[
https://issues.apache.org/jira/browse/SPARK-6160?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15228990#comment-15228990
]
Joseph K. Bradley commented on SPARK-6160:
------------------------------------------
Update: This should be done within spark.ml, rather than spark.mllib. If this
requires changes within spark.mllib, then we should move the implementation to
spark.ml (in a separate JIRA) before working on this task.
[~GayathriMurali] I don't think anyone is working on this. Let's figure out a
good API for those results within spark.ml, and then it should be clear how to
persist them (as a set of DataFrame columns, presumably). It may not make
sense to use ChiSqTestResult within spark.ml, so we should decide on a good API
there.
> ChiSqSelector should keep test statistic info
> ---------------------------------------------
>
> Key: SPARK-6160
> URL: https://issues.apache.org/jira/browse/SPARK-6160
> Project: Spark
> Issue Type: Improvement
> Components: ML
> Reporter: Joseph K. Bradley
> Priority: Minor
>
> It is useful to have the test statistics explaining selected features, but
> these data are thrown out when constructing the ChiSqSelectorModel. The data
> are expensive to recompute, so the ChiSqSelectorModel should store and expose
> them.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
