Another found on the hard drive gist, sticking it here for later, not sure that it works.

genetic_algo.py

from random import random, randint

letters="ABCDEFGHIJKLMNOPQRSTUVWXYZ"

def individual(len):
    return [letters[int(round(random()*100)%26)] for x in xrange(len)]

def population(size, len):
    return [individual(len) for x in xrange(size)]

def fitness(source, target):
    count=0
    for i in xrange(len(source)):
        if target[i]==source[i]:
            count = count+1
    return count

def grade(pop, target):
    grades=[]
    for p in pop:
        grades.append(fitness(p, target))
    return sum(grades)/float(len(grades))

def avg(grades):
    return sum(grades)/float(len(grades))

def evolve(pop, target, retain=0.2, random_select=0.05, mutate=0.01):
    grades = [(fitness(x, target),x) for x in pop]
    grades = [x[1] for x in sorted(grades)]
    length = int(len(grades) * retain)
    parents= grades[:length]
    print "Parents - ", parents


    return parents

if __name__ == "__main__":
    target = ['H','E','L','L','O']
    indsize = len(target)
    popsize = 100
    pop = population(popsize, indsize)
    print "Population: ", pop
    fitness_history = [grade(pop, target),]
    for i in xrange(100):
        pop = evolve(pop, target)
        fitness_history.append(grade(pop,target))
        print fitness_history
    #for datum in fitness_history:
        #print datum, pop