This is the official implementation of the paper "Mixture-of-Linear-Experts for Long-term Time Series Forecasting". [arXiv] [PMLR]
Please refer to the requirements.txt file for the required packages.
All datasets we used in our experiments (except Weather2K) are available at this Google Drive's shared folder. These datasets were first provided in Autoformer. Please download the datasets and put them in the dataset folder. Each dataset is an .csv file.
Weather2K dataset is available at this GitHub repository.
To run the main experiments, please run the following command:
scripts/run_all_3_seeds.shThis scripts sequentially runs the main experiments on all datasets with 3 different random seeds. The results will be saved in the logs folder.
The repository has been cleaned up for easier usage. If you want to run ablation experiments, please refer to earlier commits.
We thank the authors of the following repositories for their open-source code or dataset, which we used in our experiments:
https://github.com/zhouhaoyi/Informer2020
https://github.com/cure-lab/LTSF-Linear
https://github.com/plumprc/RTSF
https://github.com/bycnfz/weather2k
If you find our work useful, please consider citing our paper using the following BibTeX:
@inproceedings{ni2024mixture,
title={Mixture-of-Linear-Experts for Long-term Time Series Forecasting},
author={Ni, Ronghao and Lin, Zinan and Wang, Shuaiqi and Fanti, Giulia},
booktitle={International Conference on Artificial Intelligence and Statistics},
pages={4672--4680},
year={2024},
organization={PMLR}
}