PyTorch：安装和配置

向右看齐 2022-01-26 13:25 424阅读 0赞

## 安装 ##

**pip安装**

pip3 install torch torchvision

**macos还需要安装**

brew install libomp

否则出错：ImportError: dlopen(/...torch/\_C.cpython-36m-darwin.so, 9): Library not loaded: /usr/local/opt/libomp/lib/libomp.dylib  
  Referenced from: .../torch/lib/libshm.dylib  Reason: image not found

**安装固定cuda版本对应的pytorch**

pip install torch==1.7.1+cu92 torchvision==0.8.2+cu92 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch\_stable.html

其它cuda版本对应的pytorch版本查询\[[https://download.pytorch.org/whl/torch\_stable.html][https_download.pytorch.org_whl_torch_stable.html]\]  
如cuda=9.0版本时，支持的最高pytorch版本可能就是1.1.0了：[cu90/torch-1.1.0-cp36-cp36m-linux\_x86\_64.whl][cu90_torch-1.1.0-cp36-cp36m-linux_x86_64.whl]。

\[[INSTALL PYTORCH][]\]\[[INSTALLING PREVIOUS VERSIONS OF PYTORCH][]\]

[\-柚子皮-][-_-]

显卡驱动版本-cuda版本-pytorch版本需要一一对应，否则torch.cuda.is\_available()=False。

### 显卡驱动版本查询 ###

$nvidia-smi    \#Driver Version: 384.81  
Mon Jan 11 14:51:48 2021         
\+-----------------------------------------------------------------------------+  
| NVIDIA-SMI 384.81                 Driver Version: 384.81                    |  
|-------------------------------+----------------------+----------------------+  
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |  
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |  
|===============================+======================+======================|  
|   0  Tesla P100-PCIE...  On   | 00000000:04:00.0 Off |                    0 |  
| N/A   33C    P0    25W / 250W |      0MiB / 16276MiB |      0%      Default |  
\+-------------------------------+----------------------+----------------------+  
|   1  Tesla P100-PCIE...  On   | 00000000:81:00.0 Off |                    0 |  
| N/A   28C    P0    25W / 250W |      0MiB / 16276MiB |      0%      Default |  
\+-------------------------------+----------------------+----------------------+  
                                                                                 
\+-----------------------------------------------------------------------------+  
| Processes:                                                       GPU Memory |  
|  GPU       PID   Type   Process name                             Usage      |  
|=============================================================================|  
|  No running processes found                                                 |  
\+-----------------------------------------------------------------------------+

### cuda(toolkit)版本查询 ###

nvcc --version或nvcc -V，如果 nvcc 没有安装，那么用

$cat /usr/local/cuda/version.txt

CUDA Version 9.0.176

查看pytorch对应cuda的版本

import torch

print(torch.version.cuda)

**查看 cuDNN 版本**  
cat /usr/local/cuda/include/cudnn.h | grep CUDNN\_MAJOR -A 2  
如果没有，那么可能没有安装 cuDNN。

Note: NVIDIA cuDNN是用于深度神经网络的GPU加速库。

### 驱动版本和cuda(toolkit)版本对应关系 ###

<table>  Table 2. CUDA Toolkit and Compatible Driver Versions  
 <thead> 
  <tr> 
   <th colspan="1">CUDA Toolkit</th> 
   <th colspan="1">Linux x86_64 Driver Version</th> 
   <th colspan="1">Windows x86_64 Driver Version</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 11.2.0 GA</td> 
   <td colspan="1" style="vertical-align:top;">&gt;=460.27.04</td> 
   <td colspan="1" style="vertical-align:top;">&gt;=460.89</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 11.1.1 Update 1</td> 
   <td colspan="1" style="vertical-align:top;">&gt;=455.32</td> 
   <td colspan="1" style="vertical-align:top;">&gt;=456.81</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 11.1 GA</td> 
   <td colspan="1" style="vertical-align:top;">&gt;=455.23</td> 
   <td colspan="1" style="vertical-align:top;">&gt;=456.38</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 11.0.3 Update 1</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 450.51.06</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 451.82</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 11.0.2 GA</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 450.51.05</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 451.48</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 11.0.1 RC</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 450.36.06</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 451.22</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 10.2.89</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 440.33</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 441.22</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 10.1 (10.1.105 general release, and updates)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 418.39</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 418.96</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 10.0.130</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 410.48</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 411.31</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 9.2 (9.2.148 Update 1)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 396.37</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 398.26</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 9.2 (9.2.88)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 396.26</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 397.44</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 9.1 (9.1.85)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 390.46</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 391.29</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 9.0 (9.0.76)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 384.81</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 385.54</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 8.0 (8.0.61 GA2)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 375.26</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 376.51</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 8.0 (8.0.44)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 367.48</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 369.30</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 7.5 (7.5.16)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 352.31</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 353.66</td> 
  </tr> 
  <tr> 
   <td colspan="1" style="vertical-align:top;">CUDA 7.0 (7.0.28)</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 346.46</td> 
   <td colspan="1" style="vertical-align:top;">&gt;= 347.62</td> 
  </tr> 
 </tbody> 
</table>

\[官方[对照表][Link 1]\]\[[nvidia显卡，驱动以及cuda版本对应查询][nvidia_cuda]\]

\[[pytorch版本，cuda版本，系统cuda版本查询和对应关系][pytorch_cuda_cuda]\]

from: [\-柚子皮-][-_-]

ref:

[https_download.pytorch.org_whl_torch_stable.html]: https://download.pytorch.org/whl/torch_stable.html
[cu90_torch-1.1.0-cp36-cp36m-linux_x86_64.whl]: https://download.pytorch.org/whl/cu90/torch-1.1.0-cp36-cp36m-linux_x86_64.whl
[INSTALL PYTORCH]: https://pytorch.org/
[INSTALLING PREVIOUS VERSIONS OF PYTORCH]: https://pytorch.org/get-started/previous-versions/
[-_-]: https://me.csdn.net/pipisorry
[Link 1]: https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html
[nvidia_cuda]: https://blog.csdn.net/He_9520/article/details/100032803
[pytorch_cuda_cuda]: https://www.cnblogs.com/Wanggcong/p/12625540.html