moarchiving.get_archive

module documentation

Convenience create functions for any of the nondominated archives.

The appropriate number of objectives, is derived from the input arguments.

Function get_cmo_archive Return a constrained nondominated archive instance with the proper number of objectives.
Function get_mo_archive Return a nondominated archive instance with the proper number of objectives.
def get_cmo_archive(list_of_f_vals=None, list_of_g_vals=None, reference_point=None, infos=None, n_obj=None, tau=1):

Return a constrained nondominated archive instance with the proper number of objectives.

list_of_f_vals is a list of objective vectors with n_obj objectives, list_of_g_vals is a list of constraint violation vectors (or values). If list_of_f_vals is not provided, n_obj can be provided to define the number of objectives.

reference_point is used for the hypervolume computation and pruning of the archive. A list of additional information for each objective vector can be provided in infos.

def get_mo_archive(list_of_f_vals=None, reference_point=None, infos=None, n_obj=None):

Return a nondominated archive instance with the proper number of objectives.

list_of_f_vals is a list of objective vectors with n_obj objectives. If list_of_f_vals is not provided, n_obj can be provided to define the number of objectives which is by default 2.

reference_point is used for hypervolume computation and pruning of the archive. A list of additional information for each objective vector, for example the solution from which the objective values were computed, can be provided in infos.