8000 Pytorch 2.1.0 CUDA 12.x docker image missing · Issue #110643 · pytorch/pytorch · GitHub
[go: up one dir, main page]

Skip to content

Pytorch 2.1.0 CUDA 12.x docker image missing #110643

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.

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

Closed
bhack opened this issue Oct 5, 2023 · 7 comments
Closed

Pytorch 2.1.0 CUDA 12.x docker image missing #110643

bhack opened this issue Oct 5, 2023 · 7 comments
Labels
module: docker oncall: releng In support of CI and Release Engineering triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Comments

@bhack
Copy link
Contributor
bhack commented Oct 5, 2023

🚀 The feature, motivation and pitch

Please add CUDA 12.x official images.

Alternatives

No response

Additional context

#91122 (comment)

@malfet malfet added oncall: releng In support of CI and Release Engineering triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module module: docker labels Oct 6, 2023
@malfet malfet changed the title Pytorch 2.1.0 CUDA 12.x image missing Pytorch 2.1.0 CUDA 12.x docker image missing Oct 6, 2023
@malfet
Copy link
Contributor
malfet commented Oct 6, 2023

Edited the title, @bhack just to confirm, you are asking for a docker images, as otherwise wheels and conda packages indeed has been published, aren't they?

@bhack
Copy link
Contributor Author
bhack commented Oct 6, 2023

Yes official Docker images:
https://hub.docker.com/r/pytorch/pytorch/tags

@atalman
Copy link
Contributor
atalman commented Oct 6, 2023

@bhack Good call, let me publish CUDA 12.1 Docker versions

pytorchmergebot pushed a commit that referenced this issue Oct 6, 2023
Since our pipy released CUDA version is 12.1.1, Moving the Docker builds to 12.1.1. Related to : #110643
Pull Request resolved: #110703
Approved by: https://github.com/DanilBaibak
atalman added a commit to atalman/pytorch that referenced this issue Oct 6, 2023
Since our pipy released CUDA version is 12.1.1, Moving the Docker builds to 12.1.1. Related to : pytorch#110643
Pull Request resolved: pytorch#110703
Approved by: https://github.com/DanilBaibak
atalman added a commit that referenced this issue Oct 6, 2023
Since our pipy released CUDA version is 12.1.1, Moving the Docker builds to 12.1.1. Related to : #110643
Pull Request resolved: #110703
Approved by: https://github.com/DanilBaibak
@atalman
Copy link
Contributor
atalman commented Oct 6, 2023

@bhack Cuda 12.1 images have been uploaded to: https://hub.docker.com/r/pytorch/pytorch/tags
Please confirm it works for you. And we can probably close this issue

@bhack
Copy link
Contributor Author
bhack commented Oct 11, 2023

@atalman As in pytorch 2.1.0 it is mandoatory c++17 it is a annoying to patch all third party c++ pytorch extensions when building in this dev container to add explicit g++ flags.

Why we are not using a g++ version (like 11) that default c++17? Or why we have not switched to 22.04 LTS to default g++ 11 and c++17?

@malfet
Copy link
Contributor
malfet commented Nov 4, 2023

Closing this one as docker image has been published. @bhack please do not hesitate to propose a patch (or file another issue) if you believe docker container does not function as expected (or want some improvements)
But let me propose a PR to move docker to 22.04, because why not..

@malfet malfet closed this as completed Nov 4, 2023
@bhack
Copy link
Contributor Author
bhack commented Nov 4, 2023

What is required to move to 22.04 and kick-off the CI in a PR?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
module: docker oncall: releng In support of CI and Release Engineering triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
None yet
Development

No branches or pull requests

3 participants
0