Chaste Release::3.1
FourthOrderTensor.hpp
00001 /*
00002 
00003 Copyright (c) 2005-2012, University of Oxford.
00004 All rights reserved.
00005 
00006 University of Oxford means the Chancellor, Masters and Scholars of the
00007 University of Oxford, having an administrative office at Wellington
00008 Square, Oxford OX1 2JD, UK.
00009 
00010 This file is part of Chaste.
00011 
00012 Redistribution and use in source and binary forms, with or without
00013 modification, are permitted provided that the following conditions are met:
00014  * Redistributions of source code must retain the above copyright notice,
00015    this list of conditions and the following disclaimer.
00016  * Redistributions in binary form must reproduce the above copyright notice,
00017    this list of conditions and the following disclaimer in the documentation
00018    and/or other materials provided with the distribution.
00019  * Neither the name of the University of Oxford nor the names of its
00020    contributors may be used to endorse or promote products derived from this
00021    software without specific prior written permission.
00022 
00023 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
00024 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
00025 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
00026 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
00027 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
00028 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
00029 GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
00030 HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
00031 LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
00032 OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
00033 
00034 */
00035 
00036 #ifndef _FOURTHORDERTENSOR_HPP_
00037 #define _FOURTHORDERTENSOR_HPP_
00038 
00039 #include <cassert>
00040 #include <vector>
00041 
00042 #include <boost/numeric/ublas/vector.hpp>
00043 #include <boost/numeric/ublas/matrix.hpp>
00044 using namespace boost::numeric::ublas;
00045 #include "Exception.hpp"
00046 
00052 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00053 class FourthOrderTensor
00054 {
00055 private:
00056 
00057     std::vector<double> mData;  
00065     unsigned GetVectorIndex(unsigned M, unsigned N, unsigned P, unsigned Q)
00066     {
00067         assert(M<DIM1);
00068         assert(N<DIM2);
00069         assert(P<DIM3);
00070         assert(Q<DIM4);
00071         return M + DIM1*N + DIM1*DIM2*P + DIM1*DIM2*DIM3*Q;
00072     }
00073 
00074 public:
00075 
00079     FourthOrderTensor();
00080 
00089     template<unsigned CONTRACTED_DIM>
00090     void SetAsContractionOnFirstDimension(const c_matrix<double,DIM1,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<CONTRACTED_DIM,DIM2,DIM3,DIM4>& rTensor);
00091 
00092 
00101     template<unsigned CONTRACTED_DIM>
00102     void SetAsContractionOnSecondDimension(const c_matrix<double,DIM2,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<DIM1,CONTRACTED_DIM,DIM3,DIM4>& rTensor);
00103 
00112     template<unsigned CONTRACTED_DIM>
00113     void SetAsContractionOnThirdDimension(const c_matrix<double,DIM3,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<DIM1,DIM2,CONTRACTED_DIM,DIM4>& rTensor);
00114 
00123     template<unsigned CONTRACTED_DIM>
00124     void SetAsContractionOnFourthDimension(const c_matrix<double,DIM4,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<DIM1,DIM2,DIM3,CONTRACTED_DIM>& rTensor);
00125 
00134     double& operator()(unsigned M, unsigned N, unsigned P, unsigned Q);
00135 
00139     void Zero();
00140 
00144     std::vector<double>& rGetData()
00145     {
00146         return mData;
00147     }
00148 };
00149 
00151 // Implementation (lots of possibilities for the dimensions so no point with explicit instantiation)
00153 
00154 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00155 FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::FourthOrderTensor()
00156 {
00157     unsigned size = DIM1*DIM2*DIM3*DIM4;
00158     mData.resize(size, 0.0);
00159 }
00160 
00161 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00162 template<unsigned CONTRACTED_DIM>
00163 void FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::SetAsContractionOnFirstDimension(const c_matrix<double,DIM1,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<CONTRACTED_DIM,DIM2,DIM3,DIM4>& rTensor)
00164 {
00165     Zero();
00166 
00167     std::vector<double>::iterator iter = mData.begin();
00168     std::vector<double>::iterator other_tensor_iter = rTensor.rGetData().begin();
00169 
00170     for (unsigned d=0; d<DIM4; d++)
00171     {
00172         for (unsigned c=0; c<DIM3; c++)
00173         {
00174             for (unsigned b=0; b<DIM2; b++)
00175             {
00176                 for (unsigned a=0; a<DIM1; a++)
00177                 {
00178                     for (unsigned N=0; N<CONTRACTED_DIM; N++)
00179                     {
00180                         /*
00181                          * The following just does
00182                          *
00183                          * mData[GetVectorIndex(a,b,c,d)] += rMatrix(a,N) * rTensor(N,b,c,d);
00184                          *
00185                          * but more efficiently using iterators into the data vector, not
00186                          * using random access.
