AWS sagemaker medium security documentation change
Summary
Updated SageMaker training code examples to use new ModelTrainer class instead of Estimator, added Networking configuration, and corrected security_group_ids to use list format
Security assessment
The change explicitly modifies security_group_ids parameter to require list format instead of string. Incorrect security group configuration could lead to network exposure vulnerabilities. This prevents misconfiguration that might result in overly permissive access to training jobs.
Diff
diff --git a/sagemaker/latest/dg/model-access-training-data-fsx.md b/sagemaker/latest/dg/model-access-training-data-fsx.md index 9e5c0cfb4..9939581b1 100644 --- a//sagemaker/latest/dg/model-access-training-data-fsx.md +++ b//sagemaker/latest/dg/model-access-training-data-fsx.md @@ -60 +60 @@ Using the SageMaker Python SDK -To properly set the Amazon FSx file system as the data source, configure the SageMaker AI estimator classes and `FileSystemInput` using the following instruction. +To properly set the Amazon FSx file system as the data source, configure the SageMaker AI ModelTrainer classes and `FileSystemInput` using the following instruction. @@ -64 +64 @@ To properly set the Amazon FSx file system as the data source, configure the Sag - from sagemaker.inputs import FileSystemInput + from sagemaker.core.inputs import FileSystemInput @@ -77 +77 @@ When you specify `directory_path`, make sure that you provide the Amazon FSx fil - 2. Configure a SageMaker AI estimator with the VPC configuration used for the Amazon FSx file system. + 2. Configure a SageMaker AI ModelTrainer with the VPC configuration used for the Amazon FSx file system. @@ -79 +79,2 @@ When you specify `directory_path`, make sure that you provide the Amazon FSx fil - from sagemaker.estimator import Estimator + from sagemaker.train import ModelTrainer + from sagemaker.train.configs import Networking @@ -81 +82 @@ When you specify `directory_path`, make sure that you provide the Amazon FSx fil - estimator = Estimator( + model_trainer = ModelTrainer( @@ -83,0 +85 @@ When you specify `directory_path`, make sure that you provide the Amazon FSx fil + networking=Networking( @@ -85 +87,2 @@ When you specify `directory_path`, make sure that you provide the Amazon FSx fil - security_group_ids="security-group-id" + security_group_ids=["security-group-id"] + ) @@ -90 +93 @@ Make sure that the IAM role for the SageMaker training job has the permissions t - 3. Launch the training job by running the estimator.fit method with the Amazon FSx file system. + 3. Launch the training job by running the ModelTrainer.train method with the Amazon FSx file system. @@ -92 +95 @@ Make sure that the IAM role for the SageMaker training job has the permissions t - estimator.fit(train_fs) + model_trainer.train(train_fs) @@ -97 +100 @@ Make sure that the IAM role for the SageMaker training job has the permissions t -To find more code examples, see [Use File Systems as Training Inputs](https://sagemaker.readthedocs.io/en/stable/overview.html#use-file-systems-as-training-inputs) in the _SageMaker Python SDK documentation_. +To find more code examples, see [Use File Systems as Training Inputs](https://sagemaker.readthedocs.io/en/stable/) in the _SageMaker Python SDK documentation_.