Python-如果不更改内部变量的值，为什么要更新内部循环的变量？

Question

I want to set the values of matrix W in Wo. Then, I will go inside a loop and do operations with W. At the end of the loop, I want to restore it to Wo, but I am getting the opposite result, Wo is becoming Wo.

How is that possible? How can I accomplish what I want? I have also tried defining Wo as a global but no result

W = np.random.rand(5, 10)
global Wo
Wo = W # Back up for the initial values of W

for k in range(0, K):

  for j in range(0, m):       # For every hidden node

    for i in range(0, n):        # For every connection

      # Operations (W get changed)
      W = Wo

Then, if I check the value of Wo has been modified, in a way that Wo == W is always True.

What am I missing?

Thanks in advance

Answer 1

To accomplish what you want, you need a deep copy. Otherwise you only create a reference to the same memory.

import numpy as np
import copy as cp

W = np.random.rand(5, 10)
global Wo

Wo = cp.deepcopy(W) # Back up for the initial values of W

for k in range(0, K):

  for j in range(0, m):       # For every hidden node

    for i in range(0, n):        # For every connection

      # Operations (W get changed)
      W = cp.deepcopy(Wo)

Answer 2

All items in Python are references so if you change one reference other will change too

l = []
global l_global
l_global = l

for i in range(1, 10):
  l.append(i)

print(l)
print(l_global)

Does it make sense now for your case?