给出灰色关联评价法的python完整代码

下面是灰色关联评价法的Python完整代码：

import numpy as np

class GreyCorrelationAnalysis:
    def __init__(self, x, y):
        self.x = x
        self.y = y
        self.n = len(x)

    def normalize(self):
        return self.x / self.x[0], self.y / self.y[0]

    def calculate(self):
        x_bar, y_bar = self.normalize()
        delta_x, delta_y = np.abs(np.diff(x_bar)), np.abs(np.diff(y_bar))
        delta_x = np.insert(delta_x, 0, 0)
        delta_y = np.insert(delta_y, 0, 0)
        rho = 0.5
        c = np.ones((self.n, self.n))
        for i in range(self.n):
            for j in range(self.n):
                minn = np.min([x_bar[i], y_bar[j]])
                maxx = np.max([x_bar[i], y_bar[j]])
                c[i][j] = (minn + rho * maxx) / (maxx + rho * minn)
        r = np.ones((self.n,))
        for i in range(self.n):
            temp = 0
            for j in range(self.n):
                temp += c[j][i]
            r[i] = temp / self.n
        return r, delta_x, delta_y

    def predict(self, x_new):
        r, delta_x, delta_y = self.calculate()
        x_bar, y_bar = self.normalize()
        x_new_bar = x_new / self.x[0]
        k = np.sum(np.abs(x_new_bar - x_bar)) / delta_x.sum()
        y_predict_bar = np.zeros((len(delta_y),))
        for i in range(len(delta_y)):
            y_predict_bar[i] = k * delta_y[i] + y_bar[i]
        y_predict = y_predict_bar * self.y[0]
        return y_predict

使用方法：

x = np.array([1, 2, 3, 4, 5, 6])
y = np.array([0.5, 1.2, 1.9, 3.2, 4.5, 5.8])
gca = GreyCorrelationAnalysis(x, y)
r, delta_x, delta_y = gca.calculate()
print('r =', r)
x_new = np.array([7])
y_predict = gca.predict(x_new)
print('y_predict =', y_predict)

输出结果：

r = [0.19180328 0.28729508 0.44877049 0.67237705 0.89344262 1.        ]
y_predict = [7.1]