Tensorflow in Ubuntu_16.04

完整步骤

1. CUDA 8.0

下载地址 https://developer.nvidia.com/cuda-downloads

选择runfile安装类型

安装时，不安装驱动，只安装Toolkits和Samples

出现以下信息时

Missing recommended library: libGLU.so 
Missing recommended library: libX11.so 
Missing recommended library: libXi.so 
Missing recommended library: libXmu.so

命令行运行以下代码并重新安装CUDA

sudo apt-get install freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev

2. cuDNN6.0

下载地址 https://developer.nvidia.com/cudnn

解压并运行以下命令

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*

3. 配置环境

sudo gedit ~/.bashrc

在末尾添加

export LD_LIBRARY_PATH=”$LD_LIBRARY_PATH:/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64”

export CUDA_HOME=/usr/local/cuda

exportPATH="$CUDA_HOME/bin:$PATH"

保存并运行

source ~/.bashrc

4. Tensorflow

官方教程 https://www.tensorflow.org/install/install_linux

虚拟环境安装（推荐）

安装pip和虚拟环境依据python版本运行以下其中一条命令:

$ sudo apt-get install python-pip python-dev python-virtualenv # for Python 2.7
$ sudo apt-get install python3-pip python3-dev python-virtualenv # for Python 3.n

运行以下命令创建虚拟环境

$ virtualenv --system-site-packages 目标文件夹 # for Python 2.7
$ virtualenv --system-site-packages -p python3 目标文件夹 # for Python 3.n

激活虚拟环境

$ source ~/tensorflow/bin/activate # bash, sh, ksh, or zsh
$ source ~/tensorflow/bin/activate.csh  # csh or tcsh

会看到命令行变为以下形式

(tensorflow)$

如果pip版本为8.1以上，运行
```
(tensorflow)$ easy_install -U pip
```

选择合适的命令行安装

(tensorflow)$ pip install --upgrade tensorflow      # for Python 2.7
(tensorflow)$ pip3 install --upgrade tensorflow     # for Python 3.n
(tensorflow)$ pip install --upgrade tensorflow-gpu  # for Python 2.7 and GPU
(tensorflow)$ pip3 install --upgrade tensorflow-gpu # for Python 3.n and GPU

每次使用Tensorflow的时候都要激活虚拟环境

$ source ~/tensorflow/bin/activate      # bash, sh, ksh, or zsh
$ source ~/tensorflow/bin/activate.csh  # csh or tcsh

不使用Tensorflow时，运行

(tensorflow)$ deactivate

卸载Tensorflow
```
$ rm -r targetDirectory 
```

测试

  $ python
  import tensorflow as tf
  hello = tf.constant('Hello, TensorFlow!')
  sess = tf.Session()
  print(sess.run(hello))

如果安装成功，会出现如下结果

  Hello, TensorFlow!

Congratulations！

安装教程

1. CUDA 8.0

2. cuDNN6.0

3. 配置环境

4. Tensorflow

FEATURED TAGS