Uses of Class
org.tribuo.common.sgd.FMParameters
Package
Description
Provides an implementation of a classification factorization machine using Stochastic Gradient Descent.
Provides the base classes for models trained with stochastic gradient descent.
Provides an implementation of a multi-label classification factorization machine model using Stochastic Gradient Descent.
Provides an implementation of factorization machines for regression using Stochastic Gradient Descent.
-
Uses of FMParameters in org.tribuo.classification.sgd.fm
Modifier and TypeMethodDescriptionprotected FMClassificationModel
FMClassificationTrainer.createModel
(String name, ModelProvenance provenance, ImmutableFeatureMap featureMap, ImmutableOutputInfo<Label> outputInfo, FMParameters parameters) -
Uses of FMParameters in org.tribuo.common.sgd
Modifier and TypeMethodDescriptionFMParameters.copy()
protected FMParameters
AbstractFMTrainer.createParameters
(int numFeatures, int numOutputs, SplittableRandom localRNG) Constructs the trainable parameters object, in this case aFMParameters
containing a weight matrix for the feature weights and a series of weight matrices for the factorized feature representation.ModifierConstructorDescriptionprotected
AbstractFMModel
(String name, ModelProvenance provenance, ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<T> outputIDInfo, FMParameters parameters, boolean generatesProbabilities) Constructs a factorization machine model trained via SGD. -
Uses of FMParameters in org.tribuo.multilabel.sgd.fm
Modifier and TypeMethodDescriptionprotected FMMultiLabelModel
FMMultiLabelTrainer.createModel
(String name, ModelProvenance provenance, ImmutableFeatureMap featureMap, ImmutableOutputInfo<MultiLabel> outputInfo, FMParameters parameters) -
Uses of FMParameters in org.tribuo.regression.sgd.fm
Modifier and TypeMethodDescriptionprotected FMRegressionModel
FMRegressionTrainer.createModel
(String name, ModelProvenance provenance, ImmutableFeatureMap featureMap, ImmutableOutputInfo<Regressor> outputInfo, FMParameters parameters)