Regexp Tokenizer In NLP

from nltk.tokenize import RegexpTokenizer

text = "I have 2 dogs, 3 cats, and a parrot!"

# Only words (no punctuation, no numbers)
tokenizer = RegexpTokenizer(r'[A-Za-z]+')
print(tokenizer.tokenize(text))
# Output: ['I', 'have', 'dogs', 'cats', 'and', 'a', 'parrot']

# Words + numbers
tokenizer = RegexpTokenizer(r'\w+')
print(tokenizer.tokenize(text))
# Output: ['I', 'have', '2', 'dogs', '3', 'cats', 'and', 'a', 'parrot']

Query/Feedback