Remove leaky custom_gradient wrapper for softplus (fixes #2008)#2022
Open
pranithreddym wants to merge 2 commits into
Open
Remove leaky custom_gradient wrapper for softplus (fixes #2008)#2022pranithreddym wants to merge 2 commits into
pranithreddym wants to merge 2 commits into
Conversation
) The _stable_grad_softplus custom_gradient closure captured the input tensor x, which ended up in TF's gradient registry (never cleared), causing a memory leak on every call. The same issue affected sigmoid.py which had an identical copy of the wrapper. tf.nn.softplus has been numerically stable since 2019-2020: - Uses log1p via tensorflow/tensorflow@ce92fd6ae5 - Gradient uses sigmoid via tensorflow/tensorflow@862d326bf The JAX mode already used tf.nn.softplus directly; this change makes the non-JAX path consistent and removes the TODO(b/155501444) workaround. Affected files: - bijectors/softplus.py: replace if/else block with _stable_grad_softplus = tf.nn.softplus - bijectors/sigmoid.py: same removal of the duplicate custom_gradient wrapper
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The _stable_grad_softplus custom_gradient closure captured the input tensor x, which ended up in TF's gradient registry (never cleared), causing a memory leak on every call. The same issue affected sigmoid.py which had an identical copy of the wrapper.
tf.nn.softplus has been numerically stable since 2019-2020:
The JAX mode already used tf.nn.softplus directly; this change makes the non-JAX path consistent and removes the TODO(b/155501444) workaround.
Affected files: