1.2 - Train MoCoGAN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training MoCoGAN\n",
    "\n",
    "In this notebook, the process of using the class Trainer will be explained and the training of MoCoGAN will be carried on as if the `newTrain.py` script would be launched from Console."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Models and Dataloading\n",
    "First, let's import classes from `models` and `dataloading` modules.\n",
    "\n",
    "Then let's create the dataloader and define the clParameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('./mocogan')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/carlo/anaconda3/lib/python3.6/site-packages/dask/config.py:168: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.\n",
      "  data = yaml.load(f.read()) or {}\n"
     ]
    }
   ],
   "source": [
    "from mocogan.dataloading import DataLoaderFactory, Transformator\n",
    "from mocogan.models      import UCF_101, Discriminator_I, VideoDiscriminator, VideoGenerator\n",
    "\n",
    "def createModels(parameters, dict_For_Classes = None):\n",
    "\n",
    "    n_classes  = len(parameters['classes'])\n",
    "\n",
    "    n_channels = 3\n",
    "    dim_z_content=50                                            #  dimensionality of the content input, ie hidden space [default: 50]\n",
    "    dim_z_motion=10                                             #  dimensionality of the motion input [default: 10]\n",
    "    dim_z_category= n_classes if n_classes > 0 else 101         #  dimensionality of categorical input [default: 101]\n",
    "    video_lenght = 16                                           #  hyperparameter T of the mocogan [default: 16]\n",
    "    cuda = parameters['cuda']\n",
    "\n",
    "    return Discriminator_I(), VideoDiscriminator(n_channels, dim_z_category, cuda = cuda), \\\n",
    "                 VideoGenerator(n_channels, dim_z_content, dim_z_category, dim_z_motion, video_lenght, cuda, class_to_idx = dict_For_Classes)\n",
    "\n",
    "\n",
    "img_size          = 96\n",
    "max_frame         = 16\n",
    "mean              = (100.99800554447337/255, 96.7195209000943/255, 89.63882431650443/255)\n",
    "std               = (72.07041943699456/255, 70.41506399740703/255, 71.55581999303428/255)\n",
    "\n",
    "batch_size        = 16\n",
    "classes           = ['Biking', 'HorseRace', 'HorseRiding', 'Bowling', 'IceDancing']\n",
    "preprocessing     = 0\n",
    "cuda              = 1\n",
    "soft_labels       = True\n",
    "random_labels     = True\n",
    "shuffle_labels    = True\n",
    "i_wasserstein     = 5\n",
    "n_iter            = 100\n",
    "i_log_stat        = 1\n",
    "pre_train_epoch   = 0\n",
    "i_save_weights    = i_wasserstein\n",
    "i_alternate_train = 2\n",
    "\n",
    "\n",
    "\n",
    "parameters = {\n",
    "        \"cuda\"              : cuda,\n",
    "        \"batch_size\"        : batch_size,\n",
    "        \"soft_labels\"       : soft_labels,\n",
    "        \"n_iter\"            : n_iter,\n",
    "        \"classes\"           : classes,\n",
    "        \"i_log_stat\"        : i_log_stat,\n",
    "        \"pre_train_epoch\"   : pre_train_epoch,\n",
