I am interested in designing a layer of my feedforward neural network which takes as an input the vector x and gives out the same as I would get with
model =
