keras_export('keras.applications.mobilenet_v2.preprocess_input')

keras_export('keras.applications.vgg19.preprocess_input')

keras_export('keras.applications.resnet50.preprocess_input', 'keras.applications.resnet.preproce

keras_export('keras.applications.resnet_v2.preprocess_input')

dispatch.add_dispatch_support

dispatch.add_dispatch_support

tf_export(v1=["train.range_input_producer"])

dispatch.add_dispatch_support

dispatch.add_dispatch_support

tf_export("recompute_grad")

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

deprecated('2017-03-25', 'signature_fns are deprecated. For canned Estimators they are no '

deprecation.deprecated( date=None, instructions="Use standard file APIs to delete files with this pref

deprecated( None, "Please switch to remove_squeezable_dimensions from " "tf.confusion_matrix. Note tha

dispatch.add_dispatch_support

def_function.function(input_signature=[ tensor_spec.TensorSpec(shape=[None, 1, 1], dtype=dtypes.floa

dispatch.add_dispatch_support

deprecation.deprecated( None, "Queue-based input pipelines have been replaced by `tf.data`. Use " "`tf

deprecation.deprecated( None, "Queue-based input pipelines have been replaced by `tf.data`. Use " "`tf

function.defun(input_signature=[tensor_spec.TensorSpec([2], dtypes.int32)])

dispatch.add_dispatch_support

dispatch.add_dispatch_support

tf_export(v1=["train.slice_input_producer"])

deprecated('2016-12-01', 'Use `tf.losses.softmax_cross_entropy` ' 'and explicit logits computation

tf.function(input_signature=[ tf.TensorSpec([], tf.float32), tf.TensorSpec([], tf.float32) ])

tf.function(input_signature=[])

tf.function(input_signature=[tf.TensorSpec([], tf.float32)])

tf.function(input_signature=[tf.TensorSpec([], tf.float32)])

tf.function(input_signature=[tf.TensorSpec([], tf.float32)])

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

tf_export(v1=["train.string_input_producer"])

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

dispatch.add_dispatch_support

combinations.generate(combinations.combine( mode=["graph"], input_type=["dataset"], api_type

parameterized.parameters([ dict( params_shape=[2, 3, 4, 5, 6, 7], indices_shape=[2,

combinations.generate( combinations.combine( mode=["graph", "eager"], input_type=["d

mock.patch.object(resolver, 'is_running_in_gce', mock_is_running_in_gce)

parameterized.named_parameters( ("Shapes", [0], [[1, 2, 3]], "must have compatible output shapes."),

test.mock.patch.object(checkpoint_management, "time")

test_util.enable_output_all_intermediates

parameterized.parameters([ dict( doc="Single string", input=_nested_encode([u"仅今年前"]

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, 'warning', autospec=True)

combinations.generate( combinations.combine( mode=["graph"], input_type=["input_fn"],

test_util.enable_output_all_intermediates

test_util.enable_output_all_intermediates

parameterized.parameters( dict( rt_inputs=(), axis=0, error=ValueError,

parameterized.parameters( { 'rt_input': [[1, 2, 3]], 'default': [0], 'erro

parameterized.parameters([ dict( input=[b"\xFE", b"hello", b"==\xFF==", b"world"], i

parameterized.parameters([ dict( input=[b"\xFE", b"hello", b"==\xFF==", b"world"], i

parameterized.parameters([ dict( # axis=2 out of bounds: expected -2<=axis<2. rt_input=[[10,

test.mock.patch.object(logging, "warning", autospec=True)

parameterized.parameters([ dict(input=[b"\xFEED"], errors="strict", input_encoding

parameterized.parameters([ dict(inputs=[], expected=[]), dict(inputs=[23, "42"], expected=[("0", 2

parameterized.parameters([ dict(inputs=[], expected=[]), dict(inputs=[23, "42"], expected=[((0,),

parameterized.parameters([ dict( params_shape=[2, 3, 4, 5, 6, 7], indices_shape=[2,

mock.patch.object(six.moves.urllib.request, 'urlopen', mock_not_runn

mock.patch.object(six.moves.urllib.request, 'urlopen', mock_running_

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, "warning", autospec=True)

test.mock.patch.object(logging, "warning", autospec=True)

test_util.disable_xla("b/123860949") # The computation is constant folded

parameterized.parameters( [ {'inputs': (12, 8, 3)}, {'inputs': ([1, 2, 3], [7, 8, 9]

mock.patch.dict( os.environ, { 'SLURM_PROCID': '1', 'SLURM_NTASKS': '5', '

mock.patch.dict( os.environ, { 'SLURM_PROCID': '1', 'SLURM_NTASKS': '5', '

test_util.enable_output_all_intermediates

test_util.enable_output_all_intermediates

test.mock.patch.object(saver_module, "time")

mock.patch.object(resolver.TPUClusterResolver, '_request_compute_metadata', mock_request_

mock.patch.object(eager.context, "list_devices")

mock.patch.object(eager.context, 'list_devices')

mock.patch.object(eager.context, "list_devices")

mock.patch.object(eager.context, 'list_devices')

mock.patch.object(eager.context, 'list_devices')

mock.patch.object(eager.context, "list_devices")

combinations.generate( combinations.combine( mode=["graph", "eager"], input_type=["i

combinations.generate( combinations.combine( mode=["graph"], input_type=["input_fn",

mock.patch.object(resolver.TPUClusterResolver, '_request_compute_metadata', mock_request_

test_util.enable_output_all_intermediates

test_util.enable_output_all_intermediates