Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/1950#discussion_r16279835
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/tree/impl/TreePoint.scala ---
@@ -0,0 +1,202 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.mllib.tree.impl
+
+import org.apache.spark.mllib.regression.LabeledPoint
+import org.apache.spark.mllib.tree.configuration.Strategy
+import org.apache.spark.mllib.tree.model.Bin
+import org.apache.spark.rdd.RDD
+
+
+/**
+ * Internal representation of LabeledPoint for DecisionTree.
+ * This bins feature values based on a subsampled of data as follows:
+ * (a) Continuous features are binned into ranges.
+ * (b) Unordered categorical features are binned based on subsets of
feature values.
+ * "Unordered categorical features" are categorical features with low
arity used in
+ * multiclass classification.
+ * (c) Ordered categorical features are binned based on feature values.
+ * "Ordered categorical features" are categorical features with high
arity,
+ * or any categorical feature used in regression or binary
classification.
+ *
+ * @param label Label from LabeledPoint
+ * @param features Binned feature values.
+ * Same length as LabeledPoint.features, but values are
bin indices.
+ */
+private[tree] class TreePoint(val label: Double, val features: Array[Int])
extends Serializable {
+}
+
+
+private[tree] object TreePoint {
+
+ /**
+ * Convert an input dataset into its TreePoint representation,
+ * binning feature values in preparation for DecisionTree training.
+ * @param input Input dataset.
+ * @param strategy DecisionTree training info, used for dataset
metadata.
+ * @param bins Bins for features, of size (numFeatures, numBins).
+ * @return TreePoint dataset representation
+ */
+ def convertToTreeRDD(
+ input: RDD[LabeledPoint],
+ strategy: Strategy,
+ bins: Array[Array[Bin]]): RDD[TreePoint] = {
+ input.map { x =>
+ TreePoint.labeledPointToTreePoint(x,
strategy.isMulticlassClassification, bins,
+ strategy.categoricalFeaturesInfo)
+ }
+ }
+
+ /**
+ * Convert one LabeledPoint into its TreePoint representation.
+ * @param bins Bins for features, of size (numFeatures, numBins).
+ * @param categoricalFeaturesInfo Map over categorical features:
feature index --> feature arity
+ */
+ private def labeledPointToTreePoint(
+ labeledPoint: LabeledPoint,
+ isMulticlassClassification: Boolean,
+ bins: Array[Array[Bin]],
+ categoricalFeaturesInfo: Map[Int, Int]): TreePoint = {
+
+ val numFeatures = labeledPoint.features.size
+ val numBins = bins(0).size
+ val arr = new Array[Int](numFeatures)
+ var featureIndex = 0 // offset by 1 for label
+ while (featureIndex < numFeatures) {
+ val featureInfo = categoricalFeaturesInfo.get(featureIndex)
+ val isFeatureContinuous = featureInfo.isEmpty
+ if (isFeatureContinuous) {
+ arr(featureIndex) = findBin(featureIndex, labeledPoint,
isFeatureContinuous, false,
+ bins, categoricalFeaturesInfo)
+ } else {
+ val featureCategories = featureInfo.get
+ val isSpaceSufficientForAllCategoricalSplits
+ = numBins > math.pow(2, featureCategories.toInt - 1) - 1
+ val isUnorderedFeature =
+ isMulticlassClassification &&
isSpaceSufficientForAllCategoricalSplits
+ arr(featureIndex) = findBin(featureIndex, labeledPoint,
isFeatureContinuous,
+ isUnorderedFeature, bins, categoricalFeaturesInfo)
+ }
+ featureIndex += 1
+ }
+
+ new TreePoint(labeledPoint.label, arr)
+ }
+
+
+ /**
+ * Find bin for one (labeledPoint, feature).
+ *
+ * @param isUnorderedFeature (only applies if feature is categorical)
+ * @param bins Bins for features, of size (numFeatures, numBins).
+ * @param categoricalFeaturesInfo Map over categorical features:
feature index --> feature arity
+ */
+ private def findBin(
+ featureIndex: Int,
+ labeledPoint: LabeledPoint,
+ isFeatureContinuous: Boolean,
+ isUnorderedFeature: Boolean,
+ bins: Array[Array[Bin]],
+ categoricalFeaturesInfo: Map[Int, Int]): Int = {
+
+ /**
+ * Binary search helper method for continuous feature.
+ */
+ def binarySearchForBins(): Int = {
+ val binForFeatures = bins(featureIndex)
+ val feature = labeledPoint.features(featureIndex)
+ var left = 0
+ var right = binForFeatures.length - 1
+ while (left <= right) {
+ val mid = left + (right - left) / 2
+ val bin = binForFeatures(mid)
+ val lowThreshold = bin.lowSplit.threshold
+ val highThreshold = bin.highSplit.threshold
+ if ((lowThreshold < feature) && (highThreshold >= feature)) {
+ return mid
+ }
--- End diff --
`} else if {`
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
