TF 2.0 is out! Get hands-on practice at TF World, Oct 28-31. Use code TF20 for 20% off select passes. Register now

tf.contrib.tpu.infeed_enqueue_tuple

Defined in generated file: python/ops/gen_tpu_ops.py

Feeds multiple Tensor values into the computation as an XLA tuple.

tf.contrib.tpu.infeed_enqueue_tuple(
    inputs,
    shapes,
    layouts=[],
    device_ordinal=-1,
    name=None
)

Args:

  • inputs: A list of Tensor objects. A list of tensors that will be provided using the infeed mechanism.
  • shapes: A list of shapes (each a tf.TensorShape or list of ints). The shapes of each tensor in inputs.
  • layouts: An optional list of ints. Defaults to []. A vector holding the requested layout in minor-to-major sequence for all the tuple shapes, in the order the shapes appear in the "shapes" input. The layout elements for a sub-shape can be set to -1, in which case the corresponding layout will be computed by the infeed operation.
  • device_ordinal: An optional int. Defaults to -1. The TPU device to use. This should be -1 when the Op is running on a TPU device, and >= 0 when the Op is running on the CPU device.
  • name: A name for the operation (optional).

Returns:

The created Operation.