StructuredPenaltyOperator.from_penalties()#

classmethod StructuredPenaltyOperator.from_penalties(scales, penalties, eps=1e-06, validate_args=False)[source]#

Shortcut for initializing the operator from marignal scales and penalties (expensive).

Computes the eigenvalue decompositions of the marginal penalties and the mask at runtime, so this is not appropriate for repeated execution if computational efficiency is a concern.