Skip to main content

agrag.eval.answer.context_precision

context_precision(judge:DeepEvalBaseLLM, *, threshold:float = 0.5) -> BaseMetric

Build the metric for useful evidence ranked before noise.

Parameters:

  • judge (DeepEvalBaseLLM) – The judge model.
  • threshold (float) – The minimum score that counts as success.

Returns:

  • BaseMetric – The context precision metric.