    "        \"i_save_weights\"    : i_save_weights,\n",
    "        \"i_alternate_train\" : i_alternate_train,\n",
    "        \"preprocessed\"      : preprocessing,\n",
    "        \"random_labels\"     : random_labels,\n",
    "        \"shuffle_labels\"    : shuffle_labels,\n",
    "        \"i_wasserstein\"     : i_wasserstein\n",
    "    }\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "transformator = Transformator(img_size, max_frame, medium= mean, stdDev= std)\n",
    "dataloader    = DataLoaderFactory(UCF_101, transformator, preprocessing, batch_size, classes = classes).getDataLoader()\n",
    "\n",
    "dis_i, dis_v, gen = createModels(parameters, dataloader.dataset.class_to_idx)\n",
    "\n",
    "gen.init_weigths()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create the Trainer\n",
    "Let's import and use the Trainer Object to train the networks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from mocogan.trainer import Trainer\n",
    "\n",
    "trainer = Trainer(parameters, dis_i, dis_v, gen, dataloader, transformator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Start Training\n",
    "Finally, let's start training process in the `Trainer`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch [45/45] Loss_Di: 1.1356 Loss_Dv: 1.8988 Accuracy_Dv: 0.5625 Loss_Gen: 0.0000\n",
      "[1/100] (0d 0h 3m 18s) Loss_Di: 1.2626 Loss_Dv: 2.4158 Accuracy_Dv: 0.4639 Loss_Generator: 3.0266\n",
      "Batch [45/45] Loss_Di: 0.7337 Loss_Dv: 1.6259 Accuracy_Dv: 0.5625 Loss_Gen: 0.0000\n",
      "[2/100] (0d 0h 6m 48s) Loss_Di: 0.9326 Loss_Dv: 1.6695 Accuracy_Dv: 0.5694 Loss_Generator: 3.0215\n",
      "Batch [45/45] Loss_Di: 0.4908 Loss_Dv: 1.6827 Accuracy_Dv: 0.5000 Loss_Gen: 0.0000\n",
      "[3/100] (0d 0h 10m 13s) Loss_Di: 0.6240 Loss_Dv: 1.4974 Accuracy_Dv: 0.5556 Loss_Generator: 2.7066\n",
      "Batch [45/45] Loss_Di: 0.3142 Loss_Dv: 1.2262 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[4/100] (0d 0h 13m 33s) Loss_Di: 0.4946 Loss_Dv: 1.4383 Accuracy_Dv: 0.5750 Loss_Generator: 2.3793\n",
      "Batch [45/45] Loss_Di: 0.3828 Loss_Dv: 1.1352 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[5/100] (0d 0h 16m 51s) Loss_Di: 0.4236 Loss_Dv: 1.3835 Accuracy_Dv: 0.5861 Loss_Generator: 1.9519\n",
      "Batch [45/45] Loss_Di: 0.3982 Loss_Dv: 1.0709 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[6/100] (0d 0h 20m 17s) Loss_Di: 0.3825 Loss_Dv: 1.3359 Accuracy_Dv: 0.6181 Loss_Generator: 1.4841\n",
      "Batch [45/45] Loss_Di: 0.2649 Loss_Dv: 1.3203 Accuracy_Dv: 0.5000 Loss_Gen: 0.0000\n",
      "[7/100] (0d 0h 23m 37s) Loss_Di: 0.3706 Loss_Dv: 1.2702 Accuracy_Dv: 0.6389 Loss_Generator: 1.3520\n",
      "Batch [45/45] Loss_Di: 0.2513 Loss_Dv: 1.2915 Accuracy_Dv: 0.5000 Loss_Gen: 0.0000\n",
      "[8/100] (0d 0h 27m 6s) Loss_Di: 0.3706 Loss_Dv: 1.3069 Accuracy_Dv: 0.6292 Loss_Generator: 1.2160\n",
      "Batch [45/45] Loss_Di: 0.1961 Loss_Dv: 1.1792 Accuracy_Dv: 0.5625 Loss_Gen: 0.0000\n",
      "[9/100] (0d 0h 30m 27s) Loss_Di: 0.2289 Loss_Dv: 1.1188 Accuracy_Dv: 0.6389 Loss_Generator: 0.8499\n",
      "Batch [45/45] Loss_Di: 0.2169 Loss_Dv: 1.1996 Accuracy_Dv: 0.6250 Loss_Gen: 0.0000\n",
