[ 
https://issues.apache.org/jira/browse/MATH-1524?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17058758#comment-17058758
 ] 

Gilles Sadowski commented on MATH-1524:
---------------------------------------

bq. This would be a change to the entire Clusterer API

As said from the outset, the API has flaws (some reported quite some time ago, 
others reflected in recently filed in JIRA, and yet others indirectly pointed 
out by in the previous comment) and now is the time to fix it.  Everything is 
up for change.

For a start, I think that we'll need to distinguish an immutable {{Cluster}} 
(as the result of a clustering process) from a mutable {{ClusterBuilder}} (as 
used during the clustering process).


> "chooseInitialCenters" should move out from KMeansPlusPlusClusterer
> -------------------------------------------------------------------
>
>                 Key: MATH-1524
>                 URL: https://issues.apache.org/jira/browse/MATH-1524
>             Project: Commons Math
>          Issue Type: Improvement
>            Reporter: Chen Tao
>            Priority: Major
>         Attachments: centroid.png, getCenter.png
>
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> There are two reason for "chooseInitialCenters" should be move out from 
> "KMeansPlusPlusClusterer":
> # k-means++ clusterer is a special case of k-means clusterer, that k-means++ 
> initialize the cluster centers with k-means++ algorithm. Another case is 
> initialize the cluster centers with random points.
> # The mini batch k-means will reuse "chooseInitialCenters". 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to