2016-07-28 20:47:37 -07:00
# Installation on Mac OS X
2016-07-08 20:03:21 -07:00
2016-11-12 19:37:20 -08:00
Ray should work with Python 2 and Python 3. We have tested Ray on OS X 10.11.
2016-07-08 20:03:21 -07:00
2016-07-28 20:47:37 -07:00
## Dependencies
2016-07-08 20:03:21 -07:00
2016-11-12 19:37:20 -08:00
To install Ray, first install the following dependencies. We recommend using
[Anaconda ](https://www.continuum.io/downloads ).
2016-07-08 20:03:21 -07:00
```
brew update
2016-11-12 19:37:20 -08:00
brew install cmake automake autoconf libtool boost
sudo easy_install pip # If you're using Anaconda, then this is unnecessary.
pip install numpy funcsigs colorama psutil redis --ignore-installed six
pip install --upgrade git+git://github.com/cloudpipe/cloudpickle.git@0d225a4695f1f65ae1cbb2e0bbc145e10167cce4 # We use the latest version of cloudpickle because it can serialize named tuples.
2016-11-13 16:23:28 -08:00
pip install --upgrade --verbose git+git://github.com/ray-project/numbuf.git@488f881d708bc54e86ed375ee97aa94540808fa1
2016-07-08 20:03:21 -07:00
```
2016-11-12 19:37:20 -08:00
# Install Ray
2016-07-08 20:03:21 -07:00
2016-11-12 19:37:20 -08:00
Ray can be installed with pip as follows.
2016-07-08 20:03:21 -07:00
```
2016-11-12 19:37:20 -08:00
pip install --upgrade --verbose "git+git://github.com/ray-project/ray.git#egg =ray& subdirectory=lib/python"
2016-07-08 20:03:21 -07:00
```
2016-11-12 19:37:20 -08:00
Alternatively, Ray can be built from the repository as follows.
2016-07-08 20:03:21 -07:00
2016-11-12 19:37:20 -08:00
```
git clone https://github.com/ray-project/ray.git
cd lib/python
python setup.py install
```
2016-07-08 20:03:21 -07:00
2016-11-12 19:37:20 -08:00
## Test if the installation succeeded
To test if the installation was successful, try running some tests. This assumes
that you've cloned the git repository.
2016-07-08 20:03:21 -07:00
```
2016-11-12 19:37:20 -08:00
python test/runtest.py
2016-07-08 20:03:21 -07:00
```