[
https://issues.apache.org/jira/browse/SPARK-5832?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14329270#comment-14329270
]
Xiangrui Meng commented on SPARK-5832:
--------------------------------------
Maybe I understood the complexity wrongly. So it is O(nnz * k), where nnz is
the number of nonzeros in the similarity matrix and k is the number of
iterations. Is it correct? If that's the case, this is the same complexity as
the power iteration clustering we just added, and I'm okay to include it in
MLlib. (If an algorithm only works well on a single machine, we tend to not
include it in MLlib just for feature complete. Users can pick a single-machine
implementation and use it on a single worker.)
I will make a pass on your PR and it would be great if you can provide some
performance results.
> Add Affinity Propagation clustering algorithm
> ---------------------------------------------
>
> Key: SPARK-5832
> URL: https://issues.apache.org/jira/browse/SPARK-5832
> Project: Spark
> Issue Type: New Feature
> Components: MLlib
> Reporter: Liang-Chi Hsieh
> Assignee: Liang-Chi Hsieh
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
