Optimizing bogosort genetic algorithm

bogo_genetic.py

from random import sample, uniform, randrange, shuffle
from time import sleep

def breed(mother, father):
    fetus = mother[:randrange(len(mother))]
    sperm = list(filter(lambda x: x not in fetus, father))
    return fetus + sperm

def fitness(array):
    return len(list(filter(lambda x: x[0]==x[1], 
        zip(array, sorted(array))))) / float(len(array))

def wheel_choice(items):
    p = uniform(0.0, sum(map(fitness, items)))
    for item in items:
        p -= fitness(item)
        if p < 0: return item
    return item

def gen_shuffle(array):
    childs = [sample(array, len(array)) for _ in range(8)]
    mother, father = wheel_choice(childs), wheel_choice(childs)
    return breed(mother, father)

def bogo_shuffle(x):
    shuffle(x)
    return x

def bogo(x, func):
    cnt = 0
    while x != sorted(x):
        x = func(x)
        cnt += 1
    return cnt

def test(array, epoch, func):
    total = 0
    for _ in range(epoch):
        total += bogo(array[:], func)

    print ("""FUNCTION: {}
epoch: {}
array: {}
total: {}
average: {}""".format(func, epoch, array, total, total / float(epoch)))

epoch = 32
size = 8
array = sample(range(size), size)

test(array, epoch, bogo_shuffle)
test(array, epoch, gen_shuffle)