2020-11-03

python实现FR共轭梯度法求极值

python代码实现

import numpy as np
from sympy import *
import math


# 定义符号
x1, x2, t = symbols('x1, x2, t')


def func():
    # return (x1 - 1) ** 2 + (x2 - 1) ** 2
    return 1.5 * x1 ** 2 + 0.5 * x2 ** 2 - x1 * x2 - 2 * x1
    # return x1 ** 2 + 2 * x2 ** 2 - 2 * x1 * x2 - 4 * x1


def grad(data):
    # 求梯度向量,data=[data1, data2]
    f = func()
    grad_vec = [diff(f, x1), diff(f, x2)]  # 求偏导数,梯度向量
    grad = []
    for item in grad_vec:
        grad.append(item.subs(x1, data[0]).subs(x2, data[1]))
    return grad


def grad_len(grad):
    # 梯度向量的模长
    vec_len = math.sqrt(pow(grad[0], 2) + pow(grad[1], 2))
    return vec_len


def zhudian(f):
    # 求得min(t)的驻点
    t_diff = diff(f)
    t_min = solve(t_diff)
    return t_min


def main(X0, theta):
    f = func()
    grad_vec = grad(X0)
    grad_length = grad_len(grad_vec)  # 梯度向量的模长
    k = 0
    grad_list = []
    alpha_list = []
    while grad_length > theta:  # 迭代的终止条件
        k += 1
        if k == 1:
            print("迭代次数：", k, end='\t')
            p = -np.array(grad_vec)
            grad_list.append(grad_vec)
            alpha_list.append(p)
            print("负梯度方向：", p, end='\t')
            print("共轭方向：", p, end='\t')
            # 迭代
            X = np.array(X0) + t * p
            t_func = f.subs(x1, X[0]).subs(x2, X[1])
            t_min = zhudian(t_func)
            print("精确搜索到的步长：", t_min, end='\t')
            X0 = np.array(X0) + t_min * p
            print("下一个迭代点：", X0, end='\t')
            grad_vec = grad(X0)
            grad_length = grad_len(grad_vec)
            print('该点处梯度模长', grad_length)
        else:
            print("迭代次数：", k, end='\t')
            grad_list.append(grad_vec)
            alpha = grad_len(grad_list[k - 1]) ** 2 / grad_len(grad_list[k - 2]) ** 2
            print("alpha", alpha)
            p = -np.array(grad_list[k - 1]) + alpha * np.array(alpha_list[k - 2])
            alpha_list.append(p)
            print("共轭方向：", p, end='\t')
            # 迭代
            X = np.array(X0) + t * p
            t_func = f.subs(x1, X[0]).subs(x2, X[1])
            t_min = zhudian(t_func)
            print("精确搜索到的步长：", t_min, end='\t')
            X0 = np.array(X0) + t_min * p
            print("下一个迭代点：", X0, end='\t')
            grad_vec = grad(X0)
            grad_length = grad_len(grad_vec)
            print('该点处梯度模长', grad_length)



# 给定初始迭代点和阈值
main([-2, 4], 0.1)