Changes batch norm momentum defaults to 1-momentum #2100
Merged
Conversation
This file contains 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
dantreiman
changed the title
justinxzhao
approved these changes
Jun 6, 2022
w4nderlust
reviewed
Jun 6, 2022
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Outdated suggestions cannot be applied.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.
tensorflow.keras Batch Norm momentum is parameterized as 1 - torch Batch Norm momentum:
tf.keras.layers.BatchNormalization: (default momentum: 0.99)
moving_mean = moving_mean * momentum + mean(batch) * (1 - momentum)torch.nn.BatchNorm1d: (default momentum: 0.1)
x_hat = (1 - momentum) * x_hat + momentum * x_t,where
x_hatis the estimated statistic andx_tis the new observed value.Personally I agree with Tensorflow's definition, nevertheless since we heavily reference the torch docs I do think we should change defaults to match torch.
Going deeper: https://discuss.pytorch.org/t/batch-norm-momentum-default-value/11955