Azure CLI ml extension Commands

This document provides a comprehensive reference for the az ml extension, which enables you to manage your Azure Machine Learning resources directly from your command line.

`az ml job`

Manage machine learning jobs.

Subgroups

create: Create a job.
list: List jobs.
show: Show job details.
cancel: Cancel a job.
stream: Stream job logs.

Example:

az ml job create --file job.yml

`az ml environment`

Manage machine learning environments.

Subgroups

create: Create an environment.
list: List environments.
show: Show environment details.
update: Update an environment.
delete: Delete an environment.

Example:

az ml environment create --file env.yml

`az ml component`

Manage machine learning components.

Subgroups

create: Create a component.
list: List components.
show: Show component details.
update: Update a component.
delete: Delete a component.

Example:

az ml component create --file component.yml

`az ml data`

Manage machine learning data assets.

Subgroups

create: Create a data asset.
list: List data assets.
show: Show data asset details.
update: Update a data asset.
delete: Delete a data asset.

Example:

az ml data create --file data.yml

`az ml workspace`

Manage machine learning workspaces.

Subgroups

create: Create a workspace.
list: List workspaces.
show: Show workspace details.
update: Update a workspace.
delete: Delete a workspace.

Example:

az ml workspace create --name myworkspace --resource-group myrg

`az ml compute`

Manage compute resources.

Subgroups

create: Create a compute resource.
list: List compute resources.
show: Show compute resource details.
update: Update a compute resource.
delete: Delete a compute resource.

Example:

az ml compute create --type AmlCompute --name gpu-cluster --min-instances 0 --max-instances 4

`az ml endpoint`

Manage inference endpoints.

Subgroups

create: Create an endpoint.
list: List endpoints.
show: Show endpoint details.
update: Update an endpoint.
delete: Delete an endpoint.
scoring-request: Make a scoring request to an endpoint.

Example:

az ml endpoint create --file endpoint.yml

`az ml model`

Manage model assets.

Subgroups

create: Create a model asset.
list: List model assets.
show: Show model asset details.
update: Update a model asset.
delete: Delete a model asset.

Example:

az ml model create --name my-model --version 1 --path ./model/

`az ml registry`

Manage model and component registries.

Subgroups

create: Create a registry.
list: List registries.
show: Show registry details.
update: Update a registry.
delete: Delete a registry.

Example:

az ml registry create --name my-registry --resource-group myrg

`az ml asset`

Manage Azure Machine Learning assets (jobs, environments, components, data, models).

This command is an alias for specific asset types (e.g., az ml job, az ml environment).

Example:

az ml asset list --type job

`az ml connection`

Manage connections to external services.

Subgroups

create: Create a connection.
list: List connections.
show: Show connection details.
update: Update a connection.
delete: Delete a connection.

Example:

az ml connection create --name my-storage-connection --resource-group myrg --workspace-name myworkspace --type storage --connection-string "DefaultEndpointsProtocol=https;AccountName=mystorageaccount;..."

`az ml code`

Manage code assets for Azure Machine Learning.

Subgroups

create: Upload and register code.
list: List registered code assets.
show: Show details of a code asset.
delete: Delete a code asset.

Example:

az ml code create --path ./src --name my-code-asset --version 1