05_WGAN_MNIST.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# WGAN MNIST\n",
    "\n",
    "This notebook is for implementing `Wasserstein GAN (WGAN)` from the paper [Wasserstein GAN](https://arxiv.org/abs/1701.07875) with [Tensorflow](https://www.tensorflow.org). <br>\n",
    "[MNIST data](http:https://yann.lecun.com/exdb/mnist/) will be used.\n",
    "\n",
    "Reference: [hwalsuklee's Github](https://github.com/hwalsuklee/tensorflow-generative-model-collections), [Jonathan Hui's Blog](https://medium.com/@jonathan_hui/gan-wasserstein-gan-wgan-gp-6a1a2aa1b490)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ['CUDA_VISIBLE_DEVICES'] = '0'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "import random\n",
    "import datetime\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parameters\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "algorithm = WGAN_MNIST\n",
    "\n",
    "img_size   = 28\n",
    "\n",
    "batch_size = 128\n",
    "num_epoch  = 50\n",
    "\n",
    "n_latent = 100\n",
    "\n",
    "beta1 = 0.5\n",
    "\n",
    "learning_rate_g = 0.0002\n",
    "learning_rate_d = 0.0002\n",
    "\n",
    "show_result_epoch = 5\n",
    "\n",
    "date_time = datetime.datetime.now().strftime(\"%Y%m%d-%H-%M-%S\")\n",
    "\n",
    "load_model = False\n",
    "train_model = True\n",
    "\n",
    "save_path = \"./saved_models/\" + date_time + \"_\" + algorithm\n",
    "load_path = \"./saved_models/20190809-11-04-47_WGAN_MNIST/model/model\" \n",
    "\n",
    "# WGAN Parateter\n",
    "n_critic = 1\n",
    "c = 0.01 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import MNIST Dataset\n",
    "\n",
    "Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "mnist = tf.keras.datasets.mnist.load_data(path='mnist.npz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (60000, 28, 28)\n",
      "y_train shape: (60000,)\n",
      "x_test shape: (10000, 28, 28)\n",
      "y_test shape: (10000,)\n"
     ]
    }
   ],
   "source": [
    "x_train = mnist[0][0]\n",
    "y_train = mnist[0][1]\n",
    "x_test  = mnist[1][0]\n",
    "y_test  = mnist[1][1]\n",
    "\n",
    "print('x_train shape: {}'.format(x_train.shape))\n",
    "print('y_train shape: {}'.format(y_train.shape))\n",
    "print('x_test shape: {}'.format(x_test.shape))\n",
    "print('y_test shape: {}'.format(y_test.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def Generator(x, is_training, reuse=False):\n",
    "    with tf.variable_scope('Generator', reuse=reuse):\n",
    "        xavier_init_conv = tf.contrib.layers.xavier_initializer_conv2d()\n",
    "        xavier_init = tf.contrib.layers.xavier_initializer()\n",
    "        \n",
    "        # Project and Reshape \n",
    "        w1 = tf.get_variable('w1', [x.get_shape()[1], 7*7*1024], initializer=xavier_init)\n",
