SciML / SciML/NeuralOperators.jl
Implement U-FNO (U-Net Enhanced Fourier Neural Operator)
Nobody has claimed this yet.
- Dominant language
- Julia
- Stars
- 41
- Forks
- 15
- Avg merge
- 13h 14m
- Merged PRs (30d)
- 12
Description
Summary
Implement U-FNO, which adds U-Net-style encoder-decoder paths to the Fourier Neural Operator.
Reference
- Wen et al., "U-FNO -- An enhanced Fourier neural operator-based deep-learning model for multiphase flow," Advances in Water Resources, 2022. DOI: 10.1016/j.advwatres.2022.104180
Description
U-FNO augments the standard FNO with U-Net-style skip connections and encoder-decoder paths operating in physical space, in addition to the spectral-domain Fourier layers. This improves the model's ability to capture multi-scale features, particularly for complex multiphase flow applications.
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
No files, tests, or entry points are named. Start by locating the existing Fourier Neural Operator implementation and read Wen et al.'s U-FNO paper to map the encoder-decoder paths, physical-space operations, and skip connections onto the project. Done means U-FNO is implemented and its behavior is validated against the repository's existing operator tests or examples.
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