[Numeric] ANN: hmm-hmatrix - efficient Hidden Markov Model implementation

[Henning Thielemann]

On Wed, 19 Aug 2015, Dominic Steinitz wrote:

> How general are the models that it can fit? For example, a Kalman filter 
> is an example of a hidden Markov model in which the updates are linear 
> and the errors are Gaussian.

My package is limited to a finite set of states. I don't think you can use 
it for Kalman filtering and so far I do not know, whether I can generalize 
it that way. Unfortunately, I have no experience with Kalman filters.