Getting response from Rasa without any user input

[hknajjar0]

As soon as I open the UI in browser, the UI displays a message when it normally should do action_listen.

Here's server start: ubuntu@ip-172-31-27-224$ docker run -p 5005:5005 -v $(pwd):/app rasa/rasa:latest-full run -p 5005 --cors "*" --enable-api --log-file out.log --debug 2019-12-13 04:08:31 DEBUG rasa.cli.utils - Parameter 'endpoints' not set. Using default location 'endpoints.yml' instead. 2019-12-13 04:08:31 DEBUG rasa.cli.utils - Parameter 'credentials' not set. Using default location 'credentials.yml' instead. 2019-12-13 04:08:32 DEBUG rasa.core.utils - Available web server routes: /conversations/<conversation_id>/messages POST add_message /conversations/<conversation_id>/tracker/events POST append_events /webhooks/rasa GET custom_webhook_RasaChatInput.health /webhooks/rasa/webhook POST custom_webhook_RasaChatInput.receive /webhooks/rest GET custom_webhook_RestInput.health /webhooks/rest/webhook POST custom_webhook_RestInput.receive /model/test/intents POST evaluate_intents /model/test/stories POST evaluate_stories /conversations/<conversation_id>/execute POST execute_action /domain GET get_domain / GET hello /model PUT load_model /model/parse POST parse /conversations/<conversation_id>/predict POST predict /conversations/<conversation_id>/tracker/events PUT replace_events /conversations/<conversation_id>/story GET retrieve_story /conversations/<conversation_id>/tracker GET retrieve_tracker /status GET status /model/predict POST tracker_predict /model/train POST train /model DELETE unload_model /version GET version 2019-12-13 04:08:32 INFO root - Starting Rasa server on http://localhost:5005 2019-12-13 04:08:32 DEBUG rasa.core.utils - Using the default number of Sanic workers (1). 2019-12-13 04:08:32 INFO root - Enabling coroutine debugging. Loop id 94360658814392. 2019-12-13 04:08:32 DEBUG rasa.model - Extracted model to '/tmp/tmpw2eav37k'. 2019-12-13 04:08:36.810930: E tensorflow/stream_executor/cuda/cuda_driver.cc:318] failed call to cuInit: UNKNOWN ERROR (303) 2019-12-13 04:08:37 DEBUG rasa.core.tracker_store - Connected to InMemoryTrackerStore. 2019-12-13 04:08:37 DEBUG rasa.core.lock_store - Connected to lock store 'InMemoryLockStore'. 2019-12-13 04:08:37 DEBUG rasa.model - Extracted model to '/tmp/tmpgj627mrc'. 2019-12-13 04:08:37 DEBUG pykwalify.compat - Using yaml library: /build/lib/python3.6/site-packages/ruamel/yaml/init.py 2019-12-13 04:08:38 DEBUG rasa.core.nlg.generator - Instantiated NLG to 'TemplatedNaturalLanguageGenerator'.

Then I open the UI after starting node.js server 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Creating a new tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15'. 2019-12-13 04:13:33 DEBUG rasa.core.policies.memoization - Current tracker state [None, None, None, {}, {'prev_action_listen': 1.0}] 2019-12-13 04:13:33 DEBUG rasa.core.policies.memoization - There is no memorised next action 2019-12-13 04:13:33 DEBUG rasa.core.policies.ensemble - Predicted next action using policy_1_KerasPolicy 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15' 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15' 2019-12-13 04:13:33 DEBUG rasa.core.processor - Action 'utter_kenz_inquiry' ended with events '['BotUttered(text: At Kenz, our ultimate goal is to bring out the peak taste in every bean. It starts with picking the best green beans and micro-roasting each batch, one pound at a time. It's best when we know how you prepare your coffee in order to best roast your beans., data: {"elements": null, "quick_replies": null, "buttons": null, "attachment": null, "image": null, "custom": null}, metadata: {})']' 2019-12-13 04:13:33 DEBUG rasa.core.lock_store - Deleted lock for conversation '7293772f-09f8-46c3-8719-9e5a10e13c15'. 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15' 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15' 2019-12-13 04:13:33 DEBUG rasa.core.policies.memoization - Current tracker state [None, None, {}, {'prev_action_listen': 1.0}, {'prev_utter_kenz_inquiry': 1.0}] 2019-12-13 04:13:33 DEBUG rasa.core.policies.memoization - There is no memorised next action 2019-12-13 04:13:33 DEBUG rasa.core.policies.mapping_policy - There is no mapped action for the predicted intent, 'None'. 2019-12-13 04:13:33 DEBUG rasa.core.policies.ensemble - Predicted next action using policy_1_KerasPolicy 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15' 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15' 2019-12-13 04:13:33 DEBUG rasa.core.processor - Action 'action_listen' ended with events '[]' 2019-12-13 04:13:33 DEBUG rasa.core.lock_store - Deleted lock for conversation '7293772f-09f8-46c3-8719-9e5a10e13c15'. 2019-12-13 04:13:33 DEBUG rasa.core.tracker_store - Recreating tracker for id '7293772f-09f8-46c3-8719-9e5a10e13c15'

This isn't a Rasa issue because I only see this when using Assister UI. I bolded likely culprit.

Best regards, HN

[keyvan-m-sadeghi]

Hi @hknajjar0, thanks for reporting this issue. I guess this has to do with the way I've currently coded the listener. See predictUntilListen recursive call. My reasoning was that a chatbot/assistant will always be the one to lead the conversation, though I can see scenarios where this doesn't hold. Can you play with the linked code and tell me what works for you?

[hknajjar0]

Thanks for quick turnaround, Keyvan. I agree in principle. However, it seems the chatbot initiates conversation in an inconsistent way. Is there a way to hard-code it such that it always starts with utter_greet action?