Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add a condition to verify training status during image processing #20650

Merged
merged 5 commits into from
Dec 18, 2024

Conversation

shashaka
Copy link
Contributor

I have added a training condition to ensure that processing is not applied during the inference step for mix-up and random hue adjustments.

@codecov-commenter
Copy link

codecov-commenter commented Dec 16, 2024

Codecov Report

Attention: Patch coverage is 72.05882% with 19 lines in your changes missing coverage. Please review.

Project coverage is 81.89%. Comparing base (bce0f5b) to head (69307f6).

Files with missing lines Patch % Lines
...layers/preprocessing/image_preprocessing/mix_up.py 56.81% 18 Missing and 1 partial ⚠️
Additional details and impacted files
@@           Coverage Diff           @@
##           master   #20650   +/-   ##
=======================================
  Coverage   81.88%   81.89%           
=======================================
  Files         543      543           
  Lines       50609    50633   +24     
  Branches     7816     7822    +6     
=======================================
+ Hits        41443    41467   +24     
  Misses       7273     7273           
  Partials     1893     1893           
Flag Coverage Δ
keras 81.73% <70.58%> (+<0.01%) ⬆️
keras-jax 63.82% <64.70%> (+0.01%) ⬆️
keras-numpy 58.86% <64.70%> (+0.01%) ⬆️
keras-openvino 29.97% <2.94%> (-0.02%) ⬇️
keras-tensorflow 64.58% <64.70%> (+0.01%) ⬆️
keras-torch 63.65% <64.70%> (+0.01%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@google-ml-butler google-ml-butler bot added kokoro:force-run ready to pull Ready to be merged into the codebase labels Dec 17, 2024
@fchollet
Copy link
Collaborator

Thanks for the PR -- please take a look at the merge conflict 👍

# Conflicts:
#	keras/src/layers/preprocessing/image_preprocessing/random_hue.py
@google-ml-butler google-ml-butler bot removed the ready to pull Ready to be merged into the codebase label Dec 18, 2024
@shashaka
Copy link
Contributor Author

@fchollet,
I have resolved the merge conflict, updated the transform_bounding_boxes method, and added the corresponding test cases. Please let me know if any corrections are needed.

Copy link
Collaborator

@fchollet fchollet left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@google-ml-butler google-ml-butler bot added kokoro:force-run ready to pull Ready to be merged into the codebase labels Dec 18, 2024
@fchollet fchollet merged commit 4c7c4b5 into keras-team:master Dec 18, 2024
7 checks passed
@google-ml-butler google-ml-butler bot removed awaiting review ready to pull Ready to be merged into the codebase kokoro:force-run labels Dec 18, 2024
@shashaka shashaka deleted the training_condition branch December 18, 2024 23:13
shashaka added a commit to shashaka/keras that referenced this pull request Dec 20, 2024
…ras-team#20650)

* Add a condition to verify training status during image processing

* resolve merge conflict

* fix transform_bounding_boxes logic

* add transform_bounding_boxes test
fchollet pushed a commit that referenced this pull request Dec 22, 2024
* Add implementations for random_saturation

* change parse_factor method to inner method.

* Add implementations for random_color_jitter

* Fix Randomhue (#20652)

* Small fix in random hue

* use self.backend for seed

* test: add test for class weights (py_dataset adapter) (#20638)

* test: add test for class weights (py_dataset adapter)

* "call _standardize_batch from enqueuer"

m

* add more tests, handle pytorch astype issue

m

* convert to numpy to ensure consistent handling of operations

* Fix paths for pytest in contribution guide (#20655)

* Add preliminary support of OpenVINO as Keras 3 backend (#19727)

* [POC][OV] Support OpenVINO as Keras 3 backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Mark all unsupported ops from numpy space

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Mark unsupported ops in core, image, and linalg spaces

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Mark unsupported ops in math, nn, random, and rnn spaces

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix sorting imports

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Format imports

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix sorting imports

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix sorting imports

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix inference

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Remove openvino specific code in common part

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix typo

* Clean-up code

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Recover imports

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Sort imports properly

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Format source code

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Format the rest of source code

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Continue format adjustment

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add OpenVINO dependency

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix inference using OV backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Support bert_base_en_uncased and mobilenet_v3_small from Keras Hub

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Remove extra openvino specific code from layer.py

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Apply code-style formatting

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Apply code-style formatting

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix remained code-style issue

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Run tests for OpenVINO backend in GHA

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add config file for openvino backend validation

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add import test for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix error in import_test.py

