[DSD] Correctly handle shared parameters for optimizer state_dict#128685
Closed
fegin wants to merge 6 commits intogh/fegin/254/basefrom
Closed
[DSD] Correctly handle shared parameters for optimizer state_dict#128685fegin wants to merge 6 commits intogh/fegin/254/basefrom
fegin wants to merge 6 commits intogh/fegin/254/basefrom
Conversation
Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/128685
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 6 Unrelated FailuresAs of commit de9c20e with merge base 73ba432 ( NEW FAILURE - The following job has failed:
FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
BROKEN TRUNK - The following jobs 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. |
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D58573487 |
…te_dict" Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) cc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k LucasLLC MeetVadakkanchery mhorowitz [ghstack-poisoned]
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D58573487 |
…te_dict" Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) cc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k LucasLLC MeetVadakkanchery mhorowitz [ghstack-poisoned]
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D58573487 |
…te_dict" Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) cc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k LucasLLC MeetVadakkanchery mhorowitz [ghstack-poisoned]
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D58573487 |
fegin
added a commit
that referenced
this pull request
Jun 14, 2024
Pull Request resolved: #128685 Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. ghstack-source-id: 230252126 Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/)
…te_dict" * Fixes #128011 See the discussion in #128076 Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) cc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k LucasLLC MeetVadakkanchery mhorowitz [ghstack-poisoned]
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D58573487 |
…te_dict" * Fixes #128011 See the discussion in #128076 Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) cc mrshenli pritamdamania87 zhaojuanmao satgera gqchen aazzolini osalpekar jiayisuse H-Huang kwen2501 awgu penguinwu XilunWu wanchaol fduwjj wz337 tianyu-l wconstab yf225 chauhang d4l3k LucasLLC MeetVadakkanchery mhorowitz [ghstack-poisoned]
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D58573487 |
Contributor
|
@pytorchbot merge -f 'Landed internally' (Initiating merge automatically since Phabricator Diff has merged, using force because this PR might not pass merge_rules.json but landed internally) |
Collaborator
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
pytorchmergebot
pushed a commit
that referenced
this pull request
Jun 18, 2024
HSDP1 + broadcast_from_rank0 actually behaves differently from FSDP1 + broadcast_from_rank0. So we need an unittest to cover this use case. This test relies on the fix from #128446. Differential Revision: [D58621436](https://our.internmc.facebook.com/intern/diff/D58621436/) Pull Request resolved: #128755 Approved by: https://github.com/Skylion007, https://github.com/wz337 ghstack dependencies: #128685
fegin
added a commit
that referenced
this pull request
Jun 21, 2024
…28685) * Fixes #128011 See the discussion in #128076 Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) Pull Request resolved: #128685 Approved by: https://github.com/LucasLLC (cherry picked from commit 1a52791)
This was referenced Jun 21, 2024
fegin
added a commit
that referenced
this pull request
Jun 21, 2024
HSDP1 + broadcast_from_rank0 actually behaves differently from FSDP1 + broadcast_from_rank0. So we need an unittest to cover this use case. This test relies on the fix from #128446. Differential Revision: [D58621436](https://our.internmc.facebook.com/intern/diff/D58621436/) Pull Request resolved: #128755 Approved by: https://github.com/Skylion007, https://github.com/wz337 ghstack dependencies: #128685 (cherry picked from commit fe8558b)
atalman
pushed a commit
that referenced
this pull request
Jun 26, 2024
#129252) [DSD] Correctly handle shared parameters for optimizer state_dict (#128685) * Fixes #128011 See the discussion in #128076 Current implementation of `set_optimizer_state_dict()` assumes that all the fqns returned by `_get_fqns()` must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue. Differential Revision: [D58573487](https://our.internmc.facebook.com/intern/diff/D58573487/) Pull Request resolved: #128685 Approved by: https://github.com/LucasLLC (cherry picked from commit 1a52791)
atalman
pushed a commit
that referenced
this pull request
Jun 26, 2024
…129255) HSDP1 + broadcast_from_rank0 actually behaves differently from FSDP1 + broadcast_from_rank0. So we need an unittest to cover this use case. This test relies on the fix from #128446. Differential Revision: [D58621436](https://our.internmc.facebook.com/intern/diff/D58621436/) Pull Request resolved: #128755 Approved by: https://github.com/Skylion007, https://github.com/wz337 ghstack dependencies: #128685 (cherry picked from commit fe8558b)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
Fixes #128011
See the discussion in #128076
Current implementation of
set_optimizer_state_dict()assumes that all the fqns returned by_get_fqns()must exist in the optimizer state_dict. This is not true if the model has shared parameters. In such a case, only one fqn of the shared parameters will appear in the optimizer state_dict. This PR addresses the issue.Differential Revision: D58573487
cc @mrshenli @pritamdamania87 @zhaojuanmao @satgera @gqchen @aazzolini @osalpekar @jiayisuse @H-Huang @kwen2501 @awgu @penguinwu @XilunWu @wanchaol @fduwjj @wz337 @tianyu-l @wconstab @yf225 @chauhang @d4l3k @LucasLLC @MeetVadakkanchery @mhorowitz