python - 如何制作对数最佳拟合线？

Question

这是我现在拥有的代码，我有一些数据，我希望保留不确定性条。我需要的是什么scipy？

import matplotlib
import numpy as np
import matplotlib.pyplot as plt
import statistics
from statistics import stdev


y1 = [73.64, 76.47, 75.70, 71.71, 73.70, 79.39]
y2 = [219.70, 206.96, 235.31, 189.91, 256.48, 210.25]
y3 = [241.11, 271.70, 255.19, 229.67, 222.30, 200.70]
y4 = [256.59, 263.97, 262.17, 243.14, 245.42, 256.55]

y1_mean = statistics.mean(y1)
y2_mean = statistics.mean(y2)
y3_mean = statistics.mean(y3)
y4_mean = statistics.mean(y4)

y = np.array([y1_mean, y2_mean, y3_mean, y4_mean])
x = np.array([0,0.3,1.5,3])
e = np.array([stdev(y1), stdev(y2), stdev(y3), stdev(y4)])


plt.errorbar(x, y, e, linestyle = 'none', color = 'turquoise' )

plt.scatter(x, y, color = 'green')

plt.xlabel('x-coordinates')
plt.ylabel('y-coordinates')
plt.title('Sample graph')
plt.show()

我希望它是那样的，但适合我的数据：

标签： pythonnumpymatplotlibscipycurve-fitting

Answer 1

由于解释不佳，不太确定您想要什么，但我会尝试通过使用plt.semilogy()和曲线拟合来帮助您。你可以尝试plt.semilogy(x,y)看看你得到了什么，但在这个解决方案中我想拟合曲线，所以这里是编辑过的代码，希望它能帮助你或指导你解决问题：

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
import statistics as st
import scipy as sp
from scipy.optimize import curve_fit

y1 = [6,5,6,7,8]
y2 = [8,9,10,10,11]
y3 = [14,15,16,14,14]
y4 = [16,17,18,19,18]

y1_mean = statistics.mean(y1)
y2_mean = statistics.mean(y2)
y3_mean = statistics.mean(y3)
y4_mean = statistics.mean(y4)

y = np.array([y1_mean, y2_mean, y3_mean, y4_mean])
x = np.array([3,5,6,8])
e = np.array([st.stdev(y1), st.stdev(y2), st.stdev(y3), st.stdev(y4)])


def f(x,a,b,c): 
        return a*(np.square(x))+(b*x)+c

popt, pcov = curve_fit(f, x, y)

fitA = popt[0]
fitB = popt[1]
fitC = popt[0]

yFit = f(x,fitA,fitB,fitC)


plt.errorbar(x, y, e, linestyle = 'none', color = 'k', label="Points" )
plt.semilogy(x,y, color='g', linestyle="--", label="Log")
plt.semilogy(x,yFit,color='r', label="Log Fit")
plt.scatter(x, y, color = 'k')
plt.xlabel('x-coordinates')
plt.ylabel('y-coordinates')
plt.title('Sample graph')
plt.legend()
plt.show()

这就是我得到的