OptimalGainsFinder Class Reference

Compute the optimal gains that maximize SNR. More...

#include <Channels/OptimalGains.H>

Inheritance diagram for OptimalGainsFinder:

[legend]

Collaboration diagram for OptimalGainsFinder:

[legend]

List of all members.

Public Member Functions
	OptimalGainsFinder (const Image< byte > &targetMask, const Image< byte > &distractorMask, rutz::shared_ptr< ParamMap > pmap, const bool doMax=true)
	Construct with settings for the algorithm.
virtual	~OptimalGainsFinder ()
	Virtual destructor.
virtual void	visitChannelBase (ChannelBase &chan)
	Visit the given ChannelBase node.
virtual void	visitSingleChannel (SingleChannel &chan)
	Visit the given SingleChannel node.
virtual void	visitComplexChannel (ComplexChannel &chan)
	Visit the given ComplexChannel node.
virtual rutz::shared_ptr < ParamMap >	pmap () const
	Get our pmap, it will contain saliencsT and salienceD values after visit.

---

Detailed Description

Compute the optimal gains that maximize SNR.

This ChannelVisitor computes the salience of the target and distractors using the masks as cues for their locations. Based on the salience, it finds SNR and the gains; however, SNR and gains are only used internally (they are needed to compute the salience at the next level up the channel hierarchy) but are not saved here. Rather, the salienceT / salienceD info is saved. One can then use app-combineOptimalGains to compute the gains, and a PrefrontalCortexGS to use them.

Definition at line 56 of file OptimalGains.H.

---

Constructor & Destructor Documentation

OptimalGainsFinder::OptimalGainsFinder	(	const Image< byte > &	targetMask,
		const Image< byte > &	distractorMask,
		rutz::shared_ptr< ParamMap >	pmap,
		const bool	doMax = true
	)

Construct with settings for the algorithm.

Definition at line 54 of file OptimalGains.C.

OptimalGainsFinder::~OptimalGainsFinder

(

)

[virtual]

Virtual destructor.

Definition at line 66 of file OptimalGains.C.

---

Member Function Documentation

rutz::shared_ptr< ParamMap > OptimalGainsFinder::pmap

(

)

const [virtual]

Get our pmap, it will contain saliencsT and salienceD values after visit.

Definition at line 249 of file OptimalGains.C.

void OptimalGainsFinder::visitChannelBase

(

ChannelBase &

chan

)

[virtual]

Visit the given ChannelBase node.

Implements ChannelVisitor.

Definition at line 70 of file OptimalGains.C.

References ModelComponent::tagName().

void OptimalGainsFinder::visitComplexChannel

(

ComplexChannel &

chan

)

[virtual]

Visit the given ComplexChannel node.

Implements ChannelVisitor.

Definition at line 154 of file OptimalGains.C.

References ChannelFacetMap::getFacet(), ComplexChannel::getMapDims(), getMaskedMinMax(), getMaskedMinMaxAvg(), ComplexChannel::getOutput(), ChannelFacetMap::hasFacet(), Image< T >::initialized(), ComplexChannel::killCaches(), ComplexChannel::numChans(), OPTIGAIN_BG_FIRING, rescale(), rutz::shared_ptr< T >::reset(), ChannelFacetMap::setFacet(), sformat(), ComplexChannel::subChan(), and rutz::shared_ptr< T >::swap().

void OptimalGainsFinder::visitSingleChannel

(

SingleChannel &

chan

)

[virtual]

Visit the given SingleChannel node.

Implements ChannelVisitor.

Definition at line 74 of file OptimalGains.C.

References ChannelFacetMap::getFacet(), SingleChannel::getLevelSpec(), SingleChannel::getMapDims(), getMaskedMinMax(), getMaskedMinMaxAvg(), SingleChannel::getOutput(), SingleChannel::getSubmap(), ChannelFacetMap::hasFacet(), LevelSpec::indexToCS(), Image< T >::initialized(), SingleChannel::killCaches(), SingleChannel::numSubmaps(), OPTIGAIN_BG_FIRING, rescale(), rutz::shared_ptr< T >::reset(), ChannelFacetMap::setFacet(), sformat(), and ModelComponent::tagName().

---

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

• src/Channels/OptimalGains.H
• src/Channels/OptimalGains.C