The CrossValidationSamplingType enumeration is used to determine cross-validation method.
It is used by the following properties:
Value | Brief description |
0 | Kfold. K-fold cross-validation. |
1 | LeaveOneOut. Leave-one-out cross-validation. |
2 | RandomSampling. Repeated random sub-sampling cross-validation. |
Detailed description of cross-validation methods:
K-fold Cross-Validation. Source data is divided into K folds that are equal in size: one fold is used for testing, the rest K-1 are used for training. The procedure is repeated K times, each of folds is used once as a test set. The K results are output, each for one fold, the resulting accuracy assessment is an averaged value by all folds.
Use the NumberOfFolds property to set value of the K parameter.
Leave-One-Out Cross-Validation. A single value is used as a custom test data set, the other observations from source data are used for training. Training and checking the predicted value is repeated as many times as there is the number of objects with the known classification in the sample. The method does not contain additional parameters.
Repeated Random Sub-Sampling Cross-Validation. Source data is randomly divided into a training set and a test set.
Use the TrainingSetSize property to set training set size and the NumberOfRandomTests property to set the number of cross-validation procedure repetitions.
See also: