aws-sdk-sagemaker 1.187.0 → 1.188.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: 0ff45d1ca80af78f54ad7574ec4a27aabb57f2ebb78cb8d2871f2f66068b2494
4
- data.tar.gz: 38dbefd7a3cbad8adc50b30cc4f3b156e72c9dc555e91f2dbaf706f6be34851d
3
+ metadata.gz: c1cc6aa017d0a61c512a403ca520b758a79858d6ed30444a6b8fc209235a566b
4
+ data.tar.gz: 7c3ed41853cabbc93820d3ee924aae42852b605a76aae984c98b384f5c0547bb
5
5
  SHA512:
6
- metadata.gz: b76e78194cc7e824435ace5d161aa5ef41bca24fa44684a09abdd8f737b3e36af7517e26e64189a3ef4636ae8a28f6c50c7d69ce7f0338088878aec312e554c1
7
- data.tar.gz: 2a1d91bf4b47f4cf0346280826f772b896fca9cfed844bc990f60773187987dcca7fa05f6ec35491961af9175a76b787c70999d00e8650750adb586f8c931e39
6
+ metadata.gz: e12d3d897ebdc29ead6763814cecb2629a79a63585eae30c743751f0eeea93b91250ed5dfc0c807e12f9105ca162b16157664aa1144caad28d2625a97fabc5a4
7
+ data.tar.gz: 2edc4ba1e484791aebb0a59b2471042c5c54aa75caa2742ea0027dc0d8ba61488e6fefcd7c4e2fcc29c0648dafe58ee7e6ba931edd5c9653370f8ffe1742a1f9
data/CHANGELOG.md CHANGED
@@ -1,6 +1,11 @@
1
1
  Unreleased Changes
2
2
  ------------------
3
3
 
4
+ 1.188.0 (2023-06-19)
5
+ ------------------
6
+
7
+ * Feature - Amazon Sagemaker Autopilot releases CreateAutoMLJobV2 and DescribeAutoMLJobV2 for Autopilot customers with ImageClassification, TextClassification and Tabular problem type config support.
8
+
4
9
  1.187.0 (2023-06-15)
5
10
  ------------------
6
11
 
data/VERSION CHANGED
@@ -1 +1 @@
1
- 1.187.0
1
+ 1.188.0
@@ -1184,18 +1184,30 @@ module Aws::SageMaker
1184
1184
  req.send_request(options)
1185
1185
  end
1186
1186
 
1187
- # Creates an Autopilot job.
1187
+ # Creates an Autopilot job also referred to as Autopilot experiment or
1188
+ # AutoML job.
1188
1189
  #
1189
- # Find the best-performing model after you run an Autopilot job by
1190
- # calling [DescribeAutoMLJob][1].
1190
+ # Find the best-performing model after you run an AutoML job by calling
1191
+ # [DescribeAutoMLJobV2][1] (recommended) or [DescribeAutoMLJob][2].
1191
1192
  #
1192
- # For information about how to use Autopilot, see [Automate Model
1193
- # Development with Amazon SageMaker Autopilot][2].
1193
+ # <note markdown="1"> `CreateAutoMLJob` only accepts tabular input data. We recommend using
1194
+ # [CreateAutoMLJobV2][3] for all problem types. `CreateAutoMLJobV2` can
1195
+ # process the same tabular data as its previous version
1196
+ # `CreateAutoMLJob`, as well as non-tabular data for problem types such
1197
+ # as image or text classification.
1194
1198
  #
1199
+ # Find guidelines about how to migrate `CreateAutoMLJob` to
1200
+ # `CreateAutoMLJobV2` in [Migrate a CreateAutoMLJob to
1201
+ # CreateAutoMLJobV2][4].
1202
+ #
1203
+ # </note>
1195
1204
  #
1196
1205
  #
1197
- # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJob.html
1198
- # [2]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-automate-model-development.html
1206
+ #
1207
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJobV2.html
1208
+ # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJob.html
1209
+ # [3]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
1210
+ # [4]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-automate-model-development-create-experiment-api.html#autopilot-create-experiment-api-migrate-v1-v2
1199
1211
  #
1200
1212
  # @option params [required, String] :auto_ml_job_name
1201
1213
  # Identifies an Autopilot job. The name must be unique to your account
@@ -1229,15 +1241,13 @@ module Aws::SageMaker
1229
1241
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-datasets-problem-types.html#autopilot-problem-types
1230
1242
  #
1231
1243
  # @option params [Types::AutoMLJobObjective] :auto_ml_job_objective
1232
- # Defines the objective metric used to measure the predictive quality of
1233
- # an AutoML job. You provide an [AutoMLJobObjective$MetricName][1] and
1234
- # Autopilot infers whether to minimize or maximize it. For
1235
- # [CreateAutoMLJobV2][2], only `Accuracy` is supported.
1244
+ # Specifies a metric to minimize or maximize as the objective of a job.
1245
+ # If not specified, the default objective metric depends on the problem
1246
+ # type. See [AutoMLJobObjective][1] for the default values.
1236
1247
  #
1237
1248
  #
1238
1249
  #
1239
1250
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLJobObjective.html
1240
- # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
1241
1251
  #
1242
1252
  # @option params [Types::AutoMLJobConfig] :auto_ml_job_config
1243
1253
  # A collection of settings used to configure an AutoML job.
@@ -1350,24 +1360,32 @@ module Aws::SageMaker
1350
1360
  req.send_request(options)
1351
1361
  end
1352
1362
 
