Update app.py

app.py

@@ -15,7 +15,7 @@ from sklearn.metrics import f1_score
 import seaborn as sns
 
 #################### BEAM PREDICTION #########################}
-def beam_prediction_task(data_percentage, task_complexity, user_mode="light"):
+def beam_prediction_task(data_percentage, task_complexity, theme):
     # Folder naming convention based on input_type, data_percentage, and task_complexity
     raw_folder = f"images/raw_{data_percentage/100:.1f}_{task_complexity}"
     embeddings_folder = f"images/embedding_{data_percentage/100:.1f}_{task_complexity}"
@@ -24,7 +24,7 @@ def beam_prediction_task(data_percentage, task_complexity, user_mode="light"):
     raw_cm = compute_average_confusion_matrix(raw_folder)
     if raw_cm is not None:
         raw_cm_path = os.path.join(raw_folder, "confusion_matrix_raw.png")
-        plot_confusion_matrix_beamPred(raw_cm, classes=np.arange(raw_cm.shape[0]), title=f"Raw Confusion Matrix\n({data_percentage}% data, {task_complexity} beams)", save_path=raw_cm_path, user_mode=user_mode)
+        plot_confusion_matrix_beamPred(raw_cm, classes=np.arange(raw_cm.shape[0]), title=f"Raw Confusion Matrix\n({data_percentage}% data, {task_complexity} beams)", save_path=raw_cm_path, theme)
         raw_img = Image.open(raw_cm_path)
     else:
         raw_img = None
@@ -33,7 +33,7 @@ def beam_prediction_task(data_percentage, task_complexity, user_mode="light"):
     embeddings_cm = compute_average_confusion_matrix(embeddings_folder)
     if embeddings_cm is not None:
         embeddings_cm_path = os.path.join(embeddings_folder, "confusion_matrix_embeddings.png")
-        plot_confusion_matrix_beamPred(embeddings_cm, classes=np.arange(embeddings_cm.shape[0]), title=f"Embeddings Confusion Matrix\n({data_percentage}% data, {task_complexity} beams)", save_path=embeddings_cm_path, user_mode=user_mode)
+        plot_confusion_matrix_beamPred(embeddings_cm, classes=np.arange(embeddings_cm.shape[0]), title=f"Embeddings Confusion Matrix\n({data_percentage}% data, {task_complexity} beams)", save_path=embeddings_cm_path, theme)
         embeddings_img = Image.open(embeddings_cm_path)
     else:
         embeddings_img = None
@@ -59,12 +59,12 @@ def compute_f1_score(cm):
     f1 = np.nan_to_num(f1)  # Replace NaN with 0
     return np.mean(f1)  # Return the mean F1-score across all classes
 
-def plot_confusion_matrix_beamPred(cm, classes, title, save_path, user_mode="dark"):
+def plot_confusion_matrix_beamPred(cm, classes, title, save_path, theme):
     # Compute the average F1-score
     avg_f1 = compute_f1_score(cm)
 
     # Choose the color scheme based on the user's mode
-    if user_mode == 'dark':
+    if theme == 'dark':
         plt.style.use('dark_background')  # Use dark mode styling
         text_color = 'white'
         cmap = 'cividis'  # Dark-mode-friendly colormap
@@ -474,6 +474,21 @@ def process_hdf5_file(uploaded_file, percentage):
         sys.stdout = sys.__stdout__  # Reset print statements
 
 ######################## Define the Gradio interface ###############################
+# JavaScript to detect dark/light mode and pass it to Gradio as 'light' or 'dark'
+detect_theme_js = """
+<script>
+    function detectTheme() {
+        if (window.matchMedia && window.matchMedia('(prefers-color-scheme: dark)').matches) {
+            return 'dark';
+        } else {
+            return 'light';
+        }
+    }
+    document.getElementById('theme_output').value = detectTheme();
+    document.getElementById('theme_output').dispatchEvent(new Event('change'));
+</script>
+"""
+
 with gr.Blocks(css="""
     .slider-container {
         display: inline-block;
@@ -534,6 +549,10 @@ with gr.Blocks(css="""
     # Tab for Beam Prediction Task
     with gr.Tab("Beam Prediction Task"):
         #gr.Markdown("### Beam Prediction Task")
+        gr.HTML(detect_theme_js)
+    
+        # Hidden textbox to store the detected theme
+        theme_output = gr.Textbox(label="Detected Theme", elem_id="theme_output", visible=False)
 
         # Explanation section with creative spacing and minimal design
         gr.Markdown("""
@@ -563,8 +582,9 @@ with gr.Blocks(css="""
             embeddings_img_bp = gr.Image(label="Embeddings", type="pil", width=300, height=500)
 
         # Update the confusion matrices whenever sliders change
-        data_percentage_slider.change(fn=beam_prediction_task, inputs=[data_percentage_slider, task_complexity_dropdown], outputs=[raw_img_bp, embeddings_img_bp])
-        task_complexity_dropdown.change(fn=beam_prediction_task, inputs=[data_percentage_slider, task_complexity_dropdown], outputs=[raw_img_bp, embeddings_img_bp])
+        data_percentage_slider.change(fn=beam_prediction_task, inputs=[data_percentage_slider, task_complexity_dropdown, theme_output], outputs=[raw_img_bp, embeddings_img_bp])
+        task_complexity_dropdown.change(fn=beam_prediction_task, inputs=[data_percentage_slider, task_complexity_dropdown, theme_output], outputs=[raw_img_bp, embeddings_img_bp])
+        theme_output.change(fn=beam_prediction_task, inputs=[data_percentage_slider, task_complexity_dropdown, theme_output], outputs=[raw_img_bp, embeddings_img_bp])
 
         # Add a conclusion section at the bottom
         gr.Markdown("""