-
Notifications
You must be signed in to change notification settings - Fork 25.6k
Min-cut partitioner always saves tensors that are returned as-is in backward #114970
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
…ackward [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114970
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit e468749 with merge base 45e7af5 ( BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
… as-is in backward" [ghstack-poisoned]
# x would be part of cut nodes. A case where this happens is if | ||
# NestedTensor saves a offset tensor as part of the singleton int | ||
# in sizes. | ||
nx_graph.add_edge(node.name + "_out", "sink", capacity=math.inf) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Alternatively we could just update the cut-set manually
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you add a test?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Added
… as-is in backward" [ghstack-poisoned]
@Chillee should review this one |
… as-is in backward" cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx chenyang78 aakhundov kadeng [ghstack-poisoned]
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What happens today with this? An error because it can't find a valid cut?
Basically yeah, when we go to extract the modules corresponding to the input/outputs, it would error out because one of outputs of the graph is not passed as in as input anywhere. |
@pytorchbot merge |
Merge failedReason: This PR needs a If not, please add the To add a label, you can comment to pytorchbot, for example For more information, see Details for Dev Infra teamRaised by workflow job |
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
…ackward (#114970) Pull Request resolved: #114970 Approved by: https://github.com/Chillee
…119708) Pull Request resolved: #119708 Approved by: https://github.com/albanD ghstack dependencies: #114970
Stack from ghstack (oldest at bottom):
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @aakhundov @kadeng