tree 18d4e0798bb8f43cb0268554e5478e329af0946c
parent 7d48404703436e5f66e1174184cc6f30703f92c6
author Orhan Kislal <okislal@apache.org> 1565916153 -0400
committer Orhan Kislal <okislal@apache.org> 1568767185 -0400

Kmeans: Add automatic optimal cluster estimation

JIRA: MADLIB-1380

This commit adds the option to run k-means clustering algorithm for a range of
`k` values and get the optimal `k` with its associated cluster centers. It is
only supported for random and pp initial seeding options.

Closes #433

Co-authored-by: Nikhil Kak <nkak@pivotal.io>
Co-authored-by: Ekta Khanna <ekhanna@pivotal.io>