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

tf.keras.layers.TimeDistributed(
    layer, **kwargs
)

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 video samples, where each sample is a 128x128 RGB image with channels_last data format, across 10 timesteps. The batch input shape is (32, 10, 128, 128, 3).

You can then use TimeDistributed to apply a Conv2D layer to each of the 10 timesteps, independently:

inputs = tf.keras.Input(shape=(10, 128, 128, 3))
conv_2d_layer = tf.keras.layers.Conv2D(64, (3, 3))
outputs = tf.keras.layers.TimeDistributed(conv_2d_layer)(inputs)
outputs.shape
TensorShape([None, 10, 126, 126, 64])

layer a tf.keras.layers.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 tf.keras.layers.Layer instance.