Indicates how a distributed variable will be aggregated.

tf.distribute.Strategy distributes a model by making multiple copies (called "replicas") acting on different elements of the input batch in a data parallel model. When performing some variable-update operation, for example var.assign_add(x), in a model, we need to resolve how to combine the different values for x computed in the different replicas.

  • NONE: This is the default, giving an error if you use a variable-update operation with multiple replicas.
  • SUM: Add the updates across replicas.
  • MEAN: Take the arithmetic mean ("average") of the updates across replicas.
  • ONLY_FIRST_REPLICA: This is for when every replica is performing the same update, but we only want to perform the update once. Used, e.g., for the global step counter.

For example:

strategy = tf.distribute.MirroredStrategy(["GPU:0", "GPU:1"])
with strategy.scope():
  v = tf.Variable(5.0, aggregation=tf.VariableAggregation.MEAN)
def update_fn():
  return v.assign_add(1.0)
  0: <tf.Tensor: shape=(), dtype=float32, numpy=6.0>,
  1: <tf.Tensor: shape=(), dtype=float32, numpy=6.0>
  • ONLY_FIRST_TOWER: Deprecated alias for ONLY_FIRST_REPLICA.

MEAN <VariableAggregation.MEAN: 2>
NONE <VariableAggregation.NONE: 0>
SUM <VariableAggregation.SUM: 1>