[Refactoring] Import at root to enable vmap monkey-patching #500
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
facebook-github-bot
added
the
CLA Signed
|
cc @nicolas-dufour class MyModule(torch.nn.Module):
def forward(self, tensordict, tensor):
a = tensordict["a"]
return (
TensorDict(
{"a": a}, tensordict.batch_size, device=tensordict.device
),
tensor,
)
tensor = torch.randn(3).expand(4, 3)
tensordict = TensorDict({"a": torch.randn(3, 1)}, [3])
out = vmap(MyModule(), (None, 0))(tensordict, tensor)
assert out[0].shape == torch.Size([4, 3])
assert out[1].shape == torch.Size([4, 3])
assert out[0]["a"].shape == torch.Size([4, 3, 1])allowing for nn.Module with arbitrary TensorDict inputs |
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Outdated suggestions cannot be applied.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.
Description
nn.Modulevmap with tensordictsFollowing suggestions from @zou3519 we remove and add a batch dimension when vmapping over a tensordict.
Some commended code shows attempts to register TensorDIct in pytree but that failed + it requires to decompose and recompose the tensordict multiple times, while doing complex operations to keep track of the modified batch size. Besides the fact that it did not work, it also seemed suboptimal from a compute time perspective.