[RLlib] Fix error in PPO with use_kl_loss=False
#45031
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
…kl_loss=True' and otherwise throws an error. Furthermore, removed the 'kl_sampled_values' from 'PPOLeanrer.additional_update_for_module' b/c it is unused. Signed-off-by: Simon Zehnder <simon.zehnder@gmail.com>
Signed-off-by: Simon Zehnder <simon.zehnder@gmail.com>
Signed-off-by: Simon Zehnder <simon.zehnder@gmail.com>
sven1977
reviewed
Apr 30, 2024
| ) | ||
|
|
||
| # Update KL coefficient. | ||
| if config.use_kl_loss: | ||
| assert sampled_kl_values, "Sampled KL values are empty." |
sven1977
reviewed
Apr 30, 2024
| @@ -22,6 +22,7 @@ | |||
| lr=0.0003, | |||
| num_sgd_iter=6, | |||
| vf_loss_coeff=0.01, | |||
| use_kl_loss=True, | |||
There was a problem hiding this comment.
True by default, but yes, setting this explicitly here is always better :)
sven1977
approved these changes
Apr 30, 2024
sven1977
requested review from
avnishn,
ArturNiederfahrenhorst,
maxpumperla and
kouroshHakha
as code owners
harborn
pushed a commit
to harborn/ray
that referenced
this pull request
May 8, 2024
ryanaoleary
pushed a commit
to ryanaoleary/ray
that referenced
this pull request
Jun 7, 2024
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Outdated suggestions cannot be applied.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?
Using no KL loss in PPO throws an
AssertionErrorb/c thesampled_kl_valuesare empty. This PR fixes this error by moving the assertion only into the case whereuse_kl_loss=True. Furthermore, the sampled KL values were passed to thePPOLearnerthat does not use them and instead passed them over to theTorchLearnerthat has no use of them either. This argument was removed fromPPOLearner.additional_update_for_module.Related issue number
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.