这里要为仅支持 CPU 的 TensorFlow 构建一个 pip 软件包,需要调用以下命令:
$ bazel build --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --config=opt --verbose_failures //tensorflow/tools/pip_package:build_pip_package
 
大概编译15分钟后,输出错误信息如下:
ERROR: /private/var/tmp/_bazel_mazhiyong/54650f23f4cec88510e7895c92a68913/external/protobuf_archive/BUILD:659:1: C++ compilation of rule '@protobuf_archive//:python/google/protobuf/pyext/_message.so' failed (Exit 1)
external/protobuf_archive/python/google/protobuf/pyext/message.cc:547:11: error: assigning to 'char *' from incompatible type 'const char *'
      if (PyString_AsStringAndSize(key, &key_str_data, &key_str_size) != 0)
          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:82:22: note: expanded from macro 'PyString_AsStringAndSize'
       ((*(charpp) = PyUnicode_AsUTF8AndSize(ob, (sizep))) == NULL? -1: 0): \
                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:998:7: error: assigning to 'char *' from incompatible type 'const char *'
  if (PyString_AsStringAndSize(name, &field_name, &size) < 0) {
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:82:22: note: expanded from macro 'PyString_AsStringAndSize'
       ((*(charpp) = PyUnicode_AsUTF8AndSize(ob, (sizep))) == NULL? -1: 0): \
                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:1035:9: error: assigning to 'char *' from incompatible type 'const char *'
    if (PyString_AsStringAndSize(value, &enum_label, &size) < 0) {
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:82:22: note: expanded from macro 'PyString_AsStringAndSize'
       ((*(charpp) = PyUnicode_AsUTF8AndSize(ob, (sizep))) == NULL? -1: 0): \
                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:1532:16: error: assigning to 'char *' from incompatible type 'const char *'
  field_name = PyUnicode_AsUTF8AndSize(arg, &size);
               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:2048:7: warning: 'PyObject_AsReadBuffer' is deprecated [-Wdeprecated-declarations]
  if (PyObject_AsReadBuffer(arg, &data, &data_length) < 0) {
      ^
bazel-out/darwin-opt/genfiles/external/local_config_python/python_include/abstract.h:492:39: note: 'PyObject_AsReadBuffer' has been explicitly marked deprecated here
                                      Py_DEPRECATED(3.0);
                                      ^
bazel-out/darwin-opt/genfiles/external/local_config_python/python_include/pyport.h:493:54: note: expanded from macro 'Py_DEPRECATED'
#define Py_DEPRECATED(VERSION_UNUSED) __attribute__((__deprecated__))
                                                     ^
external/protobuf_archive/python/google/protobuf/pyext/message.cc:2119:7: error: assigning to 'char *' from incompatible type 'const char *'
  if (PyString_AsStringAndSize(arg, &name_data, &name_size) < 0)
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
external/protobuf_archive/python/google/protobuf/pyext/message.cc:82:22: note: expanded from macro 'PyString_AsStringAndSize'
       ((*(charpp) = PyUnicode_AsUTF8AndSize(ob, (sizep))) == NULL? -1: 0): \
                     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 warning and 5 errors generated.
Target //tensorflow/tools/pip_package:build_pip_package failed to build
Use --verbose_failures to see the command lines of failed build steps.
INFO: Elapsed time: 771.278s, Critical Path: 26.15s
INFO: 2156 processes: 2156 local.
FAILED: Build did NOT complete successfully
 
这个错误我估计是编译器与代码兼容的问题,查找了半天暂时没找到解决办法,所以就暂时搁置了TensorFlow主分支代码在Python3.7环境下的代码编译,过一段时间再试试。

TensorFlow Python3.7环境下的源码编译(三)编译的更多相关文章

  1. TensorFlow Python3.7环境下的源码编译(二)安装配置

    源代码树的根目录中包含了一个名为 configure 的 bash 脚本. $ ./configure 接下来,配置系统会给出各种询问,以确认编译时的配置参数.   一.重要参数解释 Please s ...

  2. TensorFlow Python3.7环境下的源码编译(一)环境准备

    参考: https://blog.csdn.net/yhily2008/article/details/79967118 https://tensorflow.google.cn/install/in ...

  3. TensorFlow Python2.7环境下的源码编译(三)编译

    一.源代码编译 这里要为仅支持 CPU 的 TensorFlow 构建一个 pip 软件包,需要调用以下命令: $ bazel build --cxxopt="-D_GLIBCXX_USE_ ...

  4. TensorFlow Python2.7环境下的源码编译(一)环境准备

    参考: https://blog.csdn.net/yhily2008/article/details/79967118 https://tensorflow.google.cn/install/in ...

  5. TensorFlow Python2.7环境下的源码编译(二)安装配置

    源代码树的根目录中包含了一个名为 configure 的 bash 脚本. $ ./configure 接下来,配置系统会给出各种询问,以确认编译时的配置参数.   一.重要参数解释 Do you w ...

  6. Linux环境下levelDB源码编译与安装

    1.下载源码并编译 git clone https://github.com/google/leveldb.git cd leveldb //编译源码的时候需要安装cmake,并且版本需要大于3.9, ...

  7. 23 使用环境 UsageEnvironment——Live555源码阅读

    23 使用环境 UsageEnvironment——Live555源码阅读(三)UsageEnvironment 23 使用环境 UsageEnvironment——Live555源码阅读(三)Usa ...

  8. 在docker容器中python3.5环境下使用DIGITS训练caffe模型

    ********* 此处使用的基础镜像为 nvcr.io/nvidia/digits:18.06,镜像大小为6.04GB,可从nvidia官方pull此镜像: 容器配置: CUDA:9.0 CUDNN ...

  9. 基于Docker的TensorFlow机器学习框架搭建和实例源码解读

    概述:基于Docker的TensorFlow机器学习框架搭建和实例源码解读,TensorFlow作为最火热的机器学习框架之一,Docker是的容器,可以很好的结合起来,为机器学习或者科研人员提供便捷的 ...

随机推荐

  1. XtraEditors三、LookUpEdit、GridLookUpEdit、SearchLookUpEdit

    https://documentation.devexpress.com/WindowsForms/DevExpress.XtraEditors.LookUpEditBase.class 一.Look ...

  2. 《Python核心编程》笔记

    1 python是大小写敏感的 2 遍历一个字典的键值: for a in dict_obj.keys(): print a 3 列表解析功能可以让代码很简洁,比如:squared = [x ** 2 ...

  3. (转)em重建全过程

    该问题遇到N次,被郁闷N次,特此记录以备不时之需 由于n久不用em,而本机在公司使用dhcp自动获取ip,导致ip变化,而使em启动报出ora-12514 DBD ERROR: OCIServerAt ...

  4. Java虚拟机17:互斥同步、锁优化及synchronized和volatile

    互斥同步 互斥同步(Mutual Exclusion & Synchronization)是常见的一种并发正确性保证手段.同步是指子啊多个线程并发访问共享数据时,保证共享数据在同一时刻只能被一 ...

  5. Uva1395 POJ3522 Slim Span (最小生成树)

    Description Given an undirected weighted graph G, you should find one of spanning trees specified as ...

  6. CF613D:Kingdom and its Cities(树形DP,虚树)

    Description 一个王国有n座城市,城市之间由n-1条道路相连,形成一个树结构,国王决定将一些城市设为重要城市. 这个国家有的时候会遭受外敌入侵,重要城市由于加强了防护,一定不会被占领.而非重 ...

  7. 【转】JS实现继承的几种方式

    既然要实现继承,那么首先我们得有一个父类,代码如下: // 定义一个动物类 function Animal (name) { // 属性 this.name = name || 'Animal'; / ...

  8. Docker实战(八)之Web服务与应用

    1.Apache 官方提供了名为httpd的Apache镜像,可以作为基础web服务镜像 Dockerfile(安装apache2) FROM httpd:2.4 COPY ./public-html ...

  9. Python django 404页面配置和debug=false 静态文件配置 django版本1.10.5

    django设置404页面 1.设置settings文件 DEBUG = False ALLOWED_HOSTS = ['127.0.0.1', 'localhost']或者 ALLOWED_HOST ...

  10. POJ 2250 (LCS,经典输出LCS序列 dfs)

    题目链接: http://poj.org/problem?id=2250 Compromise Time Limit: 1000MS   Memory Limit: 65536K Total Subm ...