fipy.solvers.petsc.linearCGSolver

Classes

LinearCGSolver([tolerance, ...])

Interface to the conjugate gradient (CG) solver in PETSc.

LinearPCGSolver

alias of LinearCGSolver
class fipy.solvers.petsc.linearCGSolver.LinearCGSolver(tolerance='default', absolute_tolerance=None, divergence_tolerance=None, criterion='default', iterations='default', precon='default')

Bases: PETScKrylovSolver

Interface to the conjugate gradient (CG) solver in PETSc.

Parameters:

  • tolerance (float) – Required relative error tolerance.

  • absolute_tolerance (float) – Required absolute error tolerance.

  • divergence_tolerance (float) – Required divergence error tolerance.

  • criterion ({'default', 'unscaled', 'RHS', 'matrix', 'initial', 'preconditioned', 'natural', 'legacy'}) – Interpretation of tolerance. See Convergence for more information.

  • iterations (int) – Maximum number of iterative steps to perform.

  • precon (PETScPreconditioner or str, optional) –

    Preconditioner to apply to the matrix. A value of None means to perform an unpreconditioned solve. While, for consistency with other solver suites, FiPy does implement some preconditioner objects for PETSc, you can also simply pass one of the PCType strings. (default: DefaultPreconditioner).

DEFAULT_ITERATIONS = 1000

Default maximum number of iterative steps to perform

DEFAULT_PRECONDITIONER

alias of DefaultPreconditioner

DEFAULT_TOLERANCE = 1e-05

Default tolerance for linear solves unless criterion=”legacy”

LEGACY_TOLERANCE = 1e-10

Default tolerance for linear solves if criterion=”legacy”

__repr__()

Return repr(self).

property default_tolerance

Default tolerance for linear solve

fipy.solvers.petsc.linearCGSolver.LinearPCGSolver

alias of LinearCGSolver

Last updated on Sep 02, 2025. Created using Sphinx 7.1.2.