How to create a multivariate gaussian mixture model??

Question

A. P. B. 2017-7-8

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/347839-how-to-create-a-multivariate-gaussian-mixture-model

评论： Sergio Cypress 2017-9-17

 [counts,binLocations] = imhist(X);
        stem(binLocations, counts, 'MarkerSize', 1 );
        xlim([-1 1]); 
        % inital kmeans step used to initialize EM
        K = 2;               % number of mixtures/clusters
        rng('default');
        cInd = kmeans(X(:), K,'MaxIter', 75536);
        % fit a GMM model
        options = statset('MaxIter', 75536); 
        gmm = fitgmdist(X(:), K,'Start',cInd,'CovarianceType','diagonal','Regularize',1e-5,'Options',options);

The piece of code shows how to fit a GMM to a univariate Gaussian distribution. X is and image. But how this can be extended to create a a 2 component 2 dimensional multivariate GMM?

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Sergio Cypress 2017-9-17

http://web.eecs.umich.edu/~cscott/pubs/tcem_tr.pdf this paper present EM algorithm to handle Multi GMM.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Prashant Arora 2017-7-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/347839-how-to-create-a-multivariate-gaussian-mixture-model#answer_274743

Hi Akshara,

The gmdistribution function supports multivariate gaussian distributions. Check the required dimensions of mu and sigma to create a multivariate 2 dimensional 2 component distribution.