TimeGAN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "1a32cf30",
   "metadata": {},
   "source": [
    "# TimeGAN\n",
    "\n",
    "Adapted from Stefan Jansen's book  \n",
    "[Machine Learning for Trading](https://github.com/stefan-jansen/machine-learning-for-trading/blob/main/21_gans_for_synthetic_time_series/02_TimeGAN_TF2.ipynb).  \n",
    "*TensorFlow2 implemention*  \n",
    "*stock prices*  \n",
    "\n",
    "Which in turn was based Jinsung Yoon, Daniel Jarrett, and Mihaela van der Schaar's paper  \n",
    "[Time-series Generative Adversarial Networks](https://proceedings.neurips.cc/paper/2019/file/c9efe5f26cd17ba6216bbe2a7d26d490-Paper.pdf),\n",
    "Neural Information Processing Systems (NeurIPS), 2019.  \n",
    "*TensorFlow1 implementation*  \n",
    "*Google stock data -- Open, High, Low, Close, Adj_Close, Volume*\n",
    "\n",
    "Last updated: Feb 1, 2023  \n",
    "[Original code](https://github.com/vanderschaarlab/mlforhealthlabpub/tree/main/alg/timegan) by Jinsung Yoon (jsyoon0823@gmail.com)  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fd76bfa",
   "metadata": {},
   "source": [
    "## Troubleshooting\n",
    "In Autoencoder Training Loop (M1 Mac)\n",
    "* **Error:** Graph execution error: Detected at node 'StatefulPartitionedCall_21' defined at (most recent call last)\n",
    "* **Problem:** Apple's latest version of `tensorflow-deps=2.10.0` is not compatible with latest versions of `tensorflow-macos=2.11.0` and `tensorflow-metal==0.7.0`. The latest version of `tensorflow==2.11.0` has a new optimizer API.\n",
    "* **[Temporary Fix](https://developer.apple.com/forums/thread/721619):** Use `from tensorflow.keras.optimizers.legacy import Adam` for now"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23b1ace2",
   "metadata": {},
   "source": [
    "## 1. Imports & Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "802a91d1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "00855495",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "import tensorflow as tf\n",
    "from pathlib import Path\n",
    "from tqdm import tqdm\n",
    "\n",
    "from tensorflow.keras.models import Sequential, Model\n",
    "from tensorflow.keras.layers import GRU, Dense, RNN, GRUCell, Input\n",
    "from tensorflow.keras.losses import BinaryCrossentropy, MeanSquaredError\n",
    "# from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.optimizers.legacy import Adam # solve tensorflow issue on M1 Mac\n",
    "from tensorflow.keras.callbacks import TensorBoard\n",
    "from tensorflow.keras.utils import plot_model\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "baa85f5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using GPU\n"
     ]
    }
   ],
   "source": [
    "gpu_devices = tf.config.experimental.list_physical_devices('GPU')\n",
    "if gpu_devices:\n",
    "    print('Using GPU')\n",
    "    tf.config.experimental.set_memory_growth(gpu_devices[0], True)\n",
    "else:\n",
    "    print('Using CPU')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "30121885",
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set_style('white')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a41c13ec",
   "metadata": {},
   "source": [
    "## 2. Experiment Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ec9655b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "results_path = Path('time_gan')\n",
    "if not results_path.exists():\n",
    "    results_path.mkdir()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2e151304",
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "024a9842",
   "metadata": {},
   "outputs": [],
   "source": [
    "log_dir = results_path / f'experiment_{experiment:02}'\n",
    "if not log_dir.exists():\n",
    "    log_dir.mkdir(parents=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d3cf533a",
   "metadata": {},
   "outputs": [],
   "source": [
    "hdf_store = results_path / 'TimeSeriesGAN.h5'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38fa7561",
   "metadata": {},
   "source": [
    "## 3. Prepare Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a197bbed",
   "metadata": {},
   "source": [
    "### 3.1. Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bcdda552",
   "metadata": {},
   "outputs": [],
   "source": [
    "seq_len = 24\n",
    "n_seq = 6\n",
    "batch_size = 128"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6007b3fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['Open', 'High', 'Low', 'Close', 'Adj_Close', 'Volume']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e3950a6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def select_data():\n",
    "    df = (pd.read_csv('data/GOOGLE_BIG.csv')\n",
    "          .dropna())\n",
    "    df.to_hdf(hdf_store, 'data/real')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1f4535b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "select_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f45925b7",
   "metadata": {},
   "source": [
    "### 3.2. Plot Series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b96acb31",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_hdf(hdf_store, 'data/real')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fbadad57",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3685, 6)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "98821d1d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Adj_Close</th>\n",
       "      <th>Volume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>49.676899</td>\n",
       "      <td>51.693783</td>\n",
       "      <td>47.669952</td>\n",
       "      <td>49.845802</td>\n",
       "      <td>49.845802</td>\n",
       "      <td>44994500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>50.178635</td>\n",
       "      <td>54.187561</td>\n",
       "      <td>49.925285</td>\n",
       "      <td>53.805050</td>\n",
       "      <td>53.805050</td>\n",
       "      <td>23005800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55.017166</td>\n",
       "      <td>56.373344</td>\n",
       "      <td>54.172661</td>\n",
       "      <td>54.346527</td>\n",
       "      <td>54.346527</td>\n",
       "      <td>18393200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>55.260582</td>\n",
       "      <td>55.439419</td>\n",
       "      <td>51.450363</td>\n",
       "      <td>52.096165</td>\n",
       "      <td>52.096165</td>\n",
       "      <td>15361800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>52.140873</td>\n",
       "      <td>53.651051</td>\n",
       "      <td>51.604362</td>\n",
       "      <td>52.657513</td>\n",
       "      <td>52.657513</td>\n",
       "      <td>9257400</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Open       High        Low      Close  Adj_Close    Volume\n",
       "0  49.676899  51.693783  47.669952  49.845802  49.845802  44994500\n",
       "1  50.178635  54.187561  49.925285  53.805050  53.805050  23005800\n",
       "2  55.017166  56.373344  54.172661  54.346527  54.346527  18393200\n",
       "3  55.260582  55.439419  51.450363  52.096165  52.096165  15361800\n",
       "4  52.140873  53.651051  51.604362  52.657513  52.657513   9257400"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "592a042a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1400x600 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "axes = df.div(df.iloc[0]).plot(subplots=True,\n",
    "                               figsize=(14, 6),\n",
    "                               layout=(3, 2),\n",
