MODEL_VERSIONS view¶

This Information Schema view displays a row for each machine learning model version defined in the specified (or current) database.

Columns¶

Column Name	Data Type	Description
DATABASE_NAME	TEXT	Database to which the model version belongs.
SCHEMA_NAME	TEXT	Schema to which the model version belongs.
MODEL_NAME	TEXT	Model to which the model version belongs.
MODEL_VERSION_NAME	TEXT	Name of the model version.
VERSION_ALIASES	ARRAY	List of aliases of the model version.
COMMENT	TEXT	Comment for the model version.
OWNER	TEXT	Name of the role that owns the model version.
CREATED	TIMESTAMP_LTZ	Date and time when model version was created.
LAST_ALTERED	TIMESTAMP_LTZ	Date and time when the model version was last updated.
FUNCTIONS	TEXT	Functions in the model version.
MODEL_TYPE	TEXT	Type of the model to which the model version belongs.
PYTHON_VERSION	TEXT	Version of Python required by the model version.
LANGUAGE	TEXT	Language in which the model version is implemented.
DEPENDENCIES	TEXT	Dependencies of the model version.
METADATA	TEXT	Metadata of the model version.
USERDATA	TEXT	User data of the model version.

The view only displays objects for which the current role for the session has been granted access privileges.
The view does not include model versions that have been dropped.

Retrieve the names of all model versions, functions, and the model they belong to, in the mydatabase database:

SELECT model_version_name, model_name, functions, schema_name, database_name
    FROM mydatabase.INFORMATION_SCHEMA.MODEL_VERSIONS;

Copy