| .. |
|
legacy_rnn
|
Update RNNCell._rnn_get_variable to use Variable._trainable in TF2 mode.
|
2020-06-05 23:02:56 -07:00 |
|
ops
|
Reduce Dense.__call__ overhead by ~15%
|
2020-06-01 20:22:44 -07:00 |
|
preprocessing
|
Fix a critical breakage in training argument default value in inference for
|
2020-06-25 15:26:51 -07:00 |
|
__init__.py
|
Export StringLookup and IntegerLookup layers.
|
2020-05-29 12:37:56 -07:00 |
|
advanced_activations_test.py
|
Remove passing experimental_run_tf_function in most tests.
|
2020-02-27 13:34:09 -08:00 |
|
advanced_activations.py
|
Fixes:GitHub Issue #38640
|
2020-04-27 13:01:35 -07:00 |
|
BUILD
|
Move the module test wrt to Keras private API to keras test.
|
2020-06-16 09:23:32 -07:00 |
|
convolutional_recurrent_test.py
|
Remove passing experimental_run_tf_function in most tests.
|
2020-02-27 13:34:09 -08:00 |
|
convolutional_recurrent.py
|
Make return_state as explicit kwarg in the Conv2DLSTM layer.
|
2020-06-18 15:55:19 -07:00 |
|
convolutional_test.py
|
Reduce Conv2D.__call__ overhead by 45%, Conv3D.__call__ overhead by 30%, and
|
2020-06-11 10:06:48 -07:00 |
|
convolutional_transpose_test.py
|
|
|
|
convolutional.py
|
Reduce Conv2D.__call__ eager overhead by ~20%, Conv3D.__call__ by ~35%, and
|
2020-06-12 14:54:58 -07:00 |
|
core_test.py
|
Fix Lambda layer deserialization polluting core module.
|
2020-06-05 12:50:10 -07:00 |
|
core.py
|
Fix Keras documentation
|
2020-06-17 02:12:14 -07:00 |
|
cudnn_recurrent_test.py
|
Disable failed test for now.
|
2020-05-13 09:14:15 -07:00 |
|
cudnn_recurrent.py
|
Merge pull request #26216 from lgeiger:keras-recurrent-updates
|
2020-03-16 11:13:23 -07:00 |
|
dense_attention_test.py
|
Update tests under keras.layers to use combinations.
|
2020-03-16 14:34:12 -07:00 |
|
dense_attention.py
|
Update ops.convert_to_tensor() to use the TF public symbol in Keras.
|
2020-02-13 10:36:58 -08:00 |
|
einsum_dense_test.py
|
Add an einsum layer.
|
2020-04-30 08:42:06 -07:00 |
|
einsum_dense.py
|
Add an einsum layer.
|
2020-04-30 08:42:06 -07:00 |
|
embeddings_test.py
|
Support ShardedVariable in tf.keras.layers.Embedding.
|
2020-05-21 11:06:33 -07:00 |
|
embeddings.py
|
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 |
|
gru_test.py
|
Fix RNN cell wrt deep copy.
|
2020-06-01 21:56:22 -07:00 |
|
gru_v2_test.py
|
Update tests under keras.layers to use combinations.
|
2020-03-16 14:34:12 -07:00 |
|
kernelized_test.py
|
Make RandomFourierFeatures state saveable.
|
2020-05-25 17:08:03 -07:00 |
|
kernelized.py
|
Make RandomFourierFeatures state saveable.
|
2020-05-25 17:08:03 -07:00 |
|
layers_test.py
|
Move the module test wrt to Keras private API to keras test.
|
2020-06-16 09:23:32 -07:00 |
|
local_test.py
|
Unifies the behaviors of "regular Sequential" (starts with an Input, wraps a Functional model) and "deferred Sequential" models (only gets built when it sees its input data for the first time).
|
2020-04-12 16:04:20 -07:00 |
|
local.py
|
tf.SparseTensor to tf.sparse.SparseTensor
|
2020-03-25 10:49:19 +08:00 |
|
lstm_test.py
|
Fix RNN cell wrt deep copy.
|
2020-06-01 21:56:22 -07:00 |
|
lstm_v2_test.py
|
Remove passing experimental_run_tf_function in most tests.
|
2020-02-27 13:34:09 -08:00 |
|
merge_test.py
|
Make tf.keras.layers.Add() support scalar tensors.
|
2020-04-06 21:34:18 -07:00 |
|
merge.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 |
|
noise_test.py
|
Fix crash when float64 or mixed precision used in certain layers.
|
2020-01-27 16:27:40 -08:00 |
|
noise.py
|
Fix crash when float64 or mixed precision used in certain layers.
|
2020-01-27 16:27:40 -08:00 |
|
normalization_test.py
|
Simplify Layer.add_udpate in v2 and update version_selector to use v1 inside a
|
2020-06-17 11:01:44 -07:00 |
|
normalization_v2.py
|
Fix various links in Keras docstrings.
|
2020-04-15 19:59:22 -07:00 |
|
normalization.py
|
Do non-fused mixed precision BatchNormalization in float32.
|
2020-06-09 18:06:06 -07:00 |
|
pooling_test.py
|
Update tests under keras.layers to use combinations.
|
2020-03-16 14:34:12 -07:00 |
|
pooling.py
|
Remove Layer._supports_ragged_inputs property.
|
2020-04-30 16:07:54 -07:00 |
|
recurrent_test.py
|
Remove conditional loss/graph tracing based on inputs in v2 add_update/get_updates_for API.
|
2020-04-21 15:08:33 -07:00 |
|
recurrent_v2_test.py
|
Don't use the cuDNN GPU kernel for LSTMs when inputs are RaggedTensors.
|
2020-05-29 12:50:57 -07:00 |
|
recurrent_v2.py
|
Updated documentation for GRU & LSTM
|
2020-06-09 10:43:47 +08:00 |
|
recurrent.py
|
Avoid deprecation warning related to the use of collections.Sequence. It will stop working in Python 3.8.
|
2020-06-17 14:35:15 -07:00 |
|
rnn_cell_wrapper_v2_test.py
|
Update tests under keras.layers to use combinations.
|
2020-03-16 14:34:12 -07:00 |
|
rnn_cell_wrapper_v2.py
|
Updates Keras layer __call__ to always set training, with the following priority order:
|
2020-06-22 13:13:46 -07:00 |
|
separable_convolutional_test.py
|
Split off SeparableConv and ConvTranspose tests and parameterize many test so they can be properly sharded.
|
2019-01-10 22:28:17 -08:00 |
|
serialization_test.py
|
Make regularizers API more consistent.
|
2020-05-13 14:27:05 -07:00 |
|
serialization.py
|
Export StringLookup and IntegerLookup layers.
|
2020-05-29 12:37:56 -07:00 |
|
simplernn_test.py
|
Fix RNN cell wrt deep copy.
|
2020-06-01 21:56:22 -07:00 |
|
subclassed_layers_test.py
|
Update ops.convert_to_tensor() to use the TF public symbol in Keras.
|
2020-02-13 10:36:58 -08:00 |
|
tensorflow_op_layer_test.py
|
Remove keras_tensor skipping from keras tensorflow_op_layer_test
|
2020-06-12 10:29:30 -07:00 |
|
wrappers_test.py
|
Simplify Layer.add_udpate in v2 and update version_selector to use v1 inside a
|
2020-06-17 11:01:44 -07:00 |
|
wrappers.py
|
Fix Keras documentation
|
2020-06-17 02:12:14 -07:00 |
