HeatWave ML stores machine learning models in a model catalog in
MySQL. A model catalog is a standard MySQL table named
MODEL_CATALOG. HeatWave ML creates a model
catalog for any user that creates a machine learning model.
MODEL_CATALOG table is created in a
name of the owning user.
When a user creates a model, the
routine creates the model catalog schema and table if they do
inserts the model as a row in the
MODEL_CATALOG table at the end of training.
A model catalog is accessible only to the owning user unless the user grants privileges on the model catalog to another user. This means that HeatWave ML routines can only use models that are accessible to the user running the routines. For information about granting model catalog privileges, see Section 3.7.9, “Sharing Models”.
A database administrator can manage a model catalog table as they would a regular MySQL table.
MODEL_CATALOG table has the following
A unique auto-incrementing numeric identifier for the model.
model_handle. The model handle is generated when the
ML_TRAINroutine is executed on a training dataset. The
schemaName_tableName_userName_No, as in the following example:
The format of the model handle is subject to change.
A string in JSON format containing the serialized HeatWave ML model.
The user who initiated the
ML_TRAINroutine to create the model.
A timestamp indicating when the model was created (in UNIX epoch time). A model is created when the
ML_TRAINroutine finishes executing.
The name of the column in the training table that was specified as the target column.
The name of the input table specified by the
The model object size, in bytes.
The type of model (algorithm) selected by
ML_TRAINto build the model.
The task type specified in the
The feature columns used to train the model.
The model explanation generated during training. See Section 3.7.6, “Model Explanations”. This column was added in MySQL 8.0.29.
The last time the model was accessed. HeatWave ML routines update this value to the current timestamp when accessing the model.