Knn query
editKnn query
editFinds the k nearest vectors to a query vector, as measured by a similarity metric. knn query finds nearest vectors through approximate search on indexed dense_vectors. The preferred way to do approximate kNN search is through the top level knn section of a search request. knn query is reserved for expert cases, where there is a need to combine this query with other queries.
Example request
editresp = client.indices.create( index="my-image-index", mappings={ "properties": { "image-vector": { "type": "dense_vector", "dims": 3, "index": True, "similarity": "l2_norm" }, "file-type": { "type": "keyword" }, "title": { "type": "text" } } }, ) print(resp)
response = client.indices.create( index: 'my-image-index', body: { mappings: { properties: { "image-vector": { type: 'dense_vector', dims: 3, index: true, similarity: 'l2_norm' }, "file-type": { type: 'keyword' }, title: { type: 'text' } } } } ) puts response
const response = await client.indices.create({ index: "my-image-index", mappings: { properties: { "image-vector": { type: "dense_vector", dims: 3, index: true, similarity: "l2_norm", }, "file-type": { type: "keyword", }, title: { type: "text", }, }, }, }); console.log(response);
PUT my-image-index { "mappings": { "properties": { "image-vector": { "type": "dense_vector", "dims": 3, "index": true, "similarity": "l2_norm" }, "file-type": { "type": "keyword" }, "title": { "type": "text" } } } }
-
Index your data.
resp = client.bulk( index="my-image-index", refresh=True, operations=[ { "index": { "_id": "1" } }, { "image-vector": [ 1, 5, -20 ], "file-type": "jpg", "title": "mountain lake" }, { "index": { "_id": "2" } }, { "image-vector": [ 42, 8, -15 ], "file-type": "png", "title": "frozen lake" }, { "index": { "_id": "3" } }, { "image-vector": [ 15, 11, 23 ], "file-type": "jpg", "title": "mountain lake lodge" } ], ) print(resp)
response = client.bulk( index: 'my-image-index', refresh: true, body: [ { index: { _id: '1' } }, { "image-vector": [ 1, 5, -20 ], "file-type": 'jpg', title: 'mountain lake' }, { index: { _id: '2' } }, { "image-vector": [ 42, 8, -15 ], "file-type": 'png', title: 'frozen lake' }, { index: { _id: '3' } }, { "image-vector": [ 15, 11, 23 ], "file-type": 'jpg', title: 'mountain lake lodge' } ] ) puts response
const response = await client.bulk({ index: "my-image-index", refresh: "true", operations: [ { index: { _id: "1", }, }, { "image-vector": [1, 5, -20], "file-type": "jpg", title: "mountain lake", }, { index: { _id: "2", }, }, { "image-vector": [42, 8, -15], "file-type": "png", title: "frozen lake", }, { index: { _id: "3", }, }, { "image-vector": [15, 11, 23], "file-type": "jpg", title: "mountain lake lodge", }, ], }); console.log(response);
POST my-image-index/_bulk?refresh=true { "index": { "_id": "1" } } { "image-vector": [1, 5, -20], "file-type": "jpg", "title": "mountain lake" } { "index": { "_id": "2" } } { "image-vector": [42, 8, -15], "file-type": "png", "title": "frozen lake"} { "index": { "_id": "3" } } { "image-vector": [15, 11, 23], "file-type": "jpg", "title": "mountain lake lodge" }
-
Run the search using the
knn
query, asking for the top 10 nearest vectors from each shard, and then combine shard results to get the top 3 global results.resp = client.search( index="my-image-index", size=3, query={ "knn": { "field": "image-vector", "query_vector": [ -5, 9, -12 ], "k": 10 } }, ) print(resp)
const response = await client.search({ index: "my-image-index", size: 3, query: { knn: { field: "image-vector", query_vector: [-5, 9, -12], k: 10, }, }, }); console.log(response);
POST my-image-index/_search { "size" : 3, "query" : { "knn": { "field": "image-vector", "query_vector": [-5, 9, -12], "k": 10 } } }
Top-level parameters for knn
edit-
field
-
(Required, string) The name of the vector field to search against. Must be a
dense_vector
field with indexing enabled. -
query_vector
-
(Optional, array of floats or string) Query vector. Must have the same number of dimensions as the vector field you are searching against. Must be either an array of floats or a hex-encoded byte vector. Either this or
query_vector_builder
must be provided. -
query_vector_builder
-
(Optional, object) Query vector builder. A configuration object indicating how to build a query_vector before executing the request. You must provide either a
query_vector_builder
orquery_vector
, but not both. Refer to Perform semantic search to learn more. -
k
-
(Optional, integer) The number of nearest neighbors to return from each shard. Elasticsearch collects
k
results from each shard, then merges them to find the global top results. This value must be less than or equal tonum_candidates
. Defaults tonum_candidates
. -
num_candidates
-
