errors

exception zfit.minimizers.errors.NewMinimum

Bases: Exception

Exception class for cases where a new minimum is found.

args

with_traceback()

Exception.with_traceback(tb) – set self.__traceback__ to tb and return self.

zfit.minimizers.errors.compute_errors(result, params, sigma=1, rootf=<function _rootf>, rtol=0.01)

Computes asymmetric errors of parameters by profiling the loss function in the fit result.

Parameters:

• result (FitResult) – fit result
• params (list(Parameter)) – The parameters to calculate the errors error. If None, use all parameters.
• sigma (float) – Errors are calculated with respect to sigma std deviations.
• rootf (callable) – function used to find the roots of the loss function
• rtol (float, default=0.01) – relative tolerance between the computed and the exact roots

Returns:

A OrderedDict containing as keys the parameter and as value a dict which

contains two keys ‘lower’ and ‘upper’, holding the calculated errors. Example: result[par1][‘upper’] -> the asymmetric upper error of ‘par1’

FitResult or None: a fit result is returned when a new minimum is found during the loss scan

Return type:

OrderedDict

zfit.minimizers.errors.get_crossing_value(result, params, direction, sigma, rootf, rtol)

Find the crossing point between the profiled loss function, for given parameters, and the value of errordef for a given direction (positive / negative). errordef = 1 for a chisquare fit, = 0.5 for a likelihood fit.

zfit.minimizers.errors.pll(minimizer, loss, params, values) → FitResult

Compute minimum profile likelihood for given parameters and values.