ImportError: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory

ImportError: libtorch_cuda_cu.so: cannot open shared object file: No such file or directory

这是我在搭建mmpose环境进行最终环境测试遇到的问题，记录一下

第一步：判定pytorch版本是否相符

这个其实大部分人都不会犯错，我这边检查发现torch版本配置没问题，适合cuda版本相适配的

检查mmcv版本的问题

官网这里给出的是大致版本适配，但实际上我就是这里出问题的
因为具体在这这些里面的小版本中还是要注意兼容的情况的

我的解决办法

首先我卸载了mmcv，然后重新安装
之后控制台报错

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
mmpose 1.3.2 requires mmdet<3.3.0,>=3.0.0, but you have mmdet 3.3.0 which is incompatible.
mmdet 3.3.0 requires mmcv<2.2.0,>=2.0.0rc4, but you have mmcv 2.2.0 which is incompatible.
Successfully installed mmcv-2.2.0

也就是说mmcv版本是不适配的，然后我把报错扔给大模型给出了建议的降级版本，
之后重新安装，最终成功运行
这里给出我的相关版本

(mmpose01) [class1@node01 mmpose]$ python -c "import torch; print(torch.__version__)"
2.1.1+cu118
(mmpose01) [class1@node01 mmpose]$ pip show torch torchvision torchaudio
Name: torch
Version: 2.1.1+cu118
Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration
Home-page: https://pytorch.org/
Author: PyTorch Team
Author-email: packages@pytorch.org
License: BSD-3
Location: /home/class1/.conda/envs/mmpose01/lib/python3.8/site-packages
Requires: filelock, fsspec, jinja2, networkx, sympy, triton, typing-extensions
Required-by: efficientnet-pytorch, lpips, pytorch-triton-rocm, torch-ema, torchaudio, torchvision
---
Name: torchvision
Version: 0.16.1+cu118
Summary: image and video datasets and models for torch deep learning
Home-page: https://github.com/pytorch/vision
Author: PyTorch Core Team
Author-email: soumith@pytorch.org
License: BSD
Location: /home/class1/.conda/envs/mmpose01/lib/python3.8/site-packages
Requires: numpy, pillow, requests, torch
Required-by: lpips, mmpose
---
Name: torchaudio
Version: 2.1.1+cu118
Summary: An audio package for PyTorch
Home-page: https://github.com/pytorch/audio
Author: Soumith Chintala, David Pollack, Sean Naren, Peter Goldsborough, Moto Hira, Caroline Chen, Jeff Hwang, Zhaoheng Ni, Xiaohui Zhang
Author-email: soumith@pytorch.org
License: 
Location: /home/class1/.conda/envs/mmpose01/lib/python3.8/site-packages
Requires: torch
Required-by: 
(mmpose01) [class1@node01 mmpose]$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Wed_Sep_21_10:33:58_PDT_2022
Cuda compilation tools, release 11.8, V11.8.89
Build cuda_11.8.r11.8/compiler.31833905_0
(mmpose01) [class1@node01 mmpose]$ pip list | grep mm
cumm-cu118                        0.6.3
mmcv                              2.1.0
mmdet                             3.2.0
mmengine                          0.9.0
mmpose                            1.3.2
(mmpose01) [class1@node01 mmpose]$ 

在该环境下我是可以成功运行的