You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "Hyukjin Kwon (JIRA)" <ji...@apache.org> on 2019/05/21 04:02:21 UTC

[jira] [Updated] (SPARK-18616) Pure Python Implementation of MLWritable for use in Pipeline

     [ https://issues.apache.org/jira/browse/SPARK-18616?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Hyukjin Kwon updated SPARK-18616:
---------------------------------
    Labels: bulk-closed  (was: )

> Pure Python Implementation of MLWritable for use in Pipeline
> ------------------------------------------------------------
>
>                 Key: SPARK-18616
>                 URL: https://issues.apache.org/jira/browse/SPARK-18616
>             Project: Spark
>          Issue Type: Improvement
>          Components: ML
>    Affects Versions: 2.0.2
>         Environment: pyspark
>            Reporter: Andrea Matsunaga
>            Priority: Major
>              Labels: bulk-closed
>
> When developing an estimator and model completely in python, it is possible to implement the save() function, and it works for a standalone model, but not when added to a Pipeline. The reason is that Pipeline save implementation forces the use of JavaMLWritable, thus also requiring the object to have methods that are meaningful only to Java objects. Pipelines implementation need to have a check for the type of writable object defined.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

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