Actions, resources, and condition keys for Amazon SageMaker with MLflow
Amazon SageMaker with MLflow (service prefix: sagemaker-mlflow) provides the following
service-specific operations, resources, actions, and condition keys for use in IAM permission
policies.
References:
-
Learn how to configure this service.
-
View a list of the API operations available for this service.
-
Learn how to secure this service and its resources by using IAM permission policies.
-
View the programmatic service authorization reference
for this service.
Topics
Actions defined by Amazon SageMaker with MLflow
You can specify the following actions in the Action element of an IAM
policy statement. Use policies to grant permissions to perform an operation in AWS. When
you use an action in a policy, you usually allow or deny access to the API operation or CLI
command with the same name. However, in some cases, a single action controls access to more
than one operation. Alternatively, some operations require several different actions.
| Actions | Description | Resource types (*required) | Condition keys | Access level |
|---|---|---|---|---|
Grants permission to access the MLflow UI |
Read |
|||
Grants permission to create an MLflow experiment |
Write |
|||
Grants permission to create a new model version |
Write |
|||
Grants permission to create a registered model |
Write |
|||
Grants permission to create a new run within an experiment |
Write |
|||
Grants permission to mark an MLflow experiment for deletion |
Write |
|||
Grants permission to delete a logged model in MLflow |
Write |
|||
Grants permission to delete a tag for a logged model in MLflow |
Write |
|||
Grants permission to delete a model version |
Write |
|||
Grants permission to delete a model version tag |
Write |
|||
Grants permission to delete a registered model |
Write |
|||
Grants permission to delete a registered model alias |
Write |
|||
Grants permission to delete a registered model tag |
Write |
|||
Grants permission to mark a run for deletion |
Write |
|||
Grants permission to delete a tag on a run |
Write |
|||
Grants permission to delete a trace tag in MLflow |
Write |
|||
Grants permission to delete traces in MLflow |
Write |
|||
Grants permission to end a trace in MLflow |
Write |
|||
Grants permission to set status for a logged model in MLflow |
Write |
|||
Grants permission to get a URI to download model artifacts for a specific model version |
Read |
|||
Grants permission to get metadata for an MLflow experiment |
Read |
|||
Grants permission to get metadata for an MLflow experiment by name |
Read |
|||
Grants permission to get the latest model versions |
List |
|||
Grants permission to get a logged model in MLflow |
Read |
|||
Grants permission to get a list of all values for the specified metric for a given run |
Read |
|||
Grants permission to get a model version by model name and version |
Read |
|||
Grants permission to get model version by alias in MLflow |
Read |
|||
Grants permission to get a registered model |
Read |
|||
Grants permission to get metadata, metrics, parameters, and tags for a run |
Read |
|||
Grants permission to get information about a trace in MLflow |
Read |
|||
Grants permission to list artifacts for a run |
List |
|||
Grants permission to list artifacts for a logged model in MLflow |
List |
|||
Grants permission to log a batch of metrics, parameters, and tags for a run |
Write |
|||
Grants permission to log inputs for a run |
Write |
|||
Grants permission to log params for a logged model in MLflow |
Write |
|||
Grants permission to log a metric for a run |
Write |
|||
Grants permission to log the model associated with a run |
Write |
|||
Grants permission to log outputs, such as models, for a run in MLflow |
Write |
|||
Grants permission to log a parameter tracked during a run |
Write |
|||
Grants permission to rename a registered model |
Write |
|||
Grants permission to restore an experiment marked for deletion |
Write |
|||
Grants permission to restore a deleted run |
Write |
|||
Grants permission to search for MLflow experiments |
Read |
|||
Grants permission to search for logged models in MLflow |
Read |
|||
Grants permission to search for a model version |
Read |
|||
Grants permission to search for registered models in MLflow |
Read |
|||
Grants permission to search for runs that satisfy expressions |
Read |
|||
Grants permission to search for traces in MLflow |
Read |
|||
Grants permission to set a tag on an experiment |
Write |
|||
Grants permission to set tags for a logged model in MLflow |
Write |
|||
Grants permission to set a tag for the model version |
Write |
|||
Grants permission to set a registered model alias |
Write |
|||
Grants permission to set a tag for a registered model |
Write |
|||
Grants permission to set a tag on a run |
Write |
|||
Grants permission to set a trace tag in MLflow |
Write |
|||
Grants permission to start a trace in MLflow |
Write |
|||
Grants permission to transition a model version to a particular stage |
Write |
|||
Grants permission to update the metadata for an MLflow experiment |
Write |
|||
Grants permission to update the model version |
Write |
|||
Grants permission to update a registered model |
Write |
|||
Grants permission to update run metadata |
Write |
Resource types defined by Amazon SageMaker with MLflow
The following resource types are defined by this service and can be used in the
Resource element of IAM permission policy statements.
| Resource types | ARN | Condition keys |
|---|---|---|
arn:${Partition}:sagemaker:${Region}:${Account}:mlflow-tracking-server/${MlflowTrackingServerName} |
Condition keys for Amazon SageMaker with MLflow
Amazon SageMaker with MLflow defines the following condition keys that can be used in the
Condition element of an IAM policy.
| Condition keys | Description | Type |
|---|---|---|
Filters access by a tag key and value pair |
String |
