wolffd@0: function bnet = bayes_update_params(bnet, cases, clamped)
wolffd@0: % BAYES_UPDATE_PARAMS Bayesian parameter updating given completely observed data
wolffd@0: % bnet = bayes_update_params(bnet, cases, clamped)
wolffd@0: %
wolffd@0: % If there is a missing data, you must use EM.
wolffd@0: % cases(i,m) is the value assigned to node i in case m (this can also be a cell array).
wolffd@0: % clamped(i,m) = 1 if node i was set by intervention in case m (default: clamped = zeros).
wolffd@0: % Clamped nodes are not updated.
wolffd@0: % If there is a single case, clamped is a list of the clamped nodes, not a bit vector.
wolffd@0: 
wolffd@0: 
wolffd@0: %if iscell(cases), usecell = 1; else usecell = 0; end
wolffd@0: 
wolffd@0: n = length(bnet.dag);
wolffd@0: ncases = size(cases, 2);
wolffd@0: if n ~= size(cases, 1)
wolffd@0:   error('data must be of size nnodes * ncases');
wolffd@0: end
wolffd@0: 
wolffd@0: if ncases == 1 % clamped is a list of nodes
wolffd@0:   if nargin < 3, clamped = []; end
wolffd@0:   clamp_set = clamped;
wolffd@0:   clamped = zeros(n,1);
wolffd@0:   clamped(clamp_set) = 1;
wolffd@0: else % each row of clamped is a bit vector
wolffd@0:   if nargin < 3, clamped = zeros(n,ncases); end
wolffd@0: end
wolffd@0: 
wolffd@0: for i=1:n
wolffd@0:   e = bnet.equiv_class(i);
wolffd@0:   if adjustable_CPD(bnet.CPD{e})
wolffd@0:     u = find(clamped(i,:)==0);
wolffd@0:     ps = parents(bnet.dag, i);
wolffd@0:     bnet.CPD{e} = bayes_update_params(bnet.CPD{e}, cases(i,u), cases(ps,u));
wolffd@0:   end
wolffd@0: end
wolffd@0: 
wolffd@0: