ALTER MODEL … DROP VERSION

Removes a version from the specified machine learning model.

See also:

ALTER MODEL … ADD VERSION, ALTER MODEL … MODIFY VERSION

Syntax

ALTER MODEL [ IF EXISTS ] <name> DROP VERSION <version_name>
Copy

Parameters

name

Specifies the identifier of the model. If the identifier contains spaces, special characters, or mixed-case characters, the entire identifier must be enclosed in double quotes. Identifiers enclosed in double quotes are also case-sensitive. For information on identifier syntax, see Identifier requirements.

If the identifier contains spaces or special characters, the entire string must be enclosed in double quotes. Identifiers enclosed in double quotes are also case-sensitive.

For more information, see Identifier requirements.

version_name

Specifies the identifier of the version to be removed.

If the identifier contains spaces or special characters, the entire string must be enclosed in double quotes. Identifiers enclosed in double quotes are also case-sensitive.

For more information, see Identifier requirements.

Usage notes

Aliases are alternative names for model versions. In addition to aliases you create, the following three system aliases are available.

  • DEFAULT refers to the default version of the model.

  • FIRST refers to the oldest version of the model by creation time.

  • LAST refers to the newest version of the model by creation time.

When you drop the first or last model version, the corresponding system alias, FIRST or LAST, adjusts to point to the new first or last alias.

You cannot drop the default version of a model. Change the default to a different version, if there is one, using ALTER MODEL … SET DEFAULT VERSION, then drop the unneeded version. If there is no other version to select as the default, because the model has only one version, drop the entire model.