Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

A Vector based implementation of the kmeans clustering algorithm

branch: master

This branch is 0 commits ahead and 0 commits behind master

Fetching latest commit…

Cannot retrieve the latest commit at this time

README.markdown

k-means clustering algorithm in Haskell

kmeans-vector is a Haskell library to perform the k-means clustering algorithm. It is based on the very efficient 'vector' library.

Feel free to contribute, may it be features, performance improvements, etc.

Performances

kmeans-vector is much faster than the existing kmeans package. For example, the kmeans package performs k-means on 10000 2D points with k=5 in 21.099s while kmeans-vector does the same in 6.605s. For 50000 3D points, with k=5 still, kmeans-vector performs in 47.853s while I Ctrl+C'd the kmeans version after 6 minutes and a half.

Author

This library is written and maintained by Alp Mestanogullari, alpmestan@gmail.com

Something went wrong with that request. Please try again.