[jira] [Commented] (FLINK-3128) Add Isotonic Regression To ML Library

["Fridtjof Sander (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-3128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15044880#comment-15044880 ] 

Fridtjof Sander commented on FLINK-3128:
----------------------------------------

I would like to work on this together with [~nhll]. We proposed this project as part of a course at TU Berlin (IMPRO3).
For a start, we basically just want to port the Spark implementation to Flink.

The parallel implementation of isotonic regression requires range partitioning. So we depend on [FLINK-7|https://issues.apache.org/jira/browse/FLINK-7], but the PR looks promising.

We are not sure about the relation of this to [GLM|https://issues.apache.org/jira/browse/FLINK-2013] since we do not use a gradient based solver for fitting.

> Add Isotonic Regression To ML Library
> -------------------------------------
>
>                 Key: FLINK-3128
>                 URL: https://issues.apache.org/jira/browse/FLINK-3128
>             Project: Flink
>          Issue Type: New Feature
>          Components: Machine Learning Library
>            Reporter: Fridtjof Sander
>            Priority: Minor
>
> Isotonic Regression fits a monotonically increasing function (also called isotonic function) to a plane of datapoints.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)