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.

Sign up for GitHub

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 TorchMultiCategorical.to_deterministic to return a multi-dimensional tensor instead of a list. #49098

Merged
merged 3 commits into from
Dec 6, 2024
Merged

[RLlib] Fix TorchMultiCategorical.to_deterministic to return a multi-dimensional tensor instead of a list. #49098

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
4b2e7f4
Modified deterministic distribution for 'TorchMultiCategorical' such …
simonsays1980 Dec 4, 2024
396ad95
Merge branch 'master' into fix-action-multi-distribution-deterministi…
simonsays1980 Dec 5, 2024
6897578
Apply suggestions from code review
sven1977 Dec 6, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 11 additions & 2 deletions rllib/models/torch/torch_distributions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -497,8 +497,17 @@ def from_logits(

return TorchMultiCategorical(categoricals=categoricals)

def to_deterministic(self) -> "TorchMultiDistribution":
return TorchMultiDistribution([cat.to_deterministic() for cat in self._cats])
def to_deterministic(self) -> "TorchDeterministic":
if self._cats[0].probs is not None:
probs_or_logits = nn.utils.rnn.pad_sequence(
[cat.logits.t() for cat in self._cats], padding_value=-torch.inf
)
else:
probs_or_logits = nn.utils.rnn.pad_sequence(
[cat.logits.t() for cat in self._cats], padding_value=-torch.inf
)

return TorchDeterministic(loc=torch.argmax(probs_or_logits, dim=0))


@DeveloperAPI
Expand Down
Loading