Distributed inference

Open jleinonen opened this issue 7 months ago • 5 comments

Earth2Studio Pull Request

Description

Checklist

[ ] I am familiar with the Contributing Guidelines.
[ ] New or existing tests cover these changes.
[ ] The documentation is up to date with these changes.
[ ] The CHANGELOG.md is up to date with these changes.
[ ] An issue is linked to this pull request.

Dependencies

May 20 '25 17:05 jleinonen

I am wondering whether it would make sense to support the distributed recipe on a single GPU as well. So users can test the code with a single GPU and then move to more GPUs later.

May 21 '25 14:05 swbg

/blossom-ci

May 26 '25 14:05 jleinonen

/blossom-ci

Jun 02 '25 12:06 jleinonen

Will review design of distributed utility closely and likely suggest modifications.

Jun 03 '25 16:06 dallasfoster

NOTE: It was reported to me that this had problems with PyTorch 2.7 so we should test with that and fix the problem if necessary before merging.

Jul 02 '25 14:07 jleinonen