valerionew / wordle-solver

My attempt at solving wordle puzzles using bayesian statistcs

wordle-solver

My attempt at solving wordle puzzles in HARD MODE using bayesian statistcs

If you're even lazier, check out https://github.com/lorossi/wordle-solution

italian words taken from: https://github.com/pietroppeter/wordle-it/blob/e490781af9ff34a0294e3c9da3a5c0941ab94a7b/wordle-it.js#L942-L943

Performance

ITA after "curated dictionary: average 3.6781 after curated

profile  = [1, 122, 554, 588, 209, 42, 6]
avg = profile*[1:7]'/sum(profile)
bar(profile)
title("Wordle solver guess attempts in ITA after 'curated dict' (7=not solved)")

ITA: average 3.8984 guesses before curated

profile  = [1, 158, 804, 917, 442, 138,50];
avg = profile*[1:7]'/sum(profile)
bar(profile)
title("Wordle solver guess attempts in ITA (7=not solved)")

ENG: average 3.7460 guesses

profile  = [1, 146, 821, 926, 336, 63,22];
avg = profile*[1:7]'/sum(profile)
bar(profile)
title("Wordle solver guess attempts in ENG (7=not solved)")

Notes for myself of the future

To run the profiler, run:

python -m cProfile -s tottime myscript.py