Ubuntu 18.04 安装NVIDIA驱动，配置CUDA，Tensorflow流程

参考链接

http://homeway.me/2017/08/08/setup-gpu-for-tensorflow/

 

安装过程中以root登录

文章部分安装包需要根据自己的情况进行下载，并不是傻瓜式操作。有需要可以留言或私信。可以发给你，有爱心的客官可以打赏个饭钱 ~_~。

 

1. 查看显卡配置

lspci | grep -i vga

 

2. 下载对应的显卡驱动

https://www.nvidia.com/Download/index.aspx?lang=en-us

为Nvidia驱动搭建依赖环境
 

sudo apt-get install dkms build-essential linux-headers-generic


sudo mkdir /etc/modprobe.d/

sudo touch /etc/modprobe.d/blacklist-nouveau.conf


cat >> /etc/modprobe.d/blacklist-nouveau.conf << EOF

blacklist nouveau

blacklist lbm-nouveau

options nouveau modeset=0

alias nouveau off

alias lbm-nouveau off

EOF

如果无法读写 sudo vi blacklist-nouveau.conf

echo options nouveau modeset=0 | sudo tee -a /etc/modprobe.d/nouveau-kms.conf

sudo update-initramfs -u


sudo reboot

NVIDIA-Linux-XXXX是对应的显卡驱动版本，需要修改

sudo chmod u+x NVIDIA-Linux-x86_64-390.77.run

sudo ./NVIDIA-Linux-x86_64-390.77.run

sudo reboot

3. 安装cuda

sudo dpkg -i cuda-repo-ubuntu1604-9-0-local_9.0.176-1_amd64.deb

sudo apt-key add /var/cuda-repo-9-0-local/7fa2af80.pub

sudo apt-get update

sudo apt-get install cuda -y


CUDA_HOME=/usr/local/cuda-10.0

ln -vfs $CUDA_HOME /usr/local/cuda

cat >> ~/.profile << EOF

export CUDA_HOME=${CUDA_HOME}

export LD_LIBRARY_PATH=${CUDA_HOME}/lib64

PATH=${CUDA_HOME}/bin:${PATH}

EOF

source ~/.profile

这里经验告诉我们需要再重启一下，reboot

 

测试CUDA

➜ $ cd $CUDA_HOME/samples/1_Utilities/deviceQuery

➜ $ make

➜ $ ./deviceQuery

 

4. 安装CUDA NN

tar xzvf cudnn-8.0-linux-x64-v5.1.tgz

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*

5. 安装Python和依赖环境

sudo apt-get install python-dev python3-dev python-numpy \

 python3-numpy python-six python3-six build-essential python-pip python3-pip


sudo dpkg -i cuda-repo-ubuntu1604_9.1.85-1_amd64.deb

sudo dpkg -i nvidia-machine-learning-repo-ubuntu1604_1.0.0-1_amd64.deb

sudo apt-get update


dpkg -i  nccl-repo-ubuntu1804-2.3.5-ga-cuda10.0_1-1_amd64.deb

apt-get update

6.安装tensorflow

pip3 install tensorflow-1.12.0rc0-cp36-cp36m-linux_x86_64.whl

需要重新装一下numpy

pip3 install -U numpy