import numpy as np import matplotlib as mpl from matplotlib import pyplot as plt from scipy.optimize import minimize x = np.linspace(0.0, 1.0, 30) np.random.seed(42) p = 1.4 a = 0.9 epsilon = 0.3 data = a * x**p + epsilon * (np.random.rand(x.size) - 0.5) def prediction(args): A = args[0] P = args[1] return A * x**P def objective_function(args): rms_error = np.trapz((prediction(args) - data)**2) / (max(x) - min(x)) return rms_error a_guess = 1.0 p_guess = 1.0 solution = minimize(objective_function, [a_guess, p_guess]) print(solution['x']) fig, ax = plt.subplots(figsize=(8,4)) ax.plot(x, data, 'o') ax.plot(x, prediction(solution['x'])) plt.tight_layout() plt.grid() plt.show()