Update evaluation.py

evaluation.py

@@ -156,16 +156,7 @@ def calculate_metrics(question, q_dataset, response, docs, time_taken):
     ground_truth_completeness = compute_cosine_similarity(response, ground_truth_answer)
     
     # Predicted metrics
-    predicted_metrics = {
-        "RAG_model_response": response,
-        "ground_truth": ground_truth_answer,
-        "context_relevance": context_relevance(question, docs),
-        "context_utilization": context_utilization(response, docs),
-        "completeness": compute_cosine_similarity(response, ground_truth_answer), #completeness(response, ground_truth_answer),
-        "adherence": adherence(response, docs),
-        "response_time": time_taken,
-        "ground truth completeness": ground_truth_completeness
-    }
+   
 
         # Predicted metrics
     predicted_metrics_rmse = {
@@ -218,7 +209,19 @@ def calculate_metrics(question, q_dataset, response, docs, time_taken):
         predicted_metrics_rmse["rmse"] = rmse_value  # Adding RMSE to metrics
     else:
         predicted_metrics_rmse["rmse"] = "Invalid RMSE calculation"
+        
      '''   
+     predicted_metrics = {
+        "RAG_model_response": response,
+        "ground_truth": ground_truth_answer,
+        "context_relevance": context_relevance(question, docs),
+        "context_utilization": context_utilization(response, docs),
+        "completeness": compute_cosine_similarity(response, ground_truth_answer), #completeness(response, ground_truth_answer),
+        "adherence": adherence(response, docs),
+        "response_time": time_taken,
+        "ground truth completeness": ground_truth_completeness,
+        "rmse": overall_rmse
+    }
     return predicted_metrics
 
 ''' def retrieve_ground_truths(question, dataset):