de.lmu.ifi.dbs.elki.math.dimensionsimilarity

Class DimensionSimilarityMatrix

java.lang.Object

de.lmu.ifi.dbs.elki.math.dimensionsimilarity.DimensionSimilarityMatrix

Direct Known Subclasses:

DimensionSimilarityMatrix.FullDimensionSimilarityMatrix, DimensionSimilarityMatrix.PartialDimensionSimilarityMatrix

public abstract class DimensionSimilarityMatrix
extends Object

Class representing a similarity matrix between dimensions.

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	DimensionSimilarityMatrix.FullDimensionSimilarityMatrix Complete matrix of pairwise dimension similarities.
static class	DimensionSimilarityMatrix.PartialDimensionSimilarityMatrix Partial matrix of pairwise dimension similarities.
static class	DimensionSimilarityMatrix.PrimAdapter Adapter class for running prim's algorithm.

Field Summary

Fields

Modifier and Type	Field and Description
static DimensionSimilarityMatrix.PrimAdapter	PRIM_ADAPTER Adapter class for running Prim's minimum spanning tree algorithm.
private double[]	sim Flat, symmetric storage.

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	DimensionSimilarityMatrix(int dims) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
Matrix	copyToFullMatrix() Transform linear triangle matrix into a full matrix.
abstract int	dim(int idx) Get the dimension at position idx.
double	get(int x, int y) Get the value of the given matrix position.
private int	index(int x, int y) Indexing function for triangular matrix.
static DimensionSimilarityMatrix	make(int dims) Make a full dimension similarity matrix.
static DimensionSimilarityMatrix	make(int[] dims) Make a partial dimension similarity matrix.
void	set(int x, int y, double val) Set the value of the given matrix position.
abstract int	size() Number of dimensions.
String	toString()

Methods inherited from class java.lang.Object

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

Field Detail

PRIM_ADAPTER

public static final DimensionSimilarityMatrix.PrimAdapter PRIM_ADAPTER

Adapter class for running Prim's minimum spanning tree algorithm.

sim

private final double[] sim

Flat, symmetric storage. We use a lower triangle matrix. Basic memory layout (X = undef, S = symmetric)

  X  S  S  S  S  S
  0  X  S  S  S  S
  1  2  X  S  S  S
  3  4  5  X  S  S
  6  7  8  9  X  S
 10 11 12 13 14  X
 

Constructor Detail

DimensionSimilarityMatrix

protected DimensionSimilarityMatrix(int dims)

Constructor.

Parameters:

dims - Number of dimensions to allocate.

Method Detail

size

public abstract int size()

Number of dimensions.

Returns:

Size of dimensions array.

dim

public abstract int dim(int idx)

Get the dimension at position idx.

Parameters:

idx - Position

Returns:

Dimension

set

public void set(int x,
       int y,
       double val)

Set the value of the given matrix position. Note that x == y is invalid!

Parameters:

x - X index coordinate

y - Y index coordinate

val - Value

get

public double get(int x,
         int y)

Get the value of the given matrix position. Note that x == y is invalid!

Parameters:

x - X index coordinate

y - Y index coordinate

Returns:

Value

index

private int index(int x,
        int y)

Indexing function for triangular matrix.

Parameters:

x - X coordinate

y - Y coordinate

Returns:

Array index

copyToFullMatrix

public Matrix copyToFullMatrix()

Transform linear triangle matrix into a full matrix.

Returns:

New matrix

toString

public String toString()

Overrides:

toString in class Object

make

public static DimensionSimilarityMatrix make(int dims)

Make a full dimension similarity matrix.

Parameters:

dims - Number of dimensions.

Returns:

Matrix

make

public static DimensionSimilarityMatrix make(int[] dims)

Make a partial dimension similarity matrix.

Parameters:

dims - Array of relevant dimensions

Returns:

Matrix