🤖 AI Summary
Streaming anomaly detection faces stringent constraints—including memory efficiency, single-pass processing, and constant-time complexity—yet existing methods are predominantly batch-oriented, lacking real-time capability and memory controllability. To address this gap, we propose PySAD: the first open-source, streaming-native Python framework for anomaly detection. It systematically integrates over a dozen state-of-the-art online algorithms and supports end-to-end experimental workflows, including feature projection, incremental learning, and probabilistic calibration. PySAD uniquely unifies dynamic model updating, single-sample inference, and reproducible infrastructure—including comprehensive unit tests and CI/CD pipelines—within a lightweight, modular architecture. Built upon PyOD and scikit-learn and strictly adhering to PEP8, it lowers barriers to streaming AD research and rapid prototyping. Empirical adoption across academia and industry demonstrates its effectiveness in enhancing evaluation consistency, reproducibility, and practical deployability.
📝 Abstract
PySAD is an open-source python framework for anomaly detection on streaming data. PySAD serves various state-of-the-art methods for streaming anomaly detection. The framework provides a complete set of tools to design anomaly detection experiments ranging from projectors to probability calibrators. PySAD builds upon popular open-source frameworks such as PyOD and scikit-learn. We enforce software quality by enforcing compliance with PEP8 guidelines, functional testing and using continuous integration. The source code is publicly available on this https URL.
