Package org.tribuo.clustering.kmeans

Class KMeansOptions

java.lang.Object

org.tribuo.clustering.kmeans.KMeansOptions

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Options

public class KMeansOptions
extends Object
implements com.oracle.labs.mlrg.olcut.config.Options

OLCUT Options for the K-Means implementation.

Field Summary

Fields

Modifier and Type	Field	Description
int	centroids	Number of centroids in K-Means.
DistanceType	distType	Distance function in K-Means.
KMeansTrainer.Initialisation	initialisation	Initialisation function in K-Means.
int	iterations	Iterations of the k-means algorithm.
int	numThreads	Number of computation threads in K-Means.

Fields inherited from interface com.oracle.labs.mlrg.olcut.config.Options

header

Constructor Summary

Constructors

Constructor	Description
KMeansOptions()

Method Summary

Modifier and Type	Method	Description
KMeansTrainer	getTrainer()	Gets the configured KMeansTrainer using the options in this object.

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.oracle.labs.mlrg.olcut.config.Options

getOptionsDescription

Field Details

iterations

@Option(longName="kmeans-interations",
        usage="Iterations of the k-means algorithm. Defaults to 10.")
public int iterations

Iterations of the k-means algorithm. Defaults to 10.

centroids

@Option(longName="kmeans-num-centroids",
        usage="Number of centroids in K-Means. Defaults to 10.")
public int centroids

Number of centroids in K-Means. Defaults to 10.

distType

@Option(longName="kmeans-distance-type",
        usage="The type of distance function to use for various distance calculations.")
public DistanceType distType

Distance function in K-Means. Defaults to L2 (EUCLIDEAN).

initialisation

@Option(longName="kmeans-initialisation",
        usage="Initialisation function in K-Means. Defaults to RANDOM.")
public KMeansTrainer.Initialisation initialisation

Initialisation function in K-Means. Defaults to RANDOM.

numThreads

@Option(longName="kmeans-num-threads",
        usage="Number of computation threads in K-Means. Defaults to 4.")
public int numThreads

Number of computation threads in K-Means. Defaults to 4.

Constructor Details

KMeansOptions

public KMeansOptions()

Method Details

getTrainer

public KMeansTrainer getTrainer()

Gets the configured KMeansTrainer using the options in this object.

Returns:

A KMeansTrainer.