Pre-filtering Vector Searches

Capella Operational

how-to

You can specify filters as part of a Vector Search query object to restrict the documents searched in a Search index.

About Pre-filtering

The Search Service supports pre-filtering on Vector Search queries. Pre-filtering allows you to execute vector searches over a subset of the vector index, via the means of a filter request that qualifies the subset.

Prerequisites

You have the Search Service enabled on a node in your operational cluster. For more information about how to change Services on your operational cluster, see Modify a Paid Cluster.
You have a bucket with scopes and collections in your operational cluster. For more information, see Manage Buckets.

You have created a Search Vector Index.

For more information about how to create a Search Vector Index, see Create a Search Vector Index in Quick Mode.

You can import a sample dataset to use with the procedure or examples on this page.

Go to Data Tools Import from your cluster and import the color-vector-sample sample data.

For the best results, consider using the sample Search Vector Index from Create a Search Vector Index with the Capella UI.

Procedure

To add pre-filtering to a Vector Search query:

On the Operational Clusters page, select the operational cluster where you created your Search index.
Go to Data Tools Search.
Next to your Search Vector Index, click Search.
In the Search field, enter a search query that includes a filter object with your knn object.

For more information about the filter object, see filter.
Press Enter or click Search.
(Optional) To view a document and its source collection, click a document name in the search results list.

Example: Pre-Filter A Vector Search Query For The Color "Navy"

For example, the following Vector Search query tries to find matches to a color with an RGB value of [176, 0, 176] with a minimum brightness of 70 and a maximum of 80. A pre-filter on the query will narrow the documents searched inside the Search Vector Index to documents that have a color field value that closely matches navy:

{
  "fields": ["*"],
  "query": {
    "min": 70,
    "max": 80,
    "inclusive_min": false,
    "inclusive_max": true,
    "field": "brightness"
  },
  "knn": [
    {
      "k": 10,
      "field": "colorvect_l2",
      "vector": [ 176, 0, 176 ],
      "filter": {
        "field":  "color",
        "match": "navy"
      }
    }
  ]
}