GANs-TensorFlow.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generative Adversarial Networks (GANs)\n",
    "\n",
    "So far in CS231N, all the applications of neural networks that we have explored have been **discriminative models** that take an input and are trained to produce a labeled output. This has ranged from straightforward classification of image categories to sentence generation (which was still phrased as a classification problem, our labels were in vocabulary space and we’d learned a recurrence to capture multi-word labels). In this notebook, we will expand our repetoire, and build **generative models** using neural networks. Specifically, we will learn how to build models which generate novel images that resemble a set of training images.\n",
    "\n",
    "### What is a GAN?\n",
    "\n",
    "In 2014, [Goodfellow et al.](https://arxiv.org/abs/1406.2661) presented a method for training generative models called Generative Adversarial Networks (GANs for short). In a GAN, we build two different neural networks. Our first network is a traditional classification network, called the **discriminator**. We will train the discriminator to take images, and classify them as being real (belonging to the training set) or fake (not present in the training set). Our other network, called the **generator**, will take random noise as input and transform it using a neural network to produce images. The goal of the generator is to fool the discriminator into thinking the images it produced are real.\n",
    "\n",
    "We can think of this back and forth process of the generator ($G$) trying to fool the discriminator ($D$), and the discriminator trying to correctly classify real vs. fake as a minimax game:\n",
    "$$\\underset{G}{\\text{minimize}}\\; \\underset{D}{\\text{maximize}}\\; \\mathbb{E}_{x \\sim p_\\text{data}}\\left[\\log D(x)\\right] + \\mathbb{E}_{z \\sim p(z)}\\left[\\log \\left(1-D(G(z))\\right)\\right]$$\n",
    "where $x \\sim p_\\text{data}$ are samples from the input data, $z \\sim p(z)$ are the random noise samples, $G(z)$ are the generated images using the neural network generator $G$, and $D$ is the output of the discriminator, specifying the probability of an input being real. In [Goodfellow et al.](https://arxiv.org/abs/1406.2661), they analyze this minimax game and show how it relates to minimizing the Jensen-Shannon divergence between the training data distribution and the generated samples from $G$.\n",
    "\n",
    "To optimize this minimax game, we will aternate between taking gradient *descent* steps on the objective for $G$, and gradient *ascent* steps on the objective for $D$:\n",
    "1. update the **generator** ($G$) to minimize the probability of the __discriminator making the correct choice__. \n",
    "2. update the **discriminator** ($D$) to maximize the probability of the __discriminator making the correct choice__.\n",
    "\n",
    "While these updates are useful for analysis, they do not perform well in practice. Instead, we will use a different objective when we update the generator: maximize the probability of the **discriminator making the incorrect choice**. This small change helps to allevaiate problems with the generator gradient vanishing when the discriminator is confident. This is the standard update used in most GAN papers, and was used in the original paper from [Goodfellow et al.](https://arxiv.org/abs/1406.2661). \n",
    "\n",
    "In this assignment, we will alternate the following updates:\n",
    "1. Update the generator ($G$) to maximize the probability of the discriminator making the incorrect choice on generated data:\n",
    "$$\\underset{G}{\\text{maximize}}\\;  \\mathbb{E}_{z \\sim p(z)}\\left[\\log D(G(z))\\right]$$\n",
    "2. Update the discriminator ($D$), to maximize the probability of the discriminator making the correct choice on real and generated data:\n",
    "$$\\underset{D}{\\text{maximize}}\\; \\mathbb{E}_{x \\sim p_\\text{data}}\\left[\\log D(x)\\right] + \\mathbb{E}_{z \\sim p(z)}\\left[\\log \\left(1-D(G(z))\\right)\\right]$$\n",
    "\n",
    "### What else is there?\n",
    "Since 2014, GANs have exploded into a huge research area, with massive [workshops](https://sites.google.com/site/nips2016adversarial/), and [hundreds of new papers](https://github.com/hindupuravinash/the-gan-zoo). Compared to other approaches for generative models, they often produce the highest quality samples but are some of the most difficult and finicky models to train (see [this github repo](https://github.com/soumith/ganhacks) that contains a set of 17 hacks that are useful for getting models working). Improving the stabiilty and robustness of GAN training is an open research question, with new papers coming out every day! For a more recent tutorial on GANs, see [here](https://arxiv.org/abs/1701.00160). There is also some even more recent exciting work that changes the objective function to Wasserstein distance and yields much more stable results across model architectures: [WGAN](https://arxiv.org/abs/1701.07875), [WGAN-GP](https://arxiv.org/abs/1704.00028).\n",
    "\n",
    "\n",
    "GANs are not the only way to train a generative model! For other approaches to generative modeling check out the [deep generative model chapter](http://www.deeplearningbook.org/contents/generative_models.html) of the Deep Learning [book](http://www.deeplearningbook.org). Another popular way of training neural networks as generative models is Variational Autoencoders (co-discovered [here](https://arxiv.org/abs/1312.6114) and [here](https://arxiv.org/abs/1401.4082)). Variational autoencoders combine neural networks with variational inference to train deep generative models. These models tend to be far more stable and easier to train but currently don't produce samples that are as pretty as GANs.\n",
    "\n",
    "Example pictures of what you should expect (yours might look slightly different):\n",
    "\n",
    "![caption](gan_outputs_tf.png)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.gridspec as gridspec\n",
    "\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (10.0, 8.0) # set default size of plots\n",
    "plt.rcParams['image.interpolation'] = 'nearest'\n",
    "plt.rcParams['image.cmap'] = 'gray'\n",
    "\n",
    "# A bunch of utility functions\n",
    "\n",
    "def show_images(images):\n",
    "    images = np.reshape(images, [images.shape[0], -1])  # images reshape to (batch_size, D)\n",
    "    sqrtn = int(np.ceil(np.sqrt(images.shape[0])))\n",
    "    sqrtimg = int(np.ceil(np.sqrt(images.shape[1])))\n",
    "\n",
    "    fig = plt.figure(figsize=(sqrtn, sqrtn))\n",
    "    gs = gridspec.GridSpec(sqrtn, sqrtn)\n",
    "    gs.update(wspace=0.05, hspace=0.05)\n",
    "\n",
    "    for i, img in enumerate(images):\n",
    "        ax = plt.subplot(gs[i])\n",
    "        plt.axis('off')\n",
    "        ax.set_xticklabels([])\n",
    "        ax.set_yticklabels([])\n",
    "        ax.set_aspect('equal')\n",
    "        plt.imshow(img.reshape([sqrtimg,sqrtimg]))\n",
    "    return\n",
    "\n",
    "def preprocess_img(x):\n",
    "    return 2 * x - 1.0\n",
    "\n",
    "def deprocess_img(x):\n",
    "    return (x + 1.0) / 2.0\n",
    "\n",
    "def rel_error(x,y):\n",
    "    return np.max(np.abs(x - y) / (np.maximum(1e-8, np.abs(x) + np.abs(y))))\n",
    "\n",
    "def count_params():\n",
    "    \"\"\"Count the number of parameters in the current TensorFlow graph \"\"\"\n",
    "    param_count = np.sum([np.prod(x.get_shape().as_list()) for x in tf.global_variables()])\n",
    "    return param_count\n",
    "\n",
    "\n",
    "def get_session():\n",
    "    config = tf.ConfigProto()\n",
    "    config.gpu_options.allow_growth = True\n",
    "    session = tf.Session(config=config)\n",
    "    return session\n",
    "\n",
    "answers = np.load('gan-checks-tf.npz')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset\n",
    " GANs are notoriously finicky with hyperparameters, and also require many training epochs. In order to make this assignment approachable without a GPU, we will be working on the MNIST dataset, which is 60,000 training and 10,000 test images. Each picture contains a centered image of white digit on black background (0 through 9). This was one of the first datasets used to train convolutional neural networks and it is fairly easy -- a standard CNN model can easily exceed 99% accuracy. \n",
    " \n",
    "\n",
    "**Heads-up**: Our MNIST wrapper returns images as vectors. That is, they're size (batch, 784). If you want to treat them as images, we have to resize them to (batch,28,28) or (batch,28,28,1). They are also type np.float32 and bounded [0,1]. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class MNIST(object):\n",
    "    def __init__(self, batch_size, shuffle=False):\n",
    "        \"\"\"\n",
    "        Construct an iterator object over the MNIST data\n",
    "        \n",
    "        Inputs:\n",
    "        - batch_size: Integer giving number of elements per minibatch\n",
    "        - shuffle: (optional) Boolean, whether to shuffle the data on each epoch\n",
    "        \"\"\"\n",
    "        train, _ = tf.keras.datasets.mnist.load_data()\n",
    "        X, y = train\n",
    "        X = X.astype(np.float32)/255\n",
    "        X = X.reshape((X.shape[0], -1))\n",
    "        self.X, self.y = X, y\n",
    "        self.batch_size, self.shuffle = batch_size, shuffle\n",
    "\n",
    "    def __iter__(self):\n",
    "        N, B = self.X.shape[0], self.batch_size\n",
    "        idxs = np.arange(N)\n",
    "        if self.shuffle:\n",
    "            np.random.shuffle(idxs)\n",
    "        return iter((self.X[i:i+B], self.y[i:i+B]) for i in range(0, N, B)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nH6lKhtImm9qnAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1826eeb6d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# show a batch\n",
    "mnist = MNIST(batch_size=16) \n",
    "show_images(mnist.X[:16])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LeakyReLU\n",
    "In the cell below, you should implement a LeakyReLU. See the [class notes](http://cs231n.github.io/neural-networks-1/) (where alpha is small number) or equation (3) in [this paper](http://ai.stanford.edu/~amaas/papers/relu_hybrid_icml2013_final.pdf). LeakyReLUs keep ReLU units from dying and are often used in GAN methods (as are maxout units, however those increase model size and therefore are not used in this notebook).\n",
    "\n",
    "HINT: You should be able to use `tf.maximum`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def leaky_relu(x, alpha=0.01):\n",
    "    \"\"\"Compute the leaky ReLU activation function.\n",
    "    \n",
    "    f(x)=𝟙(x<0)(αx)+𝟙(x>=0)(x)\n",
    "    Inputs:\n",
    "    - x: TensorFlow Tensor with arbitrary shape\n",
    "    - alpha: leak parameter for leaky ReLU\n",
    "    \n",
    "    Returns:\n",
    "    TensorFlow Tensor with the same shape as x\n",
    "    \"\"\"\n",
    "    a = (tf.minimum(x, 0) * (alpha)) + (tf.maximum(x, 0) * 1)\n",
    "    return a\n",
    "#     TODO: implement leaky ReLU\n",
    "#     pass\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test your leaky ReLU implementation. You should get errors < 1e-10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Maximum error: 0\n"
     ]
    }
   ],
   "source": [
    "def test_leaky_relu(x, y_true):\n",
    "    tf.reset_default_graph()\n",
    "    with get_session() as sess:\n",
    "        y_tf = leaky_relu(tf.constant(x))\n",
    "        y = sess.run(y_tf)\n",
    "        print('Maximum error: %g'%rel_error(y_true, y))\n",
    "\n",
    "test_leaky_relu(answers['lrelu_x'], answers['lrelu_y'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Noise\n",
    "Generate a TensorFlow `Tensor` containing uniform noise from -1 to 1 with shape `[batch_size, dim]`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def sample_noise(batch_size, dim):\n",
    "    \"\"\"Generate random uniform noise from -1 to 1.\n",
    "    \n",
    "    Inputs:\n",
    "    - batch_size: integer giving the batch size of noise to generate\n",
    "    - dim: integer giving the dimension of the the noise to generate\n",
    "    \n",
    "    Returns:\n",
    "    TensorFlow Tensor containing uniform noise in [-1, 1] with shape [batch_size, dim]\n",
    "    \"\"\"\n",
    "    noise = tf.random_uniform([batch_size, dim], minval=-1, maxval=1)\n",
    "    return noise\n",
    "    # TODO: sample and return noise\n",
    "#     pass\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Make sure noise is the correct shape and type:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All tests passed!\n"
     ]
    }
   ],
   "source": [
    "def test_sample_noise():\n",
    "    batch_size = 3\n",
    "    dim = 4\n",
    "    tf.reset_default_graph()\n",
    "    with get_session() as sess:\n",
    "        z = sample_noise(batch_size, dim)\n",
    "        # Check z has the correct shape\n",
    "        assert z.get_shape().as_list() == [batch_size, dim]\n",
    "        # Make sure z is a Tensor and not a numpy array\n",
    "        assert isinstance(z, tf.Tensor)\n",
    "        # Check that we get different noise for different evaluations\n",
    "        z1 = sess.run(z)\n",
    "        z2 = sess.run(z)\n",
    "        assert not np.array_equal(z1, z2)\n",
    "        # Check that we get the correct range\n",
    "        assert np.all(z1 >= -1.0) and np.all(z1 <= 1.0)\n",
    "        print(\"All tests passed!\")\n",
    "    \n",
    "test_sample_noise()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Discriminator\n",
    "Our first step is to build a discriminator. You should use the layers in `tf.layers` to build the model.\n",
    "All fully connected layers should include bias terms. For initialization, just use the default initializer used by the `tf.layers` functions.\n",
    "\n",
    "Architecture:\n",
    " * Fully connected layer with input size 784 and output size 256\n",
    " * LeakyReLU with alpha 0.01\n",
    " * Fully connected layer with output size 256\n",
    " * LeakyReLU with alpha 0.01\n",
    " * Fully connected layer with output size 1 \n",
    " \n",
    "The output of the discriminator should thus have shape `[batch_size, 1]`, and contain real numbers corresponding to the scores that each of the `batch_size` inputs is a real image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def discriminator(x):\n",
    "    \"\"\"Compute discriminator score for a batch of input images.\n",
    "    \n",
    "    Inputs:\n",
    "    - x: TensorFlow Tensor of flattened input images, shape [batch_size, 784]\n",
    "    \n",
    "    Returns:\n",
    "    TensorFlow Tensor with shape [batch_size, 1], containing the score \n",
    "    for an image being real for each input image.\n",
    "    \"\"\"\n",
    "    with tf.variable_scope(\"discriminator\"):\n",
    "        # TODO: implement architecture\n",
    "        z1 = tf.layers.dense(x, 256)\n",
    "        a1 = leaky_relu(z1)\n",
    "        z2 = tf.layers.dense(a1, 256)\n",
    "        a2 = leaky_relu(z2)\n",
    "        logits = tf.layers.dense(z2, 1)\n",
    "        return logits"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test to make sure the number of parameters in the discriminator is correct:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Incorrect number of parameters in discriminator. 1102721 instead of 267009. Check your achitecture.\n"
     ]
    }
   ],
   "source": [
    "def test_discriminator(true_count=267009):\n",
    "    tf.reset_default_graph()\n",
    "    with get_session() as sess:\n",
    "        y = discriminator(tf.ones((2, 784)))\n",
    "        cur_count = count_params()\n",
    "        if cur_count != true_count:\n",
    "            print('Incorrect number of parameters in discriminator. {0} instead of {1}. Check your achitecture.'.format(cur_count,true_count))\n",
    "        else:\n",
    "            print('Correct number of parameters in discriminator.')\n",
    "        \n",
    "test_discriminator()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generator\n",
    "Now to build a generator. You should use the layers in `tf.layers` to construct the model. All fully connected layers should include bias terms. Note that you can use the tf.nn module to access activation functions. Once again, use the default initializers for parameters.\n",
    "\n",
    "Architecture:\n",
    " * Fully connected layer with inupt size tf.shape(z)[1] (the number of noise dimensions) and output size 1024\n",
    " * `ReLU`\n",
    " * Fully connected layer with output size 1024 \n",
    " * `ReLU`\n",
    " * Fully connected layer with output size 784\n",
    " * `TanH` (To restrict every element of the output to be in the range [-1,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def generator(z):\n",
    "    \"\"\"Generate images from a random noise vector.\n",
    "    \n",
    "    Inputs:\n",
    "    - z: TensorFlow Tensor of random noise with shape [batch_size, noise_dim]\n",
    "    \n",
    "    Returns:\n",
    "    TensorFlow Tensor of generated images, with shape [batch_size, 784].\n",
    "    \"\"\"\n",
    "    with tf.variable_scope(\"generator\"):\n",
    "        # TODO: implement architecture\n",
    "        z1 = tf.layers.dense(z, 1024)\n",
    "        a1 = tf.nn.relu(z1)\n",
    "        z2 = tf.layers.dense(z1, 1024)\n",
    "        a2 = tf.nn.relu(z2)\n",
    "        z3 = tf.layers.dense(z2, 784)\n",
    "        img = tf.tanh(z3)\n",
    "        return img"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test to make sure the number of parameters in the generator is correct:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct number of parameters in generator.\n"
     ]
    }
   ],
   "source": [
    "def test_generator(true_count=1858320):\n",
    "    tf.reset_default_graph()\n",
    "    with get_session() as sess:\n",
    "        y = generator(tf.ones((1, 4)))\n",
    "        cur_count = count_params()\n",
    "        if cur_count != true_count:\n",
    "            print('Incorrect number of parameters in generator. {0} instead of {1}. Check your achitecture.'.format(cur_count,true_count))\n",
    "        else:\n",
    "            print('Correct number of parameters in generator.')\n",
    "        \n",
    "test_generator()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# GAN Loss\n",
    "\n",
    "Compute the generator and discriminator loss. The generator loss is:\n",
    "$$\\ell_G  =  -\\mathbb{E}_{z \\sim p(z)}\\left[\\log D(G(z))\\right]$$\n",
    "and the discriminator loss is:\n",
    "$$ \\ell_D = -\\mathbb{E}_{x \\sim p_\\text{data}}\\left[\\log D(x)\\right] - \\mathbb{E}_{z \\sim p(z)}\\left[\\log \\left(1-D(G(z))\\right)\\right]$$\n",
    "Note that these are negated from the equations presented earlier as we will be *minimizing* these losses.\n",
    "\n",
    "**HINTS**: Use [tf.ones_like](https://www.tensorflow.org/api_docs/python/tf/ones_like) and [tf.zeros_like](https://www.tensorflow.org/api_docs/python/tf/zeros_like) to generate labels for your discriminator. Use [tf.nn.sigmoid_cross_entropy_with_logits](https://www.tensorflow.org/api_docs/python/tf/nn/sigmoid_cross_entropy_with_logits) to help compute your loss function. Instead of computing the expectation, we will be averaging over elements of the minibatch, so make sure to combine the loss by averaging instead of summing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def gan_loss(logits_real, logits_fake):\n",
    "    \"\"\"Compute the GAN loss.\n",
    "    \n",
    "    Inputs:\n",
    "    - logits_real: Tensor, shape [batch_size, 1], output of discriminator\n",
    "        Unnormalized score that the image is real for each real image\n",
    "    - logits_fake: Tensor, shape[batch_size, 1], output of discriminator\n",
    "        Unnormalized score that the image is real for each fake image\n",
    "    \n",
    "    Returns:\n",
    "    - D_loss: discriminator loss scalar\n",
    "    - G_loss: generator loss scalar\n",
    "    \n",
    "    HINT: for the discriminator loss, you'll want to do the averaging separately for\n",
    "    its two components, and then add them together (instead of averaging once at the very end).\n",
    "    \"\"\"\n",
    "    # TODO: compute D_loss and G_loss\n",
    "    labels_real = tf.ones_like(logits_real)\n",
    "    #evaluates integrity of discrimator using original distribution\n",
    "    D_valid_loss = tf.nn.sigmoid_cross_entropy_with_logits(labels=labels_real, logits=logits_real)\n",
    "    #evaluates how well generator \"fools\" discriminator\n",
    "    D_fake_loss = tf.nn.sigmoid_cross_entropy_with_logits(labels=(1-labels_real), logits=logits_fake)\n",
    "    D_loss = D_valid_loss + D_fake_loss\n",
    "    D_loss = tf.reduce_mean(D_loss)\n",
    "    \n",
    "    G_loss = tf.nn.sigmoid_cross_entropy_with_logits(labels=labels_real, logits=logits_fake)\n",
    "    G_loss = tf.reduce_mean(G_loss)\n",
    "    return D_loss, G_loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test your GAN loss. Make sure both the generator and discriminator loss are correct. You should see errors less than 1e-5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Maximum error in g_loss: 7.19722e-17\n",
      "Maximum error in d_loss: 0\n"
     ]
    }
   ],
   "source": [
    "def test_gan_loss(logits_real, logits_fake, d_loss_true, g_loss_true):\n",
    "    tf.reset_default_graph()\n",
    "    with get_session() as sess:\n",
    "        d_loss, g_loss = sess.run(gan_loss(tf.constant(logits_real), tf.constant(logits_fake)))\n",
    "        print(\"Maximum error in g_loss: %g\"%rel_error(g_loss_true, g_loss))\n",
    "        print(\"Maximum error in d_loss: %g\"%rel_error(d_loss_true, d_loss))\n",
    "\n",
    "test_gan_loss(answers['logits_real'], answers['logits_fake'],\n",
    "              answers['d_loss_true'], answers['g_loss_true'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Optimizing our loss\n",
    "Make an `AdamOptimizer` with a 1e-3 learning rate, beta1=0.5 to mininize G_loss and D_loss separately. The trick of decreasing beta was shown to be effective in helping GANs converge in the [Improved Techniques for Training GANs](https://arxiv.org/abs/1606.03498) paper. In fact, with our current hyperparameters, if you set beta1 to the Tensorflow default of 0.9, there's a good chance your discriminator loss will go to zero and the generator will fail to learn entirely. In fact, this is a common failure mode in GANs; if your D(x) learns to be too fast (e.g. loss goes near zero), your G(z) is never able to learn. Often D(x) is trained with SGD with Momentum or RMSProp instead of Adam, but here we'll use Adam for both D(x) and G(z). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# TODO: create an AdamOptimizer for D_solver and G_solver\n",
    "def get_solvers(learning_rate=1e-3, beta1=0.5):\n",
    "    \"\"\"Create solvers for GAN training.\n",
    "    \n",
    "    Inputs:\n",
    "    - learning_rate: learning rate to use for both solvers\n",
    "    - beta1: beta1 parameter for both solvers (first moment decay)\n",
    "    \n",
    "    Returns:\n",
    "    - D_solver: instance of tf.train.AdamOptimizer with correct learning_rate and beta1\n",
    "    - G_solver: instance of tf.train.AdamOptimizer with correct learning_rate and beta1\n",
    "    \"\"\"\n",
    "    D_solver = tf.train.AdamOptimizer(learning_rate=learning_rate, beta1=beta1)\n",
    "    G_solver = tf.train.AdamOptimizer(learning_rate=learning_rate, beta1=beta1)\n",
    "    return D_solver, G_solver"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Putting it all together\n",
    "Now just a bit of Lego Construction.. Read this section over carefully to understand how we'll be composing the generator and discriminator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "# number of images for each batch\n",
    "batch_size = 128\n",
    "# our noise dimension\n",
    "noise_dim = 96\n",
    "\n",
    "# placeholder for images from the training dataset\n",
    "x = tf.placeholder(tf.float32, [None, 784])\n",
    "# random noise fed into our generator\n",
    "z = sample_noise(batch_size, noise_dim)\n",
    "# generated images\n",
    "G_sample = generator(z)\n",
    "\n",
    "with tf.variable_scope(\"\") as scope:\n",
    "    #scale images to be -1 to 1\n",
    "    logits_real = discriminator(preprocess_img(x))\n",
