Package org.tribuo.common.sgd

Provides the base classes for models trained with stochastic gradient descent.

See: Description

Interface Summary

Interface	Description
SGDObjective<T>	An interface for a loss function that can produce the loss and gradient incurred by a single prediction.

Class Summary

Class	Description
AbstractLinearSGDModel<T extends Output<T>>
AbstractLinearSGDTrainer<T extends Output<T>,U>	A trainer for a linear model which uses SGD.
AbstractSGDModel<T extends Output<T>>
AbstractSGDModel.PredAndActive	A nominal tuple used to capture the prediction and the number of active features used by the model.
AbstractSGDTrainer<T extends Output<T>,U,V extends Model<T>,X extends FeedForwardParameters>	A trainer for a model which uses SGD.

Package org.tribuo.common.sgd Description

Provides the base classes for models trained with stochastic gradient descent.