Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/8866#discussion_r48617831
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/regression/DecisionTreeRegressor.scala
---
@@ -129,6 +136,22 @@ final class DecisionTreeRegressionModel private[ml] (
rootNode.predictImpl(features).prediction
}
+ /** We need to update this function if we ever add other impurity
measures. */
+ protected def predictVariance(features: Vector): Double = {
+ rootNode.predictImpl(features).impurityStats.calculate()
+ }
+
+ override protected def transformImpl(dataset: DataFrame): DataFrame = {
--- End diff --
Thanks for pointing out that predictionCol gets checked in transform.
Unfortunately, that means that turning off predictionCol will also turn off
varianceCol. This is annoying (and probably points to a flaw in the
abstraction design), but you'll probably have to override both transform and
transformImpl to get the desired behavior.
---
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]
