oa::FnMatrix::convTranspose2d

convTranspose2d: 2D transposed convolution (learnable upsampling). input: [N, inC, H, W], weight: [inC, outC, K, K], Bias: [outC] output: [N, outC, H_out, W_out] where H_out = (H - 1) * S - 2P + K.

FnC++ public headers

Function Documentation

Matrix oa::FnMatrix::convTranspose2d(
	const Matrix & inX,
	const Matrix & inWeight,
	const Matrix & inBias,
	oa::U32 inStride,
	oa::U32 inPadding
)

convTranspose2d: 2D transposed convolution (learnable upsampling). input: [N, inC, H, W], weight: [inC, outC, K, K], Bias: [outC] output: [N, outC, H_out, W_out] where H_out = (H - 1) * S - 2P + K.

Parameters

inX

const Matrix &

—

inWeight

const Matrix &

—

inBias

const Matrix &

—

inStride

oa::U32

—

inPadding

oa::U32

—

Returns

Matrix

The declared return value.