Skip to content
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

[train] Storage: Call callbacks with new checkpoint #38456

Merged
merged 6 commits into from
Aug 17, 2023

Conversation

krfricke
Copy link
Contributor

@krfricke krfricke commented Aug 15, 2023

Why are these changes needed?

Adjust checkpoint interface to accept new-style checkpoint objects instead.

In progress reporter output:

Training saved a checkpoint for iteration 4 at: (s3)bucket/foo/bar

Related issue number

Checks

  • I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR.
  • I've run scripts/format.sh to lint the changes in this PR.
  • I've included any doc changes needed for https://docs.ray.io/en/master/.
    • I've added any new APIs to the API Reference. For example, if I added a
      method in Tune, I've added it in doc/source/tune/api/ under the
      corresponding .rst file.
  • I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
  • Testing Strategy
    • Unit tests
    • Release tests
    • This PR is not tested :(

Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
Copy link
Contributor

@justinvyu justinvyu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks! One suggestion to fix the test.

@@ -1074,7 +1074,7 @@ def clear_checkpoint(self):
self.temporary_state.restoring_from = None
self.run_metadata.invalidate_cache()

def on_checkpoint(self, checkpoint: _TrackedCheckpoint):
def on_checkpoint(self, checkpoint: Union[_TrackedCheckpoint, _TrainingResult]):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ahh I haven't been so on top of updating all of these 😄

python/ray/tune/syncer.py Show resolved Hide resolved
@ericl ericl added the @author-action-required The PR author is responsible for the next step. Remove tag to send back to the reviewer. label Aug 15, 2023
Kai Fricke added 2 commits August 16, 2023 11:56
Signed-off-by: Kai Fricke <kai@anyscale.com>
…checkpoint

# Conflicts:
#	python/ray/tune/execution/tune_controller.py
saved_iter = "?"
if trial.last_result and TRAINING_ITERATION in trial.last_result:
saved_iter = trial.last_result[TRAINING_ITERATION]

self._start_block(f"trial_{trial}_result_{saved_iter}")

if isinstance(checkpoint, Checkpoint):
loc = f"({checkpoint.filesystem.type_name}){checkpoint.path}"
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We currently have no way to restore the actual location URI, so this is the next best thing.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add an example of what this looks like in the pr description?

@krfricke krfricke removed the @author-action-required The PR author is responsible for the next step. Remove tag to send back to the reviewer. label Aug 16, 2023
@justinvyu justinvyu added the tests-ok The tagger certifies test failures are unrelated and assumes personal liability. label Aug 16, 2023
Copy link
Contributor

@justinvyu justinvyu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, lgtm. Only left some nits.

python/ray/tune/execution/tune_controller.py Outdated Show resolved Hide resolved
saved_iter = "?"
if trial.last_result and TRAINING_ITERATION in trial.last_result:
saved_iter = trial.last_result[TRAINING_ITERATION]

self._start_block(f"trial_{trial}_result_{saved_iter}")

if isinstance(checkpoint, Checkpoint):
loc = f"({checkpoint.filesystem.type_name}){checkpoint.path}"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add an example of what this looks like in the pr description?

@ericl ericl added the @author-action-required The PR author is responsible for the next step. Remove tag to send back to the reviewer. label Aug 16, 2023
@ericl
Copy link
Contributor

ericl commented Aug 16, 2023

Good to merge, but just in case you want to address the nits.

@ericl ericl removed the tests-ok The tagger certifies test failures are unrelated and assumes personal liability. label Aug 16, 2023
Kai Fricke added 2 commits August 17, 2023 11:05
@krfricke krfricke merged commit b73a496 into ray-project:master Aug 17, 2023
8 of 30 checks passed
@krfricke krfricke deleted the train/storage-on-checkpoint branch August 17, 2023 09:33
arvind-chandra pushed a commit to lmco/ray that referenced this pull request Aug 31, 2023
Adjust checkpoint interface to accept new-style checkpoint objects instead.

In progress reporter output:

```
Training saved a checkpoint for iteration 4 at: (s3)bucket/foo/bar
```

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
Adjust checkpoint interface to accept new-style checkpoint objects instead.

In progress reporter output:

```
Training saved a checkpoint for iteration 4 at: (s3)bucket/foo/bar
```

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
Labels
@author-action-required The PR author is responsible for the next step. Remove tag to send back to the reviewer.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants