aws-sdk-forecastservice 1.20.0 → 1.21.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 697d13456570bd791ef364242e9342183ce9e0fdc58035f6113b12557f9bf7a4
-  data.tar.gz: 544c6abba89f7ad18f8130d70db89d5b2caa527a1fe628c24853abf6360dc3b7
+  metadata.gz: 74a8806673eb69e824936781c81d6be28393cea09b5ffb415ccf5814a99122f3
+  data.tar.gz: 76bbd8d1464800534c1d93420bf43cef63abbfefbe6045aa44ae8eca7e9bb3bb
 SHA512:
-  metadata.gz: b0aca3a9390d06310b672ad60aa5e3ec913b2022e480ecd3f1a2f276f5790f63f7bbce5ba3290d588501f61dab0c616eb8ccc51954b52c3abc5ea302625833b1
-  data.tar.gz: '013864157c95dcae1627f12bdcc3e38313814f7cfa8705c92967ab1a905bfe7bf875ed12a3404d1fc11ffa004edc421b918cea84bf021772e9cba3ead5eea556'
+  metadata.gz: 757e9ab6daa00f01ee5c38efbf294f6b641c6d5fcbd679200f4ec66a2a73752f36fef3ab531e6b4eeb42e7da8386c223358d80329e43f018a9cffb547da31d13
+  data.tar.gz: ee2c713042268fd9d38e54924b632b9d1c6aa8eecea71f390229dff63e4e4e820b13891b2bb2d9e185beefa93645767686c44fdbaf6f2fab4606e2f53b21ced3

data/CHANGELOG.md

@@ -1,6 +1,11 @@
 Unreleased Changes
 ------------------
 
+1.21.0 (2021-06-03)
+------------------
+
+* Feature - Added optional field AutoMLOverrideStrategy to CreatePredictor API that allows users to customize AutoML strategy. If provided in CreatePredictor request, this field is visible in DescribePredictor and GetAccuracyMetrics responses.
+
 1.20.0 (2021-05-21)
 ------------------
 

data/VERSION

@@ -1 +1 @@
-1.20.0
+1.21.0

data/lib/aws-sdk-forecastservice.rb

@@ -48,6 +48,6 @@ require_relative 'aws-sdk-forecastservice/customizations'
 # @!group service
 module Aws::ForecastService
 
-  GEM_VERSION = '1.20.0'
+  GEM_VERSION = '1.21.0'
 
 end

data/lib/aws-sdk-forecastservice/client.rb

@@ -1077,6 +1077,13 @@ module Aws::ForecastService
     #   This is a good option if you aren't sure which algorithm is suitable
     #   for your training data. In this case, `PerformHPO` must be false.
     #
+    # @option params [String] :auto_ml_override_strategy
+    #   Used to overide the default AutoML strategy, which is to optimize
+    #   predictor accuracy. To apply an AutoML strategy that minimizes
+    #   training time, use `LatencyOptimized`.
+    #
+    #   This parameter is only valid for predictors trained using AutoML.
+    #
     # @option params [Boolean] :perform_hpo
     #   Whether to perform hyperparameter optimization (HPO). HPO finds
     #   optimal hyperparameter values for your training data. The process of
@@ -1176,6 +1183,7 @@ module Aws::ForecastService
     #     forecast_horizon: 1, # required
     #     forecast_types: ["ForecastType"],
     #     perform_auto_ml: false,
+    #     auto_ml_override_strategy: "LatencyOptimized", # accepts LatencyOptimized
     #     perform_hpo: false,
     #     training_parameters: {
     #       "ParameterKey" => "ParameterValue",
@@ -1942,6 +1950,7 @@ module Aws::ForecastService
     #   * {Types::DescribePredictorResponse#forecast_horizon #forecast_horizon} => Integer
     #   * {Types::DescribePredictorResponse#forecast_types #forecast_types} => Array<String>
     #   * {Types::DescribePredictorResponse#perform_auto_ml #perform_auto_ml} => Boolean
+    #   * {Types::DescribePredictorResponse#auto_ml_override_strategy #auto_ml_override_strategy} => String
     #   * {Types::DescribePredictorResponse#perform_hpo #perform_hpo} => Boolean
     #   * {Types::DescribePredictorResponse#training_parameters #training_parameters} => Hash<String,String>
     #   * {Types::DescribePredictorResponse#evaluation_parameters #evaluation_parameters} => Types::EvaluationParameters
@@ -1973,6 +1982,7 @@ module Aws::ForecastService
     #   resp.forecast_types #=> Array
     #   resp.forecast_types[0] #=> String
     #   resp.perform_auto_ml #=> Boolean
+    #   resp.auto_ml_override_strategy #=> String, one of "LatencyOptimized"
     #   resp.perform_hpo #=> Boolean
     #   resp.training_parameters #=> Hash
     #   resp.training_parameters["ParameterKey"] #=> String
@@ -2124,6 +2134,7 @@ module Aws::ForecastService
     # @return [Types::GetAccuracyMetricsResponse] Returns a {Seahorse::Client::Response response} object which responds to the following methods:
     #
     #   * {Types::GetAccuracyMetricsResponse#predictor_evaluation_results #predictor_evaluation_results} => Array<Types::EvaluationResult>
+    #   * {Types::GetAccuracyMetricsResponse#auto_ml_override_strategy #auto_ml_override_strategy} => String
     #
     # @example Request syntax with placeholder values
     #
@@ -2148,6 +2159,7 @@ module Aws::ForecastService
     #   resp.predictor_evaluation_results[0].test_windows[0].metrics.error_metrics[0].forecast_type #=> String
     #   resp.predictor_evaluation_results[0].test_windows[0].metrics.error_metrics[0].wape #=> Float
     #   resp.predictor_evaluation_results[0].test_windows[0].metrics.error_metrics[0].rmse #=> Float
+    #   resp.auto_ml_override_strategy #=> String, one of "LatencyOptimized"
     #
     # @see http://docs.aws.amazon.com/goto/WebAPI/forecast-2018-06-26/GetAccuracyMetrics AWS API Documentation
     #
@@ -2875,7 +2887,7 @@ module Aws::ForecastService
         params: params,
         config: config)
       context[:gem_name] = 'aws-sdk-forecastservice'
-      context[:gem_version] = '1.20.0'
+      context[:gem_version] = '1.21.0'
       Seahorse::Client::Request.new(handlers, context)
     end
 

data/lib/aws-sdk-forecastservice/client_api.rb

@@ -16,6 +16,7 @@ module Aws::ForecastService
     Arn = Shapes::StringShape.new(name: 'Arn')
     ArnList = Shapes::ListShape.new(name: 'ArnList')
     AttributeType = Shapes::StringShape.new(name: 'AttributeType')
+    AutoMLOverrideStrategy = Shapes::StringShape.new(name: 'AutoMLOverrideStrategy')
     Boolean = Shapes::BooleanShape.new(name: 'Boolean')
     CategoricalParameterRange = Shapes::StructureShape.new(name: 'CategoricalParameterRange')
     CategoricalParameterRanges = Shapes::ListShape.new(name: 'CategoricalParameterRanges')
@@ -261,6 +262,7 @@ module Aws::ForecastService
     CreatePredictorRequest.add_member(:forecast_horizon, Shapes::ShapeRef.new(shape: Integer, required: true, location_name: "ForecastHorizon"))
     CreatePredictorRequest.add_member(:forecast_types, Shapes::ShapeRef.new(shape: ForecastTypes, location_name: "ForecastTypes"))
     CreatePredictorRequest.add_member(:perform_auto_ml, Shapes::ShapeRef.new(shape: Boolean, location_name: "PerformAutoML"))
+    CreatePredictorRequest.add_member(:auto_ml_override_strategy, Shapes::ShapeRef.new(shape: AutoMLOverrideStrategy, location_name: "AutoMLOverrideStrategy"))
     CreatePredictorRequest.add_member(:perform_hpo, Shapes::ShapeRef.new(shape: Boolean, location_name: "PerformHPO"))
     CreatePredictorRequest.add_member(:training_parameters, Shapes::ShapeRef.new(shape: TrainingParameters, location_name: "TrainingParameters"))
     CreatePredictorRequest.add_member(:evaluation_parameters, Shapes::ShapeRef.new(shape: EvaluationParameters, location_name: "EvaluationParameters"))
@@ -430,6 +432,7 @@ module Aws::ForecastService
     DescribePredictorResponse.add_member(:forecast_horizon, Shapes::ShapeRef.new(shape: Integer, location_name: "ForecastHorizon"))
     DescribePredictorResponse.add_member(:forecast_types, Shapes::ShapeRef.new(shape: ForecastTypes, location_name: "ForecastTypes"))
     DescribePredictorResponse.add_member(:perform_auto_ml, Shapes::ShapeRef.new(shape: Boolean, location_name: "PerformAutoML"))
+    DescribePredictorResponse.add_member(:auto_ml_override_strategy, Shapes::ShapeRef.new(shape: AutoMLOverrideStrategy, location_name: "AutoMLOverrideStrategy"))
     DescribePredictorResponse.add_member(:perform_hpo, Shapes::ShapeRef.new(shape: Boolean, location_name: "PerformHPO"))
     DescribePredictorResponse.add_member(:training_parameters, Shapes::ShapeRef.new(shape: TrainingParameters, location_name: "TrainingParameters"))
     DescribePredictorResponse.add_member(:evaluation_parameters, Shapes::ShapeRef.new(shape: EvaluationParameters, location_name: "EvaluationParameters"))
@@ -527,6 +530,7 @@ module Aws::ForecastService
     GetAccuracyMetricsRequest.struct_class = Types::GetAccuracyMetricsRequest
 
