gov.sandia.cognition.math.matrix

Class NumericalDifferentiator.MatrixJacobian

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.math.matrix.NumericalDifferentiator<Vector,Vector,Matrix>

gov.sandia.cognition.math.matrix.NumericalDifferentiator.MatrixJacobian

All Implemented Interfaces:

Evaluator<Vector,Vector>, DifferentiableEvaluator<Vector,Vector,Matrix>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

Enclosing class:

NumericalDifferentiator<InputType,OutputType,DerivativeType>

public static class NumericalDifferentiator.MatrixJacobian
extends NumericalDifferentiator<Vector,Vector,Matrix>

Numerical differentiator based on a Matrix Jacobian.

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class gov.sandia.cognition.math.matrix.NumericalDifferentiator

NumericalDifferentiator.DoubleJacobian, NumericalDifferentiator.MatrixJacobian, NumericalDifferentiator.VectorJacobian

Constructor Summary

Constructors

Constructor and Description
MatrixJacobian() Default constructor
MatrixJacobian(Evaluator<? super Vector,Vector> internalFunction) Creates a new instance of VectorJacobian
MatrixJacobian(Evaluator<? super Vector,Vector> internalFunction, double delta) Create a new instance of VectorJacobian

Method Summary

Modifier and Type	Method and Description
Matrix	differentiate(Vector input) Differentiates the output with respect to the input
static Matrix	differentiate(Vector input, Evaluator<? super Vector,Vector> f) Static access to the numerical differentiation procedure.
static Matrix	differentiate(Vectorizable input, Evaluator<? super Vector,Vector> f, double h) Static access to the numerical differentiation procedure.

Methods inherited from class gov.sandia.cognition.math.matrix.NumericalDifferentiator

clone, evaluate, getDelta, getInternalFunction, setDelta, setInternalFunction

Methods inherited from class java.lang.Object

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

Constructor Detail

MatrixJacobian

public MatrixJacobian()

Default constructor

MatrixJacobian

public MatrixJacobian(Evaluator<? super Vector,Vector> internalFunction)

Creates a new instance of VectorJacobian

Parameters:

internalFunction - Internal function to numerically differencing.

MatrixJacobian

public MatrixJacobian(Evaluator<? super Vector,Vector> internalFunction,
                      double delta)

Create a new instance of VectorJacobian

Parameters:

internalFunction - Internal function to numerically differencing.

delta - Value for x-value differencing

Method Detail

differentiate

public static Matrix differentiate(Vector input,
                                   Evaluator<? super Vector,Vector> f)

Static access to the numerical differentiation procedure.

Parameters:

input - Input about which to approximate the derivative.

f - Function of which to approximate the derivative.

Returns:

Approximated Jacobian, of the same dimension as input

differentiate

public static Matrix differentiate(Vectorizable input,
                                   Evaluator<? super Vector,Vector> f,
                                   double h)

Static access to the numerical differentiation procedure.

Parameters:

input - Input about which to approximate the derivative.

f - Function of which to approximate the derivative.

h - Value for x-value differencing

Returns:

Approximated Jacobian, of the same dimension as input

differentiate

public Matrix differentiate(Vector input)

Description copied from interface: DifferentiableEvaluator

Differentiates the output with respect to the input

Parameters:

input - Input about which to compute the derivative

Returns:

Derivative of the output with respect to the given input