pop_int = convertPOP2Reseau(pop=pop,
reseau_BIT_size=r_bit_len,
reseau_amount=r_amount,
reseau_bound=X_BOUND)
fitness_input = list(map(lambda x: (x, sh_index), pop_int))
# 计算种群的适应度
pop_fitness = pool.starmap(fitness, fitness_input) # 可以放入可迭代对象，自动分配进程
# pop_fitness = list(map(lambda x: fitness(x, sh_index), pop_int))
# 调整适应度
last_fitness = get_fitness(pop_fitness)
# 根据适应度进行选择
pop = select(pop, last_fitness)
# 交叉变异
pop = C_M(pop, CROSS_RATE, DNA_SIZE, POP_SIZE, MUTATION_RATE)
print('本次最高的适应度为：' + str(np.max(pop_fitness)) + ' 耗时：' +
str(round(time.time() - t_s, 1)) + '秒！')
print('最终的网格：' +
str(pop_int[list(last_fitness).index(np.max(last_fitness))]))
result_money.append(np.max(pop_fitness))
# 训练结束，找出最好的
print('收益提高情况：' + str(result_money))
print('最终的网格：' +
str(pop_int[list(last_fitness).index(np.max(last_fitness))]))
for _ in range(N_GENERATIONS):
F_values = F(translateDNA(
pop, DNA_SIZE,
X_BOUND)) # 计算适应度的值？ compute function value by extracting DNA
# 画图相关
if 'sca' in globals(): sca.remove()
sca = plt.scatter(translateDNA(pop, DNA_SIZE, X_BOUND),
F_values,
s=200,
lw=0,
c='red',
alpha=0.5)
plt.pause(0.05)
# 进行进化
fitness = get_fitness(F_values)
print("Most fitted DNA: ", pop[np.argmax(fitness), :])
pop = select(pop, fitness, len(pop))
# pop_copy = pop.copy()
pop = C_M(pop, CROSS_RATE, DNA_SIZE, POP_SIZE, MUTATION_RATE)
# for parent in pop:
# child = crossover(parent, pop_copy)
# child = mutate(child)
# parent[:] = child # 子孙代替父辈
plt.ioff()
plt.show()