mesh-adaptation/UM2N
[NeurIPS 2024 Spotlight] Towards Universal Mesh Movement Networks
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.
Stars
19
Forks
—
Language
Python
License
MIT
Category
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.
Higher-rated alternatives
daavoo/pyntcloud
pyntcloud is a Python library for working with 3D point clouds.
yangyanli/PointCNN
PointCNN: Convolution On X-Transformed Points (NeurIPS 2018)
torch-points3d/torch-points3d
Pytorch framework for doing deep learning on point clouds.
yogeshhk/MidcurveNN
Computation of Midcurve of Thin Polygons using Neural Networks
charlesq34/pointnet2
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space