Github user manishamde commented on a diff in the pull request:
https://github.com/apache/spark/pull/2125#discussion_r16865673
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/tree/impl/DecisionTreeMetadata.scala
---
@@ -70,32 +83,48 @@ private[tree] object DecisionTreeMetadata {
case Regression => 0
}
- val maxBins = math.min(strategy.maxBins, numExamples).toInt
- val log2MaxBinsp1 = math.log(maxBins + 1) / math.log(2.0)
+ val maxPossibleBins = math.min(strategy.maxBins, numExamples).toInt
+ val log2MaxPossibleBinsp1 = math.log(maxPossibleBins + 1) /
math.log(2.0)
+ // We check the number of bins here against maxPossibleBins.
+ // This needs to be checked here instead of in Strategy since
maxPossibleBins can be modified
+ // based on the number of training examples.
val unorderedFeatures = new mutable.HashSet[Int]()
+ val numBins = Array.fill[Int](numFeatures)(maxPossibleBins)
if (numClasses > 2) {
--- End diff --
val isMulticlassClassification = numClasses > 2 might be more readable
---
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]
