[Flash Checkpoint] Integrated flash checkpoint #155
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
aoyulong
changed the title
Caozhou1995
force-pushed
the
adaptive_flash_ckpt
branch
from
d913e4e
to
60bec48
Compare
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Outdated suggestions cannot be applied.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.
This PR integrated flash checkpoint, referring to dlrover.
Dlrover's core capabilities are invoked, but FlagScale customizes file paths, data organization, and so on.
To use the save and load functions of flash checkpoint, see the following example:
from flagscale.train.checkpointing import save_checkpointfrom flagscale.train.checkpointing import load_checkpointInterfaces are compatible with megatron and dlrover, the results are shown below:
save
load
If you don't want to use flash checkpoint, just set
flash=False, the save and load functions provided by megatron are used.