vae-conditional-full-mnist.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import cm\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import json\n",
    "import os\n",
    "import feather\n",
    "from pandas.io.json import json_normalize\n",
    "from functools import partial\n",
    "from pdb import set_trace\n",
    "\n",
    "from datetime import datetime, timedelta\n",
    "import urllib.request\n",
    "\n",
    "from fastai2.vision.all import *\n",
    "from fastai2.tabular.all import *\n",
    "\n",
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "url = URLs.MNIST\n",
    "path = untar_data(url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "class PILImageBWNoised(PILImageBW): pass\n",
    "class TensorImageBWNoised(TensorImageBW): pass\n",
    "PILImageBWNoised._tensor_cls = TensorImageBWNoised\n",
    "\n",
    "class PILImageTarget(PILImageBW): pass\n",
    "class TensorImageTarget(TensorImageBW): pass\n",
    "PILImageTarget._tensor_cls = TensorImageTarget\n",
    "\n",
    "@Normalize\n",
    "def encodes(self, to:TensorImageTarget): return to\n",
    "\n",
    "class AddNoiseTransform(Transform):\n",
    "    \"Add noise to image\"\n",
    "    order = 11\n",
    "    def __init__(self, noise_factor=0.5): store_attr(self, 'noise_factor')\n",
    "    def encodes(self, o:TensorImageBWNoised): return (o + (self.noise_factor * torch.randn(*o.shape).to(o.device))).clamp(0,1)\n",
    "\n",
    "mnist = DataBlock(blocks=(ImageBlock(cls=PILImageBWNoised), CategoryBlock, ImageBlock(cls=PILImageTarget)), \n",
    "                 get_items=get_image_files,\n",
    "                 get_x= [noop, parent_label],\n",
    "                 splitter=GrandparentSplitter(train_name='training' if url == URLs.MNIST else 'train', valid_name='testing' if url == URLs.MNIST else 'valid'),\n",
    "                 batch_tfms=[AddNoiseTransform(.3), Normalize()],\n",
    "                 n_inp=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "dls = mnist.dataloaders(path, num_workers=0, bs=128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "class UnFlatten(Module):\n",
    "    def __init__(self, size): self.size = size\n",
    "    def forward(self, input):\n",
    "        return input.view(input.size(0), self.size, 7, 7)\n",
    "\n",
    "class VAE(nn.Module):\n",
    "    def __init__(self, image_channels=3, h_dim=1024, z_dim=32):\n",
    "        super(VAE, self).__init__()\n",
    "        self.z_dim = z_dim\n",
    "        kernel_size = 2\n",
    "        self.encoder = nn.Sequential(\n",
    "            ResBlock(1, 1, 16, act_cls=Swish),\n",
    "            nn.MaxPool2d(2, 2),\n",
    "            ResBlock(1, 16, 4, act_cls=Swish),\n",
    "            nn.MaxPool2d(2, 2),\n",
    "            Flatten()\n",
    "        )\n",
    "        \n",
    "        self.emb = Embedding(10, emb_sz_rule(10))\n",
    "        \n",
    "        h_dim = h_dim + emb_sz_rule(10)\n",
    "        \n",
    "        self.fc1 = nn.Linear(h_dim, z_dim)\n",
    "        self.fc2 = nn.Linear(h_dim, z_dim)\n",
    "        self.fc3 = nn.Linear(z_dim, h_dim-emb_sz_rule(10))\n",
    "        self.fc4 = nn.Linear(emb_sz_rule(10), 7*7)\n",
    "        \n",
    "        self.decoder = nn.Sequential(\n",
    "            ResBlock(1, 5, 8, ks=1, act_cls=Swish),\n",
    "            nn.Upsample(scale_factor=2),\n",
    "            ResBlock(1, 8, 16, act_cls=Swish),\n",
    "            nn.Upsample(scale_factor=2),\n",
    "            nn.Conv2d(16, 1, 3, padding=1),\n",
    "            nn.Sigmoid()\n",
    "        )\n",
    "        \n",
    "    def reparameterize(self, mu, logvar):\n",
    "        std = logvar.mul(0.5).exp_()\n",
    "        # return torch.normal(mu, std)\n",
    "        esp = torch.cuda.HalfTensor(*mu.size()).normal_()\n",
    "#         esp = torch.randn(*mu.size()).to(mu.device)\n",
    "        z = mu + std * esp\n",
    "        return z\n",
    "    \n",
    "    def bottleneck(self, h):\n",
    "        mu, logvar = self.fc1(h), self.fc2(h)\n",
    "        z = self.reparameterize(mu, logvar)\n",
    "        return z, mu, logvar\n",
    "\n",
    "    def encode(self, x, num):\n",
    "        h = self.encoder(x)\n",
    "        e = self.emb(num)\n",
    "        h = torch.cat([h,e], dim=1)\n",
    "        z, mu, logvar = self.bottleneck(h)\n",
    "        return z, mu, logvar\n",
    "\n",
    "    def decode(self, z, num):\n",
    "        z = self.fc3(z)\n",
    "        z = z.view(z.size(0), 4, 7, 7)\n",
    "        \n",
    "        num = self.fc4(self.emb(num)).view(z.size(0), 1, 7, 7)\n",
    "        z = torch.cat([z, num], dim=1)\n",
    "        \n",
    "        z = self.decoder(z)\n",
    "        return z\n",
    "\n",
    "    def forward(self, x, num):\n",
    "        num = num.long()\n",
    "        z, mu, logvar = self.encode(x, num)\n",
    "        z = self.decode(z, num)\n",
    "        return z, mu, logvar\n",
    "    \n",
    "    def inference(self, n=1, c=None):\n",
    "\n",
    "        batch_size = n\n",
    "        z = torch.randn([batch_size, self.latent_size])\n",
    "\n",
    "        recon_x = self.decoder(z, c)\n",
    "\n",
    "        return recon_x\n",
    "    \n",
    "model = VAE(image_channels=1, h_dim=196, z_dim=128).cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [],
   "source": [
    "class Concat(Module):\n",
    "    def __init__(self, dim):\n",
    "        self.dim\n",
    "    def forward(self, tensors):\n",
    "        return torch.cat(tensors, dim=self.dim)\n",
    "\n",
    "class FastVAE(Module):\n",
    "    def __init__(self, img_channels=3, z_dim=64):\n",
    "        self.body = create_body(resnet18, n_in=img_channels)\n",
    "        h_dim = num_features_model(self.body)\n",
    "        \n",
    "        self.z_dim = z_dim\n",
    "        \n",
    "        self.emb = Embedding(10, emb_sz_rule(10))\n",
    "        \n",
    "        self.encoder = nn.Sequential(*[self.body, nn.AdaptiveAvgPool2d(1), Flatten()])\n",
    "        \n",
    "        h_dim = h_dim + emb_sz_rule(10)\n",
    "        \n",
    "        self.fc_mu = nn.Linear(h_dim, z_dim)\n",
    "        self.fc_logvar = nn.Linear(h_dim, z_dim)\n",
    "        self.fc_z_to_channel = nn.Linear(z_dim, 4*7*7)\n",
    "        self.fc_emb_to_channel = nn.Linear(emb_sz_rule(10), 1*7*7)\n",
    "        \n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.Conv2d(4+1, 4, 1),\n",
    "            Swish(),\n",
    "            nn.Upsample(scale_factor=2),\n",
    "            nn.Conv2d(4, 16, 3, padding=1),\n",
    "            Swish(),\n",
    "            nn.Upsample(scale_factor=2),\n",
    "            nn.Conv2d(16, 1, 3, padding=1),\n",
    "            nn.Sigmoid()\n",
    "        )\n",
    "        \n",
    "        self.fc = nn.ModuleList([self.fc_mu, self.fc_logvar, self.fc_z_to_channel, self.fc_emb_to_channel, self.decoder])\n",
    "        \n",
    "    def reparameterize(self, mu, logvar):\n",
    "        std = logvar.mul(0.5).exp_()\n",
    "        # return torch.normal(mu, std)\n",
    "        esp = torch.cuda.HalfTensor(*mu.size()).normal_()\n",
    "        z = mu + std * esp\n",
    "        return z\n",
    "    \n",
    "    def bottleneck(self, h):\n",
    "        mu, logvar = self.fc_mu(h), self.fc_logvar(h)\n",
    "        z = self.reparameterize(mu, logvar)\n",
    "        return z, mu, logvar\n",
    "\n",
    "    def encode(self, x, num):\n",
    "        h = self.encoder(x)\n",
    "        e = self.emb(num)\n",
    "        h = torch.cat([h,e], dim=1)\n",
    "        z, mu, logvar = self.bottleneck(h)\n",
    "        return z, mu, logvar\n",
    "\n",
    "    def decode(self, z, num):\n",
    "        z = self.fc_z_to_channel(z)\n",
    "        z = z.view(z.size(0), 4, 7, 7)\n",
    "        \n",
    "        num = self.fc_emb_to_channel(self.emb(num)).view(z.size(0), 1, 7, 7)\n",
    "        z = torch.cat([z, num], dim=1)\n",
    "        \n",
    "#         t = self.adapt(z)\n",
    "#         print(t.shape)\n",
    "        z = self.decoder(z)\n",
    "        return z\n",
    "\n",
    "    def forward(self, x, num):\n",
    "        num = num.long()\n",
    "#         padding = x.new(x.size(0), 2, x.size(2), x.size(3))\n",
    "#         x = torch.cat((x, padding), 1)\n",
    "        z, mu, logvar = self.encode(x, num)\n",
    "        z = self.decode(z, num)\n",
    "        return z, mu, logvar\n",
    "    \n",
    "# model = FastVAE(img_channels=3, z_dim=64).cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "img, num, targets = dls.one_batch()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %debug\n",
    "with torch.no_grad():\n",
    "    res = model(img, num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "class BCEMetric(Metric):\n",
    "    def __init__(self): self.preds = []\n",
    "    def accumulate(self, learn):\n",
    "        self.preds.append(F.binary_cross_entropy(to_detach(learn.pred[0]), to_detach(learn.y), reduction='sum'))\n",
    "    @property\n",
    "    def value(self):\n",
    "        return np.array(self.preds).mean()\n",
    "    \n",
    "class KLDMetric(Metric):\n",
    "    def __init__(self): self.preds = []\n",
    "    def accumulate(self, learn):\n",
    "        pred, mu, logvar, _ = to_detach(learn.pred)\n",
    "        KLD = -0.5 * torch.sum(1 + logvar - mu.pow(2) - logvar.exp())\n",
    "        self.preds.append(KLD)\n",
    "    @property\n",
    "    def value(self):\n",
    "        return np.array(self.preds).mean()\n",
    "    \n",
    "class MUMetric(Metric):\n",
    "    def __init__(self): self.preds = []\n",
    "    def accumulate(self, learn):\n",
    "        pred, mu, logvar, _ = to_detach(learn.pred)\n",
    "        self.preds.append(mu.mean())\n",
    "    @property\n",
    "    def value(self):\n",
    "        return np.array(self.preds).mean()\n",
    "    \n",
    "class StdMetric(Metric):\n",
    "    def __init__(self): self.preds = []\n",
    "    def accumulate(self, learn):\n",
    "        pred, mu, logvar, _ = to_detach(learn.pred)\n",
    "        self.preds.append((logvar.exp_() ** .5).mean())\n",
    "    @property\n",
    "    def value(self):\n",
    "        return np.array(self.preds).mean()\n",
    "\n",
    "def loss_fn(output, x):\n",
    "    recon_x, mu, logvar, kl_weight = output\n",
    "    \n",
    "    BCE = F.binary_cross_entropy(recon_x, x, reduction='sum')\n",
    "\n",
    "    # see Appendix B from VAE paper:\n",
    "    # Kingma and Welling. Auto-Encoding Variational Bayes. ICLR, 2014\n",
    "    # 0.5 * sum(1 + log(sigma^2) - mu^2 - sigma^2)\n",
    "    KLD = -0.5 * torch.sum(1 + logvar - mu.pow(2) - logvar.exp())\n",
    "\n",
    "    return BCE + (kl_weight * KLD)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1fb05566ac8>]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f = combine_scheds([.3, .5, .2], [SchedCos(0,10), SchedCos(10,1), SchedNo(1,1)])\n",
    "p = torch.linspace(0.,1,100)\n",
    "plt.plot(p,[f(o) for o in p])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai2.callback.tensorboard import TensorBoardCallback\n",
    "\n",
    "# def params(m):\n",
    "#     \"Return all parameters of `m`\"\n",
    "#     return [p for p in m.parameters()]\n",
    "\n",
    "# def  _split(m): return L(m.encoder, m.fc).map(params)\n",
    "\n",
    "scheds = {'kl_weight': f}\n",
    "\n",
    "class AnnealedLossCallback(Callback):\n",
    "    def after_pred(self):\n",
    "        kl = self.learn.pred[0].new(1)\n",
    "        kl[0] = self.opt.hypers[0]['kl_weight']\n",
    "        self.learn.pred = self.learn.pred + (kl,)\n",
    "        \n",
    "class DelayedEarlyStoppingCallback(EarlyStoppingCallback):\n",
    "    def after_epoch(self):\n",
    "        if(self.pct_train > .6):\n",
    "            super().after_epoch()\n",
    "        \n",
    "model = VAE(image_channels=1, h_dim=196, z_dim=128).cuda()\n",
    "\n",
    "# shutil.rmtree('logs')\n",
    "# DelayedEarlyStoppingCallback(patience=5, monitor='kld')\n",
    "learn = Learner(dls, model, cbs=[ParamScheduler(scheds), AnnealedLossCallback(), ShowGraphCallback()], loss_func=loss_fn, metrics=[BCEMetric(), KLDMetric(), MUMetric(), StdMetric()])\n",
    "# learn = learn.to_fp16()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>bce</th>\n",
       "      <th>kld</th>\n",
       "      <th>mu</th>\n",
       "      <th>std</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>49240.941406</td>\n",
       "      <td>46473.011719</td>\n",
       "      <td>45862.12</td>\n",
       "      <td>23229.787</td>\n",
       "      <td>-0.11116075</td>\n",
       "      <td>0.4659131</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>32856.019531</td>\n",
       "      <td>31489.925781</td>\n",
       "      <td>38177.89</td>\n",
       "      <td>24235.482</td>\n",
       "      <td>-0.07615657</td>\n",
       "      <td>0.3506731</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>23071.173828</td>\n",
       "      <td>22240.945312</td>\n",
       "      <td>32334.13</td>\n",
       "      <td>23640.676</td>\n",
       "      <td>-0.05621188</td>\n",
       "      <td>0.30639923</td>\n",
       "      <td>01:36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>19643.550781</td>\n",
       "      <td>19296.556641</td>\n",
       "      <td>28576.373</td>\n",
       "      <td>21871.174</td>\n",
       "      <td>-0.044826448</td>\n",
       "      <td>0.30322415</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>18412.857422</td>\n",
       "      <td>18176.861328</td>\n",
       "      <td>26059.752</td>\n",
       "      <td>19857.637</td>\n",
       "      <td>-0.039370596</td>\n",
       "      <td>0.3239451</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>17642.796875</td>\n",
       "      <td>17483.023438</td>\n",
       "      <td>24228.111</td>\n",
       "      <td>18077.496</td>\n",
       "      <td>-0.03681837</td>\n",
       "      <td>0.35526928</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>17266.500000</td>\n",
       "      <td>17082.734375</td>\n",
       "      <td>22812.256</td>\n",
       "      <td>16614.672</td>\n",
       "      <td>-0.0350689</td>\n",
       "      <td>0.38875282</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>17159.984375</td>\n",
       "      <td>17052.056641</td>\n",
       "      <td>21714.535</td>\n",
       "      <td>15364.035</td>\n",
       "      <td>-0.03324466</td>\n",
       "      <td>0.4219596</td>\n",
       "      <td>01:43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>17305.464844</td>\n",
       "      <td>17164.544922</td>\n",
       "      <td>20832.115</td>\n",
       "      <td>14314.728</td>\n",
       "      <td>-0.0316113</td>\n",
       "      <td>0.45238984</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>17846.162109</td>\n",
       "      <td>17905.111328</td>\n",
       "      <td>20227.768</td>\n",
       "      <td>13322.954</td>\n",
       "      <td>-0.030040152</td>\n",
       "      <td>0.4841605</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>18200.382812</td>\n",
       "      <td>17985.542969</td>\n",
       "      <td>19714.867</td>\n",
       "      <td>12475.498</td>\n",
       "      <td>-0.028433962</td>\n",
       "      <td>0.51139605</td>\n",
       "      <td>01:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>18531.060547</td>\n",
       "      <td>18199.748047</td>\n",
       "      <td>19293.709</td>\n",
       "      <td>11729.384</td>\n",
       "      <td>-0.026834313</td>\n",
       "      <td>0.53669864</td>\n",
       "      <td>01:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>18892.042969</td>\n",
       "      <td>18589.982422</td>\n",
       "      <td>18981.873</td>\n",
       "      <td>11045.916</td>\n",
       "      <td>-0.02510274</td>\n",
       "      <td>0.5603049</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>18726.287109</td>\n",
       "      <td>18884.792969</td>\n",
       "      <td>18741.002</td>\n",
       "      <td>10428.921</td>\n",
       "      <td>-0.023645535</td>\n",
       "      <td>0.5814688</td>\n",
       "      <td>01:47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>18394.433594</td>\n",
       "      <td>18335.845703</td>\n",
       "      <td>18490.805</td>\n",
       "      <td>9877.984</td>\n",
       "      <td>-0.022029016</td>\n",
       "      <td>0.60179096</td>\n",
       "      <td>01:56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>18140.765625</td>\n",
       "      <td>18020.814453</td>\n",
       "      <td>18207.584</td>\n",
       "      <td>9407.456</td>\n",
       "      <td>-0.020901963</td>\n",
       "      <td>0.6201064</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>18091.671875</td>\n",
       "      <td>18154.279297</td>\n",
       "      <td>17948.178</td>\n",
       "      <td>8986.82</td>\n",
       "      <td>-0.019774174</td>\n",
       "      <td>0.637118</td>\n",
       "      <td>01:47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>18176.462891</td>\n",
       "      <td>18137.189453</td>\n",
       "      <td>17756.223</td>\n",
       "      <td>8581.059</td>\n",
       "      <td>-0.018710729</td>\n",
       "      <td>0.65259194</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>18295.609375</td>\n",
       "      <td>18146.566406</td>\n",
       "      <td>17566.422</td>\n",
       "      <td>8217.785</td>\n",
       "      <td>-0.017743694</td>\n",
       "      <td>0.6669745</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>18416.837891</td>\n",
       "      <td>18276.496094</td>\n",
       "      <td>17408.602</td>\n",
       "      <td>7880.6846</td>\n",
       "      <td>-0.016824028</td>\n",
       "      <td>0.68023556</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>18501.845703</td>\n",
       "      <td>18310.353516</td>\n",
       "      <td>17271.021</td>\n",
       "      <td>7568.4233</td>\n",
       "      <td>-0.01607045</td>\n",
       "      <td>0.6925283</td>\n",
       "      <td>01:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>18684.966797</td>\n",
       "      <td>18484.304688</td>\n",
       "      <td>17149.846</td>\n",
       "      <td>7281.1</td>\n",
       "      <td>-0.015277856</td>\n",
       "      <td>0.703915</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>18699.033203</td>\n",
       "      <td>18629.089844</td>\n",
       "      <td>17033.53</td>\n",
       "      <td>7018.3853</td>\n",
       "      <td>-0.014617006</td>\n",
       "      <td>0.71438384</td>\n",
       "      <td>01:43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>18828.423828</td>\n",
       "      <td>18808.353516</td>\n",
       "      <td>16939.357</td>\n",
       "      <td>6772.408</td>\n",
       "      <td>-0.014044423</td>\n",
       "      <td>0.7243113</td>\n",
       "      <td>01:56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>18916.183594</td>\n",
       "      <td>18833.632812</td>\n",
       "      <td>16880.916</td>\n",
       "      <td>6535.757</td>\n",
       "      <td>-0.0135288155</td>\n",
       "      <td>0.73355603</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>18927.519531</td>\n",
       "      <td>18843.722656</td>\n",
       "      <td>16823.545</td>\n",
       "      <td>6316.1646</td>\n",
       "      <td>-0.012928927</td>\n",
       "      <td>0.74219877</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>19107.839844</td>\n",
       "      <td>18970.021484</td>\n",
       "      <td>16759.793</td>\n",
       "      <td>6114.571</td>\n",
       "      <td>-0.012403288</td>\n",
       "      <td>0.7503449</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>19112.015625</td>\n",
       "      <td>19187.189453</td>\n",
       "      <td>16708.812</td>\n",
       "      <td>5925.4673</td>\n",
       "      <td>-0.011991681</td>\n",
       "      <td>0.757948</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>19278.751953</td>\n",
       "      <td>19108.406250</td>\n",
       "      <td>16662.75</td>\n",
       "      <td>5746.9697</td>\n",
       "      <td>-0.011620844</td>\n",
       "      <td>0.76508963</td>\n",
       "      <td>01:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>19334.250000</td>\n",
       "      <td>19254.333984</td>\n",
       "      <td>16628.559</td>\n",
       "      <td>5578.2725</td>\n",
       "      <td>-0.01121479</td>\n",
       "      <td>0.77184004</td>\n",
       "      <td>02:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>19369.185547</td>\n",
       "      <td>19244.664062</td>\n",
       "      <td>16604.752</td>\n",
       "      <td>5417.742</td>\n",
       "      <td>-0.010920201</td>\n",
       "      <td>0.7783062</td>\n",
       "      <td>01:58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>19582.050781</td>\n",
       "      <td>19418.056641</td>\n",
       "      <td>16588.547</td>\n",
       "      <td>5266.2197</td>\n",
       "      <td>-0.010587395</td>\n",
       "      <td>0.7844136</td>\n",
       "      <td>01:46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>19487.267578</td>\n",
       "      <td>19404.640625</td>\n",
       "      <td>16584.154</td>\n",
       "      <td>5121.171</td>\n",
       "      <td>-0.010330083</td>\n",
       "      <td>0.7902337</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>19595.662109</td>\n",
       "      <td>19522.263672</td>\n",
       "      <td>16584.564</td>\n",
       "      <td>4983.8706</td>\n",
       "      <td>-0.010049496</td>\n",
       "      <td>0.7957599</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>19679.513672</td>\n",
       "      <td>19511.314453</td>\n",
       "      <td>16582.342</td>\n",
       "      <td>4854.2563</td>\n",
       "      <td>-0.009815853</td>\n",
       "      <td>0.80100393</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>19732.765625</td>\n",
       "      <td>19644.240234</td>\n",
       "      <td>16589.66</td>\n",
       "      <td>4730.483</td>\n",
       "      <td>-0.009606945</td>\n",
       "      <td>0.805985</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>36</td>\n",
       "      <td>19872.552734</td>\n",
       "      <td>19619.966797</td>\n",
       "      <td>16587.682</td>\n",
       "      <td>4614.221</td>\n",
       "      <td>-0.009343331</td>\n",
       "      <td>0.8107152</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>19871.681641</td>\n",
       "      <td>19744.154297</td>\n",
       "      <td>16593.867</td>\n",
       "      <td>4503.0117</td>\n",
       "      <td>-0.009088758</td>\n",
       "      <td>0.8151997</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>19985.933594</td>\n",
       "      <td>19835.095703</td>\n",
       "      <td>16598.438</td>\n",
       "      <td>4397.6772</td>\n",
       "      <td>-0.008884348</td>\n",
       "      <td>0.8194714</td>\n",
       "      <td>01:43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>19963.464844</td>\n",
       "      <td>19793.027344</td>\n",
       "      <td>16604.871</td>\n",
       "      <td>4296.8867</td>\n",
       "      <td>-0.008687535</td>\n",
       "      <td>0.8235364</td>\n",
       "      <td>01:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>20097.089844</td>\n",
       "      <td>19881.146484</td>\n",
       "      <td>16611.773</td>\n",
       "      <td>4200.928</td>\n",
       "      <td>-0.008479767</td>\n",
       "      <td>0.82744706</td>\n",
       "      <td>01:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>20029.222656</td>\n",
       "      <td>19891.787109</td>\n",
       "      <td>16619.523</td>\n",
       "      <td>4109.1895</td>\n",
       "      <td>-0.008264256</td>\n",
       "      <td>0.8311836</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>42</td>\n",
       "      <td>20159.125000</td>\n",
       "      <td>19975.851562</td>\n",
       "      <td>16634.207</td>\n",
       "      <td>4020.8513</td>\n",
       "      <td>-0.008099596</td>\n",
       "      <td>0.8347578</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>43</td>\n",
       "      <td>20179.955078</td>\n",
       "      <td>20161.166016</td>\n",
       "      <td>16652.473</td>\n",
       "      <td>3936.3516</td>\n",
       "      <td>-0.007938756</td>\n",
       "      <td>0.8381847</td>\n",
       "      <td>01:46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>44</td>\n",
       "      <td>20235.347656</td>\n",
       "      <td>20106.054688</td>\n",
       "      <td>16665.643</td>\n",
       "      <td>3855.8218</td>\n",
       "      <td>-0.0077650845</td>\n",
       "      <td>0.84146696</td>\n",
       "      <td>01:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45</td>\n",
       "      <td>20336.808594</td>\n",
       "      <td>20148.199219</td>\n",
       "      <td>16681.604</td>\n",
       "      <td>3778.364</td>\n",
       "      <td>-0.0076128156</td>\n",
       "      <td>0.84462404</td>\n",
       "      <td>01:43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>46</td>\n",
       "      <td>20380.851562</td>\n",
       "      <td>20266.794922</td>\n",
       "      <td>16700.07</td>\n",
       "      <td>3704.01</td>\n",
       "      <td>-0.007462773</td>\n",
       "      <td>0.8476293</td>\n",
       "      <td>01:44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>47</td>\n",
       "      <td>20338.214844</td>\n",
       "      <td>20230.757812</td>\n",
       "      <td>16716.125</td>\n",
       "      <td>3632.7468</td>\n",
       "      <td>-0.0073044165</td>\n",
       "      <td>0.8505139</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>48</td>\n",
       "      <td>20362.421875</td>\n",
       "      <td>20197.029297</td>\n",
       "      <td>16734.139</td>\n",
       "      <td>3563.9395</td>\n",
       "      <td>-0.0071852305</td>\n",
       "      <td>0.8533131</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>49</td>\n",
       "      <td>20435.595703</td>\n",
       "      <td>20286.341797</td>\n",
       "      <td>16748.965</td>\n",
       "      <td>3498.2654</td>\n",
       "      <td>-0.007036368</td>\n",
       "      <td>0.8559996</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>20391.515625</td>\n",
       "      <td>20283.205078</td>\n",
       "      <td>16767.242</td>\n",
       "      <td>3434.6606</td>\n",
       "      <td>-0.0068925596</td>\n",
       "      <td>0.85858506</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>51</td>\n",
       "      <td>20489.152344</td>\n",
       "      <td>20335.791016</td>\n",
       "      <td>16787.52</td>\n",
       "      <td>3373.2634</td>\n",
       "      <td>-0.0067442437</td>\n",
       "      <td>0.86107683</td>\n",
       "      <td>01:44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52</td>\n",
       "      <td>20442.669922</td>\n",
       "      <td>20334.787109</td>\n",
       "      <td>16804.268</td>\n",
       "      <td>3314.423</td>\n",
       "      <td>-0.0066160266</td>\n",
       "      <td>0.86347514</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>53</td>\n",
       "      <td>20424.406250</td>\n",
       "      <td>20413.410156</td>\n",
       "      <td>16823.457</td>\n",
       "      <td>3257.5508</td>\n",
       "      <td>-0.0065384433</td>\n",
       "      <td>0.86578333</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>54</td>\n",
       "      <td>20481.056641</td>\n",
       "      <td>20392.136719</td>\n",
       "      <td>16842.826</td>\n",
       "      <td>3202.588</td>\n",
       "      <td>-0.0064284336</td>\n",
       "      <td>0.8680196</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>55</td>\n",
       "      <td>20510.832031</td>\n",
       "      <td>20360.041016</td>\n",
       "      <td>16858.82</td>\n",
       "      <td>3149.7737</td>\n",
       "      <td>-0.006338686</td>\n",
       "      <td>0.8701767</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>56</td>\n",
       "      <td>20500.126953</td>\n",
       "      <td>20359.394531</td>\n",
       "      <td>16877.102</td>\n",
       "      <td>3098.507</td>\n",
       "      <td>-0.006226639</td>\n",
       "      <td>0.872261</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>57</td>\n",
       "      <td>20474.796875</td>\n",
       "      <td>20351.599609</td>\n",
       "      <td>16889.873</td>\n",
       "      <td>3049.4707</td>\n",
       "      <td>-0.0061419117</td>\n",
       "      <td>0.874271</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>58</td>\n",
       "      <td>20563.058594</td>\n",
       "      <td>20352.013672</td>\n",
       "      <td>16904.021</td>\n",
       "      <td>3001.907</td>\n",
       "      <td>-0.006075807</td>\n",
       "      <td>0.8762062</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>59</td>\n",
       "      <td>20475.619141</td>\n",
       "      <td>20334.820312</td>\n",
       "      <td>16917.604</td>\n",
       "      <td>2955.906</td>\n",
       "      <td>-0.0059741214</td>\n",
       "      <td>0.8780782</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>20470.289062</td>\n",
       "      <td>20380.435547</td>\n",
       "      <td>16932.041</td>\n",
       "      <td>2911.3567</td>\n",
       "      <td>-0.005873791</td>\n",
       "      <td>0.8798931</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61</td>\n",
       "      <td>20499.556641</td>\n",
       "      <td>20327.839844</td>\n",
       "      <td>16943.928</td>\n",
       "      <td>2868.372</td>\n",
       "      <td>-0.00578025</td>\n",
       "      <td>0.8816544</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>62</td>\n",
       "      <td>20465.203125</td>\n",
       "      <td>20324.119141</td>\n",
       "      <td>16956.572</td>\n",
       "      <td>2826.639</td>\n",
       "      <td>-0.005692315</td>\n",
       "      <td>0.883352</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>63</td>\n",
       "      <td>20521.480469</td>\n",
       "      <td>20334.673828</td>\n",
       "      <td>16970.678</td>\n",
       "      <td>2786.0469</td>\n",
       "      <td>-0.0056139077</td>\n",
       "      <td>0.8850018</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>64</td>\n",
       "      <td>20446.615234</td>\n",
       "      <td>20430.369141</td>\n",
       "      <td>16983.574</td>\n",
       "      <td>2746.9324</td>\n",
       "      <td>-0.0055534556</td>\n",
       "      <td>0.88658303</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>65</td>\n",
       "      <td>20568.755859</td>\n",
       "      <td>20349.119141</td>\n",
       "      <td>16997.48</td>\n",
       "      <td>2708.748</td>\n",
       "      <td>-0.005465455</td>\n",
       "      <td>0.8881341</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>66</td>\n",
       "      <td>20409.593750</td>\n",
       "      <td>20245.466797</td>\n",
       "      <td>17006.936</td>\n",
       "      <td>2671.9695</td>\n",
       "      <td>-0.005378021</td>\n",
       "      <td>0.88963264</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>67</td>\n",
       "      <td>20468.000000</td>\n",
       "      <td>20300.031250</td>\n",
       "      <td>17018.078</td>\n",
       "      <td>2636.1633</td>\n",
       "      <td>-0.00531269</td>\n",
       "      <td>0.89109147</td>\n",
       "      <td>01:44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>68</td>\n",
       "      <td>20371.441406</td>\n",
       "      <td>20259.453125</td>\n",
       "      <td>17028.166</td>\n",
       "      <td>2601.425</td>\n",
       "      <td>-0.0052552726</td>\n",
       "      <td>0.8925092</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>69</td>\n",
       "      <td>20315.261719</td>\n",
       "      <td>20212.761719</td>\n",
       "      <td>17035.791</td>\n",
       "      <td>2567.8464</td>\n",
       "      <td>-0.0051695323</td>\n",
       "      <td>0.8938798</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>70</td>\n",
       "      <td>20401.882812</td>\n",
       "      <td>20270.011719</td>\n",
       "      <td>17043.568</td>\n",
       "      <td>2535.2754</td>\n",
       "      <td>-0.005084569</td>\n",
       "      <td>0.8952076</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>71</td>\n",
       "      <td>20377.847656</td>\n",
       "      <td>20243.445312</td>\n",
       "      <td>17053.338</td>\n",
       "      <td>2503.3604</td>\n",
       "      <td>-0.0050299354</td>\n",
       "      <td>0.8965093</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>72</td>\n",
       "      <td>20323.654297</td>\n",
       "      <td>20253.345703</td>\n",
       "      <td>17061.748</td>\n",
       "      <td>2472.4666</td>\n",
       "      <td>-0.0049624275</td>\n",
       "      <td>0.8977608</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>73</td>\n",
       "      <td>20326.054688</td>\n",
       "      <td>20162.033203</td>\n",
       "      <td>17066.98</td>\n",
       "      <td>2442.61</td>\n",
       "      <td>-0.004906794</td>\n",
       "      <td>0.8989849</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>74</td>\n",
       "      <td>20270.515625</td>\n",
       "      <td>20128.500000</td>\n",
       "      <td>17072.084</td>\n",
       "      <td>2413.5242</td>\n",
       "      <td>-0.004835887</td>\n",
       "      <td>0.90016913</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>20276.958984</td>\n",
       "      <td>20151.535156</td>\n",
       "      <td>17077.645</td>\n",
       "      <td>2385.195</td>\n",
       "      <td>-0.0047873296</td>\n",
       "      <td>0.9013253</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>76</td>\n",
       "      <td>20240.699219</td>\n",
       "      <td>20123.054688</td>\n",
       "      <td>17082.98</td>\n",
       "      <td>2357.5977</td>\n",
       "      <td>-0.004740937</td>\n",
       "      <td>0.9024509</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77</td>\n",
       "      <td>20251.644531</td>\n",
       "      <td>20216.908203</td>\n",
       "      <td>17088.688</td>\n",
       "      <td>2330.8105</td>\n",
       "      <td>-0.004686109</td>\n",
       "      <td>0.9035469</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>20201.945312</td>\n",
       "      <td>20131.761719</td>\n",
       "      <td>17094.059</td>\n",
       "      <td>2304.6348</td>\n",
       "      <td>-0.0046275323</td>\n",
       "      <td>0.90461195</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>79</td>\n",
       "      <td>20131.445312</td>\n",
       "      <td>20013.250000</td>\n",
       "      <td>17097.873</td>\n",
       "      <td>2279.136</td>\n",
       "      <td>-0.004586517</td>\n",
       "      <td>0.9056472</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>80</td>\n",
       "      <td>20122.619141</td>\n",
       "      <td>20012.076172</td>\n",
       "      <td>17100.838</td>\n",
       "      <td>2254.3806</td>\n",
       "      <td>-0.0045349495</td>\n",
       "      <td>0.9066536</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>81</td>\n",
       "      <td>20158.541016</td>\n",
       "      <td>19991.625000</td>\n",
       "      <td>17102.855</td>\n",
       "      <td>2230.331</td>\n",
       "      <td>-0.0044782204</td>\n",
       "      <td>0.90763974</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>82</td>\n",
       "      <td>20127.789062</td>\n",
       "      <td>20033.273438</td>\n",
       "      <td>17106.3</td>\n",
       "      <td>2206.7864</td>\n",
       "      <td>-0.004435106</td>\n",
       "      <td>0.9086035</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>83</td>\n",
       "      <td>20023.490234</td>\n",
       "      <td>19867.888672</td>\n",
       "      <td>17106.5</td>\n",
       "      <td>2183.9763</td>\n",
       "      <td>-0.00439712</td>\n",
       "      <td>0.9095389</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84</td>\n",
       "      <td>20013.205078</td>\n",
       "      <td>19907.103516</td>\n",
       "      <td>17108.496</td>\n",
       "      <td>2161.5864</td>\n",
       "      <td>-0.004353958</td>\n",
       "      <td>0.91044563</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>85</td>\n",
       "      <td>19914.453125</td>\n",
       "      <td>19852.664062</td>\n",
