You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mahout.apache.org by Rajesh Nikam <ra...@gmail.com> on 2012/11/24 15:08:24 UTC

** how to dump centroids from canopy clustering **

Hi,

I am using canopy before k-means clustering from mahout.
I want to dump the centroids that are required for other processing.

These centroids stored in
/user/hadoop/<>-canopy-centroids/clusters-0-final/part-r-00000.

I could copy using hadoop fs -get command, however this stored data in
mahout specific format - SEQ.

It this sequence file file which could be dumped using "mahout seqdumper" ?

Seqdumper just give cluster count?

I want to dump feature values for cetroids.

please let me what is the way out.

Thanks
Rajesh