MATLAB: K-means: see the initial centroids

initial centroidkmeans

Hi everyone,

Is it possible to see the randomly selected initial centroids when using the built-in kmeans function? I want to keep track of the centroids in each iteration.

Thanks.

Best Answer

Yes it is possible.

kmeans allows you to set option parameters via the statset function. In the help page kmeans there some examples on how using stateset.

Set the parameter 'OutputFcn' as shown in this tutorial, for displaying iteration results in iterative solvers: https://it.mathworks.com/help/matlab/math/output-functions.html

Related Solutions

MATLAB: How i can solve this problem? Pleas!

Your ‘T’ matrix must become a cell array. See my sample of ‘T’ for an example.

Once you do that, I would do something like this:

Y = [12.004
    13.1573 
    11.1665  
    11.8082    
    12.2129   
    12.8075   
    12.4167  
    11.5708   
    12.5798    
    13.7489];
T={'cluster2'; 'cluster1'; 'cluster3'; 'cluster9'; 'cluster0'; 'cluster7'; 'cluster4'}; % Sample Of Cell Array
[UnqClust,~,idx] = unique(T);                                                           % Unique Values & Locations
Out = Y(idx);                                                                           % Mapped Output
for k1 = 1:length(idx)
    fprintf(1, '\t%s\t%.4f\n', T{k1}, Out(k1))              % Demonstration Output (Delete Loop)
end
  cluster2  11.1665
  cluster1  13.1573
  cluster3  11.8082
  cluster9  12.4167
  cluster0  12.0040
  cluster7  12.8075
  cluster4  12.2129

MATLAB: Gmdistribution question

options = statset('Display','iter');
obj = gmdistribution.fit(X,k,options);

If that does not work, you could try the statset option OutputFcn . I cannot tell at the moment what input arguments are passed to those functions.

Best Answer

Related Solutions

MATLAB: How i can solve this problem? Pleas!

MATLAB: Gmdistribution question

Related Question