#include <mtrnnInterface.h>

Inheritance diagram for MTRNNInterface:

Public Slots
void	sendParameters (float learningRate, float momentum, float weightRange, float threshold, int iterations, int seed, int numFastNeurons, int numSlowNeurons, int ioDeltaT, int fastDeltaT, int slowDeltaT, int debuggingLevel, int plotUpdateInterval)
	Sends training file data to remotehost running module.

Public Slots inherited from Interface
void	sendAbortRequest ()
	Sends request to abort module.

void	sendStartRequest ()
	Sends request to start module.

void	sendStopRequest ()
	Sends request to stop module.

void	sendSimulationMode (int simulationMode)
	Sends simulation module to module.

Signals
void	errorReceived (int iteration, float error)

void	networkReceived (QVector< float > network)

void	errorsReceived (QVector< float > errors)

void	parametersReceived (float learningRate, float momentum, float weightRange, float threshold, int iterations, int seed, int numFastNeurons, int numSlowNeurons, int ioDeltaT, int fastDeltaT, int slowDeltaT, int debuggingLevel, int plotUpdateInterval)

Signals inherited from Interface
void	gpuListReceived (QStringList gpuList)

void	progressReceived (int progress)

void	statusReceived (int statusID)

Public Member Functions
	MTRNNInterface (GUI *pGUI)
	Constructor.

void	sendTrainingFileData (int numControlNeurons, int numLinguisticNeurons, int numVisionNeurons, int numActionNeurons, int numSequences, int sequenceWidth, float minValue, float maxValue, QVector< int > seqenceSteps, QVector< float > sequenceData, int maxSequenceSteps, int totalSequenceSteps)
	Sends training file data to remotehost running module.

void	sendTrainingRequest (QString fileName)
	Sends training request to localhost module.

void	sendTrainingRequest ()
	Sends training request to remotehost module.

void	sendSaveRequest (QString fileName)
	Sends request to module to save neural network.

void	sendNetworkRequest ()
	Sends request to module to send back neural network.

Public Member Functions inherited from Interface
	Interface (GUI *pGUI)
	Constructor.

void	sendGpuListRequest ()
	Sends request for list of GPU devices found on module server.

void	sendCpuRequest ()
	Sends CPU request to module, which will updates its execution mode to CPU.

void	sendParametersRequest ()
	Sends parameters request to module.

void	sendQuitRequest ()
	Sends request to quit module.

void	close ()
	Closes module.

void	sendGpuID (int gpuID)
	Sends ID of selected GPU device to module.

void	sendGpuIDs (QVector< int > gpuIDs)
	Sends ID of selected GPU devices to module.

	Interface ()
	Constructor.

void	sendMessage (string message)
	Sends message to the output port.

void	sendGpuList ()
	Sends a list of detected GPU devices to the output port.

void	sendParameters ()
	Sends default parameters to the output port.

void	sendStatus (int status)
	Sends module status to the output port.

bool	configure (ResourceFinder &rf)
	Configures module.

	Interface ()

void	sendMessage (string message)

void	sendGpuList ()

void	sendParameters ()

void	sendStatus (int status)

bool	configure (ResourceFinder &rf)

	Interface ()

void	sendStatus (int status)

void	sendMessage (string message)

void	sendGpuList ()

void	sendParameters ()

void	sendActivity ()
	Sends activity.

bool	configure (ResourceFinder &rf)

	Interface ()

void	sendStatus (int status)

void	sendMessage (string message)

void	sendGpuList ()

void	sendParameters (float learningRate, float momentum, float threshold, float weightRange, int iterations, int seed, int numFastNeurons, int numSlowNeurons, int ioDeltaT, int fastDeltaT, int slowDeltaT, int debuggingLevel, int portUpdateInterval)
	Sends parameters to the output port.

void	sendError (int iteration, float error)
	Sends error at particular iteration to the output port.

void	sendProgress (int progress)
	Sends training progress (in %) to the output port.

void	sendNetwork ()
	Sends the weights of neural network to the output port.

void	sendErrors ()
	Sends the final training errors the output port.

bool	configure (ResourceFinder &rf)

	Interface ()

void	sendParameters (float learningRate, int numSubIterations, int numOutputs, int iterationPause)
	Sends parameters to the output port.

void	sendStatus (int status)

void	sendMap ()
	Sends the weights of self-organising map to the output port.

void	sendMessage (string message)

void	sendGpuDeviceList ()
	Sends a list of detected GPU devices to the output port.

void	sendNumInputs ()
	Sends number of inputs to the output port.

void	sendTrainingTime ()
	Sends the total training time (in ms) to the output port.

void	sendDataPointLimits ()
	Sends the original data min and max values to the output port.

void	sendProgress (int progress)

bool	configure (ResourceFinder &rf)

	Interface ()

void	sendMessage (string message)

void	sendGpuList ()

void	sendParameters ()

void	sendStatus (int status)

bool	configure (ResourceFinder &rf)

Protected Member Functions
void	processPortData ()
	Processes data received from module.

Protected Member Functions inherited from Interface
virtual void	processGpuList ()
	Processes data about GPU devices found on server running module.

virtual void	printMessage ()
	Prints out or logs message received from module.

Additional Inherited Members
Public Attributes inherited from Interface
Image	leftCam

Image	rightCam

Image	leftMotion

Image	rightMotion

Protected Attributes inherited from Interface
GUI *	module

yarp::os::Port	inputPort

yarp::os::Port	outputPort

yarp::os::Bottle	receivedBottle

Detailed Description

Interface between MTRNN GUI and MTRNN module

Constructor & Destructor Documentation

MTRNNInterface::MTRNNInterface ( GUI * pGUI )

Constructor.

Parameters

[in] pGUI - pointer to GUI

Member Function Documentation

void MTRNNInterface::errorReceived	(	int	iteration,
		float	error
	)

signal

void MTRNNInterface::errorsReceived ( QVector< float > errors )

signal

void MTRNNInterface::networkReceived ( QVector< float > network )

signal

void MTRNNInterface::parametersReceived	(	float	learningRate,
		float	momentum,
		float	weightRange,
		float	threshold,
		int	iterations,
		int	seed,
		int	numFastNeurons,
		int	numSlowNeurons,
		int	ioDeltaT,
		int	fastDeltaT,
		int	slowDeltaT,
		int	debuggingLevel,
		int	plotUpdateInterval
	)

signal

void MTRNNInterface::processPortData ( )

protectedvirtual

Processes data received from module.

Reimplemented from Interface.

void MTRNNInterface::sendNetworkRequest ( )

Sends request to module to send back neural network.

void MTRNNInterface::sendParameters	(	float	learningRate,
		float	momentum,
		float	weightRange,
		float	threshold,
		int	iterations,
		int	seed,
		int	numFastNeurons,
		int	numSlowNeurons,
		int	ioDeltaT,
		int	fastDeltaT,
		int	slowDeltaT,
		int	debuggingLevel,
		int	plotUpdateInterval
	)

slot

Sends training file data to remotehost running module.

Parameters

[in]	learningRate	- learning rate
[in]	momentum	- momentum
[in]	weightRange	- initial range of weights
[in]	threshold	- error threshold
[in]	iterations	- number of iterations
[in]	seed	- random number generator seed
[in]	numFastNeurons	- number of fast neurons
[in]	numSlowNeurons	- number of slow neurons
[in]	ioDeltaT	- input-output delta-t value
[in]	fastDeltaT	- fast delta-t value
[in]	slowDeltaT	- slow delta-t value
[in]	debuggingLevel	- debugging level
[in]	plotUpdateInterval	- update interval of the error plot

void MTRNNInterface::sendSaveRequest ( QString fileName )

Sends request to module to save neural network.

Parameters

[in] fileName - file name used for saving of neural network

void MTRNNInterface::sendTrainingFileData	(	int	numControlNeurons,
		int	numLinguisticNeurons,
		int	numVisionNeurons,
		int	numActionNeurons,
		int	numSequences,
		int	sequenceWidth,
		float	minValue,
		float	maxValue,
		QVector< int >	seqenceSteps,
		QVector< float >	sequenceData,
		int	maxSequenceSteps,
		int	totalSequenceSteps
	)

Sends training file data to remotehost running module.

Parameters

[in]	numControlNeurons	- number of control neurons
[in]	numLinguisticNeurons	- number of linguistic neurons
[in]	numVisionNeurons	- number of vision neurons
[in]	numActionNeurons	- number of action neurons
[in]	numSequences	- number training sequences
[in]	sequenceWidth	- number of elements in each step of the sequence
[in]	minValue	- minimum value present in the original training data
[in]	maxValue	- maximum value present in the original training data
[in]	seqenceSteps	- number of steps in each sequence
[in]	seqenceData	- data containing all sequences
[in]	maxSequenceSteps	- number of steps in the longest sequences
[in]	totalSequenceSteps	- total number of steps when all sequences are joint

void MTRNNInterface::sendTrainingRequest ( QString fileName )

Sends training request to localhost module.

Parameters

[in] fileName - name of the training file

Note: This function is called when a localhost module is asked to load a training file.; Remotehost modules need to be sent training data via port using sendTrainingFileData function.

void MTRNNInterface::sendTrainingRequest ( )

Sends training request to remotehost module.

Note: This function is called after remotehost module was initialised with the training data using sendTrainingFileData function.

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

C:/Users/mpeniak/Dropbox/Aquila/trunk/aquila/include/mtrnnInterface.h
C:/Users/mpeniak/Dropbox/Aquila/trunk/aquila/src/mtrnnInterface.cpp

Public Slots

Signals

Public Member Functions

Protected Member Functions

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation