Mercurial > hg > camir-aes2014
comparison toolboxes/FullBNT-1.0.7/bnt/examples/static/sample1.m @ 0:e9a9cd732c1e tip
first hg version after svn
author | wolffd |
---|---|
date | Tue, 10 Feb 2015 15:05:51 +0000 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
-1:000000000000 | 0:e9a9cd732c1e |
---|---|
1 % Check sampling on a mixture of experts model | |
2 % | |
3 % X \ | |
4 % | | | |
5 % Q | | |
6 % | / | |
7 % Y | |
8 % | |
9 % where all arcs point down. | |
10 % We condition everything on X, so X is a root node. Q is a softmax, and Y is a linear Gaussian. | |
11 % Q is hidden, X and Y are observed. | |
12 | |
13 X = 1; | |
14 Q = 2; | |
15 Y = 3; | |
16 dag = zeros(3,3); | |
17 dag(X,[Q Y]) = 1; | |
18 dag(Q,Y) = 1; | |
19 ns = [1 2 2]; | |
20 dnodes = [2]; | |
21 bnet = mk_bnet(dag, ns, dnodes); | |
22 | |
23 x = 0.5; | |
24 bnet.CPD{1} = root_CPD(bnet, 1, x); | |
25 bnet.CPD{2} = softmax_CPD(bnet, 2); | |
26 bnet.CPD{3} = gaussian_CPD(bnet, 3); | |
27 | |
28 data_case = sample_bnet(bnet, 'evidence', {0.8, [], []}) | |
29 ll = log_lik_complete(bnet, data_case) | |
30 | |
31 data_case = sample_bnet(bnet, 'evidence', {-11, [], []}) | |
32 ll = log_lik_complete(bnet, data_case) | |
33 | |
34 |