[
https://issues.apache.org/jira/browse/IGNITE-6059?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16178169#comment-16178169
]
Yury Babak commented on IGNITE-6059:
------------------------------------
Hi Mikhail,
Actually we have two implementations of k-means, local and distributed and this
ticket about distributed version. Currently distributed k-means works with
SparseDistributedMatrix wich is row/col based. But we also have
SparseBlockDistributedMatrix. So main goal of this ticket is support
SparseBlockDistributedMatrix for distributed version of k-means and main
challenge is keep performance on the same level as of SparseDistributedMatrix.
For me this sounds difficult, on the other hand I double-checked other tickets
and this one look like the easiest for a newbie. So if you want to join us -
welcome.
Btw, as far as I know [~oignatenko] works on benchmark related
ticket(IGNITE-6123) and if would like to start with this, contact with him
about performance measurement.
I look forward to your commits.
Regards,
Yury
> Use any distributed matrix in K-Means
> -------------------------------------
>
> Key: IGNITE-6059
> URL: https://issues.apache.org/jira/browse/IGNITE-6059
> Project: Ignite
> Issue Type: Improvement
> Components: ml
> Reporter: Yury Babak
> Fix For: 2.3
>
>
> Currently k-means work only with row/col matrix.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
