[ https://issues.apache.org/jira/browse/MAHOUT-121?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12720810#action_12720810 ]
Grant Ingersoll commented on MAHOUT-121: ---------------------------------------- bq. I see, i did not even know this utils/ directory was added. I'll get it into my client and update it accordingly. Just this morning... Can you tell I've got some time for Mahout lately? ;-) bq. To avoid confusion... I suppose let me keep the 'official' version of the patch for now since I need to make some fixes. It is still substantially identical to the last couple that have been posted, so, reviewing those is valid for anyone who cares. All yours, but just so you know, this is a really hot topic for me and I've got a deadline associated with it, so I can help if need be. > Speed up distance calculations for sparse vectors > ------------------------------------------------- > > Key: MAHOUT-121 > URL: https://issues.apache.org/jira/browse/MAHOUT-121 > Project: Mahout > Issue Type: Improvement > Components: Matrix > Reporter: Shashikant Kore > Attachments: MAHOUT-121.patch, MAHOUT-121.patch, MAHOUT-121.patch, > mahout-121.patch, MAHOUT-121jfe.patch, Mahout1211.patch > > > From my mail to the Mahout mailing list. > I am working on clustering a dataset which has thousands of sparse vectors. > The complete dataset has few tens of thousands of feature items but each > vector has only couple of hundred feature items. For this, there is an > optimization in distance calculation, a link to which I found the archives of > Mahout mailing list. > http://lingpipe-blog.com/2009/03/12/speeding-up-k-means-clustering-algebra-sparse-vectors/ > I tried out this optimization. The test setup had 2000 document vectors > with few hundred items. I ran canopy generation with Euclidean distance and > t1, t2 values as 250 and 200. > > Current Canopy Generation: 28 min 15 sec. > Canopy Generation with distance optimization: 1 min 38 sec. > I know by experience that using Integer, Double objects instead of primitives > is computationally expensive. I changed the sparse vector implementation to > used primitive collections by Trove [ > http://trove4j.sourceforge.net/ ]. > Distance optimization with Trove: 59 sec > Current canopy generation with Trove: 21 min 55 sec > To sum, these two optimizations reduced cluster generation time by a 97%. > Currently, I have made the changes for Euclidean Distance, Canopy and KMeans. > > Licensing of Trove seems to be an issue which needs to be addressed. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.