java.lang.Object
es.uam.eps.ir.relison.grid.links.recommendation.algorithms.supervised.WekaMLGridSearch<U>
Type Parameters:
U - type of the users.
All Implemented Interfaces:
AlgorithmGridSearch<U>

public class WekaMLGridSearch<U>
extends java.lang.Object
implements AlgorithmGridSearch<U>
Grid search generator for LambdaMART algorithm. Differently from other algorithms, for the LambdaMART algorithm, we consider the grid to be a list of configurations.
See Also:
MachineLearningWekaRecommender
  • Field Summary

    Fields 
    Modifier and Type Field Description
    private static java.lang.String CLASSIFIER
    Identifier for the route of the jforests configuration file.
    private static java.lang.String ITER
    The number of iterations of the random forest algorithm.
    private static java.lang.String TEST
    Identifier for the route containing the test instances.
    private static java.lang.String TRAIN
    Identifier for the route containing the training instances.
    private org.ranksys.formats.parsing.Parser<U> uParser
    Parser for reading users.
  • Constructor Summary

    Constructors 
    Constructor Description
    WekaMLGridSearch​(org.ranksys.formats.parsing.Parser<U> uParser)
    Constructor.
  • Method Summary

    Modifier and Type Method Description
    java.util.Map<java.lang.String,​RecommendationAlgorithmFunction<U>> grid​(Grid grid)
    Obtains the different recommendation algorithms to execute in a grid.
    java.util.Map<java.lang.String,​java.util.function.Supplier<es.uam.eps.ir.ranksys.rec.Recommender<U,​U>>> grid​(Grid grid, FastGraph<U> graph, es.uam.eps.ir.ranksys.fast.preference.FastPreferenceData<U,​U> prefData)
    Obtains the different recommendation algorithms to execute in a grid.
    private java.util.List<org.jooq.lambda.tuple.Tuple2<java.lang.String,​java.util.function.Supplier<weka.classifiers.Classifier>>> selectClassifiers​(java.util.Map<java.lang.String,​Grid> classifiersGrid)  

    Methods inherited from class java.lang.Object

    clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
  • Field Details

    • TRAIN

      private static final java.lang.String TRAIN
      Identifier for the route containing the training instances.
      See Also:
      Constant Field Values
    • TEST

      private static final java.lang.String TEST
      Identifier for the route containing the test instances.
      See Also:
      Constant Field Values
    • CLASSIFIER

      private static final java.lang.String CLASSIFIER
      Identifier for the route of the jforests configuration file.
      See Also:
      Constant Field Values
    • ITER

      private static final java.lang.String ITER
      The number of iterations of the random forest algorithm.
      See Also:
      Constant Field Values
    • uParser

      private final org.ranksys.formats.parsing.Parser<U> uParser
      Parser for reading users.
  • Constructor Details

    • WekaMLGridSearch

      public WekaMLGridSearch​(org.ranksys.formats.parsing.Parser<U> uParser)
      Constructor.
      Parameters:
      uParser - a parser for reading users.
  • Method Details

    • grid

      public java.util.Map<java.lang.String,​RecommendationAlgorithmFunction<U>> grid​(Grid grid)
      Description copied from interface: AlgorithmGridSearch
      Obtains the different recommendation algorithms to execute in a grid.
      Specified by:
      grid in interface AlgorithmGridSearch<U>
      Parameters:
      grid - The grid for the algorithm.
      Returns:
      a map containing the different recommendations.
    • selectClassifiers

      private java.util.List<org.jooq.lambda.tuple.Tuple2<java.lang.String,​java.util.function.Supplier<weka.classifiers.Classifier>>> selectClassifiers​(java.util.Map<java.lang.String,​Grid> classifiersGrid)
    • grid

      public java.util.Map<java.lang.String,​java.util.function.Supplier<es.uam.eps.ir.ranksys.rec.Recommender<U,​U>>> grid​(Grid grid, FastGraph<U> graph, es.uam.eps.ir.ranksys.fast.preference.FastPreferenceData<U,​U> prefData)
      Description copied from interface: AlgorithmGridSearch
      Obtains the different recommendation algorithms to execute in a grid.
      Specified by:
      grid in interface AlgorithmGridSearch<U>
      Parameters:
      grid - The grid for the algorithm.
      graph - The training graph.
      prefData - The preference training data.
      Returns:
      a map containing the different recommendations.