TrainingJob.hs

{-# LANGUAGE DeriveGeneric #-}
{-# LANGUAGE DuplicateRecordFields #-}
{-# LANGUAGE NamedFieldPuns #-}
{-# LANGUAGE OverloadedStrings #-}
{-# LANGUAGE RecordWildCards #-}
{-# LANGUAGE StrictData #-}
{-# LANGUAGE NoImplicitPrelude #-}
{-# OPTIONS_GHC -fno-warn-unused-imports #-}
{-# OPTIONS_GHC -fno-warn-unused-matches #-}

-- Derived from AWS service descriptions, licensed under Apache 2.0.

-- |
-- Module      : Amazonka.SageMaker.Types.TrainingJob
-- Copyright   : (c) 2013-2023 Brendan Hay
-- License     : Mozilla Public License, v. 2.0.
-- Maintainer  : Brendan Hay
-- Stability   : auto-generated
-- Portability : non-portable (GHC extensions)
module Amazonka.SageMaker.Types.TrainingJob where

import qualified Amazonka.Core as Core
import qualified Amazonka.Core.Lens.Internal as Lens
import qualified Amazonka.Data as Data
import qualified Amazonka.Prelude as Prelude
import Amazonka.SageMaker.Types.AlgorithmSpecification
import Amazonka.SageMaker.Types.Channel
import Amazonka.SageMaker.Types.CheckpointConfig
import Amazonka.SageMaker.Types.DebugHookConfig
import Amazonka.SageMaker.Types.DebugRuleConfiguration
import Amazonka.SageMaker.Types.DebugRuleEvaluationStatus
import Amazonka.SageMaker.Types.ExperimentConfig
import Amazonka.SageMaker.Types.MetricData
import Amazonka.SageMaker.Types.ModelArtifacts
import Amazonka.SageMaker.Types.OutputDataConfig
import Amazonka.SageMaker.Types.ResourceConfig
import Amazonka.SageMaker.Types.RetryStrategy
import Amazonka.SageMaker.Types.SecondaryStatus
import Amazonka.SageMaker.Types.SecondaryStatusTransition
import Amazonka.SageMaker.Types.StoppingCondition
import Amazonka.SageMaker.Types.Tag
import Amazonka.SageMaker.Types.TensorBoardOutputConfig
import Amazonka.SageMaker.Types.TrainingJobStatus
import Amazonka.SageMaker.Types.VpcConfig

-- | Contains information about a training job.
--
-- /See:/ 'newTrainingJob' smart constructor.
data TrainingJob = TrainingJob'
  { -- | Information about the algorithm used for training, and algorithm
    -- metadata.
    algorithmSpecification :: Prelude.Maybe AlgorithmSpecification,
    -- | The Amazon Resource Name (ARN) of the job.
    autoMLJobArn :: Prelude.Maybe Prelude.Text,
    -- | The billable time in seconds.
    billableTimeInSeconds :: Prelude.Maybe Prelude.Natural,
    checkpointConfig :: Prelude.Maybe CheckpointConfig,
    -- | A timestamp that indicates when the training job was created.
    creationTime :: Prelude.Maybe Data.POSIX,
    debugHookConfig :: Prelude.Maybe DebugHookConfig,
    -- | Information about the debug rule configuration.
    debugRuleConfigurations :: Prelude.Maybe [DebugRuleConfiguration],
    -- | Information about the evaluation status of the rules for the training
    -- job.
    debugRuleEvaluationStatuses :: Prelude.Maybe [DebugRuleEvaluationStatus],
    -- | To encrypt all communications between ML compute instances in
    -- distributed training, choose @True@. Encryption provides greater
    -- security for distributed training, but training might take longer. How
    -- long it takes depends on the amount of communication between compute
    -- instances, especially if you use a deep learning algorithm in
    -- distributed training.
    enableInterContainerTrafficEncryption :: Prelude.Maybe Prelude.Bool,
    -- | When true, enables managed spot training using Amazon EC2 Spot instances
    -- to run training jobs instead of on-demand instances. For more
    -- information, see
    -- <https://docs.aws.amazon.com/sagemaker/latest/dg/model-managed-spot-training.html Managed Spot Training>.
    enableManagedSpotTraining :: Prelude.Maybe Prelude.Bool,
    -- | If the @TrainingJob@ was created with network isolation, the value is
    -- set to @true@. If network isolation is enabled, nodes can\'t communicate
    -- beyond the VPC they run in.
    enableNetworkIsolation :: Prelude.Maybe Prelude.Bool,
    -- | The environment variables to set in the Docker container.
    environment :: Prelude.Maybe (Prelude.HashMap Prelude.Text Prelude.Text),
    experimentConfig :: Prelude.Maybe ExperimentConfig,
    -- | If the training job failed, the reason it failed.
    failureReason :: Prelude.Maybe Prelude.Text,
    -- | A list of final metric values that are set when the training job
    -- completes. Used only if the training job was configured to use metrics.
    finalMetricDataList :: Prelude.Maybe [MetricData],
    -- | Algorithm-specific parameters.
    hyperParameters :: Prelude.Maybe (Prelude.HashMap Prelude.Text Prelude.Text),
    -- | An array of @Channel@ objects that describes each data input channel.
    inputDataConfig :: Prelude.Maybe (Prelude.NonEmpty Channel),
    -- | The Amazon Resource Name (ARN) of the labeling job.
    labelingJobArn :: Prelude.Maybe Prelude.Text,
    -- | A timestamp that indicates when the status of the training job was last
    -- modified.
    lastModifiedTime :: Prelude.Maybe Data.POSIX,
    -- | Information about the Amazon S3 location that is configured for storing
    -- model artifacts.
    modelArtifacts :: Prelude.Maybe ModelArtifacts,
    -- | The S3 path where model artifacts that you configured when creating the
    -- job are stored. SageMaker creates subfolders for model artifacts.
    outputDataConfig :: Prelude.Maybe OutputDataConfig,
    -- | Resources, including ML compute instances and ML storage volumes, that
    -- are configured for model training.
    resourceConfig :: Prelude.Maybe ResourceConfig,
    -- | The number of times to retry the job when the job fails due to an
    -- @InternalServerError@.
    retryStrategy :: Prelude.Maybe RetryStrategy,
    -- | The Amazon Web Services Identity and Access Management (IAM) role
    -- configured for the training job.
    roleArn :: Prelude.Maybe Prelude.Text,
    -- | Provides detailed information about the state of the training job. For
    -- detailed information about the secondary status of the training job, see
    -- @StatusMessage@ under SecondaryStatusTransition.
    --
    -- SageMaker provides primary statuses and secondary statuses that apply to
    -- each of them:
    --
    -- [InProgress]
    --     -   @Starting@ - Starting the training job.
    --
    --     -   @Downloading@ - An optional stage for algorithms that support
    --         @File@ training input mode. It indicates that data is being
    --         downloaded to the ML storage volumes.
    --
    --     -   @Training@ - Training is in progress.
    --
    --     -   @Uploading@ - Training is complete and the model artifacts are
    --         being uploaded to the S3 location.
    --
    -- [Completed]
    --     -   @Completed@ - The training job has completed.
    --
    -- [Failed]
    --     -   @Failed@ - The training job has failed. The reason for the
    --         failure is returned in the @FailureReason@ field of
    --         @DescribeTrainingJobResponse@.
    --
    -- [Stopped]
    --     -   @MaxRuntimeExceeded@ - The job stopped because it exceeded the
    --         maximum allowed runtime.
    --
    --     -   @Stopped@ - The training job has stopped.
    --
    -- [Stopping]
    --     -   @Stopping@ - Stopping the training job.
    --
    -- Valid values for @SecondaryStatus@ are subject to change.
    --
    -- We no longer support the following secondary statuses:
    --
    -- -   @LaunchingMLInstances@
    --
    -- -   @PreparingTrainingStack@
    --
    -- -   @DownloadingTrainingImage@
    secondaryStatus :: Prelude.Maybe SecondaryStatus,
    -- | A history of all of the secondary statuses that the training job has
    -- transitioned through.
    secondaryStatusTransitions :: Prelude.Maybe [SecondaryStatusTransition],
    -- | Specifies a limit to how long a model training job can run. It also
    -- specifies how long a managed Spot training job has to complete. When the
    -- job reaches the time limit, SageMaker ends the training job. Use this
    -- API to cap model training costs.
    --
    -- To stop a job, SageMaker sends the algorithm the @SIGTERM@ signal, which
    -- delays job termination for 120 seconds. Algorithms can use this
    -- 120-second window to save the model artifacts, so the results of
    -- training are not lost.
    stoppingCondition :: Prelude.Maybe StoppingCondition,
    -- | An array of key-value pairs. You can use tags to categorize your Amazon
    -- Web Services resources in different ways, for example, by purpose,
    -- owner, or environment. For more information, see
    -- <https://docs.aws.amazon.com/general/latest/gr/aws_tagging.html Tagging Amazon Web Services Resources>.
    tags :: Prelude.Maybe [Tag],
    tensorBoardOutputConfig :: Prelude.Maybe TensorBoardOutputConfig,
    -- | Indicates the time when the training job ends on training instances. You
    -- are billed for the time interval between the value of
    -- @TrainingStartTime@ and this time. For successful jobs and stopped jobs,
    -- this is the time after model artifacts are uploaded. For failed jobs,
    -- this is the time when SageMaker detects a job failure.
    trainingEndTime :: Prelude.Maybe Data.POSIX,
    -- | The Amazon Resource Name (ARN) of the training job.
    trainingJobArn :: Prelude.Maybe Prelude.Text,
    -- | The name of the training job.
    trainingJobName :: Prelude.Maybe Prelude.Text,
    -- | The status of the training job.
    --
    -- Training job statuses are:
    --
    -- -   @InProgress@ - The training is in progress.
    --
    -- -   @Completed@ - The training job has completed.
    --
    -- -   @Failed@ - The training job has failed. To see the reason for the
    --     failure, see the @FailureReason@ field in the response to a
    --     @DescribeTrainingJobResponse@ call.
    --
    -- -   @Stopping@ - The training job is stopping.
    --
    -- -   @Stopped@ - The training job has stopped.
    --
    -- For more detailed information, see @SecondaryStatus@.
    trainingJobStatus :: Prelude.Maybe TrainingJobStatus,
    -- | Indicates the time when the training job starts on training instances.
    -- You are billed for the time interval between this time and the value of
    -- @TrainingEndTime@. The start time in CloudWatch Logs might be later than
    -- this time. The difference is due to the time it takes to download the
    -- training data and to the size of the training container.
    trainingStartTime :: Prelude.Maybe Data.POSIX,
    -- | The training time in seconds.
    trainingTimeInSeconds :: Prelude.Maybe Prelude.Natural,
    -- | The Amazon Resource Name (ARN) of the associated hyperparameter tuning
    -- job if the training job was launched by a hyperparameter tuning job.
    tuningJobArn :: Prelude.Maybe Prelude.Text,
    -- | A VpcConfig object that specifies the VPC that this training job has
    -- access to. For more information, see
    -- <https://docs.aws.amazon.com/sagemaker/latest/dg/train-vpc.html Protect Training Jobs by Using an Amazon Virtual Private Cloud>.
    vpcConfig :: Prelude.Maybe VpcConfig
  }
  deriving (Prelude.Eq, Prelude.Read, Prelude.Show, Prelude.Generic)

-- |
-- Create a value of 'TrainingJob' with all optional fields omitted.
--
-- Use <https://hackage.haskell.org/package/generic-lens generic-lens> or <https://hackage.haskell.org/package/optics optics> to modify other optional fields.
--
-- The following record fields are available, with the corresponding lenses provided
-- for backwards compatibility:
--
-- 'algorithmSpecification', 'trainingJob_algorithmSpecification' - Information about the algorithm used for training, and algorithm
-- metadata.
--
-- 'autoMLJobArn', 'trainingJob_autoMLJobArn' - The Amazon Resource Name (ARN) of the job.
--
-- 'billableTimeInSeconds', 'trainingJob_billableTimeInSeconds' - The billable time in seconds.
--
-- 'checkpointConfig', 'trainingJob_checkpointConfig' - Undocumented member.
--
-- 'creationTime', 'trainingJob_creationTime' - A timestamp that indicates when the training job was created.
--
-- 'debugHookConfig', 'trainingJob_debugHookConfig' - Undocumented member.
--
-- 'debugRuleConfigurations', 'trainingJob_debugRuleConfigurations' - Information about the debug rule configuration.
--
-- 'debugRuleEvaluationStatuses', 'trainingJob_debugRuleEvaluationStatuses' - Information about the evaluation status of the rules for the training
-- job.
--
-- 'enableInterContainerTrafficEncryption', 'trainingJob_enableInterContainerTrafficEncryption' - To encrypt all communications between ML compute instances in
-- distributed training, choose @True@. Encryption provides greater
-- security for distributed training, but training might take longer. How
-- long it takes depends on the amount of communication between compute
-- instances, especially if you use a deep learning algorithm in
-- distributed training.
--
-- 'enableManagedSpotTraining', 'trainingJob_enableManagedSpotTraining' - When true, enables managed spot training using Amazon EC2 Spot instances
-- to run training jobs instead of on-demand instances. For more
-- information, see
-- <https://docs.aws.amazon.com/sagemaker/latest/dg/model-managed-spot-training.html Managed Spot Training>.
--
-- 'enableNetworkIsolation', 'trainingJob_enableNetworkIsolation' - If the @TrainingJob@ was created with network isolation, the value is
-- set to @true@. If network isolation is enabled, nodes can\'t communicate
-- beyond the VPC they run in.
--
-- 'environment', 'trainingJob_environment' - The environment variables to set in the Docker container.
--
-- 'experimentConfig', 'trainingJob_experimentConfig' - Undocumented member.
--
-- 'failureReason', 'trainingJob_failureReason' - If the training job failed, the reason it failed.
--
-- 'finalMetricDataList', 'trainingJob_finalMetricDataList' - A list of final metric values that are set when the training job
-- completes. Used only if the training job was configured to use metrics.
--
-- 'hyperParameters', 'trainingJob_hyperParameters' - Algorithm-specific parameters.
--
-- 'inputDataConfig', 'trainingJob_inputDataConfig' - An array of @Channel@ objects that describes each data input channel.
--
-- 'labelingJobArn', 'trainingJob_labelingJobArn' - The Amazon Resource Name (ARN) of the labeling job.
--
-- 'lastModifiedTime', 'trainingJob_lastModifiedTime' - A timestamp that indicates when the status of the training job was last
-- modified.
--
-- 'modelArtifacts', 'trainingJob_modelArtifacts' - Information about the Amazon S3 location that is configured for storing
-- model artifacts.
--
-- 'outputDataConfig', 'trainingJob_outputDataConfig' - The S3 path where model artifacts that you configured when creating the
-- job are stored. SageMaker creates subfolders for model artifacts.
--
-- 'resourceConfig', 'trainingJob_resourceConfig' - Resources, including ML compute instances and ML storage volumes, that
-- are configured for model training.
--
-- 'retryStrategy', 'trainingJob_retryStrategy' - The number of times to retry the job when the job fails due to an
-- @InternalServerError@.
--
-- 'roleArn', 'trainingJob_roleArn' - The Amazon Web Services Identity and Access Management (IAM) role
-- configured for the training job.
--
-- 'secondaryStatus', 'trainingJob_secondaryStatus' - Provides detailed information about the state of the training job. For
-- detailed information about the secondary status of the training job, see
-- @StatusMessage@ under SecondaryStatusTransition.
--
-- SageMaker provides primary statuses and secondary statuses that apply to
-- each of them:
--
-- [InProgress]
--     -   @Starting@ - Starting the training job.
--
--     -   @Downloading@ - An optional stage for algorithms that support
--         @File@ training input mode. It indicates that data is being
--         downloaded to the ML storage volumes.
--
--     -   @Training@ - Training is in progress.
--
--     -   @Uploading@ - Training is complete and the model artifacts are
--         being uploaded to the S3 location.
--
-- [Completed]
--     -   @Completed@ - The training job has completed.
--
-- [Failed]
--     -   @Failed@ - The training job has failed. The reason for the
--         failure is returned in the @FailureReason@ field of
--         @DescribeTrainingJobResponse@.
--
-- [Stopped]
--     -   @MaxRuntimeExceeded@ - The job stopped because it exceeded the
--         maximum allowed runtime.
--
--     -   @Stopped@ - The training job has stopped.
--
-- [Stopping]
--     -   @Stopping@ - Stopping the training job.
--
-- Valid values for @SecondaryStatus@ are subject to change.
--
-- We no longer support the following secondary statuses:
--
-- -   @LaunchingMLInstances@
--
-- -   @PreparingTrainingStack@
--
-- -   @DownloadingTrainingImage@
--
-- 'secondaryStatusTransitions', 'trainingJob_secondaryStatusTransitions' - A history of all of the secondary statuses that the training job has
-- transitioned through.
--
-- 'stoppingCondition', 'trainingJob_stoppingCondition' - Specifies a limit to how long a model training job can run. It also
-- specifies how long a managed Spot training job has to complete. When the
-- job reaches the time limit, SageMaker ends the training job. Use this
-- API to cap model training costs.
--
-- To stop a job, SageMaker sends the algorithm the @SIGTERM@ signal, which
-- delays job termination for 120 seconds. Algorithms can use this
-- 120-second window to save the model artifacts, so the results of
-- training are not lost.
--
-- 'tags', 'trainingJob_tags' - An array of key-value pairs. You can use tags to categorize your Amazon
-- Web Services resources in different ways, for example, by purpose,
-- owner, or environment. For more information, see
-- <https://docs.aws.amazon.com/general/latest/gr/aws_tagging.html Tagging Amazon Web Services Resources>.
--
-- 'tensorBoardOutputConfig', 'trainingJob_tensorBoardOutputConfig' - Undocumented member.
--
-- 'trainingEndTime', 'trainingJob_trainingEndTime' - Indicates the time when the training job ends on training instances. You
-- are billed for the time interval between the value of
-- @TrainingStartTime@ and this time. For successful jobs and stopped jobs,
-- this is the time after model artifacts are uploaded. For failed jobs,
-- this is the time when SageMaker detects a job failure.
--
-- 'trainingJobArn', 'trainingJob_trainingJobArn' - The Amazon Resource Name (ARN) of the training job.
--
-- 'trainingJobName', 'trainingJob_trainingJobName' - The name of the training job.
--
-- 'trainingJobStatus', 'trainingJob_trainingJobStatus' - The status of the training job.
--
-- Training job statuses are:
--
-- -   @InProgress@ - The training is in progress.
--
-- -   @Completed@ - The training job has completed.
--
-- -   @Failed@ - The training job has failed. To see the reason for the
--     failure, see the @FailureReason@ field in the response to a
--     @DescribeTrainingJobResponse@ call.
--
-- -   @Stopping@ - The training job is stopping.
--
-- -   @Stopped@ - The training job has stopped.
--
-- For more detailed information, see @SecondaryStatus@.
--
-- 'trainingStartTime', 'trainingJob_trainingStartTime' - Indicates the time when the training job starts on training instances.
-- You are billed for the time interval between this time and the value of
-- @TrainingEndTime@. The start time in CloudWatch Logs might be later than
-- this time. The difference is due to the time it takes to download the
-- training data and to the size of the training container.
--
-- 'trainingTimeInSeconds', 'trainingJob_trainingTimeInSeconds' - The training time in seconds.
--
-- 'tuningJobArn', 'trainingJob_tuningJobArn' - The Amazon Resource Name (ARN) of the associated hyperparameter tuning
-- job if the training job was launched by a hyperparameter tuning job.
--
-- 'vpcConfig', 'trainingJob_vpcConfig' - A VpcConfig object that specifies the VPC that this training job has
-- access to. For more information, see
-- <https://docs.aws.amazon.com/sagemaker/latest/dg/train-vpc.html Protect Training Jobs by Using an Amazon Virtual Private Cloud>.
newTrainingJob ::
  TrainingJob
newTrainingJob =
  TrainingJob'
    { algorithmSpecification =
        Prelude.Nothing,
      autoMLJobArn = Prelude.Nothing,
      billableTimeInSeconds = Prelude.Nothing,
      checkpointConfig = Prelude.Nothing,
      creationTime = Prelude.Nothing,
      debugHookConfig = Prelude.Nothing,
      debugRuleConfigurations = Prelude.Nothing,
      debugRuleEvaluationStatuses = Prelude.Nothing,
      enableInterContainerTrafficEncryption =
        Prelude.Nothing,
      enableManagedSpotTraining = Prelude.Nothing,
      enableNetworkIsolation = Prelude.Nothing,
      environment = Prelude.Nothing,
      experimentConfig = Prelude.Nothing,
      failureReason = Prelude.Nothing,
      finalMetricDataList = Prelude.Nothing,
      hyperParameters = Prelude.Nothing,
      inputDataConfig = Prelude.Nothing,
      labelingJobArn = Prelude.Nothing,
      lastModifiedTime = Prelude.Nothing,
      modelArtifacts = Prelude.Nothing,
      outputDataConfig = Prelude.Nothing,
      resourceConfig = Prelude.Nothing,
      retryStrategy = Prelude.Nothing,
      roleArn = Prelude.Nothing,
      secondaryStatus = Prelude.Nothing,
      secondaryStatusTransitions = Prelude.Nothing,
      stoppingCondition = Prelude.Nothing,
      tags = Prelude.Nothing,
      tensorBoardOutputConfig = Prelude.Nothing,
      trainingEndTime = Prelude.Nothing,
      trainingJobArn = Prelude.Nothing,
      trainingJobName = Prelude.Nothing,
      trainingJobStatus = Prelude.Nothing,
      trainingStartTime = Prelude.Nothing,
      trainingTimeInSeconds = Prelude.Nothing,
      tuningJobArn = Prelude.Nothing,
      vpcConfig = Prelude.Nothing
    }

-- | Information about the algorithm used for training, and algorithm
-- metadata.
trainingJob_algorithmSpecification :: Lens.Lens' TrainingJob (Prelude.Maybe AlgorithmSpecification)
trainingJob_algorithmSpecification = Lens.lens (\TrainingJob' {algorithmSpecification} -> algorithmSpecification) (\s@TrainingJob' {} a -> s {algorithmSpecification = a} :: TrainingJob)

-- | The Amazon Resource Name (ARN) of the job.
trainingJob_autoMLJobArn :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_autoMLJobArn = Lens.lens (\TrainingJob' {autoMLJobArn} -> autoMLJobArn) (\s@TrainingJob' {} a -> s {autoMLJobArn = a} :: TrainingJob)

-- | The billable time in seconds.
trainingJob_billableTimeInSeconds :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Natural)
trainingJob_billableTimeInSeconds = Lens.lens (\TrainingJob' {billableTimeInSeconds} -> billableTimeInSeconds) (\s@TrainingJob' {} a -> s {billableTimeInSeconds = a} :: TrainingJob)

-- | Undocumented member.
trainingJob_checkpointConfig :: Lens.Lens' TrainingJob (Prelude.Maybe CheckpointConfig)
trainingJob_checkpointConfig = Lens.lens (\TrainingJob' {checkpointConfig} -> checkpointConfig) (\s@TrainingJob' {} a -> s {checkpointConfig = a} :: TrainingJob)

