离线编译Tensorflow

2017-02-01 约 678 字预计阅读 2 分钟

概述

关于编译各种项目的源码，很多人都深有体会，因为公司都会以安全为由，不管是开发机还是 IDC 的生产服务器，一般都会直接屏蔽外网。但是很多编译工具或者方案，都需要去下载一些外部文件，这样导致在开发机上想编译个什么都会很麻烦。

当然还有一种选择就是离线编译，那么意味着必须把编译时候需要的所有依赖文件都提前下载好，让编译工具能够在编译过程找到这些必须的文件。所以，找出来依赖的外部文件，就可以做到离线编译了。

Tensorflow的编译

如同官网说的，为 Tensorflow 编译 GPU 支持最简单的方法就是使用 Docker！主机只需要安装了 NVIDIA 的驱动以及 CUDA 包，安装好 nvidia-docker，就可以编译了，当然离线编译的话，还需要解决依赖的外部文件。

因为编译 Tensorflow 的工具是 Bazel，我想大部分人都不太熟悉，包括我自己，但是要找出依赖的外部文件，还是需要稍微了解一下的。

Bazel 是谷歌内部编译 C++ 项目的工具，Tensorflow 有一个 WORKSPACE 的文件，关于他是什么含义，可以看看官网的介绍。

另外，再看看 WORKSPACE 文件的内容。

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81


workspace(name = "org_tensorflow")

http_archive(
    name = "io_bazel_rules_closure",
    sha256 = "a38539c5b5c358548e75b44141b4ab637bba7c4dc02b46b1f62a96d6433f56ae",
    strip_prefix = "rules_closure-dbb96841cc0a5fb2664c37822803b06dab20c7d1",
    urls = [
        "https://mirror.bazel.build/github.com/bazelbuild/rules_closure/archive/dbb96841cc0a5fb2664c37822803b06dab20c7d1.tar.gz",
        "https://github.com/bazelbuild/rules_closure/archive/dbb96841cc0a5fb2664c37822803b06dab20c7d1.tar.gz",  # 2018-04-13
    ],
)

load("@io_bazel_rules_closure//closure:defs.bzl", "closure_repositories")

closure_repositories()

# We must check the bazel version before trying to parse any other BUILD
# files, in case the parsing of those build files depends on the bazel
# version we require here.
load("//tensorflow:version_check.bzl", "check_bazel_version_at_least")
check_bazel_version_at_least("0.15.0")

load("//tensorflow:workspace.bzl", "tf_workspace")

load("//third_party/android:android_configure.bzl", "android_configure")
android_configure(name="local_config_android")
load("@local_config_android//:android.bzl", "android_workspace")
android_workspace()

# Please add all new TensorFlow dependencies in workspace.bzl.
tf_workspace()

new_http_archive(
    name = "inception_v1",
    build_file = "models.BUILD",
    sha256 = "7efe12a8363f09bc24d7b7a450304a15655a57a7751929b2c1593a71183bb105",
    urls = [
        "http://storage.googleapis.com/download.tensorflow.org/models/inception_v1.zip",
        "http://download.tensorflow.org/models/inception_v1.zip",
    ],
)

new_http_archive(
    name = "mobile_ssd",
    build_file = "models.BUILD",
    sha256 = "bddd81ea5c80a97adfac1c9f770e6f55cbafd7cce4d3bbe15fbeb041e6b8f3e8",
    urls = [
        "http://storage.googleapis.com/download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_android_export.zip",
        "http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_android_export.zip",
    ],
)

new_http_archive(
    name = "mobile_multibox",
    build_file = "models.BUILD",
    sha256 = "859edcddf84dddb974c36c36cfc1f74555148e9c9213dedacf1d6b613ad52b96",
    urls = [
        "http://storage.googleapis.com/download.tensorflow.org/models/mobile_multibox_v1a.zip",
        "http://download.tensorflow.org/models/mobile_multibox_v1a.zip",
    ],
)

new_http_archive(
    name = "stylize",
    build_file = "models.BUILD",
    sha256 = "3d374a730aef330424a356a8d4f04d8a54277c425e274ecb7d9c83aa912c6bfa",
    urls = [
        "http://storage.googleapis.com/download.tensorflow.org/models/stylize_v1.zip",
        "http://download.tensorflow.org/models/stylize_v1.zip",
    ],
)

new_http_archive(
    name = "speech_commands",
    build_file = "models.BUILD",
    sha256 = "c3ec4fea3158eb111f1d932336351edfe8bd515bb6e87aad4f25dbad0a600d0c",
    urls = [
        "http://storage.googleapis.com/download.tensorflow.org/models/speech_commands_v0.01.zip",
        "http://download.tensorflow.org/models/speech_commands_v0.01.zip",
    ],
)

参考资料

警告

本文最后更新于 2017年2月1日，文中内容可能已过时，请谨慎参考。

💡赞赏支持

微信打赏

支付宝打赏

目录

离线编译Tensorflow

概述

Tensorflow的编译

参考资料