tlm_adjoint.optimization

Module Contents

tlm_adjoint.optimization.minimize_scipy(forward, M0, *, manager=None, **kwargs)

Minimize a functional using scipy.optimize.minimize().

Parameters:

forwardcallable

Accepts one or more variable arguments, and returns a scalar variable defining the functional.

M0variable or Sequence[variable, …]

The initial guess.

managerEquationManager

Used to create an internal manager via EquationManager.new(). manager() is used if not supplied.

kwargs

Passed to scipy.optimize.minimize().

Returns:

Mvariable or Sequence[variable, …]

The result of the minimization

minimize_return_value

The return value from scipy.optimize.minimize().

Warning

Data is gathered onto the root process (process zero) so that the serial scipy.optimize.minimize() can be used.

class tlm_adjoint.optimization.TAOSolver(forward, space, *, solver_parameters=None, H_0_action=None, M_inv_action=None, manager=None)

Functional minimization using TAO.

Parameters:

forwardcallable

Accepts one or more variable arguments, and returns a scalar variable defining the functional.

spacespace or Sequence[space, …]

The control space.

solver_parametersMapping

TAO solver parameters.

H_0_actioncallable

Defines the initial Hessian inverse approximation. Accepts one or more variables as arguments, defining a direction, and returns a variable or a Sequence of variables defining the (conjugate of) the action of an initial Hessian inverse approximation on this direction. Arguments should not be modified.

M_inv_actioncallable

Defines a dual space norm. Accepts one or more variables as arguments, defining a direction, and returns a variable or a Sequence of variables defining the (conjugate of) the action of a Hermitian and positive definite matrix on this direction. Arguments should not be modified. H_0_action is used if not supplied.

managerEquationManager

Used to create an internal manager via EquationManager.new(). manager() is used if not supplied.

property tao

The petsc4py.PETSc.TAO used to solve the optimization problem.

solve(M)

Solve the optimization problem.

Parameters:

Mvariable or Sequence[variable, …]

Defines the initial guess, and stores the result of the minimization.

tlm_adjoint.optimization.minimize_tao(forward, M0, *args, **kwargs)

Minimize a functional using TAO.

Parameters:

forwardcallable

Accepts one or more variable arguments, and returns a scalar variable defining the functional.

M0variable or Sequence[variable, …]

The initial guess.

args, kwargs

Passed to the TAOSolver constructor.

Returns:

variable or Sequence[variable, …]

The result of the minimization.