app.py

import streamlit as st
from extractor import extract, FewDocumentsError
from summarizer import summarize
from translation import translate
import time
import cProfile
from sentence_transformers import SentenceTransformer
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import torch

@st.cache(allow_output_mutation=True)
def init():
    # Dowload required NLTK resources
    from nltk import download
    download('punkt')
    download('stopwords')

    device = "cuda" if torch.cuda.is_available() else "cpu"
    # Model for semantic searches
    search_model = SentenceTransformer('msmarco-distilbert-base-v4', device=device)
    # Model for abstraction
    summ_model = AutoModelForSeq2SeqLM.from_pretrained('t5-base')
    tokenizer = AutoTokenizer.from_pretrained('t5-base')

    return search_model, summ_model, tokenizer

def main():
    search_model, summ_model, tokenizer = init()

    st.title("AutoSumm")
    st.subheader("Lucas Antunes & Matheus Vieira")

    portuguese = st.checkbox('Traduzir para o português.')

    if portuguese:
        st.subheader("Digite o tópico sobre o qual você deseja gerar um resumo")
        query_pt = st.text_input('Digite o tópico') #text is stored in this variable
        button = st.button('Gerar resumo')
    else:
        st.subheader("Type the desired topic to generate the summary")
        query = st.text_input('Type your topic') #text is stored in this variable
        button = st.button('Generate summary')

    if 'few_documents' not in st.session_state:
        st.session_state['few_documents'] = False
        few_documents = False
    else:
        few_documents = st.session_state['few_documents']

    if button:
        start_time = time.time()
        query = translate(query_pt, 'pt', 'en') if portuguese else query
        try:
            with st.spinner('Extraindo textos relevantes...'):
                text = extract(query, search_model=search_model)
        except FewDocumentsError as e:
            few_documents = True
            st.session_state['few_documents'] = True
            st.session_state['documents'] = e.documents
            st.session_state['msg'] = e.msg
        else:

            st.info(f'(Extraction) Elapsed time: {time.time() - start_time:.2f}s')
            with st.spinner('Gerando resumo...'):
                summary = summarize(text, summ_model, tokenizer)
            st.info(f'(Total) Elapsed time: {time.time() - start_time:.2f}s')

            if portuguese:
                st.markdown(f'Seu resumo para "{query_pt}":\n\n> {translate(summary, "en", "pt")}')
            else:
                st.markdown(f'Your summary for "{query}":\n\n> {summary}')


    if few_documents:
        st.warning(st.session_state['msg'])
        if st.button('Prosseguir'):
            start_time = time.time()
            with st.spinner('Extraindo textos relevantes...'):
                text = extract(query, search_model=search_model, extracted_documents=st.session_state['documents'])
            st.info(f'(Extraction) Elapsed time: {time.time() - start_time:.2f}s')
            with st.spinner('Gerando resumo...'):
                summary = summarize(text, summ_model, tokenizer)
            st.info(f'(Total) Elapsed time: {time.time() - start_time:.2f}s')

            if portuguese:
                st.markdown(f'Seu resumo para "{query_pt}":\n\n> {translate(summary, "en", "pt")}')
            else:
                st.markdown(f'Your summary for "{query}":\n\n> {summary}')

            st.session_state['few_documents'] = False
            few_documents = False
            
if __name__ == '__main__':
    cProfile.run('main()', 'stats.txt')