Mercurial > hg > camir-aes2014
annotate toolboxes/FullBNT-1.0.7/bnt/inference/static/@gaussian_inf_engine/gaussian_inf_engine.m @ 0:e9a9cd732c1e tip
first hg version after svn
author | wolffd |
---|---|
date | Tue, 10 Feb 2015 15:05:51 +0000 |
parents | |
children |
rev | line source |
---|---|
wolffd@0 | 1 function engine = gaussian_inf_engine(bnet) |
wolffd@0 | 2 % GAUSSIAN_INF_ENGINE Computes the joint multivariate Gaussian corresponding to the bnet |
wolffd@0 | 3 % engine = gaussian_inf_engine(bnet) |
wolffd@0 | 4 % |
wolffd@0 | 5 % For details on how to compute the joint Gaussian from the bnet, see |
wolffd@0 | 6 % - "Gaussian Influence Diagrams", R. Shachter and C. R. Kenley, Management Science, 35(5):527--550, 1989. |
wolffd@0 | 7 % Once we have the Gaussian, we can apply the standard formulas for conditioning and marginalization. |
wolffd@0 | 8 |
wolffd@0 | 9 assert(isequal(bnet.cnodes, 1:length(bnet.dag))); |
wolffd@0 | 10 |
wolffd@0 | 11 [W, D, mu] = extract_params_from_gbn(bnet); |
wolffd@0 | 12 U = inv(eye(size(W)) - W')'; |
wolffd@0 | 13 Sigma = U' * D * U; |
wolffd@0 | 14 |
wolffd@0 | 15 engine.mu = mu; |
wolffd@0 | 16 engine.Sigma = Sigma; |
wolffd@0 | 17 %engine.logp = log(normal_coef(Sigma)); |
wolffd@0 | 18 |
wolffd@0 | 19 % This is where we will store the results between enter_evidence and marginal_nodes |
wolffd@0 | 20 engine.Hmu = []; |
wolffd@0 | 21 engine.HSigma = []; |
wolffd@0 | 22 engine.hnodes = []; |
wolffd@0 | 23 |
wolffd@0 | 24 engine = class(engine, 'gaussian_inf_engine', inf_engine(bnet)); |
wolffd@0 | 25 |