init_vectorstore_table

init_vectorstore_table(
  self,
  table_name: str,
  vector_size: int,
  *,
  schema_name: str

metadata_json_column

'langchain_metadata'

Parameters

Name	Type	Description
`table_name`*	`str`	The database table name.
`vector_size`*	`int`	Vector size for the embedding model to be used.
`schema_name`	`str`	Default:`'public'` The schema name. Default: "public".
`content_column`	`str`	Default:`'content'`
`embedding_column (str)` *	`unknown`
`metadata_columns`	`Optional[list[Union[Column, ColumnDict]]]`	Default:`None`
`metadata_json_column`	`str`	Default:`'langchain_metadata'`
`id_column (Union[str, Column, ColumnDict])` *	`unknown`
`overwrite_existing`	`bool`	Default:`False`
`store_metadata`	`bool`	Default:`True`

Create a table for saving of vectors to be used with PGVectorStore.

Name of the column to store document content. Default: "page_content".

Name of the column to store vector embeddings. Default: "embedding".

A list of Columns to create for custom metadata. Default: None. Optional.

The column to store extra metadata in JSON format. Default: "langchain_metadata". Optional.

Column to store ids. Default: "langchain_id" column name with data type UUID. Optional.

Whether to drop existing table. Default: False.

Whether to store metadata in the table. Default: True.