Class KernelSVMTrainer

java.lang.Object
org.tribuo.classification.sgd.kernel.KernelSVMTrainer
All Implemented Interfaces:
com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<TrainerProvenance>, Trainer<Label>, WeightedExamples
public class KernelSVMTrainer extends Object implements Trainer<Label>, WeightedExamples
A trainer for a kernelised model using the Pegasos optimiser.

The Pegasos optimiser is extremely sensitive to the lambda parameter, and this value must be tuned to get good performance.

See: 

 Shalev-Shwartz S, Singer Y, Srebro N, Cotter A
 "Pegasos: Primal Estimated Sub-Gradient Solver for SVM"
 Mathematical Programming, 2011.

  • Constructor Details

    • KernelSVMTrainer

      public KernelSVMTrainer(Kernel kernel, double lambda, int epochs, int loggingInterval, long seed)
      Constructs a trainer for a kernel SVM model.
      Parameters:
      kernel - The kernel function to use as a similarity measure.
      lambda - l2 regulariser on the support vectors.
      epochs - The number of epochs (complete passes through the training data).
      loggingInterval - Log the loss after this many iterations. If -1 don't log anything.
      seed - A seed for the random number generator, used to shuffle the examples before each epoch.

    • KernelSVMTrainer

      public KernelSVMTrainer(Kernel kernel, double lambda, int epochs, long seed)
      Constructs a trainer for a kernel SVM model. Sets the logging interval to 1000.
      Parameters:
      kernel - The kernel function to use as a similarity measure.
      lambda - l2 regulariser on the support vectors.
      epochs - The number of epochs (complete passes through the training data).
      seed - A seed for the random number generator, used to shuffle the examples before each epoch.

  • Method Details

    • postConfig

      public void postConfig()
      Used by the OLCUT configuration system, and should not be called by external code.
      Specified by:
      postConfig in interface com.oracle.labs.mlrg.olcut.config.Configurable

    • setShuffle

      public void setShuffle(boolean shuffle)
      Turn on or off shuffling of examples.

      This isn't exposed in the constructor as it defaults to on. This method should only be used for debugging.

      Parameters:
      shuffle - If true shuffle the examples, if false leave them in their current order.

    • train

      public KernelSVMModel train(Dataset<Label> examples, Map<String, com.oracle.labs.mlrg.olcut.provenance.Provenance> runProvenance)
      Description copied from interface: Trainer
      Trains a predictive model using the examples in the given data set.
      Specified by:
      train in interface Trainer<Label>
      Parameters:
      examples - the data set containing the examples.
      runProvenance - Training run specific provenance (e.g., fold number).
      Returns:
      a predictive model that can be used to generate predictions for new examples.

    • getInvocationCount

      public int getInvocationCount()
      Description copied from interface: Trainer
      The number of times this trainer instance has had it's train method invoked.

      This is used to determine how many times the trainer's RNG has been accessed to ensure replicability in the random number stream.

      Specified by:
      getInvocationCount in interface Trainer<Label>
      Returns:
      The number of train invocations.

    • toString

      public String toString()
      Overrides:
      toString in class Object

    • getProvenance

      public TrainerProvenance getProvenance()
      Specified by:
      getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<TrainerProvenance>