A Python library for anomaly detection across tabular, time series, graph, text, image and audio data, with 60+ detectors and a classic fit/predict API.
PyOD is a Python library for outlier and anomaly detection. It provides a large set of detection algorithms, from classic methods such as Isolation Forest, LOF and kNN to neural approaches such as autoencoders, VAE and DeepSVDD, all behind a consistent fit/decision_function/predict interface. Detectors are grouped by data modality, covering tabular data, time series, graph data and, through foundation-model embeddings, text, images and audio.
Beyond the classic API, version 3 adds ADEngine, an orchestration layer that profiles data, plans and builds a detector, runs detection and explains the findings. It also ships an od-expert skill for Claude Code and Codex, and an MCP server that exposes ten stateless tools for MCP-compatible LLM agents. The classic fit/predict workflow remains available for users who already know which detector they need.
The project includes utilities for synthetic data generation, evaluation metrics such as Precision@n, score combination methods for ensembles, and model saving with joblib or pickle. It is installable via pip and conda-forge.
Install the core library from PyPI
pip install pyodFit a detector and read anomaly scores
Summary generated by AI from the project's README. GitHub stats refresh daily.
Python SDK client for Nixtla's TimeGPT, a hosted pre-trained foundation model for time series forecasting and anomaly detection.
An LLM agent framework from Microsoft that automates data-driven R&D tasks such as reading papers, implementing models and factors, and running Kaggle-style experiments.
Open-source real-time infrastructure monitoring agent with per-second metrics, auto-discovery, built-in ML anomaly detection and alerting.
A curated list of open source libraries for deploying, monitoring, versioning, scaling and securing production machine learning.
AutoML library that trains and ensembles ML models for tabular, time series, and multimodal data with a few lines of Python.
Python machine learning library built on SciPy and NumPy, offering classification, regression, clustering, and model selection tools under a BSD license.