HyperIso 1.0.3

Flavour observables, Wilson coefficients and statistical inference through one shared C++ backend

HyperIso is a modular flavour-physics framework for Standard Model and beyond-the-Standard-Model calculations. The same initialization, parameter database, Wilson-coefficient pipeline and observable calculators are exposed through C++, Python, the hyperiso-ui command-line interface and the Dash graphical interface.

Getting started
Build, install and run a first calculation.

Architecture
Modules, data flow and optional backends.

Inputs and parameters
JSON defaults, YAML overrides and LHA-family files.

User interfaces
C++, Python, CLI and Dash workflows.

Statistics
Uncertainties, chi-square fits and confidence contours.

Reproducibility
Frozen release references and provenance.

Extending HyperIso
Where to add physics, bindings, tests and documentation.

C++ API index
Classes, structures, files and call graphs.

Scientific scope

HyperIso evolves the SuperIso calculation model into a layered and extensible software architecture. It supports:

Wilson-coefficient matching and running for the implemented coefficient groups;
native Standard Model, two-Higgs-doublet and supersymmetric workflows;
optional leading-order Wilson coefficients generated for generic BSM models through MARTY;
flavour-observable predictions in the implemented B, D and K decay families;
Monte-Carlo uncertainty propagation, correlated experimental inputs, maximum-likelihood fits, chi-square fits and two-dimensional confidence contours;
LHA, SLHA and FLHA input together with reproducible JSON defaults and YAML overrides.

The exact observable and coefficient catalogue is available from the generated class and enum indices. Features outside the supported release scope are listed in the repository's KNOWN_LIMITATIONS.md file.

Data flow

The diagram below summarizes how numerical inputs enter the runtime database, how model-dependent short-distance information is converted into Wilson coefficients, and how observables feed the statistical layer.

HyperIso high-level data flow

The principal user-facing C++ façades are:

HyperisoMaster for initialization, input paths and optional runtime integrations;
WilsonInterface for Wilson-coefficient matching, running and inspection;
ObservableInterface for observable selection and prediction;
StatisticInterface for uncertainty propagation, fits, scans and contours.

First Python calculation

from pyhyperiso.Common import Model, Observables, QCDOrder
from pyhyperiso.Core import HyperisoConfig, HyperisoMaster
from pyhyperiso.Observable import ObservableInterface
 
config = HyperisoConfig(model=Model.SM)
master = HyperisoMaster()
master.init(lha_file="Assets/lha/si_input.flha", config=config)
 
observables = ObservableInterface()
observables.add_observable(Observables.BR_BS_MUMU, QCDOrder.NNLO)
print(observables.compute_observable(Observables.BR_BS_MUMU))

First C++ calculation

#include <iostream>
 
#include "HyperisoMaster.h"
#include "Include.h"
#include "ObservableInterface.h"
 
int main() {
    HyperisoConfig config;
    config.model = Model::SM;
 
    HyperisoMaster master;
    master.init("Assets/lha/si_input.flha", config);
 
    ObservableInterface observables;
    observables.add_observable(Observables::BR_BS_MUMU, QCDOrder::NNLO);
    std::cout << observables.compute_observable(Observables::BR_BS_MUMU) << '\n';
}

Installation routes

Python users: python -m pip install pyhyperiso==1.0.3 once the release is published.
C++ and CLI users: configure and install the CMake project; see Getting started.
GUI users: install the Python package and the Dash requirements, then start GHyperiso/HyperisoDashGUI/app.py.

Optional backends are explicit. MARTY and SOFTSUSY are not required for standard SM calculations or for the frozen release reference suite. Enable or register them only for workflows that need them.

Publication and citation

The source repository, release assets, reproducibility metadata and CITATION.cff are intended to identify the same immutable release. When using HyperIso in scientific work, cite both the associated article and the software release tag.