Multitask Chatbot Project

Overview

This AI-powered chatbot is built with Streamlit and integrates with Groq's API. It provides a user-friendly interface for text-based interactions, document processing and database connectivity.

Table of Contents

• Overview
• Features
  • Core Functionalities
  • Document Management
• Run Locally
• Environment Variables
• Development Tools and Technologies
• Contributing

🚀 Features

Core Functionalities

• 💬 Text-based Interaction: Understands and responds to queries effectively
• 🧠 Context Awareness: Maintains conversation context for relevant responses
• 🎯 Model Selection: Automatically selects the appropriate model based on the query complexity
• 🤖 Agent-based Architecture: Utilizes LangChain's React agent for advanced reasoning and tool use

Document Management

• 📄 Document Processing: Support for various document formats including:
  • PDF files (using PyMuPDF)
  • PowerPoint presentations (python-pptx)
  • Excel spreadsheets (openpyxl)
• 📝 Text Extraction: OCR capabilities using Tesseract
• 🔍 Vector Search: Document embeddings using sentence-transformers and FAISS

🛠️ Run Locally

1. Clone the project

   git clone https://github.com/AugustinMORVAL/Multitask_Chatbot.git

2. Install dependencies

   pip install -r requirements.txt

3. Run the Chatbot

   streamlit run chatbot.py

🔑 Environment Variables

The application supports two methods for configuration:

Option 1: Using a .env file

Create a .env file in the root directory with your API keys:

GROQ_API_KEY=your-api-key-here

Option 2: Using Streamlit Secrets

For Streamlit Cloud deployment, create .streamlit/secrets.toml:

groq_api_key = "your-api-key-here"

Note: Both .env and .streamlit/secrets.toml are included in .gitignore for security.

🧰 Development Tools and Technologies

Core Dependencies

• 🎯 Streamlit: Web interface and application framework
• 🤖 LangChain: LLM integration and chain management
• 🔍 Vector Search: FAISS for efficient similarity search
• 📄 Document Processing:
  • PyMuPDF for PDF handling
  • python-pptx for PowerPoint files
  • openpyxl for Excel files
  • pytesseract for OCR
• 🔤 Embeddings: sentence-transformers for text embeddings
• 🌐 Web Search: DuckDuckGo integration
• 📊 Database: SQLAlchemy and MongoDB support

🤝 Contributing

Contributions are welcome! Please feel free to submit pull requests or open issues for bugs and feature requests.

⚠️ Disclaimer

This is a demo version with limited functionality. Not all features listed above are currently implemented.