tf.raw_ops.AccumulatorTakeGradient

Extracts the average gradient in the given ConditionalAccumulator.

View aliases

Compat aliases for migration

See Migration guide for more details.

tf.compat.v1.raw_ops.AccumulatorTakeGradient

tf.raw_ops.AccumulatorTakeGradient(
    handle, num_required, dtype, name=None
)

The op blocks until sufficient (i.e., more than num_required) gradients have been accumulated. If the accumulator has already aggregated more than num_required gradients, it returns the average of the accumulated gradients. Also automatically increments the recorded global_step in the accumulator by 1, and resets the aggregate to 0.

Args
handle	A Tensor of type mutable string. The handle to an accumulator.
num_required	A Tensor of type int32. Number of gradients required before we return an aggregate.
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 data type of accumulated gradients. Needs to correspond to the type of the accumulator.
name	A name for the operation (optional).

Returns
A Tensor of type dtype.