Jbrich95/pinnEV

Partially-Interpretable Neural Networks for Extreme Value modelling

30
/ 100
Emerging

This tool helps researchers and analysts model rare or extreme events, like wildfires or unusual weather patterns, using a combination of statistical distributions and neural networks. You provide data on an event (the 'response') and related influencing factors ('covariates'), and it outputs a model that describes the probability and characteristics of these extreme events. This is for statisticians, climate scientists, actuaries, or anyone working with extreme value analysis to better understand and predict rare occurrences.

No commits in the last 6 months.

Use this if you need to model the probability and intensity of extreme events (like severe storms, financial crises, or disease outbreaks) and want a flexible modeling approach that combines the interpretability of traditional statistics with the power of neural networks.

Not ideal if your data explicitly uses -1e10 as a valid observation value, or if you require models to train exclusively on GPUs for very large datasets.

extreme-value-analysis risk-modeling environmental-science climatology actuarial-science
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 4 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

8

Forks

1

Language

R

License

MIT

Last pushed

Jun 02, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Jbrich95/pinnEV"

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