public class FoldBasedSubsetInstruction extends Instruction
ReproducibleInstances object.
The original dataset is split into folds using the specified fold technique, e.g., 10-fold cross validation or a simple stratified split.
Then, a subset of the created folds is joined as to determine the desired subset of the original set.
For example, to obtain the training data in a 5-fold cross validation if the test data corresponds to the 3rd fold, the foldTechnique is "| Modifier and Type | Field and Description |
|---|---|
static java.lang.String |
COMMAND_NAME
Constant string to identify this instruction.
|
| Constructor and Description |
|---|
FoldBasedSubsetInstruction(java.lang.String foldTechnique,
int... outIndices)
Constructor to create a split Instruction that can be converted into json.
|
getCommand, getInputs, setCommand, setInputspublic static final java.lang.String COMMAND_NAME
public FoldBasedSubsetInstruction(java.lang.String foldTechnique,
int... outIndices)
foldTechnique - method used to compute the foldsoutIndex - index of the portion to use in the following