COBYLA¶

class COBYLA(maxiter=1000, disp=False, rhobeg=1.0, tol=None)[source]¶

Constrained Optimization By Linear Approximation optimizer.

COBYLA is a numerical optimization method for constrained problems where the derivative of the objective function is not known.

Parameters

maxiter (int) – Maximum number of function evaluations.
disp (bool) – Set to True to print convergence messages.
rhobeg (float) – Reasonable initial changes to the variables.
tol (Optional[float]) – Final accuracy in the optimization (not precisely guaranteed). This is a lower bound on the size of the trust region.

Attributes

`COBYLA.bounds_support_level`	Returns bounds support level
`COBYLA.gradient_support_level`	Returns gradient support level
`COBYLA.initial_point_support_level`	Returns initial point support level
`COBYLA.is_bounds_ignored`	Returns is bounds ignored
`COBYLA.is_bounds_required`	Returns is bounds required
`COBYLA.is_bounds_supported`	Returns is bounds supported
`COBYLA.is_gradient_ignored`	Returns is gradient ignored
`COBYLA.is_gradient_required`	Returns is gradient required
`COBYLA.is_gradient_supported`	Returns is gradient supported
`COBYLA.is_initial_point_ignored`	Returns is initial point ignored
`COBYLA.is_initial_point_required`	Returns is initial point required
`COBYLA.is_initial_point_supported`	Returns is initial point supported
`COBYLA.setting`	Return setting

Methods

`COBYLA.get_support_level`()	Return support level dictionary
`COBYLA.gradient_num_diff`(x_center, f, epsilon)	We compute the gradient with the numeric differentiation in the parallel way, around the point x_center.
`COBYLA.optimize`(num_vars, objective_function)	Perform optimization.
`COBYLA.print_options`()	Print algorithm-specific options.
`COBYLA.set_max_evals_grouped`(limit)	Set max evals grouped
`COBYLA.set_options`(**kwargs)	Sets or updates values in the options dictionary.
`COBYLA.wrap_function`(function, args)	Wrap the function to implicitly inject the args at the call of the function.