def __init__(self,
var_type,
ref_points=None,
epsilon=0.001,
weights=None,
**kwargs):
self.ref_dirs = np.array(ref_points)
self.epsilon = epsilon
self.weights = weights
set_default_if_none(var_type, kwargs)
set_if_none(kwargs, 'survival', None)
super().__init__(**kwargs)
def __init__(self, var_type="real", n_neighbors=15, **kwargs):
#set_if_none(kwargs, "crossover", DifferentialEvolutionCrossover())
set_default_if_none(var_type, kwargs)
super().__init__(**kwargs)
self.n_neighbors = n_neighbors
# initialized when problem is known
self.weights = None
self.neighbours = None
self.ideal_point = None
def __init__(self, variant="DE/rand/1/exp", CR=0.1, F=0.75, **kwargs):
set_default_if_none("real", kwargs)
super().__init__(**kwargs)
self.selection = RandomSelection()
self.crossover = DifferentialEvolutionCrossover(weight=F)
_, self.var_selection, self.var_n, self.var_mutation, = variant.split(
"/")
self.mutation = DifferentialEvolutionMutation(self.var_mutation, CR)
self.func_display_attrs = disp_single_objective
def __init__(self,
var_type,
ref_points=None,
mu=0.1,
ref_pop_size=None,
method='uniform',
p=None,
**kwargs):
"""
Parameters
----------
var_type : string
Variable type which must be real in this case
ref_points : numpy.array
Reference points to be focused on during the evolutionary computation.
mu : double
The shrink factor
ref_pop_size : int
If the structured reference lines should be based off of a different population size
than the actual population size. Default value is pop size.
p : double
If the structured reference directions should be based off of p gaps specify a p value, otherwise
reference directions will be based on the population size.
ref_sampling_method : string
Reference direction generation method. Currently only 'uniform' or 'random'.
"""
self.ref_points = ref_points
self.ref_dirs = None
self.mu = mu
self.method = method
set_default_if_none(var_type, kwargs)
set_if_none(kwargs, 'survival', None)
self.ref_pop_size = ref_pop_size
self.p = p
super().__init__(**kwargs)
def __init__(self, var_type, ref_points=None, **kwargs):
self.ref_points = ref_points
self.ref_dirs = None
set_default_if_none(var_type, kwargs)
set_if_none(kwargs, 'survival', None)
super().__init__(**kwargs)
def __init__(self, var_type, **kwargs):
set_default_if_none(var_type, kwargs)
set_if_none(kwargs, 'selection',
TournamentSelection(f_comp=comp_by_rank_and_crowding))
set_if_none(kwargs, 'survival', RankAndCrowdingSurvival())
super().__init__(**kwargs)
def __init__(self,
**kwargs):
set_default_if_none("real", kwargs)
super().__init__(**kwargs)
self.crossover = DifferentialEvolutionCrossover(prob=0.5, weight=0.75, variant="DE/rand/1", method="binomial")
def __init__(self, var_type, **kwargs):
set_if_none(kwargs, 'survival', FitnessSurvival())
set_default_if_none(var_type, kwargs)
super().__init__(**kwargs)
def __init__(self, var_type, **kwargs):
set_default_if_none(var_type, **kwargs)
super().__init__(**kwargs)
