Parameters for a learned full-matrix field coupling. More...

#include <field_coupling.h>

Inheritance diagram for dnf_composer::element::FieldCouplingParameters:

Collaboration diagram for dnf_composer::element::FieldCouplingParameters:

Public Member Functions
	FieldCouplingParameters (const ElementDimensions &inputFieldDimensions=ElementDimensions{}, const LearningRule learningRule=LearningRule::HEBB, const double scalar=1.0, const double learningRate=0.01)
	Construct FieldCoupling parameters.

bool	operator== (const FieldCouplingParameters &other) const

std::string	toString () const override

Public Member Functions inherited from dnf_composer::element::ElementSpecificParameters
	ElementSpecificParameters ()=default

void	print () const

virtual	~ElementSpecificParameters ()=default

Public Attributes
ElementDimensions	inputFieldDimensions
	Spatial dimensions of the source (input) field.

bool	isLearningActive
	If true, weights are updated each step.

double	learningRate
	Learning rate η (step size for weight updates).

LearningRule	learningRule
	Which weight update rule to use.

double	scalar
	Scaling factor applied to the coupling output.

Detailed Description

Parameters for a learned full-matrix field coupling.

Constructor & Destructor Documentation

◆ FieldCouplingParameters()

dnf_composer::element::FieldCouplingParameters::FieldCouplingParameters	(	const ElementDimensions &	inputFieldDimensions = `ElementDimensions{}`,
		const LearningRule	learningRule = `LearningRule::HEBB`,
		const double	scalar = `1.0`,
		const double	learningRate = `0.01`
	)

inlineexplicit

Construct FieldCoupling parameters.

Parameters

inputFieldDimensions	Dimensions of the source field.
learningRule	Weight update rule (default HEBB).
scalar	Output scaling factor (default 1.0).
learningRate	Learning rate η (default 0.01).

Member Function Documentation

◆ operator==()

bool dnf_composer::element::FieldCouplingParameters::operator== ( const FieldCouplingParameters & other ) const

inline

◆ toString()

std::string dnf_composer::element::FieldCouplingParameters::toString ( ) const

inlineoverridevirtual

Implements dnf_composer::element::ElementSpecificParameters.

Here is the call graph for this function:

Here is the caller graph for this function:

Member Data Documentation

◆ inputFieldDimensions

ElementDimensions dnf_composer::element::FieldCouplingParameters::inputFieldDimensions

Spatial dimensions of the source (input) field.

◆ isLearningActive

bool dnf_composer::element::FieldCouplingParameters::isLearningActive

If true, weights are updated each step.

◆ learningRate

double dnf_composer::element::FieldCouplingParameters::learningRate

Learning rate η (step size for weight updates).

◆ learningRule

LearningRule dnf_composer::element::FieldCouplingParameters::learningRule

Which weight update rule to use.

◆ scalar

double dnf_composer::element::FieldCouplingParameters::scalar

Scaling factor applied to the coupling output.

The documentation for this struct was generated from the following file:

include/elements/field_coupling.h

Public Member Functions

Public Attributes