BigQueryVectorSearch(
self,
embedding: Embeddings,
project_id: str,
dataset_name: str| Name | Type | Description |
|---|---|---|
embedding* | Embeddings | Text Embedding model to use. |
project_id* | str | GCP project. |
dataset_name* | str | BigQuery dataset to store documents and embeddings. |
table_name* | str | BigQuery table name. |
location | str | Default: 'US' |
content_field | str | Default: DEFAULT_CONTENT_COLUMN_NAME |
metadata_field | str | Default: DEFAULT_METADATA_COLUMN_NAME |
text_embedding_field | str | Default: DEFAULT_TEXT_EMBEDDING_COLUMN_NAME |
doc_id_field | str | Default: DEFAULT_DOC_ID_COLUMN_NAME |
distance_strategy | DistanceStrategy | Default: DEFAULT_DISTANCE_STRATEGY |
credentials | Credentials | Default: None |
Google Cloud BigQuery vector store.
To use, you need the following packages installed: google-cloud-bigquery
BigQuery region. Defaults to
US(multi-region).
Specifies the column to store the content.
Defaults to content.
Specifies the column to store the metadata.
Defaults to metadata.
Specifies the column to store
the embeddings vector.
Defaults to text_embedding.
Specifies the column to store the document id.
Defaults to doc_id.
Determines the strategy employed for calculating the distance between vectors in the embedding space. Defaults to EUCLIDEAN_DISTANCE. Available options are:
Custom Google Cloud credentials to use. Defaults to None.