# ScatterNd

public final class **ScatterNd** Scatters \`updates\` into a tensor of shape \`shape\` according to \`indices\`.


Scatter sparse \`updates\` according to individual values at the specified
\`indices\`. This op returns an output tensor with the \`shape\` you specify. This
op is the inverse of the [`tf.gather_nd`](https://www.tensorflow.org/api_docs/python/tf/gather_nd) operator which extracts values or slices
from a given tensor.


This operation is similar to [`tf.tensor_scatter_nd_add`](https://www.tensorflow.org/api_docs/python/tf/tensor_scatter_nd_add), except that the tensor
is zero-initialized. Calling [`tf.scatter_nd(indices, updates, shape)`](https://www.tensorflow.org/api_docs/python/tf/scatter_nd)
is identical to calling
\`tf.tensor_scatter_nd_add(tf.zeros(shape, updates.dtype), indices, updates)\`


If \`indices\` contains duplicates, the associated \`updates\` are accumulated
(summed) into the output tensor.


**WARNING** : For floating-point data types, the output may be nondeterministic.
This is because the order in which the updates are applied is nondeterministic
and when floating-point numbers are added in different orders the resulting
numerical approximation error can be slightly different. However, the output
will be deterministic if op determinism is enabled via
[`tf.config.experimental.enable_op_determinism`](https://www.tensorflow.org/api_docs/python/tf/config/experimental/enable_op_determinism).


\`indices\` is an integer tensor containing indices into the output tensor. The
last dimension of \`indices\` can be at most the rank of \`shape\`:


indices.shape\[-1\] \<= shape.rank


The last dimension of \`indices\` corresponds to indices of elements
(if \`indices.shape\[-1\] = shape.rank\`) or slices
(if \`indices.shape\[-1\] \< shape.rank\`) along dimension \`indices.shape\[-1\]\` of
\`shape\`.


\`updates\` is a tensor with shape:


indices.shape\[:-1\] + shape\[indices.shape\[-1\]:\]


The simplest form of the scatter op is to insert individual elements in
a tensor by index. Consider an example where you want to insert 4 scattered
elements in a rank-1 tensor with 8 elements.


![](https://www.tensorflow.org/images/ScatterNd1.png)


In Python, this scatter operation would look like this:

    indices = tf.constant([[4], [3], [1], [7]])
         updates = tf.constant([9, 10, 11, 12])
         shape = tf.constant([8])
         scatter = tf.scatter_nd(indices, updates, shape)
         print(scatter)
     
The resulting tensor would look like this:


\[0, 11, 0, 10, 9, 0, 0, 12\]


You can also insert entire slices of a higher rank tensor all at once. For
example, you can insert two slices in the first dimension of a rank-3 tensor
with two matrices of new values.


![](https://www.tensorflow.org/images/ScatterNd2.png)


In Python, this scatter operation would look like this:

    indices = tf.constant([[1], [3]])
         updates = tf.constant([[[5, 5, 5, 5], [6, 6, 6, 6],
                                 [7, 7, 7, 7], [8, 8, 8, 8]],
                                [[5, 5, 5, 5], [6, 6, 6, 6],
                                 [7, 7, 7, 7], [8, 8, 8, 8]]])
         shape = tf.constant([4, 4, 4])
         scatter = tf.scatter_nd(indices, updates, shape)
         print(scatter)
     
The resulting tensor would look like this:


\[\[\[0, 0, 0, 0\], \[0, 0, 0, 0\], \[0, 0, 0, 0\], \[0, 0, 0, 0\]\],
\[\[5, 5, 5, 5\], \[6, 6, 6, 6\], \[7, 7, 7, 7\], \[8, 8, 8, 8\]\],
\[\[0, 0, 0, 0\], \[0, 0, 0, 0\], \[0, 0, 0, 0\], \[0, 0, 0, 0\]\],
\[\[5, 5, 5, 5\], \[6, 6, 6, 6\], \[7, 7, 7, 7\], \[8, 8, 8, 8\]\]\]


Note that on CPU, if an out of bound index is found, an error is returned.
On GPU, if an out of bound index is found, the index is ignored.

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

<br />

### Public Methods

|---|---|
| [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<U\> | [asOutput](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/ScatterNd#asOutput())() Returns the symbolic handle of a tensor. |
| static \<U, T extends Number\> [ScatterNd](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/ScatterNd)\<U\> | [create](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/ScatterNd#create(org.tensorflow.op.Scope,%20org.tensorflow.Operand<T>,%20org.tensorflow.Operand<U>,%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\> indices, [Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)\<U\> updates, [Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)\<T\> shape) Factory method to create a class wrapping a new ScatterNd operation. |
| [Output](https://www.tensorflow.org/api_docs/java/org/tensorflow/Output)\<U\> | [output](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/ScatterNd#output())() A new tensor with the given shape and updates applied according to the indices. |

### 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)\<U\> | [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)\<U\>
**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 [ScatterNd](https://www.tensorflow.org/api_docs/java/org/tensorflow/op/core/ScatterNd)\<U\>
**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\> indices, [Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)\<U\> updates, [Operand](https://www.tensorflow.org/api_docs/java/org/tensorflow/Operand)\<T\> shape)

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

##### Parameters

| scope | current scope |
| indices | Tensor of indices. |
| updates | Values to scatter into the output tensor. |
| shape | 1-D. The shape of the output tensor. |
|---|---|

##### Returns

- a new instance of ScatterNd

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

A new tensor with the given shape and updates applied according
to the indices.