This repository has been archived by the owner on Nov 22, 2022. It is now read-only.
Inverse Sqrt Scheduler #1150
Closed
Inverse Sqrt Scheduler #1150
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Summary:
Currently, WarmupScheduler does this during the warm-up period:
lr = base_lr * current_step / warmup_steps
This diff adds the option of adding LR decay after the warm-up period:
lr = base_lr * sqrt(warmup_steps) / sqrt(current_step)
This is similar to [Fairseq's implementation](https://github.com/pytorch/fairseq/blob/master/fairseq/optim/lr_scheduler/inverse_square_root_schedule.py).
Reviewed By: ccsasuke
Differential Revision: D18491650
fbshipit-source-id: 2e24bd2355759ee8e2e0c1b0e87e600be23f81fb
facebook-github-bot
added
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D18491650 |
|
This pull request has been merged in c35d513. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Currently, WarmupScheduler does this during the warm-up period:
This diff adds the option of adding LR decay after the warm-up period:
This is similar to Fairseq's implementation.
Reviewed By: ccsasuke
Differential Revision: D18491650