-
Notifications
You must be signed in to change notification settings - Fork 3.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add support for reloading the last checkpoint saved by passing ckpt_path="last"
#12816
Merged
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
otaj
added
feature
Is an improvement or enhancement
checkpointing
Related to checkpointing
fault tolerance
callback: model checkpoint
labels
Apr 20, 2022
otaj
requested review from
williamFalcon,
tchaton,
awaelchli,
edenlightning,
Borda,
SeanNaren and
justusschock
as code owners
April 22, 2022 07:55
carmocca
approved these changes
Apr 28, 2022
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM! I've pushed some minor simplification to the code and greater simplifications to the test to narrow their scope to just these warnings.
carmocca
changed the title
Feature/last checkpoint
Add support for reloading the last checkpoint saved by passing Apr 28, 2022
ckpt_path="last"
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
otaj
commented
Apr 28, 2022
rohitgr7
approved these changes
May 4, 2022
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
great!
awaelchli
approved these changes
May 4, 2022
kaushikb11
approved these changes
May 5, 2022
12 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
callback: model checkpoint
checkpointing
Related to checkpointing
fault tolerance
feature
Is an improvement or enhancement
ready
PRs ready to be merged
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.
What does this PR do?
Implements #11912
Allows you to pass keyword
"last"
tockpt_path
argument."best"
still remains the default value ifNone
is passed, except one particular case - in methodfit
when fault tolerant training is enabled.Value
last
selects the checkpoint with the latest modified timestamp from set of all last checkpoints (currently only one, because they overwrite each other, but once #5030 is solved, it could be a different one) and all fault tolerant checkpoints. If no checkpoint in these two sets is found (due to fault tolerant training not being enabled and passingsave_last=False
toModelCheckpoint
), warning is issued and no checkpoint is loaded.What needs to be done before making it non-draft
Does your PR introduce any breaking changes? If yes, please list them.
Before submitting
PR review
Anyone in the community is welcome to review the PR.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:
Did you have fun?
Make sure you had fun coding 🙃
cc @Borda @awaelchli @ananthsub @ninginthecloud @rohitgr7 @carmocca @justusschock @jjenniferdai