    "                               title=features,\n",
    "                               legend=False,\n",
    "                               rot=0,\n",
    "                               lw=1, \n",
    "                               color='k')\n",
    "for ax in axes.flatten():\n",
    "    ax.set_xlabel('')\n",
    "\n",
    "plt.suptitle('Normalized Price Series')\n",
    "plt.gcf().tight_layout()\n",
    "sns.despine();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2ae833c",
   "metadata": {},
   "source": [
    "### 3.3. Correlation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "52b69af3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.clustermap(df.corr(),\n",
    "               annot=True,\n",
    "               fmt='.2f',\n",
    "               cmap=sns.diverging_palette(h_neg=20,\n",
    "                                          h_pos=220), center=0);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c9eac00",
   "metadata": {},
   "source": [
    "### 3.4. Normalize Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0fe29eea",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()\n",
    "scaled_data = scaler.fit_transform(df).astype(np.float32)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5aed722c",
   "metadata": {},
   "source": [
    "### 3.5. Create rolling window sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "0be3b548",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = []\n",
    "for i in range(len(df) - seq_len):\n",
    "    data.append(scaled_data[i:i + seq_len])\n",
    "\n",
    "n_windows = len(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6dedb8a",
   "metadata": {},
   "source": [
    "### 3.6. Create tf.data.Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e37369fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M1 Max\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:38:52.274108: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2023-02-03 10:38:52.274133: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    }
   ],
   "source": [
    "real_series = (tf.data.Dataset\n",
    "               .from_tensor_slices(data)\n",
    "               .shuffle(buffer_size=n_windows)\n",
    "               .batch(batch_size))\n",
    "real_series_iter = iter(real_series.repeat())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59d73ef7",
   "metadata": {},
   "source": [
    "### 3.7. Set up random series generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d6cea028",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_random_data():\n",
    "    while True:\n",
    "        yield np.random.uniform(low=0, high=1, size=(seq_len, n_seq))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6833803",
   "metadata": {},
   "source": [
    "We use the Python generator to feed a `tf.data.Dataset` that continues to call the random number generator as long as necessary and produces the desired batch size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "21e9d066",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:38:53.055126: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    }
   ],
   "source": [
    "random_series = iter(tf.data.Dataset\n",
    "                     .from_generator(make_random_data, output_types=tf.float32)\n",
    "                     .batch(batch_size)\n",
    "                     .repeat())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bfbe693f",
   "metadata": {},
   "source": [
    "## 4. TimeGAN Components"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82fe264e",
   "metadata": {},
   "source": [
    "The design of the TimeGAN components follows the author's sample code."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e45c8ec",
   "metadata": {},
   "source": [
    "### 4.1. Network Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d1292514",
   "metadata": {},
   "outputs": [],
   "source": [
    "hidden_dim = 24\n",
    "num_layers = 3"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a854b239",
   "metadata": {},
   "source": [
    "### 4.2. Set up logger"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "5993e75d",
   "metadata": {},
   "outputs": [],
   "source": [
    "writer = tf.summary.create_file_writer(log_dir.as_posix())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eabc9b8a",
   "metadata": {},
   "source": [
    "### 4.3. Input place holders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "283b7454",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = Input(shape=[seq_len, n_seq], name='RealData')\n",
    "Z = Input(shape=[seq_len, n_seq], name='RandomData')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73919f1b",
   "metadata": {},
   "source": [
    "### 4.4. RNN block generator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8304ce94",
   "metadata": {},
   "source": [
    "We keep it very simple and use a very similar architecture for all four components. For a real-world application, they should be tailored to the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "eaeec0b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_rnn(n_layers, hidden_units, output_units, name):\n",
    "    return Sequential([GRU(units=hidden_units,\n",
    "                           return_sequences=True,\n",
    "                           name=f'GRU_{i + 1}') for i in range(n_layers)] +\n",
    "                      [Dense(units=output_units,\n",
    "                             activation='sigmoid',\n",
    "                             name='OUT')], name=name)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9a4ef07",
   "metadata": {},
   "source": [
    "### 4.5. Embedder & Recovery"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b2623eef",
   "metadata": {},
   "outputs": [],
   "source": [
    "embedder = make_rnn(n_layers=3, \n",
    "                    hidden_units=hidden_dim, \n",
    "                    output_units=hidden_dim, \n",
    "                    name='Embedder')\n",
    "recovery = make_rnn(n_layers=3, \n",
    "                    hidden_units=hidden_dim, \n",
    "                    output_units=n_seq, \n",
    "                    name='Recovery')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a10fdcf",
   "metadata": {},
   "source": [
    "### 4.6. Generator & Discriminator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "157f2a37",
   "metadata": {},
   "outputs": [],
   "source": [
    "generator = make_rnn(n_layers=3, \n",
    "                     hidden_units=hidden_dim, \n",
    "                     output_units=hidden_dim, \n",
    "                     name='Generator')\n",
    "discriminator = make_rnn(n_layers=3, \n",
    "                         hidden_units=hidden_dim, \n",
    "                         output_units=1, \n",
    "                         name='Discriminator')\n",
    "supervisor = make_rnn(n_layers=2, \n",
    "                      hidden_units=hidden_dim, \n",
    "                      output_units=hidden_dim, \n",
    "                      name='Supervisor')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c223191",
   "metadata": {},
   "source": [
    "## 5. TimeGAN Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a2c474d",
   "metadata": {},
   "source": [
    "### 5.1. Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "41bf92cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_steps = 10000\n",
    "gamma = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b92c1800",
   "metadata": {},
   "source": [
    "### 5.2. Generic Loss Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f00371f2",
   "metadata": {},
   "outputs": [],
   "source": [
    "mse = MeanSquaredError()\n",
    "bce = BinaryCrossentropy()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc6d9368",
   "metadata": {},
   "source": [
    "## 6. Phase 1: Autoencoder Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91ea3c75",
   "metadata": {},
   "source": [
    "### 6.1. Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "4b69bb69",
   "metadata": {},
   "outputs": [],
   "source": [
    "H = embedder(X)\n",
    "X_tilde = recovery(H)\n",
    "\n",
    "autoencoder = Model(inputs=X,\n",
    "                    outputs=X_tilde,\n",
    "                    name='Autoencoder')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "a9dcf35d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"Autoencoder\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " RealData (InputLayer)       [(None, 24, 6)]           0         \n",
      "                                                                 \n",
      " Embedder (Sequential)       (None, 24, 24)            10104     \n",
      "                                                                 \n",
      " Recovery (Sequential)       (None, 24, 6)             10950     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 21,054\n",
      "Trainable params: 21,054\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "autoencoder.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "08f5ae76",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(autoencoder,\n",
    "           to_file=(results_path / 'autoencoder.png').as_posix(),\n",
    "           show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36348423",
   "metadata": {},
   "source": [
    "### 6.2. Autoencoder Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "dfb88a2c",
   "metadata": {},
   "outputs": [],
   "source": [
    "autoencoder_optimizer = Adam()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85f4759d",
   "metadata": {},
   "source": [
    "### 6.3. Autoencoder Training Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "c56fbd4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_autoencoder_init(x):\n",
    "    with tf.GradientTape() as tape:\n",
    "        x_tilde = autoencoder(x)\n",
    "        embedding_loss_t0 = mse(x, x_tilde)\n",
    "        e_loss_0 = 10 * tf.sqrt(embedding_loss_t0)\n",
    "\n",
    "    var_list = embedder.trainable_variables + recovery.trainable_variables\n",
    "    gradients = tape.gradient(e_loss_0, var_list)\n",
    "    autoencoder_optimizer.apply_gradients(zip(gradients, var_list))\n",
    "    return tf.sqrt(embedding_loss_t0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1b32501",
   "metadata": {},
   "source": [
    "### 6.4. Autoencoder Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "160a476f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                      | 0/10000 [00:00<?, ?it/s]2023-02-03 10:39:04.316206: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:04.611495: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:04.747177: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:04.797366: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:04.853074: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:04.903403: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:04.953955: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:05.173626: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:06.348349: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:06.422946: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:06.504608: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:06.578714: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:06.653340: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "  0%|▏                                                                            | 27/10000 [00:05<10:23, 15.99it/s]2023-02-03 10:39:08.908913: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.211082: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.260210: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.308959: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.376483: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.426699: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.478040: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.687606: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.796532: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.878400: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:09.973855: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:10.055611: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:39:10.131417: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "100%|██████████████████████████████████████████████████████████████████████████| 10000/10000 [07:40<00:00, 21.72it/s]\n"
     ]
    }
   ],
   "source": [
    "for step in tqdm(range(train_steps)):\n",
    "    X_ = next(real_series_iter)\n",
    "    step_e_loss_t0 = train_autoencoder_init(X_)\n",
    "    with writer.as_default():\n",
    "        tf.summary.scalar('Loss Autoencoder Init', step_e_loss_t0, step=step)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0d0db5a",
   "metadata": {},
   "source": [
    "GOOGLE_BIG: 10000/10000 [07:40<00:00, 21.72it/s]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b46801c2",
   "metadata": {},
   "source": [
    "### 6.5. Persist model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "6db1851a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Function `_wrapped_model` contains input name(s) RealData with unsupported characters which will be renamed to realdata in the SavedModel.\n",
      "WARNING:absl:Found untraced functions such as gru_cell_layer_call_fn, gru_cell_layer_call_and_return_conditional_losses, gru_cell_1_layer_call_fn, gru_cell_1_layer_call_and_return_conditional_losses, gru_cell_2_layer_call_fn while saving (showing 5 of 12). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: time_gan/experiment_00/autoencoder/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: time_gan/experiment_00/autoencoder/assets\n"
     ]
    }
   ],
   "source": [
    "# autoencoder.save(log_dir / 'autoencoder')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb0a8d01",
   "metadata": {},
   "source": [
    "## 7. Phase 2: Supervised training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8392aaf",
   "metadata": {},
   "source": [
    "### 7.1. Define Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "96c7018b",
   "metadata": {},
   "outputs": [],
   "source": [
    "supervisor_optimizer = Adam()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "344d8339",
   "metadata": {},
   "source": [
    "### 7.2. Train Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "6a70ef6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_supervisor(x):\n",
    "    with tf.GradientTape() as tape:\n",
    "        h = embedder(x)\n",
    "        h_hat_supervised = supervisor(h)\n",
    "        g_loss_s = mse(h[:, 1:, :], h_hat_supervised[:, :-1, :])\n",
    "\n",
    "    var_list = supervisor.trainable_variables\n",
    "    gradients = tape.gradient(g_loss_s, var_list)\n",
    "    supervisor_optimizer.apply_gradients(zip(gradients, var_list))\n",
    "    return g_loss_s"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a08dc280",
   "metadata": {},
   "source": [
    "### 7.3. Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "ffced413",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                      | 0/10000 [00:00<?, ?it/s]2023-02-03 10:48:33.675843: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:33.886859: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:33.943529: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:33.994264: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:34.046234: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:34.097347: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:34.276922: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:34.354703: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "  0%|                                                                            | 4/10000 [00:02<1:09:49,  2.39it/s]2023-02-03 10:48:35.253913: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.468466: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.520320: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.570006: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.619529: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.672868: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.742390: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:48:35.817682: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "100%|██████████████████████████████████████████████████████████████████████████| 10000/10000 [04:28<00:00, 37.29it/s]\n"
     ]
    }
   ],
   "source": [
    "for step in tqdm(range(train_steps)):\n",
    "    X_ = next(real_series_iter)\n",
    "    step_g_loss_s = train_supervisor(X_)\n",
    "    with writer.as_default():\n",
    "        tf.summary.scalar('Loss Generator Supervised Init', step_g_loss_s, step=step)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cc2ee97",
   "metadata": {},
   "source": [
    "GOOGLE_BIG: 10000/10000 [04:28<00:00, 37.29it/s]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe551b1e",
   "metadata": {},
   "source": [
    "### 7.4. Persist Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "1e5095a8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
      "WARNING:absl:Function `_wrapped_model` contains input name(s) GRU_1_input with unsupported characters which will be renamed to gru_1_input in the SavedModel.\n",
      "WARNING:absl:Found untraced functions such as gru_cell_12_layer_call_fn, gru_cell_12_layer_call_and_return_conditional_losses, gru_cell_13_layer_call_fn, gru_cell_13_layer_call_and_return_conditional_losses while saving (showing 4 of 4). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: time_gan/experiment_00/supervisor/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: time_gan/experiment_00/supervisor/assets\n"
     ]
    }
   ],
   "source": [
    "# supervisor.save(log_dir / 'supervisor')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af65109b",
   "metadata": {},
   "source": [
    "## 8. Joint Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d49c6182",
   "metadata": {},
   "source": [
    "### 8.1. Generator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3eef9c0d",
   "metadata": {},
   "source": [
    "#### 8.1.1. Adversarial Architecture - Supervised"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b75c87d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "E_hat = generator(Z)\n",
    "H_hat = supervisor(E_hat)\n",
    "Y_fake = discriminator(H_hat)\n",
    "\n",
    "adversarial_supervised = Model(inputs=Z,\n",
    "                               outputs=Y_fake,\n",
    "                               name='AdversarialNetSupervised')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "7aa5b495",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"AdversarialNetSupervised\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " RandomData (InputLayer)     [(None, 24, 6)]           0         \n",
      "                                                                 \n",
      " Generator (Sequential)      (None, 24, 24)            10104     \n",
      "                                                                 \n",
      " Supervisor (Sequential)     (None, 24, 24)            7800      \n",
      "                                                                 \n",
      " Discriminator (Sequential)  (None, 24, 1)             10825     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 28,729\n",
      "Trainable params: 28,729\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "adversarial_supervised.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "7051a04e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(adversarial_supervised, show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dcd05931",
   "metadata": {},
   "source": [
    "#### 8.1.2. Adversarial Architecture in Latent Space"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "4da66e8c",
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_fake_e = discriminator(E_hat)\n",
    "\n",
    "adversarial_emb = Model(inputs=Z,\n",
    "                    outputs=Y_fake_e,\n",
    "                    name='AdversarialNet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "d6a85da4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"AdversarialNet\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " RandomData (InputLayer)     [(None, 24, 6)]           0         \n",
      "                                                                 \n",
      " Generator (Sequential)      (None, 24, 24)            10104     \n",
      "                                                                 \n",
      " Discriminator (Sequential)  (None, 24, 1)             10825     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 20,929\n",
      "Trainable params: 20,929\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "adversarial_emb.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "5d719519",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(adversarial_emb, show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a14ae2f1",
   "metadata": {},
   "source": [
    "#### 8.1.3. Mean & Variance Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "f4214a1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_hat = recovery(H_hat)\n",
    "synthetic_data = Model(inputs=Z,\n",
    "                       outputs=X_hat,\n",
    "                       name='SyntheticData')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "2f8b5414",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"SyntheticData\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " RandomData (InputLayer)     [(None, 24, 6)]           0         \n",
      "                                                                 \n",
      " Generator (Sequential)      (None, 24, 24)            10104     \n",
      "                                                                 \n",
      " Supervisor (Sequential)     (None, 24, 24)            7800      \n",
      "                                                                 \n",
      " Recovery (Sequential)       (None, 24, 6)             10950     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 28,854\n",
      "Trainable params: 28,854\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "synthetic_data.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "7d1c8d8e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(synthetic_data, show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "f44ee900",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_generator_moment_loss(y_true, y_pred):\n",
    "    y_true_mean, y_true_var = tf.nn.moments(x=y_true, axes=[0])\n",
    "    y_pred_mean, y_pred_var = tf.nn.moments(x=y_pred, axes=[0])\n",
    "    g_loss_mean = tf.reduce_mean(tf.abs(y_true_mean - y_pred_mean))\n",
    "    g_loss_var = tf.reduce_mean(tf.abs(tf.sqrt(y_true_var + 1e-6) - tf.sqrt(y_pred_var + 1e-6)))\n",
    "    return g_loss_mean + g_loss_var"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dcbe5322",
   "metadata": {},
   "source": [
    "### 8.2. Discriminator"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1cf3fc2",
   "metadata": {},
   "source": [
    "#### 8.2.1. Architecture: Real Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "76943934",
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_real = discriminator(H)\n",
    "discriminator_model = Model(inputs=X,\n",
    "                            outputs=Y_real,\n",
    "                            name='DiscriminatorReal')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "1110cc83",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"DiscriminatorReal\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " RealData (InputLayer)       [(None, 24, 6)]           0         \n",
      "                                                                 \n",
      " Embedder (Sequential)       (None, 24, 24)            10104     \n",
      "                                                                 \n",
      " Discriminator (Sequential)  (None, 24, 1)             10825     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 20,929\n",
      "Trainable params: 20,929\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "discriminator_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "805f3e7d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_model(discriminator_model, show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "689f1224",
   "metadata": {},
   "source": [
    "### 8.3. Optimizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "ab0a4a19",
   "metadata": {},
   "outputs": [],
   "source": [
    "generator_optimizer = Adam()\n",
    "discriminator_optimizer = Adam()\n",
    "embedding_optimizer = Adam()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08b32fbe",
   "metadata": {},
   "source": [
    "### 8.4. Generator Train Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "ce03afe3",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_generator(x, z):\n",
    "    with tf.GradientTape() as tape:\n",
    "        y_fake = adversarial_supervised(z)\n",
    "        generator_loss_unsupervised = bce(y_true=tf.ones_like(y_fake),\n",
    "                                          y_pred=y_fake)\n",
    "\n",
    "        y_fake_e = adversarial_emb(z)\n",
    "        generator_loss_unsupervised_e = bce(y_true=tf.ones_like(y_fake_e),\n",
    "                                            y_pred=y_fake_e)\n",
    "        h = embedder(x)\n",
    "        h_hat_supervised = supervisor(h)\n",
    "        generator_loss_supervised = mse(h[:, 1:, :], h_hat_supervised[:, 1:, :])\n",
    "\n",
    "        x_hat = synthetic_data(z)\n",
    "        generator_moment_loss = get_generator_moment_loss(x, x_hat)\n",
    "\n",
    "        generator_loss = (generator_loss_unsupervised +\n",
    "                          generator_loss_unsupervised_e +\n",
    "                          100 * tf.sqrt(generator_loss_supervised) +\n",
    "                          100 * generator_moment_loss)\n",
    "\n",
    "    var_list = generator.trainable_variables + supervisor.trainable_variables\n",
    "    gradients = tape.gradient(generator_loss, var_list)\n",
    "    generator_optimizer.apply_gradients(zip(gradients, var_list))\n",
    "    return generator_loss_unsupervised, generator_loss_supervised, generator_moment_loss"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb6c7028",
   "metadata": {},
   "source": [
    "### 8.5. Embedding Train Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "f4ecfd32",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_embedder(x):\n",
    "    with tf.GradientTape() as tape:\n",
    "        h = embedder(x)\n",
    "        h_hat_supervised = supervisor(h)\n",
    "        generator_loss_supervised = mse(h[:, 1:, :], h_hat_supervised[:, 1:, :])\n",
    "\n",
    "        x_tilde = autoencoder(x)\n",
    "        embedding_loss_t0 = mse(x, x_tilde)\n",
    "        e_loss = 10 * tf.sqrt(embedding_loss_t0) + 0.1 * generator_loss_supervised\n",
    "\n",
    "    var_list = embedder.trainable_variables + recovery.trainable_variables\n",
    "    gradients = tape.gradient(e_loss, var_list)\n",
    "    embedding_optimizer.apply_gradients(zip(gradients, var_list))\n",
    "    return tf.sqrt(embedding_loss_t0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4874451f",
   "metadata": {},
   "source": [
    "### 8.6. Discriminator Train Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "ae0eaa22",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def get_discriminator_loss(x, z):\n",
    "    y_real = discriminator_model(x)\n",
    "    discriminator_loss_real = bce(y_true=tf.ones_like(y_real),\n",
    "                                  y_pred=y_real)\n",
    "\n",
    "    y_fake = adversarial_supervised(z)\n",
    "    discriminator_loss_fake = bce(y_true=tf.zeros_like(y_fake),\n",
    "                                  y_pred=y_fake)\n",
    "\n",
    "    y_fake_e = adversarial_emb(z)\n",
    "    discriminator_loss_fake_e = bce(y_true=tf.zeros_like(y_fake_e),\n",
    "                                    y_pred=y_fake_e)\n",
    "    return (discriminator_loss_real +\n",
    "            discriminator_loss_fake +\n",
    "            gamma * discriminator_loss_fake_e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "27dd7c80",
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_discriminator(x, z):\n",
    "    with tf.GradientTape() as tape:\n",
    "        discriminator_loss = get_discriminator_loss(x, z)\n",
    "\n",
    "    var_list = discriminator.trainable_variables\n",
    "    gradients = tape.gradient(discriminator_loss, var_list)\n",
    "    discriminator_optimizer.apply_gradients(zip(gradients, var_list))\n",
    "    return discriminator_loss"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eeea8649",
   "metadata": {},
   "source": [
    "### 8.7. Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "e7b8c2f6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:59:23.427436: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.651653: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.651835: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.651856: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.651955: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.961476: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.962282: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.963331: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:24.964291: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.096088: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.097101: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.097828: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.098855: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.242299: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.243013: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.244194: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.245318: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.379881: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.380466: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.381642: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.382432: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.521360: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.526142: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.527228: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.742479: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.743356: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:25.743781: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.039308: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.084339: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.085070: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.138099: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.142915: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.268756: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.535329: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.571951: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.673804: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.679154: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.761673: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.763399: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.765647: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.877285: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.882636: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:26.938179: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.008779: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.068590: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.177611: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.200715: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.338929: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.344346: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.484256: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.485211: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:27.619355: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:30.281656: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:59:30.850287: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:30.850336: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:30.942576: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:30.943812: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.035470: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.036584: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.131962: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.133230: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.216869: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.217785: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.301145: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.322711: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.400264: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.471699: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.473935: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.703925: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.704883: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.824249: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.824913: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:31.969990: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:32.063599: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:32.158562: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.479671: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.841258: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.841271: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.841310: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.954096: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.955116: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:34.956319: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.073506: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.074642: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.075984: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.198688: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.199936: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.201504: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.319217: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.320295: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.321581: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.433748: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.434727: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.437387: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.551894: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:35.625568: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:37.115943: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:37.890687: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:37.890727: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:37.890876: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.015009: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.016042: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.017368: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.130695: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.131750: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.132661: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.252225: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.253351: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.254456: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:59:38.366634: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.396547: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.398043: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.515928: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.517269: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.518675: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.638401: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.896399: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.896425: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:38.897656: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.109890: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.112423: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.259334: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.431733: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.433970: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.447263: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:39.625447: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0 | d_loss: 2.0486 | g_loss_u: 0.8183 | g_loss_s: 0.0042 | g_loss_v: 0.7469 | e_loss_t0: 0.0417\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:59:45.896769: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.140391: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.140450: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.140475: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.140486: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.403801: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.404734: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.406057: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.407155: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.558883: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.560008: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.561803: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.563132: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.725698: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.728289: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.730563: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.732072: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.839154: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.863663: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.887208: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.888536: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:47.989990: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.071220: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.072629: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.160603: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.226605: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.230083: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.246198: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.354679: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.417276: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.420571: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.462720: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.483856: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.625900: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.773682: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.840763: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.842519: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.845355: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.962819: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:48.964022: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.056556: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.075836: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.130106: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.144340: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.419195: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.421684: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.609020: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.619514: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.777099: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.778525: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.945972: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:49.947688: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:51.444741: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 10:59:52.001450: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.001571: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.109499: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.111265: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.215047: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.216815: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.325057: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.327091: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.430209: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.431978: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.534957: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.566181: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.649902: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.727763: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.748022: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.955941: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:52.957464: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:53.120724: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:53.122350: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:53.311329: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:53.429855: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 10:59:53.553188: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.033163: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.397642: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.397690: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.397704: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.529285: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.530873: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.532743: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.657783: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.659246: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.661166: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.806592: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.808652: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.811086: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.942573: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.944118: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:06.946015: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:07.070869: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:07.074233: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:07.078828: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:07.215894: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:07.295435: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:08.613266: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.386855: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.386879: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.386902: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.529060: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.530738: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.532606: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.637593: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.661448: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.663541: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.751126: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.807091: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.809551: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-03 11:00:09.891094: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.941204: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:09.946887: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.047092: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.054551: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.091174: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.199663: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.274492: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.323865: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.423518: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.530146: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.630691: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.646808: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.663392: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.838115: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:10.854391: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n",
      "2023-02-03 11:00:11.002758: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 1,000 | d_loss: 0.1299 | g_loss_u: 3.3986 | g_loss_s: 0.0532 | g_loss_v: 0.7612 | e_loss_t0: 0.0427\n",
      " 2,000 | d_loss: 0.1065 | g_loss_u: 5.1381 | g_loss_s: 0.1472 | g_loss_v: 0.7678 | e_loss_t0: 0.0349\n",
      " 3,000 | d_loss: 1.0576 | g_loss_u: 1.0446 | g_loss_s: 0.0542 | g_loss_v: 0.5752 | e_loss_t0: 0.0521\n",
      " 4,000 | d_loss: 0.1406 | g_loss_u: 4.6089 | g_loss_s: 0.1230 | g_loss_v: 0.5067 | e_loss_t0: 0.0417\n",
      " 5,000 | d_loss: 0.1104 | g_loss_u: 4.0239 | g_loss_s: 0.2800 | g_loss_v: 0.6347 | e_loss_t0: 0.0401\n",
      " 6,000 | d_loss: 0.1212 | g_loss_u: 4.0914 | g_loss_s: 0.2253 | g_loss_v: 0.4564 | e_loss_t0: 0.0451\n",
