Mohansree Vijayakumar
- Email: mohansreesk14@gmail.com
- Role: Lead Developer & ML Engineer
- Contributions: Full project development, model implementation, documentation
This project was developed as part of a machine learning internship focusing on financial time series forecasting with uncertainty quantification.
- Model Development: Implemented LSTM, MC Dropout, Bayesian Neural Networks, and Transformer models
- Uncertainty Quantification: Probabilistic forecasting with confidence intervals
- Feature Engineering: 21 technical indicators for financial analysis
- Hyperparameter Optimization: Optuna-based automated tuning
- Backtesting System: Strategy evaluation with uncertainty-aware position sizing
- Web Application: Interactive Streamlit dashboard
- Documentation: Comprehensive technical documentation and visualizations
This project utilizes several open-source libraries and frameworks:
- PyTorch for deep learning
- Pyro for probabilistic programming
- Optuna for hyperparameter optimization
- Streamlit for web application
- yfinance for financial data
For questions, suggestions, or collaboration opportunities:
- Email: mohansreesk14@gmail.com
- GitHub: Open an issue in the repository
Last Updated: 2025-10-14