-import random, itertools, time, ast\r
+import random, itertools, time, numpy\r
\r
source_link = "https://matthewminer.name/projects/calculators/wordle-words-left/"\r
valid_list = [\r
\r
alphabet = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z']\r
\r
-def most_used_letters(words=valid_list):\r
+def most_used_letters(words=valid_list, letters=alphabet, print_result=True):\r
'''\r
Outputs how many times each letter is used in the words array.\r
'''\r
use_each_letter = {}\r
- for i in alphabet:\r
+ for i in letters:\r
count = 0\r
for word in words:\r
for letter in word:\r
break\r
use_each_letter[i] = count\r
use_each_letter = dict(sorted(use_each_letter.items(), key=lambda item: item[1], reverse=True))\r
- print("Letter | Usage | Percent")\r
- print("----------------------------")\r
- for k in use_each_letter:\r
- print(f"{k.upper()} | {use_each_letter[k]:5} | {round((100 * use_each_letter[k]) / len(words)):2}%")\r
+ if print_result:\r
+ print("Letter | Usage | Percent")\r
+ print("----------------------------")\r
+ for k in use_each_letter:\r
+ print(f"{k.upper()} | {use_each_letter[k]:5} | {round((100 * use_each_letter[k]) / len(words)):2}%")\r
return use_each_letter\r
\r
\r
prev = time.time()\r
start = prev\r
letters_to_ignore = ['D','M'] # Don't diplay well on the watch\r
- letter_usage = most_used_letters(words=words)\r
+ letter_usage = most_used_letters(words=words, print_result=False)\r
for letter in letter_usage:\r
if (100 * letter_usage[letter])/len(words) < min_letter_occ_percent_to_consider:\r
letters_to_ignore.append(letter)\r
dict_combos_counts = dict(sorted(dict_combos_counts.items(), key=lambda item: item[1], reverse=True))\r
\r
most_common_key = next(iter(dict_combos_counts))\r
- print(f"The Most Common Combo is: {most_common_key} with {dict_combos_counts[most_common_key]} words.")\r
+ print(f"The Most Common Combo for {num_letters_in_set} letters is: {most_common_key} with {dict_combos_counts[most_common_key]} words.")\r
# print_valid_words(ast.literal_eval(most_common_key)) # Uncomment to display the text it creates\r
\r
if txt_out:\r
print(f"{key}, {item}") \r
\r
\r
-def location_of_letters(letters=alphabet, list=valid_list):\r
- print(" 1 2 3 4 5 ")\r
- print("-----------------------------------------")\r
+def location_of_letters(letters=alphabet, list=valid_list, print_result=True):\r
+ letter_location_percentages = {}\r
+ if print_result:\r
+ print(" 1 2 3 4 5 ")\r
+ print("-----------------------------------------")\r
letters = sorted(letters)\r
for letter in letters:\r
location = [0, 0, 0, 0, 0]\r
for i, char in enumerate(word):\r
if char.upper() == letter.upper():\r
location[i]+=1\r
- location = [f"{round((100 * x) / sum(location)):2}%" for x in location]\r
- print(f"{letter} : {location}")\r
+ location = [((100 * x) / sum(location)) for x in location]\r
+ letter_location_percentages[letter] = location\r
+ location_txt = [f"{round(x):2}%" for x in location]\r
+ if print_result:\r
+ print(f"{letter} : {location_txt}")\r
+ return letter_location_percentages\r
+\r
+\r
+def best_first_word(letters=alphabet, list=valid_list, print_result=True, words_to_print=None):\r
+ '''\r
+ Word_good has every word with only unique letters as keys and that values are:\r
+ 1. Take the usage of every letter, normalize the max to 100 and the min to 0.\r
+ 2. Go through each letter in the word and see how often that letter appears in that exact location.\r
+ 3. Multiply that occurrance in location with the normalized total usage.\r
+ 4. Do this for each letter and add it all together.\r
+ \r
+ Ex: SLATE\r
+ Normalized usage: S=38, L=42, A=79, T=46, E=100\r
+ S in position 1: 55%\r
+ L in position 2: 27%\r
+ A in position 3: 31%\r
+ T in position 4: 19%\r
+ E in position 5: 34%\r
+ Total = (38*55) + (42*27) + (79*31) + (46*19) + (100*35) = 10047\r
+ ''' \r
+ valid_words = list_of_valid_words(letters, list)\r
+ letter_usage = most_used_letters(words=list, letters=letters, print_result=False)\r
+ a=[[max(letter_usage.values()),1],[min(letter_usage.values()),1]]\r
+ b=[100,0]\r
+ m,b = numpy.linalg.solve(a,b).tolist()\r
+ letter_usage_normalized = {key: ((m * value) + b) for key, value in letter_usage.items()}\r
+ loc_letters = location_of_letters(letters=letters, list=list, print_result=False)\r
+ word_good = {}\r
+ valid_words_unique = [word for word in valid_words if len(word) == len(set(word))]\r
+ for word in valid_words_unique:\r
+ usage = 0\r
+ for i, char in enumerate(word):\r
+ usage += loc_letters[char][i] * letter_usage_normalized[char]\r
+ word_good[word] = round(usage)\r
+ word_good = {k: v for k, v in sorted(word_good.items(), key=lambda item: item[1], reverse=True)}\r
+ if print_result:\r
+ print("Word, Usage Value")\r
+ print("------------------")\r
+ for i,[k,v] in enumerate(word_good.items()):\r
+ if words_to_print is not None and i+1 > words_to_print:\r
+ break\r
+ print(f"{k}, {v}")\r
+ return word_good\r
\r
\r
if __name__ == "__main__":\r
my_letters = ['A', 'C', 'E', 'H', 'I', 'L', 'N', 'O', 'P', 'R', 'S', 'T']\r
#print(f"{len(list_of_valid_words(my_letters, valid_list))} Words can be made with {my_letters}")\r
- #most_used_letters()\r
- #location_of_letters(my_letters)\r
+ #most_used_letters(letters=my_letters)\r
+ #location_of_letters(letters=my_letters)\r
print_valid_words(my_letters)\r
#txt_of_all_letter_combos_differing_sizes(max = 16, min=10)\r
- #txt_of_all_letter_combos(14)
\ No newline at end of file
+ #txt_of_all_letter_combos(14)\r
+ #best_first_word(letters=my_letters, print_result=True, words_to_print=10)
\ No newline at end of file
projects / sensor-watch.git / commitdiff