Package

org.apache.spark.mllib.tree

model

Permalink

package model

Visibility
  1. Public
  2. All

Type Members

  1. class DecisionTreeModel extends Serializable with Saveable

    Permalink

    :: Experimental :: Decision tree model for classification or regression.

    :: Experimental :: Decision tree model for classification or regression. This model stores the decision tree structure and parameters.

    Annotations
    @Since( "1.0.0" ) @Experimental()
  2. class GradientBoostedTreesModel extends TreeEnsembleModel with Saveable

    Permalink

    :: Experimental :: Represents a gradient boosted trees model.

    :: Experimental :: Represents a gradient boosted trees model.

    Annotations
    @Since( "1.2.0" ) @Experimental()
  3. class InformationGainStats extends Serializable

    Permalink

    :: DeveloperApi :: Information gain statistics for each split

    :: DeveloperApi :: Information gain statistics for each split

    Annotations
    @Since( "1.0.0" ) @DeveloperApi()
  4. class Node extends Serializable with Logging

    Permalink

    :: DeveloperApi :: Node in a decision tree.

    :: DeveloperApi :: Node in a decision tree.

    About node indexing: Nodes are indexed from 1. Node 1 is the root; nodes 2, 3 are the left, right children. Node index 0 is not used.

    Annotations
    @Since( "1.0.0" ) @DeveloperApi()
  5. class Predict extends Serializable

    Permalink

    Predicted value for a node

    Predicted value for a node

    Annotations
    @Since( "1.2.0" ) @DeveloperApi()
  6. class RandomForestModel extends TreeEnsembleModel with Saveable

    Permalink

    :: Experimental :: Represents a random forest model.

    :: Experimental :: Represents a random forest model.

    Annotations
    @Since( "1.2.0" ) @Experimental()
  7. case class Split(feature: Int, threshold: Double, featureType: FeatureType, categories: List[Double]) extends Product with Serializable

    Permalink

    :: DeveloperApi :: Split applied to a feature

    :: DeveloperApi :: Split applied to a feature

    feature

    feature index

    threshold

    Threshold for continuous feature. Split left if feature <= threshold, else right.

    featureType

    type of feature -- categorical or continuous

    categories

    Split left if categorical feature value is in this set, else right.

    Annotations
    @Since( "1.0.0" ) @DeveloperApi()

Value Members

  1. object DecisionTreeModel extends Loader[DecisionTreeModel] with Logging with Serializable

    Permalink
    Annotations
    @Since( "1.3.0" )
  2. object GradientBoostedTreesModel extends Loader[GradientBoostedTreesModel] with Serializable

    Permalink

    Annotations
    @Since( "1.3.0" )
  3. object RandomForestModel extends Loader[RandomForestModel] with Serializable

    Permalink
    Annotations
    @Since( "1.3.0" )

Ungrouped