Full-matrix learned coupling between two neural fields. More...

#include <field_coupling.h>

Inheritance diagram for dnf_composer::element::FieldCoupling:

Collaboration diagram for dnf_composer::element::FieldCoupling:

Public Member Functions
void	addInput (const std::shared_ptr< Element > &inputElement, const std::string &inputComponent="output") override
	Register `inputElement` as an upstream source for this element.

void	changeDimensions (const ElementDimensions &newDimensions) override
	Resize the output field dimensions and rebuild the weight matrix. Preserves input field dimensions and clears weights. Connections are not removed — call removeInputs()/removeOutputs() first if needed.

void	changeInputDimensions (const ElementDimensions &newInputDimensions)
	Resize the input field dimensions and rebuild the weight matrix. Preserves output field dimensions and clears weights. Connections are not removed — call removeInputs()/removeOutputs() first if needed.

void	clearWeights ()
	Reset the weight matrix to all zeros.

std::shared_ptr< Element >	clone () const override

	FieldCoupling (const ElementCommonParameters &elementCommonParameters, const FieldCouplingParameters &fc_parameters)
	Construct a FieldCoupling.

FieldCouplingParameters	getParameters () const

std::string	getWeightsDirectory () const

void	init () override
	Initialize the element (called once before the simulation loop).

void	readWeights ()
	Load the weight matrix from a binary file in `weightsDirectory`.

void	setLearning (bool learning)
	Enable or disable online weight updates.

void	setLearningRate (double learningRate)

void	setParameters (const FieldCouplingParameters &fcp)

void	setWeightsDirectory (const std::string &dir)
	Set the directory used for `readWeights()` / `writeWeights()`.

void	step (double t, double deltaT) override
	Advance the element by one time step.

std::string	toString () const override

void	tryReadWeights ()
	Load weights if the file exists; log INFO in either case. Unlike `readWeights()`, this never logs an error — use it when weights may legitimately be absent (e.g. first run of a programmatic simulation).

void	writeWeights () const
	Save the current weight matrix to a binary file in `weightsDirectory`.

Public Member Functions inherited from dnf_composer::element::Element
void	buildInputCache ()
	Cache raw pointers to input component data. Call after all element init()s complete.

void	close ()

	Element (const ElementCommonParameters &parameters)
	Construct an element with the given common parameters.

std::vector< double >	getComponent (const std::string &componentName)
	Return a copy of the named component vector.

std::vector< std::string >	getComponentList () const

std::vector< double > *	getComponentPtr (const std::string &componentName)

const std::unordered_map< std::string, std::vector< double > > *	getComponents () const
	Return a read-only pointer to the full components map.

ElementCommonParameters	getElementCommonParameters () const

std::vector< std::shared_ptr< Element > >	getInputs ()

std::unordered_map< std::shared_ptr< Element >, std::string >	getInputsAndComponents ()
	Return all inputs mapped to the component name they expose.

ElementLabel	getLabel () const

int	getMaxSpatialDimension () const

std::vector< std::shared_ptr< Element > >	getOutputs ()

int	getSize () const
	Return the number of spatial samples (size = round(x_max / d_x)).

double	getStepSize () const
	Return the spatial resolution (d_x).

int	getUniqueIdentifier () const

std::string	getUniqueName () const

bool	hasInput () const

bool	hasInput (const std::string &inputElementName, const std::string &inputComponent)

bool	hasInput (int inputElementId, const std::string &inputComponent)

bool	hasOutput () const

bool	hasOutput (const std::string &outputElementName, const std::string &outputComponent)

bool	hasOutput (int outputElementId, const std::string &outputComponent)

void	print () const

void	removeInput (const std::string &inputElementId)

void	removeInput (int uniqueId)

void	removeInputs ()

void	removeOutput (const std::string &outputElementId)
	Deregister this element as an input of `outputElementId`.

void	removeOutput (int uniqueId)
	Deregister this element as an input of the element with `uniqueId`.

void	removeOutputs ()

void	setUniqueName (const std::string &name)

void	updateInput ()
	Pull data from all registered input elements into this element's components.

virtual	~Element ()=default

Protected Attributes
std::shared_ptr< Element >	input

std::shared_ptr< Element >	output

FieldCouplingParameters	parameters

std::string	weightsDirectory
	Directory used for weight serialization.

Protected Attributes inherited from dnf_composer::element::Element
ElementCommonParameters	commonParameters
	Name, label, and spatial dimensions.

std::unordered_map< std::string, std::vector< double > >	components
	Named data arrays (e.g. "output").

std::unordered_map< std::shared_ptr< Element >, std::string >	inputs
	Upstream elements and the component they expose.

std::unordered_map< std::shared_ptr< Element >, std::string >	outputs
	Downstream elements that read this element's output.

Detailed Description

Full-matrix learned coupling between two neural fields.

FieldCoupling maintains an (output_size × input_size) weight matrix W. On each step() it computes output = W * f(input) (matrix-vector product of the weight matrix with the input field's "output" component).

When learning is active (setLearning(true)), weights are updated according to the selected LearningRule (HEBB, OJA, or DELTA). Weights can be persisted to and loaded from disk via writeWeights() / readWeights().

