Update app.py

app.py

@@ -18,36 +18,38 @@ def predict_fall(movement_activity, location, day_of_week, hour_of_day, minute_o
         # Initialize all features to 0
         data = {feat: 0 for feat in feature_names}
 
-        # One-hot encode categoricals
+        # Set one-hot encoded categorical variables
         data[f'Movement Activity_{movement_activity}'] = 1
         data[f'Location_{location}'] = 1
         data[f'day_of_week_{day_of_week}'] = 1
 
-        # Set numeric features
+        # Set numerical features
         data['hour_of_day'] = hour_of_day
         data['minute_of_day'] = minute_of_day
         data['time_since_last_event'] = time_since_last_event
 
-        # Build DataFrame with columns in correct order
+        # Build DataFrame in the exact order of feature_names (very important)
         input_df = pd.DataFrame([data], columns=feature_names)
 
-        # Scale all features (must match scaler's expected input columns)
-        scaled_array = scaler.transform(input_df)
-        input_df.loc[:, feature_names] = scaled_array
+        # Pass columns in correct order to scaler
+        scaled_features = scaler.transform(input_df[feature_names])
 
-        # Predict probability of fall
-        prob_fall = model.predict_proba(input_df)[:, 1][0]
+        # Replace numerical values with scaled values (or all if scaler scales all)
+        input_df.loc[:, feature_names] = scaled_features
 
+        # Predict probability
+        pred_proba = model.predict_proba(input_df)[0, 1]
         threshold = 0.4
-        prediction = "Fall Detected" if prob_fall >= threshold else "No Fall"
+        pred_label = "Fall Detected" if pred_proba >= threshold else "No Fall"
 
-        return f"Prediction: {prediction}\nFall Probability: {prob_fall:.2f}"
+        return f"Prediction: {pred_label}\nFall Probability: {pred_proba:.2f}"
 
     except Exception as e:
         import traceback
         traceback.print_exc()
         return f"Error: {str(e)}. Check logs for details."
 
+
 # Gradio interface setup
 iface = gr.Interface(
     fn=predict_fall,