Update app.py

app.py

@@ -7,7 +7,7 @@ from PIL import Image
 MODEL_PATH = "model/best_food_model.keras"
 model = tf.keras.models.load_model(MODEL_PATH)
 
-# 2. Labels & Database (Make sure to include the full dictionary from previous turns)
+# 2. Labels & Database
 LABELS = [
     'apple_pie', 'baby_back_ribs', 'baklava', 'beef_carpaccio', 'beef_tartare', 'beet_salad', 'beignets', 
     'bibimbap', 'bread_pudding', 'breakfast_burrito', 'bruschetta', 'caesar_salad', 'cannoli', 'caprese_salad', 
@@ -24,6 +24,7 @@ LABELS = [
     'shrimp_and_grits', 'spaghetti_bolognese', 'spaghetti_carbonara', 'spring_rolls', 'steak', 'strawberry_shortcake', 
     'sushi', 'tacos', 'takoyaki', 'tiramisu', 'tuna_tartare', 'waffles'
 ]
+
 NUTRITION_DB = {
     'apple_pie': {'cal': 237, 'protein': 1.9, 'carbs': 34.0, 'fat': 11.0},
     'baby_back_ribs': {'cal': 292, 'protein': 17.0, 'carbs': 0.0, 'fat': 24.0},
@@ -128,10 +129,11 @@ NUTRITION_DB = {
     'waffles': {'cal': 291, 'protein': 8.0, 'carbs': 33.0, 'fat': 14.0}
 }
 
+# 3. Prediction Function
 def predict_nutrition(img):
     if img is None:
         return None, "Please upload an image."
-    
+        
     # Preprocessing
     img = Image.fromarray(img).resize((224, 224))
     img_array = tf.keras.preprocessing.image.img_to_array(img)
@@ -165,7 +167,7 @@ def predict_nutrition(img):
     
     return confidences, nutri_markdown
 
-# 3. Enhanced Gradio UI
+# 4. Enhanced Gradio UI
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     gr.Markdown("# 🍎 Food-101 Deep Learning Classifier")
     gr.Markdown("Identify food items and see their nutritional breakdown instantly.")
@@ -176,11 +178,9 @@ with gr.Blocks(theme=gr.themes.Soft()) as demo:
             submit_btn = gr.Button("Analyze Meal", variant="primary")
             
         with gr.Column(scale=1):
-            # Output 1: Top 3 Confidence Chart
             output_chart = gr.Label(num_top_classes=3, label="Top 3 Predictions")
-            # Output 2: Nutrition Table
             output_nutri = gr.Markdown(label="Nutrition Breakdown")
-    
+            
     submit_btn.click(
         fn=predict_nutrition, 
         inputs=input_img,