Table of Contents¶
Natural Language Processing: A Notebook-Based Introduction
This open textbook is built from interactive Jupyter notebooks. Read each chapter inline, launch it in Google Colab with one click, or download it to run locally. New to the book? Start with the Preface.
How to run a chapter
At the top of every chapter you'll find a launch bar:
- Open in Colab — run it in your browser on a free Google Colab runtime.
- View on GitHub — read the source.
- Download .ipynb — run it locally in Jupyter or VS Code.
I — Language Model Foundations¶
| Ch | Chapter | Topics |
|---|---|---|
| 1 | Tokenization & Vocabulary | Subword segmentation, vocabulary design, OOV handling, n-grams |
| 1+ | Setup Tutorial | Environment & Python package setup |
| 1+ | AI Usage Example | Responsible AI-assisted problem solving |
II — Machine Learning Foundation and NLP Tasks¶
| Ch | Chapter | Topics |
|---|---|---|
| 2 | Text Classification | Sentence classification, logistic regression |
III — Representations & Sequences¶
| Ch | Chapter | Topics |
|---|---|---|
| 3 | Embeddings & Sequence Models | Word embeddings, RNNs, token classification |
IV — Transformers & LLMs¶
| Ch | Chapter | Topics |
|---|---|---|
| 4 | Transformers & LLMs | Attention, transformers, pretrained models |
| 4+ | Transformer Illustrated | Visual, step-by-step transformer walkthrough |
V — LLM Alignment and Agentic AI¶
| Ch | Chapter | Topics |
|---|---|---|
| 5 | Fine-Tuning with LoRA/QLoRA | Supervised fine-tuning, parameter-efficient training (TRL) |
| 5+ | Agentic NLP | Building an LLM agent with LangGraph |
See the Roadmap for chapters in progress and planned.
Run everything locally¶
git clone https://github.com/mlciv/nlp-notebooks.git
cd nlp-notebooks
python3 -m venv .venv
source .venv/bin/activate
pip install jupyter
jupyter lab
The notebooks live in docs/notebooks/. Each chapter lists its own additional
dependencies in its first cells.