【CUDA】显卡驱动版本、CUDA、cudnn之间关系以及ubuntu18.04卸载安装多版本CUDA

CUDA与cuDNN关系:https://www.jianshu.com/p/622f47f94784
NVIDIA显卡驱动版本，CUDA版本，cudnn版本之间关系及如何选择:https://blog.csdn.net/iamqianrenzhan/article/details/89343601

1. 显卡驱动装最新的,因为可以向下兼容更多的CUDA版本
2. CUDA版本根据所需的Tensorflow或Pytorch的要求进行选择
3. 在根据CUDA选择对应的cudnn版本

一行命令卸载:https://blog.csdn.net/qq_33200967/article/details/80689543#CUDA_71
ubuntu18.04安装CUDA9.0:https://www.jianshu.com/p/288c536a1092
多版本切换:https://blog.csdn.net/qq_23996885/article/details/90205937
cudnn按不成功解决方法:在tgz包
https://blog.csdn.net/CAU_Ayao/article/details/83627342#1_Cuda90_14
https://blog.csdn.net/Aiolia86/article/details/80342240
检查cudnn是否安装成功:https://blog.csdn.net/wanzhen4330/article/details/81704474

cudnn安装顺序：

sudo dpkg -i libcudnn7_7.0.5.15-1+cuda9.0_amd64.deb 
sudo dpkg -i libcudnn7-dev_7.0.5.15-1+cuda9.0_amd64.deb 
sudo dpkg -i libcudnn7-doc_7.0.5.15-1+cuda9.0_amd64.deb 

https://blog.csdn.net/m0_37407756/article/details/80769952

cudnn另一种安装方法：压缩包的安装方法

sudo cp cuda/include/cudnn.h /usr/local/cuda/include/
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64/ 
sudo chmod a+r /usr/local/cuda/include/cudnn.h
sudo chmod a+r /usr/local/cuda/lib64/libcudnn*

https://blog.csdn.net/wanzhen4330/article/details/81699769

查看cuda和cudnn版本：

nvcc  -V
cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2