reset

Reset incremental normalizer model

Since R2026a

collapse all in page

Syntax

Normalizer = reset(Normalizer)

Description

Normalizer = reset(Normalizer) returns the incremental normalizer model Normalizer with reset normalization properties. The reset function resets the SumOfWeights, NumTrainingObservations, and Center properties. If ScaleData is true, the function also resets the Scale property.

example

Examples

collapse all

Open Live Script

Load Fisher's iris data set.

load fisheriris

Create a ZScoreNormalizer model object for simple-weighted normalization of streaming data. Specify to scale the data, and set the Center values as 0 and the SumOfWeights values as 1. Display the properties of the model object.

Normalizer = incrementalNormalizer("zscore",ScaleData=1, ...
    Center=1*zeros(size(meas,2),1),SumOfWeights=1);
details(Normalizer)
  incremental.preprocessing.ZScoreNormalizer with properties:

               SumOfWeights: [1 1 1 1]
                  ScaleData: 1
                     Center: [0 0 0 0]
                      Scale: [0 0 0 0]
             PredictorNames: ["x1"    "x2"    "x3"    "x4"]
                     IsWarm: 1
    NumTrainingObservations: 0
              NumPredictors: 4
               WarmupPeriod: 0
             TrainingPeriod: Inf
            UpdateFrequency: 1
      CategoricalPredictors: []

  Methods, Superclasses

Fit the model to the data.

Normalizer = fit(Normalizer,meas);

Display the property values of the fitted model.

details(Normalizer)
  incremental.preprocessing.ZScoreNormalizer with properties:

               SumOfWeights: [151 151 151 151]
                  ScaleData: 1
                     Center: [5.8046 3.0371 3.7331 1.1914]
                      Scale: [0.9493 0.4990 1.7799 0.7634]
             PredictorNames: ["x1"    "x2"    "x3"    "x4"]
                     IsWarm: 1
    NumTrainingObservations: 150
              NumPredictors: 4
               WarmupPeriod: 0
             TrainingPeriod: Inf
            UpdateFrequency: 1
      CategoricalPredictors: []

  Methods, Superclasses

The fit function updates the following properties of Normalizer: SumOfWeights, Center, Scale, and NumTrainingObserations.

Reset the model by calling the reset function.

Normalizer = reset(Normalizer);

Display the properties of the reset model.

details(Normalizer)
  incremental.preprocessing.ZScoreNormalizer with properties:

               SumOfWeights: [0 0 0 0]
                  ScaleData: 1
                     Center: [0 0 0 0]
                      Scale: [0 0 0 0]
             PredictorNames: ["x1"    "x2"    "x3"    "x4"]
                     IsWarm: 1
    NumTrainingObservations: 0
              NumPredictors: 4
               WarmupPeriod: 0
             TrainingPeriod: Inf
            UpdateFrequency: 1
      CategoricalPredictors: []

  Methods, Superclasses

The reset function resets the same properties that the fit function updates.

Input Arguments

collapse all

Incremental normalizer model, specified as a ZScoreNormalizer, ExponentiallyWeightedNormalizer, or ClassWeightedNormalizer model object. You create Normalizer by calling incrementalNormalizer.

Version History

Introduced in R2026a

See Also

| | | | | |