ubuntu下安装CUDA，cuDNN及pytorch-gpu版本的步骤教程

第一步：

安装显卡驱动：

sudo add-apt-repository ppa:graphics-drivers/ppa

sudo apt-get update

sudo apt-get install nvidia-390

sudo reboot

nvidia-smi

sudo apt-mark hold nvidia-390

cuda安装：

降级gcc g++：

sudo apt-get install gcc-4.8

sudo apt-get install g++-4.8

/usr/bin

ls –l gcc*

sudo mv gcc gcc.bak

sudo ln –s gcc-4.8 gcc

ls –l g++*

sudo mv g++ g++.bak

sudo ln –s g++-4.8 g++

gcc --version

g++ --version

安装cuda即可

sudo gedit ~/.barshrc

export PATH=/usr/local/cuda-9.1/bin${PATH:+:$PATH}}

export LD_LIBRARY_PATH=/usr/local/cuda-9.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

cudnn安装：

tar -zxvf cudnn.tgz

sudo cp /home/fin/cuda/targets/ppc64le-linux/include/cudnn.h /usr/local/cuda-9.1/include

sudo cp /home/fin/cuda/targets/ppc64le-linux/lib/libcudnn* /usr/local/cuda-9.1/lib64

sudo chmod a+r /usr/local/cuda-9.1/include/cudnn.h /usr/local/cuda-9.1/lib64/libcudnn*

更新pip

sudo chmod 777 -R anaconda3

pip install --upgrade pip

python -m pip install --upgrade pip

安装pytorch

pip install torch-0.4.0-cp36-cp36m-manylinux1_x86_64.whl

pip install torchvision