Input Table Schema
| Column Name | Data Type | Description |
|---|---|---|
| id_column | Any | Column that uniquely identifies an observation in test table. |
| input_column | INTEGER, BIGINT, SMALLINT, BYTEINT, FLOAT, DECIMAL, NUMBER | Column used as predictors (features) for model training. |
| partition_by_column | Any | Unique identifier of model for testing points in InputTable. They must be matched with partition_by_column in the model. |
| accumulate_column | Any | Input table columns to copy to the output table. |
Model Schema
See TD_GLM for a model example.
| Column Name | Data Type | Description |
|---|---|---|
| partiton_by_column | Any | Unique identifier of model. |
| attribute | SMALLINT | Numeric index that represents predictor and model metrics wherein model metrics have negative values and predictors take positive values. Intercept is specified using index 0. |
| predictor | VARCHAR | Predictor or model metric name. The maximum length is 32000 characters. |
| estimate | FLOAT | Predictor weights and numeric values of metrics. |
| value | VARCHAR | Values of metric string. The maximum length is 30 characters. |