You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Edi Bice (JIRA)" <ji...@apache.org> on 2016/04/25 16:57:12 UTC
[jira] [Commented] (SPARK-12405) Expose featureImportances on
org.apache.spark.mllib.tree.RandomForest
[ https://issues.apache.org/jira/browse/SPARK-12405?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15256434#comment-15256434 ]
Edi Bice commented on SPARK-12405:
----------------------------------
This should be reopened - it's really not a duplicate. As the title and description clearly state it's about exposing featureImportances for mllib.tree.RandomForest. Chased all the mentioned JIRAs around, looked at the commit code and all of them point to the ML RandomForest. There are still plenty of reasons to use the MLLIB.tree.RandomForest like the ability to save/load to/from disk which the ML one does not yet provide!
> Expose featureImportances on org.apache.spark.mllib.tree.RandomForest
> ---------------------------------------------------------------------
>
> Key: SPARK-12405
> URL: https://issues.apache.org/jira/browse/SPARK-12405
> Project: Spark
> Issue Type: Sub-task
> Components: MLlib
> Affects Versions: 1.5.2
> Reporter: Asim Jalis
>
> "featureImportances" is only exposed at ML not MLlib.
> It cannot be used with org.apache.spark.mllib.tree.RandomForest.
> To use it requires using RandomForestClassifier of ML to train and create a RandomForestClassificationModel.
--
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