[FSDP] Consolidate FSDP state_dict offload_to_cpu settings #86211

fegin · 2022-10-04T16:29:18Z

Stack from ghstack (oldest at bottom):

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now
have offload_to_cpu options.

Differential Revision: D40065969

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/) [ghstack-poisoned]

linux-foundation-easycla · 2022-10-04T16:29:21Z

The committers listed above are authorized under a signed CLA.

✅ login: fegin / name: Chien-Chin Huang (97498b4)

pytorch-bot · 2022-10-04T16:29:22Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/86211

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit our office hours

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

✅ No Failures

As of commit 0798b95:
💚 Looks good so far! There are no failures yet. 💚

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

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/) ghstack-source-id: 169306202 Pull Request resolved: #86211

rohan-varma

LGTM, thanks! Do we need to update the appropriate documentation anywhere?

rohan-varma · 2022-10-04T23:57:16Z

test/distributed/fsdp/test_fsdp_state_dict.py

-_UNFLATTENED_STATE_DICT_IMPLS = ["state_dict", "sharded_state_dict"]
-_FLATTENED_STATE_DICT_IMPLS = ["local_state_dict"]
+_UNFLATTENED_STATE_DICT_IMPLS = ["state_dict"]
+_FLATTENED_STATE_DICT_IMPLS = []


why is local_state_dict out of this flattened state_dict impls list? Isn't still a flattened version of state_dict?

oops, accidentally commit this change. Will revert it.

awgu · 2022-10-05T13:57:36Z

test/distributed/fsdp/test_fsdp_state_dict.py

        else:
-            config = None
+            raise ValueError("Unspported state_dict_type")


nit:

Suggested change

raise ValueError("Unspported state_dict_type")

raise ValueError("Unsupported state_dict_type")

test/distributed/fsdp/test_fsdp_state_dict.py

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/) [ghstack-poisoned]

Pull Request resolved: #86211 Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. ghstack-source-id: 169442813 Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/)

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/) [ghstack-poisoned]

Pull Request resolved: #86211 Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. ghstack-source-id: 170025910 Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/)

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/) [ghstack-poisoned]

Pull Request resolved: #86211 Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. ghstack-source-id: 170308168 Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/)

fegin · 2022-10-14T16:21:58Z

@pytorchbot merge

pytorchmergebot · 2022-10-14T16:23:23Z

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

github-actions · 2022-10-14T16:24:10Z

Hey @fegin.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

[FSDP] Consolidate FSDP state_dict offload_to_cpu settings

97498b4

Consolidate FSDP state_dict offload_to_cpu settings. All state_dict_types now have offload_to_cpu options. Differential Revision: [D40065969](https://our.internmc.facebook.com/intern/diff/D40065969/) [ghstack-poisoned]

fegin requested review from mrshenli, pritamdamania87, zhaojuanmao, rohan-varma, H-Huang, awgu, kwen2501 and mingzhe09088 as code owners October 4, 2022 16:29

pytorch-bot bot added the release notes: distributed (fsdp) release notes category label Oct 4, 2022

facebook-github-bot added cla signed oncall: distributed Add this issue/PR to distributed oncall triage queue labels Oct 4, 2022

fegin mentioned this pull request Oct 4, 2022

[FSDP] Add set_state_dict_type API to setup state_dict_type without using context manager #86243

Closed

rohan-varma approved these changes Oct 5, 2022

View reviewed changes

pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Oct 5, 2022

awgu reviewed Oct 5, 2022

View reviewed changes

fegin mentioned this pull request Oct 13, 2022

[FSDP] Introduce optim_state_dict() and load_optim_state_dict() #86912

Closed

pytorchmergebot added the Merged label Oct 14, 2022

pytorchmergebot closed this in a86278b Oct 14, 2022

facebook-github-bot deleted the gh/fegin/30/head branch June 8, 2023 17:15

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[FSDP] Consolidate FSDP state_dict offload_to_cpu settings #86211

[FSDP] Consolidate FSDP state_dict offload_to_cpu settings #86211

fegin commented Oct 4, 2022 •

edited

linux-foundation-easycla bot commented Oct 4, 2022 •

edited

pytorch-bot bot commented Oct 4, 2022 •

edited

rohan-varma left a comment

rohan-varma Oct 4, 2022

fegin Oct 5, 2022 •

edited

awgu Oct 5, 2022

fegin commented Oct 14, 2022

pytorchmergebot commented Oct 14, 2022

github-actions bot commented Oct 14, 2022

	raise ValueError("Unspported state_dict_type")
	raise ValueError("Unsupported state_dict_type")

[FSDP] Consolidate FSDP state_dict offload_to_cpu settings #86211

[FSDP] Consolidate FSDP state_dict offload_to_cpu settings #86211

Conversation

fegin commented Oct 4, 2022 • edited

linux-foundation-easycla bot commented Oct 4, 2022 • edited

pytorch-bot bot commented Oct 4, 2022 • edited

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/86211

✅ No Failures

rohan-varma left a comment

Choose a reason for hiding this comment

rohan-varma Oct 4, 2022

Choose a reason for hiding this comment

fegin Oct 5, 2022 • edited

Choose a reason for hiding this comment

awgu Oct 5, 2022

Choose a reason for hiding this comment

fegin commented Oct 14, 2022

pytorchmergebot commented Oct 14, 2022

Merge started

github-actions bot commented Oct 14, 2022

fegin commented Oct 4, 2022 •

edited

linux-foundation-easycla bot commented Oct 4, 2022 •

edited

pytorch-bot bot commented Oct 4, 2022 •

edited

fegin Oct 5, 2022 •

edited