1353
- # Creates an Amazon SageMaker AutoML job that uses non-tabular data such
1354
- # as images or text for Computer Vision or Natural Language Processing
1355
- # problems.
1363
+ # Creates an Autopilot job also referred to as Autopilot experiment or
1364
+ # AutoML job V2.
1356
1365
  #
1357
- # Find the resulting model after you run an AutoML job V2 by calling
1358
- # [DescribeAutoMLJobV2][1].
1366
+ # We recommend using [CreateAutoMLJobV2][1] for all problem types.
1367
+ # `CreateAutoMLJobV2` can process the same tabular data as its previous
1368
+ # version `CreateAutoMLJob`, as well as non-tabular data for problem
1369
+ # types such as image or text classification.
1359
1370
  #
1360
- # To create an `AutoMLJob` using tabular data, see [CreateAutoMLJob][2].
1371
+ # Find guidelines about how to migrate `CreateAutoMLJob` to
1372
+ # `CreateAutoMLJobV2` in [Migrate a CreateAutoMLJob to
1373
+ # CreateAutoMLJobV2][2].
1361
1374
  #
1362
- # <note markdown="1"> This API action is callable through SageMaker Canvas only. Calling it
1363
- # directly from the CLI or an SDK results in an error.
1375
+ # For the list of available problem types supported by
1376
+ # `CreateAutoMLJobV2`, see [AutoMLProblemTypeConfig][3].
1364
1377
  #
1365
- # </note>
1378
+ # Find the best-performing model after you run an AutoML job V2 by
1379
+ # calling [DescribeAutoMLJobV2][4]. Calling [DescribeAutoMLJob][5] on a
1380
+ # AutoML job V2 results in an error.
1366
1381
  #
1367
1382
  #
1368
1383
  #
1369
- # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJobV2.html
1370
- # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJob.html
1384
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
1385
+ # [2]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-automate-model-development-create-experiment-api.html#autopilot-create-experiment-api-migrate-v1-v2
1386
+ # [3]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLProblemTypeConfig.html
1387
+ # [4]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJobV2.html
1388
+ # [5]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DescribeAutoMLJob.html
1371
1389
  #
1372
1390
  # @option params [required, String] :auto_ml_job_name
1373
1391
  # Identifies an Autopilot job. The name must be unique to your account
@@ -1379,10 +1397,12 @@ module Aws::SageMaker
1379
1397
  # [InputDataConfig][1] supported by `CreateAutoMLJob`. The supported
1380
1398
  # formats depend on the problem type:
1381
1399
  #
1382
- # * ImageClassification: S3Prefix, `ManifestFile`,
1383
- # `AugmentedManifestFile`
1400
+ # * For Tabular problem types: `S3Prefix`, `ManifestFile`.
1401
+ #
1402
+ # * For ImageClassification: `S3Prefix`, `ManifestFile`,
1403
+ # `AugmentedManifestFile`.
1384
1404
  #
1385
- # * TextClassification: S3Prefix
1405
+ # * For TextClassification: `S3Prefix`.
1386
1406
  #
1387
1407
  #
1388
1408
  #
@@ -1396,6 +1416,13 @@ module Aws::SageMaker
1396
1416
  # Defines the configuration settings of one of the supported problem
1397
1417
  # types.
1398
1418
  #
1419
+ # <note markdown="1"> For tabular problem types, you must either specify the type of
1420
+ # supervised learning problem in `AutoMLProblemTypeConfig`
1421
+ # (`TabularJobConfig.ProblemType`) and provide the `AutoMLJobObjective`,
1422
+ # or none at all.
1423
+ #
1424
+ # </note>
1425
+ #
1399
1426
  # @option params [required, String] :role_arn
1400
1427
  # The ARN of the role that is used to access the data.
1401
1428
  #
@@ -1415,11 +1442,20 @@ module Aws::SageMaker
1415
1442
  #
1416
1443
  # @option params [Types::AutoMLJobObjective] :auto_ml_job_objective
1417
1444
  # Specifies a metric to minimize or maximize as the objective of a job.
1418
- # For [CreateAutoMLJobV2][1], only `Accuracy` is supported.
1445
+ # If not specified, the default objective metric depends on the problem
1446
+ # type. For the list of default values per problem type, see
1447
+ # [AutoMLJobObjective][1].
1448
+ #
1449
+ # <note markdown="1"> For tabular problem types, you must either provide the
1450
+ # `AutoMLJobObjective` and indicate the type of supervised learning
1451
+ # problem in `AutoMLProblemTypeConfig` (`TabularJobConfig.ProblemType`),
1452
+ # or none.
1419
1453
  #
1454
+ # </note>
1420
1455
  #
1421
1456
  #
1422
- # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
1457
+ #
1458
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLJobObjective.html
1423
1459
  #
1424
1460
  # @option params [Types::ModelDeployConfig] :model_deploy_config
1425
1461
  # Specifies how to generate the endpoint name for an automatic one-click
@@ -1429,12 +1465,9 @@ module Aws::SageMaker
1429
1465
  # This structure specifies how to split the data into train and
1430
1466
  # validation datasets.
1431
1467
  #
1432
- # If you are using the V1 API (for example `CreateAutoMLJob`) or the V2
1433
- # API for Natural Language Processing problems (for example
1434
- # `CreateAutoMLJobV2` with a `TextClassificationJobConfig` problem
1435
- # type), the validation and training datasets must contain the same
1436
- # headers. Also, for V1 API jobs, the validation dataset must be less
1437
- # than 2 GB in size.
1468
+ # The validation and training datasets must contain the same headers.
1469
+ # For jobs created by calling `CreateAutoMLJob`, the validation dataset
1470
+ # must be less than 2 GB in size.
1438
1471
  #
1439
1472
  # @return [Types::CreateAutoMLJobV2Response] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
1440
1473
  #
@@ -1478,6 +1511,26 @@ module Aws::SageMaker
1478
1511
  # content_column: "ContentColumn",
1479
1512
  # target_label_column: "TargetLabelColumn",
1480
1513
  # },
1514
+ # tabular_job_config: {
1515
+ # candidate_generation_config: {
1516
+ # algorithms_config: [
1517
+ # {
1518
+ # auto_ml_algorithms: ["xgboost"], # required, accepts xgboost, linear-learner, mlp, lightgbm, catboost, randomforest, extra-trees, nn-torch, fastai
1519
+ # },
1520
+ # ],
1521
+ # },
1522
+ # completion_criteria: {
1523
+ # max_candidates: 1,
1524
+ # max_runtime_per_training_job_in_seconds: 1,
1525
+ # max_auto_ml_job_runtime_in_seconds: 1,
1526
+ # },
1527
+ # feature_specification_s3_uri: "S3Uri",
1528
+ # mode: "AUTO", # accepts AUTO, ENSEMBLING, HYPERPARAMETER_TUNING
1529
+ # generate_candidate_definitions_only: false,
1530
+ # problem_type: "BinaryClassification", # accepts BinaryClassification, MulticlassClassification, Regression
1531
+ # target_attribute_name: "TargetAttributeName", # required
1532
+ # sample_weight_attribute_name: "SampleWeightAttributeName",
1533
+ # },
1481
1534
  # },
1482
1535
  # role_arn: "RoleArn", # required
1483
1536
  # tags: [
@@ -9950,7 +10003,12 @@ module Aws::SageMaker
9950
10003
  req.send_request(options)
9951
10004
  end
9952
10005
 
9953
- # Returns information about an Amazon SageMaker AutoML job.
10006
+ # Returns information about an AutoML job created by calling
10007
+ # [CreateAutoMLJob][1].
10008
+ #
10009
+ #
10010
+ #
10011
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJob.html
9954
10012
  #
9955
10013
  # @option params [required, String] :auto_ml_job_name
9956
10014
  # Requests information about an AutoML job using its unique name.
@@ -10079,15 +10137,15 @@ module Aws::SageMaker
10079
10137
  req.send_request(options)
10080
10138
  end
10081
10139
 
10082
- # Returns information about an Amazon SageMaker AutoML V2 job.
10140
+ # Returns information about an AutoML job V2 created by calling
10141
+ # [CreateAutoMLJobV2][1].
10083
10142
  #
10084
- # <note markdown="1"> This API action is callable through SageMaker Canvas only. Calling it
10085
- # directly from the CLI or an SDK results in an error.
10086
10143
  #
10087
- # </note>
10144
+ #
10145
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
10088
10146
  #
10089
10147
  # @option params [required, String] :auto_ml_job_name
10090
- # Requests information about an AutoML V2 job using its unique name.
10148
+ # Requests information about an AutoML job V2 using its unique name.
10091
10149
  #
10092
10150
  # @return [Types::DescribeAutoMLJobV2Response] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
10093
10151
  #
@@ -10110,6 +10168,9 @@ module Aws::SageMaker
10110
10168
  # * {Types::DescribeAutoMLJobV2Response#model_deploy_result #model_deploy_result} => Types::ModelDeployResult
10111
10169
  # * {Types::DescribeAutoMLJobV2Response#data_split_config #data_split_config} => Types::AutoMLDataSplitConfig
10112
10170
  # * {Types::DescribeAutoMLJobV2Response#security_config #security_config} => Types::AutoMLSecurityConfig
10171
+ # * {Types::DescribeAutoMLJobV2Response#auto_ml_job_artifacts #auto_ml_job_artifacts} => Types::AutoMLJobArtifacts
10172
+ # * {Types::DescribeAutoMLJobV2Response#resolved_attributes #resolved_attributes} => Types::AutoMLResolvedAttributes
10173
+ # * {Types::DescribeAutoMLJobV2Response#auto_ml_problem_type_config_name #auto_ml_problem_type_config_name} => String
10113
10174
  #
10114
10175
  # @example Request syntax with placeholder values
10115
10176
  #
@@ -10139,6 +10200,18 @@ module Aws::SageMaker
10139
10200
  # resp.auto_ml_problem_type_config.text_classification_job_config.completion_criteria.max_auto_ml_job_runtime_in_seconds #=> Integer
10140
10201
  # resp.auto_ml_problem_type_config.text_classification_job_config.content_column #=> String
10141
10202
  # resp.auto_ml_problem_type_config.text_classification_job_config.target_label_column #=> String
10203
+ # resp.auto_ml_problem_type_config.tabular_job_config.candidate_generation_config.algorithms_config #=> Array
10204
+ # resp.auto_ml_problem_type_config.tabular_job_config.candidate_generation_config.algorithms_config[0].auto_ml_algorithms #=> Array
10205
+ # resp.auto_ml_problem_type_config.tabular_job_config.candidate_generation_config.algorithms_config[0].auto_ml_algorithms[0] #=> String, one of "xgboost", "linear-learner", "mlp", "lightgbm", "catboost", "randomforest", "extra-trees", "nn-torch", "fastai"
10206
+ # resp.auto_ml_problem_type_config.tabular_job_config.completion_criteria.max_candidates #=> Integer
10207
+ # resp.auto_ml_problem_type_config.tabular_job_config.completion_criteria.max_runtime_per_training_job_in_seconds #=> Integer
10208
+ # resp.auto_ml_problem_type_config.tabular_job_config.completion_criteria.max_auto_ml_job_runtime_in_seconds #=> Integer
10209
+ # resp.auto_ml_problem_type_config.tabular_job_config.feature_specification_s3_uri #=> String
10210
+ # resp.auto_ml_problem_type_config.tabular_job_config.mode #=> String, one of "AUTO", "ENSEMBLING", "HYPERPARAMETER_TUNING"
10211
+ # resp.auto_ml_problem_type_config.tabular_job_config.generate_candidate_definitions_only #=> Boolean
10212
+ # resp.auto_ml_problem_type_config.tabular_job_config.problem_type #=> String, one of "BinaryClassification", "MulticlassClassification", "Regression"
10213
+ # resp.auto_ml_problem_type_config.tabular_job_config.target_attribute_name #=> String
10214
+ # resp.auto_ml_problem_type_config.tabular_job_config.sample_weight_attribute_name #=> String
10142
10215
  # resp.creation_time #=> Time
10143
10216
  # resp.end_time #=> Time
10144
10217
  # resp.last_modified_time #=> Time
@@ -10190,6 +10263,14 @@ module Aws::SageMaker
10190
10263
  # resp.security_config.vpc_config.security_group_ids[0] #=> String
10191
10264
  # resp.security_config.vpc_config.subnets #=> Array
10192
10265
  # resp.security_config.vpc_config.subnets[0] #=> String
10266
+ # resp.auto_ml_job_artifacts.candidate_definition_notebook_location #=> String
10267
+ # resp.auto_ml_job_artifacts.data_exploration_notebook_location #=> String
10268
+ # resp.resolved_attributes.auto_ml_job_objective.metric_name #=> String, one of "Accuracy", "MSE", "F1", "F1macro", "AUC", "RMSE", "MAE", "R2", "BalancedAccuracy", "Precision", "PrecisionMacro", "Recall", "RecallMacro"
10269
+ # resp.resolved_attributes.completion_criteria.max_candidates #=> Integer
10270
+ # resp.resolved_attributes.completion_criteria.max_runtime_per_training_job_in_seconds #=> Integer
10271
+ # resp.resolved_attributes.completion_criteria.max_auto_ml_job_runtime_in_seconds #=> Integer
10272
+ # resp.resolved_attributes.auto_ml_problem_type_resolved_attributes.tabular_resolved_attributes.problem_type #=> String, one of "BinaryClassification", "MulticlassClassification", "Regression"
10273
+ # resp.auto_ml_problem_type_config_name #=> String, one of "ImageClassification", "TextClassification", "Tabular"
10193
10274
  #
10194
10275
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeAutoMLJobV2 AWS API Documentation
10195
10276
  #
@@ -23627,7 +23708,7 @@ module Aws::SageMaker
23627
23708
  params: params,
23628
23709
  config: config)
23629
23710
  context[:gem_name] = 'aws-sdk-sagemaker'
23630
- context[:gem_version] = '1.187.0'
23711
+ context[:gem_version] = '1.188.0'
23631
23712
  Seahorse::Client::Request.new(handlers, context)
23632
23713
  end
23633
23714
 
@@ -137,7 +137,10 @@ module Aws::SageMaker
137
137
  AutoMLPartialFailureReason = Shapes::StructureShape.new(name: 'AutoMLPartialFailureReason')
138
138
  AutoMLPartialFailureReasons = Shapes::ListShape.new(name: 'AutoMLPartialFailureReasons')
139
139
  AutoMLProblemTypeConfig = Shapes::UnionShape.new(name: 'AutoMLProblemTypeConfig')
140
+ AutoMLProblemTypeConfigName = Shapes::StringShape.new(name: 'AutoMLProblemTypeConfigName')
141
+ AutoMLProblemTypeResolvedAttributes = Shapes::UnionShape.new(name: 'AutoMLProblemTypeResolvedAttributes')
140
142
  AutoMLProcessingUnit = Shapes::StringShape.new(name: 'AutoMLProcessingUnit')
143
+ AutoMLResolvedAttributes = Shapes::StructureShape.new(name: 'AutoMLResolvedAttributes')
141
144
  AutoMLS3DataSource = Shapes::StructureShape.new(name: 'AutoMLS3DataSource')
142
145
  AutoMLS3DataType = Shapes::StringShape.new(name: 'AutoMLS3DataType')
143
146
  AutoMLSecurityConfig = Shapes::StructureShape.new(name: 'AutoMLSecurityConfig')
@@ -171,6 +174,7 @@ module Aws::SageMaker
171
174
  CallbackToken = Shapes::StringShape.new(name: 'CallbackToken')
172
175
  CandidateArtifactLocations = Shapes::StructureShape.new(name: 'CandidateArtifactLocations')
173
176
  CandidateDefinitionNotebookLocation = Shapes::StringShape.new(name: 'CandidateDefinitionNotebookLocation')
177
+ CandidateGenerationConfig = Shapes::StructureShape.new(name: 'CandidateGenerationConfig')
174
178
  CandidateName = Shapes::StringShape.new(name: 'CandidateName')
175
179
  CandidateProperties = Shapes::StructureShape.new(name: 'CandidateProperties')
176
180
  CandidateSortBy = Shapes::StringShape.new(name: 'CandidateSortBy')
@@ -1786,6 +1790,8 @@ module Aws::SageMaker
1786
1790
  SuggestionQuery = Shapes::StructureShape.new(name: 'SuggestionQuery')
1787
1791
  TableFormat = Shapes::StringShape.new(name: 'TableFormat')
1788
1792
  TableName = Shapes::StringShape.new(name: 'TableName')
1793
+ TabularJobConfig = Shapes::StructureShape.new(name: 'TabularJobConfig')
1794
+ TabularResolvedAttributes = Shapes::StructureShape.new(name: 'TabularResolvedAttributes')
1789
1795
  Tag = Shapes::StructureShape.new(name: 'Tag')
1790
1796
  TagKey = Shapes::StringShape.new(name: 'TagKey')
1791
1797
  TagKeyList = Shapes::ListShape.new(name: 'TagKeyList')
@@ -2327,12 +2333,25 @@ module Aws::SageMaker
2327
2333
 
2328
2334
  AutoMLProblemTypeConfig.add_member(:image_classification_job_config, Shapes::ShapeRef.new(shape: ImageClassificationJobConfig, location_name: "ImageClassificationJobConfig"))
2329
2335
  AutoMLProblemTypeConfig.add_member(:text_classification_job_config, Shapes::ShapeRef.new(shape: TextClassificationJobConfig, location_name: "TextClassificationJobConfig"))
2336
+ AutoMLProblemTypeConfig.add_member(:tabular_job_config, Shapes::ShapeRef.new(shape: TabularJobConfig, location_name: "TabularJobConfig"))
2330
2337
  AutoMLProblemTypeConfig.add_member(:unknown, Shapes::ShapeRef.new(shape: nil, location_name: 'unknown'))
2331
2338
  AutoMLProblemTypeConfig.add_member_subclass(:image_classification_job_config, Types::AutoMLProblemTypeConfig::ImageClassificationJobConfig)
2332
2339
  AutoMLProblemTypeConfig.add_member_subclass(:text_classification_job_config, Types::AutoMLProblemTypeConfig::TextClassificationJobConfig)
2340
+ AutoMLProblemTypeConfig.add_member_subclass(:tabular_job_config, Types::AutoMLProblemTypeConfig::TabularJobConfig)
2333
2341
  AutoMLProblemTypeConfig.add_member_subclass(:unknown, Types::AutoMLProblemTypeConfig::Unknown)
2334
2342
  AutoMLProblemTypeConfig.struct_class = Types::AutoMLProblemTypeConfig
2335
2343
 
2344
+ AutoMLProblemTypeResolvedAttributes.add_member(:tabular_resolved_attributes, Shapes::ShapeRef.new(shape: TabularResolvedAttributes, location_name: "TabularResolvedAttributes"))
2345
+ AutoMLProblemTypeResolvedAttributes.add_member(:unknown, Shapes::ShapeRef.new(shape: nil, location_name: 'unknown'))
2346
+ AutoMLProblemTypeResolvedAttributes.add_member_subclass(:tabular_resolved_attributes, Types::AutoMLProblemTypeResolvedAttributes::TabularResolvedAttributes)
2347
+ AutoMLProblemTypeResolvedAttributes.add_member_subclass(:unknown, Types::AutoMLProblemTypeResolvedAttributes::Unknown)
2348
+ AutoMLProblemTypeResolvedAttributes.struct_class = Types::AutoMLProblemTypeResolvedAttributes
2349
+
2350
+ AutoMLResolvedAttributes.add_member(:auto_ml_job_objective, Shapes::ShapeRef.new(shape: AutoMLJobObjective, location_name: "AutoMLJobObjective"))
2351
+ AutoMLResolvedAttributes.add_member(:completion_criteria, Shapes::ShapeRef.new(shape: AutoMLJobCompletionCriteria, location_name: "CompletionCriteria"))
2352
+ AutoMLResolvedAttributes.add_member(:auto_ml_problem_type_resolved_attributes, Shapes::ShapeRef.new(shape: AutoMLProblemTypeResolvedAttributes, location_name: "AutoMLProblemTypeResolvedAttributes"))
2353
+ AutoMLResolvedAttributes.struct_class = Types::AutoMLResolvedAttributes
2354
+
2336
2355
  AutoMLS3DataSource.add_member(:s3_data_type, Shapes::ShapeRef.new(shape: AutoMLS3DataType, required: true, location_name: "S3DataType"))
2337
2356
  AutoMLS3DataSource.add_member(:s3_uri, Shapes::ShapeRef.new(shape: S3Uri, required: true, location_name: "S3Uri"))
2338
2357
  AutoMLS3DataSource.struct_class = Types::AutoMLS3DataSource
@@ -2421,6 +2440,9 @@ module Aws::SageMaker
2421
2440
  CandidateArtifactLocations.add_member(:model_insights, Shapes::ShapeRef.new(shape: ModelInsightsLocation, location_name: "ModelInsights"))
2422
2441
  CandidateArtifactLocations.struct_class = Types::CandidateArtifactLocations
2423
2442
 
2443
+ CandidateGenerationConfig.add_member(:algorithms_config, Shapes::ShapeRef.new(shape: AutoMLAlgorithmsConfig, location_name: "AlgorithmsConfig"))
2444
+ CandidateGenerationConfig.struct_class = Types::CandidateGenerationConfig
2445
+
2424
2446
  CandidateProperties.add_member(:candidate_artifact_locations, Shapes::ShapeRef.new(shape: CandidateArtifactLocations, location_name: "CandidateArtifactLocations"))
2425
2447
  CandidateProperties.add_member(:candidate_metrics, Shapes::ShapeRef.new(shape: MetricDataList, location_name: "CandidateMetrics"))
2426
2448
  CandidateProperties.struct_class = Types::CandidateProperties
@@ -3764,6 +3786,9 @@ module Aws::SageMaker
3764
3786
  DescribeAutoMLJobV2Response.add_member(:model_deploy_result, Shapes::ShapeRef.new(shape: ModelDeployResult, location_name: "ModelDeployResult"))
3765
3787
  DescribeAutoMLJobV2Response.add_member(:data_split_config, Shapes::ShapeRef.new(shape: AutoMLDataSplitConfig, location_name: "DataSplitConfig"))
3766
3788
  DescribeAutoMLJobV2Response.add_member(:security_config, Shapes::ShapeRef.new(shape: AutoMLSecurityConfig, location_name: "SecurityConfig"))
3789
+ DescribeAutoMLJobV2Response.add_member(:auto_ml_job_artifacts, Shapes::ShapeRef.new(shape: AutoMLJobArtifacts, location_name: "AutoMLJobArtifacts"))
3790
+ DescribeAutoMLJobV2Response.add_member(:resolved_attributes, Shapes::ShapeRef.new(shape: AutoMLResolvedAttributes, location_name: "ResolvedAttributes"))
3791
+ DescribeAutoMLJobV2Response.add_member(:auto_ml_problem_type_config_name, Shapes::ShapeRef.new(shape: AutoMLProblemTypeConfigName, location_name: "AutoMLProblemTypeConfigName"))
3767
3792
  DescribeAutoMLJobV2Response.struct_class = Types::DescribeAutoMLJobV2Response
3768
3793
 
3769
3794
  DescribeCodeRepositoryInput.add_member(:code_repository_name, Shapes::ShapeRef.new(shape: EntityName, required: true, location_name: "CodeRepositoryName"))
@@ -8200,6 +8225,19 @@ module Aws::SageMaker
8200
8225
  SuggestionQuery.add_member(:property_name_query, Shapes::ShapeRef.new(shape: PropertyNameQuery, location_name: "PropertyNameQuery"))
8201
8226
  SuggestionQuery.struct_class = Types::SuggestionQuery
8202
8227
 
8228
+ TabularJobConfig.add_member(:candidate_generation_config, Shapes::ShapeRef.new(shape: CandidateGenerationConfig, location_name: "CandidateGenerationConfig"))
8229
+ TabularJobConfig.add_member(:completion_criteria, Shapes::ShapeRef.new(shape: AutoMLJobCompletionCriteria, location_name: "CompletionCriteria"))
8230
+ TabularJobConfig.add_member(:feature_specification_s3_uri, Shapes::ShapeRef.new(shape: S3Uri, location_name: "FeatureSpecificationS3Uri"))
8231
+ TabularJobConfig.add_member(:mode, Shapes::ShapeRef.new(shape: AutoMLMode, location_name: "Mode"))
8232
+ TabularJobConfig.add_member(:generate_candidate_definitions_only, Shapes::ShapeRef.new(shape: GenerateCandidateDefinitionsOnly, location_name: "GenerateCandidateDefinitionsOnly"))
8233
+ TabularJobConfig.add_member(:problem_type, Shapes::ShapeRef.new(shape: ProblemType, location_name: "ProblemType"))
8234
+ TabularJobConfig.add_member(:target_attribute_name, Shapes::ShapeRef.new(shape: TargetAttributeName, required: true, location_name: "TargetAttributeName"))
8235
+ TabularJobConfig.add_member(:sample_weight_attribute_name, Shapes::ShapeRef.new(shape: SampleWeightAttributeName, location_name: "SampleWeightAttributeName"))
8236
+ TabularJobConfig.struct_class = Types::TabularJobConfig
8237
+
8238
+ TabularResolvedAttributes.add_member(:problem_type, Shapes::ShapeRef.new(shape: ProblemType, location_name: "ProblemType"))
8239
+ TabularResolvedAttributes.struct_class = Types::TabularResolvedAttributes
8240
+
8203
8241
  Tag.add_member(:key, Shapes::ShapeRef.new(shape: TagKey, required: true, location_name: "Key"))
8204
8242
  Tag.add_member(:value, Shapes::ShapeRef.new(shape: TagValue, required: true, location_name: "Value"))
8205
8243
  Tag.struct_class = Types::Tag
@@ -1762,8 +1762,9 @@ module Aws::SageMaker
1762
1762
  # @!attribute [rw] inference_container_definitions
1763
1763
  # The mapping of all supported processing unit (CPU, GPU, etc...) to
1764
1764
  # inference container definitions for the candidate. This field is
1765
- # populated for the V2 API only (for example, for jobs created by
1766
- # calling `CreateAutoMLJobV2`).
1765
+ # populated for the AutoML jobs V2 (for example, for jobs created by
1766
+ # calling `CreateAutoMLJobV2`) related to image or text classification
1767
+ # problem types only.
1767
1768
  # @return [Hash<String,Array<Types::AutoMLContainerDefinition>>]
1768
1769
  #
1769
1770
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/AutoMLCandidate AWS API Documentation
@@ -2035,12 +2036,9 @@ module Aws::SageMaker
2035
2036
  # This structure specifies how to split the data into train and
2036
2037
  # validation datasets.
2037
2038
  #
2038
- # If you are using the V1 API (for example `CreateAutoMLJob`) or the V2
2039
- # API for Natural Language Processing problems (for example
2040
- # `CreateAutoMLJobV2` with a `TextClassificationJobConfig` problem
2041
- # type), the validation and training datasets must contain the same
2042
- # headers. Also, for V1 API jobs, the validation dataset must be less
2043
- # than 2 GB in size.
2039
+ # The validation and training datasets must contain the same headers.
2040
+ # For jobs created by calling `CreateAutoMLJob`, the validation dataset
2041
+ # must be less than 2 GB in size.
2044
2042
  #
2045
2043
  # @!attribute [rw] validation_fraction
2046
2044
  # The validation fraction (optional) is a float that specifies the
@@ -2077,14 +2075,12 @@ module Aws::SageMaker
2077
2075
  end
2078
2076
 
