tf.raw_ops.DirectedInterleaveDataset

A substitute for InterleaveDataset on a fixed list of N datasets.

tf.raw_ops.DirectedInterleaveDataset(
    selector_input_dataset, data_input_datasets, output_types, output_shapes,
    name=None
)

Args:

  • selector_input_dataset: A Tensor of type variant. A dataset of scalar DT_INT64 elements that determines which of the N data inputs should produce the next output element.
  • data_input_datasets: A list of at least 1 Tensor objects with type variant. N datasets with the same type that will be interleaved according to the values of selector_input_dataset.
  • output_types: A list of tf.DTypes that has length >= 1.
  • output_shapes: A list of shapes (each a tf.TensorShape or list of ints) that has length >= 1.
  • name: A name for the operation (optional).

Returns:

A Tensor of type variant.