Package com.pulumi.aws.sagemaker

Class TrainingJob

  • public class TrainingJob
extends com.pulumi.resources.CustomResource
    Manages an AWS SageMaker AI Training Job. ## Example Usage ### Basic Usage 
     
 package generated_program;
 
 import com.pulumi.Context;
 import com.pulumi.Pulumi;
 import com.pulumi.core.Output;
 import com.pulumi.aws.sagemaker.TrainingJob;
 import com.pulumi.aws.sagemaker.TrainingJobArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobOutputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobResourceConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobStoppingConditionArgs;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Map;
 import java.io.File;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 
 public class App {
     public static void main(String[] args) {
         Pulumi.run(App::stack);
     }
 
     public static void stack(Context ctx) {
         var example = new TrainingJob("example", TrainingJobArgs.builder()
             .algorithmSpecification(TrainingJobAlgorithmSpecificationArgs.builder()
                 .trainingInputMode("File")
                 .trainingImage(exampleAwsSagemakerPrebuiltEcrImage.registryPath())
                 .build())
             .outputDataConfig(TrainingJobOutputDataConfigArgs.builder()
                 .s3OutputPath(String.format("s3://%s/output/", exampleAwsS3Bucket.bucket()))
                 .build())
             .resourceConfig(TrainingJobResourceConfigArgs.builder()
                 .instanceType("ml.m5.large")
                 .instanceCount(1)
                 .volumeSizeInGb(30)
                 .build())
             .stoppingCondition(TrainingJobStoppingConditionArgs.builder()
                 .maxRuntimeInSeconds(3600)
                 .build())
             .trainingJobName("example")
             .roleArn(exampleAwsIamRole.arn())
             .build());
 
     }
 }
    ### With VPC Configuration 
     
 package generated_program;
 
 import com.pulumi.Context;
 import com.pulumi.Pulumi;
 import com.pulumi.core.Output;
 import com.pulumi.aws.sagemaker.TrainingJob;
 import com.pulumi.aws.sagemaker.TrainingJobArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobOutputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobResourceConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobStoppingConditionArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobVpcConfigArgs;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Map;
 import java.io.File;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 
 public class App {
     public static void main(String[] args) {
         Pulumi.run(App::stack);
     }
 
     public static void stack(Context ctx) {
         var example = new TrainingJob("example", TrainingJobArgs.builder()
             .algorithmSpecification(TrainingJobAlgorithmSpecificationArgs.builder()
                 .trainingInputMode("File")
                 .trainingImage(exampleAwsSagemakerPrebuiltEcrImage.registryPath())
                 .build())
             .outputDataConfig(TrainingJobOutputDataConfigArgs.builder()
                 .s3OutputPath(String.format("s3://%s/output/", exampleAwsS3Bucket.bucket()))
                 .build())
             .resourceConfig(TrainingJobResourceConfigArgs.builder()
                 .instanceType("ml.m5.large")
                 .instanceCount(1)
                 .volumeSizeInGb(30)
                 .build())
             .stoppingCondition(TrainingJobStoppingConditionArgs.builder()
                 .maxRuntimeInSeconds(3600)
                 .build())
             .vpcConfig(TrainingJobVpcConfigArgs.builder()
                 .securityGroupIds(exampleAwsSecurityGroup.id())
                 .subnets(exampleAwsSubnet.id())
                 .build())
             .trainingJobName("example")
             .roleArn(exampleAwsIamRole.arn())
             .build());
 
     }
 }
    ### With Input Data and Hyperparameters 
     
 package generated_program;
 
 import com.pulumi.Context;
 import com.pulumi.Pulumi;
 import com.pulumi.core.Output;
 import com.pulumi.aws.sagemaker.TrainingJob;
 import com.pulumi.aws.sagemaker.TrainingJobArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobOutputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobResourceConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobStoppingConditionArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInputDataConfigDataSourceArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInputDataConfigDataSourceS3DataSourceArgs;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Map;
 import java.io.File;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 
 public class App {
     public static void main(String[] args) {
         Pulumi.run(App::stack);
     }
 
     public static void stack(Context ctx) {
         var example = new TrainingJob("example", TrainingJobArgs.builder()
             .algorithmSpecification(TrainingJobAlgorithmSpecificationArgs.builder()
                 .trainingInputMode("File")
                 .trainingImage(exampleAwsSagemakerPrebuiltEcrImage.registryPath())
                 .enableSagemakerMetricsTimeSeries(true)
                 .build())
             .outputDataConfig(TrainingJobOutputDataConfigArgs.builder()
                 .s3OutputPath(String.format("s3://%s/output/", exampleAwsS3Bucket.bucket()))
                 .build())
             .resourceConfig(TrainingJobResourceConfigArgs.builder()
                 .instanceType("ml.m5.large")
                 .instanceCount(1)
                 .volumeSizeInGb(30)
                 .build())
             .stoppingCondition(TrainingJobStoppingConditionArgs.builder()
                 .maxRuntimeInSeconds(3600)
                 .build())
             .inputDataConfigs(TrainingJobInputDataConfigArgs.builder()
                 .dataSource(TrainingJobInputDataConfigDataSourceArgs.builder()
                     .s3DataSource(TrainingJobInputDataConfigDataSourceS3DataSourceArgs.builder()
                         .s3DataType("S3Prefix")
                         .s3Uri(String.format("s3://%s/train/", exampleAwsS3Bucket.bucket()))
                         .build())
                     .build())
                 .channelName("train")
                 .build())
             .trainingJobName("example")
             .roleArn(exampleAwsIamRole.arn())
             .hyperParameters(Map.ofEntries(
                 Map.entry("mini_batch_size", "200"),
                 Map.entry("epochs", "10")
             ))
             .build());
 
     }
 }
    ### With Encrypted Output, Checkpoints, and TensorBoard 
     
 package generated_program;
 
 import com.pulumi.Context;
 import com.pulumi.Pulumi;
 import com.pulumi.core.Output;
 import com.pulumi.aws.sagemaker.TrainingJob;
 import com.pulumi.aws.sagemaker.TrainingJobArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobCheckpointConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobOutputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobResourceConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobStoppingConditionArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobTensorBoardOutputConfigArgs;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Map;
 import java.io.File;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 
 public class App {
     public static void main(String[] args) {
         Pulumi.run(App::stack);
     }
 
     public static void stack(Context ctx) {
         var example = new TrainingJob("example", TrainingJobArgs.builder()
             .algorithmSpecification(TrainingJobAlgorithmSpecificationArgs.builder()
                 .trainingInputMode("File")
                 .trainingImage(exampleAwsSagemakerPrebuiltEcrImage.registryPath())
                 .build())
             .checkpointConfig(TrainingJobCheckpointConfigArgs.builder()
                 .localPath("/opt/ml/checkpoints")
                 .s3Uri(String.format("s3://%s/checkpoints/", exampleAwsS3Bucket.bucket()))
                 .build())
             .outputDataConfig(TrainingJobOutputDataConfigArgs.builder()
                 .compressionType("GZIP")
                 .kmsKeyId(exampleAwsKmsKey.arn())
                 .s3OutputPath(String.format("s3://%s/output/", exampleAwsS3Bucket.bucket()))
                 .build())
             .resourceConfig(TrainingJobResourceConfigArgs.builder()
                 .instanceType("ml.m5.large")
                 .instanceCount(1)
                 .volumeSizeInGb(30)
                 .volumeKmsKeyId(exampleAwsKmsKey.arn())
                 .build())
             .stoppingCondition(TrainingJobStoppingConditionArgs.builder()
                 .maxRuntimeInSeconds(3600)
                 .build())
             .tensorBoardOutputConfig(TrainingJobTensorBoardOutputConfigArgs.builder()
                 .localPath("/opt/ml/output/tensorboard")
                 .s3OutputPath(String.format("s3://%s/tensorboard/", exampleAwsS3Bucket.bucket()))
                 .build())
             .trainingJobName("example")
             .roleArn(exampleAwsIamRole.arn())
             .build());
 
