Libra is an end-to-end, full-stack web application that enables users to:

• Visualize real-time and historical stock prices
• Predict future stock trends using deep learning
• Analyze sentiment from news and tweets
• Chat with a built-in AI stock assistant
• Manage portfolios, alerts, and more

• Secure authentication with JWT & Google OAuth
• PostgreSQL for user data
• Redis for session handling and caching
• Personalized dashboard with saved preferences

• Interactive charts (candlesticks, volume, moving averages)
• WebSocket-powered real-time updates
• Multi-stock comparisons
• Filters for 1D, 1W, 1M, YTD, etc.

• LSTM and Transformer models for time-series prediction
• Show predicted trends with confidence intervals
• Display evaluation metrics: MSE, MAE, etc.
• Optional retraining via user data uploads (admin)

• Scrape real-time financial news and tweets
• Use FinBERT or RoBERTa to detect sentiment
• Overlay sentiment analysis with stock charts

• ⭐️ Custom Watchlist
• 📈 Portfolio Tracker (simulated trading)
• 📧 Weekly Performance Emails
• 🔔 Price Alerts via Email or UI

• Ask about stock trends, news summaries, or predictions
• Get help navigating the site
• Learn how predictions were made (basic XAI)
• Powered by OpenAI or fine-tuned model

• Visual feature attributions (via SHAP, Grad-CAM)
• Transparent insight into prediction logic

/libra
├── frontend/ # Next.js frontend
│ ├── pages/
│ ├── components/
│ └── charts/
├── backend/ # FastAPI/Django backend
│ ├── routes/
│ ├── auth/
│ ├── services/
│ └── chatbot/
├── ml_model/ # Model training + serving
│ ├── train_model.ipynb
│ ├── model.pt
│ └── predict.py
├── chatbot/ # Chatbot logic
│ └── chat_handler.py
├── scripts/ # Data fetchers and cron jobs
│ └── data_scraper.py
├── docker-compose.yml # Dev orchestration
└── README.md

• Node.js & npm
• Python 3.8+
• Docker & Docker Compose
• PostgreSQL & Redis running locally

# Clone the repo
git clone https://github.com/yourusername/libra.git
cd libra

# Start Docker services
docker-compose up --build

# Frontend setup
cd frontend
npm install
npm run dev

# Backend setup
cd backend
pip install -r requirements.txt
uvicorn main:app --reload

• Frontend: Vercel
• Backend: Render, Fly.io, or AWS EC2
• Model API: TorchServe or self-hosted FastAPI route

We welcome PRs! To contribute:

1. Fork the repository
2. Create a new branch: git checkout -b feature-name
3. Commit your changes
4. Open a pull request

Licensed under the MIT License. See LICENSE for details.