Class GradientNumerical_F64

java.lang.Object

org.ddogleg.chainrule.GradientNumerical_F64

All Implemented Interfaces:

GradientNumerical<double[]>

public class GradientNumerical_F64
extends Object
implements GradientNumerical<double[]>

Implementation of GradientNumerical for double[]

Constructor Summary

Constructors

Constructor	Description
GradientNumerical_F64()

Method Summary

Modifier and Type	Method	Description
void	configure(double T)	Overrides default settings for computing numerical gradient.
void	differentiate(CRArray<double[]> dout, List<CRArray<double[]>> gradInput)	Performs numerical differentiation to compute the gradients of input and parameters.
void	setFunction(List<CRArray<double[]>> inputs, CROperatorForward<double[]> function)	Sets the function which will be differentiated and other parameters.

Methods inherited from class Object

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

Constructor Details

GradientNumerical_F64

public GradientNumerical_F64()

Method Details

configure

public void configure(double T)

Description copied from interface: GradientNumerical

Overrides default settings for computing numerical gradient.

Specified by:

configure in interface GradientNumerical<double[]>

Parameters:

T - Sampling distance used for numerical differentiation

setFunction

public void setFunction(List<CRArray<double[]>> inputs,
 CROperatorForward<double[]> function)

Description copied from interface: GradientNumerical

Sets the function which will be differentiated and other parameters.

Specified by:

setFunction in interface GradientNumerical<double[]>

Parameters:

inputs - Value of all inputs passed into the function

function - The function which is to be differentiated

differentiate

public void differentiate(CRArray<double[]> dout,
 List<CRArray<double[]>> gradInput)

Description copied from interface: GradientNumerical

Performs numerical differentiation to compute the gradients of input and parameters. The value of parameters will be the current parameters stored in the function.

Specified by:

differentiate in interface GradientNumerical<double[]>

Parameters:

dout - (Input) Output's Jacobian, computed from next layer.

gradInput - (Output) Storage for input gradients