def get_raw_scores()

in metrics/squad_v2/compute_score.py [0:0]

• 17 lines of code
• 10 McCabe index (conditional complexity)

def get_raw_scores(dataset, preds):
    exact_scores = {}
    f1_scores = {}
    for article in dataset:
        for p in article["paragraphs"]:
            for qa in p["qas"]:
                qid = qa["id"]
                gold_answers = [t for t in qa["answers"]["text"] if normalize_answer(t)]
                if not gold_answers:
                    # For unanswerable questions, only correct answer is empty string
                    gold_answers = [""]
                if qid not in preds:
                    print(f"Missing prediction for {qid}")
                    continue
                a_pred = preds[qid]
                # Take max over all gold answers
                exact_scores[qid] = max(compute_exact(a, a_pred) for a in gold_answers)
                f1_scores[qid] = max(compute_f1(a, a_pred) for a in gold_answers)
    return exact_scores, f1_scores