This repository contains:
-a small collection of simple probabilistic generative models for modeling time series data
-some MCMC algorithms (e.g., particle filter) for inference of marginal distributions at each time point
-an implementation of the forward-backward algorithm for sampling from the joint posterior distribution in time (this is a crucial step in the learning algorihthm, which is just EM)
-some experimental code for adapting the proposal distribution as the model parameters change during learning to speed up inference