from detpy.DETAlgs.base import BaseAlg
from detpy.DETAlgs.data.alg_data import MGDEData
from detpy.DETAlgs.methods.methods_de import binomial_crossing, selection
from detpy.DETAlgs.methods.methods_mgde import mgde_mutation, mgde_adapt_threshold
from detpy.models.enums.boundary_constrain import fix_boundary_constraints
[docs]
class MGDE(BaseAlg):
"""
MGDE
Links:
https://link.springer.com/article/10.1007/s10479-022-04641-3
References:
Zouache, D., Ben Abdelaziz, F. MGDE: a many-objective guided differential evolution with strengthened dominance
relation and bi-goal evolution. Ann Oper Res (2022). https://doi.org/10.1007/s10479-022-04641-3
"""
def __init__(self, params: MGDEData, db_conn=None, db_auto_write=False):
super().__init__(MGDE.__name__, params, db_conn, db_auto_write)
self.mutation_factor_f = params.mutation_factor_f
self.mutation_factor_k = params.mutation_factor_k
self.crossover_rate = params.crossover_rate
self.threshold = params.threshold
self.mu = params.mu
self.generation = 1
[docs]
def next_epoch(self):
# New population after mutation
v_pop = mgde_mutation(self._pop, self.generation, self.num_of_epochs, self.mutation_factor_f,
self.mutation_factor_k)
# Apply boundary constrains on population in place
fix_boundary_constraints(v_pop, self.boundary_constraints_fun)
# New population after crossing
u_pop = binomial_crossing(self._pop, v_pop, cr=self.crossover_rate)
# Update values before selection
u_pop.update_fitness_values(self._function.eval, self.parallel_processing)
# Select new population
new_pop = selection(self._pop, u_pop)
mgde_adapt_threshold(new_pop, self.threshold, self.mu, self._function.eval)
# Override data
self._pop = new_pop
self._epoch_number += 1