00187                          */
00188                         *iter += rMatrix(a,N) * *other_tensor_iter;
00189                         other_tensor_iter++;
00190                     }
00191 
00192                     iter++;
00193 
00194                     if (a != DIM1-1)
00195                     {
00196                         other_tensor_iter -= CONTRACTED_DIM;
00197                     }
00198                 }
00199             }
00200         }
00201     }
00202 }
00203 
00204 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00205 template<unsigned CONTRACTED_DIM>
00206 void FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::SetAsContractionOnSecondDimension(const c_matrix<double,DIM2,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<DIM1,CONTRACTED_DIM,DIM3,DIM4>& rTensor)
00207 {
00208     Zero();
00209 
00210     std::vector<double>::iterator iter = mData.begin();
00211     std::vector<double>::iterator other_tensor_iter = rTensor.rGetData().begin();
00212 
00213     for (unsigned d=0; d<DIM4; d++)
00214     {
00215         for (unsigned c=0; c<DIM3; c++)
00216         {
00217             for (unsigned b=0; b<DIM2; b++)
00218             {
00219                 for (unsigned N=0; N<CONTRACTED_DIM; N++)
00220                 {
00221                     for (unsigned a=0; a<DIM1; a++)
00222                     {
00223                         /*
00224                          * The following just does
00225                          *
00226                          * mData[GetVectorIndex(a,b,c,d)] += rMatrix(b,N) * rTensor(a,N,c,d);
00227                          *
00228                          * but more efficiently using iterators into the data vector, not
00229                          * using random access.
00230                          */
00231                         *iter += rMatrix(b,N) * *other_tensor_iter;
00232                         iter++;
00233                         other_tensor_iter++;
00234                     }
00235 
00236                     if (N != CONTRACTED_DIM-1)
00237                     {
00238                         iter -= DIM1;
00239                     }
00240                 }
00241                 if (b != DIM2-1)
00242                 {
00243                     other_tensor_iter -= CONTRACTED_DIM*DIM1;
00244                 }
00245             }
00246         }
00247     }
00248 }
00249 
00250 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00251 template<unsigned CONTRACTED_DIM>
00252 void FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::SetAsContractionOnThirdDimension(const c_matrix<double,DIM3,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<DIM1,DIM2,CONTRACTED_DIM,DIM4>& rTensor)
00253 {
00254     Zero();
00255 
00256     std::vector<double>::iterator iter = mData.begin();
00257     std::vector<double>::iterator other_tensor_iter = rTensor.rGetData().begin();
00258 
00259     for (unsigned d=0; d<DIM4; d++)
00260     {
00261         for (unsigned c=0; c<DIM3; c++)
00262         {
00263             for (unsigned N=0; N<CONTRACTED_DIM; N++)
00264             {
00265                 for (unsigned b=0; b<DIM2; b++)
00266                 {
00267                     for (unsigned a=0; a<DIM1; a++)
00268                     {
00269                         /*
00270                          * The following just does
00271                          *
00272                          * mData[GetVectorIndex(a,b,c,d)] += rMatrix(c,N) * rTensor(a,b,N,d);
00273                          *
00274                          * but more efficiently using iterators into the data vector, not
00275                          * using random access.
00276                          */
00277                         *iter += rMatrix(c,N) * *other_tensor_iter;
00278                         iter++;
00279                         other_tensor_iter++;
00280                     }
00281                 }
00282 
00283                 if (N != CONTRACTED_DIM-1)
00284                 {
00285                     iter -= DIM1*DIM2;
00286                 }
00287             }
00288 
00289             if (c != DIM3-1)
00290             {
00291                 other_tensor_iter -= CONTRACTED_DIM*DIM1*DIM2;
00292             }
00293         }
00294     }
00295 }
00296 
00297 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00298 template<unsigned CONTRACTED_DIM>
00299 void FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::SetAsContractionOnFourthDimension(const c_matrix<double,DIM4,CONTRACTED_DIM>& rMatrix, FourthOrderTensor<DIM1,DIM2,DIM3,CONTRACTED_DIM>& rTensor)
00300 {
00301     Zero();
00302 
00303     std::vector<double>::iterator iter = mData.begin();
00304     std::vector<double>::iterator other_tensor_iter = rTensor.rGetData().begin();
00305 
00306     for (unsigned d=0; d<DIM4; d++)
00307     {
00308         for (unsigned N=0; N<CONTRACTED_DIM; N++)
00309         {
00310             for (unsigned c=0; c<DIM3; c++)
00311             {
00312                 for (unsigned b=0; b<DIM2; b++)
00313                 {
00314                     for (unsigned a=0; a<DIM1; a++)
00315                     {
00316                         /*
00317                          * The following just does
00318                          *
00319                          * mData[GetVectorIndex(a,b,c,d)] += rMatrix(d,N) * rTensor(a,b,c,N);
00320                          *
00321                          * but more efficiently using iterators into the data vector, not
00322                          * using random access.
00323                          */
00324                         *iter += rMatrix(d,N) * *other_tensor_iter;
00325 
00326                         iter++;
00327                         other_tensor_iter++;
00328                     }
00329                 }
00330             }
00331 
00332             if (N != CONTRACTED_DIM-1)
00333             {
00334                 iter-= DIM1*DIM2*DIM3;
00335             }
00336         }
00337 
00338         other_tensor_iter -= CONTRACTED_DIM*DIM1*DIM2*DIM3;
00339     }
00340 }
00341 
00342 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00343 double& FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::operator()(unsigned M, unsigned N, unsigned P, unsigned Q)
00344 {
00345     assert(M<DIM1);
00346     assert(N<DIM2);
00347     assert(P<DIM3);
00348     assert(Q<DIM4);
00349 
00350     return mData[GetVectorIndex(M,N,P,Q)];
00351 }
00352 
00353 template<unsigned DIM1, unsigned DIM2, unsigned DIM3, unsigned DIM4>
00354 void FourthOrderTensor<DIM1,DIM2,DIM3,DIM4>::Zero()
00355 {
00356     for (unsigned i=0; i<mData.size(); i++)
00357     {
00358         mData[i] = 0.0;
00359     }
00360 }
00361 
00362 #endif //_FOURTHORDERTENSOR_HPP_