我使用onnx官方的代码对我的onnx模型进行了量化，之后转换成om，出现下面的错误应该怎么解决？

转换命令：atc --model=/media/ai/root_fs/xianchang_hole_static_onnx.onnx --framework=5 --output=/media/ai/root_fs/xianchang_hole_static_onnx --soc_version=Ascend310B4 --log=error

错误：

ATC run failed, Please check the detail log, Try 'atc --help' for more information

E16005: [PID: 250308] 2025-11-20-09:03:09.002.059 The model has [9] [--domain_version] fields, but only one is allowed.

        Solution: Invalid ONNX model. Modify the ONNX model. If no domain is specified on the operator node, only one domain can be specified on the model.

        TraceBack (most recent call last):

        Model parse to graph failed, graph name:xianchang_hole_static_onnx.[FUNC:ModelParseToGraph][FILE:onnx_parser.cc][LINE:936]

        ATC model parse ret fail.[FUNC:ParseGraph][FILE:omg.cc][LINE:776]

①我的onnx模型应该是没有问题的，因为他可以正常使用onnxruntime进行推理。

②在别的帖子看到的解决方案：cid:link_0 用这个库尝试了一下：代码如下：

from magiconnx import OnnxGraph

graph = OnnxGraph('/home/HwHiAiUser/Documents/yolov8_web_inference/model/xianchang_hole_static_onnx.onnx')

print(graph.inputs)

print(graph.outputs)

graph.keep_default_domain()

graph.save('basic_vsr_v1.onnx')

又出现新的错误：

Traceback (most recent call last):

  File "/home/HwHiAiUser/Documents/yolov8_web_inference/test_model/onnx_simple.py", line 3, in <module>

    graph = OnnxGraph('/home/HwHiAiUser/Documents/yolov8_web_inference/model/xianchang_hole_static_onnx.onnx')

  File "/home/HwHiAiUser/.conda/envs/3dinference/lib/python3.10/site-packages/magiconnx/utils/log.py", line 22, in wrapper

    return func(*args, **kwargs)

  File "/home/HwHiAiUser/.conda/envs/3dinference/lib/python3.10/site-packages/magiconnx/graph.py", line 36, in __init__

    self._update_ops_map(out, node)

  File "/home/HwHiAiUser/.conda/envs/3dinference/lib/python3.10/site-packages/magiconnx/graph.py", line 340, in _update_ops_map

    raise RuntimeError(

RuntimeError: This is an invalid model. Error: two nodes with same node name (images_QuantizeLinear)

所以最开始的问题应该怎么解决？