Release Version 2.41

Models

  • Added HuberRegression and TweedieRegression.

  • Added costFunction parameter to the SupportVectorRegressionGradientVariant’s and SupportVectorMachineGradientVariant’s new() constructors.

    • Added SquaredHingeLoss, MeanPoissonDeviance and MeanGammaDeviance to the list of cost functions for NeuralNetwork, FactorizationMachine and FactorizedPairwiseInteraction.

  • Added Exponent to the list of activation functions for NeuralNetwork.

  • Added Exponent to the list of binary functions for FactorizationMachine and FactorizedPairwiseInteraction.

Others

  • Added PairwiseRankingDatasetCreator.
This site uses Just the Docs, a documentation theme for Jekyll.