2079
2077
  # A channel is a named input source that training algorithms can
2080
- # consume. This channel is used for the non tabular training data of an
2081
- # AutoML job using the V2 API. For tabular training data, see [
2082
- # AutoMLChannel][1]. For more information, see [ Channel][2].
2078
+ # consume. This channel is used for AutoML jobs V2 (jobs created by
2079
+ # calling [CreateAutoMLJobV2][1]).
2083
2080
  #
2084
2081
  #
2085
2082
  #
2086
- # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLChannel.html
2087
- # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_Channel.html
2083
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
2088
2084
  #
2089
2085
  # @!attribute [rw] channel_type
2090
2086
  # The type of channel. Defines whether the data are used for training
@@ -2096,23 +2092,27 @@ module Aws::SageMaker
2096
2092
  # The content type of the data from the input source. The following
2097
2093
  # are the allowed content types for different problems:
2098
2094
  #
2099
- # * ImageClassification: `image/png`, `image/jpeg`, or `image/*`. The
2100
- # default value is `image/*`.
2095
+ # * For Tabular problem types: `text/csv;header=present` or
2096
+ # `x-application/vnd.amazon+parquet`. The default value is
2097
+ # `text/csv;header=present`.
2101
2098
  #
2102
- # * TextClassification: `text/csv;header=present` or
2099
+ # * For ImageClassification: `image/png`, `image/jpeg`, or `image/*`.
2100
+ # The default value is `image/*`.
2101
+ #
2102
+ # * For TextClassification: `text/csv;header=present` or
2103
2103
  # `x-application/vnd.amazon+parquet`. The default value is
2104
2104
  # `text/csv;header=present`.
2105
2105
  # @return [String]
2106
2106
  #
2107
2107
  # @!attribute [rw] compression_type
2108
- # The allowed compression types depend on the input format. We allow
2109
- # the compression type `Gzip` for `S3Prefix` inputs only. For all
2110
- # other inputs, the compression type should be `None`. If no
2111
- # compression type is provided, we default to `None`.
2108
+ # The allowed compression types depend on the input format and problem
2109
+ # type. We allow the compression type `Gzip` for `S3Prefix` inputs on
2110
+ # tabular data only. For all other inputs, the compression type should
2111
+ # be `None`. If no compression type is provided, we default to `None`.
2112
2112
  # @return [String]
2113
2113
  #
2114
2114
  # @!attribute [rw] data_source
2115
- # The data source for an AutoML channel.
2115
+ # The data source for an AutoML channel (Required).
2116
2116
  # @return [Types::AutoMLDataSource]
2117
2117
  #
2118
2118
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/AutoMLJobChannel AWS API Documentation
@@ -2132,7 +2132,7 @@ module Aws::SageMaker
2132
2132
  # @!attribute [rw] max_candidates
2133
2133
  # The maximum number of times a training job is allowed to run.
2134
2134
  #
2135
- # For V2 jobs (jobs created by calling `CreateAutoMLJobV2`), the
2135
+ # For job V2s (jobs created by calling `CreateAutoMLJobV2`), the
2136
2136
  # supported value is 1.
2137
2137
  # @return [Integer]
2138
2138
  #
@@ -2142,7 +2142,7 @@ module Aws::SageMaker
2142
2142
  # tuning job. For more information, see the [StoppingCondition][1]
2143
2143
  # used by the [CreateHyperParameterTuningJob][2] action.
2144
2144
  #
2145
- # For V2 jobs (jobs created by calling `CreateAutoMLJobV2`), this
2145
+ # For job V2s (jobs created by calling `CreateAutoMLJobV2`), this
2146
2146
  # field controls the runtime of the job candidate.
2147
2147
  #
2148
2148
  #
@@ -2221,7 +2221,7 @@ module Aws::SageMaker
2221
2221
  #
2222
2222
  #
2223
2223
  #
2224
- # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-suppprt
2224
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support
2225
2225
  # @return [String]
2226
2226
  #
2227
2227
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/AutoMLJobConfig AWS API Documentation
@@ -2237,8 +2237,6 @@ module Aws::SageMaker
2237
2237
  end
2238
2238
 
2239
2239
  # Specifies a metric to minimize or maximize as the objective of a job.
2240
- # V2 API jobs (for example jobs created by calling `CreateAutoMLJobV2`),
2241
- # support `Accuracy` only.
2242
2240
  #
2243
2241
  # @!attribute [rw] metric_name
2244
2242
  # The name of the objective metric used to measure the predictive
@@ -2253,11 +2251,15 @@ module Aws::SageMaker
2253
2251
  # If you do not specify a metric explicitly, the default behavior is
2254
2252
  # to automatically use:
2255
2253
  #
2256
- # * `MSE`: for regression.
2254
+ # * For tabular problem types:
2257
2255
  #
2258
- # * `F1`: for binary classification
2256
+ # * Regression: `MSE`.
2259
2257
  #
2260
- # * `Accuracy`: for multiclass classification.
2258
+ # * Binary classification: `F1`.
2259
+ #
2260
+ # * Multiclass classification: `Accuracy`.
2261
+ #
2262
+ # * For image or text classification problem types: `Accuracy`
2261
2263
  #
2262
2264
  #
2263
2265
  #
@@ -2375,28 +2377,34 @@ module Aws::SageMaker
2375
2377
  end
2376
2378
 
2377
2379
  # A collection of settings specific to the problem type used to
2378
- # configure an AutoML job using the V2 API. There must be one and only
2379
- # one config of the following type.
2380
+ # configure an AutoML job V2. There must be one and only one config of
2381
+ # the following type.
2380
2382
  #
2381
2383
  # @note AutoMLProblemTypeConfig is a union - when making an API calls you must set exactly one of the members.
2382
2384
  #
2383
2385
  # @note AutoMLProblemTypeConfig is a union - when returned from an API call exactly one value will be set and the returned type will be a subclass of AutoMLProblemTypeConfig corresponding to the set member.
2384
2386
  #
2385
2387
  # @!attribute [rw] image_classification_job_config
2386
- # Settings used to configure an AutoML job using the V2 API for the
2387
- # image classification problem type.
2388
+ # Settings used to configure an AutoML job V2 for the image
2389
+ # classification problem type.
2388
2390
  # @return [Types::ImageClassificationJobConfig]
2389
2391
  #
2390
2392
  # @!attribute [rw] text_classification_job_config
2391
- # Settings used to configure an AutoML job using the V2 API for the
2392
- # text classification problem type.
2393
+ # Settings used to configure an AutoML job V2 for the text
2394
+ # classification problem type.
2393
2395
  # @return [Types::TextClassificationJobConfig]
2394
2396
  #
2397
+ # @!attribute [rw] tabular_job_config
2398
+ # Settings used to configure an AutoML job V2 for a tabular problem
2399
+ # type (regression, classification).
2400
+ # @return [Types::TabularJobConfig]
2401
+ #
2395
2402
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/AutoMLProblemTypeConfig AWS API Documentation
2396
2403
  #
2397
2404
  class AutoMLProblemTypeConfig < Struct.new(
2398
2405
  :image_classification_job_config,
2399
2406
  :text_classification_job_config,
2407
+ :tabular_job_config,
2400
2408
  :unknown)
2401
2409
  SENSITIVE = []
2402
2410
  include Aws::Structure
@@ -2404,9 +2412,58 @@ module Aws::SageMaker
2404
2412
 
2405
2413
  class ImageClassificationJobConfig < AutoMLProblemTypeConfig; end
2406
2414
  class TextClassificationJobConfig < AutoMLProblemTypeConfig; end
2415
+ class TabularJobConfig < AutoMLProblemTypeConfig; end
2407
2416
  class Unknown < AutoMLProblemTypeConfig; end
2408
2417
  end
2409
2418
 
