Update app.py

app.py

@@ -12,16 +12,8 @@ from sklearn.metrics import confusion_matrix
 import matplotlib.pyplot as plt
 import pandas as pd
 
-# Paths to the predefined images folder
-RAW_PATH = os.path.join("images", "raw")
-EMBEDDINGS_PATH = os.path.join("images", "embeddings")
-
-# Specific values for percentage of data for training
-percentage_values = (np.arange(9) + 1)*10
-
-
-
 
+#################### BEAM PREDICTION #########################}
 def beam_prediction_task(data_percentage, task_complexity):
     # Folder naming convention based on input_type, data_percentage, and task_complexity
     raw_folder = f"images/raw_{data_percentage/100:.1f}_{task_complexity}"
@@ -92,40 +84,6 @@ def plot_confusion_matrix_beamPred(cm, classes, title, save_path):
     plt.savefig(save_path)
     plt.close()
 
-
-#def plot_confusion_matrix_beamPred(cm, classes, title, save_path):
-#    plt.figure(figsize=(8, 6))
-#    plt.imshow(cm, interpolation='nearest', cmap='coolwarm')
-#    plt.title(title)
-#    plt.colorbar()
-#    tick_marks = np.arange(len(classes))
-#    plt.xticks(tick_marks, classes, rotation=45)
-#    plt.yticks(tick_marks, classes)
-#
-#    plt.tight_layout()
-#    plt.ylabel('True label')
-#    plt.xlabel('Predicted label')
-#    plt.savefig(save_path)
-#    plt.close()
-    
-# Function to compute the average confusion matrix across CSV files in a folder
-#def compute_average_confusion_matrix(folder):
-#    confusion_matrices = []
-#   for file in os.listdir(folder):
-#        if file.endswith(".csv"):
-#            data = pd.read_csv(os.path.join(folder, file))
-#            y_true = data["Target"]
-#            y_pred = data["Top-1 Prediction"]
-#            num_labels = len(np.unique(y_true))
-#            cm = confusion_matrix(y_true, y_pred, labels=np.arange(num_labels))
-#            confusion_matrices.append(cm)
-#
-#    if confusion_matrices:
-#        avg_cm = np.mean(confusion_matrices, axis=0)
-#        return avg_cm
-#    else:
-#        return None
-
 def compute_average_confusion_matrix(folder):
     confusion_matrices = []
     max_num_labels = 0
@@ -162,10 +120,99 @@ def compute_average_confusion_matrix(folder):
     else:
         return None
 
+########################## LOS/NLOS CLASSIFICATION #############################3
 
 
+# Paths to the predefined images folder
+LOS_PATH = "images_LoS"
 
+# Define the percentage values
+percentage_values_los = np.linspace(0.1, 1, 20) * 100  # 20 percentage values
 
+# Function to compute confusion matrix and plot it
+def plot_confusion_matrix_from_csv(csv_file_path, title, save_path):
+    # Load CSV file
+    data = pd.read_csv(csv_file_path)
+    
+    # Extract ground truth and predictions
+    y_true = data['ground-truth']
+    y_pred = data['predicted']
+    
+    # Compute confusion matrix
+    cm = confusion_matrix(y_true, y_pred)
+    
+    # Plot the confusion matrix
+    plt.figure(figsize=(5, 5))
+    plt.imshow(cm, interpolation='nearest', cmap='Blues')
+    plt.title(title)
+    plt.colorbar()
+    plt.xticks([0, 1], labels=['Class 0', 'Class 1'])
+    plt.yticks([0, 1], labels=['Class 0', 'Class 1'])
+    
+    # Annotate the confusion matrix
+    thresh = cm.max() / 2
+    for i in range(cm.shape[0]):
+        for j in range(cm.shape[1]):
+            plt.text(j, i, format(cm[i, j], 'd'), ha="center", va="center", 
+                     color="white" if cm[i, j] > thresh else "black")
+    
+    plt.ylabel('True label')
+    plt.xlabel('Predicted label')
+    plt.tight_layout()
+    
+    # Save the plot as an image
+    plt.savefig(save_path)
+    plt.close()
+    
+    # Return the saved image
+    return Image.open(save_path)
+
+# Function to load confusion matrix based on percentage and input_type
+def display_confusion_matrices_los(percentage_idx):
+    percentage = percentage_values_los[percentage_idx]
+    
+    # Construct folder names
+    raw_folder = os.path.join(LOS_PATH, f"raw_{percentage/100:.3f}_los_noTraining")
+    embeddings_folder = os.path.join(LOS_PATH, f"embedding_{percentage/100:.3f}_los_noTraining")
+    
+    # Process raw confusion matrix
+    raw_csv_file = os.path.join(raw_folder, "confusion_matrix.csv")
+    raw_cm_img_path = os.path.join(raw_folder, "confusion_matrix_raw.png")
+    raw_img = plot_confusion_matrix_from_csv(raw_csv_file, 
+                                             f"Raw Confusion Matrix ({percentage:.1f}% data)", 
+                                             raw_cm_img_path)
+
+    # Process embeddings confusion matrix
+    embeddings_csv_file = os.path.join(embeddings_folder, "confusion_matrix.csv")
+    embeddings_cm_img_path = os.path.join(embeddings_folder, "confusion_matrix_embeddings.png")
+    embeddings_img = plot_confusion_matrix_from_csv(embeddings_csv_file, 
+                                                    f"Embeddings Confusion Matrix ({percentage:.1f}% data)", 
+                                                    embeddings_cm_img_path)
+
+    return raw_img, embeddings_img
+
+# Main function to handle user choice
+def handle_user_choice(choice, percentage_idx=None, uploaded_file=None):
+    if choice == "Use Predefined Data":
+        return display_confusion_matrices_los(percentage_idx)
+    elif choice == "Upload Dataset":
+        if uploaded_file is not None:
+            return process_hdf5_file(uploaded_file, percentage_idx)
+        else:
+            return "Please upload a dataset", "Please upload a dataset"
+    else:
+        return "Invalid choice", "Invalid choice"
+
+
+
+
+
+
+
+
+
+
+        
 
 # Custom class to capture print output
 class PrintCapture(io.StringIO):
@@ -410,7 +457,7 @@ def process_hdf5_file(uploaded_file, percentage_idx):
         os.chdir(original_dir)
         sys.stdout = sys.__stdout__  # Reset print statements
 
-# Define the Gradio interface
+######################## Define the Gradio interface ###############################
 with gr.Blocks(css="""
     .slider-container {
         display: inline-block;
@@ -439,17 +486,35 @@ with gr.Blocks(css="""
     # Separate Tab for LoS/NLoS Classification Task
     with gr.Tab("LoS/NLoS Classification Task"):
         gr.Markdown("### LoS/NLoS Classification Task")
-        file_input = gr.File(label="Upload HDF5 Dataset", file_types=[".h5"])
 
-        with gr.Row():
-            percentage_dropdown_los = gr.Dropdown(choices=[0, 1, 2, 3, 4, 5, 6, 7, 8], value=0, label="Percentage of Data for Training")
+        # Radio button for user choice: predefined data or upload dataset
+        choice_radio = gr.Radio(choices=["Use Predefined Data", "Upload Dataset"], label="Choose how to proceed", value="Use Predefined Data")
+        
+        # Dropdown for selecting percentage for predefined data
+        percentage_dropdown_los = gr.Dropdown(choices=list(range(20)), value=0, label="Percentage of Data for Training")
+
+        # File uploader for dataset (only visible if user chooses to upload a dataset)
+        file_input = gr.File(label="Upload HDF5 Dataset", file_types=[".h5"], visible=False)
+
+        # Confusion matrices display
         with gr.Row():
             raw_img_los = gr.Image(label="Raw Channels", type="pil", width=300, height=300)
             embeddings_img_los = gr.Image(label="Embeddings", type="pil", width=300, height=300)
             output_textbox = gr.Textbox(label="Console Output", lines=10)
 
-        # Placeholder for LoS/NLoS classification function (already implemented in your previous code)
-        file_input.change(fn=los_nlos_classification, inputs=[file_input, percentage_dropdown_los], outputs=[raw_img_los, embeddings_img_los, output_textbox])
+        # Update the file uploader visibility based on user choice
+        def toggle_file_input(choice):
+            return gr.update(visible=(choice == "Upload Dataset"))
+        
+        choice_radio.change(fn=toggle_file_input, inputs=[choice_radio], outputs=file_input)
+
+        # When user makes a choice, update the display
+        choice_radio.change(fn=handle_user_choice, inputs=[choice_radio, percentage_dropdown_los, file_input], 
+                            outputs=[raw_img_los, embeddings_img_los, output_textbox])
+
+        # When percentage slider changes (for predefined data)
+        percentage_dropdown_los.change(fn=handle_user_choice, inputs=[choice_radio, percentage_dropdown_los, file_input], 
+                                       outputs=[raw_img_los, embeddings_img_los, output_textbox])
 
 # Launch the app
 if __name__ == "__main__":