Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/7099#issuecomment-119293973
It's odd to me that there is not a natural conversion from scaled loss to
unscaled loss; the scaled loss should essentially be a loss scaled by the mean
label, which we know. The main issue is getting separate loss and
regularization terms back from Breeze.
But I'm fine with providing only the scaled loss. @feynmanliang Can you
please include a comment in the Scala doc about that?
Btw, I'm distinguishing "loss" (e.g., squared error) from "objective"
(e.g., squared error + regularization), and I hope we can stick with this
common usage throughout documentation.
---
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]
