-
Notifications
You must be signed in to change notification settings - Fork 88
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
re-enable interleaved 1f1b test #1079
base: gh/H-Huang/3/base
Are you sure you want to change the base?
Conversation
ghstack-source-id: 4a8c1cc4471dc9615d2430080b3b392bfacbb84f Pull Request resolved: #1079
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.
Thanks for enabling the CI. Left one comment.
) | ||
schedule.step_microbatches(microbatches) | ||
schedule.step_microbatches(microbatches, target_mbs=target_mbs) |
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.
Should we feed target to every rank?
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.
we don't have to, but if we do that is also allright too since if it is not the last stage then it wont calculate the loss using the target_mbs.
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.
I'd rather not rely on assumptions :)
We should test the real case (strictly speaking even microbatches
should not be fed to every rank).
torch.randn_like(microbatch) for _ in range(num_microbatches) | ||
] | ||
schedule.step_microbatches(microbatches) | ||
schedule.step_microbatches(microbatches, target_mbs=target_mbs) |
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.
Same here.
Stack from ghstack (oldest at bottom):