ImportError: libhccl.so: cannot open shared object file: No such file or directory
收藏回复举报
ImportError: libhccl.so: cannot open shared object file: No such file or directory
t('forum.solved') 已解决
新人帖
发表于2025-08-26 12:19:14
0 查看

GPU型号

true

驱动安装参考步骤

https://bbs.huaweicloud.com/blogs/440057

python环境

python==3.10.4
torch==2.1.0
torch-npu=2.1.0.post13

测试代码

import torch
import torch_npu

x = torch.randn(2, 2).npu()
y = torch.randn(2, 2).npu()
z = x.mm(y)

print(z)

运行测试代码详细错误

Traceback (most recent call last):  File "/root/develop/CosyVoice/npu_t2.py", line 1, in <module>
    import torch
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch/__init__.py", line 1382, in <module>
    from .functional import *  # noqa: F403
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch/functional.py", line 7, in <module>
    import torch.nn.functional as F
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch/nn/__init__.py", line 1, in <module>
    from .modules import *  # noqa: F403
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch/nn/modules/__init__.py", line 35, in <module>
    from .transformer import TransformerEncoder, TransformerDecoder, \
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch/nn/modules/transformer.py", line 20, in <module>
    device: torch.device = torch.device(torch._C._get_default_device()),  # torch.device('cpu'),
/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch/nn/modules/transformer.py:20: UserWarning: Failed to initialize NumPy: _ARRAY_API not found (Triggered internally at /pytorch/torch/csrc/utils/tensor_numpy.cpp:84.)
  device: torch.device = torch.device(torch._C._get_default_device()),  # torch.device('cpu'),
Traceback (most recent call last):
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch_npu/__init__.py", line 19, in <module>
    import torch_npu.npu
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch_npu/npu/__init__.py", line 128, in <module>
    from torch_npu.utils import _should_print_warning
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch_npu/utils/__init__.py", line 4, in <module>
    from torch_npu import _C
ImportError: libhccl.so: cannot open shared object file: No such file or directory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/develop/CosyVoice/npu_t2.py", line 2, in <module>
    import torch_npu
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch_npu/__init__.py", line 21, in <module>
    from torch_npu.utils._error_code import ErrCode, pta_error
  File "/root/miniconda3/envs/cosyvoice/lib/python3.10/site-packages/torch_npu/utils/__init__.py", line 4, in <module>
    from torch_npu import _C
ImportError: libhccl.so: cannot open shared object file: No such file or directory

问题1:

产生这个问题原因是驱动安装问题还是包版本问题,是否有相应的解决方案?

问题2:

是否有相关的验证demo,确定驱动是否正常,torch_npu是否正常,对应的包适配是否正常?如果有相关官方函数,开发就可以自己快速验证是哪个环节问题

我要发帖子