Update eval_utils.py

eval_utils.py

@@ -239,8 +239,8 @@ def evaluate_pcr(gold_data, pred_data):
         print(f"Micro-F1@{k} on IL-PCR test set:", f1)
 
     max_f1 = max(f1_scores)
-    index_max = f1_scores.index(max_f1) + 1
-    return {"muF1@K": f"{max_f1*100:.2f}@{index_max}"}
+    # index_max = f1_scores.index(max_f1) + 1
+    return {"muF1@K": f"{max_f1*100:.2f}"} #@{index_max}"}
 
 
 def evaluate_summ(gold_data, pred_data):
@@ -319,9 +319,9 @@ def create_output_json(evaluation_results):
         "L-NER": {"strict mF1": evaluation_results["lner"]["strict mF1"]},
         "RR": {"mF1": evaluation_results["rr"]["mF1"]},
         "CJPE": {
-            "mF1": evaluation_results["cjpe"]["mF1"],
-            "ROUGE-L": evaluation_results["cjpe"]["ROUGE-L"],
-            "BLEU": evaluation_results["cjpe"]["BLEU"],
+            "mF1": evaluation_results["cjpe-prediction"]["mF1"],
+            "ROUGE-L": evaluation_results["cjpe-explanation"]["ROUGE-L"],
+            "BLEU": evaluation_results["cjpe-explanation"]["BLEU"],
         },
         "BAIL": {"mF1": evaluation_results["bail"]["mF1"]},
         "LSI": {"mF1": evaluation_results["lsi"]["mF1"]},