import streamlit as st from streamlit.logger import get_logger from transformers import pipeline import datasets import pandas as pd from huggingface_hub import login LOGGER = get_logger(__name__) model = "sivan22/halacha-siman-seif-classifier-new" login('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') ds = datasets.load_dataset('sivan22/orach-chaim',token=True) df = ds['train'].to_pandas() def clean(s)->str: return s.replace(" ","") df['seif']= df['seif'].apply(clean) def get_predicts_local(input)->str: classifier = pipeline("text-classification",model=model,top_k=None) predicts = classifier(input) return predicts def get_predicts_online(input)->str: import requests API_URL = "https://api-inference.huggingface.co/models/" + model headers = {"Authorization": f"Bearer {'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'}"} def query(input_text): response = requests.post(API_URL, headers=headers, json='{{inputs:' +input_text+'}{wait_for_model:true}}') return response.json() predicts = query(input) return predicts def run(): st.set_page_config( page_title="Halacha classification", page_icon="", ) st.write("# חיפוש בשולחן ערוך") use_local = st.checkbox("חיפוש לא מקוון") user_input = st.text_input('כתוב כאן את שאלתך', placeholder='כמה נרות מדליקים בחנוכה') if st.button('חפש') and user_input!="": get_predicts = get_predicts_local if use_local else get_predicts_online #print(get_predicts(user_input)[0][0:5]) for prediction in get_predicts(user_input)[0][:5]: rows = df[((df["bookname"] == " שלחן ערוך - אורח חיים ") | (df["bookname"] ==" משנה ברורה")) & (df["siman"] == prediction['label'].split(' ')[0])& (df["seif"] == prediction['label'].split(' ')[1]) ] rows.sort_values(["bookname"],ascending=False, inplace=True) print(prediction['label'].split(' ')) st.write('סימן ' + str(prediction['label']), rows[['text','sek','seif','siman','bookname']]) if __name__ == "__main__": run()