[jira] [Commented] (SPARK-19054) Eliminate extra pass in NB

["Apache Spark (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-19054?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15793987#comment-15793987 ] 

Apache Spark commented on SPARK-19054:
--------------------------------------

User 'zhengruifeng' has created a pull request for this issue:
https://github.com/apache/spark/pull/16453

> Eliminate extra pass in NB
> --------------------------
>
>                 Key: SPARK-19054
>                 URL: https://issues.apache.org/jira/browse/SPARK-19054
>             Project: Spark
>          Issue Type: Improvement
>          Components: ML
>            Reporter: zhengruifeng
>            Priority: Minor
>
> {{val numClasses = getNumClasses(dataset)}} in {{NaiveBayes}} are used to check the length of thresholds only when {{positiveLabel && isDefined(thresholds)}} is true.
> In the case that {{positiveLabel}} is true and {{isDefined(thresholds)}} is false, we don't need to calculate {{getNumClasses}} which may cause a pass on the whole data.



--
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