mesh-adaptation/UM2N

[NeurIPS 2024 Spotlight] Towards Universal Mesh Movement Networks

24
/ 100
Experimental

This project helps scientific and engineering professionals accurately and efficiently solve complex Partial Differential Equations (PDEs). It takes a computational mesh as input and outputs an adapted mesh that improves the accuracy of numerical solutions without increasing computational cost. It's designed for researchers, engineers, and scientists working with simulations in fields like fluid dynamics or structural analysis.

No commits in the last 6 months.

Use this if you need to perform mesh adaptation for various PDE types and complex boundary geometries without requiring extensive re-training for each new scenario.

Not ideal if your mesh adaptation needs are simple and can be handled by conventional, less computationally intensive methods or if you prefer strictly deterministic, non-learning-based approaches.

computational-fluid-dynamics finite-element-analysis numerical-simulation scientific-computing engineering-analysis
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

19

Forks

Language

Python

License

MIT

Last pushed

Jul 16, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mesh-adaptation/UM2N"

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