{"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"colab":{"provenance":[]}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import tensorflow as tf\nimport pandas as pd\nfrom sklearn.preprocessing import LabelEncoder\nfrom sklearn.model_selection import train_test_split\nimport numpy as np\nfrom sklearn.metrics import confusion_matrix, classification_report, roc_curve, roc_auc_score\nimport matplotlib.pyplot as plt\nfrom sklearn.metrics import confusion_matrix, classification_report, roc_curve, roc_auc_score, f1_score\nfrom sklearn.metrics import confusion_matrix, classification_report, f1_score, roc_curve, roc_auc_score, accuracy_score\nfrom sklearn.preprocessing import LabelEncoder, StandardScaler\nimport seaborn as sns","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"urpzOOYBcInE","outputId":"664561bd-91cb-4a5b-a24f-b03e7b39d07f","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:04:35.990528Z","iopub.execute_input":"2025-01-17T19:04:35.991025Z","iopub.status.idle":"2025-01-17T19:04:54.616128Z","shell.execute_reply.started":"2025-01-17T19:04:35.990978Z","shell.execute_reply":"2025-01-17T19:04:54.614671Z"}},"outputs":[],"execution_count":1},{"cell_type":"code","source":"# Load the dataset\ndata = pd.read_csv(\"/kaggle/input/patientdata/paitent.csv\")  # Replace with the actual file path","metadata":{"id":"_OSH7Y0reMBn","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:04:56.728399Z","iopub.execute_input":"2025-01-17T19:04:56.728825Z","iopub.status.idle":"2025-01-17T19:04:56.972723Z","shell.execute_reply.started":"2025-01-17T19:04:56.728785Z","shell.execute_reply":"2025-01-17T19:04:56.971529Z"}},"outputs":[],"execution_count":2},{"cell_type":"code","source":"data.head()","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":342},"id":"x6L5DgCUed1a","outputId":"e8594793-820b-4d89-826b-26242fd21617","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:04:59.160608Z","iopub.execute_input":"2025-01-17T19:04:59.161077Z","iopub.status.idle":"2025-01-17T19:04:59.208266Z","shell.execute_reply.started":"2025-01-17T19:04:59.161033Z","shell.execute_reply":"2025-01-17T19:04:59.206893Z"}},"outputs":[{"execution_count":3,"output_type":"execute_result","data":{"text/plain":"   # FP1-F7     F7-T7         T7-P7     P7-O1    FP1-F3     F3-C3     C3-P3  \\\n0 -0.000059  0.000024  1.040000e-05  0.000002 -0.000016  0.000009  0.000010   \n1 -0.000063  0.000025  9.960000e-06  0.000004 -0.000016  0.000009  0.000010   \n2 -0.000065  0.000040  1.950000e-07  0.000006 -0.000014  0.000008  0.000014   \n3 -0.000067  0.000048 -6.450000e-06  0.000003 -0.000015  0.000008  0.000015   \n4 -0.000068  0.000057 -8.010000e-06  0.000001 -0.000013  0.000013  0.000016   \n\n      P3-O1    FP2-F4         F4-C4  ...   T8-P8-0         P8-O2     FZ-CZ  \\\n0 -0.000024 -0.000051 -2.090000e-05  ...  0.000008 -1.660000e-05  0.000028   \n1 -0.000025 -0.000049 -1.620000e-05  ...  0.000028 -8.010000e-06  0.000023   \n2 -0.000026 -0.000045 -1.190000e-05  ...  0.000010 -9.770000e-07  0.000023   \n3 -0.000030 -0.000043 -6.060000e-06  ...  0.000025 -5.860000e-07  0.000018   \n4 -0.000033 -0.000039  5.860000e-07  ...  0.000030 -1.370000e-06  0.000015   \n\n      CZ-PZ         P7-T7        T7-FT9      FT9-FT10       FT10-T8   T8-P8-1  \\\n0 -0.000015 -9.960000e-06  1.270000e-05  1.860000e-05  1.040000e-05  0.000008   \n1 -0.000013 -9.570000e-06  1.470000e-05  1.350000e-05 -6.060000e-06  0.000028   \n2 -0.000012  1.950000e-07  4.490000e-06  1.150000e-05  1.040000e-05  0.000010   \n3 -0.000009  6.840000e-06  1.950000e-07  2.540000e-06 -5.860000e-07  0.000025   \n4 -0.000010  8.400000e-06 -4.100000e-06 -1.950000e-07 -1.950000e-07  0.000030   \n\n        label  \n0  noseizures  \n1  noseizures  \n2  noseizures  \n3  noseizures  \n4  noseizures  \n\n[5 rows x 24 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th># FP1-F7</th>\n      <th>F7-T7</th>\n      <th>T7-P7</th>\n      <th>P7-O1</th>\n      <th>FP1-F3</th>\n      <th>F3-C3</th>\n      <th>C3-P3</th>\n      <th>P3-O1</th>\n      <th>FP2-F4</th>\n      <th>F4-C4</th>\n      <th>...</th>\n      <th>T8-P8-0</th>\n      <th>P8-O2</th>\n      <th>FZ-CZ</th>\n      <th>CZ-PZ</th>\n      <th>P7-T7</th>\n      <th>T7-FT9</th>\n      <th>FT9-FT10</th>\n      <th>FT10-T8</th>\n      <th>T8-P8-1</th>\n      <th>label</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>-0.000059</td>\n      <td>0.000024</td>\n      <td>1.040000e-05</td>\n      <td>0.000002</td>\n      <td>-0.000016</td>\n      <td>0.000009</td>\n      <td>0.000010</td>\n      <td>-0.000024</td>\n      <td>-0.000051</td>\n      <td>-2.090000e-05</td>\n      <td>...</td>\n      <td>0.000008</td>\n      <td>-1.660000e-05</td>\n      <td>0.000028</td>\n      <td>-0.000015</td>\n      <td>-9.960000e-06</td>\n      <td>1.270000e-05</td>\n      <td>1.860000e-05</td>\n      <td>1.040000e-05</td>\n      <td>0.000008</td>\n      <td>noseizures</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>-0.000063</td>\n      <td>0.000025</td>\n      <td>9.960000e-06</td>\n      <td>0.000004</td>\n      <td>-0.000016</td>\n      <td>0.000009</td>\n      <td>0.000010</td>\n      <td>-0.000025</td>\n      <td>-0.000049</td>\n      <td>-1.620000e-05</td>\n      <td>...</td>\n      <td>0.000028</td>\n      <td>-8.010000e-06</td>\n      <td>0.000023</td>\n      <td>-0.000013</td>\n      <td>-9.570000e-06</td>\n      <td>1.470000e-05</td>\n      <td>1.350000e-05</td>\n      <td>-6.060000e-06</td>\n      <td>0.000028</td>\n      <td>noseizures</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>-0.000065</td>\n      <td>0.000040</td>\n      <td>1.950000e-07</td>\n      <td>0.000006</td>\n      <td>-0.000014</td>\n      <td>0.000008</td>\n      <td>0.000014</td>\n      <td>-0.000026</td>\n      <td>-0.000045</td>\n      <td>-1.190000e-05</td>\n      <td>...</td>\n      <td>0.000010</td>\n      <td>-9.770000e-07</td>\n      <td>0.000023</td>\n      <td>-0.000012</td>\n      <td>1.950000e-07</td>\n      <td>4.490000e-06</td>\n      <td>1.150000e-05</td>\n      <td>1.040000e-05</td>\n      <td>0.000010</td>\n      <td>noseizures</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>-0.000067</td>\n      <td>0.000048</td>\n      <td>-6.450000e-06</td>\n      <td>0.000003</td>\n      <td>-0.000015</td>\n      <td>0.000008</td>\n      <td>0.000015</td>\n      <td>-0.000030</td>\n      <td>-0.000043</td>\n      <td>-6.060000e-06</td>\n      <td>...</td>\n      <td>0.000025</td>\n      <td>-5.860000e-07</td>\n      <td>0.000018</td>\n      <td>-0.000009</td>\n      <td>6.840000e-06</td>\n      <td>1.950000e-07</td>\n      <td>2.540000e-06</td>\n      <td>-5.860000e-07</td>\n      <td>0.000025</td>\n      <td>noseizures</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>-0.000068</td>\n      <td>0.000057</td>\n      <td>-8.010000e-06</td>\n      <td>0.000001</td>\n      <td>-0.000013</td>\n      <td>0.000013</td>\n      <td>0.000016</td>\n      <td>-0.000033</td>\n      <td>-0.000039</td>\n      <td>5.860000e-07</td>\n      <td>...</td>\n      <td>0.000030</td>\n      <td>-1.370000e-06</td>\n      <td>0.000015</td>\n      <td>-0.000010</td>\n      <td>8.400000e-06</td>\n      <td>-4.100000e-06</td>\n      <td>-1.950000e-07</td>\n      <td>-1.950000e-07</td>\n      <td>0.000030</td>\n      <td>noseizures</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 24 columns</p>\n</div>"},"metadata":{}}],"execution_count":3},{"cell_type":"code","source":"# Visualize the label column\nlabel_counts = data['label'].value_counts()\nlabel_counts.plot(kind='bar')\nplt.xlabel('Labels')\nplt.ylabel('Counts')\nplt.title('Label Distribution')\nplt.show()","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":532},"id":"yWB7LFF-efG4","outputId":"fb6fda20-8f8e-4713-f009-94466d89a4e7","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:02.601375Z","iopub.execute_input":"2025-01-17T19:05:02.601718Z","iopub.status.idle":"2025-01-17T19:05:03.003693Z","shell.execute_reply.started":"2025-01-17T19:05:02.601692Z","shell.execute_reply":"2025-01-17T19:05:03.002430Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":4},{"cell_type":"code","source":"# Separate features and labels\nX = data.drop(\"label\", axis=1)\ny = data[\"label\"]","metadata":{"id":"7fj-JDNUekeP","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:08.474257Z","iopub.execute_input":"2025-01-17T19:05:08.474685Z","iopub.status.idle":"2025-01-17T19:05:08.489313Z","shell.execute_reply.started":"2025-01-17T19:05:08.474655Z","shell.execute_reply":"2025-01-17T19:05:08.488072Z"}},"outputs":[],"execution_count":5},{"cell_type":"code","source":"# Convert labels to numerical values\nlabel_encoder = LabelEncoder()\ny_encoded = label_encoder.fit_transform(y)","metadata":{"id":"FAJkju9YemPV","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:11.106368Z","iopub.execute_input":"2025-01-17T19:05:11.106763Z","iopub.status.idle":"2025-01-17T19:05:11.119408Z","shell.execute_reply.started":"2025-01-17T19:05:11.106735Z","shell.execute_reply":"2025-01-17T19:05:11.117933Z"}},"outputs":[],"execution_count":6},{"cell_type":"code","source":"# Split the dataset into train and test sets\nX_train, X_test, y_train, y_test = train_test_split(X, y_encoded, test_size=0.2, random_state=42)","metadata":{"id":"xgE8jfq9es_w","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:13.957960Z","iopub.execute_input":"2025-01-17T19:05:13.958367Z","iopub.status.idle":"2025-01-17T19:05:13.976379Z","shell.execute_reply.started":"2025-01-17T19:05:13.958338Z","shell.execute_reply":"2025-01-17T19:05:13.975027Z"}},"outputs":[],"execution_count":7},{"cell_type":"code","source":"# Standardize the