Created
September 22, 2015 02:27
-
-
Save Lambdanaut/786923a53d9ec895cd73 to your computer and use it in GitHub Desktop.
This file contains hidden or 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 | |
| class MM(object): | |
| def __init__(self): | |
| # Wordlist form = ['this','is','words'] | |
| # Graph form = [((1,1)],[(2,1)],[]] | |
| # [((count, index)],[(count, index)],[]] | |
| self.wordlist = [] | |
| self.graphlist = [] | |
| def get_words(self, i): | |
| i = ''.join(c for c in i if c.isalnum() or c == ' ' or c=='.').lower() | |
| words = i.split(' ') | |
| return words | |
| def add_to_graph(self, word, next_word=None): | |
| if next_word is not None: | |
| try: | |
| next_word_i = self.wordlist.index(next_word) | |
| except ValueError: | |
| # Next word hasn't been indexed | |
| next_word_i = self.add_to_graph(next_word) | |
| try: | |
| # Word has been indexed | |
| word_i = self.wordlist.index(word) | |
| word_neighbors = self.graphlist[word_i] | |
| if next_word is not None: | |
| # Messy updating code | |
| next_word_neighbors_i = 0 | |
| for _, i in word_neighbors: | |
| if i == next_word_i: | |
| break | |
| else: | |
| next_word_neighbors_i += 1 | |
| else: | |
| # First time next_word has appeared after word | |
| word_neighbors.append([0, next_word_i]) | |
| next_word_neighbors_count, _ = word_neighbors.pop(next_word_neighbors_i) | |
| word_neighbors.append((next_word_neighbors_count + 1, next_word_i)) | |
| self.graphlist[word_i] = word_neighbors | |
| except ValueError: | |
| # Word hasn't been indexed | |
| if next_word is None: | |
| self.graphlist.append([]) | |
| else: | |
| self.graphlist.append([(1, next_word_i)]) | |
| self.wordlist.append(word) | |
| word_i = len(self.wordlist) - 1 | |
| self.graphlist[word_i].sort(reverse=True) | |
| return word_i | |
| def learn(self, i): | |
| words = self.get_words(i) | |
| for w in range(len(words)): | |
| try: | |
| next_word = words[w+1] | |
| except: break | |
| word = words[w] | |
| self.add_to_graph(word, next_word) | |
| def generate(self, starting_word=None, max_cycles=5000): | |
| generated = '' | |
| if starting_word is None: | |
| word = random.choice(self.wordlist) | |
| else: | |
| word = starting_word | |
| for cycle in range(max_cycles): | |
| generated += '{} '.format(word) | |
| try: | |
| word_i = self.wordlist.index(word) | |
| except ValueError: | |
| 'ERROR' | |
| word_neighbors = self.graphlist[word_i] | |
| try: | |
| next_word_i = int(max(0, min(len(word_neighbors)-1, random.gauss(0, len(word_neighbors)/2)))) | |
| _, best_neighbor = word_neighbors[next_word_i] | |
| except IndexError: | |
| break | |
| word = self.wordlist[best_neighbor] | |
| return generated | |
| if __name__ == '__main__': | |
| mm = MM() | |
| f = open('lambdanaut.txt', 'r') | |
| mm.learn(f.read()) | |
| while True: | |
| i = raw_input('Input?') | |
| mm.learn(i) | |
| print mm.generate() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment