langchain.evaluation.loading.load_evaluator

langchain.evaluation.loading.load_evaluator(evaluator: EvaluatorType, *, llm: Optional[BaseLanguageModel] = None, **kwargs: Any) → Union[Chain, StringEvaluator]

加载指定的评估链。

evaluator：EvaluatorType

要加载的评估器类型。

llm：BaseLanguageModel，可选

用于评估的语言模型，默认为None。

**kwargs：Any

要传递给评估器的其他关键字参数。

Chain

加载的评估链。

>>> from langchain.evaluation import load_evaluator, EvaluatorType
>>> evaluator = load_evaluator(EvaluatorType.QA)

Parameters

• evaluator (EvaluatorType) –

• llm (Optional[BaseLanguageModel]) –

• kwargs (Any) –

Return type

Union[Chain, StringEvaluator]

Examples using load_evaluator

• ANTHROPIC_API_KEY=<YOUR ANTHROPIC API KEY>

• Alternatively

• Check for the presence of a YYYY-MM-DD string.

• Correct

• Initialize the language model

• The prompt was assigned to the evaluator

• The results purely character-based, so it’s less useful when negation is concerned

• This is equivalent to loading using the enum

• You can load by enum or by raw python string

• pairwise_embedding_distance.md