from detpy.DETAlgs.base import BaseAlg
from detpy.DETAlgs.data.alg_data import COMDEData
from detpy.DETAlgs.methods.methods_comde import calculate_cr, comde_mutation
from detpy.DETAlgs.methods.methods_de import binomial_crossing, selection
from detpy.models.enums.boundary_constrain import fix_boundary_constraints
[docs]
class COMDE(BaseAlg):
"""
COMDE
Links:
https://www.sciencedirect.com/science/article/pii/S0020025512000278
References:
Mohamed, A. W., & Sabry, H. Z. (2012). Constrained optimization based on modified differential evolution algorithm.
In Information Sciences (Vol. 194, pp. 171–208). Elsevier BV. https://doi.org/10.1016/j.ins.2012.01.008
"""
def __init__(self, params: COMDEData, db_conn=None, db_auto_write=False):
super().__init__(COMDE.__name__, params, db_conn, db_auto_write)
self.mutation_factor = params.mutation_factor # F
self.crossover_rate = params.crossover_rate # Cr
[docs]
def next_epoch(self):
# Calculate not constant cr depend on generation number
cr = calculate_cr(self._epoch_number, self.num_of_epochs)
# New population after mutation
v_pop = comde_mutation(self._pop)
# 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=cr)
# 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)
# Override data
self._pop = new_pop
self._epoch_number += 1