HasGP is a library, implemented in Haskell, providing algorithms for supervised regression and classification using Gaussian processes (GPs). The standard text describing the use of GPs for machine learning can be found here.
HasGP is an experimental library, the aim of which is to explore functional programming as a means of implementing machine learning systems as opposed to the predominant imperative/object-oriented approach. It is not intended to be a replacement for the numerous libraries that employ the latter paradigm, a list of which can be found here. A detailed explanation of why the library is under development can be found in .
The library is open-source, and is available under the GPL3 license. The current release is the initial experimental release, version 0.1.
HasGP has a user manual explaining how to install and use it. It also provides an introduction to how purely functional techniques have been employed in the implementation. More detailed documentation generated by Haddock from the source code can either be generated during installation or consulted here.
 HasGP: A Haskell Library for Gaussian Process Inference, Sean B. Holden, September 2011, submitted.