semi-supervised.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "fbc981d7",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Missing Part of the Graph Structure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "deb6cf91",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cuda\n",
      "Epoch [10/200], Loss: 0.0967\n",
      "Epoch [20/200], Loss: 0.0639\n",
      "Epoch [30/200], Loss: 0.0676\n",
      "Epoch [40/200], Loss: 0.0519\n",
      "Epoch [50/200], Loss: 0.0534\n",
      "Epoch [60/200], Loss: 0.0439\n",
      "Epoch [70/200], Loss: 0.0441\n",
      "Epoch [80/200], Loss: 0.0476\n",
      "Epoch [90/200], Loss: 0.0415\n",
      "Epoch [100/200], Loss: 0.0460\n",
      "Epoch [110/200], Loss: 0.0333\n",
      "Epoch [120/200], Loss: 0.0387\n",
      "Epoch [130/200], Loss: 0.0512\n",
      "Epoch [140/200], Loss: 0.0421\n",
      "Epoch [150/200], Loss: 0.0380\n",
      "Epoch [160/200], Loss: 0.0417\n",
      "Epoch [170/200], Loss: 0.0417\n",
      "Epoch [180/200], Loss: 0.0463\n",
      "Epoch [190/200], Loss: 0.0365\n",
      "Epoch [200/200], Loss: 0.0383\n",
      "Epoch [10/500], Joint Loss: 1.3568, Feature Reconstruction Loss: 1.8327, Neighborhood Reconstruction Loss: 0.8809\n",
      "Epoch [20/500], Joint Loss: 0.8752, Feature Reconstruction Loss: 1.0364, Neighborhood Reconstruction Loss: 0.7140\n",
      "Epoch [30/500], Joint Loss: 0.6886, Feature Reconstruction Loss: 0.7505, Neighborhood Reconstruction Loss: 0.6267\n",
      "Epoch [40/500], Joint Loss: 0.5652, Feature Reconstruction Loss: 0.5853, Neighborhood Reconstruction Loss: 0.5452\n",
      "Epoch [50/500], Joint Loss: 0.4736, Feature Reconstruction Loss: 0.4915, Neighborhood Reconstruction Loss: 0.4556\n",
      "Epoch [60/500], Joint Loss: 0.4323, Feature Reconstruction Loss: 0.4357, Neighborhood Reconstruction Loss: 0.4289\n",
      "Epoch [70/500], Joint Loss: 0.3879, Feature Reconstruction Loss: 0.4013, Neighborhood Reconstruction Loss: 0.3745\n",
      "Epoch [80/500], Joint Loss: 0.3648, Feature Reconstruction Loss: 0.3746, Neighborhood Reconstruction Loss: 0.3550\n",
      "Epoch [90/500], Joint Loss: 0.3436, Feature Reconstruction Loss: 0.3559, Neighborhood Reconstruction Loss: 0.3314\n",
      "Epoch [100/500], Joint Loss: 0.3396, Feature Reconstruction Loss: 0.3404, Neighborhood Reconstruction Loss: 0.3389\n",
      "Epoch [110/500], Joint Loss: 0.3192, Feature Reconstruction Loss: 0.3336, Neighborhood Reconstruction Loss: 0.3048\n",
      "Epoch [120/500], Joint Loss: 0.3015, Feature Reconstruction Loss: 0.3122, Neighborhood Reconstruction Loss: 0.2908\n",
      "Epoch [130/500], Joint Loss: 0.3023, Feature Reconstruction Loss: 0.3037, Neighborhood Reconstruction Loss: 0.3009\n",
      "Epoch [140/500], Joint Loss: 0.2944, Feature Reconstruction Loss: 0.2977, Neighborhood Reconstruction Loss: 0.2911\n",
      "Epoch [150/500], Joint Loss: 0.2865, Feature Reconstruction Loss: 0.2882, Neighborhood Reconstruction Loss: 0.2847\n",
      "Epoch [160/500], Joint Loss: 0.2759, Feature Reconstruction Loss: 0.2816, Neighborhood Reconstruction Loss: 0.2701\n",
      "Epoch [170/500], Joint Loss: 0.2764, Feature Reconstruction Loss: 0.2744, Neighborhood Reconstruction Loss: 0.2785\n",
      "Epoch [180/500], Joint Loss: 0.2670, Feature Reconstruction Loss: 0.2706, Neighborhood Reconstruction Loss: 0.2633\n",
      "Epoch [190/500], Joint Loss: 0.2786, Feature Reconstruction Loss: 0.2635, Neighborhood Reconstruction Loss: 0.2938\n",
      "Epoch [200/500], Joint Loss: 0.2773, Feature Reconstruction Loss: 0.2622, Neighborhood Reconstruction Loss: 0.2924\n",
      "Epoch [210/500], Joint Loss: 0.2600, Feature Reconstruction Loss: 0.2480, Neighborhood Reconstruction Loss: 0.2721\n",
      "Epoch [220/500], Joint Loss: 0.2607, Feature Reconstruction Loss: 0.2482, Neighborhood Reconstruction Loss: 0.2733\n",
      "Epoch [230/500], Joint Loss: 0.2927, Feature Reconstruction Loss: 0.2436, Neighborhood Reconstruction Loss: 0.3418\n",
      "Epoch [240/500], Joint Loss: 0.2480, Feature Reconstruction Loss: 0.2308, Neighborhood Reconstruction Loss: 0.2652\n",
      "Epoch [250/500], Joint Loss: 0.2431, Feature Reconstruction Loss: 0.2280, Neighborhood Reconstruction Loss: 0.2581\n",
      "Epoch [260/500], Joint Loss: 0.2615, Feature Reconstruction Loss: 0.2298, Neighborhood Reconstruction Loss: 0.2933\n",
      "Epoch [270/500], Joint Loss: 0.2428, Feature Reconstruction Loss: 0.2153, Neighborhood Reconstruction Loss: 0.2704\n",
      "Epoch [280/500], Joint Loss: 0.2318, Feature Reconstruction Loss: 0.2160, Neighborhood Reconstruction Loss: 0.2475\n",
      "Epoch [290/500], Joint Loss: 0.2421, Feature Reconstruction Loss: 0.2185, Neighborhood Reconstruction Loss: 0.2657\n",
      "Epoch [300/500], Joint Loss: 0.2772, Feature Reconstruction Loss: 0.2107, Neighborhood Reconstruction Loss: 0.3436\n",
      "Epoch [310/500], Joint Loss: 0.2394, Feature Reconstruction Loss: 0.2065, Neighborhood Reconstruction Loss: 0.2724\n",
      "Epoch [320/500], Joint Loss: 0.2344, Feature Reconstruction Loss: 0.2034, Neighborhood Reconstruction Loss: 0.2654\n",
      "Epoch [330/500], Joint Loss: 0.2264, Feature Reconstruction Loss: 0.2098, Neighborhood Reconstruction Loss: 0.2430\n",
      "Epoch [340/500], Joint Loss: 0.2338, Feature Reconstruction Loss: 0.2011, Neighborhood Reconstruction Loss: 0.2666\n",
      "Epoch [350/500], Joint Loss: 0.2199, Feature Reconstruction Loss: 0.1974, Neighborhood Reconstruction Loss: 0.2424\n",
      "Epoch [360/500], Joint Loss: 0.2221, Feature Reconstruction Loss: 0.1925, Neighborhood Reconstruction Loss: 0.2518\n",
      "Epoch [370/500], Joint Loss: 0.2146, Feature Reconstruction Loss: 0.1856, Neighborhood Reconstruction Loss: 0.2437\n",
      "Epoch [380/500], Joint Loss: 0.2290, Feature Reconstruction Loss: 0.1924, Neighborhood Reconstruction Loss: 0.2656\n",
      "Epoch [390/500], Joint Loss: 0.2171, Feature Reconstruction Loss: 0.1818, Neighborhood Reconstruction Loss: 0.2523\n",
      "Epoch [400/500], Joint Loss: 0.2165, Feature Reconstruction Loss: 0.1863, Neighborhood Reconstruction Loss: 0.2467\n",
      "Epoch [410/500], Joint Loss: 0.2240, Feature Reconstruction Loss: 0.1822, Neighborhood Reconstruction Loss: 0.2657\n",
      "Epoch [420/500], Joint Loss: 0.2149, Feature Reconstruction Loss: 0.1837, Neighborhood Reconstruction Loss: 0.2461\n",
      "Epoch [430/500], Joint Loss: 0.2284, Feature Reconstruction Loss: 0.1791, Neighborhood Reconstruction Loss: 0.2777\n",
      "Epoch [440/500], Joint Loss: 0.2143, Feature Reconstruction Loss: 0.1852, Neighborhood Reconstruction Loss: 0.2434\n",
      "Epoch [450/500], Joint Loss: 0.2130, Feature Reconstruction Loss: 0.1777, Neighborhood Reconstruction Loss: 0.2482\n",
      "Epoch [460/500], Joint Loss: 0.2116, Feature Reconstruction Loss: 0.1767, Neighborhood Reconstruction Loss: 0.2465\n",
      "Epoch [470/500], Joint Loss: 0.2036, Feature Reconstruction Loss: 0.1710, Neighborhood Reconstruction Loss: 0.2362\n",
      "Epoch [480/500], Joint Loss: 0.2209, Feature Reconstruction Loss: 0.1774, Neighborhood Reconstruction Loss: 0.2645\n",
      "Epoch [490/500], Joint Loss: 0.2099, Feature Reconstruction Loss: 0.1712, Neighborhood Reconstruction Loss: 0.2486\n",
      "Epoch [500/500], Joint Loss: 0.2182, Feature Reconstruction Loss: 0.1742, Neighborhood Reconstruction Loss: 0.2623\n",
      "Accuracy: 0.9756\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.92      0.96       784\n",
      "           1       0.97      1.00      0.98       650\n",
      "           2       0.99      0.99      0.99       576\n",
      "           3       0.96      1.00      0.98       625\n",
      "           4       0.95      1.00      0.97       585\n",
      "           5       0.98      0.96      0.97       711\n",
      "\n",
      "    accuracy                           0.98      3931\n",
      "   macro avg       0.98      0.98      0.98      3931\n",
      "weighted avg       0.98      0.98      0.98      3931\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler, label_binarize\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.metrics import classification_report, accuracy_score, roc_curve, auc, confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(f'Using device: {device}')\n",
    "\n",
    "data_dict = np.load('dataset.npy', allow_pickle=True).item()\n",
    "all_frames = data_dict['frames']\n",
    "all_labels = data_dict['labels']\n",
    "\n",
    "scaler = StandardScaler()\n",
    "num_samples, num_timesteps, num_features = all_frames.shape\n",
    "all_frames = all_frames.reshape(num_samples * num_timesteps, num_features)\n",
    "all_frames = scaler.fit_transform(all_frames)\n",
    "all_frames = all_frames.reshape(num_samples, num_timesteps, num_features)\n",
    "\n",
    "all_frames = torch.tensor(all_frames, dtype=torch.float32).to(device)\n",
    "all_labels = torch.tensor(all_labels, dtype=torch.long).to(device)\n",
    "\n",
    "class Autoencoder(nn.Module):\n",
    "    def __init__(self, input_dim, encoding_dim):\n",
    "        super(Autoencoder, self).__init__()\n",
    "        self.encoder = nn.Sequential(\n",
    "            nn.Linear(input_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, encoding_dim)\n",
    "        )\n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.Linear(encoding_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, input_dim)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.encoder(x)\n",
    "        x = self.decoder(x)\n",
    "        return x\n",
    "    \n",
    "    def encode(self, x):\n",
    "        return self.encoder(x)\n",
    "\n",
    "input_dim = num_timesteps * num_features\n",
    "encoding_dim = 64\n",
    "learning_rate = 0.001\n",
    "num_epochs_ae = 200\n",
    "batch_size = 32\n",
    "\n",
    "autoencoder = Autoencoder(input_dim=input_dim, encoding_dim=encoding_dim).to(device)\n",
    "criterion_ae = nn.MSELoss()\n",
    "optimizer_ae = optim.Adam(autoencoder.parameters(), lr=learning_rate)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(all_frames, batch_size=batch_size, shuffle=True)\n",
    "\n",
    "for epoch in range(num_epochs_ae):\n",
    "    autoencoder.train()\n",
    "    running_loss = 0.0\n",
    "    for data in train_loader:\n",
    "        data = data.view(data.size(0), -1)\n",
    "        optimizer_ae.zero_grad()\n",
    "        output = autoencoder(data)\n",
    "        loss = criterion_ae(output, data)\n",
    "        loss.backward()\n",
    "        optimizer_ae.step()\n",
    "        running_loss += loss.item() * data.size(0)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ae}], Loss: {running_loss / len(train_loader.dataset):.4f}')\n",
    "\n",
    "for param in autoencoder.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "all_features = autoencoder.encode(all_frames.view(all_frames.size(0), -1)).detach().cpu().numpy()\n",
    "\n",
    "class MLPEncoder(nn.Module):\n",
    "    def __init__(self, in_channels, hidden_channels, out_channels):\n",
    "        super(MLPEncoder, self).__init__()\n",
    "        self.fc1 = nn.Linear(in_channels, hidden_channels)\n",
    "        self.fc2 = nn.Linear(hidden_channels, out_channels)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = F.relu(self.fc1(x))\n",
    "        z = self.fc2(x)\n",
    "        return z\n",
    "\n",
    "class FeatureDecoder(nn.Module):\n",
    "    def __init__(self, in_channels, out_channels):\n",
    "        super(FeatureDecoder, self).__init__()\n",
    "        self.fc = nn.Linear(in_channels, out_channels)\n",
    "\n",
    "    def forward(self, z):\n",
    "        return self.fc(z)\n",
    "\n",
    "class InnerProductDecoder(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(InnerProductDecoder, self).__init__()\n",
    "\n",
    "    def forward(self, z):\n",
    "        adj_pred = torch.sigmoid(torch.matmul(z, z.t()))\n",
    "        return adj_pred\n",
    "\n",
    "class FeatureTransformer(nn.Module):\n",
    "    def __init__(self, encoder, feature_decoder, adj_decoder):\n",
    "        super(FeatureTransformer, self).__init__()\n",
    "        self.encoder = encoder\n",
    "        self.feature_decoder = feature_decoder\n",
    "        self.adj_decoder = adj_decoder\n",
    "\n",
    "    def forward(self, x):\n",
    "        z = self.encoder(x)\n",
    "        x_reconstructed = self.feature_decoder(z)\n",
    "        adj_pred = self.adj_decoder(z)\n",
    "        return z, x_reconstructed, adj_pred\n",
    "\n",
    "input_dim_ft = encoding_dim\n",
    "hidden_dim_ft = 64\n",
    "output_dim_ft = 64\n",
    "batch_size_ft = 64\n",
    "num_epochs_ft = 500\n",
    "\n",
    "mlp_encoder = MLPEncoder(input_dim_ft, hidden_dim_ft, output_dim_ft).to(device)\n",
    "feature_decoder = FeatureDecoder(output_dim_ft, input_dim_ft).to(device)\n",
    "adj_decoder = InnerProductDecoder().to(device)\n",
    "\n",
    "feature_transformer = FeatureTransformer(mlp_encoder, feature_decoder, adj_decoder).to(device)\n",
    "optimizer_ft = optim.Adam(feature_transformer.parameters(), lr=learning_rate)\n",
    "\n",
    "def joint_loss_function(x, x_reconstructed, adj, adj_pred, alpha=0.5):\n",
    "    feature_reconstruction_loss = F.mse_loss(x_reconstructed, x)\n",
    "    adj_reconstruction_loss = F.binary_cross_entropy(adj_pred, adj)\n",
    "    loss = alpha * feature_reconstruction_loss + (1 - alpha) * adj_reconstruction_loss\n",
    "    return loss, feature_reconstruction_loss, adj_reconstruction_loss\n",
    "\n",
    "def create_adj_matrix(labels):\n",
    "    num_samples = len(labels)\n",
    "    adj_matrix = np.zeros((num_samples, num_samples))\n",
    "    for i in range(num_samples):\n",
    "        for j in range(num_samples):\n",
    "            if labels[i] == labels[j]:\n",
    "                adj_matrix[i, j] = 1\n",
    "    return adj_matrix\n",
    "\n",
    "class CustomDataset(torch.utils.data.Dataset):\n",
    "    def __init__(self, features, labels):\n",
    "        self.features = features\n",
    "        self.labels = labels\n",
    "        self.indices = np.arange(len(features))\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.features)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        return self.features[idx], self.labels[idx], self.indices[idx]\n",
    "\n",
    "train_features, test_features, train_labels, test_labels = train_test_split(\n",
    "    all_features, all_labels.cpu().numpy(), test_size=0.05, random_state=42, stratify=all_labels.cpu().numpy()\n",
    ")\n",
    "\n",
    "train_features = torch.tensor(train_features, dtype=torch.float32).to(device)\n",
    "test_features = torch.tensor(test_features, dtype=torch.float32).to(device)\n",
    "train_labels = torch.tensor(train_labels, dtype=torch.long).to(device)\n",
    "test_labels = torch.tensor(test_labels, dtype=torch.long).to(device)\n",
    "\n",
    "train_adj = create_adj_matrix(train_labels.cpu().numpy())\n",
    "\n",
    "train_adj_np = train_adj.copy()\n",
    "num_edges = int(np.sum(train_adj_np) / 2)\n",
    "num_disrupt = int(0 * num_edges)\n",
    "\n",
    "edges = np.array(np.triu_indices(train_adj_np.shape[0], 1)).T\n",
    "np.random.shuffle(edges)\n",
    "\n",
    "disrupted_edges = edges[:num_disrupt]\n",
    "\n",
    "for edge in disrupted_edges:\n",
    "    train_adj_np[edge[0], edge[1]] = 0\n",
    "    train_adj_np[edge[1], edge[0]] = 0\n",
    "\n",
    "train_adj = torch.tensor(train_adj_np, dtype=torch.float32).to(device)\n",
    "\n",
    "train_dataset = CustomDataset(train_features, train_labels)\n",
    "train_loader_ft = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size_ft, shuffle=True)\n",
    "\n",
    "joint_losses = []\n",
    "feature_reconstruction_losses = []\n",
    "neighborhood_reconstruction_losses = []\n",
    "\n",
    "for epoch in range(num_epochs_ft):\n",
    "    feature_transformer.train()\n",
    "    running_joint_loss = 0.0\n",
    "    running_feature_reconstruction_loss = 0.0\n",
    "    running_neighborhood_reconstruction_loss = 0.0\n",
    "    for data, labels, indices in train_loader_ft:\n",
    "        optimizer_ft.zero_grad()\n",
    "        z, x_reconstructed, adj_pred = feature_transformer(data)\n",
    "        batch_adj = train_adj[indices][:, indices]\n",
    "        joint_loss, feature_reconstruction_loss, neighborhood_reconstruction_loss = joint_loss_function(data, x_reconstructed, batch_adj, adj_pred)\n",
    "        joint_loss.backward()\n",
    "        optimizer_ft.step()\n",
    "        running_joint_loss += joint_loss.item() * data.size(0)\n",
    "        running_feature_reconstruction_loss += feature_reconstruction_loss.item() * data.size(0)\n",
    "        running_neighborhood_reconstruction_loss += neighborhood_reconstruction_loss.item() * data.size(0)\n",
    "    \n",
    "    avg_joint_loss = running_joint_loss / len(train_loader_ft.dataset)\n",
    "    avg_feature_reconstruction_loss = running_feature_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    avg_neighborhood_reconstruction_loss = running_neighborhood_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    joint_losses.append(avg_joint_loss)\n",
    "    feature_reconstruction_losses.append(avg_feature_reconstruction_loss)\n",
    "    neighborhood_reconstruction_losses.append(avg_neighborhood_reconstruction_loss)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ft}], Joint Loss: {avg_joint_loss:.4f}, Feature Reconstruction Loss: {avg_feature_reconstruction_loss:.4f}, Neighborhood Reconstruction Loss: {avg_neighborhood_reconstruction_loss:.4f}')\n",
    "\n",
    "for param in feature_transformer.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "with torch.no_grad():\n",
    "    test_reconstructed_features, _, _ = feature_transformer(test_features)\n",
    "test_reconstructed_features = test_reconstructed_features.cpu().numpy()\n",
    "\n",
    "nb_classifier = GaussianNB()\n",
    "nb_classifier.fit(test_reconstructed_features, test_labels.cpu().numpy())\n",
    "\n",
    "with torch.no_grad():\n",
    "    train_reconstructed_features, _, _ = feature_transformer(train_features)\n",
    "train_reconstructed_features = train_reconstructed_features.cpu().numpy()\n",
    "\n",
    "train_pred_labels = nb_classifier.predict(train_reconstructed_features)\n",
    "\n",
    "accuracy = accuracy_score(train_labels.cpu().numpy(), train_pred_labels)\n",
    "print(f'Accuracy: {accuracy:.4f}')\n",
    "print(classification_report(train_labels.cpu().numpy(), train_pred_labels))\n",
    "\n",
    "train_labels_binarized = label_binarize(train_labels.cpu().numpy(), classes=np.unique(train_labels.cpu().numpy()))\n",
    "n_classes = train_labels_binarized.shape[1]\n",
    "\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "\n",
    "for i in range(n_classes):\n",
    "    fpr[i], tpr[i], _ = roc_curve(train_labels_binarized[:, i], nb_classifier.predict_proba(train_reconstructed_features)[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "plt.figure()\n",
    "colors = ['aqua', 'darkorange', 'cornflowerblue', 'green', 'red', 'purple']\n",
    "for i, color in zip(range(n_classes), colors):\n",
    "    plt.plot(fpr[i], tpr[i], color=color, lw=2,\n",
    "             label=f'ROC curve of class {i} (area = {roc_auc[i]:.2f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', lw=2)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curves')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "conf_matrix = confusion_matrix(train_labels.cpu().numpy(), train_pred_labels)\n",
    "plt.figure(figsize=(10, 8))\n",
    "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()\n",
    "\n",
    "plt.figure()\n",
    "#plt.plot(range(1, num_epochs_ft + 1), joint_losses, label='Joint Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), feature_reconstruction_losses, label='Feature Reconstruction Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), neighborhood_reconstruction_losses, label='Neighborhood Reconstruction Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss Curves')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e43779cf",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "21f0344a",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cuda\n",
      "Epoch [10/200], Loss: 0.0937\n",
      "Epoch [20/200], Loss: 0.0597\n",
      "Epoch [30/200], Loss: 0.0644\n",
      "Epoch [40/200], Loss: 0.0543\n",
      "Epoch [50/200], Loss: 0.0496\n",
      "Epoch [60/200], Loss: 0.0453\n",
      "Epoch [70/200], Loss: 0.0610\n",
      "Epoch [80/200], Loss: 0.0534\n",
      "Epoch [90/200], Loss: 0.0482\n",
      "Epoch [100/200], Loss: 0.0373\n",
      "Epoch [110/200], Loss: 0.0357\n",
      "Epoch [120/200], Loss: 0.0371\n",
      "Epoch [130/200], Loss: 0.0395\n",
      "Epoch [140/200], Loss: 0.0386\n",
      "Epoch [150/200], Loss: 0.0397\n",
      "Epoch [160/200], Loss: 0.0421\n",
      "Epoch [170/200], Loss: 0.0350\n",
      "Epoch [180/200], Loss: 0.0380\n",
      "Epoch [190/200], Loss: 0.0379\n",
      "Epoch [200/200], Loss: 0.0384\n",
      "Epoch [10/500], Joint Loss: 1.7534, Feature Reconstruction Loss: 1.7506, Neighborhood Reconstruction Loss: 1.7562\n",
      "Epoch [20/500], Joint Loss: 1.2559, Feature Reconstruction Loss: 0.9940, Neighborhood Reconstruction Loss: 1.5178\n",
      "Epoch [30/500], Joint Loss: 1.0919, Feature Reconstruction Loss: 0.7040, Neighborhood Reconstruction Loss: 1.4798\n",
      "Epoch [40/500], Joint Loss: 0.9971, Feature Reconstruction Loss: 0.5349, Neighborhood Reconstruction Loss: 1.4594\n",
      "Epoch [50/500], Joint Loss: 0.9471, Feature Reconstruction Loss: 0.4469, Neighborhood Reconstruction Loss: 1.4473\n",
      "Epoch [60/500], Joint Loss: 0.9392, Feature Reconstruction Loss: 0.3851, Neighborhood Reconstruction Loss: 1.4933\n",
      "Epoch [70/500], Joint Loss: 0.8766, Feature Reconstruction Loss: 0.3424, Neighborhood Reconstruction Loss: 1.4107\n",
      "Epoch [80/500], Joint Loss: 0.8832, Feature Reconstruction Loss: 0.2997, Neighborhood Reconstruction Loss: 1.4667\n",
      "Epoch [90/500], Joint Loss: 0.8343, Feature Reconstruction Loss: 0.2716, Neighborhood Reconstruction Loss: 1.3969\n",
      "Epoch [100/500], Joint Loss: 0.8257, Feature Reconstruction Loss: 0.2575, Neighborhood Reconstruction Loss: 1.3940\n",
      "Epoch [110/500], Joint Loss: 0.7892, Feature Reconstruction Loss: 0.2342, Neighborhood Reconstruction Loss: 1.3442\n",
      "Epoch [120/500], Joint Loss: 0.8164, Feature Reconstruction Loss: 0.2221, Neighborhood Reconstruction Loss: 1.4107\n",
      "Epoch [130/500], Joint Loss: 0.7666, Feature Reconstruction Loss: 0.2124, Neighborhood Reconstruction Loss: 1.3208\n",
      "Epoch [140/500], Joint Loss: 0.7653, Feature Reconstruction Loss: 0.2119, Neighborhood Reconstruction Loss: 1.3188\n",
      "Epoch [150/500], Joint Loss: 0.7417, Feature Reconstruction Loss: 0.1952, Neighborhood Reconstruction Loss: 1.2882\n",
      "Epoch [160/500], Joint Loss: 0.7523, Feature Reconstruction Loss: 0.1918, Neighborhood Reconstruction Loss: 1.3127\n",
      "Epoch [170/500], Joint Loss: 0.7434, Feature Reconstruction Loss: 0.1858, Neighborhood Reconstruction Loss: 1.3009\n",
      "Epoch [180/500], Joint Loss: 0.7327, Feature Reconstruction Loss: 0.1840, Neighborhood Reconstruction Loss: 1.2813\n",
      "Epoch [190/500], Joint Loss: 0.7705, Feature Reconstruction Loss: 0.1813, Neighborhood Reconstruction Loss: 1.3596\n",
      "Epoch [200/500], Joint Loss: 0.7617, Feature Reconstruction Loss: 0.1774, Neighborhood Reconstruction Loss: 1.3461\n",
      "Epoch [210/500], Joint Loss: 0.7491, Feature Reconstruction Loss: 0.1752, Neighborhood Reconstruction Loss: 1.3231\n",
      "Epoch [220/500], Joint Loss: 0.7404, Feature Reconstruction Loss: 0.1750, Neighborhood Reconstruction Loss: 1.3059\n",
      "Epoch [230/500], Joint Loss: 0.7234, Feature Reconstruction Loss: 0.1754, Neighborhood Reconstruction Loss: 1.2713\n",
      "Epoch [240/500], Joint Loss: 0.7264, Feature Reconstruction Loss: 0.1731, Neighborhood Reconstruction Loss: 1.2797\n",
      "Epoch [250/500], Joint Loss: 0.7370, Feature Reconstruction Loss: 0.1682, Neighborhood Reconstruction Loss: 1.3058\n",
      "Epoch [260/500], Joint Loss: 0.7299, Feature Reconstruction Loss: 0.1662, Neighborhood Reconstruction Loss: 1.2936\n",
      "Epoch [270/500], Joint Loss: 0.7179, Feature Reconstruction Loss: 0.1642, Neighborhood Reconstruction Loss: 1.2717\n",
      "Epoch [280/500], Joint Loss: 0.7425, Feature Reconstruction Loss: 0.1671, Neighborhood Reconstruction Loss: 1.3179\n",
      "Epoch [290/500], Joint Loss: 0.7443, Feature Reconstruction Loss: 0.1646, Neighborhood Reconstruction Loss: 1.3241\n",
      "Epoch [300/500], Joint Loss: 0.7230, Feature Reconstruction Loss: 0.1614, Neighborhood Reconstruction Loss: 1.2847\n",
      "Epoch [310/500], Joint Loss: 0.7251, Feature Reconstruction Loss: 0.1578, Neighborhood Reconstruction Loss: 1.2923\n",
      "Epoch [320/500], Joint Loss: 0.7235, Feature Reconstruction Loss: 0.1571, Neighborhood Reconstruction Loss: 1.2899\n",
      "Epoch [330/500], Joint Loss: 0.7187, Feature Reconstruction Loss: 0.1550, Neighborhood Reconstruction Loss: 1.2824\n",
      "Epoch [340/500], Joint Loss: 0.7290, Feature Reconstruction Loss: 0.1667, Neighborhood Reconstruction Loss: 1.2913\n",
      "Epoch [350/500], Joint Loss: 0.7094, Feature Reconstruction Loss: 0.1545, Neighborhood Reconstruction Loss: 1.2644\n",
      "Epoch [360/500], Joint Loss: 0.7241, Feature Reconstruction Loss: 0.1536, Neighborhood Reconstruction Loss: 1.2945\n",
      "Epoch [370/500], Joint Loss: 0.7132, Feature Reconstruction Loss: 0.1507, Neighborhood Reconstruction Loss: 1.2757\n",
      "Epoch [380/500], Joint Loss: 0.7152, Feature Reconstruction Loss: 0.1614, Neighborhood Reconstruction Loss: 1.2690\n",
      "Epoch [390/500], Joint Loss: 0.7119, Feature Reconstruction Loss: 0.1498, Neighborhood Reconstruction Loss: 1.2740\n",
      "Epoch [400/500], Joint Loss: 0.7173, Feature Reconstruction Loss: 0.1499, Neighborhood Reconstruction Loss: 1.2846\n",
      "Epoch [410/500], Joint Loss: 0.6954, Feature Reconstruction Loss: 0.1493, Neighborhood Reconstruction Loss: 1.2416\n",
      "Epoch [420/500], Joint Loss: 0.7034, Feature Reconstruction Loss: 0.1498, Neighborhood Reconstruction Loss: 1.2569\n",
      "Epoch [430/500], Joint Loss: 0.6911, Feature Reconstruction Loss: 0.1540, Neighborhood Reconstruction Loss: 1.2282\n",
      "Epoch [440/500], Joint Loss: 0.7278, Feature Reconstruction Loss: 0.1499, Neighborhood Reconstruction Loss: 1.3058\n",
      "Epoch [450/500], Joint Loss: 0.6773, Feature Reconstruction Loss: 0.1459, Neighborhood Reconstruction Loss: 1.2088\n",
      "Epoch [460/500], Joint Loss: 0.7052, Feature Reconstruction Loss: 0.1451, Neighborhood Reconstruction Loss: 1.2653\n",
      "Epoch [470/500], Joint Loss: 0.6889, Feature Reconstruction Loss: 0.1462, Neighborhood Reconstruction Loss: 1.2315\n",
      "Epoch [480/500], Joint Loss: 0.6859, Feature Reconstruction Loss: 0.1434, Neighborhood Reconstruction Loss: 1.2285\n",
      "Epoch [490/500], Joint Loss: 0.7109, Feature Reconstruction Loss: 0.1407, Neighborhood Reconstruction Loss: 1.2812\n",
      "Epoch [500/500], Joint Loss: 0.7128, Feature Reconstruction Loss: 0.1375, Neighborhood Reconstruction Loss: 1.2881\n",
      "Accuracy: 0.5126\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.29      0.99      0.45       784\n",
      "           1       0.00      0.00      0.00       650\n",
      "           2       1.00      0.97      0.99       576\n",
      "           3       0.00      0.00      0.00       625\n",
      "           4       0.00      0.00      0.00       585\n",
      "           5       0.98      0.95      0.97       711\n",
      "\n",
      "    accuracy                           0.51      3931\n",
      "   macro avg       0.38      0.49      0.40      3931\n",
      "weighted avg       0.38      0.51      0.41      3931\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Qishan\\anaconda3\\envs\\exam\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Qishan\\anaconda3\\envs\\exam\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n",
      "C:\\Users\\Qishan\\anaconda3\\envs\\exam\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1509: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler, label_binarize\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.metrics import classification_report, accuracy_score, roc_curve, auc, confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(f'Using device: {device}')\n",
    "\n",
    "data_dict = np.load('dataset.npy', allow_pickle=True).item()\n",
    "all_frames = data_dict['frames']\n",
    "all_labels = data_dict['labels']\n",
    "\n",
    "scaler = StandardScaler()\n",
    "num_samples, num_timesteps, num_features = all_frames.shape\n",
    "all_frames = all_frames.reshape(num_samples * num_timesteps, num_features)\n",
    "all_frames = scaler.fit_transform(all_frames)\n",
    "all_frames = all_frames.reshape(num_samples, num_timesteps, num_features)\n",
    "\n",
    "all_frames = torch.tensor(all_frames, dtype=torch.float32).to(device)\n",
    "all_labels = torch.tensor(all_labels, dtype=torch.long).to(device)\n",
    "\n",
    "class Autoencoder(nn.Module):\n",
    "    def __init__(self, input_dim, encoding_dim):\n",
    "        super(Autoencoder, self).__init__()\n",
    "        self.encoder = nn.Sequential(\n",
    "            nn.Linear(input_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, encoding_dim)\n",
    "        )\n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.Linear(encoding_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, input_dim)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.encoder(x)\n",
    "        x = self.decoder(x)\n",
    "        return x\n",
    "    \n",
    "    def encode(self, x):\n",
    "        return self.encoder(x)\n",
    "\n",
    "input_dim = num_timesteps * num_features\n",
    "encoding_dim = 64\n",
    "learning_rate = 0.001\n",
    "num_epochs_ae = 200\n",
    "batch_size = 32\n",
    "\n",
    "autoencoder = Autoencoder(input_dim=input_dim, encoding_dim=encoding_dim).to(device)\n",
    "criterion_ae = nn.MSELoss()\n",