      "[10/100] (0d 0h 33m 49s) Loss_Di: 0.2158 Loss_Dv: 1.0941 Accuracy_Dv: 0.6597 Loss_Generator: 0.7772\n",
      "Batch [45/45] Loss_Di: 0.2022 Loss_Dv: 0.8924 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[11/100] (0d 0h 37m 17s) Loss_Di: 0.2126 Loss_Dv: 1.0746 Accuracy_Dv: 0.6667 Loss_Generator: 0.7509\n",
      "Batch [45/45] Loss_Di: 0.2244 Loss_Dv: 1.2497 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[12/100] (0d 0h 40m 41s) Loss_Di: 0.2143 Loss_Dv: 1.0224 Accuracy_Dv: 0.6736 Loss_Generator: 0.7273\n",
      "Batch [45/45] Loss_Di: 0.2050 Loss_Dv: 1.0509 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[13/100] (0d 0h 44m 9s) Loss_Di: 0.2091 Loss_Dv: 1.0154 Accuracy_Dv: 0.6917 Loss_Generator: 0.6858\n",
      "Batch [45/45] Loss_Di: 0.1974 Loss_Dv: 1.1455 Accuracy_Dv: 0.6250 Loss_Gen: 0.0000\n",
      "[14/100] (0d 0h 47m 35s) Loss_Di: 0.2071 Loss_Dv: 1.0224 Accuracy_Dv: 0.6889 Loss_Generator: 0.5796\n",
      "Batch [45/45] Loss_Di: 0.1926 Loss_Dv: 1.0476 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[15/100] (0d 0h 50m 58s) Loss_Di: 0.2111 Loss_Dv: 1.0036 Accuracy_Dv: 0.7014 Loss_Generator: 0.5630\n",
      "Batch [45/45] Loss_Di: 0.2116 Loss_Dv: 0.7957 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[16/100] (0d 0h 54m 34s) Loss_Di: 0.2052 Loss_Dv: 0.9737 Accuracy_Dv: 0.6819 Loss_Generator: 0.5869\n",
      "Batch [45/45] Loss_Di: 0.1832 Loss_Dv: 0.8678 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[17/100] (0d 0h 57m 55s) Loss_Di: 0.2041 Loss_Dv: 0.9418 Accuracy_Dv: 0.7208 Loss_Generator: 0.5403\n",
      "Batch [45/45] Loss_Di: 0.2211 Loss_Dv: 0.8109 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[18/100] (0d 1h 1m 17s) Loss_Di: 0.2040 Loss_Dv: 0.9827 Accuracy_Dv: 0.7347 Loss_Generator: 0.5219\n",
      "Batch [45/45] Loss_Di: 0.1850 Loss_Dv: 1.4705 Accuracy_Dv: 0.5000 Loss_Gen: 0.0000\n",
      "[19/100] (0d 1h 4m 46s) Loss_Di: 0.2046 Loss_Dv: 0.9549 Accuracy_Dv: 0.7292 Loss_Generator: 0.4912\n",
      "Batch [45/45] Loss_Di: 0.2060 Loss_Dv: 0.6330 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[20/100] (0d 1h 8m 9s) Loss_Di: 0.2026 Loss_Dv: 0.9373 Accuracy_Dv: 0.7403 Loss_Generator: 0.5276\n",
      "Batch [45/45] Loss_Di: 0.2308 Loss_Dv: 1.1424 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[21/100] (0d 1h 11m 37s) Loss_Di: 0.1994 Loss_Dv: 0.9745 Accuracy_Dv: 0.7056 Loss_Generator: 0.5273\n",
      "Batch [45/45] Loss_Di: 0.1992 Loss_Dv: 0.7949 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[22/100] (0d 1h 15m 2s) Loss_Di: 0.2041 Loss_Dv: 0.9043 Accuracy_Dv: 0.7236 Loss_Generator: 0.4839\n",
      "Batch [45/45] Loss_Di: 0.1940 Loss_Dv: 1.1038 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[23/100] (0d 1h 18m 24s) Loss_Di: 0.2019 Loss_Dv: 0.9036 Accuracy_Dv: 0.7222 Loss_Generator: 0.4839\n",
      "Batch [45/45] Loss_Di: 0.1916 Loss_Dv: 0.8843 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[24/100] (0d 1h 21m 46s) Loss_Di: 0.2012 Loss_Dv: 0.9202 Accuracy_Dv: 0.7278 Loss_Generator: 0.4706\n",
      "Batch [45/45] Loss_Di: 0.2085 Loss_Dv: 0.7957 Accuracy_Dv: 0.6250 Loss_Gen: 0.0000\n",
