cma.sigma_adaptation.CMAAdaptSigmaBase

class documentation

class CMAAdaptSigmaBase(object):

Known subclasses: cma.sigma_adaptation.CMAAdaptSigmaCSA, cma.sigma_adaptation.CMAAdaptSigmaDistanceProportional, cma.sigma_adaptation.CMAAdaptSigmaMedianImprovement, cma.sigma_adaptation.CMAAdaptSigmaNone, cma.sigma_adaptation.CMAAdaptSigmaTPA

View In Hierarchy

step-size adaptation base class, implement hsig (for stalling distribution update) functionality via an isotropic evolution path.

Details: hsig or _update_ps must be called before the sampling distribution is changed. _update_ps depends heavily on cma.CMAEvolutionStrategy.

Method __init__ Undocumented
Method check_consistency make consistency checks with a CMAEvolutionStrategy instance as input
Method hsig return "OK-signal" for rank-one update, True (OK) or False (stall rank-one update), based on the length of an evolution path
Method initialize_base set parameters and state variable based on dimension, mueff and possibly further options.
Method update update es.sigma
Method update2 return sigma change factor and update self.delta.
Instance Variable cs Undocumented
Instance Variable delta cumulated effect of adaptation
Instance Variable is_initialized_base Undocumented
Instance Variable ps Undocumented
Method _update_ps update the isotropic evolution path.
Instance Variable _ps_updated_iteration Undocumented
def __init__(self, *args, **kwargs):
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA, cma.sigma_adaptation.CMAAdaptSigmaDistanceProportional, cma.sigma_adaptation.CMAAdaptSigmaMedianImprovement, cma.sigma_adaptation.CMAAdaptSigmaTPA

Undocumented

def check_consistency(self, es):
overridden in cma.sigma_adaptation.CMAAdaptSigmaTPA

make consistency checks with a CMAEvolutionStrategy instance as input

def hsig(self, es):

return "OK-signal" for rank-one update, True (OK) or False (stall rank-one update), based on the length of an evolution path

def initialize_base(self, es):

set parameters and state variable based on dimension, mueff and possibly further options.

def update(self, es, **kwargs):
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA, cma.sigma_adaptation.CMAAdaptSigmaDistanceProportional, cma.sigma_adaptation.CMAAdaptSigmaMedianImprovement, cma.sigma_adaptation.CMAAdaptSigmaNone, cma.sigma_adaptation.CMAAdaptSigmaTPA

update es.sigma

Parameters
esCMAEvolutionStrategy class instance
**kwargswhatever else is needed to update es.sigma, which should be none.
def update2(self, es, **kwargs):
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA

return sigma change factor and update self.delta.

self.delta == sigma/sigma0 accumulates all past changes starting from 1.0.

Unlike update, update2 is not supposed to change attributes in es, specifically it should not change es.sigma.

cs =
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA

Undocumented

delta: int =
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA

cumulated effect of adaptation

is_initialized_base: bool =

Undocumented

ps =
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA

Undocumented

def _update_ps(self, es):
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA

update the isotropic evolution path.

Using es attributes mean, mean_old, sigma, sigma_vec, sp.weights.mueff, cp.cmean and sm.transform_inverse.

Parameters
es:CMAEvolutionStrategyUndocumented
_ps_updated_iteration =
overridden in cma.sigma_adaptation.CMAAdaptSigmaCSA

Undocumented