Scatter `updates` into an existing tensor according to `indices`.
This operation creates a new tensor by applying sparse `updates` to the passed in `tensor`. This operation is very similar to `tf.scatter_nd`, except that the updates are scattered onto an existing tensor (as opposed to a zero-tensor). If the memory for the existing tensor cannot be re-used, a copy is made and updated.
If `indices` contains duplicates, then their updates are accumulated (summed).
WARNING: The order in which updates are applied is nondeterministic, so the output will be nondeterministic if `indices` contains duplicates -- because of some numerical approximation issues, numbers summed in different order may yield different results.
`indices` is an integer tensor containing indices into a new tensor of shape `shape`. 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 into 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 scatter is to insert individual elements in a tensor by index. For example, say we want to insert 4 scattered elements in a rank-1 tensor with 8 elements.
In Python, this scatter 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) >>> print(tf.tensor_scatter_nd_update(tensor, indices, updates)) tf.Tensor([ 1 11 1 10 9 1 1 12], shape=(8,), dtype=int32)
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 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) >>> print(tf.tensor_scatter_nd_update(tensor, indices, updates).numpy()) [[[5 5 5 5] [6 6 6 6] [7 7 7 7] [8 8 8 8]] [[1 1 1 1] [1 1 1 1] [1 1 1 1] [1 1 1 1]] [[5 5 5 5] [6 6 6 6] [7 7 7 7] [8 8 8 8]] [[1 1 1 1] [1 1 1 1] [1 1 1 1] [1 1 1 1]]]
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.
Public Methods
Output<T> |
asOutput()
Returns the symbolic handle of a tensor.
|
static <T, U extends Number> TensorScatterUpdate<T> | |
Output<T> |
output()
A new tensor with the given shape and updates applied 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 TensorScatterUpdate<T> create (Scope scope, Operand<T> tensor, Operand<U> indices, Operand<T> updates)
Factory method to create a class wrapping a new TensorScatterUpdate operation.
Parameters
scope | current scope |
---|---|
tensor | Tensor to copy/update. |
indices | Index tensor. |
updates | Updates to scatter into output. |
Returns
- a new instance of TensorScatterUpdate
public Output<T> output ()
A new tensor with the given shape and updates applied according to the indices.