sihohan / DAMP

Unofficial Python implementation for "Matrix Profile XXIV: Scaling Time Series Anomaly Detection to Trillions of Datapoints and Ultra-fast Arriving Data Streams" (KDD '22)

Discord Aware Matrix Profile (DAMP)

Authors:

• Siho Han (@sihohan)
• Jihwan Min (@rtm-jihwan-min)
• Taeyeong Heo (@htyvv)
• JuI Ma (@iju298)

This repository contains an unofficial Python implementation of Discord Aware Matrix Profile (DAMP), introduced in "Matrix Profile XXIV: Scaling Time Series Anomaly Detection to Trillions of Datapoints and Ultra-fast Arriving Data Streams" (KDD '22). The official MATLAB implementation can be found here.

Project Organization

├── data
|   └── samples
|       └── BourkeStreetMall.txt
├── .gitignore
├── README.md
├── damp.py
└── utils.py

Requirements

• Python >= 3.6
• matplotlib
• numpy

Datasets

This repository includes Bourke Street Mall as the default dataset (see the data directory), which can be downloaded here.

Run

You can run the code using the following command.

python damp.py

With --enable_output, the resulting plot and DAMP values will be saved in the ./figures and ./outputs directories, respectively.

Note that the input time series and its corresponding DAMP scores on the plot are scaled for visualization purposes.

References

• Lu, Yue, et al. "Matrix Profile XXIV: Scaling Time Series Anomaly Detection to Trillions of Datapoints and Ultra-fast Arriving Data Streams." Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2022.