Update evaluation.py

evaluation.py

@@ -176,7 +176,7 @@ def calculate_metrics(question, q_dataset, response, docs, time_taken):
         "response_time": time_taken
     }
 
-    store_rmse(question, predicted_metrics, ground_truth_metrics)
+    store_rmse(question, predicted_metrics_rmse, ground_truth_metrics)
      # Now, make sure the values passed to RMSE calculation are numeric
     #predicted_completeness = predicted_metrics['completeness']
     
@@ -190,11 +190,11 @@ def calculate_metrics(question, q_dataset, response, docs, time_taken):
         '''
     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"] = rmse_value  # Adding RMSE to metrics
+        predicted_metrics["rmse"] = rmse_value  # Adding RMSE to metrics
     else:
-        predicted_metrics_rmse["rmse"] = "Invalid RMSE calculation"
+        predicted_metrics["rmse"] = "Invalid RMSE calculation"
         
-    return predicted_metrics_rmse
+    return predicted_metrics
 
 ''' def retrieve_ground_truths(question, dataset):
     for split_name, instances in dataset.items():