Negative Indexing in NumPy

Understanding Negative Indexing

In standard Python lists, you access elements using their position starting from 0. For example:

my_list = [10, 20, 30, 40, 50]
print(my_list[0])  # Output: 10
print(my_list[3])  # Output: 40

NumPy arrays extend this by allowing negative indices. A negative index -i refers to the i-th element from the end of the array.

import numpy as np

my_array = np.array([10, 20, 30, 40, 50])
print(my_array[-1])  # Output: 50
print(my_array[-3])  # Output: 30

As you can see, my_array[-1] accesses the last element (50), and my_array[-3] accesses the third element from the end (30).

Practical Applications of Negative Indexing

Negative indexing is incredibly useful in many scenarios:

1. Accessing the last n elements:

Need the last three elements? Negative indexing makes it trivial:

last_three = my_array[-3:]
print(last_three)  # Output: [30 40 50]

2. Slicing from the end:

You can combine negative indexing with slicing to extract portions of the array from the end:

middle_section = my_array[-4:-1] # elements from -4 up to (but not including) -1
print(middle_section) # Output: [20 30 40]

3. Efficiently manipulating the end of arrays:

Suppose you need to append or remove elements from the end of a large array. Negative indexing simplifies the process, avoiding the need to constantly recalculate indices.

my_array = my_array[:-1]
print(my_array) # Output: [10 20 30 40]


my_array = np.append(my_array, 60)
print(my_array) # Output: [10 20 30 40 60]

4. Multi-dimensional arrays:

Negative indexing works seamlessly with multi-dimensional NumPy arrays.

multi_array = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
print(multi_array[-1, -1])  # Output: 9 (last row, last column)
print(multi_array[:, -1])  # Output: [3 6 9] (last column)

These examples demonstrate the power and flexibility of negative indexing in NumPy. By mastering this technique, you’ll write more efficient and readable NumPy code. It’s a vital skill for any Python data scientist.