-
Notifications
You must be signed in to change notification settings - Fork 25.6k
Fix test_modules_can_be_imported #145387
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
Fix test_modules_can_be_imported #145387
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/145387
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 8dc9142 with merge base 40e27fb ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
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.
Sounds ok as a stopgap
We should discuss with @wconstab what to do with this partial dependency.
| "torch.onnx._internal.exporter._reporting", | ||
| "torch.onnx._internal.exporter._schemas", | ||
| "torch.onnx._internal.exporter._tensors", | ||
| "torch.onnx._internal.exporter._torchlib.ops", |
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.
@wconstab this is failing because this relies on the "pulp" package to run which is not a dependency. Why do we depend on this to even import the files / any details where I can read about why we are making this a dependency?
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.
It's for the file below in torch.distributed._tools.sac_ilp which was built as an optional tool for memory estimation for FSDP and uses pulp as an ILP solver. But without pulp the tool doesn't work
| from pulp import ( # type: ignore[import-untyped,import-not-found] |
|
@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 |
Make torchelastic publicly importable by raising error on import etcd lazily, [BE task, row 7](https://docs.google.com/spreadsheets/d/1TtATnLJf1rVXaBQd3X3yYqm9xNN9BIWG7QqRgrFiRRI/edit?gid=1748512924#gid=1748512924) Pull Request resolved: #145396 Approved by: https://github.com/albanD ghstack dependencies: #145387
Stack from ghstack (oldest at bottom):
test_modules_can_be_importedtest is currently failing due to a few missing private modules and this PR gets it working before I start to clean up the public allow list