pytorch_ipynb/autoencoder/ae-var.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Deep Learning Models -- A collection of various deep learning architectures, models, and tips for TensorFlow and PyTorch in Jupyter Notebooks.\n",
    "- Author: Sebastian Raschka\n",
    "- GitHub Repository: https://github.com/rasbt/deeplearning-models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sebastian Raschka \n",
      "\n",
      "CPython 3.7.1\n",
      "IPython 7.2.0\n",
      "\n",
      "torch 1.0.0\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -a 'Sebastian Raschka' -v -p torch"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Runs on CPU or GPU (if available)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Zoo -- Variational Autoencoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A simple variational autoencoder that compresses 768-pixel MNIST images down to a 15-pixel latent vector representation."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "from torch.utils.data import DataLoader\n",
    "from torchvision import datasets\n",
    "from torchvision import transforms\n",
    "\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    torch.backends.cudnn.deterministic = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Device: cuda:0\n",
      "Image batch dimensions: torch.Size([128, 1, 28, 28])\n",
      "Image label dimensions: torch.Size([128])\n"
     ]
    }
   ],
   "source": [
    "##########################\n",
    "### SETTINGS\n",
    "##########################\n",
    "\n",
    "# Device\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "print('Device:', device)\n",
    "\n",
    "# Hyperparameters\n",
    "random_seed = 0\n",
    "learning_rate = 0.001\n",
    "num_epochs = 50\n",
    "batch_size = 128\n",
    "\n",
    "# Architecture\n",
    "num_features = 784\n",
    "num_hidden_1 = 500\n",
    "num_latent = 15\n",
    "\n",
    "\n",
    "##########################\n",
    "### MNIST DATASET\n",
    "##########################\n",
    "\n",
    "# Note transforms.ToTensor() scales input images\n",
    "# to 0-1 range\n",
    "train_dataset = datasets.MNIST(root='data', \n",
    "                               train=True, \n",
    "                               transform=transforms.ToTensor(),\n",
    "                               download=True)\n",
    "\n",
    "test_dataset = datasets.MNIST(root='data', \n",
    "                              train=False, \n",
    "                              transform=transforms.ToTensor())\n",
    "\n",
    "\n",
    "train_loader = DataLoader(dataset=train_dataset, \n",
    "                          batch_size=batch_size, \n",
    "                          shuffle=True)\n",
    "\n",
    "test_loader = DataLoader(dataset=test_dataset, \n",
    "                         batch_size=batch_size, \n",
    "                         shuffle=False)\n",
    "\n",
    "# Checking the dataset\n",
    "for images, labels in train_loader:  \n",
    "    print('Image batch dimensions:', images.shape)\n",
    "    print('Image label dimensions:', labels.shape)\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "##########################\n",
    "### MODEL\n",
    "##########################\n",
    "\n",
    "class VariationalAutoencoder(torch.nn.Module):\n",
    "\n",
    "    def __init__(self, num_features, num_hidden_1, num_latent):\n",
    "        super(VariationalAutoencoder, self).__init__()\n",
    "        \n",
    "        ### ENCODER\n",
    "        self.hidden_1 = torch.nn.Linear(num_features, num_hidden_1)\n",
    "        self.z_mean = torch.nn.Linear(num_hidden_1, num_latent)\n",
    "        # in the original paper (Kingma & Welling 2015, we use\n",
    "        # have a z_mean and z_var, but the problem is that\n",
    "        # the z_var can be negative, which would cause issues\n",
    "        # in the log later. Hence we assume that latent vector\n",
    "        # has a z_mean and z_log_var component, and when we need\n",
    "        # the regular variance or std_dev, we simply use \n",
    "        # an exponential function\n",
    "        self.z_log_var = torch.nn.Linear(num_hidden_1, num_latent)\n",
    "        \n",
    "        \n",
    "        ### DECODER\n",
    "        self.linear_3 = torch.nn.Linear(num_latent, num_hidden_1)\n",
    "        self.linear_4 = torch.nn.Linear(num_hidden_1, num_features)\n",
    "\n",
    "    def reparameterize(self, z_mu, z_log_var):\n",
    "        # Sample epsilon from standard normal distribution\n",
    "        eps = torch.randn(z_mu.size(0), z_mu.size(1)).to(device)\n",
    "        # note that log(x^2) = 2*log(x); hence divide by 2 to get std_dev\n",
    "        # i.e., std_dev = exp(log(std_dev^2)/2) = exp(log(var)/2)\n",
    "        z = z_mu + eps * torch.exp(z_log_var/2.) \n",
    "        return z\n",
    "        \n",
    "    def encoder(self, features):\n",
    "        x = self.hidden_1(features)\n",
    "        x = F.leaky_relu(x, negative_slope=0.0001)\n",
    "        z_mean = self.z_mean(x)\n",
    "        z_log_var = self.z_log_var(x)\n",
    "        encoded = self.reparameterize(z_mean, z_log_var)\n",
    "        return z_mean, z_log_var, encoded\n",
    "    \n",
    "    def decoder(self, encoded):\n",
    "        x = self.linear_3(encoded)\n",
    "        x = F.leaky_relu(x, negative_slope=0.0001)\n",
    "        x = self.linear_4(x)\n",
    "        decoded = torch.sigmoid(x)\n",
    "        return decoded\n",
    "\n",
    "    def forward(self, features):\n",
    "        \n",
    "        z_mean, z_log_var, encoded = self.encoder(features)\n",
    "        decoded = self.decoder(encoded)\n",
    "        \n",
    "        return z_mean, z_log_var, encoded, decoded\n",
    "\n",
    "    \n",
    "torch.manual_seed(random_seed)\n",
    "model = VariationalAutoencoder(num_features,\n",
    "                               num_hidden_1,\n",
    "                               num_latent)\n",
    "model = model.to(device)\n",
    "    \n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 001/050 | Batch 000/469 | Cost: 70481.2422\n",
      "Epoch: 001/050 | Batch 050/469 | Cost: 27139.5547\n",
      "Epoch: 001/050 | Batch 100/469 | Cost: 22833.3730\n",
      "Epoch: 001/050 | Batch 150/469 | Cost: 19493.1523\n",
      "Epoch: 001/050 | Batch 200/469 | Cost: 18727.4688\n",
      "Epoch: 001/050 | Batch 250/469 | Cost: 18074.2676\n",
      "Epoch: 001/050 | Batch 300/469 | Cost: 16633.2852\n",
      "Epoch: 001/050 | Batch 350/469 | Cost: 17136.7852\n",
      "Epoch: 001/050 | Batch 400/469 | Cost: 16402.5293\n",
      "Epoch: 001/050 | Batch 450/469 | Cost: 16062.9814\n",
      "Epoch: 002/050 | Batch 000/469 | Cost: 16577.0840\n",
      "Epoch: 002/050 | Batch 050/469 | Cost: 15451.8242\n",
      "Epoch: 002/050 | Batch 100/469 | Cost: 15667.8535\n",
      "Epoch: 002/050 | Batch 150/469 | Cost: 15734.0801\n",
      "Epoch: 002/050 | Batch 200/469 | Cost: 15145.4365\n",
      "Epoch: 002/050 | Batch 250/469 | Cost: 15326.6953\n",
      "Epoch: 002/050 | Batch 300/469 | Cost: 15408.5801\n",
      "Epoch: 002/050 | Batch 350/469 | Cost: 15637.5430\n",
      "Epoch: 002/050 | Batch 400/469 | Cost: 14793.0332\n",
      "Epoch: 002/050 | Batch 450/469 | Cost: 15046.4414\n",
      "Epoch: 003/050 | Batch 000/469 | Cost: 14457.0537\n",
      "Epoch: 003/050 | Batch 050/469 | Cost: 14483.2910\n",
      "Epoch: 003/050 | Batch 100/469 | Cost: 14374.9258\n",
      "Epoch: 003/050 | Batch 150/469 | Cost: 13934.8672\n",
      "Epoch: 003/050 | Batch 200/469 | Cost: 15053.9336\n",
      "Epoch: 003/050 | Batch 250/469 | Cost: 14673.1025\n",
      "Epoch: 003/050 | Batch 300/469 | Cost: 14324.3916\n",
      "Epoch: 003/050 | Batch 350/469 | Cost: 14318.4229\n",
      "Epoch: 003/050 | Batch 400/469 | Cost: 14501.4912\n",
      "Epoch: 003/050 | Batch 450/469 | Cost: 13753.9082\n",
      "Epoch: 004/050 | Batch 000/469 | Cost: 15024.3789\n",
      "Epoch: 004/050 | Batch 050/469 | Cost: 14310.9219\n",
      "Epoch: 004/050 | Batch 100/469 | Cost: 14723.5176\n",
      "Epoch: 004/050 | Batch 150/469 | Cost: 15469.9473\n",
      "Epoch: 004/050 | Batch 200/469 | Cost: 14126.0586\n",
      "Epoch: 004/050 | Batch 250/469 | Cost: 14321.4062\n",
      "Epoch: 004/050 | Batch 300/469 | Cost: 13834.0576\n",
      "Epoch: 004/050 | Batch 350/469 | Cost: 14363.6494\n",
      "Epoch: 004/050 | Batch 400/469 | Cost: 14136.7422\n",
      "Epoch: 004/050 | Batch 450/469 | Cost: 13603.2012\n",
      "Epoch: 005/050 | Batch 000/469 | Cost: 14002.0479\n",
      "Epoch: 005/050 | Batch 050/469 | Cost: 14221.5488\n",
      "Epoch: 005/050 | Batch 100/469 | Cost: 13972.6787\n",
      "Epoch: 005/050 | Batch 150/469 | Cost: 13918.2402\n",
      "Epoch: 005/050 | Batch 200/469 | Cost: 13839.3809\n",
      "Epoch: 005/050 | Batch 250/469 | Cost: 14421.0020\n",
      "Epoch: 005/050 | Batch 300/469 | Cost: 14611.1816\n",
      "Epoch: 005/050 | Batch 350/469 | Cost: 13653.8027\n",
      "Epoch: 005/050 | Batch 400/469 | Cost: 13632.8047\n",
      "Epoch: 005/050 | Batch 450/469 | Cost: 13612.9375\n",
      "Epoch: 006/050 | Batch 000/469 | Cost: 13993.7344\n",
      "Epoch: 006/050 | Batch 050/469 | Cost: 13976.1006\n",
      "Epoch: 006/050 | Batch 100/469 | Cost: 14309.5527\n",
      "Epoch: 006/050 | Batch 150/469 | Cost: 13427.8916\n",
      "Epoch: 006/050 | Batch 200/469 | Cost: 13811.6260\n",
      "Epoch: 006/050 | Batch 250/469 | Cost: 14130.3496\n",
      "Epoch: 006/050 | Batch 300/469 | Cost: 12895.7324\n",
      "Epoch: 006/050 | Batch 350/469 | Cost: 13445.3213\n",
      "Epoch: 006/050 | Batch 400/469 | Cost: 13374.8242\n",
      "Epoch: 006/050 | Batch 450/469 | Cost: 13549.5098\n",
      "Epoch: 007/050 | Batch 000/469 | Cost: 13913.4043\n",
      "Epoch: 007/050 | Batch 050/469 | Cost: 13703.5654\n",
      "Epoch: 007/050 | Batch 100/469 | Cost: 14132.1758\n",
      "Epoch: 007/050 | Batch 150/469 | Cost: 14052.9814\n",
      "Epoch: 007/050 | Batch 200/469 | Cost: 13750.3535\n",
      "Epoch: 007/050 | Batch 250/469 | Cost: 14316.6953\n",
      "Epoch: 007/050 | Batch 300/469 | Cost: 13224.3281\n",
      "Epoch: 007/050 | Batch 350/469 | Cost: 14139.7979\n",
      "Epoch: 007/050 | Batch 400/469 | Cost: 13795.6016\n",
      "Epoch: 007/050 | Batch 450/469 | Cost: 13915.5020\n",
      "Epoch: 008/050 | Batch 000/469 | Cost: 13548.9512\n",
      "Epoch: 008/050 | Batch 050/469 | Cost: 13558.6338\n",
      "Epoch: 008/050 | Batch 100/469 | Cost: 13883.1074\n",
      "Epoch: 008/050 | Batch 150/469 | Cost: 13128.7617\n",
      "Epoch: 008/050 | Batch 200/469 | Cost: 13133.5879\n",
      "Epoch: 008/050 | Batch 250/469 | Cost: 13518.8672\n",
      "Epoch: 008/050 | Batch 300/469 | Cost: 13679.2324\n",
      "Epoch: 008/050 | Batch 350/469 | Cost: 13928.9824\n",
      "Epoch: 008/050 | Batch 400/469 | Cost: 14079.2256\n",
      "Epoch: 008/050 | Batch 450/469 | Cost: 13294.2021\n",
      "Epoch: 009/050 | Batch 000/469 | Cost: 13619.6504\n",
      "Epoch: 009/050 | Batch 050/469 | Cost: 13831.6201\n",
      "Epoch: 009/050 | Batch 100/469 | Cost: 13848.1406\n",
      "Epoch: 009/050 | Batch 150/469 | Cost: 14622.0889\n",
      "Epoch: 009/050 | Batch 200/469 | Cost: 13843.3887\n",
      "Epoch: 009/050 | Batch 250/469 | Cost: 13673.2441\n",
      "Epoch: 009/050 | Batch 300/469 | Cost: 13646.6543\n",
      "Epoch: 009/050 | Batch 350/469 | Cost: 13411.1816\n",
      "Epoch: 009/050 | Batch 400/469 | Cost: 14463.7988\n",
      "Epoch: 009/050 | Batch 450/469 | Cost: 13585.7891\n",
      "Epoch: 010/050 | Batch 000/469 | Cost: 13929.6816\n",
      "Epoch: 010/050 | Batch 050/469 | Cost: 13659.5176\n",
      "Epoch: 010/050 | Batch 100/469 | Cost: 13504.2568\n",
      "Epoch: 010/050 | Batch 150/469 | Cost: 13717.9434\n",
      "Epoch: 010/050 | Batch 200/469 | Cost: 13711.8818\n",
      "Epoch: 010/050 | Batch 250/469 | Cost: 13554.4062\n",
      "Epoch: 010/050 | Batch 300/469 | Cost: 13317.5156\n",
      "Epoch: 010/050 | Batch 350/469 | Cost: 13279.9912\n",
      "Epoch: 010/050 | Batch 400/469 | Cost: 13069.9648\n",
      "Epoch: 010/050 | Batch 450/469 | Cost: 13087.7695\n",
      "Epoch: 011/050 | Batch 000/469 | Cost: 13800.6113\n",
      "Epoch: 011/050 | Batch 050/469 | Cost: 13924.1973\n",
      "Epoch: 011/050 | Batch 100/469 | Cost: 13173.4414\n",
      "Epoch: 011/050 | Batch 150/469 | Cost: 13963.7402\n",
      "Epoch: 011/050 | Batch 200/469 | Cost: 13682.3281\n",
      "Epoch: 011/050 | Batch 250/469 | Cost: 13664.8027\n",
      "Epoch: 011/050 | Batch 300/469 | Cost: 14188.4707\n",
      "Epoch: 011/050 | Batch 350/469 | Cost: 13625.5840\n",
      "Epoch: 011/050 | Batch 400/469 | Cost: 13482.8643\n",
      "Epoch: 011/050 | Batch 450/469 | Cost: 13912.9238\n",
      "Epoch: 012/050 | Batch 000/469 | Cost: 13048.4648\n",
      "Epoch: 012/050 | Batch 050/469 | Cost: 13041.4395\n",
      "Epoch: 012/050 | Batch 100/469 | Cost: 13212.3662\n",
      "Epoch: 012/050 | Batch 150/469 | Cost: 13304.4463\n",
      "Epoch: 012/050 | Batch 200/469 | Cost: 13445.9287\n",
      "Epoch: 012/050 | Batch 250/469 | Cost: 13693.2676\n",
      "Epoch: 012/050 | Batch 300/469 | Cost: 13072.9004\n",
      "Epoch: 012/050 | Batch 350/469 | Cost: 13414.5361\n",
      "Epoch: 012/050 | Batch 400/469 | Cost: 13669.4121\n",
      "Epoch: 012/050 | Batch 450/469 | Cost: 13366.8633\n",
      "Epoch: 013/050 | Batch 000/469 | Cost: 13785.0518\n",
      "Epoch: 013/050 | Batch 050/469 | Cost: 13788.7734\n",
      "Epoch: 013/050 | Batch 100/469 | Cost: 13442.9023\n",
      "Epoch: 013/050 | Batch 150/469 | Cost: 13771.4902\n",
      "Epoch: 013/050 | Batch 200/469 | Cost: 13357.2217\n",
      "Epoch: 013/050 | Batch 250/469 | Cost: 13402.6758\n",
      "Epoch: 013/050 | Batch 300/469 | Cost: 13852.4033\n",
      "Epoch: 013/050 | Batch 350/469 | Cost: 13301.3457\n",
      "Epoch: 013/050 | Batch 400/469 | Cost: 13379.6172\n",
      "Epoch: 013/050 | Batch 450/469 | Cost: 14275.3047\n",
      "Epoch: 014/050 | Batch 000/469 | Cost: 13433.3906\n",
      "Epoch: 014/050 | Batch 050/469 | Cost: 13059.2354\n",
      "Epoch: 014/050 | Batch 100/469 | Cost: 14031.3721\n",
      "Epoch: 014/050 | Batch 150/469 | Cost: 13950.9883\n",
      "Epoch: 014/050 | Batch 200/469 | Cost: 13684.6611\n",
      "Epoch: 014/050 | Batch 250/469 | Cost: 13630.9336\n",
      "Epoch: 014/050 | Batch 300/469 | Cost: 13527.6230\n",
      "Epoch: 014/050 | Batch 350/469 | Cost: 13746.0527\n",
      "Epoch: 014/050 | Batch 400/469 | Cost: 13490.6982\n",
      "Epoch: 014/050 | Batch 450/469 | Cost: 13669.7402\n",
      "Epoch: 015/050 | Batch 000/469 | Cost: 13422.4238\n",
      "Epoch: 015/050 | Batch 050/469 | Cost: 13303.6855\n",
      "Epoch: 015/050 | Batch 100/469 | Cost: 13421.2900\n",
      "Epoch: 015/050 | Batch 150/469 | Cost: 13129.2764\n",
      "Epoch: 015/050 | Batch 200/469 | Cost: 13276.9336\n",
      "Epoch: 015/050 | Batch 250/469 | Cost: 13776.0889\n",
      "Epoch: 015/050 | Batch 300/469 | Cost: 13634.2188\n",
      "Epoch: 015/050 | Batch 350/469 | Cost: 13438.3828\n",
      "Epoch: 015/050 | Batch 400/469 | Cost: 13401.1045\n",
      "Epoch: 015/050 | Batch 450/469 | Cost: 13567.6631\n",
      "Epoch: 016/050 | Batch 000/469 | Cost: 13150.5625\n",
      "Epoch: 016/050 | Batch 050/469 | Cost: 13414.6553\n",
      "Epoch: 016/050 | Batch 100/469 | Cost: 12979.0029\n",
      "Epoch: 016/050 | Batch 150/469 | Cost: 13146.0801\n",
      "Epoch: 016/050 | Batch 200/469 | Cost: 13257.3301\n",
      "Epoch: 016/050 | Batch 250/469 | Cost: 14350.7471\n",
      "Epoch: 016/050 | Batch 300/469 | Cost: 13836.4316\n",
      "Epoch: 016/050 | Batch 350/469 | Cost: 13865.9902\n",
      "Epoch: 016/050 | Batch 400/469 | Cost: 13237.3877\n",
      "Epoch: 016/050 | Batch 450/469 | Cost: 13339.0303\n",
      "Epoch: 017/050 | Batch 000/469 | Cost: 13266.2793\n",
      "Epoch: 017/050 | Batch 050/469 | Cost: 13568.0957\n",
      "Epoch: 017/050 | Batch 100/469 | Cost: 12923.4482\n",
      "Epoch: 017/050 | Batch 150/469 | Cost: 14093.2070\n",
      "Epoch: 017/050 | Batch 200/469 | Cost: 13326.7510\n",
      "Epoch: 017/050 | Batch 250/469 | Cost: 13965.0625\n",
