aws-sdk-sagemaker 1.85.0 → 1.86.0

Sign up to get free protection for your applications and to get access to all the features.
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: 1a310c45462d28f9d02c0bb98f4098c776fafd38d8670c3838d9500dba494d0e
4
- data.tar.gz: 8ba796f073154abc69cb52668c6cb959bd4b564aa5c306127a18886e567562bb
3
+ metadata.gz: 70252da28bdb9fdf188757e86122657444538bbb61d8cd8ee3affc624df66fe0
4
+ data.tar.gz: 4569055d7487adf471f69404f3d0755b6b65626cb1cea5adf6e89e1f9e89254d
5
5
  SHA512:
6
- metadata.gz: 41f700775b5172af131116d4ee826d4175442dd141231e5b40cf0a7ec6062731df247ea9d4518beea4c7cd2de06c9f745589fb7de9429daca67db0fde5be8dab
7
- data.tar.gz: fa962da6c45b112dd8c940dc9ade2ddb21a14630253ca69c5b98aacce80a50bef7ff4acca8642f07c59b48be275e2065554042f135cebb6783a234b30ae66a14
6
+ metadata.gz: 295f08196ef61ed263fe7464b0040bd77c733c7c1e50e726401eecc6ccca1d16e8a944501141e1e99edc86627f38653bb71619f7fd9c5b16b1c7a79cc778c88e
7
+ data.tar.gz: 94ca0e69a233115b6e3f2a07874a974d47b2b268cbed09c9db304c36ad07e5fb329ca346a95793ea3fddd2b125fec06ddb3075e9d1b777029c3a7808ef7e4521
data/CHANGELOG.md CHANGED
@@ -1,6 +1,11 @@
1
1
  Unreleased Changes
2
2
  ------------------
3
3
 
4
+ 1.86.0 (2021-05-04)
5
+ ------------------
6
+
7
+ * Feature - Enable retrying Training and Tuning Jobs that fail with InternalServerError by setting RetryStrategy.
8
+
4
9
  1.85.0 (2021-03-30)
5
10
  ------------------
6
11
 
data/VERSION CHANGED
@@ -1 +1 @@
1
- 1.85.0
1
+ 1.86.0
@@ -49,6 +49,6 @@ require_relative 'aws-sdk-sagemaker/customizations'
49
49
  # @!group service
50
50
  module Aws::SageMaker
51
51
 
52
- GEM_VERSION = '1.85.0'
52
+ GEM_VERSION = '1.86.0'
53
53
 
54
54
  end
@@ -508,6 +508,13 @@ module Aws::SageMaker
508
508
  # artifact. For more information, see [Amazon SageMaker ML Lineage
509
509
  # Tracking][1].
510
510
  #
511
+ # <note markdown="1"> `CreateAction` can only be invoked from within an SageMaker managed
512
+ # environment. This includes SageMaker training jobs, processing jobs,
513
+ # transform jobs, and SageMaker notebooks. A call to `CreateAction` from
514
+ # outside one of these environments results in an error.
515
+ #
516
+ # </note>
517
+ #
511
518
  #
512
519
  #
513
520
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/lineage-tracking.html
@@ -952,6 +959,13 @@ module Aws::SageMaker
952
959
  # URI of a dataset and the ECR registry path of an image. For more
953
960
  # information, see [Amazon SageMaker ML Lineage Tracking][1].
954
961
  #
962
+ # <note markdown="1"> `CreateArtifact` can only be invoked from within an SageMaker managed
963
+ # environment. This includes SageMaker training jobs, processing jobs,
964
+ # transform jobs, and SageMaker notebooks. A call to `CreateArtifact`
965
+ # from outside one of these environments results in an error.
966
+ #
967
+ # </note>
968
+ #
955
969
  #
956
970
  #
957
971
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/lineage-tracking.html
@@ -1350,6 +1364,13 @@ module Aws::SageMaker
1350
1364
  # entities. Some examples are an endpoint and a model package. For more
1351
1365
  # information, see [Amazon SageMaker ML Lineage Tracking][1].
1352
1366
  #
1367
+ # <note markdown="1"> `CreateContext` can only be invoked from within an SageMaker managed
1368
+ # environment. This includes SageMaker training jobs, processing jobs,
1369
+ # transform jobs, and SageMaker notebooks. A call to `CreateContext`
1370
+ # from outside one of these environments results in an error.
1371
+ #
1372
+ # </note>
1373
+ #
1353
1374
  #
1354
1375
  #
1355
1376
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/lineage-tracking.html
@@ -1679,6 +1700,9 @@ module Aws::SageMaker
1679
1700
  # optional value. Tag keys must be unique per resource. Tags are
1680
1701
  # searchable using the `Search` API.
1681
1702
  #
1703
+ # Tags that you specify for the Domain are also added to all Apps that
1704
+ # the Domain launches.
1705
+ #
1682
1706
  # @option params [String] :app_network_access_type
1683
1707
  # Specifies the VPC used for non-EFS traffic. The default value is
1684
1708
  # `PublicInternetOnly`.
@@ -2155,10 +2179,10 @@ module Aws::SageMaker
2155
2179
  # measuring the impact of a change to one or more inputs, while keeping
2156
2180
  # the remaining inputs constant.
2157
2181
  #
2158
- # When you use Amazon SageMaker Studio or the Amazon SageMaker Python
2159
- # SDK, all experiments, trials, and trial components are automatically
2160
- # tracked, logged, and indexed. When you use the AWS SDK for Python
2161
- # (Boto), you must use the logging APIs provided by the SDK.
2182
+ # When you use SageMaker Studio or the SageMaker Python SDK, all
2183
+ # experiments, trials, and trial components are automatically tracked,
2184
+ # logged, and indexed. When you use the AWS SDK for Python (Boto), you
2185
+ # must use the logging APIs provided by the SDK.
2162
2186
  #
2163
2187
  # You can add tags to experiments, trials, trial components and then use
2164
2188
  # the Search API to search for the tags.
@@ -2736,6 +2760,9 @@ module Aws::SageMaker
2736
2760
  # s3_uri: "S3Uri", # required
2737
2761
  # local_path: "DirectoryPath",
2738
2762
  # },
2763
+ # retry_strategy: {
2764
+ # maximum_retry_attempts: 1, # required
2765
+ # },
2739
2766
  # },
2740
2767
  # training_job_definitions: [
2741
2768
  # {
@@ -2834,6 +2861,9 @@ module Aws::SageMaker
2834
2861
  # s3_uri: "S3Uri", # required
2835
2862
  # local_path: "DirectoryPath",
2836
2863
  # },
2864
+ # retry_strategy: {
2865
+ # maximum_retry_attempts: 1, # required
2866
+ # },
2837
2867
  # },
2838
2868
  # ],
2839
2869
  # warm_start_config: {
@@ -4983,8 +5013,6 @@ module Aws::SageMaker
4983
5013
  # * `OutputDataConfig` - Identifies the Amazon S3 bucket where you want
4984
5014
  # Amazon SageMaker to save the results of model training.
4985
5015
  #
4986
- #
4987
- #
4988
5016
  # * `ResourceConfig` - Identifies the resources, ML compute instances,
4989
5017
  # and ML storage volumes to deploy for model training. In distributed
4990
5018
  # training, you specify more than one instance.
@@ -5000,12 +5028,15 @@ module Aws::SageMaker
5000
5028
  #
5001
5029
  # * `StoppingCondition` - To help cap training costs, use
5002
5030
  # `MaxRuntimeInSeconds` to set a time limit for training. Use
5003
- # `MaxWaitTimeInSeconds` to specify how long you are willing to wait
5004
- # for a managed spot training job to complete.
5031
+ # `MaxWaitTimeInSeconds` to specify how long a managed spot training
5032
+ # job has to complete.
5005
5033
  #
5006
5034
  # * `Environment` - The environment variables to set in the Docker
5007
5035
  # container.
5008
5036
  #
5037
+ # * `RetryStrategy` - The number of times to retry the job when the job
5038
+ # fails due to an `InternalServerError`.
5039
+ #
5009
5040
  # For more information about Amazon SageMaker, see [How It Works][3].
5010
5041
  #
5011
5042
  #
@@ -5109,9 +5140,10 @@ module Aws::SageMaker
5109
5140
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/train-vpc.html
5110
5141
  #
5111
5142
  # @option params [required, Types::StoppingCondition] :stopping_condition
5112
- # Specifies a limit to how long a model training job can run. When the
5113
- # job reaches the time limit, Amazon SageMaker ends the training job.
5114
- # Use this API to cap model training costs.
5143
+ # Specifies a limit to how long a model training job can run. It also
5144
+ # specifies how long a managed Spot training job has to complete. When
5145
+ # the job reaches the time limit, Amazon SageMaker ends the training
5146
+ # job. Use this API to cap model training costs.
5115
5147
  #
5116
5148
  # To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
5117
5149
  # signal, which delays job termination for 120 seconds. Algorithms can
@@ -5206,6 +5238,10 @@ module Aws::SageMaker
5206
5238
  # @option params [Hash<String,String>] :environment
5207
5239
  # The environment variables to set in the Docker container.
5208
5240
  #
5241
+ # @option params [Types::RetryStrategy] :retry_strategy
5242
+ # The number of times to retry the job when the job fails due to an
5243
+ # `InternalServerError`.
5244
+ #
5209
5245
  # @return [Types::CreateTrainingJobResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
5210
5246
  #
5211
5247
  # * {Types::CreateTrainingJobResponse#training_job_arn #training_job_arn} => String
@@ -5347,6 +5383,9 @@ module Aws::SageMaker
5347
5383
  # environment: {
5348
5384
  # "TrainingEnvironmentKey" => "TrainingEnvironmentValue",
5349
5385
  # },
5386
+ # retry_strategy: {
5387
+ # maximum_retry_attempts: 1, # required
5388
+ # },
5350
5389
  # })
5351
5390
  #
5352
5391
  # @example Response structure
@@ -5572,14 +5611,14 @@ module Aws::SageMaker
5572
5611
  req.send_request(options)
5573
5612
  end
5574
5613
 
5575
- # Creates an Amazon SageMaker *trial*. A trial is a set of steps called
5576
- # *trial components* that produce a machine learning model. A trial is
5577
- # part of a single Amazon SageMaker *experiment*.
5614
+ # Creates an SageMaker *trial*. A trial is a set of steps called *trial
5615
+ # components* that produce a machine learning model. A trial is part of
5616
+ # a single SageMaker *experiment*.
5578
5617
  #
5579
- # When you use Amazon SageMaker Studio or the Amazon SageMaker Python
5580
- # SDK, all experiments, trials, and trial components are automatically
5581
- # tracked, logged, and indexed. When you use the AWS SDK for Python
5582
- # (Boto), you must use the logging APIs provided by the SDK.
5618
+ # When you use SageMaker Studio or the SageMaker Python SDK, all
5619
+ # experiments, trials, and trial components are automatically tracked,
5620
+ # logged, and indexed. When you use the AWS SDK for Python (Boto), you
5621
+ # must use the logging APIs provided by the SDK.
5583
5622
  #
5584
5623
  # You can add tags to a trial and then use the Search API to search for
5585
5624
  # the tags.
@@ -5650,19 +5689,19 @@ module Aws::SageMaker
5650
5689
  # Trial components include pre-processing jobs, training jobs, and batch
5651
5690
  # transform jobs.
5652
5691
  #
5653
- # When you use Amazon SageMaker Studio or the Amazon SageMaker Python
5654
- # SDK, all experiments, trials, and trial components are automatically
5655
- # tracked, logged, and indexed. When you use the AWS SDK for Python
5656
- # (Boto), you must use the logging APIs provided by the SDK.
5692
+ # When you use SageMaker Studio or the SageMaker Python SDK, all
5693
+ # experiments, trials, and trial components are automatically tracked,
5694
+ # logged, and indexed. When you use the AWS SDK for Python (Boto), you
5695
+ # must use the logging APIs provided by the SDK.
5657
5696
  #
5658
5697
  # You can add tags to a trial component and then use the Search API to
5659
5698
  # search for the tags.
5660
5699
  #
5661
- # <note markdown="1"> `CreateTrialComponent` can only be invoked from within an Amazon
5662
- # SageMaker managed environment. This includes Amazon SageMaker training
5663
- # jobs, processing jobs, transform jobs, and Amazon SageMaker notebooks.
5664
- # A call to `CreateTrialComponent` from outside one of these
5665
- # environments results in an error.
5700
+ # <note markdown="1"> `CreateTrialComponent` can only be invoked from within an SageMaker
5701
+ # managed environment. This includes SageMaker training jobs, processing
5702
+ # jobs, transform jobs, and SageMaker notebooks. A call to
5703
+ # `CreateTrialComponent` from outside one of these environments results
5704
+ # in an error.
5666
5705
  #
5667
5706
  # </note>
5668
5707
  #
@@ -5782,7 +5821,7 @@ module Aws::SageMaker
5782
5821
  # The ID of the associated Domain.
5783
5822
  #
5784
5823
  # @option params [required, String] :user_profile_name
5785
- # A name for the UserProfile.
5824
+ # A name for the UserProfile. This value is not case sensitive.
5786
5825
  #
5787
5826
  # @option params [String] :single_sign_on_user_identifier
5788
5827
  # A specifier for the type of value specified in SingleSignOnUserValue.
@@ -5800,6 +5839,9 @@ module Aws::SageMaker
5800
5839
  # Each tag consists of a key and an optional value. Tag keys must be
5801
5840
  # unique per resource.
5802
5841
  #
5842
+ # Tags that you specify for the User Profile are also added to all Apps
5843
+ # that the User Profile launches.
5844
+ #
5803
5845
  # @option params [Types::UserSettings] :user_settings
5804
5846
  # A collection of settings.
5805
5847
  #
@@ -6450,7 +6492,7 @@ module Aws::SageMaker
6450
6492
  req.send_request(options)
6451
6493
  end
6452
6494
 
6453
- # Deletes an Amazon SageMaker experiment. All trials associated with the
6495
+ # Deletes an SageMaker experiment. All trials associated with the
6454
6496
  # experiment must be deleted first. Use the ListTrials API to get a list
6455
6497
  # of the trials associated with the experiment.
6456
6498
  #
@@ -6845,7 +6887,10 @@ module Aws::SageMaker
6845
6887
  req.send_request(options)
6846
6888
  end
6847
6889
 
6848
- # Deletes a pipeline if there are no in-progress executions.
6890
+ # Deletes a pipeline if there are no running instances of the pipeline.
6891
+ # To delete a pipeline, you must stop all running instances of the
6892
+ # pipeline using the `StopPipelineExecution` API. When you delete a
6893
+ # pipeline, all instances of the pipeline are deleted.
6849
6894
  #
6850
6895
  # @option params [required, String] :pipeline_name
6851
6896
  # The name of the pipeline to delete.
@@ -8515,6 +8560,7 @@ module Aws::SageMaker
8515
8560
  # resp.training_job_definition.enable_managed_spot_training #=> Boolean
8516
8561
  # resp.training_job_definition.checkpoint_config.s3_uri #=> String
8517
8562
  # resp.training_job_definition.checkpoint_config.local_path #=> String
8563
+ # resp.training_job_definition.retry_strategy.maximum_retry_attempts #=> Integer
8518
8564
  # resp.training_job_definitions #=> Array
8519
8565
  # resp.training_job_definitions[0].definition_name #=> String
8520
8566
  # resp.training_job_definitions[0].tuning_objective.type #=> String, one of "Maximize", "Minimize"
@@ -8575,6 +8621,7 @@ module Aws::SageMaker
8575
8621
  # resp.training_job_definitions[0].enable_managed_spot_training #=> Boolean
8576
8622
  # resp.training_job_definitions[0].checkpoint_config.s3_uri #=> String
8577
8623
  # resp.training_job_definitions[0].checkpoint_config.local_path #=> String
8624
+ # resp.training_job_definitions[0].retry_strategy.maximum_retry_attempts #=> Integer
8578
8625
  # resp.hyper_parameter_tuning_job_status #=> String, one of "Completed", "InProgress", "Failed", "Stopped", "Stopping"
8579
8626
  # resp.creation_time #=> Time
8580
8627
  # resp.hyper_parameter_tuning_end_time #=> Time
@@ -9898,6 +9945,7 @@ module Aws::SageMaker
9898
9945
  # * {Types::DescribeTrainingJobResponse#profiler_rule_configurations #profiler_rule_configurations} => Array&lt;Types::ProfilerRuleConfiguration&gt;
9899
9946
  # * {Types::DescribeTrainingJobResponse#profiler_rule_evaluation_statuses #profiler_rule_evaluation_statuses} => Array&lt;Types::ProfilerRuleEvaluationStatus&gt;
9900
9947
  # * {Types::DescribeTrainingJobResponse#profiling_status #profiling_status} => String
9948
+ # * {Types::DescribeTrainingJobResponse#retry_strategy #retry_strategy} => Types::RetryStrategy
9901
9949
  # * {Types::DescribeTrainingJobResponse#environment #environment} => Hash&lt;String,String&gt;
9902
9950
  #
9903
9951
  # @example Request syntax with placeholder values
@@ -9915,7 +9963,7 @@ module Aws::SageMaker
9915
9963
  # resp.auto_ml_job_arn #=> String
9916
9964
  # resp.model_artifacts.s3_model_artifacts #=> String
9917
9965
  # resp.training_job_status #=> String, one of "InProgress", "Completed", "Failed", "Stopping", "Stopped"
9918
- # resp.secondary_status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating"
9966
+ # resp.secondary_status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating", "Restarting"
9919
9967
  # resp.failure_reason #=> String
9920
9968
  # resp.hyper_parameters #=> Hash
9921
9969
  # resp.hyper_parameters["HyperParameterKey"] #=> String
@@ -9960,7 +10008,7 @@ module Aws::SageMaker
9960
10008
  # resp.training_end_time #=> Time
9961
10009
  # resp.last_modified_time #=> Time
9962
10010
  # resp.secondary_status_transitions #=> Array
9963
- # resp.secondary_status_transitions[0].status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating"
10011
+ # resp.secondary_status_transitions[0].status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating", "Restarting"
9964
10012
  # resp.secondary_status_transitions[0].start_time #=> Time
9965
10013
  # resp.secondary_status_transitions[0].end_time #=> Time
9966
10014
  # resp.secondary_status_transitions[0].status_message #=> String
@@ -10023,6 +10071,7 @@ module Aws::SageMaker
10023
10071
  # resp.profiler_rule_evaluation_statuses[0].status_details #=> String
10024
10072
  # resp.profiler_rule_evaluation_statuses[0].last_modified_time #=> Time
10025
10073
  # resp.profiling_status #=> String, one of "Enabled", "Disabled"
10074
+ # resp.retry_strategy.maximum_retry_attempts #=> Integer
10026
10075
  # resp.environment #=> Hash
10027
10076
  # resp.environment["TrainingEnvironmentKey"] #=> String
10028
10077
  #
@@ -10269,7 +10318,7 @@ module Aws::SageMaker
10269
10318
  # The domain ID.
10270
10319
  #
10271
10320
  # @option params [required, String] :user_profile_name
10272
- # The user profile name.
10321
+ # The user profile name. This value is not case sensitive.
10273
10322
  #
10274
10323
  # @return [Types::DescribeUserProfileResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
10275
10324
  #
@@ -11910,7 +11959,8 @@ module Aws::SageMaker
11910
11959
  # token in the next request.
11911
11960
  #
11912
11961
  # @option params [Integer] :max_results
11913
- # The maximum number of endpoints to return in the response.
11962
+ # The maximum number of endpoints to return in the response. This value
11963
+ # defaults to 10.
11914
11964
  #
11915
11965
  # @option params [String] :name_contains
11916
11966
  # A string in endpoint names. This filter returns only endpoints whose
@@ -13979,16 +14029,17 @@ module Aws::SageMaker
13979
14029
  # <note markdown="1"> When `StatusEquals` and `MaxResults` are set at the same time, the
13980
14030
  # `MaxResults` number of training jobs are first retrieved ignoring the
13981
14031
  # `StatusEquals` parameter and then they are filtered by the
13982
- # `StatusEquals` parameter, which is returned as a response. For
13983
- # example, if `ListTrainingJobs` is invoked with the following
14032
+ # `StatusEquals` parameter, which is returned as a response.
14033
+ #
14034
+ # For example, if `ListTrainingJobs` is invoked with the following
13984
14035
  # parameters:
13985
14036
  #
13986
14037
  # `\{ ... MaxResults: 100, StatusEquals: InProgress ... \}`
13987
14038
  #
13988
- # Then, 100 trainings jobs with any status including those other than
13989
- # `InProgress` are selected first (sorted according the creation time,
13990
- # from the latest to the oldest) and those with status `InProgress` are
13991
- # returned.
14039
+ # First, 100 trainings jobs with any status, including those other than
14040
+ # `InProgress`, are selected (sorted according to the creation time,
14041
+ # from the most current to the oldest). Next, those with a status of
14042
+ # `InProgress` are returned.
13992
14043
  #
13993
14044
  # You can quickly test the API using the following AWS CLI code.
13994
14045
  #
@@ -14832,7 +14883,7 @@ module Aws::SageMaker
14832
14883
  # resp.results[0].training_job.auto_ml_job_arn #=> String
14833
14884
  # resp.results[0].training_job.model_artifacts.s3_model_artifacts #=> String
14834
14885
  # resp.results[0].training_job.training_job_status #=> String, one of "InProgress", "Completed", "Failed", "Stopping", "Stopped"
14835
- # resp.results[0].training_job.secondary_status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating"
14886
+ # resp.results[0].training_job.secondary_status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating", "Restarting"
14836
14887
  # resp.results[0].training_job.failure_reason #=> String
14837
14888
  # resp.results[0].training_job.hyper_parameters #=> Hash
14838
14889
  # resp.results[0].training_job.hyper_parameters["HyperParameterKey"] #=> String
@@ -14877,7 +14928,7 @@ module Aws::SageMaker
14877
14928
  # resp.results[0].training_job.training_end_time #=> Time
14878
14929
  # resp.results[0].training_job.last_modified_time #=> Time
14879
14930
  # resp.results[0].training_job.secondary_status_transitions #=> Array
14880
- # resp.results[0].training_job.secondary_status_transitions[0].status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating"
14931
+ # resp.results[0].training_job.secondary_status_transitions[0].status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating", "Restarting"
14881
14932
  # resp.results[0].training_job.secondary_status_transitions[0].start_time #=> Time
14882
14933
  # resp.results[0].training_job.secondary_status_transitions[0].end_time #=> Time
14883
14934
  # resp.results[0].training_job.secondary_status_transitions[0].status_message #=> String
@@ -14922,6 +14973,7 @@ module Aws::SageMaker
14922
14973
  # resp.results[0].training_job.debug_rule_evaluation_statuses[0].last_modified_time #=> Time
14923
14974
  # resp.results[0].training_job.environment #=> Hash
14924
14975
  # resp.results[0].training_job.environment["TrainingEnvironmentKey"] #=> String
14976
+ # resp.results[0].training_job.retry_strategy.maximum_retry_attempts #=> Integer
14925
14977
  # resp.results[0].training_job.tags #=> Array
14926
14978
  # resp.results[0].training_job.tags[0].key #=> String
14927
14979
  # resp.results[0].training_job.tags[0].value #=> String
@@ -15020,7 +15072,7 @@ module Aws::SageMaker
15020
15072
  # resp.results[0].trial_component.source_detail.training_job.auto_ml_job_arn #=> String
15021
15073
  # resp.results[0].trial_component.source_detail.training_job.model_artifacts.s3_model_artifacts #=> String
15022
15074
  # resp.results[0].trial_component.source_detail.training_job.training_job_status #=> String, one of "InProgress", "Completed", "Failed", "Stopping", "Stopped"
15023
- # resp.results[0].trial_component.source_detail.training_job.secondary_status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating"
15075
+ # resp.results[0].trial_component.source_detail.training_job.secondary_status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating", "Restarting"
15024
15076
  # resp.results[0].trial_component.source_detail.training_job.failure_reason #=> String
15025
15077
  # resp.results[0].trial_component.source_detail.training_job.hyper_parameters #=> Hash
15026
15078
  # resp.results[0].trial_component.source_detail.training_job.hyper_parameters["HyperParameterKey"] #=> String
@@ -15065,7 +15117,7 @@ module Aws::SageMaker
15065
15117
  # resp.results[0].trial_component.source_detail.training_job.training_end_time #=> Time
15066
15118
  # resp.results[0].trial_component.source_detail.training_job.last_modified_time #=> Time
15067
15119
  # resp.results[0].trial_component.source_detail.training_job.secondary_status_transitions #=> Array
15068
- # resp.results[0].trial_component.source_detail.training_job.secondary_status_transitions[0].status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating"
15120
+ # resp.results[0].trial_component.source_detail.training_job.secondary_status_transitions[0].status #=> String, one of "Starting", "LaunchingMLInstances", "PreparingTrainingStack", "Downloading", "DownloadingTrainingImage", "Training", "Uploading", "Stopping", "Stopped", "MaxRuntimeExceeded", "Completed", "Failed", "Interrupted", "MaxWaitTimeExceeded", "Updating", "Restarting"
15069
15121
  # resp.results[0].trial_component.source_detail.training_job.secondary_status_transitions[0].start_time #=> Time
15070
15122
  # resp.results[0].trial_component.source_detail.training_job.secondary_status_transitions[0].end_time #=> Time
15071
15123
  # resp.results[0].trial_component.source_detail.training_job.secondary_status_transitions[0].status_message #=> String
@@ -15110,6 +15162,7 @@ module Aws::SageMaker
15110
15162
  # resp.results[0].trial_component.source_detail.training_job.debug_rule_evaluation_statuses[0].last_modified_time #=> Time
15111
15163
  # resp.results[0].trial_component.source_detail.training_job.environment #=> Hash
15112
15164
  # resp.results[0].trial_component.source_detail.training_job.environment["TrainingEnvironmentKey"] #=> String
15165
+ # resp.results[0].trial_component.source_detail.training_job.retry_strategy.maximum_retry_attempts #=> Integer
15113
15166
  # resp.results[0].trial_component.source_detail.training_job.tags #=> Array
15114
15167
  # resp.results[0].trial_component.source_detail.training_job.tags[0].key #=> String
15115
15168
  # resp.results[0].trial_component.source_detail.training_job.tags[0].value #=> String
@@ -17355,7 +17408,7 @@ module Aws::SageMaker
17355
17408
  params: params,
17356
17409
  config: config)
17357
17410
  context[:gem_name] = 'aws-sdk-sagemaker'
17358
- context[:gem_version] = '1.85.0'
17411
+ context[:gem_version] = '1.86.0'
17359
17412
  Seahorse::Client::Request.new(handlers, context)
17360
17413
  end
17361
17414
 
