Enforce saving at end of training if saving option chosen #30160
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
Closed
4 tasks
|
We can certainly make this configurable as an arg, but I'm not sure it makes sense to do this |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
amyeroberts
reviewed
Apr 18, 2024
pacman100
reviewed
Apr 29, 2024
There was a problem hiding this comment.
Thank you, @muellerzr, for enabling the model to be saved at the end of training when using --save_steps strategy. Left a nit.
|
@amyeroberts good for a rereview, no functional changes happened just switched out the test logic to be more obvious what's going on |
muellerzr
force-pushed
the
muellerzr-load-best-model
branch
from
cc684e5 to
9bbb268
Compare
muellerzr
force-pushed
the
muellerzr-load-best-model
branch
from
ef951dc to
4002cb8
Compare
pacman100
approved these changes
May 16, 2024
itazap
pushed a commit
that referenced
this pull request
May 21, 2024
* Enforce saving at end of training * Fix test * Rework test * Fixup tests' * Update comment based on sourab feedback * Clean
itazap
pushed a commit
that referenced
this pull request
May 21, 2024
* Enforce saving at end of training * Fix test * Rework test * Fixup tests' * Update comment based on sourab feedback * Clean
itazap
pushed a commit
that referenced
this pull request
May 22, 2024
* Enforce saving at end of training * Fix test * Rework test * Fixup tests' * Update comment based on sourab feedback * Clean
itazap
pushed a commit
that referenced
this pull request
May 24, 2024
* Enforce saving at end of training * Fix test * Rework test * Fixup tests' * Update comment based on sourab feedback * Clean
zucchini-nlp
pushed a commit
to zucchini-nlp/transformers
that referenced
this pull request
Jun 11, 2024
…e#30160) * Enforce saving at end of training * Fix test * Rework test * Fixup tests' * Update comment based on sourab feedback * Clean
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.
What does this PR do?
Currently we save at the end of training if an
epochstrategy is chosen, but not if astepsstrategy is chosen. This can be undesireable because the end of a users training might not be saved at all, sometimes upwards of 10% as seen here: #28539This PR mimics the behavior of
epoch, by choosing to save the model at the end of training as well ifstepsis chosen.I do not believe it makes sense to make this configurable, because a user should always have access to the end-resulting model when training!
Fixes #28539
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@amyeroberts @pacman100