Array Indexing

Relax. We'll talk through this in plain words — no textbook voice.

Indexing means picking out a single element from an array by its position. Just like in Python, NumPy indices start at 0. For a 2-D array, you can pick an element using the form array[row, column].

python

import numpy as np

marks = np.array([65, 78, 90, 82])
print("First mark:", marks[0])
print("Last mark:", marks[-1])

student_scores = np.array([
    [80, 75, 90],
    [88, 92, 85]
])

print("Second student, third subject:", student_scores[1, 2])

marks[0] grabs the first element. marks[-1] grabs the last one. For a 2-D array, student_scores[1, 2] grabs the second row, third column. Since indices start at 0, row 1 is actually the second row.

You should see

First mark: 65 Last mark: 82 Second student, third subject: 85

Info

Once you've got a solid handle on indexing, you can pull exactly the values you need out of table data, image pixels, or matrix values.

Easy traps

Calling an index beyond the array's length throws an IndexError. For example, if an array only has 4 elements, marks[4] doesn't exist.

Exercise

If an orders table has column 0 as order id, column 1 as amount, and column 2 as quantity, indexing lets you pull any cell you want directly.

python

import numpy as np

orders = np.array([
    [101, 25000, 2],
    [102, 18000, 1],
    [103, 42000, 3]
])

# First order amount
print("First order amount:", orders[0, 1])

# Last order quantity
print("Last order quantity:", orders[-1, 2])

You'll know it worked when: First order amount: 25000 Last order quantity: 3