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

cpu_offload is depreciated #469

Closed
StellaAthena opened this issue Nov 21, 2021 · 2 comments · Fixed by #720
Closed

cpu_offload is depreciated #469

StellaAthena opened this issue Nov 21, 2021 · 2 comments · Fixed by #720
Assignees
@EricHallahan
Labels
bug Something isn't working

Comments

@StellaAthena
Copy link
Member

Describe the bug
Running the model gives the following warning: [2021-11-20 20:08:18,491] [WARNING] [config.py:77:_sanity_check] DeepSpeedConfig: cpu_offload is deprecated. Please use offload_optimizer.

We should update the way that our code works to use the currently supported methodology.
@StellaAthena StellaAthena added the bug Something isn't working label Nov 21, 2021
@EricHallahan
Copy link
Contributor

I actually have looked into this warning message on-and-off for a couple months now at a low priority as it has bugged me too. I'll assign myself to the task.

@EricHallahan EricHallahan self-assigned this Nov 21, 2021
@StellaAthena StellaAthena mentioned this issue Feb 25, 2022
Open
@Quentin-Anthony
Copy link
Member

Quentin-Anthony commented Nov 25, 2022 •
edited
Loading

We can just remove the cpu_offload flags in the configs. DS has moved to a offload_optimizer dict (https://www.deepspeed.ai/docs/config-json/#optimizer-offloading) instead of a single bool anyway. I'll create a PR.

@Quentin-Anthony Quentin-Anthony mentioned this issue Nov 25, 2022
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@EricHallahan EricHallahan

Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Deprecate cpu_offload zero_optimization Option EleutherAI/gpt-neox
3 participants
@Quentin-Anthony @StellaAthena @EricHallahan