Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

1.16.2 missing wheels #18378

Closed
damonallison opened this issue Nov 9, 2023 · 2 comments
Closed

1.16.2 missing wheels #18378

damonallison opened this issue Nov 9, 2023 · 2 comments

Comments

@damonallison
Copy link

Describe the issue

The latest 1.16.2 release only has a single osx x86_64 wheel on pypi.org, causing us to pin to 1.16.1. Are there plans to publish wheels for all platforms / archs?

https://pypi.org/project/onnxruntime/#files

To reproduce

https://pypi.org/project/onnxruntime/#files

Urgency

No blockers / inconvenience only

Platform

Mac

OS Version

Darwin Kernel Version 22.6.0

ONNX Runtime Installation

Other / Unknown

ONNX Runtime Version or Commit ID

1.16.2

ONNX Runtime API

Python

Architecture

X64

Execution Provider

Default CPU

Execution Provider Library Version

No response
@snnn
Copy link
Member

snnn commented Nov 9, 2023

Thanks for reporting the issue to us. We hit an internal error during the publishing . I will create an internal ticket to ask for support.

@thilomichael thilomichael mentioned this issue Nov 9, 2023
Closed
@jambayk jambayk mentioned this issue Nov 9, 2023
Merged
5 tasks
@tazarov tazarov mentioned this issue Nov 9, 2023
Closed
jambayk added a commit to microsoft/Olive that referenced this issue Nov 10, 2023
@jambayk
[LlaMA2] Sync with onnxruntime 1.16.2 (#711)
21a02c7 
## Describe your changes
onnxruntime 1.16.2 just got released. The llama example and
`OrtTransformersOptimization._replace_mha_with_gqa` are updated to sync
with it.
The calibration dataloaders for Inc smooth quant are removed since we
don't have use for them currently and they are out of date.

**Note:** onnxruntime cpu package is only available for x64 python 3.10
right now. microsoft/onnxruntime#18378

### Results On A100:
**gpu_mha**

![gpu_mha](https://github.com/microsoft/Olive/assets/94929125/00ea1adf-2848-48a6-aa69-63d718027cdf)

**gpu_gqa (no shared kv buffer)**

![gpu_gqa](https://github.com/microsoft/Olive/assets/94929125/37280d0e-b4cb-403c-a91c-dd23599b14cf)

**gpu_gqa (shared kv buffer)**

![gpu_gqa_shared_buffer](https://github.com/microsoft/Olive/assets/94929125/2b1042f8-3df6-404e-9e07-0a356c31c3ef)

## Checklist before requesting a review
- [ ] Add unit tests for this change.
- [ ] Make sure all tests can pass.
- [ ] Update documents if necessary.
- [ ] Lint and apply fixes to your code by running `lintrunner -a`
- [ ] Is this a user-facing change? If yes, give a description of this
change to be included in the release notes.

## (Optional) Issue link
@snnn
Copy link
Member

snnn commented Nov 10, 2023

Resolved.

@snnn snnn closed this as completed Nov 10, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@damonallison @snnn