def read_squad()

in distilbertqatrain.py [0:0]

• 20 lines of code
• 11 McCabe index (conditional complexity)

def read_squad(path):
    # open JSON file and load intro dictionary
    with open(path, 'rb') as f:
        squad_dict = json.load(f)

    # initialize lists for contexts, questions, and answers
    contexts = []
    questions = []
    answers = []
    # iterate through all data in squad data
    for group in squad_dict['data']:
        for passage in group['paragraphs']:
            context = passage['context']
            for qa in passage['qas']:
                question = qa['question']
                # check if we need to be extracting from 'answers' or 'plausible_answers'
                if 'plausible_answers' in qa.keys():
                    access = 'plausible_answers'
                else:
                    access = 'answers'
                for answer in qa[access]:
                    # append data to lists
                    contexts.append(context)
                    questions.append(question)
                    answers.append(answer)
    # return formatted data lists
    return contexts, questions, answers