Interface SparseLinearAlgebraProvider<E>

All Superinterfaces:

AlgorithmProvider, AutoCloseable, LinearAlgebraProvider<E>

All Known Implementing Classes:

CPUSparseLinearAlgebraProvider, DistributedLinearAlgebraProvider, EpistemeLinearAlgebraBackend, GRPCLinearAlgebraBackend, NativeCUDASparseLinearAlgebraDoubleBackend, NativeCUDASparseLinearAlgebraFloatBackend, NativeMPFRSparseLinearAlgebraBackend, NativeOpenCLSparseLinearAlgebraDoubleBackend, NativeOpenCLSparseLinearAlgebraFloatBackend

public interface SparseLinearAlgebraProvider<E>
extends LinearAlgebraProvider<E>

Marker interface for Linear Algebra Providers specialized for Sparse Matrices.

This allows specialized discovery of sparse algorithms via AlgorithmManager.

Since:

1.2

Author:

Silvere Martin-Michiellot, Gemini AI (Google DeepMind)

Method Summary

Modifier and Type	Method	Description
default Vector<E>	bicgstab(Matrix<E> a, Vector<E> b, Vector<E> x0, E tolerance, int maxIterations)	Solves Ax = b using BiCGSTAB (BiConjugate Gradient Stabilized) method.
default E	conditionNumber(Matrix<E> a)	Returns the condition number of the matrix (L2 norm).
default Vector<E>	conjugateGradient(Matrix<E> a, Vector<E> b, Vector<E> x0, E tolerance, int maxIterations)	Solves Ax = b using Conjugate Gradient method.
default Matrix<E>	fromDense(Matrix<E> a)	Converts a dense matrix to its sparse representation.
default String	getAlgorithmType()	Returns the unique category of the algorithm.
default Vector<E>	gmres(Matrix<E> a, Vector<E> b, Vector<E> x0, E tolerance, int maxIterations, int restarts)	Solves Ax = b using GMRES (Generalized Minimal Residual) method.
default int	rank(Matrix<E> a)	Returns the rank of the matrix.
default Matrix<E>	toDense(Matrix<E> a)	Converts a sparse matrix to its dense representation.

Methods inherited from interface AlgorithmProvider

description, getMetadata, isAvailable, shutdown

Methods inherited from interface LinearAlgebraProvider

acos, acosh, add, add, angle, asin, asinh, atan, atanh, cbrt, cholesky, close, configure, conjugateTranspose, cos, cosh, cross, determinant, dot, eigen, exp, getEnvironmentInfo, getName, getPriority, inverse, isCompatible, log, log10, lu, multiply, multiply, multiply, norm, normalize, pow, projection, qr, scale, score, sin, sinh, solve, solve, solve, solve, solveTriangular, sqrt, subtract, subtract, svd, tan, tanh, trace, transpose

Method Details

getAlgorithmType

default String getAlgorithmType()

Description copied from interface: AlgorithmProvider

Returns the unique category of the algorithm.

Specified by:

getAlgorithmType in interface AlgorithmProvider

Specified by:

getAlgorithmType in interface LinearAlgebraProvider<E>

bicgstab

default Vector<E> bicgstab(Matrix<E> a,
 Vector<E> b,
 Vector<E> x0,
 E tolerance,
 int maxIterations)

Solves Ax = b using BiCGSTAB (BiConjugate Gradient Stabilized) method.

conjugateGradient

default Vector<E> conjugateGradient(Matrix<E> a,
 Vector<E> b,
 Vector<E> x0,
 E tolerance,
 int maxIterations)

Solves Ax = b using Conjugate Gradient method.

gmres

default Vector<E> gmres(Matrix<E> a,
 Vector<E> b,
 Vector<E> x0,
 E tolerance,
 int maxIterations,
 int restarts)

Solves Ax = b using GMRES (Generalized Minimal Residual) method.

toDense

default Matrix<E> toDense(Matrix<E> a)

Converts a sparse matrix to its dense representation.

fromDense

default Matrix<E> fromDense(Matrix<E> a)

Converts a dense matrix to its sparse representation.

rank

default int rank(Matrix<E> a)

Description copied from interface: LinearAlgebraProvider

Returns the rank of the matrix.

Specified by:

rank in interface LinearAlgebraProvider<E>

conditionNumber

default E conditionNumber(Matrix<E> a)

Description copied from interface: LinearAlgebraProvider

Returns the condition number of the matrix (L2 norm).

Specified by:

conditionNumber in interface LinearAlgebraProvider<E>