|TensorFlow 1 version||View source on GitHub|
A layer for sequence input.
Compat aliases for migration
See Migration guide for more details.
tf.keras.experimental.SequenceFeatures( feature_columns, trainable=True, name=None, **kwargs )
feature_columns must be sequence dense columns with the same
sequence_length. The output of this method can be fed into sequence
networks, such as RNN.
The output of this method is a 3D
Tensor of shape
[batch_size, T, D].
T is the maximum sequence length for this batch, which could differ from
batch to batch.
feature_columns are given with
num_elements each, their
outputs are concatenated. So, the final
Tensor has shape
[batch_size, T, D0 + D1 + ... + Dn].
# Behavior of some cells or feature columns may depend on whether we are in # training or inference mode, e.g. applying dropout. training = True rating = sequence_numeric_column('rating') watches = sequence_categorical_column_with_identity( 'watches', num_buckets=1000) watches_embedding = embedding_column(watches, dimension=10) columns = [rating, watches_embedding] sequence_input_layer = SequenceFeatures(columns) features = tf.io.parse_example(..., features=make_parse_example_spec(columns)) sequence_input, sequence_length = sequence_input_layer( features, training=training) sequence_length_mask = tf.sequence_mask(sequence_length) rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size, training=training) rnn_layer = tf.keras.layers.RNN(rnn_cell, training=training) outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
An iterable of dense sequence columns. Valid columns are
||Boolean, whether the layer's variables will be updated via gradient descent during training.|
||Name to give to the SequenceFeatures.|
||Keyword arguments to construct a layer.|
If any of the