Have a question? Connect with the community at the TensorFlow Forum Visit Forum

SoftsignGrad

public final class SoftsignGrad

Computes softsign gradients for a softsign 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 / (1 + abs(features)) ** 2`.
static <T extends TNumber > SoftsignGrad <T>
create ( Scope scope, Operand <T> gradients, Operand <T> features)
Factory method to create a class wrapping a new SoftsignGrad operation.

Inherited Methods

Constants

public static final String OP_NAME

The name of this op, as known by TensorFlow core engine

Constant Value: "SoftsignGrad"

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 / (1 + abs(features)) ** 2`.

public static SoftsignGrad <T> create ( Scope scope, Operand <T> gradients, Operand <T> features)

Factory method to create a class wrapping a new SoftsignGrad operation.

Parameters
scope current scope
gradients The backpropagated gradients to the corresponding softsign operation.
features The features passed as input to the corresponding softsign operation.
Returns
  • a new instance of SoftsignGrad