wolffd@0: % Sigmoid Belief IOHMM
wolffd@0: % Here is the model
wolffd@0: %
wolffd@0: %  X \  X \
wolffd@0: %  | |  | |
wolffd@0: %  Q-|->Q-|-> ...
wolffd@0: %  | /  | /
wolffd@0: %  Y    Y
wolffd@0: %
wolffd@0: clear all;
wolffd@0: clc;
wolffd@0: rand('state',0); randn('state',0);
wolffd@0: X = 1; Q = 2; Y = 3;
wolffd@0: % intra time-slice graph
wolffd@0: intra=zeros(3);
wolffd@0: intra(X,[Q Y])=1;
wolffd@0: intra(Q,Y)=1;
wolffd@0: % inter time-slice graph
wolffd@0: inter=zeros(3);
wolffd@0: inter(Q,Q)=1;
wolffd@0: 
wolffd@0: ns = [1 3 1]; 
wolffd@0: dnodes = [2];
wolffd@0: eclass1 = [1 2 3];
wolffd@0: eclass2 = [1 4 3];
wolffd@0: bnet = mk_dbn(intra, inter, ns, dnodes, eclass1, eclass2);
wolffd@0: bnet.CPD{1} = root_CPD(bnet, 1);
wolffd@0: % ==========================================================
wolffd@0: bnet.CPD{2} = softmax_CPD(bnet, 2);
wolffd@0: bnet.CPD{4} = softmax_CPD(bnet, 5, 'discrete', [2]);
wolffd@0: % ==========================================================
wolffd@0: bnet.CPD{3} = gaussian_CPD(bnet, 3);
wolffd@0: 
wolffd@0: % make some data
wolffd@0: T=20;
wolffd@0: cases = cell(3, T);
wolffd@0: cases(1,:)=num2cell(round(rand(1,T)*2)+1);
wolffd@0: %cases(2,:)=num2cell(round(rand(1,T))+1);
wolffd@0: cases(3,:)=num2cell(rand(1,T));
wolffd@0: 
wolffd@0: engine = bk_inf_engine(bnet, 'exact', [1 2 3]);
wolffd@0: 
wolffd@0: % log lik before learning
wolffd@0: [engine, loglik] = enter_evidence(engine, cases);
wolffd@0: 
wolffd@0: % do learning
wolffd@0: ev=cell(1,1);
wolffd@0: ev{1}=cases;
wolffd@0: [bnet2, LL2] = learn_params_dbn_em(engine, ev, 3);