Github user yanboliang commented on the issue:
https://github.com/apache/spark/pull/18613
@felixcheung @wangmiao1981 In Scala, we set ```handleInvalid``` for both
estimator and model, although it only takes effect for model prediction. The
reason behind this is we should support pipeline training and transform, so we
need to support set model param during estimator fitting.
For R side, why I propose to set ```handleInvalid``` for ```predict``` is
there is no pipeline concept for R and in other native R algorithms like
```glm```, we set ```handleInvalid``` for ```predict```. I'm open to hear you
thoughts.
BTW, this PR is not related to the above discussion, let move corresponding
discussion to the other PR. This PR fix the bug to make ```label``` column also
can handle invalid according users' setting. Would you mind to have a look at
this one? Thanks.
---
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]
