Class FMMultiLabelOptions

All Implemented Interfaces:

public class FMMultiLabelOptions extends Object implements
CLI options for training a linear classifier.
  • Field Details

    • sgoOptions

      public GradientOptimiserOptions sgoOptions
      The gradient descent optimiser options.
    • fmEpochs

      @Option(longName="fm-epochs", usage="Number of SGD epochs.") public int fmEpochs
      Number of SGD epochs.
    • fmObjective

      @Option(longName="fm-objective", usage="Loss function.") public FMMultiLabelOptions.LossEnum fmObjective
      Loss function.
    • fmLoggingInterval

      @Option(longName="fm-logging-interval", usage="Log the objective after <int> examples.") public int fmLoggingInterval
      Log the objective after n examples.
    • fmMinibatchSize

      @Option(longName="fm-minibatch-size", usage="Minibatch size.") public int fmMinibatchSize
      Minibatch size.
    • fmFactorSize

      @Option(longName="fm-factor-size", usage="Factor size.") public int fmFactorSize
      Factor size.
    • fmVariance

      @Option(longName="fm-variance", usage="Variance of the initialization gaussian.") public double fmVariance
      Variance of the initialization gaussian.
  • Constructor Details

    • FMMultiLabelOptions

      public FMMultiLabelOptions()
  • Method Details

    • getLoss

      public MultiLabelObjective getLoss()
      Returns the loss function specified in the arguments.
      The loss function.
    • getTrainer

      public FMMultiLabelTrainer getTrainer()
      Gets the trainer as configured by these options.
      The trainer.