timegan.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eH9WWJoxzSkb"
      },
      "source": [
        "# **TimeGAN**\n",
        "Unofficial implementation in PyTorch.\n",
        "\n",
        "Jinsung Yoon, Daniel Jarrett, Mihaela van der Schaar, \"Time-series Generative Adversarial Networks,\" Neural Information Processing Systems (NeurIPS), 2019.\n",
        "\n",
        "Notebook Author: https://github.com/haufjan"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "718KZdM1zb4J"
      },
      "source": [
        "### **Import**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "ssg8LF2LzNN-"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "import time\n",
        "from itertools import chain, cycle\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "import torch\n",
        "from torch import nn, tensor\n",
        "from torch.utils.data import TensorDataset, DataLoader\n",
        "\n",
        "import sklearn\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.decomposition import PCA\n",
        "from sklearn.manifold import TSNE\n",
        "from sklearn.utils import shuffle\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "\n",
        "import matplotlib as mpl\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import tqdm\n",
        "from tqdm import tqdm"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Random Seeds**"
      ],
      "metadata": {
        "id": "826NOxKd34hw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "np.random.seed(42)\n",
        "torch.manual_seed(42)"
      ],
      "metadata": {
        "id": "3J0RX1LJ33Pd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "495ea4fb-26b8-4459-8032-a3eb4f91b206"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<torch._C.Generator at 0x7dd55d8b9150>"
            ]
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Str9BmbZ0Pga"
      },
      "source": [
        "### **Functions**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "lz68w1nGTMbc"
      },
      "outputs": [],
      "source": [
        "def loading(*files: str) -> pd.DataFrame:\n",
        "    \"\"\"\n",
        "    Load data from csv file\n",
        "    \"\"\"\n",
        "    return_list = []\n",
        "    for file in files:\n",
        "        path = f'{file}' if file.endswith('csv') else f'{file}.csv'\n",
        "        return_list.extend([pd.read_csv(path)])\n",
        "\n",
        "    return return_list if len(return_list) > 1 else return_list.pop()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "QDtCTXsFTMbc"
      },
      "outputs": [],
      "source": [
        "def preprocessing(*inputs: tuple, horizon: int, shuffle_stack: bool = True, random_state: int = None) -> np.ndarray:\n",
        "    \"\"\"\n",
        "    Conduct preprocessing: scale data, slice data into sequences and shuffle data stack\n",
        "    \"\"\"\n",
        "    return_list = []\n",
        "    for data, bool_scale in inputs:\n",
        "        #Create Minimum-Maximum scaler\n",
        "        if bool_scale:\n",
        "            scaler = MinMaxScaler().fit(data)\n",
        "            data = scaler.transform(data)\n",
        "            print('\\nMaximum values:\\n', scaler.data_max_, '\\nMinimum values:\\n', scaler.data_min_)\n",
        "\n",
        "        #Create a list holding the sequences defined by sliding window of width = horizon and stack to a 3-dimensional array (batch, horizon, feature)\n",
        "        data_stack = np.stack([data[i:i+horizon] for i in range(len(data) - horizon)])\n",
        "\n",
        "        if shuffle_stack:\n",
        "            #In TimeGAN code the dataset is mixed to make it similar to independent and identically distributed (iid)\n",
        "            data_stack = shuffle(data_stack, random_state=random_state)\n",
        "\n",
        "        return_list.extend([data_stack, scaler.data_max_, scaler.data_min_]) if bool_scale else return_list.extend([data_stack])\n",
        "\n",
        "    return return_list if len(return_list) > 1 else return_list.pop()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "u7S0g9k_xmdu"
      },
      "outputs": [],
      "source": [
        "def low_dimensional_representation(data_ori: np.ndarray, data_gen: np.ndarray, technique: str) -> np.ndarray:\n",
        "    \"\"\"\n",
        "    Transform data into low-dimensional representation\n",
        "    \"\"\"\n",
        "    assert technique in ['pca', 'tsne']\n",
        "\n",
        "    sample_no = min([1000, len(data_ori)])\n",
        "    idx = np.random.permutation(len(data_ori))[:sample_no]\n",
        "    data_ori, data_gen = np.asarray(data_ori)[idx], np.asarray(data_gen)[idx]\n",
        "    no, seq_len, dim = data_ori.shape\n",
        "\n",
        "    prep_data = np.array([np.reshape(np.mean(data_ori[i,:,:], 1), [1,seq_len]).flatten().tolist() for i in range(sample_no)])\n",
        "    prep_data_hat = np.array([np.reshape(np.mean(data_gen[i,:,:], 1), [1,seq_len]).flatten().tolist() for i in range(sample_no)])\n",
        "\n",
        "    if technique == 'pca':\n",
        "        # PCA Analysis\n",
        "        pca = PCA(n_components = 2)\n",
        "        pca.fit(prep_data)\n",
        "\n",
        "        return pca.transform(prep_data), pca.transform(prep_data_hat)\n",
        "\n",
        "    elif technique == 'tsne':\n",
        "        # Do t-SNE Analysis together\n",
        "        prep_data_final = np.concatenate((prep_data, prep_data_hat), axis = 0)\n",
        "\n",
        "        # TSNE anlaysis\n",
        "        tsne = TSNE(n_components = 2, verbose = 1, perplexity = 40, max_iter = 300)\n",
        "        tsne_result = tsne.fit_transform(prep_data_final)\n",
        "\n",
        "        return tsne_result[:sample_no], tsne_result[sample_no:]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "lTUybux8xtrt"
      },
      "outputs": [],
      "source": [
        "def plot_distribution_estimate(lowdim_data_ori: np.ndarray, lowdim_data_gen: np.ndarray, technique: str) -> None:\n",
        "    \"\"\"\n",
        "    Visualize distribution estimate via sampled data points\n",
        "    \"\"\"\n",
        "    fig, ax = plt.subplots()\n",
        "    #Use rasterized Line2D Artist here instead of scatter. Espacially benefitial when exporting the plot as pdf file (or other vector graphics).\n",
        "    ax.add_line(mpl.lines.Line2D(lowdim_data_ori[:,0], lowdim_data_ori[:,1], ls='', marker='o', markersize=3, c = 'red', alpha = 0.2, rasterized=True, label='Original'))\n",
        "    ax.add_line(mpl.lines.Line2D(lowdim_data_gen[:,0], lowdim_data_gen[:,1], ls='', marker='o', markersize=3, c = 'blue', alpha = 0.2, rasterized=True, label='Generated'))\n",
        "    ax.autoscale_view()\n",
        "    ax.set_xlabel(f'x-{technique}')\n",
        "    ax.set_ylabel(f'y-{technique}')\n",
        "    ax.legend()\n",
        "    fig.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fAihUkKATMbe"
      },
      "source": [
        "### **Load and prepare Data**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P7s7qj5E0Uaq",
        "outputId": "aa0bb498-3cfa-46fe-84ce-19d0f1f96f37"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Maximum values:\n",
            " [1.27100000e+03 1.27389001e+03 1.24902002e+03 1.26832996e+03\n",
            " 1.26832996e+03 8.27681000e+07] \n",
            "Minimum values:\n",
            " [  49.274517   50.541279   47.669952   49.681866   49.681866 7900.      ]\n"
          ]
        }
      ],
      "source": [
        "#Data set\n",
        "data = loading(os.path.join('stock_data'))\n",
        "\n",
        "#Squence length\n",
        "HORIZON = 24\n",
        "\n",
        "#Flip stock_data to make it chronological\n",
        "data = np.asarray(data)[::-1]\n",
        "\n",
        "#Preprocess data\n",
        "data_train, max_val, min_val = preprocessing((data, True), horizon=HORIZON)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "XStlLg8BTMbf",
        "outputId": "a06722ea-a17d-4034-f0ee-b1e90ffb99af"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#Visualize scaled training data\n",
        "fig, ax = plt.subplots()\n",
        "ax.plot((data - min_val)/(max_val - min_val))\n",
        "ax.set_xlim(0, len(data))\n",
        "ax.set_xlabel('Time')\n",
        "ax.set_ylabel('Magnitude')\n",
        "ax.set_title('Scaled Data')\n",
        "ax.grid(True)\n",
        "fig.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pdDKS37x0VC4"
      },
      "source": [
        "### **Device**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8zckeTt-0ZMW",
        "outputId": "863cace4-ed36-4e5c-8646-2c7c6b0db967"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Using cuda device\n"
          ]
        }
      ],
      "source": [
        "device = (\n",
        "    'cuda'\n",
        "    if torch.cuda.is_available()\n",
        "    else 'mps'\n",
        "    if torch.backends.mps.is_available()\n",
        "    else 'cpu'\n",
        ")\n",
        "print(f'Using {device} device')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kKniR--0wUJ9"
      },
      "source": [
        "### **Networks**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gLlo4PXEzs0s"
      },
      "source": [
        "##### **Embedder**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "2WD5tfQzzsSs"
      },
      "outputs": [],
      "source": [
        "class Embedder(nn.Module):\n",
        "    def __init__(self, module_name, input_features, hidden_dim, num_layers):\n",
        "        super().__init__()\n",
        "        assert module_name in ['gru', 'lstm']\n",
        "        if module_name == 'gru':\n",
        "            self.rnn = nn.GRU(input_size=input_features, hidden_size=hidden_dim, num_layers=num_layers, batch_first=True)\n",
        "        elif module_name == 'lstm':\n",
        "            self.rnn = nn.LSTM(input_size=input_features, hidden_size=hidden_dim, num_layers=num_layers, batch_first=True)\n",
        "        else:\n",
        "            raise Exception()\n",
        "        self.model = nn.Sequential(nn.Linear(hidden_dim, hidden_dim),\n",
        "                                   nn.Sigmoid())\n",
        "\n",
        "    def forward(self, x):\n",
        "        seq, _ = self.rnn(x)\n",
        "        return self.model(seq)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VVIg02nC0ETI"
      },
      "source": [
        "##### **Recovery**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "id": "0PbPeQT90Dya"
      },
      "outputs": [],
      "source": [
        "class Recovery(nn.Module):\n",
        "    def __init__(self, module_name, input_features, hidden_dim, num_layers):\n",
        "        super().__init__()\n",
        "        assert module_name in ['gru', 'lstm']\n",
        "        if module_name == 'gru':\n",
        "            self.rnn = nn.GRU(input_size=hidden_dim, hidden_size=hidden_dim, num_layers=num_layers, batch_first=True)\n",
        "        elif module_name == 'lstm':\n",
        "            self.rnn = nn.LSTM(input_size=hidden_dim, hidden_size=hidden_dim, num_layers=num_layers, batch_first=True)\n",
        "        else:\n",
        "            raise Exception()\n",
