[ https://issues.apache.org/jira/browse/SPARK-15510?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15299095#comment-15299095 ]
Miao Wang commented on SPARK-15510: ----------------------------------- I will study the code and try to solve it. Thanks! > SparkR NaiveBayes should not require label to have NominalAttribute > ------------------------------------------------------------------- > > Key: SPARK-15510 > URL: https://issues.apache.org/jira/browse/SPARK-15510 > Project: Spark > Issue Type: Improvement > Components: ML, SparkR > Reporter: Joseph K. Bradley > > Currently, SparkR's NaiveBayes API fails if the label is numeric. It works > if the label is a String. This is because NaiveBayesWrapper requires that > the input column be annotated with NominalAttribute, which is created when > Strings are indexed by RFormula. We should eliminate this restriction since > it causes failures easily, such as when trying to run NB on LibSVM datasets. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org