01_VAE_MLP_MNIST.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# VAE MLP MNIST\n",
    "\n",
    "This notebook is for implementing `Variational Auto-Encoder` with [Tensorflow](https://www.tensorflow.org). <br>\n",
    "Network architecture will be `Multi-layer perceptron` and [MNIST data](http://yann.lecun.com/exdb/mnist/) will be used. \n",
    "\n",
    "Reference: [hwalsuklee's Github](https://github.com/hwalsuklee/tensorflow-generative-model-collections)"
   ]
  },
  {
   "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 = 'VAE_MNIST'\n",
    "\n",
    "img_size   = 28\n",
    "data_size  = img_size**2\n",
    "\n",
    "num_label  = 10\n",
    "\n",
    "batch_size = 512\n",
    "num_epoch  = 20\n",
    "\n",
    "n_hidden = 256\n",
    "n_latent = 128\n",
    "\n",
    "learning_rate = 1e-3\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_VAE_MNIST/model/model\" "
   ]
  },
  {
   "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, 784)\n",
      "y_train shape: (60000,)\n",
      "x_test shape: (10000, 784)\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",
    "x_train = np.reshape(x_train, [-1, data_size])\n",
    "x_test  = np.reshape(x_test, [-1, data_size])\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": [
    "## Encoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def MLP_encoder(x, keep_prob):\n",
    "    with tf.variable_scope('MLP_encoder'):\n",
    "        xavier_init = tf.contrib.layers.xavier_initializer()\n",
    "        \n",
    "        # First hidden layer\n",
    "        w1 = tf.get_variable('w1', [x.get_shape()[1], n_hidden], initializer=xavier_init)\n",
    "        b1 = tf.get_variable('b1', [n_hidden], initializer=xavier_init)\n",
    "        \n",
    "        h1 = tf.nn.relu(tf.matmul(x,w1)+b1)\n",
    "        h1 = tf.nn.dropout(h1, keep_prob)\n",
    "        \n",
    "        # Second hidden layer\n",
    "        w2 = tf.get_variable('w2', [h1.get_shape()[1], n_hidden], initializer=xavier_init)      \n",
    "        b2 = tf.get_variable('b2', [n_hidden], initializer=xavier_init)\n",
    "        \n",
    "        h2 = tf.nn.relu(tf.matmul(h1,w2)+b2)\n",
    "        h2 = tf.nn.dropout(h2, keep_prob)\n",
    "        \n",
    "        # Output layer\n",
    "        w_mean = tf.get_variable('w_mean', [h2.get_shape()[1], n_latent], initializer=xavier_init)      \n",
    "        b_mean = tf.get_variable('b_mean', [n_latent], initializer=xavier_init)      \n",
    "        \n",
    "        w_std = tf.get_variable('w_std', [h2.get_shape()[1], n_latent], initializer=xavier_init)      \n",
    "        b_std = tf.get_variable('b_std', [n_latent], initializer=xavier_init)    \n",
    "        \n",
    "        mean = tf.matmul(h2, w_mean)+b_mean\n",
    "        std  = tf.nn.relu(tf.matmul(h2, w_std)+b_std)\n",
    "        \n",
