-
Notifications
You must be signed in to change notification settings - Fork 26.3k
[FSDP] Allow MixedPrecision to skip inputs #90620
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
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/90620
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 FailuresAs of commit 85a007b: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
rohan-varma
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, but curious what's the use case to skip inputs?
|
Thanks @rohan-varma, one model I am working on has one forward argument, which is sensitive to precision. So have to keep it in fp32, instead of converting it to bfloat16 and convert back. |
awgu
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Overall, this looks good to me. Sorry for leaving all of the comments in separate reviews instead of just one.
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
|
My bad for suggesting to un-default I think you need to add a value for
|
[ghstack-poisoned]
|
This pull request has been merged in 80542ad. |
ghstack-source-id: d9e229f Pull Request resolved: pytorch#90620
Stack from ghstack (oldest at bottom):