Javier González


The following software, used in my publications, can be downloaded from my GitHub account. Most of it is written in R, Matlab and Python. Please contact me if you have any question or comment.

GPy and GPyOpt: Bayesian optimization with Gaussian processes

I am a core developer of GPyOpt a python framework for global optimization using Gaussian processes. This software is part of GPy, the python Gaussian Process framework of the Machine Learning group of the University of Sheffield. GPyOpt is under continuous improvement and it contains standard Bayesian optimization methods as well as the last contributions of our group related to this topic. Both GPy and GPyOpt are available under the BSD 3-clause license.

dpp: sampling from determinantal point processes

dpp is a small Python package to sample from determinantal point processes. It is based on the Matlab code of Alex Kulesza

Odest: inference of parameters of dynamical systems

odest is a R-package estimating parameters of systems of Ordinary Differential equations. The method is based on a regularization approach in Reproducing kernel Hilbert Spaces.

RobustPLS: Robust high dimensional regression

RobustPLS is a Matlab package for robust high dimensional Partial Least Squares Regression.

Tools for Science dissemination

There exist some tools that make research dissemination a very easy and pleasant experience. I am particularly fan of two of them: the IPython notebook (or Jupyther, its recent language agnostic version) and R-shiny. Below, you can check some examples. New demonstrations coming soon!

  • Tutorial in Bayesian optimization with different acquisition functions [Ipython notebook].
  • Writing kernels in GPy with Theano [Ipython notebook].
  • Bayesian Optimization with GPyOpt [Ipython notebook].