Cross Validation on Training Set

If no test dataset is available, an n-fold cross validation can be carried out on the training dataset. This is started via Model → Cross Validation on Training Set in the main window

The dataset is split into n equal parts, according to the value of "Number of Cross Validations" you entered.

For each output, an "n-fold Cross Validation Output <output>" opens with n plots. Each plot is the result of an individual model training on n-1 datasets and the prediction on the dataset not used for modeling.

The of the "n-fold Cross Validation Output <Output>" window

  • Save as Bitmap

    This allows the plot in the respective window to be saved as a bitmap graphic format.

  • Close

    Closes the window.

The of the "n-fold Cross Validation Output <Output>" window

  • Online Help (F1)

    Opens the online help (<F1>).

See also

Cross Validation on Training Set