Applies softmax to a batched N-D `SparseTensor`.
The inputs represent an N-D SparseTensor with logical shape `[..., B, C]` (where `N >= 2`), and with indices sorted in the canonical lexicographic order.
This op is equivalent to applying the normal tf.nn.softmax() to each innermost
logical submatrix with shape `[B, C]`, but with the catch that the implicitly
zero elements do not participate. Specifically, the algorithm is equivalent
to the following:
(1) Applies tf.nn.softmax() to a densified view of each innermost submatrix
with shape `[B, C]`, along the size-C dimension;
(2) Masks out the original implicitly-zero locations;
(3) Renormalizes the remaining elements.
Hence, the `SparseTensor` result has exactly the same non-zero indices and shape.
Constants
| String | OP_NAME | The name of this op, as known by TensorFlow core engine |
Public Methods
| Output<T> |
asOutput()
Returns the symbolic handle of the tensor.
|
| static <T extends TNumber> SparseSoftmax<T> | |
| Output<T> |
output()
1-D.
|
Inherited Methods
Constants
public static final String OP_NAME
The name of this op, as known by TensorFlow core engine
Public Methods
public Output<T> asOutput ()
Returns the symbolic handle of the 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 SparseSoftmax<T> create (Scope scope, Operand<TInt64> spIndices, Operand<T> spValues, Operand<TInt64> spShape)
Factory method to create a class wrapping a new SparseSoftmax operation.
Parameters
| scope | current scope |
|---|---|
| spIndices | 2-D. `NNZ x R` matrix with the indices of non-empty values in a SparseTensor, in canonical ordering. |
| spValues | 1-D. `NNZ` non-empty values corresponding to `sp_indices`. |
| spShape | 1-D. Shape of the input SparseTensor. |
Returns
- a new instance of SparseSoftmax