CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers
|
Reference implementation for complex-valued GEMM in host-side code. More...
#include "cutlass/coord.h"
#include "cutlass/complex.h"
#include "cutlass/numeric_types.h"
#include "cutlass/functional.h"
#include "cutlass/numeric_conversion.h"
#include "cutlass/matrix_traits.h"
#include "cutlass/tensor_view.h"
#include "cutlass/gemm/gemm.h"
Go to the source code of this file.
Namespaces | |
cutlass | |
cutlass::reference | |
cutlass::reference::host | |
Functions | |
template<typename ElementA , typename LayoutA , typename ElementB , typename LayoutB , typename ElementC , typename LayoutC , typename ScalarType , typename ComputeType , typename ConvertOp = NumericConverter<ElementC, ScalarType>, typename InnerProductOp = multiply_add<ComputeType>> | |
void | cutlass::reference::host::GemmComplex (gemm::GemmCoord problem_size, ScalarType alpha, TensorRef< ElementA, LayoutA > tensor_a, ComplexTransform transform_a, TensorRef< ElementB, LayoutB > tensor_b, ComplexTransform transform_b, ScalarType beta, TensorRef< ElementC, LayoutC > tensor_c, ComputeType initial_accum) |
template<typename ElementA , typename LayoutA , typename ElementB , typename LayoutB , typename ElementC , typename LayoutC , typename ScalarType > | |
void | cutlass::reference::host::GemmComplex (gemm::GemmCoord problem_size, ScalarType alpha, TensorRef< ElementA, LayoutA > tensor_a, ComplexTransform transform_a, TensorRef< ElementB, LayoutB > tensor_b, ComplexTransform transform_b, ScalarType beta, TensorRef< ElementC, LayoutC > tensor_c) |