[ allennlp.data.tokenizers.whitespace_tokenizer ]
Tokenizer that assumes you've already done your own tokenization somehow and have
separated the tokens by spaces. We just split the input string on whitespace and return the
Note that we use
text.split(), which means that the amount of whitespace between the
tokens does not matter. This will never result in spaces being included as tokens.
Registered as a
Tokenizer with name "whitespace" and "just_spaces".
| @overrides | def tokenize(self, text: str) -> List[Token]