Code for spatial-temporal paper

Copyright 2015-2023 Max Shinn, Amber Hu, Laurel Turner

The following is code used to produce the figures in the paper. Because this includes code which evolved slowly over the course of eight years, many of the names of models and metrics, as well as the figure numbering, are inconsistent with the names included in the paper, so it might be hard to find what you are looking for. To make the code usable, we have created the "spatiotemporal" Python package by extracting, renaming, and documenting all of the important functions from this codebase. The code in this repository is for reference only.

The "core" directory contains all of the algorithms and models central to the paper. The "figures" directory contains code to create each of the figures in the manuscript. The "data_wrangling" directory contains code which converts data into different file formats.