    "        return mean, std"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def MLP_decoder(z, keep_prob):\n",
    "    with tf.variable_scope('MLP_decoder'):\n",
    "        xavier_init = tf.contrib.layers.xavier_initializer()\n",
    "\n",
    "        # First hidden layer\n",
    "        w1 = tf.get_variable('w1', [z.get_shape()[1], n_hidden], initializer=xavier_init)\n",
    "        b1 = tf.get_variable('b1', [n_hidden], initializer=xavier_init)\n",
    "\n",
    "        h1 = tf.nn.relu(tf.matmul(z,w1)+b1)\n",
    "        h1 = tf.nn.dropout(h1, keep_prob)\n",
    "\n",
    "        # Second hidden layer\n",
    "        w2 = tf.get_variable('w2', [h1.get_shape()[1], n_hidden], initializer=xavier_init)      \n",
    "        b2 = tf.get_variable('b2', [n_hidden], initializer=xavier_init)\n",
    "\n",
    "        h2 = tf.nn.relu(tf.matmul(h1,w2)+b2)\n",
    "        h2 = tf.nn.dropout(h2, keep_prob)\n",
    "\n",
    "        # Output layer\n",
    "        wo = tf.get_variable('wo', [h2.get_shape()[1], data_size], initializer=xavier_init)      \n",
    "        bo = tf.get_variable('bo', [data_size], initializer=xavier_init)      \n",
    "\n",
    "        output = tf.sigmoid(tf.matmul(h2, wo)+bo)\n",
    "\n",
    "        return output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## VAE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def VAE(x, keep_prob_encoder, keep_prob_decoder):\n",
    "    # Encoder\n",
    "    mean, std = MLP_encoder(x, keep_prob_encoder)\n",
    "    \n",
    "    # Sampling z\n",
    "    z = mean + std * tf.random_normal(tf.shape(mean), 0, 1, dtype=tf.float32)\n",
    "    \n",
    "    # Decoder\n",
    "    y = MLP_decoder(z, keep_prob_decoder)\n",
    "    \n",
    "    # loss\n",
    "    cross_entropy = - tf.reduce_sum(x * tf.log(1e-8 + y) + (1 - x) * tf.log(1e-8 + 1 - y), 1)\n",
    "    KL_divergence = 0.5 * tf.reduce_sum(tf.square(mean) + tf.square(std) - tf.log(1e-8 + tf.square(std)) - 1, 1)\n",
    "    \n",
    "    loss = tf.reduce_mean(cross_entropy) + tf.reduce_mean(KL_divergence)\n",
    "     \n",
    "    return loss, mean, std, z, y    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build Graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = tf.placeholder(tf.float32, shape=[None, data_size])\n",
    "x_normalize = tf.cast(x, tf.float32) / 255.0\n",
    "\n",
    "keep_prob_encoder = tf.placeholder(tf.float32)\n",
    "keep_prob_decoder = tf.placeholder(tf.float32)\n",
    "\n",
    "loss, mean, std, z, y = VAE(x_normalize, keep_prob_encoder, keep_prob_decoder) \n",
    "\n",
    "# optimization\n",
    "train_step = tf.train.AdamOptimizer(learning_rate).minimize(loss)"
   ]
  },
  {
   "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": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 1 / Loss: 207.12474\n",
      "Epoch: 2 / Loss: 186.90643\n",
      "Epoch: 3 / Loss: 170.883\n",
      "Epoch: 4 / Loss: 156.4654\n",
      "Epoch: 5 / Loss: 143.96161\n",
      "Epoch: 6 / Loss: 139.06602\n",
      "Epoch: 7 / Loss: 135.62518\n",
      "Epoch: 8 / Loss: 130.91068\n",
      "Epoch: 9 / Loss: 134.32993\n",
      "Epoch: 10 / Loss: 132.6017\n",
      "Epoch: 11 / Loss: 123.47276\n",
      "Epoch: 12 / Loss: 133.58064\n",
      "Epoch: 13 / Loss: 125.14267\n",
      "Epoch: 14 / Loss: 125.77563\n",
      "Epoch: 15 / Loss: 123.75738\n",
      "Epoch: 16 / Loss: 130.77655\n",
      "Epoch: 17 / Loss: 128.16959\n",
      "Epoch: 18 / Loss: 126.81251\n",
      "Epoch: 19 / Loss: 128.17857\n",
      "Epoch: 20 / Loss: 117.04929\n"
     ]
    }
   ],
   "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",
    "            # Run Optimizer!\n",
    "            _, loss_train = sess.run([train_step, loss], feed_dict = {x: data_x_in, keep_prob_encoder: 0.9, keep_prob_decoder: 0.9})\n",
    "\n",
    "            print(\"Batch: {} / {}\".format(j, len_data), end=\"\\r\")\n",
    "\n",
    "        print(\"Epoch: \" + str(i+1) + ' / ' + \"Loss: \" + str(loss_train))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test (Reconstruction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "num_test = 10\n",
    "\n",
    "for i in range(num_test):\n",
    "    idx = random.randint(0,x_test.shape[0])\n",
    "    \n",
    "    output, mu, sigma = sess.run([y, mean, std], feed_dict = {x: [x_test[idx,:]], keep_prob_encoder: 1, keep_prob_decoder: 1})\n",
    "\n",
    "    input_reshape  = np.reshape(x_test[idx,:], [img_size, img_size])\n",
    "    output_reshape = np.reshape(output, [img_size, img_size])\n",
    "\n",
    "    f, ax = plt.subplots(1,2)\n",
    "    ax[0].imshow(input_reshape, cmap = 'gray')\n",
    "    ax[0].axis('off')\n",
    "    ax[0].set_title('Original Image')\n",
    "\n",
    "    ax[1].imshow(output_reshape, cmap = 'gray')\n",
    "    ax[1].axis('off')\n",
    "    ax[1].set_title('Reconstructed Image')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Print Average mu & sigma"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average mu: 0.04700107127428055\n",
      "Average std: 0.9675211906433105\n"
     ]
    }
   ],
   "source": [
    "print(\"Average mu: {}\".format(np.mean(mu)))\n",
    "print(\"Average std: {}\".format(np.mean(sigma)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Changing Latent Space"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABH4AAAB3CAYAAAB19YfCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvDW2N/gAAIABJREFUeJztncmzXWXVh98oIhia0AhEIH1LSAcEAiQSIiCtdA4YqeWAkU70v3FAWaVlFWVbdhAkIZSRLgnpEwIJpCGRTsFe7PINvsry2duzDjvJvbl77/s8o3VPTrPPu97u7Px+75pw7NixIiIiIiIiIiIi/eNjY30BIiIiIiIiIiIyOnjjR0RERERERESkp3jjR0RERERERESkp3jjR0RERERERESkp3jjR0RERERERESkp3jjR0RERERERESkp3jjR0RERERERESkp3jjR0RERERERESkp5xxOj9swoQJx07n540Xjh07NuFU38PcjA7mpr2Ym/ZibtqLuWkv5qa9mJv2Ym7ai7lpL+amvQzLjYofEREREREREZGe4o0fEREREREREZGe4o0