Draft
Conversation
- Each rank uses a different LR from a log-spaced range (2 orders of magnitude) - After evaluation, all_reduce MAX to get best score across ranks - Auto-disables if world_size < 2 or > 20 - New flag: DownstreamTaskConfig.rank_max_lr (default: False) Co-authored-by: henryh <henryh@allenai.org>
|
Cursor Agent can help with this pull request. Just |
- LRs: [1e-4, 5e-4, 1e-3, 5e-3, 1e-2, 5e-2, 1e-1, 5e-1] - Auto-disable if world_size != 8 (must match number of sweep LRs) Co-authored-by: henryh <henryh@allenai.org>
Co-authored-by: henryh <henryh@allenai.org>
… low priority Co-authored-by: henryh <henryh@allenai.org>
gabrieltseng
reviewed
Mar 5, 2026
| RANK_MAX_LRS = [1e-4, 5e-4, 1e-3, 5e-3, 1e-2, 5e-2, 1e-1, 5e-1] | ||
|
|
||
|
|
||
| def get_rank_lr(rank: int, world_size: int) -> float | None: |
Collaborator
There was a problem hiding this comment.
world_size is not necessary for this function
gabrieltseng
reviewed
Mar 5, 2026
| select_final_test_miou_based_on_epoch_of_max_val_miou: bool = False, | ||
| n_bootstrap: int = 0, | ||
| bootstrap_seed: int = 42, | ||
| rank_max_lr: bool = False, |
Collaborator
There was a problem hiding this comment.
i feel like this should be true by default?
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add
rank_max_lrto enable per-rank learning rates for linear probe evaluation, providing a free learning rate sweep during in-loop evaluations.