Skip to content

Fix AdamWeightDecay for TF 2.11 #20735

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Rocketknight1 merged 3 commits into from
Dec 13, 2022
Merged

Fix AdamWeightDecay for TF 2.11 #20735

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3710b42
Fix AdamWeightDecay for TF
Rocketknight1 Dec 12, 2022
c9d00b3
Fix AdamWeightDecay for TF
Rocketknight1 Dec 12, 2022
6bf735e
make fixup
Rocketknight1 Dec 12, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion src/transformers/optimization_tf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,12 @@
import tensorflow as tf


if hasattr(tf.keras, "optimizer") and hasattr(tf.keras.optimizer, "legacy"):
Adam = tf.keras.optimizer.legacy.Adam
else:
Adam = tf.keras.optimizers.Adam


class WarmUp(tf.keras.optimizers.schedules.LearningRateSchedule):
"""
Applies a warmup schedule on a given learning rate decay schedule.
Expand Down Expand Up @@ -163,7 +169,7 @@ def create_optimizer(
return optimizer, lr_schedule


class AdamWeightDecay(tf.keras.optimizers.Adam):
class AdamWeightDecay(Adam):
"""
Adam enables L2 weight decay and clip_by_global_norm on gradients. Just adding the square of the weights to the
loss function is *not* the correct way of using L2 regularization/weight decay with Adam, since that will interact
Expand Down