[
https://issues.apache.org/jira/browse/SPARK-16957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15990195#comment-15990195
]
Yan Facai (颜发才) commented on SPARK-16957:
-
To match the other libraries, we use mean value for
[
https://issues.apache.org/jira/browse/SPARK-16957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15980268#comment-15980268
]
Yan Facai (颜发才) commented on SPARK-16957:
-
[~vlad.feinberg] Hi, I found that R's gbm uses mean
[
https://issues.apache.org/jira/browse/SPARK-16957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15960260#comment-15960260
]
Apache Spark commented on SPARK-16957:
--
User 'facaiy' has created a pull request for this issue:
[
https://issues.apache.org/jira/browse/SPARK-16957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15960177#comment-15960177
]
Yan Facai (颜发才) commented on SPARK-16957:
-
I think that it is helpful for small dataset, while
[
https://issues.apache.org/jira/browse/SPARK-16957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15441640#comment-15441640
]
Abdeali Kothari commented on SPARK-16957:
-
Hi, I'd like to begin contributing, and this seems