de.lmu.ifi.dbs.elki.distance.similarityfunction.kernel
Class KernelMatrix<O extends RealVector<O,?>>

java.lang.Object
  de.lmu.ifi.dbs.elki.logging.AbstractLoggable
      de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizable
          de.lmu.ifi.dbs.elki.distance.similarityfunction.kernel.KernelMatrix<O>

All Implemented Interfaces:

Loggable, Parameterizable

public class KernelMatrix<O extends RealVector<O,?>>
extends AbstractParameterizable

Provides a class for storing the kernel matrix and several extraction methods for convenience.

Author:

Simon Paradies

Field Summary

(package private) Matrix

kernel The kernel matrix

Fields inherited from class de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizable

optionHandler

Fields inherited from class de.lmu.ifi.dbs.elki.logging.AbstractLoggable

debug

Constructor Summary

KernelMatrix(double[][] matrixArray)
Wraps the matrixArray in a KernelMatrix

KernelMatrix(KernelFunction<O,DoubleDistance> kernelFunction, Database<O> database)
Provides a new kernel matrix.

KernelMatrix(KernelFunction<O,DoubleDistance> kernelFunction, Database<O> database, List<Integer> ids)
Provides a new kernel matrix.

KernelMatrix(Matrix matrix)
Makes a new kernel matrix from matrix.

Method Summary

static Matrix	centerKernelMatrix(KernelMatrix<? extends RealVector<?,? extends Number>> kernelMatrix) Centers the Kernel Matrix in Feature Space according to Smola et.
static Matrix	centerMatrix(Matrix matrix) Centers the matrix in feature space according to Smola et.
double	getDistance(int o1, int o2) Returns the kernel distance between the two specified objects.
Matrix	getKernel()
double	getSimilarity(int o1, int o2) Returns the kernel value of object o1 and object o2
double	getSquaredDistance(int o1, int o2) Returns the squared kernel distance between the two specified objects.
Matrix	getSubColumn(int i, List<Integer> ids) Returns the ith kernel matrix column for all objects in ids
Matrix	getSubMatrix(Collection<Integer> ids) Returns a sub kernel matrix for all objects in ids
String	toString()

Methods inherited from class de.lmu.ifi.dbs.elki.utilities.optionhandling.AbstractParameterizable

addOption, checkGlobalParameterConstraints, deleteOption, description, description, description, getAttributeSettings, getParameters, getParameterValue, getPossibleOptions, inlineDescription, isSet, setParameters, setParameters

Methods inherited from class de.lmu.ifi.dbs.elki.logging.AbstractLoggable

debugFine, debugFiner, debugFinest, exception, message, progress, progress, progress, verbose, verbose, warning

Methods inherited from class java.lang.Object

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

Field Detail

kernel

Matrix kernel

The kernel matrix

Constructor Detail

KernelMatrix

public KernelMatrix(double[][] matrixArray)

Wraps the matrixArray in a KernelMatrix

Parameters:

matrixArray - two dimensional double array

KernelMatrix

public KernelMatrix(KernelFunction<O,DoubleDistance> kernelFunction,
                    Database<O> database)

Provides a new kernel matrix.

Parameters:

kernelFunction - the kernel function used to compute the kernel matrix

database - the database for which the kernel matrix is computed

KernelMatrix

public KernelMatrix(KernelFunction<O,DoubleDistance> kernelFunction,
                    Database<O> database,
                    List<Integer> ids)

Provides a new kernel matrix.

Parameters:

kernelFunction - the kernel function used to compute the kernel matrix

database - the database that holds the objects

ids - the IDs of those objects for which the kernel matrix is computed

KernelMatrix

public KernelMatrix(Matrix matrix)

Makes a new kernel matrix from matrix.

Parameters:

matrix - a matrix

Method Detail

getDistance

public double getDistance(int o1,
                          int o2)

Returns the kernel distance between the two specified objects.

Parameters:

o1 - first ObjectID

o2 - second ObjectID

Returns:

the distance between the two objects

getKernel

public Matrix getKernel()

getSimilarity

public double getSimilarity(int o1,
                            int o2)

Returns the kernel value of object o1 and object o2

Parameters:

o1 - ID of first object

o2 - ID of second object

Returns:

the kernel value of object o1 and object o2

getSquaredDistance

public double getSquaredDistance(int o1,
                                 int o2)

Returns the squared kernel distance between the two specified objects.

Parameters:

o1 - first ObjectID

o2 - second ObjectID

Returns:

the distance between the two objects

getSubColumn

public Matrix getSubColumn(int i,
                           List<Integer> ids)

Returns the ith kernel matrix column for all objects in ids

Parameters:

i - the column which should be returned

ids - the objects

Returns:

the ith kernel matrix column for all objects in ids

getSubMatrix

public Matrix getSubMatrix(Collection<Integer> ids)

Returns a sub kernel matrix for all objects in ids

Parameters:

ids - the objects

Returns:

a sub kernel matrix for all objects in ids.

centerMatrix

public static Matrix centerMatrix(Matrix matrix)

Centers the matrix in feature space according to Smola et. Schoelkopf, Learning with Kernels p. 431 Alters the input matrix. If you still need the original matrix, use centeredMatrix = centerKernelMatrix(uncenteredMatrix.copy()) {

Parameters:

matrix - the matrix to be centered

Returns:

centered matrix (for convenience)

toString

public String toString()

Overrides:

toString in class Object

See Also:

Matrix.toString()

centerKernelMatrix

public static Matrix centerKernelMatrix(KernelMatrix<? extends RealVector<?,? extends Number>> kernelMatrix)

Centers the Kernel Matrix in Feature Space according to Smola et. Schoelkopf, Learning with Kernels p. 431 Alters the input matrix. If you still need the original matrix, use centeredMatrix = centerKernelMatrix(uncenteredMatrix.copy()) {

Parameters:

kernelMatrix - the kernel matrix to be centered

Returns:

centered kernelMatrix (for convenience)