mindspore.ops.tensor_scatter_add

View Source On Gitee
mindspore.ops.tensor_scatter_add(input_x, indices, updates)[source]

Return a new tensor by adding the values from updates in input_x indicated by indices .

\[output\left [indices \right ] = input\_x + update\]

The figure below shows an example of the computational process of tensor_scatter_add:

../../_images/TensorScatterAdd.png

Note

If the values in indices are out of the index range of the input input_x:

  • On GPU, if some values of the indices are out of bound, instead of raising an index error, the corresponding updates will not be updated to self tensor.

  • On CPU, if some values of the indices are out of bound, raising an index error.

  • On Ascend, out of bound checking is not supported, if some values of the indices are out of bound, unknown errors may be caused.

Parameters

  • input_x (Tensor) – The target tensor. The dimension of input_x must be no less than indices.shape[-1].

  • indices (Tensor) – The index of input tensor whose data type is mindspore.int32 or mindspore.int64. The rank must be at least 2.

  • updates (Tensor) – The tensor to update the input tensor, has the same type as input, and updates. Shape should be equal to indices.shape[:-1] + input_x.shape[indices.shape[-1]:].

Returns

Tensor

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> input_x = mindspore.tensor([[-0.1, 0.3, 3.6], [0.4, 0.5, -3.2]], mindspore.float32)
>>> indices = mindspore.tensor([[0, 0], [0, 0]], mindspore.int32)
>>> updates = mindspore.tensor([1.0, 2.2], mindspore.float32)
>>> output = mindspore.ops.tensor_scatter_add(input_x, indices, updates)
>>> print(output)
[[ 3.1  0.3  3.6]
 [ 0.4  0.5 -3.2]]