    "        b1 = tf.get_variable('b1', [7*7*1024], initializer=xavier_init)\n",
    "        \n",
    "        x_project = tf.matmul(x,w1)+b1\n",
    "        x_reshape = tf.reshape(x_project, (-1, 7, 7, 1024))\n",
    "        \n",
    "        # First deconv layer\n",
    "        h1 = tf.layers.conv2d_transpose(x_reshape,filters=512, kernel_size=5, strides=1, padding='SAME', kernel_initializer=xavier_init_conv)\n",
    "        h1 = tf.nn.leaky_relu(h1)\n",
    "        \n",
    "        # Second deconv layer\n",
    "        h2 = tf.layers.conv2d_transpose(h1,filters=256, kernel_size=5, strides=2, padding='SAME', kernel_initializer=xavier_init_conv)\n",
    "        h2 = tf.layers.batch_normalization(h2, training=is_training)\n",
    "        h2 = tf.nn.leaky_relu(h2)  \n",
    "        \n",
    "        # Third deconv layer \n",
    "        h3 = tf.layers.conv2d_transpose(h2, filters=128, kernel_size=5, strides=2, padding='SAME', kernel_initializer=xavier_init_conv)\n",
    "        h3 = tf.layers.batch_normalization(h3, training=is_training)\n",
    "        h3 = tf.nn.leaky_relu(h3)\n",
    "        \n",
    "        # Output layer\n",
    "        logits = tf.layers.conv2d_transpose(h3, filters=1, kernel_size=5, strides=1, padding='SAME', kernel_initializer=xavier_init_conv)     \n",
    "        output = tf.tanh(logits)\n",
    "                \n",
    "        return output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Discriminator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def Discriminator(x, is_training, reuse=False):\n",
    "    with tf.variable_scope('Discriminator', reuse=reuse):\n",
    "        xavier_init_conv = tf.contrib.layers.xavier_initializer_conv2d()\n",
    "        xavier_init = tf.contrib.layers.xavier_initializer()\n",
    "\n",
    "        # First conv layer\n",
    "        h1 = tf.layers.conv2d(x, filters=64, kernel_size=5, strides=2, padding='SAME', kernel_initializer=xavier_init_conv)\n",
    "        h1 = tf.nn.leaky_relu(h1)\n",
    "        \n",
    "        # Second conv layer\n",
    "        h2 = tf.layers.conv2d(h1, filters=128, kernel_size=5, strides=2, padding='SAME', kernel_initializer=xavier_init_conv)\n",
    "        h2 = tf.layers.batch_normalization(h2, training=is_training)\n",
    "        h2 = tf.nn.leaky_relu(h2)\n",
    "        \n",
    "        # Trhid conv layer\n",
    "        h3 = tf.layers.conv2d(h2, filters=256, kernel_size=5, strides=2, padding='SAME', kernel_initializer=xavier_init_conv)\n",
    "        h3 = tf.layers.batch_normalization(h3, training=is_training)\n",
    "        h3 = tf.nn.leaky_relu(h3)\n",
    "        \n",
    "        # Output layer\n",
    "        flatten = tf.reshape(h3, (-1, h3.get_shape()[1]*h3.get_shape()[2]*h3.get_shape()[3]))\n",
    "        \n",
    "        logit  = tf.layers.dense(flatten, 1, kernel_initializer=xavier_init)\n",