     }
 }
    ### With Managed Spot Training and Custom Metrics 
     
 package generated_program;
 
 import com.pulumi.Context;
 import com.pulumi.Pulumi;
 import com.pulumi.core.Output;
 import com.pulumi.aws.sagemaker.TrainingJob;
 import com.pulumi.aws.sagemaker.TrainingJobArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationMetricDefinitionArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobOutputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobResourceConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobRetryStrategyArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobStoppingConditionArgs;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Map;
 import java.io.File;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 
 public class App {
     public static void main(String[] args) {
         Pulumi.run(App::stack);
     }
 
     public static void stack(Context ctx) {
         var example = new TrainingJob("example", TrainingJobArgs.builder()
             .algorithmSpecification(TrainingJobAlgorithmSpecificationArgs.builder()
                 .metricDefinitions(                
                     TrainingJobAlgorithmSpecificationMetricDefinitionArgs.builder()
                         .name("train:loss")
                         .regex("loss: ([0-9\\.]+)")
                         .build(),
                     TrainingJobAlgorithmSpecificationMetricDefinitionArgs.builder()
                         .name("validation:accuracy")
                         .regex("accuracy: ([0-9\\.]+)")
                         .build())
                 .trainingInputMode("File")
                 .trainingImage(trainingImage)
                 .containerEntrypoints(                
                     "python",
                     "/opt/ml/code/train.py")
                 .containerArguments(                
                     "--epochs",
                     "10",
                     "--batch-size",
                     "128")
                 .build())
             .outputDataConfig(TrainingJobOutputDataConfigArgs.builder()
                 .s3OutputPath(String.format("s3://%s/output/", exampleAwsS3Bucket.bucket()))
                 .build())
             .resourceConfig(TrainingJobResourceConfigArgs.builder()
                 .instanceType("ml.m5.xlarge")
                 .instanceCount(1)
                 .volumeSizeInGb(50)
                 .keepAlivePeriodInSeconds(600)
                 .build())
             .retryStrategy(TrainingJobRetryStrategyArgs.builder()
                 .maximumRetryAttempts(3)
                 .build())
             .stoppingCondition(TrainingJobStoppingConditionArgs.builder()
                 .maxRuntimeInSeconds(3600)
                 .maxWaitTimeInSeconds(7200)
                 .build())
             .trainingJobName("example")
             .roleArn(exampleAwsIamRole.arn())
             .enableManagedSpotTraining(true)
             .enableNetworkIsolation(true)
             .enableInterContainerTrafficEncryption(true)
             .environment(Map.ofEntries(
                 Map.entry("MODEL_DIR", "/opt/ml/model"),
                 Map.entry("SM_LOG_LEVEL", "20")
             ))
             .hyperParameters(Map.ofEntries(
                 Map.entry("epochs", "10"),
                 Map.entry("batch_size", "128")
             ))
             .tags(Map.ofEntries(
                 Map.entry("Environment", "test"),
                 Map.entry("Workload", "training")
             ))
             .build());
 
     }
 }
    ### With Multiple Input Channels, Infrastructure Checks, and Session Tag Chaining 
     
 package generated_program;
 
 import com.pulumi.Context;
 import com.pulumi.Pulumi;
 import com.pulumi.core.Output;
 import com.pulumi.aws.sagemaker.TrainingJob;
 import com.pulumi.aws.sagemaker.TrainingJobArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobAlgorithmSpecificationArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInfraCheckConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobOutputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobResourceConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobSessionChainingConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobStoppingConditionArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInputDataConfigArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInputDataConfigDataSourceArgs;
 import com.pulumi.aws.sagemaker.inputs.TrainingJobInputDataConfigDataSourceS3DataSourceArgs;
 import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Map;
 import java.io.File;
 import java.nio.file.Files;
 import java.nio.file.Paths;
 
 public class App {
     public static void main(String[] args) {
         Pulumi.run(App::stack);
     }
 
     public static void stack(Context ctx) {
         var example = new TrainingJob("example", TrainingJobArgs.builder()
             .algorithmSpecification(TrainingJobAlgorithmSpecificationArgs.builder()
                 .trainingInputMode("File")
                 .trainingImage(exampleAwsSagemakerPrebuiltEcrImage.registryPath())
                 .build())
             .infraCheckConfig(TrainingJobInfraCheckConfigArgs.builder()
                 .enableInfraCheck(true)
                 .build())
             .outputDataConfig(TrainingJobOutputDataConfigArgs.builder()
                 .s3OutputPath(String.format("s3://%s/output/", exampleAwsS3Bucket.bucket()))
                 .build())
             .resourceConfig(TrainingJobResourceConfigArgs.builder()
                 .instanceType("ml.m5.large")
                 .instanceCount(1)
                 .volumeSizeInGb(30)
                 .build())
             .sessionChainingConfig(TrainingJobSessionChainingConfigArgs.builder()
                 .enableSessionTagChaining(true)
                 .build())
             .stoppingCondition(TrainingJobStoppingConditionArgs.builder()
                 .maxRuntimeInSeconds(3600)
                 .build())
             .inputDataConfigs(            
                 TrainingJobInputDataConfigArgs.builder()
                     .dataSource(TrainingJobInputDataConfigDataSourceArgs.builder()
                         .s3DataSource(TrainingJobInputDataConfigDataSourceS3DataSourceArgs.builder()
                             .s3DataDistributionType("FullyReplicated")
                             .s3DataType("S3Prefix")
                             .s3Uri(String.format("s3://%s/train/", exampleAwsS3Bucket.bucket()))
                             .build())
                         .build())
                     .channelName("train")
                     .contentType("text/csv")
                     .inputMode("File")
                     .build(),
                 TrainingJobInputDataConfigArgs.builder()
                     .dataSource(TrainingJobInputDataConfigDataSourceArgs.builder()
                         .s3DataSource(TrainingJobInputDataConfigDataSourceS3DataSourceArgs.builder()
                             .s3DataDistributionType("FullyReplicated")
                             .s3DataType("S3Prefix")
                             .s3Uri(String.format("s3://%s/validation/", exampleAwsS3Bucket.bucket()))
                             .build())
                         .build())
                     .channelName("validation")
                     .contentType("text/csv")
                     .inputMode("File")
                     .build())
             .trainingJobName("example")
             .roleArn(exampleAwsIamRole.arn())
             .build());
 
     }
 }
    ## Import ### Identity Schema #### Required * `trainingJobName` - (String) Name of the Training Job. #### Optional * `accountId` (String) AWS Account where this resource is managed. * `region` (String) Region where this resource is managed. Using `pulumi import`, import SageMaker AI Training Job using the `trainingJobName`. For example: ```sh $ pulumi import aws:sagemaker/trainingJob:TrainingJob example my-training-job ```

    • Constructor Detail

      • TrainingJob

        public TrainingJob​(java.lang.String name)
        Parameters:
        name - The _unique_ name of the resulting resource.

      • TrainingJob

        public TrainingJob​(java.lang.String name,
                   TrainingJobArgs args)
        Parameters:
        name - The _unique_ name of the resulting resource.
        args - The arguments to use to populate this resource's properties.

      • TrainingJob

        public TrainingJob​(java.lang.String name,
                   TrainingJobArgs args,
                   @Nullable
                   com.pulumi.resources.CustomResourceOptions options)
        Parameters:
        name - The _unique_ name of the resulting resource.
        args - The arguments to use to populate this resource's properties.
        options - A bag of options that control this resource's behavior.

    • Method Detail

      • algorithmSpecification

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

      • arn

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

      • checkpointConfig

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

      • debugHookConfig

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

      • debugRuleConfigurations

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

      • deleteModelPackagesOnDestroy

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

      • enableManagedSpotTraining

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

      • hyperParameters

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

      • infraCheckConfig

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

      • inputDataConfigs

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

      • mlflowConfig

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

      • modelPackageConfig

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

      • outputDataConfig

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

      • profilerConfig

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

      • profilerRuleConfigurations

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

      • region

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

      • resourceConfig

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

      • retryStrategy

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

      • roleArn

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

      • tags

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

      • trainingJobName

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

      • get

        public static TrainingJob get​(java.lang.String name,
                              com.pulumi.core.Output<java.lang.String> id,
                              @Nullable
                              TrainingJobState state,
                              @Nullable
                              com.pulumi.resources.CustomResourceOptions options)
        Get an existing Host resource's state with the given name, ID, and optional extra properties used to qualify the lookup.
        Parameters:
        name - The _unique_ name of the resulting resource.
        id - The _unique_ provider ID of the resource to lookup.
        state -
        options - Optional settings to control the behavior of the CustomResource.