sparse.hpp File Reference

#include <Eigen/Sparse>
#include <Eigen/src/Core/Map.h>
#include <Eigen/src/Core/Matrix.h>
#include <global_opts.h>
#include <ladel_constants.h>
#include <ladel_types.h>
#include <memory>
#include <qpalm_cxx-export.hpp>

Include dependency graph for sparse.hpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Data Structures
struct	qpalm::alloc::ladel_sparse_matrix_deleter

Namespaces
namespace	qpalm
namespace	qpalm::alloc
	RAII-based wrappers for the allocation and deallocation functions of the C API.

Typedefs
using	qpalm::index_t = Eigen::Index
	Index types for vectors and matrices.
using	qpalm::sp_index_t = ladel_int
	Index type for sparse matrices representation.
using	qpalm::sparse_mat_t = Eigen::SparseMatrix< c_float, Eigen::ColMajor, sp_index_t >
	Owning sparse matrix type.
using	qpalm::sparse_mat_view_t = Eigen::Map< const sparse_mat_t >
	Read-only view on a sparse matrix.
using	qpalm::triplet_t = Eigen::Triplet< c_float, sp_index_t >
	Type for (row, column, value) triplets for initializing sparse matrices.
using	qpalm::vec_t = Eigen::Matrix< c_float, Eigen::Dynamic, 1 >
	Owning dense vector type.
using	qpalm::borrowed_vec_t = Eigen::Map< vec_t >
	Borrowed dense vector type (vector view).
using	qpalm::const_borrowed_vec_t = Eigen::Map< const vec_t >
	Read-only borrowed dense vector type (vector view).
using	qpalm::ref_vec_t = Eigen::Ref< vec_t >
	Reference to a dense vector (vector view).
using	qpalm::const_ref_vec_t = Eigen::Ref< const vec_t >
	Read-only reference to a dense vector (vector view).
using	qpalm::ladel_sparse_matrix_ptr = std::unique_ptr< ladel_sparse_matrix, alloc::ladel_sparse_matrix_deleter >
	Smart pointer that automatically cleans up an owning ladel_sparse_matrix object.

Functions
QPALM_CXX_EXPORT ladel_sparse_matrix	qpalm::eigen_to_ladel (sparse_mat_t &mat, ladel_int symmetry=UNSYMMETRIC)
	Convert an Eigen sparse matrix to a LADEL sparse matrix, without creating a copy.
QPALM_CXX_EXPORT ladel_sparse_matrix_ptr	qpalm::eigen_to_ladel_copy (const sparse_mat_t &mat)
	Similar to eigen_to_ladel, but creates a copy of all data, in such a way that the returned matrix is completely decoupled from mat, and such that it can be reallocated and deallocated by the ladel_sparse_free and similar functions.