[
https://issues.apache.org/jira/browse/MAHOUT-121?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12740666#action_12740666
]
Grant Ingersoll commented on MAHOUT-121:
----------------------------------------
The only thing I'm missing at this point is the question from earlier about the
removal of the size comparison, then I am ready to commit.
{quote}
What about the commenting out of:
{code}
if (centroid.size() != v.size()) {
{code}
It seems like if we are going to do this in this distance measure, we should do
it in the other distance measure. We could push this out a layer via
documentation and require the caller to do the check, if required. Otherwise,
do we have some guarantee that the centroid and the vector are the same
cardinality in all cases of calls to this method? I think we do explicitly for
the Cluster calls, but that isn't the same.
{quote}
> 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
> Assignee: Grant Ingersoll
> Attachments: Canopy_Wiki_1000-2009-06-24.snapshot, doc-vector-4k,
> MAHOUT-121-cluster-distance.patch, MAHOUT-121-distance-optimization.patch,
> MAHOUT-121-new-distance-optimization.patch, mahout-121.patch,
> MAHOUT-121.patch, MAHOUT-121.patch, 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.