feature data\nscaler = StandardScaler()\nX_train_scaled = scaler.fit_transform(X_train)\nX_test_scaled = scaler.transform(X_test)","metadata":{"id":"ztTI_K_bey8G","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:16.559726Z","iopub.execute_input":"2025-01-17T19:05:16.560171Z","iopub.status.idle":"2025-01-17T19:05:16.618500Z","shell.execute_reply.started":"2025-01-17T19:05:16.560138Z","shell.execute_reply":"2025-01-17T19:05:16.616690Z"}},"outputs":[],"execution_count":8},{"cell_type":"code","source":"# Convert the data to TensorFlow tensors\nX_train_tensor = tf.constant(X_train_scaled, dtype=tf.float32)\ny_train_tensor = tf.constant(y_train, dtype=tf.int32)\nX_test_tensor = tf.constant(X_test_scaled, dtype=tf.float32)\ny_test_tensor = tf.constant(y_test, dtype=tf.int32)","metadata":{"id":"gygMp8KYe1q_","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:19.515772Z","iopub.execute_input":"2025-01-17T19:05:19.516231Z","iopub.status.idle":"2025-01-17T19:05:19.571224Z","shell.execute_reply.started":"2025-01-17T19:05:19.516195Z","shell.execute_reply":"2025-01-17T19:05:19.570010Z"}},"outputs":[],"execution_count":9},{"cell_type":"code","source":"# Define the GNN model\nclass GNNModel(tf.keras.Model):\n    def __init__(self, num_classes):\n        super(GNNModel, self).__init__()\n        self.dense1 = tf.keras.layers.Dense(16, activation='relu')\n        self.dense2 = tf.keras.layers.Dense(num_classes, activation='softmax')\n\n    def call(self, inputs):\n        x = self.dense1(inputs)\n        x = self.dense2(x)\n        return x","metadata":{"id":"oAWoftTPe5gg","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:22.264966Z","iopub.execute_input":"2025-01-17T19:05:22.265358Z","iopub.status.idle":"2025-01-17T19:05:22.359027Z","shell.execute_reply.started":"2025-01-17T19:05:22.265329Z","shell.execute_reply":"2025-01-17T19:05:22.357741Z"}},"outputs":[],"execution_count":10},{"cell_type":"code","source":"# Create the GNN model\nnum_classes = len(label_encoder.classes_)\nmodel = GNNModel(num_classes)","metadata":{"id":"903Nmn5Le6ZC","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:25.042494Z","iopub.execute_input":"2025-01-17T19:05:25.042894Z","iopub.status.idle":"2025-01-17T19:05:25.052517Z","shell.execute_reply.started":"2025-01-17T19:05:25.042865Z","shell.execute_reply":"2025-01-17T19:05:25.050960Z"}},"outputs":[],"execution_count":11},{"cell_type":"code","source":"# Define loss function and optimizer\nloss_object = tf.keras.losses.SparseCategoricalCrossentropy()\noptimizer = tf.keras.optimizers.Adam()","metadata":{"id":"56J31UCqfBOP","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:27.767737Z","iopub.execute_input":"2025-01-17T19:05:27.768237Z","iopub.status.idle":"2025-01-17T19:05:27.789713Z","shell.execute_reply.started":"2025-01-17T19:05:27.768201Z","shell.execute_reply":"2025-01-17T19:05:27.788057Z"}},"outputs":[],"execution_count":12},{"cell_type":"code","source":"# Training loop\nepochs = 50\nbatch_size = 32\nsteps_per_epoch = len(X_train_scaled) // batch_size","metadata":{"id":"h6WnaTXSfCHn","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:31.022716Z","iopub.execute_input":"2025-01-17T19:05:31.023167Z","iopub.status.idle":"2025-01-17T19:05:31.028627Z","shell.execute_reply.started":"2025-01-17T19:05:31.023130Z","shell.execute_reply":"2025-01-17T19:05:31.027064Z"}},"outputs":[],"execution_count":13},{"cell_type":"code","source":"train_loss_history = []\ntest_loss_history = []\ntrain_accuracy_history = []\ntest_accuracy_history = []","metadata":{"id":"JLKcMxyjfIIQ","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:34.089128Z","iopub.execute_input":"2025-01-17T19:05:34.089530Z","iopub.status.idle":"2025-01-17T19:05:34.094343Z","shell.execute_reply.started":"2025-01-17T19:05:34.089499Z","shell.execute_reply":"2025-01-17T19:05:34.093027Z"}},"outputs":[],"execution_count":14},{"cell_type":"code","source":"for epoch in range(epochs):\n    for step in range(steps_per_epoch):\n        start = step * batch_size\n        end = (step + 1) * batch_size\n        with tf.GradientTape() as tape:\n            predictions = model(X_train_tensor[start:end])\n            loss = loss_object(y_train_tensor[start:end], predictions)\n\n        gradients = tape.gradient(loss, model.trainable_variables)\n        optimizer.apply_gradients(zip(gradients, model.trainable_variables))\n\n    train_loss = loss_object(y_train_tensor, model(X_train_tensor))\n    test_loss = loss_object(y_test_tensor, model(X_test_tensor))\n    train_loss_history.append(train_loss.numpy())\n    test_loss_history.append(test_loss.numpy())\n\n    y_pred_train = np.argmax(model(X_train_tensor).numpy(), axis=1)\n    y_pred_test = np.argmax(model(X_test_tensor).numpy(), axis=1)\n\n    train_accuracy = accuracy_score(y_train, y_pred_train)\n    test_accuracy = accuracy_score(y_test, y_pred_test)\n\n    train_accuracy_history.append(train_accuracy)\n    test_accuracy_history.append(test_accuracy)\n\n    print(f\"Epoch {epoch + 1}/{epochs}, Train Loss: {train_loss:.4f}, Train Accuracy: {train_accuracy:.4f}, Test Loss: {test_loss:.4f}, Test Accuracy: {test_accuracy:.4f}\")\n","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"dQj1L6nbfJLj","outputId":"34722d85-5978-4630-9353-105a1cfa3a3d","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:05:37.755622Z","iopub.execute_input":"2025-01-17T19:05:37.756030Z","iopub.status.idle":"2025-01-17T19:22:15.912925Z","shell.execute_reply.started":"2025-01-17T19:05:37.755996Z","shell.execute_reply":"2025-01-17T19:22:15.911932Z"}},"outputs":[{"name":"stdout","text":"Epoch 1/50, Train Loss: 0.1680, Train Accuracy: 0.9561, Test Loss: 0.1641, Test Accuracy: 0.9567\nEpoch 2/50, Train Loss: 0.1214, Train Accuracy: 0.9662, Test Loss: 0.1153, Test Accuracy: 0.9694\nEpoch 3/50, Train Loss: 0.1065, Train Accuracy: 0.9708, Test Loss: 0.0998, Test Accuracy: 0.9741\nEpoch 4/50, Train Loss: 0.0971, Train Accuracy: 0.9738, Test Loss: 0.0910, Test Accuracy: 0.9769\nEpoch 5/50, Train Loss: 0.0905, Train Accuracy: 0.9757, Test Loss: 0.0849, Test Accuracy: 0.9780\nEpoch 6/50, Train Loss: 0.0856, Train Accuracy: 0.9768, Test Loss: 0.0804, Test Accuracy: 0.9790\nEpoch 7/50, Train Loss: 0.0817, Train Accuracy: 0.9775, Test Loss: 0.0768, Test Accuracy: 0.9803\nEpoch 8/50, Train Loss: 0.0786, Train Accuracy: 0.9783, Test Loss: 0.0742, Test Accuracy: 0.9803\nEpoch 9/50, Train Loss: 0.0761, Train Accuracy: 0.9787, Test Loss: 0.0722, Test Accuracy: 0.9809\nEpoch 10/50, Train Loss: 0.0739, Train Accuracy: 0.9794, Test Loss: 0.0706, Test Accuracy: 0.9814\nEpoch 11/50, Train Loss: 0.0721, Train Accuracy: 0.9797, Test Loss: 0.0693, Test Accuracy: 0.9814\nEpoch 12/50, Train Loss: 0.0706, Train Accuracy: 0.9798, Test Loss: 0.0681, Test Accuracy: 0.9814\nEpoch 13/50, Train Loss: 0.0693, Train Accuracy: 0.9802, Test Loss: 0.0671, Test Accuracy: 0.9814\nEpoch 14/50, Train Loss: 0.0682, Train Accuracy: 0.9804, Test Loss: 0.0662, Test Accuracy: 0.9812\nEpoch 15/50, Train Loss: 0.0672, Train Accuracy: 0.9806, Test Loss: 0.0654, Test Accuracy: 0.9817\nEpoch 16/50, Train Loss: 0.0662, Train Accuracy: 0.9807, Test Loss: 0.0646, Test Accuracy: 0.9814\nEpoch 17/50, Train Loss: 0.0653, Train Accuracy: 0.9808, Test Loss: 0.0638, Test Accuracy: 0.9821\nEpoch 18/50, Train Loss: 0.0643, Train Accuracy: 0.9809, Test Loss: 0.0631, Test Accuracy: 0.9825\nEpoch 19/50, Train Loss: 0.0634, Train Accuracy: 0.9814, Test Loss: 0.0624, Test Accuracy: 0.9822\nEpoch 20/50, Train Loss: 0.0625, Train Accuracy: 0.9816, Test Loss: 0.0618, Test Accuracy: 0.9825\nEpoch 21/50, Train Loss: 0.0618, Train Accuracy: 0.9818, Test Loss: 0.0613, Test Accuracy: 0.9828\nEpoch 22/50, Train Loss: 0.0611, Train Accuracy: 0.9819, Test Loss: 0.0608, Test Accuracy: 0.9825\nEpoch 23/50, Train Loss: 0.0604, Train Accuracy: 0.9818, Test Loss: 0.0603, Test Accuracy: 0.9822\nEpoch 24/50, Train Loss: 0.0599, Train Accuracy: 0.9820, Test Loss: 0.0600, Test Accuracy: 0.9824\nEpoch 25/50, Train Loss: 0.0594, Train Accuracy: 0.9821, Test Loss: 0.0597, Test Accuracy: 0.9825\nEpoch 26/50, Train Loss: 0.0588, Train Accuracy: 0.9822, Test Loss: 0.0596, Test Accuracy: 0.9822\nEpoch 27/50, Train Loss: 0.0584, Train Accuracy: 0.9821, Test Loss: 0.0593, Test Accuracy: 0.9824\nEpoch 28/50, Train Loss: 0.0578, Train Accuracy: 0.9822, Test Loss: 0.0589, Test Accuracy: 0.9824\nEpoch 29/50, Train Loss: 0.0574, Train Accuracy: 0.9825, Test Loss: 0.0587, Test Accuracy: 0.9827\nEpoch 30/50, Train Loss: 0.0569, Train Accuracy: 0.9828, Test Loss: 0.0583, Test Accuracy: 0.9830\nEpoch 31/50, Train Loss: 0.0565, Train Accuracy: 0.9829, Test Loss: 0.0579, Test Accuracy: 0.9830\nEpoch 32/50, Train Loss: 0.0560, Train Accuracy: 0.9831, Test Loss: 0.0575, Test Accuracy: 0.9831\nEpoch 33/50, Train Loss: 0.0556, Train Accuracy: 0.9832, Test Loss: 0.0571, Test Accuracy: 0.9834\nEpoch 34/50, Train Loss: 0.0553, Train Accuracy: 0.9832, Test Loss: 0.0568, Test Accuracy: 0.9836\nEpoch 35/50, Train Loss: 0.0550, Train Accuracy: 0.9834, Test Loss: 0.0565, Test Accuracy: 0.9834\nEpoch 36/50, Train Loss: 0.0547, Train Accuracy: 0.9833, Test Loss: 0.0563, Test Accuracy: 0.9836\nEpoch 37/50, Train Loss: 0.0543, Train Accuracy: 0.9833, Test Loss: 0.0560, Test Accuracy: 0.9839\nEpoch 38/50, Train Loss: 0.0541, Train Accuracy: 0.9834, Test Loss: 0.0559, Test Accuracy: 0.9836\nEpoch 39/50, Train Loss: 0.0537, Train Accuracy: 0.9835, Test Loss: 0.0555, Test Accuracy: 0.9836\nEpoch 40/50, Train Loss: 0.0534, Train Accuracy: 0.9837, Test Loss: 0.0554, Test Accuracy: 0.9837\nEpoch 41/50, Train Loss: 0.0532, Train Accuracy: 0.9838, Test Loss: 0.0552, Test Accuracy: 0.9834\nEpoch 