-- | A timestamp that indicates when the training job was created.
trainingJob_creationTime :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.UTCTime)
trainingJob_creationTime = Lens.lens (\TrainingJob' {creationTime} -> creationTime) (\s@TrainingJob' {} a -> s {creationTime = a} :: TrainingJob) Prelude.. Lens.mapping Data._Time

-- | Undocumented member.
trainingJob_debugHookConfig :: Lens.Lens' TrainingJob (Prelude.Maybe DebugHookConfig)
trainingJob_debugHookConfig = Lens.lens (\TrainingJob' {debugHookConfig} -> debugHookConfig) (\s@TrainingJob' {} a -> s {debugHookConfig = a} :: TrainingJob)

-- | Information about the debug rule configuration.
trainingJob_debugRuleConfigurations :: Lens.Lens' TrainingJob (Prelude.Maybe [DebugRuleConfiguration])
trainingJob_debugRuleConfigurations = Lens.lens (\TrainingJob' {debugRuleConfigurations} -> debugRuleConfigurations) (\s@TrainingJob' {} a -> s {debugRuleConfigurations = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | Information about the evaluation status of the rules for the training
-- job.
trainingJob_debugRuleEvaluationStatuses :: Lens.Lens' TrainingJob (Prelude.Maybe [DebugRuleEvaluationStatus])
trainingJob_debugRuleEvaluationStatuses = Lens.lens (\TrainingJob' {debugRuleEvaluationStatuses} -> debugRuleEvaluationStatuses) (\s@TrainingJob' {} a -> s {debugRuleEvaluationStatuses = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | To encrypt all communications between ML compute instances in
-- distributed training, choose @True@. Encryption provides greater
-- security for distributed training, but training might take longer. How
-- long it takes depends on the amount of communication between compute
-- instances, especially if you use a deep learning algorithm in
-- distributed training.
trainingJob_enableInterContainerTrafficEncryption :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Bool)
trainingJob_enableInterContainerTrafficEncryption = Lens.lens (\TrainingJob' {enableInterContainerTrafficEncryption} -> enableInterContainerTrafficEncryption) (\s@TrainingJob' {} a -> s {enableInterContainerTrafficEncryption = a} :: TrainingJob)

-- | When true, enables managed spot training using Amazon EC2 Spot instances
-- to run training jobs instead of on-demand instances. For more
-- information, see
-- <https://docs.aws.amazon.com/sagemaker/latest/dg/model-managed-spot-training.html Managed Spot Training>.
trainingJob_enableManagedSpotTraining :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Bool)
trainingJob_enableManagedSpotTraining = Lens.lens (\TrainingJob' {enableManagedSpotTraining} -> enableManagedSpotTraining) (\s@TrainingJob' {} a -> s {enableManagedSpotTraining = a} :: TrainingJob)

-- | If the @TrainingJob@ was created with network isolation, the value is
-- set to @true@. If network isolation is enabled, nodes can\'t communicate
-- beyond the VPC they run in.
trainingJob_enableNetworkIsolation :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Bool)
trainingJob_enableNetworkIsolation = Lens.lens (\TrainingJob' {enableNetworkIsolation} -> enableNetworkIsolation) (\s@TrainingJob' {} a -> s {enableNetworkIsolation = a} :: TrainingJob)

-- | The environment variables to set in the Docker container.
trainingJob_environment :: Lens.Lens' TrainingJob (Prelude.Maybe (Prelude.HashMap Prelude.Text Prelude.Text))
trainingJob_environment = Lens.lens (\TrainingJob' {environment} -> environment) (\s@TrainingJob' {} a -> s {environment = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | Undocumented member.
trainingJob_experimentConfig :: Lens.Lens' TrainingJob (Prelude.Maybe ExperimentConfig)
trainingJob_experimentConfig = Lens.lens (\TrainingJob' {experimentConfig} -> experimentConfig) (\s@TrainingJob' {} a -> s {experimentConfig = a} :: TrainingJob)

-- | If the training job failed, the reason it failed.
trainingJob_failureReason :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_failureReason = Lens.lens (\TrainingJob' {failureReason} -> failureReason) (\s@TrainingJob' {} a -> s {failureReason = a} :: TrainingJob)

-- | A list of final metric values that are set when the training job
-- completes. Used only if the training job was configured to use metrics.
trainingJob_finalMetricDataList :: Lens.Lens' TrainingJob (Prelude.Maybe [MetricData])
trainingJob_finalMetricDataList = Lens.lens (\TrainingJob' {finalMetricDataList} -> finalMetricDataList) (\s@TrainingJob' {} a -> s {finalMetricDataList = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | Algorithm-specific parameters.
trainingJob_hyperParameters :: Lens.Lens' TrainingJob (Prelude.Maybe (Prelude.HashMap Prelude.Text Prelude.Text))
trainingJob_hyperParameters = Lens.lens (\TrainingJob' {hyperParameters} -> hyperParameters) (\s@TrainingJob' {} a -> s {hyperParameters = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | An array of @Channel@ objects that describes each data input channel.
trainingJob_inputDataConfig :: Lens.Lens' TrainingJob (Prelude.Maybe (Prelude.NonEmpty Channel))
trainingJob_inputDataConfig = Lens.lens (\TrainingJob' {inputDataConfig} -> inputDataConfig) (\s@TrainingJob' {} a -> s {inputDataConfig = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | The Amazon Resource Name (ARN) of the labeling job.
trainingJob_labelingJobArn :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_labelingJobArn = Lens.lens (\TrainingJob' {labelingJobArn} -> labelingJobArn) (\s@TrainingJob' {} a -> s {labelingJobArn = a} :: TrainingJob)

-- | A timestamp that indicates when the status of the training job was last
-- modified.
trainingJob_lastModifiedTime :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.UTCTime)
trainingJob_lastModifiedTime = Lens.lens (\TrainingJob' {lastModifiedTime} -> lastModifiedTime) (\s@TrainingJob' {} a -> s {lastModifiedTime = a} :: TrainingJob) Prelude.. Lens.mapping Data._Time

-- | Information about the Amazon S3 location that is configured for storing
-- model artifacts.
trainingJob_modelArtifacts :: Lens.Lens' TrainingJob (Prelude.Maybe ModelArtifacts)
trainingJob_modelArtifacts = Lens.lens (\TrainingJob' {modelArtifacts} -> modelArtifacts) (\s@TrainingJob' {} a -> s {modelArtifacts = a} :: TrainingJob)

-- | The S3 path where model artifacts that you configured when creating the
-- job are stored. SageMaker creates subfolders for model artifacts.
trainingJob_outputDataConfig :: Lens.Lens' TrainingJob (Prelude.Maybe OutputDataConfig)
trainingJob_outputDataConfig = Lens.lens (\TrainingJob' {outputDataConfig} -> outputDataConfig) (\s@TrainingJob' {} a -> s {outputDataConfig = a} :: TrainingJob)

-- | Resources, including ML compute instances and ML storage volumes, that
-- are configured for model training.
trainingJob_resourceConfig :: Lens.Lens' TrainingJob (Prelude.Maybe ResourceConfig)
trainingJob_resourceConfig = Lens.lens (\TrainingJob' {resourceConfig} -> resourceConfig) (\s@TrainingJob' {} a -> s {resourceConfig = a} :: TrainingJob)

-- | The number of times to retry the job when the job fails due to an
-- @InternalServerError@.
trainingJob_retryStrategy :: Lens.Lens' TrainingJob (Prelude.Maybe RetryStrategy)
trainingJob_retryStrategy = Lens.lens (\TrainingJob' {retryStrategy} -> retryStrategy) (\s@TrainingJob' {} a -> s {retryStrategy = a} :: TrainingJob)

-- | The Amazon Web Services Identity and Access Management (IAM) role
-- configured for the training job.
trainingJob_roleArn :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_roleArn = Lens.lens (\TrainingJob' {roleArn} -> roleArn) (\s@TrainingJob' {} a -> s {roleArn = a} :: TrainingJob)

-- | Provides detailed information about the state of the training job. For
-- detailed information about the secondary status of the training job, see
-- @StatusMessage@ under SecondaryStatusTransition.
--
-- SageMaker provides primary statuses and secondary statuses that apply to
-- each of them:
--
-- [InProgress]
--     -   @Starting@ - Starting the training job.
--
--     -   @Downloading@ - An optional stage for algorithms that support
--         @File@ training input mode. It indicates that data is being
--         downloaded to the ML storage volumes.
--
--     -   @Training@ - Training is in progress.
--
--     -   @Uploading@ - Training is complete and the model artifacts are
--         being uploaded to the S3 location.
--
-- [Completed]
--     -   @Completed@ - The training job has completed.
--
-- [Failed]
--     -   @Failed@ - The training job has failed. The reason for the
--         failure is returned in the @FailureReason@ field of
--         @DescribeTrainingJobResponse@.
--
-- [Stopped]
--     -   @MaxRuntimeExceeded@ - The job stopped because it exceeded the
--         maximum allowed runtime.
--
--     -   @Stopped@ - The training job has stopped.
--
-- [Stopping]
--     -   @Stopping@ - Stopping the training job.
--
-- Valid values for @SecondaryStatus@ are subject to change.
--
-- We no longer support the following secondary statuses:
--
-- -   @LaunchingMLInstances@
--
-- -   @PreparingTrainingStack@
--
-- -   @DownloadingTrainingImage@
trainingJob_secondaryStatus :: Lens.Lens' TrainingJob (Prelude.Maybe SecondaryStatus)
trainingJob_secondaryStatus = Lens.lens (\TrainingJob' {secondaryStatus} -> secondaryStatus) (\s@TrainingJob' {} a -> s {secondaryStatus = a} :: TrainingJob)

-- | A history of all of the secondary statuses that the training job has
-- transitioned through.
trainingJob_secondaryStatusTransitions :: Lens.Lens' TrainingJob (Prelude.Maybe [SecondaryStatusTransition])
trainingJob_secondaryStatusTransitions = Lens.lens (\TrainingJob' {secondaryStatusTransitions} -> secondaryStatusTransitions) (\s@TrainingJob' {} a -> s {secondaryStatusTransitions = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | Specifies a limit to how long a model training job can run. It also
-- specifies how long a managed Spot training job has to complete. When the
-- job reaches the time limit, SageMaker ends the training job. Use this
-- API to cap model training costs.
--
-- To stop a job, SageMaker sends the algorithm the @SIGTERM@ signal, which
-- delays job termination for 120 seconds. Algorithms can use this
-- 120-second window to save the model artifacts, so the results of
-- training are not lost.
trainingJob_stoppingCondition :: Lens.Lens' TrainingJob (Prelude.Maybe StoppingCondition)
trainingJob_stoppingCondition = Lens.lens (\TrainingJob' {stoppingCondition} -> stoppingCondition) (\s@TrainingJob' {} a -> s {stoppingCondition = a} :: TrainingJob)

-- | An array of key-value pairs. You can use tags to categorize your Amazon
-- Web Services resources in different ways, for example, by purpose,
-- owner, or environment. For more information, see
-- <https://docs.aws.amazon.com/general/latest/gr/aws_tagging.html Tagging Amazon Web Services Resources>.
trainingJob_tags :: Lens.Lens' TrainingJob (Prelude.Maybe [Tag])
trainingJob_tags = Lens.lens (\TrainingJob' {tags} -> tags) (\s@TrainingJob' {} a -> s {tags = a} :: TrainingJob) Prelude.. Lens.mapping Lens.coerced

-- | Undocumented member.
trainingJob_tensorBoardOutputConfig :: Lens.Lens' TrainingJob (Prelude.Maybe TensorBoardOutputConfig)
trainingJob_tensorBoardOutputConfig = Lens.lens (\TrainingJob' {tensorBoardOutputConfig} -> tensorBoardOutputConfig) (\s@TrainingJob' {} a -> s {tensorBoardOutputConfig = a} :: TrainingJob)