       "      <td>17109.08</td>\n",
       "      <td>2139.8416</td>\n",
       "      <td>-0.00431451</td>\n",
       "      <td>0.9113326</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>86</td>\n",
       "      <td>20089.685547</td>\n",
       "      <td>19801.031250</td>\n",
       "      <td>17107.775</td>\n",
       "      <td>2118.7905</td>\n",
       "      <td>-0.004275813</td>\n",
       "      <td>0.9121925</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>87</td>\n",
       "      <td>19930.169922</td>\n",
       "      <td>19768.681641</td>\n",
       "      <td>17107.809</td>\n",
       "      <td>2098.0627</td>\n",
       "      <td>-0.004226692</td>\n",
       "      <td>0.91303736</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>88</td>\n",
       "      <td>19982.515625</td>\n",
       "      <td>19715.812500</td>\n",
       "      <td>17106.236</td>\n",
       "      <td>2077.9673</td>\n",
       "      <td>-0.004192111</td>\n",
       "      <td>0.91385835</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>89</td>\n",
       "      <td>19813.986328</td>\n",
       "      <td>19773.769531</td>\n",
       "      <td>17103.986</td>\n",
       "      <td>2058.5312</td>\n",
       "      <td>-0.0041462416</td>\n",
       "      <td>0.9146523</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90</td>\n",
       "      <td>19857.511719</td>\n",
       "      <td>19785.775391</td>\n",
       "      <td>17102.982</td>\n",
       "      <td>2039.4424</td>\n",
       "      <td>-0.004121549</td>\n",
       "      <td>0.9154318</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>91</td>\n",
       "      <td>19769.187500</td>\n",
       "      <td>19613.937500</td>\n",
       "      <td>17101.045</td>\n",
       "      <td>2020.708</td>\n",
       "      <td>-0.004082445</td>\n",
       "      <td>0.916191</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>92</td>\n",
       "      <td>19787.205078</td>\n",
       "      <td>19587.144531</td>\n",
       "      <td>17097.314</td>\n",
       "      <td>2002.6293</td>\n",
       "      <td>-0.004042237</td>\n",
       "      <td>0.9169333</td>\n",
       "      <td>01:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93</td>\n",
       "      <td>19708.876953</td>\n",
       "      <td>19531.421875</td>\n",
       "      <td>17091.287</td>\n",
       "      <td>1985.2267</td>\n",
       "      <td>-0.0040047406</td>\n",
       "      <td>0.917653</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>94</td>\n",
       "      <td>19631.658203</td>\n",
       "      <td>19517.750000</td>\n",
       "      <td>17086.738</td>\n",
       "      <td>1968.0547</td>\n",
       "      <td>-0.0039572776</td>\n",
       "      <td>0.9183534</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>95</td>\n",
       "      <td>19592.720703</td>\n",
       "      <td>19445.039062</td>\n",
       "      <td>17082.096</td>\n",
       "      <td>1951.2292</td>\n",
       "      <td>-0.0039250813</td>\n",
       "      <td>0.91904163</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96</td>\n",
       "      <td>19480.695312</td>\n",
       "      <td>19389.099609</td>\n",
       "      <td>17077.025</td>\n",
       "      <td>1934.8057</td>\n",
       "      <td>-0.0038821597</td>\n",
       "      <td>0.9197121</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>97</td>\n",
       "      <td>19457.406250</td>\n",
       "      <td>19334.800781</td>\n",
       "      <td>17071.176</td>\n",
       "      <td>1918.8331</td>\n",
       "      <td>-0.0038488135</td>\n",
       "      <td>0.92036396</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>98</td>\n",
       "      <td>19416.638672</td>\n",
       "      <td>19226.548828</td>\n",
       "      <td>17064.285</td>\n",
       "      <td>1903.2653</td>\n",
       "      <td>-0.0038178246</td>\n",
       "      <td>0.92100054</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>99</td>\n",
       "      <td>19356.769531</td>\n",
       "      <td>19266.505859</td>\n",
       "      <td>17058.31</td>\n",
       "      <td>1888.0242</td>\n",
       "      <td>-0.0037867427</td>\n",
       "      <td>0.9216238</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>100</td>\n",
       "      <td>19270.423828</td>\n",
       "      <td>19141.234375</td>\n",
       "      <td>17050.16</td>\n",
       "      <td>1873.3195</td>\n",
       "      <td>-0.0037406606</td>\n",
       "      <td>0.922233</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>101</td>\n",
       "      <td>19172.871094</td>\n",
       "      <td>19078.400391</td>\n",
       "      <td>17042.297</td>\n",
       "      <td>1858.8717</td>\n",
       "      <td>-0.0037236111</td>\n",
       "      <td>0.92282194</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>102</td>\n",
       "      <td>19169.808594</td>\n",
       "      <td>19078.656250</td>\n",
       "      <td>17033.83</td>\n",
       "      <td>1844.9045</td>\n",
       "      <td>-0.0037069062</td>\n",
       "      <td>0.92339474</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>103</td>\n",
       "      <td>19139.355469</td>\n",
       "      <td>18980.265625</td>\n",
       "      <td>17024.934</td>\n",
       "      <td>1831.24</td>\n",
       "      <td>-0.0036847726</td>\n",
       "      <td>0.9239571</td>\n",
       "      <td>01:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104</td>\n",
       "      <td>19066.394531</td>\n",
       "      <td>18882.779297</td>\n",
       "      <td>17014.912</td>\n",
       "      <td>1817.987</td>\n",
       "      <td>-0.0036623117</td>\n",
       "      <td>0.92450523</td>\n",
       "      <td>01:43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>105</td>\n",
       "      <td>18934.972656</td>\n",
       "      <td>18823.056641</td>\n",
       "      <td>17005.463</td>\n",
       "      <td>1804.9253</td>\n",
       "      <td>-0.0036469072</td>\n",
       "      <td>0.9250421</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>106</td>\n",
       "      <td>18888.779297</td>\n",
       "      <td>18814.871094</td>\n",
       "      <td>16995.076</td>\n",
       "      <td>1792.3789</td>\n",
       "      <td>-0.003620638</td>\n",
       "      <td>0.9255588</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>107</td>\n",
       "      <td>18872.443359</td>\n",
       "      <td>18723.216797</td>\n",
       "      <td>16984.473</td>\n",
       "      <td>1780.0948</td>\n",
       "      <td>-0.0035956176</td>\n",
       "      <td>0.92606455</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>108</td>\n",
       "      <td>18768.080078</td>\n",
       "      <td>18663.099609</td>\n",
       "      <td>16972.639</td>\n",
       "      <td>1768.2991</td>\n",
       "      <td>-0.0035878092</td>\n",
       "      <td>0.9265471</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>109</td>\n",
       "      <td>18793.630859</td>\n",
       "      <td>18542.955078</td>\n",
       "      <td>16960.518</td>\n",
       "      <td>1756.7299</td>\n",
       "      <td>-0.0035642167</td>\n",
       "      <td>0.927024</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>110</td>\n",
       "      <td>18578.509766</td>\n",
       "      <td>18480.535156</td>\n",
       "      <td>16948.527</td>\n",
       "      <td>1745.3951</td>\n",
       "      <td>-0.0035276802</td>\n",
       "      <td>0.9274865</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>111</td>\n",
       "      <td>18524.982422</td>\n",
       "      <td>18383.712891</td>\n",
       "      <td>16935.982</td>\n",
       "      <td>1734.367</td>\n",
       "      <td>-0.0035062758</td>\n",
       "      <td>0.92794096</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>112</td>\n",
       "      <td>18415.369141</td>\n",
       "      <td>18298.150391</td>\n",
       "      <td>16922.78</td>\n",
       "      <td>1723.6852</td>\n",
       "      <td>-0.0034913681</td>\n",
       "      <td>0.92837816</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>113</td>\n",
       "      <td>18342.421875</td>\n",
       "      <td>18227.941406</td>\n",
       "      <td>16909.57</td>\n",
       "      <td>1713.2476</td>\n",
       "      <td>-0.0034765091</td>\n",
       "      <td>0.9288029</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>114</td>\n",
       "      <td>18343.109375</td>\n",
       "      <td>18186.923828</td>\n",
       "      <td>16895.057</td>\n",
       "      <td>1703.3733</td>\n",
       "      <td>-0.0034624354</td>\n",
       "      <td>0.92921025</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>115</td>\n",
       "      <td>18263.548828</td>\n",
       "      <td>18149.833984</td>\n",
       "      <td>16880.76</td>\n",
       "      <td>1693.7592</td>\n",
       "      <td>-0.0034501299</td>\n",
       "      <td>0.9296092</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>116</td>\n",
       "      <td>18139.839844</td>\n",
       "      <td>18006.759766</td>\n",
       "      <td>16865.697</td>\n",
       "      <td>1684.4152</td>\n",
       "      <td>-0.00341939</td>\n",
       "      <td>0.92999905</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>117</td>\n",
       "      <td>18050.857422</td>\n",
       "      <td>18124.111328</td>\n",
       "      <td>16852.023</td>\n",
       "      <td>1675.3538</td>\n",
       "      <td>-0.0033906316</td>\n",
       "      <td>0.93037724</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>118</td>\n",
       "      <td>18154.996094</td>\n",
       "      <td>18081.177734</td>\n",
       "      <td>16839.795</td>\n",
       "      <td>1666.2411</td>\n",
       "      <td>-0.0033749575</td>\n",
       "      <td>0.9307454</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>119</td>\n",
       "      <td>17966.539062</td>\n",
       "      <td>18203.455078</td>\n",
       "      <td>16829.125</td>\n",
       "      <td>1657.3602</td>\n",
       "      <td>-0.0033552516</td>\n",
       "      <td>0.9311075</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>120</td>\n",
       "      <td>17944.406250</td>\n",
       "      <td>17761.701172</td>\n",
       "      <td>16814.705</td>\n",
       "      <td>1648.8717</td>\n",
       "      <td>-0.0033477028</td>\n",
       "      <td>0.9314516</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>121</td>\n",
       "      <td>17739.458984</td>\n",
       "      <td>18053.535156</td>\n",
       "      <td>16803.082</td>\n",
       "      <td>1640.6495</td>\n",
       "      <td>-0.0033367411</td>\n",
       "      <td>0.9317896</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>122</td>\n",
       "      <td>17736.500000</td>\n",
       "      <td>17607.421875</td>\n",
       "      <td>16787.95</td>\n",
       "      <td>1632.7753</td>\n",
       "      <td>-0.0033227007</td>\n",
       "      <td>0.932117</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>123</td>\n",
       "      <td>17712.046875</td>\n",
       "      <td>17550.888672</td>\n",
       "      <td>16772.805</td>\n",
       "      <td>1625.1687</td>\n",
       "      <td>-0.0033113246</td>\n",
       "      <td>0.93242866</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>124</td>\n",
       "      <td>17557.808594</td>\n",
       "      <td>17467.572266</td>\n",
       "      <td>16758.068</td>\n",
       "      <td>1617.6489</td>\n",
       "      <td>-0.0032924442</td>\n",
       "      <td>0.93273747</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>125</td>\n",
       "      <td>17438.033203</td>\n",
       "      <td>17326.886719</td>\n",
       "      <td>16741.7</td>\n",
       "      <td>1610.6792</td>\n",
       "      <td>-0.003266797</td>\n",
       "      <td>0.9330314</td>\n",
       "      <td>01:38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>126</td>\n",
       "      <td>17258.177734</td>\n",
       "      <td>17190.513672</td>\n",
       "      <td>16724.734</td>\n",
       "      <td>1603.9769</td>\n",
       "      <td>-0.0032547512</td>\n",
