Katherine Wu fed56de128 Remove unneeded input extraction from layer tracing. ... The inputs are used for V1 layer's losses (`layer.get_losses_for(input)`), but this code is also causing issues when tracing layers with multiple inputs. This change fixes the tracing errors. PiperOrigin-RevId: 358235121 Change-Id: I94b667d1f89be2415e046ebeb00c4adc1acdc207		2021-02-18 12:21:12 -08:00
..
saved_model	Remove unneeded input extraction from layer tracing.	2021-02-18 12:21:12 -08:00
__init__.py
BUILD	PY2 removal cleanup	2021-01-15 16:48:57 -08:00
hdf5_format.py	[*.py,tensorflow/cc/framework/cc_op_gen.cc] Rename "Arguments:" to "Args:"	2020-12-22 09:24:04 +11:00
losses_serialization_test.py	Remove passing experimental_run_tf_function in most tests.	2020-02-27 13:34:09 -08:00
metrics_serialization_test.py	Remove passing experimental_run_tf_function in most tests.	2020-02-27 13:34:09 -08:00
model_config.py	Add support for creating Keras.inputs from arbitrary TypeSpecs. (Including TypeSpecs that don't have a dtype)	2021-01-05 12:58:49 -08:00
save_test.py	Add support for shared objects in keras when (de)serializing.	2021-02-01 12:27:05 -08:00
save_weights_test.py	Remove irrelevant warnings.	2021-01-11 11:18:13 -08:00
save.py	Add support for shared objects in keras when (de)serializing.	2021-02-01 12:27:05 -08:00
saved_model_experimental_test.py	Copy tensorflow.python.saved_model.loader_impl.get_train_op to test file	2020-10-30 12:36:12 -07:00
saved_model_experimental.py	Move V1 optimizer code to a separate file optimizer_v1.py from the generic utils in optimizers.py.	2020-09-24 05:34:47 -07:00
saving_utils_test.py	Allow autograph to be applied in internal helper utility. That avoids downstream calls from being confused about whether autograph is enabled in their context.	2020-11-10 10:14:31 -08:00
saving_utils.py	Allow weights to be loaded from saved model.	2020-12-16 09:40:49 -08:00