Class MaxEntOptimizableByLabelLikelihood

    • Constructor Detail

      • MaxEntOptimizableByLabelLikelihood

        public MaxEntOptimizableByLabelLikelihood()
      • MaxEntOptimizableByLabelLikelihood

        public MaxEntOptimizableByLabelLikelihood​(InstanceList trainingSet,
                                                  MaxEnt initialClassifier)
    • Method Detail

      • getClassifier

        public MaxEnt getClassifier()
      • setParameter

        public void setParameter​(int index,
                                 double v)
        Specified by:
        setParameter in interface Optimizable
      • getValueGradientCalls

        public int getValueGradientCalls()
        Counts how many times this trainer has computed the gradient of the log probability of training labels.
      • getValueCalls

        public int getValueCalls()
        Counts how many times this trainer has computed the log probability of training labels.
      • useNoPrior

        public MaxEntOptimizableByLabelLikelihood useNoPrior()
        In some cases a prior term is implemented in the optimizer, (eg orthant-wise L-BFGS), so we occasionally want to only calculate the log likelihood.
      • setGaussianPriorVariance

        public MaxEntOptimizableByLabelLikelihood setGaussianPriorVariance​(double gaussianPriorVariance)
        Sets a parameter to prevent overtraining. A smaller variance for the prior means that feature weights are expected to hover closer to 0, so extra evidence is required to set a higher weight.
        This trainer