---

annie::RecurrentNeuron Class Reference

#include <RecurrentNeuron.h>

Inheritance diagram for annie::RecurrentNeuron:

List of all members.

Public Methods
	RecurrentNeuron (int label, bool hasBias=false)
virtual void	step ()
virtual real	getNextOutput ()
virtual real	getOutput ()
virtual real	getLastOutput ()
	The output of the neuron at the previous time step. More...
virtual const char *	getClassName ()
	Returns "RecurrentNeuron". More...
virtual int	getTime ()
virtual void	reset (real initialActivation)
virtual void	calculateNewWeights (real learningRate)
	Not yet implemented. More...
virtual void	update ()
	Not yet implemented. More...
virtual void	setDesiredOutput (real desired)
	Not yet implemented. More...
Protected Methods
virtual void	_recacheOutput ()
	See also: Neuron::_recacheOutput. More...
Protected Attributes
int	_time
	Keeps track of the "time", i.e., number of steps taken by this neuron since the last reset. More...
real	_initialOutput
	The output of the neuron at time 0. More...
real	_lastOutput
	The last output of the neuron. More...

---

Detailed Description

A neuron used for recurrent networks. These neurons have a concept of time. Thus, their output starts from an initial state and then as time progresses their output may change.

Recurrent networks allow cycles in the graph formed by connections between neurons, which are not allowed by simple multi-layer networks. For example, consider a network in which a recurrent neuron is connected to itself. Output now becomes time dependent. output(time=0) = an initial, fixed value. output(time=1) = weight_of_link * output(0) output(time=t) = weight_of_link * output(t-1) etc.

A recurrent neuron has all the features of a simple neuron and adds the concept of time, hence the RecurrentNeuron class is a sub-class of the SimpleNeuron class

---

Constructor & Destructor Documentation

annie::RecurrentNeuron::RecurrentNeuron (  int    label, bool    hasBias = false )

Creates a recurrent neuron. The default initial value is 0, thus at time=0 the output of the neuron will be 0. To change use reset Parameters: label  The label to give to the neuron hasBias  Allow this neuron to have a bias? See also: reset

---

Member Function Documentation

virtual void annie::RecurrentNeuron::_recacheOutput (    )  [protected, virtual]

See also: Neuron::_recacheOutput. Reimplemented from annie::SimpleNeuron.

virtual void annie::RecurrentNeuron::calculateNewWeights (  real    learningRate )  [virtual]

Not yet implemented.

virtual const char* annie::RecurrentNeuron::getClassName (    )  [virtual]

Returns "RecurrentNeuron". Reimplemented from annie::SimpleNeuron.

virtual real annie::RecurrentNeuron::getLastOutput (    )  [virtual]

The output of the neuron at the previous time step.

virtual real annie::RecurrentNeuron::getNextOutput (    )  [virtual]

Steps the neuron in time and returns the next output. See also: step , getOutput , getTime Returns: The output of the neuron at the next time step

virtual real annie::RecurrentNeuron::getOutput (    )  [virtual]

Returns the output of the neuron. At time 0 this will be the initial value which is set using reset(), and which is zero by default. At other times the output will be calculated Returns: the output of this neuron at the current time See also: getTime Reimplemented from annie::Neuron.

virtual int annie::RecurrentNeuron::getTime (    )  [virtual]

Returns the current time as known by the neuron. Initially 0, this value can change by calls to step and getNextOutput See also: step , getNextOutput

virtual void annie::RecurrentNeuron::reset (  real    initialActivation )  [virtual]

Resets the time to 0. Parameters: initialActivation  The activation of the neuron at time 0. The output at time 0 will be the activatation function applied on this value

virtual void annie::RecurrentNeuron::setDesiredOutput (  real    desired )  [virtual]

Not yet implemented. Reimplemented from annie::SimpleNeuron.

virtual void annie::RecurrentNeuron::step (    )  [virtual]

Updates the time in the neuron by one unit. This updation will ensure that all other neurons providing input to this one (directly by a connection or indirectly through a series of connections) also are brought up to date with the current time

virtual void annie::RecurrentNeuron::update (    )  [virtual]

Not yet implemented. Reimplemented from annie::SimpleNeuron.

---

Member Data Documentation

real annie::RecurrentNeuron::_initialOutput [protected]

The output of the neuron at time 0.

real annie::RecurrentNeuron::_lastOutput [protected]

The last output of the neuron.

int annie::RecurrentNeuron::_time [protected]

Keeps track of the "time", i.e., number of steps taken by this neuron since the last reset.

---

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

• RecurrentNeuron.h

---