Svd

English | 简体中文

Product Support Status

Product	Supported
Ascend 950PR&950DT products	×
Atlas A3 products	√
Atlas A2 products	√
Atlas 200I/500 A2 inference products	×
Atlas inference products	×
Atlas training products	×

Function Description

• API function: computes the singular value decomposition of one or more matrices.

  When the dimension of the input tensor is greater than 2, the high-dimensional tensor is treated as a batch of matrices. For an input tensor with the shape (..., M, N), the dimensions (...) before the second-to-last dimension are regarded as batch dimensions, and the singular value decomposition is computed independently for each (M, N) matrix.

• Computation formula:

$$\mathbf{i}\mathbf{n}\mathbf{p}\mathbf{u}\mathbf{t}=\mathbf{U}\times \mathrm{d}\mathrm{i}\mathrm{a}\mathrm{g}(\mathbf{s}\mathbf{i}\mathbf{g}\mathbf{m}\mathbf{a})\times {\mathbf{V}}^T$$

Parameter Name	Input/Output/Attribute	Description	Data Type	Data Format
input	Input	Tensor on which the singular value decomposition is to be performed, corresponding to input in the formula. Empty tensors are not supported. The shape has at least 2 dimensions.	FLOAT, DOUBLE, COMPLEX64, COMPLEX128	ND
sigma	Output	Output tensor, corresponding to sigma in the formula. The shape needs to be deduced based on the shape of input and the fullMatrices parameter. The data type must be the same as that of input.	FLOAT, DOUBLE, COMPLEX64, COMPLEX128	ND
U	Output	Output tensor, corresponding to U in the formula. The shape needs to be deduced based on the shape of input and the fullMatrices parameter. The data type must be the same as that of input.	FLOAT, DOUBLE, COMPLEX64, COMPLEX128	ND
V	Output	Output tensor, corresponding to V in the formula. The shape needs to be deduced based on the shape of input and the fullMatrices parameter. The data type must be the same as that of input.	FLOAT, DOUBLE, COMPLEX64, COMPLEX128	ND
compute_uv	Attribute	Optional attribute. The default value is true. Denotes whether to compute the output tensors u and v. If set to true, the output tensors sigma, u, and v are all computed. If set to false, only sigma is computed.	BOOL	true/false
full_matrices	Attribute	Optional attribute. Denotes whether to compute the output tensors u and v completely. The default value is false.	BOOL	true/false

Constraint Description

None

Invocation Description

Invocation Method	Invocation Sample	Description
aclnn invocation	test_aclnn_svd	Invokes the Svd operator through the aclnnSvd API.