sagemaker.core.jumpstart.factory.utils#
This module stores JumpStart factory utilities.
Functions
|
Returns kwargs required to call deploy on sagemaker.estimator.Model object. |
|
Returns kwargs required to instantiate sagemaker.estimator.Model object. |
|
Returns a dictionary of model info kwargs to use with JumpStart APIs. |
- sagemaker.core.jumpstart.factory.utils.get_deploy_kwargs(model_id: str, model_version: str | None = None, hub_arn: str | None = None, model_type: JumpStartModelType = JumpStartModelType.OPEN_WEIGHTS, region: str | None = None, initial_instance_count: int | None = None, instance_type: str | None = None, serializer: BaseSerializer | None = None, deserializer: BaseDeserializer | None = None, accelerator_type: str | None = None, endpoint_name: str | None = None, inference_component_name: str | None = None, tags: List[Dict[str, str | PipelineVariable]] | Dict[str, str | PipelineVariable] | None = None, kms_key: str | None = None, wait: bool | None = None, data_capture_config: DataCaptureConfig | None = None, async_inference_config: AsyncInferenceConfig | None = None, serverless_inference_config: ServerlessInferenceConfig | None = None, volume_size: int | None = None, model_data_download_timeout: int | None = None, container_startup_health_check_timeout: int | None = None, inference_recommendation_id: str | None = None, explainer_config: ExplainerConfig | None = None, tolerate_vulnerable_model: bool | None = None, tolerate_deprecated_model: bool | None = None, sagemaker_session: Session | None = None, accept_eula: bool | None = None, model_reference_arn: str | None = None, endpoint_logging: bool | None = None, resources: ResourceRequirements | None = None, managed_instance_scaling: str | None = None, endpoint_type: EndpointType | None = None, training_config_name: str | None = None, config_name: str | None = None, routing_config: Dict[str, Any] | None = None, model_access_configs: Dict[str, ModelAccessConfig] | None = None, inference_ami_version: str | None = None) JumpStartModelDeployKwargs[source]#
Returns kwargs required to call deploy on sagemaker.estimator.Model object.
- sagemaker.core.jumpstart.factory.utils.get_init_kwargs(model_id: str, model_from_estimator: bool = False, model_version: str | None = None, hub_arn: str | None = None, model_type: JumpStartModelType | None = JumpStartModelType.OPEN_WEIGHTS, tolerate_vulnerable_model: bool | None = None, tolerate_deprecated_model: bool | None = None, instance_type: str | None = None, region: str | None = None, image_uri: str | PipelineVariable | None = None, model_data: str | PipelineVariable | dict | None = None, role: str | None = None, env: Dict[str, str | PipelineVariable] | None = None, name: str | None = None, vpc_config: Dict[str, List[str | PipelineVariable]] | None = None, sagemaker_session: Session | None = None, enable_network_isolation: bool | PipelineVariable | None = None, model_kms_key: str | None = None, image_config: Dict[str, str | PipelineVariable] | None = None, source_dir: str | None = None, code_location: str | None = None, entry_point: str | None = None, container_log_level: int | PipelineVariable | None = None, dependencies: List[str] | None = None, git_config: Dict[str, str] | None = None, model_package_arn: str | None = None, training_instance_type: str | None = None, disable_instance_type_logging: bool = False, resources: ResourceRequirements | None = None, config_name: str | None = None, additional_model_data_sources: Dict[str, Any] | None = None) JumpStartModelInitKwargs[source]#
Returns kwargs required to instantiate sagemaker.estimator.Model object.
- sagemaker.core.jumpstart.factory.utils.get_model_info_default_kwargs(kwargs: JumpStartModelDeployKwargs | JumpStartModelInitKwargs | JumpStartEstimatorFitKwargs | JumpStartEstimatorInitKwargs | JumpStartEstimatorDeployKwargs, include_config_name: bool = True, include_model_version: bool = True, include_tolerate_flags: bool = True) dict[source]#
Returns a dictionary of model info kwargs to use with JumpStart APIs.