TrainingSpecification - Amazon SageMaker
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China.

TrainingSpecification

Defines how the algorithm is used for a training job.

Contents

MetricDefinitions

A list of MetricDefinition objects, which are used for parsing metrics generated by the algorithm.

Type: Array of MetricDefinition objects

Array Members: Minimum number of 0 items. Maximum number of 40 items.

Required: No

SupportedHyperParameters

A list of the HyperParameterSpecification objects, that define the supported hyperparameters. This is required if the algorithm supports automatic model tuning.>

Type: Array of HyperParameterSpecification objects

Array Members: Minimum number of 0 items. Maximum number of 100 items.

Required: No

SupportedTrainingInstanceTypes

A list of the instance types that this algorithm can use for training.

Type: Array of strings

Valid Values: ml.m4.xlarge | ml.m4.2xlarge | ml.m4.4xlarge | ml.m4.10xlarge | ml.m4.16xlarge | ml.g4dn.xlarge | ml.g4dn.2xlarge | ml.g4dn.4xlarge | ml.g4dn.8xlarge | ml.g4dn.12xlarge | ml.g4dn.16xlarge | ml.m5.large | ml.m5.xlarge | ml.m5.2xlarge | ml.m5.4xlarge | ml.m5.12xlarge | ml.m5.24xlarge | ml.c4.xlarge | ml.c4.2xlarge | ml.c4.4xlarge | ml.c4.8xlarge | ml.p2.xlarge | ml.p2.8xlarge | ml.p2.16xlarge | ml.p3.2xlarge | ml.p3.8xlarge | ml.p3.16xlarge | ml.p3dn.24xlarge | ml.p4d.24xlarge | ml.c5.xlarge | ml.c5.2xlarge | ml.c5.4xlarge | ml.c5.9xlarge | ml.c5.18xlarge | ml.c5n.xlarge | ml.c5n.2xlarge | ml.c5n.4xlarge | ml.c5n.9xlarge | ml.c5n.18xlarge

Required: Yes

SupportedTuningJobObjectiveMetrics

A list of the metrics that the algorithm emits that can be used as the objective metric in a hyperparameter tuning job.

Type: Array of HyperParameterTuningJobObjective objects

Required: No

SupportsDistributedTraining

Indicates whether the algorithm supports distributed training. If set to false, buyers can't request more than one instance during training.

Type: Boolean

Required: No

TrainingChannels

A list of ChannelSpecification objects, which specify the input sources to be used by the algorithm.

Type: Array of ChannelSpecification objects

Array Members: Minimum number of 1 item. Maximum number of 8 items.

Required: Yes

TrainingImage

The Amazon ECR registry path of the Docker image that contains the training algorithm.

Type: String

Length Constraints: Maximum length of 255.

Pattern: [\S]+

Required: Yes

TrainingImageDigest

An MD5 hash of the training algorithm that identifies the Docker image used for training.

Type: String

Length Constraints: Maximum length of 72.

Pattern: ^[Ss][Hh][Aa]256:[0-9a-fA-F]{64}$

Required: No

See Also

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