ga_1e.py

# mutation
def mutate(lineup, score, prob=0.9):
    gene = deepcopy(lineup.gene)
    # if probability not greater than 0.9
    # flip 0s
    if np.random.random(1)[0] <= prob:
        zero_idx = np.where(gene==0)[0]
        if len(zero_idx) > 0:
            idx = np.random.choice(zero_idx)
            gene[idx] = 1
            return Lineup(score=score, gene=gene)
        else:
            # if not enough 0s to flip
            # return original
            return lineup
    else:
        none_zero_idx = np.where(gene==1)[0]
        if len(none_zero_idx) > 1:
            idx = np.random.choice(none_zero_idx)
            gene[idx] = 1
            return Lineup(score=score, gene=gene)
        else:
            return lineup