      "Epoch: 017/050 | Batch 300/469 | Cost: 13380.1445\n",
      "Epoch: 017/050 | Batch 350/469 | Cost: 13277.1875\n",
      "Epoch: 017/050 | Batch 400/469 | Cost: 13872.7607\n",
      "Epoch: 017/050 | Batch 450/469 | Cost: 13272.6797\n",
      "Epoch: 018/050 | Batch 000/469 | Cost: 13048.2461\n",
      "Epoch: 018/050 | Batch 050/469 | Cost: 13508.2314\n",
      "Epoch: 018/050 | Batch 100/469 | Cost: 12814.9834\n",
      "Epoch: 018/050 | Batch 150/469 | Cost: 13623.1924\n",
      "Epoch: 018/050 | Batch 200/469 | Cost: 13246.6113\n",
      "Epoch: 018/050 | Batch 250/469 | Cost: 13471.1328\n",
      "Epoch: 018/050 | Batch 300/469 | Cost: 13271.7930\n",
      "Epoch: 018/050 | Batch 350/469 | Cost: 13494.4883\n",
      "Epoch: 018/050 | Batch 400/469 | Cost: 13280.4316\n",
      "Epoch: 018/050 | Batch 450/469 | Cost: 13408.9775\n",
      "Epoch: 019/050 | Batch 000/469 | Cost: 13347.4941\n",
      "Epoch: 019/050 | Batch 050/469 | Cost: 13403.6006\n",
      "Epoch: 019/050 | Batch 100/469 | Cost: 12944.8574\n",
      "Epoch: 019/050 | Batch 150/469 | Cost: 13410.6201\n",
      "Epoch: 019/050 | Batch 200/469 | Cost: 13398.5342\n",
      "Epoch: 019/050 | Batch 250/469 | Cost: 13786.6992\n",
      "Epoch: 019/050 | Batch 300/469 | Cost: 12185.1465\n",
      "Epoch: 019/050 | Batch 350/469 | Cost: 13143.7744\n",
      "Epoch: 019/050 | Batch 400/469 | Cost: 13101.7451\n",
      "Epoch: 019/050 | Batch 450/469 | Cost: 13410.3252\n",
      "Epoch: 020/050 | Batch 000/469 | Cost: 13459.7676\n",
      "Epoch: 020/050 | Batch 050/469 | Cost: 13551.5127\n",
      "Epoch: 020/050 | Batch 100/469 | Cost: 13246.3486\n",
      "Epoch: 020/050 | Batch 150/469 | Cost: 13524.1133\n",
      "Epoch: 020/050 | Batch 200/469 | Cost: 13695.5605\n",
      "Epoch: 020/050 | Batch 250/469 | Cost: 13447.3887\n",
      "Epoch: 020/050 | Batch 300/469 | Cost: 13389.4941\n",
      "Epoch: 020/050 | Batch 350/469 | Cost: 13180.2422\n",
      "Epoch: 020/050 | Batch 400/469 | Cost: 13606.3457\n",
      "Epoch: 020/050 | Batch 450/469 | Cost: 13646.9355\n",
      "Epoch: 021/050 | Batch 000/469 | Cost: 13557.3623\n",
      "Epoch: 021/050 | Batch 050/469 | Cost: 13147.0098\n",
      "Epoch: 021/050 | Batch 100/469 | Cost: 13287.7227\n",
      "Epoch: 021/050 | Batch 150/469 | Cost: 12849.9639\n",
      "Epoch: 021/050 | Batch 200/469 | Cost: 13058.6406\n",
      "Epoch: 021/050 | Batch 250/469 | Cost: 13192.6367\n",
      "Epoch: 021/050 | Batch 300/469 | Cost: 13393.4082\n",
      "Epoch: 021/050 | Batch 350/469 | Cost: 13834.2705\n",
      "Epoch: 021/050 | Batch 400/469 | Cost: 13503.1680\n",
      "Epoch: 021/050 | Batch 450/469 | Cost: 13592.5518\n",
      "Epoch: 022/050 | Batch 000/469 | Cost: 13658.5986\n",
      "Epoch: 022/050 | Batch 050/469 | Cost: 13389.6855\n",
      "Epoch: 022/050 | Batch 100/469 | Cost: 13313.9707\n",
      "Epoch: 022/050 | Batch 150/469 | Cost: 13508.8438\n",
      "Epoch: 022/050 | Batch 200/469 | Cost: 12984.4082\n",
      "Epoch: 022/050 | Batch 250/469 | Cost: 13159.5137\n",
      "Epoch: 022/050 | Batch 300/469 | Cost: 13195.3516\n",
      "Epoch: 022/050 | Batch 350/469 | Cost: 13606.1777\n",
      "Epoch: 022/050 | Batch 400/469 | Cost: 12865.0508\n",
      "Epoch: 022/050 | Batch 450/469 | Cost: 13227.6514\n",
      "Epoch: 023/050 | Batch 000/469 | Cost: 13067.6016\n",
      "Epoch: 023/050 | Batch 050/469 | Cost: 13425.5498\n",
      "Epoch: 023/050 | Batch 100/469 | Cost: 13016.2773\n",
      "Epoch: 023/050 | Batch 150/469 | Cost: 13322.1260\n",
      "Epoch: 023/050 | Batch 200/469 | Cost: 12861.3926\n",
      "Epoch: 023/050 | Batch 250/469 | Cost: 13000.5967\n",
      "Epoch: 023/050 | Batch 300/469 | Cost: 13761.4629\n",
      "Epoch: 023/050 | Batch 350/469 | Cost: 13482.9814\n",
      "Epoch: 023/050 | Batch 400/469 | Cost: 12838.6201\n",
      "Epoch: 023/050 | Batch 450/469 | Cost: 13252.9746\n",
      "Epoch: 024/050 | Batch 000/469 | Cost: 13445.4590\n",
      "Epoch: 024/050 | Batch 050/469 | Cost: 13583.6416\n",
      "Epoch: 024/050 | Batch 100/469 | Cost: 13507.9443\n",
      "Epoch: 024/050 | Batch 150/469 | Cost: 13385.5938\n",
      "Epoch: 024/050 | Batch 200/469 | Cost: 13271.1357\n",
      "Epoch: 024/050 | Batch 250/469 | Cost: 13643.7109\n",
      "Epoch: 024/050 | Batch 300/469 | Cost: 13713.0889\n",
      "Epoch: 024/050 | Batch 350/469 | Cost: 12844.5703\n",
      "Epoch: 024/050 | Batch 400/469 | Cost: 12984.4746\n",
      "Epoch: 024/050 | Batch 450/469 | Cost: 13015.4365\n",
      "Epoch: 025/050 | Batch 000/469 | Cost: 13575.6875\n",
      "Epoch: 025/050 | Batch 050/469 | Cost: 13195.2832\n",
      "Epoch: 025/050 | Batch 100/469 | Cost: 13478.4746\n",
      "Epoch: 025/050 | Batch 150/469 | Cost: 13194.2852\n",
      "Epoch: 025/050 | Batch 200/469 | Cost: 12877.8242\n",
      "Epoch: 025/050 | Batch 250/469 | Cost: 13061.2148\n",
      "Epoch: 025/050 | Batch 300/469 | Cost: 13397.2266\n",
      "Epoch: 025/050 | Batch 350/469 | Cost: 12763.3711\n",
      "Epoch: 025/050 | Batch 400/469 | Cost: 13262.5332\n",
      "Epoch: 025/050 | Batch 450/469 | Cost: 13390.2393\n",
      "Epoch: 026/050 | Batch 000/469 | Cost: 13211.5508\n",
      "Epoch: 026/050 | Batch 050/469 | Cost: 13458.3652\n",
      "Epoch: 026/050 | Batch 100/469 | Cost: 12846.0557\n",
      "Epoch: 026/050 | Batch 150/469 | Cost: 12842.9570\n",
      "Epoch: 026/050 | Batch 200/469 | Cost: 13594.9395\n",
      "Epoch: 026/050 | Batch 250/469 | Cost: 13021.0605\n",
      "Epoch: 026/050 | Batch 300/469 | Cost: 13126.7686\n",
      "Epoch: 026/050 | Batch 350/469 | Cost: 12951.5898\n",
      "Epoch: 026/050 | Batch 400/469 | Cost: 13600.2119\n",
      "Epoch: 026/050 | Batch 450/469 | Cost: 13313.3535\n",
      "Epoch: 027/050 | Batch 000/469 | Cost: 12835.4717\n",
      "Epoch: 027/050 | Batch 050/469 | Cost: 12731.1875\n",
      "Epoch: 027/050 | Batch 100/469 | Cost: 13234.9297\n",
      "Epoch: 027/050 | Batch 150/469 | Cost: 13105.2148\n",
      "Epoch: 027/050 | Batch 200/469 | Cost: 13234.0684\n",
      "Epoch: 027/050 | Batch 250/469 | Cost: 13147.0801\n",
      "Epoch: 027/050 | Batch 300/469 | Cost: 13271.8262\n",
      "Epoch: 027/050 | Batch 350/469 | Cost: 12936.5947\n",
      "Epoch: 027/050 | Batch 400/469 | Cost: 13336.0293\n",
      "Epoch: 027/050 | Batch 450/469 | Cost: 13387.3662\n",
      "Epoch: 028/050 | Batch 000/469 | Cost: 13452.3438\n",
      "Epoch: 028/050 | Batch 050/469 | Cost: 13245.0342\n",
      "Epoch: 028/050 | Batch 100/469 | Cost: 13007.5234\n",
      "Epoch: 028/050 | Batch 150/469 | Cost: 13068.0166\n",
      "Epoch: 028/050 | Batch 200/469 | Cost: 12575.0166\n",
      "Epoch: 028/050 | Batch 250/469 | Cost: 13051.9434\n",
      "Epoch: 028/050 | Batch 300/469 | Cost: 13185.3330\n",
      "Epoch: 028/050 | Batch 350/469 | Cost: 13587.7715\n",
      "Epoch: 028/050 | Batch 400/469 | Cost: 12877.1436\n",
      "Epoch: 028/050 | Batch 450/469 | Cost: 13305.9297\n",
      "Epoch: 029/050 | Batch 000/469 | Cost: 13244.1865\n",
      "Epoch: 029/050 | Batch 050/469 | Cost: 13002.6309\n",
      "Epoch: 029/050 | Batch 100/469 | Cost: 13432.6504\n",
      "Epoch: 029/050 | Batch 150/469 | Cost: 13128.8027\n",
      "Epoch: 029/050 | Batch 200/469 | Cost: 12879.6543\n",
      "Epoch: 029/050 | Batch 250/469 | Cost: 13248.0068\n",
      "Epoch: 029/050 | Batch 300/469 | Cost: 13176.9912\n",
      "Epoch: 029/050 | Batch 350/469 | Cost: 13055.7490\n",
      "Epoch: 029/050 | Batch 400/469 | Cost: 13092.9580\n",
      "Epoch: 029/050 | Batch 450/469 | Cost: 13179.1875\n",
      "Epoch: 030/050 | Batch 000/469 | Cost: 13205.4668\n",
      "Epoch: 030/050 | Batch 050/469 | Cost: 13425.4883\n",
      "Epoch: 030/050 | Batch 100/469 | Cost: 12924.2070\n",
      "Epoch: 030/050 | Batch 150/469 | Cost: 13293.8105\n",
      "Epoch: 030/050 | Batch 200/469 | Cost: 12805.0674\n",
      "Epoch: 030/050 | Batch 250/469 | Cost: 12823.4629\n",
      "Epoch: 030/050 | Batch 300/469 | Cost: 12680.0322\n",
      "Epoch: 030/050 | Batch 350/469 | Cost: 13412.4023\n",
      "Epoch: 030/050 | Batch 400/469 | Cost: 13796.5479\n",
