[
https://issues.apache.org/jira/browse/SPARK-8983?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joseph K. Bradley updated SPARK-8983:
-------------------------------------
Target Version/s: 1.6.0 (was: 1.5.0)
> ML Tuning Cross-Validation Improvements
> ---------------------------------------
>
> Key: SPARK-8983
> URL: https://issues.apache.org/jira/browse/SPARK-8983
> Project: Spark
> Issue Type: Umbrella
> Components: ML
> Reporter: Feynman Liang
>
> This is an umbrella for grouping together various improvements to pipelines
> tuning features, centralizing developer communication, and encouraging code
> reuse and common interfaces.
> We currently only support k-fold CV in {{CrossValidator}} while competing
> packages (e.g. [R caret|http://topepo.github.io/caret/splitting.html]) are
> much more feature rich, supporting balanced class labels, hold-out for
> time-series data, etc.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
