python人工智能算法之差分进化算法的实现

import numpy as np
from random import sample, randint, random
def evolve(xs,cr,func):
    # 变异
    vs = []
    N = len(xs[0])
    for _ in range(len(xs)):
        x = xs.pop(0)
        r1, r2, r3 = sample(xs, 3)
        xs.append(x)
        vs.append(r1 + random()*(r2-r3))
    # 交叉
    us = []
    for i in range(len(xs)):
        us.append(vs[i] if random() < cr else xs[i])
        j = randint(0, N-1)
        us[i][j] = vs[i][j]
    # 选择
    xNext = []
    for x,u in zip(xs, us):
        xNext.append(x if func(x)<func(u) else u)
    return xNext

# N为个体数；nDim为解维度；nIter为迭代次数
def de(N, nDim, xRange, cr, func, nIter):
    xs = [uniRand(*xRange, nDim) for _ in range(N)]
    for _ in range(nIter):
        xs = evolve(xs, cr, func)
    fs = [func(x) for x in xs]
    xBest = xs[np.argmin(fs)]
    msg = f"当前最优结果为{np.min(fs)}，参数为"
    msg += ", ".join([f"{x:.4f}" for x in xBest])
    print(msg)

def test(xs):
    _sum = 0.0
    for i in range(len(xs)):
        _sum = _sum + np.cos((xs[i]*i)/5)*(i+1)
    return _sum
​​​​​​​if __name__ == "__main__":
    de(20, 5, (-5,5), 0.5, test, 100)