ensemble_base

Base ensemble for optimisation: the control vector as state, objective evaluation over the members, and multilevel bookkeeping.

EnsembleOptimizationBase

Bases: BaseEnsemble

Base class for the popt ensemble

finest_level

Index of the finest (highest-fidelity) level, the last one in en_size.

num_levels

Number of fidelity levels: the length of en_size, or 1 without multilevel.

__init__(options, simulator, objective)

Parameters:

Name	Type	Description	Default
options	dict	Options for the ensemble class	required
simulator	callable	The forward simulator (e.g. flow). If None, no simulation is performed.	required
objective	callable	The objective function (e.g. npv)	required

function(x, *args, level=None, **kwargs)

Evaluate objective values for a single state vector or an ensemble.

Parameters:

Name	Type	Description	Default
x	ndarray	Control vector with shape (n_controls,) or ensemble matrix with shape (n_controls, n_perturbations).	required
level	int	Fidelity level to simulate on, for a multilevel ensemble (multilevel in the options). Default: a single control vector runs on the finest level and an ensemble is split over the levels by en_size. With level given, every column runs on that level. Negative values count from the finest level. A value on a coarser level is returned but does not update stateF, which stays the objective the optimizer works with.	None

Returns:

Type	Description
ndarray	Objective function values, or inf when the simulation crashed, so the optimizer rejects the point instead of the run ending. A crashed single-point evaluation leaves stateF at its last good value.

Raises:

Type	Description
ValueError	If x is not one- or two-dimensional.

get_bounds()

Returns:

Name	Type	Description
bounds	list	(min, max) pairs for each element in x. None is used to specify no bound.

get_cov()

Returns:

Name	Type	Description
cov	ndarray	Covariance matrix, shape (number of controls, number of controls)

get_state()

Returns:

Name	Type	Description
x	ndarray	Control vector as ndarray, shape (number of controls, number of perturbations)

levels_with_members(x)

Indices of the levels that x (a matrix, or a per-level list) puts members on.

save_stateX(state=None, path='./', filetype='npz')

Save the state vector.

Parameters:

Name	Type	Description	Default
path	str	Path to save the state vector. Default is current directory.	'./'
filetype	str	File type to save the state vector. Options are 'csv', 'npz' or 'npy'. Default is 'npz'.	'npz'