Package software.amazon.awssdk.services.sagemaker.model

Interface TrainingSpecification.Builder

All Superinterfaces:
Buildable, CopyableBuilder<TrainingSpecification.Builder,TrainingSpecification>, SdkBuilder<TrainingSpecification.Builder,TrainingSpecification>, SdkPojo
Enclosing class:
TrainingSpecification
public static interface TrainingSpecification.Builder extends SdkPojo, CopyableBuilder<TrainingSpecification.Builder,TrainingSpecification>

  • Method Details

    • trainingImage

      TrainingSpecification.Builder trainingImage(String trainingImage)

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

      Parameters:
      trainingImage - The Amazon ECR registry path of the Docker image that contains the training algorithm.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • trainingImageDigest

      TrainingSpecification.Builder trainingImageDigest(String trainingImageDigest)

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

      Parameters:
      trainingImageDigest - An MD5 hash of the training algorithm that identifies the Docker image used for training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedHyperParameters

      TrainingSpecification.Builder supportedHyperParameters(Collection<HyperParameterSpecification> supportedHyperParameters)

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

      Parameters:
      supportedHyperParameters - A list of the HyperParameterSpecification objects, that define the supported hyperparameters. This is required if the algorithm supports automatic model tuning.>
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedHyperParameters

      TrainingSpecification.Builder supportedHyperParameters(HyperParameterSpecification... supportedHyperParameters)

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

      Parameters:
      supportedHyperParameters - A list of the HyperParameterSpecification objects, that define the supported hyperparameters. This is required if the algorithm supports automatic model tuning.>
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedHyperParameters

      TrainingSpecification.Builder supportedHyperParameters(Consumer<HyperParameterSpecification.Builder>... supportedHyperParameters)

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

      This is a convenience method that creates an instance of the HyperParameterSpecification.Builder avoiding the need to create one manually via HyperParameterSpecification.builder().

      When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to supportedHyperParameters(List<HyperParameterSpecification>).

      Parameters:
      supportedHyperParameters - a consumer that will call methods on HyperParameterSpecification.Builder
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also:

    • supportedTrainingInstanceTypesWithStrings

      TrainingSpecification.Builder supportedTrainingInstanceTypesWithStrings(Collection<String> supportedTrainingInstanceTypes)

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

      Parameters:
      supportedTrainingInstanceTypes - A list of the instance types that this algorithm can use for training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedTrainingInstanceTypesWithStrings

      TrainingSpecification.Builder supportedTrainingInstanceTypesWithStrings(String... supportedTrainingInstanceTypes)

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

      Parameters:
      supportedTrainingInstanceTypes - A list of the instance types that this algorithm can use for training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedTrainingInstanceTypes

      TrainingSpecification.Builder supportedTrainingInstanceTypes(Collection<TrainingInstanceType> supportedTrainingInstanceTypes)

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

      Parameters:
      supportedTrainingInstanceTypes - A list of the instance types that this algorithm can use for training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedTrainingInstanceTypes

      TrainingSpecification.Builder supportedTrainingInstanceTypes(TrainingInstanceType... supportedTrainingInstanceTypes)

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

      Parameters:
      supportedTrainingInstanceTypes - A list of the instance types that this algorithm can use for training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportsDistributedTraining

      TrainingSpecification.Builder supportsDistributedTraining(Boolean supportsDistributedTraining)

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

      Parameters:
      supportsDistributedTraining - Indicates whether the algorithm supports distributed training. If set to false, buyers can't request more than one instance during training.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • metricDefinitions

      TrainingSpecification.Builder metricDefinitions(Collection<MetricDefinition> metricDefinitions)

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

      Parameters:
      metricDefinitions - A list of MetricDefinition objects, which are used for parsing metrics generated by the algorithm.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • metricDefinitions

      TrainingSpecification.Builder metricDefinitions(MetricDefinition... metricDefinitions)

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

      Parameters:
      metricDefinitions - A list of MetricDefinition objects, which are used for parsing metrics generated by the algorithm.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • metricDefinitions

      TrainingSpecification.Builder metricDefinitions(Consumer<MetricDefinition.Builder>... metricDefinitions)

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

      This is a convenience method that creates an instance of the MetricDefinition.Builder avoiding the need to create one manually via MetricDefinition.builder().

      When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to metricDefinitions(List<MetricDefinition>).

      Parameters:
      metricDefinitions - a consumer that will call methods on MetricDefinition.Builder
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also:

    • trainingChannels

      TrainingSpecification.Builder trainingChannels(Collection<ChannelSpecification> trainingChannels)

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

      Parameters:
      trainingChannels - A list of ChannelSpecification objects, which specify the input sources to be used by the algorithm.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • trainingChannels

      TrainingSpecification.Builder trainingChannels(ChannelSpecification... trainingChannels)

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

      Parameters:
      trainingChannels - A list of ChannelSpecification objects, which specify the input sources to be used by the algorithm.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • trainingChannels

      TrainingSpecification.Builder trainingChannels(Consumer<ChannelSpecification.Builder>... trainingChannels)

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

      This is a convenience method that creates an instance of the ChannelSpecification.Builder avoiding the need to create one manually via ChannelSpecification.builder().

      When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to trainingChannels(List<ChannelSpecification>).

      Parameters:
      trainingChannels - a consumer that will call methods on ChannelSpecification.Builder
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also:

    • supportedTuningJobObjectiveMetrics

      TrainingSpecification.Builder supportedTuningJobObjectiveMetrics(Collection<HyperParameterTuningJobObjective> supportedTuningJobObjectiveMetrics)

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

      Parameters:
      supportedTuningJobObjectiveMetrics - A list of the metrics that the algorithm emits that can be used as the objective metric in a hyperparameter tuning job.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedTuningJobObjectiveMetrics

      TrainingSpecification.Builder supportedTuningJobObjectiveMetrics(HyperParameterTuningJobObjective... supportedTuningJobObjectiveMetrics)

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

      Parameters:
      supportedTuningJobObjectiveMetrics - A list of the metrics that the algorithm emits that can be used as the objective metric in a hyperparameter tuning job.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • supportedTuningJobObjectiveMetrics

      TrainingSpecification.Builder supportedTuningJobObjectiveMetrics(Consumer<HyperParameterTuningJobObjective.Builder>... supportedTuningJobObjectiveMetrics)

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

      This is a convenience method that creates an instance of the HyperParameterTuningJobObjective.Builder avoiding the need to create one manually via HyperParameterTuningJobObjective.builder().

      When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to supportedTuningJobObjectiveMetrics(List<HyperParameterTuningJobObjective>).

      Parameters:
      supportedTuningJobObjectiveMetrics - a consumer that will call methods on HyperParameterTuningJobObjective.Builder
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also:

    • additionalS3DataSource

      TrainingSpecification.Builder additionalS3DataSource(AdditionalS3DataSource additionalS3DataSource)

      The additional data source used during the training job.

      Parameters:
      additionalS3DataSource - The additional data source used during the training job.
      Returns:
      Returns a reference to this object so that method calls can be chained together.

    • additionalS3DataSource

      default TrainingSpecification.Builder additionalS3DataSource(Consumer<AdditionalS3DataSource.Builder> additionalS3DataSource)

      The additional data source used during the training job.

      This is a convenience method that creates an instance of the AdditionalS3DataSource.Builder avoiding the need to create one manually via AdditionalS3DataSource.builder().

      When the Consumer completes, SdkBuilder.build() is called immediately and its result is passed to additionalS3DataSource(AdditionalS3DataSource).

      Parameters:
      additionalS3DataSource - a consumer that will call methods on AdditionalS3DataSource.Builder
      Returns:
      Returns a reference to this object so that method calls can be chained together.
      See Also: