import gradio as gr
import os
from transformers import pipeline
import re
import pandas as pd

def preprocess_text(text):
    text = re.sub(r'[^\u4e00-\u9fff]', '', text)    
    return text

os.environ['HF_TOKEN'] = os.environ['Century_Test']

nlp = pipeline('text-classification', model='bdsl/HanmunRoBERTa')

def predict_century(text):
    preprocessed_input = preprocess_text(text)
    result = nlp(preprocessed_input)
    result.sort(key=lambda x: x['score'], reverse=True)
    return {item['label']: item['score'] for item in result}

iface = gr.Interface(fn=predict_century, inputs="text", outputs="json")
iface.launch()