Reliability pillar – Best practices

The reliability pillar encompasses the ability of a workload to perform its intended function correctly and consistently when it’s expected to. This section includes best practices to consider during model development that includes model training, tuning, and model performance evaluation.

Best practices

• MLREL-06: Enable CI/CD/CT automation with traceability
• MLREL-07: Ensure feature consistency across training and inference
• MLREL-08: Ensure model validation with relevant data
• MLREL-09: Establish data bias detection and mitigation