F_t add factor nonnegativity checks for adjust_* #2356
Merged
Conversation
This file contains hidden or 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
reviewed
Jun 29, 2020
There was a problem hiding this comment.
Thanks for the improvements @ssnl !
Test failures are related, looks like the optimization using lerp is not supported with torchscript.
Can you revert those changes? I think it's fine to use the unfused ops, as torchscript can most probably fuse everything into a single call anyway.
Comment on lines
293
to
296
| if img1.dtype in [torch.half, torch.float32, torch.float64]: | ||
| return torch.lerp(ratio, img2, img1).clamp(0, 1).to(img1.dtype) | ||
| else: | ||
| return torch.lerp(ratio, img2.float(), img1.float()).clamp(0, 255).to(img1.dtype) |
There was a problem hiding this comment.
Looks like torchscript might not support this overload yet. Can you revert those changes?
fmassa
approved these changes
Jun 29, 2020
|
Test failures seems related, can you fix those? |
fmassa
reviewed
Jun 29, 2020
| if not _is_tensor_a_torch_image(img): | ||
| raise TypeError('tensor is not a torch image.') | ||
|
|
||
| mean = torch.mean(rgb_to_grayscale(img).to(torch.float)) | ||
| mean = rgb_to_grayscale(img).mean() |
There was a problem hiding this comment.
mean is currently not supported for uint8 types. I think this will be soon fixed in PyTorch, but in the meantime, can you revert this change as well?
fmassa
approved these changes
Jun 30, 2020
de-vri-es
pushed a commit
to fizyr-forks/torchvision
that referenced
this pull request
Aug 4, 2020
* F_t add factor nonnegativity checks for adjust_* * Update functional_tensor.py * Update functional_tensor.py * Update functional_tensor.py * Update functional_tensor.py
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.
No description provided.