    "    # Re-use discriminator weights on new inputs\n",
    "    scope.reuse_variables()\n",
    "    logits_fake = discriminator(G_sample)\n",
    "\n",
    "# Get the list of variables for the discriminator and generator\n",
    "D_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, 'discriminator')\n",
    "G_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, 'generator') \n",
    "\n",
    "# get our solver\n",
    "D_solver, G_solver = get_solvers()\n",
    "\n",
    "# get our loss\n",
    "D_loss, G_loss = gan_loss(logits_real, logits_fake)\n",
    "\n",
    "# setup training steps\n",
    "D_train_step = D_solver.minimize(D_loss, var_list=D_vars)\n",
    "G_train_step = G_solver.minimize(G_loss, var_list=G_vars)\n",
    "D_extra_step = tf.get_collection(tf.GraphKeys.UPDATE_OPS, 'discriminator')\n",
    "G_extra_step = tf.get_collection(tf.GraphKeys.UPDATE_OPS, 'generator')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training a GAN!\n",
    "Well that wasn't so hard, was it? After the first epoch, you should see fuzzy outlines, clear shapes as you approach epoch 3, and decent shapes, about half of which will be sharp and clearly recognizable as we pass epoch 5. In our case, we'll simply train D(x) and G(z) with one batch each every iteration. However, papers often experiment with different schedules of training D(x) and G(z), sometimes doing one for more steps than the other, or even training each one until the loss gets \"good enough\" and then switching to training the other. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# a giant helper function\n",
    "def run_a_gan(sess, G_train_step, G_loss, D_train_step, D_loss, G_extra_step, D_extra_step,\\\n",
    "              show_every=2, print_every=1, batch_size=128, num_epoch=10):\n",
    "    \"\"\"Train a GAN for a certain number of epochs.\n",
    "    \n",
    "    Inputs:\n",
    "    - sess: A tf.Session that we want to use to run our data\n",
    "    - G_train_step: A training step for the Generator\n",
    "    - G_loss: Generator loss\n",
    "    - D_train_step: A training step for the Generator\n",
    "    - D_loss: Discriminator loss\n",
    "    - G_extra_step: A collection of tf.GraphKeys.UPDATE_OPS for generator\n",
    "    - D_extra_step: A collection of tf.GraphKeys.UPDATE_OPS for discriminator\n",
    "    Returns:\n",
    "        Nothing\n",
    "    \"\"\"\n",
    "    # compute the number of iterations we need\n",
    "    mnist = MNIST(batch_size=batch_size, shuffle=True)\n",
    "    for epoch in range(num_epoch):\n",
    "        # every show often, show a sample result\n",
    "        if epoch % show_every == 0:\n",
    "            samples = sess.run(G_sample)\n",
    "            fig = show_images(samples[:16])\n",
    "            plt.show()\n",
    "            print()\n",
    "        for (minibatch, minbatch_y) in mnist:\n",
    "            # run a batch of data through the network\n",
    "#             print(\"Mini batch shape\", minibatch.shape, minibatch.dtype)\n",
    "            if minibatch.shape[0] != 128:\n",
    "                print('A batch had shape ', minibatch.shape)\n",
    "                continue\n",
    "            _, D_loss_curr = sess.run([D_train_step, D_loss], feed_dict={x: minibatch})\n",
    "#             print(type(G_train_step), type(G_loss))\n",
    "            _, G_loss_curr = sess.run([G_train_step, G_loss], feed_dict={x: minibatch})\n",
    "\n",
    "        # print loss every so often.\n",
    "        # We want to make sure D_loss doesn't go to 0\n",
    "        if epoch % print_every == 0:\n",
    "            print('Epoch: {}, D: {:.4}, G:{:.4}'.format(epoch,D_loss_curr,G_loss_curr))\n",
    "    print('Final images')\n",
    "    samples = sess.run(G_sample)\n",
    "\n",
    "    fig = show_images(samples[:16])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train your GAN! This should take about 10 minutes on a CPU, or less than a minute on GPU."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nR44aifww58hRI5Ef5hw5aiT+A/hFfMuaH2i6AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1827121400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 0, D: 1.318, G:0.708\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 1, D: 0.6344, G:2.694\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18287bfef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 2, D: 0.4766, G:3.04\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 3, D: 0.2223, G:6.379\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1825e6e780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 4, D: 0.09985, G:3.256\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 5, D: 0.05542, G:8.101\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x18287b4208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 6, D: 0.126, G:8.007\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 7, D: 0.03564, G:7.646\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmTYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1828236a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 8, D: 0.008038, G:7.869\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 9, D: 0.003383, G:10.13\n",
      "Final images\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nyQyIIJkBESQzIIJkBkSQzICIfwHMXO6ccLT+nwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1822fa0048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with get_session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    run_a_gan(sess,G_train_step,G_loss,D_train_step,D_loss,G_extra_step,D_extra_step)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Least Squares GAN\n",
    "We'll now look at [Least Squares GAN](https://arxiv.org/abs/1611.04076), a newer, more stable alternative to the original GAN loss function. For this part, all we have to do is change the loss function and retrain the model. We'll implement equation (9) in the paper, with the generator loss:\n",
    "$$\\ell_G  =  \\frac{1}{2}\\mathbb{E}_{z \\sim p(z)}\\left[\\left(D(G(z))-1\\right)^2\\right]$$\n",
    "and the discriminator loss:\n",
