Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/6761#issuecomment-115200955
So you get different results in scikit? Between R and scikit, do you have a
stable result for multinomial and bernoulli respectively? and do they match
mllib? then that's probably good enough.
If it's still not working I think we can hand-craft a very simple problem
where the answer can just be calculated by hand.
Thanks for your hard work on developing a test for this. It is definitely
additive.
---
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]
