Amazon SageMaker
Developer Guide

HyperParameterAlgorithmSpecification

Specifies which training algorithm to use for training jobs that a hyperparameter tuning job launches and the metrics to monitor.

Contents

MetricDefinitions

An array of MetricDefinition objects that specify the metrics that the algorithm emits.

Type: Array of MetricDefinition objects

Array Members: Minimum number of 0 items. Maximum number of 20 items.

Required: No

TrainingImage

The registry path of the Docker image that contains the training algorithm. For information about Docker registry paths for built-in algorithms, see Algorithms Provided by Amazon SageMaker: Common Parameters .

Type: String

Length Constraints: Maximum length of 255.

Required: Yes

TrainingInputMode

The input mode that the algorithm supports: File or Pipe. In File input mode, Amazon SageMaker downloads the training data from Amazon S3 to the storage volume that is attached to the training instance and mounts the directory to the Docker volume for the training container. In Pipe input mode, Amazon SageMaker streams data directly from Amazon S3 to the container.

If you specify File mode, make sure that you provision the storage volume that is attached to the training instance with enough capacity to accommodate the training data downloaded from Amazon S3, the model artifacts, and intermediate information.

For more information about input modes, see Algorithms.

Type: String

Valid Values: Pipe | File

Required: Yes

See Also

For more information about using this API in one of the language-specific AWS SDKs, see the following:

On this page: