Skip to content

Conversation

@oscarandersson8218
Copy link
Collaborator

Summary

torch.allclose() will broadcast data if neccessary. This means that Tester.run_method_and_compare_outputs() may pass even though a delegate produced an output of wrong shape.

torch.allclose() will broadcast data if neccessary. This means that
Tester.run_method_and_compare_outputs() may pass even though a delegate
produced an output of wrong shape.

Signed-off-by: Oscar Andersson <[email protected]>
Change-Id: If62907e0fc69cc28b1f5ce5f8cfd6f31b1870a1e
@pytorch-bot
Copy link

pytorch-bot bot commented Jan 10, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7589

Note: Links to docs will display an error until the docs builds have been completed.

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

❌ 1 New Failure

As of commit 9399721 with merge base a059981 (image):

NEW FAILURE - The following job has failed:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jan 10, 2025
@oscarandersson8218 oscarandersson8218 added partner: arm For backend delegation, kernels, demo, etc. from the 3rd-party partner, Arm ciflow/trunk topic: not user facing labels Jan 10, 2025
Copy link
Contributor

@digantdesai digantdesai left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I hope this didn't unearth some hidden failures :p

for i in range(len(model_output)):
model = model_output[i]
ref = ref_output[i]
assert (
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good catch.

@digantdesai
Copy link
Contributor

TestConv1d.test_qs8_conv1d_batchnorm_seq looks flaky.

@digantdesai digantdesai merged commit 4192fec into pytorch:main Jan 10, 2025
106 of 109 checks passed
YIWENX14 pushed a commit that referenced this pull request Jan 28, 2025
Make Tester check shapes of ref and model output

torch.allclose() will broadcast data if neccessary. This means that
Tester.run_method_and_compare_outputs() may pass even though a delegate
produced an output of wrong shape.


Change-Id: If62907e0fc69cc28b1f5ce5f8cfd6f31b1870a1e

Signed-off-by: Oscar Andersson <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/trunk CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. partner: arm For backend delegation, kernels, demo, etc. from the 3rd-party partner, Arm topic: not user facing

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants