oa::FnMatrix::matMulNt

matMulNt: oa::Matrix multiplication: B is [N,K] and out = A @ Bᵀ (the OA weight convention, shared with Linear/attention; NOT PyTorch-standard A with B as [K,N]). For batched/standard A use Bmm. Router resolves precision at context execution.

FnC++ public headers

Function Documentation

oa::Matrix oa::FnMatrix::matMulNt(
	const oa::Matrix & inA,
	const oa::Matrix & inB,
	oa::MatMulPrecision inPrecision = oa::MatMulPrecision::Auto
)
oa::Matrix oa::FnMatrix::matMulNt(
	const oa::Matrix & inInput,
	const oa::QuantMatrix & inWeight
)

Parameters

inA

const oa::Matrix &

—

inB

const oa::Matrix &

—

inPrecision

oa::MatMulPrecision

Default: oa::MatMulPrecision::Auto

Returns

oa::Matrix

The declared return value.