[ https://issues.apache.org/jira/browse/SPARK-22004?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16165853#comment-16165853 ]
Weichen Xu commented on SPARK-22004: ------------------------------------ I will create PR once SPARK-21086 merged. > CrossValidator, TrainValidationSplit dump sub models to disk when fitting: > Scala API > ------------------------------------------------------------------------------------ > > Key: SPARK-22004 > URL: https://issues.apache.org/jira/browse/SPARK-22004 > Project: Spark > Issue Type: Sub-task > Components: ML > Affects Versions: 2.3.0 > Reporter: Weichen Xu > > We add a parameter indicating whether to persist models to disk during > training (default = off). This will use ML persistence to dump models to a > directory so they are available later but do not consume memory. > Note: when persisting the model list, use indices as the sub-model path -- This message was sent by Atlassian JIRA (v6.4.14#64029) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org