One can find the (probabilistic) distance between two HMMs (a distance similar to the Kullback-Leibler measure) using the KullbackLeiblerDistanceCalculator
class
[5].
This class generates n
long observations sequences of length l
using one HMM and finds their probabilities using the other. Parameters n
and l
can be set using the setNbSequences
and setSequencesLength
methods.
Pay attention to the fact that this distance measure is not symetric; see the code documentation for more details.