    "optimizer_ae = optim.Adam(autoencoder.parameters(), lr=learning_rate)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(all_frames, batch_size=batch_size, shuffle=True)\n",
    "\n",
    "for epoch in range(num_epochs_ae):\n",
    "    autoencoder.train()\n",
    "    running_loss = 0.0\n",
    "    for data in train_loader:\n",
    "        data = data.view(data.size(0), -1)\n",
    "        optimizer_ae.zero_grad()\n",
    "        output = autoencoder(data)\n",
    "        loss = criterion_ae(output, data)\n",
    "        loss.backward()\n",
    "        optimizer_ae.step()\n",
    "        running_loss += loss.item() * data.size(0)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ae}], Loss: {running_loss / len(train_loader.dataset):.4f}')\n",
    "\n",
    "for param in autoencoder.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "all_features = autoencoder.encode(all_frames.view(all_frames.size(0), -1)).detach().cpu().numpy()\n",
    "\n",
    "class MLPEncoder(nn.Module):\n",
    "    def __init__(self, in_channels, hidden_channels, out_channels):\n",
    "        super(MLPEncoder, self).__init__()\n",
    "        self.fc1 = nn.Linear(in_channels, hidden_channels)\n",
    "        self.fc2 = nn.Linear(hidden_channels, out_channels)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = F.relu(self.fc1(x))\n",
    "        z = self.fc2(x)\n",
    "        return z\n",
    "\n",
    "class FeatureDecoder(nn.Module):\n",
    "    def __init__(self, in_channels, out_channels):\n",
    "        super(FeatureDecoder, self).__init__()\n",
    "        self.fc = nn.Linear(in_channels, out_channels)\n",
    "\n",
    "    def forward(self, z):\n",
    "        return self.fc(z)\n",
    "\n",
    "class InnerProductDecoder(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(InnerProductDecoder, self).__init__()\n",
    "\n",
    "    def forward(self, z):\n",
    "        adj_pred = torch.sigmoid(torch.matmul(z, z.t()))\n",
    "        return adj_pred\n",
    "\n",
    "class FeatureTransformer(nn.Module):\n",
    "    def __init__(self, encoder, feature_decoder, adj_decoder):\n",
    "        super(FeatureTransformer, self).__init__()\n",
    "        self.encoder = encoder\n",
    "        self.feature_decoder = feature_decoder\n",
    "        self.adj_decoder = adj_decoder\n",
    "\n",
    "    def forward(self, x):\n",
    "        z = self.encoder(x)\n",
    "        x_reconstructed = self.feature_decoder(z)\n",
    "        adj_pred = self.adj_decoder(z)\n",
    "        return z, x_reconstructed, adj_pred\n",
    "\n",
    "input_dim_ft = encoding_dim\n",
    "hidden_dim_ft = 64\n",
    "output_dim_ft = 64\n",
    "batch_size_ft = 64\n",
    "num_epochs_ft = 500\n",
    "\n",
    "mlp_encoder = MLPEncoder(input_dim_ft, hidden_dim_ft, output_dim_ft).to(device)\n",
    "feature_decoder = FeatureDecoder(output_dim_ft, input_dim_ft).to(device)\n",
    "adj_decoder = InnerProductDecoder().to(device)\n",
    "\n",
    "feature_transformer = FeatureTransformer(mlp_encoder, feature_decoder, adj_decoder).to(device)\n",
    "optimizer_ft = optim.Adam(feature_transformer.parameters(), lr=learning_rate)\n",
    "\n",
    "def joint_loss_function(x, x_reconstructed, adj, adj_pred, alpha=0.5):\n",
    "    feature_reconstruction_loss = F.mse_loss(x_reconstructed, x)\n",
    "    adj_reconstruction_loss = F.binary_cross_entropy(adj_pred, adj)\n",
    "    loss = alpha * feature_reconstruction_loss + (1 - alpha) * adj_reconstruction_loss\n",
    "    return loss, feature_reconstruction_loss, adj_reconstruction_loss\n",
    "\n",
    "def create_adj_matrix(labels):\n",
    "    num_samples = len(labels)\n",
    "    adj_matrix = np.zeros((num_samples, num_samples))\n",
    "    for i in range(num_samples):\n",
    "        for j in range(num_samples):\n",
    "            if labels[i] == labels[j]:\n",
    "                adj_matrix[i, j] = 1\n",
    "    return adj_matrix\n",
    "\n",
    "class CustomDataset(torch.utils.data.Dataset):\n",
    "    def __init__(self, features, labels):\n",
    "        self.features = features\n",
    "        self.labels = labels\n",
    "        self.indices = np.arange(len(features))\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.features)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        return self.features[idx], self.labels[idx], self.indices[idx]\n",
    "\n",
    "train_features, test_features, train_labels, test_labels = train_test_split(\n",
    "    all_features, all_labels.cpu().numpy(), test_size=0.05, random_state=42, stratify=all_labels.cpu().numpy()\n",
    ")\n",
    "\n",
    "train_features = torch.tensor(train_features, dtype=torch.float32).to(device)\n",
    "test_features = torch.tensor(test_features, dtype=torch.float32).to(device)\n",
    "train_labels = torch.tensor(train_labels, dtype=torch.long).to(device)\n",
    "test_labels = torch.tensor(test_labels, dtype=torch.long).to(device)\n",
    "\n",
    "train_adj = create_adj_matrix(train_labels.cpu().numpy())\n",
    "\n",
    "train_adj_np = train_adj.copy()\n",
    "num_edges = int(np.sum(train_adj_np) / 2)\n",
    "num_disrupt = int(1 * num_edges)\n",
    "\n",
    "edges = np.array(np.triu_indices(train_adj_np.shape[0], 1)).T\n",
    "np.random.shuffle(edges)\n",
    "\n",
    "disrupted_edges = edges[:num_disrupt]\n",
    "\n",
    "for edge in disrupted_edges:\n",
    "    train_adj_np[edge[0], edge[1]] = 0\n",
    "    train_adj_np[edge[1], edge[0]] = 0\n",
    "\n",
    "train_adj = torch.tensor(train_adj_np, dtype=torch.float32).to(device)\n",
    "\n",
    "train_dataset = CustomDataset(train_features, train_labels)\n",
    "train_loader_ft = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size_ft, shuffle=True)\n",
    "\n",
    "joint_losses = []\n",
    "feature_reconstruction_losses = []\n",
    "neighborhood_reconstruction_losses = []\n",
    "\n",
    "for epoch in range(num_epochs_ft):\n",
    "    feature_transformer.train()\n",
    "    running_joint_loss = 0.0\n",
    "    running_feature_reconstruction_loss = 0.0\n",
    "    running_neighborhood_reconstruction_loss = 0.0\n",
    "    for data, labels, indices in train_loader_ft:\n",
    "        optimizer_ft.zero_grad()\n",
    "        z, x_reconstructed, adj_pred = feature_transformer(data)\n",
    "        batch_adj = train_adj[indices][:, indices]\n",
    "        joint_loss, feature_reconstruction_loss, neighborhood_reconstruction_loss = joint_loss_function(data, x_reconstructed, batch_adj, adj_pred)\n",
    "        joint_loss.backward()\n",
    "        optimizer_ft.step()\n",
    "        running_joint_loss += joint_loss.item() * data.size(0)\n",
    "        running_feature_reconstruction_loss += feature_reconstruction_loss.item() * data.size(0)\n",
    "        running_neighborhood_reconstruction_loss += neighborhood_reconstruction_loss.item() * data.size(0)\n",
    "    \n",
    "    avg_joint_loss = running_joint_loss / len(train_loader_ft.dataset)\n",
    "    avg_feature_reconstruction_loss = running_feature_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    avg_neighborhood_reconstruction_loss = running_neighborhood_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    joint_losses.append(avg_joint_loss)\n",
    "    feature_reconstruction_losses.append(avg_feature_reconstruction_loss)\n",
    "    neighborhood_reconstruction_losses.append(avg_neighborhood_reconstruction_loss)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ft}], Joint Loss: {avg_joint_loss:.4f}, Feature Reconstruction Loss: {avg_feature_reconstruction_loss:.4f}, Neighborhood Reconstruction Loss: {avg_neighborhood_reconstruction_loss:.4f}')\n",
    "\n",
    "for param in feature_transformer.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "with torch.no_grad():\n",
    "    test_reconstructed_features, _, _ = feature_transformer(test_features)\n",
    "test_reconstructed_features = test_reconstructed_features.cpu().numpy()\n",
    "\n",
    "svm_classifier = SVC(probability=True)\n",
    "svm_classifier.fit(test_reconstructed_features, test_labels.cpu().numpy())\n",
    "\n",
    "with torch.no_grad():\n",
    "    train_reconstructed_features, _, _ = feature_transformer(train_features)\n",
    "train_reconstructed_features = train_reconstructed_features.cpu().numpy()\n",
    "\n",
    "train_pred_labels = svm_classifier.predict(train_reconstructed_features)\n",
    "\n",
    "accuracy = accuracy_score(train_labels.cpu().numpy(), train_pred_labels)\n",
    "print(f'Accuracy: {accuracy:.4f}')\n",
    "print(classification_report(train_labels.cpu().numpy(), train_pred_labels))\n",
    "\n",
    "train_labels_binarized = label_binarize(train_labels.cpu().numpy(), classes=np.unique(train_labels.cpu().numpy()))\n",
    "n_classes = train_labels_binarized.shape[1]\n",
    "\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "\n",
    "for i in range(n_classes):\n",
    "    fpr[i], tpr[i], _ = roc_curve(train_labels_binarized[:, i], svm_classifier.predict_proba(train_reconstructed_features)[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "plt.figure()\n",
    "colors = ['aqua', 'darkorange', 'cornflowerblue', 'green', 'red', 'purple']\n",
    "for i, color in zip(range(n_classes), colors):\n",
    "    plt.plot(fpr[i], tpr[i], color=color, lw=2,\n",
    "             label=f'ROC curve of class {i} (area = {roc_auc[i]:.2f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', lw=2)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curves')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "conf_matrix = confusion_matrix(train_labels.cpu().numpy(), train_pred_labels)\n",
    "plt.figure(figsize=(10, 10))\n",
    "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(range(1, num_epochs_ft + 1), joint_losses, label='Joint Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), feature_reconstruction_losses, label='Feature Reconstruction Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), neighborhood_reconstruction_losses, label='Neighborhood Reconstruction Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss Curves')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "743c2144",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "090368da",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cuda\n",
