The Sample Data Subsetting Processor draws a random subset of the input data set. It may be used for splitting the data randomly in training and test data for forecasting evaluation (with e.g. Decision Tree Classification Forecast Processor, Improved Linear Regression (Forecasting) Processor).
The processor can operate on any dataset.
A table with random observations pulled from the input data set according to the configuration.
As configured, the output data set contains six rows. Using the seed 10, a random sample consisting of rows 1 and 3 to 7 of the original data was chosen. Every time the workflow gets executed using this configuration and input data, the same result will appear.