Learn the essentials fast — without losing rigor.
The AFAP book series is built for students, engineers, and researchers who want a direct, example-driven path: concepts, minimal theory, practical code, and real evaluation. Book 1 is published, and the companion code is available on GitHub.
What makes AFAP different?
-
✓Speed-first structure
Only the theory you need, immediately followed by implementation and results. -
✓Leakage-safe workflows
Train/test discipline, proper cross-validation, and reproducible evaluation. -
✓Hands-on code and results
Practical scripts, notebooks, datasets, and experiment logs. -
✓Engineer-friendly explanations
No fluff. Focus on what works, why it works, and how to use it.
Published books
Book 1
Introduction to scikit-learn and Its Ecosystem
A practical and fast introduction to scikit-learn, machine learning workflows, model evaluation, pipelines, and the surrounding Python ecosystem.
Code and resources
The companion repository contains the code examples and resources for the book. The goal is simple: run the examples, reproduce the results, and then modify them for your own work.
-
✓End-to-end scripts
Train, evaluate, and export models with clean structure. -
✓Reproducible evaluation
Proper splits, metrics, cross-validation, and leakage-safe workflows. -
✓Minimal dependencies
Clean Python and scikit-learn style, easy to run and modify.
Roadmap
Upcoming AFAP books will appear here as they move from planning to writing to publication.
-
1Book 1: Introduction to scikit-learn and Its Ecosystem
Status: Published. -
2Book 2: Supervised Learning
Status: Planned / in preparation. -
3Book 3: Classification Techniques
Status: Planned / in preparation. -
4Book 4: Imbalanced Data Handling
Status: Planned / in preparation.
FAQ
What is the AFAP philosophy?
AFAP means As Fast As Possible: learn the essential theory quickly, then move directly into practical implementation and real evaluation.
Do I need to read the books in order?
Not necessarily. Each book is designed to work as a standalone guide, but reading them in order gives a more structured learning path.
Where is the code?
The code is hosted on GitHub and linked directly from this page.
Who is the series for?
Students, researchers, engineers, and developers who want practical, fast, and reproducible machine learning workflows.
No comments:
Post a Comment