Update app.py

app.py

@@ -371,21 +371,28 @@ def plot_confusion_matrix(y_true, y_pred, title):
     # Return the saved image
     return Image.open(f"{title}.png")
     
-def identical_train_test_split(output_emb, output_raw, labels, percentage):
+def identical_train_test_split(output_emb, output_raw, labels, train_percentage):
     N = output_emb.shape[0]  # Get the total number of samples
     
     # Generate the indices for shuffling and splitting
     indices = torch.randperm(N)  # Randomly shuffle the indices
     
-    # Calculate the split index
-    #percentage = percentage_values_los[percentage_idx]
-    split_index = int(N * percentage/100)
-    print(f'Training Size: {split_index}')
+    # Calculate the split index for test (10% of the data)
+    test_split_index = int(N * 0.10)
+    
+    # Test indices (first 10% of the data)
+    test_indices = indices[:test_split_index]
     
-    # Split indices into train and test
-    train_indices = indices[:split_index]  # First 80% for training
-    test_indices = indices[split_index:]   # Remaining 20% for testing
+    # Remaining indices for training
+    remaining_indices = indices[test_split_index:]
     
+    # Calculate the split index for training from the remaining 90%
+    train_split_index = int(len(remaining_indices) * train_percentage / 100)
+    print(f'Training Size: {train_split_index} out of remaining {len(remaining_indices)}')
+    
+    # Train indices (based on the provided percentage from the remaining 90%)
+    train_indices = remaining_indices[:train_split_index]
+
     # Select the same indices from both output_emb and output_raw
     train_emb = output_emb[train_indices]
     test_emb = output_emb[test_indices]
@@ -446,7 +453,7 @@ def process_hdf5_file(uploaded_file, percentage):
 
         # Step 5: Load the HDF5 file and extract the channels and labels
         with h5py.File(uploaded_file.name, 'r') as f:
-            channels = np.array(f['channels']).astype(np.float32)  # Assuming 'channels' dataset in the HDF5 file
+            channels = torch.tensor(f['channels']).float() #astype(np.float32)  # Assuming 'channels' dataset in the HDF5 file
             labels = torch.tensor(f['labels']).long()  # Assuming 'labels' dataset in the HDF5 file
         print(f"Loaded dataset with {channels.shape[0]} samples.")