Skip to content

text_field

[ allennlp.data.fields.text_field ]


A TextField represents a string of text, the kind that you might want to represent with standard word vectors, or pass through an LSTM.

TextFieldTensors#

TextFieldTensors = Dict[str, Dict[str, torch.Tensor]]

TextField Objects#

class TextField(SequenceField[TextFieldTensors]):
 | def __init__(
 |     self,
 |     tokens: List[Token],
 |     token_indexers: Dict[str, TokenIndexer]
 | ) -> None

This Field represents a list of string tokens. Before constructing this object, you need to tokenize raw strings using a Tokenizer.

Because string tokens can be represented as indexed arrays in a number of ways, we also take a dictionary of TokenIndexer objects that will be used to convert the tokens into indices. Each TokenIndexer could represent each token as a single ID, or a list of character IDs, or something else.

This field will get converted into a dictionary of arrays, one for each TokenIndexer. A SingleIdTokenIndexer produces an array of shape (num_tokens,), while a TokenCharactersIndexer produces an array of shape (num_tokens, num_characters).

count_vocab_items#

 | @overrides
 | def count_vocab_items(self, counter: Dict[str, Dict[str, int]])

index#

 | @overrides
 | def index(self, vocab: Vocabulary)

get_padding_lengths#

 | @overrides
 | def get_padding_lengths(self) -> Dict[str, int]

The TextField has a list of Tokens, and each Token gets converted into arrays by (potentially) several TokenIndexers. This method gets the max length (over tokens) associated with each of these arrays.

sequence_length#

 | @overrides
 | def sequence_length(self) -> int

as_tensor#

 | @overrides
 | def as_tensor(
 |     self,
 |     padding_lengths: Dict[str, int]
 | ) -> Dict[str, torch.Tensor]

empty_field#

 | @overrides
 | def empty_field(self)

batch_tensors#

 | @overrides
 | def batch_tensors(
 |     self,
 |     tensor_list: List[TextFieldTensors]
 | ) -> TextFieldTensors

This is creating a dict of {token_indexer_name: {token_indexer_outputs: batched_tensor}} for each token indexer used to index this field.

duplicate#

 | @overrides
 | def duplicate(self)

Overrides the behavior of duplicate so that self._token_indexers won't actually be deep-copied.

Not only would it be extremely inefficient to deep-copy the token indexers, but it also fails in many cases since some tokenizers (like those used in the 'transformers' lib) cannot actually be deep-copied.