ENH: refactor our loop implementations into a stand-alone library
Open
Nobody has claimed this yet.
01 - Enhancement
23 - Wish List
component: numpy.ufunc
component: SIMD
- Dominant language
- Python
- Stars
- 32.8k
- Forks
- 12.8k
- Avg merge
- 1d 7h
- Merged PRs (30d)
- 197
Description
We are slowly getting to a best-of-breed set of CPU inner loop functions using Unversal Intrinsics. It would be helpful to other projects if we could refactor these into a stand-alone C/C++ library, so they could be re-used in the growing number of NumPy-like libraries.
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
Review NumPy’s existing CPU inner-loop implementations and their Universal Intrinsics usage first. Define the scope and interface of a standalone C/C++ library that other NumPy-like projects could reuse; done means the refactoring and reuse boundary are agreed and implemented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- c, cpp
- Domain
- performance
- Issue type
- Refactor
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100