This is the code repository for Mathematics of Machine Learning, published by Packt.
Tivadar Danka
Mathematics of Machine Learning provides a rigorous yet accessible introduction to the mathematical underpinnings of machine learning, designed for engineers, developers, and data scientists ready to elevate their technical expertise. With this book, you’ll explore the core disciplines of linear algebra, calculus, and probability theory essential for mastering advanced machine learning concepts.
PhD mathematician turned ML engineer Tivadar Danka—known for his intuitive teaching style that has attracted 100k+ followers—guides you through complex concepts with clarity, providing the structured guidance you need to deepen your theoretical knowledge and enhance your ability to solve complex machine learning problems. Balancing theory with application, this book offers clear explanations of mathematical constructs and their direct relevance to machine learning tasks. Through practical Python examples, you’ll learn to implement and use these ideas in real-world scenarios, such as training machine learning models with gradient descent or working with vectors, matrices, and tensors.
By the end of this book, you’ll have gained the confidence to engage with advanced machine learning literature and tailor algorithms to meet specific project requirements.
- Understand core concepts of linear algebra, including matrices, eigenvalues, and decompositions
- Grasp fundamental principles of calculus, including differentiation and integration
- Explore advanced topics in multivariable calculus for optimization in high dimensions
- Master essential probability concepts like distributions, Bayes' theorem, and entropy
- Bring mathematical ideas to life through Python-based implementations
| Chapters | Colab | Kaggle | Gradient | Studio Lab |
|---|---|---|---|---|
| Part 1: Linear Algebra | ||||
|
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Part 2: Calculus | ||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Part 3: Multivariable Calculus | ||||
|
|
|
|
|
| Part 4: Probability Theory | ||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
You can run these Jupyter notebooks on remotely on cloud platforms like Google Colab, or locally on your machine. If you decide to run them virtually, you should create a virtual environment first:
virtualenv .venv
After the enviromnment is created, activate it and install the required Python packages.
source .venv/bin/activate
pip install -r requirements.txt
Tivadar Danka is a mathematician by training, a machine learning engineer by profession, and an educator by passion. After finishing his PhD in 2016 (about the arcane subject of orthogonal polynomials), he switched career paths and has been working in machine learning ever since. His work includes applying deep learning to cell microscopy images to identify and phenotype cells, creating one of the most popular open source Python packages for active learning, building a full machine learning library from scratch, and collecting about a total of 100k followers on social media, all by posting high-quality educational content.
