Watch talks from the 2019 TensorFlow Dev Summit Watch now

tf.contrib.tpu.repeat

tf.contrib.tpu.repeat(
    n,
    body,
    inputs=None,
    infeed_queue=None,
    name=None
)

Defined in tensorflow/contrib/tpu/python/tpu/training_loop.py.

Builds a training loop that executes a fixed number of iterations.

The set of loop-carried tensors correspond to inputs. body must be a function that takes and returns the values of the loop-carried tensors.

Args:

  • n: the number of loop iterations
  • body: a Python function that builds the loop body.
  • inputs: a list of initial values passed into the training loop or None (equivalent to an empty list).
  • infeed_queue: if not None, the infeed queue from which to append a tuple of arguments as inputs to condition.
  • name: (Deprecated) Does nothing.

Returns:

The final values of the loop-carried tensors.

Raises:

  • ValueError: if there is a type error.