Package org.tribuo.regression.impl

Class SkeletalIndependentRegressionSparseTrainer<T>

java.lang.Object
org.tribuo.regression.impl.SkeletalIndependentRegressionSparseTrainer<T>
All Implemented Interfaces:
com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<TrainerProvenance>, SparseTrainer<Regressor>, Trainer<Regressor>
public abstract class SkeletalIndependentRegressionSparseTrainer<T> extends Object implements SparseTrainer<Regressor>
Base class for training n independent sparse models, one per dimension. Generates the SparseVectors once to reduce allocation.

Then wraps them in an SkeletalIndependentRegressionSparseModel to provide a Regressor prediction.

It trains each model sequentially, and could be optimised to train in parallel.

  • Constructor Details

    • SkeletalIndependentRegressionSparseTrainer

      protected SkeletalIndependentRegressionSparseTrainer()
      for olcut.

  • 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

    • train

      public SkeletalIndependentRegressionSparseModel train(Dataset<Regressor> examples)
      Description copied from interface: SparseTrainer
      Trains a sparse predictive model using the examples in the given data set.
      Specified by:
      train in interface SparseTrainer<T>
      Specified by:
      train in interface Trainer<T>
      Parameters:
      examples - The data set containing the examples.
      Returns:
      A sparse predictive model that can be used to generate predictions for new examples.

    • train

      public SkeletalIndependentRegressionSparseModel train(Dataset<Regressor> examples, Map<String,com.oracle.labs.mlrg.olcut.provenance.Provenance> runProvenance)
      Description copied from interface: SparseTrainer
      Trains a sparse predictive model using the examples in the given data set.
      Specified by:
      train in interface SparseTrainer<T>
      Specified by:
      train in interface Trainer<T>
      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.

    • train

      public SkeletalIndependentRegressionSparseModel train(Dataset<Regressor> examples, Map<String,com.oracle.labs.mlrg.olcut.provenance.Provenance> runProvenance, int invocationCount)
      Description copied from interface: SparseTrainer
      Trains a predictive model using the examples in the given data set.
      Specified by:
      train in interface SparseTrainer<T>
      Specified by:
      train in interface Trainer<T>
      Parameters:
      examples - the data set containing the examples.
      runProvenance - Training run specific provenance (e.g., fold number).
      invocationCount - The state of the RNG the trainer should be set to before training
      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<T>
      Returns:
      The number of train invocations.

    • setInvocationCount

      public void setInvocationCount(int invocationCount)
      Description copied from interface: Trainer
      Set the internal state of the trainer to the provided number of invocations of the train method.

      This is used when reproducing a Tribuo-trained model by setting the state of the RNG to what it was at when Tribuo trained the original model by simulating invocations of the train method. This method should ALWAYS be overridden, and the default method is purely for compatibility.

      In a future major release this default implementation will be removed.

      Specified by:
      setInvocationCount in interface Trainer<T>
      Parameters:
      invocationCount - the number of invocations of the train method to simulate

    • createModel

      protected abstract SkeletalIndependentRegressionSparseModel createModel(Map<String,T> models, ModelProvenance provenance, ImmutableFeatureMap featureMap, ImmutableOutputInfo<Regressor> outputInfo)
      Constructs the appropriate subclass of SkeletalIndependentRegressionModel for this trainer.
      Parameters:
      models - The models to use.
      provenance - The model provenance
      featureMap - The feature map.
      outputInfo - The regression info.
      Returns:
      A subclass of IndependentRegressionModel.

    • trainDimension

      protected abstract T trainDimension(double[] outputs, SparseVector[] features, float[] weights, SplittableRandom rng)
      Trains a single dimension of the possibly multiple dimensions.
      Parameters:
      outputs - The regression targets for this dimension.
      features - The features.
      weights - The example weights.
      rng - The RNG to use.
      Returns:
      An object representing the model.

    • useBias

      protected abstract boolean useBias()
      Returns true if the SparseVector should be constructed with a bias feature.
      Returns:
      True if the trainer needs a bias.

    • getModelClassName

      protected abstract String getModelClassName()
      Returns the class name of the model that this class produces.
      Returns:
      The class name of the model.