Update evaluation.py

evaluation.py

@@ -186,13 +186,13 @@ def calculate_metrics(question, q_dataset, response, docs, time_taken):
     else:
         predicted_metrics["rmse"] = "Invalid RMSE calculation"
         '''
-    if isinstance(predicted_metrics, (int, float)) and isinstance(ground_truth_metrics, (int, float)):
+    if isinstance(predicted_metrics_rmse, (int, float)) and isinstance(ground_truth_metrics, (int, float)):
         rmse_value = compute_rmse([predicted_metrics_rmse], [ground_truth_metrics])
-        predicted_metrics["rmse"] = rmse_value  # Adding RMSE to metrics
+        predicted_metrics_rmse["rmse"] = rmse_value  # Adding RMSE to metrics
     else:
-        predicted_metrics["rmse"] = "Invalid RMSE calculation"
+        predicted_metrics_rmse["rmse"] = "Invalid RMSE calculation"
         
-    return predicted_metrics
+    return predicted_metrics_rmse
 
 ''' def retrieve_ground_truths(question, dataset):
     for split_name, instances in dataset.items():