BatchDeployment Class

Batch endpoint deployment entity.

Warning This class should not be used directly. Please use one of the child implementations, ModelBatchDeployment or PipelineComponentBatchDeployment.

Constructor

BatchDeployment(*, name: str, endpoint_name: str | None = None, description: str | None = None, tags: Dict[str, Any] | None = None, properties: Dict[str, str] | None = None, model: str | Model | None = None, code_configuration: CodeConfiguration | None = None, environment: str | Environment | None = None, compute: str | None = None, resources: ResourceConfiguration | None = None, output_file_name: str | None = None, output_action: BatchDeploymentOutputAction | str | None = None, error_threshold: int | None = None, retry_settings: BatchRetrySettings | None = None, logging_level: str | None = None, mini_batch_size: int | None = None, max_concurrency_per_instance: int | None = None, environment_variables: Dict[str, str] | None = None, code_path: PathLike | str | None = None, scoring_script: PathLike | str | None = None, instance_count: int | None = None, **kwargs: Any)

Keyword-Only Parameters

Name	Description
name	str the name of the batch deployment
description	str Description of the resource. Default value: None
tags	dict[str, str] Tag dictionary. Tags can be added, removed, and updated. Default value: None
properties	dict[str, str] The asset property dictionary. Default value: None
model	Union[str, Model] Model entity for the endpoint deployment, defaults to None Default value: None
code_configuration	CodeConfiguration defaults to None Default value: None
environment	Union[str, Environment] Environment entity for the endpoint deployment., defaults to None Default value: None
compute	str Compute target for batch inference operation. Default value: None
output_action	str or BatchDeploymentOutputAction Indicates how the output will be organized. Possible values include: "summary_only", "append_row". Defaults to "append_row" Default value: None
output_file_name	str Customized output file name for append_row output action, defaults to "predictions.csv" Default value: None
max_concurrency_per_instance	int Indicates maximum number of parallelism per instance, defaults to 1 Default value: None
error_threshold	int Error threshold, if the error count for the entire input goes above this value, the batch inference will be aborted. Range is [-1, int.MaxValue] -1 value indicates, ignore all failures during batch inference For FileDataset count of file failures For TabularDataset, this is the count of record failures, defaults to -1 Default value: None
retry_settings	BatchRetrySettings Retry settings for a batch inference operation, defaults to None Default value: None
resources	ResourceConfiguration Indicates compute configuration for the job. Default value: None
logging_level	str Logging level for batch inference operation, defaults to "info" Default value: None
mini_batch_size	int Size of the mini-batch passed to each batch invocation, defaults to 10 Default value: None
environment_variables	dict Environment variables that will be set in deployment. Default value: None
code_path	Union[str, <xref:PathLike>] Folder path to local code assets. Equivalent to code_configuration.code. Default value: None
scoring_script	Union[str, <xref:PathLike>] Scoring script name. Equivalent to code_configuration.code.scoring_script. Default value: None
instance_count	int Number of instances the interfering will run on. Equivalent to resources.instance_count. Default value: None
endpoint_name	Default value: None

Methods

dump	Dump the deployment content into a file in yaml format.

dump

Dump the deployment content into a file in yaml format.

dump(dest: str | PathLike | IO, **kwargs: Any) -> None

Parameters

Name	Description
dest Required	Union[PathLike, str, IO[AnyStr]] The destination to receive this deployment's content. Must be either a path to a local file, or an already-open file stream. If dest is a file path, a new file will be created, and an exception is raised if the file exists. If dest is an open file, the file will be written to directly, and an exception will be raised if the file is not writable.

Attributes

base_path

The base path of the resource.

Returns

Type	Description
str	The base path of the resource.

code_path

The code directory containing the scoring script.

Returns

Type	Description
Union[str, <xref:PathLike>]

creation_context

The creation context of the resource.

Returns

Type	Description
Optional[SystemData]	The creation metadata for the resource.

id

The resource ID.

Returns

Type	Description
Optional[str]	The global ID of the resource, an Azure Resource Manager (ARM) ID.

instance_count

provisioning_state

Batch deployment provisioning state, readonly.

Returns

Type	Description
Optional[str]	Batch deployment provisioning state.

scoring_script

The scoring script file path relative to the code directory.

Returns

Type	Description
Union[str, <xref:PathLike>]

type

Type of deployment.

Returns

Type	Description
str

Feedback

Was this page helpful?