Ttykuu

I am trying to better plan the class structures of a large code base in python. The project is a data science project involving data cleaning, data analysis, etc. I model all the data cleaning tasks as one class object, and all the data analysis tasks as another class object. Under the Data_Cleaning class, I will output certain processed data, and this data will in term feed into the Data_Analysis class. I want these classes to be structured as a framework for reusability. I wonder what are the pros and cons of the following condensed example, and what are the guiding principles to determine if I should instantiate the Data_Cleaning object outside or inside the Data_analysis class?

# Approach #1
class Data_Cleaning():
 def __init__(self):
 self.numba = 2

 def numba_plus_one(self) -> int:
 return (self.numba+1)

class Data_Analysis():
 def __init__(self, input_numba):
 self.input_numba = input_numba

 def numba_times_four(self) -> int:
 return (self.input_numba*4)

x = Data_Cleaning()
my_numba = x.numba_plus_one()
print(my_numba)

y = Data_Analysis(my_numba)
my_new_numba = y.numba_times_four()
print(my_new_numba)

# Approach #2
class Data_Cleaning():
 def __init__(self):
 self.numba = 2

 def numba_plus_one(self) -> int:
 return (self.numba+1)

class Data_Analysis():
 def __init__(self):
 z = Data_Cleaning()
 self.input_numba = z.numba_plus_one()

 def numba_times_four(self) -> int:
 return (self.input_numba*4)

o = Data_Analysis()
print(o.numba_times_four())