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add import_test for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add openvino specific integration tests in GHA

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Exclude coverage for OpenVINO

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* remove coverage for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Try layer tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Run layer tests for openvino backend selectively

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Mark enabled tests for openvino backend in a different way

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Update .github/workflows/actions.yml

* Fix import for BackendVariable

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix errors in layer tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add test for Elu via openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix sorted imports

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Extend testing for attention

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Update keras/src/layers/attention/attention_test.py

* Switch on activation tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on attention tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Update keras/src/layers/attention/additive_attention_test.py

* Update keras/src/layers/attention/grouped_query_attention_test.py

* Run conv tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix convolution in openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Work around constant creation for tuple

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Work around constant creation in reshape

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Run depthwise conv tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix get_ov_output for other x types

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix elu translation

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix softmax and log_softmax for None axis

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Run nn tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix numpy operations for axis to be None

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Run operation_test for openvino_backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on math_test for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on image tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on linalg test for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Extend OpenVINOKerasTensor with new built-in methods and fix shape op

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on core tests for openvino backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Use different way of OpenVINO model creation that supports call method

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Unify integration test for openvino

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Support new operations abs, mod, etc.

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add support for more operations like squeeze, max

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Try to use excluded test files list

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Apply formatting for normalization_test.py

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Correct GHA yml file

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Test that openvino backend is used

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Revert testing change in excluded test files list

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Include testing group

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Include legacy test group

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Exclude legacy group of tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Include initializers tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Skip tests for initializers group

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Remove export test group from ignore

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Include dtype_policies test group

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Reduce ignored tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix ops.cast

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add decorator for custom_gradient

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Shorten line in custom_gradient

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Ignore dtype_policy_map test

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Include callback tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on backend tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Exclude failing tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Correct paths to excluded tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on some layers tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Remove pytest.mark.openvino_backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Register mark requires_trainable_backend

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Ignore test files in a different way

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Try different way to ignore test files

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix GHA yml

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Support tuple axis for logsumexp

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on some ops tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Switch on some callbacks tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add openvino export

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Update sklearn tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add a comment to skipp numerical_test

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add custom requirements file for OpenVINO

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Add reqs of openvino installation for api changes check

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix types of Variables and switch on some variables tests

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Fix nightly code check

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>

* Make sklearn dependency optional (#20657)

* Add a condition to verify training status during image processing (#20650)

* Add a condition to verify training status during image processing

* resolve merge conflict

* fix transform_bounding_boxes logic

* add transform_bounding_boxes test

* Fix recurrent dropout for GRU. (#20656)

The simplified implementation, which used the same recurrent dropout masks for all the previous states didn't work and caused the training to not converge with large enough recurrent dropout values.

This new implementation is now the same as Keras 2. Note that recurrent dropout requires "implementation 1" to be turned on.

Fixes #20276

* Fix example title in probabilistic_metrics.py (#20662)

* Change recurrent dropout implementation for LSTM. (#20663)

This change is to make the implementation of recurrent dropout consistent with GRU (changed as of #20656 ) and Keras 2.

Also fixed a bug where the GRU fix would break when using CUDNN with a dropout and no recurrent dropout. The solution is to create multiple masks only when needed (implementation == 1).

Added coverage for the case when dropout is set and recurrent dropout is not set.

* Never pass enable_xla=False or native_serialization=False in tests (#20664)

These are invalid options in the latest version of jax2tf, they
will just immediately throw.

* Fix `PyDatasetAdapterTest::test_class_weight` test with Torch on GPU. (#20665)

The test was failing because arrays on device and on cpu were compared.

* Fix up torch GPU failing test for mix up (#20666)

We need to make sure to use get any tensors places on cpu before using
them in the tensorflow backend during preprocessing.

* Add random_color_jitter processing layer

* Add random_color_jitter test

* Update test cases

* Correct failed test case

* Correct failed test case

* Correct failed test case

---------

Signed-off-by: Kazantsev, Roman <roman.kazantsev@intel.com>
Co-authored-by: IMvision12 <88665786+IMvision12@users.noreply.github.com>
Co-authored-by: Enrico <e.durso@live.com>
Co-authored-by: Marco <mmicu.github00@gmail.com>
Co-authored-by: Roman Kazantsev <roman.kazantsev@intel.com>
Co-authored-by: Matt Watson <1389937+mattdangerw@users.noreply.github.com>
Co-authored-by: hertschuh <1091026+hertschuh@users.noreply.github.com>
Co-authored-by: Jasmine Dhantule <dhantule@google.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants