View vs Copy¶

When the contents are physically stored in another location, it is called Copy.

On the other hand, a different view of the same memory content is provided, we call it as View.

Different array objects can share the same data. NumPy has ndarray.view() method which is a new array object that looks at the same data of the original array.

Here, change in dimensions of the new array doesn’t change dimensions of the original.

import numpy as np

fruits = np.array(["Apple","Mango","Grapes","Watermelon"])

We will create basket now as a view of fruits

basket_1 = fruits.view()
basket_2 = fruits.view()

print(basket_1)
print(basket_2)

['Apple' 'Mango' 'Grapes' 'Watermelon']
['Apple' 'Mango' 'Grapes' 'Watermelon']

print("ids for the arrays are different.")
print("id for fruits is : ")
print(id(fruits))
print("id for baskets is : ")
print(id(basket_1))
print(id(basket_2))

ids for the arrays are different.
id for fruits is : 
4369559104
id for baskets is : 
4369576544
4369576064

basket_1 is fruits

False

baskets is a view of the data owned by fruits¶

basket_1.base is fruits 

True

Change a few elements of basket. It changes the elements of fruits¶

Here, we assign a new value to the first element of basket_2. You might be astonished that the list of fruits has been “automatically” changed as well. The explanation is that there has been no new assignment to basket_2, only to one of its elements.

basket_2[0] = "Strawberry"

basket_2

array(['Strawberry', 'Mango', 'Grapes', 'Watermelon'], dtype='<U10')

fruits

array(['Strawberry', 'Mango', 'Grapes', 'Watermelon'], dtype='<U10')

And this also affects basket_1¶

basket_1

array(['Strawberry', 'Mango', 'Grapes', 'Watermelon'], dtype='<U10')

Change the entire elements of basket. It does not change fruits¶

basket_1 = np.array(["Peach","Pineapple","Banana","Orange"])

basket_1

array(['Peach', 'Pineapple', 'Banana', 'Orange'], dtype='<U9')

fruits

array(['Strawberry', 'Mango', 'Grapes', 'Watermelon'], dtype='<U10')

In this case, a new memory location had been allocated for basket_1, because we have assigned a complete new list to this variable

Change the shape of basket. It does not change the shape of fruits¶

basket_2.shape = 2,2

print("basket_2: ")
print(basket_2)

basket_2: 
[['Strawberry' 'Mango']
 ['Grapes' 'Watermelon']]

print("Shape of fruits: ")
print(fruits)

Shape of fruits: 
['Strawberry' 'Mango' 'Grapes' 'Watermelon']

Slicing an array returns a view of it¶

mini_basket = fruits[2:]

mini_basket

array(['Grapes', 'Watermelon'], dtype='<U10')

fruits[3] = "Peach"

mini_basket

array(['Grapes', 'Peach'], dtype='<U10')

Introduction to NumPy