Consider the following input channels$(5$x$5$x$3),$ two$(3$x$3$x$3)$ filter volumes and two$(1$x$1)$bias$.$ A convolutional layer cross$-$correlates the input and kernels and adds a scalar bias toproduce an output. Assume that a zero padding$($P$=1),$ stride of $2$ is applied to the input volume.Illustrate the convolution layer computation and show the output$(3$x$3$x$2).$