Copyright	(c) 2013-2021 Brendan Hay
License	Mozilla Public License, v. 2.0.
Maintainer	Brendan Hay <brendan.g.hay+amazonka@gmail.com>
Stability	auto-generated
Portability	non-portable (GHC extensions)
Safe Haskell	None

Amazonka.Forecast.Types.Metrics

Description

Synopsis

Documentation

data Metrics Source #

Provides metrics that are used to evaluate the performance of a predictor. This object is part of the WindowSummary object.

See: newMetrics smart constructor.

Constructors

Metrics'

Fields

errorMetrics :: Maybe [ErrorMetric]
Provides detailed error metrics for each forecast type. Metrics include root-mean square-error (RMSE), mean absolute percentage error (MAPE), mean absolute scaled error (MASE), and weighted average percentage error (WAPE).
rmse :: Maybe Double
The root-mean-square error (RMSE).
weightedQuantileLosses :: Maybe [WeightedQuantileLoss]
An array of weighted quantile losses. Quantiles divide a probability distribution into regions of equal probability. The distribution in this case is the loss function.
averageWeightedQuantileLoss :: Maybe Double
The average value of all weighted quantile losses.

Instances

Instances details

Eq Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Methods (==) :: Metrics -> Metrics -> Bool # (/=) :: Metrics -> Metrics -> Bool #
Read Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Methods readsPrec :: Int -> ReadS Metrics # readList :: ReadS [Metrics] # readPrec :: ReadPrec Metrics # readListPrec :: ReadPrec [Metrics] #
Show Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Methods showsPrec :: Int -> Metrics -> ShowS # show :: Metrics -> String # showList :: [Metrics] -> ShowS #
Generic Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Associated Types type Rep Metrics :: Type -> Type # Methods from :: Metrics -> Rep Metrics x # to :: Rep Metrics x -> Metrics #
NFData Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Methods rnf :: Metrics -> () #
Hashable Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Methods hashWithSalt :: Int -> Metrics -> Int # hash :: Metrics -> Int #
FromJSON Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics Methods parseJSON :: Value -> Parser Metrics # parseJSONList :: Value -> Parser [Metrics] #
type Rep Metrics Source #
Instance details Defined in Amazonka.Forecast.Types.Metrics type Rep Metrics = D1 ('MetaData "Metrics" "Amazonka.Forecast.Types.Metrics" "libZSservicesZSamazonka-forecastZSamazonka-forecast" 'False) (C1 ('MetaCons "Metrics'" 'PrefixI 'True) ((S1 ('MetaSel ('Just "errorMetrics") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe [ErrorMetric])) :: S1 ('MetaSel ('Just "rmse") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe Double))) :: (S1 ('MetaSel ('Just "weightedQuantileLosses") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe [WeightedQuantileLoss])) :*: S1 ('MetaSel ('Just "averageWeightedQuantileLoss") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedStrict) (Rec0 (Maybe Double)))))

newMetrics :: Metrics Source #

Create a value of Metrics 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:errorMetrics:Metrics', metrics_errorMetrics - Provides detailed error metrics for each forecast type. Metrics include root-mean square-error (RMSE), mean absolute percentage error (MAPE), mean absolute scaled error (MASE), and weighted average percentage error (WAPE).

$sel:rmse:Metrics', metrics_rmse - The root-mean-square error (RMSE).

$sel:weightedQuantileLosses:Metrics', metrics_weightedQuantileLosses - An array of weighted quantile losses. Quantiles divide a probability distribution into regions of equal probability. The distribution in this case is the loss function.

$sel:averageWeightedQuantileLoss:Metrics', metrics_averageWeightedQuantileLoss - The average value of all weighted quantile losses.

metrics_errorMetrics :: Lens' Metrics (Maybe [ErrorMetric]) Source #

Provides detailed error metrics for each forecast type. Metrics include root-mean square-error (RMSE), mean absolute percentage error (MAPE), mean absolute scaled error (MASE), and weighted average percentage error (WAPE).

metrics_rmse :: Lens' Metrics (Maybe Double) Source #

The root-mean-square error (RMSE).

metrics_weightedQuantileLosses :: Lens' Metrics (Maybe [WeightedQuantileLoss]) Source #

An array of weighted quantile losses. Quantiles divide a probability distribution into regions of equal probability. The distribution in this case is the loss function.

metrics_averageWeightedQuantileLoss :: Lens' Metrics (Maybe Double) Source #

The average value of all weighted quantile losses.