Observabilité du modèle

Définition

XXXXXXXXX

Français

XXXXXXXXX

Anglais

Model Observability

ML observability features are more interested to understand why something is happening. That what vs. why focus is what fundamentally differentiates monitoring versus observability. For instance, let’s take a simple scenario of model performance drift. An ML monitoring stack will be able to detect the performance degradation in the model. In contrast, an ML observability stack will compare data distributions and other key indicators to help pinpoint the cause of the drift.    Robust ML observability should include: 

• Model lineage, validation, comparison

• Data quality monitoring and troubleshooting 

• Drift monitoring/troubleshooting 

• Performance monitoring/troubleshooting 

• Business impact analysis 

Source : thesequence