      "[25/100] (0d 1h 25m 12s) Loss_Di: 0.2031 Loss_Dv: 0.8756 Accuracy_Dv: 0.7167 Loss_Generator: 0.4160\n",
      "Batch [45/45] Loss_Di: 0.1989 Loss_Dv: 0.9082 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[26/100] (0d 1h 28m 34s) Loss_Di: 0.2028 Loss_Dv: 0.9035 Accuracy_Dv: 0.7111 Loss_Generator: 0.4436\n",
      "Batch [45/45] Loss_Di: 0.2286 Loss_Dv: 0.6527 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[27/100] (0d 1h 31m 59s) Loss_Di: 0.2014 Loss_Dv: 0.8526 Accuracy_Dv: 0.7514 Loss_Generator: 0.4603\n",
      "Batch [45/45] Loss_Di: 0.1935 Loss_Dv: 0.9461 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[28/100] (0d 1h 35m 27s) Loss_Di: 0.2044 Loss_Dv: 0.8542 Accuracy_Dv: 0.7569 Loss_Generator: 0.4277\n",
      "Batch [45/45] Loss_Di: 0.2388 Loss_Dv: 0.8459 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[29/100] (0d 1h 38m 56s) Loss_Di: 0.2066 Loss_Dv: 0.8607 Accuracy_Dv: 0.7403 Loss_Generator: 0.4553\n",
      "Batch [45/45] Loss_Di: 0.1768 Loss_Dv: 0.5659 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[30/100] (0d 1h 42m 22s) Loss_Di: 0.2028 Loss_Dv: 0.8297 Accuracy_Dv: 0.7750 Loss_Generator: 0.4474\n",
      "Batch [45/45] Loss_Di: 0.2584 Loss_Dv: 0.6647 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[31/100] (0d 1h 45m 55s) Loss_Di: 0.2053 Loss_Dv: 0.8653 Accuracy_Dv: 0.7431 Loss_Generator: 0.4342\n",
      "Batch [45/45] Loss_Di: 0.2406 Loss_Dv: 0.9333 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[32/100] (0d 1h 49m 17s) Loss_Di: 0.2081 Loss_Dv: 0.8057 Accuracy_Dv: 0.7819 Loss_Generator: 0.4541\n",
      "Batch [45/45] Loss_Di: 0.1634 Loss_Dv: 0.4864 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[33/100] (0d 1h 52m 46s) Loss_Di: 0.1993 Loss_Dv: 0.8187 Accuracy_Dv: 0.7667 Loss_Generator: 0.4328\n",
      "Batch [45/45] Loss_Di: 0.1812 Loss_Dv: 0.6979 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[34/100] (0d 1h 56m 15s) Loss_Di: 0.1956 Loss_Dv: 0.7886 Accuracy_Dv: 0.7986 Loss_Generator: 0.4040\n",
      "Batch [45/45] Loss_Di: 0.2535 Loss_Dv: 0.9193 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[35/100] (0d 1h 59m 37s) Loss_Di: 0.2052 Loss_Dv: 0.7932 Accuracy_Dv: 0.7764 Loss_Generator: 0.4370\n",
      "Batch [45/45] Loss_Di: 0.2330 Loss_Dv: 0.6180 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[36/100] (0d 2h 2m 58s) Loss_Di: 0.2018 Loss_Dv: 0.8255 Accuracy_Dv: 0.7611 Loss_Generator: 0.4107\n",
      "Batch [45/45] Loss_Di: 0.2343 Loss_Dv: 0.9679 Accuracy_Dv: 0.6250 Loss_Gen: 0.0000\n",
      "[37/100] (0d 2h 6m 26s) Loss_Di: 0.2005 Loss_Dv: 0.8055 Accuracy_Dv: 0.7722 Loss_Generator: 0.4313\n",
      "Batch [45/45] Loss_Di: 0.2015 Loss_Dv: 0.7086 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[38/100] (0d 2h 9m 49s) Loss_Di: 0.2062 Loss_Dv: 0.7569 Accuracy_Dv: 0.8042 Loss_Generator: 0.4162\n",
      "Batch [45/45] Loss_Di: 0.1712 Loss_Dv: 0.6637 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[39/100] (0d 2h 13m 18s) Loss_Di: 0.1953 Loss_Dv: 0.7319 Accuracy_Dv: 0.8069 Loss_Generator: 0.4238\n",
      "Batch [45/45] Loss_Di: 0.2088 Loss_Dv: 1.1316 Accuracy_Dv: 0.6250 Loss_Gen: 0.0000\n",
      "[40/100] (0d 2h 16m 46s) Loss_Di: 0.2050 Loss_Dv: 0.7821 Accuracy_Dv: 0.7819 Loss_Generator: 0.4179\n",
