Testing set in Hierarchical Classification with Cross Validation

Hello, I'm trying to apply a Hierarchical Classification, and it works with the Cross-Validation (Training-Validation set) operator. But, the problem is when I'm using the testing set, because Performance operator does not show results for each nodo. Instead, it shows like a multi-label classification, then seems like when testing it is not doing a Hierarchical Classification.
Could you please help what's the correct form to use Hierarchical Classificator, with Cross-Validation (k-folds) with 70 % for training and 30% for testing?

This is a modified example with the Iris-Dataset.