Software written in Matlab:
New!
www.variationalbayes.org contains software for learning
various models using variational approximations to Bayesian
learning
For modelling independent identically distributed (iid) data:
For modelling time series data:
[*] To learn more about Variational Bayesian Learning, see: Beal,
M. J. and Ghahramani, Z. (2002) The
Variational Bayesian EM Algorithm for Incomplete Data: with
Application to Scoring Graphical Model Structures
Other software:

DELVE:
Data for Evaluating Learning in Valid Experiments  is a
collection of datasets, an environment within which this data can be
used to assess the performance of learning methods, and a repository
for the results of such assessments. This software and the
documentation were written by the Delve Development
Group.