Preserve each Adam group's beta2 in OneCycle - #8495
Open
vineethsaivs wants to merge 1 commit into
Open
Conversation
Signed-off-by: Vineeth Sai <vineethsai4444@gmail.com>
vineethsaivs
requested review from
loadams,
tjruwase and
tohtana
as code owners
September 13, 2026 00:40
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Constructing
OneCyclesilently replaces every Adam group's beta2 with 0.99, changing the second-moment estimator before training starts.Cause: Both momentum bounds hardcode
(beta1, 0.99).Fix: Preserve each group's existing beta2 while scheduling beta1.
Test:
DS_ACCELERATOR=cpu python -m pytest tests/unit/runtime/test_lr_schedulers.py -k preserves_per_group_beta2: fails before, passes after. A six-step linear-model training loop matches Adam with the configured beta2 values through both cycle and decay. All 44 standalone scheduler cases pass. Tested on Apple M2 Pro CPU, torch 2.11.0. Changed-file pre-commit checks pass.