Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/2780#discussion_r19052548
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/tree/DecisionTree.scala ---
@@ -912,16 +913,18 @@ object DecisionTree extends Serializable with Logging
{
val numSplits = metadata.numSplits(featureIndex)
val numBins = metadata.numBins(featureIndex)
if (metadata.isContinuous(featureIndex)) {
- val numSamples = sampledInput.length
+ val featureSamples = sampledInput.map(lp =>
lp.features(featureIndex))
+ val featureSplits =
findSplitsForContinuousFeature(featureSamples,
+ metadata, featureIndex)
+
+ val numSplits = featureSplits.length
+ val numBins = numSplits + 1
+ logDebug("numSplits= " + numSplits)
splits(featureIndex) = new Array[Split](numSplits)
bins(featureIndex) = new Array[Bin](numBins)
- val featureSamples = sampledInput.map(lp =>
lp.features(featureIndex)).sorted
- val stride: Double = numSamples.toDouble /
metadata.numBins(featureIndex)
- logDebug("stride = " + stride)
+
for (splitIndex <- 0 until numSplits) {
--- End diff --
As long as this part is being modified, would you mind switching the
for-loop to a while-loop?
---
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]
