DataLab is a compact statistics package aimed at exploratory data analysis. Please visit the DataLab Web site for more information....

RBF Network - Apply Model

Command: Math -> Neural Netzwork -> RBF Network -> Apply Model...

In order to apply a trained network to unknown data some prerequisites must be met. DataLab assumes that the variables used in the model match the variables of the unknown data. This is ensured by comparing the names of the variables of the model and of the unknown data set. If no match can be established, DataLab issues a warning. Note, that renaming the variables prevents the successful application of a trained network.

After clicking Apply Model..., the user has to select from the trained networks stored previously on the disk (these trained networks can be stored after the training has been completed). Next, the user has to define into which column the results should be entered. The calculated output values are then written into the selected column of the data matrix.

Hint: In order to avoid overwriting existing data columns it is recommended that the user creates some spare data columns by increasing the number of columns of the data matrix (command Resize).

Last Update: 2012-Jul-25