The SDL Component Suite is an industry leading collection of components supporting scientific and engineering computing. Please visit the SDL Web site for more information....


Home MathPack Statis Class TMLRModel Methods LoadModel
See also: StoreModel, ApplyModel 

LoadModel

Unit: SDL_statis
Class: TMLRModel
Declaration: function LoadModel (FName: String): integer;

The method LoadModel loads the following parameters of the regression model from a binary file which has been stored by StoreModel:
  • the number of independent variables
  • the quality of fit
  • the standard deviation of the residuals
  • the F statistic of the model
  • the names of the variables
  • the regression coefficients
  • minimum and maximum of the independent variables
  • mean and variance of the variables

The method returns the following error codes:

>0 ... everything is OK, the returned number is the number of independent variables used in the model
-1 ... file not found
-2 ... model parameters are not MLR parameters (either LDA or RR)

Hint: Please note that loading a model does not load any data but only the parameters which are necessary to apply a trained model.


Last Update: 2023-Dec-05