jsat.clustering

Class SeedSelectionMethods

java.lang.Object

jsat.clustering.SeedSelectionMethods

public class SeedSelectionMethods
extends java.lang.Object

This class provides methods for sampling a data set for a set of initial points to act as the seeds for a clustering algorithm.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SeedSelectionMethods.SeedSelection

Method Summary

Modifier and Type	Method and Description
static void	selectIntialPoints(DataSet d, int[] indices, DistanceMetric dm, java.util.List<java.lang.Double> accelCache, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod) Selects seeds from a data set to use for a clustering algorithm.
static void	selectIntialPoints(DataSet d, int[] indices, DistanceMetric dm, java.util.List<java.lang.Double> accelCache, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod, boolean parallel) Selects seeds from a data set to use for a clustering algorithm.
static void	selectIntialPoints(DataSet d, int[] indices, DistanceMetric dm, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod) Selects seeds from a data set to use for a clustering algorithm.
static void	selectIntialPoints(DataSet d, int[] indices, DistanceMetric dm, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod, boolean parallel) Selects seeds from a data set to use for a clustering algorithm.
static java.util.List<Vec>	selectIntialPoints(DataSet d, int k, DistanceMetric dm, java.util.List<java.lang.Double> accelCache, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod)
static java.util.List<Vec>	selectIntialPoints(DataSet d, int k, DistanceMetric dm, java.util.List<java.lang.Double> accelCache, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod, boolean parallel) Selects seeds from a data set to use for a clustering algorithm.
static java.util.List<Vec>	selectIntialPoints(DataSet d, int k, DistanceMetric dm, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod) Selects seeds from a data set to use for a clustering algorithm.
static java.util.List<Vec>	selectIntialPoints(DataSet d, int k, DistanceMetric dm, java.util.Random rand, SeedSelectionMethods.SeedSelection selectionMethod, boolean parallel) Selects seeds from a data set to use for a clustering algorithm.

Methods inherited from class java.lang.Object

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

Method Detail

selectIntialPoints

public static java.util.List<Vec> selectIntialPoints(DataSet d,
                                                     int k,
                                                     DistanceMetric dm,
                                                     java.util.Random rand,
                                                     SeedSelectionMethods.SeedSelection selectionMethod)

Selects seeds from a data set to use for a clustering algorithm. Copies of the vectors chosen will be returned.

Parameters:

d - the data set to perform select from

k - the number of seeds to choose

dm - the distance metric to used when selecting points

rand - a source of randomness

selectionMethod - The method of seed selection to use.

Returns:

a list of the copies of the chosen vectors.

selectIntialPoints

public static java.util.List<Vec> selectIntialPoints(DataSet d,
                                                     int k,
                                                     DistanceMetric dm,
                                                     java.util.List<java.lang.Double> accelCache,
                                                     java.util.Random rand,
                                                     SeedSelectionMethods.SeedSelection selectionMethod)

Parameters:

d - the data set to perform select from

k - the number of seeds to choose

dm - the distance metric to used when selecting points

accelCache - the cache of pre-generated acceleration information for the distance metric. May be null

rand - a source of randomness

selectionMethod - The method of seed selection to use.

Returns:

a list of the copies of the chosen vectors.

selectIntialPoints

public static java.util.List<Vec> selectIntialPoints(DataSet d,
                                                     int k,
                                                     DistanceMetric dm,
                                                     java.util.Random rand,
                                                     SeedSelectionMethods.SeedSelection selectionMethod,
                                                     boolean parallel)

Selects seeds from a data set to use for a clustering algorithm. Copies of the vectors chosen will be returned.

Parameters:

d - the data set to perform select from

k - the number of seeds to choose

dm - the distance metric to used when selecting points

rand - a source of randomness

selectionMethod - The method of seed selection to use.

parallel - true if multiple threads should be used to perform clustering. false if it should be done in a single threaded manner.

Returns:

a list of the copies of the chosen vectors.

selectIntialPoints

public static java.util.List<Vec> selectIntialPoints(DataSet d,
                                                     int k,
                                                     DistanceMetric dm,
                                                     java.util.List<java.lang.Double> accelCache,
                                                     java.util.Random rand,
                                                     SeedSelectionMethods.SeedSelection selectionMethod,
                                                     boolean parallel)

Selects seeds from a data set to use for a clustering algorithm. Copies of the vectors chosen will be returned.

Parameters:

d - the data set to perform select from

k - the number of seeds to choose

dm - the distance metric to used when selecting points

accelCache - the cache of pre-generated acceleration information for the distance metric. May be null

rand - a source of randomness

selectionMethod - The method of seed selection to use.

parallel - true if multiple threads should be used to perform clustering. false if it should be done in a single threaded manner.

Returns:

a list of the copies of the chosen vectors.

selectIntialPoints

public static void selectIntialPoints(DataSet d,
                                      int[] indices,
                                      DistanceMetric dm,
                                      java.util.Random rand,
                                      SeedSelectionMethods.SeedSelection selectionMethod)

Selects seeds from a data set to use for a clustering algorithm. The indices of the chosen points will be placed in the indices array.

Parameters:

d - the data set to perform select from

indices - a storage place to note the indices that were chosen as seed. The length of the array indicates how many seeds to select.

dm - the distance metric to used when selecting points

rand - a source of randomness

selectionMethod - The method of seed selection to use.

selectIntialPoints

public static void selectIntialPoints(DataSet d,
                                      int[] indices,
                                      DistanceMetric dm,
                                      java.util.List<java.lang.Double> accelCache,
                                      java.util.Random rand,
                                      SeedSelectionMethods.SeedSelection selectionMethod)

Selects seeds from a data set to use for a clustering algorithm. The indices of the chosen points will be placed in the indices array.

Parameters:

d - the data set to perform select from

indices - a storage place to note the indices that were chosen as seed. The length of the array indicates how many seeds to select.

dm - the distance metric to used when selecting points

accelCache - the cache of pre-generated acceleration information for the distance metric. May be null

rand - a source of randomness

selectionMethod - The method of seed selection to use.

selectIntialPoints

public static void selectIntialPoints(DataSet d,
                                      int[] indices,
                                      DistanceMetric dm,
                                      java.util.Random rand,
                                      SeedSelectionMethods.SeedSelection selectionMethod,
                                      boolean parallel)

Selects seeds from a data set to use for a clustering algorithm. The indices of the chosen points will be placed in the indices array.

Parameters:

d - the data set to perform select from

indices - a storage place to note the indices that were chosen as seed. The length of the array indicates how many seeds to select.

dm - the distance metric to used when selecting points

rand - a source of randomness

selectionMethod - The method of seed selection to use.

parallel - true if multiple threads should be used to perform clustering. false if it should be done in a single threaded manner.

selectIntialPoints

public static void selectIntialPoints(DataSet d,
                                      int[] indices,
                                      DistanceMetric dm,
                                      java.util.List<java.lang.Double> accelCache,
                                      java.util.Random rand,
                                      SeedSelectionMethods.SeedSelection selectionMethod,
                                      boolean parallel)

Selects seeds from a data set to use for a clustering algorithm. The indices of the chosen points will be placed in the indices array.

Parameters:

d - the data set to perform select from

indices - a storage place to note the indices that were chosen as seed. The length of the array indicates how many seeds to select.

dm - the distance metric to used when selecting points

accelCache - the cache of pre-generated acceleration information for the distance metric. May be null

rand - a source of randomness

selectionMethod - The method of seed selection to use.

parallel - true if multiple threads should be used to perform clustering. false if it should be done in a single threaded manner.