ann.h

遗传算法和神经网络结合

/*
  YAKS, Optann, a Khepera simulator including a separate GA and ANN 
  (Genetic Algoritm, Artificial Neural Net).
  Copyright (C) 2000  Johan Carlsson ([e-mail])
  
  This program is free software; you can redistribute it and/or
  modify it under the terms of the GNU General Public License
  as published by the Free Software Foundation; either version 2
  of the License, or any later version.
  
  This program is distributed in the hope that it will be useful,
  but WITHOUT ANY WARRANTY; without even the implied warranty of
  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  GNU General Public License for more details.
  
  You should have received a copy of the GNU General Public License
  along with this program; if not, write to the Free Software
  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.

 */

#ifndef ANN_H
#define ANN_H
#include <stdlib.h>
#include <stdio.h>
#include <iostream.h>
#include "neuron.h"
#include "balloc/balloc.h"

/*! \struct valuesS ann.h
 * \brief Structure for storing doubles.
 *
 * Used in various returns.
 */
typedef struct{
  /*! \var valuesS::nodes Number of doubles contained in the values list. */
  int nodes;
  /*! \var valuesS::values An array of 'nodes' doubles. */
  double *values;
}valuesS;


/** @class Ann ann.h
 * @brief ANN - Artificial Neural Net class, contains and maintains Neurons.
 *
 * This Neural Net implementation is based on the idee that all neurons has an
 * activation priority that determines when a neuron is activated. Neurons are
 * activated in rising order so the neuron with the smallest priority get 
 * activated first. Neurons can all have unique prioritys or group of neurons 
 * can share the same priority and by doing so form a "layer".
 *
 * @see Neuron
 */
class Ann{
 public:
  Ann();
  Ann(int dontCare);
  ~Ann();
  class Ann *duplicate() const;
  void addNeuron(int iLayerPriority, NODE_T inType, ACTIVATION_T iFuncT);
  void removeNeuron(int index);
  void connectNeurons(class Neuron *from, class Neuron *to, int initWeight);
  void connectLayer(int fromPri, int toPri);
  void connectLayerWithSCN(int fromPri, int toPri, int scnPri);
  void removeConnection(class Neuron *from, class Neuron *to);
  void copyWeightsFrom(class Ann *from);
  void mutateNeurons(int bitMutateProbability);
  void mutateNeuronsWithPri(int bitMutateProbability, int pri);
  valuesS *getValuesFromNodesWithPriority(int pri);
  void activateNet();
  void resetNet();
  int8_t getLinkWeight(int neuron,int link) const;
  int8_t *getLinkWeightPointer(int neuron,int link);	
  int getNrOfLinks(int neuron) const;
  double getOutput(int index)const;
  double *getOutputPtr(int index);
  void setInput(int index,double value);
  double getInput(int index) const;
  double *getInputPointer(int index);
  double *getActivationPointer(int index);
  int getNrOfInputNodes() const;
  int getNrOfOutputNodes() const;
  int getNrOfNodes() const;
  void printNet() const;
  void saveWeights(FILE *fp) const;
  void loadWeights(FILE *fp);
  void compareAnn(class Ann *net) const;
 private:
  int  countSCNNodesWithPri(int scnPri) const;
  int  countUnUsedSCNNodesWithPri(int scnPri) const;
  int  countNodesWithPri(int cntPri) const;
  int  nextUnUsedSCNNodeWithPri(int scnPri) const;
  void addNeuron(int iLayerPriority, NODE_T inType, ACTIVATION_T iFuncT, long int id);
  void addInNode(class Neuron *node);
  void addOutNode(class Neuron *node);
  void addSCNNode(class Neuron *node);
  void sort();
  /** Pointers to the actual neurons in an unsorted order. */
  Neuron **neurons;
  /** Help pointers to output **neurons. */
  Neuron **outNodes;
  /** Help pointers to input **neurons.  */
  Neuron **inNodes; 
  /** Help pointers to **neurons sorted in activation order. */
  Neuron **actOrder;
  /** Help pointers to SCN **neurons.    */
  Neuron **SCNNodes;
  /** Total number of neurons, includes all neurons even bias. */
  int nrOfNeurons;
  /** Total number of neurons declared as input neurons. */
  int nrOfInputs;
  /** Total number of neurons declared as output neurons. */
  int nrOfOutputs;
  /** Total number of neurons declared as SCN neurons. */
  int nrOfSCNs;
};
#endif /* ANN_H */