[Backend Tester] Add tensor error statistic reporting #12809
Merged
Conversation
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
This was referenced Jul 24, 2025
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Jul 31, 2025
ghstack-source-id: 63819cb ghstack-comment-id: 3112003831 Pull-Request: pytorch#12809
This was referenced Aug 8, 2025
This was referenced Aug 12, 2025
digantdesai
reviewed
Aug 12, 2025
| actual = actual.to(torch.float64) | ||
| reference = reference.to(torch.float64) | ||
|
|
||
| if actual.shape != reference.shape: |
There was a problem hiding this comment.
Why is this not an error?
There was a problem hiding this comment.
There are some cases where in-place ops get functionalized and alter the graph outputs. This is an issue with the defunctionalization logic in ET (there's a separate issue I filed). It's not technically the backend's fault but is a real issue. I should probably just disable the affected tests for now and treat this as an error. I'll do that.
There was a problem hiding this comment.
Filed #13336 as a follow-up. Will stack this change on top.
GregoryComer
added
the
release notes: none
agrima1304
pushed a commit
to agrima1304/executorch
that referenced
this pull request
Aug 26, 2025
Report various error statistics for the test outputs, including SQNR, mean absolute error (MAE), and L2 norm. These are saved in the detail report per test case. As an example, here is the output from Core ML running MobileNet V2 (roughly formatted from csv -> sheets -> markdown): ``` Output 0 Error Max Output 0 Error MAE Output 0 Error MSD Output 0 Error L2 Output 0 SQNR 0.0005887411535 0.0001199183663 2.32E-06 0.004750485188 41.28595734 ```
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Report various error statistics for the test outputs, including SQNR, mean absolute error (MAE), and L2 norm. These are saved in the detail report per test case.
As an example, here is the output from Core ML running MobileNet V2 (roughly formatted from csv -> sheets -> markdown):