Eigen  5.0.1-dev+master
 
Loading...
Searching...
No Matches
SparseCholesky module

Detailed Description

This module currently provides two variants of the direct sparse Cholesky decomposition for selfadjoint (hermitian) positive definite matrices. They are not intended for general selfadjoint or positive semidefinite matrices. Those decompositions are accessible via the following classes:

Such problems can also be solved using the ConjugateGradient solver from the IterativeLinearSolvers module.

For symmetric or Hermitian indefinite matrices, such as KKT systems, this module also provides MultifrontalLDLT, a factorization with 1x1 and 2x2 pivots that also reveals the matrix's inertia.

#include <Eigen/SparseCholesky>

Classes

class  Eigen::MultifrontalLDLT< MatrixType_, UpLo_, Ordering_ >
 Sparse multifrontal LDL^* factorization of a symmetric or Hermitian indefinite matrix. More...
 
class  Eigen::SimplicialCholesky< MatrixType_, UpLo_, Ordering_ >
 
class  Eigen::SimplicialCholeskyBase< Derived >
 A base class for direct sparse Cholesky factorizations. More...
 
class  Eigen::SimplicialLDLT< MatrixType_, UpLo_, Ordering_ >
 A direct sparse LDLT Cholesky factorizations without square root. More...
 
class  Eigen::SimplicialLLT< MatrixType_, UpLo_, Ordering_ >
 A direct sparse LLT Cholesky factorizations. More...
 
class  Eigen::SimplicialNonHermitianLDLT< MatrixType_, UpLo_, Ordering_ >
 A direct sparse LDLT Cholesky factorizations without square root, for symmetric non-hermitian matrices. More...
 
class  Eigen::SimplicialNonHermitianLLT< MatrixType_, UpLo_, Ordering_ >
 A direct sparse LLT Cholesky factorizations, for symmetric non-hermitian matrices. More...