[SPARK-5317]Set BoostingStrategy.defaultParams With Enumeration Algo.Classification or Algo.Regression

docs/mllib-ensembles.md

@@ -456,7 +456,7 @@ val (trainingData, testData) = (splits(0), splits(1))
 
 // Train a GradientBoostedTrees model.
 //  The defaultParams for Classification use LogLoss by default.
-val boostingStrategy = BoostingStrategy.defaultParams("Classification")
+val boostingStrategy = BoostingStrategy.defaultParams(Algo.Classification)
 boostingStrategy.numIterations = 3 // Note: Use more iterations in practice.
 boostingStrategy.treeStrategy.numClassesForClassification = 2
 boostingStrategy.treeStrategy.maxDepth = 5
@@ -506,7 +506,7 @@ JavaRDD<LabeledPoint> testData = splits[1];
 
 // Train a GradientBoostedTrees model.
 //  The defaultParams for Classification use LogLoss by default.
-BoostingStrategy boostingStrategy = BoostingStrategy.defaultParams("Classification");
+BoostingStrategy boostingStrategy = BoostingStrategy.defaultParams(Algo.Classification);
 boostingStrategy.setNumIterations(3); // Note: Use more iterations in practice.
 boostingStrategy.getTreeStrategy().setNumClassesForClassification(2);
 boostingStrategy.getTreeStrategy().setMaxDepth(5);
@@ -564,7 +564,7 @@ val (trainingData, testData) = (splits(0), splits(1))
 
 // Train a GradientBoostedTrees model.
 //  The defaultParams for Regression use SquaredError by default.
-val boostingStrategy = BoostingStrategy.defaultParams("Regression")
+val boostingStrategy = BoostingStrategy.defaultParams(Algo.Regression)
 boostingStrategy.numIterations = 3 // Note: Use more iterations in practice.
 boostingStrategy.treeStrategy.maxDepth = 5
 //  Empty categoricalFeaturesInfo indicates all features are continuous.
@@ -614,7 +614,7 @@ JavaRDD<LabeledPoint> testData = splits[1];
 
 // Train a GradientBoostedTrees model.
 //  The defaultParams for Regression use SquaredError by default.
-BoostingStrategy boostingStrategy = BoostingStrategy.defaultParams("Regression");
+BoostingStrategy boostingStrategy = BoostingStrategy.defaultParams(Algo.Regression);
 boostingStrategy.setNumIterations(3); // Note: Use more iterations in practice.
 boostingStrategy.getTreeStrategy().setMaxDepth(5);
 //  Empty categoricalFeaturesInfo indicates all features are continuous.

mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/BoostingStrategy.scala

@@ -88,4 +88,25 @@ object BoostingStrategy {
         throw new IllegalArgumentException(s"$algo is not supported by boosting.")
     }
   }
+
+  /**
+   * Returns default configuration for the boosting algorithm
+   * @param algo Learning goal.  Supported:
+   *             [[org.apache.spark.mllib.tree.configuration.Algo.Classification]],
+   *             [[org.apache.spark.mllib.tree.configuration.Algo.Regression]]
+   * @return Configuration for boosting algorithm
+   */
+  def defaultParams(algo: Algo): BoostingStrategy = {
+    val treeStragtegy = Strategy.defaultStategy(algo)
+    treeStragtegy.maxDepth = 3
+    algo match {
+      case Algo.Classification =>
+        treeStragtegy.numClasses = 2
+        new BoostingStrategy(treeStragtegy, LogLoss)
+      case Algo.Regression =>
+        new BoostingStrategy(treeStragtegy, SquaredError)
+      case _ =>
+        throw new IllegalArgumentException(s"$algo is not supported by boosting.")
+    }
+  }
 }

mllib/src/main/scala/org/apache/spark/mllib/tree/configuration/Strategy.scala

@@ -181,4 +181,17 @@ object Strategy {
       new Strategy(algo = Regression, impurity = Variance, maxDepth = 10,
         numClasses = 0)
   }
+
+  /**
+   * Construct a default set of parameters for [[org.apache.spark.mllib.tree.DecisionTree]]
+   * @param algo Algo.Classification or Algo.Regression
+   */
+  def defaultStategy(algo: Algo): Strategy = algo match {
+    case Algo.Classification =>
+      new Strategy(algo = Classification, impurity = Gini, maxDepth = 10,
+        numClasses = 2)
+    case Algo.Regression =>
+      new Strategy(algo = Regression, impurity = Variance, maxDepth = 10,
+        numClasses = 0)
+  }
 }