zero_grad() between forward and backward passes generates error message #422
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
Summary: In case of Poisson sampling, we must check that all `grad_sample` attributes are set to `None` (since we forbit accumulation with Poisson sampling). This was previously done in a forward hook, resulting in an error message being thrown inadequately, in particular when `zero_grad()` is called between the forward and backward steps (see task for more context). Hence, we instead register a backward hook at the `GradSampleModule` level (to ensure it's the first hook to be called, before the hooks that populate the `grad_sample` attribute). Updated the docstring accordingly with more details. Differential Revision: D36067469 fbshipit-source-id: 583b1865e1d1e30b50f695fa3e7dc1dbb528233d
facebook-github-bot
added
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D36067469 |
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Outdated suggestions cannot be applied.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.
Summary:
In case of Poisson sampling, we must check that all
grad_sampleattributes are set toNone(since we forbit accumulation with Poisson sampling).This was previously done in a forward hook, resulting in an error message being thrown inadequately, in particular when
zero_grad()is called between the forward and backward steps (see task for more context).Hence, we instead register a backward hook at the
GradSampleModulelevel (to ensure it's the first hook to be called, before the hooks that populate thegrad_sampleattribute). Updated the docstring accordingly with more details.Differential Revision: D36067469