NumPy Copy vs View

Copy vs View in NumPy

When you slice or transform a NumPy array, you sometimes get an independent array and sometimes get a window onto the original data. The copy() method always makes a new, independent array, while view() creates a new array object that still points at the same underlying memory as the original. Knowing which one you have prevents subtle bugs where editing one array silently changes another.

The Copy Owns Its Data

import numpy as np

arr = np.array([1, 2, 3, 4, 5])
x = arr.copy()
arr[0] = 42

print(arr)
print(x)

# Output:
# [42  2  3  4  5]
# [ 1  2  3  4  5]

The View Shares Its Data

import numpy as np

arr = np.array([1, 2, 3, 4, 5])
x = arr.view()
arr[0] = 42

print(arr)
print(x)

# Output:
# [42  2  3  4  5]
# [42  2  3  4  5]

Changing arr after making the view changes x too, because a view is just another label for the same block of memory. Changing arr after copy() leaves the copy completely untouched, since it allocated its own memory the moment copy() ran.

Checking Ownership With the base Attribute

import numpy as np

arr = np.array([1, 2, 3, 4, 5])
copy_arr = arr.copy()
view_arr = arr.view()

print(copy_arr.base)
print(view_arr.base)

# Output:
# None
# [1 2 3 4 5]