Multiple module outputs and multiple calls to backward #19799
Closed
Conversation
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
Differential Revision: D15097045 Differential Version: 80783859
pytorchbot
added
the
oncall: distributed
mrshenli
approved these changes
Apr 26, 2019
Contributor
|
This pull request has been merged in 0d8a361. |
facebook-github-bot
pushed a commit
that referenced
this pull request
Apr 29, 2019
Summary: Pull Request resolved: #19897 During validation, gradient reduction is not needed, and autograd is never called. The model output will always be a detached tensor. After the new reducer was merged, this meant that it would find all model parameters unused, and kick off reduction for them. When #19799 and output where no parameters are used and it tries to kick off reduction of zeroed gradients. Test for `torch.is_grad_enabled()` and `self.training` before calling into the reducer. Reviewed By: mrshenli Differential Revision: D15118726 fbshipit-source-id: b0208f632a61cbe8110fa626fa427937b7f05924
soumith
pushed a commit
that referenced
this pull request
Apr 29, 2019
Summary: Pull Request resolved: #19897 During validation, gradient reduction is not needed, and autograd is never called. The model output will always be a detached tensor. After the new reducer was merged, this meant that it would find all model parameters unused, and kick off reduction for them. When #19799 and output where no parameters are used and it tries to kick off reduction of zeroed gradients. Test for `torch.is_grad_enabled()` and `self.training` before calling into the reducer. Reviewed By: mrshenli Differential Revision: D15118726 fbshipit-source-id: b0208f632a61cbe8110fa626fa427937b7f05924
zhangguanheng66
pushed a commit
to zhangguanheng66/pytorch
that referenced
this pull request
May 6, 2019
Summary: Pull Request resolved: pytorch#19799 A module that returns multiple outputs and where the called may end up doing multiple calls to torch.autograd.backward did not work with DistributedDataParallel. It expected the first call to torch.autograd.backward to provide gradients for ALL parameters that expect gradients and were used in computing the module output. If you have outputs with disjoint autograd graphs it is fine to call torch.autograd.backward on both and fill in the module's parameter gradients in separate chunks. With this change we delay queuing the finalizer callback until we have marked all buckets as ready, instead of queueing it the first time we receive an autograd hook. This returns the current implementation to be functionally equivalent to the DistributedDataParallel implementation before pytorch#18953 was merged. Reviewed By: mrshenli Differential Revision: D15097045 fbshipit-source-id: 2df023319713bc31e29a8b45108c78e6593fccd4
zhangguanheng66
pushed a commit
to zhangguanheng66/pytorch
that referenced
this pull request
May 6, 2019
Summary: Pull Request resolved: pytorch#19897 During validation, gradient reduction is not needed, and autograd is never called. The model output will always be a detached tensor. After the new reducer was merged, this meant that it would find all model parameters unused, and kick off reduction for them. When pytorch#19799 and output where no parameters are used and it tries to kick off reduction of zeroed gradients. Test for `torch.is_grad_enabled()` and `self.training` before calling into the reducer. Reviewed By: mrshenli Differential Revision: D15118726 fbshipit-source-id: b0208f632a61cbe8110fa626fa427937b7f05924
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.
Stack:
:black_circle: #19799 Multiple module outputs and multiple calls to backward 💛
A module that returns multiple outputs and where the called may end up
doing multiple calls to torch.autograd.backward did not work with
DistributedDataParallel. It expected the first call to
torch.autograd.backward to provide gradients for ALL parameters that
expect gradients and were used in computing the module output. If you
have outputs with disjoint autograd graphs it is fine to call
torch.autograd.backward on both and fill in the module's parameter
gradients in separate chunks.
With this change we delay queuing the finalizer callback until we have
marked all buckets as ready, instead of queueing it the first time we
receive an autograd hook. This returns the current implementation to
be functionally equivalent to the DistributedDataParallel
implementation before #18953 was merged.
Differential Revision: D15097045