       "      <td>0.9333096</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>127</td>\n",
       "      <td>17201.332031</td>\n",
       "      <td>17085.500000</td>\n",
       "      <td>16707.39</td>\n",
       "      <td>1597.5535</td>\n",
       "      <td>-0.003241909</td>\n",
       "      <td>0.93358284</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>128</td>\n",
       "      <td>17073.199219</td>\n",
       "      <td>16900.605469</td>\n",
       "      <td>16688.957</td>\n",
       "      <td>1591.4558</td>\n",
       "      <td>-0.0032302903</td>\n",
       "      <td>0.9338359</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>129</td>\n",
       "      <td>16989.226562</td>\n",
       "      <td>16815.781250</td>\n",
       "      <td>16669.938</td>\n",
       "      <td>1585.7789</td>\n",
       "      <td>-0.003213283</td>\n",
       "      <td>0.9340715</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>130</td>\n",
       "      <td>16877.552734</td>\n",
       "      <td>16673.632812</td>\n",
       "      <td>16650.287</td>\n",
       "      <td>1580.4008</td>\n",
       "      <td>-0.00319285</td>\n",
       "      <td>0.93429697</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>131</td>\n",
       "      <td>16776.291016</td>\n",
       "      <td>16521.103516</td>\n",
       "      <td>16629.246</td>\n",
       "      <td>1575.5956</td>\n",
       "      <td>-0.0031881647</td>\n",
       "      <td>0.93450487</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>132</td>\n",
       "      <td>16744.335938</td>\n",
       "      <td>16426.525391</td>\n",
       "      <td>16608.637</td>\n",
       "      <td>1570.8386</td>\n",
       "      <td>-0.003175275</td>\n",
       "      <td>0.93469346</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>133</td>\n",
       "      <td>16456.824219</td>\n",
       "      <td>16248.571289</td>\n",
       "      <td>16586.188</td>\n",
       "      <td>1566.7955</td>\n",
       "      <td>-0.0031688898</td>\n",
       "      <td>0.9348812</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>134</td>\n",
       "      <td>16367.934570</td>\n",
       "      <td>16395.505859</td>\n",
       "      <td>16565.451</td>\n",
       "      <td>1563.02</td>\n",
       "      <td>-0.0031528769</td>\n",
       "      <td>0.93506306</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>135</td>\n",
       "      <td>16241.005859</td>\n",
       "      <td>16172.378906</td>\n",
       "      <td>16543.584</td>\n",
       "      <td>1559.5565</td>\n",
       "      <td>-0.0031412542</td>\n",
       "      <td>0.9351982</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>136</td>\n",
       "      <td>16182.583984</td>\n",
       "      <td>16088.879883</td>\n",
       "      <td>16521.904</td>\n",
       "      <td>1556.2766</td>\n",
       "      <td>-0.0031216103</td>\n",
       "      <td>0.9353312</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>137</td>\n",
       "      <td>16012.148438</td>\n",
       "      <td>16311.713867</td>\n",
       "      <td>16502.197</td>\n",
       "      <td>1553.3801</td>\n",
       "      <td>-0.0031095922</td>\n",
       "      <td>0.93544674</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>138</td>\n",
       "      <td>15918.783203</td>\n",
       "      <td>15927.232422</td>\n",
       "      <td>16480.596</td>\n",
       "      <td>1550.6182</td>\n",
       "      <td>-0.0030917362</td>\n",
       "      <td>0.93546915</td>\n",
       "      <td>01:42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>139</td>\n",
       "      <td>15871.909180</td>\n",
       "      <td>15689.766602</td>\n",
       "      <td>16457.814</td>\n",
       "      <td>1548.1765</td>\n",
       "      <td>-0.003091759</td>\n",
       "      <td>0.93555516</td>\n",
       "      <td>01:43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>140</td>\n",
       "      <td>15728.170898</td>\n",
       "      <td>15899.561523</td>\n",
       "      <td>16437.22</td>\n",
       "      <td>1545.9406</td>\n",
       "      <td>-0.0030829664</td>\n",
       "      <td>0.93537766</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>141</td>\n",
       "      <td>15761.555664</td>\n",
       "      <td>15755.603516</td>\n",
       "      <td>16416.785</td>\n",
       "      <td>1543.6155</td>\n",
       "      <td>-0.0030667535</td>\n",
       "      <td>0.93535554</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>142</td>\n",
       "      <td>15863.291016</td>\n",
       "      <td>15825.731445</td>\n",
       "      <td>16396.91</td>\n",
       "      <td>1541.8397</td>\n",
       "      <td>-0.0030461675</td>\n",
       "      <td>0.93507135</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>143</td>\n",
       "      <td>15981.112305</td>\n",
       "      <td>16078.201172</td>\n",
       "      <td>16379.194</td>\n",
       "      <td>1540.3997</td>\n",
       "      <td>-0.0030298333</td>\n",
       "      <td>0.9347635</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>144</td>\n",
       "      <td>15530.649414</td>\n",
       "      <td>15534.984375</td>\n",
       "      <td>16358.285</td>\n",
       "      <td>1539.2083</td>\n",
       "      <td>-0.0030311088</td>\n",
       "      <td>0.9346925</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>145</td>\n",
       "      <td>15363.589844</td>\n",
       "      <td>15180.150391</td>\n",
       "      <td>16335.051</td>\n",
       "      <td>1538.5768</td>\n",
       "      <td>-0.0030286617</td>\n",
       "      <td>0.9347226</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>146</td>\n",
       "      <td>15138.238281</td>\n",
       "      <td>15033.293945</td>\n",
       "      <td>16311.587</td>\n",
       "      <td>1538.056</td>\n",
       "      <td>-0.00302844</td>\n",
       "      <td>0.9347428</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>147</td>\n",
       "      <td>15152.629883</td>\n",
       "      <td>14887.298828</td>\n",
       "      <td>16287.246</td>\n",
       "      <td>1538.109</td>\n",
       "      <td>-0.0030300014</td>\n",
       "      <td>0.93474674</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>148</td>\n",
       "      <td>14981.999023</td>\n",
       "      <td>14945.657227</td>\n",
       "      <td>16263.677</td>\n",
       "      <td>1538.5194</td>\n",
       "      <td>-0.0030336867</td>\n",
       "      <td>0.9347553</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149</td>\n",
       "      <td>14809.250000</td>\n",
       "      <td>14624.293945</td>\n",
       "      <td>16238.288</td>\n",
       "      <td>1539.3374</td>\n",
       "      <td>-0.00302404</td>\n",
       "      <td>0.93476117</td>\n",
       "      <td>01:39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>150</td>\n",
       "      <td>14798.066406</td>\n",
       "      <td>14552.679688</td>\n",
       "      <td>16212.612</td>\n",
       "      <td>1540.6681</td>\n",
       "      <td>-0.0030269728</td>\n",
       "      <td>0.93471056</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>151</td>\n",
       "      <td>14683.558594</td>\n",
       "      <td>14446.177734</td>\n",
       "      <td>16186.741</td>\n",
       "      <td>1542.2103</td>\n",
       "      <td>-0.0030320382</td>\n",
       "      <td>0.9346828</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>152</td>\n",
       "      <td>14535.128906</td>\n",
       "      <td>14341.307617</td>\n",
       "      <td>16160.629</td>\n",
       "      <td>1543.9924</td>\n",
       "      <td>-0.0030240507</td>\n",
       "      <td>0.9346369</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>153</td>\n",
       "      <td>14447.378906</td>\n",
       "      <td>14234.134766</td>\n",
       "      <td>16134.368</td>\n",
       "      <td>1545.8751</td>\n",
       "      <td>-0.0030176481</td>\n",
       "      <td>0.9345828</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>154</td>\n",
       "      <td>14343.466797</td>\n",
       "      <td>14361.072266</td>\n",
       "      <td>16109.218</td>\n",
       "      <td>1548.0376</td>\n",
       "      <td>-0.0030163443</td>\n",
       "      <td>0.9345233</td>\n",
       "      <td>01:40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>155</td>\n",
       "      <td>14291.593750</td>\n",
       "      <td>14138.340820</td>\n",
       "      <td>16082.942</td>\n",
       "      <td>1550.4508</td>\n",
       "      <td>-0.0030166986</td>\n",
       "      <td>0.9344637</td>\n",
       "      <td>01:46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>156</td>\n",
       "      <td>14221.550781</td>\n",
       "      <td>14043.787109</td>\n",
       "      <td>16056.433</td>\n",
       "      <td>1552.9888</td>\n",
       "      <td>-0.0030122595</td>\n",
       "      <td>0.93439597</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>157</td>\n",
       "      <td>14199.510742</td>\n",
       "      <td>14016.631836</td>\n",
       "      <td>16030.061</td>\n",
       "      <td>1555.6602</td>\n",
       "      <td>-0.0030137098</td>\n",
       "      <td>0.9343231</td>\n",
       "      <td>03:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>158</td>\n",
       "      <td>14175.935547</td>\n",
       "      <td>13967.879883</td>\n",
       "      <td>16003.479</td>\n",
       "      <td>1558.6143</td>\n",
       "      <td>-0.0030278438</td>\n",
       "      <td>0.9342464</td>\n",
       "      <td>03:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>159</td>\n",
       "      <td>14099.639648</td>\n",
       "      <td>13963.538086</td>\n",
       "      <td>15977.093</td>\n",
       "      <td>1561.6681</td>\n",
       "      <td>-0.0030275146</td>\n",
       "      <td>0.9341616</td>\n",
       "      <td>03:04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>160</td>\n",
       "      <td>14102.060547</td>\n",
       "      <td>13935.328125</td>\n",
       "      <td>15950.781</td>\n",
       "      <td>1564.764</td>\n",
       "      <td>-0.0030235897</td>\n",
       "      <td>0.9340875</td>\n",
       "      <td>03:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>161</td>\n",
       "      <td>14114.378906</td>\n",
       "      <td>13936.639648</td>\n",
       "      <td>15924.911</td>\n",
       "      <td>1567.7129</td>\n",
       "      <td>-0.0030207487</td>\n",
       "      <td>0.9340103</td>\n",
       "      <td>03:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>162</td>\n",
       "      <td>14084.943359</td>\n",
       "      <td>13925.353516</td>\n",
       "      <td>15899.29</td>\n",
       "      <td>1570.6273</td>\n",
       "      <td>-0.0030277981</td>\n",
       "      <td>0.9339443</td>\n",
       "      <td>03:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>163</td>\n",
       "      <td>14066.994141</td>\n",
       "      <td>13870.553711</td>\n",
       "      <td>15873.564</td>\n",
       "      <td>1573.5887</td>\n",
       "      <td>-0.0030303998</td>\n",
       "      <td>0.9338654</td>\n",
       "      <td>03:07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>164</td>\n",
       "      <td>14097.787109</td>\n",
       "      <td>13878.373047</td>\n",
       "      <td>15847.908</td>\n",
       "      <td>1576.8058</td>\n",
       "      <td>-0.003037305</td>\n",
       "      <td>0.9338005</td>\n",
       "      <td>02:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>165</td>\n",
       "      <td>14044.483398</td>\n",
       "      <td>13941.391602</td>\n",
       "      <td>15822.646</td>\n",
       "      <td>1580.2726</td>\n",
       "      <td>-0.0030345093</td>\n",
       "      <td>0.93353575</td>\n",
       "      <td>01:51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>166</td>\n",
       "      <td>14125.006836</td>\n",
       "      <td>13955.373047</td>\n",
       "      <td>15797.85</td>\n",
       "      <td>1583.6216</td>\n",
       "      <td>-0.003034405</td>\n",
       "      <td>0.9332654</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>167</td>\n",
       "      <td>14133.144531</td>\n",
