evidence+theory 源代码

% a multigram is a degenerate 2HHMM where the bottom level HMMs emit deterministic strings % and the the top level abstract state…

% like mgram1, except we use a durational HMM instead of an HHMM2 past = 0; words = {'the', 't', 'h', 'e'}; data = 'the'; …

% like mgram2, except we unroll the DBN so we can use smaller % state spaces for the early duration nodes: % the state spaces ar…

% Like a durational HMM, except we use soft evidence on the observed nodes. % Should give the same results as HSMM/test_mgram2. …

if 0 % Generate some sample paths bnet = mk_map_hhmm('p', 1); % assign numbers to the nodes in topological order U = 1; A = …

seed = 1; rand('state', seed); randn('state', seed); obs_model = 'unique'; % each cell has a unique label (essentially fully…

% Generate samples from the HHMM with the true params. seed = 1; rand('state', seed); randn('state', seed); discrete_obs =…

% Generate samples from the HHMM with the true params. seed = 0; rand('state', seed); randn('state', seed); discrete_obs =…

seed = 0; rand('state', seed); randn('state', seed); discrete_obs = 1; topright = 0; Qsizes = [2 4 2]; D = 3; Qnodes = …

% Compute Viterbi path discrete HMM by different methods intra = zeros(2); intra(1,2) = 1; inter = zeros(2); inter(1,1) = 1;…