tf.keras.layers.TimeDistributed

TensorFlow 1 version View source on GitHub

This wrapper allows to apply a layer to every temporal slice of an input.

Inherits From: Wrapper

The input should be at least 3D, and the dimension of index one will be considered to be the temporal dimension.

Consider a batch of 32 samples, where each sample is a sequence of 10 vectors of 16 dimensions. The batch input shape of the layer is then (32, 10, 16), and the input_shape, not including the samples dimension, is (10, 16).

You can then use TimeDistributed to apply a Dense layer to each of the 10 timesteps, independently:

# as the first layer in a model
model = Sequential()
model.add(TimeDistributed(Dense(8), input_shape=(10, 16)))
# now model.output_shape == (None, 10, 8)

The output will then have shape (32, 10, 8).

In subsequent layers, there is no need for the input_shape:

model.add(TimeDistributed(Dense(32)))
# now model.output_shape == (None, 10, 32)

The output will then have shape (32, 10, 32).

TimeDistributed can be used with arbitrary layers, not just Dense, for instance with a Conv2D layer:

model = Sequential()
model.add(TimeDistributed(Conv2D(64, (3, 3)),
                          input_shape=(10, 299, 299, 3)))

layer a layer instance.

Call arguments:

  • inputs: Input tensor.
  • training: Python boolean indicating whether the layer should behave in training mode or in inference mode. This argument is passed to the wrapped layer (only if the layer supports this argument).
  • mask: Binary tensor of shape (samples, timesteps) indicating whether a given timestep should be masked. This argument is passed to the wrapped layer (only if the layer supports this argument).

ValueError If not initialized with a Layer instance.