zfit_gaussian_model_sampling_and_maximum_likelihood_fit.py

1import zfit
2import numpy as np
3
4# Create the observable space
5obs = zfit.Space('x', limits=(-10, 10))
6
7# Define the parameters
8mu = zfit.Parameter('mu', 1.2, -4, 6)
9sigma = zfit.Parameter('sigma', 1.3, 0.1, 10)
10
11# Create a Gaussian PDF
12gauss = zfit.pdf.Gauss(mu=mu, sigma=sigma, obs=obs)
13
14# Generate data (sampling) or use your own numpy array
15data_np = np.random.normal(1.3, 1.2, size=1000)
16data = zfit.Data.from_numpy(obs=obs, array=data_np)
17
18# Create a loss function (Negative Log Likelihood)
19nll = zfit.loss.UnbinnedNLL(model=gauss, data=data)
20
21# Create a minimizer
22minimizer = zfit.minimize.Minuit()
23
24# Minimize the loss
25result = minimizer.minimize(nll)
26
27# Estimate errors
28result.errors()
29
30print(result)