Artificial Intelligence Nanodegree

Introductory Project: Diagonal Sudoku Solver

Question 1 (Naked Twins)

Q: How do we use constraint propagation to solve the naked twins problem?
A: We iterate over each of the units in the unitlist to find the twin pairs. If a pair is found then the individual values in this pair are removed from every other box in the same unit. The Naked twins problem is taken as a third constraint along with eliminate and only choice to reduce the possibilities in a given sudoku.

Question 2 (Diagonal Sudoku)

Q: How do we use constraint propagation to solve the diagonal sudoku problem?
A: To solve the diagonal sudoku problem add a new constraint as a unit to unitlist which consists of columns, rows and 3X3 box as units. Including the diagonals as units keeps the problem unchanged and only adds a new constraint in the problem definition. The rest of the code remains the same and can directly run.

Install

This project requires Python 3.

We recommend students install Anaconda, a pre-packaged Python distribution that contains all of the necessary libraries and software for this project. Please try using the environment we provided in the Anaconda lesson of the Nanodegree.

Optional: Pygame

Optionally, you can also install pygame if you want to see your visualization. If you've followed our instructions for setting up our conda environment, you should be all set.

If not, please see how to download pygame here.

Code

  • solutions.py - You'll fill this in as part of your solution.
  • solution_test.py - Do not modify this. You can test your solution by running python solution_test.py.
  • PySudoku.py - Do not modify this. This is code for visualizing your solution.
  • visualize.py - Do not modify this. This is code for visualizing your solution.

Visualizing

To visualize your solution, please only assign values to the values_dict using the assign_values function provided in solution.py

Data

The data consists of a text file of diagonal sudokus for you to solve.