Fix BeitFeatureExtractor postprocessing #19119
Merged
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
alaradirik
changed the title
NielsRogge
reviewed
Sep 20, 2022
| resized = self.resize(image=semantic_segmentation[idx], size=target_sizes[idx]) | ||
| resized_maps.append(resized) | ||
|
|
||
| semantic_segmentation = [torch.Tensor(np.array(image)).to(torch.int64) for image in resized_maps] |
There was a problem hiding this comment.
Suggested change
| semantic_segmentation = [torch.Tensor(np.array(image)).to(torch.int64) for image in resized_maps] | |
| semantic_segmentation = [torch.Tensor(np.array(image)).long() for image in resized_maps] |
This should also work :)
NielsRogge
reviewed
Sep 20, 2022
| None, predictions will not be resized. | ||
| Returns: | ||
| semantic_segmentation: `List[torch.Tensor]` of length `batch_size`, where each item is a semantic | ||
| segmentation map of shape (w, h) corresponding to the target_sizes entry (if `target_sizes` is specified). |
There was a problem hiding this comment.
To me it's a bit counterintuitive to output (w, h) if we ask the target_sizes to be (h, w)
=> so I'd use the same format for both
There was a problem hiding this comment.
Agreed, just fixed this
NielsRogge
reviewed
Sep 20, 2022
| semantic_segmentation = semantic_segmentation.numpy() | ||
|
|
||
| for idx in range(len(semantic_segmentation)): | ||
| resized = self.resize(image=semantic_segmentation[idx], size=target_sizes[idx]) |
There was a problem hiding this comment.
As discussed offline, please use nn.functional.interpolate here.
resized = nn.functional.interpolate(semantic_segmentation[idx],
size=target_sizes[idx],
mode='bilinear',
align_corners=False)
|
The documentation is not available anymore as the PR was closed or merged. |
NielsRogge
approved these changes
Sep 20, 2022
|
Hey @alaradirik, please also ping a core maintainer for review before merging PRs. |
oneraghavan
pushed a commit
to oneraghavan/transformers
that referenced
this pull request
Sep 26, 2022
* return post-processed segmentations as list, add test * use torch to resize logits * fix assertion error if no target_size is specified
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?
BeitFeatureExtractor.post_process_semantic_segmentation()assertion error when notarget_sizesargument is providedBefore submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.