fEREREREREZGe4o0fEREREREREZGe4o0fEREREREREZGe4o0fEREREREREZGe4o0fEREREREREZGe4o0fEREREREREZGecsZYX4CIiIiIjA8mTJhwQo834dixYyf0uAyGOfjYx/77f8NnnDH45wKfw9f+5z//ifjf//53xMwHn8PHzdlg2L6MP/GJTwx8PnPG57Dd//Wvf0XMPPHxLE8ymGwMkY9//OMRn3vuuQOfw3z84x//GBgzT3JiNFmHzjnnnIiZM46Df/7znxF/+OGHEbc1Nyp+RERERERERER6ijd+RERERERERER6yrixejWRdJ3Kc0gmhaRccthrxoOUsklbZxLJ7PlZ3KRt6483kSN3jab9uMnzKHkkbJ8mOc5yMyxnfclN1tfr7ZY9L3sN5d0nKqvPnn8ykvyuzWlZvz/zzDMjrtsdOA4Yn3XWWQNfw5yxTf/+978PfJySbsq+KS1ukqc6XcgHYT7YzpRhf/KTn6y8ZuLEiRF/6lOfivj888+P+Oyzz46YuWE+fv/730f817/+NeI//vGPA59PqXdmcela+w8jyw3bmbkopZRLLrkk4ssuuyziK664IuILLrhg4Pv++c9/jvjNN98cGP/2t7+N+P3334+Y+cvGTV/hPMR8cAyVUsr06dMjnj17dsTTpk2L+Morrxz4Xn/7298iPnjwYMTbt2+PeNeuXREfOXIk4iw3fRorGZmljutIKdVx85nPfCbiSZMmRXzVVVdFzFzyvX73u99FvHXr1og3bdoU8YEDByJmXscz2VxXStVKx7WI+ZwyZUrEN910U8Rz586NmHnauXNnxOvXr4947969EbfVTtQmhu2vj8M9wqJFiyK+9dZbI54zZ07Eb7zxRsRPPPFExFu2bIm4rblR8SMiIiIiIiIi0lO88SMiIiIiIiIi0lM6afWidI4y/FKq8mKelH7hhRdGTPkj5ZKZNJnvw5hSv7feeitiSlwpP963b1/lWvmaP/3pTxFT8tpWqVhmi6DEsS7vvvjiiyOmNPXTn/50xJRCXnrppRFT/kjZHj+Psn32kffeey9iSouznJVSytGjRyOmpJ8yc9owxppMysjHs/FQSnUcXHTRRRFThj958uSIKcMntDlk10EZ99tvvx0xc8PHmadSqvnkWKEcmVaYsYbfPbMGcQwwF6WUMmPGjIiZDz6P+eS4Yx/NLCuce/iczDpBmThfW0rV8sK5K6sQMtYwB9m6MnXq1IhpfSilKqvn3MXcUELMz+C8wj7O9uWY4Ppx6NChiDkeOD/Vx0AT21GbbBVZbjjPz5w5M+KFCxdWXr906dKIZ82aFfHll18+8DPIBx98EDHHAfNBy8q2bdsGPp/jg2OxPgba1O5NyCynXKe5fi9evLjy+uXLl0dMWT3nuuwz/vCHP0R8+PDhiGlN2bBhQ8TPP/98xJzHOFdl9uMuku3NOIa4F1iwYEHl9StWrIiYeeNcx/mR6w3Xf7Y188q1gHPXeLYTZRW66vssWuy4LnGuoyWPFiLmiWsP93VcVzie5MTg/M7flVx7li1bFvE111wTMX/TMN+0EHEtl48m2+Nw/ee443xFqxfH3Pz58yN+4YUXIm7rb3ai4kdEREREREREpKd440dEREREREREpKe02uqVWUUo9ab0qpSq/GrJkiUR8wR1Shv5XpSvUgLGxykHo3SbUv3du3dHTAnYX/7yl8q1UrJMmWtbpcaZnYhtQstJXUJMqTBl+ZQ50mbEHPCzKaXjcyiRpM2BudmzZ0/EL730UsT13NDy0oXcNKkuxIopbPNSqrmhnYWPn3feeRFTykrZaVadiGOIbU2pPqtLcGzRllRKbp8gbZLuZ1WhaAHiPMa5qpSqXJvSX0q6OY9lFivagJgntiEtjrSysD2Zj7qdKLN0NamGOBZkc3smz77hhhsqr6fVKKtUlFWL4mfQBsv2ZY5ZfYdty/kpG3+ldK8aHvPBOY02Fc5JtHCXUrVP8r04X9GawrbjOOXY5HtyzHFscV1nLk+0Omgp7c1Ndv0cT+z3dds3vxfXZ84fnFuYM+ac44bjg1YWVl+hjZK57yts52wPVW8H5oPzBK1bzD/zQbsrbUq0TnANq1etGnTd4wF+32zNLqWaA+6DOOdw7WEuaa/kPj37PZRd33iG7TDMbsUcvvvuuxFnvz/4u5VzGtc6jtM22eW7QNZ/mUOu1a+99trA53CuYz5YObILuVHxIyIiIiIiIiLSU7zxIyIiIiIiIiLSUzpj9cpk+FdffXXlNatXr46YsnxKtCkhzqrQZFaTzHJG2SUrgtAyxAoGpVQtL5SNdUFWmVmsKO+lfLGUqtSU1SL4GtoWKF8lzAdjXgclkszNO++8EzGrrzCufzbz1FYZH8dKlhtaiyjvLaVqKaEsm9J9th37OMksRJR0M2bbspIXLWCUypZSlT+zv7TVvpJdC+0kWeWvUqpzH/NB2wKlprQ58LXsI1k/5hxIiSvHEGPmu5TcntGmfBC2Q2ZfzPpoKdV8sC1YHYV54ntRNswqh1xX2J60smSWYz5/mNWrrfkgWf9hG7Jtd+7cWXk9bQ5sX8J5jJ9Bex8rgnGt4nuyghVzczJzUhdyk/WlLDcvv/xy5fWsGMT9HOdB7om4js2bNy9iWpZZSY/rGR9/5ZVXPvI7dJ3su9Tn6uPQflJKdS/EPUNWWZX2/Ouvvz5i7r8519HqxTmtrXbg0wHnCc7h9aqZ3O9w/8D9Eve+HGfsFxw3/GzmiTmW/6U+zjLrF3+jcE36zW9+EzGrsjF/XagQ1WWYQ+6jWOWZa8acOXMizo646AKObBERERERERGRnuKNHxERERERERGRnuKNHxERERERERGRntK6M34yny998yxXSL9wKdXSq/RH7t27N+J9+/ZFTD8zn0+vOv179