An op computes tuple mask of deduplication data from embedding core.
The deduplication data receiving from embedding core is a Tensor with type=DT_VARIANT. The tensor itself is an XLA nested tuple, whose elements are rank 1 tensors. This op is to represents types and length of these elements.
Public Methods
| Output<Integer> |
asOutput()
Returns the symbolic handle of a tensor.
|
| static ComputeDedupDataTupleMask | |
| Output<Integer> |
outputShape()
A 2-D int tensor represent mask of deduplication data tuple generated by
`XlaRecvTPUEmbeddingDeduplicationData`.
|
Inherited Methods
Public Methods
public Output<Integer> asOutput ()
Returns the symbolic handle of a tensor.
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.
public static ComputeDedupDataTupleMask create (Scope scope, String config)
Factory method to create a class wrapping a new ComputeDedupDataTupleMask operation.
Parameters
| scope | current scope |
|---|---|
| config | Serialized TPUEmbeddingConfiguration proto. |
Returns
- a new instance of ComputeDedupDataTupleMask
public Output<Integer> outputShape ()
A 2-D int tensor represent mask of deduplication data tuple generated by `XlaRecvTPUEmbeddingDeduplicationData`. The tuple has several integer and float type 1-D tensor tuple elements. The first dimenion of this output_shape 2-D tensor is tensor type of tuple elements, `0` represents integer tensor, `1` represents float tensor. The second dimension of `output_shape` gives length of each tuple element.