import numpy as np

賦值（物件引用）

a = np.array([1, 2, 3, 4])
a

array([1, 2, 3, 4])

b = a
b

array([1, 2, 3, 4])

a[2] = 0
a

array([1, 2, 0, 4])

b

array([1, 2, 0, 4])

b[1] = 0
a

array([1, 0, 0, 4])

c = b[1:3]
c

array([0, 0])

c[1] = 2
c

array([0, 2])

a

array([1, 0, 2, 4])

b
 

array([1, 0, 2, 4])

copy(物件拷貝，淺拷貝)

a = np.array([1, 2, 3, 4])
b = a.copy()
b

array([1, 2, 3, 4])

b[0] = 0
b

array([0, 2, 3, 4])

a

array([1, 2, 3, 4])

deepcopy(物件拷貝，深拷貝)

a = [1, 2, 3, 4, ['a', 'b']] #原始物件
d = copy.deepcopy(a) #物件拷貝，深拷貝
d

[1, 2, 3, 4, ['a', 'b']]

複製之間的區別

import
 
 copy
a = [1, 2, 3, 4, ['a', 'b']] #原始物件

b = a #賦值，傳物件的引用
c = copy.copy(a) #物件拷貝，淺拷貝
d = copy.deepcopy(a) #物件拷貝，深拷貝

a.append(5) #修改物件a
a[4].append('c') #修改物件a中的['a', 'b']陣列物件

print('\t\t:a = ', a)
print('物件引用\t:b = ', b)
print('copy\t\t:c = ', c)
print('deepcopy\t:d = ', d)

        :a =  [1, 2, 3, 4, ['a', 'b', 'c'], 5]
物件引用    :b =  [1, 2, 3, 4, ['a', 'b', 'c'], 5]
copy        :c =  [1, 2, 3, 4, ['a', 'b', 'c']]
deepcopy    :d =  [1, 2, 3, 4, ['a', 'b']]
 

• 物件引用只是引用了之前物件愛的索引，當原物件與賦值後的物件發生改變時，另一個物件也會發生改變
• copy 只拷貝父物件，不會拷貝物件的內部的子物件。也就是隻拷貝一維中的內容，如果一維中是列表，則是引用列表的物件而不是值
• deepcopy 深拷貝 拷貝物件的值及其子物件的值