Guessing algrorithm for Wordle written in Python
Clone the repository, install the modules in requireents.txt and run solver.py
Once the algorithm has hints for a guess, it converts them into match data (known positions, impossible positions, minimum count and if there could be any more of the letter) for every letter in the alphablet.
Then based on each letter's match data words that won't be possible are removed.
After the list of possible words is shortend a word is chosen through the choose_word function to be the new guess to then get hints for and shorten the possiblities further.
With the solution as "aroma":
⬛🟨⬛⬛⬛ First guess is always salet (statistically the best)
⬛⬛🟨⬛🟩 Second guess is decor sice there's no more of a or n but there's c, r and e somewhere else
⬛🟩🟩⬛🟩 The guesses repeat the same process as the second until it reaches the solution
🟩🟩🟩🟩🟩 The solution is found by the fourth guess
For 1000 random solutions the algorithm averaged 4.708 tries to get to the solution.
Wordle.Solver.Demo.mp4
Given a list of possible words, if you rank every word by what ratio match it is to every guess made till now and choose the one with the lowest rank, you'll get a word that adds the maximum amount of new letters and positions to the existing pool of letters.
def choose_word(guesses: set, possible_words: set, randomize: bool = False):
"""Get an optimized choice of a word to be the next guess from the possible words"""
if randomize: return random.choice(list(possible_words))
comparison = {}
# The best next guess seems to be the one that differs most from the previous guesses
# since this diversifys the letters used therefore maximizing the hints received
for word in possible_words:
for guess in guesses:
# get the similarity between the word and the guess
comparison[word] = comparison.get(word, 0) + SequenceMatcher(
None, word, guess).ratio()
# number of total letters - number of unique letters
comparison[word] = comparison.get(word, 0) + len(word) - len(set(word))
return min(comparison, key=comparison.get)Send a pull request if you'd like to make a change or open an issue. You're free to use the guessing algorithm for your own implementations.