gcloud beta vector-search collections create - create a collection

gcloud beta vector-search collections create(COLLECTION :--location=LOCATION)[--async][--data-schema=DATA_SCHEMA][--description=DESCRIPTION][--display-name=DISPLAY_NAME][--labels=[LABELS,…]][--request-id=REQUEST_ID][--schema=SCHEMA][--vector-schema=[VECTOR_SCHEMA,…]][GCLOUD_WIDE_FLAG …]

gcloudbetavector-searchcollectionscreatemy-collection--location=us-central1--display-name='My Collection'--vector-schema='{"my-embedding-field": {"denseVector": {"dimensions": 100}}}'--project=my-project

To create a collectionmy-collection in projectmy-project and locationus-central1 with data schemaand vector schema, run:

gcloudbetavector-searchcollectionscreatemy-collection--location=us-central1--display-name='My Collection'--data-schema='{"type":"object","properties":{"year":{"type":"number"},"genre":{"type":"string"},"director":{"type":"string"},"title":{"type":"string"}}}'--vector-schema='{"plot_embedding":{"denseVector":{"dimensions":3}},"genre_embedding":{"denseVector":{"dimensions":4}},"sparse_embedding":{"sparseVector":{}}}'--project=my-project

Collection resource - Identifier. name of resource The arguments in this groupcan be used to specify the attributes of this resource. (NOTE) Some attributesare not given arguments in this group but can be set in other ways.

To set theproject attribute:

• provide the argumentcollection on the command line with a fullyspecified name;
• provide the argument--project on the command line;
• set the propertycore/project.

This must be specified.

COLLECTION

ID of the collection or fully qualified identifier for the collection.

To set thecollection attribute:

• provide the argumentcollection on the command line.

This positional argument must be specified if any of the other arguments in thisgroup are specified.

--location=LOCATION

The location id of the collection resource.

To set thelocation attribute:

• provide the argumentcollection on the command line with a fullyspecified name;
• provide the argument--location on the command line.

--async

Return immediately, without waiting for the operation in progress to complete.

--data-schema=DATA_SCHEMA

JSON Schema for data. Field names must contain only alphanumeric characters,underscores, and hyphens.

--description=DESCRIPTION

User-specified description of the collection

--display-name=DISPLAY_NAME

User-specified display name of the collection

--labels=[LABELS,…]

Labels as key value pairs.

KEY

Keys must start with a lowercase character and contain only hyphens(-), underscores (_), lowercase characters, andnumbers.

VALUE

Values must contain only hyphens (-), underscores (_),lowercase characters, and numbers.

Shorthand Example:

--labels=string=string

JSON Example:

--labels='{"string": "string"}'

File Example:

--labels=path_to_file.(yaml|json)

--request-id=REQUEST_ID

An optional request ID to identify requests. Specify a unique request ID so thatif you must retry your request, the server will know to ignore the request if ithas already been completed. The server will guarantee that for at least 60minutes since the first request.

For example, consider a situation where you make an initial request and therequest times out. If you make the request again with the same request ID, theserver can check if original operation with the same request ID was received,and if so, will ignore the second request. This prevents clients fromaccidentally creating duplicate commitments.

The request ID must be a valid UUID with the exception that zero UUID is notsupported (00000000-0000-0000-0000-000000000000).

--schema=SCHEMA

Deprecated: JSON Schema for data. Please use data_schema instead.

--vector-schema=[VECTOR_SCHEMA,…]

Schema for vector fields. Only vector fields in this schema will be searchable.Field names must contain only alphanumeric characters, underscores, and hyphens.

KEY

SetsKEY value.

VALUE

SetsVALUE value.

denseVector

Dense vector field.

dimensions

Dimensionality of the vector field.

vertexEmbeddingConfig

Configuration for generating embeddings for the vector field. If not specified,the embedding field must be populated in the DataObject.

modelId

Required: ID of the embedding model to use. Seehttps://cloud.google.com/vertex-ai/generative-ai/docs/learn/models#embeddings-modelsfor the list of supported models.

taskType

Required: Task type for the embeddings.

textTemplate

Required: Text template for the input to the model. The template must containone or more references to fields in the DataObject, e.g.: "Movie Title: {title}---- Movie Plot: {plot}".

sparseVector

Sparse vector field.

Shorthand Example:

--vector-schema=string={denseVector={dimensions=int,vertexEmbeddingConfig={modelId=string,taskType=string,textTemplate=string}},sparseVector}

JSON Example:

--vector-schema='{"string": {"denseVector": {"dimensions": int, "vertexEmbeddingConfig": {"modelId": "string", "taskType": "string", "textTemplate": "string"}}, "sparseVector": {}}}'

File Example:

--vector-schema=path_to_file.(yaml|json)

Run$gcloud help for details.