tf.data.experimental.bucket_by_sequence_length

View source on GitHub

A transformation that buckets elements in a Dataset by length. (deprecated)

tf.data.experimental.bucket_by_sequence_length(
    element_length_func,
    bucket_boundaries,
    bucket_batch_sizes,
    padded_shapes=None,
    padding_values=None,
    pad_to_bucket_boundary=False,
    no_padding=False,
    drop_remainder=False
)

Elements of the Dataset are grouped together by length and then are padded and batched.

This is useful for sequence tasks in which the elements have variable length. Grouping together elements that have similar lengths reduces the total fraction of padding in a batch which increases training step efficiency.

Below is an example to bucketize the input data to the 3 buckets "[0, 3), [3, 5), [5, inf)" based on sequence length, with batch size 2.

elements = [
  [0], [1, 2, 3, 4], [5, 6, 7],
  [7, 8, 9, 10, 11], [13, 14, 15, 16, 19, 20], [21, 22]]

dataset = tf.data.Dataset.from_generator(
    lambda: elements, tf.int64, output_shapes=[None])

dataset = dataset.apply(
    tf.data.experimental.bucket_by_sequence_length(
        element_length_func=lambda elem: tf.shape(elem)[0],
        bucket_boundaries=[3, 5],
        bucket_batch_sizes=[2, 2, 2]))

for elem in dataset.as_numpy_iterator():
  print(elem)
[[1 2 3 4]
 [5 6 7 0]]
[[ 7  8  9 10 11  0]
 [13 14 15 16 19 20]]
[[ 0  0]
 [21 22]]

There is also a possibility to pad the dataset till the bucket boundary. You can also provide which value to be used while padding the data. Below example uses -1 as padding and it also shows the input data being bucketizied to two buckets "[0,3], [4,6]".

elements = [
  [0], [1, 2, 3, 4], [5, 6, 7],
  [7, 8, 9, 10, 11], [13, 14, 15, 16, 19, 20], [21, 22]]

dataset = tf.data.Dataset.from_generator(
  lambda: elements, tf.int32, output_shapes=[None])

dataset = dataset.apply(
    tf.data.experimental.bucket_by_sequence_length(
        element_length_func=lambda elem: tf.shape(elem)[0],
        bucket_boundaries=[4, 7],
        bucket_batch_sizes=[2, 2, 2],
        pad_to_bucket_boundary=True,
        padding_values=-1))

for elem in dataset.as_numpy_iterator():
  print(elem)
[[ 0 -1 -1]
 [ 5  6  7]]
[[ 1  2  3  4 -1 -1]
 [ 7  8  9 10 11 -1]]
[[21 22 -1]]
[[13 14 15 16 19 20]]

When using pad_to_bucket_boundary option, it can be seen that it is not always possible to maintain the bucket batch size. You can drop the batches that do not maintain the bucket batch size by using the option drop_remainder. Using the same input data as in the above example you get the following result.

elements = [
  [0], [1, 2, 3, 4], [5, 6, 7],
  [7, 8, 9, 10, 11], [13, 14, 15, 16, 19, 20], [21, 22]]

dataset = tf.data.Dataset.from_generator(
  lambda: elements, tf.int32, output_shapes=[None])

dataset = dataset.apply(
    tf.data.experimental.bucket_by_sequence_length(
        element_length_func=lambda elem: tf.shape(elem)[0],
        bucket_boundaries=[4, 7],
        bucket_batch_sizes=[2, 2, 2],
        pad_to_bucket_boundary=True,
        padding_values=-1,
        drop_remainder=True))

for elem in dataset.as_numpy_iterator():
  print(elem)
[[ 0 -1 -1]
 [ 5  6  7]]
[[ 1  2  3  4 -1 -1]
 [ 7  8  9 10 11 -1]]

element_length_func function from element in Dataset to tf.int32, determines the length of the element, which will determine the bucket it goes into. bucket_boundaries list<int>, upper length boundaries of the buckets. bucket_batch_sizes list<int>, batch size per bucket. Length should be len(bucket_boundaries) + 1. padded_shapes Nested structure of tf.TensorShape to pass to tf.data.Dataset.padded_batch. If not provided, will use dataset.output_shapes, which will result in variable length dimensions being padded out to the maximum length in each batch. padding_values Values to pad with, passed to tf.data.Dataset.padded_batch. Defaults to padding with 0. pad_to_bucket_boundary bool, if False, will pad dimensions with unknown size to maximum length in batch. If True, will pad dimensions with unknown size to bucket boundary minus 1 (i.e., the maximum length in each bucket), and caller must ensure that the source Dataset does not contain any elements with length longer than max(bucket_boundaries). no_padding bool, indicates whether to pad the batch features (features need to be either of type tf.sparse.SparseTensor or of same shape). drop_remainder (Optional.) A tf.bool scalar tf.Tensor, representing whether the last batch should be dropped in the case it has fewer than batch_size elements; the default behavior is not to drop the smaller batch.

ValueError if len(bucket_batch_sizes) != len(bucket_boundaries) + 1.