Tokenizer different language fast.
python setup.py bdist_wheel
twine upload dist/*
x1 = '<a>刘强东是一个著名企业家。</a> 他创建了京东。'
t = EncoderLoader.load_tokenizer('bert-base-chinese-zh_v4-10K')
print(t.tokenize(x1, mode='char'))
print(t.tokenize(x1, mode='word'))
print(t.tokenize(x1, mode='all'))