@@ -862,6 +862,7 @@ module Aws::SageMaker
862
862
  MaxRuntimePerTrainingJobInSeconds = Shapes::IntegerShape.new(name: 'MaxRuntimePerTrainingJobInSeconds')
863
863
  MaxWaitTimeInSeconds = Shapes::IntegerShape.new(name: 'MaxWaitTimeInSeconds')
864
864
  MaximumExecutionTimeoutInSeconds = Shapes::IntegerShape.new(name: 'MaximumExecutionTimeoutInSeconds')
865
+ MaximumRetryAttempts = Shapes::IntegerShape.new(name: 'MaximumRetryAttempts')
865
866
  MediaType = Shapes::StringShape.new(name: 'MediaType')
866
867
  MemberDefinition = Shapes::StructureShape.new(name: 'MemberDefinition')
867
868
  MemberDefinitions = Shapes::ListShape.new(name: 'MemberDefinitions')
@@ -1152,6 +1153,7 @@ module Aws::SageMaker
1152
1153
  ResponseMIMETypes = Shapes::ListShape.new(name: 'ResponseMIMETypes')
1153
1154
  RetentionPolicy = Shapes::StructureShape.new(name: 'RetentionPolicy')
1154
1155
  RetentionType = Shapes::StringShape.new(name: 'RetentionType')
1156
+ RetryStrategy = Shapes::StructureShape.new(name: 'RetryStrategy')
1155
1157
  RoleArn = Shapes::StringShape.new(name: 'RoleArn')
1156
1158
  RootAccess = Shapes::StringShape.new(name: 'RootAccess')
1157
1159
  RuleConfigurationName = Shapes::StringShape.new(name: 'RuleConfigurationName')
@@ -2277,6 +2279,7 @@ module Aws::SageMaker
2277
2279
  CreateTrainingJobRequest.add_member(:profiler_config, Shapes::ShapeRef.new(shape: ProfilerConfig, location_name: "ProfilerConfig"))
2278
2280
  CreateTrainingJobRequest.add_member(:profiler_rule_configurations, Shapes::ShapeRef.new(shape: ProfilerRuleConfigurations, location_name: "ProfilerRuleConfigurations"))
2279
2281
  CreateTrainingJobRequest.add_member(:environment, Shapes::ShapeRef.new(shape: TrainingEnvironmentMap, location_name: "Environment"))
2282
+ CreateTrainingJobRequest.add_member(:retry_strategy, Shapes::ShapeRef.new(shape: RetryStrategy, location_name: "RetryStrategy"))
2280
2283
  CreateTrainingJobRequest.struct_class = Types::CreateTrainingJobRequest
2281
2284
 
2282
2285
  CreateTrainingJobResponse.add_member(:training_job_arn, Shapes::ShapeRef.new(shape: TrainingJobArn, required: true, location_name: "TrainingJobArn"))
@@ -3282,6 +3285,7 @@ module Aws::SageMaker
3282
3285
  DescribeTrainingJobResponse.add_member(:profiler_rule_configurations, Shapes::ShapeRef.new(shape: ProfilerRuleConfigurations, location_name: "ProfilerRuleConfigurations"))
3283
3286
  DescribeTrainingJobResponse.add_member(:profiler_rule_evaluation_statuses, Shapes::ShapeRef.new(shape: ProfilerRuleEvaluationStatuses, location_name: "ProfilerRuleEvaluationStatuses"))
3284
3287
  DescribeTrainingJobResponse.add_member(:profiling_status, Shapes::ShapeRef.new(shape: ProfilingStatus, location_name: "ProfilingStatus"))
3288
+ DescribeTrainingJobResponse.add_member(:retry_strategy, Shapes::ShapeRef.new(shape: RetryStrategy, location_name: "RetryStrategy"))
3285
3289
  DescribeTrainingJobResponse.add_member(:environment, Shapes::ShapeRef.new(shape: TrainingEnvironmentMap, location_name: "Environment"))
3286
3290
  DescribeTrainingJobResponse.struct_class = Types::DescribeTrainingJobResponse
3287
3291
 
@@ -3753,6 +3757,7 @@ module Aws::SageMaker
3753
3757
  HyperParameterTrainingJobDefinition.add_member(:enable_inter_container_traffic_encryption, Shapes::ShapeRef.new(shape: Boolean, location_name: "EnableInterContainerTrafficEncryption"))
3754
3758
  HyperParameterTrainingJobDefinition.add_member(:enable_managed_spot_training, Shapes::ShapeRef.new(shape: Boolean, location_name: "EnableManagedSpotTraining"))
3755
3759
  HyperParameterTrainingJobDefinition.add_member(:checkpoint_config, Shapes::ShapeRef.new(shape: CheckpointConfig, location_name: "CheckpointConfig"))
3760
+ HyperParameterTrainingJobDefinition.add_member(:retry_strategy, Shapes::ShapeRef.new(shape: RetryStrategy, location_name: "RetryStrategy"))
3756
3761
  HyperParameterTrainingJobDefinition.struct_class = Types::HyperParameterTrainingJobDefinition
3757
3762
 
3758
3763
  HyperParameterTrainingJobDefinitions.member = Shapes::ShapeRef.new(shape: HyperParameterTrainingJobDefinition)
@@ -5477,6 +5482,9 @@ module Aws::SageMaker
5477
5482
  RetentionPolicy.add_member(:home_efs_file_system, Shapes::ShapeRef.new(shape: RetentionType, location_name: "HomeEfsFileSystem"))
5478
5483
  RetentionPolicy.struct_class = Types::RetentionPolicy
5479
5484
 
5485
+ RetryStrategy.add_member(:maximum_retry_attempts, Shapes::ShapeRef.new(shape: MaximumRetryAttempts, required: true, location_name: "MaximumRetryAttempts"))
5486
+ RetryStrategy.struct_class = Types::RetryStrategy
5487
+
5480
5488
  RuleParameters.key = Shapes::ShapeRef.new(shape: ConfigKey)
5481
5489
  RuleParameters.value = Shapes::ShapeRef.new(shape: ConfigValue)
5482
5490
 
@@ -5706,6 +5714,7 @@ module Aws::SageMaker
5706
5714
  TrainingJob.add_member(:tensor_board_output_config, Shapes::ShapeRef.new(shape: TensorBoardOutputConfig, location_name: "TensorBoardOutputConfig"))
5707
5715
  TrainingJob.add_member(:debug_rule_evaluation_statuses, Shapes::ShapeRef.new(shape: DebugRuleEvaluationStatuses, location_name: "DebugRuleEvaluationStatuses"))
5708
5716
  TrainingJob.add_member(:environment, Shapes::ShapeRef.new(shape: TrainingEnvironmentMap, location_name: "Environment"))
5717
+ TrainingJob.add_member(:retry_strategy, Shapes::ShapeRef.new(shape: RetryStrategy, location_name: "RetryStrategy"))
5709
5718
  TrainingJob.add_member(:tags, Shapes::ShapeRef.new(shape: TagList, location_name: "Tags"))
5710
5719
  TrainingJob.struct_class = Types::TrainingJob
5711
5720
 
