直接正则好了，简单粗暴：

pattern = re.compile(r'<td>(\d+) / (\d+)</td>')
m = pattern.findall(html_doc)
for item in m:
    print(item[0])
    print(item[1])