Class QNModel

java.lang.Object
opennlp.tools.ml.model.AbstractModel
opennlp.tools.ml.maxent.quasinewton.QNModel
All Implemented Interfaces:
MaxentModel

public class QNModel extends AbstractModel
A maximum entropy model which has been trained via the L-BFGS algorithm , which belongs to the group of Quasi Newton (QN) algorithms.
See Also:
  • Constructor Details

    • QNModel

      public QNModel(Context[] params, String[] predLabels, String[] outcomeNames)
      Initializes a QNModel with the specified parameters, predicate/feature labels, and outcome names.
      Parameters:
      params - The parameters of the model.
      predLabels - The names of the predicates used in this model.
      outcomeNames - The names of the outcomes this model predicts.
  • Method Details

    • getNumOutcomes

      public int getNumOutcomes()
      Specified by:
      getNumOutcomes in interface MaxentModel
      Overrides:
      getNumOutcomes in class AbstractModel
      Returns:
      Retrieves the number of outcomes for this model.
    • getPredIndex

      private Context getPredIndex(String predicate)
    • eval

      public double[] eval(String[] context)
      Evaluates a context.
      Parameters:
      context - An array of String names of the contextual predicates which are to be evaluated together.
      Returns:
      An array of the probabilities for each of the different outcomes, all of which sum to 1.
    • eval

      public double[] eval(String[] context, double[] probs)
      Evaluates a context.
      Parameters:
      context - An array of String names of the contextual predicates which are to be evaluated together.
      probs - An array which is populated with the probabilities for each of the different outcomes, all of which sum to 1.
      Returns:
      An array of the probabilities for each of the different outcomes, all of which sum to 1.
    • eval

      public double[] eval(String[] context, float[] values)
      Evaluates a context with the specified context values.
      Parameters:
      context - An array of String names of the contextual predicates which are to be evaluated together.
      values - The values associated with each context.
      Returns:
      An array of the probabilities for each of the different outcomes, all of which sum to 1.
    • eval

      private double[] eval(String[] context, float[] values, double[] probs)
      Evaluates which should be used during inference.
      Parameters:
      context - The predicates which have been observed at the present decision point.
      values - The weights of the predicates which have been observed at the present decision point.
      probs - The probability for outcomes.
      Returns:
      Normalized probabilities for the outcomes given the context.
    • eval

      static double[] eval(int[] context, float[] values, double[] probs, int nOutcomes, int nPredLabels, double[] parameters)
      Evaluates which should be used during training to report model accuracy.
      Parameters:
      context - The indices of the predicates which have been observed at the present decision point.
      values - The weights of the predicates which have been observed at the present decision point.
      probs - The probability for outcomes.
      nOutcomes - The number of outcomes.
      nPredLabels - The number of unique predicates.
      parameters - The model parameters.
      Returns:
      Normalized probabilities for the outcomes given the context.