Introduction to NumPy

What Is NumPy?

NumPy (Numerical Python) is the foundational library for scientific computing in Python. It provides a high-performance multidimensional array object (ndarray) and a large collection of mathematical functions that operate on those arrays without Python-level loops.

Every major data science library — Pandas, Scikit-Learn, PyTorch, TensorFlow — stores and moves data as NumPy arrays or compatible objects. Learning NumPy is learning the lingua franca of numerical Python.

Why Not Just Use Python Lists?

Python lists are flexible: they can hold objects of any type and grow dynamically. That flexibility comes at a cost. Each element in a Python list is a full Python object with a type tag, reference count, and pointer to its value — roughly 28 bytes for a single integer.

A NumPy array of 1 million float64 values uses 8 MB. The equivalent Python list uses roughly 35 MB just for the object overhead, before counting the actual values.

More importantly, NumPy operations compile down to tight C loops. A Python for loop that adds two lists element-by-element interprets bytecode on every iteration. NumPy’s vectorized addition executes a single C function call that processes all elements in one pass.

import numpy as np
import time

size = 10_000_000

# Python list approach
py_list = list(range(size))
start = time.perf_counter()
result = [x * 2 for x in py_list]
py_time = time.perf_counter() - start

# NumPy approach
np_array = np.arange(size)
start = time.perf_counter()
result = np_array * 2
np_time = time.perf_counter() - start

print(f"Python list: {py_time:.3f}s")
print(f"NumPy array: {np_time:.3f}s")
print(f"Speedup:     {py_time / np_time:.0f}x")
# Python list: 0.412s
# NumPy array: 0.008s
# Speedup:     51x

Installation

NumPy is included in most Python distributions for data work. Install it directly with pip or conda:

pip install numpy

# or with conda (includes optimized BLAS)
conda install numpy

Your First NumPy Array

import numpy as np

# Create a 1-D array from a Python list
temperatures = np.array([22.1, 23.5, 19.8, 25.0, 21.3])

print(temperatures)        # [22.1 23.5 19.8 25.  21.3]
print(temperatures.dtype)  # float64 — NumPy inferred the type
print(temperatures.shape)  # (5,) — a 1-D array with 5 elements
print(temperatures.ndim)   # 1
print(temperatures.size)   # 5 — total number of elements

# Arithmetic broadcasts across the entire array — no loop needed
celsius_to_fahrenheit = temperatures * 9 / 5 + 32
print(celsius_to_fahrenheit)  # [71.78 74.3  67.64 77.   70.34]

# Aggregations
print(f"Mean:  {temperatures.mean():.1f}")  # 22.3
print(f"Max:   {temperatures.max():.1f}")   # 25.0
print(f"Std:   {temperatures.std():.1f}")   # 1.7

Key Array Properties

Every NumPy array has three essential properties you’ll use constantly:

import numpy as np

# 2-D array (matrix) — rows × columns
matrix = np.array([
    [1, 2, 3],
    [4, 5, 6],
    [7, 8, 9]
])

print(matrix.shape)   # (3, 3)
print(matrix.ndim)    # 2
print(matrix.dtype)   # int64
print(matrix.nbytes)  # 72 — 9 elements × 8 bytes each

What’s Next

The next tutorial covers all the ways to create NumPy arrays — from sequences, zeros/ones, ranges, random values, and by reading data from files.