backwards compatibility w. v020 ckpts, fix issue with zero-1 ckpts #543
Conversation
This file contains 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
Collaborator
jeffra
commented
Nov 19, 2020
jeffra
commented
- Checkpoints from versions <= v0.2.0 serialized a class from deepspeed.pt.loss_scaler, was seeing an exception on v0.3+ code. Added patch to direct deepspeed.pt.loss_scaler to its new home.
- Discovered issue with latest zero-1 checkpoint changes related to elastic loading checkpoints and max-elems-per-comm not aligning between src/dst partitions. We don't actually have to keep the 'legacy' max-elems-per-comm value at all.
jeffra
requested review from
arashashari,
awan-10,
cli99,
conglongli,
eltonzheng,
minjiaz,
niumanar,
RezaYazdaniAminabadi,
samyam,
ShadenSmith and
tjruwase
as code owners
tjruwase
approved these changes
Nov 19, 2020
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Outdated suggestions cannot be applied.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.