AutoTS is an advanced open-source Python package for time series forecasting that enables rapid deployment of high-accuracy forecasting models at scale. It supports a wide variety of forecasting algorithms (naive, statistical, ML, deep learning) in a familiar sklearn-like interface, with robust support for ensemble modeling, automated feature search via genetic algorithms, and scalable handling of multiple time series. AutoTS also offers a range of time series transforms, probabilistic forecasts, integration with exogenous variables, and produces both point and interval predictions. Designed for analysts, data scientists, and anyone needing production-ready forecasting, it is especially suitable for finance, operations, and large-scale time series applications.
Visit Auto Ts's official website for product details and getting started.
Comprehensive API reference and usage examples for AutoTS.
Join the discussions and engage with other users of AutoTS on GitHub.