Package smile.classification

Interface Classifier<T>

Type Parameters:
T - the type of model input object.
All Superinterfaces:
Serializable, ToDoubleFunction<T>, ToIntFunction<T>
All Known Subinterfaces:
DataFrameClassifier
All Known Implementing Classes:
AbstractClassifier, AdaBoost, DecisionTree, DiscreteNaiveBayes, FLD, GradientTreeBoost, KNN, LDA, LogisticRegression, LogisticRegression.Binomial, LogisticRegression.Multinomial, Maxent, Maxent.Binomial, Maxent.Multinomial, MLP, NaiveBayes, OneVersusOne, OneVersusRest, QDA, RandomForest, RBFNetwork, RDA, SparseLogisticRegression, SparseLogisticRegression.Binomial, SparseLogisticRegression.Multinomial, SVM
public interface Classifier<T> extends ToIntFunction<T>, ToDoubleFunction<T>, Serializable
A classifier assigns an input object into one of a given number of categories. The input object is formally termed an instance, and the categories are termed classes. The instance is usually described by a vector of features, which together constitute a description of all known characteristics of the instance.

Classification normally refers to a supervised procedure, i.e. a procedure that produces an inferred function to predict the output value of new instances based on a training set of pairs consisting of an input object and a desired output value. The inferred function is called a classifier if the output is discrete or a regression function if the output is continuous.

  • Nested Class Summary

    Nested Classes
    Modifier and Type
    Interface
    Description
    static interface 
    Classifier.Trainer<T,M extends Classifier<T>>
    The classifier trainer.

  • Method Summary

    Modifier and Type
    Method
    Description
    default double
    applyAsDouble(T x)
     
    default int
    applyAsInt(T x)
     
    int[]
    classes()
    Returns the class labels.
    static <T> Classifier<T>
    ensemble(Classifier<T>... models)
    Return an ensemble of multiple base models to obtain better predictive performance.
    int
    numClasses()
    Returns the number of classes.
    default boolean
    online()
    Returns true if this is an online learner.
    default int[]
    predict(List<T> x)
    Predicts the class labels of a list of instances.
    default int[]
    predict(List<T> x, List<double[]> posteriori)
    Predicts the class labels of a list of instances.
    default int[]
    predict(Dataset<T> x)
    Predicts the class labels of a dataset.
    default int[]
    predict(Dataset<T> x, List<double[]> posteriori)
    Predicts the class labels of a dataset.
    int
    predict(T x)
    Predicts the class label of an instance.
    default int[]
    predict(T[] x)
    Predicts the class labels of an array of instances.
    default int[]
    predict(T[] x, double[][] posteriori)
    Predicts the class labels of an array of instances.
    default int
    predict(T x, double[] posteriori)
    Predicts the class label of an instance and also calculate a posteriori probabilities.
    default double
    score(T x)
    The raw prediction score.
    default boolean
    soft()
    Returns true if this is a soft classifier that can estimate the posteriori probabilities of classification.
    default void
    update(Dataset<Instance<T>> batch)
    Updates the model with a mini-batch of new samples.
    default void
    update(T[] x, int[] y)
    Updates the model with a mini-batch of new samples.
    default void
    update(T x, int y)
    Online update the classifier with a new training instance.

  • Method Details

    • numClasses

      int numClasses()
      Returns the number of classes.
      Returns:
      the number of classes.

    • classes

      int[] classes()
      Returns the class labels.
      Returns:
      the class labels.

    • predict

      int predict(T x)
      Predicts the class label of an instance.
      Parameters:
      x - the instance to be classified.
      Returns:
      the predicted class label.

    • score

      default double score(T x)
      The raw prediction score.
      Parameters:
      x - the instance to be classified.
      Returns:
      the raw prediction score.

    • applyAsInt

      default int applyAsInt(T x)
      Specified by:
      applyAsInt in interface ToIntFunction<T>

    • applyAsDouble

      default double applyAsDouble(T x)
      Specified by:
      applyAsDouble in interface ToDoubleFunction<T>

    • predict

      default int[] predict(T[] x)
      Predicts the class labels of an array of instances.
      Parameters:
      x - the instances to be classified.
      Returns:
      the predicted class labels.

    • predict

      default int[] predict(List<T> x)
      Predicts the class labels of a list of instances.
      Parameters:
      x - the instances to be classified.
      Returns:
      the predicted class labels.

    • predict

      default int[] predict(Dataset<T> x)
      Predicts the class labels of a dataset.
      Parameters:
      x - the dataset to be classified.
      Returns:
      the predicted class labels.

    • soft

      default boolean soft()
      Returns true if this is a soft classifier that can estimate the posteriori probabilities of classification.
      Returns:
      true if soft classifier.

    • predict

      default int predict(T x, double[] posteriori)
      Predicts the class label of an instance and also calculate a posteriori probabilities. Classifiers may NOT support this method since not all classification algorithms are able to calculate such a posteriori probabilities.
      Parameters:
      x - an instance to be classified.
      posteriori - a posteriori probabilities on output.
      Returns:
      the predicted class label

    • predict

      default int[] predict(T[] x, double[][] posteriori)
      Predicts the class labels of an array of instances.
      Parameters:
      x - the instances to be classified.
      posteriori - a posteriori probabilities on output.
      Returns:
      the predicted class labels.

    • predict

      default int[] predict(List<T> x, List<double[]> posteriori)
      Predicts the class labels of a list of instances.
      Parameters:
      x - the instances to be classified.
      posteriori - an empty list to store a posteriori probabilities on output.
      Returns:
      the predicted class labels.

    • predict

      default int[] predict(Dataset<T> x, List<double[]> posteriori)
      Predicts the class labels of a dataset.
      Parameters:
      x - the dataset to be classified.
      posteriori - an empty list to store a posteriori probabilities on output.
      Returns:
      the predicted class labels.

    • online

      default boolean online()
      Returns true if this is an online learner.
      Returns:
      true if online learner.

    • update

      default void update(T x, int y)
      Online update the classifier with a new training instance. In general, this method may be NOT multi-thread safe.
      Parameters:
      x - the training instance.
      y - the training label.

    • update

      default void update(T[] x, int[] y)
      Updates the model with a mini-batch of new samples.
      Parameters:
      x - the training instances.
      y - the training labels.

    • update

      default void update(Dataset<Instance<T>> batch)
      Updates the model with a mini-batch of new samples.
      Parameters:
      batch - the training instances.

    • ensemble

      @SafeVarargs static <T> Classifier<T> ensemble(Classifier<T>... models)
      Return an ensemble of multiple base models to obtain better predictive performance.
      Type Parameters:
      T - the type of model input object.
      Parameters:
      models - the base models.
      Returns:
      the ensemble model.