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

[AIR] Refactor checkpoint encoding and decoding out of Backend to framework-specific Checkpoints #28462

Open
amogkam opened this issue Sep 13, 2022 · 0 comments
Open

[AIR] Refactor checkpoint encoding and decoding out of Backend to framework-specific Checkpoints #28462

amogkam opened this issue Sep 13, 2022 · 0 comments
Labels
P2 Important issue, but not time-critical train Ray Train Related Issue

Comments

@amogkam
Copy link
Contributor

amogkam commented Sep 13, 2022

Currently, TorchBackend has special logic to be able to serialize GPU tensors, and then deserialize them back on CPU. This logic should be refactored out of the Backends and into TorchCheckpoint.

Open question is what to do about metrics (user reports a GPU tensor as a metric).
@amogkam amogkam added triage Needs triage (eg: priority, bug/not-bug, and owning component) air labels Sep 13, 2022
@anyscalesam anyscalesam added train Ray Train Related Issue and removed air labels Oct 27, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
P2 Important issue, but not time-critical train Ray Train Related Issue
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@amogkam @anyscalesam