Github user sethah commented on a diff in the pull request:
https://github.com/apache/spark/pull/16377#discussion_r95494883
--- Diff:
mllib/src/test/scala/org/apache/spark/ml/tree/impl/RandomForestSuite.scala ---
@@ -176,6 +203,18 @@ class RandomForestSuite extends SparkFunSuite with
MLlibTestSparkContext {
assert(tree.rootNode.impurity === -1.0)
--- End diff --
Hmm, my mistake. It seems that the only time that empty splits can occur
are with regression and all continuous features. Unordered features can only
occur for multiclass classification and for categorical features that have more
than one category. So, there is no way to exercise the failure case with
classification. I think we can just have the regression test.
---
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]
