Converting to float without truncation

Question

I am trying to calculate an error function with Python 2.7, it has a term such that;

...
sumNSSD = float()
sumNSSD = sumFiGi/sumGi2
return sumNSSD 

sumFiGi and sumGi2 are of type int() and sumFiGi/sumGi2 results to be something between 0 and 1, lets say 3/4

When the code evaluates 3/4 it returns 0 because of truncation, and sumNSSD becomes int() again. When I try sumNSSD = float(sumFiGi/sumGi2) it still truncates and returns 0.0

A solution that works is to type sumNSSD = float(sumFiGi) / float(sumGi2) but the actual equation is much longer and it doesn't make sense to put every term inside float(...)

There should be an obvious way for getting around the truncation, however, I did search a lot of topics but no question answers my problem.

Using Decimal() produces the same result because it evaluates what is in between the parantheses first, then truncates and returns 0

Even Python Documentation puts it this way;

>>> from decimal import *  
>>> getcontext().prec = 6  
>>> Decimal(1) / Decimal(7)  
Decimal('0.142857')  

What would be a simple solution except putting every single term into Decimal(...) or float(...)