feat(layers): Migrate RandomChoice and RandomApply layers from keras-cv to keras #20752
+515
−0
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
- Added RandomChoice and RandomApply layers to keras.layers.preprocessing.image_preprocessing. - Implemented necessary transform methods (transform_images, transform_labels, transform_bounding_boxes, transform_segmentation_masks) to comply with BaseImagePreprocessingLayer. - Updated tests to ensure compatibility with keras and fix failing test cases. - Added support for batchwise processing, auto-vectorization, and random seed control.
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #20752 +/- ##
==========================================
+ Coverage 81.94% 82.01% +0.06%
==========================================
Files 553 557 +4
Lines 51536 51972 +436
Branches 7978 8034 +56
==========================================
+ Hits 42231 42623 +392
- Misses 7363 7393 +30
- Partials 1942 1956 +14
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
innat
reviewed
Jan 13, 2025
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
…andomApply and RandomChoice layers - Updated layers to handle JAX backend-specific behavior and avoid tracer leaks. - Updated tests to use backend-agnostic Keras operations. - Skipped JAX-specific tests with TODO to investigate further. - Removed Tensorflow-specific dependencies for better backend compatibility.
innat
reviewed
Jan 14, 2025
innat
reviewed
Jan 14, 2025
keras/src/layers/preprocessing/image_preprocessing/random_apply_test.py
Outdated
Show resolved
Hide resolved
- Add handling for layers with get_random_transformation method in both RandomApply and RandomChoice preprocessing layers - Fix seed handling in RandomChoice to ensure consistent behavior across backends - Remove JAX-specific test skipmarks after addressing tracer errors
fchollet
reviewed
Jan 16, 2025
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
- Simplified docstrings. - Removed copyright headers from files. - Refactored RandomApply and RandomChoice. - Streamlined transformation logic and improved overall code organization. - Eliminated redundant backend-specific code paths.
|
@fchollet, would love your thoughts on these changes whenever you get a chance, thanks! Also please do look into the comment I've left unresolved regarding whether it should inherit from |
fchollet
reviewed
Jan 18, 2025
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
keras/src/layers/preprocessing/image_preprocessing/random_apply.py
Outdated
Show resolved
Hide resolved
|
Sure, I'll review it. Thanks for tagging me. |
innat
reviewed
Jan 18, 2025
keras/src/layers/preprocessing/image_preprocessing/random_choice.py
Outdated
Show resolved
Hide resolved
- removed unused auto_vectorize parameter from both layers - shortened docstring headers to fit on one line - added backticks around boolean keywords (True, False) in docstrings - fixed incorrect transform methods (bounding_boxes, labels, segmentation_masks) to return inputs unchanged - removed unnecessary transformation methods from RandomChoice for consistency with original implementation
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.
RandomChoiceandRandomApplylayers tokeras.layers.preprocessing.image_preprocessingtransform_images,transform_labels,transform_bounding_boxes,transform_segmentation_masks) to comply withBaseImagePreprocessingLayer