install nvidia drivers
reference:https://linuxconfig.org/how-to-install-the-nvidia-drivers-on-ubuntu-18-04-bionic-beaver-linux
下面这部分可以忽略
////安装显卡驱动
https://blog.csdn.net/jhszh418762259/article/details/52957166
////卸载已有驱动
`sudo apt-get remove --purge nvidia*`
////下载显卡驱动？
https://www.nvidia.com/Download/index.aspx?lang=en-us
install cuda8.0
//查看GPU
lspci | grep -i nvidia
//检查自己的系统版本
uname -m && cat /etc/*release
//查看自己系统内核版本
uname -r
//安装对应内核版本的 kernel headers and development packages
sudo apt-get install linux-headers-$(uname -r)
//cuda下载地址
https://developer.nvidia.com/cuda-toolkit-archive
//检查cuda8的md5
https://developer.download.nvidia.com/compute/cuda/8.0/secure/Prod2/docs/sidebar/md5sum.txt?LqqYQR5PrLV_00SwuiZG3ceD6quk8zXcomNeGenuI4zZhjKzv2LzP2oNUMiaF5QG203I3i7uF4zfxDxXmcVxVNJi2OkkhvMwLWkz7xUoTuZjqoGaY9_wtzbaWsP43ss2tL3AU8-TqrWg_BT5sJN5ir9U798
sudo dpkg -i cuda-repo-ubuntu1404-8-0-local-ga2_8.0.61-1_amd64.deb 遇到如下错误：
dpkg-deb: error: `cuda-repo-ubuntu1404-8-0-local-ga2_8.0.61-1_amd64.deb' is not a debian format archive
dpkg: error processing archive cuda-repo-ubuntu1404-8-0-local-ga2_8.0.61-1_amd64.deb (--install):
subprocess dpkg-deb --control returned error exit status 2
Errors were encountered while processing:
cuda-repo-ubuntu1404-8-0-local-ga2_8.0.61-1_amd64.deb
原因是md5不对，详见：https://devtalk.nvidia.com/default/topic/993274/bad-checksum-for-cuda-8-0-for-ubuntu-14-04/
重新下载即可。我下载了四次才成功。
开始安装：
sudo dpkg -i cuda-repo-ubuntu1404-8-0-local-ga2_8.0.61-1_amd64.deb
sudo apt-get update
sudo apt-get install cuda
如无意外就安装成功了，下面设置环境变量。
sudo vim ~/.bashrc
在后面添加：
export CUDA_HOME=/usr/local/cuda-8.0
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64:$LD_LIBRARY_PATH
export PATH=/usr/local/cuda-8.0/bin:$PATH
保存退出，刷新source ~/.bashrc。
查看版本：nvcc -V
install CUDNN
reference: https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html
2.3.1. Installing from a Tar File
Navigate to your <cudnnpath> directory containing the cuDNN Tar file.
Unzip the cuDNN package.
$ tar -xzvf cudnn-9.0-linux-x64-v7.tgz
Copy the following files into the CUDA Toolkit directory, and change the file permissions.
$ 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*
2.3.2. Installing from a Debian File
Navigate to your <cudnnpath> directory containing cuDNN Debian file.
Install the runtime library, for example:
sudo dpkg -i libcudnn7_7.0.3.11-1+cuda9.0_amd64.deb #一定要先安装这个，否则会报错类似这个：libcudnn7-dev depends on libcudnn7 (= 7.0.1.13-1+cuda8.0); however:Package libcudnn7 is not installed.
Install the developer library, for example:
sudo dpkg -i libcudnn7-dev_7.0.3.11-1+cuda9.0_amd64.deb
Install the code samples and the cuDNN Library User Guide, for example:
sudo dpkg -i libcudnn7-doc_7.0.3.11-1+cuda9.0_amd64.deb #可选
Bonus
更新nvidia之后，可能出现有些命令无法执行的情况：Failed to initialize NVML: Driver/library version mismatch.
解决方法：重启或者参考 https://comzyh.com/blog/archives/967/
参考：
https://docs.nvidia.com/cuda/pdf/CUDA_Installation_Guide_Linux.pdf
https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html
https://blog.csdn.net/sinat_19628145/article/details/60475696
https://www.jianshu.com/p/35c7fde85968
https://blog.csdn.net/jhszh418762259/article/details/52958287
https://blog.csdn.net/m0_37924639/article/details/78785699
