Repository: spark Updated Branches: refs/heads/master 27ab0b8a0 -> 657a88835
http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/main/scala/org/apache/spark/mllib/tree/RandomForest.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/mllib/tree/RandomForest.scala b/mllib/src/main/scala/org/apache/spark/mllib/tree/RandomForest.scala index 482d339..e9304b5 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/tree/RandomForest.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/tree/RandomForest.scala @@ -56,7 +56,7 @@ import org.apache.spark.util.Utils * etc. * @param numTrees If 1, then no bootstrapping is used. If > 1, then bootstrapping is done. * @param featureSubsetStrategy Number of features to consider for splits at each node. - * Supported: "auto" (default), "all", "sqrt", "log2", "onethird". + * Supported: "auto", "all", "sqrt", "log2", "onethird". * If "auto" is set, this parameter is set based on numTrees: * if numTrees == 1, set to "all"; * if numTrees > 1 (forest) set to "sqrt" for classification and @@ -269,7 +269,7 @@ object RandomForest extends Serializable with Logging { * @param strategy Parameters for training each tree in the forest. * @param numTrees Number of trees in the random forest. * @param featureSubsetStrategy Number of features to consider for splits at each node. - * Supported: "auto" (default), "all", "sqrt", "log2", "onethird". + * Supported: "auto", "all", "sqrt", "log2", "onethird". * If "auto" is set, this parameter is set based on numTrees: * if numTrees == 1, set to "all"; * if numTrees > 1 (forest) set to "sqrt". @@ -293,13 +293,13 @@ object RandomForest extends Serializable with Logging { * * @param input Training dataset: RDD of [[org.apache.spark.mllib.regression.LabeledPoint]]. * Labels should take values {0, 1, ..., numClasses-1}. - * @param numClassesForClassification number of classes for classification. + * @param numClasses number of classes for classification. * @param categoricalFeaturesInfo Map storing arity of categorical features. * E.g., an entry (n -> k) indicates that feature n is categorical * with k categories indexed from 0: {0, 1, ..., k-1}. * @param numTrees Number of trees in the random forest. * @param featureSubsetStrategy Number of features to consider for splits at each node. - * Supported: "auto" (default), "all", "sqrt", "log2", "onethird". + * Supported: "auto", "all", "sqrt", "log2", "onethird". * If "auto" is set, this parameter is set based on numTrees: * if numTrees == 1, set to "all"; * if numTrees > 1 (forest) set to "sqrt". @@ -315,7 +315,7 @@ object RandomForest extends Serializable with Logging { */ def trainClassifier( input: RDD[LabeledPoint], - numClassesForClassification: Int, + numClasses: Int, categoricalFeaturesInfo: Map[Int, Int], numTrees: Int, featureSubsetStrategy: String, @@ -325,7 +325,7 @@ object RandomForest extends Serializable with Logging { seed: Int = Utils.random.nextInt()): RandomForestModel = { val impurityType = Impurities.fromString(impurity) val strategy = new Strategy(Classification, impurityType, maxDepth, - numClassesForClassification, maxBins, Sort, categoricalFeaturesInfo) + numClasses, maxBins, Sort, categoricalFeaturesInfo) trainClassifier(input, strategy, numTrees, featureSubsetStrategy, seed) } @@ -334,7 +334,7 @@ object RandomForest extends Serializable with Logging { */ def trainClassifier( input: JavaRDD[LabeledPoint], - numClassesForClassification: Int, + numClasses: Int, categoricalFeaturesInfo: java.util.Map[java.lang.Integer, java.lang.Integer], numTrees: Int, featureSubsetStrategy: String, @@ -342,7 +342,7 @@ object RandomForest extends Serializable with Logging { maxDepth: Int, maxBins: Int, seed: Int): RandomForestModel = { - trainClassifier(input.rdd, numClassesForClassification, + trainClassifier(input.rdd, numClasses, categoricalFeaturesInfo.asInstanceOf[java.util.Map[Int, Int]].asScala.toMap, numTrees, featureSubsetStrategy, impurity, maxDepth, maxBins, seed) } @@ -355,7 +355,7 @@ object RandomForest extends Serializable with Logging { * @param strategy Parameters for training each tree in the forest. * @param numTrees Number of trees in the random forest. * @param featureSubsetStrategy Number of features to consider for splits at each node. - * Supported: "auto" (default), "all", "sqrt", "log2", "onethird". + * Supported: "auto", "all", "sqrt", "log2", "onethird". * If "auto" is set, this parameter is set based on numTrees: * if numTrees == 1, set to "all"; * if numTrees > 1 (forest) set to "onethird". @@ -384,7 +384,7 @@ object RandomForest extends Serializable with Logging { * with k categories indexed from 0: {0, 1, ..., k-1}. * @param numTrees Number of trees in the random forest. * @param featureSubsetStrategy Number of features to consider for splits at each node. - * Supported: "auto" (default), "all", "sqrt", "log2", "onethird". + * Supported: "auto", "all", "sqrt", "log2", "onethird". * If "auto" is set, this parameter is set based on numTrees: * if numTrees == 1, set to "all"; * if numTrees > 1 (forest) set to "onethird". http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/BoostingStrategy.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/BoostingStrategy.scala b/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/BoostingStrategy.scala index e703adb..cf51d04 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/BoostingStrategy.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/BoostingStrategy.scala @@ -51,7 +51,7 @@ case class BoostingStrategy( private[tree] def assertValid(): Unit = { treeStrategy.algo match { case Classification => - require(treeStrategy.numClassesForClassification == 2, + require(treeStrategy.numClasses == 2, "Only binary classification is supported for boosting.") case Regression => // nothing @@ -80,12 +80,12 @@ object BoostingStrategy { treeStrategy.maxDepth = 3 algo match { case "Classification" => - treeStrategy.numClassesForClassification = 2 + treeStrategy.numClasses = 2 new BoostingStrategy(treeStrategy, LogLoss) case "Regression" => new BoostingStrategy(treeStrategy, SquaredError) case _ => - throw new IllegalArgumentException(s"$algo is not supported by the boosting.") + throw new IllegalArgumentException(s"$algo is not supported by boosting.") } } } http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/Strategy.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/Strategy.scala b/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/Strategy.scala index d75f384..d5cd89a 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/Strategy.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/Strategy.scala @@ -37,7 +37,7 @@ import org.apache.spark.mllib.tree.configuration.QuantileStrategy._ * Supported for Regression: [[org.apache.spark.mllib.tree.impurity.Variance]]. * @param maxDepth Maximum depth of the tree. * E.g., depth 0 means 1 leaf node; depth 1 means 1 internal node + 2 leaf nodes. - * @param numClassesForClassification Number of classes for classification. + * @param numClasses Number of classes for classification. * (Ignored for regression.) * Default value is 2 (binary classification). * @param maxBins Maximum number of bins used for discretizing continuous features and @@ -73,7 +73,7 @@ class Strategy ( @BeanProperty var algo: Algo, @BeanProperty var impurity: Impurity, @BeanProperty var maxDepth: Int, - @BeanProperty var numClassesForClassification: Int = 2, + @BeanProperty var numClasses: Int = 2, @BeanProperty var maxBins: Int = 32, @BeanProperty var quantileCalculationStrategy: QuantileStrategy = Sort, @BeanProperty var categoricalFeaturesInfo: Map[Int, Int] = Map[Int, Int](), @@ -86,7 +86,7 @@ class Strategy ( @BeanProperty var checkpointInterval: Int = 10) extends Serializable { def isMulticlassClassification = - algo == Classification && numClassesForClassification > 2 + algo == Classification && numClasses > 2 def isMulticlassWithCategoricalFeatures = isMulticlassClassification && (categoricalFeaturesInfo.size > 0) @@ -97,10 +97,10 @@ class Strategy ( algo: Algo, impurity: Impurity, maxDepth: Int, - numClassesForClassification: Int, + numClasses: Int, maxBins: Int, categoricalFeaturesInfo: java.util.Map[java.lang.Integer, java.lang.Integer]) { - this(algo, impurity, maxDepth, numClassesForClassification, maxBins, Sort, + this(algo, impurity, maxDepth, numClasses, maxBins, Sort, categoricalFeaturesInfo.asInstanceOf[java.util.Map[Int, Int]].asScala.toMap) } @@ -117,8 +117,8 @@ class Strategy ( */ def setCategoricalFeaturesInfo( categoricalFeaturesInfo: java.util.Map[java.lang.Integer, java.lang.Integer]): Unit = { - setCategoricalFeaturesInfo( - categoricalFeaturesInfo.asInstanceOf[java.util.Map[Int, Int]].asScala.toMap) + this.categoricalFeaturesInfo = + categoricalFeaturesInfo.asInstanceOf[java.util.Map[Int, Int]].asScala.toMap } /** @@ -128,9 +128,9 @@ class Strategy ( private[tree] def assertValid(): Unit = { algo match { case Classification => - require(numClassesForClassification >= 2, - s"DecisionTree Strategy for Classification must have numClassesForClassification >= 2," + - s" but numClassesForClassification = $numClassesForClassification.") + require(numClasses >= 2, + s"DecisionTree Strategy for Classification must have numClasses >= 2," + + s" but numClasses = $numClasses.") require(Set(Gini, Entropy).contains(impurity), s"DecisionTree Strategy given invalid impurity for Classification: $impurity." + s" Valid settings: Gini, Entropy") @@ -160,7 +160,7 @@ class Strategy ( /** Returns a shallow copy of this instance. */ def copy: Strategy = { - new Strategy(algo, impurity, maxDepth, numClassesForClassification, maxBins, + new Strategy(algo, impurity, maxDepth, numClasses, maxBins, quantileCalculationStrategy, categoricalFeaturesInfo, minInstancesPerNode, minInfoGain, maxMemoryInMB, subsamplingRate, useNodeIdCache, checkpointDir, checkpointInterval) } @@ -176,9 +176,9 @@ object Strategy { def defaultStrategy(algo: String): Strategy = algo match { case "Classification" => new Strategy(algo = Classification, impurity = Gini, maxDepth = 10, - numClassesForClassification = 2) + numClasses = 2) case "Regression" => new Strategy(algo = Regression, impurity = Variance, maxDepth = 10, - numClassesForClassification = 0) + numClasses = 0) } } http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/main/scala/org/apache/spark/mllib/tree/impl/DecisionTreeMetadata.scala ---------------------------------------------------------------------- diff --git a/mllib/src/main/scala/org/apache/spark/mllib/tree/impl/DecisionTreeMetadata.scala b/mllib/src/main/scala/org/apache/spark/mllib/tree/impl/DecisionTreeMetadata.scala index 5bc0f26..951733f 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/tree/impl/DecisionTreeMetadata.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/tree/impl/DecisionTreeMetadata.scala @@ -110,7 +110,7 @@ private[tree] object DecisionTreeMetadata extends Logging { val numFeatures = input.take(1)(0).features.size val numExamples = input.count() val numClasses = strategy.algo match { - case Classification => strategy.numClassesForClassification + case Classification => strategy.numClasses case Regression => 0 } http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/test/scala/org/apache/spark/mllib/tree/DecisionTreeSuite.scala ---------------------------------------------------------------------- diff --git a/mllib/src/test/scala/org/apache/spark/mllib/tree/DecisionTreeSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/tree/DecisionTreeSuite.scala index 972c905..9347eaf 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/tree/DecisionTreeSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/tree/DecisionTreeSuite.scala @@ -57,7 +57,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { Classification, Gini, maxDepth = 2, - numClassesForClassification = 2, + numClasses = 2, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 2, 1-> 2)) @@ -81,7 +81,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { Classification, Gini, maxDepth = 2, - numClassesForClassification = 2, + numClasses = 2, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 3, 1 -> 3)) @@ -177,7 +177,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { Classification, Gini, maxDepth = 2, - numClassesForClassification = 100, + numClasses = 100, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 3, 1-> 3)) @@ -271,7 +271,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { Classification, Gini, maxDepth = 2, - numClassesForClassification = 100, + numClasses = 100, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 10, 1-> 10)) // 2^(10-1) - 1 > 100, so categorical features will be ordered @@ -295,7 +295,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val strategy = new Strategy( Classification, Gini, - numClassesForClassification = 2, + numClasses = 2, maxDepth = 2, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 3, 1-> 3)) @@ -377,7 +377,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { assert(arr.length === 1000) val rdd = sc.parallelize(arr) val strategy = new Strategy(Classification, Gini, maxDepth = 3, - numClassesForClassification = 2, maxBins = 100) + numClasses = 2, maxBins = 100) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) assert(!metadata.isUnordered(featureIndex = 0)) assert(!metadata.isUnordered(featureIndex = 1)) @@ -401,7 +401,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { assert(arr.length === 1000) val rdd = sc.parallelize(arr) val strategy = new Strategy(Classification, Gini, maxDepth = 3, - numClassesForClassification = 2, maxBins = 100) + numClasses = 2, maxBins = 100) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) assert(!metadata.isUnordered(featureIndex = 0)) assert(!metadata.isUnordered(featureIndex = 1)) @@ -426,7 +426,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { assert(arr.length === 1000) val rdd = sc.parallelize(arr) val strategy = new Strategy(Classification, Entropy, maxDepth = 3, - numClassesForClassification = 2, maxBins = 100) + numClasses = 2, maxBins = 100) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) assert(!metadata.isUnordered(featureIndex = 0)) assert(!metadata.isUnordered(featureIndex = 1)) @@ -451,7 +451,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { assert(arr.length === 1000) val rdd = sc.parallelize(arr) val strategy = new Strategy(Classification, Entropy, maxDepth = 3, - numClassesForClassification = 2, maxBins = 100) + numClasses = 2, maxBins = 100) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) assert(!metadata.isUnordered(featureIndex = 0)) assert(!metadata.isUnordered(featureIndex = 1)) @@ -485,7 +485,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { // Train a 1-node model val strategyOneNode = new Strategy(Classification, Entropy, maxDepth = 1, - numClassesForClassification = 2, maxBins = 100) + numClasses = 2, maxBins = 100) val modelOneNode = DecisionTree.train(rdd, strategyOneNode) val rootNode1 = modelOneNode.topNode.deepCopy() val rootNode2 = modelOneNode.topNode.deepCopy() @@ -545,7 +545,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val arr = DecisionTreeSuite.generateCategoricalDataPointsForMulticlass() val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 3, categoricalFeaturesInfo = Map(0 -> 3, 1 -> 3)) + numClasses = 3, categoricalFeaturesInfo = Map(0 -> 3, 1 -> 3)) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) assert(strategy.isMulticlassClassification) assert(metadata.isUnordered(featureIndex = 0)) @@ -568,7 +568,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { arr(3) = new LabeledPoint(1.0, Vectors.dense(3.0)) val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 2) + numClasses = 2) val model = DecisionTree.train(rdd, strategy) DecisionTreeSuite.validateClassifier(model, arr, 1.0) @@ -585,7 +585,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 2) + numClasses = 2) val model = DecisionTree.train(rdd, strategy) DecisionTreeSuite.validateClassifier(model, arr, 1.0) @@ -600,7 +600,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val arr = DecisionTreeSuite.generateCategoricalDataPointsForMulticlass() val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 3, maxBins = maxBins, + numClasses = 3, maxBins = maxBins, categoricalFeaturesInfo = Map(0 -> 3, 1 -> 3)) assert(strategy.isMulticlassClassification) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) @@ -629,7 +629,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val arr = DecisionTreeSuite.generateContinuousDataPointsForMulticlass() val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 3, maxBins = 100) + numClasses = 3, maxBins = 100) assert(strategy.isMulticlassClassification) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) @@ -650,7 +650,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val arr = DecisionTreeSuite.generateContinuousDataPointsForMulticlass() val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 3, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 3)) + numClasses = 3, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 3)) assert(strategy.isMulticlassClassification) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) assert(metadata.isUnordered(featureIndex = 0)) @@ -671,7 +671,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val arr = DecisionTreeSuite.generateCategoricalDataPointsForMulticlassForOrderedFeatures() val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 3, maxBins = 100, + numClasses = 3, maxBins = 100, categoricalFeaturesInfo = Map(0 -> 10, 1 -> 10)) assert(strategy.isMulticlassClassification) val metadata = DecisionTreeMetadata.buildMetadata(rdd, strategy) @@ -692,7 +692,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val arr = DecisionTreeSuite.generateCategoricalDataPointsForMulticlassForOrderedFeatures() val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 4, - numClassesForClassification = 3, maxBins = 10, + numClasses = 3, maxBins = 10, categoricalFeaturesInfo = Map(0 -> 10, 1 -> 10)) assert(strategy.isMulticlassClassification) @@ -708,7 +708,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, - maxDepth = 2, numClassesForClassification = 2, minInstancesPerNode = 2) + maxDepth = 2, numClasses = 2, minInstancesPerNode = 2) val model = DecisionTree.train(rdd, strategy) assert(model.topNode.isLeaf) @@ -737,7 +737,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val rdd = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxBins = 2, maxDepth = 2, categoricalFeaturesInfo = Map(0 -> 2, 1-> 2), - numClassesForClassification = 2, minInstancesPerNode = 2) + numClasses = 2, minInstancesPerNode = 2) val rootNode = DecisionTree.train(rdd, strategy).topNode @@ -755,7 +755,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val input = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 2, - numClassesForClassification = 2, minInfoGain = 1.0) + numClasses = 2, minInfoGain = 1.0) val model = DecisionTree.train(input, strategy) assert(model.topNode.isLeaf) @@ -781,7 +781,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val input = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 1, - numClassesForClassification = 2, categoricalFeaturesInfo = Map(0 -> 3)) + numClasses = 2, categoricalFeaturesInfo = Map(0 -> 3)) val metadata = DecisionTreeMetadata.buildMetadata(input, strategy) val (splits, bins) = DecisionTree.findSplitsBins(input, metadata) @@ -824,7 +824,7 @@ class DecisionTreeSuite extends FunSuite with MLlibTestSparkContext { val input = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 5, - numClassesForClassification = 2, categoricalFeaturesInfo = Map(0 -> 3)) + numClasses = 2, categoricalFeaturesInfo = Map(0 -> 3)) val metadata = DecisionTreeMetadata.buildMetadata(input, strategy) val (splits, bins) = DecisionTree.findSplitsBins(input, metadata) http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/test/scala/org/apache/spark/mllib/tree/GradientBoostedTreesSuite.scala ---------------------------------------------------------------------- diff --git a/mllib/src/test/scala/org/apache/spark/mllib/tree/GradientBoostedTreesSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/tree/GradientBoostedTreesSuite.scala index d4d54cf..3aa97e5 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/tree/GradientBoostedTreesSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/tree/GradientBoostedTreesSuite.scala @@ -100,7 +100,7 @@ class GradientBoostedTreesSuite extends FunSuite with MLlibTestSparkContext { val rdd = sc.parallelize(GradientBoostedTreesSuite.data, 2) val treeStrategy = new Strategy(algo = Classification, impurity = Variance, maxDepth = 2, - numClassesForClassification = 2, categoricalFeaturesInfo = Map.empty, + numClasses = 2, categoricalFeaturesInfo = Map.empty, subsamplingRate = subsamplingRate) val boostingStrategy = new BoostingStrategy(treeStrategy, LogLoss, numIterations, learningRate) http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/mllib/src/test/scala/org/apache/spark/mllib/tree/RandomForestSuite.scala ---------------------------------------------------------------------- diff --git a/mllib/src/test/scala/org/apache/spark/mllib/tree/RandomForestSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/tree/RandomForestSuite.scala index 90a8c2d..f7f0f20 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/tree/RandomForestSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/tree/RandomForestSuite.scala @@ -57,7 +57,7 @@ class RandomForestSuite extends FunSuite with MLlibTestSparkContext { " comparing DecisionTree vs. RandomForest(numTrees = 1)") { val categoricalFeaturesInfo = Map.empty[Int, Int] val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 2, - numClassesForClassification = 2, categoricalFeaturesInfo = categoricalFeaturesInfo) + numClasses = 2, categoricalFeaturesInfo = categoricalFeaturesInfo) binaryClassificationTestWithContinuousFeatures(strategy) } @@ -65,7 +65,7 @@ class RandomForestSuite extends FunSuite with MLlibTestSparkContext { " comparing DecisionTree vs. RandomForest(numTrees = 1)") { val categoricalFeaturesInfo = Map.empty[Int, Int] val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 2, - numClassesForClassification = 2, categoricalFeaturesInfo = categoricalFeaturesInfo, + numClasses = 2, categoricalFeaturesInfo = categoricalFeaturesInfo, useNodeIdCache = true) binaryClassificationTestWithContinuousFeatures(strategy) } @@ -93,7 +93,7 @@ class RandomForestSuite extends FunSuite with MLlibTestSparkContext { " comparing DecisionTree vs. RandomForest(numTrees = 1)") { val categoricalFeaturesInfo = Map.empty[Int, Int] val strategy = new Strategy(algo = Regression, impurity = Variance, - maxDepth = 2, maxBins = 10, numClassesForClassification = 2, + maxDepth = 2, maxBins = 10, numClasses = 2, categoricalFeaturesInfo = categoricalFeaturesInfo) regressionTestWithContinuousFeatures(strategy) } @@ -102,7 +102,7 @@ class RandomForestSuite extends FunSuite with MLlibTestSparkContext { " comparing DecisionTree vs. RandomForest(numTrees = 1)") { val categoricalFeaturesInfo = Map.empty[Int, Int] val strategy = new Strategy(algo = Regression, impurity = Variance, - maxDepth = 2, maxBins = 10, numClassesForClassification = 2, + maxDepth = 2, maxBins = 10, numClasses = 2, categoricalFeaturesInfo = categoricalFeaturesInfo, useNodeIdCache = true) regressionTestWithContinuousFeatures(strategy) } @@ -169,14 +169,14 @@ class RandomForestSuite extends FunSuite with MLlibTestSparkContext { test("Binary classification with continuous features: subsampling features") { val categoricalFeaturesInfo = Map.empty[Int, Int] val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 2, - numClassesForClassification = 2, categoricalFeaturesInfo = categoricalFeaturesInfo) + numClasses = 2, categoricalFeaturesInfo = categoricalFeaturesInfo) binaryClassificationTestWithContinuousFeaturesAndSubsampledFeatures(strategy) } test("Binary classification with continuous features and node Id cache: subsampling features") { val categoricalFeaturesInfo = Map.empty[Int, Int] val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 2, - numClassesForClassification = 2, categoricalFeaturesInfo = categoricalFeaturesInfo, + numClasses = 2, categoricalFeaturesInfo = categoricalFeaturesInfo, useNodeIdCache = true) binaryClassificationTestWithContinuousFeaturesAndSubsampledFeatures(strategy) } @@ -191,7 +191,7 @@ class RandomForestSuite extends FunSuite with MLlibTestSparkContext { val input = sc.parallelize(arr) val strategy = new Strategy(algo = Classification, impurity = Gini, maxDepth = 5, - numClassesForClassification = 3, categoricalFeaturesInfo = categoricalFeaturesInfo) + numClasses = 3, categoricalFeaturesInfo = categoricalFeaturesInfo) val model = RandomForest.trainClassifier(input, strategy, numTrees = 2, featureSubsetStrategy = "sqrt", seed = 12345) EnsembleTestHelper.validateClassifier(model, arr, 1.0) http://git-wip-us.apache.org/repos/asf/spark/blob/657a8883/python/pyspark/mllib/tree.py ---------------------------------------------------------------------- diff --git a/python/pyspark/mllib/tree.py b/python/pyspark/mllib/tree.py index 46e2539..6670247 100644 --- a/python/pyspark/mllib/tree.py +++ b/python/pyspark/mllib/tree.py @@ -250,7 +250,7 @@ class RandomForest(object): return RandomForestModel(model) @classmethod - def trainClassifier(cls, data, numClassesForClassification, categoricalFeaturesInfo, numTrees, + def trainClassifier(cls, data, numClasses, categoricalFeaturesInfo, numTrees, featureSubsetStrategy="auto", impurity="gini", maxDepth=4, maxBins=32, seed=None): """ @@ -259,7 +259,7 @@ class RandomForest(object): :param data: Training dataset: RDD of LabeledPoint. Labels should take values {0, 1, ..., numClasses-1}. - :param numClassesForClassification: number of classes for classification. + :param numClasses: number of classes for classification. :param categoricalFeaturesInfo: Map storing arity of categorical features. E.g., an entry (n -> k) indicates that feature n is categorical with k categories indexed from 0: {0, 1, ..., k-1}. @@ -320,7 +320,7 @@ class RandomForest(object): >>> model.predict(rdd).collect() [1.0, 0.0] """ - return cls._train(data, "classification", numClassesForClassification, + return cls._train(data, "classification", numClasses, categoricalFeaturesInfo, numTrees, featureSubsetStrategy, impurity, maxDepth, maxBins, seed) --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