      "Epoch [10/200], Loss: 0.0868\n",
      "Epoch [20/200], Loss: 0.0629\n",
      "Epoch [30/200], Loss: 0.0539\n",
      "Epoch [40/200], Loss: 0.0567\n",
      "Epoch [50/200], Loss: 0.0444\n",
      "Epoch [60/200], Loss: 0.0574\n",
      "Epoch [70/200], Loss: 0.0534\n",
      "Epoch [80/200], Loss: 0.0465\n",
      "Epoch [90/200], Loss: 0.0526\n",
      "Epoch [100/200], Loss: 0.0421\n",
      "Epoch [110/200], Loss: 0.0458\n",
      "Epoch [120/200], Loss: 0.0445\n",
      "Epoch [130/200], Loss: 0.0381\n",
      "Epoch [140/200], Loss: 0.0344\n",
      "Epoch [150/200], Loss: 0.0397\n",
      "Epoch [160/200], Loss: 0.0380\n",
      "Epoch [170/200], Loss: 0.0348\n",
      "Epoch [180/200], Loss: 0.0395\n",
      "Epoch [190/200], Loss: 0.0385\n",
      "Epoch [200/200], Loss: 0.0319\n",
      "Epoch [10/500], Joint Loss: 1.6711, Feature Reconstruction Loss: 1.7068, Neighborhood Reconstruction Loss: 1.6353\n",
      "Epoch [20/500], Joint Loss: 1.2641, Feature Reconstruction Loss: 0.9736, Neighborhood Reconstruction Loss: 1.5547\n",
      "Epoch [30/500], Joint Loss: 1.0838, Feature Reconstruction Loss: 0.6747, Neighborhood Reconstruction Loss: 1.4929\n",
      "Epoch [40/500], Joint Loss: 0.9694, Feature Reconstruction Loss: 0.5170, Neighborhood Reconstruction Loss: 1.4217\n",
      "Epoch [50/500], Joint Loss: 0.9083, Feature Reconstruction Loss: 0.4226, Neighborhood Reconstruction Loss: 1.3939\n",
      "Epoch [60/500], Joint Loss: 0.9008, Feature Reconstruction Loss: 0.3657, Neighborhood Reconstruction Loss: 1.4359\n",
      "Epoch [70/500], Joint Loss: 0.8449, Feature Reconstruction Loss: 0.3217, Neighborhood Reconstruction Loss: 1.3680\n",
      "Epoch [80/500], Joint Loss: 0.8475, Feature Reconstruction Loss: 0.3000, Neighborhood Reconstruction Loss: 1.3950\n",
      "Epoch [90/500], Joint Loss: 0.8231, Feature Reconstruction Loss: 0.2829, Neighborhood Reconstruction Loss: 1.3634\n",
      "Epoch [100/500], Joint Loss: 0.8100, Feature Reconstruction Loss: 0.2709, Neighborhood Reconstruction Loss: 1.3491\n",
      "Epoch [110/500], Joint Loss: 0.7941, Feature Reconstruction Loss: 0.2608, Neighborhood Reconstruction Loss: 1.3273\n",
      "Epoch [120/500], Joint Loss: 0.7806, Feature Reconstruction Loss: 0.2454, Neighborhood Reconstruction Loss: 1.3157\n",
      "Epoch [130/500], Joint Loss: 0.8064, Feature Reconstruction Loss: 0.2432, Neighborhood Reconstruction Loss: 1.3696\n",
      "Epoch [140/500], Joint Loss: 0.7927, Feature Reconstruction Loss: 0.2301, Neighborhood Reconstruction Loss: 1.3553\n",
      "Epoch [150/500], Joint Loss: 0.7866, Feature Reconstruction Loss: 0.2231, Neighborhood Reconstruction Loss: 1.3500\n",
      "Epoch [160/500], Joint Loss: 0.7769, Feature Reconstruction Loss: 0.2148, Neighborhood Reconstruction Loss: 1.3390\n",
      "Epoch [170/500], Joint Loss: 0.7617, Feature Reconstruction Loss: 0.2070, Neighborhood Reconstruction Loss: 1.3164\n",
      "Epoch [180/500], Joint Loss: 0.8126, Feature Reconstruction Loss: 0.2166, Neighborhood Reconstruction Loss: 1.4086\n",
      "Epoch [190/500], Joint Loss: 0.7506, Feature Reconstruction Loss: 0.1971, Neighborhood Reconstruction Loss: 1.3041\n",
      "Epoch [200/500], Joint Loss: 0.7398, Feature Reconstruction Loss: 0.1878, Neighborhood Reconstruction Loss: 1.2918\n",
      "Epoch [210/500], Joint Loss: 0.7246, Feature Reconstruction Loss: 0.1902, Neighborhood Reconstruction Loss: 1.2590\n",
      "Epoch [220/500], Joint Loss: 0.7526, Feature Reconstruction Loss: 0.1878, Neighborhood Reconstruction Loss: 1.3174\n",
      "Epoch [230/500], Joint Loss: 0.6992, Feature Reconstruction Loss: 0.1810, Neighborhood Reconstruction Loss: 1.2175\n",
      "Epoch [240/500], Joint Loss: 0.7225, Feature Reconstruction Loss: 0.1777, Neighborhood Reconstruction Loss: 1.2673\n",
      "Epoch [250/500], Joint Loss: 0.7462, Feature Reconstruction Loss: 0.1810, Neighborhood Reconstruction Loss: 1.3113\n",
      "Epoch [260/500], Joint Loss: 0.7265, Feature Reconstruction Loss: 0.1760, Neighborhood Reconstruction Loss: 1.2769\n",
      "Epoch [270/500], Joint Loss: 0.6856, Feature Reconstruction Loss: 0.1704, Neighborhood Reconstruction Loss: 1.2008\n",
      "Epoch [280/500], Joint Loss: 0.7468, Feature Reconstruction Loss: 0.1713, Neighborhood Reconstruction Loss: 1.3224\n",
      "Epoch [290/500], Joint Loss: 0.7260, Feature Reconstruction Loss: 0.1671, Neighborhood Reconstruction Loss: 1.2850\n",
      "Epoch [300/500], Joint Loss: 0.7107, Feature Reconstruction Loss: 0.1635, Neighborhood Reconstruction Loss: 1.2579\n",
      "Epoch [310/500], Joint Loss: 0.7114, Feature Reconstruction Loss: 0.1615, Neighborhood Reconstruction Loss: 1.2614\n",
      "Epoch [320/500], Joint Loss: 0.7126, Feature Reconstruction Loss: 0.1633, Neighborhood Reconstruction Loss: 1.2619\n",
      "Epoch [330/500], Joint Loss: 0.7379, Feature Reconstruction Loss: 0.2012, Neighborhood Reconstruction Loss: 1.2745\n",
      "Epoch [340/500], Joint Loss: 0.7459, Feature Reconstruction Loss: 0.1612, Neighborhood Reconstruction Loss: 1.3306\n",
      "Epoch [350/500], Joint Loss: 0.7234, Feature Reconstruction Loss: 0.1562, Neighborhood Reconstruction Loss: 1.2906\n",
      "Epoch [360/500], Joint Loss: 0.7110, Feature Reconstruction Loss: 0.1517, Neighborhood Reconstruction Loss: 1.2702\n",
      "Epoch [370/500], Joint Loss: 0.6859, Feature Reconstruction Loss: 0.1505, Neighborhood Reconstruction Loss: 1.2213\n",
      "Epoch [380/500], Joint Loss: 0.6808, Feature Reconstruction Loss: 0.1513, Neighborhood Reconstruction Loss: 1.2103\n",
      "Epoch [390/500], Joint Loss: 0.7169, Feature Reconstruction Loss: 0.1516, Neighborhood Reconstruction Loss: 1.2822\n",
      "Epoch [400/500], Joint Loss: 0.6933, Feature Reconstruction Loss: 0.1439, Neighborhood Reconstruction Loss: 1.2427\n",
      "Epoch [410/500], Joint Loss: 0.7065, Feature Reconstruction Loss: 0.1433, Neighborhood Reconstruction Loss: 1.2697\n",
      "Epoch [420/500], Joint Loss: 0.6949, Feature Reconstruction Loss: 0.1407, Neighborhood Reconstruction Loss: 1.2492\n",
      "Epoch [430/500], Joint Loss: 0.7235, Feature Reconstruction Loss: 0.1591, Neighborhood Reconstruction Loss: 1.2879\n",
      "Epoch [440/500], Joint Loss: 0.7134, Feature Reconstruction Loss: 0.1415, Neighborhood Reconstruction Loss: 1.2852\n",
      "Epoch [450/500], Joint Loss: 0.7025, Feature Reconstruction Loss: 0.1462, Neighborhood Reconstruction Loss: 1.2588\n",
      "Epoch [460/500], Joint Loss: 0.6828, Feature Reconstruction Loss: 0.1394, Neighborhood Reconstruction Loss: 1.2262\n",
      "Epoch [470/500], Joint Loss: 0.6722, Feature Reconstruction Loss: 0.1410, Neighborhood Reconstruction Loss: 1.2034\n",
      "Epoch [480/500], Joint Loss: 0.6615, Feature Reconstruction Loss: 0.1372, Neighborhood Reconstruction Loss: 1.1857\n",
      "Epoch [490/500], Joint Loss: 0.6823, Feature Reconstruction Loss: 0.1342, Neighborhood Reconstruction Loss: 1.2305\n",
      "Epoch [500/500], Joint Loss: 0.6862, Feature Reconstruction Loss: 0.1369, Neighborhood Reconstruction Loss: 1.2355\n",
      "Accuracy: 0.9517\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.98      0.90       784\n",
      "           1       0.98      0.96      0.97       650\n",
      "           2       1.00      0.98      0.99       576\n",
      "           3       0.99      0.93      0.96       625\n",
      "           4       0.98      0.90      0.93       585\n",
      "           5       0.99      0.95      0.97       711\n",
      "\n",
      "    accuracy                           0.95      3931\n",
      "   macro avg       0.96      0.95      0.95      3931\n",
      "weighted avg       0.96      0.95      0.95      3931\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler, label_binarize\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import classification_report, accuracy_score, roc_curve, auc, confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(f'Using device: {device}')\n",
    "\n",
    "data_dict = np.load('dataset.npy', allow_pickle=True).item()\n",
    "all_frames = data_dict['frames']\n",
    "all_labels = data_dict['labels']\n",
    "\n",
    "scaler = StandardScaler()\n",
    "num_samples, num_timesteps, num_features = all_frames.shape\n",
    "all_frames = all_frames.reshape(num_samples * num_timesteps, num_features)\n",
    "all_frames = scaler.fit_transform(all_frames)\n",
    "all_frames = all_frames.reshape(num_samples, num_timesteps, num_features)\n",
    "\n",
    "all_frames = torch.tensor(all_frames, dtype=torch.float32).to(device)\n",
    "all_labels = torch.tensor(all_labels, dtype=torch.long).to(device)\n",
    "\n",
    "class Autoencoder(nn.Module):\n",
    "    def __init__(self, input_dim, encoding_dim):\n",
    "        super(Autoencoder, self).__init__()\n",
    "        self.encoder = nn.Sequential(\n",
    "            nn.Linear(input_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, encoding_dim)\n",
    "        )\n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.Linear(encoding_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, input_dim)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.encoder(x)\n",
    "        x = self.decoder(x)\n",
    "        return x\n",
    "    \n",
    "    def encode(self, x):\n",
    "        return self.encoder(x)\n",
    "\n",
    "input_dim = num_timesteps * num_features\n",
    "encoding_dim = 64\n",
    "learning_rate = 0.001\n",
    "num_epochs_ae = 200\n",
    "batch_size = 32\n",
    "\n",
    "autoencoder = Autoencoder(input_dim=input_dim, encoding_dim=encoding_dim).to(device)\n",
    "criterion_ae = nn.MSELoss()\n",
    "optimizer_ae = optim.Adam(autoencoder.parameters(), lr=learning_rate)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(all_frames, batch_size=batch_size, shuffle=True)\n",
    "\n",
    "for epoch in range(num_epochs_ae):\n",
    "    autoencoder.train()\n",
    "    running_loss = 0.0\n",
