Update inference/predictor.py

inference/predictor.py

@@ -7,11 +7,13 @@ from src.preprocessing import clean_and_engineer
 
 MODEL_DIR = "models"
 
+
 class CreditRiskPredictor:
     def __init__(self):
         self.model_path = get_latest_file(MODEL_DIR, "credit_model")
         self.scaler_path = get_latest_file(MODEL_DIR, "scaler")
         self.columns_path = get_latest_file(MODEL_DIR, "columns")
+
         self.model = joblib.load(self.model_path)
         self.scaler = joblib.load(self.scaler_path)
         self.columns = joblib.load(self.columns_path)
@@ -47,33 +49,33 @@ class CreditRiskPredictor:
         X_scaled = self.scaler.transform(df)
 
         # -------------------------------
-        # 6. UPDATED: Use sklearn's predict_proba
+        # 6. Scorecard logic
         # -------------------------------
         probability, credit_score, rating = self._calculate_scorecard_output(X_scaled)
 
         return probability, credit_score, rating
 
     def _calculate_scorecard_output(self, X_scaled, base_score=300, scale_length=600):
-    try:
-        print("DEBUG: Model type:", type(self.model))
-        print("DEBUG: Has predict_proba:", hasattr(self.model, "predict_proba"))
-        print("DEBUG: X_scaled shape:", X_scaled.shape)
-        print("DEBUG: Any NaN:", np.isnan(X_scaled).any())
-        print("DEBUG: Any Inf:", np.isinf(X_scaled).any())
+        try:
+            print("DEBUG: Model type:", type(self.model))
+            print("DEBUG: Has predict_proba:", hasattr(self.model, "predict_proba"))
+            print("DEBUG: X_scaled shape:", X_scaled.shape)
+            print("DEBUG: Any NaN:", np.isnan(X_scaled).any())
+            print("DEBUG: Any Inf:", np.isinf(X_scaled).any())
 
-        default_probability = self.model.predict_proba(X_scaled)[:, 1]
+            default_probability = self.model.predict_proba(X_scaled)[:, 1]
 
-        non_default_probability = 1 - default_probability
-        credit_score = base_score + non_default_probability * scale_length
-        credit_score = int(credit_score[0])
-        rating = self._get_rating(credit_score)
+            non_default_probability = 1 - default_probability
+            credit_score = base_score + non_default_probability * scale_length
+            credit_score = int(credit_score[0])
 
-        return float(default_probability[0]), credit_score, rating
+            rating = self._get_rating(credit_score)
 
-    except Exception as e:
-        print("🔥 PREDICTION ERROR:", repr(e))
-        raise
+            return float(default_probability[0]), credit_score, rating
 
+        except Exception as e:
+            print("🔥 PREDICTION ERROR:", repr(e))
+            raise
 
     def _get_rating(self, score):
         if 300 <= score < 500:
@@ -85,4 +87,4 @@ class CreditRiskPredictor:
         elif 750 <= score <= 900:
             return 'Excellent'
         else:
-            return 'Undefined'
+            return 'Undefined'