编译TensorFlow源码

参考:

https://www.tensorflow.org/install/install_sources

https://github.com/tensorflow/tensorflow/blob/master/tensorflow/go/README.md

一 环境

ubuntu 16.04.2   (virtualbox 虚拟机)

二  安装 bazel

参考:https://docs.bazel.build/versions/master/install-ubuntu.html

Using Bazel custom APT repository (recommended)

1. Install JDK 8

Install JDK 8 by using:

sudo apt-get install openjdk--jdk

On Ubuntu 14.04 LTS you'll have to use a PPA:

sudo add-apt-repository ppa:webupd8team/java
sudo apt-get update && sudo apt-get install oracle-java8-installer

  

2. Add Bazel distribution URI as a package source (one time setup)

echo "deb [arch=amd64] http://storage.googleapis.com/bazel-apt stable jdk1.8" | sudo tee /etc/apt/sources.list.d/bazel.list
curl https://bazel.build/bazel-release.pub.gpg | sudo apt-key add -

If you want to install the testing version of Bazel, replace stable with testing.

3. Install and update Bazel

sudo apt-get update && sudo apt-get install bazel


Once installed, you can upgrade to a newer version of Bazel with:

sudo apt-get upgrade bazel

三 Python和Swig

sudo apt-get install python3-numpy python3-dev python3-pip python3-wheel  swig

四 下载源码及编译TensorFlow

github直接下载最新代码  https://github.com/tensorflow/tensorflow

终端切换到源码主目录,

./configure

涉及一些交互项

dell@dell-VirtualBox:~/tensorflow-master$ ./configure
WARNING: ignoring http_proxy in environment.
You have bazel 0.5. installed.
Please specify the location of python. [Default is /usr/bin/python]: /usr/bin/python3. Found possible Python library paths:
/usr/local/lib/python3./dist-packages
/usr/lib/python3/dist-packages
Please input the desired Python library path to use. Default is [/usr/local/lib/python3./dist-packages] Do you wish to build TensorFlow with jemalloc as malloc support? [Y/n]: y
jemalloc as malloc support will be enabled for TensorFlow. Do you wish to build TensorFlow with Google Cloud Platform support? [y/N]: n
No Google Cloud Platform support will be enabled for TensorFlow. Do you wish to build TensorFlow with Hadoop File System support? [y/N]: n
No Hadoop File System support will be enabled for TensorFlow. Do you wish to build TensorFlow with XLA JIT support? [y/N]: n
No XLA JIT support will be enabled for TensorFlow. Do you wish to build TensorFlow with GDR support? [y/N]: n
No 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 support? [y/N]: n
No OpenCL 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]: Add "--config=mkl" to your bazel command to build with MKL support.
Please note that MKL on MacOS or windows is still not supported.
If you would like to use a local MKL instead of downloading, please set the environment variable "TF_MKL_ROOT" every time before build.
Configuration finished

开始编译

bazel build --config opt //tensorflow:libtensorflow.so

耗时比较长,用了90多分钟。

./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ'
REGISTER_OP_GRADIENT_UNIQ(ctr, name, fn)
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ_HELPER'
REGISTER_OP_GRADIENT_UNIQ_HELPER(__COUNTER__, name, fn)
^
tensorflow/core/ops/nn_grad.cc::: note: in expansion of macro 'REGISTER_OP_GRADIENT'
REGISTER_OP_GRADIENT("MaxPool", MaxPoolGrad);
^
./tensorflow/core/framework/function.h::: warning: 'tensorflow::unused_grad_6' defined but not used [-Wunused-variable]
static bool unused_grad_##ctr = SHOULD_REGISTER_OP_GRADIENT && \
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ'
REGISTER_OP_GRADIENT_UNIQ(ctr, name, fn)
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ_HELPER'
REGISTER_OP_GRADIENT_UNIQ_HELPER(__COUNTER__, name, fn)
^
tensorflow/core/ops/nn_grad.cc::: note: in expansion of macro 'REGISTER_OP_GRADIENT'
REGISTER_OP_GRADIENT("AvgPool", AvgPoolGrad);
^
./tensorflow/core/framework/function.h::: warning: 'tensorflow::unused_grad_7' defined but not used [-Wunused-variable]
static bool unused_grad_##ctr = SHOULD_REGISTER_OP_GRADIENT && \
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ'
REGISTER_OP_GRADIENT_UNIQ(ctr, name, fn)
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ_HELPER'
REGISTER_OP_GRADIENT_UNIQ_HELPER(__COUNTER__, name, fn)
^
tensorflow/core/ops/nn_grad.cc::: note: in expansion of macro 'REGISTER_OP_GRADIENT'
REGISTER_OP_GRADIENT("MaxPoolGrad", MaxPoolGradGrad);
^
./tensorflow/core/framework/function.h::: warning: 'tensorflow::unused_grad_8' defined but not used [-Wunused-variable]
static bool unused_grad_##ctr = SHOULD_REGISTER_OP_GRADIENT && \
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ'
REGISTER_OP_GRADIENT_UNIQ(ctr, name, fn)
^
./tensorflow/core/framework/function.h::: note: in expansion of macro 'REGISTER_OP_GRADIENT_UNIQ_HELPER'
REGISTER_OP_GRADIENT_UNIQ_HELPER(__COUNTER__, name, fn)
^
tensorflow/core/ops/nn_grad.cc::: note: in expansion of macro 'REGISTER_OP_GRADIENT'
REGISTER_OP_GRADIENT("BiasAdd", BiasAddGrad);
^
Target //tensorflow:libtensorflow.so up-to-date:
bazel-bin/tensorflow/libtensorflow.so
INFO: Elapsed time: .039s, Critical Path: .77s
INFO: Build completed successfully, total actions

编译TensorFlow源码的更多相关文章

  1. Ubuntu TensorFlow 源码 Android Demo的编译运行

    Ubuntu TensorFlow 源码 Android Demo的编译运行 一. 安装 Android 的SDK和NDK SDK 配置 A:下载 国内下载地址选最新的: SDK: https://d ...

  2. tensorflow 源码编译tensorflow 1.1.0到 tensorflow 2.0,ver:1.1.0rc1、1.4.0rc1、1.14.0-rc1、2.0.0b1

    目录 tensorflow-build table 更多详细过程信息及下载: tensorflow-build tensorflow 源码编译,提升硬件加速,支持cpu加速指令,suport SSE4 ...

  3. TensorFlow源码安装

    前言 TensorFlow如果能二进制包安装,我真的不想选择自己编译,但是情况不由人,好不容易找到一台服务器,CPU不支持AVX指令集,安装的release版本运行到import tensorflow ...

  4. Tensorflow源码解析1 -- 内核架构和源码结构

    1 主流深度学习框架对比 当今的软件开发基本都是分层化和模块化的,应用层开发会基于框架层.比如开发Linux Driver会基于Linux kernel,开发Android app会基于Android ...

  5. Tensorflow[源码安装时bazel行为解析]

    0. 引言 通过源码方式安装,并进行一定程度的解读,有助于理解tensorflow源码,本文主要基于tensorflow v1.8源码,并借鉴于如何阅读TensorFlow源码. 首先,自然是需要去b ...

  6. [图解tensorflow源码] 入门准备工作附常用的矩阵计算工具[转]

    [图解tensorflow源码] 入门准备工作 附常用的矩阵计算工具[转] Link: https://www.cnblogs.com/yao62995/p/5773142.html  tensorf ...

  7. Ubuntu16.04系统Tensorflow源码安装

    最近学习Tensorflow,记录一下安装过程.目前安装的是CPU版的 1.下载tensorflow源码 tensorflow是个开源库,在github上有源码,直接在上面下载.下载地址:https: ...

  8. tensorflow源码分析

    前言: 一般来说,如果安装tensorflow主要目的是为了调试些小程序的话,只要下载相应的包,然后,直接使用pip install tensorflow即可. 但有时我们需要将Tensorflow的 ...

  9. Windows编译Nginx源码

    Windows下的Nginx战役,人不作就不会死!就像是拿着麦当劳的优惠券去买肯德基一样,别扭啊 Nginx是一款轻量级的Web 服务器.反向代理服务器.邮件服务器等等集一大串荣誉于一身的大牌人物!他 ...

随机推荐

  1. Beautifulsoup分解

    from urllib.request import Request, ProxyHandler from urllib.request import build_opener from bs4 im ...

  2. Python入门 - 时间处理

    本讲主要介绍python时间处理方式,比如获取当前的时间: import time ftime = time.strftime("%Y年%m月%d日%H点%M分%S秒", time ...

  3. Libevent 反应堆的初始化

    // 默认情况下new eventbase struct event_base * event_base_new(void) { struct event_base *base = NULL; //初 ...

  4. 使用SpringBoot开发REST服务

    本文介绍如何基于Spring Boot搭建一个简易的REST服务框架,以及如何通过自定义注解实现Rest服务鉴权 搭建框架 pom.xml 首先,引入相关依赖,数据库使用mongodb,同时使用red ...

  5. python自带库及第三方库api察看

    今天发现一个很有意思的功能,python自带了所有库的文档查看器,配置如下: 配置pydoc服务,cmd中输入如下代码: python –m pydoc –p 1234 回车后 ,使用过程中,该窗口不 ...

  6. zuul超时的解决方案

    参考http://www.coolxuewang.com/view/10 在zuul的配置文件里增加如下配置: ribbon:    ConnectTimeout: 6000    ReadTimeo ...

  7. 【下一代核心技术DevOps】:(五)微服务CI与Rancher持续集成

    1. 引言 DevOps的核心魅力是快速的持续集成交付,降低研发和实施运维之间的交互,使得传统的各种扯皮现象统统消失.最重要的是降低成本 保障产品交付可靠性. 使用Rancher作为持续集成的关键环节 ...

  8. 树的平衡 AVL Tree

    本篇随笔主要从以下三个方面介绍树的平衡: 1):BST不平衡问题 2):BST 旋转 3):AVL Tree 一:BST不平衡问题的解析 之前有提过普通BST的一些一些缺点,例如BST的高度是介于lg ...

  9. 最短路算法之Dijkstra算法通俗解释

    Dijkstra算法 说明:求解从起点到任意点的最短距离,注意该算法应用于没有负边的图. 来,看图. 用邻接矩阵表示 int[][] m = { {0, 0, 0, 0, 0, 0}, {0, 0, ...

  10. Testin云測试破7000万次:崩溃成90%手游应用质量难题

    Testin云測试破7000万次:崩溃成90%手游应用质量难题 2014/11/13 · Testin · 业界资讯 11月13日.全球最大的移动游戏.应用真机和用户云測试平台Testin云測宣布,已 ...