ResourceConfig
Describes the resources, including machine learning (ML) compute instances and ML storage volumes, to use for model training.
Types
Properties
The number of ML compute instances to use. For distributed training, provide a value greater than 1.
The configuration of a heterogeneous cluster in JSON format.
Configuration for how training job instances are placed and allocated within UltraServers. Only applicable for UltraServer capacity.
An ordered list of ML compute instance types for the training job, in priority order. SageMaker launches the training job on the first instance type in the list that has available capacity. If capacity is insufficient, SageMaker evaluates the next instance type in the preferred list. Exactly one instance type is selected for the job.
The ML compute instance type.
The duration of time in seconds to retain configured resources in a warm pool for subsequent training jobs.
The number of instances of SelectedInstanceType that the training job launched with. The job is billed for this instance type and count. Returned by DescribeTrainingJob after an instance type is selected. This field is read-only and isn't accepted in CreateTrainingJob requests.
The instance type that SageMaker selected for the job from the provided InstancePreferences. The job is billed for this instance type and count. Returned by <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeTrainingJob.html">DescribeTrainingJob</a> after an instance type is selected. This field is read-only and isn't accepted in CreateTrainingJob requests.
The Amazon Resource Name (ARN); of the training plan to use for this resource configuration.
The Amazon Web Services KMS key that SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the training job.
The size of the ML storage volume that you want to provision.