JrTA8vHWXKRZeHo7ywlPyujC+fI0BvJ3PAsJZ6HUEq1hGRWbpXtleWGXszMn0z/M59DPzu9mzyLqZTqOTI8a2BYCcexhLnhNbKP0k9fh/nk+STss6+//nrEPHeHn5GVx6ZvmR727BwIXkO9nDvzz37RhXMasjN02Ed5Xk8p1RzwnCXOMzx7hucAMOcsE5p5+flZnCf3798/8PrqHvTsO7U1N7wu9ku2IedptkMp1XmM35354FpCWP6d56HxTBLOPZwbeXYDz6nJ5sZSunemSdZ/mBuem1f/vuzjhDnj+/IME85vfB+OJ+bmyJEjEXN8ZPnoQvsPg9+Faw/nA56TV183Oe/z9RyDXMPZ7vwMjpusNDhzybWkrfus0YLfd9iZkhxfPG+Oz+N+l/MP1/lly5ZFzPxxveHc2Na91emGeaqvr9lZQPz9wJj7Mf4G4llMHDfZeXHy0TQpGc5xx7WLZ8rceuutEXNu7MJZo12G+ePvv6NHj0bMeY9jk2t+F9Z2FT8iIiIiIiIiIj3FGz8iIiIiIiIiIj2ldVYvkllIKD+ty7mz8tObNm2KmGVfWQYxs25R0pVZNQillpSMMS6lKqXsgjyMZOWn+Z3qZanZXrRxMTeZNS6zDWXyeUpWadWg7JtS9GF2oi7IKnmNmTyb/a9efpD2B1qIdu3aFTElj2w7yrhZ1pgyVeaSEld+FmPmoy45znLe1jHEfs/c8HHK5WmDKKVa3pt9ObPDsR1ovaSFiPMYc7N79+6IacGkFXJYyXDS1nyQzBbJx/ndmYtSqvMEc8DX8H0pq6eVZfLkyRFPnDgxYtord+zYETFzT2vGeLATMTeUv9fXm8w+zZwxH2x32pRnz54dMee6jRs3Rsx5krLvrCR9n8jmYM4x7KOl5DYSvp6lxLn+cx5bsGBBxMwfbdy0uLBP9DUfTcjyVP83wjHIuYt75WnTpkXM3HDMMR9c67qwzzrd1HORtRHXmMy+OmvWrIhnzpwZMfPBPZ5Wr5GB+WD+mDOOG85v3JtxTzGe567RImtTHh9z0UUXRcz7C/w91AVU/IiIiIiIiIiI9BRv/IiIiIiIiIiI9JRWW70IZViUT9el95QUU3JNiw+fQ6vYFVdcETGlrJRxUd5FmSptNFmlobqktmtQpsgc8HsNs+jQzkKJPl9Duf7ll18eMeV2fA7lj1luKJdknrLqUKV0T0qZWQoobWf7My6l2l6Z/YUyR+aGsmFaLdnutKbws1j9KOs7fa1OlFmDaIUrpTrHMR+0NjA3tHddffXVEU+aNCnizMaVVSfK8tGF9h9G9l04bmhxoK2hDtcMVjCk/XHKlCkRs7IKq1PSppJVOcyqE3U9HyT7LpmFqF4RlO1Cq9D5558f8fTp0yOeN29exKysMnfu3Ii3bNkS8XPPPRcxbapdtnCfKllFoqa54VxH+8PixYsjvv/++yPmvuDFF1+M+Be/+EXE2on+l6btkB23wMq6nMfuueeeiGlZefLJJyNev359xHXLuQyH8wnHFGPObwsXLoz45ptvjpi/e2jv5nrjWBl5smNK5syZEzHtecxHvfKwjCwcQ8wT1xiuVazEVj/Gpe2o+BERERERERER6Sne+BERERERERER6Smttnpl9i7aGmhfqEOZKqX0F198ccSUrFKGT9k+5eSsAsaKKzwZn7JvVuGpy1o//PDDgddNyVlbpeK8LkrbWfWEVTlKqcrnaFNZtGhRxKwEQTsR88TPZjUqWiFYAYdVw2hx4WvrFWG6JnPN7E+0s1GqXbes8G9WTaMVgs/hWOF44udxHDAfWf44HrpuiyRNLCtZxZRSqmOFtiFK6S+55JKIab3jXMl5jJJV2vA4frNKXn0iy03Taky0plASPH/+/IhpFWJMSx5zs3///oibrBHjgSwHw3LDPk4b99KlSyO+8cYbI6adiGsP123Kuzm/9Wm+Gg3qeeLejNYUjo877rgj4lWrVg18DtdpWu8OHDgQsbk5ebjmc725/fbbI37ooYcipu2L7f6rX/0qYtq727q/bQv19uG8zzHEdYjVu+69996IWZ2Q682aNWsi7pplpQtkFiLmg/tp7rWeeeaZiLm3lpGHYy37DcTnMDdZhe+2ouJHRERERERERKSneONHRERERERERKSntM7qlVlWKNumvatuWaG8lNYWyoN5ajotRHyccjt+Nk/J52ft2bMnYtq+du7cGfH27dsr18rP4Ht1QZrM3NCaQnvXMKvXBRdcMDCmrYUWIn5GZn+gVP9zn/tcxJs3b454w4YNEVMa/uqrr1bei5XculA1h9fF62Wbsx+zokD9b1qFKEGlJJ9ViwilwswrqxasXLky4rVr10b81FNPRbxt27aI69UMsqp5bSW7xkwCzHYrpTpH0QrJOY0yfOaGn83qNrSy0FLBz1q3bl3ErPzBfHS9Gh5pcu11azErD1133XURL1u2LOIZM2ZETOsd5cGs0sV1a8mSJRFzveC6klnyupyLk6G+3tDexdysXr164OOTJ0+OmLmhNYXVKGmvZA44zsZzPkjdvsq1hOODa8Ndd90VMa2TnDe516LVm/tCzq1dk+SfDuq2Uc4/XIu49tx9990Rc67jazlHsTrReLOpngr1tqK9i2OK+zfa7a655pqB78V88PgDGXmyqqDMDfPKtYTzm4wuzBPXJP5u4m/P7PiKLqDiR0RERERERESkp3jjR0RERERERESkp7TO6pVBCxBPsK/bfmg1oSybUjrKsvbt2xcxJd20+lCeP2/evIhpiWFMqwXl47QulVLK008/PfA62mr1YrtRFkd5L2VxrBxQStXOkMl933///YjZJm+99VbEzDGrHFGKzCpuK1asiJjWJT7nySefrFwHK4F98MEHEfNk/S7I+9jvKXmvjxtaEpg3Vt3i45QHs7oNxxyrFrGCAWX7lFTSdvHd7343YladKKXaFzhWulaJjTBPdTtCZglixUC2CS0PfJwSYo5Z2sQeeOCBiCkZ/9nPfhbxE088MfCzSune+MjgtWfy+lLyMcVxc/To0YEx15t333134PvQRsFqVC+99FLEv/zlLyPmuKxXkexyPki2DtVzw7WIr+E8xhzQosV1iPMbLRW0InHtWb9+fcSs0MZ89CUXTeEYKqWaq3pFzeNwvmIFVY4PVlmjtY9r/pYtWz7yfcZbPsgwO1E2vrgO1deA43Au4h6a+y7mj3uH8ZyPpmS54X6B6wofP3LkSMT8PcXfWX3ZW7UJznW0fXG/wBxwHPC15mPk4Xji78Ts9zh/F3ahEjdR8SMiIiIiIiIi0lO88SMiIiIiIiIi0lNabfWqy4OPQ6n2gQMH0tdTuk3J4zvvvBMxrRCUbtH6QokdZZG0SLAyDk9r53PqUG5JuSxltG2VjWXWOdpP6tUCKJ+nVJHfl7J6tgkl+ZTesZILbXW0SNxzzz0Dn3PDDTcMfM9Sqn2H1SloZWkrWW4oJ62PG8qAKb+mJei9994b+Hr2Y7YjrV60QrDiGqu4sCobpZMci6VUbXnMU9fkr8wNx8Dhw4fT13C+Yr/knJiNG7YPLZJs95tuuiliVmL70pe+FDErvTz++OOV62O/6EtFI455joFSqtVRuH6wHdjuXIfYd9nfs2pUixYtiviRRx6JeOrUqRF///vfj/jll1+uXGsfrRRsW36/UqprCec3zoOs7EgLMdcx5oZV3Dhubr755ohpkfzOd74TMatLsq/Uv0dfYB+rz+Fsa44V2vM41rjv4n6MY5NW+jvvvHNgzPHx7LPPDrye8Vb5q973mCseecAKQ1yDuQ6de+65ETN/tHexoiStr31Z10eS+jzNv7nXYv9lPlg9j8dRsH25zhNWuuW+Y7yNj1OFOWPbce/AtYp7M1qOaBumDXY8j4+RhOs813/uzbnecB/MdasTvxHH+gJERERERERERGR08MaPiIiIiIiIiEhPabXVixK5TG5ft4Px32g1orSaknC+74lK5ihTZQWq++67L+IHH3wwYlpfSinl2muvjfj555+PmDK+ttIkN5QGl1KVOVJqSul9VkWKr81sCqzYQZvDhg0bIn700UcjpnVi7ty5lfeiVYz9qG4naCOZHJhty/5aStVexNew8gll33yvzNJDqTCrpK1bty7ir3zlKxE/9NBDEVPiSmtYKVVrwMaNGwdedxesLFmlobrVi2OFNgdKU9kvmZusyhalqWxDjhvau1hRivmgbbaUUn7+859HnNkzuwavvf59M8sKrSms5pSNG+Zyx44dEe/atSviu+++O2Jai1evXh0xq7vUJcfbt2+PmP2tC2Mlg2s25dmlVL8/bca0LXD/wFwyN3wO88F1+uGHH46Y6zrtR4899ljErPxVv76+WCmGWb04DrjG8Hm0P2R9lPNhtmasWrUqYlq9WWmKlQrrds4uz11NqO97sz0x97tcM/bs2RMxKxVl9tXly5dH/PnPfz5iVsOlJY9zbt9z8VFkv1EyqzfzxMd5RAJjVsNbuHBhxD/+8Y8jZj/oy1x1umD+Dh06FDHHEKHti1Y9zplct7q8lo81nK/4W4frENcMzle0uHJf0NZ8qPgREREREREREekp3vgREREREREREekprbZ6UXpFaRstBHVZLmWhtHdl8tVTkWJR5kipNm0tlBlfcskllddTmnzOOeec9HWMNZRnsx0o5y6lKkfNTko/FbtOJvvftm1bxGvWrImYFVooyS+llNmzZ0c8ceLEiNkn2yrjI+yjtAPRflJKte34GuaW8YmOIeaYFsBvfetbETMHt9xyS8SUuJZSteWxggX7VBdgG7J9KDMtpSrlJVme+HgmDed8yn7x9NNPD3zPb3zjGxGzUiEr45VSHWuZfaUL4yarxME5rJTqfEXbSWa3zGyRjJlv2vxoAeQ10SKZVZoqJa/2VrfhdImsnetwHHDuyyzLWX+ljDurQPm1r30tYq7xrC7J15ZSypYtWwZeX18qttTHPNs3s+RxPmc7ZHmmhY/7Qn72Zz/72Yjvv//+iLlXeeaZZyrvy31kX/IxDLYXvy/nPu6zOSaY16xKDq1FtLLcdtttEXPe45rEqoj1zxtvZGsU1w/asriv4O8N2ldo1Zs1a1bEtIDRqs88jXcbXhOYM+67WBn3vPPOi5h7CuaMzznRIzFkMLR0M2b7ci1gPs4666yBr23r/KTiR0RERERERESkp3jjR0RERERERESkp7Ta6kWZW1Z9oy69pZyRdobRluhmknHKWi+44ILKa3it9epkbSeTxQ2rkkbp9unMDT+LEntKkc8///zKa2jLO/PMM0fx6kYe5oPfMbNzlVK1e1CyO1K2SMLPpsR169atEVNaXB83tE9kVUTaKnPlNTIm9fHA9spsXJltqAl8PsfK5s2bI6aljhVaGJdSteXR0le3SLWRLDfDcsZ8ZOMmizMyqwVtDj/96U8jZhWpGTNmRFyvVMhc0VLTtWp4TXND+B2zPGUy+cxSwTZ84YUXIr7ssssi/upXvxoxLZJz5sypXB8rwtFGcyrjeqwZlg+S2fWyeS/LEy1ytBZxPeSekuODFYxYua2U3JbeV9tXlrdsnWC71ysJHof7C+7fuT+mtWjBggURD6uqyr/7mo+MJnMfc8N+zN8ejDmn8YgDViTmMQjsB/Wqim21uZxumBtWU+W8RAsR98S0ZLN9aTPKfvN2bV0fa5gbVuzivMI8ZUcw8H1G4/fTSNCtuw0iIiIiIiIiItIYb/yIiIiIiIiIiPSU1lm9KIs744z/Xh6lbZRh1StHjZXdg587efLkiCmvp6y1lPy08DZJwkgmWeT3Yp7qlaPGyu5BSSVPw2dczw2lyZTIdiE3lLPztHl+x3olnyZVU0aDTD7Ox88+++zKa9j3ulDFgLlhX6SFkN+xLttuUuVnpL57ZllllRy2+aRJkyqv55hqavVoC9n8xjHEsVVKtS2yalEjlRvmnpV0WLmFlQq5DpVSlevXv0fbyfYFzFP9O3GsNal61yRn2XO4Rrz22msRMze0rNKuWkp1HHW5Uk5mq6pb2Zv0vyZ2ycwWyRzv27cvYlpWab0799xzI66PG1rvMvtEW9eejGxNqv/NPHHcZfvszJLHtYTz5MaNGyPm2sGxwjzR9lV/3yYVE7tOljfmhjFh36VFjhZi/g5htS/aV2n14tEJ9f1+loM+5aMJzBmPJuD+j21Hi/wrr7wSMfPHHHPu4njlmlRfR8ZbDjKYm+y3EsfK7t27Iz548GDE2X6xif1/LFDxIyIiIiIiIiLSU7zxIyIiIiIiIiLSU1pn9SI8Vf7yyy+PmFLcN998s/Ia2oko/cxkoCMF5XZf/OIXI+Z1160PL774YsS0rLVVhsfrpxTu0ksvjZg2PNoRSsmtCqNdKYO5ueOOOyIeZn2gxDI7vb1NUPZLqSHzQTtBvToGpYqU+zI3I2WrYj9ibpYuXTrw8botjfaJrIpIW2E/4xhinurSe8qAmZtMynsqsuosN6zWxXm5nhtWnhhrOWsTMuk8Jdm0IPC7l1L9jqyuwXbI5rcTbR9eK6+JFQjZp+pzWrYGdq0aHmXu/L5sk1Kq+aSNlPM59wsnKsvO+g77CK3PjDlHl5LbM7qQG8J2yOTvpVRtDuynbGvmjGOIcWaFy6qO8nN5TbQWcd4rpWrDrR8tcJwu5KmJTaj+N787+y/nx2xeySpu8bP5ONudc1pmySul2RqYfV4XqP9myGx4WZ44F3F9Yrsxf9n4mDp1asSsgLtnz56IaQcvJbeNky6Mm5MhO3ohOxaDfZftyMfZ7swN9/W07bE6WL3fZxbZ8UZm9eJ44m9VWhsJrZAcQxyv3Gtkx1oM+ns0UPEjIiIiIiIiItJTvPEjIiIiIiIiItJTWm31okTu4osvjvjqq6+OmFaqUqpyb57ATQkcpY2nIren9O7rX/96xHfddVfElIxt27at8l5PPfVUxF2wE/G7U85GWe78+fMjrtuJDh8+HPH+/fsjfvvttyOmreVErUWZFeLLX/5yxLfffnvEzA37SimlrF27NmLmpk1yVH7fLDccNzNmzIi4/j0oZ2SemJtMVp1VWSGUPFKu/fDDD0d88803R0ybwKuvvlp5rxdeeGHgNXUhN1mlQs5jnFdKqbYvbW6U8tKCcKL2vKzq3Re+8IWIr7322oiZm7rVllVz6lU+2k5mCcmqm5RSldJzDDE3rNJEuW9mAWOesjXw3nvvjXjOnDkRs69x7JZSyqFDhyLObBFtJbOpcAzRulv/m+OAewHK6mnPy/JE2EemT58e8Z133hkxxzXfv24ZYt/pQj6awHmiXpmR1ir2a86PXHfZPrS78jnZ+s+5a+XKlQOvgeO1DteYbF7vWs4yy1Ap1VxxvmNfpi0iqxDF+Yd5ZaXbW2+9NeLrrrsuYuY1s+CWUt3L9yU32feo/80ccAwxT7QB0erFdZtzKH9b3XLLLREzZ1u3bo2Y+eB4LyUfN+OBrEJoZmfk4+yvbHeuZ5zfuJY899xzEXPOrFvyuf53eaycKvzunPfYvhxnfJzHhhDOe/ztQlskczOscudo5UPFj4iIiIiIiIhIT/HGj4iIiIiIiIhIT2md1YvSK8qzKfeklJHSxFJKmTVrVsQLFiyIeN26dRFT8k45Ij+PUOo1d+7ciL/5zW9GTMkqT24/cOCRHOS4AAAJ6UlEQVRAxN/+9rcr7/vaa6995Ge3FVoTMhkvpYylVKXub731VsTPPvtsxK+//nrElDBmlSOYG1qZHn300YhpvaO8m5am733ve5Vr3bt378DP7gKZzPTKK6+MuG6RJO+//37Ezz//fMS051FuzxPq2S9ohWBViEceeSTiBx54IGLKymlpevzxxyvXt2/fvoi7VtUrq6ZCOxHnmFKqMm629caNGyNmm1BGmtnzmBv2BVrv7rvvvoiZG9oifvKTn1SulXPraFdSHGmaVOJgPy6llNmzZ0fMfNLyxnFD2xfHEG1x2Zh98MEHI6bMmP3jjTfeiPiJJ56oXGtWDa8LuSFZxSbO7aVU9wYcX+y/XIc479GywnWI6w1zz7FCywqtgFzvaVctpdoXumbDI5ltoG4DYb+eN29exLSOcC3h3oH9mHmiRXbRokURr169OmJWiOJcyvfkXFpKdWw2sTV3gazqVSlVexAtjFOmTImY44nzHue3d955J2KuH9dcc03ES5YsGXgdzCtzw/cvZeSqWbaVukWK6zZtJ5z72MeZM45B2vO4b6ZtmDlmNSPOVcwHx1MpzY7R6EueSsntyMwZ9xLZno/rR1YtimsS94Ecu2QsKkd1gfrcdxzu/ziGFi5cGDHtebSm0tLFOZDVr2m7rFu9rOolIiIiIiIiIiInjTd+RERERERERER6ijd+RERERERERER6SuvO+KG/jSVV6aGj/65ewpUePL4Xz3nZvn17xDx/gb5J+vpWrVoV8VVXXRUxz1agT4/nLDz22GMRP/3005Vrpc+2C55LXiPPiOBZHtm5JaVUvd183m233RYx/ZFsR5bd5dlB119/fcQ802Hy5MkRs7/wXB+eHbNmzZrKtfJMga7lJjsviucn1HNDjzHLGrLvMx8sB0ofK32v8+fPHxhPmzYtYp6ZQR/5D3/4w4jXrl1buVb6Y5v4yNtEdrYC/fdsk1KqubnwwgsjXrZsWcTZOQgcN/SI8yyumTNnRszc8HwS5uZHP/pRxL/+9a8r18p5umvnljE39Ouz3erQt801g3MR1xX2XY4bfjbXNL4nzxdibnhmzQ9+8IOIX3rppcq18rynro2bDI6b+nfieQrs1zzHgnAd4xkjhOfI8GwsnsXAtY3n+vA8rFdeeaXyvtk5aV1YewivvWmZYJ5VwrN/OLY4BnmGGT+Da09W+ph7ys2bN0fMMyB57lMpzc5c6kKeml47vy/XKLYvz4XhmsTzkLiP4Nkj2Tkk3FM888wzEW/bti1izp+lVNeYLo+bjGFns2Rl3zlWWKqd8eLFiyPm7xi+luOAOVi/fn3E3BfUz/jhtfZlvWlKVr6e44NnyvHcJJ5PxjHH34tc87kv53rDuW5Y+/dlrDQlyw33R2xH/t5cvnx5xGw37ns5JrhWcU/BeWss2l/Fj4iIiIiIiIhIT/HGj4iIiIiIiIhIT5lwOmVGEyZM+MgPowyLlgfKEWk/oU2olKqlhLJhyhwpD6e0kZ9NaRxlWZTqUapP+SNtKiy1R9lX/TNOhWPHjg3Wrp0ATXJD2S/bk1LfG2+8MeKVK1dWXs/c0LZA+TxzTqkwc8N8ZLmhrO65556LmJYulpEfJu8+FU5XbmhnY24oq166dGnEtMiVUs1NJuOmvYQSVPYL2iWy3NB+tGnTpoiZD0rvWSK8lJEr4T4W44Z2O1pRZs2aFTHtqqWUsmDBgoHPo9WEeeJ4ou2EuWFMiTZtYjt27IiYY4iWVVonSxm5MuGnKzecV9hWXBdo4+HYKKU6bliWms9jeV2OR45ZthvXBeaGcvBdu3ZF/OKLL0ZMiwTl4PX36kJuCMcQ2402HlqDSqmWoqa9K5vraH/l53Eey+Y65mbr1q0Rs2w7Y86BpXRv3Ax5bcTcZ3HtKKVafpr7NFokaXngGGL+GXPMsj05p3E/RiskxxMtFaU0s3o1YaxzU3ufiDnvlZKXDOeejWtSZg/O9gvMDffQzAePY+AawzFXyshZJtqUm9p7pn8zb2xrzoPMDcccn8P1jbmhhYj5OHDgQMTDbPdd+30zWnCO4u8b2lE5v9HezTzRuvXqq69GzDzxdw/33KNVzr3ruSEcT9xDc3zwWATuF7Jy7jyCgc/hPDZa92CG5UbFj4iIiIiIiIhIT/HGj4iIiIiIiIhIT2md1YtQIkfZMGWjlJzW/16xYkXElDxSbk8pK+VXrOpEyTxl9bR3sTpYU7ndSDEWtgjmhvYVtietD6VUbSo33HBDxJQ2Uj5H+wrbjrmhlI62Icrq9+7dG/Hbb78dMSXdo1V1oE25oXyxblmh3H7JkiURU95NaWpWNY254en2tD+wmgqr2zA3rIQ0WtWhxiI3tIpQWso5rV5xjTJgWlaYQ+aPOaNsn1ULOEfR0kVJNyWrlOQPk3d3WULM3DCm5ZT2vFKqueKY4HrD+Y1zIGXfXHtY7YPjg/JuVp2gbYjjr0+5ySpxsH/XLSvMFccXpdvMGWXcfC0/m2sGLSiMDx48GDHt3ZTqjwfpfVZ1qJRqrjLLSia353jk+pZZ72j1YsxcjrX0vimnwxaR5S3bjzNnHI98DqF1jvtj5oOWo9NRHbIruWlCZovlOOPjfH52xEVm4T8dvx/7lBuS7QuHzZvHYZ7GsppdX3NDmuSDMAdjWTFNq5eIiIiIiIiIyDjEGz8iIiIiIiIiIj2l1Vavhu+Z/t0k5vdvYvdpi4yLtElu11QW10Tm2CQ3WT7MzcD3GRgPe16WG9IkN2MpR81oU25q79no37I8kaytzc3I02S9aYJz2ujSJB/Zc9qYD9L13DShSW7ayHjITVcxN+3F3LQXc9NetHqJiIiIiIiIiIxDvPEjIiIiIiIiItJTzvjop7Sb0aqOISfHyUjhT0fVBjE3XWFYbrJ/M09jT9ttQPL/NMmN+Wsv5kZEROTkUPEjIiIiIiIiItJTvPEjIiIiIiIiItJTvPEjIiIiIiIiItJTvPEjIiIiIiIiItJTvPEjIiIiIiIiItJTJlghQURERERERESkn6j4ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKd74ERERERERERHpKf8Hk71/L3quDd4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 11 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(num_test):\n",
    "    z1 = np.random.normal(0, 1, n_latent)\n",
    "    z2 = np.random.normal(0, 1, n_latent)\n",
    "\n",
    "    output_list = []\n",
    "\n",
    "    for j in range(11):\n",
    "        z_in = (j/10) * z1 + (1 - (j/10)) * z2\n",
    "\n",
    "        output = sess.run([y], feed_dict = {z: [z_in], keep_prob_decoder: 1})\n",
    "        output_reshape = np.reshape(output, [img_size, img_size])\n",
    "\n",
    "        output_list.append(output_reshape)\n",
    "\n",
    "    f, ax = plt.subplots(1,11,figsize=(20,15))\n",
    "\n",
    "    for j in range(len(output_list)):\n",
    "        ax[j].imshow(output_list[j], cmap = 'gray')\n",
    "        ax[j].axis('off')\n",
    "\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
}