Created
June 5, 2020 19:54
-
-
Save GiliardGodoi/d132bd1039a3d1f2d7576e9a928dfaf7 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import random | |
from evol import Population, Evolution | |
CHROMO_LENGTH = 30 | |
CHROMO_PARTITION = 10 | |
TARGET_VALUE = 52 | |
POPULATION_SIZE = 100 | |
def random_chromosome(size): | |
return ''.join(random.choices(['0', '1'], k=size)) | |
def decode_chromosome(chromosome, size): | |
def bit2int(binary): | |
return int(binary, 2) | |
sequence = [chromosome[i:i+size] for i in range(0,len(chromosome), size)] | |
return tuple(map(bit2int,sequence)) | |
def function(x,y,w): | |
return (x**2 + 2*y + w) | |
def eval_chromosome(chromosome): | |
xyz = decode_chromosome(chromosome, CHROMO_PARTITION) | |
value = function(*xyz) | |
return abs(TARGET_VALUE - value) | |
def crossing_chromosome(chromo1, chromo2): | |
idxs = random.choices(range(0, len(chromo1)), k=2) | |
idx1, idx2 = min(idxs), max(idxs) | |
new_one = chromo1[:idx1] + chromo2[idx1:idx2] + chromo1[idx2:] | |
# newTwo = chromo2[:idx1] + chromo1[idx1:idx2] + chromo2[idx2:] | |
# if (new_one == chromo1): | |
# print(f"Crossing >> {new_one}") | |
# elif (new_one == chromo2): | |
# print(f"Crossing :::::::::::: {new_one}") | |
return new_one | |
def mutation(chromosome, probability=0.8): | |
flip_bit = lambda bit : '1' if bit == '0' else '0' | |
genes = list(chromosome) | |
# print(probability) | |
for i in range(0,len(chromosome)): | |
if random.random() < probability: | |
genes[i] = flip_bit(genes[i]) | |
return ''.join(genes) | |
def select_by_wheel(population : Population): | |
weights = population._individual_weights | |
selected = random.choices(population, weights=weights, k=len(population)) | |
return selected | |
def normalize_by_windowing(population): | |
value = max([chromosome.score for chromosome in population]) + 1 | |
for chromosome in population: | |
chromosome.fitness = abs(value - chromosome.score) | |
return population | |
def pick_random(population): | |
return tuple(random.choices(population, k=2)) | |
def custom_logger(population): | |
best = min(p.fitness for p in population) | |
worst = max(p.fitness for p in population) | |
length = len(population) | |
gen = population.generation | |
print(f"Best: {best} | Worst: {worst} | Length {length} | Generation: {gen}") | |
return population | |
if __name__ == "__main__": | |
pop = (Population(chromosomes=[random_chromosome(CHROMO_LENGTH) for _ in range(POPULATION_SIZE)], | |
eval_function=eval_chromosome, | |
maximize=False).evaluate()) | |
best = min(p.fitness for p in pop) | |
print("Best fitness before evolutionary ", best) | |
evo = (Evolution() | |
.survive(n=30) | |
.breed(parent_picker=pick_random, combiner=crossing_chromosome) | |
.mutate(mutation, probability=0.5) | |
.evaluate() | |
) | |
result = pop.evolve(evo, n=2000) | |
best = min(p.fitness for p in result.evaluate()) | |
print("Best fitness after ::", best) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment