using [None] to replace None when obj is not retrieved

paddlex/inference/pipelines/pp_shitu_v2/pipeline.py

@@ -106,9 +106,9 @@ class ShiTuV2Pipeline(BasePipeline):
         single_img_res = {"input_path": input_data, "input_img": raw_img, "boxes": []}
         for i, obj in enumerate(det_res["boxes"]):
             rec_scores = rec_res["score"][i]
-            rec_scores = rec_scores.tolist() if rec_scores is not None else [None]
+            rec_scores = rec_scores if rec_scores is not None else [None]
             labels = rec_res["label"][i]
-            labels = labels.tolist() if labels is not None else [None]
+            labels = labels if labels is not None else [None]
             single_img_res["boxes"].append(
                 {
                     "labels": labels,