[DDP] Remove train call to module copies #58595
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
No longer needed since this list is always of size 1. Differential Revision: [D28548426](https://our.internmc.facebook.com/intern/diff/D28548426/) [ghstack-poisoned]
rohan-varma
requested review from
H-Huang,
mingzhe09088,
mrshenli,
pritamdamania87,
wayi1 and
zhaojuanmao
as code owners
💊 CI failures summary and remediationsAs of commit 2e0f63d (more details on the Dr. CI page):
🕵️ 1 new failure recognized by patternsThe following CI failures do not appear to be due to upstream breakages:
|
facebook-github-bot
added
oncall: distributed
This was referenced May 19, 2021
rohan-varma
added a commit
that referenced
this pull request
May 19, 2021
No longer needed since this list is always of size 1. Differential Revision: [D28548426](https://our.internmc.facebook.com/intern/diff/D28548426/) ghstack-source-id: 129383153 Pull Request resolved: #58595
zhaojuanmao
reviewed
May 19, 2021
| def train(self, mode=True): | ||
| super(DistributedDataParallel, self).train(mode) | ||
| for module in self._module_copies[1:]: | ||
| module.train(mode) |
There was a problem hiding this comment.
we still need to keep train(mode) call?
There was a problem hiding this comment.
We have the call to the parent class method on L894, which does train(mode) on the module for this process. this call is for the rest of the module copies on this process, but there are none (see https://github.com/pytorch/pytorch/blob/master/torch/nn/parallel/distributed.py#L492)
zhaojuanmao
approved these changes
May 19, 2021
No longer needed since this list is always of size 1. Differential Revision: [D28548426](https://our.internmc.facebook.com/intern/diff/D28548426/) [ghstack-poisoned]
No longer needed since this list is always of size 1. Differential Revision: [D28548426](https://our.internmc.facebook.com/intern/diff/D28548426/) [ghstack-poisoned]
|
This pull request has been merged in 1d67c6d. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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 from ghstack:
No longer needed since this list is always of size 1.
Differential Revision: D28548426