genetic_sim.py

import random

def mixUp(x, y):
	
	child_x = random_child(x)
	child_y = random_child(y)

	num_rows =len(x)
	num_cols = len(x[0])

	new_child = []

	for i in range(num_rows):
		new_child.append([0]*num_cols)

	for i in range(num_rows):
		for j in range(num_cols):
			new_child[i][j] = (child_x[i][j], child_y[i][j])

	is_pure_blood = findOut(new_child)

	return is_pure_blood, new_child

def random_child(x):

	new_child = []

	num_rows =len(x)
	num_cols = len(x[0])

	for i in range(num_rows):
		new_child.append([0]*num_cols)

	for i in range(num_rows):
		for j in range(num_cols):
			coin_flip = random.random()
			if coin_flip < 0.5:
				new_child[i][j] = x[i][j][0]
			else:
				new_child[i][j] = x[i][j][1]

	return new_child

def findOut(candidate):

	num_rows =len(candidate)
	num_cols = len(candidate[0])

	for i in range(num_rows):
		for j in range(num_cols):
			if candidate[i][j][0] != candidate[i][j][1]:
				return False

	return True

def simulate(x, y, num_simulations):

	global_ans = 0

	for _ in range(num_simulations):
		ans = 0
		while True:
			ans += 1
			is_pure_blood_one, new_child_one = mixUp(x, y)
			is_pure_blood_two, new_child_two = mixUp(x, y)
			if is_pure_blood_one:
				print 'breaking after {0} generation, with x = {1}, y = {2}, child = {3}'.format(ans, x, y, new_child_one)
				break
			
			x = new_child_one
			y = new_child_two
		
		global_ans += ans

	return global_ans / float(num_simulations)