Github user hhbyyh commented on the pull request:
https://github.com/apache/spark/pull/11102#issuecomment-183625102
@mengxr Thanks for the review. Sorry for the late response, I was on a
flight.
It's great to know DataFrame.stat.sampleB. One concern is that it does not
allow `withReplacement = true`. That means oversampling is not supported.
For prediction, are you worrying about that users need to use the same
PipelineModel for training and evaluation dataset? I would propose the solution
to explicitly allow stepwise enable/disable on each stage for a PipelineModel.
Thus users can skip specific steps in a pipeline.
---
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]
