Dense vector field type

edit

The dense_vector field type stores dense vectors of numeric values. Dense vector fields are primarily used for k-nearest neighbor (kNN) search.

The dense_vector type does not support aggregations or sorting.

You add a dense_vector field as an array of numeric values based on element_type with float by default:

response = client.indices.create(
  index: 'my-index',
  body: {
    mappings: {
      properties: {
        my_vector: {
          type: 'dense_vector',
          dims: 3
        },
        my_text: {
          type: 'keyword'
        }
      }
    }
  }
)
puts response

response = client.index(
  index: 'my-index',
  id: 1,
  body: {
    my_text: 'text1',
    my_vector: [
      0.5,
      10,
      6
    ]
  }
)
puts response

response = client.index(
  index: 'my-index',
  id: 2,
  body: {
    my_text: 'text2',
    my_vector: [
      -0.5,
      10,
      10
    ]
  }
)
puts response
PUT my-index
{
  "mappings": {
    "properties": {
      "my_vector": {
        "type": "dense_vector",
        "dims": 3
      },
      "my_text" : {
        "type" : "keyword"
      }
    }
  }
}

PUT my-index/_doc/1
{
  "my_text" : "text1",
  "my_vector" : [0.5, 10, 6]
}

PUT my-index/_doc/2
{
  "my_text" : "text2",
  "my_vector" : [-0.5, 10, 10]
}

Unlike most other data types, dense vectors are always single-valued. It is not possible to store multiple values in one dense_vector field.

Index vectors for kNN search

edit

A k-nearest neighbor (kNN) search finds the k nearest vectors to a query vector, as measured by a similarity metric.

Dense vector fields can be used to rank documents in script_score queries. This lets you perform a brute-force kNN search by scanning all documents and ranking them by similarity.

In many cases, a brute-force kNN search is not efficient enough. For this reason, the dense_vector type supports indexing vectors into a specialized data structure to support fast kNN retrieval through the knn option in the search API

Indexing vectors for approximate kNN search is an expensive process. It can take substantial time to ingest documents that contain vector fields with index enabled. See k-nearest neighbor (kNN) search to learn more about the memory requirements.

You can enable indexing by setting the index parameter:

response = client.indices.create(
  index: 'my-index-2',
  body: {
    mappings: {
      properties: {
        my_vector: {
          type: 'dense_vector',
          dims: 3,
          index: true,
          similarity: 'dot_product'
        }
      }
    }
  }
)
puts response
PUT my-index-2
{
  "mappings": {
    "properties": {
      "my_vector": {
        "type": "dense_vector",
        "dims": 3,
        "index": true,
        "similarity": "dot_product" 
      }
    }
  }
}

When index is enabled, you must define the vector similarity to use in kNN search

Elasticsearch uses the HNSW algorithm to support efficient kNN search. Like most kNN algorithms, HNSW is an approximate method that sacrifices result accuracy for improved speed.

Dense vector fields cannot be indexed if they are within nested mappings.

Parameters for dense vector fields

edit

The following mapping parameters are accepted:

element_type
(Optional, string) The data type used to encode vectors. The supported data types are float (default) and byte. float indexes a 4-byte floating-point value per dimension. byte indexes a 1-byte integer value per dimension. Using byte can result in a substantially smaller index size with the trade off of lower precision. Vectors using byte require dimensions with integer values between -128 to 127, inclusive for both indexing and searching.
dims

(Required, integer) Number of vector dimensions. Can’t exceed 1024 for indexed vectors ("index": true), or 2048 for non-indexed vectors.

[preview] This functionality is in technical preview and may be changed or removed in a future release. Elastic will work to fix any issues, but features in technical preview are not subject to the support SLA of official GA features. The number of dimensions for indexed vectors can be extended up to 2048.

index
(Optional, Boolean) If true, you can search this field using the kNN search API. Defaults to false.
similarity

(Required*, string) The vector similarity metric to use in kNN search. Documents are ranked by their vector field’s similarity to the query vector. The _score of each document will be derived from the similarity, in a way that ensures scores are positive and that a larger score corresponds to a higher ranking.

* If index is true, this parameter is required.

Valid values for similarity
l2_norm
Computes similarity based on the L2 distance (also known as Euclidean distance) between the vectors. The document _score is computed as 1 / (1 + l2_norm(query, vector)^2).
dot_product

Computes the dot product of two vectors. This option provides an optimized way to perform cosine similarity. The constraints and computed score are defined by element_type.

When element_type is float, all vectors must be unit length, including both document and query vectors. The document _score is computed as (1 + dot_product(query, vector)) / 2.

When element_type is byte, all vectors must have the same length including both document and query vectors or results will be inaccurate. The document _score is computed as 0.5 + (dot_product(query, vector) / (32768 * dims)) where dims is the number of dimensions per vector.

cosine
Computes the cosine similarity. Note that the most efficient way to perform cosine similarity is to normalize all vectors to unit length, and instead use dot_product. You should only use cosine if you need to preserve the original vectors and cannot normalize them in advance. The document _score is computed as (1 + cosine(query, vector)) / 2. The cosine similarity does not allow vectors with zero magnitude, since cosine is not defined in this case.

Although they are conceptually related, the similarity parameter is different from text field similarity and accepts a distinct set of options.

index_options

(Optional, object) An optional section that configures the kNN indexing algorithm. The HNSW algorithm has two internal parameters that influence how the data structure is built. These can be adjusted to improve the accuracy of results, at the expense of slower indexing speed. When index_options is provided, all of its properties must be defined.

Properties of index_options
type
(Required, string) The type of kNN algorithm to use. Currently only hnsw is supported.
m
(Required, integer) The number of neighbors each node will be connected to in the HNSW graph. Defaults to 16.
ef_construction
(Required, integer) The number of candidates to track while assembling the list of nearest neighbors for each new node. Defaults to 100.

Synthetic _source

edit

Synthetic _source is Generally Available only for TSDB indices (indices that have index.mode set to time_series). For other indices synthetic _source is in technical preview. Features in technical preview may be changed or removed in a future release. Elastic will apply best effort to fix any issues, but features in technical preview are not subject to the support SLA of official GA features.

dense_vector fields support synthetic _source .