HingeLossNonconformityFunction (JCP API)

java.lang.Object
- se.hb.jcp.nc.ClassifierNonconformityFunctionBase
- - se.hb.jcp.nc.ClassProbabilityNonconformityFunctionBase
  - - se.hb.jcp.nc.HingeLossNonconformityFunction

All Implemented Interfaces:

java.io.Serializable, IClassifierInformation, IClassificationNonconformityFunction
```
public class HingeLossNonconformityFunction
extends ClassProbabilityNonconformityFunctionBase
implements java.io.Serializable
```
A hinge loss nonconformity function based on the predicted class probabilities given by a classifier.

Author:

anders.gidenstam(at)hb.se

See Also:

Serialized Form

Constructor Summary

Constructors
Constructor	Description
`HingeLossNonconformityFunction(double[] classes)`
`HingeLossNonconformityFunction(double[] classes, IClassProbabilityClassifier classifier)`

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method	Description
`IClassificationNonconformityFunction`	`fitNew(cern.colt.matrix.DoubleMatrix2D x, double[] y)`	Returns a new non-conformity function based on the same parameters as the current one initialized with the supplied data.

Methods inherited from class se.hb.jcp.nc.ClassProbabilityNonconformityFunctionBase
calculateNonConformityScore

Methods inherited from class se.hb.jcp.nc.ClassifierNonconformityFunctionBase
calc_nc, calc_nc, fit, fitNew, getAttributeCount, getClassifier, getLabels, isTrained, nativeStorageTemplate

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Constructor Detail
  - HingeLossNonconformityFunction
```
public HingeLossNonconformityFunction(double[] classes)
```
  - HingeLossNonconformityFunction
```
public HingeLossNonconformityFunction(double[] classes,
                                      IClassProbabilityClassifier classifier)
```
- Method Detail
  - fitNew
```
public IClassificationNonconformityFunction fitNew(cern.colt.matrix.DoubleMatrix2D x,
                                                   double[] y)
```
    Description copied from interface: IClassificationNonconformityFunction
    
    Returns a new non-conformity function based on the same parameters as the current one initialized with the supplied data.
    
    Specified by:
    
    fitNew in interface IClassificationNonconformityFunction
    
    Specified by:
    
    fitNew in class ClassifierNonconformityFunctionBase
    
    Parameters:
    
    x - the instances.
    
    y - the targets or classes of the instances.
    
    Returns:
    
    a new non-conformity function.