Python实现人工蜂群算法的示例代码

import numpy as np
uniform = np.random.uniform
srcs = [[uniform(xL, xR), 0] for _ in range(nSrc)]

def genX(srcs, index):
    xs = srcs[index][0]
    diff = np.delete(np.arange(len(srcs)), index)
    index = rand.choice(diff)
    xTest = srcs[index][0]
    d = randint(0, len(xs) - 1)     # 更改的数据维度
    r = uniform(-1, 1)              # 随机因子
    xNew = np.copy(xs)
    xNew[d] = xs[d] + r * (xs[d] - xTest[d])
    return xNew
def leadStep(srcs, func, i):
    food = srcs[i]
    xNew = genX(srcs, i)
    if func(xNew) < func(food[0]):
        srcs[i] = [xNew, 0]
    else:
        food[1] += 1

def followStep(srcs, func):
    indRange = range(len(srcs))
    fs = [func(food[0]) for food in srcs]
    ps = np.array(fs)/np.sum(fs)
    index = rand.choices(indRange, ps)[0]
    food = srcs[index]
    xNew = genX(srcs, index)
    if func(xNew) < func(food[0]):
        srcs[index] = [xNew, 0]
    else:
        food[1] += 1

def spyStep(srcs, maxTrial, xL, xR, i):
    if srcs[i][1] > maxTrial:
        srcs[i] = [uniform(xL, xR), 0]

def ABC(func, nIter, xL, xR, 
    nPop, perEm=0.5, maxTrial=50):
    nSrc = round(nPop * perEm)  # 雇佣/跟随蜂数
    nSpy = nPop - nSrc          # 侦察蜂数
    xL, xR = np.array(xL), np.array(xR)
    srcs = [[uniform(xL, xR), 0] for _ in range(nSrc)]
    for _ in range(nIter):
        for i in range(nSrc):
            leadStep(srcs, func, i)
            spyStep(srcs, maxTrial, xL, xR, i)
        for i in range(nSpy):
            followStep(srcs, func)
    best = np.argmax([func(food[0]) for food in srcs])
    xs = srcs[best][0]
    msg = "当前参数：" + ",".join([f"{x:.5f}" for x in xs])
    msg += f"\n 最佳值{test(xs)}"
    print(msg)
    return srcs[best][0]

def test(xs):
    _sum = 0.0
    for i in range(len(xs)):
        _sum = _sum + np.cos((xs[i]*i)/5)*(i+1)
    return _sum
def main():
    xL=np.full(5,-20)
    xR=np.full(5,20)
    ABC(test, 49, xL, xR, 30)
if __name__ == '__main__':
    main()