Unfortunately, the Tensorflow team stops releasing binary package for Mac OS with CUDA support since Tensorflow 1.2. This project provides off-the-shelf binary packages. Both Python 2.7 and 3.6 are supported now!

很不幸，Tensorflow团队自从1.2版本开始停止了发布 Mac OS CUDA版。本项目提供 Mac OS 上编译好、可直接安装的Tensorflow CUDA版本。本项目同时支持Python 2.7 和 3.6 了！

You can find older releases in the releases page.

你可以在Releases页面找到以前发布的版本。

First, ensure your CUDA driver and cudnn is installed properly, and copy dependencies in folder usr_local_lib to path /usr/local/lib.

首先，确保CUDA驱动和cudnn正确安装,并且将文件夹usr_local_lib中的依赖项复制到路径/usr/local/lib。

sudo mkdir /usr/local
sudo mkdir /usr/local/lib
sudo cp usr_local_lib/* /usr/local/lib/

Second, uninstall the previous tensorflow installtion by

其次，卸载之前版本的Tensorflow:

pip uninstall tensorflow
pip uninstall tensorflow-gpu

At last, uncompress and install the binary package from this project:

最后，解压并安装：

cat tensorflow* > tensorflow.whl.zip
unzip tensorflow.whl.zip
pip install tensorflow*.whl

Install Python 3.6 from Homebrew first, and then simply follow the guide for Python 2.7 and replace pip command with pip3 and python with python3.

首先从Homebrew安装Python 3.6，然后按照Python 2.7的安装步骤执行，注意将pip替换为pip3，并用python3启动python。

Enjoy!

开始使用新版Tensorflow吧!

Source code from: https://github.com/tensorflow/tensorflow