USAD - UnSupervised Anomaly Detection on multivariate time series

August 28, 2020 ยท View on GitHub

Scripts and utility programs for implementing the USAD architecture.

Implementation by: Francesco Galati.

Additional contributions: Julien Audibert, Maria A. Zuluaga.

How to cite

If you use this software, please cite the following paper as appropriate:

Audibert, J., Michiardi, P., Guyard, F., Marti, S., Zuluaga, M. A. (2020).
USAD : UnSupervised Anomaly Detection on multivariate time series.
Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, August 23-27, 2020

Requirements

  • PyTorch 1.6.0
  • CUDA 10.1 (to allow use of GPU, not compulsory)

Running the Software

All the python classes and functions strictly needed to implement the USAD architecture can be found in usad.py. An example of an application deployed with the SWaT dataset is included in USAD.ipynb.

Copyright 2020 Eurecom.

This software is released under the BSD-3 license. Please see the license file_ for details.

Publication

Audibert et al. USAD : UnSupervised Anomaly Detection on multivariate time series. 2020