This is a a quick and dirty Python package for optimal scaling. It is actually just a wrapper around an R implementation called optiscale by William G. Jacoby.
Licensed under GNU GPL version 2.
The package can be installed directly from github as follows:
pip install git+git://github.com/alubbock/pyopscale.git
It requires a working installation of R with the optiscale package installed.
Example taken from the documentation linked above.
from pyopscale.opscale import OpScale
x1 = [1,1,1,1,2,2,2,3,3,3,3,3,3]
x2 = [3,2,2,2,1,2,3,4,5,2,6,6,4]
os = OpScale()
print(os.opscale(x1, x2, rescale=False))The result of the above should be
[2.25, 2.25, 2.25, 2.25, 2.0, 2.0, 2.0, 4.5, 4.5, 4.5, 4.5, 4.5, 4.5].
Further information is available in the optiscale documentation (PDF). See the source code for a list of arguments to the opscale function.