Generic Iteratic Deepening package for python 2.7

Author: Chen Zhang
Email: [email protected]

Background

This package is originally a responce to Programming Assignment 1 of Aritifical Intelligence tough by Professor Ernest Davis with NYU.

1. The Post Correspondence Problem
2. Problem definition
3. Coding requirements

Disclaimer

If you want to use this software for academic purposes, aka assignments, please refer to the ACADEMIC INTEGRITY.

About

I tried to make the program generic. Some abstract classes are raised in the prgram, even though Python 2 does not have official support for that. The code is well-documented with docstring. It is not going to be hard to figure out the definition and implementation details of the phantom abstract classes.

• iterative_deepening.py contains a generic IterativeDeepening class good for Searchable objects. The abstract classes of Searchable and State are also implemented.
• dominos.py, is the code that solves the assignment problem. A DominoSpace (subclass of Searchable) and PostCorrespondenceState (subclass of State) are implemented, along with class Domino.
• iterative_deepening_test.py includes the unit test for the package. A commonly seen TreeNode class is defined, and is wraped up to TreeState by inheriting State defined in iterative_deepening.py.

The package provides the following search methods:

• BFS Standard Breadth-First Search w/ or w/out maxmium queue size and maximum number of states explored
• DFS Standard Depth-First Search w/ or w/out maximum number of states explored
• IterativeDeepening Standard Iterative Deepening w/ or w/out maximum number of states explored
• Search Iterative Deepening using a list of seeds initialized with BFS w/ maximum queue size.

The search methods returns the solution (None if unfound) and the error code

0 - sulution found
1 - no soluion exists
2 - solution not found within the constraints