LAD method predicts not existing classes in training data

[AlexandreMoulti]

Hi,

I tried to used LAD method and it predicts not existing classes. I discovered this issue while trying to plot the confusion matrix. Indeed I have encoded classes between 5 and 14. But LAD method predicts sometimes a class 4 which should not exist.

Have I done a mistake ? Should I reclassify this 4 class into a 5 class (less qualitative)?

Regards

[fabianp]

this package assumes that your classes are between 0 and k-1, where k is the total number of classes