Split the input dataset into K groups. For each groupTake one fold as the reserve or validation data set.Use remaining folds as the training dataset