Github user manishamde commented on the pull request:
https://github.com/apache/spark/pull/2607#issuecomment-57966546
@jkbradley
I meant multi-class classification. As you pointed out, binary
classification should be similar to the regression case but I am not sure one
can handle multi-class classification with one tree. We might have to resort to
a one-vs-all strategy. I also agree with you on the naming convention -- log
loss or negative binomial log likehood are better names.
Yes, I plan to handle weighted weak hypothesis. In fact, I needed it for
something like AdaBoost and had to remove it before submitting this PR. Do you
think it makes sense to do it along with this PR or do it in the subsequent
AdaBoost PR?
I agree about the WeightedEnsemble model. Let me add it to the TODO list.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