@@ -2912,8 +2912,8 @@ module Aws::SageMaker
2912
2912
  include Aws::Structure
2913
2913
  end
2914
2914
 
2915
- # There was a conflict when you attempted to modify an experiment,
2916
- # trial, or trial component.
2915
+ # There was a conflict when you attempted to modify a SageMaker entity
2916
+ # such as an `Experiment` or `Artifact`.
2917
2917
  #
2918
2918
  # @!attribute [rw] message
2919
2919
  # @return [String]
@@ -4494,6 +4494,9 @@ module Aws::SageMaker
4494
4494
  # Tags to associated with the Domain. Each tag consists of a key and
4495
4495
  # an optional value. Tag keys must be unique per resource. Tags are
4496
4496
  # searchable using the `Search` API.
4497
+ #
4498
+ # Tags that you specify for the Domain are also added to all Apps that
4499
+ # the Domain launches.
4497
4500
  # @return [Array<Types::Tag>]
4498
4501
  #
4499
4502
  # @!attribute [rw] app_network_access_type
@@ -5348,6 +5351,9 @@ module Aws::SageMaker
5348
5351
  # s3_uri: "S3Uri", # required
5349
5352
  # local_path: "DirectoryPath",
5350
5353
  # },
5354
+ # retry_strategy: {
5355
+ # maximum_retry_attempts: 1, # required
5356
+ # },
5351
5357
  # },
5352
5358
  # training_job_definitions: [
5353
5359
  # {
@@ -5446,6 +5452,9 @@ module Aws::SageMaker
5446
5452
  # s3_uri: "S3Uri", # required
5447
5453
  # local_path: "DirectoryPath",
5448
5454
  # },
5455
+ # retry_strategy: {
5456
+ # maximum_retry_attempts: 1, # required
5457
+ # },
5449
5458
  # },
5450
5459
  # ],
5451
5460
  # warm_start_config: {
@@ -7823,6 +7832,9 @@ module Aws::SageMaker
7823
7832
  # environment: {
7824
7833
  # "TrainingEnvironmentKey" => "TrainingEnvironmentValue",
7825
7834
  # },
7835
+ # retry_strategy: {
7836
+ # maximum_retry_attempts: 1, # required
7837
+ # },
7826
7838
  # }
7827
7839
  #
7828
7840
  # @!attribute [rw] training_job_name
@@ -7929,9 +7941,10 @@ module Aws::SageMaker
7929
7941
  # @return [Types::VpcConfig]
7930
7942
  #
7931
7943
  # @!attribute [rw] stopping_condition
7932
- # Specifies a limit to how long a model training job can run. When the
7933
- # job reaches the time limit, Amazon SageMaker ends the training job.
7934
- # Use this API to cap model training costs.
7944
+ # Specifies a limit to how long a model training job can run. It also
7945
+ # specifies how long a managed Spot training job has to complete. When
7946
+ # the job reaches the time limit, Amazon SageMaker ends the training
7947
+ # job. Use this API to cap model training costs.
7935
7948
  #
7936
7949
  # To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
7937
7950
  # signal, which delays job termination for 120 seconds. Algorithms can
@@ -8040,6 +8053,11 @@ module Aws::SageMaker
8040
8053
  # The environment variables to set in the Docker container.
8041
8054
  # @return [Hash<String,String>]
8042
8055
  #
8056
+ # @!attribute [rw] retry_strategy
8057
+ # The number of times to retry the job when the job fails due to an
8058
+ # `InternalServerError`.
8059
+ # @return [Types::RetryStrategy]
8060
+ #
8043
8061
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/CreateTrainingJobRequest AWS API Documentation
8044
8062
  #
8045
8063
  class CreateTrainingJobRequest < Struct.new(
@@ -8063,7 +8081,8 @@ module Aws::SageMaker
8063
8081
  :experiment_config,
8064
8082
  :profiler_config,
8065
8083
  :profiler_rule_configurations,
8066
- :environment)
8084
+ :environment,
8085
+ :retry_strategy)
8067
8086
  SENSITIVE = []
8068
8087
  include Aws::Structure
8069
8088
  end
@@ -8541,7 +8560,7 @@ module Aws::SageMaker
8541
8560
  # @return [String]
8542
8561
  #
8543
8562
  # @!attribute [rw] user_profile_name
8544
- # A name for the UserProfile.
8563
+ # A name for the UserProfile. This value is not case sensitive.
8545
8564
  # @return [String]
8546
8565
  #
8547
8566
  # @!attribute [rw] single_sign_on_user_identifier
@@ -8563,6 +8582,9 @@ module Aws::SageMaker
8563
8582
  # @!attribute [rw] tags
8564
8583
  # Each tag consists of a key and an optional value. Tag keys must be
8565
8584
  # unique per resource.
8585
+ #
8586
+ # Tags that you specify for the User Profile are also added to all
8587
+ # Apps that the User Profile launches.
8566
8588
  # @return [Array<Types::Tag>]
8567
8589
  #
8568
8590
  # @!attribute [rw] user_settings
@@ -14141,9 +14163,9 @@ module Aws::SageMaker
14141
14163
  #
14142
14164
  # @!attribute [rw] stopping_condition
14143
14165
  # Specifies a limit to how long a model training job can run. It also
14144
- # specifies the maximum time to wait for a spot instance. When the job
14145
- # reaches the time limit, Amazon SageMaker ends the training job. Use
14146
- # this API to cap model training costs.
14166
+ # specifies how long a managed Spot training job has to complete. When
14167
+ # the job reaches the time limit, Amazon SageMaker ends the training
14168
+ # job. Use this API to cap model training costs.
14147
14169
  #
14148
14170
  # To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
14149
14171
  # signal, which delays job termination for 120 seconds. Algorithms can
@@ -14292,6 +14314,11 @@ module Aws::SageMaker
14292
14314
  # Profiling status of a training job.
14293
14315
  # @return [String]
14294
14316
  #
14317
+ # @!attribute [rw] retry_strategy
14318
+ # The number of times to retry the job when the job fails due to an
14319
+ # `InternalServerError`.
14320
+ # @return [Types::RetryStrategy]
14321
+ #
14295
14322
  # @!attribute [rw] environment
14296
14323
  # The environment variables to set in the Docker container.
14297
14324
  # @return [Hash<String,String>]
@@ -14337,6 +14364,7 @@ module Aws::SageMaker
14337
14364
  :profiler_rule_configurations,
14338
14365
  :profiler_rule_evaluation_statuses,
14339
14366
  :profiling_status,
14367
+ :retry_strategy,
14340
14368
  :environment)
14341
14369
  SENSITIVE = []
14342
14370
  include Aws::Structure
@@ -14719,7 +14747,7 @@ module Aws::SageMaker
14719
14747
  # @return [String]
14720
14748
  #
14721
14749
  # @!attribute [rw] user_profile_name
14722
- # The user profile name.
14750
+ # The user profile name. This value is not case sensitive.
14723
14751
  # @return [String]
14724
14752
  #
14725
14753
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeUserProfileRequest AWS API Documentation
@@ -15443,7 +15471,7 @@ module Aws::SageMaker
15443
15471
  #
15444
15472
  # @!attribute [rw] s3_data_distribution_type
15445
15473
  # Whether input data distributed in Amazon S3 is fully replicated or
15446
- # sharded by an S3 key. Defauts to `FullyReplicated`
15474
+ # sharded by an S3 key. Defaults to `FullyReplicated`
15447
15475
  # @return [String]
15448
15476
  #
15449
15477
  # @!attribute [rw] features_attribute
@@ -17989,6 +18017,9 @@ module Aws::SageMaker
17989
18017
  # s3_uri: "S3Uri", # required
17990
18018
  # local_path: "DirectoryPath",
17991
18019
  # },
18020
+ # retry_strategy: {
18021
+ # maximum_retry_attempts: 1, # required
18022
+ # },
17992
18023
  # }
17993
18024
  #
17994
18025
  # @!attribute [rw] definition_name
@@ -18070,10 +18101,9 @@ module Aws::SageMaker
18070
18101
  #
18071
18102
  # @!attribute [rw] stopping_condition
18072
18103
  # Specifies a limit to how long a model hyperparameter training job
18073
- # can run. It also specifies how long you are willing to wait for a
18074
- # managed spot training job to complete. When the job reaches the a
18075
- # limit, Amazon SageMaker ends the training job. Use this API to cap
18076
- # model training costs.
18104
+ # can run. It also specifies how long a managed spot training job has
18105
+ # to complete. When the job reaches the time limit, Amazon SageMaker
18106
+ # ends the training job. Use this API to cap model training costs.
18077
18107
  # @return [Types::StoppingCondition]
18078
18108
  #
18079
18109
  # @!attribute [rw] enable_network_isolation
@@ -18105,6 +18135,11 @@ module Aws::SageMaker
18105
18135
  # training checkpoint data.
18106
18136
  # @return [Types::CheckpointConfig]
18107
18137
  #
18138
+ # @!attribute [rw] retry_strategy
18139
+ # The number of times to retry the job when the job fails due to an
18140
+ # `InternalServerError`.
18141
+ # @return [Types::RetryStrategy]
18142
+ #
18108
18143
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/HyperParameterTrainingJobDefinition AWS API Documentation
18109
18144
  #
18110
18145
  class HyperParameterTrainingJobDefinition < Struct.new(
@@ -18122,7 +18157,8 @@ module Aws::SageMaker
18122
18157
  :enable_network_isolation,
18123
18158
  :enable_inter_container_traffic_encryption,
18124
18159
  :enable_managed_spot_training,
18125
- :checkpoint_config)
18160
+ :checkpoint_config,
18161
+ :retry_strategy)
18126
18162
  SENSITIVE = []
18127
18163
  include Aws::Structure
18128
18164
  end
@@ -19526,6 +19562,9 @@ module Aws::SageMaker
19526
19562
  #
19527
19563
  # @!attribute [rw] sns_topic_arn
19528
19564
  # An Amazon Simple Notification Service (Amazon SNS) output topic ARN.
19565
+ # Provide a `SnsTopicArn` if you want to do real time chaining to
19566
+ # another streaming job and receive an Amazon SNS notifications each
19567
+ # time a data object is submitted by a worker.
19529
19568
  #
19530
19569
  # If you provide an `SnsTopicArn` in `OutputConfig`, when workers
19531
19570
  # complete labeling tasks, Ground Truth will send labeling task output
@@ -21245,7 +21284,8 @@ module Aws::SageMaker
21245
21284
  # @return [String]
21246
21285
  #
21247
21286
  # @!attribute [rw] max_results
21248
- # The maximum number of endpoints to return in the response.
21287
+ # The maximum number of endpoints to return in the response. This
21288
+ # value defaults to 10.
21249
21289
  # @return [Integer]
21250
21290
  #
21251
21291
  # @!attribute [rw] name_contains
@@ -26503,6 +26543,13 @@ module Aws::SageMaker
26503
26543
  #
26504
26544
  # @!attribute [rw] max_runtime_in_seconds
26505
26545
  # The maximum runtime allowed in seconds.
26546
+ #
26547
+ # <note markdown="1"> The `MaxRuntimeInSeconds` cannot exceed the frequency of the job.
26548
+ # For data quality and model explainability, this can be up to 3600
26549
+ # seconds for an hourly schedule. For model bias and model quality
26550
+ # hourly schedules, this can be up to 1800 seconds.
26551
+ #
26552
+ # </note>
26506
26553
  # @return [Integer]
26507
26554
  #
26508
26555
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/MonitoringStoppingCondition AWS API Documentation
@@ -29741,7 +29788,7 @@ module Aws::SageMaker
29741
29788
  include Aws::Structure
29742
29789
  end
29743
29790
 
