HCmatinvBatched

Applicable Products

Function Description

API function

• asdBlasMakeHCmatinvBatchedPlan: initializes the operator configuration corresponding to the handle.
• asdBlasHCmatinvBatched: computes the inverse of a complex matrix.

Formula:

• Computation of asdBlasHCmatinvBatched: An inverse matrix of a batch complex matrix is computed. Each complex matrix needs to meet the condition , where A is a non-singular square matrix and an n*n input square matrix, and I is an identity matrix.

• Example:

  Input A:

  [2-2i ,1-i ,1-i, 1-i
  1-i, 2-2i, 1-i ,1-i
  1-i, 1-i ,2-2i ,1-i
  1-i ,1-i, 1-i, 2-2i]
  
  [3-3i ,1-i, 1-i, 1-i
  1-i, 3-3i, 1-i ,1-i
  1-i, 1-i ,3-3i ,1-i
  1-i, 1-i, 1-i, 3-3i]

  Input n: 4

  Input batchSize: 2

  After the asdBlasHCmatinvBatched operator is called, the output Ainv is as follows:

  [0.4+0.4i, -0.1-0.1i,-0.1-0.1i ,-0.1-0.1i
  -0.1-0.1i ,0.4+0.4i ,-0.1-0.1i ,-0.1-0.1i
  -0.1-0.1i ,-0.1-0.1i,0.4+0.4i ,-0.1-0.1i
  -0.1-0.1i ,-0.1-0.1i,-0.1-0.1i,0.4+0.4i]
  
  [0.208+0.208i,-0.0417-0.0417i,-0.0417-0.0417i,-0.0417-0.0417i
  -0.0417-0.0417i,0.208+0.208i,-0.0417-0.0417i,-0.0417-0.0417i
  -0.0417-0.0417i,-0.0417-0.0417i,0.208+0.208i,-0.0417-0.0417i
  -0.0417-0.0417i,-0.0417-0.0417i,-0.0417-0.0417i,0.208+0.208i]

Function Prototype

• AspbStatus asdBlasMakeHCmatinvBatchedPlan(asdBlasHandle handle, const int64_t n, const int64_t batchSize)
• AspbStatus asdBlasHCmatinvBatched(asdBlasHandle handle, const int64_t n, aclTensor *A, const int64_t lda,aclTensor *Ainv, const int64_t lda_inv, aclTensor *info, int64_t batchSize)

Parameter Description

• asdBlasMakeHCmatinvBatchedPlan

  Parameter	Input/Output	Type	Description
  handle	Input	asdBlasHandle	Operator handle.
  n	Input	const int64_t	Number of rows in matrix A in a single batch.
  batchSize	Input	const int64_t	Number of matrices in the complex matrix inversion.

• asdBlasHCmatinvBatched

  Parameter	Input/Output	Type	Description
  handle	Input	asdBlasHandle	Operator handle.
  n	Input	const int64_t	Number of rows in matrix A.
  A	Input	aclTensor *	Matrix A in the formula, which is in row-major order and is a tensor on the device. The data type is COMPLEX32, the data format is ND, and the shape is [batch, n, n].
  lda	Input	const int64_t	Memory address offset between the left and right adjacent elements of A (currently, the offset is n).
  Ainv	Output	aclTensor *	Output inverse matrix. The data type can only be COMPLEX32, the data format can be ND, and the shape is [batch, n, n].
  lda_inv	Input	const int64_t	Memory address offset between the left and right adjacent elements of the output inverse matrix (currently, the offset is n).
  info	Input	aclTensor *	Information about the inverse result of each batch matrix. The data type can only be int32_t, the data format can be ND, and the shape is [batch, 1].
  batchSize	Input	int64_t	Number of matrices in the complex matrix inversion.

Return Value Description

For details about the return values, see Return Value.

Constraints

• asdBlasMakeHCmatinvBatchedPlan: none.
• asdBlasHCmatinvBatched
  • The supported CANN version is CANN 8.0 or later.
  • The lda, lda_inv, and info parameters are not enabled in the current version.
  • The input parameter n is less than or equal to 256.
  • The input parameter batchSize is less than or equal to 3000.

Calling Example

For details about the operator calling example, see HCmatinvBatched.