Interface XGBoostOutputConverter<T extends Output<T>>

All Superinterfaces:
Serializable
All Known Implementing Classes:
XGBoostClassificationConverter, XGBoostRegressionConverter

public interface XGBoostOutputConverter<T extends Output<T>> extends Serializable
Converts the output of XGBoost into the appropriate prediction type.
  • Method Details

    • generatesProbabilities

      boolean generatesProbabilities()
      Does this converter produce probabilities?
      Returns:
      True if it produces probabilities.
    • convertOutput

      Prediction<T> convertOutput(ImmutableOutputInfo<T> info, List<float[]> probabilities, int numValidFeatures, Example<T> example)
      Converts a list of float arrays from XGBoost Boosters into a Tribuo Prediction.
      Parameters:
      info - The output info.
      probabilities - The XGBoost output.
      numValidFeatures - The number of valid features used in the prediction.
      example - The example this prediction was generated from.
      Returns:
      The prediction object.
    • convertBatchOutput

      List<Prediction<T>> convertBatchOutput(ImmutableOutputInfo<T> info, List<float[][]> probabilities, int[] numValidFeatures, Example<T>[] examples)
      Converts a list of float arrays from XGBoost Boosters into a Tribuo Prediction.
      Parameters:
      info - The output info.
      probabilities - The XGBoost output, list dimension is across models, first array dimension is across examples, second array dimension is across outputs.
      numValidFeatures - The number of valid features used in each prediction.
      examples - The examples these predictions were generated from.
      Returns:
      The prediction object.