def crossover(population, parent1, parent2):
crossover_point = random.randint(1, 7)
print(f"Crossover will happen at point {crossover_point}")
parent1_genes = parent1.get_genes()
parent2_genes = parent2.get_genes()
child1_genes = BitSet(8)
child2_genes = BitSet(8)
for i in range(crossover_point):
child1_genes.set(i, parent1_genes.get(i))
child2_genes.set(i, parent2_genes.get(i))
for i in range(crossover_point, 8):
child1_genes.set(i, parent2_genes.get(i))
child2_genes.set(i, parent1_genes.get(i))
child1 = Chromossome()
child1.set_genes(child1_genes)
print(
f"1st child generated from crossover: {utils.format_chromossome(child1)}"
)
child2 = Chromossome()
child2.set_genes(child2_genes)
print(
f"2nd child generated from crossover: {utils.format_chromossome(child2)}"
)
population.append(child1)
population.append(child2)
def crossover(population, parent1, parent2):
# K-point crossover
# https://en.wikipedia.org/wiki/Crossover_(genetic_algorithm)#Two-point_and_k-point_crossover
k = random.randint(1, problem.N_QUEENS - 1)
crossover_points = []
for _ in range(k):
point = random.randint(1, problem.N_QUEENS - 1)
while point in crossover_points:
point = random.randint(1, problem.N_QUEENS - 1)
crossover_points.append(point)
crossover_points.sort()
crossover_points_str = ", ".join(map(str, crossover_points))
print(f"{k} crossover(s) will happen at point(s) {crossover_points_str}")
crossover_points.append(problem.N_QUEENS)
parents_genes = [parent1.get_genes(), parent2.get_genes()]
child1_genes = [None] * problem.N_QUEENS
child2_genes = [None] * problem.N_QUEENS
last_point = 0
parent_control = 0
for point in crossover_points:
child1_genes[last_point:point] = parents_genes[parent_control][last_point:point]
child2_genes[last_point:point] = parents_genes[not parent_control][last_point:point]
last_point = point
parent_control = not parent_control
child1 = Chromossome()
child1.set_genes(child1_genes)
print(f"1st child generated from crossover: {utils.format_chromossome(child1)}")
child2 = Chromossome()
child2.set_genes(child2_genes)
print(f"2nd child generated from crossover: {utils.format_chromossome(child2)}")
population.append(child1)
population.append(child2)
def f(chromossome):
route = Chromossome.get_fenotype(chromossome.get_genes())
total_weight = 0
total_vertexes = len(GRAPH)
for i in range(total_vertexes):
vertex = route[i]
next_vertex = route[i + 1]
edge_weight = GRAPH[vertex][next_vertex]
if edge_weight is None:
edge_weight = NON_ADJACENT_WEIGHT
total_weight += edge_weight
return total_weight
def f_chromossome(chromossome):
x, y = Chromossome.get_fenotype(chromossome.get_genes())
return f(x, y)
import matplotlib.pyplot as plt
import modules.problem as problem
from modules.genetics import operators
from modules.genetics import utils
from modules.genetics.chromossome import Chromossome
if __name__ == "__main__":
population = [Chromossome() for _ in range(10)]
generation = 0
population_score = problem.g_average(population)
print(
f"Generation # {generation} -> Average population score = {population_score:.3f}\n"
)
generation_plot = []
generation_plot.append(generation)
population_score_plot = []
population_score_plot.append(population_score)
while generation < 50:
parent1, parent2 = operators.selection(population)
operators.crossover(population, parent1, parent2)
operators.mutation(population)
operators.elitism(population)
generation += 1
population_score = problem.g_average(population)
import matplotlib.pyplot as plt
import modules.problem as problem
from modules.genetics import operators
from modules.genetics import utils
from modules.genetics.chromossome import Chromossome
if __name__ == "__main__":
population = []
population.append(Chromossome(4, 3))
population.append(Chromossome(2, 9))
population.append(Chromossome(9, 11))
population.append(Chromossome(0, 15))
population.append(Chromossome(5, 5))
population.append(Chromossome(14, 3))
generation = 0
population_score = problem.g_average(population)
print(
f"Generation # {generation} -> Average population score = {population_score:.3f}\n"
)
generation_plot = []
generation_plot.append(generation)
population_score_plot = []
population_score_plot.append(population_score)
while generation < 50:
parent1, parent2 = operators.selection(population)
def crossover(population, parent1, parent2):
# Order crossover
# http://mat.uab.cat/~alseda/MasterOpt/GeneticOperations.pdf
parent1_genes = parent1.get_genes()
parent2_genes = parent2.get_genes()
genes_length = len(parent1_genes)
child1_genes = [None] * genes_length
child2_genes = [None] * genes_length
parent1_subset_start = random.randint(0, genes_length - 1)
parent2_subset_start = random.randint(0, genes_length - 1)
parent1_subset_end = random.randint(parent1_subset_start + 1, genes_length)
parent2_subset_end = random.randint(parent2_subset_start + 1, genes_length)
print(
f"1st parent subset indexes: {parent1_subset_start} (inclusive) to {parent1_subset_end} (exclusive)"
)
print(
f"2nd parent subset indexes: {parent2_subset_start} (inclusive) to {parent2_subset_end} (exclusive)"
)
child1_genes[parent1_subset_start:parent1_subset_end] = parent1_genes[
parent1_subset_start:parent1_subset_end]
child2_genes[parent2_subset_start:parent2_subset_end] = parent2_genes[
parent2_subset_start:parent2_subset_end]
parent1_next_index = parent2_next_index = 0
for i in range(genes_length):
if child1_genes[i] is None:
for k in range(parent2_next_index, genes_length):
gene = parent2_genes[k]
if gene in child1_genes:
continue
child1_genes[i] = gene
parent2_next_index = k + 1
break
if child2_genes[i] is None:
for k in range(parent1_next_index, genes_length):
gene = parent1_genes[k]
if gene in child2_genes:
continue
child2_genes[i] = gene
parent1_next_index = k + 1
break
child1 = Chromossome()
child1.set_genes(child1_genes)
print(
f"1st child generated from crossover: {utils.format_chromossome(child1)}"
)
child2 = Chromossome()
child2.set_genes(child2_genes)
print(
f"2nd child generated from crossover: {utils.format_chromossome(child2)}"
)
population.append(child1)
population.append(child2)
import matplotlib.pyplot as plt
import modules.problem as problem
from modules.genetics import operators
from modules.genetics import utils
from modules.genetics.chromossome import Chromossome
if __name__ == "__main__":
population = [Chromossome() for _ in range(10)]
generation = 0
population_score = problem.g_average(population)
print(f"Generation # {generation} -> Average population score = {population_score:.3f}\n")
generation_plot = []
generation_plot.append(generation)
population_score_plot = []
population_score_plot.append(population_score)
while generation < 50:
# pylint: disable=unbalanced-tuple-unpacking
parent1, parent2 = operators.selection(population)
operators.crossover(population, parent1, parent2)
operators.mutation(population)
operators.elitism(population)
generation += 1
population_score = problem.g_average(population)