RSVP for your your local TensorFlow Everywhere event today!


A conditional accumulator for aggregating gradients.

The accumulator accepts gradients marked with local_step greater or equal to the most recent global_step known to the accumulator. The average can be extracted from the accumulator, provided sufficient gradients have been accumulated. Extracting the average automatically resets the aggregate to 0, and increments the global_step recorded by the accumulator. This is a resource version of ConditionalAccumulator that will work in TF2.0 with tf.cond version 2.

dtype A tf.DType from: tf.float32, tf.float64, tf.int32, tf.uint8, tf.int16, tf.int8, tf.complex64, tf.int64, tf.qint8, tf.quint8, tf.qint32, tf.bfloat16, tf.uint16, tf.complex128, tf.half, tf.uint32, tf.uint64. The type of the value being accumulated.
shape A tf.TensorShape or list of ints. The shape of the values, can be [], in which case shape is unknown.
container An optional string. Defaults to "". If non-empty, this accumulator is placed in the given container. Otherwise, a default container is used.
shared_name An optional string. Defaults to "". If non-empty, this accumulator will be shared under the given name across multiple sessions.
reduction_type An optional string from: "MEAN", "SUM". Defaults to "MEAN".
name A name for the operation (optional).

A Tensor of type resource.