Development

Post meeting Notes

A brief summary of what was decided during the meeting on Sunday 12 March '07

Future Version planning

In this page we list the features that will be included in the next version of OpenBayes, as well as future versions.

OpenBayes version 2 : Issues for discussion

These are the main issues that will be discussed during the skype meeting that will take place on Sunday 11 March 2007 at 2pm central America Time (9-10pm central europe time)

OpenBayes GUI

 

UML Diagrams

The UML Diagrams of OpenBayes

BNController Draft

While testing and developing Bayesian networks using OpenBayes, I needed some kind of higher level wrapper to integrate those into my "Rational Agent". BNController is my first draft of such a class, that is suppose to ease the definition of a Bayesian Network using OpenBayes, as well as providing essential functions such as load/save. Please feel free to comment - it is a first draft and definitely rough around the edges.

Continuous Distributions

For the moment only Gaussian Distributions are available and can only be used with the MCMC Inference Engine. Plans are to develop full Gaussian support for any inference engine (and learning engine) as well as mixture of gaussians which will allow to approximate any distribution by using a sum of gaussians. (Developper : Kosta Gaitanis)

Dynamic Bayesian Networks

Dynamic Bayesian Networks is a MUST for any bayesian library. Although this feature has started to be implemented, it seems that the guy responsible has left it over. Any help for developing this feature is more than welcome. Assistance will be guaranteed by myself.

EM-algorithm

An implementation of my EM-algorithm, instance of the EMLearningEngine of François + added the possibility to use augmented Bayesian nets to the MultinomialDistribution class.

Visualize your Bayes Net using pydot

This is a small script that can visualize your Bayes Net's CPT.

LearnMLParams for Jointrees using augmented bayes nets

I am trying to override the LearnMLParams of the inference.py so that it takes into account the augmented Multinomial Distribution proposed by wannes ... any ideas, suggestions, tests?

Learning

This zip folder contains the numarray and excell package, and the openbayes version with the learning algorithms. The algorithms works with those packages and versions, but I don't guarantee it works with other versions.