        "        self.model = nn.Sequential(nn.Linear(hidden_dim, input_features),\n",
        "                                   nn.Sigmoid())\n",
        "\n",
        "    def forward(self, x):\n",
        "        seq, _ = self.rnn(x)\n",
        "        return self.model(seq)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4UVXwyaG1Yf9"
      },
      "source": [
        "##### **Generator**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "tyEUX2_P1bX0"
      },
      "outputs": [],
      "source": [
        "class Generator(nn.Module):\n",
        "    def __init__(self, module_name, input_features, hidden_dim, num_layers):\n",
        "        super().__init__()\n",
        "        assert module_name in ['gru', 'lstm']\n",
        "        if module_name == 'gru':\n",
        "            self.rnn = nn.GRU(input_size=input_features, hidden_size=hidden_dim, num_layers=num_layers, batch_first=True)\n",
        "        elif module_name == 'lstm':\n",
        "            self.rnn = nn.LSTM(input_size=input_features, hidden_size=hidden_dim, num_layers=num_layers, batch_first=True)\n",
        "        else:\n",
        "            raise Exception()\n",
        "        self.model = nn.Sequential(nn.Linear(hidden_dim, hidden_dim),\n",
        "                                   nn.Sigmoid())\n",
        "\n",
        "    def forward(self, x):\n",
        "        seq, _ = self.rnn(x)\n",
        "        return self.model(seq)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6KkUQGDY4lHl"
      },
      "source": [
        "##### **Supervisor**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "id": "r3CPhuUw4qQd"
      },
      "outputs": [],
      "source": [
        "class Supervisor(nn.Module):\n",
        "    def __init__(self, module_name, hidden_dim, num_layers):\n",
        "        super().__init__()\n",
        "        assert module_name in ['gru', 'lstm']\n",
        "        if module_name == 'gru':\n",
        "            self.rnn = nn.GRU(input_size=hidden_dim, hidden_size=hidden_dim, num_layers=num_layers-1, batch_first=True)\n",
        "        elif module_name == 'lstm':\n",
        "            self.rnn = nn.LSTM(input_size=hidden_dim, hidden_size=hidden_dim, num_layers=num_layers-1, batch_first=True)\n",
        "        else:\n",
        "            raise Exception()\n",
        "        self.model = nn.Sequential(nn.Linear(hidden_dim, hidden_dim),\n",
        "                                   nn.Sigmoid())\n",
        "\n",
        "    def forward(self, x):\n",
        "        seq, _ = self.rnn(x)\n",
        "        return self.model(seq)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_iiSZvsM73xG"
      },
      "source": [
        "##### **Discriminator**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "I8YhO_bk726j"
      },
      "outputs": [],
      "source": [
        "class Discriminator(nn.Module):\n",
        "    def __init__(self, module_name, hidden_dim, num_layers):\n",
        "        super().__init__()\n",
        "        assert module_name in ['gru', 'lstm']\n",
        "        if module_name == 'gru':\n",
        "            self.rnn = nn.GRU(input_size=hidden_dim, hidden_size=hidden_dim, num_layers=num_layers, bidirectional=False, batch_first=True)\n",
        "        elif module_name == 'lstm':\n",
        "            self.rnn = nn.LSTM(input_size=hidden_dim, hidden_size=hidden_dim, num_layers=num_layers, bidirectional=False, batch_first=True)\n",
        "        else:\n",
        "            raise Exception()\n",
        "        #If bidirectional = True\n",
        "        # self.model = nn.Linear(2*hidden_dim, 1)\n",
        "        #If bidirectional = False\n",
        "        self.model = nn.Linear(hidden_dim, 1)\n",
        "\n",
        "    def forward(self, x):\n",
        "        seq, _ = self.rnn(x)\n",
        "        return self.model(seq)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UmFuEzQPJQMs"
      },
      "source": [
        "### **Loss Functions**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "id": "j0MnVNSDRz_6"
      },
      "outputs": [],
      "source": [
        "def embedder_loss(x, x_tilde):\n",
        "    return 10*torch.sqrt(nn.MSELoss()(x_tilde, x))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "XapuumlD_oii"
      },
      "outputs": [],
      "source": [
        "def supervised_loss(h, h_hat_supervise):\n",
        "    return nn.MSELoss()(h_hat_supervise[:,:-1,:], h[:,1:,:])"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def generator_loss(y_fake, y_fake_e, h, h_hat_supervise, x, x_hat):\n",
        "    gamma = 1\n",
        "    fake = torch.ones_like(y_fake, dtype=torch.float32, device=device, requires_grad=False)\n",
        "\n",
        "    #1. Unsupervised generator loss\n",
        "    g_loss_u = nn.BCEWithLogitsLoss()(y_fake, fake)\n",
        "    g_loss_u_e = nn.BCEWithLogitsLoss()(y_fake_e, fake)\n",
        "\n",
        "    #2. Supervised loss\n",
        "    g_loss_s = nn.MSELoss()(h_hat_supervise[:,:-1,:], h[:,1:,:])\n",
        "\n",
        "    #3. Two moments\n",
        "    g_loss_v1 = torch.mean(torch.abs(torch.sqrt(torch.std(x_hat, dim=0)) - torch.sqrt(torch.std(x, dim=0))))\n",
        "    g_loss_v2 = torch.mean(torch.abs(torch.mean(x_hat, dim=0) - torch.mean(x, dim=0)))\n",
        "    g_loss_v = g_loss_v1 + g_loss_v2\n",
        "\n",
        "    return g_loss_u + gamma*g_loss_u_e + 100*torch.sqrt(g_loss_s) + 100*g_loss_v"
      ],
      "metadata": {
        "id": "IZK7UyXf9xNp"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def discriminator_loss(y_real, y_fake, y_fake_e):\n",
