IndexError: Argument passed to at() was not in the map.

Deepcong

于 2024-12-27 13:32:03 发布

阅读量284

点赞数 1

CC 4.0 BY-SA版权

文章标签：深度学习机器学习人工智能

本文链接：https://blog.youkuaiyun.com/DeepCBW/article/details/144766669

torch.onnx.export(
                    pose_model,
                    (x,),
                    check_point.split('.')[0] + '.onnx',
                    input_names=["input"],
                    output_names=["output"],
                    opset_version=11,
                    dynamic_axes={
                        "images": {0: "batch"},
                        "keypoints": {0: "batch"}
                    },
                    autograd_inlining=False,
                    operator_export_type=torch.onnx.OperatorExportTypes.ONNX_ATEN_FALLBACK
                )

解决报错：

加上 autograd_inlining=False,就行了