aws-sdk-sagemaker 1.37.0 → 1.38.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/aws-sdk-sagemaker.rb +1 -1
- data/lib/aws-sdk-sagemaker/client.rb +48 -26
- data/lib/aws-sdk-sagemaker/client_api.rb +10 -0
- data/lib/aws-sdk-sagemaker/types.rb +222 -65
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 021e6103c08986b3ce6d16c9b304862da80334a1
|
4
|
+
data.tar.gz: 00610c16334164df1821f4059bc54a3fd437a41e
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 13070718845326b45cf17a97e5c1e9719cbd7f3f9278a9c605e0e631fb28144caa7843c49050f16348601bc1e50625622f5d07c809fccbbeffefb35a96f013f9
|
7
|
+
data.tar.gz: a2a084aabeeccd690c970bb6312605b78d7fff3634e821acb7d8e48467d639254c243291ca4bd150a4dacfbd91b6ae93c94870c6230138e1e3115b7fe1a7ec51
|
data/lib/aws-sdk-sagemaker.rb
CHANGED
@@ -665,7 +665,9 @@ module Aws::SageMaker
|
|
665
665
|
# and the target device the model runs on.
|
666
666
|
#
|
667
667
|
# @option params [required, Types::StoppingCondition] :stopping_condition
|
668
|
-
#
|
668
|
+
# Specifies a limit to how long a model compilation job can run. When
|
669
|
+
# the job reaches the time limit, Amazon SageMaker ends the compilation
|
670
|
+
# job. Use this API to cap model training costs.
|
669
671
|
#
|
670
672
|
# @return [Types::CreateCompilationJobResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
|
671
673
|
#
|
@@ -683,7 +685,7 @@ module Aws::SageMaker
|
|
683
685
|
# },
|
684
686
|
# output_config: { # required
|
685
687
|
# s3_output_location: "S3Uri", # required
|
686
|
-
# target_device: "lambda", # required, accepts lambda, ml_m4, ml_m5, ml_c4, ml_c5, ml_p2, ml_p3, jetson_tx1, jetson_tx2, jetson_nano, rasp3b, deeplens, rk3399, rk3288
|
688
|
+
# target_device: "lambda", # required, accepts lambda, ml_m4, ml_m5, ml_c4, ml_c5, ml_p2, ml_p3, jetson_tx1, jetson_tx2, jetson_nano, rasp3b, deeplens, rk3399, rk3288, sbe_c
|
687
689
|
# },
|
688
690
|
# stopping_condition: { # required
|
689
691
|
# max_runtime_in_seconds: 1,
|
@@ -1841,11 +1843,10 @@ module Aws::SageMaker
|
|
1841
1843
|
# showing the Jupyter server home page from the notebook instance. The
|
1842
1844
|
# console uses this API to get the URL and show the page.
|
1843
1845
|
#
|
1844
|
-
#
|
1845
|
-
#
|
1846
|
-
#
|
1847
|
-
#
|
1848
|
-
# Management user, group, or role used to access the notebook instance.
|
1846
|
+
# IAM authorization policies for this API are also enforced for every
|
1847
|
+
# HTTP request and WebSocket frame that attempts to connect to the
|
1848
|
+
# notebook instance.For example, you can restrict access to this API and
|
1849
|
+
# to the URL that it returns to a list of IP addresses that you specify.
|
1849
1850
|
# Use the `NotIpAddress` condition operator and the `aws:SourceIP`
|
1850
1851
|
# condition context key to specify the list of IP addresses that you
|
1851
1852
|
# want to have access to the notebook instance. For more information,
|
@@ -1906,9 +1907,10 @@ module Aws::SageMaker
|
|
1906
1907
|
# * `AlgorithmSpecification` - Identifies the training algorithm to use.
|
1907
1908
|
#
|
1908
1909
|
# * `HyperParameters` - Specify these algorithm-specific parameters to
|
1909
|
-
#
|
1910
|
-
#
|
1911
|
-
#
|
1910
|
+
# enable the estimation of model parameters during training.
|
1911
|
+
# Hyperparameters can be tuned to optimize this learning process. For
|
1912
|
+
# a list of hyperparameters for each training algorithm provided by
|
1913
|
+
# Amazon SageMaker, see [Algorithms][1].
|
1912
1914
|
#
|
1913
1915
|
# * `InputDataConfig` - Describes the training dataset and the Amazon S3
|
1914
1916
|
# location where it is stored.
|
@@ -1927,7 +1929,7 @@ module Aws::SageMaker
|
|
1927
1929
|
# must grant this role the necessary permissions so that Amazon
|
1928
1930
|
# SageMaker can successfully complete model training.
|
1929
1931
|
#
|
1930
|
-
# * `StoppingCondition` - Sets a
|
1932
|
+
# * `StoppingCondition` - Sets a time limit for training. Use this
|
1931
1933
|
# parameter to cap model training costs.
|
1932
1934
|
#
|
1933
1935
|
# For more information about Amazon SageMaker, see [How It Works][2].
|
@@ -2030,17 +2032,14 @@ module Aws::SageMaker
|
|
2030
2032
|
# [1]: https://docs.aws.amazon.com/sagemaker/latest/dg/train-vpc.html
|
2031
2033
|
#
|
2032
2034
|
# @option params [required, Types::StoppingCondition] :stopping_condition
|
2033
|
-
#
|
2034
|
-
#
|
2035
|
-
#
|
2036
|
-
#
|
2037
|
-
#
|
2038
|
-
#
|
2039
|
-
#
|
2040
|
-
#
|
2041
|
-
# the intermediate results of the job. This intermediate data is a valid
|
2042
|
-
# model artifact. You can use it to create a model using the
|
2043
|
-
# `CreateModel` API.
|
2035
|
+
# Specifies a limit to how long a model training job can run. When the
|
2036
|
+
# job reaches the time limit, Amazon SageMaker ends the training job.
|
2037
|
+
# Use this API to cap model training costs.
|
2038
|
+
#
|
2039
|
+
# To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
|
2040
|
+
# signal, which delays job termination for 120 seconds. Algorithms can
|
2041
|
+
# use this 120-second window to save the model artifacts, so the results
|
2042
|
+
# of training are not lost.
|
2044
2043
|
#
|
2045
2044
|
# @option params [Array<Types::Tag>] :tags
|
2046
2045
|
# An array of key-value pairs. For more information, see [Using Cost
|
@@ -2262,6 +2261,15 @@ module Aws::SageMaker
|
|
2262
2261
|
# Describes the resources, including ML instance types and ML instance
|
2263
2262
|
# count, to use for the transform job.
|
2264
2263
|
#
|
2264
|
+
# @option params [Types::DataProcessing] :data_processing
|
2265
|
+
# The data structure used for combining the input data and inference in
|
2266
|
+
# the output file. For more information, see [Batch Transform I/O
|
2267
|
+
# Join][1].
|
2268
|
+
#
|
2269
|
+
#
|
2270
|
+
#
|
2271
|
+
# [1]: http://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-io-join.html
|
2272
|
+
#
|
2265
2273
|
# @option params [Array<Types::Tag>] :tags
|
2266
2274
|
# (Optional) An array of key-value pairs. For more information, see
|
2267
2275
|
# [Using Cost Allocation Tags][1] in the *AWS Billing and Cost
|
@@ -2308,6 +2316,11 @@ module Aws::SageMaker
|
|
2308
2316
|
# instance_count: 1, # required
|
2309
2317
|
# volume_kms_key_id: "KmsKeyId",
|
2310
2318
|
# },
|
2319
|
+
# data_processing: {
|
2320
|
+
# input_filter: "JsonPath",
|
2321
|
+
# output_filter: "JsonPath",
|
2322
|
+
# join_source: "Input", # accepts Input, None
|
2323
|
+
# },
|
2311
2324
|
# tags: [
|
2312
2325
|
# {
|
2313
2326
|
# key: "TagKey", # required
|
@@ -2898,7 +2911,7 @@ module Aws::SageMaker
|
|
2898
2911
|
# resp.input_config.data_input_config #=> String
|
2899
2912
|
# resp.input_config.framework #=> String, one of "TENSORFLOW", "MXNET", "ONNX", "PYTORCH", "XGBOOST"
|
2900
2913
|
# resp.output_config.s3_output_location #=> String
|
2901
|
-
# resp.output_config.target_device #=> String, one of "lambda", "ml_m4", "ml_m5", "ml_c4", "ml_c5", "ml_p2", "ml_p3", "jetson_tx1", "jetson_tx2", "jetson_nano", "rasp3b", "deeplens", "rk3399", "rk3288"
|
2914
|
+
# resp.output_config.target_device #=> String, one of "lambda", "ml_m4", "ml_m5", "ml_c4", "ml_c5", "ml_p2", "ml_p3", "jetson_tx1", "jetson_tx2", "jetson_nano", "rasp3b", "deeplens", "rk3399", "rk3288", "sbe_c"
|
2902
2915
|
#
|
2903
2916
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeCompilationJob AWS API Documentation
|
2904
2917
|
#
|
@@ -3657,6 +3670,7 @@ module Aws::SageMaker
|
|
3657
3670
|
# * {Types::DescribeTransformJobResponse#transform_start_time #transform_start_time} => Time
|
3658
3671
|
# * {Types::DescribeTransformJobResponse#transform_end_time #transform_end_time} => Time
|
3659
3672
|
# * {Types::DescribeTransformJobResponse#labeling_job_arn #labeling_job_arn} => String
|
3673
|
+
# * {Types::DescribeTransformJobResponse#data_processing #data_processing} => Types::DataProcessing
|
3660
3674
|
#
|
3661
3675
|
# @example Request syntax with placeholder values
|
3662
3676
|
#
|
@@ -3692,6 +3706,9 @@ module Aws::SageMaker
|
|
3692
3706
|
# resp.transform_start_time #=> Time
|
3693
3707
|
# resp.transform_end_time #=> Time
|
3694
3708
|
# resp.labeling_job_arn #=> String
|
3709
|
+
# resp.data_processing.input_filter #=> String
|
3710
|
+
# resp.data_processing.output_filter #=> String
|
3711
|
+
# resp.data_processing.join_source #=> String, one of "Input", "None"
|
3695
3712
|
#
|
3696
3713
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeTransformJob AWS API Documentation
|
3697
3714
|
#
|
@@ -3999,7 +4016,7 @@ module Aws::SageMaker
|
|
3999
4016
|
# resp.compilation_job_summaries[0].creation_time #=> Time
|
4000
4017
|
# resp.compilation_job_summaries[0].compilation_start_time #=> Time
|
4001
4018
|
# resp.compilation_job_summaries[0].compilation_end_time #=> Time
|
4002
|
-
# resp.compilation_job_summaries[0].compilation_target_device #=> String, one of "lambda", "ml_m4", "ml_m5", "ml_c4", "ml_c5", "ml_p2", "ml_p3", "jetson_tx1", "jetson_tx2", "jetson_nano", "rasp3b", "deeplens", "rk3399", "rk3288"
|
4019
|
+
# resp.compilation_job_summaries[0].compilation_target_device #=> String, one of "lambda", "ml_m4", "ml_m5", "ml_c4", "ml_c5", "ml_p2", "ml_p3", "jetson_tx1", "jetson_tx2", "jetson_nano", "rasp3b", "deeplens", "rk3399", "rk3288", "sbe_c"
|
4003
4020
|
# resp.compilation_job_summaries[0].last_modified_time #=> Time
|
4004
4021
|
# resp.compilation_job_summaries[0].compilation_job_status #=> String, one of "INPROGRESS", "COMPLETED", "FAILED", "STARTING", "STOPPING", "STOPPED"
|
4005
4022
|
# resp.next_token #=> String
|
@@ -5694,7 +5711,12 @@ module Aws::SageMaker
|
|
5694
5711
|
#
|
5695
5712
|
# @option params [Integer] :volume_size_in_gb
|
5696
5713
|
# The size, in GB, of the ML storage volume to attach to the notebook
|
5697
|
-
# instance. The default value is 5 GB.
|
5714
|
+
# instance. The default value is 5 GB. ML storage volumes are encrypted,
|
5715
|
+
# so Amazon SageMaker can't determine the amount of available free
|
5716
|
+
# space on the volume. Because of this, you can increase the volume size
|
5717
|
+
# when you update a notebook instance, but you can't decrease the
|
5718
|
+
# volume size. If you want to decrease the size of the ML storage volume
|
5719
|
+
# in use, create a new notebook instance with the desired size.
|
5698
5720
|
#
|
5699
5721
|
# @option params [String] :default_code_repository
|
5700
5722
|
# The Git repository to associate with the notebook instance as its
|
@@ -5910,7 +5932,7 @@ module Aws::SageMaker
|
|
5910
5932
|
params: params,
|
5911
5933
|
config: config)
|
5912
5934
|
context[:gem_name] = 'aws-sdk-sagemaker'
|
5913
|
-
context[:gem_version] = '1.
|
5935
|
+
context[:gem_version] = '1.38.0'
|
5914
5936
|
Seahorse::Client::Request.new(handlers, context)
|
5915
5937
|
end
|
5916
5938
|
|
@@ -107,6 +107,7 @@ module Aws::SageMaker
|
|
107
107
|
CreateWorkteamResponse = Shapes::StructureShape.new(name: 'CreateWorkteamResponse')
|
108
108
|
CreationTime = Shapes::TimestampShape.new(name: 'CreationTime')
|
109
109
|
DataInputConfig = Shapes::StringShape.new(name: 'DataInputConfig')
|
110
|
+
DataProcessing = Shapes::StructureShape.new(name: 'DataProcessing')
|
110
111
|
DataSource = Shapes::StructureShape.new(name: 'DataSource')
|
111
112
|
DeleteAlgorithmInput = Shapes::StructureShape.new(name: 'DeleteAlgorithmInput')
|
112
113
|
DeleteCodeRepositoryInput = Shapes::StructureShape.new(name: 'DeleteCodeRepositoryInput')
|
@@ -227,6 +228,8 @@ module Aws::SageMaker
|
|
227
228
|
IntegerParameterRanges = Shapes::ListShape.new(name: 'IntegerParameterRanges')
|
228
229
|
JobReferenceCode = Shapes::StringShape.new(name: 'JobReferenceCode')
|
229
230
|
JobReferenceCodeContains = Shapes::StringShape.new(name: 'JobReferenceCodeContains')
|
231
|
+
JoinSource = Shapes::StringShape.new(name: 'JoinSource')
|
232
|
+
JsonPath = Shapes::StringShape.new(name: 'JsonPath')
|
230
233
|
KmsKeyId = Shapes::StringShape.new(name: 'KmsKeyId')
|
231
234
|
LabelAttributeName = Shapes::StringShape.new(name: 'LabelAttributeName')
|
232
235
|
LabelCounter = Shapes::IntegerShape.new(name: 'LabelCounter')
|
@@ -811,6 +814,7 @@ module Aws::SageMaker
|
|
811
814
|
CreateTransformJobRequest.add_member(:transform_input, Shapes::ShapeRef.new(shape: TransformInput, required: true, location_name: "TransformInput"))
|
812
815
|
CreateTransformJobRequest.add_member(:transform_output, Shapes::ShapeRef.new(shape: TransformOutput, required: true, location_name: "TransformOutput"))
|
813
816
|
CreateTransformJobRequest.add_member(:transform_resources, Shapes::ShapeRef.new(shape: TransformResources, required: true, location_name: "TransformResources"))
|
817
|
+
CreateTransformJobRequest.add_member(:data_processing, Shapes::ShapeRef.new(shape: DataProcessing, location_name: "DataProcessing"))
|
814
818
|
CreateTransformJobRequest.add_member(:tags, Shapes::ShapeRef.new(shape: TagList, location_name: "Tags"))
|
815
819
|
CreateTransformJobRequest.struct_class = Types::CreateTransformJobRequest
|
816
820
|
|
@@ -827,6 +831,11 @@ module Aws::SageMaker
|
|
827
831
|
CreateWorkteamResponse.add_member(:workteam_arn, Shapes::ShapeRef.new(shape: WorkteamArn, location_name: "WorkteamArn"))
|
828
832
|
CreateWorkteamResponse.struct_class = Types::CreateWorkteamResponse
|
829
833
|
|
834
|
+
DataProcessing.add_member(:input_filter, Shapes::ShapeRef.new(shape: JsonPath, location_name: "InputFilter"))
|
835
|
+
DataProcessing.add_member(:output_filter, Shapes::ShapeRef.new(shape: JsonPath, location_name: "OutputFilter"))
|
836
|
+
DataProcessing.add_member(:join_source, Shapes::ShapeRef.new(shape: JoinSource, location_name: "JoinSource"))
|
837
|
+
DataProcessing.struct_class = Types::DataProcessing
|
838
|
+
|
830
839
|
DataSource.add_member(:s3_data_source, Shapes::ShapeRef.new(shape: S3DataSource, location_name: "S3DataSource"))
|
831
840
|
DataSource.struct_class = Types::DataSource
|
832
841
|
|
@@ -1100,6 +1109,7 @@ module Aws::SageMaker
|
|
1100
1109
|
DescribeTransformJobResponse.add_member(:transform_start_time, Shapes::ShapeRef.new(shape: Timestamp, location_name: "TransformStartTime"))
|
1101
1110
|
DescribeTransformJobResponse.add_member(:transform_end_time, Shapes::ShapeRef.new(shape: Timestamp, location_name: "TransformEndTime"))
|
1102
1111
|
DescribeTransformJobResponse.add_member(:labeling_job_arn, Shapes::ShapeRef.new(shape: LabelingJobArn, location_name: "LabelingJobArn"))
|
1112
|
+
DescribeTransformJobResponse.add_member(:data_processing, Shapes::ShapeRef.new(shape: DataProcessing, location_name: "DataProcessing"))
|
1103
1113
|
DescribeTransformJobResponse.struct_class = Types::DescribeTransformJobResponse
|
1104
1114
|
|
1105
1115
|
DescribeWorkteamRequest.add_member(:workteam_name, Shapes::ShapeRef.new(shape: WorkteamName, required: true, location_name: "WorkteamName"))
