evaluation_data_multiple_expected_responses.yaml

- conversation_group_id: conv_group_1
  description: Basic AI assistant interaction testing faithfulness and relevancy

  conversation_metrics: []
  conversation_metrics_metadata: {}
  turns:
    - turn_id: turn_1
      query: Who are you?
      attachments: null
      response: I am an AI assistant designed to help you find information, answer questions, and assist with various tasks. How can I help you today?
      tool_calls: null
      contexts:
        - "AI assistants are software programs designed to help users with information retrieval and task completion."
        - "This AI assistant can answer questions, provide information, and help with various user requests."
      expected_response: ["I'm a human assistant. How can I help you today?",
                          "I am an AI assistant designed to help you find information, answer questions, and assist with various tasks. How can I help you today?"]
      expected_tool_calls: null
      turn_metrics:
        - ragas:context_recall
      turn_metrics_metadata:
        ragas:context_recall:
          threshold: 0.5
      verify_script: null
  setup_script: null
  cleanup_script: null