[Feature] Support fft based convolution
Open
Nobody has claimed this yet.
enhancement
low priority
performance
- Dominant language
- C++
- Stars
- 28.5k
- Forks
- 2.3k
- Avg merge
- 3d 8h
- Merged PRs (30d)
- 62
Description
It would be nice to have FFT based convolution supported in mlx. FFT bases convolution shows much better performance for large images / arrays and kernels. The FFT building blocks are already supported in mlx, so it is mostly a matter of combining them to a convolution operation.
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 locating MLX's existing FFT building blocks and convolution operation entry points. Determine the supported array and kernel cases, then define how FFT-based convolution should be exposed and validated; the issue does not name specific files or tests, so completion criteria need clarification before implementation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100