[
https://issues.apache.org/jira/browse/FLINK-1731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14607941#comment-14607941
]
ASF GitHub Bot commented on FLINK-1731:
---------------------------------------
Github user thvasilo commented on the pull request:
https://github.com/apache/flink/pull/700#issuecomment-117053448
Hello @peedeeX21 . The API does not deal with distributed models at the
moment. In the K-means case having the model distributed is overkill, as it is
highly unlikely that you will have >1000 centroids, making the model tiny, and
distributing it actually creates unnecessary overhead.
We can keep the current implementation, but in the future we should really
test against a non distributed model, which can be broadcast in a
DataSet[Seq[LabeledVector]] and compare performance.
Also, could you add an evaluate operation (EvaluateDataSetOperation) for
Kmeans (and corresponding test)? It would be parametrized as
EvaluateDataSetOperation[Kmeans, Vector, Double]
> Add kMeans clustering algorithm to machine learning library
> -----------------------------------------------------------
>
> Key: FLINK-1731
> URL: https://issues.apache.org/jira/browse/FLINK-1731
> Project: Flink
> Issue Type: New Feature
> Components: Machine Learning Library
> Reporter: Till Rohrmann
> Assignee: Peter Schrott
> Labels: ML
>
> The Flink repository already contains a kMeans implementation but it is not
> yet ported to the machine learning library. I assume that only the used data
> types have to be adapted and then it can be more or less directly moved to
> flink-ml.
> The kMeans++ [1] and the kMeans|| [2] algorithm constitute a better
> implementation because the improve the initial seeding phase to achieve near
> optimal clustering. It might be worthwhile to implement kMeans||.
> Resources:
> [1] http://ilpubs.stanford.edu:8090/778/1/2006-13.pdf
> [2] http://theory.stanford.edu/~sergei/papers/vldb12-kmpar.pdf
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
