import streamlit as st
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

# Load the tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("suriya7/bart-finetuned-text-summarization")
model = AutoModelForSeq2SeqLM.from_pretrained("suriya7/bart-finetuned-text-summarization")

def generate_summary(text):
    inputs = tokenizer([text], max_length=1024, return_tensors='pt', truncation=True)
    summary_ids = model.generate(inputs['input_ids'], max_new_tokens=100, do_sample=False)
    summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
    
    # Post-process the summary to include only specific points
    important_points = extract_important_points(summary)
    
    return important_points

def extract_important_points(summary):
    # This is a very basic example of how you might filter for key phrases
    # You can refine this with more sophisticated NLP techniques
    keywords = ["change", "request", "important", "needs", "must", "critical", "required", "suggested"]
    filtered_lines = [line for line in summary.split('. ') if any(keyword in line.lower() for keyword in keywords)]
    return '. '.join(filtered_lines)

# Streamlit interface
st.title("Text Summarization App")

# User text input
user_input = st.text_area("Enter the text you want to summarize", height=200)

if st.button("Generate Summary"):
    if user_input:
        with st.spinner("Generating summary..."):
            summary = generate_summary(user_input)
        st.subheader("Filtered Summary:")
        st.write(summary)
    else:
        st.warning("Please enter text to summarize.")

# Instructions for using the app
st.write("Enter your text in the box above and click 'Generate Summary' to get a summarized version of your text.")