Avoid class initialization modified in method in python

Question

My code is below:

import numpy as np

class xx():
   def __init__(self, x):
      self.x = x
   def main(self):
      for i in [0,1,2,3]:
         y = self.x
         y[0, i] = 0
         print(y)
z = np.array([[0.5,0.6,0.7,0.8],[1,2,3,4],[6,4,3,1]])        
xx(z).main()

My original code is far more complicated than this, so I decided to create a similar example on the issue.

I meant to only change y in the for loop, and reassign y with self.x. but it seemed that self.x was changed in the for loop too. How can I avoid self.x getting modified each time as y changes?

Could not recreate, I get your desired output as expected. Strings are immutable, and you don't explictly update self.x, so each assignment y = self.x gets the value 'David'. — jonrsharpe
– jonrsharpe, Commented Jun 1, 2021 at 20:16
well it seems your current code currently gives me the desired output. you may need to clarigy more — Ade_1
– Ade_1, Commented Jun 1, 2021 at 20:24

Baragorn · Accepted Answer · 2021-06-03 15:12:47Z

1

I would recommend using copy.deepcopy()

import copy

class xx():
   def __init__(self, x):
      self.x = x
   def main(self):
      for i in [0,1,2]:
         y = copy.deepcopy(self.x)
         y[0, i] = 0
         print(y)
z = np.array([[1,1,1],[2,2,2]])
xx(z).main()

>>>[[0 1 1]
    [2 2 2]]

>>>[[1 0 1]
    [2 2 2]]

>>>[[1 1 0]
    [2 2 2]]

edited Jun 3, 2021 at 15:12

answered Jun 1, 2021 at 20:30

Baragorn

1461 silver badge3 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Anson Over a year ago

In my original code, it's numpy array. So python does not distinguish two objects such as dataframe numpy when reassigning? The variables are pointing to the same object using '='?

Baragorn Over a year ago

I'm not exactly sure what you mean, but notice that when you wrote the line y = self.x and then y[0, i] = 0 those two lines are equivalent to self.x[0, i] = 0 , since the y variable is just a pointer to self.x, when you use copy.deepcopy() you create a new object in the memory and can modify it without overwriting self.x

Collectives™ on Stack Overflow

Avoid class initialization modified in method in python

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related