Update eval_utils.py

eval_utils.py

@@ -89,7 +89,7 @@ def evaluate_cjpe(gold_data, pred_data):
     }
     print("Explanability for ILDC Expert:", explanation_result)
     #return {**prediction_result, **explanation_result}
-    return {"mF1": f1*100, "ROUGE-L": rouge_score*100, "BLEU": bleu_score*100}
+    return {"cjpe-prediction": {"mF1": f1*100}, "cjpe-explanation": {"ROUGE-L": rouge_score*100, "BLEU": bleu_score*100}}
 
 def span2bio(txt, roles):
     roles = sorted(roles, key = lambda x:x['start'])