Make what is currently tab "Predicted text" a "sub-tab" of "Model performance". Have the confusion matrix in there too.
Retain the "Predicted text" tab, although replace the table with the predictions on the test dataset with a table with the predictions on unseen data. This would require reticulate in order to have the fitted model make the predictions on the fly.
From #32:
reticulatein order to have the fitted model make the predictions on the fly.