Package org.coordinatekit.crf.mallet.train

Interface TrainingTestSplit

All Known Implementing Classes:
SimpleTrainingTestSplit
public interface TrainingTestSplit
A container for training and test data splits.

This interfaces exposes the results of splitting a dataset into training and test sets. Both InstanceLists share the same Alphabet for features and LabelAlphabet for labels, ensuring consistent encoding across the split.

  • Method Summary

    Modifier and Type
    Method
    Description
    default int
    size()
    Returns the total number of instances across both training and test sets.
    cc.mallet.types.InstanceList
    test()
    Returns the instances to use for model evaluation.
    cc.mallet.types.InstanceList
    training()
    Returns the training instances to use for model training.

  • Method Details

    • test

      cc.mallet.types.InstanceList test()
      Returns the instances to use for model evaluation.
      Returns:
      the test instances

    • training

      cc.mallet.types.InstanceList training()
      Returns the training instances to use for model training.
      Returns:
      the training instances

    • size

      default int size()
      Returns the total number of instances across both training and test sets.
      Returns:
      the combined size of training and test sets