SciML / SciML/NeuralOperators.jl
Implement Geom-DeepONet (Point-Cloud DeepONet for 3D Geometries)
Nobody has claimed this yet.
- Dominant language
- Julia
- Stars
- 41
- Forks
- 15
- Avg merge
- 13h 14m
- Merged PRs (30d)
- 12
Description
Summary
Implement Geom-DeepONet for predicting fields on 3D parameterized geometries using point cloud representations.
Reference
- He et al., "Geom-DeepONet: A point-cloud-based deep operator network for field predictions on 3D parameterized geometries," Computer Methods in Applied Mechanics and Engineering, 2024. DOI: 10.1016/j.cma.2024.117130
Description
Geom-DeepONet encodes parameterized 3D geometries as point clouds and predicts full-field solutions on complex geometric domains. This extends DeepONet to handle geometry variation as an input, which is critical for engineering design optimization and shape sensitivity analysis.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the linked Geom-DeepONet paper and the existing NeuralOperators.jl repository to determine the model's required components. The issue names no files, tests, or entry points, so identify where related DeepONet implementations live before planning the work. Done means supporting point-cloud representations of parameterized 3D geometries and predicting full-field solutions on those domains.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- julia
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100