      "Epoch: 030/050 | Batch 450/469 | Cost: 13084.7051\n",
      "Epoch: 031/050 | Batch 000/469 | Cost: 13054.2988\n",
      "Epoch: 031/050 | Batch 050/469 | Cost: 13315.4570\n",
      "Epoch: 031/050 | Batch 100/469 | Cost: 13284.9463\n",
      "Epoch: 031/050 | Batch 150/469 | Cost: 13184.4668\n",
      "Epoch: 031/050 | Batch 200/469 | Cost: 13099.4189\n",
      "Epoch: 031/050 | Batch 250/469 | Cost: 13391.0918\n",
      "Epoch: 031/050 | Batch 300/469 | Cost: 13057.3223\n",
      "Epoch: 031/050 | Batch 350/469 | Cost: 13442.3750\n",
      "Epoch: 031/050 | Batch 400/469 | Cost: 13491.5635\n",
      "Epoch: 031/050 | Batch 450/469 | Cost: 13054.0693\n",
      "Epoch: 032/050 | Batch 000/469 | Cost: 13219.3789\n",
      "Epoch: 032/050 | Batch 050/469 | Cost: 12822.7051\n",
      "Epoch: 032/050 | Batch 100/469 | Cost: 13439.6436\n",
      "Epoch: 032/050 | Batch 150/469 | Cost: 12843.7061\n",
      "Epoch: 032/050 | Batch 200/469 | Cost: 13097.7012\n",
      "Epoch: 032/050 | Batch 250/469 | Cost: 12950.4707\n",
      "Epoch: 032/050 | Batch 300/469 | Cost: 13238.1094\n",
      "Epoch: 032/050 | Batch 350/469 | Cost: 13027.9121\n",
      "Epoch: 032/050 | Batch 400/469 | Cost: 13150.9277\n",
      "Epoch: 032/050 | Batch 450/469 | Cost: 13239.6348\n",
      "Epoch: 033/050 | Batch 000/469 | Cost: 12967.9863\n",
      "Epoch: 033/050 | Batch 050/469 | Cost: 13261.3467\n",
      "Epoch: 033/050 | Batch 100/469 | Cost: 13218.9023\n",
      "Epoch: 033/050 | Batch 150/469 | Cost: 13092.8994\n",
      "Epoch: 033/050 | Batch 200/469 | Cost: 12983.0459\n",
      "Epoch: 033/050 | Batch 250/469 | Cost: 13031.2188\n",
      "Epoch: 033/050 | Batch 300/469 | Cost: 12894.7129\n",
      "Epoch: 033/050 | Batch 350/469 | Cost: 13563.2578\n",
      "Epoch: 033/050 | Batch 400/469 | Cost: 13094.8340\n",
      "Epoch: 033/050 | Batch 450/469 | Cost: 13279.9639\n",
      "Epoch: 034/050 | Batch 000/469 | Cost: 12986.0615\n",
      "Epoch: 034/050 | Batch 050/469 | Cost: 12981.4004\n",
      "Epoch: 034/050 | Batch 100/469 | Cost: 13308.1504\n",
      "Epoch: 034/050 | Batch 150/469 | Cost: 13338.7227\n",
      "Epoch: 034/050 | Batch 200/469 | Cost: 13310.7227\n",
      "Epoch: 034/050 | Batch 250/469 | Cost: 13158.7334\n",
      "Epoch: 034/050 | Batch 300/469 | Cost: 13248.9336\n",
      "Epoch: 034/050 | Batch 350/469 | Cost: 13256.2227\n",
      "Epoch: 034/050 | Batch 400/469 | Cost: 12818.7148\n",
      "Epoch: 034/050 | Batch 450/469 | Cost: 12835.6738\n",
      "Epoch: 035/050 | Batch 000/469 | Cost: 12766.6123\n",
      "Epoch: 035/050 | Batch 050/469 | Cost: 12521.5166\n",
      "Epoch: 035/050 | Batch 100/469 | Cost: 12340.0430\n",
      "Epoch: 035/050 | Batch 150/469 | Cost: 12873.6191\n",
      "Epoch: 035/050 | Batch 200/469 | Cost: 13027.2266\n",
      "Epoch: 035/050 | Batch 250/469 | Cost: 13575.8379\n",
      "Epoch: 035/050 | Batch 300/469 | Cost: 13458.8867\n",
      "Epoch: 035/050 | Batch 350/469 | Cost: 12816.4980\n",
      "Epoch: 035/050 | Batch 400/469 | Cost: 12663.2207\n",
      "Epoch: 035/050 | Batch 450/469 | Cost: 12733.6777\n",
      "Epoch: 036/050 | Batch 000/469 | Cost: 13078.8682\n",
      "Epoch: 036/050 | Batch 050/469 | Cost: 13072.0742\n",
      "Epoch: 036/050 | Batch 100/469 | Cost: 12666.5215\n",
      "Epoch: 036/050 | Batch 150/469 | Cost: 13091.2852\n",
      "Epoch: 036/050 | Batch 200/469 | Cost: 13462.2529\n",
      "Epoch: 036/050 | Batch 250/469 | Cost: 12630.4287\n",
      "Epoch: 036/050 | Batch 300/469 | Cost: 13213.3223\n",
      "Epoch: 036/050 | Batch 350/469 | Cost: 13298.2490\n",
      "Epoch: 036/050 | Batch 400/469 | Cost: 12989.6328\n",
      "Epoch: 036/050 | Batch 450/469 | Cost: 12918.6348\n",
      "Epoch: 037/050 | Batch 000/469 | Cost: 12605.0732\n",
      "Epoch: 037/050 | Batch 050/469 | Cost: 13055.5742\n",
      "Epoch: 037/050 | Batch 100/469 | Cost: 12719.5420\n",
      "Epoch: 037/050 | Batch 150/469 | Cost: 12599.2461\n",
      "Epoch: 037/050 | Batch 200/469 | Cost: 12545.8223\n",
      "Epoch: 037/050 | Batch 250/469 | Cost: 12449.0918\n",
      "Epoch: 037/050 | Batch 300/469 | Cost: 13342.7930\n",
      "Epoch: 037/050 | Batch 350/469 | Cost: 13066.0029\n",
      "Epoch: 037/050 | Batch 400/469 | Cost: 13258.0957\n",
      "Epoch: 037/050 | Batch 450/469 | Cost: 13180.6914\n",
      "Epoch: 038/050 | Batch 000/469 | Cost: 12527.9854\n",
      "Epoch: 038/050 | Batch 050/469 | Cost: 13618.6875\n",
      "Epoch: 038/050 | Batch 100/469 | Cost: 13039.2627\n",
      "Epoch: 038/050 | Batch 150/469 | Cost: 13062.9453\n",
      "Epoch: 038/050 | Batch 200/469 | Cost: 13139.8945\n",
      "Epoch: 038/050 | Batch 250/469 | Cost: 13168.6621\n",
      "Epoch: 038/050 | Batch 300/469 | Cost: 12623.4629\n",
      "Epoch: 038/050 | Batch 350/469 | Cost: 12757.8447\n",
      "Epoch: 038/050 | Batch 400/469 | Cost: 12830.0762\n",
      "Epoch: 038/050 | Batch 450/469 | Cost: 12733.7969\n",
      "Epoch: 039/050 | Batch 000/469 | Cost: 12927.6729\n",
      "Epoch: 039/050 | Batch 050/469 | Cost: 13016.1133\n",
      "Epoch: 039/050 | Batch 100/469 | Cost: 12955.1621\n",
      "Epoch: 039/050 | Batch 150/469 | Cost: 12945.2852\n",
      "Epoch: 039/050 | Batch 200/469 | Cost: 12680.2188\n",
      "Epoch: 039/050 | Batch 250/469 | Cost: 12958.4688\n",
      "Epoch: 039/050 | Batch 300/469 | Cost: 13075.4912\n",
      "Epoch: 039/050 | Batch 350/469 | Cost: 12962.3750\n",
      "Epoch: 039/050 | Batch 400/469 | Cost: 12863.8867\n",
      "Epoch: 039/050 | Batch 450/469 | Cost: 13399.3818\n",
      "Epoch: 040/050 | Batch 000/469 | Cost: 12694.0283\n",
      "Epoch: 040/050 | Batch 050/469 | Cost: 13524.2754\n",
      "Epoch: 040/050 | Batch 100/469 | Cost: 12840.9316\n",
      "Epoch: 040/050 | Batch 150/469 | Cost: 12661.5918\n",
      "Epoch: 040/050 | Batch 200/469 | Cost: 13256.4902\n",
      "Epoch: 040/050 | Batch 250/469 | Cost: 13027.6816\n",
      "Epoch: 040/050 | Batch 300/469 | Cost: 12941.4727\n",
      "Epoch: 040/050 | Batch 350/469 | Cost: 12656.1348\n",
      "Epoch: 040/050 | Batch 400/469 | Cost: 12979.4785\n",
      "Epoch: 040/050 | Batch 450/469 | Cost: 12705.2158\n",
      "Epoch: 041/050 | Batch 000/469 | Cost: 12759.9707\n",
      "Epoch: 041/050 | Batch 050/469 | Cost: 12406.5781\n",
      "Epoch: 041/050 | Batch 100/469 | Cost: 12696.9307\n",
      "Epoch: 041/050 | Batch 150/469 | Cost: 13398.3613\n",
      "Epoch: 041/050 | Batch 200/469 | Cost: 12777.3418\n",
      "Epoch: 041/050 | Batch 250/469 | Cost: 12854.2783\n",
      "Epoch: 041/050 | Batch 300/469 | Cost: 13037.7236\n",
      "Epoch: 041/050 | Batch 350/469 | Cost: 13410.0801\n",
      "Epoch: 041/050 | Batch 400/469 | Cost: 13350.9121\n",
      "Epoch: 041/050 | Batch 450/469 | Cost: 12898.7432\n",
      "Epoch: 042/050 | Batch 000/469 | Cost: 12766.4316\n",
      "Epoch: 042/050 | Batch 050/469 | Cost: 13303.4766\n",
      "Epoch: 042/050 | Batch 100/469 | Cost: 13112.1465\n",
      "Epoch: 042/050 | Batch 150/469 | Cost: 12951.8428\n",
      "Epoch: 042/050 | Batch 200/469 | Cost: 13367.4814\n",
      "Epoch: 042/050 | Batch 250/469 | Cost: 13274.8955\n",
      "Epoch: 042/050 | Batch 300/469 | Cost: 12908.4805\n",
      "Epoch: 042/050 | Batch 350/469 | Cost: 12858.0723\n",
      "Epoch: 042/050 | Batch 400/469 | Cost: 13279.5410\n",
      "Epoch: 042/050 | Batch 450/469 | Cost: 12669.7422\n",
      "Epoch: 043/050 | Batch 000/469 | Cost: 13124.0225\n",
      "Epoch: 043/050 | Batch 050/469 | Cost: 12976.3857\n",
      "Epoch: 043/050 | Batch 100/469 | Cost: 12655.5703\n",
      "Epoch: 043/050 | Batch 150/469 | Cost: 12876.7061\n",
      "Epoch: 043/050 | Batch 200/469 | Cost: 13277.1592\n",
      "Epoch: 043/050 | Batch 250/469 | Cost: 12657.5117\n",
      "Epoch: 043/050 | Batch 300/469 | Cost: 12915.8867\n",
      "Epoch: 043/050 | Batch 350/469 | Cost: 13254.9941\n",
      "Epoch: 043/050 | Batch 400/469 | Cost: 12649.9102\n",
      "Epoch: 043/050 | Batch 450/469 | Cost: 13198.5771\n",
      "Epoch: 044/050 | Batch 000/469 | Cost: 13573.9121\n",
      "Epoch: 044/050 | Batch 050/469 | Cost: 12972.9453\n",
      "Epoch: 044/050 | Batch 100/469 | Cost: 12764.2188\n",
      "Epoch: 044/050 | Batch 150/469 | Cost: 13482.2910\n",
      "Epoch: 044/050 | Batch 200/469 | Cost: 13304.8975\n",
      "Epoch: 044/050 | Batch 250/469 | Cost: 13446.4141\n",
      "Epoch: 044/050 | Batch 300/469 | Cost: 13096.8887\n",
      "Epoch: 044/050 | Batch 350/469 | Cost: 13551.5537\n",
      "Epoch: 044/050 | Batch 400/469 | Cost: 12693.8301\n",
      "Epoch: 044/050 | Batch 450/469 | Cost: 12812.8682\n",
      "Epoch: 045/050 | Batch 000/469 | Cost: 12698.2207\n",
      "Epoch: 045/050 | Batch 050/469 | Cost: 12705.6787\n",
      "Epoch: 045/050 | Batch 100/469 | Cost: 13046.6162\n",
      "Epoch: 045/050 | Batch 150/469 | Cost: 13085.3457\n",
      "Epoch: 045/050 | Batch 200/469 | Cost: 12922.5312\n",
      "Epoch: 045/050 | Batch 250/469 | Cost: 13367.9189\n",
      "Epoch: 045/050 | Batch 300/469 | Cost: 12917.3457\n",
      "Epoch: 045/050 | Batch 350/469 | Cost: 12896.9463\n",
      "Epoch: 045/050 | Batch 400/469 | Cost: 13378.9902\n",
      "Epoch: 045/050 | Batch 450/469 | Cost: 12873.3105\n",
      "Epoch: 046/050 | Batch 000/469 | Cost: 12739.6260\n",
      "Epoch: 046/050 | Batch 050/469 | Cost: 13021.6465\n",
      "Epoch: 046/050 | Batch 100/469 | Cost: 13027.7256\n",
      "Epoch: 046/050 | Batch 150/469 | Cost: 12995.7490\n",
      "Epoch: 046/050 | Batch 200/469 | Cost: 12588.5645\n",
      "Epoch: 046/050 | Batch 250/469 | Cost: 13288.1494\n",
      "Epoch: 046/050 | Batch 300/469 | Cost: 12766.4707\n",
      "Epoch: 046/050 | Batch 350/469 | Cost: 12326.2334\n",
      "Epoch: 046/050 | Batch 400/469 | Cost: 13174.7734\n",
      "Epoch: 046/050 | Batch 450/469 | Cost: 12531.1074\n",
      "Epoch: 047/050 | Batch 000/469 | Cost: 13050.0781\n",
      "Epoch: 047/050 | Batch 050/469 | Cost: 12920.9609\n",
      "Epoch: 047/050 | Batch 100/469 | Cost: 12954.7383\n",
      "Epoch: 047/050 | Batch 150/469 | Cost: 12598.8203\n",
      "Epoch: 047/050 | Batch 200/469 | Cost: 12969.3066\n",
      "Epoch: 047/050 | Batch 250/469 | Cost: 12893.0693\n",
      "Epoch: 047/050 | Batch 300/469 | Cost: 12975.1309\n",
      "Epoch: 047/050 | Batch 350/469 | Cost: 13452.9775\n",
      "Epoch: 047/050 | Batch 400/469 | Cost: 13320.0781\n",
      "Epoch: 047/050 | Batch 450/469 | Cost: 13015.5547\n",
      "Epoch: 048/050 | Batch 000/469 | Cost: 12687.9102\n",
      "Epoch: 048/050 | Batch 050/469 | Cost: 12957.4678\n",
      "Epoch: 048/050 | Batch 100/469 | Cost: 13027.3281\n",
      "Epoch: 048/050 | Batch 150/469 | Cost: 13472.4619\n",
      "Epoch: 048/050 | Batch 200/469 | Cost: 12705.8525\n",
      "Epoch: 048/050 | Batch 250/469 | Cost: 13201.4590\n",
      "Epoch: 048/050 | Batch 300/469 | Cost: 13181.9707\n",
      "Epoch: 048/050 | Batch 350/469 | Cost: 13372.9746\n",
      "Epoch: 048/050 | Batch 400/469 | Cost: 12831.2305\n",
      "Epoch: 048/050 | Batch 450/469 | Cost: 12963.0156\n",
      "Epoch: 049/050 | Batch 000/469 | Cost: 12832.8057\n",
      "Epoch: 049/050 | Batch 050/469 | Cost: 12771.7109\n",
      "Epoch: 049/050 | Batch 100/469 | Cost: 13143.8457\n",
      "Epoch: 049/050 | Batch 150/469 | Cost: 12863.8740\n",
      "Epoch: 049/050 | Batch 200/469 | Cost: 12841.9248\n",
      "Epoch: 049/050 | Batch 250/469 | Cost: 13240.2529\n",
      "Epoch: 049/050 | Batch 300/469 | Cost: 12889.4521\n",
      "Epoch: 049/050 | Batch 350/469 | Cost: 13022.1709\n",
      "Epoch: 049/050 | Batch 400/469 | Cost: 12963.3125\n",
      "Epoch: 049/050 | Batch 450/469 | Cost: 12778.5381\n",
      "Epoch: 050/050 | Batch 000/469 | Cost: 12820.4805\n",
      "Epoch: 050/050 | Batch 050/469 | Cost: 12769.4893\n",
      "Epoch: 050/050 | Batch 100/469 | Cost: 12682.0566\n",
      "Epoch: 050/050 | Batch 150/469 | Cost: 13312.6172\n",
      "Epoch: 050/050 | Batch 200/469 | Cost: 13716.5430\n",
      "Epoch: 050/050 | Batch 250/469 | Cost: 12201.1973\n",
      "Epoch: 050/050 | Batch 300/469 | Cost: 13228.4199\n",
      "Epoch: 050/050 | Batch 350/469 | Cost: 12986.6201\n",
      "Epoch: 050/050 | Batch 400/469 | Cost: 13097.1562\n",
      "Epoch: 050/050 | Batch 450/469 | Cost: 13272.6777\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "for epoch in range(num_epochs):\n",
    "    for batch_idx, (features, targets) in enumerate(train_loader):\n",
    "        \n",
    "        # don't need labels, only the images (features)\n",
    "        features = features.view(-1, 28*28).to(device)\n",
    "\n",
    "        ### FORWARD AND BACK PROP\n",
    "        z_mean, z_log_var, encoded, decoded = model(features)\n",
    "\n",
    "        # cost = reconstruction loss + Kullback-Leibler divergence\n",
    "        kl_divergence = (0.5 * (z_mean**2 + \n",
    "                                torch.exp(z_log_var) - z_log_var - 1)).sum()\n",
    "        pixelwise_bce = F.binary_cross_entropy(decoded, features, reduction='sum')\n",
    "        cost = kl_divergence + pixelwise_bce\n",
    "        \n",
    "        optimizer.zero_grad()\n",
    "        cost.backward()\n",
    "        \n",
    "        ### UPDATE MODEL PARAMETERS\n",
    "        optimizer.step()\n",
    "        \n",
    "        ### LOGGING\n",
    "        if not batch_idx % 50:\n",
    "            print ('Epoch: %03d/%03d | Batch %03d/%03d | Cost: %.4f' \n",
    "                   %(epoch+1, num_epochs, batch_idx, \n",
    "                     len(train_loader), cost))\n",
    "            \n",
    "    print('Time elapsed: %.2f min' % ((time.time() - start_time)/60))\n",
    "    \n",
    "print('Total Training Time: %.2f min' % ((time.time() - start_time)/60))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reconstruction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x180 with 30 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "##########################\n",
    "### VISUALIZATION\n",
    "##########################\n",
    "\n",
    "n_images = 15\n",
    "image_width = 28\n",
    "\n",
    "fig, axes = plt.subplots(nrows=2, ncols=n_images, \n",
    "                         sharex=True, sharey=True, figsize=(20, 2.5))\n",
    "orig_images = features[:n_images]\n",
    "decoded_images = decoded[:n_images]\n",
    "\n",
    "for i in range(n_images):\n",
    "    for ax, img in zip(axes, [orig_images, decoded_images]):\n",
    "        curr_img = img[i].detach().to(torch.device('cpu'))\n",
    "        ax[i].imshow(curr_img.view((image_width, image_width)), cmap='binary')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate new images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlMAAABSCAYAAABwglFkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJztnXuUlXX1/1/P3JgLyCCj3CdGARERUxRN00hN0CyTlXjLlSZ5y0sX7dfK0MoyW2VZipWWZWoqLC9fzTQUNSwvgGZQgIoigqiAMMAIw9ye3x+H936eOTPAwJk555nDfq3FOsPMuTz7+VzOZ78/e+9PEIYhjuM4juM4zq5RkOsLcBzHcRzH6c74YspxHMdxHCcDfDHlOI7jOI6TAb6YchzHcRzHyQBfTDmO4ziO42SAL6Ycx3Ecx3EywBdTjuM4juM4GZDRYioIgolBELwWBMGSIAi+01kX5TiO4ziO010IdrVoZxAEhcDrwGeAFcBc4MwwDBd23uU5juM4juMkm0yUqXHAkjAM3wrDsAG4Dzilcy7LcRzHcRyne1CUwWsHActj/18BHJ7+pCAILgAuAKioqBg7cuTIDD4yd7z99tusWbMmaO9v+W5jvtgH8PLLL68Jw3Cv9N/ni427cz+F/LcxX+wDH4u4jd2C7dnYijAMd+kfcBrw+9j/zwFu3t5rxo4dG3ZXtl77Du9LvtvYne0LwzAE5oV5bKP3093Hxu5sXxj6WAzdxm5BR23MRJlaAQyJ/X8wsDKD93McJ2E0NjYCUFSUmiqCYPsOWpgWgxn/v35uamoCoLi4mIICTyh2nK5E466hoYG1a9cCsGLFCgA+9rGPAVBeXk7Pnj1zc4F5QiYz2VxgeBAENUEQlABnAI90zmU5juM4juN0D3ZZmQrDsCkIgkuBvwOFwB1hGP6v067McZyc0tLSQnFxMRCpSVKqAAoLC4FItWpoaKCurg6Ajz76CIDVq1fzn//8B4A5c+YAMHToUAAuvPBC9txzzy62wulMpHJs3rwZgE2bNrV5Tp8+faxvOLlj/fr1AGzYsAGAhx9+mGeffRaAYcOGAfDpT38agIMPPpiKigpgx+qz0z6ZbPMRhuHfgL910rU4juM4juN0OzJaTGWLMAzN4126dCkAS5YsAWD27NlUVVUBcMghhwAwYsQIBg4cCEBZWRnQPVbb8v4XLFgAwJNPPgnAhx9+SHl5OQCHH55KmDz00EPp06cPQF54gffffz8AZ511lv3ukksuAeDmm2/OyTXt7rS0tNjPim3q0aOHqRMaU3oMw5D6+noAli9PJfo+99xzPPLII61+N2jQIAAmTJjgylQ3YdWqVQBcfPHFAMydOxeA+vp6DjjgAADOP/98AE488USbm/IhJi4Mw27x/QGRcvjWW29x4403AvD4448DKaVKcVF63GeffQAoKSmhoaEBSI1xZ+fp/j3dcRzHcRwnhyRamWpubgbg3Xff5b777gPg0UcfBSJlqra21p6v1XZ1dTXjxo0D4PTTTwfgyCOPBFIr8CRSX1/PN7/5TQAeeughILKtubnZPA55euXl5fTv3x+Aq666CoDPf/7zAOy5557dyiNsaWnhvPPOs5+FYmx2B2S3vMO1a9eyZcsWAFNv9thjj6x6yIqFgtZZeemKlMZpU1OTxdK89dZbADz77LMsXrwYiOJrZOOGDRvaqFxJp6WlxdQ3qTW1tbX283vvvdfqceTIkYwdOxaI2rGkpMTU5HS7k3QfpJQ/9thj3HDDDQAsWrQIiGLiCgoKbJ7ae++9AejVq1fO5p/4/JGO+lpLS4vZpv66efPmNvFgijUqKSlhwIABQGoMQioTNUltJd5++20AJk+ezJtvvglE47Nnz57st99+ABx11FGtXtfc3GzPS9KYbG5utnZQn4rvNuka9beCgoI2GcWiq+1J5GJKA+Kee+4BYNq0aSxcmDqlRhOZBkMQBLZA0t/efPNN61T//ve/gUiCPvPMMxOVAqpJadq0afzpT38CosHcHuoQdXV1vPvuuwD8/Oc/B+Bf//oXkJLZJ0yYAEQLzCQMjG3R1NRkbRfnO99J3nGP25todrQdoNemP0LU5n//+98BuO222+zL99vf/jYABx10UM7bcUf2aSwq5XrfffflxRdfBKJFlILYCwsLc27PjoinlUNqAfXEE08AcNdddwGpNHP9XY9aiFZWVjJ+/HgAvvCFLwAwatQo+vbtC2AB/lpcJWFLSYv4yZMnA6kFsdqstLQUgMGDBwNQVVVlW/OHHnooENmUTbQQ0P3fsmWL/U5toTG2fv16W+xru3Lp0qVmtxaH6sNjxozhsMMOA2D//fcHcmPj9pCtd999NwArV660+eO4444D4PLLL2fIkFQ1I7VjfNEbX2xCbsNH1N9eeukl7rzzTgD22itV31VtMWjQIBMUtDVZXFxsdqg91SdaWlps3PXq1Qvo3G3o7iNfOI7jOI7jJJBEKlMffPABANdddx0A77zzjilRUlokKY8ZM8a8JK22V61axRtvvAFE2w233HILkPIUtR2Wy0A7rZ4VZH7DDTdsU5EqKytrJWNCytvS9eu9FJz/4IMPWnG2SZMmASm749s2SaKxsbGNNFtRUWGefJJoT1WSJxcEQZvt2O29NggC69f//e9/gaifrly50oJ4tbWQS8WiI59dWlpqiRLaDnrxxRfNQxRKwa6pqenkq+w81Eaai2677TYAZs2aZQki2rZsaWlptf0OkecbhiGvvvoqELVjv3797GeNye31m2zS0NDAxIkTASyNHjAFQAHoI0aMAFK7AaNGjQIi27PdT8MwtHEkpX/58uVWmFIq0po1a4DU2HruueeAaNyVlZWx7777AtF41nbS0KFD7XtH7ZNr9TAdJXfMnDkTSI3Fc889F8DCR8rLy7d53S0tLe0q57loS4jsmTFjBv/85z+BaL448MADgVS7pCtMLS0tNt9oe/B//0tVbFq5cqWpjVLrtndPdhZXphzHcRzHcTIgcVJFfX09P/nJT4DIK2xqajL16fLLLwfgjDPOAFJenvZ25VE0NjayevVqAKZOnQqk9l4Bbr/9dgvCGzNmTJfbsy10rX/+858BWLduXZvnVFZWAjBx4kQrsqZV99KlS3nnnXcAzFYFva5bt44XXngBwNLSr7nmGttrTppXJa8yTmVlZeKuE6L993giQzwge1vKQly1iscmKLbv+uuvB6IA0n79+lksirypJN4PiLzCMAytUKBivxYvXmzxD/L0lQyS1LIIYRhaPOI3vvENAP7xj38ArYuWSompqKiwQqQqzyIveuPGjTYupe60F6eiMZCrBBnZ9dOf/tSUADFgwAAefvhhAFOhpHy/9tprNk91tarWkcBoqU8zZsyw+65YG13nxo0b7b00tiZMmGBxpunxRL179zZlKmkJTFJHf/GLXwBRYtapp57Kt771LaBjimG87XI5z6THbb3//vtUV1cDUczhiSeeCKSU3vjcI9RGeo/58+cDqZhilVhSHGNn2pqYxZQMX7x4MU8//TQQDfC+ffvyve99D4BzzjkHiDp8HC2qCgsLrZbNZZddBmBVmBcsWGATo+qj5CLQTl/KChqPow7yiU98AoAvf/nLZq8GT//+/S1QUovOlStX2usV0K3J5corr+Tee+8FsBpcSflyluQeR5N2UlD/1JdefIs4PbttR6hfz5kzhyuvvBKA119/HYgCe7///e9zwgknALnf+tkR8aw+bQ3pcfPmzbaVpXpwX/nKV4Dk1rNZv369LW7/+te/AlH79+3b1wKtR48eDaQSA+SgpddX+uCDD1qNS0i1f3rChe5RUVFRVttb/Vlfxtdff721pxaGDz30kM2nQtuUAwYMsKDerp5HtzW+wjC0MfXaa68B8PTTT9vvZIf6X3V1tf2s74DDDz/ctp+F2jx+EkCSxmIYhpYpqyx3fa/ss88+bRZ+7WXkbuv/uUaL1yOOOMLmRFVr7927N9D6muM/x7f8IDqHsKCgwBbPujeduZWZnJ7hOI7jOI7TDUmMMiXFZc2aNRZUptoeRx55JJ/97GeBnfdmteqUilNbW2slF6RySf7NJtqiay/oXLKsPL76+nrzbnWf5syZY9t78UBYSHkn8jjlqbz00kv86le/AuAHP/gBEG275JoHH3ywze/k4ScNpSDDjssExB+3bNlicvOPf/xjIOU9K2BWnq8UuWOOOSZRXnBHWLdunSkcktMhGl8XXHABECkFSUNjZv78+cyYMQOIyq1oi27KlCmceuqpQNRHe/ToYaqGPF6Nu82bN9vWi7bGPvroI3uexrref/PmzW0Ukq5A/Vh2qo5UGIYWgK4QhHh9M/VnKeV9+vTZbm2nrqS9QGnZFQ//UE0lKYlBEFifVCJTaWlpm9pf8bCRJNVeEo2Njfzud78Dot2J+GkguWqXzkDK1JgxY+w7sqNlftQHFMSuMXbssceaEql709zc3GmJWd1rtnYcx3Ecx0kYiVGmpDgNHDjQ0ha1/33ggQd2qNCmvIfGxkZLh1QQngJKGxsbWbZsGRCdqp1tZSoMQ1ttxwNahTxkxTvNnDmT999/H2gdHKm4BQVayrOKB3THvSvtqytlNilxSfEUbKFg+aSg+yjVtD3vKJ6WK5sU//fiiy8yb948gHYLlKrtkmZ3R1Af/vWvf23lOcSQIUMsLuzjH/84EKmyLS0tdj81/nNZvkNt/Morr5iyJgVJ8RpnnHGGqUiiR48ebcaeqk8vWrTISilIfVy7dq2pT0rzlrqsMd2VhGFopSsUYC97R48ezR/+8Aeg/diUdHr27GkKd7aV1Hj1a91bzY+9evWyyvOKZ9M9Ly0ttX6n/hYvIBtXLSDVpklR8ePU19dbUWqhxIdBgwZts8p+klEfiseoqW+mnx24rcLJGksag0p02Wuvvey1KptQVFRka4vtvW+Hrn2XXuU4juM4juMACVKmtIqurKw0xUSeEUTFNxXZr1iThQsX2gpUitP8+fMte08lB+IKkBSsXGUThWHYJvagvr6+zapc3tDo0aM5+eSTgWhPvHfv3jz//PNAVP5AXsq6devalBsIw9AUEXlo8rxyeWzAtj5fdiYFqZjycuKZMfGiqdOmTQPg97//PdD6CJX0GIaioiJrC8UHbty4EUil4h999NFA1F5J8zDVx3SW5B133GF9SurNaaedZrFFGqc6GmrNmjX2vJNOOglIZcZpTEg1yJbioftbX1/fpvjmKaecAqS8W12XbG1sbDQVWTbqCJ26ujqbg5TiXV1dbVlF6k/67Gyk3m/atImf/exnQNSvVT7gnnvusbHXkZjAePZhLvtnetHUQYMGsc8++wDRzoP61UcffdTG7r59+9r4lG3x+TGJ8YuFhYX2vSZ1RZna7WW7dyfiJQ+kNOn7a3uxU01NTVYIW2VmtJ6orKy0uMX4TpXmIJUfUh/a2TZPzGJKN6ZXr15W/0mLnSVLlnDzzTcDreVzSEnm6vTtpUfqizqeLqmBpBuaLtt3NfEUTQ14VeIFLA357LPPBlJbDOm1Qurq6kzS1WShgPXm5uZ2U68VbJm0L+WjjjrKOr6+oGfNmmXbkUkg/UsvHoiuQMdbb73VFhZCX45VVVVWef/qq68GUu2mtF29h4J+n3jiCQuWvPbaa4GUI5HrhS9EC8Tp06cD0UkFdXV1Vm/pS1/6EpD64lLgvQKxtfCoq6uzxYq2vZubm21SU3+Vc5GtfltfX2/XpTGmBWF9fb3Zr4Xv66+/buNX85PmrurqajsPTY+jR4+2sZo+ruPV1Dsb9dk777yTxx9/HIicFp3vud9++21zCxuidpIzu8cee+S8T8avN169XM54PA0eUklIWuDqu2Dp0qVmh54vZ1aL4KRRUlJitqsvah5dsGCB9dn4wcBafKl/JnGRCK0Xx+kHUOs7It7v9LdHH32U7373u0DUbvrbiBEj7DQChRBNmDChjfPg23yO4ziO4zg5IHHKVEVFhak18kiLiopsJakUUHlGQRC0CiKE1Eq8X79+QHSS+axZs4BUsLZWtireecghh2RdrdHWjarurlixwlbgkmoVnBpPTZYHsnDhQqtUrHui+1VeXt5Gei8tLeXMM88EIiUu1x6lOOCAA9psSz7wwAPccccdQDJOaG8vXTq935WXl3PMMccAkXerApWHHHJIu9vKCo4dPnw4EClUs2fPtnIY2oq49tprW3mZuWDLli12fqDUYnl+Y8eO5aqrrgIi+++//35TkRX0KRt69epl/fmVV14BUqqsnnf88ccD2Wt/3dOBAwfaNWqe0QkKa9eutXZRwd333nvPgmQ1rrVtOXz4cPbff38gUo579+7dRjHPBioPc9NNN9l913l048aNA1LqjRQszUcNDQ22PSIlVcp6UVFRVvpie8UV2ytXoHmktLTU+mB6gHFlZaX12fjWrraTlDCgNj///POtrEKSKCoqMrVTiosep0yZYsq2iv/W1NSYjTqzNRsJD7uC2mXPPfdsVbkeWidYKZznj3/8I5BK/NFY1BhWOMyqVavstVKcx40bZz9nOs+4MuU4juM4jpMBiVGmREFBga0oFTvRr18/W4GreKc8pS1btpgKpb3PwYMHmxeix9/+9rdAqjidPC8FbDc1NWVd/dB1aX+3qanJYkvkyWrPv6WlxTyK+++/H0jF1EitU3yU9sPr6+vbpJgPHTrUgtiToPTEUaB1nM2bN5t9SfIK48kL8p4U43b11Veb0rCzwdN6nhSNe++91+xXcHouzgWLlxsB+M1vfmPFX+Xl6ayss846y9QXFYNcsmSJ9fWDDz641XuWlZVZCRPFrixevNjup1TAbKlwaoPKykr7bCW13H777UDqPii4Nx4/pzlFio283WHDhpmyGI9TyUXa+htvvAGkzjuTfSrloVivN954wzx5tcnee+9tz9P1qnBpthJFgiCw/pYe9wNtk2kGDx7cRu1WfFphYaHNrfHCn3pfqZFq81tuucWK0WajoGpHCYLA1OFnnnkGiO7Nhx9+aO0o1be4uNjiqLQ7oYKmPXr0aPO9kISEgn79+pmiqhJAuq758+dz3333AZGaWF9fb2NPxYFHjhwJpOZWtbtiIisrKzst0SVxiyloXT8EUlLkQQcdBGCP7T2/vQAyDZYpU6YAKRlQhwCrJk5DQ0POFhganOeddx533XUXEDW0giTDMGTmzJlAdIZWbW1tm6289C8+iILTjz76aAvoTRpaLKSjyT9Ji6n2tkbVDp2RQaPJbvTo0eZUaIEVr4XTlcQr6Wsi05brzTffbDK66kbpDMmWlhYLwFdGTUNDgx1srEWzXj937lybIEVxcbFt82e75pTubVVVlTkvsl9B5xC1kexvamqy12rBqEVVcXGxbSmp7zQ3N2d1i13zgrZECgoKrK9q/rn11lsBeOGFF2zeiQdi6/lKxFAowpAhQ9rUbOoq9P7xtoDWW4DxQ5e1iNU4im+ttjeO9P4jRowAoi2wGTNmcMQRRwCpc1L1HklAQdYKY1EiwTPPPGP9Lp40oJ+vueYaIKpvVlBQwLHHHgvA17/+dSDVxrk+2LmiosL6mrb5FNayYcMGqyWmPrhp0yY7xFgHI2u7s7S0tM3aojPn02T0CMdxHMdxnG5KIpWpdIIg2Gba4vZOwo7/Tl5GWVmZ/U4KzoYNG3Im3+paevbsyXPPPQdEK295iLW1tRYsr0roYRi28Y7isra8ZwUfXnDBBRlXeO0qtuX9JDU4sitRn4yXUlC5gWy1m5SpNWvW8MADDwDwy1/+EkhtH6i9pOZqi2j69OlWD059saqqypRH2TZnzhwgtc2u8avnjBs3zjzJXNWB69mzZ6saUnFKS0tNkZo8eTKQCsqWwqaAbr1+/fr1tkWm/hyvzZQNdI81x40cOdIUJnnvUunffPNNa1fd/9raWksK0HtJ3R87dqwpOekKUFeh7ToRhmGr7T1IzaGyI/4dAdseR/EkKIiU5vXr19vpEUriSVodJymiX/va1wBYvXq1BaPHFVHdEyW6fPjhh0BqvCpcRHWZTjjhBFO+0vtrttTVoqIiCx1QOI92WMaNG2dqleaUV155xZQs9cts1elzZcpxHMdxHCcDdqhMBUEwBPgz0B9oAW4Lw/BXQRDsCdwPDAXeBiaHYbgu0wsKw7BVzAZsv8puvKq0nlNQUNCmiq0CSVevXm1ehZ6jfeNccuONN9rqWt6w1ICCgoI21bPbU+v02L9/f04//XQAexw2bFhiSiGkI889TklJiQXiJwHd/65SFNTXFddy8MEHW5xGtuMW1P/ef/99Hn74YSBK+IiPT6GClWvWrLGxFU8AUTxHXGmF1D2VkqCz7yZNmtTh0+G7inXr1lk7697LnvHjx1tBUsXnrF271lQZ3Sfdh0GDBrVRerI9DjXP6XP3339/izX51Kc+BURlWI477jh7voqsPvroo1beQjE3Usjr6uqsv6QrRtkiCAJrL6kQVVVVFkiefvLDjlDslMbfrFmzTHmUrUlTptILjq5fv96+19T/+vbtawqwdi60C7Jp0yYrj6HA7ZKSki6f9zqCrj9ejkOPUtZWrVoFpL7nv/jFLwLbr5TeFXTkDjUB3wrDcH/gCOBrQRCMAr4DzArDcDgwa+v/HcdxHMdxdit2qEyFYfge8N7WnzcGQbAIGAScAozf+rQ7gWeB/7erFxJP1Y2XxIdUrIE82PQ0xvXr19uqXKvuXr16mfc7d+5cAEuhXLVqla1YFZuRy1IB8nSmT5/eJkslPbU3Ha24009MP+iggzjrrLOAKC20uLg4cbFSQnv7cXr37m0ZGrlEXrraorNVInn6Tz31FBB5/CeeeGLOYoY0tpYtW2bjSH1sy5Yt9vfVq1cDkQeov0PkKW7evLnNOZRKTx4zZgznn38+AJ/5zGeA3Kkb0LpIpdpZHrwyEi+++GIrtCrPf9GiReYhax6Lp15r7srVER5S53Xfy8rK7D6rvZT5NmTIELNd6fPLli2zOTk9y7GioqLNkTi5IP1sw9raWoun0X1PV1S3hdpHaunMmTMTmwktZL++QxYuXGhtG88Kvvjii4GobdVP2yt+3V6hVJFNdVXXpWuNl7OQEqzv+YKCAov5ynY28E59WhAEQ4GDgZeAflsXWoRh+F4QBBn1Ni0qli9fbmUAtACqra01ST1eGRVSnUc3VxNCRUWFbT2oVoqk3qKiIpsMFZyt984FmpyULr4zpAdMSqqfOnVqzjrUrtDeNW7cuDER6cfqN6pJJim8oKBgu9eXHvQa/7++3Gpra01m1z1QBfWBAwfm7MtJE+UxxxxjadLacl62bJndE33B6trDMLQJPH5+lsaX+qQC6ydNmmTBoknqp5WVlRa8qvM7dVLBqFGj7MtZ1cAXL15shzfrxAWNxZqaGltE5qo/697K2dyyZYt9+agNtZitrq62La3Zs2cD8PLLL1sAuhbOCh/Yb7/9bAGTBGdN11JTU2Pto7MgtZjckWOZnhRx+OGHtzl5IwmOXntoi65///7Wtlo4TZ061YK4O9IXk9Ce7REvBfTOO+8AkfByzjnndOig7q6gw6M7CIKewAPA18Mw3LATr7sgCIJ5QRDMkyebb+S7jfluH7iN+UK+25jv9oHbmC/sDjbG6ZA7GARBMamF1D1hGD649dcfBEEwYKsqNQBY1d5rwzC8DbgN4NBDDw3bew5EEmxdXZ1Jd5LOly9f3ua8qHRJEiKvpLy83J4vr1BVUUeMGGGSrSotZ7qd0lEb20OeTmlpqcnSHSEIAvMSlap9/fXX2/