Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting

Lag-Llama is the first open-source foundation model for time series forecasting!

[Tweet Thread] [Model Weights] [Colab Demo on Zero-Shot Forecasting] [GitHub] [Paper]

---

This repository houses the Lag-Llama architecture.

---

• Coming Next: Fine-tuning scripts with examples on real-world datasets and best practices in using Lag-Llama!🚀

Updates:

• 17-Feb-2024: We have released a new updated Colab Demo for zero-shot forecasting that shows how one can load time series of different formats.
• 7-Feb-2024: We released Lag-Llama, with open-source model checkpoints and a Colab Demo for zero-shot forecasting.

---

Current Features:

💫 Zero-shot forecasting on a dataset of any frequency for any prediction length, using the Colab Demo.

---

Coming Soon:

⭐ An online gradio demo where you can upload time series and get zero-shot predictions and perform finetuning.

⭐ Features for finetuning the foundation model

⭐ Features for pretraining Lag-Llama on your own large-scale data

⭐ Scripts to reproduce all results in the paper.

---

Stay Tuned!🦙

---

Citing this work

Please use the following Bibtex entry to cite Lag-Llama.

@misc{rasul2024lagllama,
      title={Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting}, 
      author={Kashif Rasul and Arjun Ashok and Andrew Robert Williams and Hena Ghonia and Rishika Bhagwatkar and Arian Khorasani and Mohammad Javad Darvishi Bayazi and George Adamopoulos and Roland Riachi and Nadhir Hassen and Marin Biloš and Sahil Garg and Anderson Schneider and Nicolas Chapados and Alexandre Drouin and Valentina Zantedeschi and Yuriy Nevmyvaka and Irina Rish},
      year={2024},
      eprint={2310.08278},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}