Add sanitize_bounding_boxes kernel/functional
#8308
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/8308
Note: Links to docs will display an error until the docs builds have been completed. ❌ 16 New Failures, 20 Unrelated FailuresAs of commit e5a5892 with merge base d1f3a7b ( NEW FAILURES - The following jobs have failed:
FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
NicolasHug
commented
Mar 12, 2024
| Returns: | ||
| out (tuple of Tensors): The subset of valid bounding boxes, and the corresponding indexing mask. | ||
| The mask can then be used to subset other tensors (e.g. labels) that are associated with the bounding boxes. | ||
| """ |
| @pytest.mark.parametrize("min_size", (1, 10)) | ||
| @pytest.mark.parametrize("labels_getter", ("default", lambda inputs: inputs["labels"], None, lambda inputs: None)) | ||
| @pytest.mark.parametrize("sample_type", (tuple, dict)) | ||
| def test_transform(self, min_size, labels_getter, sample_type): |
There was a problem hiding this comment.
That test did not change, I just moved the input generation above in a function so that it can be reused in the newly added tests below
Comment on lines
+325
to
+328
| valid = _get_sanitize_bounding_boxes_mask( | ||
| bounding_boxes, format=format, canvas_size=canvas_size, min_size=min_size | ||
| ) | ||
| bounding_boxes = bounding_boxes[valid] |
There was a problem hiding this comment.
A bit sad that these 2 lines are somewhat duplicated below in the else: block. I couldn't find a decent way to make all that work without upsetting torchscript.
| return bounding_boxes, valid | ||
|
|
||
|
|
||
| def _get_sanitize_bounding_boxes_mask( |
There was a problem hiding this comment.
The logic of this function is completely unchanged from the existing logic in the class.
pmeier
approved these changes
Mar 15, 2024
Comment on lines
+325
to
+328
| valid = _get_sanitize_bounding_boxes_mask( | ||
| bounding_boxes, format=format, canvas_size=canvas_size, min_size=min_size | ||
| ) | ||
| bounding_boxes = bounding_boxes[valid] |
NicolasHug
commented
Mar 15, 2024
| @@ -123,7 +123,7 @@ def _extract_image_targets( | |||
| if not (len(images) == len(bboxes) == len(masks) == len(labels)): | |||
| raise TypeError( | |||
| f"{type(self).__name__}() requires input sample to contain equal sized list of Images, " | |||
| "BoundingBoxeses, Masks and Labels or OneHotLabels." | |||
| "BoundingBoxes, Masks and Labels or OneHotLabels." | |||
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 21, 2024
Reviewed By: vmoens Differential Revision: D55062772 fbshipit-source-id: 677840081e6516e2d1d603853f16314d209dbf5b
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.
Addresses part of #8294
Some design discussion started on #8294 (comment). After chatting a bit offline we decided to go with Option 3 (#8294 (comment)) because:
*args)tv_tensors.wrap()cc @vfdev-5