Eigen::AMDOrdering< StorageIndex > Class Template Reference

#include <Ordering.h>

List of all members.

Public Types
typedef PermutationMatrix < Dynamic, Dynamic, StorageIndex >	PermutationType
Public Member Functions
template<typename MatrixType >
void	operator() (const MatrixType &mat, PermutationType &perm)
template<typename SrcType , unsigned int SrcUpLo>
void	operator() (const SparseSelfAdjointView< SrcType, SrcUpLo > &mat, PermutationType &perm)

---

Detailed Description

template<typename StorageIndex>
class Eigen::AMDOrdering< StorageIndex >

Functor computing the approximate minimum degree ordering If the matrix is not structurally symmetric, an ordering of A^T+A is computed

Template Parameters:

StorageIndex

The type of indices of the matrix

See also:

COLAMDOrdering

Definition at line 52 of file Ordering.h.

---

Member Typedef Documentation

template<typename StorageIndex>

typedef PermutationMatrix<Dynamic, Dynamic, StorageIndex> Eigen::AMDOrdering< StorageIndex >::PermutationType

Definition at line 55 of file Ordering.h.

---

Member Function Documentation

template<typename StorageIndex>

template<typename MatrixType >

void Eigen::AMDOrdering< StorageIndex >::operator()	(	const MatrixType &	mat,
		PermutationType &	perm
	)		[inline]

Compute the permutation vector from a sparse matrix This routine is much faster if the input matrix is column-major

Definition at line 61 of file Ordering.h.

    {
      // Compute the symmetric pattern
      SparseMatrix<typename MatrixType::Scalar, ColMajor, StorageIndex> symm;
      internal::ordering_helper_at_plus_a(mat,symm); 
    
      // Call the AMD routine 
      //m_mat.prune(keep_diag());
      internal::minimum_degree_ordering(symm, perm);
    }

template<typename StorageIndex>

template<typename SrcType , unsigned int SrcUpLo>

void Eigen::AMDOrdering< StorageIndex >::operator()	(	const SparseSelfAdjointView< SrcType, SrcUpLo > &	mat,
		PermutationType &	perm
	)		[inline]

Compute the permutation with a selfadjoint matrix

Definition at line 74 of file Ordering.h.

    { 
      SparseMatrix<typename SrcType::Scalar, ColMajor, StorageIndex> C; C = mat;
      
      // Call the AMD routine 
      // m_mat.prune(keep_diag()); //Remove the diagonal elements 
      internal::minimum_degree_ordering(C, perm);
    }

---

The documentation for this class was generated from the following file:

• Ordering.h