pygod is a Python library designed for detecting outliers in graph-structured data using advanced anomaly detection algorithms.
A Python Library for Graph Outlier Detection (Anomaly Detection)
pygod is used by data scientists and security analysts to identify anomalous nodes or edges within graph data, which is critical for applications like fraud detection, network security, and AI model security scanning. It helps uncover hidden threats or irregular patterns in complex relational data through graph outlier detection techniques.
For best results, ensure graph data is preprocessed correctly and formatted as expected by pygod. Some algorithms may require tuning hyperparameters based on the specific graph structure and domain. Large graphs may need sufficient computational resources.
Ensure Python 3.6 or higher is installed
Install pygod via pip: pip install pygod
Optionally, clone the repository for latest updates: git clone https://github.com/pygod-team/pygod.git
Navigate to the cloned directory and install dependencies if neededfrom pygod import OutlierDetector model = OutlierDetector(method='CoLA') model.fit(graph_data) scores = model.predict()
Initialize and train an outlier detection model on graph data, then predict anomaly scores.
pygod --help
Display help information and available command line options.