Computes gradients for the scaled exponential linear (Selu) operation.
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.
|
Output<T> |
backprops()
The gradients: `gradients * (outputs + scale * alpha)`
if outputs < 0, `scale * gradients` otherwise.
|
static <T extends TNumber> SeluGrad<T> |
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 Output<T> backprops ()
The gradients: `gradients * (outputs + scale * alpha)` if outputs < 0, `scale * gradients` otherwise.
public static SeluGrad<T> create (Scope scope, Operand<T> gradients, Operand<T> outputs)
Factory method to create a class wrapping a new SeluGrad operation.
Parameters
scope | current scope |
---|---|
gradients | The backpropagated gradients to the corresponding Selu operation. |
outputs | The outputs of the corresponding Selu operation. |
Returns
- a new instance of SeluGrad