ThomasMBury/deep-early-warnings-pnas

Repository to accompany the publication 'Deep learning for early warning signals of tipping points', PNAS (2021)

49
/ 100
Emerging

This project provides tools to analyze time-series data to detect early warning signs of 'tipping points' – sudden, large shifts in complex systems like ecosystems or climate. It takes historical time series measurements (e.g., climate indicators, population sizes) and outputs predictions about the likelihood of an impending critical transition. Researchers and scientists studying complex dynamic systems would use this to anticipate critical shifts.

Use this if you need to reproduce the deep learning methodology for detecting early warning signals of tipping points described in the PNAS publication.

Not ideal if you're looking for a ready-to-use package for applying these techniques to your own data, as that is covered by the 'ewstools' library.

complex-systems climate-modeling ecosystem-dynamics critical-transitions time-series-analysis
No Package No Dependents
Maintenance 6 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

65

Forks

22

Language

Python

License

Last pushed

Nov 28, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ThomasMBury/deep-early-warnings-pnas"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.