87U4bNxL6O0N5AKykpyeo217ZsVMr43XffDUT9rrq6upWnu/U9rB9LvZHsvnHjRubNmwdE/bqsrMwCgKVIdaWK0dF2lI1VVVWcccYZQDRWnnzySUuUUICybGxoaGh1TiGkvHsVAzz33HOB6DzC8vLyTvceM+mrupZhw4aZAqnz2VRA9tVXX7V55vHHH7dHvVbKlFS4zrZxV+xTX9IpEjU1NZYUoLZUPy8vL7fK+ypzUVdXZ4rG2WefDURtWVFRkag2FGVlZa3KJEA010KkLLU3zvQdo63qJUuW2By1o9CLjtJVc6p2Ka666ipLHpk2bRqQKi2Q5ZIcXfq90dDQwPTp01v9buLEibkLj9jRE4LUSPkDsCgMw1/E/vQI8OWtP38Z+L/OvzzHcRzHcZxk0xFl6ijgHGBBEASKFP4ucAMwPQiC84F3gNMyuRB599XV1Rx33HFAFIz71FNP2b63vGB5h/EU1bgKpSMpVMxMAaQFBQXmZcZLJLR3Ank2UDzJ5MmT7Syo9ED0OLq+Xr16cdlllwHw1a9+FYg8zyTEGu0M8+fPb3MkTllZmXmBuYyn0XWpMOUVV1wBpOJOFHN3/PHHA6n+JG9WsTZSqFasWGHtqiKc48ePN8VH/T9JcQpBEJiXJ3Xpk5/8pAUfKyZFJQ9eeOEF+5vO1TzssMNaqTR63ySi6+rTp48pUzrXTEUbn3zySVMfFa/R3Nxs84zOKcx1eYf2kJI6fvx4s0v9VEWBGxsbLSlCY3HIkCFcffXVQBTvlvS2hGjHQaqa4nI3btzYJig/Xn5GZXSuu+46AB577DErJZGks/naQ+1x0UUXceGFFwLZL8XR1cTLHamtNP5yefRYR7L5/glsa8Qc12kXsvULs0+fPrZtpcXBlClTbEtBAdv6EqqqqmoTYDh48GD7WYsuvX9jY6NlN+hvuVx8aEF3xRVX2AQXP4AZUteua1ZF2osuusgy9na2UniuFo7bYt9997X7oMeTTjopZ3JtHCU6aDLV2XObNm2yBInHHnsMSAWNa8JWFqUWEieffLL9TpN7UVHRNiv7J434+YMKaNWj0KIyTtLtao8ePXrYeYPa5tPYXLlypbWxvqT23ntvLrnkEiCqwJ1Eu7VgHzduHFOnTgWwg6YVQF9bW2t9Xn33yCOPbJNN3R1IPzMznrUtJ0eOQH19Pc8++ywAP/zhD4Foa7dPnz6WgNBdTmXobg51R9BiV3311ltvZeXKlQCcdlpKy8llNnD+3XHHcRzHcZwskjg3o7Cw0Fb/eqypqeGwww7rlPcvKiqytGcRryaebeRBDBs2jAcfTMX2P//88wDmKQ0ZMsQ8Zal1AwYM2OVbofv4AAAEsklEQVSaR0nzmi+99FIuvfTSXF9GuyhZ4corrwSik8tnz55tW1oqa/H666+bB6+Uf21ZDxgwwLZ+dtW7317dlySQ5GvbGYIgsO33c845B4i2FubNm2fquEIJJk+ebO3cHZSbkpIS2xbRYz6jfiklsWfPntZOUjYWLFjA3/72t1bPV9LS8ccfz+c+9zkgPxWf7oLm3r/85S9AtPUO0ZmYudzS9J7hOI7jOI6TAcl3o7qAJHoXQRBYDMqkSZNaPTq5I71I7E033WR/U1q5glKHDh1qsW3ykDqzr+WL8tMdUPsNHz4cgB/96EdASqFS7IbaI352qJNc4vGJiq1RokRNTY2VKVESiUop1NTU2C6Jj8HcoRhaFSZ94oknbO7VblMu28dnAMdxHMdxnAzYLZUpx9lV4nvyimNz8pf0bEtXoPKTwsJCU5/16CQLZWUqc0+PScFnBsdxHMdxnAzwxZTjOI7jOE4GBOmn23fphwXBauAjYE3WPnTXqaL1dX4sDMO9dvSiIAg2Aq912VV1LjttYzdvQ8h/GzvaT3cHG30sJgcfi9tgN7Exr8ciZHkxBRAEwbwwDA/N6ofuArt6nd3FPsh/GzO5TrcxOeR7P4X8t9H7ade9Npvkez+FXb9W3+ZzHMdxHMfJAF9MOY7jOI7jZEAuFlO35eAzd4Vdvc7uYh/kv42ZXKfbmBzyvZ9C/tvo/bTrXptN8r2fwi5ea9ZjphzHcRzHcfIJ3+ZzHMdxHMfJgKwtpoIgmBgEwWtBECwJguA72frcHREEwZAgCJ4JgmBREAT/C4Lgiq2//34QBO8GQfDq1n8ndeC93MYc0Vk2JtU+yH8bvZ+6jWnvk9f2bX2N25gjOtNGAMIw7PJ/QCHwJrAPUAL8BxiVjc/uwLUNAA7Z+nMv4HVgFPB94Eq3cfexMcn27Q42ej91G3cX+9zG/LFR/7KlTI0DloRh+FYYhg3AfcApWfrs7RKG4XthGL6y9eeNwCJg0C68lduYQzrJxsTaB/lvo/fTnSLfbcx3+8BtzCmdaCOQvW2+QcDy2P9XkMFFdxVBEAwFDgZe2vqrS4MgmB8EwR1BEPTZwcvdxoSQgY3dwj7Ifxu9n+72Nua7feA2JoYMbQSyt5gK2vldotIIgyDoCTwAfD0Mww3Ab4B9gY8D7wE37ugt2vmd25hlMrQx8fZB/tvo/dRtJP/tA7cxEXSCjUD2FlMrgCGx/w8GVmbps3dIEATFpG7mPWEYPggQhuEHYRg2h2HYAtxOSq7cHm5jjukEGxNtH+S/jd5P3cat5Lt94DbmnE6yEcjeYmouMDwIgpogCEqAM4BHsvTZ2yUIggD4A7AoDMNfxH4/IPa0U4H/7uCt3MYc0kk2JtY+yH8bvZ8abmP+2wduY07pRBtT7GzE+q7+A04iFS3/JnB1tj63A9f1SVKy43zg1a3/TgLuAhZs/f0jwAC3Mf9tTKp9u4ON3k/dxt3JPrcxf2wMw9AroDuO4ziO42SCV0B3HMdxHMfJAF9MOY7jOI7jZIAvphzHcRzHcTLAF1OO4ziO4zgZ4Ispx3Ecx3GcDPDFlOM4juM4Tgb4YspxHMdxHCcDfDHlOI7jOI6TAf8f7k11/UF20SUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlMAAABSCAYAAABwglFkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJztnXl8nFX1/983Jd1T6N5AkVbAbtBiqRRpKbKKhRcgO7KoLGWrSBEUZJFVFBBFQbE/NkFQsF9AEJGloCgQKEUotVCghdLSJQ3dl2SSyfP7Y/o5M1lo0kySeWZ63q9XX0kns9wz9z73uedzzj03RFGE4ziO4ziO0zKKct0Ax3Ecx3GcfMYXU47jOI7jOFngiynHcRzHcZws8MWU4ziO4zhOFvhiynEcx3EcJwt8MeU4juM4jpMFvphyHMdxHMfJgqwWUyGEQ0MIc0MIH4YQLm2tRjmO4ziO4+QLoaVFO0MIHYD3gYOBRcAM4KQoiua0XvMcx3Ecx3HiTTbK1F7Ah1EUzY+iKAH8GTiydZrlOI7jOI6TH2yTxWt3ABZm/H8RMLb+k0IIk4BJAN26ddtz6NChWXxk7vj444+pqKgIjf2t0G0sFPsAZs6cWRFFUd/6jxeKjVvzOIXCt7FQ7AO/FnEb84LN2ViHKIpa9A84Drgr4/+nAr/Z3Gv23HPPKF/Z1PYmv5dCtzGf7YuiKALeiArYRh+nW4+N+WxfFPm1GLmNeUFzbcwmzLcI2DHj/wOBxVm8X4tYuXIl48ePZ/z48Tz33HM899xz1NbWtnczHMdxHMfZSslmMTUD2DWEMDiE0BE4EXiidZrlOI7jOI6TH7Q4ZyqKopoQwmTgGaADcE8URf9rtZY1k6VLl/KFL3wBgEWLFgGwYsUK+vTp095NiRXV1dUArFu3DoAePXrQoUOHXDbJcRzHcQqSbBLQiaLo78DfW6ktjuM4juM4eUdWi6k4UF1dzRe/+EUA+vZNbQzp0aNHLpvUbtTU1ADw9ttvA/Dhhx8C8Nprr9lzevXqBcDYsWMZNGgQADvumEp169y5c3s1tUVUVVUBacVRuXArVqyga9euAGiHSHFxcQ5a2HyiKCKZTNrvACEEUwtDaHqzSD4hG5ctW8ZVV10FwDPPPAOk8hwhNTYffvhhAPbaay+g8L6HfEbXm8YtpOccPdatW7fN9lnmWC8EZHcIgaIiP0DESeOjwXEcx3EcJwvyVpmSh3DnnXfywgsvALDddtsBcOihh+asXW2NvMUlS5aYp//SSy8BsGbNGiCVJyVPUM+vra2lf//+AHTs2LFd29wcNmzYAMADDzwAwH333cfSpUuBtHcrlW3t2rUMHDgQgO9973tAqs+7dOkCxMMLlqr2+uuvA3D//febYrhx40YABgwYwKhRowA44ogjANh///2B+CttjZFMJlmxYgUADz30EAB33HEHixenNvmqj8WIESMYNmwYEI8+c1J9+NFHHwFw8803AzBjxgzLvZQypTnk5JNP5uSTT67ztxUrVvDGG28A6bk4M2qQL32te8yDDz7IFVdcAaTmHoAhQ4bw6KOPArD99tvnpoEtQHOpyJe+aIr6dn3e39SnmfdFKYyKEuj/W6o85u1iav78+QA88cQTdhEr9FOI8qtsnDZtGgC33Xab3ZT32GMPAI4//nggJb3rxj137lwglai/fv36dm1zc6moqLBF0dNPPw3ATjvtxLe//W0Avva1r9V5/vTp0/n3v/8NwD/+8Q8A9txzT9uIkEt007nvvvsAuOGGG4CUjdtsk7rcFF5ds2YN77//PgBvvvkmALvssguAha7jjCap5cuXA/Dss88ydepUAGbNmgWkxq0WhoMHDwbgyiuvBFI34nxcNDaGvgv9zCzPkg+hXLV3wYIFXHzxxQA8//zzQKoP5ajIYU0kEgDcfvvt/O53vwOw+aW2ttZsveeeewC4+uqrATjssMNivxFGm3cyndPy8nIg7STNnj07N43bAmTHO++8A8DUqVOtX5QeoXtGv3792q1fmhP6zbx+6oeb9TORSNjiVqkgGzZssOevXr0aSM9Py5cv5+OPP67Thh133JEDDzwQSM9PEh2Ki4u3aC1ReKsOx3Ecx3GcdiTvlCmtSv/0pz8BKe9BipQ8/jh7gC0hiiJTYH70ox8BqVDRCSecAKQ9fSXel5eXWzL6smXLADjwwAPZeeedgfh8PwoJHXHEERYW6NevHwAPP/wwQ4YMAdLtVd8PHDiQV155BcB+lpeXs9NOO7Vf4xth48aNFg6QMqXQ1m677cYZZ5wBYKG9xYsX86tf/QqAefPmAekk7XPOOSc2/VQf2fTyyy8DmDKxdOlSVq1aBUBpaSkAgwYN4utf/zqQDmXKA4yrfU0hr1bfQ3l5Oe+99x4ATz75JIB9DyUlJRx77LEAjB8/HsBUnjggxfutt94C4PTTTzc1W38D2HbbbYF0eE9KzcaNGxsoB0VFRXTq1AlIqeSAjYG4RA3U5sZUEik0UuEmTJjQaCFozVVxQorhiy++2EAlrq2ttXuENi2VlJQAcOqpp7abMrW5615jTgr/4sWLLVVCqpLSBhYsWGAh6crKSiClxul3vZe+k8yxp/E8btw4vvKVrwDQs2dPoOVKcjxGtuM4juM4Tp6Sd8qU4qB//etfgdTqUQU6x40bl7N2tSXr16/nzjvvBNKr5nvuuYeJEycC6RW3vKySkhLLHxo+fDgAw4YNM08r14qA8g6uu+46IFXSYYcddgCgrKwMSMetM5HtXbt25YMPPgDSCaGKi+cCeeQ33XSTbfUX5557LgDXXHONeYX6/isrK/nXv/4FYMrcI488AsAZZ5wRy40CyWSS//0vVZtXY1Iq6B577ME+++wDpD3FLl26MGHCBADbNJDr8ZcNyWTS8vV+/etfAykVSuNQHrVYt26dqcPKgxs8eHBOFZpkMmn989hjjwHpjRyaXyHdTz169DBV7dNPPwXgk08+AVJev3IBpXL07NnTcv8uueSSOp+dy77Xdbpx40YqKiqA9KaWkpKSBm2TGjVr1izLPxJ77bWX2R0HpL78/Oc/B1IbXtTmk046CUhFJ5Rnq40+UknjoBgmk0kWLFgAwKuvvgrALbfcYoqU0LyYqUJlKoz6u74TfQ9RFJliKuV8v/3240tf+hKQzrluqUIXn9HQBPqyJEdL6uvfvz+TJ08GiEUCclswY8YMG2SydeLEiQ0uAH1HRUVF7LrrrkB6p8nIkSNzfvGrfVoI/fe//wWgU6dOtvursUVU/deXlZVZiFCTgW7iuUDJ/g899JAl4X7rW98C4Gc/+xnQeGgnsz8kSStctGrVqliGEaqqqmz3bGb/QWpiGj16NJAO4VZVVdG7d28gP3coCk3IU6ZMsRuRbs5HHXWUhTDrh82WLFliY0H13XJ149L1s3jxYtvwcOmllwJ1F1G6majfTjvtNFskqu9le4cOHSxkcuqppwJw8MEHW4Kz3iOXjoGcN807r7/+ui1sdVPNXEjpe9Ju4h/84Af2N/XdX/7ylzZudfPQgk9pBXfddReQuiZ/+9vfAnVDrApN33HHHUB6k4ES0XNJTU2NLaJ+85vfADBnzhy7puSMDhgwwF6jvlW/dO7c2WxU/2WGqGXnd77zHSDl4LXWxrXcL0cdx3Ecx3HymLxRpiTZqTSAPJ1jjz3WErGVgF5bW2srdnkcLa0dkUvk/d18883mGR533HFA43ZkJklqFa9k3y5duuTcdnm/qs8iqf2ss85i7NixTb5+4cKFAFx77bXmkRx88MFAOtG1PZEHJC91yZIlFsranCIlQggNQq/ywpYtWxZLZWrBggWWcK7+23fffYFUlX15jd27dwdS3qHUiTiG96IoatYWbfXn73//e7uOpkyZAqTUHdmra1b1wj777DPb+JHrsK3U/EsuucTCy/LeRQjBlEaFP6qrq03JkkKn1IpDDz2UM888E0hXsc8Mk+Sqz6MoMnVUm1S0aQngvPPOA9KqaiYKHUlRLC8vNztUpkXhwVyjjUkKOX/5y18GUmkFI0eOBOr2QWYJDEiXf0gkEjnfGJFMJm0jjsZqZmhO86E2GvXu3duuO93vNmzYwJIlS4D0WFXU4qCDDrK0l0yVvLXGaP6sLBzHcRzHcWJI3ihT//nPf4C0dyHv8IQTTrCYqEgkEuZxzZgxA4CZM2cCqWJlqtiba0+xKVTA8tVXX+Wwww4DGq+2qxi/1JqamhrznOJSyDSKIktU/sMf/gCkvduLLrpos+3TNvMjjzwSSOUVKddBSey5yMeRByuvPYoi83iVjLs5EomEJVeq/XrdihUrTI2NwziV4nLvvffy2Wef1XlMalRtba1tBNBzhg4dGusijZ/nleqaUqkKFZ0sLi7mrLPOqvNYY+qG6Nev32b/3h6on1RC5emnn7aE+fpkVouWZz9v3jxTYVUgWAnrRx11VM7ta4yKigrLC3r88ceB9Nb3U045xVS0xpDyqo0GURTZBhnNXXFQWRcuXMjll18OpOcPlc4ZOXJko23UfVHKlBSgOJSWKS4u5phjjgHSG3LeeOMNU+s1ZnUfWbVqlc09us/16tXLcqfVxyNGjABSpS7ast9cmXIcx3Ecx8mCvFCmamtruf7664F0jFexz6FDhzZQNRKJBNOnTwfSuxa0SyqEYLk73//+9+2xOKHtq3/+85+BlBd41VVXAY1v29TKXd9N586dzVuMiyqQSCSsnIXaK89eXkV9lCd2yimnANiW/D59+tiuFZ3tlgvkxUtBGjFihCmIm1PalLfwyiuv2I445SsoB2DFihW2DV25Ap07d855f1ZUVJhtUtF0Pd1zzz2mjsqe448/3vo3F3ltLUV9es011wDpPistLbWdQKK6utr+LhVVqmW/fv1yPr8od0gKquaXxigqKrK+027TcePG2dFOyj/RmMz1eKyPlNEbb7zR7gHK/dJ8P3LkyEbVXh1RJrVHfdqzZ09TqZQTmUukNE6bNs3Uw1/84hdAWo35vDH3z3/+s857xElV7NChg+1CV4mHe++914p26sgYKVTLli0zZV/39zFjxlgOtXLGNE+19XWYF4up9evXM2fOHCA9wL/5zW8CdUMgmWeFaYKXrKnJDdKLlPPPPx+Iz5ZttVGVa/fee28AJk2a1GhZA11IGlyyo3PnzrEIDWWydOlSnnvuOSCdFK/E3JqaGrNP/bt48WJ+/OMfA+kJQInMl112Gfvttx+Q28lci0KFK4cPH26hhMbQ+FRC7CWXXGIThOzXguPDDz+0/lTIbODAgXYTa++wrfpl9913t/bLHiWNvvrqq/aYbsiPPfaYlQRQmFaLEfVnHJG9ujlrQt51112tbpQIIVgyvkoi6IDfgQMH5nwxpQX62WefDaQO7q2/oJJ9Xbt2ZcyYMQBW5qK0tNRCQBrrcVtEqSSJkuHfeustS3W44IILAKzSdeZ8r/G6Zs0aLrroIqDuvQJSpQUU5osDcppnzJhh9b+0cNhcv9TW1pqDJmdBjo7GSC7JrJyvBfBFF11kbVZpi6eeegpIpf7oPp9ZWkb3xfYeox7mcxzHcRzHyYK8UKZmz55tK1B5FVJtMtFqe8OGDaZ6fPWrXwXSydyVlZWm2mQmW+aayspKjj76aCAtWd5///1A48pZMpm0pFB50UrE79ChQ8694fp8/PHH1iaVA5CdM2fONBldGwVWrlxpipvUGhXPO/PMM2MhT8vzkVe//fbbm431x1ZNTY0V99QGiJUrV9p7qGK0ih1WVlZav0r1GDRoUFuZ0iRSzg466CC7FqVI6XqqqqqykJL6buHChSbTq49VZuCVV14xWT9uaHxpe/wf//hHIGXr+++/D9RVhm+44Qb7O6SV81wXyoW0Sjhp0iQgVYqkfohaY2316tWW/Kv+fe+99zjggAOAtDIVJ1avXm3J9c8++yyQUueViFxf4Y2iiJUrVwLps0ufeeYZU1yFVOabbropFv0oZs+eDaTCt2qX+jOzmGp9oiiysSv03dTW1jZ6TmGukF0DBgygb9++QFqtktLdtWtXXnzxRSCdavDRRx/ZBgLdD7UGaOsIlCtTjuM4juM4WRCf5fZmKC4utoKcSnDV1nhIr8oVN+3WrZt5JVKwVKZ+48aNnHPOOfa+uUZtv/baa+2oBj3205/+FIDvfve7ll8iL3PYsGG2epcdcfAo6iNPqaqqyjwKebyyb/bs2ZbzIEpKSizZU7brLMJcF5cTyjeQmtSxY0dTZjLthpTH/JOf/ARIn2s2YMAAs2nPPfes87qddtrJ7JcyV1JSkrMSFxpbQ4cONYVDyqiuxeLi4gZ5fEuXLuWKK64A0mUGpLSdf/75lv8Qh2sxE33Pl112GZAqAQCp3JL6ydePPvqoFZSV0lb/HMY40Jhaoe9dc05FRQW33HILkN6C3rdvXzt+RXlHcUBq2iOPPGIbUpTvlEwmrf2nn346kFam1qxZU+foLUjNKZqDNL/o7MlMxTkOjBo1Ckjd2zLzhKHxnLbM+6M2SEh5Vf5qt27dYmWjCCE0yCdV1GnChAk2X6oE0tq1ay26oQ1Ot956K5DKK2vL+TMvFlM777yzTdi6OWUOEA0c/SwqKrL6GUpa0+Q2ZMgQO6soDoNHku3dd9/d4DBNVbdViBLSF9JTTz1lMnT918UhBCY0Qc2aNcsu5PpnJ+n/kK5i37dvXwsHKjkyTnZB+kak0E5ZWZnZqzbPmjULSPWhbriy4/DDD7fkWL2XFhq9e/e270LvFYek3+LiYltM6PrJbJe+C018/fr1s8ObNalp59Enn3xi41+Vm+OGFsxa7Gaiify+++6zcJkqZMcxHLY51Jfz5s2zRYhCJ9tuuy0HHnhgnefFASUm33bbbXZfyEQLK+34qn8oLqQXU8lk0saxNkgo7SIO110mupddfPHFVu1bjmpjTon6bPny5SYqKASm8PXn7aiOI1oUQzqcfsghhwDw5JNPmrOu70ZO7JQpUxg3bhzQNs6bh/kcx3Ecx3GyIC+UqW7dulkSmbwmbUEePHiwefDyDt98801bcat2iLysk046abPb19sLeVI62btHjx62lVdehlSNzK3YClv26tXLVteZSYSQ8rzqn02Y6Y3p9/bwuPQZu+22m/WFwrFSXFatWmX1or7xjW8AKU9JZ/gpaVLVh6+++uqcV3SHhsrM8uXL7Zw+ff9Sqqqrq03RkTJ65ZVXmkeo8SCVsWvXrvY3KVlxUQXUjua0p6ioyOzQBgIpktOnT7frWYprHPq1uUhFfO211ywUoRpFcVMzmkKqTVlZmUUBNBZLS0tjt1kHsPMFk8mkKYFq5+67726nRegalDK8evVqS0DXPaNLly52xqSS2eOUdN4Y3bt336INHPPmzbPafQpba+NLnKl/nqDuIwMHDrT7RmZ6hNQ3/VSUaurUqRYiVKmL1pxv8mfmchzHcRzHiSFNLr1DCDsC9wMDgFpgahRFt4UQegEPA4OAj4Hjoyha2RaN7NixI9deey2QrmieWVRPSpMUnLvvvtvO8tNKXMm8J554Ys6936qqKsuDUrLj5MmTLS9D3oKKXN54443mcamKb8eOHc2OxpSmxlSDXGx9VZv69+9vXoS2+H/00UdAqtiltqCrzEAikTAP5KWXXgLSuWOXX355rPKn1Jb169c3+I5l65AhQxg7diyAxe1LS0uteGJ5eTmQTiSNoqhBon2ux62QsihbpQw3Rf3CrBs3brScxriobs1BduscsZqaGttMEbdcqSiKbA6UOh9FkZUQULkKXWMrV66sk8QNKeVt8uTJADYPK3E5FxsHZM/bb78NwKWXXmpFRaVw9ujRo4GCqnG3Zs0abr/9diBVYRtSOUTHH388EO9istlw11132XcxZcoUIP7XXRRFlvtUVlYGpPtx2LBhlj+muaVHjx6mOunn3XffDaT6XeV4FCVozftIc2bnGuAHURQNA/YGzg8hDAcuBaZHUbQrMH3T/x3HcRzHcbYqmlSmoihaAizZ9PvaEMK7wA7AkcDXNj3tD8A/gR+1SStJrzKVd6GM/RCCxUTlbZWVldnWbK28Dz/8cCAeMeIoiswjV8x7/Pjxliul1bJi+DNmzLDnqeRDpkexuRwWedG1tbWmEuXCG0kkEuZRKHdB7Zk4caKpi1JfOnToYF6+8omUo5JIJGKhTEmhkTJTXV1teWDylE866SQgdW6f7MksZaH30HeiPJCSkhI7Mmf33Xe35+eaZDJpnqL6T7kpTeUJqWyE1JBEImHqWxxsay46407FSPfdd1+bl+LGhg0bLI9LZ2MuWbKkwQ7gzVFdXW07MlXeQsfl3HLLLdb/7dWHyi099dRTgZTCq52/m1PKND47derUYK7NPIIr3/LdmkLz7rRp0+x60xl+cWf9+vUWwRDquz59+lifZd431H+690mR7dq1q72Xco9b8z6yRRl2IYRBwJeB14D+mxZaRFG0JITQr9Va1fhnA2kZXVJsIpGwA3BVp6m6utq+3N122w1Iy9NxSCosKiqyLdQK62RORFo8qK2HHHJIg9o2jS2mmvrMXNywtFiYM2eOVQBXiQQdZjlo0KBGQ1iS8+vbHJcbr9qcGcpTDan61d4zL3xRW1trC0RtNtDBx/3797cLPi7hPUhdb1pMaYuybm5du3ZttG9U+kJbzpVI2r9/f9t+ni/U1NRw3HHH1Xls2rRpseqjTKqqqnj55ZeB9FbxLVlIifo3JpVtqampsblVjmqbV5re9F3LyVi3bt0Wz+uZ1dAhdS1qLo5rX7aUv//970AqcV9OeRzug5tDGwPmzp1r/aKFoMZZ586dG4RwKysrbVOLNqFpw0vPnj2tb9vi7Npmj5oQQnfg/4ALoyhaswWvmxRCeCOE8Ia+lEKj0G0sdPvAbSwUCt3GQrcP3MZCYWuwMZNmLU9DCMWkFlIPRlH06KaHl4UQSjepUqVAeWOvjaJoKjAVYMyYMa22v1Yr0k6dOtlqW15ueXm5KVhK2G7LpMIttXGbbbZhyJAhAJaAXFlZaaeBy7OTQtWpUyeTI+VVbrPNNlskR2fjbWXTh7KvrKzMVBdty5VcG0JosPW6srLSQl/6PhROaAvPtyU2qh0q2zF//nwLZUlVW7RoEZBKsq/fBytXrrSzpZ544ok6r9tll11szDY3wbspWuNa3LhxoyVx6qfG4ZgxY6wYoDzF+fPnc/755wPYphB5xZMnT7Yx0Fq01XwjLrjgAgsVnHbaaUBamWsPttS+8vJy2/igdq9du9aSy4X6JJlM2rWYeU1qrKuvpZRvt912dl1LORgwYECD57eFjZn3gPqlYEIIDVRS2bNu3Tq7znTeWyKRYI899qjzHm1JW4/TTFTqAeDCCy9sy4+qQ0tsVB9JzV64cKEVZ5WaJGV7/fr19phKXbzwwgs8//zzALzzzjtAOjoyYsQIKz6bk6KdITWy7gbejaLo1ow/PQF8e9Pv3wb+2uqtcxzHcRzHiTnNUabGAacC74QQ3tr02I+BnwGPhBDOAD4Bjvuc17c52k6vnIwTTjihTrIhxCfPBlIqkRKVhw8fDqRyERTHV26DVI3S0tIGdjS3iF5c7B41apSVOpAnr1IWyWTSVBt5mJ9++qn14YABAwBs63JcznHTd6u+PO6440w5VHKy8vn69etnypryTsrKynj44YeB9NEYI0eOBFKFPZUbELeEWCVgP/744wA8+OCDQKpftJFACl1FRYWpk/IiTzzxRAAmTZqUN/kpujYfeOABs+O6667LZZOaRWlpqZ2luP/++wMpr1+qrxQmlY4pLy+3nJO5c+cCqetTfSi1XON09OjRVvJE31FlZaU9T3lNbd3P9cs5JJNJ+11zp2x44YUXGmyzHzVqVKurpLlGc5EU5E6dOllR5Lii+5r6buHChXZ+p8aQFO7MnClt6Prggw8s30rFgqVGnX322W1SrFM0Zzfff4DPuyMf2LrNyY76FcHjjDpTC4suXbrYIkoJc5ogOnbsSP/+/e15en1cFkqbQ6Gqo48+2s5r0+JAF0X37t1tItMuuOLiYttxokNWFcaN2w1Y/dCzZ0+OPPJIIJ14Pn36dCCVdK+JQjtNly9fbv2pZHMlN++9996tFt5rTXr27Mm5554LpA9s1saPqqoqk+AzDyc94IADALj++usB7OabD9ep+uyXv/wlkLpByZ58uPluu+22tvDRZpzM66f+HJIZcs9Mzs5cpEC6fzOfrwVadXW1LcgU9tUmjbagY8eONpbUvpUrV5rzphvt3/72NwDeffddm4N0SPkxxxxjjkAu6vG1BXPmzAHSaQJDhw6lb9++uWxSs9EY7dOnjy2CNbc0dg6jFszdu3c3gULn9qmu3/bbb98miefW5jZ7Z8dxHMdxnK2AeO+P3ArQCrxbt26MHz8egH322QdIe0j5okI1huzr16+f1VzSzg5V/U4kEublyzvs1auXqVRS6OTlxpWioiILl0ihGjFiBJDyquQ9S37u3r27hagVKpT9Xbp0iZ0CBylvXZ6f6hZJDUgkEtZm2RpHG7aEmTNnAumTF/r27WtVs/PlmlQfNLcv6pcgKSoqatZWelVFr62tNTW5vULUaqva2adPH7u2pGwobSKRSNhpE9qgNHr06NiF07MhiiJuuOGGOo9deOGFsS+JoH5UFGPs2LGm5EtZVOQG0mFqjb0xY8YwYcIEIJ0eItWqreei/J7pHMdxHMdxcky8l6lbKfnuzTdGUVGRxavrn53UFFJy8gH1nc6MGj16dJ2fhYg8+vpnCeYzUtukSGnsnnfeeZa/6NRlSxWwtiSEYLlCKpCsHM0oivJGVWwpiUTCSgMoLzcOp39sjsxNVZmnK6j/tIFHqlXv3r0tB0zKfmb+XHuPw9yPesdxHMdxnDzGlSnHcZx6aMeQdh7quJ+DDz644FWNQmdr6L9OnTrx9NNPA2l7da5rXMnsFylTHTp0sMKq+hlXfDHlOI5TD4WWf/jDH9b56Tj5QluWo3Aa4mE+x3Ecx3GcLAjNraTdKh8WwnJgPVDRbh/acvpQt507RVHUZMWzEMJaYG6btap12WIb87wPofBtbO443Rps9GsxPvi1+DlsJTYW9LUI7byYAgghvBFF0Zh2/dAW0NJ25ot9UPg2ZtNOtzE+FPpLfvMnAAAC5ElEQVQ4hcK30cdp2722PSn0cQotb6uH+RzHcRzHcbLAF1OO4ziO4zhZkIvF1NQcfGZLaGk788U+KHwbs2mn2xgfCn2cQuHb6OO07V7bnhT6OIUWtrXdc6Ycx3Ecx3EKCQ/zOY7jOI7jZEG7LaZCCIeGEOaGED4MIVzaXp/bFCGEHUMIL4YQ3g0h/C+E8P1Nj18dQvg0hPDWpn8Tm/FebmOOaC0b42ofFL6NPk7dxnrvU9D2bXqN25gjWtNGIHW4YFv/AzoA84AvAh2Bt4Hh7fHZzWhbKTB60+8lwPvAcOBq4GK3ceuxMc72bQ02+jh1G7cW+9zGwrFR/9pLmdoL+DCKovlRFCWAPwNHttNnb5YoipZEUfTmpt/XAu8CO7TgrdzGHNJKNsbWPih8G32cbhGFbmOh2wduY05pRRuB9gvz7QAszPj/IrJodFsRQhgEfBl4bdNDk0MIs0II94QQejbxcrcxJmRhY17YB4Vvo4/Trd7GQrcP3MbYkKWNQPstpho7pjtW2whDCN2B/wMujKJoDfA7YGdgD2AJ8Ium3qKRx9zGdiZLG2NvHxS+jT5O3UYK3z5wG2NBK9gItN9iahGwY8b/BwKL2+mzmySEUEzqy3wwiqJHAaIoWhZFUTKKolrg/5GSKzeH25hjWsHGWNsHhW+jj1O3cROFbh+4jTmnlWwE2m8xNQPYNYQwOITQETgReKKdPnuzhBACcDfwbhRFt2Y8XprxtG8Cs5t4K7cxh7SSjbG1DwrfRh+nhttY+PaB25hTWtHGFFuasd7Sf8BEUtny84DL2+tzm9Gu8aRkx1nAW5v+TQQeAN7Z9PgTQKnbWPg2xtW+rcFGH6du49Zkn9tYODZGUeQV0B3HcRzHcbLBK6A7juM4juNkgS+mHMdxHMdxssAXU47jOI7jOFngiynHcRzHcZws8MWU4ziO4zhOFvhiynEcx3EcJwt8MeU4juM4jpMFvphyHMdxHMfJgv8P1v+9EQGjV9UAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(10):\n",
    "\n",
    "    ##########################\n",
    "    ### RANDOM SAMPLE\n",
    "    ##########################    \n",
    "    \n",
    "    n_images = 10\n",
    "    rand_features = torch.randn(n_images, num_latent).to(device)\n",
    "    new_images = model.decoder(rand_features)\n",
    "\n",
    "    ##########################\n",
    "    ### VISUALIZATION\n",
    "    ##########################\n",
    "\n",
    "    image_width = 28\n",
    "\n",
    "    fig, axes = plt.subplots(nrows=1, ncols=n_images, figsize=(10, 2.5), sharey=True)\n",
    "    decoded_images = new_images[:n_images]\n",
    "\n",
    "    for ax, img in zip(axes, decoded_images):\n",
    "        curr_img = img.detach().to(torch.device('cpu'))\n",
    "        ax.imshow(curr_img.view((image_width, image_width)), cmap='binary')\n",
    "        \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "numpy       1.15.4\n",
      "torch       1.0.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%watermark -iv"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  },
  "toc": {
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}