        "    gamma = 1\n",
        "    valid = torch.ones_like(y_real, dtype=torch.float32, device=device, requires_grad=False)\n",
        "    fake = torch.zeros_like(y_fake, dtype=torch.float32, device=device, requires_grad=False)\n",
        "\n",
        "    d_loss_real = nn.BCEWithLogitsLoss()(y_real, valid)\n",
        "    d_loss_fake = nn.BCEWithLogitsLoss()(y_fake, fake)\n",
        "    d_loss_fake_e = nn.BCEWithLogitsLoss()(y_fake_e, fake)\n",
        "\n",
        "    return d_loss_real + d_loss_fake + d_loss_fake_e*gamma"
      ],
      "metadata": {
        "id": "ssimvFaM9y29"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xzniJZM_TQDb"
      },
      "source": [
        "### **TimeGAN**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "zm-mY4o0TOBs"
      },
      "outputs": [],
      "source": [
        "class TimeGAN(nn.Module):\n",
        "    def __init__(self,\n",
        "                 module_name: str = 'gru',\n",
        "                 input_features: int = 1,\n",
        "                 hidden_dim: int = 8,\n",
        "                 num_layers: int = 3,\n",
        "                 epochs: int = 100,\n",
        "                 batch_size: int = 128,\n",
        "                 learning_rate: float = 1e-3):\n",
        "\n",
        "        super().__init__()\n",
        "        #Parameters\n",
        "        self.module_name = module_name\n",
        "        self.input_features = input_features\n",
        "        self.hidden_dim = hidden_dim\n",
        "        self.num_layers = num_layers\n",
        "        self.epochs = epochs\n",
        "        self.batch_size = batch_size\n",
        "        self.learning_rate = learning_rate\n",
        "\n",
        "        #Networks\n",
        "        self.embedder = Embedder(module_name, input_features, hidden_dim, num_layers)\n",
        "        self.recovery = Recovery(module_name, input_features, hidden_dim, num_layers)\n",
        "        self.generator = Generator(module_name, input_features, hidden_dim, num_layers)\n",
        "        self.supervisor = Supervisor(module_name, hidden_dim, num_layers)\n",
        "        self.discriminator = Discriminator(module_name, hidden_dim, num_layers)\n",
        "\n",
        "        #Optimizers\n",
        "        self.optimizer_e = torch.optim.Adam(chain(self.embedder.parameters(), self.recovery.parameters()), lr=learning_rate)\n",
        "        self.optimizer_g = torch.optim.Adam(chain(self.generator.parameters(), self.supervisor.parameters()), lr=learning_rate)\n",
        "        self.optimizer_d = torch.optim.Adam(self.discriminator.parameters(), lr=learning_rate)\n",
        "\n",
        "        #Loss functions\n",
        "        self.embedder_loss = embedder_loss\n",
        "        self.supervised_loss = supervised_loss\n",
        "        self.generator_loss = generator_loss\n",
        "        self.discriminator_loss = discriminator_loss\n",
        "\n",
        "        #Auxiliary\n",
        "        self.fitting_time = None\n",
        "        self.losses = []\n",
        "\n",
        "    def fit(self, data_train: np.ndarray):\n",
        "        \"\"\"\n",
        "        Train TimeGAN model in three subsequent training phases\n",
        "        \"\"\"\n",
        "        self.fitting_time = time.time()\n",
        "        data_train = tensor(data_train, dtype=torch.float32, device=device)\n",
        "\n",
        "        #1. Embedding network training\n",
        "        print('Start Embedding Network Training')\n",
        "        for epoch, frame in zip(range(self.epochs), cycle(r'-\\|/-\\|/')):\n",
        "            batches_train = DataLoader(data_train, batch_size=self.batch_size, shuffle=True)\n",
        "\n",
        "            self.train()\n",
        "            loss_e = []\n",
        "            for x in batches_train:\n",
        "                self.optimizer_e.zero_grad()\n",
        "\n",
        "                h = self.embedder(x)\n",
        "                x_tilde = self.recovery(h)\n",
        "                e_loss = self.embedder_loss(x, x_tilde)\n",
        "\n",
        "                e_loss.backward()\n",
        "                self.optimizer_e.step()\n",
        "\n",
        "                loss_e.append(e_loss.item())\n",
        "\n",
        "            if (epoch + 1) % (0.1*self.epochs) == 0:\n",
        "                print('\\rEpoch', repr(epoch + 1).rjust(len(str(self.epochs))), 'of', self.epochs, '| loss_e', f'{np.mean(loss_e):12.9f}')\n",
        "            else:\n",
        "                print('\\r', frame, sep='', end='', flush=True)\n",
        "\n",
        "        print('Finished Embedding Network Training\\n')\n",
        "\n",
        "        #2. Training using only supervised loss\n",
        "        print('Start Training with Supervised Loss Only')\n",
        "        for epoch, frame in zip(range(self.epochs), cycle(r'-\\|/-\\|/')):\n",
        "            batches_train = DataLoader(data_train, batch_size=self.batch_size, shuffle=True)\n",
        "\n",
        "            self.train()\n",
        "            loss_g = []\n",
        "            for x in batches_train:\n",
        "                self.optimizer_g.zero_grad()\n",
        "\n",
        "                h = self.embedder(x)\n",
        "                h_hat_supervise = self.supervisor(h)\n",
        "\n",
        "                g_loss = self.supervised_loss(h, h_hat_supervise)\n",
        "\n",
        "                g_loss.backward()\n",
        "                self.optimizer_g.step()\n",
        "\n",
        "                loss_g.append(g_loss.item())\n",
        "\n",
        "            if (epoch + 1) % (0.1*self.epochs) == 0:\n",
        "                print('\\rEpoch', repr(epoch + 1).rjust(len(str(self.epochs))), 'of', self.epochs, '| loss_g', f'{np.mean(loss_g):12.9f}')\n",
        "            else:\n",
        "                print('\\r', frame, sep='', end='', flush=True)\n",
        "\n",
        "        print('Finished Training with Supervised Loss Only\\n')\n",
        "\n",
        "        #3. Joint training\n",
        "        print('Start Joint Training')\n",
        "        for epoch, frame in zip(range(self.epochs), cycle(r'-\\|/-\\|/')):\n",
        "            loss_g = []\n",
        "            loss_e = []\n",
        "            #Traing generator twice more than discriminator\n",
        "            for kk in range(2):\n",
        "                dataset = TensorDataset(data_train, torch.rand(data_train.shape, dtype=torch.float32, device=device))\n",
        "\n",
        "                batches_train = DataLoader(dataset, batch_size=self.batch_size, shuffle=True)\n",
        "\n",
        "                self.train()\n",
        "                for x, z in batches_train:\n",
        "                    self.optimizer_g.zero_grad()\n",
        "\n",
        "                    h = self.embedder(x)\n",
        "                    e_hat = self.generator(z)\n",
        "                    h_hat = self.supervisor(e_hat)\n",
        "                    h_hat_supervise = self.supervisor(h)\n",
        "                    x_hat = self.recovery(h_hat)\n",
        "                    y_fake = self.discriminator(h_hat)\n",
        "                    y_fake_e = self.discriminator(e_hat)\n",
        "\n",
        "                    g_loss = self.generator_loss(y_fake, y_fake_e, h, h_hat_supervise, x, x_hat)\n",
        "\n",
        "                    g_loss.backward()\n",
        "                    self.optimizer_g.step()\n",
        "\n",
        "                    loss_g.append(g_loss.item())\n",
        "\n",
        "                    self.optimizer_e.zero_grad()\n",
        "\n",
        "                    h = self.embedder(x)\n",
        "                    h_hat_supervise = self.supervisor(h)\n",
        "                    x_tilde = self.recovery(h)\n",
        "\n",
        "                    e_loss = self.embedder_loss(x, x_tilde) + 0.1*self.supervised_loss(h, h_hat_supervise)\n",
        "\n",
        "                    e_loss.backward()\n",
        "                    self.optimizer_e.step()\n",
        "\n",
        "                    loss_e.append(e_loss.item())\n",
        "\n",
        "            dataset = TensorDataset(data_train, torch.rand(data_train.shape, dtype=torch.float32, device=device))\n",
        "\n",
        "            batches_train = DataLoader(dataset, batch_size=self.batch_size, shuffle=True)\n",
        "\n",
        "            self.train()\n",
        "            loss_d = []\n",
        "            for x, z in batches_train:\n",
        "                self.optimizer_d.zero_grad()\n",
        "\n",
        "                h = self.embedder(x)\n",
        "                e_hat = self.generator(z)\n",
        "                h_hat = self.supervisor(e_hat)\n",
        "\n",
        "                y_fake = self.discriminator(h_hat)\n",
        "                y_real = self.discriminator(h)\n",
        "                y_fake_e = self.discriminator(e_hat)\n",
        "\n",
        "                d_loss = self.discriminator_loss(y_real, y_fake, y_fake_e)\n",
        "\n",
        "                loss_d.append(d_loss.item())\n",
        "\n",
        "                if d_loss > 0.15:\n",
        "                    d_loss.backward()\n",
        "                    self.optimizer_d.step()\n",
        "\n",
        "            self.losses.append([np.mean(loss_g), np.mean(loss_e), np.mean(loss_d)])\n",
        "\n",
        "            if (epoch + 1) % (0.1*self.epochs) == 0:\n",
        "                print('\\rEpoch', repr(epoch + 1).rjust(len(str(self.epochs))), 'of', self.epochs,\n",
        "                      '| loss_g', f'{np.mean(loss_g):12.9f}',\n",
        "                      '| loss_e', f'{np.mean(loss_e):12.9f}',\n",
        "                      '| loss_d', f'{np.mean(loss_d):12.9f}')\n",
        "            else:\n",
        "                print('\\r', frame, sep='', end='', flush=True)\n",
        "\n",
        "        self.fitting_time = np.round(time.time() - self.fitting_time, 3)\n",
        "        print('Finished Joint Training\\n')\n",
        "        print('\\nElapsed Training Time: ' + time.strftime('%Hh %Mmin %Ss', time.gmtime(self.fitting_time)))\n",
        "\n",
        "    def transform(self, data_shape: tuple):\n",
        "        \"\"\"\n",
        "        Generate data using trained TimeGAN model\n",
        "        \"\"\"\n",
        "        batches_z = DataLoader(torch.rand(size=data_shape, dtype=torch.float32, device=device, requires_grad=False),\n",
        "                               batch_size=1)\n",
        "\n",
        "        generated_data = []\n",
        "        self.eval()\n",
        "        with torch.no_grad():\n",
        "            for z in batches_z:\n",
        "                e_hat = self.generator(z)\n",
        "                h_hat = self.supervisor(e_hat)\n",
        "                x_hat = self.recovery(h_hat)\n",
        "\n",
        "                generated_data.append(np.squeeze(x_hat.cpu().numpy(), axis=0))\n",
        "\n",
