Package com.pulumi.aws.sagemaker.inputs

Class TrainingJobState

  • public final class TrainingJobState
extends com.pulumi.resources.ResourceArgs

    • Method Detail

      • algorithmSpecification

        public java.util.Optional<com.pulumi.core.Output<TrainingJobAlgorithmSpecificationArgs>> algorithmSpecification()
        Returns:
        Algorithm-related parameters of the training job. See `algorithmSpecification` below. Conflicts with `serverlessJobConfig`.

      • arn

        public java.util.Optional<com.pulumi.core.Output<java.lang.String>> arn()
        Returns:
        ARN of the Training Job.

      • checkpointConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobCheckpointConfigArgs>> checkpointConfig()
        Returns:
        Location of checkpoints during training. See `checkpointConfig` below. Conflicts with `serverlessJobConfig`.

      • debugHookConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobDebugHookConfigArgs>> debugHookConfig()
        Returns:
        Configuration for debugging rules. See `debugHookConfig` below. Conflicts with `serverlessJobConfig`.

      • debugRuleConfigurations

        public java.util.Optional<com.pulumi.core.Output<java.util.List<TrainingJobDebugRuleConfigurationArgs>>> debugRuleConfigurations()
        Returns:
        List of debug rule configurations. Maximum of 20. See `debugRuleConfigurations` below.

      • deleteModelPackagesOnDestroy

        public java.util.Optional<com.pulumi.core.Output<java.lang.Boolean>> deleteModelPackagesOnDestroy()
        Returns:
        Whether to delete model packages in the configured model package group when the training job is destroyed. Default is `false`.

      • deleteVpcEnisOnDestroy

        public java.util.Optional<com.pulumi.core.Output<java.lang.Boolean>> deleteVpcEnisOnDestroy()
        Returns:
        Whether to delete detached VPC ENIs SageMaker may leave behind when the training job is destroyed. Default is `false`.

      • enableInterContainerTrafficEncryption

        public java.util.Optional<com.pulumi.core.Output<java.lang.Boolean>> enableInterContainerTrafficEncryption()
        Returns:
        Whether to encrypt inter-container traffic. When enabled, communications between containers are encrypted.

      • enableManagedSpotTraining

        public java.util.Optional<com.pulumi.core.Output<java.lang.Boolean>> enableManagedSpotTraining()
        Returns:
        Whether to use managed spot training. Optimizes the cost of training by using Amazon EC2 Spot Instances. Conflicts with `serverlessJobConfig`.

      • enableNetworkIsolation

        public java.util.Optional<com.pulumi.core.Output<java.lang.Boolean>> enableNetworkIsolation()
        Returns:
        Whether to isolate the training container from the network. No inbound or outbound network calls can be made.

      • environment

        public java.util.Optional<com.pulumi.core.Output<java.util.Map<java.lang.String,​java.lang.String>>> environment()
        Returns:
        Map of environment variables to set in the training container. Maximum of 100 entries. Conflicts with `serverlessJobConfig`.

      • experimentConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobExperimentConfigArgs>> experimentConfig()
        Returns:
        Associates a SageMaker AI Experiment or Trial to the training job. See `experimentConfig` below. Conflicts with `serverlessJobConfig`.

      • hyperParameters

        public java.util.Optional<com.pulumi.core.Output<java.util.Map<java.lang.String,​java.lang.String>>> hyperParameters()
        Returns:
        Map of hyperparameters for the training algorithm. Maximum of 100 entries.

      • infraCheckConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobInfraCheckConfigArgs>> infraCheckConfig()
        Returns:
        Infrastructure health check configuration. See `infraCheckConfig` below.

      • inputDataConfigs

        public java.util.Optional<com.pulumi.core.Output<java.util.List<TrainingJobInputDataConfigArgs>>> inputDataConfigs()
        Returns:
        List of input data channel configurations for the training job. Maximum of 20. See `inputDataConfig` below.

      • mlflowConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobMlflowConfigArgs>> mlflowConfig()
        Returns:
        MLflow integration configuration. See `mlflowConfig` below.

      • modelPackageConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobModelPackageConfigArgs>> modelPackageConfig()
        Returns:
        Model package configuration. Requires `serverlessJobConfig`. See `modelPackageConfig` below.

      • outputDataConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobOutputDataConfigArgs>> outputDataConfig()
        Returns:
        Location of the output data from the training job. See `outputDataConfig` below. The following arguments are optional:

      • profilerConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobProfilerConfigArgs>> profilerConfig()
        Returns:
        Configuration for the profiler. See `profilerConfig` below. Conflicts with `serverlessJobConfig`.

      • profilerRuleConfigurations

        public java.util.Optional<com.pulumi.core.Output<java.util.List<TrainingJobProfilerRuleConfigurationArgs>>> profilerRuleConfigurations()
        Returns:
        List of profiler rule configurations. Maximum of 20. See `profilerRuleConfigurations` below. Conflicts with `serverlessJobConfig`.

      • region

        public java.util.Optional<com.pulumi.core.Output<java.lang.String>> region()
        Returns:
        Region where this resource will be [managed](https://docs.aws.amazon.com/general/latest/gr/rande.html#regional-endpoints). Defaults to the Region set in the provider configuration.

      • remoteDebugConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobRemoteDebugConfigArgs>> remoteDebugConfig()
        Returns:
        Configuration for remote debugging. See `remoteDebugConfig` below.

      • resourceConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobResourceConfigArgs>> resourceConfig()
        Returns:
        Resources for the training job, including compute instances and storage volumes. See `resourceConfig` below.

      • retryStrategy

        public java.util.Optional<com.pulumi.core.Output<TrainingJobRetryStrategyArgs>> retryStrategy()
        Returns:
        Number of times to retry the job if it fails. See `retryStrategy` below. Conflicts with `serverlessJobConfig`.

      • roleArn

        public java.util.Optional<com.pulumi.core.Output<java.lang.String>> roleArn()
        Returns:
        ARN of the IAM role that SageMaker AI assumes to perform tasks on your behalf during training.

      • serverlessJobConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobServerlessJobConfigArgs>> serverlessJobConfig()
        Returns:
        Configuration for serverless training jobs using foundation models. Conflicts with `algorithmSpecification`, `enableManagedSpotTraining`, `environment`, `retryStrategy`, `checkpointConfig`, `debugHookConfig`, `experimentConfig`, `profilerConfig`, `profilerRuleConfigurations`, and `tensorBoardOutputConfig`. See `serverlessJobConfig` below.

      • sessionChainingConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobSessionChainingConfigArgs>> sessionChainingConfig()
        Returns:
        Configuration for session tag chaining. See `sessionChainingConfig` below.

      • tags

        public java.util.Optional<com.pulumi.core.Output<java.util.Map<java.lang.String,​java.lang.String>>> tags()
        Returns:
        Map of tags to assign to the resource. If configured with a provider `defaultTags` configuration block present, tags with matching keys will overwrite those defined at the provider-level.

      • tagsAll

        public java.util.Optional<com.pulumi.core.Output<java.util.Map<java.lang.String,​java.lang.String>>> tagsAll()
        Returns:
        Map of tags assigned to the resource, including those inherited from the provider `defaultTags` configuration block.

      • tensorBoardOutputConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobTensorBoardOutputConfigArgs>> tensorBoardOutputConfig()
        Returns:
        Configuration for TensorBoard output. See `tensorBoardOutputConfig` below. Conflicts with `serverlessJobConfig`.

      • trainingJobName

        public java.util.Optional<com.pulumi.core.Output<java.lang.String>> trainingJobName()
        Returns:
        Name of the training job. Must be between 1 and 63 characters, start with a letter or number, and contain only letters, numbers, and hyphens.

      • vpcConfig

        public java.util.Optional<com.pulumi.core.Output<TrainingJobVpcConfigArgs>> vpcConfig()
        Returns:
        VPC configuration for the training job. See `vpcConfig` below.