lunny/keras
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
8,710 Commits 28 Branches 98 Tags 373 MiB
49c6ddbc42
Commit Graph

5 Commits

Author SHA1 Message Date
Francois Chollet
782da46752 Decrease numerical delta from tf.keras 2023-05-15 16:19:58 -07:00
Neel Kovelamudi
2d40cb20b9 Adds CategoryEncoding layer, bincount op, and tests (#161) 
* Adds unit normalization and tests

* Adds layer normalization and initial tests

* Fixes formatting in docstrings

* Fix type issues for JAX

* Fix nits

* Initial stash for group_normalization and spectral_normalization

* Adds spectral normalization and tests

* Adds group normalization and tests

* Formatting fixes

* Fix small nit in docstring

* Fix docstring and tests

* Adds RandomContrast and associated tests

* Remove arithmetic comment

* Adds RandomBrightness and tests

* Fix docstring and format

* Fix nits and add backend generator

* Inlines random_contrast helper

* Add bincount op

* Add CategoryEncoding layer and tests

* Fix formatting

* Fix JAX issues

* Fix JAX bincount

* Formatting and small fix

* Fix nits and docstrings

* Add args to bincount op test
 2023-05-14 00:07:43 +00:00
Jonathan Bischof
ff0db1c03f Initialize expected modules and import torch early 2023-05-13 18:06:22 +00:00
Qianli Scott Zhu
0e4bf48c5b Add distribution support for optimizer (#150) 
* Add distribution support for optimizer

* Fix lint

* Address review comments

* Refactor the structure for base optimizer and tf.

* Fix lint

* Address review comments.

* Address review comments.
 2023-05-12 15:44:03 -07:00
Qianli Scott Zhu
46378e883f Add distribution strategy support for model.fit/eval/predict. (#119) 
* Add unit/integration test for tf.distribute.

* Fix format

* Skip the test case for non-tf backend

* Fix typo

* Fix format and unit test context config.

* Address review comments.

* Add support for h5 weights loading.

* Fix test

* Add support for a -1 dimension in the `Reshape` operation. (#103)

The code to compute the output shape is now shared between the `Reshape` operation and the `Reshape` layer.

* Added ReLU activation layer (#104)

* added relu

* add relu

* added correctness test

* reformated

* updates based on review

* Fix docstring

* Added R2score (#106)

* Add meanX metrics

* All regression metrics except for root mean squared error

* Formatting issues

* Add RootMeanSquaredError

* Docstring spacing

* Line too long fix

* Add R2Score

* Docstring fixes

* Fix test

* Fix tests

* Adds RemoteMonitor Callback (#108)

* Add Remote Monitor Callback

* Add Remote Monitor Callback

* Add Remote Monitor Callback

* Add Remote Monitor

* Add wrapper layer.

* Add learning rate schedules (#102)

* Add learning rate schedules

* Some review comments

* Use fancy new serialization tests

* s/TensorFlow/backend in docstring

* Update docstrings

* More review comments

* Added LeakyReLU activation layer (#109)

* added LeakyReLu

* update docstring

* reformat

* update config

* updated test name

* Fix docstrings

* Fix init and update tests to import from correct path (#110)

* Add distribute support for tensorflow trainer.

* Revert the previous merge edit.

* Fix lint issue

* Address review comments.

* Add TPU strategy support

* Fix lint

---------

Co-authored-by: Francois Chollet <francois.chollet@gmail.com>
Co-authored-by: hertschuh <1091026+hertschuh@users.noreply.github.com>
Co-authored-by: divyasreepat <divyashreepathihalli@gmail.com>
Co-authored-by: Gabriel Rasskin <43894452+grasskin@users.noreply.github.com>
Co-authored-by: Ramesh Sampath <1437573+sampathweb@users.noreply.github.com>
Co-authored-by: Ian Stenbit <3072903+ianstenbit@users.noreply.github.com>
 2023-05-09 16:46:44 -07:00