This preview documentation is deprecated.

Discovery for Media has been renamed to media recommendations. It is now part of Vertex AI Search, and you can manage it using the Search and Conversation console. We recommend that you use the Search and Conversation console. After March 31, 2024, the Discovery Engine console will no longer be available.

For information about what's changing, see Switch from Discovery for Media to media recommendations. For more information about recommendations in Vertex AI Search, see About recommendations.

EmbeddingSpec

JSON representation
EmbeddingVector
- JSON representation

The specification that uses customized query embedding vector to do semantic document retrieval.

JSON representation
{ "embeddingVectors": [ { object (`EmbeddingVector`) } ] }

Fields

Fields
`embeddingVectors[]`	`object (EmbeddingVector)` The embedding vector used for retrieval. Limit to 1.

embeddingVectors[]

object (EmbeddingVector)

The embedding vector used for retrieval. Limit to 1.

EmbeddingVector

Embedding vector.

JSON representation
{ "fieldPath": string, "vector": [ number ] }

Fields

Fields
`fieldPath`	`string` Embedding field path in schema.
`vector[]`	`number` Query embedding vector.

fieldPath

string

Embedding field path in schema.

vector[]

number

Query embedding vector.