fgnt / padertorch

A collection of common functionality to simplify the design, training and evaluation of machine learning models based on pytorch with an emphasis on speech processing.
MIT License
71 stars 16 forks source link

minor changes and add freeze_stats argument in normalization module … #141

Closed JanekEbb closed 1 year ago

JanekEbb commented 1 year ago

… allowing to freeze norm params (scale and bias) without freezing running statistics

codecov-commenter commented 1 year ago

Codecov Report

Base: 77.13% // Head: 77.12% // Decreases project coverage by -0.01% :warning:

Coverage data is based on head (337fe2e) compared to base (c931be2). Patch coverage: 58.33% of modified lines in pull request are covered.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #141 +/- ## ========================================== - Coverage 77.13% 77.12% -0.02% ========================================== Files 46 46 Lines 3727 3729 +2 ========================================== + Hits 2875 2876 +1 - Misses 852 853 +1 ``` | [Impacted Files](https://codecov.io/gh/fgnt/padertorch/pull/141?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=None) | Coverage Δ | | |---|---|---| | [padertorch/modules/normalization.py](https://codecov.io/gh/fgnt/padertorch/pull/141/diff?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=None#diff-cGFkZXJ0b3JjaC9tb2R1bGVzL25vcm1hbGl6YXRpb24ucHk=) | `65.84% <58.33%> (-0.16%)` | :arrow_down: | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=None). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=None)

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.