... which explicitly factorizes 3D convolution into two separate and successive operations, a 2D spatial convolution ... pytorch / vision.
確定! 回上一頁