smile.regression

Interface OnlineRegression<T>

Type Parameters:

T - the type of input object

All Superinterfaces:

Regression<T>, java.io.Serializable, java.util.function.ToDoubleFunction<T>

All Known Implementing Classes:

LinearModel, MLP

public interface OnlineRegression<T>
extends Regression<T>

Regression model with online learning capability. Online learning is a model of induction that learns one instance at a time. More formally, an online algorithm proceeds in a sequence of trials.

Method Summary

Modifier and Type	Method and Description
default void	update(T[] x, double[] y) Updates the model with a (micro-)batch of new samples.
void	update(T x, double y) Online update the regression model with a new training instance.

Methods inherited from interface smile.regression.Regression

applyAsDouble, predict, predict

Method Detail

update

default void update(T[] x,
                    double[] y)

Updates the model with a (micro-)batch of new samples.

Parameters:

x - the training instances.

y - the target values.

update

void update(T x,
            double y)

Online update the regression model with a new training instance. In general, this method may be NOT multi-thread safe.

Parameters:

x - training instance.

y - response variable.