       "      <td>13980.307617</td>\n",
       "      <td>15773.641</td>\n",
       "      <td>1586.7803</td>\n",
       "      <td>-0.0030331023</td>\n",
       "      <td>0.9329991</td>\n",
       "      <td>01:56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>168</td>\n",
       "      <td>14467.569336</td>\n",
       "      <td>14242.233398</td>\n",
       "      <td>15750.941</td>\n",
       "      <td>1590.2166</td>\n",
       "      <td>-0.003027069</td>\n",
       "      <td>0.9323911</td>\n",
       "      <td>02:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>169</td>\n",
       "      <td>14222.583984</td>\n",
       "      <td>13992.085938</td>\n",
       "      <td>15727.414</td>\n",
       "      <td>1593.2498</td>\n",
       "      <td>-0.0030269923</td>\n",
       "      <td>0.9321335</td>\n",
       "      <td>01:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>170</td>\n",
       "      <td>14202.051758</td>\n",
       "      <td>14021.025391</td>\n",
       "      <td>15704.404</td>\n",
       "      <td>1596.1738</td>\n",
       "      <td>-0.0030300291</td>\n",
       "      <td>0.9318819</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171</td>\n",
       "      <td>14359.071289</td>\n",
       "      <td>14195.049805</td>\n",
       "      <td>15681.975</td>\n",
       "      <td>1599.7548</td>\n",
       "      <td>-0.0030233227</td>\n",
       "      <td>0.9312965</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>172</td>\n",
       "      <td>14422.250000</td>\n",
       "      <td>14339.732422</td>\n",
       "      <td>15660.803</td>\n",
       "      <td>1603.1221</td>\n",
       "      <td>-0.0030168903</td>\n",
       "      <td>0.93072075</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>173</td>\n",
       "      <td>14238.473633</td>\n",
       "      <td>14032.426758</td>\n",
       "      <td>15638.833</td>\n",
       "      <td>1605.7462</td>\n",
       "      <td>-0.003014662</td>\n",
       "      <td>0.9304859</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>174</td>\n",
       "      <td>14111.084961</td>\n",
       "      <td>13951.729492</td>\n",
       "      <td>15616.648</td>\n",
       "      <td>1608.3478</td>\n",
       "      <td>-0.0030136402</td>\n",
       "      <td>0.9303997</td>\n",
       "      <td>01:48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>175</td>\n",
       "      <td>14187.177734</td>\n",
       "      <td>13969.001953</td>\n",
       "      <td>15594.576</td>\n",
       "      <td>1611.1558</td>\n",
       "      <td>-0.0030116101</td>\n",
       "      <td>0.930162</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>176</td>\n",
       "      <td>14167.900391</td>\n",
       "      <td>14022.910156</td>\n",
       "      <td>15573.137</td>\n",
       "      <td>1613.8502</td>\n",
       "      <td>-0.003001807</td>\n",
       "      <td>0.9299286</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>177</td>\n",
       "      <td>14150.501953</td>\n",
       "      <td>13952.441406</td>\n",
       "      <td>15551.656</td>\n",
       "      <td>1616.403</td>\n",
       "      <td>-0.002999605</td>\n",
       "      <td>0.9298403</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>178</td>\n",
       "      <td>14080.500977</td>\n",
       "      <td>13905.959961</td>\n",
       "      <td>15530.121</td>\n",
       "      <td>1618.9662</td>\n",
       "      <td>-0.0029941346</td>\n",
       "      <td>0.9297591</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>179</td>\n",
       "      <td>14824.177734</td>\n",
       "      <td>14519.208008</td>\n",
       "      <td>15511.172</td>\n",
       "      <td>1622.5273</td>\n",
       "      <td>-0.0029865196</td>\n",
       "      <td>0.92901516</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>180</td>\n",
       "      <td>14464.829102</td>\n",
       "      <td>14293.110352</td>\n",
       "      <td>15491.576</td>\n",
       "      <td>1625.6682</td>\n",
       "      <td>-0.002983556</td>\n",
       "      <td>0.9284702</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>181</td>\n",
       "      <td>14465.952148</td>\n",
       "      <td>14288.891602</td>\n",
       "      <td>15472.171</td>\n",
       "      <td>1628.7756</td>\n",
       "      <td>-0.0029732212</td>\n",
       "      <td>0.927933</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>182</td>\n",
       "      <td>14247.627930</td>\n",
       "      <td>14053.854492</td>\n",
       "      <td>15452.404</td>\n",
       "      <td>1631.1522</td>\n",
       "      <td>-0.002964693</td>\n",
       "      <td>0.92771935</td>\n",
       "      <td>01:58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>183</td>\n",
       "      <td>14145.791016</td>\n",
       "      <td>13956.905273</td>\n",
       "      <td>15432.427</td>\n",
       "      <td>1633.4037</td>\n",
       "      <td>-0.0029585822</td>\n",
       "      <td>0.92765176</td>\n",
       "      <td>03:01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>184</td>\n",
       "      <td>14138.794922</td>\n",
       "      <td>13919.940430</td>\n",
       "      <td>15412.342</td>\n",
       "      <td>1635.7607</td>\n",
       "      <td>-0.002955253</td>\n",
       "      <td>0.9275816</td>\n",
       "      <td>03:15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>185</td>\n",
       "      <td>14113.723633</td>\n",
       "      <td>13919.291016</td>\n",
       "      <td>15392.483</td>\n",
       "      <td>1638.0751</td>\n",
       "      <td>-0.0029475112</td>\n",
       "      <td>0.92751235</td>\n",
       "      <td>03:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>186</td>\n",
       "      <td>14110.894531</td>\n",
       "      <td>13908.471680</td>\n",
       "      <td>15372.754</td>\n",
       "      <td>1640.3948</td>\n",
       "      <td>-0.002945114</td>\n",
       "      <td>0.9274422</td>\n",
       "      <td>03:05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>187</td>\n",
       "      <td>14111.275391</td>\n",
       "      <td>13897.860352</td>\n",
       "      <td>15353.1</td>\n",
       "      <td>1642.7655</td>\n",
       "      <td>-0.0029441798</td>\n",
       "      <td>0.9273724</td>\n",
       "      <td>03:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>188</td>\n",
       "      <td>14234.290039</td>\n",
       "      <td>14112.270508</td>\n",
       "      <td>15334.29</td>\n",
       "      <td>1645.5983</td>\n",
       "      <td>-0.0029354824</td>\n",
       "      <td>0.92691827</td>\n",
       "      <td>03:05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>189</td>\n",
       "      <td>14160.514648</td>\n",
       "      <td>13984.416016</td>\n",
       "      <td>15315.729</td>\n",
       "      <td>1647.6866</td>\n",
       "      <td>-0.0029294412</td>\n",
       "      <td>0.9267126</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>190</td>\n",
       "      <td>14095.900391</td>\n",
       "      <td>13920.120117</td>\n",
       "      <td>15297.04</td>\n",
       "      <td>1649.7379</td>\n",
       "      <td>-0.0029263112</td>\n",
       "      <td>0.9266477</td>\n",
       "      <td>01:48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>191</td>\n",
       "      <td>14161.728516</td>\n",
       "      <td>13979.683594</td>\n",
       "      <td>15278.7</td>\n",
       "      <td>1651.924</td>\n",
       "      <td>-0.0029222162</td>\n",
       "      <td>0.9264501</td>\n",
       "      <td>01:47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>192</td>\n",
       "      <td>14183.171875</td>\n",
       "      <td>13953.120117</td>\n",
       "      <td>15260.413</td>\n",
       "      <td>1654.088</td>\n",
       "      <td>-0.00291863</td>\n",
       "      <td>0.9263072</td>\n",
       "      <td>01:48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>193</td>\n",
       "      <td>14128.863281</td>\n",
       "      <td>13926.860352</td>\n",
       "      <td>15242.286</td>\n",
       "      <td>1656.1229</td>\n",
       "      <td>-0.0029151163</td>\n",
       "      <td>0.92625594</td>\n",
       "      <td>01:47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>194</td>\n",
       "      <td>14124.221680</td>\n",
       "      <td>13919.476562</td>\n",
       "      <td>15224.254</td>\n",
       "      <td>1658.1915</td>\n",
       "      <td>-0.0029140771</td>\n",
       "      <td>0.9262011</td>\n",
       "      <td>01:48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>195</td>\n",
       "      <td>14144.415039</td>\n",
       "      <td>13912.543945</td>\n",
       "      <td>15206.359</td>\n",
       "      <td>1660.2495</td>\n",
       "      <td>-0.002910434</td>\n",
       "      <td>0.9261203</td>\n",
       "      <td>01:52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>196</td>\n",
       "      <td>14076.261719</td>\n",
       "      <td>13940.631836</td>\n",
       "      <td>15188.755</td>\n",
       "      <td>1662.3212</td>\n",
       "      <td>-0.0029069763</td>\n",
       "      <td>0.9260272</td>\n",
       "      <td>01:50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>197</td>\n",
       "      <td>14118.953125</td>\n",
       "      <td>13925.870117</td>\n",
       "      <td>15171.247</td>\n",
       "      <td>1664.3755</td>\n",
       "      <td>-0.0029025748</td>\n",
       "      <td>0.9259105</td>\n",
       "      <td>01:49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>198</td>\n",
       "      <td>14119.511719</td>\n",
       "      <td>13947.337891</td>\n",
       "      <td>15154.008</td>\n",
       "      <td>1666.427</td>\n",
       "      <td>-0.0028982842</td>\n",
       "      <td>0.9257852</td>\n",
       "      <td>01:53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>199</td>\n",
       "      <td>14111.185547</td>\n",
       "      <td>13959.049805</td>\n",
       "      <td>15137.002</td>\n",
       "      <td>1668.4531</td>\n",
       "      <td>-0.002892826</td>\n",
       "      <td>0.9256596</td>\n",
       "      <td>01:56</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fit_one_cycle(200)\n",
    "\n",
    "learn.save(datetime.now().strftime('%Y-%m-%d %Hh%M.%S') + '.model')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x1368 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "b = learn.dls[0].one_batch()\n",
    "input_img, nums, targets = to_detach(b)\n",
    "preds, targs = learn.get_preds(dl=[b])\n",
    "imgs, _, _, _ = preds\n",
    "\n",
    "_, axes = plt.subplots(ncols=1, nrows=10, constrained_layout=True, figsize=(10, 19))\n",
    "i = 0\n",
    "for a in axes:\n",
    "    a.set_title(nums[i].cpu().item())\n",
    "    img = torch.cat([TensorBase(input_img.clamp(0,1)), TensorBase(targs), imgs], dim=3)[i].squeeze(0).cpu()\n",
    "    a.imshow(img, cmap='Greys')\n",
    "    a.axis('off')\n",
    "    i = i + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1368x1368 with 100 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %debug\n",
    "num = 10 if url == URLs.MNIST else 2\n",
    "_, axes = plt.subplots(ncols=num, nrows=10, figsize=(19,19))\n",
    "learn.model.eval()\n",
    "i = 0\n",
    "with torch.no_grad():\n",
    "    for row in axes:\n",
    "        z = torch.empty([num,128]).normal_().cuda()\n",
    "        img = learn.model.decode(z, torch.Tensor(range(num)).long().cuda()).detach().cpu()\n",
    "        for ax in row:\n",
    "            ax.imshow(img[i%num][0], cmap='Greys')\n",
    "            ax.axis('off')\n",
    "            ax.set_title(str(i%num))\n",
