ThinkBayes is the code repository accompanying Think Bayes: a book on Bayesian statistics written in a computational style. Instead of heavy focus on continuous mathematics or calculus, the book emphasizes learning Bayesian inference by writing Python programs. The project includes code examples, scripts, and environments that correspond to the chapters of the book. Learners can run the code, experiment with probability distributions, compute posterior probabilities, and understand Bayesian updating via simulation and algorithmic methods. The book and code encourage thinking in terms of discrete approximations (sums over distributions) rather than continuous integrals, making it more accessible to many programmers. Over time, the repository has been updated (including a second edition version) to reflect improved practices, corrections, and modern Python tooling.
Features
- Python code examples tied to each chapter of Think Bayes
- Scripts and modules for Bayesian updating, probability distributions, and inference
- Use of discrete approximations (sums) rather than continuous integrals
- Exercises and runnable notebooks for readers to experiment with
- Version control and updates corresponding to revised editions
- Educational focus: code-driven learning of Bayesian statistics