        "        return np.stack(generated_data)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DcWb_7_GTMbi"
      },
      "source": [
        "### **Instantiate TimeGAN Model**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "metadata": {
        "id": "2P2ZjdlUfzKc"
      },
      "outputs": [],
      "source": [
        "#Create TimeGAN model instance with specified parameters\n",
        "model = TimeGAN(input_features=data_train.shape[-1],\n",
        "                hidden_dim=24,\n",
        "                num_layers=3,\n",
        "                epochs=750,\n",
        "                batch_size=128,\n",
        "                learning_rate=1e-3).to(device)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Train TimeGAN Model**"
      ],
      "metadata": {
        "id": "DL798PQh5Xow"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 49,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W4i4ON7mf2zr",
        "outputId": "135424b5-5d39-462c-a083-c45f304fc875"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Start Embedding Network Training\n",
            "Epoch  75 of 750 | loss_e  0.170025681\n",
            "Epoch 150 of 750 | loss_e  0.097332916\n",
            "Epoch 225 of 750 | loss_e  0.056687797\n",
            "Epoch 300 of 750 | loss_e  0.034836099\n",
            "Epoch 375 of 750 | loss_e  0.049136273\n",
            "Epoch 450 of 750 | loss_e  0.029709523\n",
            "Epoch 525 of 750 | loss_e  0.027200476\n",
            "Epoch 600 of 750 | loss_e  0.026442461\n",
            "Epoch 675 of 750 | loss_e  0.027090379\n",
            "Epoch 750 of 750 | loss_e  0.027503332\n",
            "Finished Embedding Network Training\n",
            "\n",
            "Start Training with Supervised Loss Only\n",
            "Epoch  75 of 750 | loss_g  0.000507227\n",
            "Epoch 150 of 750 | loss_g  0.000462240\n",
            "Epoch 225 of 750 | loss_g  0.000447201\n",
            "Epoch 300 of 750 | loss_g  0.000437391\n",
            "Epoch 375 of 750 | loss_g  0.000422767\n",
            "Epoch 450 of 750 | loss_g  0.000402894\n",
            "Epoch 525 of 750 | loss_g  0.000386060\n",
            "Epoch 600 of 750 | loss_g  0.000373236\n",
            "Epoch 675 of 750 | loss_g  0.000362588\n",
            "Epoch 750 of 750 | loss_g  0.000353083\n",
            "Finished Training with Supervised Loss Only\n",
            "\n",
            "Start Joint Training\n",
            "Epoch  75 of 750 | loss_g  9.253394875 | loss_e  0.025150909 | loss_d  1.608011221\n",
            "Epoch 150 of 750 | loss_g  9.151359583 | loss_e  0.023238991 | loss_d  1.777325034\n",
            "Epoch 225 of 750 | loss_g  8.818188289 | loss_e  0.023283790 | loss_d  1.764497387\n",
            "Epoch 300 of 750 | loss_g  8.831943553 | loss_e  0.022563656 | loss_d  1.740718348\n",
            "Epoch 375 of 750 | loss_g 10.212156115 | loss_e  0.022349352 | loss_d  1.705706995\n",
            "Epoch 450 of 750 | loss_g  9.159603119 | loss_e  0.022897100 | loss_d  1.599260281\n",
            "Epoch 525 of 750 | loss_g  9.214948736 | loss_e  0.021646166 | loss_d  1.787576203\n",
            "Epoch 600 of 750 | loss_g 10.482996192 | loss_e  0.021014162 | loss_d  1.606250808\n",
            "Epoch 675 of 750 | loss_g  9.336295210 | loss_e  0.020520726 | loss_d  1.688432665\n",
            "Epoch 750 of 750 | loss_g  9.793820973 | loss_e  0.020620145 | loss_d  1.903221702\n",
            "Finished Joint Training\n",
            "\n",
            "\n",
            "Elapsed Training Time: 00h 23min 30s\n"
          ]
        }
      ],
      "source": [
        "#Start training\n",
        "model.fit(data_train)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0e_mOPVFwE-L"
      },
      "source": [
        "### **Generate Data**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 50,
      "metadata": {
        "id": "Bs70xucGxzKC"
      },
      "outputs": [],
      "source": [
        "#Synthesize sequences\n",
        "data_gen = model.transform(data_train.shape)\n",
        "\n",
        "#Rescale generated sequences from range (0,1) back to original feature range\n",
        "# data_gen = data_gen*(max_val - min_val) + min_val\n",
        "\n",
        "#Rescale training data\n",
        "# data_ori = data_train*(max_val - min_val) + min_val"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Evaluation**"
      ],
      "metadata": {
        "id": "XFIsxmQm5gkn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### **1. Discriminative Score**"
      ],
      "metadata": {
        "id": "oXrC_nsSG3Se"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Define post-hoc discriminator\n",
        "class discriminator(nn.Module):\n",
        "    def __init__(self, input_features, hidden_dim, epochs, batch_size, device):\n",
        "        super().__init__()\n",
        "        #Parameters\n",
        "        self.input_features = input_features\n",
        "        self.hidden_dim = hidden_dim\n",
        "        self.num_layers = 1\n",
        "        self.epochs = epochs\n",
        "        self.batch_size = batch_size\n",
        "        self.device = device\n",
        "\n",
        "        #Layers\n",
        "        self.rnn = nn.GRU(input_size=input_features, hidden_size=hidden_dim, num_layers=self.num_layers, batch_first=True)\n",
        "        self.model = nn.Linear(hidden_dim, 1)\n",
        "        self.activation = nn.Sigmoid()\n",
        "\n",
        "        #Optimizer\n",
        "        self.optimizer = torch.optim.Adam(chain(self.rnn.parameters(), self.model.parameters()))\n",
        "\n",
        "        #Loss function\n",
        "        self.loss_fn = nn.BCEWithLogitsLoss()\n",
        "\n",
        "    def forward(self, x):\n",
        "        _, d_last_states = self.rnn(x)\n",
        "        y_hat_logit = self.model(torch.swapaxes(d_last_states, 0, 1))\n",
        "        y_hat = self.activation(y_hat_logit)\n",
        "        return y_hat_logit, y_hat\n",
        "\n",
        "    def fit(self, x, x_hat):\n",
        "        \"\"\"\n",
        "        Train model on real and synthetic data and test on both to evaluate classification accuracy\n",
        "        \"\"\"\n",
        "        #Split data into train and test fractions\n",
        "        x_train, x_test, x_hat_train, x_hat_test = train_test_split(x, x_hat, test_size=0.2)\n",
        "\n",
        "        x_train, x_hat_train = tensor(x_train, dtype=torch.float32, device=self.device), tensor(x_hat_train, dtype=torch.float32, device=self.device)\n",
        "        dataset_train = TensorDataset(x_train, x_hat_train)\n",
        "\n",
        "        x_test, x_hat_test = tensor(x_test, dtype=torch.float32, device=self.device, requires_grad=False), tensor(x_hat_test, dtype=torch.float32, device=self.device, requires_grad=False)\n",
        "\n",
        "        for itt in tqdm(range(self.epochs)):\n",
        "            batches = DataLoader(dataset_train, batch_size=self.batch_size, shuffle=True)\n",
        "\n",
        "            self.train()\n",
        "            for x, x_hat in batches:\n",
        "                self.optimizer.zero_grad()\n",
        "\n",
        "                y_logit_real, _ = self.forward(x)\n",
        "                y_logit_fake, _ = self.forward(x_hat)\n",
        "\n",
        "                d_loss_real = torch.mean(self.loss_fn(y_logit_real,\n",
        "                                                      torch.ones_like(y_logit_real, dtype=torch.float32, device=self.device, requires_grad=False)))\n",
        "                d_loss_fake = torch.mean(self.loss_fn(y_logit_fake,\n",
        "                                                      torch.zeros_like(y_logit_fake, dtype=torch.float32, device=self.device, requires_grad=False)))\n",
        "\n",
        "                d_loss = d_loss_real + d_loss_fake\n",
        "\n",
        "                d_loss.backward()\n",
        "                self.optimizer.step()\n",
        "\n",
        "        self.eval()\n",
        "        with torch.no_grad():\n",
        "            _, y_pred_real = self.forward(x_test)\n",
        "            _, y_pred_fake = self.forward(x_hat_test)\n",
        "\n",
        "            y_pred_final = np.squeeze(np.concatenate((y_pred_real.cpu().detach().numpy(), y_pred_fake.cpu().detach().numpy()), axis=0))\n",
        "            y_label_final = np.concatenate((np.ones([len(y_pred_real,)]), np.zeros([len(y_pred_fake,)])), axis=0)\n",
        "\n",
        "            acc = accuracy_score(y_label_final, (y_pred_final > 0.5))\n",
        "            discriminative_score = np.abs(0.5 - acc)\n",
        "\n",
        "        return discriminative_score"
      ],
      "metadata": {
        "id": "Z6-xzunwICl2"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Define function for computing discriminative score\n",
        "def discriminative_score_metrics(ori_data: np.ndarray, generated_data: np.ndarray, device: str):\n",
        "    \"\"\"\n",
        "    Use post-hoc RNN to classify original data and synthetic data\n",
        "\n",
        "    Args:\n",
        "        - ori_data: original data\n",
        "        - generated_data: generated synthetic data\n",
        "\n",
        "    Returns:\n",
        "        - discriminative_score: np.abs(classification accuracy - 0.5)\n",
        "    \"\"\"\n",
        "    no, seq_len, dim = ori_data.shape\n",
        "\n",
        "    hidden_dim = int(dim/2)\n",
        "    iterations = 2000\n",
        "    batch_size = 128\n",
        "\n",
        "    model = discriminator(input_features=dim, hidden_dim=hidden_dim, epochs=iterations, batch_size=batch_size, device=device).to(device)\n",
        "\n",
        "    discriminative_score = model.fit(ori_data, generated_data)\n",
        "\n",
        "    return discriminative_score"
      ],
      "metadata": {
        "id": "rbS_ksTfIDRs"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Compute discriminative score\n",
        "discriminative_score_metrics(data_train, data_gen, device)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CMNgCFtoHeCI",
        "outputId": "d629b960-61a2-4f22-c97e-eace67dface0"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 2000/2000 [03:38<00:00,  9.17it/s]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.155525238744884"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### **2. Predictive Score**"
      ],
      "metadata": {
        "id": "8NbpZJsFG730"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Define post-hoc predictor\n",
        "class predictor(nn.Module):\n",
        "    def __init__(self, dim, hidden_dim, epochs, batch_size, device):\n",
        "        super().__init__()\n",
        "        #Parameters\n",
        "        self.dim = dim\n",
        "        self.hidden_dim = hidden_dim\n",
        "        self.num_layers = 1\n",
        "        self.epochs = epochs\n",
        "        self.batch_size = batch_size\n",
        "        self.device = device\n",
        "\n",
        "        #Layers\n",
        "        self.rnn = nn.GRU(input_size=dim-1, hidden_size=hidden_dim, num_layers=self.num_layers, batch_first=True)\n",
