搭建TensorFlow环境

TensorFlow GitHub

克隆TensorFlow仓库

$ sudo apt install git
$ git clone --recurse-submodules https://github.com/tensorflow/tensorflow.git

安装Bazel

Getting Started with Bazel
Ubuntu Linux
Compile Bazel from source
GitHub releases page
下载bazel-0.11.1-dist.zip，如果下载不下来，传送门

$ sudo apt install openjdk-8-jdk
$ cd bazel-0.11.1-dist/
$ ./compile.sh

将执行路径output/bazel添加到$PATH环境变量中

$ vi ~/.bashrc
export PATH=$PATH:/home/zhnd/bazel-0.11.1-dist/output
$ source ~/.bashrc

配置TensorFlow

$ cd tensorflow/
$ ./configure

创建pip包并安装

$ sudo apt install python-pip
$ sudo apt install python-numpy
$ cd tensorflow/
$ bazel build -c opt //tensorflow/tools/pip_package:build_pip_package
$ bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg
# /tmp/tensorflow_pkg目录下生成xxx.whl文件（e.g. tensorflow-1.7.0rc1-cp27-cp27mu-linux_x86_64.whl）
$ pip install /tmp/tensorflow_pkg/xxx.whl

Try your first TensorFlow program

$ python

>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> sess.run(hello)
'Hello, TensorFlow!'
>>> a = tf.constant(10)
>>> b = tf.constant(32)
>>> sess.run(a + b)
42
>>> sess.close()

参考：TensorFlow中文社区