# StopGradient

public final class **StopGradient** Stops gradient computation.


When executed in a graph, this op outputs its input tensor as-is.


When building ops to compute gradients, this op prevents the contribution of
its inputs to be taken into account. Normally, the gradient generator adds ops
to a graph to compute the derivatives of a specified 'loss' by recursively
finding out inputs that contributed to its computation. If you insert this op
in the graph it inputs are masked from the gradient generator. They are not
taken into account for computing gradients.


This is useful any time you want to compute a value with TensorFlow but need
to pretend that the value was a constant. For example, the softmax function
for a vector x can be written as

    def softmax(x):
         numerator = tf.exp(x)
         denominator = tf.reduce_sum(numerator)
         return numerator / denominator
     
This however is susceptible to overflow if the values in x are large. An alternative more stable way is to subtract the maximum of x from each of the values.

    def stable_softmax(x):
         z = x - tf.reduce_max(x)
         numerator = tf.exp(z)
         denominator = tf.reduce_sum(numerator)
         return numerator / denominator
     
However, when we backprop through the softmax to x, we dont want to backprop through the [`tf.reduce_max(x)`](https://www.tensorflow.org/api_docs/python/tf/math/reduce_max) (if the max values are not unique then the gradient could flow to the wrong input) calculation and treat that as a constant. Therefore, we should write this out as

    def stable_softmax(x):
         z = x - tf.stop_gradient(tf.reduce_max(x))
         numerator = tf.exp(z)
         denominator = tf.reduce_sum(numerator)
         return numerator / denominator
     
Some other examples include:

- The *EM* algorithm where the *M-step* should not involve backpropagation through the output of the *E-step*.
- Contrastive divergence training of Boltzmann machines where, when differentiating the energy function, the training must not backpropagate through the graph that generated the samples from the model.
- Adversarial training, where no backprop should happen through the adversarial example generation process.

<br />

<br />

<br />

<br />

### Public Methods

|---|---|
| [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<T\> | [asOutput](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/StopGradient#asOutput())() Returns the symbolic handle of a tensor. |
| static \<T\> [StopGradient](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/StopGradient)\<T\> | [create](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/StopGradient#create(org.tensorflow.op.Scope,%20org.tensorflow.Operand<T>))([Scope](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/Scope) scope, [Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)\<T\> input) Factory method to create a class wrapping a new StopGradient operation. |
| [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<T\> | [output](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/StopGradient#output())() |

### Inherited Methods

From class [org.tensorflow.op.PrimitiveOp](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/PrimitiveOp)

|---|---|
| final boolean | [equals](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/PrimitiveOp#equals(java.lang.Object))(Object obj) |
| final int | [hashCode](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/PrimitiveOp#hashCode())() |
| [Operation](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operation) | [op](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/PrimitiveOp#op())() Returns the underlying `https://www.tensorflow.org/api_docs/java/org/tensorflow/Operation` |
| final String | [toString](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/PrimitiveOp#toString())() |

From class java.lang.Object

|---|---|
| boolean | equals(Object arg0) |
| final Class\<?\> | getClass() |
| int | hashCode() |
| final void | notify() |
| final void | notifyAll() |
| String | toString() |
| final void | wait(long arg0, int arg1) |
| final void | wait(long arg0) |
| final void | wait() |

From interface [org.tensorflow.Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)

|---|---|
| abstract [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<T\> | [asOutput](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand#asOutput())() Returns the symbolic handle of a tensor. |

## Public Methods

#### public [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<T\>
**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.

<br />

#### public static [StopGradient](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/StopGradient)\<T\>
**create**
([Scope](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/Scope) scope, [Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)\<T\> input)

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

##### Parameters

| scope | current scope |
|---|---|

##### Returns

- a new instance of StopGradient

#### public [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<T\>
**output**
()

<br />