Adds sparse `updates` to an existing tensor according to `indices`.
This operation creates a new tensor by adding sparse `updates` to the passed
in `tensor`.
This operation is very similar to tf.compat.v1.scatter_nd_add
, except that the
updates are added onto an existing tensor (as opposed to a variable). If the
memory for the existing tensor cannot be re-used, a copy is made and updated.
`indices` is an integer tensor containing indices into a new tensor of shape `tensor.shape`. The last dimension of `indices` can be at most the rank of `tensor.shape`:
indices.shape[-1] <= tensor.shape.rank
indices.shape[:-1] + tensor.shape[indices.shape[-1]:]
In Python, this scatter add operation would look like this:
>>> indices = tf.constant([[4], [3], [1], [7]])
>>> updates = tf.constant([9, 10, 11, 12])
>>> tensor = tf.ones([8], dtype=tf.int32)
>>> updated = tf.tensor_scatter_nd_add(tensor, indices, updates)
>>> updated
We can also, insert entire slices of a higher rank tensor all at once. For
example, if we wanted to insert two slices in the first dimension of a
rank-3 tensor with two matrices of new values.
In Python, this scatter add operation would look like this:
>>> indices = tf.constant([[0], [2]])
>>> 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]]])
>>> tensor = tf.ones([4, 4, 4],dtype=tf.int32)
>>> updated = tf.tensor_scatter_nd_add(tensor, indices, updates)
>>> updated
Note: 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.
Public Methods
Output<T> |
asOutput()
Returns the symbolic handle of a tensor.
|
static <T, U extends Number> TensorScatterAdd<T> | |
Output<T> |
output()
A new tensor copied from tensor and updates added according to the indices.
|
Inherited Methods
Public Methods
public 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.
public static TensorScatterAdd<T> create (Scope scope, Operand<T> tensor, Operand<U> indices, Operand<T> updates)
Factory method to create a class wrapping a new TensorScatterAdd operation.
Parameters
scope | current scope |
---|---|
tensor | Tensor to copy/update. |
indices | Index tensor. |
updates | Updates to scatter into output. |
Returns
- a new instance of TensorScatterAdd
public Output<T> output ()
A new tensor copied from tensor and updates added according to the indices.