-
Notifications
You must be signed in to change notification settings - Fork 6.1k
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
[rllib] Fix storage-path related tests #38947
Conversation
Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: matthewdeng <matt@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
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.
thanks for the fix!
rllib/algorithms/algorithm.py
Outdated
@@ -262,7 +263,7 @@ class Algorithm(Trainable, AlgorithmBase): | |||
|
|||
@staticmethod | |||
def from_checkpoint( | |||
checkpoint: Union[str, Checkpoint, NewCheckpoint], | |||
checkpoint: Union[str, Checkpoint, NewCheckpoint, _TrainingResult], |
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.
can we change the places where it's using the direct output of save to instead take in the path? I feel like Algorithm.restore
is the only thing that should take in a training result, since it's paired with save
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.
Makes sense, updated
rllib/train.py
Outdated
if trial.checkpoint.path: | ||
checkpoints.append(trial.checkpoint.path) |
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.
if trial.checkpoint.path: | |
checkpoints.append(trial.checkpoint.path) | |
if trial.checkpoint: | |
checkpoints.append(trial.checkpoint) |
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.
all we use this for is a print later on. nicer to include the fs in the string
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.
Also not sure if this is necessary, but this suggestion would technically allow this to be backwards compatible (though I'm not sure what the output representation ends up looking like)
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.
Updated
Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
Leaving a reminder here to re-enable the new persistence path for relevant tests when merging master, since I disabled the flag in #38965 |
Signed-off-by: Kai Fricke <kai@anyscale.com>
tune.Tuner( | ||
tune.with_resources( | ||
experiment, ppo.PPO.default_resource_request(config) | ||
wrapped_experiment, ppo.PPO.default_resource_request(config) |
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.
Btw, I have no idea why this test started failing in this PR.
Removed the flag in this PR, ptal |
Signed-off-by: Kai Fricke <kai@anyscale.com>
Signed-off-by: Kai Fricke <kai@anyscale.com>
# Conflicts: # python/ray/tune/tests/test_api.py
Signed-off-by: Kai Fricke <kai@anyscale.com>
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, just one question: Did we change/fix the RLlib docs occurrences, where algo.save()
is used (and assumed to return a path as a str)?
Good point. I've just taken a look and there's at least one occurrence where we haven't. Will update now! |
Signed-off-by: Kai Fricke <kai@anyscale.com>
This PR fixes rllib-related tests that didn't pass changes related to the new storage context. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com>
* [train] enable new persistence mode for core and serve tests (#38938) Signed-off-by: Matthew Deng <matt@anyscale.com> * [train] New persistence mode: Update 🐠 `ML Libraries w/ Ray Client Examples (Python 3.7)` (#38923) Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [train] remove non-URI assertion (#38944) Signed-off-by: Matthew Deng <matt@anyscale.com> * [train] New persistence mode: Update 📖 `Doc tests and examples (excluding Ray AIR examples)` (#38940) Signed-off-by: Justin Yu <justinvyu@anyscale.com> Signed-off-by: Matthew Deng <matt@anyscale.com> Co-authored-by: Matthew Deng <matt@anyscale.com> * disable legacy sync config logic in trainable (#38952) Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [2.7 CI][New Persistent Mode][6/n] 📖✈️ Ray AIR examples (#38918) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> * [2.7 CI][New Persistent Mode][2/n] 📺 📖 Doc GPU tests and examples (#38905) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> * [2.7 CI][New Persistent Mode][4/n] 📺 🚂 Train GPU tests & 🚂 Datasets Train Integration GPU Tests and Examples (#38910) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> Signed-off-by: Justin Yu <justinvyu@anyscale.com> Co-authored-by: Justin Yu <justinvyu@anyscale.com> * [2.7 CI][New Persistent Mode][1/n] 📺✈️ AIR GPU tests (ray/air) & ⚡ :python: Lightning 2.0 Train GPU tests (#38903) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> Signed-off-by: Yunxuan Xiao <xiaoyunxuan1998@gmail.com> * [train] Fix broken tune tests and support ray storage (#38950) This PR re-introduces support for ray storage ray.init(storage="s3://...") and fixes a broken tune controller test. Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [train] New persistence mode: Finish migrating `xgb`, `lgbm` and `sklearn` trainers, checkpoints + tests (#38959) Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [2.7 CI][New Persistent Mode][5/n] 📖 Doc examples for external code (#38915) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> * [train][rllib] temporarily disable new persistence mode for rllib tests (#38965) Signed-off-by: Matthew Deng <matt@anyscale.com> * [2.7 CI][New Persistent Mode][8/n]✈️ AIR tests (ray/air) (#38932) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> * [tune] Storage: 🐙 🧠 Tune tests and examples {using RLlib} migration (#38895) Signed-off-by: Kai Fricke <kai@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com> * [train] Fix MosaicTrainer example and unit test (#38970) Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [air/release] Fix dreambooth example image preprocessing logic (#39020) Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [train] clean up ray.train._checkpoint imports (#38951) Signed-off-by: Matthew Deng <matt@anyscale.com> * [train] high level cleanup of Ray Train docs (#38971) Signed-off-by: Matthew Deng <matt@anyscale.com> * [wip][docs] update FrameworkPredictor examples (#38634) Signed-off-by: Matthew Deng <matt@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> * [train] Add documentation for using metadata argument to save preprocessors (#38701) * [Train] Restructure Ray Train Example Page (#38814) Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> * [air] Deprecate some fields/classes that are supposed to be gone in 2.6. (#38794) Signed-off-by: xwjiang2010 <xwjiang2010@gmail.com> * [tune/storage] Fix Tune multinode tests (#39050) Fixes multinode tests by using the new train.report() API. Signed-off-by: Kai Fricke <kai@anyscale.com> * [tune] Fix BOHB example for new storage (#38983) The new storage path does not create "empty" checkpoints per default anymore. Previously, when no checkpoint is saved, PAUSEing a trial would create a dummy checkpoint that only contains trial metadata (such as the iteration number). This is not the case anymore. Examples now have to implement checkpointing to properly restore previous state. This was also true previously - but some of our simple examples (e.g. the one in this PR) didn't implement it and still "worked". I think it's fine to keep the functionality as is and require our examples to show checkpointing implementations. This will ensure that users don't shoot their feet trying to use e.g. BOHB. Separately, BOHB was malfunctioning as trials were repeatedly PAUSED and restarted as they've never been removed from `bracket.trials_to_unpause`. @justinvyu mentioned this in the review where it was introduced and I believed at the time it wasn't necessary - turns out it is, as we can end up in a situation where a bracket is never finished because trials are constantly running. This was not caught by any tests. We should add one in a follow-up - for now we can proceed with this PR to pick onto Ray 2.7. Signed-off-by: Kai Fricke <kai@anyscale.com> * [Release Test] Fix `long_running_horovod_tune_test`. (#39012) Signed-off-by: Yunxuan Xiao <yunxuanx@anyscale.com> Signed-off-by: Yunxuan Xiao <xiaoyunxuan1998@gmail.com> * [train] New persistence mode: `StorageContext` unit tests (#39023) Signed-off-by: Justin Yu <justinvyu@anyscale.com> * [train] enable train + tune tests and examples (#39021) Signed-off-by: Matthew Deng <matt@anyscale.com> * [rllib] Fix storage-path related tests (#38947) This PR fixes rllib-related tests that didn't pass changes related to the new storage context. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com> * [train] New persistence mode: Migrate 🐙 `Tune tests and examples (medium)` (#39081) Signed-off-by: Justin Yu <justinvyu@anyscale.com> --------- Signed-off-by: Matthew Deng <matt@anyscale.com> Signed-off-by: Justin Yu <justinvyu@anyscale.com> Signed-off-by: woshiyyya <xiaoyunxuan1998@gmail.com> Signed-off-by: Yunxuan Xiao <xiaoyunxuan1998@gmail.com> Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Signed-off-by: xwjiang2010 <xwjiang2010@gmail.com> Signed-off-by: Yunxuan Xiao <yunxuanx@anyscale.com> Co-authored-by: Justin Yu <justinvyu@anyscale.com> Co-authored-by: Yunxuan Xiao <yunxuanx@anyscale.com> Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com> Co-authored-by: Eric Liang <ekhliang@gmail.com> Co-authored-by: xwjiang2010 <87673679+xwjiang2010@users.noreply.github.com>
This PR fixes rllib-related tests that didn't pass changes related to the new storage context. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com> Signed-off-by: e428265 <arvind.chandramouli@lmco.com>
This PR fixes rllib-related tests that didn't pass changes related to the new storage context. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com>
This PR fixes rllib-related tests that didn't pass changes related to the new storage context. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com> Signed-off-by: Jim Thompson <jimthompson5802@gmail.com>
This PR fixes rllib-related tests that didn't pass changes related to the new storage context. Signed-off-by: Kai Fricke <kai@anyscale.com> Signed-off-by: matthewdeng <matt@anyscale.com> Co-authored-by: matthewdeng <matt@anyscale.com> Signed-off-by: Victor <vctr.y.m@example.com>
Why are these changes needed?
This PR fixes rllib-related tests that didn't pass changes related to the new storage context.
Related issue number
Checks
git commit -s
) in this PR.scripts/format.sh
to lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/
under thecorresponding
.rst
file.