Policies

The rasa.core.policies.Policy class decides which action to take at every step in the conversation.

There are different policies to choose from, and you can include multiple policies in a single rasa.core.agent.Agent.

Note

Per default a maximum of 10 next actions can be predicted by the agent after every user message. To update this value you can set the environment variable MAX_NUMBER_OF_PREDICTIONS to the desired number of maximum predictions.

Your project’s config.yml file takes a policies key which you can use to customize the policies your assistant uses. In the example below, the last two lines show how to use a custom policy class and pass arguments to it.

policies: - name: "KerasPolicy" featurizer: - name: MaxHistoryTrackerFeaturizer max_history: 5 state_featurizer: - name: BinarySingleStateFeaturizer - name: "MemoizationPolicy" max_history: 5 - name: "FallbackPolicy" nlu_threshold: 0.4 core_threshold: 0.3 fallback_action_name: "my_fallback_action" - name: "path.to.your.policy.class" arg1: "..."

Configuring Policies

The max_history is an important hyperparameter that controls how much dialogue history the model looks at to decide which action to take next.

You can set the max_history by passing it to your policy’s Featurizer in the policy configuration yaml file.

Note

Only the MaxHistoryTrackerFeaturizer uses a max history, whereas the FullDialogueTrackerFeaturizer always looks at the full conversation history.

Data Augmentation

When you train a model, by default Rasa Core will create longer stories by randomly combining the ones in your stories files.

You can alter this behavior with the --augmentation flag, which allows you to set the augmentation_factor. The augmentation_factor determines how many augmented stories are subsampled during training.

Action Selection

At every turn, each policy defined in your configuration will predict a next action with a certain confidence level. The bot’s next action is then decided by the policy that predicts with the highest confidence.

Keras Policy

The KerasPolicy uses a neural network implemented in Keras to select the next action. The default architecture is based on an LSTM, but you can override the model_architecture method to implement your own architecture.

def model_architecture(self, input_shape: Tuple[int, int], output_shape: Tuple[int, Optional[int]]) -> tf.keras.models.Sequential: # Build Model model = Sequential() model.add(Masking(mask_value=-1, input_shape=input_shape)) model.add(LSTM(self.rnn_size, dropout=0.2)) model.add(Dense(input_dim=self.rnn_size, units=output_shape[-1])) model.add(Activation("softmax")) model.compile(loss="categorical_crossentropy", optimizer="rmsprop", metrics=["accuracy"])

Memoization Policy

The MemoizationPolicy memorizes the conversations in your training data and predicts the next action with confidence 1.0 if this exact conversation exists in the training data.

Fallback Policy

The FallbackPolicy invokes a fallback action if:

1. The intent recognition has a confidence below nlu_threshold.
2. The highest ranked intent differs in confidence from the second highest ranked intent by less than ambiguity_threshold.
3. None of the dialogue policies predict an action with confidence higher than core_threshold.

Configuration:

policies:
  - name: "FallbackPolicy"
    nlu_threshold: 0.3
    ambiguity_threshold: 0.1
    core_threshold: 0.3
    fallback_action_name: 'action_default_fallback'

Two-Stage Fallback Policy

The TwoStageFallbackPolicy is designed to handle low NLU confidence in multiple stages.

Configuration:

policies:
  - name: TwoStageFallbackPolicy
    nlu_threshold: 0.3
    ambiguity_threshold: 0.1
    core_threshold: 0.3
    fallback_core_action_name: "action_default_fallback"
    fallback_nlu_action_name: "action_default_fallback"
    deny_suggestion_intent_name: "out_of_scope"

Form Policy

The FormPolicy is an extension of the MemoizationPolicy that handles the filling of forms. Once a FormAction is called, the FormPolicy will predict the FormAction until all required slots are filled.