[tensor wrapper subclass] Add trace transform for tensor subclasses #1584
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
github-actions
bot
added
the
documentation
crcrpar
force-pushed
the
tensor_subclass_2
branch
from
515d425 to
2e763f1
Compare
Signed-off-by: Masaki Kozuki <mkozuki@nvidia.com>
to support `__torch_dispatch__`. Since it extends the behavior that is implemented in C++ level, we'd need to apply the transform to split forward and backward traces separately. Signed-off-by: Masaki Kozuki <mkozuki@nvidia.com>
crcrpar
force-pushed
the
tensor_subclass_2
branch
from
2e763f1 to
a534003
Compare
for more information, see https://pre-commit.ci
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Outdated suggestions cannot be applied.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.
What does this PR do?
Implement
my_subclass.__tensor_flatten__andMySubclass.__tensor_unflatten__MySubclass.__torch_dispatch__.torch.fxto understand the extended behavior. Currently the mapping from core aten ops to thunder.torch ops is quite optimistic; just querying the aten op name tothunder.torch.Since
__torch_dispatch__extends the behavior that is implemented in C++ level, we'd need to apply the transform to split forward and backward traces separately.