A container for your trained model that can be deployed for SageMaker inference. This can include inference code, artifacts, and metadata. The model package type can be one of the following.
-
Versioned model: A part of a model package group in Model Registry.
-
Unversioned model: Not part of a model package group and used in AWS Marketplace.
For more information, see CreateModelPackage.
Syntax
To declare this entity in your AWS CloudFormation template, use the following syntax:
JSON
{
"Type" : "AWS::SageMaker::ModelPackage",
"Properties" : {
"AdditionalInferenceSpecifications" : [ AdditionalInferenceSpecificationDefinition, ... ],
"AdditionalInferenceSpecificationsToAdd" : [ AdditionalInferenceSpecificationDefinition, ... ],
"ApprovalDescription" : String,
"CertifyForMarketplace" : Boolean,
"ClientToken" : String,
"CustomerMetadataProperties" : {Key: Value, ...},
"Domain" : String,
"DriftCheckBaselines" : DriftCheckBaselines,
"InferenceSpecification" : InferenceSpecification,
"LastModifiedTime" : String,
"MetadataProperties" : MetadataProperties,
"ModelApprovalStatus" : String,
"ModelCard" : ModelCard,
"ModelMetrics" : ModelMetrics,
"ModelPackageDescription" : String,
"ModelPackageGroupName" : String,
"ModelPackageName" : String,
"ModelPackageStatusDetails" : ModelPackageStatusDetails,
"ModelPackageVersion" : Integer,
"SamplePayloadUrl" : String,
"SecurityConfig" : SecurityConfig,
"SkipModelValidation" : String,
"SourceAlgorithmSpecification" : SourceAlgorithmSpecification,
"SourceUri" : String,
"Tags" : [ Tag, ... ],
"Task" : String,
"ValidationSpecification" : ValidationSpecification
}
}
YAML
Type: AWS::SageMaker::ModelPackage
Properties:
AdditionalInferenceSpecifications:
- AdditionalInferenceSpecificationDefinition
AdditionalInferenceSpecificationsToAdd:
- AdditionalInferenceSpecificationDefinition
ApprovalDescription: String
CertifyForMarketplace: Boolean
ClientToken: String
CustomerMetadataProperties:
Key: Value
Domain: String
DriftCheckBaselines:
DriftCheckBaselines
InferenceSpecification:
InferenceSpecification
LastModifiedTime: String
MetadataProperties:
MetadataProperties
ModelApprovalStatus: String
ModelCard:
ModelCard
ModelMetrics:
ModelMetrics
ModelPackageDescription: String
ModelPackageGroupName: String
ModelPackageName: String
ModelPackageStatusDetails:
ModelPackageStatusDetails
ModelPackageVersion: Integer
SamplePayloadUrl: String
SecurityConfig:
SecurityConfig
SkipModelValidation: String
SourceAlgorithmSpecification:
SourceAlgorithmSpecification
SourceUri: String
Tags:
- Tag
Task: String
ValidationSpecification:
ValidationSpecification
Properties
AdditionalInferenceSpecifications-
An array of additional Inference Specification objects.
Required: No
Type: Array of AdditionalInferenceSpecificationDefinition
Minimum:
1Maximum:
15Update requires: No interruption
AdditionalInferenceSpecificationsToAdd-
An array of additional Inference Specification objects to be added to the existing array. The total number of additional Inference Specification objects cannot exceed 15. Each additional Inference Specification object specifies artifacts based on this model package that can be used on inference endpoints. Generally used with SageMaker Neo to store the compiled artifacts.
Required: No
Type: Array of AdditionalInferenceSpecificationDefinition
Minimum:
1Maximum:
15Update requires: No interruption
ApprovalDescription-
A description provided when the model approval is set.
Required: No
Type: String
Pattern:
.*Maximum:
1024Update requires: No interruption
CertifyForMarketplace-
Whether the model package is to be certified to be listed on AWS Marketplace. For information about listing model packages on AWS Marketplace, see List Your Algorithm or Model Package on AWS Marketplace.
Required: No
Type: Boolean
Update requires: No interruption
ClientToken-
A unique token that guarantees that the call to this API is idempotent.
Required: No
Type: String
Pattern:
^[a-zA-Z0-9-]+$Minimum:
1Maximum:
36Update requires: Replacement
CustomerMetadataProperties-
The metadata properties for the model package.
Required: No
Type: Object of String
Pattern:
^([\p{L}\p{Z}\p{N}_.:\/=+\-@]*)${1,128}Minimum:
1Maximum:
128Update requires: No interruption
Domain-
The machine learning domain of your model package and its components. Common machine learning domains include computer vision and natural language processing.
Required: No
Type: String
Update requires: Replacement
DriftCheckBaselines-
Represents the drift check baselines that can be used when the model monitor is set using the model package.
Required: No
Type: DriftCheckBaselines
Update requires: Replacement
InferenceSpecification-
Defines how to perform inference generation after a training job is run.
Required: No
Type: InferenceSpecification
Update requires: Replacement
LastModifiedTime-
The last time the model package was modified.
Required: No
Type: String
Update requires: No interruption
MetadataProperties-
Metadata properties of the tracking entity, trial, or trial component.
Required: No
Type: MetadataProperties
Update requires: Replacement
ModelApprovalStatus-
The approval status of the model. This can be one of the following values.
-
APPROVED- The model is approved -
REJECTED- The model is rejected. -
PENDING_MANUAL_APPROVAL- The model is waiting for manual approval.
Required: No
Type: String
Allowed values:
Approved | Rejected | PendingManualApprovalUpdate requires: No interruption
-
ModelCard-
An Amazon SageMaker Model Card.
Required: No
Type: ModelCard
Update requires: Some interruptions
ModelMetrics-
Metrics for the model.
Required: No
Type: ModelMetrics
Update requires: Replacement
ModelPackageDescription-
The description of the model package.
Required: No
Type: String
Pattern:
[\p{L}\p{M}\p{Z}\p{S}\p{N}\p{P}]*Maximum:
1024Update requires: Replacement
ModelPackageGroupName-
The model group to which the model belongs.
Required: No
Type: String
Pattern:
(arn:aws[a-z\-]*:sagemaker:[a-z0-9\-]*:[0-9]{12}:[a-z\-]*\/)?([a-zA-Z0-9]([a-zA-Z0-9-]){0,62})(?<!-)$Minimum:
1Maximum:
170Update requires: Replacement
ModelPackageName-
The name of the model package. The name can be as follows:
-
For a versioned model, the name is automatically generated by SageMaker Model Registry and follows the format '
ModelPackageGroupName/ModelPackageVersion'. -
For an unversioned model, you must provide the name.
Required: No
Type: String
Pattern:
^[a-zA-Z0-9](-*[a-zA-Z0-9]){0,62}$Minimum:
1Maximum:
63Update requires: No interruption
-
ModelPackageStatusDetails-
Specifies the validation and image scan statuses of the model package.
Required: No
Type: ModelPackageStatusDetails
Update requires: No interruption
ModelPackageVersion-
The version number of a versioned model.
Required: No
Type: Integer
Minimum:
1Update requires: No interruption
SamplePayloadUrl-
The Amazon Simple Storage Service path where the sample payload are stored. This path must point to a single gzip compressed tar archive (.tar.gz suffix).
Required: No
Type: String
Pattern:
^(https|s3)://([^/]+)/?(.*)$Maximum:
1024Update requires: Replacement
SecurityConfigProperty description not available.
Required: No
Type: SecurityConfig
Update requires: Replacement
SkipModelValidation-
Indicates if you want to skip model validation.
Required: No
Type: String
Allowed values:
None | AllUpdate requires: No interruption
SourceAlgorithmSpecification-
A list of algorithms that were used to create a model package.
Required: No
Type: SourceAlgorithmSpecification
Update requires: Replacement
SourceUri-
The URI of the source for the model package.
Required: No
Type: String
Pattern:
[\p{L}\p{M}\p{Z}\p{N}\p{P}]{0,1024}Minimum:
0Maximum:
1024Update requires: Some interruptions
-
A list of the tags associated with the model package. For more information, see Tagging AWS resources in the AWS General Reference Guide.
Required: No
Type: Array of Tag
Maximum:
50Update requires: No interruption
Task-
The machine learning task your model package accomplishes. Common machine learning tasks include object detection and image classification.
Required: No
Type: String
Update requires: Replacement
ValidationSpecification-
Specifies batch transform jobs that SageMaker runs to validate your model package.
Required: No
Type: ValidationSpecification
Update requires: Replacement
Return values
Ref
When you pass the logical ID of this resource to the intrinsic Ref function, Ref returns the Amazon Resource Name (ARN) of the model package group.
For more information about using the Ref function, see Ref.
Fn::GetAtt
The Fn::GetAtt intrinsic function returns a value for a specified attribute of this type. The following are the available attributes and sample return values.
For more information about using the Fn::GetAtt intrinsic function, see Fn::GetAtt.
CreationTime-
The time that the model package was created.
ModelPackageArn-
The Amazon Resource Name (ARN) of the model package.
ModelPackageStatus-
The status of the model package. This can be one of the following values.
-
PENDING- The model package creation is pending. -
IN_PROGRESS- The model package is in the process of being created. -
COMPLETED- The model package was successfully created. -
FAILED- The model package creation failed. -
DELETING- The model package is in the process of being deleted.
-