    "    for data in train_loader:\n",
    "        data = data.view(data.size(0), -1)\n",
    "        optimizer_ae.zero_grad()\n",
    "        output = autoencoder(data)\n",
    "        loss = criterion_ae(output, data)\n",
    "        loss.backward()\n",
    "        optimizer_ae.step()\n",
    "        running_loss += loss.item() * data.size(0)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ae}], Loss: {running_loss / len(train_loader.dataset):.4f}')\n",
    "\n",
    "for param in autoencoder.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "all_features = autoencoder.encode(all_frames.view(all_frames.size(0), -1)).detach().cpu().numpy()\n",
    "\n",
    "class MLPEncoder(nn.Module):\n",
    "    def __init__(self, in_channels, hidden_channels, out_channels):\n",
    "        super(MLPEncoder, self).__init__()\n",
    "        self.fc1 = nn.Linear(in_channels, hidden_channels)\n",
    "        self.fc2 = nn.Linear(hidden_channels, out_channels)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = F.relu(self.fc1(x))\n",
    "        z = self.fc2(x)\n",
    "        return z\n",
    "\n",
    "class FeatureDecoder(nn.Module):\n",
    "    def __init__(self, in_channels, out_channels):\n",
    "        super(FeatureDecoder, self).__init__()\n",
    "        self.fc = nn.Linear(in_channels, out_channels)\n",
    "\n",
    "    def forward(self, z):\n",
    "        return self.fc(z)\n",
    "\n",
    "class InnerProductDecoder(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(InnerProductDecoder, self).__init__()\n",
    "\n",
    "    def forward(self, z):\n",
    "        adj_pred = torch.sigmoid(torch.matmul(z, z.t()))\n",
    "        return adj_pred\n",
    "\n",
    "class FeatureTransformer(nn.Module):\n",
    "    def __init__(self, encoder, feature_decoder, adj_decoder):\n",
    "        super(FeatureTransformer, self).__init__()\n",
    "        self.encoder = encoder\n",
    "        self.feature_decoder = feature_decoder\n",
    "        self.adj_decoder = adj_decoder\n",
    "\n",
    "    def forward(self, x):\n",
    "        z = self.encoder(x)\n",
    "        x_reconstructed = self.feature_decoder(z)\n",
    "        adj_pred = self.adj_decoder(z)\n",
    "        return z, x_reconstructed, adj_pred\n",
    "\n",
    "input_dim_ft = encoding_dim\n",
    "hidden_dim_ft = 64\n",
    "output_dim_ft = 64\n",
    "batch_size_ft = 64\n",
    "num_epochs_ft = 500\n",
    "\n",
    "mlp_encoder = MLPEncoder(input_dim_ft, hidden_dim_ft, output_dim_ft).to(device)\n",
    "feature_decoder = FeatureDecoder(output_dim_ft, input_dim_ft).to(device)\n",
    "adj_decoder = InnerProductDecoder().to(device)\n",
    "\n",
    "feature_transformer = FeatureTransformer(mlp_encoder, feature_decoder, adj_decoder).to(device)\n",
    "optimizer_ft = optim.Adam(feature_transformer.parameters(), lr=learning_rate)\n",
    "\n",
    "def joint_loss_function(x, x_reconstructed, adj, adj_pred, alpha=0.5):\n",
    "    feature_reconstruction_loss = F.mse_loss(x_reconstructed, x)\n",
    "    adj_reconstruction_loss = F.binary_cross_entropy(adj_pred, adj)\n",
    "    loss = alpha * feature_reconstruction_loss + (1 - alpha) * adj_reconstruction_loss\n",
    "    return loss, feature_reconstruction_loss, adj_reconstruction_loss\n",
    "\n",
    "def create_adj_matrix(labels):\n",
    "    num_samples = len(labels)\n",
    "    adj_matrix = np.zeros((num_samples, num_samples))\n",
    "    for i in range(num_samples):\n",
    "        for j in range(num_samples):\n",
    "            if labels[i] == labels[j]:\n",
    "                adj_matrix[i, j] = 1\n",
    "    return adj_matrix\n",
    "\n",
    "class CustomDataset(torch.utils.data.Dataset):\n",
    "    def __init__(self, features, labels):\n",
    "        self.features = features\n",
    "        self.labels = labels\n",
    "        self.indices = np.arange(len(features))\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.features)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        return self.features[idx], self.labels[idx], self.indices[idx]\n",
    "\n",
    "train_features, test_features, train_labels, test_labels = train_test_split(\n",
    "    all_features, all_labels.cpu().numpy(), test_size=0.05, random_state=42, stratify=all_labels.cpu().numpy()\n",
    ")\n",
    "\n",
    "train_features = torch.tensor(train_features, dtype=torch.float32).to(device)\n",
    "test_features = torch.tensor(test_features, dtype=torch.float32).to(device)\n",
    "train_labels = torch.tensor(train_labels, dtype=torch.long).to(device)\n",
    "test_labels = torch.tensor(test_labels, dtype=torch.long).to(device)\n",
    "\n",
    "train_adj = create_adj_matrix(train_labels.cpu().numpy())\n",
    "\n",
    "train_adj_np = train_adj.copy()\n",
    "num_edges = int(np.sum(train_adj_np) / 2)\n",
    "num_disrupt = int(1 * num_edges)\n",
    "\n",
    "edges = np.array(np.triu_indices(train_adj_np.shape[0], 1)).T\n",
    "np.random.shuffle(edges)\n",
    "\n",
    "disrupted_edges = edges[:num_disrupt]\n",
    "\n",
    "for edge in disrupted_edges:\n",
    "    train_adj_np[edge[0], edge[1]] = 0\n",
    "    train_adj_np[edge[1], edge[0]] = 0\n",
    "\n",
    "train_adj = torch.tensor(train_adj_np, dtype=torch.float32).to(device)\n",
    "\n",
    "train_dataset = CustomDataset(train_features, train_labels)\n",
    "train_loader_ft = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size_ft, shuffle=True)\n",
    "\n",
    "joint_losses = []\n",
    "feature_reconstruction_losses = []\n",
    "neighborhood_reconstruction_losses = []\n",
    "\n",
    "for epoch in range(num_epochs_ft):\n",
    "    feature_transformer.train()\n",
    "    running_joint_loss = 0.0\n",
    "    running_feature_reconstruction_loss = 0.0\n",
    "    running_neighborhood_reconstruction_loss = 0.0\n",
    "    for data, labels, indices in train_loader_ft:\n",
    "        optimizer_ft.zero_grad()\n",
    "        z, x_reconstructed, adj_pred = feature_transformer(data)\n",
    "        batch_adj = train_adj[indices][:, indices]\n",
    "        joint_loss, feature_reconstruction_loss, neighborhood_reconstruction_loss = joint_loss_function(data, x_reconstructed, batch_adj, adj_pred)\n",
    "        joint_loss.backward()\n",
    "        optimizer_ft.step()\n",
    "        running_joint_loss += joint_loss.item() * data.size(0)\n",
    "        running_feature_reconstruction_loss += feature_reconstruction_loss.item() * data.size(0)\n",
    "        running_neighborhood_reconstruction_loss += neighborhood_reconstruction_loss.item() * data.size(0)\n",
    "    \n",
    "    avg_joint_loss = running_joint_loss / len(train_loader_ft.dataset)\n",
    "    avg_feature_reconstruction_loss = running_feature_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    avg_neighborhood_reconstruction_loss = running_neighborhood_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    joint_losses.append(avg_joint_loss)\n",
    "    feature_reconstruction_losses.append(avg_feature_reconstruction_loss)\n",
    "    neighborhood_reconstruction_losses.append(avg_neighborhood_reconstruction_loss)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ft}], Joint Loss: {avg_joint_loss:.4f}, Feature Reconstruction Loss: {avg_feature_reconstruction_loss:.4f}, Neighborhood Reconstruction Loss: {avg_neighborhood_reconstruction_loss:.4f}')\n",
    "\n",
    "for param in feature_transformer.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "with torch.no_grad():\n",
    "    test_reconstructed_features, _, _ = feature_transformer(test_features)\n",
    "test_reconstructed_features = test_reconstructed_features.cpu().numpy()\n",
    "\n",
    "knn_classifier = KNeighborsClassifier(n_neighbors=5)\n",
    "knn_classifier.fit(test_reconstructed_features, test_labels.cpu().numpy())\n",
    "\n",
    "with torch.no_grad():\n",
    "    train_reconstructed_features, _, _ = feature_transformer(train_features)\n",
    "train_reconstructed_features = train_reconstructed_features.cpu().numpy()\n",
    "\n",
    "train_pred_labels = knn_classifier.predict(train_reconstructed_features)\n",
    "\n",
    "accuracy = accuracy_score(train_labels.cpu().numpy(), train_pred_labels)\n",
    "print(f'Accuracy: {accuracy:.4f}')\n",
    "print(classification_report(train_labels.cpu().numpy(), train_pred_labels))\n",
    "\n",
    "train_labels_binarized = label_binarize(train_labels.cpu().numpy(), classes=np.unique(train_labels.cpu().numpy()))\n",
    "n_classes = train_labels_binarized.shape[1]\n",
    "\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "\n",
    "for i in range(n_classes):\n",
    "    fpr[i], tpr[i], _ = roc_curve(train_labels_binarized[:, i], knn_classifier.predict_proba(train_reconstructed_features)[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "plt.figure()\n",
    "colors = ['aqua', 'darkorange', 'cornflowerblue', 'green', 'red', 'purple']\n",
    "for i, color in zip(range(n_classes), colors):\n",
    "    plt.plot(fpr[i], tpr[i], color=color, lw=2,\n",
    "             label=f'ROC curve of class {i} (area = {roc_auc[i]:.2f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', lw=2)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curves')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "conf_matrix = confusion_matrix(train_labels.cpu().numpy(), train_pred_labels)\n",
    "plt.figure(figsize=(10, 10))\n",
    "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(range(1, num_epochs_ft + 1), joint_losses, label='Joint Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), feature_reconstruction_losses, label='Feature Reconstruction Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), neighborhood_reconstruction_losses, label='Neighborhood Reconstruction Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss Curves')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af89ceab",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "7ebeb9c7",
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using device: cuda\n",
      "Epoch [10/200], Loss: 0.0989\n",
      "Epoch [20/200], Loss: 0.0592\n",
      "Epoch [30/200], Loss: 0.0585\n",
