[
https://issues.apache.org/jira/browse/SPARK-5804?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xiangrui Meng updated SPARK-5804:
---------------------------------
Assignee: Peter Rudenko
> Explicitly manage cache in Crossvalidation k-fold loop
> ------------------------------------------------------
>
> Key: SPARK-5804
> URL: https://issues.apache.org/jira/browse/SPARK-5804
> Project: Spark
> Issue Type: Improvement
> Components: ML
> Affects Versions: 1.3.0
> Reporter: Peter Rudenko
> Assignee: Peter Rudenko
> Priority: Minor
> Fix For: 1.3.0
>
>
> On a big dataset explicitly unpersist train and validation folds allows to
> load more data into memory in the next loop iteration. On my environment
> (single node 8Gb worker RAM, 2 GB dataset file, 3 folds for cross
> validation), saved more than 5 minutes.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
