Fusion 5.12
Product Selector
Compute the distance between a query vector and each candidate document vector. It should be added after the Solr Query stage.
This stage is used in the default query pipeline that enables Smart Answers. See Detailed Pipeline Setup for configuration guidelines.
This stage requires a Fusion AI license.
Computes and emits selected distance between Query-Document dense vectors
Set to true to skip this stage.
Default: false
A unique label for this stage.
<= 255 characters
Define a conditional script that must result in true or false. This can be used to determine if the stage should process or not.
Name of the query context key containing dense vector (as double array)
>= 1 characters
Default: query_vector
Name of the document field containing dense vector (as double array) or compressed vector (as string)
>= 1 characters
Default: compressed_document_vector_s
Should the document field with the dense vector kept and returned in the response
Default: false
One of the following functions: cosine_similarity, cosine_distance, euclidean_distance, dot_product
Default: cosine_similarity
Allowed values: cosine_similaritycosine_distanceeuclidean_distancedot_product
Name of the result document field containing vectors distance. Might be used for further results sorting.
>= 1 characters
Default: vectors_distance
Flag to indicate if this stage should throw an exception if an error occurs.
Default: false