import random
from individual import *
from problems.abstract_problem import AbstractProblem
from typing import List
from recombination.recombination_operator import RecombinationOperator
[docs]
class FlatCrossover(RecombinationOperator):
"""
FlatCrossover performs a flat crossover on a pair of parent individuals
to produce offspring individuals.
Parameters
----------
parents : list
A list containing two parent individuals.
problem : AbstractProblem
The problem instance that provides the fitness function.
"""
[docs]
def __init__(self, parents: list, problem: AbstractProblem):
"""
Initialize the FlatCrossover object.
Parameters
----------
parents : list
A list containing two parent individuals.
problem : AbstractProblem
The problem instance that provides the fitness function.
"""
self.parents = parents
self.problem = problem
[docs]
def combine(self, p1: Individual, p2: Individual, locationsource: Individual) -> Individual:
"""
Combine two parent individuals using flat crossover to produce a single offspring.
Parameters
----------
p1 : Individual
The first parent individual.
p2 : Individual
The second parent individual.
locationsource : Individual
The individual from which to copy positional information for the offspring.
Returns
-------
Individual
The resulting offspring individual.
"""
chsize = len(p1.chromosome)
child = [0 for i in range(chsize)]
for i in range(chsize):
p1_allele = p1.chromosome[i]
p2_allele = p2.chromosome[i]
if p1_allele > p2_allele:
c_max = p1_allele
c_min = p2_allele
else:
c_max = p2_allele
c_min = p1_allele
new_allele = random.uniform(c_min, c_max)
child[i] = round(new_allele, 5)
indv = Individual(p1.gen_type, p1.ch_size)
indv.position = locationsource.position
indv.neighbors_positions = locationsource.neighbors_positions
indv.chromosome = list(child)
indv.fitness_value = self.problem.f(child)
return indv
[docs]
def get_recombinations(self) -> List[Individual]:
"""
Perform the flat crossover on the parent individuals to produce offspring.
Returns
-------
List[Individual]
A list containing the offspring individuals.
"""
p1 = self.parents[0]
p2 = self.parents[1]
return [
self.combine(p1, p2, p1),
self.combine(p1, p2, p2)
]