Ubuntu 18.04 GTX1060 Nvidia驱动安装 + CUDA8.0 + cudnn5.1

转至：https://blog.csdn.net/asd85045625/article/details/80348270

安装Nvidia驱动

sudo ubuntu-drivers autoinstall #安装驱动

验证安装情况

nvidia-smi

 

安装CUDA8.0

在官网下载CUDA8.0.deb

进入下载目录后

sudo dpkg -i cuda-repo-ubuntu1704-8-0-local_9.0.176-1_amd64.deb
sudo apt-key add /var/cuda-repo-8-0-local-ga2/7fa2af80.pub
sudo apt-get update

sudo apt-get install aptitude

sudo aptitude install cuda

配置环境

sudo gedit ~/.bashrc  
在最后两行
export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

保存，然后重启电脑 sudo reboot

测试CUDA的例子，需要编译一下

cd /usr/local/cuda-8.0/samples/1_Utilities/deviceQuery
sudo make
./deviceQuery
显示Result=pass，则测试通过，安装成功

安装CUDNN

进入下载文件夹

tar -zxvf cudnn-8.0-linux-x64-v5.tgz
sudo cp cuda/include/cudnn.h /usr/local/cuda-8.0/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda-8.0/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda-8.0/lib64/libcudnn*
安装tensorflow，指定版本安装

cuda 版本 

cat /usr/local/cuda/version.txt

cudnn 版本 

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