jekim5418/DPM

Official code for DPM : A Novel Training Method for Physics-Informed Neural Networks in Extrapolation

27
/ 100
Experimental

This project helps scientists and engineers accurately predict how physical systems behave beyond the conditions they were initially trained on. It takes equations describing physical processes (like fluid dynamics or quantum mechanics) and uses a new method to train neural networks. The output is a more reliable model that can extrapolate solutions for these complex partial differential equations, especially when traditional methods struggle to predict future states or untested conditions.

No commits in the last 6 months.

Use this if you need to model complex physical phenomena governed by Partial Differential Equations and require highly accurate predictions for conditions outside your initial training data.

Not ideal if your modeling needs are purely interpolative or if you are not working with physics-informed neural networks.

physics-simulation computational-science predictive-modeling engineering-analysis partial-differential-equations
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 14 / 25

How are scores calculated?

Stars

10

Forks

3

Language

Python

License

Last pushed

Nov 02, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jekim5418/DPM"

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