bugfix

utils.py

@@ -76,6 +76,8 @@ def parse_eval_dataset(dataset:DatasetDict) -> List[BenchmarkItem]:
         benchmark_items.append(BenchmarkItem(
             task_id=row['task_id'],
             question=row['question'],
-            evaluate=EvaluateData(items=[EvaluateItem(**item) for item in row['evaluate']['items']])
+            evaluate=EvaluateData(items=[EvaluateItem(**item) for item in row['evaluate']['items']]),
+            category=row['category'],
+            level=row['level']
         ))
     return benchmark_items