wolffd@0: function fg = mk_fgraph(G, node_sizes, factors, varargin)
wolffd@0: % MK_FGRAPH Make a factor graph
wolffd@0: % fg = mk_fgraph(G, node_sizes, factors, ...)
wolffd@0: %
wolffd@0: % A factor graph is a bipartite graph, with one side containing variables,
wolffd@0: % and the other containing functions of (subsets of) these variables.
wolffd@0: % For details, see "Factor Graphs and the Sum-Product Algorithm",
wolffd@0: %  F. Kschischang and B. Frey and H-A. Loeliger,
wolffd@0: %  IEEE Trans. Info. Theory, 2001
wolffd@0: %
wolffd@0: % G(i,j) = 1 if there is an arc from variable i to factor j
wolffd@0: %
wolffd@0: % node_sizes(i) is the number of values node i can take on,
wolffd@0: %   or the length of node i if i is a continuous-valued vector.
wolffd@0: %
wolffd@0: % 'factors' is the list of factors (kernel functions)
wolffd@0: %
wolffd@0: % The list below gives optional arguments [default value in brackets].
wolffd@0: % 
wolffd@0: % equiv_class - equiv_class(i)=j  means factor node i gets its params from factors{j} [1:F]
wolffd@0: % discrete - the list of nodes which are discrete random variables [1:N]
wolffd@0: %
wolffd@0: % e.g., fg = mk_fgraph(G, [2 2], {bnet.CPD{1},bnet.CPD{2}}, 'discrete', [1 2])
wolffd@0: 
wolffd@0: fg.G = G;
wolffd@0: fg.node_sizes = node_sizes;
wolffd@0: fg.factors = factors;
wolffd@0: [fg.nvars fg.nfactors] = size(G);
wolffd@0: 
wolffd@0: % default values for parameters
wolffd@0: fg.equiv_class = 1:fg.nfactors;
wolffd@0: fg.dnodes = 1:fg.nvars;
wolffd@0: 
wolffd@0: if nargin >= 4
wolffd@0:   args = varargin;
wolffd@0:   nargs = length(args);
wolffd@0:   for i=1:2:nargs
wolffd@0:     switch args{i},
wolffd@0:      case 'equiv_class', fg.equiv_class = args{i+1}; 
wolffd@0:      case 'discrete',    fg.dnodes = args{i+1}; 
wolffd@0:      otherwise,  
wolffd@0:       error(['invalid argument name ' args{i}]);       
wolffd@0:     end
wolffd@0:   end
wolffd@0: end
wolffd@0: 
wolffd@0: % so that determine_pot_type will work...
wolffd@0: fg.utility_nodes = [];
wolffd@0: %fg.decision_nodes = [];
wolffd@0: %fg.chance_nodes = fg.nvars;
wolffd@0: 
wolffd@0: fg.dom = cell(1, fg.nfactors);
wolffd@0: for f=1:fg.nfactors
wolffd@0:   fg.dom{f} = find(G(:,f));
wolffd@0: end
wolffd@0: fg.dep = cell(1, fg.nvars);
wolffd@0: for x=1:fg.nvars
wolffd@0:   fg.dep{x} = find(G(x,:));
wolffd@0: end
wolffd@0: fg.cnodes = mysetdiff(1:fg.nvars, fg.dnodes);