Ubuntu16.04安装TensorFlow-gpu+CUDA9.0+cuDNN v7

1.准备工作

更新Ubuntu16.04源，用的是中科大的源：

cd /etc/apt/
sudo nano sources.list

把下面的这些源添加到source.list文件头部：

deb http://mirrors.ustc.edu.cn/ubuntu/ xenial main restricted universe multiverse
deb http://mirrors.ustc.edu.cn/ubuntu/ xenial-security main restricted universe multiverse
deb http://mirrors.ustc.edu.cn/ubuntu/ xenial-updates main restricted universe multiverse
deb http://mirrors.ustc.edu.cn/ubuntu/ xenial-proposed main restricted universe multiverse
deb http://mirrors.ustc.edu.cn/ubuntu/ xenial-backports main restricted universe multiverse
deb-src http://mirrors.ustc.edu.cn/ubuntu/ xenial main restricted universe multiverse
deb-src http://mirrors.ustc.edu.cn/ubuntu/ xenial-security main restricted universe multiverse
deb-src http://mirrors.ustc.edu.cn/ubuntu/ xenial-updates main restricted universe multiverse
deb-src http://mirrors.ustc.edu.cn/ubuntu/ xenial-proposed main restricted universe multiverse
deb-src http://mirrors.ustc.edu.cn/ubuntu/ xenial-backports main restricted universe multiverse

最后更新源和更新已安装的包：

sudo apt-get update
sudo apt-get upgrade

将pip源指向清华大学的源镜像：https://mirrors.tuna.tsinghua.edu.cn/help/pypi/，具体添加一个 ~/.config/pip/pip.conf 文件

cd .config
sudo mkdir pip
sudo touch pip.conf
sudo gedit pip.conf

设置为：

[global]
index-url = https://pypi.tuna.tsinghua.edu.cn/simple

2. 安装驱动

系统设置->软件更新->附加驱动->选择nvidia最新驱动->应用更改

3. 安装CUDA

下载CUDA9.0
下载完毕，进入文件的保存路径，运行如下代码：

sudo sh cuda_8.0.61_375.26_linux.run

开始安装，首先是一份长达两公里的协议，一直按回车到底以后，输入accept。其他的操作如下所示:

Do you accept the previously read EULA?
accept/decline/quit: accept
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 367.48?
(y)es/(n)o/(q)uit: n
Install the CUDA 8.0 Toolkit?
(y)es/(n)o/(q)uit: y
Enter Toolkit Location
[ default is /usr/local/cuda-8.0 ]:
Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y
Install the CUDA 8.0 Samples?
(y)es/(n)o/(q)uit: y

除了询问是否安装显卡驱动选 n ，其他地方均为y
配置环境变量：

sudo gedit ~/.bashrc

在尾部加上这么两句话:

export PATH=/usr/local/cuda-9.0/bin:$PATH

和

export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64:$LD_LIBRARY_PATH

保存后请运行:

source ~/.bashrc

编译cuda自带的Sample，指令如下:

cd /usr/local/cuda/samples/1_Utilities/deviceQuery
sudo make
./deviceQuery

结果如下:

4. 安装cuDNN

下载地址如下： https://developer.nvidia.com/rdp/cudnn-download
选择和CUDA对应的cudnn版本(CUDA9.0对应cudnn v7)
使用tar进行解压：

tar -xvf cudnn-8.0-linux-x64-v6.0.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
sudo chmod a+r /usr/local/cuda/lib64/libcudnn*

在安装Tensorflow之前，按照Tensorflow官方安装文档的说明，需要先安装一个libcupti-dev库：

sudo apt-get install libcupti-dev

安装TensorFlow

sudo apt-get install python3-pip
pip3 install tensorflow-gpu

进入Python3

import tensorflow as tf
hello = tf.constant(‘Hello, TensorFlow!’)
sess = tf.Session()
print(sess.run(hello))

5. 遇到的问题及解决

问题1：
sudo apt-get upgrade时遇到错误:
dpkg: error processing package grub-efi-amd64-signed (–configure):
subprocess installed post-installation script returned error exit status 1
Errors were encountered while processing:
grub-efi-amd64-signed
E: Sub-process /usr/bin/dpkg returned an error code (1)
解决方法：

sudo apt-get purge grub*
sudo apt-get install grub-efi
sudo apt-get autoremove
sudo update-grub

问题2：
import tensorflow时遇到错误：
ImportError: libcudart.so.8.0: cannot open shared object file: No such file or directory
解决方法：

sudo ldconfig /usr/local/cuda/lib64