[train/tune] Remove save_to_object/restore_from_object #38757
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
Signed-off-by: Kai Fricke <kai@anyscale.com>
justinvyu
approved these changes
Aug 22, 2023
added 3 commits
August 23, 2023 00:29
Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
krfricke
requested review from
sven1977,
gjoliver,
avnishn,
ArturNiederfahrenhorst,
smorad,
maxpumperla and
kouroshHakha
as code owners
arvind-chandra
pushed a commit
to lmco/ray
that referenced
this pull request
Aug 31, 2023
…8757) This PR removes `Trainable.save_to_object` and `Trainable.restore_from_object`. These methods were technically public (as they were part of a public class), but they were only used internally in practice. That's why we believe it's fine to remove them without a full deprecation cycle. The methods stem from a time when a central storage location was not guaranteed. Then, we needed to be able to ship checkpoints between nodes - e.g. from the head node to the worker nodes where the trial is executed, or between nodes, e.g. when PBT trials exploit another trial. These assumptions are now superseded by a central storage location. We require users to save checkpoints in a location which all nodes can access. We then only have to ship a thin metadata wrapper around, instead of the full serialized checkpoint data. For large checkpoints, this was always infeasible. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: e428265 <arvind.chandramouli@lmco.com>
vymao
pushed a commit
to vymao/ray
that referenced
this pull request
Oct 11, 2023
…8757) This PR removes `Trainable.save_to_object` and `Trainable.restore_from_object`. These methods were technically public (as they were part of a public class), but they were only used internally in practice. That's why we believe it's fine to remove them without a full deprecation cycle. The methods stem from a time when a central storage location was not guaranteed. Then, we needed to be able to ship checkpoints between nodes - e.g. from the head node to the worker nodes where the trial is executed, or between nodes, e.g. when PBT trials exploit another trial. These assumptions are now superseded by a central storage location. We require users to save checkpoints in a location which all nodes can access. We then only have to ship a thin metadata wrapper around, instead of the full serialized checkpoint data. For large checkpoints, this was always infeasible. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: Victor <vctr.y.m@example.com>
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.
Why are these changes needed?
This PR removes
Trainable.save_to_objectandTrainable.restore_from_object.These methods were technically public (as they were part of a public class), but they were only used internally in practice. That's why we believe it's fine to remove them without a full deprecation cycle.
The methods stem from a time when a central storage location was not guaranteed. Then, we needed to be able to ship checkpoints between nodes - e.g. from the head node to the worker nodes where the trial is executed, or between nodes, e.g. when PBT trials exploit another trial.
These assumptions are now superseded by a central storage location. We require users to save checkpoints in a location which all nodes can access. We then only have to ship a thin metadata wrapper around, instead of the full serialized checkpoint data. For large checkpoints, this was always infeasible.
Related issue number
Closes #38674
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.