IRB - Instrument Response Builder

Introduction

The instrument response builder (IRB) was written to assert the performance of an astrophysics observatory for a set of signal and background events. The output distributions range from angular and energy resolutions to effective areas and background rates.

• Input: a set of physical events (signal and background particles) as a data table.
• Output: a set of performance plots.

Description

The instrument response builder (IRB) was written to assert the performance of astrophysics observatories. It was developed in the context of CTA but can easily be extended to accommodate other observatories as well.

As input the "IRF builder" takes data tables with the reconstructed and classified events. In the typical case that the energy spectrum of the simulated events has not been generated according to a physical flux, the events have to be reweighted before they can be used to estimate the detector performance.
Given sufficient knowledge over the Monte Carlo generator settings, the IRB package provides functionality to generate an energy-dependent event weight to make the simulated events follow any desired distribution.

The built-in event selection method determines the optimal cuts on the angular distance to an assumed source and the signal-probability so that the final sensitivity according to Li & Ma (1983) is minimised.
In the end, various distributions are generated to gauge performance of the detector. The distributions range from angular and energy resolutions to effective areas and background rates. They can be stored as plots in various picture formats or as data tables in an HDF5 output file.

Dependencies

• numpy, scipy, scikit-learn, astropy, pandas, yaml, matplotlib'
• gammapy

Installation

1. git clone the repository
2. python setup.py install    
3. Test (to be finalised)

More details on the installation procedure are available on the documentation page.

Authors and Contact

• Jérémie Decock
• Tino Michael
• Thierry Stolarczyk
• Karl Kosack