app.py

import streamlit as st
from transformers import pipeline

# Load models and tokenizers using Hugging Face's pipeline
# @st.cache_data()
def load_pipelines():
    bart_pipeline = pipeline("summarization", model="facebook/bart-large-cnn")
    t5_pipeline = pipeline("summarization", model="t5-large")
    pegasus_pipeline = pipeline("summarization", model="google/pegasus-cnn_dailymail")
    
    return {
        'BART': bart_pipeline,
        'T5': t5_pipeline,
        'Pegasus': pegasus_pipeline,
    }

prompt = """
Summarize the below paragraph
"""

# Streamlit app layout
st.title("Text Summarization with Pre-trained Models (BART, T5, Pegasus)")

text_input = st.text_area("Enter text to summarize:")

if st.button("Generate Summary"):
    if text_input:
        pipelines = load_pipelines()
        summaries = {}
        for model_name, pipeline in pipelines.items():
            summary = pipeline(f"{prompt}\n{text_input}", max_length=150, min_length=50, length_penalty=2.0, num_beams=4, early_stopping=True)[0]['summary_text']
            summaries[model_name] = summary

        st.subheader("Summaries")
        for model_name, summary in summaries.items():
            st.write(f"**{model_name}**")
            st.write(summary.replace('<n>', ''))
            st.write("---")
    else:
        st.error("Please enter text to summarize.")