www.pudn.com > HMM1.zip > logist2FitRegularized.m

function [net, niter] = logist2FitRegularized(labels, features, maxIter) 
 
if nargin < 3, maxIter = 100; end 
 
[D  N] = size(features); 
weightPrior = 0.5; 
net = glm(D, 1, 'logistic', weightPrior); 
options = foptions; 
options(14) = maxIter; 
[net, options] = glmtrain(net, options, features', labels(:)); 
niter = options(14); 
%w = logist2Fit(labelsPatches(jValidPatches), features(:, jValidPatches));