def next_generation(self):
""" Modify the population for the next generation. """
child_pop = [copy.deepcopy(self.elite_ind)]
# Perform tournament selection.
for i in range(len(self.genomes) - 1):
tourn = random.sample(self.genomes, 2)
if tourn[0]['fitness'] < tourn[1]['fitness']:
child_pop.append(copy.deepcopy(tourn[0]))
else:
child_pop.append(copy.deepcopy(tourn[1]))
child_pop[-1]['id'] = self.child_id
self.child_id += 1
#Crossover
child_pop = single_point_crossover(child_pop, cross_over_prob)
# Mutate genes in the child genomes.
child_pop = random_value_mutation(child_pop, mutation_prob)
for child in child_pop:
child['generation'] = CURRENT_GEN + 1
self.genomes = child_pop
self.id_map = {
k: v
for k, v in zip([x['id'] for x in self.genomes],
[i for i in range(self.pop_size)])
}
def next_generation(self):
""" Modify the population for the next generation. """
child_pop = [copy.deepcopy(self.elite_ind)]
# Perform tournament selection.
for i in range(len(self.genomes) - 1):
tourn = random.sample(self.genomes, TOURNAMENT_SIZE)
fitness_list = []
for ind in tourn:
fitness_list.append(ind['fitness'])
winner_index = fitness_list.index(max(fitness_list))
child_pop.append(copy.deepcopy(tourn[winner_index]))
child_pop[-1]['id'] = self.child_id
self.child_id += 1
#Crossover
child_pop = single_point_crossover(child_pop, CROSS_OVER_PROB)
# Mutate genes in the child genomes.
child_pop = random_value_mutation(child_pop, MUTATION_PROB)
for child in child_pop:
child['generation'] = CURRENT_GEN + 1
self.genomes = child_pop
self.id_map = {
k: v
for k, v in zip([x['id'] for x in self.genomes],
[i for i in range(self.pop_size)])
}