Created
March 26, 2021 05:36
-
-
Save microamp/5ab2feaaf06802ecebdd6fbf8071940e to your computer and use it in GitHub Desktop.
A super simple implementation of Markov chain
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 collections | |
import random | |
def iter_words(rows): | |
return ( | |
s.strip() for row in rows for s in row.split(" ") if s.strip() != "" | |
) | |
def iter_partitions(ss, size): | |
ls = list(ss) | |
for i in range(len(ls) - size + 1): | |
yield ls[i : i + size] | |
def build_chain(partitions, prefix_len): | |
d = collections.defaultdict(list) | |
for p in partitions: | |
k = tuple(p[:prefix_len]) | |
v = p[-1] | |
d[k].append(v) | |
return d | |
def generate(chain, prefix): | |
for word in prefix: | |
yield word | |
# count = 0 | |
while True: | |
if prefix[-1].endswith("."): | |
break | |
words = chain[prefix] | |
selected = random.choice(words) | |
yield selected | |
# count += 1 | |
prefix = prefix[1:] + (selected,) | |
filename = "/home/microamp/src/github.com/microamp/microq/cmd/poetry-archive/poems.txt" | |
with open(filename, mode="r") as f: | |
prefix_len = 2 | |
words = iter_words(f.readlines()) | |
partitions = iter_partitions(words, prefix_len + 1) | |
chain = build_chain(partitions, prefix_len) | |
# most_common = sorted( | |
# ((k, len(chain[k])) for k in chain), | |
# key=lambda pair: pair[1], | |
# reverse=True, | |
# )[:30] | |
# # pprint.pprint(most_common) | |
# prefixes = [pair[0] for pair in most_common] | |
prefixes = (("through", "the"),) | |
generated = generate(chain, random.choice(prefixes)) | |
for word in generated: | |
print(word, end=" ") | |
print() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment