SagemakerTransformProps

class aws_cdk.aws_stepfunctions_tasks.SagemakerTransformProps(*, model_name, transform_input, transform_job_name, transform_output, batch_strategy=None, environment=None, integration_pattern=None, max_concurrent_transforms=None, max_payload_in_mb=None, role=None, tags=None, transform_resources=None)

Bases: object

__init__(*, model_name, transform_input, transform_job_name, transform_output, batch_strategy=None, environment=None, integration_pattern=None, max_concurrent_transforms=None, max_payload_in_mb=None, role=None, tags=None, transform_resources=None)
Parameters
  • model_name (str) – Name of the model that you want to use for the transform job.

  • transform_input (TransformInput) – Dataset to be transformed and the Amazon S3 location where it is stored.

  • transform_job_name (str) – Training Job Name.

  • transform_output (TransformOutput) – S3 location where you want Amazon SageMaker to save the results from the transform job.

  • batch_strategy (Optional[BatchStrategy]) – Number of records to include in a mini-batch for an HTTP inference request.

  • environment (Optional[Mapping[str, str]]) – Environment variables to set in the Docker container.

  • integration_pattern (Optional[ServiceIntegrationPattern]) – The service integration pattern indicates different ways to call SageMaker APIs. The valid value is either FIRE_AND_FORGET or SYNC. Default: FIRE_AND_FORGET

  • max_concurrent_transforms (Union[int, float, None]) – Maximum number of parallel requests that can be sent to each instance in a transform job.

  • max_payload_in_mb (Union[int, float, None]) – Maximum allowed size of the payload, in MB.

  • role (Optional[IRole]) – Role for thte Training Job.

  • tags (Optional[Mapping[str, str]]) – Tags to be applied to the train job.

  • transform_resources (Optional[TransformResources]) – ML compute instances for the transform job.

stability :stability: experimental

Attributes

batch_strategy

Number of records to include in a mini-batch for an HTTP inference request.

stability :stability: experimental

Return type

Optional[BatchStrategy]

environment

Environment variables to set in the Docker container.

stability :stability: experimental

Return type

Optional[Mapping[str, str]]

integration_pattern

The service integration pattern indicates different ways to call SageMaker APIs.

The valid value is either FIRE_AND_FORGET or SYNC.

default :default: FIRE_AND_FORGET

stability :stability: experimental

Return type

Optional[ServiceIntegrationPattern]

max_concurrent_transforms

Maximum number of parallel requests that can be sent to each instance in a transform job.

stability :stability: experimental

Return type

Union[int, float, None]

max_payload_in_mb

Maximum allowed size of the payload, in MB.

stability :stability: experimental

Return type

Union[int, float, None]

model_name

Name of the model that you want to use for the transform job.

stability :stability: experimental

Return type

str

role

Role for thte Training Job.

stability :stability: experimental

Return type

Optional[IRole]

tags

Tags to be applied to the train job.

stability :stability: experimental

Return type

Optional[Mapping[str, str]]

transform_input

Dataset to be transformed and the Amazon S3 location where it is stored.

stability :stability: experimental

Return type

TransformInput

transform_job_name

Training Job Name.

stability :stability: experimental

Return type

str

transform_output

S3 location where you want Amazon SageMaker to save the results from the transform job.

stability :stability: experimental

Return type

TransformOutput

transform_resources

ML compute instances for the transform job.

stability :stability: experimental

Return type

Optional[TransformResources]