<!--
{
  "availability" : [
    "macOS: 10.15.0 - 14.0.0"
  ],
  "documentType" : "symbol",
  "framework" : "CreateML",
  "identifier" : "/documentation/CreateML/MLSupportVectorClassifier/ModelParameters-swift.struct/init(validation:maxIterations:penalty:convergenceThreshold:featureRescaling:)",
  "metadataVersion" : "0.1.0",
  "role" : "Initializer",
  "symbol" : {
    "kind" : "Initializer",
    "modules" : [
      "Create ML"
    ],
    "preciseIdentifier" : "s:8CreateML25MLSupportVectorClassifierV15ModelParametersV10validation13maxIterations7penalty20convergenceThreshold16featureRescalingA2E14ValidationDataO_SiS2dSbtcfc"
  },
  "title" : "init(validation:maxIterations:penalty:convergenceThreshold:featureRescaling:)"
}
-->

# init(validation:maxIterations:penalty:convergenceThreshold:featureRescaling:)

Creates a new set of parameters.

```
init(validation: MLSupportVectorClassifier.ModelParameters.ValidationData = .split(strategy: .automatic), maxIterations: Int = 11, penalty: Double = 1.0, convergenceThreshold: Double = 0.01, featureRescaling: Bool = true)
```

## Parameters

`validation`

The data used to monitor how well the model is generalizing.

    The default is to automatically split off some data from the training set for validation.

`maxIterations`

The maximum number of passes through the data.

    The default value is 11.

`penalty`

Weight of the regularizer. The larger the penalty the less variance in the model.

    The default value is 1.0.

`convergenceThreshold`

The threshold with which to determine if the model has converged. Consider
reducing this value for higher training accuracy, but beware of overfitting.

    The default value is 0.01.

`featureRescaling`

Determines if the features should be preprocessed to ensure all features are on the
same scale.

    The default value is true.

---

Copyright &copy; 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)