Skip to main content

model_packages

Creates, updates, deletes or gets a model_package resource or lists model_packages in a region

Overview

Namemodel_packages
TypeResource
DescriptionResource Type definition for AWS::SageMaker::ModelPackage
Idawscc.sagemaker.model_packages

Fields

NameDatatypeDescription
tagsarrayAn array of key-value pairs to apply to this resource.
additional_inference_specificationsarrayAn array of additional Inference Specification objects.
certify_for_marketplacebooleanWhether to certify the model package for listing on AWS Marketplace.
client_tokenstringA unique token that guarantees that the call to this API is idempotent.
customer_metadata_propertiesobjectThe metadata properties associated with the model package versions.
domainstringThe machine learning domain of the model package you specified.
drift_check_baselinesobjectRepresents the drift check baselines that can be used when the model monitor is set using the model package.
inference_specificationobjectDetails about inference jobs that can be run with models based on this model package.
metadata_propertiesobjectMetadata properties of the tracking entity, trial, or trial component.
model_approval_statusstringThe approval status of the model package.
model_metricsobjectA structure that contains model metrics reports.
model_package_descriptionstringThe description of the model package.
model_package_group_namestringThe name of the model package group.
model_package_namestringThe name or arn of the model package.
sample_payload_urlstringThe Amazon Simple Storage Service (Amazon S3) path where the sample payload are stored pointing to single gzip compressed tar archive.
skip_model_validationstringIndicates if you want to skip model validation.
source_algorithm_specificationobjectDetails about the algorithm that was used to create the model package.
taskstringThe machine learning task your model package accomplishes.
validation_specificationobjectSpecifies configurations for one or more transform jobs that Amazon SageMaker runs to test the model package.
model_package_arnstringThe Amazon Resource Name (ARN) of the model package group.
approval_descriptionstringA description provided for the model approval.
creation_timestringThe time at which the model package was created.
last_modified_timestringThe time at which the model package was last modified.
model_package_statusstringThe current status of the model package.
model_package_versionintegerThe version of the model package.
model_package_status_detailsobjectDetails about the current status of the model package.
source_uristringThe URI of the source for the model package.
model_cardobjectThe model card associated with the model package.
security_configobjectAn optional AWS Key Management Service key to encrypt, decrypt, and re-encrypt model package information for regulated workloads with highly sensitive data.
regionstringAWS region.

For more information, see AWS::SageMaker::ModelPackage.

Methods

NameResourceAccessible byRequired Params
create_resourcemodel_packagesINSERTregion
delete_resourcemodel_packagesDELETEIdentifier, region
update_resourcemodel_packagesUPDATEIdentifier, PatchDocument, region
list_resourcesmodel_packages_list_onlySELECTregion
get_resourcemodel_packagesSELECTIdentifier, region

SELECT examples

Gets all properties from an individual model_package.

SELECT
region,
tags,
additional_inference_specifications,
certify_for_marketplace,
client_token,
customer_metadata_properties,
domain,
drift_check_baselines,
inference_specification,
metadata_properties,
model_approval_status,
model_metrics,
model_package_description,
model_package_group_name,
model_package_name,
sample_payload_url,
skip_model_validation,
source_algorithm_specification,
task,
validation_specification,
model_package_arn,
approval_description,
creation_time,
last_modified_time,
model_package_status,
model_package_version,
additional_inference_specifications_to_add,
model_package_status_details,
source_uri,
model_card,
security_config
FROM awscc.sagemaker.model_packages
WHERE
region = 'us-east-1' AND
Identifier = '{{ model_package_arn }}';

INSERT example

Use the following StackQL query and manifest file to create a new model_package resource, using stack-deploy.

/*+ create */
INSERT INTO awscc.sagemaker.model_packages (
Tags,
AdditionalInferenceSpecifications,
CertifyForMarketplace,
ClientToken,
CustomerMetadataProperties,
Domain,
DriftCheckBaselines,
InferenceSpecification,
MetadataProperties,
ModelApprovalStatus,
ModelMetrics,
ModelPackageDescription,
ModelPackageGroupName,
ModelPackageName,
SamplePayloadUrl,
SkipModelValidation,
SourceAlgorithmSpecification,
Task,
ValidationSpecification,
ApprovalDescription,
LastModifiedTime,
ModelPackageVersion,
AdditionalInferenceSpecificationsToAdd,
ModelPackageStatusDetails,
SourceUri,
ModelCard,
SecurityConfig,
region
)
SELECT
'{{ tags }}',
'{{ additional_inference_specifications }}',
'{{ certify_for_marketplace }}',
'{{ client_token }}',
'{{ customer_metadata_properties }}',
'{{ domain }}',
'{{ drift_check_baselines }}',
'{{ inference_specification }}',
'{{ metadata_properties }}',
'{{ model_approval_status }}',
'{{ model_metrics }}',
'{{ model_package_description }}',
'{{ model_package_group_name }}',
'{{ model_package_name }}',
'{{ sample_payload_url }}',
'{{ skip_model_validation }}',
'{{ source_algorithm_specification }}',
'{{ task }}',
'{{ validation_specification }}',
'{{ approval_description }}',
'{{ last_modified_time }}',
'{{ model_package_version }}',
'{{ additional_inference_specifications_to_add }}',
'{{ model_package_status_details }}',
'{{ source_uri }}',
'{{ model_card }}',
'{{ security_config }}',
'{{ region }}';

UPDATE example

Use the following StackQL query and manifest file to update a model_package resource, using stack-deploy.

/*+ update */
UPDATE awscc.sagemaker.model_packages
SET PatchDocument = string('{{ {
"Tags": tags,
"AdditionalInferenceSpecifications": additional_inference_specifications,
"CertifyForMarketplace": certify_for_marketplace,
"CustomerMetadataProperties": customer_metadata_properties,
"ModelApprovalStatus": model_approval_status,
"ModelPackageName": model_package_name,
"SkipModelValidation": skip_model_validation,
"ApprovalDescription": approval_description,
"LastModifiedTime": last_modified_time,
"ModelPackageVersion": model_package_version,
"AdditionalInferenceSpecificationsToAdd": additional_inference_specifications_to_add,
"ModelPackageStatusDetails": model_package_status_details,
"SourceUri": source_uri,
"ModelCard": model_card
} | generate_patch_document }}')
WHERE
region = '{{ region }}' AND
Identifier = '{{ model_package_arn }}';

DELETE example

/*+ delete */
DELETE FROM awscc.sagemaker.model_packages
WHERE
Identifier = '{{ model_package_arn }}' AND
region = 'us-east-1';

Permissions

To operate on the model_packages resource, the following permissions are required:

ecr:BatchGetImage,
ecr:DescribeImages,
ecr:StartImageScan,
ecr:DescribeImageScanFindings,
sagemaker:AddTags,
sagemaker:CreateModel,
sagemaker:CreateModelPackage,
sagemaker:CreateTrainingJob,
sagemaker:CreateTransformJob,
sagemaker:DescribeTransformJob,
sagemaker:DescribeModelPackage,
sagemaker:ListTags,
sagemaker:UpdateModelPackage,
iam:PassRole,
s3:GetObject,
s3:ListBucket,
kms:CreateGrant,
kms:DescribeKey,
kms:GenerateDataKey,
kms:Decrypt