View source on GitHub

Represents the composition of a set of Layers.

Inherits From: Layer

Deprecated. Please inherit from tf.keras.Model, and see its documentation for details. tf.keras.Model should be a drop-in replacement for tfe.Network in most cases, but note that track_layer is no longer necessary or supported. Instead, Layer instances are tracked on attribute assignment (see the section of tf.keras.Model's documentation on subclassing). Since the output of track_layer is often assigned to an attribute anyway, most code can be ported by simply removing the track_layer calls.

tf.keras.Model works with all TensorFlow Layer instances, including those from tf.layers, but switching to the tf.keras.layers versions along with the migration to tf.keras.Model is recommended, since it will preserve variable names. Feel free to import it with an alias to avoid excess typing :).

Network implements the Layer interface and adds convenience methods for managing sub-Layers, such as listing variables.

Layers (including other Networks) should be added via track_layer. They can then be used when overriding the method:

class TwoLayerNetwork(tfe.Network):

  def __init__(self, name):
    super(TwoLayerNetwork, self).__init__(name=name)
    self.layer_one = self.track_layer(tf.compat.v1.layers.Dense(16,
    self.layer_two = self.track_layer(tf.compat.v1.layers.Dense(1,

  def call(self, inputs):
    return self.layer_two(self.layer_one(inputs))

After constructing an object and calling the Network, a list of variables created by tracked Layers is available via Network.variables:

net = TwoLayerNetwork(name="net")
output = net(tf.ones([1, 8]))
print([ for v in net.variables])

This example prints variable names, one kernel and one bias per tf.compat.v1.layers.Dense layer:


These variables can be passed to a Saver (tf.compat.v1.train.Saver, or tf.contrib.eager.Saver when executing eagerly) to save or restore the Network, typically alongside a global step and tf.compat.v1.train.Optimizer variables when checkpointing during training.

Note that the semantics of calling a Network with graph execution (i.e. not executing eagerly) may change slightly in the future. Currently stateful ops are pruned from the graph unless they or something that depends on them is executed in a session, but this behavior is not consistent with eager execution (where stateful ops are executed eagerly). Layers from tf.layers do not depend on this pruning and so will not be affected, but Networks which rely on stateful ops being added to the graph but not executed (e.g. via custom Layers which manage stateful ops) may break with this change.

name The name to use for this Network. If specified, it must be unique in the context where this Network is first (1) added to another Network (in which case it must not share a name with other Layers added to that Network), or (2) built/called (in which case no other 'top-level' Networks may share this name). If unspecified or None, the Network will be named using its class name, with a number appended if necessary for uniqueness (e.g. MyNetwork -> 'my_network_1').

ValueError If name is not valid. Note that some naming errors will instead be raised when the Network is called.






View source

Get a contained tf.compat.v1.layers.Layer either by name or index.

name String matching one of the names of a contained Layer. Note that the names of Layers added to Networks may not be unique when doing layer sharing (i.e. adding a Layer to this Network which was already added to another Network). The lowest index Layer with a matching name will be returned.
index Integer in [0, number of layers). Layers are assigned an index by the order they are added.

A tf.compat.v1.layers.Layer object.

ValueError If neither or both of 'index' or 'name' is specified, or the lookup failed.


View source

Track a Layer in this Network.

Network requires that all Layers used in call() be tracked so that the Network can export a complete list of variables.

layer A tf.compat.v1.layers.Layer object.

The passed in layer.

RuntimeError If init has not been called.
TypeError If layer is the wrong type.
ValueError If a Layer with the same name has already been added.