📄 Document Summarizer App

Built with Streamlit, LangChain, Hugging Face Transformers, and PyTorch

---

📌 Project Overview

This is an AI-powered PDF Document Summarizer App that extracts text from PDF files and generates concise summaries using the LaMini-Flan-T5 model. The app is built with Streamlit for an interactive web interface and integrates LangChain for smart text chunking and Transformers for natural language processing.

---

🚀 Key Features

• 📄 Upload and preview PDF documents directly in-browser
• ✂️ Intelligent document chunking using LangChain
• 🤖 Summarization using LaMini-Flan-T5 from Hugging Face
• ⚡ Efficient text preprocessing to avoid token overflow
• 🧠 Built-in PDF viewer for side-by-side comparison
• 🌐 Streamlit-powered UI for fast deployment

---

🧑‍💻 Tech Stack

• Frontend: Streamlit
• NLP: Hugging Face Transformers (pipeline API), LaMini-Flan-T5
• Text Preprocessing: LangChain (RecursiveCharacterTextSplitter)
• PDF Parsing: PyPDFLoader (LangChain Community)
• Frameworks: PyTorch
• Other Tools: Base64 encoding for PDF rendering

---

📂 How It Works

1. 📤 Upload a .pdf file via the Streamlit interface
2. 📚 Text is extracted and chunked using RecursiveCharacterTextSplitter
3. 🤖 The summarization pipeline runs with T5ForConditionalGeneration
4. 📄 The original PDF and the generated summary are displayed side by side

---

🛠️ Installation & Setup

# Clone the repository
git clone https://github.com/yourusername/document-summarizer-app.git
cd document-summarizer-app

# Create a virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate  # or venv\Scripts\activate on Windows

# Install required packages
pip install -r requirements.txt