Created
April 16, 2024 08:44
-
-
Save 1ort/28459e01944fd43df39241be5f476a63 to your computer and use it in GitHub Desktop.
Simple name generation via markov's chains
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import random | |
from collections import defaultdict | |
from itertools import islice | |
def batched(iterable, n): | |
"Batch data into tuples of length n. The last batch may be shorter." | |
# batched('ABCDEFG', 3) --> ABC DEF G | |
if n < 1: | |
raise ValueError('n must be at least one') | |
it = iter(iterable) | |
while batch := tuple(islice(it, n)): | |
yield batch | |
def batched_str(s, n): | |
for b in batched(s, n): | |
yield ''.join(b) | |
class MarkovChain: | |
def __init__(self, token_max_length): | |
self.transition_matrix = defaultdict(lambda: defaultdict(int)) | |
self.token_max_length = token_max_length | |
def train(self, words): | |
for word in words: | |
word = word.lower().strip() | |
tokens = batched_str(word, self.token_max_length) | |
last_token = None | |
for token in tokens: | |
self.transition_matrix[last_token][token] += 1 | |
last_token = token | |
self.transition_matrix[last_token][None] += 1 | |
# for i in range(len(name)): | |
# current_state = name[i:i+n] | |
# next_state = name[i+n] | |
# if current_state not in self.transition_matrix: | |
# self.transition_matrix[current_state] = {} | |
# if next_state not in self.transition_matrix[current_state]: | |
# self.transition_matrix[current_state][next_state] = 0 | |
# self.transition_matrix[current_state][next_state] += 1 | |
def next_state(self, current_state): | |
current_state = current_state or None | |
weights = list(self.transition_matrix[current_state].values()) | |
next_states = list(self.transition_matrix[current_state].keys()) | |
# print(current_state, weights, next_states) | |
return random.choices(next_states, weights=weights)[0] | |
def generate_name(self, max_length=10): | |
name_parts = ["", ] | |
while sum(len(part) for part in name_parts) < max_length: | |
next_token = self.next_state(name_parts[-1]) | |
if next_token is None: | |
break | |
name_parts.append(next_token) | |
return "".join(name_parts) | |
# Example usage | |
names = [ | |
"Lagdush", | |
"Groduf", | |
"Buga", | |
"Uglush", | |
"Lurtzog", | |
"Lugduf", | |
"Grat", | |
"Gorkil", | |
"Bolga", | |
"Snakhak", | |
"Mega", | |
"Luga", | |
"Gorkil", | |
"Balagd", | |
"Grat", | |
"Rat", | |
"Agdur", | |
"Balurtz", | |
"Lagduf", | |
"Lurtzog", | |
"Feanore", | |
"Fingormin", | |
"Ithil", | |
"Celenwe", | |
"Makili", | |
"Pengoli", | |
"Throdore", | |
"Maedhrondir", | |
"Gilgali", | |
"Ebrin", | |
"Eneleg", | |
"Galadel", | |
"Alamras", | |
"Irdahil", | |
"Enlor", | |
"Elelung", | |
"Elron", | |
"Araliod", | |
"Finore", | |
"Gelmire", | |
] | |
for token_length in range(1, 2): | |
print("token length: ", token_length) | |
markov_chain = MarkovChain(token_length) | |
markov_chain.train(names) | |
for i in range(10): | |
print(markov_chain.generate_name().capitalize()) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment