Version: 2.5.x
rasa.core.policies.ted_policy TEDPolicy Objects# Transformer Embedding Dialogue (TED) Policy is described in
https://arxiv.org/abs/1910.00486 .
This policy has a pre-defined architecture, which comprises the
following steps:
Copy - concatenate user input (user intent and entities), previous system actions,
slots and active forms for each time step into an input vector to
pre-transformer embedding layer;
- feed it to transformer;
- apply a dense layer to the output of the transformer to get embeddings of a
dialogue for each time step;
- apply a dense layer to create embeddings for system actions for each time
step;
- calculate the similarity between the dialogue embedding and embedded system
actions. This step is based on the StarSpace
(https://arxiv.org/abs/1709.03856) idea.
__init__# Copy | __init__ ( featurizer : Optional [ TrackerFeaturizer ] = None , priority : int = DEFAULT_POLICY_PRIORITY , max_history : Optional [ int ] = None , model : Optional [ RasaModel ] = None , fake_features : Optional [ Dict [ Text , List [ "Features" ] ] ] = None , entity_tag_specs : Optional [ List [ EntityTagSpec ] ] = None , should_finetune : bool = False , ** kwargs : Any , , ) - > None
Declare instance variables with default values.
train# Copy | train ( training_trackers : List [ TrackerWithCachedStates ] , domain : Domain , interpreter : NaturalLanguageInterpreter , ** kwargs : Any , , ) - > None
Train the policy on given training trackers.
predict_action_probabilities# Copy | predict_action_probabilities ( tracker : DialogueStateTracker , domain : Domain , interpreter : NaturalLanguageInterpreter , ** kwargs : Any , , ) - > PolicyPrediction
Predicts the next action the bot should take after seeing the tracker.
Arguments :
tracker
- the :class:rasa.core.trackers.DialogueStateTracker
domain
- the :class:rasa.shared.core.domain.Domain
interpreter
- Interpreter which may be used by the policies to create
additional features.Returns :
The policy's prediction (e.g. the probabilities for the actions).
persist# Copy | persist ( path : Union [ Text , Path ] ) - > None
Persists the policy to a storage.
load# Copy | @ classmethod
| load ( cls , path : Union [ Text , Path ] , should_finetune : bool = False , epoch_override : int = defaults [ EPOCHS ] , ** kwargs : Any , , ) - > "TEDPolicy"
Loads a policy from the storage.
Needs to load its featurizer
TED Objects# Copy class TED ( TransformerRasaModel )
__init__# Copy | __init__ ( data_signature : Dict [ Text , Dict [ Text , List [ FeatureSignature ] ] ] , config : Dict [ Text , Any ] , max_history_featurizer_is_used : bool , label_data : RasaModelData , entity_tag_specs : Optional [ List [ EntityTagSpec ] ] ) - > None
Intializes the TED model.
Arguments :
data_signature
- the data signature of the input dataconfig
- the model configurationmax_history_featurizer_is_used
- if 'True'
only the last dialogue turn will be usedlabel_data
- the label dataentity_tag_specs
- the entity tag specifications batch_loss# Copy | batch_loss ( batch_in : Union [ Tuple [ tf . Tensor ] , Tuple [ np . ndarray ] ] ) - > tf . Tensor
Calculates the loss for the given batch.
Arguments :
Returns :
The loss of the given batch.
prepare_for_predict# Copy | prepare_for_predict ( ) - > None
Prepares the model for prediction.
batch_predict# Copy | batch_predict ( batch_in : Union [ Tuple [ tf . Tensor ] , Tuple [ np . ndarray ] ] ) - > Dict [ Text , Union [ tf . Tensor , Dict [ Text , tf . Tensor ] ] ]
Predicts the output of the given batch.
Arguments :
Returns :
The output to predict.