nDspec is a software package designed to model (primarly) X-ray astronomical data; in particular, it is meant to enable users to fit spectral, timing and (in the future) polarimetric data natively. Currently, nDspec provides the following features:
- a class to handle folding an energy-dependent model through a given X-ray instrument response matrix. We currently support existing and legacy missions (Swift/XRT, RXTE, NuSTAR, Nicer, XMM-Newton); next-generation missions like XRISM and Athena are not supported yet.
- Classes to handle the standard spectral-timing data products (power spectra, two-dimensional cross spectra, as well as their one-dimensional projections like lag-frequency spectra). Models need to be defined by the user either in the time domain (ie, an impulse response function) or in the Fourier domain (ie, a transfer functions). Additive models can be combined by the users before computing a given Fourier product.
- a small library of phenomenological spectral-timing models, currently including a one dimensional power-law, black body and gaussian, as well as example impulse response and/or transfer functions that parametrize the time dependence of these components.
The goal of the software is to provide users with an Xspec-like package, but to move beyond simple one-dimensional spectral fitting, which typically has relied on each group utilizing their own custom software. Being an open-source modular package, unlike legacy software nDspec allows users to interface with data science libraries for model handling or fitting.
The software documentation, including installation instructions, tutorials, and the full API, is found on readthedocs. You can also find notebooks discussing the features of each class in the /notebooks/ folder.
The stable version of the software can be installed via pip; if you want to use (or contribute to!) features that are still in-developement, you should install the software directly from the repository. Unit tests make use of py.test. Running the unit tests simply requires opening the folder in which users downloaded the repository, and running the command pytest in the command line.
- STINGRAY is a library that allows users to produce typical spectral-timing products from observations. A typical use would consist of using Stingray to handle the data a user might be interested in (e.g. by converting event files into data products for modelling), and nDspec to handle modelling of that data.
- Astropy is the standard library for handling FITS files in a Python environment.
- emcee implements one of the most commonly used Bayesian sampling algorithms in astronomy.
- lmfit provides functions, classes and methods for curve optimization, such as least chi-squared fitting.
nDspec is a fully open source software, so we welcome your contribution and feedback! The best way to contact the developers is through the issues page - even a simple comment on what you find useful or intuitive (or vice versa) goes a long way in helping the project. If you would like to contribute a piece of code, you are welcome to do so either by opening an issue or submitting a pull request. Before you do so, you should read the code of conduct and coding conventions.
The initial release paper is not released yet; until then, please refer to this repository if you make use of the software in your work.
All content © 2023 The Authors. The code is distributed under the MIT license; see LICENSE for details.