updated app age input handeling

app.py

@@ -40,7 +40,8 @@ def initialize_app():
     
     # 2. FIT ENCODERS (For Feature Consistency)
     print("🔤 Fitting Label Encoders...")
-    cat_cols = ['category', 'gender', 'day_of_week', 'age']
+    # UPDATED: 'age' removed from here, treated as numeric
+    cat_cols = ['category', 'gender', 'day_of_week'] 
     for c in cat_cols:
         if c in knowledge_df.columns:
             le = LabelEncoder()
@@ -101,27 +102,36 @@ def predict_and_optimize(user_input, duration, hour, day_of_week, category, foll
         def safe_encode(col, val):
             le = ENCODERS.get(col)
             if le:
-                # If value not seen, default to first class (mostly 'Unknown' or similar if handled, else 0)
+                # If value not seen, default to first class 
                 if val in le.classes_:
                     return le.transform([val])[0]
                 else:
-                    return 0 # Fallback
+                    return 0 
             return 0
 
         cat_encoded = safe_encode('category', category)
         gender_encoded = safe_encode('gender', gender)
         day_encoded = safe_encode('day_of_week', day_of_week)
-        age_encoded = safe_encode('age', age)
+        
+        # FIX: Map Age String to Numeric
+        age_map = {
+            "18-24": 21.0, 
+            "25-34": 30.0, 
+            "35-44": 40.0, 
+            "45+": 50.0
+        }
+        age_numeric = age_map.get(str(age), 25.0) # Default to 25 if unknown
         
     except Exception as e:
         return f"Encoding Error: {str(e)}", "", "", "", ""
 
     # --- 2. INITIAL PREDICTION ---
     # Feature Order MUST match model-prep.py:
-    # Embeddings + [duration, hour, followers, age, category_enc, gender_enc, day_enc]
+    # Embeddings + [duration, hour, followers, age_numeric, category_enc, gender_enc, day_enc]
     text_vec = ST_MODEL.encode([user_input], convert_to_numpy=True)
     
-    meta_vec = np.array([[duration, hour, followers, age_encoded, cat_encoded, gender_encoded, day_encoded]])
+    # Construct metadata vector
+    meta_vec = np.array([[duration, hour, followers, age_numeric, cat_encoded, gender_encoded, day_encoded]])
     
     feat_vec = np.hstack((text_vec, meta_vec))
     
@@ -176,14 +186,13 @@ def predict_and_optimize(user_input, duration, hour, day_of_week, category, foll
     [Upload duration]
     """
 
-    
     try:
         response = llm.generate_content(prompt)
         improved_idea = response.text.strip()
         
         # --- 5. RE-SCORING ---
         new_text_vec = ST_MODEL.encode([improved_idea], convert_to_numpy=True)
-        # Assuming metadata stays same for optimization check
+        # Using same metadata for the new prediction
         new_feat_vec = np.hstack((new_text_vec, meta_vec))
         
         new_log = MODEL.predict(new_feat_vec)[0]
@@ -201,13 +210,13 @@ def predict_and_optimize(user_input, duration, hour, day_of_week, category, foll
 # --- GRADIO UI ---
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     gr.Markdown("# 🚀 Viral Content Optimizer")
-    gr.Markdown("Enter your video idea and stats to predict views and get AI-powered optimizations.")
+    gr.Markdown("Enter your video idea and stats to predict views and get AI-powered optimizations based on 2025 trends.")
     
     with gr.Row():
         with gr.Column(scale=1):
             input_text = gr.Textbox(
-                label="Video Description", 
-                placeholder="POV: When you forget to mute your mic...",
+                label="Your Video Description", 
+                placeholder="e.g., POV: trying the new grimace shake #viral",
                 lines=3
             )