42/50, Train Loss: 0.0529, Train Accuracy: 0.9838, Test Loss: 0.0550, Test Accuracy: 0.9833\nEpoch 43/50, Train Loss: 0.0526, Train Accuracy: 0.9839, Test Loss: 0.0548, Test Accuracy: 0.9836\nEpoch 44/50, Train Loss: 0.0523, Train Accuracy: 0.9841, Test Loss: 0.0546, Test Accuracy: 0.9836\nEpoch 45/50, Train Loss: 0.0521, Train Accuracy: 0.9842, Test Loss: 0.0544, Test Accuracy: 0.9836\nEpoch 46/50, Train Loss: 0.0517, Train Accuracy: 0.9844, Test Loss: 0.0542, Test Accuracy: 0.9831\nEpoch 47/50, Train Loss: 0.0514, Train Accuracy: 0.9845, Test Loss: 0.0541, Test Accuracy: 0.9831\nEpoch 48/50, Train Loss: 0.0511, Train Accuracy: 0.9845, Test Loss: 0.0540, Test Accuracy: 0.9833\nEpoch 49/50, Train Loss: 0.0509, Train Accuracy: 0.9845, Test Loss: 0.0538, Test Accuracy: 0.9833\nEpoch 50/50, Train Loss: 0.0505, Train Accuracy: 0.9846, Test Loss: 0.0535, Test Accuracy: 0.9836\n","output_type":"stream"}],"execution_count":15},{"cell_type":"code","source":"# Evaluation\ny_pred_train_prob = model(X_train_tensor).numpy()\ny_pred_test_prob = model(X_test_tensor).numpy()\n\n# Convert probabilities to class indices\ny_pred_train = np.argmax(y_pred_train_prob, axis=1)\ny_pred_test = np.argmax(y_pred_test_prob, axis=1)\n","metadata":{"id":"A0AfPi7Snbd2","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:23.652102Z","iopub.execute_input":"2025-01-17T19:22:23.652465Z","iopub.status.idle":"2025-01-17T19:22:23.670148Z","shell.execute_reply.started":"2025-01-17T19:22:23.652438Z","shell.execute_reply":"2025-01-17T19:22:23.668802Z"}},"outputs":[],"execution_count":16},{"cell_type":"code","source":"# F1 Score\nf1_train = f1_score(y_train, y_pred_train, average='weighted')\nf1_test = f1_score(y_test, y_pred_test, average='weighted')\n\n# Accuracy\ntrain_accuracy = np.mean(np.equal(y_train, y_pred_train))\ntest_accuracy = np.mean(np.equal(y_test, y_pred_test))\n\n# Other metrics\nprint(\"Train F1 Score:\", f1_train)\nprint(\"Test F1 Score:\", f1_test)\n\n# Accuracy\nprint(\"Train Accuracy:\", train_accuracy)\nprint(\"Test Accuracy:\", test_accuracy)","metadata":{"id":"eJbmwL9qne2l","colab":{"base_uri":"https://localhost:8080/"},"outputId":"8e9faaa9-64d7-4b0f-eaa3-aff00d4103be","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:26.517831Z","iopub.execute_input":"2025-01-17T19:22:26.518226Z","iopub.status.idle":"2025-01-17T19:22:26.545336Z","shell.execute_reply.started":"2025-01-17T19:22:26.518198Z","shell.execute_reply":"2025-01-17T19:22:26.543991Z"}},"outputs":[{"name":"stdout","text":"Train F1 Score: 0.9846469576854522\nTest F1 Score: 0.983575829981423\nTrain Accuracy: 0.9846483927052121\nTest Accuracy: 0.9835774522858411\n","output_type":"stream"}],"execution_count":17},{"cell_type":"code","source":"# Confusion Matrix\ncm_train = confusion_matrix(y_train, y_pred_train)\ncm_test = confusion_matrix(y_test, y_pred_test)\n\n# Plot Accuracy Curve\nplt.figure(figsize=(10, 6))\nplt.plot(range(epochs), train_accuracy_history, label='Train Accuracy')\nplt.plot(range(epochs), test_accuracy_history, label='Test Accuracy')\nplt.xlabel('Epoch')\nplt.ylabel('Accuracy')\nplt.title('Accuracy Curve')\nplt.legend()\nplt.show()\n\n","metadata":{"id":"zUFwxWB1ngFN","colab":{"base_uri":"https://localhost:8080/","height":564},"outputId":"7a6c8488-db0e-45cf-8e2f-7a912324ad85","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:31.211806Z","iopub.execute_input":"2025-01-17T19:22:31.212301Z","iopub.status.idle":"2025-01-17T19:22:31.507434Z","shell.execute_reply.started":"2025-01-17T19:22:31.212265Z","shell.execute_reply":"2025-01-17T19:22:31.506065Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x600 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":18},{"cell_type":"code","source":"# Loss Curve\nplt.figure(figsize=(10, 6))\nplt.plot(train_loss_history, label='Train Loss')\nplt.plot(test_loss_history, label='Test Loss')\nplt.xlabel('Epoch')\nplt.ylabel('Loss')\nplt.title('Loss Curve')\nplt.legend()\nplt.show()","metadata":{"id":"2O0B9qJ8nmOP","colab":{"base_uri":"https://localhost:8080/","height":564},"outputId":"8d6540b0-1e78-4500-ddc4-5b445da2cd38","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:35.594773Z","iopub.execute_input":"2025-01-17T19:22:35.595197Z","iopub.status.idle":"2025-01-17T19:22:35.842448Z","shell.execute_reply.started":"2025-01-17T19:22:35.595165Z","shell.execute_reply":"2025-01-17T19:22:35.840990Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x600 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":19},{"cell_type":"code","source":"# Classification Report\nreport_train = classification_report(y_train, y_pred_train, target_names=label_encoder.classes_)\nreport_test = classification_report(y_test, y_pred_test, target_names=label_encoder.classes_)","metadata":{"id":"JPfuijeznpe1","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:38.677537Z","iopub.execute_input":"2025-01-17T19:22:38.677970Z","iopub.status.idle":"2025-01-17T19:22:38.755576Z","shell.execute_reply.started":"2025-01-17T19:22:38.677895Z","shell.execute_reply":"2025-01-17T19:22:38.754511Z"}},"outputs":[],"execution_count":20},{"cell_type":"code","source":"# Print Confusion Matrix and Classification Report\nprint(\"Train Classification Report:\")\nprint(report_train)","metadata":{"id":"KD6iQPzgnuYG","colab":{"base_uri":"https://localhost:8080/"},"outputId":"fbed9f40-9c5c-4a4b-adb3-4a115297b8d2","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:41.555719Z","iopub.execute_input":"2025-01-17T19:22:41.556111Z","iopub.status.idle":"2025-01-17T19:22:41.563433Z","shell.execute_reply.started":"2025-01-17T19:22:41.556079Z","shell.execute_reply":"2025-01-17T19:22:41.562044Z"}},"outputs":[{"name":"stdout","text":"Train Classification Report:\n              precision    recall  f1-score   support\n\n  noseizures       0.97      1.00      0.98     13492\n    seizures       1.00      0.97      0.98     13541\n\n    accuracy                           0.98     27033\n   macro avg       0.98      0.98      0.98     27033\nweighted avg       0.98      0.98      0.98     27033\n\n","output_type":"stream"}],"execution_count":21},{"cell_type":"code","source":"print(\"Test Classification Report:\")\nprint(report_test)","metadata":{"id":"aDLuvqbFnvK3","colab":{"base_uri":"https://localhost:8080/"},"outputId":"2774bd56-7cc7-4786-bd6a-3c78b9e92b06","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:45.170033Z","iopub.execute_input":"2025-01-17T19:22:45.170396Z","iopub.status.idle":"2025-01-17T19:22:45.176542Z","shell.execute_reply.started":"2025-01-17T19:22:45.170367Z","shell.execute_reply":"2025-01-17T19:22:45.175134Z"}},"outputs":[{"name":"stdout","text":"Test Classification Report:\n              precision    recall  f1-score   support\n\n  noseizures       0.98      0.99      0.98      3404\n    seizures       0.99      0.98      0.98      3355\n\n    accuracy                           0.98      6759\n   macro avg       0.98      0.98      0.98      6759\nweighted avg       0.98      0.98      0.98      6759\n\n","output_type":"stream"}],"execution_count":22},{"cell_type":"code","source":"# ROC Curve\n# Compute ROC AUC score\n#roc_auc_train = roc_auc_score(y_train, y_pred_train_prob, multi_class='ovr')\n#oc_auc_test = roc_auc_score(y_test, y_pred_test_prob, multi_class='ovr')\n\nfrom sklearn.metrics import roc_auc_score\nimport numpy as np\n\n# Assuming y_pred_train_prob and y_pred_test_prob are the predicted probabilities for each class\n# You may need to extract the probabilities for each class depending on your model's output\n# For example, if y_pred_train_prob is of shape (27443, 2), you may need to use y_pred_train_prob[:, 1] for class 1\n\n# Calculate ROC AUC score for each class\nroc_auc_train = []\nroc_auc_test = []\n\nfor i in range(num_classes):  # Replace num_classes with the actual number of classes\n    roc_auc_train.append(roc_auc_score(y_train == i, y_pred_train_prob[:, i]))\n    roc_auc_test.append(roc_auc_score(y_test == i, y_pred_test_prob[:, i]))\n\n# Print or use roc_auc_train and roc_auc_test as needed\nprint(\"ROC AUC Train:\", np.mean(roc_auc_train))\nprint(\"ROC AUC Test:\", np.mean(roc_auc_test))\n","metadata":{"id":"HV-O7OeCnxgD","colab":{"base_uri":"https://localhost:8080/"},"outputId":"96af17f3-e93a-4f28-83fe-bc72978857e5","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:48.785805Z","iopub.execute_input":"2025-01-17T19:22:48.786298Z","iopub.status.idle":"2025-01-17T19:22:48.825487Z","shell.execute_reply.started":"2025-01-17T19:22:48.786260Z","shell.execute_reply":"2025-01-17T19:22:48.824310Z"}},"outputs":[{"name":"stdout","text":"ROC AUC Train: 0.997136184584013\nROC AUC Test: 0.9968035545102545\n","output_type":"stream"}],"execution_count":23},{"cell_type":"code","source":"print(y_pred_train_prob.shape)","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"nZyfQdk626rN","outputId":"1e242774-d148-4704-8e38-e6a648685b59","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:51.917520Z","iopub.execute_input":"2025-01-17T19:22:51.917867Z","iopub.status.idle":"2025-01-17T19:22:51.923585Z","shell.execute_reply.started":"2025-01-17T19:22:51.917838Z","shell.execute_reply":"2025-01-17T19:22:51.922312Z"}},"outputs":[{"name":"stdout","text":"(27033, 2)\n","output_type":"stream"}],"execution_count":24},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nfrom sklearn.metrics import roc_curve, auc, roc_auc_score\nimport numpy as np\n\n# Assuming y_pred_train_prob and y_pred_test_prob are the predicted probabilities for each class\n# Extract the probabilities for each class\ny_pred_train_prob = model(X_train_tensor).numpy()\ny_pred_test_prob = model(X_test_tensor).numpy()\n\n# Convert probabilities to class indices\ny_pred_train = np.argmax(y_pred_train_prob, axis=1)\ny_pred_test = np.argmax(y_pred_test_prob, axis=1)\n\n# Calculate ROC AUC score for each class\nroc_auc_train = []\nroc_auc_test = []\n\nfor i in range(num_classes):  # Replace num_classes with the actual number of classes\n    roc_auc_train.append(roc_auc_score(y_train == i, y_pred_train_prob[:, i]))\n    roc_auc_test.append(roc_auc_score(y_test == i, y_pred_test_prob[:, i]))\n\n# Print or use roc_auc_train and roc_auc_test as needed\nprint(\"ROC AUC Train:\", np.mean(roc_auc_train))\nprint(\"ROC AUC Test:\", np.mean(roc_auc_test))\n\n# Plot ROC curves\nplt.figure(figsize=(10, 6))\n\nfor i in range(num_classes):\n    fpr_train, tpr_train, _ = roc_curve(y_train == i, y_pred_train_prob[:, i])\n    roc_auc_train_class = auc(fpr_train, tpr_train)\n    plt.plot(fpr_train, tpr_train, label=f'Train ROC Curve (Class {i + 1}, AUC = {roc_auc_train_class:.2f})')\n\n    fpr_test, tpr_test, _ = roc_curve(y_test == i, y_pred_test_prob[:, i])\n    roc_auc_test_class = auc(fpr_test, tpr_test)\n    plt.plot(fpr_test, tpr_test, label=f'Test ROC Curve (Class {i + 1}, AUC = {roc_auc_test_class:.2f})')\n\nplt.plot([0, 1], [0, 1], 'k--')\nplt.xlabel('False Positive Rate')\nplt.ylabel('True Positive Rate')\nplt.title('Receiver Operating Characteristic (ROC) Curve for each class')\nplt.legend()\nplt.show()\n","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":600},"id":"wY7mD5Tu8A-v","outputId":"f9306e9d-7efd-4b79-fd3d-8f5dd65473ac","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:54.321625Z","iopub.execute_input":"2025-01-17T19:22:54.322066Z","iopub.status.idle":"2025-01-17T19:22:54.677500Z","shell.execute_reply.started":"2025-01-17T19:22:54.322036Z","shell.execute_reply":"2025-01-17T19:22:54.676094Z"}},"outputs":[{"name":"stdout","text":"ROC AUC Train: 0.997136184584013\nROC AUC Test: 0.9968035545102545\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x600 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":25},{"cell_type":"code","source":"# Plot Confusion Matrix Heatmap for Train Set\nplt.figure(figsize=(15, 10))\nsns.heatmap(cm_train, annot=True, fmt=\"d\", cmap=\"Blues\", xticklabels=label_encoder.classes_, yticklabels=label_encoder.classes_)\nplt.xlabel('Predicted')\nplt.ylabel('True')\nplt.title('Confusion Matrix (Train Set)')\nplt.show()","metadata":{"id":"nF1jzlAsn3aS","colab":{"base_uri":"https://localhost:8080/","height":872},"outputId":"c9cfbf8e-31e3-4a28-bc11-aa4793c7cb36","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:22:59.348817Z","iopub.execute_input":"2025-01-17T19:22:59.349254Z","iopub.status.idle":"2025-01-17T19:22:59.647009Z","shell.execute_reply.started":"2025-01-17T19:22:59.349219Z","shell.execute_reply":"2025-01-17T19:22:59.645689Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1500x1000 with 2 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":26},{"cell_type":"code","source":"# Plot Confusion Matrix Heatmap for Test Set\nplt.figure(figsize=(15, 10))\nsns.heatmap(cm_test, annot=True, fmt=\"d\", cmap=\"Blues\", xticklabels=label_encoder.classes_, yticklabels=label_encoder.classes_)\nplt.xlabel('Predicted')\nplt.ylabel('True')\nplt.title('Confusion Matrix (Test Set)')\nplt.show()","metadata":{"id":"nRr-8Dbin_fp","colab":{"base_uri":"https://localhost:8080/","height":872},"outputId":"e23f03e2-9276-4949-fffa-076f1e14f03a","trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:23:03.641710Z","iopub.execute_input":"2025-01-17T19:23:03.642129Z","iopub.status.idle":"2025-01-17T19:23:03.908365Z","shell.execute_reply.started":"2025-01-17T19:23:03.642101Z","shell.execute_reply":"2025-01-17T19:23:03.907049Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1500x1000 with 2 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":27},{"cell_type":"code","source":"import time\nimport numpy as np\n\n# Start time for training predictions\nstart_train_time = time.time()\ny_pred_train_prob = model(X_train_tensor).numpy()\nend_train_time = time.time()\n\n# Convert probabilities to class indices for training predictions\ny_pred_train_prob = np.argmax(y_pred_train_prob, axis=1)\n\n# Calculate time taken for training predictions\ntrain_time = end_train_time - start_train_time\nprint(f\"Time taken for training predictions: {train_time:.4f} seconds\")\n\n# Start time for testing predictions\nstart_test_time = time.time()\ny_pred_test_prob = model(X_test_tensor).numpy()\nend_test_time = time.time()\n\n# Convert probabilities to class indices for testing predictions\ny_pred_test_prob = np.argmax(y_pred_test_prob, axis=1)\n\n# Calculate time taken for testing predictions\ntest_time = end_test_time - start_test_time\nprint(f\"Time taken for testing predictions: {test_time:.4f} seconds\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:23:07.955009Z","iopub.execute_input":"2025-01-17T19:23:07.955467Z","iopub.status.idle":"2025-01-17T19:23:07.978042Z","shell.execute_reply.started":"2025-01-17T19:23:07.955432Z","shell.execute_reply":"2025-01-17T19:23:07.976583Z"}},"outputs":[{"name":"stdout","text":"Time taken for training predictions: 0.0091 seconds\nTime taken for testing predictions: 0.0054 seconds\n","output_type":"stream"}],"execution_count":28},{"cell_type":"code","source":"# Parameters for Early Stopping\npatience = 5\nbest_loss = float('inf')\nwait = 0\nvalidation_split = 0.1  # Use a small validation set for faster evaluation\n\n# Split a portion of training data for validation\nval_size = int(len(X_train_tensor) * validation_split)\nX_val_tensor = X_train_tensor[:val_size]\ny_val_tensor = y_train_tensor[:val_size]\nX_train_tensor = X_train_tensor[val_size:]\ny_train_tensor = y_train_tensor[val_size:]\n\nfor epoch in range(epochs):\n    for step in range(steps_per_epoch):\n        # Mini-batch gradient updates\n        start = step * batch_size\n        end = (step + 1) * batch_size\n        X_batch = X_train_tensor[start:end]\n        y_batch = y_train_tensor[start:end]\n\n        with tf.GradientTape() as tape:\n            predictions = model(X_batch)\n            loss = loss_object(y_batch, predictions)\n\n        gradients = tape.gradient(loss, model.trainable_variables)\n        optimizer.apply_gradients(zip(gradients, model.trainable_variables))\n\n    # Evaluate on validation data\n    val_loss = loss_object(y_val_tensor, model(X_val_tensor))\n    val_predictions = np.argmax(model(X_val_tensor).numpy(), axis=1)\n    val_accuracy = accuracy_score(y_val_tensor.numpy(), val_predictions)\n\n    train_loss_history.append(loss.numpy())\n    test_loss_history.append(val_loss.numpy())\n    train_accuracy_history.append(None)  # Skipping full train accuracy for speed\n    test_accuracy_history.append(val_accuracy)\n\n    print(f\"Epoch {epoch + 1}/{epochs}, Val Loss: {val_loss:.4f}, Val Accuracy: {val_accuracy:.4f}\")\n\n    # Early stopping check\n    if val_loss < best_loss:\n        best_loss = val_loss\n        wait = 0  # Reset patience counter\n    else:\n        wait += 1\n        print(f\"No improvement in validation loss for {wait} epoch(s).\")\n        if wait >= patience:\n            print(\"Early stopping triggered. Stopping training.\")\n            break","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:24:19.198874Z","iopub.execute_input":"2025-01-17T19:24:19.199398Z","iopub.status.idle":"2025-01-17T19:26:23.878189Z","shell.execute_reply.started":"2025-01-17T19:24:19.199362Z","shell.execute_reply":"2025-01-17T19:26:23.877044Z"}},"outputs":[{"name":"stdout","text":"Epoch 1/50, Val Loss: 0.0425, Val Accuracy: 0.9874\nEpoch 2/50, Val Loss: 0.0428, Val Accuracy: 0.9874\nNo improvement in validation loss for 1 epoch(s).\nEpoch 3/50, Val Loss: 0.0430, Val Accuracy: 0.9874\nNo improvement in validation loss for 2 epoch(s).\nEpoch 4/50, Val Loss: 0.0432, Val Accuracy: 0.9871\nNo improvement in validation loss for 3 epoch(s).\nEpoch 5/50, Val Loss: 0.0434, Val Accuracy: 0.9867\nNo improvement in validation loss for 4 epoch(s).\nEpoch 6/50, Val Loss: 0.0434, Val Accuracy: 0.9871\nNo improvement in validation loss for 5 epoch(s).\nEarly stopping triggered. Stopping training.\n","output_type":"stream"}],"execution_count":29},{"cell_type":"code","source":"import time\nimport numpy as np\n\n# Start time for training predictions\nstart_train_time = time.time()\ny_pred_train_prob = model(X_train_tensor).numpy()\nend_train_time = time.time()\n\n# Convert probabilities to class indices for training predictions\ny_pred_train_prob = np.argmax(y_pred_train_prob, axis=1)\n\n# Calculate time taken for training predictions\ntrain_time = end_train_time - start_train_time\nprint(f\"Time taken for training predictions: {train_time:.4f} seconds\")\n\n# Start time for testing predictions\nstart_test_time = time.time()\ny_pred_test_prob = model(X_test_tensor).numpy()\nend_test_time = time.time()\n\n# Convert probabilities to class indices for testing predictions\ny_pred_test_prob = np.argmax(y_pred_test_prob, axis=1)\n\n# Calculate time taken for testing predictions\ntest_time = end_test_time - start_test_time\nprint(f\"Time taken for testing predictions: {test_time:.4f} seconds\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-01-17T19:26:23.879572Z","iopub.execute_input":"2025-01-17T19:26:23.880058Z","iopub.status.idle":"2025-01-17T19:26:23.901772Z","shell.execute_reply.started":"2025-01-17T19:26:23.880010Z","shell.execute_reply":"2025-01-17T19:26:23.900387Z"}},"outputs":[{"name":"stdout","text":"Time taken for training predictions: 0.0081 seconds\nTime taken for testing predictions: 0.0054 seconds\n","output_type":"stream"}],"execution_count":30}]}