👋 Introducing the Rasa Playground

Learn more here →

CALM Video Tutorial Online Now!

You have selected 0 posts.

post by Rasalene on Mar 25, 2024

We’ve created a series of educational videos on our YouTube channel. These videos provide step-by-step guides, practical tips, and insights to help you maximise the potential of the CALM Developer Edition. Check out the first 5 episodes on YouTube and stay tuned for the second round of videos. Get the developer edition and code along with our new CALM videos.

post by geeta.m.desai on Mar 25, 2024

Thank you for sharing. I am running my vocie bot with RASA-CALM but taht is very slow. Even for single greet, it takes 7-8 seconds as it hits LLM multiple times. Please see logs below. To avoid that I used both NLUCommandAdapter and LLMCommandGenerator and provided NLU training data for specific intents but that also did not help.

2024-03-25 18:41:20 INFO root - Rasa server is up and running.
2024-03-25 18:41:20 INFO root - Enabling coroutine debugging. Loop id 140248276217872.
Bot loaded. Type a message and press enter (use ‘/stop’ to exit):
Your input → Howdy
2024-03-25 18:56:12 DEBUG rasa.core.lock_store - Issuing ticket for conversation ‘8f52c34f7281409e8286f6b92ee36efb’. 
2024-03-25 18:56:12 DEBUG rasa.core.lock_store - Acquiring lock for conversation ‘8f52c34f7281409e8286f6b92ee36efb’. 
2024-03-25 18:56:12 DEBUG rasa.core.lock_store - Acquired lock for conversation ‘8f52c34f7281409e8286f6b92ee36efb’. 
2024-03-25 18:56:12 DEBUG rasa.core.tracker_store - Could not find tracker for conversation ID ‘8f52c34f7281409e8286f6b92ee36efb’. 
2024-03-25 18:56:12 DEBUG rasa.core.tracker_store - No event broker configured. Skipping streaming events.
2024-03-25 18:56:12 DEBUG rasa.core.processor - Starting a new session for conversation ID ‘8f52c34f7281409e8286f6b92ee36efb’. 
2024-03-25 18:56:12 DEBUG rasa.core.processor - [debug] processor.actions.policy_prediction action_name=action_session_start policy_name=None prediction_events=
2024-03-25 18:56:12 DEBUG rasa.core.processor - [debug] processor.actions.log action_name=action_session_start rasa_events=[SessionStarted(type_name: session_started), ActionExecuted(action: action_listen, policy: None, confidence: None)]
2024-03-25 18:56:12 DEBUG rasa.core.processor - [debug] processor.slots.log slots={'reminder_callback_pending': False}
2024-03-25 18:56:12 DEBUG rasa.engine.runner.dask - Running graph with inputs: {'message': [UserMessage(text: Howdy, sender_id: 8f52c34f7281409e8286f6b92ee36efb)], 'tracker': DialogueStateTracker(sender_id: 8f52c34f7281409e8286f6b92ee36efb)}, targets: ['run_RegexMessageHandler'] and ExecutionContext(model_id='3bf778a5905a49d7a6fc7af246dd986e', should_add_diagnostic_data=False, is_finetuning=False, node_name=None).
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=NLUMessageConverter fn=convert_user_message node_name=nlu_message_converter
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=WhitespaceTokenizer fn=process node_name=run_WhitespaceTokenizer0
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=LexicalSyntacticFeaturizer fn=process node_name=run_LexicalSyntacticFeaturizer1
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=CountVectorsFeaturizer fn=process node_name=run_CountVectorsFeaturizer2
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=DIETClassifier fn=process node_name=run_DIETClassifier3
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=EntitySynonymMapper fn=process node_name=run_EntitySynonymMapper4
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=FlowsProvider fn=provide_inference node_name=flows_provider
2024-03-25 18:56:12 DEBUG rasa.engine.graph - [debug] graph.node.running_component clazz=LLMCommandGenerator fn=process node_name=run_LLMCommandGenerator5
2024-03-25 18:56:12 DEBUG rasa.utils.log_utils - [debug] llm_command_generator.predict_commands.prompt_rendered prompt=Your task is to analyze the current conversation context and generate a list of actions to start new business processes that we call flows, to extract slots, or respond to small talk and knowledge requests.

Related topics