mindspore.ops.csr_add

mindspore.ops.csr_add(a, b, alpha, beta)[source]

Computes the linear combination of two input CSRTensors a and b.

Warning

This interface is deprecated and will be removed after version 2.9.0.

\[out = alpha * a + beta * b\]

where both \(a\) and \(b\) are CSRTensor, \(alpha\) and \(beta\) are both Tensor.

Note

The user needs to ensure that the input sparse matrix is legal. Otherwise, the behavior of the operator is undefined. For example, when there are multiple elements in the same position, the operator may return an error or fail to execute.

Parameters

  • a (CSRTensor) – Input sparse CSRTensor.

  • b (CSRTensor) – Input sparse CSRTensor.

  • alpha (Tensor) – Dense Tensor, its shape must be able to broadcast to a.

  • beta (Tensor) – Dense Tensor, its shape must be able to broadcast to b.

Returns

CSRTensor, a CSRTensor containing the following parts.

  • indptr - Indicates the start and end point for non-zero values in each row.

  • indices - The column positions of all non-zero values of the input.

  • values - The non-zero values.

  • shape - The shape of the CSRTensor.

Supported Platforms:

GPU CPU

Examples

>>> import mindspore.common.dtype as mstype
>>> from mindspore import Tensor, CSRTensor
>>> from mindspore import ops
>>> a_indptr = Tensor([0, 1, 2], dtype=mstype.int32)
>>> a_indices = Tensor([0, 1], dtype=mstype.int32)
>>> a_values = Tensor([1, 2], dtype=mstype.float32)
>>> shape = (2, 6)
>>> b_indptr = Tensor([0, 1, 2], dtype=mstype.int32)
>>> b_indices = Tensor([0, 1], dtype=mstype.int32)
>>> b_values = Tensor([1, 2], dtype=mstype.float32)
>>> alpha = Tensor(1, mstype.float32)
>>> beta = Tensor(1, mstype.float32)
>>> csra = CSRTensor(a_indptr, a_indices, a_values, shape)
>>> csrb = CSRTensor(b_indptr, b_indices, b_values, shape)
>>> out = ops.csr_add(csra, csrb, alpha, beta)
>>> print(out)
CSRTensor(shape=[2, 6], dtype=Float32,                   indptr=Tensor(shape=[3], dtype=Int32, value=[0 1 2]),                   indices=Tensor(shape=[2], dtype=Int32, value=[0 1]),                   values=Tensor(shape=[2], dtype=Float32, value=[ 2.00000000e+00  4.00000000e+00]))