Python NumPy Introduction

What is NumPy?

NumPy is the foundational library for numerical computing in Python, providing the ndarray type -- a fixed-type, multi-dimensional array that's dramatically faster and more memory-efficient than nested Python lists for numeric work, because its elements are stored contiguously in memory rather than as separate Python objects.

Example: What is NumPy?

import numpy as np
arr = np.array([1, 2, 3])  # fixed-type, contiguous ndarray
print(type(arr))

NumPy Array Attributes

Every NumPy array carries built-in metadata: .shape gives its dimensions as a tuple, .ndim gives the number of dimensions, and .size gives the total element count.

Reading these attributes is how you quickly confirm an array's structure matches what a calculation expects, before running the calculation.

Example: NumPy Array Attributes

import numpy as np
arr = np.array([[1, 2], [3, 4]])
print(arr.shape, arr.ndim, arr.size)  # dimensions, number of dimensions, total elements

Basic Mathematical Operations

NumPy applies arithmetic element-by-element across an entire array in a single vectorized operation (a + b adds every pair of corresponding elements at once), implemented in compiled C code under the hood.

This is why NumPy math on large arrays runs orders of magnitude faster than an equivalent Python for loop.

Example: Basic Mathematical Operations

import numpy as np
a = np.array([1, 2, 3])
b = np.array([10, 20, 30])
print(a + b)  # element-wise addition, vectorized in one call

Array Data Types

Unlike a Python list, which can freely mix types, every NumPy array has one fixed dtype (like int32 or float64) shared by all its elements, chosen either automatically or explicitly via the dtype argument.

Choosing a smaller dtype when precision allows can meaningfully cut an array's memory footprint.

Example: Array Data Types

import numpy as np
arr = np.array([1, 2, 3], dtype="float64")  # explicit fixed dtype for every element
print(arr.dtype)

Summary Statistics

NumPy arrays ship with built-in aggregate methods -- .sum(), .mean(), .max(), .min(), and more -- that compute across the whole array (or along a chosen axis for multi-dimensional data) far faster than manually looping and accumulating a running total in Python.

Example: Summary Statistics

import numpy as np
arr = np.array([1, 2, 3, 4])
print(arr.sum(), arr.mean(), arr.max())  # built-in aggregate methods