Azure Machine Learning data collector provides real-time logging of input and output data from models that are deployed to managed online endpoints or Kubernetes online endpoints. Azure Machine Learning stores the logged inference data in Azure blob storage. This data can then be seamlessly used for model monitoring, debugging, or auditing, thereby providing observability into the performance of your deployed models.
Now generally available, data collector provides:
- Logging of inference data to a central location (Azure Blob Storage)
- Support for managed online endpoints and Kubernetes online endpoints
- Definition at the deployment level, allowing maximum changes to its configuration
- Support for both payload and custom logging
Learn more about data collection from models in production
Learn more about collecting production data from models deployed for real-time inferencing