RoBO - a Robust Bayesian Optimization framework.

Master Branch

Installation

RoBO uses the Gaussian processes library george and the random forests library pyrfr. In order to use these libraries make sure that libeigen and swig are installed:

sudo apt-get install libeigen3-dev swig 

Download RoBO and then change into the new directory:

git clone https://github.com/automl/RoBO
cd RoBO/

Install the required dependencies.

for req in $(cat requirements.txt); do pip install $req; done

Finally install RoBO by:

python setup.py install

Documentation

You can find the documentation for RoBO here http://automl.github.io/RoBO/

Citing RoBO

To cite RoBO please reference our BayesOpt paper:

@INPROCEEDINGS{klein-bayesopt17,
author    = {A. Klein and S. Falkner and N. Mansur and F. Hutter},
title     = {RoBO: A Flexible and Robust Bayesian Optimization Framework in Python},
booktitle = {NIPS 2017 Bayesian Optimization Workshop},
year      = {2017},
month     = dec,
}