Darts: User-Friendly Modern Machine Learning for Time Series

TL;DR

Darts is a user-friendly Python library that simplifies the application of modern machine learning techniques for time series forecasting, supporting diverse models and functionalities.

Contribution

It introduces a comprehensive, easy-to-use library integrating classical and deep learning models with advanced features like probabilistic forecasting and meta-learning.

Findings

Supports multidimensional and large datasets

Enables ensemble and external data integration

Offers a unified API similar to scikit-learn

Abstract

We present Darts, a Python machine learning library for time series, with a focus on forecasting. Darts offers a variety of models, from classics such as ARIMA to state-of-the-art deep neural networks. The emphasis of the library is on offering modern machine learning functionalities, such as supporting multidimensional series, meta-learning on multiple series, training on large datasets, incorporating external data, ensembling models, and providing a rich support for probabilistic forecasting. At the same time, great care goes into the API design to make it user-friendly and easy to use. For instance, all models can be used using fit()/predict(), similar to scikit-learn.