backtracks 0.3.4

Python package to fit relative astrometry with background star motion tracks.

backtracks

Python package to fit relative astrometry with background helical motion tracks, to discern directly imaged planets :ringed_planet: from contaminant sources :dizzy: :star:

Written by Gilles Otten (@gotten), William Balmer (@wbalmer), and Tomas Stolker (@tomasstolker).

Documentation

Documentation can be found at http://backtracks.readthedocs.io.

Tutorial

A Jupyter notebook will show you how to use backtracks by reproducing the result in Nielsen et al. (2017) and Wagner et al. (2022) for the case of the former exoplanet candidate around HD 131339 A.

Details

eDR3 Distance prior summary file from this [source](https://arxiv.org/pdf/2012.05220.pdf), published in [Bailer-Jones+2021](https://arxiv.org/abs/2012.05220).

Current example (HD131399Ab) uses data from Wagner+22 and Nielsen+17. Thank you to Kevin Wagner for providing the latest astrometry!

Log-likelihood and some utility functions borrowed heavily from orbitize! (BSD 3-clause). PPF of multivariate normal borrowed from pints (BSD 3-clause).

Installation

Currently requires and python 3.9 ish and astropy, corner, dynesty, matplotlib, numpy, novas, novas_de405, and their dependencies. Note that novas is not supported on Windows. You can create a working environment using conda+pip via a few lines of code:

$ conda create python=3.9 -n backtrack
$ conda activate backtrack
$ conda install pip
$ pip install backtracks

Or, to clone the repo and install in development mode (we recommend this, as the code is a work in progress and you can easily fix bugs you will likely encounter this way):

$ conda create python=3.11 -n backtrack
$ conda activate backtrack
$ conda install pip
$ git clone https://github.com/wbalmer/backtrack.git
$ cd backtrack
$ pip install -e .

Then, test your installation:

>>> from backtracks import System