ApBsInT is short for Approximate Bayesian Inference Toolkit.
It provides a lightweight implementation of a number of algorithms for
approximate Bayesian inference, applicable to a range of probabilistic models.
Buzzwords for what is contained right now:

   * Generalized linear models
   * Sparse (generalized) linear models
   * Expectation propagation variational Bayesian inference

See INSTALL.txt for more details.
The toolbox is written in Python, with a small part in C++. A part of the
functionality is replicated in Matlab (mainly for testing).

The most current development version is always available from our
git repository:

http://github.com/mseeger/apbsint