Minuit

Conveience wrapper for the iminuit.Minuit class, with pre-defined chi-squared (source).

Constructor

minuit(fn, x, y, dy=None, dx=None, dy_low=None, dx_low=None, fn_prime=None, fn_prime_dx=1e-6, name=None, start=None, error=None, limit=None, fix=None, print_level=1, **kwargs)

• fn: function handle. f(x, a, b, c, ...)
• x: x data
• y: y data
• dy: [optional] error in y
• dx: [optional] error in x
• dy_low: [optional] if error in y is asymmetric. If not none, dy is upper error
• dx_low: [optional] if error in y is asymmetric. If not none, dx is upper error
• fn_prime: [optional] function handle for the first derivative of fn. f'(x, a, b, c, ...)
• fn_prime_dx: Spacing in x to calculate the derivative for default calculation
• name: [optional] sequence of strings. If set, use this for setting parameter names
• start: [optional] sequence of numbers. Required if the function takes an array as input or if it has the form f(x, *pars), and name is not defined. Default: 1, broadcasted to all inputs
• error: [optional] sequence of numbers. Initial step sizes.
• limit: [optional] sequence of limits that restrict the range format: [[low, high], [low, high], ...] in which a parameter is varied by minuit., with None, inf or -inf used to disable limit
• fix: [optional] sequence of booleans. Default: False
• print_level: Set the print_level
  • 0 is quiet.
  • 1 prints out at the end of MIGRAD/HESSE/MINOS.
  • 2 prints debug messages.
• kwargs: passed to Minuit.from_array_func and Minuit. To set for parameter "a" one can assign the following keywords instead of the array inputs:
  • a = initial_value
  • error_a = start_error
  • limit_a = (low, high)
  • fix_a = True

Chi-squared is as calculated here.

Functions and Attributes

Of the functionality is described in the iminuit documentation, listed here for reference:

Function/Attribute	Description
minuit.migrad(ncall, iterate)	Run Migrad minimization.
minuit.hesse(ncall)	Run HESSE algorithm to compute asymptotic errors.
minuit.minos(*parameters, cl, ncall)	Run Minos algorithm to compute confidence intervals.
minuit.values	Access parameter values via an array-like view.
minuit.errors	Access parameter parabolic errors via an array-like view.
minuit.merrors	Return a dict-like with Minos data objects.
minuit.fixed	Access whether parameters are fixed via an array-like view.
minuit.limits	Access parameter limits via a array-like view.
minuit.covariance	Return covariance matrix.
minuit.valid	Return True if the function minimum is valid.
minuit.accurate	Return True if the covariance matrix is accurate.
minuit.fval	Get function value at minimum.
minuit.nfit	Get number of fitted parameters (fixed parameters not counted).
minuit.mnprofile(vname, *, size, bound, …)	Get Minos profile over a specified interval.
minuit.draw_mnprofile(vname, *, size, bound, …)	Draw Minos profile over a specified interval (requires matplotlib).

The bfit wrapper also implements the following functions and attributes

Function/Attribute	Description
minuit.chi2	Return reduced chi squared (mintuit.fval / dof) or nan if dof <= 0
minuit.get_merrors(attribute)	Return array composed of the attribute attribute of the items in minuit.merrors
minuit.mat_lower_limit	Return minuit.get_merrors("at_lower_limit")
minuit.mat_lower_max_fcn	Return minuit.get_merrors("mat_lower_max_fcn")
minuit.mat_upper_limit	Return minuit.get_merrors("at_upper_limit")
minuit.mat_upper_max_fcn	Return minuit.get_merrors("at_upper_max_fcn")
minuit.mis_valid	Return minuit.get_merrors("is_valid")
minuit.mlower	Return minuit.get_merrors("lower")
minuit.mlower_new_min	Return minuit.get_merrors("lower_new_min")
minuit.mlower_valid	Return minuit.get_merrors("lower_valid")
minuit.mmin	Return minuit.get_merrors("min")
minuit.mname	Return minuit.get_merrors("name")
minuit.mnfcn	Return minuit.get_merrors("nfcn")
minuit.mnumber	Return minuit.get_merrors("number")
minuit.mupper	Return minuit.get_merrors("upper")
minuit.mupper_new_min	Return minuit.get_merrors("upper_new_min")
minuit.mupper_valid	Return minuit.get_merrors("upper_valid")