The output table has a set of predictions for each test point.
| Column | Data Type | Description |
|---|---|---|
| id | INTEGER | Unique id of each input record. |
| accumulate_columns | Any | Column copied from InputTable. |
| input_columns | DOUBLE PRECISION | Calculated shap value determined by model |
| label | INTEGER | The label specified for the class (for classification models only) corresponding to Responses argument. |
| tree_num | VARCHAR | Concatenated task Index and tree num from model (for tree-based models only when Detailed('true') is specified). |
| iter_num | INTEGER | The iter num of a tree (for TD_XGBoost model only when Detailed('true') is specified). |
The secondary output table contains mean absolute shapley values for each input feature.
| Column Name | Data Type | Description |
|---|---|---|
| input_columns | DOUBLE PRECISION | Calculated global shap value determined by model. |
| label | INTEGER | The label specified for the class (this column is supported ONLY for TD_DecisionForest and TD_XGBoost classification models). |