Feynman Liang created SPARK-8983:
------------------------------------
Summary: 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]
