The ghmm plugin has two different constructors:
ghmm( )
ghmm( N, B )
where the first one creates an empty hierarchical gilbert hidden Markov model
(it assumes its parameters will be loaded hereafter); and the second one creates
a GHMM with N hidden states, whose observation batch size is B, and initializes its parameters with random values, satisfying the
stochastic constrains. Each of the states is associated with a unique
integer number, ranging from to , which identifies the state. The
observations emitted by the model are either or .