I have just trained a model with Huggingface’s own Autotrain. But I am confused. I am classifying paragraphs of text into Complaint/Objection/Refund etc but from what I can tell there’s no way to get an absolute probability metric of what category the text falls into (i.e. a confidence level).
My main issue is that I don’t know how to specify a category called “Other”/“Miscellaneous” where there is low certainty that the text matches one of the main categories.