tensorflow 源码安装（for MacOS)
install tensorflow from sources for MacOS
运行tensorflow时经常遇到如下问题：
The TensorFlow library wasn’t compiled to use SSE4.1/SSE4.2 /AVX/AVX2/FMA instructions, but these are available on your machine and could speed up CPU computations
于是好奇就想尝试通过tensorflow源码安装解决这个问题
先说一下我的电脑系统版本：macOS sierra version 10.12.6 不带独显所以我装的是cpu版本的tensorflow
下面就是叙述一下安装步骤和我在安装过程中遇到的一些问题：
1. 先把tensorflow源码下载(clone the tensorflow repository)
$ git clone https://github.com/tensorflow/tensorflow
$ cd tensorflow
$ git checkout Branch # where Branch is the desired branch
2. 配置环境(prepare environment for MacOS)
在安装tensorflow之前，必须先安装bazel和python依赖包(six numpy wheel)
利用homebrew安装bazel,先安装JDK 8和homebrew, 这两个我很早就安装了就不叙述了
利用brew安装bazel如下：
brew install bazel
bazel version
brew upgrade bazel
install python dependencies
pip install six numpy wheel
3. 配置安装tensorflow
前面下载的安装源里有一个bash 脚本文件：congigure
这个文件可以帮助你识别安装tensorflow所需包的路径等问题运行如下：在tensorflow路径，
tensorflow hfes$ ./configure
here is an example excution of the configure script:
You have bazel 0.10.0-homebrew installed.
Please specify the location of python. [Default is /Users/bingxueguo/anaconda/bin/python]: /Users/bingxueguo/anaconda/bin/python
Found possible Python library paths:
/Users/bingxueguo/anaconda/lib/python3.5/site-packages
Please input the desired Python library path to use. Default is [/Users/bingxueguo/anaconda/lib/python3.5/site-packages]
/Users/bingxueguo/anaconda/lib/python3.5/site-packages
Do you wish to build TensorFlow with Google Cloud Platform support? [Y/n]: Y
Google Cloud Platform support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Hadoop File System support? [Y/n]: Y
Hadoop File System support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Amazon S3 File System support? [Y/n]: Y
Amazon S3 File System support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Apache Kafka Platform support? [y/N]: Y
Apache Kafka Platform support will be enabled for TensorFlow.
Do you wish to build TensorFlow with XLA JIT support? [y/N]: Y
XLA JIT support will be enabled for TensorFlow.
Do you wish to build TensorFlow with GDR support? [y/N]: Y
GDR support will be enabled for TensorFlow.
Do you wish to build TensorFlow with VERBS support? [y/N]: n
No VERBS support will be enabled for TensorFlow.
Do you wish to build TensorFlow with OpenCL SYCL support? [y/N]: N
No OpenCL SYCL support will be enabled for TensorFlow.
Do you wish to build TensorFlow with CUDA support? [y/N]: N
No CUDA support will be enabled for TensorFlow.
Do you wish to build TensorFlow with MPI support? [y/N]: N
No MPI support will be enabled for TensorFlow.
Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native]: -march=native
Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: N
Not configuring the WORKSPACE for Android builds.
Preconfigured Bazel build configs. You can use any of the below by adding "--config=<>" to your build command. See tools/bazel.rc for more details.
--config=mkl # Build with MKL support.
--config=monolithic # Config for mostly static monolithic build.
--config=tensorrt # Build with TensorRT support.
Configuration finished
前面的选择Y/N会影响到之后的安装是否出现问题，
一开始我在这里选择了一个y
Do you wish to build TensorFlow with VERBS support? [y/N]: n
No VERBS support will be enabled for TensorFlow.
后面出现了如下问题：
./tensorflow/contrib/verbs/rdma.h:21:10: fatal error: 'infiniband/verbs.h' file not found
#include
1 error generated.
也是搜索查询了好久很多给的是linux的解决办法，于是我又重新运行./configure选择了n才解决。
4.编译目标程序（这个只要前面配置没什么问题基本到这一步不会出什么问题，只是时间得一二十分钟）
bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package
运行这一步后也会产生一个tensorflow的安装包(.whl)在文件夹/tmp/tensorflow_pkg/下面
5. 安装pip包
官网上给的是
sudo pip install /tmp/tensorflow_pkg/tensorflow-1.5.0-py2-none-any.whl
当然这个包的名称还是要根据自己前面运行生成的，由于我也是初接触这些，也是琢磨了一下，发现直接在网页里输入
///tmp/tensorflow_pkg/ 可以找到前面生成的安装包，下载下来放在对应的文件夹然后用pip 安装即可
pip install tensorflow-1.5.0-cp35-cp35m-macosx_10_6_x86_64.whl
现在运行这个就不会出现一开始提到的问题了：
>>> import tensorflow as tf
>>> hello = tf.constant("Hello,Tensorflow")
>>> sess = tf.Session()
>>> print(sess.run(hello))
b'Hello,Tensorflow'
运行速度也会加快