linalg.h File Reference

Static linear-algebra helpers: SVD-based conditioning, block-diagonal assembly, sorted index pairs. More...

#include "math_global.h"
#include <utility>
#include <vector>
#include <Eigen/Core>
#include <Eigen/SparseCore>
#include <Eigen/SVD>
#include <QString>

Include dependency graph for linalg.h:

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

Go to the source code of this file.

Classes
class	UTILSLIB::Linalg
	Static Eigen-based linear-algebra helpers used across MATHLIB and the inverse solvers. More...

Namespaces
namespace	UTILSLIB
	Shared utilities (I/O helpers, spectral analysis, layout management, warp algorithms).

Detailed Description

Static linear-algebra helpers: SVD-based conditioning, block-diagonal assembly, sorted index pairs.

SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2026 MNE-CPP Authors

Author

Christoph Dinh chris.nosp@m.toph.nosp@m..dinh.nosp@m.@mne.nosp@m.-cpp..nosp@m.org

Since

2.0.0

Date

March 2026

UTILSLIB::Linalg collects the dense / sparse Eigen primitives that the rest of mne-cpp keeps reaching for but that are not part of Eigen itself: condition number and condition slope from a Jacobi SVD, assembly of block-diagonal sparse matrices from a stack of dense blocks (the typical "three orientations per dipole" layout used by the forward and inverse solvers), the combine_xyz reduction that turns an [x1 y1 z1 ... x_n y_n z_n] row into per-source squared magnitudes, and a stable index-value sort helper used to permute channels or sources by an arbitrary scalar criterion.

All entry points are static and side-effect free so the class never needs to be instantiated. Dense paths run through Eigen::JacobiSVD (O(min(m,n)^2 * max(m,n))) while the sparse paths are linear in the number of stored non-zeros, which keeps the helpers usable even on the source-space-sized matrices manipulated by INVERSELIB.

Definition in file linalg.h.