GS_MODEL_WAREHOUSE
GS_MODEL_WAREHOUSE stores AI engine training models, including the models and detailed description of the training process.
| Name | Data Type | Description |
|---|---|---|
| oid | oid | Hidden column |
| modelname | name | Unique constraint |
| modelowner | oid | OID of a model owner |
| createtime | timestamp without time zone | Time when a model is created |
| processedtuples | integer | Number of tuples involved in training |
| discardedtuples | integer | Number of unqualified tuples not involved in training |
| preprocesstime | real | Data preprocessing time |
| exectime | real | Training duration |
| iterations | integer | Iteration round |
| outputtype | oid | OID of the output data type |
| modeltype | text | AI operator type |
| query | text | Query statement executed to create a model |
| modeldata | bytea | Stored binary model information |
| weight | real[] | Currently, this column applies only to GD operator models. |
| hyperparametersnames | text[] | Involved hyperparameter name |
| hyperparametersvalues | text[] | Hyperparameter value |
| hyperparametersoids | oid[] | OID of the data type corresponding to a hyperparameter |
| coefnames | text[] | Model parameter |
| coefvalues | text[] | Value of a model parameter |
| coefoids | oid[] | OID of the data type corresponding to a model parameter |
| trainingscoresname | text[] | Method used to measure model performance |
| trainingscoresvalue | real[] | Value used to measure model performance |
| modeldescribe | text[] | Model description |
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.
