def __init__(self, filename):
with open(filename, "r") as f:
self.pop_size = int(f.readline())
self.cp = float(f.readline())
self.mp = float(f.readline())
self.n_iterations = int(f.readline())
self.population = Population(self.pop_size, self.cp, self.mp)
def cool_test_once():
population = Population(100, pb, 0.001)
t_end = time.time() + 10
while time.time() < t_end:
population.get_next_gen()
plt.plot(population.stats)
plt.show()
def cool_test_30():
maxims = []
for i in range(30):
pop = Population(40, pb, 0.01)
while pop.generation < 1000:
pop.get_next_gen()
maxims.append(pop.get_best())
print(str(i) + " " + str(maxims[i]))
plt.plot(maxims)
plt.show()
def __init__(self, n, generationsNR, mutationChance, populationSize, breedCuts, selectionSample=None):
self.__n = n
self.__mutationChance = float(mutationChance)
self.__generationsNR = generationsNR
self.__populationSize = populationSize
self.__generations = [Population(n, populationSize)]
self.__breedCuts = breedCuts
self.__selectionSample = selectionSample
self.__generationCounter = 1
def run(self):
best = []
p = Population(self.__n, self.__pop_size, self.__cross_prob,
self.__prob_mut)
for i in range(self.__generations):
p.iteration()
best.append(p.getPop()[0].fitness())
if i % 10 == 0:
self.signal.emit([min(best), i])
self.signal.emit([p.getPop()[0], p.getPop()[0].fitness(), "done"])
class Algorithm:
def __init__(self, filename):
with open(filename, "r") as f:
self.pop_size = int(f.readline())
self.cp = float(f.readline())
self.mp = float(f.readline())
self.n_iterations = int(f.readline())
self.population = Population(self.pop_size, self.cp, self.mp)
def iterate(self):
self.population.evaluate()
self.population.selection()
def run(self):
num_iterations = 0
while num_iterations < self.n_iterations:
self.iterate()
print("ITERATION: {} - FITTEST: {}".format(num_iterations, self.population.best().fitness_score))
num_iterations += 1
return self.population.best()
def run_plot(self):
best_fitness = []
num_iterations = 0
for e in self.population.individuals:
print(e.fitness_score)
while num_iterations < self.n_iterations:
best_fitness.append(self.population.best())
self.iterate()
if self.population.best().fitness_score == 60:
break
print("ITERATION: {} - FITTEST: {}".format(num_iterations, self.population.best().fitness_score))
num_iterations += 1
return best_fitness
def __newGeneration(self, populationList=None):
if populationList is not None:
self.__generations.append(Population(self.__n, self.__populationSize, populationList))
self.__generations.append(Population(self.__n, self.__populationSize, self.__generations[-1].getPopulation()))
def run(self, generations, n, pop_size, cross_prob, prob_mut):
p = Population(n, pop_size, cross_prob, prob_mut)
for i in range(generations):
p.iteration()
return (p.getPop()[0], p.getPop()[0].fitness())