nn_addmm_f32()

void nn_addmm_f32	(	Tensor2D_F32 *	y,
		const Tensor2D_F32 *	c,
		const Tensor2D_F32 *	x1,
		const Tensor2D_F32 *	x2
	)

Performs a matrix multiplication of two 2D floating-point tensors and adds the result to a third tensor.

nn_addmm_f32

y[i][j] = x1[i][k] * x2[k][j] + c[i][j]

Parameters

y	The result tensor.
c	The third tensor.
x1	The first tensor.
x2	The second tensor.