[
https://issues.apache.org/jira/browse/SPARK-17163?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15437285#comment-15437285
]
Joseph K. Bradley commented on SPARK-17163:
-------------------------------------------
+1 for merging the 2 now.
+1 for adding a family parameter with default “auto” as DB suggested. With
that default, most users would never even need to know about “family” Param.
regParam: I do worry about it being 0 for MLOR by default. It’d be hard to
change the default, but perhaps we should log a warning when using the
“multinomial” family.
Btw, I didn’t say before, but thanks *a lot* [~sethah] and [~dbtsai] for adding
MLOR! Few features have been so long awaited and much needed.
> Merge MLOR into a single LOR interface
> --------------------------------------
>
> Key: SPARK-17163
> URL: https://issues.apache.org/jira/browse/SPARK-17163
> Project: Spark
> Issue Type: Sub-task
> Components: ML, MLlib
> Reporter: Seth Hendrickson
>
> Before the 2.1 release, we should finalize the API for logistic regression.
> After SPARK-7159, we have both LogisticRegression and
> MultinomialLogisticRegression models. This may be confusing to users and, is
> a bit superfluous since MLOR can do basically all of what BLOR does. We
> should decide if it needs to be changed and implement those changes before 2.1
> *Update*: Seems we have decided to merge the two estimators. I changed the
> title to reflect that.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
