Last active
January 14, 2020 23:14
-
-
Save lizadaly/8ea2175c694542e734feabe369744b3d to your computer and use it in GitHub Desktop.
Very simple Markov chain example
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 Counter | |
import nltk | |
# See https://gist.github.com/lizadaly/7071e0de589883a197433951bc7314c5 for comments on the setup here | |
word_list = [] | |
[word_list.extend(nltk.corpus.gutenberg.words(f)) for f in nltk.corpus.gutenberg.fileids()] | |
cleaned_words = [w.lower() for w in word_list if w.isalnum()] | |
all_bigrams = [b for b in nltk.bigrams(cleaned_words)] | |
# Decide on a start word for the sentence | |
start_word = 'sun' | |
# Print a 10 word sentence using the set of bigrams | |
for i in range(0, 10): | |
# Find all word pairs that start with our target word | |
# e.g. "sun shines", "sun sets" | |
start_bigrams = [b for b in all_bigrams if b[0] == start_word] | |
# Pick a random one of those, e.g. "sun shines" | |
start_bigram = random.choice(start_bigrams) | |
# Print that first word ("sun") | |
print(start_bigram[0], end=" ") | |
# Set our next word to be the following word ("shines") | |
start_word = start_bigram[1] | |
# Loop will start over, finding the next two-word phrase that starts with "shines" | |
# e.g. "shines brightly" | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Thanks for this! Simple and effective!