hover.rankExpression.distance(dimension,name).md Maven / Gradle / Ivy

Go to download
Used with the
[nearestNeighbor](https://docs.vespa.ai/en/reference/query-language-reference.html#nearestneighbor)
query operator.
A number which is close to 0 when a point vector in the document is close
to a matching point vector in the query.
The document vectors and the query vector must be the same tensor
type, with one indexed dimension of size N, representing a point in an N-dimensional space.



dimension: Specifies the dimension of name.
This must be either the string `field` or the string `label`.


When using `field`, the name given must be a field with a tensor attribute of appropriate type.
Often used when the document type has only one vector field,
see [example](https://docs.vespa.ai/en/nearest-neighbor-search.html#minimal-example).


When using `label`, queries are assumed to contain a
[nearestNeighbor](https://docs.vespa.ai/en/reference/query-language-reference.html#nearestneighbor) query item
with a [label](https://docs.vespa.ai/en/reference/query-language-reference.html#label) that matches the given name.
This is useful when having multiple vector fields, where `distance()` then maps
to the nearestNeighbor operator with the field configured.
[Example](https://docs.vespa.ai/en/nearest-neighbor-search-guide.html#using-label).


name: The value of the field name or label.



The output value depends on the
[distance metric](https://docs.vespa.ai/en/reference/schema-reference.html#distance-metric) used.
The default is the Euclidean distance between the "n"-dimensional
query point "d" and the point "d" in the document tensor field:
$$ distance = \sqrt{\sum_{i=1}^n (q_i - d_i)^2} $$


When the tensor field stores multiple vectors per document, the minimum distance
between the vectors of a document and the query vector is used in the calculation above.