Skip to content

sintel-dev/aer-paper

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

aer-analysis

Replication files for AER.

L. Wong, D. Liu, L. Berti-Equille, and K. Veeramachaneni. "Time Series Anomaly Detection using Prediction-Reconstruction Mixture Errors" IEEE BigData 2022.

Usage

Experiments were made in python 3.7. To reproduce the analysis made, create a virtual environment and install required packages.

conda create --name aer-env python=3.7
conda activate aer-env
pip install -r requirements.txt

To reproduce tables and figures, use notebooks/analyze_results.ipynb

[Optional] If you want to run benchmark scripts, please download data folder.

Note: Yahoo datasets are excluded, to include which you may need to apply for the access on the official website and then add those signals to data folder.

Now you should be able to use notebooks/run_benchmark.ipynb to benchmark all pipelines and their variations. The results will be saved into the folder ./results. You can check the running progress of each pipeline in ./logs.

⚠️ It will take a very long time to run, if you run all pipelines on all datasets. You may want to customize it by selecting a subset of pipelines or datasets in analysis.py.

Resources

  • Anomaly detection in time series using Orion

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published