Update app.py

app.py

@@ -153,10 +153,20 @@ def explain(batch: Optional[BatchInputData] = None, limit: int = 100):
 # =====================================================
 
 @app.post("/metrics")
-def metrics(batch: Optional[BatchInputData] = None, y_true: Optional[List[int]] = None, limit: int = 100):
-    """Compute ROC AUC and threshold analysis, using input or NoCoDB test data."""
+def metrics(batch: Optional[BatchInputData] = None, limit: int = 100):
+    """
+    Compute ROC AUC and threshold analysis using input or NoCoDB test data.
+    Assumes the target column 'y' is boolean (True/False).
+    """
+    # Defaults in case something fails
+    roc_auc = None
+    pr_auc = None
+    thresholds = []
+    precision = []
+    recall = []
+
     try:
-        # Use provided data or fallback to test data from NoCoDB
+        # Fetch data from batch or NoCoDB
         if batch:
             X = pd.DataFrame([item.dict() for item in batch.data])
             source = "client batch"
@@ -166,15 +176,13 @@ def metrics(batch: Optional[BatchInputData] = None, y_true: Optional[List[int]]
 
         print(f"[DEBUG] Metrics called using {source} | shape={X.shape}")
 
-        # Identify and separate target variable
-        possible_targets = ["y", "target", "label"]
-        target_col = next((c for c in possible_targets if c in X.columns), None)
+        # Ensure target 'y' exists
+        if "y" not in X.columns:
+            return {"error": "No target column 'y' found in dataset."}
 
-        if target_col is not None:
-            y_true = X[target_col].astype(int).tolist()
-            X = X.drop(columns=[target_col])
-        elif y_true is None:
-            return {"error": "No target (y_true) provided or found in dataset."}
+        # Convert boolean target to integer
+        y_true = X["y"].astype(int).tolist()
+        X = X.drop(columns=["y"])
 
         # Drop ID if exists
         if "Id" in X.columns:
@@ -183,6 +191,7 @@ def metrics(batch: Optional[BatchInputData] = None, y_true: Optional[List[int]]
         # Predict probabilities
         y_prob = model.predict_proba(X)[:, 1]
 
+        # Compute metrics
         roc_auc = roc_auc_score(y_true, y_prob)
         precision, recall, thresholds = precision_recall_curve(y_true, y_prob)
         pr_auc = auc(recall, precision)
@@ -203,6 +212,7 @@ def metrics(batch: Optional[BatchInputData] = None, y_true: Optional[List[int]]
         print(traceback.format_exc())
         return {"error": str(e), "trace": traceback.format_exc()}
 
+
     
 @app.get("/coefficients")
 def coefficients():