首先下载Anaconda3-4.2.0（最好使用这个版本，本人用4.3.0尝试不成功），官方下载链接，https://www.continuum.io/downloads#windows，需要VPN，选择下载64位的3.5的版本。

一路next，安装完毕后，点击Anaconda Prompt，输入如下命令行，

注意，我们这里建议安装tensorflow1.0.0，该版本相对稳定，当然会有一些SSE加速的警告，按下面的方式忽略即可，而1.0.1会报一些Opkernel的错误。

> conda create -n tensorflow python=3.5
> activate tensorflow
> pip install http://ci.tensorflow/view/Nightly/job/nightly-win/85/DEVICE=cpu,OS=windows/artifact/cmake_build/tf_python/dist/tensorflow-1.0.0rc2-cp35-cp35m-win_amd64.whl
> python
>>> import tensorflow as tf
>>> import os
>>> os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
>>> print('TensorFlow version: {0}'.format(tf.__version__))
TensorFlow version: 1.0.0-rc2
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> print(sess.run(hello))
b'Hello, TensorFlow!'
>>> a = tf.constant(10)
>>> b = tf.constant(32)
>>> sess.run(a+b)
42
> deactivate tensorflow

TF_CPP_MIN_LOG_LEVEL:

·        0: all logs shown (that's the default setting)

·        1: filter out INFO logs

·        2: additionally filter out WARNING logs

·        3: additionally filter out ERROR logs.

reference:

https://www.tensorflow/install/install_windows

首先下载Anaconda3-4.2.0（最好使用这个版本，本人用4.3.0尝试不成功），官方下载链接，https://www.continuum.io/downloads#windows，需要VPN，选择下载64位的3.5的版本。

一路next，安装完毕后，点击Anaconda Prompt，输入如下命令行，

注意，我们这里建议安装tensorflow1.0.0，该版本相对稳定，当然会有一些SSE加速的警告，按下面的方式忽略即可，而1.0.1会报一些Opkernel的错误。

> conda create -n tensorflow python=3.5
> activate tensorflow
> pip install http://ci.tensorflow/view/Nightly/job/nightly-win/85/DEVICE=cpu,OS=windows/artifact/cmake_build/tf_python/dist/tensorflow-1.0.0rc2-cp35-cp35m-win_amd64.whl
> python
>>> import tensorflow as tf
>>> import os
>>> os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'
>>> print('TensorFlow version: {0}'.format(tf.__version__))
TensorFlow version: 1.0.0-rc2
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> print(sess.run(hello))
b'Hello, TensorFlow!'
>>> a = tf.constant(10)
>>> b = tf.constant(32)
>>> sess.run(a+b)
42
> deactivate tensorflow

TF_CPP_MIN_LOG_LEVEL:

·        0: all logs shown (that's the default setting)

·        1: filter out INFO logs

·        2: additionally filter out WARNING logs

·        3: additionally filter out ERROR logs.

reference:

https://www.tensorflow/install/install_windows