    "        output = tf.sigmoid(logit)  \n",
    "\n",
    "        return logit, output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GAN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def GAN(x, z, is_training):\n",
    "    # Generator\n",
    "    G = Generator(z, is_training)\n",
    "    \n",
    "    # Discriminator\n",
    "    D_logit_real, D_out_real = Discriminator(x, is_training)\n",
    "    D_logit_fake, D_out_fake = Discriminator(G, is_training, reuse=True)\n",
    "    \n",
    "    # get loss \n",
    "    ########################################### WGAN ###########################################\n",
    "    d_loss = -tf.reduce_mean(D_logit_real) + tf.reduce_mean(D_logit_fake)\n",
    "    g_loss = -tf.reduce_mean(D_logit_fake) \n",
    "    ############################################################################################\n",
    "\n",
    "    return d_loss, g_loss, G"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build Graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = tf.placeholder(tf.float32, shape=[None, img_size, img_size, 1])\n",
    "x_normalize = (tf.cast(x, tf.float32) - (255.0/2)) / (255.0/2)\n",
    "\n",
    "z = tf.placeholder(tf.float32, shape=[None, n_latent])\n",
    "\n",
    "is_training = tf.placeholder(tf.bool)\n",
    "\n",
    "d_loss, g_loss, G = GAN(x_normalize, z, is_training) \n",
    "\n",
    "# optimization\n",
    "trainable_variables = tf.trainable_variables()\n",
    "\n",
    "trainable_variables_d = [var for var in trainable_variables if var.name.startswith('Discriminator')]\n",
    "trainable_variables_g = [var for var in trainable_variables if var.name.startswith('Generator')]\n",
    "\n",
    "update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)\n",
    "with tf.control_dependencies(update_ops):\n",
    "    train_step_d = tf.train.AdamOptimizer(learning_rate_d).minimize(d_loss, var_list=trainable_variables_d)\n",
    "    train_step_g = tf.train.AdamOptimizer(learning_rate_g).minimize(g_loss, var_list=trainable_variables_g)\n",
    "\n",
    "##################### WGAN #####################\n",
    "for d_var in trainable_variables_d:\n",
    "    d_var = tf.clip_by_value(d_var, -c, c)\n",
    "################################################"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize variables\n",
    "config = tf.ConfigProto()\n",
    "config.gpu_options.allow_growth = True\n",
    "\n",
    "sess = tf.InteractiveSession(config=config)\n",
    "\n",
    "init = tf.global_variables_initializer()\n",
    "sess.run(init)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Saver = tf.train.Saver()\n",
    "\n",
    "if load_model == True:\n",
    "    Saver.restore(sess, load_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1 / G Loss: 330.15402 / D Loss: -670.21545\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 2 / G Loss: 785.24725 / D Loss: -1566.26807\n",
      "Epoch: 3 / G Loss: 1357.65955 / D Loss: -2711.60010\n",
      "Epoch: 4 / G Loss: 1577.32031 / D Loss: -2823.43311\n",
      "Epoch: 5 / G Loss: 611.21527 / D Loss: -1446.52039\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 6 / G Loss: 1355.44446 / D Loss: -719.04980\n",
      "Epoch: 7 / G Loss: 2044.65515 / D Loss: -3760.02051\n",
      "Epoch: 8 / G Loss: 2802.60034 / D Loss: -5100.10547\n",
      "Epoch: 9 / G Loss: 1175.83557 / D Loss: -3310.80518\n",
      "Epoch: 10 / G Loss: 1227.37366 / D Loss: -2064.41675\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 11 / G Loss: 3435.93237 / D Loss: -5707.10547\n",
      "Epoch: 12 / G Loss: 3696.28833 / D Loss: -6621.90918\n",
      "Epoch: 13 / G Loss: 4476.42578 / D Loss: -4728.26807\n",
      "Epoch: 14 / G Loss: 5736.16553 / D Loss: -7754.96289\n",
      "Epoch: 15 / G Loss: 5673.79150 / D Loss: -11315.77441\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 16 / G Loss: 7634.92334 / D Loss: -9851.17285\n",
      "Epoch: 17 / G Loss: 9561.72852 / D Loss: -15936.99902\n",
      "Epoch: 18 / G Loss: 9267.37793 / D Loss: -13444.48438\n",
      "Epoch: 19 / G Loss: 8825.69922 / D Loss: -19277.15234\n",
      "Epoch: 20 / G Loss: 9930.48340 / D Loss: -17144.47852\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 21 / G Loss: 10290.04004 / D Loss: -16380.88770\n",
      "Epoch: 22 / G Loss: 10937.06934 / D Loss: -15040.37891\n",
      "Epoch: 23 / G Loss: 10154.62207 / D Loss: -16068.07227\n",
      "Epoch: 24 / G Loss: 12295.48730 / D Loss: -17289.25781\n",
      "Epoch: 25 / G Loss: 11402.48926 / D Loss: -18873.01562\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 26 / G Loss: 13244.80176 / D Loss: -19926.41406\n",
      "Epoch: 27 / G Loss: 14286.44434 / D Loss: -22847.92578\n",
      "Epoch: 28 / G Loss: 10179.46191 / D Loss: -22409.81250\n",
      "Epoch: 29 / G Loss: 14648.28516 / D Loss: -27452.44141\n",
      "Epoch: 30 / G Loss: 13703.06738 / D Loss: -23433.27148\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 31 / G Loss: 18929.20312 / D Loss: -35151.83594\n",
      "Epoch: 32 / G Loss: 22658.90430 / D Loss: -38384.08203\n",
      "Epoch: 33 / G Loss: 14485.93750 / D Loss: -21329.91211\n",
      "Epoch: 34 / G Loss: 22826.25000 / D Loss: -38109.11719\n",
      "Epoch: 35 / G Loss: 10200.89648 / D Loss: -26774.98633\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 36 / G Loss: 19564.13672 / D Loss: -38826.32422\n",
      "Epoch: 37 / G Loss: 16683.44336 / D Loss: -29648.39062\n",
      "Epoch: 38 / G Loss: 14310.97949 / D Loss: -30960.68359\n",
      "Epoch: 39 / G Loss: 23721.05273 / D Loss: -21793.11523\n",
      "Epoch: 40 / G Loss: 11450.45801 / D Loss: -19286.83008\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 41 / G Loss: 26381.58008 / D Loss: -43444.85547\n",
      "Epoch: 42 / G Loss: 29631.07617 / D Loss: -52336.70312\n",
      "Epoch: 43 / G Loss: 31394.95312 / D Loss: -54473.32031\n",
      "Epoch: 44 / G Loss: 25429.70312 / D Loss: -47104.94922\n",
      "Epoch: 45 / G Loss: 32512.47656 / D Loss: -36634.25000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 46 / G Loss: 34842.50781 / D Loss: -58659.47656\n",
      "Epoch: 47 / G Loss: 41001.61328 / D Loss: -54539.04688\n",
      "Epoch: 48 / G Loss: 6317.50195 / D Loss: -33762.17188\n",
      "Epoch: 49 / G Loss: 26474.24414 / D Loss: -50399.87891\n",
      "Epoch: 50 / G Loss: 25855.01758 / D Loss: -41310.01172\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAABpCAYAAAAnQqjlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAFRtJREFUeJztnXeQFFXXh59jwAyYBQMKRkTl01IUs5hRlFctMUOpZY5l1vcVMyZeRVSkypxBUTFizqHK9KkgluCHoogkUcCs/f0x89vb9MywO7sTupfzVG3t7kxPhzO3b//uueeca1EU4TiO42SXRep9Ao7jOE7L8I7ccRwn43hH7jiOk3G8I3ccx8k43pE7juNkHO/IHcdxMo535I7jOBmn7h25mU0ys13rfR6NYWa9zGy8mf1iZq+YWacqHiv1NjGzNmb2SP5cIzPbqQbHzIJdtjazF8xslplNN7ORZtahisfLgk26mtn7ZvZj/udFM+taxeOl3iZxzOw/+Xuo2edc9448C5jZSsAo4N/ACsD7wMN1Pal08CZwODC13ieSIpYHhgNrA52AOcCd9TyhFDAFOJDcvbMSMBp4qK5nlBLMrAtwEPB9S/aTqo7czPqb2Vtm9l8zm21mX5lZz/zrk81smpkdFdu+t5l9ZGY/598fmNjfkWb2tZnNNLN/x5/UZraImZ1nZhPz748wsxVKnNq/gLFRFI2Moug3YCCwmZltWB1LzHcNqbRJFEV/RFF0QxRFbwJ/V9MGxUixXZ7Nt5Ofoyj6BRgKbFtFU8SvIa02mR1F0aQol0Zu5NrLutWzxHzXkEqbxLgZOBf4oyXXmaqOPE8P4BNgReABck/uLcl98YcDQ81s2fy284AjgfZAb+AEM9sfcsM54BbgMKAD0A5YPXacU4D9gR2BjsCP5IxajI2B/9U/URTNAybmX68FabRJGsiCXXYAxjbv8ppFam1iZrOB34CbgCtbeJ3lkEqbmNlBwO9RFD3T4iuMoqiuP8AkYNf83/2BL2PvbQJEwKqx12YC3Uvs6wbgv/m//wM8GHtvaXJPPR3rc6BX7P0OwJ/AYkX2ezswKPHaW0D/hdUmiWN8C+zkbaXgGJsCs4Dt3SYN2y0DnAj0XphtAiwHfAmsnTzn5vwsRvr4Ifb3rwBRFCVfWxbAzHoAg4BuQBtgCWBkfruOwGR9KIqiX8xsZmw/nYDHzOyf2Gt/A6sC3yXOaS7QNvFaW3L+z1qQRpukgdTaxczWBZ4FToui6I2yr6z5pNYm+f3MM7NhwHQz2yiKomnlXV6zSKNNBgL3RlE0qVlXlCCNrpVyeIDcxMmaURS1A4aR88FBbvJgDW1oZkuRG1qJycBeURS1j/0sGUVRsUY4Ftgstq9lgC7UdsjcVGplk6xRM7tYLqLpReCyKIrurcK1VIp6tZVFyKnZ1RvbsA7Uyia9gFPNbKqZTQXWBEaY2bnNOemsd+TLAbOiKPrNzLYCDo299wiwb35iow25J6DF3h8GXJG/6TCzlc1svxLHeQzoZmYHmNmS5IZYn0RRNL7C11MJamUTzGyJvD0A2pjZkmZmpbavMzWxi5mtDrwMDI2iaFgVrqOS1Momu5nZ/5jZombWFhhMzn/8eeUvqcXU6v7pRU71d8//TAGOo5lzUlnvyE8ELjWzOeQ61xF6I4qiseQmHx4i9ySdC0wDfs9vciO5J+/z+c+/S25SpIAoiqYDBwBXkGuAPYB+VbieSlATm+T5gtywdHVgTP7vqsXXt5Ba2eUYoDMw0Mzm6qcK11MJamWT9sCDwE/kggS6AHtGuQiwtFGrPmVmFEVT9UPOBfNjFEXNaiuWd7S3evKz0rOB9aIo+r96n08acJsUx+1SiNukkDTZJOuKfIGY2b5mtnTep30d8Cm52eGFFrdJcdwuhbhNCkmrTVp1Rw7sR873NAVYD+gXLSxDkNK4TYrjdinEbVJIKm2y0LhWHMdxWiutXZE7juO0erwjdxzHyTg1zew0s4XCjxNFUZNjqd0mhbhNipMmuyhdIPl7kUVy2vDvv0MdtX/++Ydy8LZSSGM2cUXuOI6TcdJYa8WJkVQ6iy++OAB//vlnwbZxFeQsvKjNVDOQQfvWb7XPpZZaCoB58+bV9HxKkUw0bq3BHa7IHcdxMk5Nww/dn1WI26QQt0lx3C6FuE1yuCJ3HMfJON6RO47jZBzvyB3HcTKOR604zgJYcslcuXVFZSyxxBIALLZY7taJzzHNnDmz4DWn9ZCMvll66aWB0BYUUaZ2UEtckTuO42SczCryRRddFIB11lkHgEMPzS3ksd9+uQU5OnfuDMAHH3wAwOOPPw7ACy+8AIQ41+++y63C5Cpq4UZq6/jjjwfglVdeAeDhhx8GoGPHjgC0b98eCO0v3m7Uph577DEABgwYAJSf2eikgw033BAIo7DLL78cgA8//BCAk08+GYB27doBIY/jggsuAODmm3OL/fz2W/XXz3BF7jiOk3EyF0cu5XT44YcDcOmllwKw+uq5dVzlpxLJDLTff8+tyqSn5Pfffw/A+PG55TcvueQSAD799NOCfTSVtMXBpiG7LW02kcraaaedADjwwAMBOOKIIwBo06aNzqXo52XDeIatVLo+ozY5derUUvvwOPIi1Kut6Pvr3bs3EJT1xhtvDMAyyyyjY873OY24kj70r7/+GoDNNsut2z5nzpxmn5vHkTuO47RyMqfI9VScMWMGEKIKmoueprNnzwbg0UcfBeCiiy5q2GbatGll7bNeiqJLly4AdOjQAQjKYptttgGCetS8gEYn5557LhBsUA3Sosi7desGwNFHHw0EP6ciD0qh+0Qjud122w2AZZddtmGbJ598EgijQvnTf/rpp1L7rLki10ijT58+APTrl1tDfMUVVwSCX3fChAkATJo0CYC2bdsCIVJjk002AcL9Usl5gHq1le7duwPw0ksvAbDCCiskzwsIcyFDhgwB4MEHHwSCTe655x4gRDrJe6DtmoMrcsdxnFaOd+SO4zgZJ3OulTXWWAOAyZMnF31f11NqkqoxNET8+OOPG1476qijAPjss8+atI9aDA3jk7qaTHnttdeAMPyNnQ8QwqOSpXE1VFx//fWBMAFcSertWll77bUBeOONN4DQjpIkJ8fVzi677DIA7r//fiC4pbbddtuGz77++uvz7UvfQ6nws2q7VuRG2XPPPRteu+qqqwDo2rVrqXMCwn2g60y6MP/66y8guGCuu+46AEaNGgWUdic1hXq1FYUb6t7X5Oc333wDwNVXXw3AU089BcD06dOBYAshF+cXX3wx3+cVEg2VD6BwRe44jpNxMqfI99prLwCeeeaZBW6n69LTUk9PqSSFAmmyarnllgOKT3p9++23QJgoa0xtVFNRSIk/8cQTDa/tscceQFDYUlNSglJVw4YNA4Iy33XXXQHYaqutABg9ejQAffv2LeeUmkS9FXn//v0BuOOOO3QMACZOnAjAO++8A4TRyHHHHQfA008/DYSEMyG1JoUOcPDBBwOhzSnEsdREYLUU+XbbbQfAscceC4TJNghtpMi56BjlnFIBmhyNjwKkTJtKvdqKrl1JhuojdL/MmjWrSftRX6IgCQUZrLTSSg3b/PHHH2Wdmytyx3GcVk7mUvRXWWWVBb6vp+fw4cOBoNyVcq0nYXLR2F122QWAe++9FwjhWBASOwYNGgTAiSeeCNQnseaQQw4B5lc8ugaVHxg3bhwAgwcPBuDHH38E4NdffwXCeUtt6nMKU0wq+9aAvsNk0obmP6TI5cfcfffdgRDmmkSKPO4jF7/88gtQe/tJCaochdpwvJ2q/X/00UcATJkyBYCXX34ZCCNSKWuN6rSEm67tgAMOAGCfffYBYNVVVwXCXMT777/fcEy1WfmW04rs9NVXX5X1uWRfovBf/d9YclklcEXuOI6TcTKhyONPMvmDSyE1OnToUAA+//xzoHH1/OyzzwKFyhxg0003BYKv8eyzzwZg7ty5TbuACqJjx20iH9xtt90GhGuReirFe++9B8CYMWOAUHBMvl0p+NaAIhGSUSmKXpGavfjii4Gg4DUnk1z0WnMtq622WsGxlCRTa6699logKHGdqxQxhFGC5lgaGzWUGp2pzey8885AiHx65JFHAFhzzTUbtr3lllsAeO6554DCKI+sobaga+/UqRMQ7jfZTKMZjd6qOUJzRe44jpNxMhG1oicahLK0ip0WivHed999geDjay4qpgQhZVdP2mOOOQaA22+/vehnqznrLr92r169Gl7TyEApxoqyaSo77LADEOLQNSrRvEIlqHfUiq5R9tOoQ/MH8g0rrV4jHqkqxdirvEHPnj2BEJceR35ixQ+XolJRK2qXUoTyyap9KnqlFkitxmPndX6KBmns3qx3WylyDCD0OZqD0KhDSltlPVT6WG1NbWfddddt2KcKajUVj1pxHMdp5WTCRx73QyprSihK5aabbgJKZ3yWy9tvv93wt2LQNTN/0kknAaUVeTWRv/PMM89seE1KR+qyXJKRGYrkqKQirxfKEzjooIOAoMSlHBuLglK0hqKE7rzzTiCo7jiK169GZuyCUCE5jSqkhrUAQi2RXz4+l6WoKMXyDxw4sNan1SyksE899VQAVl555aLbqdjcQw89BMDPP/8MhLhzKfr11luv4TPlKvLGcEXuOI6TcTKhyOWThsI6IvJZym8lhd5S4vvRwgBS5IpoqAd62ldC1UiV7rjjjkDw9d13330t3ne9kZ9YKvCwww4DChceEckaI4rC+PLLL4EQA63RjxYdiKOojPhiE7VAIwGNyPT7gQceqOl5xClWElkLNKQN+fB1H2ixGuUIJOO/Ze9rrrkGCPdicoEJ3auiR48eDX+/+OKLFTt/cEXuOI6TeVKtyJNRIvHXhPyR8mNXCik6KIxBby3x1cnC9yI+P5Al4mpbmZlSV8svv/x82yqWWUv6SUEqa1f1QZLf/VprrQUEZR5Xa4rLrzXK1lRUxJtvvgkEn25LKhGWiyLMlMsAwYb1iq8vRrwf0YIQmvNSXLiQ/ZTNqu+5VB6JrvfCCy8EQhSYPg9w5ZVXzrdtS3FF7jiOk3FSrcilfoplz8mHfffddwOVr3sS92fJt6djqC5x1lE2o2pDNJYJmlakxKW+IcROJ5W4YuxHjBgBwPnnnw80vRrdFltsARRfYlB+9Xoh5ahRanI+qRaoTcUr/cl3HFfp9UJRTOpbINQYSkb5KLtbGdOqS9NUNHLXnEn8mGqz5VZBLIUrcsdxnIyTakWuOhfFaoRLPY4cObKix1Rm3xVXXNHwmp6eigbQ4qpZp2PHjkBQBapLUyqyI23Iz6kR0gknnNDwnhSzfOGqaCd/pdRVU0dymjNRLZZk5iekJ+7+rLPOAkK0VS1QRrVWUoqjyJ9kFEct0eLR8oPHV/tS9Inq9VcKzVGo/4rP45WqC99cXJE7juNknFQr8gVVF5Q/sKmrdpRCykqr5EiJx33kUnWnnXYaUH9faKXo168fEOLiteZkVq5v8803B+Doo48G5vdbyy+rSIEnn3wSKH8uRcrpyCOPBEIlTHHDDTc0/N3UNV2rjRRwsTrYlZ5LUmas1KxGefGa3rqX6lG/X9d+1113ASFDN66+Kx2FpmMq21zZwfHVpEqt49pcXJE7juNknFQrchF/kutpp9lx+XOVbdVUNIOsJ7SUlVRd/Ji33nor0DoyHuPoes477zwgjDyyUi9aUSryf8ZRxq9WiCpXDcqvKVWlDFG1P7W3IUOGlHnW1Uf3Ru/evYEQoQOVG22p1sw555wDBCUuO2uUBLWNY0+i2jp77703EL6/CRMmVO2YXbt2BUK1RNkk/j1UGlfkjuM4GSfVilxRFPHaFfL7SXVoVXStTJKsc6Hax4oF11PyjDPOAMITW8h3dcoppzS8pmPUw8dXTRQVJNISddEYyfrQIr4Ci9pDY6uyaF+qIKgsYmW7yg+fVOLK2qt1pcOmIAUs/3WfPn0a3hs1ahRQ/ghWKA5bI52NNtoICHkdist/9dVXm7X/aqGMU8156HuFylUi3H///YH5VxeDsKpXNedQXJE7juNknFSvECTfZ3wGPL66PQR/bvI6pMT0W77vZCUzfe6tt94CglKPrwJeLmlb4aQUqi+i0c3BBx8MBDVbSaphE9WEUWZenJkzZwJBDUlBbr311kCIPlHsuWLpNXeSbCeKgVZdc9XYbsn9U6kVghawPTB/zLJeK3cepFu3bgBcf/31QKhlk/T/ai6hJVEZlWwrunapbq3ROn78+IZtlFvwww8/AE1fW1MZ0VopSO1R6v+TTz4BwupULZkr8BWCHMdxWjnekTuO42ScVLtWxOmnn97wt9Kx42Vmy0HDpnHjxgEwePBgIKTdV2Jhiqy4VjRRpwUztFjCoEGDKn6sathEhf+1OHZ84lrtWi6TOXPmAGFSs5SrTSgZTWGpCkEtt3DSgqi2a6UlyC5bbrklEBZuUVuRfbSowo033gg03S2xIKrRVvr27QsUdxtqElKLPei3XCZyy6jPUblgLbeo0Eu5q8aMGQOEhLtKFKNz14rjOE4rJxOKPK6apMJUvGj77bcHQmKQtpWy1gTD8OHDgaBClS7b0hT/YmRFkb/77rtASKF+5513AOjZs2fFj1UNm2giSyVrFRIIQSVpm1LKW+1EYaoq+qQ0dy0iUKklBOOkUZHLTkqU04LCGslogrx79+5AKAtcSarRVtq1awfA6NGjgTDpDWFyUn2h+ggpcZXk1cLvShaTAtfE+oABAwB4/vnngcqMToQrcsdxnFZOJhT5gtDTVL9VTF8+vHqkm6ddkSfTlDt37gyEZc7kD64ktbBJvGiWkmDk41VY4dixY4Hgt9QCxSr5WkkV1RhpVOS6j5R6r3kTlTxQKN20adOqdg7VbCtS0/JzQ/Dva3Qv1HcoeWrixIlAaEOaX5sxYwYQwhergStyx3GcVk7mFXkaSbsi1+y7/J1Sskoxlh+xkqTdJvUgjYpcaI5B0Vxapk0qtJr9Rr3bSnI+Rf/XcrSWxBW54zhOK8cVeRWot6JojOQ8ghSHioPFS5BWirTbpB6kWZGrTWywwQZAKJNRqcWCF4S3lUJckTuO47RyXJGXgVRKYzbLiqLQEm+KwVcsdTXIik1qSZoVuVAUi/zDtegvvK0U4orccRynlVNTRe44juNUHlfkjuM4Gcc7csdxnIzjHbnjOE7G8Y7ccRwn43hH7jiOk3G8I3ccx8k43pE7juNkHO/IHcdxMo535I7jOBnHO3LHcZyM4x254zhOxvGO3HEcJ+N4R+44jpNxvCN3HMfJON6RO47jZBzvyB3HcTKOd+SO4zgZxztyx3GcjOMdueM4TsbxjtxxHCfjeEfuOI6TcbwjdxzHyTjekTuO42Sc/wdPjJEd1qsaDwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "if train_model:\n",
    "    # Training\n",
    "    data_x = x_train\n",
    "    len_data = x_train.shape[0]\n",
    "\n",
    "    for i in range(num_epoch):\n",
    "        # Shuffle the data \n",
    "        np.random.shuffle(data_x)\n",
    "\n",
    "        # Making mini-batch\n",
    "        for j in range(0, len_data, batch_size):\n",
    "            if j + batch_size < len_data:\n",