Constructor & Destructor Documentation

◆ FieldCoupling()

dnf_composer::element::FieldCoupling::FieldCoupling	(	const ElementCommonParameters &	elementCommonParameters,
		const FieldCouplingParameters &	fc_parameters
	)

Construct a FieldCoupling.

Parameters

elementCommonParameters	Name, label, and dimensions of the output field.
fc_parameters	Coupling parameters (input dimensions, learning rule).

Here is the call graph for this function:

Member Function Documentation

◆ addInput()

void dnf_composer::element::FieldCoupling::addInput	(	const std::shared_ptr< Element > &	inputElement,
		const std::string &	inputComponent = `"output"`
	)

overridevirtual

Register inputElement as an upstream source for this element.

Parameters

inputElement	The element whose output will be read.
inputComponent	Which component of `inputElement` to read (default: "output").

Reimplemented from dnf_composer::element::Element.

Here is the call graph for this function:

◆ changeDimensions()

void dnf_composer::element::FieldCoupling::changeDimensions ( const ElementDimensions & newDimensions )

overridevirtual

Resize the output field dimensions and rebuild the weight matrix. Preserves input field dimensions and clears weights. Connections are not removed — call removeInputs()/removeOutputs() first if needed.

Reimplemented from dnf_composer::element::Element.

Here is the call graph for this function:

◆ changeInputDimensions()

void dnf_composer::element::FieldCoupling::changeInputDimensions ( const ElementDimensions & newInputDimensions )

Resize the input field dimensions and rebuild the weight matrix. Preserves output field dimensions and clears weights. Connections are not removed — call removeInputs()/removeOutputs() first if needed.

Here is the call graph for this function:

◆ clearWeights()

void dnf_composer::element::FieldCoupling::clearWeights ( )

Reset the weight matrix to all zeros.

◆ clone()

std::shared_ptr< Element > dnf_composer::element::FieldCoupling::clone ( ) const

overridevirtual

Implements dnf_composer::element::Element.

◆ getParameters()

FieldCouplingParameters dnf_composer::element::FieldCoupling::getParameters ( ) const

◆ getWeightsDirectory()

std::string dnf_composer::element::FieldCoupling::getWeightsDirectory ( ) const

◆ init()

void dnf_composer::element::FieldCoupling::init ( )

overridevirtual

Initialize the element (called once before the simulation loop).

Implements dnf_composer::element::Element.

Here is the caller graph for this function:

◆ readWeights()

void dnf_composer::element::FieldCoupling::readWeights ( )

Load the weight matrix from a binary file in weightsDirectory.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ setLearning()

void dnf_composer::element::FieldCoupling::setLearning ( bool learning )

Enable or disable online weight updates.

Parameters

learning True to activate learning.

◆ setLearningRate()

void dnf_composer::element::FieldCoupling::setLearningRate ( double learningRate )

◆ setParameters()

void dnf_composer::element::FieldCoupling::setParameters ( const FieldCouplingParameters & fcp )

◆ setWeightsDirectory()

void dnf_composer::element::FieldCoupling::setWeightsDirectory ( const std::string & dir )

Set the directory used for readWeights() / writeWeights().

◆ step()

void dnf_composer::element::FieldCoupling::step	(	double	t,
		double	deltaT
	)

overridevirtual

Advance the element by one time step.

Parameters

t	Current simulation time.
deltaT	Integration step size.

Implements dnf_composer::element::Element.

Here is the call graph for this function:

◆ toString()

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

overridevirtual

Implements dnf_composer::element::Element.

Here is the call graph for this function:

◆ tryReadWeights()

void dnf_composer::element::FieldCoupling::tryReadWeights ( )

Load weights if the file exists; log INFO in either case. Unlike readWeights(), this never logs an error — use it when weights may legitimately be absent (e.g. first run of a programmatic simulation).

Here is the call graph for this function:

◆ writeWeights()

void dnf_composer::element::FieldCoupling::writeWeights ( ) const

Save the current weight matrix to a binary file in weightsDirectory.

Here is the call graph for this function:

Member Data Documentation

◆ input

std::shared_ptr<Element> dnf_composer::element::FieldCoupling::input

protected

◆ output

std::shared_ptr<Element> dnf_composer::element::FieldCoupling::output

protected

◆ parameters

FieldCouplingParameters dnf_composer::element::FieldCoupling::parameters

protected

◆ weightsDirectory

std::string dnf_composer::element::FieldCoupling::weightsDirectory

protected

Directory used for weight serialization.

The documentation for this class was generated from the following files:

include/elements/field_coupling.h
src/elements/field_coupling.cpp

Public Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ FieldCoupling()

Member Function Documentation

◆ addInput()

◆ changeDimensions()

◆ changeInputDimensions()

◆ clearWeights()

◆ clone()

◆ getParameters()

◆ getWeightsDirectory()

◆ init()

◆ readWeights()

◆ setLearning()

◆ setLearningRate()

◆ setParameters()

◆ setWeightsDirectory()

◆ step()

◆ toString()

◆ tryReadWeights()

◆ writeWeights()

Member Data Documentation

◆ input

◆ output

◆ parameters

◆ weightsDirectory