 |
SPI
SLEPc PETSc Interface is used for rapid development and intuitive matrix operations similar to MatLab or NumPy style of operations
|
|
SPI
SLEPc PETSc Interface is used for rapid development and intuitive matrix operations similar to MatLab or NumPy style of operations
|
#include "SPIMat.hpp"#include "SPIprint.hpp"#include <petscviewerhdf5.h>#include <petscmath.h>#include <slepcsvd.h>Go to the source code of this file.
Namespaces | |
| SPI | |
Macros | |
| #define | USE_GPU 0 |
Functions | |
| SPIMat | SPI::operator* (const PetscScalar a, const SPIMat A) |
| a*A operation to be equivalent to A*a More... | |
| SPIMat | SPI::operator* (const SPIMat A, const PetscScalar a) |
| A*a operation to be equivalent to A*a. More... | |
| SPIVec | SPI::operator/ (const SPIVec &b, const SPIMat &A) |
| Solve linear system, Ax=b using b/A notation. More... | |
| SPIMat | SPI::operator^ (const PetscScalar a, const SPIMat &A) |
| Y=a^A operation. More... | |
| SPIVec | SPI::solve (const SPIMat &A, const SPIVec &b) |
| Solve linear system, Ax=b using solve(A,b) notation. More... | |
| SPIMat | SPI::inv (const SPIMat &A) |
| Solve linear system A*Ainv=B using MatMatSolve. More... | |
| SPIMat | SPI::eye (const PetscInt n) |
| create, form, and return identity matrix of size n More... | |
| SPIMat | SPI::zeros (const PetscInt m, const PetscInt n) |
| create, form, and return zeros matrix of size mxn More... | |
| SPIMat | SPI::diag (const SPIVec &d, const PetscInt k) |
| set diagonal of matrix More... | |
| SPIMat | SPI::kron (const SPIMat &A, const SPIMat &B) |
| set kronecker inner product of two matrices More... | |
| std::tuple< std::vector< PetscReal >, std::vector< SPIVec >, std::vector< SPIVec > > | SPI::svd (const SPIMat &A) |
| solve SVD problem of A=U*E*V^H for skinny A matrix More... | |
| SPIVec | SPI::lstsq (const SPIMat &A, SPIVec &y) |
| solve least squares problem of A*x=y for skinny A matrix using SVD More... | |
| SPIVec | SPI::lstsq (const std::vector< SPIVec > &A, SPIVec &y) |
| solve least squares problem of A*x=y for skinny A matrix (or vectors of columns vectors) using SVD More... | |
| std::tuple< PetscScalar, SPIVec, SPIVec > | SPI::eig (const SPIMat &A, const SPIMat &B, const PetscScalar target, const PetscReal tol, const PetscInt max_iter) |
| solve general eigenvalue problem of Ax = kBx and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) More... | |
| std::tuple< PetscScalar, SPIVec > | SPI::eig_right (const SPIMat &A, const SPIMat &B, const PetscScalar target, const PetscReal tol, const PetscInt max_iter) |
| solve general eigenvalue problem of Ax = kBx and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) More... | |
| std::tuple< PetscScalar, SPIVec, SPIVec > | SPI::eig_init (const SPIMat &A, const SPIMat &B, const PetscScalar target, const SPIVec &ql, const SPIVec &qr, PetscReal tol, const PetscInt max_iter) |
| solve general eigenvalue problem of Ax = kBx and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) More... | |
| std::tuple< PetscScalar, SPIVec > | SPI::eig_init_right (const SPIMat &A, const SPIMat &B, const PetscScalar target, const SPIVec &qr, PetscReal tol, const PetscInt max_iter) |
| solve general eigenvalue problem of Ax = kBx and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) More... | |
| std::tuple< std::vector< PetscScalar >, std::vector< SPIVec > > | SPI::eig_init_rights (const SPIMat &A, const SPIMat &B, const std::vector< PetscScalar > targets, const std::vector< SPIVec > &qrs, PetscReal tol, const PetscInt max_iter) |
| solve general eigenvalue problem of Ax = kBx and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) More... | |
| std::tuple< PetscScalar, SPIVec > | SPI::polyeig (const std::vector< SPIMat > &As, const PetscScalar target, const PetscReal tol, const PetscInt max_iter) |
| solve general polynomial eigenvalue problem of (A0 + A1*alpha + A2*alpha^2 + ...)*x = 0 and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) More... | |
| std::tuple< PetscScalar, SPIVec > | SPI::polyeig_init (const std::vector< SPIMat > &As, const PetscScalar target, const SPIVec &qr, const PetscReal tol, const PetscInt max_iter) |
| solve general polynomial eigenvalue problem of (A0 + A1*alpha + A2*alpha^2 + ...)*x = 0 and return a tuple of tie(PetscScalar alpha, SPIVec eig_vector) using initial subspace vector More... | |
| SPIMat | SPI::block ( const Block2D< SPIMat > Blocks) |
| set block matrices using an input array of size rows*cols. Fills rows first More... | |
| std::tuple< SPIMat, SPIMat > | SPI::meshgrid (SPIVec &x, SPIVec &y) |
| PetscInt | SPI::save (const SPIMat &A, const std::string filename) |
| save matrix to filename in binary format (see Petsc documentation for format ) Format is (from Petsc documentation): int MAT_FILE_CLASSID int number of rows int number of columns int total number of nonzeros int *number nonzeros in each row int *column indices of all nonzeros (starting index is zero) PetscScalar *values of all nonzeros More... | |
| PetscInt | SPI::save (const std::vector< SPIMat > &As, const std::string filename) |
| save matrices to filename in binary format (see Petsc documentation for format More... | |
| PetscInt | SPI::load (SPIMat &A, const std::string filename) |
| load matrix from filename from binary format (works with save(SPIMat,std::string) function More... | |
| PetscInt | SPI::load (std::vector< SPIMat > &As, const std::string filename) |
| load matrix from filename from binary format (works with save(SPIMat,std::string) function More... | |
| PetscInt | SPI::draw (const SPIMat &A) |
| draw nonzero structure of matrix More... | |
| template<class T > | |
| SPIMat | SPI::_Function_on_each_element (T(*f)(T const &), const SPIMat &A) |
| take the function of each element in a matrix, e.g. (*f)(A(i,j)) for each i,j More... | |
| SPIMat | SPI::sin (const SPIMat &A) |
| take the sin of each element in a matrix More... | |
| SPIMat | SPI::cos (const SPIMat &A) |
| take the cos of each element in a matrix More... | |
| SPIMat | SPI::acos (const SPIMat &A) |
| take the arccos of each element in a matrix More... | |
| SPIMat | SPI::tan (const SPIMat &A) |
| take the tan of each element in a matrix More... | |
| SPIMat | SPI::operator% (const SPIMat &A, const SPIMat &B) |
| take the elementwise multiplication of each element in a matrix More... | |
| SPIMat | SPI::abs (const SPIMat &A) |
| take the abs of each element in a matrix More... | |
| SPIMat | SPI::orthogonalize (const std::vector< SPIVec > &x) |
| #define USE_GPU 0 |
Definition at line 7 of file SPIMat.cpp.
1.8.17