edu.stanford.math.plex

Class PointData

java.lang.Object

edu.stanford.math.plex.PointData

Direct Known Subclasses:

DiscreteSpace, DistanceData, PointData.Discrete, PointData.NSpace, RDirectBufferData

public abstract class PointData
extends java.lang.Object

A PointData instance encodes distance information for a data set.

Version:

$Id$

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	PointData.Discrete
static class	PointData.NSpace NSpace instances also provide dimension information.
static class	PointData.Predicate Predicate instances are used weed out points that are uninteresting in a given context (e.g., density too low or high).

Field Summary

Fields

Modifier and Type	Field and Description
protected static java.util.Random	rand

Constructor Summary

Constructors

Constructor and Description
PointData()

Method Summary

Methods

Modifier and Type	Method and Description
abstract int	count() The number of points in this data set.
abstract double	distance(int p1, int p2) Distance between 2 points.
int[]	random_points(int number_of_points) Return a sorted array of distinct indices for "random points" in the metric space.

Methods inherited from class java.lang.Object

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

Field Detail

rand

protected static java.util.Random rand

Constructor Detail

PointData

public PointData()

Method Detail

distance

public abstract double distance(int p1,
              int p2)

Distance between 2 points.

Parameters:

p1 - index of the first point

p2 - index of the second point

Returns:

distance between the two points.

count

public abstract int count()

The number of points in this data set.

Indexing begins from 1, not 0, which means that the points go from 1 to count(), inclusive. It is acceptable for an implementation to allow points to be added.

Returns:

The number of points in the data set.

random_points

public int[] random_points(int number_of_points)

Return a sorted array of distinct indices for "random points" in the metric space.

Parameters:

number_of_points - How many points we want.

Returns:

An array of indices for the random points.