CUDA + CUDNN + Aanaconda + Tensorflow-GPU + Pytorch-GPU

1. Install CUDA

Ubuntu下安装NVIDIA驱动的三种方法：

1. 使用标准Ubuntu仓库进行自动安装

2. 使用PPA仓库进行自动化安装

3. 使用官方的NVIDIA驱动进行手动安装

个人认为，第一种方法操作最为简单，方便，第三种方法是最稳定，最常用的。

很多博客进行驱动安装前都需要禁用Nouveau驱动，但是我安装驱动的时候搞了很久，最后安装好驱动就是根据第一种方式并且没禁用Nouveau驱动的情况下，我也找到了开始一直安装不成功的原因，就是我的GCC和G++降级后没有升级，所以如果可以先试试把GCC和G++升级到最高版本，然后运用第一种方式安装驱动。

1. 使用标准Ubuntu 仓库进行自动安装

    sudo apt install gcc/g++
    sudo ubuntu-drivers devices
    sudo ubuntu-drivers autoinstall
   

2.使用PPA仓库进行自动安装

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

CUDA 9.0

https://developer.nvidia.com/cuda-90-download-archive?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1704&target_type=runfilelocal

先安装一些依赖/ change Chinese server

sudo apt-get install -y libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler
sudo apt-get install -y --no-install-recommends libboost-all-dev
sudo apt-get install -y libopenblas-dev liblapack-dev libatlas-base-dev
sudo apt-get install -y libgflags-dev libgoogle-glog-dev liblmdb-dev
sudo apt-get install -y git cmake build-essential
sudo apt-get install -y freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev 

sudo apt-get update

sudo chmod a+x cuda_9.0.176_384.81_linux.run
sudo sh ./cuda_9.0.176_384.81_linux.run

vim .bashrc

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64
export PATH=$PATH:/usr/local/cuda/bin
export CUDA_HOME=$CUDA_HOME:/usr/local/cuda

source ~/.bashrc

2. Install CUDNN

https://developer.nvidia.com/rdp/cudnn-archive

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

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

3. Install Anaconda/ Pycharm

https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/?C=M&O=D
sudo chmod a+x Anaconda3-4.3.0-Linux-x86_64.sh
sudo ./Anaconda3-4.3.0-Linux-x86_64.sh

https://www.jetbrains.com/pycharm/download/#section=linux
cd ./ ... /bin
./pycharm.sh
Tools——Creat Desktop Entry；

4. Install tensorflow-gpu

conda create -n cyr python=3.5
pip install tensorflow-gpu==1.10

import tensorflow as tf
print(tf.test.is_gpu_available())

5. Install pytorch-gpu

https://pytorch.org/get-started/previous-versions/

do not add -c pytroch

pip install torch-1.1.0-cp35-cp35m-linux_x86_64.whl
pip install torchvision-0.3.0-cp35-cp35m-manylinux1_x86_64.whl

import torch
torch.cuda.is_available()