paip-python

Python implementations of some of the classic AI programs from Peter Norvig's fantastic textbook "Paradigms of Artificial Intelligence Programming."

About

This is meant to be a learning resource for beginning AI programmers. Although PAIP is a fantastic book, it is no longer common for students to have a background in Lisp programming, as many universities have replaced Lisp with other languages in introductory programming and introductory artificial intelligence courses. It is my hope that making the programs from PAIP available in a commonly-taught language will provide a useful hands-on resource for beginning AI students.

The following programs are available, each with annotated source:

• General Problem Solver, means-ends analysis problem solving
• Eliza, a pattern-matching psychiatrist
• Search, a collection of search algorithms
• Logic, a library for logic programming
• Prolog, a basic Prolog interpreter
• Emycin, an expert system shell
• Othello, some game-playing strategies for the Othello board game

Unit tests and some example applications are provided for each of these; see the paip/tests and paip/examples directories or the links from the annotated sources.

Getting Started

Get the source code from GitHub or just download the latest version.

Also make sure you have Python 2.7.

• To run the examples: python run_examples.py and follow the prompts.
• To run the Prolog interpreter: ./prolog.py. Pass the -h flag for more details on its use and capabilities.
• To run the unit tests: python run_tests.py.
• To build the documentation: python build_docs.py.

Contributing

• fork on GitHub
• write code in paip/
• add unit tests in paip/tests
• make sure all tests pass: python run_tests.py
• send me a pull request

Author

These programs were written by Daniel Connelly at Georgia Tech as an independent project supervised by Professor Ashok Goel.