Feature add ZeRO support to Checkpoint in a distributed configuration #2642
Conversation
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
sadra-barikbin
force-pushed
the
feature-add-ZeRO-support-to-checkpoint-issue-2623
branch
from
40eadb4 to
ecda83e
Compare
sadra-barikbin
force-pushed
the
feature-add-ZeRO-support-to-checkpoint-issue-2623
branch
from
ecda83e to
5c8be79
Compare
vfdev-5
reviewed
Aug 16, 2022
vfdev-5
reviewed
Aug 16, 2022
vfdev-5
reviewed
Aug 16, 2022
There was a problem hiding this comment.
@sadra-barikbin thanks for the update, I left few comments to address
sadra-barikbin
force-pushed
the
feature-add-ZeRO-support-to-checkpoint-issue-2623
branch
from
5c8be79 to
f239327
Compare
|
@sadra-barikbin GPU tests are broken : https://github.com/pytorch/ignite/runs/7914418582?check_suite_focus=true |
vfdev-5
pushed a commit
to vfdev-5/ignite
that referenced
this pull request
Aug 22, 2022
…pytorch#2642) * Implement feature * Fix bug in docstring * Fix bugs and tests * Handle pytorch<1.9.0 * Fix mypy error
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Fixes #2623
Description: To call
consolidate_state_dictmethod of ZeRO optimizer on all ranks to before callingstate_dicton target rank, the one responsible for saving the checkpoint.Check list: