-
Notifications
You must be signed in to change notification settings - Fork 0
/
end_of_bayes.py
27 lines (21 loc) · 1005 Bytes
/
end_of_bayes.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
"""
1) guess from most common/necessary letters (vowels, <s>, etc)
1a) succeed
1b) fail
2) update possible words in priors based on (1a)'s inclusions and (1b)'s exclusions
repeat
3) assess if likelihood is high enough to be correct, ie play with a ratio of satisfactory probability, so it gets good enough at guessing without too much loss. add parameters: how many possible letters there are unguessed, how many possible words there are with those letters, how many limbs have been drawn
4) assess if it's worth guessing a word or a letter
4a) guess, success
4b) guess, fail, repeat
the likelihood of being correct in a whole word guess may end up saving our ass right?
if we random strings of arbitrary length but kept max loss at 6 (head, torso, four limbs), would that change it?
"""
import random
# assess ratio, return word or letter
# assess remaining possible strings
#
def executioner_begin():
dictionary = open("dictionary.txt", mode="r").readlines()
# if multiple
"The End of iCarly"