Github user dbtsai commented on the pull request:
https://github.com/apache/spark/pull/3735#issuecomment-67562831
I agree. The test is not good. I'm thinking we probably can add couple well
known dataset like iris or prostate cancer dataset into the test resource, and
we can compare the accuracy with R. At Alpine, we also have small R script as
comment in the scala test which can reproduce the coefficients. Of course, we
should test if this will be converged to the same solution from different
initial condition. I found this issue when I refactorize our internal MLOR code
to MLLIb, and I will add more tests in MLOR PR.
---
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]
