Win10环境下RTX3080使用Pytorch或Tensorflow的环境配置

小灰灰 2023-01-06 05:29 236阅读 0赞

# 报错信息 #

`GeForce RTX 3080 with CUDA capability sm_86 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 compute_37.`

我之前电脑为1650，电脑安装的pytorch版本对应的是`cuda 10.2`\# 先粘一下别人的解决方法，显然报错是因为版本不匹配

## 解决方法 ##

**环境**：Win10+Python3.6.12+CUDA11.1+Cudnn8.0.4

1.  CUDA11.1和Cudnn8.0.4从官网安
2.  使用`conda install pytorch torchvision cudatoolkit=11 -c pytorch-nightly`来安装pytorch
3.  使用tensorflow的话 可以先conda install cudatoolkit=11 然后再pip install tensorflow-gpu==1.15  
    先装好tf就无法安装cudatoolkit11，conda自带的是cudatoolkit10；  
    先装好cudatoolkit11，再用conda装tf可能会报错，用pip不会

## 注意 ##

由于需要cuda11的支持，因此向下兼容的原则不适用。

[转自][Link 1]

# 我本地环境说明 #

[如何查看电脑安装的cuda版本号  
][cuda_]

cuda信息  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0dhYnJpZWxfd2Vp_size_16_color_FFFFFF_t_70]  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0dhYnJpZWxfd2Vp_size_16_color_FFFFFF_t_70 1]

# 安装pytorch #

[pytorch配置][pytorch]我直接安装的是11.0  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0dhYnJpZWxfd2Vp_size_16_color_FFFFFF_t_70 2]

## 解决下载速度慢问题 ##

此处参考[安装与电脑CUDA匹配的pytorch版本  
][CUDA_pytorch_]

将原始命令

conda install pytorch torchvision torchaudio cudatoolkit=11.0 -c pytorch

改为

conda install pytorch torchvision torchaudio cudatoolkit=11.0

去掉了`-c pytorch`

### 可能出现的错误 ###

Collecting package metadata (current_repodata.json): done
    Solving environment: \
    The environment is inconsistent, please check the package plan carefully
    The following packages are causing the inconsistency:
    
      - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/win-64::torchaudio==0.7.0=py37
      - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/win-64::torchvision==0.8.1=py37_cu101
    |
    Warning: 2 possible package resolutions (only showing differing packages):
      - https://mirrors.ustc.edu.cn/anaconda/pkgs/main/win-64::anaconda-custom-py37_1
      - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/win-64::anaconda-custom-py37done

解决  
[The environment is inconsistent, please check the package plan carefully The following packages are][The environment is inconsistent_ please check the package plan carefully The following packages are]

[Link 1]: https://blog.csdn.net/weixin_43896241/article/details/108979744
[cuda_]: https://blog.csdn.net/Gabriel_wei/article/details/112595642
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0dhYnJpZWxfd2Vp_size_16_color_FFFFFF_t_70]: /images/20221119/259d9205468243518e11f18df33a3475.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0dhYnJpZWxfd2Vp_size_16_color_FFFFFF_t_70 1]: /images/20221119/7af2bcdbf82d498283b1afe33f5b1390.png
[pytorch]: https://pytorch.org/
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L0dhYnJpZWxfd2Vp_size_16_color_FFFFFF_t_70 2]: /images/20221119/ea370a53fa9949c59e85a57dd997fd6e.png
[CUDA_pytorch_]: https://blog.csdn.net/Gabriel_wei/article/details/109897853
[The environment is inconsistent_ please check the package plan carefully The following packages are]: https://blog.csdn.net/Gabriel_wei/article/details/112597140