cma.sigma_adaptation.CMAAdaptSigmaCSA

class documentation

class CMAAdaptSigmaCSA(CMAAdaptSigmaBase):

CSA cumulative step-size adaptation AKA path length control.

As of 2017, CSA is considered as the default step-size control method within CMA-ES.

Method	`__init__`	postpone initialization to a method call where dimension and mueff should be known.
Method	`initialize`	set parameters and state variable based on dimension, mueff and possibly further options.
Method	`update`	call `self._update_ps(es)` and update `es.sigma`.
Method	`update2`	call `self._update_ps(es)` and update self.delta.
Instance Variable	`cs`	Undocumented
Instance Variable	`damps`	Undocumented
Instance Variable	`delta`	cumulated effect of adaptation
Instance Variable	`disregard_length_setting`	Undocumented
Instance Variable	`is_initialized`	Undocumented
Instance Variable	`max_delta_log_sigma`	Undocumented
Instance Variable	`ps`	Undocumented
Method	`_update_ps`	update path with isotropic delta mean, possibly clipped.
Instance Variable	`_ps_updated_iteration`	Undocumented

Inherited from CMAAdaptSigmaBase:

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.
Instance Variable	`is_initialized_base`	Undocumented

def __init__(self, **kwargs): ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase.__init__

postpone initialization to a method call where dimension and mueff should be known.

def initialize(self, es): ¶

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

def update(self, es, **kwargs): ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase.update

call self._update_ps(es) and update es.sigma.

Legacy method replaced by update2.

def update2(self, es, **kwargs): ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase.update2

call self._update_ps(es) and update self.delta.

Return change factor of self.delta.

From input es, either attribute N or const.chiN is used and path_for_sigma_update.

cs = ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase.cs

Undocumented

damps = ¶

Undocumented

delta: int = ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase.delta

cumulated effect of adaptation

disregard_length_setting = ¶

Undocumented

is_initialized: bool = ¶

Undocumented

max_delta_log_sigma: int = ¶

Undocumented

ps = ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase.ps

Undocumented

def _update_ps(self, es): ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase._update_ps

update path with isotropic delta mean, possibly clipped.

From input argument es, the attributes isotropic_mean_shift, opts['CSA_clip_length_value'], and N are used. opts['CSA_clip_length_value'] can be a single value, the upper bound factor, such that:

max_len = sqrt(N) + opts['CSA_clip_length_value'] * N / (N+2)

or a list with a lower and an upper factor.

_ps_updated_iteration = ¶

overrides cma.sigma_adaptation.CMAAdaptSigmaBase._ps_updated_iteration

Undocumented