primeqa.mrc.metrics.nq_f1.nq_eval.compute_optimal_metrics

primeqa.mrc.metrics.nq_f1.nq_eval.compute_optimal_metrics(long_answer_stats: List[List[Union[bool, float]]], short_answer_stats: List[List[Union[bool, float]]]) → Dict[str, float]

Computes overall metrics for long and short answers for their respective optimal thresholds :param long_answer_stats: List of long answer scores. :param short_answer_stats: List of short answer scores.

Returns

Ordered Dictionary of name (string) -> score.