      "Epoch [40/200], Loss: 0.0502\n",
      "Epoch [50/200], Loss: 0.0518\n",
      "Epoch [60/200], Loss: 0.0458\n",
      "Epoch [70/200], Loss: 0.0456\n",
      "Epoch [80/200], Loss: 0.0415\n",
      "Epoch [90/200], Loss: 0.0383\n",
      "Epoch [100/200], Loss: 0.0401\n",
      "Epoch [110/200], Loss: 0.0412\n",
      "Epoch [120/200], Loss: 0.0415\n",
      "Epoch [130/200], Loss: 0.0377\n",
      "Epoch [140/200], Loss: 0.0335\n",
      "Epoch [150/200], Loss: 0.0439\n",
      "Epoch [160/200], Loss: 0.0399\n",
      "Epoch [170/200], Loss: 0.0423\n",
      "Epoch [180/200], Loss: 0.0370\n",
      "Epoch [190/200], Loss: 0.0338\n",
      "Epoch [200/200], Loss: 0.0388\n",
      "Epoch [10/500], Joint Loss: 1.6686, Feature Reconstruction Loss: 1.6437, Neighborhood Reconstruction Loss: 1.6935\n",
      "Epoch [20/500], Joint Loss: 1.2524, Feature Reconstruction Loss: 0.9393, Neighborhood Reconstruction Loss: 1.5656\n",
      "Epoch [30/500], Joint Loss: 1.0676, Feature Reconstruction Loss: 0.6468, Neighborhood Reconstruction Loss: 1.4884\n",
      "Epoch [40/500], Joint Loss: 0.9633, Feature Reconstruction Loss: 0.4999, Neighborhood Reconstruction Loss: 1.4266\n",
      "Epoch [50/500], Joint Loss: 0.8994, Feature Reconstruction Loss: 0.4052, Neighborhood Reconstruction Loss: 1.3935\n",
      "Epoch [60/500], Joint Loss: 0.8750, Feature Reconstruction Loss: 0.3523, Neighborhood Reconstruction Loss: 1.3976\n",
      "Epoch [70/500], Joint Loss: 0.8309, Feature Reconstruction Loss: 0.3109, Neighborhood Reconstruction Loss: 1.3508\n",
      "Epoch [80/500], Joint Loss: 0.8512, Feature Reconstruction Loss: 0.2791, Neighborhood Reconstruction Loss: 1.4232\n",
      "Epoch [90/500], Joint Loss: 0.7925, Feature Reconstruction Loss: 0.2561, Neighborhood Reconstruction Loss: 1.3289\n",
      "Epoch [100/500], Joint Loss: 0.7944, Feature Reconstruction Loss: 0.2392, Neighborhood Reconstruction Loss: 1.3495\n",
      "Epoch [110/500], Joint Loss: 0.7969, Feature Reconstruction Loss: 0.2243, Neighborhood Reconstruction Loss: 1.3695\n",
      "Epoch [120/500], Joint Loss: 0.7904, Feature Reconstruction Loss: 0.2131, Neighborhood Reconstruction Loss: 1.3676\n",
      "Epoch [130/500], Joint Loss: 0.7457, Feature Reconstruction Loss: 0.1997, Neighborhood Reconstruction Loss: 1.2918\n",
      "Epoch [140/500], Joint Loss: 0.7498, Feature Reconstruction Loss: 0.1903, Neighborhood Reconstruction Loss: 1.3093\n",
      "Epoch [150/500], Joint Loss: 0.7523, Feature Reconstruction Loss: 0.1789, Neighborhood Reconstruction Loss: 1.3257\n",
      "Epoch [160/500], Joint Loss: 0.7359, Feature Reconstruction Loss: 0.1765, Neighborhood Reconstruction Loss: 1.2953\n",
      "Epoch [170/500], Joint Loss: 0.6962, Feature Reconstruction Loss: 0.1677, Neighborhood Reconstruction Loss: 1.2248\n",
      "Epoch [180/500], Joint Loss: 0.7546, Feature Reconstruction Loss: 0.1703, Neighborhood Reconstruction Loss: 1.3390\n",
      "Epoch [190/500], Joint Loss: 0.7162, Feature Reconstruction Loss: 0.1599, Neighborhood Reconstruction Loss: 1.2724\n",
      "Epoch [200/500], Joint Loss: 0.7216, Feature Reconstruction Loss: 0.1542, Neighborhood Reconstruction Loss: 1.2891\n",
      "Epoch [210/500], Joint Loss: 0.7260, Feature Reconstruction Loss: 0.1542, Neighborhood Reconstruction Loss: 1.2978\n",
      "Epoch [220/500], Joint Loss: 0.7103, Feature Reconstruction Loss: 0.1489, Neighborhood Reconstruction Loss: 1.2717\n",
      "Epoch [230/500], Joint Loss: 0.6995, Feature Reconstruction Loss: 0.1447, Neighborhood Reconstruction Loss: 1.2544\n",
      "Epoch [240/500], Joint Loss: 0.7233, Feature Reconstruction Loss: 0.1386, Neighborhood Reconstruction Loss: 1.3080\n",
      "Epoch [250/500], Joint Loss: 0.7407, Feature Reconstruction Loss: 0.1569, Neighborhood Reconstruction Loss: 1.3246\n",
      "Epoch [260/500], Joint Loss: 0.7078, Feature Reconstruction Loss: 0.1406, Neighborhood Reconstruction Loss: 1.2749\n",
      "Epoch [270/500], Joint Loss: 0.6770, Feature Reconstruction Loss: 0.1357, Neighborhood Reconstruction Loss: 1.2182\n",
      "Epoch [280/500], Joint Loss: 0.6914, Feature Reconstruction Loss: 0.1357, Neighborhood Reconstruction Loss: 1.2472\n",
      "Epoch [290/500], Joint Loss: 0.7167, Feature Reconstruction Loss: 0.1370, Neighborhood Reconstruction Loss: 1.2963\n",
      "Epoch [300/500], Joint Loss: 0.7039, Feature Reconstruction Loss: 0.1370, Neighborhood Reconstruction Loss: 1.2708\n",
      "Epoch [310/500], Joint Loss: 0.6913, Feature Reconstruction Loss: 0.1328, Neighborhood Reconstruction Loss: 1.2497\n",
      "Epoch [320/500], Joint Loss: 0.6767, Feature Reconstruction Loss: 0.1291, Neighborhood Reconstruction Loss: 1.2242\n",
      "Epoch [330/500], Joint Loss: 0.7070, Feature Reconstruction Loss: 0.1263, Neighborhood Reconstruction Loss: 1.2877\n",
      "Epoch [340/500], Joint Loss: 0.6916, Feature Reconstruction Loss: 0.1244, Neighborhood Reconstruction Loss: 1.2589\n",
      "Epoch [350/500], Joint Loss: 0.6935, Feature Reconstruction Loss: 0.1226, Neighborhood Reconstruction Loss: 1.2645\n",
      "Epoch [360/500], Joint Loss: 0.6916, Feature Reconstruction Loss: 0.1314, Neighborhood Reconstruction Loss: 1.2518\n",
      "Epoch [370/500], Joint Loss: 0.6687, Feature Reconstruction Loss: 0.1221, Neighborhood Reconstruction Loss: 1.2154\n",
      "Epoch [380/500], Joint Loss: 0.7091, Feature Reconstruction Loss: 0.1272, Neighborhood Reconstruction Loss: 1.2910\n",
      "Epoch [390/500], Joint Loss: 0.6497, Feature Reconstruction Loss: 0.1182, Neighborhood Reconstruction Loss: 1.1812\n",
      "Epoch [400/500], Joint Loss: 0.6731, Feature Reconstruction Loss: 0.1203, Neighborhood Reconstruction Loss: 1.2260\n",
      "Epoch [410/500], Joint Loss: 0.7006, Feature Reconstruction Loss: 0.1190, Neighborhood Reconstruction Loss: 1.2821\n",
      "Epoch [420/500], Joint Loss: 0.6513, Feature Reconstruction Loss: 0.1171, Neighborhood Reconstruction Loss: 1.1854\n",
      "Epoch [430/500], Joint Loss: 0.6896, Feature Reconstruction Loss: 0.1198, Neighborhood Reconstruction Loss: 1.2593\n",
      "Epoch [440/500], Joint Loss: 0.6804, Feature Reconstruction Loss: 0.1201, Neighborhood Reconstruction Loss: 1.2407\n",
      "Epoch [450/500], Joint Loss: 0.6847, Feature Reconstruction Loss: 0.1250, Neighborhood Reconstruction Loss: 1.2445\n",
      "Epoch [460/500], Joint Loss: 0.6622, Feature Reconstruction Loss: 0.1182, Neighborhood Reconstruction Loss: 1.2063\n",
      "Epoch [470/500], Joint Loss: 0.6677, Feature Reconstruction Loss: 0.1207, Neighborhood Reconstruction Loss: 1.2147\n",
      "Epoch [480/500], Joint Loss: 0.7110, Feature Reconstruction Loss: 0.1191, Neighborhood Reconstruction Loss: 1.3029\n",
      "Epoch [490/500], Joint Loss: 0.6713, Feature Reconstruction Loss: 0.1230, Neighborhood Reconstruction Loss: 1.2197\n",
      "Epoch [500/500], Joint Loss: 0.6694, Feature Reconstruction Loss: 0.1209, Neighborhood Reconstruction Loss: 1.2180\n",
      "Accuracy: 0.9761\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.93      0.96       784\n",
      "           1       0.96      0.99      0.97       650\n",
      "           2       0.99      0.99      0.99       576\n",
      "           3       0.93      1.00      0.96       625\n",
      "           4       0.99      0.98      0.99       585\n",
      "           5       0.99      0.97      0.98       711\n",
      "\n",
      "    accuracy                           0.98      3931\n",
      "   macro avg       0.98      0.98      0.98      3931\n",
      "weighted avg       0.98      0.98      0.98      3931\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import StandardScaler, label_binarize\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import classification_report, accuracy_score, roc_curve, auc, confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(f'Using device: {device}')\n",
    "\n",
    "data_dict = np.load('dataset.npy', allow_pickle=True).item()\n",
    "all_frames = data_dict['frames']\n",
    "all_labels = data_dict['labels']\n",
    "\n",
    "scaler = StandardScaler()\n",
    "num_samples, num_timesteps, num_features = all_frames.shape\n",
    "all_frames = all_frames.reshape(num_samples * num_timesteps, num_features)\n",
    "all_frames = scaler.fit_transform(all_frames)\n",
    "all_frames = all_frames.reshape(num_samples, num_timesteps, num_features)\n",
    "\n",
    "all_frames = torch.tensor(all_frames, dtype=torch.float32).to(device)\n",
    "all_labels = torch.tensor(all_labels, dtype=torch.long).to(device)\n",
    "\n",
    "class Autoencoder(nn.Module):\n",
    "    def __init__(self, input_dim, encoding_dim):\n",
    "        super(Autoencoder, self).__init__()\n",
    "        self.encoder = nn.Sequential(\n",
    "            nn.Linear(input_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, encoding_dim)\n",
    "        )\n",
    "        self.decoder = nn.Sequential(\n",
    "            nn.Linear(encoding_dim, 512),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(512, input_dim)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.encoder(x)\n",
    "        x = self.decoder(x)\n",
    "        return x\n",
    "    \n",
    "    def encode(self, x):\n",
    "        return self.encoder(x)\n",
    "\n",
    "input_dim = num_timesteps * num_features\n",
    "encoding_dim = 64\n",
    "learning_rate = 0.001\n",
    "num_epochs_ae = 200\n",
    "batch_size = 32\n",
    "\n",
    "autoencoder = Autoencoder(input_dim=input_dim, encoding_dim=encoding_dim).to(device)\n",
    "criterion_ae = nn.MSELoss()\n",
    "optimizer_ae = optim.Adam(autoencoder.parameters(), lr=learning_rate)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(all_frames, batch_size=batch_size, shuffle=True)\n",
    "\n",
    "for epoch in range(num_epochs_ae):\n",
    "    autoencoder.train()\n",
    "    running_loss = 0.0\n",
    "    for data in train_loader:\n",
    "        data = data.view(data.size(0), -1)\n",
    "        optimizer_ae.zero_grad()\n",
    "        output = autoencoder(data)\n",
    "        loss = criterion_ae(output, data)\n",