|
@@ -933,7 +933,8 @@ module Aws::SageMaker
|
|
933
933
|
# @return [Hash<String,String>]
|
934
934
|
#
|
935
935
|
# @!attribute [rw] model_package_name
|
936
|
-
# The name of the model package to use
|
936
|
+
# The name or Amazon Resource Name (ARN) of the model package to use
|
937
|
+
# to create the model.
|
937
938
|
# @return [String]
|
938
939
|
#
|
939
940
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/ContainerDefinition AWS API Documentation
|
@@ -992,7 +993,7 @@ module Aws::SageMaker
|
|
992
993
|
#
|
993
994
|
# Logarithmic
|
994
995
|
#
|
995
|
-
# :
|
996
|
+
# : Hyperparameter tuning searches the values in the hyperparameter
|
996
997
|
# range by using a logarithmic scale.
|
997
998
|
#
|
998
999
|
# Logarithmic scaling works only for ranges that have only values
|
@@ -1334,7 +1335,7 @@ module Aws::SageMaker
|
|
1334
1335
|
# },
|
1335
1336
|
# output_config: { # required
|
1336
1337
|
# s3_output_location: "S3Uri", # required
|
1337
|
-
# target_device: "lambda", # required, accepts lambda, ml_m4, ml_m5, ml_c4, ml_c5, ml_p2, ml_p3, jetson_tx1, jetson_tx2, jetson_nano, rasp3b, deeplens, rk3399, rk3288
|
1338
|
+
# target_device: "lambda", # required, accepts lambda, ml_m4, ml_m5, ml_c4, ml_c5, ml_p2, ml_p3, jetson_tx1, jetson_tx2, jetson_nano, rasp3b, deeplens, rk3399, rk3288, sbe_c
|
1338
1339
|
# },
|
1339
1340
|
# stopping_condition: { # required
|
1340
1341
|
# max_runtime_in_seconds: 1,
|
@@ -1382,7 +1383,9 @@ module Aws::SageMaker
|
|
1382
1383
|
# @return [Types::OutputConfig]
|
1383
1384
|
#
|
1384
1385
|
# @!attribute [rw] stopping_condition
|
1385
|
-
#
|
1386
|
+
# Specifies a limit to how long a model compilation job can run. When
|
1387
|
+
# the job reaches the time limit, Amazon SageMaker ends the
|
1388
|
+
# compilation job. Use this API to cap model training costs.
|
1386
1389
|
# @return [Types::StoppingCondition]
|
1387
1390
|
#
|
1388
1391
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/CreateCompilationJobRequest AWS API Documentation
|
@@ -2640,17 +2643,14 @@ module Aws::SageMaker
|
|
2640
2643
|
# @return [Types::VpcConfig]
|
2641
2644
|
#
|
2642
2645
|
# @!attribute [rw] stopping_condition
|
2643
|
-
#
|
2644
|
-
#
|
2645
|
-
#
|
2646
|
-
# Algorithms might use this 120-second window to save the model
|
2647
|
-
# artifacts.
|
2646
|
+
# Specifies a limit to how long a model training job can run. When the
|
2647
|
+
# job reaches the time limit, Amazon SageMaker ends the training job.
|
2648
|
+
# Use this API to cap model training costs.
|
2648
2649
|
#
|
2649
|
-
#
|
2650
|
-
#
|
2651
|
-
#
|
2652
|
-
#
|
2653
|
-
# create a model using the `CreateModel` API.
|
2650
|
+
# To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
|
2651
|
+
# signal, which delays job termination for 120 seconds. Algorithms can
|
2652
|
+
# use this 120-second window to save the model artifacts, so the
|
2653
|
+
# results of training are not lost.
|
2654
2654
|
# @return [Types::StoppingCondition]
|
2655
2655
|
#
|
2656
2656
|
# @!attribute [rw] tags
|
@@ -2756,6 +2756,11 @@ module Aws::SageMaker
|
|
2756
2756
|
# instance_count: 1, # required
|
2757
2757
|
# volume_kms_key_id: "KmsKeyId",
|
2758
2758
|
# },
|
2759
|
+
# data_processing: {
|
2760
|
+
# input_filter: "JsonPath",
|
2761
|
+
# output_filter: "JsonPath",
|
2762
|
+
# join_source: "Input", # accepts Input, None
|
2763
|
+
# },
|
2759
2764
|
# tags: [
|
2760
2765
|
# {
|
2761
2766
|
# key: "TagKey", # required
|
@@ -2843,6 +2848,16 @@ module Aws::SageMaker
|
|
2843
2848
|
# count, to use for the transform job.
|
2844
2849
|
# @return [Types::TransformResources]
|
2845
2850
|
#
|
2851
|
+
# @!attribute [rw] data_processing
|
2852
|
+
# The data structure used for combining the input data and inference
|
2853
|
+
# in the output file. For more information, see [Batch Transform I/O
|
2854
|
+
# Join][1].
|
2855
|
+
#
|
2856
|
+
#
|
2857
|
+
#
|
2858
|
+
# [1]: http://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-io-join.html
|
2859
|
+
# @return [Types::DataProcessing]
|
2860
|
+
#
|
2846
2861
|
# @!attribute [rw] tags
|
2847
2862
|
# (Optional) An array of key-value pairs. For more information, see
|
2848
2863
|
# [Using Cost Allocation Tags][1] in the *AWS Billing and Cost
|
@@ -2865,6 +2880,7 @@ module Aws::SageMaker
|
|
2865
2880
|
:transform_input,
|
2866
2881
|
:transform_output,
|
2867
2882
|
:transform_resources,
|
2883
|
+
:data_processing,
|
2868
2884
|
:tags)
