Skip to content
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

[FSDP][7/N] Add warning about frozen params #104967

Closed
wants to merge 6 commits into from

Conversation

@pytorch-bot
Copy link

pytorch-bot bot commented Jul 11, 2023

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/104967

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 1 Unrelated Failure

As of commit a1c151e:

NEW FAILURE - The following job has failed:

UNSTABLE - The following job failed but was likely due to flakiness present on trunk and has been marked as unstable:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@pytorch-bot pytorch-bot bot added the release notes: distributed (sharded) release notes category label Jul 11, 2023
@awgu awgu added release notes: distributed (fsdp) release notes category topic: docs topic category and removed release notes: distributed (sharded) release notes category labels Jul 11, 2023
awgu added a commit to awgu/pytorch that referenced this pull request Jul 11, 2023
ghstack-source-id: 1831d65cb9d573ff26fa9d71e63807b051ad97c8
Pull Request resolved: pytorch#104967
Copy link
Member

@rohan-varma rohan-varma left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@awgu awgu added the ciflow/trunk Trigger trunk jobs on your pull request label Aug 2, 2023
awgu added a commit to awgu/pytorch that referenced this pull request Aug 2, 2023
ghstack-source-id: 4a472d3c230cfd6ecac32231989aeacfbc47ceee
Pull Request resolved: pytorch#104967
@awgu
Copy link
Contributor Author

awgu commented Aug 2, 2023

Failure looks unrelated:

ModuleNotFoundError: No module named 'torch.version'

linux-bionic-py3_8-clang8-xla / test (xla, 1, 1, linux.12xlarge)

@awgu
Copy link
Contributor Author

awgu commented Aug 2, 2023

@pytorchbot merge -f "unrelated failures"

@pytorchmergebot
Copy link
Collaborator

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

pytorchmergebot pushed a commit that referenced this pull request Aug 2, 2023
pytorchmergebot pushed a commit that referenced this pull request Aug 3, 2023
…104969)

This does some code organization improvement.
- It renames `_FSDPPolicy` to `_Policy` to show that it is not only for FSDP but for any module-level API.
- It formalizes the contract that such a policy should return something like `target_module_to_kwargs: Dict[nn.Module, Dict[str, Any]]` that maps each module to wrap to its kwargs. It does so by requiring a `_run_policy` abstract method (this time private since users do not need to care about it). Then, our auto wrapping can just call `_run_policy()` to generate the dict and do any validation or post-processing.

This PR is technically BC-breaking because it removes the public `ModuleWrapPolicy.policy`. However, I do not think anyone was using that anyway, so this is a pretty safe breakage.

Pull Request resolved: #104969
Approved by: https://github.com/rohan-varma
ghstack dependencies: #104427, #104967, #104999
pytorchmergebot pushed a commit that referenced this pull request Aug 3, 2023
This PR adds a new `CustomPolicy` that acts like the existing `lambda_auto_wrap_policy` except it (1) leverages the new auto wrapping infrastructure and (2) allows overriding FSDP kwargs for particular instances. (1) gives it access to the validation checks (like for frozen parameters), and (2) makes it as expressive as manual wrapping. This should allow us to effectively deprecate manual wrapping if desired.

The API is as follows:
```
def lambda_fn(module: nn.Module) -> Union[bool, Dict[str, Any]]:
    ...
policy = CustomPolicy(lambda_fn)
```
The `lambda_fn` can return:
- `False` or `{}` to indicate no wrapping
- `True` to indicate wrapping while inheriting the root's FSDP kwargs
- Non-empty `dict` to indicate wrapping while overriding the specified FSDP kwargs and inheriting the rest from the root

---

After this PR, the follow-up work items for auto wrapping are:
1. Add shared parameter validation
2. (Longer-term / exploratory) Add a policy that provides a reasonable auto wrapping with "minimal" user input

Pull Request resolved: #104986
Approved by: https://github.com/ezyang
ghstack dependencies: #104427, #104967, #104999, #104969
@facebook-github-bot facebook-github-bot deleted the gh/awgu/418/head branch August 6, 2023 14:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ciflow/trunk Trigger trunk jobs on your pull request Merged release notes: distributed (fsdp) release notes category topic: docs topic category
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants