{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Shin\\Anaconda3\\envs\\tensorflow_windows\\lib\\site-packages\\h5py\\__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "from tensorflow.examples.tutorials.mnist import input_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setting up the basics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting ./inputs/mnist\\train-images-idx3-ubyte.gz\n",
      "Extracting ./inputs/mnist\\train-labels-idx1-ubyte.gz\n",
      "Extracting ./inputs/mnist\\t10k-images-idx3-ubyte.gz\n",
      "Extracting ./inputs/mnist\\t10k-labels-idx1-ubyte.gz\n"
     ]
    }
   ],
   "source": [
    "mnist = input_data.read_data_sets('./inputs/mnist')\n",
    "# Resetting default graph, starting from scratch\n",
    "tf.reset_default_graph()\n",
    "\n",
    "epochs = 6000\n",
    "batch_size = 64\n",
    "n_noise = 200\n",
    "learning_rate=0.00015\n",
    "\n",
    "real_images = tf.placeholder(dtype=tf.float32, shape=[None, 28, 28, 1], name='real_images')\n",
    "noise = tf.placeholder(dtype=tf.float32, shape=[None, n_noise])\n",
    "\n",
    "# The keep_prob variable will be used by our dropout layers, which we introduce for more stable learning outcome\n",
    "keep_prob = tf.placeholder(dtype=tf.float32, name='keep_prob')\n",
    "is_training = tf.placeholder(dtype=tf.bool, name='is_training')\n",
    "\n",
    "# Leaky Relu activation\n",
    "# https://en.wikipedia.org/wiki/Rectifier_%28neural_networks%29#Potential_problems\n",
    "def lrelu(x):\n",
    "    return tf.maximum(x, tf.multiply(x, 0.2))\n",
    "\n",
    "# Binary cross entropy for descriminators\n",
    "def binary_cross_entropy(x, z):\n",
    "    eps = 1e-12\n",
    "    return (-(x * tf.log(z + eps) + (1. - x) * tf.log(1. - z + eps)))\n",
    "\n",
    "# Code by Parag Mital (github.com/pkmital/CADL)\n",
    "def montage(images):\n",
    "    if isinstance(images, list):\n",
    "        images = np.array(images)\n",
    "    img_h = images.shape[1]\n",
    "    img_w = images.shape[2]\n",
    "    n_plots = int(np.ceil(np.sqrt(images.shape[0])))\n",
    "    m = np.ones((images.shape[1] * n_plots + n_plots + 1, images.shape[2] * n_plots + n_plots + 1)) * 0.5\n",
    "    for i in range(n_plots):\n",
    "        for j in range(n_plots):\n",
    "            this_filter = i * n_plots + j\n",
    "            if this_filter < images.shape[0]:\n",
    "                this_img = images[this_filter]\n",
    "                m[1 + i + i * img_h:1 + i + (i + 1) * img_h,\n",
    "                  1 + j + j * img_w:1 + j + (j + 1) * img_w] = this_img\n",
    "    return m"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The descriminator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "ename": "IndentationError",
     "evalue": "expected an indented block (<ipython-input-7-03e088156e11>, line 51)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;36m  File \u001b[1;32m\"<ipython-input-7-03e088156e11>\"\u001b[1;36m, line \u001b[1;32m51\u001b[0m\n\u001b[1;33m    def discriminator(real_images, reuse=None, keep_prob=keep_prob):\u001b[0m\n\u001b[1;37m                                                                    ^\u001b[0m\n\u001b[1;31mIndentationError\u001b[0m\u001b[1;31m:\u001b[0m expected an indented block\n"
     ]
    }
   ],
   "source": [
    "# It takes either real or fake MNIST image 28 x 28 in grayscale\n",
    "# we use a sigmoid to make sure our output can be interpreted \n",
    "# as the probability the input image is a real MNIST character.\n",
    "def discriminator(real_images, reuse=None, keep_prob=keep_prob):\n",
    "    activation=lrelu\n",
    "    with tf.variable_scope('disc', reuse=reuse):\n",
    "        x = tf.reshape(real_images, shape=[-1, 28, 28, 1])\n",
    "        x = tf.layers.conv2d(x, kernel_size=5, filters=64, strides=2, padding='same', activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        x = tf.layers.conv2d(x, kernel_size=5, filters=64, strides=1, padding='same', activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        x = tf.layers.conv2d(x, kernel_size=5, filters=64, strides=1, padding='same', activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        x = tf.contrib.layers.flatten(x)\n",
    "        x = tf.layers.dense(x, units=128, activation=activation)\n",
    "        x = tf.layers.dense(x, units=1, activation=tf.nn.sigmoid)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def generator(z, keep_prob=keep_prob, is_training=is_training):\n",
    "    activation = lrelu\n",
    "    momentum = 0.99\n",
    "    with tf.variable_scope('gen', reuse=None):\n",
    "        x = z\n",
    "        d1 = 4\n",
    "        d2 = 1\n",
    "        x = tf.layers.dense(x, units=d1 * d1 * d2, activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        # https://www.tensorflow.org/api_docs/python/tf/contrib/layers/batch_norm\n",
    "        x = tf.contrib.layers.batch_norm(x, is_training=is_training, decay=momentum)\n",
    "        x = tf.reshape(x, shape=[-1, d1, d1, d2])\n",
    "        x = tf.image.resize_images(x, size=[7, 7])\n",
    "        x = tf.layers.conv2d_transpose(x, kernel_size=5, filters=64, strides=2, padding='same', activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        x = tf.contrib.layers.batch_norm(x, is_training=is_training, decay=momentum)\n",
    "        x = tf.layers.conv2d_transpose(x, kernel_size=5, filters=64, strides=2, padding='same', activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        x = tf.contrib.layers.batch_norm(x, is_training=is_training, decay=momentum)\n",
    "        x = tf.layers.conv2d_transpose(x, kernel_size=5, filters=64, strides=1, padding='same', activation=activation)\n",
    "        x = tf.layers.dropout(x, keep_prob)\n",
    "        x = tf.contrib.layers.batch_norm(x, is_training=is_training, decay=momentum)\n",
    "        x = tf.layers.conv2d_transpose(x, kernel_size=5, filters=1, strides=1, padding='same', activation=tf.nn.sigmoid)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loss functions and optimizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "checking units  16\n"
     ]
    }
   ],
   "source": [
    "g = generator(noise, keep_prob, is_training)\n",
    "d_real = discriminator(real_images)\n",
    "d_fake = discriminator(g, reuse=True)\n",
    "\n",
    "vars_g = [var for var in tf.trainable_variables() if 'gen' in var.name]\n",
    "vars_d = [var for var in tf.trainable_variables() if 'disc' in var.name]\n",
    "\n",
    "# Applying regularizers\n",
    "d_reg = tf.contrib.layers.apply_regularization(tf.contrib.layers.l2_regularizer(1e-6), vars_d)\n",
    "g_reg = tf.contrib.layers.apply_regularization(tf.contrib.layers.l2_regularizer(1e-6), vars_g)\n",
    "\n",
    "loss_d_real = binary_cross_entropy(tf.ones_like(d_real), d_real)\n",
    "loss_d_fake = binary_cross_entropy(tf.zeros_like(d_fake), d_fake)\n",
    "loss_g = tf.reduce_mean(binary_cross_entropy(tf.ones_like(d_fake), d_fake))\n",
    "loss_d = tf.reduce_mean(0.5 * (loss_d_real + loss_d_fake))\n",
    "update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)\n",
    "\n",
    "optimizer_d = tf.train.AdamOptimizer(learning_rate).minimize(loss_d + d_reg, var_list=vars_d)\n",
    "# optimizer_d = tf.train.RMSPropOptimizer(learning_rate=0.00015).minimize(loss_d + d_reg, var_list=vars_d)\n",
    "# optimizer_g = tf.train.RMSPropOptimizer(learning_rate=0.00015).minimize(loss_g + g_reg, var_list=vars_g)\n",
    "optimizer_g = tf.train.AdamOptimizer(learning_rate).minimize(loss_g + g_reg, var_list=vars_g)\n",
    "\n",
    "\n",
    "sess = tf.Session()\n",
    "sess.run(tf.global_variables_initializer())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Training GAN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(64, 28, 28, 1)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x24b0301f518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n",
      "(64, 28, 28, 1)\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-22-de32e5cdffbe>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[0;32m     31\u001b[0m         \u001b[0msess\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0moptimizer_d\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m{\u001b[0m\u001b[0mnoise\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreal_images\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mbatch_x\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkeep_prob\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mkeep_prob_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mis_training\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     32\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mtrain_g\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 33\u001b[1;33m         \u001b[0msess\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0moptimizer_g\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m{\u001b[0m\u001b[0mnoise\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mkeep_prob\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mkeep_prob_train\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mis_training\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m}\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     34\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     35\u001b[0m     \u001b[1;31m# Showing sample image\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\tensorflow_windows\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36mrun\u001b[1;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m    776\u001b[0m     \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    777\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[1;32m--> 778\u001b[1;33m                          run_metadata_ptr)\n\u001b[0m\u001b[0;32m    779\u001b[0m       \u001b[1;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    780\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\tensorflow_windows\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_run\u001b[1;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m    980\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    981\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[1;32m--> 982\u001b[1;33m                              feed_dict_string, options, run_metadata)\n\u001b[0m\u001b[0;32m    983\u001b[0m     \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    984\u001b[0m       \u001b[0mresults\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\tensorflow_windows\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_do_run\u001b[1;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m   1030\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1031\u001b[0m       return self._do_call(_run_fn, self._session, feed_dict, fetch_list,\n\u001b[1;32m-> 1032\u001b[1;33m                            target_list, options, run_metadata)\n\u001b[0m\u001b[0;32m   1033\u001b[0m     \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1034\u001b[0m       return self._do_call(_prun_fn, self._session, handle, feed_dict,\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\tensorflow_windows\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_do_call\u001b[1;34m(self, fn, *args)\u001b[0m\n\u001b[0;32m   1037\u001b[0m   \u001b[1;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1038\u001b[0m     \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1039\u001b[1;33m       \u001b[1;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1040\u001b[0m     \u001b[1;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1041\u001b[0m       \u001b[0mmessage\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\tensorflow_windows\\lib\\site-packages\\tensorflow\\python\\client\\session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[1;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[0;32m   1019\u001b[0m         return tf_session.TF_Run(session, options,\n\u001b[0;32m   1020\u001b[0m                                  \u001b[0mfeed_dict\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1021\u001b[1;33m                                  status, run_metadata)\n\u001b[0m\u001b[0;32m   1022\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1023\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msession\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "samples = []\n",
    "for i in range(epochs):\n",
    "    train_d = True\n",
    "    train_g = True\n",
    "    keep_prob_train = 0.6\n",
    "\n",
    "    # Creating noise\n",
    "    n = np.random.uniform(0.0, 1.0, [batch_size, n_noise]).astype(np.float32)\n",
    "    # batch = [np.reshape(b, [28, 28]) for b in mnist.train.next_batch(batch_size=batch_size)[0]]\n",
    "    batch_x, _ = mnist.train.next_batch(batch_size)\n",
    "    batch_x = np.reshape(batch_x, newshape=[-1, 28, 28, 1])\n",
    "    d_real_ls, d_fake_ls, g_ls, d_ls = sess.run(\n",
    "        [loss_d_real, loss_d_fake, loss_g, loss_d],\n",
    "        feed_dict={real_images: batch_x, noise: n, keep_prob: keep_prob_train, is_training: True}\n",
    "    )\n",
    "    d_real_ls = np.mean(d_real_ls)\n",
    "    d_fake_ls = np.mean(d_fake_ls)\n",
    "\n",
    "    g_ls = g_ls\n",
    "    d_ls = d_ls\n",
    "\n",
    "    if g_ls * 1.5 < d_ls:\n",
    "        train_g = False\n",
    "        pass\n",
    "\n",
    "    if d_ls * 2 < g_ls:\n",
    "        train_d = False\n",
    "        pass\n",
    "    if train_d:\n",
    "        sess.run(optimizer_d, feed_dict={noise: n, real_images: batch_x, keep_prob: keep_prob_train, is_training:True})\n",
    "    if train_g:\n",
    "        sess.run(optimizer_g, feed_dict={noise: n, keep_prob: keep_prob_train, is_training:True})\n",
    "        \n",
    "    # Showing sample image\n",
    "    if not i % 50:\n",
    "        gen_sample = sess.run(g, feed_dict={noise: n, keep_prob: 1.0, is_training:False})\n",
    "        imgs = [img[:,:,0] for img in gen_sample]\n",
    "        m = montage(imgs)\n",
    "        gen_sample = m\n",
    "        plt.axis('off')\n",
    "        plt.imshow(gen_sample, cmap='gray')\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}