一、预配版本信息

1、cuda=12.1.1

2、cudnn=8.9.7

3、pytorch=2.2.0

二、引用

深度学习之环境配置：【CUDA 12.1.1+cuDNN 8.9.1】最新安装教程记录 -- 20240429_torch 1.12.0对应torchvision-CSDN博客

补充：

cuda历史版本索引：

NVIDIA Developerhttps://developer.nvidia.com/                                    》Downloadshttps://developer.nvidia.com/downloads》CUDA Toolkit

》Download Nowhttps://developer.nvidia.com/cuda-downloads》Archive of Previous CUDA Releaseshttps://developer.nvidia.com/cuda-toolkit-archive

cudnn历史版本索引：

NVIDIA Developerhttps://developer.nvidia.com/Developer Download Centers | NVIDIA Developerhttps://developer.nvidia.com/downloadsHigh Performance Computing HPC SDK | NVIDIA Developerhttps://developer.nvidia.com/hpc-sdkCUDA-X GPU-Accelerated Libraries | NVIDIA Developerhttps://developer.nvidia.com/gpu-accelerated-librariesCUDA-X GPU-Accelerated Libraries | NVIDIA Developerhttps://developer.nvidia.com/gpu-accelerated-libraries#deep-learning

三、pytorch安装

pip install pytorch==2.2.0 torchvision==0.17.0 torchaudio==2.2.0 pytorch-cuda==12.1