dspy.evaluate.answer_passage_match

dspy.evaluate.answer_passage_match(example, pred, trace=None)

Return True if any passage in pred.context contains the answer(s).

Strings are normalized (and passages also use DPR normalization internally).

Parameters:

Name	Type	Description	Default
example		dspy.Example object with field answer (str or list[str]).	required
pred		dspy.Prediction object with field context (list[str]) containing passages.	required
trace		Unused; reserved for compatibility.	None

Returns:

Name	Type	Description
bool		True if any passage contains any reference answer; otherwise False.

Example

import dspy

example = dspy.Example(answer="Eiffel Tower")
pred = dspy.Prediction(context=["The Eiffel Tower is in Paris.", "..."])

answer_passage_match(example, pred)  # True

Source code in dspy/evaluate/metrics.py

def answer_passage_match(example, pred, trace=None):
    """Return True if any passage in `pred.context` contains the answer(s).

    Strings are normalized (and passages also use DPR normalization internally).

    Args:
        example: `dspy.Example` object with field `answer` (str or list[str]).
        pred: `dspy.Prediction` object with field `context` (list[str]) containing passages.
        trace: Unused; reserved for compatibility.

    Returns:
        bool: True if any passage contains any reference answer; otherwise False.

    Example:
        ```python
        import dspy

        example = dspy.Example(answer="Eiffel Tower")
        pred = dspy.Prediction(context=["The Eiffel Tower is in Paris.", "..."])

        answer_passage_match(example, pred)  # True
        ```
    """
    if isinstance(example.answer, str):
        return _passage_match(pred.context, [example.answer])
    elif isinstance(example.answer, list):
        return _passage_match(pred.context, example.answer)

    raise ValueError(f"Invalid answer type: {type(example.answer)}")

:::