You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hivemall.apache.org by "Makoto Yui (Jira)" <ji...@apache.org> on 2020/05/27 09:02:00 UTC
[jira] [Commented] (HIVEMALL-175) Support LightGBM
[ https://issues.apache.org/jira/browse/HIVEMALL-175?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17117544#comment-17117544 ]
Makoto Yui commented on HIVEMALL-175:
-------------------------------------
LightGBM does not provide Java API for training (only for prediction).
However, it seems it's possible to build Java API following the way used in mmlspark.
https://github.com/Microsoft/LightGBM/issues/157
https://github.com/Azure/mmlspark/blob/f3d624dbd3b15b64061ab8fb9b4a3f6a61f35f99/src/main/scala/com/microsoft/ml/spark/lightgbm/TrainUtils.scala#L183
> Support LightGBM
> ----------------
>
> Key: HIVEMALL-175
> URL: https://issues.apache.org/jira/browse/HIVEMALL-175
> Project: Hivemall
> Issue Type: New Feature
> Reporter: Makoto Yui
> Assignee: Makoto Yui
> Priority: Major
> Fix For: 0.6.2
>
>
> Support LightGBM using JNI in addition to xgboost.
> https://github.com/Microsoft/LightGBM
--
This message was sent by Atlassian Jira
(v8.3.4#803005)