[
https://issues.apache.org/jira/browse/SPARK-7537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14551863#comment-14551863
]
Xiangrui Meng commented on SPARK-7537:
--------------------------------------
1. toPMML should be experimental. Shall we keep 4 toPMML methods?
1. NaiveBayes.modelType -> lowercase letters
1. DistributedLDAModel.topicDistributions is not Java friendly
1. LDAOptimizer/EMLDAOptimizer/OnlineLDAOptimizer should be private.
1. Mention the Java-friendly factory method in the RankingMetrics constructor.
1. Graduate Statistics from Experimental?
1. Statistics.kernelDensity API doesn't really match kernel density APIs from
other Statistics libraries.
> Audit new public Scala APIs for MLlib 1.4
> -----------------------------------------
>
> Key: SPARK-7537
> URL: https://issues.apache.org/jira/browse/SPARK-7537
> Project: Spark
> Issue Type: Sub-task
> Components: MLlib
> Reporter: Joseph K. Bradley
> Assignee: Xiangrui Meng
>
> Audit new public Scala APIs added to MLlib in 1.4. Take note of:
> * Protected/public classes or methods. If access can be more private, then
> it should be.
> * Documentation: Missing? Bad links or formatting?
> *Make sure to check the object doc!*
--
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
