[jira] [Commented] (SPARK-1406) PMML model evaluation support via MLib

["Vincenzo Selvaggio (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-1406?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14193829#comment-14193829 ] 

Vincenzo Selvaggio commented on SPARK-1406:
-------------------------------------------

Hi, 
based on what Sean suggested I had a go at this requirement, in particular the export of models to pmml as I find useful to decouple the producer (spark) and consumer (an app) of mining models.

Attached details on the approach taken, if you think it is valid I could proceed with the implementation of the other exporter (so far only kmeans is supported). 

Also attached the pmml exported for kmeans using the compiled spark-shell.

> PMML model evaluation support via MLib
> --------------------------------------
>
>                 Key: SPARK-1406
>                 URL: https://issues.apache.org/jira/browse/SPARK-1406
>             Project: Spark
>          Issue Type: New Feature
>          Components: MLlib
>            Reporter: Thomas Darimont
>         Attachments: SPARK-1406.pdf, kmeans.xml
>
>
> It would be useful if spark would provide support the evaluation of PMML models (http://www.dmg.org/v4-2/GeneralStructure.html).
> This would allow to use analytical models that were created with a statistical modeling tool like R, SAS, SPSS, etc. with Spark (MLib) which would perform the actual model evaluation for a given input tuple. The PMML model would then just contain the "parameterization" of an analytical model.
> Other projects like JPMML-Evaluator do a similar thing.
> https://github.com/jpmml/jpmml/tree/master/pmml-evaluator



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org