      "Batch [45/45] Loss_Di: 0.1918 Loss_Dv: 0.5328 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[41/100] (0d 2h 20m 6s) Loss_Di: 0.2004 Loss_Dv: 0.7438 Accuracy_Dv: 0.8014 Loss_Generator: 0.4151\n",
      "Batch [45/45] Loss_Di: 0.2253 Loss_Dv: 0.8515 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[42/100] (0d 2h 23m 29s) Loss_Di: 0.2052 Loss_Dv: 0.7846 Accuracy_Dv: 0.8014 Loss_Generator: 0.4022\n",
      "Batch [45/45] Loss_Di: 0.1643 Loss_Dv: 0.5519 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[43/100] (0d 2h 26m 56s) Loss_Di: 0.1946 Loss_Dv: 0.7521 Accuracy_Dv: 0.7847 Loss_Generator: 0.4145\n",
      "Batch [45/45] Loss_Di: 0.1662 Loss_Dv: 0.6358 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[44/100] (0d 2h 30m 22s) Loss_Di: 0.2024 Loss_Dv: 0.6812 Accuracy_Dv: 0.8389 Loss_Generator: 0.4218\n",
      "Batch [45/45] Loss_Di: 0.2099 Loss_Dv: 0.8749 Accuracy_Dv: 0.5000 Loss_Gen: 0.0000\n",
      "[45/100] (0d 2h 33m 42s) Loss_Di: 0.1979 Loss_Dv: 0.7284 Accuracy_Dv: 0.8097 Loss_Generator: 0.4373\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch [45/45] Loss_Di: 0.1819 Loss_Dv: 0.4246 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[46/100] (0d 2h 37m 1s) Loss_Di: 0.2003 Loss_Dv: 0.7048 Accuracy_Dv: 0.8097 Loss_Generator: 0.4291\n",
      "Batch [45/45] Loss_Di: 0.1716 Loss_Dv: 0.5958 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[47/100] (0d 2h 40m 37s) Loss_Di: 0.1998 Loss_Dv: 0.7411 Accuracy_Dv: 0.8153 Loss_Generator: 0.4096\n",
      "Batch [45/45] Loss_Di: 0.2232 Loss_Dv: 0.6144 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[48/100] (0d 2h 44m 3s) Loss_Di: 0.2071 Loss_Dv: 0.6982 Accuracy_Dv: 0.8083 Loss_Generator: 0.4273\n",
      "Batch [45/45] Loss_Di: 0.2115 Loss_Dv: 0.4988 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[49/100] (0d 2h 47m 31s) Loss_Di: 0.2026 Loss_Dv: 0.6925 Accuracy_Dv: 0.8111 Loss_Generator: 0.4176\n",
      "Batch [45/45] Loss_Di: 0.1942 Loss_Dv: 0.5482 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[50/100] (0d 2h 51m 1s) Loss_Di: 0.1998 Loss_Dv: 0.6953 Accuracy_Dv: 0.8139 Loss_Generator: 0.4141\n",
      "Batch [45/45] Loss_Di: 0.2231 Loss_Dv: 0.5389 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[51/100] (0d 2h 54m 32s) Loss_Di: 0.1961 Loss_Dv: 0.6884 Accuracy_Dv: 0.8222 Loss_Generator: 0.4062\n",
      "Batch [45/45] Loss_Di: 0.2267 Loss_Dv: 0.6607 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[52/100] (0d 2h 57m 53s) Loss_Di: 0.2016 Loss_Dv: 0.6767 Accuracy_Dv: 0.8250 Loss_Generator: 0.4113\n",
      "Batch [45/45] Loss_Di: 0.1893 Loss_Dv: 0.5559 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[53/100] (0d 3h 1m 17s) Loss_Di: 0.1987 Loss_Dv: 0.6433 Accuracy_Dv: 0.8444 Loss_Generator: 0.4099\n",
      "Batch [45/45] Loss_Di: 0.2352 Loss_Dv: 0.8257 Accuracy_Dv: 0.7500 Loss_Gen: 0.0000\n",
      "[54/100] (0d 3h 4m 45s) Loss_Di: 0.1981 Loss_Dv: 0.6681 Accuracy_Dv: 0.8389 Loss_Generator: 0.4056\n",
      "Batch [45/45] Loss_Di: 0.1672 Loss_Dv: 0.6966 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[55/100] (0d 3h 8m 9s) Loss_Di: 0.1964 Loss_Dv: 0.6473 Accuracy_Dv: 0.8361 Loss_Generator: 0.3899\n",
      "Batch [45/45] Loss_Di: 0.2361 Loss_Dv: 0.5502 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[56/100] (0d 3h 11m 36s) Loss_Di: 0.1988 Loss_Dv: 0.6221 Accuracy_Dv: 0.8514 Loss_Generator: 0.4134\n",
      "Batch [45/45] Loss_Di: 0.1829 Loss_Dv: 0.4479 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[57/100] (0d 3h 15m 5s) Loss_Di: 0.2025 Loss_Dv: 0.6267 Accuracy_Dv: 0.8514 Loss_Generator: 0.4029\n",
      "Batch [45/45] Loss_Di: 0.2211 Loss_Dv: 0.4961 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[58/100] (0d 3h 18m 37s) Loss_Di: 0.1975 Loss_Dv: 0.6625 Accuracy_Dv: 0.8292 Loss_Generator: 0.4097\n",
      "Batch [45/45] Loss_Di: 0.2083 Loss_Dv: 0.6680 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[59/100] (0d 3h 21m 59s) Loss_Di: 0.1965 Loss_Dv: 0.6401 Accuracy_Dv: 0.8375 Loss_Generator: 0.4094\n",
      "Batch [45/45] Loss_Di: 0.1770 Loss_Dv: 0.4735 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[60/100] (0d 3h 25m 25s) Loss_Di: 0.1985 Loss_Dv: 0.6489 Accuracy_Dv: 0.8444 Loss_Generator: 0.4029\n",
      "Batch [45/45] Loss_Di: 0.2040 Loss_Dv: 0.5893 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[61/100] (0d 3h 28m 47s) Loss_Di: 0.2002 Loss_Dv: 0.6375 Accuracy_Dv: 0.8361 Loss_Generator: 0.4103\n",
      "Batch [45/45] Loss_Di: 0.1995 Loss_Dv: 0.6043 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[62/100] (0d 3h 32m 7s) Loss_Di: 0.2006 Loss_Dv: 0.6449 Accuracy_Dv: 0.8250 Loss_Generator: 0.4007\n",
      "Batch [45/45] Loss_Di: 0.2346 Loss_Dv: 0.5551 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[63/100] (0d 3h 35m 34s) Loss_Di: 0.1984 Loss_Dv: 0.6139 Accuracy_Dv: 0.8569 Loss_Generator: 0.3937\n",
      "Batch [45/45] Loss_Di: 0.2432 Loss_Dv: 0.5098 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[64/100] (0d 3h 39m 2s) Loss_Di: 0.1985 Loss_Dv: 0.6188 Accuracy_Dv: 0.8333 Loss_Generator: 0.3861\n",
      "Batch [45/45] Loss_Di: 0.1636 Loss_Dv: 0.4754 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[65/100] (0d 3h 42m 22s) Loss_Di: 0.1994 Loss_Dv: 0.6339 Accuracy_Dv: 0.8417 Loss_Generator: 0.3999\n",
      "Batch [45/45] Loss_Di: 0.2572 Loss_Dv: 0.6717 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[66/100] (0d 3h 46m 1s) Loss_Di: 0.1994 Loss_Dv: 0.6420 Accuracy_Dv: 0.8403 Loss_Generator: 0.4075\n",
      "Batch [45/45] Loss_Di: 0.1679 Loss_Dv: 0.6255 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[67/100] (0d 3h 49m 22s) Loss_Di: 0.1958 Loss_Dv: 0.6472 Accuracy_Dv: 0.8542 Loss_Generator: 0.3883\n",
      "Batch [45/45] Loss_Di: 0.2355 Loss_Dv: 0.6333 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[68/100] (0d 3h 52m 46s) Loss_Di: 0.2007 Loss_Dv: 0.5848 Accuracy_Dv: 0.8625 Loss_Generator: 0.4152\n",
      "Batch [45/45] Loss_Di: 0.1987 Loss_Dv: 0.5382 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[69/100] (0d 3h 56m 9s) Loss_Di: 0.1979 Loss_Dv: 0.5970 Accuracy_Dv: 0.8403 Loss_Generator: 0.4054\n",
      "Batch [45/45] Loss_Di: 0.1878 Loss_Dv: 0.5631 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[70/100] (0d 3h 59m 29s) Loss_Di: 0.2040 Loss_Dv: 0.5842 Accuracy_Dv: 0.8667 Loss_Generator: 0.4155\n",
      "Batch [45/45] Loss_Di: 0.1736 Loss_Dv: 0.4595 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[71/100] (0d 4h 2m 57s) Loss_Di: 0.1929 Loss_Dv: 0.5879 Accuracy_Dv: 0.8514 Loss_Generator: 0.4003\n",
      "Batch [45/45] Loss_Di: 0.2213 Loss_Dv: 0.6131 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[72/100] (0d 4h 6m 23s) Loss_Di: 0.1997 Loss_Dv: 0.6207 Accuracy_Dv: 0.8444 Loss_Generator: 0.4017\n",
      "Batch [45/45] Loss_Di: 0.2242 Loss_Dv: 0.8677 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[73/100] (0d 4h 9m 52s) Loss_Di: 0.1959 Loss_Dv: 0.6344 Accuracy_Dv: 0.8250 Loss_Generator: 0.3987\n",
      "Batch [45/45] Loss_Di: 0.2146 Loss_Dv: 0.5306 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[74/100] (0d 4h 13m 13s) Loss_Di: 0.1940 Loss_Dv: 0.6412 Accuracy_Dv: 0.8292 Loss_Generator: 0.3989\n",
      "Batch [45/45] Loss_Di: 0.1826 Loss_Dv: 0.5370 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[75/100] (0d 4h 16m 41s) Loss_Di: 0.1977 Loss_Dv: 0.5818 Accuracy_Dv: 0.8653 Loss_Generator: 0.3941\n",
      "Batch [45/45] Loss_Di: 0.2280 Loss_Dv: 0.5168 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[76/100] (0d 4h 20m 7s) Loss_Di: 0.1954 Loss_Dv: 0.5737 Accuracy_Dv: 0.8639 Loss_Generator: 0.6676\n",
      "Batch [45/45] Loss_Di: 0.1690 Loss_Dv: 0.5621 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[77/100] (0d 4h 23m 35s) Loss_Di: 0.1979 Loss_Dv: 0.5661 Accuracy_Dv: 0.8722 Loss_Generator: 1.0804\n",
      "Batch [45/45] Loss_Di: 0.2173 Loss_Dv: 0.6878 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[78/100] (0d 4h 27m 1s) Loss_Di: 0.1968 Loss_Dv: 0.6010 Accuracy_Dv: 0.8597 Loss_Generator: 1.0078\n",
      "Batch [45/45] Loss_Di: 0.1908 Loss_Dv: 0.4632 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[79/100] (0d 4h 30m 27s) Loss_Di: 0.1977 Loss_Dv: 0.5793 Accuracy_Dv: 0.8556 Loss_Generator: 0.6545\n",
      "Batch [45/45] Loss_Di: 0.2102 Loss_Dv: 0.7765 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[80/100] (0d 4h 33m 47s) Loss_Di: 0.1975 Loss_Dv: 0.5928 Accuracy_Dv: 0.8694 Loss_Generator: 0.6191\n",
      "Batch [45/45] Loss_Di: 0.1844 Loss_Dv: 0.4769 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[81/100] (0d 4h 37m 21s) Loss_Di: 0.1994 Loss_Dv: 0.5589 Accuracy_Dv: 0.8833 Loss_Generator: 0.5254\n",
      "Batch [45/45] Loss_Di: 0.2007 Loss_Dv: 0.3431 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[82/100] (0d 4h 40m 48s) Loss_Di: 0.1991 Loss_Dv: 0.5776 Accuracy_Dv: 0.8528 Loss_Generator: 0.5165\n",
      "Batch [45/45] Loss_Di: 0.2043 Loss_Dv: 0.6440 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[83/100] (0d 4h 44m 13s) Loss_Di: 0.2015 Loss_Dv: 0.5724 Accuracy_Dv: 0.8556 Loss_Generator: 0.4890\n",
      "Batch [45/45] Loss_Di: 0.1823 Loss_Dv: 0.3364 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[84/100] (0d 4h 47m 43s) Loss_Di: 0.2036 Loss_Dv: 0.5215 Accuracy_Dv: 0.8875 Loss_Generator: 0.4887\n",
      "Batch [45/45] Loss_Di: 0.1935 Loss_Dv: 0.8653 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[85/100] (0d 4h 51m 8s) Loss_Di: 0.2006 Loss_Dv: 0.6002 Accuracy_Dv: 0.8542 Loss_Generator: 0.4661\n",
