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][Easy] Allow ModuleWrapPolicy to take Iterable #104999

Closed
wants to merge 4 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/104999

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

✅ No Failures

As of commit 5c5cecb:
💚 Looks good so far! There are no failures yet. 💚

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

awgu added a commit to awgu/pytorch that referenced this pull request Jul 19, 2023
ghstack-source-id: 6e56f5c32da86336e4a943a49f237d448c532cad
Pull Request resolved: pytorch#104999
awgu added a commit to awgu/pytorch that referenced this pull request Jul 21, 2023
ghstack-source-id: 6e56f5c32da86336e4a943a49f237d448c532cad
Pull Request resolved: pytorch#104999
awgu added a commit to awgu/pytorch that referenced this pull request Jul 26, 2023
ghstack-source-id: 6e56f5c32da86336e4a943a49f237d448c532cad
Pull Request resolved: pytorch#104999
awgu added a commit to awgu/pytorch that referenced this pull request Jul 31, 2023
ghstack-source-id: 6e56f5c32da86336e4a943a49f237d448c532cad
Pull Request resolved: pytorch#104999
@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: d2f04431e8203e1e6c8106daa31b477b956d2572
Pull Request resolved: pytorch#104999
@awgu
Copy link
Contributor Author

awgu commented Aug 2, 2023

@pytorchbot merge

@pytorchmergebot
Copy link
Collaborator

Merge started

Your 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

Advanced Debugging
Check the merge workflow status
here

@pytorchmergebot
Copy link
Collaborator

Merge failed

Reason: 1 mandatory check(s) failed. The first few are:

Dig deeper by viewing the failures on hud

Details for Dev Infra team Raised by workflow job

Failing merge rule: Core Maintainers

@awgu
Copy link
Contributor Author

awgu commented Aug 2, 2023

@pytorchbot merge

@pytorchmergebot
Copy link
Collaborator

Merge started

Your 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

Advanced Debugging
Check the merge workflow status
here

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/421/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: improvements topic category
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants