Arguments
This function internally uses scikit-learn function DecisionTreeClassifier through teradataml Open source ML functions.
Transformed data won’t have features, rawPrediction, probability columns.
| PySpark Argument Name | Open Source Function Argument Name | Notes |
|---|---|---|
| predictionCol | Not yet available. | |
| probabilityCol | Not yet available. | |
| rawPredictionCol | Not yet available. | |
| maxDepth | max_depth | |
| maxBins | Not yet available. | |
| minInstancesPerNode | min_samples_split | PySpark takes int value (>= 1) default value is 1, teradatamlspk supports int value (>1) or float in (0.0, 1.0]. Default value is 1.0. |
| cacheNodeIds | Not yet available. | |
| checkpointInterval | Not yet available. | |
| impurity | criterion | |
| seed | random_state | |
| weightCol | class_weight | |
| leafCol | Not yet available. | |
| minWeightFractionPerNode | min_weight_fraction_leaf |