fix(flaky): Different approach to make sure loss exists#43804
Merged
tarekziade merged 4 commits intomainfrom Feb 20, 2026
Merged
fix(flaky): Different approach to make sure loss exists#43804tarekziade merged 4 commits intomainfrom
tarekziade merged 4 commits intomainfrom
Conversation
Collaborator
Author
|
run-slow: ernie4_5_moe |
Contributor
|
This comment contains models: ["models/ernie4_5_moe"] |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
tarekziade
commented
Feb 6, 2026
|
|
||
| # This is to mimic torch.testing.assert_not_close | ||
| self.assertNotAlmostEqual(include_padding_result.aux_loss.item(), result.aux_loss.item()) | ||
| # We make sure that masking can change the loss using a deterministic synthetic example. |
Collaborator
Author
There was a problem hiding this comment.
I assume calling load_balancing_loss_func is good enough to validate loss diff, and it could be its own test. But it also ignore the model() path. is this ok?
Collaborator
Author
|
may be related #43775 |
ArthurZucker
approved these changes
Feb 9, 2026
|
|
||
| # This is to mimic torch.testing.assert_not_close | ||
| self.assertNotAlmostEqual(include_padding_result.aux_loss.item(), result.aux_loss.item()) | ||
| # We make sure that masking can change the loss using a deterministic synthetic example. |
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: ernie4_5_moe |
Collaborator
Author
|
run-slow: ernie4_5_moe |
Contributor
|
This comment contains models: ["models/ernie4_5_moe"] |
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.
What does this PR do?
The difference check was returning False 40% of the times and was reproducible locally
tested with
moved to a deterministic approach