[应用开发]在线打推理将模型执行的结果从NPU拷到CPU时，AICPU报错

在线打推理将模型执行的结果从NPU拷到CPU时，AICPU报错，LOG见附件

output.len  1
  3%|███▋                                                                                                                                      | 1/38.0 [02:40<1:39:16, 160.99s/it]skip_frame       3
refine_output     Traceback (most recent call last):
  File "inference_video.py", line 245, in <module>
    output = make_inference(I0, I1, 2**args.exp-1) if args.exp else []
  File "inference_video.py", line 172, in make_inference
    middle = model.inference(I0, I1, args.scale)
  File "/home/HwHiAiUser/wubo/RIFE/model/RIFE_HDv2.py", line 206, in inference
    return self.predict(imgs, flow, training=False)
  File "/home/HwHiAiUser/wubo/RIFE/model/RIFE_HDv2.py", line 185, in predict
    print("refine_output    ", refine_output)
  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/tensor.py", line 191, in __repr__
    return torch._tensor_str._str(self)
  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/_tensor_str.py", line 337, in _str
    tensor_str = _tensor_str(self, indent)
  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/_tensor_str.py", line 234, in _tensor_str
    formatter = _Formatter(get_summarized_data(self) if summarize else self)
  File "/usr/local/python3.7.5/lib/python3.7/site-packages/torch/_tensor_str.py", line 96, in __init__
    tensor = tensor.to("cpu")
RuntimeError: ACL stream synchronize failed, error code:507018
THPModule_npu_shutdown success.

Ascend/pytorch
暂停

内容风险标识

评论 (4)

Ascend/pytorch暂停 .gitee-modal { width: 500px !important; }

内容风险标识