Github user yanboliang commented on the pull request:
https://github.com/apache/spark/pull/11549#issuecomment-197905850
@hhbyyh I vote option 3 in JIRA. We already have
```GeneralizedLinearRegression``` in Scala, so it's better to call this
implementation from SparkR directly. Due to ```GeneralizedLinearRegression```
can not handle dataset with more than 4096 features, so it should call
```LinearRegression``` and ```LogisticRegression``` with "l-bfgs" solver
currently. Actually for ```gaussian``` family, "normal" solver equals to "irls"
solver, we can unify them as "irls" and deprecated "normal" solver.
I think We should support ```GeneralizedLinearRegression```'s solver can be
switched between ```auto, irls, l-bfgs``` at Scala side in a separate PR. After
that the SparkR::glm can directly call the Scala implementation.
---
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]
