unit8co / darts

A python library for user-friendly forecasting and anomaly detection on time series.
https://unit8co.github.io/darts/
Apache License 2.0
8.11k stars 884 forks source link

Add Cyclic Boosting model #1764

Open rijkvandermeulen opened 1 year ago

rijkvandermeulen commented 1 year ago

I recently read this paper (https://arxiv.org/abs/2002.03425) where they introduced a new model (called Cyclic Boosting). It's a type of generalized additive model using a cyclic coordinate descent optimization and a boosting-like update of parameters. They claim quite nice results in terms of accuracy, performance, and explainability. And altough it can be used for other use cases as well, it was primarly designed for time series forecasting (specifically in a retail context). Do you think it would be a good idea to add this model to Darts?

Some additional material:

lbventura commented 1 year ago

I picked up this issue on the Cyclic Boosting side, will be creating a PR on it soon 🚀