      "Batch [45/45] Loss_Di: 0.1918 Loss_Dv: 0.3987 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[86/100] (0d 4h 54m 36s) Loss_Di: 0.2012 Loss_Dv: 0.5478 Accuracy_Dv: 0.8861 Loss_Generator: 0.4432\n",
      "Batch [45/45] Loss_Di: 0.2246 Loss_Dv: 0.4770 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[87/100] (0d 4h 57m 55s) Loss_Di: 0.1995 Loss_Dv: 0.5450 Accuracy_Dv: 0.8722 Loss_Generator: 0.4483\n",
      "Batch [45/45] Loss_Di: 0.1596 Loss_Dv: 0.5739 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[88/100] (0d 5h 1m 24s) Loss_Di: 0.2010 Loss_Dv: 0.5510 Accuracy_Dv: 0.8667 Loss_Generator: 0.4495\n",
      "Batch [45/45] Loss_Di: 0.1924 Loss_Dv: 0.5874 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[89/100] (0d 5h 5m 0s) Loss_Di: 0.2015 Loss_Dv: 0.5356 Accuracy_Dv: 0.8875 Loss_Generator: 0.4559\n",
      "Batch [45/45] Loss_Di: 0.1989 Loss_Dv: 0.6112 Accuracy_Dv: 0.6875 Loss_Gen: 0.0000\n",
      "[90/100] (0d 5h 8m 25s) Loss_Di: 0.2001 Loss_Dv: 0.5457 Accuracy_Dv: 0.8694 Loss_Generator: 0.4557\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch [45/45] Loss_Di: 0.2132 Loss_Dv: 0.4645 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[91/100] (0d 5h 11m 48s) Loss_Di: 0.1973 Loss_Dv: 0.5202 Accuracy_Dv: 0.8986 Loss_Generator: 0.4244\n",
      "Batch [45/45] Loss_Di: 0.1973 Loss_Dv: 0.5942 Accuracy_Dv: 0.8750 Loss_Gen: 0.0000\n",
      "[92/100] (0d 5h 15m 11s) Loss_Di: 0.1958 Loss_Dv: 0.5367 Accuracy_Dv: 0.8833 Loss_Generator: 0.4186\n",
      "Batch [45/45] Loss_Di: 0.1823 Loss_Dv: 0.3729 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[93/100] (0d 5h 18m 33s) Loss_Di: 0.1968 Loss_Dv: 0.5399 Accuracy_Dv: 0.8653 Loss_Generator: 0.4046\n",
      "Batch [45/45] Loss_Di: 0.1561 Loss_Dv: 0.4568 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[94/100] (0d 5h 21m 59s) Loss_Di: 0.2017 Loss_Dv: 0.5697 Accuracy_Dv: 0.8736 Loss_Generator: 0.4327\n",
      "Batch [45/45] Loss_Di: 0.1827 Loss_Dv: 0.7350 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[95/100] (0d 5h 25m 27s) Loss_Di: 0.1988 Loss_Dv: 0.5607 Accuracy_Dv: 0.8708 Loss_Generator: 0.4246\n",
      "Batch [45/45] Loss_Di: 0.2076 Loss_Dv: 0.6686 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[96/100] (0d 5h 28m 48s) Loss_Di: 0.1986 Loss_Dv: 0.5642 Accuracy_Dv: 0.8750 Loss_Generator: 0.4181\n",
      "Batch [45/45] Loss_Di: 0.1877 Loss_Dv: 0.3617 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[97/100] (0d 5h 32m 11s) Loss_Di: 0.1993 Loss_Dv: 0.5407 Accuracy_Dv: 0.8764 Loss_Generator: 0.4393\n",
      "Batch [45/45] Loss_Di: 0.1847 Loss_Dv: 0.3711 Accuracy_Dv: 1.0000 Loss_Gen: 0.0000\n",
      "[98/100] (0d 5h 35m 35s) Loss_Di: 0.1987 Loss_Dv: 0.5181 Accuracy_Dv: 0.8736 Loss_Generator: 0.4219\n",
      "Batch [45/45] Loss_Di: 0.2058 Loss_Dv: 0.4829 Accuracy_Dv: 0.9375 Loss_Gen: 0.0000\n",
      "[99/100] (0d 5h 38m 57s) Loss_Di: 0.1995 Loss_Dv: 0.5077 Accuracy_Dv: 0.8861 Loss_Generator: 0.4125\n",
      "Batch [45/45] Loss_Di: 0.1997 Loss_Dv: 0.8023 Accuracy_Dv: 0.8125 Loss_Gen: 0.0000\n",
      "[100/100] (0d 5h 42m 18s) Loss_Di: 0.1966 Loss_Dv: 0.5409 Accuracy_Dv: 0.8681 Loss_Generator: 0.4299\n"
     ]
    }
   ],
   "source": [
    "trainer.train()"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}