[
https://issues.apache.org/jira/browse/FLINK-1731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14541211#comment-14541211
]
Chiwan Park edited comment on FLINK-1731 at 5/13/15 2:29 AM:
-------------------------------------------------------------
Hello, [~peedeeX21].
I think you can pass the initial centroids like {{fit(centroids:
DataSet\[Vector\], fitParameters: ParameterMap}}. The fit method means that
Learner creates a model and fits it into the given input. (in this case,
centroids)
And the created model (named like {{KMeansModel}}) decides the cluster of other
points. From this approach, the initial centroids passed as a DataSet will be
better.
You can see this approach in CoCoA implementation.
(https://github.com/apache/flink/blob/master/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/classification/CoCoA.scala)
I hope that this comment helps you.
was (Author: chiwanpark):
Hello, [~peedeeX21].
I think you can pass the initial centroids like {{fit(centroids:
DataSet\[Vector\], fitParameters: ParameterMap}}. The fit method means that
Learner creates a model and fits it into the given input. (in this case,
centroids)
And the created model (named like {{KMeansModel}}) decides the cluster of other
points. From this approach, the initial centroids passed as a DataSet will be
better.
You can see this approach in CoCoA implementation.
(https://github.com/apache/flink/blob/master/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/classification/CoCoA.scala)
I hope that this comments help you.
> 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: Alexander Alexandrov
> 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)
