scatter_nd_add¶

paddle. scatter_nd_add ( x, index, updates, name=None ) [source]

Output is obtained by applying sparse addition to a single value or slice in a Tensor.

x is a Tensor with ndim \(R\) and index is a Tensor with ndim \(K\) . Thus, index has shape \([i_0, i_1, ..., i_{K-2}, Q]\) where \(Q \leq R\) . updates is a Tensor with ndim \(K - 1 + R - Q\) and its shape is \(index.shape[:-1] + x.shape[index.shape[-1]:]\) .

According to the \([i_0, i_1, ..., i_{K-2}]\) of index , add the corresponding updates slice to the x slice which is obtained by the last one dimension of index .

           Given:

* Case 1:
    x = [0, 1, 2, 3, 4, 5]
    index = [[1], [2], [3], [1]]
    updates = [9, 10, 11, 12]

  we get:

    output = [0, 22, 12, 14, 4, 5]

* Case 2:
    x = [[65, 17], [-14, -25]]
    index = [[], []]
    updates = [[[-1, -2], [1, 2]],
               [[3, 4], [-3, -4]]]
    x.shape = (2, 2)
    index.shape = (2, 0)
    updates.shape = (2, 2, 2)

  we get:

    output = [[67, 19], [-16, -27]]

          

Parameters

x (Tensor) – The x input. Its dtype should be int32, int64, float32, float64.
index (Tensor) – The index input with ndim > 1 and index.shape[-1] <= x.ndim. Its dtype should be int32 or int64 as it is used as indexes.
updates (Tensor) – The updated value of scatter_nd_add op, and it must have the same dtype as x. It must have the shape index.shape[:-1] + x.shape[index.shape[-1]:].
name (str|None) – The output tensor name. If set None, the layer will be named automatically.

Returns

The output is a tensor with the same shape and dtype as x.

Return type

output (Tensor)

Examples

           import paddle

x = paddle.rand(shape=[3, 5, 9, 10], dtype='float32')
updates = paddle.rand(shape=[3, 9, 10], dtype='float32')
index = paddle.to_tensor([[1, 1],
                        [0, 1],
                        [1, 3]], dtype='int64')

output = paddle.scatter_nd_add(x, index, updates)
print(output.shape)
# [3, 5, 9, 10]

          