edu.stanford.math.plex

Class EuclideanArrayData

java.lang.Object

edu.stanford.math.plex.PointData

edu.stanford.math.plex.PointData.NSpace

edu.stanford.math.plex.EuclideanArrayData

public class EuclideanArrayData
extends PointData.NSpace

The EuclideanArrayData class is the simplest implementation of NSpace. Intended for smallish data sets. Instances of this class are implemented at 0-based arrays of double, but the first dimension() entries are ignored, since PointData instances are indexed from 1 to count(), inclusive. This is occasionally awkward, and it is the reason that constructors that provide the data in the form of double[][], do not require "blank space" to be provided.

Version:

$Id$

Nested Class Summary

Nested classes/interfaces inherited from class edu.stanford.math.plex.PointData

PointData.Discrete, PointData.NSpace, PointData.Predicate

Field Summary

Fields inherited from class edu.stanford.math.plex.PointData

rand

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	EuclideanArrayData()
	EuclideanArrayData(double[][] data) Construct a EuclideanArrayData instance from a double[][].
	EuclideanArrayData(double[] data, int dimension) The obvious constructor for this class.
	EuclideanArrayData(int num_pts, int dimension)
	EuclideanArrayData(int num_pts, int dimension, double min, double max)

Method Summary

Methods

Modifier and Type	Method and Description
double	coordinate(int p, int i) Get coordinates.
int	count() The number of data points.
int	dimension() The number of dimensions of the space.
double	distance(int p1, int p2) Euclidean distance.
double[]	get_pt(int p, double[] vec) Get point.
protected boolean	pointsAreConsistent(double[][] data)
protected boolean	pointsAreConsistent(double[] data, int dimension)

Methods inherited from class edu.stanford.math.plex.PointData.NSpace

xi

Methods inherited from class edu.stanford.math.plex.PointData

random_points

Methods inherited from class java.lang.Object

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

Constructor Detail

EuclideanArrayData

protected EuclideanArrayData()

EuclideanArrayData

public EuclideanArrayData(double[] data,
                  int dimension)

The obvious constructor for this class.

This is more likely to be called programmatically. The first dimension entries in data are ignored.

Parameters:

data - The first dimension entries of this array will be ignored.

dimension - The number of dimensions.

EuclideanArrayData

public EuclideanArrayData(double[][] data)

Construct a EuclideanArrayData instance from a double[][].

This constructor requires taht the double[][] be of consistent dimenions; that is, data[i].length must be constant -- this will be the dimension of the instance. The instance will be initialize with data[0] being the first "point", so data.length will be count().

Parameters:

data - A "rectangular matrix" of data; data[0] is the first point.

EuclideanArrayData

public EuclideanArrayData(int num_pts,
                  int dimension)

EuclideanArrayData

public EuclideanArrayData(int num_pts,
                  int dimension,
                  double min,
                  double max)

Method Detail

dimension

public int dimension()

The number of dimensions of the space.

Specified by:

dimension in class PointData.NSpace

Returns:

the dimension of the data.

count

public int count()

The number of data points.

Specified by:

count in class PointData

Returns:

the number of points

distance

public double distance(int p1,
              int p2)

Euclidean distance.

Specified by:

distance in class PointData

Parameters:

p1 - the first point

p2 - the second point

Returns:

the Euclidean distance between p1 and p2.

coordinate

public double coordinate(int p,
                int i)

Get coordinates. Not much checking.

Specified by:

coordinate in class PointData.NSpace

Parameters:

p - the point

i - the coordinate

Returns:

The i-th coordinate of p;

get_pt

public double[] get_pt(int p,
              double[] vec)

Get point. No checking.

Parameters:

p - the point

vec - place to store point

Returns:

Coordinates of p;

pointsAreConsistent

protected boolean pointsAreConsistent(double[] data,
                          int dimension)

pointsAreConsistent

protected boolean pointsAreConsistent(double[][] data)