Github user manishamde commented on the pull request:
https://github.com/apache/spark/pull/2125#issuecomment-53971034
The ordered categorical features are not binned and the centriods are
re-calculated using the entire bin aggregate every level. I can see the
improvement in accuracy here since we are not using a subsample for centriod
calculation. However, is there a loss in performance by repeatedly performing
this calculation at every level/group?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
