Description
The Partition node is used to split a data set randomly, using a predefined seed number. A new column will be created with the value 'TEST' or 'TRAIN' for each row.
Configuration Options
Basic Configuration Options
| Setting | Description\Parameters |
|---|---|
Training Fraction |
Percent of the data set that will have the value 'TRAIN' assigned to it. |
Test Fraction |
Percent of the data set that will have the value 'TEST' assigned to it. |
Random Seed |
Number used to seed the random split between the test and training fraction. Changing this number will change the random distribution. |