Getting Error by using interactive learning

Rasa Open Source
1

when i run interactive learning i get this error , can anyone explain it to me ?: Exception: BinarySingleStateFeaturizer was not prepared before encoding

2

Can you provide a full traceback?

3

Hey,

I’m having the same issue

Traceback (most recent call last):
  File "/Users/wassim/Desktop/Com/TQA/AlfaChatbot/alfachatbot/alfachatbot/runners/core_runner/core_runner.py", line 37, in <module>
    CoreRunner.run()
  File "/Users/wassim/Desktop/Com/TQA/AlfaChatbot/alfachatbot/alfachatbot/runners/core_runner/core_runner.py", line 32, in run
    x = asyncio.run(agent.handle_text(input_message))
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/asyncio/runners.py", line 43, in run
    return loop.run_until_complete(main)
  File "uvloop/loop.pyx", line 1451, in uvloop.loop.Loop.run_until_complete
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/agent.py", line 524, in handle_text
    return await self.handle_message(msg, message_preprocessor)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/agent.py", line 446, in handle_message
    return await processor.handle_message(message)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/processor.py", line 74, in handle_message
    await self._predict_and_execute_next_action(message, tracker)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/processor.py", line 336, in _predict_and_execute_next_action
    action, policy, confidence = self.predict_next_action(tracker)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/processor.py", line 164, in predict_next_action
    probabilities, policy = self._get_next_action_probabilities(tracker)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/processor.py", line 549, in _get_next_action_probabilities
    tracker, self.domain
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/policies/ensemble.py", line 343, in probabilities_using_best_policy
    probabilities = p.predict_action_probabilities(tracker, domain)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/policies/keras_policy.py", line 252, in predict_action_probabilities
    X = self.featurizer.create_X([tracker], domain)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/featurizers.py", line 408, in create_X
    X, _ = self._featurize_states(trackers_as_states)
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/featurizers.py", line 335, in _featurize_states
    self.state_featurizer.encode(state) for state in tracker_states
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/featurizers.py", line 335, in <listcomp>
    self.state_featurizer.encode(state) for state in tracker_states
  File "/Users/wassim/anaconda/anaconda3/envs/alfa-chatbot/lib/python3.7/site-packages/rasa/core/featurizers.py", line 102, in encode
    "BinarySingleStateFeaturizer was not prepared before encoding."
Exception: BinarySingleStateFeaturizer was not prepared before encoding.

This is where the exception is being thrown

    def encode(self, state: Dict[Text, float]) -> np.ndarray:
        """Returns a binary vector indicating which features are active.

            Given a dictionary of states (e.g. 'intent_greet',
            'prev_action_listen',...) return a binary vector indicating which
            features of `self.input_features` are in the bag. NB it's a
            regular double precision float array type.

            For example with two active features out of five possible features
            this would return a vector like `[0 0 1 0 1]`

            If intent features are given with a probability, for example
            with two active features and two uncertain intents out
            of five possible features this would return a vector
            like `[0.3, 0.7, 1.0, 0, 1.0]`.

            If this is just a padding vector we set all values to `-1`.
            padding vectors are specified by a `None` or `[None]`
            value for states.
        """

        if not self.num_features:
            raise Exception(
                "BinarySingleStateFeaturizer was not prepared before encoding."
            )

        if state is None or None in state:
            return np.ones(self.num_features, dtype=np.int32) * -1

NLU Config : pipeline: "supervised_embeddings"

Policies

policies:
  - name: KerasPolicy
    epochs: 100
    max_history: 5
  - name: FallbackPolicy
    fallback_action_name: 'action_default_fallback'
  - name: MemoizationPolicy
    max_history: 5
4

Thanks for the traceback @wassimseif. What version of core is this on?

5

I was using 1.0.0rc6

6

Hm @wassimseif seems like you are using your own runner to run the code. Do you still run into the same problem using the CLI to train and run the bot?

7

This is fixed. It was just a bug in the nlu training so the core runners are not working.

About the custom runner, for production use, should i use the cli or custom runner to deploy the chatbot ?

8

I think that’s up to you :slight_smile: I only asked about the CLI because I wanted to know if it was your code or ours that was throwing the error.