PP-DocLayoutV2模型批量推理报错

Traceback (most recent call last):
  File "D:\code\rapid_doc\model\layout\rapid_layout_self\inference_engine\onnxruntime\main.py", line 75, in __call__
    return self.session.run(self.get_output_names(), input_dict)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\code\.venv\Lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 287, in run
    return self._sess.run(output_names, input_feed, run_options)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Non-zero status code returned while running ScatterND node. Name:'ScatterND.1' Status Message: updates tensor should have shape equal to indices.shape[:-1] + data.shape[indices.shape[-1]:]. updates shape: {1}, indices shape: {2}, data shape: {302,2}

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "D:\code\rapid_doc\rapid_doc_202601021446.py", line 10, in <module>
    all_results = model.__call__(img_contents=img_contents, batch_size=128)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\code\rapid_doc\model\layout\rapid_layout_self\main.py", line 50, in __call__
    results = self.model_handler(batch_imgs)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\code\rapid_doc\model\layout\rapid_layout_self\model_handler\main.py", line 34, in __call__
    return self.model_processors(img_list)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\code\rapid_doc\model\layout\rapid_layout_self\model_handler\pp_doclayout\main.py", line 54, in __call__
    batch_preds = self.session(img_inputs, scale_factor_inputs)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\code\rapid_doc\model\layout\rapid_layout_self\inference_engine\onnxruntime\main.py", line 78, in __call__
    raise ONNXRuntimeError(error_info) from e
rapid_doc.model.layout.rapid_layout_self.inference_engine.onnxruntime.main.ONNXRuntimeError: Traceback (most recent call last):
  File "D:\code\rapid_doc\model\layout\rapid_layout_self\inference_engine\onnxruntime\main.py", line 75, in __call__
    return self.session.run(self.get_output_names(), input_dict)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\code\.venv\Lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 287, in run
    return self._sess.run(output_names, input_feed, run_options)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Non-zero status code returned while running ScatterND node. Name:'ScatterND.1' Status Message: updates tensor should have shape equal to indices.shape[:-1] + data.shape[indices.shape[-1]:]. updates shape: {1}, indices shape: {2}, data shape: {302,2}

Start with model/layout/rapid_layout_self/model_handler/pp_doclayout/main.py, where batch inputs are passed to the ONNX Runtime session, then inspect inference_engine/onnxruntime/main.py and reproduce the failure with the reported batch_size=128. Done means PP-DocLayoutV2 batch inference completes without the reported ScatterND ONNXRuntimeError.