-- | Indicates the time when the training job ends on training instances. You
-- are billed for the time interval between the value of
-- @TrainingStartTime@ and this time. For successful jobs and stopped jobs,
-- this is the time after model artifacts are uploaded. For failed jobs,
-- this is the time when SageMaker detects a job failure.
trainingJob_trainingEndTime :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.UTCTime)
trainingJob_trainingEndTime = Lens.lens (\TrainingJob' {trainingEndTime} -> trainingEndTime) (\s@TrainingJob' {} a -> s {trainingEndTime = a} :: TrainingJob) Prelude.. Lens.mapping Data._Time

-- | The Amazon Resource Name (ARN) of the training job.
trainingJob_trainingJobArn :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_trainingJobArn = Lens.lens (\TrainingJob' {trainingJobArn} -> trainingJobArn) (\s@TrainingJob' {} a -> s {trainingJobArn = a} :: TrainingJob)

-- | The name of the training job.
trainingJob_trainingJobName :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_trainingJobName = Lens.lens (\TrainingJob' {trainingJobName} -> trainingJobName) (\s@TrainingJob' {} a -> s {trainingJobName = a} :: TrainingJob)

-- | The status of the training job.
--
-- Training job statuses are:
--
-- -   @InProgress@ - The training is in progress.
--
-- -   @Completed@ - The training job has completed.
--
-- -   @Failed@ - The training job has failed. To see the reason for the
--     failure, see the @FailureReason@ field in the response to a
--     @DescribeTrainingJobResponse@ call.
--
-- -   @Stopping@ - The training job is stopping.
--
-- -   @Stopped@ - The training job has stopped.
--
-- For more detailed information, see @SecondaryStatus@.
trainingJob_trainingJobStatus :: Lens.Lens' TrainingJob (Prelude.Maybe TrainingJobStatus)
trainingJob_trainingJobStatus = Lens.lens (\TrainingJob' {trainingJobStatus} -> trainingJobStatus) (\s@TrainingJob' {} a -> s {trainingJobStatus = a} :: TrainingJob)

-- | Indicates the time when the training job starts on training instances.
-- You are billed for the time interval between this time and the value of
-- @TrainingEndTime@. The start time in CloudWatch Logs might be later than
-- this time. The difference is due to the time it takes to download the
-- training data and to the size of the training container.
trainingJob_trainingStartTime :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.UTCTime)
trainingJob_trainingStartTime = Lens.lens (\TrainingJob' {trainingStartTime} -> trainingStartTime) (\s@TrainingJob' {} a -> s {trainingStartTime = a} :: TrainingJob) Prelude.. Lens.mapping Data._Time

-- | The training time in seconds.
trainingJob_trainingTimeInSeconds :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Natural)
trainingJob_trainingTimeInSeconds = Lens.lens (\TrainingJob' {trainingTimeInSeconds} -> trainingTimeInSeconds) (\s@TrainingJob' {} a -> s {trainingTimeInSeconds = a} :: TrainingJob)

-- | The Amazon Resource Name (ARN) of the associated hyperparameter tuning
-- job if the training job was launched by a hyperparameter tuning job.
trainingJob_tuningJobArn :: Lens.Lens' TrainingJob (Prelude.Maybe Prelude.Text)
trainingJob_tuningJobArn = Lens.lens (\TrainingJob' {tuningJobArn} -> tuningJobArn) (\s@TrainingJob' {} a -> s {tuningJobArn = a} :: TrainingJob)

-- | A VpcConfig object that specifies the VPC that this training job has
-- access to. For more information, see
-- <https://docs.aws.amazon.com/sagemaker/latest/dg/train-vpc.html Protect Training Jobs by Using an Amazon Virtual Private Cloud>.
trainingJob_vpcConfig :: Lens.Lens' TrainingJob (Prelude.Maybe VpcConfig)
trainingJob_vpcConfig = Lens.lens (\TrainingJob' {vpcConfig} -> vpcConfig) (\s@TrainingJob' {} a -> s {vpcConfig = a} :: TrainingJob)

instance Data.FromJSON TrainingJob where
  parseJSON =
    Data.withObject
      "TrainingJob"
      ( \x ->
          TrainingJob'
            Prelude.<$> (x Data..:? "AlgorithmSpecification")
            Prelude.<*> (x Data..:? "AutoMLJobArn")
            Prelude.<*> (x Data..:? "BillableTimeInSeconds")
            Prelude.<*> (x Data..:? "CheckpointConfig")
            Prelude.<*> (x Data..:? "CreationTime")
            Prelude.<*> (x Data..:? "DebugHookConfig")
            Prelude.<*> ( x
                            Data..:? "DebugRuleConfigurations"
                            Data..!= Prelude.mempty
                        )
            Prelude.<*> ( x
                            Data..:? "DebugRuleEvaluationStatuses"
                            Data..!= Prelude.mempty
                        )
            Prelude.<*> (x Data..:? "EnableInterContainerTrafficEncryption")
            Prelude.<*> (x Data..:? "EnableManagedSpotTraining")
            Prelude.<*> (x Data..:? "EnableNetworkIsolation")
            Prelude.<*> (x Data..:? "Environment" Data..!= Prelude.mempty)
            Prelude.<*> (x Data..:? "ExperimentConfig")
            Prelude.<*> (x Data..:? "FailureReason")
            Prelude.<*> ( x
                            Data..:? "FinalMetricDataList"
                            Data..!= Prelude.mempty
                        )
            Prelude.<*> ( x
                            Data..:? "HyperParameters"
                            Data..!= Prelude.mempty
                        )
            Prelude.<*> (x Data..:? "InputDataConfig")
            Prelude.<*> (x Data..:? "LabelingJobArn")
            Prelude.<*> (x Data..:? "LastModifiedTime")
            Prelude.<*> (x Data..:? "ModelArtifacts")
            Prelude.<*> (x Data..:? "OutputDataConfig")
            Prelude.<*> (x Data..:? "ResourceConfig")
            Prelude.<*> (x Data..:? "RetryStrategy")
            Prelude.<*> (x Data..:? "RoleArn")
            Prelude.<*> (x Data..:? "SecondaryStatus")
            Prelude.<*> ( x
                            Data..:? "SecondaryStatusTransitions"
                            Data..!= Prelude.mempty
                        )
            Prelude.<*> (x Data..:? "StoppingCondition")
            Prelude.<*> (x Data..:? "Tags" Data..!= Prelude.mempty)
            Prelude.<*> (x Data..:? "TensorBoardOutputConfig")
            Prelude.<*> (x Data..:? "TrainingEndTime")
            Prelude.<*> (x Data..:? "TrainingJobArn")
            Prelude.<*> (x Data..:? "TrainingJobName")
            Prelude.<*> (x Data..:? "TrainingJobStatus")
            Prelude.<*> (x Data..:? "TrainingStartTime")
            Prelude.<*> (x Data..:? "TrainingTimeInSeconds")
            Prelude.<*> (x Data..:? "TuningJobArn")
            Prelude.<*> (x Data..:? "VpcConfig")
      )

instance Prelude.Hashable TrainingJob where
  hashWithSalt _salt TrainingJob' {..} =
    _salt
      `Prelude.hashWithSalt` algorithmSpecification
      `Prelude.hashWithSalt` autoMLJobArn
      `Prelude.hashWithSalt` billableTimeInSeconds
      `Prelude.hashWithSalt` checkpointConfig
      `Prelude.hashWithSalt` creationTime
      `Prelude.hashWithSalt` debugHookConfig
      `Prelude.hashWithSalt` debugRuleConfigurations
      `Prelude.hashWithSalt` debugRuleEvaluationStatuses
      `Prelude.hashWithSalt` enableInterContainerTrafficEncryption
      `Prelude.hashWithSalt` enableManagedSpotTraining
      `Prelude.hashWithSalt` enableNetworkIsolation
      `Prelude.hashWithSalt` environment
      `Prelude.hashWithSalt` experimentConfig
      `Prelude.hashWithSalt` failureReason
      `Prelude.hashWithSalt` finalMetricDataList
      `Prelude.hashWithSalt` hyperParameters
      `Prelude.hashWithSalt` inputDataConfig
      `Prelude.hashWithSalt` labelingJobArn
      `Prelude.hashWithSalt` lastModifiedTime
      `Prelude.hashWithSalt` modelArtifacts
      `Prelude.hashWithSalt` outputDataConfig
      `Prelude.hashWithSalt` resourceConfig
      `Prelude.hashWithSalt` retryStrategy
      `Prelude.hashWithSalt` roleArn
      `Prelude.hashWithSalt` secondaryStatus
      `Prelude.hashWithSalt` secondaryStatusTransitions
      `Prelude.hashWithSalt` stoppingCondition
      `Prelude.hashWithSalt` tags
      `Prelude.hashWithSalt` tensorBoardOutputConfig
      `Prelude.hashWithSalt` trainingEndTime
      `Prelude.hashWithSalt` trainingJobArn
      `Prelude.hashWithSalt` trainingJobName
      `Prelude.hashWithSalt` trainingJobStatus
      `Prelude.hashWithSalt` trainingStartTime
      `Prelude.hashWithSalt` trainingTimeInSeconds
      `Prelude.hashWithSalt` tuningJobArn
      `Prelude.hashWithSalt` vpcConfig

instance Prelude.NFData TrainingJob where
  rnf TrainingJob' {..} =
    Prelude.rnf algorithmSpecification
      `Prelude.seq` Prelude.rnf autoMLJobArn
      `Prelude.seq` Prelude.rnf billableTimeInSeconds
      `Prelude.seq` Prelude.rnf checkpointConfig
      `Prelude.seq` Prelude.rnf creationTime
      `Prelude.seq` Prelude.rnf debugHookConfig
      `Prelude.seq` Prelude.rnf debugRuleConfigurations
      `Prelude.seq` Prelude.rnf debugRuleEvaluationStatuses
      `Prelude.seq` Prelude.rnf enableInterContainerTrafficEncryption
      `Prelude.seq` Prelude.rnf enableManagedSpotTraining
      `Prelude.seq` Prelude.rnf enableNetworkIsolation
      `Prelude.seq` Prelude.rnf environment
      `Prelude.seq` Prelude.rnf experimentConfig
      `Prelude.seq` Prelude.rnf failureReason
      `Prelude.seq` Prelude.rnf finalMetricDataList
      `Prelude.seq` Prelude.rnf hyperParameters
      `Prelude.seq` Prelude.rnf inputDataConfig
      `Prelude.seq` Prelude.rnf labelingJobArn
      `Prelude.seq` Prelude.rnf lastModifiedTime
      `Prelude.seq` Prelude.rnf modelArtifacts
      `Prelude.seq` Prelude.rnf outputDataConfig
      `Prelude.seq` Prelude.rnf resourceConfig
      `Prelude.seq` Prelude.rnf
        retryStrategy
      `Prelude.seq` Prelude.rnf roleArn
      `Prelude.seq` Prelude.rnf
        secondaryStatus
      `Prelude.seq` Prelude.rnf
        secondaryStatusTransitions
      `Prelude.seq` Prelude.rnf
        stoppingCondition
      `Prelude.seq` Prelude.rnf
        tags
      `Prelude.seq` Prelude.rnf
        tensorBoardOutputConfig
      `Prelude.seq` Prelude.rnf
        trainingEndTime
      `Prelude.seq` Prelude.rnf
        trainingJobArn
      `Prelude.seq` Prelude.rnf
        trainingJobName
      `Prelude.seq` Prelude.rnf
        trainingJobStatus
      `Prelude.seq` Prelude.rnf
        trainingStartTime
      `Prelude.seq` Prelude.rnf
        trainingTimeInSeconds
      `Prelude.seq` Prelude.rnf
        tuningJobArn
      `Prelude.seq` Prelude.rnf
        vpcConfig