windows7下基于Anaconda安装TensorFlow

$ anaconda search -t conda tensorflow

Packages:

     Name                      |  Version | Package Types   | Platforms       | Builds

     ------------------------- |   ------ | --------------- | --------------- | ----------

     GlaxoSmithKline/tensorflow |   0.12.0 | conda           | linux-64        | py27hb0d0e74_0

                                          : TensorFlow is a machine learning library

     HCC/tensorflow            |    1.4.0 | conda           | linux-64        | py27_1, py34_1, py34_0, py36_0, py27_0, py35_0, py35_1

                                          : Computation using data flow graphs for scalable machine learning.

     HCC/tensorflow-cpucompat  |    1.4.0 | conda           | linux-64        | py36_0, py34_0, py35_0, py27_0

                                          : Computation using data flow graphs for scalable machine learning.

     HCC/tensorflow-fma        |    1.4.0 | conda           | linux-64        | py27_1, py34_1, py34_0, py36_0, py27_0, py35_0, py35_1

                                          : Computation using data flow graphs for scalable machine learning.

     SentientPrime/tensorflow  |    0.6.0 | conda           | osx-64          | py27_0

                                          : TensorFlow helps the tensors flow

     SmartAg/tensorflow_gpu    |    1.0.1 | conda           | linux-aarch64   | 0

     aaronzs/tensorflow        |    1.4.0 | conda           | linux-64, win-64, osx-64 | py36h39705f4_0, py36h8a03e48_0, py35hc784f49_0, py36h6db853c_0, py35h2d7a08b_0, py35h1150644_0, py35h5a8cc8b_0, py35hc0f5839_0, py36hebc11a6_0, py35ha700c16_0, py36hf8f6b73_0,
py36heb185b1_0, py36he4e0f4f_0, py35hf9a0815_0, py36h2003710_0, py36h4df9c7b_0, py35h6467dd0_0, py36hd42d972_0, py35h89e3332_0

                                          : TensorFlow helps the tensors flow

$ anaconda show r/r-tensorflow

Name:    r-tensorflow

Summary: Interface to 'TensorFlow' <https://www.tensorflow.org/>,  an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations,  while the graph edges represent the multidimensional
data arrays  (tensors) communicated between them. The flexible architecture allows you to deploy computation to one or more 'CPUs' or 'GPUs' in a desktop,  server, or mobile device with a single 'API'. 'TensorFlow' was originally developed by researchers and
engineers working on the Google Brain Team  within Google's Machine Intelligence research organization for the  purposes of conducting machine learning and deep neural networks research, but the system is general enough to be applicable in a wide variety of
other domains as well.

Access:  public

Package Types:  conda

Versions:

   + 0.4.0

   + 0.8.2

   + 1.4

To install this package with conda run:

     conda install --channel https://conda.anaconda.org/r r-tensorflow

Using Anaconda API: https://api.anaconda.org
$  conda install --channel https://conda.anaconda.org/r r-tensorflow

Fetching package metadata ...............

Solving package specifications: .

Package plan for installation in environment D:\programs\Anaconda3:

The following NEW packages will be INSTALLED:

    m2w64-bwidget:           1.9.10-2

    m2w64-bzip2:             1.0.6-6

    m2w64-expat:             2.1.1-2

    m2w64-fftw:              3.3.4-6

    m2w64-flac:              1.3.1-3

    m2w64-gcc-libgfortran:   5.3.0-6

.....

关于使用GPU，传送门：

[手把手在Windows上安装TensorFlow无坑教程]

[http://blog.sina.com.cn/s/blog_14935c5880102wu86.html]

[再写博文，回顾在Windows7上安装TensorFlow-GPU的一路坑]

[http://blog.csdn.net/infovisthinker/article/details/54705826]

(可选) 启用 GPU 支持 （*linux）

如果你使用 pip 二进制包安装了开启 GPU 支持的 TensorFlow, 你必须确保系统里安装了正确的 CUDA sdk 和 CUDNN 版本. 请参间CUDA 安装教程

你还需要设置

LD_LIBRARY_PATH

和

CUDA_HOME

环境变量. 可以考虑将下面的命令添加到

~/.bash_profile

文件中, 这样每次登陆后自动生效. 注意, 下面的命令假定 CUDA 安装目录为

/usr/local/cuda

:

export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib64"
export CUDA_HOME=/usr/local/cuda