link_pot_to_cpt.m

The BNL toolbox is a set of Matlab functions for defining and estimating the parameters of a Bayesi

function  pot_to_CPT=link_pot_to_CPT(cliquetable,clq_ass_to_node,par)
%returns, for all conditional probability tables, the number of the clique from which it will be obtained
%, and the dimension nr of the clique to marginalyze over
nr_nodes=length(par);
pot_to_CPT.proj=cell(nr_nodes,1);%dimension nr of the nodes to sum over
pot_to_CPT.ind=clq_ass_to_node; %nr of clique to which node is assigned

for i=1:nr_nodes
   set=[par{i} i];  %i and its parents
   map=mysetdiff(cliquetable.ind{pot_to_CPT.ind(i)},set); %identify nodes in the same clique but not part of 'set'
   pot_to_CPT.proj{i}=find_equiv_posns(map,cliquetable.ind{pot_to_CPT.ind(i)});%find position nr of these nodes
end