wolffd@0: function mpe = find_mpe(engine, local_evidence)
wolffd@0: % FIND_MPE Find the most probable explanation of the data  
wolffd@0: % function mpe = find_mpe(engine, local_evidence
wolffd@0: %
wolffd@0: % local_evidence{i}(j) = Pr(observation at node i | S(i)=j)
wolffd@0: %
wolffd@0: % This finds the marginally most likely value for each hidden node.
wolffd@0: % It may give inconsistent results if there are ties.
wolffd@0: 
wolffd@0: [mpe, niter] = bp_mpe_mrf2(engine.mrf2.adj_mat, engine.mrf2.pot, local_evidence, ...
wolffd@0: 			   'max_iter', engine.max_iter, 'momentum', engine.momentum, ...
wolffd@0: 			   'tol', engine.tol);