import json def load_squad_json(path): with open(path, "r", encoding="utf-8") as f: data = json.load(f) samples = [] for article in data["data"]: for para in article["paragraphs"]: context = para["context"] for qa in para["qas"]: if not qa["answers"]: continue ans = qa["answers"][0] samples.append({ "context": context, "question": qa["question"], "answer_text": ans["text"] }) return samples