Skip to content

Latest commit

 

History

History
31 lines (19 loc) · 1.03 KB

setting-up-trustyai-for-your-project.adoc

File metadata and controls

31 lines (19 loc) · 1.03 KB
Raw

Setting up TrustyAI for your project

To set up model monitoring with TrustyAI for a data science project, a data scientist does the following tasks:

  • Authenticate the TrustyAI service

  • Send training data to TrustyAI for bias or data drift monitoring

  • Label your data fields (optional)

After setting up, a data scientist can create and view bias and data drift metrics for deployed models.

Note

Model monitoring with TrustyAI is only available on the ModelMesh-based multi-model serving platform. Model monitoring with TrustyAI is unavailable on the KServe-based single-model serving platform.