Function strip_html

preference_datasets.py:60–64

(x)

        range(int(len(dataset) * 0.01), len(dataset)))

    def strip_html(x):
        x['question'] = strip_html_tags(x['question'])
        for a in x['answers']:
            a['text'] = strip_html_tags(a['text'])
        return x

    dataset = dataset.map(strip_html, num_proc=64)