annotate toolboxes/FullBNT-1.0.7/KPMstats/sample.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 x = sample(p, n)
wolffd@0 2 % SAMPLE Sample from categorical distribution.
wolffd@0 3 % Returns a row vector of integers, sampled according to the probability
wolffd@0 4 % distribution p.
wolffd@0 5 % Uses the stick-breaking algorithm.
wolffd@0 6 % Much faster algorithms are also possible.
wolffd@0 7
wolffd@0 8 if nargin < 2
wolffd@0 9 n = 1;
wolffd@0 10 end
wolffd@0 11
wolffd@0 12 cdf = cumsum(p(:));
wolffd@0 13 for i = 1:n
wolffd@0 14 x(i) = sum(cdf < rand) + 1;
wolffd@0 15 end