Sudoku resolver

This project involved the development of a Python program designed to solve Sudoku puzzles using optimization techniques. The primary goal was to create an algorithm that not only accurately solves Sudoku puzzles but also does so in an efficient manner, improving both the quality and speed of the solution. The project was driven by a personal interest in enhancing my own method of solving Sudokus, with a focus on replicating and optimizing the mental strategies I use when playing the game.

The techniques used to improve the efficiency was: First, find and fill all cells that only has one possibility. Second, sort the white cells with the number of possibles values and resolve in this order. Then repeat the step 1 and two until the problem state is stable. Finally, use backtracking to resolve the rest of the cells.

The project required a deep understanding of Sudoku mechanics and the development of an algorithm capable of mimicking human problem-solving processes, breaking down the puzzle into manageable steps, and applying logical deductions systematically.

One of the main challenges of the project was to analyze and translate my personal approach to solving Sudokus into a clear, executable algorithm. This involved dissecting my mental process during gameplay, identifying patterns and strategies I intuitively use, and then coding these steps in a way that a machine could replicate. The project also included refining the algorithm to improve its efficiency, reducing the time required to solve puzzles, especially as they increase in difficulty.

Although the target audience for this project was primarily myself, the work provided valuable insights into algorithmic thinking and problem-solving in Python. It also offered a unique opportunity to bridge the gap between human intuition and machine logic, resulting in a tool that effectively mirrors the way I approach and solve Sudoku puzzles.