TREX/Sherman-Morrison
1
0
Fork 0
mirror of https://github.com/TREX-CoE/Sherman-Morrison.git synced 2024-12-26 06:15:08 +01:00
Code Issues Projects Releases Wiki Activity
8d63dd1701
Sherman-Morrison/Makefile

65 lines
1.8 KiB
Makefile
Raw Normal View History

The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
# ARCH = -xCORE-AVX2
Put more comments in Makefile.
2021-02-04 18:52:26 +01:00
## Used compilers
Add test infrastructure for datasets
2021-02-16 10:49:15 +01:00
H5CXX = h5c++
CXX = clang++
FC = flang
Put more comments in Makefile.
2021-02-04 18:52:26 +01:00
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
## Compiler flags & common obs & libs
CXXFLAGS = -O0 -debug full -traceback
FFLAGS = -O0 -debug full -traceback
FLIBS = -lstdc++
OBJS = SM_MaponiA3.o
Added QMCChem test dataset with 3 updates. Renamed files, added the Fortran program QMCChem_dataset_test.f90 to start testing the QMCChem dataset. Wrote and tested Fortran code to import the test dataset.
2021-02-12 12:04:21 +01:00
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
## Deps & objs for C++ cMaponiA3_test_3x3_3
cMaponiA3_test_3x3_3OBJ = cMaponiA3_test_3x3_3.o
## Deps & objs for Fortran fMaponiA3_test_3x3_3
fMaponiA3_test_3x3_3OBJ = SM_MaponiA3_mod.o fMaponiA3_test_3x3_3.o
## Deps & objs for Fortran fMaponiA3_test_4x4_2
fMaponiA3_test_4x4_2OBJ = SM_MaponiA3_mod.o fMaponiA3_test_4x4_2.o
Added QMCChem test dataset with 3 updates. Renamed files, added the Fortran program QMCChem_dataset_test.f90 to start testing the QMCChem dataset. Wrote and tested Fortran code to import the test dataset.
2021-02-12 12:04:21 +01:00
## Deps & objs for Fortran QMCChem_dataset_test
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
QMCChem_dataset_testOBJ = Utils_mod.o SM_MaponiA3_mod.o QMCChem_dataset_test.o
## Default build target: build everything
all: cMaponiA3_test_3x3_3 fMaponiA3_test_3x3_3 fMaponiA3_test_4x4_2 QMCChem_dataset_test
Added QMCChem test dataset with 3 updates. Renamed files, added the Fortran program QMCChem_dataset_test.f90 to start testing the QMCChem dataset. Wrote and tested Fortran code to import the test dataset.
2021-02-12 12:04:21 +01:00
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
## Compile recipes for C++
%.o: %.cpp
Prepared the main Fortran infrastructure for the C++ calls. Updated the makefile.
2021-02-04 11:39:00 +01:00
$(CXX) $(ARCH) $(CXXFLAGS) -c -o $@ $<
Added some initial data likle Makefiles and .gitignore.
2021-01-26 07:26:54 +01:00
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
## Compile recepies for Fortran
%.o: %.f90
Prepared the main Fortran infrastructure for the C++ calls. Updated the makefile.
2021-02-04 11:39:00 +01:00
$(FC) $(ARCH) $(FFLAGS) -c -o $@ $<
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
## Explicit recipe to trigger rebuild and relinking when headerfile is changed
SM_MaponiA3.o: SM_MaponiA3.cpp Helpers.hpp
$(CXX) $(ARCH) $(CXXFLAGS) -c -o $@ $<
Put more comments in Makefile.
2021-02-04 18:52:26 +01:00
## Build tagets
Prepared the example matrix of Example 8 of the paper and its decomposition in the Fortran code and made a basic call to the subroutine MYSUBROUTINE, which is bound to the C++ void function 'Sherman-Morrison();. For now compilation fails with lots of undefined references.'
2021-02-04 13:12:34 +01:00
.PHONY: all clean distclean
clean:
@rm -vf *.o *.mod
C++ redesign of data-structures - Use flat arrays - Use real type for all matrices - Merge _f MaponiA3 files
2021-02-09 13:40:52 +01:00
Prepared the example matrix of Example 8 of the paper and its decomposition in the Fortran code and made a basic call to the subroutine MYSUBROUTINE, which is bound to the C++ void function 'Sherman-Morrison();. For now compilation fails with lots of undefined references.'
2021-02-04 13:12:34 +01:00
distclean: clean
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
@rm -vf cMaponiA3_test_3x3_3 fMaponiA3_test_3x3_3 QMCChem_dataset_test Slater_* Updates.dat
Prepared the example matrix of Example 8 of the paper and its decomposition in the Fortran code and made a basic call to the subroutine MYSUBROUTINE, which is bound to the C++ void function 'Sherman-Morrison();. For now compilation fails with lots of undefined references.'
2021-02-04 13:12:34 +01:00
## Linking the C++ example program
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
cMaponiA3_test_3x3_3: $(cMaponiA3_test_3x3_3OBJ) $(OBJS)
Prepared the main Fortran infrastructure for the C++ calls. Updated the makefile.
2021-02-04 11:39:00 +01:00
$(CXX) $(ARCH) $(CXXFLAGS) -o $@ $^
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
## Linking Fortran example program calling the C++ function
fMaponiA3_test_3x3_3: $(fMaponiA3_test_3x3_3OBJ) $(OBJS)
$(FC) $(ARCH) $(FFLAGS) $(FLIBS) -o $@ $^
Added QMCChem test dataset with 3 updates. Renamed files, added the Fortran program QMCChem_dataset_test.f90 to start testing the QMCChem dataset. Wrote and tested Fortran code to import the test dataset.
2021-02-12 12:04:21 +01:00
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
fMaponiA3_test_4x4_2: $(fMaponiA3_test_4x4_2OBJ) $(OBJS)
$(FC) $(ARCH) $(FFLAGS) $(FLIBS) -o $@ $^
Add test infrastructure for datasets
2021-02-16 10:49:15 +01:00
The algorithm now works for the following 4x4 example with 2 updates: S = [1,0,1,-1; 0,1,1,0; -1,0,-1,0; 1,1,1,1] S_inv = [1,-1,1,1; 1,0,2,1; -1,1,-2,-1; -1,0,-1,0] u1 = [0,-2,0,0] u2 = [0,-1,0,0] upd_idx = [2,4] To go from Maponi's examples where the number of updates is always equal to the the dimension of the matrix, and the decomposition is always diagonal, to cases with a non-diagonal decomposition and a number of updates unequal to its size, the following changed needed to be made: * in the calculation of the {y0k} an extra inner for-loop needs to be added to make it a full matrix-vector multiplication due to the fact that A0 is not a diagonal matrix * in some places the use of the update-order vector p needs the be replaced with that of upd_idx to make sure the correct component of the ylk is selected and the proper rank-1 matrices are constructed * when a matrix is passed from Fortran to C++ with 2D adressing, it is passed in colum-major order. The passed matrix needs to be transposed before passing to C++. Doing this inside the algorithm will break compatibility with called from C/C++.
2021-02-21 18:28:08 +01:00
QMCChem_dataset_test: $(QMCChem_dataset_testOBJ) $(OBJS)
$(FC) $(ARCH) $(FFLAGS) $(FLIBS) -o $@ $^
Reference in New Issue Copy Permalink