15  NumPy

TipLearning Objectives
  • Be able to use numpy
  • Know what numpy is good for

15.1 What Is NumPy?

NumPy (Numerical Python) is a foundational Python library for numerical and scientific computing. It provides efficient data structures and functions for working with large, multi-dimensional numerical datasets. NumPy is widely used in data science, machine learning, engineering, physics, and finance, and it underpins many other libraries in the Python scientific ecosystem, such as pandas, SciPy, and scikit-learn.

You will likely need NumPy in many of the future python courses including in machine learning and data science courses, and therefore we introduce it here.

15.2 Base Data Structure: The ndarray

The core data structure in NumPy is the ndarray (N-dimensional array).

Key characteristics:

  • Stores elements of a single data type (e.g., integers, floats)
  • Supports one-dimensional (vectors), two-dimensional (matrices), and higher-dimensional arrays
  • Optimised for performance and memory efficiency

Example:

import numpy as np

a = np.array([1, 2, 3])          # 1D array
b = np.array([[1, 2], [3, 4]])   # 2D array

15.3 Common Operations

Element-wise Operations

x = np.array([1, 2, 3])
y = np.array([4, 5, 6])

x + y        # array([5, 7, 9])
x * y        # array([4, 10, 18])

Broadcasting

x * 2        # array([2, 4, 6])

Indexing and Slicing

x[0]         # 1
x[1:]        # array([2, 3])

Aggregations

x.mean()     # 2.0
x.sum()      # 6
x.max()      # 3

Linear Algebra

A = np.array([[1, 2], [3, 4]])
np.linalg.det(A)
np.linalg.inv(A)

15.4 Reading and Saving Data

Reading Data

Common methods include from text or CSV files:

data = np.loadtxt("data.csv", delimiter=",")

or from NumPy’s native binary format:

data = np.load("data.npy")

Saving Data

np.savetxt("output.csv", data, delimiter=",")
np.save("output.npy", data)

NumPy is often combined with pandas another package for more complex data ingestion tasks.

15.5 Advantages and Disadvantages

Advantages Disadvantages
Very fast for numerical operations due to optimized C-based implementation Limited support for non-numeric or mixed data
Memory-efficient compared to Python lists Less intuitive for labeled or relational data
Extensive mathematical and linear algebra functionality Steeper learning curve than basic Python lists for beginners
Integrates well with the scientific Python ecosystem

15.6 Common Alternatives and Their Use Cases

pandas Used for labeled, tabular data (DataFrames). Builds on NumPy and adds indexing, grouping, and data-cleaning tools.

SciPy Extends NumPy with advanced scientific algorithms (optimization, signal processing, statistics).

TensorFlow / PyTorch Used for large-scale numerical computation and machine learning, especially with GPUs and automatic differentiation.

Python Lists Suitable for small datasets or heterogeneous data, but inefficient for numerical computation.

‘NumPy’ Documentation

The ‘NumPy’ documentation is very good and the modules are very widely used! For this reason I have not given any more examples here. When you start coding more, you must use the resources and documentation to solve your own challenges. Use the both the documentation and what you have learnt so far in the following exercises.

https://numpy.org/doc/

ExerciseExercise 1 - Numpy Exercise

Level:

Create the following NumPy arrays:

A 1D array of integers from 0 to 9.

A 2D array of shape (3, 3) filled with zeros.

A 2D array of shape (2, 4) filled with ones.

Hint: Use np.arange() and np.zeros() / np.ones().

  1. Compute the element-wise sum of a and b given the arrays:
a = np.array([1, 2, 3, 4])
b = np.array([10, 20, 30, 40])

  1. Multiply each element of a by 3.

  2. Compute the square of each element in b.

  1. Extract the first row given the array:
arr = np.array([[1, 2, 3],
                [4, 5, 6],
                [7, 8, 9]])

  1. Extract the last column.

  2. Extract the 2x2 subarray in the bottom-right corner.

  1. Create a random 3x4 array of integers between 0 and 10 (inclusive). Then:

  2. Compute the sum of all elements.

  3. Compute the mean of each column.

  4. Find the maximum element in each row.

Use np.random.randint().

Using NumPy’s built-in functions:

  1. Create a 5x5 array with random float values between 0 and 1.

  2. Normalize the array so that its values are between 0 and 1 using min-max normalization.

  3. Find the determinant of a 3x3 subarray of your choice.

  1. Load in the student scores from student_scores.csv. Understand the data

  2. Compute the average score for each student and store it in a 1D array called student_avg.

  3. Compute the average score for each subject and store it in a 1D array called subject_avg.

  4. Find the student(s) with the highest average score.

  5. Normalize the scores array so that each subject (column) has values between 0 and 1.

  6. Bonus: Create a boolean mask array indicating which scores are above the average for their subject.

ExerciseExercise 2 - Using Numpy in Expense_calculator

Level:

Modify your expense calculator or your own function to use NumPy to load and process the csv.

15.7 Summary

Numpy is very useful for numerical calculations, that are common in many fields.