Update app.py

app.py

@@ -72,8 +72,9 @@ def predict():
     global model
     print("Received predict request at /predict endpoint")
     
-    # Check model file existence
-    model_path = 'stroke_prediction_model.pkl'
+    # Check and resolve model file path
+    model_path = os.path.join(os.getcwd(), 'stroke_prediction_model.pkl')
+    print(f"Checking model file at: {model_path}")
     if not os.path.exists(model_path):
         print(f"Model file not found at {model_path}")
         return jsonify({'success': False, 'error': f'Model file not found at {model_path}'}), 500
@@ -97,17 +98,17 @@ def predict():
         input_data = data['input']
         print(f"Processed input data: {input_data}")
 
-        # Validate and convert input data
+        # Validate and convert input data with fallback
         features = {
-            'age': float(input_data.get('age', 0)),
+            'age': float(input_data.get('age', 0)) if input_data.get('age') else 0,
             'gender': input_data.get('gender', 'Unknown'),
             'ever_married': input_data.get('ever_married', 'No'),
             'residence_type': input_data.get('residence_type', 'Unknown'),
             'work_type': input_data.get('work_type', 'Unknown'),
-            'hypertension': int(input_data.get('hypertension', 0)),
-            'heart_disease': int(input_data.get('heart_disease', 0)),
-            'avg_glucose_level': float(input_data.get('avg_glucose_level', 0)),
-            'bmi': float(input_data.get('bmi', 0)),
+            'hypertension': int(input_data.get('hypertension', 0)) if input_data.get('hypertension') else 0,
+            'heart_disease': int(input_data.get('heart_disease', 0)) if input_data.get('heart_disease') else 0,
+            'avg_glucose_level': float(input_data.get('avg_glucose_level', 0)) if input_data.get('avg_glucose_level') else 0,
+            'bmi': float(input_data.get('bmi', 0)) if input_data.get('bmi') else 0,
             'smoking_status': input_data.get('smoking_status', 'Unknown')
         }
         print(f"Converted features: {features}")
@@ -137,9 +138,13 @@ def predict():
         print(f"Aligned DataFrame columns: {df.columns.tolist()}")
 
         # Make prediction
-        probability = model.predict_proba(df)[0][1] * 100
-        risk_prediction = "Stroke Risk" if probability > 50 else "No Stroke Risk"
-        print(f"Prediction result: probability={probability}%, prediction={risk_prediction}")
+        try:
+            probability = model.predict_proba(df)[0][1] * 100
+            risk_prediction = "Stroke Risk" if probability > 50 else "No Stroke Risk"
+            print(f"Prediction result: probability={probability}%, prediction={risk_prediction}")
+        except Exception as pred_error:
+            print(f"Prediction error: {str(pred_error)}")
+            return jsonify({'success': False, 'error': f'Prediction failed: {str(pred_error)}'}), 500
 
         # Determine contributing factors
         contributing_factors = {
@@ -150,15 +155,14 @@ def predict():
         }
         print(f"Contributing factors: {contributing_factors}")
 
-        # Temporarily comment out database write to isolate the issue
-        """
+        # Attempt to store in database with fallback
         try:
             conn = sqlite3.connect('/tmp/submissions.db')
             c = conn.cursor()
             c.execute('''
                 INSERT INTO submissions (
                     timestamp, age, gender, ever_married, residence_type, work_type,
-                    hypertension, heart_disease, avg_glucose_level, bmi, smuggling_status,
+                    hypertension, heart_disease, avg_glucose_level, bmi, smoking_status,
                     probability, risk_level
                 ) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
             ''', (
@@ -179,10 +183,9 @@ def predict():
             conn.commit()
             print("Data successfully written to database")
         except Exception as db_error:
-            print(f"Database write error: {str(db_error)}")
+            print(f"Database write error (non-critical): {str(db_error)}")
         finally:
             conn.close()
-        """
 
         # Return prediction result
         return jsonify({
@@ -192,7 +195,7 @@ def predict():
             'contributingFactors': contributing_factors
         })
     except Exception as e:
-        print(f"Error during prediction: {str(e)}")
+        print(f"Unexpected error during prediction: {str(e)}")
         return jsonify({'success': False, 'error': str(e)}), 500
 
 @app.route('/admin/report', methods=['GET'])