RosettaCodeData/Task/K-means++-clustering/Euler-Math-Toolbox/k-means++-clustering-2.euler
Ingy döt Net db842d013d A-M baby
2013-04-10 21:29:02 -07:00

13 lines
544 B
Text

>load clustering.e
Functions for clustering data.
>np=5; m=3*normal(np,2);
% Spread n points randomly around these points.
>n=5000; x=m[intrandom(1,n,np)]+normal(n,2);
% The function kmeanscluster contains the algorithm. It returns the
% indices of the clusters the points contain to.
>j=kmeanscluster(x,np);
% We plot each point with a color representing its cluster.
>P=x'; ...
> plot2d(P[1],P[2],r=totalmax(abs(m))+2,color=10+j,points=1,style="."); ...
> loop 1 to k; plot2d(m[#,1],m[#,2],points=1,style="o#",add=1); end; ...
> insimg;