2419
+ # The resolved attributes specific to the problem type of an AutoML job
2420
+ # V2.
2421
+ #
2422
+ # @note AutoMLProblemTypeResolvedAttributes is a union - when returned from an API call exactly one value will be set and the returned type will be a subclass of AutoMLProblemTypeResolvedAttributes corresponding to the set member.
2423
+ #
2424
+ # @!attribute [rw] tabular_resolved_attributes
2425
+ # Defines the resolved attributes for the `TABULAR` problem type.
2426
+ # @return [Types::TabularResolvedAttributes]
2427
+ #
2428
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/AutoMLProblemTypeResolvedAttributes AWS API Documentation
2429
+ #
2430
+ class AutoMLProblemTypeResolvedAttributes < Struct.new(
2431
+ :tabular_resolved_attributes,
2432
+ :unknown)
2433
+ SENSITIVE = []
2434
+ include Aws::Structure
2435
+ include Aws::Structure::Union
2436
+
2437
+ class TabularResolvedAttributes < AutoMLProblemTypeResolvedAttributes; end
2438
+ class Unknown < AutoMLProblemTypeResolvedAttributes; end
2439
+ end
2440
+
2441
+ # The resolved attributes used to configure an AutoML job V2.
2442
+ #
2443
+ # @!attribute [rw] auto_ml_job_objective
2444
+ # Specifies a metric to minimize or maximize as the objective of a
2445
+ # job.
2446
+ # @return [Types::AutoMLJobObjective]
2447
+ #
2448
+ # @!attribute [rw] completion_criteria
2449
+ # How long a job is allowed to run, or how many candidates a job is
2450
+ # allowed to generate.
2451
+ # @return [Types::AutoMLJobCompletionCriteria]
2452
+ #
2453
+ # @!attribute [rw] auto_ml_problem_type_resolved_attributes
2454
+ # Defines the resolved attributes specific to a problem type.
2455
+ # @return [Types::AutoMLProblemTypeResolvedAttributes]
2456
+ #
2457
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/AutoMLResolvedAttributes AWS API Documentation
2458
+ #
2459
+ class AutoMLResolvedAttributes < Struct.new(
2460
+ :auto_ml_job_objective,
2461
+ :completion_criteria,
2462
+ :auto_ml_problem_type_resolved_attributes)
2463
+ SENSITIVE = []
2464
+ include Aws::Structure
2465
+ end
2466
+
2410
2467
  # Describes the Amazon S3 data source.
2411
2468
  #
2412
2469
  # @!attribute [rw] s3_data_type
@@ -2939,6 +2996,50 @@ module Aws::SageMaker
2939
2996
  include Aws::Structure
2940
2997
  end
2941
2998
 
2999
+ # Stores the configuration information for how model candidates are
3000
+ # generated using an AutoML job V2.
3001
+ #
3002
+ # @!attribute [rw] algorithms_config
3003
+ # Stores the configuration information for the selection of algorithms
3004
+ # used to train model candidates on tabular data.
3005
+ #
3006
+ # The list of available algorithms to choose from depends on the
3007
+ # training mode set in [ `TabularJobConfig.Mode` ][1].
3008
+ #
3009
+ # * `AlgorithmsConfig` should not be set in `AUTO` training mode.
3010
+ #
3011
+ # * When `AlgorithmsConfig` is provided, one `AutoMLAlgorithms`
3012
+ # attribute must be set and one only.
3013
+ #
3014
+ # If the list of algorithms provided as values for
3015
+ # `AutoMLAlgorithms` is empty, `CandidateGenerationConfig` uses the
3016
+ # full set of algorithms for the given training mode.
3017
+ #
3018
+ # * When `AlgorithmsConfig` is not provided,
3019
+ # `CandidateGenerationConfig` uses the full set of algorithms for
3020
+ # the given training mode.
3021
+ #
3022
+ # For the list of all algorithms per problem type and training mode,
3023
+ # see [ AutoMLAlgorithmConfig][2].
3024
+ #
3025
+ # For more information on each algorithm, see the [Algorithm
3026
+ # support][3] section in Autopilot developer guide.
3027
+ #
3028
+ #
3029
+ #
3030
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TabularJobConfig.html
3031
+ # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLAlgorithmConfig.html
3032
+ # [3]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support
3033
+ # @return [Array<Types::AutoMLAlgorithmConfig>]
3034
+ #
3035
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/CandidateGenerationConfig AWS API Documentation
3036
+ #
3037
+ class CandidateGenerationConfig < Struct.new(
3038
+ :algorithms_config)
3039
+ SENSITIVE = []
3040
+ include Aws::Structure
3041
+ end
3042
+
2942
3043
  # The properties of an AutoML candidate job.
2943
3044
  #
2944
3045
  # @!attribute [rw] candidate_artifact_locations
@@ -4525,15 +4626,13 @@ module Aws::SageMaker
4525
4626
  # @return [String]
4526
4627
  #
4527
4628
  # @!attribute [rw] auto_ml_job_objective
4528
- # Defines the objective metric used to measure the predictive quality
4529
- # of an AutoML job. You provide an [AutoMLJobObjective$MetricName][1]
4530
- # and Autopilot infers whether to minimize or maximize it. For
4531
- # [CreateAutoMLJobV2][2], only `Accuracy` is supported.
4629
+ # Specifies a metric to minimize or maximize as the objective of a
4630
+ # job. If not specified, the default objective metric depends on the
4631
+ # problem type. See [AutoMLJobObjective][1] for the default values.
4532
4632
  #
4533
4633
  #
4534
4634
  #
4535
4635
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLJobObjective.html
4536
- # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
4537
4636
  # @return [Types::AutoMLJobObjective]
4538
4637
  #
4539
4638
  # @!attribute [rw] auto_ml_job_config
@@ -4607,10 +4706,12 @@ module Aws::SageMaker
4607
4706
  # [InputDataConfig][1] supported by `CreateAutoMLJob`. The supported
4608
4707
  # formats depend on the problem type:
4609
4708
  #
4610
- # * ImageClassification: S3Prefix, `ManifestFile`,
4611
- # `AugmentedManifestFile`
4709
+ # * For Tabular problem types: `S3Prefix`, `ManifestFile`.
4612
4710
  #
4613
- # * TextClassification: S3Prefix
4711
+ # * For ImageClassification: `S3Prefix`, `ManifestFile`,
4712
+ # `AugmentedManifestFile`.
4713
+ #
4714
+ # * For TextClassification: `S3Prefix`.
4614
4715
  #
4615
4716
  #
4616
4717
  #
@@ -4625,6 +4726,13 @@ module Aws::SageMaker
4625
4726
  # @!attribute [rw] auto_ml_problem_type_config
4626
4727
  # Defines the configuration settings of one of the supported problem
4627
4728
  # types.
4729
+ #
4730
+ # <note markdown="1"> For tabular problem types, you must either specify the type of
4731
+ # supervised learning problem in `AutoMLProblemTypeConfig`
4732
+ # (`TabularJobConfig.ProblemType`) and provide the
4733
+ # `AutoMLJobObjective`, or none at all.
4734
+ #
4735
+ # </note>
4628
4736
  # @return [Types::AutoMLProblemTypeConfig]
4629
4737
  #
4630
4738
  # @!attribute [rw] role_arn
@@ -4649,11 +4757,20 @@ module Aws::SageMaker
4649
4757
  #
4650
4758
  # @!attribute [rw] auto_ml_job_objective
4651
4759
  # Specifies a metric to minimize or maximize as the objective of a
4652
- # job. For [CreateAutoMLJobV2][1], only `Accuracy` is supported.
4760
+ # job. If not specified, the default objective metric depends on the
4761
+ # problem type. For the list of default values per problem type, see
4762
+ # [AutoMLJobObjective][1].
4763
+ #
4764
+ # <note markdown="1"> For tabular problem types, you must either provide the
4765
+ # `AutoMLJobObjective` and indicate the type of supervised learning
4766
+ # problem in `AutoMLProblemTypeConfig`
4767
+ # (`TabularJobConfig.ProblemType`), or none.
4768
+ #
4769
+ # </note>
4653
4770
  #
4654
4771
  #
4655
4772
  #
4656
- # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateAutoMLJobV2.html
4773
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLJobObjective.html
4657
4774
  # @return [Types::AutoMLJobObjective]
4658
4775
  #
4659
4776
  # @!attribute [rw] model_deploy_config
@@ -4665,12 +4782,9 @@ module Aws::SageMaker
4665
4782
  # This structure specifies how to split the data into train and
4666
4783
  # validation datasets.
4667
4784
  #
4668
- # If you are using the V1 API (for example `CreateAutoMLJob`) or the
4669
- # V2 API for Natural Language Processing problems (for example
4670
- # `CreateAutoMLJobV2` with a `TextClassificationJobConfig` problem
4671
- # type), the validation and training datasets must contain the same
4672
- # headers. Also, for V1 API jobs, the validation dataset must be less
4673
- # than 2 GB in size.
4785
+ # The validation and training datasets must contain the same headers.
4786
+ # For jobs created by calling `CreateAutoMLJob`, the validation
4787
+ # dataset must be less than 2 GB in size.
4674
4788
  # @return [Types::AutoMLDataSplitConfig]
4675
4789
  #
4676
4790
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/CreateAutoMLJobV2Request AWS API Documentation
@@ -10636,8 +10750,7 @@ module Aws::SageMaker
10636
10750
  # @!attribute [rw] resolved_attributes
10637
10751
  # Contains `ProblemType`, `AutoMLJobObjective`, and
10638
10752
  # `CompletionCriteria`. If you do not provide these values, they are
10639
- # auto-inferred. If you do provide them, the values used are the ones
10640
- # you provide.
10753
+ # inferred.
10641
10754
  # @return [Types::ResolvedAttributes]
10642
10755
  #
10643
10756
  # @!attribute [rw] model_deploy_config
@@ -10678,7 +10791,7 @@ module Aws::SageMaker
10678
10791
  end
10679
10792
 
10680
10793
  # @!attribute [rw] auto_ml_job_name
10681
- # Requests information about an AutoML V2 job using its unique name.
10794
+ # Requests information about an AutoML job V2 using its unique name.
10682
10795
  # @return [String]
10683
10796
  #
10684
10797
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeAutoMLJobV2Request AWS API Documentation
@@ -10690,11 +10803,11 @@ module Aws::SageMaker
10690
10803
  end
10691
10804
 
10692
10805
  # @!attribute [rw] auto_ml_job_name
10693
- # Returns the name of the AutoML V2 job.
10806
+ # Returns the name of the AutoML job V2.
10694
10807
  # @return [String]
10695
10808
  #
10696
10809
  # @!attribute [rw] auto_ml_job_arn
10697
- # Returns the Amazon Resource Name (ARN) of the AutoML V2 job.
10810
+ # Returns the Amazon Resource Name (ARN) of the AutoML job V2.
10698
10811
  # @return [String]
10699
10812
  #
10700
10813
  # @!attribute [rw] auto_ml_job_input_data_config
@@ -10718,15 +10831,15 @@ module Aws::SageMaker
10718
10831
  #
10719
10832
  # @!attribute [rw] auto_ml_problem_type_config
10720
10833
  # Returns the configuration settings of the problem type set for the
10721
- # AutoML V2 job.
10834
+ # AutoML job V2.
10722
10835
  # @return [Types::AutoMLProblemTypeConfig]
10723
10836
  #
10724
10837
  # @!attribute [rw] creation_time
10725
- # Returns the creation time of the AutoML V2 job.
10838
+ # Returns the creation time of the AutoML job V2.
10726
10839
  # @return [Time]
10727
10840
  #
10728
10841
  # @!attribute [rw] end_time
10729
- # Returns the end time of the AutoML V2 job.
10842
+ # Returns the end time of the AutoML job V2.
10730
10843
  # @return [Time]
10731
10844
  #
10732
10845
  # @!attribute [rw] last_modified_time
@@ -10734,13 +10847,13 @@ module Aws::SageMaker
10734
10847
  # @return [Time]
10735
10848
  #
10736
10849
  # @!attribute [rw] failure_reason
10737
- # Returns the reason for the failure of the AutoML V2 job, when
10850
+ # Returns the reason for the failure of the AutoML job V2, when
10738
10851
  # applicable.
10739
10852
  # @return [String]
10740
10853
  #
10741
10854
  # @!attribute [rw] partial_failure_reasons
10742
- # Returns a list of reasons for partial failures within an AutoML V2
10743
- # job.
10855
+ # Returns a list of reasons for partial failures within an AutoML job
10856
+ # V2.
10744
10857
  # @return [Array<Types::AutoMLPartialFailureReason>]
10745
10858
  #
10746
10859
  # @!attribute [rw] best_candidate
@@ -10749,11 +10862,11 @@ module Aws::SageMaker
10749
10862
  # @return [Types::AutoMLCandidate]
10750
10863
  #
10751
10864
  # @!attribute [rw] auto_ml_job_status
10752
- # Returns the status of the AutoML V2 job.
10865
+ # Returns the status of the AutoML job V2.
10753
10866
  # @return [String]
10754
10867
  #
10755
10868
  # @!attribute [rw] auto_ml_job_secondary_status
10756
- # Returns the secondary status of the AutoML V2 job.
10869
+ # Returns the secondary status of the AutoML job V2.
10757
10870
  # @return [String]
10758
10871
  #
10759
10872
  # @!attribute [rw] model_deploy_config
@@ -10775,6 +10888,19 @@ module Aws::SageMaker
10775
10888
  # VPC settings.
10776
10889
  # @return [Types::AutoMLSecurityConfig]
10777
10890
  #
10891
+ # @!attribute [rw] auto_ml_job_artifacts
10892
+ # The artifacts that are generated during an AutoML job.
10893
+ # @return [Types::AutoMLJobArtifacts]
10894
+ #
10895
+ # @!attribute [rw] resolved_attributes
10896
+ # Returns the resolved attributes used by the AutoML job V2.
10897
+ # @return [Types::AutoMLResolvedAttributes]
10898
+ #
10899
+ # @!attribute [rw] auto_ml_problem_type_config_name
10900
+ # Returns the name of the problem type configuration set for the
10901
+ # AutoML job V2.
10902
+ # @return [String]
10903
+ #
10778
10904
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeAutoMLJobV2Response AWS API Documentation
10779
10905
  #
10780
10906
  class DescribeAutoMLJobV2Response < Struct.new(
@@ -10796,7 +10922,10 @@ module Aws::SageMaker
10796
10922
  :model_deploy_config,
10797
10923
  :model_deploy_result,
10798
10924
  :data_split_config,
10799
- :security_config)
10925
+ :security_config,
10926
+ :auto_ml_job_artifacts,
10927
+ :resolved_attributes,
10928
+ :auto_ml_problem_type_config_name)
10800
10929
  SENSITIVE = []
10801
10930
  include Aws::Structure
10802
10931
  end
@@ -20733,7 +20862,7 @@ module Aws::SageMaker
20733
20862
  end
20734
20863
 
20735
20864
  # Stores the configuration information for the image classification
20736
- # problem of an AutoML job using the V2 API.
20865
+ # problem of an AutoML job V2.
20737
20866
  #
20738
20867
  # @!attribute [rw] completion_criteria
20739
20868
  # How long a job is allowed to run, or how many candidates a job is
@@ -30828,6 +30957,11 @@ module Aws::SageMaker
30828
30957
  # using [TargetPlatform][1] fields. It can be used instead of
30829
30958
  # `TargetPlatform`.
30830
30959
  #
30960
+ # <note markdown="1"> Currently `ml_trn1` is available only in US East (N. Virginia)
30961
+ # Region, and `ml_inf2` is available only in US East (Ohio) Region.
30962
+ #
30963
+ # </note>
30964
+ #
30831
30965
  #
30832
30966
  #
30833
30967
  # [1]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TargetPlatform.html
@@ -34402,8 +34536,7 @@ module Aws::SageMaker
34402
34536
  #
34403
34537
  # @!attribute [rw] auto_ml_job_objective
34404
34538
  # Specifies a metric to minimize or maximize as the objective of a
34405
- # job. V2 API jobs (for example jobs created by calling
34406
- # `CreateAutoMLJobV2`), support `Accuracy` only.
34539
+ # job.
34407
34540
  # @return [Types::AutoMLJobObjective]
34408
34541
  #
34409
34542
  # @!attribute [rw] problem_type
@@ -36307,6 +36440,161 @@ module Aws::SageMaker
36307
36440
  include Aws::Structure
36308
36441
  end
36309
36442
 
36443
+ # The collection of settings used by an AutoML job V2 for the `TABULAR`
36444
+ # problem type.
36445
+ #
36446
+ # @!attribute [rw] candidate_generation_config
36447
+ # The configuration information of how model candidates are generated.
36448
+ # @return [Types::CandidateGenerationConfig]
36449
+ #
36450
+ # @!attribute [rw] completion_criteria
36451
+ # How long a job is allowed to run, or how many candidates a job is
36452
+ # allowed to generate.
36453
+ # @return [Types::AutoMLJobCompletionCriteria]
36454
+ #
36455
+ # @!attribute [rw] feature_specification_s3_uri
36456
+ # A URL to the Amazon S3 data source containing selected features from
36457
+ # the input data source to run an Autopilot job V2. You can input
36458
+ # `FeatureAttributeNames` (optional) in JSON format as shown below:
36459
+ #
36460
+ # `\{ "FeatureAttributeNames":["col1", "col2", ...] \}`.
36461
+ #
36462
+ # You can also specify the data type of the feature (optional) in the
36463
+ # format shown below:
36464
+ #
36465
+ # `\{ "FeatureDataTypes":\{"col1":"numeric", "col2":"categorical" ...
36466
+ # \} \}`
36467
+ #
36468
+ # <note markdown="1"> These column keys may not include the target column.
36469
+ #
36470
+ # </note>
36471
+ #
36472
+ # In ensembling mode, Autopilot only supports the following data
36473
+ # types: `numeric`, `categorical`, `text`, and `datetime`. In HPO
36474
+ # mode, Autopilot can support `numeric`, `categorical`, `text`,
36475
+ # `datetime`, and `sequence`.
36476
+ #
36477
+ # If only `FeatureDataTypes` is provided, the column keys (`col1`,
36478
+ # `col2`,..) should be a subset of the column names in the input data.
36479
+ #
36480
+ # If both `FeatureDataTypes` and `FeatureAttributeNames` are provided,
36481
+ # then the column keys should be a subset of the column names provided
36482
+ # in `FeatureAttributeNames`.
36483
+ #
36484
+ # The key name `FeatureAttributeNames` is fixed. The values listed in
36485
+ # `["col1", "col2", ...]` are case sensitive and should be a list of
36486
+ # strings containing unique values that are a subset of the column
36487
+ # names in the input data. The list of columns provided must not
36488
+ # include the target column.
36489
+ # @return [String]
36490
+ #
36491
+ # @!attribute [rw] mode
36492
+ # The method that Autopilot uses to train the data. You can either
36493
+ # specify the mode manually or let Autopilot choose for you based on
36494
+ # the dataset size by selecting `AUTO`. In `AUTO` mode, Autopilot
36495
+ # chooses `ENSEMBLING` for datasets smaller than 100 MB, and
36496
+ # `HYPERPARAMETER_TUNING` for larger ones.
36497
+ #
36498
+ # The `ENSEMBLING` mode uses a multi-stack ensemble model to predict
36499
+ # classification and regression tasks directly from your dataset. This
36500
+ # machine learning mode combines several base models to produce an
36501
+ # optimal predictive model. It then uses a stacking ensemble method to
36502
+ # combine predictions from contributing members. A multi-stack
36503
+ # ensemble model can provide better performance over a single model by
36504
+ # combining the predictive capabilities of multiple models. See
36505
+ # [Autopilot algorithm support][1] for a list of algorithms supported
36506
+ # by `ENSEMBLING` mode.
36507
+ #
36508
+ # The `HYPERPARAMETER_TUNING` (HPO) mode uses the best hyperparameters
36509
+ # to train the best version of a model. HPO automatically selects an
36510
+ # algorithm for the type of problem you want to solve. Then HPO finds
36511
+ # the best hyperparameters according to your objective metric. See
36512
+ # [Autopilot algorithm support][1] for a list of algorithms supported
36513
+ # by `HYPERPARAMETER_TUNING` mode.
36514
+ #
36515
+ #
36516
+ #
36517
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support
36518
+ # @return [String]
36519
+ #
36520
+ # @!attribute [rw] generate_candidate_definitions_only
36521
+ # Generates possible candidates without training the models. A model
36522
+ # candidate is a combination of data preprocessors, algorithms, and
36523
+ # algorithm parameter settings.
36524
+ # @return [Boolean]
36525
+ #
36526
+ # @!attribute [rw] problem_type
36527
+ # The type of supervised learning problem available for the model
36528
+ # candidates of the AutoML job V2. For more information, see [ Amazon
36529
+ # SageMaker Autopilot problem types][1].
36530
+ #
36531
+ #
36532
+ #
36533
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-datasets-problem-types.html#autopilot-problem-types
36534
+ # @return [String]
36535
+ #
36536
+ # @!attribute [rw] target_attribute_name
36537
+ # The name of the target variable in supervised learning, usually
36538
+ # represented by 'y'.
36539
+ # @return [String]
36540
+ #
36541
+ # @!attribute [rw] sample_weight_attribute_name
36542
+ # If specified, this column name indicates which column of the dataset
36543
+ # should be treated as sample weights for use by the objective metric
36544
+ # during the training, evaluation, and the selection of the best
36545
+ # model. This column is not considered as a predictive feature. For
36546
+ # more information on Autopilot metrics, see [Metrics and
36547
+ # validation][1].
36548
+ #
36549
+ # Sample weights should be numeric, non-negative, with larger values
36550
+ # indicating which rows are more important than others. Data points
36551
+ # that have invalid or no weight value are excluded.
36552
+ #
36553
+ # Support for sample weights is available in [Ensembling][2] mode
36554
+ # only.
36555
+ #
36556
+ #
36557
+ #
36558
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-metrics-validation.html
36559
+ # [2]: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLAlgorithmConfig.html
36560
+ # @return [String]
36561
+ #
36562
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TabularJobConfig AWS API Documentation
36563
+ #
36564
+ class TabularJobConfig < Struct.new(
36565
+ :candidate_generation_config,
36566
+ :completion_criteria,
36567
+ :feature_specification_s3_uri,
36568
+ :mode,
36569
+ :generate_candidate_definitions_only,
36570
+ :problem_type,
36571
+ :target_attribute_name,
36572
+ :sample_weight_attribute_name)
36573
+ SENSITIVE = []
36574
+ include Aws::Structure
36575
+ end
36576
+
36577
+ # The resolved attributes specific to the `TABULAR` problem type.
36578
+ #
36579
+ # @!attribute [rw] problem_type
36580
+ # The type of supervised learning problem available for the model
36581
+ # candidates of the AutoML job V2 (Binary Classification, Multiclass
36582
+ # Classification, Regression). For more information, see [ Amazon
36583
+ # SageMaker Autopilot problem types][1].
36584
+ #
36585
+ #
36586
+ #
36587
+ # [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-datasets-problem-types.html#autopilot-problem-types
36588
+ # @return [String]
36589
+ #
36590
+ # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TabularResolvedAttributes AWS API Documentation
36591
+ #
36592
+ class TabularResolvedAttributes < Struct.new(
36593
+ :problem_type)
36594
+ SENSITIVE = []
36595
+ include Aws::Structure
36596
+ end
36597
+
36310
36598
  # A tag object that consists of a key and an optional value, used to
36311
36599
  # manage metadata for SageMaker Amazon Web Services resources.
36312
36600
  #
@@ -36432,7 +36720,7 @@ module Aws::SageMaker
36432
36720
  end
36433
36721
 
36434
36722
  # Stores the configuration information for the text classification
36435
- # problem of an AutoML job using the V2 API.
36723
+ # problem of an AutoML job V2.
36436
36724
  #
36437
36725
  # @!attribute [rw] completion_criteria
36438
36726
  # How long a job is allowed to run, or how many candidates a job is
@@ -36441,12 +36729,13 @@ module Aws::SageMaker
36441
36729
  #
36442
36730
  # @!attribute [rw] content_column
36443
36731
  # The name of the column used to provide the sentences to be
36444
- # classified. It should not be the same as the target column.
36732
+ # classified. It should not be the same as the target column
36733
+ # (Required).
36445
36734
  # @return [String]
36446
36735
  #
36447
36736
  # @!attribute [rw] target_label_column
36448
36737
  # The name of the column used to provide the class labels. It should
36449
- # not be same as the content column.
36738
+ # not be same as the content column (Required).
36450
36739
  # @return [String]
36451
36740
  #
36452
36741
  # @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TextClassificationJobConfig AWS API Documentation
@@ -53,6 +53,6 @@ require_relative 'aws-sdk-sagemaker/customizations'
53
53
  # @!group service
54
54
  module Aws::SageMaker
55
55
 
56
- GEM_VERSION = '1.187.0'
56
+ GEM_VERSION = '1.188.0'
57
57
 
58
58
  end
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: 1.187.0
4
+ version: 1.188.0
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: 2023-06-15 00:00:00.000000000 Z
11
+ date: 2023-06-19 00:00:00.000000000 Z
12
12
  dependencies:
13
13
  - !ruby/object:Gem::Dependency
14
14
  name: aws-sdk-core