I am implementing a logarithmic quantizer and what I would like to do is to optimize the code as much as possible. The precise point where I would like to make a change is the l
