Module org.nasdanika.ai

Package org.nasdanika.ai

Interface FittedPredictor.Fitter<F,L,E>

All Known Implementing Classes:

AbstractDoubleFitter, AbstractMapReduceDoubleFitter, AbstractRecursiveDoubleFitter

Enclosing interface:

FittedPredictor<F,L,E>

public static interface FittedPredictor.Fitter<F,L,E>

Method Summary

Modifier and Type	Method	Description
default <G> FittedPredictor.Fitter<G,L,E>	adaptFeature(Function<G,F> featureMapper)
default <G> FittedPredictor.Fitter<G,L,E>	adaptFeatureAsync(Function<G,reactor.core.publisher.Mono<F>> featureMapper)
default <M> FittedPredictor.Fitter<F,M,E>	adaptLabel(Function<M,L> fitMapper, Function<L,M> predictMapper)
default <M> FittedPredictor.Fitter<F,M,E>	adaptLabelAsync(Function<M,reactor.core.publisher.Mono<L>> fitMapper, Function<L,reactor.core.publisher.Mono<M>> predictMapper)
default FittedPredictor.Fitter<F,L,E>	compose(FittedPredictor.Fitter<F,L,E> other, BinaryOperator<L> add, BinaryOperator<L> subtract, FittedPredictor.ErrorComputer<F,L,E> errorComputer)	Composes two predictors by fitting this one, then computing label difference between this predictor predictions and labels and fitting the next one with the difference.
default <S> FittedPredictor<F,L,E>	fit(Collection<S> samples, Function<S,F> featureMapper, Function<S,L> labelMapper)	Creates a predictor by fitting a collection of samples.
<S> reactor.core.publisher.Mono<FittedPredictor<F,L,E>>	fitAsync(reactor.core.publisher.Flux<S> samples, Function<S,reactor.core.publisher.Mono<F>> featureMapper, Function<S,reactor.core.publisher.Mono<L>> labelMapper)	Creates a predictor by fitting a flux of samples.

Method Details

fit

default <S> FittedPredictor<F,L,E> fit(Collection<S> samples,
 Function<S,F> featureMapper,
 Function<S,L> labelMapper)

Creates a predictor by fitting a collection of samples.

Type Parameters:

S -

Parameters:

samples -

featureMapper -

labelMapper -

Returns:

fitAsync

<S>
reactor.core.publisher.Mono<FittedPredictor<F,L,E>> fitAsync(reactor.core.publisher.Flux<S> samples,
 Function<S,reactor.core.publisher.Mono<F>> featureMapper,
 Function<S,reactor.core.publisher.Mono<L>> labelMapper)

Creates a predictor by fitting a flux of samples.

Type Parameters:

S -

Parameters:

samples -

featureMapper -

labelMapper -

Returns:

A mono providing a predictor. The mono can publish a predictor before the flux is finished and update the predictor with new samples from the flux.

adaptFeature

default <G> FittedPredictor.Fitter<G,L,E> adaptFeature(Function<G,F> featureMapper)

adaptFeatureAsync

default <G> FittedPredictor.Fitter<G,L,E> adaptFeatureAsync(Function<G,reactor.core.publisher.Mono<F>> featureMapper)

adaptLabel

default <M> FittedPredictor.Fitter<F,M,E> adaptLabel(Function<M,L> fitMapper,
 Function<L,M> predictMapper)

adaptLabelAsync

default <M> FittedPredictor.Fitter<F,M,E> adaptLabelAsync(Function<M,reactor.core.publisher.Mono<L>> fitMapper,
 Function<L,reactor.core.publisher.Mono<M>> predictMapper)

compose

default FittedPredictor.Fitter<F,L,E> compose(FittedPredictor.Fitter<F,L,E> other,
 BinaryOperator<L> add,
 BinaryOperator<L> subtract,
 FittedPredictor.ErrorComputer<F,L,E> errorComputer)

Composes two predictors by fitting this one, then computing label difference between this predictor predictions and labels and fitting the next one with the difference. Prediction is done by computing prediction for this one, then the other and adding the two together.

Parameters:

other -

add -

subtract -

Returns: