pages/3_🐧_分词.py

from meutils.pipe import *

from appzoo.streamlit_app import Page

import streamlit as st

from LAC import LAC as _LAC

LAC = st.experimental_singleton(_LAC)


# @st.cache(func=None, persist=False, hash_funcs={'LAC.lac.LAC': str})
# def tokenizer():
#     print('Loading tokenizer1...')
#     return LAC()

@ttl_cache(key=str)
@disk_cache()
def tokenizer(texts):
    return LAC().run(texts)  | xmap_(lambda r: list(zip(*r)))


class MyPage(Page):

    def main(self):
        with st.form("Coding"):

            texts = st.text_area("输入文本", ["为什么东北证券那么牛？"]*3 | xjoin('\n')).split("\n")

            if st.form_submit_button('开始转换'):
                _ = tokenizer(texts)

                # st.json(_)
                # st.text(_)
                # st.dataframe(pd.DataFrame(_))


                from annotated_text import annotated_text

                for text in _:
                    annotated_text(*text)


if __name__ == '__main__':
    app_title = "# 切词"
    app_info = ""
    MyPage(app_title=app_title, app_info=app_info).main()