sagemaker:CreateTrainingJob
AWS managed IAM policies that name this action, plus the ones that allow it through a wildcard such as sagemaker:*. Allow statements written with NotAction are not expanded. This is not an IAM authorization simulation.
Policies (any)
21
Allow (Action)
12
Allow (wildcard)
9
Deny (Action)
0
NotAction
1
572 policies include at least one wildcard action string (any service). A wildcard also covers actions AWS adds later, without a new policy version. Index regenerated on every deploy.
First seen
From archive historyThis action
Already present when the archive began, on 2019-02-10
Earliest appearance in any AWS managed policy tracked here, not an official AWS launch date. Matched case-insensitively, as IAM does.
Action reference
SAR-style (unofficial)Service: Amazon SageMaker
Access level
WriteDescription
Grants permission to start a model training job. After training completes, Amazon SageMaker saves the resulting model artifacts and other optional output to an Amazon S3 location that you specify
Resource types
- reserved-capacity
- training-job*
- training-plan
Allow (Action)
- AWSBatchServiceRolePolicyForSageMaker
- AWSDeepLensServiceRolePolicy
- AWSPanoramaServiceRolePolicy
- AmazonRedshiftAllCommandsFullAccess
- AmazonSageMakerCanvasFullAccess
- AmazonSageMakerFullAccess
- AmazonSageMakerModelCustomizationCoreAccess
- AmazonSageMakerServiceCatalogProductsCloudformationServiceRolePolicy
- AmazonSageMakerServiceCatalogProductsCodeBuildServiceRolePolicy
- AmazonSageMakerServiceCatalogProductsLambdaServiceRolePolicy
- SageMakerStudioProjectRoleMachineLearningPolicy
- SageMakerStudioProjectUserRolePermissionsBoundary
Allow (wildcard)
- AdministratorAccessvia *
- AmazonDataZoneSageMakerEnvironmentRolePermissionsBoundaryvia sagemaker:*
- AWSDeepRacerServiceRolePolicyvia sagemaker:*
- AWSMcpServiceActionsFullAccessvia *
- DataScientistvia sagemaker:*
- SageMakerStudioAdminIAMDefaultExecutionPolicyvia sagemaker:*, sagemaker:*Job*
- SageMakerStudioAdminIAMPermissiveExecutionPolicyvia sagemaker:*
- SageMakerStudioUserIAMDefaultExecutionPolicyvia sagemaker:*Job*
- SageMakerStudioUserIAMPermissiveExecutionPolicyvia sagemaker:*
Deny (Action)
Thanks to Ian McKay for iam-dataset (MIT), structured data derived from the AWS Service Authorization Reference. Not maintained by AWS and not guaranteed current. IAMTrail's managed policy archive is separate.
Definitions bundle regenerated on every deploy