    "            i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDIAAABoCAYAAAAD89CiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAATOUlEQVR4nO3dSWxV5R/G8bdlaplRylDGQow4YYJGNwJRF7IABDYScALjPCfqou3SOdGF4EIXKqyMU1QGQwehRcAoRE2EFmWQVjTRUihoy9j+F/fvr7/3cM/l9A7nnLfn+1k9pfftvdyHc255877nFPX09BgAAAAAAAAXFEf9AgAAAAAAAIJiIgMAAAAAADiDiQwAAAAAAOAMJjIAAAAAAIAzmMgAAAAAAADOYCIDAAAAAAA4Y+Alvs+9WcNRlKefQ1/hoC+30Jdb8tEXXYWDY8st9OUW+nILfbmFvtySti9WZAAAAAAAAGcwkQEAAAAAAJzBRAYAAAAAAHAGExkAAAAAAMAZTGQAAAAAAABnMJEBAAAAAACcwUQGAAAAAABwBhMZAAAAAADAGUxkAAAAAAAAZwyM+gXEQU9PT9o/LyoqCvmVIAj6cgt9uYW+3EFXbqEvt9CXW+jLLfTllrj2xYoMAAAAAADgDCYyAAAAAACAM/rd1pLu7m7J58+fl9zV1WU97uTJk5LPnTsnecKECZJLSkoke5fORL2Upr+gL7fQl1voyx105ZagfXV0dEjWfU2cOFEyfRUefbnFr6/Ozk7rcX7nQ/oKV9C+/I6v8vJyyfRVeP2pL1ZkAAAAAAAAZzCRAQAAAAAAnOHs1hJ99dQLFy5IPnbsmORPPvlEcl1dnTX+t99+kzxwYO/bUFVVJXnBggWShwwZktsLTrggfX388ceS6+vrrfH0Fa6+9pXp+Bo0aJBk+ioM+nKTXtLZ1tYmOZtzYXV1tWS6KowgfXmPrSNHjkimr3D59fXRRx9J9h5fhw8fljx48GDJ9FV49OWWIH3V1tZaY/T5kL7C5dfXhx9+KNl7fLnQFysyAAAAAACAM5jIAAAAAAAAznB2a4m+4urBgwclv/3225L1kia9jMYY++qreinM1q1bJc+dOzftY9B3QfrasmWLZL0k3hh7SRR9FR59uUX3deDAAcm6r5qaGslB+9q2bZvkefPmpX0M+kZv/aGr+KMvt/j1tXbtWsm6r/b2dms8fYWLvtxSqL4aGhok01f++PW1Zs0ayfr/yt7PLz0+rn2xIgMAAAAAADiDiQwAAAAAAOAMZ7aW6OUtxhjzyy+/SH7qqack79u3T3JXV5dkfbVVY+wr8+usl9Xoq8OPHj3aGj9gwICgLz2Rgva1d+9eyadPn5bs7Ut/rfvSy9b01XXpq2/0cj9j7L6efvppyfQVD2H1pc+Huq9Ro0ZZ4+nLX6aunnzySclNTU2S6So6YfXldy6kr77x9rV//37J+neNfPbV0tIimb76hr7cku++9NYDfRcn/flFX9nL1Jf+/Gpubpacqa+SkhLJuq84HV+syAAAAAAAAM5gIgMAAAAAADiDiQwAAAAAAOCMWF8jQ+/12bhxo/W9yspKyXp/zvDhwyVfddVVkq+99lprfHFx7xzOqVOnJOs9QD/++KPka665xnd8UVFRhr9FcmTT14gRIyTn2tcPP/wg+eqrr/YdT18p+exr9uzZvs8T5Piir0vTfX355ZfW96qqqiS3trZK1n3p9/i6667zfZ4gx5fu3hj68ipkV/r9PXnypGS6yl4Ufem9wz/99JNk+rq0qPvSn12zZs2yxtPXxXRfX3zxhfW96upqyUH6yvS7RpC+OL4uLeq+OB/2TRR96R6i7osVGQAAAAAAwBlMZAAAAAAAAGfEbmtJd3e35O3bt0t+7LHHrMe1tbVJ1rd7WbRokeTVq1dLnj59ujW+s7NTsr712WeffSZ5w4YNkm+44QZrvN76oJdEJW2pU659LVy4UDJ9FV7UfX366aeS9ZJgb196Kxd9pei+nnjiCetxQc6Hq1atkpypL33baX180Vdm+exKH1vTpk2zxtNVfui+GhsbJUfR1+effy55zpw51ni97Je+UqLuSy/fpq/0/PrSt4A0xr+vxYsXS9afXZn6Onz4sGT66pt89qWPr6lTp1rjg/SV6Xyoty7QV0pDQ4Nk7/lQ39a2v/bFigwAAAAAAOAMJjIAAAAAAIAzYre15MyZM5Jff/11yXo5kzHGjBkzRrJeFv/AAw9ILisrk6yXtBhjX+V15MiRkvUSnZqaGsnvvvuuNV6/tmHDhqX7qySC7uu1116THFZf27Ztk0xflxanvmprayW/88471nj6SqEvd8Spqy1btkj2ngv1a0tqV8Zk97vG448/Llkvxw3al75S/NatWyXrvrzHFn2l0Jdb4tSX/t2QvtLTfb366quSg/alP7/Gjh0rOVNf+i6TQfvSr42+UvS/Yb2VxJjC9VVfXy856r5YkQEAAAAAAJzBRAYAAAAAAHBGLLaW9PT0SN65c6fk7777TvLAgfZLXblypeRnn31W8tChQyV7l8hoxcW9czh6OZq+IvLx48cl79q1yxp/4sQJyUlb3uTX1/fffy+5kH3p5U26r/b2dsnevjo6OiTTV0oUfekrItNXevTlDt3Vjh07JEfdlf580v+GjEluV8bk3tczzzwjOdffNejr0uLU1+TJkyVn6uvkyZOS6SslU18rVqyQXKi+9O/y9NXLr689e/ZILmRfemvkpEmTJOu+9Osyhr7+E3VfU6ZMkRx1X6zIAAAAAAAAzmAiAwAAAAAAOIOJDAAAAAAA4IxYXCPj9OnTkvWt4i5cuCB55syZ1hh96xi910fv59GKiop8n1/vDyovL0/7GL0HyPuakyZIXxUVFdaYfPal94AF7aurq8v35/V39OUW3Ze+lVWmvvSt6vLZ18SJE9M+hr5Ssjm2ou6Kz64U+oq/bPpatWqV5EJ9dum96pwLe8Xp+NLXXNDoq5eLfXE+TInT+VCLoi9WZAAAAAAAAGcwkQEAAAAAAJwRydYSvSzPGGOam5sl69s06aUrlZWV1hi9fCabJTL6e3q8Xnqjl+t4l8d4/w79WTZ9VVdXW2PC6Ku7u1syffVqamqSHHVf+vZL9JWSqa/du3dLztTXjBkzJOezL317T/oK3pVeJptNV5kEObYyfXbpHvu7OPeljy36SvH2tW/fPslB+9LbkAvVF+fClHz0Fcb5kL5S4txX0OOL82FKnM6HUf9fmRUZAAAAAADAGUxkAAAAAAAAZ8Ria8nXX38tubOzU/Ktt94qef78+dYYffXUTEumg9Djz5w5I/ncuXOS9Z1NjDFmyJAhOT2nS+Lc19mzZyXTV0rQvm677TbJYfWll53RV4orfeljbdCgQdaYkpKSnJ7TFd6u6uvrJeezq6AdBjm2Bg8ebI1JSlfG0JdrwjoXFrKvJH92ZXN86c+SfPal70ZCXylB+7r99tslZ+pL0+99IftK8vmwrq5Ocpz6On/+vO9rDuP4YkUGAAAAAABwBhMZAAAAAADAGZFsLdFXODXGmO3bt0vWy5dvvvlmyaNHj7bGeJei50IvY3r//ffTvpbZs2dbY8aNGyc516XccRe3vvRzvvfee2n//Prrr7fG0FeKfo9uuukmyWH19cEHH6T9c/rqFde+9HnyiiuusMaUlZVJ7s99ebtqbGyUrLclRnEu9OvqyiuvtMYkpStj6Ms13r4aGhok01f8xPn4Wr9+vWTd16xZs6wx9JWSTV+5vl+6r3Xr1knO1NfYsWPz9vxxl+/zYT77Cnp8hdEXKzIAAAAAAIAzmMgAAAAAAADOiGRrib7CqTHG/P7775L10pOKigrJ3iuvFhcXpx3jvWKqH/24bdu2Sd6wYUPax69cudL62nsl3f7M21dLS4tk/d7PmDFDclh9bdy4Me3j6atXa2ur5Cj62rp1q2S/vu6++27r6yRdSTxux5fua9OmTWkff++991pfJ6WvTJ9duoOwutJ3ddi8eXPax99zzz3W10npypj49aWvOs+xdbE49+V3fN13333W10nqSy8pNyZYX97fxfLZV21trWT6ulic+/rqq6/SPn7VqlXW1/SVEkVfNTU1kv36Wr16tfU1dy0BAAAAAABQmMgAAAAAAADOYCIDAAAAAAA4I5JrZBw/ftz6+q+//pJcWloquby8XHI2t2jKtAdIv4YHH3xQst6juXDhQsnefT9631F/5+2rra1N8rBhwySH1ddDDz0k2a8v7766JPelj68o+nr44Ycl674WLVok+f7777fG9/fbamnt7e3W13///bfkKPryOx8uXrxYsncff1L6OnbsmPV11F35nQt1V95rZCSlK2Pi19cjjzwiWd9a784775TsvV5QkvrSv1sY49/XpEmTJEfR15IlSyR7r8eVpL6CHl+6r2x+Fwva16OPPipZ97V06VLJK1assMbTV4rf+TDffenfd/yOL93X8uXLrfFJ6st7PtT9RdGX3/G1bNkyyVH0lZz/3QEAAAAAAOcxkQEAAAAAAJwR2tYSvXRl//791ve6urokT5kyRfLo0aMlZ1qeon+2znqp7T///GONueOOOyTr21NOnTpV8po1ayQn6ZY/xtjvY3Nzs/U9v75GjRolOYq+1q5dKznJfTU1NVnfO3PmjOSo+5o2bZpk+krJ1Jd+v8I6Hx49ejTt8yf1fJhNV2EdW3R1Mf0+7t271/penPqaPn265Lfeektykvvat2+f9b2zZ89K1u/XyJEjJdNXuDIdX359cXxFR7+PP//8s/U9fT6sqKiQnKkvv1t4Bu1rwYIFkv/444+0z09fKZn6mjlzpuQ49eW9/WsYWJEBAAAAAACcwUQGAAAAAABwRmhbS7q7uyV7lw+OGTNG8vz58yWPGDEi0M/2WyKzfv16yW+++aY15uDBg5L1FV8bGxslT5w4UXKSrpRrjDt9bd++XfKECRMk01evMPp64403rDGHDh2S7Hd8jR8/XnKS+/JuV9B9zZs3T/Lw4cMD/exc+9JXmNfHV1L7oiu3uNIX58IUV/rSx9e4ceMk01evQvW1bt06yd7fDXVfeuvsN998I7msrEwyffW67LLLJAftK8j2hGz60scXfaV4t/Xr7cVz586VHEVf+vgaO3as5Cj6YkUGAAAAAABwBhMZAAAAAADAGQXdWqKXsVy4cEHyuXPnrMfNmDFDsl5iqR+nl9t4v+7s7JRcW1sr+bnnnpPc0dFhjdfbEPQSGX0XjKQtafLrSy9BMqZwfT3//POST5w4YY3360svdaKvlGz60j/L+/Ny7WvHjh2S6SslaF96CXPQvvRdhXI9viZPniw5SX0F7UpfGT+KrvSxldSujPHvy/s5RF/xkE1ferl5Nn3V1dVJzqYvvc2EvlKC9qXvZpJNXy+88IJkb196C7jenqC3tNLXxX9uDH3FRdDjS98pRG/nCKsv/bthnC69wIoMAAAAAADgDCYyAAAAAACAM0K7a4le3qKvvGqMMcOGDZO8a9cuyYMHD5Z8yy23WGN2794tuaamRvK3334rWS8/1MtgjDFm06ZNkvXyqqiXyMRFFH3p5VHevjZv3iyZvi6m+9JXDjcm9770Emo9Pmhf06ZNk0xfKZn6Ki0tlUxf0dNd6au8GxP9sZXkrZB+gvaltw0MGjRIsr4avDH0VWiF7Ku+vj7tePrKnu7r8ssvt76XTV979uyRrJe701d+hNXXzp07JdNX9nRfevuIMfbdSfQ2j7D6iuvWcFZkAAAAAAAAZzCRAQAAAAAAnMFEBgAAAAAAcEZot1/VubjYnj85cuSI5EOHDknWe7a8+3H0LSH1/iB9C5/q6mrJd911lzVe7xWL016fKGXT18GDByXTV7joyy1+fQ0YMMB6XGtrq2S/vrz0bbXoK3d05ZZc+9LXu3jllVesMfSVf3596ffXmML1VVVVJXn58uXWePq6mPeWjv8pZF96rz599U3UfVVWVkqmr0vz60tfa8sY/770tQaT2BcrMgAAAAAAgDOYyAAAAAAAAM4I7fareomnvn2mMfYSF7215N9///X9eSUlJZKXLVsm+cUXX5Q8fvx4yd7l9siskH0tXbpUsu5rwoQJkumrb+jLLbovfbtTY4yZNGmSZPqKHl25JdO5MNe+lixZIvmll16STF/Z033p2zEaQ19xpN+voOfDU6dO+f68oUOHSqav/KOv+NNbNvLdV2lpqWTd18svvyzZ9b7ce8UAAAAAACCxmMgAAAAAAADOKPK7Wur/Zfzmpeif3d3dLdm7LFDfVeHXX3+V/Oeff0quqKiwxsyZM0dyWVmZZO9VysOm/859uMJrvi4FS199RF8p9BXs6XMaHLCvlpYWyQcOHJB89OhRyfR16afOaXAeu/IuD73xxhsl01Xv0+c0mL6CDnOqL32Vfv3ZRV99f/qcBvv0pe9uZox9fNFXbk+f02D6CjqMviKSz75YkQEAAAAAAJzBRAYAAAAAAHBGQbeWILBYLG9CYPTlFvpyS+RbSxAYx5Zb6Mst9OUW+nILfbmFrSUAAAAAAMBtTGQAAAAAAABnMJEBAAAAAACcwUQGAAAAAABwBhMZAAAAAADAGUxkAAAAAAAAZzCRAQAAAAAAnMFEBgAAAAAAcAYTGQAAAAAAwBlFPT09Ub8GAAAAAACAQFiRAQAAAAAAnMFEBgAAAAAAcAYTGQAAAAAAwBlMZAAAAAAAAGcwkQEAAAAAAJzBRAYAAAAAAHDG/wB/1Nd2zzGVPgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1368x1368 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %debug\n",
    "_, axes = plt.subplots(ncols=10, nrows=1, figsize=(19,19))\n",
    "space = np.linspace(0,1,10)\n",
    "i = 0\n",
    "with torch.no_grad():\n",
    "    z = torch.empty([1,128]).normal_().cuda()\n",
    "    for ax in axes:\n",
    "        z[0][6] = space[i]\n",
    "        img = learn.model.decode(z, torch.Tensor([0]).long().cuda()).detach().cpu()\n",
    "        ax.imshow(img[0][0], cmap='Greys')\n",
    "        ax.axis('off')\n",
    "        i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn.save('vae')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define the NN architecture\n",
    "class ConvAutoencoder(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(ConvAutoencoder, self).__init__()\n",
    "        ## encoder layers ##\n",
    "        # conv layer (depth from 1 --> 16), 3x3 kernels\n",
    "        self.conv1 = nn.Conv2d(1, 16, 3, padding=1)  \n",
    "        # conv layer (depth from 16 --> 4), 3x3 kernels\n",
    "        self.conv2 = nn.Conv2d(16, 4, 3, padding=1)\n",
    "        # pooling layer to reduce x-y dims by two; kernel and stride of 2\n",
    "        self.pool = nn.AvgPool2d(2, 2)\n",
    "        \n",
    "        ## decoder layers ##\n",
    "        self.u_conv1 = nn.Conv2d(4, 16, 3, padding=1)\n",
    "        self.u_conv2 = nn.Conv2d(16, 1, 3, padding=1)\n",
    "\n",
    "    def forward(self, x):\n",
    "        ## encode ##\n",
    "        # add hidden layers with relu activation function\n",
    "        # and maxpooling after\n",
    "        x = F.relu(self.conv1(x))\n",
    "        x = self.pool(x)\n",
    "        # add second hidden layer\n",
    "        x = F.relu(self.conv2(x))\n",
    "        x = self.pool(x)  # compressed representation\n",
    "        \n",
    "        \n",
    "        \n",
    "        ## decode ##\n",
    "        # upsample, followed by a conv layer, with relu activation function  \n",
    "        # this function is called `upsample` in some PyTorch versions\n",
    "        x = F.interpolate(x, scale_factor=2, mode='nearest')\n",
    "        x = F.relu(self.u_conv1(x))\n",
    "        # upsample again, output should have a sigmoid applied\n",
    "        x = F.interpolate(x, scale_factor=2, mode='nearest')\n",
    "        x = torch.sigmoid(self.u_conv2(x))\n",
    "                \n",
    "        return x\n",
    "\n",
    "# initialize the NN\n",
    "# model = ConvAutoencoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "> \u001b[1;32m<ipython-input-122-d78c38b6bdb1>\u001b[0m(61)\u001b[0;36mloss\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32m     59 \u001b[1;33m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m     60 \u001b[1;33m        \u001b[0mset_trace\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m---> 61 \u001b[1;33m        \u001b[0mbce\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfunctional\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbinary_cross_entropy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreconstruction\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mview\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m-\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput_size\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m     62 \u001b[1;33m        \u001b[0mKLD\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m-\u001b[0m\u001b[1;36m0.5\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mlogvar\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mmu\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpow\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mlogvar\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexp\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m     63 \u001b[1;33m        \u001b[1;31m# Normalise by same number of elements as in reconstruction\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\n"
     ]
    },
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "ipdb>  x.shape\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([5, 4])\n"
     ]
    },
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "ipdb>  x.view(-1, input_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[1., 0., 0., 0.],\n",
      "        [0., 0., 0., 1.],\n",
      "        [0., 1., 0., 0.],\n",
      "        [0., 0., 0., 1.],\n",
      "        [1., 0., 0., 0.]])\n"
     ]
    },
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "ipdb>  x.view(-1, input_size).shape\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([5, 4])\n",
      "--KeyboardInterrupt--\n",
      "\n",
      "KeyboardInterrupt: Interrupted by user\n",
      "> \u001b[1;32m<ipython-input-122-d78c38b6bdb1>\u001b[0m(61)\u001b[0;36mloss\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32m     59 \u001b[1;33m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m     60 \u001b[1;33m        \u001b[0mset_trace\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m---> 61 \u001b[1;33m        \u001b[0mbce\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfunctional\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbinary_cross_entropy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreconstruction\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mview\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m-\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput_size\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m     62 \u001b[1;33m        \u001b[0mKLD\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m-\u001b[0m\u001b[1;36m0.5\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mlogvar\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mmu\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpow\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mlogvar\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexp\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\u001b[1;32m     63 \u001b[1;33m        \u001b[1;31m# Normalise by same number of elements as in reconstruction\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[0m\n"
     ]
    },
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "ipdb>  q\n"
     ]
    },
    {
     "ename": "BdbQuit",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mBdbQuit\u001b[0m                                   Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-122-d78c38b6bdb1>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m    111\u001b[0m \u001b[0mprint_examples\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mFalse\u001b[0m \u001b[1;31m#Change to true if you want to see some examples at each step!\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    112\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m30\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 113\u001b[1;33m     \u001b[0mtrain\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mepoch\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    114\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mprint_examples\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    115\u001b[0m         \u001b[0mtest\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mepoch\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-122-d78c38b6bdb1>\u001b[0m in \u001b[0;36mtrain\u001b[1;34m(epoch, batches_per_epoch, log_interval)\u001b[0m\n\u001b[0;32m     82\u001b[0m         \u001b[0moptimizer\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mzero_grad\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     83\u001b[0m         \u001b[0mrecon_batch\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmu\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mlogvar\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 84\u001b[1;33m         \u001b[0mloss\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mloss\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrecon_batch\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmu\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mlogvar\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     85\u001b[0m         \u001b[0mloss\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     86\u001b[0m         \u001b[1;31m#train_loss += loss.data[0]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-122-d78c38b6bdb1>\u001b[0m in \u001b[0;36mloss\u001b[1;34m(self, reconstruction, x, mu, logvar)\u001b[0m\n\u001b[0;32m     59\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     60\u001b[0m         \u001b[0mset_trace\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 61\u001b[1;33m         \u001b[0mbce\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfunctional\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbinary_cross_entropy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreconstruction\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mview\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m-\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput_size\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     62\u001b[0m         \u001b[0mKLD\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m-\u001b[0m\u001b[1;36m0.5\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mlogvar\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mmu\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpow\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mlogvar\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexp\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     63\u001b[0m         \u001b[1;31m# Normalise by same number of elements as in reconstruction\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-122-d78c38b6bdb1>\u001b[0m in \u001b[0;36mloss\u001b[1;34m(self, reconstruction, x, mu, logvar)\u001b[0m\n\u001b[0;32m     59\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     60\u001b[0m         \u001b[0mset_trace\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 61\u001b[1;33m         \u001b[0mbce\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnn\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfunctional\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbinary_cross_entropy\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreconstruction\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mview\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m-\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minput_size\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     62\u001b[0m         \u001b[0mKLD\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m-\u001b[0m\u001b[1;36m0.5\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m1\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mlogvar\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mmu\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpow\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m-\u001b[0m \u001b[0mlogvar\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexp\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     63\u001b[0m         \u001b[1;31m# Normalise by same number of elements as in reconstruction\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\fastai2\\lib\\bdb.py\u001b[0m in \u001b[0;36mtrace_dispatch\u001b[1;34m(self, frame, event, arg)\u001b[0m\n\u001b[0;32m     86\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[1;31m# None\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     87\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mevent\u001b[0m \u001b[1;33m==\u001b[0m \u001b[1;34m'line'\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 88\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdispatch_line\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     89\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mevent\u001b[0m \u001b[1;33m==\u001b[0m \u001b[1;34m'call'\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     90\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdispatch_call\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0marg\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Anaconda3\\envs\\fastai2\\lib\\bdb.py\u001b[0m in \u001b[0;36mdispatch_line\u001b[1;34m(self, frame)\u001b[0m\n\u001b[0;32m    111\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mstop_here\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbreak_here\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    112\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0muser_line\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 113\u001b[1;33m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mquitting\u001b[0m\u001b[1;33m:\u001b[0m \u001b[1;32mraise\u001b[0m \u001b[0mBdbQuit\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    114\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtrace_dispatch\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    115\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mBdbQuit\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "from torch.autograd import Variable\n",
    "\n",
    "import numpy as np\n",
    "representation_size = 2\n",
    "input_size = 4\n",
    "n_samples = 2000\n",
    "batch_size = 5\n",
    "n_samples_per_batch = n_samples//input_size\n",
    "\n",
    "y = np.array([i for i in range(input_size)  for _ in range(n_samples_per_batch)])\n",
    "\n",
    "d = np.identity(input_size)\n",
    "x = np.array([d[i] for i in y], dtype=np.float32)\n",
    "\n",
    "class VAE(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(VAE, self).__init__()\n",
    "        self.en1 = nn.Linear(input_size, 200)\n",
    "        self.en_mu = nn.Linear(200, representation_size)\n",
    "        self.en_std = nn.Linear(200, representation_size)\n",
    "        self.de1 = nn.Linear(representation_size, 200)\n",
    "        self.de2 = nn.Linear(200, input_size)\n",
    "        self.relu = nn.ReLU()\n",
    "        self.sigmoid = nn.Sigmoid()\n",
    "        \n",
    "    def encode(self, x):\n",
    "        \"\"\"Encode a batch of samples, and return posterior parameters for each point.\"\"\"\n",
    "        h1 = self.relu(self.en1(x))\n",
    "        return self.en_mu(h1), self.en_std(h1)\n",
    "    \n",
    "    def decode(self, z):\n",
    "        \"\"\"Decode a batch of latent variables\"\"\"\n",
    "        \n",
    "        h2 = self.relu(self.de1(z))\n",
    "        return self.sigmoid(self.de2(h2))\n",
    "    \n",
    "    def reparam(self, mu, logvar):\n",
    "        \"\"\"Reparameterisation trick to sample z values. \n",
    "        This is stochastic during training,  and returns the mode during evaluation.\"\"\"\n",
    "        \n",
    "        if self.training:\n",
    "            std = logvar.mul(0.5).exp_()\n",
    "            eps = Variable(std.data.new(std.size()).normal_())\n",
    "            return eps.mul(std).add_(mu)\n",
    "        else:\n",
    "            return mu\n",
    "            \n",
    "    \n",
    "    def forward(self, x):\n",
    "        \"\"\"Takes a batch of samples, encodes them, and then decodes them again to compare.\"\"\"\n",
    "        mu, logvar = self.encode(x.view(-1, input_size))\n",
    "        z = self.reparam(mu, logvar)\n",
    "        return self.decode(z), mu, logvar\n",
    "    \n",
    "    def loss(self, reconstruction, x, mu, logvar):\n",
    "        \"\"\"ELBO assuming entries of x are binary variables, with closed form KLD.\"\"\"\n",
    "        \n",
    "        set_trace()\n",
    "        bce = torch.nn.functional.binary_cross_entropy(reconstruction, x.view(-1, input_size))\n",
    "        KLD = -0.5 * torch.sum(1 + logvar - mu.pow(2) - logvar.exp())\n",
    "        # Normalise by same number of elements as in reconstruction\n",
    "        KLD /= x.view(-1, input_size).data.shape[0] * input_size\n",
    "\n",
    "        return bce + KLD\n",
    "    \n",
    "    def get_z(self, x):\n",
    "        \"\"\"Encode a batch of data points, x, into their z representations.\"\"\"\n",
    "        \n",
    "        mu, logvar = self.encode(x.view(-1, input_size))\n",
    "        return self.reparam(mu, logvar)\n",
    "model = VAE()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3)\n",
    "def train(epoch, batches_per_epoch = 501, log_interval=500):\n",
    "    model.train()\n",
    "    \n",
    "    ind = np.arange(x.shape[0])\n",
    "    for i in range(batches_per_epoch):\n",
    "        data = torch.from_numpy(x[np.random.choice(ind, size=batch_size)])\n",
    "        data = Variable(data, requires_grad=False)\n",
    "        optimizer.zero_grad()\n",
    "        recon_batch, mu, logvar = model(data)\n",
    "        loss = model.loss(recon_batch, data, mu, logvar)\n",
    "        loss.backward()\n",
    "        #train_loss += loss.data[0]\n",
    "        optimizer.step()\n",
    "        if (i % log_interval == 0) and (epoch % 5 ==0):\n",
    "            #Print progress\n",
    "            print('Train Epoch: {} [{}/{}]\\tLoss: {:.6f}'.format(\n",
    "                epoch, i * batch_size, batch_size*batches_per_epoch,\n",
    "                loss.data[0] / len(data)))\n",
    "\n",
    "    print('====> Epoch: {} done!'.format(\n",
    "          epoch))\n",
    "\n",
    "\n",
    "def test(epoch, batches_per_epoch=1):\n",
    "    model.eval()\n",
    "\n",
    "    ind = np.arange(x.shape[0])\n",
    "    for i in range(batches_per_epoch):\n",
    "        \n",
    "        data = torch.from_numpy(x[np.random.choice(ind, size=batch_size)])\n",
    "        data = Variable(data, requires_grad=False)\n",
    "        recon_batch, mu, logvar = model(data)\n",
    "        if i == 0:\n",
    "            n = min(data.size(0), 2)\n",
    "            print(data.view(batch_size, 2,2)[:n])\n",
    "            print(recon_batch.view(batch_size, 2,2)[:n])\n",
    "print_examples = False #Change to true if you want to see some examples at each step!\n",
    "for epoch in range(1, 30):\n",
    "    train(epoch)\n",
    "    if print_examples:\n",
    "        test(epoch)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}