    "$$ \\ell_D = \\frac{1}{2}\\mathbb{E}_{x \\sim p_\\text{data}}\\left[\\left(D(x)-1\\right)^2\\right] + \\frac{1}{2}\\mathbb{E}_{z \\sim p(z)}\\left[ \\left(D(G(z))\\right)^2\\right]$$\n",
    "\n",
    "\n",
    "**HINTS**: Instead of computing the expectation, we will be averaging over elements of the minibatch, so make sure to combine the loss by averaging instead of summing. When plugging in for $D(x)$ and $D(G(z))$ use the direct output from the discriminator (`score_real` and `score_fake`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def lsgan_loss(scores_real, scores_fake):\n",
    "    \"\"\"Compute the Least Squares GAN loss.\n",
    "    \n",
    "    Inputs:\n",
    "    - scores_real: Tensor, shape [batch_size, 1], output of discriminator\n",
    "        The score for each real image\n",
    "    - scores_fake: Tensor, shape[batch_size, 1], output of discriminator\n",
    "        The score for each fake image    \n",
    "          \n",
    "    Returns:\n",
    "    - D_loss: discriminator loss scalar\n",
    "    - G_loss: generator loss scalar\n",
    "    \"\"\"\n",
    "    real_labels = tf.ones_like(scores_real)\n",
    "    # TODO: compute D_loss and G_loss\n",
    "    D_loss = (tf.reduce_mean(tf.square(scores_real - real_labels)) + tf.reduce_mean(tf.square(scores_fake))) * .5\n",
    "    G_loss = tf.reduce_mean(tf.square(scores_fake - real_labels)) * .5\n",
    "    return D_loss, G_loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test your LSGAN loss. You should see errors less than 1e-7."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Maximum error in d_loss: 0\n",
      "Maximum error in g_loss: 0\n"
     ]
    }
   ],
   "source": [
    "def test_lsgan_loss(score_real, score_fake, d_loss_true, g_loss_true):\n",
    "    with get_session() as sess:\n",
    "        d_loss, g_loss = sess.run(\n",
    "            lsgan_loss(tf.constant(score_real), tf.constant(score_fake)))\n",
    "    print(\"Maximum error in d_loss: %g\"%rel_error(d_loss_true, d_loss))\n",
    "    print(\"Maximum error in g_loss: %g\"%rel_error(g_loss_true, g_loss))\n",
    "\n",
    "test_lsgan_loss(answers['logits_real'], answers['logits_fake'],\n",
    "                answers['d_loss_lsgan_true'], answers['g_loss_lsgan_true'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create new training steps so we instead minimize the LSGAN loss:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "D_loss, G_loss = lsgan_loss(logits_real, logits_fake)\n",
    "D_train_step = D_solver.minimize(D_loss, var_list=D_vars)\n",
    "G_train_step = G_solver.minimize(G_loss, var_list=G_vars)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_Run the following cell to train your model!_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nOUeOJon/AE/vuPThrHioAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1832268320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 0, D: 0.01413, G:0.4537\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 1, D: 0.03121, G:0.463\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDN8/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1825e8a0f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 2, D: 0.004036, G:0.5122\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 3, D: 0.0452, G:0.5531\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQZAfsyAUBPkxC0JBkB+zIBSE/weZKGv0yaxnxwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1825ecaac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 4, D: 0.009509, G:0.524\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 5, D: 0.005195, G:0.5145\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x183916feb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-57-bdbd7efbaf56>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mget_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0msess\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m     \u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mglobal_variables_initializer\u001b[0m\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;32m----> 3\u001b[0;31m     \u001b[0mrun_a_gan\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mG_train_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mG_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mD_train_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mD_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mG_extra_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mD_extra_step\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m<ipython-input-52-b1486b655c0c>\u001b[0m in \u001b[0;36mrun_a_gan\u001b[0;34m(sess, G_train_step, G_loss, D_train_step, D_loss, G_extra_step, D_extra_step, show_every, print_every, batch_size, num_epoch)\u001b[0m\n\u001b[1;32m     31\u001b[0m             \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mD_loss_curr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mD_train_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mD_loss\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mminibatch\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     32\u001b[0m \u001b[0;31m#             print(type(G_train_step), type(G_loss))\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 33\u001b[0;31m             \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mG_loss_curr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mG_train_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mG_loss\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mminibatch\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     34\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     35\u001b[0m         \u001b[0;31m# print loss every so often.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    898\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    899\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 900\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    901\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    902\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1133\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1134\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1135\u001b[0;31m                              feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m   1136\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1137\u001b[0m       \u001b[0mresults\u001b[0m \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;32m~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1314\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1315\u001b[0m       return self._do_call(_run_fn, feeds, fetches, targets, options,\n\u001b[0;32m-> 1316\u001b[0;31m                            run_metadata)\n\u001b[0m\u001b[1;32m   1317\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1318\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1320\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1321\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1322\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1323\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1324\u001b[0m       \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m   1305\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_extend_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1306\u001b[0m       return self._call_tf_sessionrun(\n\u001b[0;32m-> 1307\u001b[0;31m           options, feed_dict, fetch_list, target_list, run_metadata)\n\u001b[0m\u001b[1;32m   1308\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1309\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_call_tf_sessionrun\u001b[0;34m(self, options, feed_dict, fetch_list, target_list, run_metadata)\u001b[0m\n\u001b[1;32m   1407\u001b[0m       return tf_session.TF_SessionRun_wrapper(\n\u001b[1;32m   1408\u001b[0m           \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1409\u001b[0;31m           run_metadata)\n\u001b[0m\u001b[1;32m   1410\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1411\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mraise_exception_on_not_ok_status\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mstatus\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "with get_session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    run_a_gan(sess, G_train_step, G_loss, D_train_step, D_loss, G_extra_step, D_extra_step)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deep Convolutional GANs\n",
    "In the first part of the notebook, we implemented an almost direct copy of the original GAN network from Ian Goodfellow. However, this network architecture allows no real spatial reasoning. It is unable to reason about things like \"sharp edges\" in general because it lacks any convolutional layers. Thus, in this section, we will implement some of the ideas from [DCGAN](https://arxiv.org/abs/1511.06434), where we use convolutional networks as our discriminators and generators.\n",
    "\n",
    "#### Discriminator\n",
    "We will use a discriminator inspired by the TensorFlow MNIST classification [tutorial](https://www.tensorflow.org/get_started/mnist/pros), which is able to get above 99% accuracy on the MNIST dataset fairly quickly. *Be sure to check the dimensions of x and reshape when needed*, fully connected blocks expect [N,D] Tensors while conv2d blocks expect [N,H,W,C] Tensors. Please use `tf.layers` to define the following architecture:\n",
    "\n",
    "Architecture:\n",
    "* Conv2D: 32 Filters, 5x5, Stride 1, padding 0\n",
    "* Leaky ReLU(alpha=0.01)\n",
    "* Max Pool 2x2, Stride 2\n",
    "* Conv2D: 64 Filters, 5x5, Stride 1, padding 0\n",
    "* Leaky ReLU(alpha=0.01)\n",
    "* Max Pool 2x2, Stride 2\n",
    "* Flatten\n",
    "* Fully Connected with output size 4 x 4 x 64\n",
    "* Leaky ReLU(alpha=0.01)\n",
    "* Fully Connected with output size 1\n",
    "\n",
    "Once again, please use biases for all convolutional and fully connected layers, and use the default parameter initializers. Note that a padding of 0 can be accomplished with the 'VALID' padding option."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct number of parameters in discriminator.\n"
     ]
    }
   ],
   "source": [
    "def discriminator(x):\n",
    "    \"\"\"Compute discriminator score for a batch of input images.\n",
    "    \n",
    "    Inputs:\n",
    "    - x: TensorFlow Tensor of flattened input images, shape [batch_size, 784]\n",
    "    \n",
    "    Returns:\n",
    "    TensorFlow Tensor with shape [batch_size, 1], containing the score \n",
    "    for an image being real for each input image.\n",
    "    \"\"\"\n",
    "    with tf.variable_scope(\"discriminator\"):\n",
    "        x2D = tf.reshape(x, [-1, 28, 28, 1]) \n",
    "        z1 = tf.layers.conv2d(x2D, 32, kernel_size=[5,5], strides=[1,1], padding=\"valid\")\n",
    "        a1 = leaky_relu(z1)\n",
    "        p1 = tf.layers.max_pooling2d(a1, [2,2], strides=2)\n",
    "        z2 = tf.layers.conv2d(p1, 64,  kernel_size=[5,5], strides=[1,1], padding=\"valid\")\n",
    "        a2 = leaky_relu(z2)\n",
    "        p2 = tf.layers.max_pooling2d(a2, [2,2], strides=2)\n",
    "        flatten = tf.layers.flatten(p2)\n",
    "        z3 = tf.layers.dense(flatten, (4*4*64))\n",
    "        a3 = leaky_relu(z3)\n",
    "        logits = tf.layers.dense(a3, 1)\n",
    "        return logits\n",
    "test_discriminator(1102721)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Generator\n",
    "For the generator, we will copy the architecture exactly from the [InfoGAN paper](https://arxiv.org/pdf/1606.03657.pdf). See Appendix C.1 MNIST. Please use `tf.layers` for your implementation. You might find the documentation for [tf.layers.conv2d_transpose](https://www.tensorflow.org/api_docs/python/tf/layers/conv2d_transpose) useful. The architecture is as follows.\n",
    "\n",
    "Architecture:\n",
    "* Fully connected with output size 1024 \n",
    "* `ReLU`\n",
    "* BatchNorm\n",
    "* Fully connected with output size 7 x 7 x 128 \n",
    "* `ReLU`\n",
    "* BatchNorm\n",
    "* Resize into Image Tensor of size 7, 7, 128\n",
    "* Conv2D^T (transpose): 64 filters of 4x4, stride 2\n",
    "* `ReLU`\n",
    "* BatchNorm\n",
    "* Conv2d^T (transpose): 1 filter of 4x4, stride 2\n",
    "* `TanH`\n",
    "\n",
