hex.knn

Class KNNDistance

java.lang.Object

water.Iced<KNNDistance>

hex.knn.KNNDistance

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, water.Freezable<KNNDistance>

Direct Known Subclasses:

CosineDistance, EuclideanDistance, ManhattanDistance

public abstract class KNNDistance
extends water.Iced<KNNDistance>

Template for various distance calculation.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
double[]	values
int	valuesLength

Constructor Summary

Constructors

Constructor and Description
KNNDistance()

Method Summary

Modifier and Type	Method and Description
abstract void	calculateValues(double v1, double v2) Method to cumulate partial calculations of distance between two vectors and save it to values array.
void	initializeValues() Initialize values array to store partial calculation of distance.
abstract double	nom(double v1, double v2) Method to calculate the distance between two points from two vectors.
abstract double	result() Calculate the result from cumulated values.

Methods inherited from class water.Iced

asBytes, clone, copyOver, frozenType, read, readExternal, readJSON, reloadFromBytes, toJsonBytes, toJsonString, write, writeExternal, writeJSON

Methods inherited from class java.lang.Object

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

Field Detail

valuesLength

public int valuesLength

values

public double[] values

Constructor Detail

KNNDistance

public KNNDistance()

Method Detail

nom

public abstract double nom(double v1,
                           double v2)

Method to calculate the distance between two points from two vectors.

Parameters:

v1 - value of an item in the first vector

v2 - value of an item in the second vector

initializeValues

public void initializeValues()

Initialize values array to store partial calculation of distance.

calculateValues

public abstract void calculateValues(double v1,
                                     double v2)

Method to cumulate partial calculations of distance between two vectors and save it to values array.

Parameters:

v1 - value of an item in the first vector

v2 - value of an item in the second vector

result

public abstract double result()

Calculate the result from cumulated values.

Returns:

Final distance calculation.