AI 可观测性平台,核心是开源的 whylogs 数据画像库:用极小的统计摘要给训练集和线上流量建画像,从而做数据漂移、输入分布异常和模型退化的检测,不必回传原始数据。适合把监控嵌进既有数据管道、又对隐私敏感的场景。 ⭐ GitHub Stars: 2.8k+ | 🔗 仓库: https://github.com/whylabs/whylogs
An AI observability platform built around the open-source whylogs profiling library. It turns training sets and production traffic into compact statistical profiles, then detects data drift, anomalous input distributions and model degradation without shipping raw records around. A good fit when monitoring must sit inside an existing data pipeline and privacy constraints matter. ⭐ GitHub Stars: 2.8k+ | 🔗 Repo: https://github.com/whylabs/whylogs
WhyLabs belongs to the AI Agent Platforms category on hedirbase. Tagged with: AI可观测性,数据漂移,whylogs,模型监控.
An AI observability platform built around the open-source whylogs profiling library. It turns training sets and production traffic into compact statistical profiles, then detects data drift, anomalous