Minqi824/ADBench

Official Implement of "ADBench: Anomaly Detection Benchmark", NeurIPS 2022.

63
/ 100
Established

ADBench helps data scientists, machine learning engineers, and researchers evaluate how well different anomaly detection algorithms perform on various tabular datasets. It takes your choice of anomaly detection algorithm and dataset, then outputs performance metrics, helping you understand an algorithm's strengths and weaknesses across different anomaly types and data quality issues. This allows practitioners to select the most suitable algorithm for their specific use case.

1,008 stars. Available on PyPI.

Use this if you need to benchmark and compare anomaly detection algorithms on tabular data, or if you're developing a new anomaly detection method and want to test its performance against a comprehensive suite of existing solutions and datasets.

Not ideal if your anomaly detection needs involve non-tabular data types like time-series, graphs, or images, or if you're looking for a simple, out-of-the-box solution without deep performance analysis.

data-science machine-learning-research data-quality fraud-detection system-monitoring
Maintenance 6 / 25
Adoption 10 / 25
Maturity 25 / 25
Community 22 / 25

How are scores calculated?

Stars

1,008

Forks

151

Language

Python

License

BSD-2-Clause

Last pushed

Jan 08, 2026

Commits (30d)

0

Dependencies

16

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Minqi824/ADBench"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.