    "        loss.backward()\n",
    "        optimizer_ae.step()\n",
    "        running_loss += loss.item() * data.size(0)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ae}], Loss: {running_loss / len(train_loader.dataset):.4f}')\n",
    "\n",
    "for param in autoencoder.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "all_features = autoencoder.encode(all_frames.view(all_frames.size(0), -1)).detach().cpu().numpy()\n",
    "\n",
    "class MLPEncoder(nn.Module):\n",
    "    def __init__(self, in_channels, hidden_channels, out_channels):\n",
    "        super(MLPEncoder, self).__init__()\n",
    "        self.fc1 = nn.Linear(in_channels, hidden_channels)\n",
    "        self.fc2 = nn.Linear(hidden_channels, out_channels)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = F.relu(self.fc1(x))\n",
    "        z = self.fc2(x)\n",
    "        return z\n",
    "\n",
    "class FeatureDecoder(nn.Module):\n",
    "    def __init__(self, in_channels, out_channels):\n",
    "        super(FeatureDecoder, self).__init__()\n",
    "        self.fc = nn.Linear(in_channels, out_channels)\n",
    "\n",
    "    def forward(self, z):\n",
    "        return self.fc(z)\n",
    "\n",
    "class InnerProductDecoder(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(InnerProductDecoder, self).__init__()\n",
    "\n",
    "    def forward(self, z):\n",
    "        adj_pred = torch.sigmoid(torch.matmul(z, z.t()))\n",
    "        return adj_pred\n",
    "\n",
    "class FeatureTransformer(nn.Module):\n",
    "    def __init__(self, encoder, feature_decoder, adj_decoder):\n",
    "        super(FeatureTransformer, self).__init__()\n",
    "        self.encoder = encoder\n",
    "        self.feature_decoder = feature_decoder\n",
    "        self.adj_decoder = adj_decoder\n",
    "\n",
    "    def forward(self, x):\n",
    "        z = self.encoder(x)\n",
    "        x_reconstructed = self.feature_decoder(z)\n",
    "        adj_pred = self.adj_decoder(z)\n",
    "        return z, x_reconstructed, adj_pred\n",
    "\n",
    "input_dim_ft = encoding_dim\n",
    "hidden_dim_ft = 64\n",
    "output_dim_ft = 64\n",
    "batch_size_ft = 64\n",
    "num_epochs_ft = 500\n",
    "\n",
    "mlp_encoder = MLPEncoder(input_dim_ft, hidden_dim_ft, output_dim_ft).to(device)\n",
    "feature_decoder = FeatureDecoder(output_dim_ft, input_dim_ft).to(device)\n",
    "adj_decoder = InnerProductDecoder().to(device)\n",
    "\n",
    "feature_transformer = FeatureTransformer(mlp_encoder, feature_decoder, adj_decoder).to(device)\n",
    "optimizer_ft = optim.Adam(feature_transformer.parameters(), lr=learning_rate)\n",
    "\n",
    "def joint_loss_function(x, x_reconstructed, adj, adj_pred, alpha=0.5):\n",
    "    feature_reconstruction_loss = F.mse_loss(x_reconstructed, x)\n",
    "    adj_reconstruction_loss = F.binary_cross_entropy(adj_pred, adj)\n",
    "    loss = alpha * feature_reconstruction_loss + (1 - alpha) * adj_reconstruction_loss\n",
    "    return loss, feature_reconstruction_loss, adj_reconstruction_loss\n",
    "\n",
    "def create_adj_matrix(labels):\n",
    "    num_samples = len(labels)\n",
    "    adj_matrix = np.zeros((num_samples, num_samples))\n",
    "    for i in range(num_samples):\n",
    "        for j in range(num_samples):\n",
    "            if labels[i] == labels[j]:\n",
    "                adj_matrix[i, j] = 1\n",
    "    return adj_matrix\n",
    "\n",
    "class CustomDataset(torch.utils.data.Dataset):\n",
    "    def __init__(self, features, labels):\n",
    "        self.features = features\n",
    "        self.labels = labels\n",
    "        self.indices = np.arange(len(features))\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.features)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        return self.features[idx], self.labels[idx], self.indices[idx]\n",
    "\n",
    "train_features, test_features, train_labels, test_labels = train_test_split(\n",
    "    all_features, all_labels.cpu().numpy(), test_size=0.05, random_state=42, stratify=all_labels.cpu().numpy()\n",
    ")\n",
    "\n",
    "train_features = torch.tensor(train_features, dtype=torch.float32).to(device)\n",
    "test_features = torch.tensor(test_features, dtype=torch.float32).to(device)\n",
    "train_labels = torch.tensor(train_labels, dtype=torch.long).to(device)\n",
    "test_labels = torch.tensor(test_labels, dtype=torch.long).to(device)\n",
    "\n",
    "train_adj = create_adj_matrix(train_labels.cpu().numpy())\n",
    "\n",
    "train_adj_np = train_adj.copy()\n",
    "num_edges = int(np.sum(train_adj_np) / 2)\n",
    "num_disrupt = int(1 * num_edges)\n",
    "\n",
    "edges = np.array(np.triu_indices(train_adj_np.shape[0], 1)).T\n",
    "np.random.shuffle(edges)\n",
    "\n",
    "disrupted_edges = edges[:num_disrupt]\n",
    "\n",
    "for edge in disrupted_edges:\n",
    "    train_adj_np[edge[0], edge[1]] = 0\n",
    "    train_adj_np[edge[1], edge[0]] = 0\n",
    "\n",
    "train_adj = torch.tensor(train_adj_np, dtype=torch.float32).to(device)\n",
    "\n",
    "train_dataset = CustomDataset(train_features, train_labels)\n",
    "train_loader_ft = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size_ft, shuffle=True)\n",
    "\n",
    "joint_losses = []\n",
    "feature_reconstruction_losses = []\n",
    "neighborhood_reconstruction_losses = []\n",
    "\n",
    "for epoch in range(num_epochs_ft):\n",
    "    feature_transformer.train()\n",
    "    running_joint_loss = 0.0\n",
    "    running_feature_reconstruction_loss = 0.0\n",
    "    running_neighborhood_reconstruction_loss = 0.0\n",
    "    for data, labels, indices in train_loader_ft:\n",
    "        optimizer_ft.zero_grad()\n",
    "        z, x_reconstructed, adj_pred = feature_transformer(data)\n",
    "        batch_adj = train_adj[indices][:, indices]\n",
    "        joint_loss, feature_reconstruction_loss, neighborhood_reconstruction_loss = joint_loss_function(data, x_reconstructed, batch_adj, adj_pred)\n",
    "        joint_loss.backward()\n",
    "        optimizer_ft.step()\n",
    "        running_joint_loss += joint_loss.item() * data.size(0)\n",
    "        running_feature_reconstruction_loss += feature_reconstruction_loss.item() * data.size(0)\n",
    "        running_neighborhood_reconstruction_loss += neighborhood_reconstruction_loss.item() * data.size(0)\n",
    "    \n",
    "    avg_joint_loss = running_joint_loss / len(train_loader_ft.dataset)\n",
    "    avg_feature_reconstruction_loss = running_feature_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    avg_neighborhood_reconstruction_loss = running_neighborhood_reconstruction_loss / len(train_loader_ft.dataset)\n",
    "    joint_losses.append(avg_joint_loss)\n",
    "    feature_reconstruction_losses.append(avg_feature_reconstruction_loss)\n",
    "    neighborhood_reconstruction_losses.append(avg_neighborhood_reconstruction_loss)\n",
    "    \n",
    "    if (epoch + 1) % 10 == 0:\n",
    "        print(f'Epoch [{epoch + 1}/{num_epochs_ft}], Joint Loss: {avg_joint_loss:.4f}, Feature Reconstruction Loss: {avg_feature_reconstruction_loss:.4f}, Neighborhood Reconstruction Loss: {avg_neighborhood_reconstruction_loss:.4f}')\n",
    "\n",
    "for param in feature_transformer.parameters():\n",
    "    param.requires_grad = False\n",
    "\n",
    "with torch.no_grad():\n",
    "    test_reconstructed_features, _, _ = feature_transformer(test_features)\n",
    "test_reconstructed_features = test_reconstructed_features.cpu().numpy()\n",
    "\n",
    "rf_classifier = RandomForestClassifier(n_estimators=100, random_state=42)\n",
    "rf_classifier.fit(test_reconstructed_features, test_labels.cpu().numpy())\n",
    "\n",
    "with torch.no_grad():\n",
    "    train_reconstructed_features, _, _ = feature_transformer(train_features)\n",
    "train_reconstructed_features = train_reconstructed_features.cpu().numpy()\n",
    "\n",
    "train_pred_labels = rf_classifier.predict(train_reconstructed_features)\n",
    "\n",
    "accuracy = accuracy_score(train_labels.cpu().numpy(), train_pred_labels)\n",
    "print(f'Accuracy: {accuracy:.4f}')\n",
    "print(classification_report(train_labels.cpu().numpy(), train_pred_labels))\n",
    "\n",
    "train_labels_binarized = label_binarize(train_labels.cpu().numpy(), classes=np.unique(train_labels.cpu().numpy()))\n",
    "n_classes = train_labels_binarized.shape[1]\n",
    "\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "\n",
    "for i in range(n_classes):\n",
    "    fpr[i], tpr[i], _ = roc_curve(train_labels_binarized[:, i], rf_classifier.predict_proba(train_reconstructed_features)[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "plt.figure()\n",
    "colors = ['aqua', 'darkorange', 'cornflowerblue', 'green', 'red', 'purple']\n",
    "for i, color in zip(range(n_classes), colors):\n",
    "    plt.plot(fpr[i], tpr[i], color=color, lw=2,\n",
    "             label=f'ROC curve of class {i} (area = {roc_auc[i]:.2f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', lw=2)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curves')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "conf_matrix = confusion_matrix(train_labels.cpu().numpy(), train_pred_labels)\n",
    "plt.figure(figsize=(10, 10))\n",
    "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.show()\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(range(1, num_epochs_ft + 1), joint_losses, label='Joint Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), feature_reconstruction_losses, label='Feature Reconstruction Loss')\n",
    "plt.plot(range(1, num_epochs_ft + 1), neighborhood_reconstruction_losses, label='Neighborhood Reconstruction Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss Curves')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "exam",
   "language": "python",
   "name": "exam"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.19"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "773.333px",
    "left": "98px",
    "top": "0px",
    "width": "339.865px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}