Version: 2.x

rasa.core.featurizers.single_state_featurizer

SingleStateFeaturizer Objects

class SingleStateFeaturizer()

Base class to transform the dialogue state into an ML format.

Subclasses of SingleStateFeaturizer will decide how a bot will transform the dialogue state into a dictionary mapping an attribute to its features. Possible attributes are: INTENT, TEXT, ACTION_NAME, ACTION_TEXT, ENTITIES, SLOTS and ACTIVE_LOOP. Each attribute will be featurized into a list of rasa.utils.features.Features.

__init__

| __init__() -> None

Initialize the single state featurizer.

prepare_for_training

| prepare_for_training(domain: Domain, interpreter: NaturalLanguageInterpreter, bilou_tagging: bool = False) -> None

Gets necessary information for featurization from domain.

Arguments:

  • domain - An instance of :class:rasa.shared.core.domain.Domain.
  • interpreter - The interpreter used to encode the state
  • bilou_tagging - indicates whether BILOU tagging should be used or not

encode_state

| encode_state(state: State, interpreter: NaturalLanguageInterpreter) -> Dict[Text, List[Features]]

Encodes the given state with the help of the given interpreter.

Arguments:

  • state - The state to encode
  • interpreter - The interpreter used to encode the state

Returns:

A dictionary of state_type to list of features.

encode_entities

| encode_entities(entity_data: Dict[Text, Any], interpreter: NaturalLanguageInterpreter, bilou_tagging: bool = False) -> Dict[Text, List[Features]]

Encodes the given entity data with the help of the given interpreter.

Produce numeric entity tags for tokens.

Arguments:

  • entity_data - The dict containing the text and entity labels and locations
  • interpreter - The interpreter used to encode the state
  • bilou_tagging - indicates whether BILOU tagging should be used or not

Returns:

A dictionary of entity type to list of features.

encode_all_labels

| encode_all_labels(domain: Domain, interpreter: NaturalLanguageInterpreter) -> List[Dict[Text, List[Features]]]

Encodes all labels from the domain using the given interpreter.

Arguments:

  • domain - The domain that contains the labels.
  • interpreter - The interpreter used to encode the labels.

Returns:

A list of encoded labels.

encode_all_actions

| encode_all_actions(domain: Domain, interpreter: NaturalLanguageInterpreter) -> List[Dict[Text, List[Features]]]

Encodes all actions from the domain using the given interpreter.

This method is deprecated and will be removed in Rasa Open Source 3.0.0 . It is recommended to use encode_all_labels instead.

Arguments:

  • domain - The domain that contains the actions.
  • interpreter - The interpreter used to encode the actions.

Returns:

A list of encoded actions.

IntentTokenizerSingleStateFeaturizer Objects

class IntentTokenizerSingleStateFeaturizer(SingleStateFeaturizer)

A SingleStateFeaturizer for use with policies that predict intent labels.

encode_all_labels

| encode_all_labels(domain: Domain, interpreter: NaturalLanguageInterpreter) -> List[Dict[Text, List[Features]]]

Encodes all relevant labels from the domain using the given interpreter.

Arguments:

  • domain - The domain that contains the labels.
  • interpreter - The interpreter used to encode the labels.

Returns:

A list of encoded labels.

BinarySingleStateFeaturizer Objects

class BinarySingleStateFeaturizer(SingleStateFeaturizer)

Dialogue State featurizer which features the state as binaries.

__init__

| __init__() -> None

Creates featurizer.