Update app.py

app.py

@@ -158,6 +158,13 @@ def predict_crime_level(crime_felony, crime_misd, crime_viol, sr311_total, dob_p
     """Predicts crime level based on input features."""
     print(f"DEBUG: predict_crime_level called with inputs: {crime_felony}, {crime_misd}, {crime_viol}, {sr311_total}, {dob_permits_total}")
     
+    # Define emoji mapping
+    emoji_map = {
+        "Low": "🟢",
+        "Medium": "🟡", 
+        "High": "🔴"
+    }
+    
     if model is None:
         # Create a dummy prediction based on simple logic when model is not available
         total_crime = crime_felony + crime_misd + crime_viol
@@ -183,9 +190,12 @@ def predict_crime_level(crime_felony, crime_misd, crime_viol, sr311_total, dob_p
         if dob_permits_total > 25:
             # High construction activity might indicate development/change
             confidence["Medium"] = min(0.8, confidence.get("Medium", 0) + 0.2)
-            
-        print(f"DEBUG: Dummy prediction result: {prediction}, confidence: {confidence}")
-        return f"Predicted Crime Level: {prediction} (using fallback model)", confidence
+        
+        # Add emojis to confidence labels
+        confidence_with_emojis = {f"{level} {emoji_map[level]}": prob for level, prob in confidence.items()}
+        
+        print(f"DEBUG: Dummy prediction result: {prediction}, confidence: {confidence_with_emojis}")
+        return emoji_map[prediction], confidence_with_emojis
 
     try:
         # The model expects many more features than we have available in this interface
@@ -222,19 +232,25 @@ def predict_crime_level(crime_felony, crime_misd, crime_viol, sr311_total, dob_p
         # Determine final prediction
         prediction = max(confidence.items(), key=lambda x: x[1])[0]
         
-        print(f"DEBUG: Enhanced fallback prediction result: {prediction}, confidence: {confidence}")
-        return f"Predicted Crime Level: {prediction} (enhanced model)", confidence
+        # Add emojis to confidence labels
+        confidence_with_emojis = {f"{level} {emoji_map[level]}": prob for level, prob in confidence.items()}
+        
+        print(f"DEBUG: Enhanced fallback prediction result: {prediction}, confidence: {confidence_with_emojis}")
+        return emoji_map[prediction], confidence_with_emojis
         
     except Exception as e:
         print(f"DEBUG: Error in model prediction: {e}")
         # Even if there's an error, provide a basic prediction
         total_crime = crime_felony + crime_misd + crime_viol
         if total_crime <= 20:
-            return "Predicted Crime Level: Low (basic fallback)", {"Low": 0.6, "Medium": 0.3, "High": 0.1}
+            confidence = {"Low 🟢": 0.6, "Medium 🟡": 0.3, "High 🔴": 0.1}
+            return "🟢", confidence
         elif total_crime <= 50:
-            return "Predicted Crime Level: Medium (basic fallback)", {"Low": 0.2, "Medium": 0.6, "High": 0.2}
+            confidence = {"Low 🟢": 0.2, "Medium 🟡": 0.6, "High 🔴": 0.2}
+            return "🟡", confidence
         else:
-            return "Predicted Crime Level: High (basic fallback)", {"Low": 0.1, "Medium": 0.3, "High": 0.6}
+            confidence = {"Low 🟢": 0.1, "Medium 🟡": 0.3, "High 🔴": 0.6}
+            return "🔴", confidence
 
 def forecast_time_series(geoid, selected_metric):
     """Forecasts crime for a specific GEOID."""