GradientPerturbation#

class GradientPerturbation(func, grad, func_args, grad_args, n_red=1)[source]#

Bases: ObjectivePerturbation, ABC

A class for perturbations performed by decreasing an objective function using the gradient.

Parameters:

  • func (Callable) – The objective function to be perturbed.

  • grad (Callable) – The gradient function of the objective function.

  • func_args (List) – The arguments to be passed to the objective function.

  • grad_args (List) – The arguments to be passed to the gradient function.

  • n_red (int) – The reduction factor, by default 1.

Attributes:
funcFuncWrapper

A wrapped version of the objective function with its arguments.

gradFuncWrapper

A wrapped gradient function with its arguments.

n_redint

The number of reduction steps to perform.

_kint

Keeps track of the number of performed perturbations.

Methods

perturbation_step(x)

Perform n_red perturbation steps on the input array.

post_step(x, *args, **kwargs)

If required perform any form of step after each perturbation in each iteration.

pre_step(x, *args, **kwargs)

If required perform any form of step previous to each perturbation in each iteration.

reset()

Reset the perturbation to its initial state.
perturbation_step(x)#

Perform n_red perturbation steps on the input array.

Parameters:

x (ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]]) – The input array to be perturbed.

Return type:

ndarray

Returns:

The perturbed array after applying the reduction steps.

post_step(x, *args, **kwargs)#

If required perform any form of step after each perturbation in each iteration.

This method is intended to be overridden by subclasses to implement specific post-step logic. By default, it does nothing.

Parameters:

x (ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]]) – Current iterate.

Return type:

None

pre_step(x, *args, **kwargs)#

If required perform any form of step previous to each perturbation in each iteration.

This method is intended to be overridden by subclasses to implement specific pre-step logic. By default, it does nothing.

Parameters:

x (ndarray[tuple[Any, ...], dtype[TypeVar(_ScalarT, bound= generic)]]) – Current iterate.

Return type:

None

reset()#

Reset the perturbation to its initial state.

Return type:

None