Home
Platyrhynchos is a project centered around creating word-centric crossword generation algorithms. This can be viewed as an alternative to methods that focus on generating matrices filled with letters and mutating them to make them into crosswords.
The first method combines first improvement and best improvement local search methods to concatenate together crosswords. The choice of which one to use is made using the temperature calculated for a given turn. The second algorithm is a simulated annealing algorithm that uses the best improvement search and word removal operation. The crosswords are evaluated using a goal function that includes the number of intersections and the density of letters in the crossword.
Currently, we are investigating a "smart" word insertion method that finds the worst rows and columns and tries to improve their score.
Please get in touch with the current contributors!
We're currently implementing the doctest library as our project's default method for creating tests. You can run them using poetry run python tests.py. Remember you need to first download the doctests by running git submodule update --init --recursive in the repo root. This is where you would place more advanced descriptions of your work. Although, please add docstrings.