Decomposed Transformer with Frequency Attention for Multivariate Time Series Anomaly Detection

Unofficial Python implementation of the decomposed transformer algorithm:
Qin, Shuxin, et al. Decomposed transformer with frequency attention for multivariate time series anomaly detection. 2022 IEEE International Conference on Big Data (Big Data). IEEE, 2022.

P.S. This implementation is solely for my personal exploration, and the results obtained should not be considered indicative of the outcomes reported in the original paper！

Official implementation：DecomposedTransAD.

Get Started

1. Install Python 3.6, PyTorch >= 1.4.0.

2. Download data. You can obtain five benchmarks from Google Cloud. All the datasets are well pre-processed.

bash ./scripts/SMD.sh
bash ./scripts/MSL.sh
bash ./scripts/SMAP.sh
bash ./scripts/PSM.sh
bash ./scripts/SWAT.sh

Citation

@inproceedings{qin2022decomposed,
  title={Decomposed transformer with frequency attention for multivariate time series anomaly detection},
  author={Qin, Shuxin and Zhu, Jing and Wang, Dan and Ou, Liang and Gui, Hongxin and Tao, Gaofeng},
  booktitle={2022 IEEE International Conference on Big Data (Big Data)},
  pages={1090--1098},
  year={2022},
  organization={IEEE}
}