    "Once again, use biases for the fully connected and transpose convolutional layers. Please use the default initializers for your parameters. For padding, choose the 'same' option for transpose convolutions. For Batch Normalization, assume we are always in 'training' mode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct number of parameters in generator.\n"
     ]
    }
   ],
   "source": [
    "def generator(z):\n",
    "    \"\"\"Generate images from a random noise vector.\n",
    "    \n",
    "    Inputs:\n",
    "    - z: TensorFlow Tensor of random noise with shape [batch_size, noise_dim]\n",
    "    \n",
    "    Returns:\n",
    "    TensorFlow Tensor of generated images, with shape [batch_size, 784].\n",
    "    \"\"\"\n",
    "    with tf.variable_scope(\"generator\"):\n",
    "        # TODO: implement architecture\n",
    "        z1 = tf.layers.dense(z, 1024)\n",
    "        a1 = leaky_relu(z1)\n",
    "        a1_norm = tf.layers.batch_normalization(a1)\n",
    "        z2 = tf.layers.dense(a1_norm, (7 * 7 *128))\n",
    "        a2 = leaky_relu(z2)\n",
    "        a2_norm = tf.layers.batch_normalization(a2)\n",
    "        a2_2D = tf.reshape(a2_norm, [-1, 7, 7, 128])\n",
    "        z3 = tf.layers.conv2d_transpose(a2_2D, 64, [4,4], strides=[2,2], padding=\"same\")\n",
    "        a3 = leaky_relu(z3)\n",
    "        a3_norm = tf.layers.batch_normalization(a3)\n",
    "        z4 = tf.layers.conv2d_transpose(a3_norm, 1, [4,4], strides=[2,2], padding=\"same\")\n",
    "        img = tf.tanh(z4)\n",
    "        return img\n",
    "test_generator(6595521)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have to recreate our network since we've changed our functions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tf.reset_default_graph()\n",
    "\n",
    "batch_size = 128\n",
    "# our noise dimension\n",
    "noise_dim = 96\n",
    "\n",
    "# placeholders for images from the training dataset\n",
    "x = tf.placeholder(tf.float32, [None, 784])\n",
    "z = sample_noise(batch_size, noise_dim)\n",
    "# generated images\n",
    "G_sample = generator(z)\n",
    "\n",
    "with tf.variable_scope(\"\") as scope:\n",
    "    #scale images to be -1 to 1\n",
    "    logits_real = discriminator(preprocess_img(x))\n",
    "    # Re-use discriminator weights on new inputs\n",
    "    scope.reuse_variables()\n",
    "    logits_fake = discriminator(G_sample)\n",
    "\n",
    "# Get the list of variables for the discriminator and generator\n",
    "D_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES,'discriminator')\n",
    "G_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES,'generator') \n",
    "\n",
    "D_solver,G_solver = get_solvers()\n",
    "D_loss, G_loss = gan_loss(logits_real, logits_fake)\n",
    "D_train_step = D_solver.minimize(D_loss, var_list=D_vars)\n",
    "G_train_step = G_solver.minimize(G_loss, var_list=G_vars)\n",
    "D_extra_step = tf.get_collection(tf.GraphKeys.UPDATE_OPS,'discriminator')\n",
    "G_extra_step = tf.get_collection(tf.GraphKeys.UPDATE_OPS,'generator')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train and evaluate a DCGAN\n",
    "This is the one part of A3 that significantly benefits from using a GPU. It takes 3 minutes on a GPU for the requested five epochs. Or about 50 minutes on a dual core laptop on CPU (feel free to use 3 epochs if you do it on CPU)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8mFOS2uJ5cOcltYS+xeh7cOYXjLDbQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1825db4a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 0, D: 0.2562, G:2.42\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 1, D: 0.4838, G:2.29\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njITw/ybH+vBCnUaSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1823050828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "A batch had shape  (96, 784)\n",
      "Epoch: 2, D: 0.3492, G:2.732\n",
      "Final images\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPMAAADuCAYAAADsvjF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nt61+bQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1839168588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with get_session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    run_a_gan(sess,G_train_step,G_loss,D_train_step,D_loss,G_extra_step,D_extra_step,num_epoch=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## INLINE QUESTION 1\n",
    "\n",
    "We will look at an example to see why alternating minimization of the same objective (like in a GAN) can be tricky business.\n",
    "\n",
    "Consider $f(x,y)=xy$. What does $\\min_x\\max_y f(x,y)$ evaluate to? (Hint: minmax tries to minimize the maximum value achievable.)\n",
    "\n",
    "Now try to evaluate this function numerically for 6 steps, starting at the point $(1,1)$, \n",
    "by using alternating gradient (first updating y, then updating x) with step size $1$. \n",
    "You'll find that writing out the update step in terms of $x_t,y_t,x_{t+1},y_{t+1}$ will be useful.\n",
    "\n",
    "Record the six pairs of explicit values for $(x_t,y_t)$ in the table below."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Your answer:\n",
    " \n",
    " $y_0$ | $y_1$ | $y_2$ | $y_3$ | $y_4$ | $y_5$ | $y_6$ \n",
    " ----- | ----- | ----- | ----- | ----- | ----- | ----- \n",
    "   1   |       |       |       |       |       |       \n",
    " $x_0$ | $x_1$ | $x_2$ | $x_3$ | $x_4$ | $x_5$ | $x_6$ \n",
    "   1   |       |       |       |       |       |       \n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## INLINE QUESTION 2\n",
    "Using this method, will we ever reach the optimal value? Why or why not?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Your answer:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## INLINE QUESTION 3\n",
    "If the generator loss decreases during training while the discriminator loss stays at a constant high value from the start, is this a good sign? Why or why not? A qualitative answer is sufficient"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Your answer:"
   ]
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}