     GetAccuracyMetricsResponse.add_member(:predictor_evaluation_results, Shapes::ShapeRef.new(shape: PredictorEvaluationResults, location_name: "PredictorEvaluationResults"))
+    GetAccuracyMetricsResponse.add_member(:auto_ml_override_strategy, Shapes::ShapeRef.new(shape: AutoMLOverrideStrategy, location_name: "AutoMLOverrideStrategy"))
     GetAccuracyMetricsResponse.struct_class = Types::GetAccuracyMetricsResponse
 
     HyperParameterTuningJobConfig.add_member(:parameter_ranges, Shapes::ShapeRef.new(shape: ParameterRanges, location_name: "ParameterRanges"))

data/lib/aws-sdk-forecastservice/types.rb

@@ -782,6 +782,7 @@ module Aws::ForecastService
     #         forecast_horizon: 1, # required
     #         forecast_types: ["ForecastType"],
     #         perform_auto_ml: false,
+    #         auto_ml_override_strategy: "LatencyOptimized", # accepts LatencyOptimized
     #         perform_hpo: false,
     #         training_parameters: {
     #           "ParameterKey" => "ParameterValue",
@@ -913,6 +914,14 @@ module Aws::ForecastService
     #   false.
     #   @return [Boolean]
     #
+    # @!attribute [rw] auto_ml_override_strategy
+    #   Used to overide the default AutoML strategy, which is to optimize
+    #   predictor accuracy. To apply an AutoML strategy that minimizes
+    #   training time, use `LatencyOptimized`.
+    #
+    #   This parameter is only valid for predictors trained using AutoML.
+    #   @return [String]
+    #
     # @!attribute [rw] perform_hpo
     #   Whether to perform hyperparameter optimization (HPO). HPO finds
     #   optimal hyperparameter values for your training data. The process of
@@ -1017,6 +1026,7 @@ module Aws::ForecastService
       :forecast_horizon,
       :forecast_types,
       :perform_auto_ml,
+      :auto_ml_override_strategy,
       :perform_hpo,
       :training_parameters,
       :evaluation_parameters,
@@ -2064,6 +2074,14 @@ module Aws::ForecastService
     #   Whether the predictor is set to perform AutoML.
     #   @return [Boolean]
     #
+    # @!attribute [rw] auto_ml_override_strategy
+    #   The AutoML strategy used to train the predictor. Unless
+    #   `LatencyOptimized` is specified, the AutoML strategy optimizes
+    #   predictor accuracy.
+    #
+    #   This parameter is only valid for predictors trained using AutoML.
+    #   @return [String]
+    #
     # @!attribute [rw] perform_hpo
     #   Whether the predictor is set to perform hyperparameter optimization
     #   (HPO).
@@ -2172,6 +2190,7 @@ module Aws::ForecastService
       :forecast_horizon,
       :forecast_types,
       :perform_auto_ml,
+      :auto_ml_override_strategy,
       :perform_hpo,
       :training_parameters,
       :evaluation_parameters,
@@ -2740,10 +2759,19 @@ module Aws::ForecastService
     #   An array of results from evaluating the predictor.
     #   @return [Array<Types::EvaluationResult>]
     #
+    # @!attribute [rw] auto_ml_override_strategy
+    #   The AutoML strategy used to train the predictor. Unless
+    #   `LatencyOptimized` is specified, the AutoML strategy optimizes
+    #   predictor accuracy.
+    #
+    #   This parameter is only valid for predictors trained using AutoML.
+    #   @return [String]
+    #
     # @see http://docs.aws.amazon.com/goto/WebAPI/forecast-2018-06-26/GetAccuracyMetricsResponse AWS API Documentation
     #
     class GetAccuracyMetricsResponse < Struct.new(
-      :predictor_evaluation_results)
+      :predictor_evaluation_results,
+      :auto_ml_override_strategy)
       SENSITIVE = []
       include Aws::Structure
     end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: aws-sdk-forecastservice
 version: !ruby/object:Gem::Version
-  version: 1.20.0
+  version: 1.21.0
 platform: ruby
 authors:
 - Amazon Web Services
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-05-21 00:00:00.000000000 Z
+date: 2021-06-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: aws-sdk-core