      " 7,000 | d_loss: 0.0985 | g_loss_u: 4.0673 | g_loss_s: 0.1251 | g_loss_v: 0.5122 | e_loss_t0: 0.0402\n",
      " 8,000 | d_loss: 0.0949 | g_loss_u: 4.5053 | g_loss_s: 0.0323 | g_loss_v: 0.4036 | e_loss_t0: 0.0440\n",
      " 9,000 | d_loss: 0.0827 | g_loss_u: 4.5532 | g_loss_s: 0.0741 | g_loss_v: 0.5130 | e_loss_t0: 0.0657\n"
     ]
    }
   ],
   "source": [
    "step_g_loss_u = step_g_loss_s = step_g_loss_v = step_e_loss_t0 = step_d_loss = 0\n",
    "for step in range(train_steps):\n",
    "    # Train generator (twice as often as discriminator)\n",
    "    for kk in range(2):\n",
    "        X_ = next(real_series_iter)\n",
    "        Z_ = next(random_series)\n",
    "\n",
    "        # Train generator\n",
    "        step_g_loss_u, step_g_loss_s, step_g_loss_v = train_generator(X_, Z_)\n",
    "        # Train embedder\n",
    "        step_e_loss_t0 = train_embedder(X_)\n",
    "\n",
    "    X_ = next(real_series_iter)\n",
    "    Z_ = next(random_series)\n",
    "    step_d_loss = get_discriminator_loss(X_, Z_)\n",
    "    if step_d_loss > 0.15:\n",
    "        step_d_loss = train_discriminator(X_, Z_)\n",
    "\n",
    "    if step % 1000 == 0:\n",
    "        print(f'{step:6,.0f} | d_loss: {step_d_loss:6.4f} | g_loss_u: {step_g_loss_u:6.4f} | '\n",
    "              f'g_loss_s: {step_g_loss_s:6.4f} | g_loss_v: {step_g_loss_v:6.4f} | e_loss_t0: {step_e_loss_t0:6.4f}')\n",
    "\n",
    "    with writer.as_default():\n",
    "        tf.summary.scalar('G Loss S', step_g_loss_s, step=step)\n",
    "        tf.summary.scalar('G Loss U', step_g_loss_u, step=step)\n",
    "        tf.summary.scalar('G Loss V', step_g_loss_v, step=step)\n",
    "        tf.summary.scalar('E Loss T0', step_e_loss_t0, step=step)\n",
    "        tf.summary.scalar('D Loss', step_d_loss, step=step)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ffa8a6e7",
   "metadata": {},
   "source": [
    "GOOGLE_BIG:  \n",
    "\n",
    "| step? | G Loss S | G Loss U | G Loss V | E Loss T0 | D Loss | \n",
    "| --- | --- | --- | --- | --- | --- |\n",
    "|0 | d_loss: 2.0486 | g_loss_u: 0.8183 | g_loss_s: 0.0042 | g_loss_v: 0.7469 | e_loss_t0: 0.0417 |\n",
    "| 1,000 | d_loss: 0.1299 | g_loss_u: 3.3986 | g_loss_s: 0.0532 | g_loss_v: 0.7612 | e_loss_t0: 0.0427 |\n",
    "| 2,000 | d_loss: 0.1065 | g_loss_u: 5.1381 | g_loss_s: 0.1472 | g_loss_v: 0.7678 | e_loss_t0: 0.0349 |\n",
    "| 3,000 | d_loss: 1.0576 | g_loss_u: 1.0446 | g_loss_s: 0.0542 | g_loss_v: 0.5752 | e_loss_t0: 0.0521 |\n",
    "| 4,000 | d_loss: 0.1406 | g_loss_u: 4.6089 | g_loss_s: 0.1230 | g_loss_v: 0.5067 | e_loss_t0: 0.0417 |\n",
    "| 5,000 | d_loss: 0.1104 | g_loss_u: 4.0239 | g_loss_s: 0.2800 | g_loss_v: 0.6347 | e_loss_t0: 0.0401 |\n",
    "| 6,000 | d_loss: 0.1212 | g_loss_u: 4.0914 | g_loss_s: 0.2253 | g_loss_v: 0.4564 | e_loss_t0: 0.0451 |\n",
    "| 7,000 | d_loss: 0.0985 | g_loss_u: 4.0673 | g_loss_s: 0.1251 | g_loss_v: 0.5122 | e_loss_t0: 0.0402 |\n",
    "| 8,000 | d_loss: 0.0949 | g_loss_u: 4.5053 | g_loss_s: 0.0323 | g_loss_v: 0.4036 | e_loss_t0: 0.0440 |\n",
    "| 9,000 | d_loss: 0.0827 | g_loss_u: 4.5532 | g_loss_s: 0.0741 | g_loss_v: 0.5130 | e_loss_t0: 0.0657 |\n",
    "\n",
    "~2.5 hours"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5cb700da",
   "metadata": {},
   "source": [
    "### 8.8. Persist Synthetic Data Generator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "fd1ba98f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Compiled the loaded model, but the compiled metrics have yet to be built. `model.compile_metrics` will be empty until you train or evaluate the model.\n",
      "WARNING:absl:Function `_wrapped_model` contains input name(s) RandomData with unsupported characters which will be renamed to randomdata in the SavedModel.\n",
      "WARNING:absl:Found untraced functions such as gru_cell_6_layer_call_fn, gru_cell_6_layer_call_and_return_conditional_losses, gru_cell_7_layer_call_fn, gru_cell_7_layer_call_and_return_conditional_losses, gru_cell_8_layer_call_fn while saving (showing 5 of 16). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: time_gan/experiment_00/synthetic_data/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: time_gan/experiment_00/synthetic_data/assets\n"
     ]
    }
   ],
   "source": [
    "# synthetic_data.save(log_dir / 'synthetic_data')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fabcbe82",
   "metadata": {},
   "source": [
    "## 9. Generate Synthetic Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "53e1d760",
   "metadata": {},
   "outputs": [],
   "source": [
    "generated_data = []\n",
    "for i in range(int(n_windows / batch_size)):\n",
    "    Z_ = next(random_series)\n",
    "    d = synthetic_data(Z_)\n",
    "    generated_data.append(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "bf34c07a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "28"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(generated_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "b0e10c4f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3584, 24, 6)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "generated_data = np.array(np.vstack(generated_data))\n",
    "generated_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "7386462d",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save(log_dir / 'generated_data.npy', generated_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34c7bb78",
   "metadata": {},
   "source": [
    "### 9.1. Rescale"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "b4b52c14",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3584, 24, 6)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "generated_data = (scaler.inverse_transform(generated_data\n",
    "                                           .reshape(-1, n_seq))\n",
    "                  .reshape(-1, seq_len, n_seq))\n",
    "generated_data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d37dcb9",
   "metadata": {},
   "source": [
    "### 9.2. Persist Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "22cb095c",
   "metadata": {},
   "outputs": [],
   "source": [
    "with pd.HDFStore(hdf_store) as store:\n",
    "    store.put('data/synthetic', pd.DataFrame(generated_data.reshape(-1, n_seq),\n",
    "                                             columns=features))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c97ef71",
   "metadata": {},
   "source": [
    "### 9.3. Plot sample Series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "acd31dc3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1400x700 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=3, ncols=2, figsize=(14, 7))\n",
    "axes = axes.flatten()\n",
    "\n",
    "index = list(range(1, 25))\n",
    "synthetic = generated_data[np.random.randint(n_windows)]\n",
    "\n",
    "idx = np.random.randint(len(df) - seq_len)\n",
    "real = df.iloc[idx: idx + seq_len]\n",
    "\n",
    "for j, feature in enumerate(features):\n",
    "    (pd.DataFrame({'Real': real.iloc[:, j].values,\n",
    "                   'Synthetic': synthetic[:, j]})\n",
    "     .plot(ax=axes[j],\n",
    "           title=feature,\n",
    "           secondary_y='Synthetic', style=['-', '--'],\n",
    "           lw=1))\n",
    "sns.despine()\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fd664e09",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:TimeGANEnv]",
   "language": "python",
   "name": "conda-env-TimeGANEnv-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}