org.encog.mathutil

Class VectorAlgebra

java.lang.Object

org.encog.mathutil.VectorAlgebra

public class VectorAlgebra
extends java.lang.Object

Basic vector algebra operators. Vectors are represented as arrays of doubles. This class was created to support the calculations in the PSO algorithm. This class is thread safe. Contributed by: Geoffroy Noel https://github.com/goffer-looney

Constructor Summary

Constructors

Constructor and Description
VectorAlgebra()

Method Summary

Modifier and Type	Method and Description
void	add(double[] v1, double[] v2) v1 = v1 + v2
void	clampComponents(double[] v, double maxValue) For each components, reset their value to maxValue if their absolute value exceeds it.
void	copy(double[] dst, double[] src) dst = src Copy a vector.
double	dotProduct(double[] v1, double[] v2) Take the dot product of two vectors.
void	mul(double[] v, double k) v = k * v The components of the vector are multiplied by k.
void	mulRand(double[] v, double k) v = k * U(0,1) * v The components of the vector are multiplied by k and a random number.
void	neg(double[] v) v = -v
void	randomise(double[] v) v = U(0, 0.1)
void	randomise(double[] v, double maxValue) v = U(-1, 1) * maxValue Randomise each component of a vector to [-maxValue, maxValue].
void	sub(double[] v1, double[] v2) v1 = v1 - v2

Methods inherited from class java.lang.Object

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

Constructor Detail

VectorAlgebra

public VectorAlgebra()

Method Detail

add

public void add(double[] v1,
                double[] v2)

v1 = v1 + v2

Parameters:

v1 - an array of doubles

v2 - an array of doubles

sub

public void sub(double[] v1,
                double[] v2)

v1 = v1 - v2

Parameters:

v1 - an array of doubles

v2 - an array of doubles

neg

public void neg(double[] v)

v = -v

Parameters:

v - an array of doubles

mulRand

public void mulRand(double[] v,
                    double k)

v = k * U(0,1) * v The components of the vector are multiplied by k and a random number. A new random number is generated for each component. Thread-safety depends on Random.nextDouble()

Parameters:

v - an array of doubles.

k - a scalar.

mul

public void mul(double[] v,
                double k)

v = k * v The components of the vector are multiplied by k.

Parameters:

v - an array of doubles.

k - a scalar.

copy

public void copy(double[] dst,
                 double[] src)

dst = src Copy a vector.

Parameters:

dst - an array of doubles

src - an array of doubles

randomise

public void randomise(double[] v)

v = U(0, 0.1)

Parameters:

v - an array of doubles

randomise

public void randomise(double[] v,
                      double maxValue)

v = U(-1, 1) * maxValue Randomise each component of a vector to [-maxValue, maxValue]. thread-safety depends on Random.nextDouble().

Parameters:

v - an array of doubles

maxValue - the max value

clampComponents

public void clampComponents(double[] v,
                            double maxValue)

For each components, reset their value to maxValue if their absolute value exceeds it.

Parameters:

v - an array of doubles

maxValue - if -1 this function does nothing

dotProduct

public double dotProduct(double[] v1,
                         double[] v2)

Take the dot product of two vectors.

Parameters:

v1 - The first vector.

v2 - The second vector.

Returns:

The dot product.