Let us see how to copy arrays in Python. There are 3 ways to copy arrays :
- Simply using the assignment operator.
- Shallow Copy
- Deep Copy
Assigning the Array
We can create a copy of an array by using the assignment operator (=).
Syntax :
new_arr = old_ arr
In Python, Assignment statements do not copy objects, they create bindings between a target and an object. When we use = operator user thinks that this creates a new object; well, it doesn’t. It only creates a new variable that shares the reference of the original object.
Example:
Python3
# importing the modulefrom numpy import * # creating the first arrayarr1 = array([2, 6, 9, 4]) # displaying the identity of arr1print(id(arr1))# assigning arr1 to arr2arr2 = arr1 # displaying the identity of arr2print(id(arr2))# making a change in arr1arr1[1] = 7 # displaying the arraysprint(arr1)print(arr2) |
Output :
117854800 117854800 [2 7 9 4] [2 7 9 4]
We can see that both the arrays reference the same object.
Shallow Copy
A shallow copy means constructing a new collection object and then populating it with references to the child objects found in the original. The copying process does not recurse and therefore won’t create copies of the child objects themselves. In the case of shallow copy, a reference of the object is copied in another object. It means that any changes made to a copy of the object do reflect in the original object. We will be implementing shallow copy using the view() function.
Example :
Python3
# importing the modulefrom numpy import * # creating the first arrayarr1 = array([2, 6, 9, 4])# displaying the identity of arr1print(id(arr1))# shallow copy arr1 in arr2 using view()arr2 = arr1.view() # displaying the identity of arr2print(id(arr2)) # making a change in arr1arr1[1] = 7 # displaying the arraysprint(arr1)print(arr2) |
This time although the 2 arrays reference different objects, still on changing the value of one, the value of another also changes.
Deep Copy
Deep copy is a process in which the copying process occurs recursively. It means first constructing a new collection object and then recursively populating it with copies of the child objects found in the original. In the case of deep copy, a copy of the object is copied into another object. It means that any changes made to a copy of the object do not reflect in the original object. We will be implementing deep copy using the copy() function.
Python3
# importing the modulefrom numpy import * # creating the first arrayarr1 = array([2, 6, 9, 4])# displaying the identity of arr1print(id(arr1))# shallow copy arr1 in arr2 using copy()arr2 = arr1.copy()# displaying the identity of arr2print(id(arr2)) # making a change in arr1arr1[1] = 7 # displaying the arraysprint(arr1)print(arr2) |
Output :
121258976 125714048 [2 7 9 4] [2 6 9 4]
This time the changes made in one array are not reflected in the other array.
Deep Copy Continued
If you are dealing with NumPy matrices, then numpy.copy() will give you a deep copy. However, if your matrix is simply a list of lists then consider the below two approaches in the task of rotating an image (represented as a list of a list) 90 degrees:
Python3
import copydef rotate_matrix(image): # Copy method one copy_image_one = copy.deepcopy(image) print("Original", matrix) print("Copy of original", copy_image_one) N = len(matrix) # Part 1, reverse order within each row for row in range(N): for column in range(N): copy_image_one[row][column] = image[row][N-column-1] print("After modification") print("Original", matrix) print("Copy", copy_image_one) # Copy method two copy_image_two = [list(row) for row in copy_image_one] # Test on what happens when you remove list from the above code. # Part 2, transpose for row in range(N): for column in range(N): copy_image_two[column][row] = copy_image_one[row][column] return copy_image_twoif __name__ == "__main__": matrix = [[1, 2, 3], [4, 5, 6], [7, 8, 9]] print("Rotated image", rotate_matrix(matrix)) |
Output:
Original [[1, 2, 3], [4, 5, 6], [7, 8, 9]] Copy of original [[1, 2, 3], [4, 5, 6], [7, 8, 9]] After modification Original [[1, 2, 3], [4, 5, 6], [7, 8, 9]] Copy [[3, 2, 1], [6, 5, 4], [9, 8, 7]] Rotated image [[3, 6, 9], [2, 5, 8], [1, 4, 7]]
