-
Notifications
You must be signed in to change notification settings - Fork 21.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[CPU] Bump test_complex_2d
thresholds for LBFGS on complex64
#126358
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/126358
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 3900c22 with merge base 636e799 (): FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
I reckon that one vectorises and the other one doesn't, hence the error, but would be good to confirm it. |
rtol=4.5e-5, | ||
atol=5e-5, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is the default tolerance for complex64? Souds like it should be at that level already?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
rtol of 1.3e-6 and atol of 1e-5
ref: https://pytorch.org/docs/stable/testing.html#torch.testing.assert_close
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ho right! Sorry I confused myself with halving the bitlength. This is expected to match fp32.
The general rule here is that for single op, we expect these to hold. If you test a bunch of ops chained one after the other, then we might have to increase the tolerance yes.
https://github.com/pytorch/pytorch/blob/main/test/test_optim.py#L518-L527 for why it's not bitwise equal |
@pytorchmergebot rebase |
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
Successfully rebased |
efdd003
to
3900c22
Compare
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: Not merging any PRs at the moment because there is a merge blocking https://github.com/pytorch/pytorch/labels/ci:%20sev issue open at: Details for Dev Infra teamRaised by workflow job |
@pytorchmergebot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Is this supposed to be bitwise identical? Wasn't sure how to interpret the comment but it seems to be giving mismatches like:
on Neoverse-N2 SBSA ARM CPUs.
cc @vincentqb @jbschlosser @albanD @janeyx99 @crcrpar @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @ezyang @anjali411 @dylanbespalko @mruberry @lezcano @nikitaved @amjames