ContextRecall

class opik.evaluation.metrics.ContextRecall(model: str | OpikBaseModel | None = None, name: str = 'context_recall_metric', few_shot_examples: List[FewShotExampleContextRecall] | None = None, track: bool = True)

Bases: BaseMetric

A metric that evaluates the context recall of an input-output pair using an LLM.

This metric uses a language model to assess how well the given output incorporates the provided context for the given input. It returns a score between 0.0 and 1.0, where higher values indicate better context recall.

Parameters:

  • model – The language model to use for evaluation. Can be a string (model name) or an opik.evaluation.models.OpikBaseModel subclass instance. opik.evaluation.models.LiteLLMChatModel is used by default.

  • name – The name of the metric. Defaults to “ContextRecallMetric”.

  • few_shot_examples – A list of few-shot examples to provide to the model. If None, uses the default few-shot examples.

  • track – Whether to track the metric. Defaults to True.

Example

>>> from opik.evaluation.metrics import ContextRecall
>>> context_recall_metric = ContextRecall()
>>> result = context_recall_metric.score("What's the capital of France?", "The capital of France is Paris.", "Paris", ["France is a country in Europe."])
>>> print(result.value)
0.9
>>> print(result.reason)
The LLM's response is highly accurate, correctly identifying 'Paris' as the capital of France and aligning with the expected answer ...
score(input: str, output: str, expected_output: str, context: List[str], **ignored_kwargs: Any) → ScoreResult

Calculate the context recall score for the given input-output pair.

Parameters:

  • input – The input text to be evaluated.

  • output – The output text to be evaluated.

  • expected_output – The expected output for the given input.

  • context – A list of context strings relevant to the input.

  • **ignored_kwargs – Additional keyword arguments that are ignored.

Returns:

A ScoreResult object containing the context recall score (between 0.0 and 1.0) and a reason for the score.

Return type:

score_result.ScoreResult

async ascore(input: str, output: str, expected_output: str, context: List[str], **ignored_kwargs: Any) → ScoreResult

Asynchronously calculate the context recall score for the given input-output pair.

This method is the asynchronous version of score(). For detailed documentation, please refer to the score() method.

Parameters:

  • input – The input text to be evaluated.

  • output – The output text to be evaluated.

  • expected_output – The expected output for the given input.

  • context – A list of context strings relevant to the input.

  • **ignored_kwargs – Additional keyword arguments that are ignored.

Returns:

A ScoreResult object with the context recall score and reason.

Return type:

score_result.ScoreResult