You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Joseph K. Bradley (JIRA)" <ji...@apache.org> on 2016/11/01 17:47:59 UTC

[jira] [Resolved] (SPARK-17848) Move LabelCol datatype cast into Predictor.fit

     [ https://issues.apache.org/jira/browse/SPARK-17848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Joseph K. Bradley resolved SPARK-17848.
---------------------------------------
       Resolution: Fixed
    Fix Version/s: 2.1.0

Issue resolved by pull request 15414
[https://github.com/apache/spark/pull/15414]

> Move LabelCol datatype cast into Predictor.fit
> ----------------------------------------------
>
>                 Key: SPARK-17848
>                 URL: https://issues.apache.org/jira/browse/SPARK-17848
>             Project: Spark
>          Issue Type: Improvement
>          Components: ML
>            Reporter: zhengruifeng
>            Assignee: zhengruifeng
>             Fix For: 2.1.0
>
>
> `Classifier.getNumClasses` can not support Non-Double types, and classification algos relying on it do not support non-double labelCol (like `NavieBayes`)
> As suggested by [~sethah], it is not a reasonable way to do datatype cast everywhere. And we can make cast only happen in `Predictor`.
> [~yanboliang][~josephkb][~srowen]



--
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