1D Indexing

You have Virat Kohli's runs from 14 IPL matches stored in one NumPy array. Very often you don't want all 14 numbers — you want "the score in match 1", "the last match", or "the third-last match". Picking out one element by its position is called indexing, and it is the first skill you need before slicing, filtering and everything else in NumPy.

Definition

Indexing is selecting a single element of an array by writing its position number (index) inside square brackets, e.g. arr[3]. In a 1D array, indexes start at 0 for the first element and go up to n − 1 for the last; negative indexes count backwards from the end, so -1 is the last element.

• Position number (index) — an integer that says where the element sits, not what its value is.
• Starts at 0 — NumPy follows Python: the first element is arr[0], not arr[1]. So "match 1" is index 0, "match 5" is index 4.
• n − 1 — an array of size 14 has valid positive indexes 0 to 13. Index 14 does not exist.
• Negative index — -1 is the last element, -2 the second-last, … -n is the first. Internally NumPy converts -k to n - k.

Positions at a glance

value :   72   45   88   31   64    0  103   52   19   77   41   93   28   66
index :    0    1    2    3    4    5    6    7    8    9   10   11   12   13
neg   :  -14  -13  -12  -11  -10   -9   -8   -7   -6   -5   -4   -3   -2   -1

Every element has two valid addresses: one positive, one negative. virat_runs[6] and virat_runs[-8] are the same element (103).

Syntax

arr[i]        # element at position i (0-based)
arr[-k]       # k-th element from the end

The result of indexing a 1D array with one integer is a single scalar value (a NumPy scalar such as np.int64), not an array.

Example — first, last and any match

import numpy as np

virat_runs = np.array([72, 45, 88, 31, 64, 0, 103, 52, 19, 77, 41, 93, 28, 66])

print("Match 1  :", virat_runs[0])
print("Match 7  :", virat_runs[6])
print("Last     :", virat_runs[-1])
print("2nd last :", virat_runs[-2])
print("Size     :", virat_runs.size)
print("Same element?", virat_runs[6] == virat_runs[-8])

Match 1  : 72
Match 7  : 103
Last     : 66
2nd last : 28
Size     : 14
Same element? True

How it works: "match 7" is index 6 because counting starts at 0. virat_runs[-8] means "8 from the end": 14 - 8 = 6, the same slot.

Example — what type do you get back?

virat_runs = np.array([72, 45, 88, 31, 64, 0, 103, 52, 19, 77, 41, 93, 28, 66])

score = virat_runs[0]
print(score)
print(type(score))
print(repr(score))
print(int(score) + 1)

72
<class 'numpy.int64'>
np.int64(72)
73

How it works: one integer index gives you one value, stored as a NumPy scalar of the array's dtype (np.int64 here). print() shows just 72, while repr() (what Jupyter shows when you type the expression alone) shows np.int64(72) in NumPy 2. It behaves like a normal number in maths; use int(...) if you need a plain Python int.

Example — index out of range

virat_runs = np.array([72, 45, 88, 31, 64, 0, 103, 52, 19, 77, 41, 93, 28, 66])

try:
    print(virat_runs[14])
except IndexError as e:
    print("IndexError:", e)

IndexError: index 14 is out of bounds for axis 0 with size 14

How it works: valid indexes for 14 elements are 0…13 (or -14…-1). Asking for 14 raises an IndexError — exactly like a Python list. (Slicing, covered next, is more forgiving.)

Python list vs NumPy (1D indexing)

Common mistakes

• ❌ virat_runs[1] for "match 1". ✅ virat_runs[0] — the first element is at index 0.
• ❌ virat_runs[14] for the last of 14 matches. ✅ virat_runs[13] or, better, virat_runs[-1] (works for any length).
• ❌ virat_runs(0) with round brackets — arrays are not functions (TypeError: 'numpy.ndarray' object is not callable). ✅ Square brackets virat_runs[0].
• ❌ Using a float index like virat_runs[2.0]. ✅ Indexes must be integers — NumPy raises IndexError: only integers, slices ... are valid indices.

Interview tip

Q: What is arr[-1] and why is it useful? A: It returns the last element. Negative indexes count from the end (-1 → n-1), so you can grab the latest value without knowing the array's length — handy for "the most recent match" or "today's price".