
Calculates the crossentropy between the two selected HMM models based on observation sequences.
The crossentropy is a measure of the distance between two models λ_{1} and λ_{2}. It is defined as
D(λ_{1},λ_{2}) = 1/N (log p(O_{2}λ_{1})  log p(O_{2}λ_{2})), 
where O_{2} is an observation sequence of length N generated by model λ_{2}.
The symmetrized version is:
D_{s}(λ_{1},λ_{2}) = (D(λ_{1},λ_{2}) + D(λ_{2},λ_{1}))/2. 
© djmw, October 17, 2010