What Constitutes Good Contrastive Learning in Time-Series Forecasting?

This is the origin Pytorch implementation of Informer in the following paper: What Constitutes Good Contrastive Learning in Time-Series Forecasting?. We developed our code based on the repositories of Informer and CoST.

This repo implements the experiments of end-to-end training For end-to-end two-step training experiments, please refer to this repo

Citation

If you find this repository useful in your research, please consider citing the following paper:

@article{DBLP:journals/corr/abs-2306-12086,
  author       = {Chiyu Zhang and
                  Qi Yan and
                  Lili Meng and
                  Tristan Sylvain},
  title        = {What Constitutes Good Contrastive Learning in Time-Series Forecasting?},
  journal      = {CoRR},
  volume       = {abs/2306.12086},
  year         = {2023},
  url          = {https://doi.org/10.48550/arXiv.2306.12086},
  doi          = {10.48550/arXiv.2306.12086},
  eprinttype    = {arXiv},
  eprint       = {2306.12086},
  timestamp    = {Fri, 23 Jun 2023 15:19:11 +0200},
  biburl       = {https://dblp.org/rec/journals/corr/abs-2306-12086.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

Contact

If you have any questions, feel free to contact Chiyu Zhang through Email (zcy94@outlook.com) or Github issues.

Acknowledgments

We acknowledge the authors of the repositories of Informer and CoST.