Namespace SpiceSharp.Algebra.Solve

Classes

DenseLUSolver<T>

An base class for dense linear systems that can be solved using LU decomposition. Pivoting is controlled by the

DenseLUSolver<T>.ReorderedMatrix

A matrix that keeps everything synchronized for our solver.

DenseLUSolver<T>.ReorderedVector

A vector that keeps everything synchronized for our solver.

MarkowitzDiagonal<T>

Markowitz-count based strategy for finding a pivot. Searches the whole diagonal of the submatrix.

MarkowitzEntireMatrix<T>

Markowitz-count based strategy for finding a pivot. Search the complete submatrix.

MarkowitzQuickDiagonal<T>

Markowitz-based pivot search. Quickly search the diagonal for valid pivots.

MarkowitzSearchStrategy<T>

A template for a search strategy for finding pivots. It is used for implementing the strategy outlined by Markowitz.

MarkowitzSingleton<T>

Markowitz-count based strategy for finding a pivot. This strategy will search for singletons (rows or columns with only one element), these can be found rather cheaply.

Markowitz<T>

A search strategy based on methods outlined by Markowitz.

PivotingSolver<M, V, T>

A standard implementation for pivoting solvers.

A base class for sparse linear systems that can be solved using LU decomposition. Pivoting is controlled by the Parameters property. The implementation is optimized for sparse matrices through the ISparseMatrix<T> interface.

SparseLUSolver<T>.ReorderedMatrix

A sparse matrix that keeps both the matrix and right hand side vector synchronized for our solver.

SparseLUSolver<T>.ReorderedVector

A sparse vector that keeps both the matrix and right hand side vector synchronized for our solver.