matthiasm@8: function count = compute_counts(data, sz)
matthiasm@8: % COMPUTE_COUNTS Count the number of times each combination of discrete assignments occurs
matthiasm@8: % count = compute_counts(data, sz)
matthiasm@8: %
matthiasm@8: % data(i,t) is the value of variable i in case t
matthiasm@8: % sz(i) : values for variable i are assumed to be in [1:sz(i)]
matthiasm@8: %
matthiasm@8: % Example: to compute a transition matrix for an HMM from a sequence of labeled states:
matthiasm@8: % transmat = mk_stochastic(compute_counts([seq(1:end-1); seq(2:end)], [nstates nstates]));
matthiasm@8: 
matthiasm@8: assert(length(sz) == size(data, 1));
matthiasm@8: P = prod(sz);
matthiasm@8: indices = subv2ind(sz, data'); % each row of data' is a case 
matthiasm@8: %count = histc(indices, 1:P);
matthiasm@8: count = hist(indices, 1:P);
matthiasm@8: count = myreshape(count, sz);
matthiasm@8: