org.apache.spark.mllib.tree.model

Class DecisionTreeModel

java.lang.Object

org.apache.spark.mllib.tree.model.DecisionTreeModel

All Implemented Interfaces:

java.io.Serializable, Saveable

public class DecisionTreeModel
extends java.lang.Object
implements scala.Serializable, Saveable

Decision tree model for classification or regression. This model stores the decision tree structure and parameters. param: topNode root node param: algo algorithm type -- classification or regression

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
DecisionTreeModel(Node topNode, scala.Enumeration.Value algo)

Method Summary

Methods

Modifier and Type	Method and Description
scala.Enumeration.Value	algo()
int	depth() Get depth of tree.
protected java.lang.String	formatVersion() Current version of model save/load format.
static DecisionTreeModel	load(SparkContext sc, java.lang.String path)
int	numNodes() Get number of nodes in tree, including leaf nodes.
JavaRDD<java.lang.Object>	predict(JavaRDD<Vector> features) Predict values for the given data set using the model trained.
RDD<java.lang.Object>	predict(RDD<Vector> features) Predict values for the given data set using the model trained.
double	predict(Vector features) Predict values for a single data point using the model trained.
void	save(SparkContext sc, java.lang.String path) Save this model to the given path.
java.lang.String	toDebugString() Print the full model to a string.
Node	topNode()
java.lang.String	toString() Print a summary of the model.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

DecisionTreeModel

public DecisionTreeModel(Node topNode,
                 scala.Enumeration.Value algo)

Method Detail

load

public static DecisionTreeModel load(SparkContext sc,
                     java.lang.String path)

topNode

public Node topNode()

algo

public scala.Enumeration.Value algo()

predict

public double predict(Vector features)

Predict values for a single data point using the model trained.

Parameters:

features - array representing a single data point

Returns:

Double prediction from the trained model

predict

public RDD<java.lang.Object> predict(RDD<Vector> features)

Predict values for the given data set using the model trained.

Parameters:

features - RDD representing data points to be predicted

Returns:

RDD of predictions for each of the given data points

predict

public JavaRDD<java.lang.Object> predict(JavaRDD<Vector> features)

Predict values for the given data set using the model trained.

Parameters:

features - JavaRDD representing data points to be predicted

Returns:

JavaRDD of predictions for each of the given data points

numNodes

public int numNodes()

Get number of nodes in tree, including leaf nodes.

Returns:

(undocumented)

depth

public int depth()

Get depth of tree. E.g.: Depth 0 means 1 leaf node. Depth 1 means 1 internal node and 2 leaf nodes.

Returns:

(undocumented)

toString

public java.lang.String toString()

Print a summary of the model.

Overrides:

toString in class java.lang.Object

Returns:

(undocumented)

toDebugString

public java.lang.String toDebugString()

Print the full model to a string.

Returns:

(undocumented)

save

public void save(SparkContext sc,
        java.lang.String path)

Description copied from interface: Saveable

Save this model to the given path.

This saves: - human-readable (JSON) model metadata to path/metadata/ - Parquet formatted data to path/data/

The model may be loaded using Loader.load.

Specified by:

save in interface Saveable

Parameters:

sc - Spark context used to save model data.

path - Path specifying the directory in which to save this model. If the directory already exists, this method throws an exception.

formatVersion

protected java.lang.String formatVersion()

Description copied from interface: Saveable

Current version of model save/load format.

Specified by:

formatVersion in interface Saveable