    "                data_x_in = data_x[j : j + batch_size, :]\n",
    "            else:\n",
    "                data_x_in = data_x[j : len_data, :]\n",
    "\n",
    "            data_x_in = data_x_in.reshape((-1, img_size, img_size, 1))\n",
    "\n",
    "            sampled_z = np.random.uniform(-1, 1, size=(data_x_in.shape[0] , n_latent))\n",
    "\n",
    "            # Run Optimizer!\n",
    "            G_out = sess.run(G, feed_dict = {x: data_x_in, z: sampled_z, is_training: True})\n",
    "\n",
    "            ########################################### WGAN ###########################################\n",
    "            for iter_d in range(n_critic):\n",
    "                _, loss_d = sess.run([train_step_d, d_loss], feed_dict = {x: data_x_in, z: sampled_z, is_training: True})\n",
    "            ############################################################################################\n",
    "\n",
    "            _, loss_g = sess.run([train_step_g, g_loss], feed_dict = {x: data_x_in, z: sampled_z, is_training: True})\n",
    "\n",
    "            print(\"Batch: {} / {}\".format(j, len_data), end=\"\\r\")\n",
    "\n",
    "        # Print Progess\n",
    "        print(\"Epoch: {} / G Loss: {:.5f} / D Loss: {:.5f}\".format((i+1), loss_g, loss_d))\n",
    "\n",
    "        # Show test images \n",
    "        z_test = np.random.uniform(-1, 1, size=(5, n_latent))\n",
    "        G_out = sess.run(G, feed_dict = {z: z_test, is_training: False})\n",
    "\n",
    "        if i == 0 or (i+1) % show_result_epoch == 0:\n",
    "            f, ax = plt.subplots(1,5)\n",
    "            for j in range(5):\n",
    "                ax[j].imshow(G_out[j,:,:,0], cmap = 'gray')\n",
    "                ax[j].axis('off')\n",
    "                ax[j].set_title('Image '+str(j))\n",
    "\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Image Generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "num_test = 10\n",
    "\n",
    "img = np.zeros([img_size * num_test, img_size * num_test])\n",
    "\n",
    "z_result = np.random.uniform(-1, 1, size=(num_test**2, n_latent))\n",
    "G_result = sess.run(G, feed_dict = {z: z_result, is_training: False})\n",
    "\n",
    "for i in range(num_test**2):\n",
    "    row_num = int(i/num_test)\n",
    "    col_num = int(i%num_test)\n",
    "    \n",
    "    img[row_num * img_size : (row_num + 1) * img_size, (col_num) * img_size : (col_num + 1) * img_size] = G_result[i,:,:,0]\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.imshow(img, cmap='gray')\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "os.mkdir(save_path)\n",
    "\n",
    "Saver.save(sess, save_path + \"/model/model\")\n",
    "print(\"Model is saved in {}\".format(save_path + \"/model/model\"))"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}