GEMM  General matrixmatrix multiplication¶

pyclblas.
clblasCgemm
(order, transA, transB, M, N, K, alpha, A, offA, lda, B, offB, ldb, beta, C, offC, ldc, commandQueues, eventWaitList)¶ wraps: clblasCgemm
Matrixmatrix product of general rectangular matrices with float complex elements. Extended version. <dl class=”section note”> <dt> Note </dt> <dd> This function is not threadsafe. </dd> </dl> Matrixmatrix products:
 ( C ← α A B + β C )
 ( C ← α A^{T}B + β C )
 ( C ← α A B^{T}+ β C )
 ( C ← α A^{T}B^{T}+ β C )
Parameters:  order (clblasOrder [in]) – Row/column order.
 transA (clblasTranspose [in]) – How matrix A is to be transposed.
 transB (clblasTranspose [in]) – How matrix B is to be transposed.
 M (int [in]) – Number of rows in matrix A.
 N (int [in]) – Number of columns in matrix B.
 K (int [in]) – Number of columns in matrix A and rows in matrix B.
 alpha (complex [in]) – The factor of matrix A.
 A (pyopencl.Buffer [in]) – Buffer object storing matrix A.
 offA (int [in]) – Offset of the first element of the matrix A in the buffer object. Counted in elements.
 lda (int [in]) – Leading dimension of matrix A.
 B (pyopencl.Buffer [in]) – Buffer object storing matrix B.
 offB (int [in]) – Offset of the first element of the matrix B in the buffer object. Counted in elements.
 ldb (int [in]) – Leading dimension of matrix B.
 beta (complex [in]) – The factor of matrix C.
 C (pyopencl.Buffer [out]) – Buffer object storing matrix C.
 offC (int [in]) – Offset of the first element of the matrix C in the buffer object. Counted in elements.
 ldc (int [in]) – Leading dimension of matrix C.
 commandQueues (pyopencl.CommandQueue [in]) – OpenCL command queues. A list, tuple, or single instance of pyopencl.CommandQueue. Must not be None.
 eventWaitList (pyopencl.Event [in]) – Event wait list. A list, tuple, or single instance of pyopencl.Event. May be None.
Returns: A tuple of pyopencl.Event instances, one for each commandQueue supplied.

pyclblas.
clblasDgemm
(order, transA, transB, M, N, K, alpha, A, offA, lda, B, offB, ldb, beta, C, offC, ldc, commandQueues, eventWaitList)¶ wraps: clblasDgemm
Matrixmatrix product of general rectangular matrices with double elements. Extended version. <dl class=”section note”> <dt> Note </dt> <dd> This function is not threadsafe. </dd> </dl> Matrixmatrix products:
 ( C ← α A B + β C )
 ( C ← α A^{T}B + β C )
 ( C ← α A B^{T}+ β C )
 ( C ← α A^{T}B^{T}+ β C )
Parameters:  order (clblasOrder [in]) – Row/column order.
 transA (clblasTranspose [in]) – How matrix A is to be transposed.
 transB (clblasTranspose [in]) – How matrix B is to be transposed.
 M (int [in]) – Number of rows in matrix A.
 N (int [in]) – Number of columns in matrix B.
 K (int [in]) – Number of columns in matrix A and rows in matrix B.
 alpha (float [in]) – The factor of matrix A.
 A (pyopencl.Buffer [in]) – Buffer object storing matrix A.
 offA (int [in]) – Offset of the first element of the matrix A in the buffer object. Counted in elements.
 lda (int [in]) – Leading dimension of matrix A.
 B (pyopencl.Buffer [in]) – Buffer object storing matrix B.
 offB (int [in]) – Offset of the first element of the matrix B in the buffer object. Counted in elements.
 ldb (int [in]) – Leading dimension of matrix B.
 beta (float [in]) – The factor of matrix C.
 C (pyopencl.Buffer [out]) – Buffer object storing matrix C.
 offC (int [in]) – Offset of the first element of the matrix C in the buffer object. Counted in elements.
 ldc (int [in]) – Leading dimension of matrix C.
 commandQueues (pyopencl.CommandQueue [in]) – OpenCL command queues. A list, tuple, or single instance of pyopencl.CommandQueue. Must not be None.
 eventWaitList (pyopencl.Event [in]) – Event wait list. A list, tuple, or single instance of pyopencl.Event. May be None.
Returns: A tuple of pyopencl.Event instances, one for each commandQueue supplied.

