[Feat] Gradient sync is turned off during gradient accumulation.

[EquationWalker]

Gradient sync should be turned off during gradient accumulation.

FSDPModule.set_requires_gradient_sync
DDP.set_requires_gradient_sync

[meta-cla]

Hi @EquationWalker!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at [email protected]. Thanks!

[meta-cla]

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

[tianyu-l]

IIUC, for FSDP, there are two ways of doing this -- doing gradient sync requires extra communication but reduce the memory overhead. Usually the gradient sync should be well-hidden, and doing it by default makes the code simpler. Hence the current choice.

See more discussions here #292 (comment)