npu_sys_ctrl.cpp:82 NPU error, error code is 100002

使用mmdetection 部署并测试模型时，出现以下错误：
 File "tools/test.py", line 152, in <module>
    main()
  File "tools/test.py", line 140, in main
    args.show, args.show_dir)
  File "/home/mmdeploy-0.10.0/mmdeploy/codebase/base/task.py", line 140, in single_gpu_test
    out_dir, **kwargs)
  File "/home/mmdeploy-0.10.0/mmdeploy/codebase/mmcls/deploy/mmclassification.py", line 144, in single_gpu_test
    model, data_loader, show, out_dir, win_name=win_name, **kwargs)
  File "/root/miniconda3/envs/mmlab/lib/python3.7/site-packages/mmcls-0.25.0-py3.7.egg/mmcls/apis/test.py", line 40, in single_gpu_test
    for i, data in enumerate(data_loader):
  File "/root/miniconda3/envs/mmlab/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 355, in __iter__
    return self._get_iterator()
  File "/root/miniconda3/envs/mmlab/lib/python3.7/site-packages/torch_npu/utils/dataloader.py", line 70, in _get_iterator
    return _MultiProcessingDataLoaderIter(self)
  File "/root/miniconda3/envs/mmlab/lib/python3.7/site-packages/torch_npu/utils/dataloader.py", line 137, in __init__
    train_device_id = torch_npu.npu.current_device()
  File "/root/miniconda3/envs/mmlab/lib/python3.7/site-packages/torch_npu/npu/utils.py", line 155, in current_device
    torch_npu.npu._lazy_init()  
File "/root/miniconda3/envs/mmlab/lib/python3.7/site-packages/torch_npu/npu/utils.py", line 102, in _lazy_init
    torch_npu._C._npu_init()
RuntimeError: Initialize:/home/pytorch-v1.8.1-3.0.rc3/torch_npu/csrc/core/npu/sys_ctrl/npu_sys_ctrl.cpp:82 NPU error, error code is 100002。。

但是在python 环境中 初始化完全正常：
>>> import torch_npu 
[W OperatorEntry.cpp:121] Warning: Overriding a previously registered kernel for the same operator and the same dispatch key
  operator: aten::_has_compatible_shallow_copy_type(Tensor self, Tensor from) -> (bool)
    registered at aten/src/ATen/RegisterSchema.cpp:20
  dispatch key: Math
  previous kernel: registered at aten/src/ATen/RegisterMath.cpp:5686
       new kernel: registered at /home/l00665583/pytorch-v1.8.1-3.0.rc3/torch_npu/csrc/aten/ops/HasCompatibleShallowCopyType.cpp:37 (function registerKernel)
>>> torch_npu._C._npu_init()

当前所用环境：
python: 3.7.16
pytorch: 1.8.0a0
torch_npu: '1.8.1.rc3'
mmcv-full: 1.7.1
CANN:6.0.1

Ascend/pytorch
暂停

内容风险标识

评论 (7)

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

内容风险标识