ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting

This repository contains the official implementation for the ICLR 2026 paper: "ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting".

---

📖 Overview

ProtoTS is a novel interpretable forecasting framework that achieves both high accuracy and transparent decision-making through modeling prototypical temporal patterns across hierarchical scales.

🛠️ Installation

To set up the environment and install all dependencies, run:

pip install -r requirements.txt

🚀 Usage

To train the model or reproduce the experimental results, execute the provided training script:

bash scripts/train.sh

✍️ Citation

If you find our work or code useful in your research, please consider citing:

@article{peng2025protots,
  title={ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting},
  author={Peng, Ziheng and Ren, Shijie and Gu, Xinyue and Yang, Linxiao and Wang, Xiting and Sun, Liang},
  journal={arXiv preprint arXiv:2509.23159},
  year={2025}
}