Github user MLnick commented on a diff in the pull request:
https://github.com/apache/spark/pull/11989#discussion_r58657961
--- Diff:
mllib/src/test/scala/org/apache/spark/ml/tree/impl/RandomForestSuite.scala ---
@@ -422,6 +422,13 @@ class RandomForestSuite extends SparkFunSuite with
MLlibTestSparkContext {
checkFeatureSubsetStrategy(numTrees = 1, "log2",
(math.log(numFeatures) / math.log(2)).ceil.toInt)
checkFeatureSubsetStrategy(numTrees = 1, "onethird", (numFeatures /
3.0).ceil.toInt)
+ checkFeatureSubsetStrategy(numTrees = 1, "0.1", (0.1 *
numFeatures).ceil.toInt)
--- End diff --
Is there a particular reason these test cases differ from the Java ones? I
notice in the Java tests we also test some of the regex like ".1" and ".10" and
"0.10".
I'm wondering if we shouldn't just have a couple test cases for the regex
edge cases here (to ensure it gets translated correctly).
---
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]