(Optional, integer) The number of nearest neighbor candidates to consider per shard while doing knn search. Cannot exceed 10,000. Increasing
num_candidates
tends to improve the accuracy of the final results. Defaults to1.5 * k
ifk
is set, or1.5 * size
ifk
is not set. -
filter
-
(Optional, query object) Query to filter the documents that can match. The kNN search will return the top documents that also match this filter. The value can be a single query or a list of queries. If
filter
is not provided, all documents are allowed to match.The filter is a pre-filter, meaning that it is applied during the approximate kNN search to ensure that
num_candidates
matching documents are returned. -
similarity
-
(Optional, float) The minimum similarity required for a document to be considered a match. The similarity value calculated relates to the raw
similarity
used. Not the document score. The matched documents are then scored according tosimilarity
and the providedboost
is applied. -
boost
-
(Optional, float) Floating point number used to multiply the scores of matched documents. This value cannot be negative. Defaults to
1.0
. -
_name
-
(Optional, string) Name field to identify the query
Pre-filters and post-filters in knn query
editThere are two ways to filter documents that match a kNN query:
-
pre-filtering – filter is applied during the approximate kNN search
to ensure that
k
matching documents are returned. - post-filtering – filter is applied after the approximate kNN search completes, which results in fewer than k results, even when there are enough matching documents.
Pre-filtering is supported through the filter
parameter of the knn
query.
Also filters from aliases are applied as pre-filters.
All other filters found in the Query DSL tree are applied as post-filters.
For example, knn
query finds the top 3 documents with the nearest vectors
(k=3), which are combined with term
filter, that is
post-filtered. The final set of documents will contain only a single document
that passes the post-filter.
resp = client.search( index="my-image-index", size=10, query={ "bool": { "must": { "knn": { "field": "image-vector", "query_vector": [ -5, 9, -12 ], "k": 3 } }, "filter": { "term": { "file-type": "png" } } } }, ) print(resp)
const response = await client.search({ index: "my-image-index", size: 10, query: { bool: { must: { knn: { field: "image-vector", query_vector: [-5, 9, -12], k: 3, }, }, filter: { term: { "file-type": "png", }, }, }, }, }); console.log(response);
POST my-image-index/_search { "size" : 10, "query" : { "bool" : { "must" : { "knn": { "field": "image-vector", "query_vector": [-5, 9, -12], "k": 3 } }, "filter" : { "term" : { "file-type" : "png" } } } } }
Hybrid search with knn query
editKnn query can be used as a part of hybrid search, where knn query is combined
with other lexical queries. For example, the query below finds documents with
title
matching mountain lake
, and combines them with the top 10 documents
that have the closest image vectors to the query_vector
. The combined documents
are then scored and the top 3 top scored documents are returned.
+
resp = client.search( index="my-image-index", size=3, query={ "bool": { "should": [ { "match": { "title": { "query": "mountain lake", "boost": 1 } } }, { "knn": { "field": "image-vector", "query_vector": [ -5, 9, -12 ], "k": 10, "boost": 2 } } ] } }, ) print(resp)
const response = await client.search({ index: "my-image-index", size: 3, query: { bool: { should: [ { match: { title: { query: "mountain lake", boost: 1, }, }, }, { knn: { field: "image-vector", query_vector: [-5, 9, -12], k: 10, boost: 2, }, }, ], }, }, }); console.log(response);
POST my-image-index/_search { "size" : 3, "query": { "bool": { "should": [ { "match": { "title": { "query": "mountain lake", "boost": 1 } } }, { "knn": { "field": "image-vector", "query_vector": [-5, 9, -12], "k": 10, "boost": 2 } } ] } } }
Knn query inside a nested query
editknn
query can be used inside a nested query. The behaviour here is similar
to top level nested kNN search:
- kNN search over nested dense_vectors diversifies the top results over the top-level document
-
filter
over the top-level document metadata is supported and acts as a pre-filter -
filter
overnested
field metadata is not supported
A sample query can look like below:
{ "query" : { "nested" : { "path" : "paragraph", "query" : { "knn": { "query_vector": [ 0.45, 45 ], "field": "paragraph.vector", "num_candidates": 2 } } } } }
Knn query with aggregations
editknn
query calculates aggregations on top k
documents from each shard.
Thus, the final results from aggregations contain
k * number_of_shards
documents. This is different from
the top level knn section where aggregations are
calculated on the global top k
nearest documents.