28 lines
1.0 KiB
Markdown
28 lines
1.0 KiB
Markdown
|
---
|
||
|
|
title: CUDA12.2使用支持CUDA的pytorch
|
||
|
|
tags:
|
||
|
|
- AI
|
||
|
|
- python
|
||
|
|
- 学习笔记
|
||
|
|
categories:
|
||
|
|
- 学习笔记
|
||
|
|
index_img: /img/2024/2/17/torch.png
|
||
|
|
banner_img: /img/2024/2/17/torch.png
|
||
|
|
permalink: /2024/02/17/cuda122usepytorch/index.html
|
||
|
|
date: 2024-02-17 11:36:55
|
||
|
|
---
|
||
|
|
如果你在nvidia官网安装了最新版的CUDA(CUDA12.2/12.3),但访问 [https://download.pytorch.org/whl/cu122/torch_stable.html](https://download.pytorch.org/whl/cu122/torch_stable.html) 就会发现pytorch并没有支持此版本的CUDA
|
||
|
|
但是支持CUDA12.1的pytorch也可以支持CUDA12.2(看网上说CUDA12.3也可以支持,但没有试过)
|
||
|
|
执行
|
||
|
|
```shell
|
||
|
|
pip install torch torchvision torchaudio -f https://download.pytorch.org/whl/cu121/torch_stable.html
|
||
|
|
```
|
||
|
|
来安装支持CUDA12.1的torch
|
||
|
|
### 检查
|
||
|
|
创建一个.py文件,内容为:
|
||
|
|
```python
|
||
|
|
import torch
|
||
|
|
print(torch.cuda.is_available())
|
||
|
|
```
|
||
|
|
运行它,如果返回True,则说明PyTorch已成功启用CUDA支持
|
||
|
|
如果返回False,那么重装一下torch或重启一下服务器应该就好了
|