Package

org.apache.spark.mllib.tree

configuration

Permalink

package configuration

Visibility
  1. Public
  2. All

Type Members

  1. case class BoostingStrategy(treeStrategy: Strategy, loss: Loss, numIterations: Int = 100, learningRate: Double = 0.1, validationTol: Double = 0.001) extends Serializable with Product

    Permalink

    Configuration options for org.apache.spark.mllib.tree.GradientBoostedTrees.

    treeStrategy

    Parameters for the tree algorithm. We support regression and binary classification for boosting. Impurity setting will be ignored.

    loss

    Loss function used for minimization during gradient boosting.

    numIterations

    Number of iterations of boosting. In other words, the number of weak hypotheses used in the final model.

    learningRate

    Learning rate for shrinking the contribution of each estimator. The learning rate should be between in the interval (0, 1]

    validationTol

    validationTol is a condition which decides iteration termination when runWithValidation is used. The end of iteration is decided based on below logic: If the current loss on the validation set is greater than 0.01, the diff of validation error is compared to relative tolerance which is validationTol * (current loss on the validation set). If the current loss on the validation set is less than or equal to 0.01, the diff of validation error is compared to absolute tolerance which is validationTol * 0.01. Ignored when org.apache.spark.mllib.tree.GradientBoostedTrees.run() is used.

    Annotations
    @Since( "1.2.0" )
  2. class Strategy extends Serializable

    Permalink

    Stores all the configuration options for tree construction

    Stores all the configuration options for tree construction

    Annotations
    @Since( "1.0.0" )

Value Members

  1. object Algo extends Enumeration

    Permalink

    Enum to select the algorithm for the decision tree

    Enum to select the algorithm for the decision tree

    Annotations
    @Since( "1.0.0" )
  2. object BoostingStrategy extends Serializable

    Permalink
    Annotations
    @Since( "1.2.0" )
  3. object FeatureType extends Enumeration

    Permalink

    Enum to describe whether a feature is "continuous" or "categorical"

    Enum to describe whether a feature is "continuous" or "categorical"

    Annotations
    @Since( "1.0.0" )
  4. object QuantileStrategy extends Enumeration

    Permalink

    Enum for selecting the quantile calculation strategy

    Enum for selecting the quantile calculation strategy

    Annotations
    @Since( "1.0.0" )
  5. object Strategy extends Serializable

    Permalink
    Annotations
    @Since( "1.2.0" )

Ungrouped