Version: 3.x

rasa.nlu.extractors._crf_entity_extractor

CRFEntityExtractor Objects

class CRFEntityExtractor(EntityExtractor)

extract_entities

| extract_entities(message: Message) -> List[Dict[Text, Any]]

Extract entities from the given message using the trained model(s).

load

| @classmethod
| load(cls, meta: Dict[Text, Any], model_dir: Text, model_metadata: Metadata = None, cached_component: Optional["CRFEntityExtractor"] = None, **kwargs: Any, ,) -> "CRFEntityExtractor"

Loads trained component (see parent class for full docstring).

persist

| persist(file_name: Text, model_dir: Text) -> Optional[Dict[Text, Any]]

Persist this model into the passed directory.

Returns the metadata necessary to load the model again.