Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/6622#discussion_r31676753
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/classification/OneVsRest.scala ---
@@ -209,4 +215,12 @@ final class OneVsRest(override val uid: String)
val model = new OneVsRestModel(uid, labelAttribute.toMetadata(),
models).setParent(this)
copyValues(model)
}
+
+ override def copy(extra: ParamMap): OneVsRest = {
+ val copied = super.copy(extra).asInstanceOf[OneVsRest]
+ if (isDefined(classifier)) {
+ copied.setClassifier($(classifier).copy(extra))
--- End diff --
When writing copy for CrossValidator, I recall we decided not to copy the
extra params to the estimator param.
This case seems analogous, so we probably should not copy the extra params
to classifier here.
(I could imagine arguments for either behavior, but we should be
consistent.)
---
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]
