# yzhao062/pyod

A Python library for anomaly detection across tabular, time series, graph, text, image, and audio data. 60+ detectors, benchmark-backed ADEngine orchestration, and an agentic workflow for AI agents.

Repository: https://github.com/yzhao062/pyod
Canonical: https://ross.abutalabs.com/products/pyod
Homepage: https://pyod.dev/
Language: Python
License: BSD-2-Clause
License Family: permissive
Topics: outlier-detection, anomaly-detection, outlier-ensembles, machine-learning, data-mining, unsupervised-learning, fraud-detection, deep-learning, data-science, novelty-detection, out-of-distribution-detection, foundation-models, image-anomaly-detection, multimodal, nlp-anomaly-detection, agentic-ai, graph-anomaly-detection, time-series, time-series-anomaly-detection
Last push: 2026-08-19T03:12:27+00:00

## Health v2 (maintenance only)
Score: 98/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 98, longevity 100
- inputs: {"age_days": 3256, "days_push": 14, "days_rel": 16, "gap_med": 12, "n_releases_24m": 22}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 9977, forks 1499 (observed 2026-08-28T04:10:39.143739+00:00)

## What it is
PyOD is the most comprehensive Python library for anomaly detection, offering 60+ detectors across tabular, time series, graph, text, image, and audio data under a unified fit/predict API. Version 3 adds ADEngine orchestration, benchmark-backed planning, and agentic/MCP workflows for AI agents.

## Use cases
- detect anomalies in tabular data
- find outliers in time series
- detect fraud in transactions
- identify anomalous images
- detect outliers in graph data
- flag novelty in text documents
- run anomaly detection from an AI agent

## When to choose
- you need a wide selection of anomaly detection algorithms behind one consistent API
- you work with multiple data types (tabular, time series, graph, text, image)
- you want a mature, benchmark-backed, widely adopted library with 50M+ installs

## When to avoid
- you need supervised classification rather than unsupervised outlier detection
- you need real-time streaming anomaly detection at very low latency
- you want a turnkey GUI product rather than a Python library

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, nlp, data-science, agent-framework, mcp
- domain: machine-learning, data-science, artificial-intelligence, analytics, time-series
- platform: python, cross-platform
- tags: anomaly-detection, outlier-detection, unsupervised-learning, fraud-detection, novelty-detection, time-series, graph, multimodal, agentic-ai, benchmarking

## Member repositories
- yzhao062/pyod (main) score 98

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:39.143739+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:20:02.650474+00:00, confidence not recorded.
  - readme: https://github.com/yzhao062/pyod (fetched 2026-08-28T04:10:39.143739+00:00, sha 299918a183e2)
  - homepage: https://pyod.dev/ (fetched 2026-08-29T08:19:48.814569+00:00, sha bdc72e350364)
  - registry_pypi: https://pypi.org/pypi/pyod/json (fetched 2026-08-29T08:19:48.824348+00:00, sha 20c6e2f97340)
- Data as of 2026-08-30T08:39:29.467469+00:00.