app.py

import streamlit as st
from peft import PeftModel
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
import torch


@st.cache_resource
def load_model():
    """Load the PEFT model and tokenizer once and cache them"""
    base_model = AutoModelForSeq2SeqLM.from_pretrained("t5-small")
    peft_model = PeftModel.from_pretrained(base_model, "Lakshan2003/finetuned-t5-xsum")
    tokenizer = AutoTokenizer.from_pretrained("Lakshan2003/finetuned-t5-xsum")
    return peft_model, tokenizer


def generate_summary(text, model, tokenizer, max_length=128, min_length=30):
    """Generate summary using the PEFT model"""
    # Move model to GPU if available
    device = "cuda" if torch.cuda.is_available() else "cpu"
    model = model.to(device)

    # Prepare the input text
    prefix = "summarize: "
    input_text = prefix + text

    # Tokenize
    inputs = tokenizer(input_text, return_tensors="pt", max_length=512, truncation=True)
    inputs = {k: v.to(device) for k, v in inputs.items()}

    # Generate summary
    with torch.no_grad():
        output_ids = model.generate(
            input_ids=inputs["input_ids"],
            attention_mask=inputs["attention_mask"],
            max_length=max_length,
            min_length=min_length,
            num_beams=4,
            length_penalty=2.0,
            early_stopping=True,
            no_repeat_ngram_size=3
        )

    # Decode the summary
    summary = tokenizer.decode(output_ids[0], skip_special_tokens=True)
    return summary


def main():
    st.set_page_config(
        page_title="SummarizeAI Pro",
        page_icon="✨",
        layout="wide"
    )

    # Custom CSS
    st.markdown("""
        <style>
        .main-title {
            text-align: center;
            color: #1E88E5;
            font-size: 3rem !important;
            font-weight: 700;
            margin-bottom: 1rem;
        }
        .subtitle {
            text-align: center;
            color: #424242;
            font-size: 1.2rem !important;
            margin-bottom: 2rem;
        }
        </style>
    """, unsafe_allow_html=True)

    # App title and subtitle
    st.markdown("<h1 class='main-title'>✨ SummarizeAI Pro</h1>", unsafe_allow_html=True)
    st.markdown("<p class='subtitle'>Transform lengthy text into concise, meaningful summaries with AI</p>",
                unsafe_allow_html=True)

    # Load model and tokenizer
    with st.spinner("Loading model... (this may take a few moments)"):
        model, tokenizer = load_model()

    # Input text area
    text = st.text_area(
        "📝 Enter your text below:",
        height=200,
        placeholder="Paste your text here and let SummarizeAI Pro work its magic..."
    )

    # Create three columns for better layout
    col1, col2, col3 = st.columns([1, 1, 1])

    with col1:
        max_length = st.slider("Maximum summary length", 50, 250, 128)
    with col2:
        min_length = st.slider("Minimum summary length", 10, 100, 30)
    with col3:
        st.markdown("<br>", unsafe_allow_html=True)  # Spacing
        generate_button = st.button("✨ Generate Summary", use_container_width=True)

    if generate_button:
        if text:
            with st.spinner("✨ AI is crafting your summary..."):
                try:
                    summary = generate_summary(text, model, tokenizer,
                                               max_length=max_length,
                                               min_length=min_length)

                    st.markdown("### 📊 Summary Results")

                    # Create columns for statistics
                    stat_col1, stat_col2 = st.columns(2)
                    with stat_col1:
                        st.info(f"📄 Original text: {len(text.split())} words")
                    with stat_col2:
                        st.info(f"✂️ Summarized text: {len(summary.split())} words")

                    # Display summary in a nice box
                    st.markdown("### ✨ Generated Summary")
                    st.markdown(f"""
                        <div style="
                            padding: 20px;
                            border-radius: 10px;
                            background-color: #f0f2f6;
                            border-left: 5px solid #1E88E5;
                        ">
                            {summary}
                        </div>
                    """, unsafe_allow_html=True)

                except Exception as e:
                    st.error(f"🚫 An error occurred: {str(e)}")
        else:
            st.warning("⚠️ Please enter some text to summarize.")

    # Sidebar with enhanced styling
    st.sidebar.markdown("## 🎯 About SummarizeAI Pro")
    st.sidebar.markdown("""
        SummarizeAI Pro uses advanced AI technology powered by a PEFT-tuned T5 model 
        to generate accurate and concise summaries while preserving the key points 
        of your text.
    """)

    st.sidebar.markdown("## 📖 How to Use")
    st.sidebar.markdown("""
        1. 📝 Paste your text in the input box
        2. 🎚️ Adjust summary length with sliders
        3. 🚀 Click 'Generate Summary'
        4. ✨ Get your AI-powered summary
    """)

    # Footer
    st.markdown("""
        <div style='text-align: center; color: #666; padding: 20px;'>
            <p>Made with ❤️ by Lakshan Cooray</p>
        </div>
    """, unsafe_allow_html=True)


if __name__ == "__main__":
    main()