Azure Machine Learning

In the Ultra Transcenders books

AI-901AI-300AI-103

Each book explains Azure Machine Learning in context, with comparison tables and the common traps.

Terms in this definition

• WHERE

  Limits a SELECT, UPDATE or DELETE to just the rows meeting a condition. Omit it, and the statement hits every row.

• RAMP

  The approach used by the Cloud Adoption Framework's Manage methodology to structure cloud operations: teams, duties, processes and tools. It divides what the central platform handles from what each workload handles.

• Online management group

  Workloads that may talk directly to or from the internet, or that need no virtual network at all, go in this management group. It is a sibling of Corp and Local beneath Landing zones.

• Model monitoring

  Capability in Azure Machine Learning that, on a schedule, checks production inference data against reference data for each signal. When a metric crosses its threshold it raises an alert, by email unless Event Grid events are configured.

• Prompt flow

  Tool in Foundry and Azure Machine Learning for assembling LLM applications as graphs of nodes (LLM, prompt and Python tools), with variants and evaluation. Microsoft Agent Framework is replacing it as it is retired.

Related terms

• AmlOnlineEndpointConsoleLog

  Table in Log Analytics capturing console output from Azure Machine Learning online endpoint containers, useful when a container will not start or handles requests incorrectly.

• az ml model restore

  Command in the Azure Machine Learning CLI that un-archives a model or model version so it shows up in listings again, for rollback or auditing.

• az ml online-deployment get-logs

  Azure Machine Learning CLI command fetching container logs for a deployment, from the inference server by default or from storage-initializer, to troubleshoot issues like init() failures or missing packages.

• Azure AI Administrator

  Built-in role meant for Azure Machine Learning and Foundry hubs only, where it holds every control plane permission over Azure AI and the services it relies on. Today's Foundry resources are governed by Foundry Owner or Foundry Account Owner instead.

• Azure Container Instances

  Way to run container groups without any cluster, billed per second. It offers neither autoscaling nor spreading across zones, and Azure Machine Learning treats it as a legacy v1 deployment target.

• Azure Machine Learning designer

  Visual, drag-and-drop tool inside Azure Machine Learning studio for assembling pipelines that train models. Foundry models are browsed and deployed elsewhere.

• Azure Machine Learning SDK v2

  Python SDK currently used with Azure Machine Learning, centred on the MLClient class. It has no logging API, so MLflow is used to log metrics.

• Azure Machine Learning workspace

  Main Azure Machine Learning resource, holding models, environments, components and data assets, recording job history and pointing to compute and datastores. Anything registered there is owned by that workspace unless it is put in a registry.