-
Notifications
You must be signed in to change notification settings - Fork 1.1k
[FSDP] Make env var and dataclass flag consistent for cpu_ram_efficient_loading
#3307
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
base: main
Are you sure you want to change the base?
Conversation
Signed-off-by: SumanthRH <sumanthrh@anyscale.com>
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
not stale |
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
not stale. @muellerzr would appreciate a review on this! |
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
not stale |
cc @S1ro1 if you can take a look as you were working on that for fsdpv2 also |
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
not stale, gonna take a look! |
What does this PR do?
Makes environment variable and dataclass consistent for
cpu_ram_efficient_loading
. Currently, there's an inconsistency when the FSDP plugin is initialized in a standalone fashion: thecpu_ram_efficient_loading
flag can be set toTrue
but if the user doesn't useaccelerate launch
, then the environment variableFSDP_CPU_RAM_EFFICIENT_LOADING
would not be set. This environment variable is actually used by transformers to initialize parameters efficiently in the process group (See: #2100 (comment) for details on how this happens)I can see a previous PR #2998 that tried to make all env vars optional, but looks like this was left.
Fixes # (issue)
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
cc @muellerzr