| from nltk import pos_tag, word_tokenize | |
| def yodinator(text): | |
| text = word_tokenize(text) | |
| tagged = pos_tag(text) | |
| verbs = ("MD", "VB", "VBD", "VBG", "VBN", "VBP", "VBZ", "RB", "RBR", "RBS") | |
| adverbs = ("RB", "RBR", "RBS") | |
| first_half = [] | |
| second_half = [] | |
| for t in range(0, len(tagged)): | |
| if tagged[t][1] in verbs and not tagged[t+1][1] in verbs: | |
| first_half.extend(tagged[:t+1]) | |
| second_half.extend(tagged[t+1:]) | |
| break | |
| elif tagged[t][1] in verbs and tagged[t+1][1] in verbs: | |
| first_half.extend(tagged[:t+1]) | |
| second_half.extend(tagged[t+1:]) | |
| break | |
| added = second_half + first_half | |
| sentence = [item[0] for item in added if item[0] not in (".",",")] | |
| if "?" in sentence: | |
| sentence.append(sentence.pop(sentence.index("?"))) | |
| sentence = " ".join(sentence) | |
| return sentence |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment