I have constructed ensemble bagged tree model in the classification learner app in Matlab. The model is to predict binary variable (0 or 1). Most of the predictors are categorical. (62 records, 5 folded cross validation, 40 variables./ 2 were excluded when fitted the model). I have exported the model (named: trainedModel)
In order to interpret the model, I'm planning to use PDP (partial dependence plot). I do not know what is the right way to do it, I need your valuable help on that. I am using the following code:
Mdl = trainedModel.ClassificationEnsemble;
pd = plotPartialDependence(Mdl,{'Age','exp_val'}); %age and exp_val are ordinal variables
- the data are clean and there are no missing values
- When ran the code error message says:
Error using classreg.learning.classif.FullClassificationModel/plotPartialDependence (line 391)
Not enough input arguments.