        "        self.model = nn.Sequential(nn.Linear(self.hidden_dim, 1), nn.Sigmoid())\n",
        "\n",
        "        #Optimizer\n",
        "        self.optimizer = torch.optim.Adam(chain(self.rnn.parameters(), self.model.parameters()), 1e-3)\n",
        "\n",
        "        #Loss function\n",
        "        self.loss_fn = nn.L1Loss()\n",
        "\n",
        "    def forward(self, x):\n",
        "        p_outputs, _ = self.rnn(x)\n",
        "        return self.model(p_outputs)\n",
        "\n",
        "    def fit(self, data_train, data_test):\n",
        "        \"\"\"\n",
        "        Train model on synthetic and test on real data\n",
        "        \"\"\"\n",
        "        x_train = data_train[:,:-1,:(self.dim-1)]\n",
        "        y_train = np.reshape(data_train[:,1:,(self.dim-1)], (data_train.shape[0], data_train.shape[1]-1, 1))\n",
        "\n",
        "        x_train, y_train = tensor(x_train, dtype=torch.float32, device=self.device), tensor(y_train, dtype=torch.float32, device=self.device)\n",
        "\n",
        "        dataset = TensorDataset(x_train, y_train)\n",
        "        for itt in tqdm(range(self.epochs)):\n",
        "            batches_train = DataLoader(dataset, batch_size=self.batch_size, shuffle=True)\n",
        "\n",
        "            self.train()\n",
        "            #Mini batches, train on synthetic\n",
        "            for X, Y in batches_train:\n",
        "                self.optimizer.zero_grad()\n",
        "\n",
        "                pred_train = self.forward(X)\n",
        "                loss = self.loss_fn(pred_train, Y)\n",
        "\n",
        "                loss.backward()\n",
        "                self.optimizer.step()\n",
        "\n",
        "        #Test model on original data\n",
        "        x_test = data_test[:,:-1,:(self.dim-1)]\n",
        "        y_test = np.reshape(data_test[:,1:,(self.dim-1)], (data_test.shape[0], data_test.shape[1]-1, 1))\n",
        "\n",
        "        x_test, y_test = tensor(x_test, dtype=torch.float32, device=self.device, requires_grad=False), tensor(y_test, dtype=torch.float32, device=self.device, requires_grad=False)\n",
        "\n",
        "        MAE = 0\n",
        "        self.eval()\n",
        "        with torch.no_grad():\n",
        "            pred_test =  self.forward(x_test)\n",
        "\n",
        "            for i in range(len(pred_test)):\n",
        "                MAE =+ mean_absolute_error(y_test[i,:,:].cpu().detach().numpy(), pred_test[i,:,:].cpu().detach().numpy())\n",
        "\n",
        "        return MAE"
      ],
      "metadata": {
        "id": "_bGJNjM7Hmc9"
      },
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Define function for computing predictive score\n",
        "def predictive_score_metrics(ori_data: np.ndarray, generated_data: np.ndarray, device: str):\n",
        "    \"\"\"\n",
        "    Report the performance of Post-hoc RNN one-step ahead prediction\n",
        "\n",
        "    Args:\n",
        "        - ori_data: original data\n",
        "        - generated_data: generated synthetic data\n",
        "\n",
        "    Returns:\n",
        "        - predictive_score: MAE of the predictions on the original data\n",
        "    \"\"\"\n",
        "    no, seq_len, dim = np.asarray(ori_data).shape\n",
        "\n",
        "    hidden_dim = int(dim/2)\n",
        "    iterations = 5000\n",
        "    batch_size = 128\n",
        "\n",
        "    model = predictor(dim, hidden_dim, iterations, batch_size, device=device).to(device)\n",
        "\n",
        "    predictive_score = model.fit(generated_data, ori_data)\n",
        "\n",
        "    return predictive_score/no"
      ],
      "metadata": {
        "id": "_SaBS57IHpm6"
      },
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Compute predictive score\n",
        "predictive_score_metrics(data_train, data_gen, device)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n4tdHYqiG__J",
        "outputId": "501b2d5b-f7b7-4d97-ae74-bcda9960c3f9"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 5000/5000 [07:48<00:00, 10.67it/s]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "7.276645358127833e-06"
            ]
          },
          "metadata": {},
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##### **3. Visualization**"
      ],
      "metadata": {
        "id": "FNNOFrM8GzPy"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 58,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "R3QF8l6rTMbk",
        "outputId": "7504749a-3d7d-4090-9c5d-9e4eb9ac1bb3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[t-SNE] Computing 121 nearest neighbors...\n",
            "[t-SNE] Indexed 2000 samples in 0.000s...\n",
            "[t-SNE] Computed neighbors for 2000 samples in 0.094s...\n",
            "[t-SNE] Computed conditional probabilities for sample 1000 / 2000\n",
            "[t-SNE] Computed conditional probabilities for sample 2000 / 2000\n",
            "[t-SNE] Mean sigma: 0.032452\n",
            "[t-SNE] KL divergence after 250 iterations with early exaggeration: 53.624294\n",
            "[t-SNE] KL divergence after 300 iterations: 1.042779\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "#Plot real and generated distributions\n",
        "plot_distribution_estimate(*low_dimensional_representation(data_train, data_gen, 'pca'), 'pca')\n",
        "plot_distribution_estimate(*low_dimensional_representation(data_train, data_gen, 'tsne'), 'tsne')"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.10.15"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}