Qdrant · Schema

RecommendRequest

Recommendation request. Provides positive and negative examples of the vectors, which can be ids of points that are already stored in the collection, raw vectors, or even ids and vectors combined. Service should look for the points which are closer to positive examples and at the same time further to negative examples. The concrete way of how to compare negative and positive distances is up to the `strategy` chosen.

AIArtificial IntelligenceVector Databases

Properties

Name	Type	Description
shard_key	object	Specify in which shards to look for the points, if not specified - look in all shards
positive	array	Look for vectors closest to those
negative	array	Try to avoid vectors like this
strategy	object	How to use positive and negative examples to find the results
filter	object	Look only for points which satisfies this conditions
params	object	Additional search params
limit	integer	Max number of result to return
offset	integer	Offset of the first result to return. May be used to paginate results. Note: large offset values may cause performance issues.
with_payload	object	Select which payload to return with the response. Default is false.
with_vector	object	Options for specifying which vectors to include into response. Default is false.
score_threshold	number	Define a minimal score threshold for the result. If defined, less similar results will not be returned. Score of the returned result might be higher or smaller than the threshold depending on the Dist
using	object	Define which vector to use for recommendation, if not specified - try to use default vector
lookup_from	object	The location used to lookup vectors. If not specified - use current collection. Note: the other collection should have the same vector size as the current collection

View JSON Schema on GitHub

JSON Schema

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "#/components/schemas/RecommendRequest",
  "title": "RecommendRequest",
  "description": "Recommendation request. Provides positive and negative examples of the vectors, which can be ids of points that are already stored in the collection, raw vectors, or even ids and vectors combined.\n\nService should look for the points which are closer to positive examples and at the same time further to negative examples. The concrete way of how to compare negative and positive distances is up to the `strategy` chosen.",
  "type": "object",
  "required": [
    "limit"
  ],
  "properties": {
    "shard_key": {
      "description": "Specify in which shards to look for the points, if not specified - look in all shards",
      "anyOf": [
        {
          "$ref": "#/components/schemas/ShardKeySelector"
        },
        {
          "nullable": true
        }
      ]
    },
    "positive": {
      "description": "Look for vectors closest to those",
      "default": [],
      "type": "array",
      "items": {
        "$ref": "#/components/schemas/RecommendExample"
      }
    },
    "negative": {
      "description": "Try to avoid vectors like this",
      "default": [],
      "type": "array",
      "items": {
        "$ref": "#/components/schemas/RecommendExample"
      }
    },
    "strategy": {
      "description": "How to use positive and negative examples to find the results",
      "anyOf": [
        {
          "$ref": "#/components/schemas/RecommendStrategy"
        },
        {
          "nullable": true
        }
      ]
    },
    "filter": {
      "description": "Look only for points which satisfies this conditions",
      "anyOf": [
        {
          "$ref": "#/components/schemas/Filter"
        },
        {
          "nullable": true
        }
      ]
    },
    "params": {
      "description": "Additional search params",
      "anyOf": [
        {
          "$ref": "#/components/schemas/SearchParams"
        },
        {
          "nullable": true
        }
      ]
    },
    "limit": {
      "description": "Max number of result to return",
      "type": "integer",
      "format": "uint",
      "minimum": 1
    },
    "offset": {
      "description": "Offset of the first result to return. May be used to paginate results. Note: large offset values may cause performance issues.",
      "type": "integer",
      "format": "uint",
      "minimum": 0,
      "nullable": true
    },
    "with_payload": {
      "description": "Select which payload to return with the response. Default is false.",
      "anyOf": [
        {
          "$ref": "#/components/schemas/WithPayloadInterface"
        },
        {
          "nullable": true
        }
      ]
    },
    "with_vector": {
      "description": "Options for specifying which vectors to include into response. Default is false.",
      "default": null,
      "anyOf": [
        {
          "$ref": "#/components/schemas/WithVector"
        },
        {
          "nullable": true
        }
      ]
    },
    "score_threshold": {
      "description": "Define a minimal score threshold for the result. If defined, less similar results will not be returned. Score of the returned result might be higher or smaller than the threshold depending on the Distance function used. E.g. for cosine similarity only higher scores will be returned.",
      "type": "number",
      "format": "float",
      "nullable": true
    },
    "using": {
      "description": "Define which vector to use for recommendation, if not specified - try to use default vector",
      "default": null,
      "anyOf": [
        {
          "$ref": "#/components/schemas/UsingVector"
        },
        {
          "nullable": true
        }
      ]
    },
    "lookup_from": {
      "description": "The location used to lookup vectors. If not specified - use current collection. Note: the other collection should have the same vector size as the current collection",
      "default": null,
      "anyOf": [
        {
          "$ref": "#/components/schemas/LookupLocation"
        },
        {
          "nullable": true
        }
      ]
    }
  }
}