std :: ml::gpu_cuda
Function std::cuda_gemm
int cuda_gemm(
struct CudaBuf a,
int ta,
struct CudaBuf b,
int tb,
struct CudaBuf c,
unsigned long m,
unsigned long n,
unsigned long k,
)c[m,n] = op(a)[m,k] * op(b)[k,n] (cuBLAS sgemm)