create_training_plan ¶

def create_training_plan(
    self,
    *,
    config_overrides: Optional[SageMakerClientConfig] = None,
    training_plan_name: Optional[
        "capo_sagemaker.types.training_plan_name.TrainingPlanName"
    ] = None,
    training_plan_offering_id: Optional[
        "capo_sagemaker.types.training_plan_offering_id.TrainingPlanOfferingId"
    ] = None,
    spare_instance_count_per_ultra_server: Optional[
        "capo_sagemaker.types.spare_instance_count_per_ultra_server.SpareInstanceCountPerUltraServer"
    ] = None,
    tags: Optional["capo_sagemaker.types.tag_list.TagList"] = None,
) -> (
    "capo_sagemaker.types.create_training_plan_response.CreateTrainingPlanResponse"
):
    """<p>Creates a new training plan in SageMaker to reserve compute capacity.</p> <p>Amazon SageMaker Training Plan is a capability within SageMaker that allows customers to reserve and manage GPU capacity for large-scale AI model training. It provides a way to secure predictable access to computational resources within specific timelines and budgets, without the need to manage underlying infrastructure. </p> <p> <b>How it works</b> </p> <p>Plans can be created for specific resources such as SageMaker Training Jobs or SageMaker HyperPod clusters, automatically provisioning resources, setting up infrastructure, executing workloads, and handling infrastructure failures.</p> <p> <b>Plan creation workflow</b> </p> <ul> <li> <p>Users search for available plan offerings based on their requirements (e.g., instance type, count, start time, duration) using the <code> <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_SearchTrainingPlanOfferings.html">SearchTrainingPlanOfferings</a> </code> API operation.</p> </li> <li> <p>They create a plan that best matches their needs using the ID of the plan offering they want to use. </p> </li> <li> <p>After successful upfront payment, the plan's status becomes <code>Scheduled</code>. </p> </li> <li> <p>The plan can be used to:</p> <ul> <li> <p>Queue training jobs.</p> </li> <li> <p>Allocate to an instance group of a SageMaker HyperPod cluster. </p> </li> </ul> </li> <li> <p>When the plan start date arrives, it becomes <code>Active</code>. Based on available reserved capacity:</p> <ul> <li> <p>Training jobs are launched.</p> </li> <li> <p>Instance groups are provisioned.</p> </li> </ul> </li> </ul> <p> <b>Plan composition</b> </p> <p>A plan can consist of one or more Reserved Capacities, each defined by a specific instance type, quantity, Availability Zone, duration, and start and end times. For more information about Reserved Capacity, see <code> <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_ReservedCapacitySummary.html">ReservedCapacitySummary</a> </code>.</p>

    Args:
        training_plan_name: <p>The name of the training plan to create.</p>
        training_plan_offering_id: <p>The unique identifier of the training plan offering to use for creating this plan.</p>
        spare_instance_count_per_ultra_server: <p>Number of spare instances to reserve per UltraServer for enhanced resiliency. Default is 1.</p>
        tags: <p>An array of key-value pairs to apply to this training plan.</p>

    Raises:
        capo_sagemaker.errors.resource_in_use.ResourceInUse: <p>Resource being accessed is in use.</p>
        capo_sagemaker.errors.resource_limit_exceeded.ResourceLimitExceeded: <p> You have exceeded an SageMaker resource limit. For example, you might have too many training jobs created. </p>
        capo_sagemaker.errors.resource_not_found.ResourceNotFound: <p>Resource being access is not found.</p>
        capo_sagemaker.errors.UnknownServiceError: The service returned an error code this client does not model.
    """

    def _handler(
        req: "OperationRequest[capo_sagemaker.types.create_training_plan_request.CreateTrainingPlanRequest]",
    ) -> OperationResponse[
        "capo_sagemaker.types.create_training_plan_response.CreateTrainingPlanResponse"
    ]:
        import capo_sagemaker._operations.sage_maker.create_training_plan

        output, http_response = (
            capo_sagemaker._operations.sage_maker.create_training_plan.create_training_plan(
                req.options, req.input
            )
        )
        return OperationResponse(output=output, response=http_response)

    interceptors_, options_ = self.operation_options(
        config_overrides, OperationLogging("CreateTrainingPlan")
    )
    input_: capo_sagemaker.types.create_training_plan_request.CreateTrainingPlanRequest = {}
    if training_plan_name is not None:
        input_["training_plan_name"] = training_plan_name
    if training_plan_offering_id is not None:
        input_["training_plan_offering_id"] = training_plan_offering_id
    if spare_instance_count_per_ultra_server is not None:
        input_["spare_instance_count_per_ultra_server"] = (
            spare_instance_count_per_ultra_server
        )
    if tags is not None:
        input_["tags"] = tags

    response = execute_pipeline(
        OperationRequest(input=input_, options=options_),
        handler=_handler,
        interceptors=list(interceptors_),
    )
    response.response.close()
    return response.output