datamllab/tods

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1666 stars · 206 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

TODS is a full-stack automated machine learning system for outlier detection on multivariate time-series data, developed by DATA Lab at Rice University. It provides modules for preprocessing, feature extraction, a wide range of detection algorithms (point-wise, pattern-wise, system-wise), and AutoML pipeline search.

Use cases

• detect anomalies in multivariate time-series sensor data
• automatically build an outlier detection pipeline with AutoML
• find anomalous subsequences in time series
• detect fraudulent transactions in time-series data
• monitor blockchain data for outliers
• apply PyOD point-wise detectors to time series

When to choose

• you need automated, knowledge-free anomaly detection pipelines for time-series data
• you want access to many detection algorithms (PyOD, DeepLog, Telemanom) in one package
• you need point-wise, pattern-wise, or system-wise outlier detection scenarios

When to avoid

• you need real-time streaming anomaly detection at low latency
• you work outside Python or need actively maintained software with frequent updates
• your data is not time-series (use PyOD or general anomaly detection tools instead)

Facets

library · maturity maintenance

machine-learning data-science etl machine-learning data-science time-series analytics python windows anomaly-detection outlier-detection automl time-series automl human-in-the-loop linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="datamllab/tods")