[RLlib; Offline RL] Make data pipeline better configurable and tuneable for users. #46777
+365
−248
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
…class into separate files. Signed-off-by: simonsays1980 <[email protected]>
…gorithmConfig' and 'OfflineData' to make the data pipeline better configurable and tuneable. Tested single-and multi-learner sertups with BC. Signed-off-by: simonsays1980 <[email protected]>
sven1977
changed the title
sven1977
requested review from
sven1977 and
ArturNiederfahrenhorst
as code owners
sven1977
reviewed
Jul 24, 2024
sven1977
reviewed
Jul 24, 2024
sven1977
reviewed
Jul 24, 2024
| @@ -2470,6 +2499,8 @@ def offline_data( | |||
| self.input_read_method_kwargs = input_read_method_kwargs | |||
| if input_read_schema is not NotProvided: | |||
| self.input_read_schema = input_read_schema | |||
| if prelearner_class is not NotProvided: | |||
| self.prelearner_class = prelearner_class | |||
There was a problem hiding this comment.
Wait, where do these get assigned?
input_read_method_kwargs
map_batches_kwargs
There was a problem hiding this comment.
Up in the file. Where all attributes get default values.
There was a problem hiding this comment.
Ah, sorry, didn't see this. I think b/c it hadn't been changed in this PR. All good.
sven1977
reviewed
Jul 24, 2024
|
|
||
| class OfflinePreLearner: |
There was a problem hiding this comment.
Thanks for separating these!
sven1977
reviewed
Jul 24, 2024
Co-authored-by: Sven Mika <[email protected]> Signed-off-by: simonsays1980 <[email protected]>
sven1977
reviewed
Jul 24, 2024
Signed-off-by: simonsays1980 <[email protected]>
sven1977
approved these changes
Jul 25, 2024
…n class. Signed-off-by: simonsays1980 <[email protected]>
…defined. Signed-off-by: simonsays1980 <[email protected]>
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?
The new
Offline RL APImakes direct use ofray.data.Datasetsand theremap_batchesanditer_batchesmethods to transform and iterate over batches. These methods allow for configuration to tune the data pipeline. This PR proposes a way for users to easily pass arguments to these methods in the ´AlgorithmConfig` to tune their data pipelines.In addition the
OfflinePreLearneris moved to its own file from which users can override its behavior in its_map_episodesand__call__method to define in detail how their data should be transferred to a batch/episodes.Furthermore, some bug fixes have been made.
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.