[RLlin] Add dist_inputs to action sampler fn returns in TorchPolicyV2 #33795
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Outdated suggestions cannot be applied.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 adds the action distribution inputs to the values returned by self.action_sampler_fn in TorchPolicyV2.
Before, there where not enough values to unpack when
action_sampler_fn
was implemented correctly.I also explored unifying all
action_sampler_fn
occurrences in RLlib to four return values but I don't think that makes sense at this point. Most Policy classes will soon be deprecated anyway and we would be harddeprecating something (action_sampler_fn with three returns) that simply works for the time being.Fixes #33716