yzhao062 / pyod

94%

master: 95%

LAST BUILD ON BRANCH development

branch: SELECT

CHANGE BRANCH

x

• No branch selected
• 0.6.X
• CD-test-data-align
• John-Almardeny-COF_Yahya
• V0.5.0
• V0.7.8
• V0.7.9
• V0.8.1
• V0.8.3
• V0.8.4
• V0.8.6
• V0.8.8
• V0.8.9
• V0.9.1
• V0.9.2
• V6.0.5
• autoencoder-kwargs
• ci-python
• dependabot/pip/docs/tensorflow-1.15.2
• development
• fix-hbos-auto-bins-indexerror
• fix-refs
• issue-180-temp-fix
• jmlr
• lscp_implementation
• master
• nlp-adbench
• origin/jmlr
• pr_test
• pyproject.toml
• pythresh-v1
• refactor_combination
• remove-deprecated
• revert-224-copod
• revert-61-SOD_Yahya
• v.0.6.7
• v0.4.5
• v0.7.0
• v0.7.5
• v0.7.6

Build # 24177838518

Build Type

push

github

Committed by yzhao062

Commit Message

feat: add ADEngine Tier B lifecycle methods (run, analyze, explain, iterate, report)

Full anomaly detection lifecycle on ADEngine:
- run_detection() with timing and score summary
- analyze_results() with score distribution, top anomalies, feature importance
- explain_findings() with per-sample explanations and contributing features
- suggest_next_step() with feedback parsing and threshold adjustment payloads
- generate_report() with markdown and JSON output
- detect() reimplemented on run_detection() for Tier B compatibility
- Updated od-expert skill and example with full lifecycle
- 86 tests passing, 3 review rounds

Coverage Stats

374 of 393 new or added lines in 2 files covered. (95.17%)

12562 of 13405 relevant lines covered (93.71%)

9.37 hits per line