fitt

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


df = pd.DataFrame({'energy': [81, 121.78, 244.31, 276.398, 302.85, 344.31, 356.01, 778.9, 963.39, 1112.07, 1408.01],
                   'efficiency_meas_ring1': [0.0031, 0.0044, 0.0056, 0.0052, 0.0050, 0.0046, 0.0045, 0.0030, 0.0026, 0.0024, 0.0021], 
                   'efficiency_meas_ring2': [0.0052, 0.0105, 0.0102, 0.0095, 0.0091, 0.0085, 0.0083, 0.0056, 0.0050, 0.0046, 0.0039], 
                  'yerr_1': [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1], 
                  'yerr_2': [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1]})

def eff_fit(x, a, b, c, d, e, f, g):
    return (a + b*np.log(x) + c*np.log(x)**2 + d*np.log(x)**3 + e*np.log(x)**4 + f*np.log(x)**5 + g*np.log(x)**6)/x
    
xdata = df['energy']
ydata = df['efficiency_meas_ring1']*(1/0.012)
ydata_2 = df['efficiency_meas_ring2']*(1/0.012)

sigma_ring1 = df['yerr_1']*0.25
sigma_ring2 = df['yerr_2']*0.25

# Plot the actual data
plt.plot(xdata, ydata, ".", label="Data_ring1");
plt.plot(xdata, ydata_2, "*", label="Data_ring2");

# The actual curve fitting happens here
optimizedParameters1, pcov = curve_fit(eff_fit, xdata, ydata);
optimizedParameters2, pcov = curve_fit(eff_fit, xdata, ydata_2);

# Use the optimized parameters to plot the best fit
x = np.linspace(80,1500,10000)
plt.plot(x, eff_fit(x, *optimizedParameters1), label="fit_ring1");
plt.plot(x, eff_fit(x, *optimizedParameters2), label="fit_ring2");

plt.errorbar(xdata, ydata, sigma_ring1, fmt = 'o',color = 'orange', 
            ecolor = 'lightgreen', elinewidth = 2, capsize=5)
plt.errorbar(xdata, ydata_2, sigma_ring2, fmt = 'o',color = 'orange', 
            ecolor = 'lightgreen', elinewidth = 2, capsize=5)


# Show the graph
plt.xlabel('E[keV]')
plt.ylabel('efficiency [a.u]')
plt.legend();
plt.show();