https://blog.csdn.net/BigData_Mining/article/details/99670642

https://blog.csdn.net/deeposcar/article/details/80710842

遇见问题解决方法

问题 :
安装nvidia驱动reboot
安装cuda reboot

delete cuda 配置

https://blog.csdn.net/u012074597/article/details/80317275

• --purge选项会将配置文件、数据库等删除
  $ sudo apt-get autoremove --purge cuda

install cuda

0 install nvidia

NVIDIA 内核驱动版本与系统驱动不一致

• 输入下条命令，查看你的显卡驱动所使用的内核版本

cat /proc/driver/nvidia/version

• 使用下面的命令查看系统推荐安装哪个版本的N卡驱动
  ubuntu-drivers devices

• install nvidia 驱动
  sudo ubuntu-drivers autoinstall

• reboot
  https://blog.csdn.net/qq_40200387/article/details/90341107
  https://blog.csdn.net/deeposcar/article/details/80710842

1. install cuda

https://developer.nvidia.com/cuda-10.0-download-archive?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1804&target_type=deblocal

sudo dpkg -i cuda-repo-ubuntu1804-10-0-local-10.0.130-410.48_1.0-1_amd64.deb
sudo apt-key add /var/cuda-repo-<version>/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda

https://blog.csdn.net/u010417185/article/details/59105765

• 添加到环境变量

export PATH=/us/local/cuda-10.2/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-10.2/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

• reboot

• nvcc

install cudnn

down cuDNN Library for Linux

注意，解压后的文件夹名称为cuda ,将对应文件复制到 /usr/local中的cuda内

• 1
  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 /usr/local/cuda/lib64/libcudnn*
• 2
  sudo cp cuda/include/cudnn.h /usr/local/cuda10.2/include
  sudo cp cuda/lib64/libcudnn* /usr/local/cuda10.2/lib64
  sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda10.2/lib64/libcudnn*

developer

cuDNN Runtime Library for Ubuntu18.04 (Deb)
cuDNN Developer Library for Ubuntu18.04 (Deb)
sudo dpkg -i ***.deb

cudnn 版本问题

developer library 包含了在Ubuntu系统上开发深度学习时所需的cuDNN头文件，如果你不需要开发编译任何深度学习程序，而只是将其用于运行某些深度学习应用，那么只下载“runtime library”就足够了。
我是把3个文件都下载下来了，分别安装。

• Runtime

• Developer

• cuDNN Library for Linux

• cudnn 安装成功是否验证
  cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2

2.多版本切换和安装

https://blog.csdn.net/sinat_29963957/article/details/83108324