org.apache.spark.mllib.clustering

Class BisectingKMeansModel

Object

org.apache.spark.mllib.clustering.BisectingKMeansModel

All Implemented Interfaces:

java.io.Serializable, Logging, Saveable

public class BisectingKMeansModel
extends Object
implements scala.Serializable, Saveable, Logging

Clustering model produced by BisectingKMeans. The prediction is done level-by-level from the root node to a leaf node, and at each node among its children the closest to the input point is selected.

param: root the root node of the clustering tree

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	BisectingKMeansModel.SaveLoadV1_0$
static class	BisectingKMeansModel.SaveLoadV2_0$

Constructor Summary

Constructors

Constructor and Description
BisectingKMeansModel(org.apache.spark.mllib.clustering.ClusteringTreeNode root)

Method Summary

Modifier and Type	Method and Description
Vector[]	clusterCenters() Leaf cluster centers.
double	computeCost(JavaRDD<Vector> data) Java-friendly version of computeCost().
double	computeCost(RDD<Vector> data) Computes the sum of squared distances between the input points and their corresponding cluster centers.
double	computeCost(Vector point) Computes the squared distance between the input point and the cluster center it belongs to.
String	distanceMeasure()
int	k() Number of leaf clusters.
static BisectingKMeansModel	load(SparkContext sc, String path)
JavaRDD<Integer>	predict(JavaRDD<Vector> points) Java-friendly version of predict().
RDD<Object>	predict(RDD<Vector> points) Predicts the indices of the clusters that the input points belong to.
int	predict(Vector point) Predicts the index of the cluster that the input point belongs to.
void	save(SparkContext sc, String path) Save this model to the given path.

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.apache.spark.internal.Logging

initializeLogging, initializeLogIfNecessary, initializeLogIfNecessary, isTraceEnabled, log_, log, logDebug, logDebug, logError, logError, logInfo, logInfo, logName, logTrace, logTrace, logWarning, logWarning

Constructor Detail

BisectingKMeansModel

public BisectingKMeansModel(org.apache.spark.mllib.clustering.ClusteringTreeNode root)

Method Detail

load

public static BisectingKMeansModel load(SparkContext sc,
                                        String path)

distanceMeasure

public String distanceMeasure()

clusterCenters

public Vector[] clusterCenters()

Leaf cluster centers.

Returns:

(undocumented)

k

public int k()

Number of leaf clusters.

Returns:

(undocumented)

predict

public int predict(Vector point)

Predicts the index of the cluster that the input point belongs to.

Parameters:

point - (undocumented)

Returns:

(undocumented)

predict

public RDD<Object> predict(RDD<Vector> points)

Predicts the indices of the clusters that the input points belong to.

Parameters:

points - (undocumented)

Returns:

(undocumented)

predict

public JavaRDD<Integer> predict(JavaRDD<Vector> points)

Java-friendly version of predict().

Parameters:

points - (undocumented)

Returns:

(undocumented)

computeCost

public double computeCost(Vector point)

Computes the squared distance between the input point and the cluster center it belongs to.

Parameters:

point - (undocumented)

Returns:

(undocumented)

computeCost

public double computeCost(RDD<Vector> data)

Computes the sum of squared distances between the input points and their corresponding cluster centers.

Parameters:

data - (undocumented)

Returns:

(undocumented)

computeCost

public double computeCost(JavaRDD<Vector> data)

Java-friendly version of computeCost().

Parameters:

data - (undocumented)

Returns:

(undocumented)

save

public void save(SparkContext sc,
                 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.