| .. |
|
api
|
Export unpack_x_y_sample_weight and pack_x_y_sample_weight
|
2020-06-02 23:44:08 -07:00 |
|
applications
|
correct summing total blocks
|
2020-06-05 18:55:32 -04:00 |
|
benchmark
|
Add fused/non-fused inference/training overhead benchmarks for
|
2020-06-16 08:58:33 -07:00 |
|
datasets
|
Ensure there are test samples for imdb dataset, when maxlen is low
|
2020-06-19 07:04:19 -05:00 |
|
distribute
|
Skip data loading error in multi_worker_tutorial_test (the test does not aim to cover this).
|
2020-06-23 15:56:23 -07:00 |
|
engine
|
Merge pull request #41483 from tomerk/cherrypicks_DC0YA
|
2020-07-17 12:07:45 -07:00 |
|
estimator
|
Add a note to model_to_estimator docs about weighted metrics.
|
2020-04-22 21:41:46 -07:00 |
|
feature_column
|
Move the _BaseFeatureLayer back to Keras.
|
2020-05-20 11:34:23 -07:00 |
|
initializers
|
Initializer docstrings improvements (code formatting, etc).
|
2020-04-01 14:26:42 -07:00 |
|
integration_test
|
Fork the keras related tpu_strategy_test to keras integration test.
|
2020-06-18 19:41:18 -07:00 |
|
layers
|
Going back to forcing embedding layer variables on the CPU even within a tf.function as this is breaking some user code.
|
2020-07-22 11:45:58 -07:00 |
|
legacy_tf_layers
|
Merge pull request #35519 from ROCmSoftwarePlatform:google_upstream_rocm_test_updates_191231
|
2020-03-25 11:42:20 -07:00 |
|
mixed_precision
|
Allow tf.distribute.TPUStrategy to be used with TPUEmbedding API and ensure that LossScaleOptimizer properly rejects it.
|
2020-06-25 15:32:43 -07:00 |
|
optimizer_v2
|
[ROCm] Adding no_rocm tag to a unit-test regression.
|
2020-06-02 14:47:24 +00:00 |
|
premade
|
Stop skipping KerasTensors for Keras tests that now work with KerasTensors.
|
2020-06-15 14:26:38 -07:00 |
|
preprocessing
|
GitHub Issue #40462 fix typo
|
2020-06-20 20:37:12 -07:00 |
|
protobuf
|
Update naming of embedding in callback for TensorBoard integration.
|
2020-06-04 02:39:35 -07:00 |
|
saving
|
Roll forward of cl/316247127: Make sure compiled metrics are accessible after loading from H5 or SavedModel.
|
2020-06-19 10:22:33 -07:00 |
|
tests
|
Updates Keras layer __call__ to always set training, with the following priority order:
|
2020-06-22 13:13:46 -07:00 |
|
type
|
Update package visibility for all keras sub packages.
|
2020-03-19 09:30:52 -07:00 |
|
utils
|
Simplify Layer.add_udpate in v2 and update version_selector to use v1 inside a
|
2020-06-17 11:01:44 -07:00 |
|
wrappers
|
Update package visibility for all keras sub packages.
|
2020-03-19 09:30:52 -07:00 |
|
__init__.py
|
Update tf.keras version number.
|
2020-06-15 20:37:05 -07:00 |
|
activations_test.py
|
fix conflict
|
2020-03-11 11:09:44 -07:00 |
|
activations.py
|
Merge pull request #27814 from Bharat123rox:update-elu
|
2020-06-10 14:16:42 -07:00 |
|
backend_config_test.py
|
Update tests under Keras to use combination.
|
2020-03-06 11:55:08 -08:00 |
|
backend_config.py
|
Add dispatch support to more Python APIs.
|
2020-05-15 11:03:18 -07:00 |
|
backend_test.py
|
Merge pull request #36316 from faustomorales:fix/#35799-ctc-decode
|
2020-05-28 11:11:44 -07:00 |
|
backend.py
|
Fix critical bug with add_loss TFOpLayer graph construction
|
2020-07-15 12:18:38 -07:00 |
|
BUILD
|
This change adds WIP support for using KerasTensor objects in Keras's functional API instead of symbolic graph tf.Tensors. It is controlled by an internal behavior flag that is disabled by default and is not yet exposed in TF's APIs.
|
2020-06-05 15:47:28 -07:00 |
|
callbacks_test.py
|
Add SaveOptions/CheckpointOptions to keras.Models.save_weights and keras_call_backs.ModelCheckpoint.
|
2020-06-17 15:09:44 -07:00 |
|
callbacks_v1_test.py
|
Update tests under keras to use combinations.
|
2020-03-14 15:56:20 -07:00 |
|
callbacks_v1.py
|
Fix various links in Keras docstrings.
|
2020-04-15 19:59:22 -07:00 |
|
callbacks.py
|
Fix the api docstrings for on_*_batch_begin methods in callbacks.Callback
|
2020-06-25 15:37:32 -07:00 |
|
combinations_test.py
|
Add test combinations for Keras test.
|
2020-03-04 14:24:59 -08:00 |
|
combinations.py
|
This change adds WIP support for using KerasTensor objects in Keras's functional API instead of symbolic graph tf.Tensors. It is controlled by an internal behavior flag that is disabled by default and is not yet exposed in TF's APIs.
|
2020-06-05 15:47:28 -07:00 |
|
constraints_test.py
|
Update tests under Keras to use combination.
|
2020-03-06 11:55:08 -08:00 |
|
constraints.py
|
Improve keras.constraints docstrings.
|
2020-03-23 10:08:12 -07:00 |
|
initializers_test.py
|
Make Keras deserialization aware of enable/disable_v2_behavior switches.
|
2020-03-27 10:26:34 -07:00 |
|
keras_parameterized_test.py
|
This change adds WIP support for using KerasTensor objects in Keras's functional API instead of symbolic graph tf.Tensors. It is controlled by an internal behavior flag that is disabled by default and is not yet exposed in TF's APIs.
|
2020-06-05 15:47:28 -07:00 |
|
keras_parameterized.py
|
This change adds WIP support for using KerasTensor objects in Keras's functional API instead of symbolic graph tf.Tensors. It is controlled by an internal behavior flag that is disabled by default and is not yet exposed in TF's APIs.
|
2020-06-05 15:47:28 -07:00 |
|
losses_test.py
|
Switches keras.backend.placeholder + keras.backend.function to build a keras model when running eagerly (instead of trying to directly lift ops out of a graph into a concretefunction). Allows us to strip most of EagerDefinedExecutionFunction from the keras backend.
|
2020-05-21 11:53:44 -07:00 |
|
losses.py
|
Update losses.py
|
2020-06-11 17:47:56 -07:00 |
|
metrics_confusion_matrix_test.py
|
Fix bug in SensitivitySpecificityBase derived metrics.
|
2020-03-18 10:21:37 -07:00 |
|
metrics_correctness_test.py
|
Fix last partial batch loss regression in 2.2
|
2020-04-21 13:13:04 -07:00 |
|
metrics_functional_test.py
|
Fix SparseTopKCategoricalAccuracy for predictions with rank > 2.
|
2019-11-21 09:48:07 -08:00 |
|
metrics_test.py
|
Merge pull request #39134 from yongtang:36790-tf.metrics.Recall-backend-float64
|
2020-05-28 10:01:00 -07:00 |
|
metrics.py
|
Roll forward of cl/316247127: Make sure compiled metrics are accessible after loading from H5 or SavedModel.
|
2020-06-19 10:22:33 -07:00 |
|
models_test.py
|
Follow up to 0e1f3de50a. This change enables keras tensor tests related to updates/losses.
|
2020-06-09 01:13:30 -07:00 |
|
models.py
|
Reduce 1-layer Functional.__call__ overhead by ~10%
|
2020-06-01 16:51:20 -07:00 |
|
optimizers_test.py
|
Fix a number of deserialization error messages. While print(string, obj) is valid syntax, ValueError(string, obj) is not.
|
2020-05-04 21:25:54 -07:00 |
|
optimizers.py
|
Fix a number of deserialization error messages. While print(string, obj) is valid syntax, ValueError(string, obj) is not.
|
2020-05-04 21:25:54 -07:00 |
|
README.md
|
Branch 167812735 (#12867)
|
2017-09-06 23:52:56 -07:00 |
|
regularizers_test.py
|
Stop skipping KerasTensors for Keras tests that now work with KerasTensors.
|
2020-06-15 14:26:38 -07:00 |
|
regularizers.py
|
Merge pull request #40073 from howl-anderson:bugfix/tf_issue_37196
|
2020-06-10 10:00:46 -07:00 |
|
testing_utils.py
|
Fix tokenization tests and update testing_utils to transfer state between layer creation.
|
2020-06-19 14:34:39 -07:00 |