|
2869
2885
|
include Aws::Structure
|
2870
2886
|
end
|
@@ -2958,6 +2974,82 @@ module Aws::SageMaker
|
|
2958
2974
|
include Aws::Structure
|
2959
2975
|
end
|
2960
2976
|
|
2977
|
+
# The data structure used to combine the input data and transformed data
|
2978
|
+
# from the batch transform output into a joined dataset and to store it
|
2979
|
+
# in an output file. It also contains information on how to filter the
|
2980
|
+
# input data and the joined dataset. For more information, see [Batch
|
2981
|
+
# Transform I/O Join][1].
|
2982
|
+
#
|
2983
|
+
#
|
2984
|
+
#
|
2985
|
+
# [1]: http://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-io-join.html
|
2986
|
+
#
|
2987
|
+
# @note When making an API call, you may pass DataProcessing
|
2988
|
+
# data as a hash:
|
2989
|
+
#
|
2990
|
+
# {
|
2991
|
+
# input_filter: "JsonPath",
|
2992
|
+
# output_filter: "JsonPath",
|
2993
|
+
# join_source: "Input", # accepts Input, None
|
2994
|
+
# }
|
2995
|
+
#
|
2996
|
+
# @!attribute [rw] input_filter
|
2997
|
+
# A JSONPath expression used to select a portion of the input data to
|
2998
|
+
# pass to the algorithm. Use the `InputFilter` parameter to exclude
|
2999
|
+
# fields, such as an ID column, from the input. If you want Amazon
|
3000
|
+
# SageMaker to pass the entire input dataset to the algorithm, accept
|
3001
|
+
# the default value `$`.
|
3002
|
+
#
|
3003
|
+
# Examples: `"$"`, `"$[1:]"`, `"$.features"`
|
3004
|
+
# @return [String]
|
3005
|
+
#
|
3006
|
+
# @!attribute [rw] output_filter
|
3007
|
+
# A JSONPath expression used to select a portion of the joined dataset
|
3008
|
+
# to save in the output file for a batch transform job. If you want
|
3009
|
+
# Amazon SageMaker to store the entire input dataset in the output
|
3010
|
+
# file, leave the default value, `$`. If you specify indexes that
|
3011
|
+
# aren't within the dimension size of the joined dataset, you get an
|
3012
|
+
# error.
|
3013
|
+
#
|
3014
|
+
# Examples: `"$"`, `"$[0,5:]"`, `"$.['id','SageMakerOutput']"`
|
3015
|
+
# @return [String]
|
3016
|
+
#
|
3017
|
+
# @!attribute [rw] join_source
|
3018
|
+
# Specifies the source of the data to join with the transformed data.
|
3019
|
+
# The valid values are `None` and `Input` The default value is `None`
|
3020
|
+
# which specifies not to join the input with the transformed data. If
|
3021
|
+
# you want the batch transform job to join the original input data
|
3022
|
+
# with the transformed data, set `JoinSource` to `Input`. To join
|
3023
|
+
# input and output, the batch transform job must satisfy the
|
3024
|
+
# [Requirements for Using Batch Transform I/O Join][1].
|
3025
|
+
#
|
3026
|
+
# For JSON or JSONLines objects, such as a JSON array, Amazon
|
3027
|
+
# SageMaker adds the transformed data to the input JSON object in an
|
3028
|
+
# attribute called `SageMakerOutput`. The joined result for JSON must
|
3029
|
+
# be a key-value pair object. If the input is not a key-value pair
|
3030
|
+
# object, Amazon SageMaker creates a new JSON file. In the new JSON
|
3031
|
+
# file, and the input data is stored under the `SageMakerInput` key
|
3032
|
+
# and the results are stored in `SageMakerOutput`.
|
3033
|
+
#
|
3034
|
+
# For CSV files, Amazon SageMaker combines the transformed data with
|
3035
|
+
# the input data at the end of the input data and stores it in the
|
3036
|
+
# output file. The joined data has the joined input data followed by
|
3037
|
+
# the transformed data and the output is a CSV file.
|
3038
|
+
#
|
3039
|
+
#
|
3040
|
+
#
|
3041
|
+
# [1]: http://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-io-join.html#batch-transform-io-join-requirements
|
3042
|
+
# @return [String]
|
3043
|
+
#
|
3044
|
+
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DataProcessing AWS API Documentation
|
3045
|
+
#
|
3046
|
+
class DataProcessing < Struct.new(
|
3047
|
+
:input_filter,
|
3048
|
+
:output_filter,
|
3049
|
+
:join_source)
|
3050
|
+
include Aws::Structure
|
3051
|
+
end
|
3052
|
+
|
2961
3053
|
# Describes the location of the channel data.
|
2962
3054
|
#
|
2963
3055
|
# @note When making an API call, you may pass DataSource
|
@@ -3409,7 +3501,9 @@ module Aws::SageMaker
|
|
3409
3501
|
# @return [Time]
|
3410
3502
|
#
|
3411
3503
|
# @!attribute [rw] stopping_condition
|
3412
|
-
#
|
3504
|
+
# Specifies a limit to how long a model compilation job can run. When
|
3505
|
+
# the job reaches the time limit, Amazon SageMaker ends the
|
3506
|
+
# compilation job. Use this API to cap model training costs.
|
3413
3507
|
# @return [Types::StoppingCondition]
|
3414
3508
|
#
|
3415
3509
|
# @!attribute [rw] creation_time
|
@@ -4474,7 +4568,14 @@ module Aws::SageMaker
|
|
4474
4568
|
# @return [Types::VpcConfig]
|
4475
4569
|
#
|
4476
4570
|
# @!attribute [rw] stopping_condition
|
4477
|
-
#
|
4571
|
+
# Specifies a limit to how long a model training job can run. When the
|
4572
|
+
# job reaches the time limit, Amazon SageMaker ends the training job.
|
4573
|
+
# Use this API to cap model training costs.
|
4574
|
+
#
|
4575
|
+
# To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
|
4576
|
+
# signal, which delays job termination for 120 seconds. Algorithms can
|
4577
|
+
# use this 120-second window to save the model artifacts, so the
|
4578
|
+
# results of training are not lost.
|
4478
4579
|
# @return [Types::StoppingCondition]
|
4479
4580
|
#
|
4480
4581
|
# @!attribute [rw] creation_time
|
@@ -4535,7 +4636,7 @@ module Aws::SageMaker
|
|
4535
4636
|
# distributed training, choose `True`. Encryption provides greater
|
4536
4637
|
# security for distributed training, but training might take longer.
|
4537
4638
|
# How long it takes depends on the amount of communication between
|
4538
|
-
# compute instances, especially if you use a deep learning
|
4639
|
+
# compute instances, especially if you use a deep learning algorithms
|
4539
4640
|
# in distributed training.
|
4540
4641
|
# @return [Boolean]
|
4541
4642
|
#
|
@@ -4676,6 +4777,18 @@ module Aws::SageMaker
|
|
4676
4777
|
# labeling job that created the transform or training job.
|
4677
4778
|
# @return [String]
|
4678
4779
|
#
|
4780
|
+
# @!attribute [rw] data_processing
|
4781
|
+
# The data structure used to combine the input data and transformed
|
4782
|
+
# data from the batch transform output into a joined dataset and to
|
4783
|
+
# store it in an output file. It also contains information on how to
|
4784
|
+
# filter the input data and the joined dataset. For more information,
|
4785
|
+
# see [Batch Transform I/O Join][1].
|
4786
|
+
#
|
4787
|
+
#
|
4788
|
+
#
|
4789
|
+
# [1]: http://docs.aws.amazon.com/sagemaker/latest/dg/batch-transform-io-join.html
|
4790
|
+
# @return [Types::DataProcessing]
|
4791
|
+
#
|
4679
4792
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/DescribeTransformJobResponse AWS API Documentation
|
4680
4793
|
#
|
4681
4794
|
class DescribeTransformJobResponse < Struct.new(
|
@@ -4694,7 +4807,8 @@ module Aws::SageMaker
|
|
4694
4807
|
:creation_time,
|
4695
4808
|
:transform_start_time,
|
4696
4809
|
:transform_end_time,
|
4697
|
-
:labeling_job_arn
|
4810
|
+
:labeling_job_arn,
|
4811
|
+
:data_processing)
|
4698
4812
|
include Aws::Structure
|
4699
4813
|
end
|
4700
4814
|
|
@@ -5199,7 +5313,7 @@ module Aws::SageMaker
|
|
5199
5313
|
#
|
5200
5314
|
# * `arn:aws:lambda:eu-west-1:568282634449:function:PRE-TextMultiClass`
|
5201
5315
|
#
|
5202
|
-
# **Asia Pacific (Tokyo (ap-northeast-1):**
|
5316
|
+
# **Asia Pacific (Tokyo) (ap-northeast-1):**
|
5203
5317
|
#
|
5204
5318
|
# * `arn:aws:lambda:ap-northeast-1:477331159723:function:PRE-BoundingBox`
|
5205
5319
|
#
|
@@ -5209,7 +5323,7 @@ module Aws::SageMaker
|
|
5209
5323
|
#
|
5210
5324
|
# * `arn:aws:lambda:ap-northeast-1:477331159723:function:PRE-TextMultiClass`
|
5211
5325
|
#
|
5212
|
-
# **Asia Pacific (Sydney (ap-southeast-1):**
|
5326
|
+
# **Asia Pacific (Sydney) (ap-southeast-1):**
|
5213
5327
|
#
|
5214
5328
|
# * `arn:aws:lambda:ap-southeast-2:454466003867:function:PRE-BoundingBox`
|
5215
5329
|
#
|
@@ -5538,16 +5652,9 @@ module Aws::SageMaker
|
|
5538
5652
|
# @return [Types::ResourceConfig]
|
5539
5653
|
#
|
5540
5654
|
# @!attribute [rw] stopping_condition
|
5541
|
-
#
|
5542
|
-
#
|
5543
|
-
#
|
5544
|
-
# To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
|
5545
|
-
# signal. This delays job termination for 120 seconds. Algorithms
|
5546
|
-
# might use this 120-second window to save the model artifacts.
|
5547
|
-
#
|
5548
|
-
# When Amazon SageMaker terminates a job because the stopping
|
5549
|
-
# condition has been met, training algorithms provided by Amazon
|
5550
|
-
# SageMaker save the intermediate results of the job.
|
5655
|
+
# Specifies a limit to how long a model hyperparameter training job
|
5656
|
+
# can run. When the job reaches the time limit, Amazon SageMaker ends
|
5657
|
+
# the training job. Use this API to cap model training costs.
|
5551
5658
|
# @return [Types::StoppingCondition]
|
5552
5659
|
#
|
5553
5660
|
# @!attribute [rw] enable_network_isolation
|
@@ -6475,6 +6582,28 @@ module Aws::SageMaker
|
|
6475
6582
|
# @!attribute [rw] kms_key_id
|
6476
6583
|
# The AWS Key Management Service ID of the key used to encrypt the
|
6477
6584
|
# output data, if any.
|
6585
|
+
#
|
6586
|
+
# If you use a KMS key ID or an alias of your master key, the Amazon
|
6587
|
+
# SageMaker execution role must include permissions to call
|
6588
|
+
# `kms:Encrypt`. If you don't provide a KMS key ID, Amazon SageMaker
|
6589
|
+
# uses the default KMS key for Amazon S3 for your role's account.
|
6590
|
+
# Amazon SageMaker uses server-side encryption with KMS-managed keys
|
6591
|
+
# for `LabelingJobOutputConfig`. If you use a bucket policy with an
|
6592
|
+
# `s3:PutObject` permission that only allows objects with server-side
|
6593
|
+
# encryption, set the condition key of
|
6594
|
+
# `s3:x-amz-server-side-encryption` to `"aws:kms"`. For more
|
6595
|
+
# information, see [KMS-Managed Encryption Keys][1] in the *Amazon
|
6596
|
+
# Simple Storage Service Developer Guide.*
|
6597
|
+
#
|
6598
|
+
# The KMS key policy must grant permission to the IAM role that you
|
6599
|
+
# specify in your `CreateLabelingJob` request. For more information,
|
6600
|
+
# see [Using Key Policies in AWS KMS][2] in the *AWS Key Management
|
6601
|
+
# Service Developer Guide*.
|
6602
|
+
#
|
6603
|
+
#
|
6604
|
+
#
|
6605
|
+
# [1]: https://docs.aws.amazon.com/AmazonS3/latest/dev/UsingKMSEncryption.html
|
6606
|
+
# [2]: http://docs.aws.amazon.com/kms/latest/developerguide/key-policies.html
|
6478
6607
|
# @return [String]
|
6479
6608
|
#
|
6480
6609
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/LabelingJobOutputConfig AWS API Documentation
|
@@ -8855,7 +8984,7 @@ module Aws::SageMaker
|
|
8855
8984
|
#
|
8856
8985
|
# {
|
8857
8986
|
# s3_output_location: "S3Uri", # required
|
8858
|
-
# target_device: "lambda", # required, accepts lambda, ml_m4, ml_m5, ml_c4, ml_c5, ml_p2, ml_p3, jetson_tx1, jetson_tx2, jetson_nano, rasp3b, deeplens, rk3399, rk3288
|
8987
|
+
# target_device: "lambda", # required, accepts lambda, ml_m4, ml_m5, ml_c4, ml_c5, ml_p2, ml_p3, jetson_tx1, jetson_tx2, jetson_nano, rasp3b, deeplens, rk3399, rk3288, sbe_c
|
8859
8988
|
# }
|
8860
8989
|
#
|
8861
8990
|
# @!attribute [rw] s3_output_location
|
@@ -8909,14 +9038,22 @@ module Aws::SageMaker
|
|
8909
9038
|
#
|
8910
9039
|
# `"arn:aws:kms:us-west-2:111122223333:alias/ExampleAlias"`
|
8911
9040
|
#
|
8912
|
-
# If you
|
8913
|
-
#
|
9041
|
+
# If you use a KMS key ID or an alias of your master key, the Amazon
|
9042
|
+
# SageMaker execution role must include permissions to call
|
9043
|
+
# `kms:Encrypt`. If you don't provide a KMS key ID, Amazon SageMaker
|
9044
|
+
# uses the default KMS key for Amazon S3 for your role's account.
|
9045
|
+
# Amazon SageMaker uses server-side encryption with KMS-managed keys
|
9046
|
+
# for `OutputDataConfig`. If you use a bucket policy with an
|
9047
|
+
# `s3:PutObject` permission that only allows objects with server-side
|
9048
|
+
# encryption, set the condition key of
|
9049
|
+
# `s3:x-amz-server-side-encryption` to `"aws:kms"`. For more
|
8914
9050
|
# information, see [KMS-Managed Encryption Keys][1] in the *Amazon
|
8915
9051
|
# Simple Storage Service Developer Guide.*
|
8916
9052
|
#
|
8917
9053
|
# The KMS key policy must grant permission to the IAM role that you
|
8918
|
-
# specify in your `
|
8919
|
-
#
|
9054
|
+
# specify in your `CreateTrainingJob`, `CreateTransformJob`, or
|
9055
|
+
# `CreateHyperParameterTuningJob` requests. For more information, see
|
9056
|
+
# [Using Key Policies in AWS KMS][2] in the *AWS Key Management
|
8920
9057
|
# Service Developer Guide*.
|
8921
9058
|
#
|
8922
9059
|
#
|
@@ -9977,10 +10114,9 @@ module Aws::SageMaker
|
|
9977
10114
|
# examples, don't use status messages in if statements.
|
9978
10115
|
#
|
9979
10116
|
# To have an overview of your training job's progress, view
|
9980
|
-
# `TrainingJobStatus` and `SecondaryStatus` in
|
9981
|
-
#
|
9982
|
-
#
|
9983
|
-
# following:
|
10117
|
+
# `TrainingJobStatus` and `SecondaryStatus` in DescribeTrainingJob,
|
10118
|
+
# and `StatusMessage` together. For example, at the start of a
|
10119
|
+
# training job, you might see the following:
|
9984
10120
|
#
|
9985
10121
|
# * `TrainingJobStatus` - InProgress
|
9986
10122
|
#
|
@@ -10219,20 +10355,28 @@ module Aws::SageMaker
|
|
10219
10355
|
include Aws::Structure
|
10220
10356
|
end
|
10221
10357
|
|
10222
|
-
# Specifies how long model training
|
10223
|
-
# the limit, Amazon SageMaker ends the
|
10224
|
-
# model training
|
10358
|
+
# Specifies a limit to how long a model training or compilation job can
|
10359
|
+
# run. When the job reaches the time limit, Amazon SageMaker ends the
|
10360
|
+
# training or compilation job. Use this API to cap model training costs.
|
10225
10361
|
#
|
10226
10362
|
# To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
|
10227
|
-
# signal, which delays job termination
|
10363
|
+
# signal, which delays job termination for 120 seconds. Algorithms can
|
10228
10364
|
# use this 120-second window to save the model artifacts, so the results
|
10229
|
-
# of training
|
10365
|
+
# of training are not lost.
|
10366
|
+
#
|
10367
|
+
# The training algorithms provided by Amazon SageMaker automatically
|
10368
|
+
# save the intermediate results of a model training job when possible.
|
10369
|
+
# This attempt to save artifacts is only a best effort case as model
|
10370
|
+
# might not be in a state from which it can be saved. For example, if
|
10371
|
+
# training has just started, the model might not be ready to save. When
|
10372
|
+
# saved, this intermediate data is a valid model artifact. You can use
|
10373
|
+
# it to create a model with `CreateModel`.
|
10374
|
+
#
|
10375
|
+
# <note markdown="1"> The Neural Topic Model (NTM) currently does not support saving
|
10376
|
+
# intermediate model artifacts. When training NTMs, make sure that the
|
10377
|
+
# maximum runtime is sufficient for the training job to complete.
|
10230
10378
|
#
|
10231
|
-
#
|
10232
|
-
# the intermediate results of a model training job (it is best effort
|
10233
|
-
# case, as model might not be ready to save as some stages, for example
|
10234
|
-
# training just started). This intermediate data is a valid model
|
10235
|
-
# artifact. You can use it to create a model (`CreateModel`).
|
10379
|
+
# </note>
|
10236
10380
|
#
|
10237
10381
|
# @note When making an API call, you may pass StoppingCondition
|
10238
10382
|
# data as a hash:
|
@@ -10242,10 +10386,10 @@ module Aws::SageMaker
|
|
10242
10386
|
# }
|
10243
10387
|
#
|
10244
10388
|
# @!attribute [rw] max_runtime_in_seconds
|
10245
|
-
# The maximum length of time, in seconds, that the training
|
10246
|
-
# run. If
|
10247
|
-
# SageMaker ends the job. If value is not specified, default
|
10248
|
-
# 1 day.
|
10389
|
+
# The maximum length of time, in seconds, that the training or
|
10390
|
+
# compilation job can run. If job does not complete during this time,
|
10391
|
+
# Amazon SageMaker ends the job. If value is not specified, default
|
10392
|
+
# value is 1 day. The maximum value is 28 days.
|
10249
10393
|
# @return [Integer]
|
10250
10394
|
#
|
10251
10395
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/StoppingCondition AWS API Documentation
|
@@ -10483,7 +10627,14 @@ module Aws::SageMaker
|
|
10483
10627
|
# @return [Types::VpcConfig]
|
10484
10628
|
#
|
10485
10629
|
# @!attribute [rw] stopping_condition
|
10486
|
-
#
|
10630
|
+
# Specifies a limit to how long a model training job can run. When the
|
10631
|
+
# job reaches the time limit, Amazon SageMaker ends the training job.
|
10632
|
+
# Use this API to cap model training costs.
|
10633
|
+
#
|
10634
|
+
# To stop a job, Amazon SageMaker sends the algorithm the `SIGTERM`
|
10635
|
+
# signal, which delays job termination for 120 seconds. Algorithms can
|
10636
|
+
# use this 120-second window to save the model artifacts, so the
|
10637
|
+
# results of training are not lost.
|
10487
10638
|
# @return [Types::StoppingCondition]
|
10488
10639
|
#
|
10489
10640
|
# @!attribute [rw] creation_time
|
@@ -10661,12 +10812,13 @@ module Aws::SageMaker
|
|
10661
10812
|
# @return [Types::ResourceConfig]
|
10662
10813
|
#
|
10663
10814
|
# @!attribute [rw] stopping_condition
|
10664
|
-
#
|
10665
|
-
# training
|
10815
|
+
# Specifies a limit to how long a model training job can run. When the
|
10816
|
+
# job reaches the time limit, Amazon SageMaker ends the training job.
|
10817
|
+
# Use this API to cap model training costs.
|
10666
10818
|
#
|
10667
10819
|
# To stop a job, Amazon SageMaker sends the algorithm the SIGTERM
|
10668
|
-
# signal, which delays job termination for 120 seconds. Algorithms
|
10669
|
-
#
|
10820
|
+
# signal, which delays job termination for 120 seconds. Algorithms can
|
10821
|
+
# use this 120-second window to save the model artifacts.
|
10670
10822
|
# @return [Types::StoppingCondition]
|
10671
10823
|
#
|
10672
10824
|
# @see http://docs.aws.amazon.com/goto/WebAPI/sagemaker-2017-07-24/TrainingJobDefinition AWS API Documentation
|
@@ -11226,16 +11378,15 @@ module Aws::SageMaker
|
|
11226
11378
|
# }
|
11227
11379
|
#
|
11228
11380
|
# @!attribute [rw] instance_type
|
11229
|
-
# The ML compute instance type for the transform job.
|
11230
|
-
# built-in algorithms to transform moderately sized datasets,
|
11231
|
-
# ml.m4.xlarge or `ml.m5.large`
|
11232
|
-
# value for `InstanceType`.
|
11381
|
+
# The ML compute instance type for the transform job. If you are using
|
11382
|
+
# built-in algorithms to transform moderately sized datasets, we
|
11383
|
+
# recommend using ml.m4.xlarge or `ml.m5.large`instance types.
|
11233
11384
|
# @return [String]
|
11234
11385
|
#
|
11235
11386
|
# @!attribute [rw] instance_count
|
11236
11387
|
# The number of ML compute instances to use in the transform job. For
|
11237
|
-
# distributed transform,
|
11238
|
-
# value is `1`.
|
11388
|
+
# distributed transform jobs, specify a value greater than 1. The
|
11389
|
+
# default value is `1`.
|
11239
11390
|
# @return [Integer]
|
11240
11391
|
#
|
11241
11392
|
# @!attribute [rw] volume_kms_key_id
|
@@ -11594,7 +11745,13 @@ module Aws::SageMaker
|
|
11594
11745
|
#
|
11595
11746
|
# @!attribute [rw] volume_size_in_gb
|
11596
11747
|
# The size, in GB, of the ML storage volume to attach to the notebook
|
11597
|
-
# instance. The default value is 5 GB.
|
11748
|
+
# instance. The default value is 5 GB. ML storage volumes are
|
11749
|
+
# encrypted, so Amazon SageMaker can't determine the amount of
|
11750
|
+
# available free space on the volume. Because of this, you can
|
11751
|
+
# increase the volume size when you update a notebook instance, but
|
11752
|
+
# you can't decrease the volume size. If you want to decrease the
|
11753
|
+
# size of the ML storage volume in use, create a new notebook instance
|
11754
|
+
# with the desired size.
|
11598
11755
|
# @return [Integer]
|
11599
11756
|
#
|
11600
11757
|
# @!attribute [rw] default_code_repository
|
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.
|
4
|
+
version: 1.38.0
|
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: 2019-
|
11
|
+
date: 2019-06-11 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: aws-sdk-core
|