Reinforcement Learning Toolbox 2.0 last updated:

General Documentation Manual Tutorial Class Reference Master Thesis Examples Related Papers Downloads Links News mailto:webmaster Main Page     Class Hierarchy   Compound List   File List   Compound Members   File Members crewardmodel.h Go to the documentation of this file. // Copyright (C) 2003 // Gerhard Neumann (gneumann@gmx.net) // Stephan Neumann (sneumann@gmx.net) // // This file is part of RL Toolbox. // http://www.igi.tugraz.at/ril_toolbox // // All rights reserved. // // Redistribution and use in source and binary forms, with or without // modification, are permitted provided that the following conditions // are met: // 1. Redistributions of source code must retain the above copyright // notice, this list of conditions and the following disclaimer. // 2. Redistributions in binary form must reproduce the above copyright // notice, this list of conditions and the following disclaimer in the // documentation and/or other materials provided with the distribution. // 3. The name of the author may not be used to endorse or promote products // derived from this software without specific prior written permission. // // THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR // IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES // OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. // IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, // INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT // NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF // THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. #ifndef C_REWARDMODEL_H #define C_REWARDMODEL_H #include <map> #include <vector> #include "crewardfunction.h" #include "clearndataobject.h" #include "cbaseobjects.h" #include "cutility.h" #include "cagentlistener.h" class CAbstractFeatureStochasticEstimatedModel; class CFeatureStateRewardFunction : public CFeatureRewardFunction { protected: std::map<int, double> *rewards; public: CFeatureStateRewardFunction(CStateProperties *discretizer); virtual ~CFeatureStateRewardFunction(); virtual double getReward(int oldState, CAction *action, int newState); virtual double getReward(int state); virtual void setReward(int state, double reward); }; class CFeatureRewardModel : public CFeatureRewardFunction, public CSemiMDPRewardListener, public CActionObject, public CLearnDataObject { protected: CMyArray2D<CFeatureMap *> *rewardTable; CMyArray2D<CFeatureMap *> *visitTable; CAbstractFeatureStochasticEstimatedModel *model; bool bExternVisitSparse; double getTransitionVisits(int oldState, int action, int newState); public: CFeatureRewardModel(CActionSet *actions, CRewardFunction *function, CAbstractFeatureStochasticEstimatedModel *model, CStateModifier *discretizer); CFeatureRewardModel(CActionSet *actions, CRewardFunction *function, CStateModifier *discretizer); virtual ~CFeatureRewardModel(); virtual double getReward(int oldState, CAction *action, int newState); virtual void nextStep(CStateCollection *oldState, CAction *action, double reward, CStateCollection *newState); virtual void saveData(FILE *stream); virtual void loadData(FILE *stream); virtual void resetData(); }; class CFeatureStateRewardModel : public CFeatureRewardFunction, public CSemiMDPRewardListener, public CLearnDataObject { protected: double *rewards; double *visits; double rewardMean; int numRewards; public: CFeatureStateRewardModel(CRewardFunction *function, CStateModifier *discretizer); virtual ~CFeatureStateRewardModel(); virtual double getReward(CState *oldState, CAction *action, CState *newState); virtual double getReward(int oldState, CAction *action, int newState); virtual double getReward(int newState); virtual void nextStep(CStateCollection *oldState, CAction *action, double reward, CStateCollection *newState); virtual void saveData(FILE *stream); virtual void loadData(FILE *stream); virtual void resetData(); }; class CRewardEpisode : public CSemiMDPRewardListener { protected: std::vector<double> *rewards; public: CRewardEpisode(CRewardFunction *rewardFunction); virtual ~CRewardEpisode(); virtual void nextStep(CStateCollection *oldState, CAction *action, double reward, CStateCollection *newState); virtual void newEpisode(); int getNumRewards(); double getReward(int index); double getMeanReward(); double getLastStepsMeanReward(int Steps); virtual void saveBIN(FILE *stream); virtual void saveData(FILE *stream); virtual void loadBIN(FILE *stream); virtual void loadData(FILE *stream); }; class CRewardHistory { protected: public: CRewardHistory() {}; virtual ~CRewardHistory(){}; virtual CRewardEpisode* getEpisode(int index) = 0; virtual int getNumEpisodes() = 0; }; class CRewardHistorySubset : public CRewardHistory { protected: CRewardHistory *episodes; std::vector<int> *indices; public: CRewardHistorySubset(CRewardHistory *episodes, std::vector<int> *indices); virtual ~CRewardHistorySubset(); virtual int getNumEpisodes(); virtual CRewardEpisode* getEpisode(int index); }; class CRewardLogger : public CSemiMDPRewardListener, public CLearnDataObject, public CRewardHistory { protected: char filename[512]; FILE* file; char loadFileName[512]; int holdMemory; std::list<CRewardEpisode *> *episodes; CRewardEpisode *currentEpisode; void init(); public: CRewardLogger(CRewardFunction *rewardFunction, char* autoSavefile, int holdMemory); CRewardLogger(CRewardFunction *rewardFunction); CRewardLogger(char *loadFile, CRewardFunction *rewardFunction); virtual ~CRewardLogger(); virtual void nextStep(CStateCollection *oldState, CAction *action, double reward, CStateCollection *nextState); virtual void newEpisode(); void setAutoSaveFile(char *filename); virtual void saveBIN(FILE *stream); virtual void saveData(FILE *stream); void loadBIN(FILE *stream, int episodes = -1); virtual void loadData(FILE *stream, int episodes = -1); virtual void loadData(FILE *stream); virtual int getNumEpisodes(); virtual CRewardEpisode* getCurrentEpisode(); virtual CRewardEpisode* getEpisode(int index); void clearAutoSaveFile(); void setLoadDataFile(char *loadData); virtual void resetData(); }; class CSemiMDPLastNRewardFunction : public CRewardFunction, public CRewardEpisode { protected: double gamma; public: CSemiMDPLastNRewardFunction(CRewardFunction *rewardFunction, double gamma); virtual ~CSemiMDPLastNRewardFunction(); virtual double getReward(CStateCollection *oldState, CAction *action, CStateCollection *newState); }; #endif // REWARDMODEL_H