29744
- # Resource being accessed is in use.
29791
+ # The resource being accessed is in use.
29745
29792
  #
29746
29793
  # @!attribute [rw] message
29747
29794
  # @return [String]
@@ -29798,7 +29845,7 @@ module Aws::SageMaker
29798
29845
  include Aws::Structure
29799
29846
  end
29800
29847
 
29801
- # Resource being access is not found.
29848
+ # The resource being accessed was not found.
29802
29849
  #
29803
29850
  # @!attribute [rw] message
29804
29851
  # @return [String]
@@ -29870,6 +29917,32 @@ module Aws::SageMaker
29870
29917
  include Aws::Structure
29871
29918
  end
29872
29919
 
29920
+ # The retry strategy to use when a training job fails due to an
29921
+ # `InternalServerError`. `RetryStrategy` is specified as part of the
29922
+ # `CreateTrainingJob` and `CreateHyperParameterTuningJob` requests. You
29923
+ # can add the `StoppingCondition` parameter to the request to limit the
29924
+ # training time for the complete job.
29925
+ #
29926
+ # @note When making an API call, you may pass RetryStrategy
29927
+ # data as a hash:
29928
+ #
29929
+ # {
29930
+ # maximum_retry_attempts: 1, # required
29931
+ # }
29932
+ #
29933
+ # @!attribute [rw] maximum_retry_attempts
29934
+ # The number of times to retry the job. When the job is retried, it's
29935
+ # `SecondaryStatus` is changed to `STARTING`.
29936
+ # @return [Integer]
29937
+ #
29938
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/RetryStrategy AWS API Documentation
29939
+ #
29940
+ class RetryStrategy < Struct.new(
29941
+ :maximum_retry_attempts)
29942
+ SENSITIVE = []
29943
+ include Aws::Structure
29944
+ end
29945
+
29873
29946
  # Describes the S3 data source.
29874
29947
  #
29875
29948
  # @note When making an API call, you may pass S3DataSource
@@ -31102,11 +31175,11 @@ module Aws::SageMaker
31102
31175
  include Aws::Structure
31103
31176
  end
31104
31177
 
31105
- # Specifies a limit to how long a model training or compilation job can
31106
- # run. It also specifies how long you are willing to wait for a managed
31107
- # spot training job to complete. When the job reaches the time limit,
31108
- # Amazon SageMaker ends the training or compilation job. Use this API to
31109
- # cap model training costs.
31178
+ # Specifies a limit to how long a model training job, model compilation
31179
+ # job, or hyperparameter tuning job can run. It also specifies how long
31180
+ # a managed Spot training job has to complete. When the job reaches the
31181
+ # time limit, Amazon SageMaker ends the training or compilation job. Use
31182
+ # this API to cap model training costs.
31110
31183
  #
31111
31184
  # To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
31112
31185
  # signal, which delays job termination for 120 seconds. Algorithms can
@@ -31136,18 +31209,27 @@ module Aws::SageMaker
31136
31209
  # }
31137
31210
  #
31138
31211
  # @!attribute [rw] max_runtime_in_seconds
31139
- # The maximum length of time, in seconds, that the training or
31140
- # compilation job can run. If job does not complete during this time,
31141
- # Amazon SageMaker ends the job. If value is not specified, default
31142
- # value is 1 day. The maximum value is 28 days.
31212
+ # The maximum length of time, in seconds, that a training or
31213
+ # compilation job can run. If the job does not complete during this
31214
+ # time, Amazon SageMaker ends the job.
31215
+ #
31216
+ # When `RetryStrategy` is specified in the job request,
31217
+ # `MaxRuntimeInSeconds` specifies the maximum time for all of the
31218
+ # attempts in total, not each individual attempt.
31219
+ #
31220
+ # The default value is 1 day. The maximum value is 28 days.
31143
31221
  # @return [Integer]
31144
31222
  #
31145
31223
  # @!attribute [rw] max_wait_time_in_seconds
31146
- # The maximum length of time, in seconds, how long you are willing to
31147
- # wait for a managed spot training job to complete. It is the amount
31148
- # of time spent waiting for Spot capacity plus the amount of time the
31149
- # training job runs. It must be equal to or greater than
31150
- # `MaxRuntimeInSeconds`.
31224
+ # The maximum length of time, in seconds, that a managed Spot training
31225
+ # job has to complete. It is the amount of time spent waiting for Spot
31226
+ # capacity plus the amount of time the job can run. It must be equal
31227
+ # to or greater than `MaxRuntimeInSeconds`. If the job does not
31228
+ # complete during this time, Amazon SageMaker ends the job.
31229
+ #
31230
+ # When `RetryStrategy` is specified in the job request,
31231
+ # `MaxWaitTimeInSeconds` specifies the maximum time for all of the
31232
+ # attempts in total, not each individual attempt.
31151
31233
  # @return [Integer]
31152
31234
  #
31153
31235
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/StoppingCondition AWS API Documentation
@@ -31542,9 +31624,10 @@ module Aws::SageMaker
31542
31624
  # @return [Types::VpcConfig]
31543
31625
  #
31544
31626
  # @!attribute [rw] stopping_condition
31545
- # Specifies a limit to how long a model training job can run. When the
31546
- # job reaches the time limit, Amazon SageMaker ends the training job.
31547
- # Use this API to cap model training costs.
31627
+ # Specifies a limit to how long a model training job can run. It also
31628
+ # specifies how long a managed Spot training job has to complete. When
31629
+ # the job reaches the time limit, Amazon SageMaker ends the training
31630
+ # job. Use this API to cap model training costs.
31548
31631
  #
31549
31632
  # To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
31550
31633
  # signal, which delays job termination for 120 seconds. Algorithms can
@@ -31669,6 +31752,11 @@ module Aws::SageMaker
31669
31752
  # The environment variables to set in the Docker container.
31670
31753
  # @return [Hash<String,String>]
31671
31754
  #
31755
+ # @!attribute [rw] retry_strategy
31756
+ # The number of times to retry the job when the job fails due to an
31757
+ # `InternalServerError`.
31758
+ # @return [Types::RetryStrategy]
31759
+ #
31672
31760
  # @!attribute [rw] tags
31673
31761
  # An array of key-value pairs. You can use tags to categorize your AWS
31674
31762
  # resources in different ways, for example, by purpose, owner, or
@@ -31717,6 +31805,7 @@ module Aws::SageMaker
31717
31805
  :tensor_board_output_config,
31718
31806
  :debug_rule_evaluation_statuses,
31719
31807
  :environment,
31808
+ :retry_strategy,
31720
31809
  :tags)
31721
31810
  SENSITIVE = []
31722
31811
  include Aws::Structure
@@ -31810,9 +31899,10 @@ module Aws::SageMaker
31810
31899
  # @return [Types::ResourceConfig]
31811
31900
  #
31812
31901
  # @!attribute [rw] stopping_condition
31813
- # Specifies a limit to how long a model training job can run. When the
31814
- # job reaches the time limit, Amazon SageMaker ends the training job.
31815
- # Use this API to cap model training costs.
31902
+ # Specifies a limit to how long a model training job can run. It also
31903
+ # specifies how long a managed Spot training job has to complete. When
31904
+ # the job reaches the time limit, Amazon SageMaker ends the training
31905
+ # job. Use this API to cap model training costs.
31816
31906
  #
31817
31907
  # To stop a job, Amazon SageMaker sends the algorithm the SIGTERM
31818
31908
  # signal, which delays job termination for 120 seconds. Algorithms can
@@ -32145,7 +32235,7 @@ module Aws::SageMaker
32145
32235
  #
32146
32236
  #
32147
32237
  # [1]: https://mxnet.apache.org/api/faq/recordio
32148
- # [2]: https://www.tensorflow.org/guide/datasets#consuming_tfrecord_data
32238
+ # [2]: https://www.tensorflow.org/guide/data#consuming_tfrecord_data
32149
32239
  # @return [String]
32150
32240
  #
32151
32241
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TransformInput AWS API Documentation
metadata CHANGED
@@ -1,14 +1,14 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: aws-sdk-sagemaker
3
3
  version: !ruby/object:Gem::Version
4
- version: 1.85.0
4
+ version: 1.86.0
5
5
  platform: ruby
6
6
  authors:
7
7
  - Amazon Web Services
8
8
  autorequire:
9
9
  bindir: bin
10
10
  cert_chain: []
11
- date: 2021-03-30 00:00:00.000000000 Z
11
+ date: 2021-05-04 00:00:00.000000000 Z
12
12
  dependencies:
13
13
  - !ruby/object:Gem::Dependency
14
14
  name: aws-sdk-core
@@ -84,8 +84,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
84
84
  - !ruby/object:Gem::Version
85
85
  version: '0'
86
86
  requirements: []
87
- rubyforge_project:
88
- rubygems_version: 2.7.6.2
87
+ rubygems_version: 3.1.6
89
88
  signing_key:
90
89
  specification_version: 4
91
90
  summary: AWS SDK for Ruby - SageMaker