使用 scipy.optimize.curve_fit 我无法正确拟合 V*cos(ω*t) 函数

Question

我已经从 python 中的一个函数创建了数据，所以我认为拟合需要是完美的，但事实并非如此 我不知道为什么

import matplotlib.pyplot as plt
import numpy as np
from scipy.optimize import curve_fit 

def fun(offset, V, f, phi):
        return V * np.cos(2*np.pi*f * x + phi) + offset

    xData = np.arange(0, 1, 0.005) #time
    f = 2 #frequency
    yData = 5 * np.cos(2*np.pi*f * x)

    popt, pcov = curve_fit(fun, xData, yData, p0=[1, 1, 1])

    print(popt)

    plt.scatter(xData, yData, label="data")
    plt.plot(xData, fun(x, *popt), label="fit")
    plt.legend(loc = "upper right")
    plt.show()

plotted data and fit points

Answer 1

存在多个问题，包括您的函数定义和初始猜测。这有效：

import matplotlib.pyplot as plt
import numpy as np
from scipy.optimize import curve_fit 

def fun(xData, offset, V, f, phi):
        return V * np.cos(2*np.pi*f * xData + phi) + offset

xData = np.arange(0, 1, 0.005) #time
f = 2 #frequency
yData = 5 * np.cos(2*np.pi*f * xData)

popt, pcov = curve_fit(fun, xData, yData, p0=[0, 6, 2, 1])

print(popt)

plt.scatter(xData, yData, label="data", c='k')
plt.plot(xData, fun(xData, *popt), label="fit", c='r')
plt.legend(loc = "upper right")
plt.show()