microsoft / microsoft/onnxruntime
Suggestion: add python wheels of various providers to each release
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 21.9k
- Forks
- 4.2k
- Avg merge
- 4d 11h
- Merged PRs (30d)
- 184
Description
For example, I find that dnnl backend is a lot faster than the default backend but I have to manually compile onnxruntime, which is time consuming and error prone. It'll be a lot easier for users if developers who are familiar with code can upload wheels of various providers to each release.
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 how ONNX Runtime is manually compiled and how wheels are uploaded to each release, focusing on the provider variants described in the issue. Determine which release and build entry points would need to support multiple provider wheels; done means users can obtain those wheels from each release instead of compiling manually.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, python
- Domain
- build-system, machine-learning, release
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100