Ignore the `NoUtterancesError` when calculating pesq for a batch

veera-puthiran-14082 commented 2 months ago

What does this PR do?

Fixes #2752

Ignores the NoUtterancesError and create tensors with the available results, if a batch of size 32 is passed and 2 of them encountered an error, then for 30 pesq scores avg will be calculated and returned.
introduced a simple lambda function which filter out exception messages and keep the pesq values calculated.
since the np.array contained error messages and pesq scores, the result was in np.object dtype. it is converted to np.float32 since pesq is also calculated in float32

Checklist

[ ] Was this discussed/agreed via a Github issue? (no need for typos and docs improvements)
[x] Did you read the contributor guideline, Pull Request section?
[ ] Did you make sure to update the docs?
[ ] Did you write any new necessary tests?

PR review

Anyone in the community is free to review the PR once the tests have passed. If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

📚 Documentation preview 📚: https://torchmetrics--2753.org.readthedocs.build/en/2753/

codecov[bot] commented 2 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 35%. Comparing base (6226a53) to head (0831873). Report is 1 commits behind head on master.

:exclamation: There is a different number of reports uploaded between BASE (6226a53) and HEAD (0831873). Click for more details.

HEAD has 72 uploads less than BASE
| Flag | BASE (6226a53) | HEAD (0831873) | |------|------|------| |torch2.0.1+cpu|6|3| |python3.10|10|5| |Windows|5|3| |cpu|37|20| |torch1.11.0+cpu|2|1| |Linux|26|13| |python3.9|18|9| |torch1.12.1+cpu|2|1| |macOS|6|4| |torch2.0.1|4|2| |torch1.10.2+cpu|2|1| |torch2.4.0+cpu|2|1| |python3.11|5|3| |python3.8|4|3| |torch1.13.1+cpu|5|3| |torch2.3.1+cpu|4|2| |torch2.2.2+cpu|4|2| |torch2.4.0+cu121|2|1| |torch2.1.2+cpu|2|1| |gpu|2|0| |unittest|2|0|

Additional details and impacted files

```diff @@ Coverage Diff @@ ## master #2753 +/- ## ======================================== - Coverage 69% 35% -34% ======================================== Files 329 316 -13 Lines 18079 17919 -160 ======================================== - Hits 12501 6291 -6210 - Misses 5578 11628 +6050 ```

veera-puthiran-14082 commented 2 months ago

@Borda can this be merged ? Thanks in advance

Lightning-AI / torchmetrics