pyclblas.
clblasSgemm
(order, transA, transB, M, N, K, alpha, A, offA, lda, B, offB, ldb, beta, C, offC, ldc, commandQueues, eventWaitList)¶ wraps: clblasSgemm
Matrixmatrix product of general rectangular matrices with float elements. Extended version. <dl class=”section note”> <dt> Note </dt> <dd> This function is not threadsafe. </dd> </dl> Matrixmatrix products:
 ( C ← α A B + β C )
 ( C ← α A^{T}B + β C )
 ( C ← α A B^{T}+ β C )
 ( C ← α A^{T}B^{T}+ β C )
Parameters:  order (clblasOrder [in]) – Row/column order.
 transA (clblasTranspose [in]) – How matrix A is to be transposed.
 transB (clblasTranspose [in]) – How matrix B is to be transposed.
 M (int [in]) – Number of rows in matrix A.
 N (int [in]) – Number of columns in matrix B.
 K (int [in]) – Number of columns in matrix A and rows in matrix B.
 alpha (float [in]) – The factor of matrix A.
 A (pyopencl.Buffer [in]) – Buffer object storing matrix A.
 offA (int [in]) – Offset of the first element of the matrix A in the buffer object. Counted in elements.
 lda (int [in]) – Leading dimension of matrix A. It cannot be less than K when the order parameter is set to clblasRowMajor, or less than M when the parameter is set to clblasColumnMajor.
 B (pyopencl.Buffer [in]) – Buffer object storing matrix B.
 offB (int [in]) – Offset of the first element of the matrix B in the buffer object. Counted in elements.
 ldb (int [in]) – Leading dimension of matrix B. It cannot be less than N when the order parameter is set to clblasRowMajor, or less than K when it is set to clblasColumnMajor.
 beta (float [in]) – The factor of matrix C.
 C (pyopencl.Buffer [out]) – Buffer object storing matrix C.
 offC (int [in]) – Offset of the first element of the matrix C in the buffer object. Counted in elements.
 ldc (int [in]) – Leading dimension of matrix C. It cannot be less than N when the order parameter is set to clblasRowMajor, or less than M when it is set to clblasColumnMajorOrder.
 commandQueues (pyopencl.CommandQueue [in]) – OpenCL command queues. A list, tuple, or single instance of pyopencl.CommandQueue. Must not be None.
 eventWaitList (pyopencl.Event [in]) – Event wait list. A list, tuple, or single instance of pyopencl.Event. May be None.
Returns: A tuple of pyopencl.Event instances, one for each commandQueue supplied.

pyclblas.
clblasZgemm
(order, transA, transB, M, N, K, alpha, A, offA, lda, B, offB, ldb, beta, C, offC, ldc, commandQueues, eventWaitList)¶ wraps: clblasZgemm
Matrixmatrix product of general rectangular matrices with double complex elements. Exteneded version. <dl class=”section note”> <dt> Note </dt> <dd> This function is not threadsafe. </dd> </dl> Matrixmatrix products:
 ( C ← α A B + β C )
 ( C ← α A^{T}B + β C )
 ( C ← α A B^{T}+ β C )
 ( C ← α A^{T}B^{T}+ β C )
Parameters:  order (clblasOrder [in]) – Row/column order.
 transA (clblasTranspose [in]) – How matrix A is to be transposed.
 transB (clblasTranspose [in]) – How matrix B is to be transposed.
 M (int [in]) – Number of rows in matrix A.
 N (int [in]) – Number of columns in matrix B.
 K (int [in]) – Number of columns in matrix A and rows in matrix B.
 alpha (complex [in]) – The factor of matrix A.
 A (pyopencl.Buffer [in]) – Buffer object storing matrix A.
 offA (int [in]) – Offset of the first element of the matrix A in the buffer object. Counted in elements.
 lda (int [in]) – Leading dimension of matrix A.
 B (pyopencl.Buffer [in]) – Buffer object storing matrix B.
 offB (int [in]) – Offset of the first element of the matrix B in the buffer object. Counted in elements.
 ldb (int [in]) – Leading dimension of matrix B.
 beta (complex [in]) – The factor of matrix C.
 C (pyopencl.Buffer [out]) – Buffer object storing matrix C.
 offC (int [in]) – Offset of the first element of the matrix C in the buffer object. Counted in elements.
 ldc (int [in]) – Leading dimension of matrix C.
 commandQueues (pyopencl.CommandQueue [in]) – OpenCL command queues. A list, tuple, or single instance of pyopencl.CommandQueue. Must not be None.
 eventWaitList (pyopencl.Event [in]) – Event wait list. A list, tuple, or single instance of pyopencl.Event. May be None.
Returns: A tuple of pyopencl.Event instances, one for each commandQueue supplied.