Extracts the average sparse gradient in a SparseConditionalAccumulator.
The op will 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 will return its average of the accumulated gradients. Also automatically increments the recorded global_step in the accumulator by 1, and resets the aggregate to 0.
Constants
String | OP_NAME | The name of this op, as known by TensorFlow core engine |
Public Methods
static <T extends TType> SparseAccumulatorTakeGradient<T> | |
Output<TInt64> |
indices()
Indices of the average of the accumulated sparse gradients.
|
Output<TInt64> |
shape()
Shape of the average of the accumulated sparse gradients.
|
Output<T> |
values()
Values of the average of the accumulated sparse gradients.
|
Inherited Methods
Constants
public static final String OP_NAME
The name of this op, as known by TensorFlow core engine
Public Methods
public static SparseAccumulatorTakeGradient<T> create (Scope scope, Operand<TString> handle, Operand<TInt32> numRequired, Class<T> dtype)
Factory method to create a class wrapping a new SparseAccumulatorTakeGradient operation.
Parameters
scope | current scope |
---|---|
handle | The handle to a SparseConditionalAccumulator. |
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. |
Returns
- a new instance of SparseAccumulatorTakeGradient