Rank: 20 based on 1619 downloads (last 30 days) and 19 files submitted
photo

Mo Chen

E-mail

Personal Profile:
Professional Interests:
Machine Learning, Bayesian Statistics

 

Watch this Author's files

 

Files Posted by Mo View all
Updated   File Tags Downloads
(last 30 days)
Comments Rating
17 Oct 2013 Screenshot K-medoids K-medoids clustering algorithm Author: Mo Chen clustering, kmedoids, kmeans 88 12
  • 3.875
3.9 | 8 ratings
25 Sep 2013 Screenshot Viterbi algorithm (Belief propagation) for HMM MAP inference Viterbi algorithm (Belief propagation for directed graphical model) for HMM MAP inference Author: Mo Chen viterbi, hmm, state space, inference 36 0
  • 3.0
3.0 | 1 rating
18 Mar 2013 Log scale density function of basic distributions probability density function of basic distributions in log scale Author: Mo Chen pdf, probability, distribution, statistics, machine learning, prml 23 0
13 Mar 2012 Information Theory Toolbox Functions for Information theory, such as entropy, mutual information, KL divergence, etc Author: Mo Chen information theory, entropy, joint entropy, conditional entropy, relative entropy, kl divergence 143 7
  • 5.0
5.0 | 4 ratings
29 Feb 2012 Random number with specified probability Random number from a discrete distribution Author: Mo Chen discrete multinomial ... 20 3
  • 5.0
5.0 | 1 rating
Comments and Ratings by Mo View all
Updated File Comments Rating
29 Oct 2013 Visual Studio 2013 support for MATLAB Compiler Toolbox Visual Studio 2013 support files for MATLAB Compiler Toolbox (win32) Author: Arnaud Faucher

not working for mex

11 Mar 2013 EM algorithm for Gaussian mixture model EM algorithm for Gaussian mixture. Works on arbitray dimensions with high speed and precision. Author: Mo Chen


Hi, cjain, you have function call mu in path. It is your problem to solve, not mine.

07 Sep 2012 Functional Library Functional programming Author: Matthew Maycock

29 Aug 2012 Kmeans Kmeans implemented using accumarray Author: Wei LI

29 Aug 2012 Kmeans Kmeans implemented using accumarray Author: Wei LI

Carry on

Comments and Ratings on Mo's Files View all
Updated File Comment by Comments Rating
23 Jul 2014 EM algorithm for Gaussian mixture model EM algorithm for Gaussian mixture. Works on arbitray dimensions with high speed and precision. Author: Mo Chen Zhixin

19 Jul 2014 Information Theory Toolbox Functions for Information theory, such as entropy, mutual information, KL divergence, etc Author: Mo Chen Padmanaban, Subash

Change line 16:
Mx = sparse(idx, round(x), 1,n,k,n);

Apply the same changes to all sparse operations if the program throws the same error.

17 Jul 2014 Variational Bayesian Inference for Gaussian Mixture Model Variational Bayes method (mean field) for GMM can auto determine the number of components Author: Mo Chen Johannes

Hi,
This is a great help for me. However I also have the problem with multdimensional data.

Here is how to reproduce the error:
I create some dummy data (10 gaussian clusters in 3D)

numClusters = 10;

allClusters = [];

for ii = 1:numClusters
sigmaX = 20;
sigmaY = 20;
sigmaZ = 80;
sigmas = diag([sigmaX sigmaY sigmaZ]);
imageSize = diag([ 1000 1000 20 ]);

simGauss = sigmas * randn(3,2e2);
mu = imageSize * rand(3,1);
cluster = bsxfun(@plus, simGauss,mu);

allClusters = cat(2, allClusters, cluster);
end

label=vbgm(allClusters,20);
Error using chol
Matrix must be positive definite.

Sometimes it works, most of the time not however. I trace the problem to entries of nk becoming zero and a subsequent division by zero.
I have tried to replace the zero values with small ones, but that didn't help. Have no solution so far :-/

01 Jul 2014 Normalized Mutual Information Fully vectorized implementation normalized mutual information Author: Mo Chen dinh, bao

a

26 Jun 2014 kmeans clustering Fully vectorized kmeans algorithm. Fast yet simple (10 lines) Author: Mo Chen Canós, Antoni J.

The algorithm randomly fails, probably due to the random initialization.

For instance, the vector of 1D data X=[9.13,2.68,2.33,9.41] with k=2 is sometimes clustered (labeled) as [1 1 1 1], instead of the right values [1 2 2 1] or [2 1 1 2];

Contact us