Darts: User-Friendly Modern Machine Learning for Time Series

By J. Herzen, Francesco Lässig, Samuele Giuliano Piazzetta, T. Neuer, L'eo Tafti, Guillaume Raille, Tomas Van Pottelbergh, Marek Pasieka, Andrzej Skrodzki, Nicolas Huguenin and 9 more. Journal of machine learning research, 2021

315 citations recorded by the source corpus.

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.

Subfields: Management Science and Operations Research, Signal Processing

View Paper: https://www.semanticscholar.org/paper/22733aac53e89446aed76dd1983bf2d74567ba88

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

1. Nate Gruver
2. Andrew Gordon Wilson
3. Rajat Sen
4. Weihao Kong
5. Lintao Ma
6. Shiyu Wang
7. Shikai Qiu
8. Yichen Zhou

In Signal Processing

1. Nate Gruver
2. Andrew Gordon Wilson
3. Rajat Sen
4. Weihao Kong
5. Xiaoming Shi
6. Lintao Ma
7. Shiyu Wang
8. Shikai Qiu

Citations in the corpus

• Large Language Models Are Zero-Shot Time Series Forecasters (2023, methodology): “As we are using Figure 4: LLMT IME with base model GPT-3 or LLaMA-2 70B has the best or second best aggregated performance on several deterministic time series benchmarks [23, 18, 54] while being entirely zero-shot.”
• Time-LLM: Time Series Forecasting by Reprogramming Large Language Models (2023, methodology): “We attribute this success to our reprogramming framework being better at activating the LLM’s knowledge transfer and reasoning capabilities in a resource-efficient manner when performing time series tasks.”
• A decoder-only foundation model for time-series forecasting (2023, methodology): “To benchmark our model’s performance, we choose three groups of commonly used forecasting datasets that cover various domains, sizes, granularities, and horizon lengths: Darts (Herzen et al., 2022), Monash (Godahewa et al., 2021) and Informer datasets (Zhou et al., 2021), to test the generalization ”