Elman Class Reference

Multilayer Neural Network with context neurons (after Elman) Example of 2 hidden layer network O O O | | | H H H ----->-----+ 1:1 fixed connections (time delayed) | | |\-<-< __ | | | | \ \ \/ |l | l = lambda -> recurrent connection | | | C C C<-+ | | | | ^-^-^--<--+ H H H ----->-----+ 1:1 fixed connections (time delayed) | | |\-<-< __ | | | | \ \ \/ |l | l = lambda -> recurrent connection I I I C C C<-+ | ^-^-^--<--+. More...

#include <elman.h>

Inherits MultiLayerFFNN.

Inheritance diagram for Elman:

[legend]Collaboration diagram for Elman:

[legend]List of all members.

Public Member Functions
	Elman (double eps, const std::vector< Layer > &layers, double lambda)
virtual	~Elman ()
virtual void	init (unsigned int inputDim, unsigned int outputDim, double unit_map=0.0)
	initialisation of the network with the given number of input and output units
virtual const matrix::Matrix	process (const matrix::Matrix &input)
	passive processing of the input (this will be different for every input, since it is a recurrent network)
virtual const matrix::Matrix	learn (const matrix::Matrix &input, const matrix::Matrix &nom_output, double learnRateFactor=1)
	performs learning and returns the network output before learning
bool	store (FILE *f) const
	stores the layer binary into file stream
bool	restore (FILE *f)
	restores the layer binary from file stream
virtual paramkey	getName () const
	return the name of the object
Protected Attributes
std::vector< matrix::Matrix >	contextweights
std::vector< matrix::Matrix >	contexts
double	lambda

---

Detailed Description

Multilayer Neural Network with context neurons (after Elman) Example of 2 hidden layer network O O O | | | H H H ----->-----+ 1:1 fixed connections (time delayed) | | |\-<-< __ | | | | \ \ \/ |l | l = lambda -> recurrent connection | | | C C C<-+ | | | | ^-^-^--<--+ H H H ----->-----+ 1:1 fixed connections (time delayed) | | |\-<-< __ | | | | \ \ \/ |l | l = lambda -> recurrent connection I I I C C C<-+ | ^-^-^--<--+.

---

Constructor & Destructor Documentation

Elman (  double  eps, const std::vector< Layer > &  layers, double  lambda )  [inline]

Parameters: eps  learning rate layers  Layer description (the input layer is not specified (always linear)) lambda  self-recurrent feedback strength of context neurons

virtual ~Elman (   )  [inline, virtual]

---

Member Function Documentation

virtual paramkey getName (   )  const [inline, virtual]

return the name of the object Reimplemented from MultiLayerFFNN.

void init (  unsigned int  inputDim, unsigned int  outputDim, double  unit_map = 0.0 )  [virtual]

initialisation of the network with the given number of input and output units Reimplemented from MultiLayerFFNN.

const Matrix learn (  const matrix::Matrix &  input, const matrix::Matrix &  nom_output, double  learnRateFactor = 1 )  [virtual]

performs learning and returns the network output before learning Reimplemented from MultiLayerFFNN.

const Matrix process (  const matrix::Matrix &  input  )  [virtual]

passive processing of the input (this will be different for every input, since it is a recurrent network) Reimplemented from MultiLayerFFNN.

bool restore (  FILE *  f  )  [virtual]

restores the layer binary from file stream Reimplemented from MultiLayerFFNN.

bool store (  FILE *  f  )  const [virtual]

stores the layer binary into file stream See also: InvertableModel::response Reimplemented from MultiLayerFFNN.

---

Member Data Documentation

std::vector<matrix::Matrix> contexts [protected]

std::vector<matrix::Matrix> contextweights [protected]

double lambda [protected]

---

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

• elman.h
• elman.cpp

---