Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/6761#issuecomment-114101009
Have you ruled out the example ordering? it looks like the Spark test does
produce the same examples deterministically but might be worth just printing
what it's using to make sure it matches your expectation.
Also how did you generate the Bernoulli results? the e1071 implementation
looks like it's only doing Multinomial, but I am not any expert on that
package. Is the Multinomial result correct? then I'd expect maybe it's that
you're not actually getting the results from a Bernoulli model from this R
package? it's not just Multinomial but with counts > 1 mapped to 1.
---
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]
