Make RandomHorizontalFlip torchscriptable #2278
Closed
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
fmassa
commented
Jun 1, 2020
| @@ -500,25 +500,29 @@ def __repr__(self): | |||
| return self.__class__.__name__ + '(size={0}, padding={1})'.format(self.size, self.padding) | |||
|
|
|||
|
|
|||
| class RandomHorizontalFlip(object): | |||
| """Horizontally flip the given PIL Image randomly with a given probability. | |||
| class RandomHorizontalFlip(torch.nn.Module): | |||
There was a problem hiding this comment.
@eellison this is the part I would like your thoughts on
Closed
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Outdated suggestions cannot be applied.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.
This is the first PR in a series towards making the torchvision transforms torchscriptable.
The majority of the work has already been done in the
functional_tensor.pyfile, which for now hasn't been properly exposed to the user.The objective of this PR is twofold:
functionalandtransformsseamlessly support Tensor argumentsfunctionalandtransformsbe torchscriptableIn order to minimize code changes, I decided for now to move (one at a time) the Pillow transforms to a separate file, and in
functional.pydispatch to eitherfunctional_tensor.pyorfunctional_pil.py.Another thing that might be worth discussion: the torchvision transforms now inherit from
nn.Module. I think this makes things simpler wrt torchscript, and as such might be preferred over keeping them as before. In particular, it will make it possible to save the transforms into a standalone file, and having custom__str__implementations.cc @eellison for feedback on replacing standard classes with classes inheriting from
nn.Modulefor a more seamless support for torchscript.