evaluate

evaluate(
  self,
  examples: Sequence[dict[str, str]],
  predictions: Sequence[dict[str, str]],
  *,
  question_key: str = 'context',
  answer_key: str = 'reference',
  prediction_key: str = 'prediction',
  instruction_key: str = 'instruction',
  **kwargs: Any = {}
) -> list[dict]