probfit 1.0.2

probfit

probfit is a set of functions that helps you construct a complex fit. It’s intended to be used with iminuit. The tool includes Binned/Unbinned Likelihood estimator, \(\chi^2\) regression, Binned \(\chi^2\) estimator and Simultaneous fit estimator. Various functors for manipulating PDF such as Normalization and Convolution(with caching) and various builtin functions normally used in B physics is also provided.

from probfit import UnbinnedLH, gaussian
from iminuit import Minuit
data = np.randn(10000)
ulh = UnbinnedLH(data)
m = Minuit(ulh, mean=0.1, sigma=1.1)
m.migrad()
ulh.draw(m)

Requirement

• iminuit https://iminuit.github.com/iminuit/

• numpy http://www.numpy.org/

• matplotlib http://matplotlib.org/

Tutorial

open tutorial.ipynb in ipython notebook. You can view it online too.

Documentation

See here