Skip to content

Latest commit

 

History

History
112 lines (79 loc) · 4.97 KB

README.md

File metadata and controls

112 lines (79 loc) · 4.97 KB
Raw

Test package - /test_scripts/

This is a collection of scripts I have used to test the sudoku solver. As there were quite a few of them, I bundled them into a package to stop test_sudoku.py from getting too long.

I have also included some test data that can be used with these scripts. These can be found in /data/.

As well as the data provided by the University of Bath, I have also included a /custom/ directory containing a few puzzles I have encountered that my solver has found particularly challenging. If you want more sudokus, please see here.

Note When printing the output of an unsolvable sudoku, the functions outlined here will use 'x' instead of -1. The solver will still output a 9x9 array of -1, as per the spec, but this increases the readability of the output in the terminal.

Note The scripts in this package (other than basic tests) do not check the validity of the solutions. They only check the time it takes to solve them.

Links

basic_tests

This is based on the sudoku testing program from Uni of Bath. It tests against each set of puzzles stored in /data/ and compares answers against the solutions (also stored in /data/).

It takes the argument use_process_time. If this is True, time.process_time() will be used to measure how long the sudoku solver takes. Otherwise it will use time.perf_counter.

Usage:

import test_scripts

test_scripts.basic_tests(use_process_time=...)

all_zeros_test

Tests the solver by giving it a blank sudoku, i.e. a (9,9) NumPy array of 0s.

It takes the argument use_process_time. If this is True, time.process_time() will be used to measure how long the sudoku solver takes. Otherwise it will use time.perf_counter.

Usage:

import test_scripts

test_scripts.all_zeros_test(use_process_time=...)

already_solved_test

Tests the solver by giving it a solved sudoku. The solver should return the sudoku as it is.

It takes the argument use_process_time. If this is True, time.process_time() will be used to measure how long the sudoku solver takes. Otherwise it will use time.perf_counter.

Usage:

import test_scripts

test_scripts.already_solved_test(use_process_time=...)

load_array

Loads an array from a .npy or a .csv file. If the file does not exist, or appears to be of incorrect format to be a sudoku, raise an error.

Usage:

import test_scripts

test_scripts.load_array(filename=...)

load_txt_file

Loads an array from a .txt file. If the file does not exist, or appears to be of incorrect format to be a sudoku, raise an error.

Usage:

import test_scripts

test_scripts.load_txt_file(filename=...)

multiple_batches

Runs a test similar to basic_tests a set number of times. This is also a lot less verbose than basic_test in that it does not print the sudokus or solutions. It also assumes that the sudoku solver is correct for each solve.

It takes two arguments:

  • use_process_time - If this is True, time.process_time() will be used to measure how long the sudoku solver takes. Otherwise it will use time.perf_counter.

  • num_of_batches - The number of batches to process (each batch = 1 run of basic_tests, 60 sudokus)

Usage:

import test_scripts

test_scripts.multiple_batches(num_of_batches=..., use_process_time=...)

solve_multiple

Solves multiple sudokus stored in a 3d array and prints the results.

It takes 4 arguments:

  • sudoku_array - the 3d array of sudokus to solve
  • save_to_file - whether or not to save results to a file. If you are not interested in the sudoku solutions and only want to know the processing time, then set this to False. Setting it to True will make the function return an array of the solutions, but it will significantly increase processing time.
  • use_process_time - If this is True, time.process_time() will be used to measure how long the sudoku solver takes. Otherwise it will use time.perf_counter.
  • quit_after - The number of sudokus to stop processing after.

Usage:

import test_scripts

test_scripts.solve_multiple(sudoku_array=...,
                            save_to_file_=...,
                            use_process_time=...,
                            quit_after=...)