nn_elu2d_f16()

void nn_elu2d_f16	(	Tensor2D_F16 *	y,
		const Tensor2D_F16 *	x,
		float	alpha
	)