def initialize(self, max_population):
self.tot_fitness = 0
self.max_fitness = 0
self.max_ind = None
while self.tot_fitness == 0:
self.individuals = []
for i in range(max_population):
ind = Individual(self.test)
ind.initialize(self.alphabet)
ind.calc_fitness()
self.tot_fitness += ind.fitness
if ind.fitness > self.max_fitness:
self.max_fitness = ind.fitness
self.max_ind = ind
self.individuals.append(ind)
def synthesize_population(self):
population = Population()
populace = []
for i in range(0, config.populationSize):
gene = Individual()
gene.individual = gene.create_individual()
gene.fitness = gene.calc_fitness(gene.individual)
populace.append(gene)
population.populace = populace
return population
def mutate(self,child_1):
chromosome = Individual()
individual = child_1.individual
start_index = 1
end_index = len(individual) - 2
mutate_index = random.randrange(start_index, end_index)
individual[mutate_index] = np.random.uniform()
individual = chromosome.normalize_fracs(individual)
# update id and params
individual[0] = str(chromosome.generate_id())
chromosome.update_params(individual)
child = Individual()
child.individual = individual
child.fitness = chromosome.calc_fitness(individual)
return child
def crossover(self, individual1, individual2):
# crossover fractions
parent_1 = individual1.individual
parent_2 = individual2.individual
chromosome = Individual()
start_index = 1
end_index = len(parent_1) - 2
crossover_index = random.randrange(start_index, end_index + 1)
child_1a = parent_1[crossover_index:]
child_1b = parent_2[:crossover_index]
child_1 = child_1b + child_1a
child_2a = parent_2[crossover_index:]
child_2b = parent_1[:crossover_index]
child_2 = child_2b + child_2a
# crossover lake's veg type
if np.random.uniform() > 0.5:
tmp = child_1[len(child_1) - 1]
child_1[len(child_1) - 1] = child_2[len(child_2) - 1]
child_2[len(child_2) - 1] = tmp
# normalize fractions
child_1 = chromosome.normalize_fracs(child_1)
child_2 = chromosome.normalize_fracs(child_2)
# update id and params
child_1[0] = chromosome.generate_id()
child_2[0] = chromosome.generate_id()
update = Update()
update.update_params(child_1)
update.update_params(child_2)
# chromosome.update_params(child_1)
# chromosome.update_params(child_2)
child1 = Individual()
child2 = Individual()
child1.individual = child_1
child2.individual = child_2
child1.fitness = chromosome.calc_fitness(child_1)
child2.fitness = chromosome.calc_fitness(child_2)
return child1, child2
self.test, self.alphabet, self.mutation_rate, new_individuals
)
# print("\nNew Populaition:")
# [print(r.dna) for r in new_Population.individuals]
return new_Population
if __name__ == "__main__":
alphabet = string.digits + string.ascii_letters + string.punctuation + " "
in1 = Individual("This is a string", "abcdefghijklmnop")
in2 = Individual("This is a string", "1234567890123456")
in3 = Individual("This is a string", "IIIIIIIIIIIIIIII")
in4 = Individual("This is a string", "OOOOOOOOOOOOOOOO")
in1.calc_fitness()
in2.calc_fitness()
in3.calc_fitness()
in4.calc_fitness()
inds = [in1, in2, in3, in4]
pop = Population("This is a string", alphabet, 0.01, inds)
# pop.calc_values()
pop.initialize(20)
pop.reproduce()
# print(pop.max_population, pop.test, pop.alphabet, pop.dna_length)
# print(len(pop.individuals))
# in3, in4 = pop.crossover(pop.individuals[0], pop.individuals[1])
# print(in3.dna)
# print(in4.dna)
