CategoricalResultFeature (JPMML evaluator 1.7.3 API)

All Superinterfaces:

ResultFeature

All Known Subinterfaces:

HasAffinity, HasAffinityRanking, HasConfidence, HasProbability

All Known Implementing Classes:

AffinityDistribution, AggregateProbabilityDistribution, ClusterAffinityDistribution, ConfidenceDistribution, NeuronProbabilityDistribution, NodeScoreDistribution, ProbabilityDistribution, SimpleRuleScoreDistribution, VoteProbabilityDistribution
```
public interface CategoricalResultFeature<K>
extends ResultFeature
```
A common superinterface for categorical result features.

Classification is a supervised learning task. The set of all categories is defined by the range of categories of the target field.

Clustering is an unsupervised learning task. The set of all categories is defined by the identifiers of reference entities. For clustering models this includes all clusters. For k-nearest neighbor models this includes k nearest training instances.

See Also:

TargetField.getCategories()

- Method Summary
  
  All Methods Instance Methods Abstract Methods
  Modifier and Type Method Description
  
  Set<K> getCategories()
  Gets the set of known categories.

- Method Detail
  - getCategories
```
Set<K> getCategories()
```
    Gets the set of known categories.
    
    The set of known categories either equals the set of all categories, or is a proper subset of it.