Class RescoreTopNQuery

java.lang.Object
org.apache.lucene.search.Query
org.apache.lucene.search.RescoreTopNQuery

public class RescoreTopNQuery extends Query
A Query that re-scores another Query with a DoubleValuesSource function and cut-off the results at top N. Unlike Rescorer which does rescoring at post-collection phase, this Query does the rescoring at rewrite() phase. The reason it operates in rewrite phase is to be compatible with KNN vector query, where the results are collected upfront, but it can work with any type of Query. Unlike or FunctionScoreQuery, this Query will work even with the no-scoring ScoreMode.
WARNING: This API is experimental and might change in incompatible ways in the next release.
  • Constructor Details

    • RescoreTopNQuery

      public RescoreTopNQuery(Query query, DoubleValuesSource valuesSource, int n)
      Execute the inner Query, re-score using a customizable DoubleValueSource and trim down the result to k
      Parameters:
      query - the query to execute as initial phase
      valuesSource - the double value source to re-score
      n - the number of documents to find
      Throws:
      IllegalArgumentException - if n is less than 1
  • Method Details

    • rewrite

      public Query rewrite(IndexSearcher indexSearcher) throws IOException
      Description copied from class: Query
      Expert: called to re-write queries into primitive queries. For example, a PrefixQuery will be rewritten into a BooleanQuery that consists of TermQuerys.

      Callers are expected to call rewrite multiple times if necessary, until the rewritten query is the same as the original query.

      The rewrite process may be able to make use of IndexSearcher's executor and be executed in parallel if the executor is provided.

      Overrides:
      rewrite in class Query
      Throws:
      IOException
      See Also:
    • hashCode

      public int hashCode()
      Description copied from class: Query
      Override and implement query hash code properly in a subclass. This is required so that QueryCache works properly.
      Specified by:
      hashCode in class Query
      See Also:
    • equals

      public boolean equals(Object o)
      Description copied from class: Query
      Override and implement query instance equivalence properly in a subclass. This is required so that QueryCache works properly.

      Typically a query will be equal to another only if it's an instance of the same class and its document-filtering properties are identical to those of the other instance. Utility methods are provided for certain repetitive code.

      Specified by:
      equals in class Query
      See Also:
    • visit

      public void visit(QueryVisitor visitor)
      Description copied from class: Query
      Recurse through the query tree, visiting any child queries.
      Specified by:
      visit in class Query
      Parameters:
      visitor - a QueryVisitor to be called by each query in the tree
    • toString

      public String toString(String field)
      Description copied from class: Query
      Prints a query to a string, with field assumed to be the default field and omitted.
      Specified by:
      toString in class Query
    • createFullPrecisionRescorerQuery

      public static Query createFullPrecisionRescorerQuery(Query in, float[] targetVector, String field, int n)
      Utility method to create a new RescoreTopNQuery which uses full-precision vectors for rescoring.
      Parameters:
      in - the inner Query to rescore
      targetVector - the target vector to compute score
      field - the vector field to compute score
      n - the number of results to keep
      Returns:
      the RescoreTopNQuery
    • createLateInteractionQuery

      public static Query createLateInteractionQuery(Query in, int n, String fieldName, float[][] queryVector, VectorSimilarityFunction vectorSimilarityFunction)
      Creates a RescoreTopNQuery that computes top N results using multi-vector similarity comparisons against a late interaction field.

      Note: This query computes late interaction field similarity for the entire match-set of wrapped query, and returns a new query with only top-N hits in the match-set. This is typically useful in combining a query's results with other queries for hybrid search. To simply rerank the top N hits without scoring entire match-set, see LateInteractionRescorer.

      Parameters:
      in - the inner Query to rescore
      n - number of results to keep
      fieldName - the LateInteractionField for recomputing top N hits
      queryVector - query multi-vector to use for similarity comparisons
      vectorSimilarityFunction - function to use for vector similarity comparisons.