tabpfn-time-series 1.0.0

Zero-shot time series forecasting with TabPFNv2

TabPFN-TS

Zero-Shot Time Series Forecasting with TabPFNv2

📌 News

• 27-05-2025: 📝 New paper version and v1.0.0 release! Strong GIFT-EVAL results, new AutoSeasonalFeatures, improved CalendarFeatures.
• 27-01-2025: 🚀 Ranked 1st on GIFT-EVAL benchmark^[1]!
• 10-10-2024: 🚀 TabPFN-TS paper accepted to NeurIPS 2024 TRL and TSALM workshops!

[1] Last checked on: 10/03/2025

✨ Introduction

We demonstrate that the tabular foundation model TabPFNv2, combined with lightweight feature engineering, enables zero-shot time series forecasting for both point and probabilistic tasks. On the GIFT-EVAL benchmark, our method achieves performance on par with top-tier models across both evaluation metrics.

📖 How does it work?

Our work proposes to frame univariate time series forecasting as a tabular regression problem.

Concretely, we:

1. Transform a time series into a table
2. Extract features and add them to the table
3. Perform regression on the table using TabPFNv2
4. Use regression results as time series forecasting outputs

For more details, please refer to our paper.

👉 Why gives us a try?

• Zero-shot forecasting: this method is extremely fast and requires no training, making it highly accessible for experimenting with your own problems.
• Point and probabilistic forecasting: it provides accurate point forecasts as well as probabilistic forecasts.
• Support for exogenous variables: if you have exogenous variables, this method can seemlessly incorporate them into the forecasting model.

On top of that, thanks to tabpfn-client from Prior Labs, you won’t even need your own GPU to run fast inference with TabPFNv2. 😉 We have included tabpfn-client as the default engine in our implementation.

How to use it?

The demo should explain it all. 😉