T
- data type for average()
output@Operator(group="train") public final class AccumulatorTakeGradient<T> extends PrimitiveOp implements Operand<T>
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.
operation
Modifier and Type | Method and Description |
---|---|
Output<T> |
asOutput()
Returns the symbolic handle of a tensor.
|
Output<T> |
average()
The average of the accumulated gradients.
|
static <T> AccumulatorTakeGradient<T> |
create(Scope scope,
Operand<String> handle,
Operand<Integer> numRequired,
Class<T> dtype)
Factory method to create a class wrapping a new AccumulatorTakeGradient operation.
|
equals, hashCode, op, toString
public static <T> AccumulatorTakeGradient<T> create(Scope scope, Operand<String> handle, Operand<Integer> numRequired, Class<T> dtype)
scope
- current scopehandle
- The handle to an accumulator.numRequired
- Number of gradients required before we return an aggregate.dtype
- The data type of accumulated gradients. Needs to correspond to the type
of the accumulator.public Output<T> asOutput()
Operand
Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.
asOutput
in interface Operand<T>
OperationBuilder.addInput(Output)
Copyright © 2022. All rights reserved.