Delegate DirectILKernelGenerator.GemvKernel

Namespace

NumSharp.Backends.Kernels

Assembly

NumSharp.dll

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.

public delegate void DirectILKernelGenerator.GemvKernel(void* a, long aRowStride, void* x, void* y, long m, long k)

Parameters

a void*

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.

aRowStride long

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.

x void*

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.

y void*

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.

m long

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.

k long

Matrix·vector kernel: y[i] = dot(A[i], x) for row-major A whose rows are contiguous (K along stride 1) and start aRowStride elements apart, and contiguous x/y.