epoch [   25/   30] | d_loss: 0.1214 | g_loss: 4.0234\n",
      "Epoch [   25/   30] | d_loss: 0.1626 | g_loss: 3.1226\n",
      "Epoch [   26/   30] | d_loss: 0.3092 | g_loss: 1.3677\n",
      "Epoch [   26/   30] | d_loss: 0.1366 | g_loss: 1.8619\n",
      "Epoch [   27/   30] | d_loss: 0.1001 | g_loss: 4.0665\n",
      "Epoch [   27/   30] | d_loss: 0.1265 | g_loss: 3.1218\n",
      "Epoch [   28/   30] | d_loss: 0.4663 | g_loss: 7.2471\n",
      "Epoch [   28/   30] | d_loss: 0.1548 | g_loss: 4.7897\n",
      "Epoch [   29/   30] | d_loss: 0.1647 | g_loss: 3.5853\n",
      "Epoch [   29/   30] | d_loss: 0.2097 | g_loss: 3.3415\n",
      "Epoch [   30/   30] | d_loss: 0.1217 | g_loss: 2.0347\n",
      "Epoch [   30/   30] | d_loss: 0.4070 | g_loss: 2.0527\n"
     ]
    }
   ],
   "source": [
    "import pickle as pkl\n",
    "\n",
    "# training hyperparams\n",
    "num_epochs = 30\n",
    "\n",
    "# keep track of loss and generated, \"fake\" samples\n",
    "samples = []\n",
    "losses = []\n",
    "\n",
    "print_every = 300\n",
    "\n",
    "# Get some fixed data for sampling. These are images that are held\n",
    "# constant throughout training, and allow us to inspect the model's performance\n",
    "sample_size=16\n",
    "fixed_z = np.random.uniform(-1, 1, size=(sample_size, z_size))\n",
    "fixed_z = torch.from_numpy(fixed_z).float()\n",
    "\n",
    "# train the network\n",
    "for epoch in range(num_epochs):\n",
    "    \n",
    "    for batch_i, (real_images, _) in enumerate(train_loader):\n",
    "                \n",
    "        batch_size = real_images.size(0)\n",
    "        \n",
    "        # important rescaling step\n",
    "        real_images = scale(real_images)\n",
    "        \n",
    "        # ============================================\n",
    "        #            TRAIN THE DISCRIMINATOR\n",
    "        # ============================================\n",
    "        \n",
    "        d_optimizer.zero_grad()\n",
    "        \n",
    "        # 1. Train with real images\n",
    "\n",
    "        # Compute the discriminator losses on real images \n",
    "        if train_on_gpu:\n",
    "            real_images = real_images.cuda()\n",
    "        \n",
    "        D_real = D(real_images)\n",
    "        d_real_loss = real_loss(D_real)\n",
    "        \n",
    "        # 2. Train with fake images\n",
    "        \n",
    "        # Generate fake images\n",
    "        z = np.random.uniform(-1, 1, size=(batch_size, z_size))\n",
    "        z = torch.from_numpy(z).float()\n",
    "        # move x to GPU, if available\n",
    "        if train_on_gpu:\n",
    "            z = z.cuda()\n",
    "        fake_images = G(z)\n",
    "        \n",
    "        # Compute the discriminator losses on fake images            \n",
    "        D_fake = D(fake_images)\n",
    "        d_fake_loss = fake_loss(D_fake)\n",
    "        \n",
    "        # add up loss and perform backprop\n",
    "        d_loss = d_real_loss + d_fake_loss\n",
    "        d_loss.backward()\n",
    "        d_optimizer.step()\n",
    "        \n",
    "        \n",
    "        # =========================================\n",
    "        #            TRAIN THE GENERATOR\n",
    "        # =========================================\n",
    "        g_optimizer.zero_grad()\n",
    "        \n",
    "        # 1. Train with fake images and flipped labels\n",
    "        \n",
    "        # Generate fake images\n",
    "        z = np.random.uniform(-1, 1, size=(batch_size, z_size))\n",
    "        z = torch.from_numpy(z).float()\n",
    "        if train_on_gpu:\n",
    "            z = z.cuda()\n",
    "        fake_images = G(z)\n",
    "        \n",
    "        # Compute the discriminator losses on fake images \n",
    "        # using flipped labels!\n",
    "        D_fake = D(fake_images)\n",
    "        g_loss = real_loss(D_fake) # use real loss to flip labels\n",
    "        \n",
    "        # perform backprop\n",
    "        g_loss.backward()\n",
    "        g_optimizer.step()\n",
    "\n",
    "        # Print some loss stats\n",
    "        if batch_i % print_every == 0:\n",
    "            # append discriminator loss and generator loss\n",
    "            losses.append((d_loss.item(), g_loss.item()))\n",
    "            # print discriminator and generator loss\n",
    "            print('Epoch [{:5d}/{:5d}] | d_loss: {:6.4f} | g_loss: {:6.4f}'.format(\n",
    "                    epoch+1, num_epochs, d_loss.item(), g_loss.item()))\n",
    "\n",
    "    \n",
    "    ## AFTER EACH EPOCH##    \n",
    "    # generate and save sample, fake images\n",
    "    G.eval() # for generating samples\n",
    "    if train_on_gpu:\n",
    "        fixed_z = fixed_z.cuda()\n",
    "    samples_z = G(fixed_z)\n",
    "    samples.append(samples_z)\n",
    "    G.train() # back to training mode\n",
    "\n",
    "\n",
    "# Save training generator samples\n",
    "with open('train_samples.pkl', 'wb') as f:\n",
    "    pkl.dump(samples, f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training loss\n",
    "\n",
    "Here we'll plot the training losses for the generator and discriminator, recorded after each epoch."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x22b4a0a1b48>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "losses = np.array(losses)\n",
    "plt.plot(losses.T[0], label='Discriminator', alpha=0.5)\n",
    "plt.plot(losses.T[1], label='Generator', alpha=0.5)\n",
    "plt.title(\"Training Losses\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Generator samples from training\n",
    "\n",
    "Here we can view samples of images from the generator. We'll look at the images we saved during training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# helper function for viewing a list of passed in sample images\n",
    "def view_samples(epoch, samples):\n",
    "    fig, axes = plt.subplots(figsize=(16,4), nrows=2, ncols=8, sharey=True, sharex=True)\n",
    "    for ax, img in zip(axes.flatten(), samples[epoch]):\n",
    "        img = img.detach().cpu().numpy()\n",
    "        img = np.transpose(img, (1, 2, 0))\n",
    "        img = ((img +1)*255 / (2)).astype(np.uint8) # rescale to pixel range (0-255)\n",
    "        ax.xaxis.set_visible(False)\n",
    "        ax.yaxis.set_visible(False)\n",
    "        im = ax.imshow(img.reshape((32,32,3)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = view_samples(-1, samples)"
   ]
  },
  {
   "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}