lint

src/llamafactory/train/sft/metric.py

@@ -65,6 +65,7 @@ class ComputeAccuracy:
     r"""
     Computes accuracy and supports `batch_eval_metrics`.
     """
+
     def _dump(self) -> Optional[Dict[str, float]]:
         result = None
         if hasattr(self, "score_dict"):