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 cutility.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_UTILITY_H #define C_UTILITY_H #include <map> #include <assert.h> #include <newmat/newmat.h> //class Matrix; //class ColumnVector; int my_round(double value); double my_exp(double value); void getPseudoInverse(Matrix *J, Matrix *pinv, double lambda); /* class ColumnVector { protected: double *data; unsigned int dimensions; public: ColumnVector(unsigned int dimension, double *data = NULL); virtual ~ColumnVector(); void addVector(ColumnVector *vector); void addScalar(double scalar); void dotVector(ColumnVector *vector); void multMatrix(Matrix *matrix, ColumnVector *output); void multMatrix(Matrix *matrix); void multVector(ColumnVector *vector, Matrix *output); void multScalar(double scalar); double getLength(); void normalizeVector(); double scalarProduct(ColumnVector *vector); virtual double getDistance(ColumnVector *vector); double element(unsigned int index); void setElement(unsigned int index, double value); unsigned int getNumDimensions(); // bool isColumnVector(); double *getData(); void setVector(ColumnVector *vector); void initVector(double init); void saveASCII(FILE *stream); void loadASCII(FILE *stream); }; class Matrix { protected: double *data; unsigned int columns; unsigned int rows; public: Matrix(unsigned int rows, unsigned int columns, double *data = NULL); ~Matrix(); void addVector(ColumnVector *vector); void addScalar(double scalar); void multMatrix(Matrix *matrix, Matrix *output); void multVector(ColumnVector *vector, ColumnVector *output); void multScalar(double scalar); double element(unsigned int row, unsigned int column); void setElement(unsigned int row, unsigned int column, double value); double *getRow(unsigned int row); unsigned int nrows(); unsigned int ncols(); double *getData(); void setMatrix(Matrix *matrix); void initMatrix(double init); void saveASCII(FILE *stream); };*/ template <typename T1> class CMyArray { protected: T1 *data; int *dim; int numDim; int size; CMyArray() {}; void initialize(int numDim, int dim[]) { this->numDim = numDim; this->dim = new int[numDim]; memcpy(this->dim, dim, numDim * sizeof(int)); size = 1; for (int i = 0; i < numDim; i++) { size = size * dim[i]; } data = new T1 [size]; } public: CMyArray(int numDim, int dim[]) { CMyArray<T1>::initialize(numDim, dim); } ~CMyArray() { delete dim; delete data; } T1 get(int indices[]) { int index = 0, size = 1; for (int i = 0; i < numDim; i++) { assert(indices[i] < dim[i] && indices[i] >=0); index += indices[i] * size; size = size * dim[i]; } return data[index]; } void set(int indices[], T1 d) { int index = 0, size = 1; for (int i = 0; i < numDim; i++) { assert(indices[i] < dim[i] && indices[i] >=0); index += indices[i] * size; size = size * dim[i]; } data[index] = d; } void init(T1 initVal) { for (int i = 0; i < size; i++) { data[i] = initVal; } } int getSize() { return size; } void set1D(int index1d, T1 d) { data[index1d] = d; } T1 get1D(int index1d) { return data[index1d]; } }; class CDistributions { public: static void getGibbsDistribution(double beta, double *values, unsigned int numValues); static void getS1L0Distribution(double *values, unsigned int numValues); static double getNormalDistributionSample(double mu, double sigma); static int getSampledIndex(double *distribution, int numValues); }; template<typename T1> class CMyArray2D : public CMyArray<T1> { public: CMyArray2D(int xDim, int yDim) { int *l_dim = new int[2]; l_dim[0] = xDim; l_dim[1] = yDim; CMyArray<T1>::initialize(2, l_dim); delete l_dim; } ~CMyArray2D() { } T1 get(int xIndex, int yIndex) { int indices[2]; indices[0] = xIndex; indices[1] = yIndex; return CMyArray<T1>::get(indices); } void set(int xIndex, int yIndex, T1 d) { int indices[2]; indices[0] = xIndex; indices[1] = yIndex; CMyArray<T1>::set(indices, d); } }; template<typename T1> class CMyArray3D : public CMyArray<T1> { public: CMyArray3D(int xDim, int yDim, int zDim) : CMyArray<T1>() { int *ldim = new int[3]; ldim[0] = xDim; ldim[1] = yDim; ldim[2] = zDim; CMyArray<T1>::initialize(3, ldim); } ~CMyArray3D() { } T1 get(int xIndex, int yIndex, int zIndex) { int indices[3]; indices[0] = xIndex; indices[1] = yIndex; indices[2] = zIndex; return CMyArray<T1>::get(indices); } void set(int xIndex, int yIndex, int zIndex, T1 d) { int indices[3]; indices[0] = xIndex; indices[1] = yIndex; indices[2] = zIndex; CMyArray<T1>::set(indices, d); } }; // Maps feature indices to feature factors class CFeatureMap : public std::map<int, double> { protected: double stdValue; public: CFeatureMap(double stdValue = 0.0); double getValue(unsigned int featureIndex); }; /* class CFeatureSparse { protected: CMyArray<CFeatureList *> *sparse; double stdValue; char *sparseName; int numDim; int *dim; void initSparse(int numDim, int dim[]); CFeatureSparse(); public: CFeatureSparse(FILE *file, int numDim = 0, int *dim = NULL); CFeatureSparse(int numDim, int dim[]); virtual ~CFeatureSparse(); virtual double getFactor(int indeces[], unsigned int featureIndex); void setFactor(double factor, int indeces[], unsigned int featureIndex); void addFactor(double factor, int indeces[], unsigned int featureIndex); void loadASCII(FILE *stream); CFeature *getCFeature(int indeces[], unsigned int featureIndex); CFeature *getCFeature1D(int index1D, unsigned int featureIndex); virtual CFeatureList *getFeatureList(int indeces[]); void saveASCII(FILE *stream); }; //so there a a 2 dimensional array of feature lists. class CFeatureSparse2D : public CFeatureSparse { public: CFeatureSparse2D(FILE *file); CFeatureSparse2D(int dim1, int dim2); virtual ~CFeatureSparse2D(); virtual double getFactor(int ind1, int ind2, unsigned int featureIndex); virtual void setFactor(double factor, int ind1, int ind2, unsigned int featureIndex); virtual void addFactor(double factor, int ind1, int ind2, unsigned int featureIndex); virtual CFeatureList *getFeatureList(int ind1, int ind2); CFeature* getCFeature(int ind1, int ind2, unsigned int featureIndex); };*/ #endif