Copyright	(c) 2013-2023 Brendan Hay
License	Mozilla Public License, v. 2.0.
Maintainer	Brendan Hay
Stability	auto-generated
Portability	non-portable (GHC extensions)
Safe Haskell	Safe-Inferred
Language	Haskell2010

Amazonka.Forecast.Types.EvaluationParameters

Description

Documentation

data EvaluationParameters Source #

Parameters that define how to split a dataset into training data and testing data, and the number of iterations to perform. These parameters are specified in the predefined algorithms but you can override them in the CreatePredictor request.

See: newEvaluationParameters smart constructor.

Constructors

EvaluationParameters'

Fields

backTestWindowOffset :: Maybe Int
The point from the end of the dataset where you want to split the data for model training and testing (evaluation). Specify the value as the number of data points. The default is the value of the forecast horizon. BackTestWindowOffset can be used to mimic a past virtual forecast start date. This value must be greater than or equal to the forecast horizon and less than half of the TARGET_TIME_SERIES dataset length.
ForecastHorizon <= BackTestWindowOffset < 1/2 * TARGET_TIME_SERIES dataset length
numberOfBacktestWindows :: Maybe Int
The number of times to split the input data. The default is 1. Valid values are 1 through 5.

Instances

Instances details

FromJSON EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods parseJSON :: Value -> Parser EvaluationParameters # parseJSONList :: Value -> Parser [EvaluationParameters] #
ToJSON EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods toJSON :: EvaluationParameters -> Value # toEncoding :: EvaluationParameters -> Encoding # toJSONList :: [EvaluationParameters] -> Value # toEncodingList :: [EvaluationParameters] -> Encoding #
Generic EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Associated Types type Rep EvaluationParameters :: Type -> Type # Methods from :: EvaluationParameters -> Rep EvaluationParameters x # to :: Rep EvaluationParameters x -> EvaluationParameters #
Read EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods readsPrec :: Int -> ReadS EvaluationParameters # readList :: ReadS [EvaluationParameters] # readPrec :: ReadPrec EvaluationParameters # readListPrec :: ReadPrec [EvaluationParameters] #
Show EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods showsPrec :: Int -> EvaluationParameters -> ShowS # show :: EvaluationParameters -> String # showList :: [EvaluationParameters] -> ShowS #
NFData EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods rnf :: EvaluationParameters -> () #
Eq EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods (==) :: EvaluationParameters -> EvaluationParameters -> Bool # (/=) :: EvaluationParameters -> EvaluationParameters -> Bool #
Hashable EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters Methods hashWithSalt :: Int -> EvaluationParameters -> Int # hash :: EvaluationParameters -> Int #
type Rep EvaluationParameters Source #
Instance details Defined in Amazonka.Forecast.Types.EvaluationParameters type Rep EvaluationParameters = D1 ('MetaData "EvaluationParameters" "Amazonka.Forecast.Types.EvaluationParameters" "amazonka-forecast-2.0-HHvJwvxGrDPBJtUcnmLBqf" 'False) (C1 ('MetaCons "EvaluationParameters'" 'PrefixI 'True) (S1 ('MetaSel ('Just "backTestWindowOffset") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe Int)) :*: S1 ('MetaSel ('Just "numberOfBacktestWindows") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe Int))))

newEvaluationParameters :: EvaluationParameters Source #

Create a value of EvaluationParameters with all optional fields omitted.

Use generic-lens or optics to modify other optional fields.

The following record fields are available, with the corresponding lenses provided for backwards compatibility:

$sel:backTestWindowOffset:EvaluationParameters', evaluationParameters_backTestWindowOffset - The point from the end of the dataset where you want to split the data for model training and testing (evaluation). Specify the value as the number of data points. The default is the value of the forecast horizon. BackTestWindowOffset can be used to mimic a past virtual forecast start date. This value must be greater than or equal to the forecast horizon and less than half of the TARGET_TIME_SERIES dataset length.

ForecastHorizon <= BackTestWindowOffset < 1/2 * TARGET_TIME_SERIES dataset length

$sel:numberOfBacktestWindows:EvaluationParameters', evaluationParameters_numberOfBacktestWindows - The number of times to split the input data. The default is 1. Valid values are 1 through 5.

evaluationParameters_backTestWindowOffset :: Lens' EvaluationParameters (Maybe Int) Source #

The point from the end of the dataset where you want to split the data for model training and testing (evaluation). Specify the value as the number of data points. The default is the value of the forecast horizon. BackTestWindowOffset can be used to mimic a past virtual forecast start date. This value must be greater than or equal to the forecast horizon and less than half of the TARGET_TIME_SERIES dataset length.

ForecastHorizon <= BackTestWindowOffset < 1/2 * TARGET_TIME_SERIES dataset length

evaluationParameters_numberOfBacktestWindows :: Lens' EvaluationParameters (Maybe Int) Source #

The number of times to split the input data. The default is 1. Valid values are 1 through 5.