The routines use a common data structure 'chmm' with fields:chmm.P                  transition probability    .Pi                 initial state probability    .K                  dimension of state space (vector)    .LPtrain            final training log-posterior    .priors.Dir2d_alpha   2-D Dirichlet prior counts for Tx Probabilities           .Dir_alpha     Dirichlet prior counts for hidden Probabilities @t=0chmm.train.cyc		max number of cycles through data	  .tol		termination tolerance of likelihoodchmm.data.Xtrain	training data, 1st channel         .Ytrain        training data, 2nd channel	 .T		length of training sequence		chmm.hmmone and chmm.hmmtwo           .obsmodel	name of observation model			'Gauss'		- Gaussian			'GaussCom'	- Gaussian with common cov			'LIKE'          - observations are likelihoods	   .K		dimension of state-space	   .obsupdate	update observation model	   .updatep	update state transition matrix           .Pi 		initial state probability            .P		state transition probabilities                         (3D array b/c 2 parent nodes) 	   .init	initialised (1 or 0)           .gmmll	log-likelihood of gmm model used for initialisation           .mix		Gaussian mixture model trained on same data	   .train.init   	initialisation flag                 .obsupdate	update observation model (1 or 0)                 .pupdate	update transition matrix (1 or 0)	             .priors.Dir3d_alpha   3-D Dirichlet prior for Tx probabilities                 .Dir_alpha     Dirichlet prior for initial state probabilityFor 'Gauss' and 'GaussCom' observation models we also have:chmm.hmmone.state(k).Mu		mean vector for state k                    .Cov	mean covariance matrix for state k                    .a		ar coefficients for state k	            .priors     priors for each state                             .norm_Mu     Prior for mean: mean (1,dimension(data))                             .norm_Cov    Prior for mean: covariance                             .norm_Prec   Prior for mean: precision                             .Wish_alpha  Prior for Covariance: scale parameter                             .Wish_B      Prior for Covariance: shape matrix                             .Wish_k      Prior for Covariance: dimension of shape matrixFor 'AR' observation models we have:chmm.hmmone.state(k).p          order of AR modelchmm.hmmone.state(k).a          parameter vector for AR modelchmm.hmmone.state(k).v          noise variance for AR modelFor 'LIKE' observation models, there are no extra parameters.