- IDColumn
- Column that contains a unique identifier for each row in the input table.Mandatory when Seed argument is present so that the output of TD_TrainTestSplit is deterministic across multiple function calls.
- TrainSize
- Size of the train dataset. It accepts float values in the (0, 1) range. Default is 0.75.
- TestSize
- Size of the test dataset. It accepts float values in the (0, 1) range. Default is 0.25.
- Seed
- Seed value that controls the shuffling applied to the data before applying the split. Pass an INT for reproducible output across multiple function calls. When the argument is not specified, different runs of the query generate different outputs. The random seed value must be in the range of 0 to INT_MAX.
- StratifyColumn
- Column name that contains the labels indicating which data needs to be stratified.
- If both trainSize and testSize arguments are specified, then their sum must be equal to 1.
- TrainSize and the TestSize must be greater than the number of classes when using stratify.
- If the input table does not have an identifier column, then TD_FillRowID can be used to generate one.