ResourcePath¶

The following are the resource paths available in Olive.

ResourceType¶

class olive.resource_path.ResourceType(value)[source]¶

An enumeration.

AzureMLDatastore = 'azureml_datastore'¶

AzureMLJobOutput = 'azureml_job_output'¶

AzureMLModel = 'azureml_model'¶

AzureMLRegistryModel = 'azureml_registry_model'¶

LocalFile = 'file'¶

LocalFolder = 'folder'¶

StringName = 'string_name'¶

Each resource path is followed by a description of the path and a list of the its configuration options.

LocalFile¶

Local file resource path.

path¶

Path to the resource.

type: pathlib.Path | str

required: True

LocalFolder¶

Local folder resource path.

path¶

Path to the resource.

type: pathlib.Path | str

required: True

StringName¶

String name resource path.

name¶

Name of the resource.

type: str

required: True

AzureMLModel¶

AzureML Model resource path.

azureml_client¶

AzureML client config.

type: olive.azureml.azureml_client.AzureMLClientConfig

required: True

name¶

Name of the model.

type: str

required: True

version¶

Version of the model.

type: int | str

required: True

AzureMLClientConfig

pydantic settings olive.azureml.azureml_client.AzureMLClientConfig[source]¶

field subscription_id: str = None¶: Azure subscription id. Required if aml_config_path is not provided.

field resource_group: str = None¶: Azure resource group. Required if aml_config_path is not provided.

field workspace_name: str = None¶: Azure workspace name. Required if aml_config_path is not provided.

field aml_config_path: str = None¶: Path to AzureML config file. If provided, other fields are ignored.

field read_timeout: int = 60¶: Read timeout in seconds for HTTP requests.

field max_operation_retries: int = 3¶: Max number of retries for AzureML operations like resource creation or download.

field operation_retry_interval: int = 5¶: Initial interval in seconds between retries for AzureML operations like resource creation or download. The interval doubles after each retry.

field default_auth_params: Dict[str, Any] | None = None¶: Default auth config for AzureML client. Please refer to https://learn.microsoft.com/en-us/python/api/azure-identity/azure.identity.defaultazurecredential?view=azure-python#parameters for more details.

field keyvault_name: str | None = None¶: Name of the keyvault to use. If provided, the keyvault will be used to retrieve secrets.

get_workspace_config() → Dict[str, str][source]¶: Get the workspace config as a dict.

create_client()[source]¶: Create an MLClient instance.

create_registry_client(registry_name: str)[source]¶: Create an MLClient instance.

AzureMLDatastore¶

AzureML DataStore resource path.

azureml_client¶

AzureML client config.

type: olive.azureml.azureml_client.AzureMLClientConfig

default_value: None

datastore_name¶

Name of the datastore.

type: str

default_value: None

relative_path¶

Relative path to the resource.

type: str

default_value: None

datastore_url¶

URL of the datastore.

type: str

default_value: None

AzureMLJobOutput¶

AzureML job output resource path.

azureml_client¶

AzureML client config.

type: olive.azureml.azureml_client.AzureMLClientConfig

required: True

job_name¶

Name of the job.

type: str

required: True

output_name¶

Name of the output.

type: str

required: True

relative_path¶

Relative path to the resource.

type: str

required: True