What would be the most efficient way to implement a Pooled Standard Deviation in numpy?
is it already efficiently implemented in numpy, or any other library?
