azure_ml_sdk.services.deployment.endpoint¶

Classes¶

`OnlineDeployment`	Class for defining deployments to be used in online endpoints.
`OnlineEndpoint`	Class for creating or updating online endpoints on Azure Machine Learning.
`OnlineEndpointManager`	Class for managing online endpoints on Azure Machine Learning.

Module Contents¶

class azure_ml_sdk.services.deployment.endpoint.OnlineDeployment(ml_client: azure.ai.ml.MLClient, name: str, endpoint_name: str, config: dict | None = {})¶

Class for defining deployments to be used in online endpoints.

ml_client¶

name¶

endpoint_name¶

property status: str | None¶: Returns the provisioning state of an online deployment.

property id: str | None¶: Returns the id of the online deployment.

get() → azure.ai.ml.entities.ManagedOnlineDeployment | None¶

Return the online deployment object.

Raises:: ValueError – If the online deployment is not found in the online endpoint.
Returns:: The online deployment object.
Return type:: Optional[ManagedOnlineDeployment]

get_logs(lines: int = 100) → str¶

Retrieve the logs from the online deployment.

Parameters:: lines (int) – The maximum number of lines to tail. Defaults to 100 lines.
Returns:: the logs of the online deployment.
Return type:: str

Configure the online deployment.

Parameters:

model (str) – The name of the registered model in the workspace.
environment (str) – The name of the registered environment in the workspace.
scoring_folder (str) – The path to the folder containing the scoring script.
scoring_script (str) – The name of the scroring script (.py).
instance_type (str) – The name name of the compute instance.
instance_count (int) – The number of instances to deploy. Defaults to 1.

class azure_ml_sdk.services.deployment.endpoint.OnlineEndpoint(ml_client: azure.ai.ml.MLClient, name: str)¶

Class for creating or updating online endpoints on Azure Machine Learning.

ml_client¶

name¶

property status: str | None¶: Return the provisioning state of the online endpoint.

property scoring_uri: str | None¶: Return the scoring uri of the online endpoint.

property keys: str | None¶: Return the authentication key of the online endpoint.

property id: str | None¶: Return the id of the online endpoint.

get() → azure.ai.ml.entities.ManagedOnlineEndpoint | None¶

_summary_.

Raises:: ResourceNotFoundError – _description_
Returns:: _description_
Return type:: Optional[ManagedOnlineEndpoint]

invoke(deployment_name: str, request_file: str) → None¶

Invoke the online endpoint with a payload in a json request file.

Parameters:

deployment_name (str) – The name of the online deployment to invoke.
request_file (str) – The path to a json file containing the request.

list_deployments() → List[str]¶

List all deployments of the online endpoint.

Returns:: List of the names of the deployments of the online endpoint.
Return type:: List[str]

get_deployment(deployment_name: str) → OnlineDeployment | None¶

Get a deployment of the online endpoint.

Raises:: ValueError – If the deployment is not found in the online endpoint.
Parameters:: deployment_name (str) – The name of the deployment.
Returns:: The online deployment, if it exists.
Return type:: Optional[OnlineDeployment]

create_or_update_deployment(deployment_name: str, config: dict | None = {}) → OnlineDeployment¶

Create a deployment in the online Endpoint.

Parameters:

deployment_name (str) – The name of the deployment to create or update.
config (dict) – The dictionary containing the deployment configuration.

Raises:

ResourceNotFoundError – if the endpoint is not found in the workspace.

delete_deployment(deployment_name: str) → None¶

Delete a deployment from an online endpoint.

Parameters:: deployment_name (str) – The name of the deployment.
Raises:: ValueError – If the deployment is not found in the online endpoint.

class azure_ml_sdk.services.deployment.endpoint.OnlineEndpointManager(ml_client: azure.ai.ml.MLClient)¶

Class for managing online endpoints on Azure Machine Learning.

ml_client¶

list_endpoints() → List[str]¶

List all online endpoints in the workspace.

Returns:: A list of the names of the endpoints of the workspace.

create_endpoint(endpoint_name: str, description: str, auth_mode: str = 'key') → OnlineEndpoint¶

Create an online endpoint in the workspace.

Parameters:

endpoint_name (str) – name of the endpoint.
description (str) – description of the endpoint.
auth_mode (str) – Authentication mode. Possible values: ‘aml_token’, ‘key’. Defaults to ‘key’.

delete_endpoint(endpoint_name: str) → None¶

Delete an online endpoint in the workspace.

Parameters:: endpoint_name (str) – name of the endpoint.

get_endpoint(endpoint_name: str) → OnlineEndpoint¶

Get an online endpoint (if deployed).

Returns:: The online endpoint instance.
Return type:: OnlineEndpoint