MinKit reference

Here you can find the documentation on the functions and classes of the Minkit package.

Functions

add_pdf_src(path)	This function adds a new path where to look for user-defined PDFs.
binned_chisquare(pdf, data[, range])	Definition of the binned chi-square FCN.
binned_extended_chisquare(pdf, data)	Definition of the binned chi-square FCN for extended PDFs.
binned_extended_maximum_likelihood(pdf, data)	Definition of the binned extended maximum likelihood FCN.
binned_maximum_likelihood(pdf, data)	Definition of the binned maximum likelihood FCN.
data_plotting_arrays(data, **kwargs)	Get the values from a data sample for plotting.
minimizer(fcn, pdf, data[, minimizer, …])	Create a new minimizer to be used within a context.
pdf_from_json(obj[, backend])	Load a PDF from a JSON object.
pdf_plotting_arrays(pdf, data_values, edges)	Scale the PDF values given the data values that have already been plotted using a defined set of edges.
pdf_to_json(pdf)	Dump a PDF to a JSON-like object.
register_pdf(cl)	Decorator to register PDF classes in the PDF_REGISTRY registry.
simultaneous_minimizer(categories[, …])	Create a new object to minimizer a PDF.
sweights(pdfs, yields, data[, range, …])	Calculate the s-weights for the different provided species.
sweights_u(a, sweights[, bins, range])	Get the uncertainty associated to the s-weights related to sample a.
timer()	Create an object that, on exit, displays the time elapsed.
unbinned_extended_maximum_likelihood(pdf, data)	Definition of the unbinned extended maximum likelihood FCN.
unbinned_maximum_likelihood(pdf, data[, range])	Definition of the unbinned maximum likelihood FCN.

Classes

AddPDFs(name, pdfs, yields)	This special PDF defines the sum of many different PDFs, where each of them is multiplied by a factor.
Amoroso(name, x, a, theta, alpha, beta[, …])	Create a new PDF with the name, parameter related to the data and the argument parameters.
Argus(name, x, mu, c, p[, backend])	Create a new PDF with the name, parameter related to the data and the argument parameters.
ArrayOperations(backend, **kwargs)	Build the object to do operations within a backend.
Backend([btype])	Object used in order to do operations with objects of the minkit module.
BinnedDataSet(edges, gaps, pars, values)	A binned data set.
BinnedEvaluator(fcn, pdf, data[, constraints])	Proxy class to evaluate an FCN with a PDF on a BinnedDataSet object.
Category	Create new instance of Category(fcn, pdf, data)
Chebyshev(name, x, *coeffs[, backend])	Build the class given the name, parameter related to data and coefficients.
ConvPDFs(name, first, second[, range])	Represent the convolution of two different PDFs.
CrystalBall(name, x, c, s, a, n[, backend])	Create a new PDF with the name, parameter related to the data and the argument parameters.
DataSet(data, pars[, weights])	Definition of an unbinned data set to evaluate PDFs.
Evaluator()	Object to evaluate an FCN on a set of PDFs and data sets.
ExpPoly(name, x, k, *coeffs[, backend])	Create a new PDF with the parameters related to the data and the slope parameter.
Exponential(name, x, k[, backend])	Create a new PDF with the parameters related to the data and the slope parameter.
Formula(name, formula, pars)	Parameter representing an operation of many parameters.
Gaussian(name, x, center, sigma[, backend])	Create a new PDF with the parameters related to the data, center and standard deviation.
Landau(name, x, center, sigma[, backend])	Create a Landau PDF with the parameters related to the data, center and scale parameter.
Minimizer(evaluator)	Abstract class to serve as an API between Minkit and the different minimization methods.
MinuitMinimizer(evaluator, **minimizer_config)	Interface with the iminuit.Minuit class.
MultiPDF(name, pdfs[, arg_pars])	Base class owing many PDFs.
PDF(name, data_pars, args_pars[, backend])	Build the class from a name, a set of data parameters and argument parameters.
Parameter(name[, value, bounds, ranges, …])	Object to represent a parameter for a PDF.
ParameterBase()	Abstract class for parameter objects.
Polynomial(name, x, *coeffs[, backend])	Build the class given the name, parameter related to data and the coefficients.
PowerLaw(name, x, c, n[, backend])	Build the class given the name, parameter related to data and the coefficients.
ProdPDFs(name, pdfs)	This object represents the product of many PDFs where the data parameters are not shared among them.
Registry(*args, **kwargs)	Extension of list to hold information used in minkit.
SciPyMinimizer(evaluator, **minimizer_config)	Interface with the scipy.optimize.minimize() function.
SimultaneousEvaluator(evaluators[, constraints])	Build an object to evaluate PDFs on independent data samples.
SourcePDF(name, data_pars[, arg_pars, …])	This object defines a PDF built from source files (C++, PyOpenCL or CUDA), which depend on the backend to use.
UnbinnedEvaluator(fcn, pdf, data[, range, …])	Proxy class to evaluate an FCN with a PDF.
barray(array[, length, backend])	Array of booleans.
carray(array[, ndim, length, backend])	They can be of complex type.
darray(array[, ndim, length, backend])	Array of floats.
farray(array, dtype[, ndim, length, backend])	Array of floats.
iarray(array[, length, backend])	Array of integers.
marray(array, dtype[, length, backend])	Wrapper over the arrays to do operations in CPU or GPU devices.

Class Inheritance Diagram