readme.md

Try this:

#!/bin/python

import make as lrnr

t = lrnr.smarty()

t.learn("dogs bark")
t.learn("cats do not bark")
t.learn("dogs are warm blooded")
t.learn("cats are warm blooded")
t.learn("fish are not warm blooded")
t.learn("fish have scales")

t.think()
t.prt()
t.predPrt()

This will yeild the following table.

               bark           blooded        scales
dogs           1              1              None
cats           0              1              None
fish           None           0              1

               bark           blooded        scales
dogs           1              1              1
cats           0              1              1
fish           0.5            0              1

While the results of this underlying algorithm can are bad on small data sets, in places where the sample size is larger this fitting provides smooth toplogies in n dimensional space to fit any data.