Skip to content

The implementation of "TimeCHEAT: A Channel Harmony Strategy for Irregularly Sampled Multivariate Time Series Analysis"

License

Notifications You must be signed in to change notification settings

Alrash/TimeCHEAT

Repository files navigation

TimeCHEAT

The implementation of "TimeCHEAT: A Channel Harmony Strategy for Irregularly Sampled Multivariate Time Series Analysis" in AAAI 2025.

framework

Paper: Arxiv

Appendix: Appendix

Sample

# classification
python -u main.py -da phy12  -rf 96 -lr 1e-3 -ns 0 -e 60 -bs 64 -pa 8 -l 2 -s 42 -ns 1

# forecasting
python -u main.py -da mimiciii -rf 96 -lr 1e-3 -ns 0 -e 60 -bs 64 -pa 8 -l 2 -ds forecast -s 42 -fi 0

# imputation
python -u main.py -da phy -rf 96 -lr 1e-3 -ns 0 -e 60 -bs 64 -pa 8 -l 2 -ds impute -s 42 -stp 0.5

Citation

@inproceedings{liu2025timecheat,
  title={TimeCHEAT: A Channel Harmony Strategy for Irregularly Sampled Multivariate Time Series Analysis},
  author={Liu, Jiexi and Cao, Meng and Chen, Songcan},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={39},
  number={18},
  pages={18861--18869},
  year={2025}
}

Acknowledgement

GrapFITi

Time-Series-Library

PatchTST

Email

About

The implementation of "TimeCHEAT: A Channel Harmony Strategy for Irregularly Sampled Multivariate Time Series Analysis"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages