Amp gradient accumulation example#36601
Closed
mcarilli wants to merge 5 commits intopytorch:masterfrom
Closed
Conversation
💊 Build failures summary and remediationsAs of commit 1af3707 (more details on the Dr. CI page):
🚧 1 upstream failure:These were probably caused by upstream breakages:
Extra GitHub checks
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker. This comment has been revised 14 times. |
ngimel
approved these changes
Apr 17, 2020
Contributor
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ngimel has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
Contributor
laurentdupin
pushed a commit
to laurentdupin/pytorch
that referenced
this pull request
Apr 24, 2026
Summary: Several people have asked me about proper Amp usage with gradient accumulation. In particular, it's [unclear to people](NVIDIA/apex#439 (comment)) that you should only call `scaler.unscale_()` (if desired) and `scaler.update()` in iterations where you actually plan to step. This PR adds a minimal accumulation example. I built the docs locally and it looks free from sphinx errors, at least. Pull Request resolved: pytorch#36601 Differential Revision: D21082295 Pulled By: ngimel fbshipit-source-id: b2faa6c02b9f7e1972618a0f1d5360a03f0450ac
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
6 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Several people have asked me about proper Amp usage with gradient accumulation. In particular, it's unclear to people that you should only call
scaler.unscale_()(if desired) andscaler.update()in iterations where you actually plan to step. This PR adds a minimal accumulation example.I built the docs locally and it looks free from sphinx errors, at least.